Rebaseupstream/users/avillega/gsymdwarf-include-end_sequence-debug_line-rows-when-looking-up-addr-ranges

Created using spr 1.3.5

2697 files changed, 122727 insertions, 48163 deletions

diff --git a/.ci/generate-buildkite-pipeline-premerge b/.ci/generate-buildkite-pipeline-premerge
index 81e9246de9b5..78a9cb77ff7d 100755
--- a/.ci/generate-buildkite-pipeline-premerge
+++ b/.ci/generate-buildkite-pipeline-premerge
@@ -91,7 +91,7 @@ function add-dependencies() {
     echo "${project}"
     case ${project} in
     bolt)
-      for p in lld llvm; do
+      for p in clang lld llvm; do
         echo $p
       done
     ;;
diff --git a/.ci/monolithic-linux.sh b/.ci/monolithic-linux.sh
index b347c443da67..b00a4b984a1d 100755
--- a/.ci/monolithic-linux.sh
+++ b/.ci/monolithic-linux.sh
@@ -48,7 +48,6 @@ cmake -S "${MONOREPO_ROOT}"/llvm -B "${BUILD_DIR}" \
       -D LLVM_LIT_ARGS="-v --xunit-xml-output ${BUILD_DIR}/test-results.xml --timeout=1200 --time-tests" \
       -D LLVM_ENABLE_LLD=ON \
       -D CMAKE_CXX_FLAGS=-gmlt \
-      -D BOLT_CLANG_EXE=/usr/bin/clang \
       -D LLVM_CCACHE_BUILD=ON \
       -D MLIR_ENABLE_BINDINGS_PYTHON=ON
 
diff --git a/.github/CODEOWNERS b/.github/CODEOWNERS
index 45da8af51bb9..0f178df1d18f 100644
--- a/.github/CODEOWNERS
+++ b/.github/CODEOWNERS
@@ -35,6 +35,10 @@
 clang/lib/AST/Interp/ @tbaederr
 clang/test/AST/Interp/ @tbaederr
 
+/clang/include/clang/CIR @lanza @bcardosolopes
+/clang/lib/CIR @lanza @bcardosolopes
+/clang/tools/cir-* @lanza @bcardosolopes
+
 /lldb/ @JDevlieghere
 
 # MLIR Interfaces.
diff --git a/.github/new-prs-labeler.yml b/.github/new-prs-labeler.yml
index 9cf64417d3cb..d608ea449f1d 100644
--- a/.github/new-prs-labeler.yml
+++ b/.github/new-prs-labeler.yml
@@ -1,3 +1,6 @@
+BOLT:
+  - bolt/**/*
+
 ClangIR:
   - clang/include/clang/CIR/**/*
   - clang/lib/CIR/**/*
@@ -467,6 +470,7 @@ backend:m68k:
 
 libc++:
   - libcxx/**
+  - .github/workflows/libcxx-*
 
 libc++abi:
   - libcxxabi/**
diff --git a/.github/workflows/libcxx-build-and-test.yaml b/.github/workflows/libcxx-build-and-test.yaml
index 1e9367732e59..44a3d79c72c0 100644
--- a/.github/workflows/libcxx-build-and-test.yaml
+++ b/.github/workflows/libcxx-build-and-test.yaml
@@ -61,12 +61,10 @@ jobs:
         ]
         cc: [  'clang-19' ]
         cxx: [ 'clang++-19' ]
-        clang_tidy: [ 'ON' ]
         include:
           - config: 'generic-gcc'
             cc: 'gcc-13'
             cxx: 'g++-13'
-            clang_tidy: 'OFF'
     steps:
       - uses: actions/checkout@v4
       - name: ${{ matrix.config }}.${{ matrix.cxx }}
@@ -74,7 +72,6 @@ jobs:
         env:
           CC: ${{ matrix.cc }}
           CXX: ${{ matrix.cxx }}
-          ENABLE_CLANG_TIDY: ${{ matrix.clang_tidy }}
       - uses: actions/upload-artifact@26f96dfa697d77e81fd5907df203aa23a56210a8 # v4.3.0
         if: always()
         with:
@@ -102,20 +99,16 @@ jobs:
         ]
         cc: [ 'clang-19' ]
         cxx: [ 'clang++-19' ]
-        clang_tidy: [ 'ON' ]
         include:
           - config: 'generic-gcc-cxx11'
             cc: 'gcc-13'
             cxx: 'g++-13'
-            clang_tidy: 'OFF'
           - config: 'generic-cxx23'
             cc: 'clang-17'
             cxx: 'clang++-17'
-            clang_tidy: 'OFF'
           - config: 'generic-cxx26'
             cc: 'clang-18'
             cxx: 'clang++-18'
-            clang_tidy: 'ON'
     steps:
       - uses: actions/checkout@v4
       - name: ${{ matrix.config }}
@@ -123,7 +116,6 @@ jobs:
         env:
           CC: ${{ matrix.cc }}
           CXX: ${{ matrix.cxx }}
-          ENABLE_CLANG_TIDY: ${{ matrix.clang_tidy }}
       - uses: actions/upload-artifact@26f96dfa697d77e81fd5907df203aa23a56210a8 # v4.3.0
         if: always()  # Upload artifacts even if the build or test suite fails
         with:
@@ -188,7 +180,6 @@ jobs:
         env:
           CC: clang-19
           CXX: clang++-19
-          ENABLE_CLANG_TIDY: "OFF"
       - uses: actions/upload-artifact@26f96dfa697d77e81fd5907df203aa23a56210a8 # v4.3.0
         if: always()
         with:
diff --git a/bolt/include/bolt/Passes/BinaryPasses.h b/bolt/include/bolt/Passes/BinaryPasses.h
index 8d89ef8b5484..5d7692559eda 100644
--- a/bolt/include/bolt/Passes/BinaryPasses.h
+++ b/bolt/include/bolt/Passes/BinaryPasses.h
@@ -400,8 +400,7 @@ public:
 /// dyno stats categories.
 class PrintProgramStats : public BinaryFunctionPass {
 public:
-  explicit PrintProgramStats(const cl::opt<bool> &PrintPass)
-      : BinaryFunctionPass(PrintPass) {}
+  explicit PrintProgramStats() : BinaryFunctionPass(false) {}
 
   const char *getName() const override { return "print-stats"; }
   bool shouldPrint(const BinaryFunction &) const override { return false; }
diff --git a/bolt/include/bolt/Rewrite/RewriteInstance.h b/bolt/include/bolt/Rewrite/RewriteInstance.h
index af832b4c7c84..f4bffba96b1d 100644
--- a/bolt/include/bolt/Rewrite/RewriteInstance.h
+++ b/bolt/include/bolt/Rewrite/RewriteInstance.h
@@ -424,8 +424,12 @@ private:
 
   /// Common section names.
   static StringRef getEHFrameSectionName() { return ".eh_frame"; }
+  static StringRef getEHFrameHdrSectionName() { return ".eh_frame_hdr"; }
   static StringRef getRelaDynSectionName() { return ".rela.dyn"; }
 
+  /// FILE symbol name used for local fragments of global functions.
+  static StringRef getBOLTFileSymbolName() { return "bolt-pseudo.o"; }
+
   /// An instance of the input binary we are processing, externally owned.
   llvm::object::ELFObjectFileBase *InputFile;
 
diff --git a/bolt/lib/Profile/DataAggregator.cpp b/bolt/lib/Profile/DataAggregator.cpp
index 0b2a4e86561f..70e324cc0165 100644
--- a/bolt/lib/Profile/DataAggregator.cpp
+++ b/bolt/lib/Profile/DataAggregator.cpp
@@ -14,6 +14,7 @@
 #include "bolt/Profile/DataAggregator.h"
 #include "bolt/Core/BinaryContext.h"
 #include "bolt/Core/BinaryFunction.h"
+#include "bolt/Passes/BinaryPasses.h"
 #include "bolt/Profile/BoltAddressTranslation.h"
 #include "bolt/Profile/Heatmap.h"
 #include "bolt/Profile/YAMLProfileWriter.h"
@@ -611,6 +612,7 @@ Error DataAggregator::readProfile(BinaryContext &BC) {
         if (std::error_code EC = writeBATYAML(BC, opts::SaveProfile))
           report_error("cannot create output data file", EC);
     }
+    BC.logBOLTErrorsAndQuitOnFatal(PrintProgramStats().runOnFunctions(BC));
   }
 
   return Error::success();
diff --git a/bolt/lib/Rewrite/BinaryPassManager.cpp b/bolt/lib/Rewrite/BinaryPassManager.cpp
index be4888ccfa56..cbb7199a53dd 100644
--- a/bolt/lib/Rewrite/BinaryPassManager.cpp
+++ b/bolt/lib/Rewrite/BinaryPassManager.cpp
@@ -356,7 +356,7 @@ Error BinaryFunctionPassManager::runAllPasses(BinaryContext &BC) {
   // order they're registered.
 
   // Run this pass first to use stats for the original functions.
-  Manager.registerPass(std::make_unique<PrintProgramStats>(NeverPrint));
+  Manager.registerPass(std::make_unique<PrintProgramStats>());
 
   if (opts::PrintProfileStats)
     Manager.registerPass(std::make_unique<PrintProfileStats>(NeverPrint));
diff --git a/bolt/lib/Rewrite/BoltDiff.cpp b/bolt/lib/Rewrite/BoltDiff.cpp
index fa43b7a2f92c..74b5ca18abce 100644
--- a/bolt/lib/Rewrite/BoltDiff.cpp
+++ b/bolt/lib/Rewrite/BoltDiff.cpp
@@ -292,7 +292,7 @@ class RewriteInstanceDiff {
         }
       }
     }
-    PrintProgramStats PPS(opts::NeverPrint);
+    PrintProgramStats PPS;
     outs() << "* BOLT-DIFF: Starting print program stats pass for binary 1\n";
     RI1.BC->logBOLTErrorsAndQuitOnFatal(PPS.runOnFunctions(*RI1.BC));
     outs() << "* BOLT-DIFF: Starting print program stats pass for binary 2\n";
diff --git a/bolt/lib/Rewrite/RewriteInstance.cpp b/bolt/lib/Rewrite/RewriteInstance.cpp
index 4e0096cf988a..065260936e70 100644
--- a/bolt/lib/Rewrite/RewriteInstance.cpp
+++ b/bolt/lib/Rewrite/RewriteInstance.cpp
@@ -1725,12 +1725,6 @@ void RewriteInstance::adjustFunctionBoundaries() {
       if (!Function.isSymbolValidInScope(Symbol, SymbolSize))
         break;
 
-      // Ignore unnamed symbols. Used, for example, by debugging info on RISC-V.
-      if (BC->isRISCV() && cantFail(Symbol.getName()).empty()) {
-        ++NextSymRefI;
-        continue;
-      }
-
       // Skip basic block labels. This happens on RISC-V with linker relaxation
       // enabled because every branch needs a relocation and corresponding
       // symbol. We don't want to add such symbols as entry points.
@@ -3932,11 +3926,6 @@ void RewriteInstance::patchELFPHDRTable() {
 
   OS.seek(PHDRTableOffset);
 
-  bool ModdedGnuStack = false;
-  (void)ModdedGnuStack;
-  bool AddedSegment = false;
-  (void)AddedSegment;
-
   auto createNewTextPhdr = [&]() {
     ELF64LEPhdrTy NewPhdr;
     NewPhdr.p_type = ELF::PT_LOAD;
@@ -3952,40 +3941,53 @@ void RewriteInstance::patchELFPHDRTable() {
     NewPhdr.p_filesz = NewTextSegmentSize;
     NewPhdr.p_memsz = NewTextSegmentSize;
     NewPhdr.p_flags = ELF::PF_X | ELF::PF_R;
-    // FIXME: Currently instrumentation is experimental and the runtime data
-    // is emitted with code, thus everything needs to be writable
-    if (opts::Instrument)
+    if (opts::Instrument) {
+      // FIXME: Currently instrumentation is experimental and the runtime data
+      // is emitted with code, thus everything needs to be writable.
       NewPhdr.p_flags |= ELF::PF_W;
+    }
     NewPhdr.p_align = BC->PageAlign;
 
     return NewPhdr;
   };
 
-  auto createNewWritableSectionsPhdr = [&]() {
-    ELF64LEPhdrTy NewPhdr;
-    NewPhdr.p_type = ELF::PT_LOAD;
-    NewPhdr.p_offset = getFileOffsetForAddress(NewWritableSegmentAddress);
-    NewPhdr.p_vaddr = NewWritableSegmentAddress;
-    NewPhdr.p_paddr = NewWritableSegmentAddress;
-    NewPhdr.p_filesz = NewWritableSegmentSize;
-    NewPhdr.p_memsz = NewWritableSegmentSize;
-    NewPhdr.p_align = BC->RegularPageSize;
-    NewPhdr.p_flags = ELF::PF_R | ELF::PF_W;
-    return NewPhdr;
+  auto writeNewSegmentPhdrs = [&]() {
+    ELF64LE::Phdr NewTextPhdr = createNewTextPhdr();
+    OS.write(reinterpret_cast<const char *>(&NewTextPhdr), sizeof(NewTextPhdr));
+
+    if (NewWritableSegmentSize) {
+      ELF64LEPhdrTy NewPhdr;
+      NewPhdr.p_type = ELF::PT_LOAD;
+      NewPhdr.p_offset = getFileOffsetForAddress(NewWritableSegmentAddress);
+      NewPhdr.p_vaddr = NewWritableSegmentAddress;
+      NewPhdr.p_paddr = NewWritableSegmentAddress;
+      NewPhdr.p_filesz = NewWritableSegmentSize;
+      NewPhdr.p_memsz = NewWritableSegmentSize;
+      NewPhdr.p_align = BC->RegularPageSize;
+      NewPhdr.p_flags = ELF::PF_R | ELF::PF_W;
+      OS.write(reinterpret_cast<const char *>(&NewPhdr), sizeof(NewPhdr));
+    }
   };
 
+  bool ModdedGnuStack = false;
+  bool AddedSegment = false;
+
   // Copy existing program headers with modifications.
   for (const ELF64LE::Phdr &Phdr : cantFail(Obj.program_headers())) {
     ELF64LE::Phdr NewPhdr = Phdr;
-    if (PHDRTableAddress && Phdr.p_type == ELF::PT_PHDR) {
-      NewPhdr.p_offset = PHDRTableOffset;
-      NewPhdr.p_vaddr = PHDRTableAddress;
-      NewPhdr.p_paddr = PHDRTableAddress;
-      NewPhdr.p_filesz = sizeof(NewPhdr) * Phnum;
-      NewPhdr.p_memsz = sizeof(NewPhdr) * Phnum;
-    } else if (Phdr.p_type == ELF::PT_GNU_EH_FRAME) {
-      ErrorOr<BinarySection &> EHFrameHdrSec =
-          BC->getUniqueSectionByName(getNewSecPrefix() + ".eh_frame_hdr");
+    switch (Phdr.p_type) {
+    case ELF::PT_PHDR:
+      if (PHDRTableAddress) {
+        NewPhdr.p_offset = PHDRTableOffset;
+        NewPhdr.p_vaddr = PHDRTableAddress;
+        NewPhdr.p_paddr = PHDRTableAddress;
+        NewPhdr.p_filesz = sizeof(NewPhdr) * Phnum;
+        NewPhdr.p_memsz = sizeof(NewPhdr) * Phnum;
+      }
+      break;
+    case ELF::PT_GNU_EH_FRAME: {
+      ErrorOr<BinarySection &> EHFrameHdrSec = BC->getUniqueSectionByName(
+          getNewSecPrefix() + getEHFrameHdrSectionName());
       if (EHFrameHdrSec && EHFrameHdrSec->isAllocatable() &&
           EHFrameHdrSec->isFinalized()) {
         NewPhdr.p_offset = EHFrameHdrSec->getOutputFileOffset();
@@ -3994,37 +3996,36 @@ void RewriteInstance::patchELFPHDRTable() {
         NewPhdr.p_filesz = EHFrameHdrSec->getOutputSize();
         NewPhdr.p_memsz = EHFrameHdrSec->getOutputSize();
       }
-    } else if (opts::UseGnuStack && Phdr.p_type == ELF::PT_GNU_STACK) {
-      NewPhdr = createNewTextPhdr();
-      ModdedGnuStack = true;
-    } else if (!opts::UseGnuStack && Phdr.p_type == ELF::PT_DYNAMIC) {
-      // Insert the new header before DYNAMIC.
-      ELF64LE::Phdr NewTextPhdr = createNewTextPhdr();
-      OS.write(reinterpret_cast<const char *>(&NewTextPhdr),
-               sizeof(NewTextPhdr));
-      if (NewWritableSegmentSize) {
-        ELF64LEPhdrTy NewWritablePhdr = createNewWritableSectionsPhdr();
-        OS.write(reinterpret_cast<const char *>(&NewWritablePhdr),
-                 sizeof(NewWritablePhdr));
+      break;
+    }
+    case ELF::PT_GNU_STACK:
+      if (opts::UseGnuStack) {
+        // Overwrite the header with the new text segment header.
+        NewPhdr = createNewTextPhdr();
+        ModdedGnuStack = true;
       }
-      AddedSegment = true;
+      break;
+    case ELF::PT_DYNAMIC:
+      if (!opts::UseGnuStack) {
+        // Insert new headers before DYNAMIC.
+        writeNewSegmentPhdrs();
+        AddedSegment = true;
+      }
+      break;
     }
     OS.write(reinterpret_cast<const char *>(&NewPhdr), sizeof(NewPhdr));
   }
 
   if (!opts::UseGnuStack && !AddedSegment) {
-    // Append the new header to the end of the table.
-    ELF64LE::Phdr NewTextPhdr = createNewTextPhdr();
-    OS.write(reinterpret_cast<const char *>(&NewTextPhdr), sizeof(NewTextPhdr));
-    if (NewWritableSegmentSize) {
-      ELF64LEPhdrTy NewWritablePhdr = createNewWritableSectionsPhdr();
-      OS.write(reinterpret_cast<const char *>(&NewWritablePhdr),
-               sizeof(NewWritablePhdr));
-    }
+    // Append new headers to the end of the table.
+    writeNewSegmentPhdrs();
   }
 
-  assert((!opts::UseGnuStack || ModdedGnuStack) &&
-         "could not find GNU_STACK program header to modify");
+  if (opts::UseGnuStack && !ModdedGnuStack) {
+    BC->errs()
+        << "BOLT-ERROR: could not find PT_GNU_STACK program header to modify\n";
+    exit(1);
+  }
 }
 
 namespace {
@@ -4493,6 +4494,8 @@ void RewriteInstance::updateELFSymbolTable(
   // Symbols for the new symbol table.
   std::vector<ELFSymTy> Symbols;
 
+  bool EmittedColdFileSymbol = false;
+
   auto getNewSectionIndex = [&](uint32_t OldIndex) {
     // For dynamic symbol table, the section index could be wrong on the input,
     // and its value is ignored by the runtime if it's different from
@@ -4551,6 +4554,20 @@ void RewriteInstance::updateELFSymbolTable(
       Symbols.emplace_back(ICFSymbol);
     }
     if (Function.isSplit()) {
+      // Prepend synthetic FILE symbol to prevent local cold fragments from
+      // colliding with existing symbols with the same name.
+      if (!EmittedColdFileSymbol &&
+          FunctionSymbol.getBinding() == ELF::STB_GLOBAL) {
+        ELFSymTy FileSymbol;
+        FileSymbol.st_shndx = ELF::SHN_ABS;
+        FileSymbol.st_name = AddToStrTab(getBOLTFileSymbolName());
+        FileSymbol.st_value = 0;
+        FileSymbol.st_size = 0;
+        FileSymbol.st_other = 0;
+        FileSymbol.setBindingAndType(ELF::STB_LOCAL, ELF::STT_FILE);
+        Symbols.emplace_back(FileSymbol);
+        EmittedColdFileSymbol = true;
+      }
       for (const FunctionFragment &FF :
            Function.getLayout().getSplitFragments()) {
         if (FF.getAddress()) {
@@ -5432,6 +5449,17 @@ uint64_t RewriteInstance::getNewFunctionOrDataAddress(uint64_t OldAddress) {
   if (BD && BD->isMoved())
     return BD->getOutputAddress();
 
+  if (const BinaryFunction *BF =
+          BC->getBinaryFunctionContainingAddress(OldAddress)) {
+    if (BF->isEmitted()) {
+      BC->errs() << "BOLT-ERROR: unable to get new address corresponding to "
+                    "input address 0x"
+                 << Twine::utohexstr(OldAddress) << " in function " << *BF
+                 << ". Consider adding this function to --skip-funcs=...\n";
+      exit(1);
+    }
+  }
+
   return 0;
 }
 
@@ -5671,7 +5699,8 @@ void RewriteInstance::writeEHFrameHeader() {
       BC->AsmInfo->getCodePointerSize()));
   check_error(std::move(Er), "failed to parse EH frame");
 
-  LLVM_DEBUG(dbgs() << "BOLT: writing a new .eh_frame_hdr\n");
+  LLVM_DEBUG(dbgs() << "BOLT: writing a new " << getEHFrameHdrSectionName()
+                    << '\n');
 
   NextAvailableAddress =
       appendPadding(Out->os(), NextAvailableAddress, EHFrameHdrAlign);
@@ -5689,16 +5718,17 @@ void RewriteInstance::writeEHFrameHeader() {
   const unsigned Flags = BinarySection::getFlags(/*IsReadOnly=*/true,
                                                  /*IsText=*/false,
                                                  /*IsAllocatable=*/true);
-  BinarySection *OldEHFrameHdrSection = getSection(".eh_frame_hdr");
+  BinarySection *OldEHFrameHdrSection = getSection(getEHFrameHdrSectionName());
   if (OldEHFrameHdrSection)
-    OldEHFrameHdrSection->setOutputName(getOrgSecPrefix() + ".eh_frame_hdr");
+    OldEHFrameHdrSection->setOutputName(getOrgSecPrefix() +
+                                        getEHFrameHdrSectionName());
 
   BinarySection &EHFrameHdrSec = BC->registerOrUpdateSection(
-      getNewSecPrefix() + ".eh_frame_hdr", ELF::SHT_PROGBITS, Flags, nullptr,
-      NewEHFrameHdr.size(), /*Alignment=*/1);
+      getNewSecPrefix() + getEHFrameHdrSectionName(), ELF::SHT_PROGBITS, Flags,
+      nullptr, NewEHFrameHdr.size(), /*Alignment=*/1);
   EHFrameHdrSec.setOutputFileOffset(EHFrameHdrFileOffset);
   EHFrameHdrSec.setOutputAddress(EHFrameHdrOutputAddress);
-  EHFrameHdrSec.setOutputName(".eh_frame_hdr");
+  EHFrameHdrSec.setOutputName(getEHFrameHdrSectionName());
 
   NextAvailableAddress += EHFrameHdrSec.getOutputSize();
 
diff --git a/bolt/test/RISCV/unnamed-sym-no-entry.c b/bolt/test/RISCV/fake-label-no-entry.c
index 605bbc00aeec..bd125263101b 100644
--- a/bolt/test/RISCV/unnamed-sym-no-entry.c
+++ b/bolt/test/RISCV/fake-label-no-entry.c
@@ -5,12 +5,12 @@
 
 // RUN: %clang %cflags -g -Wl,-q -o %t %s
 
-/// Verify that the binary indeed contains an unnamed symbol at _start
+/// Verify that the binary indeed contains a fake label ".L0 " at _start.
 // RUN: llvm-readelf -s %t | FileCheck %s --check-prefix=CHECK-ELF
 // CHECK-ELF-DAG: [[#%x,START:]] {{.*}} FUNC GLOBAL DEFAULT [[#%d,SECTION:]] _start{{$}}
-// CHECK-ELF-DAG: [[#%x,START]] {{.*}} NOTYPE LOCAL DEFAULT [[#SECTION]] {{$}}
+// CHECK-ELF-DAG: [[#%x,START]] {{.*}} NOTYPE LOCAL DEFAULT [[#SECTION]] .L0 {{$}}
 
-/// Verify that BOLT did not create an extra entry point for the unnamed symbol
+/// Verify that BOLT did not create an extra entry point for the fake label.
 // RUN: llvm-bolt -o %t.bolt %t --print-cfg | FileCheck %s
 // CHECK: Binary Function "_start" after building cfg {
 // CHECK:  IsMultiEntry: 0
diff --git a/bolt/test/X86/cdsplit-symbol-names.s b/bolt/test/X86/cdsplit-symbol-names.s
index e2259276e255..e53863e22246 100644
--- a/bolt/test/X86/cdsplit-symbol-names.s
+++ b/bolt/test/X86/cdsplit-symbol-names.s
@@ -10,6 +10,7 @@
 # RUN:         --call-scale=2 --data=%t.fdata --reorder-blocks=ext-tsp
 # RUN: llvm-objdump --syms %t.bolt | FileCheck %s --check-prefix=CHECK-SYMS-WARM
 
+# CHECK-SYMS-WARM: 0000000000000000 l df *ABS* 0000000000000000 bolt-pseudo.o
 # CHECK-SYMS-WARM: .text.warm
 # CHECK-SYMS-WARM-SAME: chain.warm
 # CHECK-SYMS-WARM: .text.cold
diff --git a/bolt/test/X86/indirect-goto-pie.test b/bolt/test/X86/indirect-goto-pie.test
new file mode 100644
index 000000000000..039ff5c41d3d
--- /dev/null
+++ b/bolt/test/X86/indirect-goto-pie.test
@@ -0,0 +1,16 @@
+# Check that llvm-bolt fails to process PIC binaries with computed goto, as the
+# support is not there yet for correctly updating dynamic relocations
+# referencing code inside functions.
+
+REQUIRES: x86_64-linux
+
+RUN: %clang %S/Inputs/indirect_goto.c -o %t -fpic -pie -Wl,-q
+RUN: not llvm-bolt %t -o %t.bolt --relocs=1 --print-cfg --print-only=main \
+RUN:   |& FileCheck %s
+
+# Check that processing works if main() is skipped.
+RUN: llvm-bolt %t -o %t.bolt --relocs=1 --skip-funcs=main
+
+CHECK:  jmpq    *%rax # UNKNOWN CONTROL FLOW
+
+CHECK: BOLT-ERROR: unable to get new address
diff --git a/bolt/test/X86/pre-aggregated-perf.test b/bolt/test/X86/pre-aggregated-perf.test
index e8c3f64239a2..0bd44720f1b7 100644
--- a/bolt/test/X86/pre-aggregated-perf.test
+++ b/bolt/test/X86/pre-aggregated-perf.test
@@ -11,7 +11,14 @@ REQUIRES: system-linux
 
 RUN: yaml2obj %p/Inputs/blarge.yaml &> %t.exe
 RUN: perf2bolt %t.exe -o %t --pa -p %p/Inputs/pre-aggregated.txt -w %t.new \
-RUN:   --profile-use-dfs
+RUN:   --profile-use-dfs | FileCheck %s
+
+RUN: llvm-bolt %t.exe -data %t -o %t.null | FileCheck %s
+RUN: llvm-bolt %t.exe -data %t.new -o %t.null | FileCheck %s
+RUN: llvm-bolt %t.exe -p %p/Inputs/pre-aggregated.txt --pa -o %t.null | FileCheck %s
+
+CHECK: BOLT-INFO: 4 out of 7 functions in the binary (57.1%) have non-empty execution profile
+
 RUN: cat %t | sort | FileCheck %s -check-prefix=PERF2BOLT
 RUN: cat %t.new | FileCheck %s -check-prefix=NEWFORMAT
 
diff --git a/bolt/test/X86/shrinkwrapping-do-not-pessimize.s b/bolt/test/X86/shrinkwrapping-do-not-pessimize.s
index a57131131423..3fdd5f5e38fe 100644
--- a/bolt/test/X86/shrinkwrapping-do-not-pessimize.s
+++ b/bolt/test/X86/shrinkwrapping-do-not-pessimize.s
@@ -53,6 +53,6 @@ end_if_1:
   .size _start, .-_start
 
   .data
-rel:  .quad end_if_1
+rel:  .quad _start
 
 # CHECK:   BOLT-INFO: Shrink wrapping moved 0 spills inserting load/stores and 0 spills inserting push/pops
diff --git a/bolt/test/runtime/X86/Inputs/indirect_goto.c b/bolt/test/runtime/X86/Inputs/indirect_goto.c
deleted file mode 100644
index b781e9e03b6d..000000000000
--- a/bolt/test/runtime/X86/Inputs/indirect_goto.c
+++ /dev/null
@@ -1,18 +0,0 @@
-int main(int argc, char *argv[]) {
-  static const void *T1[] = { &&L1, &&L2 };
-  static const void *T2[] = { &&L2, &&L3 };
-
-  const void **T = (argc > 1) ? T1 : T2;
-
-  int i = 0;
-
-L0:
-  goto *T[argc];
-L1:
-  ++i;
-L2:
-  i++;
-L3:
-  i++;
-  return i;
-}
diff --git a/bolt/test/runtime/X86/indirect-goto-pie.test b/bolt/test/runtime/X86/indirect-goto-pie.test
deleted file mode 100644
index 76089fda3abf..000000000000
--- a/bolt/test/runtime/X86/indirect-goto-pie.test
+++ /dev/null
@@ -1,10 +0,0 @@
-# Check llvm-bolt processes binaries compiled from sources that use indirect goto.
-REQUIRES: x86_64-linux
-
-RUN: %clang %S/Inputs/indirect_goto.c -o %t -fpic -pie -Wl,-q
-RUN: llvm-bolt %t -o %t.bolt --relocs=1 --print-cfg --print-only=main \
-RUN:   |& FileCheck %s 
-# The test fails as we don't update corresponding dynamic relocations.
-RUN: not %t.bolt
-
-CHECK:  jmpq    *%rax # UNKNOWN CONTROL FLOW
diff --git a/clang-tools-extra/clang-tidy/ClangTidyCheck.cpp b/clang-tools-extra/clang-tidy/ClangTidyCheck.cpp
index 3e926236adb4..710b361e16c0 100644
--- a/clang-tools-extra/clang-tidy/ClangTidyCheck.cpp
+++ b/clang-tools-extra/clang-tidy/ClangTidyCheck.cpp
@@ -139,6 +139,12 @@ void ClangTidyCheck::OptionsView::storeInt(ClangTidyOptions::OptionMap &Options,
   store(Options, LocalName, llvm::itostr(Value));
 }
 
+void ClangTidyCheck::OptionsView::storeUnsigned(
+    ClangTidyOptions::OptionMap &Options, StringRef LocalName,
+    uint64_t Value) const {
+  store(Options, LocalName, llvm::utostr(Value));
+}
+
 template <>
 void ClangTidyCheck::OptionsView::store<bool>(
     ClangTidyOptions::OptionMap &Options, StringRef LocalName,
diff --git a/clang-tools-extra/clang-tidy/ClangTidyCheck.h b/clang-tools-extra/clang-tidy/ClangTidyCheck.h
index 656a2f008f6e..7427aa9bf48f 100644
--- a/clang-tools-extra/clang-tidy/ClangTidyCheck.h
+++ b/clang-tools-extra/clang-tidy/ClangTidyCheck.h
@@ -411,7 +411,10 @@ public:
     std::enable_if_t<std::is_integral_v<T>>
     store(ClangTidyOptions::OptionMap &Options, StringRef LocalName,
           T Value) const {
-      storeInt(Options, LocalName, Value);
+      if constexpr (std::is_signed_v<T>)
+        storeInt(Options, LocalName, Value);
+      else
+        storeUnsigned(Options, LocalName, Value);
     }
 
     /// Stores an option with the check-local name \p LocalName with
@@ -422,7 +425,7 @@ public:
     store(ClangTidyOptions::OptionMap &Options, StringRef LocalName,
           std::optional<T> Value) const {
       if (Value)
-        storeInt(Options, LocalName, *Value);
+        store(Options, LocalName, *Value);
       else
         store(Options, LocalName, "none");
     }
@@ -470,6 +473,8 @@ public:
     void storeInt(ClangTidyOptions::OptionMap &Options, StringRef LocalName,
                   int64_t Value) const;
 
+    void storeUnsigned(ClangTidyOptions::OptionMap &Options,
+                       StringRef LocalName, uint64_t Value) const;
 
     std::string NamePrefix;
     const ClangTidyOptions::OptionMap &CheckOptions;
diff --git a/clang-tools-extra/clang-tidy/bugprone/BugproneTidyModule.cpp b/clang-tools-extra/clang-tidy/bugprone/BugproneTidyModule.cpp
index 2931325d8b57..1b92d2e60cc1 100644
--- a/clang-tools-extra/clang-tidy/bugprone/BugproneTidyModule.cpp
+++ b/clang-tools-extra/clang-tidy/bugprone/BugproneTidyModule.cpp
@@ -54,6 +54,7 @@
 #include "PosixReturnCheck.h"
 #include "RedundantBranchConditionCheck.h"
 #include "ReservedIdentifierCheck.h"
+#include "ReturnConstRefFromParameterCheck.h"
 #include "SharedPtrArrayMismatchCheck.h"
 #include "SignalHandlerCheck.h"
 #include "SignedCharMisuseCheck.h"
@@ -137,6 +138,8 @@ public:
         "bugprone-inaccurate-erase");
     CheckFactories.registerCheck<IncorrectEnableIfCheck>(
         "bugprone-incorrect-enable-if");
+    CheckFactories.registerCheck<ReturnConstRefFromParameterCheck>(
+        "bugprone-return-const-ref-from-parameter");
     CheckFactories.registerCheck<SwitchMissingDefaultCaseCheck>(
         "bugprone-switch-missing-default-case");
     CheckFactories.registerCheck<IncDecInConditionsCheck>(
diff --git a/clang-tools-extra/clang-tidy/bugprone/CMakeLists.txt b/clang-tools-extra/clang-tidy/bugprone/CMakeLists.txt
index 081ba67efe15..2d303191f886 100644
--- a/clang-tools-extra/clang-tidy/bugprone/CMakeLists.txt
+++ b/clang-tools-extra/clang-tidy/bugprone/CMakeLists.txt
@@ -26,6 +26,7 @@ add_clang_library(clangTidyBugproneModule
   ImplicitWideningOfMultiplicationResultCheck.cpp
   InaccurateEraseCheck.cpp
   IncorrectEnableIfCheck.cpp
+  ReturnConstRefFromParameterCheck.cpp
   SuspiciousStringviewDataUsageCheck.cpp
   SwitchMissingDefaultCaseCheck.cpp
   IncDecInConditionsCheck.cpp
diff --git a/clang-tools-extra/clang-tidy/bugprone/EasilySwappableParametersCheck.cpp b/clang-tools-extra/clang-tidy/bugprone/EasilySwappableParametersCheck.cpp
index 84e99c7fafc7..10868129e76d 100644
--- a/clang-tools-extra/clang-tidy/bugprone/EasilySwappableParametersCheck.cpp
+++ b/clang-tools-extra/clang-tidy/bugprone/EasilySwappableParametersCheck.cpp
@@ -967,7 +967,8 @@ approximateStandardConversionSequence(const TheCheck &Check, QualType From,
   // Get out the qualifiers of the original type. This will always be
   // re-applied to the WorkType to ensure it is the same qualification as the
   // original From was.
-  auto QualifiersToApply = From.split().Quals.getAsOpaqueValue();
+  auto FastQualifiersToApply = static_cast<unsigned>(
+      From.split().Quals.getAsOpaqueValue() & Qualifiers::FastMask);
 
   // LValue->RValue is irrelevant for the check, because it is a thing to be
   // done at a call site, and will be performed if need be performed.
@@ -993,7 +994,7 @@ approximateStandardConversionSequence(const TheCheck &Check, QualType From,
     // "const double -> double".
     LLVM_DEBUG(llvm::dbgs()
                << "--- approximateStdConv. Conversion between numerics.\n");
-    WorkType = QualType{ToBuiltin, QualifiersToApply};
+    WorkType = QualType{ToBuiltin, FastQualifiersToApply};
   }
 
   const auto *FromEnum = WorkType->getAs<EnumType>();
@@ -1002,7 +1003,7 @@ approximateStandardConversionSequence(const TheCheck &Check, QualType From,
     // Unscoped enumerations (or enumerations in C) convert to numerics.
     LLVM_DEBUG(llvm::dbgs()
                << "--- approximateStdConv. Unscoped enum to numeric.\n");
-    WorkType = QualType{ToBuiltin, QualifiersToApply};
+    WorkType = QualType{ToBuiltin, FastQualifiersToApply};
   } else if (FromNumeric && ToEnum && ToEnum->isUnscopedEnumerationType()) {
     // Numeric types convert to enumerations only in C.
     if (Ctx.getLangOpts().CPlusPlus) {
@@ -1013,7 +1014,7 @@ approximateStandardConversionSequence(const TheCheck &Check, QualType From,
 
     LLVM_DEBUG(llvm::dbgs()
                << "--- approximateStdConv. Numeric to unscoped enum.\n");
-    WorkType = QualType{ToEnum, QualifiersToApply};
+    WorkType = QualType{ToEnum, FastQualifiersToApply};
   }
 
   // Check for pointer conversions.
@@ -1022,14 +1023,14 @@ approximateStandardConversionSequence(const TheCheck &Check, QualType From,
   if (FromPtr && ToPtr) {
     if (ToPtr->isVoidPointerType()) {
       LLVM_DEBUG(llvm::dbgs() << "--- approximateStdConv. To void pointer.\n");
-      WorkType = QualType{ToPtr, QualifiersToApply};
+      WorkType = QualType{ToPtr, FastQualifiersToApply};
     }
 
     const auto *FromRecordPtr = FromPtr->getPointeeCXXRecordDecl();
     const auto *ToRecordPtr = ToPtr->getPointeeCXXRecordDecl();
     if (isDerivedToBase(FromRecordPtr, ToRecordPtr)) {
       LLVM_DEBUG(llvm::dbgs() << "--- approximateStdConv. Derived* to Base*\n");
-      WorkType = QualType{ToPtr, QualifiersToApply};
+      WorkType = QualType{ToPtr, FastQualifiersToApply};
     }
   }
 
@@ -1039,7 +1040,7 @@ approximateStandardConversionSequence(const TheCheck &Check, QualType From,
   const auto *ToRecord = To->getAsCXXRecordDecl();
   if (isDerivedToBase(FromRecord, ToRecord)) {
     LLVM_DEBUG(llvm::dbgs() << "--- approximateStdConv. Derived To Base.\n");
-    WorkType = QualType{ToRecord->getTypeForDecl(), QualifiersToApply};
+    WorkType = QualType{ToRecord->getTypeForDecl(), FastQualifiersToApply};
   }
 
   if (Ctx.getLangOpts().CPlusPlus17 && FromPtr && ToPtr) {
@@ -1054,7 +1055,7 @@ approximateStandardConversionSequence(const TheCheck &Check, QualType From,
         !ToFunctionPtr->hasNoexceptExceptionSpec()) {
       LLVM_DEBUG(llvm::dbgs() << "--- approximateStdConv. noexcept function "
                                  "pointer to non-noexcept.\n");
-      WorkType = QualType{ToPtr, QualifiersToApply};
+      WorkType = QualType{ToPtr, FastQualifiersToApply};
     }
   }
 
diff --git a/clang-tools-extra/clang-tidy/bugprone/ReturnConstRefFromParameterCheck.cpp b/clang-tools-extra/clang-tidy/bugprone/ReturnConstRefFromParameterCheck.cpp
new file mode 100644
index 000000000000..8ae37d4f774d
--- /dev/null
+++ b/clang-tools-extra/clang-tidy/bugprone/ReturnConstRefFromParameterCheck.cpp
@@ -0,0 +1,34 @@
+//===--- ReturnConstRefFromParameterCheck.cpp - clang-tidy ----------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#include "ReturnConstRefFromParameterCheck.h"
+#include "../utils/Matchers.h"
+#include "clang/ASTMatchers/ASTMatchFinder.h"
+#include "clang/ASTMatchers/ASTMatchers.h"
+
+using namespace clang::ast_matchers;
+
+namespace clang::tidy::bugprone {
+
+void ReturnConstRefFromParameterCheck::registerMatchers(MatchFinder *Finder) {
+  Finder->addMatcher(
+      returnStmt(hasReturnValue(declRefExpr(to(parmVarDecl(hasType(
+                     hasCanonicalType(matchers::isReferenceToConst())))))))
+          .bind("ret"),
+      this);
+}
+
+void ReturnConstRefFromParameterCheck::check(
+    const MatchFinder::MatchResult &Result) {
+  const auto *R = Result.Nodes.getNodeAs<ReturnStmt>("ret");
+  diag(R->getRetValue()->getBeginLoc(),
+       "returning a constant reference parameter may cause a use-after-free "
+       "when the parameter is constructed from a temporary");
+}
+
+} // namespace clang::tidy::bugprone
diff --git a/clang-tools-extra/clang-tidy/bugprone/ReturnConstRefFromParameterCheck.h b/clang-tools-extra/clang-tidy/bugprone/ReturnConstRefFromParameterCheck.h
new file mode 100644
index 000000000000..8768d0708738
--- /dev/null
+++ b/clang-tools-extra/clang-tidy/bugprone/ReturnConstRefFromParameterCheck.h
@@ -0,0 +1,40 @@
+//===--- ReturnConstRefFromParameterCheck.h - clang-tidy --------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_RETURNCONSTREFFROMPARAMETERCHECK_H
+#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_RETURNCONSTREFFROMPARAMETERCHECK_H
+
+#include "../ClangTidyCheck.h"
+
+namespace clang::tidy::bugprone {
+
+/// Detects return statements that return a constant reference parameter as
+/// constant reference. This may cause use-after-free errors if the caller uses
+/// xvalues as arguments.
+///
+/// For the user-facing documentation see:
+/// http://clang.llvm.org/extra/clang-tidy/checks/bugprone/return-const-ref-from-parameter.html
+class ReturnConstRefFromParameterCheck : public ClangTidyCheck {
+public:
+  ReturnConstRefFromParameterCheck(StringRef Name, ClangTidyContext *Context)
+      : ClangTidyCheck(Name, Context) {}
+  void registerMatchers(ast_matchers::MatchFinder *Finder) override;
+  void check(const ast_matchers::MatchFinder::MatchResult &Result) override;
+  std::optional<TraversalKind> getCheckTraversalKind() const override {
+    // Use 'AsIs' to make sure the return type is exactly the same as the
+    // parameter type.
+    return TK_AsIs;
+  }
+  bool isLanguageVersionSupported(const LangOptions &LangOpts) const override {
+    return LangOpts.CPlusPlus;
+  }
+};
+
+} // namespace clang::tidy::bugprone
+
+#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_BUGPRONE_RETURNCONSTREFFROMPARAMETERCHECK_H
diff --git a/clang-tools-extra/clang-tidy/modernize/CMakeLists.txt b/clang-tools-extra/clang-tidy/modernize/CMakeLists.txt
index 6852db6c2ee3..8005d6e91c06 100644
--- a/clang-tools-extra/clang-tidy/modernize/CMakeLists.txt
+++ b/clang-tools-extra/clang-tidy/modernize/CMakeLists.txt
@@ -16,6 +16,7 @@ add_clang_library(clangTidyModernizeModule
   MakeSharedCheck.cpp
   MakeSmartPtrCheck.cpp
   MakeUniqueCheck.cpp
+  MinMaxUseInitializerListCheck.cpp
   ModernizeTidyModule.cpp
   PassByValueCheck.cpp
   RawStringLiteralCheck.cpp
diff --git a/clang-tools-extra/clang-tidy/modernize/MinMaxUseInitializerListCheck.cpp b/clang-tools-extra/clang-tidy/modernize/MinMaxUseInitializerListCheck.cpp
new file mode 100644
index 000000000000..45f7700463d5
--- /dev/null
+++ b/clang-tools-extra/clang-tidy/modernize/MinMaxUseInitializerListCheck.cpp
@@ -0,0 +1,271 @@
+//===--- MinMaxUseInitializerListCheck.cpp - clang-tidy -------------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#include "MinMaxUseInitializerListCheck.h"
+#include "../utils/ASTUtils.h"
+#include "../utils/LexerUtils.h"
+#include "clang/ASTMatchers/ASTMatchFinder.h"
+#include "clang/Frontend/CompilerInstance.h"
+#include "clang/Lex/Lexer.h"
+
+using namespace clang;
+
+namespace {
+
+struct FindArgsResult {
+  const Expr *First;
+  const Expr *Last;
+  const Expr *Compare;
+  SmallVector<const clang::Expr *, 2> Args;
+};
+
+} // anonymous namespace
+
+using namespace clang::ast_matchers;
+
+namespace clang::tidy::modernize {
+
+static FindArgsResult findArgs(const CallExpr *Call) {
+  FindArgsResult Result;
+  Result.First = nullptr;
+  Result.Last = nullptr;
+  Result.Compare = nullptr;
+
+  //   check if the function has initializer list argument
+  if (Call->getNumArgs() < 3) {
+    auto ArgIterator = Call->arguments().begin();
+
+    const auto *InitListExpr =
+        dyn_cast<CXXStdInitializerListExpr>(*ArgIterator);
+    const auto *InitList =
+        InitListExpr != nullptr
+            ? dyn_cast<clang::InitListExpr>(
+                  InitListExpr->getSubExpr()->IgnoreImplicit())
+            : nullptr;
+
+    if (InitList) {
+      Result.Args.append(InitList->inits().begin(), InitList->inits().end());
+      Result.First = *ArgIterator;
+      Result.Last = *ArgIterator;
+
+      // check if there is a comparison argument
+      std::advance(ArgIterator, 1);
+      if (ArgIterator != Call->arguments().end())
+        Result.Compare = *ArgIterator;
+
+      return Result;
+    }
+    Result.Args = SmallVector<const Expr *>(Call->arguments());
+  } else {
+    // if it has 3 arguments then the last will be the comparison
+    Result.Compare = *(std::next(Call->arguments().begin(), 2));
+    Result.Args = SmallVector<const Expr *>(llvm::drop_end(Call->arguments()));
+  }
+  Result.First = Result.Args.front();
+  Result.Last = Result.Args.back();
+
+  return Result;
+}
+
+static SmallVector<FixItHint>
+generateReplacements(const MatchFinder::MatchResult &Match,
+                     const CallExpr *TopCall, const FindArgsResult &Result,
+                     const bool IgnoreNonTrivialTypes,
+                     const std::uint64_t IgnoreTrivialTypesOfSizeAbove) {
+  SmallVector<FixItHint> FixItHints;
+  const SourceManager &SourceMngr = *Match.SourceManager;
+  const LangOptions &LanguageOpts = Match.Context->getLangOpts();
+
+  const QualType ResultType = TopCall->getDirectCallee()
+                                  ->getReturnType()
+                                  .getCanonicalType()
+                                  .getNonReferenceType()
+                                  .getUnqualifiedType();
+
+  // check if the type is trivial
+  const bool IsResultTypeTrivial = ResultType.isTrivialType(*Match.Context);
+
+  if ((!IsResultTypeTrivial && IgnoreNonTrivialTypes))
+    return FixItHints;
+
+  if (IsResultTypeTrivial &&
+      static_cast<std::uint64_t>(
+          Match.Context->getTypeSizeInChars(ResultType).getQuantity()) >
+          IgnoreTrivialTypesOfSizeAbove)
+    return FixItHints;
+
+  for (const Expr *Arg : Result.Args) {
+    const auto *InnerCall = dyn_cast<CallExpr>(Arg->IgnoreParenImpCasts());
+
+    // If the argument is not a nested call
+    if (!InnerCall) {
+      // check if typecast is required
+      const QualType ArgType = Arg->IgnoreParenImpCasts()
+                                   ->getType()
+                                   .getCanonicalType()
+                                   .getUnqualifiedType();
+
+      if (ArgType == ResultType)
+        continue;
+
+      const StringRef ArgText = Lexer::getSourceText(
+          CharSourceRange::getTokenRange(Arg->getSourceRange()), SourceMngr,
+          LanguageOpts);
+
+      const auto Replacement = Twine("static_cast<")
+                                   .concat(ResultType.getAsString(LanguageOpts))
+                                   .concat(">(")
+                                   .concat(ArgText)
+                                   .concat(")")
+                                   .str();
+
+      FixItHints.push_back(
+          FixItHint::CreateReplacement(Arg->getSourceRange(), Replacement));
+      continue;
+    }
+
+    const FindArgsResult InnerResult = findArgs(InnerCall);
+
+    // if the nested call doesn't have arguments skip it
+    if (!InnerResult.First || !InnerResult.Last)
+      continue;
+
+    // if the nested call is not the same as the top call
+    if (InnerCall->getDirectCallee()->getQualifiedNameAsString() !=
+        TopCall->getDirectCallee()->getQualifiedNameAsString())
+      continue;
+
+    // if the nested call doesn't have the same compare function
+    if ((Result.Compare || InnerResult.Compare) &&
+        !utils::areStatementsIdentical(Result.Compare, InnerResult.Compare,
+                                       *Match.Context))
+      continue;
+
+    // remove the function call
+    FixItHints.push_back(
+        FixItHint::CreateRemoval(InnerCall->getCallee()->getSourceRange()));
+
+    // remove the parentheses
+    const auto LParen = utils::lexer::findNextTokenSkippingComments(
+        InnerCall->getCallee()->getEndLoc(), SourceMngr, LanguageOpts);
+    if (LParen.has_value() && LParen->is(tok::l_paren))
+      FixItHints.push_back(
+          FixItHint::CreateRemoval(SourceRange(LParen->getLocation())));
+    FixItHints.push_back(
+        FixItHint::CreateRemoval(SourceRange(InnerCall->getRParenLoc())));
+
+    // if the inner call has an initializer list arg
+    if (InnerResult.First == InnerResult.Last) {
+      // remove the initializer list braces
+      FixItHints.push_back(FixItHint::CreateRemoval(
+          CharSourceRange::getTokenRange(InnerResult.First->getBeginLoc())));
+      FixItHints.push_back(FixItHint::CreateRemoval(
+          CharSourceRange::getTokenRange(InnerResult.First->getEndLoc())));
+    }
+
+    const SmallVector<FixItHint> InnerReplacements = generateReplacements(
+        Match, InnerCall, InnerResult, IgnoreNonTrivialTypes,
+        IgnoreTrivialTypesOfSizeAbove);
+
+    FixItHints.append(InnerReplacements);
+
+    if (InnerResult.Compare) {
+      // find the comma after the value arguments
+      const auto Comma = utils::lexer::findNextTokenSkippingComments(
+          InnerResult.Last->getEndLoc(), SourceMngr, LanguageOpts);
+
+      // remove the comma and the comparison
+      if (Comma.has_value() && Comma->is(tok::comma))
+        FixItHints.push_back(
+            FixItHint::CreateRemoval(SourceRange(Comma->getLocation())));
+
+      FixItHints.push_back(
+          FixItHint::CreateRemoval(InnerResult.Compare->getSourceRange()));
+    }
+  }
+
+  return FixItHints;
+}
+
+MinMaxUseInitializerListCheck::MinMaxUseInitializerListCheck(
+    StringRef Name, ClangTidyContext *Context)
+    : ClangTidyCheck(Name, Context),
+      IgnoreNonTrivialTypes(Options.get("IgnoreNonTrivialTypes", true)),
+      IgnoreTrivialTypesOfSizeAbove(
+          Options.get("IgnoreTrivialTypesOfSizeAbove", 32L)),
+      Inserter(Options.getLocalOrGlobal("IncludeStyle",
+                                        utils::IncludeSorter::IS_LLVM),
+               areDiagsSelfContained()) {}
+
+void MinMaxUseInitializerListCheck::storeOptions(
+    ClangTidyOptions::OptionMap &Opts) {
+  Options.store(Opts, "IgnoreNonTrivialTypes", IgnoreNonTrivialTypes);
+  Options.store(Opts, "IgnoreTrivialTypesOfSizeAbove",
+                IgnoreTrivialTypesOfSizeAbove);
+  Options.store(Opts, "IncludeStyle", Inserter.getStyle());
+}
+
+void MinMaxUseInitializerListCheck::registerMatchers(MatchFinder *Finder) {
+  auto CreateMatcher = [](const StringRef FunctionName) {
+    auto FuncDecl = functionDecl(hasName(FunctionName));
+    auto Expression = callExpr(callee(FuncDecl));
+
+    return callExpr(callee(FuncDecl),
+                    anyOf(hasArgument(0, Expression),
+                          hasArgument(1, Expression),
+                          hasArgument(0, cxxStdInitializerListExpr())),
+                    unless(hasParent(Expression)))
+        .bind("topCall");
+  };
+
+  Finder->addMatcher(CreateMatcher("::std::max"), this);
+  Finder->addMatcher(CreateMatcher("::std::min"), this);
+}
+
+void MinMaxUseInitializerListCheck::registerPPCallbacks(
+    const SourceManager &SM, Preprocessor *PP, Preprocessor *ModuleExpanderPP) {
+  Inserter.registerPreprocessor(PP);
+}
+
+void MinMaxUseInitializerListCheck::check(
+    const MatchFinder::MatchResult &Match) {
+
+  const auto *TopCall = Match.Nodes.getNodeAs<CallExpr>("topCall");
+
+  const FindArgsResult Result = findArgs(TopCall);
+  const SmallVector<FixItHint> Replacements =
+      generateReplacements(Match, TopCall, Result, IgnoreNonTrivialTypes,
+                           IgnoreTrivialTypesOfSizeAbove);
+
+  if (Replacements.empty())
+    return;
+
+  const DiagnosticBuilder Diagnostic =
+      diag(TopCall->getBeginLoc(),
+           "do not use nested 'std::%0' calls, use an initializer list instead")
+      << TopCall->getDirectCallee()->getName()
+      << Inserter.createIncludeInsertion(
+             Match.SourceManager->getFileID(TopCall->getBeginLoc()),
+             "<algorithm>");
+
+  // if the top call doesn't have an initializer list argument
+  if (Result.First != Result.Last) {
+    // add { and } insertions
+    Diagnostic << FixItHint::CreateInsertion(Result.First->getBeginLoc(), "{");
+
+    Diagnostic << FixItHint::CreateInsertion(
+        Lexer::getLocForEndOfToken(Result.Last->getEndLoc(), 0,
+                                   *Match.SourceManager,
+                                   Match.Context->getLangOpts()),
+        "}");
+  }
+
+  Diagnostic << Replacements;
+}
+
+} // namespace clang::tidy::modernize
diff --git a/clang-tools-extra/clang-tidy/modernize/MinMaxUseInitializerListCheck.h b/clang-tools-extra/clang-tidy/modernize/MinMaxUseInitializerListCheck.h
new file mode 100644
index 000000000000..577d12653076
--- /dev/null
+++ b/clang-tools-extra/clang-tidy/modernize/MinMaxUseInitializerListCheck.h
@@ -0,0 +1,56 @@
+//===--- MinMaxUseInitializerListCheck.h - clang-tidy -----------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_MINMAXUSEINITIALIZERLISTCHECK_H
+#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_MINMAXUSEINITIALIZERLISTCHECK_H
+
+#include "../ClangTidyCheck.h"
+#include "../utils/IncludeInserter.h"
+
+namespace clang::tidy::modernize {
+
+/// Replaces nested ``std::min`` and ``std::max`` calls with an initializer list
+/// where applicable.
+///
+/// For example:
+///
+/// \code
+///   int a = std::max(std::max(i, j), k);
+/// \endcode
+///
+/// This code is transformed to:
+///
+/// \code
+///   int a = std::max({i, j, k});
+/// \endcode
+class MinMaxUseInitializerListCheck : public ClangTidyCheck {
+public:
+  MinMaxUseInitializerListCheck(StringRef Name, ClangTidyContext *Context);
+
+  void storeOptions(ClangTidyOptions::OptionMap &Opts) override;
+  void registerMatchers(ast_matchers::MatchFinder *Finder) override;
+  void registerPPCallbacks(const SourceManager &SM, Preprocessor *PP,
+                           Preprocessor *ModuleExpanderPP) override;
+  void check(const ast_matchers::MatchFinder::MatchResult &Match) override;
+
+  bool isLanguageVersionSupported(const LangOptions &LangOpts) const override {
+    return LangOpts.CPlusPlus11;
+  }
+  std::optional<TraversalKind> getCheckTraversalKind() const override {
+    return TK_IgnoreUnlessSpelledInSource;
+  }
+
+private:
+  bool IgnoreNonTrivialTypes;
+  std::uint64_t IgnoreTrivialTypesOfSizeAbove;
+  utils::IncludeInserter Inserter;
+};
+
+} // namespace clang::tidy::modernize
+
+#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_MODERNIZE_MINMAXUSEINITIALIZERLISTCHECK_H
diff --git a/clang-tools-extra/clang-tidy/modernize/ModernizeTidyModule.cpp b/clang-tools-extra/clang-tidy/modernize/ModernizeTidyModule.cpp
index e96cf274f58c..776558433c5b 100644
--- a/clang-tools-extra/clang-tidy/modernize/ModernizeTidyModule.cpp
+++ b/clang-tools-extra/clang-tidy/modernize/ModernizeTidyModule.cpp
@@ -18,6 +18,7 @@
 #include "MacroToEnumCheck.h"
 #include "MakeSharedCheck.h"
 #include "MakeUniqueCheck.h"
+#include "MinMaxUseInitializerListCheck.h"
 #include "PassByValueCheck.h"
 #include "RawStringLiteralCheck.h"
 #include "RedundantVoidArgCheck.h"
@@ -68,6 +69,8 @@ public:
     CheckFactories.registerCheck<MacroToEnumCheck>("modernize-macro-to-enum");
     CheckFactories.registerCheck<MakeSharedCheck>("modernize-make-shared");
     CheckFactories.registerCheck<MakeUniqueCheck>("modernize-make-unique");
+    CheckFactories.registerCheck<MinMaxUseInitializerListCheck>(
+        "modernize-min-max-use-initializer-list");
     CheckFactories.registerCheck<PassByValueCheck>("modernize-pass-by-value");
     CheckFactories.registerCheck<UseDesignatedInitializersCheck>(
         "modernize-use-designated-initializers");
diff --git a/clang-tools-extra/clang-tidy/modernize/UseNullptrCheck.h b/clang-tools-extra/clang-tidy/modernize/UseNullptrCheck.h
index 6c32a4edb4ff..f1591bae4465 100644
--- a/clang-tools-extra/clang-tidy/modernize/UseNullptrCheck.h
+++ b/clang-tools-extra/clang-tidy/modernize/UseNullptrCheck.h
@@ -19,7 +19,7 @@ public:
   bool isLanguageVersionSupported(const LangOptions &LangOpts) const override {
     // FIXME this should be CPlusPlus11 but that causes test cases to
     // erroneously fail.
-    return LangOpts.CPlusPlus;
+    return LangOpts.CPlusPlus || LangOpts.C23;
   }
   void storeOptions(ClangTidyOptions::OptionMap &Opts) override;
   void registerMatchers(ast_matchers::MatchFinder *Finder) override;
diff --git a/clang-tools-extra/clang-tidy/modernize/UseStartsEndsWithCheck.cpp b/clang-tools-extra/clang-tidy/modernize/UseStartsEndsWithCheck.cpp
index 062f6e9911db..89ee45faecd7 100644
--- a/clang-tools-extra/clang-tidy/modernize/UseStartsEndsWithCheck.cpp
+++ b/clang-tools-extra/clang-tidy/modernize/UseStartsEndsWithCheck.cpp
@@ -43,7 +43,9 @@ void UseStartsEndsWithCheck::registerMatchers(MatchFinder *Finder) {
       callee(cxxMethodDecl(hasName("find")).bind("find_fun")),
       // ... on a class with a starts_with function.
       on(hasType(
-          hasCanonicalType(hasDeclaration(ClassWithStartsWithFunction)))));
+          hasCanonicalType(hasDeclaration(ClassWithStartsWithFunction)))),
+      // Bind search expression.
+      hasArgument(0, expr().bind("search_expr")));
 
   const auto RFindExpr = cxxMemberCallExpr(
       // A method call with a second argument of zero...
@@ -52,15 +54,68 @@ void UseStartsEndsWithCheck::registerMatchers(MatchFinder *Finder) {
       callee(cxxMethodDecl(hasName("rfind")).bind("find_fun")),
       // ... on a class with a starts_with function.
       on(hasType(
-          hasCanonicalType(hasDeclaration(ClassWithStartsWithFunction)))));
+          hasCanonicalType(hasDeclaration(ClassWithStartsWithFunction)))),
+      // Bind search expression.
+      hasArgument(0, expr().bind("search_expr")));
+
+  // Match a string literal and an integer or strlen() call matching the length.
+  const auto HasStringLiteralAndLengthArgs = [](const auto StringArgIndex,
+                                                const auto LengthArgIndex) {
+    return allOf(
+        hasArgument(StringArgIndex, stringLiteral().bind("string_literal_arg")),
+        hasArgument(LengthArgIndex,
+                    anyOf(integerLiteral().bind("integer_literal_size_arg"),
+                          callExpr(callee(functionDecl(parameterCountIs(1),
+                                                       hasName("strlen"))),
+                                   hasArgument(0, stringLiteral().bind(
+                                                      "strlen_arg"))))));
+  };
+
+  // Match a string variable and a call to length() or size().
+  const auto HasStringVariableAndSizeCallArgs = [](const auto StringArgIndex,
+                                                   const auto LengthArgIndex) {
+    return allOf(
+        hasArgument(StringArgIndex, declRefExpr(hasDeclaration(
+                                        decl().bind("string_var_decl")))),
+        hasArgument(LengthArgIndex,
+                    cxxMemberCallExpr(
+                        callee(cxxMethodDecl(isConst(), parameterCountIs(0),
+                                             hasAnyName("size", "length"))),
+                        on(declRefExpr(
+                            to(decl(equalsBoundNode("string_var_decl"))))))));
+  };
 
-  const auto FindOrRFindExpr =
-      cxxMemberCallExpr(anyOf(FindExpr, RFindExpr)).bind("find_expr");
+  // Match either one of the two cases above.
+  const auto HasStringAndLengthArgs =
+      [HasStringLiteralAndLengthArgs, HasStringVariableAndSizeCallArgs](
+          const auto StringArgIndex, const auto LengthArgIndex) {
+        return anyOf(
+            HasStringLiteralAndLengthArgs(StringArgIndex, LengthArgIndex),
+            HasStringVariableAndSizeCallArgs(StringArgIndex, LengthArgIndex));
+      };
+
+  const auto CompareExpr = cxxMemberCallExpr(
+      // A method call with three arguments...
+      argumentCountIs(3),
+      // ... where the first argument is zero...
+      hasArgument(0, ZeroLiteral),
+      // ... named compare...
+      callee(cxxMethodDecl(hasName("compare")).bind("find_fun")),
+      // ... on a class with a starts_with function...
+      on(hasType(
+          hasCanonicalType(hasDeclaration(ClassWithStartsWithFunction)))),
+      // ... where the third argument is some string and the second a length.
+      HasStringAndLengthArgs(2, 1),
+      // Bind search expression.
+      hasArgument(2, expr().bind("search_expr")));
 
   Finder->addMatcher(
-      // Match [=!]= with a zero on one side and a string.(r?)find on the other.
-      binaryOperator(hasAnyOperatorName("==", "!="),
-                     hasOperands(FindOrRFindExpr, ZeroLiteral))
+      // Match [=!]= with a zero on one side and (r?)find|compare on the other.
+      binaryOperator(
+          hasAnyOperatorName("==", "!="),
+          hasOperands(cxxMemberCallExpr(anyOf(FindExpr, RFindExpr, CompareExpr))
+                          .bind("find_expr"),
+                      ZeroLiteral))
           .bind("expr"),
       this);
 }
@@ -69,9 +124,28 @@ void UseStartsEndsWithCheck::check(const MatchFinder::MatchResult &Result) {
   const auto *ComparisonExpr = Result.Nodes.getNodeAs<BinaryOperator>("expr");
   const auto *FindExpr = Result.Nodes.getNodeAs<CXXMemberCallExpr>("find_expr");
   const auto *FindFun = Result.Nodes.getNodeAs<CXXMethodDecl>("find_fun");
+  const auto *SearchExpr = Result.Nodes.getNodeAs<Expr>("search_expr");
   const auto *StartsWithFunction =
       Result.Nodes.getNodeAs<CXXMethodDecl>("starts_with_fun");
 
+  const auto *StringLiteralArg =
+      Result.Nodes.getNodeAs<StringLiteral>("string_literal_arg");
+  const auto *IntegerLiteralSizeArg =
+      Result.Nodes.getNodeAs<IntegerLiteral>("integer_literal_size_arg");
+  const auto *StrlenArg = Result.Nodes.getNodeAs<StringLiteral>("strlen_arg");
+
+  // Filter out compare cases where the length does not match string literal.
+  if (StringLiteralArg && IntegerLiteralSizeArg &&
+      StringLiteralArg->getLength() !=
+          IntegerLiteralSizeArg->getValue().getZExtValue()) {
+    return;
+  }
+
+  if (StringLiteralArg && StrlenArg &&
+      StringLiteralArg->getLength() != StrlenArg->getLength()) {
+    return;
+  }
+
   if (ComparisonExpr->getBeginLoc().isMacroID()) {
     return;
   }
@@ -79,13 +153,13 @@ void UseStartsEndsWithCheck::check(const MatchFinder::MatchResult &Result) {
   const bool Neg = ComparisonExpr->getOpcode() == BO_NE;
 
   auto Diagnostic =
-      diag(FindExpr->getBeginLoc(), "use %0 instead of %1() %select{==|!=}2 0")
+      diag(FindExpr->getExprLoc(), "use %0 instead of %1() %select{==|!=}2 0")
       << StartsWithFunction->getName() << FindFun->getName() << Neg;
 
-  // Remove possible zero second argument and ' [!=]= 0' suffix.
+  // Remove possible arguments after search expression and ' [!=]= 0' suffix.
   Diagnostic << FixItHint::CreateReplacement(
       CharSourceRange::getTokenRange(
-          Lexer::getLocForEndOfToken(FindExpr->getArg(0)->getEndLoc(), 0,
+          Lexer::getLocForEndOfToken(SearchExpr->getEndLoc(), 0,
                                      *Result.SourceManager, getLangOpts()),
           ComparisonExpr->getEndLoc()),
       ")");
@@ -94,11 +168,12 @@ void UseStartsEndsWithCheck::check(const MatchFinder::MatchResult &Result) {
   Diagnostic << FixItHint::CreateRemoval(CharSourceRange::getCharRange(
       ComparisonExpr->getBeginLoc(), FindExpr->getBeginLoc()));
 
-  // Replace '(r?)find' with 'starts_with'.
+  // Replace method name by 'starts_with'.
+  // Remove possible arguments before search expression.
   Diagnostic << FixItHint::CreateReplacement(
-      CharSourceRange::getTokenRange(FindExpr->getExprLoc(),
-                                     FindExpr->getExprLoc()),
-      StartsWithFunction->getName());
+      CharSourceRange::getCharRange(FindExpr->getExprLoc(),
+                                    SearchExpr->getBeginLoc()),
+      (StartsWithFunction->getName() + "(").str());
 
   // Add possible negation '!'.
   if (Neg) {
diff --git a/clang-tools-extra/clang-tidy/modernize/UseStartsEndsWithCheck.h b/clang-tools-extra/clang-tidy/modernize/UseStartsEndsWithCheck.h
index 34e971776825..840191f32149 100644
--- a/clang-tools-extra/clang-tidy/modernize/UseStartsEndsWithCheck.h
+++ b/clang-tools-extra/clang-tidy/modernize/UseStartsEndsWithCheck.h
@@ -13,9 +13,10 @@
 
 namespace clang::tidy::modernize {
 
-/// Checks whether a ``find`` or ``rfind`` result is compared with 0 and
-/// suggests replacing with ``starts_with`` when the method exists in the class.
-/// Notably, this will work with ``std::string`` and ``std::string_view``.
+/// Checks for common roundabout ways to express ``starts_with`` and
+/// ``ends_with`` and suggests replacing with ``starts_with`` when the method is
+/// available. Notably, this will work with ``std::string`` and
+/// ``std::string_view``.
 ///
 /// For the user-facing documentation see:
 /// http://clang.llvm.org/extra/clang-tidy/checks/modernize/use-starts-ends-with.html
diff --git a/clang-tools-extra/clang-tidy/readability/AvoidReturnWithVoidValueCheck.cpp b/clang-tools-extra/clang-tidy/readability/AvoidReturnWithVoidValueCheck.cpp
index 48bca41f4a3b..f077040a3529 100644
--- a/clang-tools-extra/clang-tidy/readability/AvoidReturnWithVoidValueCheck.cpp
+++ b/clang-tools-extra/clang-tidy/readability/AvoidReturnWithVoidValueCheck.cpp
@@ -64,8 +64,11 @@ void AvoidReturnWithVoidValueCheck::check(
            << BraceInsertionHints.closingBraceFixIt();
   }
   Diag << FixItHint::CreateRemoval(VoidReturn->getReturnLoc());
-  if (!Result.Nodes.getNodeAs<FunctionDecl>("function_parent") ||
-      SurroundingBlock->body_back() != VoidReturn)
+  const auto *FunctionParent =
+      Result.Nodes.getNodeAs<FunctionDecl>("function_parent");
+  if (!FunctionParent ||
+      (SurroundingBlock && SurroundingBlock->body_back() != VoidReturn))
+    // If this is not the last statement in a function body, we add a `return`.
     Diag << FixItHint::CreateInsertion(SemicolonPos.getLocWithOffset(1),
                                        " return;", true);
 }
diff --git a/clang-tools-extra/clang-tidy/readability/CMakeLists.txt b/clang-tools-extra/clang-tidy/readability/CMakeLists.txt
index dd772d692025..41065fc8e878 100644
--- a/clang-tools-extra/clang-tidy/readability/CMakeLists.txt
+++ b/clang-tools-extra/clang-tidy/readability/CMakeLists.txt
@@ -28,6 +28,7 @@ add_clang_library(clangTidyReadabilityModule
   IsolateDeclarationCheck.cpp
   MagicNumbersCheck.cpp
   MakeMemberFunctionConstCheck.cpp
+  MathMissingParenthesesCheck.cpp
   MisleadingIndentationCheck.cpp
   MisplacedArrayIndexCheck.cpp
   NamedParameterCheck.cpp
diff --git a/clang-tools-extra/clang-tidy/readability/MathMissingParenthesesCheck.cpp b/clang-tools-extra/clang-tidy/readability/MathMissingParenthesesCheck.cpp
new file mode 100644
index 000000000000..d1e20b9074ce
--- /dev/null
+++ b/clang-tools-extra/clang-tidy/readability/MathMissingParenthesesCheck.cpp
@@ -0,0 +1,97 @@
+//===--- MathMissingParenthesesCheck.cpp - clang-tidy ---------------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#include "MathMissingParenthesesCheck.h"
+#include "clang/AST/ASTContext.h"
+#include "clang/ASTMatchers/ASTMatchFinder.h"
+#include "clang/Lex/Lexer.h"
+
+using namespace clang::ast_matchers;
+
+namespace clang::tidy::readability {
+
+void MathMissingParenthesesCheck::registerMatchers(MatchFinder *Finder) {
+  Finder->addMatcher(binaryOperator(unless(hasParent(binaryOperator())),
+                                    unless(isAssignmentOperator()),
+                                    unless(isComparisonOperator()),
+                                    unless(hasAnyOperatorName("&&", "||")),
+                                    hasDescendant(binaryOperator()))
+                         .bind("binOp"),
+                     this);
+}
+
+static int getPrecedence(const BinaryOperator *BinOp) {
+  if (!BinOp)
+    return 0;
+  switch (BinOp->getOpcode()) {
+  case BO_Mul:
+  case BO_Div:
+  case BO_Rem:
+    return 5;
+  case BO_Add:
+  case BO_Sub:
+    return 4;
+  case BO_And:
+    return 3;
+  case BO_Xor:
+    return 2;
+  case BO_Or:
+    return 1;
+  default:
+    return 0;
+  }
+}
+static void addParantheses(const BinaryOperator *BinOp,
+                           const BinaryOperator *ParentBinOp,
+                           ClangTidyCheck *Check,
+                           const clang::SourceManager &SM,
+                           const clang::LangOptions &LangOpts) {
+  if (!BinOp)
+    return;
+
+  int Precedence1 = getPrecedence(BinOp);
+  int Precedence2 = getPrecedence(ParentBinOp);
+
+  if (ParentBinOp != nullptr && Precedence1 != Precedence2) {
+    const clang::SourceLocation StartLoc = BinOp->getBeginLoc();
+    const clang::SourceLocation EndLoc =
+        clang::Lexer::getLocForEndOfToken(BinOp->getEndLoc(), 0, SM, LangOpts);
+    if (EndLoc.isInvalid())
+      return;
+
+    Check->diag(StartLoc,
+                "'%0' has higher precedence than '%1'; add parentheses to "
+                "explicitly specify the order of operations")
+        << (Precedence1 > Precedence2 ? BinOp->getOpcodeStr()
+                                      : ParentBinOp->getOpcodeStr())
+        << (Precedence1 > Precedence2 ? ParentBinOp->getOpcodeStr()
+                                      : BinOp->getOpcodeStr())
+        << FixItHint::CreateInsertion(StartLoc, "(")
+        << FixItHint::CreateInsertion(EndLoc, ")")
+        << SourceRange(StartLoc, EndLoc);
+  }
+
+  addParantheses(dyn_cast<BinaryOperator>(BinOp->getLHS()->IgnoreImpCasts()),
+                 BinOp, Check, SM, LangOpts);
+  addParantheses(dyn_cast<BinaryOperator>(BinOp->getRHS()->IgnoreImpCasts()),
+                 BinOp, Check, SM, LangOpts);
+}
+
+void MathMissingParenthesesCheck::check(
+    const MatchFinder::MatchResult &Result) {
+  const auto *BinOp = Result.Nodes.getNodeAs<BinaryOperator>("binOp");
+  std::vector<
+      std::pair<clang::SourceRange, std::pair<const clang::BinaryOperator *,
+                                              const clang::BinaryOperator *>>>
+      Insertions;
+  const SourceManager &SM = *Result.SourceManager;
+  const clang::LangOptions &LO = Result.Context->getLangOpts();
+  addParantheses(BinOp, nullptr, this, SM, LO);
+}
+
+} // namespace clang::tidy::readability
diff --git a/clang-tools-extra/clang-tidy/readability/MathMissingParenthesesCheck.h b/clang-tools-extra/clang-tidy/readability/MathMissingParenthesesCheck.h
new file mode 100644
index 000000000000..9a9d2b3cfaab
--- /dev/null
+++ b/clang-tools-extra/clang-tidy/readability/MathMissingParenthesesCheck.h
@@ -0,0 +1,34 @@
+//===--- MathMissingParenthesesCheck.h - clang-tidy -------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_MATHMISSINGPARENTHESESCHECK_H
+#define LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_MATHMISSINGPARENTHESESCHECK_H
+
+#include "../ClangTidyCheck.h"
+
+namespace clang::tidy::readability {
+
+/// Check for mising parantheses in mathematical expressions that involve
+/// operators of different priorities.
+///
+/// For the user-facing documentation see:
+/// http://clang.llvm.org/extra/clang-tidy/checks/readability/math-missing-parentheses.html
+class MathMissingParenthesesCheck : public ClangTidyCheck {
+public:
+  MathMissingParenthesesCheck(StringRef Name, ClangTidyContext *Context)
+      : ClangTidyCheck(Name, Context) {}
+  void registerMatchers(ast_matchers::MatchFinder *Finder) override;
+  void check(const ast_matchers::MatchFinder::MatchResult &Result) override;
+  std::optional<TraversalKind> getCheckTraversalKind() const override {
+    return TK_IgnoreUnlessSpelledInSource;
+  }
+};
+
+} // namespace clang::tidy::readability
+
+#endif // LLVM_CLANG_TOOLS_EXTRA_CLANG_TIDY_READABILITY_MATHMISSINGPARENTHESESCHECK_H
diff --git a/clang-tools-extra/clang-tidy/readability/ReadabilityTidyModule.cpp b/clang-tools-extra/clang-tidy/readability/ReadabilityTidyModule.cpp
index 376b84683df7..d61c0ba39658 100644
--- a/clang-tools-extra/clang-tidy/readability/ReadabilityTidyModule.cpp
+++ b/clang-tools-extra/clang-tidy/readability/ReadabilityTidyModule.cpp
@@ -32,6 +32,7 @@
 #include "IsolateDeclarationCheck.h"
 #include "MagicNumbersCheck.h"
 #include "MakeMemberFunctionConstCheck.h"
+#include "MathMissingParenthesesCheck.h"
 #include "MisleadingIndentationCheck.h"
 #include "MisplacedArrayIndexCheck.h"
 #include "NamedParameterCheck.h"
@@ -105,6 +106,8 @@ public:
         "readability-identifier-naming");
     CheckFactories.registerCheck<ImplicitBoolConversionCheck>(
         "readability-implicit-bool-conversion");
+    CheckFactories.registerCheck<MathMissingParenthesesCheck>(
+        "readability-math-missing-parentheses");
     CheckFactories.registerCheck<RedundantInlineSpecifierCheck>(
         "readability-redundant-inline-specifier");
     CheckFactories.registerCheck<InconsistentDeclarationParameterNameCheck>(
diff --git a/clang-tools-extra/clangd/CodeCompletionStrings.cpp b/clang-tools-extra/clangd/CodeCompletionStrings.cpp
index 2075e5965f18..9b4442b0bb76 100644
--- a/clang-tools-extra/clangd/CodeCompletionStrings.cpp
+++ b/clang-tools-extra/clangd/CodeCompletionStrings.cpp
@@ -253,7 +253,7 @@ void getSignature(const CodeCompletionString &CCS, std::string *Signature,
       if (!IncludeFunctionArguments &&
           ResultKind == CodeCompletionResult::RK_Declaration)
         TruncateSnippetAt.emplace(Snippet->size());
-      LLVM_FALLTHROUGH;
+      [[fallthrough]];
     case CodeCompletionString::CK_RightParen:
     case CodeCompletionString::CK_LeftBracket:
     case CodeCompletionString::CK_RightBracket:
diff --git a/clang-tools-extra/docs/ReleaseNotes.rst b/clang-tools-extra/docs/ReleaseNotes.rst
index f3f9a81f9a8e..3038d2b125f2 100644
--- a/clang-tools-extra/docs/ReleaseNotes.rst
+++ b/clang-tools-extra/docs/ReleaseNotes.rst
@@ -100,11 +100,15 @@ Improvements to clang-tidy
 - Improved :program:`run-clang-tidy.py` script. Added argument `-source-filter`
   to filter source files from the compilation database, via a RegEx. In a
   similar fashion to what `-header-filter` does for header files.
+
 - Improved :program:`check_clang_tidy.py` script. Added argument `-export-fixes`
   to aid in clang-tidy and test development.
 
-- Fixed ``--verify-config`` option not properly parsing checks when using the 
-  literal operator in the ``.clang-tidy`` config.
+- Fixed bug where big values for unsigned check options overflowed into negative values
+  when being printed with `--dump-config`.
+
+- Fixed `--verify-config` option not properly parsing checks when using the
+  literal operator in the `.clang-tidy` config.
 
 New checks
 ^^^^^^^^^^
@@ -115,6 +119,13 @@ New checks
   Detects error-prone Curiously Recurring Template Pattern usage, when the CRTP
   can be constructed outside itself and the derived class.
 
+- New :doc:`bugprone-return-const-ref-from-parameter
+  <clang-tidy/checks/bugprone/return-const-ref-from-parameter>` check.
+
+  Detects return statements that return a constant reference parameter as constant
+  reference. This may cause use-after-free errors if the caller uses xvalues as
+  arguments.
+
 - New :doc:`bugprone-suspicious-stringview-data-usage
   <clang-tidy/checks/bugprone/suspicious-stringview-data-usage>` check.
 
@@ -122,6 +133,12 @@ New checks
   to reading out-of-bounds data due to inadequate or incorrect string null
   termination.
 
+- New :doc:`modernize-min-max-use-initializer-list
+  <clang-tidy/checks/modernize/min-max-use-initializer-list>` check.
+
+  Replaces nested ``std::min`` and ``std::max`` calls with an initializer list
+  where applicable.
+
 - New :doc:`modernize-use-designated-initializers
   <clang-tidy/checks/modernize/use-designated-initializers>` check.
 
@@ -134,6 +151,12 @@ New checks
   Enforces consistent style for enumerators' initialization, covering three
   styles: none, first only, or all initialized explicitly.
 
+- New :doc:`readability-math-missing-parentheses
+  <clang-tidy/checks/readability/math-missing-parentheses>` check.
+
+  Check for missing parentheses in mathematical expressions that involve
+  operators of different priorities.
+
 - New :doc:`readability-use-std-min-max
   <clang-tidy/checks/readability/use-std-min-max>` check.
 
@@ -215,7 +238,7 @@ Changes in existing checks
 
 - Improved :doc:`google-explicit-constructor
   <clang-tidy/checks/google/explicit-constructor>` check to better handle
-  ``C++-20`` `explicit(bool)`.
+  C++20 `explicit(bool)`.
 
 - Improved :doc:`google-global-names-in-headers
   <clang-tidy/checks/google/global-names-in-headers>` check by replacing the local
@@ -228,6 +251,10 @@ Changes in existing checks
   check by ignoring other functions with same prefixes as the target specific
   functions.
 
+- Improved :doc:`linuxkernel-must-check-errs
+  <clang-tidy/checks/linuxkernel/must-check-errs>` check documentation to
+  consistently use the check's proper name.
+
 - Improved :doc:`llvm-header-guard
   <clang-tidy/checks/llvm/header-guard>` check by replacing the local
   option `HeaderFileExtensions` by the global option of the same name.
@@ -260,10 +287,18 @@ Changes in existing checks
   don't remove parentheses used in ``sizeof`` calls when they have array index
   accesses as arguments.
 
+- Improved :doc:`modernize-use-nullptr
+  <clang-tidy/checks/modernize/use-nullptr>` check to include support for C23,
+  which also has introduced the ``nullptr`` keyword.
+
 - Improved :doc:`modernize-use-override
   <clang-tidy/checks/modernize/use-override>` check to also remove any trailing
   whitespace when deleting the ``virtual`` keyword.
 
+- Improved :doc:`modernize-use-starts-ends-with
+  <clang-tidy/checks/modernize/use-starts-ends-with>` check to also handle
+  calls to ``compare`` method.
+
 - Improved :doc:`modernize-use-using <clang-tidy/checks/modernize/use-using>`
   check by adding support for detection of typedefs declared on function level.
 
@@ -311,13 +346,9 @@ Miscellaneous
 ^^^^^^^^^^^^^
 
 - Fixed incorrect formatting in :program:`clang-apply-replacements` when no
-  ``--format`` option is specified. Now :program:`clang-apply-replacements`
+  `--format` option is specified. Now :program:`clang-apply-replacements`
   applies formatting only with the option.
 
-- Fixed the :doc:`linuxkernel-must-check-errs
-  <clang-tidy/checks/linuxkernel/must-check-errs>` documentation to consistently
-  use the check's proper name.
-
 Improvements to include-fixer
 -----------------------------
 
diff --git a/clang-tools-extra/docs/clang-tidy/checks/bugprone/return-const-ref-from-parameter.rst b/clang-tools-extra/docs/clang-tidy/checks/bugprone/return-const-ref-from-parameter.rst
new file mode 100644
index 000000000000..f007dfe54999
--- /dev/null
+++ b/clang-tools-extra/docs/clang-tidy/checks/bugprone/return-const-ref-from-parameter.rst
@@ -0,0 +1,31 @@
+.. title:: clang-tidy - bugprone-return-const-ref-from-parameter
+
+bugprone-return-const-ref-from-parameter
+========================================
+
+Detects return statements that return a constant reference parameter as constant
+reference. This may cause use-after-free errors if the caller uses xvalues as
+arguments.
+
+In C++, constant reference parameters can accept xvalues which will be destructed
+after the call. When the function returns such a parameter also as constant reference,
+then the returned reference can be used after the object it refers to has been
+destroyed.
+
+Example
+-------
+
+.. code-block:: c++
+
+  struct S {
+    int v;
+    S(int);
+    ~S();
+  };
+  
+  const S &fn(const S &a) {
+    return a;
+  }
+
+  const S& s = fn(S{1});
+  s.v; // use after free
diff --git a/clang-tools-extra/docs/clang-tidy/checks/cert/env33-c.rst b/clang-tools-extra/docs/clang-tidy/checks/cert/env33-c.rst
index c5321b07f7f8..9271c9ecccc0 100644
--- a/clang-tools-extra/docs/clang-tidy/checks/cert/env33-c.rst
+++ b/clang-tools-extra/docs/clang-tidy/checks/cert/env33-c.rst
@@ -10,4 +10,4 @@ but does not actually attempt to execute a command.
 
 This check corresponds to the CERT C Coding Standard rule
 `ENV33-C. Do not call system()
-<https://www.securecoding.cert.org/confluence/pages/viewpage.action?pageId=2130132>`_.
+<https://www.securecoding.cert.org/confluence/display/c/ENV33-C.+Do+not+call+system()>`_.
diff --git a/clang-tools-extra/docs/clang-tidy/checks/list.rst b/clang-tools-extra/docs/clang-tidy/checks/list.rst
index 3a06d7c30c9b..49747ff896ba 100644
--- a/clang-tools-extra/docs/clang-tidy/checks/list.rst
+++ b/clang-tools-extra/docs/clang-tidy/checks/list.rst
@@ -120,6 +120,7 @@ Clang-Tidy Checks
    :doc:`bugprone-posix-return <bugprone/posix-return>`, "Yes"
    :doc:`bugprone-redundant-branch-condition <bugprone/redundant-branch-condition>`, "Yes"
    :doc:`bugprone-reserved-identifier <bugprone/reserved-identifier>`, "Yes"
+   :doc:`bugprone-return-const-ref-from-parameter <bugprone/return-const-ref-from-parameter>`
    :doc:`bugprone-shared-ptr-array-mismatch <bugprone/shared-ptr-array-mismatch>`, "Yes"
    :doc:`bugprone-signal-handler <bugprone/signal-handler>`,
    :doc:`bugprone-signed-char-misuse <bugprone/signed-char-misuse>`,
@@ -275,6 +276,7 @@ Clang-Tidy Checks
    :doc:`modernize-macro-to-enum <modernize/macro-to-enum>`, "Yes"
    :doc:`modernize-make-shared <modernize/make-shared>`, "Yes"
    :doc:`modernize-make-unique <modernize/make-unique>`, "Yes"
+   :doc:`modernize-min-max-use-initializer-list <modernize/min-max-use-initializer-list>`, "Yes"
    :doc:`modernize-pass-by-value <modernize/pass-by-value>`, "Yes"
    :doc:`modernize-raw-string-literal <modernize/raw-string-literal>`, "Yes"
    :doc:`modernize-redundant-void-arg <modernize/redundant-void-arg>`, "Yes"
@@ -362,6 +364,7 @@ Clang-Tidy Checks
    :doc:`readability-isolate-declaration <readability/isolate-declaration>`, "Yes"
    :doc:`readability-magic-numbers <readability/magic-numbers>`,
    :doc:`readability-make-member-function-const <readability/make-member-function-const>`, "Yes"
+   :doc:`readability-math-missing-parentheses <readability/math-missing-parentheses>`, "Yes"
    :doc:`readability-misleading-indentation <readability/misleading-indentation>`,
    :doc:`readability-misplaced-array-index <readability/misplaced-array-index>`, "Yes"
    :doc:`readability-named-parameter <readability/named-parameter>`, "Yes"
diff --git a/clang-tools-extra/docs/clang-tidy/checks/modernize/min-max-use-initializer-list.rst b/clang-tools-extra/docs/clang-tidy/checks/modernize/min-max-use-initializer-list.rst
new file mode 100644
index 000000000000..d6721a25629b
--- /dev/null
+++ b/clang-tools-extra/docs/clang-tidy/checks/modernize/min-max-use-initializer-list.rst
@@ -0,0 +1,50 @@
+.. title:: clang-tidy - modernize-min-max-use-initializer-list
+
+modernize-min-max-use-initializer-list
+======================================
+
+Replaces nested ``std::min`` and ``std::max`` calls with an initializer list 
+where applicable.
+
+For instance, consider the following code:
+
+.. code-block:: cpp
+
+  int a = std::max(std::max(i, j), k);
+
+The check will transform the above code to:
+
+.. code-block:: cpp
+
+  int a = std::max({i, j, k});
+
+Performance Considerations
+==========================
+
+While this check simplifies the code and makes it more readable, it may cause 
+performance degradation for non-trivial types due to the need to copy objects 
+into the initializer list.
+
+To avoid this, it is recommended to use `std::ref` or `std::cref` for
+non-trivial types:
+
+.. code-block:: cpp
+
+  std::string b = std::max({std::ref(i), std::ref(j), std::ref(k)});
+
+Options
+=======
+
+.. option:: IncludeStyle
+
+  A string specifying which include-style is used, `llvm` or `google`. Default
+  is `llvm`.
+
+.. option:: IgnoreNonTrivialTypes
+
+  A boolean specifying whether to ignore non-trivial types. Default is `true`.
+
+.. option:: IgnoreTrivialTypesOfSizeAbove
+
+  An integer specifying the size (in bytes) above which trivial types are
+  ignored. Default is `32`.
\ No newline at end of file
diff --git a/clang-tools-extra/docs/clang-tidy/checks/modernize/use-nullptr.rst b/clang-tools-extra/docs/clang-tidy/checks/modernize/use-nullptr.rst
index 5e1ba858adf3..25e17fee0a3d 100644
--- a/clang-tools-extra/docs/clang-tidy/checks/modernize/use-nullptr.rst
+++ b/clang-tools-extra/docs/clang-tidy/checks/modernize/use-nullptr.rst
@@ -4,7 +4,7 @@ modernize-use-nullptr
 =====================
 
 The check converts the usage of null pointer constants (e.g. ``NULL``, ``0``)
-to use the new C++11 ``nullptr`` keyword.
+to use the new C++11 and C23 ``nullptr`` keyword.
 
 Example
 -------
diff --git a/clang-tools-extra/docs/clang-tidy/checks/modernize/use-starts-ends-with.rst b/clang-tools-extra/docs/clang-tidy/checks/modernize/use-starts-ends-with.rst
index 7f8a262d2ab3..34237ede30a3 100644
--- a/clang-tools-extra/docs/clang-tidy/checks/modernize/use-starts-ends-with.rst
+++ b/clang-tools-extra/docs/clang-tidy/checks/modernize/use-starts-ends-with.rst
@@ -3,15 +3,16 @@
 modernize-use-starts-ends-with
 ==============================
 
-Checks whether a ``find`` or ``rfind`` result is compared with 0 and suggests
-replacing with ``starts_with`` when the method exists in the class. Notably,
-this will work with ``std::string`` and ``std::string_view``.
+Checks for common roundabout ways to express ``starts_with`` and ``ends_with``
+and suggests replacing with ``starts_with`` when the method is available.
+Notably, this will work with ``std::string`` and ``std::string_view``.
 
 .. code-block:: c++
 
   std::string s = "...";
   if (s.find("prefix") == 0) { /* do something */ }
   if (s.rfind("prefix", 0) == 0) { /* do something */ }
+  if (s.compare(0, strlen("prefix"), "prefix") == 0) { /* do something */ }
 
 becomes
 
@@ -20,3 +21,4 @@ becomes
   std::string s = "...";
   if (s.starts_with("prefix")) { /* do something */ }
   if (s.starts_with("prefix")) { /* do something */ }
+  if (s.starts_with("prefix")) { /* do something */ }
diff --git a/clang-tools-extra/docs/clang-tidy/checks/readability/math-missing-parentheses.rst b/clang-tools-extra/docs/clang-tidy/checks/readability/math-missing-parentheses.rst
new file mode 100644
index 000000000000..21d66daab334
--- /dev/null
+++ b/clang-tools-extra/docs/clang-tidy/checks/readability/math-missing-parentheses.rst
@@ -0,0 +1,27 @@
+.. title:: clang-tidy - readability-math-missing-parentheses
+
+readability-math-missing-parentheses
+====================================
+
+Check for missing parentheses in mathematical expressions that involve operators
+of different priorities.
+
+Parentheses in mathematical expressions clarify the order
+of operations, especially with different-priority operators. Lengthy or multiline
+expressions can obscure this order, leading to coding errors. IDEs can aid clarity
+by highlighting parentheses. Explicitly using parentheses also clarifies what the 
+developer had in mind when writing the expression. Ensuring their presence reduces
+ambiguity and errors, promoting clearer and more maintainable code.
+
+Before:
+
+.. code-block:: c++
+
+  int x = 1 + 2 * 3 - 4 / 5;
+
+
+After:
+
+.. code-block:: c++
+
+  int x = 1 + (2 * 3) - (4 / 5);
\ No newline at end of file
diff --git a/clang-tools-extra/test/clang-tidy/checkers/Inputs/Headers/string b/clang-tools-extra/test/clang-tidy/checkers/Inputs/Headers/string
index 28e2b4a231e5..d031f27beb9d 100644
--- a/clang-tools-extra/test/clang-tidy/checkers/Inputs/Headers/string
+++ b/clang-tools-extra/test/clang-tidy/checkers/Inputs/Headers/string
@@ -44,6 +44,8 @@ struct basic_string {
   int compare(const C* s) const;
   int compare(size_type pos, size_type len, const _Type&) const;
   int compare(size_type pos, size_type len, const C* s) const;
+  template<class StringViewLike>
+  int compare(size_type pos1, size_type count1, const StringViewLike& t) const;
 
   size_type find(const _Type& str, size_type pos = 0) const;
   size_type find(const C* s, size_type pos = 0) const;
@@ -129,6 +131,8 @@ bool operator!=(const char*, const std::string&);
 bool operator==(const std::wstring&, const std::wstring&);
 bool operator==(const std::wstring&, const wchar_t*);
 bool operator==(const wchar_t*, const std::wstring&);
+
+size_t strlen(const char* str);
 }
 
 #endif // _STRING_
diff --git a/clang-tools-extra/test/clang-tidy/checkers/Inputs/Headers/string.h b/clang-tools-extra/test/clang-tidy/checkers/Inputs/Headers/string.h
index 4ab7e930e4b5..af205868059a 100644
--- a/clang-tools-extra/test/clang-tidy/checkers/Inputs/Headers/string.h
+++ b/clang-tools-extra/test/clang-tidy/checkers/Inputs/Headers/string.h
@@ -12,5 +12,6 @@
 #include "stddef.h"
 
 void *memcpy(void *dest, const void *src, size_t n);
+size_t strlen(const char* str);
 
 #endif // _STRING_H_
diff --git a/clang-tools-extra/test/clang-tidy/checkers/abseil/redundant-strcat-calls.cpp b/clang-tools-extra/test/clang-tidy/checkers/abseil/redundant-strcat-calls.cpp
index ecd17bba293c..dbd354b132e2 100644
--- a/clang-tools-extra/test/clang-tidy/checkers/abseil/redundant-strcat-calls.cpp
+++ b/clang-tools-extra/test/clang-tidy/checkers/abseil/redundant-strcat-calls.cpp
@@ -1,8 +1,6 @@
 // RUN: %check_clang_tidy %s abseil-redundant-strcat-calls %t -- -- -isystem %clang_tidy_headers
 #include <string>
 
-int strlen(const char *);
-
 namespace absl {
 
 class string_view {
diff --git a/clang-tools-extra/test/clang-tidy/checkers/bugprone/return-const-ref-from-parameter.cpp b/clang-tools-extra/test/clang-tidy/checkers/bugprone/return-const-ref-from-parameter.cpp
new file mode 100644
index 000000000000..a83a019ec743
--- /dev/null
+++ b/clang-tools-extra/test/clang-tidy/checkers/bugprone/return-const-ref-from-parameter.cpp
@@ -0,0 +1,31 @@
+// RUN: %check_clang_tidy %s bugprone-return-const-ref-from-parameter %t
+
+using T = int;
+using TConst = int const;
+using TConstRef = int const&;
+
+namespace invalid {
+
+int const &f1(int const &a) { return a; }
+// CHECK-MESSAGES: :[[@LINE-1]]:38: warning: returning a constant reference parameter
+
+int const &f2(T const &a) { return a; }
+// CHECK-MESSAGES: :[[@LINE-1]]:36: warning: returning a constant reference parameter
+
+int const &f3(TConstRef a) { return a; }
+// CHECK-MESSAGES: :[[@LINE-1]]:37: warning: returning a constant reference parameter
+
+int const &f4(TConst &a) { return a; }
+// CHECK-MESSAGES: :[[@LINE-1]]:35: warning: returning a constant reference parameter
+
+} // namespace invalid
+
+namespace valid {
+
+int const &f1(int &a) { return a; }
+
+int const &f2(int &&a) { return a; }
+
+int f1(int const &a) { return a; }
+
+} // namespace valid
diff --git a/clang-tools-extra/test/clang-tidy/checkers/modernize/min-max-use-initializer-list.cpp b/clang-tools-extra/test/clang-tidy/checkers/modernize/min-max-use-initializer-list.cpp
new file mode 100644
index 000000000000..51ab9bda975f
--- /dev/null
+++ b/clang-tools-extra/test/clang-tidy/checkers/modernize/min-max-use-initializer-list.cpp
@@ -0,0 +1,305 @@
+// RUN: %check_clang_tidy %s modernize-min-max-use-initializer-list %t
+
+// CHECK-FIXES: #include <algorithm>
+namespace utils {
+template <typename T>
+T max(T a, T b) {
+  return (a < b) ? b : a;
+}
+} // namespace utils
+
+namespace std {
+template< class T >
+struct initializer_list {
+  initializer_list()=default;
+  initializer_list(T*,int){}
+  const T* begin() const {return nullptr;}
+  const T* end() const {return nullptr;}
+};
+
+template<class ForwardIt>
+ForwardIt min_element(ForwardIt first, ForwardIt last)
+{
+    if (first == last)
+        return last;
+
+    ForwardIt smallest = first;
+
+    while (++first != last)
+        if (*first < *smallest)
+            smallest = first;
+
+    return smallest;
+}
+
+template<class ForwardIt, class Compare>
+ForwardIt min_element(ForwardIt first, ForwardIt last, Compare comp)
+{
+    if (first == last)
+        return last;
+
+    ForwardIt smallest = first;
+
+    while (++first != last)
+        if (comp(*first, *smallest))
+            smallest = first;
+
+    return smallest;
+}
+
+template<class ForwardIt>
+ForwardIt max_element(ForwardIt first, ForwardIt last)
+{
+    if (first == last)
+        return last;
+
+    ForwardIt largest = first;
+
+    while (++first != last)
+        if (*largest < *first)
+            largest = first;
+
+    return largest;
+}
+
+template<class ForwardIt, class Compare>
+ForwardIt max_element(ForwardIt first, ForwardIt last, Compare comp)
+{
+    if (first == last)
+        return last;
+
+    ForwardIt largest = first;
+
+    while(++first != last)
+        if (comp(*largest, *first))
+            largest = first;
+
+    return largest;
+}
+
+template< class T >
+const T& max( const T& a, const T& b ) {
+  return (a < b) ? b : a;
+};
+
+template< class T >
+T max(std::initializer_list<T> ilist)
+{
+    return *std::max_element(ilist.begin(), ilist.end());
+}
+
+template< class T, class Compare >
+const T& max( const T& a, const T& b, Compare comp ) {
+  return (comp(a, b)) ? b : a;
+};
+
+template< class T, class Compare >
+T max(std::initializer_list<T> ilist, Compare comp) {
+    return *std::max_element(ilist.begin(), ilist.end(), comp);
+};
+
+template< class T >
+const T& min( const T& a, const T& b ) {
+  return (b < a) ? b : a;
+};
+
+template< class T >
+T min(std::initializer_list<T> ilist)
+{
+    return *std::min_element(ilist.begin(), ilist.end());
+}
+
+
+template< class T, class Compare >
+const T& min( const T& a, const T& b, Compare comp ) {
+  return (comp(b, a)) ? b : a;
+};
+
+template< class T, class Compare >
+T min(std::initializer_list<T> ilist, Compare comp) {
+    return *std::min_element(ilist.begin(), ilist.end(), comp);
+};
+
+} // namespace std
+
+using namespace std;
+
+namespace {
+bool fless_than(int a, int b) {
+return a < b;
+}
+
+bool fgreater_than(int a, int b) {
+return a > b;
+}
+auto less_than = [](int a, int b) { return a < b; };
+auto greater_than = [](int a, int b) { return a > b; };
+
+int max1 = std::max(1, std::max(2, 3));
+// CHECK-MESSAGES: :[[@LINE-1]]:12: warning: do not use nested 'std::max' calls, use an initializer list instead [modernize-min-max-use-initializer-list]
+// CHECK-FIXES: int max1 = std::max({1, 2, 3});
+
+int min1 = std::min(1, std::min(2, 3));
+// CHECK-MESSAGES: :[[@LINE-1]]:12: warning: do not use nested 'std::min' calls, use an initializer list instead [modernize-min-max-use-initializer-list]
+// CHECK-FIXES: int min1 = std::min({1, 2, 3});
+
+int max2 = std::max(1, std::max(2, std::max(3, 4)));
+// CHECK-MESSAGES: :[[@LINE-1]]:12: warning: do not use nested 'std::max' calls, use an initializer list instead [modernize-min-max-use-initializer-list]
+// CHECK-FIXES: int max2 = std::max({1, 2, 3, 4});
+
+int max2b = std::max(std::max(std::max(1, 2), std::max(3, 4)), std::max(std::max(5, 6), std::max(7, 8)));
+// CHECK-MESSAGES: :[[@LINE-1]]:13: warning: do not use nested 'std::max' calls, use an initializer list instead [modernize-min-max-use-initializer-list]
+// CHECK-FIXES: int max2b = std::max({1, 2, 3, 4, 5, 6, 7, 8});
+
+int max2c = std::max(std::max(1, std::max(2, 3)), 4);
+// CHECK-MESSAGES: :[[@LINE-1]]:13: warning: do not use nested 'std::max' calls, use an initializer list instead [modernize-min-max-use-initializer-list]
+// CHECK-FIXES: int max2c = std::max({1, 2, 3, 4});
+
+int max2d = std::max(std::max({1, 2, 3}), 4);
+// CHECK-MESSAGES: :[[@LINE-1]]:13: warning: do not use nested 'std::max' calls, use an initializer list instead [modernize-min-max-use-initializer-list]
+// CHECK-FIXES: int max2d = std::max({1, 2, 3, 4});
+
+
+int max2e = std::max(1, max(2, max(3, 4)));
+// CHECK-MESSAGES: :[[@LINE-1]]:13: warning: do not use nested 'std::max' calls, use an initializer list instead [modernize-min-max-use-initializer-list]
+// CHECK-FIXES: int max2e = std::max({1, 2, 3, 4});
+
+int min2 = std::min(1, std::min(2, std::min(3, 4)));
+// CHECK-MESSAGES: :[[@LINE-1]]:12: warning: do not use nested 'std::min' calls, use an initializer list instead [modernize-min-max-use-initializer-list]
+// CHECK-FIXES: int min2 = std::min({1, 2, 3, 4});
+
+int max3 = std::max(std::max(4, 5), std::min(2, std::min(3, 1)));
+// CHECK-MESSAGES: :[[@LINE-1]]:12: warning: do not use nested 'std::max' calls, use an initializer list instead [modernize-min-max-use-initializer-list]
+// CHECK-MESSAGES: :[[@LINE-2]]:37: warning: do not use nested 'std::min' calls, use an initializer list instead [modernize-min-max-use-initializer-list]
+// CHECK-FIXES: int max3 = std::max({4, 5, std::min({2, 3, 1})});
+
+int min3 = std::min(std::min(4, 5), std::max(2, std::max(3, 1)));
+// CHECK-MESSAGES: :[[@LINE-1]]:12: warning: do not use nested 'std::min' calls, use an initializer list instead [modernize-min-max-use-initializer-list]
+// CHECK-MESSAGES: :[[@LINE-2]]:37: warning: do not use nested 'std::max' calls, use an initializer list instead [modernize-min-max-use-initializer-list]
+// CHECK-FIXES: int min3 = std::min({4, 5, std::max({2, 3, 1})});
+
+int max4 = std::max(1, std::max(2, 3, greater_than), less_than);
+// CHECK-FIXES: int max4 = std::max(1, std::max(2, 3, greater_than), less_than);
+
+int min4 = std::min(1, std::min(2, 3, greater_than), less_than);
+// CHECK-FIXES: int min4 = std::min(1, std::min(2, 3, greater_than), less_than);
+
+int max5 = std::max(1, std::max(2, 3), less_than);
+// CHECK-FIXES: int max5 = std::max(1, std::max(2, 3), less_than);
+
+int min5 = std::min(1, std::min(2, 3), less_than);
+// CHECK-FIXES: int min5 = std::min(1, std::min(2, 3), less_than);
+
+int max6 = std::max(1, std::max(2, 3, greater_than), greater_than);
+// CHECK-MESSAGES: :[[@LINE-1]]:12: warning: do not use nested 'std::max' calls, use an initializer list instead [modernize-min-max-use-initializer-list]
+// CHECK-FIXES: int max6 = std::max({1, 2, 3 }, greater_than);
+
+int min6 = std::min(1, std::min(2, 3, greater_than), greater_than);
+// CHECK-MESSAGES: :[[@LINE-1]]:12: warning: do not use nested 'std::min' calls, use an initializer list instead [modernize-min-max-use-initializer-list]
+// CHECK-FIXES: int min6 = std::min({1, 2, 3 }, greater_than);
+
+int max7 = std::max(1, std::max(2, 3, fless_than), fgreater_than);
+// CHECK-FIXES: int max7 = std::max(1, std::max(2, 3, fless_than), fgreater_than);
+
+int min7 = std::min(1, std::min(2, 3, fless_than), fgreater_than);
+// CHECK-FIXES: int min7 = std::min(1, std::min(2, 3, fless_than), fgreater_than);
+
+int max8 = std::max(1, std::max(2, 3, fless_than), less_than);
+// CHECK-FIXES: int max8 = std::max(1, std::max(2, 3, fless_than), less_than)
+
+int min8 = std::min(1, std::min(2, 3, fless_than), less_than);
+// CHECK-FIXES: int min8 = std::min(1, std::min(2, 3, fless_than), less_than);
+
+int max9 = std::max(1, std::max(2, 3, fless_than), fless_than);
+// CHECK-MESSAGES: :[[@LINE-1]]:12: warning: do not use nested 'std::max' calls, use an initializer list instead [modernize-min-max-use-initializer-list]
+// CHECK-FIXES: int max9 = std::max({1, 2, 3 }, fless_than);
+
+int min9 = std::min(1, std::min(2, 3, fless_than), fless_than);
+// CHECK-MESSAGES: :[[@LINE-1]]:12: warning: do not use nested 'std::min' calls, use an initializer list instead [modernize-min-max-use-initializer-list]
+// CHECK-FIXES: int min9 = std::min({1, 2, 3 }, fless_than);
+
+int min10 = std::min(std::min(4, 5), std::max(2, utils::max(3, 1)));
+// CHECK-MESSAGES: :[[@LINE-1]]:13: warning: do not use nested 'std::min' calls, use an initializer list instead [modernize-min-max-use-initializer-list]
+// CHECK-FIXES: int min10 = std::min({4, 5, std::max(2, utils::max(3, 1))});
+
+int max10 = std::max({std::max(1, 2), std::max({5, 6, 1}), 2, std::min({1, 2, 4})});
+// CHECK-MESSAGES: :[[@LINE-1]]:13: warning: do not use nested 'std::max' calls, use an initializer list instead [modernize-min-max-use-initializer-list]
+// CHECK-FIXES: int max10 = std::max({1, 2, 5, 6, 1, 2, std::min({1, 2, 4})});
+
+int typecastTest = std::max(std::max<int>(0U, 0.0f), 0);
+// CHECK-MESSAGES: :[[@LINE-1]]:20: warning: do not use nested 'std::max' calls, use an initializer list instead [modernize-min-max-use-initializer-list]
+// CHECK-FIXES: int typecastTest = std::max({static_cast<int>(0U), static_cast<int>(0.0f), 0});
+
+int typecastTest1 = std::max(std::max<long>(0U, 0.0f), 0L);
+// CHECK-MESSAGES: :[[@LINE-1]]:21: warning: do not use nested 'std::max' calls, use an initializer list instead [modernize-min-max-use-initializer-list]
+// CHECK-FIXES: int typecastTest1 = std::max({static_cast<long>(0U), static_cast<long>(0.0f), 0L});
+
+int typecastTest2 = std::max(std::max<int>(10U, 20.0f), 30);
+// CHECK-MESSAGES: :[[@LINE-1]]:21: warning: do not use nested 'std::max' calls, use an initializer list instead [modernize-min-max-use-initializer-list]
+// CHECK-FIXES: int typecastTest2 = std::max({static_cast<int>(10U), static_cast<int>(20.0f), 30});
+
+int typecastTest3 = std::max(std::max<int>(0U, std::max<int>(0.0f, 1.0f)), 0);
+// CHECK-MESSAGES: :[[@LINE-1]]:21: warning: do not use nested 'std::max' calls, use an initializer list instead [modernize-min-max-use-initializer-list]
+// CHECK-FIXES: int typecastTest3 = std::max({static_cast<int>(0U), static_cast<int>(0.0f), static_cast<int>(1.0f), 0});
+
+#define max3f(a, b, c) std::max(a, std::max(b, c))
+// CHECK-FIXES: #define max3f(a, b, c) std::max(a, std::max(b, c))
+
+#define value 4545
+int macroVarMax = std::max(value, std::max(1, 2));
+// CHECK-MESSAGES: :[[@LINE-1]]:19: warning: do not use nested 'std::max' calls, use an initializer list instead [modernize-min-max-use-initializer-list]
+// CHECK-FIXES: int macroVarMax = std::max({value, 1, 2});
+
+#define value2 45U
+int macroVarMax2 = std::max(1, std::max<int>(value2, 2.0f));
+// CHECK-MESSAGES: :[[@LINE-1]]:20: warning: do not use nested 'std::max' calls, use an initializer list instead [modernize-min-max-use-initializer-list]
+// CHECK-FIXES: int macroVarMax2 = std::max({1, static_cast<int>(value2), static_cast<int>(2.0f)});
+
+// True-negative tests
+int maxTN1 = std::max(1, 2);
+// CHECK-FIXES: int maxTN1 = std::max(1, 2);
+
+int maxTN2 = std::max({1, 2, 3});
+// CHECK-FIXES: int maxTN2 = std::max({1, 2, 3});
+
+int maxTN3 = std::max({1, 2, 3}, less_than);
+// CHECK-FIXES: int maxTN3 = std::max({1, 2, 3}, less_than);
+
+// non-trivial types
+struct A {
+  int a;
+  A(int a) : a(a) {}
+  bool operator<(const A &rhs) const { return a < rhs.a; }
+};
+
+A maxNT1 = std::max(A(1), A(2));
+// CHECK-FIXES: A maxNT1 = std::max(A(1), A(2));
+
+A maxNT2 = std::max(A(1), std::max(A(2), A(3)));
+// CHECK-FIXES: A maxNT2 = std::max(A(1), std::max(A(2), A(3)));
+
+A maxNT3 = std::max(A(1), std::max(A(2), A(3)), [](const A &lhs, const A &rhs) { return lhs.a < rhs.a; });
+// CHECK-FIXES: A maxNT3 = std::max(A(1), std::max(A(2), A(3)), [](const A &lhs, const A &rhs) { return lhs.a < rhs.a; });
+
+// Trivial type with size greater than 32
+struct B {
+  // 9*4 = 36 bytes > 32 bytes
+  int a[9];
+
+  bool operator<(const B& rhs) const {
+    return a[0] < rhs.a[0];
+  }
+};
+
+B maxTT1 = std::max(B(), B());
+// CHECK-FIXES: B maxTT1 = std::max(B(), B());
+
+B maxTT2 = std::max(B(), std::max(B(), B()));
+// CHECK-FIXES: B maxTT2 = std::max(B(), std::max(B(), B()));
+
+B maxTT3 = std::max(B(), std::max(B(), B()), [](const B &lhs, const B &rhs) { return lhs.a[0] < rhs.a[0]; });
+// CHECK-FIXES: B maxTT3 = std::max(B(), std::max(B(), B()), [](const B &lhs, const B &rhs) { return lhs.a[0] < rhs.a[0]; });
+
+
+} // namespace
+
diff --git a/clang-tools-extra/test/clang-tidy/checkers/modernize/use-nullptr-c23.c b/clang-tools-extra/test/clang-tidy/checkers/modernize/use-nullptr-c23.c
new file mode 100644
index 000000000000..6fb879b91e41
--- /dev/null
+++ b/clang-tools-extra/test/clang-tidy/checkers/modernize/use-nullptr-c23.c
@@ -0,0 +1,139 @@
+// RUN: %check_clang_tidy %s modernize-use-nullptr %t -- -- -std=c23
+
+#define NULL 0
+
+void test_assignment() {
+  int *p1 = 0;
+  // CHECK-MESSAGES: :[[@LINE-1]]:13: warning: use nullptr [modernize-use-nullptr]
+  // CHECK-FIXES: int *p1 = nullptr;
+  p1 = 0;
+  // CHECK-MESSAGES: :[[@LINE-1]]:8: warning: use nullptr
+  // CHECK-FIXES: p1 = nullptr;
+
+  int *p2 = NULL;
+  // CHECK-MESSAGES: :[[@LINE-1]]:13: warning: use nullptr
+  // CHECK-FIXES: int *p2 = nullptr;
+
+  p2 = p1;
+  // CHECK-FIXES: p2 = p1;
+
+  const int null = 0;
+  int *p3 = &null;
+
+  p3 = NULL;
+  // CHECK-MESSAGES: :[[@LINE-1]]:8: warning: use nullptr
+  // CHECK-FIXES: p3 = nullptr;
+
+  int *p4 = p3;
+
+  int i1 = 0;
+
+  int i2 = NULL;
+
+  int i3 = null;
+
+  int *p5, *p6, *p7;
+  p5 = p6 = p7 = NULL;
+  // CHECK-MESSAGES: :[[@LINE-1]]:18: warning: use nullptr
+  // CHECK-FIXES: p5 = p6 = p7 = nullptr;
+}
+
+void test_function(int *p) {}
+
+void test_function_no_ptr_param(int i) {}
+
+void test_function_call() {
+  test_function(0);
+  // CHECK-MESSAGES: :[[@LINE-1]]:17: warning: use nullptr
+  // CHECK-FIXES: test_function(nullptr);
+
+  test_function(NULL);
+  // CHECK-MESSAGES: :[[@LINE-1]]:17: warning: use nullptr
+  // CHECK-FIXES: test_function(nullptr);
+
+  test_function_no_ptr_param(0);
+}
+
+char *test_function_return1() {
+  return 0;
+  // CHECK-MESSAGES: :[[@LINE-1]]:10: warning: use nullptr
+  // CHECK-FIXES: return nullptr;
+}
+
+void *test_function_return2() {
+  return NULL;
+  // CHECK-MESSAGES: :[[@LINE-1]]:10: warning: use nullptr
+  // CHECK-FIXES: return nullptr;
+}
+
+int test_function_return4() {
+  return 0;
+}
+
+int test_function_return5() {
+  return NULL;
+}
+
+int *test_function_return_cast1() {
+  return(int)0;
+  // CHECK-MESSAGES: :[[@LINE-1]]:9: warning: use nullptr
+  // CHECK-FIXES: return nullptr;
+}
+
+int *test_function_return_cast2() {
+#define RET return
+  RET(int)0;
+  // CHECK-MESSAGES: :[[@LINE-1]]:6: warning: use nullptr
+  // CHECK-FIXES: RET nullptr;
+#undef RET
+}
+
+// Test parentheses expressions resulting in a nullptr.
+int *test_parentheses_expression1() {
+  return(0);
+  // CHECK-MESSAGES: :[[@LINE-1]]:10: warning: use nullptr
+  // CHECK-FIXES: return(nullptr);
+}
+
+int *test_parentheses_expression2() {
+  return((int)(0.0f));
+  // CHECK-MESSAGES: :[[@LINE-1]]:10: warning: use nullptr
+  // CHECK-FIXES: return(nullptr);
+}
+
+int *test_nested_parentheses_expression() {
+  return((((0))));
+  // CHECK-MESSAGES: :[[@LINE-1]]:13: warning: use nullptr
+  // CHECK-FIXES: return((((nullptr))));
+}
+
+void test_const_pointers() {
+  const int *const_p1 = 0;
+  // CHECK-MESSAGES: :[[@LINE-1]]:25: warning: use nullptr
+  // CHECK-FIXES: const int *const_p1 = nullptr;
+  const int *const_p2 = NULL;
+  // CHECK-MESSAGES: :[[@LINE-1]]:25: warning: use nullptr
+  // CHECK-FIXES: const int *const_p2 = nullptr;
+  const int *const_p3 = (int)0;
+  // CHECK-MESSAGES: :[[@LINE-1]]:25: warning: use nullptr
+  // CHECK-FIXES: const int *const_p3 = nullptr;
+  const int *const_p4 = (int)0.0f;
+  // CHECK-MESSAGES: :[[@LINE-1]]:25: warning: use nullptr
+  // CHECK-FIXES: const int *const_p4 = nullptr;
+}
+
+void test_nested_implicit_cast_expr() {
+  int func0(void*, void*);
+  int func1(int, void*, void*);
+
+  (double)func1(0, 0, 0);
+  // CHECK-MESSAGES: :[[@LINE-1]]:20: warning: use nullptr
+  // CHECK-MESSAGES: :[[@LINE-2]]:23: warning: use nullptr
+  // CHECK-FIXES: (double)func1(0, nullptr, nullptr);
+  (double)func1(func0(0, 0), 0, 0);
+  // CHECK-MESSAGES: :[[@LINE-1]]:23: warning: use nullptr
+  // CHECK-MESSAGES: :[[@LINE-2]]:26: warning: use nullptr
+  // CHECK-MESSAGES: :[[@LINE-3]]:30: warning: use nullptr
+  // CHECK-MESSAGES: :[[@LINE-4]]:33: warning: use nullptr
+  // CHECK-FIXES: (double)func1(func0(nullptr, nullptr), nullptr, nullptr);
+}
diff --git a/clang-tools-extra/test/clang-tidy/checkers/modernize/use-nullptr.c b/clang-tools-extra/test/clang-tidy/checkers/modernize/use-nullptr.c
index c2ccbbd81171..1218b837199c 100644
--- a/clang-tools-extra/test/clang-tidy/checkers/modernize/use-nullptr.c
+++ b/clang-tools-extra/test/clang-tidy/checkers/modernize/use-nullptr.c
@@ -1,4 +1,4 @@
-// RUN: clang-tidy %s -checks=-*,modernize-use-nullptr -- | count 0
+// RUN: clang-tidy %s -checks=-*,modernize-use-nullptr -- -std=c17 | count 0
 
 // Note: this test expects no diagnostics, but FileCheck cannot handle that,
 // hence the use of | count 0.
diff --git a/clang-tools-extra/test/clang-tidy/checkers/modernize/use-starts-ends-with.cpp b/clang-tools-extra/test/clang-tidy/checkers/modernize/use-starts-ends-with.cpp
index 65ed9ed895bc..c5b2c86befd1 100644
--- a/clang-tools-extra/test/clang-tidy/checkers/modernize/use-starts-ends-with.cpp
+++ b/clang-tools-extra/test/clang-tidy/checkers/modernize/use-starts-ends-with.cpp
@@ -1,6 +1,7 @@
 // RUN: %check_clang_tidy -std=c++20 %s modernize-use-starts-ends-with %t -- \
 // RUN:   -- -isystem %clang_tidy_headers
 
+#include <string.h>
 #include <string>
 
 std::string foo(std::string);
@@ -158,10 +159,64 @@ void test(std::string s, std::string_view sv, sub_string ss, sub_sub_string sss,
   // CHECK-MESSAGES: :[[@LINE-1]]:{{[0-9]+}}: warning: use startsWith
   // CHECK-FIXES: puvi.startsWith("a");
 
+  s.compare(0, 1, "a") == 0;
+  // CHECK-MESSAGES: :[[@LINE-1]]:{{[0-9]+}}: warning: use starts_with instead of compare() == 0
+  // CHECK-FIXES: s.starts_with("a");
+
+  s.compare(0, 1, "a") != 0;
+  // CHECK-MESSAGES: :[[@LINE-1]]:{{[0-9]+}}: warning: use starts_with instead of compare() != 0
+  // CHECK-FIXES: !s.starts_with("a");
+
+  s.compare(0, strlen("a"), "a") == 0;
+  // CHECK-MESSAGES: :[[@LINE-1]]:{{[0-9]+}}: warning: use starts_with
+  // CHECK-FIXES: s.starts_with("a");
+
+  s.compare(0, std::strlen("a"), "a") == 0;
+  // CHECK-MESSAGES: :[[@LINE-1]]:{{[0-9]+}}: warning: use starts_with
+  // CHECK-FIXES: s.starts_with("a");
+
+  s.compare(0, std::strlen(("a")), "a") == 0;
+  // CHECK-MESSAGES: :[[@LINE-1]]:{{[0-9]+}}: warning: use starts_with
+  // CHECK-FIXES: s.starts_with("a");
+
+  s.compare(0, std::strlen(("a")), (("a"))) == 0;
+  // CHECK-MESSAGES: :[[@LINE-1]]:{{[0-9]+}}: warning: use starts_with
+  // CHECK-FIXES: s.starts_with("a");
+
+  s.compare(0, s.size(), s) == 0;
+  // CHECK-MESSAGES: :[[@LINE-1]]:{{[0-9]+}}: warning: use starts_with
+  // CHECK-FIXES: s.starts_with(s);
+
+  s.compare(0, s.length(), s) == 0;
+  // CHECK-MESSAGES: :[[@LINE-1]]:{{[0-9]+}}: warning: use starts_with
+  // CHECK-FIXES: s.starts_with(s);
+
+  0 != s.compare(0, sv.length(), sv);
+  // CHECK-MESSAGES: :[[@LINE-1]]:{{[0-9]+}}: warning: use starts_with
+  // CHECK-FIXES: s.starts_with(sv);
+
+  #define LENGTH(x) (x).length()
+  s.compare(0, LENGTH(s), s) == 0;
+  // CHECK-MESSAGES: :[[@LINE-1]]:{{[0-9]+}}: warning: use starts_with
+  // CHECK-FIXES: s.starts_with(s);
+
+  s.compare(ZERO, LENGTH(s), s) == ZERO;
+  // CHECK-MESSAGES: :[[@LINE-1]]:{{[0-9]+}}: warning: use starts_with
+  // CHECK-FIXES: s.starts_with(s);
+
+  s.compare(ZERO, LENGTH(sv), sv) != 0;
+  // CHECK-MESSAGES: :[[@LINE-1]]:{{[0-9]+}}: warning: use starts_with
+  // CHECK-FIXES: !s.starts_with(sv);
+
   // Expressions that don't trigger the check are here.
   #define EQ(x, y) ((x) == (y))
   EQ(s.find("a"), 0);
 
   #define DOTFIND(x, y) (x).find(y)
   DOTFIND(s, "a") == 0;
+
+  #define STARTS_WITH_COMPARE(x, y) (x).compare(0, (x).size(), (y))
+  STARTS_WITH_COMPARE(s, s) == 0;
+
+  s.compare(0, 1, "ab") == 0;
 }
diff --git a/clang-tools-extra/test/clang-tidy/checkers/readability/math-missing-parentheses.cpp b/clang-tools-extra/test/clang-tidy/checkers/readability/math-missing-parentheses.cpp
new file mode 100644
index 000000000000..edbe2e1c37c7
--- /dev/null
+++ b/clang-tools-extra/test/clang-tidy/checkers/readability/math-missing-parentheses.cpp
@@ -0,0 +1,120 @@
+// RUN: %check_clang_tidy %s readability-math-missing-parentheses %t
+
+#define MACRO_AND &
+#define MACRO_ADD +
+#define MACRO_OR |
+#define MACRO_MULTIPLY *
+#define MACRO_XOR ^
+#define MACRO_SUBTRACT -
+#define MACRO_DIVIDE /
+
+int foo(){
+    return 5;
+}
+
+int bar(){
+    return 4;
+}
+
+class fun{
+public:
+    int A;
+    double B;
+    fun(){
+        A = 5;
+        B = 5.4;
+    }
+};
+
+void f(){
+    //CHECK-MESSAGES: :[[@LINE+2]]:17: warning: '*' has higher precedence than '+'; add parentheses to explicitly specify the order of operations [readability-math-missing-parentheses]
+    //CHECK-FIXES: int a = 1 + (2 * 3);
+    int a = 1 + 2 * 3;
+
+    int a_negative = 1 + (2 * 3); // No warning
+
+    int b = 1 + 2 + 3; // No warning
+
+    int c = 1 * 2 * 3; // No warning
+
+    //CHECK-MESSAGES: :[[@LINE+3]]:17: warning: '*' has higher precedence than '+'; add parentheses to explicitly specify the order of operations [readability-math-missing-parentheses]
+    //CHECK-MESSAGES: :[[@LINE+2]]:25: warning: '/' has higher precedence than '-'; add parentheses to explicitly specify the order of operations [readability-math-missing-parentheses]
+    //CHECK-FIXES: int d = 1 + (2 * 3) - (4 / 5);
+    int d = 1 + 2 * 3 - 4 / 5;
+
+    int d_negative = 1 + (2 * 3) - (4 / 5); // No warning
+
+    //CHECK-MESSAGES: :[[@LINE+4]]:13: warning: '&' has higher precedence than '|'; add parentheses to explicitly specify the order of operations [readability-math-missing-parentheses]
+    //CHECK-MESSAGES: :[[@LINE+3]]:17: warning: '+' has higher precedence than '&'; add parentheses to explicitly specify the order of operations [readability-math-missing-parentheses]
+    //CHECK-MESSAGES: :[[@LINE+2]]:25: warning: '*' has higher precedence than '|'; add parentheses to explicitly specify the order of operations [readability-math-missing-parentheses]
+    //CHECK-FIXES: int e = (1 & (2 + 3)) | (4 * 5);
+    int e = 1 & 2 + 3 | 4 * 5;
+
+    int e_negative = (1 & (2 + 3)) | (4 * 5); // No warning
+
+    //CHECK-MESSAGES: :[[@LINE+2]]:13: warning: '*' has higher precedence than '+'; add parentheses to explicitly specify the order of operations [readability-math-missing-parentheses]
+    //CHECK-FIXES: int f = (1 * -2) + 4;
+    int f = 1 * -2 + 4;
+
+    int f_negative = (1 * -2) + 4; // No warning
+
+    //CHECK-MESSAGES: :[[@LINE+2]]:13: warning: '*' has higher precedence than '+'; add parentheses to explicitly specify the order of operations [readability-math-missing-parentheses]
+    //CHECK-FIXES: int g = (1 * 2 * 3) + 4 + 5;
+    int g = 1 * 2 * 3 + 4 + 5;
+
+    int g_negative = (1 * 2 * 3) + 4 + 5; // No warning
+
+    //CHECK-MESSAGES: :[[@LINE+4]]:13: warning: '&' has higher precedence than '|'; add parentheses to explicitly specify the order of operations [readability-math-missing-parentheses]
+    //CHECK-MESSAGES: :[[@LINE+3]]:19: warning: '+' has higher precedence than '&'; add parentheses to explicitly specify the order of operations [readability-math-missing-parentheses]
+    //CHECK-MESSAGES: :[[@LINE+2]]:27: warning: '*' has higher precedence than '|'; add parentheses to explicitly specify the order of operations [readability-math-missing-parentheses]
+    //CHECK-FIXES: int h = (120 & (2 + 3)) | (22 * 5);
+    int h = 120 & 2 + 3 | 22 * 5;
+
+    int h_negative = (120 & (2 + 3)) | (22 * 5); // No warning
+
+    int i = 1 & 2 & 3; // No warning
+
+    int j = 1 | 2 | 3; // No warning
+
+    int k = 1 ^ 2 ^ 3; // No warning
+
+    //CHECK-MESSAGES: :[[@LINE+2]]:13: warning: '+' has higher precedence than '^'; add parentheses to explicitly specify the order of operations [readability-math-missing-parentheses]
+    //CHECK-FIXES: int l = (1 + 2) ^ 3;
+    int l = 1 + 2 ^ 3;
+
+    int l_negative = (1 + 2) ^ 3; // No warning
+
+    //CHECK-MESSAGES: :[[@LINE+2]]:13: warning: '*' has higher precedence than '+'; add parentheses to explicitly specify the order of operations [readability-math-missing-parentheses]
+    //CHECK-FIXES: int m = (2 * foo()) + bar();
+    int m = 2 * foo() + bar();
+
+    int m_negative = (2 * foo()) + bar(); // No warning
+
+    //CHECK-MESSAGES: :[[@LINE+2]]:13: warning: '*' has higher precedence than '+'; add parentheses to explicitly specify the order of operations [readability-math-missing-parentheses]
+    //CHECK-FIXES: int n = (1.05 * foo()) + double(bar());
+    int n = 1.05 * foo() + double(bar());
+
+    int n_negative = (1.05 * foo()) + double(bar()); // No warning
+
+    //CHECK-MESSAGES: :[[@LINE+3]]:17: warning: '*' has higher precedence than '+'; add parentheses to explicitly specify the order of operations [readability-math-missing-parentheses]
+    //CHECK-FIXES: int o = 1 + (obj.A * 3) + obj.B;
+    fun obj;
+    int o = 1 + obj.A * 3 + obj.B;
+
+    int o_negative = 1 + (obj.A * 3) + obj.B; // No warning
+
+    //CHECK-MESSAGES: :[[@LINE+2]]:18: warning: '*' has higher precedence than '+'; add parentheses to explicitly specify the order of operations [readability-math-missing-parentheses]
+    //CHECK-FIXES: int p = 1U + (2 * 3);
+    int p = 1U + 2 * 3;
+
+    int p_negative = 1U + (2 * 3); // No warning
+
+    //CHECK-MESSAGES: :[[@LINE+7]]:13: warning: '+' has higher precedence than '|'; add parentheses to explicitly specify the order of operations [readability-math-missing-parentheses]
+    //CHECK-MESSAGES: :[[@LINE+6]]:25: warning: '*' has higher precedence than '+'; add parentheses to explicitly specify the order of operations [readability-math-missing-parentheses]
+    //CHECK-MESSAGES: :[[@LINE+5]]:53: warning: '&' has higher precedence than '^'; add parentheses to explicitly specify the order of operations [readability-math-missing-parentheses]
+    //CHECK-MESSAGES: :[[@LINE+4]]:53: warning: '^' has higher precedence than '|'; add parentheses to explicitly specify the order of operations [readability-math-missing-parentheses]
+    //CHECK-MESSAGES: :[[@LINE+3]]:77: warning: '-' has higher precedence than '^'; add parentheses to explicitly specify the order of operations [readability-math-missing-parentheses]
+    //CHECK-MESSAGES: :[[@LINE+2]]:94: warning: '/' has higher precedence than '-'; add parentheses to explicitly specify the order of operations [readability-math-missing-parentheses]
+    //CHECK-FIXES: int q = (1 MACRO_ADD (2 MACRO_MULTIPLY 3)) MACRO_OR ((4 MACRO_AND 5) MACRO_XOR (6 MACRO_SUBTRACT (7 MACRO_DIVIDE 8)));
+    int q = 1 MACRO_ADD 2 MACRO_MULTIPLY 3 MACRO_OR 4 MACRO_AND 5 MACRO_XOR 6 MACRO_SUBTRACT 7 MACRO_DIVIDE 8; // No warning
+}
diff --git a/clang-tools-extra/test/clang-tidy/infrastructure/Inputs/config-files/5/.clang-tidy b/clang-tools-extra/test/clang-tidy/infrastructure/Inputs/config-files/5/.clang-tidy
new file mode 100644
index 000000000000..e33f0f8bb332
--- /dev/null
+++ b/clang-tools-extra/test/clang-tidy/infrastructure/Inputs/config-files/5/.clang-tidy
@@ -0,0 +1,4 @@
+InheritParentConfig: true
+Checks: 'misc-throw-by-value-catch-by-reference'
+CheckOptions:
+  misc-throw-by-value-catch-by-reference.MaxSize: '1152921504606846976'
diff --git a/clang-tools-extra/test/clang-tidy/infrastructure/config-files.cpp b/clang-tools-extra/test/clang-tidy/infrastructure/config-files.cpp
index ab4f3becb7a9..cb0f0bc4d133 100644
--- a/clang-tools-extra/test/clang-tidy/infrastructure/config-files.cpp
+++ b/clang-tools-extra/test/clang-tidy/infrastructure/config-files.cpp
@@ -64,3 +64,11 @@
 
 // Validate that check options are printed in alphabetical order:
 // RUN: clang-tidy --checks="-*,readability-identifier-naming" --dump-config %S/Inputs/config-files/- -- | grep "readability-identifier-naming\." | sort --check
+
+// Dumped config does not overflow for unsigned options
+// RUN: clang-tidy --dump-config \
+// RUN: --checks="-*,misc-throw-by-value-catch-by-reference" \
+// RUN: -- | grep -v -q "misc-throw-by-value-catch-by-reference.MaxSize: '-1'"
+
+// RUN: clang-tidy --dump-config %S/Inputs/config-files/5/- \
+// RUN: -- | grep -q "misc-throw-by-value-catch-by-reference.MaxSize: '1152921504606846976'"
diff --git a/clang-tools-extra/test/pp-trace/pp-trace-pragma-general.cpp b/clang-tools-extra/test/pp-trace/pp-trace-pragma-general.cpp
index f01ebd1ec67d..b16ec56e321b 100644
--- a/clang-tools-extra/test/pp-trace/pp-trace-pragma-general.cpp
+++ b/clang-tools-extra/test/pp-trace/pp-trace-pragma-general.cpp
@@ -21,6 +21,12 @@ void foo() {
 
 // CHECK: ---
 // CHECK-NEXT: - Callback: PragmaDirective
+// CHECK-NEXT:   Loc: "<built-in>:{{.+}}:1"
+// CHECK-NEXT:   Introducer: PIK_HashPragma
+// CHECK-NEXT: - Callback: PragmaDirective
+// CHECK-NEXT:   Loc: "<built-in>:{{.+}}:1"
+// CHECK-NEXT:   Introducer: PIK_HashPragma
+// CHECK-NEXT: - Callback: PragmaDirective
 // CHECK-NEXT:   Loc: "{{.*}}{{[/\\]}}pp-trace-pragma-general.cpp:3:1"
 // CHECK-NEXT:   Introducer: PIK_HashPragma
 // CHECK-NEXT: - Callback: PragmaDiagnosticPush
diff --git a/clang-tools-extra/test/pp-trace/pp-trace-pragma-ms.cpp b/clang-tools-extra/test/pp-trace/pp-trace-pragma-ms.cpp
index 932b0eb93c90..f5bf9ac2b955 100644
--- a/clang-tools-extra/test/pp-trace/pp-trace-pragma-ms.cpp
+++ b/clang-tools-extra/test/pp-trace/pp-trace-pragma-ms.cpp
@@ -18,6 +18,12 @@
 
 // CHECK: ---
 // CHECK-NEXT: - Callback: PragmaDirective
+// CHECK-NEXT:   Loc: "<built-in>:{{.+}}:1"
+// CHECK-NEXT:   Introducer: PIK_HashPragma
+// CHECK-NEXT: - Callback: PragmaDirective
+// CHECK-NEXT:   Loc: "<built-in>:{{.+}}:1"
+// CHECK-NEXT:   Introducer: PIK_HashPragma
+// CHECK-NEXT: - Callback: PragmaDirective
 // CHECK-NEXT:   Loc: "{{.*}}{{[/\\]}}pp-trace-pragma-ms.cpp:3:1"
 // CHECK-NEXT:   Introducer: PIK_HashPragma
 // CHECK-NEXT: - Callback: PragmaComment
@@ -67,7 +73,7 @@
 // CHECK-NEXT:   Introducer: PIK_HashPragma
 // CHECK-NEXT: - Callback: PragmaMessage
 // CHECK-NEXT:   Loc: "{{.*}}{{[/\\]}}pp-trace-pragma-ms.cpp:13:9"
-// CHECK-NEXT:   Namespace: 
+// CHECK-NEXT:   Namespace:
 // CHECK-NEXT:   Kind: PMK_Message
 // CHECK-NEXT:   Str: message argument
 // CHECK-NEXT: - Callback: PragmaDirective
diff --git a/clang-tools-extra/test/pp-trace/pp-trace-pragma-opencl.cpp b/clang-tools-extra/test/pp-trace/pp-trace-pragma-opencl.cpp
index 31f61027994f..ed33d37eb3d5 100644
--- a/clang-tools-extra/test/pp-trace/pp-trace-pragma-opencl.cpp
+++ b/clang-tools-extra/test/pp-trace/pp-trace-pragma-opencl.cpp
@@ -6,6 +6,12 @@
 
 // CHECK: ---
 // CHECK-NEXT: - Callback: PragmaDirective
+// CHECK-NEXT:   Loc: "<built-in>:{{.+}}:1"
+// CHECK-NEXT:   Introducer: PIK_HashPragma
+// CHECK-NEXT: - Callback: PragmaDirective
+// CHECK-NEXT:   Loc: "<built-in>:{{.+}}:1"
+// CHECK-NEXT:   Introducer: PIK_HashPragma
+// CHECK-NEXT: - Callback: PragmaDirective
 // CHECK-NEXT:   Loc: "{{.*}}{{[/\\]}}pp-trace-pragma-opencl.cpp:3:1"
 // CHECK-NEXT:   Introducer: PIK_HashPragma
 // CHECK-NEXT: - Callback: PragmaOpenCLExtension
diff --git a/clang/CMakeLists.txt b/clang/CMakeLists.txt
index f092766fa19f..cf97e3c6e851 100644
--- a/clang/CMakeLists.txt
+++ b/clang/CMakeLists.txt
@@ -166,6 +166,10 @@ if(CLANG_ENABLE_LIBXML2)
 endif()
 
 if(CLANG_ENABLE_CIR)
+  if (CLANG_BUILT_STANDALONE)
+    message(FATAL_ERROR
+      "ClangIR is not yet supported in the standalone build.")
+  endif()
   if (NOT "${LLVM_ENABLE_PROJECTS}" MATCHES "MLIR|mlir")
     message(FATAL_ERROR
       "Cannot build ClangIR without MLIR in LLVM_ENABLE_PROJECTS")
diff --git a/clang/cmake/caches/Release.cmake b/clang/cmake/caches/Release.cmake
index bd1f688d61a7..c0bfcbdfc1c2 100644
--- a/clang/cmake/caches/Release.cmake
+++ b/clang/cmake/caches/Release.cmake
@@ -1,93 +1,94 @@
 # Plain options configure the first build.
 # BOOTSTRAP_* options configure the second build.
 # BOOTSTRAP_BOOTSTRAP_* options configure the third build.
+# PGO Builds have 3 stages (stage1, stage2-instrumented, stage2)
+# non-PGO Builds have 2 stages (stage1, stage2)
 
-# General Options
+
+function (set_final_stage_var name value type)
+  if (LLVM_RELEASE_ENABLE_PGO)
+    set(BOOTSTRAP_BOOTSTRAP_${name} ${value} CACHE ${type} "")
+  else()
+    set(BOOTSTRAP_${name} ${value} CACHE ${type} "")
+  endif()
+endfunction()
+
+function (set_instrument_and_final_stage_var name value type)
+  # This sets the varaible for the final stage in non-PGO builds and in
+  # the stage2-instrumented stage for PGO builds.
+  set(BOOTSTRAP_${name} ${value} CACHE ${type} "")
+  if (LLVM_RELEASE_ENABLE_PGO)
+    # Set the variable in the final stage for PGO builds.
+    set(BOOTSTRAP_BOOTSTRAP_${name} ${value} CACHE ${type} "")
+  endif()
+endfunction()
+
+# General Options:
+# If you want to override any of the LLVM_RELEASE_* variables you can set them
+# on the command line via -D, but you need to do this before you pass this
+# cache file to CMake via -C. e.g.
+#
+# cmake -D LLVM_RELEASE_ENABLE_PGO=ON -C Release.cmake
 set(LLVM_RELEASE_ENABLE_LTO THIN CACHE STRING "")
 set(LLVM_RELEASE_ENABLE_PGO OFF CACHE BOOL "")
-
+set(LLVM_RELEASE_ENABLE_RUNTIMES "compiler-rt;libcxx;libcxxabi;libunwind" CACHE STRING "")
+set(LLVM_RELEASE_ENABLE_PROJECTS "clang;lld;lldb;clang-tools-extra;bolt;polly;mlir;flang" CACHE STRING "")
+# Note we don't need to add install here, since it is one of the pre-defined
+# steps.
+set(LLVM_RELEASE_FINAL_STAGE_TARGETS "clang;package;check-all;check-llvm;check-clang" CACHE STRING "")
 set(CMAKE_BUILD_TYPE RELEASE CACHE STRING "")
 
-# Stage 1 Bootstrap Setup
+# Stage 1 Options
+set(LLVM_TARGETS_TO_BUILD Native CACHE STRING "")
 set(CLANG_ENABLE_BOOTSTRAP ON CACHE BOOL "")
+
+set(STAGE1_PROJECTS "clang")
+set(STAGE1_RUNTIMES "")
+
 if (LLVM_RELEASE_ENABLE_PGO)
+  list(APPEND STAGE1_PROJECTS "lld")
+  list(APPEND STAGE1_RUNTIMES "compiler-rt")
   set(CLANG_BOOTSTRAP_TARGETS
     generate-profdata
-    stage2
+    stage2-package
     stage2-clang
-    stage2-distribution
     stage2-install
-    stage2-install-distribution
-    stage2-install-distribution-toolchain
     stage2-check-all
     stage2-check-llvm
-    stage2-check-clang
-    stage2-test-suite CACHE STRING "")
-else()
-  set(CLANG_BOOTSTRAP_TARGETS
-    clang
-    check-all
-    check-llvm
-    check-clang
-    test-suite
-    stage3
-    stage3-clang
-    stage3-check-all
-    stage3-check-llvm
-    stage3-check-clang
-    stage3-install
-    stage3-test-suite CACHE STRING "")
-endif()
+    stage2-check-clang CACHE STRING "")
 
-# Stage 1 Options
-set(STAGE1_PROJECTS "clang")
-set(STAGE1_RUNTIMES "")
+  # Configuration for stage2-instrumented
+  set(BOOTSTRAP_CLANG_ENABLE_BOOTSTRAP ON CACHE STRING "")
+  # This enables the build targets for the final stage which is called stage2.
+  set(BOOTSTRAP_CLANG_BOOTSTRAP_TARGETS ${LLVM_RELEASE_FINAL_STAGE_TARGETS} CACHE STRING "")
+  set(BOOTSTRAP_LLVM_BUILD_INSTRUMENTED IR CACHE STRING "")
+  set(BOOTSTRAP_LLVM_ENABLE_RUNTIMES "compiler-rt" CACHE STRING "")
+  set(BOOTSTRAP_LLVM_ENABLE_PROJECTS "clang;lld" CACHE STRING "")
 
-if (LLVM_RELEASE_ENABLE_PGO)
-  list(APPEND STAGE1_PROJECTS "lld")
-  list(APPEND STAGE1_RUNTIMES "compiler-rt")
+else()
+  if (LLVM_RELEASE_ENABLE_LTO)
+    list(APPEND STAGE1_PROJECTS "lld")
+  endif()
+  # Any targets added here will be given the target name stage2-${target}, so
+  # if you want to run them you can just use:
+  # ninja -C $BUILDDIR stage2-${target}
+  set(CLANG_BOOTSTRAP_TARGETS ${LLVM_RELEASE_FINAL_STAGE_TARGETS} CACHE STRING "")
 endif()
 
+# Stage 1 Common Config
 set(LLVM_ENABLE_RUNTIMES ${STAGE1_RUNTIMES} CACHE STRING "")
 set(LLVM_ENABLE_PROJECTS ${STAGE1_PROJECTS} CACHE STRING "")
 
-set(LLVM_TARGETS_TO_BUILD Native CACHE STRING "")
-
-# Stage 2 Bootstrap Setup
-set(BOOTSTRAP_CLANG_ENABLE_BOOTSTRAP ON CACHE STRING "")
-set(BOOTSTRAP_CLANG_BOOTSTRAP_TARGETS
-  clang
-  check-all
-  check-llvm
-  check-clang CACHE STRING "")
-
-# Stage 2 Options
-set(STAGE2_PROJECTS "clang")
-set(STAGE2_RUNTIMES "")
-
-if (LLVM_RELEASE_ENABLE_LTO OR LLVM_RELEASE_ENABLE_PGO)
- list(APPEND STAGE2_PROJECTS "lld")
-endif()
-
-if (LLVM_RELEASE_ENABLE_PGO)
-  set(BOOTSTRAP_LLVM_BUILD_INSTRUMENTED IR CACHE STRING "")
-  list(APPEND STAGE2_RUNTIMES "compiler-rt")
-  set(BOOTSTRAP_LLVM_ENABLE_LTO ${LLVM_RELEASE_ENABLE_LTO})
-  if (LLVM_RELEASE_ENABLE_LTO)
-    set(BOOTSTRAP_LLVM_ENABLE_LLD ON CACHE BOOL "")
-  endif()
+# stage2-instrumented and Final Stage Config:
+# Options that need to be set in both the instrumented stage (if we are doing
+# a pgo build) and the final stage.
+set_instrument_and_final_stage_var(CMAKE_POSITION_INDEPENDENT_CODE "ON" STRING)
+set_instrument_and_final_stage_var(LLVM_ENABLE_LTO "${LLVM_RELEASE_ENABLE_LTO}" STRING)
+if (LLVM_RELEASE_ENABLE_LTO)
+  set_instrument_and_final_stage_var(LLVM_ENABLE_LLD "ON" BOOL)
 endif()
 
-set(BOOTSTRAP_LLVM_ENABLE_PROJECTS ${STAGE2_PROJECTS} CACHE STRING "")
-set(BOOTSTRAP_LLVM_ENABLE_RUNTIMES ${STAGE2_RUNTIMES} CACHE STRING "")
-if (NOT LLVM_RELEASE_ENABLE_PGO)
-  set(BOOTSTRAP_LLVM_TARGETS_TO_BUILD Native CACHE STRING "")
-endif()
+# Final Stage Config (stage2)
+set_final_stage_var(LLVM_ENABLE_RUNTIMES "${LLVM_RELEASE_ENABLE_RUNTIMES}" STRING)
+set_final_stage_var(LLVM_ENABLE_PROJECTS "${LLVM_RELEASE_ENABLE_PROJECTS}" STRING)
 
-# Stage 3 Options
-set(BOOTSTRAP_BOOTSTRAP_LLVM_ENABLE_RUNTIMES "compiler-rt;libcxx;libcxxabi;libunwind" CACHE STRING "")
-set(BOOTSTRAP_BOOTSTRAP_LLVM_ENABLE_PROJECTS "clang;lld;lldb;clang-tools-extra;bolt;polly;mlir;flang" CACHE STRING "")
-set(BOOTSTRAP_BOOTSTRAP_LLVM_ENABLE_LTO ${LLVM_RELEASE_ENABLE_LTO} CACHE STRING "")
-if (LLVM_RELEASE_ENABLE_LTO)
-  set(BOOTSTRAP_BOOTSTRAP_LLVM_ENABLE_LLD ON CACHE BOOL "")
-endif()
diff --git a/clang/cmake/caches/VectorEngine.cmake b/clang/cmake/caches/VectorEngine.cmake
index e3976f3206db..2f968a21cc40 100644
--- a/clang/cmake/caches/VectorEngine.cmake
+++ b/clang/cmake/caches/VectorEngine.cmake
@@ -40,6 +40,7 @@ set(RUNTIMES_x86_64-unknown-linux-gnu_COMPILER_RT_BUILD_CRT OFF CACHE BOOL "")
 set(RUNTIMES_x86_64-unknown-linux-gnu_COMPILER_RT_BUILD_SANITIZERS OFF CACHE BOOL "")
 set(RUNTIMES_x86_64-unknown-linux-gnu_COMPILER_RT_BUILD_XRAY OFF CACHE BOOL "")
 set(RUNTIMES_x86_64-unknown-linux-gnu_COMPILER_RT_BUILD_LIBFUZZER OFF CACHE BOOL "")
+set(RUNTIMES_x86_64-unknown-linux-gnu_COMPILER_RT_BUILD_CTX_PROFILE OFF CACHE BOOL "")
 set(RUNTIMES_x86_64-unknown-linux-gnu_COMPILER_RT_BUILD_PROFILE OFF CACHE BOOL "")
 set(RUNTIMES_x86_64-unknown-linux-gnu_COMPILER_RT_BUILD_MEMPROF OFF CACHE BOOL "")
 set(RUNTIMES_x86_64-unknown-linux-gnu_COMPILER_RT_BUILD_ORC OFF CACHE BOOL "")
@@ -52,6 +53,7 @@ set(RUNTIMES_ve-unknown-linux-gnu_COMPILER_RT_BUILD_SANITIZERS OFF CACHE BOOL ""
 set(RUNTIMES_ve-unknown-linux-gnu_COMPILER_RT_BUILD_XRAY OFF CACHE BOOL "")
 set(RUNTIMES_ve-unknown-linux-gnu_COMPILER_RT_BUILD_LIBFUZZER OFF CACHE BOOL "")
 set(RUNTIMES_ve-unknown-linux-gnu_COMPILER_RT_BUILD_PROFILE ON CACHE BOOL "")
+set(RUNTIMES_ve-unknown-linux-gnu_COMPILER_RT_BUILD_CTX_PROFILE OFF CACHE BOOL "")
 set(RUNTIMES_ve-unknown-linux-gnu_COMPILER_RT_BUILD_MEMPROF OFF CACHE BOOL "")
 set(RUNTIMES_ve-unknown-linux-gnu_COMPILER_RT_BUILD_ORC OFF CACHE BOOL "")
 set(RUNTIMES_ve-unknown-linux-gnu_COMPILER_RT_BUILD_GWP_ASAN OFF CACHE BOOL "")
diff --git a/clang/docs/LanguageExtensions.rst b/clang/docs/LanguageExtensions.rst
index 84fc4dee02fa..c2e90f4e7d58 100644
--- a/clang/docs/LanguageExtensions.rst
+++ b/clang/docs/LanguageExtensions.rst
@@ -711,6 +711,8 @@ even-odd element pair with indices ``i * 2`` and ``i * 2 + 1`` with
 power of 2, the vector is widened with neutral elements for the reduction
 at the end to the next power of 2.
 
+These reductions support both fixed-sized and scalable vector types.
+
 Example:
 
 .. code-block:: c++
@@ -1493,6 +1495,7 @@ Conditional ``explicit``                     __cpp_conditional_explicit       C+
 ``if consteval``                             __cpp_if_consteval               C++23         C++20
 ``static operator()``                        __cpp_static_call_operator       C++23         C++03
 Attributes on Lambda-Expressions                                              C++23         C++11
+Attributes on Structured Bindings            __cpp_structured_bindings        C++26         C++03
 ``= delete ("should have a reason");``       __cpp_deleted_function           C++26         C++03
 -------------------------------------------- -------------------------------- ------------- -------------
 Designated initializers (N494)                                                C99           C89
@@ -2928,7 +2931,7 @@ Query for this feature with ``__has_builtin(__builtin_dump_struct)``
 ``__builtin_shufflevector`` is used to express generic vector
 permutation/shuffle/swizzle operations.  This builtin is also very important
 for the implementation of various target-specific header files like
-``<xmmintrin.h>``.
+``<xmmintrin.h>``. This builtin can be used within constant expressions.
 
 **Syntax**:
 
@@ -2955,7 +2958,7 @@ for the implementation of various target-specific header files like
   // Concatenate every other element of 8-element vectors V1 and V2.
   __builtin_shufflevector(V1, V2, 0, 2, 4, 6, 8, 10, 12, 14)
 
-  // Shuffle v1 with some elements being undefined
+  // Shuffle v1 with some elements being undefined. Not allowed in constexpr.
   __builtin_shufflevector(v1, v1, 3, -1, 1, -1)
 
 **Description**:
@@ -2968,6 +2971,7 @@ starting with the first vector, continuing into the second vector.  Thus, if
 ``vec1`` is a 4-element vector, index 5 would refer to the second element of
 ``vec2``. An index of -1 can be used to indicate that the corresponding element
 in the returned vector is a don't care and can be optimized by the backend.
+Values of -1 are not supported in constant expressions.
 
 The result of ``__builtin_shufflevector`` is a vector with the same element
 type as ``vec1``/``vec2`` but that has an element count equal to the number of
@@ -2982,7 +2986,8 @@ Query for this feature with ``__has_builtin(__builtin_shufflevector)``.
 
 ``__builtin_convertvector`` is used to express generic vector
 type-conversion operations. The input vector and the output vector
-type must have the same number of elements.
+type must have the same number of elements. This builtin can be used within
+constant expressions.
 
 **Syntax**:
 
@@ -5572,3 +5577,25 @@ but the expression has no runtime effects.
 Type- and value-dependent expressions are not supported yet.
 
 This facility is designed to aid with testing name lookup machinery.
+
+Predefined Macros
+=================
+
+`__GCC_DESTRUCTIVE_SIZE` and `__GCC_CONSTRUCTIVE_SIZE`
+------------------------------------------------------
+Specify the mimum offset between two objects to avoid false sharing and the
+maximum size of contiguous memory to promote true sharing, respectively. These
+macros are predefined in all C and C++ language modes, but can be redefined on
+the command line with ``-D`` to specify different values as needed or can be
+undefined on the command line with ``-U`` to disable support for the feature.
+
+**Note: the values the macros expand to are not guaranteed to be stable. They
+are are affected by architectures and CPU tuning flags, can change between
+releases of Clang and will not match the values defined by other compilers such
+as GCC.**
+
+Compiling different TUs depending on these flags (including use of
+``std::hardware_constructive_interference`` or
+``std::hardware_destructive_interference``)  with different compilers, macro
+definitions, or architecture flags will lead to ODR violations and should be
+avoided.
\ No newline at end of file
diff --git a/clang/docs/LibTooling.rst b/clang/docs/LibTooling.rst
index df50dcebf9b8..87d84321ab28 100644
--- a/clang/docs/LibTooling.rst
+++ b/clang/docs/LibTooling.rst
@@ -63,15 +63,22 @@ and automatic location of the compilation database using source files paths.
   #include "llvm/Support/CommandLine.h"
 
   using namespace clang::tooling;
+  using namespace llvm;
 
   // Apply a custom category to all command-line options so that they are the
   // only ones displayed.
-  static llvm::cl::OptionCategory MyToolCategory("my-tool options");
+  static cl::OptionCategory MyToolCategory("my-tool options");
 
   int main(int argc, const char **argv) {
-    // CommonOptionsParser constructor will parse arguments and create a
-    // CompilationDatabase.  In case of error it will terminate the program.
-    CommonOptionsParser OptionsParser(argc, argv, MyToolCategory);
+    // CommonOptionsParser::create will parse arguments and create a
+    // CompilationDatabase.
+    auto ExpectedParser = CommonOptionsParser::create(argc, argv, MyToolCategory);
+    if (!ExpectedParser) {
+      // Fail gracefully for unsupported options.
+      llvm::errs() << ExpectedParser.takeError();
+      return 1;
+    }
+    CommonOptionsParser& OptionsParser = ExpectedParser.get();
 
     // Use OptionsParser.getCompilations() and OptionsParser.getSourcePathList()
     // to retrieve CompilationDatabase and the list of input file paths.
@@ -133,7 +140,12 @@ version of this example tool is also checked into the clang tree at
   static cl::extrahelp MoreHelp("\nMore help text...\n");
 
   int main(int argc, const char **argv) {
-    CommonOptionsParser OptionsParser(argc, argv, MyToolCategory);
+    auto ExpectedParser = CommonOptionsParser::create(argc, argv, MyToolCategory);
+    if (!ExpectedParser) {
+      llvm::errs() << ExpectedParser.takeError();
+      return 1;
+    }
+    CommonOptionsParser& OptionsParser = ExpectedParser.get();
     ClangTool Tool(OptionsParser.getCompilations(),
                    OptionsParser.getSourcePathList());
     return Tool.run(newFrontendActionFactory<clang::SyntaxOnlyAction>().get());
diff --git a/clang/docs/OpenMPSupport.rst b/clang/docs/OpenMPSupport.rst
index f8146bc365e8..5e63b2c0f0be 100644
--- a/clang/docs/OpenMPSupport.rst
+++ b/clang/docs/OpenMPSupport.rst
@@ -310,7 +310,9 @@ implementation.
 +------------------------------+--------------------------------------------------------------+--------------------------+-----------------------------------------------------------------------+
 | misc                         | dispatch construct and function variant argument adjustment  | :part:`worked on`        | D99537, D99679                                                        |
 +------------------------------+--------------------------------------------------------------+--------------------------+-----------------------------------------------------------------------+
-| misc                         | assume and assumes directives                                | :part:`worked on`        |                                                                       |
+| misc                         | assumes directives                                           | :part:`worked on`        |                                                                       |
++------------------------------+--------------------------------------------------------------+--------------------------+-----------------------------------------------------------------------+
+| misc                         | assume directive                                             | :part:`worked on`        |                                                                       |
 +------------------------------+--------------------------------------------------------------+--------------------------+-----------------------------------------------------------------------+
 | misc                         | nothing directive                                            | :good:`done`             | D123286                                                               |
 +------------------------------+--------------------------------------------------------------+--------------------------+-----------------------------------------------------------------------+
diff --git a/clang/docs/ReleaseNotes.rst b/clang/docs/ReleaseNotes.rst
index f339fab6e842..2d0e663d93e4 100644
--- a/clang/docs/ReleaseNotes.rst
+++ b/clang/docs/ReleaseNotes.rst
@@ -90,6 +90,18 @@ C++ Language Changes
 --------------------
 - Implemented ``_BitInt`` literal suffixes ``__wb`` or ``__WB`` as a Clang extension with ``unsigned`` modifiers also allowed. (#GH85223).
 
+C++17 Feature Support
+^^^^^^^^^^^^^^^^^^^^^
+- Clang now exposes ``__GCC_DESTRUCTIVE_SIZE`` and ``__GCC_CONSTRUCTIVE_SIZE``
+  predefined macros to support standard library implementations of
+  ``std::hardware_destructive_interference_size`` and
+  ``std::hardware_constructive_interference_size``, respectively. These macros
+  are predefined in all C and C++ language modes. The values the macros
+  expand to are not stable between releases of Clang and do not need to match
+  the values produced by GCC, so these macros should not be used from header
+  files because they may not be stable across multiple TUs (the values may vary
+  based on compiler version as well as CPU tuning). #GH60174
+
 C++20 Feature Support
 ^^^^^^^^^^^^^^^^^^^^^
 
@@ -131,6 +143,9 @@ C++2c Feature Support
 
 - Implemented `P2573R2: = delete("should have a reason"); <https://wg21.link/P2573R2>`_
 
+- Implemented `P0609R3: Attributes for Structured Bindings <https://wg21.link/P0609R3>`_
+
+- Implemented `P2748R5 Disallow Binding a Returned Glvalue to a Temporary <https://wg21.link/P2748R5>`_.
 
 Resolutions to C++ Defect Reports
 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
@@ -207,6 +222,20 @@ Non-comprehensive list of changes in this release
 - ``__typeof_unqual__`` is available in all C modes as an extension, which behaves
   like ``typeof_unqual`` from C23, similar to ``__typeof__`` and ``typeof``.
 
+- ``__builtin_reduce_{add|mul|xor|or|and|min|max}`` builtins now support scalable vectors.
+
+* Shared libraries linked with either the ``-ffast-math``, ``-Ofast``, or
+  ``-funsafe-math-optimizations`` flags will no longer enable flush-to-zero
+  floating-point mode by default. This decision can be overridden with use of
+  ``-mdaz-ftz``. This behavior now matches GCC's behavior.
+  (`#57589 <https://github.com/llvm/llvm-project/issues/57589>`_)
+
+* ``-fdenormal-fp-math=preserve-sign`` is no longer implied by ``-ffast-math``
+  on x86 systems.
+
+- Builtins ``__builtin_shufflevector()`` and ``__builtin_convertvector()`` may
+  now be used within constant expressions.
+
 New Compiler Flags
 ------------------
 - ``-fsanitize=implicit-bitfield-conversion`` checks implicit truncation and
@@ -415,6 +444,9 @@ Bug Fixes in This Version
   operator.
   Fixes (#GH83267).
 
+- Fix crash on ill-formed partial specialization with CRTP.
+  Fixes (#GH89374).
+
 - Clang now correctly generates overloads for bit-precise integer types for
   builtin operators in C++. Fixes #GH82998.
 
@@ -433,6 +465,10 @@ Bug Fixes in This Version
 
 - Fixed an assertion failure on invalid InitListExpr in C89 mode (#GH88008).
 
+- Fixed missing destructor calls when we branch from middle of an expression.
+  This could happen through a branch in stmt-expr or in an expression containing a coroutine
+  suspension. Fixes (#GH63818) (#GH88478).
+
 - Clang will no longer diagnose an erroneous non-dependent ``switch`` condition
   during instantiation, and instead will only diagnose it once, during checking
   of the function template.
@@ -560,6 +596,11 @@ Bug Fixes to C++ Support
 - Fix the Itanium mangling of lambdas defined in a member of a local class (#GH88906)
 - Fixed a crash when trying to evaluate a user-defined ``static_assert`` message whose ``size()``
   function returns a large or negative value. Fixes (#GH89407).
+- Fixed a use-after-free bug in parsing of type constraints with default arguments that involve lambdas. (#GH67235)
+- Fixed bug in which the body of a consteval lambda within a template was not parsed as within an
+  immediate function context.
+- Fix CTAD for ``std::initializer_list``. This allows ``std::initializer_list{1, 2, 3}`` to be deduced as
+  ``std::initializer_list<int>`` as intended.
 
 Bug Fixes to AST Handling
 ^^^^^^^^^^^^^^^^^^^^^^^^^
@@ -612,6 +653,9 @@ Arm and AArch64 Support
     * Arm Cortex-A78AE (cortex-a78ae).
     * Arm Cortex-A520AE (cortex-a520ae).
     * Arm Cortex-A720AE (cortex-a720ae).
+    * Arm Neoverse-N3 (neoverse-n3).
+    * Arm Neoverse-V3 (neoverse-v3).
+    * Arm Neoverse-V3AE (neoverse-v3ae).
 
 Android Support
 ^^^^^^^^^^^^^^^
@@ -706,6 +750,9 @@ clang-format
 libclang
 --------
 
+- ``clang_getSpellingLocation`` now correctly resolves macro expansions; that
+  is, it returns the spelling location instead of the expansion location.
+
 Static Analyzer
 ---------------
 
diff --git a/clang/docs/UsersManual.rst b/clang/docs/UsersManual.rst
index c464bc3a69ad..a7fc7ec87a16 100644
--- a/clang/docs/UsersManual.rst
+++ b/clang/docs/UsersManual.rst
@@ -1506,7 +1506,8 @@ floating point semantic models: precise (the default), strict, and fast.
 
    * ``-ffp-contract=fast``
 
-   Note: ``-ffast-math`` causes ``crtfastmath.o`` to be linked with code. See
+   Note: ``-ffast-math`` causes ``crtfastmath.o`` to be linked with code unless
+   ``-shared`` or ``-mno-daz-ftz`` is present. See
    :ref:`crtfastmath.o` for more details.
 
 .. option:: -fno-fast-math
@@ -1560,7 +1561,8 @@ floating point semantic models: precise (the default), strict, and fast.
      ``-ffp-contract``.
 
    Note: ``-fno-fast-math`` implies ``-fdenormal-fp-math=ieee``.
-   ``-fno-fast-math`` causes ``crtfastmath.o`` to not be linked with code.
+   ``-fno-fast-math`` causes ``crtfastmath.o`` to not be linked with code
+   unless ``-mdaz-ftz`` is present.
 
 .. option:: -fdenormal-fp-math=<value>
 
@@ -1938,10 +1940,13 @@ by using ``#pragma STDC FENV_ROUND`` with a value other than ``FE_DYNAMIC``.
 
 A note about ``crtfastmath.o``
 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
-``-ffast-math`` and ``-funsafe-math-optimizations`` cause ``crtfastmath.o`` to be
-automatically linked,  which adds a static constructor that sets the FTZ/DAZ
+``-ffast-math`` and ``-funsafe-math-optimizations`` without the ``-shared``
+option cause ``crtfastmath.o`` to be
+automatically linked, which adds a static constructor that sets the FTZ/DAZ
 bits in MXCSR, affecting not only the current compilation unit but all static
-and shared libraries included in the program.
+and shared libraries included in the program. This decision can be overridden
+by using either the flag ``-mdaz-ftz`` or ``-mno-daz-ftz`` to respectively
+link or not link ``crtfastmath.o``.
 
 .. _FLT_EVAL_METHOD:
 
@@ -2314,6 +2319,8 @@ are listed below.
    on ELF targets when using the integrated assembler. This flag currently
    only has an effect on ELF targets.
 
+.. _funique_internal_linkage_names:
+
 .. option:: -f[no]-unique-internal-linkage-names
 
    Controls whether Clang emits a unique (best-effort) symbol name for internal
@@ -2443,27 +2450,41 @@ usual build cycle when using sample profilers for optimization:
    usual build flags that you always build your application with. The only
    requirement is that DWARF debug info including source line information is
    generated. This DWARF information is important for the profiler to be able
-   to map instructions back to source line locations.
+   to map instructions back to source line locations. The usefulness of this
+   DWARF information can be improved with the ``-fdebug-info-for-profiling``
+   and ``-funique-internal-linkage-names`` options.
 
-   On Linux, ``-g`` or just ``-gline-tables-only`` is sufficient:
+   On Linux:
 
    .. code-block:: console
 
-     $ clang++ -O2 -gline-tables-only code.cc -o code
+     $ clang++ -O2 -gline-tables-only \
+       -fdebug-info-for-profiling -funique-internal-linkage-names \
+       code.cc -o code
 
    While MSVC-style targets default to CodeView debug information, DWARF debug
    information is required to generate source-level LLVM profiles. Use
    ``-gdwarf`` to include DWARF debug information:
 
-   .. code-block:: console
+   .. code-block:: winbatch
+
+     > clang-cl /O2 -gdwarf -gline-tables-only ^
+       /clang:-fdebug-info-for-profiling /clang:-funique-internal-linkage-names ^
+       code.cc /Fe:code /fuse-ld=lld /link /debug:dwarf
+
+.. note::
 
-     $ clang-cl -O2 -gdwarf -gline-tables-only coff-profile.cpp -fuse-ld=lld -link -debug:dwarf
+   :ref:`-funique-internal-linkage-names <funique_internal_linkage_names>`
+   generates unique names based on given command-line source file paths. If
+   your build system uses absolute source paths and these paths may change
+   between steps 1 and 4, then the uniqued function names may change and result
+   in unused profile data. Consider omitting this option in such cases.
 
 2. Run the executable under a sampling profiler. The specific profiler
    you use does not really matter, as long as its output can be converted
    into the format that the LLVM optimizer understands.
 
-   Two such profilers are the the Linux Perf profiler
+   Two such profilers are the Linux Perf profiler
    (https://perf.wiki.kernel.org/) and Intel's Sampling Enabling Product (SEP),
    available as part of `Intel VTune
    <https://software.intel.com/content/www/us/en/develop/tools/oneapi/components/vtune-profiler.html>`_.
@@ -2477,7 +2498,9 @@ usual build cycle when using sample profilers for optimization:
 
    .. code-block:: console
 
-     $ perf record -b ./code
+     $ perf record -b -e BR_INST_RETIRED.NEAR_TAKEN:uppp ./code
+
+   If the event above is unavailable, ``branches:u`` is probably next-best.
 
    Note the use of the ``-b`` flag. This tells Perf to use the Last Branch
    Record (LBR) to record call chains. While this is not strictly required,
@@ -2527,21 +2550,42 @@ usual build cycle when using sample profilers for optimization:
    that executes faster than the original one. Note that you are not
    required to build the code with the exact same arguments that you
    used in the first step. The only requirement is that you build the code
-   with ``-gline-tables-only`` and ``-fprofile-sample-use``.
+   with the same debug info options and ``-fprofile-sample-use``.
+
+   On Linux:
 
    .. code-block:: console
 
-     $ clang++ -O2 -gline-tables-only -fprofile-sample-use=code.prof code.cc -o code
+     $ clang++ -O2 -gline-tables-only \
+       -fdebug-info-for-profiling -funique-internal-linkage-names \
+       -fprofile-sample-use=code.prof code.cc -o code
 
-  [OPTIONAL] Sampling-based profiles can have inaccuracies or missing block/
-  edge counters. The profile inference algorithm (profi) can be used to infer
-  missing blocks and edge counts, and improve the quality of profile data.
-  Enable it with ``-fsample-profile-use-profi``.
+   On Windows:
 
-  .. code-block:: console
+   .. code-block:: winbatch
 
-    $ clang++ -O2 -gline-tables-only -fprofile-sample-use=code.prof \
-      -fsample-profile-use-profi code.cc -o code
+     > clang-cl /O2 -gdwarf -gline-tables-only ^
+       /clang:-fdebug-info-for-profiling /clang:-funique-internal-linkage-names ^
+       /fprofile-sample-use=code.prof code.cc /Fe:code /fuse-ld=lld /link /debug:dwarf
+
+   [OPTIONAL] Sampling-based profiles can have inaccuracies or missing block/
+   edge counters. The profile inference algorithm (profi) can be used to infer
+   missing blocks and edge counts, and improve the quality of profile data.
+   Enable it with ``-fsample-profile-use-profi``. For example, on Linux:
+
+   .. code-block:: console
+
+     $ clang++ -fsample-profile-use-profi -O2 -gline-tables-only \
+       -fdebug-info-for-profiling -funique-internal-linkage-names \
+       -fprofile-sample-use=code.prof code.cc -o code
+
+   On Windows:
+
+   .. code-block:: winbatch
+
+     > clang-cl /clang:-fsample-profile-use-profi /O2 -gdwarf -gline-tables-only ^
+       /clang:-fdebug-info-for-profiling /clang:-funique-internal-linkage-names ^
+       /fprofile-sample-use=code.prof code.cc /Fe:code /fuse-ld=lld /link /debug:dwarf
 
 Sample Profile Formats
 """"""""""""""""""""""
@@ -4921,6 +4965,9 @@ directory. Using the example installation above, this would mean passing
 If the user links the program with the ``clang`` or ``clang-cl`` drivers, the
 driver will pass this flag for them.
 
+The auto-linking can be disabled with -fno-rtlib-defaultlib. If that flag is
+used, pass the complete flag to required libraries as described for ASan below.
+
 If the linker cannot find the appropriate library, it will emit an error like
 this::
 
diff --git a/clang/include/clang-c/Index.h b/clang/include/clang-c/Index.h
index 7a8bd985a91f..365b607c7411 100644
--- a/clang/include/clang-c/Index.h
+++ b/clang/include/clang-c/Index.h
@@ -1644,8 +1644,9 @@ enum CXCursorKind {
   CXCursor_ObjCSelfExpr = 146,
 
   /** OpenMP 5.0 [2.1.5, Array Section].
+   * OpenACC 3.3 [2.7.1, Data Specification for Data Clauses (Sub Arrays)]
    */
-  CXCursor_OMPArraySectionExpr = 147,
+  CXCursor_ArraySectionExpr = 147,
 
   /** Represents an @available(...) check.
    */
diff --git a/clang/include/clang/APINotes/Types.h b/clang/include/clang/APINotes/Types.h
index 93bb045d6a66..026a4a431e73 100644
--- a/clang/include/clang/APINotes/Types.h
+++ b/clang/include/clang/APINotes/Types.h
@@ -675,6 +675,11 @@ class TagInfo : public CommonTypeInfo {
   LLVM_PREFERRED_TYPE(bool)
   unsigned IsFlagEnum : 1;
 
+  LLVM_PREFERRED_TYPE(bool)
+  unsigned SwiftCopyableSpecified : 1;
+  LLVM_PREFERRED_TYPE(bool)
+  unsigned SwiftCopyable : 1;
+
 public:
   std::optional<std::string> SwiftImportAs;
   std::optional<std::string> SwiftRetainOp;
@@ -682,7 +687,9 @@ public:
 
   std::optional<EnumExtensibilityKind> EnumExtensibility;
 
-  TagInfo() : HasFlagEnum(0), IsFlagEnum(0) {}
+  TagInfo()
+      : HasFlagEnum(0), IsFlagEnum(0), SwiftCopyableSpecified(false),
+        SwiftCopyable(false) {}
 
   std::optional<bool> isFlagEnum() const {
     if (HasFlagEnum)
@@ -694,6 +701,15 @@ public:
     IsFlagEnum = Value.value_or(false);
   }
 
+  std::optional<bool> isSwiftCopyable() const {
+    return SwiftCopyableSpecified ? std::optional<bool>(SwiftCopyable)
+                                  : std::nullopt;
+  }
+  void setSwiftCopyable(std::optional<bool> Value) {
+    SwiftCopyableSpecified = Value.has_value();
+    SwiftCopyable = Value.value_or(false);
+  }
+
   TagInfo &operator|=(const TagInfo &RHS) {
     static_cast<CommonTypeInfo &>(*this) |= RHS;
 
@@ -710,6 +726,9 @@ public:
     if (!EnumExtensibility)
       EnumExtensibility = RHS.EnumExtensibility;
 
+    if (!SwiftCopyableSpecified)
+      setSwiftCopyable(RHS.isSwiftCopyable());
+
     return *this;
   }
 
@@ -724,6 +743,7 @@ inline bool operator==(const TagInfo &LHS, const TagInfo &RHS) {
          LHS.SwiftRetainOp == RHS.SwiftRetainOp &&
          LHS.SwiftReleaseOp == RHS.SwiftReleaseOp &&
          LHS.isFlagEnum() == RHS.isFlagEnum() &&
+         LHS.isSwiftCopyable() == RHS.isSwiftCopyable() &&
          LHS.EnumExtensibility == RHS.EnumExtensibility;
 }
 
diff --git a/clang/include/clang/AST/ASTContext.h b/clang/include/clang/AST/ASTContext.h
index 28f8d67811f0..6dbd06251dda 100644
--- a/clang/include/clang/AST/ASTContext.h
+++ b/clang/include/clang/AST/ASTContext.h
@@ -455,7 +455,7 @@ class ASTContext : public RefCountedBase<ASTContext> {
   /// initialization of another module).
   struct PerModuleInitializers {
     llvm::SmallVector<Decl*, 4> Initializers;
-    llvm::SmallVector<uint32_t, 4> LazyInitializers;
+    llvm::SmallVector<GlobalDeclID, 4> LazyInitializers;
 
     void resolve(ASTContext &Ctx);
   };
@@ -1059,7 +1059,7 @@ public:
   /// or an ImportDecl nominating another module that has initializers.
   void addModuleInitializer(Module *M, Decl *Init);
 
-  void addLazyModuleInitializers(Module *M, ArrayRef<uint32_t> IDs);
+  void addLazyModuleInitializers(Module *M, ArrayRef<GlobalDeclID> IDs);
 
   /// Get the initializations to perform when importing a module, if any.
   ArrayRef<Decl*> getModuleInitializers(Module *M);
@@ -1127,7 +1127,8 @@ public:
   CanQualType OCLSamplerTy, OCLEventTy, OCLClkEventTy;
   CanQualType OCLQueueTy, OCLReserveIDTy;
   CanQualType IncompleteMatrixIdxTy;
-  CanQualType OMPArraySectionTy, OMPArrayShapingTy, OMPIteratorTy;
+  CanQualType ArraySectionTy;
+  CanQualType OMPArrayShapingTy, OMPIteratorTy;
 #define EXT_OPAQUE_TYPE(ExtType, Id, Ext) \
   CanQualType Id##Ty;
 #include "clang/Basic/OpenCLExtensionTypes.def"
@@ -2196,6 +2197,16 @@ public:
     return getQualifiedType(type.getUnqualifiedType(), Qs);
   }
 
+  /// \brief Return a type with the given __ptrauth qualifier.
+  QualType getPointerAuthType(QualType Ty, PointerAuthQualifier PointerAuth) {
+    assert(!Ty.getPointerAuth());
+    assert(PointerAuth);
+
+    Qualifiers Qs;
+    Qs.setPointerAuth(PointerAuth);
+    return getQualifiedType(Ty, Qs);
+  }
+
   unsigned char getFixedPointScale(QualType Ty) const;
   unsigned char getFixedPointIBits(QualType Ty) const;
   llvm::FixedPointSemantics getFixedPointSemantics(QualType Ty) const;
diff --git a/clang/include/clang/AST/ASTNodeTraverser.h b/clang/include/clang/AST/ASTNodeTraverser.h
index 216dc9eef08b..bf7c204e4ad7 100644
--- a/clang/include/clang/AST/ASTNodeTraverser.h
+++ b/clang/include/clang/AST/ASTNodeTraverser.h
@@ -844,6 +844,12 @@ public:
     }
   }
 
+  void VisitUnresolvedLookupExpr(const UnresolvedLookupExpr *E) {
+    if (E->hasExplicitTemplateArgs())
+      for (auto Arg : E->template_arguments())
+        Visit(Arg.getArgument());
+  }
+
   void VisitRequiresExpr(const RequiresExpr *E) {
     for (auto *D : E->getLocalParameters())
       Visit(D);
diff --git a/clang/include/clang/AST/AbstractBasicReader.h b/clang/include/clang/AST/AbstractBasicReader.h
index 1f2797cc7014..ab036f1d445a 100644
--- a/clang/include/clang/AST/AbstractBasicReader.h
+++ b/clang/include/clang/AST/AbstractBasicReader.h
@@ -213,9 +213,9 @@ public:
   }
 
   Qualifiers readQualifiers() {
-    static_assert(sizeof(Qualifiers().getAsOpaqueValue()) <= sizeof(uint32_t),
+    static_assert(sizeof(Qualifiers().getAsOpaqueValue()) <= sizeof(uint64_t),
                   "update this if the value size changes");
-    uint32_t value = asImpl().readUInt32();
+    uint64_t value = asImpl().readUInt64();
     return Qualifiers::fromOpaqueValue(value);
   }
 
diff --git a/clang/include/clang/AST/AbstractBasicWriter.h b/clang/include/clang/AST/AbstractBasicWriter.h
index 07afa388de2c..8e42fcaad1d3 100644
--- a/clang/include/clang/AST/AbstractBasicWriter.h
+++ b/clang/include/clang/AST/AbstractBasicWriter.h
@@ -196,9 +196,9 @@ public:
   }
 
   void writeQualifiers(Qualifiers value) {
-    static_assert(sizeof(value.getAsOpaqueValue()) <= sizeof(uint32_t),
+    static_assert(sizeof(value.getAsOpaqueValue()) <= sizeof(uint64_t),
                   "update this if the value size changes");
-    asImpl().writeUInt32(value.getAsOpaqueValue());
+    asImpl().writeUInt64(value.getAsOpaqueValue());
   }
 
   void writeExceptionSpecInfo(
diff --git a/clang/include/clang/AST/BuiltinTypes.def b/clang/include/clang/AST/BuiltinTypes.def
index c04f6f6f1271..0a36fdc5d9c0 100644
--- a/clang/include/clang/AST/BuiltinTypes.def
+++ b/clang/include/clang/AST/BuiltinTypes.def
@@ -320,7 +320,7 @@ PLACEHOLDER_TYPE(ARCUnbridgedCast, ARCUnbridgedCastTy)
 PLACEHOLDER_TYPE(IncompleteMatrixIdx, IncompleteMatrixIdxTy)
 
 // A placeholder type for OpenMP array sections.
-PLACEHOLDER_TYPE(OMPArraySection, OMPArraySectionTy)
+PLACEHOLDER_TYPE(ArraySection, ArraySectionTy)
 
 // A placeholder type for OpenMP array shaping operation.
 PLACEHOLDER_TYPE(OMPArrayShaping, OMPArrayShapingTy)
diff --git a/clang/include/clang/AST/ComputeDependence.h b/clang/include/clang/AST/ComputeDependence.h
index 7abf9141237d..6d3a51c379f9 100644
--- a/clang/include/clang/AST/ComputeDependence.h
+++ b/clang/include/clang/AST/ComputeDependence.h
@@ -94,7 +94,7 @@ class DesignatedInitExpr;
 class ParenListExpr;
 class PseudoObjectExpr;
 class AtomicExpr;
-class OMPArraySectionExpr;
+class ArraySectionExpr;
 class OMPArrayShapingExpr;
 class OMPIteratorExpr;
 class ObjCArrayLiteral;
@@ -189,7 +189,7 @@ ExprDependence computeDependence(ParenListExpr *E);
 ExprDependence computeDependence(PseudoObjectExpr *E);
 ExprDependence computeDependence(AtomicExpr *E);
 
-ExprDependence computeDependence(OMPArraySectionExpr *E);
+ExprDependence computeDependence(ArraySectionExpr *E);
 ExprDependence computeDependence(OMPArrayShapingExpr *E);
 ExprDependence computeDependence(OMPIteratorExpr *E);
 
diff --git a/clang/include/clang/AST/Decl.h b/clang/include/clang/AST/Decl.h
index 8b121896d66d..a53c27a99a8c 100644
--- a/clang/include/clang/AST/Decl.h
+++ b/clang/include/clang/AST/Decl.h
@@ -157,7 +157,7 @@ public:
                                    SourceLocation CommentLoc,
                                    PragmaMSCommentKind CommentKind,
                                    StringRef Arg);
-  static PragmaCommentDecl *CreateDeserialized(ASTContext &C, DeclID ID,
+  static PragmaCommentDecl *CreateDeserialized(ASTContext &C, GlobalDeclID ID,
                                                unsigned ArgSize);
 
   PragmaMSCommentKind getCommentKind() const { return CommentKind; }
@@ -192,7 +192,7 @@ public:
                                           SourceLocation Loc, StringRef Name,
                                           StringRef Value);
   static PragmaDetectMismatchDecl *
-  CreateDeserialized(ASTContext &C, DeclID ID, unsigned NameValueSize);
+  CreateDeserialized(ASTContext &C, GlobalDeclID ID, unsigned NameValueSize);
 
   StringRef getName() const { return getTrailingObjects<char>(); }
   StringRef getValue() const { return getTrailingObjects<char>() + ValueStart; }
@@ -518,7 +518,7 @@ public:
   static LabelDecl *Create(ASTContext &C, DeclContext *DC,
                            SourceLocation IdentL, IdentifierInfo *II,
                            SourceLocation GnuLabelL);
-  static LabelDecl *CreateDeserialized(ASTContext &C, DeclID ID);
+  static LabelDecl *CreateDeserialized(ASTContext &C, GlobalDeclID ID);
 
   LabelStmt *getStmt() const { return TheStmt; }
   void setStmt(LabelStmt *T) { TheStmt = T; }
@@ -581,7 +581,7 @@ public:
                                IdentifierInfo *Id, NamespaceDecl *PrevDecl,
                                bool Nested);
 
-  static NamespaceDecl *CreateDeserialized(ASTContext &C, DeclID ID);
+  static NamespaceDecl *CreateDeserialized(ASTContext &C, GlobalDeclID ID);
 
   using redecl_range = redeclarable_base::redecl_range;
   using redecl_iterator = redeclarable_base::redecl_iterator;
@@ -1146,7 +1146,7 @@ public:
                          const IdentifierInfo *Id, QualType T,
                          TypeSourceInfo *TInfo, StorageClass S);
 
-  static VarDecl *CreateDeserialized(ASTContext &C, DeclID ID);
+  static VarDecl *CreateDeserialized(ASTContext &C, GlobalDeclID ID);
 
   SourceRange getSourceRange() const override LLVM_READONLY;
 
@@ -1728,7 +1728,7 @@ public:
   static ImplicitParamDecl *Create(ASTContext &C, QualType T,
                                    ImplicitParamKind ParamKind);
 
-  static ImplicitParamDecl *CreateDeserialized(ASTContext &C, DeclID ID);
+  static ImplicitParamDecl *CreateDeserialized(ASTContext &C, GlobalDeclID ID);
 
   ImplicitParamDecl(ASTContext &C, DeclContext *DC, SourceLocation IdLoc,
                     const IdentifierInfo *Id, QualType Type,
@@ -1782,7 +1782,7 @@ public:
                              TypeSourceInfo *TInfo, StorageClass S,
                              Expr *DefArg);
 
-  static ParmVarDecl *CreateDeserialized(ASTContext &C, DeclID ID);
+  static ParmVarDecl *CreateDeserialized(ASTContext &C, GlobalDeclID ID);
 
   SourceRange getSourceRange() const override LLVM_READONLY;
 
@@ -2178,7 +2178,7 @@ public:
          bool hasWrittenPrototype, ConstexprSpecKind ConstexprKind,
          Expr *TrailingRequiresClause);
 
-  static FunctionDecl *CreateDeserialized(ASTContext &C, DeclID ID);
+  static FunctionDecl *CreateDeserialized(ASTContext &C, GlobalDeclID ID);
 
   DeclarationNameInfo getNameInfo() const {
     return DeclarationNameInfo(getDeclName(), getLocation(), DNLoc);
@@ -3136,7 +3136,7 @@ public:
                            TypeSourceInfo *TInfo, Expr *BW, bool Mutable,
                            InClassInitStyle InitStyle);
 
-  static FieldDecl *CreateDeserialized(ASTContext &C, DeclID ID);
+  static FieldDecl *CreateDeserialized(ASTContext &C, GlobalDeclID ID);
 
   /// Returns the index of this field within its record,
   /// as appropriate for passing to ASTRecordLayout::getFieldOffset.
@@ -3311,7 +3311,7 @@ public:
                                   SourceLocation L, IdentifierInfo *Id,
                                   QualType T, Expr *E,
                                   const llvm::APSInt &V);
-  static EnumConstantDecl *CreateDeserialized(ASTContext &C, DeclID ID);
+  static EnumConstantDecl *CreateDeserialized(ASTContext &C, GlobalDeclID ID);
 
   const Expr *getInitExpr() const { return (const Expr*) Init; }
   Expr *getInitExpr() { return (Expr*) Init; }
@@ -3357,7 +3357,7 @@ public:
                                    QualType T,
                                    llvm::MutableArrayRef<NamedDecl *> CH);
 
-  static IndirectFieldDecl *CreateDeserialized(ASTContext &C, DeclID ID);
+  static IndirectFieldDecl *CreateDeserialized(ASTContext &C, GlobalDeclID ID);
 
   using chain_iterator = ArrayRef<NamedDecl *>::const_iterator;
 
@@ -3542,7 +3542,7 @@ public:
   static TypedefDecl *Create(ASTContext &C, DeclContext *DC,
                              SourceLocation StartLoc, SourceLocation IdLoc,
                              const IdentifierInfo *Id, TypeSourceInfo *TInfo);
-  static TypedefDecl *CreateDeserialized(ASTContext &C, DeclID ID);
+  static TypedefDecl *CreateDeserialized(ASTContext &C, GlobalDeclID ID);
 
   SourceRange getSourceRange() const override LLVM_READONLY;
 
@@ -3567,7 +3567,7 @@ public:
   static TypeAliasDecl *Create(ASTContext &C, DeclContext *DC,
                                SourceLocation StartLoc, SourceLocation IdLoc,
                                const IdentifierInfo *Id, TypeSourceInfo *TInfo);
-  static TypeAliasDecl *CreateDeserialized(ASTContext &C, DeclID ID);
+  static TypeAliasDecl *CreateDeserialized(ASTContext &C, GlobalDeclID ID);
 
   SourceRange getSourceRange() const override LLVM_READONLY;
 
@@ -3977,7 +3977,7 @@ public:
                           IdentifierInfo *Id, EnumDecl *PrevDecl,
                           bool IsScoped, bool IsScopedUsingClassTag,
                           bool IsFixed);
-  static EnumDecl *CreateDeserialized(ASTContext &C, DeclID ID);
+  static EnumDecl *CreateDeserialized(ASTContext &C, GlobalDeclID ID);
 
   /// Overrides to provide correct range when there's an enum-base specifier
   /// with forward declarations.
@@ -4182,7 +4182,7 @@ public:
   static RecordDecl *Create(const ASTContext &C, TagKind TK, DeclContext *DC,
                             SourceLocation StartLoc, SourceLocation IdLoc,
                             IdentifierInfo *Id, RecordDecl* PrevDecl = nullptr);
-  static RecordDecl *CreateDeserialized(const ASTContext &C, DeclID ID);
+  static RecordDecl *CreateDeserialized(const ASTContext &C, GlobalDeclID ID);
 
   RecordDecl *getPreviousDecl() {
     return cast_or_null<RecordDecl>(
@@ -4433,7 +4433,7 @@ public:
                                   StringLiteral *Str, SourceLocation AsmLoc,
                                   SourceLocation RParenLoc);
 
-  static FileScopeAsmDecl *CreateDeserialized(ASTContext &C, DeclID ID);
+  static FileScopeAsmDecl *CreateDeserialized(ASTContext &C, GlobalDeclID ID);
 
   SourceLocation getAsmLoc() const { return getLocation(); }
   SourceLocation getRParenLoc() const { return RParenLoc; }
@@ -4469,7 +4469,7 @@ class TopLevelStmtDecl : public Decl, public DeclContext {
 
 public:
   static TopLevelStmtDecl *Create(ASTContext &C, Stmt *Statement);
-  static TopLevelStmtDecl *CreateDeserialized(ASTContext &C, DeclID ID);
+  static TopLevelStmtDecl *CreateDeserialized(ASTContext &C, GlobalDeclID ID);
 
   SourceRange getSourceRange() const override LLVM_READONLY;
   Stmt *getStmt() { return Statement; }
@@ -4563,7 +4563,7 @@ protected:
 
 public:
   static BlockDecl *Create(ASTContext &C, DeclContext *DC, SourceLocation L);
-  static BlockDecl *CreateDeserialized(ASTContext &C, DeclID ID);
+  static BlockDecl *CreateDeserialized(ASTContext &C, GlobalDeclID ID);
 
   SourceLocation getCaretLocation() const { return getLocation(); }
 
@@ -4717,7 +4717,7 @@ public:
 
   static CapturedDecl *Create(ASTContext &C, DeclContext *DC,
                               unsigned NumParams);
-  static CapturedDecl *CreateDeserialized(ASTContext &C, DeclID ID,
+  static CapturedDecl *CreateDeserialized(ASTContext &C, GlobalDeclID ID,
                                           unsigned NumParams);
 
   Stmt *getBody() const override;
@@ -4851,7 +4851,7 @@ public:
                                     SourceLocation EndLoc);
 
   /// Create a new, deserialized module import declaration.
-  static ImportDecl *CreateDeserialized(ASTContext &C, DeclID ID,
+  static ImportDecl *CreateDeserialized(ASTContext &C, GlobalDeclID ID,
                                         unsigned NumLocations);
 
   /// Retrieve the module that was imported by the import declaration.
@@ -4892,7 +4892,7 @@ private:
 public:
   static ExportDecl *Create(ASTContext &C, DeclContext *DC,
                             SourceLocation ExportLoc);
-  static ExportDecl *CreateDeserialized(ASTContext &C, DeclID ID);
+  static ExportDecl *CreateDeserialized(ASTContext &C, GlobalDeclID ID);
 
   SourceLocation getExportLoc() const { return getLocation(); }
   SourceLocation getRBraceLoc() const { return RBraceLoc; }
@@ -4931,7 +4931,7 @@ class EmptyDecl : public Decl {
 public:
   static EmptyDecl *Create(ASTContext &C, DeclContext *DC,
                            SourceLocation L);
-  static EmptyDecl *CreateDeserialized(ASTContext &C, DeclID ID);
+  static EmptyDecl *CreateDeserialized(ASTContext &C, GlobalDeclID ID);
 
   static bool classof(const Decl *D) { return classofKind(D->getKind()); }
   static bool classofKind(Kind K) { return K == Empty; }
@@ -4957,7 +4957,7 @@ public:
                                 bool CBuffer, SourceLocation KwLoc,
                                 IdentifierInfo *ID, SourceLocation IDLoc,
                                 SourceLocation LBrace);
-  static HLSLBufferDecl *CreateDeserialized(ASTContext &C, DeclID ID);
+  static HLSLBufferDecl *CreateDeserialized(ASTContext &C, GlobalDeclID ID);
 
   SourceRange getSourceRange() const override LLVM_READONLY {
     return SourceRange(getLocStart(), RBraceLoc);
diff --git a/clang/include/clang/AST/DeclBase.h b/clang/include/clang/AST/DeclBase.h
index 161e14fc8969..e43e812cd945 100644
--- a/clang/include/clang/AST/DeclBase.h
+++ b/clang/include/clang/AST/DeclBase.h
@@ -15,6 +15,7 @@
 
 #include "clang/AST/ASTDumperUtils.h"
 #include "clang/AST/AttrIterator.h"
+#include "clang/AST/DeclID.h"
 #include "clang/AST/DeclarationName.h"
 #include "clang/AST/SelectorLocationsKind.h"
 #include "clang/Basic/IdentifierTable.h"
@@ -349,8 +350,6 @@ protected:
   LLVM_PREFERRED_TYPE(Linkage)
   mutable unsigned CacheValidAndLinkage : 3;
 
-  using DeclID = uint32_t;
-
   /// Allocate memory for a deserialized declaration.
   ///
   /// This routine must be used to allocate memory for any declaration that is
@@ -360,7 +359,7 @@ protected:
   /// \param Ctx The context in which we will allocate memory.
   /// \param ID The global ID of the deserialized declaration.
   /// \param Extra The amount of extra space to allocate after the object.
-  void *operator new(std::size_t Size, const ASTContext &Ctx, DeclID ID,
+  void *operator new(std::size_t Size, const ASTContext &Ctx, GlobalDeclID ID,
                      std::size_t Extra = 0);
 
   /// Allocate memory for a non-deserialized declaration.
@@ -778,10 +777,10 @@ public:
 
   /// Retrieve the global declaration ID associated with this
   /// declaration, which specifies where this Decl was loaded from.
-  unsigned getGlobalID() const {
+  GlobalDeclID getGlobalID() const {
     if (isFromASTFile())
-      return *((const unsigned*)this - 1);
-    return 0;
+      return (*((const GlobalDeclID *)this - 1));
+    return GlobalDeclID();
   }
 
   /// Retrieve the global ID of the module that owns this particular
diff --git a/clang/include/clang/AST/DeclCXX.h b/clang/include/clang/AST/DeclCXX.h
index a7644d2a19d2..fb52ac804849 100644
--- a/clang/include/clang/AST/DeclCXX.h
+++ b/clang/include/clang/AST/DeclCXX.h
@@ -120,7 +120,7 @@ public:
     return new (C, DC) AccessSpecDecl(AS, DC, ASLoc, ColonLoc);
   }
 
-  static AccessSpecDecl *CreateDeserialized(ASTContext &C, DeclID ID);
+  static AccessSpecDecl *CreateDeserialized(ASTContext &C, GlobalDeclID ID);
 
   // Implement isa/cast/dyncast/etc.
   static bool classof(const Decl *D) { return classofKind(D->getKind()); }
@@ -579,7 +579,8 @@ public:
                                      TypeSourceInfo *Info, SourceLocation Loc,
                                      unsigned DependencyKind, bool IsGeneric,
                                      LambdaCaptureDefault CaptureDefault);
-  static CXXRecordDecl *CreateDeserialized(const ASTContext &C, DeclID ID);
+  static CXXRecordDecl *CreateDeserialized(const ASTContext &C,
+                                           GlobalDeclID ID);
 
   bool isDynamicClass() const {
     return data().Polymorphic || data().NumVBases != 0;
@@ -1980,7 +1981,8 @@ public:
          CXXConstructorDecl *Ctor = nullptr,
          DeductionCandidate Kind = DeductionCandidate::Normal);
 
-  static CXXDeductionGuideDecl *CreateDeserialized(ASTContext &C, DeclID ID);
+  static CXXDeductionGuideDecl *CreateDeserialized(ASTContext &C,
+                                                   GlobalDeclID ID);
 
   ExplicitSpecifier getExplicitSpecifier() { return ExplicitSpec; }
   const ExplicitSpecifier getExplicitSpecifier() const { return ExplicitSpec; }
@@ -2035,7 +2037,8 @@ public:
   static RequiresExprBodyDecl *Create(ASTContext &C, DeclContext *DC,
                                       SourceLocation StartLoc);
 
-  static RequiresExprBodyDecl *CreateDeserialized(ASTContext &C, DeclID ID);
+  static RequiresExprBodyDecl *CreateDeserialized(ASTContext &C,
+                                                  GlobalDeclID ID);
 
   // Implement isa/cast/dyncast/etc.
   static bool classof(const Decl *D) { return classofKind(D->getKind()); }
@@ -2078,7 +2081,7 @@ public:
          ConstexprSpecKind ConstexprKind, SourceLocation EndLocation,
          Expr *TrailingRequiresClause = nullptr);
 
-  static CXXMethodDecl *CreateDeserialized(ASTContext &C, DeclID ID);
+  static CXXMethodDecl *CreateDeserialized(ASTContext &C, GlobalDeclID ID);
 
   bool isStatic() const;
   bool isInstance() const { return !isStatic(); }
@@ -2579,7 +2582,7 @@ public:
   friend class ASTDeclWriter;
   friend TrailingObjects;
 
-  static CXXConstructorDecl *CreateDeserialized(ASTContext &C, DeclID ID,
+  static CXXConstructorDecl *CreateDeserialized(ASTContext &C, GlobalDeclID ID,
                                                 uint64_t AllocKind);
   static CXXConstructorDecl *
   Create(ASTContext &C, CXXRecordDecl *RD, SourceLocation StartLoc,
@@ -2822,7 +2825,7 @@ public:
          bool UsesFPIntrin, bool isInline, bool isImplicitlyDeclared,
          ConstexprSpecKind ConstexprKind,
          Expr *TrailingRequiresClause = nullptr);
-  static CXXDestructorDecl *CreateDeserialized(ASTContext & C, DeclID ID);
+  static CXXDestructorDecl *CreateDeserialized(ASTContext &C, GlobalDeclID ID);
 
   void setOperatorDelete(FunctionDecl *OD, Expr *ThisArg);
 
@@ -2881,7 +2884,7 @@ public:
          bool UsesFPIntrin, bool isInline, ExplicitSpecifier ES,
          ConstexprSpecKind ConstexprKind, SourceLocation EndLocation,
          Expr *TrailingRequiresClause = nullptr);
-  static CXXConversionDecl *CreateDeserialized(ASTContext &C, DeclID ID);
+  static CXXConversionDecl *CreateDeserialized(ASTContext &C, GlobalDeclID ID);
 
   ExplicitSpecifier getExplicitSpecifier() {
     return getCanonicalDecl()->ExplicitSpec;
@@ -2948,7 +2951,7 @@ public:
                                  SourceLocation ExternLoc,
                                  SourceLocation LangLoc,
                                  LinkageSpecLanguageIDs Lang, bool HasBraces);
-  static LinkageSpecDecl *CreateDeserialized(ASTContext &C, DeclID ID);
+  static LinkageSpecDecl *CreateDeserialized(ASTContext &C, GlobalDeclID ID);
 
   /// Return the language specified by this linkage specification.
   LinkageSpecLanguageIDs getLanguage() const {
@@ -3096,7 +3099,7 @@ public:
                                     SourceLocation IdentLoc,
                                     NamedDecl *Nominated,
                                     DeclContext *CommonAncestor);
-  static UsingDirectiveDecl *CreateDeserialized(ASTContext &C, DeclID ID);
+  static UsingDirectiveDecl *CreateDeserialized(ASTContext &C, GlobalDeclID ID);
 
   SourceRange getSourceRange() const override LLVM_READONLY {
     return SourceRange(UsingLoc, getLocation());
@@ -3157,7 +3160,7 @@ public:
                                     SourceLocation IdentLoc,
                                     NamedDecl *Namespace);
 
-  static NamespaceAliasDecl *CreateDeserialized(ASTContext &C, DeclID ID);
+  static NamespaceAliasDecl *CreateDeserialized(ASTContext &C, GlobalDeclID ID);
 
   using redecl_range = redeclarable_base::redecl_range;
   using redecl_iterator = redeclarable_base::redecl_iterator;
@@ -3254,7 +3257,7 @@ public:
         LifetimeExtendedTemporaryDecl(Temp, EDec, Mangling);
   }
   static LifetimeExtendedTemporaryDecl *CreateDeserialized(ASTContext &C,
-                                                           DeclID ID) {
+                                                           GlobalDeclID ID) {
     return new (C, ID) LifetimeExtendedTemporaryDecl(EmptyShell{});
   }
 
@@ -3357,7 +3360,7 @@ public:
         UsingShadowDecl(UsingShadow, C, DC, Loc, Name, Introducer, Target);
   }
 
-  static UsingShadowDecl *CreateDeserialized(ASTContext &C, DeclID ID);
+  static UsingShadowDecl *CreateDeserialized(ASTContext &C, GlobalDeclID ID);
 
   using redecl_range = redeclarable_base::redecl_range;
   using redecl_iterator = redeclarable_base::redecl_iterator;
@@ -3566,7 +3569,7 @@ public:
                            const DeclarationNameInfo &NameInfo,
                            bool HasTypenameKeyword);
 
-  static UsingDecl *CreateDeserialized(ASTContext &C, DeclID ID);
+  static UsingDecl *CreateDeserialized(ASTContext &C, GlobalDeclID ID);
 
   SourceRange getSourceRange() const override LLVM_READONLY;
 
@@ -3645,7 +3648,7 @@ public:
                                             UsingDecl *Using, NamedDecl *Target,
                                             bool IsVirtual);
   static ConstructorUsingShadowDecl *CreateDeserialized(ASTContext &C,
-                                                        DeclID ID);
+                                                        GlobalDeclID ID);
 
   /// Override the UsingShadowDecl's getIntroducer, returning the UsingDecl that
   /// introduced this.
@@ -3757,7 +3760,7 @@ public:
                                SourceLocation UsingL, SourceLocation EnumL,
                                SourceLocation NameL, TypeSourceInfo *EnumType);
 
-  static UsingEnumDecl *CreateDeserialized(ASTContext &C, DeclID ID);
+  static UsingEnumDecl *CreateDeserialized(ASTContext &C, GlobalDeclID ID);
 
   SourceRange getSourceRange() const override LLVM_READONLY;
 
@@ -3830,7 +3833,7 @@ public:
                                NamedDecl *InstantiatedFrom,
                                ArrayRef<NamedDecl *> UsingDecls);
 
-  static UsingPackDecl *CreateDeserialized(ASTContext &C, DeclID ID,
+  static UsingPackDecl *CreateDeserialized(ASTContext &C, GlobalDeclID ID,
                                            unsigned NumExpansions);
 
   SourceRange getSourceRange() const override LLVM_READONLY {
@@ -3923,8 +3926,8 @@ public:
            NestedNameSpecifierLoc QualifierLoc,
            const DeclarationNameInfo &NameInfo, SourceLocation EllipsisLoc);
 
-  static UnresolvedUsingValueDecl *
-  CreateDeserialized(ASTContext &C, DeclID ID);
+  static UnresolvedUsingValueDecl *CreateDeserialized(ASTContext &C,
+                                                      GlobalDeclID ID);
 
   SourceRange getSourceRange() const override LLVM_READONLY;
 
@@ -4014,8 +4017,8 @@ public:
            SourceLocation TargetNameLoc, DeclarationName TargetName,
            SourceLocation EllipsisLoc);
 
-  static UnresolvedUsingTypenameDecl *
-  CreateDeserialized(ASTContext &C, DeclID ID);
+  static UnresolvedUsingTypenameDecl *CreateDeserialized(ASTContext &C,
+                                                         GlobalDeclID ID);
 
   /// Retrieves the canonical declaration of this declaration.
   UnresolvedUsingTypenameDecl *getCanonicalDecl() override {
@@ -4045,7 +4048,7 @@ public:
                                              SourceLocation Loc,
                                              DeclarationName Name);
   static UnresolvedUsingIfExistsDecl *CreateDeserialized(ASTContext &Ctx,
-                                                         DeclID ID);
+                                                         GlobalDeclID ID);
 
   static bool classof(const Decl *D) { return classofKind(D->getKind()); }
   static bool classofKind(Kind K) { return K == Decl::UnresolvedUsingIfExists; }
@@ -4073,7 +4076,7 @@ public:
                                   SourceLocation StaticAssertLoc,
                                   Expr *AssertExpr, Expr *Message,
                                   SourceLocation RParenLoc, bool Failed);
-  static StaticAssertDecl *CreateDeserialized(ASTContext &C, DeclID ID);
+  static StaticAssertDecl *CreateDeserialized(ASTContext &C, GlobalDeclID ID);
 
   Expr *getAssertExpr() { return AssertExprAndFailed.getPointer(); }
   const Expr *getAssertExpr() const { return AssertExprAndFailed.getPointer(); }
@@ -4120,7 +4123,7 @@ public:
 
   static BindingDecl *Create(ASTContext &C, DeclContext *DC,
                              SourceLocation IdLoc, IdentifierInfo *Id);
-  static BindingDecl *CreateDeserialized(ASTContext &C, DeclID ID);
+  static BindingDecl *CreateDeserialized(ASTContext &C, GlobalDeclID ID);
 
   /// Get the expression to which this declaration is bound. This may be null
   /// in two different cases: while parsing the initializer for the
@@ -4189,7 +4192,7 @@ public:
                                    QualType T, TypeSourceInfo *TInfo,
                                    StorageClass S,
                                    ArrayRef<BindingDecl *> Bindings);
-  static DecompositionDecl *CreateDeserialized(ASTContext &C, DeclID ID,
+  static DecompositionDecl *CreateDeserialized(ASTContext &C, GlobalDeclID ID,
                                                unsigned NumBindings);
 
   ArrayRef<BindingDecl *> bindings() const {
@@ -4246,7 +4249,7 @@ public:
                                 SourceLocation L, DeclarationName N, QualType T,
                                 TypeSourceInfo *TInfo, SourceLocation StartL,
                                 IdentifierInfo *Getter, IdentifierInfo *Setter);
-  static MSPropertyDecl *CreateDeserialized(ASTContext &C, DeclID ID);
+  static MSPropertyDecl *CreateDeserialized(ASTContext &C, GlobalDeclID ID);
 
   static bool classof(const Decl *D) { return D->getKind() == MSProperty; }
 
@@ -4300,7 +4303,7 @@ private:
   MSGuidDecl(DeclContext *DC, QualType T, Parts P);
 
   static MSGuidDecl *Create(const ASTContext &C, QualType T, Parts P);
-  static MSGuidDecl *CreateDeserialized(ASTContext &C, DeclID ID);
+  static MSGuidDecl *CreateDeserialized(ASTContext &C, GlobalDeclID ID);
 
   // Only ASTContext::getMSGuidDecl and deserialization create these.
   friend class ASTContext;
@@ -4353,7 +4356,7 @@ class UnnamedGlobalConstantDecl : public ValueDecl,
   static UnnamedGlobalConstantDecl *Create(const ASTContext &C, QualType T,
                                            const APValue &APVal);
   static UnnamedGlobalConstantDecl *CreateDeserialized(ASTContext &C,
-                                                       DeclID ID);
+                                                       GlobalDeclID ID);
 
   // Only ASTContext::getUnnamedGlobalConstantDecl and deserialization create
   // these.
diff --git a/clang/include/clang/AST/DeclContextInternals.h b/clang/include/clang/AST/DeclContextInternals.h
index c4734ab57895..e169c4859219 100644
--- a/clang/include/clang/AST/DeclContextInternals.h
+++ b/clang/include/clang/AST/DeclContextInternals.h
@@ -42,11 +42,12 @@ class StoredDeclsList {
   /// external declarations.
   DeclsAndHasExternalTy Data;
 
-  template<typename Fn>
-  void erase_if(Fn ShouldErase) {
+  template <typename Fn> DeclListNode::Decls *erase_if(Fn ShouldErase) {
     Decls List = Data.getPointer();
+
     if (!List)
-      return;
+      return nullptr;
+
     ASTContext &C = getASTContext();
     DeclListNode::Decls NewHead = nullptr;
     DeclListNode::Decls *NewLast = nullptr;
@@ -79,6 +80,17 @@ class StoredDeclsList {
     Data.setPointer(NewHead);
 
     assert(llvm::none_of(getLookupResult(), ShouldErase) && "Still exists!");
+
+    if (!Data.getPointer())
+      // All declarations are erased.
+      return nullptr;
+    else if (NewHead.is<NamedDecl *>())
+      // The list only contains a declaration, the header itself.
+      return (DeclListNode::Decls *)&Data;
+    else {
+      assert(NewLast && NewLast->is<NamedDecl *>() && "Not the tail?");
+      return NewLast;
+    }
   }
 
   void erase(NamedDecl *ND) {
@@ -160,12 +172,16 @@ public:
 
   void replaceExternalDecls(ArrayRef<NamedDecl*> Decls) {
     // Remove all declarations that are either external or are replaced with
-    // external declarations.
-    erase_if([Decls](NamedDecl *ND) {
+    // external declarations with higher visibilities.
+    DeclListNode::Decls *Tail = erase_if([Decls](NamedDecl *ND) {
       if (ND->isFromASTFile())
         return true;
+      // FIXME: Can we get rid of this loop completely?
       for (NamedDecl *D : Decls)
-        if (D->declarationReplaces(ND, /*IsKnownNewer=*/false))
+        // Only replace the local declaration if the external declaration has
+        // higher visibilities.
+        if (D->getModuleOwnershipKind() <= ND->getModuleOwnershipKind() &&
+            D->declarationReplaces(ND, /*IsKnownNewer=*/false))
           return true;
       return false;
     });
@@ -185,24 +201,15 @@ public:
       DeclsAsList = Node;
     }
 
-    DeclListNode::Decls Head = Data.getPointer();
-    if (Head.isNull()) {
+    if (!Data.getPointer()) {
       Data.setPointer(DeclsAsList);
       return;
     }
 
-    // Find the end of the existing list.
-    // FIXME: It would be possible to preserve information from erase_if to
-    // avoid this rescan looking for the end of the list.
-    DeclListNode::Decls *Tail = &Head;
-    while (DeclListNode *Node = Tail->dyn_cast<DeclListNode *>())
-      Tail = &Node->Rest;
-
     // Append the Decls.
     DeclListNode *Node = C.AllocateDeclListNode(Tail->get<NamedDecl *>());
     Node->Rest = DeclsAsList;
     *Tail = Node;
-    Data.setPointer(Head);
   }
 
   /// Return the list of all the decls.
diff --git a/clang/include/clang/AST/DeclFriend.h b/clang/include/clang/AST/DeclFriend.h
index b56627a5337d..9789282f351a 100644
--- a/clang/include/clang/AST/DeclFriend.h
+++ b/clang/include/clang/AST/DeclFriend.h
@@ -112,7 +112,7 @@ public:
   Create(ASTContext &C, DeclContext *DC, SourceLocation L, FriendUnion Friend_,
          SourceLocation FriendL,
          ArrayRef<TemplateParameterList *> FriendTypeTPLists = std::nullopt);
-  static FriendDecl *CreateDeserialized(ASTContext &C, DeclID ID,
+  static FriendDecl *CreateDeserialized(ASTContext &C, GlobalDeclID ID,
                                         unsigned FriendTypeNumTPLists);
 
   /// If this friend declaration names an (untemplated but possibly
diff --git a/clang/include/clang/AST/DeclID.h b/clang/include/clang/AST/DeclID.h
new file mode 100644
index 000000000000..614ba06b6386
--- /dev/null
+++ b/clang/include/clang/AST/DeclID.h
@@ -0,0 +1,227 @@
+//===--- DeclID.h - ID number for deserialized declarations  ----*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file defines DeclID class family to describe the deserialized
+// declarations. The DeclID is widely used in AST via LazyDeclPtr, or calls to
+// `ExternalASTSource::getExternalDecl`. It will be helpful for type safety to
+// require the use of `DeclID` to explicit.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_CLANG_AST_DECLID_H
+#define LLVM_CLANG_AST_DECLID_H
+
+#include "llvm/ADT/DenseMapInfo.h"
+#include "llvm/ADT/iterator.h"
+
+namespace clang {
+
+/// Predefined declaration IDs.
+///
+/// These declaration IDs correspond to predefined declarations in the AST
+/// context, such as the NULL declaration ID. Such declarations are never
+/// actually serialized, since they will be built by the AST context when
+/// it is created.
+enum PredefinedDeclIDs {
+  /// The NULL declaration.
+  PREDEF_DECL_NULL_ID = 0,
+
+  /// The translation unit.
+  PREDEF_DECL_TRANSLATION_UNIT_ID = 1,
+
+  /// The Objective-C 'id' type.
+  PREDEF_DECL_OBJC_ID_ID = 2,
+
+  /// The Objective-C 'SEL' type.
+  PREDEF_DECL_OBJC_SEL_ID = 3,
+
+  /// The Objective-C 'Class' type.
+  PREDEF_DECL_OBJC_CLASS_ID = 4,
+
+  /// The Objective-C 'Protocol' type.
+  PREDEF_DECL_OBJC_PROTOCOL_ID = 5,
+
+  /// The signed 128-bit integer type.
+  PREDEF_DECL_INT_128_ID = 6,
+
+  /// The unsigned 128-bit integer type.
+  PREDEF_DECL_UNSIGNED_INT_128_ID = 7,
+
+  /// The internal 'instancetype' typedef.
+  PREDEF_DECL_OBJC_INSTANCETYPE_ID = 8,
+
+  /// The internal '__builtin_va_list' typedef.
+  PREDEF_DECL_BUILTIN_VA_LIST_ID = 9,
+
+  /// The internal '__va_list_tag' struct, if any.
+  PREDEF_DECL_VA_LIST_TAG = 10,
+
+  /// The internal '__builtin_ms_va_list' typedef.
+  PREDEF_DECL_BUILTIN_MS_VA_LIST_ID = 11,
+
+  /// The predeclared '_GUID' struct.
+  PREDEF_DECL_BUILTIN_MS_GUID_ID = 12,
+
+  /// The extern "C" context.
+  PREDEF_DECL_EXTERN_C_CONTEXT_ID = 13,
+
+  /// The internal '__make_integer_seq' template.
+  PREDEF_DECL_MAKE_INTEGER_SEQ_ID = 14,
+
+  /// The internal '__NSConstantString' typedef.
+  PREDEF_DECL_CF_CONSTANT_STRING_ID = 15,
+
+  /// The internal '__NSConstantString' tag type.
+  PREDEF_DECL_CF_CONSTANT_STRING_TAG_ID = 16,
+
+  /// The internal '__type_pack_element' template.
+  PREDEF_DECL_TYPE_PACK_ELEMENT_ID = 17,
+};
+
+/// The number of declaration IDs that are predefined.
+///
+/// For more information about predefined declarations, see the
+/// \c PredefinedDeclIDs type and the PREDEF_DECL_*_ID constants.
+const unsigned int NUM_PREDEF_DECL_IDS = 18;
+
+/// GlobalDeclID means DeclID in the current ASTContext and LocalDeclID means
+/// DeclID specific to a certain ModuleFile. Specially, in ASTWriter, the
+/// LocalDeclID to the ModuleFile been writting is equal to the GlobalDeclID.
+/// Outside the serializer, all the DeclID been used should be GlobalDeclID.
+/// We can translate a LocalDeclID to the GlobalDeclID by
+/// `ASTReader::getGlobalDeclID()`.
+
+class DeclIDBase {
+public:
+  /// An ID number that refers to a declaration in an AST file.
+  ///
+  /// The ID numbers of declarations are consecutive (in order of
+  /// discovery), with values below NUM_PREDEF_DECL_IDS being reserved.
+  /// At the start of a chain of precompiled headers, declaration ID 1 is
+  /// used for the translation unit declaration.
+  ///
+  /// DeclID should only be used directly in serialization. All other users
+  /// should use LocalDeclID or GlobalDeclID.
+  using DeclID = uint32_t;
+
+protected:
+  DeclIDBase() : ID(PREDEF_DECL_NULL_ID) {}
+  explicit DeclIDBase(DeclID ID) : ID(ID) {}
+
+public:
+  DeclID get() const { return ID; }
+
+  explicit operator DeclID() const { return ID; }
+
+  explicit operator PredefinedDeclIDs() const { return (PredefinedDeclIDs)ID; }
+
+  bool isValid() const { return ID != PREDEF_DECL_NULL_ID; }
+
+  bool isInvalid() const { return ID == PREDEF_DECL_NULL_ID; }
+
+  friend bool operator==(const DeclIDBase &LHS, const DeclIDBase &RHS) {
+    return LHS.ID == RHS.ID;
+  }
+  friend bool operator!=(const DeclIDBase &LHS, const DeclIDBase &RHS) {
+    return LHS.ID != RHS.ID;
+  }
+  // We may sort the decl ID.
+  friend bool operator<(const DeclIDBase &LHS, const DeclIDBase &RHS) {
+    return LHS.ID < RHS.ID;
+  }
+  friend bool operator>(const DeclIDBase &LHS, const DeclIDBase &RHS) {
+    return LHS.ID > RHS.ID;
+  }
+  friend bool operator<=(const DeclIDBase &LHS, const DeclIDBase &RHS) {
+    return LHS.ID <= RHS.ID;
+  }
+  friend bool operator>=(const DeclIDBase &LHS, const DeclIDBase &RHS) {
+    return LHS.ID >= RHS.ID;
+  }
+
+protected:
+  DeclID ID;
+};
+
+class LocalDeclID : public DeclIDBase {
+  using Base = DeclIDBase;
+
+public:
+  LocalDeclID() : Base() {}
+  LocalDeclID(PredefinedDeclIDs ID) : Base(ID) {}
+  explicit LocalDeclID(DeclID ID) : Base(ID) {}
+
+  LocalDeclID &operator++() {
+    ++ID;
+    return *this;
+  }
+
+  LocalDeclID operator++(int) {
+    LocalDeclID Ret = *this;
+    ++(*this);
+    return Ret;
+  }
+};
+
+class GlobalDeclID : public DeclIDBase {
+  using Base = DeclIDBase;
+
+public:
+  GlobalDeclID() : Base() {}
+  explicit GlobalDeclID(DeclID ID) : Base(ID) {}
+
+  // For DeclIDIterator<GlobalDeclID> to be able to convert a GlobalDeclID
+  // to a LocalDeclID.
+  explicit operator LocalDeclID() const { return LocalDeclID(this->ID); }
+};
+
+/// A helper iterator adaptor to convert the iterators to
+/// `SmallVector<SomeDeclID>` to the iterators to `SmallVector<OtherDeclID>`.
+template <class FromTy, class ToTy>
+class DeclIDIterator
+    : public llvm::iterator_adaptor_base<DeclIDIterator<FromTy, ToTy>,
+                                         const FromTy *,
+                                         std::forward_iterator_tag, ToTy> {
+public:
+  DeclIDIterator() : DeclIDIterator::iterator_adaptor_base(nullptr) {}
+
+  DeclIDIterator(const FromTy *ID)
+      : DeclIDIterator::iterator_adaptor_base(ID) {}
+
+  ToTy operator*() const { return ToTy(*this->I); }
+
+  bool operator==(const DeclIDIterator &RHS) const { return this->I == RHS.I; }
+};
+
+} // namespace clang
+
+namespace llvm {
+template <> struct DenseMapInfo<clang::GlobalDeclID> {
+  using GlobalDeclID = clang::GlobalDeclID;
+  using DeclID = GlobalDeclID::DeclID;
+
+  static GlobalDeclID getEmptyKey() {
+    return GlobalDeclID(DenseMapInfo<DeclID>::getEmptyKey());
+  }
+
+  static GlobalDeclID getTombstoneKey() {
+    return GlobalDeclID(DenseMapInfo<DeclID>::getTombstoneKey());
+  }
+
+  static unsigned getHashValue(const GlobalDeclID &Key) {
+    return DenseMapInfo<DeclID>::getHashValue(Key.get());
+  }
+
+  static bool isEqual(const GlobalDeclID &L, const GlobalDeclID &R) {
+    return L == R;
+  }
+};
+
+} // namespace llvm
+
+#endif
diff --git a/clang/include/clang/AST/DeclObjC.h b/clang/include/clang/AST/DeclObjC.h
index 7780afa6f1cf..d2cc61ca19f8 100644
--- a/clang/include/clang/AST/DeclObjC.h
+++ b/clang/include/clang/AST/DeclObjC.h
@@ -236,7 +236,7 @@ public:
          ObjCImplementationControl impControl = ObjCImplementationControl::None,
          bool HasRelatedResultType = false);
 
-  static ObjCMethodDecl *CreateDeserialized(ASTContext &C, DeclID ID);
+  static ObjCMethodDecl *CreateDeserialized(ASTContext &C, GlobalDeclID ID);
 
   ObjCMethodDecl *getCanonicalDecl() override;
   const ObjCMethodDecl *getCanonicalDecl() const {
@@ -614,7 +614,8 @@ public:
                                    IdentifierInfo *name,
                                    SourceLocation colonLoc,
                                    TypeSourceInfo *boundInfo);
-  static ObjCTypeParamDecl *CreateDeserialized(ASTContext &ctx, DeclID ID);
+  static ObjCTypeParamDecl *CreateDeserialized(ASTContext &ctx,
+                                               GlobalDeclID ID);
 
   SourceRange getSourceRange() const override LLVM_READONLY;
 
@@ -789,7 +790,7 @@ public:
                                   TypeSourceInfo *TSI,
                                   PropertyControl propControl = None);
 
-  static ObjCPropertyDecl *CreateDeserialized(ASTContext &C, DeclID ID);
+  static ObjCPropertyDecl *CreateDeserialized(ASTContext &C, GlobalDeclID ID);
 
   SourceLocation getAtLoc() const { return AtLoc; }
   void setAtLoc(SourceLocation L) { AtLoc = L; }
@@ -1279,7 +1280,8 @@ public:
          ObjCInterfaceDecl *PrevDecl,
          SourceLocation ClassLoc = SourceLocation(), bool isInternal = false);
 
-  static ObjCInterfaceDecl *CreateDeserialized(const ASTContext &C, DeclID ID);
+  static ObjCInterfaceDecl *CreateDeserialized(const ASTContext &C,
+                                               GlobalDeclID ID);
 
   /// Retrieve the type parameters of this class.
   ///
@@ -1969,7 +1971,7 @@ public:
                               TypeSourceInfo *TInfo, AccessControl ac,
                               Expr *BW = nullptr, bool synthesized = false);
 
-  static ObjCIvarDecl *CreateDeserialized(ASTContext &C, DeclID ID);
+  static ObjCIvarDecl *CreateDeserialized(ASTContext &C, GlobalDeclID ID);
 
   /// Return the class interface that this ivar is logically contained
   /// in; this is either the interface where the ivar was declared, or the
@@ -2039,7 +2041,8 @@ public:
                                      SourceLocation IdLoc, IdentifierInfo *Id,
                                      QualType T, Expr *BW);
 
-  static ObjCAtDefsFieldDecl *CreateDeserialized(ASTContext &C, DeclID ID);
+  static ObjCAtDefsFieldDecl *CreateDeserialized(ASTContext &C,
+                                                 GlobalDeclID ID);
 
   // Implement isa/cast/dyncast/etc.
   static bool classof(const Decl *D) { return classofKind(D->getKind()); }
@@ -2142,7 +2145,7 @@ public:
                                   SourceLocation atStartLoc,
                                   ObjCProtocolDecl *PrevDecl);
 
-  static ObjCProtocolDecl *CreateDeserialized(ASTContext &C, DeclID ID);
+  static ObjCProtocolDecl *CreateDeserialized(ASTContext &C, GlobalDeclID ID);
 
   const ObjCProtocolList &getReferencedProtocols() const {
     assert(hasDefinition() && "No definition available!");
@@ -2361,7 +2364,7 @@ public:
          ObjCTypeParamList *typeParamList,
          SourceLocation IvarLBraceLoc = SourceLocation(),
          SourceLocation IvarRBraceLoc = SourceLocation());
-  static ObjCCategoryDecl *CreateDeserialized(ASTContext &C, DeclID ID);
+  static ObjCCategoryDecl *CreateDeserialized(ASTContext &C, GlobalDeclID ID);
 
   ObjCInterfaceDecl *getClassInterface() { return ClassInterface; }
   const ObjCInterfaceDecl *getClassInterface() const { return ClassInterface; }
@@ -2558,7 +2561,8 @@ public:
   Create(ASTContext &C, DeclContext *DC, const IdentifierInfo *Id,
          ObjCInterfaceDecl *classInterface, SourceLocation nameLoc,
          SourceLocation atStartLoc, SourceLocation CategoryNameLoc);
-  static ObjCCategoryImplDecl *CreateDeserialized(ASTContext &C, DeclID ID);
+  static ObjCCategoryImplDecl *CreateDeserialized(ASTContext &C,
+                                                  GlobalDeclID ID);
 
   ObjCCategoryDecl *getCategoryDecl() const;
 
@@ -2640,7 +2644,8 @@ public:
                                         SourceLocation IvarLBraceLoc=SourceLocation(),
                                         SourceLocation IvarRBraceLoc=SourceLocation());
 
-  static ObjCImplementationDecl *CreateDeserialized(ASTContext &C, DeclID ID);
+  static ObjCImplementationDecl *CreateDeserialized(ASTContext &C,
+                                                    GlobalDeclID ID);
 
   /// init_iterator - Iterates through the ivar initializer list.
   using init_iterator = CXXCtorInitializer **;
@@ -2780,7 +2785,7 @@ public:
                                          ObjCInterfaceDecl* aliasedClass);
 
   static ObjCCompatibleAliasDecl *CreateDeserialized(ASTContext &C,
-                                                     DeclID ID);
+                                                     GlobalDeclID ID);
 
   const ObjCInterfaceDecl *getClassInterface() const { return AliasedClass; }
   ObjCInterfaceDecl *getClassInterface() { return AliasedClass; }
@@ -2851,7 +2856,8 @@ public:
                                       ObjCIvarDecl *ivarDecl,
                                       SourceLocation ivarLoc);
 
-  static ObjCPropertyImplDecl *CreateDeserialized(ASTContext &C, DeclID ID);
+  static ObjCPropertyImplDecl *CreateDeserialized(ASTContext &C,
+                                                  GlobalDeclID ID);
 
   SourceRange getSourceRange() const override LLVM_READONLY;
 
diff --git a/clang/include/clang/AST/DeclOpenMP.h b/clang/include/clang/AST/DeclOpenMP.h
index c7ede7f2157f..e542c3c8e66b 100644
--- a/clang/include/clang/AST/DeclOpenMP.h
+++ b/clang/include/clang/AST/DeclOpenMP.h
@@ -59,7 +59,7 @@ protected:
   }
 
   template <typename T, typename... Params>
-  static T *createEmptyDirective(const ASTContext &C, unsigned ID,
+  static T *createEmptyDirective(const ASTContext &C, GlobalDeclID ID,
                                  unsigned NumClauses, unsigned NumChildren,
                                  Params &&... P) {
     auto *Inst = new (C, ID, size(NumClauses, NumChildren))
@@ -133,7 +133,7 @@ public:
                                       SourceLocation L,
                                       ArrayRef<Expr *> VL);
   static OMPThreadPrivateDecl *CreateDeserialized(ASTContext &C,
-                                                  DeclID ID, unsigned N);
+                                                  GlobalDeclID ID, unsigned N);
 
   typedef MutableArrayRef<Expr *>::iterator varlist_iterator;
   typedef ArrayRef<const Expr *>::iterator varlist_const_iterator;
@@ -214,7 +214,7 @@ public:
          QualType T, OMPDeclareReductionDecl *PrevDeclInScope);
   /// Create deserialized declare reduction node.
   static OMPDeclareReductionDecl *CreateDeserialized(ASTContext &C,
-                                                     DeclID ID);
+                                                     GlobalDeclID ID);
 
   /// Get combiner expression of the declare reduction construct.
   Expr *getCombiner() { return Combiner; }
@@ -318,8 +318,8 @@ public:
                                       ArrayRef<OMPClause *> Clauses,
                                       OMPDeclareMapperDecl *PrevDeclInScope);
   /// Creates deserialized declare mapper node.
-  static OMPDeclareMapperDecl *CreateDeserialized(ASTContext &C, DeclID ID,
-                                                  unsigned N);
+  static OMPDeclareMapperDecl *CreateDeserialized(ASTContext &C,
+                                                  GlobalDeclID ID, unsigned N);
 
   using clauselist_iterator = MutableArrayRef<OMPClause *>::iterator;
   using clauselist_const_iterator = ArrayRef<const OMPClause *>::iterator;
@@ -397,7 +397,8 @@ public:
                                      IdentifierInfo *Id, QualType T,
                                      SourceLocation StartLoc);
 
-  static OMPCapturedExprDecl *CreateDeserialized(ASTContext &C, DeclID ID);
+  static OMPCapturedExprDecl *CreateDeserialized(ASTContext &C,
+                                                 GlobalDeclID ID);
 
   SourceRange getSourceRange() const override LLVM_READONLY;
 
@@ -427,7 +428,7 @@ public:
   static OMPRequiresDecl *Create(ASTContext &C, DeclContext *DC,
                                  SourceLocation L, ArrayRef<OMPClause *> CL);
   /// Create deserialized requires node.
-  static OMPRequiresDecl *CreateDeserialized(ASTContext &C, DeclID ID,
+  static OMPRequiresDecl *CreateDeserialized(ASTContext &C, GlobalDeclID ID,
                                              unsigned N);
 
   using clauselist_iterator = MutableArrayRef<OMPClause *>::iterator;
@@ -495,7 +496,7 @@ public:
   static OMPAllocateDecl *Create(ASTContext &C, DeclContext *DC,
                                  SourceLocation L, ArrayRef<Expr *> VL,
                                  ArrayRef<OMPClause *> CL);
-  static OMPAllocateDecl *CreateDeserialized(ASTContext &C, DeclID ID,
+  static OMPAllocateDecl *CreateDeserialized(ASTContext &C, GlobalDeclID ID,
                                              unsigned NVars, unsigned NClauses);
 
   typedef MutableArrayRef<Expr *>::iterator varlist_iterator;
diff --git a/clang/include/clang/AST/DeclTemplate.h b/clang/include/clang/AST/DeclTemplate.h
index e2afff8d4450..3ee03eebdb8c 100644
--- a/clang/include/clang/AST/DeclTemplate.h
+++ b/clang/include/clang/AST/DeclTemplate.h
@@ -797,7 +797,7 @@ protected:
     ///
     /// The first value in the array is the number of specializations/partial
     /// specializations that follow.
-    uint32_t *LazySpecializations = nullptr;
+    GlobalDeclID *LazySpecializations = nullptr;
 
     /// The set of "injected" template arguments used within this
     /// template.
@@ -1087,7 +1087,8 @@ public:
                                       NamedDecl *Decl);
 
   /// Create an empty function template node.
-  static FunctionTemplateDecl *CreateDeserialized(ASTContext &C, DeclID ID);
+  static FunctionTemplateDecl *CreateDeserialized(ASTContext &C,
+                                                  GlobalDeclID ID);
 
   // Implement isa/cast/dyncast support
   static bool classof(const Decl *D) { return classofKind(D->getKind()); }
@@ -1204,9 +1205,9 @@ public:
          bool Typename, bool ParameterPack, bool HasTypeConstraint = false,
          std::optional<unsigned> NumExpanded = std::nullopt);
   static TemplateTypeParmDecl *CreateDeserialized(const ASTContext &C,
-                                                  DeclID ID);
+                                                  GlobalDeclID ID);
   static TemplateTypeParmDecl *CreateDeserialized(const ASTContext &C,
-                                                  DeclID ID,
+                                                  GlobalDeclID ID,
                                                   bool HasTypeConstraint);
 
   /// Whether this template type parameter was declared with
@@ -1413,11 +1414,10 @@ public:
          QualType T, TypeSourceInfo *TInfo, ArrayRef<QualType> ExpandedTypes,
          ArrayRef<TypeSourceInfo *> ExpandedTInfos);
 
+  static NonTypeTemplateParmDecl *
+  CreateDeserialized(ASTContext &C, GlobalDeclID ID, bool HasTypeConstraint);
   static NonTypeTemplateParmDecl *CreateDeserialized(ASTContext &C,
-                                                     DeclID ID,
-                                                     bool HasTypeConstraint);
-  static NonTypeTemplateParmDecl *CreateDeserialized(ASTContext &C,
-                                                     DeclID ID,
+                                                     GlobalDeclID ID,
                                                      unsigned NumExpandedTypes,
                                                      bool HasTypeConstraint);
 
@@ -1632,10 +1632,9 @@ public:
          ArrayRef<TemplateParameterList *> Expansions);
 
   static TemplateTemplateParmDecl *CreateDeserialized(ASTContext &C,
-                                                      DeclID ID);
-  static TemplateTemplateParmDecl *CreateDeserialized(ASTContext &C,
-                                                      DeclID ID,
-                                                      unsigned NumExpansions);
+                                                      GlobalDeclID ID);
+  static TemplateTemplateParmDecl *
+  CreateDeserialized(ASTContext &C, GlobalDeclID ID, unsigned NumExpansions);
 
   using TemplateParmPosition::getDepth;
   using TemplateParmPosition::setDepth;
@@ -1857,8 +1856,8 @@ public:
          ClassTemplateDecl *SpecializedTemplate,
          ArrayRef<TemplateArgument> Args,
          ClassTemplateSpecializationDecl *PrevDecl);
-  static ClassTemplateSpecializationDecl *
-  CreateDeserialized(ASTContext &C, DeclID ID);
+  static ClassTemplateSpecializationDecl *CreateDeserialized(ASTContext &C,
+                                                             GlobalDeclID ID);
 
   void getNameForDiagnostic(raw_ostream &OS, const PrintingPolicy &Policy,
                             bool Qualified) const override;
@@ -2110,7 +2109,7 @@ public:
          ClassTemplatePartialSpecializationDecl *PrevDecl);
 
   static ClassTemplatePartialSpecializationDecl *
-  CreateDeserialized(ASTContext &C, DeclID ID);
+  CreateDeserialized(ASTContext &C, GlobalDeclID ID);
 
   ClassTemplatePartialSpecializationDecl *getMostRecentDecl() {
     return cast<ClassTemplatePartialSpecializationDecl>(
@@ -2306,7 +2305,7 @@ public:
                                    NamedDecl *Decl);
 
   /// Create an empty class template node.
-  static ClassTemplateDecl *CreateDeserialized(ASTContext &C, DeclID ID);
+  static ClassTemplateDecl *CreateDeserialized(ASTContext &C, GlobalDeclID ID);
 
   /// Return the specialization with the provided arguments if it exists,
   /// otherwise return the insertion point.
@@ -2472,7 +2471,7 @@ public:
          MutableArrayRef<TemplateParameterList *> Params, FriendUnion Friend,
          SourceLocation FriendLoc);
 
-  static FriendTemplateDecl *CreateDeserialized(ASTContext &C, DeclID ID);
+  static FriendTemplateDecl *CreateDeserialized(ASTContext &C, GlobalDeclID ID);
 
   /// If this friend declaration names a templated type (or
   /// a dependent member type of a templated type), return that
@@ -2573,7 +2572,8 @@ public:
                                        NamedDecl *Decl);
 
   /// Create an empty alias template node.
-  static TypeAliasTemplateDecl *CreateDeserialized(ASTContext &C, DeclID ID);
+  static TypeAliasTemplateDecl *CreateDeserialized(ASTContext &C,
+                                                   GlobalDeclID ID);
 
   // Implement isa/cast/dyncast support
   static bool classof(const Decl *D) { return classofKind(D->getKind()); }
@@ -2670,7 +2670,7 @@ public:
          TypeSourceInfo *TInfo, StorageClass S,
          ArrayRef<TemplateArgument> Args);
   static VarTemplateSpecializationDecl *CreateDeserialized(ASTContext &C,
-                                                           DeclID ID);
+                                                           GlobalDeclID ID);
 
   void getNameForDiagnostic(raw_ostream &OS, const PrintingPolicy &Policy,
                             bool Qualified) const override;
@@ -2900,8 +2900,8 @@ public:
          TypeSourceInfo *TInfo, StorageClass S, ArrayRef<TemplateArgument> Args,
          const TemplateArgumentListInfo &ArgInfos);
 
-  static VarTemplatePartialSpecializationDecl *CreateDeserialized(ASTContext &C,
-                                                                  DeclID ID);
+  static VarTemplatePartialSpecializationDecl *
+  CreateDeserialized(ASTContext &C, GlobalDeclID ID);
 
   VarTemplatePartialSpecializationDecl *getMostRecentDecl() {
     return cast<VarTemplatePartialSpecializationDecl>(
@@ -3078,7 +3078,7 @@ public:
                                  VarDecl *Decl);
 
   /// Create an empty variable template node.
-  static VarTemplateDecl *CreateDeserialized(ASTContext &C, DeclID ID);
+  static VarTemplateDecl *CreateDeserialized(ASTContext &C, GlobalDeclID ID);
 
   /// Return the specialization with the provided arguments if it exists,
   /// otherwise return the insertion point.
@@ -3183,7 +3183,7 @@ public:
                              SourceLocation L, DeclarationName Name,
                              TemplateParameterList *Params,
                              Expr *ConstraintExpr);
-  static ConceptDecl *CreateDeserialized(ASTContext &C, DeclID ID);
+  static ConceptDecl *CreateDeserialized(ASTContext &C, GlobalDeclID ID);
 
   Expr *getConstraintExpr() const {
     return ConstraintExpr;
@@ -3232,7 +3232,7 @@ public:
   Create(const ASTContext &C, DeclContext *DC, SourceLocation SL,
          ArrayRef<TemplateArgument> ConvertedArgs);
   static ImplicitConceptSpecializationDecl *
-  CreateDeserialized(const ASTContext &C, DeclID ID,
+  CreateDeserialized(const ASTContext &C, GlobalDeclID ID,
                      unsigned NumTemplateArgs);
 
   ArrayRef<TemplateArgument> getTemplateArguments() const {
@@ -3275,7 +3275,7 @@ private:
   static TemplateParamObjectDecl *Create(const ASTContext &C, QualType T,
                                          const APValue &V);
   static TemplateParamObjectDecl *CreateDeserialized(ASTContext &C,
-                                                     DeclID ID);
+                                                     GlobalDeclID ID);
 
   /// Only ASTContext::getTemplateParamObjectDecl and deserialization
   /// create these.
diff --git a/clang/include/clang/AST/Expr.h b/clang/include/clang/AST/Expr.h
index 2bfefeabc348..f2bf667636dc 100644
--- a/clang/include/clang/AST/Expr.h
+++ b/clang/include/clang/AST/Expr.h
@@ -6610,6 +6610,275 @@ public:
 
 };
 
+/// This class represents BOTH the OpenMP Array Section and OpenACC 'subarray',
+/// with a boolean differentiator.
+/// OpenMP 5.0 [2.1.5, Array Sections].
+/// To specify an array section in an OpenMP construct, array subscript
+/// expressions are extended with the following syntax:
+/// \code
+/// [ lower-bound : length : stride ]
+/// [ lower-bound : length : ]
+/// [ lower-bound : length ]
+/// [ lower-bound : : stride ]
+/// [ lower-bound : : ]
+/// [ lower-bound : ]
+/// [ : length : stride ]
+/// [ : length : ]
+/// [ : length ]
+/// [ : : stride ]
+/// [ : : ]
+/// [ : ]
+/// \endcode
+/// The array section must be a subset of the original array.
+/// Array sections are allowed on multidimensional arrays. Base language array
+/// subscript expressions can be used to specify length-one dimensions of
+/// multidimensional array sections.
+/// Each of the lower-bound, length, and stride expressions if specified must be
+/// an integral type expressions of the base language. When evaluated
+/// they represent a set of integer values as follows:
+/// \code
+/// { lower-bound, lower-bound + stride, lower-bound + 2 * stride,... ,
+/// lower-bound + ((length - 1) * stride) }
+/// \endcode
+/// The lower-bound and length must evaluate to non-negative integers.
+/// The stride must evaluate to a positive integer.
+/// When the size of the array dimension is not known, the length must be
+/// specified explicitly.
+/// When the stride is absent it defaults to 1.
+/// When the length is absent it defaults to ⌈(size − lower-bound)/stride⌉,
+/// where size is the size of the array dimension. When the lower-bound is
+/// absent it defaults to 0.
+///
+///
+/// OpenACC 3.3 [2.7.1 Data Specification in Data Clauses]
+/// In C and C++, a subarray is an array name followed by an extended array
+/// range specification in brackets, with start and length, such as
+///
+/// AA[2:n]
+///
+/// If the lower bound is missing, zero is used. If the length is missing and
+/// the array has known size, the size of the array is used; otherwise the
+/// length is required. The subarray AA[2:n] means elements AA[2], AA[3], . . .
+/// , AA[2+n-1]. In C and C++, a two dimensional array may be declared in at
+/// least four ways:
+///
+/// -Statically-sized array: float AA[100][200];
+/// -Pointer to statically sized rows: typedef float row[200]; row* BB;
+/// -Statically-sized array of pointers: float* CC[200];
+/// -Pointer to pointers: float** DD;
+///
+/// Each dimension may be statically sized, or a pointer to dynamically
+/// allocated memory. Each of these may be included in a data clause using
+/// subarray notation to specify a rectangular array:
+///
+/// -AA[2:n][0:200]
+/// -BB[2:n][0:m]
+/// -CC[2:n][0:m]
+/// -DD[2:n][0:m]
+///
+/// Multidimensional rectangular subarrays in C and C++ may be specified for any
+/// array with any combination of statically-sized or dynamically-allocated
+/// dimensions. For statically sized dimensions, all dimensions except the first
+/// must specify the whole extent to preserve the contiguous data restriction,
+/// discussed below. For dynamically allocated dimensions, the implementation
+/// will allocate pointers in device memory corresponding to the pointers in
+/// local memory and will fill in those pointers as appropriate.
+///
+/// In Fortran, a subarray is an array name followed by a comma-separated list
+/// of range specifications in parentheses, with lower and upper bound
+/// subscripts, such as
+///
+/// arr(1:high,low:100)
+///
+/// If either the lower or upper bounds are missing, the declared or allocated
+/// bounds of the array, if known, are used. All dimensions except the last must
+/// specify the whole extent, to preserve the contiguous data restriction,
+/// discussed below.
+///
+/// Restrictions
+///
+/// -In Fortran, the upper bound for the last dimension of an assumed-size dummy
+/// array must be specified.
+///
+/// -In C and C++, the length for dynamically allocated dimensions of an array
+/// must be explicitly specified.
+///
+/// -In C and C++, modifying pointers in pointer arrays during the data
+/// lifetime, either on the host or on the device, may result in undefined
+/// behavior.
+///
+/// -If a subarray  appears in a data clause, the implementation may choose to
+/// allocate memory for only that subarray on the accelerator.
+///
+/// -In Fortran, array pointers may appear, but pointer association is not
+/// preserved in device memory.
+///
+/// -Any array or subarray in a data clause, including Fortran array pointers,
+/// must be a contiguous section of memory, except for dynamic multidimensional
+/// C arrays.
+///
+/// -In C and C++, if a variable or array of composite type appears, all the
+/// data members of the struct or class are allocated and copied, as
+/// appropriate. If a composite member is a pointer type, the data addressed by
+/// that pointer are not implicitly copied.
+///
+/// -In Fortran, if a variable or array of composite type appears, all the
+/// members of that derived type are allocated and copied, as appropriate. If
+/// any member has the allocatable or pointer attribute, the data accessed
+/// through that member are not copied.
+///
+/// -If an expression is used in a subscript or subarray expression in a clause
+/// on a data construct, the same value is used when copying data at the end of
+/// the data region, even if the values of variables in the expression change
+/// during the data region.
+class ArraySectionExpr : public Expr {
+  friend class ASTStmtReader;
+  friend class ASTStmtWriter;
+
+public:
+  enum ArraySectionType { OMPArraySection, OpenACCArraySection };
+
+private:
+  enum {
+    BASE,
+    LOWER_BOUND,
+    LENGTH,
+    STRIDE,
+    END_EXPR,
+    OPENACC_END_EXPR = STRIDE
+  };
+
+  ArraySectionType ASType = OMPArraySection;
+  Stmt *SubExprs[END_EXPR] = {nullptr};
+  SourceLocation ColonLocFirst;
+  SourceLocation ColonLocSecond;
+  SourceLocation RBracketLoc;
+
+public:
+  // Constructor for OMP array sections, which include a 'stride'.
+  ArraySectionExpr(Expr *Base, Expr *LowerBound, Expr *Length, Expr *Stride,
+                   QualType Type, ExprValueKind VK, ExprObjectKind OK,
+                   SourceLocation ColonLocFirst, SourceLocation ColonLocSecond,
+                   SourceLocation RBracketLoc)
+      : Expr(ArraySectionExprClass, Type, VK, OK), ASType(OMPArraySection),
+        ColonLocFirst(ColonLocFirst), ColonLocSecond(ColonLocSecond),
+        RBracketLoc(RBracketLoc) {
+    setBase(Base);
+    setLowerBound(LowerBound);
+    setLength(Length);
+    setStride(Stride);
+    setDependence(computeDependence(this));
+  }
+
+  // Constructor for OpenACC sub-arrays, which do not permit a 'stride'.
+  ArraySectionExpr(Expr *Base, Expr *LowerBound, Expr *Length, QualType Type,
+                   ExprValueKind VK, ExprObjectKind OK, SourceLocation ColonLoc,
+                   SourceLocation RBracketLoc)
+      : Expr(ArraySectionExprClass, Type, VK, OK), ASType(OpenACCArraySection),
+        ColonLocFirst(ColonLoc), RBracketLoc(RBracketLoc) {
+    setBase(Base);
+    setLowerBound(LowerBound);
+    setLength(Length);
+    setDependence(computeDependence(this));
+  }
+
+  /// Create an empty array section expression.
+  explicit ArraySectionExpr(EmptyShell Shell)
+      : Expr(ArraySectionExprClass, Shell) {}
+
+  /// Return original type of the base expression for array section.
+  static QualType getBaseOriginalType(const Expr *Base);
+
+  static bool classof(const Stmt *T) {
+    return T->getStmtClass() == ArraySectionExprClass;
+  }
+
+  bool isOMPArraySection() const { return ASType == OMPArraySection; }
+  bool isOpenACCArraySection() const { return ASType == OpenACCArraySection; }
+
+  /// Get base of the array section.
+  Expr *getBase() { return cast<Expr>(SubExprs[BASE]); }
+  const Expr *getBase() const { return cast<Expr>(SubExprs[BASE]); }
+
+  /// Get lower bound of array section.
+  Expr *getLowerBound() { return cast_or_null<Expr>(SubExprs[LOWER_BOUND]); }
+  const Expr *getLowerBound() const {
+    return cast_or_null<Expr>(SubExprs[LOWER_BOUND]);
+  }
+
+  /// Get length of array section.
+  Expr *getLength() { return cast_or_null<Expr>(SubExprs[LENGTH]); }
+  const Expr *getLength() const { return cast_or_null<Expr>(SubExprs[LENGTH]); }
+
+  /// Get stride of array section.
+  Expr *getStride() {
+    assert(ASType != OpenACCArraySection &&
+           "Stride not valid in OpenACC subarrays");
+    return cast_or_null<Expr>(SubExprs[STRIDE]);
+  }
+
+  const Expr *getStride() const {
+    assert(ASType != OpenACCArraySection &&
+           "Stride not valid in OpenACC subarrays");
+    return cast_or_null<Expr>(SubExprs[STRIDE]);
+  }
+
+  SourceLocation getBeginLoc() const LLVM_READONLY {
+    return getBase()->getBeginLoc();
+  }
+  SourceLocation getEndLoc() const LLVM_READONLY { return RBracketLoc; }
+
+  SourceLocation getColonLocFirst() const { return ColonLocFirst; }
+  SourceLocation getColonLocSecond() const {
+    assert(ASType != OpenACCArraySection &&
+           "second colon for stride not valid in OpenACC subarrays");
+    return ColonLocSecond;
+  }
+  SourceLocation getRBracketLoc() const { return RBracketLoc; }
+
+  SourceLocation getExprLoc() const LLVM_READONLY {
+    return getBase()->getExprLoc();
+  }
+
+  child_range children() {
+    return child_range(
+        &SubExprs[BASE],
+        &SubExprs[ASType == OMPArraySection ? END_EXPR : OPENACC_END_EXPR]);
+  }
+
+  const_child_range children() const {
+    return const_child_range(
+        &SubExprs[BASE],
+        &SubExprs[ASType == OMPArraySection ? END_EXPR : OPENACC_END_EXPR]);
+  }
+
+private:
+  /// Set base of the array section.
+  void setBase(Expr *E) { SubExprs[BASE] = E; }
+
+  /// Set lower bound of the array section.
+  void setLowerBound(Expr *E) { SubExprs[LOWER_BOUND] = E; }
+
+  /// Set length of the array section.
+  void setLength(Expr *E) { SubExprs[LENGTH] = E; }
+
+  /// Set length of the array section.
+  void setStride(Expr *E) {
+    assert(ASType != OpenACCArraySection &&
+           "Stride not valid in OpenACC subarrays");
+    SubExprs[STRIDE] = E;
+  }
+
+  void setColonLocFirst(SourceLocation L) { ColonLocFirst = L; }
+
+  void setColonLocSecond(SourceLocation L) {
+    assert(ASType != OpenACCArraySection &&
+           "second colon for stride not valid in OpenACC subarrays");
+    ColonLocSecond = L;
+  }
+  void setRBracketLoc(SourceLocation L) { RBracketLoc = L; }
+};
+
 /// Frontend produces RecoveryExprs on semantic errors that prevent creating
 /// other well-formed expressions. E.g. when type-checking of a binary operator
 /// fails, we cannot produce a BinaryOperator expression. Instead, we can choose
diff --git a/clang/include/clang/AST/ExprOpenMP.h b/clang/include/clang/AST/ExprOpenMP.h
index be5b1f3fdd11..54a0c203f656 100644
--- a/clang/include/clang/AST/ExprOpenMP.h
+++ b/clang/include/clang/AST/ExprOpenMP.h
@@ -17,130 +17,6 @@
 #include "clang/AST/Expr.h"
 
 namespace clang {
-/// OpenMP 5.0 [2.1.5, Array Sections].
-/// To specify an array section in an OpenMP construct, array subscript
-/// expressions are extended with the following syntax:
-/// \code
-/// [ lower-bound : length : stride ]
-/// [ lower-bound : length : ]
-/// [ lower-bound : length ]
-/// [ lower-bound : : stride ]
-/// [ lower-bound : : ]
-/// [ lower-bound : ]
-/// [ : length : stride ]
-/// [ : length : ]
-/// [ : length ]
-/// [ : : stride ]
-/// [ : : ]
-/// [ : ]
-/// \endcode
-/// The array section must be a subset of the original array.
-/// Array sections are allowed on multidimensional arrays. Base language array
-/// subscript expressions can be used to specify length-one dimensions of
-/// multidimensional array sections.
-/// Each of the lower-bound, length, and stride expressions if specified must be
-/// an integral type expressions of the base language. When evaluated
-/// they represent a set of integer values as follows:
-/// \code
-/// { lower-bound, lower-bound + stride, lower-bound + 2 * stride,... ,
-/// lower-bound + ((length - 1) * stride) }
-/// \endcode
-/// The lower-bound and length must evaluate to non-negative integers.
-/// The stride must evaluate to a positive integer.
-/// When the size of the array dimension is not known, the length must be
-/// specified explicitly.
-/// When the stride is absent it defaults to 1.
-/// When the length is absent it defaults to ⌈(size − lower-bound)/stride⌉,
-/// where size is the size of the array dimension. When the lower-bound is
-/// absent it defaults to 0.
-class OMPArraySectionExpr : public Expr {
-  enum { BASE, LOWER_BOUND, LENGTH, STRIDE, END_EXPR };
-  Stmt *SubExprs[END_EXPR];
-  SourceLocation ColonLocFirst;
-  SourceLocation ColonLocSecond;
-  SourceLocation RBracketLoc;
-
-public:
-  OMPArraySectionExpr(Expr *Base, Expr *LowerBound, Expr *Length, Expr *Stride,
-                      QualType Type, ExprValueKind VK, ExprObjectKind OK,
-                      SourceLocation ColonLocFirst,
-                      SourceLocation ColonLocSecond, SourceLocation RBracketLoc)
-      : Expr(OMPArraySectionExprClass, Type, VK, OK),
-        ColonLocFirst(ColonLocFirst), ColonLocSecond(ColonLocSecond),
-        RBracketLoc(RBracketLoc) {
-    SubExprs[BASE] = Base;
-    SubExprs[LOWER_BOUND] = LowerBound;
-    SubExprs[LENGTH] = Length;
-    SubExprs[STRIDE] = Stride;
-    setDependence(computeDependence(this));
-  }
-
-  /// Create an empty array section expression.
-  explicit OMPArraySectionExpr(EmptyShell Shell)
-      : Expr(OMPArraySectionExprClass, Shell) {}
-
-  /// An array section can be written only as Base[LowerBound:Length].
-
-  /// Get base of the array section.
-  Expr *getBase() { return cast<Expr>(SubExprs[BASE]); }
-  const Expr *getBase() const { return cast<Expr>(SubExprs[BASE]); }
-  /// Set base of the array section.
-  void setBase(Expr *E) { SubExprs[BASE] = E; }
-
-  /// Return original type of the base expression for array section.
-  static QualType getBaseOriginalType(const Expr *Base);
-
-  /// Get lower bound of array section.
-  Expr *getLowerBound() { return cast_or_null<Expr>(SubExprs[LOWER_BOUND]); }
-  const Expr *getLowerBound() const {
-    return cast_or_null<Expr>(SubExprs[LOWER_BOUND]);
-  }
-  /// Set lower bound of the array section.
-  void setLowerBound(Expr *E) { SubExprs[LOWER_BOUND] = E; }
-
-  /// Get length of array section.
-  Expr *getLength() { return cast_or_null<Expr>(SubExprs[LENGTH]); }
-  const Expr *getLength() const { return cast_or_null<Expr>(SubExprs[LENGTH]); }
-  /// Set length of the array section.
-  void setLength(Expr *E) { SubExprs[LENGTH] = E; }
-
-  /// Get stride of array section.
-  Expr *getStride() { return cast_or_null<Expr>(SubExprs[STRIDE]); }
-  const Expr *getStride() const { return cast_or_null<Expr>(SubExprs[STRIDE]); }
-  /// Set length of the array section.
-  void setStride(Expr *E) { SubExprs[STRIDE] = E; }
-
-  SourceLocation getBeginLoc() const LLVM_READONLY {
-    return getBase()->getBeginLoc();
-  }
-  SourceLocation getEndLoc() const LLVM_READONLY { return RBracketLoc; }
-
-  SourceLocation getColonLocFirst() const { return ColonLocFirst; }
-  void setColonLocFirst(SourceLocation L) { ColonLocFirst = L; }
-
-  SourceLocation getColonLocSecond() const { return ColonLocSecond; }
-  void setColonLocSecond(SourceLocation L) { ColonLocSecond = L; }
-
-  SourceLocation getRBracketLoc() const { return RBracketLoc; }
-  void setRBracketLoc(SourceLocation L) { RBracketLoc = L; }
-
-  SourceLocation getExprLoc() const LLVM_READONLY {
-    return getBase()->getExprLoc();
-  }
-
-  static bool classof(const Stmt *T) {
-    return T->getStmtClass() == OMPArraySectionExprClass;
-  }
-
-  child_range children() {
-    return child_range(&SubExprs[BASE], &SubExprs[END_EXPR]);
-  }
-
-  const_child_range children() const {
-    return const_child_range(&SubExprs[BASE], &SubExprs[END_EXPR]);
-  }
-};
-
 /// An explicit cast in C or a C-style cast in C++, which uses the syntax
 /// ([s1][s2]...[sn])expr. For example: @c ([3][3])f.
 class OMPArrayShapingExpr final
diff --git a/clang/include/clang/AST/ExternalASTSource.h b/clang/include/clang/AST/ExternalASTSource.h
index 230c83943c22..385c32edbae0 100644
--- a/clang/include/clang/AST/ExternalASTSource.h
+++ b/clang/include/clang/AST/ExternalASTSource.h
@@ -99,7 +99,7 @@ public:
   /// passes back decl sets as VisibleDeclaration objects.
   ///
   /// The default implementation of this method is a no-op.
-  virtual Decl *GetExternalDecl(uint32_t ID);
+  virtual Decl *GetExternalDecl(GlobalDeclID ID);
 
   /// Resolve a selector ID into a selector.
   ///
@@ -375,7 +375,7 @@ public:
     if (isOffset()) {
       assert(Source &&
              "Cannot deserialize a lazy pointer without an AST source");
-      Ptr = reinterpret_cast<uint64_t>((Source->*Get)(Ptr >> 1));
+      Ptr = reinterpret_cast<uint64_t>((Source->*Get)(OffsT(Ptr >> 1)));
     }
     return reinterpret_cast<T*>(Ptr);
   }
@@ -579,7 +579,7 @@ using LazyDeclStmtPtr =
 
 /// A lazy pointer to a declaration.
 using LazyDeclPtr =
-    LazyOffsetPtr<Decl, uint32_t, &ExternalASTSource::GetExternalDecl>;
+    LazyOffsetPtr<Decl, GlobalDeclID, &ExternalASTSource::GetExternalDecl>;
 
 /// A lazy pointer to a set of CXXCtorInitializers.
 using LazyCXXCtorInitializersPtr =
diff --git a/clang/include/clang/AST/RecursiveASTVisitor.h b/clang/include/clang/AST/RecursiveASTVisitor.h
index 7eb92e304a38..f9b145b4e86a 100644
--- a/clang/include/clang/AST/RecursiveASTVisitor.h
+++ b/clang/include/clang/AST/RecursiveASTVisitor.h
@@ -2740,7 +2740,7 @@ DEF_TRAVERSE_STMT(CXXMemberCallExpr, {})
 DEF_TRAVERSE_STMT(AddrLabelExpr, {})
 DEF_TRAVERSE_STMT(ArraySubscriptExpr, {})
 DEF_TRAVERSE_STMT(MatrixSubscriptExpr, {})
-DEF_TRAVERSE_STMT(OMPArraySectionExpr, {})
+DEF_TRAVERSE_STMT(ArraySectionExpr, {})
 DEF_TRAVERSE_STMT(OMPArrayShapingExpr, {})
 DEF_TRAVERSE_STMT(OMPIteratorExpr, {})
 
diff --git a/clang/include/clang/AST/Type.h b/clang/include/clang/AST/Type.h
index 99f45d518c79..e6643469e0b3 100644
--- a/clang/include/clang/AST/Type.h
+++ b/clang/include/clang/AST/Type.h
@@ -25,8 +25,10 @@
 #include "clang/Basic/Diagnostic.h"
 #include "clang/Basic/ExceptionSpecificationType.h"
 #include "clang/Basic/LLVM.h"
+#include "clang/Basic/LangOptions.h"
 #include "clang/Basic/Linkage.h"
 #include "clang/Basic/PartialDiagnostic.h"
+#include "clang/Basic/PointerAuthOptions.h"
 #include "clang/Basic/SourceLocation.h"
 #include "clang/Basic/Specifiers.h"
 #include "clang/Basic/Visibility.h"
@@ -139,6 +141,174 @@ using CanQualType = CanQual<Type>;
 #define TYPE(Class, Base) class Class##Type;
 #include "clang/AST/TypeNodes.inc"
 
+/// Pointer-authentication qualifiers.
+class PointerAuthQualifier {
+  enum : uint32_t {
+    EnabledShift = 0,
+    EnabledBits = 1,
+    EnabledMask = 1 << EnabledShift,
+    AddressDiscriminatedShift = EnabledShift + EnabledBits,
+    AddressDiscriminatedBits = 1,
+    AddressDiscriminatedMask = 1 << AddressDiscriminatedShift,
+    AuthenticationModeShift =
+        AddressDiscriminatedShift + AddressDiscriminatedBits,
+    AuthenticationModeBits = 2,
+    AuthenticationModeMask = ((1 << AuthenticationModeBits) - 1)
+                             << AuthenticationModeShift,
+    IsaPointerShift = AuthenticationModeShift + AuthenticationModeBits,
+    IsaPointerBits = 1,
+    IsaPointerMask = ((1 << IsaPointerBits) - 1) << IsaPointerShift,
+    AuthenticatesNullValuesShift = IsaPointerShift + IsaPointerBits,
+    AuthenticatesNullValuesBits = 1,
+    AuthenticatesNullValuesMask = ((1 << AuthenticatesNullValuesBits) - 1)
+                                  << AuthenticatesNullValuesShift,
+    KeyShift = AuthenticatesNullValuesShift + AuthenticatesNullValuesBits,
+    KeyBits = 10,
+    KeyMask = ((1 << KeyBits) - 1) << KeyShift,
+    DiscriminatorShift = KeyShift + KeyBits,
+    DiscriminatorBits = 16,
+    DiscriminatorMask = ((1u << DiscriminatorBits) - 1) << DiscriminatorShift,
+  };
+
+  // bits:     |0      |1      |2..3              |4          |
+  //           |Enabled|Address|AuthenticationMode|ISA pointer|
+  // bits:     |5                |6..15|   16...31   |
+  //           |AuthenticatesNull|Key  |Discriminator|
+  uint32_t Data = 0;
+
+  // The following static assertions check that each of the 32 bits is present
+  // exactly in one of the constants.
+  static_assert((EnabledBits + AddressDiscriminatedBits +
+                 AuthenticationModeBits + IsaPointerBits +
+                 AuthenticatesNullValuesBits + KeyBits + DiscriminatorBits) ==
+                    32,
+                "PointerAuthQualifier should be exactly 32 bits");
+  static_assert((EnabledMask + AddressDiscriminatedMask +
+                 AuthenticationModeMask + IsaPointerMask +
+                 AuthenticatesNullValuesMask + KeyMask + DiscriminatorMask) ==
+                    0xFFFFFFFF,
+                "All masks should cover the entire bits");
+  static_assert((EnabledMask ^ AddressDiscriminatedMask ^
+                 AuthenticationModeMask ^ IsaPointerMask ^
+                 AuthenticatesNullValuesMask ^ KeyMask ^ DiscriminatorMask) ==
+                    0xFFFFFFFF,
+                "All masks should cover the entire bits");
+
+  PointerAuthQualifier(unsigned Key, bool IsAddressDiscriminated,
+                       unsigned ExtraDiscriminator,
+                       PointerAuthenticationMode AuthenticationMode,
+                       bool IsIsaPointer, bool AuthenticatesNullValues)
+      : Data(EnabledMask |
+             (IsAddressDiscriminated
+                  ? llvm::to_underlying(AddressDiscriminatedMask)
+                  : 0) |
+             (Key << KeyShift) |
+             (llvm::to_underlying(AuthenticationMode)
+              << AuthenticationModeShift) |
+             (ExtraDiscriminator << DiscriminatorShift) |
+             (IsIsaPointer << IsaPointerShift) |
+             (AuthenticatesNullValues << AuthenticatesNullValuesShift)) {
+    assert(Key <= KeyNoneInternal);
+    assert(ExtraDiscriminator <= MaxDiscriminator);
+    assert((Data == 0) ==
+           (getAuthenticationMode() == PointerAuthenticationMode::None));
+  }
+
+public:
+  enum {
+    KeyNoneInternal = (1u << KeyBits) - 1,
+
+    /// The maximum supported pointer-authentication key.
+    MaxKey = KeyNoneInternal - 1,
+
+    /// The maximum supported pointer-authentication discriminator.
+    MaxDiscriminator = (1u << DiscriminatorBits) - 1
+  };
+
+public:
+  PointerAuthQualifier() = default;
+
+  static PointerAuthQualifier
+  Create(unsigned Key, bool IsAddressDiscriminated, unsigned ExtraDiscriminator,
+         PointerAuthenticationMode AuthenticationMode, bool IsIsaPointer,
+         bool AuthenticatesNullValues) {
+    if (Key == PointerAuthKeyNone)
+      Key = KeyNoneInternal;
+    assert(Key <= KeyNoneInternal && "out-of-range key value");
+    return PointerAuthQualifier(Key, IsAddressDiscriminated, ExtraDiscriminator,
+                                AuthenticationMode, IsIsaPointer,
+                                AuthenticatesNullValues);
+  }
+
+  bool isPresent() const {
+    assert((Data == 0) ==
+           (getAuthenticationMode() == PointerAuthenticationMode::None));
+    return Data != 0;
+  }
+
+  explicit operator bool() const { return isPresent(); }
+
+  unsigned getKey() const {
+    assert(isPresent());
+    return (Data & KeyMask) >> KeyShift;
+  }
+
+  bool hasKeyNone() const { return isPresent() && getKey() == KeyNoneInternal; }
+
+  bool isAddressDiscriminated() const {
+    assert(isPresent());
+    return (Data & AddressDiscriminatedMask) >> AddressDiscriminatedShift;
+  }
+
+  unsigned getExtraDiscriminator() const {
+    assert(isPresent());
+    return (Data >> DiscriminatorShift);
+  }
+
+  PointerAuthenticationMode getAuthenticationMode() const {
+    return PointerAuthenticationMode((Data & AuthenticationModeMask) >>
+                                     AuthenticationModeShift);
+  }
+
+  bool isIsaPointer() const {
+    assert(isPresent());
+    return (Data & IsaPointerMask) >> IsaPointerShift;
+  }
+
+  bool authenticatesNullValues() const {
+    assert(isPresent());
+    return (Data & AuthenticatesNullValuesMask) >> AuthenticatesNullValuesShift;
+  }
+
+  PointerAuthQualifier withoutKeyNone() const {
+    return hasKeyNone() ? PointerAuthQualifier() : *this;
+  }
+
+  friend bool operator==(PointerAuthQualifier Lhs, PointerAuthQualifier Rhs) {
+    return Lhs.Data == Rhs.Data;
+  }
+  friend bool operator!=(PointerAuthQualifier Lhs, PointerAuthQualifier Rhs) {
+    return Lhs.Data != Rhs.Data;
+  }
+
+  bool isEquivalent(PointerAuthQualifier Other) const {
+    return withoutKeyNone() == Other.withoutKeyNone();
+  }
+
+  uint32_t getAsOpaqueValue() const { return Data; }
+
+  // Deserialize pointer-auth qualifiers from an opaque representation.
+  static PointerAuthQualifier fromOpaqueValue(uint32_t Opaque) {
+    PointerAuthQualifier Result;
+    Result.Data = Opaque;
+    assert((Result.Data == 0) ==
+           (Result.getAuthenticationMode() == PointerAuthenticationMode::None));
+    return Result;
+  }
+
+  void Profile(llvm::FoldingSetNodeID &ID) const { ID.AddInteger(Data); }
+};
+
 /// The collection of all-type qualifiers we support.
 /// Clang supports five independent qualifiers:
 /// * C99: const, volatile, and restrict
@@ -147,8 +317,9 @@ using CanQualType = CanQual<Type>;
 /// * Objective C: the GC attributes (none, weak, or strong)
 class Qualifiers {
 public:
-  enum TQ { // NOTE: These flags must be kept in sync with DeclSpec::TQ.
-    Const    = 0x1,
+  enum TQ : uint64_t {
+    // NOTE: These flags must be kept in sync with DeclSpec::TQ.
+    Const = 0x1,
     Restrict = 0x2,
     Volatile = 0x4,
     CVRMask = Const | Volatile | Restrict
@@ -182,7 +353,7 @@ public:
     OCL_Autoreleasing
   };
 
-  enum {
+  enum : uint64_t {
     /// The maximum supported address space number.
     /// 23 bits should be enough for anyone.
     MaxAddressSpace = 0x7fffffu,
@@ -197,16 +368,25 @@ public:
   /// Returns the common set of qualifiers while removing them from
   /// the given sets.
   static Qualifiers removeCommonQualifiers(Qualifiers &L, Qualifiers &R) {
+    Qualifiers Q;
+    PointerAuthQualifier LPtrAuth = L.getPointerAuth();
+    if (LPtrAuth.isPresent() &&
+        LPtrAuth.getKey() != PointerAuthQualifier::KeyNoneInternal &&
+        LPtrAuth == R.getPointerAuth()) {
+      Q.setPointerAuth(LPtrAuth);
+      PointerAuthQualifier Empty;
+      L.setPointerAuth(Empty);
+      R.setPointerAuth(Empty);
+    }
+
     // If both are only CVR-qualified, bit operations are sufficient.
     if (!(L.Mask & ~CVRMask) && !(R.Mask & ~CVRMask)) {
-      Qualifiers Q;
       Q.Mask = L.Mask & R.Mask;
       L.Mask &= ~Q.Mask;
       R.Mask &= ~Q.Mask;
       return Q;
     }
 
-    Qualifiers Q;
     unsigned CommonCRV = L.getCVRQualifiers() & R.getCVRQualifiers();
     Q.addCVRQualifiers(CommonCRV);
     L.removeCVRQualifiers(CommonCRV);
@@ -251,16 +431,14 @@ public:
   }
 
   // Deserialize qualifiers from an opaque representation.
-  static Qualifiers fromOpaqueValue(unsigned opaque) {
+  static Qualifiers fromOpaqueValue(uint64_t opaque) {
     Qualifiers Qs;
     Qs.Mask = opaque;
     return Qs;
   }
 
   // Serialize these qualifiers into an opaque representation.
-  unsigned getAsOpaqueValue() const {
-    return Mask;
-  }
+  uint64_t getAsOpaqueValue() const { return Mask; }
 
   bool hasConst() const { return Mask & Const; }
   bool hasOnlyConst() const { return Mask == Const; }
@@ -302,7 +480,7 @@ public:
   }
   void removeCVRQualifiers(unsigned mask) {
     assert(!(mask & ~CVRMask) && "bitmask contains non-CVR bits");
-    Mask &= ~mask;
+    Mask &= ~static_cast<uint64_t>(mask);
   }
   void removeCVRQualifiers() {
     removeCVRQualifiers(CVRMask);
@@ -407,6 +585,20 @@ public:
     setAddressSpace(space);
   }
 
+  bool hasPointerAuth() const { return Mask & PtrAuthMask; }
+  PointerAuthQualifier getPointerAuth() const {
+    return PointerAuthQualifier::fromOpaqueValue(Mask >> PtrAuthShift);
+  }
+  void setPointerAuth(PointerAuthQualifier Q) {
+    Mask = (Mask & ~PtrAuthMask) |
+           (uint64_t(Q.getAsOpaqueValue()) << PtrAuthShift);
+  }
+  void removePointerAuth() { Mask &= ~PtrAuthMask; }
+  void addPointerAuth(PointerAuthQualifier Q) {
+    assert(Q.isPresent());
+    setPointerAuth(Q);
+  }
+
   // Fast qualifiers are those that can be allocated directly
   // on a QualType object.
   bool hasFastQualifiers() const { return getFastQualifiers(); }
@@ -417,7 +609,7 @@ public:
   }
   void removeFastQualifiers(unsigned mask) {
     assert(!(mask & ~FastMask) && "bitmask contains non-fast qualifier bits");
-    Mask &= ~mask;
+    Mask &= ~static_cast<uint64_t>(mask);
   }
   void removeFastQualifiers() {
     removeFastQualifiers(FastMask);
@@ -454,6 +646,8 @@ public:
         addObjCGCAttr(Q.getObjCGCAttr());
       if (Q.hasObjCLifetime())
         addObjCLifetime(Q.getObjCLifetime());
+      if (Q.hasPointerAuth())
+        addPointerAuth(Q.getPointerAuth());
     }
   }
 
@@ -471,6 +665,8 @@ public:
         removeObjCLifetime();
       if (getAddressSpace() == Q.getAddressSpace())
         removeAddressSpace();
+      if (getPointerAuth() == Q.getPointerAuth())
+        removePointerAuth();
     }
   }
 
@@ -483,6 +679,8 @@ public:
            !hasObjCGCAttr() || !qs.hasObjCGCAttr());
     assert(getObjCLifetime() == qs.getObjCLifetime() ||
            !hasObjCLifetime() || !qs.hasObjCLifetime());
+    assert(!hasPointerAuth() || !qs.hasPointerAuth() ||
+           getPointerAuth() == qs.getPointerAuth());
     Mask |= qs.Mask;
   }
 
@@ -536,6 +734,8 @@ public:
            // be changed.
            (getObjCGCAttr() == other.getObjCGCAttr() || !hasObjCGCAttr() ||
             !other.hasObjCGCAttr()) &&
+           // Pointer-auth qualifiers must match exactly.
+           getPointerAuth() == other.getPointerAuth() &&
            // ObjC lifetime qualifiers must match exactly.
            getObjCLifetime() == other.getObjCLifetime() &&
            // CVR qualifiers may subset.
@@ -605,24 +805,26 @@ public:
   void print(raw_ostream &OS, const PrintingPolicy &Policy,
              bool appendSpaceIfNonEmpty = false) const;
 
-  void Profile(llvm::FoldingSetNodeID &ID) const {
-    ID.AddInteger(Mask);
-  }
+  void Profile(llvm::FoldingSetNodeID &ID) const { ID.AddInteger(Mask); }
 
 private:
-  // bits:     |0 1 2|3|4 .. 5|6  ..  8|9   ...   31|
-  //           |C R V|U|GCAttr|Lifetime|AddressSpace|
-  uint32_t Mask = 0;
-
-  static const uint32_t UMask = 0x8;
-  static const uint32_t UShift = 3;
-  static const uint32_t GCAttrMask = 0x30;
-  static const uint32_t GCAttrShift = 4;
-  static const uint32_t LifetimeMask = 0x1C0;
-  static const uint32_t LifetimeShift = 6;
-  static const uint32_t AddressSpaceMask =
+  // bits:     |0 1 2|3|4 .. 5|6  ..  8|9   ...   31|32 ... 63|
+  //           |C R V|U|GCAttr|Lifetime|AddressSpace| PtrAuth |
+  uint64_t Mask = 0;
+  static_assert(sizeof(PointerAuthQualifier) == sizeof(uint32_t),
+                "PointerAuthQualifier must be 32 bits");
+
+  static constexpr uint64_t UMask = 0x8;
+  static constexpr uint64_t UShift = 3;
+  static constexpr uint64_t GCAttrMask = 0x30;
+  static constexpr uint64_t GCAttrShift = 4;
+  static constexpr uint64_t LifetimeMask = 0x1C0;
+  static constexpr uint64_t LifetimeShift = 6;
+  static constexpr uint64_t AddressSpaceMask =
       ~(CVRMask | UMask | GCAttrMask | LifetimeMask);
-  static const uint32_t AddressSpaceShift = 9;
+  static constexpr uint64_t AddressSpaceShift = 9;
+  static constexpr uint64_t PtrAuthShift = 32;
+  static constexpr uint64_t PtrAuthMask = uint64_t(0xffffffff) << PtrAuthShift;
 };
 
 class QualifiersAndAtomic {
@@ -1242,6 +1444,10 @@ public:
   // true when Type is objc's weak and weak is enabled but ARC isn't.
   bool isNonWeakInMRRWithObjCWeak(const ASTContext &Context) const;
 
+  PointerAuthQualifier getPointerAuth() const {
+    return getQualifiers().getPointerAuth();
+  }
+
   enum PrimitiveDefaultInitializeKind {
     /// The type does not fall into any of the following categories. Note that
     /// this case is zero-valued so that values of this enum can be used as a
@@ -2172,6 +2378,10 @@ public:
   /// 'riscv_rvv_vector_bits' type attribute as VectorType.
   QualType getRVVEltType(const ASTContext &Ctx) const;
 
+  /// Returns the representative type for the element of a sizeless vector
+  /// builtin type.
+  QualType getSizelessVectorEltType(const ASTContext &Ctx) const;
+
   /// Types are partitioned into 3 broad categories (C99 6.2.5p1):
   /// object types, function types, and incomplete types.
 
diff --git a/clang/include/clang/Analysis/FlowSensitive/DataflowEnvironment.h b/clang/include/clang/Analysis/FlowSensitive/DataflowEnvironment.h
index d50dba35f826..cdf89c7def2c 100644
--- a/clang/include/clang/Analysis/FlowSensitive/DataflowEnvironment.h
+++ b/clang/include/clang/Analysis/FlowSensitive/DataflowEnvironment.h
@@ -244,6 +244,21 @@ public:
                           Environment::ValueModel &Model,
                           ExprJoinBehavior ExprBehavior);
 
+  /// Returns a value that approximates both `Val1` and `Val2`, or null if no
+  /// such value can be produced.
+  ///
+  /// `Env1` and `Env2` can be used to query child values and path condition
+  /// implications of `Val1` and `Val2` respectively. The joined value will be
+  /// produced in `JoinedEnv`.
+  ///
+  /// Requirements:
+  ///
+  ///  `Val1` and `Val2` must model values of type `Type`.
+  static Value *joinValues(QualType Ty, Value *Val1, const Environment &Env1,
+                           Value *Val2, const Environment &Env2,
+                           Environment &JoinedEnv,
+                           Environment::ValueModel &Model);
+
   /// Widens the environment point-wise, using `PrevEnv` as needed to inform the
   /// approximation.
   ///
diff --git a/clang/include/clang/Analysis/FlowSensitive/Transfer.h b/clang/include/clang/Analysis/FlowSensitive/Transfer.h
index ed148250d8eb..940025e02100 100644
--- a/clang/include/clang/Analysis/FlowSensitive/Transfer.h
+++ b/clang/include/clang/Analysis/FlowSensitive/Transfer.h
@@ -53,7 +53,8 @@ private:
 /// Requirements:
 ///
 ///  `S` must not be `ParenExpr` or `ExprWithCleanups`.
-void transfer(const StmtToEnvMap &StmtToEnv, const Stmt &S, Environment &Env);
+void transfer(const StmtToEnvMap &StmtToEnv, const Stmt &S, Environment &Env,
+              Environment::ValueModel &Model);
 
 } // namespace dataflow
 } // namespace clang
diff --git a/clang/include/clang/Basic/Attr.td b/clang/include/clang/Basic/Attr.td
index dc87a8c6f022..97e06fe7d2e6 100644
--- a/clang/include/clang/Basic/Attr.td
+++ b/clang/include/clang/Basic/Attr.td
@@ -368,8 +368,8 @@ class Clang<string name, bit allowInC = 1, int version = 1>
   bit AllowInC = allowInC;
 }
 
-// HLSL Semantic spellings
-class HLSLSemantic<string name> : Spelling<name, "HLSLSemantic">;
+// HLSL Annotation spellings
+class HLSLAnnotation<string name> : Spelling<name, "HLSLAnnotation">;
 
 class Accessor<string name, list<Spelling> spellings> {
   string Name = name;
@@ -3211,7 +3211,7 @@ def ObjCRequiresPropertyDefs : InheritableAttr {
 def Unused : InheritableAttr {
   let Spellings = [CXX11<"", "maybe_unused", 201603>, GCC<"unused">,
                    C23<"", "maybe_unused", 202106>];
-  let Subjects = SubjectList<[Var, ObjCIvar, Type, Enum, EnumConstant, Label,
+  let Subjects = SubjectList<[Var, Binding, ObjCIvar, Type, Enum, EnumConstant, Label,
                               Field, ObjCMethod, FunctionLike]>;
   let Documentation = [WarnMaybeUnusedDocs];
 }
@@ -4358,14 +4358,14 @@ def HLSLNumThreads: InheritableAttr {
 }
 
 def HLSLSV_GroupIndex: HLSLAnnotationAttr {
-  let Spellings = [HLSLSemantic<"SV_GroupIndex">];
+  let Spellings = [HLSLAnnotation<"SV_GroupIndex">];
   let Subjects = SubjectList<[ParmVar, GlobalVar]>;
   let LangOpts = [HLSL];
   let Documentation = [HLSLSV_GroupIndexDocs];
 }
 
 def HLSLResourceBinding: InheritableAttr {
-  let Spellings = [HLSLSemantic<"register">];
+  let Spellings = [HLSLAnnotation<"register">];
   let Subjects = SubjectList<[HLSLBufferObj, ExternalGlobalVar]>;
   let LangOpts = [HLSL];
   let Args = [StringArgument<"Slot">, StringArgument<"Space", 1>];
@@ -4373,7 +4373,7 @@ def HLSLResourceBinding: InheritableAttr {
 }
 
 def HLSLSV_DispatchThreadID: HLSLAnnotationAttr {
-  let Spellings = [HLSLSemantic<"SV_DispatchThreadID">];
+  let Spellings = [HLSLAnnotation<"SV_DispatchThreadID">];
   let Subjects = SubjectList<[ParmVar, Field]>;
   let LangOpts = [HLSL];
   let Documentation = [HLSLSV_DispatchThreadIDDocs];
diff --git a/clang/include/clang/Basic/AttributeCommonInfo.h b/clang/include/clang/Basic/AttributeCommonInfo.h
index ef2ddf525c98..5f024b4b5fd7 100644
--- a/clang/include/clang/Basic/AttributeCommonInfo.h
+++ b/clang/include/clang/Basic/AttributeCommonInfo.h
@@ -52,8 +52,8 @@ public:
     /// Context-sensitive version of a keyword attribute.
     AS_ContextSensitiveKeyword,
 
-    /// <vardecl> : <semantic>
-    AS_HLSLSemantic,
+    /// <vardecl> : <annotation>
+    AS_HLSLAnnotation,
 
     /// The attibute has no source code manifestation and is only created
     /// implicitly.
@@ -120,7 +120,7 @@ public:
     }
     static Form Pragma() { return AS_Pragma; }
     static Form ContextSensitiveKeyword() { return AS_ContextSensitiveKeyword; }
-    static Form HLSLSemantic() { return AS_HLSLSemantic; }
+    static Form HLSLAnnotation() { return AS_HLSLAnnotation; }
     static Form Implicit() { return AS_Implicit; }
 
   private:
diff --git a/clang/include/clang/Basic/DiagnosticFrontendKinds.td b/clang/include/clang/Basic/DiagnosticFrontendKinds.td
index 14b08d4927ec..fcffadacc8e6 100644
--- a/clang/include/clang/Basic/DiagnosticFrontendKinds.td
+++ b/clang/include/clang/Basic/DiagnosticFrontendKinds.td
@@ -370,4 +370,7 @@ def warn_missing_symbol_graph_dir : Warning<
   "Missing symbol graph output directory, defaulting to working directory">,
   InGroup<ExtractAPIMisuse>;
 
+def err_ast_action_on_llvm_ir : Error<
+  "cannot apply AST actions to LLVM IR file '%0'">,
+  DefaultFatal;
 }
diff --git a/clang/include/clang/Basic/DiagnosticParseKinds.td b/clang/include/clang/Basic/DiagnosticParseKinds.td
index 38174cf3549f..fdffb35ea0d9 100644
--- a/clang/include/clang/Basic/DiagnosticParseKinds.td
+++ b/clang/include/clang/Basic/DiagnosticParseKinds.td
@@ -478,6 +478,15 @@ def ext_decomp_decl_empty : ExtWarn<
   "ISO C++17 does not allow a decomposition group to be empty">,
   InGroup<DiagGroup<"empty-decomposition">>;
 
+// C++26 structured bindings
+def ext_decl_attrs_on_binding : ExtWarn<
+  "an attribute specifier sequence attached to a structured binding declaration "
+  "is a C++2c extension">, InGroup<CXX26>;
+def warn_cxx23_compat_decl_attrs_on_binding : Warning<
+  "an attribute specifier sequence attached to a structured binding declaration "
+  "is incompatible with C++ standards before C++2c">,
+  InGroup<CXXPre26Compat>, DefaultIgnore;
+
 /// Objective-C parser diagnostics
 def err_expected_minus_or_plus : Error<
   "method type specifier must start with '-' or '+'">;
diff --git a/clang/include/clang/Basic/DiagnosticSemaKinds.td b/clang/include/clang/Basic/DiagnosticSemaKinds.td
index 63e951daec74..f72d5c252b86 100644
--- a/clang/include/clang/Basic/DiagnosticSemaKinds.td
+++ b/clang/include/clang/Basic/DiagnosticSemaKinds.td
@@ -3755,14 +3755,12 @@ def err_sme_definition_using_za_in_non_sme_target : Error<
 def err_sme_definition_using_zt0_in_non_sme2_target : Error<
   "function using ZT0 state requires 'sme2'">;
 def warn_sme_streaming_pass_return_vl_to_non_streaming : Warning<
-  "passing a VL-dependent argument to/from a function that has a different"
-  " streaming-mode. The streaming and non-streaming vector lengths may be"
-  " different">,
+  "%select{returning|passing}0 a VL-dependent argument %select{from|to}0 a function with a different"
+  " streaming-mode is undefined behaviour when the streaming and non-streaming vector lengths are different at runtime">,
   InGroup<AArch64SMEAttributes>, DefaultIgnore;
 def warn_sme_locally_streaming_has_vl_args_returns : Warning<
-  "passing/returning a VL-dependent argument to/from a __arm_locally_streaming"
-  " function. The streaming and non-streaming vector"
-  " lengths may be different">,
+  "%select{returning|passing}0 a VL-dependent argument %select{from|to}0 a locally streaming function is undefined"
+  " behaviour when the streaming and non-streaming vector lengths are different at runtime">,
   InGroup<AArch64SMEAttributes>, DefaultIgnore;
 def err_conflicting_attributes_arm_state : Error<
   "conflicting attributes for state '%0'">;
@@ -9903,6 +9901,9 @@ def warn_format_invalid_annotation : Warning<
 def warn_format_P_no_precision : Warning<
   "using '%%P' format specifier without precision">,
   InGroup<Format>;
+def warn_format_P_with_objc_pointer : Warning<
+  "using '%%P' format specifier with an Objective-C pointer results in dumping runtime object structure, not object value">,
+  InGroup<Format>;
 def warn_printf_ignored_flag: Warning<
   "flag '%0' is ignored when flag '%1' is present">,
   InGroup<Format>;
@@ -9952,6 +9953,8 @@ def warn_ret_stack_addr_ref : Warning<
 def warn_ret_local_temp_addr_ref : Warning<
   "returning %select{address of|reference to}0 local temporary object">,
   InGroup<ReturnStackAddress>;
+def err_ret_local_temp_ref : Error<
+  "returning reference to local temporary object">;
 def warn_ret_addr_label : Warning<
   "returning address of label, which is local">,
   InGroup<ReturnStackAddress>;
@@ -10330,9 +10333,13 @@ def err_shufflevector_nonconstant_argument : Error<
 def err_shufflevector_argument_too_large : Error<
   "index for __builtin_shufflevector must be less than the total number "
   "of vector elements">;
+def err_shufflevector_minus_one_is_undefined_behavior_constexpr : Error<
+  "index for __builtin_shufflevector not within the bounds of the input vectors; index of -1 found at position %0 not permitted in a constexpr context.">;
 
 def err_convertvector_non_vector : Error<
   "first argument to __builtin_convertvector must be a vector">;
+def err_convertvector_constexpr_unsupported_vector_cast : Error<
+  "unsupported vector cast from %0 to %1 in a constant expression.">;
 def err_builtin_non_vector_type : Error<
   "%0 argument to %1 must be of vector type">;
 def err_convertvector_incompatible_vector : Error<
@@ -11163,7 +11170,7 @@ def err_omp_declare_mapper_redefinition : Error<
   "redefinition of user-defined mapper for type %0 with name %1">;
 def err_omp_invalid_mapper: Error<
   "cannot find a valid user-defined mapper for type %0 with name %1">;
-def err_omp_array_section_use : Error<"OpenMP array section is not allowed here">;
+def err_array_section_use : Error<"%select{OpenACC sub-array|OpenMP array section}0 is not allowed here">;
 def err_omp_array_shaping_use : Error<"OpenMP array shaping operation is not allowed here">;
 def err_omp_iterator_use : Error<"OpenMP iterator is not allowed here">;
 def err_omp_typecheck_section_value : Error<
diff --git a/clang/include/clang/Basic/FileManager.h b/clang/include/clang/Basic/FileManager.h
index 2245fd78bfc9..8b4206e52cd4 100644
--- a/clang/include/clang/Basic/FileManager.h
+++ b/clang/include/clang/Basic/FileManager.h
@@ -114,6 +114,12 @@ class FileManager : public RefCountedBase<FileManager> {
   ///
   unsigned NextFileUID;
 
+  /// Statistics gathered during the lifetime of the FileManager.
+  unsigned NumDirLookups = 0;
+  unsigned NumFileLookups = 0;
+  unsigned NumDirCacheMisses = 0;
+  unsigned NumFileCacheMisses = 0;
+
   // Caching.
   std::unique_ptr<FileSystemStatCache> StatCache;
 
@@ -341,6 +347,10 @@ private:
 
 public:
   void PrintStats() const;
+
+  /// Import statistics from a child FileManager and add them to this current
+  /// FileManager.
+  void AddStats(const FileManager &Other);
 };
 
 } // end namespace clang
diff --git a/clang/include/clang/Basic/LangOptions.h b/clang/include/clang/Basic/LangOptions.h
index 75562284ec7d..e2a2aa71b880 100644
--- a/clang/include/clang/Basic/LangOptions.h
+++ b/clang/include/clang/Basic/LangOptions.h
@@ -57,6 +57,13 @@ enum class ShaderStage {
   Invalid,
 };
 
+enum class PointerAuthenticationMode : unsigned {
+  None,
+  Strip,
+  SignAndStrip,
+  SignAndAuth
+};
+
 /// Bitfields of LangOptions, split out from LangOptions in order to ensure that
 /// this large collection of bitfields is a trivial class type.
 class LangOptionsBase {
@@ -878,6 +885,8 @@ public:
   /// Return difference with the given option set.
   FPOptionsOverride getChangesFrom(const FPOptions &Base) const;
 
+  void applyChanges(FPOptionsOverride FPO);
+
   // We can define most of the accessors automatically:
 #define OPTION(NAME, TYPE, WIDTH, PREVIOUS)                                    \
   TYPE get##NAME() const {                                                     \
@@ -959,6 +968,11 @@ public:
       setAllowFPContractAcrossStatement();
   }
 
+  void setDisallowOptimizations() {
+    setFPPreciseEnabled(true);
+    setDisallowFPContract();
+  }
+
   storage_type getAsOpaqueInt() const {
     return (static_cast<storage_type>(Options.getAsOpaqueInt())
             << FPOptions::StorageBitSize) |
@@ -1015,6 +1029,10 @@ inline FPOptionsOverride FPOptions::getChangesFrom(const FPOptions &Base) const
   return getChangesSlow(Base);
 }
 
+inline void FPOptions::applyChanges(FPOptionsOverride FPO) {
+  *this = FPO.applyOverrides(*this);
+}
+
 /// Describes the kind of translation unit being processed.
 enum TranslationUnitKind {
   /// The translation unit is a complete translation unit.
diff --git a/clang/include/clang/Basic/PointerAuthOptions.h b/clang/include/clang/Basic/PointerAuthOptions.h
new file mode 100644
index 000000000000..e5cdcc31ebfb
--- /dev/null
+++ b/clang/include/clang/Basic/PointerAuthOptions.h
@@ -0,0 +1,23 @@
+//===--- PointerAuthOptions.h -----------------------------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+//  This file defines options for configuring pointer-auth technologies
+//  like ARMv8.3.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_CLANG_BASIC_POINTERAUTHOPTIONS_H
+#define LLVM_CLANG_BASIC_POINTERAUTHOPTIONS_H
+
+namespace clang {
+
+constexpr unsigned PointerAuthKeyNone = -1;
+
+} // end namespace clang
+
+#endif
diff --git a/clang/include/clang/Basic/StmtNodes.td b/clang/include/clang/Basic/StmtNodes.td
index b4e3ae573b95..305f19daa4a9 100644
--- a/clang/include/clang/Basic/StmtNodes.td
+++ b/clang/include/clang/Basic/StmtNodes.td
@@ -71,7 +71,7 @@ def OffsetOfExpr : StmtNode<Expr>;
 def UnaryExprOrTypeTraitExpr : StmtNode<Expr>;
 def ArraySubscriptExpr : StmtNode<Expr>;
 def MatrixSubscriptExpr : StmtNode<Expr>;
-def OMPArraySectionExpr : StmtNode<Expr>;
+def ArraySectionExpr : StmtNode<Expr>;
 def OMPIteratorExpr : StmtNode<Expr>;
 def CallExpr : StmtNode<Expr>;
 def MemberExpr : StmtNode<Expr>;
diff --git a/clang/include/clang/Basic/TargetInfo.h b/clang/include/clang/Basic/TargetInfo.h
index e1ef7454f016..3ced2e7397a7 100644
--- a/clang/include/clang/Basic/TargetInfo.h
+++ b/clang/include/clang/Basic/TargetInfo.h
@@ -40,6 +40,7 @@
 #include <cassert>
 #include <optional>
 #include <string>
+#include <utility>
 #include <vector>
 
 namespace llvm {
@@ -1792,6 +1793,15 @@ public:
   /// Whether to support HIP image/texture API's.
   virtual bool hasHIPImageSupport() const { return true; }
 
+  /// The first value in the pair is the minimum offset between two objects to
+  /// avoid false sharing (destructive interference). The second value in the
+  /// pair is maximum size of contiguous memory to promote true sharing
+  /// (constructive interference). Neither of these values are considered part
+  /// of the ABI and can be changed by targets at any time.
+  virtual std::pair<unsigned, unsigned> hardwareInterferenceSizes() const {
+    return std::make_pair(64, 64);
+  }
+
 protected:
   /// Copy type and layout related info.
   void copyAuxTarget(const TargetInfo *Aux);
diff --git a/clang/include/clang/Basic/arm_neon.td b/clang/include/clang/Basic/arm_neon.td
index 6d655c39360d..6390ba3f9fe5 100644
--- a/clang/include/clang/Basic/arm_neon.td
+++ b/clang/include/clang/Basic/arm_neon.td
@@ -275,7 +275,7 @@ def OP_VCVT_BF16_F32_HI_A32
                            (call "vget_low", $p0))>;
 
 def OP_CVT_F32_BF16
-    : Op<(bitcast "R", (op "<<", (bitcast "int32_t", $p0),
+    : Op<(bitcast "R", (op "<<", (cast "int32_t", (bitcast "int16_t", $p0)),
                                  (literal "int32_t", "16")))>;
 
 //===----------------------------------------------------------------------===//
diff --git a/clang/include/clang/Basic/arm_sve.td b/clang/include/clang/Basic/arm_sve.td
index 6cc249837d3f..15340ebb62b3 100644
--- a/clang/include/clang/Basic/arm_sve.td
+++ b/clang/include/clang/Basic/arm_sve.td
@@ -1961,19 +1961,20 @@ def SVPSEL_D : SInst<"svpsel_lane_b64", "PPPm", "Pl", MergeNone, "", [IsStreamin
 
 // Standalone sve2.1 builtins
 let TargetGuard = "sve2p1" in {
-def SVORQV   : SInst<"svorqv[_{d}]", "{Pd", "csilUcUsUiUl", MergeNone, "aarch64_sve_orqv", [IsReductionQV]>;
-def SVEORQV  : SInst<"sveorqv[_{d}]", "{Pd", "csilUcUsUiUl", MergeNone, "aarch64_sve_eorqv", [IsReductionQV]>;
-def SVADDQV  : SInst<"svaddqv[_{d}]", "{Pd", "hfdcsilUcUsUiUl", MergeNone, "aarch64_sve_addqv", [IsReductionQV]>;
-def SVANDQV  : SInst<"svandqv[_{d}]", "{Pd", "csilUcUsUiUl", MergeNone, "aarch64_sve_andqv", [IsReductionQV]>;
-def SVSMAXQV : SInst<"svmaxqv[_{d}]", "{Pd", "csil", MergeNone, "aarch64_sve_smaxqv", [IsReductionQV]>;
-def SVUMAXQV : SInst<"svmaxqv[_{d}]", "{Pd", "UcUsUiUl", MergeNone, "aarch64_sve_umaxqv", [IsReductionQV]>;
-def SVSMINQV : SInst<"svminqv[_{d}]", "{Pd", "csil", MergeNone, "aarch64_sve_sminqv", [IsReductionQV]>;
-def SVUMINQV : SInst<"svminqv[_{d}]", "{Pd", "UcUsUiUl", MergeNone, "aarch64_sve_uminqv", [IsReductionQV]>;
-
-def SVFMAXNMQV: SInst<"svmaxnmqv[_{d}]", "{Pd", "hfd", MergeNone, "aarch64_sve_fmaxnmqv", [IsReductionQV]>;
-def SVFMINNMQV: SInst<"svminnmqv[_{d}]", "{Pd", "hfd", MergeNone, "aarch64_sve_fminnmqv", [IsReductionQV]>;
-def SVFMAXQV: SInst<"svmaxqv[_{d}]", "{Pd", "hfd", MergeNone, "aarch64_sve_fmaxqv", [IsReductionQV]>;
-def SVFMINQV: SInst<"svminqv[_{d}]", "{Pd", "hfd", MergeNone, "aarch64_sve_fminqv", [IsReductionQV]>;
+def SVORQV   : SInst<"svorqv[_{d}]",  "{Pd", "csilUcUsUiUl", MergeNone, "aarch64_sve_orqv",   [IsReductionQV]>;
+def SVEORQV  : SInst<"sveorqv[_{d}]", "{Pd", "csilUcUsUiUl", MergeNone, "aarch64_sve_eorqv",  [IsReductionQV]>;
+def SVADDQV  : SInst<"svaddqv[_{d}]", "{Pd", "csilUcUsUiUl", MergeNone, "aarch64_sve_addqv",  [IsReductionQV]>;
+def SVANDQV  : SInst<"svandqv[_{d}]", "{Pd", "csilUcUsUiUl", MergeNone, "aarch64_sve_andqv",  [IsReductionQV]>;
+def SVSMAXQV : SInst<"svmaxqv[_{d}]", "{Pd", "csil",         MergeNone, "aarch64_sve_smaxqv", [IsReductionQV]>;
+def SVUMAXQV : SInst<"svmaxqv[_{d}]", "{Pd", "UcUsUiUl",     MergeNone, "aarch64_sve_umaxqv", [IsReductionQV]>;
+def SVSMINQV : SInst<"svminqv[_{d}]", "{Pd", "csil",         MergeNone, "aarch64_sve_sminqv", [IsReductionQV]>;
+def SVUMINQV : SInst<"svminqv[_{d}]", "{Pd", "UcUsUiUl",     MergeNone, "aarch64_sve_uminqv", [IsReductionQV]>;
+
+def SVFADDQV   : SInst<"svaddqv[_{d}]",   "{Pd", "hfd", MergeNone, "aarch64_sve_faddqv",   [IsReductionQV]>;
+def SVFMAXNMQV : SInst<"svmaxnmqv[_{d}]", "{Pd", "hfd", MergeNone, "aarch64_sve_fmaxnmqv", [IsReductionQV]>;
+def SVFMINNMQV : SInst<"svminnmqv[_{d}]", "{Pd", "hfd", MergeNone, "aarch64_sve_fminnmqv", [IsReductionQV]>;
+def SVFMAXQV   : SInst<"svmaxqv[_{d}]",   "{Pd", "hfd", MergeNone, "aarch64_sve_fmaxqv",   [IsReductionQV]>;
+def SVFMINQV   : SInst<"svminqv[_{d}]",   "{Pd", "hfd", MergeNone, "aarch64_sve_fminqv",   [IsReductionQV]>;
 }
 
 let TargetGuard = "sve2p1|sme2" in {
diff --git a/clang/include/clang/Basic/riscv_vector.td b/clang/include/clang/Basic/riscv_vector.td
index 76ed544f3b2b..cca4367751b9 100644
--- a/clang/include/clang/Basic/riscv_vector.td
+++ b/clang/include/clang/Basic/riscv_vector.td
@@ -14,14 +14,14 @@
 
 include "riscv_vector_common.td"
 
-defvar TypeList = ["c","s","i","l","x","f","d"];
+defvar TypeList = ["c","s","i","l","x","f","d","y"];
 defvar EEWList = [["8", "(Log2EEW:3)"],
                   ["16", "(Log2EEW:4)"],
                   ["32", "(Log2EEW:5)"],
                   ["64", "(Log2EEW:6)"]];
 
 class IsFloat<string type> {
-  bit val = !or(!eq(type, "x"), !eq(type, "f"), !eq(type, "d"));
+  bit val = !or(!eq(type, "x"), !eq(type, "f"), !eq(type, "d"), !eq(type, "y"));
 }
 
 let SupportOverloading = false,
@@ -118,7 +118,8 @@ multiclass RVVIndexedLoad<string op> {
         defvar eew_type = eew_list[1];
         let Name = op # eew # "_v", IRName = op, MaskedIRName = op # "_mask",
             RequiredFeatures = !if(!eq(type, "x"), ["Zvfhmin"],
-                                                   []<string>) in {
+                                   !if(!eq(type, "y"), ["Zvfbfmin"],
+                                                   []<string>)) in {
           def: RVVOutOp1Builtin<"v", "vPCe" # eew_type # "Uv", type>;
             if !not(IsFloat<type>.val) then {
               def: RVVOutOp1Builtin<"Uv", "UvPCUe" # eew_type # "Uv", type>;
@@ -129,7 +130,8 @@ multiclass RVVIndexedLoad<string op> {
       defvar eew64_type = "(Log2EEW:6)";
       let Name = op # eew64 # "_v", IRName = op, MaskedIRName = op # "_mask",
           RequiredFeatures = !if(!eq(type, "x"), ["Zvfhmin", "RV64"],
-                                                 ["RV64"]) in {
+                                 !if(!eq(type, "y"), ["Zvfbfmin", "RV64"],
+                                                 ["RV64"])) in {
           def: RVVOutOp1Builtin<"v", "vPCe" # eew64_type # "Uv", type>;
             if !not(IsFloat<type>.val) then {
               def: RVVOutOp1Builtin<"Uv", "UvPCUe" # eew64_type # "Uv", type>;
@@ -223,7 +225,8 @@ multiclass RVVIndexedStore<string op> {
           defvar eew_type = eew_list[1];
           let Name = op # eew  # "_v", IRName = op, MaskedIRName = op # "_mask",
           RequiredFeatures = !if(!eq(type, "x"), ["Zvfhmin"],
-                                                 []<string>) in  {
+                                 !if(!eq(type, "y"), ["Zvfbfmin"],
+                                                 []<string>)) in  {
             def : RVVBuiltin<"v", "0Pe" # eew_type # "Uvv", type>;
             if !not(IsFloat<type>.val) then {
               def : RVVBuiltin<"Uv", "0PUe" # eew_type # "UvUv", type>;
@@ -234,7 +237,8 @@ multiclass RVVIndexedStore<string op> {
         defvar eew64_type = "(Log2EEW:6)";
         let Name = op # eew64  # "_v", IRName = op, MaskedIRName = op # "_mask",
             RequiredFeatures = !if(!eq(type, "x"), ["Zvfhmin", "RV64"],
-                                                   ["RV64"]) in  {
+                                   !if(!eq(type, "y"), ["Zvfbfmin", "RV64"],
+                                                   ["RV64"])) in  {
           def : RVVBuiltin<"v", "0Pe" # eew64_type # "Uvv", type>;
           if !not(IsFloat<type>.val) then {
             def : RVVBuiltin<"Uv", "0PUe" # eew64_type # "UvUv", type>;
@@ -358,6 +362,10 @@ multiclass RVVNonTupleVCreateBuiltin<int dst_lmul, list<int> src_lmul_list> {
     def vcreate # src_v # dst_v : RVVBuiltin<src_v # dst_v,
                                              dst_v # src_s,
                                              "csilxfd">;
+    let RequiredFeatures = ["Zvfbfmin"] in
+      def vcreate_bf16 # src_v # dst_v : RVVBuiltin<src_v # dst_v,
+                                                    dst_v # src_s,
+                                                    "y", dst_v>;
 
     defvar src_uv = FixedVString<src_lmul, num, "Uv">.V;
     defvar src_us = FixedVString<src_lmul, num, "Uv">.S;
@@ -683,6 +691,8 @@ defm vle8: RVVVLEBuiltin<["c"]>;
 defm vle16: RVVVLEBuiltin<["s"]>;
 let Name = "vle16_v", RequiredFeatures = ["Zvfhmin"] in
   defm vle16_h: RVVVLEBuiltin<["x"]>;
+let Name = "vle16_v", RequiredFeatures = ["Zvfbfmin"] in
+  defm vle16_bf16 : RVVVLEBuiltin<["y"]>;
 defm vle32: RVVVLEBuiltin<["i","f"]>;
 defm vle64: RVVVLEBuiltin<["l","d"]>;
 
@@ -691,6 +701,8 @@ defm vse8 : RVVVSEBuiltin<["c"]>;
 defm vse16: RVVVSEBuiltin<["s"]>;
 let Name = "vse16_v", RequiredFeatures = ["Zvfhmin"] in
   defm vse16_h: RVVVSEBuiltin<["x"]>;
+let Name = "vse16_v", RequiredFeatures = ["Zvfbfmin"] in
+  defm vse16_bf16: RVVVSEBuiltin<["y"]>;
 defm vse32: RVVVSEBuiltin<["i","f"]>;
 defm vse64: RVVVSEBuiltin<["l","d"]>;
 
@@ -699,6 +711,8 @@ defm vlse8: RVVVLSEBuiltin<["c"]>;
 defm vlse16: RVVVLSEBuiltin<["s"]>;
 let Name = "vlse16_v", RequiredFeatures = ["Zvfhmin"] in
   defm vlse16_h: RVVVLSEBuiltin<["x"]>;
+let Name = "vlse16_v", RequiredFeatures = ["Zvfbfmin"] in
+  defm vlse16_bf16: RVVVLSEBuiltin<["y"]>;
 defm vlse32: RVVVLSEBuiltin<["i","f"]>;
 defm vlse64: RVVVLSEBuiltin<["l","d"]>;
 
@@ -706,6 +720,8 @@ defm vsse8 : RVVVSSEBuiltin<["c"]>;
 defm vsse16: RVVVSSEBuiltin<["s"]>;
 let Name = "vsse16_v", RequiredFeatures = ["Zvfhmin"] in
   defm vsse16_h: RVVVSSEBuiltin<["x"]>;
+let Name = "vsse16_v", RequiredFeatures = ["Zvfbfmin"] in
+  defm vsse16_bf: RVVVSSEBuiltin<["y"]>;
 defm vsse32: RVVVSSEBuiltin<["i","f"]>;
 defm vsse64: RVVVSSEBuiltin<["l","d"]>;
 
@@ -721,6 +737,8 @@ defm vle8ff: RVVVLEFFBuiltin<["c"]>;
 defm vle16ff: RVVVLEFFBuiltin<["s"]>;
 let Name = "vle16ff_v", RequiredFeatures = ["Zvfhmin"] in
   defm vle16ff: RVVVLEFFBuiltin<["x"]>;
+let Name = "vle16ff_v", RequiredFeatures = ["Zvfbfmin"] in
+  defm vle16ff: RVVVLEFFBuiltin<["y"]>;
 defm vle32ff: RVVVLEFFBuiltin<["i", "f"]>;
 defm vle64ff: RVVVLEFFBuiltin<["l", "d"]>;
 
@@ -732,14 +750,16 @@ multiclass RVVUnitStridedSegLoadTuple<string op> {
                        !eq(type, "l") : "64",
                        !eq(type, "x") : "16",
                        !eq(type, "f") : "32",
-                       !eq(type, "d") : "64");
+                       !eq(type, "d") : "64",
+                       !eq(type, "y") : "16");
       foreach nf = NFList in {
         let Name = op # nf # "e" # eew # "_v",
             IRName = op # nf,
             MaskedIRName = op # nf # "_mask",
             NF = nf,
             RequiredFeatures = !if(!eq(type, "x"), ["Zvfhmin"],
-                                                   []<string>),
+                                   !if(!eq(type, "y"), ["Zvfbfmin"],
+                                                   []<string>)),
             ManualCodegen = [{
     {
       llvm::Type *ElementVectorType = cast<StructType>(ResultType)->elements()[0];
@@ -793,7 +813,8 @@ multiclass RVVUnitStridedSegStoreTuple<string op> {
                        !eq(type, "l") : "64",
                        !eq(type, "x") : "16",
                        !eq(type, "f") : "32",
-                       !eq(type, "d") : "64");
+                       !eq(type, "d") : "64",
+                       !eq(type, "y") : "16");
       foreach nf = NFList in {
       let Name = op # nf # "e" # eew # "_v",
           IRName = op # nf,
@@ -801,7 +822,8 @@ multiclass RVVUnitStridedSegStoreTuple<string op> {
           NF = nf,
           HasMaskedOffOperand = false,
           RequiredFeatures = !if(!eq(type, "x"), ["Zvfhmin"],
-                                                 []<string>),
+                                 !if(!eq(type, "y"), ["Zvfbfmin"],
+                                                 []<string>)),
           ManualCodegen = [{
     {
       // Masked
@@ -846,14 +868,16 @@ multiclass RVVUnitStridedSegLoadFFTuple<string op> {
                        !eq(type, "l") : "64",
                        !eq(type, "x") : "16",
                        !eq(type, "f") : "32",
-                       !eq(type, "d") : "64");
+                       !eq(type, "d") : "64",
+                       !eq(type, "y") : "16");
       foreach nf = NFList in {
         let Name = op # nf # "e" # eew # "ff_v",
             IRName = op # nf # "ff",
             MaskedIRName = op # nf # "ff_mask",
             NF = nf,
             RequiredFeatures = !if(!eq(type, "x"), ["Zvfhmin"],
-                                                   []<string>),
+                                   !if(!eq(type, "y"), ["Zvfbfmin"],
+                                                   []<string>)),
             ManualCodegen = [{
     {
       llvm::Type *ElementVectorType = cast<StructType>(ResultType)->elements()[0];
@@ -921,14 +945,16 @@ multiclass RVVStridedSegLoadTuple<string op> {
                        !eq(type, "l") : "64",
                        !eq(type, "x") : "16",
                        !eq(type, "f") : "32",
-                       !eq(type, "d") : "64");
+                       !eq(type, "d") : "64",
+                       !eq(type, "y") : "16");
       foreach nf = NFList in {
         let Name = op # nf # "e" # eew # "_v",
             IRName = op # nf,
             MaskedIRName = op # nf # "_mask",
             NF = nf,
             RequiredFeatures = !if(!eq(type, "x"), ["Zvfhmin"],
-                                                   []<string>),
+                                   !if(!eq(type, "y"), ["Zvfbfmin"],
+                                                   []<string>)),
             ManualCodegen = [{
     {
       llvm::Type *ElementVectorType = cast<StructType>(ResultType)->elements()[0];
@@ -983,7 +1009,8 @@ multiclass RVVStridedSegStoreTuple<string op> {
                        !eq(type, "l") : "64",
                        !eq(type, "x") : "16",
                        !eq(type, "f") : "32",
-                       !eq(type, "d") : "64");
+                       !eq(type, "d") : "64",
+                       !eq(type, "y") : "16");
       foreach nf = NFList in {
         let Name = op # nf # "e" # eew # "_v",
             IRName = op # nf,
@@ -992,7 +1019,8 @@ multiclass RVVStridedSegStoreTuple<string op> {
             HasMaskedOffOperand = false,
             MaskedPolicyScheme = NonePolicy,
             RequiredFeatures = !if(!eq(type, "x"), ["Zvfhmin"],
-                                                   []<string>),
+                                   !if(!eq(type, "y"), ["Zvfbfmin"],
+                                                   []<string>)),
             ManualCodegen = [{
     {
       // Masked
@@ -1041,7 +1069,8 @@ multiclass RVVIndexedSegLoadTuple<string op> {
             MaskedIRName = op # nf # "_mask",
             NF = nf,
             RequiredFeatures = !if(!eq(type, "x"), ["Zvfhmin"],
-                                                   []<string>),
+                                   !if(!eq(type, "y"), ["Zvfbfmin"],
+                                                   []<string>)),
             ManualCodegen = [{
     {
       llvm::Type *ElementVectorType = cast<StructType>(ResultType)->elements()[0];
@@ -1104,7 +1133,8 @@ multiclass RVVIndexedSegStoreTuple<string op> {
             HasMaskedOffOperand = false,
             MaskedPolicyScheme = NonePolicy,
             RequiredFeatures = !if(!eq(type, "x"), ["Zvfhmin"],
-                                                   []<string>),
+                                   !if(!eq(type, "y"), ["Zvfbfmin"],
+                                                   []<string>)),
             ManualCodegen = [{
     {
       // Masked
@@ -2308,6 +2338,12 @@ let HasMasked = false, HasVL = false, IRName = "" in {
       def vreinterpret_h_i : RVVBuiltin<"vFv", "Fvv", "s", "Fv">;
       def vreinterpret_h_u : RVVBuiltin<"UvFv", "FvUv", "s", "Fv">;
     }
+    let RequiredFeatures = ["Zvfbfmin"] in {
+      def vreinterpret_i_bf16 : RVVBuiltin<"vIv", "Ivv", "y", "Iv">;
+      def vreinterpret_u_bf16 : RVVBuiltin<"vUv", "Uvv", "y", "Uv">;
+      def vreinterpret_bf16_i : RVVBuiltin<"Ivv", "vIv", "y", "v">;
+      def vreinterpret_bf16_u : RVVBuiltin<"Uvv", "vUv", "y", "v">;
+    }
 
     // Reinterpret between different SEW under the same LMUL
     foreach dst_sew = ["(FixedSEW:8)", "(FixedSEW:16)", "(FixedSEW:32)",
@@ -2372,12 +2408,16 @@ let HasMasked = false, HasVL = false, IRName = "" in {
         return llvm::PoisonValue::get(ResultType);
       }] in {
     def vundefined : RVVBuiltin<"v", "v", "csilxfd">;
+    let RequiredFeatures = ["Zvfbfmin"] in
+      def vundefined_bf16 : RVVBuiltin<"v", "v", "y">;
     def vundefined_u : RVVBuiltin<"Uv", "Uv", "csil">;
 
     foreach nf = NFList in {
       let NF = nf in {
         defvar T = "(Tuple:" # nf # ")";
         def : RVVBuiltin<T # "v", T # "v", "csilxfd">;
+        let RequiredFeatures = ["Zvfbfmin"] in
+          def : RVVBuiltin<T # "v", T # "v", "y">;
         def : RVVBuiltin<T # "Uv", T # "Uv", "csil">;
       }
     }
@@ -2396,6 +2436,9 @@ let HasMasked = false, HasVL = false, IRName = "" in {
                         "(SFixedLog2LMUL:0)", "(SFixedLog2LMUL:1)", "(SFixedLog2LMUL:2)"] in {
       def vlmul_trunc # dst_lmul : RVVBuiltin<"v" # dst_lmul # "v",
                                               dst_lmul # "vv", "csilxfd", dst_lmul # "v">;
+      let RequiredFeatures = ["Zvfbfmin"] in
+        def vlmul_trunc_bf16 # dst_lmul : RVVBuiltin<"v" # dst_lmul # "v",
+                                                     dst_lmul # "vv", "y", dst_lmul # "v">;
       def vlmul_trunc_u # dst_lmul : RVVBuiltin<"Uv" # dst_lmul # "Uv",
                                                 dst_lmul # "UvUv", "csil", dst_lmul # "Uv">;
     }
@@ -2414,6 +2457,9 @@ let HasMasked = false, HasVL = false, IRName = "" in {
                         "(LFixedLog2LMUL:1)", "(LFixedLog2LMUL:2)", "(LFixedLog2LMUL:3)"] in {
       def vlmul_ext # dst_lmul : RVVBuiltin<"v" # dst_lmul # "v",
                                             dst_lmul # "vv", "csilxfd", dst_lmul # "v">;
+      let RequiredFeatures = ["Zvfbfmin"] in
+        def vlmul_ext_bf16 # dst_lmul : RVVBuiltin<"v" # dst_lmul # "v",
+                                                   dst_lmul # "vv", "y", dst_lmul # "v">;
       def vlmul_ext_u # dst_lmul : RVVBuiltin<"Uv" # dst_lmul # "Uv",
                                               dst_lmul # "UvUv", "csil", dst_lmul # "Uv">;
     }
@@ -2441,12 +2487,12 @@ let HasMasked = false, HasVL = false, IRName = "" in {
       }
       }] in {
     foreach dst_lmul = ["(SFixedLog2LMUL:0)", "(SFixedLog2LMUL:1)", "(SFixedLog2LMUL:2)"] in {
-      def : RVVBuiltin<"v" # dst_lmul # "v", dst_lmul # "vvKz", "csilxfd", dst_lmul # "v">;
+      def : RVVBuiltin<"v" # dst_lmul # "v", dst_lmul # "vvKz", "csilxfdy", dst_lmul # "v">;
       def : RVVBuiltin<"Uv" # dst_lmul # "Uv", dst_lmul # "UvUvKz", "csil", dst_lmul # "Uv">;
     }
     foreach nf = NFList in {
       defvar T = "(Tuple:" # nf # ")";
-      def : RVVBuiltin<T # "vv", "v" # T # "vKz", "csilxfd", "v">;
+      def : RVVBuiltin<T # "vv", "v" # T # "vKz", "csilxfdy", "v">;
       def : RVVBuiltin<T # "UvUv", "Uv" # T # "UvKz", "csil", "Uv">;
     }
   }
@@ -2474,11 +2520,15 @@ let HasMasked = false, HasVL = false, IRName = "" in {
       }] in {
     foreach dst_lmul = ["(LFixedLog2LMUL:1)", "(LFixedLog2LMUL:2)", "(LFixedLog2LMUL:3)"] in {
       def : RVVBuiltin<"v" # dst_lmul # "v", dst_lmul # "v" # dst_lmul # "vKzv", "csilxfd">;
+      let RequiredFeatures = ["Zvfbfmin"] in
+        def : RVVBuiltin<"v" # dst_lmul # "v", dst_lmul # "v" # dst_lmul # "vKzv", "y">;
       def : RVVBuiltin<"Uv" # dst_lmul # "Uv", dst_lmul # "Uv" # dst_lmul #"UvKzUv", "csil">;
     }
     foreach nf = NFList in {
       defvar T = "(Tuple:" # nf # ")";
       def : RVVBuiltin<"v" # T # "v", T # "v" # T # "vKzv", "csilxfd">;
+      let RequiredFeatures = ["Zvfbfmin"] in
+        def : RVVBuiltin<"v" # T # "v", T # "v" # T # "vKzv", "y">;
       def : RVVBuiltin<"Uv" # T # "Uv", T # "Uv" # T # "UvKzUv", "csil">;
     }
   }
@@ -2523,7 +2573,9 @@ let HasMasked = false, HasVL = false, IRName = "" in {
         defvar T = "(Tuple:" # nf # ")";
         defvar V = VString<nf, /*signed=*/true>.S;
         defvar UV = VString<nf, /*signed=*/false>.S;
-        def : RVVBuiltin<T # "v", T # "v" # V, "csilxfd">;
+        def : RVVBuiltin<T # "v", T # "v" # V, "csilxfdy">;
+        let RequiredFeatures = ["Zvfbfmin"] in
+          def : RVVBuiltin<T # "v", T # "v" # V, "y">;
         def : RVVBuiltin<T # "Uv", T # "Uv" # UV, "csil">;
       }
     }
@@ -2549,8 +2601,7 @@ multiclass RVVOutBuiltinSetZvk<bit HasVV = 1, bit HasVS = 1> {
 
   if HasVS then {
     foreach vs2_lmul = ["(SEFixedLog2LMUL:-1)", "(SEFixedLog2LMUL:0)",
-                        "(SEFixedLog2LMUL:1)", "(SEFixedLog2LMUL:2)",
-                        "(SEFixedLog2LMUL:3)"] in {
+                        "(SEFixedLog2LMUL:1)", "(SEFixedLog2LMUL:2)"] in {
     defvar name = NAME # !if(!eq(NAME, "vaesz"), "", "_vs");
     let OverloadedName = name, IRName = NAME # "_vs", Name = NAME # "_vs",
         IntrinsicTypes = [-1, 1] in
diff --git a/clang/include/clang/CIR/CMakeLists.txt b/clang/include/clang/CIR/CMakeLists.txt
index e69de29bb2d1..f8d6f407a03d 100644
--- a/clang/include/clang/CIR/CMakeLists.txt
+++ b/clang/include/clang/CIR/CMakeLists.txt
@@ -0,0 +1,6 @@
+set(MLIR_INCLUDE_DIR ${LLVM_MAIN_SRC_DIR}/../mlir/include ) # --includedir
+set(MLIR_TABLEGEN_OUTPUT_DIR ${CMAKE_BINARY_DIR}/tools/mlir/include)
+include_directories(${MLIR_INCLUDE_DIR})
+include_directories(${MLIR_TABLEGEN_OUTPUT_DIR})
+
+add_subdirectory(Dialect)
diff --git a/clang/include/clang/CIR/Dialect/CMakeLists.txt b/clang/include/clang/CIR/Dialect/CMakeLists.txt
new file mode 100644
index 000000000000..f33061b2d87c
--- /dev/null
+++ b/clang/include/clang/CIR/Dialect/CMakeLists.txt
@@ -0,0 +1 @@
+add_subdirectory(IR)
diff --git a/clang/include/clang/CIR/Dialect/IR/CIRDialect.h b/clang/include/clang/CIR/Dialect/IR/CIRDialect.h
new file mode 100644
index 000000000000..d53e5d1663d6
--- /dev/null
+++ b/clang/include/clang/CIR/Dialect/IR/CIRDialect.h
@@ -0,0 +1,16 @@
+//===- CIRDialect.h - CIR dialect -------------------------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file declares the CIR dialect.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_CLANG_CIR_DIALECT_IR_CIRDIALECT_H
+#define LLVM_CLANG_CIR_DIALECT_IR_CIRDIALECT_H
+
+#endif // LLVM_CLANG_CIR_DIALECT_IR_CIRDIALECT_H
diff --git a/clang/include/clang/CIR/Dialect/IR/CIRDialect.td b/clang/include/clang/CIR/Dialect/IR/CIRDialect.td
new file mode 100644
index 000000000000..69d6e9774942
--- /dev/null
+++ b/clang/include/clang/CIR/Dialect/IR/CIRDialect.td
@@ -0,0 +1,44 @@
+//===- CIRDialect.td - CIR dialect -------------------------*- tablegen -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file declares the CIR dialect.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_CLANG_CIR_DIALECT_IR_CIRDIALECT
+#define LLVM_CLANG_CIR_DIALECT_IR_CIRDIALECT
+
+include "mlir/IR/OpBase.td"
+
+def CIR_Dialect : Dialect {
+  let name = "cir";
+
+  // A short one-line summary of our dialect.
+  let summary = "A high-level dialect for analyzing and optimizing Clang "
+                "supported languages";
+
+  let cppNamespace = "::mlir::cir";
+
+  let useDefaultAttributePrinterParser = 0;
+  let useDefaultTypePrinterParser = 0;
+
+  let extraClassDeclaration = [{
+    void registerAttributes();
+    void registerTypes();
+
+    Type parseType(DialectAsmParser &parser) const override;
+    void printType(Type type, DialectAsmPrinter &printer) const override;
+
+    Attribute parseAttribute(DialectAsmParser &parser,
+                             Type type) const override;
+
+    void printAttribute(Attribute attr, DialectAsmPrinter &os) const override;
+  }];
+}
+
+#endif // LLVM_CLANG_CIR_DIALECT_IR_CIRDIALECT
diff --git a/clang/include/clang/CIR/Dialect/IR/CIROps.td b/clang/include/clang/CIR/Dialect/IR/CIROps.td
new file mode 100644
index 000000000000..7311c8db783e
--- /dev/null
+++ b/clang/include/clang/CIR/Dialect/IR/CIROps.td
@@ -0,0 +1,19 @@
+//===-- CIROps.td - CIR dialect definition -----------------*- tablegen -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+///
+/// \file
+/// Definition of the CIR dialect
+///
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_CLANG_CIR_DIALECT_IR_CIROPS
+#define LLVM_CLANG_CIR_DIALECT_IR_CIROPS
+
+include "clang/CIR/Dialect/IR/CIRDialect.td"
+
+#endif // LLVM_CLANG_CIR_DIALECT_IR_CIROPS
diff --git a/clang/include/clang/CIR/Dialect/IR/CMakeLists.txt b/clang/include/clang/CIR/Dialect/IR/CMakeLists.txt
new file mode 100644
index 000000000000..28ae30dab8df
--- /dev/null
+++ b/clang/include/clang/CIR/Dialect/IR/CMakeLists.txt
@@ -0,0 +1,16 @@
+# This replicates part of the add_mlir_dialect cmake function from MLIR that
+# cannot be used here. This happens because it expects to be run inside MLIR
+# directory which is not the case for CIR (and also FIR, both have similar
+# workarounds).
+
+# Equivalent to add_mlir_dialect(CIROps cir)
+set(LLVM_TARGET_DEFINITIONS CIROps.td)
+mlir_tablegen(CIROps.h.inc -gen-op-decls)
+mlir_tablegen(CIROps.cpp.inc -gen-op-defs)
+mlir_tablegen(CIROpsTypes.h.inc -gen-typedef-decls)
+mlir_tablegen(CIROpsTypes.cpp.inc -gen-typedef-defs)
+mlir_tablegen(CIROpsDialect.h.inc -gen-dialect-decls)
+mlir_tablegen(CIROpsDialect.cpp.inc -gen-dialect-defs)
+add_public_tablegen_target(MLIRCIROpsIncGen)
+add_dependencies(mlir-headers MLIRCIROpsIncGen)
+
diff --git a/clang/include/clang/Driver/Options.td b/clang/include/clang/Driver/Options.td
index 3c55aceecdde..263d1edf141a 100644
--- a/clang/include/clang/Driver/Options.td
+++ b/clang/include/clang/Driver/Options.td
@@ -2615,6 +2615,11 @@ defm protect_parens : BoolFOption<"protect-parens",
           "floating-point expressions are evaluated">,
   NegFlag<SetFalse>>;
 
+defm daz_ftz : SimpleMFlag<"daz-ftz",
+  "Globally set", "Do not globally set",
+  " the denormals-are-zero (DAZ) and flush-to-zero (FTZ) bits in the "
+  "floating-point control register on program startup">;
+
 def ffor_scope : Flag<["-"], "ffor-scope">, Group<f_Group>;
 def fno_for_scope : Flag<["-"], "fno-for-scope">, Group<f_Group>;
 
@@ -4876,6 +4881,8 @@ def msimd128 : Flag<["-"], "msimd128">, Group<m_wasm_Features_Group>;
 def mno_simd128 : Flag<["-"], "mno-simd128">, Group<m_wasm_Features_Group>;
 def mrelaxed_simd : Flag<["-"], "mrelaxed-simd">, Group<m_wasm_Features_Group>;
 def mno_relaxed_simd : Flag<["-"], "mno-relaxed-simd">, Group<m_wasm_Features_Group>;
+def mhalf_precision : Flag<["-"], "mhalf-precision">, Group<m_wasm_Features_Group>;
+def mno_half_precision : Flag<["-"], "mno-half-precision">, Group<m_wasm_Features_Group>;
 def mnontrapping_fptoint : Flag<["-"], "mnontrapping-fptoint">, Group<m_wasm_Features_Group>;
 def mno_nontrapping_fptoint : Flag<["-"], "mno-nontrapping-fptoint">, Group<m_wasm_Features_Group>;
 def msign_ext : Flag<["-"], "msign-ext">, Group<m_wasm_Features_Group>;
@@ -5508,6 +5515,14 @@ def fno_rtlib_add_rpath: Flag<["-"], "fno-rtlib-add-rpath">,
   Visibility<[ClangOption, FlangOption]>,
   HelpText<"Do not add -rpath with architecture-specific resource directory to the linker flags. "
   "When --hip-link is specified, do not add -rpath with HIP runtime library directory to the linker flags">;
+def frtlib_defaultlib : Flag<["-"], "frtlib-defaultlib">,
+                      Visibility<[ClangOption, CLOption]>,
+                      Group<f_Group>,
+                      HelpText<"On Windows, emit /defaultlib: directives to link compiler-rt libraries (default)">;
+def fno_rtlib_defaultlib : Flag<["-"], "fno-rtlib-defaultlib">,
+                      Visibility<[ClangOption, CLOption]>,
+                      Group<f_Group>,
+                      HelpText<"On Windows, do not emit /defaultlib: directives to link compiler-rt libraries">;
 def offload_add_rpath: Flag<["--"], "offload-add-rpath">,
   Flags<[NoArgumentUnused]>,
   Alias<frtlib_add_rpath>;
@@ -6574,12 +6589,6 @@ def J : JoinedOrSeparate<["-"], "J">,
   Group<gfortran_Group>,
   Alias<module_dir>;
 
-let Visibility = [FlangOption] in {
-def no_fortran_main : Flag<["-"], "fno-fortran-main">,
-  Visibility<[FlangOption]>, Group<f_Group>,
-  HelpText<"Do not include Fortran_main.a (provided by Flang) when linking">;
-} // let Visibility = [ FlangOption ]
-
 //===----------------------------------------------------------------------===//
 // FC1 Options
 //===----------------------------------------------------------------------===//
diff --git a/clang/include/clang/ExtractAPI/API.h b/clang/include/clang/ExtractAPI/API.h
index 92cacf65c7d6..d323e1668a72 100644
--- a/clang/include/clang/ExtractAPI/API.h
+++ b/clang/include/clang/ExtractAPI/API.h
@@ -208,20 +208,20 @@ struct APIRecord {
     RK_ClassTemplate,
     RK_ClassTemplateSpecialization,
     RK_ClassTemplatePartialSpecialization,
-    RK_LastRecordContext,
-    RK_GlobalFunction,
-    RK_GlobalFunctionTemplate,
-    RK_GlobalFunctionTemplateSpecialization,
+    RK_StructField,
+    RK_UnionField,
+    RK_CXXField,
+    RK_StaticField,
+    RK_CXXFieldTemplate,
     RK_GlobalVariable,
     RK_GlobalVariableTemplate,
     RK_GlobalVariableTemplateSpecialization,
     RK_GlobalVariableTemplatePartialSpecialization,
+    RK_LastRecordContext,
+    RK_GlobalFunction,
+    RK_GlobalFunctionTemplate,
+    RK_GlobalFunctionTemplateSpecialization,
     RK_EnumConstant,
-    RK_StructField,
-    RK_UnionField,
-    RK_StaticField,
-    RK_CXXField,
-    RK_CXXFieldTemplate,
     RK_Concept,
     RK_CXXStaticMethod,
     RK_CXXInstanceMethod,
@@ -321,6 +321,10 @@ public:
 
   RecordContext(APIRecord::RecordKind Kind) : Kind(Kind) {}
 
+  /// Append \p Other children chain into ours and empty out Other's record
+  /// chain.
+  void stealRecordChain(RecordContext &Other);
+
   APIRecord::RecordKind getKind() const { return Kind; }
 
   struct record_iterator {
@@ -370,6 +374,7 @@ private:
   APIRecord::RecordKind Kind;
   mutable APIRecord *First = nullptr;
   mutable APIRecord *Last = nullptr;
+  bool IsWellFormed() const;
 
 protected:
   friend class APISet;
@@ -475,7 +480,7 @@ struct GlobalFunctionTemplateSpecializationRecord : GlobalFunctionRecord {
 };
 
 /// This holds information associated with global functions.
-struct GlobalVariableRecord : APIRecord {
+struct GlobalVariableRecord : APIRecord, RecordContext {
   GlobalVariableRecord(StringRef USR, StringRef Name, SymbolReference Parent,
                        PresumedLoc Loc, AvailabilityInfo Availability,
                        LinkageInfo Linkage, const DocComment &Comment,
@@ -483,23 +488,28 @@ struct GlobalVariableRecord : APIRecord {
                        DeclarationFragments SubHeading, bool IsFromSystemHeader)
       : APIRecord(RK_GlobalVariable, USR, Name, Parent, Loc,
                   std::move(Availability), Linkage, Comment, Declaration,
-                  SubHeading, IsFromSystemHeader) {}
+                  SubHeading, IsFromSystemHeader),
+        RecordContext(RK_GlobalVariable) {}
 
   GlobalVariableRecord(RecordKind Kind, StringRef USR, StringRef Name,
-                       SymbolReference Parent,
-
-                       PresumedLoc Loc, AvailabilityInfo Availability,
-                       LinkageInfo Linkage, const DocComment &Comment,
+                       SymbolReference Parent, PresumedLoc Loc,
+                       AvailabilityInfo Availability, LinkageInfo Linkage,
+                       const DocComment &Comment,
                        DeclarationFragments Declaration,
                        DeclarationFragments SubHeading, bool IsFromSystemHeader)
       : APIRecord(Kind, USR, Name, Parent, Loc, std::move(Availability),
                   Linkage, Comment, Declaration, SubHeading,
-                  IsFromSystemHeader) {}
+                  IsFromSystemHeader),
+        RecordContext(Kind) {}
 
   static bool classof(const APIRecord *Record) {
     return classofKind(Record->getKind());
   }
-  static bool classofKind(RecordKind K) { return K == RK_GlobalVariable; }
+  static bool classofKind(RecordKind K) {
+    return K == RK_GlobalVariable || K == RK_GlobalVariableTemplate ||
+           K == RK_GlobalVariableTemplateSpecialization ||
+           K == RK_GlobalVariableTemplatePartialSpecialization;
+  }
 
 private:
   virtual void anchor();
@@ -591,20 +601,47 @@ private:
   virtual void anchor();
 };
 
+struct TagRecord : APIRecord, RecordContext {
+  TagRecord(RecordKind Kind, StringRef USR, StringRef Name,
+            SymbolReference Parent, PresumedLoc Loc,
+            AvailabilityInfo Availability, const DocComment &Comment,
+            DeclarationFragments Declaration, DeclarationFragments SubHeading,
+            bool IsFromSystemHeader, bool IsEmbeddedInVarDeclarator,
+            AccessControl Access = AccessControl())
+      : APIRecord(Kind, USR, Name, Parent, Loc, std::move(Availability),
+                  LinkageInfo::none(), Comment, Declaration, SubHeading,
+                  IsFromSystemHeader, std::move(Access)),
+        RecordContext(Kind),
+        IsEmbeddedInVarDeclarator(IsEmbeddedInVarDeclarator){};
+
+  static bool classof(const APIRecord *Record) {
+    return classofKind(Record->getKind());
+  }
+  static bool classofKind(RecordKind K) {
+    return K == RK_Struct || K == RK_Union || K == RK_Enum;
+  }
+
+  bool IsEmbeddedInVarDeclarator;
+
+  virtual ~TagRecord() = 0;
+};
+
 /// This holds information associated with enums.
-struct EnumRecord : APIRecord, RecordContext {
+struct EnumRecord : TagRecord {
   EnumRecord(StringRef USR, StringRef Name, SymbolReference Parent,
              PresumedLoc Loc, AvailabilityInfo Availability,
              const DocComment &Comment, DeclarationFragments Declaration,
-             DeclarationFragments SubHeading, bool IsFromSystemHeader)
-      : APIRecord(RK_Enum, USR, Name, Parent, Loc, std::move(Availability),
-                  LinkageInfo::none(), Comment, Declaration, SubHeading,
-                  IsFromSystemHeader),
-        RecordContext(RK_Enum) {}
+             DeclarationFragments SubHeading, bool IsFromSystemHeader,
+             bool IsEmbeddedInVarDeclarator,
+             AccessControl Access = AccessControl())
+      : TagRecord(RK_Enum, USR, Name, Parent, Loc, std::move(Availability),
+                  Comment, Declaration, SubHeading, IsFromSystemHeader,
+                  IsEmbeddedInVarDeclarator, std::move(Access)) {}
 
   static bool classof(const APIRecord *Record) {
     return classofKind(Record->getKind());
   }
+
   static bool classofKind(RecordKind K) { return K == RK_Enum; }
 
 private:
@@ -612,7 +649,7 @@ private:
 };
 
 /// This holds information associated with struct or union fields fields.
-struct RecordFieldRecord : APIRecord {
+struct RecordFieldRecord : APIRecord, RecordContext {
   RecordFieldRecord(RecordKind Kind, StringRef USR, StringRef Name,
                     SymbolReference Parent, PresumedLoc Loc,
                     AvailabilityInfo Availability, const DocComment &Comment,
@@ -620,7 +657,8 @@ struct RecordFieldRecord : APIRecord {
                     DeclarationFragments SubHeading, bool IsFromSystemHeader)
       : APIRecord(Kind, USR, Name, Parent, Loc, std::move(Availability),
                   LinkageInfo::none(), Comment, Declaration, SubHeading,
-                  IsFromSystemHeader) {}
+                  IsFromSystemHeader),
+        RecordContext(Kind) {}
 
   static bool classof(const APIRecord *Record) {
     return classofKind(Record->getKind());
@@ -633,16 +671,17 @@ struct RecordFieldRecord : APIRecord {
 };
 
 /// This holds information associated with structs and unions.
-struct RecordRecord : APIRecord, RecordContext {
+struct RecordRecord : TagRecord {
   RecordRecord(RecordKind Kind, StringRef USR, StringRef Name,
                SymbolReference Parent, PresumedLoc Loc,
                AvailabilityInfo Availability, const DocComment &Comment,
                DeclarationFragments Declaration,
-               DeclarationFragments SubHeading, bool IsFromSystemHeader)
-      : APIRecord(Kind, USR, Name, Parent, Loc, std::move(Availability),
-                  LinkageInfo::none(), Comment, Declaration, SubHeading,
-                  IsFromSystemHeader),
-        RecordContext(Kind) {}
+               DeclarationFragments SubHeading, bool IsFromSystemHeader,
+               bool IsEmbeddedInVarDeclarator,
+               AccessControl Access = AccessControl())
+      : TagRecord(Kind, USR, Name, Parent, Loc, std::move(Availability),
+                  Comment, Declaration, SubHeading, IsFromSystemHeader,
+                  IsEmbeddedInVarDeclarator, std::move(Access)) {}
 
   static bool classof(const APIRecord *Record) {
     return classofKind(Record->getKind());
@@ -651,6 +690,8 @@ struct RecordRecord : APIRecord, RecordContext {
     return K == RK_Struct || K == RK_Union;
   }
 
+  bool isAnonymousWithNoTypedef() { return Name.empty(); }
+
   virtual ~RecordRecord() = 0;
 };
 
@@ -676,9 +717,11 @@ struct StructRecord : RecordRecord {
   StructRecord(StringRef USR, StringRef Name, SymbolReference Parent,
                PresumedLoc Loc, AvailabilityInfo Availability,
                const DocComment &Comment, DeclarationFragments Declaration,
-               DeclarationFragments SubHeading, bool IsFromSystemHeader)
+               DeclarationFragments SubHeading, bool IsFromSystemHeader,
+               bool IsEmbeddedInVarDeclarator)
       : RecordRecord(RK_Struct, USR, Name, Parent, Loc, std::move(Availability),
-                     Comment, Declaration, SubHeading, IsFromSystemHeader) {}
+                     Comment, Declaration, SubHeading, IsFromSystemHeader,
+                     IsEmbeddedInVarDeclarator) {}
 
   static bool classof(const APIRecord *Record) {
     return classofKind(Record->getKind());
@@ -711,9 +754,11 @@ struct UnionRecord : RecordRecord {
   UnionRecord(StringRef USR, StringRef Name, SymbolReference Parent,
               PresumedLoc Loc, AvailabilityInfo Availability,
               const DocComment &Comment, DeclarationFragments Declaration,
-              DeclarationFragments SubHeading, bool IsFromSystemHeader)
+              DeclarationFragments SubHeading, bool IsFromSystemHeader,
+              bool IsEmbeddedInVarDeclarator)
       : RecordRecord(RK_Union, USR, Name, Parent, Loc, std::move(Availability),
-                     Comment, Declaration, SubHeading, IsFromSystemHeader) {}
+                     Comment, Declaration, SubHeading, IsFromSystemHeader,
+                     IsEmbeddedInVarDeclarator) {}
 
   static bool classof(const APIRecord *Record) {
     return classofKind(Record->getKind());
@@ -724,7 +769,7 @@ private:
   virtual void anchor();
 };
 
-struct CXXFieldRecord : APIRecord {
+struct CXXFieldRecord : APIRecord, RecordContext {
   CXXFieldRecord(StringRef USR, StringRef Name, SymbolReference Parent,
                  PresumedLoc Loc, AvailabilityInfo Availability,
                  const DocComment &Comment, DeclarationFragments Declaration,
@@ -732,7 +777,8 @@ struct CXXFieldRecord : APIRecord {
                  bool IsFromSystemHeader)
       : APIRecord(RK_CXXField, USR, Name, Parent, Loc, std::move(Availability),
                   LinkageInfo::none(), Comment, Declaration, SubHeading,
-                  IsFromSystemHeader, std::move(Access)) {}
+                  IsFromSystemHeader, std::move(Access)),
+        RecordContext(RK_CXXField) {}
 
   CXXFieldRecord(RecordKind Kind, StringRef USR, StringRef Name,
                  SymbolReference Parent, PresumedLoc Loc,
@@ -742,7 +788,8 @@ struct CXXFieldRecord : APIRecord {
                  bool IsFromSystemHeader)
       : APIRecord(Kind, USR, Name, Parent, Loc, std::move(Availability),
                   LinkageInfo::none(), Comment, Declaration, SubHeading,
-                  IsFromSystemHeader, std::move(Access)) {}
+                  IsFromSystemHeader, std::move(Access)),
+        RecordContext(Kind) {}
 
   static bool classof(const APIRecord *Record) {
     return classofKind(Record->getKind());
@@ -1118,18 +1165,18 @@ struct ObjCContainerRecord : APIRecord, RecordContext {
   virtual ~ObjCContainerRecord() = 0;
 };
 
-struct CXXClassRecord : APIRecord, RecordContext {
+struct CXXClassRecord : RecordRecord {
   SmallVector<SymbolReference> Bases;
 
   CXXClassRecord(StringRef USR, StringRef Name, SymbolReference Parent,
                  PresumedLoc Loc, AvailabilityInfo Availability,
                  const DocComment &Comment, DeclarationFragments Declaration,
                  DeclarationFragments SubHeading, RecordKind Kind,
-                 AccessControl Access, bool IsFromSystemHeader)
-      : APIRecord(Kind, USR, Name, Parent, Loc, std::move(Availability),
-                  LinkageInfo::none(), Comment, Declaration, SubHeading,
-                  IsFromSystemHeader, std::move(Access)),
-        RecordContext(Kind) {}
+                 AccessControl Access, bool IsFromSystemHeader,
+                 bool IsEmbeddedInVarDeclarator = false)
+      : RecordRecord(Kind, USR, Name, Parent, Loc, std::move(Availability),
+                     Comment, Declaration, SubHeading, IsFromSystemHeader,
+                     IsEmbeddedInVarDeclarator, std::move(Access)) {}
 
   static bool classof(const APIRecord *Record) {
     return classofKind(Record->getKind());
diff --git a/clang/include/clang/ExtractAPI/APIRecords.inc b/clang/include/clang/ExtractAPI/APIRecords.inc
index 15fee809656d..4cda4ef2f9be 100644
--- a/clang/include/clang/ExtractAPI/APIRecords.inc
+++ b/clang/include/clang/ExtractAPI/APIRecords.inc
@@ -35,10 +35,11 @@ CONCRETE_RECORD(GlobalVariableTemplateSpecializationRecord,
 CONCRETE_RECORD(GlobalVariableTemplatePartialSpecializationRecord,
                 GlobalVariableRecord,
                 RK_GlobalVariableTemplatePartialSpecialization)
+ABSTRACT_RECORD(TagRecord, APIRecord)
 CONCRETE_RECORD(EnumConstantRecord, APIRecord, RK_EnumConstant)
-CONCRETE_RECORD(EnumRecord, APIRecord, RK_Enum)
+CONCRETE_RECORD(EnumRecord, TagRecord, RK_Enum)
 ABSTRACT_RECORD(RecordFieldRecord, APIRecord)
-ABSTRACT_RECORD(RecordRecord, APIRecord)
+ABSTRACT_RECORD(RecordRecord, TagRecord)
 CONCRETE_RECORD(StructFieldRecord, RecordFieldRecord, RK_StructField)
 CONCRETE_RECORD(StructRecord, APIRecord, RK_Struct)
 CONCRETE_RECORD(UnionFieldRecord, RecordFieldRecord, RK_UnionField)
@@ -99,5 +100,16 @@ RECORD_CONTEXT(ClassTemplateSpecializationRecord,
                RK_ClassTemplateSpecialization)
 RECORD_CONTEXT(ClassTemplatePartialSpecializationRecord,
                RK_ClassTemplatePartialSpecialization)
+RECORD_CONTEXT(StructFieldRecord, RK_StructField)
+RECORD_CONTEXT(UnionFieldRecord, RK_UnionField)
+RECORD_CONTEXT(CXXFieldRecord, RK_CXXField)
+RECORD_CONTEXT(StaticFieldRecord, RK_StaticField)
+RECORD_CONTEXT(CXXFieldTemplateRecord, RK_CXXFieldTemplate)
+RECORD_CONTEXT(GlobalVariableRecord, RK_GlobalVariable)
+RECORD_CONTEXT(GlobalVariableTemplateRecord, RK_GlobalVariableTemplate)
+RECORD_CONTEXT(GlobalVariableTemplateSpecializationRecord,
+               RK_GlobalVariableTemplateSpecialization)
+RECORD_CONTEXT(GlobalVariableTemplatePartialSpecializationRecord,
+               RK_GlobalVariableTemplatePartialSpecialization)
 
 #undef RECORD_CONTEXT
diff --git a/clang/include/clang/ExtractAPI/DeclarationFragments.h b/clang/include/clang/ExtractAPI/DeclarationFragments.h
index 94392c185165..535da90b9828 100644
--- a/clang/include/clang/ExtractAPI/DeclarationFragments.h
+++ b/clang/include/clang/ExtractAPI/DeclarationFragments.h
@@ -27,6 +27,8 @@
 #include "clang/AST/TypeLoc.h"
 #include "clang/Basic/Specifiers.h"
 #include "clang/Lex/MacroInfo.h"
+#include <iterator>
+#include <utility>
 #include <vector>
 
 namespace clang {
@@ -113,28 +115,26 @@ public:
 
   ConstFragmentIterator cend() const { return Fragments.cend(); }
 
-  // Add a new Fragment at an arbitrary offset.
-  DeclarationFragments &insert(FragmentIterator It, StringRef Spelling,
-                               FragmentKind Kind,
-                               StringRef PreciseIdentifier = "",
-                               const Decl *Declaration = nullptr) {
-    Fragments.insert(It,
-                     Fragment(Spelling, Kind, PreciseIdentifier, Declaration));
-    return *this;
+  /// Prepend another DeclarationFragments to the beginning.
+  ///
+  /// \returns a reference to the DeclarationFragments object itself after
+  /// appending to chain up consecutive operations.
+  DeclarationFragments &prepend(DeclarationFragments Other) {
+    return insert(begin(), std::move(Other));
   }
 
-  DeclarationFragments &insert(FragmentIterator It,
-                               DeclarationFragments &&Other) {
-    Fragments.insert(It, std::make_move_iterator(Other.Fragments.begin()),
-                     std::make_move_iterator(Other.Fragments.end()));
-    Other.Fragments.clear();
-    return *this;
+  /// Append another DeclarationFragments to the end.
+  ///
+  /// \returns a reference to the DeclarationFragments object itself after
+  /// appending to chain up consecutive operations.
+  DeclarationFragments &append(DeclarationFragments Other) {
+    return insert(end(), std::move(Other));
   }
 
   /// Append a new Fragment to the end of the Fragments.
   ///
   /// \returns a reference to the DeclarationFragments object itself after
-  /// appending to chain up consecutive appends.
+  /// appending to chain up consecutive operations.
   DeclarationFragments &append(StringRef Spelling, FragmentKind Kind,
                                StringRef PreciseIdentifier = "",
                                const Decl *Declaration = nullptr) {
@@ -149,18 +149,48 @@ public:
     return *this;
   }
 
-  /// Append another DeclarationFragments to the end.
-  ///
-  /// Note: \p Other is moved from and cannot be used after a call to this
-  /// method.
+  /// Inserts another DeclarationFragments at \p It.
   ///
   /// \returns a reference to the DeclarationFragments object itself after
-  /// appending to chain up consecutive appends.
-  DeclarationFragments &append(DeclarationFragments &&Other) {
-    Fragments.insert(Fragments.end(),
-                     std::make_move_iterator(Other.Fragments.begin()),
-                     std::make_move_iterator(Other.Fragments.end()));
-    Other.Fragments.clear();
+  /// appending to chain up consecutive operations.
+  DeclarationFragments &insert(FragmentIterator It,
+                               DeclarationFragments Other) {
+    if (Other.Fragments.empty())
+      return *this;
+
+    if (Fragments.empty()) {
+      Fragments = std::move(Other.Fragments);
+      return *this;
+    }
+
+    const auto &OtherFrags = Other.Fragments;
+    auto ToInsertBegin = std::make_move_iterator(Other.begin());
+    auto ToInsertEnd = std::make_move_iterator(Other.end());
+
+    // If we aren't inserting at the end let's make sure that we merge their
+    // last fragment with It if both are text fragments.
+    if (It != end() && It->Kind == FragmentKind::Text &&
+        OtherFrags.back().Kind == FragmentKind::Text) {
+      auto &TheirBackSpelling = OtherFrags.back().Spelling;
+      It->Spelling.reserve(It->Spelling.size() + TheirBackSpelling.size());
+      It->Spelling.insert(It->Spelling.begin(), TheirBackSpelling.begin(),
+                          TheirBackSpelling.end());
+      --ToInsertEnd;
+    }
+
+    // If we aren't inserting at the beginning we want to merge their first
+    // fragment with the fragment before It if both are text fragments.
+    if (It != begin() && std::prev(It)->Kind == FragmentKind::Text &&
+        OtherFrags.front().Kind == FragmentKind::Text) {
+      auto PrevIt = std::prev(It);
+      auto &TheirFrontSpelling = OtherFrags.front().Spelling;
+      PrevIt->Spelling.reserve(PrevIt->Spelling.size() +
+                               TheirFrontSpelling.size());
+      PrevIt->Spelling.append(TheirFrontSpelling);
+      ++ToInsertBegin;
+    }
+
+    Fragments.insert(It, ToInsertBegin, ToInsertEnd);
     return *this;
   }
 
@@ -177,13 +207,13 @@ public:
   /// Append a text Fragment of a space character.
   ///
   /// \returns a reference to the DeclarationFragments object itself after
-  /// appending to chain up consecutive appends.
+  /// appending to chain up consecutive operations.
   DeclarationFragments &appendSpace();
 
   /// Append a text Fragment of a semicolon character.
   ///
   /// \returns a reference to the DeclarationFragments object itself after
-  /// appending to chain up consecutive appends.
+  /// appending to chain up consecutive operations.
   DeclarationFragments &appendSemicolon();
 
   /// Removes a trailing semicolon character if present.
diff --git a/clang/include/clang/ExtractAPI/ExtractAPIVisitor.h b/clang/include/clang/ExtractAPI/ExtractAPIVisitor.h
index 4cb866892b5d..97cc457ea2a9 100644
--- a/clang/include/clang/ExtractAPI/ExtractAPIVisitor.h
+++ b/clang/include/clang/ExtractAPI/ExtractAPIVisitor.h
@@ -224,6 +224,29 @@ protected:
 
     return API.createSymbolReference(Name, USR, getOwningModuleName(D));
   }
+
+  bool isEmbeddedInVarDeclarator(const TagDecl &D) {
+    return D.getName().empty() && getTypedefName(&D).empty() &&
+           D.isEmbeddedInDeclarator();
+  }
+
+  void maybeMergeWithAnonymousTag(const DeclaratorDecl &D,
+                                  RecordContext *NewRecordContext) {
+    if (!NewRecordContext)
+      return;
+    auto *Tag = D.getType()->getAsTagDecl();
+    SmallString<128> TagUSR;
+    clang::index::generateUSRForDecl(Tag, TagUSR);
+    if (auto *Record = llvm::dyn_cast_if_present<TagRecord>(
+            API.findRecordForUSR(TagUSR))) {
+      if (Record->IsEmbeddedInVarDeclarator) {
+        NewRecordContext->stealRecordChain(*Record);
+        auto *NewRecord = cast<APIRecord>(NewRecordContext);
+        if (NewRecord->Comment.empty())
+          NewRecord->Comment = Record->Comment;
+      }
+    }
+  }
 };
 
 template <typename Derived>
@@ -273,12 +296,18 @@ bool ExtractAPIVisitorBase<Derived>::VisitVarDecl(const VarDecl *Decl) {
         USR, Name, createHierarchyInformationForDecl(*Decl), Loc,
         AvailabilityInfo::createFromDecl(Decl), Linkage, Comment, Declaration,
         SubHeading, Access, isInSystemHeader(Decl));
-  } else
+  } else {
     // Add the global variable record to the API set.
-    API.createRecord<GlobalVariableRecord>(
+    auto *NewRecord = API.createRecord<GlobalVariableRecord>(
         USR, Name, createHierarchyInformationForDecl(*Decl), Loc,
         AvailabilityInfo::createFromDecl(Decl), Linkage, Comment, Declaration,
         SubHeading, isInSystemHeader(Decl));
+
+    // If this global variable has a non typedef'd anonymous tag type let's
+    // pretend the type's child records are under us in the hierarchy.
+    maybeMergeWithAnonymousTag(*Decl, NewRecord);
+  }
+
   return true;
 }
 
@@ -364,7 +393,7 @@ bool ExtractAPIVisitorBase<Derived>::VisitEnumDecl(const EnumDecl *Decl) {
   if (Name.empty()) {
     llvm::raw_svector_ostream OS(QualifiedNameBuffer);
     Decl->printQualifiedName(OS);
-    Name = QualifiedNameBuffer.str();
+    Name = QualifiedNameBuffer;
   }
 
   SmallString<128> USR;
@@ -385,7 +414,7 @@ bool ExtractAPIVisitorBase<Derived>::VisitEnumDecl(const EnumDecl *Decl) {
   auto *ER = API.createRecord<EnumRecord>(
       USR, Name, createHierarchyInformationForDecl(*Decl), Loc,
       AvailabilityInfo::createFromDecl(Decl), Comment, Declaration, SubHeading,
-      isInSystemHeader(Decl));
+      isInSystemHeader(Decl), isEmbeddedInVarDeclarator(*Decl));
 
   // Now collect information about the enumerators in this enum.
   getDerivedExtractAPIVisitor().recordEnumConstants(ER, Decl->enumerators());
@@ -510,16 +539,10 @@ bool ExtractAPIVisitorBase<Derived>::VisitRecordDecl(const RecordDecl *Decl) {
   if (!getDerivedExtractAPIVisitor().shouldDeclBeIncluded(Decl))
     return true;
 
-  SmallString<128> QualifiedNameBuffer;
   // Collect symbol information.
   StringRef Name = Decl->getName();
   if (Name.empty())
     Name = getTypedefName(Decl);
-  if (Name.empty()) {
-    llvm::raw_svector_ostream OS(QualifiedNameBuffer);
-    Decl->printQualifiedName(OS);
-    Name = QualifiedNameBuffer.str();
-  }
 
   SmallString<128> USR;
   index::generateUSRForDecl(Decl, USR);
@@ -541,12 +564,12 @@ bool ExtractAPIVisitorBase<Derived>::VisitRecordDecl(const RecordDecl *Decl) {
     API.createRecord<UnionRecord>(
         USR, Name, createHierarchyInformationForDecl(*Decl), Loc,
         AvailabilityInfo::createFromDecl(Decl), Comment, Declaration,
-        SubHeading, isInSystemHeader(Decl));
+        SubHeading, isInSystemHeader(Decl), isEmbeddedInVarDeclarator(*Decl));
   else
     API.createRecord<StructRecord>(
         USR, Name, createHierarchyInformationForDecl(*Decl), Loc,
         AvailabilityInfo::createFromDecl(Decl), Comment, Declaration,
-        SubHeading, isInSystemHeader(Decl));
+        SubHeading, isInSystemHeader(Decl), isEmbeddedInVarDeclarator(*Decl));
 
   return true;
 }
@@ -559,6 +582,9 @@ bool ExtractAPIVisitorBase<Derived>::VisitCXXRecordDecl(
     return true;
 
   StringRef Name = Decl->getName();
+  if (Name.empty())
+    Name = getTypedefName(Decl);
+
   SmallString<128> USR;
   index::generateUSRForDecl(Decl, USR);
   PresumedLoc Loc =
@@ -585,8 +611,7 @@ bool ExtractAPIVisitorBase<Derived>::VisitCXXRecordDecl(
   CXXClassRecord *Record;
   if (Decl->getDescribedClassTemplate()) {
     // Inject template fragments before class fragments.
-    Declaration.insert(
-        Declaration.begin(),
+    Declaration.prepend(
         DeclarationFragmentsBuilder::getFragmentsForRedeclarableTemplate(
             Decl->getDescribedClassTemplate()));
     Record = API.createRecord<ClassTemplateRecord>(
@@ -598,7 +623,8 @@ bool ExtractAPIVisitorBase<Derived>::VisitCXXRecordDecl(
     Record = API.createRecord<CXXClassRecord>(
         USR, Name, createHierarchyInformationForDecl(*Decl), Loc,
         AvailabilityInfo::createFromDecl(Decl), Comment, Declaration,
-        SubHeading, Kind, Access, isInSystemHeader(Decl));
+        SubHeading, Kind, Access, isInSystemHeader(Decl),
+        isEmbeddedInVarDeclarator(*Decl));
 
   Record->Bases = getBases(Decl);
 
@@ -1075,18 +1101,17 @@ bool ExtractAPIVisitorBase<Derived>::VisitTypedefNameDecl(
   // If the underlying type was defined as part of the typedef modify it's
   // fragments directly and pretend the typedef doesn't exist.
   if (auto *TagDecl = Decl->getUnderlyingType()->getAsTagDecl()) {
-    if (TagDecl->getName() == Decl->getName() &&
-        TagDecl->isEmbeddedInDeclarator() && TagDecl->isCompleteDefinition()) {
+    if (TagDecl->isEmbeddedInDeclarator() && TagDecl->isCompleteDefinition() &&
+        Decl->getName() == TagDecl->getName()) {
       SmallString<128> TagUSR;
       index::generateUSRForDecl(TagDecl, TagUSR);
       if (auto *Record = API.findRecordForUSR(TagUSR)) {
         DeclarationFragments LeadingFragments;
         LeadingFragments.append("typedef",
-                                DeclarationFragments::FragmentKind::Keyword, "",
-                                nullptr);
+                                DeclarationFragments::FragmentKind::Keyword);
         LeadingFragments.appendSpace();
         Record->Declaration.removeTrailingSemicolon()
-            .insert(Record->Declaration.begin(), std::move(LeadingFragments))
+            .prepend(std::move(LeadingFragments))
             .append(" { ... } ", DeclarationFragments::FragmentKind::Text)
             .append(Name, DeclarationFragments::FragmentKind::Identifier)
             .appendSemicolon();
@@ -1221,26 +1246,31 @@ bool ExtractAPIVisitorBase<Derived>::VisitFieldDecl(const FieldDecl *Decl) {
   DeclarationFragments SubHeading =
       DeclarationFragmentsBuilder::getSubHeading(Decl);
 
+  RecordContext *NewRecord = nullptr;
   if (isa<CXXRecordDecl>(Decl->getDeclContext())) {
     AccessControl Access = DeclarationFragmentsBuilder::getAccessControl(Decl);
 
-    API.createRecord<CXXFieldRecord>(
+    NewRecord = API.createRecord<CXXFieldRecord>(
         USR, Name, createHierarchyInformationForDecl(*Decl), Loc,
         AvailabilityInfo::createFromDecl(Decl), Comment, Declaration,
         SubHeading, Access, isInSystemHeader(Decl));
   } else if (auto *RD = dyn_cast<RecordDecl>(Decl->getDeclContext())) {
     if (RD->isUnion())
-      API.createRecord<UnionFieldRecord>(
+      NewRecord = API.createRecord<UnionFieldRecord>(
           USR, Name, createHierarchyInformationForDecl(*Decl), Loc,
           AvailabilityInfo::createFromDecl(Decl), Comment, Declaration,
           SubHeading, isInSystemHeader(Decl));
     else
-      API.createRecord<StructFieldRecord>(
+      NewRecord = API.createRecord<StructFieldRecord>(
           USR, Name, createHierarchyInformationForDecl(*Decl), Loc,
           AvailabilityInfo::createFromDecl(Decl), Comment, Declaration,
           SubHeading, isInSystemHeader(Decl));
   }
 
+  // If this field has a non typedef'd anonymous tag type let's pretend the
+  // type's child records are under us in the hierarchy.
+  maybeMergeWithAnonymousTag(*Decl, NewRecord);
+
   return true;
 }
 
diff --git a/clang/include/clang/Frontend/ASTUnit.h b/clang/include/clang/Frontend/ASTUnit.h
index a2c1b25dd224..080844893c13 100644
--- a/clang/include/clang/Frontend/ASTUnit.h
+++ b/clang/include/clang/Frontend/ASTUnit.h
@@ -241,7 +241,7 @@ private:
 
   /// A list of the serialization ID numbers for each of the top-level
   /// declarations parsed within the precompiled preamble.
-  std::vector<serialization::DeclID> TopLevelDeclsInPreamble;
+  std::vector<LocalDeclID> TopLevelDeclsInPreamble;
 
   /// Whether we should be caching code-completion results.
   bool ShouldCacheCodeCompletionResults : 1;
diff --git a/clang/include/clang/Frontend/MultiplexConsumer.h b/clang/include/clang/Frontend/MultiplexConsumer.h
index 7f8d2858b386..f29c8e92fded 100644
--- a/clang/include/clang/Frontend/MultiplexConsumer.h
+++ b/clang/include/clang/Frontend/MultiplexConsumer.h
@@ -35,7 +35,7 @@ public:
   void IdentifierRead(serialization::IdentID ID, IdentifierInfo *II) override;
   void MacroRead(serialization::MacroID ID, MacroInfo *MI) override;
   void TypeRead(serialization::TypeIdx Idx, QualType T) override;
-  void DeclRead(serialization::DeclID ID, const Decl *D) override;
+  void DeclRead(GlobalDeclID ID, const Decl *D) override;
   void SelectorRead(serialization::SelectorID iD, Selector Sel) override;
   void MacroDefinitionRead(serialization::PreprocessedEntityID,
                            MacroDefinitionRecord *MD) override;
diff --git a/clang/include/clang/Lex/HeaderSearch.h b/clang/include/clang/Lex/HeaderSearch.h
index c5f90ef4cb36..5ac63dddd4d4 100644
--- a/clang/include/clang/Lex/HeaderSearch.h
+++ b/clang/include/clang/Lex/HeaderSearch.h
@@ -56,6 +56,12 @@ class TargetInfo;
 /// The preprocessor keeps track of this information for each
 /// file that is \#included.
 struct HeaderFileInfo {
+  // TODO: Whether the file was included is not a property of the file itself.
+  // It's a preprocessor state, move it there.
+  /// True if this file has been included (or imported) **locally**.
+  LLVM_PREFERRED_TYPE(bool)
+  unsigned IsLocallyIncluded : 1;
+
   // TODO: Whether the file was imported is not a property of the file itself.
   // It's a preprocessor state, move it there.
   /// True if this is a \#import'd file.
@@ -135,10 +141,10 @@ struct HeaderFileInfo {
   StringRef Framework;
 
   HeaderFileInfo()
-      : isImport(false), isPragmaOnce(false), DirInfo(SrcMgr::C_User),
-        External(false), isModuleHeader(false), isTextualModuleHeader(false),
-        isCompilingModuleHeader(false), Resolved(false),
-        IndexHeaderMapHeader(false), IsValid(false) {}
+      : IsLocallyIncluded(false), isImport(false), isPragmaOnce(false),
+        DirInfo(SrcMgr::C_User), External(false), isModuleHeader(false),
+        isTextualModuleHeader(false), isCompilingModuleHeader(false),
+        Resolved(false), IndexHeaderMapHeader(false), IsValid(false) {}
 
   /// Retrieve the controlling macro for this header file, if
   /// any.
diff --git a/clang/include/clang/Parse/Parser.h b/clang/include/clang/Parse/Parser.h
index d3bb04ff7a2c..fb117bf04087 100644
--- a/clang/include/clang/Parse/Parser.h
+++ b/clang/include/clang/Parse/Parser.h
@@ -313,7 +313,15 @@ class Parser : public CodeCompletionHandler {
   /// top-level declaration is finished.
   SmallVector<TemplateIdAnnotation *, 16> TemplateIds;
 
+  /// Don't destroy template annotations in MaybeDestroyTemplateIds even if
+  /// we're at the end of a declaration. Instead, we defer the destruction until
+  /// after a top-level declaration.
+  /// Use DelayTemplateIdDestructionRAII rather than setting it directly.
+  bool DelayTemplateIdDestruction = false;
+
   void MaybeDestroyTemplateIds() {
+    if (DelayTemplateIdDestruction)
+      return;
     if (!TemplateIds.empty() &&
         (Tok.is(tok::eof) || !PP.mightHavePendingAnnotationTokens()))
       DestroyTemplateIds();
@@ -329,6 +337,22 @@ class Parser : public CodeCompletionHandler {
     ~DestroyTemplateIdAnnotationsRAIIObj() { Self.MaybeDestroyTemplateIds(); }
   };
 
+  struct DelayTemplateIdDestructionRAII {
+    Parser &Self;
+    bool PrevDelayTemplateIdDestruction;
+
+    DelayTemplateIdDestructionRAII(Parser &Self,
+                                   bool DelayTemplateIdDestruction) noexcept
+        : Self(Self),
+          PrevDelayTemplateIdDestruction(Self.DelayTemplateIdDestruction) {
+      Self.DelayTemplateIdDestruction = DelayTemplateIdDestruction;
+    }
+
+    ~DelayTemplateIdDestructionRAII() noexcept {
+      Self.DelayTemplateIdDestruction = PrevDelayTemplateIdDestruction;
+    }
+  };
+
   /// Identifiers which have been declared within a tentative parse.
   SmallVector<const IdentifierInfo *, 8> TentativelyDeclaredIdentifiers;
 
@@ -2967,25 +2991,25 @@ private:
       Sema::AttributeCompletion Completion = Sema::AttributeCompletion::None,
       const IdentifierInfo *EnclosingScope = nullptr);
 
-  void MaybeParseHLSLSemantics(Declarator &D,
-                               SourceLocation *EndLoc = nullptr) {
-    assert(getLangOpts().HLSL && "MaybeParseHLSLSemantics is for HLSL only");
+  void MaybeParseHLSLAnnotations(Declarator &D,
+                                 SourceLocation *EndLoc = nullptr) {
+    assert(getLangOpts().HLSL && "MaybeParseHLSLAnnotations is for HLSL only");
     if (Tok.is(tok::colon)) {
       ParsedAttributes Attrs(AttrFactory);
-      ParseHLSLSemantics(Attrs, EndLoc);
+      ParseHLSLAnnotations(Attrs, EndLoc);
       D.takeAttributes(Attrs);
     }
   }
 
-  void MaybeParseHLSLSemantics(ParsedAttributes &Attrs,
-                               SourceLocation *EndLoc = nullptr) {
-    assert(getLangOpts().HLSL && "MaybeParseHLSLSemantics is for HLSL only");
+  void MaybeParseHLSLAnnotations(ParsedAttributes &Attrs,
+                                 SourceLocation *EndLoc = nullptr) {
+    assert(getLangOpts().HLSL && "MaybeParseHLSLAnnotations is for HLSL only");
     if (getLangOpts().HLSL && Tok.is(tok::colon))
-      ParseHLSLSemantics(Attrs, EndLoc);
+      ParseHLSLAnnotations(Attrs, EndLoc);
   }
 
-  void ParseHLSLSemantics(ParsedAttributes &Attrs,
-                          SourceLocation *EndLoc = nullptr);
+  void ParseHLSLAnnotations(ParsedAttributes &Attrs,
+                            SourceLocation *EndLoc = nullptr);
   Decl *ParseHLSLBuffer(SourceLocation &DeclEnd);
 
   void MaybeParseMicrosoftAttributes(ParsedAttributes &Attrs) {
diff --git a/clang/include/clang/Sema/DeclSpec.h b/clang/include/clang/Sema/DeclSpec.h
index c9eecdafe62c..760c7980be52 100644
--- a/clang/include/clang/Sema/DeclSpec.h
+++ b/clang/include/clang/Sema/DeclSpec.h
@@ -36,6 +36,7 @@
 #include "llvm/ADT/SmallVector.h"
 #include "llvm/Support/Compiler.h"
 #include "llvm/Support/ErrorHandling.h"
+#include <optional>
 
 namespace clang {
   class ASTContext;
@@ -1790,6 +1791,7 @@ public:
   struct Binding {
     IdentifierInfo *Name;
     SourceLocation NameLoc;
+    std::optional<ParsedAttributes> Attrs;
   };
 
 private:
@@ -2339,10 +2341,10 @@ public:
   }
 
   /// Set the decomposition bindings for this declarator.
-  void
-  setDecompositionBindings(SourceLocation LSquareLoc,
-                           ArrayRef<DecompositionDeclarator::Binding> Bindings,
-                           SourceLocation RSquareLoc);
+  void setDecompositionBindings(
+      SourceLocation LSquareLoc,
+      MutableArrayRef<DecompositionDeclarator::Binding> Bindings,
+      SourceLocation RSquareLoc);
 
   /// AddTypeInfo - Add a chunk to this declarator. Also extend the range to
   /// EndLoc, which should be the last token of the chunk.
diff --git a/clang/include/clang/Sema/MultiplexExternalSemaSource.h b/clang/include/clang/Sema/MultiplexExternalSemaSource.h
index 2bf91cb5212c..238fb398b7d1 100644
--- a/clang/include/clang/Sema/MultiplexExternalSemaSource.h
+++ b/clang/include/clang/Sema/MultiplexExternalSemaSource.h
@@ -65,7 +65,7 @@ public:
 
   /// Resolve a declaration ID into a declaration, potentially
   /// building a new declaration.
-  Decl *GetExternalDecl(uint32_t ID) override;
+  Decl *GetExternalDecl(GlobalDeclID ID) override;
 
   /// Complete the redeclaration chain if it's been extended since the
   /// previous generation of the AST source.
diff --git a/clang/include/clang/Sema/ParsedAttr.h b/clang/include/clang/Sema/ParsedAttr.h
index 25a5fa05b21c..8368d9ce6146 100644
--- a/clang/include/clang/Sema/ParsedAttr.h
+++ b/clang/include/clang/Sema/ParsedAttr.h
@@ -948,6 +948,7 @@ public:
   ParsedAttributes(AttributeFactory &factory) : pool(factory) {}
   ParsedAttributes(const ParsedAttributes &) = delete;
   ParsedAttributes &operator=(const ParsedAttributes &) = delete;
+  ParsedAttributes(ParsedAttributes &&G) = default;
 
   AttributePool &getPool() const { return pool; }
 
diff --git a/clang/include/clang/Sema/Sema.h b/clang/include/clang/Sema/Sema.h
index 64607b91acbf..1ca523ec88c2 100644
--- a/clang/include/clang/Sema/Sema.h
+++ b/clang/include/clang/Sema/Sema.h
@@ -3084,6 +3084,7 @@ public:
   Decl *ActOnStartOfFunctionDef(Scope *S, Decl *D,
                                 SkipBodyInfo *SkipBody = nullptr,
                                 FnBodyKind BodyKind = FnBodyKind::Other);
+  void applyFunctionAttributesBeforeParsingBody(Decl *FD);
 
   /// Determine whether we can delay parsing the body of a function or
   /// function template until it is used, assuming we don't care about emitting
diff --git a/clang/include/clang/Sema/SemaOpenACC.h b/clang/include/clang/Sema/SemaOpenACC.h
index ea28617f79b8..da19503c2902 100644
--- a/clang/include/clang/Sema/SemaOpenACC.h
+++ b/clang/include/clang/Sema/SemaOpenACC.h
@@ -193,6 +193,12 @@ public:
   /// conversions and diagnostics to 'int'.
   ExprResult ActOnIntExpr(OpenACCDirectiveKind DK, OpenACCClauseKind CK,
                           SourceLocation Loc, Expr *IntExpr);
+
+  /// Checks and creates an Array Section used in an OpenACC construct/clause.
+  ExprResult ActOnArraySectionExpr(Expr *Base, SourceLocation LBLoc,
+                                   Expr *LowerBound,
+                                   SourceLocation ColonLocFirst, Expr *Length,
+                                   SourceLocation RBLoc);
 };
 
 } // namespace clang
diff --git a/clang/include/clang/Serialization/ASTBitCodes.h b/clang/include/clang/Serialization/ASTBitCodes.h
index 500098dd3dab..a8df5a0bda08 100644
--- a/clang/include/clang/Serialization/ASTBitCodes.h
+++ b/clang/include/clang/Serialization/ASTBitCodes.h
@@ -17,6 +17,7 @@
 #ifndef LLVM_CLANG_SERIALIZATION_ASTBITCODES_H
 #define LLVM_CLANG_SERIALIZATION_ASTBITCODES_H
 
+#include "clang/AST/DeclID.h"
 #include "clang/AST/DeclarationName.h"
 #include "clang/AST/Type.h"
 #include "clang/Basic/IdentifierTable.h"
@@ -59,18 +60,9 @@ const unsigned VERSION_MINOR = 1;
 /// and start at 1. 0 is reserved for NULL.
 using IdentifierID = uint32_t;
 
-/// An ID number that refers to a declaration in an AST file.
-///
-/// The ID numbers of declarations are consecutive (in order of
-/// discovery), with values below NUM_PREDEF_DECL_IDS being reserved.
-/// At the start of a chain of precompiled headers, declaration ID 1 is
-/// used for the translation unit declaration.
-using DeclID = uint32_t;
-
-// FIXME: Turn these into classes so we can have some type safety when
-// we go from local ID to global and vice-versa.
-using LocalDeclID = DeclID;
-using GlobalDeclID = DeclID;
+/// An ID number that refers to a declaration in an AST file. See the comments
+/// in DeclIDBase for details.
+using DeclID = DeclIDBase::DeclID;
 
 /// An ID number that refers to a type in an AST file.
 ///
@@ -981,8 +973,8 @@ enum PredefinedTypeIDs {
   /// OpenCL reserve_id type.
   PREDEF_TYPE_RESERVE_ID_ID = 41,
 
-  /// The placeholder type for OpenMP array section.
-  PREDEF_TYPE_OMP_ARRAY_SECTION = 42,
+  /// The placeholder type for an array section.
+  PREDEF_TYPE_ARRAY_SECTION = 42,
 
   /// The '__float128' type
   PREDEF_TYPE_FLOAT128_ID = 43,
@@ -1165,74 +1157,6 @@ enum SpecialTypeIDs {
 /// The number of special type IDs.
 const unsigned NumSpecialTypeIDs = 8;
 
-/// Predefined declaration IDs.
-///
-/// These declaration IDs correspond to predefined declarations in the AST
-/// context, such as the NULL declaration ID. Such declarations are never
-/// actually serialized, since they will be built by the AST context when
-/// it is created.
-enum PredefinedDeclIDs {
-  /// The NULL declaration.
-  PREDEF_DECL_NULL_ID = 0,
-
-  /// The translation unit.
-  PREDEF_DECL_TRANSLATION_UNIT_ID = 1,
-
-  /// The Objective-C 'id' type.
-  PREDEF_DECL_OBJC_ID_ID = 2,
-
-  /// The Objective-C 'SEL' type.
-  PREDEF_DECL_OBJC_SEL_ID = 3,
-
-  /// The Objective-C 'Class' type.
-  PREDEF_DECL_OBJC_CLASS_ID = 4,
-
-  /// The Objective-C 'Protocol' type.
-  PREDEF_DECL_OBJC_PROTOCOL_ID = 5,
-
-  /// The signed 128-bit integer type.
-  PREDEF_DECL_INT_128_ID = 6,
-
-  /// The unsigned 128-bit integer type.
-  PREDEF_DECL_UNSIGNED_INT_128_ID = 7,
-
-  /// The internal 'instancetype' typedef.
-  PREDEF_DECL_OBJC_INSTANCETYPE_ID = 8,
-
-  /// The internal '__builtin_va_list' typedef.
-  PREDEF_DECL_BUILTIN_VA_LIST_ID = 9,
-
-  /// The internal '__va_list_tag' struct, if any.
-  PREDEF_DECL_VA_LIST_TAG = 10,
-
-  /// The internal '__builtin_ms_va_list' typedef.
-  PREDEF_DECL_BUILTIN_MS_VA_LIST_ID = 11,
-
-  /// The predeclared '_GUID' struct.
-  PREDEF_DECL_BUILTIN_MS_GUID_ID = 12,
-
-  /// The extern "C" context.
-  PREDEF_DECL_EXTERN_C_CONTEXT_ID = 13,
-
-  /// The internal '__make_integer_seq' template.
-  PREDEF_DECL_MAKE_INTEGER_SEQ_ID = 14,
-
-  /// The internal '__NSConstantString' typedef.
-  PREDEF_DECL_CF_CONSTANT_STRING_ID = 15,
-
-  /// The internal '__NSConstantString' tag type.
-  PREDEF_DECL_CF_CONSTANT_STRING_TAG_ID = 16,
-
-  /// The internal '__type_pack_element' template.
-  PREDEF_DECL_TYPE_PACK_ELEMENT_ID = 17,
-};
-
-/// The number of declaration IDs that are predefined.
-///
-/// For more information about predefined declarations, see the
-/// \c PredefinedDeclIDs type and the PREDEF_DECL_*_ID constants.
-const unsigned int NUM_PREDEF_DECL_IDS = 18;
-
 /// Record of updates for a declaration that was modified after
 /// being deserialized. This can occur within DECLTYPES_BLOCK_ID.
 const unsigned int DECL_UPDATES = 49;
@@ -2002,7 +1926,7 @@ enum StmtCode {
   STMT_OMP_TARGET_TEAMS_GENERIC_LOOP_DIRECTIVE,
   STMT_OMP_PARALLEL_GENERIC_LOOP_DIRECTIVE,
   STMT_OMP_TARGET_PARALLEL_GENERIC_LOOP_DIRECTIVE,
-  EXPR_OMP_ARRAY_SECTION,
+  EXPR_ARRAY_SECTION,
   EXPR_OMP_ARRAY_SHAPING,
   EXPR_OMP_ITERATOR,
 
@@ -2056,39 +1980,10 @@ enum CtorInitializerType {
 /// Kinds of cleanup objects owned by ExprWithCleanups.
 enum CleanupObjectKind { COK_Block, COK_CompoundLiteral };
 
-/// Describes the redeclarations of a declaration.
-struct LocalRedeclarationsInfo {
-  // The ID of the first declaration
-  DeclID FirstID;
-
-  // Offset into the array of redeclaration chains.
-  unsigned Offset;
-
-  friend bool operator<(const LocalRedeclarationsInfo &X,
-                        const LocalRedeclarationsInfo &Y) {
-    return X.FirstID < Y.FirstID;
-  }
-
-  friend bool operator>(const LocalRedeclarationsInfo &X,
-                        const LocalRedeclarationsInfo &Y) {
-    return X.FirstID > Y.FirstID;
-  }
-
-  friend bool operator<=(const LocalRedeclarationsInfo &X,
-                         const LocalRedeclarationsInfo &Y) {
-    return X.FirstID <= Y.FirstID;
-  }
-
-  friend bool operator>=(const LocalRedeclarationsInfo &X,
-                         const LocalRedeclarationsInfo &Y) {
-    return X.FirstID >= Y.FirstID;
-  }
-};
-
 /// Describes the categories of an Objective-C class.
 struct ObjCCategoriesInfo {
   // The ID of the definition
-  DeclID DefinitionID;
+  LocalDeclID DefinitionID;
 
   // Offset into the array of category lists.
   unsigned Offset;
diff --git a/clang/include/clang/Serialization/ASTDeserializationListener.h b/clang/include/clang/Serialization/ASTDeserializationListener.h
index f3a01a4b9731..3ab7f1a91843 100644
--- a/clang/include/clang/Serialization/ASTDeserializationListener.h
+++ b/clang/include/clang/Serialization/ASTDeserializationListener.h
@@ -44,7 +44,7 @@ public:
   ///        unqualified.
   virtual void TypeRead(serialization::TypeIdx Idx, QualType T) { }
   /// A decl was deserialized from the AST file.
-  virtual void DeclRead(serialization::DeclID ID, const Decl *D) { }
+  virtual void DeclRead(GlobalDeclID ID, const Decl *D) {}
   /// A selector was read from the AST file.
   virtual void SelectorRead(serialization::SelectorID iD, Selector Sel) {}
   /// A macro definition was read from the AST file.
diff --git a/clang/include/clang/Serialization/ASTReader.h b/clang/include/clang/Serialization/ASTReader.h
index 1cd8b6a357cb..64f1ebc117b3 100644
--- a/clang/include/clang/Serialization/ASTReader.h
+++ b/clang/include/clang/Serialization/ASTReader.h
@@ -501,10 +501,7 @@ private:
   /// = I + 1 has already been loaded.
   llvm::PagedVector<Decl *> DeclsLoaded;
 
-  static_assert(std::is_same_v<serialization::DeclID, Decl::DeclID>);
-
-  using GlobalDeclMapType =
-      ContinuousRangeMap<serialization::DeclID, ModuleFile *, 4>;
+  using GlobalDeclMapType = ContinuousRangeMap<GlobalDeclID, ModuleFile *, 4>;
 
   /// Mapping from global declaration IDs to the module in which the
   /// declaration resides.
@@ -512,16 +509,15 @@ private:
 
   using FileOffset = std::pair<ModuleFile *, uint64_t>;
   using FileOffsetsTy = SmallVector<FileOffset, 2>;
-  using DeclUpdateOffsetsMap =
-      llvm::DenseMap<serialization::DeclID, FileOffsetsTy>;
+  using DeclUpdateOffsetsMap = llvm::DenseMap<GlobalDeclID, FileOffsetsTy>;
 
   /// Declarations that have modifications residing in a later file
   /// in the chain.
   DeclUpdateOffsetsMap DeclUpdateOffsets;
 
-  using DelayedNamespaceOffsetMapTy = llvm::DenseMap<
-      serialization::DeclID,
-      std::pair</*LexicalOffset*/ uint64_t, /*VisibleOffset*/ uint64_t>>;
+  using DelayedNamespaceOffsetMapTy =
+      llvm::DenseMap<GlobalDeclID, std::pair</*LexicalOffset*/ uint64_t,
+                                             /*VisibleOffset*/ uint64_t>>;
 
   /// Mapping from global declaration IDs to the lexical and visible block
   /// offset for delayed namespace in reduced BMI.
@@ -535,13 +531,12 @@ private:
 
   struct PendingUpdateRecord {
     Decl *D;
-    serialization::GlobalDeclID ID;
+    GlobalDeclID ID;
 
     // Whether the declaration was just deserialized.
     bool JustLoaded;
 
-    PendingUpdateRecord(serialization::GlobalDeclID ID, Decl *D,
-                        bool JustLoaded)
+    PendingUpdateRecord(GlobalDeclID ID, Decl *D, bool JustLoaded)
         : D(D), ID(ID), JustLoaded(JustLoaded) {}
   };
 
@@ -594,10 +589,10 @@ private:
 
   struct FileDeclsInfo {
     ModuleFile *Mod = nullptr;
-    ArrayRef<serialization::LocalDeclID> Decls;
+    ArrayRef<LocalDeclID> Decls;
 
     FileDeclsInfo() = default;
-    FileDeclsInfo(ModuleFile *Mod, ArrayRef<serialization::LocalDeclID> Decls)
+    FileDeclsInfo(ModuleFile *Mod, ArrayRef<LocalDeclID> Decls)
         : Mod(Mod), Decls(Decls) {}
   };
 
@@ -606,7 +601,11 @@ private:
 
   /// An array of lexical contents of a declaration context, as a sequence of
   /// Decl::Kind, DeclID pairs.
-  using LexicalContents = ArrayRef<llvm::support::unaligned_uint32_t>;
+  using unalighed_decl_id_t =
+      llvm::support::detail::packed_endian_specific_integral<
+          serialization::DeclID, llvm::endianness::native,
+          llvm::support::unaligned>;
+  using LexicalContents = ArrayRef<unalighed_decl_id_t>;
 
   /// Map from a DeclContext to its lexical contents.
   llvm::DenseMap<const DeclContext*, std::pair<ModuleFile*, LexicalContents>>
@@ -631,8 +630,7 @@ private:
 
   /// Updates to the visible declarations of declaration contexts that
   /// haven't been loaded yet.
-  llvm::DenseMap<serialization::DeclID, DeclContextVisibleUpdates>
-      PendingVisibleUpdates;
+  llvm::DenseMap<GlobalDeclID, DeclContextVisibleUpdates> PendingVisibleUpdates;
 
   /// The set of C++ or Objective-C classes that have forward
   /// declarations that have not yet been linked to their definitions.
@@ -658,7 +656,7 @@ private:
   /// Read the record that describes the visible contents of a DC.
   bool ReadVisibleDeclContextStorage(ModuleFile &M,
                                      llvm::BitstreamCursor &Cursor,
-                                     uint64_t Offset, serialization::DeclID ID);
+                                     uint64_t Offset, GlobalDeclID ID);
 
   /// A vector containing identifiers that have already been
   /// loaded.
@@ -811,21 +809,26 @@ private:
   /// This contains the data loaded from all EAGERLY_DESERIALIZED_DECLS blocks
   /// in the chain. The referenced declarations are deserialized and passed to
   /// the consumer eagerly.
-  SmallVector<serialization::DeclID, 16> EagerlyDeserializedDecls;
+  SmallVector<GlobalDeclID, 16> EagerlyDeserializedDecls;
 
   /// The IDs of all tentative definitions stored in the chain.
   ///
   /// Sema keeps track of all tentative definitions in a TU because it has to
   /// complete them and pass them on to CodeGen. Thus, tentative definitions in
   /// the PCH chain must be eagerly deserialized.
-  SmallVector<serialization::DeclID, 16> TentativeDefinitions;
+  SmallVector<GlobalDeclID, 16> TentativeDefinitions;
 
   /// The IDs of all CXXRecordDecls stored in the chain whose VTables are
   /// used.
   ///
   /// CodeGen has to emit VTables for these records, so they have to be eagerly
   /// deserialized.
-  SmallVector<serialization::DeclID, 64> VTableUses;
+  struct VTableUse {
+    GlobalDeclID ID;
+    SourceLocation::UIntTy RawLoc;
+    bool Used;
+  };
+  SmallVector<VTableUse> VTableUses;
 
   /// A snapshot of the pending instantiations in the chain.
   ///
@@ -833,7 +836,11 @@ private:
   /// end of the TU. It consists of a pair of values for every pending
   /// instantiation where the first value is the ID of the decl and the second
   /// is the instantiation location.
-  SmallVector<serialization::DeclID, 64> PendingInstantiations;
+  struct PendingInstantiation {
+    GlobalDeclID ID;
+    SourceLocation::UIntTy RawLoc;
+  };
+  SmallVector<PendingInstantiation, 64> PendingInstantiations;
 
   //@}
 
@@ -843,11 +850,11 @@ private:
 
   /// A snapshot of Sema's unused file-scoped variable tracking, for
   /// generating warnings.
-  SmallVector<serialization::DeclID, 16> UnusedFileScopedDecls;
+  SmallVector<GlobalDeclID, 16> UnusedFileScopedDecls;
 
   /// A list of all the delegating constructors we've seen, to diagnose
   /// cycles.
-  SmallVector<serialization::DeclID, 4> DelegatingCtorDecls;
+  SmallVector<GlobalDeclID, 4> DelegatingCtorDecls;
 
   /// Method selectors used in a @selector expression. Used for
   /// implementation of -Wselector.
@@ -860,7 +867,7 @@ private:
   /// The IDs of type aliases for ext_vectors that exist in the chain.
   ///
   /// Used by Sema for finding sugared names for ext_vectors in diagnostics.
-  SmallVector<serialization::DeclID, 4> ExtVectorDecls;
+  SmallVector<GlobalDeclID, 4> ExtVectorDecls;
 
   //@}
 
@@ -871,7 +878,7 @@ private:
   /// The IDs of all potentially unused typedef names in the chain.
   ///
   /// Sema tracks these to emit warnings.
-  SmallVector<serialization::DeclID, 16> UnusedLocalTypedefNameCandidates;
+  SmallVector<GlobalDeclID, 16> UnusedLocalTypedefNameCandidates;
 
   /// Our current depth in #pragma cuda force_host_device begin/end
   /// macros.
@@ -880,7 +887,7 @@ private:
   /// The IDs of the declarations Sema stores directly.
   ///
   /// Sema tracks a few important decls, such as namespace std, directly.
-  SmallVector<serialization::DeclID, 4> SemaDeclRefs;
+  SmallVector<GlobalDeclID, 4> SemaDeclRefs;
 
   /// The IDs of the types ASTContext stores directly.
   ///
@@ -891,7 +898,7 @@ private:
   ///
   /// The AST context tracks a few important decls, currently cudaConfigureCall,
   /// directly.
-  SmallVector<serialization::DeclID, 2> CUDASpecialDeclRefs;
+  SmallVector<GlobalDeclID, 2> CUDASpecialDeclRefs;
 
   /// The floating point pragma option settings.
   SmallVector<uint64_t, 1> FPPragmaOptions;
@@ -940,11 +947,15 @@ private:
   llvm::DenseMap<const Decl *, std::set<std::string>> OpenCLDeclExtMap;
 
   /// A list of the namespaces we've seen.
-  SmallVector<serialization::DeclID, 4> KnownNamespaces;
+  SmallVector<GlobalDeclID, 4> KnownNamespaces;
 
   /// A list of undefined decls with internal linkage followed by the
   /// SourceLocation of a matching ODR-use.
-  SmallVector<serialization::DeclID, 8> UndefinedButUsed;
+  struct UndefinedButUsedDecl {
+    GlobalDeclID ID;
+    SourceLocation::UIntTy RawLoc;
+  };
+  SmallVector<UndefinedButUsedDecl, 8> UndefinedButUsed;
 
   /// Delete expressions to analyze at the end of translation unit.
   SmallVector<uint64_t, 8> DelayedDeleteExprs;
@@ -956,7 +967,7 @@ private:
   /// The IDs of all decls to be checked for deferred diags.
   ///
   /// Sema tracks these to emit deferred diags.
-  llvm::SmallSetVector<serialization::DeclID, 4> DeclsToCheckForDeferredDiags;
+  llvm::SmallSetVector<GlobalDeclID, 4> DeclsToCheckForDeferredDiags;
 
 private:
   struct ImportedSubmodule {
@@ -1093,8 +1104,8 @@ private:
   ///
   /// The declarations on the identifier chain for these identifiers will be
   /// loaded once the recursive loading has completed.
-  llvm::MapVector<IdentifierInfo *, SmallVector<uint32_t, 4>>
-    PendingIdentifierInfos;
+  llvm::MapVector<IdentifierInfo *, SmallVector<GlobalDeclID, 4>>
+      PendingIdentifierInfos;
 
   /// The set of lookup results that we have faked in order to support
   /// merging of partially deserialized decls but that we have not yet removed.
@@ -1138,8 +1149,8 @@ private:
   /// been loaded but its DeclContext was not set yet.
   struct PendingDeclContextInfo {
     Decl *D;
-    serialization::GlobalDeclID SemaDC;
-    serialization::GlobalDeclID LexicalDC;
+    GlobalDeclID SemaDC;
+    GlobalDeclID LexicalDC;
   };
 
   /// The set of Decls that have been loaded but their DeclContexts are
@@ -1220,8 +1231,7 @@ private:
   /// module is loaded.
   SmallVector<ObjCInterfaceDecl *, 16> ObjCClassesLoaded;
 
-  using KeyDeclsMap =
-      llvm::DenseMap<Decl *, SmallVector<serialization::DeclID, 2>>;
+  using KeyDeclsMap = llvm::DenseMap<Decl *, SmallVector<GlobalDeclID, 2>>;
 
   /// A mapping from canonical declarations to the set of global
   /// declaration IDs for key declaration that have been merged with that
@@ -1430,7 +1440,7 @@ private:
   QualType readTypeRecord(unsigned Index);
   RecordLocation TypeCursorForIndex(unsigned Index);
   void LoadedDecl(unsigned Index, Decl *D);
-  Decl *ReadDeclRecord(serialization::DeclID ID);
+  Decl *ReadDeclRecord(GlobalDeclID ID);
   void markIncompleteDeclChain(Decl *D);
 
   /// Returns the most recent declaration of a declaration (which must be
@@ -1438,11 +1448,10 @@ private:
   /// merged into its redecl chain.
   Decl *getMostRecentExistingDecl(Decl *D);
 
-  RecordLocation DeclCursorForID(serialization::DeclID ID,
-                                 SourceLocation &Location);
+  RecordLocation DeclCursorForID(GlobalDeclID ID, SourceLocation &Location);
   void loadDeclUpdateRecords(PendingUpdateRecord &Record);
   void loadPendingDeclChain(Decl *D, uint64_t LocalOffset);
-  void loadObjCCategories(serialization::GlobalDeclID ID, ObjCInterfaceDecl *D,
+  void loadObjCCategories(GlobalDeclID ID, ObjCInterfaceDecl *D,
                           unsigned PreviousGeneration = 0);
 
   RecordLocation getLocalBitOffset(uint64_t GlobalOffset);
@@ -1477,11 +1486,10 @@ private:
                                unsigned ClientLoadCapabilities);
 
 public:
-  class ModuleDeclIterator
-      : public llvm::iterator_adaptor_base<
-            ModuleDeclIterator, const serialization::LocalDeclID *,
-            std::random_access_iterator_tag, const Decl *, ptrdiff_t,
-            const Decl *, const Decl *> {
+  class ModuleDeclIterator : public llvm::iterator_adaptor_base<
+                                 ModuleDeclIterator, const LocalDeclID *,
+                                 std::random_access_iterator_tag, const Decl *,
+                                 ptrdiff_t, const Decl *, const Decl *> {
     ASTReader *Reader = nullptr;
     ModuleFile *Mod = nullptr;
 
@@ -1489,7 +1497,7 @@ public:
     ModuleDeclIterator() : iterator_adaptor_base(nullptr) {}
 
     ModuleDeclIterator(ASTReader *Reader, ModuleFile *Mod,
-                       const serialization::LocalDeclID *Pos)
+                       const LocalDeclID *Pos)
         : iterator_adaptor_base(Pos), Reader(Reader), Mod(Mod) {}
 
     value_type operator*() const {
@@ -1517,9 +1525,8 @@ private:
 
   void pushExternalDeclIntoScope(NamedDecl *D, DeclarationName Name);
 
-  void addPendingDeclContextInfo(Decl *D,
-                                 serialization::GlobalDeclID SemaDC,
-                                 serialization::GlobalDeclID LexicalDC) {
+  void addPendingDeclContextInfo(Decl *D, GlobalDeclID SemaDC,
+                                 GlobalDeclID LexicalDC) {
     assert(D);
     PendingDeclContextInfo Info = { D, SemaDC, LexicalDC };
     PendingDeclContextInfos.push_back(Info);
@@ -1897,38 +1904,36 @@ public:
 
   /// Map from a local declaration ID within a given module to a
   /// global declaration ID.
-  serialization::DeclID getGlobalDeclID(ModuleFile &F,
-                                      serialization::LocalDeclID LocalID) const;
+  GlobalDeclID getGlobalDeclID(ModuleFile &F, LocalDeclID LocalID) const;
 
   /// Returns true if global DeclID \p ID originated from module \p M.
-  bool isDeclIDFromModule(serialization::GlobalDeclID ID, ModuleFile &M) const;
+  bool isDeclIDFromModule(GlobalDeclID ID, ModuleFile &M) const;
 
   /// Retrieve the module file that owns the given declaration, or NULL
   /// if the declaration is not from a module file.
   ModuleFile *getOwningModuleFile(const Decl *D);
 
   /// Returns the source location for the decl \p ID.
-  SourceLocation getSourceLocationForDeclID(serialization::GlobalDeclID ID);
+  SourceLocation getSourceLocationForDeclID(GlobalDeclID ID);
 
   /// Resolve a declaration ID into a declaration, potentially
   /// building a new declaration.
-  Decl *GetDecl(serialization::DeclID ID);
-  Decl *GetExternalDecl(uint32_t ID) override;
+  Decl *GetDecl(GlobalDeclID ID);
+  Decl *GetExternalDecl(GlobalDeclID ID) override;
 
   /// Resolve a declaration ID into a declaration. Return 0 if it's not
   /// been loaded yet.
-  Decl *GetExistingDecl(serialization::DeclID ID);
+  Decl *GetExistingDecl(GlobalDeclID ID);
 
   /// Reads a declaration with the given local ID in the given module.
-  Decl *GetLocalDecl(ModuleFile &F, uint32_t LocalID) {
+  Decl *GetLocalDecl(ModuleFile &F, LocalDeclID LocalID) {
     return GetDecl(getGlobalDeclID(F, LocalID));
   }
 
   /// Reads a declaration with the given local ID in the given module.
   ///
   /// \returns The requested declaration, casted to the given return type.
-  template<typename T>
-  T *GetLocalDeclAs(ModuleFile &F, uint32_t LocalID) {
+  template <typename T> T *GetLocalDeclAs(ModuleFile &F, LocalDeclID LocalID) {
     return cast_or_null<T>(GetLocalDecl(F, LocalID));
   }
 
@@ -1937,16 +1942,15 @@ public:
   ///
   /// \returns the global ID of the given declaration as known in the given
   /// module file.
-  serialization::DeclID
-  mapGlobalIDToModuleFileGlobalID(ModuleFile &M,
-                                  serialization::DeclID GlobalID);
+  LocalDeclID mapGlobalIDToModuleFileGlobalID(ModuleFile &M,
+                                              GlobalDeclID GlobalID);
 
   /// Reads a declaration ID from the given position in a record in the
   /// given module.
   ///
   /// \returns The declaration ID read from the record, adjusted to a global ID.
-  serialization::DeclID ReadDeclID(ModuleFile &F, const RecordData &Record,
-                                   unsigned &Idx);
+  GlobalDeclID ReadDeclID(ModuleFile &F, const RecordData &Record,
+                          unsigned &Idx);
 
   /// Reads a declaration from the given position in a record in the
   /// given module.
@@ -2121,7 +2125,7 @@ public:
 
   void SetIdentifierInfo(unsigned ID, IdentifierInfo *II);
   void SetGloballyVisibleDecls(IdentifierInfo *II,
-                               const SmallVectorImpl<uint32_t> &DeclIDs,
+                               const SmallVectorImpl<GlobalDeclID> &DeclIDs,
                                SmallVectorImpl<Decl *> *Decls = nullptr);
 
   /// Report a diagnostic.
@@ -2363,7 +2367,7 @@ public:
 
   // Contains the IDs for declarations that were requested before we have
   // access to a Sema object.
-  SmallVector<uint64_t, 16> PreloadedDeclIDs;
+  SmallVector<GlobalDeclID, 16> PreloadedDeclIDs;
 
   /// Retrieve the semantic analysis object used to analyze the
   /// translation unit in which the precompiled header is being
diff --git a/clang/include/clang/Serialization/ASTRecordReader.h b/clang/include/clang/Serialization/ASTRecordReader.h
index 7dd1140106e4..06b80f266a94 100644
--- a/clang/include/clang/Serialization/ASTRecordReader.h
+++ b/clang/include/clang/Serialization/ASTRecordReader.h
@@ -103,13 +103,6 @@ public:
                                                  DC);
   }
 
-  /// Read the record that describes the visible contents of a DC.
-  bool readVisibleDeclContextStorage(uint64_t Offset,
-                                     serialization::DeclID ID) {
-    return Reader->ReadVisibleDeclContextStorage(*F, F->DeclsCursor, Offset,
-                                                 ID);
-  }
-
   ExplicitSpecifier readExplicitSpec() {
     uint64_t Kind = readInt();
     bool HasExpr = Kind & 0x1;
@@ -143,8 +136,7 @@ public:
   /// Reads a declaration with the given local ID in the given module.
   ///
   /// \returns The requested declaration, casted to the given return type.
-  template<typename T>
-  T *GetLocalDeclAs(uint32_t LocalID) {
+  template <typename T> T *GetLocalDeclAs(LocalDeclID LocalID) {
     return cast_or_null<T>(Reader->GetLocalDecl(*F, LocalID));
   }
 
@@ -190,9 +182,7 @@ public:
   /// Reads a declaration ID from the given position in this record.
   ///
   /// \returns The declaration ID read from the record, adjusted to a global ID.
-  serialization::DeclID readDeclID() {
-    return Reader->ReadDeclID(*F, Record, Idx);
-  }
+  GlobalDeclID readDeclID() { return Reader->ReadDeclID(*F, Record, Idx); }
 
   /// Reads a declaration from the given position in a record in the
   /// given module, advancing Idx.
diff --git a/clang/include/clang/Serialization/ASTWriter.h b/clang/include/clang/Serialization/ASTWriter.h
index 13b4ad4ad295..6c45b7348b85 100644
--- a/clang/include/clang/Serialization/ASTWriter.h
+++ b/clang/include/clang/Serialization/ASTWriter.h
@@ -212,10 +212,10 @@ private:
   llvm::SmallVector<NamespaceDecl *, 16> DelayedNamespace;
 
   /// The first ID number we can use for our own declarations.
-  serialization::DeclID FirstDeclID = serialization::NUM_PREDEF_DECL_IDS;
+  LocalDeclID FirstDeclID = LocalDeclID(clang::NUM_PREDEF_DECL_IDS);
 
   /// The decl ID that will be assigned to the next new decl.
-  serialization::DeclID NextDeclID = FirstDeclID;
+  LocalDeclID NextDeclID = FirstDeclID;
 
   /// Map that provides the ID numbers of each declaration within
   /// the output stream, as well as those deserialized from a chained PCH.
@@ -223,7 +223,7 @@ private:
   /// The ID numbers of declarations are consecutive (in order of
   /// discovery) and start at 2. 1 is reserved for the translation
   /// unit, while 0 is reserved for NULL.
-  llvm::DenseMap<const Decl *, serialization::DeclID> DeclIDs;
+  llvm::DenseMap<const Decl *, LocalDeclID> DeclIDs;
 
   /// Offset of each declaration in the bitstream, indexed by
   /// the declaration's ID.
@@ -233,9 +233,8 @@ private:
   /// are relative to this value.
   uint64_t DeclTypesBlockStartOffset = 0;
 
-  /// Sorted (by file offset) vector of pairs of file offset/DeclID.
-  using LocDeclIDsTy =
-      SmallVector<std::pair<unsigned, serialization::DeclID>, 64>;
+  /// Sorted (by file offset) vector of pairs of file offset/LocalDeclID.
+  using LocDeclIDsTy = SmallVector<std::pair<unsigned, LocalDeclID>, 64>;
   struct DeclIDInFileInfo {
     LocDeclIDsTy DeclIDs;
 
@@ -250,7 +249,7 @@ private:
   /// that it contains.
   FileDeclIDsTy FileDeclIDs;
 
-  void associateDeclWithFile(const Decl *D, serialization::DeclID);
+  void associateDeclWithFile(const Decl *D, LocalDeclID);
 
   /// The first ID number we can use for our own types.
   serialization::TypeID FirstTypeID = serialization::NUM_PREDEF_TYPE_IDS;
@@ -421,8 +420,8 @@ private:
   /// headers. The declarations themselves are stored as declaration
   /// IDs, since they will be written out to an EAGERLY_DESERIALIZED_DECLS
   /// record.
-  SmallVector<serialization::DeclID, 16> EagerlyDeserializedDecls;
-  SmallVector<serialization::DeclID, 16> ModularCodegenDecls;
+  RecordData EagerlyDeserializedDecls;
+  RecordData ModularCodegenDecls;
 
   /// DeclContexts that have received extensions since their serialized
   /// form.
@@ -709,7 +708,7 @@ public:
       return false;
     auto I = DeclIDs.find(D);
     return (I == DeclIDs.end() ||
-            I->second >= serialization::NUM_PREDEF_DECL_IDS);
+            I->second.get() >= clang::NUM_PREDEF_DECL_IDS);
   };
 
   /// Emit a reference to a declaration.
@@ -717,12 +716,13 @@ public:
   // Emit a reference to a declaration if the declaration was emitted.
   void AddEmittedDeclRef(const Decl *D, RecordDataImpl &Record);
 
-  /// Force a declaration to be emitted and get its ID.
-  serialization::DeclID GetDeclRef(const Decl *D);
+  /// Force a declaration to be emitted and get its local ID to the module file
+  /// been writing.
+  LocalDeclID GetDeclRef(const Decl *D);
 
-  /// Determine the declaration ID of an already-emitted
+  /// Determine the local declaration ID of an already-emitted
   /// declaration.
-  serialization::DeclID getDeclID(const Decl *D);
+  LocalDeclID getDeclID(const Decl *D);
 
   /// Whether or not the declaration got emitted. If not, it wouldn't be
   /// emitted.
diff --git a/clang/include/clang/Serialization/ModuleFile.h b/clang/include/clang/Serialization/ModuleFile.h
index bc0aa89966c2..25f644e76edb 100644
--- a/clang/include/clang/Serialization/ModuleFile.h
+++ b/clang/include/clang/Serialization/ModuleFile.h
@@ -462,7 +462,7 @@ public:
   serialization::DeclID BaseDeclID = 0;
 
   /// Remapping table for declaration IDs in this module.
-  ContinuousRangeMap<uint32_t, int, 2> DeclRemap;
+  ContinuousRangeMap<serialization::DeclID, int, 2> DeclRemap;
 
   /// Mapping from the module files that this module file depends on
   /// to the base declaration ID for that module as it is understood within this
@@ -474,7 +474,7 @@ public:
   llvm::DenseMap<ModuleFile *, serialization::DeclID> GlobalToLocalDeclIDs;
 
   /// Array of file-level DeclIDs sorted by file.
-  const serialization::DeclID *FileSortedDecls = nullptr;
+  const LocalDeclID *FileSortedDecls = nullptr;
   unsigned NumFileSortedDecls = 0;
 
   /// Array of category list location information within this
diff --git a/clang/include/clang/StaticAnalyzer/Core/PathSensitive/Store.h b/clang/include/clang/StaticAnalyzer/Core/PathSensitive/Store.h
index fac0c04ae2ca..ef23b160a3c0 100644
--- a/clang/include/clang/StaticAnalyzer/Core/PathSensitive/Store.h
+++ b/clang/include/clang/StaticAnalyzer/Core/PathSensitive/Store.h
@@ -225,15 +225,11 @@ public:
   ///   invalidated. This should include any regions explicitly invalidated
   ///   even if they do not currently have bindings. Pass \c NULL if this
   ///   information will not be used.
-  virtual StoreRef invalidateRegions(Store store,
-                                  ArrayRef<SVal> Values,
-                                  const Expr *E, unsigned Count,
-                                  const LocationContext *LCtx,
-                                  const CallEvent *Call,
-                                  InvalidatedSymbols &IS,
-                                  RegionAndSymbolInvalidationTraits &ITraits,
-                                  InvalidatedRegions *InvalidatedTopLevel,
-                                  InvalidatedRegions *Invalidated) = 0;
+  virtual StoreRef invalidateRegions(
+      Store store, ArrayRef<SVal> Values, const Expr *Ex, unsigned Count,
+      const LocationContext *LCtx, const CallEvent *Call,
+      InvalidatedSymbols &IS, RegionAndSymbolInvalidationTraits &ITraits,
+      InvalidatedRegions *TopLevelRegions, InvalidatedRegions *Invalidated) = 0;
 
   /// enterStackFrame - Let the StoreManager to do something when execution
   /// engine is about to execute into a callee.
diff --git a/clang/include/clang/Support/RISCVVIntrinsicUtils.h b/clang/include/clang/Support/RISCVVIntrinsicUtils.h
index ef9d6c15724b..97493bae5656 100644
--- a/clang/include/clang/Support/RISCVVIntrinsicUtils.h
+++ b/clang/include/clang/Support/RISCVVIntrinsicUtils.h
@@ -502,7 +502,8 @@ enum RVVRequire : uint32_t {
   RVV_REQ_Zvksed = 1 << 14,
   RVV_REQ_Zvksh = 1 << 15,
   RVV_REQ_Zvfbfwma = 1 << 16,
-  RVV_REQ_Experimental = 1 << 17,
+  RVV_REQ_Zvfbfmin = 1 << 17,
+  RVV_REQ_Experimental = 1 << 18,
 
   LLVM_MARK_AS_BITMASK_ENUM(RVV_REQ_Experimental)
 };
diff --git a/clang/lib/APINotes/APINotesFormat.h b/clang/lib/APINotes/APINotesFormat.h
index 615314c46f09..97e630e97fdc 100644
--- a/clang/lib/APINotes/APINotesFormat.h
+++ b/clang/lib/APINotes/APINotesFormat.h
@@ -24,7 +24,10 @@ const uint16_t VERSION_MAJOR = 0;
 /// API notes file minor version number.
 ///
 /// When the format changes IN ANY WAY, this number should be incremented.
-const uint16_t VERSION_MINOR = 25; // SwiftImportAs
+const uint16_t VERSION_MINOR = 26; // SwiftCopyable
+
+const uint8_t kSwiftCopyable = 1;
+const uint8_t kSwiftNonCopyable = 2;
 
 using IdentifierID = llvm::PointerEmbeddedInt<unsigned, 31>;
 using IdentifierIDField = llvm::BCVBR<16>;
diff --git a/clang/lib/APINotes/APINotesReader.cpp b/clang/lib/APINotes/APINotesReader.cpp
index dfc3beb6fa13..b60ca685f62c 100644
--- a/clang/lib/APINotes/APINotesReader.cpp
+++ b/clang/lib/APINotes/APINotesReader.cpp
@@ -527,6 +527,13 @@ public:
       Info.EnumExtensibility =
           static_cast<EnumExtensibilityKind>((Payload & 0x3) - 1);
 
+    uint8_t Copyable =
+        endian::readNext<uint8_t, llvm::endianness::little>(Data);
+    if (Copyable == kSwiftNonCopyable)
+      Info.setSwiftCopyable(std::optional(false));
+    else if (Copyable == kSwiftCopyable)
+      Info.setSwiftCopyable(std::optional(true));
+
     unsigned ImportAsLength =
         endian::readNext<uint16_t, llvm::endianness::little>(Data);
     if (ImportAsLength > 0) {
diff --git a/clang/lib/APINotes/APINotesWriter.cpp b/clang/lib/APINotes/APINotesWriter.cpp
index e3f5d102fcd0..3e6159763150 100644
--- a/clang/lib/APINotes/APINotesWriter.cpp
+++ b/clang/lib/APINotes/APINotesWriter.cpp
@@ -1128,7 +1128,7 @@ public:
     return 2 + (TI.SwiftImportAs ? TI.SwiftImportAs->size() : 0) +
            2 + (TI.SwiftRetainOp ? TI.SwiftRetainOp->size() : 0) +
            2 + (TI.SwiftReleaseOp ? TI.SwiftReleaseOp->size() : 0) +
-           1 + getCommonTypeInfoSize(TI);
+           2 + getCommonTypeInfoSize(TI);
   }
 
   void emitUnversionedInfo(raw_ostream &OS, const TagInfo &TI) {
@@ -1146,6 +1146,11 @@ public:
 
     writer.write<uint8_t>(Flags);
 
+    if (auto Copyable = TI.isSwiftCopyable())
+      writer.write<uint8_t>(*Copyable ? kSwiftCopyable : kSwiftNonCopyable);
+    else
+      writer.write<uint8_t>(0);
+
     if (auto ImportAs = TI.SwiftImportAs) {
       writer.write<uint16_t>(ImportAs->size() + 1);
       OS.write(ImportAs->c_str(), ImportAs->size());
diff --git a/clang/lib/APINotes/APINotesYAMLCompiler.cpp b/clang/lib/APINotes/APINotesYAMLCompiler.cpp
index 57d6da7a1775..2295d769d344 100644
--- a/clang/lib/APINotes/APINotesYAMLCompiler.cpp
+++ b/clang/lib/APINotes/APINotesYAMLCompiler.cpp
@@ -419,6 +419,7 @@ struct Tag {
   std::optional<EnumExtensibilityKind> EnumExtensibility;
   std::optional<bool> FlagEnum;
   std::optional<EnumConvenienceAliasKind> EnumConvenienceKind;
+  std::optional<bool> SwiftCopyable;
 };
 
 typedef std::vector<Tag> TagsSeq;
@@ -452,6 +453,7 @@ template <> struct MappingTraits<Tag> {
     IO.mapOptional("EnumExtensibility", T.EnumExtensibility);
     IO.mapOptional("FlagEnum", T.FlagEnum);
     IO.mapOptional("EnumKind", T.EnumConvenienceKind);
+    IO.mapOptional("SwiftCopyable", T.SwiftCopyable);
   }
 };
 } // namespace yaml
@@ -1009,6 +1011,9 @@ public:
       if (Tag.SwiftReleaseOp)
         TI.SwiftReleaseOp = Tag.SwiftReleaseOp;
 
+      if (Tag.SwiftCopyable)
+        TI.setSwiftCopyable(Tag.SwiftCopyable);
+
       if (Tag.EnumConvenienceKind) {
         if (Tag.EnumExtensibility) {
           emitError(
diff --git a/clang/lib/AST/ASTContext.cpp b/clang/lib/AST/ASTContext.cpp
index b36fb5523af5..cbf4932aff9a 100644
--- a/clang/lib/AST/ASTContext.cpp
+++ b/clang/lib/AST/ASTContext.cpp
@@ -1083,7 +1083,8 @@ void ASTContext::addModuleInitializer(Module *M, Decl *D) {
   Inits->Initializers.push_back(D);
 }
 
-void ASTContext::addLazyModuleInitializers(Module *M, ArrayRef<uint32_t> IDs) {
+void ASTContext::addLazyModuleInitializers(Module *M,
+                                           ArrayRef<GlobalDeclID> IDs) {
   auto *&Inits = ModuleInitializers[M];
   if (!Inits)
     Inits = new (*this) PerModuleInitializers;
@@ -1320,16 +1321,14 @@ void ASTContext::InitBuiltinTypes(const TargetInfo &Target,
 
   // Placeholder type for OMP array sections.
   if (LangOpts.OpenMP) {
-    InitBuiltinType(OMPArraySectionTy, BuiltinType::OMPArraySection);
+    InitBuiltinType(ArraySectionTy, BuiltinType::ArraySection);
     InitBuiltinType(OMPArrayShapingTy, BuiltinType::OMPArrayShaping);
     InitBuiltinType(OMPIteratorTy, BuiltinType::OMPIterator);
   }
-  // Placeholder type for OpenACC array sections.
-  if (LangOpts.OpenACC) {
-    // FIXME: Once we implement OpenACC array sections in Sema, this will either
-    // be combined with the OpenMP type, or given its own type. In the meantime,
-    // just use the OpenMP type so that parsing can work.
-    InitBuiltinType(OMPArraySectionTy, BuiltinType::OMPArraySection);
+  // Placeholder type for OpenACC array sections, if we are ALSO in OMP mode,
+  // don't bother, as we're just using the same type as OMP.
+  if (LangOpts.OpenACC && !LangOpts.OpenMP) {
+    InitBuiltinType(ArraySectionTy, BuiltinType::ArraySection);
   }
   if (LangOpts.MatrixTypes)
     InitBuiltinType(IncompleteMatrixIdxTy, BuiltinType::IncompleteMatrixIdx);
diff --git a/clang/lib/AST/ComputeDependence.cpp b/clang/lib/AST/ComputeDependence.cpp
index 5ec3013fabba..bad8e75b2f87 100644
--- a/clang/lib/AST/ComputeDependence.cpp
+++ b/clang/lib/AST/ComputeDependence.cpp
@@ -443,12 +443,17 @@ ExprDependence clang::computeDependence(ObjCIndirectCopyRestoreExpr *E) {
   return E->getSubExpr()->getDependence();
 }
 
-ExprDependence clang::computeDependence(OMPArraySectionExpr *E) {
+ExprDependence clang::computeDependence(ArraySectionExpr *E) {
   auto D = E->getBase()->getDependence();
   if (auto *LB = E->getLowerBound())
     D |= LB->getDependence();
   if (auto *Len = E->getLength())
     D |= Len->getDependence();
+
+  if (E->isOMPArraySection()) {
+    if (auto *Stride = E->getStride())
+      D |= Stride->getDependence();
+  }
   return D;
 }
 
diff --git a/clang/lib/AST/Decl.cpp b/clang/lib/AST/Decl.cpp
index 474e0ccde5bb..e7e95c16b697 100644
--- a/clang/lib/AST/Decl.cpp
+++ b/clang/lib/AST/Decl.cpp
@@ -2151,7 +2151,7 @@ VarDecl *VarDecl::Create(ASTContext &C, DeclContext *DC, SourceLocation StartL,
   return new (C, DC) VarDecl(Var, C, DC, StartL, IdL, Id, T, TInfo, S);
 }
 
-VarDecl *VarDecl::CreateDeserialized(ASTContext &C, Decl::DeclID ID) {
+VarDecl *VarDecl::CreateDeserialized(ASTContext &C, GlobalDeclID ID) {
   return new (C, ID)
       VarDecl(Var, C, nullptr, SourceLocation(), SourceLocation(), nullptr,
               QualType(), nullptr, SC_None);
@@ -2929,7 +2929,7 @@ QualType ParmVarDecl::getOriginalType() const {
   return T;
 }
 
-ParmVarDecl *ParmVarDecl::CreateDeserialized(ASTContext &C, Decl::DeclID ID) {
+ParmVarDecl *ParmVarDecl::CreateDeserialized(ASTContext &C, GlobalDeclID ID) {
   return new (C, ID)
       ParmVarDecl(ParmVar, C, nullptr, SourceLocation(), SourceLocation(),
                   nullptr, QualType(), nullptr, SC_None, nullptr);
@@ -4553,7 +4553,7 @@ FieldDecl *FieldDecl::Create(const ASTContext &C, DeclContext *DC,
                                BW, Mutable, InitStyle);
 }
 
-FieldDecl *FieldDecl::CreateDeserialized(ASTContext &C, Decl::DeclID ID) {
+FieldDecl *FieldDecl::CreateDeserialized(ASTContext &C, GlobalDeclID ID) {
   return new (C, ID) FieldDecl(Field, nullptr, SourceLocation(),
                                SourceLocation(), nullptr, QualType(), nullptr,
                                nullptr, false, ICIS_NoInit);
@@ -4863,7 +4863,7 @@ EnumDecl *EnumDecl::Create(ASTContext &C, DeclContext *DC,
   return Enum;
 }
 
-EnumDecl *EnumDecl::CreateDeserialized(ASTContext &C, Decl::DeclID ID) {
+EnumDecl *EnumDecl::CreateDeserialized(ASTContext &C, GlobalDeclID ID) {
   EnumDecl *Enum =
       new (C, ID) EnumDecl(C, nullptr, SourceLocation(), SourceLocation(),
                            nullptr, nullptr, false, false, false);
@@ -5025,7 +5025,8 @@ RecordDecl *RecordDecl::Create(const ASTContext &C, TagKind TK, DeclContext *DC,
   return R;
 }
 
-RecordDecl *RecordDecl::CreateDeserialized(const ASTContext &C, Decl::DeclID ID) {
+RecordDecl *RecordDecl::CreateDeserialized(const ASTContext &C,
+                                           GlobalDeclID ID) {
   RecordDecl *R = new (C, ID)
       RecordDecl(Record, TagTypeKind::Struct, C, nullptr, SourceLocation(),
                  SourceLocation(), nullptr, nullptr);
@@ -5297,7 +5298,7 @@ PragmaCommentDecl *PragmaCommentDecl::Create(const ASTContext &C,
 }
 
 PragmaCommentDecl *PragmaCommentDecl::CreateDeserialized(ASTContext &C,
-                                                         Decl::DeclID ID,
+                                                         GlobalDeclID ID,
                                                          unsigned ArgSize) {
   return new (C, ID, additionalSizeToAlloc<char>(ArgSize + 1))
       PragmaCommentDecl(nullptr, SourceLocation(), PCK_Unknown);
@@ -5322,7 +5323,7 @@ PragmaDetectMismatchDecl::Create(const ASTContext &C, TranslationUnitDecl *DC,
 }
 
 PragmaDetectMismatchDecl *
-PragmaDetectMismatchDecl::CreateDeserialized(ASTContext &C, Decl::DeclID ID,
+PragmaDetectMismatchDecl::CreateDeserialized(ASTContext &C, GlobalDeclID ID,
                                              unsigned NameValueSize) {
   return new (C, ID, additionalSizeToAlloc<char>(NameValueSize + 1))
       PragmaDetectMismatchDecl(nullptr, SourceLocation(), 0);
@@ -5349,7 +5350,7 @@ LabelDecl *LabelDecl::Create(ASTContext &C, DeclContext *DC,
   return new (C, DC) LabelDecl(DC, IdentL, II, nullptr, GnuLabelL);
 }
 
-LabelDecl *LabelDecl::CreateDeserialized(ASTContext &C, Decl::DeclID ID) {
+LabelDecl *LabelDecl::CreateDeserialized(ASTContext &C, GlobalDeclID ID) {
   return new (C, ID) LabelDecl(nullptr, SourceLocation(), nullptr, nullptr,
                                SourceLocation());
 }
@@ -5390,7 +5391,7 @@ ImplicitParamDecl *ImplicitParamDecl::Create(ASTContext &C, QualType Type,
 }
 
 ImplicitParamDecl *ImplicitParamDecl::CreateDeserialized(ASTContext &C,
-                                                         Decl::DeclID ID) {
+                                                         GlobalDeclID ID) {
   return new (C, ID) ImplicitParamDecl(C, QualType(), ImplicitParamKind::Other);
 }
 
@@ -5408,7 +5409,7 @@ FunctionDecl::Create(ASTContext &C, DeclContext *DC, SourceLocation StartLoc,
   return New;
 }
 
-FunctionDecl *FunctionDecl::CreateDeserialized(ASTContext &C, Decl::DeclID ID) {
+FunctionDecl *FunctionDecl::CreateDeserialized(ASTContext &C, GlobalDeclID ID) {
   return new (C, ID) FunctionDecl(
       Function, C, nullptr, SourceLocation(), DeclarationNameInfo(), QualType(),
       nullptr, SC_None, false, false, ConstexprSpecKind::Unspecified, nullptr);
@@ -5418,7 +5419,7 @@ BlockDecl *BlockDecl::Create(ASTContext &C, DeclContext *DC, SourceLocation L) {
   return new (C, DC) BlockDecl(DC, L);
 }
 
-BlockDecl *BlockDecl::CreateDeserialized(ASTContext &C, Decl::DeclID ID) {
+BlockDecl *BlockDecl::CreateDeserialized(ASTContext &C, GlobalDeclID ID) {
   return new (C, ID) BlockDecl(nullptr, SourceLocation());
 }
 
@@ -5432,7 +5433,7 @@ CapturedDecl *CapturedDecl::Create(ASTContext &C, DeclContext *DC,
       CapturedDecl(DC, NumParams);
 }
 
-CapturedDecl *CapturedDecl::CreateDeserialized(ASTContext &C, Decl::DeclID ID,
+CapturedDecl *CapturedDecl::CreateDeserialized(ASTContext &C, GlobalDeclID ID,
                                                unsigned NumParams) {
   return new (C, ID, additionalSizeToAlloc<ImplicitParamDecl *>(NumParams))
       CapturedDecl(nullptr, NumParams);
@@ -5458,8 +5459,8 @@ EnumConstantDecl *EnumConstantDecl::Create(ASTContext &C, EnumDecl *CD,
   return new (C, CD) EnumConstantDecl(C, CD, L, Id, T, E, V);
 }
 
-EnumConstantDecl *
-EnumConstantDecl::CreateDeserialized(ASTContext &C, Decl::DeclID ID) {
+EnumConstantDecl *EnumConstantDecl::CreateDeserialized(ASTContext &C,
+                                                       GlobalDeclID ID) {
   return new (C, ID) EnumConstantDecl(C, nullptr, SourceLocation(), nullptr,
                                       QualType(), nullptr, llvm::APSInt());
 }
@@ -5486,7 +5487,7 @@ IndirectFieldDecl::Create(ASTContext &C, DeclContext *DC, SourceLocation L,
 }
 
 IndirectFieldDecl *IndirectFieldDecl::CreateDeserialized(ASTContext &C,
-                                                         Decl::DeclID ID) {
+                                                         GlobalDeclID ID) {
   return new (C, ID)
       IndirectFieldDecl(C, nullptr, SourceLocation(), DeclarationName(),
                         QualType(), std::nullopt);
@@ -5547,7 +5548,7 @@ bool TypedefNameDecl::isTransparentTagSlow() const {
   return isTransparent;
 }
 
-TypedefDecl *TypedefDecl::CreateDeserialized(ASTContext &C, Decl::DeclID ID) {
+TypedefDecl *TypedefDecl::CreateDeserialized(ASTContext &C, GlobalDeclID ID) {
   return new (C, ID) TypedefDecl(C, nullptr, SourceLocation(), SourceLocation(),
                                  nullptr, nullptr);
 }
@@ -5560,7 +5561,8 @@ TypeAliasDecl *TypeAliasDecl::Create(ASTContext &C, DeclContext *DC,
   return new (C, DC) TypeAliasDecl(C, DC, StartLoc, IdLoc, Id, TInfo);
 }
 
-TypeAliasDecl *TypeAliasDecl::CreateDeserialized(ASTContext &C, Decl::DeclID ID) {
+TypeAliasDecl *TypeAliasDecl::CreateDeserialized(ASTContext &C,
+                                                 GlobalDeclID ID) {
   return new (C, ID) TypeAliasDecl(C, nullptr, SourceLocation(),
                                    SourceLocation(), nullptr, nullptr);
 }
@@ -5591,7 +5593,7 @@ FileScopeAsmDecl *FileScopeAsmDecl::Create(ASTContext &C, DeclContext *DC,
 }
 
 FileScopeAsmDecl *FileScopeAsmDecl::CreateDeserialized(ASTContext &C,
-                                                       Decl::DeclID ID) {
+                                                       GlobalDeclID ID) {
   return new (C, ID) FileScopeAsmDecl(nullptr, nullptr, SourceLocation(),
                                       SourceLocation());
 }
@@ -5609,7 +5611,7 @@ TopLevelStmtDecl *TopLevelStmtDecl::Create(ASTContext &C, Stmt *Statement) {
 }
 
 TopLevelStmtDecl *TopLevelStmtDecl::CreateDeserialized(ASTContext &C,
-                                                       Decl::DeclID ID) {
+                                                       GlobalDeclID ID) {
   return new (C, ID)
       TopLevelStmtDecl(/*DC=*/nullptr, SourceLocation(), /*S=*/nullptr);
 }
@@ -5630,7 +5632,7 @@ EmptyDecl *EmptyDecl::Create(ASTContext &C, DeclContext *DC, SourceLocation L) {
   return new (C, DC) EmptyDecl(DC, L);
 }
 
-EmptyDecl *EmptyDecl::CreateDeserialized(ASTContext &C, Decl::DeclID ID) {
+EmptyDecl *EmptyDecl::CreateDeserialized(ASTContext &C, GlobalDeclID ID) {
   return new (C, ID) EmptyDecl(nullptr, SourceLocation());
 }
 
@@ -5663,7 +5665,8 @@ HLSLBufferDecl *HLSLBufferDecl::Create(ASTContext &C,
   return Result;
 }
 
-HLSLBufferDecl *HLSLBufferDecl::CreateDeserialized(ASTContext &C, Decl::DeclID ID) {
+HLSLBufferDecl *HLSLBufferDecl::CreateDeserialized(ASTContext &C,
+                                                   GlobalDeclID ID) {
   return new (C, ID) HLSLBufferDecl(nullptr, false, SourceLocation(), nullptr,
                                     SourceLocation(), SourceLocation());
 }
@@ -5719,7 +5722,7 @@ ImportDecl *ImportDecl::CreateImplicit(ASTContext &C, DeclContext *DC,
   return Import;
 }
 
-ImportDecl *ImportDecl::CreateDeserialized(ASTContext &C, Decl::DeclID ID,
+ImportDecl *ImportDecl::CreateDeserialized(ASTContext &C, GlobalDeclID ID,
                                            unsigned NumLocations) {
   return new (C, ID, additionalSizeToAlloc<SourceLocation>(NumLocations))
       ImportDecl(EmptyShell());
@@ -5752,6 +5755,6 @@ ExportDecl *ExportDecl::Create(ASTContext &C, DeclContext *DC,
   return new (C, DC) ExportDecl(DC, ExportLoc);
 }
 
-ExportDecl *ExportDecl::CreateDeserialized(ASTContext &C, Decl::DeclID ID) {
+ExportDecl *ExportDecl::CreateDeserialized(ASTContext &C, GlobalDeclID ID) {
   return new (C, ID) ExportDecl(nullptr, SourceLocation());
 }
diff --git a/clang/lib/AST/DeclBase.cpp b/clang/lib/AST/DeclBase.cpp
index 7cb6b31c541f..f341c74cf86e 100644
--- a/clang/lib/AST/DeclBase.cpp
+++ b/clang/lib/AST/DeclBase.cpp
@@ -71,7 +71,7 @@ void Decl::updateOutOfDate(IdentifierInfo &II) const {
 #include "clang/AST/DeclNodes.inc"
 
 void *Decl::operator new(std::size_t Size, const ASTContext &Context,
-                         Decl::DeclID ID, std::size_t Extra) {
+                         GlobalDeclID ID, std::size_t Extra) {
   // Allocate an extra 8 bytes worth of storage, which ensures that the
   // resulting pointer will still be 8-byte aligned.
   static_assert(sizeof(unsigned) * 2 >= alignof(Decl),
@@ -85,7 +85,7 @@ void *Decl::operator new(std::size_t Size, const ASTContext &Context,
   PrefixPtr[0] = 0;
 
   // Store the global declaration ID in the second 4 bytes.
-  PrefixPtr[1] = ID;
+  PrefixPtr[1] = ID.get();
 
   return Result;
 }
@@ -1115,7 +1115,9 @@ int64_t Decl::getID() const {
 
 const FunctionType *Decl::getFunctionType(bool BlocksToo) const {
   QualType Ty;
-  if (const auto *D = dyn_cast<ValueDecl>(this))
+  if (const auto *D = dyn_cast<BindingDecl>(this))
+    return nullptr;
+  else if (const auto *D = dyn_cast<ValueDecl>(this))
     Ty = D->getType();
   else if (const auto *D = dyn_cast<TypedefNameDecl>(this))
     Ty = D->getUnderlyingType();
diff --git a/clang/lib/AST/DeclCXX.cpp b/clang/lib/AST/DeclCXX.cpp
index 426c52620510..75c441293d62 100644
--- a/clang/lib/AST/DeclCXX.cpp
+++ b/clang/lib/AST/DeclCXX.cpp
@@ -57,7 +57,8 @@ using namespace clang;
 
 void AccessSpecDecl::anchor() {}
 
-AccessSpecDecl *AccessSpecDecl::CreateDeserialized(ASTContext &C, Decl::DeclID ID) {
+AccessSpecDecl *AccessSpecDecl::CreateDeserialized(ASTContext &C,
+                                                   GlobalDeclID ID) {
   return new (C, ID) AccessSpecDecl(EmptyShell());
 }
 
@@ -68,7 +69,7 @@ void LazyASTUnresolvedSet::getFromExternalSource(ASTContext &C) const {
 
   for (ASTUnresolvedSet::iterator I = Impl.begin(); I != Impl.end(); ++I)
     I.setDecl(cast<NamedDecl>(Source->GetExternalDecl(
-        reinterpret_cast<uintptr_t>(I.getDecl()) >> 2)));
+        GlobalDeclID(reinterpret_cast<uintptr_t>(I.getDecl()) >> 2))));
   Impl.Decls.setLazy(false);
 }
 
@@ -160,8 +161,8 @@ CXXRecordDecl::CreateLambda(const ASTContext &C, DeclContext *DC,
   return R;
 }
 
-CXXRecordDecl *
-CXXRecordDecl::CreateDeserialized(const ASTContext &C, Decl::DeclID ID) {
+CXXRecordDecl *CXXRecordDecl::CreateDeserialized(const ASTContext &C,
+                                                 GlobalDeclID ID) {
   auto *R = new (C, ID)
       CXXRecordDecl(CXXRecord, TagTypeKind::Struct, C, nullptr,
                     SourceLocation(), SourceLocation(), nullptr, nullptr);
@@ -2162,8 +2163,8 @@ CXXDeductionGuideDecl *CXXDeductionGuideDecl::Create(
                                            TInfo, EndLocation, Ctor, Kind);
 }
 
-CXXDeductionGuideDecl *CXXDeductionGuideDecl::CreateDeserialized(ASTContext &C,
-                                                                 Decl::DeclID ID) {
+CXXDeductionGuideDecl *
+CXXDeductionGuideDecl::CreateDeserialized(ASTContext &C, GlobalDeclID ID) {
   return new (C, ID) CXXDeductionGuideDecl(
       C, nullptr, SourceLocation(), ExplicitSpecifier(), DeclarationNameInfo(),
       QualType(), nullptr, SourceLocation(), nullptr,
@@ -2175,8 +2176,8 @@ RequiresExprBodyDecl *RequiresExprBodyDecl::Create(
   return new (C, DC) RequiresExprBodyDecl(C, DC, StartLoc);
 }
 
-RequiresExprBodyDecl *RequiresExprBodyDecl::CreateDeserialized(ASTContext &C,
-                                                               Decl::DeclID ID) {
+RequiresExprBodyDecl *
+RequiresExprBodyDecl::CreateDeserialized(ASTContext &C, GlobalDeclID ID) {
   return new (C, ID) RequiresExprBodyDecl(C, nullptr, SourceLocation());
 }
 
@@ -2281,7 +2282,8 @@ CXXMethodDecl::Create(ASTContext &C, CXXRecordDecl *RD, SourceLocation StartLoc,
       isInline, ConstexprKind, EndLocation, TrailingRequiresClause);
 }
 
-CXXMethodDecl *CXXMethodDecl::CreateDeserialized(ASTContext &C, Decl::DeclID ID) {
+CXXMethodDecl *CXXMethodDecl::CreateDeserialized(ASTContext &C,
+                                                 GlobalDeclID ID) {
   return new (C, ID) CXXMethodDecl(
       CXXMethod, C, nullptr, SourceLocation(), DeclarationNameInfo(),
       QualType(), nullptr, SC_None, false, false,
@@ -2699,7 +2701,7 @@ CXXConstructorDecl::CXXConstructorDecl(
 void CXXConstructorDecl::anchor() {}
 
 CXXConstructorDecl *CXXConstructorDecl::CreateDeserialized(ASTContext &C,
-                                                           Decl::DeclID ID,
+                                                           GlobalDeclID ID,
                                                            uint64_t AllocKind) {
   bool hasTrailingExplicit = static_cast<bool>(AllocKind & TAKHasTailExplicit);
   bool isInheritingConstructor =
@@ -2845,8 +2847,8 @@ bool CXXConstructorDecl::isSpecializationCopyingObject() const {
 
 void CXXDestructorDecl::anchor() {}
 
-CXXDestructorDecl *
-CXXDestructorDecl::CreateDeserialized(ASTContext &C, Decl::DeclID ID) {
+CXXDestructorDecl *CXXDestructorDecl::CreateDeserialized(ASTContext &C,
+                                                         GlobalDeclID ID) {
   return new (C, ID) CXXDestructorDecl(
       C, nullptr, SourceLocation(), DeclarationNameInfo(), QualType(), nullptr,
       false, false, false, ConstexprSpecKind::Unspecified, nullptr);
@@ -2877,8 +2879,8 @@ void CXXDestructorDecl::setOperatorDelete(FunctionDecl *OD, Expr *ThisArg) {
 
 void CXXConversionDecl::anchor() {}
 
-CXXConversionDecl *
-CXXConversionDecl::CreateDeserialized(ASTContext &C, Decl::DeclID ID) {
+CXXConversionDecl *CXXConversionDecl::CreateDeserialized(ASTContext &C,
+                                                         GlobalDeclID ID) {
   return new (C, ID) CXXConversionDecl(
       C, nullptr, SourceLocation(), DeclarationNameInfo(), QualType(), nullptr,
       false, false, ExplicitSpecifier(), ConstexprSpecKind::Unspecified,
@@ -2924,7 +2926,7 @@ LinkageSpecDecl *LinkageSpecDecl::Create(ASTContext &C, DeclContext *DC,
 }
 
 LinkageSpecDecl *LinkageSpecDecl::CreateDeserialized(ASTContext &C,
-                                                     Decl::DeclID ID) {
+                                                     GlobalDeclID ID) {
   return new (C, ID)
       LinkageSpecDecl(nullptr, SourceLocation(), SourceLocation(),
                       LinkageSpecLanguageIDs::C, false);
@@ -2946,7 +2948,7 @@ UsingDirectiveDecl *UsingDirectiveDecl::Create(ASTContext &C, DeclContext *DC,
 }
 
 UsingDirectiveDecl *UsingDirectiveDecl::CreateDeserialized(ASTContext &C,
-                                                           Decl::DeclID ID) {
+                                                           GlobalDeclID ID) {
   return new (C, ID) UsingDirectiveDecl(nullptr, SourceLocation(),
                                         SourceLocation(),
                                         NestedNameSpecifierLoc(),
@@ -2985,7 +2987,8 @@ NamespaceDecl *NamespaceDecl::Create(ASTContext &C, DeclContext *DC,
       NamespaceDecl(C, DC, Inline, StartLoc, IdLoc, Id, PrevDecl, Nested);
 }
 
-NamespaceDecl *NamespaceDecl::CreateDeserialized(ASTContext &C, Decl::DeclID ID) {
+NamespaceDecl *NamespaceDecl::CreateDeserialized(ASTContext &C,
+                                                 GlobalDeclID ID) {
   return new (C, ID) NamespaceDecl(C, nullptr, false, SourceLocation(),
                                    SourceLocation(), nullptr, nullptr, false);
 }
@@ -3046,8 +3049,8 @@ NamespaceAliasDecl *NamespaceAliasDecl::Create(ASTContext &C, DeclContext *DC,
                                         QualifierLoc, IdentLoc, Namespace);
 }
 
-NamespaceAliasDecl *
-NamespaceAliasDecl::CreateDeserialized(ASTContext &C, Decl::DeclID ID) {
+NamespaceAliasDecl *NamespaceAliasDecl::CreateDeserialized(ASTContext &C,
+                                                           GlobalDeclID ID) {
   return new (C, ID) NamespaceAliasDecl(C, nullptr, SourceLocation(),
                                         SourceLocation(), nullptr,
                                         NestedNameSpecifierLoc(),
@@ -3102,8 +3105,8 @@ UsingShadowDecl::UsingShadowDecl(Kind K, ASTContext &C, EmptyShell Empty)
     : NamedDecl(K, nullptr, SourceLocation(), DeclarationName()),
       redeclarable_base(C) {}
 
-UsingShadowDecl *
-UsingShadowDecl::CreateDeserialized(ASTContext &C, Decl::DeclID ID) {
+UsingShadowDecl *UsingShadowDecl::CreateDeserialized(ASTContext &C,
+                                                     GlobalDeclID ID) {
   return new (C, ID) UsingShadowDecl(UsingShadow, C, EmptyShell());
 }
 
@@ -3126,7 +3129,7 @@ ConstructorUsingShadowDecl::Create(ASTContext &C, DeclContext *DC,
 }
 
 ConstructorUsingShadowDecl *
-ConstructorUsingShadowDecl::CreateDeserialized(ASTContext &C, Decl::DeclID ID) {
+ConstructorUsingShadowDecl::CreateDeserialized(ASTContext &C, GlobalDeclID ID) {
   return new (C, ID) ConstructorUsingShadowDecl(C, EmptyShell());
 }
 
@@ -3174,7 +3177,7 @@ UsingDecl *UsingDecl::Create(ASTContext &C, DeclContext *DC, SourceLocation UL,
   return new (C, DC) UsingDecl(DC, UL, QualifierLoc, NameInfo, HasTypename);
 }
 
-UsingDecl *UsingDecl::CreateDeserialized(ASTContext &C, Decl::DeclID ID) {
+UsingDecl *UsingDecl::CreateDeserialized(ASTContext &C, GlobalDeclID ID) {
   return new (C, ID) UsingDecl(nullptr, SourceLocation(),
                                NestedNameSpecifierLoc(), DeclarationNameInfo(),
                                false);
@@ -3198,7 +3201,8 @@ UsingEnumDecl *UsingEnumDecl::Create(ASTContext &C, DeclContext *DC,
       UsingEnumDecl(DC, EnumType->getType()->getAsTagDecl()->getDeclName(), UL, EL, NL, EnumType);
 }
 
-UsingEnumDecl *UsingEnumDecl::CreateDeserialized(ASTContext &C, Decl::DeclID ID) {
+UsingEnumDecl *UsingEnumDecl::CreateDeserialized(ASTContext &C,
+                                                 GlobalDeclID ID) {
   return new (C, ID)
       UsingEnumDecl(nullptr, DeclarationName(), SourceLocation(),
                     SourceLocation(), SourceLocation(), nullptr);
@@ -3217,7 +3221,7 @@ UsingPackDecl *UsingPackDecl::Create(ASTContext &C, DeclContext *DC,
   return new (C, DC, Extra) UsingPackDecl(DC, InstantiatedFrom, UsingDecls);
 }
 
-UsingPackDecl *UsingPackDecl::CreateDeserialized(ASTContext &C, Decl::DeclID ID,
+UsingPackDecl *UsingPackDecl::CreateDeserialized(ASTContext &C, GlobalDeclID ID,
                                                  unsigned NumExpansions) {
   size_t Extra = additionalSizeToAlloc<NamedDecl *>(NumExpansions);
   auto *Result =
@@ -3243,7 +3247,7 @@ UnresolvedUsingValueDecl::Create(ASTContext &C, DeclContext *DC,
 }
 
 UnresolvedUsingValueDecl *
-UnresolvedUsingValueDecl::CreateDeserialized(ASTContext &C, Decl::DeclID ID) {
+UnresolvedUsingValueDecl::CreateDeserialized(ASTContext &C, GlobalDeclID ID) {
   return new (C, ID) UnresolvedUsingValueDecl(nullptr, QualType(),
                                               SourceLocation(),
                                               NestedNameSpecifierLoc(),
@@ -3273,7 +3277,8 @@ UnresolvedUsingTypenameDecl::Create(ASTContext &C, DeclContext *DC,
 }
 
 UnresolvedUsingTypenameDecl *
-UnresolvedUsingTypenameDecl::CreateDeserialized(ASTContext &C, Decl::DeclID ID) {
+UnresolvedUsingTypenameDecl::CreateDeserialized(ASTContext &C,
+                                                GlobalDeclID ID) {
   return new (C, ID) UnresolvedUsingTypenameDecl(
       nullptr, SourceLocation(), SourceLocation(), NestedNameSpecifierLoc(),
       SourceLocation(), nullptr, SourceLocation());
@@ -3286,7 +3291,8 @@ UnresolvedUsingIfExistsDecl::Create(ASTContext &Ctx, DeclContext *DC,
 }
 
 UnresolvedUsingIfExistsDecl *
-UnresolvedUsingIfExistsDecl::CreateDeserialized(ASTContext &Ctx, Decl::DeclID ID) {
+UnresolvedUsingIfExistsDecl::CreateDeserialized(ASTContext &Ctx,
+                                                GlobalDeclID ID) {
   return new (Ctx, ID)
       UnresolvedUsingIfExistsDecl(nullptr, SourceLocation(), DeclarationName());
 }
@@ -3310,7 +3316,7 @@ StaticAssertDecl *StaticAssertDecl::Create(ASTContext &C, DeclContext *DC,
 }
 
 StaticAssertDecl *StaticAssertDecl::CreateDeserialized(ASTContext &C,
-                                                       Decl::DeclID ID) {
+                                                       GlobalDeclID ID) {
   return new (C, ID) StaticAssertDecl(nullptr, SourceLocation(), nullptr,
                                       nullptr, SourceLocation(), false);
 }
@@ -3332,7 +3338,7 @@ BindingDecl *BindingDecl::Create(ASTContext &C, DeclContext *DC,
   return new (C, DC) BindingDecl(DC, IdLoc, Id);
 }
 
-BindingDecl *BindingDecl::CreateDeserialized(ASTContext &C, Decl::DeclID ID) {
+BindingDecl *BindingDecl::CreateDeserialized(ASTContext &C, GlobalDeclID ID) {
   return new (C, ID) BindingDecl(nullptr, SourceLocation(), nullptr);
 }
 
@@ -3363,7 +3369,7 @@ DecompositionDecl *DecompositionDecl::Create(ASTContext &C, DeclContext *DC,
 }
 
 DecompositionDecl *DecompositionDecl::CreateDeserialized(ASTContext &C,
-                                                         Decl::DeclID ID,
+                                                         GlobalDeclID ID,
                                                          unsigned NumBindings) {
   size_t Extra = additionalSizeToAlloc<BindingDecl *>(NumBindings);
   auto *Result = new (C, ID, Extra)
@@ -3402,7 +3408,7 @@ MSPropertyDecl *MSPropertyDecl::Create(ASTContext &C, DeclContext *DC,
 }
 
 MSPropertyDecl *MSPropertyDecl::CreateDeserialized(ASTContext &C,
-                                                   Decl::DeclID ID) {
+                                                   GlobalDeclID ID) {
   return new (C, ID) MSPropertyDecl(nullptr, SourceLocation(),
                                     DeclarationName(), QualType(), nullptr,
                                     SourceLocation(), nullptr, nullptr);
@@ -3419,7 +3425,7 @@ MSGuidDecl *MSGuidDecl::Create(const ASTContext &C, QualType T, Parts P) {
   return new (C, DC) MSGuidDecl(DC, T, P);
 }
 
-MSGuidDecl *MSGuidDecl::CreateDeserialized(ASTContext &C, Decl::DeclID ID) {
+MSGuidDecl *MSGuidDecl::CreateDeserialized(ASTContext &C, GlobalDeclID ID) {
   return new (C, ID) MSGuidDecl(nullptr, QualType(), Parts());
 }
 
@@ -3529,7 +3535,7 @@ UnnamedGlobalConstantDecl::Create(const ASTContext &C, QualType T,
 }
 
 UnnamedGlobalConstantDecl *
-UnnamedGlobalConstantDecl::CreateDeserialized(ASTContext &C, Decl::DeclID ID) {
+UnnamedGlobalConstantDecl::CreateDeserialized(ASTContext &C, GlobalDeclID ID) {
   return new (C, ID)
       UnnamedGlobalConstantDecl(C, nullptr, QualType(), APValue());
 }
diff --git a/clang/lib/AST/DeclFriend.cpp b/clang/lib/AST/DeclFriend.cpp
index 1fabf8aa80c2..04b9b93699f3 100644
--- a/clang/lib/AST/DeclFriend.cpp
+++ b/clang/lib/AST/DeclFriend.cpp
@@ -62,7 +62,7 @@ FriendDecl *FriendDecl::Create(ASTContext &C, DeclContext *DC,
   return FD;
 }
 
-FriendDecl *FriendDecl::CreateDeserialized(ASTContext &C, Decl::DeclID ID,
+FriendDecl *FriendDecl::CreateDeserialized(ASTContext &C, GlobalDeclID ID,
                                            unsigned FriendTypeNumTPLists) {
   std::size_t Extra =
       additionalSizeToAlloc<TemplateParameterList *>(FriendTypeNumTPLists);
diff --git a/clang/lib/AST/DeclObjC.cpp b/clang/lib/AST/DeclObjC.cpp
index d4275eea0582..83062b0e6887 100644
--- a/clang/lib/AST/DeclObjC.cpp
+++ b/clang/lib/AST/DeclObjC.cpp
@@ -862,7 +862,8 @@ ObjCMethodDecl *ObjCMethodDecl::Create(
       isImplicitlyDeclared, isDefined, impControl, HasRelatedResultType);
 }
 
-ObjCMethodDecl *ObjCMethodDecl::CreateDeserialized(ASTContext &C, Decl::DeclID ID) {
+ObjCMethodDecl *ObjCMethodDecl::CreateDeserialized(ASTContext &C,
+                                                   GlobalDeclID ID) {
   return new (C, ID) ObjCMethodDecl(SourceLocation(), SourceLocation(),
                                     Selector(), QualType(), nullptr, nullptr);
 }
@@ -1486,7 +1487,7 @@ ObjCTypeParamDecl *ObjCTypeParamDecl::Create(ASTContext &ctx, DeclContext *dc,
 }
 
 ObjCTypeParamDecl *ObjCTypeParamDecl::CreateDeserialized(ASTContext &ctx,
-                                                         Decl::DeclID ID) {
+                                                         GlobalDeclID ID) {
   return new (ctx, ID) ObjCTypeParamDecl(ctx, nullptr,
                                          ObjCTypeParamVariance::Invariant,
                                          SourceLocation(), 0, SourceLocation(),
@@ -1551,7 +1552,7 @@ ObjCInterfaceDecl *ObjCInterfaceDecl::Create(
 }
 
 ObjCInterfaceDecl *ObjCInterfaceDecl::CreateDeserialized(const ASTContext &C,
-                                                         Decl::DeclID ID) {
+                                                         GlobalDeclID ID) {
   auto *Result = new (C, ID)
       ObjCInterfaceDecl(C, nullptr, SourceLocation(), nullptr, nullptr,
                         SourceLocation(), nullptr, false);
@@ -1865,7 +1866,7 @@ ObjCIvarDecl *ObjCIvarDecl::Create(ASTContext &C, ObjCContainerDecl *DC,
                                   synthesized);
 }
 
-ObjCIvarDecl *ObjCIvarDecl::CreateDeserialized(ASTContext &C, Decl::DeclID ID) {
+ObjCIvarDecl *ObjCIvarDecl::CreateDeserialized(ASTContext &C, GlobalDeclID ID) {
   return new (C, ID) ObjCIvarDecl(nullptr, SourceLocation(), SourceLocation(),
                                   nullptr, QualType(), nullptr,
                                   ObjCIvarDecl::None, nullptr, false);
@@ -1914,7 +1915,7 @@ ObjCAtDefsFieldDecl
 }
 
 ObjCAtDefsFieldDecl *ObjCAtDefsFieldDecl::CreateDeserialized(ASTContext &C,
-                                                             Decl::DeclID ID) {
+                                                             GlobalDeclID ID) {
   return new (C, ID) ObjCAtDefsFieldDecl(nullptr, SourceLocation(),
                                          SourceLocation(), nullptr, QualType(),
                                          nullptr);
@@ -1949,7 +1950,7 @@ ObjCProtocolDecl *ObjCProtocolDecl::Create(ASTContext &C, DeclContext *DC,
 }
 
 ObjCProtocolDecl *ObjCProtocolDecl::CreateDeserialized(ASTContext &C,
-                                                       Decl::DeclID ID) {
+                                                       GlobalDeclID ID) {
   ObjCProtocolDecl *Result =
       new (C, ID) ObjCProtocolDecl(C, nullptr, nullptr, SourceLocation(),
                                    SourceLocation(), nullptr);
@@ -2148,7 +2149,7 @@ ObjCCategoryDecl *ObjCCategoryDecl::Create(
 }
 
 ObjCCategoryDecl *ObjCCategoryDecl::CreateDeserialized(ASTContext &C,
-                                                       Decl::DeclID ID) {
+                                                       GlobalDeclID ID) {
   return new (C, ID) ObjCCategoryDecl(nullptr, SourceLocation(),
                                       SourceLocation(), SourceLocation(),
                                       nullptr, nullptr, nullptr);
@@ -2188,8 +2189,8 @@ ObjCCategoryImplDecl *ObjCCategoryImplDecl::Create(
                                           atStartLoc, CategoryNameLoc);
 }
 
-ObjCCategoryImplDecl *ObjCCategoryImplDecl::CreateDeserialized(ASTContext &C,
-                                                               Decl::DeclID ID) {
+ObjCCategoryImplDecl *
+ObjCCategoryImplDecl::CreateDeserialized(ASTContext &C, GlobalDeclID ID) {
   return new (C, ID) ObjCCategoryImplDecl(nullptr, nullptr, nullptr,
                                           SourceLocation(), SourceLocation(),
                                           SourceLocation());
@@ -2296,7 +2297,7 @@ ObjCImplementationDecl::Create(ASTContext &C, DeclContext *DC,
 }
 
 ObjCImplementationDecl *
-ObjCImplementationDecl::CreateDeserialized(ASTContext &C, Decl::DeclID ID) {
+ObjCImplementationDecl::CreateDeserialized(ASTContext &C, GlobalDeclID ID) {
   return new (C, ID) ObjCImplementationDecl(nullptr, nullptr, nullptr,
                                             SourceLocation(), SourceLocation());
 }
@@ -2339,7 +2340,7 @@ ObjCCompatibleAliasDecl::Create(ASTContext &C, DeclContext *DC,
 }
 
 ObjCCompatibleAliasDecl *
-ObjCCompatibleAliasDecl::CreateDeserialized(ASTContext &C, Decl::DeclID ID) {
+ObjCCompatibleAliasDecl::CreateDeserialized(ASTContext &C, GlobalDeclID ID) {
   return new (C, ID) ObjCCompatibleAliasDecl(nullptr, SourceLocation(),
                                              nullptr, nullptr);
 }
@@ -2360,7 +2361,7 @@ ObjCPropertyDecl::Create(ASTContext &C, DeclContext *DC, SourceLocation L,
 }
 
 ObjCPropertyDecl *ObjCPropertyDecl::CreateDeserialized(ASTContext &C,
-                                                       Decl::DeclID ID) {
+                                                       GlobalDeclID ID) {
   return new (C, ID) ObjCPropertyDecl(nullptr, SourceLocation(), nullptr,
                                       SourceLocation(), SourceLocation(),
                                       QualType(), nullptr, None);
@@ -2392,8 +2393,8 @@ ObjCPropertyImplDecl *ObjCPropertyImplDecl::Create(ASTContext &C,
                                           ivarLoc);
 }
 
-ObjCPropertyImplDecl *ObjCPropertyImplDecl::CreateDeserialized(ASTContext &C,
-                                                               Decl::DeclID ID) {
+ObjCPropertyImplDecl *
+ObjCPropertyImplDecl::CreateDeserialized(ASTContext &C, GlobalDeclID ID) {
   return new (C, ID) ObjCPropertyImplDecl(nullptr, SourceLocation(),
                                           SourceLocation(), nullptr, Dynamic,
                                           nullptr, SourceLocation());
diff --git a/clang/lib/AST/DeclOpenMP.cpp b/clang/lib/AST/DeclOpenMP.cpp
index b178a15aab5f..81ca48e60942 100644
--- a/clang/lib/AST/DeclOpenMP.cpp
+++ b/clang/lib/AST/DeclOpenMP.cpp
@@ -36,7 +36,7 @@ OMPThreadPrivateDecl *OMPThreadPrivateDecl::Create(ASTContext &C,
 }
 
 OMPThreadPrivateDecl *OMPThreadPrivateDecl::CreateDeserialized(ASTContext &C,
-                                                               Decl::DeclID ID,
+                                                               GlobalDeclID ID,
                                                                unsigned N) {
   return OMPDeclarativeDirective::createEmptyDirective<OMPThreadPrivateDecl>(
       C, ID, 0, N);
@@ -63,7 +63,8 @@ OMPAllocateDecl *OMPAllocateDecl::Create(ASTContext &C, DeclContext *DC,
   return D;
 }
 
-OMPAllocateDecl *OMPAllocateDecl::CreateDeserialized(ASTContext &C, Decl::DeclID ID,
+OMPAllocateDecl *OMPAllocateDecl::CreateDeserialized(ASTContext &C,
+                                                     GlobalDeclID ID,
                                                      unsigned NVars,
                                                      unsigned NClauses) {
   return OMPDeclarativeDirective::createEmptyDirective<OMPAllocateDecl>(
@@ -89,7 +90,8 @@ OMPRequiresDecl *OMPRequiresDecl::Create(ASTContext &C, DeclContext *DC,
                                                                    L);
 }
 
-OMPRequiresDecl *OMPRequiresDecl::CreateDeserialized(ASTContext &C, Decl::DeclID ID,
+OMPRequiresDecl *OMPRequiresDecl::CreateDeserialized(ASTContext &C,
+                                                     GlobalDeclID ID,
                                                      unsigned N) {
   return OMPDeclarativeDirective::createEmptyDirective<OMPRequiresDecl>(
       C, ID, N, 0, SourceLocation());
@@ -117,7 +119,7 @@ OMPDeclareReductionDecl *OMPDeclareReductionDecl::Create(
 }
 
 OMPDeclareReductionDecl *
-OMPDeclareReductionDecl::CreateDeserialized(ASTContext &C, Decl::DeclID ID) {
+OMPDeclareReductionDecl::CreateDeserialized(ASTContext &C, GlobalDeclID ID) {
   return new (C, ID) OMPDeclareReductionDecl(
       OMPDeclareReduction, /*DC=*/nullptr, SourceLocation(), DeclarationName(),
       QualType(), /*PrevDeclInScope=*/nullptr);
@@ -148,7 +150,7 @@ OMPDeclareMapperDecl *OMPDeclareMapperDecl::Create(
 }
 
 OMPDeclareMapperDecl *OMPDeclareMapperDecl::CreateDeserialized(ASTContext &C,
-                                                               Decl::DeclID ID,
+                                                               GlobalDeclID ID,
                                                                unsigned N) {
   return OMPDeclarativeDirective::createEmptyDirective<OMPDeclareMapperDecl>(
       C, ID, N, 1, SourceLocation(), DeclarationName(), QualType(),
@@ -179,7 +181,7 @@ OMPCapturedExprDecl *OMPCapturedExprDecl::Create(ASTContext &C, DeclContext *DC,
 }
 
 OMPCapturedExprDecl *OMPCapturedExprDecl::CreateDeserialized(ASTContext &C,
-                                                             Decl::DeclID ID) {
+                                                             GlobalDeclID ID) {
   return new (C, ID) OMPCapturedExprDecl(C, nullptr, nullptr, QualType(),
                                          /*TInfo=*/nullptr, SourceLocation());
 }
diff --git a/clang/lib/AST/DeclTemplate.cpp b/clang/lib/AST/DeclTemplate.cpp
index 0ba271c3e04e..d27a30e0c5fc 100644
--- a/clang/lib/AST/DeclTemplate.cpp
+++ b/clang/lib/AST/DeclTemplate.cpp
@@ -337,9 +337,10 @@ void RedeclarableTemplateDecl::loadLazySpecializationsImpl() const {
   CommonBase *CommonBasePtr = getMostRecentDecl()->getCommonPtr();
   if (CommonBasePtr->LazySpecializations) {
     ASTContext &Context = getASTContext();
-    uint32_t *Specs = CommonBasePtr->LazySpecializations;
+    GlobalDeclID *Specs = CommonBasePtr->LazySpecializations;
     CommonBasePtr->LazySpecializations = nullptr;
-    for (uint32_t I = 0, N = *Specs++; I != N; ++I)
+    unsigned SpecSize = (*Specs++).get();
+    for (unsigned I = 0; I != SpecSize; ++I)
       (void)Context.getExternalSource()->GetExternalDecl(Specs[I]);
   }
 }
@@ -417,8 +418,8 @@ FunctionTemplateDecl::Create(ASTContext &C, DeclContext *DC, SourceLocation L,
   return TD;
 }
 
-FunctionTemplateDecl *FunctionTemplateDecl::CreateDeserialized(ASTContext &C,
-                                                               Decl::DeclID ID) {
+FunctionTemplateDecl *
+FunctionTemplateDecl::CreateDeserialized(ASTContext &C, GlobalDeclID ID) {
   return new (C, ID) FunctionTemplateDecl(C, nullptr, SourceLocation(),
                                           DeclarationName(), nullptr, nullptr);
 }
@@ -503,7 +504,7 @@ ClassTemplateDecl *ClassTemplateDecl::Create(ASTContext &C, DeclContext *DC,
 }
 
 ClassTemplateDecl *ClassTemplateDecl::CreateDeserialized(ASTContext &C,
-                                                         Decl::DeclID ID) {
+                                                         GlobalDeclID ID) {
   return new (C, ID) ClassTemplateDecl(C, nullptr, SourceLocation(),
                                        DeclarationName(), nullptr, nullptr);
 }
@@ -652,14 +653,14 @@ TemplateTypeParmDecl *TemplateTypeParmDecl::Create(
 }
 
 TemplateTypeParmDecl *
-TemplateTypeParmDecl::CreateDeserialized(const ASTContext &C, Decl::DeclID ID) {
+TemplateTypeParmDecl::CreateDeserialized(const ASTContext &C, GlobalDeclID ID) {
   return new (C, ID)
       TemplateTypeParmDecl(nullptr, SourceLocation(), SourceLocation(), nullptr,
                            false, false, std::nullopt);
 }
 
 TemplateTypeParmDecl *
-TemplateTypeParmDecl::CreateDeserialized(const ASTContext &C, Decl::DeclID ID,
+TemplateTypeParmDecl::CreateDeserialized(const ASTContext &C, GlobalDeclID ID,
                                          bool HasTypeConstraint) {
   return new (C, ID,
               additionalSizeToAlloc<TypeConstraint>(HasTypeConstraint ? 1 : 0))
@@ -759,7 +760,7 @@ NonTypeTemplateParmDecl *NonTypeTemplateParmDecl::Create(
 }
 
 NonTypeTemplateParmDecl *
-NonTypeTemplateParmDecl::CreateDeserialized(ASTContext &C, Decl::DeclID ID,
+NonTypeTemplateParmDecl::CreateDeserialized(ASTContext &C, GlobalDeclID ID,
                                             bool HasTypeConstraint) {
   return new (C, ID, additionalSizeToAlloc<std::pair<QualType,
                                                      TypeSourceInfo *>,
@@ -770,7 +771,7 @@ NonTypeTemplateParmDecl::CreateDeserialized(ASTContext &C, Decl::DeclID ID,
 }
 
 NonTypeTemplateParmDecl *
-NonTypeTemplateParmDecl::CreateDeserialized(ASTContext &C, Decl::DeclID ID,
+NonTypeTemplateParmDecl::CreateDeserialized(ASTContext &C, GlobalDeclID ID,
                                             unsigned NumExpandedTypes,
                                             bool HasTypeConstraint) {
   auto *NTTP =
@@ -836,13 +837,13 @@ TemplateTemplateParmDecl::Create(const ASTContext &C, DeclContext *DC,
 }
 
 TemplateTemplateParmDecl *
-TemplateTemplateParmDecl::CreateDeserialized(ASTContext &C, Decl::DeclID ID) {
+TemplateTemplateParmDecl::CreateDeserialized(ASTContext &C, GlobalDeclID ID) {
   return new (C, ID) TemplateTemplateParmDecl(nullptr, SourceLocation(), 0, 0,
                                               false, nullptr, false, nullptr);
 }
 
 TemplateTemplateParmDecl *
-TemplateTemplateParmDecl::CreateDeserialized(ASTContext &C, Decl::DeclID ID,
+TemplateTemplateParmDecl::CreateDeserialized(ASTContext &C, GlobalDeclID ID,
                                              unsigned NumExpansions) {
   auto *TTP =
       new (C, ID, additionalSizeToAlloc<TemplateParameterList *>(NumExpansions))
@@ -949,7 +950,7 @@ ClassTemplateSpecializationDecl::Create(ASTContext &Context, TagKind TK,
 
 ClassTemplateSpecializationDecl *
 ClassTemplateSpecializationDecl::CreateDeserialized(ASTContext &C,
-                                                    Decl::DeclID ID) {
+                                                    GlobalDeclID ID) {
   auto *Result =
     new (C, ID) ClassTemplateSpecializationDecl(C, ClassTemplateSpecialization);
   Result->setMayHaveOutOfDateDef(false);
@@ -1035,8 +1036,7 @@ ConceptDecl *ConceptDecl::Create(ASTContext &C, DeclContext *DC,
   return TD;
 }
 
-ConceptDecl *ConceptDecl::CreateDeserialized(ASTContext &C,
-                                             Decl::DeclID ID) {
+ConceptDecl *ConceptDecl::CreateDeserialized(ASTContext &C, GlobalDeclID ID) {
   ConceptDecl *Result = new (C, ID) ConceptDecl(nullptr, SourceLocation(),
                                                 DeclarationName(),
                                                 nullptr, nullptr);
@@ -1070,7 +1070,7 @@ ImplicitConceptSpecializationDecl *ImplicitConceptSpecializationDecl::Create(
 
 ImplicitConceptSpecializationDecl *
 ImplicitConceptSpecializationDecl::CreateDeserialized(
-    const ASTContext &C, Decl::DeclID ID, unsigned NumTemplateArgs) {
+    const ASTContext &C, GlobalDeclID ID, unsigned NumTemplateArgs) {
   return new (C, ID, additionalSizeToAlloc<TemplateArgument>(NumTemplateArgs))
       ImplicitConceptSpecializationDecl(EmptyShell{}, NumTemplateArgs);
 }
@@ -1133,7 +1133,7 @@ Create(ASTContext &Context, TagKind TK,DeclContext *DC,
 
 ClassTemplatePartialSpecializationDecl *
 ClassTemplatePartialSpecializationDecl::CreateDeserialized(ASTContext &C,
-                                                           Decl::DeclID ID) {
+                                                           GlobalDeclID ID) {
   auto *Result = new (C, ID) ClassTemplatePartialSpecializationDecl(C);
   Result->setMayHaveOutOfDateDef(false);
   return Result;
@@ -1160,7 +1160,7 @@ FriendTemplateDecl::Create(ASTContext &Context, DeclContext *DC,
 }
 
 FriendTemplateDecl *FriendTemplateDecl::CreateDeserialized(ASTContext &C,
-                                                           Decl::DeclID ID) {
+                                                           GlobalDeclID ID) {
   return new (C, ID) FriendTemplateDecl(EmptyShell());
 }
 
@@ -1179,8 +1179,8 @@ TypeAliasTemplateDecl::Create(ASTContext &C, DeclContext *DC, SourceLocation L,
   return TD;
 }
 
-TypeAliasTemplateDecl *TypeAliasTemplateDecl::CreateDeserialized(ASTContext &C,
-                                                                 Decl::DeclID ID) {
+TypeAliasTemplateDecl *
+TypeAliasTemplateDecl::CreateDeserialized(ASTContext &C, GlobalDeclID ID) {
   return new (C, ID) TypeAliasTemplateDecl(C, nullptr, SourceLocation(),
                                            DeclarationName(), nullptr, nullptr);
 }
@@ -1218,7 +1218,7 @@ VarTemplateDecl *VarTemplateDecl::Create(ASTContext &C, DeclContext *DC,
 }
 
 VarTemplateDecl *VarTemplateDecl::CreateDeserialized(ASTContext &C,
-                                                     Decl::DeclID ID) {
+                                                     GlobalDeclID ID) {
   return new (C, ID) VarTemplateDecl(C, nullptr, SourceLocation(),
                                      DeclarationName(), nullptr, nullptr);
 }
@@ -1340,7 +1340,8 @@ VarTemplateSpecializationDecl *VarTemplateSpecializationDecl::Create(
 }
 
 VarTemplateSpecializationDecl *
-VarTemplateSpecializationDecl::CreateDeserialized(ASTContext &C, Decl::DeclID ID) {
+VarTemplateSpecializationDecl::CreateDeserialized(ASTContext &C,
+                                                  GlobalDeclID ID) {
   return new (C, ID)
       VarTemplateSpecializationDecl(VarTemplateSpecialization, C);
 }
@@ -1432,7 +1433,7 @@ VarTemplatePartialSpecializationDecl::Create(
 
 VarTemplatePartialSpecializationDecl *
 VarTemplatePartialSpecializationDecl::CreateDeserialized(ASTContext &C,
-                                                         Decl::DeclID ID) {
+                                                         GlobalDeclID ID) {
   return new (C, ID) VarTemplatePartialSpecializationDecl(C);
 }
 
@@ -1546,7 +1547,7 @@ TemplateParamObjectDecl *TemplateParamObjectDecl::Create(const ASTContext &C,
 }
 
 TemplateParamObjectDecl *
-TemplateParamObjectDecl::CreateDeserialized(ASTContext &C, Decl::DeclID ID) {
+TemplateParamObjectDecl::CreateDeserialized(ASTContext &C, GlobalDeclID ID) {
   auto *TPOD = new (C, ID) TemplateParamObjectDecl(nullptr, QualType(), APValue());
   C.addDestruction(&TPOD->Value);
   return TPOD;
diff --git a/clang/lib/AST/Expr.cpp b/clang/lib/AST/Expr.cpp
index 9eec7edc9d1a..63dcdb919c71 100644
--- a/clang/lib/AST/Expr.cpp
+++ b/clang/lib/AST/Expr.cpp
@@ -3680,7 +3680,7 @@ bool Expr::HasSideEffects(const ASTContext &Ctx,
   case ParenExprClass:
   case ArraySubscriptExprClass:
   case MatrixSubscriptExprClass:
-  case OMPArraySectionExprClass:
+  case ArraySectionExprClass:
   case OMPArrayShapingExprClass:
   case OMPIteratorExprClass:
   case MemberExprClass:
@@ -5060,9 +5060,9 @@ QualType AtomicExpr::getValueType() const {
   return T;
 }
 
-QualType OMPArraySectionExpr::getBaseOriginalType(const Expr *Base) {
+QualType ArraySectionExpr::getBaseOriginalType(const Expr *Base) {
   unsigned ArraySectionCount = 0;
-  while (auto *OASE = dyn_cast<OMPArraySectionExpr>(Base->IgnoreParens())) {
+  while (auto *OASE = dyn_cast<ArraySectionExpr>(Base->IgnoreParens())) {
     Base = OASE->getBase();
     ++ArraySectionCount;
   }
diff --git a/clang/lib/AST/ExprClassification.cpp b/clang/lib/AST/ExprClassification.cpp
index 7026fca8554c..2bb8f9aeedc7 100644
--- a/clang/lib/AST/ExprClassification.cpp
+++ b/clang/lib/AST/ExprClassification.cpp
@@ -145,7 +145,7 @@ static Cl::Kinds ClassifyInternal(ASTContext &Ctx, const Expr *E) {
   case Expr::FunctionParmPackExprClass:
   case Expr::MSPropertyRefExprClass:
   case Expr::MSPropertySubscriptExprClass:
-  case Expr::OMPArraySectionExprClass:
+  case Expr::ArraySectionExprClass:
   case Expr::OMPArrayShapingExprClass:
   case Expr::OMPIteratorExprClass:
     return Cl::CL_LValue;
diff --git a/clang/lib/AST/ExprConstant.cpp b/clang/lib/AST/ExprConstant.cpp
index de3c2a63913e..f1aa19e4409e 100644
--- a/clang/lib/AST/ExprConstant.cpp
+++ b/clang/lib/AST/ExprConstant.cpp
@@ -2706,7 +2706,11 @@ static bool checkFloatingPointResult(EvalInfo &Info, const Expr *E,
 static bool HandleFloatToFloatCast(EvalInfo &Info, const Expr *E,
                                    QualType SrcType, QualType DestType,
                                    APFloat &Result) {
-  assert(isa<CastExpr>(E) || isa<CompoundAssignOperator>(E));
+  assert((isa<CastExpr>(E) || isa<CompoundAssignOperator>(E) ||
+          isa<ConvertVectorExpr>(E)) &&
+         "HandleFloatToFloatCast has been checked with only CastExpr, "
+         "CompoundAssignOperator and ConvertVectorExpr. Please either validate "
+         "the new expression or address the root cause of this usage.");
   llvm::RoundingMode RM = getActiveRoundingMode(Info, E);
   APFloat::opStatus St;
   APFloat Value = Result;
@@ -9237,9 +9241,10 @@ bool PointerExprEvaluator::VisitCastExpr(const CastExpr *E) {
       bool HasValidResult = !Result.InvalidBase && !Result.Designator.Invalid &&
                             !Result.IsNullPtr;
       bool VoidPtrCastMaybeOK =
-          HasValidResult &&
-          Info.Ctx.hasSameUnqualifiedType(Result.Designator.getType(Info.Ctx),
-                                          E->getType()->getPointeeType());
+          Result.IsNullPtr ||
+          (HasValidResult &&
+           Info.Ctx.hasSimilarType(Result.Designator.getType(Info.Ctx),
+                                   E->getType()->getPointeeType()));
       // 1. We'll allow it in std::allocator::allocate, and anything which that
       //    calls.
       // 2. HACK 2022-03-28: Work around an issue with libstdc++'s
@@ -10709,8 +10714,11 @@ namespace {
     bool VisitUnaryImag(const UnaryOperator *E);
     bool VisitBinaryOperator(const BinaryOperator *E);
     bool VisitUnaryOperator(const UnaryOperator *E);
+    bool VisitConvertVectorExpr(const ConvertVectorExpr *E);
+    bool VisitShuffleVectorExpr(const ShuffleVectorExpr *E);
+
     // FIXME: Missing: conditional operator (for GNU
-    //                 conditional select), shufflevector, ExtVectorElementExpr
+    //                 conditional select), ExtVectorElementExpr
   };
 } // end anonymous namespace
 
@@ -10961,6 +10969,122 @@ bool VectorExprEvaluator::VisitUnaryOperator(const UnaryOperator *E) {
   return Success(APValue(ResultElements.data(), ResultElements.size()), E);
 }
 
+static bool handleVectorElementCast(EvalInfo &Info, const FPOptions FPO,
+                                    const Expr *E, QualType SourceTy,
+                                    QualType DestTy, APValue const &Original,
+                                    APValue &Result) {
+  if (SourceTy->isIntegerType()) {
+    if (DestTy->isRealFloatingType()) {
+      Result = APValue(APFloat(0.0));
+      return HandleIntToFloatCast(Info, E, FPO, SourceTy, Original.getInt(),
+                                  DestTy, Result.getFloat());
+    }
+    if (DestTy->isIntegerType()) {
+      Result = APValue(
+          HandleIntToIntCast(Info, E, DestTy, SourceTy, Original.getInt()));
+      return true;
+    }
+  } else if (SourceTy->isRealFloatingType()) {
+    if (DestTy->isRealFloatingType()) {
+      Result = Original;
+      return HandleFloatToFloatCast(Info, E, SourceTy, DestTy,
+                                    Result.getFloat());
+    }
+    if (DestTy->isIntegerType()) {
+      Result = APValue(APSInt());
+      return HandleFloatToIntCast(Info, E, SourceTy, Original.getFloat(),
+                                  DestTy, Result.getInt());
+    }
+  }
+
+  Info.FFDiag(E, diag::err_convertvector_constexpr_unsupported_vector_cast)
+      << SourceTy << DestTy;
+  return false;
+}
+
+bool VectorExprEvaluator::VisitConvertVectorExpr(const ConvertVectorExpr *E) {
+  APValue Source;
+  QualType SourceVecType = E->getSrcExpr()->getType();
+  if (!EvaluateAsRValue(Info, E->getSrcExpr(), Source))
+    return false;
+
+  QualType DestTy = E->getType()->castAs<VectorType>()->getElementType();
+  QualType SourceTy = SourceVecType->castAs<VectorType>()->getElementType();
+
+  const FPOptions FPO = E->getFPFeaturesInEffect(Info.Ctx.getLangOpts());
+
+  auto SourceLen = Source.getVectorLength();
+  SmallVector<APValue, 4> ResultElements;
+  ResultElements.reserve(SourceLen);
+  for (unsigned EltNum = 0; EltNum < SourceLen; ++EltNum) {
+    APValue Elt;
+    if (!handleVectorElementCast(Info, FPO, E, SourceTy, DestTy,
+                                 Source.getVectorElt(EltNum), Elt))
+      return false;
+    ResultElements.push_back(std::move(Elt));
+  }
+
+  return Success(APValue(ResultElements.data(), ResultElements.size()), E);
+}
+
+static bool handleVectorShuffle(EvalInfo &Info, const ShuffleVectorExpr *E,
+                                QualType ElemType, APValue const &VecVal1,
+                                APValue const &VecVal2, unsigned EltNum,
+                                APValue &Result) {
+  unsigned const TotalElementsInInputVector1 = VecVal1.getVectorLength();
+  unsigned const TotalElementsInInputVector2 = VecVal2.getVectorLength();
+
+  APSInt IndexVal = E->getShuffleMaskIdx(Info.Ctx, EltNum);
+  int64_t index = IndexVal.getExtValue();
+  // The spec says that -1 should be treated as undef for optimizations,
+  // but in constexpr we'd have to produce an APValue::Indeterminate,
+  // which is prohibited from being a top-level constant value. Emit a
+  // diagnostic instead.
+  if (index == -1) {
+    Info.FFDiag(
+        E, diag::err_shufflevector_minus_one_is_undefined_behavior_constexpr)
+        << EltNum;
+    return false;
+  }
+
+  if (index < 0 ||
+      index >= TotalElementsInInputVector1 + TotalElementsInInputVector2)
+    llvm_unreachable("Out of bounds shuffle index");
+
+  if (index >= TotalElementsInInputVector1)
+    Result = VecVal2.getVectorElt(index - TotalElementsInInputVector1);
+  else
+    Result = VecVal1.getVectorElt(index);
+  return true;
+}
+
+bool VectorExprEvaluator::VisitShuffleVectorExpr(const ShuffleVectorExpr *E) {
+  APValue VecVal1;
+  const Expr *Vec1 = E->getExpr(0);
+  if (!EvaluateAsRValue(Info, Vec1, VecVal1))
+    return false;
+  APValue VecVal2;
+  const Expr *Vec2 = E->getExpr(1);
+  if (!EvaluateAsRValue(Info, Vec2, VecVal2))
+    return false;
+
+  VectorType const *DestVecTy = E->getType()->castAs<VectorType>();
+  QualType DestElTy = DestVecTy->getElementType();
+
+  auto TotalElementsInOutputVector = DestVecTy->getNumElements();
+
+  SmallVector<APValue, 4> ResultElements;
+  ResultElements.reserve(TotalElementsInOutputVector);
+  for (unsigned EltNum = 0; EltNum < TotalElementsInOutputVector; ++EltNum) {
+    APValue Elt;
+    if (!handleVectorShuffle(Info, E, DestElTy, VecVal1, VecVal2, EltNum, Elt))
+      return false;
+    ResultElements.push_back(std::move(Elt));
+  }
+
+  return Success(APValue(ResultElements.data(), ResultElements.size()), E);
+}
+
 //===----------------------------------------------------------------------===//
 // Array Evaluation
 //===----------------------------------------------------------------------===//
@@ -16130,7 +16254,7 @@ static ICEDiag CheckICE(const Expr* E, const ASTContext &Ctx) {
   case Expr::StringLiteralClass:
   case Expr::ArraySubscriptExprClass:
   case Expr::MatrixSubscriptExprClass:
-  case Expr::OMPArraySectionExprClass:
+  case Expr::ArraySectionExprClass:
   case Expr::OMPArrayShapingExprClass:
   case Expr::OMPIteratorExprClass:
   case Expr::MemberExprClass:
diff --git a/clang/lib/AST/ExternalASTSource.cpp b/clang/lib/AST/ExternalASTSource.cpp
index 090ef02aa422..e96a47496851 100644
--- a/clang/lib/AST/ExternalASTSource.cpp
+++ b/clang/lib/AST/ExternalASTSource.cpp
@@ -68,9 +68,7 @@ bool ExternalASTSource::layoutRecordType(
   return false;
 }
 
-Decl *ExternalASTSource::GetExternalDecl(uint32_t ID) {
-  return nullptr;
-}
+Decl *ExternalASTSource::GetExternalDecl(GlobalDeclID ID) { return nullptr; }
 
 Selector ExternalASTSource::GetExternalSelector(uint32_t ID) {
   return Selector();
diff --git a/clang/lib/AST/Interp/ByteCodeExprGen.cpp b/clang/lib/AST/Interp/ByteCodeExprGen.cpp
index 8cd0c198d9a8..17f95e7f3cac 100644
--- a/clang/lib/AST/Interp/ByteCodeExprGen.cpp
+++ b/clang/lib/AST/Interp/ByteCodeExprGen.cpp
@@ -110,18 +110,37 @@ bool ByteCodeExprGen<Emitter>::VisitCastExpr(const CastExpr *CE) {
     if (!this->visit(SubExpr))
       return false;
 
-    unsigned DerivedOffset = collectBaseOffset(getRecordTy(CE->getType()),
-                                               getRecordTy(SubExpr->getType()));
+    const auto extractRecordDecl = [](QualType Ty) -> const CXXRecordDecl * {
+      if (const auto *PT = dyn_cast<PointerType>(Ty))
+        return PT->getPointeeType()->getAsCXXRecordDecl();
+      return Ty->getAsCXXRecordDecl();
+    };
+
+    // FIXME: We can express a series of non-virtual casts as a single
+    // GetPtrBasePop op.
+    QualType CurType = SubExpr->getType();
+    for (const CXXBaseSpecifier *B : CE->path()) {
+      if (B->isVirtual()) {
+        if (!this->emitGetPtrVirtBasePop(extractRecordDecl(B->getType()), CE))
+          return false;
+        CurType = B->getType();
+      } else {
+        unsigned DerivedOffset = collectBaseOffset(B->getType(), CurType);
+        if (!this->emitGetPtrBasePop(DerivedOffset, CE))
+          return false;
+        CurType = B->getType();
+      }
+    }
 
-    return this->emitGetPtrBasePop(DerivedOffset, CE);
+    return true;
   }
 
   case CK_BaseToDerived: {
     if (!this->visit(SubExpr))
       return false;
 
-    unsigned DerivedOffset = collectBaseOffset(getRecordTy(SubExpr->getType()),
-                                               getRecordTy(CE->getType()));
+    unsigned DerivedOffset =
+        collectBaseOffset(SubExpr->getType(), CE->getType());
 
     return this->emitGetPtrDerivedPop(DerivedOffset, CE);
   }
@@ -193,6 +212,13 @@ bool ByteCodeExprGen<Emitter>::VisitCastExpr(const CastExpr *CE) {
     if (!this->visit(SubExpr))
       return false;
 
+    // If SubExpr doesn't result in a pointer, make it one.
+    if (PrimType FromT = classifyPrim(SubExpr->getType()); FromT != PT_Ptr) {
+      assert(isPtrType(FromT));
+      if (!this->emitDecayPtr(FromT, PT_Ptr, CE))
+        return false;
+    }
+
     PrimType T = classifyPrim(CE->getType());
     if (T == PT_IntAP)
       return this->emitCastPointerIntegralAP(Ctx.getBitWidth(CE->getType()),
@@ -905,8 +931,31 @@ bool ByteCodeExprGen<Emitter>::VisitImplicitValueInitExpr(const ImplicitValueIni
   if (std::optional<PrimType> T = classify(QT))
     return this->visitZeroInitializer(*T, QT, E);
 
-  if (QT->isRecordType())
-    return false;
+  if (QT->isRecordType()) {
+    const RecordDecl *RD = QT->getAsRecordDecl();
+    assert(RD);
+    if (RD->isInvalidDecl())
+      return false;
+    if (RD->isUnion()) {
+      // C++11 [dcl.init]p5: If T is a (possibly cv-qualified) union type, the
+      // object's first non-static named data member is zero-initialized
+      // FIXME
+      return false;
+    }
+
+    if (const auto *CXXRD = dyn_cast<CXXRecordDecl>(RD);
+        CXXRD && CXXRD->getNumVBases() > 0) {
+      // TODO: Diagnose.
+      return false;
+    }
+
+    const Record *R = getRecord(QT);
+    if (!R)
+      return false;
+
+    assert(Initializing);
+    return this->visitZeroRecordInitializer(R, E);
+  }
 
   if (QT->isIncompleteArrayType())
     return true;
@@ -981,122 +1030,98 @@ bool ByteCodeExprGen<Emitter>::VisitArraySubscriptExpr(
 
 template <class Emitter>
 bool ByteCodeExprGen<Emitter>::visitInitList(ArrayRef<const Expr *> Inits,
+                                             const Expr *ArrayFiller,
                                              const Expr *E) {
-  assert(E->getType()->isRecordType());
-  const Record *R = getRecord(E->getType());
+  if (E->getType()->isVoidType())
+    return this->emitInvalid(E);
 
-  if (Inits.size() == 1 && E->getType() == Inits[0]->getType()) {
-    return this->visitInitializer(Inits[0]);
+  // Handle discarding first.
+  if (DiscardResult) {
+    for (const Expr *Init : Inits) {
+      if (!this->discard(Init))
+        return false;
+    }
+    return true;
   }
 
-  unsigned InitIndex = 0;
-  for (const Expr *Init : Inits) {
-    // Skip unnamed bitfields.
-    while (InitIndex < R->getNumFields() &&
-           R->getField(InitIndex)->Decl->isUnnamedBitField())
-      ++InitIndex;
+  // Primitive values.
+  if (std::optional<PrimType> T = classify(E->getType())) {
+    assert(!DiscardResult);
+    if (Inits.size() == 0)
+      return this->visitZeroInitializer(*T, E->getType(), E);
+    assert(Inits.size() == 1);
+    return this->delegate(Inits[0]);
+  }
 
-    if (!this->emitDupPtr(E))
-      return false;
+  QualType T = E->getType();
+  if (T->isRecordType()) {
+    const Record *R = getRecord(E->getType());
 
-    if (std::optional<PrimType> T = classify(Init)) {
-      const Record::Field *FieldToInit = R->getField(InitIndex);
-      if (!this->visit(Init))
-        return false;
+    if (Inits.size() == 1 && E->getType() == Inits[0]->getType()) {
+      return this->visitInitializer(Inits[0]);
+    }
 
-      if (FieldToInit->isBitField()) {
-        if (!this->emitInitBitField(*T, FieldToInit, E))
-          return false;
-      } else {
-        if (!this->emitInitField(*T, FieldToInit->Offset, E))
-          return false;
-      }
+    unsigned InitIndex = 0;
+    for (const Expr *Init : Inits) {
+      // Skip unnamed bitfields.
+      while (InitIndex < R->getNumFields() &&
+             R->getField(InitIndex)->Decl->isUnnamedBitField())
+        ++InitIndex;
 
-      if (!this->emitPopPtr(E))
+      if (!this->emitDupPtr(E))
         return false;
-      ++InitIndex;
-    } else {
-      // Initializer for a direct base class.
-      if (const Record::Base *B = R->getBase(Init->getType())) {
-        if (!this->emitGetPtrBasePop(B->Offset, Init))
-          return false;
 
-        if (!this->visitInitializer(Init))
-          return false;
-
-        if (!this->emitFinishInitPop(E))
-          return false;
-        // Base initializers don't increase InitIndex, since they don't count
-        // into the Record's fields.
-      } else {
+      if (std::optional<PrimType> T = classify(Init)) {
         const Record::Field *FieldToInit = R->getField(InitIndex);
-        // Non-primitive case. Get a pointer to the field-to-initialize
-        // on the stack and recurse into visitInitializer().
-        if (!this->emitGetPtrField(FieldToInit->Offset, Init))
+        if (!this->visit(Init))
           return false;
 
-        if (!this->visitInitializer(Init))
-          return false;
+        if (FieldToInit->isBitField()) {
+          if (!this->emitInitBitField(*T, FieldToInit, E))
+            return false;
+        } else {
+          if (!this->emitInitField(*T, FieldToInit->Offset, E))
+            return false;
+        }
 
         if (!this->emitPopPtr(E))
           return false;
         ++InitIndex;
-      }
-    }
-  }
-  return true;
-}
+      } else {
+        // Initializer for a direct base class.
+        if (const Record::Base *B = R->getBase(Init->getType())) {
+          if (!this->emitGetPtrBasePop(B->Offset, Init))
+            return false;
 
-/// Pointer to the array(not the element!) must be on the stack when calling
-/// this.
-template <class Emitter>
-bool ByteCodeExprGen<Emitter>::visitArrayElemInit(unsigned ElemIndex,
-                                                  const Expr *Init) {
-  if (std::optional<PrimType> T = classify(Init->getType())) {
-    // Visit the primitive element like normal.
-    if (!this->visit(Init))
-      return false;
-    return this->emitInitElem(*T, ElemIndex, Init);
-  }
+          if (!this->visitInitializer(Init))
+            return false;
 
-  // Advance the pointer currently on the stack to the given
-  // dimension.
-  if (!this->emitConstUint32(ElemIndex, Init))
-    return false;
-  if (!this->emitArrayElemPtrUint32(Init))
-    return false;
-  if (!this->visitInitializer(Init))
-    return false;
-  return this->emitFinishInitPop(Init);
-}
+          if (!this->emitFinishInitPop(E))
+            return false;
+          // Base initializers don't increase InitIndex, since they don't count
+          // into the Record's fields.
+        } else {
+          const Record::Field *FieldToInit = R->getField(InitIndex);
+          // Non-primitive case. Get a pointer to the field-to-initialize
+          // on the stack and recurse into visitInitializer().
+          if (!this->emitGetPtrField(FieldToInit->Offset, Init))
+            return false;
 
-template <class Emitter>
-bool ByteCodeExprGen<Emitter>::VisitInitListExpr(const InitListExpr *E) {
-  // Handle discarding first.
-  if (DiscardResult) {
-    for (const Expr *Init : E->inits()) {
-      if (!this->discard(Init))
-        return false;
+          if (!this->visitInitializer(Init))
+            return false;
+
+          if (!this->emitPopPtr(E))
+            return false;
+          ++InitIndex;
+        }
+      }
     }
     return true;
   }
 
-  // Primitive values.
-  if (std::optional<PrimType> T = classify(E->getType())) {
-    assert(!DiscardResult);
-    if (E->getNumInits() == 0)
-      return this->visitZeroInitializer(*T, E->getType(), E);
-    assert(E->getNumInits() == 1);
-    return this->delegate(E->inits()[0]);
-  }
-
-  QualType T = E->getType();
-  if (T->isRecordType())
-    return this->visitInitList(E->inits(), E);
-
   if (T->isArrayType()) {
     unsigned ElementIndex = 0;
-    for (const Expr *Init : E->inits()) {
+    for (const Expr *Init : Inits) {
       if (!this->visitArrayElemInit(ElementIndex, Init))
         return false;
       ++ElementIndex;
@@ -1104,13 +1129,13 @@ bool ByteCodeExprGen<Emitter>::VisitInitListExpr(const InitListExpr *E) {
 
     // Expand the filler expression.
     // FIXME: This should go away.
-    if (const Expr *Filler = E->getArrayFiller()) {
+    if (ArrayFiller) {
       const ConstantArrayType *CAT =
           Ctx.getASTContext().getAsConstantArrayType(E->getType());
       uint64_t NumElems = CAT->getZExtSize();
 
       for (; ElementIndex != NumElems; ++ElementIndex) {
-        if (!this->visitArrayElemInit(ElementIndex, Filler))
+        if (!this->visitArrayElemInit(ElementIndex, ArrayFiller))
           return false;
       }
     }
@@ -1119,10 +1144,10 @@ bool ByteCodeExprGen<Emitter>::VisitInitListExpr(const InitListExpr *E) {
   }
 
   if (const auto *ComplexTy = E->getType()->getAs<ComplexType>()) {
-    unsigned NumInits = E->getNumInits();
+    unsigned NumInits = Inits.size();
 
     if (NumInits == 1)
-      return this->delegate(E->inits()[0]);
+      return this->delegate(Inits[0]);
 
     QualType ElemQT = ComplexTy->getElementType();
     PrimType ElemT = classifyPrim(ElemQT);
@@ -1136,7 +1161,7 @@ bool ByteCodeExprGen<Emitter>::VisitInitListExpr(const InitListExpr *E) {
       }
     } else if (NumInits == 2) {
       unsigned InitIndex = 0;
-      for (const Expr *Init : E->inits()) {
+      for (const Expr *Init : Inits) {
         if (!this->visit(Init))
           return false;
 
@@ -1150,14 +1175,14 @@ bool ByteCodeExprGen<Emitter>::VisitInitListExpr(const InitListExpr *E) {
 
   if (const auto *VecT = E->getType()->getAs<VectorType>()) {
     unsigned NumVecElements = VecT->getNumElements();
-    assert(NumVecElements >= E->getNumInits());
+    assert(NumVecElements >= Inits.size());
 
     QualType ElemQT = VecT->getElementType();
     PrimType ElemT = classifyPrim(ElemQT);
 
     // All initializer elements.
     unsigned InitIndex = 0;
-    for (const Expr *Init : E->inits()) {
+    for (const Expr *Init : Inits) {
       if (!this->visit(Init))
         return false;
 
@@ -1179,19 +1204,38 @@ bool ByteCodeExprGen<Emitter>::VisitInitListExpr(const InitListExpr *E) {
   return false;
 }
 
+/// Pointer to the array(not the element!) must be on the stack when calling
+/// this.
 template <class Emitter>
-bool ByteCodeExprGen<Emitter>::VisitCXXParenListInitExpr(
-    const CXXParenListInitExpr *E) {
-  if (DiscardResult) {
-    for (const Expr *Init : E->getInitExprs()) {
-      if (!this->discard(Init))
-        return false;
-    }
-    return true;
+bool ByteCodeExprGen<Emitter>::visitArrayElemInit(unsigned ElemIndex,
+                                                  const Expr *Init) {
+  if (std::optional<PrimType> T = classify(Init->getType())) {
+    // Visit the primitive element like normal.
+    if (!this->visit(Init))
+      return false;
+    return this->emitInitElem(*T, ElemIndex, Init);
   }
 
-  assert(E->getType()->isRecordType());
-  return this->visitInitList(E->getInitExprs(), E);
+  // Advance the pointer currently on the stack to the given
+  // dimension.
+  if (!this->emitConstUint32(ElemIndex, Init))
+    return false;
+  if (!this->emitArrayElemPtrUint32(Init))
+    return false;
+  if (!this->visitInitializer(Init))
+    return false;
+  return this->emitFinishInitPop(Init);
+}
+
+template <class Emitter>
+bool ByteCodeExprGen<Emitter>::VisitInitListExpr(const InitListExpr *E) {
+  return this->visitInitList(E->inits(), E->getArrayFiller(), E);
+}
+
+template <class Emitter>
+bool ByteCodeExprGen<Emitter>::VisitCXXParenListInitExpr(
+    const CXXParenListInitExpr *E) {
+  return this->visitInitList(E->getInitExprs(), E->getArrayFiller(), E);
 }
 
 template <class Emitter>
@@ -1314,6 +1358,20 @@ bool ByteCodeExprGen<Emitter>::VisitUnaryExprOrTypeTraitExpr(
     assert(E->getTypeOfArgument()->isSizelessVectorType());
   }
 
+  if (Kind == UETT_VecStep) {
+    if (const auto *VT = E->getTypeOfArgument()->getAs<VectorType>()) {
+      unsigned N = VT->getNumElements();
+
+      // The vec_step built-in functions that take a 3-component
+      // vector return 4. (OpenCL 1.1 spec 6.11.12)
+      if (N == 3)
+        N = 4;
+
+      return this->emitConst(N, E);
+    }
+    return this->emitConst(1, E);
+  }
+
   return false;
 }
 
@@ -2321,8 +2379,7 @@ bool ByteCodeExprGen<Emitter>::VisitCXXUuidofExpr(const CXXUuidofExpr *E) {
   if (!this->emitGetPtrGlobal(*GlobalIndex, E))
     return false;
 
-  const Record *R = this->getRecord(E->getType());
-  assert(R);
+  assert(this->getRecord(E->getType()));
 
   const APValue &V = E->getGuidDecl()->getAsAPValue();
   if (V.getKind() == APValue::None)
@@ -2330,41 +2387,8 @@ bool ByteCodeExprGen<Emitter>::VisitCXXUuidofExpr(const CXXUuidofExpr *E) {
 
   assert(V.isStruct());
   assert(V.getStructNumBases() == 0);
-  // FIXME: This could be useful in visitAPValue, too.
-  for (unsigned I = 0, N = V.getStructNumFields(); I != N; ++I) {
-    const APValue &F = V.getStructField(I);
-    const Record::Field *RF = R->getField(I);
-
-    if (F.isInt()) {
-      PrimType T = classifyPrim(RF->Decl->getType());
-      if (!this->visitAPValue(F, T, E))
-        return false;
-      if (!this->emitInitField(T, RF->Offset, E))
-        return false;
-    } else if (F.isArray()) {
-      assert(RF->Desc->isPrimitiveArray());
-      const auto *ArrType = RF->Decl->getType()->getAsArrayTypeUnsafe();
-      PrimType ElemT = classifyPrim(ArrType->getElementType());
-      assert(ArrType);
-
-      if (!this->emitDupPtr(E))
-        return false;
-      if (!this->emitGetPtrField(RF->Offset, E))
-        return false;
-
-      for (unsigned A = 0, AN = F.getArraySize(); A != AN; ++A) {
-        if (!this->visitAPValue(F.getArrayInitializedElt(A), ElemT, E))
-          return false;
-        if (!this->emitInitElem(ElemT, A, E))
-          return false;
-      }
-
-      if (!this->emitPopPtr(E))
-        return false;
-    } else {
-      assert(false && "I don't think this should be possible");
-    }
-  }
+  if (!this->visitAPValueInitializer(V, E))
+    return false;
 
   return this->emitFinishInit(E);
 }
@@ -2930,6 +2954,54 @@ bool ByteCodeExprGen<Emitter>::visitAPValue(const APValue &Val,
 }
 
 template <class Emitter>
+bool ByteCodeExprGen<Emitter>::visitAPValueInitializer(const APValue &Val,
+                                                       const Expr *E) {
+  if (Val.isStruct()) {
+    const Record *R = this->getRecord(E->getType());
+    assert(R);
+
+    for (unsigned I = 0, N = Val.getStructNumFields(); I != N; ++I) {
+      const APValue &F = Val.getStructField(I);
+      const Record::Field *RF = R->getField(I);
+
+      if (F.isInt()) {
+        PrimType T = classifyPrim(RF->Decl->getType());
+        if (!this->visitAPValue(F, T, E))
+          return false;
+        if (!this->emitInitField(T, RF->Offset, E))
+          return false;
+      } else if (F.isArray()) {
+        assert(RF->Desc->isPrimitiveArray());
+        const auto *ArrType = RF->Decl->getType()->getAsArrayTypeUnsafe();
+        PrimType ElemT = classifyPrim(ArrType->getElementType());
+        assert(ArrType);
+
+        if (!this->emitDupPtr(E))
+          return false;
+        if (!this->emitGetPtrField(RF->Offset, E))
+          return false;
+
+        for (unsigned A = 0, AN = F.getArraySize(); A != AN; ++A) {
+          if (!this->visitAPValue(F.getArrayInitializedElt(A), ElemT, E))
+            return false;
+          if (!this->emitInitElem(ElemT, A, E))
+            return false;
+        }
+
+        if (!this->emitPopPtr(E))
+          return false;
+      } else {
+        assert(false && "I don't think this should be possible");
+      }
+    }
+    return true;
+  }
+  // TODO: Other types.
+
+  return false;
+}
+
+template <class Emitter>
 bool ByteCodeExprGen<Emitter>::VisitBuiltinCallExpr(const CallExpr *E) {
   const Function *Func = getFunction(E->getDirectCallee());
   if (!Func)
@@ -3450,9 +3522,17 @@ bool ByteCodeExprGen<Emitter>::VisitDeclRefExpr(const DeclRefExpr *E) {
   } else if (const auto *FuncDecl = dyn_cast<FunctionDecl>(D)) {
     const Function *F = getFunction(FuncDecl);
     return F && this->emitGetFnPtr(F, E);
-  } else if (isa<TemplateParamObjectDecl>(D)) {
-    if (std::optional<unsigned> Index = P.getOrCreateGlobal(D))
-      return this->emitGetPtrGlobal(*Index, E);
+  } else if (const auto *TPOD = dyn_cast<TemplateParamObjectDecl>(D)) {
+    if (std::optional<unsigned> Index = P.getOrCreateGlobal(D)) {
+      if (!this->emitGetPtrGlobal(*Index, E))
+        return false;
+      if (std::optional<PrimType> T = classify(E->getType())) {
+        if (!this->visitAPValue(TPOD->getValue(), *T, E))
+          return false;
+        return this->emitInitGlobal(*T, *Index, E);
+      }
+      return this->visitAPValueInitializer(TPOD->getValue(), E);
+    }
     return false;
   }
 
@@ -3529,35 +3609,17 @@ void ByteCodeExprGen<Emitter>::emitCleanup() {
 
 template <class Emitter>
 unsigned
-ByteCodeExprGen<Emitter>::collectBaseOffset(const RecordType *BaseType,
-                                            const RecordType *DerivedType) {
-  assert(BaseType);
-  assert(DerivedType);
-  const auto *FinalDecl = cast<CXXRecordDecl>(BaseType->getDecl());
-  const RecordDecl *CurDecl = DerivedType->getDecl();
-  const Record *CurRecord = getRecord(CurDecl);
-  assert(CurDecl && FinalDecl);
-
-  unsigned OffsetSum = 0;
-  for (;;) {
-    assert(CurRecord->getNumBases() > 0);
-    // One level up
-    for (const Record::Base &B : CurRecord->bases()) {
-      const auto *BaseDecl = cast<CXXRecordDecl>(B.Decl);
-
-      if (BaseDecl == FinalDecl || BaseDecl->isDerivedFrom(FinalDecl)) {
-        OffsetSum += B.Offset;
-        CurRecord = B.R;
-        CurDecl = BaseDecl;
-        break;
-      }
-    }
-    if (CurDecl == FinalDecl)
-      break;
-  }
+ByteCodeExprGen<Emitter>::collectBaseOffset(const QualType BaseType,
+                                            const QualType DerivedType) {
+  const auto extractRecordDecl = [](QualType Ty) -> const CXXRecordDecl * {
+    if (const auto *PT = dyn_cast<PointerType>(Ty))
+      return PT->getPointeeType()->getAsCXXRecordDecl();
+    return Ty->getAsCXXRecordDecl();
+  };
+  const CXXRecordDecl *BaseDecl = extractRecordDecl(BaseType);
+  const CXXRecordDecl *DerivedDecl = extractRecordDecl(DerivedType);
 
-  assert(OffsetSum > 0);
-  return OffsetSum;
+  return Ctx.collectBaseOffset(BaseDecl, DerivedDecl);
 }
 
 /// Emit casts from a PrimType to another PrimType.
diff --git a/clang/lib/AST/Interp/ByteCodeExprGen.h b/clang/lib/AST/Interp/ByteCodeExprGen.h
index 7e9dc8631fc0..a89e37c67aa6 100644
--- a/clang/lib/AST/Interp/ByteCodeExprGen.h
+++ b/clang/lib/AST/Interp/ByteCodeExprGen.h
@@ -181,6 +181,7 @@ protected:
   bool visitVarDecl(const VarDecl *VD);
   /// Visit an APValue.
   bool visitAPValue(const APValue &Val, PrimType ValType, const Expr *E);
+  bool visitAPValueInitializer(const APValue &Val, const Expr *E);
 
   /// Visits an expression and converts it to a boolean.
   bool visitBool(const Expr *E);
@@ -224,7 +225,8 @@ protected:
     return this->emitFinishInitPop(I);
   }
 
-  bool visitInitList(ArrayRef<const Expr *> Inits, const Expr *E);
+  bool visitInitList(ArrayRef<const Expr *> Inits, const Expr *ArrayFiller,
+                     const Expr *E);
   bool visitArrayElemInit(unsigned ElemIndex, const Expr *Init);
 
   /// Creates a local primitive value.
@@ -283,8 +285,8 @@ private:
 
   bool emitRecordDestruction(const Record *R);
   bool emitDestruction(const Descriptor *Desc);
-  unsigned collectBaseOffset(const RecordType *BaseType,
-                             const RecordType *DerivedType);
+  unsigned collectBaseOffset(const QualType BaseType,
+                             const QualType DerivedType);
 
 protected:
   /// Variable to storage mapping.
diff --git a/clang/lib/AST/Interp/ByteCodeStmtGen.cpp b/clang/lib/AST/Interp/ByteCodeStmtGen.cpp
index 36dab6252ece..ff91baf595f1 100644
--- a/clang/lib/AST/Interp/ByteCodeStmtGen.cpp
+++ b/clang/lib/AST/Interp/ByteCodeStmtGen.cpp
@@ -189,14 +189,23 @@ bool ByteCodeStmtGen<Emitter>::visitFunc(const FunctionDecl *F) {
         if (!emitFieldInitializer(F, F->Offset, InitExpr))
           return false;
       } else if (const Type *Base = Init->getBaseClass()) {
-        // Base class initializer.
-        // Get This Base and call initializer on it.
         const auto *BaseDecl = Base->getAsCXXRecordDecl();
         assert(BaseDecl);
-        const Record::Base *B = R->getBase(BaseDecl);
-        assert(B);
-        if (!this->emitGetPtrThisBase(B->Offset, InitExpr))
-          return false;
+
+        if (Init->isBaseVirtual()) {
+          assert(R->getVirtualBase(BaseDecl));
+          if (!this->emitGetPtrThisVirtBase(BaseDecl, InitExpr))
+            return false;
+
+        } else {
+          // Base class initializer.
+          // Get This Base and call initializer on it.
+          const Record::Base *B = R->getBase(BaseDecl);
+          assert(B);
+          if (!this->emitGetPtrThisBase(B->Offset, InitExpr))
+            return false;
+        }
+
         if (!this->visitInitializer(InitExpr))
           return false;
         if (!this->emitFinishInitPop(InitExpr))
@@ -323,7 +332,8 @@ bool ByteCodeStmtGen<Emitter>::visitCompoundStmt(
 template <class Emitter>
 bool ByteCodeStmtGen<Emitter>::visitDeclStmt(const DeclStmt *DS) {
   for (auto *D : DS->decls()) {
-    if (isa<StaticAssertDecl, TagDecl, TypedefNameDecl, UsingEnumDecl>(D))
+    if (isa<StaticAssertDecl, TagDecl, TypedefNameDecl, UsingEnumDecl,
+            FunctionDecl>(D))
       continue;
 
     const auto *VD = dyn_cast<VarDecl>(D);
diff --git a/clang/lib/AST/Interp/Context.cpp b/clang/lib/AST/Interp/Context.cpp
index 274178837bf0..d51a57e5e92e 100644
--- a/clang/lib/AST/Interp/Context.cpp
+++ b/clang/lib/AST/Interp/Context.cpp
@@ -262,3 +262,36 @@ const Function *Context::getOrCreateFunction(const FunctionDecl *FD) {
 
   return Func;
 }
+
+unsigned Context::collectBaseOffset(const RecordDecl *BaseDecl,
+                                    const RecordDecl *DerivedDecl) const {
+  assert(BaseDecl);
+  assert(DerivedDecl);
+  const auto *FinalDecl = cast<CXXRecordDecl>(BaseDecl);
+  const RecordDecl *CurDecl = DerivedDecl;
+  const Record *CurRecord = P->getOrCreateRecord(CurDecl);
+  assert(CurDecl && FinalDecl);
+
+  unsigned OffsetSum = 0;
+  for (;;) {
+    assert(CurRecord->getNumBases() > 0);
+    // One level up
+    for (const Record::Base &B : CurRecord->bases()) {
+      const auto *BaseDecl = cast<CXXRecordDecl>(B.Decl);
+
+      if (BaseDecl == FinalDecl || BaseDecl->isDerivedFrom(FinalDecl)) {
+        OffsetSum += B.Offset;
+        CurRecord = B.R;
+        CurDecl = BaseDecl;
+        break;
+      }
+    }
+    if (CurDecl == FinalDecl)
+      break;
+
+    // break;
+  }
+
+  assert(OffsetSum > 0);
+  return OffsetSum;
+}
diff --git a/clang/lib/AST/Interp/Context.h b/clang/lib/AST/Interp/Context.h
index 23c439ad8912..360e9499d084 100644
--- a/clang/lib/AST/Interp/Context.h
+++ b/clang/lib/AST/Interp/Context.h
@@ -104,6 +104,9 @@ public:
   /// Returns the program. This is only needed for unittests.
   Program &getProgram() const { return *P.get(); }
 
+  unsigned collectBaseOffset(const RecordDecl *BaseDecl,
+                             const RecordDecl *DerivedDecl) const;
+
 private:
   /// Runs a function.
   bool Run(State &Parent, const Function *Func, APValue &Result);
diff --git a/clang/lib/AST/Interp/Descriptor.cpp b/clang/lib/AST/Interp/Descriptor.cpp
index a4ccc0236d29..954c58c8cb37 100644
--- a/clang/lib/AST/Interp/Descriptor.cpp
+++ b/clang/lib/AST/Interp/Descriptor.cpp
@@ -136,28 +136,66 @@ static void moveArrayDesc(Block *B, const std::byte *Src, std::byte *Dst,
   }
 }
 
+static void initField(Block *B, std::byte *Ptr, bool IsConst, bool IsMutable,
+                      bool IsActive, const Descriptor *D,
+                      unsigned FieldOffset) {
+  bool IsUnion = false; // FIXME
+  auto *Desc = reinterpret_cast<InlineDescriptor *>(Ptr + FieldOffset) - 1;
+  Desc->Offset = FieldOffset;
+  Desc->Desc = D;
+  Desc->IsInitialized = D->IsArray;
+  Desc->IsBase = false;
+  Desc->IsActive = IsActive && !IsUnion;
+  Desc->IsConst = IsConst || D->IsConst;
+  Desc->IsFieldMutable = IsMutable || D->IsMutable;
+
+  if (auto Fn = D->CtorFn)
+    Fn(B, Ptr + FieldOffset, Desc->IsConst, Desc->IsFieldMutable,
+       Desc->IsActive, D);
+}
+
+static void initBase(Block *B, std::byte *Ptr, bool IsConst, bool IsMutable,
+                     bool IsActive, const Descriptor *D, unsigned FieldOffset,
+                     bool IsVirtualBase) {
+  assert(D);
+  assert(D->ElemRecord);
+
+  bool IsUnion = D->ElemRecord->isUnion();
+  auto *Desc = reinterpret_cast<InlineDescriptor *>(Ptr + FieldOffset) - 1;
+  Desc->Offset = FieldOffset;
+  Desc->Desc = D;
+  Desc->IsInitialized = D->IsArray;
+  Desc->IsBase = true;
+  Desc->IsActive = IsActive && !IsUnion;
+  Desc->IsConst = IsConst || D->IsConst;
+  Desc->IsFieldMutable = IsMutable || D->IsMutable;
+
+  for (const auto &V : D->ElemRecord->bases())
+    initBase(B, Ptr + FieldOffset, IsConst, IsMutable, IsActive, V.Desc,
+             V.Offset, false);
+  for (const auto &F : D->ElemRecord->fields())
+    initField(B, Ptr + FieldOffset, IsConst, IsMutable, IsActive, F.Desc,
+              F.Offset);
+
+  // If this is initializing a virtual base, we do NOT want to consider its
+  // virtual bases, those are already flattened into the parent record when
+  // creating it.
+  if (IsVirtualBase)
+    return;
+
+  for (const auto &V : D->ElemRecord->virtual_bases())
+    initBase(B, Ptr + FieldOffset, IsConst, IsMutable, IsActive, V.Desc,
+             V.Offset, true);
+}
+
 static void ctorRecord(Block *B, std::byte *Ptr, bool IsConst, bool IsMutable,
                        bool IsActive, const Descriptor *D) {
-  const bool IsUnion = D->ElemRecord->isUnion();
-  auto CtorSub = [=](unsigned SubOff, const Descriptor *F, bool IsBase) {
-    auto *Desc = reinterpret_cast<InlineDescriptor *>(Ptr + SubOff) - 1;
-    Desc->Offset = SubOff;
-    Desc->Desc = F;
-    Desc->IsInitialized = F->IsArray && !IsBase;
-    Desc->IsBase = IsBase;
-    Desc->IsActive = IsActive && !IsUnion;
-    Desc->IsConst = IsConst || F->IsConst;
-    Desc->IsFieldMutable = IsMutable || F->IsMutable;
-    if (auto Fn = F->CtorFn)
-      Fn(B, Ptr + SubOff, Desc->IsConst, Desc->IsFieldMutable, Desc->IsActive,
-         F);
-  };
-  for (const auto &B : D->ElemRecord->bases())
-    CtorSub(B.Offset, B.Desc, /*isBase=*/true);
+  for (const auto &V : D->ElemRecord->bases())
+    initBase(B, Ptr, IsConst, IsMutable, IsActive, V.Desc, V.Offset, false);
   for (const auto &F : D->ElemRecord->fields())
-    CtorSub(F.Offset, F.Desc, /*isBase=*/false);
+    initField(B, Ptr, IsConst, IsMutable, IsActive, F.Desc, F.Offset);
   for (const auto &V : D->ElemRecord->virtual_bases())
-    CtorSub(V.Offset, V.Desc, /*isBase=*/true);
+    initBase(B, Ptr, IsConst, IsMutable, IsActive, V.Desc, V.Offset, true);
 }
 
 static void dtorRecord(Block *B, std::byte *Ptr, const Descriptor *D) {
diff --git a/clang/lib/AST/Interp/Descriptor.h b/clang/lib/AST/Interp/Descriptor.h
index c386fc8ac7b0..cd20495c259c 100644
--- a/clang/lib/AST/Interp/Descriptor.h
+++ b/clang/lib/AST/Interp/Descriptor.h
@@ -82,6 +82,9 @@ struct InlineDescriptor {
   InlineDescriptor(const Descriptor *D)
       : Offset(sizeof(InlineDescriptor)), IsConst(false), IsInitialized(false),
         IsBase(false), IsActive(false), IsFieldMutable(false), Desc(D) {}
+
+  void dump() const { dump(llvm::errs()); }
+  void dump(llvm::raw_ostream &OS) const;
 };
 
 /// Describes a memory block created by an allocation site.
diff --git a/clang/lib/AST/Interp/Disasm.cpp b/clang/lib/AST/Interp/Disasm.cpp
index d127f33223e8..ccdc96a79436 100644
--- a/clang/lib/AST/Interp/Disasm.cpp
+++ b/clang/lib/AST/Interp/Disasm.cpp
@@ -200,7 +200,7 @@ LLVM_DUMP_METHOD void Descriptor::dump(llvm::raw_ostream &OS) const {
     OS << " primitive";
 
   if (isZeroSizeArray())
-    OS << " zero-size-arrary";
+    OS << " zero-size-array";
   else if (isUnknownSizeArray())
     OS << " unknown-size-array";
 
@@ -208,6 +208,25 @@ LLVM_DUMP_METHOD void Descriptor::dump(llvm::raw_ostream &OS) const {
     OS << " dummy";
 }
 
+LLVM_DUMP_METHOD void InlineDescriptor::dump(llvm::raw_ostream &OS) const {
+  {
+    ColorScope SC(OS, true, {llvm::raw_ostream::BLUE, true});
+    OS << "InlineDescriptor " << (const void *)this << "\n";
+  }
+  OS << "Offset: " << Offset << "\n";
+  OS << "IsConst: " << IsConst << "\n";
+  OS << "IsInitialized: " << IsInitialized << "\n";
+  OS << "IsBase: " << IsBase << "\n";
+  OS << "IsActive: " << IsActive << "\n";
+  OS << "IsFieldMutable: " << IsFieldMutable << "\n";
+  OS << "Desc: ";
+  if (Desc)
+    Desc->dump(OS);
+  else
+    OS << "nullptr";
+  OS << "\n";
+}
+
 LLVM_DUMP_METHOD void InterpFrame::dump(llvm::raw_ostream &OS,
                                         unsigned Indent) const {
   unsigned Spaces = Indent * 2;
@@ -251,8 +270,6 @@ LLVM_DUMP_METHOD void Record::dump(llvm::raw_ostream &OS, unsigned Indentation,
     ++I;
   }
 
-  // FIXME: Virtual bases.
-
   I = 0;
   for (const Record::Field &F : fields()) {
     OS.indent(Indent) << "- Field " << I << ": ";
@@ -263,6 +280,14 @@ LLVM_DUMP_METHOD void Record::dump(llvm::raw_ostream &OS, unsigned Indentation,
     OS << ". Offset " << (Offset + F.Offset) << "\n";
     ++I;
   }
+
+  I = 0;
+  for (const Record::Base &B : virtual_bases()) {
+    OS.indent(Indent) << "- Virtual Base " << I << ". Offset "
+                      << (Offset + B.Offset) << "\n";
+    B.R->dump(OS, Indentation + 1, Offset + B.Offset);
+    ++I;
+  }
 }
 
 LLVM_DUMP_METHOD void Block::dump(llvm::raw_ostream &OS) const {
diff --git a/clang/lib/AST/Interp/Interp.h b/clang/lib/AST/Interp/Interp.h
index 9283f697c007..9da0286deada 100644
--- a/clang/lib/AST/Interp/Interp.h
+++ b/clang/lib/AST/Interp/Interp.h
@@ -1355,20 +1355,26 @@ inline bool VirtBaseHelper(InterpState &S, CodePtr OpPC, const RecordDecl *Decl,
   while (Base.isBaseClass())
     Base = Base.getBase();
 
-  auto *Field = Base.getRecord()->getVirtualBase(Decl);
-  S.Stk.push<Pointer>(Base.atField(Field->Offset));
+  const Record::Base *VirtBase = Base.getRecord()->getVirtualBase(Decl);
+  S.Stk.push<Pointer>(Base.atField(VirtBase->Offset));
   return true;
 }
 
-inline bool GetPtrVirtBase(InterpState &S, CodePtr OpPC, const RecordDecl *D) {
+inline bool GetPtrVirtBasePop(InterpState &S, CodePtr OpPC,
+                              const RecordDecl *D) {
+  assert(D);
   const Pointer &Ptr = S.Stk.pop<Pointer>();
   if (!CheckNull(S, OpPC, Ptr, CSK_Base))
     return false;
+  if (Ptr.isDummy()) // FIXME: Once we have type info for dummy pointers, this
+                     // needs to go.
+    return false;
   return VirtBaseHelper(S, OpPC, D, Ptr);
 }
 
 inline bool GetPtrThisVirtBase(InterpState &S, CodePtr OpPC,
                                const RecordDecl *D) {
+  assert(D);
   if (S.checkingPotentialConstantExpression())
     return false;
   const Pointer &This = S.Current->getThis();
diff --git a/clang/lib/AST/Interp/Opcodes.td b/clang/lib/AST/Interp/Opcodes.td
index 742785b28eb4..2a97b978b523 100644
--- a/clang/lib/AST/Interp/Opcodes.td
+++ b/clang/lib/AST/Interp/Opcodes.td
@@ -336,7 +336,7 @@ def GetPtrDerivedPop : Opcode {
 }
 
 // [Pointer] -> [Pointer]
-def GetPtrVirtBase : Opcode {
+def GetPtrVirtBasePop : Opcode {
   // RecordDecl of base class.
   let Args = [ArgRecordDecl];
 }
diff --git a/clang/lib/AST/Interp/Program.cpp b/clang/lib/AST/Interp/Program.cpp
index 3773e0662f78..02075c20cf55 100644
--- a/clang/lib/AST/Interp/Program.cpp
+++ b/clang/lib/AST/Interp/Program.cpp
@@ -173,7 +173,8 @@ std::optional<unsigned> Program::createGlobal(const ValueDecl *VD,
   if (const auto *Var = dyn_cast<VarDecl>(VD)) {
     IsStatic = Context::shouldBeGloballyIndexed(VD);
     IsExtern = Var->hasExternalStorage();
-  } else if (isa<UnnamedGlobalConstantDecl, MSGuidDecl>(VD)) {
+  } else if (isa<UnnamedGlobalConstantDecl, MSGuidDecl,
+                 TemplateParamObjectDecl>(VD)) {
     IsStatic = true;
     IsExtern = false;
   } else {
diff --git a/clang/lib/AST/ItaniumMangle.cpp b/clang/lib/AST/ItaniumMangle.cpp
index 106c69dd5bee..ed9e6eeb36c7 100644
--- a/clang/lib/AST/ItaniumMangle.cpp
+++ b/clang/lib/AST/ItaniumMangle.cpp
@@ -4715,7 +4715,7 @@ recurse:
   case Expr::MSPropertySubscriptExprClass:
   case Expr::TypoExprClass: // This should no longer exist in the AST by now.
   case Expr::RecoveryExprClass:
-  case Expr::OMPArraySectionExprClass:
+  case Expr::ArraySectionExprClass:
   case Expr::OMPArrayShapingExprClass:
   case Expr::OMPIteratorExprClass:
   case Expr::CXXInheritedCtorInitExprClass:
diff --git a/clang/lib/AST/NSAPI.cpp b/clang/lib/AST/NSAPI.cpp
index ecc56c13fb75..6f586173edb0 100644
--- a/clang/lib/AST/NSAPI.cpp
+++ b/clang/lib/AST/NSAPI.cpp
@@ -462,7 +462,7 @@ NSAPI::getNSNumberFactoryMethodKind(QualType T) const {
   case BuiltinType::PseudoObject:
   case BuiltinType::BuiltinFn:
   case BuiltinType::IncompleteMatrixIdx:
-  case BuiltinType::OMPArraySection:
+  case BuiltinType::ArraySection:
   case BuiltinType::OMPArrayShaping:
   case BuiltinType::OMPIterator:
   case BuiltinType::BFloat16:
diff --git a/clang/lib/AST/StmtPrinter.cpp b/clang/lib/AST/StmtPrinter.cpp
index 5855ab3141ed..f010d36513a4 100644
--- a/clang/lib/AST/StmtPrinter.cpp
+++ b/clang/lib/AST/StmtPrinter.cpp
@@ -1521,7 +1521,7 @@ void StmtPrinter::VisitMatrixSubscriptExpr(MatrixSubscriptExpr *Node) {
   OS << "]";
 }
 
-void StmtPrinter::VisitOMPArraySectionExpr(OMPArraySectionExpr *Node) {
+void StmtPrinter::VisitArraySectionExpr(ArraySectionExpr *Node) {
   PrintExpr(Node->getBase());
   OS << "[";
   if (Node->getLowerBound())
@@ -1531,7 +1531,7 @@ void StmtPrinter::VisitOMPArraySectionExpr(OMPArraySectionExpr *Node) {
     if (Node->getLength())
       PrintExpr(Node->getLength());
   }
-  if (Node->getColonLocSecond().isValid()) {
+  if (Node->isOMPArraySection() && Node->getColonLocSecond().isValid()) {
     OS << ":";
     if (Node->getStride())
       PrintExpr(Node->getStride());
diff --git a/clang/lib/AST/StmtProfile.cpp b/clang/lib/AST/StmtProfile.cpp
index c81724f84dd9..a95f5c6103e2 100644
--- a/clang/lib/AST/StmtProfile.cpp
+++ b/clang/lib/AST/StmtProfile.cpp
@@ -1435,7 +1435,7 @@ void StmtProfiler::VisitMatrixSubscriptExpr(const MatrixSubscriptExpr *S) {
   VisitExpr(S);
 }
 
-void StmtProfiler::VisitOMPArraySectionExpr(const OMPArraySectionExpr *S) {
+void StmtProfiler::VisitArraySectionExpr(const ArraySectionExpr *S) {
   VisitExpr(S);
 }
 
diff --git a/clang/lib/AST/Type.cpp b/clang/lib/AST/Type.cpp
index cb22c91a12aa..68e81f45b4c2 100644
--- a/clang/lib/AST/Type.cpp
+++ b/clang/lib/AST/Type.cpp
@@ -2510,6 +2510,18 @@ bool Type::isSveVLSBuiltinType() const {
   return false;
 }
 
+QualType Type::getSizelessVectorEltType(const ASTContext &Ctx) const {
+  assert(isSizelessVectorType() && "Must be sizeless vector type");
+  // Currently supports SVE and RVV
+  if (isSVESizelessBuiltinType())
+    return getSveEltType(Ctx);
+
+  if (isRVVSizelessBuiltinType())
+    return getRVVEltType(Ctx);
+
+  llvm_unreachable("Unhandled type");
+}
+
 QualType Type::getSveEltType(const ASTContext &Ctx) const {
   assert(isSveVLSBuiltinType() && "unsupported type!");
 
@@ -3413,8 +3425,8 @@ StringRef BuiltinType::getName(const PrintingPolicy &Policy) const {
     return "reserve_id_t";
   case IncompleteMatrixIdx:
     return "<incomplete matrix index type>";
-  case OMPArraySection:
-    return "<OpenMP array section type>";
+  case ArraySection:
+    return "<array section type>";
   case OMPArrayShaping:
     return "<OpenMP array shaping type>";
   case OMPIterator:
@@ -4710,7 +4722,7 @@ bool Type::canHaveNullability(bool ResultIfUnknown) const {
     case BuiltinType::BuiltinFn:
     case BuiltinType::NullPtr:
     case BuiltinType::IncompleteMatrixIdx:
-    case BuiltinType::OMPArraySection:
+    case BuiltinType::ArraySection:
     case BuiltinType::OMPArrayShaping:
     case BuiltinType::OMPIterator:
       return false;
diff --git a/clang/lib/AST/TypeLoc.cpp b/clang/lib/AST/TypeLoc.cpp
index 21e152f6aea8..ce45b47d5cfe 100644
--- a/clang/lib/AST/TypeLoc.cpp
+++ b/clang/lib/AST/TypeLoc.cpp
@@ -429,7 +429,7 @@ TypeSpecifierType BuiltinTypeLoc::getWrittenTypeSpec() const {
 #include "clang/Basic/WebAssemblyReferenceTypes.def"
   case BuiltinType::BuiltinFn:
   case BuiltinType::IncompleteMatrixIdx:
-  case BuiltinType::OMPArraySection:
+  case BuiltinType::ArraySection:
   case BuiltinType::OMPArrayShaping:
   case BuiltinType::OMPIterator:
     return TST_unspecified;
diff --git a/clang/lib/Analysis/FlowSensitive/ASTOps.cpp b/clang/lib/Analysis/FlowSensitive/ASTOps.cpp
index 619bf772bba5..bd1676583ecc 100644
--- a/clang/lib/Analysis/FlowSensitive/ASTOps.cpp
+++ b/clang/lib/Analysis/FlowSensitive/ASTOps.cpp
@@ -33,12 +33,20 @@ namespace clang::dataflow {
 
 const Expr &ignoreCFGOmittedNodes(const Expr &E) {
   const Expr *Current = &E;
-  if (auto *EWC = dyn_cast<ExprWithCleanups>(Current)) {
-    Current = EWC->getSubExpr();
+  const Expr *Last = nullptr;
+  while (Current != Last) {
+    Last = Current;
+    if (auto *EWC = dyn_cast<ExprWithCleanups>(Current)) {
+      Current = EWC->getSubExpr();
+      assert(Current != nullptr);
+    }
+    if (auto *CE = dyn_cast<ConstantExpr>(Current)) {
+      Current = CE->getSubExpr();
+      assert(Current != nullptr);
+    }
+    Current = Current->IgnoreParens();
     assert(Current != nullptr);
   }
-  Current = Current->IgnoreParens();
-  assert(Current != nullptr);
   return *Current;
 }
 
diff --git a/clang/lib/Analysis/FlowSensitive/DataflowEnvironment.cpp b/clang/lib/Analysis/FlowSensitive/DataflowEnvironment.cpp
index 05395e07a7a6..d79e73440289 100644
--- a/clang/lib/Analysis/FlowSensitive/DataflowEnvironment.cpp
+++ b/clang/lib/Analysis/FlowSensitive/DataflowEnvironment.cpp
@@ -157,7 +157,13 @@ static WidenResult widenDistinctValues(QualType Type, Value &Prev,
                                        Value &Current, Environment &CurrentEnv,
                                        Environment::ValueModel &Model) {
   // Boolean-model widening.
-  if (auto *PrevBool = dyn_cast<BoolValue>(&Prev)) {
+  if (isa<BoolValue>(Prev) && isa<BoolValue>(Current)) {
+    // FIXME: Checking both values should be unnecessary, but we can currently
+    // end up with `BoolValue`s in integer-typed variables. See comment in
+    // `joinDistinctValues()` for details.
+    auto &PrevBool = cast<BoolValue>(Prev);
+    auto &CurBool = cast<BoolValue>(Current);
+
     if (isa<TopBoolValue>(Prev))
       // Safe to return `Prev` here, because Top is never dependent on the
       // environment.
@@ -166,13 +172,12 @@ static WidenResult widenDistinctValues(QualType Type, Value &Prev,
     // We may need to widen to Top, but before we do so, check whether both
     // values are implied to be either true or false in the current environment.
     // In that case, we can simply return a literal instead.
-    auto &CurBool = cast<BoolValue>(Current);
-    bool TruePrev = PrevEnv.proves(PrevBool->formula());
+    bool TruePrev = PrevEnv.proves(PrevBool.formula());
     bool TrueCur = CurrentEnv.proves(CurBool.formula());
     if (TruePrev && TrueCur)
       return {&CurrentEnv.getBoolLiteralValue(true), LatticeEffect::Unchanged};
     if (!TruePrev && !TrueCur &&
-        PrevEnv.proves(PrevEnv.arena().makeNot(PrevBool->formula())) &&
+        PrevEnv.proves(PrevEnv.arena().makeNot(PrevBool.formula())) &&
         CurrentEnv.proves(CurrentEnv.arena().makeNot(CurBool.formula())))
       return {&CurrentEnv.getBoolLiteralValue(false), LatticeEffect::Unchanged};
 
@@ -237,13 +242,8 @@ joinLocToVal(const llvm::MapVector<const StorageLocation *, Value *> &LocToVal,
       continue;
     assert(It->second != nullptr);
 
-    if (areEquivalentValues(*Val, *It->second)) {
-      Result.insert({Loc, Val});
-      continue;
-    }
-
-    if (Value *JoinedVal = joinDistinctValues(
-            Loc->getType(), *Val, Env1, *It->second, Env2, JoinedEnv, Model)) {
+    if (Value *JoinedVal = Environment::joinValues(
+            Loc->getType(), Val, Env1, It->second, Env2, JoinedEnv, Model)) {
       Result.insert({Loc, JoinedVal});
     }
   }
@@ -338,6 +338,18 @@ public:
     }
   }
 
+  bool TraverseDecl(Decl *D) {
+    // Don't traverse nested record or function declarations.
+    // - We won't be analyzing code contained in these anyway
+    // - We don't model fields that are used only in these nested declaration,
+    //   so trying to propagate a result object to initializers of such fields
+    //   would cause an error.
+    if (isa_and_nonnull<RecordDecl>(D) || isa_and_nonnull<FunctionDecl>(D))
+      return true;
+
+    return RecursiveASTVisitor<ResultObjectVisitor>::TraverseDecl(D);
+  }
+
   bool TraverseBindingDecl(BindingDecl *BD) {
     // `RecursiveASTVisitor` doesn't traverse holding variables for
     // `BindingDecl`s by itself, so we need to tell it to.
@@ -775,27 +787,16 @@ Environment Environment::join(const Environment &EnvA, const Environment &EnvB,
   JoinedEnv.LocForRecordReturnVal = EnvA.LocForRecordReturnVal;
   JoinedEnv.ThisPointeeLoc = EnvA.ThisPointeeLoc;
 
-  if (EnvA.ReturnVal == nullptr || EnvB.ReturnVal == nullptr) {
-    // `ReturnVal` might not always get set -- for example if we have a return
-    // statement of the form `return some_other_func()` and we decide not to
-    // analyze `some_other_func()`.
-    // In this case, we can't say anything about the joined return value -- we
-    // don't simply want to propagate the return value that we do have, because
-    // it might not be the correct one.
-    // This occurs for example in the test `ContextSensitiveMutualRecursion`.
+  if (EnvA.CallStack.empty()) {
     JoinedEnv.ReturnVal = nullptr;
-  } else if (areEquivalentValues(*EnvA.ReturnVal, *EnvB.ReturnVal)) {
-    JoinedEnv.ReturnVal = EnvA.ReturnVal;
   } else {
-    assert(!EnvA.CallStack.empty());
     // FIXME: Make `CallStack` a vector of `FunctionDecl` so we don't need this
     // cast.
     auto *Func = dyn_cast<FunctionDecl>(EnvA.CallStack.back());
     assert(Func != nullptr);
-    if (Value *JoinedVal =
-            joinDistinctValues(Func->getReturnType(), *EnvA.ReturnVal, EnvA,
-                               *EnvB.ReturnVal, EnvB, JoinedEnv, Model))
-      JoinedEnv.ReturnVal = JoinedVal;
+    JoinedEnv.ReturnVal =
+        joinValues(Func->getReturnType(), EnvA.ReturnVal, EnvA, EnvB.ReturnVal,
+                   EnvB, JoinedEnv, Model);
   }
 
   if (EnvA.ReturnLoc == EnvB.ReturnLoc)
@@ -821,6 +822,24 @@ Environment Environment::join(const Environment &EnvA, const Environment &EnvB,
   return JoinedEnv;
 }
 
+Value *Environment::joinValues(QualType Ty, Value *Val1,
+                               const Environment &Env1, Value *Val2,
+                               const Environment &Env2, Environment &JoinedEnv,
+                               Environment::ValueModel &Model) {
+  if (Val1 == nullptr || Val2 == nullptr)
+    // We can't say anything about the joined value -- even if one of the values
+    // is non-null, we don't want to simply propagate it, because it would be
+    // too specific: Because the other value is null, that means we have no
+    // information at all about the value (i.e. the value is unconstrained).
+    return nullptr;
+
+  if (areEquivalentValues(*Val1, *Val2))
+    // Arbitrarily return one of the two values.
+    return Val1;
+
+  return joinDistinctValues(Ty, *Val1, Env1, *Val2, Env2, JoinedEnv, Model);
+}
+
 StorageLocation &Environment::createStorageLocation(QualType Type) {
   return DACtx->createStorageLocation(Type);
 }
diff --git a/clang/lib/Analysis/FlowSensitive/Transfer.cpp b/clang/lib/Analysis/FlowSensitive/Transfer.cpp
index 2771c8b2e37e..fd224aeb79b1 100644
--- a/clang/lib/Analysis/FlowSensitive/Transfer.cpp
+++ b/clang/lib/Analysis/FlowSensitive/Transfer.cpp
@@ -41,7 +41,11 @@ namespace dataflow {
 
 const Environment *StmtToEnvMap::getEnvironment(const Stmt &S) const {
   auto BlockIt = ACFG.getStmtToBlock().find(&ignoreCFGOmittedNodes(S));
-  assert(BlockIt != ACFG.getStmtToBlock().end());
+  if (BlockIt == ACFG.getStmtToBlock().end()) {
+    assert(false);
+    // Return null to avoid dereferencing the end iterator in non-assert builds.
+    return nullptr;
+  }
   if (!ACFG.isBlockReachable(*BlockIt->getSecond()))
     return nullptr;
   if (BlockIt->getSecond()->getBlockID() == CurBlockID)
@@ -124,8 +128,9 @@ namespace {
 
 class TransferVisitor : public ConstStmtVisitor<TransferVisitor> {
 public:
-  TransferVisitor(const StmtToEnvMap &StmtToEnv, Environment &Env)
-      : StmtToEnv(StmtToEnv), Env(Env) {}
+  TransferVisitor(const StmtToEnvMap &StmtToEnv, Environment &Env,
+                  Environment::ValueModel &Model)
+      : StmtToEnv(StmtToEnv), Env(Env), Model(Model) {}
 
   void VisitBinaryOperator(const BinaryOperator *S) {
     const Expr *LHS = S->getLHS();
@@ -641,17 +646,42 @@ public:
   }
 
   void VisitConditionalOperator(const ConditionalOperator *S) {
-    // FIXME: Revisit this once flow conditions are added to the framework. For
-    // `a = b ? c : d` we can add `b => a == c && !b => a == d` to the flow
-    // condition.
-    // When we do this, we will need to retrieve the values of the operands from
-    // the environments for the basic blocks they are computed in, in a similar
-    // way to how this is done for short-circuited logical operators in
-    // `getLogicOperatorSubExprValue()`.
-    if (S->isGLValue())
-      Env.setStorageLocation(*S, Env.createObject(S->getType()));
-    else if (!S->getType()->isRecordType()) {
-      if (Value *Val = Env.createValue(S->getType()))
+    const Environment *TrueEnv = StmtToEnv.getEnvironment(*S->getTrueExpr());
+    const Environment *FalseEnv = StmtToEnv.getEnvironment(*S->getFalseExpr());
+
+    if (TrueEnv == nullptr || FalseEnv == nullptr) {
+      // If the true or false branch is dead, we may not have an environment for
+      // it. We could handle this specifically by forwarding the value or
+      // location of the live branch, but this case is rare enough that this
+      // probably isn't worth the additional complexity.
+      return;
+    }
+
+    if (S->isGLValue()) {
+      StorageLocation *TrueLoc = TrueEnv->getStorageLocation(*S->getTrueExpr());
+      StorageLocation *FalseLoc =
+          FalseEnv->getStorageLocation(*S->getFalseExpr());
+      if (TrueLoc == FalseLoc && TrueLoc != nullptr)
+        Env.setStorageLocation(*S, *TrueLoc);
+    } else if (!S->getType()->isRecordType()) {
+      // The conditional operator can evaluate to either of the values of the
+      // two branches. To model this, join these two values together to yield
+      // the result of the conditional operator.
+      // Note: Most joins happen in `computeBlockInputState()`, but this case is
+      // different:
+      // - `computeBlockInputState()` (which in turn calls `Environment::join()`
+      //   joins values associated with the _same_ expression or storage
+      //   location, then associates the joined value with that expression or
+      //   storage location. This join has nothing to do with transfer --
+      //   instead, it joins together the results of performing transfer on two
+      //   different blocks.
+      // - Here, we join values associated with _different_ expressions (the
+      //   true and false branch), then associate the joined value with a third
+      //   expression (the conditional operator itself). This join is what it
+      //   means to perform transfer on the conditional operator.
+      if (Value *Val = Environment::joinValues(
+              S->getType(), TrueEnv->getValue(*S->getTrueExpr()), *TrueEnv,
+              FalseEnv->getValue(*S->getFalseExpr()), *FalseEnv, Env, Model))
         Env.setValue(*S, *Val);
     }
   }
@@ -810,12 +840,14 @@ private:
 
   const StmtToEnvMap &StmtToEnv;
   Environment &Env;
+  Environment::ValueModel &Model;
 };
 
 } // namespace
 
-void transfer(const StmtToEnvMap &StmtToEnv, const Stmt &S, Environment &Env) {
-  TransferVisitor(StmtToEnv, Env).Visit(&S);
+void transfer(const StmtToEnvMap &StmtToEnv, const Stmt &S, Environment &Env,
+              Environment::ValueModel &Model) {
+  TransferVisitor(StmtToEnv, Env, Model).Visit(&S);
 }
 
 } // namespace dataflow
diff --git a/clang/lib/Analysis/FlowSensitive/TypeErasedDataflowAnalysis.cpp b/clang/lib/Analysis/FlowSensitive/TypeErasedDataflowAnalysis.cpp
index 71d5c1a6c4f4..12eff4dd4b78 100644
--- a/clang/lib/Analysis/FlowSensitive/TypeErasedDataflowAnalysis.cpp
+++ b/clang/lib/Analysis/FlowSensitive/TypeErasedDataflowAnalysis.cpp
@@ -316,7 +316,7 @@ builtinTransferStatement(unsigned CurBlockID, const CFGStmt &Elt,
   const Stmt *S = Elt.getStmt();
   assert(S != nullptr);
   transfer(StmtToEnvMap(AC.ACFG, AC.BlockStates, CurBlockID, InputState), *S,
-           InputState.Env);
+           InputState.Env, AC.Analysis);
 }
 
 /// Built-in transfer function for `CFGInitializer`.
@@ -452,7 +452,7 @@ transferCFGBlock(const CFGBlock &Block, AnalysisContext &AC,
       // terminator condition, but not as a `CFGElement`. The condition of an if
       // statement is one such example.
       transfer(StmtToEnvMap(AC.ACFG, AC.BlockStates, Block.getBlockID(), State),
-               *TerminatorCond, State.Env);
+               *TerminatorCond, State.Env, AC.Analysis);
 
     // If the transfer function didn't produce a value, create an atom so that
     // we have *some* value for the condition expression. This ensures that
diff --git a/clang/lib/Basic/FileManager.cpp b/clang/lib/Basic/FileManager.cpp
index cd520a6375e0..143c04309d07 100644
--- a/clang/lib/Basic/FileManager.cpp
+++ b/clang/lib/Basic/FileManager.cpp
@@ -39,12 +39,6 @@ using namespace clang;
 
 #define DEBUG_TYPE "file-search"
 
-ALWAYS_ENABLED_STATISTIC(NumDirLookups, "Number of directory lookups.");
-ALWAYS_ENABLED_STATISTIC(NumFileLookups, "Number of file lookups.");
-ALWAYS_ENABLED_STATISTIC(NumDirCacheMisses,
-                         "Number of directory cache misses.");
-ALWAYS_ENABLED_STATISTIC(NumFileCacheMisses, "Number of file cache misses.");
-
 //===----------------------------------------------------------------------===//
 // Common logic.
 //===----------------------------------------------------------------------===//
@@ -656,6 +650,14 @@ StringRef FileManager::getCanonicalName(const void *Entry, StringRef Name) {
   return CanonicalName;
 }
 
+void FileManager::AddStats(const FileManager &Other) {
+  assert(&Other != this && "Collecting stats into the same FileManager");
+  NumDirLookups += Other.NumDirLookups;
+  NumFileLookups += Other.NumFileLookups;
+  NumDirCacheMisses += Other.NumDirCacheMisses;
+  NumFileCacheMisses += Other.NumFileCacheMisses;
+}
+
 void FileManager::PrintStats() const {
   llvm::errs() << "\n*** File Manager Stats:\n";
   llvm::errs() << UniqueRealFiles.size() << " real files found, "
diff --git a/clang/lib/Basic/Targets/ARM.h b/clang/lib/Basic/Targets/ARM.h
index e69adbe75473..df9855a52e61 100644
--- a/clang/lib/Basic/Targets/ARM.h
+++ b/clang/lib/Basic/Targets/ARM.h
@@ -225,6 +225,10 @@ public:
   bool hasBitIntType() const override { return true; }
 
   const char *getBFloat16Mangling() const override { return "u6__bf16"; };
+
+  std::pair<unsigned, unsigned> hardwareInterferenceSizes() const override {
+    return std::make_pair(getTriple().isArch64Bit() ? 256 : 64, 64);
+  }
 };
 
 class LLVM_LIBRARY_VISIBILITY ARMleTargetInfo : public ARMTargetInfo {
diff --git a/clang/lib/Basic/Targets/AVR.h b/clang/lib/Basic/Targets/AVR.h
index 9376c46cd98c..feeb04f37eeb 100644
--- a/clang/lib/Basic/Targets/AVR.h
+++ b/clang/lib/Basic/Targets/AVR.h
@@ -175,6 +175,10 @@ public:
   std::optional<std::string> handleAsmEscapedChar(char EscChar) const override;
   StringRef getABI() const override { return ABI; }
 
+  std::pair<unsigned, unsigned> hardwareInterferenceSizes() const override {
+    return std::make_pair(32, 32);
+  }
+
 protected:
   std::string CPU;
   StringRef ABI;
diff --git a/clang/lib/Basic/Targets/BPF.h b/clang/lib/Basic/Targets/BPF.h
index 489f29fc4fea..d19b37dd4df7 100644
--- a/clang/lib/Basic/Targets/BPF.h
+++ b/clang/lib/Basic/Targets/BPF.h
@@ -113,6 +113,10 @@ public:
     StringRef CPUName(Name);
     return isValidCPUName(CPUName);
   }
+
+  std::pair<unsigned, unsigned> hardwareInterferenceSizes() const override {
+    return std::make_pair(32, 32);
+  }
 };
 } // namespace targets
 } // namespace clang
diff --git a/clang/lib/Basic/Targets/M68k.h b/clang/lib/Basic/Targets/M68k.h
index 7ffa901127e5..b732add77e03 100644
--- a/clang/lib/Basic/Targets/M68k.h
+++ b/clang/lib/Basic/Targets/M68k.h
@@ -56,6 +56,10 @@ public:
   BuiltinVaListKind getBuiltinVaListKind() const override;
   bool setCPU(const std::string &Name) override;
   CallingConvCheckResult checkCallingConvention(CallingConv CC) const override;
+
+  std::pair<unsigned, unsigned> hardwareInterferenceSizes() const override {
+    return std::make_pair(32, 32);
+  }
 };
 
 } // namespace targets
diff --git a/clang/lib/Basic/Targets/Mips.h b/clang/lib/Basic/Targets/Mips.h
index 0d6e4b4d0808..730deb674aa5 100644
--- a/clang/lib/Basic/Targets/Mips.h
+++ b/clang/lib/Basic/Targets/Mips.h
@@ -431,6 +431,10 @@ public:
 
   bool validateTarget(DiagnosticsEngine &Diags) const override;
   bool hasBitIntType() const override { return true; }
+
+  std::pair<unsigned, unsigned> hardwareInterferenceSizes() const override {
+    return std::make_pair(32, 32);
+  }
 };
 } // namespace targets
 } // namespace clang
diff --git a/clang/lib/Basic/Targets/PPC.h b/clang/lib/Basic/Targets/PPC.h
index 60bc1dec8f95..cd0f08dfb3bc 100644
--- a/clang/lib/Basic/Targets/PPC.h
+++ b/clang/lib/Basic/Targets/PPC.h
@@ -423,6 +423,10 @@ public:
     // This is the ELF definition
     return TargetInfo::PowerABIBuiltinVaList;
   }
+
+  std::pair<unsigned, unsigned> hardwareInterferenceSizes() const override {
+    return std::make_pair(32, 32);
+  }
 };
 
 // Note: ABI differences may eventually require us to have a separate
@@ -503,6 +507,10 @@ public:
       return CCCR_Warning;
     }
   }
+
+  std::pair<unsigned, unsigned> hardwareInterferenceSizes() const override {
+    return std::make_pair(128, 128);
+  }
 };
 
 class LLVM_LIBRARY_VISIBILITY AIXPPC32TargetInfo :
diff --git a/clang/lib/Basic/Targets/RISCV.h b/clang/lib/Basic/Targets/RISCV.h
index 78580b5b1c10..d0e9cdc6da07 100644
--- a/clang/lib/Basic/Targets/RISCV.h
+++ b/clang/lib/Basic/Targets/RISCV.h
@@ -16,7 +16,7 @@
 #include "clang/Basic/TargetInfo.h"
 #include "clang/Basic/TargetOptions.h"
 #include "llvm/Support/Compiler.h"
-#include "llvm/Support/RISCVISAInfo.h"
+#include "llvm/TargetParser/RISCVISAInfo.h"
 #include "llvm/TargetParser/Triple.h"
 #include <optional>
 
@@ -122,6 +122,10 @@ public:
   void fillValidTuneCPUList(SmallVectorImpl<StringRef> &Values) const override;
   bool supportsTargetAttributeTune() const override { return true; }
   ParsedTargetAttr parseTargetAttr(StringRef Str) const override;
+
+  std::pair<unsigned, unsigned> hardwareInterferenceSizes() const override {
+    return std::make_pair(32, 32);
+  }
 };
 class LLVM_LIBRARY_VISIBILITY RISCV32TargetInfo : public RISCVTargetInfo {
 public:
diff --git a/clang/lib/Basic/Targets/Sparc.h b/clang/lib/Basic/Targets/Sparc.h
index 214fef88e1dc..3357bee33e1a 100644
--- a/clang/lib/Basic/Targets/Sparc.h
+++ b/clang/lib/Basic/Targets/Sparc.h
@@ -140,6 +140,10 @@ public:
     CPU = getCPUKind(Name);
     return CPU != CK_GENERIC;
   }
+
+  std::pair<unsigned, unsigned> hardwareInterferenceSizes() const override {
+    return std::make_pair(32, 32);
+  }
 };
 
 // SPARC v8 is the 32-bit mode selected by Triple::sparc.
diff --git a/clang/lib/Basic/Targets/SystemZ.h b/clang/lib/Basic/Targets/SystemZ.h
index 8e302acd51b8..73d3aa01a043 100644
--- a/clang/lib/Basic/Targets/SystemZ.h
+++ b/clang/lib/Basic/Targets/SystemZ.h
@@ -220,6 +220,10 @@ public:
   int getEHDataRegisterNumber(unsigned RegNo) const override {
     return RegNo < 4 ? 6 + RegNo : -1;
   }
+
+  std::pair<unsigned, unsigned> hardwareInterferenceSizes() const override {
+    return std::make_pair(256, 256);
+  }
 };
 } // namespace targets
 } // namespace clang
diff --git a/clang/lib/Basic/Targets/WebAssembly.cpp b/clang/lib/Basic/Targets/WebAssembly.cpp
index f1c925d90cb6..3d76411f890a 100644
--- a/clang/lib/Basic/Targets/WebAssembly.cpp
+++ b/clang/lib/Basic/Targets/WebAssembly.cpp
@@ -47,6 +47,7 @@ bool WebAssemblyTargetInfo::hasFeature(StringRef Feature) const {
   return llvm::StringSwitch<bool>(Feature)
       .Case("simd128", SIMDLevel >= SIMD128)
       .Case("relaxed-simd", SIMDLevel >= RelaxedSIMD)
+      .Case("half-precision", HasHalfPrecision)
       .Case("nontrapping-fptoint", HasNontrappingFPToInt)
       .Case("sign-ext", HasSignExt)
       .Case("exception-handling", HasExceptionHandling)
@@ -148,18 +149,19 @@ bool WebAssemblyTargetInfo::initFeatureMap(
     llvm::StringMap<bool> &Features, DiagnosticsEngine &Diags, StringRef CPU,
     const std::vector<std::string> &FeaturesVec) const {
   if (CPU == "bleeding-edge") {
-    Features["nontrapping-fptoint"] = true;
-    Features["sign-ext"] = true;
-    Features["bulk-memory"] = true;
     Features["atomics"] = true;
+    Features["bulk-memory"] = true;
+    Features["multimemory"] = true;
     Features["mutable-globals"] = true;
-    Features["tail-call"] = true;
+    Features["nontrapping-fptoint"] = true;
     Features["reference-types"] = true;
-    Features["multimemory"] = true;
+    Features["sign-ext"] = true;
+    Features["tail-call"] = true;
+    Features["half-precision"] = true;
     setSIMDLevel(Features, SIMD128, true);
   } else if (CPU == "generic") {
-    Features["sign-ext"] = true;
     Features["mutable-globals"] = true;
+    Features["sign-ext"] = true;
   }
 
   return TargetInfo::initFeatureMap(Features, Diags, CPU, FeaturesVec);
@@ -216,6 +218,15 @@ bool WebAssemblyTargetInfo::handleTargetFeatures(
       HasBulkMemory = false;
       continue;
     }
+    if (Feature == "+half-precision") {
+      SIMDLevel = std::max(SIMDLevel, SIMD128);
+      HasHalfPrecision = true;
+      continue;
+    }
+    if (Feature == "-half-precision") {
+      HasHalfPrecision = false;
+      continue;
+    }
     if (Feature == "+atomics") {
       HasAtomics = true;
       continue;
diff --git a/clang/lib/Basic/Targets/WebAssembly.h b/clang/lib/Basic/Targets/WebAssembly.h
index 5568aa28eaef..e4c18879182e 100644
--- a/clang/lib/Basic/Targets/WebAssembly.h
+++ b/clang/lib/Basic/Targets/WebAssembly.h
@@ -64,6 +64,7 @@ class LLVM_LIBRARY_VISIBILITY WebAssemblyTargetInfo : public TargetInfo {
   bool HasReferenceTypes = false;
   bool HasExtendedConst = false;
   bool HasMultiMemory = false;
+  bool HasHalfPrecision = false;
 
   std::string ABI;
 
diff --git a/clang/lib/CIR/CMakeLists.txt b/clang/lib/CIR/CMakeLists.txt
index e69de29bb2d1..d2ff200e0da5 100644
--- a/clang/lib/CIR/CMakeLists.txt
+++ b/clang/lib/CIR/CMakeLists.txt
@@ -0,0 +1,4 @@
+include_directories(${LLVM_MAIN_SRC_DIR}/../mlir/include)
+include_directories(${CMAKE_BINARY_DIR}/tools/mlir/include)
+
+add_subdirectory(Dialect)
diff --git a/clang/lib/CIR/Dialect/CMakeLists.txt b/clang/lib/CIR/Dialect/CMakeLists.txt
new file mode 100644
index 000000000000..f33061b2d87c
--- /dev/null
+++ b/clang/lib/CIR/Dialect/CMakeLists.txt
@@ -0,0 +1 @@
+add_subdirectory(IR)
diff --git a/clang/lib/CIR/Dialect/IR/CIRDialect.cpp b/clang/lib/CIR/Dialect/IR/CIRDialect.cpp
new file mode 100644
index 000000000000..c2829c3ff2af
--- /dev/null
+++ b/clang/lib/CIR/Dialect/IR/CIRDialect.cpp
@@ -0,0 +1,13 @@
+//===- CIRDialect.cpp - MLIR CIR ops implementation -----------------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file implements the CIR dialect and its operations.
+//
+//===----------------------------------------------------------------------===//
+
+#include <clang/CIR/Dialect/IR/CIRDialect.h>
diff --git a/clang/lib/CIR/Dialect/IR/CMakeLists.txt b/clang/lib/CIR/Dialect/IR/CMakeLists.txt
new file mode 100644
index 000000000000..0d7476b55570
--- /dev/null
+++ b/clang/lib/CIR/Dialect/IR/CMakeLists.txt
@@ -0,0 +1,3 @@
+add_clang_library(MLIRCIR
+  CIRDialect.cpp
+  )
diff --git a/clang/lib/CodeGen/CGBuiltin.cpp b/clang/lib/CodeGen/CGBuiltin.cpp
index 7e5f2edfc732..a370734e00d3 100644
--- a/clang/lib/CodeGen/CGBuiltin.cpp
+++ b/clang/lib/CodeGen/CGBuiltin.cpp
@@ -822,8 +822,9 @@ CodeGenFunction::evaluateOrEmitBuiltinObjectSize(const Expr *E, unsigned Type,
   return ConstantInt::get(ResType, ObjectSize, /*isSigned=*/true);
 }
 
-const FieldDecl *CodeGenFunction::FindFlexibleArrayMemberField(
-    ASTContext &Ctx, const RecordDecl *RD, StringRef Name, uint64_t &Offset) {
+const FieldDecl *CodeGenFunction::FindFlexibleArrayMemberFieldAndOffset(
+    ASTContext &Ctx, const RecordDecl *RD, const FieldDecl *FAMDecl,
+    uint64_t &Offset) {
   const LangOptions::StrictFlexArraysLevelKind StrictFlexArraysLevel =
       getLangOpts().getStrictFlexArraysLevel();
   uint32_t FieldNo = 0;
@@ -832,7 +833,7 @@ const FieldDecl *CodeGenFunction::FindFlexibleArrayMemberField(
     return nullptr;
 
   for (const FieldDecl *FD : RD->fields()) {
-    if ((Name.empty() || FD->getNameAsString() == Name) &&
+    if ((!FAMDecl || FD == FAMDecl) &&
         Decl::isFlexibleArrayMemberLike(
             Ctx, FD, FD->getType(), StrictFlexArraysLevel,
             /*IgnoreTemplateOrMacroSubstitution=*/true)) {
@@ -843,8 +844,8 @@ const FieldDecl *CodeGenFunction::FindFlexibleArrayMemberField(
 
     QualType Ty = FD->getType();
     if (Ty->isRecordType()) {
-      if (const FieldDecl *Field = FindFlexibleArrayMemberField(
-              Ctx, Ty->getAsRecordDecl(), Name, Offset)) {
+      if (const FieldDecl *Field = FindFlexibleArrayMemberFieldAndOffset(
+              Ctx, Ty->getAsRecordDecl(), FAMDecl, Offset)) {
         const ASTRecordLayout &Layout = Ctx.getASTRecordLayout(RD);
         Offset += Layout.getFieldOffset(FieldNo);
         return Field;
@@ -930,12 +931,14 @@ CodeGenFunction::emitFlexibleArrayMemberSize(const Expr *E, unsigned Type,
 
   // Get the flexible array member Decl.
   const RecordDecl *OuterRD = nullptr;
-  std::string FAMName;
+  const FieldDecl *FAMDecl = nullptr;
   if (const auto *ME = dyn_cast<MemberExpr>(Base)) {
     // Check if \p Base is referencing the FAM itself.
     const ValueDecl *VD = ME->getMemberDecl();
     OuterRD = VD->getDeclContext()->getOuterLexicalRecordContext();
-    FAMName = VD->getNameAsString();
+    FAMDecl = dyn_cast<FieldDecl>(VD);
+    if (!FAMDecl)
+      return nullptr;
   } else if (const auto *DRE = dyn_cast<DeclRefExpr>(Base)) {
     // Check if we're pointing to the whole struct.
     QualType Ty = DRE->getDecl()->getType();
@@ -974,9 +977,11 @@ CodeGenFunction::emitFlexibleArrayMemberSize(const Expr *E, unsigned Type,
   if (!OuterRD)
     return nullptr;
 
+  // We call FindFlexibleArrayMemberAndOffset even if FAMDecl is non-null to
+  // get its offset.
   uint64_t Offset = 0;
-  const FieldDecl *FAMDecl =
-      FindFlexibleArrayMemberField(Ctx, OuterRD, FAMName, Offset);
+  FAMDecl =
+      FindFlexibleArrayMemberFieldAndOffset(Ctx, OuterRD, FAMDecl, Offset);
   Offset = Ctx.toCharUnitsFromBits(Offset).getQuantity();
 
   if (!FAMDecl || !FAMDecl->getType()->isCountAttributedType())
@@ -3623,7 +3628,7 @@ RValue CodeGenFunction::EmitBuiltinExpr(const GlobalDecl GD, unsigned BuiltinID,
     // frexpl instead of legalizing this type in the BE.
     if (&getTarget().getLongDoubleFormat() == &llvm::APFloat::PPCDoubleDouble())
       break;
-    LLVM_FALLTHROUGH;
+    [[fallthrough]];
   }
   case Builtin::BI__builtin_frexp:
   case Builtin::BI__builtin_frexpf:
@@ -3880,9 +3885,12 @@ RValue CodeGenFunction::EmitBuiltinExpr(const GlobalDecl GD, unsigned BuiltinID,
   }
 
   case Builtin::BI__builtin_reduce_max: {
-    auto GetIntrinsicID = [](QualType QT) {
+    auto GetIntrinsicID = [this](QualType QT) {
       if (auto *VecTy = QT->getAs<VectorType>())
         QT = VecTy->getElementType();
+      else if (QT->isSizelessVectorType())
+        QT = QT->getSizelessVectorEltType(CGM.getContext());
+
       if (QT->isSignedIntegerType())
         return llvm::Intrinsic::vector_reduce_smax;
       if (QT->isUnsignedIntegerType())
@@ -3895,9 +3903,12 @@ RValue CodeGenFunction::EmitBuiltinExpr(const GlobalDecl GD, unsigned BuiltinID,
   }
 
   case Builtin::BI__builtin_reduce_min: {
-    auto GetIntrinsicID = [](QualType QT) {
+    auto GetIntrinsicID = [this](QualType QT) {
       if (auto *VecTy = QT->getAs<VectorType>())
         QT = VecTy->getElementType();
+      else if (QT->isSizelessVectorType())
+        QT = QT->getSizelessVectorEltType(CGM.getContext());
+
       if (QT->isSignedIntegerType())
         return llvm::Intrinsic::vector_reduce_smin;
       if (QT->isUnsignedIntegerType())
@@ -5361,7 +5372,7 @@ RValue CodeGenFunction::EmitBuiltinExpr(const GlobalDecl GD, unsigned BuiltinID,
     case Builtin::BI__builtin_ptrauth_auth_and_resign:
       if (Args[4]->getType()->isPointerTy())
         Args[4] = Builder.CreatePtrToInt(Args[4], IntPtrTy);
-      LLVM_FALLTHROUGH;
+      [[fallthrough]];
 
     case Builtin::BI__builtin_ptrauth_auth:
     case Builtin::BI__builtin_ptrauth_sign_unauthenticated:
@@ -18845,7 +18856,7 @@ Value *CodeGenFunction::EmitAMDGPUBuiltinExpr(unsigned BuiltinID,
     case AMDGPU::BI__builtin_amdgcn_wmma_f16_16x16x16_f16_w32_gfx12:
     case AMDGPU::BI__builtin_amdgcn_wmma_f16_16x16x16_f16_w64_gfx12:
       AppendFalseForOpselArg = true;
-      LLVM_FALLTHROUGH;
+      [[fallthrough]];
     case AMDGPU::BI__builtin_amdgcn_wmma_f16_16x16x16_f16_w32:
     case AMDGPU::BI__builtin_amdgcn_wmma_f16_16x16x16_f16_w64:
       ArgsForMatchingMatrixTypes = {2, 0}; // CD, AB
@@ -18854,7 +18865,7 @@ Value *CodeGenFunction::EmitAMDGPUBuiltinExpr(unsigned BuiltinID,
     case AMDGPU::BI__builtin_amdgcn_wmma_bf16_16x16x16_bf16_w32_gfx12:
     case AMDGPU::BI__builtin_amdgcn_wmma_bf16_16x16x16_bf16_w64_gfx12:
       AppendFalseForOpselArg = true;
-      LLVM_FALLTHROUGH;
+      [[fallthrough]];
     case AMDGPU::BI__builtin_amdgcn_wmma_bf16_16x16x16_bf16_w32:
     case AMDGPU::BI__builtin_amdgcn_wmma_bf16_16x16x16_bf16_w64:
       ArgsForMatchingMatrixTypes = {2, 0}; // CD, AB
diff --git a/clang/lib/CodeGen/CGCall.cpp b/clang/lib/CodeGen/CGCall.cpp
index 6c33cc17621f..69548902dc43 100644
--- a/clang/lib/CodeGen/CGCall.cpp
+++ b/clang/lib/CodeGen/CGCall.cpp
@@ -1585,6 +1585,11 @@ bool CodeGenModule::ReturnTypeUsesSRet(const CGFunctionInfo &FI) {
   return RI.isIndirect() || (RI.isInAlloca() && RI.getInAllocaSRet());
 }
 
+bool CodeGenModule::ReturnTypeHasInReg(const CGFunctionInfo &FI) {
+  const auto &RI = FI.getReturnInfo();
+  return RI.getInReg();
+}
+
 bool CodeGenModule::ReturnSlotInterferesWithArgs(const CGFunctionInfo &FI) {
   return ReturnTypeUsesSRet(FI) &&
          getTargetCodeGenInfo().doesReturnSlotInterfereWithArgs();
@@ -4693,11 +4698,11 @@ void CodeGenFunction::EmitCallArg(CallArgList &args, const Expr *E,
     AggValueSlot Slot = args.isUsingInAlloca()
         ? createPlaceholderSlot(*this, type) : CreateAggTemp(type, "agg.tmp");
 
-    bool DestroyedInCallee = true, NeedsEHCleanup = true;
+    bool DestroyedInCallee = true, NeedsCleanup = true;
     if (const auto *RD = type->getAsCXXRecordDecl())
       DestroyedInCallee = RD->hasNonTrivialDestructor();
     else
-      NeedsEHCleanup = needsEHCleanup(type.isDestructedType());
+      NeedsCleanup = type.isDestructedType();
 
     if (DestroyedInCallee)
       Slot.setExternallyDestructed();
@@ -4706,14 +4711,15 @@ void CodeGenFunction::EmitCallArg(CallArgList &args, const Expr *E,
     RValue RV = Slot.asRValue();
     args.add(RV, type);
 
-    if (DestroyedInCallee && NeedsEHCleanup) {
+    if (DestroyedInCallee && NeedsCleanup) {
       // Create a no-op GEP between the placeholder and the cleanup so we can
       // RAUW it successfully.  It also serves as a marker of the first
       // instruction where the cleanup is active.
-      pushFullExprCleanup<DestroyUnpassedArg>(EHCleanup, Slot.getAddress(),
-                                              type);
+      pushFullExprCleanup<DestroyUnpassedArg>(NormalAndEHCleanup,
+                                              Slot.getAddress(), type);
       // This unreachable is a temporary marker which will be removed later.
-      llvm::Instruction *IsActive = Builder.CreateUnreachable();
+      llvm::Instruction *IsActive =
+          Builder.CreateFlagLoad(llvm::Constant::getNullValue(Int8PtrTy));
       args.addArgCleanupDeactivation(EHStack.stable_begin(), IsActive);
     }
     return;
diff --git a/clang/lib/CodeGen/CGCleanup.cpp b/clang/lib/CodeGen/CGCleanup.cpp
index e6f8e6873004..469e0363b744 100644
--- a/clang/lib/CodeGen/CGCleanup.cpp
+++ b/clang/lib/CodeGen/CGCleanup.cpp
@@ -634,12 +634,19 @@ static void destroyOptimisticNormalEntry(CodeGenFunction &CGF,
 /// Pops a cleanup block.  If the block includes a normal cleanup, the
 /// current insertion point is threaded through the cleanup, as are
 /// any branch fixups on the cleanup.
-void CodeGenFunction::PopCleanupBlock(bool FallthroughIsBranchThrough) {
+void CodeGenFunction::PopCleanupBlock(bool FallthroughIsBranchThrough,
+                                      bool ForDeactivation) {
   assert(!EHStack.empty() && "cleanup stack is empty!");
   assert(isa<EHCleanupScope>(*EHStack.begin()) && "top not a cleanup!");
   EHCleanupScope &Scope = cast<EHCleanupScope>(*EHStack.begin());
   assert(Scope.getFixupDepth() <= EHStack.getNumBranchFixups());
 
+  // If we are deactivating a normal cleanup, we need to pretend that the
+  // fallthrough is unreachable. We restore this IP before returning.
+  CGBuilderTy::InsertPoint NormalDeactivateOrigIP;
+  if (ForDeactivation && (Scope.isNormalCleanup() || !getLangOpts().EHAsynch)) {
+    NormalDeactivateOrigIP = Builder.saveAndClearIP();
+  }
   // Remember activation information.
   bool IsActive = Scope.isActive();
   Address NormalActiveFlag =
@@ -667,7 +674,8 @@ void CodeGenFunction::PopCleanupBlock(bool FallthroughIsBranchThrough) {
 
   // - whether there's a fallthrough
   llvm::BasicBlock *FallthroughSource = Builder.GetInsertBlock();
-  bool HasFallthrough = (FallthroughSource != nullptr && IsActive);
+  bool HasFallthrough =
+      FallthroughSource != nullptr && (IsActive || HasExistingBranches);
 
   // Branch-through fall-throughs leave the insertion point set to the
   // end of the last cleanup, which points to the current scope.  The
@@ -692,7 +700,11 @@ void CodeGenFunction::PopCleanupBlock(bool FallthroughIsBranchThrough) {
 
   // If we have a prebranched fallthrough into an inactive normal
   // cleanup, rewrite it so that it leads to the appropriate place.
-  if (Scope.isNormalCleanup() && HasPrebranchedFallthrough && !IsActive) {
+  if (Scope.isNormalCleanup() && HasPrebranchedFallthrough &&
+      !RequiresNormalCleanup) {
+    // FIXME: Come up with a program which would need forwarding prebranched
+    // fallthrough and add tests. Otherwise delete this and assert against it.
+    assert(!IsActive);
     llvm::BasicBlock *prebranchDest;
 
     // If the prebranch is semantically branching through the next
@@ -724,6 +736,8 @@ void CodeGenFunction::PopCleanupBlock(bool FallthroughIsBranchThrough) {
     EHStack.popCleanup(); // safe because there are no fixups
     assert(EHStack.getNumBranchFixups() == 0 ||
            EHStack.hasNormalCleanups());
+    if (NormalDeactivateOrigIP.isSet())
+      Builder.restoreIP(NormalDeactivateOrigIP);
     return;
   }
 
@@ -760,11 +774,19 @@ void CodeGenFunction::PopCleanupBlock(bool FallthroughIsBranchThrough) {
   if (!RequiresNormalCleanup) {
     // Mark CPP scope end for passed-by-value Arg temp
     //   per Windows ABI which is "normally" Cleanup in callee
-    if (IsEHa && getInvokeDest() && Builder.GetInsertBlock()) {
-      if (Personality.isMSVCXXPersonality())
+    if (IsEHa && getInvokeDest()) {
+      // If we are deactivating a normal cleanup then we don't have a
+      // fallthrough. Restore original IP to emit CPP scope ends in the correct
+      // block.
+      if (NormalDeactivateOrigIP.isSet())
+        Builder.restoreIP(NormalDeactivateOrigIP);
+      if (Personality.isMSVCXXPersonality() && Builder.GetInsertBlock())
         EmitSehCppScopeEnd();
+      if (NormalDeactivateOrigIP.isSet())
+        NormalDeactivateOrigIP = Builder.saveAndClearIP();
     }
     destroyOptimisticNormalEntry(*this, Scope);
+    Scope.MarkEmitted();
     EHStack.popCleanup();
   } else {
     // If we have a fallthrough and no other need for the cleanup,
@@ -781,6 +803,7 @@ void CodeGenFunction::PopCleanupBlock(bool FallthroughIsBranchThrough) {
       }
 
       destroyOptimisticNormalEntry(*this, Scope);
+      Scope.MarkEmitted();
       EHStack.popCleanup();
 
       EmitCleanup(*this, Fn, cleanupFlags, NormalActiveFlag);
@@ -916,6 +939,7 @@ void CodeGenFunction::PopCleanupBlock(bool FallthroughIsBranchThrough) {
       }
 
       // IV.  Pop the cleanup and emit it.
+      Scope.MarkEmitted();
       EHStack.popCleanup();
       assert(EHStack.hasNormalCleanups() == HasEnclosingCleanups);
 
@@ -984,6 +1008,8 @@ void CodeGenFunction::PopCleanupBlock(bool FallthroughIsBranchThrough) {
     }
   }
 
+  if (NormalDeactivateOrigIP.isSet())
+    Builder.restoreIP(NormalDeactivateOrigIP);
   assert(EHStack.hasNormalCleanups() || EHStack.getNumBranchFixups() == 0);
 
   // Emit the EH cleanup if required.
@@ -1143,25 +1169,6 @@ void CodeGenFunction::EmitBranchThroughCleanup(JumpDest Dest) {
   Builder.ClearInsertionPoint();
 }
 
-static bool IsUsedAsNormalCleanup(EHScopeStack &EHStack,
-                                  EHScopeStack::stable_iterator C) {
-  // If we needed a normal block for any reason, that counts.
-  if (cast<EHCleanupScope>(*EHStack.find(C)).getNormalBlock())
-    return true;
-
-  // Check whether any enclosed cleanups were needed.
-  for (EHScopeStack::stable_iterator
-         I = EHStack.getInnermostNormalCleanup();
-         I != C; ) {
-    assert(C.strictlyEncloses(I));
-    EHCleanupScope &S = cast<EHCleanupScope>(*EHStack.find(I));
-    if (S.getNormalBlock()) return true;
-    I = S.getEnclosingNormalCleanup();
-  }
-
-  return false;
-}
-
 static bool IsUsedAsEHCleanup(EHScopeStack &EHStack,
                               EHScopeStack::stable_iterator cleanup) {
   // If we needed an EH block for any reason, that counts.
@@ -1210,8 +1217,7 @@ static void SetupCleanupBlockActivation(CodeGenFunction &CGF,
   // Calculate whether the cleanup was used:
 
   //   - as a normal cleanup
-  if (Scope.isNormalCleanup() &&
-      (isActivatedInConditional || IsUsedAsNormalCleanup(CGF.EHStack, C))) {
+  if (Scope.isNormalCleanup()) {
     Scope.setTestFlagInNormalCleanup();
     needFlag = true;
   }
@@ -1224,13 +1230,16 @@ static void SetupCleanupBlockActivation(CodeGenFunction &CGF,
   }
 
   // If it hasn't yet been used as either, we're done.
-  if (!needFlag) return;
+  if (!needFlag)
+    return;
 
   Address var = Scope.getActiveFlag();
   if (!var.isValid()) {
+    CodeGenFunction::AllocaTrackerRAII AllocaTracker(CGF);
     var = CGF.CreateTempAlloca(CGF.Builder.getInt1Ty(), CharUnits::One(),
                                "cleanup.isactive");
     Scope.setActiveFlag(var);
+    Scope.AddAuxAllocas(AllocaTracker.Take());
 
     assert(dominatingIP && "no existing variable and no dominating IP!");
 
@@ -1273,17 +1282,8 @@ void CodeGenFunction::DeactivateCleanupBlock(EHScopeStack::stable_iterator C,
   // to the current RunCleanupsScope.
   if (C == EHStack.stable_begin() &&
       CurrentCleanupScopeDepth.strictlyEncloses(C)) {
-    // Per comment below, checking EHAsynch is not really necessary
-    // it's there to assure zero-impact w/o EHAsynch option
-    if (!Scope.isNormalCleanup() && getLangOpts().EHAsynch) {
-      PopCleanupBlock();
-    } else {
-      // If it's a normal cleanup, we need to pretend that the
-      // fallthrough is unreachable.
-      CGBuilderTy::InsertPoint SavedIP = Builder.saveAndClearIP();
-      PopCleanupBlock();
-      Builder.restoreIP(SavedIP);
-    }
+    PopCleanupBlock(/*FallthroughIsBranchThrough=*/false,
+                    /*ForDeactivation=*/true);
     return;
   }
 
diff --git a/clang/lib/CodeGen/CGCleanup.h b/clang/lib/CodeGen/CGCleanup.h
index 03e4a29d7b3d..c73c97146abc 100644
--- a/clang/lib/CodeGen/CGCleanup.h
+++ b/clang/lib/CodeGen/CGCleanup.h
@@ -16,8 +16,11 @@
 #include "EHScopeStack.h"
 
 #include "Address.h"
+#include "llvm/ADT/STLExtras.h"
+#include "llvm/ADT/SetVector.h"
 #include "llvm/ADT/SmallPtrSet.h"
 #include "llvm/ADT/SmallVector.h"
+#include "llvm/IR/Instruction.h"
 
 namespace llvm {
 class BasicBlock;
@@ -266,6 +269,51 @@ class alignas(8) EHCleanupScope : public EHScope {
   };
   mutable struct ExtInfo *ExtInfo;
 
+  /// Erases auxillary allocas and their usages for an unused cleanup.
+  /// Cleanups should mark these allocas as 'used' if the cleanup is
+  /// emitted, otherwise these instructions would be erased.
+  struct AuxillaryAllocas {
+    SmallVector<llvm::Instruction *, 1> AuxAllocas;
+    bool used = false;
+
+    // Records a potentially unused instruction to be erased later.
+    void Add(llvm::AllocaInst *Alloca) { AuxAllocas.push_back(Alloca); }
+
+    // Mark all recorded instructions as used. These will not be erased later.
+    void MarkUsed() {
+      used = true;
+      AuxAllocas.clear();
+    }
+
+    ~AuxillaryAllocas() {
+      if (used)
+        return;
+      llvm::SetVector<llvm::Instruction *> Uses;
+      for (auto *Inst : llvm::reverse(AuxAllocas))
+        CollectUses(Inst, Uses);
+      // Delete uses in the reverse order of insertion.
+      for (auto *I : llvm::reverse(Uses))
+        I->eraseFromParent();
+    }
+
+  private:
+    void CollectUses(llvm::Instruction *I,
+                     llvm::SetVector<llvm::Instruction *> &Uses) {
+      if (!I || !Uses.insert(I))
+        return;
+      for (auto *User : I->users())
+        CollectUses(cast<llvm::Instruction>(User), Uses);
+    }
+  };
+  mutable struct AuxillaryAllocas *AuxAllocas;
+
+  AuxillaryAllocas &getAuxillaryAllocas() {
+    if (!AuxAllocas) {
+      AuxAllocas = new struct AuxillaryAllocas();
+    }
+    return *AuxAllocas;
+  }
+
   /// The number of fixups required by enclosing scopes (not including
   /// this one).  If this is the top cleanup scope, all the fixups
   /// from this index onwards belong to this scope.
@@ -298,7 +346,7 @@ public:
                  EHScopeStack::stable_iterator enclosingEH)
       : EHScope(EHScope::Cleanup, enclosingEH),
         EnclosingNormal(enclosingNormal), NormalBlock(nullptr),
-        ActiveFlag(Address::invalid()), ExtInfo(nullptr),
+        ActiveFlag(Address::invalid()), ExtInfo(nullptr), AuxAllocas(nullptr),
         FixupDepth(fixupDepth) {
     CleanupBits.IsNormalCleanup = isNormal;
     CleanupBits.IsEHCleanup = isEH;
@@ -312,8 +360,15 @@ public:
   }
 
   void Destroy() {
+    if (AuxAllocas)
+      delete AuxAllocas;
     delete ExtInfo;
   }
+  void AddAuxAllocas(llvm::SmallVector<llvm::AllocaInst *> Allocas) {
+    for (auto *Alloca : Allocas)
+      getAuxillaryAllocas().Add(Alloca);
+  }
+  void MarkEmitted() { getAuxillaryAllocas().MarkUsed(); }
   // Objects of EHCleanupScope are not destructed. Use Destroy().
   ~EHCleanupScope() = delete;
 
diff --git a/clang/lib/CodeGen/CGCoroutine.cpp b/clang/lib/CodeGen/CGCoroutine.cpp
index 93ca711f716f..567e85a02dc6 100644
--- a/clang/lib/CodeGen/CGCoroutine.cpp
+++ b/clang/lib/CodeGen/CGCoroutine.cpp
@@ -413,10 +413,8 @@ llvm::Function *
 CodeGenFunction::generateAwaitSuspendWrapper(Twine const &CoroName,
                                              Twine const &SuspendPointName,
                                              CoroutineSuspendExpr const &S) {
-  std::string FuncName = "__await_suspend_wrapper_";
-  FuncName += CoroName.str();
-  FuncName += '_';
-  FuncName += SuspendPointName.str();
+  std::string FuncName =
+      (CoroName + ".__await_suspend_wrapper__" + SuspendPointName).str();
 
   ASTContext &C = getContext();
 
diff --git a/clang/lib/CodeGen/CGDebugInfo.cpp b/clang/lib/CodeGen/CGDebugInfo.cpp
index 539ded5cca5e..787db3504874 100644
--- a/clang/lib/CodeGen/CGDebugInfo.cpp
+++ b/clang/lib/CodeGen/CGDebugInfo.cpp
@@ -1372,7 +1372,26 @@ llvm::DIType *CGDebugInfo::CreateType(const TemplateSpecializationType *Ty,
 
   SourceLocation Loc = AliasDecl->getLocation();
 
-  if (CGM.getCodeGenOpts().DebugTemplateAlias) {
+  if (CGM.getCodeGenOpts().DebugTemplateAlias &&
+      // The TemplateSpecializationType doesn't contain any instantiation
+      // information; dependent template arguments can't be resolved. For now,
+      // fall back to DW_TAG_typedefs for template aliases that are
+      // instantiation dependent, e.g.:
+      // ```
+      // template <int>
+      // using A = int;
+      //
+      // template<int I>
+      // struct S {
+      //   using AA = A<I>; // Instantiation dependent.
+      //   AA aa;
+      // };
+      //
+      // S<0> s;
+      // ```
+      // S::AA's underlying type A<I> is dependent on I so will be emitted as a
+      // DW_TAG_typedef.
+      !Ty->isInstantiationDependentType()) {
     auto ArgVector = ::GetTemplateArgs(TD, Ty);
     TemplateArgs Args = {TD->getTemplateParameters(), ArgVector};
 
diff --git a/clang/lib/CodeGen/CGDecl.cpp b/clang/lib/CodeGen/CGDecl.cpp
index ce6d6d895607..9cc67cdbe424 100644
--- a/clang/lib/CodeGen/CGDecl.cpp
+++ b/clang/lib/CodeGen/CGDecl.cpp
@@ -19,6 +19,7 @@
 #include "CodeGenFunction.h"
 #include "CodeGenModule.h"
 #include "ConstantEmitter.h"
+#include "EHScopeStack.h"
 #include "PatternInit.h"
 #include "TargetInfo.h"
 #include "clang/AST/ASTContext.h"
@@ -35,6 +36,7 @@
 #include "llvm/Analysis/ValueTracking.h"
 #include "llvm/IR/DataLayout.h"
 #include "llvm/IR/GlobalVariable.h"
+#include "llvm/IR/Instructions.h"
 #include "llvm/IR/Intrinsics.h"
 #include "llvm/IR/Type.h"
 #include <optional>
@@ -2201,6 +2203,27 @@ void CodeGenFunction::pushDestroy(CleanupKind cleanupKind, Address addr,
                                      destroyer, useEHCleanupForArray);
 }
 
+// Pushes a destroy and defers its deactivation until its
+// CleanupDeactivationScope is exited.
+void CodeGenFunction::pushDestroyAndDeferDeactivation(
+    QualType::DestructionKind dtorKind, Address addr, QualType type) {
+  assert(dtorKind && "cannot push destructor for trivial type");
+
+  CleanupKind cleanupKind = getCleanupKind(dtorKind);
+  pushDestroyAndDeferDeactivation(
+      cleanupKind, addr, type, getDestroyer(dtorKind), cleanupKind & EHCleanup);
+}
+
+void CodeGenFunction::pushDestroyAndDeferDeactivation(
+    CleanupKind cleanupKind, Address addr, QualType type, Destroyer *destroyer,
+    bool useEHCleanupForArray) {
+  llvm::Instruction *DominatingIP =
+      Builder.CreateFlagLoad(llvm::Constant::getNullValue(Int8PtrTy));
+  pushDestroy(cleanupKind, addr, type, destroyer, useEHCleanupForArray);
+  DeferredDeactivationCleanupStack.push_back(
+      {EHStack.stable_begin(), DominatingIP});
+}
+
 void CodeGenFunction::pushStackRestore(CleanupKind Kind, Address SPMem) {
   EHStack.pushCleanup<CallStackRestore>(Kind, SPMem);
 }
@@ -2217,39 +2240,48 @@ void CodeGenFunction::pushLifetimeExtendedDestroy(CleanupKind cleanupKind,
   // If we're not in a conditional branch, we don't need to bother generating a
   // conditional cleanup.
   if (!isInConditionalBranch()) {
-    // Push an EH-only cleanup for the object now.
     // FIXME: When popping normal cleanups, we need to keep this EH cleanup
     // around in case a temporary's destructor throws an exception.
-    if (cleanupKind & EHCleanup)
-      EHStack.pushCleanup<DestroyObject>(
-          static_cast<CleanupKind>(cleanupKind & ~NormalCleanup), addr, type,
-          destroyer, useEHCleanupForArray);
 
+    // Add the cleanup to the EHStack. After the full-expr, this would be
+    // deactivated before being popped from the stack.
+    pushDestroyAndDeferDeactivation(cleanupKind, addr, type, destroyer,
+                                    useEHCleanupForArray);
+
+    // Since this is lifetime-extended, push it once again to the EHStack after
+    // the full expression.
     return pushCleanupAfterFullExprWithActiveFlag<DestroyObject>(
-        cleanupKind, Address::invalid(), addr, type, destroyer, useEHCleanupForArray);
+        cleanupKind, Address::invalid(), addr, type, destroyer,
+        useEHCleanupForArray);
   }
 
   // Otherwise, we should only destroy the object if it's been initialized.
-  // Re-use the active flag and saved address across both the EH and end of
-  // scope cleanups.
 
-  using SavedType = typename DominatingValue<Address>::saved_type;
   using ConditionalCleanupType =
       EHScopeStack::ConditionalCleanup<DestroyObject, Address, QualType,
                                        Destroyer *, bool>;
-
-  Address ActiveFlag = createCleanupActiveFlag();
-  SavedType SavedAddr = saveValueInCond(addr);
-
-  if (cleanupKind & EHCleanup) {
-    EHStack.pushCleanup<ConditionalCleanupType>(
-        static_cast<CleanupKind>(cleanupKind & ~NormalCleanup), SavedAddr, type,
-        destroyer, useEHCleanupForArray);
-    initFullExprCleanupWithFlag(ActiveFlag);
-  }
-
+  DominatingValue<Address>::saved_type SavedAddr = saveValueInCond(addr);
+
+  // Remember to emit cleanup if we branch-out before end of full-expression
+  // (eg: through stmt-expr or coro suspensions).
+  AllocaTrackerRAII DeactivationAllocas(*this);
+  Address ActiveFlagForDeactivation = createCleanupActiveFlag();
+
+  pushCleanupAndDeferDeactivation<ConditionalCleanupType>(
+      cleanupKind, SavedAddr, type, destroyer, useEHCleanupForArray);
+  initFullExprCleanupWithFlag(ActiveFlagForDeactivation);
+  EHCleanupScope &cleanup = cast<EHCleanupScope>(*EHStack.begin());
+  // Erase the active flag if the cleanup was not emitted.
+  cleanup.AddAuxAllocas(std::move(DeactivationAllocas).Take());
+
+  // Since this is lifetime-extended, push it once again to the EHStack after
+  // the full expression.
+  // The previous active flag would always be 'false' due to forced deferred
+  // deactivation. Use a separate flag for lifetime-extension to correctly
+  // remember if this branch was taken and the object was initialized.
+  Address ActiveFlagForLifetimeExt = createCleanupActiveFlag();
   pushCleanupAfterFullExprWithActiveFlag<ConditionalCleanupType>(
-      cleanupKind, ActiveFlag, SavedAddr, type, destroyer,
+      cleanupKind, ActiveFlagForLifetimeExt, SavedAddr, type, destroyer,
       useEHCleanupForArray);
 }
 
@@ -2442,9 +2474,9 @@ namespace {
   };
 } // end anonymous namespace
 
-/// pushIrregularPartialArrayCleanup - Push an EH cleanup to destroy
-/// already-constructed elements of the given array.  The cleanup
-/// may be popped with DeactivateCleanupBlock or PopCleanupBlock.
+/// pushIrregularPartialArrayCleanup - Push a NormalAndEHCleanup to
+/// destroy already-constructed elements of the given array.  The cleanup may be
+/// popped with DeactivateCleanupBlock or PopCleanupBlock.
 ///
 /// \param elementType - the immediate element type of the array;
 ///   possibly still an array type
@@ -2453,10 +2485,9 @@ void CodeGenFunction::pushIrregularPartialArrayCleanup(llvm::Value *arrayBegin,
                                                        QualType elementType,
                                                        CharUnits elementAlign,
                                                        Destroyer *destroyer) {
-  pushFullExprCleanup<IrregularPartialArrayDestroy>(EHCleanup,
-                                                    arrayBegin, arrayEndPointer,
-                                                    elementType, elementAlign,
-                                                    destroyer);
+  pushFullExprCleanup<IrregularPartialArrayDestroy>(
+      NormalAndEHCleanup, arrayBegin, arrayEndPointer, elementType,
+      elementAlign, destroyer);
 }
 
 /// pushRegularPartialArrayCleanup - Push an EH cleanup to destroy
diff --git a/clang/lib/CodeGen/CGExpr.cpp b/clang/lib/CodeGen/CGExpr.cpp
index 931cb391342e..d96c7bb1e568 100644
--- a/clang/lib/CodeGen/CGExpr.cpp
+++ b/clang/lib/CodeGen/CGExpr.cpp
@@ -115,10 +115,16 @@ RawAddress CodeGenFunction::CreateTempAlloca(llvm::Type *Ty, CharUnits Align,
 llvm::AllocaInst *CodeGenFunction::CreateTempAlloca(llvm::Type *Ty,
                                                     const Twine &Name,
                                                     llvm::Value *ArraySize) {
+  llvm::AllocaInst *Alloca;
   if (ArraySize)
-    return Builder.CreateAlloca(Ty, ArraySize, Name);
-  return new llvm::AllocaInst(Ty, CGM.getDataLayout().getAllocaAddrSpace(),
-                              ArraySize, Name, AllocaInsertPt);
+    Alloca = Builder.CreateAlloca(Ty, ArraySize, Name);
+  else
+    Alloca = new llvm::AllocaInst(Ty, CGM.getDataLayout().getAllocaAddrSpace(),
+                                  ArraySize, Name, AllocaInsertPt);
+  if (Allocas) {
+    Allocas->Add(Alloca);
+  }
+  return Alloca;
 }
 
 /// CreateDefaultAlignTempAlloca - This creates an alloca with the
@@ -1621,8 +1627,8 @@ LValue CodeGenFunction::EmitLValueHelper(const Expr *E,
     return EmitArraySubscriptExpr(cast<ArraySubscriptExpr>(E));
   case Expr::MatrixSubscriptExprClass:
     return EmitMatrixSubscriptExpr(cast<MatrixSubscriptExpr>(E));
-  case Expr::OMPArraySectionExprClass:
-    return EmitOMPArraySectionExpr(cast<OMPArraySectionExpr>(E));
+  case Expr::ArraySectionExprClass:
+    return EmitArraySectionExpr(cast<ArraySectionExpr>(E));
   case Expr::ExtVectorElementExprClass:
     return EmitExtVectorElementExpr(cast<ExtVectorElementExpr>(E));
   case Expr::CXXThisExprClass:
@@ -4363,8 +4369,8 @@ static Address emitOMPArraySectionBase(CodeGenFunction &CGF, const Expr *Base,
                                        QualType BaseTy, QualType ElTy,
                                        bool IsLowerBound) {
   LValue BaseLVal;
-  if (auto *ASE = dyn_cast<OMPArraySectionExpr>(Base->IgnoreParenImpCasts())) {
-    BaseLVal = CGF.EmitOMPArraySectionExpr(ASE, IsLowerBound);
+  if (auto *ASE = dyn_cast<ArraySectionExpr>(Base->IgnoreParenImpCasts())) {
+    BaseLVal = CGF.EmitArraySectionExpr(ASE, IsLowerBound);
     if (BaseTy->isArrayType()) {
       Address Addr = BaseLVal.getAddress(CGF);
       BaseInfo = BaseLVal.getBaseInfo();
@@ -4396,9 +4402,13 @@ static Address emitOMPArraySectionBase(CodeGenFunction &CGF, const Expr *Base,
   return CGF.EmitPointerWithAlignment(Base, &BaseInfo, &TBAAInfo);
 }
 
-LValue CodeGenFunction::EmitOMPArraySectionExpr(const OMPArraySectionExpr *E,
-                                                bool IsLowerBound) {
-  QualType BaseTy = OMPArraySectionExpr::getBaseOriginalType(E->getBase());
+LValue CodeGenFunction::EmitArraySectionExpr(const ArraySectionExpr *E,
+                                             bool IsLowerBound) {
+
+  assert(!E->isOpenACCArraySection() &&
+         "OpenACC Array section codegen not implemented");
+
+  QualType BaseTy = ArraySectionExpr::getBaseOriginalType(E->getBase());
   QualType ResultExprTy;
   if (auto *AT = getContext().getAsArrayType(BaseTy))
     ResultExprTy = AT->getElementType();
diff --git a/clang/lib/CodeGen/CGExprAgg.cpp b/clang/lib/CodeGen/CGExprAgg.cpp
index 355fec42be44..44d476976a55 100644
--- a/clang/lib/CodeGen/CGExprAgg.cpp
+++ b/clang/lib/CodeGen/CGExprAgg.cpp
@@ -15,6 +15,7 @@
 #include "CodeGenFunction.h"
 #include "CodeGenModule.h"
 #include "ConstantEmitter.h"
+#include "EHScopeStack.h"
 #include "TargetInfo.h"
 #include "clang/AST/ASTContext.h"
 #include "clang/AST/Attr.h"
@@ -24,6 +25,7 @@
 #include "llvm/IR/Constants.h"
 #include "llvm/IR/Function.h"
 #include "llvm/IR/GlobalVariable.h"
+#include "llvm/IR/Instruction.h"
 #include "llvm/IR/IntrinsicInst.h"
 #include "llvm/IR/Intrinsics.h"
 using namespace clang;
@@ -558,24 +560,27 @@ void AggExprEmitter::EmitArrayInit(Address DestPtr, llvm::ArrayType *AType,
   // For that, we'll need an EH cleanup.
   QualType::DestructionKind dtorKind = elementType.isDestructedType();
   Address endOfInit = Address::invalid();
-  EHScopeStack::stable_iterator cleanup;
-  llvm::Instruction *cleanupDominator = nullptr;
-  if (CGF.needsEHCleanup(dtorKind)) {
+  CodeGenFunction::CleanupDeactivationScope deactivation(CGF);
+
+  if (dtorKind) {
+    CodeGenFunction::AllocaTrackerRAII allocaTracker(CGF);
     // In principle we could tell the cleanup where we are more
     // directly, but the control flow can get so varied here that it
     // would actually be quite complex.  Therefore we go through an
     // alloca.
+    llvm::Instruction *dominatingIP =
+        Builder.CreateFlagLoad(llvm::ConstantInt::getNullValue(CGF.Int8PtrTy));
     endOfInit = CGF.CreateTempAlloca(begin->getType(), CGF.getPointerAlign(),
                                      "arrayinit.endOfInit");
-    cleanupDominator = Builder.CreateStore(begin, endOfInit);
+    Builder.CreateStore(begin, endOfInit);
     CGF.pushIrregularPartialArrayCleanup(begin, endOfInit, elementType,
                                          elementAlign,
                                          CGF.getDestroyer(dtorKind));
-    cleanup = CGF.EHStack.stable_begin();
+    cast<EHCleanupScope>(*CGF.EHStack.find(CGF.EHStack.stable_begin()))
+        .AddAuxAllocas(allocaTracker.Take());
 
-  // Otherwise, remember that we didn't need a cleanup.
-  } else {
-    dtorKind = QualType::DK_none;
+    CGF.DeferredDeactivationCleanupStack.push_back(
+        {CGF.EHStack.stable_begin(), dominatingIP});
   }
 
   llvm::Value *one = llvm::ConstantInt::get(CGF.SizeTy, 1);
@@ -671,9 +676,6 @@ void AggExprEmitter::EmitArrayInit(Address DestPtr, llvm::ArrayType *AType,
 
     CGF.EmitBlock(endBB);
   }
-
-  // Leave the partial-array cleanup if we entered one.
-  if (dtorKind) CGF.DeactivateCleanupBlock(cleanup, cleanupDominator);
 }
 
 //===----------------------------------------------------------------------===//
@@ -1374,9 +1376,8 @@ AggExprEmitter::VisitLambdaExpr(LambdaExpr *E) {
   LValue SlotLV = CGF.MakeAddrLValue(Slot.getAddress(), E->getType());
 
   // We'll need to enter cleanup scopes in case any of the element
-  // initializers throws an exception.
-  SmallVector<EHScopeStack::stable_iterator, 16> Cleanups;
-  llvm::Instruction *CleanupDominator = nullptr;
+  // initializers throws an exception or contains branch out of the expressions.
+  CodeGenFunction::CleanupDeactivationScope scope(CGF);
 
   CXXRecordDecl::field_iterator CurField = E->getLambdaClass()->field_begin();
   for (LambdaExpr::const_capture_init_iterator i = E->capture_init_begin(),
@@ -1395,28 +1396,12 @@ AggExprEmitter::VisitLambdaExpr(LambdaExpr *E) {
     if (QualType::DestructionKind DtorKind =
             CurField->getType().isDestructedType()) {
       assert(LV.isSimple());
-      if (CGF.needsEHCleanup(DtorKind)) {
-        if (!CleanupDominator)
-          CleanupDominator = CGF.Builder.CreateAlignedLoad(
-              CGF.Int8Ty,
-              llvm::Constant::getNullValue(CGF.Int8PtrTy),
-              CharUnits::One()); // placeholder
-
-        CGF.pushDestroy(EHCleanup, LV.getAddress(CGF), CurField->getType(),
-                        CGF.getDestroyer(DtorKind), false);
-        Cleanups.push_back(CGF.EHStack.stable_begin());
-      }
+      if (DtorKind)
+        CGF.pushDestroyAndDeferDeactivation(
+            NormalAndEHCleanup, LV.getAddress(CGF), CurField->getType(),
+            CGF.getDestroyer(DtorKind), false);
     }
   }
-
-  // Deactivate all the partial cleanups in reverse order, which
-  // generally means popping them.
-  for (unsigned i = Cleanups.size(); i != 0; --i)
-    CGF.DeactivateCleanupBlock(Cleanups[i-1], CleanupDominator);
-
-  // Destroy the placeholder if we made one.
-  if (CleanupDominator)
-    CleanupDominator->eraseFromParent();
 }
 
 void AggExprEmitter::VisitExprWithCleanups(ExprWithCleanups *E) {
@@ -1705,14 +1690,7 @@ void AggExprEmitter::VisitCXXParenListOrInitListExpr(
   // We'll need to enter cleanup scopes in case any of the element
   // initializers throws an exception.
   SmallVector<EHScopeStack::stable_iterator, 16> cleanups;
-  llvm::Instruction *cleanupDominator = nullptr;
-  auto addCleanup = [&](const EHScopeStack::stable_iterator &cleanup) {
-    cleanups.push_back(cleanup);
-    if (!cleanupDominator) // create placeholder once needed
-      cleanupDominator = CGF.Builder.CreateAlignedLoad(
-          CGF.Int8Ty, llvm::Constant::getNullValue(CGF.Int8PtrTy),
-          CharUnits::One());
-  };
+  CodeGenFunction::CleanupDeactivationScope DeactivateCleanups(CGF);
 
   unsigned curInitIndex = 0;
 
@@ -1735,10 +1713,8 @@ void AggExprEmitter::VisitCXXParenListOrInitListExpr(
       CGF.EmitAggExpr(InitExprs[curInitIndex++], AggSlot);
 
       if (QualType::DestructionKind dtorKind =
-              Base.getType().isDestructedType()) {
-        CGF.pushDestroy(dtorKind, V, Base.getType());
-        addCleanup(CGF.EHStack.stable_begin());
-      }
+              Base.getType().isDestructedType())
+        CGF.pushDestroyAndDeferDeactivation(dtorKind, V, Base.getType());
     }
   }
 
@@ -1815,10 +1791,10 @@ void AggExprEmitter::VisitCXXParenListOrInitListExpr(
     if (QualType::DestructionKind dtorKind
           = field->getType().isDestructedType()) {
       assert(LV.isSimple());
-      if (CGF.needsEHCleanup(dtorKind)) {
-        CGF.pushDestroy(EHCleanup, LV.getAddress(CGF), field->getType(),
-                        CGF.getDestroyer(dtorKind), false);
-        addCleanup(CGF.EHStack.stable_begin());
+      if (dtorKind) {
+        CGF.pushDestroyAndDeferDeactivation(
+            NormalAndEHCleanup, LV.getAddress(CGF), field->getType(),
+            CGF.getDestroyer(dtorKind), false);
         pushedCleanup = true;
       }
     }
@@ -1831,17 +1807,6 @@ void AggExprEmitter::VisitCXXParenListOrInitListExpr(
         if (GEP->use_empty())
           GEP->eraseFromParent();
   }
-
-  // Deactivate all the partial cleanups in reverse order, which
-  // generally means popping them.
-  assert((cleanupDominator || cleanups.empty()) &&
-         "Missing cleanupDominator before deactivating cleanup blocks");
-  for (unsigned i = cleanups.size(); i != 0; --i)
-    CGF.DeactivateCleanupBlock(cleanups[i-1], cleanupDominator);
-
-  // Destroy the placeholder if we made one.
-  if (cleanupDominator)
-    cleanupDominator->eraseFromParent();
 }
 
 void AggExprEmitter::VisitArrayInitLoopExpr(const ArrayInitLoopExpr *E,
diff --git a/clang/lib/CodeGen/CGExprCXX.cpp b/clang/lib/CodeGen/CGExprCXX.cpp
index 673ccef84d67..c18c36d3f3f3 100644
--- a/clang/lib/CodeGen/CGExprCXX.cpp
+++ b/clang/lib/CodeGen/CGExprCXX.cpp
@@ -1008,8 +1008,8 @@ void CodeGenFunction::EmitNewArrayInitializer(
   const Expr *Init = E->getInitializer();
   Address EndOfInit = Address::invalid();
   QualType::DestructionKind DtorKind = ElementType.isDestructedType();
-  EHScopeStack::stable_iterator Cleanup;
-  llvm::Instruction *CleanupDominator = nullptr;
+  CleanupDeactivationScope deactivation(*this);
+  bool pushedCleanup = false;
 
   CharUnits ElementSize = getContext().getTypeSizeInChars(ElementType);
   CharUnits ElementAlign =
@@ -1105,19 +1105,24 @@ void CodeGenFunction::EmitNewArrayInitializer(
     }
 
     // Enter a partial-destruction Cleanup if necessary.
-    if (needsEHCleanup(DtorKind)) {
+    if (DtorKind) {
+      AllocaTrackerRAII AllocaTracker(*this);
       // In principle we could tell the Cleanup where we are more
       // directly, but the control flow can get so varied here that it
       // would actually be quite complex.  Therefore we go through an
       // alloca.
+      llvm::Instruction *DominatingIP =
+          Builder.CreateFlagLoad(llvm::ConstantInt::getNullValue(Int8PtrTy));
       EndOfInit = CreateTempAlloca(BeginPtr.getType(), getPointerAlign(),
                                    "array.init.end");
-      CleanupDominator =
-          Builder.CreateStore(BeginPtr.emitRawPointer(*this), EndOfInit);
       pushIrregularPartialArrayCleanup(BeginPtr.emitRawPointer(*this),
                                        EndOfInit, ElementType, ElementAlign,
                                        getDestroyer(DtorKind));
-      Cleanup = EHStack.stable_begin();
+      cast<EHCleanupScope>(*EHStack.find(EHStack.stable_begin()))
+          .AddAuxAllocas(AllocaTracker.Take());
+      DeferredDeactivationCleanupStack.push_back(
+          {EHStack.stable_begin(), DominatingIP});
+      pushedCleanup = true;
     }
 
     CharUnits StartAlign = CurPtr.getAlignment();
@@ -1164,9 +1169,6 @@ void CodeGenFunction::EmitNewArrayInitializer(
   // initialization.
   llvm::ConstantInt *ConstNum = dyn_cast<llvm::ConstantInt>(NumElements);
   if (ConstNum && ConstNum->getZExtValue() <= InitListElements) {
-    // If there was a Cleanup, deactivate it.
-    if (CleanupDominator)
-      DeactivateCleanupBlock(Cleanup, CleanupDominator);
     return;
   }
 
@@ -1281,13 +1283,14 @@ void CodeGenFunction::EmitNewArrayInitializer(
     Builder.CreateStore(CurPtr.emitRawPointer(*this), EndOfInit);
 
   // Enter a partial-destruction Cleanup if necessary.
-  if (!CleanupDominator && needsEHCleanup(DtorKind)) {
-    llvm::Value *BeginPtrRaw = BeginPtr.emitRawPointer(*this);
-    llvm::Value *CurPtrRaw = CurPtr.emitRawPointer(*this);
-    pushRegularPartialArrayCleanup(BeginPtrRaw, CurPtrRaw, ElementType,
+  if (!pushedCleanup && needsEHCleanup(DtorKind)) {
+    llvm::Instruction *DominatingIP =
+        Builder.CreateFlagLoad(llvm::ConstantInt::getNullValue(Int8PtrTy));
+    pushRegularPartialArrayCleanup(BeginPtr.emitRawPointer(*this),
+                                   CurPtr.emitRawPointer(*this), ElementType,
                                    ElementAlign, getDestroyer(DtorKind));
-    Cleanup = EHStack.stable_begin();
-    CleanupDominator = Builder.CreateUnreachable();
+    DeferredDeactivationCleanupStack.push_back(
+        {EHStack.stable_begin(), DominatingIP});
   }
 
   // Emit the initializer into this element.
@@ -1295,10 +1298,7 @@ void CodeGenFunction::EmitNewArrayInitializer(
                           AggValueSlot::DoesNotOverlap);
 
   // Leave the Cleanup if we entered one.
-  if (CleanupDominator) {
-    DeactivateCleanupBlock(Cleanup, CleanupDominator);
-    CleanupDominator->eraseFromParent();
-  }
+  deactivation.ForceDeactivate();
 
   // Advance to the next element by adjusting the pointer type as necessary.
   llvm::Value *NextPtr = Builder.CreateConstInBoundsGEP1_32(
diff --git a/clang/lib/CodeGen/CGExprConstant.cpp b/clang/lib/CodeGen/CGExprConstant.cpp
index c924660c5a91..94962091116a 100644
--- a/clang/lib/CodeGen/CGExprConstant.cpp
+++ b/clang/lib/CodeGen/CGExprConstant.cpp
@@ -564,12 +564,13 @@ class ConstStructBuilder {
 
 public:
   static llvm::Constant *BuildStruct(ConstantEmitter &Emitter,
-                                     InitListExpr *ILE, QualType StructTy);
+                                     const InitListExpr *ILE,
+                                     QualType StructTy);
   static llvm::Constant *BuildStruct(ConstantEmitter &Emitter,
                                      const APValue &Value, QualType ValTy);
   static bool UpdateStruct(ConstantEmitter &Emitter,
                            ConstantAggregateBuilder &Const, CharUnits Offset,
-                           InitListExpr *Updater);
+                           const InitListExpr *Updater);
 
 private:
   ConstStructBuilder(ConstantEmitter &Emitter,
@@ -586,7 +587,7 @@ private:
   bool AppendBitField(const FieldDecl *Field, uint64_t FieldOffset,
                       llvm::ConstantInt *InitExpr, bool AllowOverwrite = false);
 
-  bool Build(InitListExpr *ILE, bool AllowOverwrite);
+  bool Build(const InitListExpr *ILE, bool AllowOverwrite);
   bool Build(const APValue &Val, const RecordDecl *RD, bool IsPrimaryBase,
              const CXXRecordDecl *VTableClass, CharUnits BaseOffset);
   llvm::Constant *Finalize(QualType Ty);
@@ -635,7 +636,7 @@ bool ConstStructBuilder::AppendBitField(
 static bool EmitDesignatedInitUpdater(ConstantEmitter &Emitter,
                                       ConstantAggregateBuilder &Const,
                                       CharUnits Offset, QualType Type,
-                                      InitListExpr *Updater) {
+                                      const InitListExpr *Updater) {
   if (Type->isRecordType())
     return ConstStructBuilder::UpdateStruct(Emitter, Const, Offset, Updater);
 
@@ -647,7 +648,7 @@ static bool EmitDesignatedInitUpdater(ConstantEmitter &Emitter,
   llvm::Type *ElemTy = Emitter.CGM.getTypes().ConvertTypeForMem(ElemType);
 
   llvm::Constant *FillC = nullptr;
-  if (Expr *Filler = Updater->getArrayFiller()) {
+  if (const Expr *Filler = Updater->getArrayFiller()) {
     if (!isa<NoInitExpr>(Filler)) {
       FillC = Emitter.tryEmitAbstractForMemory(Filler, ElemType);
       if (!FillC)
@@ -658,7 +659,7 @@ static bool EmitDesignatedInitUpdater(ConstantEmitter &Emitter,
   unsigned NumElementsToUpdate =
       FillC ? CAT->getZExtSize() : Updater->getNumInits();
   for (unsigned I = 0; I != NumElementsToUpdate; ++I, Offset += ElemSize) {
-    Expr *Init = nullptr;
+    const Expr *Init = nullptr;
     if (I < Updater->getNumInits())
       Init = Updater->getInit(I);
 
@@ -667,7 +668,7 @@ static bool EmitDesignatedInitUpdater(ConstantEmitter &Emitter,
         return false;
     } else if (!Init || isa<NoInitExpr>(Init)) {
       continue;
-    } else if (InitListExpr *ChildILE = dyn_cast<InitListExpr>(Init)) {
+    } else if (const auto *ChildILE = dyn_cast<InitListExpr>(Init)) {
       if (!EmitDesignatedInitUpdater(Emitter, Const, Offset, ElemType,
                                      ChildILE))
         return false;
@@ -683,7 +684,7 @@ static bool EmitDesignatedInitUpdater(ConstantEmitter &Emitter,
   return true;
 }
 
-bool ConstStructBuilder::Build(InitListExpr *ILE, bool AllowOverwrite) {
+bool ConstStructBuilder::Build(const InitListExpr *ILE, bool AllowOverwrite) {
   RecordDecl *RD = ILE->getType()->castAs<RecordType>()->getDecl();
   const ASTRecordLayout &Layout = CGM.getContext().getASTRecordLayout(RD);
 
@@ -711,7 +712,7 @@ bool ConstStructBuilder::Build(InitListExpr *ILE, bool AllowOverwrite) {
 
     // Get the initializer.  A struct can include fields without initializers,
     // we just use explicit null values for them.
-    Expr *Init = nullptr;
+    const Expr *Init = nullptr;
     if (ElementNo < ILE->getNumInits())
       Init = ILE->getInit(ElementNo++);
     if (Init && isa<NoInitExpr>(Init))
@@ -879,7 +880,7 @@ llvm::Constant *ConstStructBuilder::Finalize(QualType Type) {
 }
 
 llvm::Constant *ConstStructBuilder::BuildStruct(ConstantEmitter &Emitter,
-                                                InitListExpr *ILE,
+                                                const InitListExpr *ILE,
                                                 QualType ValTy) {
   ConstantAggregateBuilder Const(Emitter.CGM);
   ConstStructBuilder Builder(Emitter, Const, CharUnits::Zero());
@@ -906,7 +907,8 @@ llvm::Constant *ConstStructBuilder::BuildStruct(ConstantEmitter &Emitter,
 
 bool ConstStructBuilder::UpdateStruct(ConstantEmitter &Emitter,
                                       ConstantAggregateBuilder &Const,
-                                      CharUnits Offset, InitListExpr *Updater) {
+                                      CharUnits Offset,
+                                      const InitListExpr *Updater) {
   return ConstStructBuilder(Emitter, Const, Offset)
       .Build(Updater, /*AllowOverwrite*/ true);
 }
@@ -1013,8 +1015,8 @@ EmitArrayConstant(CodeGenModule &CGM, llvm::ArrayType *DesiredType,
 //
 // Constant folding is currently missing support for a few features supported
 // here: CK_ToUnion, CK_ReinterpretMemberPointer, and DesignatedInitUpdateExpr.
-class ConstExprEmitter :
-  public StmtVisitor<ConstExprEmitter, llvm::Constant*, QualType> {
+class ConstExprEmitter
+    : public ConstStmtVisitor<ConstExprEmitter, llvm::Constant *, QualType> {
   CodeGenModule &CGM;
   ConstantEmitter &Emitter;
   llvm::LLVMContext &VMContext;
@@ -1027,43 +1029,42 @@ public:
   //                            Visitor Methods
   //===--------------------------------------------------------------------===//
 
-  llvm::Constant *VisitStmt(Stmt *S, QualType T) {
-    return nullptr;
-  }
+  llvm::Constant *VisitStmt(const Stmt *S, QualType T) { return nullptr; }
 
-  llvm::Constant *VisitConstantExpr(ConstantExpr *CE, QualType T) {
+  llvm::Constant *VisitConstantExpr(const ConstantExpr *CE, QualType T) {
     if (llvm::Constant *Result = Emitter.tryEmitConstantExpr(CE))
       return Result;
     return Visit(CE->getSubExpr(), T);
   }
 
-  llvm::Constant *VisitParenExpr(ParenExpr *PE, QualType T) {
+  llvm::Constant *VisitParenExpr(const ParenExpr *PE, QualType T) {
     return Visit(PE->getSubExpr(), T);
   }
 
   llvm::Constant *
-  VisitSubstNonTypeTemplateParmExpr(SubstNonTypeTemplateParmExpr *PE,
+  VisitSubstNonTypeTemplateParmExpr(const SubstNonTypeTemplateParmExpr *PE,
                                     QualType T) {
     return Visit(PE->getReplacement(), T);
   }
 
-  llvm::Constant *VisitGenericSelectionExpr(GenericSelectionExpr *GE,
+  llvm::Constant *VisitGenericSelectionExpr(const GenericSelectionExpr *GE,
                                             QualType T) {
     return Visit(GE->getResultExpr(), T);
   }
 
-  llvm::Constant *VisitChooseExpr(ChooseExpr *CE, QualType T) {
+  llvm::Constant *VisitChooseExpr(const ChooseExpr *CE, QualType T) {
     return Visit(CE->getChosenSubExpr(), T);
   }
 
-  llvm::Constant *VisitCompoundLiteralExpr(CompoundLiteralExpr *E, QualType T) {
+  llvm::Constant *VisitCompoundLiteralExpr(const CompoundLiteralExpr *E,
+                                           QualType T) {
     return Visit(E->getInitializer(), T);
   }
 
-  llvm::Constant *VisitCastExpr(CastExpr *E, QualType destType) {
+  llvm::Constant *VisitCastExpr(const CastExpr *E, QualType destType) {
     if (const auto *ECE = dyn_cast<ExplicitCastExpr>(E))
       CGM.EmitExplicitCastExprType(ECE, Emitter.CGF);
-    Expr *subExpr = E->getSubExpr();
+    const Expr *subExpr = E->getSubExpr();
 
     switch (E->getCastKind()) {
     case CK_ToUnion: {
@@ -1117,7 +1118,8 @@ public:
       // interesting conversions should be done in Evaluate().  But as a
       // special case, allow compound literals to support the gcc extension
       // allowing "struct x {int x;} x = (struct x) {};".
-      if (auto *E = dyn_cast<CompoundLiteralExpr>(subExpr->IgnoreParens()))
+      if (const auto *E =
+              dyn_cast<CompoundLiteralExpr>(subExpr->IgnoreParens()))
         return Visit(E->getInitializer(), destType);
       return nullptr;
     }
@@ -1232,21 +1234,22 @@ public:
     llvm_unreachable("Invalid CastKind");
   }
 
-  llvm::Constant *VisitCXXDefaultInitExpr(CXXDefaultInitExpr *DIE, QualType T) {
+  llvm::Constant *VisitCXXDefaultInitExpr(const CXXDefaultInitExpr *DIE,
+                                          QualType T) {
     // No need for a DefaultInitExprScope: we don't handle 'this' in a
     // constant expression.
     return Visit(DIE->getExpr(), T);
   }
 
-  llvm::Constant *VisitExprWithCleanups(ExprWithCleanups *E, QualType T) {
+  llvm::Constant *VisitExprWithCleanups(const ExprWithCleanups *E, QualType T) {
     return Visit(E->getSubExpr(), T);
   }
 
-  llvm::Constant *VisitIntegerLiteral(IntegerLiteral *I, QualType T) {
+  llvm::Constant *VisitIntegerLiteral(const IntegerLiteral *I, QualType T) {
     return llvm::ConstantInt::get(CGM.getLLVMContext(), I->getValue());
   }
 
-  llvm::Constant *EmitArrayInitialization(InitListExpr *ILE, QualType T) {
+  llvm::Constant *EmitArrayInitialization(const InitListExpr *ILE, QualType T) {
     auto *CAT = CGM.getContext().getAsConstantArrayType(ILE->getType());
     assert(CAT && "can't emit array init for non-constant-bound array");
     unsigned NumInitElements = ILE->getNumInits();
@@ -1260,7 +1263,7 @@ public:
 
     // Initialize remaining array elements.
     llvm::Constant *fillC = nullptr;
-    if (Expr *filler = ILE->getArrayFiller()) {
+    if (const Expr *filler = ILE->getArrayFiller()) {
       fillC = Emitter.tryEmitAbstractForMemory(filler, EltType);
       if (!fillC)
         return nullptr;
@@ -1275,7 +1278,7 @@ public:
 
     llvm::Type *CommonElementType = nullptr;
     for (unsigned i = 0; i < NumInitableElts; ++i) {
-      Expr *Init = ILE->getInit(i);
+      const Expr *Init = ILE->getInit(i);
       llvm::Constant *C = Emitter.tryEmitPrivateForMemory(Init, EltType);
       if (!C)
         return nullptr;
@@ -1292,16 +1295,17 @@ public:
                              fillC);
   }
 
-  llvm::Constant *EmitRecordInitialization(InitListExpr *ILE, QualType T) {
+  llvm::Constant *EmitRecordInitialization(const InitListExpr *ILE,
+                                           QualType T) {
     return ConstStructBuilder::BuildStruct(Emitter, ILE, T);
   }
 
-  llvm::Constant *VisitImplicitValueInitExpr(ImplicitValueInitExpr* E,
+  llvm::Constant *VisitImplicitValueInitExpr(const ImplicitValueInitExpr *E,
                                              QualType T) {
     return CGM.EmitNullConstant(T);
   }
 
-  llvm::Constant *VisitInitListExpr(InitListExpr *ILE, QualType T) {
+  llvm::Constant *VisitInitListExpr(const InitListExpr *ILE, QualType T) {
     if (ILE->isTransparent())
       return Visit(ILE->getInit(0), T);
 
@@ -1314,8 +1318,9 @@ public:
     return nullptr;
   }
 
-  llvm::Constant *VisitDesignatedInitUpdateExpr(DesignatedInitUpdateExpr *E,
-                                                QualType destType) {
+  llvm::Constant *
+  VisitDesignatedInitUpdateExpr(const DesignatedInitUpdateExpr *E,
+                                QualType destType) {
     auto C = Visit(E->getBase(), destType);
     if (!C)
       return nullptr;
@@ -1329,12 +1334,13 @@ public:
 
     llvm::Type *ValTy = CGM.getTypes().ConvertType(destType);
     bool HasFlexibleArray = false;
-    if (auto *RT = destType->getAs<RecordType>())
+    if (const auto *RT = destType->getAs<RecordType>())
       HasFlexibleArray = RT->getDecl()->hasFlexibleArrayMember();
     return Const.build(ValTy, HasFlexibleArray);
   }
 
-  llvm::Constant *VisitCXXConstructExpr(CXXConstructExpr *E, QualType Ty) {
+  llvm::Constant *VisitCXXConstructExpr(const CXXConstructExpr *E,
+                                        QualType Ty) {
     if (!E->getConstructor()->isTrivial())
       return nullptr;
 
@@ -1344,13 +1350,13 @@ public:
       assert(E->getConstructor()->isCopyOrMoveConstructor() &&
              "trivial ctor has argument but isn't a copy/move ctor");
 
-      Expr *Arg = E->getArg(0);
+      const Expr *Arg = E->getArg(0);
       assert(CGM.getContext().hasSameUnqualifiedType(Ty, Arg->getType()) &&
              "argument to copy ctor is of wrong type");
 
       // Look through the temporary; it's just converting the value to an
       // lvalue to pass it to the constructor.
-      if (auto *MTE = dyn_cast<MaterializeTemporaryExpr>(Arg))
+      if (const auto *MTE = dyn_cast<MaterializeTemporaryExpr>(Arg))
         return Visit(MTE->getSubExpr(), Ty);
       // Don't try to support arbitrary lvalue-to-rvalue conversions for now.
       return nullptr;
@@ -1359,12 +1365,12 @@ public:
     return CGM.EmitNullConstant(Ty);
   }
 
-  llvm::Constant *VisitStringLiteral(StringLiteral *E, QualType T) {
+  llvm::Constant *VisitStringLiteral(const StringLiteral *E, QualType T) {
     // This is a string literal initializing an array in an initializer.
     return CGM.GetConstantArrayFromStringLiteral(E);
   }
 
-  llvm::Constant *VisitObjCEncodeExpr(ObjCEncodeExpr *E, QualType T) {
+  llvm::Constant *VisitObjCEncodeExpr(const ObjCEncodeExpr *E, QualType T) {
     // This must be an @encode initializing an array in a static initializer.
     // Don't emit it as the address of the string, emit the string data itself
     // as an inline array.
@@ -1383,14 +1389,14 @@ public:
     return Visit(E->getSubExpr(), T);
   }
 
-  llvm::Constant *VisitUnaryMinus(UnaryOperator *U, QualType T) {
+  llvm::Constant *VisitUnaryMinus(const UnaryOperator *U, QualType T) {
     if (llvm::Constant *C = Visit(U->getSubExpr(), T))
       if (auto *CI = dyn_cast<llvm::ConstantInt>(C))
         return llvm::ConstantInt::get(CGM.getLLVMContext(), -CI->getValue());
     return nullptr;
   }
 
-  llvm::Constant *VisitPackIndexingExpr(PackIndexingExpr *E, QualType T) {
+  llvm::Constant *VisitPackIndexingExpr(const PackIndexingExpr *E, QualType T) {
     return Visit(E->getSelectedExpr(), T);
   }
 
@@ -1696,8 +1702,7 @@ llvm::Constant *ConstantEmitter::tryEmitPrivateForVarInit(const VarDecl &D) {
 
   if (!destType->isReferenceType()) {
     QualType nonMemoryDestType = getNonMemoryType(CGM, destType);
-    if (llvm::Constant *C = ConstExprEmitter(*this).Visit(const_cast<Expr *>(E),
-                                                          nonMemoryDestType))
+    if (llvm::Constant *C = ConstExprEmitter(*this).Visit(E, nonMemoryDestType))
       return emitForMemory(C, destType);
   }
 
@@ -1777,8 +1782,7 @@ llvm::Constant *ConstantEmitter::tryEmitPrivate(const Expr *E,
   assert(!destType->isVoidType() && "can't emit a void constant");
 
   if (!destType->isReferenceType())
-    if (llvm::Constant *C =
-            ConstExprEmitter(*this).Visit(const_cast<Expr *>(E), destType))
+    if (llvm::Constant *C = ConstExprEmitter(*this).Visit(E, destType))
       return C;
 
   Expr::EvalResult Result;
@@ -2022,7 +2026,7 @@ ConstantLValue
 ConstantLValueEmitter::VisitObjCBoxedExpr(const ObjCBoxedExpr *E) {
   assert(E->isExpressibleAsConstantInitializer() &&
          "this boxed expression can't be emitted as a compile-time constant");
-  auto *SL = cast<StringLiteral>(E->getSubExpr()->IgnoreParenCasts());
+  const auto *SL = cast<StringLiteral>(E->getSubExpr()->IgnoreParenCasts());
   return emitConstantObjCStringLiteral(SL, E->getType(), CGM);
 }
 
@@ -2048,12 +2052,12 @@ ConstantLValueEmitter::VisitCallExpr(const CallExpr *E) {
       builtin != Builtin::BI__builtin___NSStringMakeConstantString)
     return nullptr;
 
-  auto literal = cast<StringLiteral>(E->getArg(0)->IgnoreParenCasts());
+  const auto *Literal = cast<StringLiteral>(E->getArg(0)->IgnoreParenCasts());
   if (builtin == Builtin::BI__builtin___NSStringMakeConstantString) {
-    return CGM.getObjCRuntime().GenerateConstantString(literal);
+    return CGM.getObjCRuntime().GenerateConstantString(Literal);
   } else {
     // FIXME: need to deal with UCN conversion issues.
-    return CGM.GetAddrOfConstantCFString(literal);
+    return CGM.GetAddrOfConstantCFString(Literal);
   }
 }
 
diff --git a/clang/lib/CodeGen/CGExprScalar.cpp b/clang/lib/CodeGen/CGExprScalar.cpp
index 40a5cd20c3d7..af48e8d2b839 100644
--- a/clang/lib/CodeGen/CGExprScalar.cpp
+++ b/clang/lib/CodeGen/CGExprScalar.cpp
@@ -2330,7 +2330,7 @@ Value *ScalarExprEmitter::VisitCastExpr(CastExpr *CE) {
     }
 
     // Perform VLAT <-> VLST bitcast through memory.
-    // TODO: since the llvm.experimental.vector.{insert,extract} intrinsics
+    // TODO: since the llvm.vector.{insert,extract} intrinsics
     //       require the element types of the vectors to be the same, we
     //       need to keep this around for bitcasts between VLAT <-> VLST where
     //       the element types of the vectors are not the same, until we figure
diff --git a/clang/lib/CodeGen/CGLoopInfo.cpp b/clang/lib/CodeGen/CGLoopInfo.cpp
index 72d1471021ac..0d4800b90a2f 100644
--- a/clang/lib/CodeGen/CGLoopInfo.cpp
+++ b/clang/lib/CodeGen/CGLoopInfo.cpp
@@ -673,8 +673,6 @@ void LoopInfoStack::push(BasicBlock *Header, clang::ASTContext &Ctx,
         setPipelineDisabled(true);
         break;
       case LoopHintAttr::UnrollCount:
-        setUnrollState(LoopAttributes::Disable);
-        break;
       case LoopHintAttr::UnrollAndJamCount:
       case LoopHintAttr::VectorizeWidth:
       case LoopHintAttr::InterleaveCount:
diff --git a/clang/lib/CodeGen/CGObjCGNU.cpp b/clang/lib/CodeGen/CGObjCGNU.cpp
index 4e7f777ba1d9..43dd38659518 100644
--- a/clang/lib/CodeGen/CGObjCGNU.cpp
+++ b/clang/lib/CodeGen/CGObjCGNU.cpp
@@ -2905,23 +2905,29 @@ CGObjCGNU::GenerateMessageSend(CodeGenFunction &CGF,
       break;
     case CodeGenOptions::Mixed:
     case CodeGenOptions::NonLegacy:
+      StringRef name = "objc_msgSend";
       if (CGM.ReturnTypeUsesFPRet(ResultType)) {
-        imp =
-            CGM.CreateRuntimeFunction(llvm::FunctionType::get(IdTy, IdTy, true),
-                                      "objc_msgSend_fpret")
-                .getCallee();
+        name = "objc_msgSend_fpret";
       } else if (CGM.ReturnTypeUsesSRet(MSI.CallInfo)) {
-        // The actual types here don't matter - we're going to bitcast the
-        // function anyway
-        imp =
-            CGM.CreateRuntimeFunction(llvm::FunctionType::get(IdTy, IdTy, true),
-                                      "objc_msgSend_stret")
-                .getCallee();
-      } else {
-        imp = CGM.CreateRuntimeFunction(
-                     llvm::FunctionType::get(IdTy, IdTy, true), "objc_msgSend")
-                  .getCallee();
+        name = "objc_msgSend_stret";
+
+        // The address of the memory block is be passed in x8 for POD type,
+        // or in x0 for non-POD type (marked as inreg).
+        bool shouldCheckForInReg =
+            CGM.getContext()
+                .getTargetInfo()
+                .getTriple()
+                .isWindowsMSVCEnvironment() &&
+            CGM.getContext().getTargetInfo().getTriple().isAArch64();
+        if (shouldCheckForInReg && CGM.ReturnTypeHasInReg(MSI.CallInfo)) {
+          name = "objc_msgSend_stret2";
+        }
       }
+      // The actual types here don't matter - we're going to bitcast the
+      // function anyway
+      imp = CGM.CreateRuntimeFunction(llvm::FunctionType::get(IdTy, IdTy, true),
+                                      name)
+                .getCallee();
     }
 
   // Reset the receiver in case the lookup modified it
diff --git a/clang/lib/CodeGen/CGOpenMPRuntime.cpp b/clang/lib/CodeGen/CGOpenMPRuntime.cpp
index 2ae11e129c75..e39c7c58d278 100644
--- a/clang/lib/CodeGen/CGOpenMPRuntime.cpp
+++ b/clang/lib/CodeGen/CGOpenMPRuntime.cpp
@@ -742,8 +742,8 @@ LValue ReductionCodeGen::emitSharedLValue(CodeGenFunction &CGF, const Expr *E) {
 
 LValue ReductionCodeGen::emitSharedLValueUB(CodeGenFunction &CGF,
                                             const Expr *E) {
-  if (const auto *OASE = dyn_cast<OMPArraySectionExpr>(E))
-    return CGF.EmitOMPArraySectionExpr(OASE, /*IsLowerBound=*/false);
+  if (const auto *OASE = dyn_cast<ArraySectionExpr>(E))
+    return CGF.EmitArraySectionExpr(OASE, /*IsLowerBound=*/false);
   return LValue();
 }
 
@@ -800,7 +800,7 @@ void ReductionCodeGen::emitSharedOrigLValue(CodeGenFunction &CGF, unsigned N) {
 
 void ReductionCodeGen::emitAggregateType(CodeGenFunction &CGF, unsigned N) {
   QualType PrivateType = getPrivateType(N);
-  bool AsArraySection = isa<OMPArraySectionExpr>(ClausesData[N].Ref);
+  bool AsArraySection = isa<ArraySectionExpr>(ClausesData[N].Ref);
   if (!PrivateType->isVariablyModifiedType()) {
     Sizes.emplace_back(
         CGF.getTypeSize(OrigAddresses[N].first.getType().getNonReferenceType()),
@@ -941,9 +941,9 @@ static Address castToBase(CodeGenFunction &CGF, QualType BaseTy, QualType ElTy,
 
 static const VarDecl *getBaseDecl(const Expr *Ref, const DeclRefExpr *&DE) {
   const VarDecl *OrigVD = nullptr;
-  if (const auto *OASE = dyn_cast<OMPArraySectionExpr>(Ref)) {
+  if (const auto *OASE = dyn_cast<ArraySectionExpr>(Ref)) {
     const Expr *Base = OASE->getBase()->IgnoreParenImpCasts();
-    while (const auto *TempOASE = dyn_cast<OMPArraySectionExpr>(Base))
+    while (const auto *TempOASE = dyn_cast<ArraySectionExpr>(Base))
       Base = TempOASE->getBase()->IgnoreParenImpCasts();
     while (const auto *TempASE = dyn_cast<ArraySubscriptExpr>(Base))
       Base = TempASE->getBase()->IgnoreParenImpCasts();
@@ -3570,9 +3570,8 @@ getPointerAndSize(CodeGenFunction &CGF, const Expr *E) {
       SizeVal = CGF.Builder.CreateNUWMul(SizeVal, Sz);
     }
   } else if (const auto *ASE =
-                 dyn_cast<OMPArraySectionExpr>(E->IgnoreParenImpCasts())) {
-    LValue UpAddrLVal =
-        CGF.EmitOMPArraySectionExpr(ASE, /*IsLowerBound=*/false);
+                 dyn_cast<ArraySectionExpr>(E->IgnoreParenImpCasts())) {
+    LValue UpAddrLVal = CGF.EmitArraySectionExpr(ASE, /*IsLowerBound=*/false);
     Address UpAddrAddress = UpAddrLVal.getAddress(CGF);
     llvm::Value *UpAddr = CGF.Builder.CreateConstGEP1_32(
         UpAddrAddress.getElementType(), UpAddrAddress.emitRawPointer(CGF),
@@ -6672,8 +6671,8 @@ private:
     // Given that an array section is considered a built-in type, we need to
     // do the calculation based on the length of the section instead of relying
     // on CGF.getTypeSize(E->getType()).
-    if (const auto *OAE = dyn_cast<OMPArraySectionExpr>(E)) {
-      QualType BaseTy = OMPArraySectionExpr::getBaseOriginalType(
+    if (const auto *OAE = dyn_cast<ArraySectionExpr>(E)) {
+      QualType BaseTy = ArraySectionExpr::getBaseOriginalType(
                             OAE->getBase()->IgnoreParenImpCasts())
                             .getCanonicalType();
 
@@ -6779,7 +6778,7 @@ private:
   /// Return true if the provided expression is a final array section. A
   /// final array section, is one whose length can't be proved to be one.
   bool isFinalArraySectionExpression(const Expr *E) const {
-    const auto *OASE = dyn_cast<OMPArraySectionExpr>(E);
+    const auto *OASE = dyn_cast<ArraySectionExpr>(E);
 
     // It is not an array section and therefore not a unity-size one.
     if (!OASE)
@@ -6795,7 +6794,7 @@ private:
     // for this dimension. Also, we should always expect a length if the
     // base type is pointer.
     if (!Length) {
-      QualType BaseQTy = OMPArraySectionExpr::getBaseOriginalType(
+      QualType BaseQTy = ArraySectionExpr::getBaseOriginalType(
                              OASE->getBase()->IgnoreParenImpCasts())
                              .getCanonicalType();
       if (const auto *ATy = dyn_cast<ConstantArrayType>(BaseQTy.getTypePtr()))
@@ -7027,7 +7026,7 @@ private:
     Address BP = Address::invalid();
     const Expr *AssocExpr = I->getAssociatedExpression();
     const auto *AE = dyn_cast<ArraySubscriptExpr>(AssocExpr);
-    const auto *OASE = dyn_cast<OMPArraySectionExpr>(AssocExpr);
+    const auto *OASE = dyn_cast<ArraySectionExpr>(AssocExpr);
     const auto *OAShE = dyn_cast<OMPArrayShapingExpr>(AssocExpr);
 
     if (isa<MemberExpr>(AssocExpr)) {
@@ -7179,14 +7178,14 @@ private:
       // special treatment for array sections given that they are built-in
       // types.
       const auto *OASE =
-          dyn_cast<OMPArraySectionExpr>(I->getAssociatedExpression());
+          dyn_cast<ArraySectionExpr>(I->getAssociatedExpression());
       const auto *OAShE =
           dyn_cast<OMPArrayShapingExpr>(I->getAssociatedExpression());
       const auto *UO = dyn_cast<UnaryOperator>(I->getAssociatedExpression());
       const auto *BO = dyn_cast<BinaryOperator>(I->getAssociatedExpression());
       bool IsPointer =
           OAShE ||
-          (OASE && OMPArraySectionExpr::getBaseOriginalType(OASE)
+          (OASE && ArraySectionExpr::getBaseOriginalType(OASE)
                        .getCanonicalType()
                        ->isAnyPointerType()) ||
           I->getAssociatedExpression()->getType()->isAnyPointerType();
@@ -7207,7 +7206,7 @@ private:
         assert((Next == CE ||
                 isa<MemberExpr>(Next->getAssociatedExpression()) ||
                 isa<ArraySubscriptExpr>(Next->getAssociatedExpression()) ||
-                isa<OMPArraySectionExpr>(Next->getAssociatedExpression()) ||
+                isa<ArraySectionExpr>(Next->getAssociatedExpression()) ||
                 isa<OMPArrayShapingExpr>(Next->getAssociatedExpression()) ||
                 isa<UnaryOperator>(Next->getAssociatedExpression()) ||
                 isa<BinaryOperator>(Next->getAssociatedExpression())) &&
@@ -7439,7 +7438,7 @@ private:
             PartialStruct.LowestElem = {FieldIndex, LowestElem};
             if (IsFinalArraySection) {
               Address HB =
-                  CGF.EmitOMPArraySectionExpr(OASE, /*IsLowerBound=*/false)
+                  CGF.EmitArraySectionExpr(OASE, /*IsLowerBound=*/false)
                       .getAddress(CGF);
               PartialStruct.HighestElem = {FieldIndex, HB};
             } else {
@@ -7452,7 +7451,7 @@ private:
           } else if (FieldIndex > PartialStruct.HighestElem.first) {
             if (IsFinalArraySection) {
               Address HB =
-                  CGF.EmitOMPArraySectionExpr(OASE, /*IsLowerBound=*/false)
+                  CGF.EmitArraySectionExpr(OASE, /*IsLowerBound=*/false)
                       .getAddress(CGF);
               PartialStruct.HighestElem = {FieldIndex, HB};
             } else {
@@ -7510,12 +7509,12 @@ private:
     for (const OMPClauseMappableExprCommon::MappableComponent &Component :
          Components) {
       const Expr *AssocExpr = Component.getAssociatedExpression();
-      const auto *OASE = dyn_cast<OMPArraySectionExpr>(AssocExpr);
+      const auto *OASE = dyn_cast<ArraySectionExpr>(AssocExpr);
 
       if (!OASE)
         continue;
 
-      QualType Ty = OMPArraySectionExpr::getBaseOriginalType(OASE->getBase());
+      QualType Ty = ArraySectionExpr::getBaseOriginalType(OASE->getBase());
       auto *CAT = Context.getAsConstantArrayType(Ty);
       auto *VAT = Context.getAsVariableArrayType(Ty);
 
@@ -7589,7 +7588,7 @@ private:
         continue;
       }
 
-      const auto *OASE = dyn_cast<OMPArraySectionExpr>(AssocExpr);
+      const auto *OASE = dyn_cast<ArraySectionExpr>(AssocExpr);
 
       if (!OASE)
         continue;
@@ -8780,7 +8779,7 @@ static ValueDecl *getDeclFromThisExpr(const Expr *E) {
   if (!E)
     return nullptr;
 
-  if (const auto *OASE = dyn_cast<OMPArraySectionExpr>(E->IgnoreParenCasts()))
+  if (const auto *OASE = dyn_cast<ArraySectionExpr>(E->IgnoreParenCasts()))
     if (const MemberExpr *ME =
             dyn_cast<MemberExpr>(OASE->getBase()->IgnoreParenImpCasts()))
       return ME->getMemberDecl();
diff --git a/clang/lib/CodeGen/CGOpenMPRuntimeGPU.cpp b/clang/lib/CodeGen/CGOpenMPRuntimeGPU.cpp
index eb716520e5ff..87496c8e488c 100644
--- a/clang/lib/CodeGen/CGOpenMPRuntimeGPU.cpp
+++ b/clang/lib/CodeGen/CGOpenMPRuntimeGPU.cpp
@@ -92,9 +92,9 @@ static const ValueDecl *getPrivateItem(const Expr *RefExpr) {
     while (const auto *TempASE = dyn_cast<ArraySubscriptExpr>(Base))
       Base = TempASE->getBase()->IgnoreParenImpCasts();
     RefExpr = Base;
-  } else if (auto *OASE = dyn_cast<OMPArraySectionExpr>(RefExpr)) {
+  } else if (auto *OASE = dyn_cast<ArraySectionExpr>(RefExpr)) {
     const Expr *Base = OASE->getBase()->IgnoreParenImpCasts();
-    while (const auto *TempOASE = dyn_cast<OMPArraySectionExpr>(Base))
+    while (const auto *TempOASE = dyn_cast<ArraySectionExpr>(Base))
       Base = TempOASE->getBase()->IgnoreParenImpCasts();
     while (const auto *TempASE = dyn_cast<ArraySubscriptExpr>(Base))
       Base = TempASE->getBase()->IgnoreParenImpCasts();
diff --git a/clang/lib/CodeGen/CGStmtOpenMP.cpp b/clang/lib/CodeGen/CGStmtOpenMP.cpp
index a0a8a07c76ba..ef3aa3a8e0dc 100644
--- a/clang/lib/CodeGen/CGStmtOpenMP.cpp
+++ b/clang/lib/CodeGen/CGStmtOpenMP.cpp
@@ -1256,7 +1256,7 @@ void CodeGenFunction::EmitOMPReductionClauseInit(
     const auto *LHSVD = cast<VarDecl>(cast<DeclRefExpr>(*ILHS)->getDecl());
     const auto *RHSVD = cast<VarDecl>(cast<DeclRefExpr>(*IRHS)->getDecl());
     QualType Type = PrivateVD->getType();
-    bool isaOMPArraySectionExpr = isa<OMPArraySectionExpr>(IRef);
+    bool isaOMPArraySectionExpr = isa<ArraySectionExpr>(IRef);
     if (isaOMPArraySectionExpr && Type->isVariablyModifiedType()) {
       // Store the address of the original variable associated with the LHS
       // implicit variable.
@@ -7289,7 +7289,7 @@ void CodeGenFunction::EmitOMPUseDevicePtrClause(
 
 static const VarDecl *getBaseDecl(const Expr *Ref) {
   const Expr *Base = Ref->IgnoreParenImpCasts();
-  while (const auto *OASE = dyn_cast<OMPArraySectionExpr>(Base))
+  while (const auto *OASE = dyn_cast<ArraySectionExpr>(Base))
     Base = OASE->getBase()->IgnoreParenImpCasts();
   while (const auto *ASE = dyn_cast<ArraySubscriptExpr>(Base))
     Base = ASE->getBase()->IgnoreParenImpCasts();
diff --git a/clang/lib/CodeGen/CodeGenFunction.cpp b/clang/lib/CodeGen/CodeGenFunction.cpp
index 86a6ddd80cc1..87766a758311 100644
--- a/clang/lib/CodeGen/CodeGenFunction.cpp
+++ b/clang/lib/CodeGen/CodeGenFunction.cpp
@@ -91,6 +91,8 @@ CodeGenFunction::CodeGenFunction(CodeGenModule &cgm, bool suppressNewContext)
 
 CodeGenFunction::~CodeGenFunction() {
   assert(LifetimeExtendedCleanupStack.empty() && "failed to emit a cleanup");
+  assert(DeferredDeactivationCleanupStack.empty() &&
+         "missed to deactivate a cleanup");
 
   if (getLangOpts().OpenMP && CurFn)
     CGM.getOpenMPRuntime().functionFinished(*this);
@@ -346,6 +348,10 @@ static void EmitIfUsed(CodeGenFunction &CGF, llvm::BasicBlock *BB) {
 void CodeGenFunction::FinishFunction(SourceLocation EndLoc) {
   assert(BreakContinueStack.empty() &&
          "mismatched push/pop in break/continue stack!");
+  assert(LifetimeExtendedCleanupStack.empty() &&
+         "mismatched push/pop of cleanups in EHStack!");
+  assert(DeferredDeactivationCleanupStack.empty() &&
+         "mismatched activate/deactivate of cleanups!");
 
   bool OnlySimpleReturnStmts = NumSimpleReturnExprs > 0
     && NumSimpleReturnExprs == NumReturnExprs
diff --git a/clang/lib/CodeGen/CodeGenFunction.h b/clang/lib/CodeGen/CodeGenFunction.h
index ff1873325d40..6e7417fc7f52 100644
--- a/clang/lib/CodeGen/CodeGenFunction.h
+++ b/clang/lib/CodeGen/CodeGenFunction.h
@@ -39,6 +39,7 @@
 #include "llvm/ADT/MapVector.h"
 #include "llvm/ADT/SmallVector.h"
 #include "llvm/Frontend/OpenMP/OMPIRBuilder.h"
+#include "llvm/IR/Instructions.h"
 #include "llvm/IR/ValueHandle.h"
 #include "llvm/Support/Debug.h"
 #include "llvm/Transforms/Utils/SanitizerStats.h"
@@ -670,6 +671,51 @@ public:
 
   EHScopeStack EHStack;
   llvm::SmallVector<char, 256> LifetimeExtendedCleanupStack;
+
+  // A stack of cleanups which were added to EHStack but have to be deactivated
+  // later before being popped or emitted. These are usually deactivated on
+  // exiting a `CleanupDeactivationScope` scope. For instance, after a
+  // full-expr.
+  //
+  // These are specially useful for correctly emitting cleanups while
+  // encountering branches out of expression (through stmt-expr or coroutine
+  // suspensions).
+  struct DeferredDeactivateCleanup {
+    EHScopeStack::stable_iterator Cleanup;
+    llvm::Instruction *DominatingIP;
+  };
+  llvm::SmallVector<DeferredDeactivateCleanup> DeferredDeactivationCleanupStack;
+
+  // Enters a new scope for capturing cleanups which are deferred to be
+  // deactivated, all of which will be deactivated once the scope is exited.
+  struct CleanupDeactivationScope {
+    CodeGenFunction &CGF;
+    size_t OldDeactivateCleanupStackSize;
+    bool Deactivated;
+    CleanupDeactivationScope(CodeGenFunction &CGF)
+        : CGF(CGF), OldDeactivateCleanupStackSize(
+                        CGF.DeferredDeactivationCleanupStack.size()),
+          Deactivated(false) {}
+
+    void ForceDeactivate() {
+      assert(!Deactivated && "Deactivating already deactivated scope");
+      auto &Stack = CGF.DeferredDeactivationCleanupStack;
+      for (size_t I = Stack.size(); I > OldDeactivateCleanupStackSize; I--) {
+        CGF.DeactivateCleanupBlock(Stack[I - 1].Cleanup,
+                                   Stack[I - 1].DominatingIP);
+        Stack[I - 1].DominatingIP->eraseFromParent();
+      }
+      Stack.resize(OldDeactivateCleanupStackSize);
+      Deactivated = true;
+    }
+
+    ~CleanupDeactivationScope() {
+      if (Deactivated)
+        return;
+      ForceDeactivate();
+    }
+  };
+
   llvm::SmallVector<const JumpDest *, 2> SEHTryEpilogueStack;
 
   llvm::Instruction *CurrentFuncletPad = nullptr;
@@ -875,6 +921,19 @@ public:
       new (Buffer + sizeof(Header) + sizeof(T)) RawAddress(ActiveFlag);
   }
 
+  // Push a cleanup onto EHStack and deactivate it later. It is usually
+  // deactivated when exiting a `CleanupDeactivationScope` (for example: after a
+  // full expression).
+  template <class T, class... As>
+  void pushCleanupAndDeferDeactivation(CleanupKind Kind, As... A) {
+    // Placeholder dominating IP for this cleanup.
+    llvm::Instruction *DominatingIP =
+        Builder.CreateFlagLoad(llvm::Constant::getNullValue(Int8PtrTy));
+    EHStack.pushCleanup<T>(Kind, A...);
+    DeferredDeactivationCleanupStack.push_back(
+        {EHStack.stable_begin(), DominatingIP});
+  }
+
   /// Set up the last cleanup that was pushed as a conditional
   /// full-expression cleanup.
   void initFullExprCleanup() {
@@ -898,7 +957,8 @@ public:
 
   /// PopCleanupBlock - Will pop the cleanup entry on the stack and
   /// process all branch fixups.
-  void PopCleanupBlock(bool FallThroughIsBranchThrough = false);
+  void PopCleanupBlock(bool FallThroughIsBranchThrough = false,
+                       bool ForDeactivation = false);
 
   /// DeactivateCleanupBlock - Deactivates the given cleanup block.
   /// The block cannot be reactivated.  Pops it if it's the top of the
@@ -926,6 +986,7 @@ public:
   class RunCleanupsScope {
     EHScopeStack::stable_iterator CleanupStackDepth, OldCleanupScopeDepth;
     size_t LifetimeExtendedCleanupStackSize;
+    CleanupDeactivationScope DeactivateCleanups;
     bool OldDidCallStackSave;
   protected:
     bool PerformCleanup;
@@ -940,8 +1001,7 @@ public:
   public:
     /// Enter a new cleanup scope.
     explicit RunCleanupsScope(CodeGenFunction &CGF)
-      : PerformCleanup(true), CGF(CGF)
-    {
+        : DeactivateCleanups(CGF), PerformCleanup(true), CGF(CGF) {
       CleanupStackDepth = CGF.EHStack.stable_begin();
       LifetimeExtendedCleanupStackSize =
           CGF.LifetimeExtendedCleanupStack.size();
@@ -971,6 +1031,7 @@ public:
     void ForceCleanup(std::initializer_list<llvm::Value**> ValuesToReload = {}) {
       assert(PerformCleanup && "Already forced cleanup");
       CGF.DidCallStackSave = OldDidCallStackSave;
+      DeactivateCleanups.ForceDeactivate();
       CGF.PopCleanupBlocks(CleanupStackDepth, LifetimeExtendedCleanupStackSize,
                            ValuesToReload);
       PerformCleanup = false;
@@ -2160,6 +2221,11 @@ public:
                      Address addr, QualType type);
   void pushDestroy(CleanupKind kind, Address addr, QualType type,
                    Destroyer *destroyer, bool useEHCleanupForArray);
+  void pushDestroyAndDeferDeactivation(QualType::DestructionKind dtorKind,
+                                       Address addr, QualType type);
+  void pushDestroyAndDeferDeactivation(CleanupKind cleanupKind, Address addr,
+                                       QualType type, Destroyer *destroyer,
+                                       bool useEHCleanupForArray);
   void pushLifetimeExtendedDestroy(CleanupKind kind, Address addr,
                                    QualType type, Destroyer *destroyer,
                                    bool useEHCleanupForArray);
@@ -2698,6 +2764,33 @@ public:
                             TBAAAccessInfo *TBAAInfo = nullptr);
   LValue EmitLoadOfPointerLValue(Address Ptr, const PointerType *PtrTy);
 
+private:
+  struct AllocaTracker {
+    void Add(llvm::AllocaInst *I) { Allocas.push_back(I); }
+    llvm::SmallVector<llvm::AllocaInst *> Take() { return std::move(Allocas); }
+
+  private:
+    llvm::SmallVector<llvm::AllocaInst *> Allocas;
+  };
+  AllocaTracker *Allocas = nullptr;
+
+public:
+  // Captures all the allocas created during the scope of its RAII object.
+  struct AllocaTrackerRAII {
+    AllocaTrackerRAII(CodeGenFunction &CGF)
+        : CGF(CGF), OldTracker(CGF.Allocas) {
+      CGF.Allocas = &Tracker;
+    }
+    ~AllocaTrackerRAII() { CGF.Allocas = OldTracker; }
+
+    llvm::SmallVector<llvm::AllocaInst *> Take() { return Tracker.Take(); }
+
+  private:
+    CodeGenFunction &CGF;
+    AllocaTracker *OldTracker;
+    AllocaTracker Tracker;
+  };
+
   /// CreateTempAlloca - This creates an alloca and inserts it into the entry
   /// block if \p ArraySize is nullptr, otherwise inserts it at the current
   /// insertion point of the builder. The caller is responsible for setting an
@@ -3204,12 +3297,12 @@ public:
                            llvm::Value *Index, QualType IndexType,
                            QualType IndexedType, bool Accessed);
 
-  // Find a struct's flexible array member. It may be embedded inside multiple
-  // sub-structs, but must still be the last field.
-  const FieldDecl *FindFlexibleArrayMemberField(ASTContext &Ctx,
-                                                const RecordDecl *RD,
-                                                StringRef Name,
-                                                uint64_t &Offset);
+  // Find a struct's flexible array member and get its offset. It may be
+  // embedded inside multiple sub-structs, but must still be the last field.
+  const FieldDecl *
+  FindFlexibleArrayMemberFieldAndOffset(ASTContext &Ctx, const RecordDecl *RD,
+                                        const FieldDecl *FAMDecl,
+                                        uint64_t &Offset);
 
   /// Find the FieldDecl specified in a FAM's "counted_by" attribute. Returns
   /// \p nullptr if either the attribute or the field doesn't exist.
@@ -4169,8 +4262,8 @@ public:
   LValue EmitArraySubscriptExpr(const ArraySubscriptExpr *E,
                                 bool Accessed = false);
   LValue EmitMatrixSubscriptExpr(const MatrixSubscriptExpr *E);
-  LValue EmitOMPArraySectionExpr(const OMPArraySectionExpr *E,
-                                 bool IsLowerBound = true);
+  LValue EmitArraySectionExpr(const ArraySectionExpr *E,
+                              bool IsLowerBound = true);
   LValue EmitExtVectorElementExpr(const ExtVectorElementExpr *E);
   LValue EmitMemberExpr(const MemberExpr *E);
   LValue EmitObjCIsaExpr(const ObjCIsaExpr *E);
diff --git a/clang/lib/CodeGen/CodeGenModule.cpp b/clang/lib/CodeGen/CodeGenModule.cpp
index 0c447b20cef4..d085e735ecb4 100644
--- a/clang/lib/CodeGen/CodeGenModule.cpp
+++ b/clang/lib/CodeGen/CodeGenModule.cpp
@@ -68,9 +68,9 @@
 #include "llvm/Support/CommandLine.h"
 #include "llvm/Support/ConvertUTF.h"
 #include "llvm/Support/ErrorHandling.h"
-#include "llvm/Support/RISCVISAInfo.h"
 #include "llvm/Support/TimeProfiler.h"
 #include "llvm/Support/xxhash.h"
+#include "llvm/TargetParser/RISCVISAInfo.h"
 #include "llvm/TargetParser/Triple.h"
 #include "llvm/TargetParser/X86TargetParser.h"
 #include <optional>
diff --git a/clang/lib/CodeGen/CodeGenModule.h b/clang/lib/CodeGen/CodeGenModule.h
index 1cc447765e2c..be43a18fc608 100644
--- a/clang/lib/CodeGen/CodeGenModule.h
+++ b/clang/lib/CodeGen/CodeGenModule.h
@@ -1241,6 +1241,9 @@ public:
   /// Return true iff the given type uses 'sret' when used as a return type.
   bool ReturnTypeUsesSRet(const CGFunctionInfo &FI);
 
+  /// Return true iff the given type has `inreg` set.
+  bool ReturnTypeHasInReg(const CGFunctionInfo &FI);
+
   /// Return true iff the given type uses an argument slot when 'sret' is used
   /// as a return type.
   bool ReturnSlotInterferesWithArgs(const CGFunctionInfo &FI);
diff --git a/clang/lib/CodeGen/CodeGenTypes.cpp b/clang/lib/CodeGen/CodeGenTypes.cpp
index 1568b6e6275b..e8d75eda029e 100644
--- a/clang/lib/CodeGen/CodeGenTypes.cpp
+++ b/clang/lib/CodeGen/CodeGenTypes.cpp
@@ -409,7 +409,7 @@ llvm::Type *CodeGenTypes::ConvertType(QualType T) {
       break;
     case BuiltinType::LongDouble:
       LongDoubleReferenced = true;
-      LLVM_FALLTHROUGH;
+      [[fallthrough]];
     case BuiltinType::BFloat16:
     case BuiltinType::Float:
     case BuiltinType::Double:
diff --git a/clang/lib/Driver/Driver.cpp b/clang/lib/Driver/Driver.cpp
index 0da92001e08c..76b7b9fdfb4f 100644
--- a/clang/lib/Driver/Driver.cpp
+++ b/clang/lib/Driver/Driver.cpp
@@ -87,12 +87,12 @@
 #include "llvm/Support/PrettyStackTrace.h"
 #include "llvm/Support/Process.h"
 #include "llvm/Support/Program.h"
-#include "llvm/Support/RISCVISAInfo.h"
 #include "llvm/Support/Regex.h"
 #include "llvm/Support/StringSaver.h"
 #include "llvm/Support/VirtualFileSystem.h"
 #include "llvm/Support/raw_ostream.h"
 #include "llvm/TargetParser/Host.h"
+#include "llvm/TargetParser/RISCVISAInfo.h"
 #include <cstdlib> // ::getenv
 #include <map>
 #include <memory>
diff --git a/clang/lib/Driver/SanitizerArgs.cpp b/clang/lib/Driver/SanitizerArgs.cpp
index 8bfe9f02a091..6a4f2548c0bf 100644
--- a/clang/lib/Driver/SanitizerArgs.cpp
+++ b/clang/lib/Driver/SanitizerArgs.cpp
@@ -1192,7 +1192,9 @@ void SanitizerArgs::addArgs(const ToolChain &TC, const llvm::opt::ArgList &Args,
                           BinaryMetadataIgnorelistFiles);
   }
 
-  if (TC.getTriple().isOSWindows() && needsUbsanRt()) {
+  if (TC.getTriple().isOSWindows() && needsUbsanRt() &&
+      Args.hasFlag(options::OPT_frtlib_defaultlib,
+                   options::OPT_fno_rtlib_defaultlib, true)) {
     // Instruct the code generator to embed linker directives in the object file
     // that cause the required runtime libraries to be linked.
     CmdArgs.push_back(
@@ -1203,7 +1205,9 @@ void SanitizerArgs::addArgs(const ToolChain &TC, const llvm::opt::ArgList &Args,
           "--dependent-lib=" +
           TC.getCompilerRTBasename(Args, "ubsan_standalone_cxx")));
   }
-  if (TC.getTriple().isOSWindows() && needsStatsRt()) {
+  if (TC.getTriple().isOSWindows() && needsStatsRt() &&
+      Args.hasFlag(options::OPT_frtlib_defaultlib,
+                   options::OPT_fno_rtlib_defaultlib, true)) {
     CmdArgs.push_back(Args.MakeArgString(
         "--dependent-lib=" + TC.getCompilerRTBasename(Args, "stats_client")));
 
diff --git a/clang/lib/Driver/ToolChain.cpp b/clang/lib/Driver/ToolChain.cpp
index 237092ed07e5..341d6202a9ca 100644
--- a/clang/lib/Driver/ToolChain.cpp
+++ b/clang/lib/Driver/ToolChain.cpp
@@ -1307,9 +1307,14 @@ void ToolChain::AddCCKextLibArgs(const ArgList &Args,
 
 bool ToolChain::isFastMathRuntimeAvailable(const ArgList &Args,
                                            std::string &Path) const {
+  // Don't implicitly link in mode-changing libraries in a shared library, since
+  // this can have very deleterious effects. See the various links from
+  // https://github.com/llvm/llvm-project/issues/57589 for more information.
+  bool Default = !Args.hasArgNoClaim(options::OPT_shared);
+
   // Do not check for -fno-fast-math or -fno-unsafe-math when -Ofast passed
   // (to keep the linker options consistent with gcc and clang itself).
-  if (!isOptimizationLevelFast(Args)) {
+  if (Default && !isOptimizationLevelFast(Args)) {
     // Check if -ffast-math or -funsafe-math.
     Arg *A =
       Args.getLastArg(options::OPT_ffast_math, options::OPT_fno_fast_math,
@@ -1318,8 +1323,14 @@ bool ToolChain::isFastMathRuntimeAvailable(const ArgList &Args,
 
     if (!A || A->getOption().getID() == options::OPT_fno_fast_math ||
         A->getOption().getID() == options::OPT_fno_unsafe_math_optimizations)
-      return false;
+      Default = false;
   }
+
+  // Whatever decision came as a result of the above implicit settings, either
+  // -mdaz-ftz or -mno-daz-ftz is capable of overriding it.
+  if (!Args.hasFlag(options::OPT_mdaz_ftz, options::OPT_mno_daz_ftz, Default))
+    return false;
+
   // If crtfastmath.o exists add it to the arguments.
   Path = GetFilePath("crtfastmath.o");
   return (Path != "crtfastmath.o"); // Not found.
diff --git a/clang/lib/Driver/ToolChains/AIX.cpp b/clang/lib/Driver/ToolChains/AIX.cpp
index c1b350893b37..aab98506adb9 100644
--- a/clang/lib/Driver/ToolChains/AIX.cpp
+++ b/clang/lib/Driver/ToolChains/AIX.cpp
@@ -376,9 +376,7 @@ void AIX::AddOpenMPIncludeArgs(const ArgList &DriverArgs,
       addSystemInclude(DriverArgs, CC1Args, PathOpenMP.str());
       break;
     case Driver::OMPRT_IOMP5:
-      LLVM_FALLTHROUGH;
     case Driver::OMPRT_GOMP:
-      LLVM_FALLTHROUGH;
     case Driver::OMPRT_Unknown:
       // Unknown / unsupported include paths.
       break;
diff --git a/clang/lib/Driver/ToolChains/AMDGPU.cpp b/clang/lib/Driver/ToolChains/AMDGPU.cpp
index 4e6362a0f406..07965b487ea7 100644
--- a/clang/lib/Driver/ToolChains/AMDGPU.cpp
+++ b/clang/lib/Driver/ToolChains/AMDGPU.cpp
@@ -617,8 +617,7 @@ void amdgpu::Linker::ConstructJob(Compilation &C, const JobAction &JA,
                                   const InputInfoList &Inputs,
                                   const ArgList &Args,
                                   const char *LinkingOutput) const {
-
-  std::string Linker = getToolChain().GetProgramPath(getShortName());
+  std::string Linker = getToolChain().GetLinkerPath();
   ArgStringList CmdArgs;
   CmdArgs.push_back("--no-undefined");
   CmdArgs.push_back("-shared");
diff --git a/clang/lib/Driver/ToolChains/Arch/RISCV.cpp b/clang/lib/Driver/ToolChains/Arch/RISCV.cpp
index 96b3cc3bb8ff..2e2bce849467 100644
--- a/clang/lib/Driver/ToolChains/Arch/RISCV.cpp
+++ b/clang/lib/Driver/ToolChains/Arch/RISCV.cpp
@@ -15,9 +15,9 @@
 #include "clang/Driver/Options.h"
 #include "llvm/Option/ArgList.h"
 #include "llvm/Support/Error.h"
-#include "llvm/Support/RISCVISAInfo.h"
 #include "llvm/Support/raw_ostream.h"
 #include "llvm/TargetParser/Host.h"
+#include "llvm/TargetParser/RISCVISAInfo.h"
 #include "llvm/TargetParser/RISCVTargetParser.h"
 
 using namespace clang::driver;
diff --git a/clang/lib/Driver/ToolChains/Clang.cpp b/clang/lib/Driver/ToolChains/Clang.cpp
index 5894a48e0e37..651a2b5aac36 100644
--- a/clang/lib/Driver/ToolChains/Clang.cpp
+++ b/clang/lib/Driver/ToolChains/Clang.cpp
@@ -54,11 +54,11 @@
 #include "llvm/Support/FileSystem.h"
 #include "llvm/Support/Path.h"
 #include "llvm/Support/Process.h"
-#include "llvm/Support/RISCVISAInfo.h"
 #include "llvm/Support/YAMLParser.h"
 #include "llvm/TargetParser/ARMTargetParserCommon.h"
 #include "llvm/TargetParser/Host.h"
 #include "llvm/TargetParser/LoongArchTargetParser.h"
+#include "llvm/TargetParser/RISCVISAInfo.h"
 #include "llvm/TargetParser/RISCVTargetParser.h"
 #include <cctype>
 
@@ -637,7 +637,9 @@ static void addPGOAndCoverageFlags(const ToolChain &TC, Compilation &C,
                                            ProfileGenerateArg->getValue()));
     // The default is to use Clang Instrumentation.
     CmdArgs.push_back("-fprofile-instrument=clang");
-    if (TC.getTriple().isWindowsMSVCEnvironment()) {
+    if (TC.getTriple().isWindowsMSVCEnvironment() &&
+        Args.hasFlag(options::OPT_frtlib_defaultlib,
+                     options::OPT_fno_rtlib_defaultlib, true)) {
       // Add dependent lib for clang_rt.profile
       CmdArgs.push_back(Args.MakeArgString(
           "--dependent-lib=" + TC.getCompilerRTBasename(Args, "profile")));
@@ -656,7 +658,9 @@ static void addPGOAndCoverageFlags(const ToolChain &TC, Compilation &C,
     CmdArgs.push_back("-fprofile-instrument=csllvm");
   }
   if (PGOGenArg) {
-    if (TC.getTriple().isWindowsMSVCEnvironment()) {
+    if (TC.getTriple().isWindowsMSVCEnvironment() &&
+        Args.hasFlag(options::OPT_frtlib_defaultlib,
+                     options::OPT_fno_rtlib_defaultlib, true)) {
       // Add dependent lib for clang_rt.profile
       CmdArgs.push_back(Args.MakeArgString(
           "--dependent-lib=" + TC.getCompilerRTBasename(Args, "profile")));
@@ -831,46 +835,6 @@ static void addPGOAndCoverageFlags(const ToolChain &TC, Compilation &C,
   }
 }
 
-/// Check whether the given input tree contains any compilation actions.
-static bool ContainsCompileAction(const Action *A) {
-  if (isa<CompileJobAction>(A) || isa<BackendJobAction>(A))
-    return true;
-
-  return llvm::any_of(A->inputs(), ContainsCompileAction);
-}
-
-/// Check if -relax-all should be passed to the internal assembler.
-/// This is done by default when compiling non-assembler source with -O0.
-static bool UseRelaxAll(Compilation &C, const ArgList &Args) {
-  bool RelaxDefault = true;
-
-  if (Arg *A = Args.getLastArg(options::OPT_O_Group))
-    RelaxDefault = A->getOption().matches(options::OPT_O0);
-
-  // RISC-V requires an indirect jump for offsets larger than 1MiB. This cannot
-  // be done by assembler branch relaxation as it needs a free temporary
-  // register. Because of this, branch relaxation is handled by a MachineIR
-  // pass before the assembler. Forcing assembler branch relaxation for -O0
-  // makes the MachineIR branch relaxation inaccurate and it will miss cases
-  // where an indirect branch is necessary. To avoid this issue we are
-  // sacrificing the compile time improvement of using -mrelax-all for -O0.
-  if (C.getDefaultToolChain().getTriple().isRISCV())
-    RelaxDefault = false;
-
-  if (RelaxDefault) {
-    RelaxDefault = false;
-    for (const auto &Act : C.getActions()) {
-      if (ContainsCompileAction(Act)) {
-        RelaxDefault = true;
-        break;
-      }
-    }
-  }
-
-  return Args.hasFlag(options::OPT_mrelax_all, options::OPT_mno_relax_all,
-                      RelaxDefault);
-}
-
 static void
 RenderDebugEnablingArgs(const ArgList &Args, ArgStringList &CmdArgs,
                         llvm::codegenoptions::DebugInfoKind DebugInfoKind,
@@ -2468,8 +2432,16 @@ static void CollectArgsForIntegratedAssembler(Compilation &C,
                                               const ArgList &Args,
                                               ArgStringList &CmdArgs,
                                               const Driver &D) {
-  if (UseRelaxAll(C, Args))
-    CmdArgs.push_back("-mrelax-all");
+  // Default to -mno-relax-all.
+  //
+  // Note: RISC-V requires an indirect jump for offsets larger than 1MiB. This
+  // cannot be done by assembler branch relaxation as it needs a free temporary
+  // register. Because of this, branch relaxation is handled by a MachineIR pass
+  // before the assembler. Forcing assembler branch relaxation for -O0 makes the
+  // MachineIR branch relaxation inaccurate and it will miss cases where an
+  // indirect branch is necessary.
+  Args.addOptInFlag(CmdArgs, options::OPT_mrelax_all,
+                    options::OPT_mno_relax_all);
 
   // Only default to -mincremental-linker-compatible if we think we are
   // targeting the MSVC linker.
diff --git a/clang/lib/Driver/ToolChains/CommonArgs.cpp b/clang/lib/Driver/ToolChains/CommonArgs.cpp
index b65b96db16bd..fec11c7e716f 100644
--- a/clang/lib/Driver/ToolChains/CommonArgs.cpp
+++ b/clang/lib/Driver/ToolChains/CommonArgs.cpp
@@ -1191,118 +1191,10 @@ bool tools::addOpenMPRuntime(const Compilation &C, ArgStringList &CmdArgs,
   return true;
 }
 
-/// Determines if --whole-archive is active in the list of arguments.
-static bool isWholeArchivePresent(const ArgList &Args) {
-  bool WholeArchiveActive = false;
-  for (auto *Arg : Args.filtered(options::OPT_Wl_COMMA)) {
-    if (Arg) {
-      for (StringRef ArgValue : Arg->getValues()) {
-        if (ArgValue == "--whole-archive")
-          WholeArchiveActive = true;
-        if (ArgValue == "--no-whole-archive")
-          WholeArchiveActive = false;
-      }
-    }
-  }
-
-  return WholeArchiveActive;
-}
-
-/// Determine if driver is invoked to create a shared object library (-static)
-static bool isSharedLinkage(const ArgList &Args) {
-  return Args.hasArg(options::OPT_shared);
-}
-
-/// Determine if driver is invoked to create a static object library (-shared)
-static bool isStaticLinkage(const ArgList &Args) {
-  return Args.hasArg(options::OPT_static);
-}
-
-/// Add Fortran runtime libs for MSVC
-static void addFortranRuntimeLibsMSVC(const ArgList &Args,
-                                      llvm::opt::ArgStringList &CmdArgs) {
-  unsigned RTOptionID = options::OPT__SLASH_MT;
-  if (auto *rtl = Args.getLastArg(options::OPT_fms_runtime_lib_EQ)) {
-    RTOptionID = llvm::StringSwitch<unsigned>(rtl->getValue())
-                     .Case("static", options::OPT__SLASH_MT)
-                     .Case("static_dbg", options::OPT__SLASH_MTd)
-                     .Case("dll", options::OPT__SLASH_MD)
-                     .Case("dll_dbg", options::OPT__SLASH_MDd)
-                     .Default(options::OPT__SLASH_MT);
-  }
-  switch (RTOptionID) {
-  case options::OPT__SLASH_MT:
-    CmdArgs.push_back("/WHOLEARCHIVE:Fortran_main.static.lib");
-    break;
-  case options::OPT__SLASH_MTd:
-    CmdArgs.push_back("/WHOLEARCHIVE:Fortran_main.static_dbg.lib");
-    break;
-  case options::OPT__SLASH_MD:
-    CmdArgs.push_back("/WHOLEARCHIVE:Fortran_main.dynamic.lib");
-    break;
-  case options::OPT__SLASH_MDd:
-    CmdArgs.push_back("/WHOLEARCHIVE:Fortran_main.dynamic_dbg.lib");
-    break;
-  }
-}
-
-// Add FortranMain runtime lib
-static void addFortranMain(const ToolChain &TC, const ArgList &Args,
-                           llvm::opt::ArgStringList &CmdArgs) {
-  // 0. Shared-library linkage
-  // If we are attempting to link a library, we should not add
-  // -lFortran_main.a to the link line, as the `main` symbol is not
-  // required for a library and should also be provided by one of
-  // the translation units of the code that this shared library
-  // will be linked against eventually.
-  if (isSharedLinkage(Args) || isStaticLinkage(Args)) {
-    return;
-  }
-
-  // 1. MSVC
-  if (TC.getTriple().isKnownWindowsMSVCEnvironment()) {
-    addFortranRuntimeLibsMSVC(Args, CmdArgs);
-    return;
-  }
-
-  // 2. GNU and similar
-  const Driver &D = TC.getDriver();
-  const char *FortranMainLinkFlag = "-lFortran_main";
-
-  // Warn if the user added `-lFortran_main` - this library is an implementation
-  // detail of Flang and should be handled automaticaly by the driver.
-  for (const char *arg : CmdArgs) {
-    if (strncmp(arg, FortranMainLinkFlag, strlen(FortranMainLinkFlag)) == 0)
-      D.Diag(diag::warn_drv_deprecated_custom)
-          << FortranMainLinkFlag
-          << "see the Flang driver documentation for correct usage";
-  }
-
-  // The --whole-archive option needs to be part of the link line to make
-  // sure that the main() function from Fortran_main.a is pulled in by the
-  // linker. However, it shouldn't be used if it's already active.
-  // TODO: Find an equivalent of `--whole-archive` for Darwin and AIX.
-  if (!isWholeArchivePresent(Args) && !TC.getTriple().isMacOSX() &&
-      !TC.getTriple().isOSAIX()) {
-    CmdArgs.push_back("--whole-archive");
-    CmdArgs.push_back(FortranMainLinkFlag);
-    CmdArgs.push_back("--no-whole-archive");
-    return;
-  }
-
-  CmdArgs.push_back(FortranMainLinkFlag);
-}
-
 /// Add Fortran runtime libs
 void tools::addFortranRuntimeLibs(const ToolChain &TC, const ArgList &Args,
                                   llvm::opt::ArgStringList &CmdArgs) {
-  // 1. Link FortranMain
-  // FortranMain depends on FortranRuntime, so needs to be listed first. If
-  // -fno-fortran-main has been passed, skip linking Fortran_main.a
-  if (!Args.hasArg(options::OPT_no_fortran_main))
-    addFortranMain(TC, Args, CmdArgs);
-
-  // 2. Link FortranRuntime and FortranDecimal
+  // Link FortranRuntime and FortranDecimal
   // These are handled earlier on Windows by telling the frontend driver to
   // add the correct libraries to link against as dependents in the object
   // file.
diff --git a/clang/lib/Driver/ToolChains/Flang.cpp b/clang/lib/Driver/ToolChains/Flang.cpp
index abe0b9316760..8955b9fb653c 100644
--- a/clang/lib/Driver/ToolChains/Flang.cpp
+++ b/clang/lib/Driver/ToolChains/Flang.cpp
@@ -15,7 +15,7 @@
 #include "llvm/Frontend/Debug/Options.h"
 #include "llvm/Support/FileSystem.h"
 #include "llvm/Support/Path.h"
-#include "llvm/Support/RISCVISAInfo.h"
+#include "llvm/TargetParser/RISCVISAInfo.h"
 #include "llvm/TargetParser/RISCVTargetParser.h"
 
 #include <cassert>
@@ -282,7 +282,6 @@ static void processVSRuntimeLibrary(const ToolChain &TC, const ArgList &Args,
   assert(TC.getTriple().isKnownWindowsMSVCEnvironment() &&
          "can only add VS runtime library on Windows!");
   // if -fno-fortran-main has been passed, skip linking Fortran_main.a
-  bool LinkFortranMain = !Args.hasArg(options::OPT_no_fortran_main);
   if (TC.getTriple().isKnownWindowsMSVCEnvironment()) {
     CmdArgs.push_back(Args.MakeArgString(
         "--dependent-lib=" + TC.getCompilerRTBasename(Args, "builtins")));
@@ -300,8 +299,6 @@ static void processVSRuntimeLibrary(const ToolChain &TC, const ArgList &Args,
   case options::OPT__SLASH_MT:
     CmdArgs.push_back("-D_MT");
     CmdArgs.push_back("--dependent-lib=libcmt");
-    if (LinkFortranMain)
-      CmdArgs.push_back("--dependent-lib=Fortran_main.static.lib");
     CmdArgs.push_back("--dependent-lib=FortranRuntime.static.lib");
     CmdArgs.push_back("--dependent-lib=FortranDecimal.static.lib");
     break;
@@ -309,8 +306,6 @@ static void processVSRuntimeLibrary(const ToolChain &TC, const ArgList &Args,
     CmdArgs.push_back("-D_MT");
     CmdArgs.push_back("-D_DEBUG");
     CmdArgs.push_back("--dependent-lib=libcmtd");
-    if (LinkFortranMain)
-      CmdArgs.push_back("--dependent-lib=Fortran_main.static_dbg.lib");
     CmdArgs.push_back("--dependent-lib=FortranRuntime.static_dbg.lib");
     CmdArgs.push_back("--dependent-lib=FortranDecimal.static_dbg.lib");
     break;
@@ -318,8 +313,6 @@ static void processVSRuntimeLibrary(const ToolChain &TC, const ArgList &Args,
     CmdArgs.push_back("-D_MT");
     CmdArgs.push_back("-D_DLL");
     CmdArgs.push_back("--dependent-lib=msvcrt");
-    if (LinkFortranMain)
-      CmdArgs.push_back("--dependent-lib=Fortran_main.dynamic.lib");
     CmdArgs.push_back("--dependent-lib=FortranRuntime.dynamic.lib");
     CmdArgs.push_back("--dependent-lib=FortranDecimal.dynamic.lib");
     break;
@@ -328,8 +321,6 @@ static void processVSRuntimeLibrary(const ToolChain &TC, const ArgList &Args,
     CmdArgs.push_back("-D_DEBUG");
     CmdArgs.push_back("-D_DLL");
     CmdArgs.push_back("--dependent-lib=msvcrtd");
-    if (LinkFortranMain)
-      CmdArgs.push_back("--dependent-lib=Fortran_main.dynamic_dbg.lib");
     CmdArgs.push_back("--dependent-lib=FortranRuntime.dynamic_dbg.lib");
     CmdArgs.push_back("--dependent-lib=FortranDecimal.dynamic_dbg.lib");
     break;
diff --git a/clang/lib/Driver/ToolChains/Gnu.cpp b/clang/lib/Driver/ToolChains/Gnu.cpp
index dedbfac6cb25..9849c59685cc 100644
--- a/clang/lib/Driver/ToolChains/Gnu.cpp
+++ b/clang/lib/Driver/ToolChains/Gnu.cpp
@@ -30,8 +30,8 @@
 #include "llvm/Option/ArgList.h"
 #include "llvm/Support/CodeGen.h"
 #include "llvm/Support/Path.h"
-#include "llvm/Support/RISCVISAInfo.h"
 #include "llvm/Support/VirtualFileSystem.h"
+#include "llvm/TargetParser/RISCVISAInfo.h"
 #include "llvm/TargetParser/TargetParser.h"
 #include <system_error>
 
@@ -1796,9 +1796,7 @@ selectRISCVMultilib(const MultilibSet &RISCVMultilibSet, StringRef Arch,
       }
       auto &MLConfigISAInfo = *MLConfigParseResult;
 
-      const llvm::RISCVISAInfo::OrderedExtensionMap &MLConfigArchExts =
-          MLConfigISAInfo->getExtensions();
-      for (auto MLConfigArchExt : MLConfigArchExts) {
+      for (auto &MLConfigArchExt : MLConfigISAInfo->getExtensions()) {
         auto ExtName = MLConfigArchExt.first;
         NewMultilib.flag(Twine("-", ExtName).str());
 
diff --git a/clang/lib/Driver/ToolChains/Linux.cpp b/clang/lib/Driver/ToolChains/Linux.cpp
index fb65881061ef..db2c20d7b461 100644
--- a/clang/lib/Driver/ToolChains/Linux.cpp
+++ b/clang/lib/Driver/ToolChains/Linux.cpp
@@ -842,25 +842,6 @@ void Linux::addProfileRTLibs(const llvm::opt::ArgList &Args,
   ToolChain::addProfileRTLibs(Args, CmdArgs);
 }
 
-llvm::DenormalMode
-Linux::getDefaultDenormalModeForType(const llvm::opt::ArgList &DriverArgs,
-                                     const JobAction &JA,
-                                     const llvm::fltSemantics *FPType) const {
-  switch (getTriple().getArch()) {
-  case llvm::Triple::x86:
-  case llvm::Triple::x86_64: {
-    std::string Unused;
-    // DAZ and FTZ are turned on in crtfastmath.o
-    if (!DriverArgs.hasArg(options::OPT_nostdlib, options::OPT_nostartfiles) &&
-        isFastMathRuntimeAvailable(DriverArgs, Unused))
-      return llvm::DenormalMode::getPreserveSign();
-    return llvm::DenormalMode::getIEEE();
-  }
-  default:
-    return llvm::DenormalMode::getIEEE();
-  }
-}
-
 void Linux::addExtraOpts(llvm::opt::ArgStringList &CmdArgs) const {
   for (const auto &Opt : ExtraOpts)
     CmdArgs.push_back(Opt.c_str());
diff --git a/clang/lib/Driver/ToolChains/Linux.h b/clang/lib/Driver/ToolChains/Linux.h
index 524391743090..2d9e674e50a6 100644
--- a/clang/lib/Driver/ToolChains/Linux.h
+++ b/clang/lib/Driver/ToolChains/Linux.h
@@ -59,10 +59,6 @@ public:
 
   std::vector<std::string> ExtraOpts;
 
-  llvm::DenormalMode getDefaultDenormalModeForType(
-      const llvm::opt::ArgList &DriverArgs, const JobAction &JA,
-      const llvm::fltSemantics *FPType = nullptr) const override;
-
   const char *getDefaultLinker() const override;
 
 protected:
diff --git a/clang/lib/ExtractAPI/API.cpp b/clang/lib/ExtractAPI/API.cpp
index 5a62c5deb240..96bef967d859 100644
--- a/clang/lib/ExtractAPI/API.cpp
+++ b/clang/lib/ExtractAPI/API.cpp
@@ -54,7 +54,28 @@ RecordContext *APIRecord::castToRecordContext(const APIRecord *Record) {
   }
 }
 
+bool RecordContext::IsWellFormed() const {
+  // Check that First and Last are both null or both non-null.
+  return (First == nullptr) == (Last == nullptr);
+}
+
+void RecordContext::stealRecordChain(RecordContext &Other) {
+  assert(IsWellFormed());
+  // If we don't have an empty chain append Other's chain into ours.
+  if (First)
+    Last->NextInContext = Other.First;
+  else
+    First = Other.First;
+
+  Last = Other.Last;
+
+  // Delete Other's chain to ensure we don't accidentally traverse it.
+  Other.First = nullptr;
+  Other.Last = nullptr;
+}
+
 void RecordContext::addToRecordChain(APIRecord *Record) const {
+  assert(IsWellFormed());
   if (!First) {
     First = Record;
     Last = Record;
@@ -95,6 +116,7 @@ SymbolReference APISet::createSymbolReference(StringRef Name, StringRef USR,
 }
 
 APIRecord::~APIRecord() {}
+TagRecord::~TagRecord() {}
 RecordRecord::~RecordRecord() {}
 RecordFieldRecord::~RecordFieldRecord() {}
 ObjCContainerRecord::~ObjCContainerRecord() {}
diff --git a/clang/lib/ExtractAPI/DeclarationFragments.cpp b/clang/lib/ExtractAPI/DeclarationFragments.cpp
index 0a243120b7c0..9bf7950888db 100644
--- a/clang/lib/ExtractAPI/DeclarationFragments.cpp
+++ b/clang/lib/ExtractAPI/DeclarationFragments.cpp
@@ -396,7 +396,8 @@ DeclarationFragments DeclarationFragmentsBuilder::getFragmentsForType(
     const TagDecl *Decl = TagTy->getDecl();
     // Anonymous decl, skip this fragment.
     if (Decl->getName().empty())
-      return Fragments;
+      return Fragments.append("{ ... }",
+                              DeclarationFragments::FragmentKind::Text);
     SmallString<128> TagUSR;
     clang::index::generateUSRForDecl(Decl, TagUSR);
     return Fragments.append(Decl->getName(),
@@ -743,11 +744,16 @@ DeclarationFragmentsBuilder::getFragmentsForEnum(const EnumDecl *EnumDecl) {
 
   QualType IntegerType = EnumDecl->getIntegerType();
   if (!IntegerType.isNull())
-    Fragments.append(": ", DeclarationFragments::FragmentKind::Text)
+    Fragments.appendSpace()
+        .append(": ", DeclarationFragments::FragmentKind::Text)
         .append(
             getFragmentsForType(IntegerType, EnumDecl->getASTContext(), After))
         .append(std::move(After));
 
+  if (EnumDecl->getName().empty())
+    Fragments.appendSpace().append("{ ... }",
+                                   DeclarationFragments::FragmentKind::Text);
+
   return Fragments.appendSemicolon();
 }
 
@@ -778,9 +784,12 @@ DeclarationFragments DeclarationFragmentsBuilder::getFragmentsForRecordDecl(
   else
     Fragments.append("struct", DeclarationFragments::FragmentKind::Keyword);
 
+  Fragments.appendSpace();
   if (!Record->getName().empty())
-    Fragments.appendSpace().append(
-        Record->getName(), DeclarationFragments::FragmentKind::Identifier);
+    Fragments.append(Record->getName(),
+                     DeclarationFragments::FragmentKind::Identifier);
+  else
+    Fragments.append("{ ... }", DeclarationFragments::FragmentKind::Text);
 
   return Fragments.appendSemicolon();
 }
diff --git a/clang/lib/ExtractAPI/Serialization/SymbolGraphSerializer.cpp b/clang/lib/ExtractAPI/Serialization/SymbolGraphSerializer.cpp
index 57f966c8b2be..34278b5d40c4 100644
--- a/clang/lib/ExtractAPI/Serialization/SymbolGraphSerializer.cpp
+++ b/clang/lib/ExtractAPI/Serialization/SymbolGraphSerializer.cpp
@@ -164,27 +164,29 @@ std::optional<Array> serializeAvailability(const AvailabilityInfo &Avail) {
   if (Avail.isDefault())
     return std::nullopt;
 
-  Object Availability;
   Array AvailabilityArray;
-  Availability["domain"] = Avail.Domain;
-  serializeObject(Availability, "introduced",
-                  serializeSemanticVersion(Avail.Introduced));
-  serializeObject(Availability, "deprecated",
-                  serializeSemanticVersion(Avail.Deprecated));
-  serializeObject(Availability, "obsoleted",
-                  serializeSemanticVersion(Avail.Obsoleted));
+
   if (Avail.isUnconditionallyDeprecated()) {
     Object UnconditionallyDeprecated;
     UnconditionallyDeprecated["domain"] = "*";
     UnconditionallyDeprecated["isUnconditionallyDeprecated"] = true;
     AvailabilityArray.emplace_back(std::move(UnconditionallyDeprecated));
   }
-  if (Avail.isUnconditionallyUnavailable()) {
-    Object UnconditionallyUnavailable;
-    UnconditionallyUnavailable["domain"] = "*";
-    UnconditionallyUnavailable["isUnconditionallyUnavailable"] = true;
-    AvailabilityArray.emplace_back(std::move(UnconditionallyUnavailable));
+  Object Availability;
+
+  Availability["domain"] = Avail.Domain;
+
+  if (Avail.isUnavailable()) {
+    Availability["isUnconditionallyUnavailable"] = true;
+  } else {
+    serializeObject(Availability, "introduced",
+                    serializeSemanticVersion(Avail.Introduced));
+    serializeObject(Availability, "deprecated",
+                    serializeSemanticVersion(Avail.Deprecated));
+    serializeObject(Availability, "obsoleted",
+                    serializeSemanticVersion(Avail.Obsoleted));
   }
+
   AvailabilityArray.emplace_back(std::move(Availability));
   return AvailabilityArray;
 }
@@ -665,6 +667,14 @@ bool SymbolGraphSerializer::shouldSkip(const APIRecord *Record) const {
   if (Record->Availability.isUnconditionallyUnavailable())
     return true;
 
+  // Filter out symbols without a name as we can generate correct symbol graphs
+  // for them. In practice these are anonymous record types that aren't attached
+  // to a declaration.
+  if (auto *Tag = dyn_cast<TagRecord>(Record)) {
+    if (Tag->IsEmbeddedInVarDeclarator)
+      return true;
+  }
+
   // Filter out symbols prefixed with an underscored as they are understood to
   // be symbols clients should not use.
   if (Record->Name.starts_with("_"))
diff --git a/clang/lib/Format/Format.cpp b/clang/lib/Format/Format.cpp
index ccb2c9190e2e..c8d8ec3afbd9 100644
--- a/clang/lib/Format/Format.cpp
+++ b/clang/lib/Format/Format.cpp
@@ -807,7 +807,6 @@ template <> struct MappingTraits<FormatStyle> {
         FormatStyle PredefinedStyle;
         if (getPredefinedStyle(StyleName, Style.Language, &PredefinedStyle) &&
             Style == PredefinedStyle) {
-          IO.mapOptional("# BasedOnStyle", StyleName);
           BasedOnStyle = StyleName;
           break;
         }
@@ -3117,6 +3116,7 @@ static void sortCppIncludes(const FormatStyle &Style,
     return;
   }
 
+  const auto OldCursor = Cursor ? *Cursor : 0;
   std::string result;
   for (unsigned Index : Indices) {
     if (!result.empty()) {
@@ -3140,6 +3140,8 @@ static void sortCppIncludes(const FormatStyle &Style,
   // the entire range of blocks. Otherwise, no replacement is generated.
   if (replaceCRLF(result) == replaceCRLF(std::string(Code.substr(
                                  IncludesBeginOffset, IncludesBlockSize)))) {
+    if (Cursor)
+      *Cursor = OldCursor;
     return;
   }
 
diff --git a/clang/lib/Format/FormatToken.h b/clang/lib/Format/FormatToken.h
index f651e6228c20..28b6488e54a4 100644
--- a/clang/lib/Format/FormatToken.h
+++ b/clang/lib/Format/FormatToken.h
@@ -1623,10 +1623,10 @@ struct AdditionalKeywords {
   IdentifierInfo *kw_then;
 
   /// Returns \c true if \p Tok is a keyword or an identifier.
-  bool isWordLike(const FormatToken &Tok) const {
+  bool isWordLike(const FormatToken &Tok, bool IsVerilog = true) const {
     // getIdentifierinfo returns non-null for keywords as well as identifiers.
     return Tok.Tok.getIdentifierInfo() &&
-           !Tok.isOneOf(kw_verilogHash, kw_verilogHashHash, kw_apostrophe);
+           (!IsVerilog || !isVerilogKeywordSymbol(Tok));
   }
 
   /// Returns \c true if \p Tok is a true JavaScript identifier, returns
@@ -1755,6 +1755,10 @@ struct AdditionalKeywords {
     }
   }
 
+  bool isVerilogKeywordSymbol(const FormatToken &Tok) const {
+    return Tok.isOneOf(kw_verilogHash, kw_verilogHashHash, kw_apostrophe);
+  }
+
   bool isVerilogWordOperator(const FormatToken &Tok) const {
     return Tok.isOneOf(kw_before, kw_intersect, kw_dist, kw_iff, kw_inside,
                        kw_with);
diff --git a/clang/lib/Format/TokenAnnotator.cpp b/clang/lib/Format/TokenAnnotator.cpp
index cdfb4256e41d..d366ae2080bc 100644
--- a/clang/lib/Format/TokenAnnotator.cpp
+++ b/clang/lib/Format/TokenAnnotator.cpp
@@ -4780,9 +4780,14 @@ bool TokenAnnotator::spaceRequiredBefore(const AnnotatedLine &Line,
   if (Left.Finalized)
     return Right.hasWhitespaceBefore();
 
+  const bool IsVerilog = Style.isVerilog();
+  assert(!IsVerilog || !IsCpp);
+
   // Never ever merge two words.
-  if (Keywords.isWordLike(Right) && Keywords.isWordLike(Left))
+  if (Keywords.isWordLike(Right, IsVerilog) &&
+      Keywords.isWordLike(Left, IsVerilog)) {
     return true;
+  }
 
   // Leave a space between * and /* to avoid C4138 `comment end` found outside
   // of comment.
@@ -4834,10 +4839,8 @@ bool TokenAnnotator::spaceRequiredBefore(const AnnotatedLine &Line,
         Right.is(TT_TemplateOpener)) {
       return true;
     }
-    if (Left.is(tok::identifier) && Right.is(tok::numeric_constant) &&
-        Right.TokenText[0] == '.') {
-      return false;
-    }
+    if (Left.Tok.getIdentifierInfo() && Right.is(tok::numeric_constant))
+      return Right.TokenText[0] != '.';
   } else if (Style.isProto()) {
     if (Right.is(tok::period) &&
         Left.isOneOf(Keywords.kw_optional, Keywords.kw_required,
@@ -5065,12 +5068,10 @@ bool TokenAnnotator::spaceRequiredBefore(const AnnotatedLine &Line,
         Right.is(TT_TemplateOpener)) {
       return true;
     }
-  } else if (Style.isVerilog()) {
+  } else if (IsVerilog) {
     // An escaped identifier ends with whitespace.
-    if (Style.isVerilog() && Left.is(tok::identifier) &&
-        Left.TokenText[0] == '\\') {
+    if (Left.is(tok::identifier) && Left.TokenText[0] == '\\')
       return true;
-    }
     // Add space between things in a primitive's state table unless in a
     // transition like `(0?)`.
     if ((Left.is(TT_VerilogTableItem) &&
@@ -5266,21 +5267,11 @@ bool TokenAnnotator::spaceRequiredBefore(const AnnotatedLine &Line,
     return true;
   }
   if (Left.is(TT_UnaryOperator)) {
-    if (Right.isNot(tok::l_paren)) {
-      // The alternative operators for ~ and ! are "compl" and "not".
-      // If they are used instead, we do not want to combine them with
-      // the token to the right, unless that is a left paren.
-      if (Left.is(tok::exclaim) && Left.TokenText == "not")
-        return true;
-      if (Left.is(tok::tilde) && Left.TokenText == "compl")
-        return true;
-      // Lambda captures allow for a lone &, so "&]" needs to be properly
-      // handled.
-      if (Left.is(tok::amp) && Right.is(tok::r_square))
-        return Style.SpacesInSquareBrackets;
-    }
-    return (Style.SpaceAfterLogicalNot && Left.is(tok::exclaim)) ||
-           Right.is(TT_BinaryOperator);
+    // Lambda captures allow for a lone &, so "&]" needs to be properly
+    // handled.
+    if (Left.is(tok::amp) && Right.is(tok::r_square))
+      return Style.SpacesInSquareBrackets;
+    return Style.SpaceAfterLogicalNot && Left.is(tok::exclaim);
   }
 
   // If the next token is a binary operator or a selector name, we have
diff --git a/clang/lib/Format/UnwrappedLineParser.cpp b/clang/lib/Format/UnwrappedLineParser.cpp
index 603268f771ac..3a263955a6a8 100644
--- a/clang/lib/Format/UnwrappedLineParser.cpp
+++ b/clang/lib/Format/UnwrappedLineParser.cpp
@@ -534,11 +534,11 @@ void UnwrappedLineParser::calculateBraceTypes(bool ExpectClassBody) {
     case tok::r_brace:
       if (LBraceStack.empty())
         break;
-      if (LBraceStack.back().Tok->is(BK_Unknown)) {
+      if (auto *LBrace = LBraceStack.back().Tok; LBrace->is(BK_Unknown)) {
         bool ProbablyBracedList = false;
         if (Style.Language == FormatStyle::LK_Proto) {
           ProbablyBracedList = NextTok->isOneOf(tok::comma, tok::r_square);
-        } else {
+        } else if (LBrace->isNot(TT_EnumLBrace)) {
           // Using OriginalColumn to distinguish between ObjC methods and
           // binary operators is a bit hacky.
           bool NextIsObjCMethod = NextTok->isOneOf(tok::plus, tok::minus) &&
@@ -552,7 +552,7 @@ void UnwrappedLineParser::calculateBraceTypes(bool ExpectClassBody) {
 
           // If we already marked the opening brace as braced list, the closing
           // must also be part of it.
-          ProbablyBracedList = LBraceStack.back().Tok->is(TT_BracedListLBrace);
+          ProbablyBracedList = LBrace->is(TT_BracedListLBrace);
 
           ProbablyBracedList = ProbablyBracedList ||
                                (Style.isJavaScript() &&
@@ -608,13 +608,9 @@ void UnwrappedLineParser::calculateBraceTypes(bool ExpectClassBody) {
             ProbablyBracedList = true;
           }
         }
-        if (ProbablyBracedList) {
-          Tok->setBlockKind(BK_BracedInit);
-          LBraceStack.back().Tok->setBlockKind(BK_BracedInit);
-        } else {
-          Tok->setBlockKind(BK_Block);
-          LBraceStack.back().Tok->setBlockKind(BK_Block);
-        }
+        const auto BlockKind = ProbablyBracedList ? BK_BracedInit : BK_Block;
+        Tok->setBlockKind(BlockKind);
+        LBrace->setBlockKind(BlockKind);
       }
       LBraceStack.pop_back();
       break;
@@ -819,8 +815,11 @@ FormatToken *UnwrappedLineParser::parseBlock(bool MustBeDeclaration,
     return IfLBrace;
   }
 
-  if (FormatTok->is(tok::r_brace) && Tok->is(TT_NamespaceLBrace))
-    FormatTok->setFinalizedType(TT_NamespaceRBrace);
+  if (FormatTok->is(tok::r_brace)) {
+    FormatTok->setBlockKind(BK_Block);
+    if (Tok->is(TT_NamespaceLBrace))
+      FormatTok->setFinalizedType(TT_NamespaceRBrace);
+  }
 
   const bool IsFunctionRBrace =
       FormatTok->is(tok::r_brace) && Tok->is(TT_FunctionLBrace);
@@ -2415,6 +2414,7 @@ bool UnwrappedLineParser::tryToParseChildBlock() {
 }
 
 bool UnwrappedLineParser::parseBracedList(bool IsAngleBracket, bool IsEnum) {
+  assert(!IsAngleBracket || !IsEnum);
   bool HasError = false;
 
   // FIXME: Once we have an expression parser in the UnwrappedLineParser,
@@ -2437,8 +2437,11 @@ bool UnwrappedLineParser::parseBracedList(bool IsAngleBracket, bool IsEnum) {
       }
     }
     if (FormatTok->is(IsAngleBracket ? tok::greater : tok::r_brace)) {
-      if (IsEnum && !Style.AllowShortEnumsOnASingleLine)
-        addUnwrappedLine();
+      if (IsEnum) {
+        FormatTok->setBlockKind(BK_Block);
+        if (!Style.AllowShortEnumsOnASingleLine)
+          addUnwrappedLine();
+      }
       nextToken();
       return !HasError;
     }
@@ -3910,6 +3913,8 @@ void UnwrappedLineParser::parseRecord(bool ParseAsExpr) {
   const FormatToken &InitialToken = *FormatTok;
   nextToken();
 
+  const FormatToken *ClassName = nullptr;
+  bool IsDerived = false;
   auto IsNonMacroIdentifier = [](const FormatToken *Tok) {
     return Tok->is(tok::identifier) && Tok->TokenText != Tok->TokenText.upper();
   };
@@ -3934,15 +3939,35 @@ void UnwrappedLineParser::parseRecord(bool ParseAsExpr) {
     }
     if (FormatTok->is(tok::l_square) && handleCppAttributes())
       continue;
+    const auto *Previous = FormatTok;
     nextToken();
-    // We can have macros in between 'class' and the class name.
-    if (!IsNonMacroIdentifier(FormatTok->Previous) &&
-        FormatTok->is(tok::l_paren)) {
-      parseParens();
+    switch (FormatTok->Tok.getKind()) {
+    case tok::l_paren:
+      // We can have macros in between 'class' and the class name.
+      if (!IsNonMacroIdentifier(Previous))
+        parseParens();
+      break;
+    case tok::coloncolon:
+      break;
+    default:
+      if (!ClassName && Previous->is(tok::identifier))
+        ClassName = Previous;
     }
   }
 
+  auto IsListInitialization = [&] {
+    if (!ClassName || IsDerived)
+      return false;
+    assert(FormatTok->is(tok::l_brace));
+    const auto *Prev = FormatTok->getPreviousNonComment();
+    assert(Prev);
+    return Prev != ClassName && Prev->is(tok::identifier) &&
+           Prev->isNot(Keywords.kw_final) && tryToParseBracedList();
+  };
+
   if (FormatTok->isOneOf(tok::colon, tok::less)) {
+    if (FormatTok->is(tok::colon))
+      IsDerived = true;
     int AngleNestingLevel = 0;
     do {
       if (FormatTok->is(tok::less))
@@ -3955,6 +3980,8 @@ void UnwrappedLineParser::parseRecord(bool ParseAsExpr) {
         break;
       }
       if (FormatTok->is(tok::l_brace)) {
+        if (AngleNestingLevel == 0 && IsListInitialization())
+          return;
         calculateBraceTypes(/*ExpectClassBody=*/true);
         if (!tryToParseBracedList())
           break;
@@ -3999,6 +4026,8 @@ void UnwrappedLineParser::parseRecord(bool ParseAsExpr) {
     }
   };
   if (FormatTok->is(tok::l_brace)) {
+    if (IsListInitialization())
+      return;
     auto [OpenBraceType, ClosingBraceType] = GetBraceTypes(InitialToken);
     FormatTok->setFinalizedType(OpenBraceType);
     if (ParseAsExpr) {
diff --git a/clang/lib/Format/WhitespaceManager.cpp b/clang/lib/Format/WhitespaceManager.cpp
index 4f822807dd98..44fd807ec27e 100644
--- a/clang/lib/Format/WhitespaceManager.cpp
+++ b/clang/lib/Format/WhitespaceManager.cpp
@@ -128,11 +128,14 @@ const tooling::Replacements &WhitespaceManager::generateReplacements() {
 void WhitespaceManager::calculateLineBreakInformation() {
   Changes[0].PreviousEndOfTokenColumn = 0;
   Change *LastOutsideTokenChange = &Changes[0];
-  for (unsigned i = 1, e = Changes.size(); i != e; ++i) {
+  for (unsigned I = 1, e = Changes.size(); I != e; ++I) {
+    auto &C = Changes[I];
+    auto &P = Changes[I - 1];
+    auto &PrevTokLength = P.TokenLength;
     SourceLocation OriginalWhitespaceStart =
-        Changes[i].OriginalWhitespaceRange.getBegin();
+        C.OriginalWhitespaceRange.getBegin();
     SourceLocation PreviousOriginalWhitespaceEnd =
-        Changes[i - 1].OriginalWhitespaceRange.getEnd();
+        P.OriginalWhitespaceRange.getEnd();
     unsigned OriginalWhitespaceStartOffset =
         SourceMgr.getFileOffset(OriginalWhitespaceStart);
     unsigned PreviousOriginalWhitespaceEndOffset =
@@ -167,31 +170,28 @@ void WhitespaceManager::calculateLineBreakInformation() {
     // line of the token.
     auto NewlinePos = Text.find_first_of('\n');
     if (NewlinePos == StringRef::npos) {
-      Changes[i - 1].TokenLength = OriginalWhitespaceStartOffset -
-                                   PreviousOriginalWhitespaceEndOffset +
-                                   Changes[i].PreviousLinePostfix.size() +
-                                   Changes[i - 1].CurrentLinePrefix.size();
+      PrevTokLength = OriginalWhitespaceStartOffset -
+                      PreviousOriginalWhitespaceEndOffset +
+                      C.PreviousLinePostfix.size() + P.CurrentLinePrefix.size();
+      if (!P.IsInsideToken)
+        PrevTokLength = std::min(PrevTokLength, P.Tok->ColumnWidth);
     } else {
-      Changes[i - 1].TokenLength =
-          NewlinePos + Changes[i - 1].CurrentLinePrefix.size();
+      PrevTokLength = NewlinePos + P.CurrentLinePrefix.size();
     }
 
     // If there are multiple changes in this token, sum up all the changes until
     // the end of the line.
-    if (Changes[i - 1].IsInsideToken && Changes[i - 1].NewlinesBefore == 0) {
-      LastOutsideTokenChange->TokenLength +=
-          Changes[i - 1].TokenLength + Changes[i - 1].Spaces;
-    } else {
-      LastOutsideTokenChange = &Changes[i - 1];
-    }
+    if (P.IsInsideToken && P.NewlinesBefore == 0)
+      LastOutsideTokenChange->TokenLength += PrevTokLength + P.Spaces;
+    else
+      LastOutsideTokenChange = &P;
 
-    Changes[i].PreviousEndOfTokenColumn =
-        Changes[i - 1].StartOfTokenColumn + Changes[i - 1].TokenLength;
+    C.PreviousEndOfTokenColumn = P.StartOfTokenColumn + PrevTokLength;
 
-    Changes[i - 1].IsTrailingComment =
-        (Changes[i].NewlinesBefore > 0 || Changes[i].Tok->is(tok::eof) ||
-         (Changes[i].IsInsideToken && Changes[i].Tok->is(tok::comment))) &&
-        Changes[i - 1].Tok->is(tok::comment) &&
+    P.IsTrailingComment =
+        (C.NewlinesBefore > 0 || C.Tok->is(tok::eof) ||
+         (C.IsInsideToken && C.Tok->is(tok::comment))) &&
+        P.Tok->is(tok::comment) &&
         // FIXME: This is a dirty hack. The problem is that
         // BreakableLineCommentSection does comment reflow changes and here is
         // the aligning of trailing comments. Consider the case where we reflow
diff --git a/clang/lib/Frontend/ASTUnit.cpp b/clang/lib/Frontend/ASTUnit.cpp
index 3610a08831e7..1b93588553a2 100644
--- a/clang/lib/Frontend/ASTUnit.cpp
+++ b/clang/lib/Frontend/ASTUnit.cpp
@@ -1067,7 +1067,7 @@ public:
 
   std::vector<Decl *> takeTopLevelDecls() { return std::move(TopLevelDecls); }
 
-  std::vector<serialization::DeclID> takeTopLevelDeclIDs() {
+  std::vector<LocalDeclID> takeTopLevelDeclIDs() {
     return std::move(TopLevelDeclIDs);
   }
 
@@ -1101,7 +1101,7 @@ public:
 private:
   unsigned Hash = 0;
   std::vector<Decl *> TopLevelDecls;
-  std::vector<serialization::DeclID> TopLevelDeclIDs;
+  std::vector<LocalDeclID> TopLevelDeclIDs;
   llvm::SmallVector<ASTUnit::StandaloneDiagnostic, 4> PreambleDiags;
 };
 
@@ -1467,11 +1467,12 @@ void ASTUnit::RealizeTopLevelDeclsFromPreamble() {
 
   std::vector<Decl *> Resolved;
   Resolved.reserve(TopLevelDeclsInPreamble.size());
-  ExternalASTSource &Source = *getASTContext().getExternalSource();
+  // The module file of the preamble.
+  serialization::ModuleFile &MF = Reader->getModuleManager().getPrimaryModule();
   for (const auto TopLevelDecl : TopLevelDeclsInPreamble) {
     // Resolve the declaration ID to an actual declaration, possibly
     // deserializing the declaration in the process.
-    if (Decl *D = Source.GetExternalDecl(TopLevelDecl))
+    if (Decl *D = Reader->GetDecl(Reader->getGlobalDeclID(MF, TopLevelDecl)))
       Resolved.push_back(D);
   }
   TopLevelDeclsInPreamble.clear();
diff --git a/clang/lib/Frontend/CompilerInstance.cpp b/clang/lib/Frontend/CompilerInstance.cpp
index 6e3baf838644..66a45b888f15 100644
--- a/clang/lib/Frontend/CompilerInstance.cpp
+++ b/clang/lib/Frontend/CompilerInstance.cpp
@@ -1293,6 +1293,10 @@ compileModuleImpl(CompilerInstance &ImportingInstance, SourceLocation ImportLoc,
                                             diag::remark_module_build_done)
     << ModuleName;
 
+  // Propagate the statistics to the parent FileManager.
+  if (!FrontendOpts.ModulesShareFileManager)
+    ImportingInstance.getFileManager().AddStats(Instance.getFileManager());
+
   if (Crashed) {
     // Clear the ASTConsumer if it hasn't been already, in case it owns streams
     // that must be closed before clearing output files.
diff --git a/clang/lib/Frontend/FrontendAction.cpp b/clang/lib/Frontend/FrontendAction.cpp
index b7c9967316f0..9ae7664b4b49 100644
--- a/clang/lib/Frontend/FrontendAction.cpp
+++ b/clang/lib/Frontend/FrontendAction.cpp
@@ -80,7 +80,7 @@ public:
     if (Previous)
       Previous->TypeRead(Idx, T);
   }
-  void DeclRead(serialization::DeclID ID, const Decl *D) override {
+  void DeclRead(GlobalDeclID ID, const Decl *D) override {
     if (Previous)
       Previous->DeclRead(ID, D);
   }
@@ -102,7 +102,7 @@ public:
                                    bool DeletePrevious)
       : DelegatingDeserializationListener(Previous, DeletePrevious) {}
 
-  void DeclRead(serialization::DeclID ID, const Decl *D) override {
+  void DeclRead(GlobalDeclID ID, const Decl *D) override {
     llvm::outs() << "PCH DECL: " << D->getDeclKindName();
     if (const NamedDecl *ND = dyn_cast<NamedDecl>(D)) {
       llvm::outs() << " - ";
@@ -128,7 +128,7 @@ public:
       : DelegatingDeserializationListener(Previous, DeletePrevious), Ctx(Ctx),
         NamesToCheck(NamesToCheck) {}
 
-  void DeclRead(serialization::DeclID ID, const Decl *D) override {
+  void DeclRead(GlobalDeclID ID, const Decl *D) override {
     if (const NamedDecl *ND = dyn_cast<NamedDecl>(D))
       if (NamesToCheck.find(ND->getNameAsString()) != NamesToCheck.end()) {
         unsigned DiagID
@@ -757,8 +757,11 @@ bool FrontendAction::BeginSourceFile(CompilerInstance &CI,
 
   // IR files bypass the rest of initialization.
   if (Input.getKind().getLanguage() == Language::LLVM_IR) {
-    assert(hasIRSupport() &&
-           "This action does not have IR file support!");
+    if (!hasIRSupport()) {
+      CI.getDiagnostics().Report(diag::err_ast_action_on_llvm_ir)
+          << Input.getFile();
+      return false;
+    }
 
     // Inform the diagnostic client we are processing a source file.
     CI.getDiagnosticClient().BeginSourceFile(CI.getLangOpts(), nullptr);
diff --git a/clang/lib/Frontend/InitPreprocessor.cpp b/clang/lib/Frontend/InitPreprocessor.cpp
index 4f44c3b7b89d..c1d209466ffe 100644
--- a/clang/lib/Frontend/InitPreprocessor.cpp
+++ b/clang/lib/Frontend/InitPreprocessor.cpp
@@ -389,8 +389,7 @@ static void InitializeStandardPredefinedMacros(const TargetInfo &TI,
                         Twine((unsigned)LangOpts.getHLSLVersion()));
 
     if (LangOpts.NativeHalfType)
-      Builder.defineMacro("__HLSL_ENABLE_16_BIT",
-                          Twine((unsigned)LangOpts.getHLSLVersion()));
+      Builder.defineMacro("__HLSL_ENABLE_16_BIT", "1");
 
     // Shader target information
     // "enums" for shader stages
@@ -704,7 +703,7 @@ static void InitializeCPlusPlusFeatureTestMacros(const LangOptions &LangOpts,
     Builder.defineMacro("__cpp_nested_namespace_definitions", "201411L");
     Builder.defineMacro("__cpp_variadic_using", "201611L");
     Builder.defineMacro("__cpp_aggregate_bases", "201603L");
-    Builder.defineMacro("__cpp_structured_bindings", "201606L");
+    Builder.defineMacro("__cpp_structured_bindings", "202403L");
     Builder.defineMacro("__cpp_nontype_template_args",
                         "201411L"); // (not latest)
     Builder.defineMacro("__cpp_fold_expressions", "201603L");
@@ -1309,6 +1308,16 @@ static void InitializePredefinedMacros(const TargetInfo &TI,
     Builder.defineMacro("__GCC_ATOMIC_TEST_AND_SET_TRUEVAL", "1");
   }
 
+  // GCC defines these macros in both C and C++ modes despite them being needed
+  // mostly for STL implementations in C++.
+  auto [Destructive, Constructive] = TI.hardwareInterferenceSizes();
+  Builder.defineMacro("__GCC_DESTRUCTIVE_SIZE", Twine(Destructive));
+  Builder.defineMacro("__GCC_CONSTRUCTIVE_SIZE", Twine(Constructive));
+  // We need to use push_macro to allow users to redefine these macros from the
+  // command line with -D and not issue a -Wmacro-redefined warning.
+  Builder.append("#pragma push_macro(\"__GCC_DESTRUCTIVE_SIZE\")");
+  Builder.append("#pragma push_macro(\"__GCC_CONSTRUCTIVE_SIZE\")");
+
   auto addLockFreeMacros = [&](const llvm::Twine &Prefix) {
     // Used by libc++ and libstdc++ to implement ATOMIC_<foo>_LOCK_FREE.
 #define DEFINE_LOCK_FREE_MACRO(TYPE, Type)                                     \
diff --git a/clang/lib/Frontend/MultiplexConsumer.cpp b/clang/lib/Frontend/MultiplexConsumer.cpp
index 744ea70cc24d..c74bfd86195f 100644
--- a/clang/lib/Frontend/MultiplexConsumer.cpp
+++ b/clang/lib/Frontend/MultiplexConsumer.cpp
@@ -52,8 +52,8 @@ void MultiplexASTDeserializationListener::TypeRead(
     Listeners[i]->TypeRead(Idx, T);
 }
 
-void MultiplexASTDeserializationListener::DeclRead(
-    serialization::DeclID ID, const Decl *D) {
+void MultiplexASTDeserializationListener::DeclRead(GlobalDeclID ID,
+                                                   const Decl *D) {
   for (size_t i = 0, e = Listeners.size(); i != e; ++i)
     Listeners[i]->DeclRead(ID, D);
 }
diff --git a/clang/lib/Headers/cpuid.h b/clang/lib/Headers/cpuid.h
index 0bb9912b465f..bb7692efb78f 100644
--- a/clang/lib/Headers/cpuid.h
+++ b/clang/lib/Headers/cpuid.h
@@ -10,7 +10,7 @@
 #ifndef __CPUID_H
 #define __CPUID_H
 
-#if !(__x86_64__ || __i386__)
+#if !defined(__x86_64__) && !defined(__i386__)
 #error this header is for x86 only
 #endif
 
@@ -256,7 +256,7 @@
 #define bit_AVX10_256   0x00020000
 #define bit_AVX10_512   0x00040000
 
-#if __i386__
+#ifdef __i386__
 #define __cpuid(__leaf, __eax, __ebx, __ecx, __edx) \
     __asm("cpuid" : "=a"(__eax), "=b" (__ebx), "=c"(__ecx), "=d"(__edx) \
                   : "0"(__leaf))
@@ -285,7 +285,7 @@ static __inline unsigned int __get_cpuid_max (unsigned int __leaf,
                                               unsigned int *__sig)
 {
     unsigned int __eax, __ebx, __ecx, __edx;
-#if __i386__
+#ifdef __i386__
     int __cpuid_supported;
 
     __asm("  pushfl\n"
diff --git a/clang/lib/Lex/HeaderSearch.cpp b/clang/lib/Lex/HeaderSearch.cpp
index 0632882b2961..574723b33866 100644
--- a/clang/lib/Lex/HeaderSearch.cpp
+++ b/clang/lib/Lex/HeaderSearch.cpp
@@ -1574,6 +1574,7 @@ bool HeaderSearch::ShouldEnterIncludeFile(Preprocessor &PP,
     }
   }
 
+  FileInfo.IsLocallyIncluded = true;
   IsFirstIncludeOfFile = PP.markIncluded(File);
   return true;
 }
diff --git a/clang/lib/Parse/ParseCXXInlineMethods.cpp b/clang/lib/Parse/ParseCXXInlineMethods.cpp
index d054eda279b8..a26568dfd6aa 100644
--- a/clang/lib/Parse/ParseCXXInlineMethods.cpp
+++ b/clang/lib/Parse/ParseCXXInlineMethods.cpp
@@ -603,6 +603,8 @@ void Parser::ParseLexedMethodDef(LexedMethod &LM) {
   // to be re-used for method bodies as well.
   ParseScope FnScope(this, Scope::FnScope | Scope::DeclScope |
                                Scope::CompoundStmtScope);
+  Sema::FPFeaturesStateRAII SaveFPFeatures(Actions);
+
   Actions.ActOnStartOfFunctionDef(getCurScope(), LM.D);
 
   if (Tok.is(tok::kw_try)) {
diff --git a/clang/lib/Parse/ParseDecl.cpp b/clang/lib/Parse/ParseDecl.cpp
index 5f26b5a9e46b..a7846e102a43 100644
--- a/clang/lib/Parse/ParseDecl.cpp
+++ b/clang/lib/Parse/ParseDecl.cpp
@@ -2222,7 +2222,7 @@ Parser::DeclGroupPtrTy Parser::ParseDeclGroup(ParsingDeclSpec &DS,
   }
 
   if (getLangOpts().HLSL)
-    MaybeParseHLSLSemantics(D);
+    MaybeParseHLSLAnnotations(D);
 
   if (Tok.is(tok::kw_requires))
     ParseTrailingRequiresClause(D);
@@ -2469,7 +2469,7 @@ Parser::DeclGroupPtrTy Parser::ParseDeclGroup(ParsingDeclSpec &DS,
     ParseDeclarator(D);
 
     if (getLangOpts().HLSL)
-      MaybeParseHLSLSemantics(D);
+      MaybeParseHLSLAnnotations(D);
 
     if (!D.isInvalidType()) {
       // C++2a [dcl.decl]p1
@@ -7038,18 +7038,23 @@ void Parser::ParseDirectDeclarator(Declarator &D) {
 void Parser::ParseDecompositionDeclarator(Declarator &D) {
   assert(Tok.is(tok::l_square));
 
+  TentativeParsingAction PA(*this);
+  BalancedDelimiterTracker T(*this, tok::l_square);
+  T.consumeOpen();
+
+  if (isCXX11AttributeSpecifier())
+    DiagnoseAndSkipCXX11Attributes();
+
   // If this doesn't look like a structured binding, maybe it's a misplaced
   // array declarator.
-  // FIXME: Consume the l_square first so we don't need extra lookahead for
-  // this.
-  if (!(NextToken().is(tok::identifier) &&
-        GetLookAheadToken(2).isOneOf(tok::comma, tok::r_square)) &&
-      !(NextToken().is(tok::r_square) &&
-        GetLookAheadToken(2).isOneOf(tok::equal, tok::l_brace)))
+  if (!(Tok.is(tok::identifier) &&
+        NextToken().isOneOf(tok::comma, tok::r_square, tok::kw_alignas,
+                            tok::l_square)) &&
+      !(Tok.is(tok::r_square) &&
+        NextToken().isOneOf(tok::equal, tok::l_brace))) {
+    PA.Revert();
     return ParseMisplacedBracketDeclarator(D);
-
-  BalancedDelimiterTracker T(*this, tok::l_square);
-  T.consumeOpen();
+  }
 
   SmallVector<DecompositionDeclarator::Binding, 32> Bindings;
   while (Tok.isNot(tok::r_square)) {
@@ -7074,13 +7079,27 @@ void Parser::ParseDecompositionDeclarator(Declarator &D) {
       }
     }
 
+    if (isCXX11AttributeSpecifier())
+      DiagnoseAndSkipCXX11Attributes();
+
     if (Tok.isNot(tok::identifier)) {
       Diag(Tok, diag::err_expected) << tok::identifier;
       break;
     }
 
-    Bindings.push_back({Tok.getIdentifierInfo(), Tok.getLocation()});
+    IdentifierInfo *II = Tok.getIdentifierInfo();
+    SourceLocation Loc = Tok.getLocation();
     ConsumeToken();
+
+    ParsedAttributes Attrs(AttrFactory);
+    if (isCXX11AttributeSpecifier()) {
+      Diag(Tok, getLangOpts().CPlusPlus26
+                    ? diag::warn_cxx23_compat_decl_attrs_on_binding
+                    : diag::ext_decl_attrs_on_binding);
+      MaybeParseCXX11Attributes(Attrs);
+    }
+
+    Bindings.push_back({II, Loc, std::move(Attrs)});
   }
 
   if (Tok.isNot(tok::r_square))
@@ -7095,6 +7114,8 @@ void Parser::ParseDecompositionDeclarator(Declarator &D) {
     T.consumeClose();
   }
 
+  PA.Commit();
+
   return D.setDecompositionBindings(T.getOpenLocation(), Bindings,
                                     T.getCloseLocation());
 }
@@ -7699,7 +7720,7 @@ void Parser::ParseParameterDeclarationClause(
     // Parse GNU attributes, if present.
     MaybeParseGNUAttributes(ParmDeclarator);
     if (getLangOpts().HLSL)
-      MaybeParseHLSLSemantics(DS.getAttributes());
+      MaybeParseHLSLAnnotations(DS.getAttributes());
 
     if (Tok.is(tok::kw_requires)) {
       // User tried to define a requires clause in a parameter declaration,
diff --git a/clang/lib/Parse/ParseExpr.cpp b/clang/lib/Parse/ParseExpr.cpp
index 32d96f81c4c8..7d6febb04a82 100644
--- a/clang/lib/Parse/ParseExpr.cpp
+++ b/clang/lib/Parse/ParseExpr.cpp
@@ -31,6 +31,7 @@
 #include "clang/Sema/ParsedTemplate.h"
 #include "clang/Sema/Scope.h"
 #include "clang/Sema/SemaCUDA.h"
+#include "clang/Sema/SemaOpenACC.h"
 #include "clang/Sema/SemaOpenMP.h"
 #include "clang/Sema/SemaSYCL.h"
 #include "clang/Sema/TypoCorrection.h"
@@ -2070,15 +2071,22 @@ Parser::ParsePostfixExpressionSuffix(ExprResult LHS) {
       if (!LHS.isInvalid() && !HasError && !Length.isInvalid() &&
           !Stride.isInvalid() && Tok.is(tok::r_square)) {
         if (ColonLocFirst.isValid() || ColonLocSecond.isValid()) {
-          // FIXME: OpenACC hasn't implemented Sema/Array section handling at a
-          // semantic level yet. For now, just reuse the OpenMP implementation
-          // as it gets the parsing/type management mostly right, and we can
-          // replace this call to ActOnOpenACCArraySectionExpr in the future.
-          // Eventually we'll genericize the OPenMPArraySectionExpr type as
-          // well.
-          LHS = Actions.OpenMP().ActOnOMPArraySectionExpr(
-              LHS.get(), Loc, ArgExprs.empty() ? nullptr : ArgExprs[0],
-              ColonLocFirst, ColonLocSecond, Length.get(), Stride.get(), RLoc);
+          // Like above, AllowOpenACCArraySections is 'more specific' and only
+          // enabled when actively parsing a 'var' in a 'var-list' during
+          // clause/'cache' construct parsing, so it is more specific. So we
+          // should do it first, so that the correct node gets created.
+          if (AllowOpenACCArraySections) {
+            assert(!Stride.isUsable() && !ColonLocSecond.isValid() &&
+                   "Stride/second colon not allowed for OpenACC");
+            LHS = Actions.OpenACC().ActOnArraySectionExpr(
+                LHS.get(), Loc, ArgExprs.empty() ? nullptr : ArgExprs[0],
+                ColonLocFirst, Length.get(), RLoc);
+          } else {
+            LHS = Actions.OpenMP().ActOnOMPArraySectionExpr(
+                LHS.get(), Loc, ArgExprs.empty() ? nullptr : ArgExprs[0],
+                ColonLocFirst, ColonLocSecond, Length.get(), Stride.get(),
+                RLoc);
+          }
         } else {
           LHS = Actions.ActOnArraySubscriptExpr(getCurScope(), LHS.get(), Loc,
                                                 ArgExprs, RLoc);
diff --git a/clang/lib/Parse/ParseHLSL.cpp b/clang/lib/Parse/ParseHLSL.cpp
index d97985d42369..f4cbece31f18 100644
--- a/clang/lib/Parse/ParseHLSL.cpp
+++ b/clang/lib/Parse/ParseHLSL.cpp
@@ -63,7 +63,7 @@ Decl *Parser::ParseHLSLBuffer(SourceLocation &DeclEnd) {
   SourceLocation IdentifierLoc = ConsumeToken();
 
   ParsedAttributes Attrs(AttrFactory);
-  MaybeParseHLSLSemantics(Attrs, nullptr);
+  MaybeParseHLSLAnnotations(Attrs, nullptr);
 
   ParseScope BufferScope(this, Scope::DeclScope);
   BalancedDelimiterTracker T(*this, tok::l_brace);
@@ -118,12 +118,10 @@ static void fixSeparateAttrArgAndNumber(StringRef ArgStr, SourceLocation ArgLoc,
   Slot = IdentifierLoc::create(Ctx, ArgLoc, PP.getIdentifierInfo(FixedArg));
 }
 
-void Parser::ParseHLSLSemantics(ParsedAttributes &Attrs,
-                                SourceLocation *EndLoc) {
-  // FIXME: HLSLSemantic is shared for Semantic and resource binding which is
-  // confusing. Need a better name to avoid misunderstanding. Issue
-  // https://github.com/llvm/llvm-project/issues/57882
-  assert(Tok.is(tok::colon) && "Not a HLSL Semantic");
+void Parser::ParseHLSLAnnotations(ParsedAttributes &Attrs,
+                                  SourceLocation *EndLoc) {
+
+  assert(Tok.is(tok::colon) && "Not a HLSL Annotation");
   ConsumeToken();
 
   IdentifierInfo *II = nullptr;
@@ -141,7 +139,7 @@ void Parser::ParseHLSLSemantics(ParsedAttributes &Attrs,
   if (EndLoc)
     *EndLoc = Tok.getLocation();
   ParsedAttr::Kind AttrKind =
-      ParsedAttr::getParsedKind(II, nullptr, ParsedAttr::AS_HLSLSemantic);
+      ParsedAttr::getParsedKind(II, nullptr, ParsedAttr::AS_HLSLAnnotation);
 
   ArgsVector ArgExprs;
   switch (AttrKind) {
@@ -192,10 +190,10 @@ void Parser::ParseHLSLSemantics(ParsedAttributes &Attrs,
   case ParsedAttr::AT_HLSLSV_DispatchThreadID:
     break;
   default:
-    llvm_unreachable("invalid HLSL Semantic");
+    llvm_unreachable("invalid HLSL Annotation");
     break;
   }
 
   Attrs.addNew(II, Loc, nullptr, SourceLocation(), ArgExprs.data(),
-               ArgExprs.size(), ParsedAttr::Form::HLSLSemantic());
+               ArgExprs.size(), ParsedAttr::Form::HLSLAnnotation());
 }
diff --git a/clang/lib/Parse/ParseObjc.cpp b/clang/lib/Parse/ParseObjc.cpp
index 671dcb71e51a..8e54fe012c55 100644
--- a/clang/lib/Parse/ParseObjc.cpp
+++ b/clang/lib/Parse/ParseObjc.cpp
@@ -3736,6 +3736,7 @@ void Parser::ParseLexedObjCMethodDefs(LexedMethod &LM, bool parseMethod) {
   ParseScope BodyScope(this, (parseMethod ? Scope::ObjCMethodScope : 0) |
                                  Scope::FnScope | Scope::DeclScope |
                                  Scope::CompoundStmtScope);
+  Sema::FPFeaturesStateRAII SaveFPFeatures(Actions);
 
   // Tell the actions module that we have entered a method or c-function definition
   // with the specified Declarator for the method/function.
diff --git a/clang/lib/Parse/ParseOpenACC.cpp b/clang/lib/Parse/ParseOpenACC.cpp
index 8a18fca8064e..29326f5d993a 100644
--- a/clang/lib/Parse/ParseOpenACC.cpp
+++ b/clang/lib/Parse/ParseOpenACC.cpp
@@ -327,7 +327,7 @@ OpenACCReductionOperator ParseReductionOperator(Parser &P) {
       return OpenACCReductionOperator::Max;
     if (ReductionKindTok.getIdentifierInfo()->isStr("min"))
       return OpenACCReductionOperator::Min;
-    LLVM_FALLTHROUGH;
+    [[fallthrough]];
   default:
     P.Diag(ReductionKindTok, diag::err_acc_invalid_reduction_operator);
     return OpenACCReductionOperator::Invalid;
@@ -945,7 +945,7 @@ Parser::OpenACCClauseParseResult Parser::ParseOpenACCClauseParams(
       // the 'update' clause, so we have to handle it here.  U se an assert to
       // make sure we get the right differentiator.
       assert(DirKind == OpenACCDirectiveKind::Update);
-      LLVM_FALLTHROUGH;
+      [[fallthrough]];
     case OpenACCClauseKind::Attach:
     case OpenACCClauseKind::Copy:
     case OpenACCClauseKind::Delete:
diff --git a/clang/lib/Parse/ParseTemplate.cpp b/clang/lib/Parse/ParseTemplate.cpp
index b07ce451e878..665253a6674d 100644
--- a/clang/lib/Parse/ParseTemplate.cpp
+++ b/clang/lib/Parse/ParseTemplate.cpp
@@ -733,7 +733,12 @@ NamedDecl *Parser::ParseTypeParameter(unsigned Depth, unsigned Position) {
   // we introduce the type parameter into the local scope.
   SourceLocation EqualLoc;
   ParsedType DefaultArg;
+  std::optional<DelayTemplateIdDestructionRAII> DontDestructTemplateIds;
   if (TryConsumeToken(tok::equal, EqualLoc)) {
+    // The default argument might contain a lambda declaration; avoid destroying
+    // parsed template ids at the end of that declaration because they can be
+    // used in a type constraint later.
+    DontDestructTemplateIds.emplace(*this, /*DelayTemplateIdDestruction=*/true);
     // The default argument may declare template parameters, notably
     // if it contains a generic lambda, so we need to increase
     // the template depth as these parameters would not be instantiated
diff --git a/clang/lib/Parse/Parser.cpp b/clang/lib/Parse/Parser.cpp
index ef46fc74cedc..adcbe5858bc7 100644
--- a/clang/lib/Parse/Parser.cpp
+++ b/clang/lib/Parse/Parser.cpp
@@ -1497,6 +1497,8 @@ Decl *Parser::ParseFunctionDefinition(ParsingDeclarator &D,
     return Actions.ActOnFinishFunctionBody(Res, nullptr, false);
   }
 
+  Sema::FPFeaturesStateRAII SaveFPFeatures(Actions);
+
   if (Tok.is(tok::kw_try))
     return ParseFunctionTryBlock(Res, BodyScope);
 
diff --git a/clang/lib/Sema/DeclSpec.cpp b/clang/lib/Sema/DeclSpec.cpp
index b79683bb32a6..5f63c857c430 100644
--- a/clang/lib/Sema/DeclSpec.cpp
+++ b/clang/lib/Sema/DeclSpec.cpp
@@ -293,7 +293,7 @@ DeclaratorChunk DeclaratorChunk::getFunction(bool hasProto,
 
 void Declarator::setDecompositionBindings(
     SourceLocation LSquareLoc,
-    ArrayRef<DecompositionDeclarator::Binding> Bindings,
+    MutableArrayRef<DecompositionDeclarator::Binding> Bindings,
     SourceLocation RSquareLoc) {
   assert(!hasName() && "declarator given multiple names!");
 
@@ -317,7 +317,7 @@ void Declarator::setDecompositionBindings(
           new DecompositionDeclarator::Binding[Bindings.size()];
       BindingGroup.DeleteBindings = true;
     }
-    std::uninitialized_copy(Bindings.begin(), Bindings.end(),
+    std::uninitialized_move(Bindings.begin(), Bindings.end(),
                             BindingGroup.Bindings);
   }
 }
diff --git a/clang/lib/Sema/JumpDiagnostics.cpp b/clang/lib/Sema/JumpDiagnostics.cpp
index ce6211c23218..8af36d5c24e3 100644
--- a/clang/lib/Sema/JumpDiagnostics.cpp
+++ b/clang/lib/Sema/JumpDiagnostics.cpp
@@ -180,7 +180,8 @@ static ScopePair GetDiagForGotoScopeDecl(Sema &S, const Decl *D) {
     }
 
     const Expr *Init = VD->getInit();
-    if (S.Context.getLangOpts().CPlusPlus && VD->hasLocalStorage() && Init) {
+    if (S.Context.getLangOpts().CPlusPlus && VD->hasLocalStorage() && Init &&
+        !Init->containsErrors()) {
       // C++11 [stmt.dcl]p3:
       //   A program that jumps from a point where a variable with automatic
       //   storage duration is not in scope to a point where it is in scope
diff --git a/clang/lib/Sema/MultiplexExternalSemaSource.cpp b/clang/lib/Sema/MultiplexExternalSemaSource.cpp
index 058e22cb2b81..79e656eb4b7e 100644
--- a/clang/lib/Sema/MultiplexExternalSemaSource.cpp
+++ b/clang/lib/Sema/MultiplexExternalSemaSource.cpp
@@ -46,7 +46,7 @@ void MultiplexExternalSemaSource::AddSource(ExternalSemaSource *Source) {
 // ExternalASTSource.
 //===----------------------------------------------------------------------===//
 
-Decl *MultiplexExternalSemaSource::GetExternalDecl(uint32_t ID) {
+Decl *MultiplexExternalSemaSource::GetExternalDecl(GlobalDeclID ID) {
   for(size_t i = 0; i < Sources.size(); ++i)
     if (Decl *Result = Sources[i]->GetExternalDecl(ID))
       return Result;
diff --git a/clang/lib/Sema/SemaAPINotes.cpp b/clang/lib/Sema/SemaAPINotes.cpp
index 4c445f28bba8..c5998aca0d72 100644
--- a/clang/lib/Sema/SemaAPINotes.cpp
+++ b/clang/lib/Sema/SemaAPINotes.cpp
@@ -594,6 +594,11 @@ static void ProcessAPINotes(Sema &S, TagDecl *D, const api_notes::TagInfo &Info,
     D->addAttr(
         SwiftAttrAttr::Create(S.Context, "release:" + ReleaseOp.value()));
 
+  if (auto Copyable = Info.isSwiftCopyable()) {
+    if (!*Copyable)
+      D->addAttr(SwiftAttrAttr::Create(S.Context, "~Copyable"));
+  }
+
   if (auto Extensibility = Info.EnumExtensibility) {
     using api_notes::EnumExtensibilityKind;
     bool ShouldAddAttribute = (*Extensibility != EnumExtensibilityKind::None);
diff --git a/clang/lib/Sema/SemaChecking.cpp b/clang/lib/Sema/SemaChecking.cpp
index 51757f4cf727..cf8840c63024 100644
--- a/clang/lib/Sema/SemaChecking.cpp
+++ b/clang/lib/Sema/SemaChecking.cpp
@@ -3164,13 +3164,20 @@ Sema::CheckBuiltinFunctionCall(FunctionDecl *FDecl, unsigned BuiltinID,
 
     const Expr *Arg = TheCall->getArg(0);
     const auto *TyA = Arg->getType()->getAs<VectorType>();
-    if (!TyA) {
+
+    QualType ElTy;
+    if (TyA)
+      ElTy = TyA->getElementType();
+    else if (Arg->getType()->isSizelessVectorType())
+      ElTy = Arg->getType()->getSizelessVectorEltType(Context);
+
+    if (ElTy.isNull()) {
       Diag(Arg->getBeginLoc(), diag::err_builtin_invalid_arg_type)
           << 1 << /* vector ty*/ 4 << Arg->getType();
       return ExprError();
     }
 
-    TheCall->setType(TyA->getElementType());
+    TheCall->setType(ElTy);
     break;
   }
 
@@ -3186,12 +3193,20 @@ Sema::CheckBuiltinFunctionCall(FunctionDecl *FDecl, unsigned BuiltinID,
 
     const Expr *Arg = TheCall->getArg(0);
     const auto *TyA = Arg->getType()->getAs<VectorType>();
-    if (!TyA || !TyA->getElementType()->isIntegerType()) {
+
+    QualType ElTy;
+    if (TyA)
+      ElTy = TyA->getElementType();
+    else if (Arg->getType()->isSizelessVectorType())
+      ElTy = Arg->getType()->getSizelessVectorEltType(Context);
+
+    if (ElTy.isNull() || !ElTy->isIntegerType()) {
       Diag(Arg->getBeginLoc(), diag::err_builtin_invalid_arg_type)
           << 1  << /* vector of integers */ 6 << Arg->getType();
       return ExprError();
     }
-    TheCall->setType(TyA->getElementType());
+
+    TheCall->setType(ElTy);
     break;
   }
 
@@ -7953,7 +7968,8 @@ void Sema::checkCall(NamedDecl *FDecl, const FunctionProtoType *Proto,
     // For variadic functions, we may have more args than parameters.
     // For some K&R functions, we may have less args than parameters.
     const auto N = std::min<unsigned>(Proto->getNumParams(), Args.size());
-    bool AnyScalableArgsOrRet = Proto->getReturnType()->isSizelessVectorType();
+    bool IsScalableRet = Proto->getReturnType()->isSizelessVectorType();
+    bool IsScalableArg = false;
     for (unsigned ArgIdx = 0; ArgIdx < N; ++ArgIdx) {
       // Args[ArgIdx] can be null in malformed code.
       if (const Expr *Arg = Args[ArgIdx]) {
@@ -7968,7 +7984,7 @@ void Sema::checkCall(NamedDecl *FDecl, const FunctionProtoType *Proto,
 
         QualType ParamTy = Proto->getParamType(ArgIdx);
         if (ParamTy->isSizelessVectorType())
-          AnyScalableArgsOrRet = true;
+          IsScalableArg = true;
         QualType ArgTy = Arg->getType();
         CheckArgAlignment(Arg->getExprLoc(), FDecl, std::to_string(ArgIdx + 1),
                           ArgTy, ParamTy);
@@ -7993,7 +8009,8 @@ void Sema::checkCall(NamedDecl *FDecl, const FunctionProtoType *Proto,
     // arguments or return values, then warn the user that the streaming and
     // non-streaming vector lengths may be different.
     const auto *CallerFD = dyn_cast<FunctionDecl>(CurContext);
-    if (CallerFD && (!FD || !FD->getBuiltinID()) && AnyScalableArgsOrRet) {
+    if (CallerFD && (!FD || !FD->getBuiltinID()) &&
+        (IsScalableArg || IsScalableRet)) {
       bool IsCalleeStreaming =
           ExtInfo.AArch64SMEAttributes & FunctionType::SME_PStateSMEnabledMask;
       bool IsCalleeStreamingCompatible =
@@ -8002,8 +8019,14 @@ void Sema::checkCall(NamedDecl *FDecl, const FunctionProtoType *Proto,
       ArmStreamingType CallerFnType = getArmStreamingFnType(CallerFD);
       if (!IsCalleeStreamingCompatible &&
           (CallerFnType == ArmStreamingCompatible ||
-           ((CallerFnType == ArmStreaming) ^ IsCalleeStreaming)))
-        Diag(Loc, diag::warn_sme_streaming_pass_return_vl_to_non_streaming);
+           ((CallerFnType == ArmStreaming) ^ IsCalleeStreaming))) {
+        if (IsScalableArg)
+          Diag(Loc, diag::warn_sme_streaming_pass_return_vl_to_non_streaming)
+              << /*IsArg=*/true;
+        if (IsScalableRet)
+          Diag(Loc, diag::warn_sme_streaming_pass_return_vl_to_non_streaming)
+              << /*IsArg=*/false;
+      }
     }
 
     FunctionType::ArmStateValue CalleeArmZAState =
@@ -12536,6 +12559,17 @@ CheckPrintfHandler::checkFormatExpr(const analyze_printf::PrintfSpecifier &FS,
     return true;
   }
 
+  // Diagnose attempts to use '%P' with ObjC object types, which will result in
+  // dumping raw class data (like is-a pointer), not actual data.
+  if (FS.getConversionSpecifier().getKind() == ConversionSpecifier::PArg &&
+      ExprTy->isObjCObjectPointerType()) {
+    const CharSourceRange &CSR =
+        getSpecifierRange(StartSpecifier, SpecifierLen);
+    EmitFormatDiagnostic(S.PDiag(diag::warn_format_P_with_objc_pointer),
+                         E->getExprLoc(), false, CSR);
+    return true;
+  }
+
   ArgType::MatchKind ImplicitMatch = ArgType::NoMatch;
   ArgType::MatchKind Match = AT.matchesType(S.Context, ExprTy);
   ArgType::MatchKind OrigMatch = Match;
@@ -18716,8 +18750,10 @@ void Sema::CheckArrayAccess(const Expr *expr) {
         expr = cast<MemberExpr>(expr)->getBase();
         break;
       }
-      case Stmt::OMPArraySectionExprClass: {
-        const OMPArraySectionExpr *ASE = cast<OMPArraySectionExpr>(expr);
+      case Stmt::ArraySectionExprClass: {
+        const ArraySectionExpr *ASE = cast<ArraySectionExpr>(expr);
+        // FIXME: We should probably be checking all of the elements to the
+        // 'length' here as well.
         if (ASE->getLowerBound())
           CheckArrayAccess(ASE->getBase(), ASE->getLowerBound(),
                            /*ASE=*/nullptr, AllowOnePastEnd > 0);
diff --git a/clang/lib/Sema/SemaDecl.cpp b/clang/lib/Sema/SemaDecl.cpp
index 35eac93e324d..671752b56e01 100644
--- a/clang/lib/Sema/SemaDecl.cpp
+++ b/clang/lib/Sema/SemaDecl.cpp
@@ -1974,7 +1974,7 @@ static bool ShouldDiagnoseUnusedDecl(const LangOptions &LangOpts,
     // it is, by the bindings' expressions).
     bool IsAllPlaceholders = true;
     for (const auto *BD : DD->bindings()) {
-      if (BD->isReferenced())
+      if (BD->isReferenced() || BD->hasAttr<UnusedAttr>())
         return false;
       IsAllPlaceholders = IsAllPlaceholders && BD->isPlaceholderVar(LangOpts);
     }
@@ -12417,12 +12417,16 @@ bool Sema::CheckFunctionDeclaration(Scope *S, FunctionDecl *NewFD,
     bool UsesZT0 = Attr && Attr->isNewZT0();
 
     if (NewFD->hasAttr<ArmLocallyStreamingAttr>()) {
-      if (NewFD->getReturnType()->isSizelessVectorType() ||
-          llvm::any_of(NewFD->parameters(), [](ParmVarDecl *P) {
+      if (NewFD->getReturnType()->isSizelessVectorType())
+        Diag(NewFD->getLocation(),
+             diag::warn_sme_locally_streaming_has_vl_args_returns)
+            << /*IsArg=*/false;
+      if (llvm::any_of(NewFD->parameters(), [](ParmVarDecl *P) {
             return P->getOriginalType()->isSizelessVectorType();
           }))
         Diag(NewFD->getLocation(),
-             diag::warn_sme_locally_streaming_has_vl_args_returns);
+             diag::warn_sme_locally_streaming_has_vl_args_returns)
+            << /*IsArg=*/true;
     }
     if (const auto *FPT = NewFD->getType()->getAs<FunctionProtoType>()) {
       FunctionProtoType::ExtProtoInfo EPI = FPT->getExtProtoInfo();
@@ -13498,16 +13502,18 @@ void Sema::checkNonTrivialCUnion(QualType QT, SourceLocation Loc,
 void Sema::AddInitializerToDecl(Decl *RealDecl, Expr *Init, bool DirectInit) {
   // If there is no declaration, there was an error parsing it.  Just ignore
   // the initializer.
-  if (!RealDecl || RealDecl->isInvalidDecl()) {
+  if (!RealDecl) {
     CorrectDelayedTyposInExpr(Init, dyn_cast_or_null<VarDecl>(RealDecl));
     return;
   }
 
-  if (CXXMethodDecl *Method = dyn_cast<CXXMethodDecl>(RealDecl)) {
-    // Pure-specifiers are handled in ActOnPureSpecifier.
-    Diag(Method->getLocation(), diag::err_member_function_initialization)
-      << Method->getDeclName() << Init->getSourceRange();
-    Method->setInvalidDecl();
+  if (auto *Method = dyn_cast<CXXMethodDecl>(RealDecl)) {
+    if (!Method->isInvalidDecl()) {
+      // Pure-specifiers are handled in ActOnPureSpecifier.
+      Diag(Method->getLocation(), diag::err_member_function_initialization)
+          << Method->getDeclName() << Init->getSourceRange();
+      Method->setInvalidDecl();
+    }
     return;
   }
 
@@ -13519,6 +13525,15 @@ void Sema::AddInitializerToDecl(Decl *RealDecl, Expr *Init, bool DirectInit) {
     return;
   }
 
+  if (VDecl->isInvalidDecl()) {
+    CorrectDelayedTyposInExpr(Init, VDecl);
+    ExprResult Recovery =
+        CreateRecoveryExpr(Init->getBeginLoc(), Init->getEndLoc(), {Init});
+    if (Expr *E = Recovery.get())
+      VDecl->setInit(E);
+    return;
+  }
+
   // WebAssembly tables can't be used to initialise a variable.
   if (Init && !Init->getType().isNull() &&
       Init->getType()->isWebAssemblyTableType()) {
@@ -15899,6 +15914,11 @@ Decl *Sema::ActOnStartOfFunctionDef(Scope *FnBodyScope, Decl *D,
     FD->setInvalidDecl();
     return D;
   }
+
+  // Some function attributes (like OptimizeNoneAttr) need actions before
+  // parsing body started.
+  applyFunctionAttributesBeforeParsingBody(D);
+
   // We want to attach documentation to original Decl (which might be
   // a function template).
   ActOnDocumentableDecl(D);
@@ -15910,6 +15930,20 @@ Decl *Sema::ActOnStartOfFunctionDef(Scope *FnBodyScope, Decl *D,
   return D;
 }
 
+void Sema::applyFunctionAttributesBeforeParsingBody(Decl *FD) {
+  if (!FD || FD->isInvalidDecl())
+    return;
+  if (auto *TD = dyn_cast<FunctionTemplateDecl>(FD))
+    FD = TD->getTemplatedDecl();
+  if (FD && FD->hasAttr<OptimizeNoneAttr>()) {
+    FPOptionsOverride FPO;
+    FPO.setDisallowOptimizations();
+    CurFPFeatures.applyChanges(FPO);
+    FpPragmaStack.CurrentValue =
+        CurFPFeatures.getChangesFrom(FPOptions(LangOpts));
+  }
+}
+
 /// Given the set of return statements within a function body,
 /// compute the variables that are subject to the named return value
 /// optimization.
diff --git a/clang/lib/Sema/SemaDeclCXX.cpp b/clang/lib/Sema/SemaDeclCXX.cpp
index abdbc9d8830c..338b0ec1e099 100644
--- a/clang/lib/Sema/SemaDeclCXX.cpp
+++ b/clang/lib/Sema/SemaDeclCXX.cpp
@@ -910,6 +910,8 @@ Sema::ActOnDecompositionDeclarator(Scope *S, Declarator &D,
 
     auto *BD = BindingDecl::Create(Context, DC, B.NameLoc, VarName);
 
+    ProcessDeclAttributeList(S, BD, *B.Attrs);
+
     // Find the shadowed declaration before filtering for scope.
     NamedDecl *ShadowedDecl = D.getCXXScopeSpec().isEmpty()
                                   ? getShadowedDeclaration(BD, Previous)
@@ -12052,11 +12054,17 @@ bool Sema::isStdInitializerList(QualType Ty, QualType *Element) {
 
     Template = Specialization->getSpecializedTemplate();
     Arguments = Specialization->getTemplateArgs().data();
-  } else if (const TemplateSpecializationType *TST =
-                 Ty->getAs<TemplateSpecializationType>()) {
-    Template = dyn_cast_or_null<ClassTemplateDecl>(
-        TST->getTemplateName().getAsTemplateDecl());
-    Arguments = TST->template_arguments().begin();
+  } else {
+    const TemplateSpecializationType *TST = nullptr;
+    if (auto *ICN = Ty->getAs<InjectedClassNameType>())
+      TST = ICN->getInjectedTST();
+    else
+      TST = Ty->getAs<TemplateSpecializationType>();
+    if (TST) {
+      Template = dyn_cast_or_null<ClassTemplateDecl>(
+          TST->getTemplateName().getAsTemplateDecl());
+      Arguments = TST->template_arguments().begin();
+    }
   }
   if (!Template)
     return false;
diff --git a/clang/lib/Sema/SemaDeclObjC.cpp b/clang/lib/Sema/SemaDeclObjC.cpp
index 74d6f0700b0e..934ba174a426 100644
--- a/clang/lib/Sema/SemaDeclObjC.cpp
+++ b/clang/lib/Sema/SemaDeclObjC.cpp
@@ -494,6 +494,10 @@ void Sema::ActOnStartOfObjCMethodDef(Scope *FnBodyScope, Decl *D) {
       }
     }
   }
+
+  // Some function attributes (like OptimizeNoneAttr) need actions before
+  // parsing body started.
+  applyFunctionAttributesBeforeParsingBody(D);
 }
 
 namespace {
diff --git a/clang/lib/Sema/SemaExceptionSpec.cpp b/clang/lib/Sema/SemaExceptionSpec.cpp
index 00384f9dc16a..c9dd6bb2413e 100644
--- a/clang/lib/Sema/SemaExceptionSpec.cpp
+++ b/clang/lib/Sema/SemaExceptionSpec.cpp
@@ -1314,7 +1314,7 @@ CanThrowResult Sema::canThrow(const Stmt *S) {
     // Some might be dependent for other reasons.
   case Expr::ArraySubscriptExprClass:
   case Expr::MatrixSubscriptExprClass:
-  case Expr::OMPArraySectionExprClass:
+  case Expr::ArraySectionExprClass:
   case Expr::OMPArrayShapingExprClass:
   case Expr::OMPIteratorExprClass:
   case Expr::BinaryOperatorClass:
diff --git a/clang/lib/Sema/SemaExpr.cpp b/clang/lib/Sema/SemaExpr.cpp
index 5c861467bc10..50f92c496a53 100644
--- a/clang/lib/Sema/SemaExpr.cpp
+++ b/clang/lib/Sema/SemaExpr.cpp
@@ -5069,11 +5069,18 @@ ExprResult Sema::ActOnArraySubscriptExpr(Scope *S, Expr *base,
                                          SourceLocation rbLoc) {
 
   if (base && !base->getType().isNull() &&
-      base->hasPlaceholderType(BuiltinType::OMPArraySection))
-    return OpenMP().ActOnOMPArraySectionExpr(base, lbLoc, ArgExprs.front(),
-                                             SourceLocation(), SourceLocation(),
-                                             /*Length*/ nullptr,
-                                             /*Stride=*/nullptr, rbLoc);
+      base->hasPlaceholderType(BuiltinType::ArraySection)) {
+    auto *AS = cast<ArraySectionExpr>(base);
+    if (AS->isOMPArraySection())
+      return OpenMP().ActOnOMPArraySectionExpr(
+          base, lbLoc, ArgExprs.front(), SourceLocation(), SourceLocation(),
+          /*Length*/ nullptr,
+          /*Stride=*/nullptr, rbLoc);
+
+    return OpenACC().ActOnArraySectionExpr(base, lbLoc, ArgExprs.front(),
+                                           SourceLocation(), /*Length*/ nullptr,
+                                           rbLoc);
+  }
 
   // Since this might be a postfix expression, get rid of ParenListExprs.
   if (isa<ParenListExpr>(base)) {
@@ -6361,7 +6368,7 @@ static bool isPlaceholderToRemoveAsArg(QualType type) {
   case BuiltinType::BoundMember:
   case BuiltinType::BuiltinFn:
   case BuiltinType::IncompleteMatrixIdx:
-  case BuiltinType::OMPArraySection:
+  case BuiltinType::ArraySection:
   case BuiltinType::OMPArrayShaping:
   case BuiltinType::OMPIterator:
     return true;
@@ -21343,8 +21350,9 @@ ExprResult Sema::CheckPlaceholderExpr(Expr *E) {
     return ExprError();
 
   // Expressions of unknown type.
-  case BuiltinType::OMPArraySection:
-    Diag(E->getBeginLoc(), diag::err_omp_array_section_use);
+  case BuiltinType::ArraySection:
+    Diag(E->getBeginLoc(), diag::err_array_section_use)
+        << cast<ArraySectionExpr>(E)->isOMPArraySection();
     return ExprError();
 
   // Expressions of unknown type.
diff --git a/clang/lib/Sema/SemaInit.cpp b/clang/lib/Sema/SemaInit.cpp
index 793e16df1789..7d9eaf672046 100644
--- a/clang/lib/Sema/SemaInit.cpp
+++ b/clang/lib/Sema/SemaInit.cpp
@@ -7753,9 +7753,9 @@ static void visitLocalsRetainedByReferenceBinding(IndirectLocalPath &Path,
     break;
   }
 
-  case Stmt::OMPArraySectionExprClass: {
+  case Stmt::ArraySectionExprClass: {
     visitLocalsRetainedByInitializer(Path,
-                                     cast<OMPArraySectionExpr>(Init)->getBase(),
+                                     cast<ArraySectionExpr>(Init)->getBase(),
                                      Visit, true, EnableLifetimeWarnings);
     break;
   }
@@ -8340,8 +8340,17 @@ void Sema::checkInitializerLifetime(const InitializedEntity &Entity,
             << Entity.getType()->isReferenceType() << CLE->getInitializer() << 2
             << DiagRange;
       } else {
-        Diag(DiagLoc, diag::warn_ret_local_temp_addr_ref)
-         << Entity.getType()->isReferenceType() << DiagRange;
+        // P2748R5: Disallow Binding a Returned Glvalue to a Temporary.
+        // [stmt.return]/p6: In a function whose return type is a reference,
+        // other than an invented function for std::is_convertible ([meta.rel]),
+        // a return statement that binds the returned reference to a temporary
+        // expression ([class.temporary]) is ill-formed.
+        if (getLangOpts().CPlusPlus26 && Entity.getType()->isReferenceType())
+          Diag(DiagLoc, diag::err_ret_local_temp_ref)
+              << Entity.getType()->isReferenceType() << DiagRange;
+        else
+          Diag(DiagLoc, diag::warn_ret_local_temp_addr_ref)
+              << Entity.getType()->isReferenceType() << DiagRange;
       }
       break;
     }
@@ -10790,8 +10799,6 @@ QualType Sema::DeduceTemplateSpecializationFromInitializer(
   // FIXME: Perform "exact type" matching first, per CWG discussion?
   //        Or implement this via an implied 'T(T) -> T' deduction guide?
 
-  // FIXME: Do we need/want a std::initializer_list<T> special case?
-
   // Look up deduction guides, including those synthesized from constructors.
   //
   // C++1z [over.match.class.deduct]p1:
diff --git a/clang/lib/Sema/SemaOpenACC.cpp b/clang/lib/Sema/SemaOpenACC.cpp
index ba69e71e30a1..d5cfe82a5d70 100644
--- a/clang/lib/Sema/SemaOpenACC.cpp
+++ b/clang/lib/Sema/SemaOpenACC.cpp
@@ -423,6 +423,21 @@ ExprResult SemaOpenACC::ActOnIntExpr(OpenACCDirectiveKind DK,
   return IntExpr;
 }
 
+ExprResult SemaOpenACC::ActOnArraySectionExpr(Expr *Base, SourceLocation LBLoc,
+                                              Expr *LowerBound,
+                                              SourceLocation ColonLoc,
+                                              Expr *Length,
+                                              SourceLocation RBLoc) {
+  ASTContext &Context = getASTContext();
+
+  // TODO OpenACC: We likely have to reproduce a lot of the same logic from the
+  // OMP version of this, but at the moment we don't have a good way to test it,
+  // so for now we'll just create the node.
+  return new (Context)
+      ArraySectionExpr(Base, LowerBound, Length, Context.ArraySectionTy,
+                       VK_LValue, OK_Ordinary, ColonLoc, RBLoc);
+}
+
 bool SemaOpenACC::ActOnStartStmtDirective(OpenACCDirectiveKind K,
                                           SourceLocation StartLoc) {
   return diagnoseConstructAppertainment(*this, K, StartLoc, /*IsStmt=*/true);
diff --git a/clang/lib/Sema/SemaOpenMP.cpp b/clang/lib/Sema/SemaOpenMP.cpp
index 5ba09926acf2..cee8da495c54 100644
--- a/clang/lib/Sema/SemaOpenMP.cpp
+++ b/clang/lib/Sema/SemaOpenMP.cpp
@@ -2230,7 +2230,7 @@ bool SemaOpenMP::isOpenMPCapturedByRef(const ValueDecl *D, unsigned Level,
               dyn_cast<UnaryOperator>(Last->getAssociatedExpression());
           if ((UO && UO->getOpcode() == UO_Deref) ||
               isa<ArraySubscriptExpr>(Last->getAssociatedExpression()) ||
-              isa<OMPArraySectionExpr>(Last->getAssociatedExpression()) ||
+              isa<ArraySectionExpr>(Last->getAssociatedExpression()) ||
               isa<MemberExpr>(EI->getAssociatedExpression()) ||
               isa<OMPArrayShapingExpr>(Last->getAssociatedExpression())) {
             IsVariableAssociatedWithSection = true;
@@ -3884,7 +3884,7 @@ public:
                                     MappableComponent &MC) {
                                return MC.getAssociatedDeclaration() ==
                                           nullptr &&
-                                      (isa<OMPArraySectionExpr>(
+                                      (isa<ArraySectionExpr>(
                                            MC.getAssociatedExpression()) ||
                                        isa<OMPArrayShapingExpr>(
                                            MC.getAssociatedExpression()) ||
@@ -4062,7 +4062,7 @@ public:
                   // Do both expressions have the same kind?
                   if (CCI->getAssociatedExpression()->getStmtClass() !=
                       SC.getAssociatedExpression()->getStmtClass())
-                    if (!((isa<OMPArraySectionExpr>(
+                    if (!((isa<ArraySectionExpr>(
                                SC.getAssociatedExpression()) ||
                            isa<OMPArrayShapingExpr>(
                                SC.getAssociatedExpression())) &&
@@ -5428,9 +5428,9 @@ static std::pair<ValueDecl *, bool> getPrivateItem(Sema &S, Expr *&RefExpr,
         Base = TempASE->getBase()->IgnoreParenImpCasts();
       RefExpr = Base;
       IsArrayExpr = ArraySubscript;
-    } else if (auto *OASE = dyn_cast_or_null<OMPArraySectionExpr>(RefExpr)) {
+    } else if (auto *OASE = dyn_cast_or_null<ArraySectionExpr>(RefExpr)) {
       Expr *Base = OASE->getBase()->IgnoreParenImpCasts();
-      while (auto *TempOASE = dyn_cast<OMPArraySectionExpr>(Base))
+      while (auto *TempOASE = dyn_cast<ArraySectionExpr>(Base))
         Base = TempOASE->getBase()->IgnoreParenImpCasts();
       while (auto *TempASE = dyn_cast<ArraySubscriptExpr>(Base))
         Base = TempASE->getBase()->IgnoreParenImpCasts();
@@ -6060,10 +6060,10 @@ processImplicitMapsWithDefaultMappers(Sema &S, DSAStackTy *Stack,
       // Array section - need to check for the mapping of the array section
       // element.
       QualType CanonType = E->getType().getCanonicalType();
-      if (CanonType->isSpecificBuiltinType(BuiltinType::OMPArraySection)) {
-        const auto *OASE = cast<OMPArraySectionExpr>(E->IgnoreParenImpCasts());
+      if (CanonType->isSpecificBuiltinType(BuiltinType::ArraySection)) {
+        const auto *OASE = cast<ArraySectionExpr>(E->IgnoreParenImpCasts());
         QualType BaseType =
-            OMPArraySectionExpr::getBaseOriginalType(OASE->getBase());
+            ArraySectionExpr::getBaseOriginalType(OASE->getBase());
         QualType ElemType;
         if (const auto *ATy = BaseType->getAsArrayTypeUnsafe())
           ElemType = ATy->getElementType();
@@ -19513,7 +19513,7 @@ struct ReductionData {
 } // namespace
 
 static bool checkOMPArraySectionConstantForReduction(
-    ASTContext &Context, const OMPArraySectionExpr *OASE, bool &SingleElement,
+    ASTContext &Context, const ArraySectionExpr *OASE, bool &SingleElement,
     SmallVectorImpl<llvm::APSInt> &ArraySizes) {
   const Expr *Length = OASE->getLength();
   if (Length == nullptr) {
@@ -19540,7 +19540,7 @@ static bool checkOMPArraySectionConstantForReduction(
 
   // We require length = 1 for all array sections except the right-most to
   // guarantee that the memory region is contiguous and has no holes in it.
-  while (const auto *TempOASE = dyn_cast<OMPArraySectionExpr>(Base)) {
+  while (const auto *TempOASE = dyn_cast<ArraySectionExpr>(Base)) {
     Length = TempOASE->getLength();
     if (Length == nullptr) {
       // For array sections of the form [1:] or [:], we would need to analyze
@@ -19745,12 +19745,12 @@ static bool actOnOMPReductionKindClause(
     Expr *TaskgroupDescriptor = nullptr;
     QualType Type;
     auto *ASE = dyn_cast<ArraySubscriptExpr>(RefExpr->IgnoreParens());
-    auto *OASE = dyn_cast<OMPArraySectionExpr>(RefExpr->IgnoreParens());
+    auto *OASE = dyn_cast<ArraySectionExpr>(RefExpr->IgnoreParens());
     if (ASE) {
       Type = ASE->getType().getNonReferenceType();
     } else if (OASE) {
       QualType BaseType =
-          OMPArraySectionExpr::getBaseOriginalType(OASE->getBase());
+          ArraySectionExpr::getBaseOriginalType(OASE->getBase());
       if (const auto *ATy = BaseType->getAsArrayTypeUnsafe())
         Type = ATy->getElementType();
       else
@@ -21284,10 +21284,10 @@ OMPClause *SemaOpenMP::ActOnOpenMPDependClause(
           // List items used in depend clauses cannot be zero-length array
           // sections.
           QualType ExprTy = RefExpr->getType().getNonReferenceType();
-          const auto *OASE = dyn_cast<OMPArraySectionExpr>(SimpleExpr);
+          const auto *OASE = dyn_cast<ArraySectionExpr>(SimpleExpr);
           if (OASE) {
             QualType BaseType =
-                OMPArraySectionExpr::getBaseOriginalType(OASE->getBase());
+                ArraySectionExpr::getBaseOriginalType(OASE->getBase());
             if (BaseType.isNull())
               return nullptr;
             if (const auto *ATy = BaseType->getAsArrayTypeUnsafe())
@@ -21346,7 +21346,7 @@ OMPClause *SemaOpenMP::ActOnOpenMPDependClause(
             Res = SemaRef.CreateBuiltinUnaryOp(ELoc, UO_AddrOf,
                                                RefExpr->IgnoreParenImpCasts());
           }
-          if (!Res.isUsable() && !isa<OMPArraySectionExpr>(SimpleExpr) &&
+          if (!Res.isUsable() && !isa<ArraySectionExpr>(SimpleExpr) &&
               !isa<OMPArrayShapingExpr>(SimpleExpr)) {
             Diag(ELoc, diag::err_omp_expected_addressable_lvalue_or_array_item)
                 << (getLangOpts().OpenMP >= 50 ? 1 : 0)
@@ -21447,7 +21447,7 @@ static bool checkTypeMappable(SourceLocation SL, SourceRange SR, Sema &SemaRef,
 static bool checkArrayExpressionDoesNotReferToWholeSize(Sema &SemaRef,
                                                         const Expr *E,
                                                         QualType BaseQTy) {
-  const auto *OASE = dyn_cast<OMPArraySectionExpr>(E);
+  const auto *OASE = dyn_cast<ArraySectionExpr>(E);
 
   // If this is an array subscript, it refers to the whole size if the size of
   // the dimension is constant and equals 1. Also, an array section assumes the
@@ -21505,7 +21505,7 @@ static bool checkArrayExpressionDoesNotReferToWholeSize(Sema &SemaRef,
 static bool checkArrayExpressionDoesNotReferToUnitySize(Sema &SemaRef,
                                                         const Expr *E,
                                                         QualType BaseQTy) {
-  const auto *OASE = dyn_cast<OMPArraySectionExpr>(E);
+  const auto *OASE = dyn_cast<ArraySectionExpr>(E);
 
   // An array subscript always refer to a single element. Also, an array section
   // assumes the format of an array subscript if no colon is used.
@@ -21720,14 +21720,14 @@ public:
     return RelevantExpr || Visit(E);
   }
 
-  bool VisitOMPArraySectionExpr(OMPArraySectionExpr *OASE) {
+  bool VisitArraySectionExpr(ArraySectionExpr *OASE) {
     // After OMP 5.0  Array section in reduction clause will be implicitly
     // mapped
     assert(!(SemaRef.getLangOpts().OpenMP < 50 && NoDiagnose) &&
            "Array sections cannot be implicitly mapped.");
     Expr *E = OASE->getBase()->IgnoreParenImpCasts();
     QualType CurType =
-        OMPArraySectionExpr::getBaseOriginalType(E).getCanonicalType();
+        ArraySectionExpr::getBaseOriginalType(E).getCanonicalType();
 
     // OpenMP 4.5 [2.15.5.1, map Clause, Restrictions, C++, p.1]
     //  If the type of a list item is a reference to a type T then the type
@@ -21900,7 +21900,7 @@ static const Expr *checkMapClauseExpressionBase(
       auto CE = CurComponents.rend();
       for (; CI != CE; ++CI) {
         const auto *OASE =
-            dyn_cast<OMPArraySectionExpr>(CI->getAssociatedExpression());
+            dyn_cast<ArraySectionExpr>(CI->getAssociatedExpression());
         if (!OASE)
           continue;
         if (OASE && OASE->getLength())
@@ -21970,10 +21970,10 @@ static bool checkMapConflicts(
           //  variable in map clauses of the same construct.
           if (CurrentRegionOnly &&
               (isa<ArraySubscriptExpr>(CI->getAssociatedExpression()) ||
-               isa<OMPArraySectionExpr>(CI->getAssociatedExpression()) ||
+               isa<ArraySectionExpr>(CI->getAssociatedExpression()) ||
                isa<OMPArrayShapingExpr>(CI->getAssociatedExpression())) &&
               (isa<ArraySubscriptExpr>(SI->getAssociatedExpression()) ||
-               isa<OMPArraySectionExpr>(SI->getAssociatedExpression()) ||
+               isa<ArraySectionExpr>(SI->getAssociatedExpression()) ||
                isa<OMPArrayShapingExpr>(SI->getAssociatedExpression()))) {
             SemaRef.Diag(CI->getAssociatedExpression()->getExprLoc(),
                          diag::err_omp_multiple_array_items_in_map_clause)
@@ -22001,11 +22001,10 @@ static bool checkMapConflicts(
           if (const auto *ASE =
                   dyn_cast<ArraySubscriptExpr>(SI->getAssociatedExpression())) {
             Type = ASE->getBase()->IgnoreParenImpCasts()->getType();
-          } else if (const auto *OASE = dyn_cast<OMPArraySectionExpr>(
+          } else if (const auto *OASE = dyn_cast<ArraySectionExpr>(
                          SI->getAssociatedExpression())) {
             const Expr *E = OASE->getBase()->IgnoreParenImpCasts();
-            Type =
-                OMPArraySectionExpr::getBaseOriginalType(E).getCanonicalType();
+            Type = ArraySectionExpr::getBaseOriginalType(E).getCanonicalType();
           } else if (const auto *OASE = dyn_cast<OMPArrayShapingExpr>(
                          SI->getAssociatedExpression())) {
             Type = OASE->getBase()->getType()->getPointeeType();
@@ -22480,13 +22479,13 @@ static void checkMappableExpressionList(
     (void)I;
     QualType Type;
     auto *ASE = dyn_cast<ArraySubscriptExpr>(VE->IgnoreParens());
-    auto *OASE = dyn_cast<OMPArraySectionExpr>(VE->IgnoreParens());
+    auto *OASE = dyn_cast<ArraySectionExpr>(VE->IgnoreParens());
     auto *OAShE = dyn_cast<OMPArrayShapingExpr>(VE->IgnoreParens());
     if (ASE) {
       Type = ASE->getType().getNonReferenceType();
     } else if (OASE) {
       QualType BaseType =
-          OMPArraySectionExpr::getBaseOriginalType(OASE->getBase());
+          ArraySectionExpr::getBaseOriginalType(OASE->getBase());
       if (const auto *ATy = BaseType->getAsArrayTypeUnsafe())
         Type = ATy->getElementType();
       else
@@ -23955,7 +23954,7 @@ SemaOpenMP::ActOnOpenMPUseDeviceAddrClause(ArrayRef<Expr *> VarList,
     MVLI.VarBaseDeclarations.push_back(D);
     MVLI.VarComponents.emplace_back();
     Expr *Component = SimpleRefExpr;
-    if (VD && (isa<OMPArraySectionExpr>(RefExpr->IgnoreParenImpCasts()) ||
+    if (VD && (isa<ArraySectionExpr>(RefExpr->IgnoreParenImpCasts()) ||
                isa<ArraySubscriptExpr>(RefExpr->IgnoreParenImpCasts())))
       Component =
           SemaRef.DefaultFunctionArrayLvalueConversion(SimpleRefExpr).get();
@@ -24105,7 +24104,7 @@ SemaOpenMP::ActOnOpenMPHasDeviceAddrClause(ArrayRef<Expr *> VarList,
     // against other clauses later on.
     Expr *Component = SimpleRefExpr;
     auto *VD = dyn_cast<VarDecl>(D);
-    if (VD && (isa<OMPArraySectionExpr>(RefExpr->IgnoreParenImpCasts()) ||
+    if (VD && (isa<ArraySectionExpr>(RefExpr->IgnoreParenImpCasts()) ||
                isa<ArraySubscriptExpr>(RefExpr->IgnoreParenImpCasts())))
       Component =
           SemaRef.DefaultFunctionArrayLvalueConversion(SimpleRefExpr).get();
@@ -24519,7 +24518,7 @@ OMPClause *SemaOpenMP::ActOnOpenMPAffinityClause(
       Sema::TentativeAnalysisScope Trap(SemaRef);
       Res = SemaRef.CreateBuiltinUnaryOp(ELoc, UO_AddrOf, SimpleExpr);
     }
-    if (!Res.isUsable() && !isa<OMPArraySectionExpr>(SimpleExpr) &&
+    if (!Res.isUsable() && !isa<ArraySectionExpr>(SimpleExpr) &&
         !isa<OMPArrayShapingExpr>(SimpleExpr)) {
       Diag(ELoc, diag::err_omp_expected_addressable_lvalue_or_array_item)
           << 1 << 0 << RefExpr->getSourceRange();
@@ -24632,7 +24631,7 @@ ExprResult SemaOpenMP::ActOnOMPArraySectionExpr(
     Expr *Stride, SourceLocation RBLoc) {
   ASTContext &Context = getASTContext();
   if (Base->hasPlaceholderType() &&
-      !Base->hasPlaceholderType(BuiltinType::OMPArraySection)) {
+      !Base->hasPlaceholderType(BuiltinType::ArraySection)) {
     ExprResult Result = SemaRef.CheckPlaceholderExpr(Base);
     if (Result.isInvalid())
       return ExprError();
@@ -24672,13 +24671,13 @@ ExprResult SemaOpenMP::ActOnOMPArraySectionExpr(
        (LowerBound->isTypeDependent() || LowerBound->isValueDependent())) ||
       (Length && (Length->isTypeDependent() || Length->isValueDependent())) ||
       (Stride && (Stride->isTypeDependent() || Stride->isValueDependent()))) {
-    return new (Context) OMPArraySectionExpr(
+    return new (Context) ArraySectionExpr(
         Base, LowerBound, Length, Stride, Context.DependentTy, VK_LValue,
         OK_Ordinary, ColonLocFirst, ColonLocSecond, RBLoc);
   }
 
   // Perform default conversions.
-  QualType OriginalTy = OMPArraySectionExpr::getBaseOriginalType(Base);
+  QualType OriginalTy = ArraySectionExpr::getBaseOriginalType(Base);
   QualType ResultTy;
   if (OriginalTy->isAnyPointerType()) {
     ResultTy = OriginalTy->getPointeeType();
@@ -24801,14 +24800,14 @@ ExprResult SemaOpenMP::ActOnOMPArraySectionExpr(
     }
   }
 
-  if (!Base->hasPlaceholderType(BuiltinType::OMPArraySection)) {
+  if (!Base->hasPlaceholderType(BuiltinType::ArraySection)) {
     ExprResult Result = SemaRef.DefaultFunctionArrayLvalueConversion(Base);
     if (Result.isInvalid())
       return ExprError();
     Base = Result.get();
   }
-  return new (Context) OMPArraySectionExpr(
-      Base, LowerBound, Length, Stride, Context.OMPArraySectionTy, VK_LValue,
+  return new (Context) ArraySectionExpr(
+      Base, LowerBound, Length, Stride, Context.ArraySectionTy, VK_LValue,
       OK_Ordinary, ColonLocFirst, ColonLocSecond, RBLoc);
 }
 
diff --git a/clang/lib/Sema/SemaRISCVVectorLookup.cpp b/clang/lib/Sema/SemaRISCVVectorLookup.cpp
index bf89a4ac51af..26e13e87b1d6 100644
--- a/clang/lib/Sema/SemaRISCVVectorLookup.cpp
+++ b/clang/lib/Sema/SemaRISCVVectorLookup.cpp
@@ -216,6 +216,7 @@ void RISCVIntrinsicManagerImpl::ConstructRVVIntrinsics(
       {"zvksed", RVV_REQ_Zvksed},
       {"zvksh", RVV_REQ_Zvksh},
       {"zvfbfwma", RVV_REQ_Zvfbfwma},
+      {"zvfbfmin", RVV_REQ_Zvfbfmin},
       {"experimental", RVV_REQ_Experimental}};
 
   // Construction of RVVIntrinsicRecords need to sync with createRVVIntrinsics
diff --git a/clang/lib/Sema/SemaStmtAttr.cpp b/clang/lib/Sema/SemaStmtAttr.cpp
index 9d44c22c8ddc..1c84830b6ddd 100644
--- a/clang/lib/Sema/SemaStmtAttr.cpp
+++ b/clang/lib/Sema/SemaStmtAttr.cpp
@@ -109,16 +109,14 @@ static Attr *handleLoopHintAttr(Sema &S, Stmt *St, const ParsedAttr &A,
     SetHints(LoopHintAttr::Unroll, LoopHintAttr::Disable);
   } else if (PragmaName == "unroll") {
     // #pragma unroll N
-    if (ValueExpr && !ValueExpr->isValueDependent()) {
-      llvm::APSInt ValueAPS;
-      ExprResult R = S.VerifyIntegerConstantExpression(ValueExpr, &ValueAPS);
-      assert(!R.isInvalid() && "unroll count value must be a valid value, it's "
-                               "should be checked in Sema::CheckLoopHintExpr");
-      (void)R;
-      // The values of 0 and 1 block any unrolling of the loop.
-      if (ValueAPS.isZero() || ValueAPS.isOne())
-        SetHints(LoopHintAttr::UnrollCount, LoopHintAttr::Disable);
-      else
+    if (ValueExpr) {
+      if (!ValueExpr->isValueDependent()) {
+        auto Value = ValueExpr->EvaluateKnownConstInt(S.getASTContext());
+        if (Value.isZero() || Value.isOne())
+          SetHints(LoopHintAttr::Unroll, LoopHintAttr::Disable);
+        else
+          SetHints(LoopHintAttr::UnrollCount, LoopHintAttr::Numeric);
+      } else
         SetHints(LoopHintAttr::UnrollCount, LoopHintAttr::Numeric);
     } else
       SetHints(LoopHintAttr::Unroll, LoopHintAttr::Enable);
diff --git a/clang/lib/Sema/SemaTemplate.cpp b/clang/lib/Sema/SemaTemplate.cpp
index 4bda31ba67c0..bbcb7c33a985 100644
--- a/clang/lib/Sema/SemaTemplate.cpp
+++ b/clang/lib/Sema/SemaTemplate.cpp
@@ -9460,6 +9460,7 @@ DeclResult Sema::ActOnClassTemplateSpecialization(
       Diag(TemplateNameLoc, diag::err_partial_spec_fully_specialized)
         << ClassTemplate->getDeclName();
       isPartialSpecialization = false;
+      Invalid = true;
     }
   }
 
@@ -9675,6 +9676,7 @@ DeclResult Sema::ActOnClassTemplateSpecialization(
   if (SkipBody && SkipBody->ShouldSkip)
     return SkipBody->Previous;
 
+  Specialization->setInvalidDecl(Invalid);
   return Specialization;
 }
 
diff --git a/clang/lib/Sema/SemaTemplateDeduction.cpp b/clang/lib/Sema/SemaTemplateDeduction.cpp
index 0b6375001f53..c3815bca0385 100644
--- a/clang/lib/Sema/SemaTemplateDeduction.cpp
+++ b/clang/lib/Sema/SemaTemplateDeduction.cpp
@@ -1914,6 +1914,9 @@ static TemplateDeductionResult DeduceTemplateArgumentsByTypeMatch(
       if (!S.isCompleteType(Info.getLocation(), A))
         return Result;
 
+      if (getCanonicalRD(A)->isInvalidDecl())
+        return Result;
+
       // Reset the incorrectly deduced argument from above.
       Deduced = DeducedOrig;
 
diff --git a/clang/lib/Sema/SemaTemplateInstantiate.cpp b/clang/lib/Sema/SemaTemplateInstantiate.cpp
index 98d5c7cb3a8a..3a9fd906b7af 100644
--- a/clang/lib/Sema/SemaTemplateInstantiate.cpp
+++ b/clang/lib/Sema/SemaTemplateInstantiate.cpp
@@ -2151,13 +2151,25 @@ TemplateInstantiator::TransformLoopHintAttr(const LoopHintAttr *LH) {
 
   // Generate error if there is a problem with the value.
   if (getSema().CheckLoopHintExpr(TransformedExpr, LH->getLocation(),
-                                  LH->getOption() == LoopHintAttr::UnrollCount))
+                                  LH->getSemanticSpelling() ==
+                                      LoopHintAttr::Pragma_unroll))
     return LH;
 
+  LoopHintAttr::OptionType Option = LH->getOption();
+  LoopHintAttr::LoopHintState State = LH->getState();
+
+  llvm::APSInt ValueAPS =
+      TransformedExpr->EvaluateKnownConstInt(getSema().getASTContext());
+  // The values of 0 and 1 block any unrolling of the loop.
+  if (ValueAPS.isZero() || ValueAPS.isOne()) {
+    Option = LoopHintAttr::Unroll;
+    State = LoopHintAttr::Disable;
+  }
+
   // Create new LoopHintValueAttr with integral expression in place of the
   // non-type template parameter.
-  return LoopHintAttr::CreateImplicit(getSema().Context, LH->getOption(),
-                                      LH->getState(), TransformedExpr, *LH);
+  return LoopHintAttr::CreateImplicit(getSema().Context, Option, State,
+                                      TransformedExpr, *LH);
 }
 const NoInlineAttr *TemplateInstantiator::TransformStmtNoInlineAttr(
     const Stmt *OrigS, const Stmt *InstS, const NoInlineAttr *A) {
diff --git a/clang/lib/Sema/SemaTemplateInstantiateDecl.cpp b/clang/lib/Sema/SemaTemplateInstantiateDecl.cpp
index 787a485e0b2f..d544cfac55ba 100644
--- a/clang/lib/Sema/SemaTemplateInstantiateDecl.cpp
+++ b/clang/lib/Sema/SemaTemplateInstantiateDecl.cpp
@@ -5184,6 +5184,7 @@ void Sema::InstantiateFunctionDefinition(SourceLocation PointOfInstantiation,
 
     ParmVarDecl *Parm = Function->getParamDecl(0);
     TypeSourceInfo *NewParmSI = IR.TransformType(Parm->getTypeSourceInfo());
+    assert(NewParmSI && "Type transformation failed.");
     Parm->setType(NewParmSI->getType());
     Parm->setTypeSourceInfo(NewParmSI);
   };
diff --git a/clang/lib/Sema/TreeTransform.h b/clang/lib/Sema/TreeTransform.h
index 9404be5a46f3..f47bc219e6fa 100644
--- a/clang/lib/Sema/TreeTransform.h
+++ b/clang/lib/Sema/TreeTransform.h
@@ -2784,15 +2784,23 @@ public:
   ///
   /// By default, performs semantic analysis to build the new expression.
   /// Subclasses may override this routine to provide different behavior.
-  ExprResult RebuildOMPArraySectionExpr(Expr *Base, SourceLocation LBracketLoc,
-                                        Expr *LowerBound,
-                                        SourceLocation ColonLocFirst,
-                                        SourceLocation ColonLocSecond,
-                                        Expr *Length, Expr *Stride,
-                                        SourceLocation RBracketLoc) {
-    return getSema().OpenMP().ActOnOMPArraySectionExpr(
-        Base, LBracketLoc, LowerBound, ColonLocFirst, ColonLocSecond, Length,
-        Stride, RBracketLoc);
+  ExprResult RebuildArraySectionExpr(bool IsOMPArraySection, Expr *Base,
+                                     SourceLocation LBracketLoc,
+                                     Expr *LowerBound,
+                                     SourceLocation ColonLocFirst,
+                                     SourceLocation ColonLocSecond,
+                                     Expr *Length, Expr *Stride,
+                                     SourceLocation RBracketLoc) {
+    if (IsOMPArraySection)
+      return getSema().OpenMP().ActOnOMPArraySectionExpr(
+          Base, LBracketLoc, LowerBound, ColonLocFirst, ColonLocSecond, Length,
+          Stride, RBracketLoc);
+
+    assert(Stride == nullptr && !ColonLocSecond.isValid() &&
+           "Stride/second colon not allowed for OpenACC");
+
+    return getSema().OpenACC().ActOnArraySectionExpr(
+        Base, LBracketLoc, LowerBound, ColonLocFirst, Length, RBracketLoc);
   }
 
   /// Build a new array shaping expression.
@@ -11742,7 +11750,7 @@ TreeTransform<Derived>::TransformMatrixSubscriptExpr(MatrixSubscriptExpr *E) {
 
 template <typename Derived>
 ExprResult
-TreeTransform<Derived>::TransformOMPArraySectionExpr(OMPArraySectionExpr *E) {
+TreeTransform<Derived>::TransformArraySectionExpr(ArraySectionExpr *E) {
   ExprResult Base = getDerived().TransformExpr(E->getBase());
   if (Base.isInvalid())
     return ExprError();
@@ -11762,20 +11770,25 @@ TreeTransform<Derived>::TransformOMPArraySectionExpr(OMPArraySectionExpr *E) {
   }
 
   ExprResult Stride;
-  if (Expr *Str = E->getStride()) {
-    Stride = getDerived().TransformExpr(Str);
-    if (Stride.isInvalid())
-      return ExprError();
+  if (E->isOMPArraySection()) {
+    if (Expr *Str = E->getStride()) {
+      Stride = getDerived().TransformExpr(Str);
+      if (Stride.isInvalid())
+        return ExprError();
+    }
   }
 
   if (!getDerived().AlwaysRebuild() && Base.get() == E->getBase() &&
-      LowerBound.get() == E->getLowerBound() && Length.get() == E->getLength())
+      LowerBound.get() == E->getLowerBound() &&
+      Length.get() == E->getLength() &&
+      (E->isOpenACCArraySection() || Stride.get() == E->getStride()))
     return E;
 
-  return getDerived().RebuildOMPArraySectionExpr(
-      Base.get(), E->getBase()->getEndLoc(), LowerBound.get(),
-      E->getColonLocFirst(), E->getColonLocSecond(), Length.get(), Stride.get(),
-      E->getRBracketLoc());
+  return getDerived().RebuildArraySectionExpr(
+      E->isOMPArraySection(), Base.get(), E->getBase()->getEndLoc(),
+      LowerBound.get(), E->getColonLocFirst(),
+      E->isOMPArraySection() ? E->getColonLocSecond() : SourceLocation{},
+      Length.get(), Stride.get(), E->getRBracketLoc());
 }
 
 template <typename Derived>
@@ -14186,6 +14199,8 @@ TreeTransform<Derived>::TransformLambdaExpr(LambdaExpr *E) {
   // FIXME: Sema's lambda-building mechanism expects us to push an expression
   // evaluation context even if we're not transforming the function body.
   getSema().PushExpressionEvaluationContext(
+      E->getCallOperator()->isConsteval() ?
+      Sema::ExpressionEvaluationContext::ImmediateFunctionContext :
       Sema::ExpressionEvaluationContext::PotentiallyEvaluated);
 
   Sema::CodeSynthesisContext C;
diff --git a/clang/lib/Serialization/ASTCommon.cpp b/clang/lib/Serialization/ASTCommon.cpp
index f8d54c0c3989..e017f5bdb488 100644
--- a/clang/lib/Serialization/ASTCommon.cpp
+++ b/clang/lib/Serialization/ASTCommon.cpp
@@ -261,8 +261,8 @@ serialization::TypeIdxFromBuiltin(const BuiltinType *BT) {
   case BuiltinType::IncompleteMatrixIdx:
     ID = PREDEF_TYPE_INCOMPLETE_MATRIX_IDX;
     break;
-  case BuiltinType::OMPArraySection:
-    ID = PREDEF_TYPE_OMP_ARRAY_SECTION;
+  case BuiltinType::ArraySection:
+    ID = PREDEF_TYPE_ARRAY_SECTION;
     break;
   case BuiltinType::OMPArrayShaping:
     ID = PREDEF_TYPE_OMP_ARRAY_SHAPING;
diff --git a/clang/lib/Serialization/ASTReader.cpp b/clang/lib/Serialization/ASTReader.cpp
index d64925676df7..0ef57a3ea804 100644
--- a/clang/lib/Serialization/ASTReader.cpp
+++ b/clang/lib/Serialization/ASTReader.cpp
@@ -954,14 +954,16 @@ ASTSelectorLookupTrait::ReadData(Selector, const unsigned char* d,
   // Load instance methods
   for (unsigned I = 0; I != NumInstanceMethods; ++I) {
     if (ObjCMethodDecl *Method = Reader.GetLocalDeclAs<ObjCMethodDecl>(
-            F, endian::readNext<uint32_t, llvm::endianness::little>(d)))
+            F,
+            LocalDeclID(endian::readNext<DeclID, llvm::endianness::little>(d))))
       Result.Instance.push_back(Method);
   }
 
   // Load factory methods
   for (unsigned I = 0; I != NumFactoryMethods; ++I) {
     if (ObjCMethodDecl *Method = Reader.GetLocalDeclAs<ObjCMethodDecl>(
-            F, endian::readNext<uint32_t, llvm::endianness::little>(d)))
+            F,
+            LocalDeclID(endian::readNext<DeclID, llvm::endianness::little>(d))))
       Result.Factory.push_back(Method);
   }
 
@@ -1088,10 +1090,11 @@ IdentifierInfo *ASTIdentifierLookupTrait::ReadData(const internal_key_type& k,
   // Read all of the declarations visible at global scope with this
   // name.
   if (DataLen > 0) {
-    SmallVector<uint32_t, 4> DeclIDs;
-    for (; DataLen > 0; DataLen -= 4)
+    SmallVector<GlobalDeclID, 4> DeclIDs;
+    for (; DataLen > 0; DataLen -= sizeof(DeclID))
       DeclIDs.push_back(Reader.getGlobalDeclID(
-          F, endian::readNext<uint32_t, llvm::endianness::little>(d)));
+          F,
+          LocalDeclID(endian::readNext<DeclID, llvm::endianness::little>(d))));
     Reader.SetGloballyVisibleDecls(II, DeclIDs);
   }
 
@@ -1211,8 +1214,8 @@ void ASTDeclContextNameLookupTrait::ReadDataInto(internal_key_type,
                                                  data_type_builder &Val) {
   using namespace llvm::support;
 
-  for (unsigned NumDecls = DataLen / 4; NumDecls; --NumDecls) {
-    uint32_t LocalID = endian::readNext<uint32_t, llvm::endianness::little>(d);
+  for (unsigned NumDecls = DataLen / sizeof(DeclID); NumDecls; --NumDecls) {
+    LocalDeclID LocalID(endian::readNext<DeclID, llvm::endianness::little>(d));
     Val.insert(Reader.getGlobalDeclID(F, LocalID));
   }
 }
@@ -1259,9 +1262,8 @@ bool ASTReader::ReadLexicalDeclContextStorage(ModuleFile &M,
   if (!Lex.first) {
     Lex = std::make_pair(
         &M, llvm::ArrayRef(
-                reinterpret_cast<const llvm::support::unaligned_uint32_t *>(
-                    Blob.data()),
-                Blob.size() / 4));
+                reinterpret_cast<const unalighed_decl_id_t *>(Blob.data()),
+                Blob.size() / sizeof(DeclID)));
   }
   DC->setHasExternalLexicalStorage(true);
   return false;
@@ -1270,7 +1272,7 @@ bool ASTReader::ReadLexicalDeclContextStorage(ModuleFile &M,
 bool ASTReader::ReadVisibleDeclContextStorage(ModuleFile &M,
                                               BitstreamCursor &Cursor,
                                               uint64_t Offset,
-                                              DeclID ID) {
+                                              GlobalDeclID ID) {
   assert(Offset != 0);
 
   SavedStreamPosition SavedPosition(Cursor);
@@ -1653,7 +1655,7 @@ bool ASTReader::ReadSLocEntry(int ID) {
 
     unsigned NumFileDecls = Record[7];
     if (NumFileDecls && ContextObj) {
-      const DeclID *FirstDecl = F->FileSortedDecls + Record[6];
+      const LocalDeclID *FirstDecl = F->FileSortedDecls + Record[6];
       assert(F->FileSortedDecls && "FILE_SORTED_DECLS not encountered yet ?");
       FileDeclIDs[FID] =
           FileDeclsInfo(F, llvm::ArrayRef(FirstDecl, NumFileDecls));
@@ -3369,8 +3371,8 @@ llvm::Error ASTReader::ReadASTBlock(ModuleFile &F,
       if (F.LocalNumDecls > 0) {
         // Introduce the global -> local mapping for declarations within this
         // module.
-        GlobalDeclMap.insert(
-          std::make_pair(getTotalNumDecls() + NUM_PREDEF_DECL_IDS, &F));
+        GlobalDeclMap.insert(std::make_pair(
+            GlobalDeclID(getTotalNumDecls() + NUM_PREDEF_DECL_IDS), &F));
 
         // Introduce the local -> global mapping for declarations within this
         // module.
@@ -3389,9 +3391,8 @@ llvm::Error ASTReader::ReadASTBlock(ModuleFile &F,
     case TU_UPDATE_LEXICAL: {
       DeclContext *TU = ContextObj->getTranslationUnitDecl();
       LexicalContents Contents(
-          reinterpret_cast<const llvm::support::unaligned_uint32_t *>(
-              Blob.data()),
-          static_cast<unsigned int>(Blob.size() / 4));
+          reinterpret_cast<const unalighed_decl_id_t *>(Blob.data()),
+          static_cast<unsigned int>(Blob.size() / sizeof(DeclID)));
       TULexicalDecls.push_back(std::make_pair(&F, Contents));
       TU->setHasExternalLexicalStorage(true);
       break;
@@ -3399,7 +3400,7 @@ llvm::Error ASTReader::ReadASTBlock(ModuleFile &F,
 
     case UPDATE_VISIBLE: {
       unsigned Idx = 0;
-      serialization::DeclID ID = ReadDeclID(F, Record, Idx);
+      GlobalDeclID ID = ReadDeclID(F, Record, Idx);
       auto *Data = (const unsigned char*)Blob.data();
       PendingVisibleUpdates[ID].push_back(PendingVisibleUpdate{&F, Data});
       // If we've already loaded the decl, perform the updates when we finish
@@ -3460,7 +3461,8 @@ llvm::Error ASTReader::ReadASTBlock(ModuleFile &F,
       // FIXME: Skip reading this record if our ASTConsumer doesn't care
       // about "interesting" decls (for instance, if we're building a module).
       for (unsigned I = 0, N = Record.size(); I != N; ++I)
-        EagerlyDeserializedDecls.push_back(getGlobalDeclID(F, Record[I]));
+        EagerlyDeserializedDecls.push_back(
+            getGlobalDeclID(F, LocalDeclID(Record[I])));
       break;
 
     case MODULAR_CODEGEN_DECLS:
@@ -3469,7 +3471,8 @@ llvm::Error ASTReader::ReadASTBlock(ModuleFile &F,
       if (F.Kind == MK_MainFile ||
           getContext().getLangOpts().BuildingPCHWithObjectFile)
         for (unsigned I = 0, N = Record.size(); I != N; ++I)
-          EagerlyDeserializedDecls.push_back(getGlobalDeclID(F, Record[I]));
+          EagerlyDeserializedDecls.push_back(
+              getGlobalDeclID(F, LocalDeclID(Record[I])));
       break;
 
     case SPECIAL_TYPES:
@@ -3501,12 +3504,14 @@ llvm::Error ASTReader::ReadASTBlock(ModuleFile &F,
 
     case UNUSED_FILESCOPED_DECLS:
       for (unsigned I = 0, N = Record.size(); I != N; ++I)
-        UnusedFileScopedDecls.push_back(getGlobalDeclID(F, Record[I]));
+        UnusedFileScopedDecls.push_back(
+            getGlobalDeclID(F, LocalDeclID(Record[I])));
       break;
 
     case DELEGATING_CTORS:
       for (unsigned I = 0, N = Record.size(); I != N; ++I)
-        DelegatingCtorDecls.push_back(getGlobalDeclID(F, Record[I]));
+        DelegatingCtorDecls.push_back(
+            getGlobalDeclID(F, LocalDeclID(Record[I])));
       break;
 
     case WEAK_UNDECLARED_IDENTIFIERS:
@@ -3614,7 +3619,7 @@ llvm::Error ASTReader::ReadASTBlock(ModuleFile &F,
       break;
 
     case FILE_SORTED_DECLS:
-      F.FileSortedDecls = (const DeclID *)Blob.data();
+      F.FileSortedDecls = (const LocalDeclID *)Blob.data();
       F.NumFileSortedDecls = Record[0];
       break;
 
@@ -3669,7 +3674,7 @@ llvm::Error ASTReader::ReadASTBlock(ModuleFile &F,
 
     case EXT_VECTOR_DECLS:
       for (unsigned I = 0, N = Record.size(); I != N; ++I)
-        ExtVectorDecls.push_back(getGlobalDeclID(F, Record[I]));
+        ExtVectorDecls.push_back(getGlobalDeclID(F, LocalDeclID(Record[I])));
       break;
 
     case VTABLE_USES:
@@ -3683,18 +3688,14 @@ llvm::Error ASTReader::ReadASTBlock(ModuleFile &F,
       VTableUses.clear();
 
       for (unsigned Idx = 0, N = Record.size(); Idx != N; /* In loop */) {
-        VTableUses.push_back(getGlobalDeclID(F, Record[Idx++]));
         VTableUses.push_back(
-          ReadSourceLocation(F, Record, Idx).getRawEncoding());
-        VTableUses.push_back(Record[Idx++]);
+            {getGlobalDeclID(F, LocalDeclID(Record[Idx++])),
+             ReadSourceLocation(F, Record, Idx).getRawEncoding(),
+             (bool)Record[Idx++]});
       }
       break;
 
     case PENDING_IMPLICIT_INSTANTIATIONS:
-      if (PendingInstantiations.size() % 2 != 0)
-        return llvm::createStringError(
-            std::errc::illegal_byte_sequence,
-            "Invalid existing PendingInstantiations");
 
       if (Record.size() % 2 != 0)
         return llvm::createStringError(
@@ -3702,9 +3703,9 @@ llvm::Error ASTReader::ReadASTBlock(ModuleFile &F,
             "Invalid PENDING_IMPLICIT_INSTANTIATIONS block");
 
       for (unsigned I = 0, N = Record.size(); I != N; /* in loop */) {
-        PendingInstantiations.push_back(getGlobalDeclID(F, Record[I++]));
         PendingInstantiations.push_back(
-          ReadSourceLocation(F, Record, I).getRawEncoding());
+            {getGlobalDeclID(F, LocalDeclID(Record[I++])),
+             ReadSourceLocation(F, Record, I).getRawEncoding()});
       }
       break;
 
@@ -3713,7 +3714,7 @@ llvm::Error ASTReader::ReadASTBlock(ModuleFile &F,
         return llvm::createStringError(std::errc::illegal_byte_sequence,
                                        "Invalid SEMA_DECL_REFS block");
       for (unsigned I = 0, N = Record.size(); I != N; ++I)
-        SemaDeclRefs.push_back(getGlobalDeclID(F, Record[I]));
+        SemaDeclRefs.push_back(getGlobalDeclID(F, LocalDeclID(Record[I])));
       break;
 
     case PPD_ENTITIES_OFFSETS: {
@@ -3772,7 +3773,7 @@ llvm::Error ASTReader::ReadASTBlock(ModuleFile &F,
             std::errc::illegal_byte_sequence,
             "invalid DECL_UPDATE_OFFSETS block in AST file");
       for (unsigned I = 0, N = Record.size(); I != N; I += 2) {
-        GlobalDeclID ID = getGlobalDeclID(F, Record[I]);
+        GlobalDeclID ID = getGlobalDeclID(F, LocalDeclID(Record[I]));
         DeclUpdateOffsets[ID].push_back(std::make_pair(&F, Record[I + 1]));
 
         // If we've already loaded the decl, perform the updates when we finish
@@ -3790,7 +3791,7 @@ llvm::Error ASTReader::ReadASTBlock(ModuleFile &F,
             "invalid DELAYED_NAMESPACE_LEXICAL_VISIBLE_RECORD block in AST "
             "file");
       for (unsigned I = 0, N = Record.size(); I != N; I += 3) {
-        GlobalDeclID ID = getGlobalDeclID(F, Record[I]);
+        GlobalDeclID ID = getGlobalDeclID(F, LocalDeclID(Record[I]));
 
         uint64_t BaseOffset = F.DeclsBlockStartOffset;
         assert(BaseOffset && "Invalid DeclsBlockStartOffset for module file!");
@@ -3825,7 +3826,8 @@ llvm::Error ASTReader::ReadASTBlock(ModuleFile &F,
       // FIXME: Modules will have trouble with this.
       CUDASpecialDeclRefs.clear();
       for (unsigned I = 0, N = Record.size(); I != N; ++I)
-        CUDASpecialDeclRefs.push_back(getGlobalDeclID(F, Record[I]));
+        CUDASpecialDeclRefs.push_back(
+            getGlobalDeclID(F, LocalDeclID(Record[I])));
       break;
 
     case HEADER_SEARCH_TABLE:
@@ -3866,32 +3868,30 @@ llvm::Error ASTReader::ReadASTBlock(ModuleFile &F,
 
     case TENTATIVE_DEFINITIONS:
       for (unsigned I = 0, N = Record.size(); I != N; ++I)
-        TentativeDefinitions.push_back(getGlobalDeclID(F, Record[I]));
+        TentativeDefinitions.push_back(
+            getGlobalDeclID(F, LocalDeclID(Record[I])));
       break;
 
     case KNOWN_NAMESPACES:
       for (unsigned I = 0, N = Record.size(); I != N; ++I)
-        KnownNamespaces.push_back(getGlobalDeclID(F, Record[I]));
+        KnownNamespaces.push_back(getGlobalDeclID(F, LocalDeclID(Record[I])));
       break;
 
     case UNDEFINED_BUT_USED:
-      if (UndefinedButUsed.size() % 2 != 0)
-        return llvm::createStringError(std::errc::illegal_byte_sequence,
-                                       "Invalid existing UndefinedButUsed");
-
       if (Record.size() % 2 != 0)
         return llvm::createStringError(std::errc::illegal_byte_sequence,
                                        "invalid undefined-but-used record");
       for (unsigned I = 0, N = Record.size(); I != N; /* in loop */) {
-        UndefinedButUsed.push_back(getGlobalDeclID(F, Record[I++]));
         UndefinedButUsed.push_back(
-            ReadSourceLocation(F, Record, I).getRawEncoding());
+            {getGlobalDeclID(F, LocalDeclID(Record[I++])),
+             ReadSourceLocation(F, Record, I).getRawEncoding()});
       }
       break;
 
     case DELETE_EXPRS_TO_ANALYZE:
       for (unsigned I = 0, N = Record.size(); I != N;) {
-        DelayedDeleteExprs.push_back(getGlobalDeclID(F, Record[I++]));
+        DelayedDeleteExprs.push_back(
+            getGlobalDeclID(F, LocalDeclID(Record[I++])).get());
         const uint64_t Count = Record[I++];
         DelayedDeleteExprs.push_back(Count);
         for (uint64_t C = 0; C < Count; ++C) {
@@ -3976,7 +3976,7 @@ llvm::Error ASTReader::ReadASTBlock(ModuleFile &F,
     case UNUSED_LOCAL_TYPEDEF_NAME_CANDIDATES:
       for (unsigned I = 0, N = Record.size(); I != N; ++I)
         UnusedLocalTypedefNameCandidates.push_back(
-            getGlobalDeclID(F, Record[I]));
+            getGlobalDeclID(F, LocalDeclID(Record[I])));
       break;
 
     case CUDA_PRAGMA_FORCE_HOST_DEVICE_DEPTH:
@@ -4032,7 +4032,8 @@ llvm::Error ASTReader::ReadASTBlock(ModuleFile &F,
 
     case DECLS_TO_CHECK_FOR_DEFERRED_DIAGS:
       for (unsigned I = 0, N = Record.size(); I != N; ++I)
-        DeclsToCheckForDeferredDiags.insert(getGlobalDeclID(F, Record[I]));
+        DeclsToCheckForDeferredDiags.insert(
+            getGlobalDeclID(F, LocalDeclID(Record[I])));
       break;
     }
   }
@@ -4655,9 +4656,8 @@ ASTReader::ASTReadResult ASTReader::ReadAST(StringRef FileName, ModuleKind Type,
   // that we load any additional categories.
   if (ContextObj) {
     for (unsigned I = 0, N = ObjCClassesLoaded.size(); I != N; ++I) {
-      loadObjCCategories(ObjCClassesLoaded[I]->getGlobalID(),
-                         ObjCClassesLoaded[I],
-                         PreviousGeneration);
+      loadObjCCategories(GlobalDeclID(ObjCClassesLoaded[I]->getGlobalID()),
+                         ObjCClassesLoaded[I], PreviousGeneration);
     }
   }
 
@@ -5101,8 +5101,9 @@ void ASTReader::InitializeContext() {
 
   // If there's a listener, notify them that we "read" the translation unit.
   if (DeserializationListener)
-    DeserializationListener->DeclRead(PREDEF_DECL_TRANSLATION_UNIT_ID,
-                                      Context.getTranslationUnitDecl());
+    DeserializationListener->DeclRead(
+        GlobalDeclID(PREDEF_DECL_TRANSLATION_UNIT_ID),
+        Context.getTranslationUnitDecl());
 
   // FIXME: Find a better way to deal with collisions between these
   // built-in types. Right now, we just ignore the problem.
@@ -6010,9 +6011,9 @@ llvm::Error ASTReader::ReadSubmoduleBlock(ModuleFile &F,
     case SUBMODULE_INITIALIZERS: {
       if (!ContextObj)
         break;
-      SmallVector<uint32_t, 16> Inits;
+      SmallVector<GlobalDeclID, 16> Inits;
       for (auto &ID : Record)
-        Inits.push_back(getGlobalDeclID(F, ID));
+        Inits.push_back(getGlobalDeclID(F, LocalDeclID(ID)));
       ContextObj->addLazyModuleInitializers(CurrentModule, Inits);
       break;
     }
@@ -7384,11 +7385,11 @@ QualType ASTReader::GetType(TypeID ID) {
     case PREDEF_TYPE_INCOMPLETE_MATRIX_IDX:
       T = Context.IncompleteMatrixIdxTy;
       break;
-    case PREDEF_TYPE_OMP_ARRAY_SECTION:
-      T = Context.OMPArraySectionTy;
+    case PREDEF_TYPE_ARRAY_SECTION:
+      T = Context.ArraySectionTy;
       break;
     case PREDEF_TYPE_OMP_ARRAY_SHAPING:
-      T = Context.OMPArraySectionTy;
+      T = Context.OMPArrayShapingTy;
       break;
     case PREDEF_TYPE_OMP_ITERATOR:
       T = Context.OMPIteratorTy;
@@ -7517,9 +7518,7 @@ ASTRecordReader::readASTTemplateArgumentListInfo() {
   return ASTTemplateArgumentListInfo::Create(getContext(), Result);
 }
 
-Decl *ASTReader::GetExternalDecl(uint32_t ID) {
-  return GetDecl(ID);
-}
+Decl *ASTReader::GetExternalDecl(GlobalDeclID ID) { return GetDecl(ID); }
 
 void ASTReader::CompleteRedeclChain(const Decl *D) {
   if (NumCurrentElementsDeserializing) {
@@ -7652,44 +7651,45 @@ CXXBaseSpecifier *ASTReader::GetExternalCXXBaseSpecifiers(uint64_t Offset) {
   return Bases;
 }
 
-serialization::DeclID
-ASTReader::getGlobalDeclID(ModuleFile &F, LocalDeclID LocalID) const {
-  if (LocalID < NUM_PREDEF_DECL_IDS)
-    return LocalID;
+GlobalDeclID ASTReader::getGlobalDeclID(ModuleFile &F,
+                                        LocalDeclID LocalID) const {
+  DeclID ID = LocalID.get();
+  if (ID < NUM_PREDEF_DECL_IDS)
+    return GlobalDeclID(ID);
 
   if (!F.ModuleOffsetMap.empty())
     ReadModuleOffsetMap(F);
 
-  ContinuousRangeMap<uint32_t, int, 2>::iterator I
-    = F.DeclRemap.find(LocalID - NUM_PREDEF_DECL_IDS);
+  ContinuousRangeMap<DeclID, int, 2>::iterator I =
+      F.DeclRemap.find(ID - NUM_PREDEF_DECL_IDS);
   assert(I != F.DeclRemap.end() && "Invalid index into decl index remap");
 
-  return LocalID + I->second;
+  return GlobalDeclID(ID + I->second);
 }
 
-bool ASTReader::isDeclIDFromModule(serialization::GlobalDeclID ID,
-                                   ModuleFile &M) const {
+bool ASTReader::isDeclIDFromModule(GlobalDeclID ID, ModuleFile &M) const {
   // Predefined decls aren't from any module.
-  if (ID < NUM_PREDEF_DECL_IDS)
+  if (ID.get() < NUM_PREDEF_DECL_IDS)
     return false;
 
-  return ID - NUM_PREDEF_DECL_IDS >= M.BaseDeclID &&
-         ID - NUM_PREDEF_DECL_IDS < M.BaseDeclID + M.LocalNumDecls;
+  return ID.get() - NUM_PREDEF_DECL_IDS >= M.BaseDeclID &&
+         ID.get() - NUM_PREDEF_DECL_IDS < M.BaseDeclID + M.LocalNumDecls;
 }
 
 ModuleFile *ASTReader::getOwningModuleFile(const Decl *D) {
   if (!D->isFromASTFile())
     return nullptr;
-  GlobalDeclMapType::const_iterator I = GlobalDeclMap.find(D->getGlobalID());
+  GlobalDeclMapType::const_iterator I =
+      GlobalDeclMap.find(GlobalDeclID(D->getGlobalID()));
   assert(I != GlobalDeclMap.end() && "Corrupted global declaration map");
   return I->second;
 }
 
 SourceLocation ASTReader::getSourceLocationForDeclID(GlobalDeclID ID) {
-  if (ID < NUM_PREDEF_DECL_IDS)
+  if (ID.get() < NUM_PREDEF_DECL_IDS)
     return SourceLocation();
 
-  unsigned Index = ID - NUM_PREDEF_DECL_IDS;
+  unsigned Index = ID.get() - NUM_PREDEF_DECL_IDS;
 
   if (Index > DeclsLoaded.size()) {
     Error("declaration ID out-of-range for AST file");
@@ -7763,9 +7763,9 @@ static Decl *getPredefinedDecl(ASTContext &Context, PredefinedDeclIDs ID) {
   llvm_unreachable("PredefinedDeclIDs unknown enum value");
 }
 
-Decl *ASTReader::GetExistingDecl(DeclID ID) {
+Decl *ASTReader::GetExistingDecl(GlobalDeclID ID) {
   assert(ContextObj && "reading decl with no AST context");
-  if (ID < NUM_PREDEF_DECL_IDS) {
+  if (ID.get() < NUM_PREDEF_DECL_IDS) {
     Decl *D = getPredefinedDecl(*ContextObj, (PredefinedDeclIDs)ID);
     if (D) {
       // Track that we have merged the declaration with ID \p ID into the
@@ -7777,7 +7777,7 @@ Decl *ASTReader::GetExistingDecl(DeclID ID) {
     return D;
   }
 
-  unsigned Index = ID - NUM_PREDEF_DECL_IDS;
+  unsigned Index = ID.get() - NUM_PREDEF_DECL_IDS;
 
   if (Index >= DeclsLoaded.size()) {
     assert(0 && "declaration ID out-of-range for AST file");
@@ -7788,11 +7788,11 @@ Decl *ASTReader::GetExistingDecl(DeclID ID) {
   return DeclsLoaded[Index];
 }
 
-Decl *ASTReader::GetDecl(DeclID ID) {
-  if (ID < NUM_PREDEF_DECL_IDS)
+Decl *ASTReader::GetDecl(GlobalDeclID ID) {
+  if (ID.get() < NUM_PREDEF_DECL_IDS)
     return GetExistingDecl(ID);
 
-  unsigned Index = ID - NUM_PREDEF_DECL_IDS;
+  unsigned Index = ID.get() - NUM_PREDEF_DECL_IDS;
 
   if (Index >= DeclsLoaded.size()) {
     assert(0 && "declaration ID out-of-range for AST file");
@@ -7809,32 +7809,32 @@ Decl *ASTReader::GetDecl(DeclID ID) {
   return DeclsLoaded[Index];
 }
 
-DeclID ASTReader::mapGlobalIDToModuleFileGlobalID(ModuleFile &M,
-                                                  DeclID GlobalID) {
-  if (GlobalID < NUM_PREDEF_DECL_IDS)
-    return GlobalID;
+LocalDeclID ASTReader::mapGlobalIDToModuleFileGlobalID(ModuleFile &M,
+                                                       GlobalDeclID GlobalID) {
+  DeclID ID = GlobalID.get();
+  if (ID < NUM_PREDEF_DECL_IDS)
+    return LocalDeclID(ID);
 
   GlobalDeclMapType::const_iterator I = GlobalDeclMap.find(GlobalID);
   assert(I != GlobalDeclMap.end() && "Corrupted global declaration map");
   ModuleFile *Owner = I->second;
 
-  llvm::DenseMap<ModuleFile *, serialization::DeclID>::iterator Pos
-    = M.GlobalToLocalDeclIDs.find(Owner);
+  llvm::DenseMap<ModuleFile *, DeclID>::iterator Pos =
+      M.GlobalToLocalDeclIDs.find(Owner);
   if (Pos == M.GlobalToLocalDeclIDs.end())
-    return 0;
+    return LocalDeclID();
 
-  return GlobalID - Owner->BaseDeclID + Pos->second;
+  return LocalDeclID(ID - Owner->BaseDeclID + Pos->second);
 }
 
-serialization::DeclID ASTReader::ReadDeclID(ModuleFile &F,
-                                            const RecordData &Record,
-                                            unsigned &Idx) {
+GlobalDeclID ASTReader::ReadDeclID(ModuleFile &F, const RecordData &Record,
+                                   unsigned &Idx) {
   if (Idx >= Record.size()) {
     Error("Corrupted AST file");
-    return 0;
+    return GlobalDeclID(0);
   }
 
-  return getGlobalDeclID(F, Record[Idx++]);
+  return getGlobalDeclID(F, LocalDeclID(Record[Idx++]));
 }
 
 /// Resolve the offset of a statement into a statement.
@@ -7870,7 +7870,7 @@ void ASTReader::FindExternalLexicalDecls(
       if (!IsKindWeWant(K))
         continue;
 
-      auto ID = (serialization::DeclID)+LexicalDecls[I + 1];
+      auto ID = (DeclID) + LexicalDecls[I + 1];
 
       // Don't add predefined declarations to the lexical context more
       // than once.
@@ -7881,7 +7881,7 @@ void ASTReader::FindExternalLexicalDecls(
         PredefsVisited[ID] = true;
       }
 
-      if (Decl *D = GetLocalDecl(*M, ID)) {
+      if (Decl *D = GetLocalDecl(*M, LocalDeclID(ID))) {
         assert(D->getKind() == K && "wrong kind for lexical decl");
         if (!DC->isDeclInLexicalTraversal(D))
           Decls.push_back(D);
@@ -7952,7 +7952,7 @@ void ASTReader::FindFileRegionDecls(FileID File,
   SourceLocation EndLoc = BeginLoc.getLocWithOffset(Length);
 
   DeclIDComp DIDComp(*this, *DInfo.Mod);
-  ArrayRef<serialization::LocalDeclID>::iterator BeginIt =
+  ArrayRef<LocalDeclID>::iterator BeginIt =
       llvm::lower_bound(DInfo.Decls, BeginLoc, DIDComp);
   if (BeginIt != DInfo.Decls.begin())
     --BeginIt;
@@ -7965,13 +7965,12 @@ void ASTReader::FindFileRegionDecls(FileID File,
              ->isTopLevelDeclInObjCContainer())
     --BeginIt;
 
-  ArrayRef<serialization::LocalDeclID>::iterator EndIt =
+  ArrayRef<LocalDeclID>::iterator EndIt =
       llvm::upper_bound(DInfo.Decls, EndLoc, DIDComp);
   if (EndIt != DInfo.Decls.end())
     ++EndIt;
 
-  for (ArrayRef<serialization::LocalDeclID>::iterator
-         DIt = BeginIt; DIt != EndIt; ++DIt)
+  for (ArrayRef<LocalDeclID>::iterator DIt = BeginIt; DIt != EndIt; ++DIt)
     Decls.push_back(GetDecl(getGlobalDeclID(*DInfo.Mod, *DIt)));
 }
 
@@ -7992,7 +7991,8 @@ ASTReader::FindExternalVisibleDeclsByName(const DeclContext *DC,
   // Load the list of declarations.
   SmallVector<NamedDecl *, 64> Decls;
   llvm::SmallPtrSet<NamedDecl *, 8> Found;
-  for (DeclID ID : It->second.Table.find(Name)) {
+
+  for (GlobalDeclID ID : It->second.Table.find(Name)) {
     NamedDecl *ND = cast<NamedDecl>(GetDecl(ID));
     if (ND->getDeclName() == Name && Found.insert(ND).second)
       Decls.push_back(ND);
@@ -8013,7 +8013,7 @@ void ASTReader::completeVisibleDeclsMap(const DeclContext *DC) {
 
   DeclsMap Decls;
 
-  for (DeclID ID : It->second.Table.findAll()) {
+  for (GlobalDeclID ID : It->second.Table.findAll()) {
     NamedDecl *ND = cast<NamedDecl>(GetDecl(ID));
     Decls[ND->getDeclName()].push_back(ND);
   }
@@ -8164,8 +8164,12 @@ dumpModuleIDMap(StringRef Name,
   llvm::errs() << Name << ":\n";
   for (typename MapType::const_iterator I = Map.begin(), IEnd = Map.end();
        I != IEnd; ++I) {
-    llvm::errs() << "  " << I->first << " -> " << I->second->FileName
-      << "\n";
+    uint64_t ID = 0;
+    if constexpr (std::is_integral_v<Key>)
+      ID = I->first;
+    else /*GlobalDeclID*/
+      ID = I->first.get();
+    llvm::errs() << "  " << ID << " -> " << I->second->FileName << "\n";
   }
 }
 
@@ -8211,7 +8215,7 @@ void ASTReader::InitializeSema(Sema &S) {
 
   // Makes sure any declarations that were deserialized "too early"
   // still get added to the identifier's declaration chains.
-  for (uint64_t ID : PreloadedDeclIDs) {
+  for (GlobalDeclID ID : PreloadedDeclIDs) {
     NamedDecl *D = cast<NamedDecl>(GetDecl(ID));
     pushExternalDeclIntoScope(D, D->getDeclName());
   }
@@ -8240,11 +8244,11 @@ void ASTReader::UpdateSema() {
     assert(SemaDeclRefs.size() % 3 == 0);
     for (unsigned I = 0; I != SemaDeclRefs.size(); I += 3) {
       if (!SemaObj->StdNamespace)
-        SemaObj->StdNamespace = SemaDeclRefs[I];
+        SemaObj->StdNamespace = SemaDeclRefs[I].get();
       if (!SemaObj->StdBadAlloc)
-        SemaObj->StdBadAlloc = SemaDeclRefs[I+1];
+        SemaObj->StdBadAlloc = SemaDeclRefs[I + 1].get();
       if (!SemaObj->StdAlignValT)
-        SemaObj->StdAlignValT = SemaDeclRefs[I+2];
+        SemaObj->StdAlignValT = SemaDeclRefs[I + 2].get();
     }
     SemaDeclRefs.clear();
   }
@@ -8617,18 +8621,20 @@ void ASTReader::ReadKnownNamespaces(
 void ASTReader::ReadUndefinedButUsed(
     llvm::MapVector<NamedDecl *, SourceLocation> &Undefined) {
   for (unsigned Idx = 0, N = UndefinedButUsed.size(); Idx != N;) {
-    NamedDecl *D = cast<NamedDecl>(GetDecl(UndefinedButUsed[Idx++]));
-    SourceLocation Loc =
-        SourceLocation::getFromRawEncoding(UndefinedButUsed[Idx++]);
+    UndefinedButUsedDecl &U = UndefinedButUsed[Idx++];
+    NamedDecl *D = cast<NamedDecl>(GetDecl(U.ID));
+    SourceLocation Loc = SourceLocation::getFromRawEncoding(U.RawLoc);
     Undefined.insert(std::make_pair(D, Loc));
   }
+  UndefinedButUsed.clear();
 }
 
 void ASTReader::ReadMismatchingDeleteExpressions(llvm::MapVector<
     FieldDecl *, llvm::SmallVector<std::pair<SourceLocation, bool>, 4>> &
                                                      Exprs) {
   for (unsigned Idx = 0, N = DelayedDeleteExprs.size(); Idx != N;) {
-    FieldDecl *FD = cast<FieldDecl>(GetDecl(DelayedDeleteExprs[Idx++]));
+    FieldDecl *FD =
+        cast<FieldDecl>(GetDecl(GlobalDeclID(DelayedDeleteExprs[Idx++])));
     uint64_t Count = DelayedDeleteExprs[Idx++];
     for (uint64_t C = 0; C < Count; ++C) {
       SourceLocation DeleteLoc =
@@ -8742,9 +8748,10 @@ void ASTReader::ReadWeakUndeclaredIdentifiers(
 void ASTReader::ReadUsedVTables(SmallVectorImpl<ExternalVTableUse> &VTables) {
   for (unsigned Idx = 0, N = VTableUses.size(); Idx < N; /* In loop */) {
     ExternalVTableUse VT;
-    VT.Record = dyn_cast_or_null<CXXRecordDecl>(GetDecl(VTableUses[Idx++]));
-    VT.Location = SourceLocation::getFromRawEncoding(VTableUses[Idx++]);
-    VT.DefinitionRequired = VTableUses[Idx++];
+    VTableUse &TableInfo = VTableUses[Idx++];
+    VT.Record = dyn_cast_or_null<CXXRecordDecl>(GetDecl(TableInfo.ID));
+    VT.Location = SourceLocation::getFromRawEncoding(TableInfo.RawLoc);
+    VT.DefinitionRequired = TableInfo.Used;
     VTables.push_back(VT);
   }
 
@@ -8754,9 +8761,9 @@ void ASTReader::ReadUsedVTables(SmallVectorImpl<ExternalVTableUse> &VTables) {
 void ASTReader::ReadPendingInstantiations(
        SmallVectorImpl<std::pair<ValueDecl *, SourceLocation>> &Pending) {
   for (unsigned Idx = 0, N = PendingInstantiations.size(); Idx < N;) {
-    ValueDecl *D = cast<ValueDecl>(GetDecl(PendingInstantiations[Idx++]));
-    SourceLocation Loc
-      = SourceLocation::getFromRawEncoding(PendingInstantiations[Idx++]);
+    PendingInstantiation &Inst = PendingInstantiations[Idx++];
+    ValueDecl *D = cast<ValueDecl>(GetDecl(Inst.ID));
+    SourceLocation Loc = SourceLocation::getFromRawEncoding(Inst.RawLoc);
 
     Pending.push_back(std::make_pair(D, Loc));
   }
@@ -8771,11 +8778,11 @@ void ASTReader::ReadLateParsedTemplates(
     RecordDataImpl &LateParsed = LPT.second;
     for (unsigned Idx = 0, N = LateParsed.size(); Idx < N;
          /* In loop */) {
-      FunctionDecl *FD =
-          cast<FunctionDecl>(GetLocalDecl(*FMod, LateParsed[Idx++]));
+      FunctionDecl *FD = cast<FunctionDecl>(
+          GetLocalDecl(*FMod, LocalDeclID(LateParsed[Idx++])));
 
       auto LT = std::make_unique<LateParsedTemplate>();
-      LT->D = GetLocalDecl(*FMod, LateParsed[Idx++]);
+      LT->D = GetLocalDecl(*FMod, LocalDeclID(LateParsed[Idx++]));
       LT->FPO = FPOptions::getFromOpaqueInt(LateParsed[Idx++]);
 
       ModuleFile *F = getOwningModuleFile(LT->D);
@@ -8833,10 +8840,9 @@ void ASTReader::SetIdentifierInfo(IdentifierID ID, IdentifierInfo *II) {
 /// \param Decls if non-null, this vector will be populated with the set of
 /// deserialized declarations. These declarations will not be pushed into
 /// scope.
-void
-ASTReader::SetGloballyVisibleDecls(IdentifierInfo *II,
-                              const SmallVectorImpl<uint32_t> &DeclIDs,
-                                   SmallVectorImpl<Decl *> *Decls) {
+void ASTReader::SetGloballyVisibleDecls(
+    IdentifierInfo *II, const SmallVectorImpl<GlobalDeclID> &DeclIDs,
+    SmallVectorImpl<Decl *> *Decls) {
   if (NumCurrentElementsDeserializing && !Decls) {
     PendingIdentifierInfos[II].append(DeclIDs.begin(), DeclIDs.end());
     return;
@@ -9184,9 +9190,9 @@ void ASTRecordReader::readUnresolvedSet(LazyASTUnresolvedSet &Set) {
   unsigned NumDecls = readInt();
   Set.reserve(getContext(), NumDecls);
   while (NumDecls--) {
-    DeclID ID = readDeclID();
+    GlobalDeclID ID = readDeclID();
     AccessSpecifier AS = (AccessSpecifier) readInt();
-    Set.addLazyDecl(getContext(), ID, AS);
+    Set.addLazyDecl(getContext(), ID.get(), AS);
   }
 }
 
@@ -9560,7 +9566,7 @@ void ASTReader::finishPendingActions() {
 
     while (!PendingIdentifierInfos.empty()) {
       IdentifierInfo *II = PendingIdentifierInfos.back().first;
-      SmallVector<uint32_t, 4> DeclIDs =
+      SmallVector<GlobalDeclID, 4> DeclIDs =
           std::move(PendingIdentifierInfos.back().second);
       PendingIdentifierInfos.pop_back();
 
diff --git a/clang/lib/Serialization/ASTReaderDecl.cpp b/clang/lib/Serialization/ASTReaderDecl.cpp
index 74d40f7da34c..744f11de88c2 100644
--- a/clang/lib/Serialization/ASTReaderDecl.cpp
+++ b/clang/lib/Serialization/ASTReaderDecl.cpp
@@ -84,14 +84,14 @@ namespace clang {
     ASTReader &Reader;
     ASTRecordReader &Record;
     ASTReader::RecordLocation Loc;
-    const DeclID ThisDeclID;
+    const GlobalDeclID ThisDeclID;
     const SourceLocation ThisDeclLoc;
 
     using RecordData = ASTReader::RecordData;
 
     TypeID DeferredTypeID = 0;
     unsigned AnonymousDeclNumber = 0;
-    GlobalDeclID NamedDeclForTagDecl = 0;
+    GlobalDeclID NamedDeclForTagDecl = GlobalDeclID();
     IdentifierInfo *TypedefNameForLinkage = nullptr;
 
     ///A flag to carry the information for a decl from the entity is
@@ -124,15 +124,13 @@ namespace clang {
       return Record.readTypeSourceInfo();
     }
 
-    serialization::DeclID readDeclID() {
-      return Record.readDeclID();
-    }
+    GlobalDeclID readDeclID() { return Record.readDeclID(); }
 
     std::string readString() {
       return Record.readString();
     }
 
-    void readDeclIDList(SmallVectorImpl<DeclID> &IDs) {
+    void readDeclIDList(SmallVectorImpl<GlobalDeclID> &IDs) {
       for (unsigned I = 0, Size = Record.readInt(); I != Size; ++I)
         IDs.push_back(readDeclID());
     }
@@ -258,14 +256,14 @@ namespace clang {
 
   public:
     ASTDeclReader(ASTReader &Reader, ASTRecordReader &Record,
-                  ASTReader::RecordLocation Loc,
-                  DeclID thisDeclID, SourceLocation ThisDeclLoc)
+                  ASTReader::RecordLocation Loc, GlobalDeclID thisDeclID,
+                  SourceLocation ThisDeclLoc)
         : Reader(Reader), Record(Record), Loc(Loc), ThisDeclID(thisDeclID),
           ThisDeclLoc(ThisDeclLoc) {}
 
-    template <typename T> static
-    void AddLazySpecializations(T *D,
-                                SmallVectorImpl<serialization::DeclID>& IDs) {
+    template <typename T>
+    static void AddLazySpecializations(T *D,
+                                       SmallVectorImpl<GlobalDeclID> &IDs) {
       if (IDs.empty())
         return;
 
@@ -275,13 +273,14 @@ namespace clang {
       auto *&LazySpecializations = D->getCommonPtr()->LazySpecializations;
 
       if (auto &Old = LazySpecializations) {
-        IDs.insert(IDs.end(), Old + 1, Old + 1 + Old[0]);
+        IDs.insert(IDs.end(), Old + 1, Old + 1 + Old[0].get());
         llvm::sort(IDs);
         IDs.erase(std::unique(IDs.begin(), IDs.end()), IDs.end());
       }
 
-      auto *Result = new (C) serialization::DeclID[1 + IDs.size()];
-      *Result = IDs.size();
+      auto *Result = new (C) GlobalDeclID[1 + IDs.size()];
+      *Result = GlobalDeclID(IDs.size());
+
       std::copy(IDs.begin(), IDs.end(), Result + 1);
 
       LazySpecializations = Result;
@@ -315,7 +314,7 @@ namespace clang {
     void ReadFunctionDefinition(FunctionDecl *FD);
     void Visit(Decl *D);
 
-    void UpdateDecl(Decl *D, SmallVectorImpl<serialization::DeclID> &);
+    void UpdateDecl(Decl *D, SmallVectorImpl<GlobalDeclID> &);
 
     static void setNextObjCCategory(ObjCCategoryDecl *Cat,
                                     ObjCCategoryDecl *Next) {
@@ -557,7 +556,7 @@ void ASTDeclReader::Visit(Decl *D) {
 
     // If this is a tag declaration with a typedef name for linkage, it's safe
     // to load that typedef now.
-    if (NamedDeclForTagDecl)
+    if (NamedDeclForTagDecl.isValid())
       cast<TagDecl>(D)->TypedefNameDeclOrQualifier =
           cast<TypedefNameDecl>(Reader.GetDecl(NamedDeclForTagDecl));
   } else if (auto *ID = dyn_cast<ObjCInterfaceDecl>(D)) {
@@ -601,8 +600,8 @@ void ASTDeclReader::VisitDecl(Decl *D) {
     // placeholder.
     GlobalDeclID SemaDCIDForTemplateParmDecl = readDeclID();
     GlobalDeclID LexicalDCIDForTemplateParmDecl =
-        HasStandaloneLexicalDC ? readDeclID() : 0;
-    if (!LexicalDCIDForTemplateParmDecl)
+        HasStandaloneLexicalDC ? readDeclID() : GlobalDeclID();
+    if (LexicalDCIDForTemplateParmDecl.isInvalid())
       LexicalDCIDForTemplateParmDecl = SemaDCIDForTemplateParmDecl;
     Reader.addPendingDeclContextInfo(D,
                                      SemaDCIDForTemplateParmDecl,
@@ -1848,7 +1847,7 @@ void ASTDeclReader::VisitNamespaceDecl(NamespaceDecl *D) {
   // this namespace; loading it might load a later declaration of the
   // same namespace, and we have an invariant that older declarations
   // get merged before newer ones try to merge.
-  GlobalDeclID AnonNamespace = 0;
+  GlobalDeclID AnonNamespace;
   if (Redecl.getFirstID() == ThisDeclID) {
     AnonNamespace = readDeclID();
   } else {
@@ -1859,7 +1858,7 @@ void ASTDeclReader::VisitNamespaceDecl(NamespaceDecl *D) {
 
   mergeRedeclarable(D, Redecl);
 
-  if (AnonNamespace) {
+  if (AnonNamespace.isValid()) {
     // Each module has its own anonymous namespace, which is disjoint from
     // any other module's anonymous namespaces, so don't attach the anonymous
     // namespace at all.
@@ -2019,7 +2018,7 @@ void ASTDeclReader::ReadCXXDefinitionData(
     if (Data.NumVBases)
       Data.VBases = ReadGlobalOffset();
 
-    Data.FirstFriend = readDeclID();
+    Data.FirstFriend = readDeclID().get();
   } else {
     using Capture = LambdaCapture;
 
@@ -2278,12 +2277,12 @@ ASTDeclReader::VisitCXXRecordDeclImpl(CXXRecordDecl *D) {
   // Lazily load the key function to avoid deserializing every method so we can
   // compute it.
   if (WasDefinition) {
-    DeclID KeyFn = readDeclID();
-    if (KeyFn && D->isCompleteDefinition())
+    GlobalDeclID KeyFn = readDeclID();
+    if (KeyFn.get() && D->isCompleteDefinition())
       // FIXME: This is wrong for the ARM ABI, where some other module may have
       // made this function no longer be a key function. We need an update
       // record or similar for that case.
-      C.KeyFunctions[D] = KeyFn;
+      C.KeyFunctions[D] = KeyFn.get();
   }
 
   return Redecl;
@@ -2372,7 +2371,7 @@ void ASTDeclReader::VisitFriendDecl(FriendDecl *D) {
   for (unsigned i = 0; i != D->NumTPLists; ++i)
     D->getTrailingObjects<TemplateParameterList *>()[i] =
         Record.readTemplateParameterList();
-  D->NextFriend = readDeclID();
+  D->NextFriend = readDeclID().get();
   D->UnsupportedFriend = (Record.readInt() != 0);
   D->FriendLoc = readSourceLocation();
 }
@@ -2457,7 +2456,7 @@ void ASTDeclReader::VisitClassTemplateDecl(ClassTemplateDecl *D) {
   if (ThisDeclID == Redecl.getFirstID()) {
     // This ClassTemplateDecl owns a CommonPtr; read it to keep track of all of
     // the specializations.
-    SmallVector<serialization::DeclID, 32> SpecIDs;
+    SmallVector<GlobalDeclID, 32> SpecIDs;
     readDeclIDList(SpecIDs);
     ASTDeclReader::AddLazySpecializations(D, SpecIDs);
   }
@@ -2485,7 +2484,7 @@ void ASTDeclReader::VisitVarTemplateDecl(VarTemplateDecl *D) {
   if (ThisDeclID == Redecl.getFirstID()) {
     // This VarTemplateDecl owns a CommonPtr; read it to keep track of all of
     // the specializations.
-    SmallVector<serialization::DeclID, 32> SpecIDs;
+    SmallVector<GlobalDeclID, 32> SpecIDs;
     readDeclIDList(SpecIDs);
     ASTDeclReader::AddLazySpecializations(D, SpecIDs);
   }
@@ -2587,7 +2586,7 @@ void ASTDeclReader::VisitFunctionTemplateDecl(FunctionTemplateDecl *D) {
 
   if (ThisDeclID == Redecl.getFirstID()) {
     // This FunctionTemplateDecl owns a CommonPtr; read it.
-    SmallVector<serialization::DeclID, 32> SpecIDs;
+    SmallVector<GlobalDeclID, 32> SpecIDs;
     readDeclIDList(SpecIDs);
     ASTDeclReader::AddLazySpecializations(D, SpecIDs);
   }
@@ -2783,7 +2782,7 @@ ASTDeclReader::VisitDeclContext(DeclContext *DC) {
 template <typename T>
 ASTDeclReader::RedeclarableResult
 ASTDeclReader::VisitRedeclarable(Redeclarable<T> *D) {
-  DeclID FirstDeclID = readDeclID();
+  GlobalDeclID FirstDeclID = readDeclID();
   Decl *MergeWith = nullptr;
 
   bool IsKeyDecl = ThisDeclID == FirstDeclID;
@@ -2791,9 +2790,9 @@ ASTDeclReader::VisitRedeclarable(Redeclarable<T> *D) {
 
   uint64_t RedeclOffset = 0;
 
-  // 0 indicates that this declaration was the only declaration of its entity,
-  // and is used for space optimization.
-  if (FirstDeclID == 0) {
+  // invalid FirstDeclID  indicates that this declaration was the only
+  // declaration of its entity, and is used for space optimization.
+  if (FirstDeclID.isInvalid()) {
     FirstDeclID = ThisDeclID;
     IsKeyDecl = true;
     IsFirstLocalDecl = true;
@@ -2922,9 +2921,9 @@ void ASTDeclReader::mergeTemplatePattern(RedeclarableTemplateDecl *D,
                                          bool IsKeyDecl) {
   auto *DPattern = D->getTemplatedDecl();
   auto *ExistingPattern = Existing->getTemplatedDecl();
-  RedeclarableResult Result(/*MergeWith*/ ExistingPattern,
-                            DPattern->getCanonicalDecl()->getGlobalID(),
-                            IsKeyDecl);
+  RedeclarableResult Result(
+      /*MergeWith*/ ExistingPattern,
+      GlobalDeclID(DPattern->getCanonicalDecl()->getGlobalID()), IsKeyDecl);
 
   if (auto *DClass = dyn_cast<CXXRecordDecl>(DPattern)) {
     // Merge with any existing definition.
@@ -3079,14 +3078,14 @@ void ASTDeclReader::VisitOMPDeclareReductionDecl(OMPDeclareReductionDecl *D) {
   Expr *Init = Record.readExpr();
   auto IK = static_cast<OMPDeclareReductionInitKind>(Record.readInt());
   D->setInitializer(Init, IK);
-  D->PrevDeclInScope = readDeclID();
+  D->PrevDeclInScope = readDeclID().get();
 }
 
 void ASTDeclReader::VisitOMPDeclareMapperDecl(OMPDeclareMapperDecl *D) {
   Record.readOMPChildren(D->Data);
   VisitValueDecl(D);
   D->VarName = Record.readDeclarationName();
-  D->PrevDeclInScope = readDeclID();
+  D->PrevDeclInScope = readDeclID().get();
 }
 
 void ASTDeclReader::VisitOMPCapturedExprDecl(OMPCapturedExprDecl *D) {
@@ -3140,7 +3139,7 @@ public:
 
   OMPTraitInfo *readOMPTraitInfo() { return Reader.readOMPTraitInfo(); }
 
-  template <typename T> T *GetLocalDeclAs(uint32_t LocalID) {
+  template <typename T> T *GetLocalDeclAs(LocalDeclID LocalID) {
     return Reader.GetLocalDeclAs<T>(LocalID);
   }
 };
@@ -3243,13 +3242,13 @@ bool ASTReader::isConsumerInterestedIn(Decl *D) {
 }
 
 /// Get the correct cursor and offset for loading a declaration.
-ASTReader::RecordLocation
-ASTReader::DeclCursorForID(DeclID ID, SourceLocation &Loc) {
+ASTReader::RecordLocation ASTReader::DeclCursorForID(GlobalDeclID ID,
+                                                     SourceLocation &Loc) {
   GlobalDeclMapType::iterator I = GlobalDeclMap.find(ID);
   assert(I != GlobalDeclMap.end() && "Corrupted global declaration map");
   ModuleFile *M = I->second;
   const DeclOffset &DOffs =
-      M->DeclOffsets[ID - M->BaseDeclID - NUM_PREDEF_DECL_IDS];
+      M->DeclOffsets[ID.get() - M->BaseDeclID - NUM_PREDEF_DECL_IDS];
   Loc = TranslateSourceLocation(*M, DOffs.getLocation());
   return RecordLocation(M, DOffs.getBitOffset(M->DeclsBlockStartOffset));
 }
@@ -3792,8 +3791,8 @@ void ASTReader::markIncompleteDeclChain(Decl *D) {
 }
 
 /// Read the declaration at the given offset from the AST file.
-Decl *ASTReader::ReadDeclRecord(DeclID ID) {
-  unsigned Index = ID - NUM_PREDEF_DECL_IDS;
+Decl *ASTReader::ReadDeclRecord(GlobalDeclID ID) {
+  unsigned Index = ID.get() - NUM_PREDEF_DECL_IDS;
   SourceLocation DeclLoc;
   RecordLocation Loc = DeclCursorForID(ID, DeclLoc);
   llvm::BitstreamCursor &DeclsCursor = Loc.F->DeclsCursor;
@@ -3827,6 +3826,7 @@ Decl *ASTReader::ReadDeclRecord(DeclID ID) {
     llvm::report_fatal_error(
         Twine("ASTReader::readDeclRecord failed reading decl code: ") +
         toString(MaybeDeclCode.takeError()));
+
   switch ((DeclCode)MaybeDeclCode.get()) {
   case DECL_CONTEXT_LEXICAL:
   case DECL_CONTEXT_VISIBLE:
@@ -3953,9 +3953,8 @@ Decl *ASTReader::ReadDeclRecord(DeclID ID) {
   }
   case DECL_EXPANDED_NON_TYPE_TEMPLATE_PARM_PACK: {
     bool HasTypeConstraint = Record.readInt();
-    D = NonTypeTemplateParmDecl::CreateDeserialized(Context, ID,
-                                                    Record.readInt(),
-                                                    HasTypeConstraint);
+    D = NonTypeTemplateParmDecl::CreateDeserialized(
+        Context, ID, Record.readInt(), HasTypeConstraint);
     break;
   }
   case DECL_TEMPLATE_TEMPLATE_PARM:
@@ -4202,12 +4201,12 @@ void ASTReader::loadDeclUpdateRecords(PendingUpdateRecord &Record) {
   // The declaration may have been modified by files later in the chain.
   // If this is the case, read the record containing the updates from each file
   // and pass it to ASTDeclReader to make the modifications.
-  serialization::GlobalDeclID ID = Record.ID;
+  GlobalDeclID ID = Record.ID;
   Decl *D = Record.D;
   ProcessingUpdatesRAIIObj ProcessingUpdates(*this);
   DeclUpdateOffsetsMap::iterator UpdI = DeclUpdateOffsets.find(ID);
 
-  SmallVector<serialization::DeclID, 8> PendingLazySpecializationIDs;
+  SmallVector<GlobalDeclID, 8> PendingLazySpecializationIDs;
 
   if (UpdI != DeclUpdateOffsets.end()) {
     auto UpdateOffsets = std::move(UpdI->second);
@@ -4327,7 +4326,7 @@ void ASTReader::loadPendingDeclChain(Decl *FirstLocal, uint64_t LocalOffset) {
   // we should instead generate one loop per kind and dispatch up-front?
   Decl *MostRecent = FirstLocal;
   for (unsigned I = 0, N = Record.size(); I != N; ++I) {
-    auto *D = GetLocalDecl(*M, Record[N - I - 1]);
+    auto *D = GetLocalDecl(*M, LocalDeclID(Record[N - I - 1]));
     ASTDeclReader::attachPreviousDecl(*this, D, MostRecent, CanonDecl);
     MostRecent = D;
   }
@@ -4344,7 +4343,7 @@ namespace {
     llvm::SmallPtrSetImpl<ObjCCategoryDecl *> &Deserialized;
     ObjCCategoryDecl *Tail = nullptr;
     llvm::DenseMap<DeclarationName, ObjCCategoryDecl *> NameCategoryMap;
-    serialization::GlobalDeclID InterfaceID;
+    GlobalDeclID InterfaceID;
     unsigned PreviousGeneration;
 
     void add(ObjCCategoryDecl *Cat) {
@@ -4386,11 +4385,10 @@ namespace {
     }
 
   public:
-    ObjCCategoriesVisitor(ASTReader &Reader,
-                          ObjCInterfaceDecl *Interface,
-                          llvm::SmallPtrSetImpl<ObjCCategoryDecl *> &Deserialized,
-                          serialization::GlobalDeclID InterfaceID,
-                          unsigned PreviousGeneration)
+    ObjCCategoriesVisitor(
+        ASTReader &Reader, ObjCInterfaceDecl *Interface,
+        llvm::SmallPtrSetImpl<ObjCCategoryDecl *> &Deserialized,
+        GlobalDeclID InterfaceID, unsigned PreviousGeneration)
         : Reader(Reader), Interface(Interface), Deserialized(Deserialized),
           InterfaceID(InterfaceID), PreviousGeneration(PreviousGeneration) {
       // Populate the name -> category map with the set of known categories.
@@ -4412,8 +4410,9 @@ namespace {
       // Map global ID of the definition down to the local ID used in this
       // module file. If there is no such mapping, we'll find nothing here
       // (or in any module it imports).
-      DeclID LocalID = Reader.mapGlobalIDToModuleFileGlobalID(M, InterfaceID);
-      if (!LocalID)
+      LocalDeclID LocalID =
+          Reader.mapGlobalIDToModuleFileGlobalID(M, InterfaceID);
+      if (LocalID.isInvalid())
         return true;
 
       // Perform a binary search to find the local redeclarations for this
@@ -4437,15 +4436,14 @@ namespace {
       M.ObjCCategories[Offset++] = 0; // Don't try to deserialize again
       for (unsigned I = 0; I != N; ++I)
         add(cast_or_null<ObjCCategoryDecl>(
-              Reader.GetLocalDecl(M, M.ObjCCategories[Offset++])));
+            Reader.GetLocalDecl(M, LocalDeclID(M.ObjCCategories[Offset++]))));
       return true;
     }
   };
 
 } // namespace
 
-void ASTReader::loadObjCCategories(serialization::GlobalDeclID ID,
-                                   ObjCInterfaceDecl *D,
+void ASTReader::loadObjCCategories(GlobalDeclID ID, ObjCInterfaceDecl *D,
                                    unsigned PreviousGeneration) {
   ObjCCategoriesVisitor Visitor(*this, D, CategoriesDeserialized, ID,
                                 PreviousGeneration);
@@ -4473,8 +4471,9 @@ static void forAllLaterRedecls(DeclT *D, Fn F) {
   }
 }
 
-void ASTDeclReader::UpdateDecl(Decl *D,
-   llvm::SmallVectorImpl<serialization::DeclID> &PendingLazySpecializationIDs) {
+void ASTDeclReader::UpdateDecl(
+    Decl *D,
+    llvm::SmallVectorImpl<GlobalDeclID> &PendingLazySpecializationIDs) {
   while (Record.getIdx() < Record.size()) {
     switch ((DeclUpdateKind)Record.readInt()) {
     case UPD_CXX_ADDED_IMPLICIT_MEMBER: {
diff --git a/clang/lib/Serialization/ASTReaderInternals.h b/clang/lib/Serialization/ASTReaderInternals.h
index 25a46ddabcb7..49268ad5251d 100644
--- a/clang/lib/Serialization/ASTReaderInternals.h
+++ b/clang/lib/Serialization/ASTReaderInternals.h
@@ -49,15 +49,15 @@ public:
   static const int MaxTables = 4;
 
   /// The lookup result is a list of global declaration IDs.
-  using data_type = SmallVector<DeclID, 4>;
+  using data_type = SmallVector<GlobalDeclID, 4>;
 
   struct data_type_builder {
     data_type &Data;
-    llvm::DenseSet<DeclID> Found;
+    llvm::DenseSet<GlobalDeclID> Found;
 
     data_type_builder(data_type &D) : Data(D) {}
 
-    void insert(DeclID ID) {
+    void insert(GlobalDeclID ID) {
       // Just use a linear scan unless we have more than a few IDs.
       if (Found.empty() && !Data.empty()) {
         if (Data.size() <= 4) {
@@ -108,7 +108,7 @@ public:
 
   static void MergeDataInto(const data_type &From, data_type_builder &To) {
     To.Data.reserve(To.Data.size() + From.size());
-    for (DeclID ID : From)
+    for (GlobalDeclID ID : From)
       To.insert(ID);
   }
 
diff --git a/clang/lib/Serialization/ASTReaderStmt.cpp b/clang/lib/Serialization/ASTReaderStmt.cpp
index baded0fe1983..7d3930022a69 100644
--- a/clang/lib/Serialization/ASTReaderStmt.cpp
+++ b/clang/lib/Serialization/ASTReaderStmt.cpp
@@ -956,14 +956,22 @@ void ASTStmtReader::VisitMatrixSubscriptExpr(MatrixSubscriptExpr *E) {
   E->setRBracketLoc(readSourceLocation());
 }
 
-void ASTStmtReader::VisitOMPArraySectionExpr(OMPArraySectionExpr *E) {
+void ASTStmtReader::VisitArraySectionExpr(ArraySectionExpr *E) {
   VisitExpr(E);
+  E->ASType = Record.readEnum<ArraySectionExpr::ArraySectionType>();
+
   E->setBase(Record.readSubExpr());
   E->setLowerBound(Record.readSubExpr());
   E->setLength(Record.readSubExpr());
-  E->setStride(Record.readSubExpr());
+
+  if (E->isOMPArraySection())
+    E->setStride(Record.readSubExpr());
+
   E->setColonLocFirst(readSourceLocation());
-  E->setColonLocSecond(readSourceLocation());
+
+  if (E->isOMPArraySection())
+    E->setColonLocSecond(readSourceLocation());
+
   E->setRBracketLoc(readSourceLocation());
 }
 
@@ -3090,8 +3098,8 @@ Stmt *ASTReader::ReadStmtFromStream(ModuleFile &F) {
       S = new (Context) MatrixSubscriptExpr(Empty);
       break;
 
-    case EXPR_OMP_ARRAY_SECTION:
-      S = new (Context) OMPArraySectionExpr(Empty);
+    case EXPR_ARRAY_SECTION:
+      S = new (Context) ArraySectionExpr(Empty);
       break;
 
     case EXPR_OMP_ARRAY_SHAPING:
diff --git a/clang/lib/Serialization/ASTWriter.cpp b/clang/lib/Serialization/ASTWriter.cpp
index 018b854652a4..0408eeb6a95b 100644
--- a/clang/lib/Serialization/ASTWriter.cpp
+++ b/clang/lib/Serialization/ASTWriter.cpp
@@ -171,34 +171,9 @@ GetAffectingModuleMaps(const Preprocessor &PP, Module *RootModule) {
            .ModulesPruneNonAffectingModuleMaps)
     return std::nullopt;
 
-  SmallVector<const Module *> ModulesToProcess{RootModule};
-
   const HeaderSearch &HS = PP.getHeaderSearchInfo();
-
-  SmallVector<OptionalFileEntryRef, 16> FilesByUID;
-  HS.getFileMgr().GetUniqueIDMapping(FilesByUID);
-
-  if (FilesByUID.size() > HS.header_file_size())
-    FilesByUID.resize(HS.header_file_size());
-
-  for (unsigned UID = 0, LastUID = FilesByUID.size(); UID != LastUID; ++UID) {
-    OptionalFileEntryRef File = FilesByUID[UID];
-    if (!File)
-      continue;
-
-    const HeaderFileInfo *HFI = HS.getExistingLocalFileInfo(*File);
-    if (!HFI || (HFI->isModuleHeader && !HFI->isCompilingModuleHeader))
-      continue;
-
-    for (const auto &KH : HS.findResolvedModulesForHeader(*File)) {
-      if (!KH.getModule())
-        continue;
-      ModulesToProcess.push_back(KH.getModule());
-    }
-  }
-
   const ModuleMap &MM = HS.getModuleMap();
-  SourceManager &SourceMgr = PP.getSourceManager();
+  const SourceManager &SourceMgr = PP.getSourceManager();
 
   std::set<const FileEntry *> ModuleMaps;
   auto CollectIncludingModuleMaps = [&](FileID FID, FileEntryRef F) {
@@ -233,12 +208,48 @@ GetAffectingModuleMaps(const Preprocessor &PP, Module *RootModule) {
     }
   };
 
-  for (const Module *CurrentModule : ModulesToProcess) {
+  // Handle all the affecting modules referenced from the root module.
+
+  std::queue<const Module *> Q;
+  Q.push(RootModule);
+  while (!Q.empty()) {
+    const Module *CurrentModule = Q.front();
+    Q.pop();
+
     CollectIncludingMapsFromAncestors(CurrentModule);
     for (const Module *ImportedModule : CurrentModule->Imports)
       CollectIncludingMapsFromAncestors(ImportedModule);
     for (const Module *UndeclaredModule : CurrentModule->UndeclaredUses)
       CollectIncludingMapsFromAncestors(UndeclaredModule);
+
+    for (auto *M : CurrentModule->submodules())
+      Q.push(M);
+  }
+
+  // Handle textually-included headers that belong to other modules.
+
+  SmallVector<OptionalFileEntryRef, 16> FilesByUID;
+  HS.getFileMgr().GetUniqueIDMapping(FilesByUID);
+
+  if (FilesByUID.size() > HS.header_file_size())
+    FilesByUID.resize(HS.header_file_size());
+
+  for (unsigned UID = 0, LastUID = FilesByUID.size(); UID != LastUID; ++UID) {
+    OptionalFileEntryRef File = FilesByUID[UID];
+    if (!File)
+      continue;
+
+    const HeaderFileInfo *HFI = HS.getExistingLocalFileInfo(*File);
+    if (!HFI)
+      continue; // We have no information on this being a header file.
+    if (!HFI->isCompilingModuleHeader && HFI->isModuleHeader)
+      continue; // Modular header, handled in the above module-based loop.
+    if (!HFI->isCompilingModuleHeader && !HFI->IsLocallyIncluded)
+      continue; // Non-modular header not included locally is not affecting.
+
+    for (const auto &KH : HS.findResolvedModulesForHeader(*File))
+      if (const Module *M = KH.getModule())
+        CollectIncludingMapsFromAncestors(M);
   }
 
   return ModuleMaps;
@@ -2053,14 +2064,13 @@ void ASTWriter::WriteHeaderSearch(const HeaderSearch &HS) {
     if (!File)
       continue;
 
-    // Get the file info. Skip emitting this file if we have no information on
-    // it as a header file (in which case HFI will be null) or if it hasn't
-    // changed since it was loaded. Also skip it if it's for a modular header
-    // from a different module; in that case, we rely on the module(s)
-    // containing the header to provide this information.
     const HeaderFileInfo *HFI = HS.getExistingLocalFileInfo(*File);
-    if (!HFI || (HFI->isModuleHeader && !HFI->isCompilingModuleHeader))
-      continue;
+    if (!HFI)
+      continue; // We have no information on this being a header file.
+    if (!HFI->isCompilingModuleHeader && HFI->isModuleHeader)
+      continue; // Header file info is tracked by the owning module file.
+    if (!HFI->isCompilingModuleHeader && !PP->alreadyIncluded(*File))
+      continue; // Non-modular header not included is not needed.
 
     // Massage the file path into an appropriate form.
     StringRef Filename = File->getName();
@@ -3033,7 +3043,7 @@ void ASTWriter::WriteSubmodules(Module *WritingModule) {
     RecordData Inits;
     for (Decl *D : Context->getModuleInitializers(Mod))
       if (wasDeclEmitted(D))
-        Inits.push_back(GetDeclRef(D));
+        AddDeclRef(D, Inits);
     if (!Inits.empty())
       Stream.EmitRecord(SUBMODULE_INITIALIZERS, Inits);
 
@@ -3210,13 +3220,13 @@ uint64_t ASTWriter::WriteDeclContextLexicalBlock(ASTContext &Context,
     return 0;
 
   uint64_t Offset = Stream.GetCurrentBitNo();
-  SmallVector<uint32_t, 128> KindDeclPairs;
+  SmallVector<DeclID, 128> KindDeclPairs;
   for (const auto *D : DC->decls()) {
     if (DoneWritingDeclsAndTypes && !wasDeclEmitted(D))
       continue;
 
     KindDeclPairs.push_back(D->getKind());
-    KindDeclPairs.push_back(GetDeclRef(D));
+    KindDeclPairs.push_back(GetDeclRef(D).get());
   }
 
   ++NumLexicalDeclContexts;
@@ -3251,7 +3261,7 @@ void ASTWriter::WriteTypeDeclOffsets() {
   unsigned DeclOffsetAbbrev = Stream.EmitAbbrev(std::move(Abbrev));
   {
     RecordData::value_type Record[] = {DECL_OFFSET, DeclOffsets.size(),
-                                       FirstDeclID - NUM_PREDEF_DECL_IDS};
+                                       FirstDeclID.get() - NUM_PREDEF_DECL_IDS};
     Stream.EmitRecordWithBlob(DeclOffsetAbbrev, Record, bytes(DeclOffsets));
   }
 }
@@ -3272,7 +3282,7 @@ void ASTWriter::WriteFileDeclIDsMap() {
     Info.FirstDeclIndex = FileGroupedDeclIDs.size();
     llvm::stable_sort(Info.DeclIDs);
     for (auto &LocDeclEntry : Info.DeclIDs)
-      FileGroupedDeclIDs.push_back(LocDeclEntry.second);
+      FileGroupedDeclIDs.push_back(LocDeclEntry.second.get());
   }
 
   auto Abbrev = std::make_shared<BitCodeAbbrev>();
@@ -3348,11 +3358,11 @@ public:
     for (const ObjCMethodList *Method = &Methods.Instance; Method;
          Method = Method->getNext())
       if (ShouldWriteMethodListNode(Method))
-        DataLen += 4;
+        DataLen += sizeof(DeclID);
     for (const ObjCMethodList *Method = &Methods.Factory; Method;
          Method = Method->getNext())
       if (ShouldWriteMethodListNode(Method))
-        DataLen += 4;
+        DataLen += sizeof(DeclID);
     return emitULEBKeyDataLength(KeyLen, DataLen, Out);
   }
 
@@ -3410,11 +3420,11 @@ public:
     for (const ObjCMethodList *Method = &Methods.Instance; Method;
          Method = Method->getNext())
       if (ShouldWriteMethodListNode(Method))
-        LE.write<uint32_t>(Writer.getDeclID(Method->getMethod()));
+        LE.write<DeclID>((DeclID)Writer.getDeclID(Method->getMethod()));
     for (const ObjCMethodList *Method = &Methods.Factory; Method;
          Method = Method->getNext())
       if (ShouldWriteMethodListNode(Method))
-        LE.write<uint32_t>(Writer.getDeclID(Method->getMethod()));
+        LE.write<DeclID>((DeclID)Writer.getDeclID(Method->getMethod()));
 
     assert(Out.tell() - Start == DataLen && "Data length is wrong");
   }
@@ -3618,7 +3628,6 @@ class ASTIdentifierTableTrait {
   /// doesn't check whether the name has macros defined; use PublicMacroIterator
   /// to check that.
   bool isInterestingIdentifier(const IdentifierInfo *II, uint64_t MacroOffset) {
-    II->getObjCOrBuiltinID();
     bool IsInteresting =
         II->getNotableIdentifierID() !=
             tok::NotableIdentifierKind::not_notable ||
@@ -3687,7 +3696,8 @@ public:
         DataLen += 4; // MacroDirectives offset.
 
       if (NeedDecls)
-        DataLen += std::distance(IdResolver.begin(II), IdResolver.end()) * 4;
+        DataLen += std::distance(IdResolver.begin(II), IdResolver.end()) *
+                   sizeof(DeclID);
     }
     return emitULEBKeyDataLength(KeyLen, DataLen, Out);
   }
@@ -3733,8 +3743,8 @@ public:
       // Only emit declarations that aren't from a chained PCH, though.
       SmallVector<NamedDecl *, 16> Decls(IdResolver.decls(II));
       for (NamedDecl *D : llvm::reverse(Decls))
-        LE.write<uint32_t>(
-            Writer.getDeclID(getDeclForLocalLookup(PP.getLangOpts(), D)));
+        LE.write<DeclID>((DeclID)Writer.getDeclID(
+            getDeclForLocalLookup(PP.getLangOpts(), D)));
     }
   }
 };
@@ -3850,7 +3860,7 @@ namespace {
 // Trait used for the on-disk hash table used in the method pool.
 class ASTDeclContextNameLookupTrait {
   ASTWriter &Writer;
-  llvm::SmallVector<DeclID, 64> DeclIDs;
+  llvm::SmallVector<LocalDeclID, 64> DeclIDs;
 
 public:
   using key_type = DeclarationNameKey;
@@ -3883,7 +3893,10 @@ public:
 
   data_type ImportData(const reader::ASTDeclContextNameLookupTrait::data_type &FromReader) {
     unsigned Start = DeclIDs.size();
-    llvm::append_range(DeclIDs, FromReader);
+    DeclIDs.insert(
+        DeclIDs.end(),
+        DeclIDIterator<GlobalDeclID, LocalDeclID>(FromReader.begin()),
+        DeclIDIterator<GlobalDeclID, LocalDeclID>(FromReader.end()));
     return std::make_pair(Start, DeclIDs.size());
   }
 
@@ -3928,8 +3941,8 @@ public:
       break;
     }
 
-    // 4 bytes for each DeclID.
-    unsigned DataLen = 4 * (Lookup.second - Lookup.first);
+    // length of DeclIDs.
+    unsigned DataLen = sizeof(DeclID) * (Lookup.second - Lookup.first);
 
     return emitULEBKeyDataLength(KeyLen, DataLen, Out);
   }
@@ -3972,7 +3985,7 @@ public:
     endian::Writer LE(Out, llvm::endianness::little);
     uint64_t Start = Out.tell(); (void)Start;
     for (unsigned I = Lookup.first, N = Lookup.second; I != N; ++I)
-      LE.write<uint32_t>(DeclIDs[I]);
+      LE.write<DeclID>((DeclID)DeclIDs[I]);
     assert(Out.tell() - Start == DataLen && "Data length is wrong");
   }
 };
@@ -4306,7 +4319,8 @@ void ASTWriter::WriteDeclContextVisibleUpdate(const DeclContext *DC) {
     DC = cast<DeclContext>(Chain->getKeyDeclaration(cast<Decl>(DC)));
 
   // Write the lookup table
-  RecordData::value_type Record[] = {UPDATE_VISIBLE, getDeclID(cast<Decl>(DC))};
+  RecordData::value_type Record[] = {UPDATE_VISIBLE,
+                                     getDeclID(cast<Decl>(DC)).get()};
   Stream.EmitRecordWithBlob(UpdateVisibleAbbrev, Record, LookupTable);
 }
 
@@ -4360,7 +4374,7 @@ void ASTWriter::WriteObjCCategories() {
            Cat = Class->known_categories_begin(),
            CatEnd = Class->known_categories_end();
          Cat != CatEnd; ++Cat, ++Size) {
-      assert(getDeclID(*Cat) != 0 && "Bogus category");
+      assert(getDeclID(*Cat).isValid() && "Bogus category");
       AddDeclRef(*Cat, Categories);
     }
 
@@ -5078,7 +5092,7 @@ void ASTWriter::WriteSpecialDeclRecords(Sema &SemaRef) {
       if (!D || !wasDeclEmitted(D))
         SemaDeclRefs.push_back(0);
       else
-        SemaDeclRefs.push_back(getDeclID(D));
+        AddDeclRef(D, SemaDeclRefs);
     };
 
     AddEmittedDeclRefOrZero(SemaRef.getStdNamespace());
@@ -5089,10 +5103,10 @@ void ASTWriter::WriteSpecialDeclRecords(Sema &SemaRef) {
     Stream.EmitRecord(SEMA_DECL_REFS, SemaDeclRefs);
 
   // Write the record containing decls to be checked for deferred diags.
-  SmallVector<serialization::DeclID, 64> DeclsToCheckForDeferredDiags;
+  RecordData DeclsToCheckForDeferredDiags;
   for (auto *D : SemaRef.DeclsToCheckForDeferredDiags)
     if (wasDeclEmitted(D))
-      DeclsToCheckForDeferredDiags.push_back(getDeclID(D));
+      AddDeclRef(D, DeclsToCheckForDeferredDiags);
   if (!DeclsToCheckForDeferredDiags.empty())
     Stream.EmitRecord(DECLS_TO_CHECK_FOR_DEFERRED_DIAGS,
         DeclsToCheckForDeferredDiags);
@@ -5462,7 +5476,7 @@ void ASTWriter::WriteDeclAndTypes(ASTContext &Context) {
     if (VisibleOffset)
       VisibleOffset -= DeclTypesBlockStartOffset;
 
-    DelayedNamespaceRecord.push_back(getDeclID(NS));
+    AddDeclRef(NS, DelayedNamespaceRecord);
     DelayedNamespaceRecord.push_back(LexicalOffset);
     DelayedNamespaceRecord.push_back(VisibleOffset);
   }
@@ -5486,7 +5500,7 @@ void ASTWriter::WriteDeclAndTypes(ASTContext &Context) {
   const TranslationUnitDecl *TU = Context.getTranslationUnitDecl();
   // Create a lexical update block containing all of the declarations in the
   // translation unit that do not come from other AST files.
-  SmallVector<uint32_t, 128> NewGlobalKindDeclPairs;
+  SmallVector<DeclID, 128> NewGlobalKindDeclPairs;
   for (const auto *D : TU->noload_decls()) {
     if (D->isFromASTFile())
       continue;
@@ -5496,7 +5510,7 @@ void ASTWriter::WriteDeclAndTypes(ASTContext &Context) {
       continue;
 
     NewGlobalKindDeclPairs.push_back(D->getKind());
-    NewGlobalKindDeclPairs.push_back(GetDeclRef(D));
+    NewGlobalKindDeclPairs.push_back(GetDeclRef(D).get());
   }
 
   auto Abv = std::make_shared<llvm::BitCodeAbbrev>();
@@ -5557,7 +5571,7 @@ void ASTWriter::WriteDeclUpdatesBlocks(RecordDataImpl &OffsetsRecord) {
       case UPD_CXX_ADDED_TEMPLATE_SPECIALIZATION:
       case UPD_CXX_ADDED_ANONYMOUS_NAMESPACE:
         assert(Update.getDecl() && "no decl to add?");
-        Record.push_back(GetDeclRef(Update.getDecl()));
+        Record.AddDeclRef(Update.getDecl());
         break;
 
       case UPD_CXX_ADDED_FUNCTION_DEFINITION:
@@ -5698,7 +5712,7 @@ void ASTWriter::WriteDeclUpdatesBlocks(RecordDataImpl &OffsetsRecord) {
       }
     }
 
-    OffsetsRecord.push_back(GetDeclRef(D));
+    AddDeclRef(D, OffsetsRecord);
     OffsetsRecord.push_back(Record.Emit(DECL_UPDATES));
   }
 }
@@ -5963,18 +5977,18 @@ void ASTWriter::AddEmittedDeclRef(const Decl *D, RecordDataImpl &Record) {
   if (!wasDeclEmitted(D))
     return;
 
-  Record.push_back(GetDeclRef(D));
+  Record.push_back(GetDeclRef(D).get());
 }
 
 void ASTWriter::AddDeclRef(const Decl *D, RecordDataImpl &Record) {
-  Record.push_back(GetDeclRef(D));
+  Record.push_back(GetDeclRef(D).get());
 }
 
-DeclID ASTWriter::GetDeclRef(const Decl *D) {
+LocalDeclID ASTWriter::GetDeclRef(const Decl *D) {
   assert(WritingAST && "Cannot request a declaration ID before AST writing");
 
   if (!D) {
-    return 0;
+    return LocalDeclID();
   }
 
   // If the DeclUpdate from the GMF gets touched, emit it.
@@ -5988,14 +6002,14 @@ DeclID ASTWriter::GetDeclRef(const Decl *D) {
   // If D comes from an AST file, its declaration ID is already known and
   // fixed.
   if (D->isFromASTFile())
-    return D->getGlobalID();
+    return LocalDeclID(D->getGlobalID());
 
   assert(!(reinterpret_cast<uintptr_t>(D) & 0x01) && "Invalid decl pointer");
-  DeclID &ID = DeclIDs[D];
-  if (ID == 0) {
+  LocalDeclID &ID = DeclIDs[D];
+  if (ID.isInvalid()) {
     if (DoneWritingDeclsAndTypes) {
       assert(0 && "New decl seen after serializing all the decls to emit!");
-      return 0;
+      return LocalDeclID();
     }
 
     // We haven't seen this declaration before. Give it a new ID and
@@ -6007,14 +6021,14 @@ DeclID ASTWriter::GetDeclRef(const Decl *D) {
   return ID;
 }
 
-DeclID ASTWriter::getDeclID(const Decl *D) {
+LocalDeclID ASTWriter::getDeclID(const Decl *D) {
   if (!D)
-    return 0;
+    return LocalDeclID();
 
   // If D comes from an AST file, its declaration ID is already known and
   // fixed.
   if (D->isFromASTFile())
-    return D->getGlobalID();
+    return LocalDeclID(D->getGlobalID());
 
   assert(DeclIDs.contains(D) && "Declaration not emitted!");
   return DeclIDs[D];
@@ -6035,8 +6049,8 @@ bool ASTWriter::wasDeclEmitted(const Decl *D) const {
   return Emitted;
 }
 
-void ASTWriter::associateDeclWithFile(const Decl *D, DeclID ID) {
-  assert(ID);
+void ASTWriter::associateDeclWithFile(const Decl *D, LocalDeclID ID) {
+  assert(ID.isValid());
   assert(D);
 
   SourceLocation Loc = D->getLocation();
@@ -6068,7 +6082,7 @@ void ASTWriter::associateDeclWithFile(const Decl *D, DeclID ID) {
   if (!Info)
     Info = std::make_unique<DeclIDInFileInfo>();
 
-  std::pair<unsigned, serialization::DeclID> LocDecl(Offset, ID);
+  std::pair<unsigned, LocalDeclID> LocDecl(Offset, ID);
   LocDeclIDsTy &Decls = Info->DeclIDs;
   Decls.push_back(LocDecl);
 }
@@ -6338,7 +6352,7 @@ void ASTRecordWriter::AddCXXDefinitionData(const CXXRecordDecl *D) {
       Writer->Context->getLangOpts().ModulesDebugInfo && !D->isDependentType();
   Record->push_back(ModulesDebugInfo);
   if (ModulesDebugInfo)
-    Writer->ModularCodegenDecls.push_back(Writer->GetDeclRef(D));
+    Writer->AddDeclRef(D, Writer->ModularCodegenDecls);
 
   // IsLambda bit is already saved.
 
@@ -6442,7 +6456,7 @@ void ASTWriter::ReaderInitialized(ASTReader *Reader) {
 
   // Note, this will get called multiple times, once one the reader starts up
   // and again each time it's done reading a PCH or module.
-  FirstDeclID = NUM_PREDEF_DECL_IDS + Chain->getTotalNumDecls();
+  FirstDeclID = LocalDeclID(NUM_PREDEF_DECL_IDS + Chain->getTotalNumDecls());
   FirstTypeID = NUM_PREDEF_TYPE_IDS + Chain->getTotalNumTypes();
   FirstIdentID = NUM_PREDEF_IDENT_IDS + Chain->getTotalNumIdentifiers();
   FirstMacroID = NUM_PREDEF_MACRO_IDS + Chain->getTotalNumMacros();
diff --git a/clang/lib/Serialization/ASTWriterDecl.cpp b/clang/lib/Serialization/ASTWriterDecl.cpp
index c6db107e0ca4..0edc4feda3ef 100644
--- a/clang/lib/Serialization/ASTWriterDecl.cpp
+++ b/clang/lib/Serialization/ASTWriterDecl.cpp
@@ -223,9 +223,9 @@ namespace clang {
         assert(!Common->LazySpecializations);
       }
 
-      ArrayRef<DeclID> LazySpecializations;
+      ArrayRef<GlobalDeclID> LazySpecializations;
       if (auto *LS = Common->LazySpecializations)
-        LazySpecializations = llvm::ArrayRef(LS + 1, LS[0]);
+        LazySpecializations = llvm::ArrayRef(LS + 1, LS[0].get());
 
       // Add a slot to the record for the number of specializations.
       unsigned I = Record.size();
@@ -243,7 +243,9 @@ namespace clang {
         assert(D->isCanonicalDecl() && "non-canonical decl in set");
         AddFirstDeclFromEachModule(D, /*IncludeLocal*/true);
       }
-      Record.append(LazySpecializations.begin(), LazySpecializations.end());
+      Record.append(
+          DeclIDIterator<GlobalDeclID, DeclID>(LazySpecializations.begin()),
+          DeclIDIterator<GlobalDeclID, DeclID>(LazySpecializations.end()));
 
       // Update the size entry we added earlier.
       Record[I] = Record.size() - I - 1;
@@ -1166,7 +1168,7 @@ void ASTDeclWriter::VisitVarDecl(VarDecl *D) {
   Record.push_back(VarDeclBits);
 
   if (ModulesCodegen)
-    Writer.ModularCodegenDecls.push_back(Writer.GetDeclRef(D));
+    Writer.AddDeclRef(D, Writer.ModularCodegenDecls);
 
   if (D->hasAttr<BlocksAttr>()) {
     BlockVarCopyInit Init = Writer.Context->getBlockVarCopyInit(D);
@@ -2786,10 +2788,10 @@ void ASTWriter::WriteDecl(ASTContext &Context, Decl *D) {
                                       "serializing");
 
   // Determine the ID for this declaration.
-  serialization::DeclID ID;
+  LocalDeclID ID;
   assert(!D->isFromASTFile() && "should not be emitting imported decl");
-  serialization::DeclID &IDR = DeclIDs[D];
-  if (IDR == 0)
+  LocalDeclID &IDR = DeclIDs[D];
+  if (IDR.isInvalid())
     IDR = NextDeclID++;
 
   ID = IDR;
@@ -2807,7 +2809,7 @@ void ASTWriter::WriteDecl(ASTContext &Context, Decl *D) {
 
   // Record the offset for this declaration
   SourceLocation Loc = D->getLocation();
-  unsigned Index = ID - FirstDeclID;
+  unsigned Index = ID.get() - FirstDeclID.get();
   if (DeclOffsets.size() == Index)
     DeclOffsets.emplace_back(getAdjustedLocation(Loc), Offset,
                              DeclTypesBlockStartOffset);
@@ -2827,7 +2829,7 @@ void ASTWriter::WriteDecl(ASTContext &Context, Decl *D) {
   // Note declarations that should be deserialized eagerly so that we can add
   // them to a record in the AST file later.
   if (isRequiredDecl(D, Context, WritingModule))
-    EagerlyDeserializedDecls.push_back(ID);
+    AddDeclRef(D, EagerlyDeserializedDecls);
 }
 
 void ASTRecordWriter::AddFunctionDefinition(const FunctionDecl *FD) {
@@ -2863,7 +2865,7 @@ void ASTRecordWriter::AddFunctionDefinition(const FunctionDecl *FD) {
   }
   Record->push_back(ModulesCodegen);
   if (ModulesCodegen)
-    Writer->ModularCodegenDecls.push_back(Writer->GetDeclRef(FD));
+    Writer->AddDeclRef(FD, Writer->ModularCodegenDecls);
   if (auto *CD = dyn_cast<CXXConstructorDecl>(FD)) {
     Record->push_back(CD->getNumCtorInitializers());
     if (CD->getNumCtorInitializers())
diff --git a/clang/lib/Serialization/ASTWriterStmt.cpp b/clang/lib/Serialization/ASTWriterStmt.cpp
index cd5f733baf76..39aec31b6d87 100644
--- a/clang/lib/Serialization/ASTWriterStmt.cpp
+++ b/clang/lib/Serialization/ASTWriterStmt.cpp
@@ -880,16 +880,21 @@ void ASTStmtWriter::VisitMatrixSubscriptExpr(MatrixSubscriptExpr *E) {
   Code = serialization::EXPR_ARRAY_SUBSCRIPT;
 }
 
-void ASTStmtWriter::VisitOMPArraySectionExpr(OMPArraySectionExpr *E) {
+void ASTStmtWriter::VisitArraySectionExpr(ArraySectionExpr *E) {
   VisitExpr(E);
+  Record.writeEnum(E->ASType);
   Record.AddStmt(E->getBase());
   Record.AddStmt(E->getLowerBound());
   Record.AddStmt(E->getLength());
-  Record.AddStmt(E->getStride());
+  if (E->isOMPArraySection())
+    Record.AddStmt(E->getStride());
   Record.AddSourceLocation(E->getColonLocFirst());
-  Record.AddSourceLocation(E->getColonLocSecond());
+
+  if (E->isOMPArraySection())
+    Record.AddSourceLocation(E->getColonLocSecond());
+
   Record.AddSourceLocation(E->getRBracketLoc());
-  Code = serialization::EXPR_OMP_ARRAY_SECTION;
+  Code = serialization::EXPR_ARRAY_SECTION;
 }
 
 void ASTStmtWriter::VisitOMPArrayShapingExpr(OMPArrayShapingExpr *E) {
diff --git a/clang/lib/StaticAnalyzer/Checkers/DereferenceChecker.cpp b/clang/lib/StaticAnalyzer/Checkers/DereferenceChecker.cpp
index a678c3827e7f..1cebfbbee77d 100644
--- a/clang/lib/StaticAnalyzer/Checkers/DereferenceChecker.cpp
+++ b/clang/lib/StaticAnalyzer/Checkers/DereferenceChecker.cpp
@@ -188,9 +188,9 @@ void DereferenceChecker::reportBug(DerefKind K, ProgramStateRef State,
     os << DerefStr1;
     break;
   }
-  case Stmt::OMPArraySectionExprClass: {
+  case Stmt::ArraySectionExprClass: {
     os << "Array access";
-    const OMPArraySectionExpr *AE = cast<OMPArraySectionExpr>(S);
+    const ArraySectionExpr *AE = cast<ArraySectionExpr>(S);
     AddDerefSource(os, Ranges, AE->getBase()->IgnoreParenCasts(),
                    State.get(), N->getLocationContext());
     os << DerefStr1;
diff --git a/clang/lib/StaticAnalyzer/Checkers/IdenticalExprChecker.cpp b/clang/lib/StaticAnalyzer/Checkers/IdenticalExprChecker.cpp
index 1cf81b54e77d..7ac34ef8164e 100644
--- a/clang/lib/StaticAnalyzer/Checkers/IdenticalExprChecker.cpp
+++ b/clang/lib/StaticAnalyzer/Checkers/IdenticalExprChecker.cpp
@@ -350,7 +350,7 @@ static bool isIdenticalStmt(const ASTContext &Ctx, const Stmt *Stmt1,
     return false;
   case Stmt::CallExprClass:
   case Stmt::ArraySubscriptExprClass:
-  case Stmt::OMPArraySectionExprClass:
+  case Stmt::ArraySectionExprClass:
   case Stmt::OMPArrayShapingExprClass:
   case Stmt::OMPIteratorExprClass:
   case Stmt::ImplicitCastExprClass:
diff --git a/clang/lib/StaticAnalyzer/Checkers/StreamChecker.cpp b/clang/lib/StaticAnalyzer/Checkers/StreamChecker.cpp
index bd495cd0f971..a0aa2316a7b4 100644
--- a/clang/lib/StaticAnalyzer/Checkers/StreamChecker.cpp
+++ b/clang/lib/StaticAnalyzer/Checkers/StreamChecker.cpp
@@ -600,7 +600,7 @@ struct StreamOperationEvaluator {
   SValBuilder &SVB;
   const ASTContext &ACtx;
 
-  SymbolRef StreamSym;
+  SymbolRef StreamSym = nullptr;
   const StreamState *SS = nullptr;
   const CallExpr *CE = nullptr;
   StreamErrorState NewES;
diff --git a/clang/lib/StaticAnalyzer/Checkers/Taint.cpp b/clang/lib/StaticAnalyzer/Checkers/Taint.cpp
index 4edb671753bf..6362c82b009d 100644
--- a/clang/lib/StaticAnalyzer/Checkers/Taint.cpp
+++ b/clang/lib/StaticAnalyzer/Checkers/Taint.cpp
@@ -216,21 +216,17 @@ std::vector<SymbolRef> taint::getTaintedSymbolsImpl(ProgramStateRef State,
   std::vector<SymbolRef> TaintedSymbols;
   if (!Reg)
     return TaintedSymbols;
-  // Element region (array element) is tainted if either the base or the offset
-  // are tainted.
+
+  // Element region (array element) is tainted if the offset is tainted.
   if (const ElementRegion *ER = dyn_cast<ElementRegion>(Reg)) {
     std::vector<SymbolRef> TaintedIndex =
         getTaintedSymbolsImpl(State, ER->getIndex(), K, returnFirstOnly);
     llvm::append_range(TaintedSymbols, TaintedIndex);
     if (returnFirstOnly && !TaintedSymbols.empty())
       return TaintedSymbols; // return early if needed
-    std::vector<SymbolRef> TaintedSuperRegion =
-        getTaintedSymbolsImpl(State, ER->getSuperRegion(), K, returnFirstOnly);
-    llvm::append_range(TaintedSymbols, TaintedSuperRegion);
-    if (returnFirstOnly && !TaintedSymbols.empty())
-      return TaintedSymbols; // return early if needed
   }
 
+  // Symbolic region is tainted if the corresponding symbol is tainted.
   if (const SymbolicRegion *SR = dyn_cast<SymbolicRegion>(Reg)) {
     std::vector<SymbolRef> TaintedRegions =
         getTaintedSymbolsImpl(State, SR->getSymbol(), K, returnFirstOnly);
@@ -239,6 +235,8 @@ std::vector<SymbolRef> taint::getTaintedSymbolsImpl(ProgramStateRef State,
       return TaintedSymbols; // return early if needed
   }
 
+  // Any subregion (including Element and Symbolic regions) is tainted if its
+  // super-region is tainted.
   if (const SubRegion *ER = dyn_cast<SubRegion>(Reg)) {
     std::vector<SymbolRef> TaintedSubRegions =
         getTaintedSymbolsImpl(State, ER->getSuperRegion(), K, returnFirstOnly);
@@ -318,4 +316,4 @@ std::vector<SymbolRef> taint::getTaintedSymbolsImpl(ProgramStateRef State,
     }
   }
   return TaintedSymbols;
-}
\ No newline at end of file
+}
diff --git a/clang/lib/StaticAnalyzer/Checkers/WebKit/PtrTypesSemantics.h b/clang/lib/StaticAnalyzer/Checkers/WebKit/PtrTypesSemantics.h
index 9ed8e7cab6ab..ec1db1cc3358 100644
--- a/clang/lib/StaticAnalyzer/Checkers/WebKit/PtrTypesSemantics.h
+++ b/clang/lib/StaticAnalyzer/Checkers/WebKit/PtrTypesSemantics.h
@@ -50,6 +50,9 @@ std::optional<bool> isUncounted(const clang::CXXRecordDecl* Class);
 /// class, false if not, std::nullopt if inconclusive.
 std::optional<bool> isUncountedPtr(const clang::Type* T);
 
+/// \returns true if Name is a RefPtr, Ref, or its variant, false if not.
+bool isRefType(const std::string &Name);
+
 /// \returns true if \p F creates ref-countable object from uncounted parameter,
 /// false if not.
 bool isCtorOfRefCounted(const clang::FunctionDecl *F);
diff --git a/clang/lib/StaticAnalyzer/Checkers/WebKit/UncountedCallArgsChecker.cpp b/clang/lib/StaticAnalyzer/Checkers/WebKit/UncountedCallArgsChecker.cpp
index 8b41a949fd67..ae494de58da3 100644
--- a/clang/lib/StaticAnalyzer/Checkers/WebKit/UncountedCallArgsChecker.cpp
+++ b/clang/lib/StaticAnalyzer/Checkers/WebKit/UncountedCallArgsChecker.cpp
@@ -53,6 +53,13 @@ public:
       bool shouldVisitTemplateInstantiations() const { return true; }
       bool shouldVisitImplicitCode() const { return false; }
 
+      bool TraverseClassTemplateDecl(ClassTemplateDecl *Decl) {
+        if (isRefType(safeGetName(Decl)))
+          return true;
+        return RecursiveASTVisitor<LocalVisitor>::TraverseClassTemplateDecl(
+            Decl);
+      }
+
       bool VisitCallExpr(const CallExpr *CE) {
         Checker->visitCallExpr(CE);
         return true;
diff --git a/clang/lib/StaticAnalyzer/Core/ExprEngine.cpp b/clang/lib/StaticAnalyzer/Core/ExprEngine.cpp
index 09c69f9612d9..0b1edf3e5c96 100644
--- a/clang/lib/StaticAnalyzer/Core/ExprEngine.cpp
+++ b/clang/lib/StaticAnalyzer/Core/ExprEngine.cpp
@@ -1948,7 +1948,7 @@ void ExprEngine::Visit(const Stmt *S, ExplodedNode *Pred,
     case Stmt::CXXPseudoDestructorExprClass:
     case Stmt::SubstNonTypeTemplateParmExprClass:
     case Stmt::CXXNullPtrLiteralExprClass:
-    case Stmt::OMPArraySectionExprClass:
+    case Stmt::ArraySectionExprClass:
     case Stmt::OMPArrayShapingExprClass:
     case Stmt::OMPIteratorExprClass:
     case Stmt::SYCLUniqueStableNameExprClass:
diff --git a/clang/lib/Tooling/DependencyScanning/DependencyScanningWorker.cpp b/clang/lib/Tooling/DependencyScanning/DependencyScanningWorker.cpp
index 32850f5eea92..0c047b6c5da2 100644
--- a/clang/lib/Tooling/DependencyScanning/DependencyScanningWorker.cpp
+++ b/clang/lib/Tooling/DependencyScanning/DependencyScanningWorker.cpp
@@ -439,6 +439,9 @@ public:
     if (Result)
       setLastCC1Arguments(std::move(OriginalInvocation));
 
+    // Propagate the statistics to the parent FileManager.
+    DriverFileMgr->AddStats(ScanInstance.getFileManager());
+
     return Result;
   }
 
diff --git a/clang/test/APINotes/Inputs/Headers/SwiftImportAs.apinotes b/clang/test/APINotes/Inputs/Headers/SwiftImportAs.apinotes
index 5dbb83cab86b..b0eead42869a 100644
--- a/clang/test/APINotes/Inputs/Headers/SwiftImportAs.apinotes
+++ b/clang/test/APINotes/Inputs/Headers/SwiftImportAs.apinotes
@@ -7,3 +7,7 @@ Tags:
   SwiftImportAs: reference
   SwiftReleaseOp: RCRelease
   SwiftRetainOp: RCRetain
+- Name: NonCopyableType
+  SwiftCopyable: false
+- Name: CopyableType
+  SwiftCopyable: true
diff --git a/clang/test/APINotes/Inputs/Headers/SwiftImportAs.h b/clang/test/APINotes/Inputs/Headers/SwiftImportAs.h
index 82b8a6749c4f..a8f6d0248eae 100644
--- a/clang/test/APINotes/Inputs/Headers/SwiftImportAs.h
+++ b/clang/test/APINotes/Inputs/Headers/SwiftImportAs.h
@@ -4,3 +4,6 @@ struct RefCountedType { int value; };
 
 inline void RCRetain(RefCountedType *x) { x->value++; }
 inline void RCRelease(RefCountedType *x) { x->value--; }
+
+struct NonCopyableType { int value; };
+struct CopyableType { int value; };
diff --git a/clang/test/APINotes/swift-import-as.cpp b/clang/test/APINotes/swift-import-as.cpp
index 904857e58593..103cf02f431a 100644
--- a/clang/test/APINotes/swift-import-as.cpp
+++ b/clang/test/APINotes/swift-import-as.cpp
@@ -2,6 +2,8 @@
 // RUN: %clang_cc1 -fmodules -fblocks -fimplicit-module-maps -fmodules-cache-path=%t/ModulesCache -fdisable-module-hash -fapinotes-modules -fsyntax-only -I %S/Inputs/Headers %s -x c++
 // RUN: %clang_cc1 -fmodules -fblocks -fimplicit-module-maps -fmodules-cache-path=%t/ModulesCache -fdisable-module-hash -fapinotes-modules -fsyntax-only -I %S/Inputs/Headers %s -x c++ -ast-dump -ast-dump-filter ImmortalRefType | FileCheck -check-prefix=CHECK-IMMORTAL %s
 // RUN: %clang_cc1 -fmodules -fblocks -fimplicit-module-maps -fmodules-cache-path=%t/ModulesCache -fdisable-module-hash -fapinotes-modules -fsyntax-only -I %S/Inputs/Headers %s -x c++ -ast-dump -ast-dump-filter RefCountedType | FileCheck -check-prefix=CHECK-REF-COUNTED %s
+// RUN: %clang_cc1 -fmodules -fblocks -fimplicit-module-maps -fmodules-cache-path=%t/ModulesCache -fdisable-module-hash -fapinotes-modules -fsyntax-only -I %S/Inputs/Headers %s -x c++ -ast-dump -ast-dump-filter NonCopyableType | FileCheck -check-prefix=CHECK-NON-COPYABLE %s
+// RUN: %clang_cc1 -fmodules -fblocks -fimplicit-module-maps -fmodules-cache-path=%t/ModulesCache -fdisable-module-hash -fapinotes-modules -fsyntax-only -I %S/Inputs/Headers %s -x c++ -ast-dump -ast-dump-filter CopyableType | FileCheck -check-prefix=CHECK-COPYABLE %s
 
 #include <SwiftImportAs.h>
 
@@ -14,3 +16,11 @@
 // CHECK-REF-COUNTED: SwiftAttrAttr {{.+}} <<invalid sloc>> "import_reference"
 // CHECK-REF-COUNTED: SwiftAttrAttr {{.+}} <<invalid sloc>> "retain:RCRetain"
 // CHECK-REF-COUNTED: SwiftAttrAttr {{.+}} <<invalid sloc>> "release:RCRelease"
+
+// CHECK-NON-COPYABLE: Dumping NonCopyableType:
+// CHECK-NON-COPYABLE-NEXT: CXXRecordDecl {{.+}} imported in SwiftImportAs {{.+}} struct NonCopyableType
+// CHECK-NON-COPYABLE: SwiftAttrAttr {{.+}} <<invalid sloc>> "~Copyable"
+
+// CHECK-COPYABLE: Dumping CopyableType:
+// CHECK-COPYABLE-NEXT: CXXRecordDecl {{.+}} imported in SwiftImportAs {{.+}} struct CopyableType
+// CHECK-COPYABLE-NOT: SwiftAttrAttr
diff --git a/clang/test/AST/Interp/c.c b/clang/test/AST/Interp/c.c
index a5951158ed0e..207da5fe8126 100644
--- a/clang/test/AST/Interp/c.c
+++ b/clang/test/AST/Interp/c.c
@@ -263,3 +263,10 @@ const int *p = &b;
 const __int128 K = (__int128)(int*)0;
 const unsigned __int128 KU = (unsigned __int128)(int*)0;
 #endif
+
+
+int test3(void) {
+  int a[2];
+  a[0] = test3; // all-error {{incompatible pointer to integer conversion assigning to 'int' from 'int (void)'}}
+  return 0;
+}
diff --git a/clang/test/AST/Interp/cxx23.cpp b/clang/test/AST/Interp/cxx23.cpp
index f0325eef6d87..13cc9f43febc 100644
--- a/clang/test/AST/Interp/cxx23.cpp
+++ b/clang/test/AST/Interp/cxx23.cpp
@@ -141,3 +141,19 @@ struct check_ice {
     };
 };
 static_assert(check_ice<42>::x == 42);
+
+
+namespace VirtualBases {
+  namespace One {
+    struct U { int n; };
+    struct V : U { int n; };
+    struct A : virtual V { int n; };
+    struct Aa { int n; };
+    struct B : virtual A, Aa {};
+    struct C : virtual A, Aa {};
+    struct D : B, C {};
+
+    /// Calls the constructor of D.
+    D d;
+  }
+}
diff --git a/clang/test/AST/Interp/functions.cpp b/clang/test/AST/Interp/functions.cpp
index f9bb5d53634e..a5bb9f1a19aa 100644
--- a/clang/test/AST/Interp/functions.cpp
+++ b/clang/test/AST/Interp/functions.cpp
@@ -601,3 +601,19 @@ namespace FromIntegral {
                                            // both-warning {{variable length arrays}}
 #endif
 }
+
+namespace {
+  template <typename T> using id = T;
+  template <typename T>
+  constexpr void g() {
+    constexpr id<void (T)> f;
+  }
+
+  static_assert((g<int>(), true), "");
+}
+
+namespace {
+  /// The InitListExpr here is of void type.
+  void bir [[clang::annotate("B", {1, 2, 3, 4})]] (); // both-error {{'annotate' attribute requires parameter 1 to be a constant expression}} \
+                                                      // both-note {{subexpression not valid in a constant expression}}
+}
diff --git a/clang/test/AST/Interp/opencl.cl b/clang/test/AST/Interp/opencl.cl
new file mode 100644
index 000000000000..b9ba4f8b9b55
--- /dev/null
+++ b/clang/test/AST/Interp/opencl.cl
@@ -0,0 +1,32 @@
+// RUN: %clang_cc1 -fsyntax-only -verify=ref,both %s
+// RUN: %clang_cc1 -fsyntax-only -verify=expected,both %s -fexperimental-new-constant-interpreter
+
+// both-no-diagnostics
+
+typedef int int2 __attribute__((ext_vector_type(2)));
+typedef int int3 __attribute__((ext_vector_type(3)));
+typedef int int4 __attribute__((ext_vector_type(4)));
+typedef int int8 __attribute__((ext_vector_type(8)));
+typedef int int16 __attribute__((ext_vector_type(16)));
+
+void foo(int3 arg1, int8 arg2) {
+  int4 auto1;
+  int16 *auto2;
+  int auto3;
+  int2 auto4;
+  struct S *incomplete1;
+
+  int res1[vec_step(arg1) == 4 ? 1 : -1];
+  int res2[vec_step(arg2) == 8 ? 1 : -1];
+  int res3[vec_step(auto1) == 4 ? 1 : -1];
+  int res4[vec_step(*auto2) == 16 ? 1 : -1];
+  int res5[vec_step(auto3) == 1 ? 1 : -1];
+  int res6[vec_step(auto4) == 2 ? 1 : -1];
+  int res7[vec_step(int2) == 2 ? 1 : -1];
+  int res8[vec_step(int3) == 4 ? 1 : -1];
+  int res9[vec_step(int4) == 4 ? 1 : -1];
+  int res10[vec_step(int8) == 8 ? 1 : -1];
+  int res11[vec_step(int16) == 16 ? 1 : -1];
+  int res12[vec_step(void) == 1 ? 1 : -1];
+}
+
diff --git a/clang/test/AST/Interp/records.cpp b/clang/test/AST/Interp/records.cpp
index 3e52354a4a10..771e5adfca34 100644
--- a/clang/test/AST/Interp/records.cpp
+++ b/clang/test/AST/Interp/records.cpp
@@ -90,8 +90,7 @@ struct Ints2 {
   int a = 10;
   int b;
 };
-constexpr Ints2 ints22; // both-error {{without a user-provided default constructor}} \
-                        // expected-error {{must be initialized by a constant expression}}
+constexpr Ints2 ints22; // both-error {{without a user-provided default constructor}}
 
 constexpr Ints2 I2 = Ints2{12, 25};
 static_assert(I2.a == 12, "");
@@ -1031,6 +1030,12 @@ namespace ParenInit {
                      // both-note {{required by 'constinit' specifier}} \
                      // both-note {{reference to temporary is not a constant expression}} \
                      // both-note {{temporary created here}}
+
+
+  /// Initializing an array.
+  constexpr void bar(int i, int j) {
+    int arr[4](i, j);
+  }
 }
 #endif
 
@@ -1330,3 +1335,108 @@ namespace UnnamedBitFields {
   static_assert(a.f == 1.0, "");
   static_assert(a.c == 'a', "");
 }
+
+/// FIXME: This still doesn't work in the new interpreter because
+/// we lack type information for dummy pointers.
+namespace VirtualBases {
+  /// This used to crash.
+  namespace One {
+    class A {
+    protected:
+      int x;
+    };
+    class B : public virtual A {
+    public:
+      int getX() { return x; } // ref-note {{declared here}}
+    };
+
+    class DV : virtual public B{};
+
+    void foo() {
+      DV b;
+      int a[b.getX()]; // both-warning {{variable length arrays}} \
+                       // ref-note {{non-constexpr function 'getX' cannot be used}}
+    }
+  }
+
+  namespace Two {
+    struct U { int n; };
+    struct A : virtual U { int n; };
+    struct B : A {};
+    B a;
+    static_assert((U*)(A*)(&a) == (U*)(&a), "");
+
+    struct C : virtual A {};
+    struct D : B, C {};
+    D d;
+    constexpr B *p = &d;
+    constexpr C *q = &d;
+    static_assert((A*)p == (A*)q, ""); // both-error {{failed}}
+  }
+
+  namespace Three {
+    struct U { int n; };
+    struct V : U { int n; };
+    struct A : virtual V { int n; };
+    struct Aa { int n; };
+    struct B : virtual A, Aa {};
+
+    struct C : virtual A, Aa {};
+
+    struct D : B, C {};
+
+    D d;
+
+    constexpr B *p = &d;
+    constexpr C *q = &d;
+
+    static_assert((void*)p != (void*)q, "");
+    static_assert((A*)p == (A*)q, "");
+    static_assert((Aa*)p != (Aa*)q, "");
+
+    constexpr V *v = p;
+    constexpr V *w = q;
+    constexpr V *x = (A*)p;
+    static_assert(v == w, "");
+    static_assert(v == x, "");
+
+    static_assert((U*)&d == p, "");
+    static_assert((U*)&d == q, "");
+    static_assert((U*)&d == v, "");
+    static_assert((U*)&d == w, "");
+    static_assert((U*)&d == x, "");
+
+    struct X {};
+    struct Y1 : virtual X {};
+    struct Y2 : X {};
+    struct Z : Y1, Y2 {};
+    Z z;
+    static_assert((X*)(Y1*)&z != (X*)(Y2*)&z, "");
+  }
+}
+
+namespace ZeroInit {
+  struct S3 {
+    S3() = default;
+    S3(const S3&) = default;
+    S3(S3&&) = default;
+    constexpr S3(int n) : n(n) {}
+    int n;
+  };
+  constexpr S3 s3d; // both-error {{default initialization of an object of const type 'const S3' without a user-provided default constructor}}
+  static_assert(s3d.n == 0, "");
+}
+
+namespace {
+#if __cplusplus >= 202002L
+  struct C {
+    template <unsigned N> constexpr C(const char (&)[N]) : n(N) {}
+    unsigned n;
+  };
+  template <C c>
+  constexpr auto operator""_c() { return c.n; }
+
+  constexpr auto waldo = "abc"_c;
+  static_assert(waldo == 4, "");
+#endif
+}
diff --git a/clang/test/AST/ast-dump-fpfeatures.cpp b/clang/test/AST/ast-dump-fpfeatures.cpp
index da0011602a72..68144e31a930 100644
--- a/clang/test/AST/ast-dump-fpfeatures.cpp
+++ b/clang/test/AST/ast-dump-fpfeatures.cpp
@@ -1,10 +1,10 @@
 // Test without serialization:
-// RUN: %clang_cc1 -fsyntax-only -triple x86_64-pc-linux -std=c++11 -ast-dump %s \
+// RUN: %clang_cc1 -fsyntax-only -triple x86_64-pc-linux -std=c++11 -fcxx-exceptions -ast-dump %s \
 // RUN: | FileCheck --strict-whitespace %s
 
 // Test with serialization:
-// RUN: %clang_cc1 -triple x86_64-pc-linux -emit-pch -o %t %s
-// RUN: %clang_cc1 -x c++ -triple x86_64-pc-linux -include-pch %t -ast-dump-all /dev/null \
+// RUN: %clang_cc1 -triple x86_64-pc-linux -emit-pch -fcxx-exceptions -o %t %s
+// RUN: %clang_cc1 -x c++ -triple x86_64-pc-linux -include-pch %t -fcxx-exceptions -ast-dump-all /dev/null \
 // RUN: | sed -e "s/ <undeserialized declarations>//" -e "s/ imported//" \
 // RUN: | FileCheck --strict-whitespace %s
 
@@ -187,3 +187,65 @@ float func_18(float x, float y) {
 // CHECK:         CompoundStmt {{.*}} ConstRoundingMode=downward
 // CHECK:           ReturnStmt
 // CHECK:             BinaryOperator {{.*}} ConstRoundingMode=downward
+
+#pragma float_control(precise, off)
+
+__attribute__((optnone))
+float func_19(float x, float y) {
+  return x + y;
+}
+
+// CHECK-LABEL: FunctionDecl {{.*}} func_19 'float (float, float)'
+// CHECK:         CompoundStmt {{.*}} MathErrno=1
+// CHECK:           ReturnStmt
+// CHECK:             BinaryOperator {{.*}} 'float' '+' ConstRoundingMode=downward MathErrno=1
+
+__attribute__((optnone))
+float func_20(float x, float y) try {
+  return x + y;
+} catch (...) {
+  return 1.0;
+}
+
+// CHECK-LABEL: FunctionDecl {{.*}} func_20 'float (float, float)'
+// CHECK:         CompoundStmt {{.*}} ConstRoundingMode=downward MathErrno=1
+// CHECK:           ReturnStmt
+// CHECK:             BinaryOperator {{.*}} 'float' '+' ConstRoundingMode=downward MathErrno=1
+
+struct C21 {
+  C21(float x, float y);
+  __attribute__((optnone)) float a_method(float x, float y) {
+    return x * y;
+  }
+  float member;
+};
+
+// CHECK-LABEL: CXXMethodDecl {{.*}} a_method 'float (float, float)'
+// CHECK:         CompoundStmt {{.*}} ConstRoundingMode=downward MathErrno=1
+// CHECK:           ReturnStmt
+// CHECK:             BinaryOperator {{.*}} 'float' '*' ConstRoundingMode=downward MathErrno=1
+
+__attribute__((optnone)) C21::C21(float x, float y) : member(x + y) {}
+
+// CHECK-LABEL: CXXConstructorDecl {{.*}} C21 'void (float, float)'
+// CHECK:         CXXCtorInitializer {{.*}} 'member' 'float'
+// CHECK:           BinaryOperator {{.*}} 'float' '+' ConstRoundingMode=downward MathErrno=1
+
+template <typename T>
+__attribute__((optnone)) T func_22(T x, T y) {
+  return x + y;
+}
+
+// CHECK-LABEL: FunctionTemplateDecl {{.*}} func_22
+// CHECK:         FunctionDecl {{.*}} func_22 'T (T, T)'
+// CHECK:           CompoundStmt {{.*}} ConstRoundingMode=downward MathErrno=1
+// CHECK:             ReturnStmt
+// CHECK:               BinaryOperator {{.*}} '+' ConstRoundingMode=downward MathErrno=1
+// CHECK:         FunctionDecl {{.*}} func_22 'float (float, float)'
+// CHECK:           CompoundStmt {{.*}} ConstRoundingMode=downward MathErrno=1
+// CHECK:             ReturnStmt
+// CHECK:               BinaryOperator {{.*}} 'float' '+' ConstRoundingMode=downward MathErrno=1
+
+float func_23(float x, float y) {
+  return func_22(x, y);
+}
\ No newline at end of file
diff --git a/clang/test/AST/ast-dump-fpfeatures.m b/clang/test/AST/ast-dump-fpfeatures.m
new file mode 100644
index 000000000000..cf77529a7568
--- /dev/null
+++ b/clang/test/AST/ast-dump-fpfeatures.m
@@ -0,0 +1,29 @@
+// Test without serialization:
+// RUN: %clang_cc1 -fsyntax-only -triple x86_64-pc-linux -ast-dump %s \
+// RUN: | FileCheck --strict-whitespace %s
+
+// Test with serialization:
+// RUN: %clang_cc1 -triple x86_64-pc-linux -emit-pch -o %t %s
+// RUN: %clang_cc1 -x objective-c -triple x86_64-pc-linux -include-pch %t -ast-dump-all /dev/null \
+// RUN: | sed -e "s/ <undeserialized declarations>//" -e "s/ imported//" \
+// RUN: | FileCheck --strict-whitespace %s
+
+
+@interface Adder
+- (float) sum: (float)x with: (float)y __attribute((optnone));
+@end
+
+#pragma float_control(precise, off)
+
+@implementation Adder
+- (float) sum: (float)x with: (float)y __attribute((optnone)) {
+  return x + y;
+}
+
+@end
+
+// CHECK-LABEL: ObjCImplementationDecl {{.*}} Adder
+// CHECK:         ObjCMethodDecl {{.*}} - sum:with: 'float'
+// CHECK:           CompoundStmt {{.*}} MathErrno=1
+// CHECK-NEXT:        ReturnStmt
+// CHECK-NEXT:          BinaryOperator {{.*}} 'float' '+' MathErrno=1
diff --git a/clang/test/AST/ast-dump-late-parsing.cpp b/clang/test/AST/ast-dump-late-parsing.cpp
new file mode 100644
index 000000000000..760664efc5f1
--- /dev/null
+++ b/clang/test/AST/ast-dump-late-parsing.cpp
@@ -0,0 +1,24 @@
+// RUN: %clang_cc1 -fsyntax-only -triple x86_64-pc-linux -std=c++11 -fcxx-exceptions -fdelayed-template-parsing -ast-dump %s \
+// RUN: | FileCheck %s
+
+#pragma STDC FENV_ROUND FE_DOWNWARD
+#pragma float_control(precise, off)
+
+template <typename T>
+__attribute__((optnone)) T func_22(T x, T y) {
+  return x + y;
+}
+
+// CHECK-LABEL: FunctionTemplateDecl {{.*}} func_22
+// CHECK:         FunctionDecl {{.*}} func_22 'T (T, T)'
+// CHECK:           CompoundStmt {{.*}} ConstRoundingMode=downward MathErrno=1
+// CHECK:             ReturnStmt
+// CHECK:               BinaryOperator {{.*}} '+' ConstRoundingMode=downward MathErrno=1
+// CHECK:         FunctionDecl {{.*}} func_22 'float (float, float)'
+// CHECK:           CompoundStmt {{.*}} ConstRoundingMode=downward MathErrno=1
+// CHECK:             ReturnStmt
+// CHECK:               BinaryOperator {{.*}} 'float' '+' ConstRoundingMode=downward MathErrno=1
+
+float func_23(float x, float y) {
+  return func_22(x, y);
+}
diff --git a/clang/test/AST/ast-dump-macro-json.c b/clang/test/AST/ast-dump-macro-json.c
index 96f4be6fec3d..fb9b4118b4f1 100644
--- a/clang/test/AST/ast-dump-macro-json.c
+++ b/clang/test/AST/ast-dump-macro-json.c
@@ -132,7 +132,7 @@ void BLAP(foo, __COUNTER__)(void);
 // CHECK-NEXT:   "spellingLoc": {
 // CHECK-NEXT:    "offset": {{[0-9]+}},
 // CHECK-NEXT:    "file": "<scratch space>",
-// CHECK-NEXT:    "line": 3,
+// CHECK-NEXT:    "line": 5,
 // CHECK-NEXT:    "col": 1,
 // CHECK-NEXT:    "tokLen": 4
 // CHECK-NEXT:   },
@@ -169,7 +169,7 @@ void BLAP(foo, __COUNTER__)(void);
 // CHECK-NEXT:   "spellingLoc": {
 // CHECK-NEXT:    "offset": {{[0-9]+}},
 // CHECK-NEXT:    "file": "<scratch space>",
-// CHECK-NEXT:    "line": 5,
+// CHECK-NEXT:    "line": 7,
 // CHECK-NEXT:    "col": 1,
 // CHECK-NEXT:    "tokLen": 4
 // CHECK-NEXT:   },
diff --git a/clang/test/AST/ast-dump-pragma-unroll.cpp b/clang/test/AST/ast-dump-pragma-unroll.cpp
new file mode 100644
index 000000000000..f9c254b803ff
--- /dev/null
+++ b/clang/test/AST/ast-dump-pragma-unroll.cpp
@@ -0,0 +1,31 @@
+// RUN: %clang_cc1 -triple x86_64-unknown-unknown -ast-dump %s | FileCheck %s
+
+using size_t = unsigned long long;
+
+// CHECK: LoopHintAttr {{.*}} Implicit unroll UnrollCount Numeric
+// CHECK: LoopHintAttr {{.*}} Implicit unroll UnrollCount Numeric
+// CHECK: LoopHintAttr {{.*}} Implicit unroll Unroll Disable
+// CHECK: LoopHintAttr {{.*}} Implicit unroll Unroll Disable
+template <bool Flag>
+int value_dependent(int n) {
+  constexpr int N = 100;
+  auto init = [=]() { return Flag ? n : 0UL; };
+  auto cond = [=](size_t ix) { return Flag ? ix != 0 : ix < 10; };
+  auto iter = [=](size_t ix) {
+    return Flag ? ix & ~(1ULL << __builtin_clzll(ix)) : ix + 1;
+  };
+
+#pragma unroll Flag ? 1 : N
+  for (size_t ix = init(); cond(ix); ix = iter(ix)) {
+    n *= n;
+  }
+#pragma unroll Flag ? 0 : N
+  for (size_t ix = init(); cond(ix); ix = iter(ix)) {
+    n *= n;
+  }
+  return n;
+}
+
+void test_value_dependent(int n) {
+  value_dependent<true>(n);
+}
diff --git a/clang/test/AST/ast-dump-recovery.cpp b/clang/test/AST/ast-dump-recovery.cpp
index cfb013585ad7..77527743fe85 100644
--- a/clang/test/AST/ast-dump-recovery.cpp
+++ b/clang/test/AST/ast-dump-recovery.cpp
@@ -413,6 +413,14 @@ void RecoveryExprForInvalidDecls(Unknown InvalidDecl) {
   // CHECK-NEXT: `-RecoveryExpr {{.*}} '<dependent type>'
 }
 
+void InitializerOfInvalidDecl() {
+  int ValidDecl;
+  Unkown InvalidDecl = ValidDecl;
+  // CHECK:      VarDecl {{.*}} invalid InvalidDecl
+  // CHECK-NEXT: `-RecoveryExpr {{.*}} '<dependent type>' contains-errors
+  // CHECK-NEXT:   `-DeclRefExpr {{.*}} 'int' lvalue Var {{.*}} 'ValidDecl'
+}
+
 void RecoverToAnInvalidDecl() {
   Unknown* foo; // invalid decl
   goo; // the typo was correct to the invalid foo.
diff --git a/clang/test/AST/ast-dump-template-json-win32-mangler-crash.cpp b/clang/test/AST/ast-dump-template-json-win32-mangler-crash.cpp
index cf740516db6f..5ac55d269dce 100644
--- a/clang/test/AST/ast-dump-template-json-win32-mangler-crash.cpp
+++ b/clang/test/AST/ast-dump-template-json-win32-mangler-crash.cpp
@@ -1846,6 +1846,42 @@ int main()
 // CHECK-NEXT:              "kind": "VarTemplateDecl",
 // CHECK-NEXT:              "name": "is_const_v"
 // CHECK-NEXT:             }
+// CHECK-NEXT:            ],
+// CHECK-NEXT:            "inner": [
+// CHECK-NEXT:             {
+// CHECK-NEXT:              "kind": "TemplateArgument",
+// CHECK-NEXT:              "type": {
+// CHECK-NEXT:               "qualType": "const _Ty"
+// CHECK-NEXT:              },
+// CHECK-NEXT:              "inner": [
+// CHECK-NEXT:               {
+// CHECK-NEXT:                "id": "0x{{.*}}",
+// CHECK-NEXT:                "kind": "QualType",
+// CHECK-NEXT:                "type": {
+// CHECK-NEXT:                 "qualType": "const _Ty"
+// CHECK-NEXT:                },
+// CHECK-NEXT:                "qualifiers": "const",
+// CHECK-NEXT:                "inner": [
+// CHECK-NEXT:                 {
+// CHECK-NEXT:                  "id": "0x{{.*}}",
+// CHECK-NEXT:                  "kind": "TemplateTypeParmType",
+// CHECK-NEXT:                  "type": {
+// CHECK-NEXT:                   "qualType": "_Ty"
+// CHECK-NEXT:                  },
+// CHECK-NEXT:                  "isDependent": true,
+// CHECK-NEXT:                  "isInstantiationDependent": true,
+// CHECK-NEXT:                  "depth": 0,
+// CHECK-NEXT:                  "index": 0,
+// CHECK-NEXT:                  "decl": {
+// CHECK-NEXT:                   "id": "0x{{.*}}",
+// CHECK-NEXT:                   "kind": "TemplateTypeParmDecl",
+// CHECK-NEXT:                   "name": "_Ty"
+// CHECK-NEXT:                  }
+// CHECK-NEXT:                 }
+// CHECK-NEXT:                ]
+// CHECK-NEXT:               }
+// CHECK-NEXT:              ]
+// CHECK-NEXT:             }
 // CHECK-NEXT:            ]
 // CHECK-NEXT:           }
 // CHECK-NEXT:          ]
@@ -1900,6 +1936,32 @@ int main()
 // CHECK-NEXT:              "kind": "VarTemplateDecl",
 // CHECK-NEXT:              "name": "is_reference_v"
 // CHECK-NEXT:             }
+// CHECK-NEXT:            ],
+// CHECK-NEXT:            "inner": [
+// CHECK-NEXT:             {
+// CHECK-NEXT:              "kind": "TemplateArgument",
+// CHECK-NEXT:              "type": {
+// CHECK-NEXT:               "qualType": "_Ty"
+// CHECK-NEXT:              },
+// CHECK-NEXT:              "inner": [
+// CHECK-NEXT:               {
+// CHECK-NEXT:                "id": "0x{{.*}}",
+// CHECK-NEXT:                "kind": "TemplateTypeParmType",
+// CHECK-NEXT:                "type": {
+// CHECK-NEXT:                 "qualType": "_Ty"
+// CHECK-NEXT:                },
+// CHECK-NEXT:                "isDependent": true,
+// CHECK-NEXT:                "isInstantiationDependent": true,
+// CHECK-NEXT:                "depth": 0,
+// CHECK-NEXT:                "index": 0,
+// CHECK-NEXT:                "decl": {
+// CHECK-NEXT:                 "id": "0x{{.*}}",
+// CHECK-NEXT:                 "kind": "TemplateTypeParmDecl",
+// CHECK-NEXT:                 "name": "_Ty"
+// CHECK-NEXT:                }
+// CHECK-NEXT:               }
+// CHECK-NEXT:              ]
+// CHECK-NEXT:             }
 // CHECK-NEXT:            ]
 // CHECK-NEXT:           }
 // CHECK-NEXT:          ]
@@ -2565,6 +2627,32 @@ int main()
 // CHECK-NEXT:            "kind": "VarTemplateDecl",
 // CHECK-NEXT:            "name": "is_function_v"
 // CHECK-NEXT:           }
+// CHECK-NEXT:          ],
+// CHECK-NEXT:          "inner": [
+// CHECK-NEXT:           {
+// CHECK-NEXT:            "kind": "TemplateArgument",
+// CHECK-NEXT:            "type": {
+// CHECK-NEXT:             "qualType": "_Ty1"
+// CHECK-NEXT:            },
+// CHECK-NEXT:            "inner": [
+// CHECK-NEXT:             {
+// CHECK-NEXT:              "id": "0x{{.*}}",
+// CHECK-NEXT:              "kind": "TemplateTypeParmType",
+// CHECK-NEXT:              "type": {
+// CHECK-NEXT:               "qualType": "_Ty1"
+// CHECK-NEXT:              },
+// CHECK-NEXT:              "isDependent": true,
+// CHECK-NEXT:              "isInstantiationDependent": true,
+// CHECK-NEXT:              "depth": 0,
+// CHECK-NEXT:              "index": 0,
+// CHECK-NEXT:              "decl": {
+// CHECK-NEXT:               "id": "0x{{.*}}",
+// CHECK-NEXT:               "kind": "TemplateTypeParmDecl",
+// CHECK-NEXT:               "name": "_Ty1"
+// CHECK-NEXT:              }
+// CHECK-NEXT:             }
+// CHECK-NEXT:            ]
+// CHECK-NEXT:           }
 // CHECK-NEXT:          ]
 // CHECK-NEXT:         }
 // CHECK-NEXT:        ]
diff --git a/clang/test/AST/ast-dump-templates.cpp b/clang/test/AST/ast-dump-templates.cpp
index d25ef36dd4d3..9fcafbcbcc46 100644
--- a/clang/test/AST/ast-dump-templates.cpp
+++ b/clang/test/AST/ast-dump-templates.cpp
@@ -104,3 +104,17 @@ void (*q)() = f<>;
 // CHECK1: template<> void f<0L>()
 // CHECK1: template<> void f<0U>()
 }
+
+namespace test6 {
+template <class D>
+constexpr bool C = true;
+
+template <class Key>
+void func() {
+  C<Key>;
+// DUMP:      UnresolvedLookupExpr {{.*}} '<dependent type>' lvalue (no ADL) = 'C'
+// DUMP-NEXT: `-TemplateArgument type 'Key'
+// DUMP-NEXT:   `-TemplateTypeParmType {{.*}} 'Key' dependent depth 0 index 0
+// DUMP-NEXT:     `-TemplateTypeParm {{.*}} 'Key'
+}
+}
diff --git a/clang/test/Analysis/Checkers/WebKit/call-args-regression-traverse-decl-crash.cpp b/clang/test/Analysis/Checkers/WebKit/call-args-regression-traverse-decl-crash.cpp
new file mode 100644
index 000000000000..3d8e822025f6
--- /dev/null
+++ b/clang/test/Analysis/Checkers/WebKit/call-args-regression-traverse-decl-crash.cpp
@@ -0,0 +1,7 @@
+// RUN: %clang_analyze_cc1 -analyzer-checker=alpha.webkit.UncountedCallArgsChecker -verify %s
+// expected-no-diagnostics
+
+template <class Class> struct T;
+template <template <class> class Class, class Type>
+struct T<Class<Type>>
+{ };
diff --git a/clang/test/Analysis/Checkers/WebKit/call-args.cpp b/clang/test/Analysis/Checkers/WebKit/call-args.cpp
index f2e1f9bc5a24..2a4b6bb1f106 100644
--- a/clang/test/Analysis/Checkers/WebKit/call-args.cpp
+++ b/clang/test/Analysis/Checkers/WebKit/call-args.cpp
@@ -32,7 +32,7 @@ namespace ref_counted {
   void consume_ref_counted(Ref<RefCountable>) {}
 
   void foo() {
-    consume_refcntbl(provide_ref_counted().get());
+    consume_refcntbl(provide_ref_counted().ptr());
     // no warning
   }
 }
diff --git a/clang/test/Analysis/Checkers/WebKit/mock-types.h b/clang/test/Analysis/Checkers/WebKit/mock-types.h
index aab99197dfa4..c27ea9baaf3b 100644
--- a/clang/test/Analysis/Checkers/WebKit/mock-types.h
+++ b/clang/test/Analysis/Checkers/WebKit/mock-types.h
@@ -1,24 +1,61 @@
 #ifndef mock_types_1103988513531
 #define mock_types_1103988513531
 
-template <typename T> struct Ref {
-  T *t;
+template<typename T>
+struct RawPtrTraits {
+  using StorageType = T*;
 
-  Ref() : t{} {};
-  Ref(T &t)
-    : t(t) {
-    if (t)
-      t->ref();
+  template<typename U>
+  static T* exchange(StorageType& ptr, U&& newValue)
+  {
+    StorageType oldValue = static_cast<StorageType&&>(ptr);
+    ptr = static_cast<U&&>(newValue);
+    return oldValue;
   }
-  ~Ref() {
-    if (t)
-      t->deref();
+
+  static void swap(StorageType& a, StorageType& b)
+  {
+    StorageType temp = static_cast<StorageType&&>(a);
+    a = static_cast<StorageType&&>(b);
+    b = static_cast<StorageType&&>(temp);
   }
-  T *get() { return t; }
-  T *ptr() { return t; }
-  T *operator->() { return t; }
-  operator const T &() const { return *t; }
-  operator T &() { return *t; }
+  static T* unwrap(const StorageType& ptr) { return ptr; }
+};
+
+template<typename T> struct DefaultRefDerefTraits {
+  static T* refIfNotNull(T* ptr)
+  {
+    if (ptr)
+      ptr->ref();
+    return ptr;
+  }
+
+  static T& ref(T& ref)
+  {
+    ref.ref();
+    return ref;
+  }
+
+  static void derefIfNotNull(T* ptr)
+  {
+    if (ptr)
+      ptr->deref();
+  }
+};
+
+template <typename T, typename PtrTraits = RawPtrTraits<T>, typename RefDerefTraits = DefaultRefDerefTraits<T>> struct Ref {
+  typename PtrTraits::StorageType t;
+
+  Ref() : t{} {};
+  Ref(T &t) : t(RefDerefTraits::refIfNotNull(t)) { }
+  Ref(const Ref& o) : t(RefDerefTraits::refIfNotNull(PtrTraits::unwrap(o.t))) { }
+  ~Ref() { RefDerefTraits::derefIfNotNull(PtrTraits::exchange(t, nullptr)); }
+  T &get() { return *PtrTraits::unwrap(t); }
+  T *ptr() { return PtrTraits::unwrap(t); }
+  T *operator->() { return PtrTraits::unwrap(t); }
+  operator const T &() const { return *PtrTraits::unwrap(t); }
+  operator T &() { return *PtrTraits::unwrap(t); }
+  T* leakRef() { PtrTraits::exchange(t, nullptr); }
 };
 
 template <typename T> struct RefPtr {
diff --git a/clang/test/Analysis/html_diagnostics/relevant_lines/multifile.c b/clang/test/Analysis/html_diagnostics/relevant_lines/multifile.c
index 3abffd609b5b..1998c9383d9d 100644
--- a/clang/test/Analysis/html_diagnostics/relevant_lines/multifile.c
+++ b/clang/test/Analysis/html_diagnostics/relevant_lines/multifile.c
@@ -11,4 +11,4 @@ int f(int coin) {
 // RUN: rm -rf %t.output
 // RUN: %clang_analyze_cc1 -analyze -analyzer-checker=core -analyzer-output html -o %t.output %s
 // RUN: cat %t.output/* | FileCheck %s --match-full-lines
-// CHECK: var relevant_lines = {"1": {"3": 1, "4": 1, "5": 1, "6": 1}, "3": {"3": 1, "4": 1, "5": 1, "6": 1, "7": 1}};
+// CHECK: var relevant_lines = {"1": {"3": 1, "4": 1, "5": 1, "6": 1}, "4": {"3": 1, "4": 1, "5": 1, "6": 1, "7": 1}};
diff --git a/clang/test/CXX/drs/cwg2149.cpp b/clang/test/CXX/drs/cwg2149.cpp
new file mode 100644
index 000000000000..8416e42cbd69
--- /dev/null
+++ b/clang/test/CXX/drs/cwg2149.cpp
@@ -0,0 +1,77 @@
+// RUN: %clang_cc1 -std=c++98 -triple x86_64-unknown-unknown %s -verify=expected,cxx98 -fexceptions -fcxx-exceptions -pedantic-errors -ast-dump | FileCheck %s --check-prefixes CXX98
+// RUN: %clang_cc1 -std=c++11 -triple x86_64-unknown-unknown %s -verify=expected,since-cxx11 -fexceptions -fcxx-exceptions -pedantic-errors
+// RUN: %clang_cc1 -std=c++14 -triple x86_64-unknown-unknown %s -verify=expected,since-cxx11 -fexceptions -fcxx-exceptions -pedantic-errors
+// RUN: %clang_cc1 -std=c++17 -triple x86_64-unknown-unknown %s -verify=expected,since-cxx11 -fexceptions -fcxx-exceptions -pedantic-errors
+// RUN: %clang_cc1 -std=c++20 -triple x86_64-unknown-unknown %s -verify=expected,since-cxx11 -fexceptions -fcxx-exceptions -pedantic-errors
+// RUN: %clang_cc1 -std=c++23 -triple x86_64-unknown-unknown %s -verify=expected,since-cxx11 -fexceptions -fcxx-exceptions -pedantic-errors
+// RUN: %clang_cc1 -std=c++2c -triple x86_64-unknown-unknown %s -verify=expected,since-cxx11 -fexceptions -fcxx-exceptions -pedantic-errors
+
+#if __cplusplus == 199711L
+#define static_assert(...) __extension__ _Static_assert(__VA_ARGS__)
+// cxx98-error@-1 {{variadic macros are a C99 feature}}
+#endif
+
+namespace cwg2149 { // cwg2149: 3.1
+#if __cplusplus <= 201103L
+struct X { int i, j, k; };
+#else
+struct X { int i, j, k = 42; };
+#endif
+
+template<int N> 
+void f1(const X(&)[N]); // #cwg2149-f1
+
+template<int N>
+void f2(const X(&)[N][2]); // #cwg2149-f2
+
+void f() {
+  X a[] = { 1, 2, 3, 4, 5, 6 };
+  static_assert(sizeof(a) / sizeof(X) == 2, "");
+  X b[2] = { { 1, 2, 3 }, { 4, 5, 6 } };
+  X c[][2] = { 1, 2, 3, 4, 5, 6 };
+  static_assert(sizeof(c) / sizeof(X[2]) == 1, "");
+  
+  #if __cplusplus >= 201103L
+  constexpr X ca[] = { 1, 2, 3, 4, 5, 6 };
+  constexpr X cb[2] = { { 1, 2, 3 }, { 4, 5, 6 } };
+  static_assert(ca[0].i == cb[0].i, "");
+  static_assert(ca[0].j == cb[0].j, "");
+  static_assert(ca[0].k == cb[0].k, "");
+  static_assert(ca[1].i == cb[1].i, "");
+  static_assert(ca[1].j == cb[1].j, "");
+  static_assert(ca[1].k == cb[1].k, "");
+
+  f1({ 1, 2, 3, 4, 5, 6 });
+  // since-cxx11-error@-1 {{no matching function for call to 'f1'}}
+  //   since-cxx11-note@#cwg2149-f1 {{candidate function [with N = 6] not viable: no known conversion from 'int' to 'const X' for 1st argument}}
+  f2({ 1, 2, 3, 4, 5, 6 });
+  // since-cxx11-error@-1 {{no matching function for call to 'f2'}}
+  //   since-cxx11-note@#cwg2149-f2 {{candidate function [with N = 6] not viable: no known conversion from 'int' to 'const X[2]' for 1st argument}}
+  #endif
+}
+} // namespace cwg2149
+
+// Constant evaluation is not powerful enough in 98 mode to check for equality
+// via static_assert, even with constant folding enabled.
+
+// CXX98:       VarDecl {{.+}} a 'X[2]'
+// CXX98-NEXT:  `-InitListExpr {{.+}} 'X[2]'
+// CXX98-NEXT:    |-InitListExpr {{.+}} 'X':'cwg2149::X'
+// CXX98-NEXT:    | |-IntegerLiteral {{.+}} 'int' 1
+// CXX98-NEXT:    | |-IntegerLiteral {{.+}} 'int' 2
+// CXX98-NEXT:    | `-IntegerLiteral {{.+}} 'int' 3
+// CXX98-NEXT:    `-InitListExpr {{.+}} 'X':'cwg2149::X'
+// CXX98-NEXT:      |-IntegerLiteral {{.+}} 'int' 4
+// CXX98-NEXT:      |-IntegerLiteral {{.+}} 'int' 5
+// CXX98-NEXT:      `-IntegerLiteral {{.+}} 'int' 6
+
+// CXX98:       VarDecl {{.+}} b 'X[2]'
+// CXX98-NEXT:  `-InitListExpr {{.+}} 'X[2]'
+// CXX98-NEXT:    |-InitListExpr {{.+}} 'X':'cwg2149::X'
+// CXX98-NEXT:    | |-IntegerLiteral {{.+}} 'int' 1
+// CXX98-NEXT:    | |-IntegerLiteral {{.+}} 'int' 2
+// CXX98-NEXT:    | `-IntegerLiteral {{.+}} 'int' 3
+// CXX98-NEXT:    `-InitListExpr {{.+}} 'X':'cwg2149::X'
+// CXX98-NEXT:      |-IntegerLiteral {{.+}} 'int' 4
+// CXX98-NEXT:      |-IntegerLiteral {{.+}} 'int' 5
+// CXX98-NEXT:      `-IntegerLiteral {{.+}} 'int' 6
diff --git a/clang/test/CXX/drs/cwg650.cpp b/clang/test/CXX/drs/cwg650.cpp
index dcb844095b05..33ea179986e3 100644
--- a/clang/test/CXX/drs/cwg650.cpp
+++ b/clang/test/CXX/drs/cwg650.cpp
@@ -4,7 +4,7 @@
 // RUN: %clang_cc1 -std=c++17 %s -triple x86_64-linux-gnu -emit-llvm -o - -fexceptions -fcxx-exceptions -pedantic-errors | llvm-cxxfilt -n | FileCheck %s --check-prefixes CHECK
 // RUN: %clang_cc1 -std=c++20 %s -triple x86_64-linux-gnu -emit-llvm -o - -fexceptions -fcxx-exceptions -pedantic-errors | llvm-cxxfilt -n | FileCheck %s --check-prefixes CHECK
 // RUN: %clang_cc1 -std=c++23 %s -triple x86_64-linux-gnu -emit-llvm -o - -fexceptions -fcxx-exceptions -pedantic-errors | llvm-cxxfilt -n | FileCheck %s --check-prefixes CHECK
-// RUN: %clang_cc1 -std=c++2c %s -triple x86_64-linux-gnu -emit-llvm -o - -fexceptions -fcxx-exceptions -pedantic-errors | llvm-cxxfilt -n | FileCheck %s --check-prefixes CHECK
+// We aren't testing this since C++26 because of P2748R5 "Disallow Binding a Returned Glvalue to a Temporary".
 
 #if __cplusplus == 199711L
 #define NOTHROW throw()
diff --git a/clang/test/CXX/drs/dr20xx.cpp b/clang/test/CXX/drs/dr20xx.cpp
index 291a77e0cc71..9797097acce7 100644
--- a/clang/test/CXX/drs/dr20xx.cpp
+++ b/clang/test/CXX/drs/dr20xx.cpp
@@ -90,7 +90,7 @@ namespace cwg2026 { // cwg2026: 11
   }
 }
 
-namespace cwg2049 { // cwg2049: 18 drafting P2308R1
+namespace cwg2049 { // cwg2049: 18
 #if __cplusplus >= 202302L
 template <int* x = {}> struct X {};
 X<> a;
diff --git a/clang/test/CXX/drs/dr21xx.cpp b/clang/test/CXX/drs/dr21xx.cpp
index 4fab10c279aa..082deb42e4fa 100644
--- a/clang/test/CXX/drs/dr21xx.cpp
+++ b/clang/test/CXX/drs/dr21xx.cpp
@@ -175,6 +175,8 @@ void foo() {
 }
 }
 
+// cwg2149 is in cwg2149.cpp
+
 namespace cwg2157 { // cwg2157: 11
 #if __cplusplus >= 201103L
   enum E : int;
diff --git a/clang/test/CXX/drs/dr24xx.cpp b/clang/test/CXX/drs/dr24xx.cpp
index 5ffaebda68c1..9f876cd87083 100644
--- a/clang/test/CXX/drs/dr24xx.cpp
+++ b/clang/test/CXX/drs/dr24xx.cpp
@@ -45,7 +45,7 @@ void fallthrough(int n) {
 #endif
 }
 
-namespace cwg2450 { // cwg2450: 18 review P2308R1
+namespace cwg2450 { // cwg2450: 18
 #if __cplusplus >= 202302L
 struct S {int a;};
 template <S s>
@@ -59,7 +59,7 @@ f<{.a= 0}>();
 #endif
 }
 
-namespace cwg2459 { // cwg2459: 18 drafting P2308R1
+namespace cwg2459 { // cwg2459: 18
 #if __cplusplus >= 202302L
 struct A {
   constexpr A(float) {}
diff --git a/clang/test/CXX/drs/dr25xx.cpp b/clang/test/CXX/drs/dr25xx.cpp
index 62b2a0a088cc..8bca58f44944 100644
--- a/clang/test/CXX/drs/dr25xx.cpp
+++ b/clang/test/CXX/drs/dr25xx.cpp
@@ -130,12 +130,14 @@ struct D3 : B {
 #endif
 
 #if __cplusplus >= 202302L
-namespace cwg2561 { // cwg2561: 18 review 2023-11-09
+namespace cwg2561 { // cwg2561: no tentatively ready 2024-03-18
 struct C {
     constexpr C(auto) { }
 };
 void foo() {
     constexpr auto b = [](this C) { return 1; };
+    // FIXME: closure type shouldn't have a conversion function to function
+    //        pointer, because explicit object parameter is present. 
     constexpr int (*fp)(C) = b;
     static_assert(fp(1) == 1);
     static_assert((&decltype(b)::operator())(1) == 1);
diff --git a/clang/test/CXX/drs/dr28xx.cpp b/clang/test/CXX/drs/dr28xx.cpp
index 4d9b0c76758d..be35d366bdd6 100644
--- a/clang/test/CXX/drs/dr28xx.cpp
+++ b/clang/test/CXX/drs/dr28xx.cpp
@@ -10,7 +10,15 @@
 // expected-no-diagnostics
 #endif
 
-namespace cwg2847 { // cwg2847: 19
+namespace cwg2819 { // cwg2819: 19 tentatively ready 2023-12-01
+#if __cpp_constexpr >= 202306L
+  constexpr void* p = nullptr;
+  constexpr int* q = static_cast<int*>(p);
+  static_assert(q == nullptr);
+#endif
+}
+
+namespace cwg2847 { // cwg2847: 19 review 2024-03-01
 
 #if __cplusplus >= 202002L
 
@@ -59,7 +67,7 @@ void B<int>::g() requires true;
 
 } // namespace cwg2847
 
-namespace cwg2858 { // cwg2858: 19
+namespace cwg2858 { // cwg2858: 19 tentatively ready 2024-04-05
 
 #if __cplusplus > 202302L
 
diff --git a/clang/test/CXX/expr/expr.const/p5-26.cpp b/clang/test/CXX/expr/expr.const/p5-26.cpp
index 3624b1e5a3e3..7513b11c09aa 100644
--- a/clang/test/CXX/expr/expr.const/p5-26.cpp
+++ b/clang/test/CXX/expr/expr.const/p5-26.cpp
@@ -37,3 +37,10 @@ void err() {
                                                 // cxx23-note {{cast from 'void *' is not allowed in a constant expression in C++ standards before C++2c}} \
                                                 // cxx26-note {{cast from 'void *' is not allowed in a constant expression because the pointed object type 'T' is not similar to the target type 'S'}}
 }
+
+int* p;
+constexpr int** pp = &p;
+constexpr void* vp = pp;
+constexpr auto cvp = static_cast<const int* volatile*>(vp);
+// cxx23-error@-1 {{constant expression}}
+// cxx23-note@-2 {{cast from 'void *' is not allowed in a constant expression}}
diff --git a/clang/test/CXX/stmt.stmt/stmt.return/p6.cpp b/clang/test/CXX/stmt.stmt/stmt.return/p6.cpp
new file mode 100644
index 000000000000..c192b0c8112a
--- /dev/null
+++ b/clang/test/CXX/stmt.stmt/stmt.return/p6.cpp
@@ -0,0 +1,25 @@
+// RUN: %clang_cc1 -std=c++26 -fsyntax-only -verify %s
+
+auto&& f1() {
+  return 42; // expected-error{{returning reference to local temporary object}}
+}
+const double& f2() {
+  static int x = 42;
+  return x; // expected-error{{returning reference to local temporary object}}
+}
+auto&& id(auto&& r) {
+  return static_cast<decltype(r)&&>(r);
+}
+auto&& f3() {
+  return id(42);        // OK, but probably a bug
+}
+
+void unevaluated() {
+  using a = decltype ([] () -> const int & {
+    const int &i = 0; // expected-note {{binding reference variable 'i' here}}
+    return i; // expected-error{{returning reference to local temporary object}}
+} ());
+}
+
+static_assert(__is_convertible(int, const int &));
+static_assert(__is_nothrow_convertible(int, const int &));
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vcreate.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vcreate.c
new file mode 100644
index 000000000000..8c8e1cdfb658
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vcreate.c
@@ -0,0 +1,477 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 8 x bfloat> @test_vcreate_v_bf16m1_bf16m2(
+// CHECK-RV64-SAME: <vscale x 4 x bfloat> [[V0:%.*]], <vscale x 4 x bfloat> [[V1:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x bfloat> @llvm.vector.insert.nxv8bf16.nxv4bf16(<vscale x 8 x bfloat> poison, <vscale x 4 x bfloat> [[V0]], i64 0)
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = call <vscale x 8 x bfloat> @llvm.vector.insert.nxv8bf16.nxv4bf16(<vscale x 8 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[V1]], i64 4)
+// CHECK-RV64-NEXT:    ret <vscale x 8 x bfloat> [[TMP1]]
+//
+vbfloat16m2_t test_vcreate_v_bf16m1_bf16m2(vbfloat16m1_t v0, vbfloat16m1_t v1) {
+  return __riscv_vcreate_v_bf16m1_bf16m2(v0, v1);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 16 x bfloat> @test_vcreate_v_bf16m1_bf16m4(
+// CHECK-RV64-SAME: <vscale x 4 x bfloat> [[V0:%.*]], <vscale x 4 x bfloat> [[V1:%.*]], <vscale x 4 x bfloat> [[V2:%.*]], <vscale x 4 x bfloat> [[V3:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x bfloat> @llvm.vector.insert.nxv16bf16.nxv4bf16(<vscale x 16 x bfloat> poison, <vscale x 4 x bfloat> [[V0]], i64 0)
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = call <vscale x 16 x bfloat> @llvm.vector.insert.nxv16bf16.nxv4bf16(<vscale x 16 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[V1]], i64 4)
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call <vscale x 16 x bfloat> @llvm.vector.insert.nxv16bf16.nxv4bf16(<vscale x 16 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[V2]], i64 8)
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call <vscale x 16 x bfloat> @llvm.vector.insert.nxv16bf16.nxv4bf16(<vscale x 16 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[V3]], i64 12)
+// CHECK-RV64-NEXT:    ret <vscale x 16 x bfloat> [[TMP3]]
+//
+vbfloat16m4_t test_vcreate_v_bf16m1_bf16m4(vbfloat16m1_t v0, vbfloat16m1_t v1,
+                                           vbfloat16m1_t v2, vbfloat16m1_t v3) {
+  return __riscv_vcreate_v_bf16m1_bf16m4(v0, v1, v2, v3);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 32 x bfloat> @test_vcreate_v_bf16m1_bf16m8(
+// CHECK-RV64-SAME: <vscale x 4 x bfloat> [[V0:%.*]], <vscale x 4 x bfloat> [[V1:%.*]], <vscale x 4 x bfloat> [[V2:%.*]], <vscale x 4 x bfloat> [[V3:%.*]], <vscale x 4 x bfloat> [[V4:%.*]], <vscale x 4 x bfloat> [[V5:%.*]], <vscale x 4 x bfloat> [[V6:%.*]], <vscale x 4 x bfloat> [[V7:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x bfloat> @llvm.vector.insert.nxv32bf16.nxv4bf16(<vscale x 32 x bfloat> poison, <vscale x 4 x bfloat> [[V0]], i64 0)
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = call <vscale x 32 x bfloat> @llvm.vector.insert.nxv32bf16.nxv4bf16(<vscale x 32 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[V1]], i64 4)
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call <vscale x 32 x bfloat> @llvm.vector.insert.nxv32bf16.nxv4bf16(<vscale x 32 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[V2]], i64 8)
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call <vscale x 32 x bfloat> @llvm.vector.insert.nxv32bf16.nxv4bf16(<vscale x 32 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[V3]], i64 12)
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call <vscale x 32 x bfloat> @llvm.vector.insert.nxv32bf16.nxv4bf16(<vscale x 32 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[V4]], i64 16)
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call <vscale x 32 x bfloat> @llvm.vector.insert.nxv32bf16.nxv4bf16(<vscale x 32 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[V5]], i64 20)
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call <vscale x 32 x bfloat> @llvm.vector.insert.nxv32bf16.nxv4bf16(<vscale x 32 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[V6]], i64 24)
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call <vscale x 32 x bfloat> @llvm.vector.insert.nxv32bf16.nxv4bf16(<vscale x 32 x bfloat> [[TMP6]], <vscale x 4 x bfloat> [[V7]], i64 28)
+// CHECK-RV64-NEXT:    ret <vscale x 32 x bfloat> [[TMP7]]
+//
+vbfloat16m8_t test_vcreate_v_bf16m1_bf16m8(vbfloat16m1_t v0, vbfloat16m1_t v1,
+                                           vbfloat16m1_t v2, vbfloat16m1_t v3,
+                                           vbfloat16m1_t v4, vbfloat16m1_t v5,
+                                           vbfloat16m1_t v6, vbfloat16m1_t v7) {
+  return __riscv_vcreate_v_bf16m1_bf16m8(v0, v1, v2, v3, v4, v5, v6, v7);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 16 x bfloat> @test_vcreate_v_bf16m2_bf16m4(
+// CHECK-RV64-SAME: <vscale x 8 x bfloat> [[V0:%.*]], <vscale x 8 x bfloat> [[V1:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x bfloat> @llvm.vector.insert.nxv16bf16.nxv8bf16(<vscale x 16 x bfloat> poison, <vscale x 8 x bfloat> [[V0]], i64 0)
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = call <vscale x 16 x bfloat> @llvm.vector.insert.nxv16bf16.nxv8bf16(<vscale x 16 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[V1]], i64 8)
+// CHECK-RV64-NEXT:    ret <vscale x 16 x bfloat> [[TMP1]]
+//
+vbfloat16m4_t test_vcreate_v_bf16m2_bf16m4(vbfloat16m2_t v0, vbfloat16m2_t v1) {
+  return __riscv_vcreate_v_bf16m2_bf16m4(v0, v1);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 32 x bfloat> @test_vcreate_v_bf16m2_bf16m8(
+// CHECK-RV64-SAME: <vscale x 8 x bfloat> [[V0:%.*]], <vscale x 8 x bfloat> [[V1:%.*]], <vscale x 8 x bfloat> [[V2:%.*]], <vscale x 8 x bfloat> [[V3:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x bfloat> @llvm.vector.insert.nxv32bf16.nxv8bf16(<vscale x 32 x bfloat> poison, <vscale x 8 x bfloat> [[V0]], i64 0)
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = call <vscale x 32 x bfloat> @llvm.vector.insert.nxv32bf16.nxv8bf16(<vscale x 32 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[V1]], i64 8)
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call <vscale x 32 x bfloat> @llvm.vector.insert.nxv32bf16.nxv8bf16(<vscale x 32 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[V2]], i64 16)
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call <vscale x 32 x bfloat> @llvm.vector.insert.nxv32bf16.nxv8bf16(<vscale x 32 x bfloat> [[TMP2]], <vscale x 8 x bfloat> [[V3]], i64 24)
+// CHECK-RV64-NEXT:    ret <vscale x 32 x bfloat> [[TMP3]]
+//
+vbfloat16m8_t test_vcreate_v_bf16m2_bf16m8(vbfloat16m2_t v0, vbfloat16m2_t v1,
+                                           vbfloat16m2_t v2, vbfloat16m2_t v3) {
+  return __riscv_vcreate_v_bf16m2_bf16m8(v0, v1, v2, v3);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 32 x bfloat> @test_vcreate_v_bf16m4_bf16m8(
+// CHECK-RV64-SAME: <vscale x 16 x bfloat> [[V0:%.*]], <vscale x 16 x bfloat> [[V1:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x bfloat> @llvm.vector.insert.nxv32bf16.nxv16bf16(<vscale x 32 x bfloat> poison, <vscale x 16 x bfloat> [[V0]], i64 0)
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = call <vscale x 32 x bfloat> @llvm.vector.insert.nxv32bf16.nxv16bf16(<vscale x 32 x bfloat> [[TMP0]], <vscale x 16 x bfloat> [[V1]], i64 16)
+// CHECK-RV64-NEXT:    ret <vscale x 32 x bfloat> [[TMP1]]
+//
+vbfloat16m8_t test_vcreate_v_bf16m4_bf16m8(vbfloat16m4_t v0, vbfloat16m4_t v1) {
+  return __riscv_vcreate_v_bf16m4_bf16m8(v0, v1);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vcreate_v_bf16mf4x2(
+// CHECK-RV64-SAME: <vscale x 1 x bfloat> [[V0:%.*]], <vscale x 1 x bfloat> [[V1:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } poison, <vscale x 1 x bfloat> [[V0]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]], <vscale x 1 x bfloat> [[V1]], 1
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP1]]
+//
+vbfloat16mf4x2_t test_vcreate_v_bf16mf4x2(vbfloat16mf4_t v0,
+                                          vbfloat16mf4_t v1) {
+  return __riscv_vcreate_v_bf16mf4x2(v0, v1);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vcreate_v_bf16mf4x3(
+// CHECK-RV64-SAME: <vscale x 1 x bfloat> [[V0:%.*]], <vscale x 1 x bfloat> [[V1:%.*]], <vscale x 1 x bfloat> [[V2:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } poison, <vscale x 1 x bfloat> [[V0]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]], <vscale x 1 x bfloat> [[V1]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP1]], <vscale x 1 x bfloat> [[V2]], 2
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP2]]
+//
+vbfloat16mf4x3_t test_vcreate_v_bf16mf4x3(vbfloat16mf4_t v0, vbfloat16mf4_t v1,
+                                          vbfloat16mf4_t v2) {
+  return __riscv_vcreate_v_bf16mf4x3(v0, v1, v2);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vcreate_v_bf16mf4x4(
+// CHECK-RV64-SAME: <vscale x 1 x bfloat> [[V0:%.*]], <vscale x 1 x bfloat> [[V1:%.*]], <vscale x 1 x bfloat> [[V2:%.*]], <vscale x 1 x bfloat> [[V3:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } poison, <vscale x 1 x bfloat> [[V0]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]], <vscale x 1 x bfloat> [[V1]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP1]], <vscale x 1 x bfloat> [[V2]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP2]], <vscale x 1 x bfloat> [[V3]], 3
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP3]]
+//
+vbfloat16mf4x4_t test_vcreate_v_bf16mf4x4(vbfloat16mf4_t v0, vbfloat16mf4_t v1,
+                                          vbfloat16mf4_t v2,
+                                          vbfloat16mf4_t v3) {
+  return __riscv_vcreate_v_bf16mf4x4(v0, v1, v2, v3);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vcreate_v_bf16mf4x5(
+// CHECK-RV64-SAME: <vscale x 1 x bfloat> [[V0:%.*]], <vscale x 1 x bfloat> [[V1:%.*]], <vscale x 1 x bfloat> [[V2:%.*]], <vscale x 1 x bfloat> [[V3:%.*]], <vscale x 1 x bfloat> [[V4:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } poison, <vscale x 1 x bfloat> [[V0]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]], <vscale x 1 x bfloat> [[V1]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP1]], <vscale x 1 x bfloat> [[V2]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP2]], <vscale x 1 x bfloat> [[V3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP3]], <vscale x 1 x bfloat> [[V4]], 4
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP4]]
+//
+vbfloat16mf4x5_t test_vcreate_v_bf16mf4x5(vbfloat16mf4_t v0, vbfloat16mf4_t v1,
+                                          vbfloat16mf4_t v2, vbfloat16mf4_t v3,
+                                          vbfloat16mf4_t v4) {
+  return __riscv_vcreate_v_bf16mf4x5(v0, v1, v2, v3, v4);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vcreate_v_bf16mf4x6(
+// CHECK-RV64-SAME: <vscale x 1 x bfloat> [[V0:%.*]], <vscale x 1 x bfloat> [[V1:%.*]], <vscale x 1 x bfloat> [[V2:%.*]], <vscale x 1 x bfloat> [[V3:%.*]], <vscale x 1 x bfloat> [[V4:%.*]], <vscale x 1 x bfloat> [[V5:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } poison, <vscale x 1 x bfloat> [[V0]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]], <vscale x 1 x bfloat> [[V1]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP1]], <vscale x 1 x bfloat> [[V2]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP2]], <vscale x 1 x bfloat> [[V3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP3]], <vscale x 1 x bfloat> [[V4]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP4]], <vscale x 1 x bfloat> [[V5]], 5
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP5]]
+//
+vbfloat16mf4x6_t test_vcreate_v_bf16mf4x6(vbfloat16mf4_t v0, vbfloat16mf4_t v1,
+                                          vbfloat16mf4_t v2, vbfloat16mf4_t v3,
+                                          vbfloat16mf4_t v4,
+                                          vbfloat16mf4_t v5) {
+  return __riscv_vcreate_v_bf16mf4x6(v0, v1, v2, v3, v4, v5);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vcreate_v_bf16mf4x7(
+// CHECK-RV64-SAME: <vscale x 1 x bfloat> [[V0:%.*]], <vscale x 1 x bfloat> [[V1:%.*]], <vscale x 1 x bfloat> [[V2:%.*]], <vscale x 1 x bfloat> [[V3:%.*]], <vscale x 1 x bfloat> [[V4:%.*]], <vscale x 1 x bfloat> [[V5:%.*]], <vscale x 1 x bfloat> [[V6:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } poison, <vscale x 1 x bfloat> [[V0]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]], <vscale x 1 x bfloat> [[V1]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP1]], <vscale x 1 x bfloat> [[V2]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP2]], <vscale x 1 x bfloat> [[V3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP3]], <vscale x 1 x bfloat> [[V4]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP4]], <vscale x 1 x bfloat> [[V5]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP5]], <vscale x 1 x bfloat> [[V6]], 6
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP6]]
+//
+vbfloat16mf4x7_t test_vcreate_v_bf16mf4x7(vbfloat16mf4_t v0, vbfloat16mf4_t v1,
+                                          vbfloat16mf4_t v2, vbfloat16mf4_t v3,
+                                          vbfloat16mf4_t v4, vbfloat16mf4_t v5,
+                                          vbfloat16mf4_t v6) {
+  return __riscv_vcreate_v_bf16mf4x7(v0, v1, v2, v3, v4, v5, v6);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vcreate_v_bf16mf4x8(
+// CHECK-RV64-SAME: <vscale x 1 x bfloat> [[V0:%.*]], <vscale x 1 x bfloat> [[V1:%.*]], <vscale x 1 x bfloat> [[V2:%.*]], <vscale x 1 x bfloat> [[V3:%.*]], <vscale x 1 x bfloat> [[V4:%.*]], <vscale x 1 x bfloat> [[V5:%.*]], <vscale x 1 x bfloat> [[V6:%.*]], <vscale x 1 x bfloat> [[V7:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } poison, <vscale x 1 x bfloat> [[V0]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]], <vscale x 1 x bfloat> [[V1]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP1]], <vscale x 1 x bfloat> [[V2]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP2]], <vscale x 1 x bfloat> [[V3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP3]], <vscale x 1 x bfloat> [[V4]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP4]], <vscale x 1 x bfloat> [[V5]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP5]], <vscale x 1 x bfloat> [[V6]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP6]], <vscale x 1 x bfloat> [[V7]], 7
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP7]]
+//
+vbfloat16mf4x8_t test_vcreate_v_bf16mf4x8(vbfloat16mf4_t v0, vbfloat16mf4_t v1,
+                                          vbfloat16mf4_t v2, vbfloat16mf4_t v3,
+                                          vbfloat16mf4_t v4, vbfloat16mf4_t v5,
+                                          vbfloat16mf4_t v6,
+                                          vbfloat16mf4_t v7) {
+  return __riscv_vcreate_v_bf16mf4x8(v0, v1, v2, v3, v4, v5, v6, v7);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vcreate_v_bf16mf2x2(
+// CHECK-RV64-SAME: <vscale x 2 x bfloat> [[V0:%.*]], <vscale x 2 x bfloat> [[V1:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } poison, <vscale x 2 x bfloat> [[V0]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]], <vscale x 2 x bfloat> [[V1]], 1
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP1]]
+//
+vbfloat16mf2x2_t test_vcreate_v_bf16mf2x2(vbfloat16mf2_t v0,
+                                          vbfloat16mf2_t v1) {
+  return __riscv_vcreate_v_bf16mf2x2(v0, v1);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vcreate_v_bf16mf2x3(
+// CHECK-RV64-SAME: <vscale x 2 x bfloat> [[V0:%.*]], <vscale x 2 x bfloat> [[V1:%.*]], <vscale x 2 x bfloat> [[V2:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } poison, <vscale x 2 x bfloat> [[V0]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]], <vscale x 2 x bfloat> [[V1]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP1]], <vscale x 2 x bfloat> [[V2]], 2
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP2]]
+//
+vbfloat16mf2x3_t test_vcreate_v_bf16mf2x3(vbfloat16mf2_t v0, vbfloat16mf2_t v1,
+                                          vbfloat16mf2_t v2) {
+  return __riscv_vcreate_v_bf16mf2x3(v0, v1, v2);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vcreate_v_bf16mf2x4(
+// CHECK-RV64-SAME: <vscale x 2 x bfloat> [[V0:%.*]], <vscale x 2 x bfloat> [[V1:%.*]], <vscale x 2 x bfloat> [[V2:%.*]], <vscale x 2 x bfloat> [[V3:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } poison, <vscale x 2 x bfloat> [[V0]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]], <vscale x 2 x bfloat> [[V1]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP1]], <vscale x 2 x bfloat> [[V2]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP2]], <vscale x 2 x bfloat> [[V3]], 3
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP3]]
+//
+vbfloat16mf2x4_t test_vcreate_v_bf16mf2x4(vbfloat16mf2_t v0, vbfloat16mf2_t v1,
+                                          vbfloat16mf2_t v2,
+                                          vbfloat16mf2_t v3) {
+  return __riscv_vcreate_v_bf16mf2x4(v0, v1, v2, v3);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vcreate_v_bf16mf2x5(
+// CHECK-RV64-SAME: <vscale x 2 x bfloat> [[V0:%.*]], <vscale x 2 x bfloat> [[V1:%.*]], <vscale x 2 x bfloat> [[V2:%.*]], <vscale x 2 x bfloat> [[V3:%.*]], <vscale x 2 x bfloat> [[V4:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } poison, <vscale x 2 x bfloat> [[V0]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]], <vscale x 2 x bfloat> [[V1]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP1]], <vscale x 2 x bfloat> [[V2]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP2]], <vscale x 2 x bfloat> [[V3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP3]], <vscale x 2 x bfloat> [[V4]], 4
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP4]]
+//
+vbfloat16mf2x5_t test_vcreate_v_bf16mf2x5(vbfloat16mf2_t v0, vbfloat16mf2_t v1,
+                                          vbfloat16mf2_t v2, vbfloat16mf2_t v3,
+                                          vbfloat16mf2_t v4) {
+  return __riscv_vcreate_v_bf16mf2x5(v0, v1, v2, v3, v4);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vcreate_v_bf16mf2x6(
+// CHECK-RV64-SAME: <vscale x 2 x bfloat> [[V0:%.*]], <vscale x 2 x bfloat> [[V1:%.*]], <vscale x 2 x bfloat> [[V2:%.*]], <vscale x 2 x bfloat> [[V3:%.*]], <vscale x 2 x bfloat> [[V4:%.*]], <vscale x 2 x bfloat> [[V5:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } poison, <vscale x 2 x bfloat> [[V0]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]], <vscale x 2 x bfloat> [[V1]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP1]], <vscale x 2 x bfloat> [[V2]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP2]], <vscale x 2 x bfloat> [[V3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP3]], <vscale x 2 x bfloat> [[V4]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP4]], <vscale x 2 x bfloat> [[V5]], 5
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP5]]
+//
+vbfloat16mf2x6_t test_vcreate_v_bf16mf2x6(vbfloat16mf2_t v0, vbfloat16mf2_t v1,
+                                          vbfloat16mf2_t v2, vbfloat16mf2_t v3,
+                                          vbfloat16mf2_t v4,
+                                          vbfloat16mf2_t v5) {
+  return __riscv_vcreate_v_bf16mf2x6(v0, v1, v2, v3, v4, v5);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vcreate_v_bf16mf2x7(
+// CHECK-RV64-SAME: <vscale x 2 x bfloat> [[V0:%.*]], <vscale x 2 x bfloat> [[V1:%.*]], <vscale x 2 x bfloat> [[V2:%.*]], <vscale x 2 x bfloat> [[V3:%.*]], <vscale x 2 x bfloat> [[V4:%.*]], <vscale x 2 x bfloat> [[V5:%.*]], <vscale x 2 x bfloat> [[V6:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } poison, <vscale x 2 x bfloat> [[V0]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]], <vscale x 2 x bfloat> [[V1]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP1]], <vscale x 2 x bfloat> [[V2]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP2]], <vscale x 2 x bfloat> [[V3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP3]], <vscale x 2 x bfloat> [[V4]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP4]], <vscale x 2 x bfloat> [[V5]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP5]], <vscale x 2 x bfloat> [[V6]], 6
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP6]]
+//
+vbfloat16mf2x7_t test_vcreate_v_bf16mf2x7(vbfloat16mf2_t v0, vbfloat16mf2_t v1,
+                                          vbfloat16mf2_t v2, vbfloat16mf2_t v3,
+                                          vbfloat16mf2_t v4, vbfloat16mf2_t v5,
+                                          vbfloat16mf2_t v6) {
+  return __riscv_vcreate_v_bf16mf2x7(v0, v1, v2, v3, v4, v5, v6);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vcreate_v_bf16mf2x8(
+// CHECK-RV64-SAME: <vscale x 2 x bfloat> [[V0:%.*]], <vscale x 2 x bfloat> [[V1:%.*]], <vscale x 2 x bfloat> [[V2:%.*]], <vscale x 2 x bfloat> [[V3:%.*]], <vscale x 2 x bfloat> [[V4:%.*]], <vscale x 2 x bfloat> [[V5:%.*]], <vscale x 2 x bfloat> [[V6:%.*]], <vscale x 2 x bfloat> [[V7:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } poison, <vscale x 2 x bfloat> [[V0]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]], <vscale x 2 x bfloat> [[V1]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP1]], <vscale x 2 x bfloat> [[V2]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP2]], <vscale x 2 x bfloat> [[V3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP3]], <vscale x 2 x bfloat> [[V4]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP4]], <vscale x 2 x bfloat> [[V5]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP5]], <vscale x 2 x bfloat> [[V6]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP6]], <vscale x 2 x bfloat> [[V7]], 7
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP7]]
+//
+vbfloat16mf2x8_t test_vcreate_v_bf16mf2x8(vbfloat16mf2_t v0, vbfloat16mf2_t v1,
+                                          vbfloat16mf2_t v2, vbfloat16mf2_t v3,
+                                          vbfloat16mf2_t v4, vbfloat16mf2_t v5,
+                                          vbfloat16mf2_t v6,
+                                          vbfloat16mf2_t v7) {
+  return __riscv_vcreate_v_bf16mf2x8(v0, v1, v2, v3, v4, v5, v6, v7);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vcreate_v_bf16m1x2(
+// CHECK-RV64-SAME: <vscale x 4 x bfloat> [[V0:%.*]], <vscale x 4 x bfloat> [[V1:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } poison, <vscale x 4 x bfloat> [[V0]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]], <vscale x 4 x bfloat> [[V1]], 1
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP1]]
+//
+vbfloat16m1x2_t test_vcreate_v_bf16m1x2(vbfloat16m1_t v0, vbfloat16m1_t v1) {
+  return __riscv_vcreate_v_bf16m1x2(v0, v1);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vcreate_v_bf16m1x3(
+// CHECK-RV64-SAME: <vscale x 4 x bfloat> [[V0:%.*]], <vscale x 4 x bfloat> [[V1:%.*]], <vscale x 4 x bfloat> [[V2:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } poison, <vscale x 4 x bfloat> [[V0]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]], <vscale x 4 x bfloat> [[V1]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP1]], <vscale x 4 x bfloat> [[V2]], 2
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP2]]
+//
+vbfloat16m1x3_t test_vcreate_v_bf16m1x3(vbfloat16m1_t v0, vbfloat16m1_t v1,
+                                        vbfloat16m1_t v2) {
+  return __riscv_vcreate_v_bf16m1x3(v0, v1, v2);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vcreate_v_bf16m1x4(
+// CHECK-RV64-SAME: <vscale x 4 x bfloat> [[V0:%.*]], <vscale x 4 x bfloat> [[V1:%.*]], <vscale x 4 x bfloat> [[V2:%.*]], <vscale x 4 x bfloat> [[V3:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } poison, <vscale x 4 x bfloat> [[V0]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]], <vscale x 4 x bfloat> [[V1]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP1]], <vscale x 4 x bfloat> [[V2]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP2]], <vscale x 4 x bfloat> [[V3]], 3
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP3]]
+//
+vbfloat16m1x4_t test_vcreate_v_bf16m1x4(vbfloat16m1_t v0, vbfloat16m1_t v1,
+                                        vbfloat16m1_t v2, vbfloat16m1_t v3) {
+  return __riscv_vcreate_v_bf16m1x4(v0, v1, v2, v3);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vcreate_v_bf16m1x5(
+// CHECK-RV64-SAME: <vscale x 4 x bfloat> [[V0:%.*]], <vscale x 4 x bfloat> [[V1:%.*]], <vscale x 4 x bfloat> [[V2:%.*]], <vscale x 4 x bfloat> [[V3:%.*]], <vscale x 4 x bfloat> [[V4:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } poison, <vscale x 4 x bfloat> [[V0]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]], <vscale x 4 x bfloat> [[V1]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP1]], <vscale x 4 x bfloat> [[V2]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP2]], <vscale x 4 x bfloat> [[V3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP3]], <vscale x 4 x bfloat> [[V4]], 4
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP4]]
+//
+vbfloat16m1x5_t test_vcreate_v_bf16m1x5(vbfloat16m1_t v0, vbfloat16m1_t v1,
+                                        vbfloat16m1_t v2, vbfloat16m1_t v3,
+                                        vbfloat16m1_t v4) {
+  return __riscv_vcreate_v_bf16m1x5(v0, v1, v2, v3, v4);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vcreate_v_bf16m1x6(
+// CHECK-RV64-SAME: <vscale x 4 x bfloat> [[V0:%.*]], <vscale x 4 x bfloat> [[V1:%.*]], <vscale x 4 x bfloat> [[V2:%.*]], <vscale x 4 x bfloat> [[V3:%.*]], <vscale x 4 x bfloat> [[V4:%.*]], <vscale x 4 x bfloat> [[V5:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } poison, <vscale x 4 x bfloat> [[V0]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]], <vscale x 4 x bfloat> [[V1]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP1]], <vscale x 4 x bfloat> [[V2]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP2]], <vscale x 4 x bfloat> [[V3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP3]], <vscale x 4 x bfloat> [[V4]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP4]], <vscale x 4 x bfloat> [[V5]], 5
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP5]]
+//
+vbfloat16m1x6_t test_vcreate_v_bf16m1x6(vbfloat16m1_t v0, vbfloat16m1_t v1,
+                                        vbfloat16m1_t v2, vbfloat16m1_t v3,
+                                        vbfloat16m1_t v4, vbfloat16m1_t v5) {
+  return __riscv_vcreate_v_bf16m1x6(v0, v1, v2, v3, v4, v5);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vcreate_v_bf16m1x7(
+// CHECK-RV64-SAME: <vscale x 4 x bfloat> [[V0:%.*]], <vscale x 4 x bfloat> [[V1:%.*]], <vscale x 4 x bfloat> [[V2:%.*]], <vscale x 4 x bfloat> [[V3:%.*]], <vscale x 4 x bfloat> [[V4:%.*]], <vscale x 4 x bfloat> [[V5:%.*]], <vscale x 4 x bfloat> [[V6:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } poison, <vscale x 4 x bfloat> [[V0]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]], <vscale x 4 x bfloat> [[V1]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP1]], <vscale x 4 x bfloat> [[V2]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP2]], <vscale x 4 x bfloat> [[V3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP3]], <vscale x 4 x bfloat> [[V4]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP4]], <vscale x 4 x bfloat> [[V5]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP5]], <vscale x 4 x bfloat> [[V6]], 6
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP6]]
+//
+vbfloat16m1x7_t test_vcreate_v_bf16m1x7(vbfloat16m1_t v0, vbfloat16m1_t v1,
+                                        vbfloat16m1_t v2, vbfloat16m1_t v3,
+                                        vbfloat16m1_t v4, vbfloat16m1_t v5,
+                                        vbfloat16m1_t v6) {
+  return __riscv_vcreate_v_bf16m1x7(v0, v1, v2, v3, v4, v5, v6);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vcreate_v_bf16m1x8(
+// CHECK-RV64-SAME: <vscale x 4 x bfloat> [[V0:%.*]], <vscale x 4 x bfloat> [[V1:%.*]], <vscale x 4 x bfloat> [[V2:%.*]], <vscale x 4 x bfloat> [[V3:%.*]], <vscale x 4 x bfloat> [[V4:%.*]], <vscale x 4 x bfloat> [[V5:%.*]], <vscale x 4 x bfloat> [[V6:%.*]], <vscale x 4 x bfloat> [[V7:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } poison, <vscale x 4 x bfloat> [[V0]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]], <vscale x 4 x bfloat> [[V1]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP1]], <vscale x 4 x bfloat> [[V2]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP2]], <vscale x 4 x bfloat> [[V3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP3]], <vscale x 4 x bfloat> [[V4]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP4]], <vscale x 4 x bfloat> [[V5]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP5]], <vscale x 4 x bfloat> [[V6]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP6]], <vscale x 4 x bfloat> [[V7]], 7
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP7]]
+//
+vbfloat16m1x8_t test_vcreate_v_bf16m1x8(vbfloat16m1_t v0, vbfloat16m1_t v1,
+                                        vbfloat16m1_t v2, vbfloat16m1_t v3,
+                                        vbfloat16m1_t v4, vbfloat16m1_t v5,
+                                        vbfloat16m1_t v6, vbfloat16m1_t v7) {
+  return __riscv_vcreate_v_bf16m1x8(v0, v1, v2, v3, v4, v5, v6, v7);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vcreate_v_bf16m2x2(
+// CHECK-RV64-SAME: <vscale x 8 x bfloat> [[V0:%.*]], <vscale x 8 x bfloat> [[V1:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } poison, <vscale x 8 x bfloat> [[V0]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP0]], <vscale x 8 x bfloat> [[V1]], 1
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP1]]
+//
+vbfloat16m2x2_t test_vcreate_v_bf16m2x2(vbfloat16m2_t v0, vbfloat16m2_t v1) {
+  return __riscv_vcreate_v_bf16m2x2(v0, v1);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vcreate_v_bf16m2x3(
+// CHECK-RV64-SAME: <vscale x 8 x bfloat> [[V0:%.*]], <vscale x 8 x bfloat> [[V1:%.*]], <vscale x 8 x bfloat> [[V2:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } poison, <vscale x 8 x bfloat> [[V0]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP0]], <vscale x 8 x bfloat> [[V1]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP1]], <vscale x 8 x bfloat> [[V2]], 2
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP2]]
+//
+vbfloat16m2x3_t test_vcreate_v_bf16m2x3(vbfloat16m2_t v0, vbfloat16m2_t v1,
+                                        vbfloat16m2_t v2) {
+  return __riscv_vcreate_v_bf16m2x3(v0, v1, v2);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vcreate_v_bf16m2x4(
+// CHECK-RV64-SAME: <vscale x 8 x bfloat> [[V0:%.*]], <vscale x 8 x bfloat> [[V1:%.*]], <vscale x 8 x bfloat> [[V2:%.*]], <vscale x 8 x bfloat> [[V3:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } poison, <vscale x 8 x bfloat> [[V0]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP0]], <vscale x 8 x bfloat> [[V1]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP1]], <vscale x 8 x bfloat> [[V2]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP2]], <vscale x 8 x bfloat> [[V3]], 3
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP3]]
+//
+vbfloat16m2x4_t test_vcreate_v_bf16m2x4(vbfloat16m2_t v0, vbfloat16m2_t v1,
+                                        vbfloat16m2_t v2, vbfloat16m2_t v3) {
+  return __riscv_vcreate_v_bf16m2x4(v0, v1, v2, v3);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @test_vcreate_v_bf16m4x2(
+// CHECK-RV64-SAME: <vscale x 16 x bfloat> [[V0:%.*]], <vscale x 16 x bfloat> [[V1:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = insertvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } poison, <vscale x 16 x bfloat> [[V0]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = insertvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[TMP0]], <vscale x 16 x bfloat> [[V1]], 1
+// CHECK-RV64-NEXT:    ret { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[TMP1]]
+//
+vbfloat16m4x2_t test_vcreate_v_bf16m4x2(vbfloat16m4_t v0, vbfloat16m4_t v1) {
+  return __riscv_vcreate_v_bf16m4x2(v0, v1);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vget.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vget.c
new file mode 100644
index 000000000000..11a385dbe5c9
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vget.c
@@ -0,0 +1,333 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 4 x bfloat> @test_vget_v_bf16m2_bf16m1(
+// CHECK-RV64-SAME: <vscale x 8 x bfloat> [[SRC:%.*]], i64 noundef [[INDEX:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x bfloat> @llvm.vector.extract.nxv4bf16.nxv8bf16(<vscale x 8 x bfloat> [[SRC]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 4 x bfloat> [[TMP0]]
+//
+vbfloat16m1_t test_vget_v_bf16m2_bf16m1(vbfloat16m2_t src, size_t index) {
+  return __riscv_vget_v_bf16m2_bf16m1(src, 0);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 4 x bfloat> @test_vget_v_bf16m4_bf16m1(
+// CHECK-RV64-SAME: <vscale x 16 x bfloat> [[SRC:%.*]], i64 noundef [[INDEX:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x bfloat> @llvm.vector.extract.nxv4bf16.nxv16bf16(<vscale x 16 x bfloat> [[SRC]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 4 x bfloat> [[TMP0]]
+//
+vbfloat16m1_t test_vget_v_bf16m4_bf16m1(vbfloat16m4_t src, size_t index) {
+  return __riscv_vget_v_bf16m4_bf16m1(src, 0);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 4 x bfloat> @test_vget_v_bf16m8_bf16m1(
+// CHECK-RV64-SAME: <vscale x 32 x bfloat> [[SRC:%.*]], i64 noundef [[INDEX:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x bfloat> @llvm.vector.extract.nxv4bf16.nxv32bf16(<vscale x 32 x bfloat> [[SRC]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 4 x bfloat> [[TMP0]]
+//
+vbfloat16m1_t test_vget_v_bf16m8_bf16m1(vbfloat16m8_t src, size_t index) {
+  return __riscv_vget_v_bf16m8_bf16m1(src, 0);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 8 x bfloat> @test_vget_v_bf16m4_bf16m2(
+// CHECK-RV64-SAME: <vscale x 16 x bfloat> [[SRC:%.*]], i64 noundef [[INDEX:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x bfloat> @llvm.vector.extract.nxv8bf16.nxv16bf16(<vscale x 16 x bfloat> [[SRC]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 8 x bfloat> [[TMP0]]
+//
+vbfloat16m2_t test_vget_v_bf16m4_bf16m2(vbfloat16m4_t src, size_t index) {
+  return __riscv_vget_v_bf16m4_bf16m2(src, 0);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 8 x bfloat> @test_vget_v_bf16m8_bf16m2(
+// CHECK-RV64-SAME: <vscale x 32 x bfloat> [[SRC:%.*]], i64 noundef [[INDEX:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x bfloat> @llvm.vector.extract.nxv8bf16.nxv32bf16(<vscale x 32 x bfloat> [[SRC]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 8 x bfloat> [[TMP0]]
+//
+vbfloat16m2_t test_vget_v_bf16m8_bf16m2(vbfloat16m8_t src, size_t index) {
+  return __riscv_vget_v_bf16m8_bf16m2(src, 0);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 16 x bfloat> @test_vget_v_bf16m8_bf16m4(
+// CHECK-RV64-SAME: <vscale x 32 x bfloat> [[SRC:%.*]], i64 noundef [[INDEX:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x bfloat> @llvm.vector.extract.nxv16bf16.nxv32bf16(<vscale x 32 x bfloat> [[SRC]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 16 x bfloat> [[TMP0]]
+//
+vbfloat16m4_t test_vget_v_bf16m8_bf16m4(vbfloat16m8_t src, size_t index) {
+  return __riscv_vget_v_bf16m8_bf16m4(src, 0);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 1 x bfloat> @test_vget_v_bf16mf4x2_bf16mf4(
+// CHECK-RV64-SAME: { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[SRC:%.*]], i64 noundef [[INDEX:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[SRC]], 0
+// CHECK-RV64-NEXT:    ret <vscale x 1 x bfloat> [[TMP0]]
+//
+vbfloat16mf4_t test_vget_v_bf16mf4x2_bf16mf4(vbfloat16mf4x2_t src,
+                                             size_t index) {
+  return __riscv_vget_v_bf16mf4x2_bf16mf4(src, 0);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 1 x bfloat> @test_vget_v_bf16mf4x3_bf16mf4(
+// CHECK-RV64-SAME: { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[SRC:%.*]], i64 noundef [[INDEX:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[SRC]], 0
+// CHECK-RV64-NEXT:    ret <vscale x 1 x bfloat> [[TMP0]]
+//
+vbfloat16mf4_t test_vget_v_bf16mf4x3_bf16mf4(vbfloat16mf4x3_t src,
+                                             size_t index) {
+  return __riscv_vget_v_bf16mf4x3_bf16mf4(src, 0);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 1 x bfloat> @test_vget_v_bf16mf4x4_bf16mf4(
+// CHECK-RV64-SAME: { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[SRC:%.*]], i64 noundef [[INDEX:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[SRC]], 0
+// CHECK-RV64-NEXT:    ret <vscale x 1 x bfloat> [[TMP0]]
+//
+vbfloat16mf4_t test_vget_v_bf16mf4x4_bf16mf4(vbfloat16mf4x4_t src,
+                                             size_t index) {
+  return __riscv_vget_v_bf16mf4x4_bf16mf4(src, 0);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 1 x bfloat> @test_vget_v_bf16mf4x5_bf16mf4(
+// CHECK-RV64-SAME: { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[SRC:%.*]], i64 noundef [[INDEX:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[SRC]], 0
+// CHECK-RV64-NEXT:    ret <vscale x 1 x bfloat> [[TMP0]]
+//
+vbfloat16mf4_t test_vget_v_bf16mf4x5_bf16mf4(vbfloat16mf4x5_t src,
+                                             size_t index) {
+  return __riscv_vget_v_bf16mf4x5_bf16mf4(src, 0);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 1 x bfloat> @test_vget_v_bf16mf4x6_bf16mf4(
+// CHECK-RV64-SAME: { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[SRC:%.*]], i64 noundef [[INDEX:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[SRC]], 0
+// CHECK-RV64-NEXT:    ret <vscale x 1 x bfloat> [[TMP0]]
+//
+vbfloat16mf4_t test_vget_v_bf16mf4x6_bf16mf4(vbfloat16mf4x6_t src,
+                                             size_t index) {
+  return __riscv_vget_v_bf16mf4x6_bf16mf4(src, 0);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 1 x bfloat> @test_vget_v_bf16mf4x7_bf16mf4(
+// CHECK-RV64-SAME: { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[SRC:%.*]], i64 noundef [[INDEX:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[SRC]], 0
+// CHECK-RV64-NEXT:    ret <vscale x 1 x bfloat> [[TMP0]]
+//
+vbfloat16mf4_t test_vget_v_bf16mf4x7_bf16mf4(vbfloat16mf4x7_t src,
+                                             size_t index) {
+  return __riscv_vget_v_bf16mf4x7_bf16mf4(src, 0);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 1 x bfloat> @test_vget_v_bf16mf4x8_bf16mf4(
+// CHECK-RV64-SAME: { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[SRC:%.*]], i64 noundef [[INDEX:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[SRC]], 0
+// CHECK-RV64-NEXT:    ret <vscale x 1 x bfloat> [[TMP0]]
+//
+vbfloat16mf4_t test_vget_v_bf16mf4x8_bf16mf4(vbfloat16mf4x8_t src,
+                                             size_t index) {
+  return __riscv_vget_v_bf16mf4x8_bf16mf4(src, 0);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 2 x bfloat> @test_vget_v_bf16mf2x2_bf16mf2(
+// CHECK-RV64-SAME: { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[SRC:%.*]], i64 noundef [[INDEX:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[SRC]], 0
+// CHECK-RV64-NEXT:    ret <vscale x 2 x bfloat> [[TMP0]]
+//
+vbfloat16mf2_t test_vget_v_bf16mf2x2_bf16mf2(vbfloat16mf2x2_t src,
+                                             size_t index) {
+  return __riscv_vget_v_bf16mf2x2_bf16mf2(src, 0);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 2 x bfloat> @test_vget_v_bf16mf2x3_bf16mf2(
+// CHECK-RV64-SAME: { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[SRC:%.*]], i64 noundef [[INDEX:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[SRC]], 0
+// CHECK-RV64-NEXT:    ret <vscale x 2 x bfloat> [[TMP0]]
+//
+vbfloat16mf2_t test_vget_v_bf16mf2x3_bf16mf2(vbfloat16mf2x3_t src,
+                                             size_t index) {
+  return __riscv_vget_v_bf16mf2x3_bf16mf2(src, 0);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 2 x bfloat> @test_vget_v_bf16mf2x4_bf16mf2(
+// CHECK-RV64-SAME: { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[SRC:%.*]], i64 noundef [[INDEX:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[SRC]], 0
+// CHECK-RV64-NEXT:    ret <vscale x 2 x bfloat> [[TMP0]]
+//
+vbfloat16mf2_t test_vget_v_bf16mf2x4_bf16mf2(vbfloat16mf2x4_t src,
+                                             size_t index) {
+  return __riscv_vget_v_bf16mf2x4_bf16mf2(src, 0);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 2 x bfloat> @test_vget_v_bf16mf2x5_bf16mf2(
+// CHECK-RV64-SAME: { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[SRC:%.*]], i64 noundef [[INDEX:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[SRC]], 0
+// CHECK-RV64-NEXT:    ret <vscale x 2 x bfloat> [[TMP0]]
+//
+vbfloat16mf2_t test_vget_v_bf16mf2x5_bf16mf2(vbfloat16mf2x5_t src,
+                                             size_t index) {
+  return __riscv_vget_v_bf16mf2x5_bf16mf2(src, 0);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 2 x bfloat> @test_vget_v_bf16mf2x6_bf16mf2(
+// CHECK-RV64-SAME: { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[SRC:%.*]], i64 noundef [[INDEX:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[SRC]], 0
+// CHECK-RV64-NEXT:    ret <vscale x 2 x bfloat> [[TMP0]]
+//
+vbfloat16mf2_t test_vget_v_bf16mf2x6_bf16mf2(vbfloat16mf2x6_t src,
+                                             size_t index) {
+  return __riscv_vget_v_bf16mf2x6_bf16mf2(src, 0);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 2 x bfloat> @test_vget_v_bf16mf2x7_bf16mf2(
+// CHECK-RV64-SAME: { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[SRC:%.*]], i64 noundef [[INDEX:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[SRC]], 0
+// CHECK-RV64-NEXT:    ret <vscale x 2 x bfloat> [[TMP0]]
+//
+vbfloat16mf2_t test_vget_v_bf16mf2x7_bf16mf2(vbfloat16mf2x7_t src,
+                                             size_t index) {
+  return __riscv_vget_v_bf16mf2x7_bf16mf2(src, 0);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 2 x bfloat> @test_vget_v_bf16mf2x8_bf16mf2(
+// CHECK-RV64-SAME: { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[SRC:%.*]], i64 noundef [[INDEX:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[SRC]], 0
+// CHECK-RV64-NEXT:    ret <vscale x 2 x bfloat> [[TMP0]]
+//
+vbfloat16mf2_t test_vget_v_bf16mf2x8_bf16mf2(vbfloat16mf2x8_t src,
+                                             size_t index) {
+  return __riscv_vget_v_bf16mf2x8_bf16mf2(src, 0);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 4 x bfloat> @test_vget_v_bf16m1x2_bf16m1(
+// CHECK-RV64-SAME: { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[SRC:%.*]], i64 noundef [[INDEX:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[SRC]], 0
+// CHECK-RV64-NEXT:    ret <vscale x 4 x bfloat> [[TMP0]]
+//
+vbfloat16m1_t test_vget_v_bf16m1x2_bf16m1(vbfloat16m1x2_t src, size_t index) {
+  return __riscv_vget_v_bf16m1x2_bf16m1(src, 0);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 4 x bfloat> @test_vget_v_bf16m1x3_bf16m1(
+// CHECK-RV64-SAME: { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[SRC:%.*]], i64 noundef [[INDEX:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[SRC]], 0
+// CHECK-RV64-NEXT:    ret <vscale x 4 x bfloat> [[TMP0]]
+//
+vbfloat16m1_t test_vget_v_bf16m1x3_bf16m1(vbfloat16m1x3_t src, size_t index) {
+  return __riscv_vget_v_bf16m1x3_bf16m1(src, 0);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 4 x bfloat> @test_vget_v_bf16m1x4_bf16m1(
+// CHECK-RV64-SAME: { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[SRC:%.*]], i64 noundef [[INDEX:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[SRC]], 0
+// CHECK-RV64-NEXT:    ret <vscale x 4 x bfloat> [[TMP0]]
+//
+vbfloat16m1_t test_vget_v_bf16m1x4_bf16m1(vbfloat16m1x4_t src, size_t index) {
+  return __riscv_vget_v_bf16m1x4_bf16m1(src, 0);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 4 x bfloat> @test_vget_v_bf16m1x5_bf16m1(
+// CHECK-RV64-SAME: { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[SRC:%.*]], i64 noundef [[INDEX:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[SRC]], 0
+// CHECK-RV64-NEXT:    ret <vscale x 4 x bfloat> [[TMP0]]
+//
+vbfloat16m1_t test_vget_v_bf16m1x5_bf16m1(vbfloat16m1x5_t src, size_t index) {
+  return __riscv_vget_v_bf16m1x5_bf16m1(src, 0);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 4 x bfloat> @test_vget_v_bf16m1x6_bf16m1(
+// CHECK-RV64-SAME: { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[SRC:%.*]], i64 noundef [[INDEX:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[SRC]], 0
+// CHECK-RV64-NEXT:    ret <vscale x 4 x bfloat> [[TMP0]]
+//
+vbfloat16m1_t test_vget_v_bf16m1x6_bf16m1(vbfloat16m1x6_t src, size_t index) {
+  return __riscv_vget_v_bf16m1x6_bf16m1(src, 0);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 4 x bfloat> @test_vget_v_bf16m1x7_bf16m1(
+// CHECK-RV64-SAME: { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[SRC:%.*]], i64 noundef [[INDEX:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[SRC]], 0
+// CHECK-RV64-NEXT:    ret <vscale x 4 x bfloat> [[TMP0]]
+//
+vbfloat16m1_t test_vget_v_bf16m1x7_bf16m1(vbfloat16m1x7_t src, size_t index) {
+  return __riscv_vget_v_bf16m1x7_bf16m1(src, 0);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 4 x bfloat> @test_vget_v_bf16m1x8_bf16m1(
+// CHECK-RV64-SAME: { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[SRC:%.*]], i64 noundef [[INDEX:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[SRC]], 0
+// CHECK-RV64-NEXT:    ret <vscale x 4 x bfloat> [[TMP0]]
+//
+vbfloat16m1_t test_vget_v_bf16m1x8_bf16m1(vbfloat16m1x8_t src, size_t index) {
+  return __riscv_vget_v_bf16m1x8_bf16m1(src, 0);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 8 x bfloat> @test_vget_v_bf16m2x2_bf16m2(
+// CHECK-RV64-SAME: { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[SRC:%.*]], i64 noundef [[INDEX:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[SRC]], 0
+// CHECK-RV64-NEXT:    ret <vscale x 8 x bfloat> [[TMP0]]
+//
+vbfloat16m2_t test_vget_v_bf16m2x2_bf16m2(vbfloat16m2x2_t src, size_t index) {
+  return __riscv_vget_v_bf16m2x2_bf16m2(src, 0);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 8 x bfloat> @test_vget_v_bf16m2x3_bf16m2(
+// CHECK-RV64-SAME: { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[SRC:%.*]], i64 noundef [[INDEX:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[SRC]], 0
+// CHECK-RV64-NEXT:    ret <vscale x 8 x bfloat> [[TMP0]]
+//
+vbfloat16m2_t test_vget_v_bf16m2x3_bf16m2(vbfloat16m2x3_t src, size_t index) {
+  return __riscv_vget_v_bf16m2x3_bf16m2(src, 0);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 8 x bfloat> @test_vget_v_bf16m2x4_bf16m2(
+// CHECK-RV64-SAME: { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[SRC:%.*]], i64 noundef [[INDEX:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[SRC]], 0
+// CHECK-RV64-NEXT:    ret <vscale x 8 x bfloat> [[TMP0]]
+//
+vbfloat16m2_t test_vget_v_bf16m2x4_bf16m2(vbfloat16m2x4_t src, size_t index) {
+  return __riscv_vget_v_bf16m2x4_bf16m2(src, 0);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 16 x bfloat> @test_vget_v_bf16m4x2_bf16m4(
+// CHECK-RV64-SAME: { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[SRC:%.*]], i64 noundef [[INDEX:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[SRC]], 0
+// CHECK-RV64-NEXT:    ret <vscale x 16 x bfloat> [[TMP0]]
+//
+vbfloat16m4_t test_vget_v_bf16m4x2_bf16m4(vbfloat16m4x2_t src, size_t index) {
+  return __riscv_vget_v_bf16m4x2_bf16m4(src, 0);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vle16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vle16.c
new file mode 100644
index 000000000000..6f54ddf3f87c
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vle16.c
@@ -0,0 +1,132 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 1 x bfloat> @test_vle16_v_bf16mf4(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x bfloat> @llvm.riscv.vle.nxv1bf16.i64(<vscale x 1 x bfloat> poison, ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret <vscale x 1 x bfloat> [[TMP0]]
+//
+vbfloat16mf4_t test_vle16_v_bf16mf4(const __bf16 *rs1, size_t vl) {
+  return __riscv_vle16_v_bf16mf4(rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 2 x bfloat> @test_vle16_v_bf16mf2(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x bfloat> @llvm.riscv.vle.nxv2bf16.i64(<vscale x 2 x bfloat> poison, ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret <vscale x 2 x bfloat> [[TMP0]]
+//
+vbfloat16mf2_t test_vle16_v_bf16mf2(const __bf16 *rs1, size_t vl) {
+  return __riscv_vle16_v_bf16mf2(rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 4 x bfloat> @test_vle16_v_bf16m1(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x bfloat> @llvm.riscv.vle.nxv4bf16.i64(<vscale x 4 x bfloat> poison, ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret <vscale x 4 x bfloat> [[TMP0]]
+//
+vbfloat16m1_t test_vle16_v_bf16m1(const __bf16 *rs1, size_t vl) {
+  return __riscv_vle16_v_bf16m1(rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 8 x bfloat> @test_vle16_v_bf16m2(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x bfloat> @llvm.riscv.vle.nxv8bf16.i64(<vscale x 8 x bfloat> poison, ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret <vscale x 8 x bfloat> [[TMP0]]
+//
+vbfloat16m2_t test_vle16_v_bf16m2(const __bf16 *rs1, size_t vl) {
+  return __riscv_vle16_v_bf16m2(rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 16 x bfloat> @test_vle16_v_bf16m4(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x bfloat> @llvm.riscv.vle.nxv16bf16.i64(<vscale x 16 x bfloat> poison, ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret <vscale x 16 x bfloat> [[TMP0]]
+//
+vbfloat16m4_t test_vle16_v_bf16m4(const __bf16 *rs1, size_t vl) {
+  return __riscv_vle16_v_bf16m4(rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 32 x bfloat> @test_vle16_v_bf16m8(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x bfloat> @llvm.riscv.vle.nxv32bf16.i64(<vscale x 32 x bfloat> poison, ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret <vscale x 32 x bfloat> [[TMP0]]
+//
+vbfloat16m8_t test_vle16_v_bf16m8(const __bf16 *rs1, size_t vl) {
+  return __riscv_vle16_v_bf16m8(rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 1 x bfloat> @test_vle16_v_bf16mf4_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x bfloat> @llvm.riscv.vle.mask.nxv1bf16.i64(<vscale x 1 x bfloat> poison, ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret <vscale x 1 x bfloat> [[TMP0]]
+//
+vbfloat16mf4_t test_vle16_v_bf16mf4_m(vbool64_t vm, const __bf16 *rs1,
+                                      size_t vl) {
+  return __riscv_vle16_v_bf16mf4_m(vm, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 2 x bfloat> @test_vle16_v_bf16mf2_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x bfloat> @llvm.riscv.vle.mask.nxv2bf16.i64(<vscale x 2 x bfloat> poison, ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret <vscale x 2 x bfloat> [[TMP0]]
+//
+vbfloat16mf2_t test_vle16_v_bf16mf2_m(vbool32_t vm, const __bf16 *rs1,
+                                      size_t vl) {
+  return __riscv_vle16_v_bf16mf2_m(vm, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 4 x bfloat> @test_vle16_v_bf16m1_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x bfloat> @llvm.riscv.vle.mask.nxv4bf16.i64(<vscale x 4 x bfloat> poison, ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret <vscale x 4 x bfloat> [[TMP0]]
+//
+vbfloat16m1_t test_vle16_v_bf16m1_m(vbool16_t vm, const __bf16 *rs1,
+                                    size_t vl) {
+  return __riscv_vle16_v_bf16m1_m(vm, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 8 x bfloat> @test_vle16_v_bf16m2_m(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x bfloat> @llvm.riscv.vle.mask.nxv8bf16.i64(<vscale x 8 x bfloat> poison, ptr [[RS1]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret <vscale x 8 x bfloat> [[TMP0]]
+//
+vbfloat16m2_t test_vle16_v_bf16m2_m(vbool8_t vm, const __bf16 *rs1, size_t vl) {
+  return __riscv_vle16_v_bf16m2_m(vm, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 16 x bfloat> @test_vle16_v_bf16m4_m(
+// CHECK-RV64-SAME: <vscale x 16 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x bfloat> @llvm.riscv.vle.mask.nxv16bf16.i64(<vscale x 16 x bfloat> poison, ptr [[RS1]], <vscale x 16 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret <vscale x 16 x bfloat> [[TMP0]]
+//
+vbfloat16m4_t test_vle16_v_bf16m4_m(vbool4_t vm, const __bf16 *rs1, size_t vl) {
+  return __riscv_vle16_v_bf16m4_m(vm, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 32 x bfloat> @test_vle16_v_bf16m8_m(
+// CHECK-RV64-SAME: <vscale x 32 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x bfloat> @llvm.riscv.vle.mask.nxv32bf16.i64(<vscale x 32 x bfloat> poison, ptr [[RS1]], <vscale x 32 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret <vscale x 32 x bfloat> [[TMP0]]
+//
+vbfloat16m8_t test_vle16_v_bf16m8_m(vbool2_t vm, const __bf16 *rs1, size_t vl) {
+  return __riscv_vle16_v_bf16m8_m(vm, rs1, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vle16ff.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vle16ff.c
new file mode 100644
index 000000000000..f73cfe73bae0
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vle16ff.c
@@ -0,0 +1,177 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 1 x bfloat> @test_vle16ff_v_bf16mf4(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, i64 } @llvm.riscv.vleff.nxv1bf16.i64(<vscale x 1 x bfloat> poison, ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    store i64 [[TMP2]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret <vscale x 1 x bfloat> [[TMP1]]
+//
+vbfloat16mf4_t test_vle16ff_v_bf16mf4(const __bf16 *rs1, size_t *new_vl,
+                                      size_t vl) {
+  return __riscv_vle16ff_v_bf16mf4(rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 2 x bfloat> @test_vle16ff_v_bf16mf2(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, i64 } @llvm.riscv.vleff.nxv2bf16.i64(<vscale x 2 x bfloat> poison, ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    store i64 [[TMP2]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret <vscale x 2 x bfloat> [[TMP1]]
+//
+vbfloat16mf2_t test_vle16ff_v_bf16mf2(const __bf16 *rs1, size_t *new_vl,
+                                      size_t vl) {
+  return __riscv_vle16ff_v_bf16mf2(rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 4 x bfloat> @test_vle16ff_v_bf16m1(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, i64 } @llvm.riscv.vleff.nxv4bf16.i64(<vscale x 4 x bfloat> poison, ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    store i64 [[TMP2]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret <vscale x 4 x bfloat> [[TMP1]]
+//
+vbfloat16m1_t test_vle16ff_v_bf16m1(const __bf16 *rs1, size_t *new_vl,
+                                    size_t vl) {
+  return __riscv_vle16ff_v_bf16m1(rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 8 x bfloat> @test_vle16ff_v_bf16m2(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 8 x bfloat>, i64 } @llvm.riscv.vleff.nxv8bf16.i64(<vscale x 8 x bfloat> poison, ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    store i64 [[TMP2]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret <vscale x 8 x bfloat> [[TMP1]]
+//
+vbfloat16m2_t test_vle16ff_v_bf16m2(const __bf16 *rs1, size_t *new_vl,
+                                    size_t vl) {
+  return __riscv_vle16ff_v_bf16m2(rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 16 x bfloat> @test_vle16ff_v_bf16m4(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 16 x bfloat>, i64 } @llvm.riscv.vleff.nxv16bf16.i64(<vscale x 16 x bfloat> poison, ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 16 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 16 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    store i64 [[TMP2]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret <vscale x 16 x bfloat> [[TMP1]]
+//
+vbfloat16m4_t test_vle16ff_v_bf16m4(const __bf16 *rs1, size_t *new_vl,
+                                    size_t vl) {
+  return __riscv_vle16ff_v_bf16m4(rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 32 x bfloat> @test_vle16ff_v_bf16m8(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 32 x bfloat>, i64 } @llvm.riscv.vleff.nxv32bf16.i64(<vscale x 32 x bfloat> poison, ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 32 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 32 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    store i64 [[TMP2]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret <vscale x 32 x bfloat> [[TMP1]]
+//
+vbfloat16m8_t test_vle16ff_v_bf16m8(const __bf16 *rs1, size_t *new_vl,
+                                    size_t vl) {
+  return __riscv_vle16ff_v_bf16m8(rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 1 x bfloat> @test_vle16ff_v_bf16mf4_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, i64 } @llvm.riscv.vleff.mask.nxv1bf16.i64(<vscale x 1 x bfloat> poison, ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    store i64 [[TMP2]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret <vscale x 1 x bfloat> [[TMP1]]
+//
+vbfloat16mf4_t test_vle16ff_v_bf16mf4_m(vbool64_t vm, const __bf16 *rs1,
+                                        size_t *new_vl, size_t vl) {
+  return __riscv_vle16ff_v_bf16mf4_m(vm, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 2 x bfloat> @test_vle16ff_v_bf16mf2_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, i64 } @llvm.riscv.vleff.mask.nxv2bf16.i64(<vscale x 2 x bfloat> poison, ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    store i64 [[TMP2]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret <vscale x 2 x bfloat> [[TMP1]]
+//
+vbfloat16mf2_t test_vle16ff_v_bf16mf2_m(vbool32_t vm, const __bf16 *rs1,
+                                        size_t *new_vl, size_t vl) {
+  return __riscv_vle16ff_v_bf16mf2_m(vm, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 4 x bfloat> @test_vle16ff_v_bf16m1_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, i64 } @llvm.riscv.vleff.mask.nxv4bf16.i64(<vscale x 4 x bfloat> poison, ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    store i64 [[TMP2]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret <vscale x 4 x bfloat> [[TMP1]]
+//
+vbfloat16m1_t test_vle16ff_v_bf16m1_m(vbool16_t vm, const __bf16 *rs1,
+                                      size_t *new_vl, size_t vl) {
+  return __riscv_vle16ff_v_bf16m1_m(vm, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 8 x bfloat> @test_vle16ff_v_bf16m2_m(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 8 x bfloat>, i64 } @llvm.riscv.vleff.mask.nxv8bf16.i64(<vscale x 8 x bfloat> poison, ptr [[RS1]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    store i64 [[TMP2]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret <vscale x 8 x bfloat> [[TMP1]]
+//
+vbfloat16m2_t test_vle16ff_v_bf16m2_m(vbool8_t vm, const __bf16 *rs1,
+                                      size_t *new_vl, size_t vl) {
+  return __riscv_vle16ff_v_bf16m2_m(vm, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 16 x bfloat> @test_vle16ff_v_bf16m4_m(
+// CHECK-RV64-SAME: <vscale x 16 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 16 x bfloat>, i64 } @llvm.riscv.vleff.mask.nxv16bf16.i64(<vscale x 16 x bfloat> poison, ptr [[RS1]], <vscale x 16 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 16 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 16 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    store i64 [[TMP2]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret <vscale x 16 x bfloat> [[TMP1]]
+//
+vbfloat16m4_t test_vle16ff_v_bf16m4_m(vbool4_t vm, const __bf16 *rs1,
+                                      size_t *new_vl, size_t vl) {
+  return __riscv_vle16ff_v_bf16m4_m(vm, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 32 x bfloat> @test_vle16ff_v_bf16m8_m(
+// CHECK-RV64-SAME: <vscale x 32 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 32 x bfloat>, i64 } @llvm.riscv.vleff.mask.nxv32bf16.i64(<vscale x 32 x bfloat> poison, ptr [[RS1]], <vscale x 32 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 32 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 32 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    store i64 [[TMP2]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret <vscale x 32 x bfloat> [[TMP1]]
+//
+vbfloat16m8_t test_vle16ff_v_bf16m8_m(vbool2_t vm, const __bf16 *rs1,
+                                      size_t *new_vl, size_t vl) {
+  return __riscv_vle16ff_v_bf16m8_m(vm, rs1, new_vl, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vlmul_ext_v.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vlmul_ext_v.c
new file mode 100644
index 000000000000..3ab73c512316
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vlmul_ext_v.c
@@ -0,0 +1,159 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 2 x bfloat> @test_vlmul_ext_v_bf16mf4_bf16mf2(
+// CHECK-RV64-SAME: <vscale x 1 x bfloat> [[VALUE:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x bfloat> @llvm.vector.insert.nxv2bf16.nxv1bf16(<vscale x 2 x bfloat> poison, <vscale x 1 x bfloat> [[VALUE]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 2 x bfloat> [[TMP0]]
+//
+vbfloat16mf2_t test_vlmul_ext_v_bf16mf4_bf16mf2(vbfloat16mf4_t value) {
+  return __riscv_vlmul_ext_v_bf16mf4_bf16mf2(value);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 4 x bfloat> @test_vlmul_ext_v_bf16mf4_bf16m1(
+// CHECK-RV64-SAME: <vscale x 1 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x bfloat> @llvm.vector.insert.nxv4bf16.nxv1bf16(<vscale x 4 x bfloat> poison, <vscale x 1 x bfloat> [[VALUE]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 4 x bfloat> [[TMP0]]
+//
+vbfloat16m1_t test_vlmul_ext_v_bf16mf4_bf16m1(vbfloat16mf4_t value) {
+  return __riscv_vlmul_ext_v_bf16mf4_bf16m1(value);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 8 x bfloat> @test_vlmul_ext_v_bf16mf4_bf16m2(
+// CHECK-RV64-SAME: <vscale x 1 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x bfloat> @llvm.vector.insert.nxv8bf16.nxv1bf16(<vscale x 8 x bfloat> poison, <vscale x 1 x bfloat> [[VALUE]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 8 x bfloat> [[TMP0]]
+//
+vbfloat16m2_t test_vlmul_ext_v_bf16mf4_bf16m2(vbfloat16mf4_t value) {
+  return __riscv_vlmul_ext_v_bf16mf4_bf16m2(value);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 16 x bfloat> @test_vlmul_ext_v_bf16mf4_bf16m4(
+// CHECK-RV64-SAME: <vscale x 1 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x bfloat> @llvm.vector.insert.nxv16bf16.nxv1bf16(<vscale x 16 x bfloat> poison, <vscale x 1 x bfloat> [[VALUE]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 16 x bfloat> [[TMP0]]
+//
+vbfloat16m4_t test_vlmul_ext_v_bf16mf4_bf16m4(vbfloat16mf4_t value) {
+  return __riscv_vlmul_ext_v_bf16mf4_bf16m4(value);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 32 x bfloat> @test_vlmul_ext_v_bf16mf4_bf16m8(
+// CHECK-RV64-SAME: <vscale x 1 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x bfloat> @llvm.vector.insert.nxv32bf16.nxv1bf16(<vscale x 32 x bfloat> poison, <vscale x 1 x bfloat> [[VALUE]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 32 x bfloat> [[TMP0]]
+//
+vbfloat16m8_t test_vlmul_ext_v_bf16mf4_bf16m8(vbfloat16mf4_t value) {
+  return __riscv_vlmul_ext_v_bf16mf4_bf16m8(value);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 4 x bfloat> @test_vlmul_ext_v_bf16mf2_bf16m1(
+// CHECK-RV64-SAME: <vscale x 2 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x bfloat> @llvm.vector.insert.nxv4bf16.nxv2bf16(<vscale x 4 x bfloat> poison, <vscale x 2 x bfloat> [[VALUE]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 4 x bfloat> [[TMP0]]
+//
+vbfloat16m1_t test_vlmul_ext_v_bf16mf2_bf16m1(vbfloat16mf2_t value) {
+  return __riscv_vlmul_ext_v_bf16mf2_bf16m1(value);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 8 x bfloat> @test_vlmul_ext_v_bf16mf2_bf16m2(
+// CHECK-RV64-SAME: <vscale x 2 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x bfloat> @llvm.vector.insert.nxv8bf16.nxv2bf16(<vscale x 8 x bfloat> poison, <vscale x 2 x bfloat> [[VALUE]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 8 x bfloat> [[TMP0]]
+//
+vbfloat16m2_t test_vlmul_ext_v_bf16mf2_bf16m2(vbfloat16mf2_t value) {
+  return __riscv_vlmul_ext_v_bf16mf2_bf16m2(value);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 16 x bfloat> @test_vlmul_ext_v_bf16mf2_bf16m4(
+// CHECK-RV64-SAME: <vscale x 2 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x bfloat> @llvm.vector.insert.nxv16bf16.nxv2bf16(<vscale x 16 x bfloat> poison, <vscale x 2 x bfloat> [[VALUE]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 16 x bfloat> [[TMP0]]
+//
+vbfloat16m4_t test_vlmul_ext_v_bf16mf2_bf16m4(vbfloat16mf2_t value) {
+  return __riscv_vlmul_ext_v_bf16mf2_bf16m4(value);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 32 x bfloat> @test_vlmul_ext_v_bf16mf2_bf16m8(
+// CHECK-RV64-SAME: <vscale x 2 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x bfloat> @llvm.vector.insert.nxv32bf16.nxv2bf16(<vscale x 32 x bfloat> poison, <vscale x 2 x bfloat> [[VALUE]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 32 x bfloat> [[TMP0]]
+//
+vbfloat16m8_t test_vlmul_ext_v_bf16mf2_bf16m8(vbfloat16mf2_t value) {
+  return __riscv_vlmul_ext_v_bf16mf2_bf16m8(value);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 8 x bfloat> @test_vlmul_ext_v_bf16m1_bf16m2(
+// CHECK-RV64-SAME: <vscale x 4 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x bfloat> @llvm.vector.insert.nxv8bf16.nxv4bf16(<vscale x 8 x bfloat> poison, <vscale x 4 x bfloat> [[VALUE]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 8 x bfloat> [[TMP0]]
+//
+vbfloat16m2_t test_vlmul_ext_v_bf16m1_bf16m2(vbfloat16m1_t value) {
+  return __riscv_vlmul_ext_v_bf16m1_bf16m2(value);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 16 x bfloat> @test_vlmul_ext_v_bf16m1_bf16m4(
+// CHECK-RV64-SAME: <vscale x 4 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x bfloat> @llvm.vector.insert.nxv16bf16.nxv4bf16(<vscale x 16 x bfloat> poison, <vscale x 4 x bfloat> [[VALUE]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 16 x bfloat> [[TMP0]]
+//
+vbfloat16m4_t test_vlmul_ext_v_bf16m1_bf16m4(vbfloat16m1_t value) {
+  return __riscv_vlmul_ext_v_bf16m1_bf16m4(value);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 32 x bfloat> @test_vlmul_ext_v_bf16m1_bf16m8(
+// CHECK-RV64-SAME: <vscale x 4 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x bfloat> @llvm.vector.insert.nxv32bf16.nxv4bf16(<vscale x 32 x bfloat> poison, <vscale x 4 x bfloat> [[VALUE]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 32 x bfloat> [[TMP0]]
+//
+vbfloat16m8_t test_vlmul_ext_v_bf16m1_bf16m8(vbfloat16m1_t value) {
+  return __riscv_vlmul_ext_v_bf16m1_bf16m8(value);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 16 x bfloat> @test_vlmul_ext_v_bf16m2_bf16m4(
+// CHECK-RV64-SAME: <vscale x 8 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x bfloat> @llvm.vector.insert.nxv16bf16.nxv8bf16(<vscale x 16 x bfloat> poison, <vscale x 8 x bfloat> [[VALUE]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 16 x bfloat> [[TMP0]]
+//
+vbfloat16m4_t test_vlmul_ext_v_bf16m2_bf16m4(vbfloat16m2_t value) {
+  return __riscv_vlmul_ext_v_bf16m2_bf16m4(value);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 32 x bfloat> @test_vlmul_ext_v_bf16m2_bf16m8(
+// CHECK-RV64-SAME: <vscale x 8 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x bfloat> @llvm.vector.insert.nxv32bf16.nxv8bf16(<vscale x 32 x bfloat> poison, <vscale x 8 x bfloat> [[VALUE]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 32 x bfloat> [[TMP0]]
+//
+vbfloat16m8_t test_vlmul_ext_v_bf16m2_bf16m8(vbfloat16m2_t value) {
+  return __riscv_vlmul_ext_v_bf16m2_bf16m8(value);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 32 x bfloat> @test_vlmul_ext_v_bf16m4_bf16m8(
+// CHECK-RV64-SAME: <vscale x 16 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x bfloat> @llvm.vector.insert.nxv32bf16.nxv16bf16(<vscale x 32 x bfloat> poison, <vscale x 16 x bfloat> [[VALUE]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 32 x bfloat> [[TMP0]]
+//
+vbfloat16m8_t test_vlmul_ext_v_bf16m4_bf16m8(vbfloat16m4_t value) {
+  return __riscv_vlmul_ext_v_bf16m4_bf16m8(value);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vlmul_trunc_v.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vlmul_trunc_v.c
new file mode 100644
index 000000000000..478b1e556113
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vlmul_trunc_v.c
@@ -0,0 +1,159 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 1 x bfloat> @test_vlmul_trunc_v_bf16mf2_bf16mf4(
+// CHECK-RV64-SAME: <vscale x 2 x bfloat> [[VALUE:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x bfloat> @llvm.vector.extract.nxv1bf16.nxv2bf16(<vscale x 2 x bfloat> [[VALUE]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 1 x bfloat> [[TMP0]]
+//
+vbfloat16mf4_t test_vlmul_trunc_v_bf16mf2_bf16mf4(vbfloat16mf2_t value) {
+  return __riscv_vlmul_trunc_v_bf16mf2_bf16mf4(value);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 1 x bfloat> @test_vlmul_trunc_v_bf16m1_bf16mf4(
+// CHECK-RV64-SAME: <vscale x 4 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x bfloat> @llvm.vector.extract.nxv1bf16.nxv4bf16(<vscale x 4 x bfloat> [[VALUE]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 1 x bfloat> [[TMP0]]
+//
+vbfloat16mf4_t test_vlmul_trunc_v_bf16m1_bf16mf4(vbfloat16m1_t value) {
+  return __riscv_vlmul_trunc_v_bf16m1_bf16mf4(value);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 2 x bfloat> @test_vlmul_trunc_v_bf16m1_bf16mf2(
+// CHECK-RV64-SAME: <vscale x 4 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x bfloat> @llvm.vector.extract.nxv2bf16.nxv4bf16(<vscale x 4 x bfloat> [[VALUE]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 2 x bfloat> [[TMP0]]
+//
+vbfloat16mf2_t test_vlmul_trunc_v_bf16m1_bf16mf2(vbfloat16m1_t value) {
+  return __riscv_vlmul_trunc_v_bf16m1_bf16mf2(value);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 1 x bfloat> @test_vlmul_trunc_v_bf16m2_bf16mf4(
+// CHECK-RV64-SAME: <vscale x 8 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x bfloat> @llvm.vector.extract.nxv1bf16.nxv8bf16(<vscale x 8 x bfloat> [[VALUE]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 1 x bfloat> [[TMP0]]
+//
+vbfloat16mf4_t test_vlmul_trunc_v_bf16m2_bf16mf4(vbfloat16m2_t value) {
+  return __riscv_vlmul_trunc_v_bf16m2_bf16mf4(value);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 2 x bfloat> @test_vlmul_trunc_v_bf16m2_bf16mf2(
+// CHECK-RV64-SAME: <vscale x 8 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x bfloat> @llvm.vector.extract.nxv2bf16.nxv8bf16(<vscale x 8 x bfloat> [[VALUE]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 2 x bfloat> [[TMP0]]
+//
+vbfloat16mf2_t test_vlmul_trunc_v_bf16m2_bf16mf2(vbfloat16m2_t value) {
+  return __riscv_vlmul_trunc_v_bf16m2_bf16mf2(value);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 4 x bfloat> @test_vlmul_trunc_v_bf16m2_bf16m1(
+// CHECK-RV64-SAME: <vscale x 8 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x bfloat> @llvm.vector.extract.nxv4bf16.nxv8bf16(<vscale x 8 x bfloat> [[VALUE]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 4 x bfloat> [[TMP0]]
+//
+vbfloat16m1_t test_vlmul_trunc_v_bf16m2_bf16m1(vbfloat16m2_t value) {
+  return __riscv_vlmul_trunc_v_bf16m2_bf16m1(value);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 1 x bfloat> @test_vlmul_trunc_v_bf16m4_bf16mf4(
+// CHECK-RV64-SAME: <vscale x 16 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x bfloat> @llvm.vector.extract.nxv1bf16.nxv16bf16(<vscale x 16 x bfloat> [[VALUE]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 1 x bfloat> [[TMP0]]
+//
+vbfloat16mf4_t test_vlmul_trunc_v_bf16m4_bf16mf4(vbfloat16m4_t value) {
+  return __riscv_vlmul_trunc_v_bf16m4_bf16mf4(value);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 2 x bfloat> @test_vlmul_trunc_v_bf16m4_bf16mf2(
+// CHECK-RV64-SAME: <vscale x 16 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x bfloat> @llvm.vector.extract.nxv2bf16.nxv16bf16(<vscale x 16 x bfloat> [[VALUE]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 2 x bfloat> [[TMP0]]
+//
+vbfloat16mf2_t test_vlmul_trunc_v_bf16m4_bf16mf2(vbfloat16m4_t value) {
+  return __riscv_vlmul_trunc_v_bf16m4_bf16mf2(value);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 4 x bfloat> @test_vlmul_trunc_v_bf16m4_bf16m1(
+// CHECK-RV64-SAME: <vscale x 16 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x bfloat> @llvm.vector.extract.nxv4bf16.nxv16bf16(<vscale x 16 x bfloat> [[VALUE]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 4 x bfloat> [[TMP0]]
+//
+vbfloat16m1_t test_vlmul_trunc_v_bf16m4_bf16m1(vbfloat16m4_t value) {
+  return __riscv_vlmul_trunc_v_bf16m4_bf16m1(value);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 8 x bfloat> @test_vlmul_trunc_v_bf16m4_bf16m2(
+// CHECK-RV64-SAME: <vscale x 16 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x bfloat> @llvm.vector.extract.nxv8bf16.nxv16bf16(<vscale x 16 x bfloat> [[VALUE]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 8 x bfloat> [[TMP0]]
+//
+vbfloat16m2_t test_vlmul_trunc_v_bf16m4_bf16m2(vbfloat16m4_t value) {
+  return __riscv_vlmul_trunc_v_bf16m4_bf16m2(value);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 1 x bfloat> @test_vlmul_trunc_v_bf16m8_bf16mf4(
+// CHECK-RV64-SAME: <vscale x 32 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x bfloat> @llvm.vector.extract.nxv1bf16.nxv32bf16(<vscale x 32 x bfloat> [[VALUE]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 1 x bfloat> [[TMP0]]
+//
+vbfloat16mf4_t test_vlmul_trunc_v_bf16m8_bf16mf4(vbfloat16m8_t value) {
+  return __riscv_vlmul_trunc_v_bf16m8_bf16mf4(value);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 2 x bfloat> @test_vlmul_trunc_v_bf16m8_bf16mf2(
+// CHECK-RV64-SAME: <vscale x 32 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x bfloat> @llvm.vector.extract.nxv2bf16.nxv32bf16(<vscale x 32 x bfloat> [[VALUE]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 2 x bfloat> [[TMP0]]
+//
+vbfloat16mf2_t test_vlmul_trunc_v_bf16m8_bf16mf2(vbfloat16m8_t value) {
+  return __riscv_vlmul_trunc_v_bf16m8_bf16mf2(value);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 4 x bfloat> @test_vlmul_trunc_v_bf16m8_bf16m1(
+// CHECK-RV64-SAME: <vscale x 32 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x bfloat> @llvm.vector.extract.nxv4bf16.nxv32bf16(<vscale x 32 x bfloat> [[VALUE]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 4 x bfloat> [[TMP0]]
+//
+vbfloat16m1_t test_vlmul_trunc_v_bf16m8_bf16m1(vbfloat16m8_t value) {
+  return __riscv_vlmul_trunc_v_bf16m8_bf16m1(value);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 8 x bfloat> @test_vlmul_trunc_v_bf16m8_bf16m2(
+// CHECK-RV64-SAME: <vscale x 32 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x bfloat> @llvm.vector.extract.nxv8bf16.nxv32bf16(<vscale x 32 x bfloat> [[VALUE]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 8 x bfloat> [[TMP0]]
+//
+vbfloat16m2_t test_vlmul_trunc_v_bf16m8_bf16m2(vbfloat16m8_t value) {
+  return __riscv_vlmul_trunc_v_bf16m8_bf16m2(value);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 16 x bfloat> @test_vlmul_trunc_v_bf16m8_bf16m4(
+// CHECK-RV64-SAME: <vscale x 32 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x bfloat> @llvm.vector.extract.nxv16bf16.nxv32bf16(<vscale x 32 x bfloat> [[VALUE]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 16 x bfloat> [[TMP0]]
+//
+vbfloat16m4_t test_vlmul_trunc_v_bf16m8_bf16m4(vbfloat16m8_t value) {
+  return __riscv_vlmul_trunc_v_bf16m8_bf16m4(value);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vloxei16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vloxei16.c
new file mode 100644
index 000000000000..147c820da9a6
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vloxei16.c
@@ -0,0 +1,141 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 1 x bfloat> @test_vloxei16_v_bf16mf4(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x bfloat> @llvm.riscv.vloxei.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> poison, ptr [[RS1]], <vscale x 1 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret <vscale x 1 x bfloat> [[TMP0]]
+//
+vbfloat16mf4_t test_vloxei16_v_bf16mf4(const __bf16 *rs1, vuint16mf4_t rs2,
+                                       size_t vl) {
+  return __riscv_vloxei16_v_bf16mf4(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 2 x bfloat> @test_vloxei16_v_bf16mf2(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x bfloat> @llvm.riscv.vloxei.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> poison, ptr [[RS1]], <vscale x 2 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret <vscale x 2 x bfloat> [[TMP0]]
+//
+vbfloat16mf2_t test_vloxei16_v_bf16mf2(const __bf16 *rs1, vuint16mf2_t rs2,
+                                       size_t vl) {
+  return __riscv_vloxei16_v_bf16mf2(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 4 x bfloat> @test_vloxei16_v_bf16m1(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x bfloat> @llvm.riscv.vloxei.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> poison, ptr [[RS1]], <vscale x 4 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret <vscale x 4 x bfloat> [[TMP0]]
+//
+vbfloat16m1_t test_vloxei16_v_bf16m1(const __bf16 *rs1, vuint16m1_t rs2,
+                                     size_t vl) {
+  return __riscv_vloxei16_v_bf16m1(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 8 x bfloat> @test_vloxei16_v_bf16m2(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x bfloat> @llvm.riscv.vloxei.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> poison, ptr [[RS1]], <vscale x 8 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret <vscale x 8 x bfloat> [[TMP0]]
+//
+vbfloat16m2_t test_vloxei16_v_bf16m2(const __bf16 *rs1, vuint16m2_t rs2,
+                                     size_t vl) {
+  return __riscv_vloxei16_v_bf16m2(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 16 x bfloat> @test_vloxei16_v_bf16m4(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 16 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x bfloat> @llvm.riscv.vloxei.nxv16bf16.nxv16i16.i64(<vscale x 16 x bfloat> poison, ptr [[RS1]], <vscale x 16 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret <vscale x 16 x bfloat> [[TMP0]]
+//
+vbfloat16m4_t test_vloxei16_v_bf16m4(const __bf16 *rs1, vuint16m4_t rs2,
+                                     size_t vl) {
+  return __riscv_vloxei16_v_bf16m4(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 32 x bfloat> @test_vloxei16_v_bf16m8(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 32 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x bfloat> @llvm.riscv.vloxei.nxv32bf16.nxv32i16.i64(<vscale x 32 x bfloat> poison, ptr [[RS1]], <vscale x 32 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret <vscale x 32 x bfloat> [[TMP0]]
+//
+vbfloat16m8_t test_vloxei16_v_bf16m8(const __bf16 *rs1, vuint16m8_t rs2,
+                                     size_t vl) {
+  return __riscv_vloxei16_v_bf16m8(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 1 x bfloat> @test_vloxei16_v_bf16mf4_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x bfloat> @llvm.riscv.vloxei.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> poison, ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret <vscale x 1 x bfloat> [[TMP0]]
+//
+vbfloat16mf4_t test_vloxei16_v_bf16mf4_m(vbool64_t vm, const __bf16 *rs1,
+                                         vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vloxei16_v_bf16mf4_m(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 2 x bfloat> @test_vloxei16_v_bf16mf2_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x bfloat> @llvm.riscv.vloxei.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> poison, ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret <vscale x 2 x bfloat> [[TMP0]]
+//
+vbfloat16mf2_t test_vloxei16_v_bf16mf2_m(vbool32_t vm, const __bf16 *rs1,
+                                         vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vloxei16_v_bf16mf2_m(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 4 x bfloat> @test_vloxei16_v_bf16m1_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x bfloat> @llvm.riscv.vloxei.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> poison, ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret <vscale x 4 x bfloat> [[TMP0]]
+//
+vbfloat16m1_t test_vloxei16_v_bf16m1_m(vbool16_t vm, const __bf16 *rs1,
+                                       vuint16m1_t rs2, size_t vl) {
+  return __riscv_vloxei16_v_bf16m1_m(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 8 x bfloat> @test_vloxei16_v_bf16m2_m(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x bfloat> @llvm.riscv.vloxei.mask.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> poison, ptr [[RS1]], <vscale x 8 x i16> [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret <vscale x 8 x bfloat> [[TMP0]]
+//
+vbfloat16m2_t test_vloxei16_v_bf16m2_m(vbool8_t vm, const __bf16 *rs1,
+                                       vuint16m2_t rs2, size_t vl) {
+  return __riscv_vloxei16_v_bf16m2_m(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 16 x bfloat> @test_vloxei16_v_bf16m4_m(
+// CHECK-RV64-SAME: <vscale x 16 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 16 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x bfloat> @llvm.riscv.vloxei.mask.nxv16bf16.nxv16i16.i64(<vscale x 16 x bfloat> poison, ptr [[RS1]], <vscale x 16 x i16> [[RS2]], <vscale x 16 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret <vscale x 16 x bfloat> [[TMP0]]
+//
+vbfloat16m4_t test_vloxei16_v_bf16m4_m(vbool4_t vm, const __bf16 *rs1,
+                                       vuint16m4_t rs2, size_t vl) {
+  return __riscv_vloxei16_v_bf16m4_m(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 32 x bfloat> @test_vloxei16_v_bf16m8_m(
+// CHECK-RV64-SAME: <vscale x 32 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 32 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x bfloat> @llvm.riscv.vloxei.mask.nxv32bf16.nxv32i16.i64(<vscale x 32 x bfloat> poison, ptr [[RS1]], <vscale x 32 x i16> [[RS2]], <vscale x 32 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret <vscale x 32 x bfloat> [[TMP0]]
+//
+vbfloat16m8_t test_vloxei16_v_bf16m8_m(vbool2_t vm, const __bf16 *rs1,
+                                       vuint16m8_t rs2, size_t vl) {
+  return __riscv_vloxei16_v_bf16m8_m(vm, rs1, rs2, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vloxseg2ei16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vloxseg2ei16.c
new file mode 100644
index 000000000000..c35a6eb68171
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vloxseg2ei16.c
@@ -0,0 +1,121 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vloxseg2ei16_v_bf16mf4x2(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vloxseg2.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], <vscale x 1 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x2_t test_vloxseg2ei16_v_bf16mf4x2(const __bf16 *rs1,
+                                               vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vloxseg2ei16_v_bf16mf4x2(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vloxseg2ei16_v_bf16mf2x2(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vloxseg2.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], <vscale x 2 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x2_t test_vloxseg2ei16_v_bf16mf2x2(const __bf16 *rs1,
+                                               vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vloxseg2ei16_v_bf16mf2x2(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vloxseg2ei16_v_bf16m1x2(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vloxseg2.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], <vscale x 4 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x2_t test_vloxseg2ei16_v_bf16m1x2(const __bf16 *rs1, vuint16m1_t rs2,
+                                             size_t vl) {
+  return __riscv_vloxseg2ei16_v_bf16m1x2(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vloxseg2ei16_v_bf16m2x2(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vloxseg2.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> poison, <vscale x 8 x bfloat> poison, ptr [[RS1]], <vscale x 8 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP0]]
+//
+vbfloat16m2x2_t test_vloxseg2ei16_v_bf16m2x2(const __bf16 *rs1, vuint16m2_t rs2,
+                                             size_t vl) {
+  return __riscv_vloxseg2ei16_v_bf16m2x2(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @test_vloxseg2ei16_v_bf16m4x2(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 16 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @llvm.riscv.vloxseg2.nxv16bf16.nxv16i16.i64(<vscale x 16 x bfloat> poison, <vscale x 16 x bfloat> poison, ptr [[RS1]], <vscale x 16 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[TMP0]]
+//
+vbfloat16m4x2_t test_vloxseg2ei16_v_bf16m4x2(const __bf16 *rs1, vuint16m4_t rs2,
+                                             size_t vl) {
+  return __riscv_vloxseg2ei16_v_bf16m4x2(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vloxseg2ei16_v_bf16mf4x2_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vloxseg2.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x2_t test_vloxseg2ei16_v_bf16mf4x2_m(vbool64_t vm,
+                                                 const __bf16 *rs1,
+                                                 vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vloxseg2ei16_v_bf16mf4x2_m(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vloxseg2ei16_v_bf16mf2x2_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vloxseg2.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x2_t test_vloxseg2ei16_v_bf16mf2x2_m(vbool32_t vm,
+                                                 const __bf16 *rs1,
+                                                 vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vloxseg2ei16_v_bf16mf2x2_m(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vloxseg2ei16_v_bf16m1x2_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vloxseg2.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x2_t test_vloxseg2ei16_v_bf16m1x2_m(vbool16_t vm, const __bf16 *rs1,
+                                               vuint16m1_t rs2, size_t vl) {
+  return __riscv_vloxseg2ei16_v_bf16m1x2_m(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vloxseg2ei16_v_bf16m2x2_m(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vloxseg2.mask.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> poison, <vscale x 8 x bfloat> poison, ptr [[RS1]], <vscale x 8 x i16> [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP0]]
+//
+vbfloat16m2x2_t test_vloxseg2ei16_v_bf16m2x2_m(vbool8_t vm, const __bf16 *rs1,
+                                               vuint16m2_t rs2, size_t vl) {
+  return __riscv_vloxseg2ei16_v_bf16m2x2_m(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @test_vloxseg2ei16_v_bf16m4x2_m(
+// CHECK-RV64-SAME: <vscale x 16 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 16 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @llvm.riscv.vloxseg2.mask.nxv16bf16.nxv16i16.i64(<vscale x 16 x bfloat> poison, <vscale x 16 x bfloat> poison, ptr [[RS1]], <vscale x 16 x i16> [[RS2]], <vscale x 16 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[TMP0]]
+//
+vbfloat16m4x2_t test_vloxseg2ei16_v_bf16m4x2_m(vbool4_t vm, const __bf16 *rs1,
+                                               vuint16m4_t rs2, size_t vl) {
+  return __riscv_vloxseg2ei16_v_bf16m4x2_m(vm, rs1, rs2, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vloxseg3ei16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vloxseg3ei16.c
new file mode 100644
index 000000000000..00c14b562bc5
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vloxseg3ei16.c
@@ -0,0 +1,99 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vloxseg3ei16_v_bf16mf4x3(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vloxseg3.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], <vscale x 1 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x3_t test_vloxseg3ei16_v_bf16mf4x3(const __bf16 *rs1,
+                                               vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vloxseg3ei16_v_bf16mf4x3(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vloxseg3ei16_v_bf16mf2x3(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vloxseg3.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], <vscale x 2 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x3_t test_vloxseg3ei16_v_bf16mf2x3(const __bf16 *rs1,
+                                               vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vloxseg3ei16_v_bf16mf2x3(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vloxseg3ei16_v_bf16m1x3(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vloxseg3.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], <vscale x 4 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x3_t test_vloxseg3ei16_v_bf16m1x3(const __bf16 *rs1, vuint16m1_t rs2,
+                                             size_t vl) {
+  return __riscv_vloxseg3ei16_v_bf16m1x3(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vloxseg3ei16_v_bf16m2x3(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vloxseg3.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> poison, <vscale x 8 x bfloat> poison, <vscale x 8 x bfloat> poison, ptr [[RS1]], <vscale x 8 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP0]]
+//
+vbfloat16m2x3_t test_vloxseg3ei16_v_bf16m2x3(const __bf16 *rs1, vuint16m2_t rs2,
+                                             size_t vl) {
+  return __riscv_vloxseg3ei16_v_bf16m2x3(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vloxseg3ei16_v_bf16mf4x3_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vloxseg3.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x3_t test_vloxseg3ei16_v_bf16mf4x3_m(vbool64_t vm,
+                                                 const __bf16 *rs1,
+                                                 vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vloxseg3ei16_v_bf16mf4x3_m(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vloxseg3ei16_v_bf16mf2x3_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vloxseg3.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x3_t test_vloxseg3ei16_v_bf16mf2x3_m(vbool32_t vm,
+                                                 const __bf16 *rs1,
+                                                 vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vloxseg3ei16_v_bf16mf2x3_m(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vloxseg3ei16_v_bf16m1x3_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vloxseg3.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x3_t test_vloxseg3ei16_v_bf16m1x3_m(vbool16_t vm, const __bf16 *rs1,
+                                               vuint16m1_t rs2, size_t vl) {
+  return __riscv_vloxseg3ei16_v_bf16m1x3_m(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vloxseg3ei16_v_bf16m2x3_m(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vloxseg3.mask.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> poison, <vscale x 8 x bfloat> poison, <vscale x 8 x bfloat> poison, ptr [[RS1]], <vscale x 8 x i16> [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP0]]
+//
+vbfloat16m2x3_t test_vloxseg3ei16_v_bf16m2x3_m(vbool8_t vm, const __bf16 *rs1,
+                                               vuint16m2_t rs2, size_t vl) {
+  return __riscv_vloxseg3ei16_v_bf16m2x3_m(vm, rs1, rs2, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vloxseg4ei16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vloxseg4ei16.c
new file mode 100644
index 000000000000..9289638807d6
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vloxseg4ei16.c
@@ -0,0 +1,99 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vloxseg4ei16_v_bf16mf4x4(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vloxseg4.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], <vscale x 1 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x4_t test_vloxseg4ei16_v_bf16mf4x4(const __bf16 *rs1,
+                                               vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vloxseg4ei16_v_bf16mf4x4(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vloxseg4ei16_v_bf16mf2x4(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vloxseg4.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], <vscale x 2 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x4_t test_vloxseg4ei16_v_bf16mf2x4(const __bf16 *rs1,
+                                               vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vloxseg4ei16_v_bf16mf2x4(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vloxseg4ei16_v_bf16m1x4(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vloxseg4.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], <vscale x 4 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x4_t test_vloxseg4ei16_v_bf16m1x4(const __bf16 *rs1, vuint16m1_t rs2,
+                                             size_t vl) {
+  return __riscv_vloxseg4ei16_v_bf16m1x4(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vloxseg4ei16_v_bf16m2x4(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vloxseg4.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> poison, <vscale x 8 x bfloat> poison, <vscale x 8 x bfloat> poison, <vscale x 8 x bfloat> poison, ptr [[RS1]], <vscale x 8 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP0]]
+//
+vbfloat16m2x4_t test_vloxseg4ei16_v_bf16m2x4(const __bf16 *rs1, vuint16m2_t rs2,
+                                             size_t vl) {
+  return __riscv_vloxseg4ei16_v_bf16m2x4(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vloxseg4ei16_v_bf16mf4x4_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vloxseg4.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x4_t test_vloxseg4ei16_v_bf16mf4x4_m(vbool64_t vm,
+                                                 const __bf16 *rs1,
+                                                 vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vloxseg4ei16_v_bf16mf4x4_m(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vloxseg4ei16_v_bf16mf2x4_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vloxseg4.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x4_t test_vloxseg4ei16_v_bf16mf2x4_m(vbool32_t vm,
+                                                 const __bf16 *rs1,
+                                                 vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vloxseg4ei16_v_bf16mf2x4_m(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vloxseg4ei16_v_bf16m1x4_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vloxseg4.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x4_t test_vloxseg4ei16_v_bf16m1x4_m(vbool16_t vm, const __bf16 *rs1,
+                                               vuint16m1_t rs2, size_t vl) {
+  return __riscv_vloxseg4ei16_v_bf16m1x4_m(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vloxseg4ei16_v_bf16m2x4_m(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vloxseg4.mask.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> poison, <vscale x 8 x bfloat> poison, <vscale x 8 x bfloat> poison, <vscale x 8 x bfloat> poison, ptr [[RS1]], <vscale x 8 x i16> [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP0]]
+//
+vbfloat16m2x4_t test_vloxseg4ei16_v_bf16m2x4_m(vbool8_t vm, const __bf16 *rs1,
+                                               vuint16m2_t rs2, size_t vl) {
+  return __riscv_vloxseg4ei16_v_bf16m2x4_m(vm, rs1, rs2, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vloxseg5ei16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vloxseg5ei16.c
new file mode 100644
index 000000000000..04c210d7a796
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vloxseg5ei16.c
@@ -0,0 +1,77 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vloxseg5ei16_v_bf16mf4x5(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vloxseg5.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], <vscale x 1 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x5_t test_vloxseg5ei16_v_bf16mf4x5(const __bf16 *rs1,
+                                               vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vloxseg5ei16_v_bf16mf4x5(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vloxseg5ei16_v_bf16mf2x5(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vloxseg5.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], <vscale x 2 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x5_t test_vloxseg5ei16_v_bf16mf2x5(const __bf16 *rs1,
+                                               vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vloxseg5ei16_v_bf16mf2x5(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vloxseg5ei16_v_bf16m1x5(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vloxseg5.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], <vscale x 4 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x5_t test_vloxseg5ei16_v_bf16m1x5(const __bf16 *rs1, vuint16m1_t rs2,
+                                             size_t vl) {
+  return __riscv_vloxseg5ei16_v_bf16m1x5(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vloxseg5ei16_v_bf16mf4x5_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vloxseg5.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x5_t test_vloxseg5ei16_v_bf16mf4x5_m(vbool64_t vm,
+                                                 const __bf16 *rs1,
+                                                 vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vloxseg5ei16_v_bf16mf4x5_m(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vloxseg5ei16_v_bf16mf2x5_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vloxseg5.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x5_t test_vloxseg5ei16_v_bf16mf2x5_m(vbool32_t vm,
+                                                 const __bf16 *rs1,
+                                                 vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vloxseg5ei16_v_bf16mf2x5_m(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vloxseg5ei16_v_bf16m1x5_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vloxseg5.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x5_t test_vloxseg5ei16_v_bf16m1x5_m(vbool16_t vm, const __bf16 *rs1,
+                                               vuint16m1_t rs2, size_t vl) {
+  return __riscv_vloxseg5ei16_v_bf16m1x5_m(vm, rs1, rs2, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vloxseg6ei16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vloxseg6ei16.c
new file mode 100644
index 000000000000..7ef2249630d9
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vloxseg6ei16.c
@@ -0,0 +1,77 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vloxseg6ei16_v_bf16mf4x6(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vloxseg6.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], <vscale x 1 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x6_t test_vloxseg6ei16_v_bf16mf4x6(const __bf16 *rs1,
+                                               vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vloxseg6ei16_v_bf16mf4x6(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vloxseg6ei16_v_bf16mf2x6(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vloxseg6.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], <vscale x 2 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x6_t test_vloxseg6ei16_v_bf16mf2x6(const __bf16 *rs1,
+                                               vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vloxseg6ei16_v_bf16mf2x6(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vloxseg6ei16_v_bf16m1x6(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vloxseg6.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], <vscale x 4 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x6_t test_vloxseg6ei16_v_bf16m1x6(const __bf16 *rs1, vuint16m1_t rs2,
+                                             size_t vl) {
+  return __riscv_vloxseg6ei16_v_bf16m1x6(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vloxseg6ei16_v_bf16mf4x6_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vloxseg6.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x6_t test_vloxseg6ei16_v_bf16mf4x6_m(vbool64_t vm,
+                                                 const __bf16 *rs1,
+                                                 vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vloxseg6ei16_v_bf16mf4x6_m(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vloxseg6ei16_v_bf16mf2x6_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vloxseg6.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x6_t test_vloxseg6ei16_v_bf16mf2x6_m(vbool32_t vm,
+                                                 const __bf16 *rs1,
+                                                 vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vloxseg6ei16_v_bf16mf2x6_m(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vloxseg6ei16_v_bf16m1x6_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vloxseg6.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x6_t test_vloxseg6ei16_v_bf16m1x6_m(vbool16_t vm, const __bf16 *rs1,
+                                               vuint16m1_t rs2, size_t vl) {
+  return __riscv_vloxseg6ei16_v_bf16m1x6_m(vm, rs1, rs2, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vloxseg7ei16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vloxseg7ei16.c
new file mode 100644
index 000000000000..b1d5ee604018
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vloxseg7ei16.c
@@ -0,0 +1,77 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vloxseg7ei16_v_bf16mf4x7(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vloxseg7.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], <vscale x 1 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x7_t test_vloxseg7ei16_v_bf16mf4x7(const __bf16 *rs1,
+                                               vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vloxseg7ei16_v_bf16mf4x7(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vloxseg7ei16_v_bf16mf2x7(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vloxseg7.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], <vscale x 2 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x7_t test_vloxseg7ei16_v_bf16mf2x7(const __bf16 *rs1,
+                                               vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vloxseg7ei16_v_bf16mf2x7(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vloxseg7ei16_v_bf16m1x7(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vloxseg7.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], <vscale x 4 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x7_t test_vloxseg7ei16_v_bf16m1x7(const __bf16 *rs1, vuint16m1_t rs2,
+                                             size_t vl) {
+  return __riscv_vloxseg7ei16_v_bf16m1x7(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vloxseg7ei16_v_bf16mf4x7_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vloxseg7.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x7_t test_vloxseg7ei16_v_bf16mf4x7_m(vbool64_t vm,
+                                                 const __bf16 *rs1,
+                                                 vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vloxseg7ei16_v_bf16mf4x7_m(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vloxseg7ei16_v_bf16mf2x7_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vloxseg7.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x7_t test_vloxseg7ei16_v_bf16mf2x7_m(vbool32_t vm,
+                                                 const __bf16 *rs1,
+                                                 vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vloxseg7ei16_v_bf16mf2x7_m(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vloxseg7ei16_v_bf16m1x7_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vloxseg7.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x7_t test_vloxseg7ei16_v_bf16m1x7_m(vbool16_t vm, const __bf16 *rs1,
+                                               vuint16m1_t rs2, size_t vl) {
+  return __riscv_vloxseg7ei16_v_bf16m1x7_m(vm, rs1, rs2, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vloxseg8ei16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vloxseg8ei16.c
new file mode 100644
index 000000000000..1e1c762e0d0b
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vloxseg8ei16.c
@@ -0,0 +1,77 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vloxseg8ei16_v_bf16mf4x8(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vloxseg8.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], <vscale x 1 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x8_t test_vloxseg8ei16_v_bf16mf4x8(const __bf16 *rs1,
+                                               vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vloxseg8ei16_v_bf16mf4x8(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vloxseg8ei16_v_bf16mf2x8(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vloxseg8.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], <vscale x 2 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x8_t test_vloxseg8ei16_v_bf16mf2x8(const __bf16 *rs1,
+                                               vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vloxseg8ei16_v_bf16mf2x8(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vloxseg8ei16_v_bf16m1x8(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vloxseg8.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], <vscale x 4 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x8_t test_vloxseg8ei16_v_bf16m1x8(const __bf16 *rs1, vuint16m1_t rs2,
+                                             size_t vl) {
+  return __riscv_vloxseg8ei16_v_bf16m1x8(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vloxseg8ei16_v_bf16mf4x8_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vloxseg8.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x8_t test_vloxseg8ei16_v_bf16mf4x8_m(vbool64_t vm,
+                                                 const __bf16 *rs1,
+                                                 vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vloxseg8ei16_v_bf16mf4x8_m(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vloxseg8ei16_v_bf16mf2x8_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vloxseg8.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x8_t test_vloxseg8ei16_v_bf16mf2x8_m(vbool32_t vm,
+                                                 const __bf16 *rs1,
+                                                 vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vloxseg8ei16_v_bf16mf2x8_m(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vloxseg8ei16_v_bf16m1x8_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vloxseg8.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x8_t test_vloxseg8ei16_v_bf16m1x8_m(vbool16_t vm, const __bf16 *rs1,
+                                               vuint16m1_t rs2, size_t vl) {
+  return __riscv_vloxseg8ei16_v_bf16m1x8_m(vm, rs1, rs2, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vlse16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vlse16.c
new file mode 100644
index 000000000000..c5c98f3abc52
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vlse16.c
@@ -0,0 +1,141 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 1 x bfloat> @test_vlse16_v_bf16mf4(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x bfloat> @llvm.riscv.vlse.nxv1bf16.i64(<vscale x 1 x bfloat> poison, ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret <vscale x 1 x bfloat> [[TMP0]]
+//
+vbfloat16mf4_t test_vlse16_v_bf16mf4(const __bf16 *rs1, ptrdiff_t rs2,
+                                     size_t vl) {
+  return __riscv_vlse16_v_bf16mf4(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 2 x bfloat> @test_vlse16_v_bf16mf2(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x bfloat> @llvm.riscv.vlse.nxv2bf16.i64(<vscale x 2 x bfloat> poison, ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret <vscale x 2 x bfloat> [[TMP0]]
+//
+vbfloat16mf2_t test_vlse16_v_bf16mf2(const __bf16 *rs1, ptrdiff_t rs2,
+                                     size_t vl) {
+  return __riscv_vlse16_v_bf16mf2(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 4 x bfloat> @test_vlse16_v_bf16m1(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x bfloat> @llvm.riscv.vlse.nxv4bf16.i64(<vscale x 4 x bfloat> poison, ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret <vscale x 4 x bfloat> [[TMP0]]
+//
+vbfloat16m1_t test_vlse16_v_bf16m1(const __bf16 *rs1, ptrdiff_t rs2,
+                                   size_t vl) {
+  return __riscv_vlse16_v_bf16m1(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 8 x bfloat> @test_vlse16_v_bf16m2(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x bfloat> @llvm.riscv.vlse.nxv8bf16.i64(<vscale x 8 x bfloat> poison, ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret <vscale x 8 x bfloat> [[TMP0]]
+//
+vbfloat16m2_t test_vlse16_v_bf16m2(const __bf16 *rs1, ptrdiff_t rs2,
+                                   size_t vl) {
+  return __riscv_vlse16_v_bf16m2(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 16 x bfloat> @test_vlse16_v_bf16m4(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x bfloat> @llvm.riscv.vlse.nxv16bf16.i64(<vscale x 16 x bfloat> poison, ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret <vscale x 16 x bfloat> [[TMP0]]
+//
+vbfloat16m4_t test_vlse16_v_bf16m4(const __bf16 *rs1, ptrdiff_t rs2,
+                                   size_t vl) {
+  return __riscv_vlse16_v_bf16m4(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 32 x bfloat> @test_vlse16_v_bf16m8(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x bfloat> @llvm.riscv.vlse.nxv32bf16.i64(<vscale x 32 x bfloat> poison, ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret <vscale x 32 x bfloat> [[TMP0]]
+//
+vbfloat16m8_t test_vlse16_v_bf16m8(const __bf16 *rs1, ptrdiff_t rs2,
+                                   size_t vl) {
+  return __riscv_vlse16_v_bf16m8(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 1 x bfloat> @test_vlse16_v_bf16mf4_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x bfloat> @llvm.riscv.vlse.mask.nxv1bf16.i64(<vscale x 1 x bfloat> poison, ptr [[RS1]], i64 [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret <vscale x 1 x bfloat> [[TMP0]]
+//
+vbfloat16mf4_t test_vlse16_v_bf16mf4_m(vbool64_t vm, const __bf16 *rs1,
+                                       ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlse16_v_bf16mf4_m(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 2 x bfloat> @test_vlse16_v_bf16mf2_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x bfloat> @llvm.riscv.vlse.mask.nxv2bf16.i64(<vscale x 2 x bfloat> poison, ptr [[RS1]], i64 [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret <vscale x 2 x bfloat> [[TMP0]]
+//
+vbfloat16mf2_t test_vlse16_v_bf16mf2_m(vbool32_t vm, const __bf16 *rs1,
+                                       ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlse16_v_bf16mf2_m(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 4 x bfloat> @test_vlse16_v_bf16m1_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x bfloat> @llvm.riscv.vlse.mask.nxv4bf16.i64(<vscale x 4 x bfloat> poison, ptr [[RS1]], i64 [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret <vscale x 4 x bfloat> [[TMP0]]
+//
+vbfloat16m1_t test_vlse16_v_bf16m1_m(vbool16_t vm, const __bf16 *rs1,
+                                     ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlse16_v_bf16m1_m(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 8 x bfloat> @test_vlse16_v_bf16m2_m(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x bfloat> @llvm.riscv.vlse.mask.nxv8bf16.i64(<vscale x 8 x bfloat> poison, ptr [[RS1]], i64 [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret <vscale x 8 x bfloat> [[TMP0]]
+//
+vbfloat16m2_t test_vlse16_v_bf16m2_m(vbool8_t vm, const __bf16 *rs1,
+                                     ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlse16_v_bf16m2_m(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 16 x bfloat> @test_vlse16_v_bf16m4_m(
+// CHECK-RV64-SAME: <vscale x 16 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x bfloat> @llvm.riscv.vlse.mask.nxv16bf16.i64(<vscale x 16 x bfloat> poison, ptr [[RS1]], i64 [[RS2]], <vscale x 16 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret <vscale x 16 x bfloat> [[TMP0]]
+//
+vbfloat16m4_t test_vlse16_v_bf16m4_m(vbool4_t vm, const __bf16 *rs1,
+                                     ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlse16_v_bf16m4_m(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 32 x bfloat> @test_vlse16_v_bf16m8_m(
+// CHECK-RV64-SAME: <vscale x 32 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x bfloat> @llvm.riscv.vlse.mask.nxv32bf16.i64(<vscale x 32 x bfloat> poison, ptr [[RS1]], i64 [[RS2]], <vscale x 32 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret <vscale x 32 x bfloat> [[TMP0]]
+//
+vbfloat16m8_t test_vlse16_v_bf16m8_m(vbool2_t vm, const __bf16 *rs1,
+                                     ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlse16_v_bf16m8_m(vm, rs1, rs2, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vlseg2e16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vlseg2e16.c
new file mode 100644
index 000000000000..6ea8a1b4a429
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vlseg2e16.c
@@ -0,0 +1,114 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg2e16_v_bf16mf4x2(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlseg2.nxv1bf16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x2_t test_vlseg2e16_v_bf16mf4x2(const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg2e16_v_bf16mf4x2(rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg2e16_v_bf16mf2x2(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlseg2.nxv2bf16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x2_t test_vlseg2e16_v_bf16mf2x2(const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg2e16_v_bf16mf2x2(rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg2e16_v_bf16m1x2(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlseg2.nxv4bf16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x2_t test_vlseg2e16_v_bf16m1x2(const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg2e16_v_bf16m1x2(rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vlseg2e16_v_bf16m2x2(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vlseg2.nxv8bf16.i64(<vscale x 8 x bfloat> poison, <vscale x 8 x bfloat> poison, ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP0]]
+//
+vbfloat16m2x2_t test_vlseg2e16_v_bf16m2x2(const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg2e16_v_bf16m2x2(rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @test_vlseg2e16_v_bf16m4x2(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @llvm.riscv.vlseg2.nxv16bf16.i64(<vscale x 16 x bfloat> poison, <vscale x 16 x bfloat> poison, ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[TMP0]]
+//
+vbfloat16m4x2_t test_vlseg2e16_v_bf16m4x2(const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg2e16_v_bf16m4x2(rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg2e16_v_bf16mf4x2_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlseg2.mask.nxv1bf16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x2_t test_vlseg2e16_v_bf16mf4x2_m(vbool64_t vm, const __bf16 *rs1,
+                                              size_t vl) {
+  return __riscv_vlseg2e16_v_bf16mf4x2_m(vm, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg2e16_v_bf16mf2x2_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlseg2.mask.nxv2bf16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x2_t test_vlseg2e16_v_bf16mf2x2_m(vbool32_t vm, const __bf16 *rs1,
+                                              size_t vl) {
+  return __riscv_vlseg2e16_v_bf16mf2x2_m(vm, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg2e16_v_bf16m1x2_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlseg2.mask.nxv4bf16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x2_t test_vlseg2e16_v_bf16m1x2_m(vbool16_t vm, const __bf16 *rs1,
+                                            size_t vl) {
+  return __riscv_vlseg2e16_v_bf16m1x2_m(vm, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vlseg2e16_v_bf16m2x2_m(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vlseg2.mask.nxv8bf16.i64(<vscale x 8 x bfloat> poison, <vscale x 8 x bfloat> poison, ptr [[RS1]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP0]]
+//
+vbfloat16m2x2_t test_vlseg2e16_v_bf16m2x2_m(vbool8_t vm, const __bf16 *rs1,
+                                            size_t vl) {
+  return __riscv_vlseg2e16_v_bf16m2x2_m(vm, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @test_vlseg2e16_v_bf16m4x2_m(
+// CHECK-RV64-SAME: <vscale x 16 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @llvm.riscv.vlseg2.mask.nxv16bf16.i64(<vscale x 16 x bfloat> poison, <vscale x 16 x bfloat> poison, ptr [[RS1]], <vscale x 16 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[TMP0]]
+//
+vbfloat16m4x2_t test_vlseg2e16_v_bf16m4x2_m(vbool4_t vm, const __bf16 *rs1,
+                                            size_t vl) {
+  return __riscv_vlseg2e16_v_bf16m4x2_m(vm, rs1, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vlseg2e16ff.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vlseg2e16ff.c
new file mode 100644
index 000000000000..cecf6f115394
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vlseg2e16ff.c
@@ -0,0 +1,179 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg2e16ff_v_bf16mf4x2(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } @llvm.riscv.vlseg2ff.nxv1bf16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } poison, <vscale x 1 x bfloat> [[TMP1]], 0
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 2
+// CHECK-RV64-NEXT:    store i64 [[TMP5]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP4]]
+//
+vbfloat16mf4x2_t test_vlseg2e16ff_v_bf16mf4x2(const __bf16 *rs1, size_t *new_vl,
+                                              size_t vl) {
+  return __riscv_vlseg2e16ff_v_bf16mf4x2(rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg2e16ff_v_bf16mf2x2(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } @llvm.riscv.vlseg2ff.nxv2bf16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } poison, <vscale x 2 x bfloat> [[TMP1]], 0
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 2
+// CHECK-RV64-NEXT:    store i64 [[TMP5]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP4]]
+//
+vbfloat16mf2x2_t test_vlseg2e16ff_v_bf16mf2x2(const __bf16 *rs1, size_t *new_vl,
+                                              size_t vl) {
+  return __riscv_vlseg2e16ff_v_bf16mf2x2(rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg2e16ff_v_bf16m1x2(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } @llvm.riscv.vlseg2ff.nxv4bf16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } poison, <vscale x 4 x bfloat> [[TMP1]], 0
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 2
+// CHECK-RV64-NEXT:    store i64 [[TMP5]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP4]]
+//
+vbfloat16m1x2_t test_vlseg2e16ff_v_bf16m1x2(const __bf16 *rs1, size_t *new_vl,
+                                            size_t vl) {
+  return __riscv_vlseg2e16ff_v_bf16m1x2(rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vlseg2e16ff_v_bf16m2x2(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } @llvm.riscv.vlseg2ff.nxv8bf16.i64(<vscale x 8 x bfloat> poison, <vscale x 8 x bfloat> poison, ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } poison, <vscale x 8 x bfloat> [[TMP1]], 0
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP2]], <vscale x 8 x bfloat> [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP0]], 2
+// CHECK-RV64-NEXT:    store i64 [[TMP5]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP4]]
+//
+vbfloat16m2x2_t test_vlseg2e16ff_v_bf16m2x2(const __bf16 *rs1, size_t *new_vl,
+                                            size_t vl) {
+  return __riscv_vlseg2e16ff_v_bf16m2x2(rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @test_vlseg2e16ff_v_bf16m4x2(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 16 x bfloat>, <vscale x 16 x bfloat>, i64 } @llvm.riscv.vlseg2ff.nxv16bf16.i64(<vscale x 16 x bfloat> poison, <vscale x 16 x bfloat> poison, ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = insertvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } poison, <vscale x 16 x bfloat> [[TMP1]], 0
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[TMP2]], <vscale x 16 x bfloat> [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat>, i64 } [[TMP0]], 2
+// CHECK-RV64-NEXT:    store i64 [[TMP5]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[TMP4]]
+//
+vbfloat16m4x2_t test_vlseg2e16ff_v_bf16m4x2(const __bf16 *rs1, size_t *new_vl,
+                                            size_t vl) {
+  return __riscv_vlseg2e16ff_v_bf16m4x2(rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg2e16ff_v_bf16mf4x2_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } @llvm.riscv.vlseg2ff.mask.nxv1bf16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } poison, <vscale x 1 x bfloat> [[TMP1]], 0
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 2
+// CHECK-RV64-NEXT:    store i64 [[TMP5]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP4]]
+//
+vbfloat16mf4x2_t test_vlseg2e16ff_v_bf16mf4x2_m(vbool64_t vm, const __bf16 *rs1,
+                                                size_t *new_vl, size_t vl) {
+  return __riscv_vlseg2e16ff_v_bf16mf4x2_m(vm, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg2e16ff_v_bf16mf2x2_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } @llvm.riscv.vlseg2ff.mask.nxv2bf16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } poison, <vscale x 2 x bfloat> [[TMP1]], 0
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 2
+// CHECK-RV64-NEXT:    store i64 [[TMP5]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP4]]
+//
+vbfloat16mf2x2_t test_vlseg2e16ff_v_bf16mf2x2_m(vbool32_t vm, const __bf16 *rs1,
+                                                size_t *new_vl, size_t vl) {
+  return __riscv_vlseg2e16ff_v_bf16mf2x2_m(vm, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg2e16ff_v_bf16m1x2_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } @llvm.riscv.vlseg2ff.mask.nxv4bf16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } poison, <vscale x 4 x bfloat> [[TMP1]], 0
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 2
+// CHECK-RV64-NEXT:    store i64 [[TMP5]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP4]]
+//
+vbfloat16m1x2_t test_vlseg2e16ff_v_bf16m1x2_m(vbool16_t vm, const __bf16 *rs1,
+                                              size_t *new_vl, size_t vl) {
+  return __riscv_vlseg2e16ff_v_bf16m1x2_m(vm, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vlseg2e16ff_v_bf16m2x2_m(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } @llvm.riscv.vlseg2ff.mask.nxv8bf16.i64(<vscale x 8 x bfloat> poison, <vscale x 8 x bfloat> poison, ptr [[RS1]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } poison, <vscale x 8 x bfloat> [[TMP1]], 0
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP2]], <vscale x 8 x bfloat> [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP0]], 2
+// CHECK-RV64-NEXT:    store i64 [[TMP5]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP4]]
+//
+vbfloat16m2x2_t test_vlseg2e16ff_v_bf16m2x2_m(vbool8_t vm, const __bf16 *rs1,
+                                              size_t *new_vl, size_t vl) {
+  return __riscv_vlseg2e16ff_v_bf16m2x2_m(vm, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @test_vlseg2e16ff_v_bf16m4x2_m(
+// CHECK-RV64-SAME: <vscale x 16 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 16 x bfloat>, <vscale x 16 x bfloat>, i64 } @llvm.riscv.vlseg2ff.mask.nxv16bf16.i64(<vscale x 16 x bfloat> poison, <vscale x 16 x bfloat> poison, ptr [[RS1]], <vscale x 16 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = insertvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } poison, <vscale x 16 x bfloat> [[TMP1]], 0
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[TMP2]], <vscale x 16 x bfloat> [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat>, i64 } [[TMP0]], 2
+// CHECK-RV64-NEXT:    store i64 [[TMP5]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[TMP4]]
+//
+vbfloat16m4x2_t test_vlseg2e16ff_v_bf16m4x2_m(vbool4_t vm, const __bf16 *rs1,
+                                              size_t *new_vl, size_t vl) {
+  return __riscv_vlseg2e16ff_v_bf16m4x2_m(vm, rs1, new_vl, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vlseg3e16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vlseg3e16.c
new file mode 100644
index 000000000000..2c8d679d8d63
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vlseg3e16.c
@@ -0,0 +1,93 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg3e16_v_bf16mf4x3(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlseg3.nxv1bf16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x3_t test_vlseg3e16_v_bf16mf4x3(const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg3e16_v_bf16mf4x3(rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg3e16_v_bf16mf2x3(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlseg3.nxv2bf16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x3_t test_vlseg3e16_v_bf16mf2x3(const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg3e16_v_bf16mf2x3(rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg3e16_v_bf16m1x3(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlseg3.nxv4bf16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x3_t test_vlseg3e16_v_bf16m1x3(const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg3e16_v_bf16m1x3(rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vlseg3e16_v_bf16m2x3(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vlseg3.nxv8bf16.i64(<vscale x 8 x bfloat> poison, <vscale x 8 x bfloat> poison, <vscale x 8 x bfloat> poison, ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP0]]
+//
+vbfloat16m2x3_t test_vlseg3e16_v_bf16m2x3(const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg3e16_v_bf16m2x3(rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg3e16_v_bf16mf4x3_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlseg3.mask.nxv1bf16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x3_t test_vlseg3e16_v_bf16mf4x3_m(vbool64_t vm, const __bf16 *rs1,
+                                              size_t vl) {
+  return __riscv_vlseg3e16_v_bf16mf4x3_m(vm, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg3e16_v_bf16mf2x3_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlseg3.mask.nxv2bf16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x3_t test_vlseg3e16_v_bf16mf2x3_m(vbool32_t vm, const __bf16 *rs1,
+                                              size_t vl) {
+  return __riscv_vlseg3e16_v_bf16mf2x3_m(vm, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg3e16_v_bf16m1x3_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlseg3.mask.nxv4bf16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x3_t test_vlseg3e16_v_bf16m1x3_m(vbool16_t vm, const __bf16 *rs1,
+                                            size_t vl) {
+  return __riscv_vlseg3e16_v_bf16m1x3_m(vm, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vlseg3e16_v_bf16m2x3_m(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vlseg3.mask.nxv8bf16.i64(<vscale x 8 x bfloat> poison, <vscale x 8 x bfloat> poison, <vscale x 8 x bfloat> poison, ptr [[RS1]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP0]]
+//
+vbfloat16m2x3_t test_vlseg3e16_v_bf16m2x3_m(vbool8_t vm, const __bf16 *rs1,
+                                            size_t vl) {
+  return __riscv_vlseg3e16_v_bf16m2x3_m(vm, rs1, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vlseg3e16ff.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vlseg3e16ff.c
new file mode 100644
index 000000000000..e7660a74c93d
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vlseg3e16ff.c
@@ -0,0 +1,161 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg3e16ff_v_bf16mf4x3(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } @llvm.riscv.vlseg3ff.nxv1bf16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } poison, <vscale x 1 x bfloat> [[TMP1]], 0
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 2
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], 2
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 3
+// CHECK-RV64-NEXT:    store i64 [[TMP7]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP6]]
+//
+vbfloat16mf4x3_t test_vlseg3e16ff_v_bf16mf4x3(const __bf16 *rs1, size_t *new_vl,
+                                              size_t vl) {
+  return __riscv_vlseg3e16ff_v_bf16mf4x3(rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg3e16ff_v_bf16mf2x3(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } @llvm.riscv.vlseg3ff.nxv2bf16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } poison, <vscale x 2 x bfloat> [[TMP1]], 0
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 2
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], 2
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 3
+// CHECK-RV64-NEXT:    store i64 [[TMP7]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP6]]
+//
+vbfloat16mf2x3_t test_vlseg3e16ff_v_bf16mf2x3(const __bf16 *rs1, size_t *new_vl,
+                                              size_t vl) {
+  return __riscv_vlseg3e16ff_v_bf16mf2x3(rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg3e16ff_v_bf16m1x3(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } @llvm.riscv.vlseg3ff.nxv4bf16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } poison, <vscale x 4 x bfloat> [[TMP1]], 0
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 2
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], 2
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 3
+// CHECK-RV64-NEXT:    store i64 [[TMP7]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP6]]
+//
+vbfloat16m1x3_t test_vlseg3e16ff_v_bf16m1x3(const __bf16 *rs1, size_t *new_vl,
+                                            size_t vl) {
+  return __riscv_vlseg3e16ff_v_bf16m1x3(rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vlseg3e16ff_v_bf16m2x3(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } @llvm.riscv.vlseg3ff.nxv8bf16.i64(<vscale x 8 x bfloat> poison, <vscale x 8 x bfloat> poison, <vscale x 8 x bfloat> poison, ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } poison, <vscale x 8 x bfloat> [[TMP1]], 0
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP2]], <vscale x 8 x bfloat> [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP0]], 2
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP4]], <vscale x 8 x bfloat> [[TMP5]], 2
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP0]], 3
+// CHECK-RV64-NEXT:    store i64 [[TMP7]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP6]]
+//
+vbfloat16m2x3_t test_vlseg3e16ff_v_bf16m2x3(const __bf16 *rs1, size_t *new_vl,
+                                            size_t vl) {
+  return __riscv_vlseg3e16ff_v_bf16m2x3(rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg3e16ff_v_bf16mf4x3_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } @llvm.riscv.vlseg3ff.mask.nxv1bf16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } poison, <vscale x 1 x bfloat> [[TMP1]], 0
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 2
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], 2
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 3
+// CHECK-RV64-NEXT:    store i64 [[TMP7]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP6]]
+//
+vbfloat16mf4x3_t test_vlseg3e16ff_v_bf16mf4x3_m(vbool64_t vm, const __bf16 *rs1,
+                                                size_t *new_vl, size_t vl) {
+  return __riscv_vlseg3e16ff_v_bf16mf4x3_m(vm, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg3e16ff_v_bf16mf2x3_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } @llvm.riscv.vlseg3ff.mask.nxv2bf16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } poison, <vscale x 2 x bfloat> [[TMP1]], 0
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 2
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], 2
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 3
+// CHECK-RV64-NEXT:    store i64 [[TMP7]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP6]]
+//
+vbfloat16mf2x3_t test_vlseg3e16ff_v_bf16mf2x3_m(vbool32_t vm, const __bf16 *rs1,
+                                                size_t *new_vl, size_t vl) {
+  return __riscv_vlseg3e16ff_v_bf16mf2x3_m(vm, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg3e16ff_v_bf16m1x3_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } @llvm.riscv.vlseg3ff.mask.nxv4bf16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } poison, <vscale x 4 x bfloat> [[TMP1]], 0
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 2
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], 2
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 3
+// CHECK-RV64-NEXT:    store i64 [[TMP7]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP6]]
+//
+vbfloat16m1x3_t test_vlseg3e16ff_v_bf16m1x3_m(vbool16_t vm, const __bf16 *rs1,
+                                              size_t *new_vl, size_t vl) {
+  return __riscv_vlseg3e16ff_v_bf16m1x3_m(vm, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vlseg3e16ff_v_bf16m2x3_m(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } @llvm.riscv.vlseg3ff.mask.nxv8bf16.i64(<vscale x 8 x bfloat> poison, <vscale x 8 x bfloat> poison, <vscale x 8 x bfloat> poison, ptr [[RS1]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } poison, <vscale x 8 x bfloat> [[TMP1]], 0
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP2]], <vscale x 8 x bfloat> [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP0]], 2
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP4]], <vscale x 8 x bfloat> [[TMP5]], 2
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP0]], 3
+// CHECK-RV64-NEXT:    store i64 [[TMP7]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP6]]
+//
+vbfloat16m2x3_t test_vlseg3e16ff_v_bf16m2x3_m(vbool8_t vm, const __bf16 *rs1,
+                                              size_t *new_vl, size_t vl) {
+  return __riscv_vlseg3e16ff_v_bf16m2x3_m(vm, rs1, new_vl, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vlseg4e16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vlseg4e16.c
new file mode 100644
index 000000000000..c7480e10028a
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vlseg4e16.c
@@ -0,0 +1,93 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg4e16_v_bf16mf4x4(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlseg4.nxv1bf16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x4_t test_vlseg4e16_v_bf16mf4x4(const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg4e16_v_bf16mf4x4(rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg4e16_v_bf16mf2x4(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlseg4.nxv2bf16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x4_t test_vlseg4e16_v_bf16mf2x4(const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg4e16_v_bf16mf2x4(rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg4e16_v_bf16m1x4(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlseg4.nxv4bf16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x4_t test_vlseg4e16_v_bf16m1x4(const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg4e16_v_bf16m1x4(rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vlseg4e16_v_bf16m2x4(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vlseg4.nxv8bf16.i64(<vscale x 8 x bfloat> poison, <vscale x 8 x bfloat> poison, <vscale x 8 x bfloat> poison, <vscale x 8 x bfloat> poison, ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP0]]
+//
+vbfloat16m2x4_t test_vlseg4e16_v_bf16m2x4(const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg4e16_v_bf16m2x4(rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg4e16_v_bf16mf4x4_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlseg4.mask.nxv1bf16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x4_t test_vlseg4e16_v_bf16mf4x4_m(vbool64_t vm, const __bf16 *rs1,
+                                              size_t vl) {
+  return __riscv_vlseg4e16_v_bf16mf4x4_m(vm, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg4e16_v_bf16mf2x4_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlseg4.mask.nxv2bf16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x4_t test_vlseg4e16_v_bf16mf2x4_m(vbool32_t vm, const __bf16 *rs1,
+                                              size_t vl) {
+  return __riscv_vlseg4e16_v_bf16mf2x4_m(vm, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg4e16_v_bf16m1x4_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlseg4.mask.nxv4bf16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x4_t test_vlseg4e16_v_bf16m1x4_m(vbool16_t vm, const __bf16 *rs1,
+                                            size_t vl) {
+  return __riscv_vlseg4e16_v_bf16m1x4_m(vm, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vlseg4e16_v_bf16m2x4_m(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vlseg4.mask.nxv8bf16.i64(<vscale x 8 x bfloat> poison, <vscale x 8 x bfloat> poison, <vscale x 8 x bfloat> poison, <vscale x 8 x bfloat> poison, ptr [[RS1]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP0]]
+//
+vbfloat16m2x4_t test_vlseg4e16_v_bf16m2x4_m(vbool8_t vm, const __bf16 *rs1,
+                                            size_t vl) {
+  return __riscv_vlseg4e16_v_bf16m2x4_m(vm, rs1, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vlseg4e16ff.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vlseg4e16ff.c
new file mode 100644
index 000000000000..de1e40c98729
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vlseg4e16ff.c
@@ -0,0 +1,177 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg4e16ff_v_bf16mf4x4(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } @llvm.riscv.vlseg4ff.nxv1bf16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } poison, <vscale x 1 x bfloat> [[TMP1]], 0
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 2
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], 2
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 3
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP6]], <vscale x 1 x bfloat> [[TMP7]], 3
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 4
+// CHECK-RV64-NEXT:    store i64 [[TMP9]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP8]]
+//
+vbfloat16mf4x4_t test_vlseg4e16ff_v_bf16mf4x4(const __bf16 *rs1, size_t *new_vl,
+                                              size_t vl) {
+  return __riscv_vlseg4e16ff_v_bf16mf4x4(rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg4e16ff_v_bf16mf2x4(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } @llvm.riscv.vlseg4ff.nxv2bf16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } poison, <vscale x 2 x bfloat> [[TMP1]], 0
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 2
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], 2
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 3
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP6]], <vscale x 2 x bfloat> [[TMP7]], 3
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 4
+// CHECK-RV64-NEXT:    store i64 [[TMP9]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP8]]
+//
+vbfloat16mf2x4_t test_vlseg4e16ff_v_bf16mf2x4(const __bf16 *rs1, size_t *new_vl,
+                                              size_t vl) {
+  return __riscv_vlseg4e16ff_v_bf16mf2x4(rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg4e16ff_v_bf16m1x4(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } @llvm.riscv.vlseg4ff.nxv4bf16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } poison, <vscale x 4 x bfloat> [[TMP1]], 0
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 2
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], 2
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 3
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP6]], <vscale x 4 x bfloat> [[TMP7]], 3
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 4
+// CHECK-RV64-NEXT:    store i64 [[TMP9]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP8]]
+//
+vbfloat16m1x4_t test_vlseg4e16ff_v_bf16m1x4(const __bf16 *rs1, size_t *new_vl,
+                                            size_t vl) {
+  return __riscv_vlseg4e16ff_v_bf16m1x4(rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vlseg4e16ff_v_bf16m2x4(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } @llvm.riscv.vlseg4ff.nxv8bf16.i64(<vscale x 8 x bfloat> poison, <vscale x 8 x bfloat> poison, <vscale x 8 x bfloat> poison, <vscale x 8 x bfloat> poison, ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } poison, <vscale x 8 x bfloat> [[TMP1]], 0
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP2]], <vscale x 8 x bfloat> [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP0]], 2
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP4]], <vscale x 8 x bfloat> [[TMP5]], 2
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP0]], 3
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP6]], <vscale x 8 x bfloat> [[TMP7]], 3
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP0]], 4
+// CHECK-RV64-NEXT:    store i64 [[TMP9]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP8]]
+//
+vbfloat16m2x4_t test_vlseg4e16ff_v_bf16m2x4(const __bf16 *rs1, size_t *new_vl,
+                                            size_t vl) {
+  return __riscv_vlseg4e16ff_v_bf16m2x4(rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg4e16ff_v_bf16mf4x4_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } @llvm.riscv.vlseg4ff.mask.nxv1bf16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } poison, <vscale x 1 x bfloat> [[TMP1]], 0
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 2
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], 2
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 3
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP6]], <vscale x 1 x bfloat> [[TMP7]], 3
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 4
+// CHECK-RV64-NEXT:    store i64 [[TMP9]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP8]]
+//
+vbfloat16mf4x4_t test_vlseg4e16ff_v_bf16mf4x4_m(vbool64_t vm, const __bf16 *rs1,
+                                                size_t *new_vl, size_t vl) {
+  return __riscv_vlseg4e16ff_v_bf16mf4x4_m(vm, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg4e16ff_v_bf16mf2x4_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } @llvm.riscv.vlseg4ff.mask.nxv2bf16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } poison, <vscale x 2 x bfloat> [[TMP1]], 0
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 2
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], 2
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 3
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP6]], <vscale x 2 x bfloat> [[TMP7]], 3
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 4
+// CHECK-RV64-NEXT:    store i64 [[TMP9]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP8]]
+//
+vbfloat16mf2x4_t test_vlseg4e16ff_v_bf16mf2x4_m(vbool32_t vm, const __bf16 *rs1,
+                                                size_t *new_vl, size_t vl) {
+  return __riscv_vlseg4e16ff_v_bf16mf2x4_m(vm, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg4e16ff_v_bf16m1x4_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } @llvm.riscv.vlseg4ff.mask.nxv4bf16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } poison, <vscale x 4 x bfloat> [[TMP1]], 0
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 2
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], 2
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 3
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP6]], <vscale x 4 x bfloat> [[TMP7]], 3
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 4
+// CHECK-RV64-NEXT:    store i64 [[TMP9]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP8]]
+//
+vbfloat16m1x4_t test_vlseg4e16ff_v_bf16m1x4_m(vbool16_t vm, const __bf16 *rs1,
+                                              size_t *new_vl, size_t vl) {
+  return __riscv_vlseg4e16ff_v_bf16m1x4_m(vm, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vlseg4e16ff_v_bf16m2x4_m(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } @llvm.riscv.vlseg4ff.mask.nxv8bf16.i64(<vscale x 8 x bfloat> poison, <vscale x 8 x bfloat> poison, <vscale x 8 x bfloat> poison, <vscale x 8 x bfloat> poison, ptr [[RS1]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } poison, <vscale x 8 x bfloat> [[TMP1]], 0
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP2]], <vscale x 8 x bfloat> [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP0]], 2
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP4]], <vscale x 8 x bfloat> [[TMP5]], 2
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP0]], 3
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP6]], <vscale x 8 x bfloat> [[TMP7]], 3
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP0]], 4
+// CHECK-RV64-NEXT:    store i64 [[TMP9]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP8]]
+//
+vbfloat16m2x4_t test_vlseg4e16ff_v_bf16m2x4_m(vbool8_t vm, const __bf16 *rs1,
+                                              size_t *new_vl, size_t vl) {
+  return __riscv_vlseg4e16ff_v_bf16m2x4_m(vm, rs1, new_vl, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vlseg5e16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vlseg5e16.c
new file mode 100644
index 000000000000..08d4af94ee54
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vlseg5e16.c
@@ -0,0 +1,72 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg5e16_v_bf16mf4x5(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlseg5.nxv1bf16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x5_t test_vlseg5e16_v_bf16mf4x5(const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg5e16_v_bf16mf4x5(rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg5e16_v_bf16mf2x5(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlseg5.nxv2bf16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x5_t test_vlseg5e16_v_bf16mf2x5(const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg5e16_v_bf16mf2x5(rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg5e16_v_bf16m1x5(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlseg5.nxv4bf16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x5_t test_vlseg5e16_v_bf16m1x5(const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg5e16_v_bf16m1x5(rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg5e16_v_bf16mf4x5_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlseg5.mask.nxv1bf16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x5_t test_vlseg5e16_v_bf16mf4x5_m(vbool64_t vm, const __bf16 *rs1,
+                                              size_t vl) {
+  return __riscv_vlseg5e16_v_bf16mf4x5_m(vm, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg5e16_v_bf16mf2x5_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlseg5.mask.nxv2bf16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x5_t test_vlseg5e16_v_bf16mf2x5_m(vbool32_t vm, const __bf16 *rs1,
+                                              size_t vl) {
+  return __riscv_vlseg5e16_v_bf16mf2x5_m(vm, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg5e16_v_bf16m1x5_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlseg5.mask.nxv4bf16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x5_t test_vlseg5e16_v_bf16m1x5_m(vbool16_t vm, const __bf16 *rs1,
+                                            size_t vl) {
+  return __riscv_vlseg5e16_v_bf16m1x5_m(vm, rs1, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vlseg5e16ff.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vlseg5e16ff.c
new file mode 100644
index 000000000000..b19b51a28eda
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vlseg5e16ff.c
@@ -0,0 +1,147 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg5e16ff_v_bf16mf4x5(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } @llvm.riscv.vlseg5ff.nxv1bf16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } poison, <vscale x 1 x bfloat> [[TMP1]], 0
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 2
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], 2
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 3
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP6]], <vscale x 1 x bfloat> [[TMP7]], 3
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 4
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP8]], <vscale x 1 x bfloat> [[TMP9]], 4
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 5
+// CHECK-RV64-NEXT:    store i64 [[TMP11]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP10]]
+//
+vbfloat16mf4x5_t test_vlseg5e16ff_v_bf16mf4x5(const __bf16 *rs1, size_t *new_vl,
+                                              size_t vl) {
+  return __riscv_vlseg5e16ff_v_bf16mf4x5(rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg5e16ff_v_bf16mf2x5(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } @llvm.riscv.vlseg5ff.nxv2bf16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } poison, <vscale x 2 x bfloat> [[TMP1]], 0
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 2
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], 2
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 3
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP6]], <vscale x 2 x bfloat> [[TMP7]], 3
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 4
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP8]], <vscale x 2 x bfloat> [[TMP9]], 4
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 5
+// CHECK-RV64-NEXT:    store i64 [[TMP11]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP10]]
+//
+vbfloat16mf2x5_t test_vlseg5e16ff_v_bf16mf2x5(const __bf16 *rs1, size_t *new_vl,
+                                              size_t vl) {
+  return __riscv_vlseg5e16ff_v_bf16mf2x5(rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg5e16ff_v_bf16m1x5(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } @llvm.riscv.vlseg5ff.nxv4bf16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } poison, <vscale x 4 x bfloat> [[TMP1]], 0
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 2
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], 2
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 3
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP6]], <vscale x 4 x bfloat> [[TMP7]], 3
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 4
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP8]], <vscale x 4 x bfloat> [[TMP9]], 4
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 5
+// CHECK-RV64-NEXT:    store i64 [[TMP11]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP10]]
+//
+vbfloat16m1x5_t test_vlseg5e16ff_v_bf16m1x5(const __bf16 *rs1, size_t *new_vl,
+                                            size_t vl) {
+  return __riscv_vlseg5e16ff_v_bf16m1x5(rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg5e16ff_v_bf16mf4x5_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } @llvm.riscv.vlseg5ff.mask.nxv1bf16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } poison, <vscale x 1 x bfloat> [[TMP1]], 0
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 2
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], 2
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 3
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP6]], <vscale x 1 x bfloat> [[TMP7]], 3
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 4
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP8]], <vscale x 1 x bfloat> [[TMP9]], 4
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 5
+// CHECK-RV64-NEXT:    store i64 [[TMP11]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP10]]
+//
+vbfloat16mf4x5_t test_vlseg5e16ff_v_bf16mf4x5_m(vbool64_t vm, const __bf16 *rs1,
+                                                size_t *new_vl, size_t vl) {
+  return __riscv_vlseg5e16ff_v_bf16mf4x5_m(vm, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg5e16ff_v_bf16mf2x5_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } @llvm.riscv.vlseg5ff.mask.nxv2bf16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } poison, <vscale x 2 x bfloat> [[TMP1]], 0
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 2
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], 2
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 3
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP6]], <vscale x 2 x bfloat> [[TMP7]], 3
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 4
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP8]], <vscale x 2 x bfloat> [[TMP9]], 4
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 5
+// CHECK-RV64-NEXT:    store i64 [[TMP11]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP10]]
+//
+vbfloat16mf2x5_t test_vlseg5e16ff_v_bf16mf2x5_m(vbool32_t vm, const __bf16 *rs1,
+                                                size_t *new_vl, size_t vl) {
+  return __riscv_vlseg5e16ff_v_bf16mf2x5_m(vm, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg5e16ff_v_bf16m1x5_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } @llvm.riscv.vlseg5ff.mask.nxv4bf16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } poison, <vscale x 4 x bfloat> [[TMP1]], 0
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 2
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], 2
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 3
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP6]], <vscale x 4 x bfloat> [[TMP7]], 3
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 4
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP8]], <vscale x 4 x bfloat> [[TMP9]], 4
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 5
+// CHECK-RV64-NEXT:    store i64 [[TMP11]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP10]]
+//
+vbfloat16m1x5_t test_vlseg5e16ff_v_bf16m1x5_m(vbool16_t vm, const __bf16 *rs1,
+                                              size_t *new_vl, size_t vl) {
+  return __riscv_vlseg5e16ff_v_bf16m1x5_m(vm, rs1, new_vl, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vlseg6e16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vlseg6e16.c
new file mode 100644
index 000000000000..018c18098122
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vlseg6e16.c
@@ -0,0 +1,72 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg6e16_v_bf16mf4x6(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlseg6.nxv1bf16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x6_t test_vlseg6e16_v_bf16mf4x6(const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg6e16_v_bf16mf4x6(rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg6e16_v_bf16mf2x6(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlseg6.nxv2bf16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x6_t test_vlseg6e16_v_bf16mf2x6(const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg6e16_v_bf16mf2x6(rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg6e16_v_bf16m1x6(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlseg6.nxv4bf16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x6_t test_vlseg6e16_v_bf16m1x6(const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg6e16_v_bf16m1x6(rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg6e16_v_bf16mf4x6_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlseg6.mask.nxv1bf16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x6_t test_vlseg6e16_v_bf16mf4x6_m(vbool64_t vm, const __bf16 *rs1,
+                                              size_t vl) {
+  return __riscv_vlseg6e16_v_bf16mf4x6_m(vm, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg6e16_v_bf16mf2x6_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlseg6.mask.nxv2bf16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x6_t test_vlseg6e16_v_bf16mf2x6_m(vbool32_t vm, const __bf16 *rs1,
+                                              size_t vl) {
+  return __riscv_vlseg6e16_v_bf16mf2x6_m(vm, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg6e16_v_bf16m1x6_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlseg6.mask.nxv4bf16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x6_t test_vlseg6e16_v_bf16m1x6_m(vbool16_t vm, const __bf16 *rs1,
+                                            size_t vl) {
+  return __riscv_vlseg6e16_v_bf16m1x6_m(vm, rs1, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vlseg6e16ff.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vlseg6e16ff.c
new file mode 100644
index 000000000000..2104f8e083f2
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vlseg6e16ff.c
@@ -0,0 +1,159 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg6e16ff_v_bf16mf4x6(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } @llvm.riscv.vlseg6ff.nxv1bf16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } poison, <vscale x 1 x bfloat> [[TMP1]], 0
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 2
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], 2
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 3
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP6]], <vscale x 1 x bfloat> [[TMP7]], 3
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 4
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP8]], <vscale x 1 x bfloat> [[TMP9]], 4
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 5
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP10]], <vscale x 1 x bfloat> [[TMP11]], 5
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 6
+// CHECK-RV64-NEXT:    store i64 [[TMP13]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP12]]
+//
+vbfloat16mf4x6_t test_vlseg6e16ff_v_bf16mf4x6(const __bf16 *rs1, size_t *new_vl,
+                                              size_t vl) {
+  return __riscv_vlseg6e16ff_v_bf16mf4x6(rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg6e16ff_v_bf16mf2x6(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } @llvm.riscv.vlseg6ff.nxv2bf16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } poison, <vscale x 2 x bfloat> [[TMP1]], 0
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 2
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], 2
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 3
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP6]], <vscale x 2 x bfloat> [[TMP7]], 3
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 4
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP8]], <vscale x 2 x bfloat> [[TMP9]], 4
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 5
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP10]], <vscale x 2 x bfloat> [[TMP11]], 5
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 6
+// CHECK-RV64-NEXT:    store i64 [[TMP13]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP12]]
+//
+vbfloat16mf2x6_t test_vlseg6e16ff_v_bf16mf2x6(const __bf16 *rs1, size_t *new_vl,
+                                              size_t vl) {
+  return __riscv_vlseg6e16ff_v_bf16mf2x6(rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg6e16ff_v_bf16m1x6(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } @llvm.riscv.vlseg6ff.nxv4bf16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } poison, <vscale x 4 x bfloat> [[TMP1]], 0
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 2
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], 2
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 3
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP6]], <vscale x 4 x bfloat> [[TMP7]], 3
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 4
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP8]], <vscale x 4 x bfloat> [[TMP9]], 4
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 5
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP10]], <vscale x 4 x bfloat> [[TMP11]], 5
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 6
+// CHECK-RV64-NEXT:    store i64 [[TMP13]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP12]]
+//
+vbfloat16m1x6_t test_vlseg6e16ff_v_bf16m1x6(const __bf16 *rs1, size_t *new_vl,
+                                            size_t vl) {
+  return __riscv_vlseg6e16ff_v_bf16m1x6(rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg6e16ff_v_bf16mf4x6_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } @llvm.riscv.vlseg6ff.mask.nxv1bf16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } poison, <vscale x 1 x bfloat> [[TMP1]], 0
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 2
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], 2
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 3
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP6]], <vscale x 1 x bfloat> [[TMP7]], 3
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 4
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP8]], <vscale x 1 x bfloat> [[TMP9]], 4
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 5
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP10]], <vscale x 1 x bfloat> [[TMP11]], 5
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 6
+// CHECK-RV64-NEXT:    store i64 [[TMP13]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP12]]
+//
+vbfloat16mf4x6_t test_vlseg6e16ff_v_bf16mf4x6_m(vbool64_t vm, const __bf16 *rs1,
+                                                size_t *new_vl, size_t vl) {
+  return __riscv_vlseg6e16ff_v_bf16mf4x6_m(vm, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg6e16ff_v_bf16mf2x6_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } @llvm.riscv.vlseg6ff.mask.nxv2bf16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } poison, <vscale x 2 x bfloat> [[TMP1]], 0
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 2
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], 2
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 3
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP6]], <vscale x 2 x bfloat> [[TMP7]], 3
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 4
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP8]], <vscale x 2 x bfloat> [[TMP9]], 4
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 5
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP10]], <vscale x 2 x bfloat> [[TMP11]], 5
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 6
+// CHECK-RV64-NEXT:    store i64 [[TMP13]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP12]]
+//
+vbfloat16mf2x6_t test_vlseg6e16ff_v_bf16mf2x6_m(vbool32_t vm, const __bf16 *rs1,
+                                                size_t *new_vl, size_t vl) {
+  return __riscv_vlseg6e16ff_v_bf16mf2x6_m(vm, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg6e16ff_v_bf16m1x6_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } @llvm.riscv.vlseg6ff.mask.nxv4bf16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } poison, <vscale x 4 x bfloat> [[TMP1]], 0
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 2
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], 2
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 3
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP6]], <vscale x 4 x bfloat> [[TMP7]], 3
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 4
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP8]], <vscale x 4 x bfloat> [[TMP9]], 4
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 5
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP10]], <vscale x 4 x bfloat> [[TMP11]], 5
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 6
+// CHECK-RV64-NEXT:    store i64 [[TMP13]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP12]]
+//
+vbfloat16m1x6_t test_vlseg6e16ff_v_bf16m1x6_m(vbool16_t vm, const __bf16 *rs1,
+                                              size_t *new_vl, size_t vl) {
+  return __riscv_vlseg6e16ff_v_bf16m1x6_m(vm, rs1, new_vl, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vlseg7e16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vlseg7e16.c
new file mode 100644
index 000000000000..b15ef25d0c1f
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vlseg7e16.c
@@ -0,0 +1,72 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg7e16_v_bf16mf4x7(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlseg7.nxv1bf16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x7_t test_vlseg7e16_v_bf16mf4x7(const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg7e16_v_bf16mf4x7(rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg7e16_v_bf16mf2x7(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlseg7.nxv2bf16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x7_t test_vlseg7e16_v_bf16mf2x7(const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg7e16_v_bf16mf2x7(rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg7e16_v_bf16m1x7(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlseg7.nxv4bf16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x7_t test_vlseg7e16_v_bf16m1x7(const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg7e16_v_bf16m1x7(rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg7e16_v_bf16mf4x7_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlseg7.mask.nxv1bf16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x7_t test_vlseg7e16_v_bf16mf4x7_m(vbool64_t vm, const __bf16 *rs1,
+                                              size_t vl) {
+  return __riscv_vlseg7e16_v_bf16mf4x7_m(vm, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg7e16_v_bf16mf2x7_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlseg7.mask.nxv2bf16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x7_t test_vlseg7e16_v_bf16mf2x7_m(vbool32_t vm, const __bf16 *rs1,
+                                              size_t vl) {
+  return __riscv_vlseg7e16_v_bf16mf2x7_m(vm, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg7e16_v_bf16m1x7_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlseg7.mask.nxv4bf16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x7_t test_vlseg7e16_v_bf16m1x7_m(vbool16_t vm, const __bf16 *rs1,
+                                            size_t vl) {
+  return __riscv_vlseg7e16_v_bf16m1x7_m(vm, rs1, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vlseg7e16ff.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vlseg7e16ff.c
new file mode 100644
index 000000000000..42361b6375cb
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vlseg7e16ff.c
@@ -0,0 +1,171 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg7e16ff_v_bf16mf4x7(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } @llvm.riscv.vlseg7ff.nxv1bf16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } poison, <vscale x 1 x bfloat> [[TMP1]], 0
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 2
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], 2
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 3
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP6]], <vscale x 1 x bfloat> [[TMP7]], 3
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 4
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP8]], <vscale x 1 x bfloat> [[TMP9]], 4
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 5
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP10]], <vscale x 1 x bfloat> [[TMP11]], 5
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 6
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP12]], <vscale x 1 x bfloat> [[TMP13]], 6
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 7
+// CHECK-RV64-NEXT:    store i64 [[TMP15]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP14]]
+//
+vbfloat16mf4x7_t test_vlseg7e16ff_v_bf16mf4x7(const __bf16 *rs1, size_t *new_vl,
+                                              size_t vl) {
+  return __riscv_vlseg7e16ff_v_bf16mf4x7(rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg7e16ff_v_bf16mf2x7(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } @llvm.riscv.vlseg7ff.nxv2bf16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } poison, <vscale x 2 x bfloat> [[TMP1]], 0
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 2
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], 2
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 3
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP6]], <vscale x 2 x bfloat> [[TMP7]], 3
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 4
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP8]], <vscale x 2 x bfloat> [[TMP9]], 4
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 5
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP10]], <vscale x 2 x bfloat> [[TMP11]], 5
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 6
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP12]], <vscale x 2 x bfloat> [[TMP13]], 6
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 7
+// CHECK-RV64-NEXT:    store i64 [[TMP15]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP14]]
+//
+vbfloat16mf2x7_t test_vlseg7e16ff_v_bf16mf2x7(const __bf16 *rs1, size_t *new_vl,
+                                              size_t vl) {
+  return __riscv_vlseg7e16ff_v_bf16mf2x7(rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg7e16ff_v_bf16m1x7(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } @llvm.riscv.vlseg7ff.nxv4bf16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } poison, <vscale x 4 x bfloat> [[TMP1]], 0
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 2
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], 2
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 3
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP6]], <vscale x 4 x bfloat> [[TMP7]], 3
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 4
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP8]], <vscale x 4 x bfloat> [[TMP9]], 4
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 5
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP10]], <vscale x 4 x bfloat> [[TMP11]], 5
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 6
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP12]], <vscale x 4 x bfloat> [[TMP13]], 6
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 7
+// CHECK-RV64-NEXT:    store i64 [[TMP15]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP14]]
+//
+vbfloat16m1x7_t test_vlseg7e16ff_v_bf16m1x7(const __bf16 *rs1, size_t *new_vl,
+                                            size_t vl) {
+  return __riscv_vlseg7e16ff_v_bf16m1x7(rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg7e16ff_v_bf16mf4x7_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } @llvm.riscv.vlseg7ff.mask.nxv1bf16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } poison, <vscale x 1 x bfloat> [[TMP1]], 0
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 2
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], 2
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 3
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP6]], <vscale x 1 x bfloat> [[TMP7]], 3
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 4
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP8]], <vscale x 1 x bfloat> [[TMP9]], 4
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 5
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP10]], <vscale x 1 x bfloat> [[TMP11]], 5
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 6
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP12]], <vscale x 1 x bfloat> [[TMP13]], 6
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 7
+// CHECK-RV64-NEXT:    store i64 [[TMP15]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP14]]
+//
+vbfloat16mf4x7_t test_vlseg7e16ff_v_bf16mf4x7_m(vbool64_t vm, const __bf16 *rs1,
+                                                size_t *new_vl, size_t vl) {
+  return __riscv_vlseg7e16ff_v_bf16mf4x7_m(vm, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg7e16ff_v_bf16mf2x7_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } @llvm.riscv.vlseg7ff.mask.nxv2bf16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } poison, <vscale x 2 x bfloat> [[TMP1]], 0
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 2
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], 2
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 3
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP6]], <vscale x 2 x bfloat> [[TMP7]], 3
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 4
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP8]], <vscale x 2 x bfloat> [[TMP9]], 4
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 5
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP10]], <vscale x 2 x bfloat> [[TMP11]], 5
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 6
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP12]], <vscale x 2 x bfloat> [[TMP13]], 6
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 7
+// CHECK-RV64-NEXT:    store i64 [[TMP15]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP14]]
+//
+vbfloat16mf2x7_t test_vlseg7e16ff_v_bf16mf2x7_m(vbool32_t vm, const __bf16 *rs1,
+                                                size_t *new_vl, size_t vl) {
+  return __riscv_vlseg7e16ff_v_bf16mf2x7_m(vm, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg7e16ff_v_bf16m1x7_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } @llvm.riscv.vlseg7ff.mask.nxv4bf16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } poison, <vscale x 4 x bfloat> [[TMP1]], 0
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 2
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], 2
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 3
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP6]], <vscale x 4 x bfloat> [[TMP7]], 3
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 4
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP8]], <vscale x 4 x bfloat> [[TMP9]], 4
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 5
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP10]], <vscale x 4 x bfloat> [[TMP11]], 5
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 6
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP12]], <vscale x 4 x bfloat> [[TMP13]], 6
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 7
+// CHECK-RV64-NEXT:    store i64 [[TMP15]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP14]]
+//
+vbfloat16m1x7_t test_vlseg7e16ff_v_bf16m1x7_m(vbool16_t vm, const __bf16 *rs1,
+                                              size_t *new_vl, size_t vl) {
+  return __riscv_vlseg7e16ff_v_bf16m1x7_m(vm, rs1, new_vl, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vlseg8e16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vlseg8e16.c
new file mode 100644
index 000000000000..7f113a21e9bf
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vlseg8e16.c
@@ -0,0 +1,72 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg8e16_v_bf16mf4x8(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlseg8.nxv1bf16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x8_t test_vlseg8e16_v_bf16mf4x8(const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg8e16_v_bf16mf4x8(rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg8e16_v_bf16mf2x8(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlseg8.nxv2bf16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x8_t test_vlseg8e16_v_bf16mf2x8(const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg8e16_v_bf16mf2x8(rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg8e16_v_bf16m1x8(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlseg8.nxv4bf16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x8_t test_vlseg8e16_v_bf16m1x8(const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg8e16_v_bf16m1x8(rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg8e16_v_bf16mf4x8_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlseg8.mask.nxv1bf16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x8_t test_vlseg8e16_v_bf16mf4x8_m(vbool64_t vm, const __bf16 *rs1,
+                                              size_t vl) {
+  return __riscv_vlseg8e16_v_bf16mf4x8_m(vm, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg8e16_v_bf16mf2x8_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlseg8.mask.nxv2bf16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x8_t test_vlseg8e16_v_bf16mf2x8_m(vbool32_t vm, const __bf16 *rs1,
+                                              size_t vl) {
+  return __riscv_vlseg8e16_v_bf16mf2x8_m(vm, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg8e16_v_bf16m1x8_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlseg8.mask.nxv4bf16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x8_t test_vlseg8e16_v_bf16m1x8_m(vbool16_t vm, const __bf16 *rs1,
+                                            size_t vl) {
+  return __riscv_vlseg8e16_v_bf16m1x8_m(vm, rs1, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vlseg8e16ff.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vlseg8e16ff.c
new file mode 100644
index 000000000000..c9cf4ba0243c
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vlseg8e16ff.c
@@ -0,0 +1,183 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg8e16ff_v_bf16mf4x8(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } @llvm.riscv.vlseg8ff.nxv1bf16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } poison, <vscale x 1 x bfloat> [[TMP1]], 0
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 2
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], 2
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 3
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP6]], <vscale x 1 x bfloat> [[TMP7]], 3
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 4
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP8]], <vscale x 1 x bfloat> [[TMP9]], 4
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 5
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP10]], <vscale x 1 x bfloat> [[TMP11]], 5
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 6
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP12]], <vscale x 1 x bfloat> [[TMP13]], 6
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 7
+// CHECK-RV64-NEXT:    [[TMP16:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP14]], <vscale x 1 x bfloat> [[TMP15]], 7
+// CHECK-RV64-NEXT:    [[TMP17:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 8
+// CHECK-RV64-NEXT:    store i64 [[TMP17]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP16]]
+//
+vbfloat16mf4x8_t test_vlseg8e16ff_v_bf16mf4x8(const __bf16 *rs1, size_t *new_vl,
+                                              size_t vl) {
+  return __riscv_vlseg8e16ff_v_bf16mf4x8(rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg8e16ff_v_bf16mf2x8(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } @llvm.riscv.vlseg8ff.nxv2bf16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } poison, <vscale x 2 x bfloat> [[TMP1]], 0
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 2
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], 2
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 3
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP6]], <vscale x 2 x bfloat> [[TMP7]], 3
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 4
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP8]], <vscale x 2 x bfloat> [[TMP9]], 4
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 5
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP10]], <vscale x 2 x bfloat> [[TMP11]], 5
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 6
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP12]], <vscale x 2 x bfloat> [[TMP13]], 6
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 7
+// CHECK-RV64-NEXT:    [[TMP16:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP14]], <vscale x 2 x bfloat> [[TMP15]], 7
+// CHECK-RV64-NEXT:    [[TMP17:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 8
+// CHECK-RV64-NEXT:    store i64 [[TMP17]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP16]]
+//
+vbfloat16mf2x8_t test_vlseg8e16ff_v_bf16mf2x8(const __bf16 *rs1, size_t *new_vl,
+                                              size_t vl) {
+  return __riscv_vlseg8e16ff_v_bf16mf2x8(rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg8e16ff_v_bf16m1x8(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } @llvm.riscv.vlseg8ff.nxv4bf16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } poison, <vscale x 4 x bfloat> [[TMP1]], 0
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 2
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], 2
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 3
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP6]], <vscale x 4 x bfloat> [[TMP7]], 3
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 4
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP8]], <vscale x 4 x bfloat> [[TMP9]], 4
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 5
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP10]], <vscale x 4 x bfloat> [[TMP11]], 5
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 6
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP12]], <vscale x 4 x bfloat> [[TMP13]], 6
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 7
+// CHECK-RV64-NEXT:    [[TMP16:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP14]], <vscale x 4 x bfloat> [[TMP15]], 7
+// CHECK-RV64-NEXT:    [[TMP17:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 8
+// CHECK-RV64-NEXT:    store i64 [[TMP17]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP16]]
+//
+vbfloat16m1x8_t test_vlseg8e16ff_v_bf16m1x8(const __bf16 *rs1, size_t *new_vl,
+                                            size_t vl) {
+  return __riscv_vlseg8e16ff_v_bf16m1x8(rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg8e16ff_v_bf16mf4x8_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } @llvm.riscv.vlseg8ff.mask.nxv1bf16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } poison, <vscale x 1 x bfloat> [[TMP1]], 0
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 2
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], 2
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 3
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP6]], <vscale x 1 x bfloat> [[TMP7]], 3
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 4
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP8]], <vscale x 1 x bfloat> [[TMP9]], 4
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 5
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP10]], <vscale x 1 x bfloat> [[TMP11]], 5
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 6
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP12]], <vscale x 1 x bfloat> [[TMP13]], 6
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 7
+// CHECK-RV64-NEXT:    [[TMP16:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP14]], <vscale x 1 x bfloat> [[TMP15]], 7
+// CHECK-RV64-NEXT:    [[TMP17:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 8
+// CHECK-RV64-NEXT:    store i64 [[TMP17]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP16]]
+//
+vbfloat16mf4x8_t test_vlseg8e16ff_v_bf16mf4x8_m(vbool64_t vm, const __bf16 *rs1,
+                                                size_t *new_vl, size_t vl) {
+  return __riscv_vlseg8e16ff_v_bf16mf4x8_m(vm, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg8e16ff_v_bf16mf2x8_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } @llvm.riscv.vlseg8ff.mask.nxv2bf16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } poison, <vscale x 2 x bfloat> [[TMP1]], 0
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 2
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], 2
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 3
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP6]], <vscale x 2 x bfloat> [[TMP7]], 3
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 4
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP8]], <vscale x 2 x bfloat> [[TMP9]], 4
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 5
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP10]], <vscale x 2 x bfloat> [[TMP11]], 5
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 6
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP12]], <vscale x 2 x bfloat> [[TMP13]], 6
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 7
+// CHECK-RV64-NEXT:    [[TMP16:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP14]], <vscale x 2 x bfloat> [[TMP15]], 7
+// CHECK-RV64-NEXT:    [[TMP17:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 8
+// CHECK-RV64-NEXT:    store i64 [[TMP17]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP16]]
+//
+vbfloat16mf2x8_t test_vlseg8e16ff_v_bf16mf2x8_m(vbool32_t vm, const __bf16 *rs1,
+                                                size_t *new_vl, size_t vl) {
+  return __riscv_vlseg8e16ff_v_bf16mf2x8_m(vm, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg8e16ff_v_bf16m1x8_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } @llvm.riscv.vlseg8ff.mask.nxv4bf16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } poison, <vscale x 4 x bfloat> [[TMP1]], 0
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 2
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], 2
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 3
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP6]], <vscale x 4 x bfloat> [[TMP7]], 3
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 4
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP8]], <vscale x 4 x bfloat> [[TMP9]], 4
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 5
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP10]], <vscale x 4 x bfloat> [[TMP11]], 5
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 6
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP12]], <vscale x 4 x bfloat> [[TMP13]], 6
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 7
+// CHECK-RV64-NEXT:    [[TMP16:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP14]], <vscale x 4 x bfloat> [[TMP15]], 7
+// CHECK-RV64-NEXT:    [[TMP17:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 8
+// CHECK-RV64-NEXT:    store i64 [[TMP17]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP16]]
+//
+vbfloat16m1x8_t test_vlseg8e16ff_v_bf16m1x8_m(vbool16_t vm, const __bf16 *rs1,
+                                              size_t *new_vl, size_t vl) {
+  return __riscv_vlseg8e16ff_v_bf16m1x8_m(vm, rs1, new_vl, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vlsseg2e16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vlsseg2e16.c
new file mode 100644
index 000000000000..92b5c2948fa3
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vlsseg2e16.c
@@ -0,0 +1,119 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlsseg2e16_v_bf16mf4x2(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlsseg2.nxv1bf16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x2_t test_vlsseg2e16_v_bf16mf4x2(const __bf16 *rs1, ptrdiff_t rs2,
+                                             size_t vl) {
+  return __riscv_vlsseg2e16_v_bf16mf4x2(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlsseg2e16_v_bf16mf2x2(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlsseg2.nxv2bf16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x2_t test_vlsseg2e16_v_bf16mf2x2(const __bf16 *rs1, ptrdiff_t rs2,
+                                             size_t vl) {
+  return __riscv_vlsseg2e16_v_bf16mf2x2(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlsseg2e16_v_bf16m1x2(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlsseg2.nxv4bf16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x2_t test_vlsseg2e16_v_bf16m1x2(const __bf16 *rs1, ptrdiff_t rs2,
+                                           size_t vl) {
+  return __riscv_vlsseg2e16_v_bf16m1x2(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vlsseg2e16_v_bf16m2x2(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vlsseg2.nxv8bf16.i64(<vscale x 8 x bfloat> poison, <vscale x 8 x bfloat> poison, ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP0]]
+//
+vbfloat16m2x2_t test_vlsseg2e16_v_bf16m2x2(const __bf16 *rs1, ptrdiff_t rs2,
+                                           size_t vl) {
+  return __riscv_vlsseg2e16_v_bf16m2x2(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @test_vlsseg2e16_v_bf16m4x2(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @llvm.riscv.vlsseg2.nxv16bf16.i64(<vscale x 16 x bfloat> poison, <vscale x 16 x bfloat> poison, ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[TMP0]]
+//
+vbfloat16m4x2_t test_vlsseg2e16_v_bf16m4x2(const __bf16 *rs1, ptrdiff_t rs2,
+                                           size_t vl) {
+  return __riscv_vlsseg2e16_v_bf16m4x2(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlsseg2e16_v_bf16mf4x2_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlsseg2.mask.nxv1bf16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], i64 [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x2_t test_vlsseg2e16_v_bf16mf4x2_m(vbool64_t vm, const __bf16 *rs1,
+                                               ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg2e16_v_bf16mf4x2_m(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlsseg2e16_v_bf16mf2x2_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlsseg2.mask.nxv2bf16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], i64 [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x2_t test_vlsseg2e16_v_bf16mf2x2_m(vbool32_t vm, const __bf16 *rs1,
+                                               ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg2e16_v_bf16mf2x2_m(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlsseg2e16_v_bf16m1x2_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlsseg2.mask.nxv4bf16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], i64 [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x2_t test_vlsseg2e16_v_bf16m1x2_m(vbool16_t vm, const __bf16 *rs1,
+                                             ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg2e16_v_bf16m1x2_m(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vlsseg2e16_v_bf16m2x2_m(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vlsseg2.mask.nxv8bf16.i64(<vscale x 8 x bfloat> poison, <vscale x 8 x bfloat> poison, ptr [[RS1]], i64 [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP0]]
+//
+vbfloat16m2x2_t test_vlsseg2e16_v_bf16m2x2_m(vbool8_t vm, const __bf16 *rs1,
+                                             ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg2e16_v_bf16m2x2_m(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @test_vlsseg2e16_v_bf16m4x2_m(
+// CHECK-RV64-SAME: <vscale x 16 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @llvm.riscv.vlsseg2.mask.nxv16bf16.i64(<vscale x 16 x bfloat> poison, <vscale x 16 x bfloat> poison, ptr [[RS1]], i64 [[RS2]], <vscale x 16 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[TMP0]]
+//
+vbfloat16m4x2_t test_vlsseg2e16_v_bf16m4x2_m(vbool4_t vm, const __bf16 *rs1,
+                                             ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg2e16_v_bf16m4x2_m(vm, rs1, rs2, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vlsseg3e16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vlsseg3e16.c
new file mode 100644
index 000000000000..d834ead8365c
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vlsseg3e16.c
@@ -0,0 +1,97 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlsseg3e16_v_bf16mf4x3(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlsseg3.nxv1bf16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x3_t test_vlsseg3e16_v_bf16mf4x3(const __bf16 *rs1, ptrdiff_t rs2,
+                                             size_t vl) {
+  return __riscv_vlsseg3e16_v_bf16mf4x3(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlsseg3e16_v_bf16mf2x3(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlsseg3.nxv2bf16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x3_t test_vlsseg3e16_v_bf16mf2x3(const __bf16 *rs1, ptrdiff_t rs2,
+                                             size_t vl) {
+  return __riscv_vlsseg3e16_v_bf16mf2x3(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlsseg3e16_v_bf16m1x3(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlsseg3.nxv4bf16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x3_t test_vlsseg3e16_v_bf16m1x3(const __bf16 *rs1, ptrdiff_t rs2,
+                                           size_t vl) {
+  return __riscv_vlsseg3e16_v_bf16m1x3(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vlsseg3e16_v_bf16m2x3(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vlsseg3.nxv8bf16.i64(<vscale x 8 x bfloat> poison, <vscale x 8 x bfloat> poison, <vscale x 8 x bfloat> poison, ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP0]]
+//
+vbfloat16m2x3_t test_vlsseg3e16_v_bf16m2x3(const __bf16 *rs1, ptrdiff_t rs2,
+                                           size_t vl) {
+  return __riscv_vlsseg3e16_v_bf16m2x3(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlsseg3e16_v_bf16mf4x3_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlsseg3.mask.nxv1bf16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], i64 [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x3_t test_vlsseg3e16_v_bf16mf4x3_m(vbool64_t vm, const __bf16 *rs1,
+                                               ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg3e16_v_bf16mf4x3_m(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlsseg3e16_v_bf16mf2x3_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlsseg3.mask.nxv2bf16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], i64 [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x3_t test_vlsseg3e16_v_bf16mf2x3_m(vbool32_t vm, const __bf16 *rs1,
+                                               ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg3e16_v_bf16mf2x3_m(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlsseg3e16_v_bf16m1x3_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlsseg3.mask.nxv4bf16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], i64 [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x3_t test_vlsseg3e16_v_bf16m1x3_m(vbool16_t vm, const __bf16 *rs1,
+                                             ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg3e16_v_bf16m1x3_m(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vlsseg3e16_v_bf16m2x3_m(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vlsseg3.mask.nxv8bf16.i64(<vscale x 8 x bfloat> poison, <vscale x 8 x bfloat> poison, <vscale x 8 x bfloat> poison, ptr [[RS1]], i64 [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP0]]
+//
+vbfloat16m2x3_t test_vlsseg3e16_v_bf16m2x3_m(vbool8_t vm, const __bf16 *rs1,
+                                             ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg3e16_v_bf16m2x3_m(vm, rs1, rs2, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vlsseg4e16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vlsseg4e16.c
new file mode 100644
index 000000000000..d59822b0fb35
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vlsseg4e16.c
@@ -0,0 +1,97 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlsseg4e16_v_bf16mf4x4(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlsseg4.nxv1bf16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x4_t test_vlsseg4e16_v_bf16mf4x4(const __bf16 *rs1, ptrdiff_t rs2,
+                                             size_t vl) {
+  return __riscv_vlsseg4e16_v_bf16mf4x4(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlsseg4e16_v_bf16mf2x4(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlsseg4.nxv2bf16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x4_t test_vlsseg4e16_v_bf16mf2x4(const __bf16 *rs1, ptrdiff_t rs2,
+                                             size_t vl) {
+  return __riscv_vlsseg4e16_v_bf16mf2x4(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlsseg4e16_v_bf16m1x4(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlsseg4.nxv4bf16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x4_t test_vlsseg4e16_v_bf16m1x4(const __bf16 *rs1, ptrdiff_t rs2,
+                                           size_t vl) {
+  return __riscv_vlsseg4e16_v_bf16m1x4(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vlsseg4e16_v_bf16m2x4(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vlsseg4.nxv8bf16.i64(<vscale x 8 x bfloat> poison, <vscale x 8 x bfloat> poison, <vscale x 8 x bfloat> poison, <vscale x 8 x bfloat> poison, ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP0]]
+//
+vbfloat16m2x4_t test_vlsseg4e16_v_bf16m2x4(const __bf16 *rs1, ptrdiff_t rs2,
+                                           size_t vl) {
+  return __riscv_vlsseg4e16_v_bf16m2x4(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlsseg4e16_v_bf16mf4x4_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlsseg4.mask.nxv1bf16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], i64 [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x4_t test_vlsseg4e16_v_bf16mf4x4_m(vbool64_t vm, const __bf16 *rs1,
+                                               ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg4e16_v_bf16mf4x4_m(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlsseg4e16_v_bf16mf2x4_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlsseg4.mask.nxv2bf16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], i64 [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x4_t test_vlsseg4e16_v_bf16mf2x4_m(vbool32_t vm, const __bf16 *rs1,
+                                               ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg4e16_v_bf16mf2x4_m(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlsseg4e16_v_bf16m1x4_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlsseg4.mask.nxv4bf16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], i64 [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x4_t test_vlsseg4e16_v_bf16m1x4_m(vbool16_t vm, const __bf16 *rs1,
+                                             ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg4e16_v_bf16m1x4_m(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vlsseg4e16_v_bf16m2x4_m(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vlsseg4.mask.nxv8bf16.i64(<vscale x 8 x bfloat> poison, <vscale x 8 x bfloat> poison, <vscale x 8 x bfloat> poison, <vscale x 8 x bfloat> poison, ptr [[RS1]], i64 [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP0]]
+//
+vbfloat16m2x4_t test_vlsseg4e16_v_bf16m2x4_m(vbool8_t vm, const __bf16 *rs1,
+                                             ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg4e16_v_bf16m2x4_m(vm, rs1, rs2, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vlsseg5e16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vlsseg5e16.c
new file mode 100644
index 000000000000..a8c653585b3b
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vlsseg5e16.c
@@ -0,0 +1,75 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlsseg5e16_v_bf16mf4x5(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlsseg5.nxv1bf16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x5_t test_vlsseg5e16_v_bf16mf4x5(const __bf16 *rs1, ptrdiff_t rs2,
+                                             size_t vl) {
+  return __riscv_vlsseg5e16_v_bf16mf4x5(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlsseg5e16_v_bf16mf2x5(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlsseg5.nxv2bf16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x5_t test_vlsseg5e16_v_bf16mf2x5(const __bf16 *rs1, ptrdiff_t rs2,
+                                             size_t vl) {
+  return __riscv_vlsseg5e16_v_bf16mf2x5(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlsseg5e16_v_bf16m1x5(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlsseg5.nxv4bf16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x5_t test_vlsseg5e16_v_bf16m1x5(const __bf16 *rs1, ptrdiff_t rs2,
+                                           size_t vl) {
+  return __riscv_vlsseg5e16_v_bf16m1x5(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlsseg5e16_v_bf16mf4x5_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlsseg5.mask.nxv1bf16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], i64 [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x5_t test_vlsseg5e16_v_bf16mf4x5_m(vbool64_t vm, const __bf16 *rs1,
+                                               ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg5e16_v_bf16mf4x5_m(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlsseg5e16_v_bf16mf2x5_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlsseg5.mask.nxv2bf16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], i64 [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x5_t test_vlsseg5e16_v_bf16mf2x5_m(vbool32_t vm, const __bf16 *rs1,
+                                               ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg5e16_v_bf16mf2x5_m(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlsseg5e16_v_bf16m1x5_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlsseg5.mask.nxv4bf16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], i64 [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x5_t test_vlsseg5e16_v_bf16m1x5_m(vbool16_t vm, const __bf16 *rs1,
+                                             ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg5e16_v_bf16m1x5_m(vm, rs1, rs2, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vlsseg6e16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vlsseg6e16.c
new file mode 100644
index 000000000000..1271f7b4eb76
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vlsseg6e16.c
@@ -0,0 +1,75 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlsseg6e16_v_bf16mf4x6(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlsseg6.nxv1bf16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x6_t test_vlsseg6e16_v_bf16mf4x6(const __bf16 *rs1, ptrdiff_t rs2,
+                                             size_t vl) {
+  return __riscv_vlsseg6e16_v_bf16mf4x6(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlsseg6e16_v_bf16mf2x6(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlsseg6.nxv2bf16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x6_t test_vlsseg6e16_v_bf16mf2x6(const __bf16 *rs1, ptrdiff_t rs2,
+                                             size_t vl) {
+  return __riscv_vlsseg6e16_v_bf16mf2x6(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlsseg6e16_v_bf16m1x6(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlsseg6.nxv4bf16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x6_t test_vlsseg6e16_v_bf16m1x6(const __bf16 *rs1, ptrdiff_t rs2,
+                                           size_t vl) {
+  return __riscv_vlsseg6e16_v_bf16m1x6(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlsseg6e16_v_bf16mf4x6_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlsseg6.mask.nxv1bf16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], i64 [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x6_t test_vlsseg6e16_v_bf16mf4x6_m(vbool64_t vm, const __bf16 *rs1,
+                                               ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg6e16_v_bf16mf4x6_m(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlsseg6e16_v_bf16mf2x6_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlsseg6.mask.nxv2bf16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], i64 [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x6_t test_vlsseg6e16_v_bf16mf2x6_m(vbool32_t vm, const __bf16 *rs1,
+                                               ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg6e16_v_bf16mf2x6_m(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlsseg6e16_v_bf16m1x6_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlsseg6.mask.nxv4bf16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], i64 [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x6_t test_vlsseg6e16_v_bf16m1x6_m(vbool16_t vm, const __bf16 *rs1,
+                                             ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg6e16_v_bf16m1x6_m(vm, rs1, rs2, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vlsseg7e16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vlsseg7e16.c
new file mode 100644
index 000000000000..f86541cc27f9
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vlsseg7e16.c
@@ -0,0 +1,75 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlsseg7e16_v_bf16mf4x7(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlsseg7.nxv1bf16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x7_t test_vlsseg7e16_v_bf16mf4x7(const __bf16 *rs1, ptrdiff_t rs2,
+                                             size_t vl) {
+  return __riscv_vlsseg7e16_v_bf16mf4x7(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlsseg7e16_v_bf16mf2x7(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlsseg7.nxv2bf16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x7_t test_vlsseg7e16_v_bf16mf2x7(const __bf16 *rs1, ptrdiff_t rs2,
+                                             size_t vl) {
+  return __riscv_vlsseg7e16_v_bf16mf2x7(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlsseg7e16_v_bf16m1x7(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlsseg7.nxv4bf16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x7_t test_vlsseg7e16_v_bf16m1x7(const __bf16 *rs1, ptrdiff_t rs2,
+                                           size_t vl) {
+  return __riscv_vlsseg7e16_v_bf16m1x7(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlsseg7e16_v_bf16mf4x7_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlsseg7.mask.nxv1bf16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], i64 [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x7_t test_vlsseg7e16_v_bf16mf4x7_m(vbool64_t vm, const __bf16 *rs1,
+                                               ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg7e16_v_bf16mf4x7_m(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlsseg7e16_v_bf16mf2x7_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlsseg7.mask.nxv2bf16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], i64 [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x7_t test_vlsseg7e16_v_bf16mf2x7_m(vbool32_t vm, const __bf16 *rs1,
+                                               ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg7e16_v_bf16mf2x7_m(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlsseg7e16_v_bf16m1x7_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlsseg7.mask.nxv4bf16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], i64 [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x7_t test_vlsseg7e16_v_bf16m1x7_m(vbool16_t vm, const __bf16 *rs1,
+                                             ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg7e16_v_bf16m1x7_m(vm, rs1, rs2, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vlsseg8e16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vlsseg8e16.c
new file mode 100644
index 000000000000..071ff57a56f9
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vlsseg8e16.c
@@ -0,0 +1,75 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlsseg8e16_v_bf16mf4x8(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlsseg8.nxv1bf16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x8_t test_vlsseg8e16_v_bf16mf4x8(const __bf16 *rs1, ptrdiff_t rs2,
+                                             size_t vl) {
+  return __riscv_vlsseg8e16_v_bf16mf4x8(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlsseg8e16_v_bf16mf2x8(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlsseg8.nxv2bf16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x8_t test_vlsseg8e16_v_bf16mf2x8(const __bf16 *rs1, ptrdiff_t rs2,
+                                             size_t vl) {
+  return __riscv_vlsseg8e16_v_bf16mf2x8(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlsseg8e16_v_bf16m1x8(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlsseg8.nxv4bf16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x8_t test_vlsseg8e16_v_bf16m1x8(const __bf16 *rs1, ptrdiff_t rs2,
+                                           size_t vl) {
+  return __riscv_vlsseg8e16_v_bf16m1x8(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlsseg8e16_v_bf16mf4x8_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlsseg8.mask.nxv1bf16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], i64 [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x8_t test_vlsseg8e16_v_bf16mf4x8_m(vbool64_t vm, const __bf16 *rs1,
+                                               ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg8e16_v_bf16mf4x8_m(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlsseg8e16_v_bf16mf2x8_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlsseg8.mask.nxv2bf16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], i64 [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x8_t test_vlsseg8e16_v_bf16mf2x8_m(vbool32_t vm, const __bf16 *rs1,
+                                               ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg8e16_v_bf16mf2x8_m(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlsseg8e16_v_bf16m1x8_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlsseg8.mask.nxv4bf16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], i64 [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x8_t test_vlsseg8e16_v_bf16m1x8_m(vbool16_t vm, const __bf16 *rs1,
+                                             ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg8e16_v_bf16m1x8_m(vm, rs1, rs2, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vluxei16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vluxei16.c
new file mode 100644
index 000000000000..cb51dc0abbb8
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vluxei16.c
@@ -0,0 +1,141 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 1 x bfloat> @test_vluxei16_v_bf16mf4(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x bfloat> @llvm.riscv.vluxei.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> poison, ptr [[RS1]], <vscale x 1 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret <vscale x 1 x bfloat> [[TMP0]]
+//
+vbfloat16mf4_t test_vluxei16_v_bf16mf4(const __bf16 *rs1, vuint16mf4_t rs2,
+                                       size_t vl) {
+  return __riscv_vluxei16_v_bf16mf4(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 2 x bfloat> @test_vluxei16_v_bf16mf2(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x bfloat> @llvm.riscv.vluxei.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> poison, ptr [[RS1]], <vscale x 2 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret <vscale x 2 x bfloat> [[TMP0]]
+//
+vbfloat16mf2_t test_vluxei16_v_bf16mf2(const __bf16 *rs1, vuint16mf2_t rs2,
+                                       size_t vl) {
+  return __riscv_vluxei16_v_bf16mf2(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 4 x bfloat> @test_vluxei16_v_bf16m1(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x bfloat> @llvm.riscv.vluxei.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> poison, ptr [[RS1]], <vscale x 4 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret <vscale x 4 x bfloat> [[TMP0]]
+//
+vbfloat16m1_t test_vluxei16_v_bf16m1(const __bf16 *rs1, vuint16m1_t rs2,
+                                     size_t vl) {
+  return __riscv_vluxei16_v_bf16m1(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 8 x bfloat> @test_vluxei16_v_bf16m2(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x bfloat> @llvm.riscv.vluxei.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> poison, ptr [[RS1]], <vscale x 8 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret <vscale x 8 x bfloat> [[TMP0]]
+//
+vbfloat16m2_t test_vluxei16_v_bf16m2(const __bf16 *rs1, vuint16m2_t rs2,
+                                     size_t vl) {
+  return __riscv_vluxei16_v_bf16m2(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 16 x bfloat> @test_vluxei16_v_bf16m4(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 16 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x bfloat> @llvm.riscv.vluxei.nxv16bf16.nxv16i16.i64(<vscale x 16 x bfloat> poison, ptr [[RS1]], <vscale x 16 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret <vscale x 16 x bfloat> [[TMP0]]
+//
+vbfloat16m4_t test_vluxei16_v_bf16m4(const __bf16 *rs1, vuint16m4_t rs2,
+                                     size_t vl) {
+  return __riscv_vluxei16_v_bf16m4(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 32 x bfloat> @test_vluxei16_v_bf16m8(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 32 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x bfloat> @llvm.riscv.vluxei.nxv32bf16.nxv32i16.i64(<vscale x 32 x bfloat> poison, ptr [[RS1]], <vscale x 32 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret <vscale x 32 x bfloat> [[TMP0]]
+//
+vbfloat16m8_t test_vluxei16_v_bf16m8(const __bf16 *rs1, vuint16m8_t rs2,
+                                     size_t vl) {
+  return __riscv_vluxei16_v_bf16m8(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 1 x bfloat> @test_vluxei16_v_bf16mf4_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x bfloat> @llvm.riscv.vluxei.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> poison, ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret <vscale x 1 x bfloat> [[TMP0]]
+//
+vbfloat16mf4_t test_vluxei16_v_bf16mf4_m(vbool64_t vm, const __bf16 *rs1,
+                                         vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vluxei16_v_bf16mf4_m(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 2 x bfloat> @test_vluxei16_v_bf16mf2_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x bfloat> @llvm.riscv.vluxei.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> poison, ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret <vscale x 2 x bfloat> [[TMP0]]
+//
+vbfloat16mf2_t test_vluxei16_v_bf16mf2_m(vbool32_t vm, const __bf16 *rs1,
+                                         vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vluxei16_v_bf16mf2_m(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 4 x bfloat> @test_vluxei16_v_bf16m1_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x bfloat> @llvm.riscv.vluxei.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> poison, ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret <vscale x 4 x bfloat> [[TMP0]]
+//
+vbfloat16m1_t test_vluxei16_v_bf16m1_m(vbool16_t vm, const __bf16 *rs1,
+                                       vuint16m1_t rs2, size_t vl) {
+  return __riscv_vluxei16_v_bf16m1_m(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 8 x bfloat> @test_vluxei16_v_bf16m2_m(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x bfloat> @llvm.riscv.vluxei.mask.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> poison, ptr [[RS1]], <vscale x 8 x i16> [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret <vscale x 8 x bfloat> [[TMP0]]
+//
+vbfloat16m2_t test_vluxei16_v_bf16m2_m(vbool8_t vm, const __bf16 *rs1,
+                                       vuint16m2_t rs2, size_t vl) {
+  return __riscv_vluxei16_v_bf16m2_m(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 16 x bfloat> @test_vluxei16_v_bf16m4_m(
+// CHECK-RV64-SAME: <vscale x 16 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 16 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x bfloat> @llvm.riscv.vluxei.mask.nxv16bf16.nxv16i16.i64(<vscale x 16 x bfloat> poison, ptr [[RS1]], <vscale x 16 x i16> [[RS2]], <vscale x 16 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret <vscale x 16 x bfloat> [[TMP0]]
+//
+vbfloat16m4_t test_vluxei16_v_bf16m4_m(vbool4_t vm, const __bf16 *rs1,
+                                       vuint16m4_t rs2, size_t vl) {
+  return __riscv_vluxei16_v_bf16m4_m(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 32 x bfloat> @test_vluxei16_v_bf16m8_m(
+// CHECK-RV64-SAME: <vscale x 32 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 32 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x bfloat> @llvm.riscv.vluxei.mask.nxv32bf16.nxv32i16.i64(<vscale x 32 x bfloat> poison, ptr [[RS1]], <vscale x 32 x i16> [[RS2]], <vscale x 32 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret <vscale x 32 x bfloat> [[TMP0]]
+//
+vbfloat16m8_t test_vluxei16_v_bf16m8_m(vbool2_t vm, const __bf16 *rs1,
+                                       vuint16m8_t rs2, size_t vl) {
+  return __riscv_vluxei16_v_bf16m8_m(vm, rs1, rs2, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vluxseg2ei16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vluxseg2ei16.c
new file mode 100644
index 000000000000..983c99364485
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vluxseg2ei16.c
@@ -0,0 +1,121 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vluxseg2ei16_v_bf16mf4x2(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vluxseg2.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], <vscale x 1 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x2_t test_vluxseg2ei16_v_bf16mf4x2(const __bf16 *rs1,
+                                               vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vluxseg2ei16_v_bf16mf4x2(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vluxseg2ei16_v_bf16mf2x2(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vluxseg2.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], <vscale x 2 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x2_t test_vluxseg2ei16_v_bf16mf2x2(const __bf16 *rs1,
+                                               vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vluxseg2ei16_v_bf16mf2x2(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vluxseg2ei16_v_bf16m1x2(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vluxseg2.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], <vscale x 4 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x2_t test_vluxseg2ei16_v_bf16m1x2(const __bf16 *rs1, vuint16m1_t rs2,
+                                             size_t vl) {
+  return __riscv_vluxseg2ei16_v_bf16m1x2(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vluxseg2ei16_v_bf16m2x2(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vluxseg2.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> poison, <vscale x 8 x bfloat> poison, ptr [[RS1]], <vscale x 8 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP0]]
+//
+vbfloat16m2x2_t test_vluxseg2ei16_v_bf16m2x2(const __bf16 *rs1, vuint16m2_t rs2,
+                                             size_t vl) {
+  return __riscv_vluxseg2ei16_v_bf16m2x2(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @test_vluxseg2ei16_v_bf16m4x2(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 16 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @llvm.riscv.vluxseg2.nxv16bf16.nxv16i16.i64(<vscale x 16 x bfloat> poison, <vscale x 16 x bfloat> poison, ptr [[RS1]], <vscale x 16 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[TMP0]]
+//
+vbfloat16m4x2_t test_vluxseg2ei16_v_bf16m4x2(const __bf16 *rs1, vuint16m4_t rs2,
+                                             size_t vl) {
+  return __riscv_vluxseg2ei16_v_bf16m4x2(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vluxseg2ei16_v_bf16mf4x2_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vluxseg2.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x2_t test_vluxseg2ei16_v_bf16mf4x2_m(vbool64_t vm,
+                                                 const __bf16 *rs1,
+                                                 vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vluxseg2ei16_v_bf16mf4x2_m(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vluxseg2ei16_v_bf16mf2x2_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vluxseg2.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x2_t test_vluxseg2ei16_v_bf16mf2x2_m(vbool32_t vm,
+                                                 const __bf16 *rs1,
+                                                 vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vluxseg2ei16_v_bf16mf2x2_m(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vluxseg2ei16_v_bf16m1x2_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vluxseg2.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x2_t test_vluxseg2ei16_v_bf16m1x2_m(vbool16_t vm, const __bf16 *rs1,
+                                               vuint16m1_t rs2, size_t vl) {
+  return __riscv_vluxseg2ei16_v_bf16m1x2_m(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vluxseg2ei16_v_bf16m2x2_m(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vluxseg2.mask.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> poison, <vscale x 8 x bfloat> poison, ptr [[RS1]], <vscale x 8 x i16> [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP0]]
+//
+vbfloat16m2x2_t test_vluxseg2ei16_v_bf16m2x2_m(vbool8_t vm, const __bf16 *rs1,
+                                               vuint16m2_t rs2, size_t vl) {
+  return __riscv_vluxseg2ei16_v_bf16m2x2_m(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @test_vluxseg2ei16_v_bf16m4x2_m(
+// CHECK-RV64-SAME: <vscale x 16 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 16 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @llvm.riscv.vluxseg2.mask.nxv16bf16.nxv16i16.i64(<vscale x 16 x bfloat> poison, <vscale x 16 x bfloat> poison, ptr [[RS1]], <vscale x 16 x i16> [[RS2]], <vscale x 16 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[TMP0]]
+//
+vbfloat16m4x2_t test_vluxseg2ei16_v_bf16m4x2_m(vbool4_t vm, const __bf16 *rs1,
+                                               vuint16m4_t rs2, size_t vl) {
+  return __riscv_vluxseg2ei16_v_bf16m4x2_m(vm, rs1, rs2, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vluxseg3ei16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vluxseg3ei16.c
new file mode 100644
index 000000000000..8d89952bea2b
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vluxseg3ei16.c
@@ -0,0 +1,99 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vluxseg3ei16_v_bf16mf4x3(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vluxseg3.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], <vscale x 1 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x3_t test_vluxseg3ei16_v_bf16mf4x3(const __bf16 *rs1,
+                                               vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vluxseg3ei16_v_bf16mf4x3(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vluxseg3ei16_v_bf16mf2x3(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vluxseg3.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], <vscale x 2 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x3_t test_vluxseg3ei16_v_bf16mf2x3(const __bf16 *rs1,
+                                               vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vluxseg3ei16_v_bf16mf2x3(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vluxseg3ei16_v_bf16m1x3(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vluxseg3.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], <vscale x 4 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x3_t test_vluxseg3ei16_v_bf16m1x3(const __bf16 *rs1, vuint16m1_t rs2,
+                                             size_t vl) {
+  return __riscv_vluxseg3ei16_v_bf16m1x3(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vluxseg3ei16_v_bf16m2x3(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vluxseg3.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> poison, <vscale x 8 x bfloat> poison, <vscale x 8 x bfloat> poison, ptr [[RS1]], <vscale x 8 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP0]]
+//
+vbfloat16m2x3_t test_vluxseg3ei16_v_bf16m2x3(const __bf16 *rs1, vuint16m2_t rs2,
+                                             size_t vl) {
+  return __riscv_vluxseg3ei16_v_bf16m2x3(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vluxseg3ei16_v_bf16mf4x3_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vluxseg3.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x3_t test_vluxseg3ei16_v_bf16mf4x3_m(vbool64_t vm,
+                                                 const __bf16 *rs1,
+                                                 vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vluxseg3ei16_v_bf16mf4x3_m(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vluxseg3ei16_v_bf16mf2x3_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vluxseg3.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x3_t test_vluxseg3ei16_v_bf16mf2x3_m(vbool32_t vm,
+                                                 const __bf16 *rs1,
+                                                 vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vluxseg3ei16_v_bf16mf2x3_m(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vluxseg3ei16_v_bf16m1x3_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vluxseg3.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x3_t test_vluxseg3ei16_v_bf16m1x3_m(vbool16_t vm, const __bf16 *rs1,
+                                               vuint16m1_t rs2, size_t vl) {
+  return __riscv_vluxseg3ei16_v_bf16m1x3_m(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vluxseg3ei16_v_bf16m2x3_m(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vluxseg3.mask.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> poison, <vscale x 8 x bfloat> poison, <vscale x 8 x bfloat> poison, ptr [[RS1]], <vscale x 8 x i16> [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP0]]
+//
+vbfloat16m2x3_t test_vluxseg3ei16_v_bf16m2x3_m(vbool8_t vm, const __bf16 *rs1,
+                                               vuint16m2_t rs2, size_t vl) {
+  return __riscv_vluxseg3ei16_v_bf16m2x3_m(vm, rs1, rs2, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vluxseg4ei16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vluxseg4ei16.c
new file mode 100644
index 000000000000..2751b1245f16
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vluxseg4ei16.c
@@ -0,0 +1,99 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vluxseg4ei16_v_bf16mf4x4(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vluxseg4.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], <vscale x 1 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x4_t test_vluxseg4ei16_v_bf16mf4x4(const __bf16 *rs1,
+                                               vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vluxseg4ei16_v_bf16mf4x4(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vluxseg4ei16_v_bf16mf2x4(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vluxseg4.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], <vscale x 2 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x4_t test_vluxseg4ei16_v_bf16mf2x4(const __bf16 *rs1,
+                                               vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vluxseg4ei16_v_bf16mf2x4(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vluxseg4ei16_v_bf16m1x4(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vluxseg4.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], <vscale x 4 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x4_t test_vluxseg4ei16_v_bf16m1x4(const __bf16 *rs1, vuint16m1_t rs2,
+                                             size_t vl) {
+  return __riscv_vluxseg4ei16_v_bf16m1x4(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vluxseg4ei16_v_bf16m2x4(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vluxseg4.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> poison, <vscale x 8 x bfloat> poison, <vscale x 8 x bfloat> poison, <vscale x 8 x bfloat> poison, ptr [[RS1]], <vscale x 8 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP0]]
+//
+vbfloat16m2x4_t test_vluxseg4ei16_v_bf16m2x4(const __bf16 *rs1, vuint16m2_t rs2,
+                                             size_t vl) {
+  return __riscv_vluxseg4ei16_v_bf16m2x4(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vluxseg4ei16_v_bf16mf4x4_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vluxseg4.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x4_t test_vluxseg4ei16_v_bf16mf4x4_m(vbool64_t vm,
+                                                 const __bf16 *rs1,
+                                                 vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vluxseg4ei16_v_bf16mf4x4_m(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vluxseg4ei16_v_bf16mf2x4_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vluxseg4.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x4_t test_vluxseg4ei16_v_bf16mf2x4_m(vbool32_t vm,
+                                                 const __bf16 *rs1,
+                                                 vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vluxseg4ei16_v_bf16mf2x4_m(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vluxseg4ei16_v_bf16m1x4_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vluxseg4.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x4_t test_vluxseg4ei16_v_bf16m1x4_m(vbool16_t vm, const __bf16 *rs1,
+                                               vuint16m1_t rs2, size_t vl) {
+  return __riscv_vluxseg4ei16_v_bf16m1x4_m(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vluxseg4ei16_v_bf16m2x4_m(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vluxseg4.mask.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> poison, <vscale x 8 x bfloat> poison, <vscale x 8 x bfloat> poison, <vscale x 8 x bfloat> poison, ptr [[RS1]], <vscale x 8 x i16> [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP0]]
+//
+vbfloat16m2x4_t test_vluxseg4ei16_v_bf16m2x4_m(vbool8_t vm, const __bf16 *rs1,
+                                               vuint16m2_t rs2, size_t vl) {
+  return __riscv_vluxseg4ei16_v_bf16m2x4_m(vm, rs1, rs2, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vluxseg5ei16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vluxseg5ei16.c
new file mode 100644
index 000000000000..9cd84e561dfa
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vluxseg5ei16.c
@@ -0,0 +1,77 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vluxseg5ei16_v_bf16mf4x5(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vluxseg5.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], <vscale x 1 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x5_t test_vluxseg5ei16_v_bf16mf4x5(const __bf16 *rs1,
+                                               vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vluxseg5ei16_v_bf16mf4x5(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vluxseg5ei16_v_bf16mf2x5(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vluxseg5.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], <vscale x 2 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x5_t test_vluxseg5ei16_v_bf16mf2x5(const __bf16 *rs1,
+                                               vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vluxseg5ei16_v_bf16mf2x5(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vluxseg5ei16_v_bf16m1x5(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vluxseg5.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], <vscale x 4 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x5_t test_vluxseg5ei16_v_bf16m1x5(const __bf16 *rs1, vuint16m1_t rs2,
+                                             size_t vl) {
+  return __riscv_vluxseg5ei16_v_bf16m1x5(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vluxseg5ei16_v_bf16mf4x5_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vluxseg5.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x5_t test_vluxseg5ei16_v_bf16mf4x5_m(vbool64_t vm,
+                                                 const __bf16 *rs1,
+                                                 vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vluxseg5ei16_v_bf16mf4x5_m(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vluxseg5ei16_v_bf16mf2x5_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vluxseg5.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x5_t test_vluxseg5ei16_v_bf16mf2x5_m(vbool32_t vm,
+                                                 const __bf16 *rs1,
+                                                 vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vluxseg5ei16_v_bf16mf2x5_m(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vluxseg5ei16_v_bf16m1x5_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vluxseg5.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x5_t test_vluxseg5ei16_v_bf16m1x5_m(vbool16_t vm, const __bf16 *rs1,
+                                               vuint16m1_t rs2, size_t vl) {
+  return __riscv_vluxseg5ei16_v_bf16m1x5_m(vm, rs1, rs2, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vluxseg6ei16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vluxseg6ei16.c
new file mode 100644
index 000000000000..95ce415c871c
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vluxseg6ei16.c
@@ -0,0 +1,77 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vluxseg6ei16_v_bf16mf4x6(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vluxseg6.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], <vscale x 1 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x6_t test_vluxseg6ei16_v_bf16mf4x6(const __bf16 *rs1,
+                                               vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vluxseg6ei16_v_bf16mf4x6(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vluxseg6ei16_v_bf16mf2x6(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vluxseg6.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], <vscale x 2 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x6_t test_vluxseg6ei16_v_bf16mf2x6(const __bf16 *rs1,
+                                               vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vluxseg6ei16_v_bf16mf2x6(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vluxseg6ei16_v_bf16m1x6(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vluxseg6.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], <vscale x 4 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x6_t test_vluxseg6ei16_v_bf16m1x6(const __bf16 *rs1, vuint16m1_t rs2,
+                                             size_t vl) {
+  return __riscv_vluxseg6ei16_v_bf16m1x6(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vluxseg6ei16_v_bf16mf4x6_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vluxseg6.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x6_t test_vluxseg6ei16_v_bf16mf4x6_m(vbool64_t vm,
+                                                 const __bf16 *rs1,
+                                                 vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vluxseg6ei16_v_bf16mf4x6_m(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vluxseg6ei16_v_bf16mf2x6_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vluxseg6.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x6_t test_vluxseg6ei16_v_bf16mf2x6_m(vbool32_t vm,
+                                                 const __bf16 *rs1,
+                                                 vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vluxseg6ei16_v_bf16mf2x6_m(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vluxseg6ei16_v_bf16m1x6_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vluxseg6.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x6_t test_vluxseg6ei16_v_bf16m1x6_m(vbool16_t vm, const __bf16 *rs1,
+                                               vuint16m1_t rs2, size_t vl) {
+  return __riscv_vluxseg6ei16_v_bf16m1x6_m(vm, rs1, rs2, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vluxseg7ei16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vluxseg7ei16.c
new file mode 100644
index 000000000000..c1b84c434e4e
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vluxseg7ei16.c
@@ -0,0 +1,77 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vluxseg7ei16_v_bf16mf4x7(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vluxseg7.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], <vscale x 1 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x7_t test_vluxseg7ei16_v_bf16mf4x7(const __bf16 *rs1,
+                                               vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vluxseg7ei16_v_bf16mf4x7(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vluxseg7ei16_v_bf16mf2x7(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vluxseg7.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], <vscale x 2 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x7_t test_vluxseg7ei16_v_bf16mf2x7(const __bf16 *rs1,
+                                               vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vluxseg7ei16_v_bf16mf2x7(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vluxseg7ei16_v_bf16m1x7(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vluxseg7.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], <vscale x 4 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x7_t test_vluxseg7ei16_v_bf16m1x7(const __bf16 *rs1, vuint16m1_t rs2,
+                                             size_t vl) {
+  return __riscv_vluxseg7ei16_v_bf16m1x7(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vluxseg7ei16_v_bf16mf4x7_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vluxseg7.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x7_t test_vluxseg7ei16_v_bf16mf4x7_m(vbool64_t vm,
+                                                 const __bf16 *rs1,
+                                                 vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vluxseg7ei16_v_bf16mf4x7_m(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vluxseg7ei16_v_bf16mf2x7_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vluxseg7.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x7_t test_vluxseg7ei16_v_bf16mf2x7_m(vbool32_t vm,
+                                                 const __bf16 *rs1,
+                                                 vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vluxseg7ei16_v_bf16mf2x7_m(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vluxseg7ei16_v_bf16m1x7_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vluxseg7.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x7_t test_vluxseg7ei16_v_bf16m1x7_m(vbool16_t vm, const __bf16 *rs1,
+                                               vuint16m1_t rs2, size_t vl) {
+  return __riscv_vluxseg7ei16_v_bf16m1x7_m(vm, rs1, rs2, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vluxseg8ei16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vluxseg8ei16.c
new file mode 100644
index 000000000000..32e5ab40e40a
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vluxseg8ei16.c
@@ -0,0 +1,77 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vluxseg8ei16_v_bf16mf4x8(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vluxseg8.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], <vscale x 1 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x8_t test_vluxseg8ei16_v_bf16mf4x8(const __bf16 *rs1,
+                                               vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vluxseg8ei16_v_bf16mf4x8(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vluxseg8ei16_v_bf16mf2x8(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vluxseg8.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], <vscale x 2 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x8_t test_vluxseg8ei16_v_bf16mf2x8(const __bf16 *rs1,
+                                               vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vluxseg8ei16_v_bf16mf2x8(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vluxseg8ei16_v_bf16m1x8(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vluxseg8.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], <vscale x 4 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x8_t test_vluxseg8ei16_v_bf16m1x8(const __bf16 *rs1, vuint16m1_t rs2,
+                                             size_t vl) {
+  return __riscv_vluxseg8ei16_v_bf16m1x8(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vluxseg8ei16_v_bf16mf4x8_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vluxseg8.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x8_t test_vluxseg8ei16_v_bf16mf4x8_m(vbool64_t vm,
+                                                 const __bf16 *rs1,
+                                                 vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vluxseg8ei16_v_bf16mf4x8_m(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vluxseg8ei16_v_bf16mf2x8_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vluxseg8.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x8_t test_vluxseg8ei16_v_bf16mf2x8_m(vbool32_t vm,
+                                                 const __bf16 *rs1,
+                                                 vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vluxseg8ei16_v_bf16mf2x8_m(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vluxseg8ei16_v_bf16m1x8_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vluxseg8.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x8_t test_vluxseg8ei16_v_bf16m1x8_m(vbool16_t vm, const __bf16 *rs1,
+                                               vuint16m1_t rs2, size_t vl) {
+  return __riscv_vluxseg8ei16_v_bf16m1x8_m(vm, rs1, rs2, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vreinterpret.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vreinterpret.c
new file mode 100644
index 000000000000..30120be47154
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vreinterpret.c
@@ -0,0 +1,249 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 1 x bfloat> @test_vreinterpret_v_i16mf4_bf16mf4(
+// CHECK-RV64-SAME: <vscale x 1 x i16> [[SRC:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = bitcast <vscale x 1 x i16> [[SRC]] to <vscale x 1 x bfloat>
+// CHECK-RV64-NEXT:    ret <vscale x 1 x bfloat> [[TMP0]]
+//
+vbfloat16mf4_t test_vreinterpret_v_i16mf4_bf16mf4(vint16mf4_t src) {
+  return __riscv_vreinterpret_v_i16mf4_bf16mf4(src);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 2 x bfloat> @test_vreinterpret_v_i16mf2_bf16mf2(
+// CHECK-RV64-SAME: <vscale x 2 x i16> [[SRC:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = bitcast <vscale x 2 x i16> [[SRC]] to <vscale x 2 x bfloat>
+// CHECK-RV64-NEXT:    ret <vscale x 2 x bfloat> [[TMP0]]
+//
+vbfloat16mf2_t test_vreinterpret_v_i16mf2_bf16mf2(vint16mf2_t src) {
+  return __riscv_vreinterpret_v_i16mf2_bf16mf2(src);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 4 x bfloat> @test_vreinterpret_v_i16m1_bf16m1(
+// CHECK-RV64-SAME: <vscale x 4 x i16> [[SRC:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = bitcast <vscale x 4 x i16> [[SRC]] to <vscale x 4 x bfloat>
+// CHECK-RV64-NEXT:    ret <vscale x 4 x bfloat> [[TMP0]]
+//
+vbfloat16m1_t test_vreinterpret_v_i16m1_bf16m1(vint16m1_t src) {
+  return __riscv_vreinterpret_v_i16m1_bf16m1(src);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 8 x bfloat> @test_vreinterpret_v_i16m2_bf16m2(
+// CHECK-RV64-SAME: <vscale x 8 x i16> [[SRC:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = bitcast <vscale x 8 x i16> [[SRC]] to <vscale x 8 x bfloat>
+// CHECK-RV64-NEXT:    ret <vscale x 8 x bfloat> [[TMP0]]
+//
+vbfloat16m2_t test_vreinterpret_v_i16m2_bf16m2(vint16m2_t src) {
+  return __riscv_vreinterpret_v_i16m2_bf16m2(src);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 16 x bfloat> @test_vreinterpret_v_i16m4_bf16m4(
+// CHECK-RV64-SAME: <vscale x 16 x i16> [[SRC:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = bitcast <vscale x 16 x i16> [[SRC]] to <vscale x 16 x bfloat>
+// CHECK-RV64-NEXT:    ret <vscale x 16 x bfloat> [[TMP0]]
+//
+vbfloat16m4_t test_vreinterpret_v_i16m4_bf16m4(vint16m4_t src) {
+  return __riscv_vreinterpret_v_i16m4_bf16m4(src);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 32 x bfloat> @test_vreinterpret_v_i16m8_bf16m8(
+// CHECK-RV64-SAME: <vscale x 32 x i16> [[SRC:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = bitcast <vscale x 32 x i16> [[SRC]] to <vscale x 32 x bfloat>
+// CHECK-RV64-NEXT:    ret <vscale x 32 x bfloat> [[TMP0]]
+//
+vbfloat16m8_t test_vreinterpret_v_i16m8_bf16m8(vint16m8_t src) {
+  return __riscv_vreinterpret_v_i16m8_bf16m8(src);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 1 x bfloat> @test_vreinterpret_v_u16mf4_bf16mf4(
+// CHECK-RV64-SAME: <vscale x 1 x i16> [[SRC:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = bitcast <vscale x 1 x i16> [[SRC]] to <vscale x 1 x bfloat>
+// CHECK-RV64-NEXT:    ret <vscale x 1 x bfloat> [[TMP0]]
+//
+vbfloat16mf4_t test_vreinterpret_v_u16mf4_bf16mf4(vuint16mf4_t src) {
+  return __riscv_vreinterpret_v_u16mf4_bf16mf4(src);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 2 x bfloat> @test_vreinterpret_v_u16mf2_bf16mf2(
+// CHECK-RV64-SAME: <vscale x 2 x i16> [[SRC:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = bitcast <vscale x 2 x i16> [[SRC]] to <vscale x 2 x bfloat>
+// CHECK-RV64-NEXT:    ret <vscale x 2 x bfloat> [[TMP0]]
+//
+vbfloat16mf2_t test_vreinterpret_v_u16mf2_bf16mf2(vuint16mf2_t src) {
+  return __riscv_vreinterpret_v_u16mf2_bf16mf2(src);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 4 x bfloat> @test_vreinterpret_v_u16m1_bf16m1(
+// CHECK-RV64-SAME: <vscale x 4 x i16> [[SRC:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = bitcast <vscale x 4 x i16> [[SRC]] to <vscale x 4 x bfloat>
+// CHECK-RV64-NEXT:    ret <vscale x 4 x bfloat> [[TMP0]]
+//
+vbfloat16m1_t test_vreinterpret_v_u16m1_bf16m1(vuint16m1_t src) {
+  return __riscv_vreinterpret_v_u16m1_bf16m1(src);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 8 x bfloat> @test_vreinterpret_v_u16m2_bf16m2(
+// CHECK-RV64-SAME: <vscale x 8 x i16> [[SRC:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = bitcast <vscale x 8 x i16> [[SRC]] to <vscale x 8 x bfloat>
+// CHECK-RV64-NEXT:    ret <vscale x 8 x bfloat> [[TMP0]]
+//
+vbfloat16m2_t test_vreinterpret_v_u16m2_bf16m2(vuint16m2_t src) {
+  return __riscv_vreinterpret_v_u16m2_bf16m2(src);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 16 x bfloat> @test_vreinterpret_v_u16m4_bf16m4(
+// CHECK-RV64-SAME: <vscale x 16 x i16> [[SRC:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = bitcast <vscale x 16 x i16> [[SRC]] to <vscale x 16 x bfloat>
+// CHECK-RV64-NEXT:    ret <vscale x 16 x bfloat> [[TMP0]]
+//
+vbfloat16m4_t test_vreinterpret_v_u16m4_bf16m4(vuint16m4_t src) {
+  return __riscv_vreinterpret_v_u16m4_bf16m4(src);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 32 x bfloat> @test_vreinterpret_v_u16m8_bf16m8(
+// CHECK-RV64-SAME: <vscale x 32 x i16> [[SRC:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = bitcast <vscale x 32 x i16> [[SRC]] to <vscale x 32 x bfloat>
+// CHECK-RV64-NEXT:    ret <vscale x 32 x bfloat> [[TMP0]]
+//
+vbfloat16m8_t test_vreinterpret_v_u16m8_bf16m8(vuint16m8_t src) {
+  return __riscv_vreinterpret_v_u16m8_bf16m8(src);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 1 x i16> @test_vreinterpret_v_bf16mf4_i16mf4(
+// CHECK-RV64-SAME: <vscale x 1 x bfloat> [[SRC:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = bitcast <vscale x 1 x bfloat> [[SRC]] to <vscale x 1 x i16>
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
+//
+vint16mf4_t test_vreinterpret_v_bf16mf4_i16mf4(vbfloat16mf4_t src) {
+  return __riscv_vreinterpret_v_bf16mf4_i16mf4(src);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 2 x i16> @test_vreinterpret_v_bf16mf2_i16mf2(
+// CHECK-RV64-SAME: <vscale x 2 x bfloat> [[SRC:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = bitcast <vscale x 2 x bfloat> [[SRC]] to <vscale x 2 x i16>
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
+//
+vint16mf2_t test_vreinterpret_v_bf16mf2_i16mf2(vbfloat16mf2_t src) {
+  return __riscv_vreinterpret_v_bf16mf2_i16mf2(src);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 4 x i16> @test_vreinterpret_v_bf16m1_i16m1(
+// CHECK-RV64-SAME: <vscale x 4 x bfloat> [[SRC:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = bitcast <vscale x 4 x bfloat> [[SRC]] to <vscale x 4 x i16>
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
+//
+vint16m1_t test_vreinterpret_v_bf16m1_i16m1(vbfloat16m1_t src) {
+  return __riscv_vreinterpret_v_bf16m1_i16m1(src);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 8 x i16> @test_vreinterpret_v_bf16m2_i16m2(
+// CHECK-RV64-SAME: <vscale x 8 x bfloat> [[SRC:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = bitcast <vscale x 8 x bfloat> [[SRC]] to <vscale x 8 x i16>
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
+//
+vint16m2_t test_vreinterpret_v_bf16m2_i16m2(vbfloat16m2_t src) {
+  return __riscv_vreinterpret_v_bf16m2_i16m2(src);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 16 x i16> @test_vreinterpret_v_bf16m4_i16m4(
+// CHECK-RV64-SAME: <vscale x 16 x bfloat> [[SRC:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = bitcast <vscale x 16 x bfloat> [[SRC]] to <vscale x 16 x i16>
+// CHECK-RV64-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
+//
+vint16m4_t test_vreinterpret_v_bf16m4_i16m4(vbfloat16m4_t src) {
+  return __riscv_vreinterpret_v_bf16m4_i16m4(src);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 32 x i16> @test_vreinterpret_v_bf16m8_i16m8(
+// CHECK-RV64-SAME: <vscale x 32 x bfloat> [[SRC:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = bitcast <vscale x 32 x bfloat> [[SRC]] to <vscale x 32 x i16>
+// CHECK-RV64-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
+//
+vint16m8_t test_vreinterpret_v_bf16m8_i16m8(vbfloat16m8_t src) {
+  return __riscv_vreinterpret_v_bf16m8_i16m8(src);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 1 x i16> @test_vreinterpret_v_bf16mf4_u16mf4(
+// CHECK-RV64-SAME: <vscale x 1 x bfloat> [[SRC:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = bitcast <vscale x 1 x bfloat> [[SRC]] to <vscale x 1 x i16>
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
+//
+vuint16mf4_t test_vreinterpret_v_bf16mf4_u16mf4(vbfloat16mf4_t src) {
+  return __riscv_vreinterpret_v_bf16mf4_u16mf4(src);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 2 x i16> @test_vreinterpret_v_bf16mf2_u16mf2(
+// CHECK-RV64-SAME: <vscale x 2 x bfloat> [[SRC:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = bitcast <vscale x 2 x bfloat> [[SRC]] to <vscale x 2 x i16>
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
+//
+vuint16mf2_t test_vreinterpret_v_bf16mf2_u16mf2(vbfloat16mf2_t src) {
+  return __riscv_vreinterpret_v_bf16mf2_u16mf2(src);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 4 x i16> @test_vreinterpret_v_bf16m1_u16m1(
+// CHECK-RV64-SAME: <vscale x 4 x bfloat> [[SRC:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = bitcast <vscale x 4 x bfloat> [[SRC]] to <vscale x 4 x i16>
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
+//
+vuint16m1_t test_vreinterpret_v_bf16m1_u16m1(vbfloat16m1_t src) {
+  return __riscv_vreinterpret_v_bf16m1_u16m1(src);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 8 x i16> @test_vreinterpret_v_bf16m2_u16m2(
+// CHECK-RV64-SAME: <vscale x 8 x bfloat> [[SRC:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = bitcast <vscale x 8 x bfloat> [[SRC]] to <vscale x 8 x i16>
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
+//
+vuint16m2_t test_vreinterpret_v_bf16m2_u16m2(vbfloat16m2_t src) {
+  return __riscv_vreinterpret_v_bf16m2_u16m2(src);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 16 x i16> @test_vreinterpret_v_bf16m4_u16m4(
+// CHECK-RV64-SAME: <vscale x 16 x bfloat> [[SRC:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = bitcast <vscale x 16 x bfloat> [[SRC]] to <vscale x 16 x i16>
+// CHECK-RV64-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
+//
+vuint16m4_t test_vreinterpret_v_bf16m4_u16m4(vbfloat16m4_t src) {
+  return __riscv_vreinterpret_v_bf16m4_u16m4(src);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 32 x i16> @test_vreinterpret_v_bf16m8_u16m8(
+// CHECK-RV64-SAME: <vscale x 32 x bfloat> [[SRC:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = bitcast <vscale x 32 x bfloat> [[SRC]] to <vscale x 32 x i16>
+// CHECK-RV64-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
+//
+vuint16m8_t test_vreinterpret_v_bf16m8_u16m8(vbfloat16m8_t src) {
+  return __riscv_vreinterpret_v_bf16m8_u16m8(src);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vse16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vse16.c
new file mode 100644
index 000000000000..4bfc5b7350cb
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vse16.c
@@ -0,0 +1,135 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local void @test_vse16_v_bf16mf4(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 1 x bfloat> [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vse.nxv1bf16.i64(<vscale x 1 x bfloat> [[VS3]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vse16_v_bf16mf4(__bf16 *rs1, vbfloat16mf4_t vs3, size_t vl) {
+  return __riscv_vse16_v_bf16mf4(rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vse16_v_bf16mf2(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 2 x bfloat> [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vse.nxv2bf16.i64(<vscale x 2 x bfloat> [[VS3]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vse16_v_bf16mf2(__bf16 *rs1, vbfloat16mf2_t vs3, size_t vl) {
+  return __riscv_vse16_v_bf16mf2(rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vse16_v_bf16m1(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 4 x bfloat> [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vse.nxv4bf16.i64(<vscale x 4 x bfloat> [[VS3]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vse16_v_bf16m1(__bf16 *rs1, vbfloat16m1_t vs3, size_t vl) {
+  return __riscv_vse16_v_bf16m1(rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vse16_v_bf16m2(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 8 x bfloat> [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vse.nxv8bf16.i64(<vscale x 8 x bfloat> [[VS3]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vse16_v_bf16m2(__bf16 *rs1, vbfloat16m2_t vs3, size_t vl) {
+  return __riscv_vse16_v_bf16m2(rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vse16_v_bf16m4(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 16 x bfloat> [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vse.nxv16bf16.i64(<vscale x 16 x bfloat> [[VS3]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vse16_v_bf16m4(__bf16 *rs1, vbfloat16m4_t vs3, size_t vl) {
+  return __riscv_vse16_v_bf16m4(rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vse16_v_bf16m8(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 32 x bfloat> [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vse.nxv32bf16.i64(<vscale x 32 x bfloat> [[VS3]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vse16_v_bf16m8(__bf16 *rs1, vbfloat16m8_t vs3, size_t vl) {
+  return __riscv_vse16_v_bf16m8(rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vse16_v_bf16mf4_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x bfloat> [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vse.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[VS3]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vse16_v_bf16mf4_m(vbool64_t vm, __bf16 *rs1, vbfloat16mf4_t vs3,
+                            size_t vl) {
+  return __riscv_vse16_v_bf16mf4_m(vm, rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vse16_v_bf16mf2_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x bfloat> [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vse.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[VS3]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vse16_v_bf16mf2_m(vbool32_t vm, __bf16 *rs1, vbfloat16mf2_t vs3,
+                            size_t vl) {
+  return __riscv_vse16_v_bf16mf2_m(vm, rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vse16_v_bf16m1_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x bfloat> [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vse.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[VS3]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vse16_v_bf16m1_m(vbool16_t vm, __bf16 *rs1, vbfloat16m1_t vs3,
+                           size_t vl) {
+  return __riscv_vse16_v_bf16m1_m(vm, rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vse16_v_bf16m2_m(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 8 x bfloat> [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vse.mask.nxv8bf16.i64(<vscale x 8 x bfloat> [[VS3]], ptr [[RS1]], <vscale x 8 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vse16_v_bf16m2_m(vbool8_t vm, __bf16 *rs1, vbfloat16m2_t vs3,
+                           size_t vl) {
+  return __riscv_vse16_v_bf16m2_m(vm, rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vse16_v_bf16m4_m(
+// CHECK-RV64-SAME: <vscale x 16 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 16 x bfloat> [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vse.mask.nxv16bf16.i64(<vscale x 16 x bfloat> [[VS3]], ptr [[RS1]], <vscale x 16 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vse16_v_bf16m4_m(vbool4_t vm, __bf16 *rs1, vbfloat16m4_t vs3,
+                           size_t vl) {
+  return __riscv_vse16_v_bf16m4_m(vm, rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vse16_v_bf16m8_m(
+// CHECK-RV64-SAME: <vscale x 32 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 32 x bfloat> [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vse.mask.nxv32bf16.i64(<vscale x 32 x bfloat> [[VS3]], ptr [[RS1]], <vscale x 32 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vse16_v_bf16m8_m(vbool2_t vm, __bf16 *rs1, vbfloat16m8_t vs3,
+                           size_t vl) {
+  return __riscv_vse16_v_bf16m8_m(vm, rs1, vs3, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vset.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vset.c
new file mode 100644
index 000000000000..779c24bd6469
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vset.c
@@ -0,0 +1,364 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 8 x bfloat> @test_vset_v_bf16m1_bf16m2(
+// CHECK-RV64-SAME: <vscale x 8 x bfloat> [[DEST:%.*]], i64 noundef [[INDEX:%.*]], <vscale x 4 x bfloat> [[VALUE:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x bfloat> @llvm.vector.insert.nxv8bf16.nxv4bf16(<vscale x 8 x bfloat> [[DEST]], <vscale x 4 x bfloat> [[VALUE]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 8 x bfloat> [[TMP0]]
+//
+vbfloat16m2_t test_vset_v_bf16m1_bf16m2(vbfloat16m2_t dest, size_t index,
+                                        vbfloat16m1_t value) {
+  return __riscv_vset_v_bf16m1_bf16m2(dest, 0, value);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 16 x bfloat> @test_vset_v_bf16m1_bf16m4(
+// CHECK-RV64-SAME: <vscale x 16 x bfloat> [[DEST:%.*]], i64 noundef [[INDEX:%.*]], <vscale x 4 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x bfloat> @llvm.vector.insert.nxv16bf16.nxv4bf16(<vscale x 16 x bfloat> [[DEST]], <vscale x 4 x bfloat> [[VALUE]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 16 x bfloat> [[TMP0]]
+//
+vbfloat16m4_t test_vset_v_bf16m1_bf16m4(vbfloat16m4_t dest, size_t index,
+                                        vbfloat16m1_t value) {
+  return __riscv_vset_v_bf16m1_bf16m4(dest, 0, value);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 16 x bfloat> @test_vset_v_bf16m2_bf16m4(
+// CHECK-RV64-SAME: <vscale x 16 x bfloat> [[DEST:%.*]], i64 noundef [[INDEX:%.*]], <vscale x 8 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x bfloat> @llvm.vector.insert.nxv16bf16.nxv8bf16(<vscale x 16 x bfloat> [[DEST]], <vscale x 8 x bfloat> [[VALUE]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 16 x bfloat> [[TMP0]]
+//
+vbfloat16m4_t test_vset_v_bf16m2_bf16m4(vbfloat16m4_t dest, size_t index,
+                                        vbfloat16m2_t value) {
+  return __riscv_vset_v_bf16m2_bf16m4(dest, 0, value);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 32 x bfloat> @test_vset_v_bf16m1_bf16m8(
+// CHECK-RV64-SAME: <vscale x 32 x bfloat> [[DEST:%.*]], i64 noundef [[INDEX:%.*]], <vscale x 4 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x bfloat> @llvm.vector.insert.nxv32bf16.nxv4bf16(<vscale x 32 x bfloat> [[DEST]], <vscale x 4 x bfloat> [[VALUE]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 32 x bfloat> [[TMP0]]
+//
+vbfloat16m8_t test_vset_v_bf16m1_bf16m8(vbfloat16m8_t dest, size_t index,
+                                        vbfloat16m1_t value) {
+  return __riscv_vset_v_bf16m1_bf16m8(dest, 0, value);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 32 x bfloat> @test_vset_v_bf16m2_bf16m8(
+// CHECK-RV64-SAME: <vscale x 32 x bfloat> [[DEST:%.*]], i64 noundef [[INDEX:%.*]], <vscale x 8 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x bfloat> @llvm.vector.insert.nxv32bf16.nxv8bf16(<vscale x 32 x bfloat> [[DEST]], <vscale x 8 x bfloat> [[VALUE]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 32 x bfloat> [[TMP0]]
+//
+vbfloat16m8_t test_vset_v_bf16m2_bf16m8(vbfloat16m8_t dest, size_t index,
+                                        vbfloat16m2_t value) {
+  return __riscv_vset_v_bf16m2_bf16m8(dest, 0, value);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 32 x bfloat> @test_vset_v_bf16m4_bf16m8(
+// CHECK-RV64-SAME: <vscale x 32 x bfloat> [[DEST:%.*]], i64 noundef [[INDEX:%.*]], <vscale x 16 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x bfloat> @llvm.vector.insert.nxv32bf16.nxv16bf16(<vscale x 32 x bfloat> [[DEST]], <vscale x 16 x bfloat> [[VALUE]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 32 x bfloat> [[TMP0]]
+//
+vbfloat16m8_t test_vset_v_bf16m4_bf16m8(vbfloat16m8_t dest, size_t index,
+                                        vbfloat16m4_t value) {
+  return __riscv_vset_v_bf16m4_bf16m8(dest, 0, value);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vset_v_bf16mf4_bf16mf4x2(
+// CHECK-RV64-SAME: { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[DEST:%.*]], i64 noundef [[INDEX:%.*]], <vscale x 1 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[DEST]], <vscale x 1 x bfloat> [[VALUE]], 0
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x2_t test_vset_v_bf16mf4_bf16mf4x2(vbfloat16mf4x2_t dest,
+                                               size_t index,
+                                               vbfloat16mf4_t value) {
+  return __riscv_vset_v_bf16mf4_bf16mf4x2(dest, 0, value);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vset_v_bf16mf4_bf16mf4x3(
+// CHECK-RV64-SAME: { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[DEST:%.*]], i64 noundef [[INDEX:%.*]], <vscale x 1 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[DEST]], <vscale x 1 x bfloat> [[VALUE]], 0
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x3_t test_vset_v_bf16mf4_bf16mf4x3(vbfloat16mf4x3_t dest,
+                                               size_t index,
+                                               vbfloat16mf4_t value) {
+  return __riscv_vset_v_bf16mf4_bf16mf4x3(dest, 0, value);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vset_v_bf16mf4_bf16mf4x4(
+// CHECK-RV64-SAME: { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[DEST:%.*]], i64 noundef [[INDEX:%.*]], <vscale x 1 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[DEST]], <vscale x 1 x bfloat> [[VALUE]], 0
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x4_t test_vset_v_bf16mf4_bf16mf4x4(vbfloat16mf4x4_t dest,
+                                               size_t index,
+                                               vbfloat16mf4_t value) {
+  return __riscv_vset_v_bf16mf4_bf16mf4x4(dest, 0, value);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vset_v_bf16mf4_bf16mf4x5(
+// CHECK-RV64-SAME: { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[DEST:%.*]], i64 noundef [[INDEX:%.*]], <vscale x 1 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[DEST]], <vscale x 1 x bfloat> [[VALUE]], 0
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x5_t test_vset_v_bf16mf4_bf16mf4x5(vbfloat16mf4x5_t dest,
+                                               size_t index,
+                                               vbfloat16mf4_t value) {
+  return __riscv_vset_v_bf16mf4_bf16mf4x5(dest, 0, value);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vset_v_bf16mf4_bf16mf4x6(
+// CHECK-RV64-SAME: { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[DEST:%.*]], i64 noundef [[INDEX:%.*]], <vscale x 1 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[DEST]], <vscale x 1 x bfloat> [[VALUE]], 0
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x6_t test_vset_v_bf16mf4_bf16mf4x6(vbfloat16mf4x6_t dest,
+                                               size_t index,
+                                               vbfloat16mf4_t value) {
+  return __riscv_vset_v_bf16mf4_bf16mf4x6(dest, 0, value);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vset_v_bf16mf4_bf16mf4x7(
+// CHECK-RV64-SAME: { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[DEST:%.*]], i64 noundef [[INDEX:%.*]], <vscale x 1 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[DEST]], <vscale x 1 x bfloat> [[VALUE]], 0
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x7_t test_vset_v_bf16mf4_bf16mf4x7(vbfloat16mf4x7_t dest,
+                                               size_t index,
+                                               vbfloat16mf4_t value) {
+  return __riscv_vset_v_bf16mf4_bf16mf4x7(dest, 0, value);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vset_v_bf16mf4_bf16mf4x8(
+// CHECK-RV64-SAME: { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[DEST:%.*]], i64 noundef [[INDEX:%.*]], <vscale x 1 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[DEST]], <vscale x 1 x bfloat> [[VALUE]], 0
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x8_t test_vset_v_bf16mf4_bf16mf4x8(vbfloat16mf4x8_t dest,
+                                               size_t index,
+                                               vbfloat16mf4_t value) {
+  return __riscv_vset_v_bf16mf4_bf16mf4x8(dest, 0, value);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vset_v_bf16mf2_bf16mf2x2(
+// CHECK-RV64-SAME: { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[DEST:%.*]], i64 noundef [[INDEX:%.*]], <vscale x 2 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[DEST]], <vscale x 2 x bfloat> [[VALUE]], 0
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x2_t test_vset_v_bf16mf2_bf16mf2x2(vbfloat16mf2x2_t dest,
+                                               size_t index,
+                                               vbfloat16mf2_t value) {
+  return __riscv_vset_v_bf16mf2_bf16mf2x2(dest, 0, value);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vset_v_bf16mf2_bf16mf2x3(
+// CHECK-RV64-SAME: { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[DEST:%.*]], i64 noundef [[INDEX:%.*]], <vscale x 2 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[DEST]], <vscale x 2 x bfloat> [[VALUE]], 0
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x3_t test_vset_v_bf16mf2_bf16mf2x3(vbfloat16mf2x3_t dest,
+                                               size_t index,
+                                               vbfloat16mf2_t value) {
+  return __riscv_vset_v_bf16mf2_bf16mf2x3(dest, 0, value);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vset_v_bf16mf2_bf16mf2x4(
+// CHECK-RV64-SAME: { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[DEST:%.*]], i64 noundef [[INDEX:%.*]], <vscale x 2 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[DEST]], <vscale x 2 x bfloat> [[VALUE]], 0
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x4_t test_vset_v_bf16mf2_bf16mf2x4(vbfloat16mf2x4_t dest,
+                                               size_t index,
+                                               vbfloat16mf2_t value) {
+  return __riscv_vset_v_bf16mf2_bf16mf2x4(dest, 0, value);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vset_v_bf16mf2_bf16mf2x5(
+// CHECK-RV64-SAME: { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[DEST:%.*]], i64 noundef [[INDEX:%.*]], <vscale x 2 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[DEST]], <vscale x 2 x bfloat> [[VALUE]], 0
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x5_t test_vset_v_bf16mf2_bf16mf2x5(vbfloat16mf2x5_t dest,
+                                               size_t index,
+                                               vbfloat16mf2_t value) {
+  return __riscv_vset_v_bf16mf2_bf16mf2x5(dest, 0, value);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vset_v_bf16mf2_bf16mf2x6(
+// CHECK-RV64-SAME: { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[DEST:%.*]], i64 noundef [[INDEX:%.*]], <vscale x 2 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[DEST]], <vscale x 2 x bfloat> [[VALUE]], 0
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x6_t test_vset_v_bf16mf2_bf16mf2x6(vbfloat16mf2x6_t dest,
+                                               size_t index,
+                                               vbfloat16mf2_t value) {
+  return __riscv_vset_v_bf16mf2_bf16mf2x6(dest, 0, value);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vset_v_bf16mf2_bf16mf2x7(
+// CHECK-RV64-SAME: { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[DEST:%.*]], i64 noundef [[INDEX:%.*]], <vscale x 2 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[DEST]], <vscale x 2 x bfloat> [[VALUE]], 0
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x7_t test_vset_v_bf16mf2_bf16mf2x7(vbfloat16mf2x7_t dest,
+                                               size_t index,
+                                               vbfloat16mf2_t value) {
+  return __riscv_vset_v_bf16mf2_bf16mf2x7(dest, 0, value);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vset_v_bf16mf2_bf16mf2x8(
+// CHECK-RV64-SAME: { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[DEST:%.*]], i64 noundef [[INDEX:%.*]], <vscale x 2 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[DEST]], <vscale x 2 x bfloat> [[VALUE]], 0
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x8_t test_vset_v_bf16mf2_bf16mf2x8(vbfloat16mf2x8_t dest,
+                                               size_t index,
+                                               vbfloat16mf2_t value) {
+  return __riscv_vset_v_bf16mf2_bf16mf2x8(dest, 0, value);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vset_v_bf16m1_bf16m1x2(
+// CHECK-RV64-SAME: { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[DEST:%.*]], i64 noundef [[INDEX:%.*]], <vscale x 4 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[DEST]], <vscale x 4 x bfloat> [[VALUE]], 0
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x2_t test_vset_v_bf16m1_bf16m1x2(vbfloat16m1x2_t dest, size_t index,
+                                            vbfloat16m1_t value) {
+  return __riscv_vset_v_bf16m1_bf16m1x2(dest, 0, value);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vset_v_bf16m1_bf16m1x3(
+// CHECK-RV64-SAME: { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[DEST:%.*]], i64 noundef [[INDEX:%.*]], <vscale x 4 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[DEST]], <vscale x 4 x bfloat> [[VALUE]], 0
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x3_t test_vset_v_bf16m1_bf16m1x3(vbfloat16m1x3_t dest, size_t index,
+                                            vbfloat16m1_t value) {
+  return __riscv_vset_v_bf16m1_bf16m1x3(dest, 0, value);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vset_v_bf16m1_bf16m1x4(
+// CHECK-RV64-SAME: { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[DEST:%.*]], i64 noundef [[INDEX:%.*]], <vscale x 4 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[DEST]], <vscale x 4 x bfloat> [[VALUE]], 0
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x4_t test_vset_v_bf16m1_bf16m1x4(vbfloat16m1x4_t dest, size_t index,
+                                            vbfloat16m1_t value) {
+  return __riscv_vset_v_bf16m1_bf16m1x4(dest, 0, value);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vset_v_bf16m1_bf16m1x5(
+// CHECK-RV64-SAME: { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[DEST:%.*]], i64 noundef [[INDEX:%.*]], <vscale x 4 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[DEST]], <vscale x 4 x bfloat> [[VALUE]], 0
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x5_t test_vset_v_bf16m1_bf16m1x5(vbfloat16m1x5_t dest, size_t index,
+                                            vbfloat16m1_t value) {
+  return __riscv_vset_v_bf16m1_bf16m1x5(dest, 0, value);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vset_v_bf16m1_bf16m1x6(
+// CHECK-RV64-SAME: { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[DEST:%.*]], i64 noundef [[INDEX:%.*]], <vscale x 4 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[DEST]], <vscale x 4 x bfloat> [[VALUE]], 0
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x6_t test_vset_v_bf16m1_bf16m1x6(vbfloat16m1x6_t dest, size_t index,
+                                            vbfloat16m1_t value) {
+  return __riscv_vset_v_bf16m1_bf16m1x6(dest, 0, value);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vset_v_bf16m1_bf16m1x7(
+// CHECK-RV64-SAME: { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[DEST:%.*]], i64 noundef [[INDEX:%.*]], <vscale x 4 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[DEST]], <vscale x 4 x bfloat> [[VALUE]], 0
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x7_t test_vset_v_bf16m1_bf16m1x7(vbfloat16m1x7_t dest, size_t index,
+                                            vbfloat16m1_t value) {
+  return __riscv_vset_v_bf16m1_bf16m1x7(dest, 0, value);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vset_v_bf16m1_bf16m1x8(
+// CHECK-RV64-SAME: { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[DEST:%.*]], i64 noundef [[INDEX:%.*]], <vscale x 4 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[DEST]], <vscale x 4 x bfloat> [[VALUE]], 0
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x8_t test_vset_v_bf16m1_bf16m1x8(vbfloat16m1x8_t dest, size_t index,
+                                            vbfloat16m1_t value) {
+  return __riscv_vset_v_bf16m1_bf16m1x8(dest, 0, value);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vset_v_bf16m2_bf16m2x2(
+// CHECK-RV64-SAME: { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[DEST:%.*]], i64 noundef [[INDEX:%.*]], <vscale x 8 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[DEST]], <vscale x 8 x bfloat> [[VALUE]], 0
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP0]]
+//
+vbfloat16m2x2_t test_vset_v_bf16m2_bf16m2x2(vbfloat16m2x2_t dest, size_t index,
+                                            vbfloat16m2_t value) {
+  return __riscv_vset_v_bf16m2_bf16m2x2(dest, 0, value);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vset_v_bf16m2_bf16m2x3(
+// CHECK-RV64-SAME: { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[DEST:%.*]], i64 noundef [[INDEX:%.*]], <vscale x 8 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[DEST]], <vscale x 8 x bfloat> [[VALUE]], 0
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP0]]
+//
+vbfloat16m2x3_t test_vset_v_bf16m2_bf16m2x3(vbfloat16m2x3_t dest, size_t index,
+                                            vbfloat16m2_t value) {
+  return __riscv_vset_v_bf16m2_bf16m2x3(dest, 0, value);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vset_v_bf16m2_bf16m2x4(
+// CHECK-RV64-SAME: { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[DEST:%.*]], i64 noundef [[INDEX:%.*]], <vscale x 8 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[DEST]], <vscale x 8 x bfloat> [[VALUE]], 0
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP0]]
+//
+vbfloat16m2x4_t test_vset_v_bf16m2_bf16m2x4(vbfloat16m2x4_t dest, size_t index,
+                                            vbfloat16m2_t value) {
+  return __riscv_vset_v_bf16m2_bf16m2x4(dest, 0, value);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @test_vset_v_bf16m4_bf16m4x2(
+// CHECK-RV64-SAME: { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[DEST:%.*]], i64 noundef [[INDEX:%.*]], <vscale x 16 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = insertvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[DEST]], <vscale x 16 x bfloat> [[VALUE]], 0
+// CHECK-RV64-NEXT:    ret { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[TMP0]]
+//
+vbfloat16m4x2_t test_vset_v_bf16m4_bf16m4x2(vbfloat16m4x2_t dest, size_t index,
+                                            vbfloat16m4_t value) {
+  return __riscv_vset_v_bf16m4_bf16m4x2(dest, 0, value);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vsoxei16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vsoxei16.c
new file mode 100644
index 000000000000..dc3c25f4cbe8
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vsoxei16.c
@@ -0,0 +1,141 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxei16_v_bf16mf4(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], <vscale x 1 x bfloat> [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxei.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[VS3]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxei16_v_bf16mf4(__bf16 *rs1, vuint16mf4_t rs2, vbfloat16mf4_t vs3,
+                             size_t vl) {
+  return __riscv_vsoxei16_v_bf16mf4(rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxei16_v_bf16mf2(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], <vscale x 2 x bfloat> [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxei.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[VS3]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxei16_v_bf16mf2(__bf16 *rs1, vuint16mf2_t rs2, vbfloat16mf2_t vs3,
+                             size_t vl) {
+  return __riscv_vsoxei16_v_bf16mf2(rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxei16_v_bf16m1(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], <vscale x 4 x bfloat> [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxei.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[VS3]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxei16_v_bf16m1(__bf16 *rs1, vuint16m1_t rs2, vbfloat16m1_t vs3,
+                            size_t vl) {
+  return __riscv_vsoxei16_v_bf16m1(rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxei16_v_bf16m2(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[RS2:%.*]], <vscale x 8 x bfloat> [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxei.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> [[VS3]], ptr [[RS1]], <vscale x 8 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxei16_v_bf16m2(__bf16 *rs1, vuint16m2_t rs2, vbfloat16m2_t vs3,
+                            size_t vl) {
+  return __riscv_vsoxei16_v_bf16m2(rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxei16_v_bf16m4(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 16 x i16> [[RS2:%.*]], <vscale x 16 x bfloat> [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxei.nxv16bf16.nxv16i16.i64(<vscale x 16 x bfloat> [[VS3]], ptr [[RS1]], <vscale x 16 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxei16_v_bf16m4(__bf16 *rs1, vuint16m4_t rs2, vbfloat16m4_t vs3,
+                            size_t vl) {
+  return __riscv_vsoxei16_v_bf16m4(rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxei16_v_bf16m8(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 32 x i16> [[RS2:%.*]], <vscale x 32 x bfloat> [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxei.nxv32bf16.nxv32i16.i64(<vscale x 32 x bfloat> [[VS3]], ptr [[RS1]], <vscale x 32 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxei16_v_bf16m8(__bf16 *rs1, vuint16m8_t rs2, vbfloat16m8_t vs3,
+                            size_t vl) {
+  return __riscv_vsoxei16_v_bf16m8(rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxei16_v_bf16mf4_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], <vscale x 1 x bfloat> [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxei.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[VS3]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxei16_v_bf16mf4_m(vbool64_t vm, __bf16 *rs1, vuint16mf4_t rs2,
+                               vbfloat16mf4_t vs3, size_t vl) {
+  return __riscv_vsoxei16_v_bf16mf4_m(vm, rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxei16_v_bf16mf2_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], <vscale x 2 x bfloat> [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxei.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[VS3]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxei16_v_bf16mf2_m(vbool32_t vm, __bf16 *rs1, vuint16mf2_t rs2,
+                               vbfloat16mf2_t vs3, size_t vl) {
+  return __riscv_vsoxei16_v_bf16mf2_m(vm, rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxei16_v_bf16m1_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], <vscale x 4 x bfloat> [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxei.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[VS3]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxei16_v_bf16m1_m(vbool16_t vm, __bf16 *rs1, vuint16m1_t rs2,
+                              vbfloat16m1_t vs3, size_t vl) {
+  return __riscv_vsoxei16_v_bf16m1_m(vm, rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxei16_v_bf16m2_m(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[RS2:%.*]], <vscale x 8 x bfloat> [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxei.mask.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> [[VS3]], ptr [[RS1]], <vscale x 8 x i16> [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxei16_v_bf16m2_m(vbool8_t vm, __bf16 *rs1, vuint16m2_t rs2,
+                              vbfloat16m2_t vs3, size_t vl) {
+  return __riscv_vsoxei16_v_bf16m2_m(vm, rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxei16_v_bf16m4_m(
+// CHECK-RV64-SAME: <vscale x 16 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 16 x i16> [[RS2:%.*]], <vscale x 16 x bfloat> [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxei.mask.nxv16bf16.nxv16i16.i64(<vscale x 16 x bfloat> [[VS3]], ptr [[RS1]], <vscale x 16 x i16> [[RS2]], <vscale x 16 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxei16_v_bf16m4_m(vbool4_t vm, __bf16 *rs1, vuint16m4_t rs2,
+                              vbfloat16m4_t vs3, size_t vl) {
+  return __riscv_vsoxei16_v_bf16m4_m(vm, rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxei16_v_bf16m8_m(
+// CHECK-RV64-SAME: <vscale x 32 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 32 x i16> [[RS2:%.*]], <vscale x 32 x bfloat> [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxei.mask.nxv32bf16.nxv32i16.i64(<vscale x 32 x bfloat> [[VS3]], ptr [[RS1]], <vscale x 32 x i16> [[RS2]], <vscale x 32 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxei16_v_bf16m8_m(vbool2_t vm, __bf16 *rs1, vuint16m8_t rs2,
+                              vbfloat16m8_t vs3, size_t vl) {
+  return __riscv_vsoxei16_v_bf16m8_m(vm, rs1, rs2, vs3, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vsoxseg2ei16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vsoxseg2ei16.c
new file mode 100644
index 000000000000..a98ca6ae63a6
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vsoxseg2ei16.c
@@ -0,0 +1,141 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxseg2ei16_v_bf16mf4x2(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[VS2:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxseg2.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 1 x i16> [[VS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxseg2ei16_v_bf16mf4x2(__bf16 *rs1, vuint16mf4_t vs2,
+                                   vbfloat16mf4x2_t vs3, size_t vl) {
+  return __riscv_vsoxseg2ei16_v_bf16mf4x2(rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxseg2ei16_v_bf16mf2x2(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[VS2:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxseg2.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 2 x i16> [[VS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxseg2ei16_v_bf16mf2x2(__bf16 *rs1, vuint16mf2_t vs2,
+                                   vbfloat16mf2x2_t vs3, size_t vl) {
+  return __riscv_vsoxseg2ei16_v_bf16mf2x2(rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxseg2ei16_v_bf16m1x2(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[VS2:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxseg2.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 4 x i16> [[VS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxseg2ei16_v_bf16m1x2(__bf16 *rs1, vuint16m1_t vs2,
+                                  vbfloat16m1x2_t vs3, size_t vl) {
+  return __riscv_vsoxseg2ei16_v_bf16m1x2(rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxseg2ei16_v_bf16m2x2(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[VS2:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxseg2.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 8 x i16> [[VS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxseg2ei16_v_bf16m2x2(__bf16 *rs1, vuint16m2_t vs2,
+                                  vbfloat16m2x2_t vs3, size_t vl) {
+  return __riscv_vsoxseg2ei16_v_bf16m2x2(rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxseg2ei16_v_bf16m4x2(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 16 x i16> [[VS2:%.*]], { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxseg2.nxv16bf16.nxv16i16.i64(<vscale x 16 x bfloat> [[TMP0]], <vscale x 16 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 16 x i16> [[VS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxseg2ei16_v_bf16m4x2(__bf16 *rs1, vuint16m4_t vs2,
+                                  vbfloat16m4x2_t vs3, size_t vl) {
+  return __riscv_vsoxseg2ei16_v_bf16m4x2(rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxseg2ei16_v_bf16mf4x2_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[VS2:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxseg2.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 1 x i16> [[VS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxseg2ei16_v_bf16mf4x2_m(vbool64_t vm, __bf16 *rs1,
+                                     vuint16mf4_t vs2, vbfloat16mf4x2_t vs3,
+                                     size_t vl) {
+  return __riscv_vsoxseg2ei16_v_bf16mf4x2_m(vm, rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxseg2ei16_v_bf16mf2x2_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[VS2:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxseg2.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 2 x i16> [[VS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxseg2ei16_v_bf16mf2x2_m(vbool32_t vm, __bf16 *rs1,
+                                     vuint16mf2_t vs2, vbfloat16mf2x2_t vs3,
+                                     size_t vl) {
+  return __riscv_vsoxseg2ei16_v_bf16mf2x2_m(vm, rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxseg2ei16_v_bf16m1x2_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[VS2:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxseg2.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 4 x i16> [[VS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxseg2ei16_v_bf16m1x2_m(vbool16_t vm, __bf16 *rs1, vuint16m1_t vs2,
+                                    vbfloat16m1x2_t vs3, size_t vl) {
+  return __riscv_vsoxseg2ei16_v_bf16m1x2_m(vm, rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxseg2ei16_v_bf16m2x2_m(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[VS2:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxseg2.mask.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 8 x i16> [[VS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxseg2ei16_v_bf16m2x2_m(vbool8_t vm, __bf16 *rs1, vuint16m2_t vs2,
+                                    vbfloat16m2x2_t vs3, size_t vl) {
+  return __riscv_vsoxseg2ei16_v_bf16m2x2_m(vm, rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxseg2ei16_v_bf16m4x2_m(
+// CHECK-RV64-SAME: <vscale x 16 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 16 x i16> [[VS2:%.*]], { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxseg2.mask.nxv16bf16.nxv16i16.i64(<vscale x 16 x bfloat> [[TMP0]], <vscale x 16 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 16 x i16> [[VS2]], <vscale x 16 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxseg2ei16_v_bf16m4x2_m(vbool4_t vm, __bf16 *rs1, vuint16m4_t vs2,
+                                    vbfloat16m4x2_t vs3, size_t vl) {
+  return __riscv_vsoxseg2ei16_v_bf16m4x2_m(vm, rs1, vs2, vs3, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vsoxseg3ei16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vsoxseg3ei16.c
new file mode 100644
index 000000000000..0f08957b30a7
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vsoxseg3ei16.c
@@ -0,0 +1,123 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxseg3ei16_v_bf16mf4x3(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[VS2:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxseg3.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 1 x i16> [[VS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxseg3ei16_v_bf16mf4x3(__bf16 *rs1, vuint16mf4_t vs2,
+                                   vbfloat16mf4x3_t vs3, size_t vl) {
+  return __riscv_vsoxseg3ei16_v_bf16mf4x3(rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxseg3ei16_v_bf16mf2x3(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[VS2:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxseg3.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 2 x i16> [[VS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxseg3ei16_v_bf16mf2x3(__bf16 *rs1, vuint16mf2_t vs2,
+                                   vbfloat16mf2x3_t vs3, size_t vl) {
+  return __riscv_vsoxseg3ei16_v_bf16mf2x3(rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxseg3ei16_v_bf16m1x3(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[VS2:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxseg3.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 4 x i16> [[VS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxseg3ei16_v_bf16m1x3(__bf16 *rs1, vuint16m1_t vs2,
+                                  vbfloat16m1x3_t vs3, size_t vl) {
+  return __riscv_vsoxseg3ei16_v_bf16m1x3(rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxseg3ei16_v_bf16m2x3(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[VS2:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxseg3.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 8 x i16> [[VS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxseg3ei16_v_bf16m2x3(__bf16 *rs1, vuint16m2_t vs2,
+                                  vbfloat16m2x3_t vs3, size_t vl) {
+  return __riscv_vsoxseg3ei16_v_bf16m2x3(rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxseg3ei16_v_bf16mf4x3_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[VS2:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxseg3.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 1 x i16> [[VS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxseg3ei16_v_bf16mf4x3_m(vbool64_t vm, __bf16 *rs1,
+                                     vuint16mf4_t vs2, vbfloat16mf4x3_t vs3,
+                                     size_t vl) {
+  return __riscv_vsoxseg3ei16_v_bf16mf4x3_m(vm, rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxseg3ei16_v_bf16mf2x3_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[VS2:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxseg3.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 2 x i16> [[VS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxseg3ei16_v_bf16mf2x3_m(vbool32_t vm, __bf16 *rs1,
+                                     vuint16mf2_t vs2, vbfloat16mf2x3_t vs3,
+                                     size_t vl) {
+  return __riscv_vsoxseg3ei16_v_bf16mf2x3_m(vm, rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxseg3ei16_v_bf16m1x3_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[VS2:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxseg3.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 4 x i16> [[VS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxseg3ei16_v_bf16m1x3_m(vbool16_t vm, __bf16 *rs1, vuint16m1_t vs2,
+                                    vbfloat16m1x3_t vs3, size_t vl) {
+  return __riscv_vsoxseg3ei16_v_bf16m1x3_m(vm, rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxseg3ei16_v_bf16m2x3_m(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[VS2:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxseg3.mask.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 8 x i16> [[VS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxseg3ei16_v_bf16m2x3_m(vbool8_t vm, __bf16 *rs1, vuint16m2_t vs2,
+                                    vbfloat16m2x3_t vs3, size_t vl) {
+  return __riscv_vsoxseg3ei16_v_bf16m2x3_m(vm, rs1, vs2, vs3, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vsoxseg4ei16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vsoxseg4ei16.c
new file mode 100644
index 000000000000..0b29625998b0
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vsoxseg4ei16.c
@@ -0,0 +1,131 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxseg4ei16_v_bf16mf4x4(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[VS2:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxseg4.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 1 x i16> [[VS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxseg4ei16_v_bf16mf4x4(__bf16 *rs1, vuint16mf4_t vs2,
+                                   vbfloat16mf4x4_t vs3, size_t vl) {
+  return __riscv_vsoxseg4ei16_v_bf16mf4x4(rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxseg4ei16_v_bf16mf2x4(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[VS2:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxseg4.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 2 x i16> [[VS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxseg4ei16_v_bf16mf2x4(__bf16 *rs1, vuint16mf2_t vs2,
+                                   vbfloat16mf2x4_t vs3, size_t vl) {
+  return __riscv_vsoxseg4ei16_v_bf16mf2x4(rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxseg4ei16_v_bf16m1x4(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[VS2:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxseg4.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 4 x i16> [[VS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxseg4ei16_v_bf16m1x4(__bf16 *rs1, vuint16m1_t vs2,
+                                  vbfloat16m1x4_t vs3, size_t vl) {
+  return __riscv_vsoxseg4ei16_v_bf16m1x4(rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxseg4ei16_v_bf16m2x4(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[VS2:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxseg4.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], <vscale x 8 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 8 x i16> [[VS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxseg4ei16_v_bf16m2x4(__bf16 *rs1, vuint16m2_t vs2,
+                                  vbfloat16m2x4_t vs3, size_t vl) {
+  return __riscv_vsoxseg4ei16_v_bf16m2x4(rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxseg4ei16_v_bf16mf4x4_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[VS2:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxseg4.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 1 x i16> [[VS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxseg4ei16_v_bf16mf4x4_m(vbool64_t vm, __bf16 *rs1,
+                                     vuint16mf4_t vs2, vbfloat16mf4x4_t vs3,
+                                     size_t vl) {
+  return __riscv_vsoxseg4ei16_v_bf16mf4x4_m(vm, rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxseg4ei16_v_bf16mf2x4_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[VS2:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxseg4.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 2 x i16> [[VS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxseg4ei16_v_bf16mf2x4_m(vbool32_t vm, __bf16 *rs1,
+                                     vuint16mf2_t vs2, vbfloat16mf2x4_t vs3,
+                                     size_t vl) {
+  return __riscv_vsoxseg4ei16_v_bf16mf2x4_m(vm, rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxseg4ei16_v_bf16m1x4_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[VS2:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxseg4.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 4 x i16> [[VS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxseg4ei16_v_bf16m1x4_m(vbool16_t vm, __bf16 *rs1, vuint16m1_t vs2,
+                                    vbfloat16m1x4_t vs3, size_t vl) {
+  return __riscv_vsoxseg4ei16_v_bf16m1x4_m(vm, rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxseg4ei16_v_bf16m2x4_m(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[VS2:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxseg4.mask.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], <vscale x 8 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 8 x i16> [[VS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxseg4ei16_v_bf16m2x4_m(vbool8_t vm, __bf16 *rs1, vuint16m2_t vs2,
+                                    vbfloat16m2x4_t vs3, size_t vl) {
+  return __riscv_vsoxseg4ei16_v_bf16m2x4_m(vm, rs1, vs2, vs3, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vsoxseg5ei16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vsoxseg5ei16.c
new file mode 100644
index 000000000000..0fc4d57918a4
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vsoxseg5ei16.c
@@ -0,0 +1,107 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxseg5ei16_v_bf16mf4x5(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[VS2:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxseg5.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 1 x i16> [[VS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxseg5ei16_v_bf16mf4x5(__bf16 *rs1, vuint16mf4_t vs2,
+                                   vbfloat16mf4x5_t vs3, size_t vl) {
+  return __riscv_vsoxseg5ei16_v_bf16mf4x5(rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxseg5ei16_v_bf16mf2x5(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[VS2:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxseg5.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 2 x i16> [[VS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxseg5ei16_v_bf16mf2x5(__bf16 *rs1, vuint16mf2_t vs2,
+                                   vbfloat16mf2x5_t vs3, size_t vl) {
+  return __riscv_vsoxseg5ei16_v_bf16mf2x5(rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxseg5ei16_v_bf16m1x5(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[VS2:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxseg5.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 4 x i16> [[VS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxseg5ei16_v_bf16m1x5(__bf16 *rs1, vuint16m1_t vs2,
+                                  vbfloat16m1x5_t vs3, size_t vl) {
+  return __riscv_vsoxseg5ei16_v_bf16m1x5(rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxseg5ei16_v_bf16mf4x5_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[VS2:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxseg5.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 1 x i16> [[VS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxseg5ei16_v_bf16mf4x5_m(vbool64_t vm, __bf16 *rs1,
+                                     vuint16mf4_t vs2, vbfloat16mf4x5_t vs3,
+                                     size_t vl) {
+  return __riscv_vsoxseg5ei16_v_bf16mf4x5_m(vm, rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxseg5ei16_v_bf16mf2x5_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[VS2:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxseg5.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 2 x i16> [[VS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxseg5ei16_v_bf16mf2x5_m(vbool32_t vm, __bf16 *rs1,
+                                     vuint16mf2_t vs2, vbfloat16mf2x5_t vs3,
+                                     size_t vl) {
+  return __riscv_vsoxseg5ei16_v_bf16mf2x5_m(vm, rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxseg5ei16_v_bf16m1x5_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[VS2:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxseg5.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 4 x i16> [[VS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxseg5ei16_v_bf16m1x5_m(vbool16_t vm, __bf16 *rs1, vuint16m1_t vs2,
+                                    vbfloat16m1x5_t vs3, size_t vl) {
+  return __riscv_vsoxseg5ei16_v_bf16m1x5_m(vm, rs1, vs2, vs3, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vsoxseg6ei16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vsoxseg6ei16.c
new file mode 100644
index 000000000000..e0c628c7eeec
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vsoxseg6ei16.c
@@ -0,0 +1,113 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxseg6ei16_v_bf16mf4x6(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[VS2:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxseg6.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 1 x i16> [[VS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxseg6ei16_v_bf16mf4x6(__bf16 *rs1, vuint16mf4_t vs2,
+                                   vbfloat16mf4x6_t vs3, size_t vl) {
+  return __riscv_vsoxseg6ei16_v_bf16mf4x6(rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxseg6ei16_v_bf16mf2x6(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[VS2:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxseg6.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 2 x i16> [[VS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxseg6ei16_v_bf16mf2x6(__bf16 *rs1, vuint16mf2_t vs2,
+                                   vbfloat16mf2x6_t vs3, size_t vl) {
+  return __riscv_vsoxseg6ei16_v_bf16mf2x6(rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxseg6ei16_v_bf16m1x6(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[VS2:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxseg6.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 4 x i16> [[VS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxseg6ei16_v_bf16m1x6(__bf16 *rs1, vuint16m1_t vs2,
+                                  vbfloat16m1x6_t vs3, size_t vl) {
+  return __riscv_vsoxseg6ei16_v_bf16m1x6(rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxseg6ei16_v_bf16mf4x6_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[VS2:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxseg6.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 1 x i16> [[VS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxseg6ei16_v_bf16mf4x6_m(vbool64_t vm, __bf16 *rs1,
+                                     vuint16mf4_t vs2, vbfloat16mf4x6_t vs3,
+                                     size_t vl) {
+  return __riscv_vsoxseg6ei16_v_bf16mf4x6_m(vm, rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxseg6ei16_v_bf16mf2x6_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[VS2:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxseg6.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 2 x i16> [[VS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxseg6ei16_v_bf16mf2x6_m(vbool32_t vm, __bf16 *rs1,
+                                     vuint16mf2_t vs2, vbfloat16mf2x6_t vs3,
+                                     size_t vl) {
+  return __riscv_vsoxseg6ei16_v_bf16mf2x6_m(vm, rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxseg6ei16_v_bf16m1x6_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[VS2:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxseg6.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 4 x i16> [[VS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxseg6ei16_v_bf16m1x6_m(vbool16_t vm, __bf16 *rs1, vuint16m1_t vs2,
+                                    vbfloat16m1x6_t vs3, size_t vl) {
+  return __riscv_vsoxseg6ei16_v_bf16m1x6_m(vm, rs1, vs2, vs3, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vsoxseg7ei16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vsoxseg7ei16.c
new file mode 100644
index 000000000000..4ae28ba5cedd
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vsoxseg7ei16.c
@@ -0,0 +1,119 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxseg7ei16_v_bf16mf4x7(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[VS2:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 6
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxseg7.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 1 x i16> [[VS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxseg7ei16_v_bf16mf4x7(__bf16 *rs1, vuint16mf4_t vs2,
+                                   vbfloat16mf4x7_t vs3, size_t vl) {
+  return __riscv_vsoxseg7ei16_v_bf16mf4x7(rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxseg7ei16_v_bf16mf2x7(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[VS2:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 6
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxseg7.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 2 x i16> [[VS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxseg7ei16_v_bf16mf2x7(__bf16 *rs1, vuint16mf2_t vs2,
+                                   vbfloat16mf2x7_t vs3, size_t vl) {
+  return __riscv_vsoxseg7ei16_v_bf16mf2x7(rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxseg7ei16_v_bf16m1x7(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[VS2:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 6
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxseg7.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 4 x i16> [[VS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxseg7ei16_v_bf16m1x7(__bf16 *rs1, vuint16m1_t vs2,
+                                  vbfloat16m1x7_t vs3, size_t vl) {
+  return __riscv_vsoxseg7ei16_v_bf16m1x7(rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxseg7ei16_v_bf16mf4x7_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[VS2:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 6
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxseg7.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 1 x i16> [[VS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxseg7ei16_v_bf16mf4x7_m(vbool64_t vm, __bf16 *rs1,
+                                     vuint16mf4_t vs2, vbfloat16mf4x7_t vs3,
+                                     size_t vl) {
+  return __riscv_vsoxseg7ei16_v_bf16mf4x7_m(vm, rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxseg7ei16_v_bf16mf2x7_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[VS2:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 6
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxseg7.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 2 x i16> [[VS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxseg7ei16_v_bf16mf2x7_m(vbool32_t vm, __bf16 *rs1,
+                                     vuint16mf2_t vs2, vbfloat16mf2x7_t vs3,
+                                     size_t vl) {
+  return __riscv_vsoxseg7ei16_v_bf16mf2x7_m(vm, rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxseg7ei16_v_bf16m1x7_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[VS2:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 6
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxseg7.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 4 x i16> [[VS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxseg7ei16_v_bf16m1x7_m(vbool16_t vm, __bf16 *rs1, vuint16m1_t vs2,
+                                    vbfloat16m1x7_t vs3, size_t vl) {
+  return __riscv_vsoxseg7ei16_v_bf16m1x7_m(vm, rs1, vs2, vs3, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vsoxseg8ei16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vsoxseg8ei16.c
new file mode 100644
index 000000000000..942a3239934a
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vsoxseg8ei16.c
@@ -0,0 +1,125 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxseg8ei16_v_bf16mf4x8(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[VS2:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 7
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxseg8.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], <vscale x 1 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 1 x i16> [[VS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxseg8ei16_v_bf16mf4x8(__bf16 *rs1, vuint16mf4_t vs2,
+                                   vbfloat16mf4x8_t vs3, size_t vl) {
+  return __riscv_vsoxseg8ei16_v_bf16mf4x8(rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxseg8ei16_v_bf16mf2x8(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[VS2:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 7
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxseg8.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], <vscale x 2 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 2 x i16> [[VS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxseg8ei16_v_bf16mf2x8(__bf16 *rs1, vuint16mf2_t vs2,
+                                   vbfloat16mf2x8_t vs3, size_t vl) {
+  return __riscv_vsoxseg8ei16_v_bf16mf2x8(rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxseg8ei16_v_bf16m1x8(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[VS2:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 7
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxseg8.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], <vscale x 4 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 4 x i16> [[VS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxseg8ei16_v_bf16m1x8(__bf16 *rs1, vuint16m1_t vs2,
+                                  vbfloat16m1x8_t vs3, size_t vl) {
+  return __riscv_vsoxseg8ei16_v_bf16m1x8(rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxseg8ei16_v_bf16mf4x8_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[VS2:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 7
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxseg8.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], <vscale x 1 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 1 x i16> [[VS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxseg8ei16_v_bf16mf4x8_m(vbool64_t vm, __bf16 *rs1,
+                                     vuint16mf4_t vs2, vbfloat16mf4x8_t vs3,
+                                     size_t vl) {
+  return __riscv_vsoxseg8ei16_v_bf16mf4x8_m(vm, rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxseg8ei16_v_bf16mf2x8_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[VS2:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 7
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxseg8.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], <vscale x 2 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 2 x i16> [[VS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxseg8ei16_v_bf16mf2x8_m(vbool32_t vm, __bf16 *rs1,
+                                     vuint16mf2_t vs2, vbfloat16mf2x8_t vs3,
+                                     size_t vl) {
+  return __riscv_vsoxseg8ei16_v_bf16mf2x8_m(vm, rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxseg8ei16_v_bf16m1x8_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[VS2:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 7
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxseg8.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], <vscale x 4 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 4 x i16> [[VS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxseg8ei16_v_bf16m1x8_m(vbool16_t vm, __bf16 *rs1, vuint16m1_t vs2,
+                                    vbfloat16m1x8_t vs3, size_t vl) {
+  return __riscv_vsoxseg8ei16_v_bf16m1x8_m(vm, rs1, vs2, vs3, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vsse16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vsse16.c
new file mode 100644
index 000000000000..13eb05bb3b0c
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vsse16.c
@@ -0,0 +1,141 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsse16_v_bf16mf4(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], <vscale x 1 x bfloat> [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsse.nxv1bf16.i64(<vscale x 1 x bfloat> [[VS3]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsse16_v_bf16mf4(__bf16 *rs1, ptrdiff_t rs2, vbfloat16mf4_t vs3,
+                           size_t vl) {
+  return __riscv_vsse16_v_bf16mf4(rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsse16_v_bf16mf2(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], <vscale x 2 x bfloat> [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsse.nxv2bf16.i64(<vscale x 2 x bfloat> [[VS3]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsse16_v_bf16mf2(__bf16 *rs1, ptrdiff_t rs2, vbfloat16mf2_t vs3,
+                           size_t vl) {
+  return __riscv_vsse16_v_bf16mf2(rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsse16_v_bf16m1(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], <vscale x 4 x bfloat> [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsse.nxv4bf16.i64(<vscale x 4 x bfloat> [[VS3]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsse16_v_bf16m1(__bf16 *rs1, ptrdiff_t rs2, vbfloat16m1_t vs3,
+                          size_t vl) {
+  return __riscv_vsse16_v_bf16m1(rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsse16_v_bf16m2(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], <vscale x 8 x bfloat> [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsse.nxv8bf16.i64(<vscale x 8 x bfloat> [[VS3]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsse16_v_bf16m2(__bf16 *rs1, ptrdiff_t rs2, vbfloat16m2_t vs3,
+                          size_t vl) {
+  return __riscv_vsse16_v_bf16m2(rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsse16_v_bf16m4(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], <vscale x 16 x bfloat> [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsse.nxv16bf16.i64(<vscale x 16 x bfloat> [[VS3]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsse16_v_bf16m4(__bf16 *rs1, ptrdiff_t rs2, vbfloat16m4_t vs3,
+                          size_t vl) {
+  return __riscv_vsse16_v_bf16m4(rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsse16_v_bf16m8(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], <vscale x 32 x bfloat> [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsse.nxv32bf16.i64(<vscale x 32 x bfloat> [[VS3]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsse16_v_bf16m8(__bf16 *rs1, ptrdiff_t rs2, vbfloat16m8_t vs3,
+                          size_t vl) {
+  return __riscv_vsse16_v_bf16m8(rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsse16_v_bf16mf4_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], <vscale x 1 x bfloat> [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsse.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[VS3]], ptr [[RS1]], i64 [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsse16_v_bf16mf4_m(vbool64_t vm, __bf16 *rs1, ptrdiff_t rs2,
+                             vbfloat16mf4_t vs3, size_t vl) {
+  return __riscv_vsse16_v_bf16mf4_m(vm, rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsse16_v_bf16mf2_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], <vscale x 2 x bfloat> [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsse.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[VS3]], ptr [[RS1]], i64 [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsse16_v_bf16mf2_m(vbool32_t vm, __bf16 *rs1, ptrdiff_t rs2,
+                             vbfloat16mf2_t vs3, size_t vl) {
+  return __riscv_vsse16_v_bf16mf2_m(vm, rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsse16_v_bf16m1_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], <vscale x 4 x bfloat> [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsse.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[VS3]], ptr [[RS1]], i64 [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsse16_v_bf16m1_m(vbool16_t vm, __bf16 *rs1, ptrdiff_t rs2,
+                            vbfloat16m1_t vs3, size_t vl) {
+  return __riscv_vsse16_v_bf16m1_m(vm, rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsse16_v_bf16m2_m(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], <vscale x 8 x bfloat> [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsse.mask.nxv8bf16.i64(<vscale x 8 x bfloat> [[VS3]], ptr [[RS1]], i64 [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsse16_v_bf16m2_m(vbool8_t vm, __bf16 *rs1, ptrdiff_t rs2,
+                            vbfloat16m2_t vs3, size_t vl) {
+  return __riscv_vsse16_v_bf16m2_m(vm, rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsse16_v_bf16m4_m(
+// CHECK-RV64-SAME: <vscale x 16 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], <vscale x 16 x bfloat> [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsse.mask.nxv16bf16.i64(<vscale x 16 x bfloat> [[VS3]], ptr [[RS1]], i64 [[RS2]], <vscale x 16 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsse16_v_bf16m4_m(vbool4_t vm, __bf16 *rs1, ptrdiff_t rs2,
+                            vbfloat16m4_t vs3, size_t vl) {
+  return __riscv_vsse16_v_bf16m4_m(vm, rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsse16_v_bf16m8_m(
+// CHECK-RV64-SAME: <vscale x 32 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], <vscale x 32 x bfloat> [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsse.mask.nxv32bf16.i64(<vscale x 32 x bfloat> [[VS3]], ptr [[RS1]], i64 [[RS2]], <vscale x 32 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsse16_v_bf16m8_m(vbool2_t vm, __bf16 *rs1, ptrdiff_t rs2,
+                            vbfloat16m8_t vs3, size_t vl) {
+  return __riscv_vsse16_v_bf16m8_m(vm, rs1, rs2, vs3, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vsseg2e16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vsseg2e16.c
new file mode 100644
index 000000000000..a6384b154239
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vsseg2e16.c
@@ -0,0 +1,134 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsseg2e16_v_bf16mf4x2(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsseg2.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsseg2e16_v_bf16mf4x2(__bf16 *rs1, vbfloat16mf4x2_t vs3, size_t vl) {
+  return __riscv_vsseg2e16_v_bf16mf4x2(rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsseg2e16_v_bf16mf2x2(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsseg2.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsseg2e16_v_bf16mf2x2(__bf16 *rs1, vbfloat16mf2x2_t vs3, size_t vl) {
+  return __riscv_vsseg2e16_v_bf16mf2x2(rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsseg2e16_v_bf16m1x2(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsseg2.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsseg2e16_v_bf16m1x2(__bf16 *rs1, vbfloat16m1x2_t vs3, size_t vl) {
+  return __riscv_vsseg2e16_v_bf16m1x2(rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsseg2e16_v_bf16m2x2(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsseg2.nxv8bf16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsseg2e16_v_bf16m2x2(__bf16 *rs1, vbfloat16m2x2_t vs3, size_t vl) {
+  return __riscv_vsseg2e16_v_bf16m2x2(rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsseg2e16_v_bf16m4x2(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsseg2.nxv16bf16.i64(<vscale x 16 x bfloat> [[TMP0]], <vscale x 16 x bfloat> [[TMP1]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsseg2e16_v_bf16m4x2(__bf16 *rs1, vbfloat16m4x2_t vs3, size_t vl) {
+  return __riscv_vsseg2e16_v_bf16m4x2(rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsseg2e16_v_bf16mf4x2_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsseg2.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsseg2e16_v_bf16mf4x2_m(vbool64_t vm, __bf16 *rs1,
+                                  vbfloat16mf4x2_t vs3, size_t vl) {
+  return __riscv_vsseg2e16_v_bf16mf4x2_m(vm, rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsseg2e16_v_bf16mf2x2_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsseg2.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsseg2e16_v_bf16mf2x2_m(vbool32_t vm, __bf16 *rs1,
+                                  vbfloat16mf2x2_t vs3, size_t vl) {
+  return __riscv_vsseg2e16_v_bf16mf2x2_m(vm, rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsseg2e16_v_bf16m1x2_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsseg2.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsseg2e16_v_bf16m1x2_m(vbool16_t vm, __bf16 *rs1, vbfloat16m1x2_t vs3,
+                                 size_t vl) {
+  return __riscv_vsseg2e16_v_bf16m1x2_m(vm, rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsseg2e16_v_bf16m2x2_m(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsseg2.mask.nxv8bf16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 8 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsseg2e16_v_bf16m2x2_m(vbool8_t vm, __bf16 *rs1, vbfloat16m2x2_t vs3,
+                                 size_t vl) {
+  return __riscv_vsseg2e16_v_bf16m2x2_m(vm, rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsseg2e16_v_bf16m4x2_m(
+// CHECK-RV64-SAME: <vscale x 16 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsseg2.mask.nxv16bf16.i64(<vscale x 16 x bfloat> [[TMP0]], <vscale x 16 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 16 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsseg2e16_v_bf16m4x2_m(vbool4_t vm, __bf16 *rs1, vbfloat16m4x2_t vs3,
+                                 size_t vl) {
+  return __riscv_vsseg2e16_v_bf16m4x2_m(vm, rs1, vs3, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vsseg3e16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vsseg3e16.c
new file mode 100644
index 000000000000..b04177c8db4d
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vsseg3e16.c
@@ -0,0 +1,117 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsseg3e16_v_bf16mf4x3(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsseg3.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsseg3e16_v_bf16mf4x3(__bf16 *rs1, vbfloat16mf4x3_t vs3, size_t vl) {
+  return __riscv_vsseg3e16_v_bf16mf4x3(rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsseg3e16_v_bf16mf2x3(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsseg3.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsseg3e16_v_bf16mf2x3(__bf16 *rs1, vbfloat16mf2x3_t vs3, size_t vl) {
+  return __riscv_vsseg3e16_v_bf16mf2x3(rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsseg3e16_v_bf16m1x3(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsseg3.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsseg3e16_v_bf16m1x3(__bf16 *rs1, vbfloat16m1x3_t vs3, size_t vl) {
+  return __riscv_vsseg3e16_v_bf16m1x3(rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsseg3e16_v_bf16m2x3(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsseg3.nxv8bf16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsseg3e16_v_bf16m2x3(__bf16 *rs1, vbfloat16m2x3_t vs3, size_t vl) {
+  return __riscv_vsseg3e16_v_bf16m2x3(rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsseg3e16_v_bf16mf4x3_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsseg3.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsseg3e16_v_bf16mf4x3_m(vbool64_t vm, __bf16 *rs1,
+                                  vbfloat16mf4x3_t vs3, size_t vl) {
+  return __riscv_vsseg3e16_v_bf16mf4x3_m(vm, rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsseg3e16_v_bf16mf2x3_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsseg3.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsseg3e16_v_bf16mf2x3_m(vbool32_t vm, __bf16 *rs1,
+                                  vbfloat16mf2x3_t vs3, size_t vl) {
+  return __riscv_vsseg3e16_v_bf16mf2x3_m(vm, rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsseg3e16_v_bf16m1x3_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsseg3.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsseg3e16_v_bf16m1x3_m(vbool16_t vm, __bf16 *rs1, vbfloat16m1x3_t vs3,
+                                 size_t vl) {
+  return __riscv_vsseg3e16_v_bf16m1x3_m(vm, rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsseg3e16_v_bf16m2x3_m(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsseg3.mask.nxv8bf16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 8 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsseg3e16_v_bf16m2x3_m(vbool8_t vm, __bf16 *rs1, vbfloat16m2x3_t vs3,
+                                 size_t vl) {
+  return __riscv_vsseg3e16_v_bf16m2x3_m(vm, rs1, vs3, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vsseg4e16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vsseg4e16.c
new file mode 100644
index 000000000000..3745c3af566e
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vsseg4e16.c
@@ -0,0 +1,125 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsseg4e16_v_bf16mf4x4(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsseg4.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsseg4e16_v_bf16mf4x4(__bf16 *rs1, vbfloat16mf4x4_t vs3, size_t vl) {
+  return __riscv_vsseg4e16_v_bf16mf4x4(rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsseg4e16_v_bf16mf2x4(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsseg4.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsseg4e16_v_bf16mf2x4(__bf16 *rs1, vbfloat16mf2x4_t vs3, size_t vl) {
+  return __riscv_vsseg4e16_v_bf16mf2x4(rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsseg4e16_v_bf16m1x4(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsseg4.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsseg4e16_v_bf16m1x4(__bf16 *rs1, vbfloat16m1x4_t vs3, size_t vl) {
+  return __riscv_vsseg4e16_v_bf16m1x4(rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsseg4e16_v_bf16m2x4(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsseg4.nxv8bf16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], <vscale x 8 x bfloat> [[TMP3]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsseg4e16_v_bf16m2x4(__bf16 *rs1, vbfloat16m2x4_t vs3, size_t vl) {
+  return __riscv_vsseg4e16_v_bf16m2x4(rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsseg4e16_v_bf16mf4x4_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsseg4.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsseg4e16_v_bf16mf4x4_m(vbool64_t vm, __bf16 *rs1,
+                                  vbfloat16mf4x4_t vs3, size_t vl) {
+  return __riscv_vsseg4e16_v_bf16mf4x4_m(vm, rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsseg4e16_v_bf16mf2x4_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsseg4.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsseg4e16_v_bf16mf2x4_m(vbool32_t vm, __bf16 *rs1,
+                                  vbfloat16mf2x4_t vs3, size_t vl) {
+  return __riscv_vsseg4e16_v_bf16mf2x4_m(vm, rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsseg4e16_v_bf16m1x4_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsseg4.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsseg4e16_v_bf16m1x4_m(vbool16_t vm, __bf16 *rs1, vbfloat16m1x4_t vs3,
+                                 size_t vl) {
+  return __riscv_vsseg4e16_v_bf16m1x4_m(vm, rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsseg4e16_v_bf16m2x4_m(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsseg4.mask.nxv8bf16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], <vscale x 8 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 8 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsseg4e16_v_bf16m2x4_m(vbool8_t vm, __bf16 *rs1, vbfloat16m2x4_t vs3,
+                                 size_t vl) {
+  return __riscv_vsseg4e16_v_bf16m2x4_m(vm, rs1, vs3, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vsseg5e16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vsseg5e16.c
new file mode 100644
index 000000000000..8aa59fbb521c
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vsseg5e16.c
@@ -0,0 +1,102 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsseg5e16_v_bf16mf4x5(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsseg5.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsseg5e16_v_bf16mf4x5(__bf16 *rs1, vbfloat16mf4x5_t vs3, size_t vl) {
+  return __riscv_vsseg5e16_v_bf16mf4x5(rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsseg5e16_v_bf16mf2x5(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsseg5.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsseg5e16_v_bf16mf2x5(__bf16 *rs1, vbfloat16mf2x5_t vs3, size_t vl) {
+  return __riscv_vsseg5e16_v_bf16mf2x5(rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsseg5e16_v_bf16m1x5(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsseg5.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsseg5e16_v_bf16m1x5(__bf16 *rs1, vbfloat16m1x5_t vs3, size_t vl) {
+  return __riscv_vsseg5e16_v_bf16m1x5(rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsseg5e16_v_bf16mf4x5_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsseg5.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsseg5e16_v_bf16mf4x5_m(vbool64_t vm, __bf16 *rs1,
+                                  vbfloat16mf4x5_t vs3, size_t vl) {
+  return __riscv_vsseg5e16_v_bf16mf4x5_m(vm, rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsseg5e16_v_bf16mf2x5_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsseg5.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsseg5e16_v_bf16mf2x5_m(vbool32_t vm, __bf16 *rs1,
+                                  vbfloat16mf2x5_t vs3, size_t vl) {
+  return __riscv_vsseg5e16_v_bf16mf2x5_m(vm, rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsseg5e16_v_bf16m1x5_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsseg5.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsseg5e16_v_bf16m1x5_m(vbool16_t vm, __bf16 *rs1, vbfloat16m1x5_t vs3,
+                                 size_t vl) {
+  return __riscv_vsseg5e16_v_bf16m1x5_m(vm, rs1, vs3, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vsseg6e16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vsseg6e16.c
new file mode 100644
index 000000000000..a2b3fd5e1102
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vsseg6e16.c
@@ -0,0 +1,108 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsseg6e16_v_bf16mf4x6(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsseg6.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsseg6e16_v_bf16mf4x6(__bf16 *rs1, vbfloat16mf4x6_t vs3, size_t vl) {
+  return __riscv_vsseg6e16_v_bf16mf4x6(rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsseg6e16_v_bf16mf2x6(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsseg6.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsseg6e16_v_bf16mf2x6(__bf16 *rs1, vbfloat16mf2x6_t vs3, size_t vl) {
+  return __riscv_vsseg6e16_v_bf16mf2x6(rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsseg6e16_v_bf16m1x6(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsseg6.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsseg6e16_v_bf16m1x6(__bf16 *rs1, vbfloat16m1x6_t vs3, size_t vl) {
+  return __riscv_vsseg6e16_v_bf16m1x6(rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsseg6e16_v_bf16mf4x6_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsseg6.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsseg6e16_v_bf16mf4x6_m(vbool64_t vm, __bf16 *rs1,
+                                  vbfloat16mf4x6_t vs3, size_t vl) {
+  return __riscv_vsseg6e16_v_bf16mf4x6_m(vm, rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsseg6e16_v_bf16mf2x6_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsseg6.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsseg6e16_v_bf16mf2x6_m(vbool32_t vm, __bf16 *rs1,
+                                  vbfloat16mf2x6_t vs3, size_t vl) {
+  return __riscv_vsseg6e16_v_bf16mf2x6_m(vm, rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsseg6e16_v_bf16m1x6_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsseg6.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsseg6e16_v_bf16m1x6_m(vbool16_t vm, __bf16 *rs1, vbfloat16m1x6_t vs3,
+                                 size_t vl) {
+  return __riscv_vsseg6e16_v_bf16m1x6_m(vm, rs1, vs3, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vsseg7e16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vsseg7e16.c
new file mode 100644
index 000000000000..366476917765
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vsseg7e16.c
@@ -0,0 +1,114 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsseg7e16_v_bf16mf4x7(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 6
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsseg7.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsseg7e16_v_bf16mf4x7(__bf16 *rs1, vbfloat16mf4x7_t vs3, size_t vl) {
+  return __riscv_vsseg7e16_v_bf16mf4x7(rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsseg7e16_v_bf16mf2x7(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 6
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsseg7.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsseg7e16_v_bf16mf2x7(__bf16 *rs1, vbfloat16mf2x7_t vs3, size_t vl) {
+  return __riscv_vsseg7e16_v_bf16mf2x7(rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsseg7e16_v_bf16m1x7(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 6
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsseg7.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsseg7e16_v_bf16m1x7(__bf16 *rs1, vbfloat16m1x7_t vs3, size_t vl) {
+  return __riscv_vsseg7e16_v_bf16m1x7(rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsseg7e16_v_bf16mf4x7_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 6
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsseg7.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsseg7e16_v_bf16mf4x7_m(vbool64_t vm, __bf16 *rs1,
+                                  vbfloat16mf4x7_t vs3, size_t vl) {
+  return __riscv_vsseg7e16_v_bf16mf4x7_m(vm, rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsseg7e16_v_bf16mf2x7_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 6
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsseg7.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsseg7e16_v_bf16mf2x7_m(vbool32_t vm, __bf16 *rs1,
+                                  vbfloat16mf2x7_t vs3, size_t vl) {
+  return __riscv_vsseg7e16_v_bf16mf2x7_m(vm, rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsseg7e16_v_bf16m1x7_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 6
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsseg7.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsseg7e16_v_bf16m1x7_m(vbool16_t vm, __bf16 *rs1, vbfloat16m1x7_t vs3,
+                                 size_t vl) {
+  return __riscv_vsseg7e16_v_bf16m1x7_m(vm, rs1, vs3, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vsseg8e16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vsseg8e16.c
new file mode 100644
index 000000000000..2c6717726e72
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vsseg8e16.c
@@ -0,0 +1,120 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsseg8e16_v_bf16mf4x8(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 7
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsseg8.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], <vscale x 1 x bfloat> [[TMP7]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsseg8e16_v_bf16mf4x8(__bf16 *rs1, vbfloat16mf4x8_t vs3, size_t vl) {
+  return __riscv_vsseg8e16_v_bf16mf4x8(rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsseg8e16_v_bf16mf2x8(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 7
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsseg8.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], <vscale x 2 x bfloat> [[TMP7]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsseg8e16_v_bf16mf2x8(__bf16 *rs1, vbfloat16mf2x8_t vs3, size_t vl) {
+  return __riscv_vsseg8e16_v_bf16mf2x8(rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsseg8e16_v_bf16m1x8(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 7
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsseg8.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], <vscale x 4 x bfloat> [[TMP7]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsseg8e16_v_bf16m1x8(__bf16 *rs1, vbfloat16m1x8_t vs3, size_t vl) {
+  return __riscv_vsseg8e16_v_bf16m1x8(rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsseg8e16_v_bf16mf4x8_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 7
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsseg8.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], <vscale x 1 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsseg8e16_v_bf16mf4x8_m(vbool64_t vm, __bf16 *rs1,
+                                  vbfloat16mf4x8_t vs3, size_t vl) {
+  return __riscv_vsseg8e16_v_bf16mf4x8_m(vm, rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsseg8e16_v_bf16mf2x8_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 7
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsseg8.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], <vscale x 2 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsseg8e16_v_bf16mf2x8_m(vbool32_t vm, __bf16 *rs1,
+                                  vbfloat16mf2x8_t vs3, size_t vl) {
+  return __riscv_vsseg8e16_v_bf16mf2x8_m(vm, rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsseg8e16_v_bf16m1x8_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 7
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsseg8.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], <vscale x 4 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsseg8e16_v_bf16m1x8_m(vbool16_t vm, __bf16 *rs1, vbfloat16m1x8_t vs3,
+                                 size_t vl) {
+  return __riscv_vsseg8e16_v_bf16m1x8_m(vm, rs1, vs3, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vssseg2e16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vssseg2e16.c
new file mode 100644
index 000000000000..a6aacb536854
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vssseg2e16.c
@@ -0,0 +1,139 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local void @test_vssseg2e16_v_bf16mf4x2(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vssseg2.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vssseg2e16_v_bf16mf4x2(__bf16 *rs1, ptrdiff_t rs2,
+                                 vbfloat16mf4x2_t vs3, size_t vl) {
+  return __riscv_vssseg2e16_v_bf16mf4x2(rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vssseg2e16_v_bf16mf2x2(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vssseg2.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vssseg2e16_v_bf16mf2x2(__bf16 *rs1, ptrdiff_t rs2,
+                                 vbfloat16mf2x2_t vs3, size_t vl) {
+  return __riscv_vssseg2e16_v_bf16mf2x2(rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vssseg2e16_v_bf16m1x2(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vssseg2.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vssseg2e16_v_bf16m1x2(__bf16 *rs1, ptrdiff_t rs2, vbfloat16m1x2_t vs3,
+                                size_t vl) {
+  return __riscv_vssseg2e16_v_bf16m1x2(rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vssseg2e16_v_bf16m2x2(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vssseg2.nxv8bf16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vssseg2e16_v_bf16m2x2(__bf16 *rs1, ptrdiff_t rs2, vbfloat16m2x2_t vs3,
+                                size_t vl) {
+  return __riscv_vssseg2e16_v_bf16m2x2(rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vssseg2e16_v_bf16m4x2(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vssseg2.nxv16bf16.i64(<vscale x 16 x bfloat> [[TMP0]], <vscale x 16 x bfloat> [[TMP1]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vssseg2e16_v_bf16m4x2(__bf16 *rs1, ptrdiff_t rs2, vbfloat16m4x2_t vs3,
+                                size_t vl) {
+  return __riscv_vssseg2e16_v_bf16m4x2(rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vssseg2e16_v_bf16mf4x2_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vssseg2.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], ptr [[RS1]], i64 [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vssseg2e16_v_bf16mf4x2_m(vbool64_t vm, __bf16 *rs1, ptrdiff_t rs2,
+                                   vbfloat16mf4x2_t vs3, size_t vl) {
+  return __riscv_vssseg2e16_v_bf16mf4x2_m(vm, rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vssseg2e16_v_bf16mf2x2_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vssseg2.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], ptr [[RS1]], i64 [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vssseg2e16_v_bf16mf2x2_m(vbool32_t vm, __bf16 *rs1, ptrdiff_t rs2,
+                                   vbfloat16mf2x2_t vs3, size_t vl) {
+  return __riscv_vssseg2e16_v_bf16mf2x2_m(vm, rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vssseg2e16_v_bf16m1x2_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vssseg2.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], ptr [[RS1]], i64 [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vssseg2e16_v_bf16m1x2_m(vbool16_t vm, __bf16 *rs1, ptrdiff_t rs2,
+                                  vbfloat16m1x2_t vs3, size_t vl) {
+  return __riscv_vssseg2e16_v_bf16m1x2_m(vm, rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vssseg2e16_v_bf16m2x2_m(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vssseg2.mask.nxv8bf16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], ptr [[RS1]], i64 [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vssseg2e16_v_bf16m2x2_m(vbool8_t vm, __bf16 *rs1, ptrdiff_t rs2,
+                                  vbfloat16m2x2_t vs3, size_t vl) {
+  return __riscv_vssseg2e16_v_bf16m2x2_m(vm, rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vssseg2e16_v_bf16m4x2_m(
+// CHECK-RV64-SAME: <vscale x 16 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vssseg2.mask.nxv16bf16.i64(<vscale x 16 x bfloat> [[TMP0]], <vscale x 16 x bfloat> [[TMP1]], ptr [[RS1]], i64 [[RS2]], <vscale x 16 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vssseg2e16_v_bf16m4x2_m(vbool4_t vm, __bf16 *rs1, ptrdiff_t rs2,
+                                  vbfloat16m4x2_t vs3, size_t vl) {
+  return __riscv_vssseg2e16_v_bf16m4x2_m(vm, rs1, rs2, vs3, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vssseg3e16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vssseg3e16.c
new file mode 100644
index 000000000000..4ce7ef9bbf04
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vssseg3e16.c
@@ -0,0 +1,121 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local void @test_vssseg3e16_v_bf16mf4x3(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vssseg3.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vssseg3e16_v_bf16mf4x3(__bf16 *rs1, ptrdiff_t rs2,
+                                 vbfloat16mf4x3_t vs3, size_t vl) {
+  return __riscv_vssseg3e16_v_bf16mf4x3(rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vssseg3e16_v_bf16mf2x3(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vssseg3.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vssseg3e16_v_bf16mf2x3(__bf16 *rs1, ptrdiff_t rs2,
+                                 vbfloat16mf2x3_t vs3, size_t vl) {
+  return __riscv_vssseg3e16_v_bf16mf2x3(rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vssseg3e16_v_bf16m1x3(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vssseg3.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vssseg3e16_v_bf16m1x3(__bf16 *rs1, ptrdiff_t rs2, vbfloat16m1x3_t vs3,
+                                size_t vl) {
+  return __riscv_vssseg3e16_v_bf16m1x3(rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vssseg3e16_v_bf16m2x3(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vssseg3.nxv8bf16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vssseg3e16_v_bf16m2x3(__bf16 *rs1, ptrdiff_t rs2, vbfloat16m2x3_t vs3,
+                                size_t vl) {
+  return __riscv_vssseg3e16_v_bf16m2x3(rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vssseg3e16_v_bf16mf4x3_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vssseg3.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], ptr [[RS1]], i64 [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vssseg3e16_v_bf16mf4x3_m(vbool64_t vm, __bf16 *rs1, ptrdiff_t rs2,
+                                   vbfloat16mf4x3_t vs3, size_t vl) {
+  return __riscv_vssseg3e16_v_bf16mf4x3_m(vm, rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vssseg3e16_v_bf16mf2x3_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vssseg3.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], ptr [[RS1]], i64 [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vssseg3e16_v_bf16mf2x3_m(vbool32_t vm, __bf16 *rs1, ptrdiff_t rs2,
+                                   vbfloat16mf2x3_t vs3, size_t vl) {
+  return __riscv_vssseg3e16_v_bf16mf2x3_m(vm, rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vssseg3e16_v_bf16m1x3_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vssseg3.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], ptr [[RS1]], i64 [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vssseg3e16_v_bf16m1x3_m(vbool16_t vm, __bf16 *rs1, ptrdiff_t rs2,
+                                  vbfloat16m1x3_t vs3, size_t vl) {
+  return __riscv_vssseg3e16_v_bf16m1x3_m(vm, rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vssseg3e16_v_bf16m2x3_m(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vssseg3.mask.nxv8bf16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], ptr [[RS1]], i64 [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vssseg3e16_v_bf16m2x3_m(vbool8_t vm, __bf16 *rs1, ptrdiff_t rs2,
+                                  vbfloat16m2x3_t vs3, size_t vl) {
+  return __riscv_vssseg3e16_v_bf16m2x3_m(vm, rs1, rs2, vs3, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vssseg4e16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vssseg4e16.c
new file mode 100644
index 000000000000..f8d980f8946f
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vssseg4e16.c
@@ -0,0 +1,129 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local void @test_vssseg4e16_v_bf16mf4x4(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vssseg4.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vssseg4e16_v_bf16mf4x4(__bf16 *rs1, ptrdiff_t rs2,
+                                 vbfloat16mf4x4_t vs3, size_t vl) {
+  return __riscv_vssseg4e16_v_bf16mf4x4(rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vssseg4e16_v_bf16mf2x4(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vssseg4.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vssseg4e16_v_bf16mf2x4(__bf16 *rs1, ptrdiff_t rs2,
+                                 vbfloat16mf2x4_t vs3, size_t vl) {
+  return __riscv_vssseg4e16_v_bf16mf2x4(rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vssseg4e16_v_bf16m1x4(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vssseg4.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vssseg4e16_v_bf16m1x4(__bf16 *rs1, ptrdiff_t rs2, vbfloat16m1x4_t vs3,
+                                size_t vl) {
+  return __riscv_vssseg4e16_v_bf16m1x4(rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vssseg4e16_v_bf16m2x4(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vssseg4.nxv8bf16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], <vscale x 8 x bfloat> [[TMP3]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vssseg4e16_v_bf16m2x4(__bf16 *rs1, ptrdiff_t rs2, vbfloat16m2x4_t vs3,
+                                size_t vl) {
+  return __riscv_vssseg4e16_v_bf16m2x4(rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vssseg4e16_v_bf16mf4x4_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vssseg4.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], ptr [[RS1]], i64 [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vssseg4e16_v_bf16mf4x4_m(vbool64_t vm, __bf16 *rs1, ptrdiff_t rs2,
+                                   vbfloat16mf4x4_t vs3, size_t vl) {
+  return __riscv_vssseg4e16_v_bf16mf4x4_m(vm, rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vssseg4e16_v_bf16mf2x4_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vssseg4.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], ptr [[RS1]], i64 [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vssseg4e16_v_bf16mf2x4_m(vbool32_t vm, __bf16 *rs1, ptrdiff_t rs2,
+                                   vbfloat16mf2x4_t vs3, size_t vl) {
+  return __riscv_vssseg4e16_v_bf16mf2x4_m(vm, rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vssseg4e16_v_bf16m1x4_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vssseg4.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], ptr [[RS1]], i64 [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vssseg4e16_v_bf16m1x4_m(vbool16_t vm, __bf16 *rs1, ptrdiff_t rs2,
+                                  vbfloat16m1x4_t vs3, size_t vl) {
+  return __riscv_vssseg4e16_v_bf16m1x4_m(vm, rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vssseg4e16_v_bf16m2x4_m(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vssseg4.mask.nxv8bf16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], <vscale x 8 x bfloat> [[TMP3]], ptr [[RS1]], i64 [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vssseg4e16_v_bf16m2x4_m(vbool8_t vm, __bf16 *rs1, ptrdiff_t rs2,
+                                  vbfloat16m2x4_t vs3, size_t vl) {
+  return __riscv_vssseg4e16_v_bf16m2x4_m(vm, rs1, rs2, vs3, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vssseg5e16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vssseg5e16.c
new file mode 100644
index 000000000000..115cdf480d97
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vssseg5e16.c
@@ -0,0 +1,105 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local void @test_vssseg5e16_v_bf16mf4x5(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vssseg5.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vssseg5e16_v_bf16mf4x5(__bf16 *rs1, ptrdiff_t rs2,
+                                 vbfloat16mf4x5_t vs3, size_t vl) {
+  return __riscv_vssseg5e16_v_bf16mf4x5(rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vssseg5e16_v_bf16mf2x5(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vssseg5.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vssseg5e16_v_bf16mf2x5(__bf16 *rs1, ptrdiff_t rs2,
+                                 vbfloat16mf2x5_t vs3, size_t vl) {
+  return __riscv_vssseg5e16_v_bf16mf2x5(rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vssseg5e16_v_bf16m1x5(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vssseg5.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vssseg5e16_v_bf16m1x5(__bf16 *rs1, ptrdiff_t rs2, vbfloat16m1x5_t vs3,
+                                size_t vl) {
+  return __riscv_vssseg5e16_v_bf16m1x5(rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vssseg5e16_v_bf16mf4x5_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vssseg5.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], ptr [[RS1]], i64 [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vssseg5e16_v_bf16mf4x5_m(vbool64_t vm, __bf16 *rs1, ptrdiff_t rs2,
+                                   vbfloat16mf4x5_t vs3, size_t vl) {
+  return __riscv_vssseg5e16_v_bf16mf4x5_m(vm, rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vssseg5e16_v_bf16mf2x5_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vssseg5.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], ptr [[RS1]], i64 [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vssseg5e16_v_bf16mf2x5_m(vbool32_t vm, __bf16 *rs1, ptrdiff_t rs2,
+                                   vbfloat16mf2x5_t vs3, size_t vl) {
+  return __riscv_vssseg5e16_v_bf16mf2x5_m(vm, rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vssseg5e16_v_bf16m1x5_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vssseg5.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], ptr [[RS1]], i64 [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vssseg5e16_v_bf16m1x5_m(vbool16_t vm, __bf16 *rs1, ptrdiff_t rs2,
+                                  vbfloat16m1x5_t vs3, size_t vl) {
+  return __riscv_vssseg5e16_v_bf16m1x5_m(vm, rs1, rs2, vs3, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vssseg6e16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vssseg6e16.c
new file mode 100644
index 000000000000..c74f19905a4d
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vssseg6e16.c
@@ -0,0 +1,111 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local void @test_vssseg6e16_v_bf16mf4x6(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vssseg6.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vssseg6e16_v_bf16mf4x6(__bf16 *rs1, ptrdiff_t rs2,
+                                 vbfloat16mf4x6_t vs3, size_t vl) {
+  return __riscv_vssseg6e16_v_bf16mf4x6(rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vssseg6e16_v_bf16mf2x6(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vssseg6.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vssseg6e16_v_bf16mf2x6(__bf16 *rs1, ptrdiff_t rs2,
+                                 vbfloat16mf2x6_t vs3, size_t vl) {
+  return __riscv_vssseg6e16_v_bf16mf2x6(rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vssseg6e16_v_bf16m1x6(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vssseg6.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vssseg6e16_v_bf16m1x6(__bf16 *rs1, ptrdiff_t rs2, vbfloat16m1x6_t vs3,
+                                size_t vl) {
+  return __riscv_vssseg6e16_v_bf16m1x6(rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vssseg6e16_v_bf16mf4x6_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vssseg6.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], ptr [[RS1]], i64 [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vssseg6e16_v_bf16mf4x6_m(vbool64_t vm, __bf16 *rs1, ptrdiff_t rs2,
+                                   vbfloat16mf4x6_t vs3, size_t vl) {
+  return __riscv_vssseg6e16_v_bf16mf4x6_m(vm, rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vssseg6e16_v_bf16mf2x6_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vssseg6.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], ptr [[RS1]], i64 [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vssseg6e16_v_bf16mf2x6_m(vbool32_t vm, __bf16 *rs1, ptrdiff_t rs2,
+                                   vbfloat16mf2x6_t vs3, size_t vl) {
+  return __riscv_vssseg6e16_v_bf16mf2x6_m(vm, rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vssseg6e16_v_bf16m1x6_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vssseg6.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], ptr [[RS1]], i64 [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vssseg6e16_v_bf16m1x6_m(vbool16_t vm, __bf16 *rs1, ptrdiff_t rs2,
+                                  vbfloat16m1x6_t vs3, size_t vl) {
+  return __riscv_vssseg6e16_v_bf16m1x6_m(vm, rs1, rs2, vs3, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vssseg7e16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vssseg7e16.c
new file mode 100644
index 000000000000..d1ff80f684a4
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vssseg7e16.c
@@ -0,0 +1,117 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local void @test_vssseg7e16_v_bf16mf4x7(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 6
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vssseg7.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vssseg7e16_v_bf16mf4x7(__bf16 *rs1, ptrdiff_t rs2,
+                                 vbfloat16mf4x7_t vs3, size_t vl) {
+  return __riscv_vssseg7e16_v_bf16mf4x7(rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vssseg7e16_v_bf16mf2x7(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 6
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vssseg7.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vssseg7e16_v_bf16mf2x7(__bf16 *rs1, ptrdiff_t rs2,
+                                 vbfloat16mf2x7_t vs3, size_t vl) {
+  return __riscv_vssseg7e16_v_bf16mf2x7(rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vssseg7e16_v_bf16m1x7(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 6
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vssseg7.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vssseg7e16_v_bf16m1x7(__bf16 *rs1, ptrdiff_t rs2, vbfloat16m1x7_t vs3,
+                                size_t vl) {
+  return __riscv_vssseg7e16_v_bf16m1x7(rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vssseg7e16_v_bf16mf4x7_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 6
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vssseg7.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], ptr [[RS1]], i64 [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vssseg7e16_v_bf16mf4x7_m(vbool64_t vm, __bf16 *rs1, ptrdiff_t rs2,
+                                   vbfloat16mf4x7_t vs3, size_t vl) {
+  return __riscv_vssseg7e16_v_bf16mf4x7_m(vm, rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vssseg7e16_v_bf16mf2x7_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 6
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vssseg7.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], ptr [[RS1]], i64 [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vssseg7e16_v_bf16mf2x7_m(vbool32_t vm, __bf16 *rs1, ptrdiff_t rs2,
+                                   vbfloat16mf2x7_t vs3, size_t vl) {
+  return __riscv_vssseg7e16_v_bf16mf2x7_m(vm, rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vssseg7e16_v_bf16m1x7_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 6
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vssseg7.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], ptr [[RS1]], i64 [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vssseg7e16_v_bf16m1x7_m(vbool16_t vm, __bf16 *rs1, ptrdiff_t rs2,
+                                  vbfloat16m1x7_t vs3, size_t vl) {
+  return __riscv_vssseg7e16_v_bf16m1x7_m(vm, rs1, rs2, vs3, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vssseg8e16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vssseg8e16.c
new file mode 100644
index 000000000000..e80ff10dab50
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vssseg8e16.c
@@ -0,0 +1,123 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local void @test_vssseg8e16_v_bf16mf4x8(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 7
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vssseg8.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], <vscale x 1 x bfloat> [[TMP7]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vssseg8e16_v_bf16mf4x8(__bf16 *rs1, ptrdiff_t rs2,
+                                 vbfloat16mf4x8_t vs3, size_t vl) {
+  return __riscv_vssseg8e16_v_bf16mf4x8(rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vssseg8e16_v_bf16mf2x8(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 7
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vssseg8.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], <vscale x 2 x bfloat> [[TMP7]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vssseg8e16_v_bf16mf2x8(__bf16 *rs1, ptrdiff_t rs2,
+                                 vbfloat16mf2x8_t vs3, size_t vl) {
+  return __riscv_vssseg8e16_v_bf16mf2x8(rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vssseg8e16_v_bf16m1x8(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 7
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vssseg8.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], <vscale x 4 x bfloat> [[TMP7]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vssseg8e16_v_bf16m1x8(__bf16 *rs1, ptrdiff_t rs2, vbfloat16m1x8_t vs3,
+                                size_t vl) {
+  return __riscv_vssseg8e16_v_bf16m1x8(rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vssseg8e16_v_bf16mf4x8_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 7
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vssseg8.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], <vscale x 1 x bfloat> [[TMP7]], ptr [[RS1]], i64 [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vssseg8e16_v_bf16mf4x8_m(vbool64_t vm, __bf16 *rs1, ptrdiff_t rs2,
+                                   vbfloat16mf4x8_t vs3, size_t vl) {
+  return __riscv_vssseg8e16_v_bf16mf4x8_m(vm, rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vssseg8e16_v_bf16mf2x8_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 7
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vssseg8.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], <vscale x 2 x bfloat> [[TMP7]], ptr [[RS1]], i64 [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vssseg8e16_v_bf16mf2x8_m(vbool32_t vm, __bf16 *rs1, ptrdiff_t rs2,
+                                   vbfloat16mf2x8_t vs3, size_t vl) {
+  return __riscv_vssseg8e16_v_bf16mf2x8_m(vm, rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vssseg8e16_v_bf16m1x8_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 7
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vssseg8.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], <vscale x 4 x bfloat> [[TMP7]], ptr [[RS1]], i64 [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vssseg8e16_v_bf16m1x8_m(vbool16_t vm, __bf16 *rs1, ptrdiff_t rs2,
+                                  vbfloat16m1x8_t vs3, size_t vl) {
+  return __riscv_vssseg8e16_v_bf16m1x8_m(vm, rs1, rs2, vs3, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vsuxei16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vsuxei16.c
new file mode 100644
index 000000000000..5bbff4314620
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vsuxei16.c
@@ -0,0 +1,141 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxei16_v_bf16mf4(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], <vscale x 1 x bfloat> [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxei.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[VS3]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxei16_v_bf16mf4(__bf16 *rs1, vuint16mf4_t rs2, vbfloat16mf4_t vs3,
+                             size_t vl) {
+  return __riscv_vsuxei16_v_bf16mf4(rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxei16_v_bf16mf2(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], <vscale x 2 x bfloat> [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxei.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[VS3]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxei16_v_bf16mf2(__bf16 *rs1, vuint16mf2_t rs2, vbfloat16mf2_t vs3,
+                             size_t vl) {
+  return __riscv_vsuxei16_v_bf16mf2(rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxei16_v_bf16m1(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], <vscale x 4 x bfloat> [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxei.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[VS3]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxei16_v_bf16m1(__bf16 *rs1, vuint16m1_t rs2, vbfloat16m1_t vs3,
+                            size_t vl) {
+  return __riscv_vsuxei16_v_bf16m1(rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxei16_v_bf16m2(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[RS2:%.*]], <vscale x 8 x bfloat> [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxei.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> [[VS3]], ptr [[RS1]], <vscale x 8 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxei16_v_bf16m2(__bf16 *rs1, vuint16m2_t rs2, vbfloat16m2_t vs3,
+                            size_t vl) {
+  return __riscv_vsuxei16_v_bf16m2(rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxei16_v_bf16m4(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 16 x i16> [[RS2:%.*]], <vscale x 16 x bfloat> [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxei.nxv16bf16.nxv16i16.i64(<vscale x 16 x bfloat> [[VS3]], ptr [[RS1]], <vscale x 16 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxei16_v_bf16m4(__bf16 *rs1, vuint16m4_t rs2, vbfloat16m4_t vs3,
+                            size_t vl) {
+  return __riscv_vsuxei16_v_bf16m4(rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxei16_v_bf16m8(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 32 x i16> [[RS2:%.*]], <vscale x 32 x bfloat> [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxei.nxv32bf16.nxv32i16.i64(<vscale x 32 x bfloat> [[VS3]], ptr [[RS1]], <vscale x 32 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxei16_v_bf16m8(__bf16 *rs1, vuint16m8_t rs2, vbfloat16m8_t vs3,
+                            size_t vl) {
+  return __riscv_vsuxei16_v_bf16m8(rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxei16_v_bf16mf4_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], <vscale x 1 x bfloat> [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxei.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[VS3]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxei16_v_bf16mf4_m(vbool64_t vm, __bf16 *rs1, vuint16mf4_t rs2,
+                               vbfloat16mf4_t vs3, size_t vl) {
+  return __riscv_vsuxei16_v_bf16mf4_m(vm, rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxei16_v_bf16mf2_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], <vscale x 2 x bfloat> [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxei.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[VS3]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxei16_v_bf16mf2_m(vbool32_t vm, __bf16 *rs1, vuint16mf2_t rs2,
+                               vbfloat16mf2_t vs3, size_t vl) {
+  return __riscv_vsuxei16_v_bf16mf2_m(vm, rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxei16_v_bf16m1_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], <vscale x 4 x bfloat> [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxei.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[VS3]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxei16_v_bf16m1_m(vbool16_t vm, __bf16 *rs1, vuint16m1_t rs2,
+                              vbfloat16m1_t vs3, size_t vl) {
+  return __riscv_vsuxei16_v_bf16m1_m(vm, rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxei16_v_bf16m2_m(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[RS2:%.*]], <vscale x 8 x bfloat> [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxei.mask.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> [[VS3]], ptr [[RS1]], <vscale x 8 x i16> [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxei16_v_bf16m2_m(vbool8_t vm, __bf16 *rs1, vuint16m2_t rs2,
+                              vbfloat16m2_t vs3, size_t vl) {
+  return __riscv_vsuxei16_v_bf16m2_m(vm, rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxei16_v_bf16m4_m(
+// CHECK-RV64-SAME: <vscale x 16 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 16 x i16> [[RS2:%.*]], <vscale x 16 x bfloat> [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxei.mask.nxv16bf16.nxv16i16.i64(<vscale x 16 x bfloat> [[VS3]], ptr [[RS1]], <vscale x 16 x i16> [[RS2]], <vscale x 16 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxei16_v_bf16m4_m(vbool4_t vm, __bf16 *rs1, vuint16m4_t rs2,
+                              vbfloat16m4_t vs3, size_t vl) {
+  return __riscv_vsuxei16_v_bf16m4_m(vm, rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxei16_v_bf16m8_m(
+// CHECK-RV64-SAME: <vscale x 32 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 32 x i16> [[RS2:%.*]], <vscale x 32 x bfloat> [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxei.mask.nxv32bf16.nxv32i16.i64(<vscale x 32 x bfloat> [[VS3]], ptr [[RS1]], <vscale x 32 x i16> [[RS2]], <vscale x 32 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxei16_v_bf16m8_m(vbool2_t vm, __bf16 *rs1, vuint16m8_t rs2,
+                              vbfloat16m8_t vs3, size_t vl) {
+  return __riscv_vsuxei16_v_bf16m8_m(vm, rs1, rs2, vs3, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vsuxseg2ei16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vsuxseg2ei16.c
new file mode 100644
index 000000000000..27db36fa9531
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vsuxseg2ei16.c
@@ -0,0 +1,141 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxseg2ei16_v_bf16mf4x2(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[VS2:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxseg2.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 1 x i16> [[VS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxseg2ei16_v_bf16mf4x2(__bf16 *rs1, vuint16mf4_t vs2,
+                                   vbfloat16mf4x2_t vs3, size_t vl) {
+  return __riscv_vsuxseg2ei16_v_bf16mf4x2(rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxseg2ei16_v_bf16mf2x2(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[VS2:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxseg2.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 2 x i16> [[VS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxseg2ei16_v_bf16mf2x2(__bf16 *rs1, vuint16mf2_t vs2,
+                                   vbfloat16mf2x2_t vs3, size_t vl) {
+  return __riscv_vsuxseg2ei16_v_bf16mf2x2(rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxseg2ei16_v_bf16m1x2(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[VS2:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxseg2.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 4 x i16> [[VS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxseg2ei16_v_bf16m1x2(__bf16 *rs1, vuint16m1_t vs2,
+                                  vbfloat16m1x2_t vs3, size_t vl) {
+  return __riscv_vsuxseg2ei16_v_bf16m1x2(rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxseg2ei16_v_bf16m2x2(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[VS2:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxseg2.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 8 x i16> [[VS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxseg2ei16_v_bf16m2x2(__bf16 *rs1, vuint16m2_t vs2,
+                                  vbfloat16m2x2_t vs3, size_t vl) {
+  return __riscv_vsuxseg2ei16_v_bf16m2x2(rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxseg2ei16_v_bf16m4x2(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 16 x i16> [[VS2:%.*]], { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxseg2.nxv16bf16.nxv16i16.i64(<vscale x 16 x bfloat> [[TMP0]], <vscale x 16 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 16 x i16> [[VS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxseg2ei16_v_bf16m4x2(__bf16 *rs1, vuint16m4_t vs2,
+                                  vbfloat16m4x2_t vs3, size_t vl) {
+  return __riscv_vsuxseg2ei16_v_bf16m4x2(rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxseg2ei16_v_bf16mf4x2_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[VS2:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxseg2.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 1 x i16> [[VS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxseg2ei16_v_bf16mf4x2_m(vbool64_t vm, __bf16 *rs1,
+                                     vuint16mf4_t vs2, vbfloat16mf4x2_t vs3,
+                                     size_t vl) {
+  return __riscv_vsuxseg2ei16_v_bf16mf4x2_m(vm, rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxseg2ei16_v_bf16mf2x2_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[VS2:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxseg2.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 2 x i16> [[VS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxseg2ei16_v_bf16mf2x2_m(vbool32_t vm, __bf16 *rs1,
+                                     vuint16mf2_t vs2, vbfloat16mf2x2_t vs3,
+                                     size_t vl) {
+  return __riscv_vsuxseg2ei16_v_bf16mf2x2_m(vm, rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxseg2ei16_v_bf16m1x2_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[VS2:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxseg2.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 4 x i16> [[VS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxseg2ei16_v_bf16m1x2_m(vbool16_t vm, __bf16 *rs1, vuint16m1_t vs2,
+                                    vbfloat16m1x2_t vs3, size_t vl) {
+  return __riscv_vsuxseg2ei16_v_bf16m1x2_m(vm, rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxseg2ei16_v_bf16m2x2_m(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[VS2:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxseg2.mask.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 8 x i16> [[VS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxseg2ei16_v_bf16m2x2_m(vbool8_t vm, __bf16 *rs1, vuint16m2_t vs2,
+                                    vbfloat16m2x2_t vs3, size_t vl) {
+  return __riscv_vsuxseg2ei16_v_bf16m2x2_m(vm, rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxseg2ei16_v_bf16m4x2_m(
+// CHECK-RV64-SAME: <vscale x 16 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 16 x i16> [[VS2:%.*]], { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxseg2.mask.nxv16bf16.nxv16i16.i64(<vscale x 16 x bfloat> [[TMP0]], <vscale x 16 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 16 x i16> [[VS2]], <vscale x 16 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxseg2ei16_v_bf16m4x2_m(vbool4_t vm, __bf16 *rs1, vuint16m4_t vs2,
+                                    vbfloat16m4x2_t vs3, size_t vl) {
+  return __riscv_vsuxseg2ei16_v_bf16m4x2_m(vm, rs1, vs2, vs3, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vsuxseg3ei16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vsuxseg3ei16.c
new file mode 100644
index 000000000000..9c4823736895
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vsuxseg3ei16.c
@@ -0,0 +1,123 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxseg3ei16_v_bf16mf4x3(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[VS2:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxseg3.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 1 x i16> [[VS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxseg3ei16_v_bf16mf4x3(__bf16 *rs1, vuint16mf4_t vs2,
+                                   vbfloat16mf4x3_t vs3, size_t vl) {
+  return __riscv_vsuxseg3ei16_v_bf16mf4x3(rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxseg3ei16_v_bf16mf2x3(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[VS2:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxseg3.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 2 x i16> [[VS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxseg3ei16_v_bf16mf2x3(__bf16 *rs1, vuint16mf2_t vs2,
+                                   vbfloat16mf2x3_t vs3, size_t vl) {
+  return __riscv_vsuxseg3ei16_v_bf16mf2x3(rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxseg3ei16_v_bf16m1x3(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[VS2:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxseg3.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 4 x i16> [[VS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxseg3ei16_v_bf16m1x3(__bf16 *rs1, vuint16m1_t vs2,
+                                  vbfloat16m1x3_t vs3, size_t vl) {
+  return __riscv_vsuxseg3ei16_v_bf16m1x3(rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxseg3ei16_v_bf16m2x3(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[VS2:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxseg3.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 8 x i16> [[VS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxseg3ei16_v_bf16m2x3(__bf16 *rs1, vuint16m2_t vs2,
+                                  vbfloat16m2x3_t vs3, size_t vl) {
+  return __riscv_vsuxseg3ei16_v_bf16m2x3(rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxseg3ei16_v_bf16mf4x3_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[VS2:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxseg3.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 1 x i16> [[VS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxseg3ei16_v_bf16mf4x3_m(vbool64_t vm, __bf16 *rs1,
+                                     vuint16mf4_t vs2, vbfloat16mf4x3_t vs3,
+                                     size_t vl) {
+  return __riscv_vsuxseg3ei16_v_bf16mf4x3_m(vm, rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxseg3ei16_v_bf16mf2x3_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[VS2:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxseg3.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 2 x i16> [[VS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxseg3ei16_v_bf16mf2x3_m(vbool32_t vm, __bf16 *rs1,
+                                     vuint16mf2_t vs2, vbfloat16mf2x3_t vs3,
+                                     size_t vl) {
+  return __riscv_vsuxseg3ei16_v_bf16mf2x3_m(vm, rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxseg3ei16_v_bf16m1x3_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[VS2:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxseg3.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 4 x i16> [[VS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxseg3ei16_v_bf16m1x3_m(vbool16_t vm, __bf16 *rs1, vuint16m1_t vs2,
+                                    vbfloat16m1x3_t vs3, size_t vl) {
+  return __riscv_vsuxseg3ei16_v_bf16m1x3_m(vm, rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxseg3ei16_v_bf16m2x3_m(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[VS2:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxseg3.mask.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 8 x i16> [[VS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxseg3ei16_v_bf16m2x3_m(vbool8_t vm, __bf16 *rs1, vuint16m2_t vs2,
+                                    vbfloat16m2x3_t vs3, size_t vl) {
+  return __riscv_vsuxseg3ei16_v_bf16m2x3_m(vm, rs1, vs2, vs3, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vsuxseg4ei16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vsuxseg4ei16.c
new file mode 100644
index 000000000000..63b13ec6a32f
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vsuxseg4ei16.c
@@ -0,0 +1,131 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxseg4ei16_v_bf16mf4x4(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[VS2:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxseg4.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 1 x i16> [[VS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxseg4ei16_v_bf16mf4x4(__bf16 *rs1, vuint16mf4_t vs2,
+                                   vbfloat16mf4x4_t vs3, size_t vl) {
+  return __riscv_vsuxseg4ei16_v_bf16mf4x4(rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxseg4ei16_v_bf16mf2x4(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[VS2:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxseg4.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 2 x i16> [[VS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxseg4ei16_v_bf16mf2x4(__bf16 *rs1, vuint16mf2_t vs2,
+                                   vbfloat16mf2x4_t vs3, size_t vl) {
+  return __riscv_vsuxseg4ei16_v_bf16mf2x4(rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxseg4ei16_v_bf16m1x4(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[VS2:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxseg4.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 4 x i16> [[VS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxseg4ei16_v_bf16m1x4(__bf16 *rs1, vuint16m1_t vs2,
+                                  vbfloat16m1x4_t vs3, size_t vl) {
+  return __riscv_vsuxseg4ei16_v_bf16m1x4(rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxseg4ei16_v_bf16m2x4(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[VS2:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxseg4.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], <vscale x 8 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 8 x i16> [[VS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxseg4ei16_v_bf16m2x4(__bf16 *rs1, vuint16m2_t vs2,
+                                  vbfloat16m2x4_t vs3, size_t vl) {
+  return __riscv_vsuxseg4ei16_v_bf16m2x4(rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxseg4ei16_v_bf16mf4x4_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[VS2:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxseg4.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 1 x i16> [[VS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxseg4ei16_v_bf16mf4x4_m(vbool64_t vm, __bf16 *rs1,
+                                     vuint16mf4_t vs2, vbfloat16mf4x4_t vs3,
+                                     size_t vl) {
+  return __riscv_vsuxseg4ei16_v_bf16mf4x4_m(vm, rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxseg4ei16_v_bf16mf2x4_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[VS2:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxseg4.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 2 x i16> [[VS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxseg4ei16_v_bf16mf2x4_m(vbool32_t vm, __bf16 *rs1,
+                                     vuint16mf2_t vs2, vbfloat16mf2x4_t vs3,
+                                     size_t vl) {
+  return __riscv_vsuxseg4ei16_v_bf16mf2x4_m(vm, rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxseg4ei16_v_bf16m1x4_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[VS2:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxseg4.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 4 x i16> [[VS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxseg4ei16_v_bf16m1x4_m(vbool16_t vm, __bf16 *rs1, vuint16m1_t vs2,
+                                    vbfloat16m1x4_t vs3, size_t vl) {
+  return __riscv_vsuxseg4ei16_v_bf16m1x4_m(vm, rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxseg4ei16_v_bf16m2x4_m(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[VS2:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxseg4.mask.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], <vscale x 8 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 8 x i16> [[VS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxseg4ei16_v_bf16m2x4_m(vbool8_t vm, __bf16 *rs1, vuint16m2_t vs2,
+                                    vbfloat16m2x4_t vs3, size_t vl) {
+  return __riscv_vsuxseg4ei16_v_bf16m2x4_m(vm, rs1, vs2, vs3, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vsuxseg5ei16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vsuxseg5ei16.c
new file mode 100644
index 000000000000..4d93f05b0a10
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vsuxseg5ei16.c
@@ -0,0 +1,107 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxseg5ei16_v_bf16mf4x5(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[VS2:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxseg5.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 1 x i16> [[VS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxseg5ei16_v_bf16mf4x5(__bf16 *rs1, vuint16mf4_t vs2,
+                                   vbfloat16mf4x5_t vs3, size_t vl) {
+  return __riscv_vsuxseg5ei16_v_bf16mf4x5(rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxseg5ei16_v_bf16mf2x5(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[VS2:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxseg5.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 2 x i16> [[VS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxseg5ei16_v_bf16mf2x5(__bf16 *rs1, vuint16mf2_t vs2,
+                                   vbfloat16mf2x5_t vs3, size_t vl) {
+  return __riscv_vsuxseg5ei16_v_bf16mf2x5(rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxseg5ei16_v_bf16m1x5(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[VS2:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxseg5.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 4 x i16> [[VS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxseg5ei16_v_bf16m1x5(__bf16 *rs1, vuint16m1_t vs2,
+                                  vbfloat16m1x5_t vs3, size_t vl) {
+  return __riscv_vsuxseg5ei16_v_bf16m1x5(rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxseg5ei16_v_bf16mf4x5_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[VS2:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxseg5.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 1 x i16> [[VS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxseg5ei16_v_bf16mf4x5_m(vbool64_t vm, __bf16 *rs1,
+                                     vuint16mf4_t vs2, vbfloat16mf4x5_t vs3,
+                                     size_t vl) {
+  return __riscv_vsuxseg5ei16_v_bf16mf4x5_m(vm, rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxseg5ei16_v_bf16mf2x5_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[VS2:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxseg5.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 2 x i16> [[VS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxseg5ei16_v_bf16mf2x5_m(vbool32_t vm, __bf16 *rs1,
+                                     vuint16mf2_t vs2, vbfloat16mf2x5_t vs3,
+                                     size_t vl) {
+  return __riscv_vsuxseg5ei16_v_bf16mf2x5_m(vm, rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxseg5ei16_v_bf16m1x5_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[VS2:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxseg5.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 4 x i16> [[VS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxseg5ei16_v_bf16m1x5_m(vbool16_t vm, __bf16 *rs1, vuint16m1_t vs2,
+                                    vbfloat16m1x5_t vs3, size_t vl) {
+  return __riscv_vsuxseg5ei16_v_bf16m1x5_m(vm, rs1, vs2, vs3, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vsuxseg6ei16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vsuxseg6ei16.c
new file mode 100644
index 000000000000..0b6ef26c3751
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vsuxseg6ei16.c
@@ -0,0 +1,113 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxseg6ei16_v_bf16mf4x6(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[VS2:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxseg6.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 1 x i16> [[VS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxseg6ei16_v_bf16mf4x6(__bf16 *rs1, vuint16mf4_t vs2,
+                                   vbfloat16mf4x6_t vs3, size_t vl) {
+  return __riscv_vsuxseg6ei16_v_bf16mf4x6(rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxseg6ei16_v_bf16mf2x6(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[VS2:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxseg6.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 2 x i16> [[VS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxseg6ei16_v_bf16mf2x6(__bf16 *rs1, vuint16mf2_t vs2,
+                                   vbfloat16mf2x6_t vs3, size_t vl) {
+  return __riscv_vsuxseg6ei16_v_bf16mf2x6(rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxseg6ei16_v_bf16m1x6(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[VS2:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxseg6.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 4 x i16> [[VS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxseg6ei16_v_bf16m1x6(__bf16 *rs1, vuint16m1_t vs2,
+                                  vbfloat16m1x6_t vs3, size_t vl) {
+  return __riscv_vsuxseg6ei16_v_bf16m1x6(rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxseg6ei16_v_bf16mf4x6_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[VS2:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxseg6.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 1 x i16> [[VS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxseg6ei16_v_bf16mf4x6_m(vbool64_t vm, __bf16 *rs1,
+                                     vuint16mf4_t vs2, vbfloat16mf4x6_t vs3,
+                                     size_t vl) {
+  return __riscv_vsuxseg6ei16_v_bf16mf4x6_m(vm, rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxseg6ei16_v_bf16mf2x6_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[VS2:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxseg6.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 2 x i16> [[VS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxseg6ei16_v_bf16mf2x6_m(vbool32_t vm, __bf16 *rs1,
+                                     vuint16mf2_t vs2, vbfloat16mf2x6_t vs3,
+                                     size_t vl) {
+  return __riscv_vsuxseg6ei16_v_bf16mf2x6_m(vm, rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxseg6ei16_v_bf16m1x6_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[VS2:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxseg6.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 4 x i16> [[VS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxseg6ei16_v_bf16m1x6_m(vbool16_t vm, __bf16 *rs1, vuint16m1_t vs2,
+                                    vbfloat16m1x6_t vs3, size_t vl) {
+  return __riscv_vsuxseg6ei16_v_bf16m1x6_m(vm, rs1, vs2, vs3, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vsuxseg7ei16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vsuxseg7ei16.c
new file mode 100644
index 000000000000..10debde6ef60
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vsuxseg7ei16.c
@@ -0,0 +1,119 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxseg7ei16_v_bf16mf4x7(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[VS2:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 6
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxseg7.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 1 x i16> [[VS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxseg7ei16_v_bf16mf4x7(__bf16 *rs1, vuint16mf4_t vs2,
+                                   vbfloat16mf4x7_t vs3, size_t vl) {
+  return __riscv_vsuxseg7ei16_v_bf16mf4x7(rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxseg7ei16_v_bf16mf2x7(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[VS2:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 6
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxseg7.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 2 x i16> [[VS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxseg7ei16_v_bf16mf2x7(__bf16 *rs1, vuint16mf2_t vs2,
+                                   vbfloat16mf2x7_t vs3, size_t vl) {
+  return __riscv_vsuxseg7ei16_v_bf16mf2x7(rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxseg7ei16_v_bf16m1x7(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[VS2:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 6
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxseg7.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 4 x i16> [[VS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxseg7ei16_v_bf16m1x7(__bf16 *rs1, vuint16m1_t vs2,
+                                  vbfloat16m1x7_t vs3, size_t vl) {
+  return __riscv_vsuxseg7ei16_v_bf16m1x7(rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxseg7ei16_v_bf16mf4x7_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[VS2:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 6
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxseg7.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 1 x i16> [[VS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxseg7ei16_v_bf16mf4x7_m(vbool64_t vm, __bf16 *rs1,
+                                     vuint16mf4_t vs2, vbfloat16mf4x7_t vs3,
+                                     size_t vl) {
+  return __riscv_vsuxseg7ei16_v_bf16mf4x7_m(vm, rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxseg7ei16_v_bf16mf2x7_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[VS2:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 6
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxseg7.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 2 x i16> [[VS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxseg7ei16_v_bf16mf2x7_m(vbool32_t vm, __bf16 *rs1,
+                                     vuint16mf2_t vs2, vbfloat16mf2x7_t vs3,
+                                     size_t vl) {
+  return __riscv_vsuxseg7ei16_v_bf16mf2x7_m(vm, rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxseg7ei16_v_bf16m1x7_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[VS2:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 6
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxseg7.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 4 x i16> [[VS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxseg7ei16_v_bf16m1x7_m(vbool16_t vm, __bf16 *rs1, vuint16m1_t vs2,
+                                    vbfloat16m1x7_t vs3, size_t vl) {
+  return __riscv_vsuxseg7ei16_v_bf16m1x7_m(vm, rs1, vs2, vs3, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vsuxseg8ei16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vsuxseg8ei16.c
new file mode 100644
index 000000000000..175c7c478173
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vsuxseg8ei16.c
@@ -0,0 +1,125 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxseg8ei16_v_bf16mf4x8(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[VS2:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 7
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxseg8.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], <vscale x 1 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 1 x i16> [[VS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxseg8ei16_v_bf16mf4x8(__bf16 *rs1, vuint16mf4_t vs2,
+                                   vbfloat16mf4x8_t vs3, size_t vl) {
+  return __riscv_vsuxseg8ei16_v_bf16mf4x8(rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxseg8ei16_v_bf16mf2x8(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[VS2:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 7
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxseg8.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], <vscale x 2 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 2 x i16> [[VS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxseg8ei16_v_bf16mf2x8(__bf16 *rs1, vuint16mf2_t vs2,
+                                   vbfloat16mf2x8_t vs3, size_t vl) {
+  return __riscv_vsuxseg8ei16_v_bf16mf2x8(rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxseg8ei16_v_bf16m1x8(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[VS2:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 7
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxseg8.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], <vscale x 4 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 4 x i16> [[VS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxseg8ei16_v_bf16m1x8(__bf16 *rs1, vuint16m1_t vs2,
+                                  vbfloat16m1x8_t vs3, size_t vl) {
+  return __riscv_vsuxseg8ei16_v_bf16m1x8(rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxseg8ei16_v_bf16mf4x8_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[VS2:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 7
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxseg8.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], <vscale x 1 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 1 x i16> [[VS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxseg8ei16_v_bf16mf4x8_m(vbool64_t vm, __bf16 *rs1,
+                                     vuint16mf4_t vs2, vbfloat16mf4x8_t vs3,
+                                     size_t vl) {
+  return __riscv_vsuxseg8ei16_v_bf16mf4x8_m(vm, rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxseg8ei16_v_bf16mf2x8_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[VS2:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 7
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxseg8.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], <vscale x 2 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 2 x i16> [[VS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxseg8ei16_v_bf16mf2x8_m(vbool32_t vm, __bf16 *rs1,
+                                     vuint16mf2_t vs2, vbfloat16mf2x8_t vs3,
+                                     size_t vl) {
+  return __riscv_vsuxseg8ei16_v_bf16mf2x8_m(vm, rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxseg8ei16_v_bf16m1x8_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[VS2:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 7
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxseg8.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], <vscale x 4 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 4 x i16> [[VS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxseg8ei16_v_bf16m1x8_m(vbool16_t vm, __bf16 *rs1, vuint16m1_t vs2,
+                                    vbfloat16m1x8_t vs3, size_t vl) {
+  return __riscv_vsuxseg8ei16_v_bf16m1x8_m(vm, rs1, vs2, vs3, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vundefined.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vundefined.c
new file mode 100644
index 000000000000..5a2c8731b955
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/bfloat16/vundefined.c
@@ -0,0 +1,280 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 1 x bfloat> @test_vundefined_bf16mf4(
+// CHECK-RV64-SAME: ) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    ret <vscale x 1 x bfloat> poison
+//
+vbfloat16mf4_t test_vundefined_bf16mf4() {
+  return __riscv_vundefined_bf16mf4();
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 2 x bfloat> @test_vundefined_bf16mf2(
+// CHECK-RV64-SAME: ) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    ret <vscale x 2 x bfloat> poison
+//
+vbfloat16mf2_t test_vundefined_bf16mf2() {
+  return __riscv_vundefined_bf16mf2();
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 4 x bfloat> @test_vundefined_bf16m1(
+// CHECK-RV64-SAME: ) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    ret <vscale x 4 x bfloat> poison
+//
+vbfloat16m1_t test_vundefined_bf16m1() { return __riscv_vundefined_bf16m1(); }
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 8 x bfloat> @test_vundefined_bf16m2(
+// CHECK-RV64-SAME: ) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    ret <vscale x 8 x bfloat> poison
+//
+vbfloat16m2_t test_vundefined_bf16m2() { return __riscv_vundefined_bf16m2(); }
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 16 x bfloat> @test_vundefined_bf16m4(
+// CHECK-RV64-SAME: ) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    ret <vscale x 16 x bfloat> poison
+//
+vbfloat16m4_t test_vundefined_bf16m4() { return __riscv_vundefined_bf16m4(); }
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 32 x bfloat> @test_vundefined_bf16m8(
+// CHECK-RV64-SAME: ) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    ret <vscale x 32 x bfloat> poison
+//
+vbfloat16m8_t test_vundefined_bf16m8() { return __riscv_vundefined_bf16m8(); }
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vundefined_bf16mf4x2(
+// CHECK-RV64-SAME: ) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } poison
+//
+vbfloat16mf4x2_t test_vundefined_bf16mf4x2() {
+  return __riscv_vundefined_bf16mf4x2();
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vundefined_bf16mf4x3(
+// CHECK-RV64-SAME: ) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } poison
+//
+vbfloat16mf4x3_t test_vundefined_bf16mf4x3() {
+  return __riscv_vundefined_bf16mf4x3();
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vundefined_bf16mf4x4(
+// CHECK-RV64-SAME: ) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } poison
+//
+vbfloat16mf4x4_t test_vundefined_bf16mf4x4() {
+  return __riscv_vundefined_bf16mf4x4();
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vundefined_bf16mf4x5(
+// CHECK-RV64-SAME: ) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } poison
+//
+vbfloat16mf4x5_t test_vundefined_bf16mf4x5() {
+  return __riscv_vundefined_bf16mf4x5();
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vundefined_bf16mf4x6(
+// CHECK-RV64-SAME: ) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } poison
+//
+vbfloat16mf4x6_t test_vundefined_bf16mf4x6() {
+  return __riscv_vundefined_bf16mf4x6();
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vundefined_bf16mf4x7(
+// CHECK-RV64-SAME: ) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } poison
+//
+vbfloat16mf4x7_t test_vundefined_bf16mf4x7() {
+  return __riscv_vundefined_bf16mf4x7();
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vundefined_bf16mf4x8(
+// CHECK-RV64-SAME: ) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } poison
+//
+vbfloat16mf4x8_t test_vundefined_bf16mf4x8() {
+  return __riscv_vundefined_bf16mf4x8();
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vundefined_bf16mf2x2(
+// CHECK-RV64-SAME: ) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } poison
+//
+vbfloat16mf2x2_t test_vundefined_bf16mf2x2() {
+  return __riscv_vundefined_bf16mf2x2();
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vundefined_bf16mf2x3(
+// CHECK-RV64-SAME: ) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } poison
+//
+vbfloat16mf2x3_t test_vundefined_bf16mf2x3() {
+  return __riscv_vundefined_bf16mf2x3();
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vundefined_bf16mf2x4(
+// CHECK-RV64-SAME: ) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } poison
+//
+vbfloat16mf2x4_t test_vundefined_bf16mf2x4() {
+  return __riscv_vundefined_bf16mf2x4();
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vundefined_bf16mf2x5(
+// CHECK-RV64-SAME: ) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } poison
+//
+vbfloat16mf2x5_t test_vundefined_bf16mf2x5() {
+  return __riscv_vundefined_bf16mf2x5();
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vundefined_bf16mf2x6(
+// CHECK-RV64-SAME: ) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } poison
+//
+vbfloat16mf2x6_t test_vundefined_bf16mf2x6() {
+  return __riscv_vundefined_bf16mf2x6();
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vundefined_bf16mf2x7(
+// CHECK-RV64-SAME: ) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } poison
+//
+vbfloat16mf2x7_t test_vundefined_bf16mf2x7() {
+  return __riscv_vundefined_bf16mf2x7();
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vundefined_bf16mf2x8(
+// CHECK-RV64-SAME: ) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } poison
+//
+vbfloat16mf2x8_t test_vundefined_bf16mf2x8() {
+  return __riscv_vundefined_bf16mf2x8();
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vundefined_bf16m1x2(
+// CHECK-RV64-SAME: ) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } poison
+//
+vbfloat16m1x2_t test_vundefined_bf16m1x2() {
+  return __riscv_vundefined_bf16m1x2();
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vundefined_bf16m1x3(
+// CHECK-RV64-SAME: ) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } poison
+//
+vbfloat16m1x3_t test_vundefined_bf16m1x3() {
+  return __riscv_vundefined_bf16m1x3();
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vundefined_bf16m1x4(
+// CHECK-RV64-SAME: ) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } poison
+//
+vbfloat16m1x4_t test_vundefined_bf16m1x4() {
+  return __riscv_vundefined_bf16m1x4();
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vundefined_bf16m1x5(
+// CHECK-RV64-SAME: ) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } poison
+//
+vbfloat16m1x5_t test_vundefined_bf16m1x5() {
+  return __riscv_vundefined_bf16m1x5();
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vundefined_bf16m1x6(
+// CHECK-RV64-SAME: ) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } poison
+//
+vbfloat16m1x6_t test_vundefined_bf16m1x6() {
+  return __riscv_vundefined_bf16m1x6();
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vundefined_bf16m1x7(
+// CHECK-RV64-SAME: ) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } poison
+//
+vbfloat16m1x7_t test_vundefined_bf16m1x7() {
+  return __riscv_vundefined_bf16m1x7();
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vundefined_bf16m1x8(
+// CHECK-RV64-SAME: ) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } poison
+//
+vbfloat16m1x8_t test_vundefined_bf16m1x8() {
+  return __riscv_vundefined_bf16m1x8();
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vundefined_bf16m2x2(
+// CHECK-RV64-SAME: ) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } poison
+//
+vbfloat16m2x2_t test_vundefined_bf16m2x2() {
+  return __riscv_vundefined_bf16m2x2();
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vundefined_bf16m2x3(
+// CHECK-RV64-SAME: ) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } poison
+//
+vbfloat16m2x3_t test_vundefined_bf16m2x3() {
+  return __riscv_vundefined_bf16m2x3();
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vundefined_bf16m2x4(
+// CHECK-RV64-SAME: ) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } poison
+//
+vbfloat16m2x4_t test_vundefined_bf16m2x4() {
+  return __riscv_vundefined_bf16m2x4();
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @test_vundefined_bf16m4x2(
+// CHECK-RV64-SAME: ) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    ret { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } poison
+//
+vbfloat16m4x2_t test_vundefined_bf16m4x2() {
+  return __riscv_vundefined_bf16m4x2();
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/vaesdf.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/vaesdf.c
index 76a9ddc0d529..3e37ac4b7749 100644
--- a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/vaesdf.c
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/vaesdf.c
@@ -206,13 +206,3 @@ vuint32m8_t test_vaesdf_vv_u32m8(vuint32m8_t vd, vuint32m8_t vs2, size_t vl) {
   return __riscv_vaesdf_vv_u32m8(vd, vs2, vl);
 }
 
-// CHECK-RV64-LABEL: define dso_local <vscale x 16 x i32> @test_vaesdf_vs_u32m8_u32m8
-// CHECK-RV64-SAME: (<vscale x 16 x i32> [[VD:%.*]], <vscale x 16 x i32> [[VS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
-// CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vaesdf.vs.nxv16i32.nxv16i32.i64(<vscale x 16 x i32> [[VD]], <vscale x 16 x i32> [[VS2]], i64 [[VL]], i64 3)
-// CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
-//
-vuint32m8_t test_vaesdf_vs_u32m8_u32m8(vuint32m8_t vd, vuint32m8_t vs2, size_t vl) {
-  return __riscv_vaesdf_vs_u32m8_u32m8(vd, vs2, vl);
-}
-
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/vaesdm.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/vaesdm.c
index 468c3f18378d..c29c1e983fce 100644
--- a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/vaesdm.c
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/vaesdm.c
@@ -206,13 +206,3 @@ vuint32m8_t test_vaesdm_vv_u32m8(vuint32m8_t vd, vuint32m8_t vs2, size_t vl) {
   return __riscv_vaesdm_vv_u32m8(vd, vs2, vl);
 }
 
-// CHECK-RV64-LABEL: define dso_local <vscale x 16 x i32> @test_vaesdm_vs_u32m8_u32m8
-// CHECK-RV64-SAME: (<vscale x 16 x i32> [[VD:%.*]], <vscale x 16 x i32> [[VS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
-// CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vaesdm.vs.nxv16i32.nxv16i32.i64(<vscale x 16 x i32> [[VD]], <vscale x 16 x i32> [[VS2]], i64 [[VL]], i64 3)
-// CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
-//
-vuint32m8_t test_vaesdm_vs_u32m8_u32m8(vuint32m8_t vd, vuint32m8_t vs2, size_t vl) {
-  return __riscv_vaesdm_vs_u32m8_u32m8(vd, vs2, vl);
-}
-
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/vaesef.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/vaesef.c
index bc6a17e4b6f0..7ed9f5c36e98 100644
--- a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/vaesef.c
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/vaesef.c
@@ -206,13 +206,3 @@ vuint32m8_t test_vaesef_vv_u32m8(vuint32m8_t vd, vuint32m8_t vs2, size_t vl) {
   return __riscv_vaesef_vv_u32m8(vd, vs2, vl);
 }
 
-// CHECK-RV64-LABEL: define dso_local <vscale x 16 x i32> @test_vaesef_vs_u32m8_u32m8
-// CHECK-RV64-SAME: (<vscale x 16 x i32> [[VD:%.*]], <vscale x 16 x i32> [[VS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
-// CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vaesef.vs.nxv16i32.nxv16i32.i64(<vscale x 16 x i32> [[VD]], <vscale x 16 x i32> [[VS2]], i64 [[VL]], i64 3)
-// CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
-//
-vuint32m8_t test_vaesef_vs_u32m8_u32m8(vuint32m8_t vd, vuint32m8_t vs2, size_t vl) {
-  return __riscv_vaesef_vs_u32m8_u32m8(vd, vs2, vl);
-}
-
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/vaesem.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/vaesem.c
index e0e1662b76f9..34eec0037491 100644
--- a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/vaesem.c
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/vaesem.c
@@ -206,13 +206,3 @@ vuint32m8_t test_vaesem_vv_u32m8(vuint32m8_t vd, vuint32m8_t vs2, size_t vl) {
   return __riscv_vaesem_vv_u32m8(vd, vs2, vl);
 }
 
-// CHECK-RV64-LABEL: define dso_local <vscale x 16 x i32> @test_vaesem_vs_u32m8_u32m8
-// CHECK-RV64-SAME: (<vscale x 16 x i32> [[VD:%.*]], <vscale x 16 x i32> [[VS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
-// CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vaesem.vs.nxv16i32.nxv16i32.i64(<vscale x 16 x i32> [[VD]], <vscale x 16 x i32> [[VS2]], i64 [[VL]], i64 3)
-// CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
-//
-vuint32m8_t test_vaesem_vs_u32m8_u32m8(vuint32m8_t vd, vuint32m8_t vs2, size_t vl) {
-  return __riscv_vaesem_vs_u32m8_u32m8(vd, vs2, vl);
-}
-
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/vaesz.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/vaesz.c
index 0700b60fa639..bfe0b0602717 100644
--- a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/vaesz.c
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/vaesz.c
@@ -156,13 +156,3 @@ vuint32m8_t test_vaesz_vs_u32m4_u32m8(vuint32m8_t vd, vuint32m4_t vs2, size_t vl
   return __riscv_vaesz_vs_u32m4_u32m8(vd, vs2, vl);
 }
 
-// CHECK-RV64-LABEL: define dso_local <vscale x 16 x i32> @test_vaesz_vs_u32m8_u32m8
-// CHECK-RV64-SAME: (<vscale x 16 x i32> [[VD:%.*]], <vscale x 16 x i32> [[VS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
-// CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vaesz.vs.nxv16i32.nxv16i32.i64(<vscale x 16 x i32> [[VD]], <vscale x 16 x i32> [[VS2]], i64 [[VL]], i64 3)
-// CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
-//
-vuint32m8_t test_vaesz_vs_u32m8_u32m8(vuint32m8_t vd, vuint32m8_t vs2, size_t vl) {
-  return __riscv_vaesz_vs_u32m8_u32m8(vd, vs2, vl);
-}
-
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/vsm4r.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/vsm4r.c
index f9d855a72d28..fe9090225596 100644
--- a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/vsm4r.c
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/non-overloaded/vsm4r.c
@@ -206,13 +206,3 @@ vuint32m8_t test_vsm4r_vv_u32m8(vuint32m8_t vd, vuint32m8_t vs2, size_t vl) {
   return __riscv_vsm4r_vv_u32m8(vd, vs2, vl);
 }
 
-// CHECK-RV64-LABEL: define dso_local <vscale x 16 x i32> @test_vsm4r_vs_u32m8_u32m8
-// CHECK-RV64-SAME: (<vscale x 16 x i32> [[VD:%.*]], <vscale x 16 x i32> [[VS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
-// CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vsm4r.vs.nxv16i32.nxv16i32.i64(<vscale x 16 x i32> [[VD]], <vscale x 16 x i32> [[VS2]], i64 [[VL]], i64 3)
-// CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
-//
-vuint32m8_t test_vsm4r_vs_u32m8_u32m8(vuint32m8_t vd, vuint32m8_t vs2, size_t vl) {
-  return __riscv_vsm4r_vs_u32m8_u32m8(vd, vs2, vl);
-}
-
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vget.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vget.c
new file mode 100644
index 000000000000..7f8ddf75156b
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vget.c
@@ -0,0 +1,333 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 4 x bfloat> @test_vget_v_bf16m2_bf16m1(
+// CHECK-RV64-SAME: <vscale x 8 x bfloat> [[SRC:%.*]], i64 noundef [[INDEX:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x bfloat> @llvm.vector.extract.nxv4bf16.nxv8bf16(<vscale x 8 x bfloat> [[SRC]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 4 x bfloat> [[TMP0]]
+//
+vbfloat16m1_t test_vget_v_bf16m2_bf16m1(vbfloat16m2_t src, size_t index) {
+  return __riscv_vget_bf16m1(src, 0);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 4 x bfloat> @test_vget_v_bf16m4_bf16m1(
+// CHECK-RV64-SAME: <vscale x 16 x bfloat> [[SRC:%.*]], i64 noundef [[INDEX:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x bfloat> @llvm.vector.extract.nxv4bf16.nxv16bf16(<vscale x 16 x bfloat> [[SRC]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 4 x bfloat> [[TMP0]]
+//
+vbfloat16m1_t test_vget_v_bf16m4_bf16m1(vbfloat16m4_t src, size_t index) {
+  return __riscv_vget_bf16m1(src, 0);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 4 x bfloat> @test_vget_v_bf16m8_bf16m1(
+// CHECK-RV64-SAME: <vscale x 32 x bfloat> [[SRC:%.*]], i64 noundef [[INDEX:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x bfloat> @llvm.vector.extract.nxv4bf16.nxv32bf16(<vscale x 32 x bfloat> [[SRC]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 4 x bfloat> [[TMP0]]
+//
+vbfloat16m1_t test_vget_v_bf16m8_bf16m1(vbfloat16m8_t src, size_t index) {
+  return __riscv_vget_bf16m1(src, 0);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 8 x bfloat> @test_vget_v_bf16m4_bf16m2(
+// CHECK-RV64-SAME: <vscale x 16 x bfloat> [[SRC:%.*]], i64 noundef [[INDEX:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x bfloat> @llvm.vector.extract.nxv8bf16.nxv16bf16(<vscale x 16 x bfloat> [[SRC]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 8 x bfloat> [[TMP0]]
+//
+vbfloat16m2_t test_vget_v_bf16m4_bf16m2(vbfloat16m4_t src, size_t index) {
+  return __riscv_vget_bf16m2(src, 0);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 8 x bfloat> @test_vget_v_bf16m8_bf16m2(
+// CHECK-RV64-SAME: <vscale x 32 x bfloat> [[SRC:%.*]], i64 noundef [[INDEX:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x bfloat> @llvm.vector.extract.nxv8bf16.nxv32bf16(<vscale x 32 x bfloat> [[SRC]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 8 x bfloat> [[TMP0]]
+//
+vbfloat16m2_t test_vget_v_bf16m8_bf16m2(vbfloat16m8_t src, size_t index) {
+  return __riscv_vget_bf16m2(src, 0);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 16 x bfloat> @test_vget_v_bf16m8_bf16m4(
+// CHECK-RV64-SAME: <vscale x 32 x bfloat> [[SRC:%.*]], i64 noundef [[INDEX:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x bfloat> @llvm.vector.extract.nxv16bf16.nxv32bf16(<vscale x 32 x bfloat> [[SRC]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 16 x bfloat> [[TMP0]]
+//
+vbfloat16m4_t test_vget_v_bf16m8_bf16m4(vbfloat16m8_t src, size_t index) {
+  return __riscv_vget_bf16m4(src, 0);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 1 x bfloat> @test_vget_v_bf16mf4x2_bf16mf4(
+// CHECK-RV64-SAME: { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[SRC:%.*]], i64 noundef [[INDEX:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[SRC]], 0
+// CHECK-RV64-NEXT:    ret <vscale x 1 x bfloat> [[TMP0]]
+//
+vbfloat16mf4_t test_vget_v_bf16mf4x2_bf16mf4(vbfloat16mf4x2_t src,
+                                             size_t index) {
+  return __riscv_vget_bf16mf4(src, 0);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 1 x bfloat> @test_vget_v_bf16mf4x3_bf16mf4(
+// CHECK-RV64-SAME: { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[SRC:%.*]], i64 noundef [[INDEX:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[SRC]], 0
+// CHECK-RV64-NEXT:    ret <vscale x 1 x bfloat> [[TMP0]]
+//
+vbfloat16mf4_t test_vget_v_bf16mf4x3_bf16mf4(vbfloat16mf4x3_t src,
+                                             size_t index) {
+  return __riscv_vget_bf16mf4(src, 0);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 1 x bfloat> @test_vget_v_bf16mf4x4_bf16mf4(
+// CHECK-RV64-SAME: { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[SRC:%.*]], i64 noundef [[INDEX:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[SRC]], 0
+// CHECK-RV64-NEXT:    ret <vscale x 1 x bfloat> [[TMP0]]
+//
+vbfloat16mf4_t test_vget_v_bf16mf4x4_bf16mf4(vbfloat16mf4x4_t src,
+                                             size_t index) {
+  return __riscv_vget_bf16mf4(src, 0);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 1 x bfloat> @test_vget_v_bf16mf4x5_bf16mf4(
+// CHECK-RV64-SAME: { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[SRC:%.*]], i64 noundef [[INDEX:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[SRC]], 0
+// CHECK-RV64-NEXT:    ret <vscale x 1 x bfloat> [[TMP0]]
+//
+vbfloat16mf4_t test_vget_v_bf16mf4x5_bf16mf4(vbfloat16mf4x5_t src,
+                                             size_t index) {
+  return __riscv_vget_bf16mf4(src, 0);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 1 x bfloat> @test_vget_v_bf16mf4x6_bf16mf4(
+// CHECK-RV64-SAME: { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[SRC:%.*]], i64 noundef [[INDEX:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[SRC]], 0
+// CHECK-RV64-NEXT:    ret <vscale x 1 x bfloat> [[TMP0]]
+//
+vbfloat16mf4_t test_vget_v_bf16mf4x6_bf16mf4(vbfloat16mf4x6_t src,
+                                             size_t index) {
+  return __riscv_vget_bf16mf4(src, 0);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 1 x bfloat> @test_vget_v_bf16mf4x7_bf16mf4(
+// CHECK-RV64-SAME: { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[SRC:%.*]], i64 noundef [[INDEX:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[SRC]], 0
+// CHECK-RV64-NEXT:    ret <vscale x 1 x bfloat> [[TMP0]]
+//
+vbfloat16mf4_t test_vget_v_bf16mf4x7_bf16mf4(vbfloat16mf4x7_t src,
+                                             size_t index) {
+  return __riscv_vget_bf16mf4(src, 0);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 1 x bfloat> @test_vget_v_bf16mf4x8_bf16mf4(
+// CHECK-RV64-SAME: { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[SRC:%.*]], i64 noundef [[INDEX:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[SRC]], 0
+// CHECK-RV64-NEXT:    ret <vscale x 1 x bfloat> [[TMP0]]
+//
+vbfloat16mf4_t test_vget_v_bf16mf4x8_bf16mf4(vbfloat16mf4x8_t src,
+                                             size_t index) {
+  return __riscv_vget_bf16mf4(src, 0);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 2 x bfloat> @test_vget_v_bf16mf2x2_bf16mf2(
+// CHECK-RV64-SAME: { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[SRC:%.*]], i64 noundef [[INDEX:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[SRC]], 0
+// CHECK-RV64-NEXT:    ret <vscale x 2 x bfloat> [[TMP0]]
+//
+vbfloat16mf2_t test_vget_v_bf16mf2x2_bf16mf2(vbfloat16mf2x2_t src,
+                                             size_t index) {
+  return __riscv_vget_bf16mf2(src, 0);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 2 x bfloat> @test_vget_v_bf16mf2x3_bf16mf2(
+// CHECK-RV64-SAME: { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[SRC:%.*]], i64 noundef [[INDEX:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[SRC]], 0
+// CHECK-RV64-NEXT:    ret <vscale x 2 x bfloat> [[TMP0]]
+//
+vbfloat16mf2_t test_vget_v_bf16mf2x3_bf16mf2(vbfloat16mf2x3_t src,
+                                             size_t index) {
+  return __riscv_vget_bf16mf2(src, 0);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 2 x bfloat> @test_vget_v_bf16mf2x4_bf16mf2(
+// CHECK-RV64-SAME: { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[SRC:%.*]], i64 noundef [[INDEX:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[SRC]], 0
+// CHECK-RV64-NEXT:    ret <vscale x 2 x bfloat> [[TMP0]]
+//
+vbfloat16mf2_t test_vget_v_bf16mf2x4_bf16mf2(vbfloat16mf2x4_t src,
+                                             size_t index) {
+  return __riscv_vget_bf16mf2(src, 0);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 2 x bfloat> @test_vget_v_bf16mf2x5_bf16mf2(
+// CHECK-RV64-SAME: { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[SRC:%.*]], i64 noundef [[INDEX:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[SRC]], 0
+// CHECK-RV64-NEXT:    ret <vscale x 2 x bfloat> [[TMP0]]
+//
+vbfloat16mf2_t test_vget_v_bf16mf2x5_bf16mf2(vbfloat16mf2x5_t src,
+                                             size_t index) {
+  return __riscv_vget_bf16mf2(src, 0);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 2 x bfloat> @test_vget_v_bf16mf2x6_bf16mf2(
+// CHECK-RV64-SAME: { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[SRC:%.*]], i64 noundef [[INDEX:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[SRC]], 0
+// CHECK-RV64-NEXT:    ret <vscale x 2 x bfloat> [[TMP0]]
+//
+vbfloat16mf2_t test_vget_v_bf16mf2x6_bf16mf2(vbfloat16mf2x6_t src,
+                                             size_t index) {
+  return __riscv_vget_bf16mf2(src, 0);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 2 x bfloat> @test_vget_v_bf16mf2x7_bf16mf2(
+// CHECK-RV64-SAME: { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[SRC:%.*]], i64 noundef [[INDEX:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[SRC]], 0
+// CHECK-RV64-NEXT:    ret <vscale x 2 x bfloat> [[TMP0]]
+//
+vbfloat16mf2_t test_vget_v_bf16mf2x7_bf16mf2(vbfloat16mf2x7_t src,
+                                             size_t index) {
+  return __riscv_vget_bf16mf2(src, 0);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 2 x bfloat> @test_vget_v_bf16mf2x8_bf16mf2(
+// CHECK-RV64-SAME: { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[SRC:%.*]], i64 noundef [[INDEX:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[SRC]], 0
+// CHECK-RV64-NEXT:    ret <vscale x 2 x bfloat> [[TMP0]]
+//
+vbfloat16mf2_t test_vget_v_bf16mf2x8_bf16mf2(vbfloat16mf2x8_t src,
+                                             size_t index) {
+  return __riscv_vget_bf16mf2(src, 0);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 4 x bfloat> @test_vget_v_bf16m1x2_bf16m1(
+// CHECK-RV64-SAME: { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[SRC:%.*]], i64 noundef [[INDEX:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[SRC]], 0
+// CHECK-RV64-NEXT:    ret <vscale x 4 x bfloat> [[TMP0]]
+//
+vbfloat16m1_t test_vget_v_bf16m1x2_bf16m1(vbfloat16m1x2_t src, size_t index) {
+  return __riscv_vget_bf16m1(src, 0);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 4 x bfloat> @test_vget_v_bf16m1x3_bf16m1(
+// CHECK-RV64-SAME: { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[SRC:%.*]], i64 noundef [[INDEX:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[SRC]], 0
+// CHECK-RV64-NEXT:    ret <vscale x 4 x bfloat> [[TMP0]]
+//
+vbfloat16m1_t test_vget_v_bf16m1x3_bf16m1(vbfloat16m1x3_t src, size_t index) {
+  return __riscv_vget_bf16m1(src, 0);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 4 x bfloat> @test_vget_v_bf16m1x4_bf16m1(
+// CHECK-RV64-SAME: { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[SRC:%.*]], i64 noundef [[INDEX:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[SRC]], 0
+// CHECK-RV64-NEXT:    ret <vscale x 4 x bfloat> [[TMP0]]
+//
+vbfloat16m1_t test_vget_v_bf16m1x4_bf16m1(vbfloat16m1x4_t src, size_t index) {
+  return __riscv_vget_bf16m1(src, 0);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 4 x bfloat> @test_vget_v_bf16m1x5_bf16m1(
+// CHECK-RV64-SAME: { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[SRC:%.*]], i64 noundef [[INDEX:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[SRC]], 0
+// CHECK-RV64-NEXT:    ret <vscale x 4 x bfloat> [[TMP0]]
+//
+vbfloat16m1_t test_vget_v_bf16m1x5_bf16m1(vbfloat16m1x5_t src, size_t index) {
+  return __riscv_vget_bf16m1(src, 0);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 4 x bfloat> @test_vget_v_bf16m1x6_bf16m1(
+// CHECK-RV64-SAME: { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[SRC:%.*]], i64 noundef [[INDEX:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[SRC]], 0
+// CHECK-RV64-NEXT:    ret <vscale x 4 x bfloat> [[TMP0]]
+//
+vbfloat16m1_t test_vget_v_bf16m1x6_bf16m1(vbfloat16m1x6_t src, size_t index) {
+  return __riscv_vget_bf16m1(src, 0);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 4 x bfloat> @test_vget_v_bf16m1x7_bf16m1(
+// CHECK-RV64-SAME: { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[SRC:%.*]], i64 noundef [[INDEX:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[SRC]], 0
+// CHECK-RV64-NEXT:    ret <vscale x 4 x bfloat> [[TMP0]]
+//
+vbfloat16m1_t test_vget_v_bf16m1x7_bf16m1(vbfloat16m1x7_t src, size_t index) {
+  return __riscv_vget_bf16m1(src, 0);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 4 x bfloat> @test_vget_v_bf16m1x8_bf16m1(
+// CHECK-RV64-SAME: { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[SRC:%.*]], i64 noundef [[INDEX:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[SRC]], 0
+// CHECK-RV64-NEXT:    ret <vscale x 4 x bfloat> [[TMP0]]
+//
+vbfloat16m1_t test_vget_v_bf16m1x8_bf16m1(vbfloat16m1x8_t src, size_t index) {
+  return __riscv_vget_bf16m1(src, 0);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 8 x bfloat> @test_vget_v_bf16m2x2_bf16m2(
+// CHECK-RV64-SAME: { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[SRC:%.*]], i64 noundef [[INDEX:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[SRC]], 0
+// CHECK-RV64-NEXT:    ret <vscale x 8 x bfloat> [[TMP0]]
+//
+vbfloat16m2_t test_vget_v_bf16m2x2_bf16m2(vbfloat16m2x2_t src, size_t index) {
+  return __riscv_vget_bf16m2(src, 0);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 8 x bfloat> @test_vget_v_bf16m2x3_bf16m2(
+// CHECK-RV64-SAME: { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[SRC:%.*]], i64 noundef [[INDEX:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[SRC]], 0
+// CHECK-RV64-NEXT:    ret <vscale x 8 x bfloat> [[TMP0]]
+//
+vbfloat16m2_t test_vget_v_bf16m2x3_bf16m2(vbfloat16m2x3_t src, size_t index) {
+  return __riscv_vget_bf16m2(src, 0);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 8 x bfloat> @test_vget_v_bf16m2x4_bf16m2(
+// CHECK-RV64-SAME: { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[SRC:%.*]], i64 noundef [[INDEX:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[SRC]], 0
+// CHECK-RV64-NEXT:    ret <vscale x 8 x bfloat> [[TMP0]]
+//
+vbfloat16m2_t test_vget_v_bf16m2x4_bf16m2(vbfloat16m2x4_t src, size_t index) {
+  return __riscv_vget_bf16m2(src, 0);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 16 x bfloat> @test_vget_v_bf16m4x2_bf16m4(
+// CHECK-RV64-SAME: { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[SRC:%.*]], i64 noundef [[INDEX:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[SRC]], 0
+// CHECK-RV64-NEXT:    ret <vscale x 16 x bfloat> [[TMP0]]
+//
+vbfloat16m4_t test_vget_v_bf16m4x2_bf16m4(vbfloat16m4x2_t src, size_t index) {
+  return __riscv_vget_bf16m4(src, 0);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vle16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vle16.c
new file mode 100644
index 000000000000..765e9d834638
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vle16.c
@@ -0,0 +1,72 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 1 x bfloat> @test_vle16_v_bf16mf4_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x bfloat> @llvm.riscv.vle.mask.nxv1bf16.i64(<vscale x 1 x bfloat> poison, ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret <vscale x 1 x bfloat> [[TMP0]]
+//
+vbfloat16mf4_t test_vle16_v_bf16mf4_m(vbool64_t vm, const __bf16 *rs1,
+                                      size_t vl) {
+  return __riscv_vle16(vm, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 2 x bfloat> @test_vle16_v_bf16mf2_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x bfloat> @llvm.riscv.vle.mask.nxv2bf16.i64(<vscale x 2 x bfloat> poison, ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret <vscale x 2 x bfloat> [[TMP0]]
+//
+vbfloat16mf2_t test_vle16_v_bf16mf2_m(vbool32_t vm, const __bf16 *rs1,
+                                      size_t vl) {
+  return __riscv_vle16(vm, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 4 x bfloat> @test_vle16_v_bf16m1_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x bfloat> @llvm.riscv.vle.mask.nxv4bf16.i64(<vscale x 4 x bfloat> poison, ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret <vscale x 4 x bfloat> [[TMP0]]
+//
+vbfloat16m1_t test_vle16_v_bf16m1_m(vbool16_t vm, const __bf16 *rs1,
+                                    size_t vl) {
+  return __riscv_vle16(vm, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 8 x bfloat> @test_vle16_v_bf16m2_m(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x bfloat> @llvm.riscv.vle.mask.nxv8bf16.i64(<vscale x 8 x bfloat> poison, ptr [[RS1]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret <vscale x 8 x bfloat> [[TMP0]]
+//
+vbfloat16m2_t test_vle16_v_bf16m2_m(vbool8_t vm, const __bf16 *rs1, size_t vl) {
+  return __riscv_vle16(vm, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 16 x bfloat> @test_vle16_v_bf16m4_m(
+// CHECK-RV64-SAME: <vscale x 16 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x bfloat> @llvm.riscv.vle.mask.nxv16bf16.i64(<vscale x 16 x bfloat> poison, ptr [[RS1]], <vscale x 16 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret <vscale x 16 x bfloat> [[TMP0]]
+//
+vbfloat16m4_t test_vle16_v_bf16m4_m(vbool4_t vm, const __bf16 *rs1, size_t vl) {
+  return __riscv_vle16(vm, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 32 x bfloat> @test_vle16_v_bf16m8_m(
+// CHECK-RV64-SAME: <vscale x 32 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x bfloat> @llvm.riscv.vle.mask.nxv32bf16.i64(<vscale x 32 x bfloat> poison, ptr [[RS1]], <vscale x 32 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret <vscale x 32 x bfloat> [[TMP0]]
+//
+vbfloat16m8_t test_vle16_v_bf16m8_m(vbool2_t vm, const __bf16 *rs1, size_t vl) {
+  return __riscv_vle16(vm, rs1, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vle16ff.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vle16ff.c
new file mode 100644
index 000000000000..9ef6a091b9d5
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vle16ff.c
@@ -0,0 +1,93 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 1 x bfloat> @test_vle16ff_v_bf16mf4_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, i64 } @llvm.riscv.vleff.mask.nxv1bf16.i64(<vscale x 1 x bfloat> poison, ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    store i64 [[TMP2]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret <vscale x 1 x bfloat> [[TMP1]]
+//
+vbfloat16mf4_t test_vle16ff_v_bf16mf4_m(vbool64_t vm, const __bf16 *rs1,
+                                        size_t *new_vl, size_t vl) {
+  return __riscv_vle16ff(vm, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 2 x bfloat> @test_vle16ff_v_bf16mf2_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, i64 } @llvm.riscv.vleff.mask.nxv2bf16.i64(<vscale x 2 x bfloat> poison, ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    store i64 [[TMP2]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret <vscale x 2 x bfloat> [[TMP1]]
+//
+vbfloat16mf2_t test_vle16ff_v_bf16mf2_m(vbool32_t vm, const __bf16 *rs1,
+                                        size_t *new_vl, size_t vl) {
+  return __riscv_vle16ff(vm, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 4 x bfloat> @test_vle16ff_v_bf16m1_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, i64 } @llvm.riscv.vleff.mask.nxv4bf16.i64(<vscale x 4 x bfloat> poison, ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    store i64 [[TMP2]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret <vscale x 4 x bfloat> [[TMP1]]
+//
+vbfloat16m1_t test_vle16ff_v_bf16m1_m(vbool16_t vm, const __bf16 *rs1,
+                                      size_t *new_vl, size_t vl) {
+  return __riscv_vle16ff(vm, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 8 x bfloat> @test_vle16ff_v_bf16m2_m(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 8 x bfloat>, i64 } @llvm.riscv.vleff.mask.nxv8bf16.i64(<vscale x 8 x bfloat> poison, ptr [[RS1]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    store i64 [[TMP2]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret <vscale x 8 x bfloat> [[TMP1]]
+//
+vbfloat16m2_t test_vle16ff_v_bf16m2_m(vbool8_t vm, const __bf16 *rs1,
+                                      size_t *new_vl, size_t vl) {
+  return __riscv_vle16ff(vm, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 16 x bfloat> @test_vle16ff_v_bf16m4_m(
+// CHECK-RV64-SAME: <vscale x 16 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 16 x bfloat>, i64 } @llvm.riscv.vleff.mask.nxv16bf16.i64(<vscale x 16 x bfloat> poison, ptr [[RS1]], <vscale x 16 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 16 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 16 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    store i64 [[TMP2]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret <vscale x 16 x bfloat> [[TMP1]]
+//
+vbfloat16m4_t test_vle16ff_v_bf16m4_m(vbool4_t vm, const __bf16 *rs1,
+                                      size_t *new_vl, size_t vl) {
+  return __riscv_vle16ff(vm, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 32 x bfloat> @test_vle16ff_v_bf16m8_m(
+// CHECK-RV64-SAME: <vscale x 32 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 32 x bfloat>, i64 } @llvm.riscv.vleff.mask.nxv32bf16.i64(<vscale x 32 x bfloat> poison, ptr [[RS1]], <vscale x 32 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 32 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 32 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    store i64 [[TMP2]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret <vscale x 32 x bfloat> [[TMP1]]
+//
+vbfloat16m8_t test_vle16ff_v_bf16m8_m(vbool2_t vm, const __bf16 *rs1,
+                                      size_t *new_vl, size_t vl) {
+  return __riscv_vle16ff(vm, rs1, new_vl, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vlmul_ext_v.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vlmul_ext_v.c
new file mode 100644
index 000000000000..6a4ef411cfa5
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vlmul_ext_v.c
@@ -0,0 +1,159 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 2 x bfloat> @test_vlmul_ext_v_bf16mf4_bf16mf2(
+// CHECK-RV64-SAME: <vscale x 1 x bfloat> [[VALUE:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x bfloat> @llvm.vector.insert.nxv2bf16.nxv1bf16(<vscale x 2 x bfloat> poison, <vscale x 1 x bfloat> [[VALUE]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 2 x bfloat> [[TMP0]]
+//
+vbfloat16mf2_t test_vlmul_ext_v_bf16mf4_bf16mf2(vbfloat16mf4_t value) {
+  return __riscv_vlmul_ext_bf16mf2(value);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 4 x bfloat> @test_vlmul_ext_v_bf16mf4_bf16m1(
+// CHECK-RV64-SAME: <vscale x 1 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x bfloat> @llvm.vector.insert.nxv4bf16.nxv1bf16(<vscale x 4 x bfloat> poison, <vscale x 1 x bfloat> [[VALUE]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 4 x bfloat> [[TMP0]]
+//
+vbfloat16m1_t test_vlmul_ext_v_bf16mf4_bf16m1(vbfloat16mf4_t value) {
+  return __riscv_vlmul_ext_bf16m1(value);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 8 x bfloat> @test_vlmul_ext_v_bf16mf4_bf16m2(
+// CHECK-RV64-SAME: <vscale x 1 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x bfloat> @llvm.vector.insert.nxv8bf16.nxv1bf16(<vscale x 8 x bfloat> poison, <vscale x 1 x bfloat> [[VALUE]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 8 x bfloat> [[TMP0]]
+//
+vbfloat16m2_t test_vlmul_ext_v_bf16mf4_bf16m2(vbfloat16mf4_t value) {
+  return __riscv_vlmul_ext_bf16m2(value);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 16 x bfloat> @test_vlmul_ext_v_bf16mf4_bf16m4(
+// CHECK-RV64-SAME: <vscale x 1 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x bfloat> @llvm.vector.insert.nxv16bf16.nxv1bf16(<vscale x 16 x bfloat> poison, <vscale x 1 x bfloat> [[VALUE]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 16 x bfloat> [[TMP0]]
+//
+vbfloat16m4_t test_vlmul_ext_v_bf16mf4_bf16m4(vbfloat16mf4_t value) {
+  return __riscv_vlmul_ext_bf16m4(value);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 32 x bfloat> @test_vlmul_ext_v_bf16mf4_bf16m8(
+// CHECK-RV64-SAME: <vscale x 1 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x bfloat> @llvm.vector.insert.nxv32bf16.nxv1bf16(<vscale x 32 x bfloat> poison, <vscale x 1 x bfloat> [[VALUE]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 32 x bfloat> [[TMP0]]
+//
+vbfloat16m8_t test_vlmul_ext_v_bf16mf4_bf16m8(vbfloat16mf4_t value) {
+  return __riscv_vlmul_ext_bf16m8(value);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 4 x bfloat> @test_vlmul_ext_v_bf16mf2_bf16m1(
+// CHECK-RV64-SAME: <vscale x 2 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x bfloat> @llvm.vector.insert.nxv4bf16.nxv2bf16(<vscale x 4 x bfloat> poison, <vscale x 2 x bfloat> [[VALUE]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 4 x bfloat> [[TMP0]]
+//
+vbfloat16m1_t test_vlmul_ext_v_bf16mf2_bf16m1(vbfloat16mf2_t value) {
+  return __riscv_vlmul_ext_bf16m1(value);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 8 x bfloat> @test_vlmul_ext_v_bf16mf2_bf16m2(
+// CHECK-RV64-SAME: <vscale x 2 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x bfloat> @llvm.vector.insert.nxv8bf16.nxv2bf16(<vscale x 8 x bfloat> poison, <vscale x 2 x bfloat> [[VALUE]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 8 x bfloat> [[TMP0]]
+//
+vbfloat16m2_t test_vlmul_ext_v_bf16mf2_bf16m2(vbfloat16mf2_t value) {
+  return __riscv_vlmul_ext_bf16m2(value);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 16 x bfloat> @test_vlmul_ext_v_bf16mf2_bf16m4(
+// CHECK-RV64-SAME: <vscale x 2 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x bfloat> @llvm.vector.insert.nxv16bf16.nxv2bf16(<vscale x 16 x bfloat> poison, <vscale x 2 x bfloat> [[VALUE]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 16 x bfloat> [[TMP0]]
+//
+vbfloat16m4_t test_vlmul_ext_v_bf16mf2_bf16m4(vbfloat16mf2_t value) {
+  return __riscv_vlmul_ext_bf16m4(value);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 32 x bfloat> @test_vlmul_ext_v_bf16mf2_bf16m8(
+// CHECK-RV64-SAME: <vscale x 2 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x bfloat> @llvm.vector.insert.nxv32bf16.nxv2bf16(<vscale x 32 x bfloat> poison, <vscale x 2 x bfloat> [[VALUE]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 32 x bfloat> [[TMP0]]
+//
+vbfloat16m8_t test_vlmul_ext_v_bf16mf2_bf16m8(vbfloat16mf2_t value) {
+  return __riscv_vlmul_ext_bf16m8(value);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 8 x bfloat> @test_vlmul_ext_v_bf16m1_bf16m2(
+// CHECK-RV64-SAME: <vscale x 4 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x bfloat> @llvm.vector.insert.nxv8bf16.nxv4bf16(<vscale x 8 x bfloat> poison, <vscale x 4 x bfloat> [[VALUE]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 8 x bfloat> [[TMP0]]
+//
+vbfloat16m2_t test_vlmul_ext_v_bf16m1_bf16m2(vbfloat16m1_t value) {
+  return __riscv_vlmul_ext_bf16m2(value);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 16 x bfloat> @test_vlmul_ext_v_bf16m1_bf16m4(
+// CHECK-RV64-SAME: <vscale x 4 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x bfloat> @llvm.vector.insert.nxv16bf16.nxv4bf16(<vscale x 16 x bfloat> poison, <vscale x 4 x bfloat> [[VALUE]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 16 x bfloat> [[TMP0]]
+//
+vbfloat16m4_t test_vlmul_ext_v_bf16m1_bf16m4(vbfloat16m1_t value) {
+  return __riscv_vlmul_ext_bf16m4(value);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 32 x bfloat> @test_vlmul_ext_v_bf16m1_bf16m8(
+// CHECK-RV64-SAME: <vscale x 4 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x bfloat> @llvm.vector.insert.nxv32bf16.nxv4bf16(<vscale x 32 x bfloat> poison, <vscale x 4 x bfloat> [[VALUE]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 32 x bfloat> [[TMP0]]
+//
+vbfloat16m8_t test_vlmul_ext_v_bf16m1_bf16m8(vbfloat16m1_t value) {
+  return __riscv_vlmul_ext_bf16m8(value);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 16 x bfloat> @test_vlmul_ext_v_bf16m2_bf16m4(
+// CHECK-RV64-SAME: <vscale x 8 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x bfloat> @llvm.vector.insert.nxv16bf16.nxv8bf16(<vscale x 16 x bfloat> poison, <vscale x 8 x bfloat> [[VALUE]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 16 x bfloat> [[TMP0]]
+//
+vbfloat16m4_t test_vlmul_ext_v_bf16m2_bf16m4(vbfloat16m2_t value) {
+  return __riscv_vlmul_ext_bf16m4(value);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 32 x bfloat> @test_vlmul_ext_v_bf16m2_bf16m8(
+// CHECK-RV64-SAME: <vscale x 8 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x bfloat> @llvm.vector.insert.nxv32bf16.nxv8bf16(<vscale x 32 x bfloat> poison, <vscale x 8 x bfloat> [[VALUE]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 32 x bfloat> [[TMP0]]
+//
+vbfloat16m8_t test_vlmul_ext_v_bf16m2_bf16m8(vbfloat16m2_t value) {
+  return __riscv_vlmul_ext_bf16m8(value);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 32 x bfloat> @test_vlmul_ext_v_bf16m4_bf16m8(
+// CHECK-RV64-SAME: <vscale x 16 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x bfloat> @llvm.vector.insert.nxv32bf16.nxv16bf16(<vscale x 32 x bfloat> poison, <vscale x 16 x bfloat> [[VALUE]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 32 x bfloat> [[TMP0]]
+//
+vbfloat16m8_t test_vlmul_ext_v_bf16m4_bf16m8(vbfloat16m4_t value) {
+  return __riscv_vlmul_ext_bf16m8(value);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vlmul_trunc_v.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vlmul_trunc_v.c
new file mode 100644
index 000000000000..04ca0ac94259
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vlmul_trunc_v.c
@@ -0,0 +1,159 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 1 x bfloat> @test_vlmul_trunc_v_bf16mf2_bf16mf4(
+// CHECK-RV64-SAME: <vscale x 2 x bfloat> [[VALUE:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x bfloat> @llvm.vector.extract.nxv1bf16.nxv2bf16(<vscale x 2 x bfloat> [[VALUE]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 1 x bfloat> [[TMP0]]
+//
+vbfloat16mf4_t test_vlmul_trunc_v_bf16mf2_bf16mf4(vbfloat16mf2_t value) {
+  return __riscv_vlmul_trunc_bf16mf4(value);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 1 x bfloat> @test_vlmul_trunc_v_bf16m1_bf16mf4(
+// CHECK-RV64-SAME: <vscale x 4 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x bfloat> @llvm.vector.extract.nxv1bf16.nxv4bf16(<vscale x 4 x bfloat> [[VALUE]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 1 x bfloat> [[TMP0]]
+//
+vbfloat16mf4_t test_vlmul_trunc_v_bf16m1_bf16mf4(vbfloat16m1_t value) {
+  return __riscv_vlmul_trunc_bf16mf4(value);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 2 x bfloat> @test_vlmul_trunc_v_bf16m1_bf16mf2(
+// CHECK-RV64-SAME: <vscale x 4 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x bfloat> @llvm.vector.extract.nxv2bf16.nxv4bf16(<vscale x 4 x bfloat> [[VALUE]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 2 x bfloat> [[TMP0]]
+//
+vbfloat16mf2_t test_vlmul_trunc_v_bf16m1_bf16mf2(vbfloat16m1_t value) {
+  return __riscv_vlmul_trunc_bf16mf2(value);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 1 x bfloat> @test_vlmul_trunc_v_bf16m2_bf16mf4(
+// CHECK-RV64-SAME: <vscale x 8 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x bfloat> @llvm.vector.extract.nxv1bf16.nxv8bf16(<vscale x 8 x bfloat> [[VALUE]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 1 x bfloat> [[TMP0]]
+//
+vbfloat16mf4_t test_vlmul_trunc_v_bf16m2_bf16mf4(vbfloat16m2_t value) {
+  return __riscv_vlmul_trunc_bf16mf4(value);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 2 x bfloat> @test_vlmul_trunc_v_bf16m2_bf16mf2(
+// CHECK-RV64-SAME: <vscale x 8 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x bfloat> @llvm.vector.extract.nxv2bf16.nxv8bf16(<vscale x 8 x bfloat> [[VALUE]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 2 x bfloat> [[TMP0]]
+//
+vbfloat16mf2_t test_vlmul_trunc_v_bf16m2_bf16mf2(vbfloat16m2_t value) {
+  return __riscv_vlmul_trunc_bf16mf2(value);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 4 x bfloat> @test_vlmul_trunc_v_bf16m2_bf16m1(
+// CHECK-RV64-SAME: <vscale x 8 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x bfloat> @llvm.vector.extract.nxv4bf16.nxv8bf16(<vscale x 8 x bfloat> [[VALUE]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 4 x bfloat> [[TMP0]]
+//
+vbfloat16m1_t test_vlmul_trunc_v_bf16m2_bf16m1(vbfloat16m2_t value) {
+  return __riscv_vlmul_trunc_bf16m1(value);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 1 x bfloat> @test_vlmul_trunc_v_bf16m4_bf16mf4(
+// CHECK-RV64-SAME: <vscale x 16 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x bfloat> @llvm.vector.extract.nxv1bf16.nxv16bf16(<vscale x 16 x bfloat> [[VALUE]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 1 x bfloat> [[TMP0]]
+//
+vbfloat16mf4_t test_vlmul_trunc_v_bf16m4_bf16mf4(vbfloat16m4_t value) {
+  return __riscv_vlmul_trunc_bf16mf4(value);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 2 x bfloat> @test_vlmul_trunc_v_bf16m4_bf16mf2(
+// CHECK-RV64-SAME: <vscale x 16 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x bfloat> @llvm.vector.extract.nxv2bf16.nxv16bf16(<vscale x 16 x bfloat> [[VALUE]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 2 x bfloat> [[TMP0]]
+//
+vbfloat16mf2_t test_vlmul_trunc_v_bf16m4_bf16mf2(vbfloat16m4_t value) {
+  return __riscv_vlmul_trunc_bf16mf2(value);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 4 x bfloat> @test_vlmul_trunc_v_bf16m4_bf16m1(
+// CHECK-RV64-SAME: <vscale x 16 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x bfloat> @llvm.vector.extract.nxv4bf16.nxv16bf16(<vscale x 16 x bfloat> [[VALUE]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 4 x bfloat> [[TMP0]]
+//
+vbfloat16m1_t test_vlmul_trunc_v_bf16m4_bf16m1(vbfloat16m4_t value) {
+  return __riscv_vlmul_trunc_bf16m1(value);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 8 x bfloat> @test_vlmul_trunc_v_bf16m4_bf16m2(
+// CHECK-RV64-SAME: <vscale x 16 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x bfloat> @llvm.vector.extract.nxv8bf16.nxv16bf16(<vscale x 16 x bfloat> [[VALUE]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 8 x bfloat> [[TMP0]]
+//
+vbfloat16m2_t test_vlmul_trunc_v_bf16m4_bf16m2(vbfloat16m4_t value) {
+  return __riscv_vlmul_trunc_bf16m2(value);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 1 x bfloat> @test_vlmul_trunc_v_bf16m8_bf16mf4(
+// CHECK-RV64-SAME: <vscale x 32 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x bfloat> @llvm.vector.extract.nxv1bf16.nxv32bf16(<vscale x 32 x bfloat> [[VALUE]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 1 x bfloat> [[TMP0]]
+//
+vbfloat16mf4_t test_vlmul_trunc_v_bf16m8_bf16mf4(vbfloat16m8_t value) {
+  return __riscv_vlmul_trunc_bf16mf4(value);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 2 x bfloat> @test_vlmul_trunc_v_bf16m8_bf16mf2(
+// CHECK-RV64-SAME: <vscale x 32 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x bfloat> @llvm.vector.extract.nxv2bf16.nxv32bf16(<vscale x 32 x bfloat> [[VALUE]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 2 x bfloat> [[TMP0]]
+//
+vbfloat16mf2_t test_vlmul_trunc_v_bf16m8_bf16mf2(vbfloat16m8_t value) {
+  return __riscv_vlmul_trunc_bf16mf2(value);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 4 x bfloat> @test_vlmul_trunc_v_bf16m8_bf16m1(
+// CHECK-RV64-SAME: <vscale x 32 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x bfloat> @llvm.vector.extract.nxv4bf16.nxv32bf16(<vscale x 32 x bfloat> [[VALUE]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 4 x bfloat> [[TMP0]]
+//
+vbfloat16m1_t test_vlmul_trunc_v_bf16m8_bf16m1(vbfloat16m8_t value) {
+  return __riscv_vlmul_trunc_bf16m1(value);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 8 x bfloat> @test_vlmul_trunc_v_bf16m8_bf16m2(
+// CHECK-RV64-SAME: <vscale x 32 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x bfloat> @llvm.vector.extract.nxv8bf16.nxv32bf16(<vscale x 32 x bfloat> [[VALUE]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 8 x bfloat> [[TMP0]]
+//
+vbfloat16m2_t test_vlmul_trunc_v_bf16m8_bf16m2(vbfloat16m8_t value) {
+  return __riscv_vlmul_trunc_bf16m2(value);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 16 x bfloat> @test_vlmul_trunc_v_bf16m8_bf16m4(
+// CHECK-RV64-SAME: <vscale x 32 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x bfloat> @llvm.vector.extract.nxv16bf16.nxv32bf16(<vscale x 32 x bfloat> [[VALUE]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 16 x bfloat> [[TMP0]]
+//
+vbfloat16m4_t test_vlmul_trunc_v_bf16m8_bf16m4(vbfloat16m8_t value) {
+  return __riscv_vlmul_trunc_bf16m4(value);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vloxei16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vloxei16.c
new file mode 100644
index 000000000000..f57636a7c74c
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vloxei16.c
@@ -0,0 +1,141 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 1 x bfloat> @test_vloxei16_v_bf16mf4(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x bfloat> @llvm.riscv.vloxei.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> poison, ptr [[RS1]], <vscale x 1 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret <vscale x 1 x bfloat> [[TMP0]]
+//
+vbfloat16mf4_t test_vloxei16_v_bf16mf4(const __bf16 *rs1, vuint16mf4_t rs2,
+                                       size_t vl) {
+  return __riscv_vloxei16(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 2 x bfloat> @test_vloxei16_v_bf16mf2(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x bfloat> @llvm.riscv.vloxei.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> poison, ptr [[RS1]], <vscale x 2 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret <vscale x 2 x bfloat> [[TMP0]]
+//
+vbfloat16mf2_t test_vloxei16_v_bf16mf2(const __bf16 *rs1, vuint16mf2_t rs2,
+                                       size_t vl) {
+  return __riscv_vloxei16(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 4 x bfloat> @test_vloxei16_v_bf16m1(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x bfloat> @llvm.riscv.vloxei.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> poison, ptr [[RS1]], <vscale x 4 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret <vscale x 4 x bfloat> [[TMP0]]
+//
+vbfloat16m1_t test_vloxei16_v_bf16m1(const __bf16 *rs1, vuint16m1_t rs2,
+                                     size_t vl) {
+  return __riscv_vloxei16(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 8 x bfloat> @test_vloxei16_v_bf16m2(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x bfloat> @llvm.riscv.vloxei.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> poison, ptr [[RS1]], <vscale x 8 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret <vscale x 8 x bfloat> [[TMP0]]
+//
+vbfloat16m2_t test_vloxei16_v_bf16m2(const __bf16 *rs1, vuint16m2_t rs2,
+                                     size_t vl) {
+  return __riscv_vloxei16(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 16 x bfloat> @test_vloxei16_v_bf16m4(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 16 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x bfloat> @llvm.riscv.vloxei.nxv16bf16.nxv16i16.i64(<vscale x 16 x bfloat> poison, ptr [[RS1]], <vscale x 16 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret <vscale x 16 x bfloat> [[TMP0]]
+//
+vbfloat16m4_t test_vloxei16_v_bf16m4(const __bf16 *rs1, vuint16m4_t rs2,
+                                     size_t vl) {
+  return __riscv_vloxei16(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 32 x bfloat> @test_vloxei16_v_bf16m8(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 32 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x bfloat> @llvm.riscv.vloxei.nxv32bf16.nxv32i16.i64(<vscale x 32 x bfloat> poison, ptr [[RS1]], <vscale x 32 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret <vscale x 32 x bfloat> [[TMP0]]
+//
+vbfloat16m8_t test_vloxei16_v_bf16m8(const __bf16 *rs1, vuint16m8_t rs2,
+                                     size_t vl) {
+  return __riscv_vloxei16(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 1 x bfloat> @test_vloxei16_v_bf16mf4_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x bfloat> @llvm.riscv.vloxei.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> poison, ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret <vscale x 1 x bfloat> [[TMP0]]
+//
+vbfloat16mf4_t test_vloxei16_v_bf16mf4_m(vbool64_t vm, const __bf16 *rs1,
+                                         vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vloxei16(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 2 x bfloat> @test_vloxei16_v_bf16mf2_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x bfloat> @llvm.riscv.vloxei.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> poison, ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret <vscale x 2 x bfloat> [[TMP0]]
+//
+vbfloat16mf2_t test_vloxei16_v_bf16mf2_m(vbool32_t vm, const __bf16 *rs1,
+                                         vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vloxei16(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 4 x bfloat> @test_vloxei16_v_bf16m1_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x bfloat> @llvm.riscv.vloxei.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> poison, ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret <vscale x 4 x bfloat> [[TMP0]]
+//
+vbfloat16m1_t test_vloxei16_v_bf16m1_m(vbool16_t vm, const __bf16 *rs1,
+                                       vuint16m1_t rs2, size_t vl) {
+  return __riscv_vloxei16(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 8 x bfloat> @test_vloxei16_v_bf16m2_m(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x bfloat> @llvm.riscv.vloxei.mask.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> poison, ptr [[RS1]], <vscale x 8 x i16> [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret <vscale x 8 x bfloat> [[TMP0]]
+//
+vbfloat16m2_t test_vloxei16_v_bf16m2_m(vbool8_t vm, const __bf16 *rs1,
+                                       vuint16m2_t rs2, size_t vl) {
+  return __riscv_vloxei16(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 16 x bfloat> @test_vloxei16_v_bf16m4_m(
+// CHECK-RV64-SAME: <vscale x 16 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 16 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x bfloat> @llvm.riscv.vloxei.mask.nxv16bf16.nxv16i16.i64(<vscale x 16 x bfloat> poison, ptr [[RS1]], <vscale x 16 x i16> [[RS2]], <vscale x 16 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret <vscale x 16 x bfloat> [[TMP0]]
+//
+vbfloat16m4_t test_vloxei16_v_bf16m4_m(vbool4_t vm, const __bf16 *rs1,
+                                       vuint16m4_t rs2, size_t vl) {
+  return __riscv_vloxei16(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 32 x bfloat> @test_vloxei16_v_bf16m8_m(
+// CHECK-RV64-SAME: <vscale x 32 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 32 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x bfloat> @llvm.riscv.vloxei.mask.nxv32bf16.nxv32i16.i64(<vscale x 32 x bfloat> poison, ptr [[RS1]], <vscale x 32 x i16> [[RS2]], <vscale x 32 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret <vscale x 32 x bfloat> [[TMP0]]
+//
+vbfloat16m8_t test_vloxei16_v_bf16m8_m(vbool2_t vm, const __bf16 *rs1,
+                                       vuint16m8_t rs2, size_t vl) {
+  return __riscv_vloxei16(vm, rs1, rs2, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vloxseg2ei16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vloxseg2ei16.c
new file mode 100644
index 000000000000..e43a37959da5
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vloxseg2ei16.c
@@ -0,0 +1,121 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vloxseg2ei16_v_bf16mf4x2(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vloxseg2.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], <vscale x 1 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x2_t test_vloxseg2ei16_v_bf16mf4x2(const __bf16 *rs1,
+                                               vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vloxseg2ei16(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vloxseg2ei16_v_bf16mf2x2(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vloxseg2.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], <vscale x 2 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x2_t test_vloxseg2ei16_v_bf16mf2x2(const __bf16 *rs1,
+                                               vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vloxseg2ei16(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vloxseg2ei16_v_bf16m1x2(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vloxseg2.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], <vscale x 4 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x2_t test_vloxseg2ei16_v_bf16m1x2(const __bf16 *rs1, vuint16m1_t rs2,
+                                             size_t vl) {
+  return __riscv_vloxseg2ei16(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vloxseg2ei16_v_bf16m2x2(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vloxseg2.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> poison, <vscale x 8 x bfloat> poison, ptr [[RS1]], <vscale x 8 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP0]]
+//
+vbfloat16m2x2_t test_vloxseg2ei16_v_bf16m2x2(const __bf16 *rs1, vuint16m2_t rs2,
+                                             size_t vl) {
+  return __riscv_vloxseg2ei16(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @test_vloxseg2ei16_v_bf16m4x2(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 16 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @llvm.riscv.vloxseg2.nxv16bf16.nxv16i16.i64(<vscale x 16 x bfloat> poison, <vscale x 16 x bfloat> poison, ptr [[RS1]], <vscale x 16 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[TMP0]]
+//
+vbfloat16m4x2_t test_vloxseg2ei16_v_bf16m4x2(const __bf16 *rs1, vuint16m4_t rs2,
+                                             size_t vl) {
+  return __riscv_vloxseg2ei16(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vloxseg2ei16_v_bf16mf4x2_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vloxseg2.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x2_t test_vloxseg2ei16_v_bf16mf4x2_m(vbool64_t vm,
+                                                 const __bf16 *rs1,
+                                                 vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vloxseg2ei16(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vloxseg2ei16_v_bf16mf2x2_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vloxseg2.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x2_t test_vloxseg2ei16_v_bf16mf2x2_m(vbool32_t vm,
+                                                 const __bf16 *rs1,
+                                                 vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vloxseg2ei16(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vloxseg2ei16_v_bf16m1x2_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vloxseg2.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x2_t test_vloxseg2ei16_v_bf16m1x2_m(vbool16_t vm, const __bf16 *rs1,
+                                               vuint16m1_t rs2, size_t vl) {
+  return __riscv_vloxseg2ei16(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vloxseg2ei16_v_bf16m2x2_m(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vloxseg2.mask.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> poison, <vscale x 8 x bfloat> poison, ptr [[RS1]], <vscale x 8 x i16> [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP0]]
+//
+vbfloat16m2x2_t test_vloxseg2ei16_v_bf16m2x2_m(vbool8_t vm, const __bf16 *rs1,
+                                               vuint16m2_t rs2, size_t vl) {
+  return __riscv_vloxseg2ei16(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @test_vloxseg2ei16_v_bf16m4x2_m(
+// CHECK-RV64-SAME: <vscale x 16 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 16 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @llvm.riscv.vloxseg2.mask.nxv16bf16.nxv16i16.i64(<vscale x 16 x bfloat> poison, <vscale x 16 x bfloat> poison, ptr [[RS1]], <vscale x 16 x i16> [[RS2]], <vscale x 16 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[TMP0]]
+//
+vbfloat16m4x2_t test_vloxseg2ei16_v_bf16m4x2_m(vbool4_t vm, const __bf16 *rs1,
+                                               vuint16m4_t rs2, size_t vl) {
+  return __riscv_vloxseg2ei16(vm, rs1, rs2, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vloxseg3ei16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vloxseg3ei16.c
new file mode 100644
index 000000000000..52950e897a2f
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vloxseg3ei16.c
@@ -0,0 +1,99 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vloxseg3ei16_v_bf16mf4x3(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vloxseg3.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], <vscale x 1 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x3_t test_vloxseg3ei16_v_bf16mf4x3(const __bf16 *rs1,
+                                               vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vloxseg3ei16(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vloxseg3ei16_v_bf16mf2x3(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vloxseg3.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], <vscale x 2 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x3_t test_vloxseg3ei16_v_bf16mf2x3(const __bf16 *rs1,
+                                               vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vloxseg3ei16(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vloxseg3ei16_v_bf16m1x3(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vloxseg3.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], <vscale x 4 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x3_t test_vloxseg3ei16_v_bf16m1x3(const __bf16 *rs1, vuint16m1_t rs2,
+                                             size_t vl) {
+  return __riscv_vloxseg3ei16(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vloxseg3ei16_v_bf16m2x3(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vloxseg3.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> poison, <vscale x 8 x bfloat> poison, <vscale x 8 x bfloat> poison, ptr [[RS1]], <vscale x 8 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP0]]
+//
+vbfloat16m2x3_t test_vloxseg3ei16_v_bf16m2x3(const __bf16 *rs1, vuint16m2_t rs2,
+                                             size_t vl) {
+  return __riscv_vloxseg3ei16(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vloxseg3ei16_v_bf16mf4x3_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vloxseg3.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x3_t test_vloxseg3ei16_v_bf16mf4x3_m(vbool64_t vm,
+                                                 const __bf16 *rs1,
+                                                 vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vloxseg3ei16(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vloxseg3ei16_v_bf16mf2x3_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vloxseg3.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x3_t test_vloxseg3ei16_v_bf16mf2x3_m(vbool32_t vm,
+                                                 const __bf16 *rs1,
+                                                 vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vloxseg3ei16(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vloxseg3ei16_v_bf16m1x3_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vloxseg3.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x3_t test_vloxseg3ei16_v_bf16m1x3_m(vbool16_t vm, const __bf16 *rs1,
+                                               vuint16m1_t rs2, size_t vl) {
+  return __riscv_vloxseg3ei16(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vloxseg3ei16_v_bf16m2x3_m(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vloxseg3.mask.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> poison, <vscale x 8 x bfloat> poison, <vscale x 8 x bfloat> poison, ptr [[RS1]], <vscale x 8 x i16> [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP0]]
+//
+vbfloat16m2x3_t test_vloxseg3ei16_v_bf16m2x3_m(vbool8_t vm, const __bf16 *rs1,
+                                               vuint16m2_t rs2, size_t vl) {
+  return __riscv_vloxseg3ei16(vm, rs1, rs2, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vloxseg4ei16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vloxseg4ei16.c
new file mode 100644
index 000000000000..3351c42c25c1
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vloxseg4ei16.c
@@ -0,0 +1,99 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vloxseg4ei16_v_bf16mf4x4(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vloxseg4.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], <vscale x 1 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x4_t test_vloxseg4ei16_v_bf16mf4x4(const __bf16 *rs1,
+                                               vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vloxseg4ei16(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vloxseg4ei16_v_bf16mf2x4(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vloxseg4.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], <vscale x 2 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x4_t test_vloxseg4ei16_v_bf16mf2x4(const __bf16 *rs1,
+                                               vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vloxseg4ei16(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vloxseg4ei16_v_bf16m1x4(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vloxseg4.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], <vscale x 4 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x4_t test_vloxseg4ei16_v_bf16m1x4(const __bf16 *rs1, vuint16m1_t rs2,
+                                             size_t vl) {
+  return __riscv_vloxseg4ei16(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vloxseg4ei16_v_bf16m2x4(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vloxseg4.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> poison, <vscale x 8 x bfloat> poison, <vscale x 8 x bfloat> poison, <vscale x 8 x bfloat> poison, ptr [[RS1]], <vscale x 8 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP0]]
+//
+vbfloat16m2x4_t test_vloxseg4ei16_v_bf16m2x4(const __bf16 *rs1, vuint16m2_t rs2,
+                                             size_t vl) {
+  return __riscv_vloxseg4ei16(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vloxseg4ei16_v_bf16mf4x4_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vloxseg4.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x4_t test_vloxseg4ei16_v_bf16mf4x4_m(vbool64_t vm,
+                                                 const __bf16 *rs1,
+                                                 vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vloxseg4ei16(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vloxseg4ei16_v_bf16mf2x4_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vloxseg4.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x4_t test_vloxseg4ei16_v_bf16mf2x4_m(vbool32_t vm,
+                                                 const __bf16 *rs1,
+                                                 vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vloxseg4ei16(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vloxseg4ei16_v_bf16m1x4_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vloxseg4.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x4_t test_vloxseg4ei16_v_bf16m1x4_m(vbool16_t vm, const __bf16 *rs1,
+                                               vuint16m1_t rs2, size_t vl) {
+  return __riscv_vloxseg4ei16(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vloxseg4ei16_v_bf16m2x4_m(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vloxseg4.mask.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> poison, <vscale x 8 x bfloat> poison, <vscale x 8 x bfloat> poison, <vscale x 8 x bfloat> poison, ptr [[RS1]], <vscale x 8 x i16> [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP0]]
+//
+vbfloat16m2x4_t test_vloxseg4ei16_v_bf16m2x4_m(vbool8_t vm, const __bf16 *rs1,
+                                               vuint16m2_t rs2, size_t vl) {
+  return __riscv_vloxseg4ei16(vm, rs1, rs2, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vloxseg5ei16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vloxseg5ei16.c
new file mode 100644
index 000000000000..4651c0ecad85
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vloxseg5ei16.c
@@ -0,0 +1,77 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vloxseg5ei16_v_bf16mf4x5(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vloxseg5.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], <vscale x 1 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x5_t test_vloxseg5ei16_v_bf16mf4x5(const __bf16 *rs1,
+                                               vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vloxseg5ei16(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vloxseg5ei16_v_bf16mf2x5(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vloxseg5.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], <vscale x 2 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x5_t test_vloxseg5ei16_v_bf16mf2x5(const __bf16 *rs1,
+                                               vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vloxseg5ei16(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vloxseg5ei16_v_bf16m1x5(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vloxseg5.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], <vscale x 4 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x5_t test_vloxseg5ei16_v_bf16m1x5(const __bf16 *rs1, vuint16m1_t rs2,
+                                             size_t vl) {
+  return __riscv_vloxseg5ei16(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vloxseg5ei16_v_bf16mf4x5_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vloxseg5.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x5_t test_vloxseg5ei16_v_bf16mf4x5_m(vbool64_t vm,
+                                                 const __bf16 *rs1,
+                                                 vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vloxseg5ei16(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vloxseg5ei16_v_bf16mf2x5_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vloxseg5.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x5_t test_vloxseg5ei16_v_bf16mf2x5_m(vbool32_t vm,
+                                                 const __bf16 *rs1,
+                                                 vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vloxseg5ei16(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vloxseg5ei16_v_bf16m1x5_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vloxseg5.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x5_t test_vloxseg5ei16_v_bf16m1x5_m(vbool16_t vm, const __bf16 *rs1,
+                                               vuint16m1_t rs2, size_t vl) {
+  return __riscv_vloxseg5ei16(vm, rs1, rs2, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vloxseg6ei16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vloxseg6ei16.c
new file mode 100644
index 000000000000..c70b924c5cab
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vloxseg6ei16.c
@@ -0,0 +1,77 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vloxseg6ei16_v_bf16mf4x6(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vloxseg6.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], <vscale x 1 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x6_t test_vloxseg6ei16_v_bf16mf4x6(const __bf16 *rs1,
+                                               vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vloxseg6ei16(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vloxseg6ei16_v_bf16mf2x6(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vloxseg6.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], <vscale x 2 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x6_t test_vloxseg6ei16_v_bf16mf2x6(const __bf16 *rs1,
+                                               vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vloxseg6ei16(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vloxseg6ei16_v_bf16m1x6(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vloxseg6.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], <vscale x 4 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x6_t test_vloxseg6ei16_v_bf16m1x6(const __bf16 *rs1, vuint16m1_t rs2,
+                                             size_t vl) {
+  return __riscv_vloxseg6ei16(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vloxseg6ei16_v_bf16mf4x6_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vloxseg6.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x6_t test_vloxseg6ei16_v_bf16mf4x6_m(vbool64_t vm,
+                                                 const __bf16 *rs1,
+                                                 vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vloxseg6ei16(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vloxseg6ei16_v_bf16mf2x6_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vloxseg6.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x6_t test_vloxseg6ei16_v_bf16mf2x6_m(vbool32_t vm,
+                                                 const __bf16 *rs1,
+                                                 vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vloxseg6ei16(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vloxseg6ei16_v_bf16m1x6_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vloxseg6.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x6_t test_vloxseg6ei16_v_bf16m1x6_m(vbool16_t vm, const __bf16 *rs1,
+                                               vuint16m1_t rs2, size_t vl) {
+  return __riscv_vloxseg6ei16(vm, rs1, rs2, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vloxseg7ei16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vloxseg7ei16.c
new file mode 100644
index 000000000000..985b0053362f
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vloxseg7ei16.c
@@ -0,0 +1,77 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vloxseg7ei16_v_bf16mf4x7(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vloxseg7.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], <vscale x 1 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x7_t test_vloxseg7ei16_v_bf16mf4x7(const __bf16 *rs1,
+                                               vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vloxseg7ei16(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vloxseg7ei16_v_bf16mf2x7(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vloxseg7.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], <vscale x 2 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x7_t test_vloxseg7ei16_v_bf16mf2x7(const __bf16 *rs1,
+                                               vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vloxseg7ei16(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vloxseg7ei16_v_bf16m1x7(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vloxseg7.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], <vscale x 4 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x7_t test_vloxseg7ei16_v_bf16m1x7(const __bf16 *rs1, vuint16m1_t rs2,
+                                             size_t vl) {
+  return __riscv_vloxseg7ei16(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vloxseg7ei16_v_bf16mf4x7_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vloxseg7.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x7_t test_vloxseg7ei16_v_bf16mf4x7_m(vbool64_t vm,
+                                                 const __bf16 *rs1,
+                                                 vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vloxseg7ei16(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vloxseg7ei16_v_bf16mf2x7_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vloxseg7.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x7_t test_vloxseg7ei16_v_bf16mf2x7_m(vbool32_t vm,
+                                                 const __bf16 *rs1,
+                                                 vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vloxseg7ei16(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vloxseg7ei16_v_bf16m1x7_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vloxseg7.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x7_t test_vloxseg7ei16_v_bf16m1x7_m(vbool16_t vm, const __bf16 *rs1,
+                                               vuint16m1_t rs2, size_t vl) {
+  return __riscv_vloxseg7ei16(vm, rs1, rs2, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vloxseg8ei16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vloxseg8ei16.c
new file mode 100644
index 000000000000..413773b1657a
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vloxseg8ei16.c
@@ -0,0 +1,77 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vloxseg8ei16_v_bf16mf4x8(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vloxseg8.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], <vscale x 1 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x8_t test_vloxseg8ei16_v_bf16mf4x8(const __bf16 *rs1,
+                                               vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vloxseg8ei16(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vloxseg8ei16_v_bf16mf2x8(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vloxseg8.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], <vscale x 2 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x8_t test_vloxseg8ei16_v_bf16mf2x8(const __bf16 *rs1,
+                                               vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vloxseg8ei16(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vloxseg8ei16_v_bf16m1x8(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vloxseg8.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], <vscale x 4 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x8_t test_vloxseg8ei16_v_bf16m1x8(const __bf16 *rs1, vuint16m1_t rs2,
+                                             size_t vl) {
+  return __riscv_vloxseg8ei16(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vloxseg8ei16_v_bf16mf4x8_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vloxseg8.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x8_t test_vloxseg8ei16_v_bf16mf4x8_m(vbool64_t vm,
+                                                 const __bf16 *rs1,
+                                                 vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vloxseg8ei16(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vloxseg8ei16_v_bf16mf2x8_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vloxseg8.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x8_t test_vloxseg8ei16_v_bf16mf2x8_m(vbool32_t vm,
+                                                 const __bf16 *rs1,
+                                                 vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vloxseg8ei16(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vloxseg8ei16_v_bf16m1x8_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vloxseg8.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x8_t test_vloxseg8ei16_v_bf16m1x8_m(vbool16_t vm, const __bf16 *rs1,
+                                               vuint16m1_t rs2, size_t vl) {
+  return __riscv_vloxseg8ei16(vm, rs1, rs2, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vlse16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vlse16.c
new file mode 100644
index 000000000000..300100a94371
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vlse16.c
@@ -0,0 +1,75 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 1 x bfloat> @test_vlse16_v_bf16mf4_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x bfloat> @llvm.riscv.vlse.mask.nxv1bf16.i64(<vscale x 1 x bfloat> poison, ptr [[RS1]], i64 [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret <vscale x 1 x bfloat> [[TMP0]]
+//
+vbfloat16mf4_t test_vlse16_v_bf16mf4_m(vbool64_t vm, const __bf16 *rs1,
+                                       ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlse16(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 2 x bfloat> @test_vlse16_v_bf16mf2_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x bfloat> @llvm.riscv.vlse.mask.nxv2bf16.i64(<vscale x 2 x bfloat> poison, ptr [[RS1]], i64 [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret <vscale x 2 x bfloat> [[TMP0]]
+//
+vbfloat16mf2_t test_vlse16_v_bf16mf2_m(vbool32_t vm, const __bf16 *rs1,
+                                       ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlse16(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 4 x bfloat> @test_vlse16_v_bf16m1_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x bfloat> @llvm.riscv.vlse.mask.nxv4bf16.i64(<vscale x 4 x bfloat> poison, ptr [[RS1]], i64 [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret <vscale x 4 x bfloat> [[TMP0]]
+//
+vbfloat16m1_t test_vlse16_v_bf16m1_m(vbool16_t vm, const __bf16 *rs1,
+                                     ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlse16(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 8 x bfloat> @test_vlse16_v_bf16m2_m(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x bfloat> @llvm.riscv.vlse.mask.nxv8bf16.i64(<vscale x 8 x bfloat> poison, ptr [[RS1]], i64 [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret <vscale x 8 x bfloat> [[TMP0]]
+//
+vbfloat16m2_t test_vlse16_v_bf16m2_m(vbool8_t vm, const __bf16 *rs1,
+                                     ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlse16(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 16 x bfloat> @test_vlse16_v_bf16m4_m(
+// CHECK-RV64-SAME: <vscale x 16 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x bfloat> @llvm.riscv.vlse.mask.nxv16bf16.i64(<vscale x 16 x bfloat> poison, ptr [[RS1]], i64 [[RS2]], <vscale x 16 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret <vscale x 16 x bfloat> [[TMP0]]
+//
+vbfloat16m4_t test_vlse16_v_bf16m4_m(vbool4_t vm, const __bf16 *rs1,
+                                     ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlse16(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 32 x bfloat> @test_vlse16_v_bf16m8_m(
+// CHECK-RV64-SAME: <vscale x 32 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x bfloat> @llvm.riscv.vlse.mask.nxv32bf16.i64(<vscale x 32 x bfloat> poison, ptr [[RS1]], i64 [[RS2]], <vscale x 32 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret <vscale x 32 x bfloat> [[TMP0]]
+//
+vbfloat16m8_t test_vlse16_v_bf16m8_m(vbool2_t vm, const __bf16 *rs1,
+                                     ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlse16(vm, rs1, rs2, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vlseg2e16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vlseg2e16.c
new file mode 100644
index 000000000000..2304cc3a7cdd
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vlseg2e16.c
@@ -0,0 +1,64 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg2e16_v_bf16mf4x2_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlseg2.mask.nxv1bf16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x2_t test_vlseg2e16_v_bf16mf4x2_m(vbool64_t vm, const __bf16 *rs1,
+                                              size_t vl) {
+  return __riscv_vlseg2e16(vm, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg2e16_v_bf16mf2x2_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlseg2.mask.nxv2bf16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x2_t test_vlseg2e16_v_bf16mf2x2_m(vbool32_t vm, const __bf16 *rs1,
+                                              size_t vl) {
+  return __riscv_vlseg2e16(vm, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg2e16_v_bf16m1x2_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlseg2.mask.nxv4bf16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x2_t test_vlseg2e16_v_bf16m1x2_m(vbool16_t vm, const __bf16 *rs1,
+                                            size_t vl) {
+  return __riscv_vlseg2e16(vm, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vlseg2e16_v_bf16m2x2_m(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vlseg2.mask.nxv8bf16.i64(<vscale x 8 x bfloat> poison, <vscale x 8 x bfloat> poison, ptr [[RS1]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP0]]
+//
+vbfloat16m2x2_t test_vlseg2e16_v_bf16m2x2_m(vbool8_t vm, const __bf16 *rs1,
+                                            size_t vl) {
+  return __riscv_vlseg2e16(vm, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @test_vlseg2e16_v_bf16m4x2_m(
+// CHECK-RV64-SAME: <vscale x 16 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @llvm.riscv.vlseg2.mask.nxv16bf16.i64(<vscale x 16 x bfloat> poison, <vscale x 16 x bfloat> poison, ptr [[RS1]], <vscale x 16 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[TMP0]]
+//
+vbfloat16m4x2_t test_vlseg2e16_v_bf16m4x2_m(vbool4_t vm, const __bf16 *rs1,
+                                            size_t vl) {
+  return __riscv_vlseg2e16(vm, rs1, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vlseg2e16ff.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vlseg2e16ff.c
new file mode 100644
index 000000000000..f645d5872c26
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vlseg2e16ff.c
@@ -0,0 +1,94 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg2e16ff_v_bf16mf4x2_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } @llvm.riscv.vlseg2ff.mask.nxv1bf16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } poison, <vscale x 1 x bfloat> [[TMP1]], 0
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 2
+// CHECK-RV64-NEXT:    store i64 [[TMP5]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP4]]
+//
+vbfloat16mf4x2_t test_vlseg2e16ff_v_bf16mf4x2_m(vbool64_t vm, const __bf16 *rs1,
+                                                size_t *new_vl, size_t vl) {
+  return __riscv_vlseg2e16ff(vm, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg2e16ff_v_bf16mf2x2_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } @llvm.riscv.vlseg2ff.mask.nxv2bf16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } poison, <vscale x 2 x bfloat> [[TMP1]], 0
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 2
+// CHECK-RV64-NEXT:    store i64 [[TMP5]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP4]]
+//
+vbfloat16mf2x2_t test_vlseg2e16ff_v_bf16mf2x2_m(vbool32_t vm, const __bf16 *rs1,
+                                                size_t *new_vl, size_t vl) {
+  return __riscv_vlseg2e16ff(vm, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg2e16ff_v_bf16m1x2_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } @llvm.riscv.vlseg2ff.mask.nxv4bf16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } poison, <vscale x 4 x bfloat> [[TMP1]], 0
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 2
+// CHECK-RV64-NEXT:    store i64 [[TMP5]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP4]]
+//
+vbfloat16m1x2_t test_vlseg2e16ff_v_bf16m1x2_m(vbool16_t vm, const __bf16 *rs1,
+                                              size_t *new_vl, size_t vl) {
+  return __riscv_vlseg2e16ff(vm, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vlseg2e16ff_v_bf16m2x2_m(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } @llvm.riscv.vlseg2ff.mask.nxv8bf16.i64(<vscale x 8 x bfloat> poison, <vscale x 8 x bfloat> poison, ptr [[RS1]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } poison, <vscale x 8 x bfloat> [[TMP1]], 0
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP2]], <vscale x 8 x bfloat> [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP0]], 2
+// CHECK-RV64-NEXT:    store i64 [[TMP5]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP4]]
+//
+vbfloat16m2x2_t test_vlseg2e16ff_v_bf16m2x2_m(vbool8_t vm, const __bf16 *rs1,
+                                              size_t *new_vl, size_t vl) {
+  return __riscv_vlseg2e16ff(vm, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @test_vlseg2e16ff_v_bf16m4x2_m(
+// CHECK-RV64-SAME: <vscale x 16 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 16 x bfloat>, <vscale x 16 x bfloat>, i64 } @llvm.riscv.vlseg2ff.mask.nxv16bf16.i64(<vscale x 16 x bfloat> poison, <vscale x 16 x bfloat> poison, ptr [[RS1]], <vscale x 16 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = insertvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } poison, <vscale x 16 x bfloat> [[TMP1]], 0
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[TMP2]], <vscale x 16 x bfloat> [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat>, i64 } [[TMP0]], 2
+// CHECK-RV64-NEXT:    store i64 [[TMP5]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[TMP4]]
+//
+vbfloat16m4x2_t test_vlseg2e16ff_v_bf16m4x2_m(vbool4_t vm, const __bf16 *rs1,
+                                              size_t *new_vl, size_t vl) {
+  return __riscv_vlseg2e16ff(vm, rs1, new_vl, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vlseg3e16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vlseg3e16.c
new file mode 100644
index 000000000000..0ce174b2402a
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vlseg3e16.c
@@ -0,0 +1,53 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg3e16_v_bf16mf4x3_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlseg3.mask.nxv1bf16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x3_t test_vlseg3e16_v_bf16mf4x3_m(vbool64_t vm, const __bf16 *rs1,
+                                              size_t vl) {
+  return __riscv_vlseg3e16(vm, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg3e16_v_bf16mf2x3_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlseg3.mask.nxv2bf16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x3_t test_vlseg3e16_v_bf16mf2x3_m(vbool32_t vm, const __bf16 *rs1,
+                                              size_t vl) {
+  return __riscv_vlseg3e16(vm, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg3e16_v_bf16m1x3_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlseg3.mask.nxv4bf16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x3_t test_vlseg3e16_v_bf16m1x3_m(vbool16_t vm, const __bf16 *rs1,
+                                            size_t vl) {
+  return __riscv_vlseg3e16(vm, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vlseg3e16_v_bf16m2x3_m(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vlseg3.mask.nxv8bf16.i64(<vscale x 8 x bfloat> poison, <vscale x 8 x bfloat> poison, <vscale x 8 x bfloat> poison, ptr [[RS1]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP0]]
+//
+vbfloat16m2x3_t test_vlseg3e16_v_bf16m2x3_m(vbool8_t vm, const __bf16 *rs1,
+                                            size_t vl) {
+  return __riscv_vlseg3e16(vm, rs1, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vlseg3e16ff.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vlseg3e16ff.c
new file mode 100644
index 000000000000..3545a54199a7
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vlseg3e16ff.c
@@ -0,0 +1,85 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg3e16ff_v_bf16mf4x3_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } @llvm.riscv.vlseg3ff.mask.nxv1bf16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } poison, <vscale x 1 x bfloat> [[TMP1]], 0
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 2
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], 2
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 3
+// CHECK-RV64-NEXT:    store i64 [[TMP7]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP6]]
+//
+vbfloat16mf4x3_t test_vlseg3e16ff_v_bf16mf4x3_m(vbool64_t vm, const __bf16 *rs1,
+                                                size_t *new_vl, size_t vl) {
+  return __riscv_vlseg3e16ff(vm, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg3e16ff_v_bf16mf2x3_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } @llvm.riscv.vlseg3ff.mask.nxv2bf16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } poison, <vscale x 2 x bfloat> [[TMP1]], 0
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 2
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], 2
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 3
+// CHECK-RV64-NEXT:    store i64 [[TMP7]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP6]]
+//
+vbfloat16mf2x3_t test_vlseg3e16ff_v_bf16mf2x3_m(vbool32_t vm, const __bf16 *rs1,
+                                                size_t *new_vl, size_t vl) {
+  return __riscv_vlseg3e16ff(vm, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg3e16ff_v_bf16m1x3_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } @llvm.riscv.vlseg3ff.mask.nxv4bf16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } poison, <vscale x 4 x bfloat> [[TMP1]], 0
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 2
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], 2
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 3
+// CHECK-RV64-NEXT:    store i64 [[TMP7]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP6]]
+//
+vbfloat16m1x3_t test_vlseg3e16ff_v_bf16m1x3_m(vbool16_t vm, const __bf16 *rs1,
+                                              size_t *new_vl, size_t vl) {
+  return __riscv_vlseg3e16ff(vm, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vlseg3e16ff_v_bf16m2x3_m(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } @llvm.riscv.vlseg3ff.mask.nxv8bf16.i64(<vscale x 8 x bfloat> poison, <vscale x 8 x bfloat> poison, <vscale x 8 x bfloat> poison, ptr [[RS1]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } poison, <vscale x 8 x bfloat> [[TMP1]], 0
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP2]], <vscale x 8 x bfloat> [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP0]], 2
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP4]], <vscale x 8 x bfloat> [[TMP5]], 2
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP0]], 3
+// CHECK-RV64-NEXT:    store i64 [[TMP7]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP6]]
+//
+vbfloat16m2x3_t test_vlseg3e16ff_v_bf16m2x3_m(vbool8_t vm, const __bf16 *rs1,
+                                              size_t *new_vl, size_t vl) {
+  return __riscv_vlseg3e16ff(vm, rs1, new_vl, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vlseg4e16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vlseg4e16.c
new file mode 100644
index 000000000000..c17b1eab338b
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vlseg4e16.c
@@ -0,0 +1,53 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg4e16_v_bf16mf4x4_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlseg4.mask.nxv1bf16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x4_t test_vlseg4e16_v_bf16mf4x4_m(vbool64_t vm, const __bf16 *rs1,
+                                              size_t vl) {
+  return __riscv_vlseg4e16(vm, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg4e16_v_bf16mf2x4_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlseg4.mask.nxv2bf16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x4_t test_vlseg4e16_v_bf16mf2x4_m(vbool32_t vm, const __bf16 *rs1,
+                                              size_t vl) {
+  return __riscv_vlseg4e16(vm, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg4e16_v_bf16m1x4_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlseg4.mask.nxv4bf16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x4_t test_vlseg4e16_v_bf16m1x4_m(vbool16_t vm, const __bf16 *rs1,
+                                            size_t vl) {
+  return __riscv_vlseg4e16(vm, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vlseg4e16_v_bf16m2x4_m(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vlseg4.mask.nxv8bf16.i64(<vscale x 8 x bfloat> poison, <vscale x 8 x bfloat> poison, <vscale x 8 x bfloat> poison, <vscale x 8 x bfloat> poison, ptr [[RS1]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP0]]
+//
+vbfloat16m2x4_t test_vlseg4e16_v_bf16m2x4_m(vbool8_t vm, const __bf16 *rs1,
+                                            size_t vl) {
+  return __riscv_vlseg4e16(vm, rs1, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vlseg4e16ff.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vlseg4e16ff.c
new file mode 100644
index 000000000000..3aadacfd62eb
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vlseg4e16ff.c
@@ -0,0 +1,93 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg4e16ff_v_bf16mf4x4_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } @llvm.riscv.vlseg4ff.mask.nxv1bf16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } poison, <vscale x 1 x bfloat> [[TMP1]], 0
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 2
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], 2
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 3
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP6]], <vscale x 1 x bfloat> [[TMP7]], 3
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 4
+// CHECK-RV64-NEXT:    store i64 [[TMP9]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP8]]
+//
+vbfloat16mf4x4_t test_vlseg4e16ff_v_bf16mf4x4_m(vbool64_t vm, const __bf16 *rs1,
+                                                size_t *new_vl, size_t vl) {
+  return __riscv_vlseg4e16ff(vm, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg4e16ff_v_bf16mf2x4_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } @llvm.riscv.vlseg4ff.mask.nxv2bf16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } poison, <vscale x 2 x bfloat> [[TMP1]], 0
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 2
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], 2
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 3
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP6]], <vscale x 2 x bfloat> [[TMP7]], 3
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 4
+// CHECK-RV64-NEXT:    store i64 [[TMP9]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP8]]
+//
+vbfloat16mf2x4_t test_vlseg4e16ff_v_bf16mf2x4_m(vbool32_t vm, const __bf16 *rs1,
+                                                size_t *new_vl, size_t vl) {
+  return __riscv_vlseg4e16ff(vm, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg4e16ff_v_bf16m1x4_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } @llvm.riscv.vlseg4ff.mask.nxv4bf16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } poison, <vscale x 4 x bfloat> [[TMP1]], 0
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 2
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], 2
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 3
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP6]], <vscale x 4 x bfloat> [[TMP7]], 3
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 4
+// CHECK-RV64-NEXT:    store i64 [[TMP9]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP8]]
+//
+vbfloat16m1x4_t test_vlseg4e16ff_v_bf16m1x4_m(vbool16_t vm, const __bf16 *rs1,
+                                              size_t *new_vl, size_t vl) {
+  return __riscv_vlseg4e16ff(vm, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vlseg4e16ff_v_bf16m2x4_m(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } @llvm.riscv.vlseg4ff.mask.nxv8bf16.i64(<vscale x 8 x bfloat> poison, <vscale x 8 x bfloat> poison, <vscale x 8 x bfloat> poison, <vscale x 8 x bfloat> poison, ptr [[RS1]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } poison, <vscale x 8 x bfloat> [[TMP1]], 0
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP2]], <vscale x 8 x bfloat> [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP0]], 2
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP4]], <vscale x 8 x bfloat> [[TMP5]], 2
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP0]], 3
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP6]], <vscale x 8 x bfloat> [[TMP7]], 3
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP0]], 4
+// CHECK-RV64-NEXT:    store i64 [[TMP9]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP8]]
+//
+vbfloat16m2x4_t test_vlseg4e16ff_v_bf16m2x4_m(vbool8_t vm, const __bf16 *rs1,
+                                              size_t *new_vl, size_t vl) {
+  return __riscv_vlseg4e16ff(vm, rs1, new_vl, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vlseg5e16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vlseg5e16.c
new file mode 100644
index 000000000000..029916fde042
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vlseg5e16.c
@@ -0,0 +1,42 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg5e16_v_bf16mf4x5_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlseg5.mask.nxv1bf16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x5_t test_vlseg5e16_v_bf16mf4x5_m(vbool64_t vm, const __bf16 *rs1,
+                                              size_t vl) {
+  return __riscv_vlseg5e16(vm, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg5e16_v_bf16mf2x5_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlseg5.mask.nxv2bf16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x5_t test_vlseg5e16_v_bf16mf2x5_m(vbool32_t vm, const __bf16 *rs1,
+                                              size_t vl) {
+  return __riscv_vlseg5e16(vm, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg5e16_v_bf16m1x5_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlseg5.mask.nxv4bf16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x5_t test_vlseg5e16_v_bf16m1x5_m(vbool16_t vm, const __bf16 *rs1,
+                                            size_t vl) {
+  return __riscv_vlseg5e16(vm, rs1, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vlseg5e16ff.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vlseg5e16ff.c
new file mode 100644
index 000000000000..d1c991ed2cee
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vlseg5e16ff.c
@@ -0,0 +1,78 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg5e16ff_v_bf16mf4x5_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } @llvm.riscv.vlseg5ff.mask.nxv1bf16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } poison, <vscale x 1 x bfloat> [[TMP1]], 0
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 2
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], 2
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 3
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP6]], <vscale x 1 x bfloat> [[TMP7]], 3
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 4
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP8]], <vscale x 1 x bfloat> [[TMP9]], 4
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 5
+// CHECK-RV64-NEXT:    store i64 [[TMP11]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP10]]
+//
+vbfloat16mf4x5_t test_vlseg5e16ff_v_bf16mf4x5_m(vbool64_t vm, const __bf16 *rs1,
+                                                size_t *new_vl, size_t vl) {
+  return __riscv_vlseg5e16ff(vm, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg5e16ff_v_bf16mf2x5_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } @llvm.riscv.vlseg5ff.mask.nxv2bf16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } poison, <vscale x 2 x bfloat> [[TMP1]], 0
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 2
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], 2
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 3
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP6]], <vscale x 2 x bfloat> [[TMP7]], 3
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 4
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP8]], <vscale x 2 x bfloat> [[TMP9]], 4
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 5
+// CHECK-RV64-NEXT:    store i64 [[TMP11]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP10]]
+//
+vbfloat16mf2x5_t test_vlseg5e16ff_v_bf16mf2x5_m(vbool32_t vm, const __bf16 *rs1,
+                                                size_t *new_vl, size_t vl) {
+  return __riscv_vlseg5e16ff(vm, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg5e16ff_v_bf16m1x5_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } @llvm.riscv.vlseg5ff.mask.nxv4bf16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } poison, <vscale x 4 x bfloat> [[TMP1]], 0
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 2
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], 2
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 3
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP6]], <vscale x 4 x bfloat> [[TMP7]], 3
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 4
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP8]], <vscale x 4 x bfloat> [[TMP9]], 4
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 5
+// CHECK-RV64-NEXT:    store i64 [[TMP11]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP10]]
+//
+vbfloat16m1x5_t test_vlseg5e16ff_v_bf16m1x5_m(vbool16_t vm, const __bf16 *rs1,
+                                              size_t *new_vl, size_t vl) {
+  return __riscv_vlseg5e16ff(vm, rs1, new_vl, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vlseg6e16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vlseg6e16.c
new file mode 100644
index 000000000000..33ecd52a6e66
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vlseg6e16.c
@@ -0,0 +1,42 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg6e16_v_bf16mf4x6_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlseg6.mask.nxv1bf16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x6_t test_vlseg6e16_v_bf16mf4x6_m(vbool64_t vm, const __bf16 *rs1,
+                                              size_t vl) {
+  return __riscv_vlseg6e16(vm, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg6e16_v_bf16mf2x6_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlseg6.mask.nxv2bf16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x6_t test_vlseg6e16_v_bf16mf2x6_m(vbool32_t vm, const __bf16 *rs1,
+                                              size_t vl) {
+  return __riscv_vlseg6e16(vm, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg6e16_v_bf16m1x6_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlseg6.mask.nxv4bf16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x6_t test_vlseg6e16_v_bf16m1x6_m(vbool16_t vm, const __bf16 *rs1,
+                                            size_t vl) {
+  return __riscv_vlseg6e16(vm, rs1, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vlseg6e16ff.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vlseg6e16ff.c
new file mode 100644
index 000000000000..5bc58996f14b
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vlseg6e16ff.c
@@ -0,0 +1,84 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg6e16ff_v_bf16mf4x6_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } @llvm.riscv.vlseg6ff.mask.nxv1bf16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } poison, <vscale x 1 x bfloat> [[TMP1]], 0
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 2
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], 2
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 3
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP6]], <vscale x 1 x bfloat> [[TMP7]], 3
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 4
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP8]], <vscale x 1 x bfloat> [[TMP9]], 4
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 5
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP10]], <vscale x 1 x bfloat> [[TMP11]], 5
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 6
+// CHECK-RV64-NEXT:    store i64 [[TMP13]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP12]]
+//
+vbfloat16mf4x6_t test_vlseg6e16ff_v_bf16mf4x6_m(vbool64_t vm, const __bf16 *rs1,
+                                                size_t *new_vl, size_t vl) {
+  return __riscv_vlseg6e16ff(vm, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg6e16ff_v_bf16mf2x6_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } @llvm.riscv.vlseg6ff.mask.nxv2bf16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } poison, <vscale x 2 x bfloat> [[TMP1]], 0
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 2
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], 2
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 3
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP6]], <vscale x 2 x bfloat> [[TMP7]], 3
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 4
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP8]], <vscale x 2 x bfloat> [[TMP9]], 4
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 5
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP10]], <vscale x 2 x bfloat> [[TMP11]], 5
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 6
+// CHECK-RV64-NEXT:    store i64 [[TMP13]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP12]]
+//
+vbfloat16mf2x6_t test_vlseg6e16ff_v_bf16mf2x6_m(vbool32_t vm, const __bf16 *rs1,
+                                                size_t *new_vl, size_t vl) {
+  return __riscv_vlseg6e16ff(vm, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg6e16ff_v_bf16m1x6_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } @llvm.riscv.vlseg6ff.mask.nxv4bf16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } poison, <vscale x 4 x bfloat> [[TMP1]], 0
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 2
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], 2
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 3
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP6]], <vscale x 4 x bfloat> [[TMP7]], 3
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 4
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP8]], <vscale x 4 x bfloat> [[TMP9]], 4
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 5
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP10]], <vscale x 4 x bfloat> [[TMP11]], 5
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 6
+// CHECK-RV64-NEXT:    store i64 [[TMP13]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP12]]
+//
+vbfloat16m1x6_t test_vlseg6e16ff_v_bf16m1x6_m(vbool16_t vm, const __bf16 *rs1,
+                                              size_t *new_vl, size_t vl) {
+  return __riscv_vlseg6e16ff(vm, rs1, new_vl, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vlseg7e16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vlseg7e16.c
new file mode 100644
index 000000000000..9804f5edac61
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vlseg7e16.c
@@ -0,0 +1,42 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg7e16_v_bf16mf4x7_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlseg7.mask.nxv1bf16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x7_t test_vlseg7e16_v_bf16mf4x7_m(vbool64_t vm, const __bf16 *rs1,
+                                              size_t vl) {
+  return __riscv_vlseg7e16(vm, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg7e16_v_bf16mf2x7_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlseg7.mask.nxv2bf16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x7_t test_vlseg7e16_v_bf16mf2x7_m(vbool32_t vm, const __bf16 *rs1,
+                                              size_t vl) {
+  return __riscv_vlseg7e16(vm, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg7e16_v_bf16m1x7_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlseg7.mask.nxv4bf16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x7_t test_vlseg7e16_v_bf16m1x7_m(vbool16_t vm, const __bf16 *rs1,
+                                            size_t vl) {
+  return __riscv_vlseg7e16(vm, rs1, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vlseg7e16ff.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vlseg7e16ff.c
new file mode 100644
index 000000000000..646c177806cf
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vlseg7e16ff.c
@@ -0,0 +1,90 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg7e16ff_v_bf16mf4x7_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } @llvm.riscv.vlseg7ff.mask.nxv1bf16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } poison, <vscale x 1 x bfloat> [[TMP1]], 0
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 2
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], 2
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 3
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP6]], <vscale x 1 x bfloat> [[TMP7]], 3
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 4
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP8]], <vscale x 1 x bfloat> [[TMP9]], 4
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 5
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP10]], <vscale x 1 x bfloat> [[TMP11]], 5
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 6
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP12]], <vscale x 1 x bfloat> [[TMP13]], 6
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 7
+// CHECK-RV64-NEXT:    store i64 [[TMP15]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP14]]
+//
+vbfloat16mf4x7_t test_vlseg7e16ff_v_bf16mf4x7_m(vbool64_t vm, const __bf16 *rs1,
+                                                size_t *new_vl, size_t vl) {
+  return __riscv_vlseg7e16ff(vm, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg7e16ff_v_bf16mf2x7_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } @llvm.riscv.vlseg7ff.mask.nxv2bf16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } poison, <vscale x 2 x bfloat> [[TMP1]], 0
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 2
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], 2
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 3
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP6]], <vscale x 2 x bfloat> [[TMP7]], 3
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 4
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP8]], <vscale x 2 x bfloat> [[TMP9]], 4
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 5
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP10]], <vscale x 2 x bfloat> [[TMP11]], 5
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 6
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP12]], <vscale x 2 x bfloat> [[TMP13]], 6
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 7
+// CHECK-RV64-NEXT:    store i64 [[TMP15]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP14]]
+//
+vbfloat16mf2x7_t test_vlseg7e16ff_v_bf16mf2x7_m(vbool32_t vm, const __bf16 *rs1,
+                                                size_t *new_vl, size_t vl) {
+  return __riscv_vlseg7e16ff(vm, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg7e16ff_v_bf16m1x7_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } @llvm.riscv.vlseg7ff.mask.nxv4bf16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } poison, <vscale x 4 x bfloat> [[TMP1]], 0
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 2
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], 2
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 3
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP6]], <vscale x 4 x bfloat> [[TMP7]], 3
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 4
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP8]], <vscale x 4 x bfloat> [[TMP9]], 4
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 5
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP10]], <vscale x 4 x bfloat> [[TMP11]], 5
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 6
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP12]], <vscale x 4 x bfloat> [[TMP13]], 6
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 7
+// CHECK-RV64-NEXT:    store i64 [[TMP15]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP14]]
+//
+vbfloat16m1x7_t test_vlseg7e16ff_v_bf16m1x7_m(vbool16_t vm, const __bf16 *rs1,
+                                              size_t *new_vl, size_t vl) {
+  return __riscv_vlseg7e16ff(vm, rs1, new_vl, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vlseg8e16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vlseg8e16.c
new file mode 100644
index 000000000000..a969317f246d
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vlseg8e16.c
@@ -0,0 +1,42 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg8e16_v_bf16mf4x8_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlseg8.mask.nxv1bf16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x8_t test_vlseg8e16_v_bf16mf4x8_m(vbool64_t vm, const __bf16 *rs1,
+                                              size_t vl) {
+  return __riscv_vlseg8e16(vm, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg8e16_v_bf16mf2x8_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlseg8.mask.nxv2bf16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x8_t test_vlseg8e16_v_bf16mf2x8_m(vbool32_t vm, const __bf16 *rs1,
+                                              size_t vl) {
+  return __riscv_vlseg8e16(vm, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg8e16_v_bf16m1x8_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlseg8.mask.nxv4bf16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x8_t test_vlseg8e16_v_bf16m1x8_m(vbool16_t vm, const __bf16 *rs1,
+                                            size_t vl) {
+  return __riscv_vlseg8e16(vm, rs1, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vlseg8e16ff.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vlseg8e16ff.c
new file mode 100644
index 000000000000..bec9006b8afc
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vlseg8e16ff.c
@@ -0,0 +1,96 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg8e16ff_v_bf16mf4x8_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } @llvm.riscv.vlseg8ff.mask.nxv1bf16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } poison, <vscale x 1 x bfloat> [[TMP1]], 0
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 2
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], 2
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 3
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP6]], <vscale x 1 x bfloat> [[TMP7]], 3
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 4
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP8]], <vscale x 1 x bfloat> [[TMP9]], 4
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 5
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP10]], <vscale x 1 x bfloat> [[TMP11]], 5
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 6
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP12]], <vscale x 1 x bfloat> [[TMP13]], 6
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 7
+// CHECK-RV64-NEXT:    [[TMP16:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP14]], <vscale x 1 x bfloat> [[TMP15]], 7
+// CHECK-RV64-NEXT:    [[TMP17:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP0]], 8
+// CHECK-RV64-NEXT:    store i64 [[TMP17]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP16]]
+//
+vbfloat16mf4x8_t test_vlseg8e16ff_v_bf16mf4x8_m(vbool64_t vm, const __bf16 *rs1,
+                                                size_t *new_vl, size_t vl) {
+  return __riscv_vlseg8e16ff(vm, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg8e16ff_v_bf16mf2x8_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } @llvm.riscv.vlseg8ff.mask.nxv2bf16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } poison, <vscale x 2 x bfloat> [[TMP1]], 0
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 2
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], 2
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 3
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP6]], <vscale x 2 x bfloat> [[TMP7]], 3
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 4
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP8]], <vscale x 2 x bfloat> [[TMP9]], 4
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 5
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP10]], <vscale x 2 x bfloat> [[TMP11]], 5
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 6
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP12]], <vscale x 2 x bfloat> [[TMP13]], 6
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 7
+// CHECK-RV64-NEXT:    [[TMP16:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP14]], <vscale x 2 x bfloat> [[TMP15]], 7
+// CHECK-RV64-NEXT:    [[TMP17:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP0]], 8
+// CHECK-RV64-NEXT:    store i64 [[TMP17]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP16]]
+//
+vbfloat16mf2x8_t test_vlseg8e16ff_v_bf16mf2x8_m(vbool32_t vm, const __bf16 *rs1,
+                                                size_t *new_vl, size_t vl) {
+  return __riscv_vlseg8e16ff(vm, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg8e16ff_v_bf16m1x8_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } @llvm.riscv.vlseg8ff.mask.nxv4bf16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } poison, <vscale x 4 x bfloat> [[TMP1]], 0
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 2
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], 2
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 3
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP6]], <vscale x 4 x bfloat> [[TMP7]], 3
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 4
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP8]], <vscale x 4 x bfloat> [[TMP9]], 4
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 5
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP10]], <vscale x 4 x bfloat> [[TMP11]], 5
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 6
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP12]], <vscale x 4 x bfloat> [[TMP13]], 6
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 7
+// CHECK-RV64-NEXT:    [[TMP16:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP14]], <vscale x 4 x bfloat> [[TMP15]], 7
+// CHECK-RV64-NEXT:    [[TMP17:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP0]], 8
+// CHECK-RV64-NEXT:    store i64 [[TMP17]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP16]]
+//
+vbfloat16m1x8_t test_vlseg8e16ff_v_bf16m1x8_m(vbool16_t vm, const __bf16 *rs1,
+                                              size_t *new_vl, size_t vl) {
+  return __riscv_vlseg8e16ff(vm, rs1, new_vl, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vlsseg2e16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vlsseg2e16.c
new file mode 100644
index 000000000000..ba5cf455287e
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vlsseg2e16.c
@@ -0,0 +1,64 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlsseg2e16_v_bf16mf4x2_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlsseg2.mask.nxv1bf16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], i64 [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x2_t test_vlsseg2e16_v_bf16mf4x2_m(vbool64_t vm, const __bf16 *rs1,
+                                               ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg2e16(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlsseg2e16_v_bf16mf2x2_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlsseg2.mask.nxv2bf16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], i64 [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x2_t test_vlsseg2e16_v_bf16mf2x2_m(vbool32_t vm, const __bf16 *rs1,
+                                               ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg2e16(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlsseg2e16_v_bf16m1x2_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlsseg2.mask.nxv4bf16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], i64 [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x2_t test_vlsseg2e16_v_bf16m1x2_m(vbool16_t vm, const __bf16 *rs1,
+                                             ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg2e16(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vlsseg2e16_v_bf16m2x2_m(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vlsseg2.mask.nxv8bf16.i64(<vscale x 8 x bfloat> poison, <vscale x 8 x bfloat> poison, ptr [[RS1]], i64 [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP0]]
+//
+vbfloat16m2x2_t test_vlsseg2e16_v_bf16m2x2_m(vbool8_t vm, const __bf16 *rs1,
+                                             ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg2e16(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @test_vlsseg2e16_v_bf16m4x2_m(
+// CHECK-RV64-SAME: <vscale x 16 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @llvm.riscv.vlsseg2.mask.nxv16bf16.i64(<vscale x 16 x bfloat> poison, <vscale x 16 x bfloat> poison, ptr [[RS1]], i64 [[RS2]], <vscale x 16 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[TMP0]]
+//
+vbfloat16m4x2_t test_vlsseg2e16_v_bf16m4x2_m(vbool4_t vm, const __bf16 *rs1,
+                                             ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg2e16(vm, rs1, rs2, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vlsseg3e16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vlsseg3e16.c
new file mode 100644
index 000000000000..0cdad1c52725
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vlsseg3e16.c
@@ -0,0 +1,53 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlsseg3e16_v_bf16mf4x3_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlsseg3.mask.nxv1bf16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], i64 [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x3_t test_vlsseg3e16_v_bf16mf4x3_m(vbool64_t vm, const __bf16 *rs1,
+                                               ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg3e16(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlsseg3e16_v_bf16mf2x3_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlsseg3.mask.nxv2bf16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], i64 [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x3_t test_vlsseg3e16_v_bf16mf2x3_m(vbool32_t vm, const __bf16 *rs1,
+                                               ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg3e16(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlsseg3e16_v_bf16m1x3_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlsseg3.mask.nxv4bf16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], i64 [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x3_t test_vlsseg3e16_v_bf16m1x3_m(vbool16_t vm, const __bf16 *rs1,
+                                             ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg3e16(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vlsseg3e16_v_bf16m2x3_m(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vlsseg3.mask.nxv8bf16.i64(<vscale x 8 x bfloat> poison, <vscale x 8 x bfloat> poison, <vscale x 8 x bfloat> poison, ptr [[RS1]], i64 [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP0]]
+//
+vbfloat16m2x3_t test_vlsseg3e16_v_bf16m2x3_m(vbool8_t vm, const __bf16 *rs1,
+                                             ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg3e16(vm, rs1, rs2, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vlsseg4e16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vlsseg4e16.c
new file mode 100644
index 000000000000..d45e46fd9326
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vlsseg4e16.c
@@ -0,0 +1,53 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlsseg4e16_v_bf16mf4x4_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlsseg4.mask.nxv1bf16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], i64 [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x4_t test_vlsseg4e16_v_bf16mf4x4_m(vbool64_t vm, const __bf16 *rs1,
+                                               ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg4e16(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlsseg4e16_v_bf16mf2x4_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlsseg4.mask.nxv2bf16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], i64 [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x4_t test_vlsseg4e16_v_bf16mf2x4_m(vbool32_t vm, const __bf16 *rs1,
+                                               ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg4e16(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlsseg4e16_v_bf16m1x4_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlsseg4.mask.nxv4bf16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], i64 [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x4_t test_vlsseg4e16_v_bf16m1x4_m(vbool16_t vm, const __bf16 *rs1,
+                                             ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg4e16(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vlsseg4e16_v_bf16m2x4_m(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vlsseg4.mask.nxv8bf16.i64(<vscale x 8 x bfloat> poison, <vscale x 8 x bfloat> poison, <vscale x 8 x bfloat> poison, <vscale x 8 x bfloat> poison, ptr [[RS1]], i64 [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP0]]
+//
+vbfloat16m2x4_t test_vlsseg4e16_v_bf16m2x4_m(vbool8_t vm, const __bf16 *rs1,
+                                             ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg4e16(vm, rs1, rs2, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vlsseg5e16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vlsseg5e16.c
new file mode 100644
index 000000000000..d7484e5e6e6d
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vlsseg5e16.c
@@ -0,0 +1,42 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlsseg5e16_v_bf16mf4x5_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlsseg5.mask.nxv1bf16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], i64 [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x5_t test_vlsseg5e16_v_bf16mf4x5_m(vbool64_t vm, const __bf16 *rs1,
+                                               ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg5e16(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlsseg5e16_v_bf16mf2x5_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlsseg5.mask.nxv2bf16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], i64 [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x5_t test_vlsseg5e16_v_bf16mf2x5_m(vbool32_t vm, const __bf16 *rs1,
+                                               ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg5e16(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlsseg5e16_v_bf16m1x5_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlsseg5.mask.nxv4bf16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], i64 [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x5_t test_vlsseg5e16_v_bf16m1x5_m(vbool16_t vm, const __bf16 *rs1,
+                                             ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg5e16(vm, rs1, rs2, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vlsseg6e16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vlsseg6e16.c
new file mode 100644
index 000000000000..9bbdc186f0f5
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vlsseg6e16.c
@@ -0,0 +1,42 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlsseg6e16_v_bf16mf4x6_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlsseg6.mask.nxv1bf16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], i64 [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x6_t test_vlsseg6e16_v_bf16mf4x6_m(vbool64_t vm, const __bf16 *rs1,
+                                               ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg6e16(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlsseg6e16_v_bf16mf2x6_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlsseg6.mask.nxv2bf16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], i64 [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x6_t test_vlsseg6e16_v_bf16mf2x6_m(vbool32_t vm, const __bf16 *rs1,
+                                               ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg6e16(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlsseg6e16_v_bf16m1x6_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlsseg6.mask.nxv4bf16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], i64 [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x6_t test_vlsseg6e16_v_bf16m1x6_m(vbool16_t vm, const __bf16 *rs1,
+                                             ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg6e16(vm, rs1, rs2, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vlsseg7e16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vlsseg7e16.c
new file mode 100644
index 000000000000..364d664525fd
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vlsseg7e16.c
@@ -0,0 +1,42 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlsseg7e16_v_bf16mf4x7_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlsseg7.mask.nxv1bf16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], i64 [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x7_t test_vlsseg7e16_v_bf16mf4x7_m(vbool64_t vm, const __bf16 *rs1,
+                                               ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg7e16(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlsseg7e16_v_bf16mf2x7_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlsseg7.mask.nxv2bf16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], i64 [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x7_t test_vlsseg7e16_v_bf16mf2x7_m(vbool32_t vm, const __bf16 *rs1,
+                                               ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg7e16(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlsseg7e16_v_bf16m1x7_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlsseg7.mask.nxv4bf16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], i64 [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x7_t test_vlsseg7e16_v_bf16m1x7_m(vbool16_t vm, const __bf16 *rs1,
+                                             ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg7e16(vm, rs1, rs2, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vlsseg8e16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vlsseg8e16.c
new file mode 100644
index 000000000000..0d6a391b2ea7
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vlsseg8e16.c
@@ -0,0 +1,42 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlsseg8e16_v_bf16mf4x8_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlsseg8.mask.nxv1bf16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], i64 [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x8_t test_vlsseg8e16_v_bf16mf4x8_m(vbool64_t vm, const __bf16 *rs1,
+                                               ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg8e16(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlsseg8e16_v_bf16mf2x8_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlsseg8.mask.nxv2bf16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], i64 [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x8_t test_vlsseg8e16_v_bf16mf2x8_m(vbool32_t vm, const __bf16 *rs1,
+                                               ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg8e16(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlsseg8e16_v_bf16m1x8_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlsseg8.mask.nxv4bf16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], i64 [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x8_t test_vlsseg8e16_v_bf16m1x8_m(vbool16_t vm, const __bf16 *rs1,
+                                             ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg8e16(vm, rs1, rs2, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vluxei16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vluxei16.c
new file mode 100644
index 000000000000..265482129740
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vluxei16.c
@@ -0,0 +1,141 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 1 x bfloat> @test_vluxei16_v_bf16mf4(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x bfloat> @llvm.riscv.vluxei.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> poison, ptr [[RS1]], <vscale x 1 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret <vscale x 1 x bfloat> [[TMP0]]
+//
+vbfloat16mf4_t test_vluxei16_v_bf16mf4(const __bf16 *rs1, vuint16mf4_t rs2,
+                                       size_t vl) {
+  return __riscv_vluxei16(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 2 x bfloat> @test_vluxei16_v_bf16mf2(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x bfloat> @llvm.riscv.vluxei.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> poison, ptr [[RS1]], <vscale x 2 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret <vscale x 2 x bfloat> [[TMP0]]
+//
+vbfloat16mf2_t test_vluxei16_v_bf16mf2(const __bf16 *rs1, vuint16mf2_t rs2,
+                                       size_t vl) {
+  return __riscv_vluxei16(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 4 x bfloat> @test_vluxei16_v_bf16m1(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x bfloat> @llvm.riscv.vluxei.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> poison, ptr [[RS1]], <vscale x 4 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret <vscale x 4 x bfloat> [[TMP0]]
+//
+vbfloat16m1_t test_vluxei16_v_bf16m1(const __bf16 *rs1, vuint16m1_t rs2,
+                                     size_t vl) {
+  return __riscv_vluxei16(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 8 x bfloat> @test_vluxei16_v_bf16m2(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x bfloat> @llvm.riscv.vluxei.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> poison, ptr [[RS1]], <vscale x 8 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret <vscale x 8 x bfloat> [[TMP0]]
+//
+vbfloat16m2_t test_vluxei16_v_bf16m2(const __bf16 *rs1, vuint16m2_t rs2,
+                                     size_t vl) {
+  return __riscv_vluxei16(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 16 x bfloat> @test_vluxei16_v_bf16m4(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 16 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x bfloat> @llvm.riscv.vluxei.nxv16bf16.nxv16i16.i64(<vscale x 16 x bfloat> poison, ptr [[RS1]], <vscale x 16 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret <vscale x 16 x bfloat> [[TMP0]]
+//
+vbfloat16m4_t test_vluxei16_v_bf16m4(const __bf16 *rs1, vuint16m4_t rs2,
+                                     size_t vl) {
+  return __riscv_vluxei16(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 32 x bfloat> @test_vluxei16_v_bf16m8(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 32 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x bfloat> @llvm.riscv.vluxei.nxv32bf16.nxv32i16.i64(<vscale x 32 x bfloat> poison, ptr [[RS1]], <vscale x 32 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret <vscale x 32 x bfloat> [[TMP0]]
+//
+vbfloat16m8_t test_vluxei16_v_bf16m8(const __bf16 *rs1, vuint16m8_t rs2,
+                                     size_t vl) {
+  return __riscv_vluxei16(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 1 x bfloat> @test_vluxei16_v_bf16mf4_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x bfloat> @llvm.riscv.vluxei.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> poison, ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret <vscale x 1 x bfloat> [[TMP0]]
+//
+vbfloat16mf4_t test_vluxei16_v_bf16mf4_m(vbool64_t vm, const __bf16 *rs1,
+                                         vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vluxei16(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 2 x bfloat> @test_vluxei16_v_bf16mf2_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x bfloat> @llvm.riscv.vluxei.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> poison, ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret <vscale x 2 x bfloat> [[TMP0]]
+//
+vbfloat16mf2_t test_vluxei16_v_bf16mf2_m(vbool32_t vm, const __bf16 *rs1,
+                                         vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vluxei16(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 4 x bfloat> @test_vluxei16_v_bf16m1_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x bfloat> @llvm.riscv.vluxei.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> poison, ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret <vscale x 4 x bfloat> [[TMP0]]
+//
+vbfloat16m1_t test_vluxei16_v_bf16m1_m(vbool16_t vm, const __bf16 *rs1,
+                                       vuint16m1_t rs2, size_t vl) {
+  return __riscv_vluxei16(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 8 x bfloat> @test_vluxei16_v_bf16m2_m(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x bfloat> @llvm.riscv.vluxei.mask.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> poison, ptr [[RS1]], <vscale x 8 x i16> [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret <vscale x 8 x bfloat> [[TMP0]]
+//
+vbfloat16m2_t test_vluxei16_v_bf16m2_m(vbool8_t vm, const __bf16 *rs1,
+                                       vuint16m2_t rs2, size_t vl) {
+  return __riscv_vluxei16(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 16 x bfloat> @test_vluxei16_v_bf16m4_m(
+// CHECK-RV64-SAME: <vscale x 16 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 16 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x bfloat> @llvm.riscv.vluxei.mask.nxv16bf16.nxv16i16.i64(<vscale x 16 x bfloat> poison, ptr [[RS1]], <vscale x 16 x i16> [[RS2]], <vscale x 16 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret <vscale x 16 x bfloat> [[TMP0]]
+//
+vbfloat16m4_t test_vluxei16_v_bf16m4_m(vbool4_t vm, const __bf16 *rs1,
+                                       vuint16m4_t rs2, size_t vl) {
+  return __riscv_vluxei16(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 32 x bfloat> @test_vluxei16_v_bf16m8_m(
+// CHECK-RV64-SAME: <vscale x 32 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 32 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x bfloat> @llvm.riscv.vluxei.mask.nxv32bf16.nxv32i16.i64(<vscale x 32 x bfloat> poison, ptr [[RS1]], <vscale x 32 x i16> [[RS2]], <vscale x 32 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret <vscale x 32 x bfloat> [[TMP0]]
+//
+vbfloat16m8_t test_vluxei16_v_bf16m8_m(vbool2_t vm, const __bf16 *rs1,
+                                       vuint16m8_t rs2, size_t vl) {
+  return __riscv_vluxei16(vm, rs1, rs2, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vluxseg2ei16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vluxseg2ei16.c
new file mode 100644
index 000000000000..b862aae053da
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vluxseg2ei16.c
@@ -0,0 +1,121 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vluxseg2ei16_v_bf16mf4x2(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vluxseg2.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], <vscale x 1 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x2_t test_vluxseg2ei16_v_bf16mf4x2(const __bf16 *rs1,
+                                               vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vluxseg2ei16(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vluxseg2ei16_v_bf16mf2x2(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vluxseg2.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], <vscale x 2 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x2_t test_vluxseg2ei16_v_bf16mf2x2(const __bf16 *rs1,
+                                               vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vluxseg2ei16(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vluxseg2ei16_v_bf16m1x2(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vluxseg2.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], <vscale x 4 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x2_t test_vluxseg2ei16_v_bf16m1x2(const __bf16 *rs1, vuint16m1_t rs2,
+                                             size_t vl) {
+  return __riscv_vluxseg2ei16(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vluxseg2ei16_v_bf16m2x2(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vluxseg2.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> poison, <vscale x 8 x bfloat> poison, ptr [[RS1]], <vscale x 8 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP0]]
+//
+vbfloat16m2x2_t test_vluxseg2ei16_v_bf16m2x2(const __bf16 *rs1, vuint16m2_t rs2,
+                                             size_t vl) {
+  return __riscv_vluxseg2ei16(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @test_vluxseg2ei16_v_bf16m4x2(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 16 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @llvm.riscv.vluxseg2.nxv16bf16.nxv16i16.i64(<vscale x 16 x bfloat> poison, <vscale x 16 x bfloat> poison, ptr [[RS1]], <vscale x 16 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[TMP0]]
+//
+vbfloat16m4x2_t test_vluxseg2ei16_v_bf16m4x2(const __bf16 *rs1, vuint16m4_t rs2,
+                                             size_t vl) {
+  return __riscv_vluxseg2ei16(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vluxseg2ei16_v_bf16mf4x2_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vluxseg2.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x2_t test_vluxseg2ei16_v_bf16mf4x2_m(vbool64_t vm,
+                                                 const __bf16 *rs1,
+                                                 vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vluxseg2ei16(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vluxseg2ei16_v_bf16mf2x2_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vluxseg2.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x2_t test_vluxseg2ei16_v_bf16mf2x2_m(vbool32_t vm,
+                                                 const __bf16 *rs1,
+                                                 vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vluxseg2ei16(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vluxseg2ei16_v_bf16m1x2_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vluxseg2.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x2_t test_vluxseg2ei16_v_bf16m1x2_m(vbool16_t vm, const __bf16 *rs1,
+                                               vuint16m1_t rs2, size_t vl) {
+  return __riscv_vluxseg2ei16(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vluxseg2ei16_v_bf16m2x2_m(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vluxseg2.mask.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> poison, <vscale x 8 x bfloat> poison, ptr [[RS1]], <vscale x 8 x i16> [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP0]]
+//
+vbfloat16m2x2_t test_vluxseg2ei16_v_bf16m2x2_m(vbool8_t vm, const __bf16 *rs1,
+                                               vuint16m2_t rs2, size_t vl) {
+  return __riscv_vluxseg2ei16(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @test_vluxseg2ei16_v_bf16m4x2_m(
+// CHECK-RV64-SAME: <vscale x 16 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 16 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @llvm.riscv.vluxseg2.mask.nxv16bf16.nxv16i16.i64(<vscale x 16 x bfloat> poison, <vscale x 16 x bfloat> poison, ptr [[RS1]], <vscale x 16 x i16> [[RS2]], <vscale x 16 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[TMP0]]
+//
+vbfloat16m4x2_t test_vluxseg2ei16_v_bf16m4x2_m(vbool4_t vm, const __bf16 *rs1,
+                                               vuint16m4_t rs2, size_t vl) {
+  return __riscv_vluxseg2ei16(vm, rs1, rs2, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vluxseg3ei16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vluxseg3ei16.c
new file mode 100644
index 000000000000..372523aebc99
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vluxseg3ei16.c
@@ -0,0 +1,99 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vluxseg3ei16_v_bf16mf4x3(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vluxseg3.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], <vscale x 1 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x3_t test_vluxseg3ei16_v_bf16mf4x3(const __bf16 *rs1,
+                                               vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vluxseg3ei16(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vluxseg3ei16_v_bf16mf2x3(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vluxseg3.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], <vscale x 2 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x3_t test_vluxseg3ei16_v_bf16mf2x3(const __bf16 *rs1,
+                                               vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vluxseg3ei16(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vluxseg3ei16_v_bf16m1x3(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vluxseg3.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], <vscale x 4 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x3_t test_vluxseg3ei16_v_bf16m1x3(const __bf16 *rs1, vuint16m1_t rs2,
+                                             size_t vl) {
+  return __riscv_vluxseg3ei16(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vluxseg3ei16_v_bf16m2x3(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vluxseg3.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> poison, <vscale x 8 x bfloat> poison, <vscale x 8 x bfloat> poison, ptr [[RS1]], <vscale x 8 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP0]]
+//
+vbfloat16m2x3_t test_vluxseg3ei16_v_bf16m2x3(const __bf16 *rs1, vuint16m2_t rs2,
+                                             size_t vl) {
+  return __riscv_vluxseg3ei16(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vluxseg3ei16_v_bf16mf4x3_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vluxseg3.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x3_t test_vluxseg3ei16_v_bf16mf4x3_m(vbool64_t vm,
+                                                 const __bf16 *rs1,
+                                                 vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vluxseg3ei16(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vluxseg3ei16_v_bf16mf2x3_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vluxseg3.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x3_t test_vluxseg3ei16_v_bf16mf2x3_m(vbool32_t vm,
+                                                 const __bf16 *rs1,
+                                                 vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vluxseg3ei16(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vluxseg3ei16_v_bf16m1x3_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vluxseg3.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x3_t test_vluxseg3ei16_v_bf16m1x3_m(vbool16_t vm, const __bf16 *rs1,
+                                               vuint16m1_t rs2, size_t vl) {
+  return __riscv_vluxseg3ei16(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vluxseg3ei16_v_bf16m2x3_m(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vluxseg3.mask.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> poison, <vscale x 8 x bfloat> poison, <vscale x 8 x bfloat> poison, ptr [[RS1]], <vscale x 8 x i16> [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP0]]
+//
+vbfloat16m2x3_t test_vluxseg3ei16_v_bf16m2x3_m(vbool8_t vm, const __bf16 *rs1,
+                                               vuint16m2_t rs2, size_t vl) {
+  return __riscv_vluxseg3ei16(vm, rs1, rs2, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vluxseg4ei16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vluxseg4ei16.c
new file mode 100644
index 000000000000..dda3615cd204
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vluxseg4ei16.c
@@ -0,0 +1,99 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vluxseg4ei16_v_bf16mf4x4(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vluxseg4.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], <vscale x 1 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x4_t test_vluxseg4ei16_v_bf16mf4x4(const __bf16 *rs1,
+                                               vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vluxseg4ei16(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vluxseg4ei16_v_bf16mf2x4(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vluxseg4.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], <vscale x 2 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x4_t test_vluxseg4ei16_v_bf16mf2x4(const __bf16 *rs1,
+                                               vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vluxseg4ei16(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vluxseg4ei16_v_bf16m1x4(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vluxseg4.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], <vscale x 4 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x4_t test_vluxseg4ei16_v_bf16m1x4(const __bf16 *rs1, vuint16m1_t rs2,
+                                             size_t vl) {
+  return __riscv_vluxseg4ei16(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vluxseg4ei16_v_bf16m2x4(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vluxseg4.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> poison, <vscale x 8 x bfloat> poison, <vscale x 8 x bfloat> poison, <vscale x 8 x bfloat> poison, ptr [[RS1]], <vscale x 8 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP0]]
+//
+vbfloat16m2x4_t test_vluxseg4ei16_v_bf16m2x4(const __bf16 *rs1, vuint16m2_t rs2,
+                                             size_t vl) {
+  return __riscv_vluxseg4ei16(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vluxseg4ei16_v_bf16mf4x4_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vluxseg4.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x4_t test_vluxseg4ei16_v_bf16mf4x4_m(vbool64_t vm,
+                                                 const __bf16 *rs1,
+                                                 vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vluxseg4ei16(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vluxseg4ei16_v_bf16mf2x4_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vluxseg4.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x4_t test_vluxseg4ei16_v_bf16mf2x4_m(vbool32_t vm,
+                                                 const __bf16 *rs1,
+                                                 vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vluxseg4ei16(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vluxseg4ei16_v_bf16m1x4_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vluxseg4.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x4_t test_vluxseg4ei16_v_bf16m1x4_m(vbool16_t vm, const __bf16 *rs1,
+                                               vuint16m1_t rs2, size_t vl) {
+  return __riscv_vluxseg4ei16(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vluxseg4ei16_v_bf16m2x4_m(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vluxseg4.mask.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> poison, <vscale x 8 x bfloat> poison, <vscale x 8 x bfloat> poison, <vscale x 8 x bfloat> poison, ptr [[RS1]], <vscale x 8 x i16> [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP0]]
+//
+vbfloat16m2x4_t test_vluxseg4ei16_v_bf16m2x4_m(vbool8_t vm, const __bf16 *rs1,
+                                               vuint16m2_t rs2, size_t vl) {
+  return __riscv_vluxseg4ei16(vm, rs1, rs2, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vluxseg5ei16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vluxseg5ei16.c
new file mode 100644
index 000000000000..52dbe05328d8
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vluxseg5ei16.c
@@ -0,0 +1,77 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vluxseg5ei16_v_bf16mf4x5(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vluxseg5.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], <vscale x 1 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x5_t test_vluxseg5ei16_v_bf16mf4x5(const __bf16 *rs1,
+                                               vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vluxseg5ei16(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vluxseg5ei16_v_bf16mf2x5(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vluxseg5.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], <vscale x 2 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x5_t test_vluxseg5ei16_v_bf16mf2x5(const __bf16 *rs1,
+                                               vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vluxseg5ei16(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vluxseg5ei16_v_bf16m1x5(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vluxseg5.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], <vscale x 4 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x5_t test_vluxseg5ei16_v_bf16m1x5(const __bf16 *rs1, vuint16m1_t rs2,
+                                             size_t vl) {
+  return __riscv_vluxseg5ei16(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vluxseg5ei16_v_bf16mf4x5_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vluxseg5.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x5_t test_vluxseg5ei16_v_bf16mf4x5_m(vbool64_t vm,
+                                                 const __bf16 *rs1,
+                                                 vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vluxseg5ei16(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vluxseg5ei16_v_bf16mf2x5_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vluxseg5.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x5_t test_vluxseg5ei16_v_bf16mf2x5_m(vbool32_t vm,
+                                                 const __bf16 *rs1,
+                                                 vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vluxseg5ei16(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vluxseg5ei16_v_bf16m1x5_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vluxseg5.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x5_t test_vluxseg5ei16_v_bf16m1x5_m(vbool16_t vm, const __bf16 *rs1,
+                                               vuint16m1_t rs2, size_t vl) {
+  return __riscv_vluxseg5ei16(vm, rs1, rs2, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vluxseg6ei16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vluxseg6ei16.c
new file mode 100644
index 000000000000..40cb79b3684e
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vluxseg6ei16.c
@@ -0,0 +1,77 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vluxseg6ei16_v_bf16mf4x6(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vluxseg6.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], <vscale x 1 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x6_t test_vluxseg6ei16_v_bf16mf4x6(const __bf16 *rs1,
+                                               vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vluxseg6ei16(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vluxseg6ei16_v_bf16mf2x6(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vluxseg6.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], <vscale x 2 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x6_t test_vluxseg6ei16_v_bf16mf2x6(const __bf16 *rs1,
+                                               vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vluxseg6ei16(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vluxseg6ei16_v_bf16m1x6(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vluxseg6.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], <vscale x 4 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x6_t test_vluxseg6ei16_v_bf16m1x6(const __bf16 *rs1, vuint16m1_t rs2,
+                                             size_t vl) {
+  return __riscv_vluxseg6ei16(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vluxseg6ei16_v_bf16mf4x6_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vluxseg6.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x6_t test_vluxseg6ei16_v_bf16mf4x6_m(vbool64_t vm,
+                                                 const __bf16 *rs1,
+                                                 vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vluxseg6ei16(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vluxseg6ei16_v_bf16mf2x6_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vluxseg6.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x6_t test_vluxseg6ei16_v_bf16mf2x6_m(vbool32_t vm,
+                                                 const __bf16 *rs1,
+                                                 vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vluxseg6ei16(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vluxseg6ei16_v_bf16m1x6_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vluxseg6.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x6_t test_vluxseg6ei16_v_bf16m1x6_m(vbool16_t vm, const __bf16 *rs1,
+                                               vuint16m1_t rs2, size_t vl) {
+  return __riscv_vluxseg6ei16(vm, rs1, rs2, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vluxseg7ei16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vluxseg7ei16.c
new file mode 100644
index 000000000000..00bea1db8879
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vluxseg7ei16.c
@@ -0,0 +1,77 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vluxseg7ei16_v_bf16mf4x7(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vluxseg7.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], <vscale x 1 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x7_t test_vluxseg7ei16_v_bf16mf4x7(const __bf16 *rs1,
+                                               vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vluxseg7ei16(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vluxseg7ei16_v_bf16mf2x7(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vluxseg7.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], <vscale x 2 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x7_t test_vluxseg7ei16_v_bf16mf2x7(const __bf16 *rs1,
+                                               vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vluxseg7ei16(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vluxseg7ei16_v_bf16m1x7(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vluxseg7.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], <vscale x 4 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x7_t test_vluxseg7ei16_v_bf16m1x7(const __bf16 *rs1, vuint16m1_t rs2,
+                                             size_t vl) {
+  return __riscv_vluxseg7ei16(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vluxseg7ei16_v_bf16mf4x7_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vluxseg7.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x7_t test_vluxseg7ei16_v_bf16mf4x7_m(vbool64_t vm,
+                                                 const __bf16 *rs1,
+                                                 vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vluxseg7ei16(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vluxseg7ei16_v_bf16mf2x7_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vluxseg7.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x7_t test_vluxseg7ei16_v_bf16mf2x7_m(vbool32_t vm,
+                                                 const __bf16 *rs1,
+                                                 vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vluxseg7ei16(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vluxseg7ei16_v_bf16m1x7_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vluxseg7.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x7_t test_vluxseg7ei16_v_bf16m1x7_m(vbool16_t vm, const __bf16 *rs1,
+                                               vuint16m1_t rs2, size_t vl) {
+  return __riscv_vluxseg7ei16(vm, rs1, rs2, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vluxseg8ei16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vluxseg8ei16.c
new file mode 100644
index 000000000000..29fa54d07399
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vluxseg8ei16.c
@@ -0,0 +1,77 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vluxseg8ei16_v_bf16mf4x8(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vluxseg8.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], <vscale x 1 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x8_t test_vluxseg8ei16_v_bf16mf4x8(const __bf16 *rs1,
+                                               vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vluxseg8ei16(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vluxseg8ei16_v_bf16mf2x8(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vluxseg8.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], <vscale x 2 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x8_t test_vluxseg8ei16_v_bf16mf2x8(const __bf16 *rs1,
+                                               vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vluxseg8ei16(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vluxseg8ei16_v_bf16m1x8(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vluxseg8.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], <vscale x 4 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x8_t test_vluxseg8ei16_v_bf16m1x8(const __bf16 *rs1, vuint16m1_t rs2,
+                                             size_t vl) {
+  return __riscv_vluxseg8ei16(rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vluxseg8ei16_v_bf16mf4x8_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vluxseg8.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, <vscale x 1 x bfloat> poison, ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x8_t test_vluxseg8ei16_v_bf16mf4x8_m(vbool64_t vm,
+                                                 const __bf16 *rs1,
+                                                 vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vluxseg8ei16(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vluxseg8ei16_v_bf16mf2x8_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vluxseg8.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, <vscale x 2 x bfloat> poison, ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x8_t test_vluxseg8ei16_v_bf16mf2x8_m(vbool32_t vm,
+                                                 const __bf16 *rs1,
+                                                 vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vluxseg8ei16(vm, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vluxseg8ei16_v_bf16m1x8_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vluxseg8.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, <vscale x 4 x bfloat> poison, ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 3)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x8_t test_vluxseg8ei16_v_bf16m1x8_m(vbool16_t vm, const __bf16 *rs1,
+                                               vuint16m1_t rs2, size_t vl) {
+  return __riscv_vluxseg8ei16(vm, rs1, rs2, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vreinterpret.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vreinterpret.c
new file mode 100644
index 000000000000..2e14f42fa346
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vreinterpret.c
@@ -0,0 +1,249 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 1 x bfloat> @test_vreinterpret_v_i16mf4_bf16mf4(
+// CHECK-RV64-SAME: <vscale x 1 x i16> [[SRC:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = bitcast <vscale x 1 x i16> [[SRC]] to <vscale x 1 x bfloat>
+// CHECK-RV64-NEXT:    ret <vscale x 1 x bfloat> [[TMP0]]
+//
+vbfloat16mf4_t test_vreinterpret_v_i16mf4_bf16mf4(vint16mf4_t src) {
+  return __riscv_vreinterpret_bf16mf4(src);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 2 x bfloat> @test_vreinterpret_v_i16mf2_bf16mf2(
+// CHECK-RV64-SAME: <vscale x 2 x i16> [[SRC:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = bitcast <vscale x 2 x i16> [[SRC]] to <vscale x 2 x bfloat>
+// CHECK-RV64-NEXT:    ret <vscale x 2 x bfloat> [[TMP0]]
+//
+vbfloat16mf2_t test_vreinterpret_v_i16mf2_bf16mf2(vint16mf2_t src) {
+  return __riscv_vreinterpret_bf16mf2(src);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 4 x bfloat> @test_vreinterpret_v_i16m1_bf16m1(
+// CHECK-RV64-SAME: <vscale x 4 x i16> [[SRC:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = bitcast <vscale x 4 x i16> [[SRC]] to <vscale x 4 x bfloat>
+// CHECK-RV64-NEXT:    ret <vscale x 4 x bfloat> [[TMP0]]
+//
+vbfloat16m1_t test_vreinterpret_v_i16m1_bf16m1(vint16m1_t src) {
+  return __riscv_vreinterpret_bf16m1(src);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 8 x bfloat> @test_vreinterpret_v_i16m2_bf16m2(
+// CHECK-RV64-SAME: <vscale x 8 x i16> [[SRC:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = bitcast <vscale x 8 x i16> [[SRC]] to <vscale x 8 x bfloat>
+// CHECK-RV64-NEXT:    ret <vscale x 8 x bfloat> [[TMP0]]
+//
+vbfloat16m2_t test_vreinterpret_v_i16m2_bf16m2(vint16m2_t src) {
+  return __riscv_vreinterpret_bf16m2(src);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 16 x bfloat> @test_vreinterpret_v_i16m4_bf16m4(
+// CHECK-RV64-SAME: <vscale x 16 x i16> [[SRC:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = bitcast <vscale x 16 x i16> [[SRC]] to <vscale x 16 x bfloat>
+// CHECK-RV64-NEXT:    ret <vscale x 16 x bfloat> [[TMP0]]
+//
+vbfloat16m4_t test_vreinterpret_v_i16m4_bf16m4(vint16m4_t src) {
+  return __riscv_vreinterpret_bf16m4(src);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 32 x bfloat> @test_vreinterpret_v_i16m8_bf16m8(
+// CHECK-RV64-SAME: <vscale x 32 x i16> [[SRC:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = bitcast <vscale x 32 x i16> [[SRC]] to <vscale x 32 x bfloat>
+// CHECK-RV64-NEXT:    ret <vscale x 32 x bfloat> [[TMP0]]
+//
+vbfloat16m8_t test_vreinterpret_v_i16m8_bf16m8(vint16m8_t src) {
+  return __riscv_vreinterpret_bf16m8(src);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 1 x bfloat> @test_vreinterpret_v_u16mf4_bf16mf4(
+// CHECK-RV64-SAME: <vscale x 1 x i16> [[SRC:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = bitcast <vscale x 1 x i16> [[SRC]] to <vscale x 1 x bfloat>
+// CHECK-RV64-NEXT:    ret <vscale x 1 x bfloat> [[TMP0]]
+//
+vbfloat16mf4_t test_vreinterpret_v_u16mf4_bf16mf4(vuint16mf4_t src) {
+  return __riscv_vreinterpret_bf16mf4(src);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 2 x bfloat> @test_vreinterpret_v_u16mf2_bf16mf2(
+// CHECK-RV64-SAME: <vscale x 2 x i16> [[SRC:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = bitcast <vscale x 2 x i16> [[SRC]] to <vscale x 2 x bfloat>
+// CHECK-RV64-NEXT:    ret <vscale x 2 x bfloat> [[TMP0]]
+//
+vbfloat16mf2_t test_vreinterpret_v_u16mf2_bf16mf2(vuint16mf2_t src) {
+  return __riscv_vreinterpret_bf16mf2(src);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 4 x bfloat> @test_vreinterpret_v_u16m1_bf16m1(
+// CHECK-RV64-SAME: <vscale x 4 x i16> [[SRC:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = bitcast <vscale x 4 x i16> [[SRC]] to <vscale x 4 x bfloat>
+// CHECK-RV64-NEXT:    ret <vscale x 4 x bfloat> [[TMP0]]
+//
+vbfloat16m1_t test_vreinterpret_v_u16m1_bf16m1(vuint16m1_t src) {
+  return __riscv_vreinterpret_bf16m1(src);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 8 x bfloat> @test_vreinterpret_v_u16m2_bf16m2(
+// CHECK-RV64-SAME: <vscale x 8 x i16> [[SRC:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = bitcast <vscale x 8 x i16> [[SRC]] to <vscale x 8 x bfloat>
+// CHECK-RV64-NEXT:    ret <vscale x 8 x bfloat> [[TMP0]]
+//
+vbfloat16m2_t test_vreinterpret_v_u16m2_bf16m2(vuint16m2_t src) {
+  return __riscv_vreinterpret_bf16m2(src);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 16 x bfloat> @test_vreinterpret_v_u16m4_bf16m4(
+// CHECK-RV64-SAME: <vscale x 16 x i16> [[SRC:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = bitcast <vscale x 16 x i16> [[SRC]] to <vscale x 16 x bfloat>
+// CHECK-RV64-NEXT:    ret <vscale x 16 x bfloat> [[TMP0]]
+//
+vbfloat16m4_t test_vreinterpret_v_u16m4_bf16m4(vuint16m4_t src) {
+  return __riscv_vreinterpret_bf16m4(src);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 32 x bfloat> @test_vreinterpret_v_u16m8_bf16m8(
+// CHECK-RV64-SAME: <vscale x 32 x i16> [[SRC:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = bitcast <vscale x 32 x i16> [[SRC]] to <vscale x 32 x bfloat>
+// CHECK-RV64-NEXT:    ret <vscale x 32 x bfloat> [[TMP0]]
+//
+vbfloat16m8_t test_vreinterpret_v_u16m8_bf16m8(vuint16m8_t src) {
+  return __riscv_vreinterpret_bf16m8(src);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 1 x i16> @test_vreinterpret_v_bf16mf4_i16mf4(
+// CHECK-RV64-SAME: <vscale x 1 x bfloat> [[SRC:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = bitcast <vscale x 1 x bfloat> [[SRC]] to <vscale x 1 x i16>
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
+//
+vint16mf4_t test_vreinterpret_v_bf16mf4_i16mf4(vbfloat16mf4_t src) {
+  return __riscv_vreinterpret_i16mf4(src);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 2 x i16> @test_vreinterpret_v_bf16mf2_i16mf2(
+// CHECK-RV64-SAME: <vscale x 2 x bfloat> [[SRC:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = bitcast <vscale x 2 x bfloat> [[SRC]] to <vscale x 2 x i16>
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
+//
+vint16mf2_t test_vreinterpret_v_bf16mf2_i16mf2(vbfloat16mf2_t src) {
+  return __riscv_vreinterpret_i16mf2(src);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 4 x i16> @test_vreinterpret_v_bf16m1_i16m1(
+// CHECK-RV64-SAME: <vscale x 4 x bfloat> [[SRC:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = bitcast <vscale x 4 x bfloat> [[SRC]] to <vscale x 4 x i16>
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
+//
+vint16m1_t test_vreinterpret_v_bf16m1_i16m1(vbfloat16m1_t src) {
+  return __riscv_vreinterpret_i16m1(src);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 8 x i16> @test_vreinterpret_v_bf16m2_i16m2(
+// CHECK-RV64-SAME: <vscale x 8 x bfloat> [[SRC:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = bitcast <vscale x 8 x bfloat> [[SRC]] to <vscale x 8 x i16>
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
+//
+vint16m2_t test_vreinterpret_v_bf16m2_i16m2(vbfloat16m2_t src) {
+  return __riscv_vreinterpret_i16m2(src);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 16 x i16> @test_vreinterpret_v_bf16m4_i16m4(
+// CHECK-RV64-SAME: <vscale x 16 x bfloat> [[SRC:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = bitcast <vscale x 16 x bfloat> [[SRC]] to <vscale x 16 x i16>
+// CHECK-RV64-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
+//
+vint16m4_t test_vreinterpret_v_bf16m4_i16m4(vbfloat16m4_t src) {
+  return __riscv_vreinterpret_i16m4(src);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 32 x i16> @test_vreinterpret_v_bf16m8_i16m8(
+// CHECK-RV64-SAME: <vscale x 32 x bfloat> [[SRC:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = bitcast <vscale x 32 x bfloat> [[SRC]] to <vscale x 32 x i16>
+// CHECK-RV64-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
+//
+vint16m8_t test_vreinterpret_v_bf16m8_i16m8(vbfloat16m8_t src) {
+  return __riscv_vreinterpret_i16m8(src);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 1 x i16> @test_vreinterpret_v_bf16mf4_u16mf4(
+// CHECK-RV64-SAME: <vscale x 1 x bfloat> [[SRC:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = bitcast <vscale x 1 x bfloat> [[SRC]] to <vscale x 1 x i16>
+// CHECK-RV64-NEXT:    ret <vscale x 1 x i16> [[TMP0]]
+//
+vuint16mf4_t test_vreinterpret_v_bf16mf4_u16mf4(vbfloat16mf4_t src) {
+  return __riscv_vreinterpret_u16mf4(src);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 2 x i16> @test_vreinterpret_v_bf16mf2_u16mf2(
+// CHECK-RV64-SAME: <vscale x 2 x bfloat> [[SRC:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = bitcast <vscale x 2 x bfloat> [[SRC]] to <vscale x 2 x i16>
+// CHECK-RV64-NEXT:    ret <vscale x 2 x i16> [[TMP0]]
+//
+vuint16mf2_t test_vreinterpret_v_bf16mf2_u16mf2(vbfloat16mf2_t src) {
+  return __riscv_vreinterpret_u16mf2(src);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 4 x i16> @test_vreinterpret_v_bf16m1_u16m1(
+// CHECK-RV64-SAME: <vscale x 4 x bfloat> [[SRC:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = bitcast <vscale x 4 x bfloat> [[SRC]] to <vscale x 4 x i16>
+// CHECK-RV64-NEXT:    ret <vscale x 4 x i16> [[TMP0]]
+//
+vuint16m1_t test_vreinterpret_v_bf16m1_u16m1(vbfloat16m1_t src) {
+  return __riscv_vreinterpret_u16m1(src);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 8 x i16> @test_vreinterpret_v_bf16m2_u16m2(
+// CHECK-RV64-SAME: <vscale x 8 x bfloat> [[SRC:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = bitcast <vscale x 8 x bfloat> [[SRC]] to <vscale x 8 x i16>
+// CHECK-RV64-NEXT:    ret <vscale x 8 x i16> [[TMP0]]
+//
+vuint16m2_t test_vreinterpret_v_bf16m2_u16m2(vbfloat16m2_t src) {
+  return __riscv_vreinterpret_u16m2(src);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 16 x i16> @test_vreinterpret_v_bf16m4_u16m4(
+// CHECK-RV64-SAME: <vscale x 16 x bfloat> [[SRC:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = bitcast <vscale x 16 x bfloat> [[SRC]] to <vscale x 16 x i16>
+// CHECK-RV64-NEXT:    ret <vscale x 16 x i16> [[TMP0]]
+//
+vuint16m4_t test_vreinterpret_v_bf16m4_u16m4(vbfloat16m4_t src) {
+  return __riscv_vreinterpret_u16m4(src);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 32 x i16> @test_vreinterpret_v_bf16m8_u16m8(
+// CHECK-RV64-SAME: <vscale x 32 x bfloat> [[SRC:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = bitcast <vscale x 32 x bfloat> [[SRC]] to <vscale x 32 x i16>
+// CHECK-RV64-NEXT:    ret <vscale x 32 x i16> [[TMP0]]
+//
+vuint16m8_t test_vreinterpret_v_bf16m8_u16m8(vbfloat16m8_t src) {
+  return __riscv_vreinterpret_u16m8(src);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vse16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vse16.c
new file mode 100644
index 000000000000..0fe9076b1f10
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vse16.c
@@ -0,0 +1,135 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local void @test_vse16_v_bf16mf4(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 1 x bfloat> [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vse.nxv1bf16.i64(<vscale x 1 x bfloat> [[VS3]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vse16_v_bf16mf4(__bf16 *rs1, vbfloat16mf4_t vs3, size_t vl) {
+  return __riscv_vse16(rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vse16_v_bf16mf2(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 2 x bfloat> [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vse.nxv2bf16.i64(<vscale x 2 x bfloat> [[VS3]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vse16_v_bf16mf2(__bf16 *rs1, vbfloat16mf2_t vs3, size_t vl) {
+  return __riscv_vse16(rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vse16_v_bf16m1(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 4 x bfloat> [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vse.nxv4bf16.i64(<vscale x 4 x bfloat> [[VS3]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vse16_v_bf16m1(__bf16 *rs1, vbfloat16m1_t vs3, size_t vl) {
+  return __riscv_vse16(rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vse16_v_bf16m2(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 8 x bfloat> [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vse.nxv8bf16.i64(<vscale x 8 x bfloat> [[VS3]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vse16_v_bf16m2(__bf16 *rs1, vbfloat16m2_t vs3, size_t vl) {
+  return __riscv_vse16(rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vse16_v_bf16m4(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 16 x bfloat> [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vse.nxv16bf16.i64(<vscale x 16 x bfloat> [[VS3]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vse16_v_bf16m4(__bf16 *rs1, vbfloat16m4_t vs3, size_t vl) {
+  return __riscv_vse16(rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vse16_v_bf16m8(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 32 x bfloat> [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vse.nxv32bf16.i64(<vscale x 32 x bfloat> [[VS3]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vse16_v_bf16m8(__bf16 *rs1, vbfloat16m8_t vs3, size_t vl) {
+  return __riscv_vse16(rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vse16_v_bf16mf4_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x bfloat> [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vse.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[VS3]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vse16_v_bf16mf4_m(vbool64_t vm, __bf16 *rs1, vbfloat16mf4_t vs3,
+                            size_t vl) {
+  return __riscv_vse16(vm, rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vse16_v_bf16mf2_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x bfloat> [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vse.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[VS3]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vse16_v_bf16mf2_m(vbool32_t vm, __bf16 *rs1, vbfloat16mf2_t vs3,
+                            size_t vl) {
+  return __riscv_vse16(vm, rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vse16_v_bf16m1_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x bfloat> [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vse.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[VS3]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vse16_v_bf16m1_m(vbool16_t vm, __bf16 *rs1, vbfloat16m1_t vs3,
+                           size_t vl) {
+  return __riscv_vse16(vm, rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vse16_v_bf16m2_m(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 8 x bfloat> [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vse.mask.nxv8bf16.i64(<vscale x 8 x bfloat> [[VS3]], ptr [[RS1]], <vscale x 8 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vse16_v_bf16m2_m(vbool8_t vm, __bf16 *rs1, vbfloat16m2_t vs3,
+                           size_t vl) {
+  return __riscv_vse16(vm, rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vse16_v_bf16m4_m(
+// CHECK-RV64-SAME: <vscale x 16 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 16 x bfloat> [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vse.mask.nxv16bf16.i64(<vscale x 16 x bfloat> [[VS3]], ptr [[RS1]], <vscale x 16 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vse16_v_bf16m4_m(vbool4_t vm, __bf16 *rs1, vbfloat16m4_t vs3,
+                           size_t vl) {
+  return __riscv_vse16(vm, rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vse16_v_bf16m8_m(
+// CHECK-RV64-SAME: <vscale x 32 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 32 x bfloat> [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vse.mask.nxv32bf16.i64(<vscale x 32 x bfloat> [[VS3]], ptr [[RS1]], <vscale x 32 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vse16_v_bf16m8_m(vbool2_t vm, __bf16 *rs1, vbfloat16m8_t vs3,
+                           size_t vl) {
+  return __riscv_vse16(vm, rs1, vs3, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vset.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vset.c
new file mode 100644
index 000000000000..67fb436bf22a
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vset.c
@@ -0,0 +1,364 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 8 x bfloat> @test_vset_v_bf16m1_bf16m2(
+// CHECK-RV64-SAME: <vscale x 8 x bfloat> [[DEST:%.*]], i64 noundef [[INDEX:%.*]], <vscale x 4 x bfloat> [[VALUE:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x bfloat> @llvm.vector.insert.nxv8bf16.nxv4bf16(<vscale x 8 x bfloat> [[DEST]], <vscale x 4 x bfloat> [[VALUE]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 8 x bfloat> [[TMP0]]
+//
+vbfloat16m2_t test_vset_v_bf16m1_bf16m2(vbfloat16m2_t dest, size_t index,
+                                        vbfloat16m1_t value) {
+  return __riscv_vset(dest, 0, value);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 16 x bfloat> @test_vset_v_bf16m1_bf16m4(
+// CHECK-RV64-SAME: <vscale x 16 x bfloat> [[DEST:%.*]], i64 noundef [[INDEX:%.*]], <vscale x 4 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x bfloat> @llvm.vector.insert.nxv16bf16.nxv4bf16(<vscale x 16 x bfloat> [[DEST]], <vscale x 4 x bfloat> [[VALUE]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 16 x bfloat> [[TMP0]]
+//
+vbfloat16m4_t test_vset_v_bf16m1_bf16m4(vbfloat16m4_t dest, size_t index,
+                                        vbfloat16m1_t value) {
+  return __riscv_vset(dest, 0, value);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 16 x bfloat> @test_vset_v_bf16m2_bf16m4(
+// CHECK-RV64-SAME: <vscale x 16 x bfloat> [[DEST:%.*]], i64 noundef [[INDEX:%.*]], <vscale x 8 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x bfloat> @llvm.vector.insert.nxv16bf16.nxv8bf16(<vscale x 16 x bfloat> [[DEST]], <vscale x 8 x bfloat> [[VALUE]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 16 x bfloat> [[TMP0]]
+//
+vbfloat16m4_t test_vset_v_bf16m2_bf16m4(vbfloat16m4_t dest, size_t index,
+                                        vbfloat16m2_t value) {
+  return __riscv_vset(dest, 0, value);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 32 x bfloat> @test_vset_v_bf16m1_bf16m8(
+// CHECK-RV64-SAME: <vscale x 32 x bfloat> [[DEST:%.*]], i64 noundef [[INDEX:%.*]], <vscale x 4 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x bfloat> @llvm.vector.insert.nxv32bf16.nxv4bf16(<vscale x 32 x bfloat> [[DEST]], <vscale x 4 x bfloat> [[VALUE]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 32 x bfloat> [[TMP0]]
+//
+vbfloat16m8_t test_vset_v_bf16m1_bf16m8(vbfloat16m8_t dest, size_t index,
+                                        vbfloat16m1_t value) {
+  return __riscv_vset(dest, 0, value);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 32 x bfloat> @test_vset_v_bf16m2_bf16m8(
+// CHECK-RV64-SAME: <vscale x 32 x bfloat> [[DEST:%.*]], i64 noundef [[INDEX:%.*]], <vscale x 8 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x bfloat> @llvm.vector.insert.nxv32bf16.nxv8bf16(<vscale x 32 x bfloat> [[DEST]], <vscale x 8 x bfloat> [[VALUE]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 32 x bfloat> [[TMP0]]
+//
+vbfloat16m8_t test_vset_v_bf16m2_bf16m8(vbfloat16m8_t dest, size_t index,
+                                        vbfloat16m2_t value) {
+  return __riscv_vset(dest, 0, value);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 32 x bfloat> @test_vset_v_bf16m4_bf16m8(
+// CHECK-RV64-SAME: <vscale x 32 x bfloat> [[DEST:%.*]], i64 noundef [[INDEX:%.*]], <vscale x 16 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x bfloat> @llvm.vector.insert.nxv32bf16.nxv16bf16(<vscale x 32 x bfloat> [[DEST]], <vscale x 16 x bfloat> [[VALUE]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 32 x bfloat> [[TMP0]]
+//
+vbfloat16m8_t test_vset_v_bf16m4_bf16m8(vbfloat16m8_t dest, size_t index,
+                                        vbfloat16m4_t value) {
+  return __riscv_vset(dest, 0, value);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vset_v_bf16mf4_bf16mf4x2(
+// CHECK-RV64-SAME: { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[DEST:%.*]], i64 noundef [[INDEX:%.*]], <vscale x 1 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[DEST]], <vscale x 1 x bfloat> [[VALUE]], 0
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x2_t test_vset_v_bf16mf4_bf16mf4x2(vbfloat16mf4x2_t dest,
+                                               size_t index,
+                                               vbfloat16mf4_t value) {
+  return __riscv_vset(dest, 0, value);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vset_v_bf16mf4_bf16mf4x3(
+// CHECK-RV64-SAME: { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[DEST:%.*]], i64 noundef [[INDEX:%.*]], <vscale x 1 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[DEST]], <vscale x 1 x bfloat> [[VALUE]], 0
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x3_t test_vset_v_bf16mf4_bf16mf4x3(vbfloat16mf4x3_t dest,
+                                               size_t index,
+                                               vbfloat16mf4_t value) {
+  return __riscv_vset(dest, 0, value);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vset_v_bf16mf4_bf16mf4x4(
+// CHECK-RV64-SAME: { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[DEST:%.*]], i64 noundef [[INDEX:%.*]], <vscale x 1 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[DEST]], <vscale x 1 x bfloat> [[VALUE]], 0
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x4_t test_vset_v_bf16mf4_bf16mf4x4(vbfloat16mf4x4_t dest,
+                                               size_t index,
+                                               vbfloat16mf4_t value) {
+  return __riscv_vset(dest, 0, value);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vset_v_bf16mf4_bf16mf4x5(
+// CHECK-RV64-SAME: { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[DEST:%.*]], i64 noundef [[INDEX:%.*]], <vscale x 1 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[DEST]], <vscale x 1 x bfloat> [[VALUE]], 0
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x5_t test_vset_v_bf16mf4_bf16mf4x5(vbfloat16mf4x5_t dest,
+                                               size_t index,
+                                               vbfloat16mf4_t value) {
+  return __riscv_vset(dest, 0, value);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vset_v_bf16mf4_bf16mf4x6(
+// CHECK-RV64-SAME: { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[DEST:%.*]], i64 noundef [[INDEX:%.*]], <vscale x 1 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[DEST]], <vscale x 1 x bfloat> [[VALUE]], 0
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x6_t test_vset_v_bf16mf4_bf16mf4x6(vbfloat16mf4x6_t dest,
+                                               size_t index,
+                                               vbfloat16mf4_t value) {
+  return __riscv_vset(dest, 0, value);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vset_v_bf16mf4_bf16mf4x7(
+// CHECK-RV64-SAME: { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[DEST:%.*]], i64 noundef [[INDEX:%.*]], <vscale x 1 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[DEST]], <vscale x 1 x bfloat> [[VALUE]], 0
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x7_t test_vset_v_bf16mf4_bf16mf4x7(vbfloat16mf4x7_t dest,
+                                               size_t index,
+                                               vbfloat16mf4_t value) {
+  return __riscv_vset(dest, 0, value);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vset_v_bf16mf4_bf16mf4x8(
+// CHECK-RV64-SAME: { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[DEST:%.*]], i64 noundef [[INDEX:%.*]], <vscale x 1 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[DEST]], <vscale x 1 x bfloat> [[VALUE]], 0
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP0]]
+//
+vbfloat16mf4x8_t test_vset_v_bf16mf4_bf16mf4x8(vbfloat16mf4x8_t dest,
+                                               size_t index,
+                                               vbfloat16mf4_t value) {
+  return __riscv_vset(dest, 0, value);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vset_v_bf16mf2_bf16mf2x2(
+// CHECK-RV64-SAME: { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[DEST:%.*]], i64 noundef [[INDEX:%.*]], <vscale x 2 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[DEST]], <vscale x 2 x bfloat> [[VALUE]], 0
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x2_t test_vset_v_bf16mf2_bf16mf2x2(vbfloat16mf2x2_t dest,
+                                               size_t index,
+                                               vbfloat16mf2_t value) {
+  return __riscv_vset(dest, 0, value);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vset_v_bf16mf2_bf16mf2x3(
+// CHECK-RV64-SAME: { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[DEST:%.*]], i64 noundef [[INDEX:%.*]], <vscale x 2 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[DEST]], <vscale x 2 x bfloat> [[VALUE]], 0
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x3_t test_vset_v_bf16mf2_bf16mf2x3(vbfloat16mf2x3_t dest,
+                                               size_t index,
+                                               vbfloat16mf2_t value) {
+  return __riscv_vset(dest, 0, value);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vset_v_bf16mf2_bf16mf2x4(
+// CHECK-RV64-SAME: { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[DEST:%.*]], i64 noundef [[INDEX:%.*]], <vscale x 2 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[DEST]], <vscale x 2 x bfloat> [[VALUE]], 0
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x4_t test_vset_v_bf16mf2_bf16mf2x4(vbfloat16mf2x4_t dest,
+                                               size_t index,
+                                               vbfloat16mf2_t value) {
+  return __riscv_vset(dest, 0, value);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vset_v_bf16mf2_bf16mf2x5(
+// CHECK-RV64-SAME: { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[DEST:%.*]], i64 noundef [[INDEX:%.*]], <vscale x 2 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[DEST]], <vscale x 2 x bfloat> [[VALUE]], 0
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x5_t test_vset_v_bf16mf2_bf16mf2x5(vbfloat16mf2x5_t dest,
+                                               size_t index,
+                                               vbfloat16mf2_t value) {
+  return __riscv_vset(dest, 0, value);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vset_v_bf16mf2_bf16mf2x6(
+// CHECK-RV64-SAME: { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[DEST:%.*]], i64 noundef [[INDEX:%.*]], <vscale x 2 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[DEST]], <vscale x 2 x bfloat> [[VALUE]], 0
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x6_t test_vset_v_bf16mf2_bf16mf2x6(vbfloat16mf2x6_t dest,
+                                               size_t index,
+                                               vbfloat16mf2_t value) {
+  return __riscv_vset(dest, 0, value);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vset_v_bf16mf2_bf16mf2x7(
+// CHECK-RV64-SAME: { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[DEST:%.*]], i64 noundef [[INDEX:%.*]], <vscale x 2 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[DEST]], <vscale x 2 x bfloat> [[VALUE]], 0
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x7_t test_vset_v_bf16mf2_bf16mf2x7(vbfloat16mf2x7_t dest,
+                                               size_t index,
+                                               vbfloat16mf2_t value) {
+  return __riscv_vset(dest, 0, value);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vset_v_bf16mf2_bf16mf2x8(
+// CHECK-RV64-SAME: { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[DEST:%.*]], i64 noundef [[INDEX:%.*]], <vscale x 2 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[DEST]], <vscale x 2 x bfloat> [[VALUE]], 0
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP0]]
+//
+vbfloat16mf2x8_t test_vset_v_bf16mf2_bf16mf2x8(vbfloat16mf2x8_t dest,
+                                               size_t index,
+                                               vbfloat16mf2_t value) {
+  return __riscv_vset(dest, 0, value);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vset_v_bf16m1_bf16m1x2(
+// CHECK-RV64-SAME: { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[DEST:%.*]], i64 noundef [[INDEX:%.*]], <vscale x 4 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[DEST]], <vscale x 4 x bfloat> [[VALUE]], 0
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x2_t test_vset_v_bf16m1_bf16m1x2(vbfloat16m1x2_t dest, size_t index,
+                                            vbfloat16m1_t value) {
+  return __riscv_vset(dest, 0, value);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vset_v_bf16m1_bf16m1x3(
+// CHECK-RV64-SAME: { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[DEST:%.*]], i64 noundef [[INDEX:%.*]], <vscale x 4 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[DEST]], <vscale x 4 x bfloat> [[VALUE]], 0
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x3_t test_vset_v_bf16m1_bf16m1x3(vbfloat16m1x3_t dest, size_t index,
+                                            vbfloat16m1_t value) {
+  return __riscv_vset(dest, 0, value);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vset_v_bf16m1_bf16m1x4(
+// CHECK-RV64-SAME: { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[DEST:%.*]], i64 noundef [[INDEX:%.*]], <vscale x 4 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[DEST]], <vscale x 4 x bfloat> [[VALUE]], 0
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x4_t test_vset_v_bf16m1_bf16m1x4(vbfloat16m1x4_t dest, size_t index,
+                                            vbfloat16m1_t value) {
+  return __riscv_vset(dest, 0, value);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vset_v_bf16m1_bf16m1x5(
+// CHECK-RV64-SAME: { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[DEST:%.*]], i64 noundef [[INDEX:%.*]], <vscale x 4 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[DEST]], <vscale x 4 x bfloat> [[VALUE]], 0
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x5_t test_vset_v_bf16m1_bf16m1x5(vbfloat16m1x5_t dest, size_t index,
+                                            vbfloat16m1_t value) {
+  return __riscv_vset(dest, 0, value);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vset_v_bf16m1_bf16m1x6(
+// CHECK-RV64-SAME: { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[DEST:%.*]], i64 noundef [[INDEX:%.*]], <vscale x 4 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[DEST]], <vscale x 4 x bfloat> [[VALUE]], 0
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x6_t test_vset_v_bf16m1_bf16m1x6(vbfloat16m1x6_t dest, size_t index,
+                                            vbfloat16m1_t value) {
+  return __riscv_vset(dest, 0, value);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vset_v_bf16m1_bf16m1x7(
+// CHECK-RV64-SAME: { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[DEST:%.*]], i64 noundef [[INDEX:%.*]], <vscale x 4 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[DEST]], <vscale x 4 x bfloat> [[VALUE]], 0
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x7_t test_vset_v_bf16m1_bf16m1x7(vbfloat16m1x7_t dest, size_t index,
+                                            vbfloat16m1_t value) {
+  return __riscv_vset(dest, 0, value);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vset_v_bf16m1_bf16m1x8(
+// CHECK-RV64-SAME: { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[DEST:%.*]], i64 noundef [[INDEX:%.*]], <vscale x 4 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[DEST]], <vscale x 4 x bfloat> [[VALUE]], 0
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP0]]
+//
+vbfloat16m1x8_t test_vset_v_bf16m1_bf16m1x8(vbfloat16m1x8_t dest, size_t index,
+                                            vbfloat16m1_t value) {
+  return __riscv_vset(dest, 0, value);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vset_v_bf16m2_bf16m2x2(
+// CHECK-RV64-SAME: { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[DEST:%.*]], i64 noundef [[INDEX:%.*]], <vscale x 8 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[DEST]], <vscale x 8 x bfloat> [[VALUE]], 0
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP0]]
+//
+vbfloat16m2x2_t test_vset_v_bf16m2_bf16m2x2(vbfloat16m2x2_t dest, size_t index,
+                                            vbfloat16m2_t value) {
+  return __riscv_vset(dest, 0, value);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vset_v_bf16m2_bf16m2x3(
+// CHECK-RV64-SAME: { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[DEST:%.*]], i64 noundef [[INDEX:%.*]], <vscale x 8 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[DEST]], <vscale x 8 x bfloat> [[VALUE]], 0
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP0]]
+//
+vbfloat16m2x3_t test_vset_v_bf16m2_bf16m2x3(vbfloat16m2x3_t dest, size_t index,
+                                            vbfloat16m2_t value) {
+  return __riscv_vset(dest, 0, value);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vset_v_bf16m2_bf16m2x4(
+// CHECK-RV64-SAME: { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[DEST:%.*]], i64 noundef [[INDEX:%.*]], <vscale x 8 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[DEST]], <vscale x 8 x bfloat> [[VALUE]], 0
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP0]]
+//
+vbfloat16m2x4_t test_vset_v_bf16m2_bf16m2x4(vbfloat16m2x4_t dest, size_t index,
+                                            vbfloat16m2_t value) {
+  return __riscv_vset(dest, 0, value);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @test_vset_v_bf16m4_bf16m4x2(
+// CHECK-RV64-SAME: { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[DEST:%.*]], i64 noundef [[INDEX:%.*]], <vscale x 16 x bfloat> [[VALUE:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = insertvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[DEST]], <vscale x 16 x bfloat> [[VALUE]], 0
+// CHECK-RV64-NEXT:    ret { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[TMP0]]
+//
+vbfloat16m4x2_t test_vset_v_bf16m4_bf16m4x2(vbfloat16m4x2_t dest, size_t index,
+                                            vbfloat16m4_t value) {
+  return __riscv_vset(dest, 0, value);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vsoxei16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vsoxei16.c
new file mode 100644
index 000000000000..a374e019665c
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vsoxei16.c
@@ -0,0 +1,141 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxei16_v_bf16mf4(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], <vscale x 1 x bfloat> [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxei.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[VS3]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxei16_v_bf16mf4(__bf16 *rs1, vuint16mf4_t rs2, vbfloat16mf4_t vs3,
+                             size_t vl) {
+  return __riscv_vsoxei16(rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxei16_v_bf16mf2(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], <vscale x 2 x bfloat> [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxei.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[VS3]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxei16_v_bf16mf2(__bf16 *rs1, vuint16mf2_t rs2, vbfloat16mf2_t vs3,
+                             size_t vl) {
+  return __riscv_vsoxei16(rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxei16_v_bf16m1(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], <vscale x 4 x bfloat> [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxei.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[VS3]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxei16_v_bf16m1(__bf16 *rs1, vuint16m1_t rs2, vbfloat16m1_t vs3,
+                            size_t vl) {
+  return __riscv_vsoxei16(rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxei16_v_bf16m2(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[RS2:%.*]], <vscale x 8 x bfloat> [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxei.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> [[VS3]], ptr [[RS1]], <vscale x 8 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxei16_v_bf16m2(__bf16 *rs1, vuint16m2_t rs2, vbfloat16m2_t vs3,
+                            size_t vl) {
+  return __riscv_vsoxei16(rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxei16_v_bf16m4(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 16 x i16> [[RS2:%.*]], <vscale x 16 x bfloat> [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxei.nxv16bf16.nxv16i16.i64(<vscale x 16 x bfloat> [[VS3]], ptr [[RS1]], <vscale x 16 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxei16_v_bf16m4(__bf16 *rs1, vuint16m4_t rs2, vbfloat16m4_t vs3,
+                            size_t vl) {
+  return __riscv_vsoxei16(rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxei16_v_bf16m8(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 32 x i16> [[RS2:%.*]], <vscale x 32 x bfloat> [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxei.nxv32bf16.nxv32i16.i64(<vscale x 32 x bfloat> [[VS3]], ptr [[RS1]], <vscale x 32 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxei16_v_bf16m8(__bf16 *rs1, vuint16m8_t rs2, vbfloat16m8_t vs3,
+                            size_t vl) {
+  return __riscv_vsoxei16(rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxei16_v_bf16mf4_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], <vscale x 1 x bfloat> [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxei.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[VS3]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxei16_v_bf16mf4_m(vbool64_t vm, __bf16 *rs1, vuint16mf4_t rs2,
+                               vbfloat16mf4_t vs3, size_t vl) {
+  return __riscv_vsoxei16(vm, rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxei16_v_bf16mf2_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], <vscale x 2 x bfloat> [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxei.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[VS3]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxei16_v_bf16mf2_m(vbool32_t vm, __bf16 *rs1, vuint16mf2_t rs2,
+                               vbfloat16mf2_t vs3, size_t vl) {
+  return __riscv_vsoxei16(vm, rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxei16_v_bf16m1_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], <vscale x 4 x bfloat> [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxei.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[VS3]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxei16_v_bf16m1_m(vbool16_t vm, __bf16 *rs1, vuint16m1_t rs2,
+                              vbfloat16m1_t vs3, size_t vl) {
+  return __riscv_vsoxei16(vm, rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxei16_v_bf16m2_m(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[RS2:%.*]], <vscale x 8 x bfloat> [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxei.mask.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> [[VS3]], ptr [[RS1]], <vscale x 8 x i16> [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxei16_v_bf16m2_m(vbool8_t vm, __bf16 *rs1, vuint16m2_t rs2,
+                              vbfloat16m2_t vs3, size_t vl) {
+  return __riscv_vsoxei16(vm, rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxei16_v_bf16m4_m(
+// CHECK-RV64-SAME: <vscale x 16 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 16 x i16> [[RS2:%.*]], <vscale x 16 x bfloat> [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxei.mask.nxv16bf16.nxv16i16.i64(<vscale x 16 x bfloat> [[VS3]], ptr [[RS1]], <vscale x 16 x i16> [[RS2]], <vscale x 16 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxei16_v_bf16m4_m(vbool4_t vm, __bf16 *rs1, vuint16m4_t rs2,
+                              vbfloat16m4_t vs3, size_t vl) {
+  return __riscv_vsoxei16(vm, rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxei16_v_bf16m8_m(
+// CHECK-RV64-SAME: <vscale x 32 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 32 x i16> [[RS2:%.*]], <vscale x 32 x bfloat> [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxei.mask.nxv32bf16.nxv32i16.i64(<vscale x 32 x bfloat> [[VS3]], ptr [[RS1]], <vscale x 32 x i16> [[RS2]], <vscale x 32 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxei16_v_bf16m8_m(vbool2_t vm, __bf16 *rs1, vuint16m8_t rs2,
+                              vbfloat16m8_t vs3, size_t vl) {
+  return __riscv_vsoxei16(vm, rs1, rs2, vs3, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vsoxseg2ei16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vsoxseg2ei16.c
new file mode 100644
index 000000000000..088ef855cd1c
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vsoxseg2ei16.c
@@ -0,0 +1,141 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxseg2ei16_v_bf16mf4x2(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[VS2:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxseg2.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 1 x i16> [[VS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxseg2ei16_v_bf16mf4x2(__bf16 *rs1, vuint16mf4_t vs2,
+                                   vbfloat16mf4x2_t vs3, size_t vl) {
+  return __riscv_vsoxseg2ei16(rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxseg2ei16_v_bf16mf2x2(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[VS2:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxseg2.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 2 x i16> [[VS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxseg2ei16_v_bf16mf2x2(__bf16 *rs1, vuint16mf2_t vs2,
+                                   vbfloat16mf2x2_t vs3, size_t vl) {
+  return __riscv_vsoxseg2ei16(rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxseg2ei16_v_bf16m1x2(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[VS2:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxseg2.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 4 x i16> [[VS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxseg2ei16_v_bf16m1x2(__bf16 *rs1, vuint16m1_t vs2,
+                                  vbfloat16m1x2_t vs3, size_t vl) {
+  return __riscv_vsoxseg2ei16(rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxseg2ei16_v_bf16m2x2(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[VS2:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxseg2.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 8 x i16> [[VS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxseg2ei16_v_bf16m2x2(__bf16 *rs1, vuint16m2_t vs2,
+                                  vbfloat16m2x2_t vs3, size_t vl) {
+  return __riscv_vsoxseg2ei16(rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxseg2ei16_v_bf16m4x2(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 16 x i16> [[VS2:%.*]], { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxseg2.nxv16bf16.nxv16i16.i64(<vscale x 16 x bfloat> [[TMP0]], <vscale x 16 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 16 x i16> [[VS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxseg2ei16_v_bf16m4x2(__bf16 *rs1, vuint16m4_t vs2,
+                                  vbfloat16m4x2_t vs3, size_t vl) {
+  return __riscv_vsoxseg2ei16(rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxseg2ei16_v_bf16mf4x2_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[VS2:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxseg2.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 1 x i16> [[VS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxseg2ei16_v_bf16mf4x2_m(vbool64_t vm, __bf16 *rs1,
+                                     vuint16mf4_t vs2, vbfloat16mf4x2_t vs3,
+                                     size_t vl) {
+  return __riscv_vsoxseg2ei16(vm, rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxseg2ei16_v_bf16mf2x2_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[VS2:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxseg2.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 2 x i16> [[VS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxseg2ei16_v_bf16mf2x2_m(vbool32_t vm, __bf16 *rs1,
+                                     vuint16mf2_t vs2, vbfloat16mf2x2_t vs3,
+                                     size_t vl) {
+  return __riscv_vsoxseg2ei16(vm, rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxseg2ei16_v_bf16m1x2_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[VS2:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxseg2.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 4 x i16> [[VS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxseg2ei16_v_bf16m1x2_m(vbool16_t vm, __bf16 *rs1, vuint16m1_t vs2,
+                                    vbfloat16m1x2_t vs3, size_t vl) {
+  return __riscv_vsoxseg2ei16(vm, rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxseg2ei16_v_bf16m2x2_m(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[VS2:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxseg2.mask.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 8 x i16> [[VS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxseg2ei16_v_bf16m2x2_m(vbool8_t vm, __bf16 *rs1, vuint16m2_t vs2,
+                                    vbfloat16m2x2_t vs3, size_t vl) {
+  return __riscv_vsoxseg2ei16(vm, rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxseg2ei16_v_bf16m4x2_m(
+// CHECK-RV64-SAME: <vscale x 16 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 16 x i16> [[VS2:%.*]], { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxseg2.mask.nxv16bf16.nxv16i16.i64(<vscale x 16 x bfloat> [[TMP0]], <vscale x 16 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 16 x i16> [[VS2]], <vscale x 16 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxseg2ei16_v_bf16m4x2_m(vbool4_t vm, __bf16 *rs1, vuint16m4_t vs2,
+                                    vbfloat16m4x2_t vs3, size_t vl) {
+  return __riscv_vsoxseg2ei16(vm, rs1, vs2, vs3, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vsoxseg3ei16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vsoxseg3ei16.c
new file mode 100644
index 000000000000..aee2c3ce96af
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vsoxseg3ei16.c
@@ -0,0 +1,123 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxseg3ei16_v_bf16mf4x3(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[VS2:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxseg3.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 1 x i16> [[VS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxseg3ei16_v_bf16mf4x3(__bf16 *rs1, vuint16mf4_t vs2,
+                                   vbfloat16mf4x3_t vs3, size_t vl) {
+  return __riscv_vsoxseg3ei16(rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxseg3ei16_v_bf16mf2x3(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[VS2:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxseg3.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 2 x i16> [[VS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxseg3ei16_v_bf16mf2x3(__bf16 *rs1, vuint16mf2_t vs2,
+                                   vbfloat16mf2x3_t vs3, size_t vl) {
+  return __riscv_vsoxseg3ei16(rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxseg3ei16_v_bf16m1x3(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[VS2:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxseg3.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 4 x i16> [[VS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxseg3ei16_v_bf16m1x3(__bf16 *rs1, vuint16m1_t vs2,
+                                  vbfloat16m1x3_t vs3, size_t vl) {
+  return __riscv_vsoxseg3ei16(rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxseg3ei16_v_bf16m2x3(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[VS2:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxseg3.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 8 x i16> [[VS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxseg3ei16_v_bf16m2x3(__bf16 *rs1, vuint16m2_t vs2,
+                                  vbfloat16m2x3_t vs3, size_t vl) {
+  return __riscv_vsoxseg3ei16(rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxseg3ei16_v_bf16mf4x3_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[VS2:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxseg3.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 1 x i16> [[VS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxseg3ei16_v_bf16mf4x3_m(vbool64_t vm, __bf16 *rs1,
+                                     vuint16mf4_t vs2, vbfloat16mf4x3_t vs3,
+                                     size_t vl) {
+  return __riscv_vsoxseg3ei16(vm, rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxseg3ei16_v_bf16mf2x3_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[VS2:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxseg3.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 2 x i16> [[VS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxseg3ei16_v_bf16mf2x3_m(vbool32_t vm, __bf16 *rs1,
+                                     vuint16mf2_t vs2, vbfloat16mf2x3_t vs3,
+                                     size_t vl) {
+  return __riscv_vsoxseg3ei16(vm, rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxseg3ei16_v_bf16m1x3_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[VS2:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxseg3.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 4 x i16> [[VS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxseg3ei16_v_bf16m1x3_m(vbool16_t vm, __bf16 *rs1, vuint16m1_t vs2,
+                                    vbfloat16m1x3_t vs3, size_t vl) {
+  return __riscv_vsoxseg3ei16(vm, rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxseg3ei16_v_bf16m2x3_m(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[VS2:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxseg3.mask.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 8 x i16> [[VS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxseg3ei16_v_bf16m2x3_m(vbool8_t vm, __bf16 *rs1, vuint16m2_t vs2,
+                                    vbfloat16m2x3_t vs3, size_t vl) {
+  return __riscv_vsoxseg3ei16(vm, rs1, vs2, vs3, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vsoxseg4ei16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vsoxseg4ei16.c
new file mode 100644
index 000000000000..4d2232129c4e
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vsoxseg4ei16.c
@@ -0,0 +1,131 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxseg4ei16_v_bf16mf4x4(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[VS2:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxseg4.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 1 x i16> [[VS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxseg4ei16_v_bf16mf4x4(__bf16 *rs1, vuint16mf4_t vs2,
+                                   vbfloat16mf4x4_t vs3, size_t vl) {
+  return __riscv_vsoxseg4ei16(rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxseg4ei16_v_bf16mf2x4(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[VS2:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxseg4.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 2 x i16> [[VS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxseg4ei16_v_bf16mf2x4(__bf16 *rs1, vuint16mf2_t vs2,
+                                   vbfloat16mf2x4_t vs3, size_t vl) {
+  return __riscv_vsoxseg4ei16(rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxseg4ei16_v_bf16m1x4(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[VS2:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxseg4.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 4 x i16> [[VS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxseg4ei16_v_bf16m1x4(__bf16 *rs1, vuint16m1_t vs2,
+                                  vbfloat16m1x4_t vs3, size_t vl) {
+  return __riscv_vsoxseg4ei16(rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxseg4ei16_v_bf16m2x4(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[VS2:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxseg4.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], <vscale x 8 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 8 x i16> [[VS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxseg4ei16_v_bf16m2x4(__bf16 *rs1, vuint16m2_t vs2,
+                                  vbfloat16m2x4_t vs3, size_t vl) {
+  return __riscv_vsoxseg4ei16(rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxseg4ei16_v_bf16mf4x4_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[VS2:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxseg4.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 1 x i16> [[VS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxseg4ei16_v_bf16mf4x4_m(vbool64_t vm, __bf16 *rs1,
+                                     vuint16mf4_t vs2, vbfloat16mf4x4_t vs3,
+                                     size_t vl) {
+  return __riscv_vsoxseg4ei16(vm, rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxseg4ei16_v_bf16mf2x4_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[VS2:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxseg4.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 2 x i16> [[VS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxseg4ei16_v_bf16mf2x4_m(vbool32_t vm, __bf16 *rs1,
+                                     vuint16mf2_t vs2, vbfloat16mf2x4_t vs3,
+                                     size_t vl) {
+  return __riscv_vsoxseg4ei16(vm, rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxseg4ei16_v_bf16m1x4_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[VS2:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxseg4.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 4 x i16> [[VS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxseg4ei16_v_bf16m1x4_m(vbool16_t vm, __bf16 *rs1, vuint16m1_t vs2,
+                                    vbfloat16m1x4_t vs3, size_t vl) {
+  return __riscv_vsoxseg4ei16(vm, rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxseg4ei16_v_bf16m2x4_m(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[VS2:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxseg4.mask.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], <vscale x 8 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 8 x i16> [[VS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxseg4ei16_v_bf16m2x4_m(vbool8_t vm, __bf16 *rs1, vuint16m2_t vs2,
+                                    vbfloat16m2x4_t vs3, size_t vl) {
+  return __riscv_vsoxseg4ei16(vm, rs1, vs2, vs3, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vsoxseg5ei16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vsoxseg5ei16.c
new file mode 100644
index 000000000000..5975ed0cc179
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vsoxseg5ei16.c
@@ -0,0 +1,107 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxseg5ei16_v_bf16mf4x5(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[VS2:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxseg5.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 1 x i16> [[VS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxseg5ei16_v_bf16mf4x5(__bf16 *rs1, vuint16mf4_t vs2,
+                                   vbfloat16mf4x5_t vs3, size_t vl) {
+  return __riscv_vsoxseg5ei16(rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxseg5ei16_v_bf16mf2x5(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[VS2:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxseg5.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 2 x i16> [[VS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxseg5ei16_v_bf16mf2x5(__bf16 *rs1, vuint16mf2_t vs2,
+                                   vbfloat16mf2x5_t vs3, size_t vl) {
+  return __riscv_vsoxseg5ei16(rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxseg5ei16_v_bf16m1x5(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[VS2:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxseg5.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 4 x i16> [[VS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxseg5ei16_v_bf16m1x5(__bf16 *rs1, vuint16m1_t vs2,
+                                  vbfloat16m1x5_t vs3, size_t vl) {
+  return __riscv_vsoxseg5ei16(rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxseg5ei16_v_bf16mf4x5_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[VS2:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxseg5.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 1 x i16> [[VS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxseg5ei16_v_bf16mf4x5_m(vbool64_t vm, __bf16 *rs1,
+                                     vuint16mf4_t vs2, vbfloat16mf4x5_t vs3,
+                                     size_t vl) {
+  return __riscv_vsoxseg5ei16(vm, rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxseg5ei16_v_bf16mf2x5_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[VS2:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxseg5.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 2 x i16> [[VS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxseg5ei16_v_bf16mf2x5_m(vbool32_t vm, __bf16 *rs1,
+                                     vuint16mf2_t vs2, vbfloat16mf2x5_t vs3,
+                                     size_t vl) {
+  return __riscv_vsoxseg5ei16(vm, rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxseg5ei16_v_bf16m1x5_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[VS2:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxseg5.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 4 x i16> [[VS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxseg5ei16_v_bf16m1x5_m(vbool16_t vm, __bf16 *rs1, vuint16m1_t vs2,
+                                    vbfloat16m1x5_t vs3, size_t vl) {
+  return __riscv_vsoxseg5ei16(vm, rs1, vs2, vs3, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vsoxseg6ei16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vsoxseg6ei16.c
new file mode 100644
index 000000000000..cca77d6dc668
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vsoxseg6ei16.c
@@ -0,0 +1,113 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxseg6ei16_v_bf16mf4x6(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[VS2:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxseg6.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 1 x i16> [[VS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxseg6ei16_v_bf16mf4x6(__bf16 *rs1, vuint16mf4_t vs2,
+                                   vbfloat16mf4x6_t vs3, size_t vl) {
+  return __riscv_vsoxseg6ei16(rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxseg6ei16_v_bf16mf2x6(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[VS2:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxseg6.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 2 x i16> [[VS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxseg6ei16_v_bf16mf2x6(__bf16 *rs1, vuint16mf2_t vs2,
+                                   vbfloat16mf2x6_t vs3, size_t vl) {
+  return __riscv_vsoxseg6ei16(rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxseg6ei16_v_bf16m1x6(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[VS2:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxseg6.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 4 x i16> [[VS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxseg6ei16_v_bf16m1x6(__bf16 *rs1, vuint16m1_t vs2,
+                                  vbfloat16m1x6_t vs3, size_t vl) {
+  return __riscv_vsoxseg6ei16(rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxseg6ei16_v_bf16mf4x6_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[VS2:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxseg6.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 1 x i16> [[VS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxseg6ei16_v_bf16mf4x6_m(vbool64_t vm, __bf16 *rs1,
+                                     vuint16mf4_t vs2, vbfloat16mf4x6_t vs3,
+                                     size_t vl) {
+  return __riscv_vsoxseg6ei16(vm, rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxseg6ei16_v_bf16mf2x6_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[VS2:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxseg6.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 2 x i16> [[VS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxseg6ei16_v_bf16mf2x6_m(vbool32_t vm, __bf16 *rs1,
+                                     vuint16mf2_t vs2, vbfloat16mf2x6_t vs3,
+                                     size_t vl) {
+  return __riscv_vsoxseg6ei16(vm, rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxseg6ei16_v_bf16m1x6_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[VS2:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxseg6.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 4 x i16> [[VS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxseg6ei16_v_bf16m1x6_m(vbool16_t vm, __bf16 *rs1, vuint16m1_t vs2,
+                                    vbfloat16m1x6_t vs3, size_t vl) {
+  return __riscv_vsoxseg6ei16(vm, rs1, vs2, vs3, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vsoxseg7ei16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vsoxseg7ei16.c
new file mode 100644
index 000000000000..f258f74f4b25
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vsoxseg7ei16.c
@@ -0,0 +1,119 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxseg7ei16_v_bf16mf4x7(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[VS2:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 6
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxseg7.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 1 x i16> [[VS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxseg7ei16_v_bf16mf4x7(__bf16 *rs1, vuint16mf4_t vs2,
+                                   vbfloat16mf4x7_t vs3, size_t vl) {
+  return __riscv_vsoxseg7ei16(rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxseg7ei16_v_bf16mf2x7(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[VS2:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 6
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxseg7.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 2 x i16> [[VS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxseg7ei16_v_bf16mf2x7(__bf16 *rs1, vuint16mf2_t vs2,
+                                   vbfloat16mf2x7_t vs3, size_t vl) {
+  return __riscv_vsoxseg7ei16(rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxseg7ei16_v_bf16m1x7(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[VS2:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 6
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxseg7.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 4 x i16> [[VS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxseg7ei16_v_bf16m1x7(__bf16 *rs1, vuint16m1_t vs2,
+                                  vbfloat16m1x7_t vs3, size_t vl) {
+  return __riscv_vsoxseg7ei16(rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxseg7ei16_v_bf16mf4x7_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[VS2:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 6
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxseg7.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 1 x i16> [[VS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxseg7ei16_v_bf16mf4x7_m(vbool64_t vm, __bf16 *rs1,
+                                     vuint16mf4_t vs2, vbfloat16mf4x7_t vs3,
+                                     size_t vl) {
+  return __riscv_vsoxseg7ei16(vm, rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxseg7ei16_v_bf16mf2x7_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[VS2:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 6
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxseg7.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 2 x i16> [[VS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxseg7ei16_v_bf16mf2x7_m(vbool32_t vm, __bf16 *rs1,
+                                     vuint16mf2_t vs2, vbfloat16mf2x7_t vs3,
+                                     size_t vl) {
+  return __riscv_vsoxseg7ei16(vm, rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxseg7ei16_v_bf16m1x7_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[VS2:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 6
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxseg7.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 4 x i16> [[VS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxseg7ei16_v_bf16m1x7_m(vbool16_t vm, __bf16 *rs1, vuint16m1_t vs2,
+                                    vbfloat16m1x7_t vs3, size_t vl) {
+  return __riscv_vsoxseg7ei16(vm, rs1, vs2, vs3, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vsoxseg8ei16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vsoxseg8ei16.c
new file mode 100644
index 000000000000..8f718ac3b730
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vsoxseg8ei16.c
@@ -0,0 +1,125 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxseg8ei16_v_bf16mf4x8(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[VS2:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 7
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxseg8.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], <vscale x 1 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 1 x i16> [[VS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxseg8ei16_v_bf16mf4x8(__bf16 *rs1, vuint16mf4_t vs2,
+                                   vbfloat16mf4x8_t vs3, size_t vl) {
+  return __riscv_vsoxseg8ei16(rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxseg8ei16_v_bf16mf2x8(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[VS2:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 7
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxseg8.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], <vscale x 2 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 2 x i16> [[VS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxseg8ei16_v_bf16mf2x8(__bf16 *rs1, vuint16mf2_t vs2,
+                                   vbfloat16mf2x8_t vs3, size_t vl) {
+  return __riscv_vsoxseg8ei16(rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxseg8ei16_v_bf16m1x8(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[VS2:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 7
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxseg8.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], <vscale x 4 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 4 x i16> [[VS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxseg8ei16_v_bf16m1x8(__bf16 *rs1, vuint16m1_t vs2,
+                                  vbfloat16m1x8_t vs3, size_t vl) {
+  return __riscv_vsoxseg8ei16(rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxseg8ei16_v_bf16mf4x8_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[VS2:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 7
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxseg8.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], <vscale x 1 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 1 x i16> [[VS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxseg8ei16_v_bf16mf4x8_m(vbool64_t vm, __bf16 *rs1,
+                                     vuint16mf4_t vs2, vbfloat16mf4x8_t vs3,
+                                     size_t vl) {
+  return __riscv_vsoxseg8ei16(vm, rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxseg8ei16_v_bf16mf2x8_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[VS2:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 7
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxseg8.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], <vscale x 2 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 2 x i16> [[VS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxseg8ei16_v_bf16mf2x8_m(vbool32_t vm, __bf16 *rs1,
+                                     vuint16mf2_t vs2, vbfloat16mf2x8_t vs3,
+                                     size_t vl) {
+  return __riscv_vsoxseg8ei16(vm, rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsoxseg8ei16_v_bf16m1x8_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[VS2:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 7
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsoxseg8.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], <vscale x 4 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 4 x i16> [[VS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsoxseg8ei16_v_bf16m1x8_m(vbool16_t vm, __bf16 *rs1, vuint16m1_t vs2,
+                                    vbfloat16m1x8_t vs3, size_t vl) {
+  return __riscv_vsoxseg8ei16(vm, rs1, vs2, vs3, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vsse16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vsse16.c
new file mode 100644
index 000000000000..103d2bff3228
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vsse16.c
@@ -0,0 +1,141 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsse16_v_bf16mf4(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], <vscale x 1 x bfloat> [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsse.nxv1bf16.i64(<vscale x 1 x bfloat> [[VS3]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsse16_v_bf16mf4(__bf16 *rs1, ptrdiff_t rs2, vbfloat16mf4_t vs3,
+                           size_t vl) {
+  return __riscv_vsse16(rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsse16_v_bf16mf2(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], <vscale x 2 x bfloat> [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsse.nxv2bf16.i64(<vscale x 2 x bfloat> [[VS3]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsse16_v_bf16mf2(__bf16 *rs1, ptrdiff_t rs2, vbfloat16mf2_t vs3,
+                           size_t vl) {
+  return __riscv_vsse16(rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsse16_v_bf16m1(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], <vscale x 4 x bfloat> [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsse.nxv4bf16.i64(<vscale x 4 x bfloat> [[VS3]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsse16_v_bf16m1(__bf16 *rs1, ptrdiff_t rs2, vbfloat16m1_t vs3,
+                          size_t vl) {
+  return __riscv_vsse16(rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsse16_v_bf16m2(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], <vscale x 8 x bfloat> [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsse.nxv8bf16.i64(<vscale x 8 x bfloat> [[VS3]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsse16_v_bf16m2(__bf16 *rs1, ptrdiff_t rs2, vbfloat16m2_t vs3,
+                          size_t vl) {
+  return __riscv_vsse16(rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsse16_v_bf16m4(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], <vscale x 16 x bfloat> [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsse.nxv16bf16.i64(<vscale x 16 x bfloat> [[VS3]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsse16_v_bf16m4(__bf16 *rs1, ptrdiff_t rs2, vbfloat16m4_t vs3,
+                          size_t vl) {
+  return __riscv_vsse16(rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsse16_v_bf16m8(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], <vscale x 32 x bfloat> [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsse.nxv32bf16.i64(<vscale x 32 x bfloat> [[VS3]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsse16_v_bf16m8(__bf16 *rs1, ptrdiff_t rs2, vbfloat16m8_t vs3,
+                          size_t vl) {
+  return __riscv_vsse16(rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsse16_v_bf16mf4_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], <vscale x 1 x bfloat> [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsse.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[VS3]], ptr [[RS1]], i64 [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsse16_v_bf16mf4_m(vbool64_t vm, __bf16 *rs1, ptrdiff_t rs2,
+                             vbfloat16mf4_t vs3, size_t vl) {
+  return __riscv_vsse16(vm, rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsse16_v_bf16mf2_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], <vscale x 2 x bfloat> [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsse.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[VS3]], ptr [[RS1]], i64 [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsse16_v_bf16mf2_m(vbool32_t vm, __bf16 *rs1, ptrdiff_t rs2,
+                             vbfloat16mf2_t vs3, size_t vl) {
+  return __riscv_vsse16(vm, rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsse16_v_bf16m1_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], <vscale x 4 x bfloat> [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsse.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[VS3]], ptr [[RS1]], i64 [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsse16_v_bf16m1_m(vbool16_t vm, __bf16 *rs1, ptrdiff_t rs2,
+                            vbfloat16m1_t vs3, size_t vl) {
+  return __riscv_vsse16(vm, rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsse16_v_bf16m2_m(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], <vscale x 8 x bfloat> [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsse.mask.nxv8bf16.i64(<vscale x 8 x bfloat> [[VS3]], ptr [[RS1]], i64 [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsse16_v_bf16m2_m(vbool8_t vm, __bf16 *rs1, ptrdiff_t rs2,
+                            vbfloat16m2_t vs3, size_t vl) {
+  return __riscv_vsse16(vm, rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsse16_v_bf16m4_m(
+// CHECK-RV64-SAME: <vscale x 16 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], <vscale x 16 x bfloat> [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsse.mask.nxv16bf16.i64(<vscale x 16 x bfloat> [[VS3]], ptr [[RS1]], i64 [[RS2]], <vscale x 16 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsse16_v_bf16m4_m(vbool4_t vm, __bf16 *rs1, ptrdiff_t rs2,
+                            vbfloat16m4_t vs3, size_t vl) {
+  return __riscv_vsse16(vm, rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsse16_v_bf16m8_m(
+// CHECK-RV64-SAME: <vscale x 32 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], <vscale x 32 x bfloat> [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsse.mask.nxv32bf16.i64(<vscale x 32 x bfloat> [[VS3]], ptr [[RS1]], i64 [[RS2]], <vscale x 32 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsse16_v_bf16m8_m(vbool2_t vm, __bf16 *rs1, ptrdiff_t rs2,
+                            vbfloat16m8_t vs3, size_t vl) {
+  return __riscv_vsse16(vm, rs1, rs2, vs3, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vsseg2e16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vsseg2e16.c
new file mode 100644
index 000000000000..c302d828fd05
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vsseg2e16.c
@@ -0,0 +1,134 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsseg2e16_v_bf16mf4x2(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsseg2.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsseg2e16_v_bf16mf4x2(__bf16 *rs1, vbfloat16mf4x2_t vs3, size_t vl) {
+  return __riscv_vsseg2e16(rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsseg2e16_v_bf16mf2x2(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsseg2.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsseg2e16_v_bf16mf2x2(__bf16 *rs1, vbfloat16mf2x2_t vs3, size_t vl) {
+  return __riscv_vsseg2e16(rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsseg2e16_v_bf16m1x2(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsseg2.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsseg2e16_v_bf16m1x2(__bf16 *rs1, vbfloat16m1x2_t vs3, size_t vl) {
+  return __riscv_vsseg2e16(rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsseg2e16_v_bf16m2x2(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsseg2.nxv8bf16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsseg2e16_v_bf16m2x2(__bf16 *rs1, vbfloat16m2x2_t vs3, size_t vl) {
+  return __riscv_vsseg2e16(rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsseg2e16_v_bf16m4x2(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsseg2.nxv16bf16.i64(<vscale x 16 x bfloat> [[TMP0]], <vscale x 16 x bfloat> [[TMP1]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsseg2e16_v_bf16m4x2(__bf16 *rs1, vbfloat16m4x2_t vs3, size_t vl) {
+  return __riscv_vsseg2e16(rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsseg2e16_v_bf16mf4x2_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsseg2.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsseg2e16_v_bf16mf4x2_m(vbool64_t vm, __bf16 *rs1,
+                                  vbfloat16mf4x2_t vs3, size_t vl) {
+  return __riscv_vsseg2e16(vm, rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsseg2e16_v_bf16mf2x2_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsseg2.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsseg2e16_v_bf16mf2x2_m(vbool32_t vm, __bf16 *rs1,
+                                  vbfloat16mf2x2_t vs3, size_t vl) {
+  return __riscv_vsseg2e16(vm, rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsseg2e16_v_bf16m1x2_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsseg2.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsseg2e16_v_bf16m1x2_m(vbool16_t vm, __bf16 *rs1, vbfloat16m1x2_t vs3,
+                                 size_t vl) {
+  return __riscv_vsseg2e16(vm, rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsseg2e16_v_bf16m2x2_m(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsseg2.mask.nxv8bf16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 8 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsseg2e16_v_bf16m2x2_m(vbool8_t vm, __bf16 *rs1, vbfloat16m2x2_t vs3,
+                                 size_t vl) {
+  return __riscv_vsseg2e16(vm, rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsseg2e16_v_bf16m4x2_m(
+// CHECK-RV64-SAME: <vscale x 16 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsseg2.mask.nxv16bf16.i64(<vscale x 16 x bfloat> [[TMP0]], <vscale x 16 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 16 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsseg2e16_v_bf16m4x2_m(vbool4_t vm, __bf16 *rs1, vbfloat16m4x2_t vs3,
+                                 size_t vl) {
+  return __riscv_vsseg2e16(vm, rs1, vs3, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vsseg3e16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vsseg3e16.c
new file mode 100644
index 000000000000..644d2145dc11
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vsseg3e16.c
@@ -0,0 +1,117 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsseg3e16_v_bf16mf4x3(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsseg3.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsseg3e16_v_bf16mf4x3(__bf16 *rs1, vbfloat16mf4x3_t vs3, size_t vl) {
+  return __riscv_vsseg3e16(rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsseg3e16_v_bf16mf2x3(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsseg3.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsseg3e16_v_bf16mf2x3(__bf16 *rs1, vbfloat16mf2x3_t vs3, size_t vl) {
+  return __riscv_vsseg3e16(rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsseg3e16_v_bf16m1x3(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsseg3.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsseg3e16_v_bf16m1x3(__bf16 *rs1, vbfloat16m1x3_t vs3, size_t vl) {
+  return __riscv_vsseg3e16(rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsseg3e16_v_bf16m2x3(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsseg3.nxv8bf16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsseg3e16_v_bf16m2x3(__bf16 *rs1, vbfloat16m2x3_t vs3, size_t vl) {
+  return __riscv_vsseg3e16(rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsseg3e16_v_bf16mf4x3_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsseg3.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsseg3e16_v_bf16mf4x3_m(vbool64_t vm, __bf16 *rs1,
+                                  vbfloat16mf4x3_t vs3, size_t vl) {
+  return __riscv_vsseg3e16(vm, rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsseg3e16_v_bf16mf2x3_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsseg3.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsseg3e16_v_bf16mf2x3_m(vbool32_t vm, __bf16 *rs1,
+                                  vbfloat16mf2x3_t vs3, size_t vl) {
+  return __riscv_vsseg3e16(vm, rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsseg3e16_v_bf16m1x3_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsseg3.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsseg3e16_v_bf16m1x3_m(vbool16_t vm, __bf16 *rs1, vbfloat16m1x3_t vs3,
+                                 size_t vl) {
+  return __riscv_vsseg3e16(vm, rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsseg3e16_v_bf16m2x3_m(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsseg3.mask.nxv8bf16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 8 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsseg3e16_v_bf16m2x3_m(vbool8_t vm, __bf16 *rs1, vbfloat16m2x3_t vs3,
+                                 size_t vl) {
+  return __riscv_vsseg3e16(vm, rs1, vs3, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vsseg4e16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vsseg4e16.c
new file mode 100644
index 000000000000..7dbd05772785
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vsseg4e16.c
@@ -0,0 +1,125 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsseg4e16_v_bf16mf4x4(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsseg4.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsseg4e16_v_bf16mf4x4(__bf16 *rs1, vbfloat16mf4x4_t vs3, size_t vl) {
+  return __riscv_vsseg4e16(rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsseg4e16_v_bf16mf2x4(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsseg4.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsseg4e16_v_bf16mf2x4(__bf16 *rs1, vbfloat16mf2x4_t vs3, size_t vl) {
+  return __riscv_vsseg4e16(rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsseg4e16_v_bf16m1x4(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsseg4.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsseg4e16_v_bf16m1x4(__bf16 *rs1, vbfloat16m1x4_t vs3, size_t vl) {
+  return __riscv_vsseg4e16(rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsseg4e16_v_bf16m2x4(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsseg4.nxv8bf16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], <vscale x 8 x bfloat> [[TMP3]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsseg4e16_v_bf16m2x4(__bf16 *rs1, vbfloat16m2x4_t vs3, size_t vl) {
+  return __riscv_vsseg4e16(rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsseg4e16_v_bf16mf4x4_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsseg4.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsseg4e16_v_bf16mf4x4_m(vbool64_t vm, __bf16 *rs1,
+                                  vbfloat16mf4x4_t vs3, size_t vl) {
+  return __riscv_vsseg4e16(vm, rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsseg4e16_v_bf16mf2x4_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsseg4.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsseg4e16_v_bf16mf2x4_m(vbool32_t vm, __bf16 *rs1,
+                                  vbfloat16mf2x4_t vs3, size_t vl) {
+  return __riscv_vsseg4e16(vm, rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsseg4e16_v_bf16m1x4_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsseg4.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsseg4e16_v_bf16m1x4_m(vbool16_t vm, __bf16 *rs1, vbfloat16m1x4_t vs3,
+                                 size_t vl) {
+  return __riscv_vsseg4e16(vm, rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsseg4e16_v_bf16m2x4_m(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsseg4.mask.nxv8bf16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], <vscale x 8 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 8 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsseg4e16_v_bf16m2x4_m(vbool8_t vm, __bf16 *rs1, vbfloat16m2x4_t vs3,
+                                 size_t vl) {
+  return __riscv_vsseg4e16(vm, rs1, vs3, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vsseg5e16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vsseg5e16.c
new file mode 100644
index 000000000000..be8a12c05078
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vsseg5e16.c
@@ -0,0 +1,102 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsseg5e16_v_bf16mf4x5(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsseg5.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsseg5e16_v_bf16mf4x5(__bf16 *rs1, vbfloat16mf4x5_t vs3, size_t vl) {
+  return __riscv_vsseg5e16(rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsseg5e16_v_bf16mf2x5(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsseg5.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsseg5e16_v_bf16mf2x5(__bf16 *rs1, vbfloat16mf2x5_t vs3, size_t vl) {
+  return __riscv_vsseg5e16(rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsseg5e16_v_bf16m1x5(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsseg5.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsseg5e16_v_bf16m1x5(__bf16 *rs1, vbfloat16m1x5_t vs3, size_t vl) {
+  return __riscv_vsseg5e16(rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsseg5e16_v_bf16mf4x5_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsseg5.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsseg5e16_v_bf16mf4x5_m(vbool64_t vm, __bf16 *rs1,
+                                  vbfloat16mf4x5_t vs3, size_t vl) {
+  return __riscv_vsseg5e16(vm, rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsseg5e16_v_bf16mf2x5_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsseg5.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsseg5e16_v_bf16mf2x5_m(vbool32_t vm, __bf16 *rs1,
+                                  vbfloat16mf2x5_t vs3, size_t vl) {
+  return __riscv_vsseg5e16(vm, rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsseg5e16_v_bf16m1x5_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsseg5.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsseg5e16_v_bf16m1x5_m(vbool16_t vm, __bf16 *rs1, vbfloat16m1x5_t vs3,
+                                 size_t vl) {
+  return __riscv_vsseg5e16(vm, rs1, vs3, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vsseg6e16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vsseg6e16.c
new file mode 100644
index 000000000000..4de257158927
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vsseg6e16.c
@@ -0,0 +1,108 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsseg6e16_v_bf16mf4x6(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsseg6.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsseg6e16_v_bf16mf4x6(__bf16 *rs1, vbfloat16mf4x6_t vs3, size_t vl) {
+  return __riscv_vsseg6e16(rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsseg6e16_v_bf16mf2x6(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsseg6.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsseg6e16_v_bf16mf2x6(__bf16 *rs1, vbfloat16mf2x6_t vs3, size_t vl) {
+  return __riscv_vsseg6e16(rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsseg6e16_v_bf16m1x6(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsseg6.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsseg6e16_v_bf16m1x6(__bf16 *rs1, vbfloat16m1x6_t vs3, size_t vl) {
+  return __riscv_vsseg6e16(rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsseg6e16_v_bf16mf4x6_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsseg6.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsseg6e16_v_bf16mf4x6_m(vbool64_t vm, __bf16 *rs1,
+                                  vbfloat16mf4x6_t vs3, size_t vl) {
+  return __riscv_vsseg6e16(vm, rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsseg6e16_v_bf16mf2x6_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsseg6.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsseg6e16_v_bf16mf2x6_m(vbool32_t vm, __bf16 *rs1,
+                                  vbfloat16mf2x6_t vs3, size_t vl) {
+  return __riscv_vsseg6e16(vm, rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsseg6e16_v_bf16m1x6_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsseg6.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsseg6e16_v_bf16m1x6_m(vbool16_t vm, __bf16 *rs1, vbfloat16m1x6_t vs3,
+                                 size_t vl) {
+  return __riscv_vsseg6e16(vm, rs1, vs3, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vsseg7e16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vsseg7e16.c
new file mode 100644
index 000000000000..24252fe47940
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vsseg7e16.c
@@ -0,0 +1,114 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsseg7e16_v_bf16mf4x7(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 6
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsseg7.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsseg7e16_v_bf16mf4x7(__bf16 *rs1, vbfloat16mf4x7_t vs3, size_t vl) {
+  return __riscv_vsseg7e16(rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsseg7e16_v_bf16mf2x7(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 6
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsseg7.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsseg7e16_v_bf16mf2x7(__bf16 *rs1, vbfloat16mf2x7_t vs3, size_t vl) {
+  return __riscv_vsseg7e16(rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsseg7e16_v_bf16m1x7(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 6
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsseg7.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsseg7e16_v_bf16m1x7(__bf16 *rs1, vbfloat16m1x7_t vs3, size_t vl) {
+  return __riscv_vsseg7e16(rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsseg7e16_v_bf16mf4x7_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 6
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsseg7.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsseg7e16_v_bf16mf4x7_m(vbool64_t vm, __bf16 *rs1,
+                                  vbfloat16mf4x7_t vs3, size_t vl) {
+  return __riscv_vsseg7e16(vm, rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsseg7e16_v_bf16mf2x7_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 6
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsseg7.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsseg7e16_v_bf16mf2x7_m(vbool32_t vm, __bf16 *rs1,
+                                  vbfloat16mf2x7_t vs3, size_t vl) {
+  return __riscv_vsseg7e16(vm, rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsseg7e16_v_bf16m1x7_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 6
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsseg7.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsseg7e16_v_bf16m1x7_m(vbool16_t vm, __bf16 *rs1, vbfloat16m1x7_t vs3,
+                                 size_t vl) {
+  return __riscv_vsseg7e16(vm, rs1, vs3, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vsseg8e16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vsseg8e16.c
new file mode 100644
index 000000000000..c6f5e5db5d9c
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vsseg8e16.c
@@ -0,0 +1,120 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsseg8e16_v_bf16mf4x8(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 7
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsseg8.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], <vscale x 1 x bfloat> [[TMP7]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsseg8e16_v_bf16mf4x8(__bf16 *rs1, vbfloat16mf4x8_t vs3, size_t vl) {
+  return __riscv_vsseg8e16(rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsseg8e16_v_bf16mf2x8(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 7
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsseg8.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], <vscale x 2 x bfloat> [[TMP7]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsseg8e16_v_bf16mf2x8(__bf16 *rs1, vbfloat16mf2x8_t vs3, size_t vl) {
+  return __riscv_vsseg8e16(rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsseg8e16_v_bf16m1x8(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 7
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsseg8.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], <vscale x 4 x bfloat> [[TMP7]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsseg8e16_v_bf16m1x8(__bf16 *rs1, vbfloat16m1x8_t vs3, size_t vl) {
+  return __riscv_vsseg8e16(rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsseg8e16_v_bf16mf4x8_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 7
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsseg8.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], <vscale x 1 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsseg8e16_v_bf16mf4x8_m(vbool64_t vm, __bf16 *rs1,
+                                  vbfloat16mf4x8_t vs3, size_t vl) {
+  return __riscv_vsseg8e16(vm, rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsseg8e16_v_bf16mf2x8_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 7
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsseg8.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], <vscale x 2 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsseg8e16_v_bf16mf2x8_m(vbool32_t vm, __bf16 *rs1,
+                                  vbfloat16mf2x8_t vs3, size_t vl) {
+  return __riscv_vsseg8e16(vm, rs1, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsseg8e16_v_bf16m1x8_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 7
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsseg8.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], <vscale x 4 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsseg8e16_v_bf16m1x8_m(vbool16_t vm, __bf16 *rs1, vbfloat16m1x8_t vs3,
+                                 size_t vl) {
+  return __riscv_vsseg8e16(vm, rs1, vs3, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vssseg2e16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vssseg2e16.c
new file mode 100644
index 000000000000..ee6d701667dd
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vssseg2e16.c
@@ -0,0 +1,139 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local void @test_vssseg2e16_v_bf16mf4x2(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vssseg2.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vssseg2e16_v_bf16mf4x2(__bf16 *rs1, ptrdiff_t rs2,
+                                 vbfloat16mf4x2_t vs3, size_t vl) {
+  return __riscv_vssseg2e16(rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vssseg2e16_v_bf16mf2x2(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vssseg2.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vssseg2e16_v_bf16mf2x2(__bf16 *rs1, ptrdiff_t rs2,
+                                 vbfloat16mf2x2_t vs3, size_t vl) {
+  return __riscv_vssseg2e16(rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vssseg2e16_v_bf16m1x2(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vssseg2.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vssseg2e16_v_bf16m1x2(__bf16 *rs1, ptrdiff_t rs2, vbfloat16m1x2_t vs3,
+                                size_t vl) {
+  return __riscv_vssseg2e16(rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vssseg2e16_v_bf16m2x2(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vssseg2.nxv8bf16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vssseg2e16_v_bf16m2x2(__bf16 *rs1, ptrdiff_t rs2, vbfloat16m2x2_t vs3,
+                                size_t vl) {
+  return __riscv_vssseg2e16(rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vssseg2e16_v_bf16m4x2(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vssseg2.nxv16bf16.i64(<vscale x 16 x bfloat> [[TMP0]], <vscale x 16 x bfloat> [[TMP1]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vssseg2e16_v_bf16m4x2(__bf16 *rs1, ptrdiff_t rs2, vbfloat16m4x2_t vs3,
+                                size_t vl) {
+  return __riscv_vssseg2e16(rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vssseg2e16_v_bf16mf4x2_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vssseg2.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], ptr [[RS1]], i64 [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vssseg2e16_v_bf16mf4x2_m(vbool64_t vm, __bf16 *rs1, ptrdiff_t rs2,
+                                   vbfloat16mf4x2_t vs3, size_t vl) {
+  return __riscv_vssseg2e16(vm, rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vssseg2e16_v_bf16mf2x2_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vssseg2.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], ptr [[RS1]], i64 [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vssseg2e16_v_bf16mf2x2_m(vbool32_t vm, __bf16 *rs1, ptrdiff_t rs2,
+                                   vbfloat16mf2x2_t vs3, size_t vl) {
+  return __riscv_vssseg2e16(vm, rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vssseg2e16_v_bf16m1x2_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vssseg2.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], ptr [[RS1]], i64 [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vssseg2e16_v_bf16m1x2_m(vbool16_t vm, __bf16 *rs1, ptrdiff_t rs2,
+                                  vbfloat16m1x2_t vs3, size_t vl) {
+  return __riscv_vssseg2e16(vm, rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vssseg2e16_v_bf16m2x2_m(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vssseg2.mask.nxv8bf16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], ptr [[RS1]], i64 [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vssseg2e16_v_bf16m2x2_m(vbool8_t vm, __bf16 *rs1, ptrdiff_t rs2,
+                                  vbfloat16m2x2_t vs3, size_t vl) {
+  return __riscv_vssseg2e16(vm, rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vssseg2e16_v_bf16m4x2_m(
+// CHECK-RV64-SAME: <vscale x 16 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vssseg2.mask.nxv16bf16.i64(<vscale x 16 x bfloat> [[TMP0]], <vscale x 16 x bfloat> [[TMP1]], ptr [[RS1]], i64 [[RS2]], <vscale x 16 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vssseg2e16_v_bf16m4x2_m(vbool4_t vm, __bf16 *rs1, ptrdiff_t rs2,
+                                  vbfloat16m4x2_t vs3, size_t vl) {
+  return __riscv_vssseg2e16(vm, rs1, rs2, vs3, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vssseg3e16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vssseg3e16.c
new file mode 100644
index 000000000000..c8562ff45364
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vssseg3e16.c
@@ -0,0 +1,121 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local void @test_vssseg3e16_v_bf16mf4x3(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vssseg3.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vssseg3e16_v_bf16mf4x3(__bf16 *rs1, ptrdiff_t rs2,
+                                 vbfloat16mf4x3_t vs3, size_t vl) {
+  return __riscv_vssseg3e16(rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vssseg3e16_v_bf16mf2x3(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vssseg3.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vssseg3e16_v_bf16mf2x3(__bf16 *rs1, ptrdiff_t rs2,
+                                 vbfloat16mf2x3_t vs3, size_t vl) {
+  return __riscv_vssseg3e16(rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vssseg3e16_v_bf16m1x3(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vssseg3.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vssseg3e16_v_bf16m1x3(__bf16 *rs1, ptrdiff_t rs2, vbfloat16m1x3_t vs3,
+                                size_t vl) {
+  return __riscv_vssseg3e16(rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vssseg3e16_v_bf16m2x3(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vssseg3.nxv8bf16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vssseg3e16_v_bf16m2x3(__bf16 *rs1, ptrdiff_t rs2, vbfloat16m2x3_t vs3,
+                                size_t vl) {
+  return __riscv_vssseg3e16(rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vssseg3e16_v_bf16mf4x3_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vssseg3.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], ptr [[RS1]], i64 [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vssseg3e16_v_bf16mf4x3_m(vbool64_t vm, __bf16 *rs1, ptrdiff_t rs2,
+                                   vbfloat16mf4x3_t vs3, size_t vl) {
+  return __riscv_vssseg3e16(vm, rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vssseg3e16_v_bf16mf2x3_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vssseg3.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], ptr [[RS1]], i64 [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vssseg3e16_v_bf16mf2x3_m(vbool32_t vm, __bf16 *rs1, ptrdiff_t rs2,
+                                   vbfloat16mf2x3_t vs3, size_t vl) {
+  return __riscv_vssseg3e16(vm, rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vssseg3e16_v_bf16m1x3_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vssseg3.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], ptr [[RS1]], i64 [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vssseg3e16_v_bf16m1x3_m(vbool16_t vm, __bf16 *rs1, ptrdiff_t rs2,
+                                  vbfloat16m1x3_t vs3, size_t vl) {
+  return __riscv_vssseg3e16(vm, rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vssseg3e16_v_bf16m2x3_m(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vssseg3.mask.nxv8bf16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], ptr [[RS1]], i64 [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vssseg3e16_v_bf16m2x3_m(vbool8_t vm, __bf16 *rs1, ptrdiff_t rs2,
+                                  vbfloat16m2x3_t vs3, size_t vl) {
+  return __riscv_vssseg3e16(vm, rs1, rs2, vs3, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vssseg4e16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vssseg4e16.c
new file mode 100644
index 000000000000..64069bdc03ba
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vssseg4e16.c
@@ -0,0 +1,129 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local void @test_vssseg4e16_v_bf16mf4x4(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vssseg4.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vssseg4e16_v_bf16mf4x4(__bf16 *rs1, ptrdiff_t rs2,
+                                 vbfloat16mf4x4_t vs3, size_t vl) {
+  return __riscv_vssseg4e16(rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vssseg4e16_v_bf16mf2x4(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vssseg4.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vssseg4e16_v_bf16mf2x4(__bf16 *rs1, ptrdiff_t rs2,
+                                 vbfloat16mf2x4_t vs3, size_t vl) {
+  return __riscv_vssseg4e16(rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vssseg4e16_v_bf16m1x4(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vssseg4.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vssseg4e16_v_bf16m1x4(__bf16 *rs1, ptrdiff_t rs2, vbfloat16m1x4_t vs3,
+                                size_t vl) {
+  return __riscv_vssseg4e16(rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vssseg4e16_v_bf16m2x4(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vssseg4.nxv8bf16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], <vscale x 8 x bfloat> [[TMP3]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vssseg4e16_v_bf16m2x4(__bf16 *rs1, ptrdiff_t rs2, vbfloat16m2x4_t vs3,
+                                size_t vl) {
+  return __riscv_vssseg4e16(rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vssseg4e16_v_bf16mf4x4_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vssseg4.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], ptr [[RS1]], i64 [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vssseg4e16_v_bf16mf4x4_m(vbool64_t vm, __bf16 *rs1, ptrdiff_t rs2,
+                                   vbfloat16mf4x4_t vs3, size_t vl) {
+  return __riscv_vssseg4e16(vm, rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vssseg4e16_v_bf16mf2x4_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vssseg4.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], ptr [[RS1]], i64 [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vssseg4e16_v_bf16mf2x4_m(vbool32_t vm, __bf16 *rs1, ptrdiff_t rs2,
+                                   vbfloat16mf2x4_t vs3, size_t vl) {
+  return __riscv_vssseg4e16(vm, rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vssseg4e16_v_bf16m1x4_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vssseg4.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], ptr [[RS1]], i64 [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vssseg4e16_v_bf16m1x4_m(vbool16_t vm, __bf16 *rs1, ptrdiff_t rs2,
+                                  vbfloat16m1x4_t vs3, size_t vl) {
+  return __riscv_vssseg4e16(vm, rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vssseg4e16_v_bf16m2x4_m(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vssseg4.mask.nxv8bf16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], <vscale x 8 x bfloat> [[TMP3]], ptr [[RS1]], i64 [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vssseg4e16_v_bf16m2x4_m(vbool8_t vm, __bf16 *rs1, ptrdiff_t rs2,
+                                  vbfloat16m2x4_t vs3, size_t vl) {
+  return __riscv_vssseg4e16(vm, rs1, rs2, vs3, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vssseg5e16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vssseg5e16.c
new file mode 100644
index 000000000000..00cc1066474b
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vssseg5e16.c
@@ -0,0 +1,105 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local void @test_vssseg5e16_v_bf16mf4x5(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vssseg5.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vssseg5e16_v_bf16mf4x5(__bf16 *rs1, ptrdiff_t rs2,
+                                 vbfloat16mf4x5_t vs3, size_t vl) {
+  return __riscv_vssseg5e16(rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vssseg5e16_v_bf16mf2x5(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vssseg5.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vssseg5e16_v_bf16mf2x5(__bf16 *rs1, ptrdiff_t rs2,
+                                 vbfloat16mf2x5_t vs3, size_t vl) {
+  return __riscv_vssseg5e16(rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vssseg5e16_v_bf16m1x5(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vssseg5.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vssseg5e16_v_bf16m1x5(__bf16 *rs1, ptrdiff_t rs2, vbfloat16m1x5_t vs3,
+                                size_t vl) {
+  return __riscv_vssseg5e16(rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vssseg5e16_v_bf16mf4x5_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vssseg5.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], ptr [[RS1]], i64 [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vssseg5e16_v_bf16mf4x5_m(vbool64_t vm, __bf16 *rs1, ptrdiff_t rs2,
+                                   vbfloat16mf4x5_t vs3, size_t vl) {
+  return __riscv_vssseg5e16(vm, rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vssseg5e16_v_bf16mf2x5_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vssseg5.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], ptr [[RS1]], i64 [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vssseg5e16_v_bf16mf2x5_m(vbool32_t vm, __bf16 *rs1, ptrdiff_t rs2,
+                                   vbfloat16mf2x5_t vs3, size_t vl) {
+  return __riscv_vssseg5e16(vm, rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vssseg5e16_v_bf16m1x5_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vssseg5.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], ptr [[RS1]], i64 [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vssseg5e16_v_bf16m1x5_m(vbool16_t vm, __bf16 *rs1, ptrdiff_t rs2,
+                                  vbfloat16m1x5_t vs3, size_t vl) {
+  return __riscv_vssseg5e16(vm, rs1, rs2, vs3, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vssseg6e16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vssseg6e16.c
new file mode 100644
index 000000000000..8be7e9fe56aa
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vssseg6e16.c
@@ -0,0 +1,111 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local void @test_vssseg6e16_v_bf16mf4x6(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vssseg6.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vssseg6e16_v_bf16mf4x6(__bf16 *rs1, ptrdiff_t rs2,
+                                 vbfloat16mf4x6_t vs3, size_t vl) {
+  return __riscv_vssseg6e16(rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vssseg6e16_v_bf16mf2x6(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vssseg6.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vssseg6e16_v_bf16mf2x6(__bf16 *rs1, ptrdiff_t rs2,
+                                 vbfloat16mf2x6_t vs3, size_t vl) {
+  return __riscv_vssseg6e16(rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vssseg6e16_v_bf16m1x6(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vssseg6.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vssseg6e16_v_bf16m1x6(__bf16 *rs1, ptrdiff_t rs2, vbfloat16m1x6_t vs3,
+                                size_t vl) {
+  return __riscv_vssseg6e16(rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vssseg6e16_v_bf16mf4x6_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vssseg6.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], ptr [[RS1]], i64 [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vssseg6e16_v_bf16mf4x6_m(vbool64_t vm, __bf16 *rs1, ptrdiff_t rs2,
+                                   vbfloat16mf4x6_t vs3, size_t vl) {
+  return __riscv_vssseg6e16(vm, rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vssseg6e16_v_bf16mf2x6_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vssseg6.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], ptr [[RS1]], i64 [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vssseg6e16_v_bf16mf2x6_m(vbool32_t vm, __bf16 *rs1, ptrdiff_t rs2,
+                                   vbfloat16mf2x6_t vs3, size_t vl) {
+  return __riscv_vssseg6e16(vm, rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vssseg6e16_v_bf16m1x6_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vssseg6.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], ptr [[RS1]], i64 [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vssseg6e16_v_bf16m1x6_m(vbool16_t vm, __bf16 *rs1, ptrdiff_t rs2,
+                                  vbfloat16m1x6_t vs3, size_t vl) {
+  return __riscv_vssseg6e16(vm, rs1, rs2, vs3, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vssseg7e16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vssseg7e16.c
new file mode 100644
index 000000000000..28c17de2ea1d
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vssseg7e16.c
@@ -0,0 +1,117 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local void @test_vssseg7e16_v_bf16mf4x7(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 6
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vssseg7.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vssseg7e16_v_bf16mf4x7(__bf16 *rs1, ptrdiff_t rs2,
+                                 vbfloat16mf4x7_t vs3, size_t vl) {
+  return __riscv_vssseg7e16(rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vssseg7e16_v_bf16mf2x7(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 6
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vssseg7.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vssseg7e16_v_bf16mf2x7(__bf16 *rs1, ptrdiff_t rs2,
+                                 vbfloat16mf2x7_t vs3, size_t vl) {
+  return __riscv_vssseg7e16(rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vssseg7e16_v_bf16m1x7(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 6
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vssseg7.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vssseg7e16_v_bf16m1x7(__bf16 *rs1, ptrdiff_t rs2, vbfloat16m1x7_t vs3,
+                                size_t vl) {
+  return __riscv_vssseg7e16(rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vssseg7e16_v_bf16mf4x7_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 6
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vssseg7.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], ptr [[RS1]], i64 [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vssseg7e16_v_bf16mf4x7_m(vbool64_t vm, __bf16 *rs1, ptrdiff_t rs2,
+                                   vbfloat16mf4x7_t vs3, size_t vl) {
+  return __riscv_vssseg7e16(vm, rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vssseg7e16_v_bf16mf2x7_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 6
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vssseg7.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], ptr [[RS1]], i64 [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vssseg7e16_v_bf16mf2x7_m(vbool32_t vm, __bf16 *rs1, ptrdiff_t rs2,
+                                   vbfloat16mf2x7_t vs3, size_t vl) {
+  return __riscv_vssseg7e16(vm, rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vssseg7e16_v_bf16m1x7_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 6
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vssseg7.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], ptr [[RS1]], i64 [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vssseg7e16_v_bf16m1x7_m(vbool16_t vm, __bf16 *rs1, ptrdiff_t rs2,
+                                  vbfloat16m1x7_t vs3, size_t vl) {
+  return __riscv_vssseg7e16(vm, rs1, rs2, vs3, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vssseg8e16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vssseg8e16.c
new file mode 100644
index 000000000000..90f0e46cf976
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vssseg8e16.c
@@ -0,0 +1,123 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local void @test_vssseg8e16_v_bf16mf4x8(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 7
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vssseg8.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], <vscale x 1 x bfloat> [[TMP7]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vssseg8e16_v_bf16mf4x8(__bf16 *rs1, ptrdiff_t rs2,
+                                 vbfloat16mf4x8_t vs3, size_t vl) {
+  return __riscv_vssseg8e16(rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vssseg8e16_v_bf16mf2x8(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 7
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vssseg8.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], <vscale x 2 x bfloat> [[TMP7]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vssseg8e16_v_bf16mf2x8(__bf16 *rs1, ptrdiff_t rs2,
+                                 vbfloat16mf2x8_t vs3, size_t vl) {
+  return __riscv_vssseg8e16(rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vssseg8e16_v_bf16m1x8(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 7
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vssseg8.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], <vscale x 4 x bfloat> [[TMP7]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vssseg8e16_v_bf16m1x8(__bf16 *rs1, ptrdiff_t rs2, vbfloat16m1x8_t vs3,
+                                size_t vl) {
+  return __riscv_vssseg8e16(rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vssseg8e16_v_bf16mf4x8_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 7
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vssseg8.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], <vscale x 1 x bfloat> [[TMP7]], ptr [[RS1]], i64 [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vssseg8e16_v_bf16mf4x8_m(vbool64_t vm, __bf16 *rs1, ptrdiff_t rs2,
+                                   vbfloat16mf4x8_t vs3, size_t vl) {
+  return __riscv_vssseg8e16(vm, rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vssseg8e16_v_bf16mf2x8_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 7
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vssseg8.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], <vscale x 2 x bfloat> [[TMP7]], ptr [[RS1]], i64 [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vssseg8e16_v_bf16mf2x8_m(vbool32_t vm, __bf16 *rs1, ptrdiff_t rs2,
+                                   vbfloat16mf2x8_t vs3, size_t vl) {
+  return __riscv_vssseg8e16(vm, rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vssseg8e16_v_bf16m1x8_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 7
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vssseg8.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], <vscale x 4 x bfloat> [[TMP7]], ptr [[RS1]], i64 [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vssseg8e16_v_bf16m1x8_m(vbool16_t vm, __bf16 *rs1, ptrdiff_t rs2,
+                                  vbfloat16m1x8_t vs3, size_t vl) {
+  return __riscv_vssseg8e16(vm, rs1, rs2, vs3, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vsuxei16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vsuxei16.c
new file mode 100644
index 000000000000..ef8083df5cfb
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vsuxei16.c
@@ -0,0 +1,141 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxei16_v_bf16mf4(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], <vscale x 1 x bfloat> [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxei.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[VS3]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxei16_v_bf16mf4(__bf16 *rs1, vuint16mf4_t rs2, vbfloat16mf4_t vs3,
+                             size_t vl) {
+  return __riscv_vsuxei16(rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxei16_v_bf16mf2(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], <vscale x 2 x bfloat> [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxei.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[VS3]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxei16_v_bf16mf2(__bf16 *rs1, vuint16mf2_t rs2, vbfloat16mf2_t vs3,
+                             size_t vl) {
+  return __riscv_vsuxei16(rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxei16_v_bf16m1(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], <vscale x 4 x bfloat> [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxei.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[VS3]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxei16_v_bf16m1(__bf16 *rs1, vuint16m1_t rs2, vbfloat16m1_t vs3,
+                            size_t vl) {
+  return __riscv_vsuxei16(rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxei16_v_bf16m2(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[RS2:%.*]], <vscale x 8 x bfloat> [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxei.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> [[VS3]], ptr [[RS1]], <vscale x 8 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxei16_v_bf16m2(__bf16 *rs1, vuint16m2_t rs2, vbfloat16m2_t vs3,
+                            size_t vl) {
+  return __riscv_vsuxei16(rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxei16_v_bf16m4(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 16 x i16> [[RS2:%.*]], <vscale x 16 x bfloat> [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxei.nxv16bf16.nxv16i16.i64(<vscale x 16 x bfloat> [[VS3]], ptr [[RS1]], <vscale x 16 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxei16_v_bf16m4(__bf16 *rs1, vuint16m4_t rs2, vbfloat16m4_t vs3,
+                            size_t vl) {
+  return __riscv_vsuxei16(rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxei16_v_bf16m8(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 32 x i16> [[RS2:%.*]], <vscale x 32 x bfloat> [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxei.nxv32bf16.nxv32i16.i64(<vscale x 32 x bfloat> [[VS3]], ptr [[RS1]], <vscale x 32 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxei16_v_bf16m8(__bf16 *rs1, vuint16m8_t rs2, vbfloat16m8_t vs3,
+                            size_t vl) {
+  return __riscv_vsuxei16(rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxei16_v_bf16mf4_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], <vscale x 1 x bfloat> [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxei.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[VS3]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxei16_v_bf16mf4_m(vbool64_t vm, __bf16 *rs1, vuint16mf4_t rs2,
+                               vbfloat16mf4_t vs3, size_t vl) {
+  return __riscv_vsuxei16(vm, rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxei16_v_bf16mf2_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], <vscale x 2 x bfloat> [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxei.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[VS3]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxei16_v_bf16mf2_m(vbool32_t vm, __bf16 *rs1, vuint16mf2_t rs2,
+                               vbfloat16mf2_t vs3, size_t vl) {
+  return __riscv_vsuxei16(vm, rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxei16_v_bf16m1_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], <vscale x 4 x bfloat> [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxei.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[VS3]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxei16_v_bf16m1_m(vbool16_t vm, __bf16 *rs1, vuint16m1_t rs2,
+                              vbfloat16m1_t vs3, size_t vl) {
+  return __riscv_vsuxei16(vm, rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxei16_v_bf16m2_m(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[RS2:%.*]], <vscale x 8 x bfloat> [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxei.mask.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> [[VS3]], ptr [[RS1]], <vscale x 8 x i16> [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxei16_v_bf16m2_m(vbool8_t vm, __bf16 *rs1, vuint16m2_t rs2,
+                              vbfloat16m2_t vs3, size_t vl) {
+  return __riscv_vsuxei16(vm, rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxei16_v_bf16m4_m(
+// CHECK-RV64-SAME: <vscale x 16 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 16 x i16> [[RS2:%.*]], <vscale x 16 x bfloat> [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxei.mask.nxv16bf16.nxv16i16.i64(<vscale x 16 x bfloat> [[VS3]], ptr [[RS1]], <vscale x 16 x i16> [[RS2]], <vscale x 16 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxei16_v_bf16m4_m(vbool4_t vm, __bf16 *rs1, vuint16m4_t rs2,
+                              vbfloat16m4_t vs3, size_t vl) {
+  return __riscv_vsuxei16(vm, rs1, rs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxei16_v_bf16m8_m(
+// CHECK-RV64-SAME: <vscale x 32 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 32 x i16> [[RS2:%.*]], <vscale x 32 x bfloat> [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxei.mask.nxv32bf16.nxv32i16.i64(<vscale x 32 x bfloat> [[VS3]], ptr [[RS1]], <vscale x 32 x i16> [[RS2]], <vscale x 32 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxei16_v_bf16m8_m(vbool2_t vm, __bf16 *rs1, vuint16m8_t rs2,
+                              vbfloat16m8_t vs3, size_t vl) {
+  return __riscv_vsuxei16(vm, rs1, rs2, vs3, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vsuxseg2ei16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vsuxseg2ei16.c
new file mode 100644
index 000000000000..0ca942972369
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vsuxseg2ei16.c
@@ -0,0 +1,141 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxseg2ei16_v_bf16mf4x2(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[VS2:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxseg2.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 1 x i16> [[VS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxseg2ei16_v_bf16mf4x2(__bf16 *rs1, vuint16mf4_t vs2,
+                                   vbfloat16mf4x2_t vs3, size_t vl) {
+  return __riscv_vsuxseg2ei16(rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxseg2ei16_v_bf16mf2x2(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[VS2:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxseg2.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 2 x i16> [[VS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxseg2ei16_v_bf16mf2x2(__bf16 *rs1, vuint16mf2_t vs2,
+                                   vbfloat16mf2x2_t vs3, size_t vl) {
+  return __riscv_vsuxseg2ei16(rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxseg2ei16_v_bf16m1x2(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[VS2:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxseg2.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 4 x i16> [[VS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxseg2ei16_v_bf16m1x2(__bf16 *rs1, vuint16m1_t vs2,
+                                  vbfloat16m1x2_t vs3, size_t vl) {
+  return __riscv_vsuxseg2ei16(rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxseg2ei16_v_bf16m2x2(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[VS2:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxseg2.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 8 x i16> [[VS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxseg2ei16_v_bf16m2x2(__bf16 *rs1, vuint16m2_t vs2,
+                                  vbfloat16m2x2_t vs3, size_t vl) {
+  return __riscv_vsuxseg2ei16(rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxseg2ei16_v_bf16m4x2(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 16 x i16> [[VS2:%.*]], { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxseg2.nxv16bf16.nxv16i16.i64(<vscale x 16 x bfloat> [[TMP0]], <vscale x 16 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 16 x i16> [[VS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxseg2ei16_v_bf16m4x2(__bf16 *rs1, vuint16m4_t vs2,
+                                  vbfloat16m4x2_t vs3, size_t vl) {
+  return __riscv_vsuxseg2ei16(rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxseg2ei16_v_bf16mf4x2_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[VS2:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxseg2.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 1 x i16> [[VS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxseg2ei16_v_bf16mf4x2_m(vbool64_t vm, __bf16 *rs1,
+                                     vuint16mf4_t vs2, vbfloat16mf4x2_t vs3,
+                                     size_t vl) {
+  return __riscv_vsuxseg2ei16(vm, rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxseg2ei16_v_bf16mf2x2_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[VS2:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxseg2.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 2 x i16> [[VS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxseg2ei16_v_bf16mf2x2_m(vbool32_t vm, __bf16 *rs1,
+                                     vuint16mf2_t vs2, vbfloat16mf2x2_t vs3,
+                                     size_t vl) {
+  return __riscv_vsuxseg2ei16(vm, rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxseg2ei16_v_bf16m1x2_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[VS2:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxseg2.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 4 x i16> [[VS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxseg2ei16_v_bf16m1x2_m(vbool16_t vm, __bf16 *rs1, vuint16m1_t vs2,
+                                    vbfloat16m1x2_t vs3, size_t vl) {
+  return __riscv_vsuxseg2ei16(vm, rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxseg2ei16_v_bf16m2x2_m(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[VS2:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxseg2.mask.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 8 x i16> [[VS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxseg2ei16_v_bf16m2x2_m(vbool8_t vm, __bf16 *rs1, vuint16m2_t vs2,
+                                    vbfloat16m2x2_t vs3, size_t vl) {
+  return __riscv_vsuxseg2ei16(vm, rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxseg2ei16_v_bf16m4x2_m(
+// CHECK-RV64-SAME: <vscale x 16 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 16 x i16> [[VS2:%.*]], { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxseg2.mask.nxv16bf16.nxv16i16.i64(<vscale x 16 x bfloat> [[TMP0]], <vscale x 16 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 16 x i16> [[VS2]], <vscale x 16 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxseg2ei16_v_bf16m4x2_m(vbool4_t vm, __bf16 *rs1, vuint16m4_t vs2,
+                                    vbfloat16m4x2_t vs3, size_t vl) {
+  return __riscv_vsuxseg2ei16(vm, rs1, vs2, vs3, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vsuxseg3ei16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vsuxseg3ei16.c
new file mode 100644
index 000000000000..873af1b1a1d8
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vsuxseg3ei16.c
@@ -0,0 +1,123 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxseg3ei16_v_bf16mf4x3(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[VS2:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxseg3.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 1 x i16> [[VS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxseg3ei16_v_bf16mf4x3(__bf16 *rs1, vuint16mf4_t vs2,
+                                   vbfloat16mf4x3_t vs3, size_t vl) {
+  return __riscv_vsuxseg3ei16(rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxseg3ei16_v_bf16mf2x3(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[VS2:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxseg3.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 2 x i16> [[VS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxseg3ei16_v_bf16mf2x3(__bf16 *rs1, vuint16mf2_t vs2,
+                                   vbfloat16mf2x3_t vs3, size_t vl) {
+  return __riscv_vsuxseg3ei16(rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxseg3ei16_v_bf16m1x3(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[VS2:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxseg3.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 4 x i16> [[VS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxseg3ei16_v_bf16m1x3(__bf16 *rs1, vuint16m1_t vs2,
+                                  vbfloat16m1x3_t vs3, size_t vl) {
+  return __riscv_vsuxseg3ei16(rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxseg3ei16_v_bf16m2x3(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[VS2:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxseg3.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 8 x i16> [[VS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxseg3ei16_v_bf16m2x3(__bf16 *rs1, vuint16m2_t vs2,
+                                  vbfloat16m2x3_t vs3, size_t vl) {
+  return __riscv_vsuxseg3ei16(rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxseg3ei16_v_bf16mf4x3_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[VS2:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxseg3.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 1 x i16> [[VS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxseg3ei16_v_bf16mf4x3_m(vbool64_t vm, __bf16 *rs1,
+                                     vuint16mf4_t vs2, vbfloat16mf4x3_t vs3,
+                                     size_t vl) {
+  return __riscv_vsuxseg3ei16(vm, rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxseg3ei16_v_bf16mf2x3_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[VS2:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxseg3.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 2 x i16> [[VS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxseg3ei16_v_bf16mf2x3_m(vbool32_t vm, __bf16 *rs1,
+                                     vuint16mf2_t vs2, vbfloat16mf2x3_t vs3,
+                                     size_t vl) {
+  return __riscv_vsuxseg3ei16(vm, rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxseg3ei16_v_bf16m1x3_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[VS2:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxseg3.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 4 x i16> [[VS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxseg3ei16_v_bf16m1x3_m(vbool16_t vm, __bf16 *rs1, vuint16m1_t vs2,
+                                    vbfloat16m1x3_t vs3, size_t vl) {
+  return __riscv_vsuxseg3ei16(vm, rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxseg3ei16_v_bf16m2x3_m(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[VS2:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxseg3.mask.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 8 x i16> [[VS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxseg3ei16_v_bf16m2x3_m(vbool8_t vm, __bf16 *rs1, vuint16m2_t vs2,
+                                    vbfloat16m2x3_t vs3, size_t vl) {
+  return __riscv_vsuxseg3ei16(vm, rs1, vs2, vs3, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vsuxseg4ei16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vsuxseg4ei16.c
new file mode 100644
index 000000000000..f0c497e10797
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vsuxseg4ei16.c
@@ -0,0 +1,131 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxseg4ei16_v_bf16mf4x4(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[VS2:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxseg4.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 1 x i16> [[VS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxseg4ei16_v_bf16mf4x4(__bf16 *rs1, vuint16mf4_t vs2,
+                                   vbfloat16mf4x4_t vs3, size_t vl) {
+  return __riscv_vsuxseg4ei16(rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxseg4ei16_v_bf16mf2x4(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[VS2:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxseg4.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 2 x i16> [[VS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxseg4ei16_v_bf16mf2x4(__bf16 *rs1, vuint16mf2_t vs2,
+                                   vbfloat16mf2x4_t vs3, size_t vl) {
+  return __riscv_vsuxseg4ei16(rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxseg4ei16_v_bf16m1x4(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[VS2:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxseg4.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 4 x i16> [[VS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxseg4ei16_v_bf16m1x4(__bf16 *rs1, vuint16m1_t vs2,
+                                  vbfloat16m1x4_t vs3, size_t vl) {
+  return __riscv_vsuxseg4ei16(rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxseg4ei16_v_bf16m2x4(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[VS2:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxseg4.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], <vscale x 8 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 8 x i16> [[VS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxseg4ei16_v_bf16m2x4(__bf16 *rs1, vuint16m2_t vs2,
+                                  vbfloat16m2x4_t vs3, size_t vl) {
+  return __riscv_vsuxseg4ei16(rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxseg4ei16_v_bf16mf4x4_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[VS2:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxseg4.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 1 x i16> [[VS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxseg4ei16_v_bf16mf4x4_m(vbool64_t vm, __bf16 *rs1,
+                                     vuint16mf4_t vs2, vbfloat16mf4x4_t vs3,
+                                     size_t vl) {
+  return __riscv_vsuxseg4ei16(vm, rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxseg4ei16_v_bf16mf2x4_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[VS2:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxseg4.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 2 x i16> [[VS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxseg4ei16_v_bf16mf2x4_m(vbool32_t vm, __bf16 *rs1,
+                                     vuint16mf2_t vs2, vbfloat16mf2x4_t vs3,
+                                     size_t vl) {
+  return __riscv_vsuxseg4ei16(vm, rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxseg4ei16_v_bf16m1x4_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[VS2:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxseg4.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 4 x i16> [[VS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxseg4ei16_v_bf16m1x4_m(vbool16_t vm, __bf16 *rs1, vuint16m1_t vs2,
+                                    vbfloat16m1x4_t vs3, size_t vl) {
+  return __riscv_vsuxseg4ei16(vm, rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxseg4ei16_v_bf16m2x4_m(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[VS2:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxseg4.mask.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], <vscale x 8 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 8 x i16> [[VS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxseg4ei16_v_bf16m2x4_m(vbool8_t vm, __bf16 *rs1, vuint16m2_t vs2,
+                                    vbfloat16m2x4_t vs3, size_t vl) {
+  return __riscv_vsuxseg4ei16(vm, rs1, vs2, vs3, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vsuxseg5ei16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vsuxseg5ei16.c
new file mode 100644
index 000000000000..6a2ce44b869d
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vsuxseg5ei16.c
@@ -0,0 +1,107 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxseg5ei16_v_bf16mf4x5(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[VS2:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxseg5.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 1 x i16> [[VS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxseg5ei16_v_bf16mf4x5(__bf16 *rs1, vuint16mf4_t vs2,
+                                   vbfloat16mf4x5_t vs3, size_t vl) {
+  return __riscv_vsuxseg5ei16(rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxseg5ei16_v_bf16mf2x5(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[VS2:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxseg5.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 2 x i16> [[VS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxseg5ei16_v_bf16mf2x5(__bf16 *rs1, vuint16mf2_t vs2,
+                                   vbfloat16mf2x5_t vs3, size_t vl) {
+  return __riscv_vsuxseg5ei16(rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxseg5ei16_v_bf16m1x5(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[VS2:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxseg5.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 4 x i16> [[VS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxseg5ei16_v_bf16m1x5(__bf16 *rs1, vuint16m1_t vs2,
+                                  vbfloat16m1x5_t vs3, size_t vl) {
+  return __riscv_vsuxseg5ei16(rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxseg5ei16_v_bf16mf4x5_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[VS2:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxseg5.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 1 x i16> [[VS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxseg5ei16_v_bf16mf4x5_m(vbool64_t vm, __bf16 *rs1,
+                                     vuint16mf4_t vs2, vbfloat16mf4x5_t vs3,
+                                     size_t vl) {
+  return __riscv_vsuxseg5ei16(vm, rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxseg5ei16_v_bf16mf2x5_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[VS2:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxseg5.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 2 x i16> [[VS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxseg5ei16_v_bf16mf2x5_m(vbool32_t vm, __bf16 *rs1,
+                                     vuint16mf2_t vs2, vbfloat16mf2x5_t vs3,
+                                     size_t vl) {
+  return __riscv_vsuxseg5ei16(vm, rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxseg5ei16_v_bf16m1x5_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[VS2:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxseg5.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 4 x i16> [[VS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxseg5ei16_v_bf16m1x5_m(vbool16_t vm, __bf16 *rs1, vuint16m1_t vs2,
+                                    vbfloat16m1x5_t vs3, size_t vl) {
+  return __riscv_vsuxseg5ei16(vm, rs1, vs2, vs3, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vsuxseg6ei16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vsuxseg6ei16.c
new file mode 100644
index 000000000000..84827b645e34
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vsuxseg6ei16.c
@@ -0,0 +1,113 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxseg6ei16_v_bf16mf4x6(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[VS2:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxseg6.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 1 x i16> [[VS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxseg6ei16_v_bf16mf4x6(__bf16 *rs1, vuint16mf4_t vs2,
+                                   vbfloat16mf4x6_t vs3, size_t vl) {
+  return __riscv_vsuxseg6ei16(rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxseg6ei16_v_bf16mf2x6(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[VS2:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxseg6.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 2 x i16> [[VS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxseg6ei16_v_bf16mf2x6(__bf16 *rs1, vuint16mf2_t vs2,
+                                   vbfloat16mf2x6_t vs3, size_t vl) {
+  return __riscv_vsuxseg6ei16(rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxseg6ei16_v_bf16m1x6(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[VS2:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxseg6.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 4 x i16> [[VS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxseg6ei16_v_bf16m1x6(__bf16 *rs1, vuint16m1_t vs2,
+                                  vbfloat16m1x6_t vs3, size_t vl) {
+  return __riscv_vsuxseg6ei16(rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxseg6ei16_v_bf16mf4x6_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[VS2:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxseg6.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 1 x i16> [[VS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxseg6ei16_v_bf16mf4x6_m(vbool64_t vm, __bf16 *rs1,
+                                     vuint16mf4_t vs2, vbfloat16mf4x6_t vs3,
+                                     size_t vl) {
+  return __riscv_vsuxseg6ei16(vm, rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxseg6ei16_v_bf16mf2x6_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[VS2:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxseg6.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 2 x i16> [[VS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxseg6ei16_v_bf16mf2x6_m(vbool32_t vm, __bf16 *rs1,
+                                     vuint16mf2_t vs2, vbfloat16mf2x6_t vs3,
+                                     size_t vl) {
+  return __riscv_vsuxseg6ei16(vm, rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxseg6ei16_v_bf16m1x6_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[VS2:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxseg6.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 4 x i16> [[VS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxseg6ei16_v_bf16m1x6_m(vbool16_t vm, __bf16 *rs1, vuint16m1_t vs2,
+                                    vbfloat16m1x6_t vs3, size_t vl) {
+  return __riscv_vsuxseg6ei16(vm, rs1, vs2, vs3, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vsuxseg7ei16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vsuxseg7ei16.c
new file mode 100644
index 000000000000..e8c00aaebe86
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vsuxseg7ei16.c
@@ -0,0 +1,119 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxseg7ei16_v_bf16mf4x7(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[VS2:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 6
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxseg7.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 1 x i16> [[VS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxseg7ei16_v_bf16mf4x7(__bf16 *rs1, vuint16mf4_t vs2,
+                                   vbfloat16mf4x7_t vs3, size_t vl) {
+  return __riscv_vsuxseg7ei16(rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxseg7ei16_v_bf16mf2x7(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[VS2:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 6
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxseg7.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 2 x i16> [[VS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxseg7ei16_v_bf16mf2x7(__bf16 *rs1, vuint16mf2_t vs2,
+                                   vbfloat16mf2x7_t vs3, size_t vl) {
+  return __riscv_vsuxseg7ei16(rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxseg7ei16_v_bf16m1x7(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[VS2:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 6
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxseg7.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 4 x i16> [[VS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxseg7ei16_v_bf16m1x7(__bf16 *rs1, vuint16m1_t vs2,
+                                  vbfloat16m1x7_t vs3, size_t vl) {
+  return __riscv_vsuxseg7ei16(rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxseg7ei16_v_bf16mf4x7_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[VS2:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 6
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxseg7.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 1 x i16> [[VS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxseg7ei16_v_bf16mf4x7_m(vbool64_t vm, __bf16 *rs1,
+                                     vuint16mf4_t vs2, vbfloat16mf4x7_t vs3,
+                                     size_t vl) {
+  return __riscv_vsuxseg7ei16(vm, rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxseg7ei16_v_bf16mf2x7_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[VS2:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 6
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxseg7.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 2 x i16> [[VS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxseg7ei16_v_bf16mf2x7_m(vbool32_t vm, __bf16 *rs1,
+                                     vuint16mf2_t vs2, vbfloat16mf2x7_t vs3,
+                                     size_t vl) {
+  return __riscv_vsuxseg7ei16(vm, rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxseg7ei16_v_bf16m1x7_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[VS2:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 6
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxseg7.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 4 x i16> [[VS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxseg7ei16_v_bf16m1x7_m(vbool16_t vm, __bf16 *rs1, vuint16m1_t vs2,
+                                    vbfloat16m1x7_t vs3, size_t vl) {
+  return __riscv_vsuxseg7ei16(vm, rs1, vs2, vs3, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vsuxseg8ei16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vsuxseg8ei16.c
new file mode 100644
index 000000000000..ff96cef336ea
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/bfloat16/vsuxseg8ei16.c
@@ -0,0 +1,125 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxseg8ei16_v_bf16mf4x8(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[VS2:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 7
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxseg8.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], <vscale x 1 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 1 x i16> [[VS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxseg8ei16_v_bf16mf4x8(__bf16 *rs1, vuint16mf4_t vs2,
+                                   vbfloat16mf4x8_t vs3, size_t vl) {
+  return __riscv_vsuxseg8ei16(rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxseg8ei16_v_bf16mf2x8(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[VS2:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 7
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxseg8.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], <vscale x 2 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 2 x i16> [[VS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxseg8ei16_v_bf16mf2x8(__bf16 *rs1, vuint16mf2_t vs2,
+                                   vbfloat16mf2x8_t vs3, size_t vl) {
+  return __riscv_vsuxseg8ei16(rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxseg8ei16_v_bf16m1x8(
+// CHECK-RV64-SAME: ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[VS2:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 7
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxseg8.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], <vscale x 4 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 4 x i16> [[VS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxseg8ei16_v_bf16m1x8(__bf16 *rs1, vuint16m1_t vs2,
+                                  vbfloat16m1x8_t vs3, size_t vl) {
+  return __riscv_vsuxseg8ei16(rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxseg8ei16_v_bf16mf4x8_m(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[VS2:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VS3]], 7
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxseg8.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], <vscale x 1 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 1 x i16> [[VS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxseg8ei16_v_bf16mf4x8_m(vbool64_t vm, __bf16 *rs1,
+                                     vuint16mf4_t vs2, vbfloat16mf4x8_t vs3,
+                                     size_t vl) {
+  return __riscv_vsuxseg8ei16(vm, rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxseg8ei16_v_bf16mf2x8_m(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[VS2:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VS3]], 7
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxseg8.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], <vscale x 2 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 2 x i16> [[VS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxseg8ei16_v_bf16mf2x8_m(vbool32_t vm, __bf16 *rs1,
+                                     vuint16mf2_t vs2, vbfloat16mf2x8_t vs3,
+                                     size_t vl) {
+  return __riscv_vsuxseg8ei16(vm, rs1, vs2, vs3, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local void @test_vsuxseg8ei16_v_bf16m1x8_m(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[VS2:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VS3]], 7
+// CHECK-RV64-NEXT:    call void @llvm.riscv.vsuxseg8.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], <vscale x 4 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 4 x i16> [[VS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret void
+//
+void test_vsuxseg8ei16_v_bf16m1x8_m(vbool16_t vm, __bf16 *rs1, vuint16m1_t vs2,
+                                    vbfloat16m1x8_t vs3, size_t vl) {
+  return __riscv_vsuxseg8ei16(vm, rs1, vs2, vs3, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/vaesdf.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/vaesdf.c
index 9c91d2a1192d..ec3cd1a5a696 100644
--- a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/vaesdf.c
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/vaesdf.c
@@ -206,13 +206,3 @@ vuint32m8_t test_vaesdf_vv_u32m8(vuint32m8_t vd, vuint32m8_t vs2, size_t vl) {
   return __riscv_vaesdf_vv(vd, vs2, vl);
 }
 
-// CHECK-RV64-LABEL: define dso_local <vscale x 16 x i32> @test_vaesdf_vs_u32m8_u32m8
-// CHECK-RV64-SAME: (<vscale x 16 x i32> [[VD:%.*]], <vscale x 16 x i32> [[VS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
-// CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vaesdf.vs.nxv16i32.nxv16i32.i64(<vscale x 16 x i32> [[VD]], <vscale x 16 x i32> [[VS2]], i64 [[VL]], i64 3)
-// CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
-//
-vuint32m8_t test_vaesdf_vs_u32m8_u32m8(vuint32m8_t vd, vuint32m8_t vs2, size_t vl) {
-  return __riscv_vaesdf_vs(vd, vs2, vl);
-}
-
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/vaesdm.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/vaesdm.c
index dac9eb38713b..85d452543db2 100644
--- a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/vaesdm.c
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/vaesdm.c
@@ -206,13 +206,3 @@ vuint32m8_t test_vaesdm_vv_u32m8(vuint32m8_t vd, vuint32m8_t vs2, size_t vl) {
   return __riscv_vaesdm_vv(vd, vs2, vl);
 }
 
-// CHECK-RV64-LABEL: define dso_local <vscale x 16 x i32> @test_vaesdm_vs_u32m8_u32m8
-// CHECK-RV64-SAME: (<vscale x 16 x i32> [[VD:%.*]], <vscale x 16 x i32> [[VS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
-// CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vaesdm.vs.nxv16i32.nxv16i32.i64(<vscale x 16 x i32> [[VD]], <vscale x 16 x i32> [[VS2]], i64 [[VL]], i64 3)
-// CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
-//
-vuint32m8_t test_vaesdm_vs_u32m8_u32m8(vuint32m8_t vd, vuint32m8_t vs2, size_t vl) {
-  return __riscv_vaesdm_vs(vd, vs2, vl);
-}
-
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/vaesef.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/vaesef.c
index 01875afa525e..6727ff24b83d 100644
--- a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/vaesef.c
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/vaesef.c
@@ -206,13 +206,3 @@ vuint32m8_t test_vaesef_vv_u32m8(vuint32m8_t vd, vuint32m8_t vs2, size_t vl) {
   return __riscv_vaesef_vv(vd, vs2, vl);
 }
 
-// CHECK-RV64-LABEL: define dso_local <vscale x 16 x i32> @test_vaesef_vs_u32m8_u32m8
-// CHECK-RV64-SAME: (<vscale x 16 x i32> [[VD:%.*]], <vscale x 16 x i32> [[VS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
-// CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vaesef.vs.nxv16i32.nxv16i32.i64(<vscale x 16 x i32> [[VD]], <vscale x 16 x i32> [[VS2]], i64 [[VL]], i64 3)
-// CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
-//
-vuint32m8_t test_vaesef_vs_u32m8_u32m8(vuint32m8_t vd, vuint32m8_t vs2, size_t vl) {
-  return __riscv_vaesef_vs(vd, vs2, vl);
-}
-
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/vaesem.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/vaesem.c
index e9a0add21c92..6ebfd5ff4962 100644
--- a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/vaesem.c
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/vaesem.c
@@ -206,13 +206,3 @@ vuint32m8_t test_vaesem_vv_u32m8(vuint32m8_t vd, vuint32m8_t vs2, size_t vl) {
   return __riscv_vaesem_vv(vd, vs2, vl);
 }
 
-// CHECK-RV64-LABEL: define dso_local <vscale x 16 x i32> @test_vaesem_vs_u32m8_u32m8
-// CHECK-RV64-SAME: (<vscale x 16 x i32> [[VD:%.*]], <vscale x 16 x i32> [[VS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
-// CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vaesem.vs.nxv16i32.nxv16i32.i64(<vscale x 16 x i32> [[VD]], <vscale x 16 x i32> [[VS2]], i64 [[VL]], i64 3)
-// CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
-//
-vuint32m8_t test_vaesem_vs_u32m8_u32m8(vuint32m8_t vd, vuint32m8_t vs2, size_t vl) {
-  return __riscv_vaesem_vs(vd, vs2, vl);
-}
-
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/vaesz.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/vaesz.c
index c80e4ebce5e5..fe657a603d7b 100644
--- a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/vaesz.c
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/vaesz.c
@@ -156,13 +156,3 @@ vuint32m8_t test_vaesz_vs_u32m4_u32m8(vuint32m8_t vd, vuint32m4_t vs2, size_t vl
   return __riscv_vaesz(vd, vs2, vl);
 }
 
-// CHECK-RV64-LABEL: define dso_local <vscale x 16 x i32> @test_vaesz_vs_u32m8_u32m8
-// CHECK-RV64-SAME: (<vscale x 16 x i32> [[VD:%.*]], <vscale x 16 x i32> [[VS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
-// CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vaesz.vs.nxv16i32.nxv16i32.i64(<vscale x 16 x i32> [[VD]], <vscale x 16 x i32> [[VS2]], i64 [[VL]], i64 3)
-// CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
-//
-vuint32m8_t test_vaesz_vs_u32m8_u32m8(vuint32m8_t vd, vuint32m8_t vs2, size_t vl) {
-  return __riscv_vaesz(vd, vs2, vl);
-}
-
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/vsm4r.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/vsm4r.c
index cfc8532835ea..82f275af3e50 100644
--- a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/vsm4r.c
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/non-policy/overloaded/vsm4r.c
@@ -206,13 +206,3 @@ vuint32m8_t test_vsm4r_vv_u32m8(vuint32m8_t vd, vuint32m8_t vs2, size_t vl) {
   return __riscv_vsm4r_vv(vd, vs2, vl);
 }
 
-// CHECK-RV64-LABEL: define dso_local <vscale x 16 x i32> @test_vsm4r_vs_u32m8_u32m8
-// CHECK-RV64-SAME: (<vscale x 16 x i32> [[VD:%.*]], <vscale x 16 x i32> [[VS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
-// CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vsm4r.vs.nxv16i32.nxv16i32.i64(<vscale x 16 x i32> [[VD]], <vscale x 16 x i32> [[VS2]], i64 [[VL]], i64 3)
-// CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
-//
-vuint32m8_t test_vsm4r_vs_u32m8_u32m8(vuint32m8_t vd, vuint32m8_t vs2, size_t vl) {
-  return __riscv_vsm4r_vs(vd, vs2, vl);
-}
-
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vle16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vle16.c
new file mode 100644
index 000000000000..53d8b4d625e7
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vle16.c
@@ -0,0 +1,249 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 1 x bfloat> @test_vle16_v_bf16mf4_tu(
+// CHECK-RV64-SAME: <vscale x 1 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x bfloat> @llvm.riscv.vle.nxv1bf16.i64(<vscale x 1 x bfloat> [[VD]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret <vscale x 1 x bfloat> [[TMP0]]
+//
+vbfloat16mf4_t test_vle16_v_bf16mf4_tu(vbfloat16mf4_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vle16_v_bf16mf4_tu(vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 2 x bfloat> @test_vle16_v_bf16mf2_tu(
+// CHECK-RV64-SAME: <vscale x 2 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x bfloat> @llvm.riscv.vle.nxv2bf16.i64(<vscale x 2 x bfloat> [[VD]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret <vscale x 2 x bfloat> [[TMP0]]
+//
+vbfloat16mf2_t test_vle16_v_bf16mf2_tu(vbfloat16mf2_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vle16_v_bf16mf2_tu(vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 4 x bfloat> @test_vle16_v_bf16m1_tu(
+// CHECK-RV64-SAME: <vscale x 4 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x bfloat> @llvm.riscv.vle.nxv4bf16.i64(<vscale x 4 x bfloat> [[VD]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret <vscale x 4 x bfloat> [[TMP0]]
+//
+vbfloat16m1_t test_vle16_v_bf16m1_tu(vbfloat16m1_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vle16_v_bf16m1_tu(vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 8 x bfloat> @test_vle16_v_bf16m2_tu(
+// CHECK-RV64-SAME: <vscale x 8 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x bfloat> @llvm.riscv.vle.nxv8bf16.i64(<vscale x 8 x bfloat> [[VD]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret <vscale x 8 x bfloat> [[TMP0]]
+//
+vbfloat16m2_t test_vle16_v_bf16m2_tu(vbfloat16m2_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vle16_v_bf16m2_tu(vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 16 x bfloat> @test_vle16_v_bf16m4_tu(
+// CHECK-RV64-SAME: <vscale x 16 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x bfloat> @llvm.riscv.vle.nxv16bf16.i64(<vscale x 16 x bfloat> [[VD]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret <vscale x 16 x bfloat> [[TMP0]]
+//
+vbfloat16m4_t test_vle16_v_bf16m4_tu(vbfloat16m4_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vle16_v_bf16m4_tu(vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 32 x bfloat> @test_vle16_v_bf16m8_tu(
+// CHECK-RV64-SAME: <vscale x 32 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x bfloat> @llvm.riscv.vle.nxv32bf16.i64(<vscale x 32 x bfloat> [[VD]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret <vscale x 32 x bfloat> [[TMP0]]
+//
+vbfloat16m8_t test_vle16_v_bf16m8_tu(vbfloat16m8_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vle16_v_bf16m8_tu(vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 1 x bfloat> @test_vle16_v_bf16mf4_tum(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], <vscale x 1 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x bfloat> @llvm.riscv.vle.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[VD]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret <vscale x 1 x bfloat> [[TMP0]]
+//
+vbfloat16mf4_t test_vle16_v_bf16mf4_tum(vbool64_t vm, vbfloat16mf4_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vle16_v_bf16mf4_tum(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 2 x bfloat> @test_vle16_v_bf16mf2_tum(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], <vscale x 2 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x bfloat> @llvm.riscv.vle.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[VD]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret <vscale x 2 x bfloat> [[TMP0]]
+//
+vbfloat16mf2_t test_vle16_v_bf16mf2_tum(vbool32_t vm, vbfloat16mf2_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vle16_v_bf16mf2_tum(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 4 x bfloat> @test_vle16_v_bf16m1_tum(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], <vscale x 4 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x bfloat> @llvm.riscv.vle.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[VD]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret <vscale x 4 x bfloat> [[TMP0]]
+//
+vbfloat16m1_t test_vle16_v_bf16m1_tum(vbool16_t vm, vbfloat16m1_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vle16_v_bf16m1_tum(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 8 x bfloat> @test_vle16_v_bf16m2_tum(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], <vscale x 8 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x bfloat> @llvm.riscv.vle.mask.nxv8bf16.i64(<vscale x 8 x bfloat> [[VD]], ptr [[RS1]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret <vscale x 8 x bfloat> [[TMP0]]
+//
+vbfloat16m2_t test_vle16_v_bf16m2_tum(vbool8_t vm, vbfloat16m2_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vle16_v_bf16m2_tum(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 16 x bfloat> @test_vle16_v_bf16m4_tum(
+// CHECK-RV64-SAME: <vscale x 16 x i1> [[VM:%.*]], <vscale x 16 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x bfloat> @llvm.riscv.vle.mask.nxv16bf16.i64(<vscale x 16 x bfloat> [[VD]], ptr [[RS1]], <vscale x 16 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret <vscale x 16 x bfloat> [[TMP0]]
+//
+vbfloat16m4_t test_vle16_v_bf16m4_tum(vbool4_t vm, vbfloat16m4_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vle16_v_bf16m4_tum(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 32 x bfloat> @test_vle16_v_bf16m8_tum(
+// CHECK-RV64-SAME: <vscale x 32 x i1> [[VM:%.*]], <vscale x 32 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x bfloat> @llvm.riscv.vle.mask.nxv32bf16.i64(<vscale x 32 x bfloat> [[VD]], ptr [[RS1]], <vscale x 32 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret <vscale x 32 x bfloat> [[TMP0]]
+//
+vbfloat16m8_t test_vle16_v_bf16m8_tum(vbool2_t vm, vbfloat16m8_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vle16_v_bf16m8_tum(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 1 x bfloat> @test_vle16_v_bf16mf4_tumu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], <vscale x 1 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x bfloat> @llvm.riscv.vle.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[VD]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 1 x bfloat> [[TMP0]]
+//
+vbfloat16mf4_t test_vle16_v_bf16mf4_tumu(vbool64_t vm, vbfloat16mf4_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vle16_v_bf16mf4_tumu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 2 x bfloat> @test_vle16_v_bf16mf2_tumu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], <vscale x 2 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x bfloat> @llvm.riscv.vle.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[VD]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 2 x bfloat> [[TMP0]]
+//
+vbfloat16mf2_t test_vle16_v_bf16mf2_tumu(vbool32_t vm, vbfloat16mf2_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vle16_v_bf16mf2_tumu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 4 x bfloat> @test_vle16_v_bf16m1_tumu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], <vscale x 4 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x bfloat> @llvm.riscv.vle.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[VD]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 4 x bfloat> [[TMP0]]
+//
+vbfloat16m1_t test_vle16_v_bf16m1_tumu(vbool16_t vm, vbfloat16m1_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vle16_v_bf16m1_tumu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 8 x bfloat> @test_vle16_v_bf16m2_tumu(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], <vscale x 8 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x bfloat> @llvm.riscv.vle.mask.nxv8bf16.i64(<vscale x 8 x bfloat> [[VD]], ptr [[RS1]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 8 x bfloat> [[TMP0]]
+//
+vbfloat16m2_t test_vle16_v_bf16m2_tumu(vbool8_t vm, vbfloat16m2_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vle16_v_bf16m2_tumu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 16 x bfloat> @test_vle16_v_bf16m4_tumu(
+// CHECK-RV64-SAME: <vscale x 16 x i1> [[VM:%.*]], <vscale x 16 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x bfloat> @llvm.riscv.vle.mask.nxv16bf16.i64(<vscale x 16 x bfloat> [[VD]], ptr [[RS1]], <vscale x 16 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 16 x bfloat> [[TMP0]]
+//
+vbfloat16m4_t test_vle16_v_bf16m4_tumu(vbool4_t vm, vbfloat16m4_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vle16_v_bf16m4_tumu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 32 x bfloat> @test_vle16_v_bf16m8_tumu(
+// CHECK-RV64-SAME: <vscale x 32 x i1> [[VM:%.*]], <vscale x 32 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x bfloat> @llvm.riscv.vle.mask.nxv32bf16.i64(<vscale x 32 x bfloat> [[VD]], ptr [[RS1]], <vscale x 32 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 32 x bfloat> [[TMP0]]
+//
+vbfloat16m8_t test_vle16_v_bf16m8_tumu(vbool2_t vm, vbfloat16m8_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vle16_v_bf16m8_tumu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 1 x bfloat> @test_vle16_v_bf16mf4_mu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], <vscale x 1 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x bfloat> @llvm.riscv.vle.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[VD]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret <vscale x 1 x bfloat> [[TMP0]]
+//
+vbfloat16mf4_t test_vle16_v_bf16mf4_mu(vbool64_t vm, vbfloat16mf4_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vle16_v_bf16mf4_mu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 2 x bfloat> @test_vle16_v_bf16mf2_mu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], <vscale x 2 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x bfloat> @llvm.riscv.vle.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[VD]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret <vscale x 2 x bfloat> [[TMP0]]
+//
+vbfloat16mf2_t test_vle16_v_bf16mf2_mu(vbool32_t vm, vbfloat16mf2_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vle16_v_bf16mf2_mu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 4 x bfloat> @test_vle16_v_bf16m1_mu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], <vscale x 4 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x bfloat> @llvm.riscv.vle.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[VD]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret <vscale x 4 x bfloat> [[TMP0]]
+//
+vbfloat16m1_t test_vle16_v_bf16m1_mu(vbool16_t vm, vbfloat16m1_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vle16_v_bf16m1_mu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 8 x bfloat> @test_vle16_v_bf16m2_mu(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], <vscale x 8 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x bfloat> @llvm.riscv.vle.mask.nxv8bf16.i64(<vscale x 8 x bfloat> [[VD]], ptr [[RS1]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret <vscale x 8 x bfloat> [[TMP0]]
+//
+vbfloat16m2_t test_vle16_v_bf16m2_mu(vbool8_t vm, vbfloat16m2_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vle16_v_bf16m2_mu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 16 x bfloat> @test_vle16_v_bf16m4_mu(
+// CHECK-RV64-SAME: <vscale x 16 x i1> [[VM:%.*]], <vscale x 16 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x bfloat> @llvm.riscv.vle.mask.nxv16bf16.i64(<vscale x 16 x bfloat> [[VD]], ptr [[RS1]], <vscale x 16 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret <vscale x 16 x bfloat> [[TMP0]]
+//
+vbfloat16m4_t test_vle16_v_bf16m4_mu(vbool4_t vm, vbfloat16m4_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vle16_v_bf16m4_mu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 32 x bfloat> @test_vle16_v_bf16m8_mu(
+// CHECK-RV64-SAME: <vscale x 32 x i1> [[VM:%.*]], <vscale x 32 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x bfloat> @llvm.riscv.vle.mask.nxv32bf16.i64(<vscale x 32 x bfloat> [[VD]], ptr [[RS1]], <vscale x 32 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret <vscale x 32 x bfloat> [[TMP0]]
+//
+vbfloat16m8_t test_vle16_v_bf16m8_mu(vbool2_t vm, vbfloat16m8_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vle16_v_bf16m8_mu(vm, vd, rs1, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vle16ff.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vle16ff.c
new file mode 100644
index 000000000000..b7e689e00ada
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vle16ff.c
@@ -0,0 +1,321 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 1 x bfloat> @test_vle16ff_v_bf16mf4_tu(
+// CHECK-RV64-SAME: <vscale x 1 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, i64 } @llvm.riscv.vleff.nxv1bf16.i64(<vscale x 1 x bfloat> [[VD]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    store i64 [[TMP2]], ptr [[NEW_VL]], align 2
+// CHECK-RV64-NEXT:    ret <vscale x 1 x bfloat> [[TMP1]]
+//
+vbfloat16mf4_t test_vle16ff_v_bf16mf4_tu(vbfloat16mf4_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vle16ff_v_bf16mf4_tu(vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 2 x bfloat> @test_vle16ff_v_bf16mf2_tu(
+// CHECK-RV64-SAME: <vscale x 2 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, i64 } @llvm.riscv.vleff.nxv2bf16.i64(<vscale x 2 x bfloat> [[VD]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    store i64 [[TMP2]], ptr [[NEW_VL]], align 2
+// CHECK-RV64-NEXT:    ret <vscale x 2 x bfloat> [[TMP1]]
+//
+vbfloat16mf2_t test_vle16ff_v_bf16mf2_tu(vbfloat16mf2_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vle16ff_v_bf16mf2_tu(vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 4 x bfloat> @test_vle16ff_v_bf16m1_tu(
+// CHECK-RV64-SAME: <vscale x 4 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, i64 } @llvm.riscv.vleff.nxv4bf16.i64(<vscale x 4 x bfloat> [[VD]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    store i64 [[TMP2]], ptr [[NEW_VL]], align 2
+// CHECK-RV64-NEXT:    ret <vscale x 4 x bfloat> [[TMP1]]
+//
+vbfloat16m1_t test_vle16ff_v_bf16m1_tu(vbfloat16m1_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vle16ff_v_bf16m1_tu(vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 8 x bfloat> @test_vle16ff_v_bf16m2_tu(
+// CHECK-RV64-SAME: <vscale x 8 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 8 x bfloat>, i64 } @llvm.riscv.vleff.nxv8bf16.i64(<vscale x 8 x bfloat> [[VD]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    store i64 [[TMP2]], ptr [[NEW_VL]], align 2
+// CHECK-RV64-NEXT:    ret <vscale x 8 x bfloat> [[TMP1]]
+//
+vbfloat16m2_t test_vle16ff_v_bf16m2_tu(vbfloat16m2_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vle16ff_v_bf16m2_tu(vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 16 x bfloat> @test_vle16ff_v_bf16m4_tu(
+// CHECK-RV64-SAME: <vscale x 16 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 16 x bfloat>, i64 } @llvm.riscv.vleff.nxv16bf16.i64(<vscale x 16 x bfloat> [[VD]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 16 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 16 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    store i64 [[TMP2]], ptr [[NEW_VL]], align 2
+// CHECK-RV64-NEXT:    ret <vscale x 16 x bfloat> [[TMP1]]
+//
+vbfloat16m4_t test_vle16ff_v_bf16m4_tu(vbfloat16m4_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vle16ff_v_bf16m4_tu(vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 32 x bfloat> @test_vle16ff_v_bf16m8_tu(
+// CHECK-RV64-SAME: <vscale x 32 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 32 x bfloat>, i64 } @llvm.riscv.vleff.nxv32bf16.i64(<vscale x 32 x bfloat> [[VD]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 32 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 32 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    store i64 [[TMP2]], ptr [[NEW_VL]], align 2
+// CHECK-RV64-NEXT:    ret <vscale x 32 x bfloat> [[TMP1]]
+//
+vbfloat16m8_t test_vle16ff_v_bf16m8_tu(vbfloat16m8_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vle16ff_v_bf16m8_tu(vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 1 x bfloat> @test_vle16ff_v_bf16mf4_tum(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], <vscale x 1 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, i64 } @llvm.riscv.vleff.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[VD]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    store i64 [[TMP2]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret <vscale x 1 x bfloat> [[TMP1]]
+//
+vbfloat16mf4_t test_vle16ff_v_bf16mf4_tum(vbool64_t vm, vbfloat16mf4_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vle16ff_v_bf16mf4_tum(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 2 x bfloat> @test_vle16ff_v_bf16mf2_tum(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], <vscale x 2 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, i64 } @llvm.riscv.vleff.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[VD]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    store i64 [[TMP2]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret <vscale x 2 x bfloat> [[TMP1]]
+//
+vbfloat16mf2_t test_vle16ff_v_bf16mf2_tum(vbool32_t vm, vbfloat16mf2_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vle16ff_v_bf16mf2_tum(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 4 x bfloat> @test_vle16ff_v_bf16m1_tum(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], <vscale x 4 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, i64 } @llvm.riscv.vleff.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[VD]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    store i64 [[TMP2]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret <vscale x 4 x bfloat> [[TMP1]]
+//
+vbfloat16m1_t test_vle16ff_v_bf16m1_tum(vbool16_t vm, vbfloat16m1_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vle16ff_v_bf16m1_tum(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 8 x bfloat> @test_vle16ff_v_bf16m2_tum(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], <vscale x 8 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 8 x bfloat>, i64 } @llvm.riscv.vleff.mask.nxv8bf16.i64(<vscale x 8 x bfloat> [[VD]], ptr [[RS1]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    store i64 [[TMP2]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret <vscale x 8 x bfloat> [[TMP1]]
+//
+vbfloat16m2_t test_vle16ff_v_bf16m2_tum(vbool8_t vm, vbfloat16m2_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vle16ff_v_bf16m2_tum(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 16 x bfloat> @test_vle16ff_v_bf16m4_tum(
+// CHECK-RV64-SAME: <vscale x 16 x i1> [[VM:%.*]], <vscale x 16 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 16 x bfloat>, i64 } @llvm.riscv.vleff.mask.nxv16bf16.i64(<vscale x 16 x bfloat> [[VD]], ptr [[RS1]], <vscale x 16 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 16 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 16 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    store i64 [[TMP2]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret <vscale x 16 x bfloat> [[TMP1]]
+//
+vbfloat16m4_t test_vle16ff_v_bf16m4_tum(vbool4_t vm, vbfloat16m4_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vle16ff_v_bf16m4_tum(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 32 x bfloat> @test_vle16ff_v_bf16m8_tum(
+// CHECK-RV64-SAME: <vscale x 32 x i1> [[VM:%.*]], <vscale x 32 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 32 x bfloat>, i64 } @llvm.riscv.vleff.mask.nxv32bf16.i64(<vscale x 32 x bfloat> [[VD]], ptr [[RS1]], <vscale x 32 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 32 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 32 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    store i64 [[TMP2]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret <vscale x 32 x bfloat> [[TMP1]]
+//
+vbfloat16m8_t test_vle16ff_v_bf16m8_tum(vbool2_t vm, vbfloat16m8_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vle16ff_v_bf16m8_tum(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 1 x bfloat> @test_vle16ff_v_bf16mf4_tumu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], <vscale x 1 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, i64 } @llvm.riscv.vleff.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[VD]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    store i64 [[TMP2]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret <vscale x 1 x bfloat> [[TMP1]]
+//
+vbfloat16mf4_t test_vle16ff_v_bf16mf4_tumu(vbool64_t vm, vbfloat16mf4_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vle16ff_v_bf16mf4_tumu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 2 x bfloat> @test_vle16ff_v_bf16mf2_tumu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], <vscale x 2 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, i64 } @llvm.riscv.vleff.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[VD]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    store i64 [[TMP2]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret <vscale x 2 x bfloat> [[TMP1]]
+//
+vbfloat16mf2_t test_vle16ff_v_bf16mf2_tumu(vbool32_t vm, vbfloat16mf2_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vle16ff_v_bf16mf2_tumu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 4 x bfloat> @test_vle16ff_v_bf16m1_tumu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], <vscale x 4 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, i64 } @llvm.riscv.vleff.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[VD]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    store i64 [[TMP2]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret <vscale x 4 x bfloat> [[TMP1]]
+//
+vbfloat16m1_t test_vle16ff_v_bf16m1_tumu(vbool16_t vm, vbfloat16m1_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vle16ff_v_bf16m1_tumu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 8 x bfloat> @test_vle16ff_v_bf16m2_tumu(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], <vscale x 8 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 8 x bfloat>, i64 } @llvm.riscv.vleff.mask.nxv8bf16.i64(<vscale x 8 x bfloat> [[VD]], ptr [[RS1]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    store i64 [[TMP2]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret <vscale x 8 x bfloat> [[TMP1]]
+//
+vbfloat16m2_t test_vle16ff_v_bf16m2_tumu(vbool8_t vm, vbfloat16m2_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vle16ff_v_bf16m2_tumu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 16 x bfloat> @test_vle16ff_v_bf16m4_tumu(
+// CHECK-RV64-SAME: <vscale x 16 x i1> [[VM:%.*]], <vscale x 16 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 16 x bfloat>, i64 } @llvm.riscv.vleff.mask.nxv16bf16.i64(<vscale x 16 x bfloat> [[VD]], ptr [[RS1]], <vscale x 16 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 16 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 16 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    store i64 [[TMP2]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret <vscale x 16 x bfloat> [[TMP1]]
+//
+vbfloat16m4_t test_vle16ff_v_bf16m4_tumu(vbool4_t vm, vbfloat16m4_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vle16ff_v_bf16m4_tumu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 32 x bfloat> @test_vle16ff_v_bf16m8_tumu(
+// CHECK-RV64-SAME: <vscale x 32 x i1> [[VM:%.*]], <vscale x 32 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 32 x bfloat>, i64 } @llvm.riscv.vleff.mask.nxv32bf16.i64(<vscale x 32 x bfloat> [[VD]], ptr [[RS1]], <vscale x 32 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 32 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 32 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    store i64 [[TMP2]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret <vscale x 32 x bfloat> [[TMP1]]
+//
+vbfloat16m8_t test_vle16ff_v_bf16m8_tumu(vbool2_t vm, vbfloat16m8_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vle16ff_v_bf16m8_tumu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 1 x bfloat> @test_vle16ff_v_bf16mf4_mu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], <vscale x 1 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, i64 } @llvm.riscv.vleff.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[VD]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    store i64 [[TMP2]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret <vscale x 1 x bfloat> [[TMP1]]
+//
+vbfloat16mf4_t test_vle16ff_v_bf16mf4_mu(vbool64_t vm, vbfloat16mf4_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vle16ff_v_bf16mf4_mu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 2 x bfloat> @test_vle16ff_v_bf16mf2_mu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], <vscale x 2 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, i64 } @llvm.riscv.vleff.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[VD]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    store i64 [[TMP2]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret <vscale x 2 x bfloat> [[TMP1]]
+//
+vbfloat16mf2_t test_vle16ff_v_bf16mf2_mu(vbool32_t vm, vbfloat16mf2_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vle16ff_v_bf16mf2_mu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 4 x bfloat> @test_vle16ff_v_bf16m1_mu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], <vscale x 4 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, i64 } @llvm.riscv.vleff.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[VD]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    store i64 [[TMP2]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret <vscale x 4 x bfloat> [[TMP1]]
+//
+vbfloat16m1_t test_vle16ff_v_bf16m1_mu(vbool16_t vm, vbfloat16m1_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vle16ff_v_bf16m1_mu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 8 x bfloat> @test_vle16ff_v_bf16m2_mu(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], <vscale x 8 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 8 x bfloat>, i64 } @llvm.riscv.vleff.mask.nxv8bf16.i64(<vscale x 8 x bfloat> [[VD]], ptr [[RS1]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    store i64 [[TMP2]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret <vscale x 8 x bfloat> [[TMP1]]
+//
+vbfloat16m2_t test_vle16ff_v_bf16m2_mu(vbool8_t vm, vbfloat16m2_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vle16ff_v_bf16m2_mu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 16 x bfloat> @test_vle16ff_v_bf16m4_mu(
+// CHECK-RV64-SAME: <vscale x 16 x i1> [[VM:%.*]], <vscale x 16 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 16 x bfloat>, i64 } @llvm.riscv.vleff.mask.nxv16bf16.i64(<vscale x 16 x bfloat> [[VD]], ptr [[RS1]], <vscale x 16 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 16 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 16 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    store i64 [[TMP2]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret <vscale x 16 x bfloat> [[TMP1]]
+//
+vbfloat16m4_t test_vle16ff_v_bf16m4_mu(vbool4_t vm, vbfloat16m4_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vle16ff_v_bf16m4_mu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 32 x bfloat> @test_vle16ff_v_bf16m8_mu(
+// CHECK-RV64-SAME: <vscale x 32 x i1> [[VM:%.*]], <vscale x 32 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 32 x bfloat>, i64 } @llvm.riscv.vleff.mask.nxv32bf16.i64(<vscale x 32 x bfloat> [[VD]], ptr [[RS1]], <vscale x 32 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 32 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 32 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    store i64 [[TMP2]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret <vscale x 32 x bfloat> [[TMP1]]
+//
+vbfloat16m8_t test_vle16ff_v_bf16m8_mu(vbool2_t vm, vbfloat16m8_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vle16ff_v_bf16m8_mu(vm, vd, rs1, new_vl, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vloxei16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vloxei16.c
new file mode 100644
index 000000000000..c0cacbf48d01
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vloxei16.c
@@ -0,0 +1,249 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 1 x bfloat> @test_vloxei16_v_bf16mf4_tu(
+// CHECK-RV64-SAME: <vscale x 1 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x bfloat> @llvm.riscv.vloxei.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[VD]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret <vscale x 1 x bfloat> [[TMP0]]
+//
+vbfloat16mf4_t test_vloxei16_v_bf16mf4_tu(vbfloat16mf4_t vd, const __bf16 *rs1, vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vloxei16_v_bf16mf4_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 2 x bfloat> @test_vloxei16_v_bf16mf2_tu(
+// CHECK-RV64-SAME: <vscale x 2 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x bfloat> @llvm.riscv.vloxei.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[VD]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret <vscale x 2 x bfloat> [[TMP0]]
+//
+vbfloat16mf2_t test_vloxei16_v_bf16mf2_tu(vbfloat16mf2_t vd, const __bf16 *rs1, vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vloxei16_v_bf16mf2_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 4 x bfloat> @test_vloxei16_v_bf16m1_tu(
+// CHECK-RV64-SAME: <vscale x 4 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x bfloat> @llvm.riscv.vloxei.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[VD]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret <vscale x 4 x bfloat> [[TMP0]]
+//
+vbfloat16m1_t test_vloxei16_v_bf16m1_tu(vbfloat16m1_t vd, const __bf16 *rs1, vuint16m1_t rs2, size_t vl) {
+  return __riscv_vloxei16_v_bf16m1_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 8 x bfloat> @test_vloxei16_v_bf16m2_tu(
+// CHECK-RV64-SAME: <vscale x 8 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x bfloat> @llvm.riscv.vloxei.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> [[VD]], ptr [[RS1]], <vscale x 8 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret <vscale x 8 x bfloat> [[TMP0]]
+//
+vbfloat16m2_t test_vloxei16_v_bf16m2_tu(vbfloat16m2_t vd, const __bf16 *rs1, vuint16m2_t rs2, size_t vl) {
+  return __riscv_vloxei16_v_bf16m2_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 16 x bfloat> @test_vloxei16_v_bf16m4_tu(
+// CHECK-RV64-SAME: <vscale x 16 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 16 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x bfloat> @llvm.riscv.vloxei.nxv16bf16.nxv16i16.i64(<vscale x 16 x bfloat> [[VD]], ptr [[RS1]], <vscale x 16 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret <vscale x 16 x bfloat> [[TMP0]]
+//
+vbfloat16m4_t test_vloxei16_v_bf16m4_tu(vbfloat16m4_t vd, const __bf16 *rs1, vuint16m4_t rs2, size_t vl) {
+  return __riscv_vloxei16_v_bf16m4_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 32 x bfloat> @test_vloxei16_v_bf16m8_tu(
+// CHECK-RV64-SAME: <vscale x 32 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 32 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x bfloat> @llvm.riscv.vloxei.nxv32bf16.nxv32i16.i64(<vscale x 32 x bfloat> [[VD]], ptr [[RS1]], <vscale x 32 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret <vscale x 32 x bfloat> [[TMP0]]
+//
+vbfloat16m8_t test_vloxei16_v_bf16m8_tu(vbfloat16m8_t vd, const __bf16 *rs1, vuint16m8_t rs2, size_t vl) {
+  return __riscv_vloxei16_v_bf16m8_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 1 x bfloat> @test_vloxei16_v_bf16mf4_tum(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], <vscale x 1 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x bfloat> @llvm.riscv.vloxei.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[VD]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret <vscale x 1 x bfloat> [[TMP0]]
+//
+vbfloat16mf4_t test_vloxei16_v_bf16mf4_tum(vbool64_t vm, vbfloat16mf4_t vd, const __bf16 *rs1, vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vloxei16_v_bf16mf4_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 2 x bfloat> @test_vloxei16_v_bf16mf2_tum(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], <vscale x 2 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x bfloat> @llvm.riscv.vloxei.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[VD]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret <vscale x 2 x bfloat> [[TMP0]]
+//
+vbfloat16mf2_t test_vloxei16_v_bf16mf2_tum(vbool32_t vm, vbfloat16mf2_t vd, const __bf16 *rs1, vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vloxei16_v_bf16mf2_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 4 x bfloat> @test_vloxei16_v_bf16m1_tum(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], <vscale x 4 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x bfloat> @llvm.riscv.vloxei.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[VD]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret <vscale x 4 x bfloat> [[TMP0]]
+//
+vbfloat16m1_t test_vloxei16_v_bf16m1_tum(vbool16_t vm, vbfloat16m1_t vd, const __bf16 *rs1, vuint16m1_t rs2, size_t vl) {
+  return __riscv_vloxei16_v_bf16m1_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 8 x bfloat> @test_vloxei16_v_bf16m2_tum(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], <vscale x 8 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x bfloat> @llvm.riscv.vloxei.mask.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> [[VD]], ptr [[RS1]], <vscale x 8 x i16> [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret <vscale x 8 x bfloat> [[TMP0]]
+//
+vbfloat16m2_t test_vloxei16_v_bf16m2_tum(vbool8_t vm, vbfloat16m2_t vd, const __bf16 *rs1, vuint16m2_t rs2, size_t vl) {
+  return __riscv_vloxei16_v_bf16m2_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 16 x bfloat> @test_vloxei16_v_bf16m4_tum(
+// CHECK-RV64-SAME: <vscale x 16 x i1> [[VM:%.*]], <vscale x 16 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 16 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x bfloat> @llvm.riscv.vloxei.mask.nxv16bf16.nxv16i16.i64(<vscale x 16 x bfloat> [[VD]], ptr [[RS1]], <vscale x 16 x i16> [[RS2]], <vscale x 16 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret <vscale x 16 x bfloat> [[TMP0]]
+//
+vbfloat16m4_t test_vloxei16_v_bf16m4_tum(vbool4_t vm, vbfloat16m4_t vd, const __bf16 *rs1, vuint16m4_t rs2, size_t vl) {
+  return __riscv_vloxei16_v_bf16m4_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 32 x bfloat> @test_vloxei16_v_bf16m8_tum(
+// CHECK-RV64-SAME: <vscale x 32 x i1> [[VM:%.*]], <vscale x 32 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 32 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x bfloat> @llvm.riscv.vloxei.mask.nxv32bf16.nxv32i16.i64(<vscale x 32 x bfloat> [[VD]], ptr [[RS1]], <vscale x 32 x i16> [[RS2]], <vscale x 32 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret <vscale x 32 x bfloat> [[TMP0]]
+//
+vbfloat16m8_t test_vloxei16_v_bf16m8_tum(vbool2_t vm, vbfloat16m8_t vd, const __bf16 *rs1, vuint16m8_t rs2, size_t vl) {
+  return __riscv_vloxei16_v_bf16m8_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 1 x bfloat> @test_vloxei16_v_bf16mf4_tumu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], <vscale x 1 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x bfloat> @llvm.riscv.vloxei.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[VD]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 1 x bfloat> [[TMP0]]
+//
+vbfloat16mf4_t test_vloxei16_v_bf16mf4_tumu(vbool64_t vm, vbfloat16mf4_t vd, const __bf16 *rs1, vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vloxei16_v_bf16mf4_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 2 x bfloat> @test_vloxei16_v_bf16mf2_tumu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], <vscale x 2 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x bfloat> @llvm.riscv.vloxei.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[VD]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 2 x bfloat> [[TMP0]]
+//
+vbfloat16mf2_t test_vloxei16_v_bf16mf2_tumu(vbool32_t vm, vbfloat16mf2_t vd, const __bf16 *rs1, vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vloxei16_v_bf16mf2_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 4 x bfloat> @test_vloxei16_v_bf16m1_tumu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], <vscale x 4 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x bfloat> @llvm.riscv.vloxei.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[VD]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 4 x bfloat> [[TMP0]]
+//
+vbfloat16m1_t test_vloxei16_v_bf16m1_tumu(vbool16_t vm, vbfloat16m1_t vd, const __bf16 *rs1, vuint16m1_t rs2, size_t vl) {
+  return __riscv_vloxei16_v_bf16m1_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 8 x bfloat> @test_vloxei16_v_bf16m2_tumu(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], <vscale x 8 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x bfloat> @llvm.riscv.vloxei.mask.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> [[VD]], ptr [[RS1]], <vscale x 8 x i16> [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 8 x bfloat> [[TMP0]]
+//
+vbfloat16m2_t test_vloxei16_v_bf16m2_tumu(vbool8_t vm, vbfloat16m2_t vd, const __bf16 *rs1, vuint16m2_t rs2, size_t vl) {
+  return __riscv_vloxei16_v_bf16m2_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 16 x bfloat> @test_vloxei16_v_bf16m4_tumu(
+// CHECK-RV64-SAME: <vscale x 16 x i1> [[VM:%.*]], <vscale x 16 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 16 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x bfloat> @llvm.riscv.vloxei.mask.nxv16bf16.nxv16i16.i64(<vscale x 16 x bfloat> [[VD]], ptr [[RS1]], <vscale x 16 x i16> [[RS2]], <vscale x 16 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 16 x bfloat> [[TMP0]]
+//
+vbfloat16m4_t test_vloxei16_v_bf16m4_tumu(vbool4_t vm, vbfloat16m4_t vd, const __bf16 *rs1, vuint16m4_t rs2, size_t vl) {
+  return __riscv_vloxei16_v_bf16m4_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 32 x bfloat> @test_vloxei16_v_bf16m8_tumu(
+// CHECK-RV64-SAME: <vscale x 32 x i1> [[VM:%.*]], <vscale x 32 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 32 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x bfloat> @llvm.riscv.vloxei.mask.nxv32bf16.nxv32i16.i64(<vscale x 32 x bfloat> [[VD]], ptr [[RS1]], <vscale x 32 x i16> [[RS2]], <vscale x 32 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 32 x bfloat> [[TMP0]]
+//
+vbfloat16m8_t test_vloxei16_v_bf16m8_tumu(vbool2_t vm, vbfloat16m8_t vd, const __bf16 *rs1, vuint16m8_t rs2, size_t vl) {
+  return __riscv_vloxei16_v_bf16m8_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 1 x bfloat> @test_vloxei16_v_bf16mf4_mu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], <vscale x 1 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x bfloat> @llvm.riscv.vloxei.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[VD]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret <vscale x 1 x bfloat> [[TMP0]]
+//
+vbfloat16mf4_t test_vloxei16_v_bf16mf4_mu(vbool64_t vm, vbfloat16mf4_t vd, const __bf16 *rs1, vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vloxei16_v_bf16mf4_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 2 x bfloat> @test_vloxei16_v_bf16mf2_mu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], <vscale x 2 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x bfloat> @llvm.riscv.vloxei.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[VD]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret <vscale x 2 x bfloat> [[TMP0]]
+//
+vbfloat16mf2_t test_vloxei16_v_bf16mf2_mu(vbool32_t vm, vbfloat16mf2_t vd, const __bf16 *rs1, vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vloxei16_v_bf16mf2_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 4 x bfloat> @test_vloxei16_v_bf16m1_mu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], <vscale x 4 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x bfloat> @llvm.riscv.vloxei.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[VD]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret <vscale x 4 x bfloat> [[TMP0]]
+//
+vbfloat16m1_t test_vloxei16_v_bf16m1_mu(vbool16_t vm, vbfloat16m1_t vd, const __bf16 *rs1, vuint16m1_t rs2, size_t vl) {
+  return __riscv_vloxei16_v_bf16m1_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 8 x bfloat> @test_vloxei16_v_bf16m2_mu(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], <vscale x 8 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x bfloat> @llvm.riscv.vloxei.mask.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> [[VD]], ptr [[RS1]], <vscale x 8 x i16> [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret <vscale x 8 x bfloat> [[TMP0]]
+//
+vbfloat16m2_t test_vloxei16_v_bf16m2_mu(vbool8_t vm, vbfloat16m2_t vd, const __bf16 *rs1, vuint16m2_t rs2, size_t vl) {
+  return __riscv_vloxei16_v_bf16m2_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 16 x bfloat> @test_vloxei16_v_bf16m4_mu(
+// CHECK-RV64-SAME: <vscale x 16 x i1> [[VM:%.*]], <vscale x 16 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 16 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x bfloat> @llvm.riscv.vloxei.mask.nxv16bf16.nxv16i16.i64(<vscale x 16 x bfloat> [[VD]], ptr [[RS1]], <vscale x 16 x i16> [[RS2]], <vscale x 16 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret <vscale x 16 x bfloat> [[TMP0]]
+//
+vbfloat16m4_t test_vloxei16_v_bf16m4_mu(vbool4_t vm, vbfloat16m4_t vd, const __bf16 *rs1, vuint16m4_t rs2, size_t vl) {
+  return __riscv_vloxei16_v_bf16m4_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 32 x bfloat> @test_vloxei16_v_bf16m8_mu(
+// CHECK-RV64-SAME: <vscale x 32 x i1> [[VM:%.*]], <vscale x 32 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 32 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x bfloat> @llvm.riscv.vloxei.mask.nxv32bf16.nxv32i16.i64(<vscale x 32 x bfloat> [[VD]], ptr [[RS1]], <vscale x 32 x i16> [[RS2]], <vscale x 32 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret <vscale x 32 x bfloat> [[TMP0]]
+//
+vbfloat16m8_t test_vloxei16_v_bf16m8_mu(vbool2_t vm, vbfloat16m8_t vd, const __bf16 *rs1, vuint16m8_t rs2, size_t vl) {
+  return __riscv_vloxei16_v_bf16m8_mu(vm, vd, rs1, rs2, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vloxseg2ei16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vloxseg2ei16.c
new file mode 100644
index 000000000000..085a3784f87e
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vloxseg2ei16.c
@@ -0,0 +1,249 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vloxseg2ei16_v_bf16mf4x2_tu(
+// CHECK-RV64-SAME: { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vloxseg2.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP2]]
+//
+vbfloat16mf4x2_t test_vloxseg2ei16_v_bf16mf4x2_tu(vbfloat16mf4x2_t vd, const __bf16 *rs1, vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vloxseg2ei16_v_bf16mf4x2_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vloxseg2ei16_v_bf16mf2x2_tu(
+// CHECK-RV64-SAME: { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vloxseg2.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP2]]
+//
+vbfloat16mf2x2_t test_vloxseg2ei16_v_bf16mf2x2_tu(vbfloat16mf2x2_t vd, const __bf16 *rs1, vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vloxseg2ei16_v_bf16mf2x2_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vloxseg2ei16_v_bf16m1x2_tu(
+// CHECK-RV64-SAME: { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vloxseg2.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP2]]
+//
+vbfloat16m1x2_t test_vloxseg2ei16_v_bf16m1x2_tu(vbfloat16m1x2_t vd, const __bf16 *rs1, vuint16m1_t rs2, size_t vl) {
+  return __riscv_vloxseg2ei16_v_bf16m1x2_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vloxseg2ei16_v_bf16m2x2_tu(
+// CHECK-RV64-SAME: { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vloxseg2.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 8 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP2]]
+//
+vbfloat16m2x2_t test_vloxseg2ei16_v_bf16m2x2_tu(vbfloat16m2x2_t vd, const __bf16 *rs1, vuint16m2_t rs2, size_t vl) {
+  return __riscv_vloxseg2ei16_v_bf16m2x2_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @test_vloxseg2ei16_v_bf16m4x2_tu(
+// CHECK-RV64-SAME: { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 16 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @llvm.riscv.vloxseg2.nxv16bf16.nxv16i16.i64(<vscale x 16 x bfloat> [[TMP0]], <vscale x 16 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 16 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[TMP2]]
+//
+vbfloat16m4x2_t test_vloxseg2ei16_v_bf16m4x2_tu(vbfloat16m4x2_t vd, const __bf16 *rs1, vuint16m4_t rs2, size_t vl) {
+  return __riscv_vloxseg2ei16_v_bf16m4x2_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vloxseg2ei16_v_bf16mf4x2_tum(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vloxseg2.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP2]]
+//
+vbfloat16mf4x2_t test_vloxseg2ei16_v_bf16mf4x2_tum(vbool64_t vm, vbfloat16mf4x2_t vd, const __bf16 *rs1, vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vloxseg2ei16_v_bf16mf4x2_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vloxseg2ei16_v_bf16mf2x2_tum(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vloxseg2.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP2]]
+//
+vbfloat16mf2x2_t test_vloxseg2ei16_v_bf16mf2x2_tum(vbool32_t vm, vbfloat16mf2x2_t vd, const __bf16 *rs1, vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vloxseg2ei16_v_bf16mf2x2_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vloxseg2ei16_v_bf16m1x2_tum(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vloxseg2.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP2]]
+//
+vbfloat16m1x2_t test_vloxseg2ei16_v_bf16m1x2_tum(vbool16_t vm, vbfloat16m1x2_t vd, const __bf16 *rs1, vuint16m1_t rs2, size_t vl) {
+  return __riscv_vloxseg2ei16_v_bf16m1x2_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vloxseg2ei16_v_bf16m2x2_tum(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vloxseg2.mask.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 8 x i16> [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP2]]
+//
+vbfloat16m2x2_t test_vloxseg2ei16_v_bf16m2x2_tum(vbool8_t vm, vbfloat16m2x2_t vd, const __bf16 *rs1, vuint16m2_t rs2, size_t vl) {
+  return __riscv_vloxseg2ei16_v_bf16m2x2_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @test_vloxseg2ei16_v_bf16m4x2_tum(
+// CHECK-RV64-SAME: <vscale x 16 x i1> [[VM:%.*]], { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 16 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @llvm.riscv.vloxseg2.mask.nxv16bf16.nxv16i16.i64(<vscale x 16 x bfloat> [[TMP0]], <vscale x 16 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 16 x i16> [[RS2]], <vscale x 16 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[TMP2]]
+//
+vbfloat16m4x2_t test_vloxseg2ei16_v_bf16m4x2_tum(vbool4_t vm, vbfloat16m4x2_t vd, const __bf16 *rs1, vuint16m4_t rs2, size_t vl) {
+  return __riscv_vloxseg2ei16_v_bf16m4x2_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vloxseg2ei16_v_bf16mf4x2_tumu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vloxseg2.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP2]]
+//
+vbfloat16mf4x2_t test_vloxseg2ei16_v_bf16mf4x2_tumu(vbool64_t vm, vbfloat16mf4x2_t vd, const __bf16 *rs1, vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vloxseg2ei16_v_bf16mf4x2_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vloxseg2ei16_v_bf16mf2x2_tumu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vloxseg2.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP2]]
+//
+vbfloat16mf2x2_t test_vloxseg2ei16_v_bf16mf2x2_tumu(vbool32_t vm, vbfloat16mf2x2_t vd, const __bf16 *rs1, vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vloxseg2ei16_v_bf16mf2x2_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vloxseg2ei16_v_bf16m1x2_tumu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vloxseg2.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP2]]
+//
+vbfloat16m1x2_t test_vloxseg2ei16_v_bf16m1x2_tumu(vbool16_t vm, vbfloat16m1x2_t vd, const __bf16 *rs1, vuint16m1_t rs2, size_t vl) {
+  return __riscv_vloxseg2ei16_v_bf16m1x2_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vloxseg2ei16_v_bf16m2x2_tumu(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vloxseg2.mask.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 8 x i16> [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP2]]
+//
+vbfloat16m2x2_t test_vloxseg2ei16_v_bf16m2x2_tumu(vbool8_t vm, vbfloat16m2x2_t vd, const __bf16 *rs1, vuint16m2_t rs2, size_t vl) {
+  return __riscv_vloxseg2ei16_v_bf16m2x2_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @test_vloxseg2ei16_v_bf16m4x2_tumu(
+// CHECK-RV64-SAME: <vscale x 16 x i1> [[VM:%.*]], { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 16 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @llvm.riscv.vloxseg2.mask.nxv16bf16.nxv16i16.i64(<vscale x 16 x bfloat> [[TMP0]], <vscale x 16 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 16 x i16> [[RS2]], <vscale x 16 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[TMP2]]
+//
+vbfloat16m4x2_t test_vloxseg2ei16_v_bf16m4x2_tumu(vbool4_t vm, vbfloat16m4x2_t vd, const __bf16 *rs1, vuint16m4_t rs2, size_t vl) {
+  return __riscv_vloxseg2ei16_v_bf16m4x2_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vloxseg2ei16_v_bf16mf4x2_mu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vloxseg2.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP2]]
+//
+vbfloat16mf4x2_t test_vloxseg2ei16_v_bf16mf4x2_mu(vbool64_t vm, vbfloat16mf4x2_t vd, const __bf16 *rs1, vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vloxseg2ei16_v_bf16mf4x2_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vloxseg2ei16_v_bf16mf2x2_mu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vloxseg2.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP2]]
+//
+vbfloat16mf2x2_t test_vloxseg2ei16_v_bf16mf2x2_mu(vbool32_t vm, vbfloat16mf2x2_t vd, const __bf16 *rs1, vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vloxseg2ei16_v_bf16mf2x2_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vloxseg2ei16_v_bf16m1x2_mu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vloxseg2.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP2]]
+//
+vbfloat16m1x2_t test_vloxseg2ei16_v_bf16m1x2_mu(vbool16_t vm, vbfloat16m1x2_t vd, const __bf16 *rs1, vuint16m1_t rs2, size_t vl) {
+  return __riscv_vloxseg2ei16_v_bf16m1x2_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vloxseg2ei16_v_bf16m2x2_mu(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vloxseg2.mask.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 8 x i16> [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP2]]
+//
+vbfloat16m2x2_t test_vloxseg2ei16_v_bf16m2x2_mu(vbool8_t vm, vbfloat16m2x2_t vd, const __bf16 *rs1, vuint16m2_t rs2, size_t vl) {
+  return __riscv_vloxseg2ei16_v_bf16m2x2_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @test_vloxseg2ei16_v_bf16m4x2_mu(
+// CHECK-RV64-SAME: <vscale x 16 x i1> [[VM:%.*]], { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 16 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @llvm.riscv.vloxseg2.mask.nxv16bf16.nxv16i16.i64(<vscale x 16 x bfloat> [[TMP0]], <vscale x 16 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 16 x i16> [[RS2]], <vscale x 16 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[TMP2]]
+//
+vbfloat16m4x2_t test_vloxseg2ei16_v_bf16m4x2_mu(vbool4_t vm, vbfloat16m4x2_t vd, const __bf16 *rs1, vuint16m4_t rs2, size_t vl) {
+  return __riscv_vloxseg2ei16_v_bf16m4x2_mu(vm, vd, rs1, rs2, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vloxseg3ei16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vloxseg3ei16.c
new file mode 100644
index 000000000000..43d21bd63478
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vloxseg3ei16.c
@@ -0,0 +1,217 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vloxseg3ei16_v_bf16mf4x3_tu(
+// CHECK-RV64-SAME: { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vloxseg3.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP3]]
+//
+vbfloat16mf4x3_t test_vloxseg3ei16_v_bf16mf4x3_tu(vbfloat16mf4x3_t vd, const __bf16 *rs1, vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vloxseg3ei16_v_bf16mf4x3_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vloxseg3ei16_v_bf16mf2x3_tu(
+// CHECK-RV64-SAME: { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vloxseg3.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP3]]
+//
+vbfloat16mf2x3_t test_vloxseg3ei16_v_bf16mf2x3_tu(vbfloat16mf2x3_t vd, const __bf16 *rs1, vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vloxseg3ei16_v_bf16mf2x3_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vloxseg3ei16_v_bf16m1x3_tu(
+// CHECK-RV64-SAME: { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vloxseg3.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP3]]
+//
+vbfloat16m1x3_t test_vloxseg3ei16_v_bf16m1x3_tu(vbfloat16m1x3_t vd, const __bf16 *rs1, vuint16m1_t rs2, size_t vl) {
+  return __riscv_vloxseg3ei16_v_bf16m1x3_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vloxseg3ei16_v_bf16m2x3_tu(
+// CHECK-RV64-SAME: { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vloxseg3.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 8 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP3]]
+//
+vbfloat16m2x3_t test_vloxseg3ei16_v_bf16m2x3_tu(vbfloat16m2x3_t vd, const __bf16 *rs1, vuint16m2_t rs2, size_t vl) {
+  return __riscv_vloxseg3ei16_v_bf16m2x3_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vloxseg3ei16_v_bf16mf4x3_tum(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vloxseg3.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP3]]
+//
+vbfloat16mf4x3_t test_vloxseg3ei16_v_bf16mf4x3_tum(vbool64_t vm, vbfloat16mf4x3_t vd, const __bf16 *rs1, vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vloxseg3ei16_v_bf16mf4x3_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vloxseg3ei16_v_bf16mf2x3_tum(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vloxseg3.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP3]]
+//
+vbfloat16mf2x3_t test_vloxseg3ei16_v_bf16mf2x3_tum(vbool32_t vm, vbfloat16mf2x3_t vd, const __bf16 *rs1, vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vloxseg3ei16_v_bf16mf2x3_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vloxseg3ei16_v_bf16m1x3_tum(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vloxseg3.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP3]]
+//
+vbfloat16m1x3_t test_vloxseg3ei16_v_bf16m1x3_tum(vbool16_t vm, vbfloat16m1x3_t vd, const __bf16 *rs1, vuint16m1_t rs2, size_t vl) {
+  return __riscv_vloxseg3ei16_v_bf16m1x3_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vloxseg3ei16_v_bf16m2x3_tum(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vloxseg3.mask.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 8 x i16> [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP3]]
+//
+vbfloat16m2x3_t test_vloxseg3ei16_v_bf16m2x3_tum(vbool8_t vm, vbfloat16m2x3_t vd, const __bf16 *rs1, vuint16m2_t rs2, size_t vl) {
+  return __riscv_vloxseg3ei16_v_bf16m2x3_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vloxseg3ei16_v_bf16mf4x3_tumu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vloxseg3.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP3]]
+//
+vbfloat16mf4x3_t test_vloxseg3ei16_v_bf16mf4x3_tumu(vbool64_t vm, vbfloat16mf4x3_t vd, const __bf16 *rs1, vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vloxseg3ei16_v_bf16mf4x3_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vloxseg3ei16_v_bf16mf2x3_tumu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vloxseg3.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP3]]
+//
+vbfloat16mf2x3_t test_vloxseg3ei16_v_bf16mf2x3_tumu(vbool32_t vm, vbfloat16mf2x3_t vd, const __bf16 *rs1, vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vloxseg3ei16_v_bf16mf2x3_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vloxseg3ei16_v_bf16m1x3_tumu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vloxseg3.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP3]]
+//
+vbfloat16m1x3_t test_vloxseg3ei16_v_bf16m1x3_tumu(vbool16_t vm, vbfloat16m1x3_t vd, const __bf16 *rs1, vuint16m1_t rs2, size_t vl) {
+  return __riscv_vloxseg3ei16_v_bf16m1x3_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vloxseg3ei16_v_bf16m2x3_tumu(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vloxseg3.mask.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 8 x i16> [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP3]]
+//
+vbfloat16m2x3_t test_vloxseg3ei16_v_bf16m2x3_tumu(vbool8_t vm, vbfloat16m2x3_t vd, const __bf16 *rs1, vuint16m2_t rs2, size_t vl) {
+  return __riscv_vloxseg3ei16_v_bf16m2x3_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vloxseg3ei16_v_bf16mf4x3_mu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vloxseg3.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP3]]
+//
+vbfloat16mf4x3_t test_vloxseg3ei16_v_bf16mf4x3_mu(vbool64_t vm, vbfloat16mf4x3_t vd, const __bf16 *rs1, vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vloxseg3ei16_v_bf16mf4x3_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vloxseg3ei16_v_bf16mf2x3_mu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vloxseg3.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP3]]
+//
+vbfloat16mf2x3_t test_vloxseg3ei16_v_bf16mf2x3_mu(vbool32_t vm, vbfloat16mf2x3_t vd, const __bf16 *rs1, vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vloxseg3ei16_v_bf16mf2x3_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vloxseg3ei16_v_bf16m1x3_mu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vloxseg3.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP3]]
+//
+vbfloat16m1x3_t test_vloxseg3ei16_v_bf16m1x3_mu(vbool16_t vm, vbfloat16m1x3_t vd, const __bf16 *rs1, vuint16m1_t rs2, size_t vl) {
+  return __riscv_vloxseg3ei16_v_bf16m1x3_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vloxseg3ei16_v_bf16m2x3_mu(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vloxseg3.mask.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 8 x i16> [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP3]]
+//
+vbfloat16m2x3_t test_vloxseg3ei16_v_bf16m2x3_mu(vbool8_t vm, vbfloat16m2x3_t vd, const __bf16 *rs1, vuint16m2_t rs2, size_t vl) {
+  return __riscv_vloxseg3ei16_v_bf16m2x3_mu(vm, vd, rs1, rs2, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vloxseg4ei16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vloxseg4ei16.c
new file mode 100644
index 000000000000..43c25cf7d570
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vloxseg4ei16.c
@@ -0,0 +1,233 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vloxseg4ei16_v_bf16mf4x4_tu(
+// CHECK-RV64-SAME: { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vloxseg4.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP4]]
+//
+vbfloat16mf4x4_t test_vloxseg4ei16_v_bf16mf4x4_tu(vbfloat16mf4x4_t vd, const __bf16 *rs1, vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vloxseg4ei16_v_bf16mf4x4_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vloxseg4ei16_v_bf16mf2x4_tu(
+// CHECK-RV64-SAME: { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vloxseg4.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP4]]
+//
+vbfloat16mf2x4_t test_vloxseg4ei16_v_bf16mf2x4_tu(vbfloat16mf2x4_t vd, const __bf16 *rs1, vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vloxseg4ei16_v_bf16mf2x4_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vloxseg4ei16_v_bf16m1x4_tu(
+// CHECK-RV64-SAME: { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vloxseg4.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP4]]
+//
+vbfloat16m1x4_t test_vloxseg4ei16_v_bf16m1x4_tu(vbfloat16m1x4_t vd, const __bf16 *rs1, vuint16m1_t rs2, size_t vl) {
+  return __riscv_vloxseg4ei16_v_bf16m1x4_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vloxseg4ei16_v_bf16m2x4_tu(
+// CHECK-RV64-SAME: { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vloxseg4.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], <vscale x 8 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 8 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP4]]
+//
+vbfloat16m2x4_t test_vloxseg4ei16_v_bf16m2x4_tu(vbfloat16m2x4_t vd, const __bf16 *rs1, vuint16m2_t rs2, size_t vl) {
+  return __riscv_vloxseg4ei16_v_bf16m2x4_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vloxseg4ei16_v_bf16mf4x4_tum(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vloxseg4.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP4]]
+//
+vbfloat16mf4x4_t test_vloxseg4ei16_v_bf16mf4x4_tum(vbool64_t vm, vbfloat16mf4x4_t vd, const __bf16 *rs1, vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vloxseg4ei16_v_bf16mf4x4_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vloxseg4ei16_v_bf16mf2x4_tum(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vloxseg4.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP4]]
+//
+vbfloat16mf2x4_t test_vloxseg4ei16_v_bf16mf2x4_tum(vbool32_t vm, vbfloat16mf2x4_t vd, const __bf16 *rs1, vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vloxseg4ei16_v_bf16mf2x4_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vloxseg4ei16_v_bf16m1x4_tum(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vloxseg4.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP4]]
+//
+vbfloat16m1x4_t test_vloxseg4ei16_v_bf16m1x4_tum(vbool16_t vm, vbfloat16m1x4_t vd, const __bf16 *rs1, vuint16m1_t rs2, size_t vl) {
+  return __riscv_vloxseg4ei16_v_bf16m1x4_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vloxseg4ei16_v_bf16m2x4_tum(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vloxseg4.mask.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], <vscale x 8 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 8 x i16> [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP4]]
+//
+vbfloat16m2x4_t test_vloxseg4ei16_v_bf16m2x4_tum(vbool8_t vm, vbfloat16m2x4_t vd, const __bf16 *rs1, vuint16m2_t rs2, size_t vl) {
+  return __riscv_vloxseg4ei16_v_bf16m2x4_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vloxseg4ei16_v_bf16mf4x4_tumu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vloxseg4.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP4]]
+//
+vbfloat16mf4x4_t test_vloxseg4ei16_v_bf16mf4x4_tumu(vbool64_t vm, vbfloat16mf4x4_t vd, const __bf16 *rs1, vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vloxseg4ei16_v_bf16mf4x4_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vloxseg4ei16_v_bf16mf2x4_tumu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vloxseg4.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP4]]
+//
+vbfloat16mf2x4_t test_vloxseg4ei16_v_bf16mf2x4_tumu(vbool32_t vm, vbfloat16mf2x4_t vd, const __bf16 *rs1, vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vloxseg4ei16_v_bf16mf2x4_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vloxseg4ei16_v_bf16m1x4_tumu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vloxseg4.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP4]]
+//
+vbfloat16m1x4_t test_vloxseg4ei16_v_bf16m1x4_tumu(vbool16_t vm, vbfloat16m1x4_t vd, const __bf16 *rs1, vuint16m1_t rs2, size_t vl) {
+  return __riscv_vloxseg4ei16_v_bf16m1x4_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vloxseg4ei16_v_bf16m2x4_tumu(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vloxseg4.mask.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], <vscale x 8 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 8 x i16> [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP4]]
+//
+vbfloat16m2x4_t test_vloxseg4ei16_v_bf16m2x4_tumu(vbool8_t vm, vbfloat16m2x4_t vd, const __bf16 *rs1, vuint16m2_t rs2, size_t vl) {
+  return __riscv_vloxseg4ei16_v_bf16m2x4_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vloxseg4ei16_v_bf16mf4x4_mu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vloxseg4.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP4]]
+//
+vbfloat16mf4x4_t test_vloxseg4ei16_v_bf16mf4x4_mu(vbool64_t vm, vbfloat16mf4x4_t vd, const __bf16 *rs1, vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vloxseg4ei16_v_bf16mf4x4_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vloxseg4ei16_v_bf16mf2x4_mu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vloxseg4.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP4]]
+//
+vbfloat16mf2x4_t test_vloxseg4ei16_v_bf16mf2x4_mu(vbool32_t vm, vbfloat16mf2x4_t vd, const __bf16 *rs1, vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vloxseg4ei16_v_bf16mf2x4_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vloxseg4ei16_v_bf16m1x4_mu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vloxseg4.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP4]]
+//
+vbfloat16m1x4_t test_vloxseg4ei16_v_bf16m1x4_mu(vbool16_t vm, vbfloat16m1x4_t vd, const __bf16 *rs1, vuint16m1_t rs2, size_t vl) {
+  return __riscv_vloxseg4ei16_v_bf16m1x4_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vloxseg4ei16_v_bf16m2x4_mu(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vloxseg4.mask.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], <vscale x 8 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 8 x i16> [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP4]]
+//
+vbfloat16m2x4_t test_vloxseg4ei16_v_bf16m2x4_mu(vbool8_t vm, vbfloat16m2x4_t vd, const __bf16 *rs1, vuint16m2_t rs2, size_t vl) {
+  return __riscv_vloxseg4ei16_v_bf16m2x4_mu(vm, vd, rs1, rs2, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vloxseg5ei16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vloxseg5ei16.c
new file mode 100644
index 000000000000..9177193a2234
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vloxseg5ei16.c
@@ -0,0 +1,189 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vloxseg5ei16_v_bf16mf4x5_tu(
+// CHECK-RV64-SAME: { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vloxseg5.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP5]]
+//
+vbfloat16mf4x5_t test_vloxseg5ei16_v_bf16mf4x5_tu(vbfloat16mf4x5_t vd, const __bf16 *rs1, vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vloxseg5ei16_v_bf16mf4x5_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vloxseg5ei16_v_bf16mf2x5_tu(
+// CHECK-RV64-SAME: { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vloxseg5.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP5]]
+//
+vbfloat16mf2x5_t test_vloxseg5ei16_v_bf16mf2x5_tu(vbfloat16mf2x5_t vd, const __bf16 *rs1, vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vloxseg5ei16_v_bf16mf2x5_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vloxseg5ei16_v_bf16m1x5_tu(
+// CHECK-RV64-SAME: { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vloxseg5.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP5]]
+//
+vbfloat16m1x5_t test_vloxseg5ei16_v_bf16m1x5_tu(vbfloat16m1x5_t vd, const __bf16 *rs1, vuint16m1_t rs2, size_t vl) {
+  return __riscv_vloxseg5ei16_v_bf16m1x5_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vloxseg5ei16_v_bf16mf4x5_tum(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vloxseg5.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP5]]
+//
+vbfloat16mf4x5_t test_vloxseg5ei16_v_bf16mf4x5_tum(vbool64_t vm, vbfloat16mf4x5_t vd, const __bf16 *rs1, vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vloxseg5ei16_v_bf16mf4x5_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vloxseg5ei16_v_bf16mf2x5_tum(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vloxseg5.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP5]]
+//
+vbfloat16mf2x5_t test_vloxseg5ei16_v_bf16mf2x5_tum(vbool32_t vm, vbfloat16mf2x5_t vd, const __bf16 *rs1, vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vloxseg5ei16_v_bf16mf2x5_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vloxseg5ei16_v_bf16m1x5_tum(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vloxseg5.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP5]]
+//
+vbfloat16m1x5_t test_vloxseg5ei16_v_bf16m1x5_tum(vbool16_t vm, vbfloat16m1x5_t vd, const __bf16 *rs1, vuint16m1_t rs2, size_t vl) {
+  return __riscv_vloxseg5ei16_v_bf16m1x5_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vloxseg5ei16_v_bf16mf4x5_tumu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vloxseg5.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP5]]
+//
+vbfloat16mf4x5_t test_vloxseg5ei16_v_bf16mf4x5_tumu(vbool64_t vm, vbfloat16mf4x5_t vd, const __bf16 *rs1, vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vloxseg5ei16_v_bf16mf4x5_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vloxseg5ei16_v_bf16mf2x5_tumu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vloxseg5.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP5]]
+//
+vbfloat16mf2x5_t test_vloxseg5ei16_v_bf16mf2x5_tumu(vbool32_t vm, vbfloat16mf2x5_t vd, const __bf16 *rs1, vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vloxseg5ei16_v_bf16mf2x5_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vloxseg5ei16_v_bf16m1x5_tumu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vloxseg5.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP5]]
+//
+vbfloat16m1x5_t test_vloxseg5ei16_v_bf16m1x5_tumu(vbool16_t vm, vbfloat16m1x5_t vd, const __bf16 *rs1, vuint16m1_t rs2, size_t vl) {
+  return __riscv_vloxseg5ei16_v_bf16m1x5_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vloxseg5ei16_v_bf16mf4x5_mu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vloxseg5.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP5]]
+//
+vbfloat16mf4x5_t test_vloxseg5ei16_v_bf16mf4x5_mu(vbool64_t vm, vbfloat16mf4x5_t vd, const __bf16 *rs1, vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vloxseg5ei16_v_bf16mf4x5_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vloxseg5ei16_v_bf16mf2x5_mu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vloxseg5.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP5]]
+//
+vbfloat16mf2x5_t test_vloxseg5ei16_v_bf16mf2x5_mu(vbool32_t vm, vbfloat16mf2x5_t vd, const __bf16 *rs1, vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vloxseg5ei16_v_bf16mf2x5_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vloxseg5ei16_v_bf16m1x5_mu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vloxseg5.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP5]]
+//
+vbfloat16m1x5_t test_vloxseg5ei16_v_bf16m1x5_mu(vbool16_t vm, vbfloat16m1x5_t vd, const __bf16 *rs1, vuint16m1_t rs2, size_t vl) {
+  return __riscv_vloxseg5ei16_v_bf16m1x5_mu(vm, vd, rs1, rs2, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vloxseg6ei16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vloxseg6ei16.c
new file mode 100644
index 000000000000..6e6ef3a5a254
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vloxseg6ei16.c
@@ -0,0 +1,201 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vloxseg6ei16_v_bf16mf4x6_tu(
+// CHECK-RV64-SAME: { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vloxseg6.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP6]]
+//
+vbfloat16mf4x6_t test_vloxseg6ei16_v_bf16mf4x6_tu(vbfloat16mf4x6_t vd, const __bf16 *rs1, vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vloxseg6ei16_v_bf16mf4x6_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vloxseg6ei16_v_bf16mf2x6_tu(
+// CHECK-RV64-SAME: { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vloxseg6.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP6]]
+//
+vbfloat16mf2x6_t test_vloxseg6ei16_v_bf16mf2x6_tu(vbfloat16mf2x6_t vd, const __bf16 *rs1, vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vloxseg6ei16_v_bf16mf2x6_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vloxseg6ei16_v_bf16m1x6_tu(
+// CHECK-RV64-SAME: { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vloxseg6.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP6]]
+//
+vbfloat16m1x6_t test_vloxseg6ei16_v_bf16m1x6_tu(vbfloat16m1x6_t vd, const __bf16 *rs1, vuint16m1_t rs2, size_t vl) {
+  return __riscv_vloxseg6ei16_v_bf16m1x6_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vloxseg6ei16_v_bf16mf4x6_tum(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vloxseg6.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP6]]
+//
+vbfloat16mf4x6_t test_vloxseg6ei16_v_bf16mf4x6_tum(vbool64_t vm, vbfloat16mf4x6_t vd, const __bf16 *rs1, vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vloxseg6ei16_v_bf16mf4x6_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vloxseg6ei16_v_bf16mf2x6_tum(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vloxseg6.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP6]]
+//
+vbfloat16mf2x6_t test_vloxseg6ei16_v_bf16mf2x6_tum(vbool32_t vm, vbfloat16mf2x6_t vd, const __bf16 *rs1, vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vloxseg6ei16_v_bf16mf2x6_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vloxseg6ei16_v_bf16m1x6_tum(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vloxseg6.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP6]]
+//
+vbfloat16m1x6_t test_vloxseg6ei16_v_bf16m1x6_tum(vbool16_t vm, vbfloat16m1x6_t vd, const __bf16 *rs1, vuint16m1_t rs2, size_t vl) {
+  return __riscv_vloxseg6ei16_v_bf16m1x6_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vloxseg6ei16_v_bf16mf4x6_tumu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vloxseg6.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP6]]
+//
+vbfloat16mf4x6_t test_vloxseg6ei16_v_bf16mf4x6_tumu(vbool64_t vm, vbfloat16mf4x6_t vd, const __bf16 *rs1, vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vloxseg6ei16_v_bf16mf4x6_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vloxseg6ei16_v_bf16mf2x6_tumu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vloxseg6.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP6]]
+//
+vbfloat16mf2x6_t test_vloxseg6ei16_v_bf16mf2x6_tumu(vbool32_t vm, vbfloat16mf2x6_t vd, const __bf16 *rs1, vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vloxseg6ei16_v_bf16mf2x6_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vloxseg6ei16_v_bf16m1x6_tumu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vloxseg6.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP6]]
+//
+vbfloat16m1x6_t test_vloxseg6ei16_v_bf16m1x6_tumu(vbool16_t vm, vbfloat16m1x6_t vd, const __bf16 *rs1, vuint16m1_t rs2, size_t vl) {
+  return __riscv_vloxseg6ei16_v_bf16m1x6_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vloxseg6ei16_v_bf16mf4x6_mu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vloxseg6.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP6]]
+//
+vbfloat16mf4x6_t test_vloxseg6ei16_v_bf16mf4x6_mu(vbool64_t vm, vbfloat16mf4x6_t vd, const __bf16 *rs1, vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vloxseg6ei16_v_bf16mf4x6_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vloxseg6ei16_v_bf16mf2x6_mu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vloxseg6.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP6]]
+//
+vbfloat16mf2x6_t test_vloxseg6ei16_v_bf16mf2x6_mu(vbool32_t vm, vbfloat16mf2x6_t vd, const __bf16 *rs1, vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vloxseg6ei16_v_bf16mf2x6_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vloxseg6ei16_v_bf16m1x6_mu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vloxseg6.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP6]]
+//
+vbfloat16m1x6_t test_vloxseg6ei16_v_bf16m1x6_mu(vbool16_t vm, vbfloat16m1x6_t vd, const __bf16 *rs1, vuint16m1_t rs2, size_t vl) {
+  return __riscv_vloxseg6ei16_v_bf16m1x6_mu(vm, vd, rs1, rs2, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vloxseg7ei16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vloxseg7ei16.c
new file mode 100644
index 000000000000..08b013399d37
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vloxseg7ei16.c
@@ -0,0 +1,213 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vloxseg7ei16_v_bf16mf4x7_tu(
+// CHECK-RV64-SAME: { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vloxseg7.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP7]]
+//
+vbfloat16mf4x7_t test_vloxseg7ei16_v_bf16mf4x7_tu(vbfloat16mf4x7_t vd, const __bf16 *rs1, vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vloxseg7ei16_v_bf16mf4x7_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vloxseg7ei16_v_bf16mf2x7_tu(
+// CHECK-RV64-SAME: { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vloxseg7.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP7]]
+//
+vbfloat16mf2x7_t test_vloxseg7ei16_v_bf16mf2x7_tu(vbfloat16mf2x7_t vd, const __bf16 *rs1, vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vloxseg7ei16_v_bf16mf2x7_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vloxseg7ei16_v_bf16m1x7_tu(
+// CHECK-RV64-SAME: { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vloxseg7.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP7]]
+//
+vbfloat16m1x7_t test_vloxseg7ei16_v_bf16m1x7_tu(vbfloat16m1x7_t vd, const __bf16 *rs1, vuint16m1_t rs2, size_t vl) {
+  return __riscv_vloxseg7ei16_v_bf16m1x7_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vloxseg7ei16_v_bf16mf4x7_tum(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vloxseg7.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP7]]
+//
+vbfloat16mf4x7_t test_vloxseg7ei16_v_bf16mf4x7_tum(vbool64_t vm, vbfloat16mf4x7_t vd, const __bf16 *rs1, vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vloxseg7ei16_v_bf16mf4x7_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vloxseg7ei16_v_bf16mf2x7_tum(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vloxseg7.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP7]]
+//
+vbfloat16mf2x7_t test_vloxseg7ei16_v_bf16mf2x7_tum(vbool32_t vm, vbfloat16mf2x7_t vd, const __bf16 *rs1, vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vloxseg7ei16_v_bf16mf2x7_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vloxseg7ei16_v_bf16m1x7_tum(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vloxseg7.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP7]]
+//
+vbfloat16m1x7_t test_vloxseg7ei16_v_bf16m1x7_tum(vbool16_t vm, vbfloat16m1x7_t vd, const __bf16 *rs1, vuint16m1_t rs2, size_t vl) {
+  return __riscv_vloxseg7ei16_v_bf16m1x7_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vloxseg7ei16_v_bf16mf4x7_tumu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vloxseg7.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP7]]
+//
+vbfloat16mf4x7_t test_vloxseg7ei16_v_bf16mf4x7_tumu(vbool64_t vm, vbfloat16mf4x7_t vd, const __bf16 *rs1, vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vloxseg7ei16_v_bf16mf4x7_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vloxseg7ei16_v_bf16mf2x7_tumu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vloxseg7.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP7]]
+//
+vbfloat16mf2x7_t test_vloxseg7ei16_v_bf16mf2x7_tumu(vbool32_t vm, vbfloat16mf2x7_t vd, const __bf16 *rs1, vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vloxseg7ei16_v_bf16mf2x7_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vloxseg7ei16_v_bf16m1x7_tumu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vloxseg7.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP7]]
+//
+vbfloat16m1x7_t test_vloxseg7ei16_v_bf16m1x7_tumu(vbool16_t vm, vbfloat16m1x7_t vd, const __bf16 *rs1, vuint16m1_t rs2, size_t vl) {
+  return __riscv_vloxseg7ei16_v_bf16m1x7_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vloxseg7ei16_v_bf16mf4x7_mu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vloxseg7.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP7]]
+//
+vbfloat16mf4x7_t test_vloxseg7ei16_v_bf16mf4x7_mu(vbool64_t vm, vbfloat16mf4x7_t vd, const __bf16 *rs1, vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vloxseg7ei16_v_bf16mf4x7_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vloxseg7ei16_v_bf16mf2x7_mu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vloxseg7.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP7]]
+//
+vbfloat16mf2x7_t test_vloxseg7ei16_v_bf16mf2x7_mu(vbool32_t vm, vbfloat16mf2x7_t vd, const __bf16 *rs1, vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vloxseg7ei16_v_bf16mf2x7_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vloxseg7ei16_v_bf16m1x7_mu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vloxseg7.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP7]]
+//
+vbfloat16m1x7_t test_vloxseg7ei16_v_bf16m1x7_mu(vbool16_t vm, vbfloat16m1x7_t vd, const __bf16 *rs1, vuint16m1_t rs2, size_t vl) {
+  return __riscv_vloxseg7ei16_v_bf16m1x7_mu(vm, vd, rs1, rs2, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vloxseg8ei16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vloxseg8ei16.c
new file mode 100644
index 000000000000..aa738359fec3
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vloxseg8ei16.c
@@ -0,0 +1,225 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vloxseg8ei16_v_bf16mf4x8_tu(
+// CHECK-RV64-SAME: { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vloxseg8.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], <vscale x 1 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP8]]
+//
+vbfloat16mf4x8_t test_vloxseg8ei16_v_bf16mf4x8_tu(vbfloat16mf4x8_t vd, const __bf16 *rs1, vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vloxseg8ei16_v_bf16mf4x8_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vloxseg8ei16_v_bf16mf2x8_tu(
+// CHECK-RV64-SAME: { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vloxseg8.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], <vscale x 2 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP8]]
+//
+vbfloat16mf2x8_t test_vloxseg8ei16_v_bf16mf2x8_tu(vbfloat16mf2x8_t vd, const __bf16 *rs1, vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vloxseg8ei16_v_bf16mf2x8_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vloxseg8ei16_v_bf16m1x8_tu(
+// CHECK-RV64-SAME: { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vloxseg8.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], <vscale x 4 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP8]]
+//
+vbfloat16m1x8_t test_vloxseg8ei16_v_bf16m1x8_tu(vbfloat16m1x8_t vd, const __bf16 *rs1, vuint16m1_t rs2, size_t vl) {
+  return __riscv_vloxseg8ei16_v_bf16m1x8_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vloxseg8ei16_v_bf16mf4x8_tum(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vloxseg8.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], <vscale x 1 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP8]]
+//
+vbfloat16mf4x8_t test_vloxseg8ei16_v_bf16mf4x8_tum(vbool64_t vm, vbfloat16mf4x8_t vd, const __bf16 *rs1, vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vloxseg8ei16_v_bf16mf4x8_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vloxseg8ei16_v_bf16mf2x8_tum(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vloxseg8.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], <vscale x 2 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP8]]
+//
+vbfloat16mf2x8_t test_vloxseg8ei16_v_bf16mf2x8_tum(vbool32_t vm, vbfloat16mf2x8_t vd, const __bf16 *rs1, vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vloxseg8ei16_v_bf16mf2x8_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vloxseg8ei16_v_bf16m1x8_tum(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vloxseg8.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], <vscale x 4 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP8]]
+//
+vbfloat16m1x8_t test_vloxseg8ei16_v_bf16m1x8_tum(vbool16_t vm, vbfloat16m1x8_t vd, const __bf16 *rs1, vuint16m1_t rs2, size_t vl) {
+  return __riscv_vloxseg8ei16_v_bf16m1x8_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vloxseg8ei16_v_bf16mf4x8_tumu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vloxseg8.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], <vscale x 1 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP8]]
+//
+vbfloat16mf4x8_t test_vloxseg8ei16_v_bf16mf4x8_tumu(vbool64_t vm, vbfloat16mf4x8_t vd, const __bf16 *rs1, vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vloxseg8ei16_v_bf16mf4x8_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vloxseg8ei16_v_bf16mf2x8_tumu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vloxseg8.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], <vscale x 2 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP8]]
+//
+vbfloat16mf2x8_t test_vloxseg8ei16_v_bf16mf2x8_tumu(vbool32_t vm, vbfloat16mf2x8_t vd, const __bf16 *rs1, vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vloxseg8ei16_v_bf16mf2x8_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vloxseg8ei16_v_bf16m1x8_tumu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vloxseg8.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], <vscale x 4 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP8]]
+//
+vbfloat16m1x8_t test_vloxseg8ei16_v_bf16m1x8_tumu(vbool16_t vm, vbfloat16m1x8_t vd, const __bf16 *rs1, vuint16m1_t rs2, size_t vl) {
+  return __riscv_vloxseg8ei16_v_bf16m1x8_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vloxseg8ei16_v_bf16mf4x8_mu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vloxseg8.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], <vscale x 1 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP8]]
+//
+vbfloat16mf4x8_t test_vloxseg8ei16_v_bf16mf4x8_mu(vbool64_t vm, vbfloat16mf4x8_t vd, const __bf16 *rs1, vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vloxseg8ei16_v_bf16mf4x8_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vloxseg8ei16_v_bf16mf2x8_mu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vloxseg8.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], <vscale x 2 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP8]]
+//
+vbfloat16mf2x8_t test_vloxseg8ei16_v_bf16mf2x8_mu(vbool32_t vm, vbfloat16mf2x8_t vd, const __bf16 *rs1, vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vloxseg8ei16_v_bf16mf2x8_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vloxseg8ei16_v_bf16m1x8_mu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vloxseg8.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], <vscale x 4 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP8]]
+//
+vbfloat16m1x8_t test_vloxseg8ei16_v_bf16m1x8_mu(vbool16_t vm, vbfloat16m1x8_t vd, const __bf16 *rs1, vuint16m1_t rs2, size_t vl) {
+  return __riscv_vloxseg8ei16_v_bf16m1x8_mu(vm, vd, rs1, rs2, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vlse16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vlse16.c
new file mode 100644
index 000000000000..150d9031b24c
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vlse16.c
@@ -0,0 +1,249 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 1 x bfloat> @test_vlse16_v_bf16mf4_tu(
+// CHECK-RV64-SAME: <vscale x 1 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x bfloat> @llvm.riscv.vlse.nxv1bf16.i64(<vscale x 1 x bfloat> [[VD]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret <vscale x 1 x bfloat> [[TMP0]]
+//
+vbfloat16mf4_t test_vlse16_v_bf16mf4_tu(vbfloat16mf4_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlse16_v_bf16mf4_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 2 x bfloat> @test_vlse16_v_bf16mf2_tu(
+// CHECK-RV64-SAME: <vscale x 2 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x bfloat> @llvm.riscv.vlse.nxv2bf16.i64(<vscale x 2 x bfloat> [[VD]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret <vscale x 2 x bfloat> [[TMP0]]
+//
+vbfloat16mf2_t test_vlse16_v_bf16mf2_tu(vbfloat16mf2_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlse16_v_bf16mf2_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 4 x bfloat> @test_vlse16_v_bf16m1_tu(
+// CHECK-RV64-SAME: <vscale x 4 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x bfloat> @llvm.riscv.vlse.nxv4bf16.i64(<vscale x 4 x bfloat> [[VD]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret <vscale x 4 x bfloat> [[TMP0]]
+//
+vbfloat16m1_t test_vlse16_v_bf16m1_tu(vbfloat16m1_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlse16_v_bf16m1_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 8 x bfloat> @test_vlse16_v_bf16m2_tu(
+// CHECK-RV64-SAME: <vscale x 8 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x bfloat> @llvm.riscv.vlse.nxv8bf16.i64(<vscale x 8 x bfloat> [[VD]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret <vscale x 8 x bfloat> [[TMP0]]
+//
+vbfloat16m2_t test_vlse16_v_bf16m2_tu(vbfloat16m2_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlse16_v_bf16m2_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 16 x bfloat> @test_vlse16_v_bf16m4_tu(
+// CHECK-RV64-SAME: <vscale x 16 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x bfloat> @llvm.riscv.vlse.nxv16bf16.i64(<vscale x 16 x bfloat> [[VD]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret <vscale x 16 x bfloat> [[TMP0]]
+//
+vbfloat16m4_t test_vlse16_v_bf16m4_tu(vbfloat16m4_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlse16_v_bf16m4_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 32 x bfloat> @test_vlse16_v_bf16m8_tu(
+// CHECK-RV64-SAME: <vscale x 32 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x bfloat> @llvm.riscv.vlse.nxv32bf16.i64(<vscale x 32 x bfloat> [[VD]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret <vscale x 32 x bfloat> [[TMP0]]
+//
+vbfloat16m8_t test_vlse16_v_bf16m8_tu(vbfloat16m8_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlse16_v_bf16m8_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 1 x bfloat> @test_vlse16_v_bf16mf4_tum(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], <vscale x 1 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x bfloat> @llvm.riscv.vlse.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[VD]], ptr [[RS1]], i64 [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret <vscale x 1 x bfloat> [[TMP0]]
+//
+vbfloat16mf4_t test_vlse16_v_bf16mf4_tum(vbool64_t vm, vbfloat16mf4_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlse16_v_bf16mf4_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 2 x bfloat> @test_vlse16_v_bf16mf2_tum(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], <vscale x 2 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x bfloat> @llvm.riscv.vlse.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[VD]], ptr [[RS1]], i64 [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret <vscale x 2 x bfloat> [[TMP0]]
+//
+vbfloat16mf2_t test_vlse16_v_bf16mf2_tum(vbool32_t vm, vbfloat16mf2_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlse16_v_bf16mf2_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 4 x bfloat> @test_vlse16_v_bf16m1_tum(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], <vscale x 4 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x bfloat> @llvm.riscv.vlse.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[VD]], ptr [[RS1]], i64 [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret <vscale x 4 x bfloat> [[TMP0]]
+//
+vbfloat16m1_t test_vlse16_v_bf16m1_tum(vbool16_t vm, vbfloat16m1_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlse16_v_bf16m1_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 8 x bfloat> @test_vlse16_v_bf16m2_tum(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], <vscale x 8 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x bfloat> @llvm.riscv.vlse.mask.nxv8bf16.i64(<vscale x 8 x bfloat> [[VD]], ptr [[RS1]], i64 [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret <vscale x 8 x bfloat> [[TMP0]]
+//
+vbfloat16m2_t test_vlse16_v_bf16m2_tum(vbool8_t vm, vbfloat16m2_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlse16_v_bf16m2_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 16 x bfloat> @test_vlse16_v_bf16m4_tum(
+// CHECK-RV64-SAME: <vscale x 16 x i1> [[VM:%.*]], <vscale x 16 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x bfloat> @llvm.riscv.vlse.mask.nxv16bf16.i64(<vscale x 16 x bfloat> [[VD]], ptr [[RS1]], i64 [[RS2]], <vscale x 16 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret <vscale x 16 x bfloat> [[TMP0]]
+//
+vbfloat16m4_t test_vlse16_v_bf16m4_tum(vbool4_t vm, vbfloat16m4_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlse16_v_bf16m4_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 32 x bfloat> @test_vlse16_v_bf16m8_tum(
+// CHECK-RV64-SAME: <vscale x 32 x i1> [[VM:%.*]], <vscale x 32 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x bfloat> @llvm.riscv.vlse.mask.nxv32bf16.i64(<vscale x 32 x bfloat> [[VD]], ptr [[RS1]], i64 [[RS2]], <vscale x 32 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret <vscale x 32 x bfloat> [[TMP0]]
+//
+vbfloat16m8_t test_vlse16_v_bf16m8_tum(vbool2_t vm, vbfloat16m8_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlse16_v_bf16m8_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 1 x bfloat> @test_vlse16_v_bf16mf4_tumu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], <vscale x 1 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x bfloat> @llvm.riscv.vlse.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[VD]], ptr [[RS1]], i64 [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 1 x bfloat> [[TMP0]]
+//
+vbfloat16mf4_t test_vlse16_v_bf16mf4_tumu(vbool64_t vm, vbfloat16mf4_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlse16_v_bf16mf4_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 2 x bfloat> @test_vlse16_v_bf16mf2_tumu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], <vscale x 2 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x bfloat> @llvm.riscv.vlse.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[VD]], ptr [[RS1]], i64 [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 2 x bfloat> [[TMP0]]
+//
+vbfloat16mf2_t test_vlse16_v_bf16mf2_tumu(vbool32_t vm, vbfloat16mf2_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlse16_v_bf16mf2_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 4 x bfloat> @test_vlse16_v_bf16m1_tumu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], <vscale x 4 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x bfloat> @llvm.riscv.vlse.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[VD]], ptr [[RS1]], i64 [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 4 x bfloat> [[TMP0]]
+//
+vbfloat16m1_t test_vlse16_v_bf16m1_tumu(vbool16_t vm, vbfloat16m1_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlse16_v_bf16m1_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 8 x bfloat> @test_vlse16_v_bf16m2_tumu(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], <vscale x 8 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x bfloat> @llvm.riscv.vlse.mask.nxv8bf16.i64(<vscale x 8 x bfloat> [[VD]], ptr [[RS1]], i64 [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 8 x bfloat> [[TMP0]]
+//
+vbfloat16m2_t test_vlse16_v_bf16m2_tumu(vbool8_t vm, vbfloat16m2_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlse16_v_bf16m2_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 16 x bfloat> @test_vlse16_v_bf16m4_tumu(
+// CHECK-RV64-SAME: <vscale x 16 x i1> [[VM:%.*]], <vscale x 16 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x bfloat> @llvm.riscv.vlse.mask.nxv16bf16.i64(<vscale x 16 x bfloat> [[VD]], ptr [[RS1]], i64 [[RS2]], <vscale x 16 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 16 x bfloat> [[TMP0]]
+//
+vbfloat16m4_t test_vlse16_v_bf16m4_tumu(vbool4_t vm, vbfloat16m4_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlse16_v_bf16m4_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 32 x bfloat> @test_vlse16_v_bf16m8_tumu(
+// CHECK-RV64-SAME: <vscale x 32 x i1> [[VM:%.*]], <vscale x 32 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x bfloat> @llvm.riscv.vlse.mask.nxv32bf16.i64(<vscale x 32 x bfloat> [[VD]], ptr [[RS1]], i64 [[RS2]], <vscale x 32 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 32 x bfloat> [[TMP0]]
+//
+vbfloat16m8_t test_vlse16_v_bf16m8_tumu(vbool2_t vm, vbfloat16m8_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlse16_v_bf16m8_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 1 x bfloat> @test_vlse16_v_bf16mf4_mu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], <vscale x 1 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x bfloat> @llvm.riscv.vlse.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[VD]], ptr [[RS1]], i64 [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret <vscale x 1 x bfloat> [[TMP0]]
+//
+vbfloat16mf4_t test_vlse16_v_bf16mf4_mu(vbool64_t vm, vbfloat16mf4_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlse16_v_bf16mf4_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 2 x bfloat> @test_vlse16_v_bf16mf2_mu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], <vscale x 2 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x bfloat> @llvm.riscv.vlse.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[VD]], ptr [[RS1]], i64 [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret <vscale x 2 x bfloat> [[TMP0]]
+//
+vbfloat16mf2_t test_vlse16_v_bf16mf2_mu(vbool32_t vm, vbfloat16mf2_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlse16_v_bf16mf2_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 4 x bfloat> @test_vlse16_v_bf16m1_mu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], <vscale x 4 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x bfloat> @llvm.riscv.vlse.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[VD]], ptr [[RS1]], i64 [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret <vscale x 4 x bfloat> [[TMP0]]
+//
+vbfloat16m1_t test_vlse16_v_bf16m1_mu(vbool16_t vm, vbfloat16m1_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlse16_v_bf16m1_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 8 x bfloat> @test_vlse16_v_bf16m2_mu(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], <vscale x 8 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x bfloat> @llvm.riscv.vlse.mask.nxv8bf16.i64(<vscale x 8 x bfloat> [[VD]], ptr [[RS1]], i64 [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret <vscale x 8 x bfloat> [[TMP0]]
+//
+vbfloat16m2_t test_vlse16_v_bf16m2_mu(vbool8_t vm, vbfloat16m2_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlse16_v_bf16m2_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 16 x bfloat> @test_vlse16_v_bf16m4_mu(
+// CHECK-RV64-SAME: <vscale x 16 x i1> [[VM:%.*]], <vscale x 16 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x bfloat> @llvm.riscv.vlse.mask.nxv16bf16.i64(<vscale x 16 x bfloat> [[VD]], ptr [[RS1]], i64 [[RS2]], <vscale x 16 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret <vscale x 16 x bfloat> [[TMP0]]
+//
+vbfloat16m4_t test_vlse16_v_bf16m4_mu(vbool4_t vm, vbfloat16m4_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlse16_v_bf16m4_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 32 x bfloat> @test_vlse16_v_bf16m8_mu(
+// CHECK-RV64-SAME: <vscale x 32 x i1> [[VM:%.*]], <vscale x 32 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x bfloat> @llvm.riscv.vlse.mask.nxv32bf16.i64(<vscale x 32 x bfloat> [[VD]], ptr [[RS1]], i64 [[RS2]], <vscale x 32 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret <vscale x 32 x bfloat> [[TMP0]]
+//
+vbfloat16m8_t test_vlse16_v_bf16m8_mu(vbool2_t vm, vbfloat16m8_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlse16_v_bf16m8_mu(vm, vd, rs1, rs2, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vlseg2e16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vlseg2e16.c
new file mode 100644
index 000000000000..61328eca7ce0
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vlseg2e16.c
@@ -0,0 +1,249 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg2e16_v_bf16mf4x2_tu(
+// CHECK-RV64-SAME: { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlseg2.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP2]]
+//
+vbfloat16mf4x2_t test_vlseg2e16_v_bf16mf4x2_tu(vbfloat16mf4x2_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg2e16_v_bf16mf4x2_tu(vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg2e16_v_bf16mf2x2_tu(
+// CHECK-RV64-SAME: { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlseg2.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP2]]
+//
+vbfloat16mf2x2_t test_vlseg2e16_v_bf16mf2x2_tu(vbfloat16mf2x2_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg2e16_v_bf16mf2x2_tu(vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg2e16_v_bf16m1x2_tu(
+// CHECK-RV64-SAME: { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlseg2.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP2]]
+//
+vbfloat16m1x2_t test_vlseg2e16_v_bf16m1x2_tu(vbfloat16m1x2_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg2e16_v_bf16m1x2_tu(vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vlseg2e16_v_bf16m2x2_tu(
+// CHECK-RV64-SAME: { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vlseg2.nxv8bf16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP2]]
+//
+vbfloat16m2x2_t test_vlseg2e16_v_bf16m2x2_tu(vbfloat16m2x2_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg2e16_v_bf16m2x2_tu(vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @test_vlseg2e16_v_bf16m4x2_tu(
+// CHECK-RV64-SAME: { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @llvm.riscv.vlseg2.nxv16bf16.i64(<vscale x 16 x bfloat> [[TMP0]], <vscale x 16 x bfloat> [[TMP1]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[TMP2]]
+//
+vbfloat16m4x2_t test_vlseg2e16_v_bf16m4x2_tu(vbfloat16m4x2_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg2e16_v_bf16m4x2_tu(vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg2e16_v_bf16mf4x2_tum(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlseg2.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP2]]
+//
+vbfloat16mf4x2_t test_vlseg2e16_v_bf16mf4x2_tum(vbool64_t vm, vbfloat16mf4x2_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg2e16_v_bf16mf4x2_tum(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg2e16_v_bf16mf2x2_tum(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlseg2.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP2]]
+//
+vbfloat16mf2x2_t test_vlseg2e16_v_bf16mf2x2_tum(vbool32_t vm, vbfloat16mf2x2_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg2e16_v_bf16mf2x2_tum(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg2e16_v_bf16m1x2_tum(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlseg2.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP2]]
+//
+vbfloat16m1x2_t test_vlseg2e16_v_bf16m1x2_tum(vbool16_t vm, vbfloat16m1x2_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg2e16_v_bf16m1x2_tum(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vlseg2e16_v_bf16m2x2_tum(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vlseg2.mask.nxv8bf16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP2]]
+//
+vbfloat16m2x2_t test_vlseg2e16_v_bf16m2x2_tum(vbool8_t vm, vbfloat16m2x2_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg2e16_v_bf16m2x2_tum(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @test_vlseg2e16_v_bf16m4x2_tum(
+// CHECK-RV64-SAME: <vscale x 16 x i1> [[VM:%.*]], { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @llvm.riscv.vlseg2.mask.nxv16bf16.i64(<vscale x 16 x bfloat> [[TMP0]], <vscale x 16 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 16 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[TMP2]]
+//
+vbfloat16m4x2_t test_vlseg2e16_v_bf16m4x2_tum(vbool4_t vm, vbfloat16m4x2_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg2e16_v_bf16m4x2_tum(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg2e16_v_bf16mf4x2_tumu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlseg2.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP2]]
+//
+vbfloat16mf4x2_t test_vlseg2e16_v_bf16mf4x2_tumu(vbool64_t vm, vbfloat16mf4x2_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg2e16_v_bf16mf4x2_tumu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg2e16_v_bf16mf2x2_tumu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlseg2.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP2]]
+//
+vbfloat16mf2x2_t test_vlseg2e16_v_bf16mf2x2_tumu(vbool32_t vm, vbfloat16mf2x2_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg2e16_v_bf16mf2x2_tumu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg2e16_v_bf16m1x2_tumu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlseg2.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP2]]
+//
+vbfloat16m1x2_t test_vlseg2e16_v_bf16m1x2_tumu(vbool16_t vm, vbfloat16m1x2_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg2e16_v_bf16m1x2_tumu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vlseg2e16_v_bf16m2x2_tumu(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vlseg2.mask.nxv8bf16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP2]]
+//
+vbfloat16m2x2_t test_vlseg2e16_v_bf16m2x2_tumu(vbool8_t vm, vbfloat16m2x2_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg2e16_v_bf16m2x2_tumu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @test_vlseg2e16_v_bf16m4x2_tumu(
+// CHECK-RV64-SAME: <vscale x 16 x i1> [[VM:%.*]], { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @llvm.riscv.vlseg2.mask.nxv16bf16.i64(<vscale x 16 x bfloat> [[TMP0]], <vscale x 16 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 16 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[TMP2]]
+//
+vbfloat16m4x2_t test_vlseg2e16_v_bf16m4x2_tumu(vbool4_t vm, vbfloat16m4x2_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg2e16_v_bf16m4x2_tumu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg2e16_v_bf16mf4x2_mu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlseg2.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP2]]
+//
+vbfloat16mf4x2_t test_vlseg2e16_v_bf16mf4x2_mu(vbool64_t vm, vbfloat16mf4x2_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg2e16_v_bf16mf4x2_mu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg2e16_v_bf16mf2x2_mu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlseg2.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP2]]
+//
+vbfloat16mf2x2_t test_vlseg2e16_v_bf16mf2x2_mu(vbool32_t vm, vbfloat16mf2x2_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg2e16_v_bf16mf2x2_mu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg2e16_v_bf16m1x2_mu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlseg2.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP2]]
+//
+vbfloat16m1x2_t test_vlseg2e16_v_bf16m1x2_mu(vbool16_t vm, vbfloat16m1x2_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg2e16_v_bf16m1x2_mu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vlseg2e16_v_bf16m2x2_mu(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vlseg2.mask.nxv8bf16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP2]]
+//
+vbfloat16m2x2_t test_vlseg2e16_v_bf16m2x2_mu(vbool8_t vm, vbfloat16m2x2_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg2e16_v_bf16m2x2_mu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @test_vlseg2e16_v_bf16m4x2_mu(
+// CHECK-RV64-SAME: <vscale x 16 x i1> [[VM:%.*]], { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @llvm.riscv.vlseg2.mask.nxv16bf16.i64(<vscale x 16 x bfloat> [[TMP0]], <vscale x 16 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 16 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[TMP2]]
+//
+vbfloat16m4x2_t test_vlseg2e16_v_bf16m4x2_mu(vbool4_t vm, vbfloat16m4x2_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg2e16_v_bf16m4x2_mu(vm, vd, rs1, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vlseg2e16ff.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vlseg2e16ff.c
new file mode 100644
index 000000000000..8a0bfb50c909
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vlseg2e16ff.c
@@ -0,0 +1,369 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg2e16ff_v_bf16mf4x2_tu(
+// CHECK-RV64-SAME: { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } @llvm.riscv.vlseg2ff.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP2]], 0
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } poison, <vscale x 1 x bfloat> [[TMP3]], 0
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP2]], 1
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], 1
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP2]], 2
+// CHECK-RV64-NEXT:    store i64 [[TMP7]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP6]]
+//
+vbfloat16mf4x2_t test_vlseg2e16ff_v_bf16mf4x2_tu(vbfloat16mf4x2_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vlseg2e16ff_v_bf16mf4x2_tu(vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg2e16ff_v_bf16mf2x2_tu(
+// CHECK-RV64-SAME: { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } @llvm.riscv.vlseg2ff.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP2]], 0
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } poison, <vscale x 2 x bfloat> [[TMP3]], 0
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP2]], 1
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], 1
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP2]], 2
+// CHECK-RV64-NEXT:    store i64 [[TMP7]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP6]]
+//
+vbfloat16mf2x2_t test_vlseg2e16ff_v_bf16mf2x2_tu(vbfloat16mf2x2_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vlseg2e16ff_v_bf16mf2x2_tu(vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg2e16ff_v_bf16m1x2_tu(
+// CHECK-RV64-SAME: { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } @llvm.riscv.vlseg2ff.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP2]], 0
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } poison, <vscale x 4 x bfloat> [[TMP3]], 0
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP2]], 1
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], 1
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP2]], 2
+// CHECK-RV64-NEXT:    store i64 [[TMP7]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP6]]
+//
+vbfloat16m1x2_t test_vlseg2e16ff_v_bf16m1x2_tu(vbfloat16m1x2_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vlseg2e16ff_v_bf16m1x2_tu(vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vlseg2e16ff_v_bf16m2x2_tu(
+// CHECK-RV64-SAME: { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } @llvm.riscv.vlseg2ff.nxv8bf16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP2]], 0
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } poison, <vscale x 8 x bfloat> [[TMP3]], 0
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP2]], 1
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP4]], <vscale x 8 x bfloat> [[TMP5]], 1
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP2]], 2
+// CHECK-RV64-NEXT:    store i64 [[TMP7]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP6]]
+//
+vbfloat16m2x2_t test_vlseg2e16ff_v_bf16m2x2_tu(vbfloat16m2x2_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vlseg2e16ff_v_bf16m2x2_tu(vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @test_vlseg2e16ff_v_bf16m4x2_tu(
+// CHECK-RV64-SAME: { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 16 x bfloat>, <vscale x 16 x bfloat>, i64 } @llvm.riscv.vlseg2ff.nxv16bf16.i64(<vscale x 16 x bfloat> [[TMP0]], <vscale x 16 x bfloat> [[TMP1]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat>, i64 } [[TMP2]], 0
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } poison, <vscale x 16 x bfloat> [[TMP3]], 0
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat>, i64 } [[TMP2]], 1
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[TMP4]], <vscale x 16 x bfloat> [[TMP5]], 1
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat>, i64 } [[TMP2]], 2
+// CHECK-RV64-NEXT:    store i64 [[TMP7]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[TMP6]]
+//
+vbfloat16m4x2_t test_vlseg2e16ff_v_bf16m4x2_tu(vbfloat16m4x2_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vlseg2e16ff_v_bf16m4x2_tu(vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg2e16ff_v_bf16mf4x2_tum(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } @llvm.riscv.vlseg2ff.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP2]], 0
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } poison, <vscale x 1 x bfloat> [[TMP3]], 0
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP2]], 1
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], 1
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP2]], 2
+// CHECK-RV64-NEXT:    store i64 [[TMP7]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP6]]
+//
+vbfloat16mf4x2_t test_vlseg2e16ff_v_bf16mf4x2_tum(vbool64_t vm, vbfloat16mf4x2_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vlseg2e16ff_v_bf16mf4x2_tum(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg2e16ff_v_bf16mf2x2_tum(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } @llvm.riscv.vlseg2ff.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP2]], 0
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } poison, <vscale x 2 x bfloat> [[TMP3]], 0
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP2]], 1
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], 1
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP2]], 2
+// CHECK-RV64-NEXT:    store i64 [[TMP7]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP6]]
+//
+vbfloat16mf2x2_t test_vlseg2e16ff_v_bf16mf2x2_tum(vbool32_t vm, vbfloat16mf2x2_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vlseg2e16ff_v_bf16mf2x2_tum(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg2e16ff_v_bf16m1x2_tum(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } @llvm.riscv.vlseg2ff.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP2]], 0
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } poison, <vscale x 4 x bfloat> [[TMP3]], 0
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP2]], 1
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], 1
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP2]], 2
+// CHECK-RV64-NEXT:    store i64 [[TMP7]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP6]]
+//
+vbfloat16m1x2_t test_vlseg2e16ff_v_bf16m1x2_tum(vbool16_t vm, vbfloat16m1x2_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vlseg2e16ff_v_bf16m1x2_tum(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vlseg2e16ff_v_bf16m2x2_tum(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } @llvm.riscv.vlseg2ff.mask.nxv8bf16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP2]], 0
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } poison, <vscale x 8 x bfloat> [[TMP3]], 0
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP2]], 1
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP4]], <vscale x 8 x bfloat> [[TMP5]], 1
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP2]], 2
+// CHECK-RV64-NEXT:    store i64 [[TMP7]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP6]]
+//
+vbfloat16m2x2_t test_vlseg2e16ff_v_bf16m2x2_tum(vbool8_t vm, vbfloat16m2x2_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vlseg2e16ff_v_bf16m2x2_tum(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @test_vlseg2e16ff_v_bf16m4x2_tum(
+// CHECK-RV64-SAME: <vscale x 16 x i1> [[VM:%.*]], { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 16 x bfloat>, <vscale x 16 x bfloat>, i64 } @llvm.riscv.vlseg2ff.mask.nxv16bf16.i64(<vscale x 16 x bfloat> [[TMP0]], <vscale x 16 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 16 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat>, i64 } [[TMP2]], 0
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } poison, <vscale x 16 x bfloat> [[TMP3]], 0
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat>, i64 } [[TMP2]], 1
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[TMP4]], <vscale x 16 x bfloat> [[TMP5]], 1
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat>, i64 } [[TMP2]], 2
+// CHECK-RV64-NEXT:    store i64 [[TMP7]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[TMP6]]
+//
+vbfloat16m4x2_t test_vlseg2e16ff_v_bf16m4x2_tum(vbool4_t vm, vbfloat16m4x2_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vlseg2e16ff_v_bf16m4x2_tum(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg2e16ff_v_bf16mf4x2_tumu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } @llvm.riscv.vlseg2ff.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP2]], 0
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } poison, <vscale x 1 x bfloat> [[TMP3]], 0
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP2]], 1
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], 1
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP2]], 2
+// CHECK-RV64-NEXT:    store i64 [[TMP7]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP6]]
+//
+vbfloat16mf4x2_t test_vlseg2e16ff_v_bf16mf4x2_tumu(vbool64_t vm, vbfloat16mf4x2_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vlseg2e16ff_v_bf16mf4x2_tumu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg2e16ff_v_bf16mf2x2_tumu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } @llvm.riscv.vlseg2ff.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP2]], 0
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } poison, <vscale x 2 x bfloat> [[TMP3]], 0
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP2]], 1
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], 1
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP2]], 2
+// CHECK-RV64-NEXT:    store i64 [[TMP7]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP6]]
+//
+vbfloat16mf2x2_t test_vlseg2e16ff_v_bf16mf2x2_tumu(vbool32_t vm, vbfloat16mf2x2_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vlseg2e16ff_v_bf16mf2x2_tumu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg2e16ff_v_bf16m1x2_tumu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } @llvm.riscv.vlseg2ff.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP2]], 0
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } poison, <vscale x 4 x bfloat> [[TMP3]], 0
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP2]], 1
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], 1
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP2]], 2
+// CHECK-RV64-NEXT:    store i64 [[TMP7]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP6]]
+//
+vbfloat16m1x2_t test_vlseg2e16ff_v_bf16m1x2_tumu(vbool16_t vm, vbfloat16m1x2_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vlseg2e16ff_v_bf16m1x2_tumu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vlseg2e16ff_v_bf16m2x2_tumu(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } @llvm.riscv.vlseg2ff.mask.nxv8bf16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP2]], 0
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } poison, <vscale x 8 x bfloat> [[TMP3]], 0
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP2]], 1
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP4]], <vscale x 8 x bfloat> [[TMP5]], 1
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP2]], 2
+// CHECK-RV64-NEXT:    store i64 [[TMP7]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP6]]
+//
+vbfloat16m2x2_t test_vlseg2e16ff_v_bf16m2x2_tumu(vbool8_t vm, vbfloat16m2x2_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vlseg2e16ff_v_bf16m2x2_tumu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @test_vlseg2e16ff_v_bf16m4x2_tumu(
+// CHECK-RV64-SAME: <vscale x 16 x i1> [[VM:%.*]], { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 16 x bfloat>, <vscale x 16 x bfloat>, i64 } @llvm.riscv.vlseg2ff.mask.nxv16bf16.i64(<vscale x 16 x bfloat> [[TMP0]], <vscale x 16 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 16 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat>, i64 } [[TMP2]], 0
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } poison, <vscale x 16 x bfloat> [[TMP3]], 0
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat>, i64 } [[TMP2]], 1
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[TMP4]], <vscale x 16 x bfloat> [[TMP5]], 1
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat>, i64 } [[TMP2]], 2
+// CHECK-RV64-NEXT:    store i64 [[TMP7]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[TMP6]]
+//
+vbfloat16m4x2_t test_vlseg2e16ff_v_bf16m4x2_tumu(vbool4_t vm, vbfloat16m4x2_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vlseg2e16ff_v_bf16m4x2_tumu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg2e16ff_v_bf16mf4x2_mu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } @llvm.riscv.vlseg2ff.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP2]], 0
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } poison, <vscale x 1 x bfloat> [[TMP3]], 0
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP2]], 1
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], 1
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP2]], 2
+// CHECK-RV64-NEXT:    store i64 [[TMP7]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP6]]
+//
+vbfloat16mf4x2_t test_vlseg2e16ff_v_bf16mf4x2_mu(vbool64_t vm, vbfloat16mf4x2_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vlseg2e16ff_v_bf16mf4x2_mu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg2e16ff_v_bf16mf2x2_mu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } @llvm.riscv.vlseg2ff.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP2]], 0
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } poison, <vscale x 2 x bfloat> [[TMP3]], 0
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP2]], 1
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], 1
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP2]], 2
+// CHECK-RV64-NEXT:    store i64 [[TMP7]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP6]]
+//
+vbfloat16mf2x2_t test_vlseg2e16ff_v_bf16mf2x2_mu(vbool32_t vm, vbfloat16mf2x2_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vlseg2e16ff_v_bf16mf2x2_mu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg2e16ff_v_bf16m1x2_mu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } @llvm.riscv.vlseg2ff.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP2]], 0
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } poison, <vscale x 4 x bfloat> [[TMP3]], 0
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP2]], 1
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], 1
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP2]], 2
+// CHECK-RV64-NEXT:    store i64 [[TMP7]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP6]]
+//
+vbfloat16m1x2_t test_vlseg2e16ff_v_bf16m1x2_mu(vbool16_t vm, vbfloat16m1x2_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vlseg2e16ff_v_bf16m1x2_mu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vlseg2e16ff_v_bf16m2x2_mu(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } @llvm.riscv.vlseg2ff.mask.nxv8bf16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP2]], 0
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } poison, <vscale x 8 x bfloat> [[TMP3]], 0
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP2]], 1
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP4]], <vscale x 8 x bfloat> [[TMP5]], 1
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP2]], 2
+// CHECK-RV64-NEXT:    store i64 [[TMP7]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP6]]
+//
+vbfloat16m2x2_t test_vlseg2e16ff_v_bf16m2x2_mu(vbool8_t vm, vbfloat16m2x2_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vlseg2e16ff_v_bf16m2x2_mu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @test_vlseg2e16ff_v_bf16m4x2_mu(
+// CHECK-RV64-SAME: <vscale x 16 x i1> [[VM:%.*]], { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 16 x bfloat>, <vscale x 16 x bfloat>, i64 } @llvm.riscv.vlseg2ff.mask.nxv16bf16.i64(<vscale x 16 x bfloat> [[TMP0]], <vscale x 16 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 16 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat>, i64 } [[TMP2]], 0
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } poison, <vscale x 16 x bfloat> [[TMP3]], 0
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat>, i64 } [[TMP2]], 1
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[TMP4]], <vscale x 16 x bfloat> [[TMP5]], 1
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat>, i64 } [[TMP2]], 2
+// CHECK-RV64-NEXT:    store i64 [[TMP7]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[TMP6]]
+//
+vbfloat16m4x2_t test_vlseg2e16ff_v_bf16m4x2_mu(vbool4_t vm, vbfloat16m4x2_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vlseg2e16ff_v_bf16m4x2_mu(vm, vd, rs1, new_vl, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vlseg3e16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vlseg3e16.c
new file mode 100644
index 000000000000..10ec33ec6fec
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vlseg3e16.c
@@ -0,0 +1,217 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg3e16_v_bf16mf4x3_tu(
+// CHECK-RV64-SAME: { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlseg3.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP3]]
+//
+vbfloat16mf4x3_t test_vlseg3e16_v_bf16mf4x3_tu(vbfloat16mf4x3_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg3e16_v_bf16mf4x3_tu(vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg3e16_v_bf16mf2x3_tu(
+// CHECK-RV64-SAME: { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlseg3.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP3]]
+//
+vbfloat16mf2x3_t test_vlseg3e16_v_bf16mf2x3_tu(vbfloat16mf2x3_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg3e16_v_bf16mf2x3_tu(vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg3e16_v_bf16m1x3_tu(
+// CHECK-RV64-SAME: { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlseg3.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP3]]
+//
+vbfloat16m1x3_t test_vlseg3e16_v_bf16m1x3_tu(vbfloat16m1x3_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg3e16_v_bf16m1x3_tu(vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vlseg3e16_v_bf16m2x3_tu(
+// CHECK-RV64-SAME: { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vlseg3.nxv8bf16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP3]]
+//
+vbfloat16m2x3_t test_vlseg3e16_v_bf16m2x3_tu(vbfloat16m2x3_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg3e16_v_bf16m2x3_tu(vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg3e16_v_bf16mf4x3_tum(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlseg3.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP3]]
+//
+vbfloat16mf4x3_t test_vlseg3e16_v_bf16mf4x3_tum(vbool64_t vm, vbfloat16mf4x3_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg3e16_v_bf16mf4x3_tum(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg3e16_v_bf16mf2x3_tum(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlseg3.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP3]]
+//
+vbfloat16mf2x3_t test_vlseg3e16_v_bf16mf2x3_tum(vbool32_t vm, vbfloat16mf2x3_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg3e16_v_bf16mf2x3_tum(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg3e16_v_bf16m1x3_tum(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlseg3.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP3]]
+//
+vbfloat16m1x3_t test_vlseg3e16_v_bf16m1x3_tum(vbool16_t vm, vbfloat16m1x3_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg3e16_v_bf16m1x3_tum(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vlseg3e16_v_bf16m2x3_tum(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vlseg3.mask.nxv8bf16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP3]]
+//
+vbfloat16m2x3_t test_vlseg3e16_v_bf16m2x3_tum(vbool8_t vm, vbfloat16m2x3_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg3e16_v_bf16m2x3_tum(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg3e16_v_bf16mf4x3_tumu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlseg3.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP3]]
+//
+vbfloat16mf4x3_t test_vlseg3e16_v_bf16mf4x3_tumu(vbool64_t vm, vbfloat16mf4x3_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg3e16_v_bf16mf4x3_tumu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg3e16_v_bf16mf2x3_tumu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlseg3.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP3]]
+//
+vbfloat16mf2x3_t test_vlseg3e16_v_bf16mf2x3_tumu(vbool32_t vm, vbfloat16mf2x3_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg3e16_v_bf16mf2x3_tumu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg3e16_v_bf16m1x3_tumu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlseg3.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP3]]
+//
+vbfloat16m1x3_t test_vlseg3e16_v_bf16m1x3_tumu(vbool16_t vm, vbfloat16m1x3_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg3e16_v_bf16m1x3_tumu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vlseg3e16_v_bf16m2x3_tumu(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vlseg3.mask.nxv8bf16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP3]]
+//
+vbfloat16m2x3_t test_vlseg3e16_v_bf16m2x3_tumu(vbool8_t vm, vbfloat16m2x3_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg3e16_v_bf16m2x3_tumu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg3e16_v_bf16mf4x3_mu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlseg3.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP3]]
+//
+vbfloat16mf4x3_t test_vlseg3e16_v_bf16mf4x3_mu(vbool64_t vm, vbfloat16mf4x3_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg3e16_v_bf16mf4x3_mu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg3e16_v_bf16mf2x3_mu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlseg3.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP3]]
+//
+vbfloat16mf2x3_t test_vlseg3e16_v_bf16mf2x3_mu(vbool32_t vm, vbfloat16mf2x3_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg3e16_v_bf16mf2x3_mu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg3e16_v_bf16m1x3_mu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlseg3.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP3]]
+//
+vbfloat16m1x3_t test_vlseg3e16_v_bf16m1x3_mu(vbool16_t vm, vbfloat16m1x3_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg3e16_v_bf16m1x3_mu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vlseg3e16_v_bf16m2x3_mu(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vlseg3.mask.nxv8bf16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP3]]
+//
+vbfloat16m2x3_t test_vlseg3e16_v_bf16m2x3_mu(vbool8_t vm, vbfloat16m2x3_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg3e16_v_bf16m2x3_mu(vm, vd, rs1, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vlseg3e16ff.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vlseg3e16ff.c
new file mode 100644
index 000000000000..03cef3d745a7
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vlseg3e16ff.c
@@ -0,0 +1,345 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg3e16ff_v_bf16mf4x3_tu(
+// CHECK-RV64-SAME: { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } @llvm.riscv.vlseg3ff.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP3]], 0
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } poison, <vscale x 1 x bfloat> [[TMP4]], 0
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], 1
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP3]], 2
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP7]], <vscale x 1 x bfloat> [[TMP8]], 2
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP3]], 3
+// CHECK-RV64-NEXT:    store i64 [[TMP10]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP9]]
+//
+vbfloat16mf4x3_t test_vlseg3e16ff_v_bf16mf4x3_tu(vbfloat16mf4x3_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vlseg3e16ff_v_bf16mf4x3_tu(vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg3e16ff_v_bf16mf2x3_tu(
+// CHECK-RV64-SAME: { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } @llvm.riscv.vlseg3ff.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP3]], 0
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } poison, <vscale x 2 x bfloat> [[TMP4]], 0
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], 1
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP3]], 2
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP7]], <vscale x 2 x bfloat> [[TMP8]], 2
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP3]], 3
+// CHECK-RV64-NEXT:    store i64 [[TMP10]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP9]]
+//
+vbfloat16mf2x3_t test_vlseg3e16ff_v_bf16mf2x3_tu(vbfloat16mf2x3_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vlseg3e16ff_v_bf16mf2x3_tu(vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg3e16ff_v_bf16m1x3_tu(
+// CHECK-RV64-SAME: { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } @llvm.riscv.vlseg3ff.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP3]], 0
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } poison, <vscale x 4 x bfloat> [[TMP4]], 0
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], 1
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP3]], 2
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP7]], <vscale x 4 x bfloat> [[TMP8]], 2
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP3]], 3
+// CHECK-RV64-NEXT:    store i64 [[TMP10]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP9]]
+//
+vbfloat16m1x3_t test_vlseg3e16ff_v_bf16m1x3_tu(vbfloat16m1x3_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vlseg3e16ff_v_bf16m1x3_tu(vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vlseg3e16ff_v_bf16m2x3_tu(
+// CHECK-RV64-SAME: { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } @llvm.riscv.vlseg3ff.nxv8bf16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP3]], 0
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } poison, <vscale x 8 x bfloat> [[TMP4]], 0
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP5]], <vscale x 8 x bfloat> [[TMP6]], 1
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP3]], 2
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP7]], <vscale x 8 x bfloat> [[TMP8]], 2
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP3]], 3
+// CHECK-RV64-NEXT:    store i64 [[TMP10]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP9]]
+//
+vbfloat16m2x3_t test_vlseg3e16ff_v_bf16m2x3_tu(vbfloat16m2x3_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vlseg3e16ff_v_bf16m2x3_tu(vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg3e16ff_v_bf16mf4x3_tum(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } @llvm.riscv.vlseg3ff.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP3]], 0
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } poison, <vscale x 1 x bfloat> [[TMP4]], 0
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], 1
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP3]], 2
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP7]], <vscale x 1 x bfloat> [[TMP8]], 2
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP3]], 3
+// CHECK-RV64-NEXT:    store i64 [[TMP10]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP9]]
+//
+vbfloat16mf4x3_t test_vlseg3e16ff_v_bf16mf4x3_tum(vbool64_t vm, vbfloat16mf4x3_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vlseg3e16ff_v_bf16mf4x3_tum(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg3e16ff_v_bf16mf2x3_tum(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } @llvm.riscv.vlseg3ff.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP3]], 0
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } poison, <vscale x 2 x bfloat> [[TMP4]], 0
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], 1
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP3]], 2
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP7]], <vscale x 2 x bfloat> [[TMP8]], 2
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP3]], 3
+// CHECK-RV64-NEXT:    store i64 [[TMP10]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP9]]
+//
+vbfloat16mf2x3_t test_vlseg3e16ff_v_bf16mf2x3_tum(vbool32_t vm, vbfloat16mf2x3_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vlseg3e16ff_v_bf16mf2x3_tum(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg3e16ff_v_bf16m1x3_tum(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } @llvm.riscv.vlseg3ff.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP3]], 0
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } poison, <vscale x 4 x bfloat> [[TMP4]], 0
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], 1
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP3]], 2
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP7]], <vscale x 4 x bfloat> [[TMP8]], 2
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP3]], 3
+// CHECK-RV64-NEXT:    store i64 [[TMP10]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP9]]
+//
+vbfloat16m1x3_t test_vlseg3e16ff_v_bf16m1x3_tum(vbool16_t vm, vbfloat16m1x3_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vlseg3e16ff_v_bf16m1x3_tum(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vlseg3e16ff_v_bf16m2x3_tum(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } @llvm.riscv.vlseg3ff.mask.nxv8bf16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP3]], 0
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } poison, <vscale x 8 x bfloat> [[TMP4]], 0
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP5]], <vscale x 8 x bfloat> [[TMP6]], 1
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP3]], 2
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP7]], <vscale x 8 x bfloat> [[TMP8]], 2
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP3]], 3
+// CHECK-RV64-NEXT:    store i64 [[TMP10]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP9]]
+//
+vbfloat16m2x3_t test_vlseg3e16ff_v_bf16m2x3_tum(vbool8_t vm, vbfloat16m2x3_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vlseg3e16ff_v_bf16m2x3_tum(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg3e16ff_v_bf16mf4x3_tumu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } @llvm.riscv.vlseg3ff.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP3]], 0
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } poison, <vscale x 1 x bfloat> [[TMP4]], 0
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], 1
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP3]], 2
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP7]], <vscale x 1 x bfloat> [[TMP8]], 2
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP3]], 3
+// CHECK-RV64-NEXT:    store i64 [[TMP10]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP9]]
+//
+vbfloat16mf4x3_t test_vlseg3e16ff_v_bf16mf4x3_tumu(vbool64_t vm, vbfloat16mf4x3_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vlseg3e16ff_v_bf16mf4x3_tumu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg3e16ff_v_bf16mf2x3_tumu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } @llvm.riscv.vlseg3ff.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP3]], 0
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } poison, <vscale x 2 x bfloat> [[TMP4]], 0
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], 1
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP3]], 2
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP7]], <vscale x 2 x bfloat> [[TMP8]], 2
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP3]], 3
+// CHECK-RV64-NEXT:    store i64 [[TMP10]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP9]]
+//
+vbfloat16mf2x3_t test_vlseg3e16ff_v_bf16mf2x3_tumu(vbool32_t vm, vbfloat16mf2x3_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vlseg3e16ff_v_bf16mf2x3_tumu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg3e16ff_v_bf16m1x3_tumu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } @llvm.riscv.vlseg3ff.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP3]], 0
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } poison, <vscale x 4 x bfloat> [[TMP4]], 0
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], 1
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP3]], 2
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP7]], <vscale x 4 x bfloat> [[TMP8]], 2
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP3]], 3
+// CHECK-RV64-NEXT:    store i64 [[TMP10]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP9]]
+//
+vbfloat16m1x3_t test_vlseg3e16ff_v_bf16m1x3_tumu(vbool16_t vm, vbfloat16m1x3_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vlseg3e16ff_v_bf16m1x3_tumu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vlseg3e16ff_v_bf16m2x3_tumu(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } @llvm.riscv.vlseg3ff.mask.nxv8bf16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP3]], 0
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } poison, <vscale x 8 x bfloat> [[TMP4]], 0
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP5]], <vscale x 8 x bfloat> [[TMP6]], 1
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP3]], 2
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP7]], <vscale x 8 x bfloat> [[TMP8]], 2
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP3]], 3
+// CHECK-RV64-NEXT:    store i64 [[TMP10]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP9]]
+//
+vbfloat16m2x3_t test_vlseg3e16ff_v_bf16m2x3_tumu(vbool8_t vm, vbfloat16m2x3_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vlseg3e16ff_v_bf16m2x3_tumu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg3e16ff_v_bf16mf4x3_mu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } @llvm.riscv.vlseg3ff.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP3]], 0
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } poison, <vscale x 1 x bfloat> [[TMP4]], 0
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], 1
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP3]], 2
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP7]], <vscale x 1 x bfloat> [[TMP8]], 2
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP3]], 3
+// CHECK-RV64-NEXT:    store i64 [[TMP10]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP9]]
+//
+vbfloat16mf4x3_t test_vlseg3e16ff_v_bf16mf4x3_mu(vbool64_t vm, vbfloat16mf4x3_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vlseg3e16ff_v_bf16mf4x3_mu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg3e16ff_v_bf16mf2x3_mu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } @llvm.riscv.vlseg3ff.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP3]], 0
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } poison, <vscale x 2 x bfloat> [[TMP4]], 0
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], 1
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP3]], 2
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP7]], <vscale x 2 x bfloat> [[TMP8]], 2
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP3]], 3
+// CHECK-RV64-NEXT:    store i64 [[TMP10]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP9]]
+//
+vbfloat16mf2x3_t test_vlseg3e16ff_v_bf16mf2x3_mu(vbool32_t vm, vbfloat16mf2x3_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vlseg3e16ff_v_bf16mf2x3_mu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg3e16ff_v_bf16m1x3_mu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } @llvm.riscv.vlseg3ff.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP3]], 0
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } poison, <vscale x 4 x bfloat> [[TMP4]], 0
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], 1
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP3]], 2
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP7]], <vscale x 4 x bfloat> [[TMP8]], 2
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP3]], 3
+// CHECK-RV64-NEXT:    store i64 [[TMP10]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP9]]
+//
+vbfloat16m1x3_t test_vlseg3e16ff_v_bf16m1x3_mu(vbool16_t vm, vbfloat16m1x3_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vlseg3e16ff_v_bf16m1x3_mu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vlseg3e16ff_v_bf16m2x3_mu(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } @llvm.riscv.vlseg3ff.mask.nxv8bf16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP3]], 0
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } poison, <vscale x 8 x bfloat> [[TMP4]], 0
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP5]], <vscale x 8 x bfloat> [[TMP6]], 1
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP3]], 2
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP7]], <vscale x 8 x bfloat> [[TMP8]], 2
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP3]], 3
+// CHECK-RV64-NEXT:    store i64 [[TMP10]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP9]]
+//
+vbfloat16m2x3_t test_vlseg3e16ff_v_bf16m2x3_mu(vbool8_t vm, vbfloat16m2x3_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vlseg3e16ff_v_bf16m2x3_mu(vm, vd, rs1, new_vl, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vlseg4e16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vlseg4e16.c
new file mode 100644
index 000000000000..770458e69ecf
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vlseg4e16.c
@@ -0,0 +1,233 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg4e16_v_bf16mf4x4_tu(
+// CHECK-RV64-SAME: { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlseg4.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP4]]
+//
+vbfloat16mf4x4_t test_vlseg4e16_v_bf16mf4x4_tu(vbfloat16mf4x4_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg4e16_v_bf16mf4x4_tu(vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg4e16_v_bf16mf2x4_tu(
+// CHECK-RV64-SAME: { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlseg4.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP4]]
+//
+vbfloat16mf2x4_t test_vlseg4e16_v_bf16mf2x4_tu(vbfloat16mf2x4_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg4e16_v_bf16mf2x4_tu(vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg4e16_v_bf16m1x4_tu(
+// CHECK-RV64-SAME: { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlseg4.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP4]]
+//
+vbfloat16m1x4_t test_vlseg4e16_v_bf16m1x4_tu(vbfloat16m1x4_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg4e16_v_bf16m1x4_tu(vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vlseg4e16_v_bf16m2x4_tu(
+// CHECK-RV64-SAME: { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vlseg4.nxv8bf16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], <vscale x 8 x bfloat> [[TMP3]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP4]]
+//
+vbfloat16m2x4_t test_vlseg4e16_v_bf16m2x4_tu(vbfloat16m2x4_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg4e16_v_bf16m2x4_tu(vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg4e16_v_bf16mf4x4_tum(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlseg4.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP4]]
+//
+vbfloat16mf4x4_t test_vlseg4e16_v_bf16mf4x4_tum(vbool64_t vm, vbfloat16mf4x4_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg4e16_v_bf16mf4x4_tum(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg4e16_v_bf16mf2x4_tum(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlseg4.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP4]]
+//
+vbfloat16mf2x4_t test_vlseg4e16_v_bf16mf2x4_tum(vbool32_t vm, vbfloat16mf2x4_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg4e16_v_bf16mf2x4_tum(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg4e16_v_bf16m1x4_tum(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlseg4.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP4]]
+//
+vbfloat16m1x4_t test_vlseg4e16_v_bf16m1x4_tum(vbool16_t vm, vbfloat16m1x4_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg4e16_v_bf16m1x4_tum(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vlseg4e16_v_bf16m2x4_tum(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vlseg4.mask.nxv8bf16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], <vscale x 8 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP4]]
+//
+vbfloat16m2x4_t test_vlseg4e16_v_bf16m2x4_tum(vbool8_t vm, vbfloat16m2x4_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg4e16_v_bf16m2x4_tum(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg4e16_v_bf16mf4x4_tumu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlseg4.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP4]]
+//
+vbfloat16mf4x4_t test_vlseg4e16_v_bf16mf4x4_tumu(vbool64_t vm, vbfloat16mf4x4_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg4e16_v_bf16mf4x4_tumu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg4e16_v_bf16mf2x4_tumu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlseg4.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP4]]
+//
+vbfloat16mf2x4_t test_vlseg4e16_v_bf16mf2x4_tumu(vbool32_t vm, vbfloat16mf2x4_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg4e16_v_bf16mf2x4_tumu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg4e16_v_bf16m1x4_tumu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlseg4.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP4]]
+//
+vbfloat16m1x4_t test_vlseg4e16_v_bf16m1x4_tumu(vbool16_t vm, vbfloat16m1x4_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg4e16_v_bf16m1x4_tumu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vlseg4e16_v_bf16m2x4_tumu(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vlseg4.mask.nxv8bf16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], <vscale x 8 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP4]]
+//
+vbfloat16m2x4_t test_vlseg4e16_v_bf16m2x4_tumu(vbool8_t vm, vbfloat16m2x4_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg4e16_v_bf16m2x4_tumu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg4e16_v_bf16mf4x4_mu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlseg4.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP4]]
+//
+vbfloat16mf4x4_t test_vlseg4e16_v_bf16mf4x4_mu(vbool64_t vm, vbfloat16mf4x4_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg4e16_v_bf16mf4x4_mu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg4e16_v_bf16mf2x4_mu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlseg4.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP4]]
+//
+vbfloat16mf2x4_t test_vlseg4e16_v_bf16mf2x4_mu(vbool32_t vm, vbfloat16mf2x4_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg4e16_v_bf16mf2x4_mu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg4e16_v_bf16m1x4_mu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlseg4.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP4]]
+//
+vbfloat16m1x4_t test_vlseg4e16_v_bf16m1x4_mu(vbool16_t vm, vbfloat16m1x4_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg4e16_v_bf16m1x4_mu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vlseg4e16_v_bf16m2x4_mu(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vlseg4.mask.nxv8bf16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], <vscale x 8 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP4]]
+//
+vbfloat16m2x4_t test_vlseg4e16_v_bf16m2x4_mu(vbool8_t vm, vbfloat16m2x4_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg4e16_v_bf16m2x4_mu(vm, vd, rs1, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vlseg4e16ff.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vlseg4e16ff.c
new file mode 100644
index 000000000000..6be407d66567
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vlseg4e16ff.c
@@ -0,0 +1,393 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg4e16ff_v_bf16mf4x4_tu(
+// CHECK-RV64-SAME: { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } @llvm.riscv.vlseg4ff.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP4]], 0
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } poison, <vscale x 1 x bfloat> [[TMP5]], 0
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP4]], 1
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP6]], <vscale x 1 x bfloat> [[TMP7]], 1
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP4]], 2
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP8]], <vscale x 1 x bfloat> [[TMP9]], 2
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP4]], 3
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP10]], <vscale x 1 x bfloat> [[TMP11]], 3
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP4]], 4
+// CHECK-RV64-NEXT:    store i64 [[TMP13]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP12]]
+//
+vbfloat16mf4x4_t test_vlseg4e16ff_v_bf16mf4x4_tu(vbfloat16mf4x4_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vlseg4e16ff_v_bf16mf4x4_tu(vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg4e16ff_v_bf16mf2x4_tu(
+// CHECK-RV64-SAME: { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } @llvm.riscv.vlseg4ff.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP4]], 0
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } poison, <vscale x 2 x bfloat> [[TMP5]], 0
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP4]], 1
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP6]], <vscale x 2 x bfloat> [[TMP7]], 1
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP4]], 2
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP8]], <vscale x 2 x bfloat> [[TMP9]], 2
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP4]], 3
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP10]], <vscale x 2 x bfloat> [[TMP11]], 3
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP4]], 4
+// CHECK-RV64-NEXT:    store i64 [[TMP13]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP12]]
+//
+vbfloat16mf2x4_t test_vlseg4e16ff_v_bf16mf2x4_tu(vbfloat16mf2x4_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vlseg4e16ff_v_bf16mf2x4_tu(vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg4e16ff_v_bf16m1x4_tu(
+// CHECK-RV64-SAME: { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } @llvm.riscv.vlseg4ff.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP4]], 0
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } poison, <vscale x 4 x bfloat> [[TMP5]], 0
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP4]], 1
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP6]], <vscale x 4 x bfloat> [[TMP7]], 1
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP4]], 2
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP8]], <vscale x 4 x bfloat> [[TMP9]], 2
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP4]], 3
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP10]], <vscale x 4 x bfloat> [[TMP11]], 3
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP4]], 4
+// CHECK-RV64-NEXT:    store i64 [[TMP13]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP12]]
+//
+vbfloat16m1x4_t test_vlseg4e16ff_v_bf16m1x4_tu(vbfloat16m1x4_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vlseg4e16ff_v_bf16m1x4_tu(vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vlseg4e16ff_v_bf16m2x4_tu(
+// CHECK-RV64-SAME: { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } @llvm.riscv.vlseg4ff.nxv8bf16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], <vscale x 8 x bfloat> [[TMP3]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP4]], 0
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } poison, <vscale x 8 x bfloat> [[TMP5]], 0
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP4]], 1
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP6]], <vscale x 8 x bfloat> [[TMP7]], 1
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP4]], 2
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP8]], <vscale x 8 x bfloat> [[TMP9]], 2
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP4]], 3
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP10]], <vscale x 8 x bfloat> [[TMP11]], 3
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP4]], 4
+// CHECK-RV64-NEXT:    store i64 [[TMP13]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP12]]
+//
+vbfloat16m2x4_t test_vlseg4e16ff_v_bf16m2x4_tu(vbfloat16m2x4_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vlseg4e16ff_v_bf16m2x4_tu(vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg4e16ff_v_bf16mf4x4_tum(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } @llvm.riscv.vlseg4ff.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP4]], 0
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } poison, <vscale x 1 x bfloat> [[TMP5]], 0
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP4]], 1
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP6]], <vscale x 1 x bfloat> [[TMP7]], 1
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP4]], 2
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP8]], <vscale x 1 x bfloat> [[TMP9]], 2
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP4]], 3
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP10]], <vscale x 1 x bfloat> [[TMP11]], 3
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP4]], 4
+// CHECK-RV64-NEXT:    store i64 [[TMP13]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP12]]
+//
+vbfloat16mf4x4_t test_vlseg4e16ff_v_bf16mf4x4_tum(vbool64_t vm, vbfloat16mf4x4_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vlseg4e16ff_v_bf16mf4x4_tum(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg4e16ff_v_bf16mf2x4_tum(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } @llvm.riscv.vlseg4ff.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP4]], 0
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } poison, <vscale x 2 x bfloat> [[TMP5]], 0
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP4]], 1
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP6]], <vscale x 2 x bfloat> [[TMP7]], 1
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP4]], 2
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP8]], <vscale x 2 x bfloat> [[TMP9]], 2
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP4]], 3
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP10]], <vscale x 2 x bfloat> [[TMP11]], 3
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP4]], 4
+// CHECK-RV64-NEXT:    store i64 [[TMP13]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP12]]
+//
+vbfloat16mf2x4_t test_vlseg4e16ff_v_bf16mf2x4_tum(vbool32_t vm, vbfloat16mf2x4_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vlseg4e16ff_v_bf16mf2x4_tum(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg4e16ff_v_bf16m1x4_tum(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } @llvm.riscv.vlseg4ff.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP4]], 0
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } poison, <vscale x 4 x bfloat> [[TMP5]], 0
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP4]], 1
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP6]], <vscale x 4 x bfloat> [[TMP7]], 1
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP4]], 2
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP8]], <vscale x 4 x bfloat> [[TMP9]], 2
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP4]], 3
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP10]], <vscale x 4 x bfloat> [[TMP11]], 3
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP4]], 4
+// CHECK-RV64-NEXT:    store i64 [[TMP13]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP12]]
+//
+vbfloat16m1x4_t test_vlseg4e16ff_v_bf16m1x4_tum(vbool16_t vm, vbfloat16m1x4_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vlseg4e16ff_v_bf16m1x4_tum(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vlseg4e16ff_v_bf16m2x4_tum(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } @llvm.riscv.vlseg4ff.mask.nxv8bf16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], <vscale x 8 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP4]], 0
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } poison, <vscale x 8 x bfloat> [[TMP5]], 0
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP4]], 1
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP6]], <vscale x 8 x bfloat> [[TMP7]], 1
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP4]], 2
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP8]], <vscale x 8 x bfloat> [[TMP9]], 2
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP4]], 3
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP10]], <vscale x 8 x bfloat> [[TMP11]], 3
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP4]], 4
+// CHECK-RV64-NEXT:    store i64 [[TMP13]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP12]]
+//
+vbfloat16m2x4_t test_vlseg4e16ff_v_bf16m2x4_tum(vbool8_t vm, vbfloat16m2x4_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vlseg4e16ff_v_bf16m2x4_tum(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg4e16ff_v_bf16mf4x4_tumu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } @llvm.riscv.vlseg4ff.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP4]], 0
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } poison, <vscale x 1 x bfloat> [[TMP5]], 0
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP4]], 1
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP6]], <vscale x 1 x bfloat> [[TMP7]], 1
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP4]], 2
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP8]], <vscale x 1 x bfloat> [[TMP9]], 2
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP4]], 3
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP10]], <vscale x 1 x bfloat> [[TMP11]], 3
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP4]], 4
+// CHECK-RV64-NEXT:    store i64 [[TMP13]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP12]]
+//
+vbfloat16mf4x4_t test_vlseg4e16ff_v_bf16mf4x4_tumu(vbool64_t vm, vbfloat16mf4x4_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vlseg4e16ff_v_bf16mf4x4_tumu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg4e16ff_v_bf16mf2x4_tumu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } @llvm.riscv.vlseg4ff.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP4]], 0
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } poison, <vscale x 2 x bfloat> [[TMP5]], 0
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP4]], 1
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP6]], <vscale x 2 x bfloat> [[TMP7]], 1
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP4]], 2
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP8]], <vscale x 2 x bfloat> [[TMP9]], 2
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP4]], 3
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP10]], <vscale x 2 x bfloat> [[TMP11]], 3
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP4]], 4
+// CHECK-RV64-NEXT:    store i64 [[TMP13]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP12]]
+//
+vbfloat16mf2x4_t test_vlseg4e16ff_v_bf16mf2x4_tumu(vbool32_t vm, vbfloat16mf2x4_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vlseg4e16ff_v_bf16mf2x4_tumu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg4e16ff_v_bf16m1x4_tumu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } @llvm.riscv.vlseg4ff.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP4]], 0
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } poison, <vscale x 4 x bfloat> [[TMP5]], 0
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP4]], 1
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP6]], <vscale x 4 x bfloat> [[TMP7]], 1
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP4]], 2
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP8]], <vscale x 4 x bfloat> [[TMP9]], 2
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP4]], 3
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP10]], <vscale x 4 x bfloat> [[TMP11]], 3
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP4]], 4
+// CHECK-RV64-NEXT:    store i64 [[TMP13]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP12]]
+//
+vbfloat16m1x4_t test_vlseg4e16ff_v_bf16m1x4_tumu(vbool16_t vm, vbfloat16m1x4_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vlseg4e16ff_v_bf16m1x4_tumu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vlseg4e16ff_v_bf16m2x4_tumu(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } @llvm.riscv.vlseg4ff.mask.nxv8bf16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], <vscale x 8 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP4]], 0
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } poison, <vscale x 8 x bfloat> [[TMP5]], 0
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP4]], 1
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP6]], <vscale x 8 x bfloat> [[TMP7]], 1
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP4]], 2
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP8]], <vscale x 8 x bfloat> [[TMP9]], 2
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP4]], 3
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP10]], <vscale x 8 x bfloat> [[TMP11]], 3
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP4]], 4
+// CHECK-RV64-NEXT:    store i64 [[TMP13]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP12]]
+//
+vbfloat16m2x4_t test_vlseg4e16ff_v_bf16m2x4_tumu(vbool8_t vm, vbfloat16m2x4_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vlseg4e16ff_v_bf16m2x4_tumu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg4e16ff_v_bf16mf4x4_mu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } @llvm.riscv.vlseg4ff.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP4]], 0
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } poison, <vscale x 1 x bfloat> [[TMP5]], 0
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP4]], 1
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP6]], <vscale x 1 x bfloat> [[TMP7]], 1
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP4]], 2
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP8]], <vscale x 1 x bfloat> [[TMP9]], 2
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP4]], 3
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP10]], <vscale x 1 x bfloat> [[TMP11]], 3
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP4]], 4
+// CHECK-RV64-NEXT:    store i64 [[TMP13]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP12]]
+//
+vbfloat16mf4x4_t test_vlseg4e16ff_v_bf16mf4x4_mu(vbool64_t vm, vbfloat16mf4x4_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vlseg4e16ff_v_bf16mf4x4_mu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg4e16ff_v_bf16mf2x4_mu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } @llvm.riscv.vlseg4ff.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP4]], 0
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } poison, <vscale x 2 x bfloat> [[TMP5]], 0
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP4]], 1
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP6]], <vscale x 2 x bfloat> [[TMP7]], 1
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP4]], 2
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP8]], <vscale x 2 x bfloat> [[TMP9]], 2
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP4]], 3
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP10]], <vscale x 2 x bfloat> [[TMP11]], 3
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP4]], 4
+// CHECK-RV64-NEXT:    store i64 [[TMP13]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP12]]
+//
+vbfloat16mf2x4_t test_vlseg4e16ff_v_bf16mf2x4_mu(vbool32_t vm, vbfloat16mf2x4_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vlseg4e16ff_v_bf16mf2x4_mu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg4e16ff_v_bf16m1x4_mu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } @llvm.riscv.vlseg4ff.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP4]], 0
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } poison, <vscale x 4 x bfloat> [[TMP5]], 0
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP4]], 1
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP6]], <vscale x 4 x bfloat> [[TMP7]], 1
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP4]], 2
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP8]], <vscale x 4 x bfloat> [[TMP9]], 2
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP4]], 3
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP10]], <vscale x 4 x bfloat> [[TMP11]], 3
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP4]], 4
+// CHECK-RV64-NEXT:    store i64 [[TMP13]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP12]]
+//
+vbfloat16m1x4_t test_vlseg4e16ff_v_bf16m1x4_mu(vbool16_t vm, vbfloat16m1x4_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vlseg4e16ff_v_bf16m1x4_mu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vlseg4e16ff_v_bf16m2x4_mu(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } @llvm.riscv.vlseg4ff.mask.nxv8bf16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], <vscale x 8 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP4]], 0
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } poison, <vscale x 8 x bfloat> [[TMP5]], 0
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP4]], 1
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP6]], <vscale x 8 x bfloat> [[TMP7]], 1
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP4]], 2
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP8]], <vscale x 8 x bfloat> [[TMP9]], 2
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP4]], 3
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP10]], <vscale x 8 x bfloat> [[TMP11]], 3
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP4]], 4
+// CHECK-RV64-NEXT:    store i64 [[TMP13]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP12]]
+//
+vbfloat16m2x4_t test_vlseg4e16ff_v_bf16m2x4_mu(vbool8_t vm, vbfloat16m2x4_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vlseg4e16ff_v_bf16m2x4_mu(vm, vd, rs1, new_vl, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vlseg5e16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vlseg5e16.c
new file mode 100644
index 000000000000..7ce59eb05eff
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vlseg5e16.c
@@ -0,0 +1,189 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg5e16_v_bf16mf4x5_tu(
+// CHECK-RV64-SAME: { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlseg5.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP5]]
+//
+vbfloat16mf4x5_t test_vlseg5e16_v_bf16mf4x5_tu(vbfloat16mf4x5_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg5e16_v_bf16mf4x5_tu(vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg5e16_v_bf16mf2x5_tu(
+// CHECK-RV64-SAME: { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlseg5.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP5]]
+//
+vbfloat16mf2x5_t test_vlseg5e16_v_bf16mf2x5_tu(vbfloat16mf2x5_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg5e16_v_bf16mf2x5_tu(vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg5e16_v_bf16m1x5_tu(
+// CHECK-RV64-SAME: { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlseg5.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP5]]
+//
+vbfloat16m1x5_t test_vlseg5e16_v_bf16m1x5_tu(vbfloat16m1x5_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg5e16_v_bf16m1x5_tu(vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg5e16_v_bf16mf4x5_tum(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlseg5.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP5]]
+//
+vbfloat16mf4x5_t test_vlseg5e16_v_bf16mf4x5_tum(vbool64_t vm, vbfloat16mf4x5_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg5e16_v_bf16mf4x5_tum(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg5e16_v_bf16mf2x5_tum(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlseg5.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP5]]
+//
+vbfloat16mf2x5_t test_vlseg5e16_v_bf16mf2x5_tum(vbool32_t vm, vbfloat16mf2x5_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg5e16_v_bf16mf2x5_tum(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg5e16_v_bf16m1x5_tum(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlseg5.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP5]]
+//
+vbfloat16m1x5_t test_vlseg5e16_v_bf16m1x5_tum(vbool16_t vm, vbfloat16m1x5_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg5e16_v_bf16m1x5_tum(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg5e16_v_bf16mf4x5_tumu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlseg5.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP5]]
+//
+vbfloat16mf4x5_t test_vlseg5e16_v_bf16mf4x5_tumu(vbool64_t vm, vbfloat16mf4x5_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg5e16_v_bf16mf4x5_tumu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg5e16_v_bf16mf2x5_tumu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlseg5.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP5]]
+//
+vbfloat16mf2x5_t test_vlseg5e16_v_bf16mf2x5_tumu(vbool32_t vm, vbfloat16mf2x5_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg5e16_v_bf16mf2x5_tumu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg5e16_v_bf16m1x5_tumu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlseg5.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP5]]
+//
+vbfloat16m1x5_t test_vlseg5e16_v_bf16m1x5_tumu(vbool16_t vm, vbfloat16m1x5_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg5e16_v_bf16m1x5_tumu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg5e16_v_bf16mf4x5_mu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlseg5.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP5]]
+//
+vbfloat16mf4x5_t test_vlseg5e16_v_bf16mf4x5_mu(vbool64_t vm, vbfloat16mf4x5_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg5e16_v_bf16mf4x5_mu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg5e16_v_bf16mf2x5_mu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlseg5.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP5]]
+//
+vbfloat16mf2x5_t test_vlseg5e16_v_bf16mf2x5_mu(vbool32_t vm, vbfloat16mf2x5_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg5e16_v_bf16mf2x5_mu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg5e16_v_bf16m1x5_mu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlseg5.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP5]]
+//
+vbfloat16m1x5_t test_vlseg5e16_v_bf16m1x5_mu(vbool16_t vm, vbfloat16m1x5_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg5e16_v_bf16m1x5_mu(vm, vd, rs1, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vlseg5e16ff.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vlseg5e16ff.c
new file mode 100644
index 000000000000..08b00d68fb96
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vlseg5e16ff.c
@@ -0,0 +1,333 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg5e16ff_v_bf16mf4x5_tu(
+// CHECK-RV64-SAME: { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } @llvm.riscv.vlseg5ff.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP5]], 0
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } poison, <vscale x 1 x bfloat> [[TMP6]], 0
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP5]], 1
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP7]], <vscale x 1 x bfloat> [[TMP8]], 1
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP5]], 2
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP9]], <vscale x 1 x bfloat> [[TMP10]], 2
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP5]], 3
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP11]], <vscale x 1 x bfloat> [[TMP12]], 3
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP5]], 4
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP13]], <vscale x 1 x bfloat> [[TMP14]], 4
+// CHECK-RV64-NEXT:    [[TMP16:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP5]], 5
+// CHECK-RV64-NEXT:    store i64 [[TMP16]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP15]]
+//
+vbfloat16mf4x5_t test_vlseg5e16ff_v_bf16mf4x5_tu(vbfloat16mf4x5_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vlseg5e16ff_v_bf16mf4x5_tu(vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg5e16ff_v_bf16mf2x5_tu(
+// CHECK-RV64-SAME: { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } @llvm.riscv.vlseg5ff.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP5]], 0
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } poison, <vscale x 2 x bfloat> [[TMP6]], 0
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP5]], 1
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP7]], <vscale x 2 x bfloat> [[TMP8]], 1
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP5]], 2
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP9]], <vscale x 2 x bfloat> [[TMP10]], 2
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP5]], 3
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP11]], <vscale x 2 x bfloat> [[TMP12]], 3
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP5]], 4
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP13]], <vscale x 2 x bfloat> [[TMP14]], 4
+// CHECK-RV64-NEXT:    [[TMP16:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP5]], 5
+// CHECK-RV64-NEXT:    store i64 [[TMP16]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP15]]
+//
+vbfloat16mf2x5_t test_vlseg5e16ff_v_bf16mf2x5_tu(vbfloat16mf2x5_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vlseg5e16ff_v_bf16mf2x5_tu(vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg5e16ff_v_bf16m1x5_tu(
+// CHECK-RV64-SAME: { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } @llvm.riscv.vlseg5ff.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP5]], 0
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } poison, <vscale x 4 x bfloat> [[TMP6]], 0
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP5]], 1
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP7]], <vscale x 4 x bfloat> [[TMP8]], 1
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP5]], 2
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP9]], <vscale x 4 x bfloat> [[TMP10]], 2
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP5]], 3
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP11]], <vscale x 4 x bfloat> [[TMP12]], 3
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP5]], 4
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP13]], <vscale x 4 x bfloat> [[TMP14]], 4
+// CHECK-RV64-NEXT:    [[TMP16:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP5]], 5
+// CHECK-RV64-NEXT:    store i64 [[TMP16]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP15]]
+//
+vbfloat16m1x5_t test_vlseg5e16ff_v_bf16m1x5_tu(vbfloat16m1x5_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vlseg5e16ff_v_bf16m1x5_tu(vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg5e16ff_v_bf16mf4x5_tum(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } @llvm.riscv.vlseg5ff.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP5]], 0
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } poison, <vscale x 1 x bfloat> [[TMP6]], 0
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP5]], 1
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP7]], <vscale x 1 x bfloat> [[TMP8]], 1
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP5]], 2
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP9]], <vscale x 1 x bfloat> [[TMP10]], 2
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP5]], 3
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP11]], <vscale x 1 x bfloat> [[TMP12]], 3
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP5]], 4
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP13]], <vscale x 1 x bfloat> [[TMP14]], 4
+// CHECK-RV64-NEXT:    [[TMP16:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP5]], 5
+// CHECK-RV64-NEXT:    store i64 [[TMP16]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP15]]
+//
+vbfloat16mf4x5_t test_vlseg5e16ff_v_bf16mf4x5_tum(vbool64_t vm, vbfloat16mf4x5_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vlseg5e16ff_v_bf16mf4x5_tum(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg5e16ff_v_bf16mf2x5_tum(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } @llvm.riscv.vlseg5ff.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP5]], 0
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } poison, <vscale x 2 x bfloat> [[TMP6]], 0
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP5]], 1
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP7]], <vscale x 2 x bfloat> [[TMP8]], 1
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP5]], 2
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP9]], <vscale x 2 x bfloat> [[TMP10]], 2
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP5]], 3
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP11]], <vscale x 2 x bfloat> [[TMP12]], 3
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP5]], 4
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP13]], <vscale x 2 x bfloat> [[TMP14]], 4
+// CHECK-RV64-NEXT:    [[TMP16:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP5]], 5
+// CHECK-RV64-NEXT:    store i64 [[TMP16]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP15]]
+//
+vbfloat16mf2x5_t test_vlseg5e16ff_v_bf16mf2x5_tum(vbool32_t vm, vbfloat16mf2x5_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vlseg5e16ff_v_bf16mf2x5_tum(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg5e16ff_v_bf16m1x5_tum(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } @llvm.riscv.vlseg5ff.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP5]], 0
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } poison, <vscale x 4 x bfloat> [[TMP6]], 0
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP5]], 1
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP7]], <vscale x 4 x bfloat> [[TMP8]], 1
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP5]], 2
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP9]], <vscale x 4 x bfloat> [[TMP10]], 2
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP5]], 3
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP11]], <vscale x 4 x bfloat> [[TMP12]], 3
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP5]], 4
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP13]], <vscale x 4 x bfloat> [[TMP14]], 4
+// CHECK-RV64-NEXT:    [[TMP16:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP5]], 5
+// CHECK-RV64-NEXT:    store i64 [[TMP16]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP15]]
+//
+vbfloat16m1x5_t test_vlseg5e16ff_v_bf16m1x5_tum(vbool16_t vm, vbfloat16m1x5_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vlseg5e16ff_v_bf16m1x5_tum(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg5e16ff_v_bf16mf4x5_tumu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } @llvm.riscv.vlseg5ff.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP5]], 0
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } poison, <vscale x 1 x bfloat> [[TMP6]], 0
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP5]], 1
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP7]], <vscale x 1 x bfloat> [[TMP8]], 1
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP5]], 2
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP9]], <vscale x 1 x bfloat> [[TMP10]], 2
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP5]], 3
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP11]], <vscale x 1 x bfloat> [[TMP12]], 3
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP5]], 4
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP13]], <vscale x 1 x bfloat> [[TMP14]], 4
+// CHECK-RV64-NEXT:    [[TMP16:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP5]], 5
+// CHECK-RV64-NEXT:    store i64 [[TMP16]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP15]]
+//
+vbfloat16mf4x5_t test_vlseg5e16ff_v_bf16mf4x5_tumu(vbool64_t vm, vbfloat16mf4x5_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vlseg5e16ff_v_bf16mf4x5_tumu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg5e16ff_v_bf16mf2x5_tumu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } @llvm.riscv.vlseg5ff.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP5]], 0
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } poison, <vscale x 2 x bfloat> [[TMP6]], 0
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP5]], 1
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP7]], <vscale x 2 x bfloat> [[TMP8]], 1
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP5]], 2
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP9]], <vscale x 2 x bfloat> [[TMP10]], 2
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP5]], 3
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP11]], <vscale x 2 x bfloat> [[TMP12]], 3
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP5]], 4
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP13]], <vscale x 2 x bfloat> [[TMP14]], 4
+// CHECK-RV64-NEXT:    [[TMP16:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP5]], 5
+// CHECK-RV64-NEXT:    store i64 [[TMP16]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP15]]
+//
+vbfloat16mf2x5_t test_vlseg5e16ff_v_bf16mf2x5_tumu(vbool32_t vm, vbfloat16mf2x5_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vlseg5e16ff_v_bf16mf2x5_tumu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg5e16ff_v_bf16m1x5_tumu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } @llvm.riscv.vlseg5ff.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP5]], 0
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } poison, <vscale x 4 x bfloat> [[TMP6]], 0
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP5]], 1
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP7]], <vscale x 4 x bfloat> [[TMP8]], 1
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP5]], 2
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP9]], <vscale x 4 x bfloat> [[TMP10]], 2
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP5]], 3
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP11]], <vscale x 4 x bfloat> [[TMP12]], 3
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP5]], 4
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP13]], <vscale x 4 x bfloat> [[TMP14]], 4
+// CHECK-RV64-NEXT:    [[TMP16:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP5]], 5
+// CHECK-RV64-NEXT:    store i64 [[TMP16]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP15]]
+//
+vbfloat16m1x5_t test_vlseg5e16ff_v_bf16m1x5_tumu(vbool16_t vm, vbfloat16m1x5_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vlseg5e16ff_v_bf16m1x5_tumu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg5e16ff_v_bf16mf4x5_mu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } @llvm.riscv.vlseg5ff.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP5]], 0
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } poison, <vscale x 1 x bfloat> [[TMP6]], 0
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP5]], 1
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP7]], <vscale x 1 x bfloat> [[TMP8]], 1
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP5]], 2
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP9]], <vscale x 1 x bfloat> [[TMP10]], 2
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP5]], 3
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP11]], <vscale x 1 x bfloat> [[TMP12]], 3
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP5]], 4
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP13]], <vscale x 1 x bfloat> [[TMP14]], 4
+// CHECK-RV64-NEXT:    [[TMP16:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP5]], 5
+// CHECK-RV64-NEXT:    store i64 [[TMP16]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP15]]
+//
+vbfloat16mf4x5_t test_vlseg5e16ff_v_bf16mf4x5_mu(vbool64_t vm, vbfloat16mf4x5_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vlseg5e16ff_v_bf16mf4x5_mu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg5e16ff_v_bf16mf2x5_mu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } @llvm.riscv.vlseg5ff.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP5]], 0
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } poison, <vscale x 2 x bfloat> [[TMP6]], 0
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP5]], 1
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP7]], <vscale x 2 x bfloat> [[TMP8]], 1
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP5]], 2
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP9]], <vscale x 2 x bfloat> [[TMP10]], 2
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP5]], 3
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP11]], <vscale x 2 x bfloat> [[TMP12]], 3
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP5]], 4
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP13]], <vscale x 2 x bfloat> [[TMP14]], 4
+// CHECK-RV64-NEXT:    [[TMP16:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP5]], 5
+// CHECK-RV64-NEXT:    store i64 [[TMP16]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP15]]
+//
+vbfloat16mf2x5_t test_vlseg5e16ff_v_bf16mf2x5_mu(vbool32_t vm, vbfloat16mf2x5_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vlseg5e16ff_v_bf16mf2x5_mu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg5e16ff_v_bf16m1x5_mu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } @llvm.riscv.vlseg5ff.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP5]], 0
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } poison, <vscale x 4 x bfloat> [[TMP6]], 0
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP5]], 1
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP7]], <vscale x 4 x bfloat> [[TMP8]], 1
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP5]], 2
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP9]], <vscale x 4 x bfloat> [[TMP10]], 2
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP5]], 3
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP11]], <vscale x 4 x bfloat> [[TMP12]], 3
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP5]], 4
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP13]], <vscale x 4 x bfloat> [[TMP14]], 4
+// CHECK-RV64-NEXT:    [[TMP16:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP5]], 5
+// CHECK-RV64-NEXT:    store i64 [[TMP16]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP15]]
+//
+vbfloat16m1x5_t test_vlseg5e16ff_v_bf16m1x5_mu(vbool16_t vm, vbfloat16m1x5_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vlseg5e16ff_v_bf16m1x5_mu(vm, vd, rs1, new_vl, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vlseg6e16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vlseg6e16.c
new file mode 100644
index 000000000000..1bcf360b86fe
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vlseg6e16.c
@@ -0,0 +1,201 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg6e16_v_bf16mf4x6_tu(
+// CHECK-RV64-SAME: { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlseg6.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP6]]
+//
+vbfloat16mf4x6_t test_vlseg6e16_v_bf16mf4x6_tu(vbfloat16mf4x6_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg6e16_v_bf16mf4x6_tu(vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg6e16_v_bf16mf2x6_tu(
+// CHECK-RV64-SAME: { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlseg6.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP6]]
+//
+vbfloat16mf2x6_t test_vlseg6e16_v_bf16mf2x6_tu(vbfloat16mf2x6_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg6e16_v_bf16mf2x6_tu(vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg6e16_v_bf16m1x6_tu(
+// CHECK-RV64-SAME: { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlseg6.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP6]]
+//
+vbfloat16m1x6_t test_vlseg6e16_v_bf16m1x6_tu(vbfloat16m1x6_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg6e16_v_bf16m1x6_tu(vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg6e16_v_bf16mf4x6_tum(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlseg6.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP6]]
+//
+vbfloat16mf4x6_t test_vlseg6e16_v_bf16mf4x6_tum(vbool64_t vm, vbfloat16mf4x6_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg6e16_v_bf16mf4x6_tum(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg6e16_v_bf16mf2x6_tum(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlseg6.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP6]]
+//
+vbfloat16mf2x6_t test_vlseg6e16_v_bf16mf2x6_tum(vbool32_t vm, vbfloat16mf2x6_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg6e16_v_bf16mf2x6_tum(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg6e16_v_bf16m1x6_tum(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlseg6.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP6]]
+//
+vbfloat16m1x6_t test_vlseg6e16_v_bf16m1x6_tum(vbool16_t vm, vbfloat16m1x6_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg6e16_v_bf16m1x6_tum(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg6e16_v_bf16mf4x6_tumu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlseg6.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP6]]
+//
+vbfloat16mf4x6_t test_vlseg6e16_v_bf16mf4x6_tumu(vbool64_t vm, vbfloat16mf4x6_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg6e16_v_bf16mf4x6_tumu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg6e16_v_bf16mf2x6_tumu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlseg6.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP6]]
+//
+vbfloat16mf2x6_t test_vlseg6e16_v_bf16mf2x6_tumu(vbool32_t vm, vbfloat16mf2x6_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg6e16_v_bf16mf2x6_tumu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg6e16_v_bf16m1x6_tumu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlseg6.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP6]]
+//
+vbfloat16m1x6_t test_vlseg6e16_v_bf16m1x6_tumu(vbool16_t vm, vbfloat16m1x6_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg6e16_v_bf16m1x6_tumu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg6e16_v_bf16mf4x6_mu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlseg6.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP6]]
+//
+vbfloat16mf4x6_t test_vlseg6e16_v_bf16mf4x6_mu(vbool64_t vm, vbfloat16mf4x6_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg6e16_v_bf16mf4x6_mu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg6e16_v_bf16mf2x6_mu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlseg6.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP6]]
+//
+vbfloat16mf2x6_t test_vlseg6e16_v_bf16mf2x6_mu(vbool32_t vm, vbfloat16mf2x6_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg6e16_v_bf16mf2x6_mu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg6e16_v_bf16m1x6_mu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlseg6.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP6]]
+//
+vbfloat16m1x6_t test_vlseg6e16_v_bf16m1x6_mu(vbool16_t vm, vbfloat16m1x6_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg6e16_v_bf16m1x6_mu(vm, vd, rs1, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vlseg6e16ff.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vlseg6e16ff.c
new file mode 100644
index 000000000000..ab2a169cf2a0
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vlseg6e16ff.c
@@ -0,0 +1,369 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg6e16ff_v_bf16mf4x6_tu(
+// CHECK-RV64-SAME: { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } @llvm.riscv.vlseg6ff.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP6]], 0
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } poison, <vscale x 1 x bfloat> [[TMP7]], 0
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP6]], 1
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP8]], <vscale x 1 x bfloat> [[TMP9]], 1
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP6]], 2
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP10]], <vscale x 1 x bfloat> [[TMP11]], 2
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP6]], 3
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP12]], <vscale x 1 x bfloat> [[TMP13]], 3
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP6]], 4
+// CHECK-RV64-NEXT:    [[TMP16:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP14]], <vscale x 1 x bfloat> [[TMP15]], 4
+// CHECK-RV64-NEXT:    [[TMP17:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP6]], 5
+// CHECK-RV64-NEXT:    [[TMP18:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP16]], <vscale x 1 x bfloat> [[TMP17]], 5
+// CHECK-RV64-NEXT:    [[TMP19:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP6]], 6
+// CHECK-RV64-NEXT:    store i64 [[TMP19]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP18]]
+//
+vbfloat16mf4x6_t test_vlseg6e16ff_v_bf16mf4x6_tu(vbfloat16mf4x6_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vlseg6e16ff_v_bf16mf4x6_tu(vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg6e16ff_v_bf16mf2x6_tu(
+// CHECK-RV64-SAME: { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } @llvm.riscv.vlseg6ff.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP6]], 0
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } poison, <vscale x 2 x bfloat> [[TMP7]], 0
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP6]], 1
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP8]], <vscale x 2 x bfloat> [[TMP9]], 1
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP6]], 2
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP10]], <vscale x 2 x bfloat> [[TMP11]], 2
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP6]], 3
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP12]], <vscale x 2 x bfloat> [[TMP13]], 3
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP6]], 4
+// CHECK-RV64-NEXT:    [[TMP16:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP14]], <vscale x 2 x bfloat> [[TMP15]], 4
+// CHECK-RV64-NEXT:    [[TMP17:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP6]], 5
+// CHECK-RV64-NEXT:    [[TMP18:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP16]], <vscale x 2 x bfloat> [[TMP17]], 5
+// CHECK-RV64-NEXT:    [[TMP19:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP6]], 6
+// CHECK-RV64-NEXT:    store i64 [[TMP19]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP18]]
+//
+vbfloat16mf2x6_t test_vlseg6e16ff_v_bf16mf2x6_tu(vbfloat16mf2x6_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vlseg6e16ff_v_bf16mf2x6_tu(vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg6e16ff_v_bf16m1x6_tu(
+// CHECK-RV64-SAME: { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } @llvm.riscv.vlseg6ff.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP6]], 0
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } poison, <vscale x 4 x bfloat> [[TMP7]], 0
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP6]], 1
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP8]], <vscale x 4 x bfloat> [[TMP9]], 1
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP6]], 2
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP10]], <vscale x 4 x bfloat> [[TMP11]], 2
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP6]], 3
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP12]], <vscale x 4 x bfloat> [[TMP13]], 3
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP6]], 4
+// CHECK-RV64-NEXT:    [[TMP16:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP14]], <vscale x 4 x bfloat> [[TMP15]], 4
+// CHECK-RV64-NEXT:    [[TMP17:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP6]], 5
+// CHECK-RV64-NEXT:    [[TMP18:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP16]], <vscale x 4 x bfloat> [[TMP17]], 5
+// CHECK-RV64-NEXT:    [[TMP19:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP6]], 6
+// CHECK-RV64-NEXT:    store i64 [[TMP19]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP18]]
+//
+vbfloat16m1x6_t test_vlseg6e16ff_v_bf16m1x6_tu(vbfloat16m1x6_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vlseg6e16ff_v_bf16m1x6_tu(vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg6e16ff_v_bf16mf4x6_tum(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } @llvm.riscv.vlseg6ff.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP6]], 0
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } poison, <vscale x 1 x bfloat> [[TMP7]], 0
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP6]], 1
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP8]], <vscale x 1 x bfloat> [[TMP9]], 1
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP6]], 2
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP10]], <vscale x 1 x bfloat> [[TMP11]], 2
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP6]], 3
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP12]], <vscale x 1 x bfloat> [[TMP13]], 3
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP6]], 4
+// CHECK-RV64-NEXT:    [[TMP16:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP14]], <vscale x 1 x bfloat> [[TMP15]], 4
+// CHECK-RV64-NEXT:    [[TMP17:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP6]], 5
+// CHECK-RV64-NEXT:    [[TMP18:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP16]], <vscale x 1 x bfloat> [[TMP17]], 5
+// CHECK-RV64-NEXT:    [[TMP19:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP6]], 6
+// CHECK-RV64-NEXT:    store i64 [[TMP19]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP18]]
+//
+vbfloat16mf4x6_t test_vlseg6e16ff_v_bf16mf4x6_tum(vbool64_t vm, vbfloat16mf4x6_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vlseg6e16ff_v_bf16mf4x6_tum(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg6e16ff_v_bf16mf2x6_tum(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } @llvm.riscv.vlseg6ff.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP6]], 0
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } poison, <vscale x 2 x bfloat> [[TMP7]], 0
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP6]], 1
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP8]], <vscale x 2 x bfloat> [[TMP9]], 1
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP6]], 2
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP10]], <vscale x 2 x bfloat> [[TMP11]], 2
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP6]], 3
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP12]], <vscale x 2 x bfloat> [[TMP13]], 3
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP6]], 4
+// CHECK-RV64-NEXT:    [[TMP16:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP14]], <vscale x 2 x bfloat> [[TMP15]], 4
+// CHECK-RV64-NEXT:    [[TMP17:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP6]], 5
+// CHECK-RV64-NEXT:    [[TMP18:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP16]], <vscale x 2 x bfloat> [[TMP17]], 5
+// CHECK-RV64-NEXT:    [[TMP19:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP6]], 6
+// CHECK-RV64-NEXT:    store i64 [[TMP19]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP18]]
+//
+vbfloat16mf2x6_t test_vlseg6e16ff_v_bf16mf2x6_tum(vbool32_t vm, vbfloat16mf2x6_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vlseg6e16ff_v_bf16mf2x6_tum(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg6e16ff_v_bf16m1x6_tum(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } @llvm.riscv.vlseg6ff.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP6]], 0
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } poison, <vscale x 4 x bfloat> [[TMP7]], 0
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP6]], 1
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP8]], <vscale x 4 x bfloat> [[TMP9]], 1
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP6]], 2
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP10]], <vscale x 4 x bfloat> [[TMP11]], 2
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP6]], 3
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP12]], <vscale x 4 x bfloat> [[TMP13]], 3
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP6]], 4
+// CHECK-RV64-NEXT:    [[TMP16:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP14]], <vscale x 4 x bfloat> [[TMP15]], 4
+// CHECK-RV64-NEXT:    [[TMP17:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP6]], 5
+// CHECK-RV64-NEXT:    [[TMP18:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP16]], <vscale x 4 x bfloat> [[TMP17]], 5
+// CHECK-RV64-NEXT:    [[TMP19:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP6]], 6
+// CHECK-RV64-NEXT:    store i64 [[TMP19]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP18]]
+//
+vbfloat16m1x6_t test_vlseg6e16ff_v_bf16m1x6_tum(vbool16_t vm, vbfloat16m1x6_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vlseg6e16ff_v_bf16m1x6_tum(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg6e16ff_v_bf16mf4x6_tumu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } @llvm.riscv.vlseg6ff.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP6]], 0
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } poison, <vscale x 1 x bfloat> [[TMP7]], 0
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP6]], 1
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP8]], <vscale x 1 x bfloat> [[TMP9]], 1
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP6]], 2
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP10]], <vscale x 1 x bfloat> [[TMP11]], 2
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP6]], 3
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP12]], <vscale x 1 x bfloat> [[TMP13]], 3
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP6]], 4
+// CHECK-RV64-NEXT:    [[TMP16:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP14]], <vscale x 1 x bfloat> [[TMP15]], 4
+// CHECK-RV64-NEXT:    [[TMP17:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP6]], 5
+// CHECK-RV64-NEXT:    [[TMP18:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP16]], <vscale x 1 x bfloat> [[TMP17]], 5
+// CHECK-RV64-NEXT:    [[TMP19:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP6]], 6
+// CHECK-RV64-NEXT:    store i64 [[TMP19]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP18]]
+//
+vbfloat16mf4x6_t test_vlseg6e16ff_v_bf16mf4x6_tumu(vbool64_t vm, vbfloat16mf4x6_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vlseg6e16ff_v_bf16mf4x6_tumu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg6e16ff_v_bf16mf2x6_tumu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } @llvm.riscv.vlseg6ff.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP6]], 0
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } poison, <vscale x 2 x bfloat> [[TMP7]], 0
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP6]], 1
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP8]], <vscale x 2 x bfloat> [[TMP9]], 1
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP6]], 2
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP10]], <vscale x 2 x bfloat> [[TMP11]], 2
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP6]], 3
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP12]], <vscale x 2 x bfloat> [[TMP13]], 3
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP6]], 4
+// CHECK-RV64-NEXT:    [[TMP16:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP14]], <vscale x 2 x bfloat> [[TMP15]], 4
+// CHECK-RV64-NEXT:    [[TMP17:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP6]], 5
+// CHECK-RV64-NEXT:    [[TMP18:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP16]], <vscale x 2 x bfloat> [[TMP17]], 5
+// CHECK-RV64-NEXT:    [[TMP19:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP6]], 6
+// CHECK-RV64-NEXT:    store i64 [[TMP19]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP18]]
+//
+vbfloat16mf2x6_t test_vlseg6e16ff_v_bf16mf2x6_tumu(vbool32_t vm, vbfloat16mf2x6_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vlseg6e16ff_v_bf16mf2x6_tumu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg6e16ff_v_bf16m1x6_tumu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } @llvm.riscv.vlseg6ff.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP6]], 0
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } poison, <vscale x 4 x bfloat> [[TMP7]], 0
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP6]], 1
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP8]], <vscale x 4 x bfloat> [[TMP9]], 1
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP6]], 2
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP10]], <vscale x 4 x bfloat> [[TMP11]], 2
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP6]], 3
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP12]], <vscale x 4 x bfloat> [[TMP13]], 3
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP6]], 4
+// CHECK-RV64-NEXT:    [[TMP16:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP14]], <vscale x 4 x bfloat> [[TMP15]], 4
+// CHECK-RV64-NEXT:    [[TMP17:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP6]], 5
+// CHECK-RV64-NEXT:    [[TMP18:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP16]], <vscale x 4 x bfloat> [[TMP17]], 5
+// CHECK-RV64-NEXT:    [[TMP19:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP6]], 6
+// CHECK-RV64-NEXT:    store i64 [[TMP19]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP18]]
+//
+vbfloat16m1x6_t test_vlseg6e16ff_v_bf16m1x6_tumu(vbool16_t vm, vbfloat16m1x6_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vlseg6e16ff_v_bf16m1x6_tumu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg6e16ff_v_bf16mf4x6_mu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } @llvm.riscv.vlseg6ff.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP6]], 0
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } poison, <vscale x 1 x bfloat> [[TMP7]], 0
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP6]], 1
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP8]], <vscale x 1 x bfloat> [[TMP9]], 1
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP6]], 2
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP10]], <vscale x 1 x bfloat> [[TMP11]], 2
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP6]], 3
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP12]], <vscale x 1 x bfloat> [[TMP13]], 3
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP6]], 4
+// CHECK-RV64-NEXT:    [[TMP16:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP14]], <vscale x 1 x bfloat> [[TMP15]], 4
+// CHECK-RV64-NEXT:    [[TMP17:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP6]], 5
+// CHECK-RV64-NEXT:    [[TMP18:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP16]], <vscale x 1 x bfloat> [[TMP17]], 5
+// CHECK-RV64-NEXT:    [[TMP19:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP6]], 6
+// CHECK-RV64-NEXT:    store i64 [[TMP19]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP18]]
+//
+vbfloat16mf4x6_t test_vlseg6e16ff_v_bf16mf4x6_mu(vbool64_t vm, vbfloat16mf4x6_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vlseg6e16ff_v_bf16mf4x6_mu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg6e16ff_v_bf16mf2x6_mu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } @llvm.riscv.vlseg6ff.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP6]], 0
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } poison, <vscale x 2 x bfloat> [[TMP7]], 0
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP6]], 1
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP8]], <vscale x 2 x bfloat> [[TMP9]], 1
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP6]], 2
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP10]], <vscale x 2 x bfloat> [[TMP11]], 2
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP6]], 3
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP12]], <vscale x 2 x bfloat> [[TMP13]], 3
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP6]], 4
+// CHECK-RV64-NEXT:    [[TMP16:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP14]], <vscale x 2 x bfloat> [[TMP15]], 4
+// CHECK-RV64-NEXT:    [[TMP17:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP6]], 5
+// CHECK-RV64-NEXT:    [[TMP18:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP16]], <vscale x 2 x bfloat> [[TMP17]], 5
+// CHECK-RV64-NEXT:    [[TMP19:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP6]], 6
+// CHECK-RV64-NEXT:    store i64 [[TMP19]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP18]]
+//
+vbfloat16mf2x6_t test_vlseg6e16ff_v_bf16mf2x6_mu(vbool32_t vm, vbfloat16mf2x6_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vlseg6e16ff_v_bf16mf2x6_mu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg6e16ff_v_bf16m1x6_mu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } @llvm.riscv.vlseg6ff.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP6]], 0
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } poison, <vscale x 4 x bfloat> [[TMP7]], 0
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP6]], 1
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP8]], <vscale x 4 x bfloat> [[TMP9]], 1
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP6]], 2
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP10]], <vscale x 4 x bfloat> [[TMP11]], 2
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP6]], 3
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP12]], <vscale x 4 x bfloat> [[TMP13]], 3
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP6]], 4
+// CHECK-RV64-NEXT:    [[TMP16:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP14]], <vscale x 4 x bfloat> [[TMP15]], 4
+// CHECK-RV64-NEXT:    [[TMP17:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP6]], 5
+// CHECK-RV64-NEXT:    [[TMP18:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP16]], <vscale x 4 x bfloat> [[TMP17]], 5
+// CHECK-RV64-NEXT:    [[TMP19:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP6]], 6
+// CHECK-RV64-NEXT:    store i64 [[TMP19]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP18]]
+//
+vbfloat16m1x6_t test_vlseg6e16ff_v_bf16m1x6_mu(vbool16_t vm, vbfloat16m1x6_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vlseg6e16ff_v_bf16m1x6_mu(vm, vd, rs1, new_vl, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vlseg7e16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vlseg7e16.c
new file mode 100644
index 000000000000..0147a14abcd3
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vlseg7e16.c
@@ -0,0 +1,213 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg7e16_v_bf16mf4x7_tu(
+// CHECK-RV64-SAME: { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlseg7.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP7]]
+//
+vbfloat16mf4x7_t test_vlseg7e16_v_bf16mf4x7_tu(vbfloat16mf4x7_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg7e16_v_bf16mf4x7_tu(vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg7e16_v_bf16mf2x7_tu(
+// CHECK-RV64-SAME: { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlseg7.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP7]]
+//
+vbfloat16mf2x7_t test_vlseg7e16_v_bf16mf2x7_tu(vbfloat16mf2x7_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg7e16_v_bf16mf2x7_tu(vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg7e16_v_bf16m1x7_tu(
+// CHECK-RV64-SAME: { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlseg7.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP7]]
+//
+vbfloat16m1x7_t test_vlseg7e16_v_bf16m1x7_tu(vbfloat16m1x7_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg7e16_v_bf16m1x7_tu(vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg7e16_v_bf16mf4x7_tum(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlseg7.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP7]]
+//
+vbfloat16mf4x7_t test_vlseg7e16_v_bf16mf4x7_tum(vbool64_t vm, vbfloat16mf4x7_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg7e16_v_bf16mf4x7_tum(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg7e16_v_bf16mf2x7_tum(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlseg7.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP7]]
+//
+vbfloat16mf2x7_t test_vlseg7e16_v_bf16mf2x7_tum(vbool32_t vm, vbfloat16mf2x7_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg7e16_v_bf16mf2x7_tum(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg7e16_v_bf16m1x7_tum(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlseg7.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP7]]
+//
+vbfloat16m1x7_t test_vlseg7e16_v_bf16m1x7_tum(vbool16_t vm, vbfloat16m1x7_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg7e16_v_bf16m1x7_tum(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg7e16_v_bf16mf4x7_tumu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlseg7.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP7]]
+//
+vbfloat16mf4x7_t test_vlseg7e16_v_bf16mf4x7_tumu(vbool64_t vm, vbfloat16mf4x7_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg7e16_v_bf16mf4x7_tumu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg7e16_v_bf16mf2x7_tumu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlseg7.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP7]]
+//
+vbfloat16mf2x7_t test_vlseg7e16_v_bf16mf2x7_tumu(vbool32_t vm, vbfloat16mf2x7_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg7e16_v_bf16mf2x7_tumu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg7e16_v_bf16m1x7_tumu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlseg7.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP7]]
+//
+vbfloat16m1x7_t test_vlseg7e16_v_bf16m1x7_tumu(vbool16_t vm, vbfloat16m1x7_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg7e16_v_bf16m1x7_tumu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg7e16_v_bf16mf4x7_mu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlseg7.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP7]]
+//
+vbfloat16mf4x7_t test_vlseg7e16_v_bf16mf4x7_mu(vbool64_t vm, vbfloat16mf4x7_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg7e16_v_bf16mf4x7_mu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg7e16_v_bf16mf2x7_mu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlseg7.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP7]]
+//
+vbfloat16mf2x7_t test_vlseg7e16_v_bf16mf2x7_mu(vbool32_t vm, vbfloat16mf2x7_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg7e16_v_bf16mf2x7_mu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg7e16_v_bf16m1x7_mu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlseg7.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP7]]
+//
+vbfloat16m1x7_t test_vlseg7e16_v_bf16m1x7_mu(vbool16_t vm, vbfloat16m1x7_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg7e16_v_bf16m1x7_mu(vm, vd, rs1, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vlseg7e16ff.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vlseg7e16ff.c
new file mode 100644
index 000000000000..1eee11324dba
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vlseg7e16ff.c
@@ -0,0 +1,405 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg7e16ff_v_bf16mf4x7_tu(
+// CHECK-RV64-SAME: { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } @llvm.riscv.vlseg7ff.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP7]], 0
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } poison, <vscale x 1 x bfloat> [[TMP8]], 0
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP7]], 1
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP9]], <vscale x 1 x bfloat> [[TMP10]], 1
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP7]], 2
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP11]], <vscale x 1 x bfloat> [[TMP12]], 2
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP7]], 3
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP13]], <vscale x 1 x bfloat> [[TMP14]], 3
+// CHECK-RV64-NEXT:    [[TMP16:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP7]], 4
+// CHECK-RV64-NEXT:    [[TMP17:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP15]], <vscale x 1 x bfloat> [[TMP16]], 4
+// CHECK-RV64-NEXT:    [[TMP18:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP7]], 5
+// CHECK-RV64-NEXT:    [[TMP19:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP17]], <vscale x 1 x bfloat> [[TMP18]], 5
+// CHECK-RV64-NEXT:    [[TMP20:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP7]], 6
+// CHECK-RV64-NEXT:    [[TMP21:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP19]], <vscale x 1 x bfloat> [[TMP20]], 6
+// CHECK-RV64-NEXT:    [[TMP22:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP7]], 7
+// CHECK-RV64-NEXT:    store i64 [[TMP22]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP21]]
+//
+vbfloat16mf4x7_t test_vlseg7e16ff_v_bf16mf4x7_tu(vbfloat16mf4x7_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vlseg7e16ff_v_bf16mf4x7_tu(vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg7e16ff_v_bf16mf2x7_tu(
+// CHECK-RV64-SAME: { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } @llvm.riscv.vlseg7ff.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP7]], 0
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } poison, <vscale x 2 x bfloat> [[TMP8]], 0
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP7]], 1
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP9]], <vscale x 2 x bfloat> [[TMP10]], 1
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP7]], 2
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP11]], <vscale x 2 x bfloat> [[TMP12]], 2
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP7]], 3
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP13]], <vscale x 2 x bfloat> [[TMP14]], 3
+// CHECK-RV64-NEXT:    [[TMP16:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP7]], 4
+// CHECK-RV64-NEXT:    [[TMP17:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP15]], <vscale x 2 x bfloat> [[TMP16]], 4
+// CHECK-RV64-NEXT:    [[TMP18:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP7]], 5
+// CHECK-RV64-NEXT:    [[TMP19:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP17]], <vscale x 2 x bfloat> [[TMP18]], 5
+// CHECK-RV64-NEXT:    [[TMP20:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP7]], 6
+// CHECK-RV64-NEXT:    [[TMP21:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP19]], <vscale x 2 x bfloat> [[TMP20]], 6
+// CHECK-RV64-NEXT:    [[TMP22:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP7]], 7
+// CHECK-RV64-NEXT:    store i64 [[TMP22]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP21]]
+//
+vbfloat16mf2x7_t test_vlseg7e16ff_v_bf16mf2x7_tu(vbfloat16mf2x7_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vlseg7e16ff_v_bf16mf2x7_tu(vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg7e16ff_v_bf16m1x7_tu(
+// CHECK-RV64-SAME: { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } @llvm.riscv.vlseg7ff.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP7]], 0
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } poison, <vscale x 4 x bfloat> [[TMP8]], 0
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP7]], 1
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP9]], <vscale x 4 x bfloat> [[TMP10]], 1
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP7]], 2
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP11]], <vscale x 4 x bfloat> [[TMP12]], 2
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP7]], 3
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP13]], <vscale x 4 x bfloat> [[TMP14]], 3
+// CHECK-RV64-NEXT:    [[TMP16:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP7]], 4
+// CHECK-RV64-NEXT:    [[TMP17:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP15]], <vscale x 4 x bfloat> [[TMP16]], 4
+// CHECK-RV64-NEXT:    [[TMP18:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP7]], 5
+// CHECK-RV64-NEXT:    [[TMP19:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP17]], <vscale x 4 x bfloat> [[TMP18]], 5
+// CHECK-RV64-NEXT:    [[TMP20:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP7]], 6
+// CHECK-RV64-NEXT:    [[TMP21:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP19]], <vscale x 4 x bfloat> [[TMP20]], 6
+// CHECK-RV64-NEXT:    [[TMP22:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP7]], 7
+// CHECK-RV64-NEXT:    store i64 [[TMP22]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP21]]
+//
+vbfloat16m1x7_t test_vlseg7e16ff_v_bf16m1x7_tu(vbfloat16m1x7_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vlseg7e16ff_v_bf16m1x7_tu(vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg7e16ff_v_bf16mf4x7_tum(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } @llvm.riscv.vlseg7ff.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP7]], 0
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } poison, <vscale x 1 x bfloat> [[TMP8]], 0
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP7]], 1
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP9]], <vscale x 1 x bfloat> [[TMP10]], 1
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP7]], 2
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP11]], <vscale x 1 x bfloat> [[TMP12]], 2
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP7]], 3
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP13]], <vscale x 1 x bfloat> [[TMP14]], 3
+// CHECK-RV64-NEXT:    [[TMP16:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP7]], 4
+// CHECK-RV64-NEXT:    [[TMP17:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP15]], <vscale x 1 x bfloat> [[TMP16]], 4
+// CHECK-RV64-NEXT:    [[TMP18:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP7]], 5
+// CHECK-RV64-NEXT:    [[TMP19:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP17]], <vscale x 1 x bfloat> [[TMP18]], 5
+// CHECK-RV64-NEXT:    [[TMP20:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP7]], 6
+// CHECK-RV64-NEXT:    [[TMP21:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP19]], <vscale x 1 x bfloat> [[TMP20]], 6
+// CHECK-RV64-NEXT:    [[TMP22:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP7]], 7
+// CHECK-RV64-NEXT:    store i64 [[TMP22]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP21]]
+//
+vbfloat16mf4x7_t test_vlseg7e16ff_v_bf16mf4x7_tum(vbool64_t vm, vbfloat16mf4x7_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vlseg7e16ff_v_bf16mf4x7_tum(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg7e16ff_v_bf16mf2x7_tum(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } @llvm.riscv.vlseg7ff.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP7]], 0
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } poison, <vscale x 2 x bfloat> [[TMP8]], 0
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP7]], 1
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP9]], <vscale x 2 x bfloat> [[TMP10]], 1
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP7]], 2
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP11]], <vscale x 2 x bfloat> [[TMP12]], 2
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP7]], 3
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP13]], <vscale x 2 x bfloat> [[TMP14]], 3
+// CHECK-RV64-NEXT:    [[TMP16:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP7]], 4
+// CHECK-RV64-NEXT:    [[TMP17:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP15]], <vscale x 2 x bfloat> [[TMP16]], 4
+// CHECK-RV64-NEXT:    [[TMP18:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP7]], 5
+// CHECK-RV64-NEXT:    [[TMP19:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP17]], <vscale x 2 x bfloat> [[TMP18]], 5
+// CHECK-RV64-NEXT:    [[TMP20:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP7]], 6
+// CHECK-RV64-NEXT:    [[TMP21:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP19]], <vscale x 2 x bfloat> [[TMP20]], 6
+// CHECK-RV64-NEXT:    [[TMP22:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP7]], 7
+// CHECK-RV64-NEXT:    store i64 [[TMP22]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP21]]
+//
+vbfloat16mf2x7_t test_vlseg7e16ff_v_bf16mf2x7_tum(vbool32_t vm, vbfloat16mf2x7_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vlseg7e16ff_v_bf16mf2x7_tum(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg7e16ff_v_bf16m1x7_tum(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } @llvm.riscv.vlseg7ff.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP7]], 0
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } poison, <vscale x 4 x bfloat> [[TMP8]], 0
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP7]], 1
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP9]], <vscale x 4 x bfloat> [[TMP10]], 1
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP7]], 2
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP11]], <vscale x 4 x bfloat> [[TMP12]], 2
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP7]], 3
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP13]], <vscale x 4 x bfloat> [[TMP14]], 3
+// CHECK-RV64-NEXT:    [[TMP16:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP7]], 4
+// CHECK-RV64-NEXT:    [[TMP17:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP15]], <vscale x 4 x bfloat> [[TMP16]], 4
+// CHECK-RV64-NEXT:    [[TMP18:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP7]], 5
+// CHECK-RV64-NEXT:    [[TMP19:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP17]], <vscale x 4 x bfloat> [[TMP18]], 5
+// CHECK-RV64-NEXT:    [[TMP20:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP7]], 6
+// CHECK-RV64-NEXT:    [[TMP21:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP19]], <vscale x 4 x bfloat> [[TMP20]], 6
+// CHECK-RV64-NEXT:    [[TMP22:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP7]], 7
+// CHECK-RV64-NEXT:    store i64 [[TMP22]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP21]]
+//
+vbfloat16m1x7_t test_vlseg7e16ff_v_bf16m1x7_tum(vbool16_t vm, vbfloat16m1x7_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vlseg7e16ff_v_bf16m1x7_tum(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg7e16ff_v_bf16mf4x7_tumu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } @llvm.riscv.vlseg7ff.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP7]], 0
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } poison, <vscale x 1 x bfloat> [[TMP8]], 0
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP7]], 1
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP9]], <vscale x 1 x bfloat> [[TMP10]], 1
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP7]], 2
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP11]], <vscale x 1 x bfloat> [[TMP12]], 2
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP7]], 3
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP13]], <vscale x 1 x bfloat> [[TMP14]], 3
+// CHECK-RV64-NEXT:    [[TMP16:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP7]], 4
+// CHECK-RV64-NEXT:    [[TMP17:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP15]], <vscale x 1 x bfloat> [[TMP16]], 4
+// CHECK-RV64-NEXT:    [[TMP18:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP7]], 5
+// CHECK-RV64-NEXT:    [[TMP19:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP17]], <vscale x 1 x bfloat> [[TMP18]], 5
+// CHECK-RV64-NEXT:    [[TMP20:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP7]], 6
+// CHECK-RV64-NEXT:    [[TMP21:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP19]], <vscale x 1 x bfloat> [[TMP20]], 6
+// CHECK-RV64-NEXT:    [[TMP22:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP7]], 7
+// CHECK-RV64-NEXT:    store i64 [[TMP22]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP21]]
+//
+vbfloat16mf4x7_t test_vlseg7e16ff_v_bf16mf4x7_tumu(vbool64_t vm, vbfloat16mf4x7_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vlseg7e16ff_v_bf16mf4x7_tumu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg7e16ff_v_bf16mf2x7_tumu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } @llvm.riscv.vlseg7ff.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP7]], 0
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } poison, <vscale x 2 x bfloat> [[TMP8]], 0
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP7]], 1
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP9]], <vscale x 2 x bfloat> [[TMP10]], 1
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP7]], 2
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP11]], <vscale x 2 x bfloat> [[TMP12]], 2
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP7]], 3
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP13]], <vscale x 2 x bfloat> [[TMP14]], 3
+// CHECK-RV64-NEXT:    [[TMP16:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP7]], 4
+// CHECK-RV64-NEXT:    [[TMP17:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP15]], <vscale x 2 x bfloat> [[TMP16]], 4
+// CHECK-RV64-NEXT:    [[TMP18:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP7]], 5
+// CHECK-RV64-NEXT:    [[TMP19:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP17]], <vscale x 2 x bfloat> [[TMP18]], 5
+// CHECK-RV64-NEXT:    [[TMP20:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP7]], 6
+// CHECK-RV64-NEXT:    [[TMP21:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP19]], <vscale x 2 x bfloat> [[TMP20]], 6
+// CHECK-RV64-NEXT:    [[TMP22:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP7]], 7
+// CHECK-RV64-NEXT:    store i64 [[TMP22]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP21]]
+//
+vbfloat16mf2x7_t test_vlseg7e16ff_v_bf16mf2x7_tumu(vbool32_t vm, vbfloat16mf2x7_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vlseg7e16ff_v_bf16mf2x7_tumu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg7e16ff_v_bf16m1x7_tumu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } @llvm.riscv.vlseg7ff.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP7]], 0
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } poison, <vscale x 4 x bfloat> [[TMP8]], 0
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP7]], 1
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP9]], <vscale x 4 x bfloat> [[TMP10]], 1
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP7]], 2
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP11]], <vscale x 4 x bfloat> [[TMP12]], 2
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP7]], 3
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP13]], <vscale x 4 x bfloat> [[TMP14]], 3
+// CHECK-RV64-NEXT:    [[TMP16:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP7]], 4
+// CHECK-RV64-NEXT:    [[TMP17:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP15]], <vscale x 4 x bfloat> [[TMP16]], 4
+// CHECK-RV64-NEXT:    [[TMP18:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP7]], 5
+// CHECK-RV64-NEXT:    [[TMP19:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP17]], <vscale x 4 x bfloat> [[TMP18]], 5
+// CHECK-RV64-NEXT:    [[TMP20:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP7]], 6
+// CHECK-RV64-NEXT:    [[TMP21:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP19]], <vscale x 4 x bfloat> [[TMP20]], 6
+// CHECK-RV64-NEXT:    [[TMP22:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP7]], 7
+// CHECK-RV64-NEXT:    store i64 [[TMP22]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP21]]
+//
+vbfloat16m1x7_t test_vlseg7e16ff_v_bf16m1x7_tumu(vbool16_t vm, vbfloat16m1x7_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vlseg7e16ff_v_bf16m1x7_tumu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg7e16ff_v_bf16mf4x7_mu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } @llvm.riscv.vlseg7ff.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP7]], 0
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } poison, <vscale x 1 x bfloat> [[TMP8]], 0
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP7]], 1
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP9]], <vscale x 1 x bfloat> [[TMP10]], 1
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP7]], 2
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP11]], <vscale x 1 x bfloat> [[TMP12]], 2
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP7]], 3
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP13]], <vscale x 1 x bfloat> [[TMP14]], 3
+// CHECK-RV64-NEXT:    [[TMP16:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP7]], 4
+// CHECK-RV64-NEXT:    [[TMP17:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP15]], <vscale x 1 x bfloat> [[TMP16]], 4
+// CHECK-RV64-NEXT:    [[TMP18:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP7]], 5
+// CHECK-RV64-NEXT:    [[TMP19:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP17]], <vscale x 1 x bfloat> [[TMP18]], 5
+// CHECK-RV64-NEXT:    [[TMP20:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP7]], 6
+// CHECK-RV64-NEXT:    [[TMP21:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP19]], <vscale x 1 x bfloat> [[TMP20]], 6
+// CHECK-RV64-NEXT:    [[TMP22:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP7]], 7
+// CHECK-RV64-NEXT:    store i64 [[TMP22]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP21]]
+//
+vbfloat16mf4x7_t test_vlseg7e16ff_v_bf16mf4x7_mu(vbool64_t vm, vbfloat16mf4x7_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vlseg7e16ff_v_bf16mf4x7_mu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg7e16ff_v_bf16mf2x7_mu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } @llvm.riscv.vlseg7ff.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP7]], 0
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } poison, <vscale x 2 x bfloat> [[TMP8]], 0
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP7]], 1
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP9]], <vscale x 2 x bfloat> [[TMP10]], 1
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP7]], 2
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP11]], <vscale x 2 x bfloat> [[TMP12]], 2
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP7]], 3
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP13]], <vscale x 2 x bfloat> [[TMP14]], 3
+// CHECK-RV64-NEXT:    [[TMP16:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP7]], 4
+// CHECK-RV64-NEXT:    [[TMP17:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP15]], <vscale x 2 x bfloat> [[TMP16]], 4
+// CHECK-RV64-NEXT:    [[TMP18:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP7]], 5
+// CHECK-RV64-NEXT:    [[TMP19:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP17]], <vscale x 2 x bfloat> [[TMP18]], 5
+// CHECK-RV64-NEXT:    [[TMP20:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP7]], 6
+// CHECK-RV64-NEXT:    [[TMP21:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP19]], <vscale x 2 x bfloat> [[TMP20]], 6
+// CHECK-RV64-NEXT:    [[TMP22:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP7]], 7
+// CHECK-RV64-NEXT:    store i64 [[TMP22]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP21]]
+//
+vbfloat16mf2x7_t test_vlseg7e16ff_v_bf16mf2x7_mu(vbool32_t vm, vbfloat16mf2x7_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vlseg7e16ff_v_bf16mf2x7_mu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg7e16ff_v_bf16m1x7_mu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } @llvm.riscv.vlseg7ff.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP7]], 0
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } poison, <vscale x 4 x bfloat> [[TMP8]], 0
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP7]], 1
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP9]], <vscale x 4 x bfloat> [[TMP10]], 1
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP7]], 2
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP11]], <vscale x 4 x bfloat> [[TMP12]], 2
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP7]], 3
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP13]], <vscale x 4 x bfloat> [[TMP14]], 3
+// CHECK-RV64-NEXT:    [[TMP16:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP7]], 4
+// CHECK-RV64-NEXT:    [[TMP17:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP15]], <vscale x 4 x bfloat> [[TMP16]], 4
+// CHECK-RV64-NEXT:    [[TMP18:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP7]], 5
+// CHECK-RV64-NEXT:    [[TMP19:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP17]], <vscale x 4 x bfloat> [[TMP18]], 5
+// CHECK-RV64-NEXT:    [[TMP20:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP7]], 6
+// CHECK-RV64-NEXT:    [[TMP21:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP19]], <vscale x 4 x bfloat> [[TMP20]], 6
+// CHECK-RV64-NEXT:    [[TMP22:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP7]], 7
+// CHECK-RV64-NEXT:    store i64 [[TMP22]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP21]]
+//
+vbfloat16m1x7_t test_vlseg7e16ff_v_bf16m1x7_mu(vbool16_t vm, vbfloat16m1x7_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vlseg7e16ff_v_bf16m1x7_mu(vm, vd, rs1, new_vl, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vlseg8e16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vlseg8e16.c
new file mode 100644
index 000000000000..b11f79f4e6a4
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vlseg8e16.c
@@ -0,0 +1,225 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg8e16_v_bf16mf4x8_tu(
+// CHECK-RV64-SAME: { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlseg8.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], <vscale x 1 x bfloat> [[TMP7]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP8]]
+//
+vbfloat16mf4x8_t test_vlseg8e16_v_bf16mf4x8_tu(vbfloat16mf4x8_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg8e16_v_bf16mf4x8_tu(vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg8e16_v_bf16mf2x8_tu(
+// CHECK-RV64-SAME: { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlseg8.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], <vscale x 2 x bfloat> [[TMP7]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP8]]
+//
+vbfloat16mf2x8_t test_vlseg8e16_v_bf16mf2x8_tu(vbfloat16mf2x8_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg8e16_v_bf16mf2x8_tu(vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg8e16_v_bf16m1x8_tu(
+// CHECK-RV64-SAME: { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlseg8.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], <vscale x 4 x bfloat> [[TMP7]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP8]]
+//
+vbfloat16m1x8_t test_vlseg8e16_v_bf16m1x8_tu(vbfloat16m1x8_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg8e16_v_bf16m1x8_tu(vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg8e16_v_bf16mf4x8_tum(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlseg8.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], <vscale x 1 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP8]]
+//
+vbfloat16mf4x8_t test_vlseg8e16_v_bf16mf4x8_tum(vbool64_t vm, vbfloat16mf4x8_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg8e16_v_bf16mf4x8_tum(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg8e16_v_bf16mf2x8_tum(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlseg8.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], <vscale x 2 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP8]]
+//
+vbfloat16mf2x8_t test_vlseg8e16_v_bf16mf2x8_tum(vbool32_t vm, vbfloat16mf2x8_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg8e16_v_bf16mf2x8_tum(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg8e16_v_bf16m1x8_tum(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlseg8.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], <vscale x 4 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP8]]
+//
+vbfloat16m1x8_t test_vlseg8e16_v_bf16m1x8_tum(vbool16_t vm, vbfloat16m1x8_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg8e16_v_bf16m1x8_tum(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg8e16_v_bf16mf4x8_tumu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlseg8.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], <vscale x 1 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP8]]
+//
+vbfloat16mf4x8_t test_vlseg8e16_v_bf16mf4x8_tumu(vbool64_t vm, vbfloat16mf4x8_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg8e16_v_bf16mf4x8_tumu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg8e16_v_bf16mf2x8_tumu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlseg8.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], <vscale x 2 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP8]]
+//
+vbfloat16mf2x8_t test_vlseg8e16_v_bf16mf2x8_tumu(vbool32_t vm, vbfloat16mf2x8_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg8e16_v_bf16mf2x8_tumu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg8e16_v_bf16m1x8_tumu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlseg8.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], <vscale x 4 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP8]]
+//
+vbfloat16m1x8_t test_vlseg8e16_v_bf16m1x8_tumu(vbool16_t vm, vbfloat16m1x8_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg8e16_v_bf16m1x8_tumu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg8e16_v_bf16mf4x8_mu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlseg8.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], <vscale x 1 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP8]]
+//
+vbfloat16mf4x8_t test_vlseg8e16_v_bf16mf4x8_mu(vbool64_t vm, vbfloat16mf4x8_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg8e16_v_bf16mf4x8_mu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg8e16_v_bf16mf2x8_mu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlseg8.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], <vscale x 2 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP8]]
+//
+vbfloat16mf2x8_t test_vlseg8e16_v_bf16mf2x8_mu(vbool32_t vm, vbfloat16mf2x8_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg8e16_v_bf16mf2x8_mu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg8e16_v_bf16m1x8_mu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlseg8.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], <vscale x 4 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP8]]
+//
+vbfloat16m1x8_t test_vlseg8e16_v_bf16m1x8_mu(vbool16_t vm, vbfloat16m1x8_t vd, const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg8e16_v_bf16m1x8_mu(vm, vd, rs1, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vlseg8e16ff.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vlseg8e16ff.c
new file mode 100644
index 000000000000..4f5e3868f78c
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vlseg8e16ff.c
@@ -0,0 +1,441 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg8e16ff_v_bf16mf4x8_tu(
+// CHECK-RV64-SAME: { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } @llvm.riscv.vlseg8ff.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], <vscale x 1 x bfloat> [[TMP7]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP8]], 0
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } poison, <vscale x 1 x bfloat> [[TMP9]], 0
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP8]], 1
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP10]], <vscale x 1 x bfloat> [[TMP11]], 1
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP8]], 2
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP12]], <vscale x 1 x bfloat> [[TMP13]], 2
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP8]], 3
+// CHECK-RV64-NEXT:    [[TMP16:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP14]], <vscale x 1 x bfloat> [[TMP15]], 3
+// CHECK-RV64-NEXT:    [[TMP17:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP8]], 4
+// CHECK-RV64-NEXT:    [[TMP18:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP16]], <vscale x 1 x bfloat> [[TMP17]], 4
+// CHECK-RV64-NEXT:    [[TMP19:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP8]], 5
+// CHECK-RV64-NEXT:    [[TMP20:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP18]], <vscale x 1 x bfloat> [[TMP19]], 5
+// CHECK-RV64-NEXT:    [[TMP21:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP8]], 6
+// CHECK-RV64-NEXT:    [[TMP22:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP20]], <vscale x 1 x bfloat> [[TMP21]], 6
+// CHECK-RV64-NEXT:    [[TMP23:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP8]], 7
+// CHECK-RV64-NEXT:    [[TMP24:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP22]], <vscale x 1 x bfloat> [[TMP23]], 7
+// CHECK-RV64-NEXT:    [[TMP25:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP8]], 8
+// CHECK-RV64-NEXT:    store i64 [[TMP25]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP24]]
+//
+vbfloat16mf4x8_t test_vlseg8e16ff_v_bf16mf4x8_tu(vbfloat16mf4x8_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vlseg8e16ff_v_bf16mf4x8_tu(vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg8e16ff_v_bf16mf2x8_tu(
+// CHECK-RV64-SAME: { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } @llvm.riscv.vlseg8ff.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], <vscale x 2 x bfloat> [[TMP7]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP8]], 0
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } poison, <vscale x 2 x bfloat> [[TMP9]], 0
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP8]], 1
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP10]], <vscale x 2 x bfloat> [[TMP11]], 1
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP8]], 2
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP12]], <vscale x 2 x bfloat> [[TMP13]], 2
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP8]], 3
+// CHECK-RV64-NEXT:    [[TMP16:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP14]], <vscale x 2 x bfloat> [[TMP15]], 3
+// CHECK-RV64-NEXT:    [[TMP17:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP8]], 4
+// CHECK-RV64-NEXT:    [[TMP18:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP16]], <vscale x 2 x bfloat> [[TMP17]], 4
+// CHECK-RV64-NEXT:    [[TMP19:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP8]], 5
+// CHECK-RV64-NEXT:    [[TMP20:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP18]], <vscale x 2 x bfloat> [[TMP19]], 5
+// CHECK-RV64-NEXT:    [[TMP21:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP8]], 6
+// CHECK-RV64-NEXT:    [[TMP22:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP20]], <vscale x 2 x bfloat> [[TMP21]], 6
+// CHECK-RV64-NEXT:    [[TMP23:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP8]], 7
+// CHECK-RV64-NEXT:    [[TMP24:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP22]], <vscale x 2 x bfloat> [[TMP23]], 7
+// CHECK-RV64-NEXT:    [[TMP25:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP8]], 8
+// CHECK-RV64-NEXT:    store i64 [[TMP25]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP24]]
+//
+vbfloat16mf2x8_t test_vlseg8e16ff_v_bf16mf2x8_tu(vbfloat16mf2x8_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vlseg8e16ff_v_bf16mf2x8_tu(vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg8e16ff_v_bf16m1x8_tu(
+// CHECK-RV64-SAME: { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } @llvm.riscv.vlseg8ff.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], <vscale x 4 x bfloat> [[TMP7]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP8]], 0
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } poison, <vscale x 4 x bfloat> [[TMP9]], 0
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP8]], 1
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP10]], <vscale x 4 x bfloat> [[TMP11]], 1
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP8]], 2
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP12]], <vscale x 4 x bfloat> [[TMP13]], 2
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP8]], 3
+// CHECK-RV64-NEXT:    [[TMP16:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP14]], <vscale x 4 x bfloat> [[TMP15]], 3
+// CHECK-RV64-NEXT:    [[TMP17:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP8]], 4
+// CHECK-RV64-NEXT:    [[TMP18:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP16]], <vscale x 4 x bfloat> [[TMP17]], 4
+// CHECK-RV64-NEXT:    [[TMP19:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP8]], 5
+// CHECK-RV64-NEXT:    [[TMP20:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP18]], <vscale x 4 x bfloat> [[TMP19]], 5
+// CHECK-RV64-NEXT:    [[TMP21:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP8]], 6
+// CHECK-RV64-NEXT:    [[TMP22:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP20]], <vscale x 4 x bfloat> [[TMP21]], 6
+// CHECK-RV64-NEXT:    [[TMP23:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP8]], 7
+// CHECK-RV64-NEXT:    [[TMP24:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP22]], <vscale x 4 x bfloat> [[TMP23]], 7
+// CHECK-RV64-NEXT:    [[TMP25:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP8]], 8
+// CHECK-RV64-NEXT:    store i64 [[TMP25]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP24]]
+//
+vbfloat16m1x8_t test_vlseg8e16ff_v_bf16m1x8_tu(vbfloat16m1x8_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vlseg8e16ff_v_bf16m1x8_tu(vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg8e16ff_v_bf16mf4x8_tum(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } @llvm.riscv.vlseg8ff.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], <vscale x 1 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP8]], 0
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } poison, <vscale x 1 x bfloat> [[TMP9]], 0
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP8]], 1
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP10]], <vscale x 1 x bfloat> [[TMP11]], 1
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP8]], 2
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP12]], <vscale x 1 x bfloat> [[TMP13]], 2
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP8]], 3
+// CHECK-RV64-NEXT:    [[TMP16:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP14]], <vscale x 1 x bfloat> [[TMP15]], 3
+// CHECK-RV64-NEXT:    [[TMP17:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP8]], 4
+// CHECK-RV64-NEXT:    [[TMP18:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP16]], <vscale x 1 x bfloat> [[TMP17]], 4
+// CHECK-RV64-NEXT:    [[TMP19:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP8]], 5
+// CHECK-RV64-NEXT:    [[TMP20:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP18]], <vscale x 1 x bfloat> [[TMP19]], 5
+// CHECK-RV64-NEXT:    [[TMP21:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP8]], 6
+// CHECK-RV64-NEXT:    [[TMP22:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP20]], <vscale x 1 x bfloat> [[TMP21]], 6
+// CHECK-RV64-NEXT:    [[TMP23:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP8]], 7
+// CHECK-RV64-NEXT:    [[TMP24:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP22]], <vscale x 1 x bfloat> [[TMP23]], 7
+// CHECK-RV64-NEXT:    [[TMP25:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP8]], 8
+// CHECK-RV64-NEXT:    store i64 [[TMP25]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP24]]
+//
+vbfloat16mf4x8_t test_vlseg8e16ff_v_bf16mf4x8_tum(vbool64_t vm, vbfloat16mf4x8_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vlseg8e16ff_v_bf16mf4x8_tum(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg8e16ff_v_bf16mf2x8_tum(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } @llvm.riscv.vlseg8ff.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], <vscale x 2 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP8]], 0
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } poison, <vscale x 2 x bfloat> [[TMP9]], 0
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP8]], 1
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP10]], <vscale x 2 x bfloat> [[TMP11]], 1
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP8]], 2
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP12]], <vscale x 2 x bfloat> [[TMP13]], 2
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP8]], 3
+// CHECK-RV64-NEXT:    [[TMP16:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP14]], <vscale x 2 x bfloat> [[TMP15]], 3
+// CHECK-RV64-NEXT:    [[TMP17:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP8]], 4
+// CHECK-RV64-NEXT:    [[TMP18:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP16]], <vscale x 2 x bfloat> [[TMP17]], 4
+// CHECK-RV64-NEXT:    [[TMP19:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP8]], 5
+// CHECK-RV64-NEXT:    [[TMP20:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP18]], <vscale x 2 x bfloat> [[TMP19]], 5
+// CHECK-RV64-NEXT:    [[TMP21:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP8]], 6
+// CHECK-RV64-NEXT:    [[TMP22:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP20]], <vscale x 2 x bfloat> [[TMP21]], 6
+// CHECK-RV64-NEXT:    [[TMP23:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP8]], 7
+// CHECK-RV64-NEXT:    [[TMP24:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP22]], <vscale x 2 x bfloat> [[TMP23]], 7
+// CHECK-RV64-NEXT:    [[TMP25:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP8]], 8
+// CHECK-RV64-NEXT:    store i64 [[TMP25]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP24]]
+//
+vbfloat16mf2x8_t test_vlseg8e16ff_v_bf16mf2x8_tum(vbool32_t vm, vbfloat16mf2x8_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vlseg8e16ff_v_bf16mf2x8_tum(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg8e16ff_v_bf16m1x8_tum(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } @llvm.riscv.vlseg8ff.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], <vscale x 4 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP8]], 0
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } poison, <vscale x 4 x bfloat> [[TMP9]], 0
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP8]], 1
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP10]], <vscale x 4 x bfloat> [[TMP11]], 1
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP8]], 2
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP12]], <vscale x 4 x bfloat> [[TMP13]], 2
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP8]], 3
+// CHECK-RV64-NEXT:    [[TMP16:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP14]], <vscale x 4 x bfloat> [[TMP15]], 3
+// CHECK-RV64-NEXT:    [[TMP17:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP8]], 4
+// CHECK-RV64-NEXT:    [[TMP18:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP16]], <vscale x 4 x bfloat> [[TMP17]], 4
+// CHECK-RV64-NEXT:    [[TMP19:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP8]], 5
+// CHECK-RV64-NEXT:    [[TMP20:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP18]], <vscale x 4 x bfloat> [[TMP19]], 5
+// CHECK-RV64-NEXT:    [[TMP21:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP8]], 6
+// CHECK-RV64-NEXT:    [[TMP22:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP20]], <vscale x 4 x bfloat> [[TMP21]], 6
+// CHECK-RV64-NEXT:    [[TMP23:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP8]], 7
+// CHECK-RV64-NEXT:    [[TMP24:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP22]], <vscale x 4 x bfloat> [[TMP23]], 7
+// CHECK-RV64-NEXT:    [[TMP25:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP8]], 8
+// CHECK-RV64-NEXT:    store i64 [[TMP25]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP24]]
+//
+vbfloat16m1x8_t test_vlseg8e16ff_v_bf16m1x8_tum(vbool16_t vm, vbfloat16m1x8_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vlseg8e16ff_v_bf16m1x8_tum(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg8e16ff_v_bf16mf4x8_tumu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } @llvm.riscv.vlseg8ff.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], <vscale x 1 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP8]], 0
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } poison, <vscale x 1 x bfloat> [[TMP9]], 0
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP8]], 1
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP10]], <vscale x 1 x bfloat> [[TMP11]], 1
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP8]], 2
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP12]], <vscale x 1 x bfloat> [[TMP13]], 2
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP8]], 3
+// CHECK-RV64-NEXT:    [[TMP16:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP14]], <vscale x 1 x bfloat> [[TMP15]], 3
+// CHECK-RV64-NEXT:    [[TMP17:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP8]], 4
+// CHECK-RV64-NEXT:    [[TMP18:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP16]], <vscale x 1 x bfloat> [[TMP17]], 4
+// CHECK-RV64-NEXT:    [[TMP19:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP8]], 5
+// CHECK-RV64-NEXT:    [[TMP20:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP18]], <vscale x 1 x bfloat> [[TMP19]], 5
+// CHECK-RV64-NEXT:    [[TMP21:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP8]], 6
+// CHECK-RV64-NEXT:    [[TMP22:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP20]], <vscale x 1 x bfloat> [[TMP21]], 6
+// CHECK-RV64-NEXT:    [[TMP23:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP8]], 7
+// CHECK-RV64-NEXT:    [[TMP24:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP22]], <vscale x 1 x bfloat> [[TMP23]], 7
+// CHECK-RV64-NEXT:    [[TMP25:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP8]], 8
+// CHECK-RV64-NEXT:    store i64 [[TMP25]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP24]]
+//
+vbfloat16mf4x8_t test_vlseg8e16ff_v_bf16mf4x8_tumu(vbool64_t vm, vbfloat16mf4x8_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vlseg8e16ff_v_bf16mf4x8_tumu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg8e16ff_v_bf16mf2x8_tumu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } @llvm.riscv.vlseg8ff.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], <vscale x 2 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP8]], 0
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } poison, <vscale x 2 x bfloat> [[TMP9]], 0
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP8]], 1
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP10]], <vscale x 2 x bfloat> [[TMP11]], 1
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP8]], 2
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP12]], <vscale x 2 x bfloat> [[TMP13]], 2
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP8]], 3
+// CHECK-RV64-NEXT:    [[TMP16:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP14]], <vscale x 2 x bfloat> [[TMP15]], 3
+// CHECK-RV64-NEXT:    [[TMP17:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP8]], 4
+// CHECK-RV64-NEXT:    [[TMP18:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP16]], <vscale x 2 x bfloat> [[TMP17]], 4
+// CHECK-RV64-NEXT:    [[TMP19:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP8]], 5
+// CHECK-RV64-NEXT:    [[TMP20:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP18]], <vscale x 2 x bfloat> [[TMP19]], 5
+// CHECK-RV64-NEXT:    [[TMP21:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP8]], 6
+// CHECK-RV64-NEXT:    [[TMP22:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP20]], <vscale x 2 x bfloat> [[TMP21]], 6
+// CHECK-RV64-NEXT:    [[TMP23:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP8]], 7
+// CHECK-RV64-NEXT:    [[TMP24:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP22]], <vscale x 2 x bfloat> [[TMP23]], 7
+// CHECK-RV64-NEXT:    [[TMP25:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP8]], 8
+// CHECK-RV64-NEXT:    store i64 [[TMP25]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP24]]
+//
+vbfloat16mf2x8_t test_vlseg8e16ff_v_bf16mf2x8_tumu(vbool32_t vm, vbfloat16mf2x8_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vlseg8e16ff_v_bf16mf2x8_tumu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg8e16ff_v_bf16m1x8_tumu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } @llvm.riscv.vlseg8ff.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], <vscale x 4 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP8]], 0
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } poison, <vscale x 4 x bfloat> [[TMP9]], 0
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP8]], 1
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP10]], <vscale x 4 x bfloat> [[TMP11]], 1
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP8]], 2
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP12]], <vscale x 4 x bfloat> [[TMP13]], 2
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP8]], 3
+// CHECK-RV64-NEXT:    [[TMP16:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP14]], <vscale x 4 x bfloat> [[TMP15]], 3
+// CHECK-RV64-NEXT:    [[TMP17:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP8]], 4
+// CHECK-RV64-NEXT:    [[TMP18:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP16]], <vscale x 4 x bfloat> [[TMP17]], 4
+// CHECK-RV64-NEXT:    [[TMP19:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP8]], 5
+// CHECK-RV64-NEXT:    [[TMP20:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP18]], <vscale x 4 x bfloat> [[TMP19]], 5
+// CHECK-RV64-NEXT:    [[TMP21:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP8]], 6
+// CHECK-RV64-NEXT:    [[TMP22:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP20]], <vscale x 4 x bfloat> [[TMP21]], 6
+// CHECK-RV64-NEXT:    [[TMP23:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP8]], 7
+// CHECK-RV64-NEXT:    [[TMP24:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP22]], <vscale x 4 x bfloat> [[TMP23]], 7
+// CHECK-RV64-NEXT:    [[TMP25:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP8]], 8
+// CHECK-RV64-NEXT:    store i64 [[TMP25]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP24]]
+//
+vbfloat16m1x8_t test_vlseg8e16ff_v_bf16m1x8_tumu(vbool16_t vm, vbfloat16m1x8_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vlseg8e16ff_v_bf16m1x8_tumu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg8e16ff_v_bf16mf4x8_mu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } @llvm.riscv.vlseg8ff.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], <vscale x 1 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP8]], 0
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } poison, <vscale x 1 x bfloat> [[TMP9]], 0
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP8]], 1
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP10]], <vscale x 1 x bfloat> [[TMP11]], 1
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP8]], 2
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP12]], <vscale x 1 x bfloat> [[TMP13]], 2
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP8]], 3
+// CHECK-RV64-NEXT:    [[TMP16:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP14]], <vscale x 1 x bfloat> [[TMP15]], 3
+// CHECK-RV64-NEXT:    [[TMP17:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP8]], 4
+// CHECK-RV64-NEXT:    [[TMP18:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP16]], <vscale x 1 x bfloat> [[TMP17]], 4
+// CHECK-RV64-NEXT:    [[TMP19:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP8]], 5
+// CHECK-RV64-NEXT:    [[TMP20:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP18]], <vscale x 1 x bfloat> [[TMP19]], 5
+// CHECK-RV64-NEXT:    [[TMP21:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP8]], 6
+// CHECK-RV64-NEXT:    [[TMP22:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP20]], <vscale x 1 x bfloat> [[TMP21]], 6
+// CHECK-RV64-NEXT:    [[TMP23:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP8]], 7
+// CHECK-RV64-NEXT:    [[TMP24:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP22]], <vscale x 1 x bfloat> [[TMP23]], 7
+// CHECK-RV64-NEXT:    [[TMP25:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP8]], 8
+// CHECK-RV64-NEXT:    store i64 [[TMP25]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP24]]
+//
+vbfloat16mf4x8_t test_vlseg8e16ff_v_bf16mf4x8_mu(vbool64_t vm, vbfloat16mf4x8_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vlseg8e16ff_v_bf16mf4x8_mu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg8e16ff_v_bf16mf2x8_mu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } @llvm.riscv.vlseg8ff.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], <vscale x 2 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP8]], 0
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } poison, <vscale x 2 x bfloat> [[TMP9]], 0
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP8]], 1
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP10]], <vscale x 2 x bfloat> [[TMP11]], 1
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP8]], 2
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP12]], <vscale x 2 x bfloat> [[TMP13]], 2
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP8]], 3
+// CHECK-RV64-NEXT:    [[TMP16:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP14]], <vscale x 2 x bfloat> [[TMP15]], 3
+// CHECK-RV64-NEXT:    [[TMP17:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP8]], 4
+// CHECK-RV64-NEXT:    [[TMP18:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP16]], <vscale x 2 x bfloat> [[TMP17]], 4
+// CHECK-RV64-NEXT:    [[TMP19:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP8]], 5
+// CHECK-RV64-NEXT:    [[TMP20:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP18]], <vscale x 2 x bfloat> [[TMP19]], 5
+// CHECK-RV64-NEXT:    [[TMP21:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP8]], 6
+// CHECK-RV64-NEXT:    [[TMP22:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP20]], <vscale x 2 x bfloat> [[TMP21]], 6
+// CHECK-RV64-NEXT:    [[TMP23:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP8]], 7
+// CHECK-RV64-NEXT:    [[TMP24:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP22]], <vscale x 2 x bfloat> [[TMP23]], 7
+// CHECK-RV64-NEXT:    [[TMP25:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP8]], 8
+// CHECK-RV64-NEXT:    store i64 [[TMP25]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP24]]
+//
+vbfloat16mf2x8_t test_vlseg8e16ff_v_bf16mf2x8_mu(vbool32_t vm, vbfloat16mf2x8_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vlseg8e16ff_v_bf16mf2x8_mu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg8e16ff_v_bf16m1x8_mu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } @llvm.riscv.vlseg8ff.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], <vscale x 4 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP8]], 0
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } poison, <vscale x 4 x bfloat> [[TMP9]], 0
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP8]], 1
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP10]], <vscale x 4 x bfloat> [[TMP11]], 1
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP8]], 2
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP12]], <vscale x 4 x bfloat> [[TMP13]], 2
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP8]], 3
+// CHECK-RV64-NEXT:    [[TMP16:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP14]], <vscale x 4 x bfloat> [[TMP15]], 3
+// CHECK-RV64-NEXT:    [[TMP17:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP8]], 4
+// CHECK-RV64-NEXT:    [[TMP18:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP16]], <vscale x 4 x bfloat> [[TMP17]], 4
+// CHECK-RV64-NEXT:    [[TMP19:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP8]], 5
+// CHECK-RV64-NEXT:    [[TMP20:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP18]], <vscale x 4 x bfloat> [[TMP19]], 5
+// CHECK-RV64-NEXT:    [[TMP21:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP8]], 6
+// CHECK-RV64-NEXT:    [[TMP22:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP20]], <vscale x 4 x bfloat> [[TMP21]], 6
+// CHECK-RV64-NEXT:    [[TMP23:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP8]], 7
+// CHECK-RV64-NEXT:    [[TMP24:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP22]], <vscale x 4 x bfloat> [[TMP23]], 7
+// CHECK-RV64-NEXT:    [[TMP25:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP8]], 8
+// CHECK-RV64-NEXT:    store i64 [[TMP25]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP24]]
+//
+vbfloat16m1x8_t test_vlseg8e16ff_v_bf16m1x8_mu(vbool16_t vm, vbfloat16m1x8_t vd, const __bf16 *rs1, size_t *new_vl, size_t vl) {
+  return __riscv_vlseg8e16ff_v_bf16m1x8_mu(vm, vd, rs1, new_vl, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vlsseg2e16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vlsseg2e16.c
new file mode 100644
index 000000000000..9ab83506e3a4
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vlsseg2e16.c
@@ -0,0 +1,249 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlsseg2e16_v_bf16mf4x2_tu(
+// CHECK-RV64-SAME: { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlsseg2.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP2]]
+//
+vbfloat16mf4x2_t test_vlsseg2e16_v_bf16mf4x2_tu(vbfloat16mf4x2_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg2e16_v_bf16mf4x2_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlsseg2e16_v_bf16mf2x2_tu(
+// CHECK-RV64-SAME: { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlsseg2.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP2]]
+//
+vbfloat16mf2x2_t test_vlsseg2e16_v_bf16mf2x2_tu(vbfloat16mf2x2_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg2e16_v_bf16mf2x2_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlsseg2e16_v_bf16m1x2_tu(
+// CHECK-RV64-SAME: { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlsseg2.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP2]]
+//
+vbfloat16m1x2_t test_vlsseg2e16_v_bf16m1x2_tu(vbfloat16m1x2_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg2e16_v_bf16m1x2_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vlsseg2e16_v_bf16m2x2_tu(
+// CHECK-RV64-SAME: { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vlsseg2.nxv8bf16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP2]]
+//
+vbfloat16m2x2_t test_vlsseg2e16_v_bf16m2x2_tu(vbfloat16m2x2_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg2e16_v_bf16m2x2_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @test_vlsseg2e16_v_bf16m4x2_tu(
+// CHECK-RV64-SAME: { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @llvm.riscv.vlsseg2.nxv16bf16.i64(<vscale x 16 x bfloat> [[TMP0]], <vscale x 16 x bfloat> [[TMP1]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[TMP2]]
+//
+vbfloat16m4x2_t test_vlsseg2e16_v_bf16m4x2_tu(vbfloat16m4x2_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg2e16_v_bf16m4x2_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlsseg2e16_v_bf16mf4x2_tum(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlsseg2.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], ptr [[RS1]], i64 [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP2]]
+//
+vbfloat16mf4x2_t test_vlsseg2e16_v_bf16mf4x2_tum(vbool64_t vm, vbfloat16mf4x2_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg2e16_v_bf16mf4x2_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlsseg2e16_v_bf16mf2x2_tum(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlsseg2.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], ptr [[RS1]], i64 [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP2]]
+//
+vbfloat16mf2x2_t test_vlsseg2e16_v_bf16mf2x2_tum(vbool32_t vm, vbfloat16mf2x2_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg2e16_v_bf16mf2x2_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlsseg2e16_v_bf16m1x2_tum(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlsseg2.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], ptr [[RS1]], i64 [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP2]]
+//
+vbfloat16m1x2_t test_vlsseg2e16_v_bf16m1x2_tum(vbool16_t vm, vbfloat16m1x2_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg2e16_v_bf16m1x2_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vlsseg2e16_v_bf16m2x2_tum(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vlsseg2.mask.nxv8bf16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], ptr [[RS1]], i64 [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP2]]
+//
+vbfloat16m2x2_t test_vlsseg2e16_v_bf16m2x2_tum(vbool8_t vm, vbfloat16m2x2_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg2e16_v_bf16m2x2_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @test_vlsseg2e16_v_bf16m4x2_tum(
+// CHECK-RV64-SAME: <vscale x 16 x i1> [[VM:%.*]], { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @llvm.riscv.vlsseg2.mask.nxv16bf16.i64(<vscale x 16 x bfloat> [[TMP0]], <vscale x 16 x bfloat> [[TMP1]], ptr [[RS1]], i64 [[RS2]], <vscale x 16 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[TMP2]]
+//
+vbfloat16m4x2_t test_vlsseg2e16_v_bf16m4x2_tum(vbool4_t vm, vbfloat16m4x2_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg2e16_v_bf16m4x2_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlsseg2e16_v_bf16mf4x2_tumu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlsseg2.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], ptr [[RS1]], i64 [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP2]]
+//
+vbfloat16mf4x2_t test_vlsseg2e16_v_bf16mf4x2_tumu(vbool64_t vm, vbfloat16mf4x2_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg2e16_v_bf16mf4x2_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlsseg2e16_v_bf16mf2x2_tumu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlsseg2.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], ptr [[RS1]], i64 [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP2]]
+//
+vbfloat16mf2x2_t test_vlsseg2e16_v_bf16mf2x2_tumu(vbool32_t vm, vbfloat16mf2x2_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg2e16_v_bf16mf2x2_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlsseg2e16_v_bf16m1x2_tumu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlsseg2.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], ptr [[RS1]], i64 [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP2]]
+//
+vbfloat16m1x2_t test_vlsseg2e16_v_bf16m1x2_tumu(vbool16_t vm, vbfloat16m1x2_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg2e16_v_bf16m1x2_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vlsseg2e16_v_bf16m2x2_tumu(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vlsseg2.mask.nxv8bf16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], ptr [[RS1]], i64 [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP2]]
+//
+vbfloat16m2x2_t test_vlsseg2e16_v_bf16m2x2_tumu(vbool8_t vm, vbfloat16m2x2_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg2e16_v_bf16m2x2_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @test_vlsseg2e16_v_bf16m4x2_tumu(
+// CHECK-RV64-SAME: <vscale x 16 x i1> [[VM:%.*]], { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @llvm.riscv.vlsseg2.mask.nxv16bf16.i64(<vscale x 16 x bfloat> [[TMP0]], <vscale x 16 x bfloat> [[TMP1]], ptr [[RS1]], i64 [[RS2]], <vscale x 16 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[TMP2]]
+//
+vbfloat16m4x2_t test_vlsseg2e16_v_bf16m4x2_tumu(vbool4_t vm, vbfloat16m4x2_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg2e16_v_bf16m4x2_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlsseg2e16_v_bf16mf4x2_mu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlsseg2.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], ptr [[RS1]], i64 [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP2]]
+//
+vbfloat16mf4x2_t test_vlsseg2e16_v_bf16mf4x2_mu(vbool64_t vm, vbfloat16mf4x2_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg2e16_v_bf16mf4x2_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlsseg2e16_v_bf16mf2x2_mu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlsseg2.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], ptr [[RS1]], i64 [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP2]]
+//
+vbfloat16mf2x2_t test_vlsseg2e16_v_bf16mf2x2_mu(vbool32_t vm, vbfloat16mf2x2_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg2e16_v_bf16mf2x2_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlsseg2e16_v_bf16m1x2_mu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlsseg2.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], ptr [[RS1]], i64 [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP2]]
+//
+vbfloat16m1x2_t test_vlsseg2e16_v_bf16m1x2_mu(vbool16_t vm, vbfloat16m1x2_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg2e16_v_bf16m1x2_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vlsseg2e16_v_bf16m2x2_mu(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vlsseg2.mask.nxv8bf16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], ptr [[RS1]], i64 [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP2]]
+//
+vbfloat16m2x2_t test_vlsseg2e16_v_bf16m2x2_mu(vbool8_t vm, vbfloat16m2x2_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg2e16_v_bf16m2x2_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @test_vlsseg2e16_v_bf16m4x2_mu(
+// CHECK-RV64-SAME: <vscale x 16 x i1> [[VM:%.*]], { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @llvm.riscv.vlsseg2.mask.nxv16bf16.i64(<vscale x 16 x bfloat> [[TMP0]], <vscale x 16 x bfloat> [[TMP1]], ptr [[RS1]], i64 [[RS2]], <vscale x 16 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[TMP2]]
+//
+vbfloat16m4x2_t test_vlsseg2e16_v_bf16m4x2_mu(vbool4_t vm, vbfloat16m4x2_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg2e16_v_bf16m4x2_mu(vm, vd, rs1, rs2, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vlsseg3e16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vlsseg3e16.c
new file mode 100644
index 000000000000..8e02291afd1c
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vlsseg3e16.c
@@ -0,0 +1,217 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlsseg3e16_v_bf16mf4x3_tu(
+// CHECK-RV64-SAME: { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlsseg3.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP3]]
+//
+vbfloat16mf4x3_t test_vlsseg3e16_v_bf16mf4x3_tu(vbfloat16mf4x3_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg3e16_v_bf16mf4x3_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlsseg3e16_v_bf16mf2x3_tu(
+// CHECK-RV64-SAME: { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlsseg3.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP3]]
+//
+vbfloat16mf2x3_t test_vlsseg3e16_v_bf16mf2x3_tu(vbfloat16mf2x3_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg3e16_v_bf16mf2x3_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlsseg3e16_v_bf16m1x3_tu(
+// CHECK-RV64-SAME: { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlsseg3.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP3]]
+//
+vbfloat16m1x3_t test_vlsseg3e16_v_bf16m1x3_tu(vbfloat16m1x3_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg3e16_v_bf16m1x3_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vlsseg3e16_v_bf16m2x3_tu(
+// CHECK-RV64-SAME: { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vlsseg3.nxv8bf16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP3]]
+//
+vbfloat16m2x3_t test_vlsseg3e16_v_bf16m2x3_tu(vbfloat16m2x3_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg3e16_v_bf16m2x3_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlsseg3e16_v_bf16mf4x3_tum(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlsseg3.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], ptr [[RS1]], i64 [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP3]]
+//
+vbfloat16mf4x3_t test_vlsseg3e16_v_bf16mf4x3_tum(vbool64_t vm, vbfloat16mf4x3_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg3e16_v_bf16mf4x3_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlsseg3e16_v_bf16mf2x3_tum(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlsseg3.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], ptr [[RS1]], i64 [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP3]]
+//
+vbfloat16mf2x3_t test_vlsseg3e16_v_bf16mf2x3_tum(vbool32_t vm, vbfloat16mf2x3_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg3e16_v_bf16mf2x3_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlsseg3e16_v_bf16m1x3_tum(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlsseg3.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], ptr [[RS1]], i64 [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP3]]
+//
+vbfloat16m1x3_t test_vlsseg3e16_v_bf16m1x3_tum(vbool16_t vm, vbfloat16m1x3_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg3e16_v_bf16m1x3_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vlsseg3e16_v_bf16m2x3_tum(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vlsseg3.mask.nxv8bf16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], ptr [[RS1]], i64 [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP3]]
+//
+vbfloat16m2x3_t test_vlsseg3e16_v_bf16m2x3_tum(vbool8_t vm, vbfloat16m2x3_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg3e16_v_bf16m2x3_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlsseg3e16_v_bf16mf4x3_tumu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlsseg3.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], ptr [[RS1]], i64 [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP3]]
+//
+vbfloat16mf4x3_t test_vlsseg3e16_v_bf16mf4x3_tumu(vbool64_t vm, vbfloat16mf4x3_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg3e16_v_bf16mf4x3_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlsseg3e16_v_bf16mf2x3_tumu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlsseg3.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], ptr [[RS1]], i64 [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP3]]
+//
+vbfloat16mf2x3_t test_vlsseg3e16_v_bf16mf2x3_tumu(vbool32_t vm, vbfloat16mf2x3_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg3e16_v_bf16mf2x3_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlsseg3e16_v_bf16m1x3_tumu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlsseg3.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], ptr [[RS1]], i64 [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP3]]
+//
+vbfloat16m1x3_t test_vlsseg3e16_v_bf16m1x3_tumu(vbool16_t vm, vbfloat16m1x3_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg3e16_v_bf16m1x3_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vlsseg3e16_v_bf16m2x3_tumu(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vlsseg3.mask.nxv8bf16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], ptr [[RS1]], i64 [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP3]]
+//
+vbfloat16m2x3_t test_vlsseg3e16_v_bf16m2x3_tumu(vbool8_t vm, vbfloat16m2x3_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg3e16_v_bf16m2x3_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlsseg3e16_v_bf16mf4x3_mu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlsseg3.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], ptr [[RS1]], i64 [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP3]]
+//
+vbfloat16mf4x3_t test_vlsseg3e16_v_bf16mf4x3_mu(vbool64_t vm, vbfloat16mf4x3_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg3e16_v_bf16mf4x3_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlsseg3e16_v_bf16mf2x3_mu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlsseg3.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], ptr [[RS1]], i64 [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP3]]
+//
+vbfloat16mf2x3_t test_vlsseg3e16_v_bf16mf2x3_mu(vbool32_t vm, vbfloat16mf2x3_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg3e16_v_bf16mf2x3_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlsseg3e16_v_bf16m1x3_mu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlsseg3.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], ptr [[RS1]], i64 [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP3]]
+//
+vbfloat16m1x3_t test_vlsseg3e16_v_bf16m1x3_mu(vbool16_t vm, vbfloat16m1x3_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg3e16_v_bf16m1x3_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vlsseg3e16_v_bf16m2x3_mu(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vlsseg3.mask.nxv8bf16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], ptr [[RS1]], i64 [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP3]]
+//
+vbfloat16m2x3_t test_vlsseg3e16_v_bf16m2x3_mu(vbool8_t vm, vbfloat16m2x3_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg3e16_v_bf16m2x3_mu(vm, vd, rs1, rs2, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vlsseg4e16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vlsseg4e16.c
new file mode 100644
index 000000000000..c9c57d5156e2
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vlsseg4e16.c
@@ -0,0 +1,233 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlsseg4e16_v_bf16mf4x4_tu(
+// CHECK-RV64-SAME: { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlsseg4.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP4]]
+//
+vbfloat16mf4x4_t test_vlsseg4e16_v_bf16mf4x4_tu(vbfloat16mf4x4_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg4e16_v_bf16mf4x4_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlsseg4e16_v_bf16mf2x4_tu(
+// CHECK-RV64-SAME: { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlsseg4.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP4]]
+//
+vbfloat16mf2x4_t test_vlsseg4e16_v_bf16mf2x4_tu(vbfloat16mf2x4_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg4e16_v_bf16mf2x4_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlsseg4e16_v_bf16m1x4_tu(
+// CHECK-RV64-SAME: { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlsseg4.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP4]]
+//
+vbfloat16m1x4_t test_vlsseg4e16_v_bf16m1x4_tu(vbfloat16m1x4_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg4e16_v_bf16m1x4_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vlsseg4e16_v_bf16m2x4_tu(
+// CHECK-RV64-SAME: { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vlsseg4.nxv8bf16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], <vscale x 8 x bfloat> [[TMP3]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP4]]
+//
+vbfloat16m2x4_t test_vlsseg4e16_v_bf16m2x4_tu(vbfloat16m2x4_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg4e16_v_bf16m2x4_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlsseg4e16_v_bf16mf4x4_tum(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlsseg4.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], ptr [[RS1]], i64 [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP4]]
+//
+vbfloat16mf4x4_t test_vlsseg4e16_v_bf16mf4x4_tum(vbool64_t vm, vbfloat16mf4x4_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg4e16_v_bf16mf4x4_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlsseg4e16_v_bf16mf2x4_tum(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlsseg4.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], ptr [[RS1]], i64 [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP4]]
+//
+vbfloat16mf2x4_t test_vlsseg4e16_v_bf16mf2x4_tum(vbool32_t vm, vbfloat16mf2x4_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg4e16_v_bf16mf2x4_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlsseg4e16_v_bf16m1x4_tum(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlsseg4.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], ptr [[RS1]], i64 [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP4]]
+//
+vbfloat16m1x4_t test_vlsseg4e16_v_bf16m1x4_tum(vbool16_t vm, vbfloat16m1x4_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg4e16_v_bf16m1x4_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vlsseg4e16_v_bf16m2x4_tum(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vlsseg4.mask.nxv8bf16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], <vscale x 8 x bfloat> [[TMP3]], ptr [[RS1]], i64 [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP4]]
+//
+vbfloat16m2x4_t test_vlsseg4e16_v_bf16m2x4_tum(vbool8_t vm, vbfloat16m2x4_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg4e16_v_bf16m2x4_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlsseg4e16_v_bf16mf4x4_tumu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlsseg4.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], ptr [[RS1]], i64 [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP4]]
+//
+vbfloat16mf4x4_t test_vlsseg4e16_v_bf16mf4x4_tumu(vbool64_t vm, vbfloat16mf4x4_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg4e16_v_bf16mf4x4_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlsseg4e16_v_bf16mf2x4_tumu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlsseg4.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], ptr [[RS1]], i64 [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP4]]
+//
+vbfloat16mf2x4_t test_vlsseg4e16_v_bf16mf2x4_tumu(vbool32_t vm, vbfloat16mf2x4_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg4e16_v_bf16mf2x4_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlsseg4e16_v_bf16m1x4_tumu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlsseg4.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], ptr [[RS1]], i64 [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP4]]
+//
+vbfloat16m1x4_t test_vlsseg4e16_v_bf16m1x4_tumu(vbool16_t vm, vbfloat16m1x4_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg4e16_v_bf16m1x4_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vlsseg4e16_v_bf16m2x4_tumu(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vlsseg4.mask.nxv8bf16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], <vscale x 8 x bfloat> [[TMP3]], ptr [[RS1]], i64 [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP4]]
+//
+vbfloat16m2x4_t test_vlsseg4e16_v_bf16m2x4_tumu(vbool8_t vm, vbfloat16m2x4_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg4e16_v_bf16m2x4_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlsseg4e16_v_bf16mf4x4_mu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlsseg4.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], ptr [[RS1]], i64 [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP4]]
+//
+vbfloat16mf4x4_t test_vlsseg4e16_v_bf16mf4x4_mu(vbool64_t vm, vbfloat16mf4x4_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg4e16_v_bf16mf4x4_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlsseg4e16_v_bf16mf2x4_mu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlsseg4.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], ptr [[RS1]], i64 [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP4]]
+//
+vbfloat16mf2x4_t test_vlsseg4e16_v_bf16mf2x4_mu(vbool32_t vm, vbfloat16mf2x4_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg4e16_v_bf16mf2x4_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlsseg4e16_v_bf16m1x4_mu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlsseg4.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], ptr [[RS1]], i64 [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP4]]
+//
+vbfloat16m1x4_t test_vlsseg4e16_v_bf16m1x4_mu(vbool16_t vm, vbfloat16m1x4_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg4e16_v_bf16m1x4_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vlsseg4e16_v_bf16m2x4_mu(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vlsseg4.mask.nxv8bf16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], <vscale x 8 x bfloat> [[TMP3]], ptr [[RS1]], i64 [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP4]]
+//
+vbfloat16m2x4_t test_vlsseg4e16_v_bf16m2x4_mu(vbool8_t vm, vbfloat16m2x4_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg4e16_v_bf16m2x4_mu(vm, vd, rs1, rs2, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vlsseg5e16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vlsseg5e16.c
new file mode 100644
index 000000000000..2f4f7e6f0f79
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vlsseg5e16.c
@@ -0,0 +1,189 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlsseg5e16_v_bf16mf4x5_tu(
+// CHECK-RV64-SAME: { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlsseg5.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP5]]
+//
+vbfloat16mf4x5_t test_vlsseg5e16_v_bf16mf4x5_tu(vbfloat16mf4x5_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg5e16_v_bf16mf4x5_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlsseg5e16_v_bf16mf2x5_tu(
+// CHECK-RV64-SAME: { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlsseg5.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP5]]
+//
+vbfloat16mf2x5_t test_vlsseg5e16_v_bf16mf2x5_tu(vbfloat16mf2x5_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg5e16_v_bf16mf2x5_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlsseg5e16_v_bf16m1x5_tu(
+// CHECK-RV64-SAME: { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlsseg5.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP5]]
+//
+vbfloat16m1x5_t test_vlsseg5e16_v_bf16m1x5_tu(vbfloat16m1x5_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg5e16_v_bf16m1x5_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlsseg5e16_v_bf16mf4x5_tum(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlsseg5.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], ptr [[RS1]], i64 [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP5]]
+//
+vbfloat16mf4x5_t test_vlsseg5e16_v_bf16mf4x5_tum(vbool64_t vm, vbfloat16mf4x5_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg5e16_v_bf16mf4x5_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlsseg5e16_v_bf16mf2x5_tum(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlsseg5.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], ptr [[RS1]], i64 [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP5]]
+//
+vbfloat16mf2x5_t test_vlsseg5e16_v_bf16mf2x5_tum(vbool32_t vm, vbfloat16mf2x5_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg5e16_v_bf16mf2x5_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlsseg5e16_v_bf16m1x5_tum(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlsseg5.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], ptr [[RS1]], i64 [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP5]]
+//
+vbfloat16m1x5_t test_vlsseg5e16_v_bf16m1x5_tum(vbool16_t vm, vbfloat16m1x5_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg5e16_v_bf16m1x5_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlsseg5e16_v_bf16mf4x5_tumu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlsseg5.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], ptr [[RS1]], i64 [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP5]]
+//
+vbfloat16mf4x5_t test_vlsseg5e16_v_bf16mf4x5_tumu(vbool64_t vm, vbfloat16mf4x5_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg5e16_v_bf16mf4x5_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlsseg5e16_v_bf16mf2x5_tumu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlsseg5.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], ptr [[RS1]], i64 [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP5]]
+//
+vbfloat16mf2x5_t test_vlsseg5e16_v_bf16mf2x5_tumu(vbool32_t vm, vbfloat16mf2x5_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg5e16_v_bf16mf2x5_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlsseg5e16_v_bf16m1x5_tumu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlsseg5.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], ptr [[RS1]], i64 [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP5]]
+//
+vbfloat16m1x5_t test_vlsseg5e16_v_bf16m1x5_tumu(vbool16_t vm, vbfloat16m1x5_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg5e16_v_bf16m1x5_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlsseg5e16_v_bf16mf4x5_mu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlsseg5.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], ptr [[RS1]], i64 [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP5]]
+//
+vbfloat16mf4x5_t test_vlsseg5e16_v_bf16mf4x5_mu(vbool64_t vm, vbfloat16mf4x5_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg5e16_v_bf16mf4x5_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlsseg5e16_v_bf16mf2x5_mu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlsseg5.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], ptr [[RS1]], i64 [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP5]]
+//
+vbfloat16mf2x5_t test_vlsseg5e16_v_bf16mf2x5_mu(vbool32_t vm, vbfloat16mf2x5_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg5e16_v_bf16mf2x5_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlsseg5e16_v_bf16m1x5_mu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlsseg5.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], ptr [[RS1]], i64 [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP5]]
+//
+vbfloat16m1x5_t test_vlsseg5e16_v_bf16m1x5_mu(vbool16_t vm, vbfloat16m1x5_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg5e16_v_bf16m1x5_mu(vm, vd, rs1, rs2, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vlsseg6e16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vlsseg6e16.c
new file mode 100644
index 000000000000..5dd8adc57c69
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vlsseg6e16.c
@@ -0,0 +1,201 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlsseg6e16_v_bf16mf4x6_tu(
+// CHECK-RV64-SAME: { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlsseg6.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP6]]
+//
+vbfloat16mf4x6_t test_vlsseg6e16_v_bf16mf4x6_tu(vbfloat16mf4x6_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg6e16_v_bf16mf4x6_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlsseg6e16_v_bf16mf2x6_tu(
+// CHECK-RV64-SAME: { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlsseg6.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP6]]
+//
+vbfloat16mf2x6_t test_vlsseg6e16_v_bf16mf2x6_tu(vbfloat16mf2x6_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg6e16_v_bf16mf2x6_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlsseg6e16_v_bf16m1x6_tu(
+// CHECK-RV64-SAME: { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlsseg6.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP6]]
+//
+vbfloat16m1x6_t test_vlsseg6e16_v_bf16m1x6_tu(vbfloat16m1x6_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg6e16_v_bf16m1x6_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlsseg6e16_v_bf16mf4x6_tum(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlsseg6.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], ptr [[RS1]], i64 [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP6]]
+//
+vbfloat16mf4x6_t test_vlsseg6e16_v_bf16mf4x6_tum(vbool64_t vm, vbfloat16mf4x6_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg6e16_v_bf16mf4x6_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlsseg6e16_v_bf16mf2x6_tum(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlsseg6.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], ptr [[RS1]], i64 [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP6]]
+//
+vbfloat16mf2x6_t test_vlsseg6e16_v_bf16mf2x6_tum(vbool32_t vm, vbfloat16mf2x6_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg6e16_v_bf16mf2x6_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlsseg6e16_v_bf16m1x6_tum(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlsseg6.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], ptr [[RS1]], i64 [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP6]]
+//
+vbfloat16m1x6_t test_vlsseg6e16_v_bf16m1x6_tum(vbool16_t vm, vbfloat16m1x6_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg6e16_v_bf16m1x6_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlsseg6e16_v_bf16mf4x6_tumu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlsseg6.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], ptr [[RS1]], i64 [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP6]]
+//
+vbfloat16mf4x6_t test_vlsseg6e16_v_bf16mf4x6_tumu(vbool64_t vm, vbfloat16mf4x6_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg6e16_v_bf16mf4x6_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlsseg6e16_v_bf16mf2x6_tumu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlsseg6.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], ptr [[RS1]], i64 [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP6]]
+//
+vbfloat16mf2x6_t test_vlsseg6e16_v_bf16mf2x6_tumu(vbool32_t vm, vbfloat16mf2x6_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg6e16_v_bf16mf2x6_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlsseg6e16_v_bf16m1x6_tumu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlsseg6.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], ptr [[RS1]], i64 [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP6]]
+//
+vbfloat16m1x6_t test_vlsseg6e16_v_bf16m1x6_tumu(vbool16_t vm, vbfloat16m1x6_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg6e16_v_bf16m1x6_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlsseg6e16_v_bf16mf4x6_mu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlsseg6.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], ptr [[RS1]], i64 [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP6]]
+//
+vbfloat16mf4x6_t test_vlsseg6e16_v_bf16mf4x6_mu(vbool64_t vm, vbfloat16mf4x6_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg6e16_v_bf16mf4x6_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlsseg6e16_v_bf16mf2x6_mu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlsseg6.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], ptr [[RS1]], i64 [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP6]]
+//
+vbfloat16mf2x6_t test_vlsseg6e16_v_bf16mf2x6_mu(vbool32_t vm, vbfloat16mf2x6_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg6e16_v_bf16mf2x6_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlsseg6e16_v_bf16m1x6_mu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlsseg6.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], ptr [[RS1]], i64 [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP6]]
+//
+vbfloat16m1x6_t test_vlsseg6e16_v_bf16m1x6_mu(vbool16_t vm, vbfloat16m1x6_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg6e16_v_bf16m1x6_mu(vm, vd, rs1, rs2, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vlsseg7e16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vlsseg7e16.c
new file mode 100644
index 000000000000..dd6acd4906cb
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vlsseg7e16.c
@@ -0,0 +1,213 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlsseg7e16_v_bf16mf4x7_tu(
+// CHECK-RV64-SAME: { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlsseg7.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP7]]
+//
+vbfloat16mf4x7_t test_vlsseg7e16_v_bf16mf4x7_tu(vbfloat16mf4x7_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg7e16_v_bf16mf4x7_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlsseg7e16_v_bf16mf2x7_tu(
+// CHECK-RV64-SAME: { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlsseg7.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP7]]
+//
+vbfloat16mf2x7_t test_vlsseg7e16_v_bf16mf2x7_tu(vbfloat16mf2x7_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg7e16_v_bf16mf2x7_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlsseg7e16_v_bf16m1x7_tu(
+// CHECK-RV64-SAME: { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlsseg7.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP7]]
+//
+vbfloat16m1x7_t test_vlsseg7e16_v_bf16m1x7_tu(vbfloat16m1x7_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg7e16_v_bf16m1x7_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlsseg7e16_v_bf16mf4x7_tum(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlsseg7.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], ptr [[RS1]], i64 [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP7]]
+//
+vbfloat16mf4x7_t test_vlsseg7e16_v_bf16mf4x7_tum(vbool64_t vm, vbfloat16mf4x7_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg7e16_v_bf16mf4x7_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlsseg7e16_v_bf16mf2x7_tum(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlsseg7.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], ptr [[RS1]], i64 [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP7]]
+//
+vbfloat16mf2x7_t test_vlsseg7e16_v_bf16mf2x7_tum(vbool32_t vm, vbfloat16mf2x7_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg7e16_v_bf16mf2x7_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlsseg7e16_v_bf16m1x7_tum(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlsseg7.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], ptr [[RS1]], i64 [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP7]]
+//
+vbfloat16m1x7_t test_vlsseg7e16_v_bf16m1x7_tum(vbool16_t vm, vbfloat16m1x7_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg7e16_v_bf16m1x7_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlsseg7e16_v_bf16mf4x7_tumu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlsseg7.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], ptr [[RS1]], i64 [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP7]]
+//
+vbfloat16mf4x7_t test_vlsseg7e16_v_bf16mf4x7_tumu(vbool64_t vm, vbfloat16mf4x7_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg7e16_v_bf16mf4x7_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlsseg7e16_v_bf16mf2x7_tumu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlsseg7.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], ptr [[RS1]], i64 [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP7]]
+//
+vbfloat16mf2x7_t test_vlsseg7e16_v_bf16mf2x7_tumu(vbool32_t vm, vbfloat16mf2x7_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg7e16_v_bf16mf2x7_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlsseg7e16_v_bf16m1x7_tumu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlsseg7.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], ptr [[RS1]], i64 [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP7]]
+//
+vbfloat16m1x7_t test_vlsseg7e16_v_bf16m1x7_tumu(vbool16_t vm, vbfloat16m1x7_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg7e16_v_bf16m1x7_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlsseg7e16_v_bf16mf4x7_mu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlsseg7.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], ptr [[RS1]], i64 [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP7]]
+//
+vbfloat16mf4x7_t test_vlsseg7e16_v_bf16mf4x7_mu(vbool64_t vm, vbfloat16mf4x7_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg7e16_v_bf16mf4x7_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlsseg7e16_v_bf16mf2x7_mu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlsseg7.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], ptr [[RS1]], i64 [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP7]]
+//
+vbfloat16mf2x7_t test_vlsseg7e16_v_bf16mf2x7_mu(vbool32_t vm, vbfloat16mf2x7_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg7e16_v_bf16mf2x7_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlsseg7e16_v_bf16m1x7_mu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlsseg7.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], ptr [[RS1]], i64 [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP7]]
+//
+vbfloat16m1x7_t test_vlsseg7e16_v_bf16m1x7_mu(vbool16_t vm, vbfloat16m1x7_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg7e16_v_bf16m1x7_mu(vm, vd, rs1, rs2, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vlsseg8e16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vlsseg8e16.c
new file mode 100644
index 000000000000..9973765dc654
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vlsseg8e16.c
@@ -0,0 +1,225 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlsseg8e16_v_bf16mf4x8_tu(
+// CHECK-RV64-SAME: { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlsseg8.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], <vscale x 1 x bfloat> [[TMP7]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP8]]
+//
+vbfloat16mf4x8_t test_vlsseg8e16_v_bf16mf4x8_tu(vbfloat16mf4x8_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg8e16_v_bf16mf4x8_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlsseg8e16_v_bf16mf2x8_tu(
+// CHECK-RV64-SAME: { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlsseg8.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], <vscale x 2 x bfloat> [[TMP7]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP8]]
+//
+vbfloat16mf2x8_t test_vlsseg8e16_v_bf16mf2x8_tu(vbfloat16mf2x8_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg8e16_v_bf16mf2x8_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlsseg8e16_v_bf16m1x8_tu(
+// CHECK-RV64-SAME: { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlsseg8.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], <vscale x 4 x bfloat> [[TMP7]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP8]]
+//
+vbfloat16m1x8_t test_vlsseg8e16_v_bf16m1x8_tu(vbfloat16m1x8_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg8e16_v_bf16m1x8_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlsseg8e16_v_bf16mf4x8_tum(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlsseg8.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], <vscale x 1 x bfloat> [[TMP7]], ptr [[RS1]], i64 [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP8]]
+//
+vbfloat16mf4x8_t test_vlsseg8e16_v_bf16mf4x8_tum(vbool64_t vm, vbfloat16mf4x8_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg8e16_v_bf16mf4x8_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlsseg8e16_v_bf16mf2x8_tum(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlsseg8.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], <vscale x 2 x bfloat> [[TMP7]], ptr [[RS1]], i64 [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP8]]
+//
+vbfloat16mf2x8_t test_vlsseg8e16_v_bf16mf2x8_tum(vbool32_t vm, vbfloat16mf2x8_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg8e16_v_bf16mf2x8_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlsseg8e16_v_bf16m1x8_tum(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlsseg8.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], <vscale x 4 x bfloat> [[TMP7]], ptr [[RS1]], i64 [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP8]]
+//
+vbfloat16m1x8_t test_vlsseg8e16_v_bf16m1x8_tum(vbool16_t vm, vbfloat16m1x8_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg8e16_v_bf16m1x8_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlsseg8e16_v_bf16mf4x8_tumu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlsseg8.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], <vscale x 1 x bfloat> [[TMP7]], ptr [[RS1]], i64 [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP8]]
+//
+vbfloat16mf4x8_t test_vlsseg8e16_v_bf16mf4x8_tumu(vbool64_t vm, vbfloat16mf4x8_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg8e16_v_bf16mf4x8_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlsseg8e16_v_bf16mf2x8_tumu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlsseg8.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], <vscale x 2 x bfloat> [[TMP7]], ptr [[RS1]], i64 [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP8]]
+//
+vbfloat16mf2x8_t test_vlsseg8e16_v_bf16mf2x8_tumu(vbool32_t vm, vbfloat16mf2x8_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg8e16_v_bf16mf2x8_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlsseg8e16_v_bf16m1x8_tumu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlsseg8.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], <vscale x 4 x bfloat> [[TMP7]], ptr [[RS1]], i64 [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP8]]
+//
+vbfloat16m1x8_t test_vlsseg8e16_v_bf16m1x8_tumu(vbool16_t vm, vbfloat16m1x8_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg8e16_v_bf16m1x8_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlsseg8e16_v_bf16mf4x8_mu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlsseg8.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], <vscale x 1 x bfloat> [[TMP7]], ptr [[RS1]], i64 [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP8]]
+//
+vbfloat16mf4x8_t test_vlsseg8e16_v_bf16mf4x8_mu(vbool64_t vm, vbfloat16mf4x8_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg8e16_v_bf16mf4x8_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlsseg8e16_v_bf16mf2x8_mu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlsseg8.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], <vscale x 2 x bfloat> [[TMP7]], ptr [[RS1]], i64 [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP8]]
+//
+vbfloat16mf2x8_t test_vlsseg8e16_v_bf16mf2x8_mu(vbool32_t vm, vbfloat16mf2x8_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg8e16_v_bf16mf2x8_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlsseg8e16_v_bf16m1x8_mu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlsseg8.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], <vscale x 4 x bfloat> [[TMP7]], ptr [[RS1]], i64 [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP8]]
+//
+vbfloat16m1x8_t test_vlsseg8e16_v_bf16m1x8_mu(vbool16_t vm, vbfloat16m1x8_t vd, const __bf16 *rs1, ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg8e16_v_bf16m1x8_mu(vm, vd, rs1, rs2, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vluxei16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vluxei16.c
new file mode 100644
index 000000000000..490f594671e5
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vluxei16.c
@@ -0,0 +1,249 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 1 x bfloat> @test_vluxei16_v_bf16mf4_tu(
+// CHECK-RV64-SAME: <vscale x 1 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x bfloat> @llvm.riscv.vluxei.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[VD]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret <vscale x 1 x bfloat> [[TMP0]]
+//
+vbfloat16mf4_t test_vluxei16_v_bf16mf4_tu(vbfloat16mf4_t vd, const __bf16 *rs1, vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vluxei16_v_bf16mf4_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 2 x bfloat> @test_vluxei16_v_bf16mf2_tu(
+// CHECK-RV64-SAME: <vscale x 2 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x bfloat> @llvm.riscv.vluxei.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[VD]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret <vscale x 2 x bfloat> [[TMP0]]
+//
+vbfloat16mf2_t test_vluxei16_v_bf16mf2_tu(vbfloat16mf2_t vd, const __bf16 *rs1, vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vluxei16_v_bf16mf2_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 4 x bfloat> @test_vluxei16_v_bf16m1_tu(
+// CHECK-RV64-SAME: <vscale x 4 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x bfloat> @llvm.riscv.vluxei.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[VD]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret <vscale x 4 x bfloat> [[TMP0]]
+//
+vbfloat16m1_t test_vluxei16_v_bf16m1_tu(vbfloat16m1_t vd, const __bf16 *rs1, vuint16m1_t rs2, size_t vl) {
+  return __riscv_vluxei16_v_bf16m1_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 8 x bfloat> @test_vluxei16_v_bf16m2_tu(
+// CHECK-RV64-SAME: <vscale x 8 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x bfloat> @llvm.riscv.vluxei.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> [[VD]], ptr [[RS1]], <vscale x 8 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret <vscale x 8 x bfloat> [[TMP0]]
+//
+vbfloat16m2_t test_vluxei16_v_bf16m2_tu(vbfloat16m2_t vd, const __bf16 *rs1, vuint16m2_t rs2, size_t vl) {
+  return __riscv_vluxei16_v_bf16m2_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 16 x bfloat> @test_vluxei16_v_bf16m4_tu(
+// CHECK-RV64-SAME: <vscale x 16 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 16 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x bfloat> @llvm.riscv.vluxei.nxv16bf16.nxv16i16.i64(<vscale x 16 x bfloat> [[VD]], ptr [[RS1]], <vscale x 16 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret <vscale x 16 x bfloat> [[TMP0]]
+//
+vbfloat16m4_t test_vluxei16_v_bf16m4_tu(vbfloat16m4_t vd, const __bf16 *rs1, vuint16m4_t rs2, size_t vl) {
+  return __riscv_vluxei16_v_bf16m4_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 32 x bfloat> @test_vluxei16_v_bf16m8_tu(
+// CHECK-RV64-SAME: <vscale x 32 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 32 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x bfloat> @llvm.riscv.vluxei.nxv32bf16.nxv32i16.i64(<vscale x 32 x bfloat> [[VD]], ptr [[RS1]], <vscale x 32 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret <vscale x 32 x bfloat> [[TMP0]]
+//
+vbfloat16m8_t test_vluxei16_v_bf16m8_tu(vbfloat16m8_t vd, const __bf16 *rs1, vuint16m8_t rs2, size_t vl) {
+  return __riscv_vluxei16_v_bf16m8_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 1 x bfloat> @test_vluxei16_v_bf16mf4_tum(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], <vscale x 1 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x bfloat> @llvm.riscv.vluxei.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[VD]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret <vscale x 1 x bfloat> [[TMP0]]
+//
+vbfloat16mf4_t test_vluxei16_v_bf16mf4_tum(vbool64_t vm, vbfloat16mf4_t vd, const __bf16 *rs1, vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vluxei16_v_bf16mf4_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 2 x bfloat> @test_vluxei16_v_bf16mf2_tum(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], <vscale x 2 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x bfloat> @llvm.riscv.vluxei.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[VD]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret <vscale x 2 x bfloat> [[TMP0]]
+//
+vbfloat16mf2_t test_vluxei16_v_bf16mf2_tum(vbool32_t vm, vbfloat16mf2_t vd, const __bf16 *rs1, vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vluxei16_v_bf16mf2_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 4 x bfloat> @test_vluxei16_v_bf16m1_tum(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], <vscale x 4 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x bfloat> @llvm.riscv.vluxei.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[VD]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret <vscale x 4 x bfloat> [[TMP0]]
+//
+vbfloat16m1_t test_vluxei16_v_bf16m1_tum(vbool16_t vm, vbfloat16m1_t vd, const __bf16 *rs1, vuint16m1_t rs2, size_t vl) {
+  return __riscv_vluxei16_v_bf16m1_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 8 x bfloat> @test_vluxei16_v_bf16m2_tum(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], <vscale x 8 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x bfloat> @llvm.riscv.vluxei.mask.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> [[VD]], ptr [[RS1]], <vscale x 8 x i16> [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret <vscale x 8 x bfloat> [[TMP0]]
+//
+vbfloat16m2_t test_vluxei16_v_bf16m2_tum(vbool8_t vm, vbfloat16m2_t vd, const __bf16 *rs1, vuint16m2_t rs2, size_t vl) {
+  return __riscv_vluxei16_v_bf16m2_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 16 x bfloat> @test_vluxei16_v_bf16m4_tum(
+// CHECK-RV64-SAME: <vscale x 16 x i1> [[VM:%.*]], <vscale x 16 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 16 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x bfloat> @llvm.riscv.vluxei.mask.nxv16bf16.nxv16i16.i64(<vscale x 16 x bfloat> [[VD]], ptr [[RS1]], <vscale x 16 x i16> [[RS2]], <vscale x 16 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret <vscale x 16 x bfloat> [[TMP0]]
+//
+vbfloat16m4_t test_vluxei16_v_bf16m4_tum(vbool4_t vm, vbfloat16m4_t vd, const __bf16 *rs1, vuint16m4_t rs2, size_t vl) {
+  return __riscv_vluxei16_v_bf16m4_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 32 x bfloat> @test_vluxei16_v_bf16m8_tum(
+// CHECK-RV64-SAME: <vscale x 32 x i1> [[VM:%.*]], <vscale x 32 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 32 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x bfloat> @llvm.riscv.vluxei.mask.nxv32bf16.nxv32i16.i64(<vscale x 32 x bfloat> [[VD]], ptr [[RS1]], <vscale x 32 x i16> [[RS2]], <vscale x 32 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret <vscale x 32 x bfloat> [[TMP0]]
+//
+vbfloat16m8_t test_vluxei16_v_bf16m8_tum(vbool2_t vm, vbfloat16m8_t vd, const __bf16 *rs1, vuint16m8_t rs2, size_t vl) {
+  return __riscv_vluxei16_v_bf16m8_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 1 x bfloat> @test_vluxei16_v_bf16mf4_tumu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], <vscale x 1 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x bfloat> @llvm.riscv.vluxei.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[VD]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 1 x bfloat> [[TMP0]]
+//
+vbfloat16mf4_t test_vluxei16_v_bf16mf4_tumu(vbool64_t vm, vbfloat16mf4_t vd, const __bf16 *rs1, vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vluxei16_v_bf16mf4_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 2 x bfloat> @test_vluxei16_v_bf16mf2_tumu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], <vscale x 2 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x bfloat> @llvm.riscv.vluxei.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[VD]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 2 x bfloat> [[TMP0]]
+//
+vbfloat16mf2_t test_vluxei16_v_bf16mf2_tumu(vbool32_t vm, vbfloat16mf2_t vd, const __bf16 *rs1, vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vluxei16_v_bf16mf2_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 4 x bfloat> @test_vluxei16_v_bf16m1_tumu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], <vscale x 4 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x bfloat> @llvm.riscv.vluxei.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[VD]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 4 x bfloat> [[TMP0]]
+//
+vbfloat16m1_t test_vluxei16_v_bf16m1_tumu(vbool16_t vm, vbfloat16m1_t vd, const __bf16 *rs1, vuint16m1_t rs2, size_t vl) {
+  return __riscv_vluxei16_v_bf16m1_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 8 x bfloat> @test_vluxei16_v_bf16m2_tumu(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], <vscale x 8 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x bfloat> @llvm.riscv.vluxei.mask.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> [[VD]], ptr [[RS1]], <vscale x 8 x i16> [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 8 x bfloat> [[TMP0]]
+//
+vbfloat16m2_t test_vluxei16_v_bf16m2_tumu(vbool8_t vm, vbfloat16m2_t vd, const __bf16 *rs1, vuint16m2_t rs2, size_t vl) {
+  return __riscv_vluxei16_v_bf16m2_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 16 x bfloat> @test_vluxei16_v_bf16m4_tumu(
+// CHECK-RV64-SAME: <vscale x 16 x i1> [[VM:%.*]], <vscale x 16 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 16 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x bfloat> @llvm.riscv.vluxei.mask.nxv16bf16.nxv16i16.i64(<vscale x 16 x bfloat> [[VD]], ptr [[RS1]], <vscale x 16 x i16> [[RS2]], <vscale x 16 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 16 x bfloat> [[TMP0]]
+//
+vbfloat16m4_t test_vluxei16_v_bf16m4_tumu(vbool4_t vm, vbfloat16m4_t vd, const __bf16 *rs1, vuint16m4_t rs2, size_t vl) {
+  return __riscv_vluxei16_v_bf16m4_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 32 x bfloat> @test_vluxei16_v_bf16m8_tumu(
+// CHECK-RV64-SAME: <vscale x 32 x i1> [[VM:%.*]], <vscale x 32 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 32 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x bfloat> @llvm.riscv.vluxei.mask.nxv32bf16.nxv32i16.i64(<vscale x 32 x bfloat> [[VD]], ptr [[RS1]], <vscale x 32 x i16> [[RS2]], <vscale x 32 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 32 x bfloat> [[TMP0]]
+//
+vbfloat16m8_t test_vluxei16_v_bf16m8_tumu(vbool2_t vm, vbfloat16m8_t vd, const __bf16 *rs1, vuint16m8_t rs2, size_t vl) {
+  return __riscv_vluxei16_v_bf16m8_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 1 x bfloat> @test_vluxei16_v_bf16mf4_mu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], <vscale x 1 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x bfloat> @llvm.riscv.vluxei.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[VD]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret <vscale x 1 x bfloat> [[TMP0]]
+//
+vbfloat16mf4_t test_vluxei16_v_bf16mf4_mu(vbool64_t vm, vbfloat16mf4_t vd, const __bf16 *rs1, vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vluxei16_v_bf16mf4_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 2 x bfloat> @test_vluxei16_v_bf16mf2_mu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], <vscale x 2 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x bfloat> @llvm.riscv.vluxei.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[VD]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret <vscale x 2 x bfloat> [[TMP0]]
+//
+vbfloat16mf2_t test_vluxei16_v_bf16mf2_mu(vbool32_t vm, vbfloat16mf2_t vd, const __bf16 *rs1, vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vluxei16_v_bf16mf2_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 4 x bfloat> @test_vluxei16_v_bf16m1_mu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], <vscale x 4 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x bfloat> @llvm.riscv.vluxei.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[VD]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret <vscale x 4 x bfloat> [[TMP0]]
+//
+vbfloat16m1_t test_vluxei16_v_bf16m1_mu(vbool16_t vm, vbfloat16m1_t vd, const __bf16 *rs1, vuint16m1_t rs2, size_t vl) {
+  return __riscv_vluxei16_v_bf16m1_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 8 x bfloat> @test_vluxei16_v_bf16m2_mu(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], <vscale x 8 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x bfloat> @llvm.riscv.vluxei.mask.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> [[VD]], ptr [[RS1]], <vscale x 8 x i16> [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret <vscale x 8 x bfloat> [[TMP0]]
+//
+vbfloat16m2_t test_vluxei16_v_bf16m2_mu(vbool8_t vm, vbfloat16m2_t vd, const __bf16 *rs1, vuint16m2_t rs2, size_t vl) {
+  return __riscv_vluxei16_v_bf16m2_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 16 x bfloat> @test_vluxei16_v_bf16m4_mu(
+// CHECK-RV64-SAME: <vscale x 16 x i1> [[VM:%.*]], <vscale x 16 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 16 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x bfloat> @llvm.riscv.vluxei.mask.nxv16bf16.nxv16i16.i64(<vscale x 16 x bfloat> [[VD]], ptr [[RS1]], <vscale x 16 x i16> [[RS2]], <vscale x 16 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret <vscale x 16 x bfloat> [[TMP0]]
+//
+vbfloat16m4_t test_vluxei16_v_bf16m4_mu(vbool4_t vm, vbfloat16m4_t vd, const __bf16 *rs1, vuint16m4_t rs2, size_t vl) {
+  return __riscv_vluxei16_v_bf16m4_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 32 x bfloat> @test_vluxei16_v_bf16m8_mu(
+// CHECK-RV64-SAME: <vscale x 32 x i1> [[VM:%.*]], <vscale x 32 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 32 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x bfloat> @llvm.riscv.vluxei.mask.nxv32bf16.nxv32i16.i64(<vscale x 32 x bfloat> [[VD]], ptr [[RS1]], <vscale x 32 x i16> [[RS2]], <vscale x 32 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret <vscale x 32 x bfloat> [[TMP0]]
+//
+vbfloat16m8_t test_vluxei16_v_bf16m8_mu(vbool2_t vm, vbfloat16m8_t vd, const __bf16 *rs1, vuint16m8_t rs2, size_t vl) {
+  return __riscv_vluxei16_v_bf16m8_mu(vm, vd, rs1, rs2, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vluxseg2ei16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vluxseg2ei16.c
new file mode 100644
index 000000000000..c363102b8c2b
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vluxseg2ei16.c
@@ -0,0 +1,249 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vluxseg2ei16_v_bf16mf4x2_tu(
+// CHECK-RV64-SAME: { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vluxseg2.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP2]]
+//
+vbfloat16mf4x2_t test_vluxseg2ei16_v_bf16mf4x2_tu(vbfloat16mf4x2_t vd, const __bf16 *rs1, vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vluxseg2ei16_v_bf16mf4x2_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vluxseg2ei16_v_bf16mf2x2_tu(
+// CHECK-RV64-SAME: { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vluxseg2.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP2]]
+//
+vbfloat16mf2x2_t test_vluxseg2ei16_v_bf16mf2x2_tu(vbfloat16mf2x2_t vd, const __bf16 *rs1, vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vluxseg2ei16_v_bf16mf2x2_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vluxseg2ei16_v_bf16m1x2_tu(
+// CHECK-RV64-SAME: { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vluxseg2.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP2]]
+//
+vbfloat16m1x2_t test_vluxseg2ei16_v_bf16m1x2_tu(vbfloat16m1x2_t vd, const __bf16 *rs1, vuint16m1_t rs2, size_t vl) {
+  return __riscv_vluxseg2ei16_v_bf16m1x2_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vluxseg2ei16_v_bf16m2x2_tu(
+// CHECK-RV64-SAME: { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vluxseg2.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 8 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP2]]
+//
+vbfloat16m2x2_t test_vluxseg2ei16_v_bf16m2x2_tu(vbfloat16m2x2_t vd, const __bf16 *rs1, vuint16m2_t rs2, size_t vl) {
+  return __riscv_vluxseg2ei16_v_bf16m2x2_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @test_vluxseg2ei16_v_bf16m4x2_tu(
+// CHECK-RV64-SAME: { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 16 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @llvm.riscv.vluxseg2.nxv16bf16.nxv16i16.i64(<vscale x 16 x bfloat> [[TMP0]], <vscale x 16 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 16 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[TMP2]]
+//
+vbfloat16m4x2_t test_vluxseg2ei16_v_bf16m4x2_tu(vbfloat16m4x2_t vd, const __bf16 *rs1, vuint16m4_t rs2, size_t vl) {
+  return __riscv_vluxseg2ei16_v_bf16m4x2_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vluxseg2ei16_v_bf16mf4x2_tum(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vluxseg2.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP2]]
+//
+vbfloat16mf4x2_t test_vluxseg2ei16_v_bf16mf4x2_tum(vbool64_t vm, vbfloat16mf4x2_t vd, const __bf16 *rs1, vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vluxseg2ei16_v_bf16mf4x2_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vluxseg2ei16_v_bf16mf2x2_tum(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vluxseg2.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP2]]
+//
+vbfloat16mf2x2_t test_vluxseg2ei16_v_bf16mf2x2_tum(vbool32_t vm, vbfloat16mf2x2_t vd, const __bf16 *rs1, vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vluxseg2ei16_v_bf16mf2x2_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vluxseg2ei16_v_bf16m1x2_tum(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vluxseg2.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP2]]
+//
+vbfloat16m1x2_t test_vluxseg2ei16_v_bf16m1x2_tum(vbool16_t vm, vbfloat16m1x2_t vd, const __bf16 *rs1, vuint16m1_t rs2, size_t vl) {
+  return __riscv_vluxseg2ei16_v_bf16m1x2_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vluxseg2ei16_v_bf16m2x2_tum(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vluxseg2.mask.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 8 x i16> [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP2]]
+//
+vbfloat16m2x2_t test_vluxseg2ei16_v_bf16m2x2_tum(vbool8_t vm, vbfloat16m2x2_t vd, const __bf16 *rs1, vuint16m2_t rs2, size_t vl) {
+  return __riscv_vluxseg2ei16_v_bf16m2x2_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @test_vluxseg2ei16_v_bf16m4x2_tum(
+// CHECK-RV64-SAME: <vscale x 16 x i1> [[VM:%.*]], { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 16 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @llvm.riscv.vluxseg2.mask.nxv16bf16.nxv16i16.i64(<vscale x 16 x bfloat> [[TMP0]], <vscale x 16 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 16 x i16> [[RS2]], <vscale x 16 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[TMP2]]
+//
+vbfloat16m4x2_t test_vluxseg2ei16_v_bf16m4x2_tum(vbool4_t vm, vbfloat16m4x2_t vd, const __bf16 *rs1, vuint16m4_t rs2, size_t vl) {
+  return __riscv_vluxseg2ei16_v_bf16m4x2_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vluxseg2ei16_v_bf16mf4x2_tumu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vluxseg2.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP2]]
+//
+vbfloat16mf4x2_t test_vluxseg2ei16_v_bf16mf4x2_tumu(vbool64_t vm, vbfloat16mf4x2_t vd, const __bf16 *rs1, vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vluxseg2ei16_v_bf16mf4x2_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vluxseg2ei16_v_bf16mf2x2_tumu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vluxseg2.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP2]]
+//
+vbfloat16mf2x2_t test_vluxseg2ei16_v_bf16mf2x2_tumu(vbool32_t vm, vbfloat16mf2x2_t vd, const __bf16 *rs1, vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vluxseg2ei16_v_bf16mf2x2_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vluxseg2ei16_v_bf16m1x2_tumu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vluxseg2.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP2]]
+//
+vbfloat16m1x2_t test_vluxseg2ei16_v_bf16m1x2_tumu(vbool16_t vm, vbfloat16m1x2_t vd, const __bf16 *rs1, vuint16m1_t rs2, size_t vl) {
+  return __riscv_vluxseg2ei16_v_bf16m1x2_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vluxseg2ei16_v_bf16m2x2_tumu(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vluxseg2.mask.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 8 x i16> [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP2]]
+//
+vbfloat16m2x2_t test_vluxseg2ei16_v_bf16m2x2_tumu(vbool8_t vm, vbfloat16m2x2_t vd, const __bf16 *rs1, vuint16m2_t rs2, size_t vl) {
+  return __riscv_vluxseg2ei16_v_bf16m2x2_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @test_vluxseg2ei16_v_bf16m4x2_tumu(
+// CHECK-RV64-SAME: <vscale x 16 x i1> [[VM:%.*]], { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 16 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @llvm.riscv.vluxseg2.mask.nxv16bf16.nxv16i16.i64(<vscale x 16 x bfloat> [[TMP0]], <vscale x 16 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 16 x i16> [[RS2]], <vscale x 16 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[TMP2]]
+//
+vbfloat16m4x2_t test_vluxseg2ei16_v_bf16m4x2_tumu(vbool4_t vm, vbfloat16m4x2_t vd, const __bf16 *rs1, vuint16m4_t rs2, size_t vl) {
+  return __riscv_vluxseg2ei16_v_bf16m4x2_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vluxseg2ei16_v_bf16mf4x2_mu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vluxseg2.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP2]]
+//
+vbfloat16mf4x2_t test_vluxseg2ei16_v_bf16mf4x2_mu(vbool64_t vm, vbfloat16mf4x2_t vd, const __bf16 *rs1, vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vluxseg2ei16_v_bf16mf4x2_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vluxseg2ei16_v_bf16mf2x2_mu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vluxseg2.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP2]]
+//
+vbfloat16mf2x2_t test_vluxseg2ei16_v_bf16mf2x2_mu(vbool32_t vm, vbfloat16mf2x2_t vd, const __bf16 *rs1, vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vluxseg2ei16_v_bf16mf2x2_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vluxseg2ei16_v_bf16m1x2_mu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vluxseg2.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP2]]
+//
+vbfloat16m1x2_t test_vluxseg2ei16_v_bf16m1x2_mu(vbool16_t vm, vbfloat16m1x2_t vd, const __bf16 *rs1, vuint16m1_t rs2, size_t vl) {
+  return __riscv_vluxseg2ei16_v_bf16m1x2_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vluxseg2ei16_v_bf16m2x2_mu(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vluxseg2.mask.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 8 x i16> [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP2]]
+//
+vbfloat16m2x2_t test_vluxseg2ei16_v_bf16m2x2_mu(vbool8_t vm, vbfloat16m2x2_t vd, const __bf16 *rs1, vuint16m2_t rs2, size_t vl) {
+  return __riscv_vluxseg2ei16_v_bf16m2x2_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @test_vluxseg2ei16_v_bf16m4x2_mu(
+// CHECK-RV64-SAME: <vscale x 16 x i1> [[VM:%.*]], { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 16 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @llvm.riscv.vluxseg2.mask.nxv16bf16.nxv16i16.i64(<vscale x 16 x bfloat> [[TMP0]], <vscale x 16 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 16 x i16> [[RS2]], <vscale x 16 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[TMP2]]
+//
+vbfloat16m4x2_t test_vluxseg2ei16_v_bf16m4x2_mu(vbool4_t vm, vbfloat16m4x2_t vd, const __bf16 *rs1, vuint16m4_t rs2, size_t vl) {
+  return __riscv_vluxseg2ei16_v_bf16m4x2_mu(vm, vd, rs1, rs2, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vluxseg3ei16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vluxseg3ei16.c
new file mode 100644
index 000000000000..b93fd2b5afd4
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vluxseg3ei16.c
@@ -0,0 +1,217 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vluxseg3ei16_v_bf16mf4x3_tu(
+// CHECK-RV64-SAME: { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vluxseg3.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP3]]
+//
+vbfloat16mf4x3_t test_vluxseg3ei16_v_bf16mf4x3_tu(vbfloat16mf4x3_t vd, const __bf16 *rs1, vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vluxseg3ei16_v_bf16mf4x3_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vluxseg3ei16_v_bf16mf2x3_tu(
+// CHECK-RV64-SAME: { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vluxseg3.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP3]]
+//
+vbfloat16mf2x3_t test_vluxseg3ei16_v_bf16mf2x3_tu(vbfloat16mf2x3_t vd, const __bf16 *rs1, vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vluxseg3ei16_v_bf16mf2x3_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vluxseg3ei16_v_bf16m1x3_tu(
+// CHECK-RV64-SAME: { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vluxseg3.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP3]]
+//
+vbfloat16m1x3_t test_vluxseg3ei16_v_bf16m1x3_tu(vbfloat16m1x3_t vd, const __bf16 *rs1, vuint16m1_t rs2, size_t vl) {
+  return __riscv_vluxseg3ei16_v_bf16m1x3_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vluxseg3ei16_v_bf16m2x3_tu(
+// CHECK-RV64-SAME: { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vluxseg3.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 8 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP3]]
+//
+vbfloat16m2x3_t test_vluxseg3ei16_v_bf16m2x3_tu(vbfloat16m2x3_t vd, const __bf16 *rs1, vuint16m2_t rs2, size_t vl) {
+  return __riscv_vluxseg3ei16_v_bf16m2x3_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vluxseg3ei16_v_bf16mf4x3_tum(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vluxseg3.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP3]]
+//
+vbfloat16mf4x3_t test_vluxseg3ei16_v_bf16mf4x3_tum(vbool64_t vm, vbfloat16mf4x3_t vd, const __bf16 *rs1, vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vluxseg3ei16_v_bf16mf4x3_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vluxseg3ei16_v_bf16mf2x3_tum(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vluxseg3.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP3]]
+//
+vbfloat16mf2x3_t test_vluxseg3ei16_v_bf16mf2x3_tum(vbool32_t vm, vbfloat16mf2x3_t vd, const __bf16 *rs1, vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vluxseg3ei16_v_bf16mf2x3_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vluxseg3ei16_v_bf16m1x3_tum(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vluxseg3.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP3]]
+//
+vbfloat16m1x3_t test_vluxseg3ei16_v_bf16m1x3_tum(vbool16_t vm, vbfloat16m1x3_t vd, const __bf16 *rs1, vuint16m1_t rs2, size_t vl) {
+  return __riscv_vluxseg3ei16_v_bf16m1x3_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vluxseg3ei16_v_bf16m2x3_tum(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vluxseg3.mask.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 8 x i16> [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP3]]
+//
+vbfloat16m2x3_t test_vluxseg3ei16_v_bf16m2x3_tum(vbool8_t vm, vbfloat16m2x3_t vd, const __bf16 *rs1, vuint16m2_t rs2, size_t vl) {
+  return __riscv_vluxseg3ei16_v_bf16m2x3_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vluxseg3ei16_v_bf16mf4x3_tumu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vluxseg3.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP3]]
+//
+vbfloat16mf4x3_t test_vluxseg3ei16_v_bf16mf4x3_tumu(vbool64_t vm, vbfloat16mf4x3_t vd, const __bf16 *rs1, vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vluxseg3ei16_v_bf16mf4x3_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vluxseg3ei16_v_bf16mf2x3_tumu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vluxseg3.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP3]]
+//
+vbfloat16mf2x3_t test_vluxseg3ei16_v_bf16mf2x3_tumu(vbool32_t vm, vbfloat16mf2x3_t vd, const __bf16 *rs1, vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vluxseg3ei16_v_bf16mf2x3_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vluxseg3ei16_v_bf16m1x3_tumu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vluxseg3.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP3]]
+//
+vbfloat16m1x3_t test_vluxseg3ei16_v_bf16m1x3_tumu(vbool16_t vm, vbfloat16m1x3_t vd, const __bf16 *rs1, vuint16m1_t rs2, size_t vl) {
+  return __riscv_vluxseg3ei16_v_bf16m1x3_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vluxseg3ei16_v_bf16m2x3_tumu(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vluxseg3.mask.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 8 x i16> [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP3]]
+//
+vbfloat16m2x3_t test_vluxseg3ei16_v_bf16m2x3_tumu(vbool8_t vm, vbfloat16m2x3_t vd, const __bf16 *rs1, vuint16m2_t rs2, size_t vl) {
+  return __riscv_vluxseg3ei16_v_bf16m2x3_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vluxseg3ei16_v_bf16mf4x3_mu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vluxseg3.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP3]]
+//
+vbfloat16mf4x3_t test_vluxseg3ei16_v_bf16mf4x3_mu(vbool64_t vm, vbfloat16mf4x3_t vd, const __bf16 *rs1, vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vluxseg3ei16_v_bf16mf4x3_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vluxseg3ei16_v_bf16mf2x3_mu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vluxseg3.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP3]]
+//
+vbfloat16mf2x3_t test_vluxseg3ei16_v_bf16mf2x3_mu(vbool32_t vm, vbfloat16mf2x3_t vd, const __bf16 *rs1, vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vluxseg3ei16_v_bf16mf2x3_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vluxseg3ei16_v_bf16m1x3_mu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vluxseg3.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP3]]
+//
+vbfloat16m1x3_t test_vluxseg3ei16_v_bf16m1x3_mu(vbool16_t vm, vbfloat16m1x3_t vd, const __bf16 *rs1, vuint16m1_t rs2, size_t vl) {
+  return __riscv_vluxseg3ei16_v_bf16m1x3_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vluxseg3ei16_v_bf16m2x3_mu(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vluxseg3.mask.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 8 x i16> [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP3]]
+//
+vbfloat16m2x3_t test_vluxseg3ei16_v_bf16m2x3_mu(vbool8_t vm, vbfloat16m2x3_t vd, const __bf16 *rs1, vuint16m2_t rs2, size_t vl) {
+  return __riscv_vluxseg3ei16_v_bf16m2x3_mu(vm, vd, rs1, rs2, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vluxseg4ei16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vluxseg4ei16.c
new file mode 100644
index 000000000000..ba48a9e1d5f5
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vluxseg4ei16.c
@@ -0,0 +1,233 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vluxseg4ei16_v_bf16mf4x4_tu(
+// CHECK-RV64-SAME: { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vluxseg4.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP4]]
+//
+vbfloat16mf4x4_t test_vluxseg4ei16_v_bf16mf4x4_tu(vbfloat16mf4x4_t vd, const __bf16 *rs1, vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vluxseg4ei16_v_bf16mf4x4_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vluxseg4ei16_v_bf16mf2x4_tu(
+// CHECK-RV64-SAME: { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vluxseg4.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP4]]
+//
+vbfloat16mf2x4_t test_vluxseg4ei16_v_bf16mf2x4_tu(vbfloat16mf2x4_t vd, const __bf16 *rs1, vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vluxseg4ei16_v_bf16mf2x4_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vluxseg4ei16_v_bf16m1x4_tu(
+// CHECK-RV64-SAME: { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vluxseg4.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP4]]
+//
+vbfloat16m1x4_t test_vluxseg4ei16_v_bf16m1x4_tu(vbfloat16m1x4_t vd, const __bf16 *rs1, vuint16m1_t rs2, size_t vl) {
+  return __riscv_vluxseg4ei16_v_bf16m1x4_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vluxseg4ei16_v_bf16m2x4_tu(
+// CHECK-RV64-SAME: { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vluxseg4.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], <vscale x 8 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 8 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP4]]
+//
+vbfloat16m2x4_t test_vluxseg4ei16_v_bf16m2x4_tu(vbfloat16m2x4_t vd, const __bf16 *rs1, vuint16m2_t rs2, size_t vl) {
+  return __riscv_vluxseg4ei16_v_bf16m2x4_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vluxseg4ei16_v_bf16mf4x4_tum(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vluxseg4.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP4]]
+//
+vbfloat16mf4x4_t test_vluxseg4ei16_v_bf16mf4x4_tum(vbool64_t vm, vbfloat16mf4x4_t vd, const __bf16 *rs1, vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vluxseg4ei16_v_bf16mf4x4_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vluxseg4ei16_v_bf16mf2x4_tum(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vluxseg4.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP4]]
+//
+vbfloat16mf2x4_t test_vluxseg4ei16_v_bf16mf2x4_tum(vbool32_t vm, vbfloat16mf2x4_t vd, const __bf16 *rs1, vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vluxseg4ei16_v_bf16mf2x4_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vluxseg4ei16_v_bf16m1x4_tum(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vluxseg4.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP4]]
+//
+vbfloat16m1x4_t test_vluxseg4ei16_v_bf16m1x4_tum(vbool16_t vm, vbfloat16m1x4_t vd, const __bf16 *rs1, vuint16m1_t rs2, size_t vl) {
+  return __riscv_vluxseg4ei16_v_bf16m1x4_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vluxseg4ei16_v_bf16m2x4_tum(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vluxseg4.mask.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], <vscale x 8 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 8 x i16> [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP4]]
+//
+vbfloat16m2x4_t test_vluxseg4ei16_v_bf16m2x4_tum(vbool8_t vm, vbfloat16m2x4_t vd, const __bf16 *rs1, vuint16m2_t rs2, size_t vl) {
+  return __riscv_vluxseg4ei16_v_bf16m2x4_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vluxseg4ei16_v_bf16mf4x4_tumu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vluxseg4.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP4]]
+//
+vbfloat16mf4x4_t test_vluxseg4ei16_v_bf16mf4x4_tumu(vbool64_t vm, vbfloat16mf4x4_t vd, const __bf16 *rs1, vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vluxseg4ei16_v_bf16mf4x4_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vluxseg4ei16_v_bf16mf2x4_tumu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vluxseg4.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP4]]
+//
+vbfloat16mf2x4_t test_vluxseg4ei16_v_bf16mf2x4_tumu(vbool32_t vm, vbfloat16mf2x4_t vd, const __bf16 *rs1, vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vluxseg4ei16_v_bf16mf2x4_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vluxseg4ei16_v_bf16m1x4_tumu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vluxseg4.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP4]]
+//
+vbfloat16m1x4_t test_vluxseg4ei16_v_bf16m1x4_tumu(vbool16_t vm, vbfloat16m1x4_t vd, const __bf16 *rs1, vuint16m1_t rs2, size_t vl) {
+  return __riscv_vluxseg4ei16_v_bf16m1x4_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vluxseg4ei16_v_bf16m2x4_tumu(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vluxseg4.mask.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], <vscale x 8 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 8 x i16> [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP4]]
+//
+vbfloat16m2x4_t test_vluxseg4ei16_v_bf16m2x4_tumu(vbool8_t vm, vbfloat16m2x4_t vd, const __bf16 *rs1, vuint16m2_t rs2, size_t vl) {
+  return __riscv_vluxseg4ei16_v_bf16m2x4_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vluxseg4ei16_v_bf16mf4x4_mu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vluxseg4.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP4]]
+//
+vbfloat16mf4x4_t test_vluxseg4ei16_v_bf16mf4x4_mu(vbool64_t vm, vbfloat16mf4x4_t vd, const __bf16 *rs1, vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vluxseg4ei16_v_bf16mf4x4_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vluxseg4ei16_v_bf16mf2x4_mu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vluxseg4.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP4]]
+//
+vbfloat16mf2x4_t test_vluxseg4ei16_v_bf16mf2x4_mu(vbool32_t vm, vbfloat16mf2x4_t vd, const __bf16 *rs1, vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vluxseg4ei16_v_bf16mf2x4_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vluxseg4ei16_v_bf16m1x4_mu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vluxseg4.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP4]]
+//
+vbfloat16m1x4_t test_vluxseg4ei16_v_bf16m1x4_mu(vbool16_t vm, vbfloat16m1x4_t vd, const __bf16 *rs1, vuint16m1_t rs2, size_t vl) {
+  return __riscv_vluxseg4ei16_v_bf16m1x4_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vluxseg4ei16_v_bf16m2x4_mu(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vluxseg4.mask.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], <vscale x 8 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 8 x i16> [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP4]]
+//
+vbfloat16m2x4_t test_vluxseg4ei16_v_bf16m2x4_mu(vbool8_t vm, vbfloat16m2x4_t vd, const __bf16 *rs1, vuint16m2_t rs2, size_t vl) {
+  return __riscv_vluxseg4ei16_v_bf16m2x4_mu(vm, vd, rs1, rs2, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vluxseg5ei16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vluxseg5ei16.c
new file mode 100644
index 000000000000..9e8443b135a6
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vluxseg5ei16.c
@@ -0,0 +1,189 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vluxseg5ei16_v_bf16mf4x5_tu(
+// CHECK-RV64-SAME: { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vluxseg5.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP5]]
+//
+vbfloat16mf4x5_t test_vluxseg5ei16_v_bf16mf4x5_tu(vbfloat16mf4x5_t vd, const __bf16 *rs1, vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vluxseg5ei16_v_bf16mf4x5_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vluxseg5ei16_v_bf16mf2x5_tu(
+// CHECK-RV64-SAME: { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vluxseg5.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP5]]
+//
+vbfloat16mf2x5_t test_vluxseg5ei16_v_bf16mf2x5_tu(vbfloat16mf2x5_t vd, const __bf16 *rs1, vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vluxseg5ei16_v_bf16mf2x5_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vluxseg5ei16_v_bf16m1x5_tu(
+// CHECK-RV64-SAME: { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vluxseg5.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP5]]
+//
+vbfloat16m1x5_t test_vluxseg5ei16_v_bf16m1x5_tu(vbfloat16m1x5_t vd, const __bf16 *rs1, vuint16m1_t rs2, size_t vl) {
+  return __riscv_vluxseg5ei16_v_bf16m1x5_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vluxseg5ei16_v_bf16mf4x5_tum(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vluxseg5.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP5]]
+//
+vbfloat16mf4x5_t test_vluxseg5ei16_v_bf16mf4x5_tum(vbool64_t vm, vbfloat16mf4x5_t vd, const __bf16 *rs1, vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vluxseg5ei16_v_bf16mf4x5_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vluxseg5ei16_v_bf16mf2x5_tum(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vluxseg5.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP5]]
+//
+vbfloat16mf2x5_t test_vluxseg5ei16_v_bf16mf2x5_tum(vbool32_t vm, vbfloat16mf2x5_t vd, const __bf16 *rs1, vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vluxseg5ei16_v_bf16mf2x5_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vluxseg5ei16_v_bf16m1x5_tum(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vluxseg5.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP5]]
+//
+vbfloat16m1x5_t test_vluxseg5ei16_v_bf16m1x5_tum(vbool16_t vm, vbfloat16m1x5_t vd, const __bf16 *rs1, vuint16m1_t rs2, size_t vl) {
+  return __riscv_vluxseg5ei16_v_bf16m1x5_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vluxseg5ei16_v_bf16mf4x5_tumu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vluxseg5.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP5]]
+//
+vbfloat16mf4x5_t test_vluxseg5ei16_v_bf16mf4x5_tumu(vbool64_t vm, vbfloat16mf4x5_t vd, const __bf16 *rs1, vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vluxseg5ei16_v_bf16mf4x5_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vluxseg5ei16_v_bf16mf2x5_tumu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vluxseg5.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP5]]
+//
+vbfloat16mf2x5_t test_vluxseg5ei16_v_bf16mf2x5_tumu(vbool32_t vm, vbfloat16mf2x5_t vd, const __bf16 *rs1, vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vluxseg5ei16_v_bf16mf2x5_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vluxseg5ei16_v_bf16m1x5_tumu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vluxseg5.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP5]]
+//
+vbfloat16m1x5_t test_vluxseg5ei16_v_bf16m1x5_tumu(vbool16_t vm, vbfloat16m1x5_t vd, const __bf16 *rs1, vuint16m1_t rs2, size_t vl) {
+  return __riscv_vluxseg5ei16_v_bf16m1x5_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vluxseg5ei16_v_bf16mf4x5_mu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vluxseg5.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP5]]
+//
+vbfloat16mf4x5_t test_vluxseg5ei16_v_bf16mf4x5_mu(vbool64_t vm, vbfloat16mf4x5_t vd, const __bf16 *rs1, vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vluxseg5ei16_v_bf16mf4x5_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vluxseg5ei16_v_bf16mf2x5_mu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vluxseg5.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP5]]
+//
+vbfloat16mf2x5_t test_vluxseg5ei16_v_bf16mf2x5_mu(vbool32_t vm, vbfloat16mf2x5_t vd, const __bf16 *rs1, vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vluxseg5ei16_v_bf16mf2x5_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vluxseg5ei16_v_bf16m1x5_mu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vluxseg5.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP5]]
+//
+vbfloat16m1x5_t test_vluxseg5ei16_v_bf16m1x5_mu(vbool16_t vm, vbfloat16m1x5_t vd, const __bf16 *rs1, vuint16m1_t rs2, size_t vl) {
+  return __riscv_vluxseg5ei16_v_bf16m1x5_mu(vm, vd, rs1, rs2, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vluxseg6ei16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vluxseg6ei16.c
new file mode 100644
index 000000000000..635baa5378ba
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vluxseg6ei16.c
@@ -0,0 +1,201 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vluxseg6ei16_v_bf16mf4x6_tu(
+// CHECK-RV64-SAME: { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vluxseg6.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP6]]
+//
+vbfloat16mf4x6_t test_vluxseg6ei16_v_bf16mf4x6_tu(vbfloat16mf4x6_t vd, const __bf16 *rs1, vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vluxseg6ei16_v_bf16mf4x6_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vluxseg6ei16_v_bf16mf2x6_tu(
+// CHECK-RV64-SAME: { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vluxseg6.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP6]]
+//
+vbfloat16mf2x6_t test_vluxseg6ei16_v_bf16mf2x6_tu(vbfloat16mf2x6_t vd, const __bf16 *rs1, vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vluxseg6ei16_v_bf16mf2x6_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vluxseg6ei16_v_bf16m1x6_tu(
+// CHECK-RV64-SAME: { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vluxseg6.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP6]]
+//
+vbfloat16m1x6_t test_vluxseg6ei16_v_bf16m1x6_tu(vbfloat16m1x6_t vd, const __bf16 *rs1, vuint16m1_t rs2, size_t vl) {
+  return __riscv_vluxseg6ei16_v_bf16m1x6_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vluxseg6ei16_v_bf16mf4x6_tum(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vluxseg6.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP6]]
+//
+vbfloat16mf4x6_t test_vluxseg6ei16_v_bf16mf4x6_tum(vbool64_t vm, vbfloat16mf4x6_t vd, const __bf16 *rs1, vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vluxseg6ei16_v_bf16mf4x6_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vluxseg6ei16_v_bf16mf2x6_tum(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vluxseg6.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP6]]
+//
+vbfloat16mf2x6_t test_vluxseg6ei16_v_bf16mf2x6_tum(vbool32_t vm, vbfloat16mf2x6_t vd, const __bf16 *rs1, vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vluxseg6ei16_v_bf16mf2x6_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vluxseg6ei16_v_bf16m1x6_tum(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vluxseg6.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP6]]
+//
+vbfloat16m1x6_t test_vluxseg6ei16_v_bf16m1x6_tum(vbool16_t vm, vbfloat16m1x6_t vd, const __bf16 *rs1, vuint16m1_t rs2, size_t vl) {
+  return __riscv_vluxseg6ei16_v_bf16m1x6_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vluxseg6ei16_v_bf16mf4x6_tumu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vluxseg6.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP6]]
+//
+vbfloat16mf4x6_t test_vluxseg6ei16_v_bf16mf4x6_tumu(vbool64_t vm, vbfloat16mf4x6_t vd, const __bf16 *rs1, vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vluxseg6ei16_v_bf16mf4x6_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vluxseg6ei16_v_bf16mf2x6_tumu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vluxseg6.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP6]]
+//
+vbfloat16mf2x6_t test_vluxseg6ei16_v_bf16mf2x6_tumu(vbool32_t vm, vbfloat16mf2x6_t vd, const __bf16 *rs1, vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vluxseg6ei16_v_bf16mf2x6_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vluxseg6ei16_v_bf16m1x6_tumu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vluxseg6.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP6]]
+//
+vbfloat16m1x6_t test_vluxseg6ei16_v_bf16m1x6_tumu(vbool16_t vm, vbfloat16m1x6_t vd, const __bf16 *rs1, vuint16m1_t rs2, size_t vl) {
+  return __riscv_vluxseg6ei16_v_bf16m1x6_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vluxseg6ei16_v_bf16mf4x6_mu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vluxseg6.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP6]]
+//
+vbfloat16mf4x6_t test_vluxseg6ei16_v_bf16mf4x6_mu(vbool64_t vm, vbfloat16mf4x6_t vd, const __bf16 *rs1, vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vluxseg6ei16_v_bf16mf4x6_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vluxseg6ei16_v_bf16mf2x6_mu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vluxseg6.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP6]]
+//
+vbfloat16mf2x6_t test_vluxseg6ei16_v_bf16mf2x6_mu(vbool32_t vm, vbfloat16mf2x6_t vd, const __bf16 *rs1, vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vluxseg6ei16_v_bf16mf2x6_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vluxseg6ei16_v_bf16m1x6_mu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vluxseg6.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP6]]
+//
+vbfloat16m1x6_t test_vluxseg6ei16_v_bf16m1x6_mu(vbool16_t vm, vbfloat16m1x6_t vd, const __bf16 *rs1, vuint16m1_t rs2, size_t vl) {
+  return __riscv_vluxseg6ei16_v_bf16m1x6_mu(vm, vd, rs1, rs2, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vluxseg7ei16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vluxseg7ei16.c
new file mode 100644
index 000000000000..69045bc79853
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vluxseg7ei16.c
@@ -0,0 +1,213 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vluxseg7ei16_v_bf16mf4x7_tu(
+// CHECK-RV64-SAME: { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vluxseg7.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP7]]
+//
+vbfloat16mf4x7_t test_vluxseg7ei16_v_bf16mf4x7_tu(vbfloat16mf4x7_t vd, const __bf16 *rs1, vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vluxseg7ei16_v_bf16mf4x7_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vluxseg7ei16_v_bf16mf2x7_tu(
+// CHECK-RV64-SAME: { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vluxseg7.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP7]]
+//
+vbfloat16mf2x7_t test_vluxseg7ei16_v_bf16mf2x7_tu(vbfloat16mf2x7_t vd, const __bf16 *rs1, vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vluxseg7ei16_v_bf16mf2x7_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vluxseg7ei16_v_bf16m1x7_tu(
+// CHECK-RV64-SAME: { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vluxseg7.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP7]]
+//
+vbfloat16m1x7_t test_vluxseg7ei16_v_bf16m1x7_tu(vbfloat16m1x7_t vd, const __bf16 *rs1, vuint16m1_t rs2, size_t vl) {
+  return __riscv_vluxseg7ei16_v_bf16m1x7_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vluxseg7ei16_v_bf16mf4x7_tum(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vluxseg7.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP7]]
+//
+vbfloat16mf4x7_t test_vluxseg7ei16_v_bf16mf4x7_tum(vbool64_t vm, vbfloat16mf4x7_t vd, const __bf16 *rs1, vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vluxseg7ei16_v_bf16mf4x7_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vluxseg7ei16_v_bf16mf2x7_tum(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vluxseg7.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP7]]
+//
+vbfloat16mf2x7_t test_vluxseg7ei16_v_bf16mf2x7_tum(vbool32_t vm, vbfloat16mf2x7_t vd, const __bf16 *rs1, vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vluxseg7ei16_v_bf16mf2x7_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vluxseg7ei16_v_bf16m1x7_tum(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vluxseg7.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP7]]
+//
+vbfloat16m1x7_t test_vluxseg7ei16_v_bf16m1x7_tum(vbool16_t vm, vbfloat16m1x7_t vd, const __bf16 *rs1, vuint16m1_t rs2, size_t vl) {
+  return __riscv_vluxseg7ei16_v_bf16m1x7_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vluxseg7ei16_v_bf16mf4x7_tumu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vluxseg7.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP7]]
+//
+vbfloat16mf4x7_t test_vluxseg7ei16_v_bf16mf4x7_tumu(vbool64_t vm, vbfloat16mf4x7_t vd, const __bf16 *rs1, vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vluxseg7ei16_v_bf16mf4x7_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vluxseg7ei16_v_bf16mf2x7_tumu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vluxseg7.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP7]]
+//
+vbfloat16mf2x7_t test_vluxseg7ei16_v_bf16mf2x7_tumu(vbool32_t vm, vbfloat16mf2x7_t vd, const __bf16 *rs1, vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vluxseg7ei16_v_bf16mf2x7_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vluxseg7ei16_v_bf16m1x7_tumu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vluxseg7.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP7]]
+//
+vbfloat16m1x7_t test_vluxseg7ei16_v_bf16m1x7_tumu(vbool16_t vm, vbfloat16m1x7_t vd, const __bf16 *rs1, vuint16m1_t rs2, size_t vl) {
+  return __riscv_vluxseg7ei16_v_bf16m1x7_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vluxseg7ei16_v_bf16mf4x7_mu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vluxseg7.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP7]]
+//
+vbfloat16mf4x7_t test_vluxseg7ei16_v_bf16mf4x7_mu(vbool64_t vm, vbfloat16mf4x7_t vd, const __bf16 *rs1, vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vluxseg7ei16_v_bf16mf4x7_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vluxseg7ei16_v_bf16mf2x7_mu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vluxseg7.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP7]]
+//
+vbfloat16mf2x7_t test_vluxseg7ei16_v_bf16mf2x7_mu(vbool32_t vm, vbfloat16mf2x7_t vd, const __bf16 *rs1, vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vluxseg7ei16_v_bf16mf2x7_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vluxseg7ei16_v_bf16m1x7_mu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vluxseg7.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP7]]
+//
+vbfloat16m1x7_t test_vluxseg7ei16_v_bf16m1x7_mu(vbool16_t vm, vbfloat16m1x7_t vd, const __bf16 *rs1, vuint16m1_t rs2, size_t vl) {
+  return __riscv_vluxseg7ei16_v_bf16m1x7_mu(vm, vd, rs1, rs2, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vluxseg8ei16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vluxseg8ei16.c
new file mode 100644
index 000000000000..17cb9dd26a75
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/bfloat16/vluxseg8ei16.c
@@ -0,0 +1,225 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vluxseg8ei16_v_bf16mf4x8_tu(
+// CHECK-RV64-SAME: { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vluxseg8.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], <vscale x 1 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP8]]
+//
+vbfloat16mf4x8_t test_vluxseg8ei16_v_bf16mf4x8_tu(vbfloat16mf4x8_t vd, const __bf16 *rs1, vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vluxseg8ei16_v_bf16mf4x8_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vluxseg8ei16_v_bf16mf2x8_tu(
+// CHECK-RV64-SAME: { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vluxseg8.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], <vscale x 2 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP8]]
+//
+vbfloat16mf2x8_t test_vluxseg8ei16_v_bf16mf2x8_tu(vbfloat16mf2x8_t vd, const __bf16 *rs1, vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vluxseg8ei16_v_bf16mf2x8_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vluxseg8ei16_v_bf16m1x8_tu(
+// CHECK-RV64-SAME: { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vluxseg8.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], <vscale x 4 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP8]]
+//
+vbfloat16m1x8_t test_vluxseg8ei16_v_bf16m1x8_tu(vbfloat16m1x8_t vd, const __bf16 *rs1, vuint16m1_t rs2, size_t vl) {
+  return __riscv_vluxseg8ei16_v_bf16m1x8_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vluxseg8ei16_v_bf16mf4x8_tum(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vluxseg8.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], <vscale x 1 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP8]]
+//
+vbfloat16mf4x8_t test_vluxseg8ei16_v_bf16mf4x8_tum(vbool64_t vm, vbfloat16mf4x8_t vd, const __bf16 *rs1, vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vluxseg8ei16_v_bf16mf4x8_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vluxseg8ei16_v_bf16mf2x8_tum(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vluxseg8.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], <vscale x 2 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP8]]
+//
+vbfloat16mf2x8_t test_vluxseg8ei16_v_bf16mf2x8_tum(vbool32_t vm, vbfloat16mf2x8_t vd, const __bf16 *rs1, vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vluxseg8ei16_v_bf16mf2x8_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vluxseg8ei16_v_bf16m1x8_tum(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vluxseg8.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], <vscale x 4 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP8]]
+//
+vbfloat16m1x8_t test_vluxseg8ei16_v_bf16m1x8_tum(vbool16_t vm, vbfloat16m1x8_t vd, const __bf16 *rs1, vuint16m1_t rs2, size_t vl) {
+  return __riscv_vluxseg8ei16_v_bf16m1x8_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vluxseg8ei16_v_bf16mf4x8_tumu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vluxseg8.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], <vscale x 1 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP8]]
+//
+vbfloat16mf4x8_t test_vluxseg8ei16_v_bf16mf4x8_tumu(vbool64_t vm, vbfloat16mf4x8_t vd, const __bf16 *rs1, vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vluxseg8ei16_v_bf16mf4x8_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vluxseg8ei16_v_bf16mf2x8_tumu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vluxseg8.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], <vscale x 2 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP8]]
+//
+vbfloat16mf2x8_t test_vluxseg8ei16_v_bf16mf2x8_tumu(vbool32_t vm, vbfloat16mf2x8_t vd, const __bf16 *rs1, vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vluxseg8ei16_v_bf16mf2x8_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vluxseg8ei16_v_bf16m1x8_tumu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vluxseg8.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], <vscale x 4 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP8]]
+//
+vbfloat16m1x8_t test_vluxseg8ei16_v_bf16m1x8_tumu(vbool16_t vm, vbfloat16m1x8_t vd, const __bf16 *rs1, vuint16m1_t rs2, size_t vl) {
+  return __riscv_vluxseg8ei16_v_bf16m1x8_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vluxseg8ei16_v_bf16mf4x8_mu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vluxseg8.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], <vscale x 1 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP8]]
+//
+vbfloat16mf4x8_t test_vluxseg8ei16_v_bf16mf4x8_mu(vbool64_t vm, vbfloat16mf4x8_t vd, const __bf16 *rs1, vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vluxseg8ei16_v_bf16mf4x8_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vluxseg8ei16_v_bf16mf2x8_mu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vluxseg8.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], <vscale x 2 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP8]]
+//
+vbfloat16mf2x8_t test_vluxseg8ei16_v_bf16mf2x8_mu(vbool32_t vm, vbfloat16mf2x8_t vd, const __bf16 *rs1, vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vluxseg8ei16_v_bf16mf2x8_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vluxseg8ei16_v_bf16m1x8_mu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vluxseg8.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], <vscale x 4 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP8]]
+//
+vbfloat16m1x8_t test_vluxseg8ei16_v_bf16m1x8_mu(vbool16_t vm, vbfloat16m1x8_t vd, const __bf16 *rs1, vuint16m1_t rs2, size_t vl) {
+  return __riscv_vluxseg8ei16_v_bf16m1x8_mu(vm, vd, rs1, rs2, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/vaesdf.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/vaesdf.c
index 8c18e323f7ce..27457c8f3af9 100644
--- a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/vaesdf.c
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/vaesdf.c
@@ -206,13 +206,3 @@ vuint32m8_t test_vaesdf_vv_u32m8_tu(vuint32m8_t vd, vuint32m8_t vs2, size_t vl)
   return __riscv_vaesdf_vv_u32m8_tu(vd, vs2, vl);
 }
 
-// CHECK-RV64-LABEL: define dso_local <vscale x 16 x i32> @test_vaesdf_vs_u32m8_u32m8_tu
-// CHECK-RV64-SAME: (<vscale x 16 x i32> [[VD:%.*]], <vscale x 16 x i32> [[VS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
-// CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vaesdf.vs.nxv16i32.nxv16i32.i64(<vscale x 16 x i32> [[VD]], <vscale x 16 x i32> [[VS2]], i64 [[VL]], i64 2)
-// CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
-//
-vuint32m8_t test_vaesdf_vs_u32m8_u32m8_tu(vuint32m8_t vd, vuint32m8_t vs2, size_t vl) {
-  return __riscv_vaesdf_vs_u32m8_u32m8_tu(vd, vs2, vl);
-}
-
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/vaesdm.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/vaesdm.c
index 7566be80e9d9..e23194eac3dc 100644
--- a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/vaesdm.c
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/vaesdm.c
@@ -206,13 +206,3 @@ vuint32m8_t test_vaesdm_vv_u32m8_tu(vuint32m8_t vd, vuint32m8_t vs2, size_t vl)
   return __riscv_vaesdm_vv_u32m8_tu(vd, vs2, vl);
 }
 
-// CHECK-RV64-LABEL: define dso_local <vscale x 16 x i32> @test_vaesdm_vs_u32m8_u32m8_tu
-// CHECK-RV64-SAME: (<vscale x 16 x i32> [[VD:%.*]], <vscale x 16 x i32> [[VS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
-// CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vaesdm.vs.nxv16i32.nxv16i32.i64(<vscale x 16 x i32> [[VD]], <vscale x 16 x i32> [[VS2]], i64 [[VL]], i64 2)
-// CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
-//
-vuint32m8_t test_vaesdm_vs_u32m8_u32m8_tu(vuint32m8_t vd, vuint32m8_t vs2, size_t vl) {
-  return __riscv_vaesdm_vs_u32m8_u32m8_tu(vd, vs2, vl);
-}
-
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/vaesef.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/vaesef.c
index ddeed6e78be1..4e99777b57fb 100644
--- a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/vaesef.c
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/vaesef.c
@@ -206,13 +206,3 @@ vuint32m8_t test_vaesef_vv_u32m8_tu(vuint32m8_t vd, vuint32m8_t vs2, size_t vl)
   return __riscv_vaesef_vv_u32m8_tu(vd, vs2, vl);
 }
 
-// CHECK-RV64-LABEL: define dso_local <vscale x 16 x i32> @test_vaesef_vs_u32m8_u32m8_tu
-// CHECK-RV64-SAME: (<vscale x 16 x i32> [[VD:%.*]], <vscale x 16 x i32> [[VS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
-// CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vaesef.vs.nxv16i32.nxv16i32.i64(<vscale x 16 x i32> [[VD]], <vscale x 16 x i32> [[VS2]], i64 [[VL]], i64 2)
-// CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
-//
-vuint32m8_t test_vaesef_vs_u32m8_u32m8_tu(vuint32m8_t vd, vuint32m8_t vs2, size_t vl) {
-  return __riscv_vaesef_vs_u32m8_u32m8_tu(vd, vs2, vl);
-}
-
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/vaesem.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/vaesem.c
index 2bd6350798f6..7488792effd4 100644
--- a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/vaesem.c
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/vaesem.c
@@ -206,13 +206,3 @@ vuint32m8_t test_vaesem_vv_u32m8_tu(vuint32m8_t vd, vuint32m8_t vs2, size_t vl)
   return __riscv_vaesem_vv_u32m8_tu(vd, vs2, vl);
 }
 
-// CHECK-RV64-LABEL: define dso_local <vscale x 16 x i32> @test_vaesem_vs_u32m8_u32m8_tu
-// CHECK-RV64-SAME: (<vscale x 16 x i32> [[VD:%.*]], <vscale x 16 x i32> [[VS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
-// CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vaesem.vs.nxv16i32.nxv16i32.i64(<vscale x 16 x i32> [[VD]], <vscale x 16 x i32> [[VS2]], i64 [[VL]], i64 2)
-// CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
-//
-vuint32m8_t test_vaesem_vs_u32m8_u32m8_tu(vuint32m8_t vd, vuint32m8_t vs2, size_t vl) {
-  return __riscv_vaesem_vs_u32m8_u32m8_tu(vd, vs2, vl);
-}
-
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/vaesz.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/vaesz.c
index 101efd751f86..08291a2cbca6 100644
--- a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/vaesz.c
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/vaesz.c
@@ -156,13 +156,3 @@ vuint32m8_t test_vaesz_vs_u32m4_u32m8_tu(vuint32m8_t vd, vuint32m4_t vs2, size_t
   return __riscv_vaesz_vs_u32m4_u32m8_tu(vd, vs2, vl);
 }
 
-// CHECK-RV64-LABEL: define dso_local <vscale x 16 x i32> @test_vaesz_vs_u32m8_u32m8_tu
-// CHECK-RV64-SAME: (<vscale x 16 x i32> [[VD:%.*]], <vscale x 16 x i32> [[VS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
-// CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vaesz.vs.nxv16i32.nxv16i32.i64(<vscale x 16 x i32> [[VD]], <vscale x 16 x i32> [[VS2]], i64 [[VL]], i64 2)
-// CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
-//
-vuint32m8_t test_vaesz_vs_u32m8_u32m8_tu(vuint32m8_t vd, vuint32m8_t vs2, size_t vl) {
-  return __riscv_vaesz_vs_u32m8_u32m8_tu(vd, vs2, vl);
-}
-
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/vsm4r.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/vsm4r.c
index c58f8d22d460..11faf7f36d23 100644
--- a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/vsm4r.c
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/non-overloaded/vsm4r.c
@@ -206,13 +206,3 @@ vuint32m8_t test_vsm4r_vv_u32m8_tu(vuint32m8_t vd, vuint32m8_t vs2, size_t vl) {
   return __riscv_vsm4r_vv_u32m8_tu(vd, vs2, vl);
 }
 
-// CHECK-RV64-LABEL: define dso_local <vscale x 16 x i32> @test_vsm4r_vs_u32m8_u32m8_tu
-// CHECK-RV64-SAME: (<vscale x 16 x i32> [[VD:%.*]], <vscale x 16 x i32> [[VS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
-// CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vsm4r.vs.nxv16i32.nxv16i32.i64(<vscale x 16 x i32> [[VD]], <vscale x 16 x i32> [[VS2]], i64 [[VL]], i64 2)
-// CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
-//
-vuint32m8_t test_vsm4r_vs_u32m8_u32m8_tu(vuint32m8_t vd, vuint32m8_t vs2, size_t vl) {
-  return __riscv_vsm4r_vs_u32m8_u32m8_tu(vd, vs2, vl);
-}
-
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vle16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vle16.c
new file mode 100644
index 000000000000..bc7996306cc3
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vle16.c
@@ -0,0 +1,273 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 1 x bfloat> @test_vle16_v_bf16mf4_tu(
+// CHECK-RV64-SAME: <vscale x 1 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x bfloat> @llvm.riscv.vle.nxv1bf16.i64(<vscale x 1 x bfloat> [[VD]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret <vscale x 1 x bfloat> [[TMP0]]
+//
+vbfloat16mf4_t test_vle16_v_bf16mf4_tu(vbfloat16mf4_t vd, const __bf16 *rs1,
+                                       size_t vl) {
+  return __riscv_vle16_tu(vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 2 x bfloat> @test_vle16_v_bf16mf2_tu(
+// CHECK-RV64-SAME: <vscale x 2 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x bfloat> @llvm.riscv.vle.nxv2bf16.i64(<vscale x 2 x bfloat> [[VD]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret <vscale x 2 x bfloat> [[TMP0]]
+//
+vbfloat16mf2_t test_vle16_v_bf16mf2_tu(vbfloat16mf2_t vd, const __bf16 *rs1,
+                                       size_t vl) {
+  return __riscv_vle16_tu(vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 4 x bfloat> @test_vle16_v_bf16m1_tu(
+// CHECK-RV64-SAME: <vscale x 4 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x bfloat> @llvm.riscv.vle.nxv4bf16.i64(<vscale x 4 x bfloat> [[VD]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret <vscale x 4 x bfloat> [[TMP0]]
+//
+vbfloat16m1_t test_vle16_v_bf16m1_tu(vbfloat16m1_t vd, const __bf16 *rs1,
+                                     size_t vl) {
+  return __riscv_vle16_tu(vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 8 x bfloat> @test_vle16_v_bf16m2_tu(
+// CHECK-RV64-SAME: <vscale x 8 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x bfloat> @llvm.riscv.vle.nxv8bf16.i64(<vscale x 8 x bfloat> [[VD]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret <vscale x 8 x bfloat> [[TMP0]]
+//
+vbfloat16m2_t test_vle16_v_bf16m2_tu(vbfloat16m2_t vd, const __bf16 *rs1,
+                                     size_t vl) {
+  return __riscv_vle16_tu(vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 16 x bfloat> @test_vle16_v_bf16m4_tu(
+// CHECK-RV64-SAME: <vscale x 16 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x bfloat> @llvm.riscv.vle.nxv16bf16.i64(<vscale x 16 x bfloat> [[VD]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret <vscale x 16 x bfloat> [[TMP0]]
+//
+vbfloat16m4_t test_vle16_v_bf16m4_tu(vbfloat16m4_t vd, const __bf16 *rs1,
+                                     size_t vl) {
+  return __riscv_vle16_tu(vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 32 x bfloat> @test_vle16_v_bf16m8_tu(
+// CHECK-RV64-SAME: <vscale x 32 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x bfloat> @llvm.riscv.vle.nxv32bf16.i64(<vscale x 32 x bfloat> [[VD]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret <vscale x 32 x bfloat> [[TMP0]]
+//
+vbfloat16m8_t test_vle16_v_bf16m8_tu(vbfloat16m8_t vd, const __bf16 *rs1,
+                                     size_t vl) {
+  return __riscv_vle16_tu(vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 1 x bfloat> @test_vle16_v_bf16mf4_tum(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], <vscale x 1 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x bfloat> @llvm.riscv.vle.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[VD]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret <vscale x 1 x bfloat> [[TMP0]]
+//
+vbfloat16mf4_t test_vle16_v_bf16mf4_tum(vbool64_t vm, vbfloat16mf4_t vd,
+                                        const __bf16 *rs1, size_t vl) {
+  return __riscv_vle16_tum(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 2 x bfloat> @test_vle16_v_bf16mf2_tum(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], <vscale x 2 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x bfloat> @llvm.riscv.vle.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[VD]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret <vscale x 2 x bfloat> [[TMP0]]
+//
+vbfloat16mf2_t test_vle16_v_bf16mf2_tum(vbool32_t vm, vbfloat16mf2_t vd,
+                                        const __bf16 *rs1, size_t vl) {
+  return __riscv_vle16_tum(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 4 x bfloat> @test_vle16_v_bf16m1_tum(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], <vscale x 4 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x bfloat> @llvm.riscv.vle.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[VD]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret <vscale x 4 x bfloat> [[TMP0]]
+//
+vbfloat16m1_t test_vle16_v_bf16m1_tum(vbool16_t vm, vbfloat16m1_t vd,
+                                      const __bf16 *rs1, size_t vl) {
+  return __riscv_vle16_tum(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 8 x bfloat> @test_vle16_v_bf16m2_tum(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], <vscale x 8 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x bfloat> @llvm.riscv.vle.mask.nxv8bf16.i64(<vscale x 8 x bfloat> [[VD]], ptr [[RS1]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret <vscale x 8 x bfloat> [[TMP0]]
+//
+vbfloat16m2_t test_vle16_v_bf16m2_tum(vbool8_t vm, vbfloat16m2_t vd,
+                                      const __bf16 *rs1, size_t vl) {
+  return __riscv_vle16_tum(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 16 x bfloat> @test_vle16_v_bf16m4_tum(
+// CHECK-RV64-SAME: <vscale x 16 x i1> [[VM:%.*]], <vscale x 16 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x bfloat> @llvm.riscv.vle.mask.nxv16bf16.i64(<vscale x 16 x bfloat> [[VD]], ptr [[RS1]], <vscale x 16 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret <vscale x 16 x bfloat> [[TMP0]]
+//
+vbfloat16m4_t test_vle16_v_bf16m4_tum(vbool4_t vm, vbfloat16m4_t vd,
+                                      const __bf16 *rs1, size_t vl) {
+  return __riscv_vle16_tum(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 32 x bfloat> @test_vle16_v_bf16m8_tum(
+// CHECK-RV64-SAME: <vscale x 32 x i1> [[VM:%.*]], <vscale x 32 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x bfloat> @llvm.riscv.vle.mask.nxv32bf16.i64(<vscale x 32 x bfloat> [[VD]], ptr [[RS1]], <vscale x 32 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret <vscale x 32 x bfloat> [[TMP0]]
+//
+vbfloat16m8_t test_vle16_v_bf16m8_tum(vbool2_t vm, vbfloat16m8_t vd,
+                                      const __bf16 *rs1, size_t vl) {
+  return __riscv_vle16_tum(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 1 x bfloat> @test_vle16_v_bf16mf4_tumu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], <vscale x 1 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x bfloat> @llvm.riscv.vle.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[VD]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 1 x bfloat> [[TMP0]]
+//
+vbfloat16mf4_t test_vle16_v_bf16mf4_tumu(vbool64_t vm, vbfloat16mf4_t vd,
+                                         const __bf16 *rs1, size_t vl) {
+  return __riscv_vle16_tumu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 2 x bfloat> @test_vle16_v_bf16mf2_tumu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], <vscale x 2 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x bfloat> @llvm.riscv.vle.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[VD]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 2 x bfloat> [[TMP0]]
+//
+vbfloat16mf2_t test_vle16_v_bf16mf2_tumu(vbool32_t vm, vbfloat16mf2_t vd,
+                                         const __bf16 *rs1, size_t vl) {
+  return __riscv_vle16_tumu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 4 x bfloat> @test_vle16_v_bf16m1_tumu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], <vscale x 4 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x bfloat> @llvm.riscv.vle.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[VD]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 4 x bfloat> [[TMP0]]
+//
+vbfloat16m1_t test_vle16_v_bf16m1_tumu(vbool16_t vm, vbfloat16m1_t vd,
+                                       const __bf16 *rs1, size_t vl) {
+  return __riscv_vle16_tumu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 8 x bfloat> @test_vle16_v_bf16m2_tumu(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], <vscale x 8 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x bfloat> @llvm.riscv.vle.mask.nxv8bf16.i64(<vscale x 8 x bfloat> [[VD]], ptr [[RS1]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 8 x bfloat> [[TMP0]]
+//
+vbfloat16m2_t test_vle16_v_bf16m2_tumu(vbool8_t vm, vbfloat16m2_t vd,
+                                       const __bf16 *rs1, size_t vl) {
+  return __riscv_vle16_tumu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 16 x bfloat> @test_vle16_v_bf16m4_tumu(
+// CHECK-RV64-SAME: <vscale x 16 x i1> [[VM:%.*]], <vscale x 16 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x bfloat> @llvm.riscv.vle.mask.nxv16bf16.i64(<vscale x 16 x bfloat> [[VD]], ptr [[RS1]], <vscale x 16 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 16 x bfloat> [[TMP0]]
+//
+vbfloat16m4_t test_vle16_v_bf16m4_tumu(vbool4_t vm, vbfloat16m4_t vd,
+                                       const __bf16 *rs1, size_t vl) {
+  return __riscv_vle16_tumu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 32 x bfloat> @test_vle16_v_bf16m8_tumu(
+// CHECK-RV64-SAME: <vscale x 32 x i1> [[VM:%.*]], <vscale x 32 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x bfloat> @llvm.riscv.vle.mask.nxv32bf16.i64(<vscale x 32 x bfloat> [[VD]], ptr [[RS1]], <vscale x 32 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 32 x bfloat> [[TMP0]]
+//
+vbfloat16m8_t test_vle16_v_bf16m8_tumu(vbool2_t vm, vbfloat16m8_t vd,
+                                       const __bf16 *rs1, size_t vl) {
+  return __riscv_vle16_tumu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 1 x bfloat> @test_vle16_v_bf16mf4_mu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], <vscale x 1 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x bfloat> @llvm.riscv.vle.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[VD]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret <vscale x 1 x bfloat> [[TMP0]]
+//
+vbfloat16mf4_t test_vle16_v_bf16mf4_mu(vbool64_t vm, vbfloat16mf4_t vd,
+                                       const __bf16 *rs1, size_t vl) {
+  return __riscv_vle16_mu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 2 x bfloat> @test_vle16_v_bf16mf2_mu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], <vscale x 2 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x bfloat> @llvm.riscv.vle.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[VD]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret <vscale x 2 x bfloat> [[TMP0]]
+//
+vbfloat16mf2_t test_vle16_v_bf16mf2_mu(vbool32_t vm, vbfloat16mf2_t vd,
+                                       const __bf16 *rs1, size_t vl) {
+  return __riscv_vle16_mu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 4 x bfloat> @test_vle16_v_bf16m1_mu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], <vscale x 4 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x bfloat> @llvm.riscv.vle.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[VD]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret <vscale x 4 x bfloat> [[TMP0]]
+//
+vbfloat16m1_t test_vle16_v_bf16m1_mu(vbool16_t vm, vbfloat16m1_t vd,
+                                     const __bf16 *rs1, size_t vl) {
+  return __riscv_vle16_mu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 8 x bfloat> @test_vle16_v_bf16m2_mu(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], <vscale x 8 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x bfloat> @llvm.riscv.vle.mask.nxv8bf16.i64(<vscale x 8 x bfloat> [[VD]], ptr [[RS1]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret <vscale x 8 x bfloat> [[TMP0]]
+//
+vbfloat16m2_t test_vle16_v_bf16m2_mu(vbool8_t vm, vbfloat16m2_t vd,
+                                     const __bf16 *rs1, size_t vl) {
+  return __riscv_vle16_mu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 16 x bfloat> @test_vle16_v_bf16m4_mu(
+// CHECK-RV64-SAME: <vscale x 16 x i1> [[VM:%.*]], <vscale x 16 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x bfloat> @llvm.riscv.vle.mask.nxv16bf16.i64(<vscale x 16 x bfloat> [[VD]], ptr [[RS1]], <vscale x 16 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret <vscale x 16 x bfloat> [[TMP0]]
+//
+vbfloat16m4_t test_vle16_v_bf16m4_mu(vbool4_t vm, vbfloat16m4_t vd,
+                                     const __bf16 *rs1, size_t vl) {
+  return __riscv_vle16_mu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 32 x bfloat> @test_vle16_v_bf16m8_mu(
+// CHECK-RV64-SAME: <vscale x 32 x i1> [[VM:%.*]], <vscale x 32 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x bfloat> @llvm.riscv.vle.mask.nxv32bf16.i64(<vscale x 32 x bfloat> [[VD]], ptr [[RS1]], <vscale x 32 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret <vscale x 32 x bfloat> [[TMP0]]
+//
+vbfloat16m8_t test_vle16_v_bf16m8_mu(vbool2_t vm, vbfloat16m8_t vd,
+                                     const __bf16 *rs1, size_t vl) {
+  return __riscv_vle16_mu(vm, vd, rs1, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vle16ff.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vle16ff.c
new file mode 100644
index 000000000000..75ad618e4262
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vle16ff.c
@@ -0,0 +1,363 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 1 x bfloat> @test_vle16ff_v_bf16mf4_tu(
+// CHECK-RV64-SAME: <vscale x 1 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, i64 } @llvm.riscv.vleff.nxv1bf16.i64(<vscale x 1 x bfloat> [[VD]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    store i64 [[TMP2]], ptr [[NEW_VL]], align 2
+// CHECK-RV64-NEXT:    ret <vscale x 1 x bfloat> [[TMP1]]
+//
+vbfloat16mf4_t test_vle16ff_v_bf16mf4_tu(vbfloat16mf4_t vd, const __bf16 *rs1,
+                                         size_t *new_vl, size_t vl) {
+  return __riscv_vle16ff_tu(vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 2 x bfloat> @test_vle16ff_v_bf16mf2_tu(
+// CHECK-RV64-SAME: <vscale x 2 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, i64 } @llvm.riscv.vleff.nxv2bf16.i64(<vscale x 2 x bfloat> [[VD]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    store i64 [[TMP2]], ptr [[NEW_VL]], align 2
+// CHECK-RV64-NEXT:    ret <vscale x 2 x bfloat> [[TMP1]]
+//
+vbfloat16mf2_t test_vle16ff_v_bf16mf2_tu(vbfloat16mf2_t vd, const __bf16 *rs1,
+                                         size_t *new_vl, size_t vl) {
+  return __riscv_vle16ff_tu(vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 4 x bfloat> @test_vle16ff_v_bf16m1_tu(
+// CHECK-RV64-SAME: <vscale x 4 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, i64 } @llvm.riscv.vleff.nxv4bf16.i64(<vscale x 4 x bfloat> [[VD]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    store i64 [[TMP2]], ptr [[NEW_VL]], align 2
+// CHECK-RV64-NEXT:    ret <vscale x 4 x bfloat> [[TMP1]]
+//
+vbfloat16m1_t test_vle16ff_v_bf16m1_tu(vbfloat16m1_t vd, const __bf16 *rs1,
+                                       size_t *new_vl, size_t vl) {
+  return __riscv_vle16ff_tu(vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 8 x bfloat> @test_vle16ff_v_bf16m2_tu(
+// CHECK-RV64-SAME: <vscale x 8 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 8 x bfloat>, i64 } @llvm.riscv.vleff.nxv8bf16.i64(<vscale x 8 x bfloat> [[VD]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    store i64 [[TMP2]], ptr [[NEW_VL]], align 2
+// CHECK-RV64-NEXT:    ret <vscale x 8 x bfloat> [[TMP1]]
+//
+vbfloat16m2_t test_vle16ff_v_bf16m2_tu(vbfloat16m2_t vd, const __bf16 *rs1,
+                                       size_t *new_vl, size_t vl) {
+  return __riscv_vle16ff_tu(vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 16 x bfloat> @test_vle16ff_v_bf16m4_tu(
+// CHECK-RV64-SAME: <vscale x 16 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 16 x bfloat>, i64 } @llvm.riscv.vleff.nxv16bf16.i64(<vscale x 16 x bfloat> [[VD]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 16 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 16 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    store i64 [[TMP2]], ptr [[NEW_VL]], align 2
+// CHECK-RV64-NEXT:    ret <vscale x 16 x bfloat> [[TMP1]]
+//
+vbfloat16m4_t test_vle16ff_v_bf16m4_tu(vbfloat16m4_t vd, const __bf16 *rs1,
+                                       size_t *new_vl, size_t vl) {
+  return __riscv_vle16ff_tu(vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 32 x bfloat> @test_vle16ff_v_bf16m8_tu(
+// CHECK-RV64-SAME: <vscale x 32 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 32 x bfloat>, i64 } @llvm.riscv.vleff.nxv32bf16.i64(<vscale x 32 x bfloat> [[VD]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 32 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 32 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    store i64 [[TMP2]], ptr [[NEW_VL]], align 2
+// CHECK-RV64-NEXT:    ret <vscale x 32 x bfloat> [[TMP1]]
+//
+vbfloat16m8_t test_vle16ff_v_bf16m8_tu(vbfloat16m8_t vd, const __bf16 *rs1,
+                                       size_t *new_vl, size_t vl) {
+  return __riscv_vle16ff_tu(vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 1 x bfloat> @test_vle16ff_v_bf16mf4_tum(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], <vscale x 1 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, i64 } @llvm.riscv.vleff.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[VD]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    store i64 [[TMP2]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret <vscale x 1 x bfloat> [[TMP1]]
+//
+vbfloat16mf4_t test_vle16ff_v_bf16mf4_tum(vbool64_t vm, vbfloat16mf4_t vd,
+                                          const __bf16 *rs1, size_t *new_vl,
+                                          size_t vl) {
+  return __riscv_vle16ff_tum(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 2 x bfloat> @test_vle16ff_v_bf16mf2_tum(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], <vscale x 2 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, i64 } @llvm.riscv.vleff.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[VD]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    store i64 [[TMP2]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret <vscale x 2 x bfloat> [[TMP1]]
+//
+vbfloat16mf2_t test_vle16ff_v_bf16mf2_tum(vbool32_t vm, vbfloat16mf2_t vd,
+                                          const __bf16 *rs1, size_t *new_vl,
+                                          size_t vl) {
+  return __riscv_vle16ff_tum(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 4 x bfloat> @test_vle16ff_v_bf16m1_tum(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], <vscale x 4 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, i64 } @llvm.riscv.vleff.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[VD]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    store i64 [[TMP2]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret <vscale x 4 x bfloat> [[TMP1]]
+//
+vbfloat16m1_t test_vle16ff_v_bf16m1_tum(vbool16_t vm, vbfloat16m1_t vd,
+                                        const __bf16 *rs1, size_t *new_vl,
+                                        size_t vl) {
+  return __riscv_vle16ff_tum(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 8 x bfloat> @test_vle16ff_v_bf16m2_tum(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], <vscale x 8 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 8 x bfloat>, i64 } @llvm.riscv.vleff.mask.nxv8bf16.i64(<vscale x 8 x bfloat> [[VD]], ptr [[RS1]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    store i64 [[TMP2]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret <vscale x 8 x bfloat> [[TMP1]]
+//
+vbfloat16m2_t test_vle16ff_v_bf16m2_tum(vbool8_t vm, vbfloat16m2_t vd,
+                                        const __bf16 *rs1, size_t *new_vl,
+                                        size_t vl) {
+  return __riscv_vle16ff_tum(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 16 x bfloat> @test_vle16ff_v_bf16m4_tum(
+// CHECK-RV64-SAME: <vscale x 16 x i1> [[VM:%.*]], <vscale x 16 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 16 x bfloat>, i64 } @llvm.riscv.vleff.mask.nxv16bf16.i64(<vscale x 16 x bfloat> [[VD]], ptr [[RS1]], <vscale x 16 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 16 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 16 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    store i64 [[TMP2]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret <vscale x 16 x bfloat> [[TMP1]]
+//
+vbfloat16m4_t test_vle16ff_v_bf16m4_tum(vbool4_t vm, vbfloat16m4_t vd,
+                                        const __bf16 *rs1, size_t *new_vl,
+                                        size_t vl) {
+  return __riscv_vle16ff_tum(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 32 x bfloat> @test_vle16ff_v_bf16m8_tum(
+// CHECK-RV64-SAME: <vscale x 32 x i1> [[VM:%.*]], <vscale x 32 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 32 x bfloat>, i64 } @llvm.riscv.vleff.mask.nxv32bf16.i64(<vscale x 32 x bfloat> [[VD]], ptr [[RS1]], <vscale x 32 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 32 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 32 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    store i64 [[TMP2]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret <vscale x 32 x bfloat> [[TMP1]]
+//
+vbfloat16m8_t test_vle16ff_v_bf16m8_tum(vbool2_t vm, vbfloat16m8_t vd,
+                                        const __bf16 *rs1, size_t *new_vl,
+                                        size_t vl) {
+  return __riscv_vle16ff_tum(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 1 x bfloat> @test_vle16ff_v_bf16mf4_tumu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], <vscale x 1 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, i64 } @llvm.riscv.vleff.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[VD]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    store i64 [[TMP2]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret <vscale x 1 x bfloat> [[TMP1]]
+//
+vbfloat16mf4_t test_vle16ff_v_bf16mf4_tumu(vbool64_t vm, vbfloat16mf4_t vd,
+                                           const __bf16 *rs1, size_t *new_vl,
+                                           size_t vl) {
+  return __riscv_vle16ff_tumu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 2 x bfloat> @test_vle16ff_v_bf16mf2_tumu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], <vscale x 2 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, i64 } @llvm.riscv.vleff.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[VD]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    store i64 [[TMP2]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret <vscale x 2 x bfloat> [[TMP1]]
+//
+vbfloat16mf2_t test_vle16ff_v_bf16mf2_tumu(vbool32_t vm, vbfloat16mf2_t vd,
+                                           const __bf16 *rs1, size_t *new_vl,
+                                           size_t vl) {
+  return __riscv_vle16ff_tumu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 4 x bfloat> @test_vle16ff_v_bf16m1_tumu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], <vscale x 4 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, i64 } @llvm.riscv.vleff.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[VD]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    store i64 [[TMP2]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret <vscale x 4 x bfloat> [[TMP1]]
+//
+vbfloat16m1_t test_vle16ff_v_bf16m1_tumu(vbool16_t vm, vbfloat16m1_t vd,
+                                         const __bf16 *rs1, size_t *new_vl,
+                                         size_t vl) {
+  return __riscv_vle16ff_tumu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 8 x bfloat> @test_vle16ff_v_bf16m2_tumu(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], <vscale x 8 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 8 x bfloat>, i64 } @llvm.riscv.vleff.mask.nxv8bf16.i64(<vscale x 8 x bfloat> [[VD]], ptr [[RS1]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    store i64 [[TMP2]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret <vscale x 8 x bfloat> [[TMP1]]
+//
+vbfloat16m2_t test_vle16ff_v_bf16m2_tumu(vbool8_t vm, vbfloat16m2_t vd,
+                                         const __bf16 *rs1, size_t *new_vl,
+                                         size_t vl) {
+  return __riscv_vle16ff_tumu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 16 x bfloat> @test_vle16ff_v_bf16m4_tumu(
+// CHECK-RV64-SAME: <vscale x 16 x i1> [[VM:%.*]], <vscale x 16 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 16 x bfloat>, i64 } @llvm.riscv.vleff.mask.nxv16bf16.i64(<vscale x 16 x bfloat> [[VD]], ptr [[RS1]], <vscale x 16 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 16 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 16 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    store i64 [[TMP2]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret <vscale x 16 x bfloat> [[TMP1]]
+//
+vbfloat16m4_t test_vle16ff_v_bf16m4_tumu(vbool4_t vm, vbfloat16m4_t vd,
+                                         const __bf16 *rs1, size_t *new_vl,
+                                         size_t vl) {
+  return __riscv_vle16ff_tumu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 32 x bfloat> @test_vle16ff_v_bf16m8_tumu(
+// CHECK-RV64-SAME: <vscale x 32 x i1> [[VM:%.*]], <vscale x 32 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 32 x bfloat>, i64 } @llvm.riscv.vleff.mask.nxv32bf16.i64(<vscale x 32 x bfloat> [[VD]], ptr [[RS1]], <vscale x 32 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 32 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 32 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    store i64 [[TMP2]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret <vscale x 32 x bfloat> [[TMP1]]
+//
+vbfloat16m8_t test_vle16ff_v_bf16m8_tumu(vbool2_t vm, vbfloat16m8_t vd,
+                                         const __bf16 *rs1, size_t *new_vl,
+                                         size_t vl) {
+  return __riscv_vle16ff_tumu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 1 x bfloat> @test_vle16ff_v_bf16mf4_mu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], <vscale x 1 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 1 x bfloat>, i64 } @llvm.riscv.vleff.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[VD]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    store i64 [[TMP2]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret <vscale x 1 x bfloat> [[TMP1]]
+//
+vbfloat16mf4_t test_vle16ff_v_bf16mf4_mu(vbool64_t vm, vbfloat16mf4_t vd,
+                                         const __bf16 *rs1, size_t *new_vl,
+                                         size_t vl) {
+  return __riscv_vle16ff_mu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 2 x bfloat> @test_vle16ff_v_bf16mf2_mu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], <vscale x 2 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 2 x bfloat>, i64 } @llvm.riscv.vleff.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[VD]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    store i64 [[TMP2]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret <vscale x 2 x bfloat> [[TMP1]]
+//
+vbfloat16mf2_t test_vle16ff_v_bf16mf2_mu(vbool32_t vm, vbfloat16mf2_t vd,
+                                         const __bf16 *rs1, size_t *new_vl,
+                                         size_t vl) {
+  return __riscv_vle16ff_mu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 4 x bfloat> @test_vle16ff_v_bf16m1_mu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], <vscale x 4 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 4 x bfloat>, i64 } @llvm.riscv.vleff.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[VD]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    store i64 [[TMP2]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret <vscale x 4 x bfloat> [[TMP1]]
+//
+vbfloat16m1_t test_vle16ff_v_bf16m1_mu(vbool16_t vm, vbfloat16m1_t vd,
+                                       const __bf16 *rs1, size_t *new_vl,
+                                       size_t vl) {
+  return __riscv_vle16ff_mu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 8 x bfloat> @test_vle16ff_v_bf16m2_mu(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], <vscale x 8 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 8 x bfloat>, i64 } @llvm.riscv.vleff.mask.nxv8bf16.i64(<vscale x 8 x bfloat> [[VD]], ptr [[RS1]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    store i64 [[TMP2]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret <vscale x 8 x bfloat> [[TMP1]]
+//
+vbfloat16m2_t test_vle16ff_v_bf16m2_mu(vbool8_t vm, vbfloat16m2_t vd,
+                                       const __bf16 *rs1, size_t *new_vl,
+                                       size_t vl) {
+  return __riscv_vle16ff_mu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 16 x bfloat> @test_vle16ff_v_bf16m4_mu(
+// CHECK-RV64-SAME: <vscale x 16 x i1> [[VM:%.*]], <vscale x 16 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 16 x bfloat>, i64 } @llvm.riscv.vleff.mask.nxv16bf16.i64(<vscale x 16 x bfloat> [[VD]], ptr [[RS1]], <vscale x 16 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 16 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 16 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    store i64 [[TMP2]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret <vscale x 16 x bfloat> [[TMP1]]
+//
+vbfloat16m4_t test_vle16ff_v_bf16m4_mu(vbool4_t vm, vbfloat16m4_t vd,
+                                       const __bf16 *rs1, size_t *new_vl,
+                                       size_t vl) {
+  return __riscv_vle16ff_mu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 32 x bfloat> @test_vle16ff_v_bf16m8_mu(
+// CHECK-RV64-SAME: <vscale x 32 x i1> [[VM:%.*]], <vscale x 32 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call { <vscale x 32 x bfloat>, i64 } @llvm.riscv.vleff.mask.nxv32bf16.i64(<vscale x 32 x bfloat> [[VD]], ptr [[RS1]], <vscale x 32 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 32 x bfloat>, i64 } [[TMP0]], 0
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 32 x bfloat>, i64 } [[TMP0]], 1
+// CHECK-RV64-NEXT:    store i64 [[TMP2]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret <vscale x 32 x bfloat> [[TMP1]]
+//
+vbfloat16m8_t test_vle16ff_v_bf16m8_mu(vbool2_t vm, vbfloat16m8_t vd,
+                                       const __bf16 *rs1, size_t *new_vl,
+                                       size_t vl) {
+  return __riscv_vle16ff_mu(vm, vd, rs1, new_vl, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vloxei16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vloxei16.c
new file mode 100644
index 000000000000..389031e75c74
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vloxei16.c
@@ -0,0 +1,291 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 1 x bfloat> @test_vloxei16_v_bf16mf4_tu(
+// CHECK-RV64-SAME: <vscale x 1 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x bfloat> @llvm.riscv.vloxei.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[VD]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret <vscale x 1 x bfloat> [[TMP0]]
+//
+vbfloat16mf4_t test_vloxei16_v_bf16mf4_tu(vbfloat16mf4_t vd, const __bf16 *rs1,
+                                          vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vloxei16_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 2 x bfloat> @test_vloxei16_v_bf16mf2_tu(
+// CHECK-RV64-SAME: <vscale x 2 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x bfloat> @llvm.riscv.vloxei.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[VD]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret <vscale x 2 x bfloat> [[TMP0]]
+//
+vbfloat16mf2_t test_vloxei16_v_bf16mf2_tu(vbfloat16mf2_t vd, const __bf16 *rs1,
+                                          vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vloxei16_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 4 x bfloat> @test_vloxei16_v_bf16m1_tu(
+// CHECK-RV64-SAME: <vscale x 4 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x bfloat> @llvm.riscv.vloxei.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[VD]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret <vscale x 4 x bfloat> [[TMP0]]
+//
+vbfloat16m1_t test_vloxei16_v_bf16m1_tu(vbfloat16m1_t vd, const __bf16 *rs1,
+                                        vuint16m1_t rs2, size_t vl) {
+  return __riscv_vloxei16_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 8 x bfloat> @test_vloxei16_v_bf16m2_tu(
+// CHECK-RV64-SAME: <vscale x 8 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x bfloat> @llvm.riscv.vloxei.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> [[VD]], ptr [[RS1]], <vscale x 8 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret <vscale x 8 x bfloat> [[TMP0]]
+//
+vbfloat16m2_t test_vloxei16_v_bf16m2_tu(vbfloat16m2_t vd, const __bf16 *rs1,
+                                        vuint16m2_t rs2, size_t vl) {
+  return __riscv_vloxei16_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 16 x bfloat> @test_vloxei16_v_bf16m4_tu(
+// CHECK-RV64-SAME: <vscale x 16 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 16 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x bfloat> @llvm.riscv.vloxei.nxv16bf16.nxv16i16.i64(<vscale x 16 x bfloat> [[VD]], ptr [[RS1]], <vscale x 16 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret <vscale x 16 x bfloat> [[TMP0]]
+//
+vbfloat16m4_t test_vloxei16_v_bf16m4_tu(vbfloat16m4_t vd, const __bf16 *rs1,
+                                        vuint16m4_t rs2, size_t vl) {
+  return __riscv_vloxei16_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 32 x bfloat> @test_vloxei16_v_bf16m8_tu(
+// CHECK-RV64-SAME: <vscale x 32 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 32 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x bfloat> @llvm.riscv.vloxei.nxv32bf16.nxv32i16.i64(<vscale x 32 x bfloat> [[VD]], ptr [[RS1]], <vscale x 32 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret <vscale x 32 x bfloat> [[TMP0]]
+//
+vbfloat16m8_t test_vloxei16_v_bf16m8_tu(vbfloat16m8_t vd, const __bf16 *rs1,
+                                        vuint16m8_t rs2, size_t vl) {
+  return __riscv_vloxei16_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 1 x bfloat> @test_vloxei16_v_bf16mf4_tum(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], <vscale x 1 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x bfloat> @llvm.riscv.vloxei.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[VD]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret <vscale x 1 x bfloat> [[TMP0]]
+//
+vbfloat16mf4_t test_vloxei16_v_bf16mf4_tum(vbool64_t vm, vbfloat16mf4_t vd,
+                                           const __bf16 *rs1, vuint16mf4_t rs2,
+                                           size_t vl) {
+  return __riscv_vloxei16_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 2 x bfloat> @test_vloxei16_v_bf16mf2_tum(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], <vscale x 2 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x bfloat> @llvm.riscv.vloxei.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[VD]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret <vscale x 2 x bfloat> [[TMP0]]
+//
+vbfloat16mf2_t test_vloxei16_v_bf16mf2_tum(vbool32_t vm, vbfloat16mf2_t vd,
+                                           const __bf16 *rs1, vuint16mf2_t rs2,
+                                           size_t vl) {
+  return __riscv_vloxei16_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 4 x bfloat> @test_vloxei16_v_bf16m1_tum(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], <vscale x 4 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x bfloat> @llvm.riscv.vloxei.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[VD]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret <vscale x 4 x bfloat> [[TMP0]]
+//
+vbfloat16m1_t test_vloxei16_v_bf16m1_tum(vbool16_t vm, vbfloat16m1_t vd,
+                                         const __bf16 *rs1, vuint16m1_t rs2,
+                                         size_t vl) {
+  return __riscv_vloxei16_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 8 x bfloat> @test_vloxei16_v_bf16m2_tum(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], <vscale x 8 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x bfloat> @llvm.riscv.vloxei.mask.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> [[VD]], ptr [[RS1]], <vscale x 8 x i16> [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret <vscale x 8 x bfloat> [[TMP0]]
+//
+vbfloat16m2_t test_vloxei16_v_bf16m2_tum(vbool8_t vm, vbfloat16m2_t vd,
+                                         const __bf16 *rs1, vuint16m2_t rs2,
+                                         size_t vl) {
+  return __riscv_vloxei16_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 16 x bfloat> @test_vloxei16_v_bf16m4_tum(
+// CHECK-RV64-SAME: <vscale x 16 x i1> [[VM:%.*]], <vscale x 16 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 16 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x bfloat> @llvm.riscv.vloxei.mask.nxv16bf16.nxv16i16.i64(<vscale x 16 x bfloat> [[VD]], ptr [[RS1]], <vscale x 16 x i16> [[RS2]], <vscale x 16 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret <vscale x 16 x bfloat> [[TMP0]]
+//
+vbfloat16m4_t test_vloxei16_v_bf16m4_tum(vbool4_t vm, vbfloat16m4_t vd,
+                                         const __bf16 *rs1, vuint16m4_t rs2,
+                                         size_t vl) {
+  return __riscv_vloxei16_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 32 x bfloat> @test_vloxei16_v_bf16m8_tum(
+// CHECK-RV64-SAME: <vscale x 32 x i1> [[VM:%.*]], <vscale x 32 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 32 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x bfloat> @llvm.riscv.vloxei.mask.nxv32bf16.nxv32i16.i64(<vscale x 32 x bfloat> [[VD]], ptr [[RS1]], <vscale x 32 x i16> [[RS2]], <vscale x 32 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret <vscale x 32 x bfloat> [[TMP0]]
+//
+vbfloat16m8_t test_vloxei16_v_bf16m8_tum(vbool2_t vm, vbfloat16m8_t vd,
+                                         const __bf16 *rs1, vuint16m8_t rs2,
+                                         size_t vl) {
+  return __riscv_vloxei16_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 1 x bfloat> @test_vloxei16_v_bf16mf4_tumu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], <vscale x 1 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x bfloat> @llvm.riscv.vloxei.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[VD]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 1 x bfloat> [[TMP0]]
+//
+vbfloat16mf4_t test_vloxei16_v_bf16mf4_tumu(vbool64_t vm, vbfloat16mf4_t vd,
+                                            const __bf16 *rs1, vuint16mf4_t rs2,
+                                            size_t vl) {
+  return __riscv_vloxei16_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 2 x bfloat> @test_vloxei16_v_bf16mf2_tumu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], <vscale x 2 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x bfloat> @llvm.riscv.vloxei.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[VD]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 2 x bfloat> [[TMP0]]
+//
+vbfloat16mf2_t test_vloxei16_v_bf16mf2_tumu(vbool32_t vm, vbfloat16mf2_t vd,
+                                            const __bf16 *rs1, vuint16mf2_t rs2,
+                                            size_t vl) {
+  return __riscv_vloxei16_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 4 x bfloat> @test_vloxei16_v_bf16m1_tumu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], <vscale x 4 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x bfloat> @llvm.riscv.vloxei.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[VD]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 4 x bfloat> [[TMP0]]
+//
+vbfloat16m1_t test_vloxei16_v_bf16m1_tumu(vbool16_t vm, vbfloat16m1_t vd,
+                                          const __bf16 *rs1, vuint16m1_t rs2,
+                                          size_t vl) {
+  return __riscv_vloxei16_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 8 x bfloat> @test_vloxei16_v_bf16m2_tumu(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], <vscale x 8 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x bfloat> @llvm.riscv.vloxei.mask.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> [[VD]], ptr [[RS1]], <vscale x 8 x i16> [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 8 x bfloat> [[TMP0]]
+//
+vbfloat16m2_t test_vloxei16_v_bf16m2_tumu(vbool8_t vm, vbfloat16m2_t vd,
+                                          const __bf16 *rs1, vuint16m2_t rs2,
+                                          size_t vl) {
+  return __riscv_vloxei16_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 16 x bfloat> @test_vloxei16_v_bf16m4_tumu(
+// CHECK-RV64-SAME: <vscale x 16 x i1> [[VM:%.*]], <vscale x 16 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 16 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x bfloat> @llvm.riscv.vloxei.mask.nxv16bf16.nxv16i16.i64(<vscale x 16 x bfloat> [[VD]], ptr [[RS1]], <vscale x 16 x i16> [[RS2]], <vscale x 16 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 16 x bfloat> [[TMP0]]
+//
+vbfloat16m4_t test_vloxei16_v_bf16m4_tumu(vbool4_t vm, vbfloat16m4_t vd,
+                                          const __bf16 *rs1, vuint16m4_t rs2,
+                                          size_t vl) {
+  return __riscv_vloxei16_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 32 x bfloat> @test_vloxei16_v_bf16m8_tumu(
+// CHECK-RV64-SAME: <vscale x 32 x i1> [[VM:%.*]], <vscale x 32 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 32 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x bfloat> @llvm.riscv.vloxei.mask.nxv32bf16.nxv32i16.i64(<vscale x 32 x bfloat> [[VD]], ptr [[RS1]], <vscale x 32 x i16> [[RS2]], <vscale x 32 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 32 x bfloat> [[TMP0]]
+//
+vbfloat16m8_t test_vloxei16_v_bf16m8_tumu(vbool2_t vm, vbfloat16m8_t vd,
+                                          const __bf16 *rs1, vuint16m8_t rs2,
+                                          size_t vl) {
+  return __riscv_vloxei16_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 1 x bfloat> @test_vloxei16_v_bf16mf4_mu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], <vscale x 1 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x bfloat> @llvm.riscv.vloxei.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[VD]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret <vscale x 1 x bfloat> [[TMP0]]
+//
+vbfloat16mf4_t test_vloxei16_v_bf16mf4_mu(vbool64_t vm, vbfloat16mf4_t vd,
+                                          const __bf16 *rs1, vuint16mf4_t rs2,
+                                          size_t vl) {
+  return __riscv_vloxei16_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 2 x bfloat> @test_vloxei16_v_bf16mf2_mu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], <vscale x 2 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x bfloat> @llvm.riscv.vloxei.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[VD]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret <vscale x 2 x bfloat> [[TMP0]]
+//
+vbfloat16mf2_t test_vloxei16_v_bf16mf2_mu(vbool32_t vm, vbfloat16mf2_t vd,
+                                          const __bf16 *rs1, vuint16mf2_t rs2,
+                                          size_t vl) {
+  return __riscv_vloxei16_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 4 x bfloat> @test_vloxei16_v_bf16m1_mu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], <vscale x 4 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x bfloat> @llvm.riscv.vloxei.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[VD]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret <vscale x 4 x bfloat> [[TMP0]]
+//
+vbfloat16m1_t test_vloxei16_v_bf16m1_mu(vbool16_t vm, vbfloat16m1_t vd,
+                                        const __bf16 *rs1, vuint16m1_t rs2,
+                                        size_t vl) {
+  return __riscv_vloxei16_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 8 x bfloat> @test_vloxei16_v_bf16m2_mu(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], <vscale x 8 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x bfloat> @llvm.riscv.vloxei.mask.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> [[VD]], ptr [[RS1]], <vscale x 8 x i16> [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret <vscale x 8 x bfloat> [[TMP0]]
+//
+vbfloat16m2_t test_vloxei16_v_bf16m2_mu(vbool8_t vm, vbfloat16m2_t vd,
+                                        const __bf16 *rs1, vuint16m2_t rs2,
+                                        size_t vl) {
+  return __riscv_vloxei16_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 16 x bfloat> @test_vloxei16_v_bf16m4_mu(
+// CHECK-RV64-SAME: <vscale x 16 x i1> [[VM:%.*]], <vscale x 16 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 16 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x bfloat> @llvm.riscv.vloxei.mask.nxv16bf16.nxv16i16.i64(<vscale x 16 x bfloat> [[VD]], ptr [[RS1]], <vscale x 16 x i16> [[RS2]], <vscale x 16 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret <vscale x 16 x bfloat> [[TMP0]]
+//
+vbfloat16m4_t test_vloxei16_v_bf16m4_mu(vbool4_t vm, vbfloat16m4_t vd,
+                                        const __bf16 *rs1, vuint16m4_t rs2,
+                                        size_t vl) {
+  return __riscv_vloxei16_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 32 x bfloat> @test_vloxei16_v_bf16m8_mu(
+// CHECK-RV64-SAME: <vscale x 32 x i1> [[VM:%.*]], <vscale x 32 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 32 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x bfloat> @llvm.riscv.vloxei.mask.nxv32bf16.nxv32i16.i64(<vscale x 32 x bfloat> [[VD]], ptr [[RS1]], <vscale x 32 x i16> [[RS2]], <vscale x 32 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret <vscale x 32 x bfloat> [[TMP0]]
+//
+vbfloat16m8_t test_vloxei16_v_bf16m8_mu(vbool2_t vm, vbfloat16m8_t vd,
+                                        const __bf16 *rs1, vuint16m8_t rs2,
+                                        size_t vl) {
+  return __riscv_vloxei16_mu(vm, vd, rs1, rs2, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vloxseg2ei16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vloxseg2ei16.c
new file mode 100644
index 000000000000..eab9cb4d632e
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vloxseg2ei16.c
@@ -0,0 +1,306 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vloxseg2ei16_v_bf16mf4x2_tu(
+// CHECK-RV64-SAME: { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vloxseg2.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP2]]
+//
+vbfloat16mf4x2_t test_vloxseg2ei16_v_bf16mf4x2_tu(vbfloat16mf4x2_t vd,
+                                                  const __bf16 *rs1,
+                                                  vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vloxseg2ei16_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vloxseg2ei16_v_bf16mf2x2_tu(
+// CHECK-RV64-SAME: { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vloxseg2.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP2]]
+//
+vbfloat16mf2x2_t test_vloxseg2ei16_v_bf16mf2x2_tu(vbfloat16mf2x2_t vd,
+                                                  const __bf16 *rs1,
+                                                  vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vloxseg2ei16_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vloxseg2ei16_v_bf16m1x2_tu(
+// CHECK-RV64-SAME: { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vloxseg2.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP2]]
+//
+vbfloat16m1x2_t test_vloxseg2ei16_v_bf16m1x2_tu(vbfloat16m1x2_t vd,
+                                                const __bf16 *rs1,
+                                                vuint16m1_t rs2, size_t vl) {
+  return __riscv_vloxseg2ei16_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vloxseg2ei16_v_bf16m2x2_tu(
+// CHECK-RV64-SAME: { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vloxseg2.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 8 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP2]]
+//
+vbfloat16m2x2_t test_vloxseg2ei16_v_bf16m2x2_tu(vbfloat16m2x2_t vd,
+                                                const __bf16 *rs1,
+                                                vuint16m2_t rs2, size_t vl) {
+  return __riscv_vloxseg2ei16_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @test_vloxseg2ei16_v_bf16m4x2_tu(
+// CHECK-RV64-SAME: { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 16 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @llvm.riscv.vloxseg2.nxv16bf16.nxv16i16.i64(<vscale x 16 x bfloat> [[TMP0]], <vscale x 16 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 16 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[TMP2]]
+//
+vbfloat16m4x2_t test_vloxseg2ei16_v_bf16m4x2_tu(vbfloat16m4x2_t vd,
+                                                const __bf16 *rs1,
+                                                vuint16m4_t rs2, size_t vl) {
+  return __riscv_vloxseg2ei16_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vloxseg2ei16_v_bf16mf4x2_tum(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vloxseg2.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP2]]
+//
+vbfloat16mf4x2_t test_vloxseg2ei16_v_bf16mf4x2_tum(vbool64_t vm,
+                                                   vbfloat16mf4x2_t vd,
+                                                   const __bf16 *rs1,
+                                                   vuint16mf4_t rs2,
+                                                   size_t vl) {
+  return __riscv_vloxseg2ei16_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vloxseg2ei16_v_bf16mf2x2_tum(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vloxseg2.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP2]]
+//
+vbfloat16mf2x2_t test_vloxseg2ei16_v_bf16mf2x2_tum(vbool32_t vm,
+                                                   vbfloat16mf2x2_t vd,
+                                                   const __bf16 *rs1,
+                                                   vuint16mf2_t rs2,
+                                                   size_t vl) {
+  return __riscv_vloxseg2ei16_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vloxseg2ei16_v_bf16m1x2_tum(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vloxseg2.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP2]]
+//
+vbfloat16m1x2_t test_vloxseg2ei16_v_bf16m1x2_tum(vbool16_t vm,
+                                                 vbfloat16m1x2_t vd,
+                                                 const __bf16 *rs1,
+                                                 vuint16m1_t rs2, size_t vl) {
+  return __riscv_vloxseg2ei16_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vloxseg2ei16_v_bf16m2x2_tum(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vloxseg2.mask.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 8 x i16> [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP2]]
+//
+vbfloat16m2x2_t test_vloxseg2ei16_v_bf16m2x2_tum(vbool8_t vm,
+                                                 vbfloat16m2x2_t vd,
+                                                 const __bf16 *rs1,
+                                                 vuint16m2_t rs2, size_t vl) {
+  return __riscv_vloxseg2ei16_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @test_vloxseg2ei16_v_bf16m4x2_tum(
+// CHECK-RV64-SAME: <vscale x 16 x i1> [[VM:%.*]], { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 16 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @llvm.riscv.vloxseg2.mask.nxv16bf16.nxv16i16.i64(<vscale x 16 x bfloat> [[TMP0]], <vscale x 16 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 16 x i16> [[RS2]], <vscale x 16 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[TMP2]]
+//
+vbfloat16m4x2_t test_vloxseg2ei16_v_bf16m4x2_tum(vbool4_t vm,
+                                                 vbfloat16m4x2_t vd,
+                                                 const __bf16 *rs1,
+                                                 vuint16m4_t rs2, size_t vl) {
+  return __riscv_vloxseg2ei16_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vloxseg2ei16_v_bf16mf4x2_tumu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vloxseg2.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP2]]
+//
+vbfloat16mf4x2_t test_vloxseg2ei16_v_bf16mf4x2_tumu(vbool64_t vm,
+                                                    vbfloat16mf4x2_t vd,
+                                                    const __bf16 *rs1,
+                                                    vuint16mf4_t rs2,
+                                                    size_t vl) {
+  return __riscv_vloxseg2ei16_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vloxseg2ei16_v_bf16mf2x2_tumu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vloxseg2.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP2]]
+//
+vbfloat16mf2x2_t test_vloxseg2ei16_v_bf16mf2x2_tumu(vbool32_t vm,
+                                                    vbfloat16mf2x2_t vd,
+                                                    const __bf16 *rs1,
+                                                    vuint16mf2_t rs2,
+                                                    size_t vl) {
+  return __riscv_vloxseg2ei16_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vloxseg2ei16_v_bf16m1x2_tumu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vloxseg2.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP2]]
+//
+vbfloat16m1x2_t test_vloxseg2ei16_v_bf16m1x2_tumu(vbool16_t vm,
+                                                  vbfloat16m1x2_t vd,
+                                                  const __bf16 *rs1,
+                                                  vuint16m1_t rs2, size_t vl) {
+  return __riscv_vloxseg2ei16_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vloxseg2ei16_v_bf16m2x2_tumu(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vloxseg2.mask.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 8 x i16> [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP2]]
+//
+vbfloat16m2x2_t test_vloxseg2ei16_v_bf16m2x2_tumu(vbool8_t vm,
+                                                  vbfloat16m2x2_t vd,
+                                                  const __bf16 *rs1,
+                                                  vuint16m2_t rs2, size_t vl) {
+  return __riscv_vloxseg2ei16_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @test_vloxseg2ei16_v_bf16m4x2_tumu(
+// CHECK-RV64-SAME: <vscale x 16 x i1> [[VM:%.*]], { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 16 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @llvm.riscv.vloxseg2.mask.nxv16bf16.nxv16i16.i64(<vscale x 16 x bfloat> [[TMP0]], <vscale x 16 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 16 x i16> [[RS2]], <vscale x 16 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[TMP2]]
+//
+vbfloat16m4x2_t test_vloxseg2ei16_v_bf16m4x2_tumu(vbool4_t vm,
+                                                  vbfloat16m4x2_t vd,
+                                                  const __bf16 *rs1,
+                                                  vuint16m4_t rs2, size_t vl) {
+  return __riscv_vloxseg2ei16_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vloxseg2ei16_v_bf16mf4x2_mu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vloxseg2.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP2]]
+//
+vbfloat16mf4x2_t test_vloxseg2ei16_v_bf16mf4x2_mu(vbool64_t vm,
+                                                  vbfloat16mf4x2_t vd,
+                                                  const __bf16 *rs1,
+                                                  vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vloxseg2ei16_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vloxseg2ei16_v_bf16mf2x2_mu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vloxseg2.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP2]]
+//
+vbfloat16mf2x2_t test_vloxseg2ei16_v_bf16mf2x2_mu(vbool32_t vm,
+                                                  vbfloat16mf2x2_t vd,
+                                                  const __bf16 *rs1,
+                                                  vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vloxseg2ei16_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vloxseg2ei16_v_bf16m1x2_mu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vloxseg2.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP2]]
+//
+vbfloat16m1x2_t test_vloxseg2ei16_v_bf16m1x2_mu(vbool16_t vm,
+                                                vbfloat16m1x2_t vd,
+                                                const __bf16 *rs1,
+                                                vuint16m1_t rs2, size_t vl) {
+  return __riscv_vloxseg2ei16_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vloxseg2ei16_v_bf16m2x2_mu(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vloxseg2.mask.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 8 x i16> [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP2]]
+//
+vbfloat16m2x2_t test_vloxseg2ei16_v_bf16m2x2_mu(vbool8_t vm, vbfloat16m2x2_t vd,
+                                                const __bf16 *rs1,
+                                                vuint16m2_t rs2, size_t vl) {
+  return __riscv_vloxseg2ei16_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @test_vloxseg2ei16_v_bf16m4x2_mu(
+// CHECK-RV64-SAME: <vscale x 16 x i1> [[VM:%.*]], { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 16 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @llvm.riscv.vloxseg2.mask.nxv16bf16.nxv16i16.i64(<vscale x 16 x bfloat> [[TMP0]], <vscale x 16 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 16 x i16> [[RS2]], <vscale x 16 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[TMP2]]
+//
+vbfloat16m4x2_t test_vloxseg2ei16_v_bf16m4x2_mu(vbool4_t vm, vbfloat16m4x2_t vd,
+                                                const __bf16 *rs1,
+                                                vuint16m4_t rs2, size_t vl) {
+  return __riscv_vloxseg2ei16_mu(vm, vd, rs1, rs2, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vloxseg3ei16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vloxseg3ei16.c
new file mode 100644
index 000000000000..ba5294df1254
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vloxseg3ei16.c
@@ -0,0 +1,264 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vloxseg3ei16_v_bf16mf4x3_tu(
+// CHECK-RV64-SAME: { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vloxseg3.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP3]]
+//
+vbfloat16mf4x3_t test_vloxseg3ei16_v_bf16mf4x3_tu(vbfloat16mf4x3_t vd,
+                                                  const __bf16 *rs1,
+                                                  vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vloxseg3ei16_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vloxseg3ei16_v_bf16mf2x3_tu(
+// CHECK-RV64-SAME: { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vloxseg3.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP3]]
+//
+vbfloat16mf2x3_t test_vloxseg3ei16_v_bf16mf2x3_tu(vbfloat16mf2x3_t vd,
+                                                  const __bf16 *rs1,
+                                                  vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vloxseg3ei16_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vloxseg3ei16_v_bf16m1x3_tu(
+// CHECK-RV64-SAME: { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vloxseg3.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP3]]
+//
+vbfloat16m1x3_t test_vloxseg3ei16_v_bf16m1x3_tu(vbfloat16m1x3_t vd,
+                                                const __bf16 *rs1,
+                                                vuint16m1_t rs2, size_t vl) {
+  return __riscv_vloxseg3ei16_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vloxseg3ei16_v_bf16m2x3_tu(
+// CHECK-RV64-SAME: { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vloxseg3.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 8 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP3]]
+//
+vbfloat16m2x3_t test_vloxseg3ei16_v_bf16m2x3_tu(vbfloat16m2x3_t vd,
+                                                const __bf16 *rs1,
+                                                vuint16m2_t rs2, size_t vl) {
+  return __riscv_vloxseg3ei16_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vloxseg3ei16_v_bf16mf4x3_tum(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vloxseg3.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP3]]
+//
+vbfloat16mf4x3_t test_vloxseg3ei16_v_bf16mf4x3_tum(vbool64_t vm,
+                                                   vbfloat16mf4x3_t vd,
+                                                   const __bf16 *rs1,
+                                                   vuint16mf4_t rs2,
+                                                   size_t vl) {
+  return __riscv_vloxseg3ei16_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vloxseg3ei16_v_bf16mf2x3_tum(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vloxseg3.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP3]]
+//
+vbfloat16mf2x3_t test_vloxseg3ei16_v_bf16mf2x3_tum(vbool32_t vm,
+                                                   vbfloat16mf2x3_t vd,
+                                                   const __bf16 *rs1,
+                                                   vuint16mf2_t rs2,
+                                                   size_t vl) {
+  return __riscv_vloxseg3ei16_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vloxseg3ei16_v_bf16m1x3_tum(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vloxseg3.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP3]]
+//
+vbfloat16m1x3_t test_vloxseg3ei16_v_bf16m1x3_tum(vbool16_t vm,
+                                                 vbfloat16m1x3_t vd,
+                                                 const __bf16 *rs1,
+                                                 vuint16m1_t rs2, size_t vl) {
+  return __riscv_vloxseg3ei16_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vloxseg3ei16_v_bf16m2x3_tum(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vloxseg3.mask.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 8 x i16> [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP3]]
+//
+vbfloat16m2x3_t test_vloxseg3ei16_v_bf16m2x3_tum(vbool8_t vm,
+                                                 vbfloat16m2x3_t vd,
+                                                 const __bf16 *rs1,
+                                                 vuint16m2_t rs2, size_t vl) {
+  return __riscv_vloxseg3ei16_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vloxseg3ei16_v_bf16mf4x3_tumu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vloxseg3.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP3]]
+//
+vbfloat16mf4x3_t test_vloxseg3ei16_v_bf16mf4x3_tumu(vbool64_t vm,
+                                                    vbfloat16mf4x3_t vd,
+                                                    const __bf16 *rs1,
+                                                    vuint16mf4_t rs2,
+                                                    size_t vl) {
+  return __riscv_vloxseg3ei16_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vloxseg3ei16_v_bf16mf2x3_tumu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vloxseg3.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP3]]
+//
+vbfloat16mf2x3_t test_vloxseg3ei16_v_bf16mf2x3_tumu(vbool32_t vm,
+                                                    vbfloat16mf2x3_t vd,
+                                                    const __bf16 *rs1,
+                                                    vuint16mf2_t rs2,
+                                                    size_t vl) {
+  return __riscv_vloxseg3ei16_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vloxseg3ei16_v_bf16m1x3_tumu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vloxseg3.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP3]]
+//
+vbfloat16m1x3_t test_vloxseg3ei16_v_bf16m1x3_tumu(vbool16_t vm,
+                                                  vbfloat16m1x3_t vd,
+                                                  const __bf16 *rs1,
+                                                  vuint16m1_t rs2, size_t vl) {
+  return __riscv_vloxseg3ei16_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vloxseg3ei16_v_bf16m2x3_tumu(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vloxseg3.mask.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 8 x i16> [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP3]]
+//
+vbfloat16m2x3_t test_vloxseg3ei16_v_bf16m2x3_tumu(vbool8_t vm,
+                                                  vbfloat16m2x3_t vd,
+                                                  const __bf16 *rs1,
+                                                  vuint16m2_t rs2, size_t vl) {
+  return __riscv_vloxseg3ei16_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vloxseg3ei16_v_bf16mf4x3_mu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vloxseg3.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP3]]
+//
+vbfloat16mf4x3_t test_vloxseg3ei16_v_bf16mf4x3_mu(vbool64_t vm,
+                                                  vbfloat16mf4x3_t vd,
+                                                  const __bf16 *rs1,
+                                                  vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vloxseg3ei16_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vloxseg3ei16_v_bf16mf2x3_mu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vloxseg3.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP3]]
+//
+vbfloat16mf2x3_t test_vloxseg3ei16_v_bf16mf2x3_mu(vbool32_t vm,
+                                                  vbfloat16mf2x3_t vd,
+                                                  const __bf16 *rs1,
+                                                  vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vloxseg3ei16_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vloxseg3ei16_v_bf16m1x3_mu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vloxseg3.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP3]]
+//
+vbfloat16m1x3_t test_vloxseg3ei16_v_bf16m1x3_mu(vbool16_t vm,
+                                                vbfloat16m1x3_t vd,
+                                                const __bf16 *rs1,
+                                                vuint16m1_t rs2, size_t vl) {
+  return __riscv_vloxseg3ei16_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vloxseg3ei16_v_bf16m2x3_mu(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vloxseg3.mask.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 8 x i16> [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP3]]
+//
+vbfloat16m2x3_t test_vloxseg3ei16_v_bf16m2x3_mu(vbool8_t vm, vbfloat16m2x3_t vd,
+                                                const __bf16 *rs1,
+                                                vuint16m2_t rs2, size_t vl) {
+  return __riscv_vloxseg3ei16_mu(vm, vd, rs1, rs2, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vloxseg4ei16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vloxseg4ei16.c
new file mode 100644
index 000000000000..94e96b9875a1
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vloxseg4ei16.c
@@ -0,0 +1,280 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vloxseg4ei16_v_bf16mf4x4_tu(
+// CHECK-RV64-SAME: { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vloxseg4.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP4]]
+//
+vbfloat16mf4x4_t test_vloxseg4ei16_v_bf16mf4x4_tu(vbfloat16mf4x4_t vd,
+                                                  const __bf16 *rs1,
+                                                  vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vloxseg4ei16_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vloxseg4ei16_v_bf16mf2x4_tu(
+// CHECK-RV64-SAME: { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vloxseg4.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP4]]
+//
+vbfloat16mf2x4_t test_vloxseg4ei16_v_bf16mf2x4_tu(vbfloat16mf2x4_t vd,
+                                                  const __bf16 *rs1,
+                                                  vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vloxseg4ei16_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vloxseg4ei16_v_bf16m1x4_tu(
+// CHECK-RV64-SAME: { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vloxseg4.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP4]]
+//
+vbfloat16m1x4_t test_vloxseg4ei16_v_bf16m1x4_tu(vbfloat16m1x4_t vd,
+                                                const __bf16 *rs1,
+                                                vuint16m1_t rs2, size_t vl) {
+  return __riscv_vloxseg4ei16_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vloxseg4ei16_v_bf16m2x4_tu(
+// CHECK-RV64-SAME: { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vloxseg4.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], <vscale x 8 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 8 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP4]]
+//
+vbfloat16m2x4_t test_vloxseg4ei16_v_bf16m2x4_tu(vbfloat16m2x4_t vd,
+                                                const __bf16 *rs1,
+                                                vuint16m2_t rs2, size_t vl) {
+  return __riscv_vloxseg4ei16_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vloxseg4ei16_v_bf16mf4x4_tum(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vloxseg4.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP4]]
+//
+vbfloat16mf4x4_t test_vloxseg4ei16_v_bf16mf4x4_tum(vbool64_t vm,
+                                                   vbfloat16mf4x4_t vd,
+                                                   const __bf16 *rs1,
+                                                   vuint16mf4_t rs2,
+                                                   size_t vl) {
+  return __riscv_vloxseg4ei16_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vloxseg4ei16_v_bf16mf2x4_tum(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vloxseg4.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP4]]
+//
+vbfloat16mf2x4_t test_vloxseg4ei16_v_bf16mf2x4_tum(vbool32_t vm,
+                                                   vbfloat16mf2x4_t vd,
+                                                   const __bf16 *rs1,
+                                                   vuint16mf2_t rs2,
+                                                   size_t vl) {
+  return __riscv_vloxseg4ei16_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vloxseg4ei16_v_bf16m1x4_tum(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vloxseg4.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP4]]
+//
+vbfloat16m1x4_t test_vloxseg4ei16_v_bf16m1x4_tum(vbool16_t vm,
+                                                 vbfloat16m1x4_t vd,
+                                                 const __bf16 *rs1,
+                                                 vuint16m1_t rs2, size_t vl) {
+  return __riscv_vloxseg4ei16_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vloxseg4ei16_v_bf16m2x4_tum(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vloxseg4.mask.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], <vscale x 8 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 8 x i16> [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP4]]
+//
+vbfloat16m2x4_t test_vloxseg4ei16_v_bf16m2x4_tum(vbool8_t vm,
+                                                 vbfloat16m2x4_t vd,
+                                                 const __bf16 *rs1,
+                                                 vuint16m2_t rs2, size_t vl) {
+  return __riscv_vloxseg4ei16_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vloxseg4ei16_v_bf16mf4x4_tumu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vloxseg4.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP4]]
+//
+vbfloat16mf4x4_t test_vloxseg4ei16_v_bf16mf4x4_tumu(vbool64_t vm,
+                                                    vbfloat16mf4x4_t vd,
+                                                    const __bf16 *rs1,
+                                                    vuint16mf4_t rs2,
+                                                    size_t vl) {
+  return __riscv_vloxseg4ei16_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vloxseg4ei16_v_bf16mf2x4_tumu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vloxseg4.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP4]]
+//
+vbfloat16mf2x4_t test_vloxseg4ei16_v_bf16mf2x4_tumu(vbool32_t vm,
+                                                    vbfloat16mf2x4_t vd,
+                                                    const __bf16 *rs1,
+                                                    vuint16mf2_t rs2,
+                                                    size_t vl) {
+  return __riscv_vloxseg4ei16_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vloxseg4ei16_v_bf16m1x4_tumu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vloxseg4.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP4]]
+//
+vbfloat16m1x4_t test_vloxseg4ei16_v_bf16m1x4_tumu(vbool16_t vm,
+                                                  vbfloat16m1x4_t vd,
+                                                  const __bf16 *rs1,
+                                                  vuint16m1_t rs2, size_t vl) {
+  return __riscv_vloxseg4ei16_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vloxseg4ei16_v_bf16m2x4_tumu(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vloxseg4.mask.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], <vscale x 8 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 8 x i16> [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP4]]
+//
+vbfloat16m2x4_t test_vloxseg4ei16_v_bf16m2x4_tumu(vbool8_t vm,
+                                                  vbfloat16m2x4_t vd,
+                                                  const __bf16 *rs1,
+                                                  vuint16m2_t rs2, size_t vl) {
+  return __riscv_vloxseg4ei16_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vloxseg4ei16_v_bf16mf4x4_mu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vloxseg4.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP4]]
+//
+vbfloat16mf4x4_t test_vloxseg4ei16_v_bf16mf4x4_mu(vbool64_t vm,
+                                                  vbfloat16mf4x4_t vd,
+                                                  const __bf16 *rs1,
+                                                  vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vloxseg4ei16_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vloxseg4ei16_v_bf16mf2x4_mu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vloxseg4.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP4]]
+//
+vbfloat16mf2x4_t test_vloxseg4ei16_v_bf16mf2x4_mu(vbool32_t vm,
+                                                  vbfloat16mf2x4_t vd,
+                                                  const __bf16 *rs1,
+                                                  vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vloxseg4ei16_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vloxseg4ei16_v_bf16m1x4_mu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vloxseg4.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP4]]
+//
+vbfloat16m1x4_t test_vloxseg4ei16_v_bf16m1x4_mu(vbool16_t vm,
+                                                vbfloat16m1x4_t vd,
+                                                const __bf16 *rs1,
+                                                vuint16m1_t rs2, size_t vl) {
+  return __riscv_vloxseg4ei16_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vloxseg4ei16_v_bf16m2x4_mu(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vloxseg4.mask.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], <vscale x 8 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 8 x i16> [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP4]]
+//
+vbfloat16m2x4_t test_vloxseg4ei16_v_bf16m2x4_mu(vbool8_t vm, vbfloat16m2x4_t vd,
+                                                const __bf16 *rs1,
+                                                vuint16m2_t rs2, size_t vl) {
+  return __riscv_vloxseg4ei16_mu(vm, vd, rs1, rs2, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vloxseg5ei16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vloxseg5ei16.c
new file mode 100644
index 000000000000..54f1a6626d48
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vloxseg5ei16.c
@@ -0,0 +1,226 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vloxseg5ei16_v_bf16mf4x5_tu(
+// CHECK-RV64-SAME: { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vloxseg5.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP5]]
+//
+vbfloat16mf4x5_t test_vloxseg5ei16_v_bf16mf4x5_tu(vbfloat16mf4x5_t vd,
+                                                  const __bf16 *rs1,
+                                                  vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vloxseg5ei16_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vloxseg5ei16_v_bf16mf2x5_tu(
+// CHECK-RV64-SAME: { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vloxseg5.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP5]]
+//
+vbfloat16mf2x5_t test_vloxseg5ei16_v_bf16mf2x5_tu(vbfloat16mf2x5_t vd,
+                                                  const __bf16 *rs1,
+                                                  vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vloxseg5ei16_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vloxseg5ei16_v_bf16m1x5_tu(
+// CHECK-RV64-SAME: { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vloxseg5.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP5]]
+//
+vbfloat16m1x5_t test_vloxseg5ei16_v_bf16m1x5_tu(vbfloat16m1x5_t vd,
+                                                const __bf16 *rs1,
+                                                vuint16m1_t rs2, size_t vl) {
+  return __riscv_vloxseg5ei16_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vloxseg5ei16_v_bf16mf4x5_tum(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vloxseg5.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP5]]
+//
+vbfloat16mf4x5_t test_vloxseg5ei16_v_bf16mf4x5_tum(vbool64_t vm,
+                                                   vbfloat16mf4x5_t vd,
+                                                   const __bf16 *rs1,
+                                                   vuint16mf4_t rs2,
+                                                   size_t vl) {
+  return __riscv_vloxseg5ei16_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vloxseg5ei16_v_bf16mf2x5_tum(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vloxseg5.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP5]]
+//
+vbfloat16mf2x5_t test_vloxseg5ei16_v_bf16mf2x5_tum(vbool32_t vm,
+                                                   vbfloat16mf2x5_t vd,
+                                                   const __bf16 *rs1,
+                                                   vuint16mf2_t rs2,
+                                                   size_t vl) {
+  return __riscv_vloxseg5ei16_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vloxseg5ei16_v_bf16m1x5_tum(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vloxseg5.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP5]]
+//
+vbfloat16m1x5_t test_vloxseg5ei16_v_bf16m1x5_tum(vbool16_t vm,
+                                                 vbfloat16m1x5_t vd,
+                                                 const __bf16 *rs1,
+                                                 vuint16m1_t rs2, size_t vl) {
+  return __riscv_vloxseg5ei16_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vloxseg5ei16_v_bf16mf4x5_tumu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vloxseg5.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP5]]
+//
+vbfloat16mf4x5_t test_vloxseg5ei16_v_bf16mf4x5_tumu(vbool64_t vm,
+                                                    vbfloat16mf4x5_t vd,
+                                                    const __bf16 *rs1,
+                                                    vuint16mf4_t rs2,
+                                                    size_t vl) {
+  return __riscv_vloxseg5ei16_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vloxseg5ei16_v_bf16mf2x5_tumu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vloxseg5.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP5]]
+//
+vbfloat16mf2x5_t test_vloxseg5ei16_v_bf16mf2x5_tumu(vbool32_t vm,
+                                                    vbfloat16mf2x5_t vd,
+                                                    const __bf16 *rs1,
+                                                    vuint16mf2_t rs2,
+                                                    size_t vl) {
+  return __riscv_vloxseg5ei16_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vloxseg5ei16_v_bf16m1x5_tumu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vloxseg5.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP5]]
+//
+vbfloat16m1x5_t test_vloxseg5ei16_v_bf16m1x5_tumu(vbool16_t vm,
+                                                  vbfloat16m1x5_t vd,
+                                                  const __bf16 *rs1,
+                                                  vuint16m1_t rs2, size_t vl) {
+  return __riscv_vloxseg5ei16_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vloxseg5ei16_v_bf16mf4x5_mu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vloxseg5.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP5]]
+//
+vbfloat16mf4x5_t test_vloxseg5ei16_v_bf16mf4x5_mu(vbool64_t vm,
+                                                  vbfloat16mf4x5_t vd,
+                                                  const __bf16 *rs1,
+                                                  vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vloxseg5ei16_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vloxseg5ei16_v_bf16mf2x5_mu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vloxseg5.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP5]]
+//
+vbfloat16mf2x5_t test_vloxseg5ei16_v_bf16mf2x5_mu(vbool32_t vm,
+                                                  vbfloat16mf2x5_t vd,
+                                                  const __bf16 *rs1,
+                                                  vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vloxseg5ei16_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vloxseg5ei16_v_bf16m1x5_mu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vloxseg5.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP5]]
+//
+vbfloat16m1x5_t test_vloxseg5ei16_v_bf16m1x5_mu(vbool16_t vm,
+                                                vbfloat16m1x5_t vd,
+                                                const __bf16 *rs1,
+                                                vuint16m1_t rs2, size_t vl) {
+  return __riscv_vloxseg5ei16_mu(vm, vd, rs1, rs2, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vloxseg6ei16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vloxseg6ei16.c
new file mode 100644
index 000000000000..a1ee2af0de3d
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vloxseg6ei16.c
@@ -0,0 +1,238 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vloxseg6ei16_v_bf16mf4x6_tu(
+// CHECK-RV64-SAME: { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vloxseg6.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP6]]
+//
+vbfloat16mf4x6_t test_vloxseg6ei16_v_bf16mf4x6_tu(vbfloat16mf4x6_t vd,
+                                                  const __bf16 *rs1,
+                                                  vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vloxseg6ei16_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vloxseg6ei16_v_bf16mf2x6_tu(
+// CHECK-RV64-SAME: { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vloxseg6.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP6]]
+//
+vbfloat16mf2x6_t test_vloxseg6ei16_v_bf16mf2x6_tu(vbfloat16mf2x6_t vd,
+                                                  const __bf16 *rs1,
+                                                  vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vloxseg6ei16_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vloxseg6ei16_v_bf16m1x6_tu(
+// CHECK-RV64-SAME: { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vloxseg6.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP6]]
+//
+vbfloat16m1x6_t test_vloxseg6ei16_v_bf16m1x6_tu(vbfloat16m1x6_t vd,
+                                                const __bf16 *rs1,
+                                                vuint16m1_t rs2, size_t vl) {
+  return __riscv_vloxseg6ei16_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vloxseg6ei16_v_bf16mf4x6_tum(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vloxseg6.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP6]]
+//
+vbfloat16mf4x6_t test_vloxseg6ei16_v_bf16mf4x6_tum(vbool64_t vm,
+                                                   vbfloat16mf4x6_t vd,
+                                                   const __bf16 *rs1,
+                                                   vuint16mf4_t rs2,
+                                                   size_t vl) {
+  return __riscv_vloxseg6ei16_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vloxseg6ei16_v_bf16mf2x6_tum(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vloxseg6.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP6]]
+//
+vbfloat16mf2x6_t test_vloxseg6ei16_v_bf16mf2x6_tum(vbool32_t vm,
+                                                   vbfloat16mf2x6_t vd,
+                                                   const __bf16 *rs1,
+                                                   vuint16mf2_t rs2,
+                                                   size_t vl) {
+  return __riscv_vloxseg6ei16_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vloxseg6ei16_v_bf16m1x6_tum(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vloxseg6.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP6]]
+//
+vbfloat16m1x6_t test_vloxseg6ei16_v_bf16m1x6_tum(vbool16_t vm,
+                                                 vbfloat16m1x6_t vd,
+                                                 const __bf16 *rs1,
+                                                 vuint16m1_t rs2, size_t vl) {
+  return __riscv_vloxseg6ei16_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vloxseg6ei16_v_bf16mf4x6_tumu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vloxseg6.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP6]]
+//
+vbfloat16mf4x6_t test_vloxseg6ei16_v_bf16mf4x6_tumu(vbool64_t vm,
+                                                    vbfloat16mf4x6_t vd,
+                                                    const __bf16 *rs1,
+                                                    vuint16mf4_t rs2,
+                                                    size_t vl) {
+  return __riscv_vloxseg6ei16_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vloxseg6ei16_v_bf16mf2x6_tumu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vloxseg6.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP6]]
+//
+vbfloat16mf2x6_t test_vloxseg6ei16_v_bf16mf2x6_tumu(vbool32_t vm,
+                                                    vbfloat16mf2x6_t vd,
+                                                    const __bf16 *rs1,
+                                                    vuint16mf2_t rs2,
+                                                    size_t vl) {
+  return __riscv_vloxseg6ei16_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vloxseg6ei16_v_bf16m1x6_tumu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vloxseg6.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP6]]
+//
+vbfloat16m1x6_t test_vloxseg6ei16_v_bf16m1x6_tumu(vbool16_t vm,
+                                                  vbfloat16m1x6_t vd,
+                                                  const __bf16 *rs1,
+                                                  vuint16m1_t rs2, size_t vl) {
+  return __riscv_vloxseg6ei16_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vloxseg6ei16_v_bf16mf4x6_mu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vloxseg6.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP6]]
+//
+vbfloat16mf4x6_t test_vloxseg6ei16_v_bf16mf4x6_mu(vbool64_t vm,
+                                                  vbfloat16mf4x6_t vd,
+                                                  const __bf16 *rs1,
+                                                  vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vloxseg6ei16_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vloxseg6ei16_v_bf16mf2x6_mu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vloxseg6.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP6]]
+//
+vbfloat16mf2x6_t test_vloxseg6ei16_v_bf16mf2x6_mu(vbool32_t vm,
+                                                  vbfloat16mf2x6_t vd,
+                                                  const __bf16 *rs1,
+                                                  vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vloxseg6ei16_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vloxseg6ei16_v_bf16m1x6_mu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vloxseg6.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP6]]
+//
+vbfloat16m1x6_t test_vloxseg6ei16_v_bf16m1x6_mu(vbool16_t vm,
+                                                vbfloat16m1x6_t vd,
+                                                const __bf16 *rs1,
+                                                vuint16m1_t rs2, size_t vl) {
+  return __riscv_vloxseg6ei16_mu(vm, vd, rs1, rs2, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vloxseg7ei16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vloxseg7ei16.c
new file mode 100644
index 000000000000..6b9cd5cf0526
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vloxseg7ei16.c
@@ -0,0 +1,250 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vloxseg7ei16_v_bf16mf4x7_tu(
+// CHECK-RV64-SAME: { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vloxseg7.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP7]]
+//
+vbfloat16mf4x7_t test_vloxseg7ei16_v_bf16mf4x7_tu(vbfloat16mf4x7_t vd,
+                                                  const __bf16 *rs1,
+                                                  vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vloxseg7ei16_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vloxseg7ei16_v_bf16mf2x7_tu(
+// CHECK-RV64-SAME: { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vloxseg7.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP7]]
+//
+vbfloat16mf2x7_t test_vloxseg7ei16_v_bf16mf2x7_tu(vbfloat16mf2x7_t vd,
+                                                  const __bf16 *rs1,
+                                                  vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vloxseg7ei16_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vloxseg7ei16_v_bf16m1x7_tu(
+// CHECK-RV64-SAME: { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vloxseg7.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP7]]
+//
+vbfloat16m1x7_t test_vloxseg7ei16_v_bf16m1x7_tu(vbfloat16m1x7_t vd,
+                                                const __bf16 *rs1,
+                                                vuint16m1_t rs2, size_t vl) {
+  return __riscv_vloxseg7ei16_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vloxseg7ei16_v_bf16mf4x7_tum(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vloxseg7.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP7]]
+//
+vbfloat16mf4x7_t test_vloxseg7ei16_v_bf16mf4x7_tum(vbool64_t vm,
+                                                   vbfloat16mf4x7_t vd,
+                                                   const __bf16 *rs1,
+                                                   vuint16mf4_t rs2,
+                                                   size_t vl) {
+  return __riscv_vloxseg7ei16_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vloxseg7ei16_v_bf16mf2x7_tum(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vloxseg7.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP7]]
+//
+vbfloat16mf2x7_t test_vloxseg7ei16_v_bf16mf2x7_tum(vbool32_t vm,
+                                                   vbfloat16mf2x7_t vd,
+                                                   const __bf16 *rs1,
+                                                   vuint16mf2_t rs2,
+                                                   size_t vl) {
+  return __riscv_vloxseg7ei16_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vloxseg7ei16_v_bf16m1x7_tum(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vloxseg7.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP7]]
+//
+vbfloat16m1x7_t test_vloxseg7ei16_v_bf16m1x7_tum(vbool16_t vm,
+                                                 vbfloat16m1x7_t vd,
+                                                 const __bf16 *rs1,
+                                                 vuint16m1_t rs2, size_t vl) {
+  return __riscv_vloxseg7ei16_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vloxseg7ei16_v_bf16mf4x7_tumu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vloxseg7.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP7]]
+//
+vbfloat16mf4x7_t test_vloxseg7ei16_v_bf16mf4x7_tumu(vbool64_t vm,
+                                                    vbfloat16mf4x7_t vd,
+                                                    const __bf16 *rs1,
+                                                    vuint16mf4_t rs2,
+                                                    size_t vl) {
+  return __riscv_vloxseg7ei16_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vloxseg7ei16_v_bf16mf2x7_tumu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vloxseg7.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP7]]
+//
+vbfloat16mf2x7_t test_vloxseg7ei16_v_bf16mf2x7_tumu(vbool32_t vm,
+                                                    vbfloat16mf2x7_t vd,
+                                                    const __bf16 *rs1,
+                                                    vuint16mf2_t rs2,
+                                                    size_t vl) {
+  return __riscv_vloxseg7ei16_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vloxseg7ei16_v_bf16m1x7_tumu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vloxseg7.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP7]]
+//
+vbfloat16m1x7_t test_vloxseg7ei16_v_bf16m1x7_tumu(vbool16_t vm,
+                                                  vbfloat16m1x7_t vd,
+                                                  const __bf16 *rs1,
+                                                  vuint16m1_t rs2, size_t vl) {
+  return __riscv_vloxseg7ei16_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vloxseg7ei16_v_bf16mf4x7_mu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vloxseg7.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP7]]
+//
+vbfloat16mf4x7_t test_vloxseg7ei16_v_bf16mf4x7_mu(vbool64_t vm,
+                                                  vbfloat16mf4x7_t vd,
+                                                  const __bf16 *rs1,
+                                                  vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vloxseg7ei16_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vloxseg7ei16_v_bf16mf2x7_mu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vloxseg7.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP7]]
+//
+vbfloat16mf2x7_t test_vloxseg7ei16_v_bf16mf2x7_mu(vbool32_t vm,
+                                                  vbfloat16mf2x7_t vd,
+                                                  const __bf16 *rs1,
+                                                  vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vloxseg7ei16_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vloxseg7ei16_v_bf16m1x7_mu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vloxseg7.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP7]]
+//
+vbfloat16m1x7_t test_vloxseg7ei16_v_bf16m1x7_mu(vbool16_t vm,
+                                                vbfloat16m1x7_t vd,
+                                                const __bf16 *rs1,
+                                                vuint16m1_t rs2, size_t vl) {
+  return __riscv_vloxseg7ei16_mu(vm, vd, rs1, rs2, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vloxseg8ei16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vloxseg8ei16.c
new file mode 100644
index 000000000000..f743b83d83c0
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vloxseg8ei16.c
@@ -0,0 +1,262 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vloxseg8ei16_v_bf16mf4x8_tu(
+// CHECK-RV64-SAME: { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vloxseg8.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], <vscale x 1 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP8]]
+//
+vbfloat16mf4x8_t test_vloxseg8ei16_v_bf16mf4x8_tu(vbfloat16mf4x8_t vd,
+                                                  const __bf16 *rs1,
+                                                  vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vloxseg8ei16_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vloxseg8ei16_v_bf16mf2x8_tu(
+// CHECK-RV64-SAME: { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vloxseg8.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], <vscale x 2 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP8]]
+//
+vbfloat16mf2x8_t test_vloxseg8ei16_v_bf16mf2x8_tu(vbfloat16mf2x8_t vd,
+                                                  const __bf16 *rs1,
+                                                  vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vloxseg8ei16_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vloxseg8ei16_v_bf16m1x8_tu(
+// CHECK-RV64-SAME: { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vloxseg8.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], <vscale x 4 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP8]]
+//
+vbfloat16m1x8_t test_vloxseg8ei16_v_bf16m1x8_tu(vbfloat16m1x8_t vd,
+                                                const __bf16 *rs1,
+                                                vuint16m1_t rs2, size_t vl) {
+  return __riscv_vloxseg8ei16_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vloxseg8ei16_v_bf16mf4x8_tum(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vloxseg8.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], <vscale x 1 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP8]]
+//
+vbfloat16mf4x8_t test_vloxseg8ei16_v_bf16mf4x8_tum(vbool64_t vm,
+                                                   vbfloat16mf4x8_t vd,
+                                                   const __bf16 *rs1,
+                                                   vuint16mf4_t rs2,
+                                                   size_t vl) {
+  return __riscv_vloxseg8ei16_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vloxseg8ei16_v_bf16mf2x8_tum(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vloxseg8.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], <vscale x 2 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP8]]
+//
+vbfloat16mf2x8_t test_vloxseg8ei16_v_bf16mf2x8_tum(vbool32_t vm,
+                                                   vbfloat16mf2x8_t vd,
+                                                   const __bf16 *rs1,
+                                                   vuint16mf2_t rs2,
+                                                   size_t vl) {
+  return __riscv_vloxseg8ei16_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vloxseg8ei16_v_bf16m1x8_tum(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vloxseg8.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], <vscale x 4 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP8]]
+//
+vbfloat16m1x8_t test_vloxseg8ei16_v_bf16m1x8_tum(vbool16_t vm,
+                                                 vbfloat16m1x8_t vd,
+                                                 const __bf16 *rs1,
+                                                 vuint16m1_t rs2, size_t vl) {
+  return __riscv_vloxseg8ei16_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vloxseg8ei16_v_bf16mf4x8_tumu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vloxseg8.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], <vscale x 1 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP8]]
+//
+vbfloat16mf4x8_t test_vloxseg8ei16_v_bf16mf4x8_tumu(vbool64_t vm,
+                                                    vbfloat16mf4x8_t vd,
+                                                    const __bf16 *rs1,
+                                                    vuint16mf4_t rs2,
+                                                    size_t vl) {
+  return __riscv_vloxseg8ei16_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vloxseg8ei16_v_bf16mf2x8_tumu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vloxseg8.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], <vscale x 2 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP8]]
+//
+vbfloat16mf2x8_t test_vloxseg8ei16_v_bf16mf2x8_tumu(vbool32_t vm,
+                                                    vbfloat16mf2x8_t vd,
+                                                    const __bf16 *rs1,
+                                                    vuint16mf2_t rs2,
+                                                    size_t vl) {
+  return __riscv_vloxseg8ei16_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vloxseg8ei16_v_bf16m1x8_tumu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vloxseg8.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], <vscale x 4 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP8]]
+//
+vbfloat16m1x8_t test_vloxseg8ei16_v_bf16m1x8_tumu(vbool16_t vm,
+                                                  vbfloat16m1x8_t vd,
+                                                  const __bf16 *rs1,
+                                                  vuint16m1_t rs2, size_t vl) {
+  return __riscv_vloxseg8ei16_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vloxseg8ei16_v_bf16mf4x8_mu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vloxseg8.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], <vscale x 1 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP8]]
+//
+vbfloat16mf4x8_t test_vloxseg8ei16_v_bf16mf4x8_mu(vbool64_t vm,
+                                                  vbfloat16mf4x8_t vd,
+                                                  const __bf16 *rs1,
+                                                  vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vloxseg8ei16_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vloxseg8ei16_v_bf16mf2x8_mu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vloxseg8.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], <vscale x 2 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP8]]
+//
+vbfloat16mf2x8_t test_vloxseg8ei16_v_bf16mf2x8_mu(vbool32_t vm,
+                                                  vbfloat16mf2x8_t vd,
+                                                  const __bf16 *rs1,
+                                                  vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vloxseg8ei16_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vloxseg8ei16_v_bf16m1x8_mu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vloxseg8.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], <vscale x 4 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP8]]
+//
+vbfloat16m1x8_t test_vloxseg8ei16_v_bf16m1x8_mu(vbool16_t vm,
+                                                vbfloat16m1x8_t vd,
+                                                const __bf16 *rs1,
+                                                vuint16m1_t rs2, size_t vl) {
+  return __riscv_vloxseg8ei16_mu(vm, vd, rs1, rs2, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vlse16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vlse16.c
new file mode 100644
index 000000000000..95652db60be0
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vlse16.c
@@ -0,0 +1,291 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 1 x bfloat> @test_vlse16_v_bf16mf4_tu(
+// CHECK-RV64-SAME: <vscale x 1 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x bfloat> @llvm.riscv.vlse.nxv1bf16.i64(<vscale x 1 x bfloat> [[VD]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret <vscale x 1 x bfloat> [[TMP0]]
+//
+vbfloat16mf4_t test_vlse16_v_bf16mf4_tu(vbfloat16mf4_t vd, const __bf16 *rs1,
+                                        ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlse16_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 2 x bfloat> @test_vlse16_v_bf16mf2_tu(
+// CHECK-RV64-SAME: <vscale x 2 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x bfloat> @llvm.riscv.vlse.nxv2bf16.i64(<vscale x 2 x bfloat> [[VD]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret <vscale x 2 x bfloat> [[TMP0]]
+//
+vbfloat16mf2_t test_vlse16_v_bf16mf2_tu(vbfloat16mf2_t vd, const __bf16 *rs1,
+                                        ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlse16_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 4 x bfloat> @test_vlse16_v_bf16m1_tu(
+// CHECK-RV64-SAME: <vscale x 4 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x bfloat> @llvm.riscv.vlse.nxv4bf16.i64(<vscale x 4 x bfloat> [[VD]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret <vscale x 4 x bfloat> [[TMP0]]
+//
+vbfloat16m1_t test_vlse16_v_bf16m1_tu(vbfloat16m1_t vd, const __bf16 *rs1,
+                                      ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlse16_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 8 x bfloat> @test_vlse16_v_bf16m2_tu(
+// CHECK-RV64-SAME: <vscale x 8 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x bfloat> @llvm.riscv.vlse.nxv8bf16.i64(<vscale x 8 x bfloat> [[VD]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret <vscale x 8 x bfloat> [[TMP0]]
+//
+vbfloat16m2_t test_vlse16_v_bf16m2_tu(vbfloat16m2_t vd, const __bf16 *rs1,
+                                      ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlse16_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 16 x bfloat> @test_vlse16_v_bf16m4_tu(
+// CHECK-RV64-SAME: <vscale x 16 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x bfloat> @llvm.riscv.vlse.nxv16bf16.i64(<vscale x 16 x bfloat> [[VD]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret <vscale x 16 x bfloat> [[TMP0]]
+//
+vbfloat16m4_t test_vlse16_v_bf16m4_tu(vbfloat16m4_t vd, const __bf16 *rs1,
+                                      ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlse16_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 32 x bfloat> @test_vlse16_v_bf16m8_tu(
+// CHECK-RV64-SAME: <vscale x 32 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x bfloat> @llvm.riscv.vlse.nxv32bf16.i64(<vscale x 32 x bfloat> [[VD]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret <vscale x 32 x bfloat> [[TMP0]]
+//
+vbfloat16m8_t test_vlse16_v_bf16m8_tu(vbfloat16m8_t vd, const __bf16 *rs1,
+                                      ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlse16_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 1 x bfloat> @test_vlse16_v_bf16mf4_tum(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], <vscale x 1 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x bfloat> @llvm.riscv.vlse.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[VD]], ptr [[RS1]], i64 [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret <vscale x 1 x bfloat> [[TMP0]]
+//
+vbfloat16mf4_t test_vlse16_v_bf16mf4_tum(vbool64_t vm, vbfloat16mf4_t vd,
+                                         const __bf16 *rs1, ptrdiff_t rs2,
+                                         size_t vl) {
+  return __riscv_vlse16_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 2 x bfloat> @test_vlse16_v_bf16mf2_tum(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], <vscale x 2 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x bfloat> @llvm.riscv.vlse.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[VD]], ptr [[RS1]], i64 [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret <vscale x 2 x bfloat> [[TMP0]]
+//
+vbfloat16mf2_t test_vlse16_v_bf16mf2_tum(vbool32_t vm, vbfloat16mf2_t vd,
+                                         const __bf16 *rs1, ptrdiff_t rs2,
+                                         size_t vl) {
+  return __riscv_vlse16_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 4 x bfloat> @test_vlse16_v_bf16m1_tum(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], <vscale x 4 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x bfloat> @llvm.riscv.vlse.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[VD]], ptr [[RS1]], i64 [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret <vscale x 4 x bfloat> [[TMP0]]
+//
+vbfloat16m1_t test_vlse16_v_bf16m1_tum(vbool16_t vm, vbfloat16m1_t vd,
+                                       const __bf16 *rs1, ptrdiff_t rs2,
+                                       size_t vl) {
+  return __riscv_vlse16_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 8 x bfloat> @test_vlse16_v_bf16m2_tum(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], <vscale x 8 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x bfloat> @llvm.riscv.vlse.mask.nxv8bf16.i64(<vscale x 8 x bfloat> [[VD]], ptr [[RS1]], i64 [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret <vscale x 8 x bfloat> [[TMP0]]
+//
+vbfloat16m2_t test_vlse16_v_bf16m2_tum(vbool8_t vm, vbfloat16m2_t vd,
+                                       const __bf16 *rs1, ptrdiff_t rs2,
+                                       size_t vl) {
+  return __riscv_vlse16_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 16 x bfloat> @test_vlse16_v_bf16m4_tum(
+// CHECK-RV64-SAME: <vscale x 16 x i1> [[VM:%.*]], <vscale x 16 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x bfloat> @llvm.riscv.vlse.mask.nxv16bf16.i64(<vscale x 16 x bfloat> [[VD]], ptr [[RS1]], i64 [[RS2]], <vscale x 16 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret <vscale x 16 x bfloat> [[TMP0]]
+//
+vbfloat16m4_t test_vlse16_v_bf16m4_tum(vbool4_t vm, vbfloat16m4_t vd,
+                                       const __bf16 *rs1, ptrdiff_t rs2,
+                                       size_t vl) {
+  return __riscv_vlse16_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 32 x bfloat> @test_vlse16_v_bf16m8_tum(
+// CHECK-RV64-SAME: <vscale x 32 x i1> [[VM:%.*]], <vscale x 32 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x bfloat> @llvm.riscv.vlse.mask.nxv32bf16.i64(<vscale x 32 x bfloat> [[VD]], ptr [[RS1]], i64 [[RS2]], <vscale x 32 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret <vscale x 32 x bfloat> [[TMP0]]
+//
+vbfloat16m8_t test_vlse16_v_bf16m8_tum(vbool2_t vm, vbfloat16m8_t vd,
+                                       const __bf16 *rs1, ptrdiff_t rs2,
+                                       size_t vl) {
+  return __riscv_vlse16_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 1 x bfloat> @test_vlse16_v_bf16mf4_tumu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], <vscale x 1 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x bfloat> @llvm.riscv.vlse.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[VD]], ptr [[RS1]], i64 [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 1 x bfloat> [[TMP0]]
+//
+vbfloat16mf4_t test_vlse16_v_bf16mf4_tumu(vbool64_t vm, vbfloat16mf4_t vd,
+                                          const __bf16 *rs1, ptrdiff_t rs2,
+                                          size_t vl) {
+  return __riscv_vlse16_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 2 x bfloat> @test_vlse16_v_bf16mf2_tumu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], <vscale x 2 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x bfloat> @llvm.riscv.vlse.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[VD]], ptr [[RS1]], i64 [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 2 x bfloat> [[TMP0]]
+//
+vbfloat16mf2_t test_vlse16_v_bf16mf2_tumu(vbool32_t vm, vbfloat16mf2_t vd,
+                                          const __bf16 *rs1, ptrdiff_t rs2,
+                                          size_t vl) {
+  return __riscv_vlse16_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 4 x bfloat> @test_vlse16_v_bf16m1_tumu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], <vscale x 4 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x bfloat> @llvm.riscv.vlse.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[VD]], ptr [[RS1]], i64 [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 4 x bfloat> [[TMP0]]
+//
+vbfloat16m1_t test_vlse16_v_bf16m1_tumu(vbool16_t vm, vbfloat16m1_t vd,
+                                        const __bf16 *rs1, ptrdiff_t rs2,
+                                        size_t vl) {
+  return __riscv_vlse16_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 8 x bfloat> @test_vlse16_v_bf16m2_tumu(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], <vscale x 8 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x bfloat> @llvm.riscv.vlse.mask.nxv8bf16.i64(<vscale x 8 x bfloat> [[VD]], ptr [[RS1]], i64 [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 8 x bfloat> [[TMP0]]
+//
+vbfloat16m2_t test_vlse16_v_bf16m2_tumu(vbool8_t vm, vbfloat16m2_t vd,
+                                        const __bf16 *rs1, ptrdiff_t rs2,
+                                        size_t vl) {
+  return __riscv_vlse16_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 16 x bfloat> @test_vlse16_v_bf16m4_tumu(
+// CHECK-RV64-SAME: <vscale x 16 x i1> [[VM:%.*]], <vscale x 16 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x bfloat> @llvm.riscv.vlse.mask.nxv16bf16.i64(<vscale x 16 x bfloat> [[VD]], ptr [[RS1]], i64 [[RS2]], <vscale x 16 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 16 x bfloat> [[TMP0]]
+//
+vbfloat16m4_t test_vlse16_v_bf16m4_tumu(vbool4_t vm, vbfloat16m4_t vd,
+                                        const __bf16 *rs1, ptrdiff_t rs2,
+                                        size_t vl) {
+  return __riscv_vlse16_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 32 x bfloat> @test_vlse16_v_bf16m8_tumu(
+// CHECK-RV64-SAME: <vscale x 32 x i1> [[VM:%.*]], <vscale x 32 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x bfloat> @llvm.riscv.vlse.mask.nxv32bf16.i64(<vscale x 32 x bfloat> [[VD]], ptr [[RS1]], i64 [[RS2]], <vscale x 32 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 32 x bfloat> [[TMP0]]
+//
+vbfloat16m8_t test_vlse16_v_bf16m8_tumu(vbool2_t vm, vbfloat16m8_t vd,
+                                        const __bf16 *rs1, ptrdiff_t rs2,
+                                        size_t vl) {
+  return __riscv_vlse16_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 1 x bfloat> @test_vlse16_v_bf16mf4_mu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], <vscale x 1 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x bfloat> @llvm.riscv.vlse.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[VD]], ptr [[RS1]], i64 [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret <vscale x 1 x bfloat> [[TMP0]]
+//
+vbfloat16mf4_t test_vlse16_v_bf16mf4_mu(vbool64_t vm, vbfloat16mf4_t vd,
+                                        const __bf16 *rs1, ptrdiff_t rs2,
+                                        size_t vl) {
+  return __riscv_vlse16_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 2 x bfloat> @test_vlse16_v_bf16mf2_mu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], <vscale x 2 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x bfloat> @llvm.riscv.vlse.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[VD]], ptr [[RS1]], i64 [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret <vscale x 2 x bfloat> [[TMP0]]
+//
+vbfloat16mf2_t test_vlse16_v_bf16mf2_mu(vbool32_t vm, vbfloat16mf2_t vd,
+                                        const __bf16 *rs1, ptrdiff_t rs2,
+                                        size_t vl) {
+  return __riscv_vlse16_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 4 x bfloat> @test_vlse16_v_bf16m1_mu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], <vscale x 4 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x bfloat> @llvm.riscv.vlse.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[VD]], ptr [[RS1]], i64 [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret <vscale x 4 x bfloat> [[TMP0]]
+//
+vbfloat16m1_t test_vlse16_v_bf16m1_mu(vbool16_t vm, vbfloat16m1_t vd,
+                                      const __bf16 *rs1, ptrdiff_t rs2,
+                                      size_t vl) {
+  return __riscv_vlse16_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 8 x bfloat> @test_vlse16_v_bf16m2_mu(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], <vscale x 8 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x bfloat> @llvm.riscv.vlse.mask.nxv8bf16.i64(<vscale x 8 x bfloat> [[VD]], ptr [[RS1]], i64 [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret <vscale x 8 x bfloat> [[TMP0]]
+//
+vbfloat16m2_t test_vlse16_v_bf16m2_mu(vbool8_t vm, vbfloat16m2_t vd,
+                                      const __bf16 *rs1, ptrdiff_t rs2,
+                                      size_t vl) {
+  return __riscv_vlse16_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 16 x bfloat> @test_vlse16_v_bf16m4_mu(
+// CHECK-RV64-SAME: <vscale x 16 x i1> [[VM:%.*]], <vscale x 16 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x bfloat> @llvm.riscv.vlse.mask.nxv16bf16.i64(<vscale x 16 x bfloat> [[VD]], ptr [[RS1]], i64 [[RS2]], <vscale x 16 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret <vscale x 16 x bfloat> [[TMP0]]
+//
+vbfloat16m4_t test_vlse16_v_bf16m4_mu(vbool4_t vm, vbfloat16m4_t vd,
+                                      const __bf16 *rs1, ptrdiff_t rs2,
+                                      size_t vl) {
+  return __riscv_vlse16_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 32 x bfloat> @test_vlse16_v_bf16m8_mu(
+// CHECK-RV64-SAME: <vscale x 32 x i1> [[VM:%.*]], <vscale x 32 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x bfloat> @llvm.riscv.vlse.mask.nxv32bf16.i64(<vscale x 32 x bfloat> [[VD]], ptr [[RS1]], i64 [[RS2]], <vscale x 32 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret <vscale x 32 x bfloat> [[TMP0]]
+//
+vbfloat16m8_t test_vlse16_v_bf16m8_mu(vbool2_t vm, vbfloat16m8_t vd,
+                                      const __bf16 *rs1, ptrdiff_t rs2,
+                                      size_t vl) {
+  return __riscv_vlse16_mu(vm, vd, rs1, rs2, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vlseg2e16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vlseg2e16.c
new file mode 100644
index 000000000000..2d15ad7713e0
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vlseg2e16.c
@@ -0,0 +1,275 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg2e16_v_bf16mf4x2_tu(
+// CHECK-RV64-SAME: { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlseg2.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP2]]
+//
+vbfloat16mf4x2_t test_vlseg2e16_v_bf16mf4x2_tu(vbfloat16mf4x2_t vd,
+                                               const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg2e16_tu(vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg2e16_v_bf16mf2x2_tu(
+// CHECK-RV64-SAME: { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlseg2.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP2]]
+//
+vbfloat16mf2x2_t test_vlseg2e16_v_bf16mf2x2_tu(vbfloat16mf2x2_t vd,
+                                               const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg2e16_tu(vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg2e16_v_bf16m1x2_tu(
+// CHECK-RV64-SAME: { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlseg2.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP2]]
+//
+vbfloat16m1x2_t test_vlseg2e16_v_bf16m1x2_tu(vbfloat16m1x2_t vd,
+                                             const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg2e16_tu(vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vlseg2e16_v_bf16m2x2_tu(
+// CHECK-RV64-SAME: { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vlseg2.nxv8bf16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP2]]
+//
+vbfloat16m2x2_t test_vlseg2e16_v_bf16m2x2_tu(vbfloat16m2x2_t vd,
+                                             const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg2e16_tu(vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @test_vlseg2e16_v_bf16m4x2_tu(
+// CHECK-RV64-SAME: { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @llvm.riscv.vlseg2.nxv16bf16.i64(<vscale x 16 x bfloat> [[TMP0]], <vscale x 16 x bfloat> [[TMP1]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[TMP2]]
+//
+vbfloat16m4x2_t test_vlseg2e16_v_bf16m4x2_tu(vbfloat16m4x2_t vd,
+                                             const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg2e16_tu(vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg2e16_v_bf16mf4x2_tum(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlseg2.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP2]]
+//
+vbfloat16mf4x2_t test_vlseg2e16_v_bf16mf4x2_tum(vbool64_t vm,
+                                                vbfloat16mf4x2_t vd,
+                                                const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg2e16_tum(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg2e16_v_bf16mf2x2_tum(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlseg2.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP2]]
+//
+vbfloat16mf2x2_t test_vlseg2e16_v_bf16mf2x2_tum(vbool32_t vm,
+                                                vbfloat16mf2x2_t vd,
+                                                const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg2e16_tum(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg2e16_v_bf16m1x2_tum(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlseg2.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP2]]
+//
+vbfloat16m1x2_t test_vlseg2e16_v_bf16m1x2_tum(vbool16_t vm, vbfloat16m1x2_t vd,
+                                              const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg2e16_tum(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vlseg2e16_v_bf16m2x2_tum(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vlseg2.mask.nxv8bf16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP2]]
+//
+vbfloat16m2x2_t test_vlseg2e16_v_bf16m2x2_tum(vbool8_t vm, vbfloat16m2x2_t vd,
+                                              const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg2e16_tum(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @test_vlseg2e16_v_bf16m4x2_tum(
+// CHECK-RV64-SAME: <vscale x 16 x i1> [[VM:%.*]], { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @llvm.riscv.vlseg2.mask.nxv16bf16.i64(<vscale x 16 x bfloat> [[TMP0]], <vscale x 16 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 16 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[TMP2]]
+//
+vbfloat16m4x2_t test_vlseg2e16_v_bf16m4x2_tum(vbool4_t vm, vbfloat16m4x2_t vd,
+                                              const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg2e16_tum(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg2e16_v_bf16mf4x2_tumu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlseg2.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP2]]
+//
+vbfloat16mf4x2_t test_vlseg2e16_v_bf16mf4x2_tumu(vbool64_t vm,
+                                                 vbfloat16mf4x2_t vd,
+                                                 const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg2e16_tumu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg2e16_v_bf16mf2x2_tumu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlseg2.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP2]]
+//
+vbfloat16mf2x2_t test_vlseg2e16_v_bf16mf2x2_tumu(vbool32_t vm,
+                                                 vbfloat16mf2x2_t vd,
+                                                 const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg2e16_tumu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg2e16_v_bf16m1x2_tumu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlseg2.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP2]]
+//
+vbfloat16m1x2_t test_vlseg2e16_v_bf16m1x2_tumu(vbool16_t vm, vbfloat16m1x2_t vd,
+                                               const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg2e16_tumu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vlseg2e16_v_bf16m2x2_tumu(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vlseg2.mask.nxv8bf16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP2]]
+//
+vbfloat16m2x2_t test_vlseg2e16_v_bf16m2x2_tumu(vbool8_t vm, vbfloat16m2x2_t vd,
+                                               const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg2e16_tumu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @test_vlseg2e16_v_bf16m4x2_tumu(
+// CHECK-RV64-SAME: <vscale x 16 x i1> [[VM:%.*]], { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @llvm.riscv.vlseg2.mask.nxv16bf16.i64(<vscale x 16 x bfloat> [[TMP0]], <vscale x 16 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 16 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[TMP2]]
+//
+vbfloat16m4x2_t test_vlseg2e16_v_bf16m4x2_tumu(vbool4_t vm, vbfloat16m4x2_t vd,
+                                               const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg2e16_tumu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg2e16_v_bf16mf4x2_mu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlseg2.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP2]]
+//
+vbfloat16mf4x2_t test_vlseg2e16_v_bf16mf4x2_mu(vbool64_t vm,
+                                               vbfloat16mf4x2_t vd,
+                                               const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg2e16_mu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg2e16_v_bf16mf2x2_mu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlseg2.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP2]]
+//
+vbfloat16mf2x2_t test_vlseg2e16_v_bf16mf2x2_mu(vbool32_t vm,
+                                               vbfloat16mf2x2_t vd,
+                                               const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg2e16_mu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg2e16_v_bf16m1x2_mu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlseg2.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP2]]
+//
+vbfloat16m1x2_t test_vlseg2e16_v_bf16m1x2_mu(vbool16_t vm, vbfloat16m1x2_t vd,
+                                             const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg2e16_mu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vlseg2e16_v_bf16m2x2_mu(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vlseg2.mask.nxv8bf16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP2]]
+//
+vbfloat16m2x2_t test_vlseg2e16_v_bf16m2x2_mu(vbool8_t vm, vbfloat16m2x2_t vd,
+                                             const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg2e16_mu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @test_vlseg2e16_v_bf16m4x2_mu(
+// CHECK-RV64-SAME: <vscale x 16 x i1> [[VM:%.*]], { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @llvm.riscv.vlseg2.mask.nxv16bf16.i64(<vscale x 16 x bfloat> [[TMP0]], <vscale x 16 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 16 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[TMP2]]
+//
+vbfloat16m4x2_t test_vlseg2e16_v_bf16m4x2_mu(vbool4_t vm, vbfloat16m4x2_t vd,
+                                             const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg2e16_mu(vm, vd, rs1, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vlseg2e16ff.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vlseg2e16ff.c
new file mode 100644
index 000000000000..d08cb9254e1e
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vlseg2e16ff.c
@@ -0,0 +1,419 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg2e16ff_v_bf16mf4x2_tu(
+// CHECK-RV64-SAME: { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } @llvm.riscv.vlseg2ff.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP2]], 0
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } poison, <vscale x 1 x bfloat> [[TMP3]], 0
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP2]], 1
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], 1
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP2]], 2
+// CHECK-RV64-NEXT:    store i64 [[TMP7]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP6]]
+//
+vbfloat16mf4x2_t test_vlseg2e16ff_v_bf16mf4x2_tu(vbfloat16mf4x2_t vd,
+                                                 const __bf16 *rs1,
+                                                 size_t *new_vl, size_t vl) {
+  return __riscv_vlseg2e16ff_tu(vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg2e16ff_v_bf16mf2x2_tu(
+// CHECK-RV64-SAME: { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } @llvm.riscv.vlseg2ff.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP2]], 0
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } poison, <vscale x 2 x bfloat> [[TMP3]], 0
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP2]], 1
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], 1
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP2]], 2
+// CHECK-RV64-NEXT:    store i64 [[TMP7]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP6]]
+//
+vbfloat16mf2x2_t test_vlseg2e16ff_v_bf16mf2x2_tu(vbfloat16mf2x2_t vd,
+                                                 const __bf16 *rs1,
+                                                 size_t *new_vl, size_t vl) {
+  return __riscv_vlseg2e16ff_tu(vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg2e16ff_v_bf16m1x2_tu(
+// CHECK-RV64-SAME: { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } @llvm.riscv.vlseg2ff.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP2]], 0
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } poison, <vscale x 4 x bfloat> [[TMP3]], 0
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP2]], 1
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], 1
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP2]], 2
+// CHECK-RV64-NEXT:    store i64 [[TMP7]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP6]]
+//
+vbfloat16m1x2_t test_vlseg2e16ff_v_bf16m1x2_tu(vbfloat16m1x2_t vd,
+                                               const __bf16 *rs1,
+                                               size_t *new_vl, size_t vl) {
+  return __riscv_vlseg2e16ff_tu(vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vlseg2e16ff_v_bf16m2x2_tu(
+// CHECK-RV64-SAME: { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } @llvm.riscv.vlseg2ff.nxv8bf16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP2]], 0
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } poison, <vscale x 8 x bfloat> [[TMP3]], 0
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP2]], 1
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP4]], <vscale x 8 x bfloat> [[TMP5]], 1
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP2]], 2
+// CHECK-RV64-NEXT:    store i64 [[TMP7]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP6]]
+//
+vbfloat16m2x2_t test_vlseg2e16ff_v_bf16m2x2_tu(vbfloat16m2x2_t vd,
+                                               const __bf16 *rs1,
+                                               size_t *new_vl, size_t vl) {
+  return __riscv_vlseg2e16ff_tu(vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @test_vlseg2e16ff_v_bf16m4x2_tu(
+// CHECK-RV64-SAME: { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 16 x bfloat>, <vscale x 16 x bfloat>, i64 } @llvm.riscv.vlseg2ff.nxv16bf16.i64(<vscale x 16 x bfloat> [[TMP0]], <vscale x 16 x bfloat> [[TMP1]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat>, i64 } [[TMP2]], 0
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } poison, <vscale x 16 x bfloat> [[TMP3]], 0
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat>, i64 } [[TMP2]], 1
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[TMP4]], <vscale x 16 x bfloat> [[TMP5]], 1
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat>, i64 } [[TMP2]], 2
+// CHECK-RV64-NEXT:    store i64 [[TMP7]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[TMP6]]
+//
+vbfloat16m4x2_t test_vlseg2e16ff_v_bf16m4x2_tu(vbfloat16m4x2_t vd,
+                                               const __bf16 *rs1,
+                                               size_t *new_vl, size_t vl) {
+  return __riscv_vlseg2e16ff_tu(vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg2e16ff_v_bf16mf4x2_tum(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } @llvm.riscv.vlseg2ff.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP2]], 0
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } poison, <vscale x 1 x bfloat> [[TMP3]], 0
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP2]], 1
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], 1
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP2]], 2
+// CHECK-RV64-NEXT:    store i64 [[TMP7]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP6]]
+//
+vbfloat16mf4x2_t test_vlseg2e16ff_v_bf16mf4x2_tum(vbool64_t vm,
+                                                  vbfloat16mf4x2_t vd,
+                                                  const __bf16 *rs1,
+                                                  size_t *new_vl, size_t vl) {
+  return __riscv_vlseg2e16ff_tum(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg2e16ff_v_bf16mf2x2_tum(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } @llvm.riscv.vlseg2ff.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP2]], 0
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } poison, <vscale x 2 x bfloat> [[TMP3]], 0
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP2]], 1
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], 1
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP2]], 2
+// CHECK-RV64-NEXT:    store i64 [[TMP7]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP6]]
+//
+vbfloat16mf2x2_t test_vlseg2e16ff_v_bf16mf2x2_tum(vbool32_t vm,
+                                                  vbfloat16mf2x2_t vd,
+                                                  const __bf16 *rs1,
+                                                  size_t *new_vl, size_t vl) {
+  return __riscv_vlseg2e16ff_tum(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg2e16ff_v_bf16m1x2_tum(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } @llvm.riscv.vlseg2ff.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP2]], 0
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } poison, <vscale x 4 x bfloat> [[TMP3]], 0
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP2]], 1
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], 1
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP2]], 2
+// CHECK-RV64-NEXT:    store i64 [[TMP7]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP6]]
+//
+vbfloat16m1x2_t test_vlseg2e16ff_v_bf16m1x2_tum(vbool16_t vm,
+                                                vbfloat16m1x2_t vd,
+                                                const __bf16 *rs1,
+                                                size_t *new_vl, size_t vl) {
+  return __riscv_vlseg2e16ff_tum(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vlseg2e16ff_v_bf16m2x2_tum(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } @llvm.riscv.vlseg2ff.mask.nxv8bf16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP2]], 0
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } poison, <vscale x 8 x bfloat> [[TMP3]], 0
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP2]], 1
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP4]], <vscale x 8 x bfloat> [[TMP5]], 1
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP2]], 2
+// CHECK-RV64-NEXT:    store i64 [[TMP7]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP6]]
+//
+vbfloat16m2x2_t test_vlseg2e16ff_v_bf16m2x2_tum(vbool8_t vm, vbfloat16m2x2_t vd,
+                                                const __bf16 *rs1,
+                                                size_t *new_vl, size_t vl) {
+  return __riscv_vlseg2e16ff_tum(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @test_vlseg2e16ff_v_bf16m4x2_tum(
+// CHECK-RV64-SAME: <vscale x 16 x i1> [[VM:%.*]], { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 16 x bfloat>, <vscale x 16 x bfloat>, i64 } @llvm.riscv.vlseg2ff.mask.nxv16bf16.i64(<vscale x 16 x bfloat> [[TMP0]], <vscale x 16 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 16 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat>, i64 } [[TMP2]], 0
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } poison, <vscale x 16 x bfloat> [[TMP3]], 0
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat>, i64 } [[TMP2]], 1
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[TMP4]], <vscale x 16 x bfloat> [[TMP5]], 1
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat>, i64 } [[TMP2]], 2
+// CHECK-RV64-NEXT:    store i64 [[TMP7]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[TMP6]]
+//
+vbfloat16m4x2_t test_vlseg2e16ff_v_bf16m4x2_tum(vbool4_t vm, vbfloat16m4x2_t vd,
+                                                const __bf16 *rs1,
+                                                size_t *new_vl, size_t vl) {
+  return __riscv_vlseg2e16ff_tum(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg2e16ff_v_bf16mf4x2_tumu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } @llvm.riscv.vlseg2ff.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP2]], 0
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } poison, <vscale x 1 x bfloat> [[TMP3]], 0
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP2]], 1
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], 1
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP2]], 2
+// CHECK-RV64-NEXT:    store i64 [[TMP7]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP6]]
+//
+vbfloat16mf4x2_t test_vlseg2e16ff_v_bf16mf4x2_tumu(vbool64_t vm,
+                                                   vbfloat16mf4x2_t vd,
+                                                   const __bf16 *rs1,
+                                                   size_t *new_vl, size_t vl) {
+  return __riscv_vlseg2e16ff_tumu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg2e16ff_v_bf16mf2x2_tumu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } @llvm.riscv.vlseg2ff.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP2]], 0
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } poison, <vscale x 2 x bfloat> [[TMP3]], 0
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP2]], 1
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], 1
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP2]], 2
+// CHECK-RV64-NEXT:    store i64 [[TMP7]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP6]]
+//
+vbfloat16mf2x2_t test_vlseg2e16ff_v_bf16mf2x2_tumu(vbool32_t vm,
+                                                   vbfloat16mf2x2_t vd,
+                                                   const __bf16 *rs1,
+                                                   size_t *new_vl, size_t vl) {
+  return __riscv_vlseg2e16ff_tumu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg2e16ff_v_bf16m1x2_tumu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } @llvm.riscv.vlseg2ff.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP2]], 0
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } poison, <vscale x 4 x bfloat> [[TMP3]], 0
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP2]], 1
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], 1
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP2]], 2
+// CHECK-RV64-NEXT:    store i64 [[TMP7]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP6]]
+//
+vbfloat16m1x2_t test_vlseg2e16ff_v_bf16m1x2_tumu(vbool16_t vm,
+                                                 vbfloat16m1x2_t vd,
+                                                 const __bf16 *rs1,
+                                                 size_t *new_vl, size_t vl) {
+  return __riscv_vlseg2e16ff_tumu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vlseg2e16ff_v_bf16m2x2_tumu(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } @llvm.riscv.vlseg2ff.mask.nxv8bf16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP2]], 0
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } poison, <vscale x 8 x bfloat> [[TMP3]], 0
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP2]], 1
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP4]], <vscale x 8 x bfloat> [[TMP5]], 1
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP2]], 2
+// CHECK-RV64-NEXT:    store i64 [[TMP7]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP6]]
+//
+vbfloat16m2x2_t test_vlseg2e16ff_v_bf16m2x2_tumu(vbool8_t vm,
+                                                 vbfloat16m2x2_t vd,
+                                                 const __bf16 *rs1,
+                                                 size_t *new_vl, size_t vl) {
+  return __riscv_vlseg2e16ff_tumu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @test_vlseg2e16ff_v_bf16m4x2_tumu(
+// CHECK-RV64-SAME: <vscale x 16 x i1> [[VM:%.*]], { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 16 x bfloat>, <vscale x 16 x bfloat>, i64 } @llvm.riscv.vlseg2ff.mask.nxv16bf16.i64(<vscale x 16 x bfloat> [[TMP0]], <vscale x 16 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 16 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat>, i64 } [[TMP2]], 0
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } poison, <vscale x 16 x bfloat> [[TMP3]], 0
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat>, i64 } [[TMP2]], 1
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[TMP4]], <vscale x 16 x bfloat> [[TMP5]], 1
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat>, i64 } [[TMP2]], 2
+// CHECK-RV64-NEXT:    store i64 [[TMP7]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[TMP6]]
+//
+vbfloat16m4x2_t test_vlseg2e16ff_v_bf16m4x2_tumu(vbool4_t vm,
+                                                 vbfloat16m4x2_t vd,
+                                                 const __bf16 *rs1,
+                                                 size_t *new_vl, size_t vl) {
+  return __riscv_vlseg2e16ff_tumu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg2e16ff_v_bf16mf4x2_mu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } @llvm.riscv.vlseg2ff.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP2]], 0
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } poison, <vscale x 1 x bfloat> [[TMP3]], 0
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP2]], 1
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], 1
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP2]], 2
+// CHECK-RV64-NEXT:    store i64 [[TMP7]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP6]]
+//
+vbfloat16mf4x2_t test_vlseg2e16ff_v_bf16mf4x2_mu(vbool64_t vm,
+                                                 vbfloat16mf4x2_t vd,
+                                                 const __bf16 *rs1,
+                                                 size_t *new_vl, size_t vl) {
+  return __riscv_vlseg2e16ff_mu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg2e16ff_v_bf16mf2x2_mu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } @llvm.riscv.vlseg2ff.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP2]], 0
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } poison, <vscale x 2 x bfloat> [[TMP3]], 0
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP2]], 1
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], 1
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP2]], 2
+// CHECK-RV64-NEXT:    store i64 [[TMP7]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP6]]
+//
+vbfloat16mf2x2_t test_vlseg2e16ff_v_bf16mf2x2_mu(vbool32_t vm,
+                                                 vbfloat16mf2x2_t vd,
+                                                 const __bf16 *rs1,
+                                                 size_t *new_vl, size_t vl) {
+  return __riscv_vlseg2e16ff_mu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg2e16ff_v_bf16m1x2_mu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } @llvm.riscv.vlseg2ff.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP2]], 0
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } poison, <vscale x 4 x bfloat> [[TMP3]], 0
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP2]], 1
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], 1
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP2]], 2
+// CHECK-RV64-NEXT:    store i64 [[TMP7]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP6]]
+//
+vbfloat16m1x2_t test_vlseg2e16ff_v_bf16m1x2_mu(vbool16_t vm, vbfloat16m1x2_t vd,
+                                               const __bf16 *rs1,
+                                               size_t *new_vl, size_t vl) {
+  return __riscv_vlseg2e16ff_mu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vlseg2e16ff_v_bf16m2x2_mu(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } @llvm.riscv.vlseg2ff.mask.nxv8bf16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP2]], 0
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } poison, <vscale x 8 x bfloat> [[TMP3]], 0
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP2]], 1
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP4]], <vscale x 8 x bfloat> [[TMP5]], 1
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP2]], 2
+// CHECK-RV64-NEXT:    store i64 [[TMP7]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP6]]
+//
+vbfloat16m2x2_t test_vlseg2e16ff_v_bf16m2x2_mu(vbool8_t vm, vbfloat16m2x2_t vd,
+                                               const __bf16 *rs1,
+                                               size_t *new_vl, size_t vl) {
+  return __riscv_vlseg2e16ff_mu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @test_vlseg2e16ff_v_bf16m4x2_mu(
+// CHECK-RV64-SAME: <vscale x 16 x i1> [[VM:%.*]], { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 16 x bfloat>, <vscale x 16 x bfloat>, i64 } @llvm.riscv.vlseg2ff.mask.nxv16bf16.i64(<vscale x 16 x bfloat> [[TMP0]], <vscale x 16 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 16 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat>, i64 } [[TMP2]], 0
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = insertvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } poison, <vscale x 16 x bfloat> [[TMP3]], 0
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat>, i64 } [[TMP2]], 1
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[TMP4]], <vscale x 16 x bfloat> [[TMP5]], 1
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat>, i64 } [[TMP2]], 2
+// CHECK-RV64-NEXT:    store i64 [[TMP7]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[TMP6]]
+//
+vbfloat16m4x2_t test_vlseg2e16ff_v_bf16m4x2_mu(vbool4_t vm, vbfloat16m4x2_t vd,
+                                               const __bf16 *rs1,
+                                               size_t *new_vl, size_t vl) {
+  return __riscv_vlseg2e16ff_mu(vm, vd, rs1, new_vl, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vlseg3e16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vlseg3e16.c
new file mode 100644
index 000000000000..336e8d0b8983
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vlseg3e16.c
@@ -0,0 +1,239 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg3e16_v_bf16mf4x3_tu(
+// CHECK-RV64-SAME: { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlseg3.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP3]]
+//
+vbfloat16mf4x3_t test_vlseg3e16_v_bf16mf4x3_tu(vbfloat16mf4x3_t vd,
+                                               const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg3e16_tu(vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg3e16_v_bf16mf2x3_tu(
+// CHECK-RV64-SAME: { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlseg3.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP3]]
+//
+vbfloat16mf2x3_t test_vlseg3e16_v_bf16mf2x3_tu(vbfloat16mf2x3_t vd,
+                                               const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg3e16_tu(vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg3e16_v_bf16m1x3_tu(
+// CHECK-RV64-SAME: { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlseg3.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP3]]
+//
+vbfloat16m1x3_t test_vlseg3e16_v_bf16m1x3_tu(vbfloat16m1x3_t vd,
+                                             const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg3e16_tu(vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vlseg3e16_v_bf16m2x3_tu(
+// CHECK-RV64-SAME: { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vlseg3.nxv8bf16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP3]]
+//
+vbfloat16m2x3_t test_vlseg3e16_v_bf16m2x3_tu(vbfloat16m2x3_t vd,
+                                             const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg3e16_tu(vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg3e16_v_bf16mf4x3_tum(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlseg3.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP3]]
+//
+vbfloat16mf4x3_t test_vlseg3e16_v_bf16mf4x3_tum(vbool64_t vm,
+                                                vbfloat16mf4x3_t vd,
+                                                const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg3e16_tum(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg3e16_v_bf16mf2x3_tum(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlseg3.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP3]]
+//
+vbfloat16mf2x3_t test_vlseg3e16_v_bf16mf2x3_tum(vbool32_t vm,
+                                                vbfloat16mf2x3_t vd,
+                                                const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg3e16_tum(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg3e16_v_bf16m1x3_tum(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlseg3.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP3]]
+//
+vbfloat16m1x3_t test_vlseg3e16_v_bf16m1x3_tum(vbool16_t vm, vbfloat16m1x3_t vd,
+                                              const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg3e16_tum(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vlseg3e16_v_bf16m2x3_tum(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vlseg3.mask.nxv8bf16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP3]]
+//
+vbfloat16m2x3_t test_vlseg3e16_v_bf16m2x3_tum(vbool8_t vm, vbfloat16m2x3_t vd,
+                                              const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg3e16_tum(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg3e16_v_bf16mf4x3_tumu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlseg3.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP3]]
+//
+vbfloat16mf4x3_t test_vlseg3e16_v_bf16mf4x3_tumu(vbool64_t vm,
+                                                 vbfloat16mf4x3_t vd,
+                                                 const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg3e16_tumu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg3e16_v_bf16mf2x3_tumu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlseg3.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP3]]
+//
+vbfloat16mf2x3_t test_vlseg3e16_v_bf16mf2x3_tumu(vbool32_t vm,
+                                                 vbfloat16mf2x3_t vd,
+                                                 const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg3e16_tumu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg3e16_v_bf16m1x3_tumu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlseg3.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP3]]
+//
+vbfloat16m1x3_t test_vlseg3e16_v_bf16m1x3_tumu(vbool16_t vm, vbfloat16m1x3_t vd,
+                                               const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg3e16_tumu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vlseg3e16_v_bf16m2x3_tumu(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vlseg3.mask.nxv8bf16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP3]]
+//
+vbfloat16m2x3_t test_vlseg3e16_v_bf16m2x3_tumu(vbool8_t vm, vbfloat16m2x3_t vd,
+                                               const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg3e16_tumu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg3e16_v_bf16mf4x3_mu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlseg3.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP3]]
+//
+vbfloat16mf4x3_t test_vlseg3e16_v_bf16mf4x3_mu(vbool64_t vm,
+                                               vbfloat16mf4x3_t vd,
+                                               const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg3e16_mu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg3e16_v_bf16mf2x3_mu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlseg3.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP3]]
+//
+vbfloat16mf2x3_t test_vlseg3e16_v_bf16mf2x3_mu(vbool32_t vm,
+                                               vbfloat16mf2x3_t vd,
+                                               const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg3e16_mu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg3e16_v_bf16m1x3_mu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlseg3.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP3]]
+//
+vbfloat16m1x3_t test_vlseg3e16_v_bf16m1x3_mu(vbool16_t vm, vbfloat16m1x3_t vd,
+                                             const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg3e16_mu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vlseg3e16_v_bf16m2x3_mu(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vlseg3.mask.nxv8bf16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP3]]
+//
+vbfloat16m2x3_t test_vlseg3e16_v_bf16m2x3_mu(vbool8_t vm, vbfloat16m2x3_t vd,
+                                             const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg3e16_mu(vm, vd, rs1, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vlseg3e16ff.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vlseg3e16ff.c
new file mode 100644
index 000000000000..7dc19f9199c6
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vlseg3e16ff.c
@@ -0,0 +1,386 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg3e16ff_v_bf16mf4x3_tu(
+// CHECK-RV64-SAME: { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } @llvm.riscv.vlseg3ff.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP3]], 0
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } poison, <vscale x 1 x bfloat> [[TMP4]], 0
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], 1
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP3]], 2
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP7]], <vscale x 1 x bfloat> [[TMP8]], 2
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP3]], 3
+// CHECK-RV64-NEXT:    store i64 [[TMP10]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP9]]
+//
+vbfloat16mf4x3_t test_vlseg3e16ff_v_bf16mf4x3_tu(vbfloat16mf4x3_t vd,
+                                                 const __bf16 *rs1,
+                                                 size_t *new_vl, size_t vl) {
+  return __riscv_vlseg3e16ff_tu(vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg3e16ff_v_bf16mf2x3_tu(
+// CHECK-RV64-SAME: { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } @llvm.riscv.vlseg3ff.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP3]], 0
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } poison, <vscale x 2 x bfloat> [[TMP4]], 0
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], 1
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP3]], 2
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP7]], <vscale x 2 x bfloat> [[TMP8]], 2
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP3]], 3
+// CHECK-RV64-NEXT:    store i64 [[TMP10]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP9]]
+//
+vbfloat16mf2x3_t test_vlseg3e16ff_v_bf16mf2x3_tu(vbfloat16mf2x3_t vd,
+                                                 const __bf16 *rs1,
+                                                 size_t *new_vl, size_t vl) {
+  return __riscv_vlseg3e16ff_tu(vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg3e16ff_v_bf16m1x3_tu(
+// CHECK-RV64-SAME: { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } @llvm.riscv.vlseg3ff.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP3]], 0
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } poison, <vscale x 4 x bfloat> [[TMP4]], 0
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], 1
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP3]], 2
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP7]], <vscale x 4 x bfloat> [[TMP8]], 2
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP3]], 3
+// CHECK-RV64-NEXT:    store i64 [[TMP10]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP9]]
+//
+vbfloat16m1x3_t test_vlseg3e16ff_v_bf16m1x3_tu(vbfloat16m1x3_t vd,
+                                               const __bf16 *rs1,
+                                               size_t *new_vl, size_t vl) {
+  return __riscv_vlseg3e16ff_tu(vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vlseg3e16ff_v_bf16m2x3_tu(
+// CHECK-RV64-SAME: { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } @llvm.riscv.vlseg3ff.nxv8bf16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP3]], 0
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } poison, <vscale x 8 x bfloat> [[TMP4]], 0
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP5]], <vscale x 8 x bfloat> [[TMP6]], 1
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP3]], 2
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP7]], <vscale x 8 x bfloat> [[TMP8]], 2
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP3]], 3
+// CHECK-RV64-NEXT:    store i64 [[TMP10]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP9]]
+//
+vbfloat16m2x3_t test_vlseg3e16ff_v_bf16m2x3_tu(vbfloat16m2x3_t vd,
+                                               const __bf16 *rs1,
+                                               size_t *new_vl, size_t vl) {
+  return __riscv_vlseg3e16ff_tu(vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg3e16ff_v_bf16mf4x3_tum(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } @llvm.riscv.vlseg3ff.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP3]], 0
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } poison, <vscale x 1 x bfloat> [[TMP4]], 0
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], 1
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP3]], 2
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP7]], <vscale x 1 x bfloat> [[TMP8]], 2
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP3]], 3
+// CHECK-RV64-NEXT:    store i64 [[TMP10]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP9]]
+//
+vbfloat16mf4x3_t test_vlseg3e16ff_v_bf16mf4x3_tum(vbool64_t vm,
+                                                  vbfloat16mf4x3_t vd,
+                                                  const __bf16 *rs1,
+                                                  size_t *new_vl, size_t vl) {
+  return __riscv_vlseg3e16ff_tum(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg3e16ff_v_bf16mf2x3_tum(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } @llvm.riscv.vlseg3ff.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP3]], 0
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } poison, <vscale x 2 x bfloat> [[TMP4]], 0
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], 1
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP3]], 2
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP7]], <vscale x 2 x bfloat> [[TMP8]], 2
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP3]], 3
+// CHECK-RV64-NEXT:    store i64 [[TMP10]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP9]]
+//
+vbfloat16mf2x3_t test_vlseg3e16ff_v_bf16mf2x3_tum(vbool32_t vm,
+                                                  vbfloat16mf2x3_t vd,
+                                                  const __bf16 *rs1,
+                                                  size_t *new_vl, size_t vl) {
+  return __riscv_vlseg3e16ff_tum(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg3e16ff_v_bf16m1x3_tum(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } @llvm.riscv.vlseg3ff.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP3]], 0
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } poison, <vscale x 4 x bfloat> [[TMP4]], 0
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], 1
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP3]], 2
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP7]], <vscale x 4 x bfloat> [[TMP8]], 2
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP3]], 3
+// CHECK-RV64-NEXT:    store i64 [[TMP10]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP9]]
+//
+vbfloat16m1x3_t test_vlseg3e16ff_v_bf16m1x3_tum(vbool16_t vm,
+                                                vbfloat16m1x3_t vd,
+                                                const __bf16 *rs1,
+                                                size_t *new_vl, size_t vl) {
+  return __riscv_vlseg3e16ff_tum(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vlseg3e16ff_v_bf16m2x3_tum(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } @llvm.riscv.vlseg3ff.mask.nxv8bf16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP3]], 0
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } poison, <vscale x 8 x bfloat> [[TMP4]], 0
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP5]], <vscale x 8 x bfloat> [[TMP6]], 1
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP3]], 2
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP7]], <vscale x 8 x bfloat> [[TMP8]], 2
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP3]], 3
+// CHECK-RV64-NEXT:    store i64 [[TMP10]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP9]]
+//
+vbfloat16m2x3_t test_vlseg3e16ff_v_bf16m2x3_tum(vbool8_t vm, vbfloat16m2x3_t vd,
+                                                const __bf16 *rs1,
+                                                size_t *new_vl, size_t vl) {
+  return __riscv_vlseg3e16ff_tum(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg3e16ff_v_bf16mf4x3_tumu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } @llvm.riscv.vlseg3ff.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP3]], 0
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } poison, <vscale x 1 x bfloat> [[TMP4]], 0
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], 1
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP3]], 2
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP7]], <vscale x 1 x bfloat> [[TMP8]], 2
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP3]], 3
+// CHECK-RV64-NEXT:    store i64 [[TMP10]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP9]]
+//
+vbfloat16mf4x3_t test_vlseg3e16ff_v_bf16mf4x3_tumu(vbool64_t vm,
+                                                   vbfloat16mf4x3_t vd,
+                                                   const __bf16 *rs1,
+                                                   size_t *new_vl, size_t vl) {
+  return __riscv_vlseg3e16ff_tumu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg3e16ff_v_bf16mf2x3_tumu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } @llvm.riscv.vlseg3ff.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP3]], 0
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } poison, <vscale x 2 x bfloat> [[TMP4]], 0
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], 1
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP3]], 2
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP7]], <vscale x 2 x bfloat> [[TMP8]], 2
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP3]], 3
+// CHECK-RV64-NEXT:    store i64 [[TMP10]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP9]]
+//
+vbfloat16mf2x3_t test_vlseg3e16ff_v_bf16mf2x3_tumu(vbool32_t vm,
+                                                   vbfloat16mf2x3_t vd,
+                                                   const __bf16 *rs1,
+                                                   size_t *new_vl, size_t vl) {
+  return __riscv_vlseg3e16ff_tumu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg3e16ff_v_bf16m1x3_tumu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } @llvm.riscv.vlseg3ff.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP3]], 0
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } poison, <vscale x 4 x bfloat> [[TMP4]], 0
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], 1
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP3]], 2
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP7]], <vscale x 4 x bfloat> [[TMP8]], 2
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP3]], 3
+// CHECK-RV64-NEXT:    store i64 [[TMP10]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP9]]
+//
+vbfloat16m1x3_t test_vlseg3e16ff_v_bf16m1x3_tumu(vbool16_t vm,
+                                                 vbfloat16m1x3_t vd,
+                                                 const __bf16 *rs1,
+                                                 size_t *new_vl, size_t vl) {
+  return __riscv_vlseg3e16ff_tumu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vlseg3e16ff_v_bf16m2x3_tumu(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } @llvm.riscv.vlseg3ff.mask.nxv8bf16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP3]], 0
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } poison, <vscale x 8 x bfloat> [[TMP4]], 0
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP5]], <vscale x 8 x bfloat> [[TMP6]], 1
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP3]], 2
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP7]], <vscale x 8 x bfloat> [[TMP8]], 2
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP3]], 3
+// CHECK-RV64-NEXT:    store i64 [[TMP10]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP9]]
+//
+vbfloat16m2x3_t test_vlseg3e16ff_v_bf16m2x3_tumu(vbool8_t vm,
+                                                 vbfloat16m2x3_t vd,
+                                                 const __bf16 *rs1,
+                                                 size_t *new_vl, size_t vl) {
+  return __riscv_vlseg3e16ff_tumu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg3e16ff_v_bf16mf4x3_mu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } @llvm.riscv.vlseg3ff.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP3]], 0
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } poison, <vscale x 1 x bfloat> [[TMP4]], 0
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], 1
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP3]], 2
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP7]], <vscale x 1 x bfloat> [[TMP8]], 2
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP3]], 3
+// CHECK-RV64-NEXT:    store i64 [[TMP10]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP9]]
+//
+vbfloat16mf4x3_t test_vlseg3e16ff_v_bf16mf4x3_mu(vbool64_t vm,
+                                                 vbfloat16mf4x3_t vd,
+                                                 const __bf16 *rs1,
+                                                 size_t *new_vl, size_t vl) {
+  return __riscv_vlseg3e16ff_mu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg3e16ff_v_bf16mf2x3_mu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } @llvm.riscv.vlseg3ff.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP3]], 0
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } poison, <vscale x 2 x bfloat> [[TMP4]], 0
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], 1
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP3]], 2
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP7]], <vscale x 2 x bfloat> [[TMP8]], 2
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP3]], 3
+// CHECK-RV64-NEXT:    store i64 [[TMP10]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP9]]
+//
+vbfloat16mf2x3_t test_vlseg3e16ff_v_bf16mf2x3_mu(vbool32_t vm,
+                                                 vbfloat16mf2x3_t vd,
+                                                 const __bf16 *rs1,
+                                                 size_t *new_vl, size_t vl) {
+  return __riscv_vlseg3e16ff_mu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg3e16ff_v_bf16m1x3_mu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } @llvm.riscv.vlseg3ff.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP3]], 0
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } poison, <vscale x 4 x bfloat> [[TMP4]], 0
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], 1
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP3]], 2
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP7]], <vscale x 4 x bfloat> [[TMP8]], 2
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP3]], 3
+// CHECK-RV64-NEXT:    store i64 [[TMP10]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP9]]
+//
+vbfloat16m1x3_t test_vlseg3e16ff_v_bf16m1x3_mu(vbool16_t vm, vbfloat16m1x3_t vd,
+                                               const __bf16 *rs1,
+                                               size_t *new_vl, size_t vl) {
+  return __riscv_vlseg3e16ff_mu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vlseg3e16ff_v_bf16m2x3_mu(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } @llvm.riscv.vlseg3ff.mask.nxv8bf16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP3]], 0
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } poison, <vscale x 8 x bfloat> [[TMP4]], 0
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP3]], 1
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP5]], <vscale x 8 x bfloat> [[TMP6]], 1
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP3]], 2
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP7]], <vscale x 8 x bfloat> [[TMP8]], 2
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP3]], 3
+// CHECK-RV64-NEXT:    store i64 [[TMP10]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP9]]
+//
+vbfloat16m2x3_t test_vlseg3e16ff_v_bf16m2x3_mu(vbool8_t vm, vbfloat16m2x3_t vd,
+                                               const __bf16 *rs1,
+                                               size_t *new_vl, size_t vl) {
+  return __riscv_vlseg3e16ff_mu(vm, vd, rs1, new_vl, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vlseg4e16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vlseg4e16.c
new file mode 100644
index 000000000000..ddb335cc242e
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vlseg4e16.c
@@ -0,0 +1,255 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg4e16_v_bf16mf4x4_tu(
+// CHECK-RV64-SAME: { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlseg4.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP4]]
+//
+vbfloat16mf4x4_t test_vlseg4e16_v_bf16mf4x4_tu(vbfloat16mf4x4_t vd,
+                                               const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg4e16_tu(vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg4e16_v_bf16mf2x4_tu(
+// CHECK-RV64-SAME: { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlseg4.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP4]]
+//
+vbfloat16mf2x4_t test_vlseg4e16_v_bf16mf2x4_tu(vbfloat16mf2x4_t vd,
+                                               const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg4e16_tu(vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg4e16_v_bf16m1x4_tu(
+// CHECK-RV64-SAME: { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlseg4.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP4]]
+//
+vbfloat16m1x4_t test_vlseg4e16_v_bf16m1x4_tu(vbfloat16m1x4_t vd,
+                                             const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg4e16_tu(vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vlseg4e16_v_bf16m2x4_tu(
+// CHECK-RV64-SAME: { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vlseg4.nxv8bf16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], <vscale x 8 x bfloat> [[TMP3]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP4]]
+//
+vbfloat16m2x4_t test_vlseg4e16_v_bf16m2x4_tu(vbfloat16m2x4_t vd,
+                                             const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg4e16_tu(vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg4e16_v_bf16mf4x4_tum(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlseg4.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP4]]
+//
+vbfloat16mf4x4_t test_vlseg4e16_v_bf16mf4x4_tum(vbool64_t vm,
+                                                vbfloat16mf4x4_t vd,
+                                                const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg4e16_tum(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg4e16_v_bf16mf2x4_tum(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlseg4.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP4]]
+//
+vbfloat16mf2x4_t test_vlseg4e16_v_bf16mf2x4_tum(vbool32_t vm,
+                                                vbfloat16mf2x4_t vd,
+                                                const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg4e16_tum(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg4e16_v_bf16m1x4_tum(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlseg4.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP4]]
+//
+vbfloat16m1x4_t test_vlseg4e16_v_bf16m1x4_tum(vbool16_t vm, vbfloat16m1x4_t vd,
+                                              const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg4e16_tum(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vlseg4e16_v_bf16m2x4_tum(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vlseg4.mask.nxv8bf16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], <vscale x 8 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP4]]
+//
+vbfloat16m2x4_t test_vlseg4e16_v_bf16m2x4_tum(vbool8_t vm, vbfloat16m2x4_t vd,
+                                              const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg4e16_tum(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg4e16_v_bf16mf4x4_tumu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlseg4.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP4]]
+//
+vbfloat16mf4x4_t test_vlseg4e16_v_bf16mf4x4_tumu(vbool64_t vm,
+                                                 vbfloat16mf4x4_t vd,
+                                                 const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg4e16_tumu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg4e16_v_bf16mf2x4_tumu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlseg4.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP4]]
+//
+vbfloat16mf2x4_t test_vlseg4e16_v_bf16mf2x4_tumu(vbool32_t vm,
+                                                 vbfloat16mf2x4_t vd,
+                                                 const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg4e16_tumu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg4e16_v_bf16m1x4_tumu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlseg4.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP4]]
+//
+vbfloat16m1x4_t test_vlseg4e16_v_bf16m1x4_tumu(vbool16_t vm, vbfloat16m1x4_t vd,
+                                               const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg4e16_tumu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vlseg4e16_v_bf16m2x4_tumu(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vlseg4.mask.nxv8bf16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], <vscale x 8 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP4]]
+//
+vbfloat16m2x4_t test_vlseg4e16_v_bf16m2x4_tumu(vbool8_t vm, vbfloat16m2x4_t vd,
+                                               const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg4e16_tumu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg4e16_v_bf16mf4x4_mu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlseg4.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP4]]
+//
+vbfloat16mf4x4_t test_vlseg4e16_v_bf16mf4x4_mu(vbool64_t vm,
+                                               vbfloat16mf4x4_t vd,
+                                               const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg4e16_mu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg4e16_v_bf16mf2x4_mu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlseg4.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP4]]
+//
+vbfloat16mf2x4_t test_vlseg4e16_v_bf16mf2x4_mu(vbool32_t vm,
+                                               vbfloat16mf2x4_t vd,
+                                               const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg4e16_mu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg4e16_v_bf16m1x4_mu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlseg4.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP4]]
+//
+vbfloat16m1x4_t test_vlseg4e16_v_bf16m1x4_mu(vbool16_t vm, vbfloat16m1x4_t vd,
+                                             const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg4e16_mu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vlseg4e16_v_bf16m2x4_mu(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vlseg4.mask.nxv8bf16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], <vscale x 8 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP4]]
+//
+vbfloat16m2x4_t test_vlseg4e16_v_bf16m2x4_mu(vbool8_t vm, vbfloat16m2x4_t vd,
+                                             const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg4e16_mu(vm, vd, rs1, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vlseg4e16ff.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vlseg4e16ff.c
new file mode 100644
index 000000000000..b16766c184d1
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vlseg4e16ff.c
@@ -0,0 +1,434 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg4e16ff_v_bf16mf4x4_tu(
+// CHECK-RV64-SAME: { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } @llvm.riscv.vlseg4ff.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP4]], 0
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } poison, <vscale x 1 x bfloat> [[TMP5]], 0
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP4]], 1
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP6]], <vscale x 1 x bfloat> [[TMP7]], 1
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP4]], 2
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP8]], <vscale x 1 x bfloat> [[TMP9]], 2
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP4]], 3
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP10]], <vscale x 1 x bfloat> [[TMP11]], 3
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP4]], 4
+// CHECK-RV64-NEXT:    store i64 [[TMP13]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP12]]
+//
+vbfloat16mf4x4_t test_vlseg4e16ff_v_bf16mf4x4_tu(vbfloat16mf4x4_t vd,
+                                                 const __bf16 *rs1,
+                                                 size_t *new_vl, size_t vl) {
+  return __riscv_vlseg4e16ff_tu(vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg4e16ff_v_bf16mf2x4_tu(
+// CHECK-RV64-SAME: { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } @llvm.riscv.vlseg4ff.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP4]], 0
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } poison, <vscale x 2 x bfloat> [[TMP5]], 0
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP4]], 1
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP6]], <vscale x 2 x bfloat> [[TMP7]], 1
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP4]], 2
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP8]], <vscale x 2 x bfloat> [[TMP9]], 2
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP4]], 3
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP10]], <vscale x 2 x bfloat> [[TMP11]], 3
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP4]], 4
+// CHECK-RV64-NEXT:    store i64 [[TMP13]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP12]]
+//
+vbfloat16mf2x4_t test_vlseg4e16ff_v_bf16mf2x4_tu(vbfloat16mf2x4_t vd,
+                                                 const __bf16 *rs1,
+                                                 size_t *new_vl, size_t vl) {
+  return __riscv_vlseg4e16ff_tu(vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg4e16ff_v_bf16m1x4_tu(
+// CHECK-RV64-SAME: { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } @llvm.riscv.vlseg4ff.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP4]], 0
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } poison, <vscale x 4 x bfloat> [[TMP5]], 0
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP4]], 1
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP6]], <vscale x 4 x bfloat> [[TMP7]], 1
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP4]], 2
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP8]], <vscale x 4 x bfloat> [[TMP9]], 2
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP4]], 3
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP10]], <vscale x 4 x bfloat> [[TMP11]], 3
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP4]], 4
+// CHECK-RV64-NEXT:    store i64 [[TMP13]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP12]]
+//
+vbfloat16m1x4_t test_vlseg4e16ff_v_bf16m1x4_tu(vbfloat16m1x4_t vd,
+                                               const __bf16 *rs1,
+                                               size_t *new_vl, size_t vl) {
+  return __riscv_vlseg4e16ff_tu(vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vlseg4e16ff_v_bf16m2x4_tu(
+// CHECK-RV64-SAME: { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } @llvm.riscv.vlseg4ff.nxv8bf16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], <vscale x 8 x bfloat> [[TMP3]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP4]], 0
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } poison, <vscale x 8 x bfloat> [[TMP5]], 0
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP4]], 1
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP6]], <vscale x 8 x bfloat> [[TMP7]], 1
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP4]], 2
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP8]], <vscale x 8 x bfloat> [[TMP9]], 2
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP4]], 3
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP10]], <vscale x 8 x bfloat> [[TMP11]], 3
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP4]], 4
+// CHECK-RV64-NEXT:    store i64 [[TMP13]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP12]]
+//
+vbfloat16m2x4_t test_vlseg4e16ff_v_bf16m2x4_tu(vbfloat16m2x4_t vd,
+                                               const __bf16 *rs1,
+                                               size_t *new_vl, size_t vl) {
+  return __riscv_vlseg4e16ff_tu(vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg4e16ff_v_bf16mf4x4_tum(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } @llvm.riscv.vlseg4ff.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP4]], 0
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } poison, <vscale x 1 x bfloat> [[TMP5]], 0
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP4]], 1
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP6]], <vscale x 1 x bfloat> [[TMP7]], 1
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP4]], 2
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP8]], <vscale x 1 x bfloat> [[TMP9]], 2
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP4]], 3
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP10]], <vscale x 1 x bfloat> [[TMP11]], 3
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP4]], 4
+// CHECK-RV64-NEXT:    store i64 [[TMP13]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP12]]
+//
+vbfloat16mf4x4_t test_vlseg4e16ff_v_bf16mf4x4_tum(vbool64_t vm,
+                                                  vbfloat16mf4x4_t vd,
+                                                  const __bf16 *rs1,
+                                                  size_t *new_vl, size_t vl) {
+  return __riscv_vlseg4e16ff_tum(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg4e16ff_v_bf16mf2x4_tum(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } @llvm.riscv.vlseg4ff.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP4]], 0
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } poison, <vscale x 2 x bfloat> [[TMP5]], 0
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP4]], 1
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP6]], <vscale x 2 x bfloat> [[TMP7]], 1
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP4]], 2
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP8]], <vscale x 2 x bfloat> [[TMP9]], 2
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP4]], 3
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP10]], <vscale x 2 x bfloat> [[TMP11]], 3
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP4]], 4
+// CHECK-RV64-NEXT:    store i64 [[TMP13]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP12]]
+//
+vbfloat16mf2x4_t test_vlseg4e16ff_v_bf16mf2x4_tum(vbool32_t vm,
+                                                  vbfloat16mf2x4_t vd,
+                                                  const __bf16 *rs1,
+                                                  size_t *new_vl, size_t vl) {
+  return __riscv_vlseg4e16ff_tum(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg4e16ff_v_bf16m1x4_tum(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } @llvm.riscv.vlseg4ff.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP4]], 0
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } poison, <vscale x 4 x bfloat> [[TMP5]], 0
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP4]], 1
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP6]], <vscale x 4 x bfloat> [[TMP7]], 1
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP4]], 2
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP8]], <vscale x 4 x bfloat> [[TMP9]], 2
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP4]], 3
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP10]], <vscale x 4 x bfloat> [[TMP11]], 3
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP4]], 4
+// CHECK-RV64-NEXT:    store i64 [[TMP13]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP12]]
+//
+vbfloat16m1x4_t test_vlseg4e16ff_v_bf16m1x4_tum(vbool16_t vm,
+                                                vbfloat16m1x4_t vd,
+                                                const __bf16 *rs1,
+                                                size_t *new_vl, size_t vl) {
+  return __riscv_vlseg4e16ff_tum(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vlseg4e16ff_v_bf16m2x4_tum(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } @llvm.riscv.vlseg4ff.mask.nxv8bf16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], <vscale x 8 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP4]], 0
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } poison, <vscale x 8 x bfloat> [[TMP5]], 0
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP4]], 1
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP6]], <vscale x 8 x bfloat> [[TMP7]], 1
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP4]], 2
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP8]], <vscale x 8 x bfloat> [[TMP9]], 2
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP4]], 3
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP10]], <vscale x 8 x bfloat> [[TMP11]], 3
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP4]], 4
+// CHECK-RV64-NEXT:    store i64 [[TMP13]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP12]]
+//
+vbfloat16m2x4_t test_vlseg4e16ff_v_bf16m2x4_tum(vbool8_t vm, vbfloat16m2x4_t vd,
+                                                const __bf16 *rs1,
+                                                size_t *new_vl, size_t vl) {
+  return __riscv_vlseg4e16ff_tum(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg4e16ff_v_bf16mf4x4_tumu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } @llvm.riscv.vlseg4ff.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP4]], 0
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } poison, <vscale x 1 x bfloat> [[TMP5]], 0
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP4]], 1
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP6]], <vscale x 1 x bfloat> [[TMP7]], 1
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP4]], 2
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP8]], <vscale x 1 x bfloat> [[TMP9]], 2
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP4]], 3
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP10]], <vscale x 1 x bfloat> [[TMP11]], 3
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP4]], 4
+// CHECK-RV64-NEXT:    store i64 [[TMP13]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP12]]
+//
+vbfloat16mf4x4_t test_vlseg4e16ff_v_bf16mf4x4_tumu(vbool64_t vm,
+                                                   vbfloat16mf4x4_t vd,
+                                                   const __bf16 *rs1,
+                                                   size_t *new_vl, size_t vl) {
+  return __riscv_vlseg4e16ff_tumu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg4e16ff_v_bf16mf2x4_tumu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } @llvm.riscv.vlseg4ff.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP4]], 0
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } poison, <vscale x 2 x bfloat> [[TMP5]], 0
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP4]], 1
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP6]], <vscale x 2 x bfloat> [[TMP7]], 1
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP4]], 2
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP8]], <vscale x 2 x bfloat> [[TMP9]], 2
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP4]], 3
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP10]], <vscale x 2 x bfloat> [[TMP11]], 3
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP4]], 4
+// CHECK-RV64-NEXT:    store i64 [[TMP13]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP12]]
+//
+vbfloat16mf2x4_t test_vlseg4e16ff_v_bf16mf2x4_tumu(vbool32_t vm,
+                                                   vbfloat16mf2x4_t vd,
+                                                   const __bf16 *rs1,
+                                                   size_t *new_vl, size_t vl) {
+  return __riscv_vlseg4e16ff_tumu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg4e16ff_v_bf16m1x4_tumu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } @llvm.riscv.vlseg4ff.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP4]], 0
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } poison, <vscale x 4 x bfloat> [[TMP5]], 0
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP4]], 1
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP6]], <vscale x 4 x bfloat> [[TMP7]], 1
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP4]], 2
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP8]], <vscale x 4 x bfloat> [[TMP9]], 2
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP4]], 3
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP10]], <vscale x 4 x bfloat> [[TMP11]], 3
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP4]], 4
+// CHECK-RV64-NEXT:    store i64 [[TMP13]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP12]]
+//
+vbfloat16m1x4_t test_vlseg4e16ff_v_bf16m1x4_tumu(vbool16_t vm,
+                                                 vbfloat16m1x4_t vd,
+                                                 const __bf16 *rs1,
+                                                 size_t *new_vl, size_t vl) {
+  return __riscv_vlseg4e16ff_tumu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vlseg4e16ff_v_bf16m2x4_tumu(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } @llvm.riscv.vlseg4ff.mask.nxv8bf16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], <vscale x 8 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP4]], 0
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } poison, <vscale x 8 x bfloat> [[TMP5]], 0
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP4]], 1
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP6]], <vscale x 8 x bfloat> [[TMP7]], 1
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP4]], 2
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP8]], <vscale x 8 x bfloat> [[TMP9]], 2
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP4]], 3
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP10]], <vscale x 8 x bfloat> [[TMP11]], 3
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP4]], 4
+// CHECK-RV64-NEXT:    store i64 [[TMP13]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP12]]
+//
+vbfloat16m2x4_t test_vlseg4e16ff_v_bf16m2x4_tumu(vbool8_t vm,
+                                                 vbfloat16m2x4_t vd,
+                                                 const __bf16 *rs1,
+                                                 size_t *new_vl, size_t vl) {
+  return __riscv_vlseg4e16ff_tumu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg4e16ff_v_bf16mf4x4_mu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } @llvm.riscv.vlseg4ff.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP4]], 0
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } poison, <vscale x 1 x bfloat> [[TMP5]], 0
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP4]], 1
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP6]], <vscale x 1 x bfloat> [[TMP7]], 1
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP4]], 2
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP8]], <vscale x 1 x bfloat> [[TMP9]], 2
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP4]], 3
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP10]], <vscale x 1 x bfloat> [[TMP11]], 3
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP4]], 4
+// CHECK-RV64-NEXT:    store i64 [[TMP13]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP12]]
+//
+vbfloat16mf4x4_t test_vlseg4e16ff_v_bf16mf4x4_mu(vbool64_t vm,
+                                                 vbfloat16mf4x4_t vd,
+                                                 const __bf16 *rs1,
+                                                 size_t *new_vl, size_t vl) {
+  return __riscv_vlseg4e16ff_mu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg4e16ff_v_bf16mf2x4_mu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } @llvm.riscv.vlseg4ff.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP4]], 0
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } poison, <vscale x 2 x bfloat> [[TMP5]], 0
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP4]], 1
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP6]], <vscale x 2 x bfloat> [[TMP7]], 1
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP4]], 2
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP8]], <vscale x 2 x bfloat> [[TMP9]], 2
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP4]], 3
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP10]], <vscale x 2 x bfloat> [[TMP11]], 3
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP4]], 4
+// CHECK-RV64-NEXT:    store i64 [[TMP13]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP12]]
+//
+vbfloat16mf2x4_t test_vlseg4e16ff_v_bf16mf2x4_mu(vbool32_t vm,
+                                                 vbfloat16mf2x4_t vd,
+                                                 const __bf16 *rs1,
+                                                 size_t *new_vl, size_t vl) {
+  return __riscv_vlseg4e16ff_mu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg4e16ff_v_bf16m1x4_mu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } @llvm.riscv.vlseg4ff.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP4]], 0
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } poison, <vscale x 4 x bfloat> [[TMP5]], 0
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP4]], 1
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP6]], <vscale x 4 x bfloat> [[TMP7]], 1
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP4]], 2
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP8]], <vscale x 4 x bfloat> [[TMP9]], 2
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP4]], 3
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP10]], <vscale x 4 x bfloat> [[TMP11]], 3
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP4]], 4
+// CHECK-RV64-NEXT:    store i64 [[TMP13]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP12]]
+//
+vbfloat16m1x4_t test_vlseg4e16ff_v_bf16m1x4_mu(vbool16_t vm, vbfloat16m1x4_t vd,
+                                               const __bf16 *rs1,
+                                               size_t *new_vl, size_t vl) {
+  return __riscv_vlseg4e16ff_mu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vlseg4e16ff_v_bf16m2x4_mu(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } @llvm.riscv.vlseg4ff.mask.nxv8bf16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], <vscale x 8 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP4]], 0
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } poison, <vscale x 8 x bfloat> [[TMP5]], 0
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP4]], 1
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP6]], <vscale x 8 x bfloat> [[TMP7]], 1
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP4]], 2
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP8]], <vscale x 8 x bfloat> [[TMP9]], 2
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP4]], 3
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP10]], <vscale x 8 x bfloat> [[TMP11]], 3
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i64 } [[TMP4]], 4
+// CHECK-RV64-NEXT:    store i64 [[TMP13]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP12]]
+//
+vbfloat16m2x4_t test_vlseg4e16ff_v_bf16m2x4_mu(vbool8_t vm, vbfloat16m2x4_t vd,
+                                               const __bf16 *rs1,
+                                               size_t *new_vl, size_t vl) {
+  return __riscv_vlseg4e16ff_mu(vm, vd, rs1, new_vl, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vlseg5e16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vlseg5e16.c
new file mode 100644
index 000000000000..8b8fb4457de2
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vlseg5e16.c
@@ -0,0 +1,207 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg5e16_v_bf16mf4x5_tu(
+// CHECK-RV64-SAME: { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlseg5.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP5]]
+//
+vbfloat16mf4x5_t test_vlseg5e16_v_bf16mf4x5_tu(vbfloat16mf4x5_t vd,
+                                               const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg5e16_tu(vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg5e16_v_bf16mf2x5_tu(
+// CHECK-RV64-SAME: { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlseg5.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP5]]
+//
+vbfloat16mf2x5_t test_vlseg5e16_v_bf16mf2x5_tu(vbfloat16mf2x5_t vd,
+                                               const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg5e16_tu(vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg5e16_v_bf16m1x5_tu(
+// CHECK-RV64-SAME: { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlseg5.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP5]]
+//
+vbfloat16m1x5_t test_vlseg5e16_v_bf16m1x5_tu(vbfloat16m1x5_t vd,
+                                             const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg5e16_tu(vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg5e16_v_bf16mf4x5_tum(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlseg5.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP5]]
+//
+vbfloat16mf4x5_t test_vlseg5e16_v_bf16mf4x5_tum(vbool64_t vm,
+                                                vbfloat16mf4x5_t vd,
+                                                const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg5e16_tum(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg5e16_v_bf16mf2x5_tum(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlseg5.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP5]]
+//
+vbfloat16mf2x5_t test_vlseg5e16_v_bf16mf2x5_tum(vbool32_t vm,
+                                                vbfloat16mf2x5_t vd,
+                                                const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg5e16_tum(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg5e16_v_bf16m1x5_tum(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlseg5.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP5]]
+//
+vbfloat16m1x5_t test_vlseg5e16_v_bf16m1x5_tum(vbool16_t vm, vbfloat16m1x5_t vd,
+                                              const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg5e16_tum(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg5e16_v_bf16mf4x5_tumu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlseg5.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP5]]
+//
+vbfloat16mf4x5_t test_vlseg5e16_v_bf16mf4x5_tumu(vbool64_t vm,
+                                                 vbfloat16mf4x5_t vd,
+                                                 const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg5e16_tumu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg5e16_v_bf16mf2x5_tumu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlseg5.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP5]]
+//
+vbfloat16mf2x5_t test_vlseg5e16_v_bf16mf2x5_tumu(vbool32_t vm,
+                                                 vbfloat16mf2x5_t vd,
+                                                 const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg5e16_tumu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg5e16_v_bf16m1x5_tumu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlseg5.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP5]]
+//
+vbfloat16m1x5_t test_vlseg5e16_v_bf16m1x5_tumu(vbool16_t vm, vbfloat16m1x5_t vd,
+                                               const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg5e16_tumu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg5e16_v_bf16mf4x5_mu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlseg5.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP5]]
+//
+vbfloat16mf4x5_t test_vlseg5e16_v_bf16mf4x5_mu(vbool64_t vm,
+                                               vbfloat16mf4x5_t vd,
+                                               const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg5e16_mu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg5e16_v_bf16mf2x5_mu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlseg5.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP5]]
+//
+vbfloat16mf2x5_t test_vlseg5e16_v_bf16mf2x5_mu(vbool32_t vm,
+                                               vbfloat16mf2x5_t vd,
+                                               const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg5e16_mu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg5e16_v_bf16m1x5_mu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlseg5.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP5]]
+//
+vbfloat16m1x5_t test_vlseg5e16_v_bf16m1x5_mu(vbool16_t vm, vbfloat16m1x5_t vd,
+                                             const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg5e16_mu(vm, vd, rs1, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vlseg5e16ff.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vlseg5e16ff.c
new file mode 100644
index 000000000000..7aab7e9e205d
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vlseg5e16ff.c
@@ -0,0 +1,365 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg5e16ff_v_bf16mf4x5_tu(
+// CHECK-RV64-SAME: { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } @llvm.riscv.vlseg5ff.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP5]], 0
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } poison, <vscale x 1 x bfloat> [[TMP6]], 0
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP5]], 1
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP7]], <vscale x 1 x bfloat> [[TMP8]], 1
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP5]], 2
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP9]], <vscale x 1 x bfloat> [[TMP10]], 2
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP5]], 3
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP11]], <vscale x 1 x bfloat> [[TMP12]], 3
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP5]], 4
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP13]], <vscale x 1 x bfloat> [[TMP14]], 4
+// CHECK-RV64-NEXT:    [[TMP16:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP5]], 5
+// CHECK-RV64-NEXT:    store i64 [[TMP16]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP15]]
+//
+vbfloat16mf4x5_t test_vlseg5e16ff_v_bf16mf4x5_tu(vbfloat16mf4x5_t vd,
+                                                 const __bf16 *rs1,
+                                                 size_t *new_vl, size_t vl) {
+  return __riscv_vlseg5e16ff_tu(vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg5e16ff_v_bf16mf2x5_tu(
+// CHECK-RV64-SAME: { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } @llvm.riscv.vlseg5ff.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP5]], 0
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } poison, <vscale x 2 x bfloat> [[TMP6]], 0
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP5]], 1
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP7]], <vscale x 2 x bfloat> [[TMP8]], 1
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP5]], 2
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP9]], <vscale x 2 x bfloat> [[TMP10]], 2
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP5]], 3
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP11]], <vscale x 2 x bfloat> [[TMP12]], 3
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP5]], 4
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP13]], <vscale x 2 x bfloat> [[TMP14]], 4
+// CHECK-RV64-NEXT:    [[TMP16:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP5]], 5
+// CHECK-RV64-NEXT:    store i64 [[TMP16]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP15]]
+//
+vbfloat16mf2x5_t test_vlseg5e16ff_v_bf16mf2x5_tu(vbfloat16mf2x5_t vd,
+                                                 const __bf16 *rs1,
+                                                 size_t *new_vl, size_t vl) {
+  return __riscv_vlseg5e16ff_tu(vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg5e16ff_v_bf16m1x5_tu(
+// CHECK-RV64-SAME: { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } @llvm.riscv.vlseg5ff.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP5]], 0
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } poison, <vscale x 4 x bfloat> [[TMP6]], 0
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP5]], 1
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP7]], <vscale x 4 x bfloat> [[TMP8]], 1
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP5]], 2
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP9]], <vscale x 4 x bfloat> [[TMP10]], 2
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP5]], 3
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP11]], <vscale x 4 x bfloat> [[TMP12]], 3
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP5]], 4
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP13]], <vscale x 4 x bfloat> [[TMP14]], 4
+// CHECK-RV64-NEXT:    [[TMP16:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP5]], 5
+// CHECK-RV64-NEXT:    store i64 [[TMP16]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP15]]
+//
+vbfloat16m1x5_t test_vlseg5e16ff_v_bf16m1x5_tu(vbfloat16m1x5_t vd,
+                                               const __bf16 *rs1,
+                                               size_t *new_vl, size_t vl) {
+  return __riscv_vlseg5e16ff_tu(vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg5e16ff_v_bf16mf4x5_tum(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } @llvm.riscv.vlseg5ff.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP5]], 0
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } poison, <vscale x 1 x bfloat> [[TMP6]], 0
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP5]], 1
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP7]], <vscale x 1 x bfloat> [[TMP8]], 1
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP5]], 2
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP9]], <vscale x 1 x bfloat> [[TMP10]], 2
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP5]], 3
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP11]], <vscale x 1 x bfloat> [[TMP12]], 3
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP5]], 4
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP13]], <vscale x 1 x bfloat> [[TMP14]], 4
+// CHECK-RV64-NEXT:    [[TMP16:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP5]], 5
+// CHECK-RV64-NEXT:    store i64 [[TMP16]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP15]]
+//
+vbfloat16mf4x5_t test_vlseg5e16ff_v_bf16mf4x5_tum(vbool64_t vm,
+                                                  vbfloat16mf4x5_t vd,
+                                                  const __bf16 *rs1,
+                                                  size_t *new_vl, size_t vl) {
+  return __riscv_vlseg5e16ff_tum(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg5e16ff_v_bf16mf2x5_tum(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } @llvm.riscv.vlseg5ff.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP5]], 0
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } poison, <vscale x 2 x bfloat> [[TMP6]], 0
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP5]], 1
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP7]], <vscale x 2 x bfloat> [[TMP8]], 1
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP5]], 2
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP9]], <vscale x 2 x bfloat> [[TMP10]], 2
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP5]], 3
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP11]], <vscale x 2 x bfloat> [[TMP12]], 3
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP5]], 4
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP13]], <vscale x 2 x bfloat> [[TMP14]], 4
+// CHECK-RV64-NEXT:    [[TMP16:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP5]], 5
+// CHECK-RV64-NEXT:    store i64 [[TMP16]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP15]]
+//
+vbfloat16mf2x5_t test_vlseg5e16ff_v_bf16mf2x5_tum(vbool32_t vm,
+                                                  vbfloat16mf2x5_t vd,
+                                                  const __bf16 *rs1,
+                                                  size_t *new_vl, size_t vl) {
+  return __riscv_vlseg5e16ff_tum(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg5e16ff_v_bf16m1x5_tum(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } @llvm.riscv.vlseg5ff.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP5]], 0
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } poison, <vscale x 4 x bfloat> [[TMP6]], 0
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP5]], 1
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP7]], <vscale x 4 x bfloat> [[TMP8]], 1
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP5]], 2
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP9]], <vscale x 4 x bfloat> [[TMP10]], 2
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP5]], 3
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP11]], <vscale x 4 x bfloat> [[TMP12]], 3
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP5]], 4
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP13]], <vscale x 4 x bfloat> [[TMP14]], 4
+// CHECK-RV64-NEXT:    [[TMP16:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP5]], 5
+// CHECK-RV64-NEXT:    store i64 [[TMP16]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP15]]
+//
+vbfloat16m1x5_t test_vlseg5e16ff_v_bf16m1x5_tum(vbool16_t vm,
+                                                vbfloat16m1x5_t vd,
+                                                const __bf16 *rs1,
+                                                size_t *new_vl, size_t vl) {
+  return __riscv_vlseg5e16ff_tum(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg5e16ff_v_bf16mf4x5_tumu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } @llvm.riscv.vlseg5ff.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP5]], 0
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } poison, <vscale x 1 x bfloat> [[TMP6]], 0
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP5]], 1
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP7]], <vscale x 1 x bfloat> [[TMP8]], 1
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP5]], 2
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP9]], <vscale x 1 x bfloat> [[TMP10]], 2
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP5]], 3
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP11]], <vscale x 1 x bfloat> [[TMP12]], 3
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP5]], 4
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP13]], <vscale x 1 x bfloat> [[TMP14]], 4
+// CHECK-RV64-NEXT:    [[TMP16:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP5]], 5
+// CHECK-RV64-NEXT:    store i64 [[TMP16]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP15]]
+//
+vbfloat16mf4x5_t test_vlseg5e16ff_v_bf16mf4x5_tumu(vbool64_t vm,
+                                                   vbfloat16mf4x5_t vd,
+                                                   const __bf16 *rs1,
+                                                   size_t *new_vl, size_t vl) {
+  return __riscv_vlseg5e16ff_tumu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg5e16ff_v_bf16mf2x5_tumu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } @llvm.riscv.vlseg5ff.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP5]], 0
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } poison, <vscale x 2 x bfloat> [[TMP6]], 0
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP5]], 1
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP7]], <vscale x 2 x bfloat> [[TMP8]], 1
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP5]], 2
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP9]], <vscale x 2 x bfloat> [[TMP10]], 2
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP5]], 3
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP11]], <vscale x 2 x bfloat> [[TMP12]], 3
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP5]], 4
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP13]], <vscale x 2 x bfloat> [[TMP14]], 4
+// CHECK-RV64-NEXT:    [[TMP16:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP5]], 5
+// CHECK-RV64-NEXT:    store i64 [[TMP16]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP15]]
+//
+vbfloat16mf2x5_t test_vlseg5e16ff_v_bf16mf2x5_tumu(vbool32_t vm,
+                                                   vbfloat16mf2x5_t vd,
+                                                   const __bf16 *rs1,
+                                                   size_t *new_vl, size_t vl) {
+  return __riscv_vlseg5e16ff_tumu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg5e16ff_v_bf16m1x5_tumu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } @llvm.riscv.vlseg5ff.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP5]], 0
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } poison, <vscale x 4 x bfloat> [[TMP6]], 0
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP5]], 1
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP7]], <vscale x 4 x bfloat> [[TMP8]], 1
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP5]], 2
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP9]], <vscale x 4 x bfloat> [[TMP10]], 2
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP5]], 3
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP11]], <vscale x 4 x bfloat> [[TMP12]], 3
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP5]], 4
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP13]], <vscale x 4 x bfloat> [[TMP14]], 4
+// CHECK-RV64-NEXT:    [[TMP16:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP5]], 5
+// CHECK-RV64-NEXT:    store i64 [[TMP16]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP15]]
+//
+vbfloat16m1x5_t test_vlseg5e16ff_v_bf16m1x5_tumu(vbool16_t vm,
+                                                 vbfloat16m1x5_t vd,
+                                                 const __bf16 *rs1,
+                                                 size_t *new_vl, size_t vl) {
+  return __riscv_vlseg5e16ff_tumu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg5e16ff_v_bf16mf4x5_mu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } @llvm.riscv.vlseg5ff.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP5]], 0
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } poison, <vscale x 1 x bfloat> [[TMP6]], 0
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP5]], 1
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP7]], <vscale x 1 x bfloat> [[TMP8]], 1
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP5]], 2
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP9]], <vscale x 1 x bfloat> [[TMP10]], 2
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP5]], 3
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP11]], <vscale x 1 x bfloat> [[TMP12]], 3
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP5]], 4
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP13]], <vscale x 1 x bfloat> [[TMP14]], 4
+// CHECK-RV64-NEXT:    [[TMP16:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP5]], 5
+// CHECK-RV64-NEXT:    store i64 [[TMP16]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP15]]
+//
+vbfloat16mf4x5_t test_vlseg5e16ff_v_bf16mf4x5_mu(vbool64_t vm,
+                                                 vbfloat16mf4x5_t vd,
+                                                 const __bf16 *rs1,
+                                                 size_t *new_vl, size_t vl) {
+  return __riscv_vlseg5e16ff_mu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg5e16ff_v_bf16mf2x5_mu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } @llvm.riscv.vlseg5ff.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP5]], 0
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } poison, <vscale x 2 x bfloat> [[TMP6]], 0
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP5]], 1
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP7]], <vscale x 2 x bfloat> [[TMP8]], 1
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP5]], 2
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP9]], <vscale x 2 x bfloat> [[TMP10]], 2
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP5]], 3
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP11]], <vscale x 2 x bfloat> [[TMP12]], 3
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP5]], 4
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP13]], <vscale x 2 x bfloat> [[TMP14]], 4
+// CHECK-RV64-NEXT:    [[TMP16:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP5]], 5
+// CHECK-RV64-NEXT:    store i64 [[TMP16]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP15]]
+//
+vbfloat16mf2x5_t test_vlseg5e16ff_v_bf16mf2x5_mu(vbool32_t vm,
+                                                 vbfloat16mf2x5_t vd,
+                                                 const __bf16 *rs1,
+                                                 size_t *new_vl, size_t vl) {
+  return __riscv_vlseg5e16ff_mu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg5e16ff_v_bf16m1x5_mu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } @llvm.riscv.vlseg5ff.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP5]], 0
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } poison, <vscale x 4 x bfloat> [[TMP6]], 0
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP5]], 1
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP7]], <vscale x 4 x bfloat> [[TMP8]], 1
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP5]], 2
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP9]], <vscale x 4 x bfloat> [[TMP10]], 2
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP5]], 3
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP11]], <vscale x 4 x bfloat> [[TMP12]], 3
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP5]], 4
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP13]], <vscale x 4 x bfloat> [[TMP14]], 4
+// CHECK-RV64-NEXT:    [[TMP16:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP5]], 5
+// CHECK-RV64-NEXT:    store i64 [[TMP16]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP15]]
+//
+vbfloat16m1x5_t test_vlseg5e16ff_v_bf16m1x5_mu(vbool16_t vm, vbfloat16m1x5_t vd,
+                                               const __bf16 *rs1,
+                                               size_t *new_vl, size_t vl) {
+  return __riscv_vlseg5e16ff_mu(vm, vd, rs1, new_vl, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vlseg6e16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vlseg6e16.c
new file mode 100644
index 000000000000..916997df7eb2
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vlseg6e16.c
@@ -0,0 +1,219 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg6e16_v_bf16mf4x6_tu(
+// CHECK-RV64-SAME: { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlseg6.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP6]]
+//
+vbfloat16mf4x6_t test_vlseg6e16_v_bf16mf4x6_tu(vbfloat16mf4x6_t vd,
+                                               const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg6e16_tu(vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg6e16_v_bf16mf2x6_tu(
+// CHECK-RV64-SAME: { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlseg6.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP6]]
+//
+vbfloat16mf2x6_t test_vlseg6e16_v_bf16mf2x6_tu(vbfloat16mf2x6_t vd,
+                                               const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg6e16_tu(vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg6e16_v_bf16m1x6_tu(
+// CHECK-RV64-SAME: { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlseg6.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP6]]
+//
+vbfloat16m1x6_t test_vlseg6e16_v_bf16m1x6_tu(vbfloat16m1x6_t vd,
+                                             const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg6e16_tu(vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg6e16_v_bf16mf4x6_tum(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlseg6.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP6]]
+//
+vbfloat16mf4x6_t test_vlseg6e16_v_bf16mf4x6_tum(vbool64_t vm,
+                                                vbfloat16mf4x6_t vd,
+                                                const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg6e16_tum(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg6e16_v_bf16mf2x6_tum(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlseg6.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP6]]
+//
+vbfloat16mf2x6_t test_vlseg6e16_v_bf16mf2x6_tum(vbool32_t vm,
+                                                vbfloat16mf2x6_t vd,
+                                                const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg6e16_tum(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg6e16_v_bf16m1x6_tum(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlseg6.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP6]]
+//
+vbfloat16m1x6_t test_vlseg6e16_v_bf16m1x6_tum(vbool16_t vm, vbfloat16m1x6_t vd,
+                                              const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg6e16_tum(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg6e16_v_bf16mf4x6_tumu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlseg6.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP6]]
+//
+vbfloat16mf4x6_t test_vlseg6e16_v_bf16mf4x6_tumu(vbool64_t vm,
+                                                 vbfloat16mf4x6_t vd,
+                                                 const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg6e16_tumu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg6e16_v_bf16mf2x6_tumu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlseg6.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP6]]
+//
+vbfloat16mf2x6_t test_vlseg6e16_v_bf16mf2x6_tumu(vbool32_t vm,
+                                                 vbfloat16mf2x6_t vd,
+                                                 const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg6e16_tumu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg6e16_v_bf16m1x6_tumu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlseg6.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP6]]
+//
+vbfloat16m1x6_t test_vlseg6e16_v_bf16m1x6_tumu(vbool16_t vm, vbfloat16m1x6_t vd,
+                                               const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg6e16_tumu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg6e16_v_bf16mf4x6_mu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlseg6.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP6]]
+//
+vbfloat16mf4x6_t test_vlseg6e16_v_bf16mf4x6_mu(vbool64_t vm,
+                                               vbfloat16mf4x6_t vd,
+                                               const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg6e16_mu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg6e16_v_bf16mf2x6_mu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlseg6.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP6]]
+//
+vbfloat16mf2x6_t test_vlseg6e16_v_bf16mf2x6_mu(vbool32_t vm,
+                                               vbfloat16mf2x6_t vd,
+                                               const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg6e16_mu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg6e16_v_bf16m1x6_mu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlseg6.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP6]]
+//
+vbfloat16m1x6_t test_vlseg6e16_v_bf16m1x6_mu(vbool16_t vm, vbfloat16m1x6_t vd,
+                                             const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg6e16_mu(vm, vd, rs1, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vlseg6e16ff.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vlseg6e16ff.c
new file mode 100644
index 000000000000..7d269acd4a1a
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vlseg6e16ff.c
@@ -0,0 +1,401 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg6e16ff_v_bf16mf4x6_tu(
+// CHECK-RV64-SAME: { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } @llvm.riscv.vlseg6ff.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP6]], 0
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } poison, <vscale x 1 x bfloat> [[TMP7]], 0
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP6]], 1
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP8]], <vscale x 1 x bfloat> [[TMP9]], 1
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP6]], 2
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP10]], <vscale x 1 x bfloat> [[TMP11]], 2
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP6]], 3
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP12]], <vscale x 1 x bfloat> [[TMP13]], 3
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP6]], 4
+// CHECK-RV64-NEXT:    [[TMP16:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP14]], <vscale x 1 x bfloat> [[TMP15]], 4
+// CHECK-RV64-NEXT:    [[TMP17:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP6]], 5
+// CHECK-RV64-NEXT:    [[TMP18:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP16]], <vscale x 1 x bfloat> [[TMP17]], 5
+// CHECK-RV64-NEXT:    [[TMP19:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP6]], 6
+// CHECK-RV64-NEXT:    store i64 [[TMP19]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP18]]
+//
+vbfloat16mf4x6_t test_vlseg6e16ff_v_bf16mf4x6_tu(vbfloat16mf4x6_t vd,
+                                                 const __bf16 *rs1,
+                                                 size_t *new_vl, size_t vl) {
+  return __riscv_vlseg6e16ff_tu(vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg6e16ff_v_bf16mf2x6_tu(
+// CHECK-RV64-SAME: { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } @llvm.riscv.vlseg6ff.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP6]], 0
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } poison, <vscale x 2 x bfloat> [[TMP7]], 0
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP6]], 1
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP8]], <vscale x 2 x bfloat> [[TMP9]], 1
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP6]], 2
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP10]], <vscale x 2 x bfloat> [[TMP11]], 2
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP6]], 3
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP12]], <vscale x 2 x bfloat> [[TMP13]], 3
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP6]], 4
+// CHECK-RV64-NEXT:    [[TMP16:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP14]], <vscale x 2 x bfloat> [[TMP15]], 4
+// CHECK-RV64-NEXT:    [[TMP17:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP6]], 5
+// CHECK-RV64-NEXT:    [[TMP18:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP16]], <vscale x 2 x bfloat> [[TMP17]], 5
+// CHECK-RV64-NEXT:    [[TMP19:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP6]], 6
+// CHECK-RV64-NEXT:    store i64 [[TMP19]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP18]]
+//
+vbfloat16mf2x6_t test_vlseg6e16ff_v_bf16mf2x6_tu(vbfloat16mf2x6_t vd,
+                                                 const __bf16 *rs1,
+                                                 size_t *new_vl, size_t vl) {
+  return __riscv_vlseg6e16ff_tu(vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg6e16ff_v_bf16m1x6_tu(
+// CHECK-RV64-SAME: { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } @llvm.riscv.vlseg6ff.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP6]], 0
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } poison, <vscale x 4 x bfloat> [[TMP7]], 0
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP6]], 1
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP8]], <vscale x 4 x bfloat> [[TMP9]], 1
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP6]], 2
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP10]], <vscale x 4 x bfloat> [[TMP11]], 2
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP6]], 3
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP12]], <vscale x 4 x bfloat> [[TMP13]], 3
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP6]], 4
+// CHECK-RV64-NEXT:    [[TMP16:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP14]], <vscale x 4 x bfloat> [[TMP15]], 4
+// CHECK-RV64-NEXT:    [[TMP17:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP6]], 5
+// CHECK-RV64-NEXT:    [[TMP18:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP16]], <vscale x 4 x bfloat> [[TMP17]], 5
+// CHECK-RV64-NEXT:    [[TMP19:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP6]], 6
+// CHECK-RV64-NEXT:    store i64 [[TMP19]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP18]]
+//
+vbfloat16m1x6_t test_vlseg6e16ff_v_bf16m1x6_tu(vbfloat16m1x6_t vd,
+                                               const __bf16 *rs1,
+                                               size_t *new_vl, size_t vl) {
+  return __riscv_vlseg6e16ff_tu(vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg6e16ff_v_bf16mf4x6_tum(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } @llvm.riscv.vlseg6ff.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP6]], 0
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } poison, <vscale x 1 x bfloat> [[TMP7]], 0
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP6]], 1
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP8]], <vscale x 1 x bfloat> [[TMP9]], 1
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP6]], 2
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP10]], <vscale x 1 x bfloat> [[TMP11]], 2
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP6]], 3
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP12]], <vscale x 1 x bfloat> [[TMP13]], 3
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP6]], 4
+// CHECK-RV64-NEXT:    [[TMP16:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP14]], <vscale x 1 x bfloat> [[TMP15]], 4
+// CHECK-RV64-NEXT:    [[TMP17:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP6]], 5
+// CHECK-RV64-NEXT:    [[TMP18:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP16]], <vscale x 1 x bfloat> [[TMP17]], 5
+// CHECK-RV64-NEXT:    [[TMP19:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP6]], 6
+// CHECK-RV64-NEXT:    store i64 [[TMP19]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP18]]
+//
+vbfloat16mf4x6_t test_vlseg6e16ff_v_bf16mf4x6_tum(vbool64_t vm,
+                                                  vbfloat16mf4x6_t vd,
+                                                  const __bf16 *rs1,
+                                                  size_t *new_vl, size_t vl) {
+  return __riscv_vlseg6e16ff_tum(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg6e16ff_v_bf16mf2x6_tum(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } @llvm.riscv.vlseg6ff.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP6]], 0
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } poison, <vscale x 2 x bfloat> [[TMP7]], 0
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP6]], 1
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP8]], <vscale x 2 x bfloat> [[TMP9]], 1
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP6]], 2
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP10]], <vscale x 2 x bfloat> [[TMP11]], 2
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP6]], 3
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP12]], <vscale x 2 x bfloat> [[TMP13]], 3
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP6]], 4
+// CHECK-RV64-NEXT:    [[TMP16:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP14]], <vscale x 2 x bfloat> [[TMP15]], 4
+// CHECK-RV64-NEXT:    [[TMP17:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP6]], 5
+// CHECK-RV64-NEXT:    [[TMP18:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP16]], <vscale x 2 x bfloat> [[TMP17]], 5
+// CHECK-RV64-NEXT:    [[TMP19:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP6]], 6
+// CHECK-RV64-NEXT:    store i64 [[TMP19]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP18]]
+//
+vbfloat16mf2x6_t test_vlseg6e16ff_v_bf16mf2x6_tum(vbool32_t vm,
+                                                  vbfloat16mf2x6_t vd,
+                                                  const __bf16 *rs1,
+                                                  size_t *new_vl, size_t vl) {
+  return __riscv_vlseg6e16ff_tum(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg6e16ff_v_bf16m1x6_tum(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } @llvm.riscv.vlseg6ff.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP6]], 0
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } poison, <vscale x 4 x bfloat> [[TMP7]], 0
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP6]], 1
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP8]], <vscale x 4 x bfloat> [[TMP9]], 1
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP6]], 2
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP10]], <vscale x 4 x bfloat> [[TMP11]], 2
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP6]], 3
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP12]], <vscale x 4 x bfloat> [[TMP13]], 3
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP6]], 4
+// CHECK-RV64-NEXT:    [[TMP16:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP14]], <vscale x 4 x bfloat> [[TMP15]], 4
+// CHECK-RV64-NEXT:    [[TMP17:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP6]], 5
+// CHECK-RV64-NEXT:    [[TMP18:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP16]], <vscale x 4 x bfloat> [[TMP17]], 5
+// CHECK-RV64-NEXT:    [[TMP19:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP6]], 6
+// CHECK-RV64-NEXT:    store i64 [[TMP19]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP18]]
+//
+vbfloat16m1x6_t test_vlseg6e16ff_v_bf16m1x6_tum(vbool16_t vm,
+                                                vbfloat16m1x6_t vd,
+                                                const __bf16 *rs1,
+                                                size_t *new_vl, size_t vl) {
+  return __riscv_vlseg6e16ff_tum(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg6e16ff_v_bf16mf4x6_tumu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } @llvm.riscv.vlseg6ff.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP6]], 0
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } poison, <vscale x 1 x bfloat> [[TMP7]], 0
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP6]], 1
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP8]], <vscale x 1 x bfloat> [[TMP9]], 1
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP6]], 2
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP10]], <vscale x 1 x bfloat> [[TMP11]], 2
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP6]], 3
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP12]], <vscale x 1 x bfloat> [[TMP13]], 3
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP6]], 4
+// CHECK-RV64-NEXT:    [[TMP16:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP14]], <vscale x 1 x bfloat> [[TMP15]], 4
+// CHECK-RV64-NEXT:    [[TMP17:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP6]], 5
+// CHECK-RV64-NEXT:    [[TMP18:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP16]], <vscale x 1 x bfloat> [[TMP17]], 5
+// CHECK-RV64-NEXT:    [[TMP19:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP6]], 6
+// CHECK-RV64-NEXT:    store i64 [[TMP19]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP18]]
+//
+vbfloat16mf4x6_t test_vlseg6e16ff_v_bf16mf4x6_tumu(vbool64_t vm,
+                                                   vbfloat16mf4x6_t vd,
+                                                   const __bf16 *rs1,
+                                                   size_t *new_vl, size_t vl) {
+  return __riscv_vlseg6e16ff_tumu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg6e16ff_v_bf16mf2x6_tumu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } @llvm.riscv.vlseg6ff.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP6]], 0
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } poison, <vscale x 2 x bfloat> [[TMP7]], 0
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP6]], 1
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP8]], <vscale x 2 x bfloat> [[TMP9]], 1
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP6]], 2
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP10]], <vscale x 2 x bfloat> [[TMP11]], 2
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP6]], 3
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP12]], <vscale x 2 x bfloat> [[TMP13]], 3
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP6]], 4
+// CHECK-RV64-NEXT:    [[TMP16:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP14]], <vscale x 2 x bfloat> [[TMP15]], 4
+// CHECK-RV64-NEXT:    [[TMP17:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP6]], 5
+// CHECK-RV64-NEXT:    [[TMP18:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP16]], <vscale x 2 x bfloat> [[TMP17]], 5
+// CHECK-RV64-NEXT:    [[TMP19:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP6]], 6
+// CHECK-RV64-NEXT:    store i64 [[TMP19]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP18]]
+//
+vbfloat16mf2x6_t test_vlseg6e16ff_v_bf16mf2x6_tumu(vbool32_t vm,
+                                                   vbfloat16mf2x6_t vd,
+                                                   const __bf16 *rs1,
+                                                   size_t *new_vl, size_t vl) {
+  return __riscv_vlseg6e16ff_tumu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg6e16ff_v_bf16m1x6_tumu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } @llvm.riscv.vlseg6ff.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP6]], 0
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } poison, <vscale x 4 x bfloat> [[TMP7]], 0
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP6]], 1
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP8]], <vscale x 4 x bfloat> [[TMP9]], 1
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP6]], 2
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP10]], <vscale x 4 x bfloat> [[TMP11]], 2
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP6]], 3
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP12]], <vscale x 4 x bfloat> [[TMP13]], 3
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP6]], 4
+// CHECK-RV64-NEXT:    [[TMP16:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP14]], <vscale x 4 x bfloat> [[TMP15]], 4
+// CHECK-RV64-NEXT:    [[TMP17:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP6]], 5
+// CHECK-RV64-NEXT:    [[TMP18:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP16]], <vscale x 4 x bfloat> [[TMP17]], 5
+// CHECK-RV64-NEXT:    [[TMP19:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP6]], 6
+// CHECK-RV64-NEXT:    store i64 [[TMP19]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP18]]
+//
+vbfloat16m1x6_t test_vlseg6e16ff_v_bf16m1x6_tumu(vbool16_t vm,
+                                                 vbfloat16m1x6_t vd,
+                                                 const __bf16 *rs1,
+                                                 size_t *new_vl, size_t vl) {
+  return __riscv_vlseg6e16ff_tumu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg6e16ff_v_bf16mf4x6_mu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } @llvm.riscv.vlseg6ff.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP6]], 0
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } poison, <vscale x 1 x bfloat> [[TMP7]], 0
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP6]], 1
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP8]], <vscale x 1 x bfloat> [[TMP9]], 1
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP6]], 2
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP10]], <vscale x 1 x bfloat> [[TMP11]], 2
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP6]], 3
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP12]], <vscale x 1 x bfloat> [[TMP13]], 3
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP6]], 4
+// CHECK-RV64-NEXT:    [[TMP16:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP14]], <vscale x 1 x bfloat> [[TMP15]], 4
+// CHECK-RV64-NEXT:    [[TMP17:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP6]], 5
+// CHECK-RV64-NEXT:    [[TMP18:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP16]], <vscale x 1 x bfloat> [[TMP17]], 5
+// CHECK-RV64-NEXT:    [[TMP19:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP6]], 6
+// CHECK-RV64-NEXT:    store i64 [[TMP19]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP18]]
+//
+vbfloat16mf4x6_t test_vlseg6e16ff_v_bf16mf4x6_mu(vbool64_t vm,
+                                                 vbfloat16mf4x6_t vd,
+                                                 const __bf16 *rs1,
+                                                 size_t *new_vl, size_t vl) {
+  return __riscv_vlseg6e16ff_mu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg6e16ff_v_bf16mf2x6_mu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } @llvm.riscv.vlseg6ff.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP6]], 0
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } poison, <vscale x 2 x bfloat> [[TMP7]], 0
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP6]], 1
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP8]], <vscale x 2 x bfloat> [[TMP9]], 1
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP6]], 2
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP10]], <vscale x 2 x bfloat> [[TMP11]], 2
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP6]], 3
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP12]], <vscale x 2 x bfloat> [[TMP13]], 3
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP6]], 4
+// CHECK-RV64-NEXT:    [[TMP16:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP14]], <vscale x 2 x bfloat> [[TMP15]], 4
+// CHECK-RV64-NEXT:    [[TMP17:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP6]], 5
+// CHECK-RV64-NEXT:    [[TMP18:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP16]], <vscale x 2 x bfloat> [[TMP17]], 5
+// CHECK-RV64-NEXT:    [[TMP19:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP6]], 6
+// CHECK-RV64-NEXT:    store i64 [[TMP19]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP18]]
+//
+vbfloat16mf2x6_t test_vlseg6e16ff_v_bf16mf2x6_mu(vbool32_t vm,
+                                                 vbfloat16mf2x6_t vd,
+                                                 const __bf16 *rs1,
+                                                 size_t *new_vl, size_t vl) {
+  return __riscv_vlseg6e16ff_mu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg6e16ff_v_bf16m1x6_mu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } @llvm.riscv.vlseg6ff.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP6]], 0
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } poison, <vscale x 4 x bfloat> [[TMP7]], 0
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP6]], 1
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP8]], <vscale x 4 x bfloat> [[TMP9]], 1
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP6]], 2
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP10]], <vscale x 4 x bfloat> [[TMP11]], 2
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP6]], 3
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP12]], <vscale x 4 x bfloat> [[TMP13]], 3
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP6]], 4
+// CHECK-RV64-NEXT:    [[TMP16:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP14]], <vscale x 4 x bfloat> [[TMP15]], 4
+// CHECK-RV64-NEXT:    [[TMP17:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP6]], 5
+// CHECK-RV64-NEXT:    [[TMP18:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP16]], <vscale x 4 x bfloat> [[TMP17]], 5
+// CHECK-RV64-NEXT:    [[TMP19:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP6]], 6
+// CHECK-RV64-NEXT:    store i64 [[TMP19]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP18]]
+//
+vbfloat16m1x6_t test_vlseg6e16ff_v_bf16m1x6_mu(vbool16_t vm, vbfloat16m1x6_t vd,
+                                               const __bf16 *rs1,
+                                               size_t *new_vl, size_t vl) {
+  return __riscv_vlseg6e16ff_mu(vm, vd, rs1, new_vl, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vlseg7e16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vlseg7e16.c
new file mode 100644
index 000000000000..d664f6c24e4b
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vlseg7e16.c
@@ -0,0 +1,231 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg7e16_v_bf16mf4x7_tu(
+// CHECK-RV64-SAME: { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlseg7.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP7]]
+//
+vbfloat16mf4x7_t test_vlseg7e16_v_bf16mf4x7_tu(vbfloat16mf4x7_t vd,
+                                               const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg7e16_tu(vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg7e16_v_bf16mf2x7_tu(
+// CHECK-RV64-SAME: { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlseg7.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP7]]
+//
+vbfloat16mf2x7_t test_vlseg7e16_v_bf16mf2x7_tu(vbfloat16mf2x7_t vd,
+                                               const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg7e16_tu(vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg7e16_v_bf16m1x7_tu(
+// CHECK-RV64-SAME: { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlseg7.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP7]]
+//
+vbfloat16m1x7_t test_vlseg7e16_v_bf16m1x7_tu(vbfloat16m1x7_t vd,
+                                             const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg7e16_tu(vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg7e16_v_bf16mf4x7_tum(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlseg7.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP7]]
+//
+vbfloat16mf4x7_t test_vlseg7e16_v_bf16mf4x7_tum(vbool64_t vm,
+                                                vbfloat16mf4x7_t vd,
+                                                const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg7e16_tum(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg7e16_v_bf16mf2x7_tum(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlseg7.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP7]]
+//
+vbfloat16mf2x7_t test_vlseg7e16_v_bf16mf2x7_tum(vbool32_t vm,
+                                                vbfloat16mf2x7_t vd,
+                                                const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg7e16_tum(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg7e16_v_bf16m1x7_tum(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlseg7.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP7]]
+//
+vbfloat16m1x7_t test_vlseg7e16_v_bf16m1x7_tum(vbool16_t vm, vbfloat16m1x7_t vd,
+                                              const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg7e16_tum(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg7e16_v_bf16mf4x7_tumu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlseg7.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP7]]
+//
+vbfloat16mf4x7_t test_vlseg7e16_v_bf16mf4x7_tumu(vbool64_t vm,
+                                                 vbfloat16mf4x7_t vd,
+                                                 const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg7e16_tumu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg7e16_v_bf16mf2x7_tumu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlseg7.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP7]]
+//
+vbfloat16mf2x7_t test_vlseg7e16_v_bf16mf2x7_tumu(vbool32_t vm,
+                                                 vbfloat16mf2x7_t vd,
+                                                 const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg7e16_tumu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg7e16_v_bf16m1x7_tumu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlseg7.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP7]]
+//
+vbfloat16m1x7_t test_vlseg7e16_v_bf16m1x7_tumu(vbool16_t vm, vbfloat16m1x7_t vd,
+                                               const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg7e16_tumu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg7e16_v_bf16mf4x7_mu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlseg7.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP7]]
+//
+vbfloat16mf4x7_t test_vlseg7e16_v_bf16mf4x7_mu(vbool64_t vm,
+                                               vbfloat16mf4x7_t vd,
+                                               const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg7e16_mu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg7e16_v_bf16mf2x7_mu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlseg7.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP7]]
+//
+vbfloat16mf2x7_t test_vlseg7e16_v_bf16mf2x7_mu(vbool32_t vm,
+                                               vbfloat16mf2x7_t vd,
+                                               const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg7e16_mu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg7e16_v_bf16m1x7_mu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlseg7.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP7]]
+//
+vbfloat16m1x7_t test_vlseg7e16_v_bf16m1x7_mu(vbool16_t vm, vbfloat16m1x7_t vd,
+                                             const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg7e16_mu(vm, vd, rs1, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vlseg7e16ff.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vlseg7e16ff.c
new file mode 100644
index 000000000000..bf7920a7f613
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vlseg7e16ff.c
@@ -0,0 +1,437 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg7e16ff_v_bf16mf4x7_tu(
+// CHECK-RV64-SAME: { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } @llvm.riscv.vlseg7ff.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP7]], 0
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } poison, <vscale x 1 x bfloat> [[TMP8]], 0
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP7]], 1
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP9]], <vscale x 1 x bfloat> [[TMP10]], 1
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP7]], 2
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP11]], <vscale x 1 x bfloat> [[TMP12]], 2
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP7]], 3
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP13]], <vscale x 1 x bfloat> [[TMP14]], 3
+// CHECK-RV64-NEXT:    [[TMP16:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP7]], 4
+// CHECK-RV64-NEXT:    [[TMP17:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP15]], <vscale x 1 x bfloat> [[TMP16]], 4
+// CHECK-RV64-NEXT:    [[TMP18:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP7]], 5
+// CHECK-RV64-NEXT:    [[TMP19:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP17]], <vscale x 1 x bfloat> [[TMP18]], 5
+// CHECK-RV64-NEXT:    [[TMP20:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP7]], 6
+// CHECK-RV64-NEXT:    [[TMP21:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP19]], <vscale x 1 x bfloat> [[TMP20]], 6
+// CHECK-RV64-NEXT:    [[TMP22:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP7]], 7
+// CHECK-RV64-NEXT:    store i64 [[TMP22]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP21]]
+//
+vbfloat16mf4x7_t test_vlseg7e16ff_v_bf16mf4x7_tu(vbfloat16mf4x7_t vd,
+                                                 const __bf16 *rs1,
+                                                 size_t *new_vl, size_t vl) {
+  return __riscv_vlseg7e16ff_tu(vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg7e16ff_v_bf16mf2x7_tu(
+// CHECK-RV64-SAME: { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } @llvm.riscv.vlseg7ff.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP7]], 0
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } poison, <vscale x 2 x bfloat> [[TMP8]], 0
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP7]], 1
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP9]], <vscale x 2 x bfloat> [[TMP10]], 1
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP7]], 2
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP11]], <vscale x 2 x bfloat> [[TMP12]], 2
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP7]], 3
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP13]], <vscale x 2 x bfloat> [[TMP14]], 3
+// CHECK-RV64-NEXT:    [[TMP16:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP7]], 4
+// CHECK-RV64-NEXT:    [[TMP17:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP15]], <vscale x 2 x bfloat> [[TMP16]], 4
+// CHECK-RV64-NEXT:    [[TMP18:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP7]], 5
+// CHECK-RV64-NEXT:    [[TMP19:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP17]], <vscale x 2 x bfloat> [[TMP18]], 5
+// CHECK-RV64-NEXT:    [[TMP20:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP7]], 6
+// CHECK-RV64-NEXT:    [[TMP21:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP19]], <vscale x 2 x bfloat> [[TMP20]], 6
+// CHECK-RV64-NEXT:    [[TMP22:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP7]], 7
+// CHECK-RV64-NEXT:    store i64 [[TMP22]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP21]]
+//
+vbfloat16mf2x7_t test_vlseg7e16ff_v_bf16mf2x7_tu(vbfloat16mf2x7_t vd,
+                                                 const __bf16 *rs1,
+                                                 size_t *new_vl, size_t vl) {
+  return __riscv_vlseg7e16ff_tu(vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg7e16ff_v_bf16m1x7_tu(
+// CHECK-RV64-SAME: { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } @llvm.riscv.vlseg7ff.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP7]], 0
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } poison, <vscale x 4 x bfloat> [[TMP8]], 0
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP7]], 1
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP9]], <vscale x 4 x bfloat> [[TMP10]], 1
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP7]], 2
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP11]], <vscale x 4 x bfloat> [[TMP12]], 2
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP7]], 3
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP13]], <vscale x 4 x bfloat> [[TMP14]], 3
+// CHECK-RV64-NEXT:    [[TMP16:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP7]], 4
+// CHECK-RV64-NEXT:    [[TMP17:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP15]], <vscale x 4 x bfloat> [[TMP16]], 4
+// CHECK-RV64-NEXT:    [[TMP18:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP7]], 5
+// CHECK-RV64-NEXT:    [[TMP19:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP17]], <vscale x 4 x bfloat> [[TMP18]], 5
+// CHECK-RV64-NEXT:    [[TMP20:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP7]], 6
+// CHECK-RV64-NEXT:    [[TMP21:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP19]], <vscale x 4 x bfloat> [[TMP20]], 6
+// CHECK-RV64-NEXT:    [[TMP22:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP7]], 7
+// CHECK-RV64-NEXT:    store i64 [[TMP22]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP21]]
+//
+vbfloat16m1x7_t test_vlseg7e16ff_v_bf16m1x7_tu(vbfloat16m1x7_t vd,
+                                               const __bf16 *rs1,
+                                               size_t *new_vl, size_t vl) {
+  return __riscv_vlseg7e16ff_tu(vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg7e16ff_v_bf16mf4x7_tum(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } @llvm.riscv.vlseg7ff.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP7]], 0
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } poison, <vscale x 1 x bfloat> [[TMP8]], 0
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP7]], 1
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP9]], <vscale x 1 x bfloat> [[TMP10]], 1
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP7]], 2
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP11]], <vscale x 1 x bfloat> [[TMP12]], 2
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP7]], 3
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP13]], <vscale x 1 x bfloat> [[TMP14]], 3
+// CHECK-RV64-NEXT:    [[TMP16:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP7]], 4
+// CHECK-RV64-NEXT:    [[TMP17:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP15]], <vscale x 1 x bfloat> [[TMP16]], 4
+// CHECK-RV64-NEXT:    [[TMP18:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP7]], 5
+// CHECK-RV64-NEXT:    [[TMP19:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP17]], <vscale x 1 x bfloat> [[TMP18]], 5
+// CHECK-RV64-NEXT:    [[TMP20:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP7]], 6
+// CHECK-RV64-NEXT:    [[TMP21:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP19]], <vscale x 1 x bfloat> [[TMP20]], 6
+// CHECK-RV64-NEXT:    [[TMP22:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP7]], 7
+// CHECK-RV64-NEXT:    store i64 [[TMP22]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP21]]
+//
+vbfloat16mf4x7_t test_vlseg7e16ff_v_bf16mf4x7_tum(vbool64_t vm,
+                                                  vbfloat16mf4x7_t vd,
+                                                  const __bf16 *rs1,
+                                                  size_t *new_vl, size_t vl) {
+  return __riscv_vlseg7e16ff_tum(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg7e16ff_v_bf16mf2x7_tum(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } @llvm.riscv.vlseg7ff.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP7]], 0
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } poison, <vscale x 2 x bfloat> [[TMP8]], 0
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP7]], 1
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP9]], <vscale x 2 x bfloat> [[TMP10]], 1
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP7]], 2
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP11]], <vscale x 2 x bfloat> [[TMP12]], 2
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP7]], 3
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP13]], <vscale x 2 x bfloat> [[TMP14]], 3
+// CHECK-RV64-NEXT:    [[TMP16:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP7]], 4
+// CHECK-RV64-NEXT:    [[TMP17:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP15]], <vscale x 2 x bfloat> [[TMP16]], 4
+// CHECK-RV64-NEXT:    [[TMP18:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP7]], 5
+// CHECK-RV64-NEXT:    [[TMP19:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP17]], <vscale x 2 x bfloat> [[TMP18]], 5
+// CHECK-RV64-NEXT:    [[TMP20:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP7]], 6
+// CHECK-RV64-NEXT:    [[TMP21:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP19]], <vscale x 2 x bfloat> [[TMP20]], 6
+// CHECK-RV64-NEXT:    [[TMP22:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP7]], 7
+// CHECK-RV64-NEXT:    store i64 [[TMP22]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP21]]
+//
+vbfloat16mf2x7_t test_vlseg7e16ff_v_bf16mf2x7_tum(vbool32_t vm,
+                                                  vbfloat16mf2x7_t vd,
+                                                  const __bf16 *rs1,
+                                                  size_t *new_vl, size_t vl) {
+  return __riscv_vlseg7e16ff_tum(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg7e16ff_v_bf16m1x7_tum(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } @llvm.riscv.vlseg7ff.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP7]], 0
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } poison, <vscale x 4 x bfloat> [[TMP8]], 0
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP7]], 1
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP9]], <vscale x 4 x bfloat> [[TMP10]], 1
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP7]], 2
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP11]], <vscale x 4 x bfloat> [[TMP12]], 2
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP7]], 3
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP13]], <vscale x 4 x bfloat> [[TMP14]], 3
+// CHECK-RV64-NEXT:    [[TMP16:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP7]], 4
+// CHECK-RV64-NEXT:    [[TMP17:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP15]], <vscale x 4 x bfloat> [[TMP16]], 4
+// CHECK-RV64-NEXT:    [[TMP18:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP7]], 5
+// CHECK-RV64-NEXT:    [[TMP19:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP17]], <vscale x 4 x bfloat> [[TMP18]], 5
+// CHECK-RV64-NEXT:    [[TMP20:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP7]], 6
+// CHECK-RV64-NEXT:    [[TMP21:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP19]], <vscale x 4 x bfloat> [[TMP20]], 6
+// CHECK-RV64-NEXT:    [[TMP22:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP7]], 7
+// CHECK-RV64-NEXT:    store i64 [[TMP22]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP21]]
+//
+vbfloat16m1x7_t test_vlseg7e16ff_v_bf16m1x7_tum(vbool16_t vm,
+                                                vbfloat16m1x7_t vd,
+                                                const __bf16 *rs1,
+                                                size_t *new_vl, size_t vl) {
+  return __riscv_vlseg7e16ff_tum(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg7e16ff_v_bf16mf4x7_tumu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } @llvm.riscv.vlseg7ff.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP7]], 0
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } poison, <vscale x 1 x bfloat> [[TMP8]], 0
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP7]], 1
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP9]], <vscale x 1 x bfloat> [[TMP10]], 1
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP7]], 2
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP11]], <vscale x 1 x bfloat> [[TMP12]], 2
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP7]], 3
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP13]], <vscale x 1 x bfloat> [[TMP14]], 3
+// CHECK-RV64-NEXT:    [[TMP16:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP7]], 4
+// CHECK-RV64-NEXT:    [[TMP17:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP15]], <vscale x 1 x bfloat> [[TMP16]], 4
+// CHECK-RV64-NEXT:    [[TMP18:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP7]], 5
+// CHECK-RV64-NEXT:    [[TMP19:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP17]], <vscale x 1 x bfloat> [[TMP18]], 5
+// CHECK-RV64-NEXT:    [[TMP20:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP7]], 6
+// CHECK-RV64-NEXT:    [[TMP21:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP19]], <vscale x 1 x bfloat> [[TMP20]], 6
+// CHECK-RV64-NEXT:    [[TMP22:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP7]], 7
+// CHECK-RV64-NEXT:    store i64 [[TMP22]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP21]]
+//
+vbfloat16mf4x7_t test_vlseg7e16ff_v_bf16mf4x7_tumu(vbool64_t vm,
+                                                   vbfloat16mf4x7_t vd,
+                                                   const __bf16 *rs1,
+                                                   size_t *new_vl, size_t vl) {
+  return __riscv_vlseg7e16ff_tumu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg7e16ff_v_bf16mf2x7_tumu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } @llvm.riscv.vlseg7ff.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP7]], 0
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } poison, <vscale x 2 x bfloat> [[TMP8]], 0
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP7]], 1
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP9]], <vscale x 2 x bfloat> [[TMP10]], 1
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP7]], 2
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP11]], <vscale x 2 x bfloat> [[TMP12]], 2
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP7]], 3
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP13]], <vscale x 2 x bfloat> [[TMP14]], 3
+// CHECK-RV64-NEXT:    [[TMP16:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP7]], 4
+// CHECK-RV64-NEXT:    [[TMP17:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP15]], <vscale x 2 x bfloat> [[TMP16]], 4
+// CHECK-RV64-NEXT:    [[TMP18:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP7]], 5
+// CHECK-RV64-NEXT:    [[TMP19:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP17]], <vscale x 2 x bfloat> [[TMP18]], 5
+// CHECK-RV64-NEXT:    [[TMP20:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP7]], 6
+// CHECK-RV64-NEXT:    [[TMP21:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP19]], <vscale x 2 x bfloat> [[TMP20]], 6
+// CHECK-RV64-NEXT:    [[TMP22:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP7]], 7
+// CHECK-RV64-NEXT:    store i64 [[TMP22]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP21]]
+//
+vbfloat16mf2x7_t test_vlseg7e16ff_v_bf16mf2x7_tumu(vbool32_t vm,
+                                                   vbfloat16mf2x7_t vd,
+                                                   const __bf16 *rs1,
+                                                   size_t *new_vl, size_t vl) {
+  return __riscv_vlseg7e16ff_tumu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg7e16ff_v_bf16m1x7_tumu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } @llvm.riscv.vlseg7ff.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP7]], 0
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } poison, <vscale x 4 x bfloat> [[TMP8]], 0
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP7]], 1
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP9]], <vscale x 4 x bfloat> [[TMP10]], 1
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP7]], 2
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP11]], <vscale x 4 x bfloat> [[TMP12]], 2
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP7]], 3
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP13]], <vscale x 4 x bfloat> [[TMP14]], 3
+// CHECK-RV64-NEXT:    [[TMP16:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP7]], 4
+// CHECK-RV64-NEXT:    [[TMP17:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP15]], <vscale x 4 x bfloat> [[TMP16]], 4
+// CHECK-RV64-NEXT:    [[TMP18:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP7]], 5
+// CHECK-RV64-NEXT:    [[TMP19:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP17]], <vscale x 4 x bfloat> [[TMP18]], 5
+// CHECK-RV64-NEXT:    [[TMP20:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP7]], 6
+// CHECK-RV64-NEXT:    [[TMP21:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP19]], <vscale x 4 x bfloat> [[TMP20]], 6
+// CHECK-RV64-NEXT:    [[TMP22:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP7]], 7
+// CHECK-RV64-NEXT:    store i64 [[TMP22]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP21]]
+//
+vbfloat16m1x7_t test_vlseg7e16ff_v_bf16m1x7_tumu(vbool16_t vm,
+                                                 vbfloat16m1x7_t vd,
+                                                 const __bf16 *rs1,
+                                                 size_t *new_vl, size_t vl) {
+  return __riscv_vlseg7e16ff_tumu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg7e16ff_v_bf16mf4x7_mu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } @llvm.riscv.vlseg7ff.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP7]], 0
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } poison, <vscale x 1 x bfloat> [[TMP8]], 0
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP7]], 1
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP9]], <vscale x 1 x bfloat> [[TMP10]], 1
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP7]], 2
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP11]], <vscale x 1 x bfloat> [[TMP12]], 2
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP7]], 3
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP13]], <vscale x 1 x bfloat> [[TMP14]], 3
+// CHECK-RV64-NEXT:    [[TMP16:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP7]], 4
+// CHECK-RV64-NEXT:    [[TMP17:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP15]], <vscale x 1 x bfloat> [[TMP16]], 4
+// CHECK-RV64-NEXT:    [[TMP18:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP7]], 5
+// CHECK-RV64-NEXT:    [[TMP19:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP17]], <vscale x 1 x bfloat> [[TMP18]], 5
+// CHECK-RV64-NEXT:    [[TMP20:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP7]], 6
+// CHECK-RV64-NEXT:    [[TMP21:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP19]], <vscale x 1 x bfloat> [[TMP20]], 6
+// CHECK-RV64-NEXT:    [[TMP22:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP7]], 7
+// CHECK-RV64-NEXT:    store i64 [[TMP22]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP21]]
+//
+vbfloat16mf4x7_t test_vlseg7e16ff_v_bf16mf4x7_mu(vbool64_t vm,
+                                                 vbfloat16mf4x7_t vd,
+                                                 const __bf16 *rs1,
+                                                 size_t *new_vl, size_t vl) {
+  return __riscv_vlseg7e16ff_mu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg7e16ff_v_bf16mf2x7_mu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } @llvm.riscv.vlseg7ff.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP7]], 0
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } poison, <vscale x 2 x bfloat> [[TMP8]], 0
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP7]], 1
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP9]], <vscale x 2 x bfloat> [[TMP10]], 1
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP7]], 2
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP11]], <vscale x 2 x bfloat> [[TMP12]], 2
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP7]], 3
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP13]], <vscale x 2 x bfloat> [[TMP14]], 3
+// CHECK-RV64-NEXT:    [[TMP16:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP7]], 4
+// CHECK-RV64-NEXT:    [[TMP17:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP15]], <vscale x 2 x bfloat> [[TMP16]], 4
+// CHECK-RV64-NEXT:    [[TMP18:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP7]], 5
+// CHECK-RV64-NEXT:    [[TMP19:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP17]], <vscale x 2 x bfloat> [[TMP18]], 5
+// CHECK-RV64-NEXT:    [[TMP20:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP7]], 6
+// CHECK-RV64-NEXT:    [[TMP21:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP19]], <vscale x 2 x bfloat> [[TMP20]], 6
+// CHECK-RV64-NEXT:    [[TMP22:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP7]], 7
+// CHECK-RV64-NEXT:    store i64 [[TMP22]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP21]]
+//
+vbfloat16mf2x7_t test_vlseg7e16ff_v_bf16mf2x7_mu(vbool32_t vm,
+                                                 vbfloat16mf2x7_t vd,
+                                                 const __bf16 *rs1,
+                                                 size_t *new_vl, size_t vl) {
+  return __riscv_vlseg7e16ff_mu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg7e16ff_v_bf16m1x7_mu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } @llvm.riscv.vlseg7ff.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP7]], 0
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } poison, <vscale x 4 x bfloat> [[TMP8]], 0
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP7]], 1
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP9]], <vscale x 4 x bfloat> [[TMP10]], 1
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP7]], 2
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP11]], <vscale x 4 x bfloat> [[TMP12]], 2
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP7]], 3
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP13]], <vscale x 4 x bfloat> [[TMP14]], 3
+// CHECK-RV64-NEXT:    [[TMP16:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP7]], 4
+// CHECK-RV64-NEXT:    [[TMP17:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP15]], <vscale x 4 x bfloat> [[TMP16]], 4
+// CHECK-RV64-NEXT:    [[TMP18:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP7]], 5
+// CHECK-RV64-NEXT:    [[TMP19:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP17]], <vscale x 4 x bfloat> [[TMP18]], 5
+// CHECK-RV64-NEXT:    [[TMP20:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP7]], 6
+// CHECK-RV64-NEXT:    [[TMP21:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP19]], <vscale x 4 x bfloat> [[TMP20]], 6
+// CHECK-RV64-NEXT:    [[TMP22:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP7]], 7
+// CHECK-RV64-NEXT:    store i64 [[TMP22]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP21]]
+//
+vbfloat16m1x7_t test_vlseg7e16ff_v_bf16m1x7_mu(vbool16_t vm, vbfloat16m1x7_t vd,
+                                               const __bf16 *rs1,
+                                               size_t *new_vl, size_t vl) {
+  return __riscv_vlseg7e16ff_mu(vm, vd, rs1, new_vl, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vlseg8e16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vlseg8e16.c
new file mode 100644
index 000000000000..66d7b64611f9
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vlseg8e16.c
@@ -0,0 +1,243 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg8e16_v_bf16mf4x8_tu(
+// CHECK-RV64-SAME: { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlseg8.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], <vscale x 1 x bfloat> [[TMP7]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP8]]
+//
+vbfloat16mf4x8_t test_vlseg8e16_v_bf16mf4x8_tu(vbfloat16mf4x8_t vd,
+                                               const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg8e16_tu(vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg8e16_v_bf16mf2x8_tu(
+// CHECK-RV64-SAME: { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlseg8.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], <vscale x 2 x bfloat> [[TMP7]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP8]]
+//
+vbfloat16mf2x8_t test_vlseg8e16_v_bf16mf2x8_tu(vbfloat16mf2x8_t vd,
+                                               const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg8e16_tu(vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg8e16_v_bf16m1x8_tu(
+// CHECK-RV64-SAME: { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlseg8.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], <vscale x 4 x bfloat> [[TMP7]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP8]]
+//
+vbfloat16m1x8_t test_vlseg8e16_v_bf16m1x8_tu(vbfloat16m1x8_t vd,
+                                             const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg8e16_tu(vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg8e16_v_bf16mf4x8_tum(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlseg8.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], <vscale x 1 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP8]]
+//
+vbfloat16mf4x8_t test_vlseg8e16_v_bf16mf4x8_tum(vbool64_t vm,
+                                                vbfloat16mf4x8_t vd,
+                                                const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg8e16_tum(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg8e16_v_bf16mf2x8_tum(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlseg8.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], <vscale x 2 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP8]]
+//
+vbfloat16mf2x8_t test_vlseg8e16_v_bf16mf2x8_tum(vbool32_t vm,
+                                                vbfloat16mf2x8_t vd,
+                                                const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg8e16_tum(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg8e16_v_bf16m1x8_tum(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlseg8.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], <vscale x 4 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP8]]
+//
+vbfloat16m1x8_t test_vlseg8e16_v_bf16m1x8_tum(vbool16_t vm, vbfloat16m1x8_t vd,
+                                              const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg8e16_tum(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg8e16_v_bf16mf4x8_tumu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlseg8.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], <vscale x 1 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP8]]
+//
+vbfloat16mf4x8_t test_vlseg8e16_v_bf16mf4x8_tumu(vbool64_t vm,
+                                                 vbfloat16mf4x8_t vd,
+                                                 const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg8e16_tumu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg8e16_v_bf16mf2x8_tumu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlseg8.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], <vscale x 2 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP8]]
+//
+vbfloat16mf2x8_t test_vlseg8e16_v_bf16mf2x8_tumu(vbool32_t vm,
+                                                 vbfloat16mf2x8_t vd,
+                                                 const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg8e16_tumu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg8e16_v_bf16m1x8_tumu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlseg8.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], <vscale x 4 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP8]]
+//
+vbfloat16m1x8_t test_vlseg8e16_v_bf16m1x8_tumu(vbool16_t vm, vbfloat16m1x8_t vd,
+                                               const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg8e16_tumu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg8e16_v_bf16mf4x8_mu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlseg8.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], <vscale x 1 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP8]]
+//
+vbfloat16mf4x8_t test_vlseg8e16_v_bf16mf4x8_mu(vbool64_t vm,
+                                               vbfloat16mf4x8_t vd,
+                                               const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg8e16_mu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg8e16_v_bf16mf2x8_mu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlseg8.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], <vscale x 2 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP8]]
+//
+vbfloat16mf2x8_t test_vlseg8e16_v_bf16mf2x8_mu(vbool32_t vm,
+                                               vbfloat16mf2x8_t vd,
+                                               const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg8e16_mu(vm, vd, rs1, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg8e16_v_bf16m1x8_mu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlseg8.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], <vscale x 4 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP8]]
+//
+vbfloat16m1x8_t test_vlseg8e16_v_bf16m1x8_mu(vbool16_t vm, vbfloat16m1x8_t vd,
+                                             const __bf16 *rs1, size_t vl) {
+  return __riscv_vlseg8e16_mu(vm, vd, rs1, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vlseg8e16ff.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vlseg8e16ff.c
new file mode 100644
index 000000000000..049984355afe
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vlseg8e16ff.c
@@ -0,0 +1,473 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg8e16ff_v_bf16mf4x8_tu(
+// CHECK-RV64-SAME: { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } @llvm.riscv.vlseg8ff.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], <vscale x 1 x bfloat> [[TMP7]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP8]], 0
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } poison, <vscale x 1 x bfloat> [[TMP9]], 0
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP8]], 1
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP10]], <vscale x 1 x bfloat> [[TMP11]], 1
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP8]], 2
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP12]], <vscale x 1 x bfloat> [[TMP13]], 2
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP8]], 3
+// CHECK-RV64-NEXT:    [[TMP16:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP14]], <vscale x 1 x bfloat> [[TMP15]], 3
+// CHECK-RV64-NEXT:    [[TMP17:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP8]], 4
+// CHECK-RV64-NEXT:    [[TMP18:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP16]], <vscale x 1 x bfloat> [[TMP17]], 4
+// CHECK-RV64-NEXT:    [[TMP19:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP8]], 5
+// CHECK-RV64-NEXT:    [[TMP20:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP18]], <vscale x 1 x bfloat> [[TMP19]], 5
+// CHECK-RV64-NEXT:    [[TMP21:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP8]], 6
+// CHECK-RV64-NEXT:    [[TMP22:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP20]], <vscale x 1 x bfloat> [[TMP21]], 6
+// CHECK-RV64-NEXT:    [[TMP23:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP8]], 7
+// CHECK-RV64-NEXT:    [[TMP24:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP22]], <vscale x 1 x bfloat> [[TMP23]], 7
+// CHECK-RV64-NEXT:    [[TMP25:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP8]], 8
+// CHECK-RV64-NEXT:    store i64 [[TMP25]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP24]]
+//
+vbfloat16mf4x8_t test_vlseg8e16ff_v_bf16mf4x8_tu(vbfloat16mf4x8_t vd,
+                                                 const __bf16 *rs1,
+                                                 size_t *new_vl, size_t vl) {
+  return __riscv_vlseg8e16ff_tu(vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg8e16ff_v_bf16mf2x8_tu(
+// CHECK-RV64-SAME: { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } @llvm.riscv.vlseg8ff.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], <vscale x 2 x bfloat> [[TMP7]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP8]], 0
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } poison, <vscale x 2 x bfloat> [[TMP9]], 0
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP8]], 1
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP10]], <vscale x 2 x bfloat> [[TMP11]], 1
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP8]], 2
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP12]], <vscale x 2 x bfloat> [[TMP13]], 2
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP8]], 3
+// CHECK-RV64-NEXT:    [[TMP16:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP14]], <vscale x 2 x bfloat> [[TMP15]], 3
+// CHECK-RV64-NEXT:    [[TMP17:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP8]], 4
+// CHECK-RV64-NEXT:    [[TMP18:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP16]], <vscale x 2 x bfloat> [[TMP17]], 4
+// CHECK-RV64-NEXT:    [[TMP19:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP8]], 5
+// CHECK-RV64-NEXT:    [[TMP20:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP18]], <vscale x 2 x bfloat> [[TMP19]], 5
+// CHECK-RV64-NEXT:    [[TMP21:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP8]], 6
+// CHECK-RV64-NEXT:    [[TMP22:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP20]], <vscale x 2 x bfloat> [[TMP21]], 6
+// CHECK-RV64-NEXT:    [[TMP23:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP8]], 7
+// CHECK-RV64-NEXT:    [[TMP24:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP22]], <vscale x 2 x bfloat> [[TMP23]], 7
+// CHECK-RV64-NEXT:    [[TMP25:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP8]], 8
+// CHECK-RV64-NEXT:    store i64 [[TMP25]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP24]]
+//
+vbfloat16mf2x8_t test_vlseg8e16ff_v_bf16mf2x8_tu(vbfloat16mf2x8_t vd,
+                                                 const __bf16 *rs1,
+                                                 size_t *new_vl, size_t vl) {
+  return __riscv_vlseg8e16ff_tu(vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg8e16ff_v_bf16m1x8_tu(
+// CHECK-RV64-SAME: { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } @llvm.riscv.vlseg8ff.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], <vscale x 4 x bfloat> [[TMP7]], ptr [[RS1]], i64 [[VL]])
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP8]], 0
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } poison, <vscale x 4 x bfloat> [[TMP9]], 0
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP8]], 1
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP10]], <vscale x 4 x bfloat> [[TMP11]], 1
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP8]], 2
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP12]], <vscale x 4 x bfloat> [[TMP13]], 2
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP8]], 3
+// CHECK-RV64-NEXT:    [[TMP16:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP14]], <vscale x 4 x bfloat> [[TMP15]], 3
+// CHECK-RV64-NEXT:    [[TMP17:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP8]], 4
+// CHECK-RV64-NEXT:    [[TMP18:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP16]], <vscale x 4 x bfloat> [[TMP17]], 4
+// CHECK-RV64-NEXT:    [[TMP19:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP8]], 5
+// CHECK-RV64-NEXT:    [[TMP20:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP18]], <vscale x 4 x bfloat> [[TMP19]], 5
+// CHECK-RV64-NEXT:    [[TMP21:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP8]], 6
+// CHECK-RV64-NEXT:    [[TMP22:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP20]], <vscale x 4 x bfloat> [[TMP21]], 6
+// CHECK-RV64-NEXT:    [[TMP23:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP8]], 7
+// CHECK-RV64-NEXT:    [[TMP24:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP22]], <vscale x 4 x bfloat> [[TMP23]], 7
+// CHECK-RV64-NEXT:    [[TMP25:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP8]], 8
+// CHECK-RV64-NEXT:    store i64 [[TMP25]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP24]]
+//
+vbfloat16m1x8_t test_vlseg8e16ff_v_bf16m1x8_tu(vbfloat16m1x8_t vd,
+                                               const __bf16 *rs1,
+                                               size_t *new_vl, size_t vl) {
+  return __riscv_vlseg8e16ff_tu(vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg8e16ff_v_bf16mf4x8_tum(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } @llvm.riscv.vlseg8ff.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], <vscale x 1 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP8]], 0
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } poison, <vscale x 1 x bfloat> [[TMP9]], 0
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP8]], 1
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP10]], <vscale x 1 x bfloat> [[TMP11]], 1
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP8]], 2
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP12]], <vscale x 1 x bfloat> [[TMP13]], 2
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP8]], 3
+// CHECK-RV64-NEXT:    [[TMP16:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP14]], <vscale x 1 x bfloat> [[TMP15]], 3
+// CHECK-RV64-NEXT:    [[TMP17:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP8]], 4
+// CHECK-RV64-NEXT:    [[TMP18:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP16]], <vscale x 1 x bfloat> [[TMP17]], 4
+// CHECK-RV64-NEXT:    [[TMP19:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP8]], 5
+// CHECK-RV64-NEXT:    [[TMP20:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP18]], <vscale x 1 x bfloat> [[TMP19]], 5
+// CHECK-RV64-NEXT:    [[TMP21:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP8]], 6
+// CHECK-RV64-NEXT:    [[TMP22:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP20]], <vscale x 1 x bfloat> [[TMP21]], 6
+// CHECK-RV64-NEXT:    [[TMP23:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP8]], 7
+// CHECK-RV64-NEXT:    [[TMP24:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP22]], <vscale x 1 x bfloat> [[TMP23]], 7
+// CHECK-RV64-NEXT:    [[TMP25:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP8]], 8
+// CHECK-RV64-NEXT:    store i64 [[TMP25]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP24]]
+//
+vbfloat16mf4x8_t test_vlseg8e16ff_v_bf16mf4x8_tum(vbool64_t vm,
+                                                  vbfloat16mf4x8_t vd,
+                                                  const __bf16 *rs1,
+                                                  size_t *new_vl, size_t vl) {
+  return __riscv_vlseg8e16ff_tum(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg8e16ff_v_bf16mf2x8_tum(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } @llvm.riscv.vlseg8ff.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], <vscale x 2 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP8]], 0
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } poison, <vscale x 2 x bfloat> [[TMP9]], 0
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP8]], 1
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP10]], <vscale x 2 x bfloat> [[TMP11]], 1
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP8]], 2
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP12]], <vscale x 2 x bfloat> [[TMP13]], 2
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP8]], 3
+// CHECK-RV64-NEXT:    [[TMP16:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP14]], <vscale x 2 x bfloat> [[TMP15]], 3
+// CHECK-RV64-NEXT:    [[TMP17:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP8]], 4
+// CHECK-RV64-NEXT:    [[TMP18:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP16]], <vscale x 2 x bfloat> [[TMP17]], 4
+// CHECK-RV64-NEXT:    [[TMP19:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP8]], 5
+// CHECK-RV64-NEXT:    [[TMP20:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP18]], <vscale x 2 x bfloat> [[TMP19]], 5
+// CHECK-RV64-NEXT:    [[TMP21:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP8]], 6
+// CHECK-RV64-NEXT:    [[TMP22:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP20]], <vscale x 2 x bfloat> [[TMP21]], 6
+// CHECK-RV64-NEXT:    [[TMP23:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP8]], 7
+// CHECK-RV64-NEXT:    [[TMP24:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP22]], <vscale x 2 x bfloat> [[TMP23]], 7
+// CHECK-RV64-NEXT:    [[TMP25:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP8]], 8
+// CHECK-RV64-NEXT:    store i64 [[TMP25]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP24]]
+//
+vbfloat16mf2x8_t test_vlseg8e16ff_v_bf16mf2x8_tum(vbool32_t vm,
+                                                  vbfloat16mf2x8_t vd,
+                                                  const __bf16 *rs1,
+                                                  size_t *new_vl, size_t vl) {
+  return __riscv_vlseg8e16ff_tum(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg8e16ff_v_bf16m1x8_tum(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } @llvm.riscv.vlseg8ff.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], <vscale x 4 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP8]], 0
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } poison, <vscale x 4 x bfloat> [[TMP9]], 0
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP8]], 1
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP10]], <vscale x 4 x bfloat> [[TMP11]], 1
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP8]], 2
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP12]], <vscale x 4 x bfloat> [[TMP13]], 2
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP8]], 3
+// CHECK-RV64-NEXT:    [[TMP16:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP14]], <vscale x 4 x bfloat> [[TMP15]], 3
+// CHECK-RV64-NEXT:    [[TMP17:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP8]], 4
+// CHECK-RV64-NEXT:    [[TMP18:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP16]], <vscale x 4 x bfloat> [[TMP17]], 4
+// CHECK-RV64-NEXT:    [[TMP19:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP8]], 5
+// CHECK-RV64-NEXT:    [[TMP20:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP18]], <vscale x 4 x bfloat> [[TMP19]], 5
+// CHECK-RV64-NEXT:    [[TMP21:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP8]], 6
+// CHECK-RV64-NEXT:    [[TMP22:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP20]], <vscale x 4 x bfloat> [[TMP21]], 6
+// CHECK-RV64-NEXT:    [[TMP23:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP8]], 7
+// CHECK-RV64-NEXT:    [[TMP24:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP22]], <vscale x 4 x bfloat> [[TMP23]], 7
+// CHECK-RV64-NEXT:    [[TMP25:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP8]], 8
+// CHECK-RV64-NEXT:    store i64 [[TMP25]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP24]]
+//
+vbfloat16m1x8_t test_vlseg8e16ff_v_bf16m1x8_tum(vbool16_t vm,
+                                                vbfloat16m1x8_t vd,
+                                                const __bf16 *rs1,
+                                                size_t *new_vl, size_t vl) {
+  return __riscv_vlseg8e16ff_tum(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg8e16ff_v_bf16mf4x8_tumu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } @llvm.riscv.vlseg8ff.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], <vscale x 1 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP8]], 0
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } poison, <vscale x 1 x bfloat> [[TMP9]], 0
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP8]], 1
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP10]], <vscale x 1 x bfloat> [[TMP11]], 1
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP8]], 2
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP12]], <vscale x 1 x bfloat> [[TMP13]], 2
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP8]], 3
+// CHECK-RV64-NEXT:    [[TMP16:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP14]], <vscale x 1 x bfloat> [[TMP15]], 3
+// CHECK-RV64-NEXT:    [[TMP17:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP8]], 4
+// CHECK-RV64-NEXT:    [[TMP18:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP16]], <vscale x 1 x bfloat> [[TMP17]], 4
+// CHECK-RV64-NEXT:    [[TMP19:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP8]], 5
+// CHECK-RV64-NEXT:    [[TMP20:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP18]], <vscale x 1 x bfloat> [[TMP19]], 5
+// CHECK-RV64-NEXT:    [[TMP21:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP8]], 6
+// CHECK-RV64-NEXT:    [[TMP22:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP20]], <vscale x 1 x bfloat> [[TMP21]], 6
+// CHECK-RV64-NEXT:    [[TMP23:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP8]], 7
+// CHECK-RV64-NEXT:    [[TMP24:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP22]], <vscale x 1 x bfloat> [[TMP23]], 7
+// CHECK-RV64-NEXT:    [[TMP25:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP8]], 8
+// CHECK-RV64-NEXT:    store i64 [[TMP25]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP24]]
+//
+vbfloat16mf4x8_t test_vlseg8e16ff_v_bf16mf4x8_tumu(vbool64_t vm,
+                                                   vbfloat16mf4x8_t vd,
+                                                   const __bf16 *rs1,
+                                                   size_t *new_vl, size_t vl) {
+  return __riscv_vlseg8e16ff_tumu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg8e16ff_v_bf16mf2x8_tumu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } @llvm.riscv.vlseg8ff.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], <vscale x 2 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP8]], 0
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } poison, <vscale x 2 x bfloat> [[TMP9]], 0
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP8]], 1
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP10]], <vscale x 2 x bfloat> [[TMP11]], 1
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP8]], 2
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP12]], <vscale x 2 x bfloat> [[TMP13]], 2
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP8]], 3
+// CHECK-RV64-NEXT:    [[TMP16:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP14]], <vscale x 2 x bfloat> [[TMP15]], 3
+// CHECK-RV64-NEXT:    [[TMP17:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP8]], 4
+// CHECK-RV64-NEXT:    [[TMP18:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP16]], <vscale x 2 x bfloat> [[TMP17]], 4
+// CHECK-RV64-NEXT:    [[TMP19:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP8]], 5
+// CHECK-RV64-NEXT:    [[TMP20:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP18]], <vscale x 2 x bfloat> [[TMP19]], 5
+// CHECK-RV64-NEXT:    [[TMP21:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP8]], 6
+// CHECK-RV64-NEXT:    [[TMP22:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP20]], <vscale x 2 x bfloat> [[TMP21]], 6
+// CHECK-RV64-NEXT:    [[TMP23:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP8]], 7
+// CHECK-RV64-NEXT:    [[TMP24:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP22]], <vscale x 2 x bfloat> [[TMP23]], 7
+// CHECK-RV64-NEXT:    [[TMP25:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP8]], 8
+// CHECK-RV64-NEXT:    store i64 [[TMP25]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP24]]
+//
+vbfloat16mf2x8_t test_vlseg8e16ff_v_bf16mf2x8_tumu(vbool32_t vm,
+                                                   vbfloat16mf2x8_t vd,
+                                                   const __bf16 *rs1,
+                                                   size_t *new_vl, size_t vl) {
+  return __riscv_vlseg8e16ff_tumu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg8e16ff_v_bf16m1x8_tumu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } @llvm.riscv.vlseg8ff.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], <vscale x 4 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP8]], 0
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } poison, <vscale x 4 x bfloat> [[TMP9]], 0
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP8]], 1
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP10]], <vscale x 4 x bfloat> [[TMP11]], 1
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP8]], 2
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP12]], <vscale x 4 x bfloat> [[TMP13]], 2
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP8]], 3
+// CHECK-RV64-NEXT:    [[TMP16:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP14]], <vscale x 4 x bfloat> [[TMP15]], 3
+// CHECK-RV64-NEXT:    [[TMP17:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP8]], 4
+// CHECK-RV64-NEXT:    [[TMP18:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP16]], <vscale x 4 x bfloat> [[TMP17]], 4
+// CHECK-RV64-NEXT:    [[TMP19:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP8]], 5
+// CHECK-RV64-NEXT:    [[TMP20:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP18]], <vscale x 4 x bfloat> [[TMP19]], 5
+// CHECK-RV64-NEXT:    [[TMP21:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP8]], 6
+// CHECK-RV64-NEXT:    [[TMP22:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP20]], <vscale x 4 x bfloat> [[TMP21]], 6
+// CHECK-RV64-NEXT:    [[TMP23:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP8]], 7
+// CHECK-RV64-NEXT:    [[TMP24:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP22]], <vscale x 4 x bfloat> [[TMP23]], 7
+// CHECK-RV64-NEXT:    [[TMP25:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP8]], 8
+// CHECK-RV64-NEXT:    store i64 [[TMP25]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP24]]
+//
+vbfloat16m1x8_t test_vlseg8e16ff_v_bf16m1x8_tumu(vbool16_t vm,
+                                                 vbfloat16m1x8_t vd,
+                                                 const __bf16 *rs1,
+                                                 size_t *new_vl, size_t vl) {
+  return __riscv_vlseg8e16ff_tumu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlseg8e16ff_v_bf16mf4x8_mu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } @llvm.riscv.vlseg8ff.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], <vscale x 1 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP8]], 0
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } poison, <vscale x 1 x bfloat> [[TMP9]], 0
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP8]], 1
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP10]], <vscale x 1 x bfloat> [[TMP11]], 1
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP8]], 2
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP12]], <vscale x 1 x bfloat> [[TMP13]], 2
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP8]], 3
+// CHECK-RV64-NEXT:    [[TMP16:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP14]], <vscale x 1 x bfloat> [[TMP15]], 3
+// CHECK-RV64-NEXT:    [[TMP17:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP8]], 4
+// CHECK-RV64-NEXT:    [[TMP18:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP16]], <vscale x 1 x bfloat> [[TMP17]], 4
+// CHECK-RV64-NEXT:    [[TMP19:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP8]], 5
+// CHECK-RV64-NEXT:    [[TMP20:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP18]], <vscale x 1 x bfloat> [[TMP19]], 5
+// CHECK-RV64-NEXT:    [[TMP21:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP8]], 6
+// CHECK-RV64-NEXT:    [[TMP22:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP20]], <vscale x 1 x bfloat> [[TMP21]], 6
+// CHECK-RV64-NEXT:    [[TMP23:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP8]], 7
+// CHECK-RV64-NEXT:    [[TMP24:%.*]] = insertvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP22]], <vscale x 1 x bfloat> [[TMP23]], 7
+// CHECK-RV64-NEXT:    [[TMP25:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i64 } [[TMP8]], 8
+// CHECK-RV64-NEXT:    store i64 [[TMP25]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP24]]
+//
+vbfloat16mf4x8_t test_vlseg8e16ff_v_bf16mf4x8_mu(vbool64_t vm,
+                                                 vbfloat16mf4x8_t vd,
+                                                 const __bf16 *rs1,
+                                                 size_t *new_vl, size_t vl) {
+  return __riscv_vlseg8e16ff_mu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlseg8e16ff_v_bf16mf2x8_mu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } @llvm.riscv.vlseg8ff.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], <vscale x 2 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP8]], 0
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } poison, <vscale x 2 x bfloat> [[TMP9]], 0
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP8]], 1
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP10]], <vscale x 2 x bfloat> [[TMP11]], 1
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP8]], 2
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP12]], <vscale x 2 x bfloat> [[TMP13]], 2
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP8]], 3
+// CHECK-RV64-NEXT:    [[TMP16:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP14]], <vscale x 2 x bfloat> [[TMP15]], 3
+// CHECK-RV64-NEXT:    [[TMP17:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP8]], 4
+// CHECK-RV64-NEXT:    [[TMP18:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP16]], <vscale x 2 x bfloat> [[TMP17]], 4
+// CHECK-RV64-NEXT:    [[TMP19:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP8]], 5
+// CHECK-RV64-NEXT:    [[TMP20:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP18]], <vscale x 2 x bfloat> [[TMP19]], 5
+// CHECK-RV64-NEXT:    [[TMP21:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP8]], 6
+// CHECK-RV64-NEXT:    [[TMP22:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP20]], <vscale x 2 x bfloat> [[TMP21]], 6
+// CHECK-RV64-NEXT:    [[TMP23:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP8]], 7
+// CHECK-RV64-NEXT:    [[TMP24:%.*]] = insertvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP22]], <vscale x 2 x bfloat> [[TMP23]], 7
+// CHECK-RV64-NEXT:    [[TMP25:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i64 } [[TMP8]], 8
+// CHECK-RV64-NEXT:    store i64 [[TMP25]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP24]]
+//
+vbfloat16mf2x8_t test_vlseg8e16ff_v_bf16mf2x8_mu(vbool32_t vm,
+                                                 vbfloat16mf2x8_t vd,
+                                                 const __bf16 *rs1,
+                                                 size_t *new_vl, size_t vl) {
+  return __riscv_vlseg8e16ff_mu(vm, vd, rs1, new_vl, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlseg8e16ff_v_bf16m1x8_mu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], ptr noundef [[NEW_VL:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } @llvm.riscv.vlseg8ff.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], <vscale x 4 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    [[TMP9:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP8]], 0
+// CHECK-RV64-NEXT:    [[TMP10:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } poison, <vscale x 4 x bfloat> [[TMP9]], 0
+// CHECK-RV64-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP8]], 1
+// CHECK-RV64-NEXT:    [[TMP12:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP10]], <vscale x 4 x bfloat> [[TMP11]], 1
+// CHECK-RV64-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP8]], 2
+// CHECK-RV64-NEXT:    [[TMP14:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP12]], <vscale x 4 x bfloat> [[TMP13]], 2
+// CHECK-RV64-NEXT:    [[TMP15:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP8]], 3
+// CHECK-RV64-NEXT:    [[TMP16:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP14]], <vscale x 4 x bfloat> [[TMP15]], 3
+// CHECK-RV64-NEXT:    [[TMP17:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP8]], 4
+// CHECK-RV64-NEXT:    [[TMP18:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP16]], <vscale x 4 x bfloat> [[TMP17]], 4
+// CHECK-RV64-NEXT:    [[TMP19:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP8]], 5
+// CHECK-RV64-NEXT:    [[TMP20:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP18]], <vscale x 4 x bfloat> [[TMP19]], 5
+// CHECK-RV64-NEXT:    [[TMP21:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP8]], 6
+// CHECK-RV64-NEXT:    [[TMP22:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP20]], <vscale x 4 x bfloat> [[TMP21]], 6
+// CHECK-RV64-NEXT:    [[TMP23:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP8]], 7
+// CHECK-RV64-NEXT:    [[TMP24:%.*]] = insertvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP22]], <vscale x 4 x bfloat> [[TMP23]], 7
+// CHECK-RV64-NEXT:    [[TMP25:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i64 } [[TMP8]], 8
+// CHECK-RV64-NEXT:    store i64 [[TMP25]], ptr [[NEW_VL]], align 8
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP24]]
+//
+vbfloat16m1x8_t test_vlseg8e16ff_v_bf16m1x8_mu(vbool16_t vm, vbfloat16m1x8_t vd,
+                                               const __bf16 *rs1,
+                                               size_t *new_vl, size_t vl) {
+  return __riscv_vlseg8e16ff_mu(vm, vd, rs1, new_vl, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vlsseg2e16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vlsseg2e16.c
new file mode 100644
index 000000000000..ad0cc42d2304
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vlsseg2e16.c
@@ -0,0 +1,296 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlsseg2e16_v_bf16mf4x2_tu(
+// CHECK-RV64-SAME: { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlsseg2.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP2]]
+//
+vbfloat16mf4x2_t test_vlsseg2e16_v_bf16mf4x2_tu(vbfloat16mf4x2_t vd,
+                                                const __bf16 *rs1,
+                                                ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg2e16_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlsseg2e16_v_bf16mf2x2_tu(
+// CHECK-RV64-SAME: { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlsseg2.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP2]]
+//
+vbfloat16mf2x2_t test_vlsseg2e16_v_bf16mf2x2_tu(vbfloat16mf2x2_t vd,
+                                                const __bf16 *rs1,
+                                                ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg2e16_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlsseg2e16_v_bf16m1x2_tu(
+// CHECK-RV64-SAME: { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlsseg2.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP2]]
+//
+vbfloat16m1x2_t test_vlsseg2e16_v_bf16m1x2_tu(vbfloat16m1x2_t vd,
+                                              const __bf16 *rs1, ptrdiff_t rs2,
+                                              size_t vl) {
+  return __riscv_vlsseg2e16_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vlsseg2e16_v_bf16m2x2_tu(
+// CHECK-RV64-SAME: { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vlsseg2.nxv8bf16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP2]]
+//
+vbfloat16m2x2_t test_vlsseg2e16_v_bf16m2x2_tu(vbfloat16m2x2_t vd,
+                                              const __bf16 *rs1, ptrdiff_t rs2,
+                                              size_t vl) {
+  return __riscv_vlsseg2e16_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @test_vlsseg2e16_v_bf16m4x2_tu(
+// CHECK-RV64-SAME: { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @llvm.riscv.vlsseg2.nxv16bf16.i64(<vscale x 16 x bfloat> [[TMP0]], <vscale x 16 x bfloat> [[TMP1]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[TMP2]]
+//
+vbfloat16m4x2_t test_vlsseg2e16_v_bf16m4x2_tu(vbfloat16m4x2_t vd,
+                                              const __bf16 *rs1, ptrdiff_t rs2,
+                                              size_t vl) {
+  return __riscv_vlsseg2e16_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlsseg2e16_v_bf16mf4x2_tum(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlsseg2.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], ptr [[RS1]], i64 [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP2]]
+//
+vbfloat16mf4x2_t test_vlsseg2e16_v_bf16mf4x2_tum(vbool64_t vm,
+                                                 vbfloat16mf4x2_t vd,
+                                                 const __bf16 *rs1,
+                                                 ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg2e16_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlsseg2e16_v_bf16mf2x2_tum(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlsseg2.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], ptr [[RS1]], i64 [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP2]]
+//
+vbfloat16mf2x2_t test_vlsseg2e16_v_bf16mf2x2_tum(vbool32_t vm,
+                                                 vbfloat16mf2x2_t vd,
+                                                 const __bf16 *rs1,
+                                                 ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg2e16_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlsseg2e16_v_bf16m1x2_tum(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlsseg2.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], ptr [[RS1]], i64 [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP2]]
+//
+vbfloat16m1x2_t test_vlsseg2e16_v_bf16m1x2_tum(vbool16_t vm, vbfloat16m1x2_t vd,
+                                               const __bf16 *rs1, ptrdiff_t rs2,
+                                               size_t vl) {
+  return __riscv_vlsseg2e16_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vlsseg2e16_v_bf16m2x2_tum(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vlsseg2.mask.nxv8bf16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], ptr [[RS1]], i64 [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP2]]
+//
+vbfloat16m2x2_t test_vlsseg2e16_v_bf16m2x2_tum(vbool8_t vm, vbfloat16m2x2_t vd,
+                                               const __bf16 *rs1, ptrdiff_t rs2,
+                                               size_t vl) {
+  return __riscv_vlsseg2e16_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @test_vlsseg2e16_v_bf16m4x2_tum(
+// CHECK-RV64-SAME: <vscale x 16 x i1> [[VM:%.*]], { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @llvm.riscv.vlsseg2.mask.nxv16bf16.i64(<vscale x 16 x bfloat> [[TMP0]], <vscale x 16 x bfloat> [[TMP1]], ptr [[RS1]], i64 [[RS2]], <vscale x 16 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[TMP2]]
+//
+vbfloat16m4x2_t test_vlsseg2e16_v_bf16m4x2_tum(vbool4_t vm, vbfloat16m4x2_t vd,
+                                               const __bf16 *rs1, ptrdiff_t rs2,
+                                               size_t vl) {
+  return __riscv_vlsseg2e16_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlsseg2e16_v_bf16mf4x2_tumu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlsseg2.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], ptr [[RS1]], i64 [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP2]]
+//
+vbfloat16mf4x2_t test_vlsseg2e16_v_bf16mf4x2_tumu(vbool64_t vm,
+                                                  vbfloat16mf4x2_t vd,
+                                                  const __bf16 *rs1,
+                                                  ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg2e16_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlsseg2e16_v_bf16mf2x2_tumu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlsseg2.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], ptr [[RS1]], i64 [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP2]]
+//
+vbfloat16mf2x2_t test_vlsseg2e16_v_bf16mf2x2_tumu(vbool32_t vm,
+                                                  vbfloat16mf2x2_t vd,
+                                                  const __bf16 *rs1,
+                                                  ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg2e16_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlsseg2e16_v_bf16m1x2_tumu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlsseg2.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], ptr [[RS1]], i64 [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP2]]
+//
+vbfloat16m1x2_t test_vlsseg2e16_v_bf16m1x2_tumu(vbool16_t vm,
+                                                vbfloat16m1x2_t vd,
+                                                const __bf16 *rs1,
+                                                ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg2e16_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vlsseg2e16_v_bf16m2x2_tumu(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vlsseg2.mask.nxv8bf16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], ptr [[RS1]], i64 [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP2]]
+//
+vbfloat16m2x2_t test_vlsseg2e16_v_bf16m2x2_tumu(vbool8_t vm, vbfloat16m2x2_t vd,
+                                                const __bf16 *rs1,
+                                                ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg2e16_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @test_vlsseg2e16_v_bf16m4x2_tumu(
+// CHECK-RV64-SAME: <vscale x 16 x i1> [[VM:%.*]], { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @llvm.riscv.vlsseg2.mask.nxv16bf16.i64(<vscale x 16 x bfloat> [[TMP0]], <vscale x 16 x bfloat> [[TMP1]], ptr [[RS1]], i64 [[RS2]], <vscale x 16 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[TMP2]]
+//
+vbfloat16m4x2_t test_vlsseg2e16_v_bf16m4x2_tumu(vbool4_t vm, vbfloat16m4x2_t vd,
+                                                const __bf16 *rs1,
+                                                ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg2e16_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlsseg2e16_v_bf16mf4x2_mu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlsseg2.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], ptr [[RS1]], i64 [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP2]]
+//
+vbfloat16mf4x2_t test_vlsseg2e16_v_bf16mf4x2_mu(vbool64_t vm,
+                                                vbfloat16mf4x2_t vd,
+                                                const __bf16 *rs1,
+                                                ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg2e16_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlsseg2e16_v_bf16mf2x2_mu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlsseg2.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], ptr [[RS1]], i64 [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP2]]
+//
+vbfloat16mf2x2_t test_vlsseg2e16_v_bf16mf2x2_mu(vbool32_t vm,
+                                                vbfloat16mf2x2_t vd,
+                                                const __bf16 *rs1,
+                                                ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg2e16_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlsseg2e16_v_bf16m1x2_mu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlsseg2.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], ptr [[RS1]], i64 [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP2]]
+//
+vbfloat16m1x2_t test_vlsseg2e16_v_bf16m1x2_mu(vbool16_t vm, vbfloat16m1x2_t vd,
+                                              const __bf16 *rs1, ptrdiff_t rs2,
+                                              size_t vl) {
+  return __riscv_vlsseg2e16_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vlsseg2e16_v_bf16m2x2_mu(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vlsseg2.mask.nxv8bf16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], ptr [[RS1]], i64 [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP2]]
+//
+vbfloat16m2x2_t test_vlsseg2e16_v_bf16m2x2_mu(vbool8_t vm, vbfloat16m2x2_t vd,
+                                              const __bf16 *rs1, ptrdiff_t rs2,
+                                              size_t vl) {
+  return __riscv_vlsseg2e16_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @test_vlsseg2e16_v_bf16m4x2_mu(
+// CHECK-RV64-SAME: <vscale x 16 x i1> [[VM:%.*]], { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @llvm.riscv.vlsseg2.mask.nxv16bf16.i64(<vscale x 16 x bfloat> [[TMP0]], <vscale x 16 x bfloat> [[TMP1]], ptr [[RS1]], i64 [[RS2]], <vscale x 16 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[TMP2]]
+//
+vbfloat16m4x2_t test_vlsseg2e16_v_bf16m4x2_mu(vbool4_t vm, vbfloat16m4x2_t vd,
+                                              const __bf16 *rs1, ptrdiff_t rs2,
+                                              size_t vl) {
+  return __riscv_vlsseg2e16_mu(vm, vd, rs1, rs2, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vlsseg3e16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vlsseg3e16.c
new file mode 100644
index 000000000000..d9866c24dcdd
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vlsseg3e16.c
@@ -0,0 +1,256 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlsseg3e16_v_bf16mf4x3_tu(
+// CHECK-RV64-SAME: { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlsseg3.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP3]]
+//
+vbfloat16mf4x3_t test_vlsseg3e16_v_bf16mf4x3_tu(vbfloat16mf4x3_t vd,
+                                                const __bf16 *rs1,
+                                                ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg3e16_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlsseg3e16_v_bf16mf2x3_tu(
+// CHECK-RV64-SAME: { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlsseg3.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP3]]
+//
+vbfloat16mf2x3_t test_vlsseg3e16_v_bf16mf2x3_tu(vbfloat16mf2x3_t vd,
+                                                const __bf16 *rs1,
+                                                ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg3e16_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlsseg3e16_v_bf16m1x3_tu(
+// CHECK-RV64-SAME: { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlsseg3.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP3]]
+//
+vbfloat16m1x3_t test_vlsseg3e16_v_bf16m1x3_tu(vbfloat16m1x3_t vd,
+                                              const __bf16 *rs1, ptrdiff_t rs2,
+                                              size_t vl) {
+  return __riscv_vlsseg3e16_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vlsseg3e16_v_bf16m2x3_tu(
+// CHECK-RV64-SAME: { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vlsseg3.nxv8bf16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP3]]
+//
+vbfloat16m2x3_t test_vlsseg3e16_v_bf16m2x3_tu(vbfloat16m2x3_t vd,
+                                              const __bf16 *rs1, ptrdiff_t rs2,
+                                              size_t vl) {
+  return __riscv_vlsseg3e16_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlsseg3e16_v_bf16mf4x3_tum(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlsseg3.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], ptr [[RS1]], i64 [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP3]]
+//
+vbfloat16mf4x3_t test_vlsseg3e16_v_bf16mf4x3_tum(vbool64_t vm,
+                                                 vbfloat16mf4x3_t vd,
+                                                 const __bf16 *rs1,
+                                                 ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg3e16_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlsseg3e16_v_bf16mf2x3_tum(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlsseg3.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], ptr [[RS1]], i64 [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP3]]
+//
+vbfloat16mf2x3_t test_vlsseg3e16_v_bf16mf2x3_tum(vbool32_t vm,
+                                                 vbfloat16mf2x3_t vd,
+                                                 const __bf16 *rs1,
+                                                 ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg3e16_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlsseg3e16_v_bf16m1x3_tum(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlsseg3.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], ptr [[RS1]], i64 [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP3]]
+//
+vbfloat16m1x3_t test_vlsseg3e16_v_bf16m1x3_tum(vbool16_t vm, vbfloat16m1x3_t vd,
+                                               const __bf16 *rs1, ptrdiff_t rs2,
+                                               size_t vl) {
+  return __riscv_vlsseg3e16_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vlsseg3e16_v_bf16m2x3_tum(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vlsseg3.mask.nxv8bf16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], ptr [[RS1]], i64 [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP3]]
+//
+vbfloat16m2x3_t test_vlsseg3e16_v_bf16m2x3_tum(vbool8_t vm, vbfloat16m2x3_t vd,
+                                               const __bf16 *rs1, ptrdiff_t rs2,
+                                               size_t vl) {
+  return __riscv_vlsseg3e16_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlsseg3e16_v_bf16mf4x3_tumu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlsseg3.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], ptr [[RS1]], i64 [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP3]]
+//
+vbfloat16mf4x3_t test_vlsseg3e16_v_bf16mf4x3_tumu(vbool64_t vm,
+                                                  vbfloat16mf4x3_t vd,
+                                                  const __bf16 *rs1,
+                                                  ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg3e16_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlsseg3e16_v_bf16mf2x3_tumu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlsseg3.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], ptr [[RS1]], i64 [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP3]]
+//
+vbfloat16mf2x3_t test_vlsseg3e16_v_bf16mf2x3_tumu(vbool32_t vm,
+                                                  vbfloat16mf2x3_t vd,
+                                                  const __bf16 *rs1,
+                                                  ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg3e16_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlsseg3e16_v_bf16m1x3_tumu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlsseg3.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], ptr [[RS1]], i64 [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP3]]
+//
+vbfloat16m1x3_t test_vlsseg3e16_v_bf16m1x3_tumu(vbool16_t vm,
+                                                vbfloat16m1x3_t vd,
+                                                const __bf16 *rs1,
+                                                ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg3e16_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vlsseg3e16_v_bf16m2x3_tumu(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vlsseg3.mask.nxv8bf16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], ptr [[RS1]], i64 [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP3]]
+//
+vbfloat16m2x3_t test_vlsseg3e16_v_bf16m2x3_tumu(vbool8_t vm, vbfloat16m2x3_t vd,
+                                                const __bf16 *rs1,
+                                                ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg3e16_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlsseg3e16_v_bf16mf4x3_mu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlsseg3.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], ptr [[RS1]], i64 [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP3]]
+//
+vbfloat16mf4x3_t test_vlsseg3e16_v_bf16mf4x3_mu(vbool64_t vm,
+                                                vbfloat16mf4x3_t vd,
+                                                const __bf16 *rs1,
+                                                ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg3e16_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlsseg3e16_v_bf16mf2x3_mu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlsseg3.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], ptr [[RS1]], i64 [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP3]]
+//
+vbfloat16mf2x3_t test_vlsseg3e16_v_bf16mf2x3_mu(vbool32_t vm,
+                                                vbfloat16mf2x3_t vd,
+                                                const __bf16 *rs1,
+                                                ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg3e16_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlsseg3e16_v_bf16m1x3_mu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlsseg3.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], ptr [[RS1]], i64 [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP3]]
+//
+vbfloat16m1x3_t test_vlsseg3e16_v_bf16m1x3_mu(vbool16_t vm, vbfloat16m1x3_t vd,
+                                              const __bf16 *rs1, ptrdiff_t rs2,
+                                              size_t vl) {
+  return __riscv_vlsseg3e16_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vlsseg3e16_v_bf16m2x3_mu(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vlsseg3.mask.nxv8bf16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], ptr [[RS1]], i64 [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP3]]
+//
+vbfloat16m2x3_t test_vlsseg3e16_v_bf16m2x3_mu(vbool8_t vm, vbfloat16m2x3_t vd,
+                                              const __bf16 *rs1, ptrdiff_t rs2,
+                                              size_t vl) {
+  return __riscv_vlsseg3e16_mu(vm, vd, rs1, rs2, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vlsseg4e16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vlsseg4e16.c
new file mode 100644
index 000000000000..ceb8e7412d06
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vlsseg4e16.c
@@ -0,0 +1,272 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlsseg4e16_v_bf16mf4x4_tu(
+// CHECK-RV64-SAME: { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlsseg4.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP4]]
+//
+vbfloat16mf4x4_t test_vlsseg4e16_v_bf16mf4x4_tu(vbfloat16mf4x4_t vd,
+                                                const __bf16 *rs1,
+                                                ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg4e16_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlsseg4e16_v_bf16mf2x4_tu(
+// CHECK-RV64-SAME: { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlsseg4.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP4]]
+//
+vbfloat16mf2x4_t test_vlsseg4e16_v_bf16mf2x4_tu(vbfloat16mf2x4_t vd,
+                                                const __bf16 *rs1,
+                                                ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg4e16_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlsseg4e16_v_bf16m1x4_tu(
+// CHECK-RV64-SAME: { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlsseg4.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP4]]
+//
+vbfloat16m1x4_t test_vlsseg4e16_v_bf16m1x4_tu(vbfloat16m1x4_t vd,
+                                              const __bf16 *rs1, ptrdiff_t rs2,
+                                              size_t vl) {
+  return __riscv_vlsseg4e16_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vlsseg4e16_v_bf16m2x4_tu(
+// CHECK-RV64-SAME: { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vlsseg4.nxv8bf16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], <vscale x 8 x bfloat> [[TMP3]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP4]]
+//
+vbfloat16m2x4_t test_vlsseg4e16_v_bf16m2x4_tu(vbfloat16m2x4_t vd,
+                                              const __bf16 *rs1, ptrdiff_t rs2,
+                                              size_t vl) {
+  return __riscv_vlsseg4e16_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlsseg4e16_v_bf16mf4x4_tum(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlsseg4.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], ptr [[RS1]], i64 [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP4]]
+//
+vbfloat16mf4x4_t test_vlsseg4e16_v_bf16mf4x4_tum(vbool64_t vm,
+                                                 vbfloat16mf4x4_t vd,
+                                                 const __bf16 *rs1,
+                                                 ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg4e16_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlsseg4e16_v_bf16mf2x4_tum(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlsseg4.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], ptr [[RS1]], i64 [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP4]]
+//
+vbfloat16mf2x4_t test_vlsseg4e16_v_bf16mf2x4_tum(vbool32_t vm,
+                                                 vbfloat16mf2x4_t vd,
+                                                 const __bf16 *rs1,
+                                                 ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg4e16_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlsseg4e16_v_bf16m1x4_tum(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlsseg4.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], ptr [[RS1]], i64 [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP4]]
+//
+vbfloat16m1x4_t test_vlsseg4e16_v_bf16m1x4_tum(vbool16_t vm, vbfloat16m1x4_t vd,
+                                               const __bf16 *rs1, ptrdiff_t rs2,
+                                               size_t vl) {
+  return __riscv_vlsseg4e16_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vlsseg4e16_v_bf16m2x4_tum(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vlsseg4.mask.nxv8bf16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], <vscale x 8 x bfloat> [[TMP3]], ptr [[RS1]], i64 [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP4]]
+//
+vbfloat16m2x4_t test_vlsseg4e16_v_bf16m2x4_tum(vbool8_t vm, vbfloat16m2x4_t vd,
+                                               const __bf16 *rs1, ptrdiff_t rs2,
+                                               size_t vl) {
+  return __riscv_vlsseg4e16_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlsseg4e16_v_bf16mf4x4_tumu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlsseg4.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], ptr [[RS1]], i64 [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP4]]
+//
+vbfloat16mf4x4_t test_vlsseg4e16_v_bf16mf4x4_tumu(vbool64_t vm,
+                                                  vbfloat16mf4x4_t vd,
+                                                  const __bf16 *rs1,
+                                                  ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg4e16_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlsseg4e16_v_bf16mf2x4_tumu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlsseg4.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], ptr [[RS1]], i64 [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP4]]
+//
+vbfloat16mf2x4_t test_vlsseg4e16_v_bf16mf2x4_tumu(vbool32_t vm,
+                                                  vbfloat16mf2x4_t vd,
+                                                  const __bf16 *rs1,
+                                                  ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg4e16_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlsseg4e16_v_bf16m1x4_tumu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlsseg4.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], ptr [[RS1]], i64 [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP4]]
+//
+vbfloat16m1x4_t test_vlsseg4e16_v_bf16m1x4_tumu(vbool16_t vm,
+                                                vbfloat16m1x4_t vd,
+                                                const __bf16 *rs1,
+                                                ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg4e16_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vlsseg4e16_v_bf16m2x4_tumu(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vlsseg4.mask.nxv8bf16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], <vscale x 8 x bfloat> [[TMP3]], ptr [[RS1]], i64 [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP4]]
+//
+vbfloat16m2x4_t test_vlsseg4e16_v_bf16m2x4_tumu(vbool8_t vm, vbfloat16m2x4_t vd,
+                                                const __bf16 *rs1,
+                                                ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg4e16_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlsseg4e16_v_bf16mf4x4_mu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlsseg4.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], ptr [[RS1]], i64 [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP4]]
+//
+vbfloat16mf4x4_t test_vlsseg4e16_v_bf16mf4x4_mu(vbool64_t vm,
+                                                vbfloat16mf4x4_t vd,
+                                                const __bf16 *rs1,
+                                                ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg4e16_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlsseg4e16_v_bf16mf2x4_mu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlsseg4.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], ptr [[RS1]], i64 [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP4]]
+//
+vbfloat16mf2x4_t test_vlsseg4e16_v_bf16mf2x4_mu(vbool32_t vm,
+                                                vbfloat16mf2x4_t vd,
+                                                const __bf16 *rs1,
+                                                ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg4e16_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlsseg4e16_v_bf16m1x4_mu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlsseg4.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], ptr [[RS1]], i64 [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP4]]
+//
+vbfloat16m1x4_t test_vlsseg4e16_v_bf16m1x4_mu(vbool16_t vm, vbfloat16m1x4_t vd,
+                                              const __bf16 *rs1, ptrdiff_t rs2,
+                                              size_t vl) {
+  return __riscv_vlsseg4e16_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vlsseg4e16_v_bf16m2x4_mu(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vlsseg4.mask.nxv8bf16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], <vscale x 8 x bfloat> [[TMP3]], ptr [[RS1]], i64 [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP4]]
+//
+vbfloat16m2x4_t test_vlsseg4e16_v_bf16m2x4_mu(vbool8_t vm, vbfloat16m2x4_t vd,
+                                              const __bf16 *rs1, ptrdiff_t rs2,
+                                              size_t vl) {
+  return __riscv_vlsseg4e16_mu(vm, vd, rs1, rs2, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vlsseg5e16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vlsseg5e16.c
new file mode 100644
index 000000000000..ff43061fc70d
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vlsseg5e16.c
@@ -0,0 +1,220 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlsseg5e16_v_bf16mf4x5_tu(
+// CHECK-RV64-SAME: { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlsseg5.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP5]]
+//
+vbfloat16mf4x5_t test_vlsseg5e16_v_bf16mf4x5_tu(vbfloat16mf4x5_t vd,
+                                                const __bf16 *rs1,
+                                                ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg5e16_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlsseg5e16_v_bf16mf2x5_tu(
+// CHECK-RV64-SAME: { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlsseg5.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP5]]
+//
+vbfloat16mf2x5_t test_vlsseg5e16_v_bf16mf2x5_tu(vbfloat16mf2x5_t vd,
+                                                const __bf16 *rs1,
+                                                ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg5e16_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlsseg5e16_v_bf16m1x5_tu(
+// CHECK-RV64-SAME: { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlsseg5.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP5]]
+//
+vbfloat16m1x5_t test_vlsseg5e16_v_bf16m1x5_tu(vbfloat16m1x5_t vd,
+                                              const __bf16 *rs1, ptrdiff_t rs2,
+                                              size_t vl) {
+  return __riscv_vlsseg5e16_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlsseg5e16_v_bf16mf4x5_tum(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlsseg5.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], ptr [[RS1]], i64 [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP5]]
+//
+vbfloat16mf4x5_t test_vlsseg5e16_v_bf16mf4x5_tum(vbool64_t vm,
+                                                 vbfloat16mf4x5_t vd,
+                                                 const __bf16 *rs1,
+                                                 ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg5e16_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlsseg5e16_v_bf16mf2x5_tum(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlsseg5.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], ptr [[RS1]], i64 [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP5]]
+//
+vbfloat16mf2x5_t test_vlsseg5e16_v_bf16mf2x5_tum(vbool32_t vm,
+                                                 vbfloat16mf2x5_t vd,
+                                                 const __bf16 *rs1,
+                                                 ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg5e16_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlsseg5e16_v_bf16m1x5_tum(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlsseg5.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], ptr [[RS1]], i64 [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP5]]
+//
+vbfloat16m1x5_t test_vlsseg5e16_v_bf16m1x5_tum(vbool16_t vm, vbfloat16m1x5_t vd,
+                                               const __bf16 *rs1, ptrdiff_t rs2,
+                                               size_t vl) {
+  return __riscv_vlsseg5e16_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlsseg5e16_v_bf16mf4x5_tumu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlsseg5.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], ptr [[RS1]], i64 [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP5]]
+//
+vbfloat16mf4x5_t test_vlsseg5e16_v_bf16mf4x5_tumu(vbool64_t vm,
+                                                  vbfloat16mf4x5_t vd,
+                                                  const __bf16 *rs1,
+                                                  ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg5e16_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlsseg5e16_v_bf16mf2x5_tumu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlsseg5.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], ptr [[RS1]], i64 [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP5]]
+//
+vbfloat16mf2x5_t test_vlsseg5e16_v_bf16mf2x5_tumu(vbool32_t vm,
+                                                  vbfloat16mf2x5_t vd,
+                                                  const __bf16 *rs1,
+                                                  ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg5e16_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlsseg5e16_v_bf16m1x5_tumu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlsseg5.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], ptr [[RS1]], i64 [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP5]]
+//
+vbfloat16m1x5_t test_vlsseg5e16_v_bf16m1x5_tumu(vbool16_t vm,
+                                                vbfloat16m1x5_t vd,
+                                                const __bf16 *rs1,
+                                                ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg5e16_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlsseg5e16_v_bf16mf4x5_mu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlsseg5.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], ptr [[RS1]], i64 [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP5]]
+//
+vbfloat16mf4x5_t test_vlsseg5e16_v_bf16mf4x5_mu(vbool64_t vm,
+                                                vbfloat16mf4x5_t vd,
+                                                const __bf16 *rs1,
+                                                ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg5e16_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlsseg5e16_v_bf16mf2x5_mu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlsseg5.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], ptr [[RS1]], i64 [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP5]]
+//
+vbfloat16mf2x5_t test_vlsseg5e16_v_bf16mf2x5_mu(vbool32_t vm,
+                                                vbfloat16mf2x5_t vd,
+                                                const __bf16 *rs1,
+                                                ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg5e16_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlsseg5e16_v_bf16m1x5_mu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlsseg5.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], ptr [[RS1]], i64 [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP5]]
+//
+vbfloat16m1x5_t test_vlsseg5e16_v_bf16m1x5_mu(vbool16_t vm, vbfloat16m1x5_t vd,
+                                              const __bf16 *rs1, ptrdiff_t rs2,
+                                              size_t vl) {
+  return __riscv_vlsseg5e16_mu(vm, vd, rs1, rs2, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vlsseg6e16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vlsseg6e16.c
new file mode 100644
index 000000000000..4220e8e07839
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vlsseg6e16.c
@@ -0,0 +1,232 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlsseg6e16_v_bf16mf4x6_tu(
+// CHECK-RV64-SAME: { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlsseg6.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP6]]
+//
+vbfloat16mf4x6_t test_vlsseg6e16_v_bf16mf4x6_tu(vbfloat16mf4x6_t vd,
+                                                const __bf16 *rs1,
+                                                ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg6e16_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlsseg6e16_v_bf16mf2x6_tu(
+// CHECK-RV64-SAME: { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlsseg6.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP6]]
+//
+vbfloat16mf2x6_t test_vlsseg6e16_v_bf16mf2x6_tu(vbfloat16mf2x6_t vd,
+                                                const __bf16 *rs1,
+                                                ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg6e16_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlsseg6e16_v_bf16m1x6_tu(
+// CHECK-RV64-SAME: { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlsseg6.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP6]]
+//
+vbfloat16m1x6_t test_vlsseg6e16_v_bf16m1x6_tu(vbfloat16m1x6_t vd,
+                                              const __bf16 *rs1, ptrdiff_t rs2,
+                                              size_t vl) {
+  return __riscv_vlsseg6e16_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlsseg6e16_v_bf16mf4x6_tum(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlsseg6.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], ptr [[RS1]], i64 [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP6]]
+//
+vbfloat16mf4x6_t test_vlsseg6e16_v_bf16mf4x6_tum(vbool64_t vm,
+                                                 vbfloat16mf4x6_t vd,
+                                                 const __bf16 *rs1,
+                                                 ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg6e16_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlsseg6e16_v_bf16mf2x6_tum(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlsseg6.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], ptr [[RS1]], i64 [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP6]]
+//
+vbfloat16mf2x6_t test_vlsseg6e16_v_bf16mf2x6_tum(vbool32_t vm,
+                                                 vbfloat16mf2x6_t vd,
+                                                 const __bf16 *rs1,
+                                                 ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg6e16_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlsseg6e16_v_bf16m1x6_tum(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlsseg6.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], ptr [[RS1]], i64 [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP6]]
+//
+vbfloat16m1x6_t test_vlsseg6e16_v_bf16m1x6_tum(vbool16_t vm, vbfloat16m1x6_t vd,
+                                               const __bf16 *rs1, ptrdiff_t rs2,
+                                               size_t vl) {
+  return __riscv_vlsseg6e16_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlsseg6e16_v_bf16mf4x6_tumu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlsseg6.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], ptr [[RS1]], i64 [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP6]]
+//
+vbfloat16mf4x6_t test_vlsseg6e16_v_bf16mf4x6_tumu(vbool64_t vm,
+                                                  vbfloat16mf4x6_t vd,
+                                                  const __bf16 *rs1,
+                                                  ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg6e16_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlsseg6e16_v_bf16mf2x6_tumu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlsseg6.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], ptr [[RS1]], i64 [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP6]]
+//
+vbfloat16mf2x6_t test_vlsseg6e16_v_bf16mf2x6_tumu(vbool32_t vm,
+                                                  vbfloat16mf2x6_t vd,
+                                                  const __bf16 *rs1,
+                                                  ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg6e16_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlsseg6e16_v_bf16m1x6_tumu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlsseg6.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], ptr [[RS1]], i64 [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP6]]
+//
+vbfloat16m1x6_t test_vlsseg6e16_v_bf16m1x6_tumu(vbool16_t vm,
+                                                vbfloat16m1x6_t vd,
+                                                const __bf16 *rs1,
+                                                ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg6e16_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlsseg6e16_v_bf16mf4x6_mu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlsseg6.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], ptr [[RS1]], i64 [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP6]]
+//
+vbfloat16mf4x6_t test_vlsseg6e16_v_bf16mf4x6_mu(vbool64_t vm,
+                                                vbfloat16mf4x6_t vd,
+                                                const __bf16 *rs1,
+                                                ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg6e16_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlsseg6e16_v_bf16mf2x6_mu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlsseg6.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], ptr [[RS1]], i64 [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP6]]
+//
+vbfloat16mf2x6_t test_vlsseg6e16_v_bf16mf2x6_mu(vbool32_t vm,
+                                                vbfloat16mf2x6_t vd,
+                                                const __bf16 *rs1,
+                                                ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg6e16_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlsseg6e16_v_bf16m1x6_mu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlsseg6.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], ptr [[RS1]], i64 [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP6]]
+//
+vbfloat16m1x6_t test_vlsseg6e16_v_bf16m1x6_mu(vbool16_t vm, vbfloat16m1x6_t vd,
+                                              const __bf16 *rs1, ptrdiff_t rs2,
+                                              size_t vl) {
+  return __riscv_vlsseg6e16_mu(vm, vd, rs1, rs2, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vlsseg7e16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vlsseg7e16.c
new file mode 100644
index 000000000000..45705a6dae78
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vlsseg7e16.c
@@ -0,0 +1,244 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlsseg7e16_v_bf16mf4x7_tu(
+// CHECK-RV64-SAME: { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlsseg7.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP7]]
+//
+vbfloat16mf4x7_t test_vlsseg7e16_v_bf16mf4x7_tu(vbfloat16mf4x7_t vd,
+                                                const __bf16 *rs1,
+                                                ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg7e16_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlsseg7e16_v_bf16mf2x7_tu(
+// CHECK-RV64-SAME: { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlsseg7.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP7]]
+//
+vbfloat16mf2x7_t test_vlsseg7e16_v_bf16mf2x7_tu(vbfloat16mf2x7_t vd,
+                                                const __bf16 *rs1,
+                                                ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg7e16_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlsseg7e16_v_bf16m1x7_tu(
+// CHECK-RV64-SAME: { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlsseg7.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP7]]
+//
+vbfloat16m1x7_t test_vlsseg7e16_v_bf16m1x7_tu(vbfloat16m1x7_t vd,
+                                              const __bf16 *rs1, ptrdiff_t rs2,
+                                              size_t vl) {
+  return __riscv_vlsseg7e16_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlsseg7e16_v_bf16mf4x7_tum(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlsseg7.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], ptr [[RS1]], i64 [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP7]]
+//
+vbfloat16mf4x7_t test_vlsseg7e16_v_bf16mf4x7_tum(vbool64_t vm,
+                                                 vbfloat16mf4x7_t vd,
+                                                 const __bf16 *rs1,
+                                                 ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg7e16_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlsseg7e16_v_bf16mf2x7_tum(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlsseg7.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], ptr [[RS1]], i64 [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP7]]
+//
+vbfloat16mf2x7_t test_vlsseg7e16_v_bf16mf2x7_tum(vbool32_t vm,
+                                                 vbfloat16mf2x7_t vd,
+                                                 const __bf16 *rs1,
+                                                 ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg7e16_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlsseg7e16_v_bf16m1x7_tum(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlsseg7.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], ptr [[RS1]], i64 [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP7]]
+//
+vbfloat16m1x7_t test_vlsseg7e16_v_bf16m1x7_tum(vbool16_t vm, vbfloat16m1x7_t vd,
+                                               const __bf16 *rs1, ptrdiff_t rs2,
+                                               size_t vl) {
+  return __riscv_vlsseg7e16_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlsseg7e16_v_bf16mf4x7_tumu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlsseg7.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], ptr [[RS1]], i64 [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP7]]
+//
+vbfloat16mf4x7_t test_vlsseg7e16_v_bf16mf4x7_tumu(vbool64_t vm,
+                                                  vbfloat16mf4x7_t vd,
+                                                  const __bf16 *rs1,
+                                                  ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg7e16_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlsseg7e16_v_bf16mf2x7_tumu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlsseg7.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], ptr [[RS1]], i64 [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP7]]
+//
+vbfloat16mf2x7_t test_vlsseg7e16_v_bf16mf2x7_tumu(vbool32_t vm,
+                                                  vbfloat16mf2x7_t vd,
+                                                  const __bf16 *rs1,
+                                                  ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg7e16_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlsseg7e16_v_bf16m1x7_tumu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlsseg7.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], ptr [[RS1]], i64 [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP7]]
+//
+vbfloat16m1x7_t test_vlsseg7e16_v_bf16m1x7_tumu(vbool16_t vm,
+                                                vbfloat16m1x7_t vd,
+                                                const __bf16 *rs1,
+                                                ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg7e16_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlsseg7e16_v_bf16mf4x7_mu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlsseg7.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], ptr [[RS1]], i64 [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP7]]
+//
+vbfloat16mf4x7_t test_vlsseg7e16_v_bf16mf4x7_mu(vbool64_t vm,
+                                                vbfloat16mf4x7_t vd,
+                                                const __bf16 *rs1,
+                                                ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg7e16_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlsseg7e16_v_bf16mf2x7_mu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlsseg7.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], ptr [[RS1]], i64 [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP7]]
+//
+vbfloat16mf2x7_t test_vlsseg7e16_v_bf16mf2x7_mu(vbool32_t vm,
+                                                vbfloat16mf2x7_t vd,
+                                                const __bf16 *rs1,
+                                                ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg7e16_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlsseg7e16_v_bf16m1x7_mu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlsseg7.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], ptr [[RS1]], i64 [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP7]]
+//
+vbfloat16m1x7_t test_vlsseg7e16_v_bf16m1x7_mu(vbool16_t vm, vbfloat16m1x7_t vd,
+                                              const __bf16 *rs1, ptrdiff_t rs2,
+                                              size_t vl) {
+  return __riscv_vlsseg7e16_mu(vm, vd, rs1, rs2, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vlsseg8e16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vlsseg8e16.c
new file mode 100644
index 000000000000..dd3a5fd36734
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vlsseg8e16.c
@@ -0,0 +1,256 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlsseg8e16_v_bf16mf4x8_tu(
+// CHECK-RV64-SAME: { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlsseg8.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], <vscale x 1 x bfloat> [[TMP7]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP8]]
+//
+vbfloat16mf4x8_t test_vlsseg8e16_v_bf16mf4x8_tu(vbfloat16mf4x8_t vd,
+                                                const __bf16 *rs1,
+                                                ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg8e16_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlsseg8e16_v_bf16mf2x8_tu(
+// CHECK-RV64-SAME: { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlsseg8.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], <vscale x 2 x bfloat> [[TMP7]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP8]]
+//
+vbfloat16mf2x8_t test_vlsseg8e16_v_bf16mf2x8_tu(vbfloat16mf2x8_t vd,
+                                                const __bf16 *rs1,
+                                                ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg8e16_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlsseg8e16_v_bf16m1x8_tu(
+// CHECK-RV64-SAME: { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlsseg8.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], <vscale x 4 x bfloat> [[TMP7]], ptr [[RS1]], i64 [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP8]]
+//
+vbfloat16m1x8_t test_vlsseg8e16_v_bf16m1x8_tu(vbfloat16m1x8_t vd,
+                                              const __bf16 *rs1, ptrdiff_t rs2,
+                                              size_t vl) {
+  return __riscv_vlsseg8e16_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlsseg8e16_v_bf16mf4x8_tum(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlsseg8.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], <vscale x 1 x bfloat> [[TMP7]], ptr [[RS1]], i64 [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP8]]
+//
+vbfloat16mf4x8_t test_vlsseg8e16_v_bf16mf4x8_tum(vbool64_t vm,
+                                                 vbfloat16mf4x8_t vd,
+                                                 const __bf16 *rs1,
+                                                 ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg8e16_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlsseg8e16_v_bf16mf2x8_tum(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlsseg8.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], <vscale x 2 x bfloat> [[TMP7]], ptr [[RS1]], i64 [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP8]]
+//
+vbfloat16mf2x8_t test_vlsseg8e16_v_bf16mf2x8_tum(vbool32_t vm,
+                                                 vbfloat16mf2x8_t vd,
+                                                 const __bf16 *rs1,
+                                                 ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg8e16_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlsseg8e16_v_bf16m1x8_tum(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlsseg8.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], <vscale x 4 x bfloat> [[TMP7]], ptr [[RS1]], i64 [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP8]]
+//
+vbfloat16m1x8_t test_vlsseg8e16_v_bf16m1x8_tum(vbool16_t vm, vbfloat16m1x8_t vd,
+                                               const __bf16 *rs1, ptrdiff_t rs2,
+                                               size_t vl) {
+  return __riscv_vlsseg8e16_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlsseg8e16_v_bf16mf4x8_tumu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlsseg8.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], <vscale x 1 x bfloat> [[TMP7]], ptr [[RS1]], i64 [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP8]]
+//
+vbfloat16mf4x8_t test_vlsseg8e16_v_bf16mf4x8_tumu(vbool64_t vm,
+                                                  vbfloat16mf4x8_t vd,
+                                                  const __bf16 *rs1,
+                                                  ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg8e16_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlsseg8e16_v_bf16mf2x8_tumu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlsseg8.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], <vscale x 2 x bfloat> [[TMP7]], ptr [[RS1]], i64 [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP8]]
+//
+vbfloat16mf2x8_t test_vlsseg8e16_v_bf16mf2x8_tumu(vbool32_t vm,
+                                                  vbfloat16mf2x8_t vd,
+                                                  const __bf16 *rs1,
+                                                  ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg8e16_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlsseg8e16_v_bf16m1x8_tumu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlsseg8.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], <vscale x 4 x bfloat> [[TMP7]], ptr [[RS1]], i64 [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP8]]
+//
+vbfloat16m1x8_t test_vlsseg8e16_v_bf16m1x8_tumu(vbool16_t vm,
+                                                vbfloat16m1x8_t vd,
+                                                const __bf16 *rs1,
+                                                ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg8e16_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vlsseg8e16_v_bf16mf4x8_mu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vlsseg8.mask.nxv1bf16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], <vscale x 1 x bfloat> [[TMP7]], ptr [[RS1]], i64 [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP8]]
+//
+vbfloat16mf4x8_t test_vlsseg8e16_v_bf16mf4x8_mu(vbool64_t vm,
+                                                vbfloat16mf4x8_t vd,
+                                                const __bf16 *rs1,
+                                                ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg8e16_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vlsseg8e16_v_bf16mf2x8_mu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vlsseg8.mask.nxv2bf16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], <vscale x 2 x bfloat> [[TMP7]], ptr [[RS1]], i64 [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP8]]
+//
+vbfloat16mf2x8_t test_vlsseg8e16_v_bf16mf2x8_mu(vbool32_t vm,
+                                                vbfloat16mf2x8_t vd,
+                                                const __bf16 *rs1,
+                                                ptrdiff_t rs2, size_t vl) {
+  return __riscv_vlsseg8e16_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vlsseg8e16_v_bf16m1x8_mu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], i64 noundef [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vlsseg8.mask.nxv4bf16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], <vscale x 4 x bfloat> [[TMP7]], ptr [[RS1]], i64 [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP8]]
+//
+vbfloat16m1x8_t test_vlsseg8e16_v_bf16m1x8_mu(vbool16_t vm, vbfloat16m1x8_t vd,
+                                              const __bf16 *rs1, ptrdiff_t rs2,
+                                              size_t vl) {
+  return __riscv_vlsseg8e16_mu(vm, vd, rs1, rs2, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vluxei16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vluxei16.c
new file mode 100644
index 000000000000..10e7fb561461
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vluxei16.c
@@ -0,0 +1,291 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 1 x bfloat> @test_vluxei16_v_bf16mf4_tu(
+// CHECK-RV64-SAME: <vscale x 1 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x bfloat> @llvm.riscv.vluxei.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[VD]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret <vscale x 1 x bfloat> [[TMP0]]
+//
+vbfloat16mf4_t test_vluxei16_v_bf16mf4_tu(vbfloat16mf4_t vd, const __bf16 *rs1,
+                                          vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vluxei16_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 2 x bfloat> @test_vluxei16_v_bf16mf2_tu(
+// CHECK-RV64-SAME: <vscale x 2 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x bfloat> @llvm.riscv.vluxei.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[VD]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret <vscale x 2 x bfloat> [[TMP0]]
+//
+vbfloat16mf2_t test_vluxei16_v_bf16mf2_tu(vbfloat16mf2_t vd, const __bf16 *rs1,
+                                          vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vluxei16_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 4 x bfloat> @test_vluxei16_v_bf16m1_tu(
+// CHECK-RV64-SAME: <vscale x 4 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x bfloat> @llvm.riscv.vluxei.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[VD]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret <vscale x 4 x bfloat> [[TMP0]]
+//
+vbfloat16m1_t test_vluxei16_v_bf16m1_tu(vbfloat16m1_t vd, const __bf16 *rs1,
+                                        vuint16m1_t rs2, size_t vl) {
+  return __riscv_vluxei16_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 8 x bfloat> @test_vluxei16_v_bf16m2_tu(
+// CHECK-RV64-SAME: <vscale x 8 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x bfloat> @llvm.riscv.vluxei.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> [[VD]], ptr [[RS1]], <vscale x 8 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret <vscale x 8 x bfloat> [[TMP0]]
+//
+vbfloat16m2_t test_vluxei16_v_bf16m2_tu(vbfloat16m2_t vd, const __bf16 *rs1,
+                                        vuint16m2_t rs2, size_t vl) {
+  return __riscv_vluxei16_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 16 x bfloat> @test_vluxei16_v_bf16m4_tu(
+// CHECK-RV64-SAME: <vscale x 16 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 16 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x bfloat> @llvm.riscv.vluxei.nxv16bf16.nxv16i16.i64(<vscale x 16 x bfloat> [[VD]], ptr [[RS1]], <vscale x 16 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret <vscale x 16 x bfloat> [[TMP0]]
+//
+vbfloat16m4_t test_vluxei16_v_bf16m4_tu(vbfloat16m4_t vd, const __bf16 *rs1,
+                                        vuint16m4_t rs2, size_t vl) {
+  return __riscv_vluxei16_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 32 x bfloat> @test_vluxei16_v_bf16m8_tu(
+// CHECK-RV64-SAME: <vscale x 32 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 32 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x bfloat> @llvm.riscv.vluxei.nxv32bf16.nxv32i16.i64(<vscale x 32 x bfloat> [[VD]], ptr [[RS1]], <vscale x 32 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret <vscale x 32 x bfloat> [[TMP0]]
+//
+vbfloat16m8_t test_vluxei16_v_bf16m8_tu(vbfloat16m8_t vd, const __bf16 *rs1,
+                                        vuint16m8_t rs2, size_t vl) {
+  return __riscv_vluxei16_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 1 x bfloat> @test_vluxei16_v_bf16mf4_tum(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], <vscale x 1 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x bfloat> @llvm.riscv.vluxei.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[VD]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret <vscale x 1 x bfloat> [[TMP0]]
+//
+vbfloat16mf4_t test_vluxei16_v_bf16mf4_tum(vbool64_t vm, vbfloat16mf4_t vd,
+                                           const __bf16 *rs1, vuint16mf4_t rs2,
+                                           size_t vl) {
+  return __riscv_vluxei16_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 2 x bfloat> @test_vluxei16_v_bf16mf2_tum(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], <vscale x 2 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x bfloat> @llvm.riscv.vluxei.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[VD]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret <vscale x 2 x bfloat> [[TMP0]]
+//
+vbfloat16mf2_t test_vluxei16_v_bf16mf2_tum(vbool32_t vm, vbfloat16mf2_t vd,
+                                           const __bf16 *rs1, vuint16mf2_t rs2,
+                                           size_t vl) {
+  return __riscv_vluxei16_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 4 x bfloat> @test_vluxei16_v_bf16m1_tum(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], <vscale x 4 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x bfloat> @llvm.riscv.vluxei.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[VD]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret <vscale x 4 x bfloat> [[TMP0]]
+//
+vbfloat16m1_t test_vluxei16_v_bf16m1_tum(vbool16_t vm, vbfloat16m1_t vd,
+                                         const __bf16 *rs1, vuint16m1_t rs2,
+                                         size_t vl) {
+  return __riscv_vluxei16_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 8 x bfloat> @test_vluxei16_v_bf16m2_tum(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], <vscale x 8 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x bfloat> @llvm.riscv.vluxei.mask.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> [[VD]], ptr [[RS1]], <vscale x 8 x i16> [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret <vscale x 8 x bfloat> [[TMP0]]
+//
+vbfloat16m2_t test_vluxei16_v_bf16m2_tum(vbool8_t vm, vbfloat16m2_t vd,
+                                         const __bf16 *rs1, vuint16m2_t rs2,
+                                         size_t vl) {
+  return __riscv_vluxei16_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 16 x bfloat> @test_vluxei16_v_bf16m4_tum(
+// CHECK-RV64-SAME: <vscale x 16 x i1> [[VM:%.*]], <vscale x 16 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 16 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x bfloat> @llvm.riscv.vluxei.mask.nxv16bf16.nxv16i16.i64(<vscale x 16 x bfloat> [[VD]], ptr [[RS1]], <vscale x 16 x i16> [[RS2]], <vscale x 16 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret <vscale x 16 x bfloat> [[TMP0]]
+//
+vbfloat16m4_t test_vluxei16_v_bf16m4_tum(vbool4_t vm, vbfloat16m4_t vd,
+                                         const __bf16 *rs1, vuint16m4_t rs2,
+                                         size_t vl) {
+  return __riscv_vluxei16_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 32 x bfloat> @test_vluxei16_v_bf16m8_tum(
+// CHECK-RV64-SAME: <vscale x 32 x i1> [[VM:%.*]], <vscale x 32 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 32 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x bfloat> @llvm.riscv.vluxei.mask.nxv32bf16.nxv32i16.i64(<vscale x 32 x bfloat> [[VD]], ptr [[RS1]], <vscale x 32 x i16> [[RS2]], <vscale x 32 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret <vscale x 32 x bfloat> [[TMP0]]
+//
+vbfloat16m8_t test_vluxei16_v_bf16m8_tum(vbool2_t vm, vbfloat16m8_t vd,
+                                         const __bf16 *rs1, vuint16m8_t rs2,
+                                         size_t vl) {
+  return __riscv_vluxei16_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 1 x bfloat> @test_vluxei16_v_bf16mf4_tumu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], <vscale x 1 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x bfloat> @llvm.riscv.vluxei.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[VD]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 1 x bfloat> [[TMP0]]
+//
+vbfloat16mf4_t test_vluxei16_v_bf16mf4_tumu(vbool64_t vm, vbfloat16mf4_t vd,
+                                            const __bf16 *rs1, vuint16mf4_t rs2,
+                                            size_t vl) {
+  return __riscv_vluxei16_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 2 x bfloat> @test_vluxei16_v_bf16mf2_tumu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], <vscale x 2 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x bfloat> @llvm.riscv.vluxei.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[VD]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 2 x bfloat> [[TMP0]]
+//
+vbfloat16mf2_t test_vluxei16_v_bf16mf2_tumu(vbool32_t vm, vbfloat16mf2_t vd,
+                                            const __bf16 *rs1, vuint16mf2_t rs2,
+                                            size_t vl) {
+  return __riscv_vluxei16_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 4 x bfloat> @test_vluxei16_v_bf16m1_tumu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], <vscale x 4 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x bfloat> @llvm.riscv.vluxei.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[VD]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 4 x bfloat> [[TMP0]]
+//
+vbfloat16m1_t test_vluxei16_v_bf16m1_tumu(vbool16_t vm, vbfloat16m1_t vd,
+                                          const __bf16 *rs1, vuint16m1_t rs2,
+                                          size_t vl) {
+  return __riscv_vluxei16_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 8 x bfloat> @test_vluxei16_v_bf16m2_tumu(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], <vscale x 8 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x bfloat> @llvm.riscv.vluxei.mask.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> [[VD]], ptr [[RS1]], <vscale x 8 x i16> [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 8 x bfloat> [[TMP0]]
+//
+vbfloat16m2_t test_vluxei16_v_bf16m2_tumu(vbool8_t vm, vbfloat16m2_t vd,
+                                          const __bf16 *rs1, vuint16m2_t rs2,
+                                          size_t vl) {
+  return __riscv_vluxei16_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 16 x bfloat> @test_vluxei16_v_bf16m4_tumu(
+// CHECK-RV64-SAME: <vscale x 16 x i1> [[VM:%.*]], <vscale x 16 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 16 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x bfloat> @llvm.riscv.vluxei.mask.nxv16bf16.nxv16i16.i64(<vscale x 16 x bfloat> [[VD]], ptr [[RS1]], <vscale x 16 x i16> [[RS2]], <vscale x 16 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 16 x bfloat> [[TMP0]]
+//
+vbfloat16m4_t test_vluxei16_v_bf16m4_tumu(vbool4_t vm, vbfloat16m4_t vd,
+                                          const __bf16 *rs1, vuint16m4_t rs2,
+                                          size_t vl) {
+  return __riscv_vluxei16_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 32 x bfloat> @test_vluxei16_v_bf16m8_tumu(
+// CHECK-RV64-SAME: <vscale x 32 x i1> [[VM:%.*]], <vscale x 32 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 32 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x bfloat> @llvm.riscv.vluxei.mask.nxv32bf16.nxv32i16.i64(<vscale x 32 x bfloat> [[VD]], ptr [[RS1]], <vscale x 32 x i16> [[RS2]], <vscale x 32 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret <vscale x 32 x bfloat> [[TMP0]]
+//
+vbfloat16m8_t test_vluxei16_v_bf16m8_tumu(vbool2_t vm, vbfloat16m8_t vd,
+                                          const __bf16 *rs1, vuint16m8_t rs2,
+                                          size_t vl) {
+  return __riscv_vluxei16_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 1 x bfloat> @test_vluxei16_v_bf16mf4_mu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], <vscale x 1 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 1 x bfloat> @llvm.riscv.vluxei.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[VD]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret <vscale x 1 x bfloat> [[TMP0]]
+//
+vbfloat16mf4_t test_vluxei16_v_bf16mf4_mu(vbool64_t vm, vbfloat16mf4_t vd,
+                                          const __bf16 *rs1, vuint16mf4_t rs2,
+                                          size_t vl) {
+  return __riscv_vluxei16_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 2 x bfloat> @test_vluxei16_v_bf16mf2_mu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], <vscale x 2 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 2 x bfloat> @llvm.riscv.vluxei.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[VD]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret <vscale x 2 x bfloat> [[TMP0]]
+//
+vbfloat16mf2_t test_vluxei16_v_bf16mf2_mu(vbool32_t vm, vbfloat16mf2_t vd,
+                                          const __bf16 *rs1, vuint16mf2_t rs2,
+                                          size_t vl) {
+  return __riscv_vluxei16_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 4 x bfloat> @test_vluxei16_v_bf16m1_mu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], <vscale x 4 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x bfloat> @llvm.riscv.vluxei.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[VD]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret <vscale x 4 x bfloat> [[TMP0]]
+//
+vbfloat16m1_t test_vluxei16_v_bf16m1_mu(vbool16_t vm, vbfloat16m1_t vd,
+                                        const __bf16 *rs1, vuint16m1_t rs2,
+                                        size_t vl) {
+  return __riscv_vluxei16_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 8 x bfloat> @test_vluxei16_v_bf16m2_mu(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], <vscale x 8 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 8 x bfloat> @llvm.riscv.vluxei.mask.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> [[VD]], ptr [[RS1]], <vscale x 8 x i16> [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret <vscale x 8 x bfloat> [[TMP0]]
+//
+vbfloat16m2_t test_vluxei16_v_bf16m2_mu(vbool8_t vm, vbfloat16m2_t vd,
+                                        const __bf16 *rs1, vuint16m2_t rs2,
+                                        size_t vl) {
+  return __riscv_vluxei16_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 16 x bfloat> @test_vluxei16_v_bf16m4_mu(
+// CHECK-RV64-SAME: <vscale x 16 x i1> [[VM:%.*]], <vscale x 16 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 16 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x bfloat> @llvm.riscv.vluxei.mask.nxv16bf16.nxv16i16.i64(<vscale x 16 x bfloat> [[VD]], ptr [[RS1]], <vscale x 16 x i16> [[RS2]], <vscale x 16 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret <vscale x 16 x bfloat> [[TMP0]]
+//
+vbfloat16m4_t test_vluxei16_v_bf16m4_mu(vbool4_t vm, vbfloat16m4_t vd,
+                                        const __bf16 *rs1, vuint16m4_t rs2,
+                                        size_t vl) {
+  return __riscv_vluxei16_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local <vscale x 32 x bfloat> @test_vluxei16_v_bf16m8_mu(
+// CHECK-RV64-SAME: <vscale x 32 x i1> [[VM:%.*]], <vscale x 32 x bfloat> [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 32 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 32 x bfloat> @llvm.riscv.vluxei.mask.nxv32bf16.nxv32i16.i64(<vscale x 32 x bfloat> [[VD]], ptr [[RS1]], <vscale x 32 x i16> [[RS2]], <vscale x 32 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret <vscale x 32 x bfloat> [[TMP0]]
+//
+vbfloat16m8_t test_vluxei16_v_bf16m8_mu(vbool2_t vm, vbfloat16m8_t vd,
+                                        const __bf16 *rs1, vuint16m8_t rs2,
+                                        size_t vl) {
+  return __riscv_vluxei16_mu(vm, vd, rs1, rs2, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vluxseg2ei16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vluxseg2ei16.c
new file mode 100644
index 000000000000..40b94c5d4076
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vluxseg2ei16.c
@@ -0,0 +1,306 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vluxseg2ei16_v_bf16mf4x2_tu(
+// CHECK-RV64-SAME: { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vluxseg2.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP2]]
+//
+vbfloat16mf4x2_t test_vluxseg2ei16_v_bf16mf4x2_tu(vbfloat16mf4x2_t vd,
+                                                  const __bf16 *rs1,
+                                                  vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vluxseg2ei16_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vluxseg2ei16_v_bf16mf2x2_tu(
+// CHECK-RV64-SAME: { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vluxseg2.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP2]]
+//
+vbfloat16mf2x2_t test_vluxseg2ei16_v_bf16mf2x2_tu(vbfloat16mf2x2_t vd,
+                                                  const __bf16 *rs1,
+                                                  vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vluxseg2ei16_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vluxseg2ei16_v_bf16m1x2_tu(
+// CHECK-RV64-SAME: { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vluxseg2.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP2]]
+//
+vbfloat16m1x2_t test_vluxseg2ei16_v_bf16m1x2_tu(vbfloat16m1x2_t vd,
+                                                const __bf16 *rs1,
+                                                vuint16m1_t rs2, size_t vl) {
+  return __riscv_vluxseg2ei16_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vluxseg2ei16_v_bf16m2x2_tu(
+// CHECK-RV64-SAME: { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vluxseg2.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 8 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP2]]
+//
+vbfloat16m2x2_t test_vluxseg2ei16_v_bf16m2x2_tu(vbfloat16m2x2_t vd,
+                                                const __bf16 *rs1,
+                                                vuint16m2_t rs2, size_t vl) {
+  return __riscv_vluxseg2ei16_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @test_vluxseg2ei16_v_bf16m4x2_tu(
+// CHECK-RV64-SAME: { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 16 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @llvm.riscv.vluxseg2.nxv16bf16.nxv16i16.i64(<vscale x 16 x bfloat> [[TMP0]], <vscale x 16 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 16 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[TMP2]]
+//
+vbfloat16m4x2_t test_vluxseg2ei16_v_bf16m4x2_tu(vbfloat16m4x2_t vd,
+                                                const __bf16 *rs1,
+                                                vuint16m4_t rs2, size_t vl) {
+  return __riscv_vluxseg2ei16_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vluxseg2ei16_v_bf16mf4x2_tum(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vluxseg2.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP2]]
+//
+vbfloat16mf4x2_t test_vluxseg2ei16_v_bf16mf4x2_tum(vbool64_t vm,
+                                                   vbfloat16mf4x2_t vd,
+                                                   const __bf16 *rs1,
+                                                   vuint16mf4_t rs2,
+                                                   size_t vl) {
+  return __riscv_vluxseg2ei16_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vluxseg2ei16_v_bf16mf2x2_tum(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vluxseg2.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP2]]
+//
+vbfloat16mf2x2_t test_vluxseg2ei16_v_bf16mf2x2_tum(vbool32_t vm,
+                                                   vbfloat16mf2x2_t vd,
+                                                   const __bf16 *rs1,
+                                                   vuint16mf2_t rs2,
+                                                   size_t vl) {
+  return __riscv_vluxseg2ei16_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vluxseg2ei16_v_bf16m1x2_tum(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vluxseg2.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP2]]
+//
+vbfloat16m1x2_t test_vluxseg2ei16_v_bf16m1x2_tum(vbool16_t vm,
+                                                 vbfloat16m1x2_t vd,
+                                                 const __bf16 *rs1,
+                                                 vuint16m1_t rs2, size_t vl) {
+  return __riscv_vluxseg2ei16_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vluxseg2ei16_v_bf16m2x2_tum(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vluxseg2.mask.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 8 x i16> [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP2]]
+//
+vbfloat16m2x2_t test_vluxseg2ei16_v_bf16m2x2_tum(vbool8_t vm,
+                                                 vbfloat16m2x2_t vd,
+                                                 const __bf16 *rs1,
+                                                 vuint16m2_t rs2, size_t vl) {
+  return __riscv_vluxseg2ei16_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @test_vluxseg2ei16_v_bf16m4x2_tum(
+// CHECK-RV64-SAME: <vscale x 16 x i1> [[VM:%.*]], { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 16 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @llvm.riscv.vluxseg2.mask.nxv16bf16.nxv16i16.i64(<vscale x 16 x bfloat> [[TMP0]], <vscale x 16 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 16 x i16> [[RS2]], <vscale x 16 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[TMP2]]
+//
+vbfloat16m4x2_t test_vluxseg2ei16_v_bf16m4x2_tum(vbool4_t vm,
+                                                 vbfloat16m4x2_t vd,
+                                                 const __bf16 *rs1,
+                                                 vuint16m4_t rs2, size_t vl) {
+  return __riscv_vluxseg2ei16_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vluxseg2ei16_v_bf16mf4x2_tumu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vluxseg2.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP2]]
+//
+vbfloat16mf4x2_t test_vluxseg2ei16_v_bf16mf4x2_tumu(vbool64_t vm,
+                                                    vbfloat16mf4x2_t vd,
+                                                    const __bf16 *rs1,
+                                                    vuint16mf4_t rs2,
+                                                    size_t vl) {
+  return __riscv_vluxseg2ei16_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vluxseg2ei16_v_bf16mf2x2_tumu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vluxseg2.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP2]]
+//
+vbfloat16mf2x2_t test_vluxseg2ei16_v_bf16mf2x2_tumu(vbool32_t vm,
+                                                    vbfloat16mf2x2_t vd,
+                                                    const __bf16 *rs1,
+                                                    vuint16mf2_t rs2,
+                                                    size_t vl) {
+  return __riscv_vluxseg2ei16_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vluxseg2ei16_v_bf16m1x2_tumu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vluxseg2.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP2]]
+//
+vbfloat16m1x2_t test_vluxseg2ei16_v_bf16m1x2_tumu(vbool16_t vm,
+                                                  vbfloat16m1x2_t vd,
+                                                  const __bf16 *rs1,
+                                                  vuint16m1_t rs2, size_t vl) {
+  return __riscv_vluxseg2ei16_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vluxseg2ei16_v_bf16m2x2_tumu(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vluxseg2.mask.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 8 x i16> [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP2]]
+//
+vbfloat16m2x2_t test_vluxseg2ei16_v_bf16m2x2_tumu(vbool8_t vm,
+                                                  vbfloat16m2x2_t vd,
+                                                  const __bf16 *rs1,
+                                                  vuint16m2_t rs2, size_t vl) {
+  return __riscv_vluxseg2ei16_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @test_vluxseg2ei16_v_bf16m4x2_tumu(
+// CHECK-RV64-SAME: <vscale x 16 x i1> [[VM:%.*]], { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 16 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @llvm.riscv.vluxseg2.mask.nxv16bf16.nxv16i16.i64(<vscale x 16 x bfloat> [[TMP0]], <vscale x 16 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 16 x i16> [[RS2]], <vscale x 16 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[TMP2]]
+//
+vbfloat16m4x2_t test_vluxseg2ei16_v_bf16m4x2_tumu(vbool4_t vm,
+                                                  vbfloat16m4x2_t vd,
+                                                  const __bf16 *rs1,
+                                                  vuint16m4_t rs2, size_t vl) {
+  return __riscv_vluxseg2ei16_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vluxseg2ei16_v_bf16mf4x2_mu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vluxseg2.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP2]]
+//
+vbfloat16mf4x2_t test_vluxseg2ei16_v_bf16mf4x2_mu(vbool64_t vm,
+                                                  vbfloat16mf4x2_t vd,
+                                                  const __bf16 *rs1,
+                                                  vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vluxseg2ei16_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vluxseg2ei16_v_bf16mf2x2_mu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vluxseg2.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP2]]
+//
+vbfloat16mf2x2_t test_vluxseg2ei16_v_bf16mf2x2_mu(vbool32_t vm,
+                                                  vbfloat16mf2x2_t vd,
+                                                  const __bf16 *rs1,
+                                                  vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vluxseg2ei16_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vluxseg2ei16_v_bf16m1x2_mu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vluxseg2.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP2]]
+//
+vbfloat16m1x2_t test_vluxseg2ei16_v_bf16m1x2_mu(vbool16_t vm,
+                                                vbfloat16m1x2_t vd,
+                                                const __bf16 *rs1,
+                                                vuint16m1_t rs2, size_t vl) {
+  return __riscv_vluxseg2ei16_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vluxseg2ei16_v_bf16m2x2_mu(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vluxseg2.mask.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 8 x i16> [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP2]]
+//
+vbfloat16m2x2_t test_vluxseg2ei16_v_bf16m2x2_mu(vbool8_t vm, vbfloat16m2x2_t vd,
+                                                const __bf16 *rs1,
+                                                vuint16m2_t rs2, size_t vl) {
+  return __riscv_vluxseg2ei16_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @test_vluxseg2ei16_v_bf16m4x2_mu(
+// CHECK-RV64-SAME: <vscale x 16 x i1> [[VM:%.*]], { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 16 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = call { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } @llvm.riscv.vluxseg2.mask.nxv16bf16.nxv16i16.i64(<vscale x 16 x bfloat> [[TMP0]], <vscale x 16 x bfloat> [[TMP1]], ptr [[RS1]], <vscale x 16 x i16> [[RS2]], <vscale x 16 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 16 x bfloat>, <vscale x 16 x bfloat> } [[TMP2]]
+//
+vbfloat16m4x2_t test_vluxseg2ei16_v_bf16m4x2_mu(vbool4_t vm, vbfloat16m4x2_t vd,
+                                                const __bf16 *rs1,
+                                                vuint16m4_t rs2, size_t vl) {
+  return __riscv_vluxseg2ei16_mu(vm, vd, rs1, rs2, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vluxseg3ei16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vluxseg3ei16.c
new file mode 100644
index 000000000000..305e811f197b
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vluxseg3ei16.c
@@ -0,0 +1,264 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vluxseg3ei16_v_bf16mf4x3_tu(
+// CHECK-RV64-SAME: { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vluxseg3.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP3]]
+//
+vbfloat16mf4x3_t test_vluxseg3ei16_v_bf16mf4x3_tu(vbfloat16mf4x3_t vd,
+                                                  const __bf16 *rs1,
+                                                  vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vluxseg3ei16_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vluxseg3ei16_v_bf16mf2x3_tu(
+// CHECK-RV64-SAME: { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vluxseg3.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP3]]
+//
+vbfloat16mf2x3_t test_vluxseg3ei16_v_bf16mf2x3_tu(vbfloat16mf2x3_t vd,
+                                                  const __bf16 *rs1,
+                                                  vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vluxseg3ei16_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vluxseg3ei16_v_bf16m1x3_tu(
+// CHECK-RV64-SAME: { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vluxseg3.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP3]]
+//
+vbfloat16m1x3_t test_vluxseg3ei16_v_bf16m1x3_tu(vbfloat16m1x3_t vd,
+                                                const __bf16 *rs1,
+                                                vuint16m1_t rs2, size_t vl) {
+  return __riscv_vluxseg3ei16_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vluxseg3ei16_v_bf16m2x3_tu(
+// CHECK-RV64-SAME: { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vluxseg3.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 8 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP3]]
+//
+vbfloat16m2x3_t test_vluxseg3ei16_v_bf16m2x3_tu(vbfloat16m2x3_t vd,
+                                                const __bf16 *rs1,
+                                                vuint16m2_t rs2, size_t vl) {
+  return __riscv_vluxseg3ei16_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vluxseg3ei16_v_bf16mf4x3_tum(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vluxseg3.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP3]]
+//
+vbfloat16mf4x3_t test_vluxseg3ei16_v_bf16mf4x3_tum(vbool64_t vm,
+                                                   vbfloat16mf4x3_t vd,
+                                                   const __bf16 *rs1,
+                                                   vuint16mf4_t rs2,
+                                                   size_t vl) {
+  return __riscv_vluxseg3ei16_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vluxseg3ei16_v_bf16mf2x3_tum(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vluxseg3.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP3]]
+//
+vbfloat16mf2x3_t test_vluxseg3ei16_v_bf16mf2x3_tum(vbool32_t vm,
+                                                   vbfloat16mf2x3_t vd,
+                                                   const __bf16 *rs1,
+                                                   vuint16mf2_t rs2,
+                                                   size_t vl) {
+  return __riscv_vluxseg3ei16_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vluxseg3ei16_v_bf16m1x3_tum(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vluxseg3.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP3]]
+//
+vbfloat16m1x3_t test_vluxseg3ei16_v_bf16m1x3_tum(vbool16_t vm,
+                                                 vbfloat16m1x3_t vd,
+                                                 const __bf16 *rs1,
+                                                 vuint16m1_t rs2, size_t vl) {
+  return __riscv_vluxseg3ei16_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vluxseg3ei16_v_bf16m2x3_tum(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vluxseg3.mask.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 8 x i16> [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP3]]
+//
+vbfloat16m2x3_t test_vluxseg3ei16_v_bf16m2x3_tum(vbool8_t vm,
+                                                 vbfloat16m2x3_t vd,
+                                                 const __bf16 *rs1,
+                                                 vuint16m2_t rs2, size_t vl) {
+  return __riscv_vluxseg3ei16_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vluxseg3ei16_v_bf16mf4x3_tumu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vluxseg3.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP3]]
+//
+vbfloat16mf4x3_t test_vluxseg3ei16_v_bf16mf4x3_tumu(vbool64_t vm,
+                                                    vbfloat16mf4x3_t vd,
+                                                    const __bf16 *rs1,
+                                                    vuint16mf4_t rs2,
+                                                    size_t vl) {
+  return __riscv_vluxseg3ei16_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vluxseg3ei16_v_bf16mf2x3_tumu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vluxseg3.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP3]]
+//
+vbfloat16mf2x3_t test_vluxseg3ei16_v_bf16mf2x3_tumu(vbool32_t vm,
+                                                    vbfloat16mf2x3_t vd,
+                                                    const __bf16 *rs1,
+                                                    vuint16mf2_t rs2,
+                                                    size_t vl) {
+  return __riscv_vluxseg3ei16_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vluxseg3ei16_v_bf16m1x3_tumu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vluxseg3.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP3]]
+//
+vbfloat16m1x3_t test_vluxseg3ei16_v_bf16m1x3_tumu(vbool16_t vm,
+                                                  vbfloat16m1x3_t vd,
+                                                  const __bf16 *rs1,
+                                                  vuint16m1_t rs2, size_t vl) {
+  return __riscv_vluxseg3ei16_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vluxseg3ei16_v_bf16m2x3_tumu(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vluxseg3.mask.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 8 x i16> [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP3]]
+//
+vbfloat16m2x3_t test_vluxseg3ei16_v_bf16m2x3_tumu(vbool8_t vm,
+                                                  vbfloat16m2x3_t vd,
+                                                  const __bf16 *rs1,
+                                                  vuint16m2_t rs2, size_t vl) {
+  return __riscv_vluxseg3ei16_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vluxseg3ei16_v_bf16mf4x3_mu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vluxseg3.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP3]]
+//
+vbfloat16mf4x3_t test_vluxseg3ei16_v_bf16mf4x3_mu(vbool64_t vm,
+                                                  vbfloat16mf4x3_t vd,
+                                                  const __bf16 *rs1,
+                                                  vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vluxseg3ei16_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vluxseg3ei16_v_bf16mf2x3_mu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vluxseg3.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP3]]
+//
+vbfloat16mf2x3_t test_vluxseg3ei16_v_bf16mf2x3_mu(vbool32_t vm,
+                                                  vbfloat16mf2x3_t vd,
+                                                  const __bf16 *rs1,
+                                                  vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vluxseg3ei16_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vluxseg3ei16_v_bf16m1x3_mu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vluxseg3.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP3]]
+//
+vbfloat16m1x3_t test_vluxseg3ei16_v_bf16m1x3_mu(vbool16_t vm,
+                                                vbfloat16m1x3_t vd,
+                                                const __bf16 *rs1,
+                                                vuint16m1_t rs2, size_t vl) {
+  return __riscv_vluxseg3ei16_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vluxseg3ei16_v_bf16m2x3_mu(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vluxseg3.mask.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], ptr [[RS1]], <vscale x 8 x i16> [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP3]]
+//
+vbfloat16m2x3_t test_vluxseg3ei16_v_bf16m2x3_mu(vbool8_t vm, vbfloat16m2x3_t vd,
+                                                const __bf16 *rs1,
+                                                vuint16m2_t rs2, size_t vl) {
+  return __riscv_vluxseg3ei16_mu(vm, vd, rs1, rs2, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vluxseg4ei16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vluxseg4ei16.c
new file mode 100644
index 000000000000..eeb7834b44ef
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vluxseg4ei16.c
@@ -0,0 +1,280 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vluxseg4ei16_v_bf16mf4x4_tu(
+// CHECK-RV64-SAME: { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vluxseg4.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP4]]
+//
+vbfloat16mf4x4_t test_vluxseg4ei16_v_bf16mf4x4_tu(vbfloat16mf4x4_t vd,
+                                                  const __bf16 *rs1,
+                                                  vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vluxseg4ei16_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vluxseg4ei16_v_bf16mf2x4_tu(
+// CHECK-RV64-SAME: { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vluxseg4.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP4]]
+//
+vbfloat16mf2x4_t test_vluxseg4ei16_v_bf16mf2x4_tu(vbfloat16mf2x4_t vd,
+                                                  const __bf16 *rs1,
+                                                  vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vluxseg4ei16_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vluxseg4ei16_v_bf16m1x4_tu(
+// CHECK-RV64-SAME: { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vluxseg4.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP4]]
+//
+vbfloat16m1x4_t test_vluxseg4ei16_v_bf16m1x4_tu(vbfloat16m1x4_t vd,
+                                                const __bf16 *rs1,
+                                                vuint16m1_t rs2, size_t vl) {
+  return __riscv_vluxseg4ei16_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vluxseg4ei16_v_bf16m2x4_tu(
+// CHECK-RV64-SAME: { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vluxseg4.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], <vscale x 8 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 8 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP4]]
+//
+vbfloat16m2x4_t test_vluxseg4ei16_v_bf16m2x4_tu(vbfloat16m2x4_t vd,
+                                                const __bf16 *rs1,
+                                                vuint16m2_t rs2, size_t vl) {
+  return __riscv_vluxseg4ei16_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vluxseg4ei16_v_bf16mf4x4_tum(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vluxseg4.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP4]]
+//
+vbfloat16mf4x4_t test_vluxseg4ei16_v_bf16mf4x4_tum(vbool64_t vm,
+                                                   vbfloat16mf4x4_t vd,
+                                                   const __bf16 *rs1,
+                                                   vuint16mf4_t rs2,
+                                                   size_t vl) {
+  return __riscv_vluxseg4ei16_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vluxseg4ei16_v_bf16mf2x4_tum(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vluxseg4.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP4]]
+//
+vbfloat16mf2x4_t test_vluxseg4ei16_v_bf16mf2x4_tum(vbool32_t vm,
+                                                   vbfloat16mf2x4_t vd,
+                                                   const __bf16 *rs1,
+                                                   vuint16mf2_t rs2,
+                                                   size_t vl) {
+  return __riscv_vluxseg4ei16_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vluxseg4ei16_v_bf16m1x4_tum(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vluxseg4.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP4]]
+//
+vbfloat16m1x4_t test_vluxseg4ei16_v_bf16m1x4_tum(vbool16_t vm,
+                                                 vbfloat16m1x4_t vd,
+                                                 const __bf16 *rs1,
+                                                 vuint16m1_t rs2, size_t vl) {
+  return __riscv_vluxseg4ei16_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vluxseg4ei16_v_bf16m2x4_tum(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vluxseg4.mask.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], <vscale x 8 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 8 x i16> [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP4]]
+//
+vbfloat16m2x4_t test_vluxseg4ei16_v_bf16m2x4_tum(vbool8_t vm,
+                                                 vbfloat16m2x4_t vd,
+                                                 const __bf16 *rs1,
+                                                 vuint16m2_t rs2, size_t vl) {
+  return __riscv_vluxseg4ei16_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vluxseg4ei16_v_bf16mf4x4_tumu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vluxseg4.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP4]]
+//
+vbfloat16mf4x4_t test_vluxseg4ei16_v_bf16mf4x4_tumu(vbool64_t vm,
+                                                    vbfloat16mf4x4_t vd,
+                                                    const __bf16 *rs1,
+                                                    vuint16mf4_t rs2,
+                                                    size_t vl) {
+  return __riscv_vluxseg4ei16_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vluxseg4ei16_v_bf16mf2x4_tumu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vluxseg4.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP4]]
+//
+vbfloat16mf2x4_t test_vluxseg4ei16_v_bf16mf2x4_tumu(vbool32_t vm,
+                                                    vbfloat16mf2x4_t vd,
+                                                    const __bf16 *rs1,
+                                                    vuint16mf2_t rs2,
+                                                    size_t vl) {
+  return __riscv_vluxseg4ei16_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vluxseg4ei16_v_bf16m1x4_tumu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vluxseg4.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP4]]
+//
+vbfloat16m1x4_t test_vluxseg4ei16_v_bf16m1x4_tumu(vbool16_t vm,
+                                                  vbfloat16m1x4_t vd,
+                                                  const __bf16 *rs1,
+                                                  vuint16m1_t rs2, size_t vl) {
+  return __riscv_vluxseg4ei16_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vluxseg4ei16_v_bf16m2x4_tumu(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vluxseg4.mask.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], <vscale x 8 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 8 x i16> [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP4]]
+//
+vbfloat16m2x4_t test_vluxseg4ei16_v_bf16m2x4_tumu(vbool8_t vm,
+                                                  vbfloat16m2x4_t vd,
+                                                  const __bf16 *rs1,
+                                                  vuint16m2_t rs2, size_t vl) {
+  return __riscv_vluxseg4ei16_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vluxseg4ei16_v_bf16mf4x4_mu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vluxseg4.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP4]]
+//
+vbfloat16mf4x4_t test_vluxseg4ei16_v_bf16mf4x4_mu(vbool64_t vm,
+                                                  vbfloat16mf4x4_t vd,
+                                                  const __bf16 *rs1,
+                                                  vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vluxseg4ei16_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vluxseg4ei16_v_bf16mf2x4_mu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vluxseg4.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP4]]
+//
+vbfloat16mf2x4_t test_vluxseg4ei16_v_bf16mf2x4_mu(vbool32_t vm,
+                                                  vbfloat16mf2x4_t vd,
+                                                  const __bf16 *rs1,
+                                                  vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vluxseg4ei16_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vluxseg4ei16_v_bf16m1x4_mu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vluxseg4.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP4]]
+//
+vbfloat16m1x4_t test_vluxseg4ei16_v_bf16m1x4_mu(vbool16_t vm,
+                                                vbfloat16m1x4_t vd,
+                                                const __bf16 *rs1,
+                                                vuint16m1_t rs2, size_t vl) {
+  return __riscv_vluxseg4ei16_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @test_vluxseg4ei16_v_bf16m2x4_mu(
+// CHECK-RV64-SAME: <vscale x 8 x i1> [[VM:%.*]], { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 8 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = call { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } @llvm.riscv.vluxseg4.mask.nxv8bf16.nxv8i16.i64(<vscale x 8 x bfloat> [[TMP0]], <vscale x 8 x bfloat> [[TMP1]], <vscale x 8 x bfloat> [[TMP2]], <vscale x 8 x bfloat> [[TMP3]], ptr [[RS1]], <vscale x 8 x i16> [[RS2]], <vscale x 8 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat>, <vscale x 8 x bfloat> } [[TMP4]]
+//
+vbfloat16m2x4_t test_vluxseg4ei16_v_bf16m2x4_mu(vbool8_t vm, vbfloat16m2x4_t vd,
+                                                const __bf16 *rs1,
+                                                vuint16m2_t rs2, size_t vl) {
+  return __riscv_vluxseg4ei16_mu(vm, vd, rs1, rs2, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vluxseg5ei16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vluxseg5ei16.c
new file mode 100644
index 000000000000..0efa49935450
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vluxseg5ei16.c
@@ -0,0 +1,226 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vluxseg5ei16_v_bf16mf4x5_tu(
+// CHECK-RV64-SAME: { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vluxseg5.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP5]]
+//
+vbfloat16mf4x5_t test_vluxseg5ei16_v_bf16mf4x5_tu(vbfloat16mf4x5_t vd,
+                                                  const __bf16 *rs1,
+                                                  vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vluxseg5ei16_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vluxseg5ei16_v_bf16mf2x5_tu(
+// CHECK-RV64-SAME: { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vluxseg5.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP5]]
+//
+vbfloat16mf2x5_t test_vluxseg5ei16_v_bf16mf2x5_tu(vbfloat16mf2x5_t vd,
+                                                  const __bf16 *rs1,
+                                                  vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vluxseg5ei16_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vluxseg5ei16_v_bf16m1x5_tu(
+// CHECK-RV64-SAME: { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vluxseg5.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP5]]
+//
+vbfloat16m1x5_t test_vluxseg5ei16_v_bf16m1x5_tu(vbfloat16m1x5_t vd,
+                                                const __bf16 *rs1,
+                                                vuint16m1_t rs2, size_t vl) {
+  return __riscv_vluxseg5ei16_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vluxseg5ei16_v_bf16mf4x5_tum(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vluxseg5.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP5]]
+//
+vbfloat16mf4x5_t test_vluxseg5ei16_v_bf16mf4x5_tum(vbool64_t vm,
+                                                   vbfloat16mf4x5_t vd,
+                                                   const __bf16 *rs1,
+                                                   vuint16mf4_t rs2,
+                                                   size_t vl) {
+  return __riscv_vluxseg5ei16_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vluxseg5ei16_v_bf16mf2x5_tum(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vluxseg5.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP5]]
+//
+vbfloat16mf2x5_t test_vluxseg5ei16_v_bf16mf2x5_tum(vbool32_t vm,
+                                                   vbfloat16mf2x5_t vd,
+                                                   const __bf16 *rs1,
+                                                   vuint16mf2_t rs2,
+                                                   size_t vl) {
+  return __riscv_vluxseg5ei16_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vluxseg5ei16_v_bf16m1x5_tum(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vluxseg5.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP5]]
+//
+vbfloat16m1x5_t test_vluxseg5ei16_v_bf16m1x5_tum(vbool16_t vm,
+                                                 vbfloat16m1x5_t vd,
+                                                 const __bf16 *rs1,
+                                                 vuint16m1_t rs2, size_t vl) {
+  return __riscv_vluxseg5ei16_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vluxseg5ei16_v_bf16mf4x5_tumu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vluxseg5.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP5]]
+//
+vbfloat16mf4x5_t test_vluxseg5ei16_v_bf16mf4x5_tumu(vbool64_t vm,
+                                                    vbfloat16mf4x5_t vd,
+                                                    const __bf16 *rs1,
+                                                    vuint16mf4_t rs2,
+                                                    size_t vl) {
+  return __riscv_vluxseg5ei16_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vluxseg5ei16_v_bf16mf2x5_tumu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vluxseg5.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP5]]
+//
+vbfloat16mf2x5_t test_vluxseg5ei16_v_bf16mf2x5_tumu(vbool32_t vm,
+                                                    vbfloat16mf2x5_t vd,
+                                                    const __bf16 *rs1,
+                                                    vuint16mf2_t rs2,
+                                                    size_t vl) {
+  return __riscv_vluxseg5ei16_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vluxseg5ei16_v_bf16m1x5_tumu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vluxseg5.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP5]]
+//
+vbfloat16m1x5_t test_vluxseg5ei16_v_bf16m1x5_tumu(vbool16_t vm,
+                                                  vbfloat16m1x5_t vd,
+                                                  const __bf16 *rs1,
+                                                  vuint16m1_t rs2, size_t vl) {
+  return __riscv_vluxseg5ei16_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vluxseg5ei16_v_bf16mf4x5_mu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vluxseg5.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP5]]
+//
+vbfloat16mf4x5_t test_vluxseg5ei16_v_bf16mf4x5_mu(vbool64_t vm,
+                                                  vbfloat16mf4x5_t vd,
+                                                  const __bf16 *rs1,
+                                                  vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vluxseg5ei16_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vluxseg5ei16_v_bf16mf2x5_mu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vluxseg5.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP5]]
+//
+vbfloat16mf2x5_t test_vluxseg5ei16_v_bf16mf2x5_mu(vbool32_t vm,
+                                                  vbfloat16mf2x5_t vd,
+                                                  const __bf16 *rs1,
+                                                  vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vluxseg5ei16_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vluxseg5ei16_v_bf16m1x5_mu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vluxseg5.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP5]]
+//
+vbfloat16m1x5_t test_vluxseg5ei16_v_bf16m1x5_mu(vbool16_t vm,
+                                                vbfloat16m1x5_t vd,
+                                                const __bf16 *rs1,
+                                                vuint16m1_t rs2, size_t vl) {
+  return __riscv_vluxseg5ei16_mu(vm, vd, rs1, rs2, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vluxseg6ei16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vluxseg6ei16.c
new file mode 100644
index 000000000000..42f09322f7b7
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vluxseg6ei16.c
@@ -0,0 +1,238 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vluxseg6ei16_v_bf16mf4x6_tu(
+// CHECK-RV64-SAME: { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vluxseg6.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP6]]
+//
+vbfloat16mf4x6_t test_vluxseg6ei16_v_bf16mf4x6_tu(vbfloat16mf4x6_t vd,
+                                                  const __bf16 *rs1,
+                                                  vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vluxseg6ei16_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vluxseg6ei16_v_bf16mf2x6_tu(
+// CHECK-RV64-SAME: { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vluxseg6.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP6]]
+//
+vbfloat16mf2x6_t test_vluxseg6ei16_v_bf16mf2x6_tu(vbfloat16mf2x6_t vd,
+                                                  const __bf16 *rs1,
+                                                  vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vluxseg6ei16_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vluxseg6ei16_v_bf16m1x6_tu(
+// CHECK-RV64-SAME: { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vluxseg6.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP6]]
+//
+vbfloat16m1x6_t test_vluxseg6ei16_v_bf16m1x6_tu(vbfloat16m1x6_t vd,
+                                                const __bf16 *rs1,
+                                                vuint16m1_t rs2, size_t vl) {
+  return __riscv_vluxseg6ei16_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vluxseg6ei16_v_bf16mf4x6_tum(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vluxseg6.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP6]]
+//
+vbfloat16mf4x6_t test_vluxseg6ei16_v_bf16mf4x6_tum(vbool64_t vm,
+                                                   vbfloat16mf4x6_t vd,
+                                                   const __bf16 *rs1,
+                                                   vuint16mf4_t rs2,
+                                                   size_t vl) {
+  return __riscv_vluxseg6ei16_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vluxseg6ei16_v_bf16mf2x6_tum(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vluxseg6.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP6]]
+//
+vbfloat16mf2x6_t test_vluxseg6ei16_v_bf16mf2x6_tum(vbool32_t vm,
+                                                   vbfloat16mf2x6_t vd,
+                                                   const __bf16 *rs1,
+                                                   vuint16mf2_t rs2,
+                                                   size_t vl) {
+  return __riscv_vluxseg6ei16_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vluxseg6ei16_v_bf16m1x6_tum(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vluxseg6.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP6]]
+//
+vbfloat16m1x6_t test_vluxseg6ei16_v_bf16m1x6_tum(vbool16_t vm,
+                                                 vbfloat16m1x6_t vd,
+                                                 const __bf16 *rs1,
+                                                 vuint16m1_t rs2, size_t vl) {
+  return __riscv_vluxseg6ei16_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vluxseg6ei16_v_bf16mf4x6_tumu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vluxseg6.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP6]]
+//
+vbfloat16mf4x6_t test_vluxseg6ei16_v_bf16mf4x6_tumu(vbool64_t vm,
+                                                    vbfloat16mf4x6_t vd,
+                                                    const __bf16 *rs1,
+                                                    vuint16mf4_t rs2,
+                                                    size_t vl) {
+  return __riscv_vluxseg6ei16_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vluxseg6ei16_v_bf16mf2x6_tumu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vluxseg6.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP6]]
+//
+vbfloat16mf2x6_t test_vluxseg6ei16_v_bf16mf2x6_tumu(vbool32_t vm,
+                                                    vbfloat16mf2x6_t vd,
+                                                    const __bf16 *rs1,
+                                                    vuint16mf2_t rs2,
+                                                    size_t vl) {
+  return __riscv_vluxseg6ei16_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vluxseg6ei16_v_bf16m1x6_tumu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vluxseg6.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP6]]
+//
+vbfloat16m1x6_t test_vluxseg6ei16_v_bf16m1x6_tumu(vbool16_t vm,
+                                                  vbfloat16m1x6_t vd,
+                                                  const __bf16 *rs1,
+                                                  vuint16m1_t rs2, size_t vl) {
+  return __riscv_vluxseg6ei16_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vluxseg6ei16_v_bf16mf4x6_mu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vluxseg6.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP6]]
+//
+vbfloat16mf4x6_t test_vluxseg6ei16_v_bf16mf4x6_mu(vbool64_t vm,
+                                                  vbfloat16mf4x6_t vd,
+                                                  const __bf16 *rs1,
+                                                  vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vluxseg6ei16_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vluxseg6ei16_v_bf16mf2x6_mu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vluxseg6.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP6]]
+//
+vbfloat16mf2x6_t test_vluxseg6ei16_v_bf16mf2x6_mu(vbool32_t vm,
+                                                  vbfloat16mf2x6_t vd,
+                                                  const __bf16 *rs1,
+                                                  vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vluxseg6ei16_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vluxseg6ei16_v_bf16m1x6_mu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vluxseg6.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP6]]
+//
+vbfloat16m1x6_t test_vluxseg6ei16_v_bf16m1x6_mu(vbool16_t vm,
+                                                vbfloat16m1x6_t vd,
+                                                const __bf16 *rs1,
+                                                vuint16m1_t rs2, size_t vl) {
+  return __riscv_vluxseg6ei16_mu(vm, vd, rs1, rs2, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vluxseg7ei16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vluxseg7ei16.c
new file mode 100644
index 000000000000..7a6f6eef4e0b
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vluxseg7ei16.c
@@ -0,0 +1,250 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vluxseg7ei16_v_bf16mf4x7_tu(
+// CHECK-RV64-SAME: { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vluxseg7.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP7]]
+//
+vbfloat16mf4x7_t test_vluxseg7ei16_v_bf16mf4x7_tu(vbfloat16mf4x7_t vd,
+                                                  const __bf16 *rs1,
+                                                  vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vluxseg7ei16_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vluxseg7ei16_v_bf16mf2x7_tu(
+// CHECK-RV64-SAME: { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vluxseg7.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP7]]
+//
+vbfloat16mf2x7_t test_vluxseg7ei16_v_bf16mf2x7_tu(vbfloat16mf2x7_t vd,
+                                                  const __bf16 *rs1,
+                                                  vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vluxseg7ei16_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vluxseg7ei16_v_bf16m1x7_tu(
+// CHECK-RV64-SAME: { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vluxseg7.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP7]]
+//
+vbfloat16m1x7_t test_vluxseg7ei16_v_bf16m1x7_tu(vbfloat16m1x7_t vd,
+                                                const __bf16 *rs1,
+                                                vuint16m1_t rs2, size_t vl) {
+  return __riscv_vluxseg7ei16_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vluxseg7ei16_v_bf16mf4x7_tum(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vluxseg7.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP7]]
+//
+vbfloat16mf4x7_t test_vluxseg7ei16_v_bf16mf4x7_tum(vbool64_t vm,
+                                                   vbfloat16mf4x7_t vd,
+                                                   const __bf16 *rs1,
+                                                   vuint16mf4_t rs2,
+                                                   size_t vl) {
+  return __riscv_vluxseg7ei16_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vluxseg7ei16_v_bf16mf2x7_tum(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vluxseg7.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP7]]
+//
+vbfloat16mf2x7_t test_vluxseg7ei16_v_bf16mf2x7_tum(vbool32_t vm,
+                                                   vbfloat16mf2x7_t vd,
+                                                   const __bf16 *rs1,
+                                                   vuint16mf2_t rs2,
+                                                   size_t vl) {
+  return __riscv_vluxseg7ei16_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vluxseg7ei16_v_bf16m1x7_tum(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vluxseg7.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP7]]
+//
+vbfloat16m1x7_t test_vluxseg7ei16_v_bf16m1x7_tum(vbool16_t vm,
+                                                 vbfloat16m1x7_t vd,
+                                                 const __bf16 *rs1,
+                                                 vuint16m1_t rs2, size_t vl) {
+  return __riscv_vluxseg7ei16_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vluxseg7ei16_v_bf16mf4x7_tumu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vluxseg7.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP7]]
+//
+vbfloat16mf4x7_t test_vluxseg7ei16_v_bf16mf4x7_tumu(vbool64_t vm,
+                                                    vbfloat16mf4x7_t vd,
+                                                    const __bf16 *rs1,
+                                                    vuint16mf4_t rs2,
+                                                    size_t vl) {
+  return __riscv_vluxseg7ei16_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vluxseg7ei16_v_bf16mf2x7_tumu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vluxseg7.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP7]]
+//
+vbfloat16mf2x7_t test_vluxseg7ei16_v_bf16mf2x7_tumu(vbool32_t vm,
+                                                    vbfloat16mf2x7_t vd,
+                                                    const __bf16 *rs1,
+                                                    vuint16mf2_t rs2,
+                                                    size_t vl) {
+  return __riscv_vluxseg7ei16_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vluxseg7ei16_v_bf16m1x7_tumu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vluxseg7.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP7]]
+//
+vbfloat16m1x7_t test_vluxseg7ei16_v_bf16m1x7_tumu(vbool16_t vm,
+                                                  vbfloat16m1x7_t vd,
+                                                  const __bf16 *rs1,
+                                                  vuint16m1_t rs2, size_t vl) {
+  return __riscv_vluxseg7ei16_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vluxseg7ei16_v_bf16mf4x7_mu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vluxseg7.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP7]]
+//
+vbfloat16mf4x7_t test_vluxseg7ei16_v_bf16mf4x7_mu(vbool64_t vm,
+                                                  vbfloat16mf4x7_t vd,
+                                                  const __bf16 *rs1,
+                                                  vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vluxseg7ei16_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vluxseg7ei16_v_bf16mf2x7_mu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vluxseg7.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP7]]
+//
+vbfloat16mf2x7_t test_vluxseg7ei16_v_bf16mf2x7_mu(vbool32_t vm,
+                                                  vbfloat16mf2x7_t vd,
+                                                  const __bf16 *rs1,
+                                                  vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vluxseg7ei16_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vluxseg7ei16_v_bf16m1x7_mu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vluxseg7.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP7]]
+//
+vbfloat16m1x7_t test_vluxseg7ei16_v_bf16m1x7_mu(vbool16_t vm,
+                                                vbfloat16m1x7_t vd,
+                                                const __bf16 *rs1,
+                                                vuint16m1_t rs2, size_t vl) {
+  return __riscv_vluxseg7ei16_mu(vm, vd, rs1, rs2, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vluxseg8ei16.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vluxseg8ei16.c
new file mode 100644
index 000000000000..57207ccad17e
--- /dev/null
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/bfloat16/vluxseg8ei16.c
@@ -0,0 +1,262 @@
+// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py UTC_ARGS: --version 4
+// REQUIRES: riscv-registered-target
+// RUN: %clang_cc1 -triple riscv64 -target-feature +v \
+// RUN:   -target-feature +experimental-zvfbfmin \
+// RUN:   -target-feature +experimental-zvfbfwma -disable-O0-optnone \
+// RUN:   -emit-llvm %s -o - | opt -S -passes=mem2reg | \
+// RUN:   FileCheck --check-prefix=CHECK-RV64 %s
+
+#include <riscv_vector.h>
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vluxseg8ei16_v_bf16mf4x8_tu(
+// CHECK-RV64-SAME: { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0:[0-9]+]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vluxseg8.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], <vscale x 1 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP8]]
+//
+vbfloat16mf4x8_t test_vluxseg8ei16_v_bf16mf4x8_tu(vbfloat16mf4x8_t vd,
+                                                  const __bf16 *rs1,
+                                                  vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vluxseg8ei16_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vluxseg8ei16_v_bf16mf2x8_tu(
+// CHECK-RV64-SAME: { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vluxseg8.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], <vscale x 2 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP8]]
+//
+vbfloat16mf2x8_t test_vluxseg8ei16_v_bf16mf2x8_tu(vbfloat16mf2x8_t vd,
+                                                  const __bf16 *rs1,
+                                                  vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vluxseg8ei16_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vluxseg8ei16_v_bf16m1x8_tu(
+// CHECK-RV64-SAME: { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vluxseg8.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], <vscale x 4 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], i64 [[VL]])
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP8]]
+//
+vbfloat16m1x8_t test_vluxseg8ei16_v_bf16m1x8_tu(vbfloat16m1x8_t vd,
+                                                const __bf16 *rs1,
+                                                vuint16m1_t rs2, size_t vl) {
+  return __riscv_vluxseg8ei16_tu(vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vluxseg8ei16_v_bf16mf4x8_tum(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vluxseg8.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], <vscale x 1 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP8]]
+//
+vbfloat16mf4x8_t test_vluxseg8ei16_v_bf16mf4x8_tum(vbool64_t vm,
+                                                   vbfloat16mf4x8_t vd,
+                                                   const __bf16 *rs1,
+                                                   vuint16mf4_t rs2,
+                                                   size_t vl) {
+  return __riscv_vluxseg8ei16_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vluxseg8ei16_v_bf16mf2x8_tum(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vluxseg8.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], <vscale x 2 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP8]]
+//
+vbfloat16mf2x8_t test_vluxseg8ei16_v_bf16mf2x8_tum(vbool32_t vm,
+                                                   vbfloat16mf2x8_t vd,
+                                                   const __bf16 *rs1,
+                                                   vuint16mf2_t rs2,
+                                                   size_t vl) {
+  return __riscv_vluxseg8ei16_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vluxseg8ei16_v_bf16m1x8_tum(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vluxseg8.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], <vscale x 4 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 2)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP8]]
+//
+vbfloat16m1x8_t test_vluxseg8ei16_v_bf16m1x8_tum(vbool16_t vm,
+                                                 vbfloat16m1x8_t vd,
+                                                 const __bf16 *rs1,
+                                                 vuint16m1_t rs2, size_t vl) {
+  return __riscv_vluxseg8ei16_tum(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vluxseg8ei16_v_bf16mf4x8_tumu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vluxseg8.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], <vscale x 1 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP8]]
+//
+vbfloat16mf4x8_t test_vluxseg8ei16_v_bf16mf4x8_tumu(vbool64_t vm,
+                                                    vbfloat16mf4x8_t vd,
+                                                    const __bf16 *rs1,
+                                                    vuint16mf4_t rs2,
+                                                    size_t vl) {
+  return __riscv_vluxseg8ei16_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vluxseg8ei16_v_bf16mf2x8_tumu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vluxseg8.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], <vscale x 2 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP8]]
+//
+vbfloat16mf2x8_t test_vluxseg8ei16_v_bf16mf2x8_tumu(vbool32_t vm,
+                                                    vbfloat16mf2x8_t vd,
+                                                    const __bf16 *rs1,
+                                                    vuint16mf2_t rs2,
+                                                    size_t vl) {
+  return __riscv_vluxseg8ei16_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vluxseg8ei16_v_bf16m1x8_tumu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vluxseg8.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], <vscale x 4 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 0)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP8]]
+//
+vbfloat16m1x8_t test_vluxseg8ei16_v_bf16m1x8_tumu(vbool16_t vm,
+                                                  vbfloat16m1x8_t vd,
+                                                  const __bf16 *rs1,
+                                                  vuint16m1_t rs2, size_t vl) {
+  return __riscv_vluxseg8ei16_tumu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @test_vluxseg8ei16_v_bf16mf4x8_mu(
+// CHECK-RV64-SAME: <vscale x 1 x i1> [[VM:%.*]], { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 1 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } @llvm.riscv.vluxseg8.mask.nxv1bf16.nxv1i16.i64(<vscale x 1 x bfloat> [[TMP0]], <vscale x 1 x bfloat> [[TMP1]], <vscale x 1 x bfloat> [[TMP2]], <vscale x 1 x bfloat> [[TMP3]], <vscale x 1 x bfloat> [[TMP4]], <vscale x 1 x bfloat> [[TMP5]], <vscale x 1 x bfloat> [[TMP6]], <vscale x 1 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 1 x i16> [[RS2]], <vscale x 1 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat>, <vscale x 1 x bfloat> } [[TMP8]]
+//
+vbfloat16mf4x8_t test_vluxseg8ei16_v_bf16mf4x8_mu(vbool64_t vm,
+                                                  vbfloat16mf4x8_t vd,
+                                                  const __bf16 *rs1,
+                                                  vuint16mf4_t rs2, size_t vl) {
+  return __riscv_vluxseg8ei16_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @test_vluxseg8ei16_v_bf16mf2x8_mu(
+// CHECK-RV64-SAME: <vscale x 2 x i1> [[VM:%.*]], { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 2 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } @llvm.riscv.vluxseg8.mask.nxv2bf16.nxv2i16.i64(<vscale x 2 x bfloat> [[TMP0]], <vscale x 2 x bfloat> [[TMP1]], <vscale x 2 x bfloat> [[TMP2]], <vscale x 2 x bfloat> [[TMP3]], <vscale x 2 x bfloat> [[TMP4]], <vscale x 2 x bfloat> [[TMP5]], <vscale x 2 x bfloat> [[TMP6]], <vscale x 2 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 2 x i16> [[RS2]], <vscale x 2 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat>, <vscale x 2 x bfloat> } [[TMP8]]
+//
+vbfloat16mf2x8_t test_vluxseg8ei16_v_bf16mf2x8_mu(vbool32_t vm,
+                                                  vbfloat16mf2x8_t vd,
+                                                  const __bf16 *rs1,
+                                                  vuint16mf2_t rs2, size_t vl) {
+  return __riscv_vluxseg8ei16_mu(vm, vd, rs1, rs2, vl);
+}
+
+// CHECK-RV64-LABEL: define dso_local { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @test_vluxseg8ei16_v_bf16m1x8_mu(
+// CHECK-RV64-SAME: <vscale x 4 x i1> [[VM:%.*]], { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD:%.*]], ptr noundef [[RS1:%.*]], <vscale x 4 x i16> [[RS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
+// CHECK-RV64-NEXT:  entry:
+// CHECK-RV64-NEXT:    [[TMP0:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 0
+// CHECK-RV64-NEXT:    [[TMP1:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 1
+// CHECK-RV64-NEXT:    [[TMP2:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 2
+// CHECK-RV64-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 3
+// CHECK-RV64-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 4
+// CHECK-RV64-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 5
+// CHECK-RV64-NEXT:    [[TMP6:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 6
+// CHECK-RV64-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[VD]], 7
+// CHECK-RV64-NEXT:    [[TMP8:%.*]] = call { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } @llvm.riscv.vluxseg8.mask.nxv4bf16.nxv4i16.i64(<vscale x 4 x bfloat> [[TMP0]], <vscale x 4 x bfloat> [[TMP1]], <vscale x 4 x bfloat> [[TMP2]], <vscale x 4 x bfloat> [[TMP3]], <vscale x 4 x bfloat> [[TMP4]], <vscale x 4 x bfloat> [[TMP5]], <vscale x 4 x bfloat> [[TMP6]], <vscale x 4 x bfloat> [[TMP7]], ptr [[RS1]], <vscale x 4 x i16> [[RS2]], <vscale x 4 x i1> [[VM]], i64 [[VL]], i64 1)
+// CHECK-RV64-NEXT:    ret { <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat>, <vscale x 4 x bfloat> } [[TMP8]]
+//
+vbfloat16m1x8_t test_vluxseg8ei16_v_bf16m1x8_mu(vbool16_t vm,
+                                                vbfloat16m1x8_t vd,
+                                                const __bf16 *rs1,
+                                                vuint16m1_t rs2, size_t vl) {
+  return __riscv_vluxseg8ei16_mu(vm, vd, rs1, rs2, vl);
+}
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/vaesdf.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/vaesdf.c
index 34fd46465c34..646854e671d7 100644
--- a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/vaesdf.c
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/vaesdf.c
@@ -206,13 +206,3 @@ vuint32m8_t test_vaesdf_vv_u32m8_tu(vuint32m8_t vd, vuint32m8_t vs2, size_t vl)
   return __riscv_vaesdf_vv_tu(vd, vs2, vl);
 }
 
-// CHECK-RV64-LABEL: define dso_local <vscale x 16 x i32> @test_vaesdf_vs_u32m8_u32m8_tu
-// CHECK-RV64-SAME: (<vscale x 16 x i32> [[VD:%.*]], <vscale x 16 x i32> [[VS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
-// CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vaesdf.vs.nxv16i32.nxv16i32.i64(<vscale x 16 x i32> [[VD]], <vscale x 16 x i32> [[VS2]], i64 [[VL]], i64 2)
-// CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
-//
-vuint32m8_t test_vaesdf_vs_u32m8_u32m8_tu(vuint32m8_t vd, vuint32m8_t vs2, size_t vl) {
-  return __riscv_vaesdf_vs_tu(vd, vs2, vl);
-}
-
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/vaesdm.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/vaesdm.c
index 2d0f8e7cafc7..90668f3f191b 100644
--- a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/vaesdm.c
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/vaesdm.c
@@ -206,13 +206,3 @@ vuint32m8_t test_vaesdm_vv_u32m8_tu(vuint32m8_t vd, vuint32m8_t vs2, size_t vl)
   return __riscv_vaesdm_vv_tu(vd, vs2, vl);
 }
 
-// CHECK-RV64-LABEL: define dso_local <vscale x 16 x i32> @test_vaesdm_vs_u32m8_u32m8_tu
-// CHECK-RV64-SAME: (<vscale x 16 x i32> [[VD:%.*]], <vscale x 16 x i32> [[VS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
-// CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vaesdm.vs.nxv16i32.nxv16i32.i64(<vscale x 16 x i32> [[VD]], <vscale x 16 x i32> [[VS2]], i64 [[VL]], i64 2)
-// CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
-//
-vuint32m8_t test_vaesdm_vs_u32m8_u32m8_tu(vuint32m8_t vd, vuint32m8_t vs2, size_t vl) {
-  return __riscv_vaesdm_vs_tu(vd, vs2, vl);
-}
-
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/vaesef.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/vaesef.c
index 26518b9a709f..8bb3e69233ad 100644
--- a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/vaesef.c
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/vaesef.c
@@ -206,13 +206,3 @@ vuint32m8_t test_vaesef_vv_u32m8_tu(vuint32m8_t vd, vuint32m8_t vs2, size_t vl)
   return __riscv_vaesef_vv_tu(vd, vs2, vl);
 }
 
-// CHECK-RV64-LABEL: define dso_local <vscale x 16 x i32> @test_vaesef_vs_u32m8_u32m8_tu
-// CHECK-RV64-SAME: (<vscale x 16 x i32> [[VD:%.*]], <vscale x 16 x i32> [[VS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
-// CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vaesef.vs.nxv16i32.nxv16i32.i64(<vscale x 16 x i32> [[VD]], <vscale x 16 x i32> [[VS2]], i64 [[VL]], i64 2)
-// CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
-//
-vuint32m8_t test_vaesef_vs_u32m8_u32m8_tu(vuint32m8_t vd, vuint32m8_t vs2, size_t vl) {
-  return __riscv_vaesef_vs_tu(vd, vs2, vl);
-}
-
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/vaesem.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/vaesem.c
index 67e13665eea8..5a83675aa0e6 100644
--- a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/vaesem.c
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/vaesem.c
@@ -206,13 +206,3 @@ vuint32m8_t test_vaesem_vv_u32m8_tu(vuint32m8_t vd, vuint32m8_t vs2, size_t vl)
   return __riscv_vaesem_vv_tu(vd, vs2, vl);
 }
 
-// CHECK-RV64-LABEL: define dso_local <vscale x 16 x i32> @test_vaesem_vs_u32m8_u32m8_tu
-// CHECK-RV64-SAME: (<vscale x 16 x i32> [[VD:%.*]], <vscale x 16 x i32> [[VS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
-// CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vaesem.vs.nxv16i32.nxv16i32.i64(<vscale x 16 x i32> [[VD]], <vscale x 16 x i32> [[VS2]], i64 [[VL]], i64 2)
-// CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
-//
-vuint32m8_t test_vaesem_vs_u32m8_u32m8_tu(vuint32m8_t vd, vuint32m8_t vs2, size_t vl) {
-  return __riscv_vaesem_vs_tu(vd, vs2, vl);
-}
-
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/vaesz.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/vaesz.c
index 71698175d7b6..afc7827a18a1 100644
--- a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/vaesz.c
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/vaesz.c
@@ -156,13 +156,3 @@ vuint32m8_t test_vaesz_vs_u32m4_u32m8_tu(vuint32m8_t vd, vuint32m4_t vs2, size_t
   return __riscv_vaesz_tu(vd, vs2, vl);
 }
 
-// CHECK-RV64-LABEL: define dso_local <vscale x 16 x i32> @test_vaesz_vs_u32m8_u32m8_tu
-// CHECK-RV64-SAME: (<vscale x 16 x i32> [[VD:%.*]], <vscale x 16 x i32> [[VS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
-// CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vaesz.vs.nxv16i32.nxv16i32.i64(<vscale x 16 x i32> [[VD]], <vscale x 16 x i32> [[VS2]], i64 [[VL]], i64 2)
-// CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
-//
-vuint32m8_t test_vaesz_vs_u32m8_u32m8_tu(vuint32m8_t vd, vuint32m8_t vs2, size_t vl) {
-  return __riscv_vaesz_tu(vd, vs2, vl);
-}
-
diff --git a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/vsm4r.c b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/vsm4r.c
index 0df390bd0724..7b70c4e2a790 100644
--- a/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/vsm4r.c
+++ b/clang/test/CodeGen/RISCV/rvv-intrinsics-autogenerated/policy/overloaded/vsm4r.c
@@ -206,13 +206,3 @@ vuint32m8_t test_vsm4r_vv_u32m8_tu(vuint32m8_t vd, vuint32m8_t vs2, size_t vl) {
   return __riscv_vsm4r_vv_tu(vd, vs2, vl);
 }
 
-// CHECK-RV64-LABEL: define dso_local <vscale x 16 x i32> @test_vsm4r_vs_u32m8_u32m8_tu
-// CHECK-RV64-SAME: (<vscale x 16 x i32> [[VD:%.*]], <vscale x 16 x i32> [[VS2:%.*]], i64 noundef [[VL:%.*]]) #[[ATTR0]] {
-// CHECK-RV64-NEXT:  entry:
-// CHECK-RV64-NEXT:    [[TMP0:%.*]] = call <vscale x 16 x i32> @llvm.riscv.vsm4r.vs.nxv16i32.nxv16i32.i64(<vscale x 16 x i32> [[VD]], <vscale x 16 x i32> [[VS2]], i64 [[VL]], i64 2)
-// CHECK-RV64-NEXT:    ret <vscale x 16 x i32> [[TMP0]]
-//
-vuint32m8_t test_vsm4r_vs_u32m8_u32m8_tu(vuint32m8_t vd, vuint32m8_t vs2, size_t vl) {
-  return __riscv_vsm4r_vs_tu(vd, vs2, vl);
-}
-
diff --git a/clang/test/CodeGen/SystemZ/builtins-systemz-zvector.c b/clang/test/CodeGen/SystemZ/builtins-systemz-zvector.c
index 48d775b88878..33f3dce73bae 100644
--- a/clang/test/CodeGen/SystemZ/builtins-systemz-zvector.c
+++ b/clang/test/CodeGen/SystemZ/builtins-systemz-zvector.c
@@ -2489,78 +2489,78 @@ void test_integer(void) {
   // CHECK-ASM: vno
 
   vuc = vec_cntlz(vsc);
-  // CHECK: call <16 x i8> @llvm.ctlz.v16i8(<16 x i8> %{{.*}}, i1 false)
+  // CHECK: call range(i8 0, 9) <16 x i8> @llvm.ctlz.v16i8(<16 x i8> %{{.*}}, i1 false)
   // CHECK-ASM: vclzb
   vuc = vec_cntlz(vuc);
-  // CHECK: call <16 x i8> @llvm.ctlz.v16i8(<16 x i8> %{{.*}}, i1 false)
+  // CHECK: call range(i8 0, 9) <16 x i8> @llvm.ctlz.v16i8(<16 x i8> %{{.*}}, i1 false)
   // CHECK-ASM: vclzb
   vus = vec_cntlz(vss);
-  // CHECK: call <8 x i16> @llvm.ctlz.v8i16(<8 x i16> %{{.*}}, i1 false)
+  // CHECK: call range(i16 0, 17) <8 x i16> @llvm.ctlz.v8i16(<8 x i16> %{{.*}}, i1 false)
   // CHECK-ASM: vclzh
   vus = vec_cntlz(vus);
-  // CHECK: call <8 x i16> @llvm.ctlz.v8i16(<8 x i16> %{{.*}}, i1 false)
+  // CHECK: call range(i16 0, 17) <8 x i16> @llvm.ctlz.v8i16(<8 x i16> %{{.*}}, i1 false)
   // CHECK-ASM: vclzh
   vui = vec_cntlz(vsi);
-  // CHECK: call <4 x i32> @llvm.ctlz.v4i32(<4 x i32> %{{.*}}, i1 false)
+  // CHECK: call range(i32 0, 33) <4 x i32> @llvm.ctlz.v4i32(<4 x i32> %{{.*}}, i1 false)
   // CHECK-ASM: vclzf
   vui = vec_cntlz(vui);
-  // CHECK: call <4 x i32> @llvm.ctlz.v4i32(<4 x i32> %{{.*}}, i1 false)
+  // CHECK: call range(i32 0, 33) <4 x i32> @llvm.ctlz.v4i32(<4 x i32> %{{.*}}, i1 false)
   // CHECK-ASM: vclzf
   vul = vec_cntlz(vsl);
-  // CHECK: call <2 x i64> @llvm.ctlz.v2i64(<2 x i64> %{{.*}}, i1 false)
+  // CHECK: call range(i64 0, 65) <2 x i64> @llvm.ctlz.v2i64(<2 x i64> %{{.*}}, i1 false)
   // CHECK-ASM: vclzg
   vul = vec_cntlz(vul);
-  // CHECK: call <2 x i64> @llvm.ctlz.v2i64(<2 x i64> %{{.*}}, i1 false)
+  // CHECK: call range(i64 0, 65) <2 x i64> @llvm.ctlz.v2i64(<2 x i64> %{{.*}}, i1 false)
   // CHECK-ASM: vclzg
 
   vuc = vec_cnttz(vsc);
-  // CHECK: call <16 x i8> @llvm.cttz.v16i8(<16 x i8> %{{.*}}, i1 false)
+  // CHECK: call range(i8 0, 9) <16 x i8> @llvm.cttz.v16i8(<16 x i8> %{{.*}}, i1 false)
   // CHECK-ASM: vctzb
   vuc = vec_cnttz(vuc);
-  // CHECK: call <16 x i8> @llvm.cttz.v16i8(<16 x i8> %{{.*}}, i1 false)
+  // CHECK: call range(i8 0, 9) <16 x i8> @llvm.cttz.v16i8(<16 x i8> %{{.*}}, i1 false)
   // CHECK-ASM: vctzb
   vus = vec_cnttz(vss);
-  // CHECK: call <8 x i16> @llvm.cttz.v8i16(<8 x i16> %{{.*}}, i1 false)
+  // CHECK: call range(i16 0, 17) <8 x i16> @llvm.cttz.v8i16(<8 x i16> %{{.*}}, i1 false)
   // CHECK-ASM: vctzh
   vus = vec_cnttz(vus);
-  // CHECK: call <8 x i16> @llvm.cttz.v8i16(<8 x i16> %{{.*}}, i1 false)
+  // CHECK: call range(i16 0, 17) <8 x i16> @llvm.cttz.v8i16(<8 x i16> %{{.*}}, i1 false)
   // CHECK-ASM: vctzh
   vui = vec_cnttz(vsi);
-  // CHECK: call <4 x i32> @llvm.cttz.v4i32(<4 x i32> %{{.*}}, i1 false)
+  // CHECK: call range(i32 0, 33) <4 x i32> @llvm.cttz.v4i32(<4 x i32> %{{.*}}, i1 false)
   // CHECK-ASM: vctzf
   vui = vec_cnttz(vui);
-  // CHECK: call <4 x i32> @llvm.cttz.v4i32(<4 x i32> %{{.*}}, i1 false)
+  // CHECK: call range(i32 0, 33) <4 x i32> @llvm.cttz.v4i32(<4 x i32> %{{.*}}, i1 false)
   // CHECK-ASM: vctzf
   vul = vec_cnttz(vsl);
-  // CHECK: call <2 x i64> @llvm.cttz.v2i64(<2 x i64> %{{.*}}, i1 false)
+  // CHECK: call range(i64 0, 65) <2 x i64> @llvm.cttz.v2i64(<2 x i64> %{{.*}}, i1 false)
   // CHECK-ASM: vctzg
   vul = vec_cnttz(vul);
-  // CHECK: call <2 x i64> @llvm.cttz.v2i64(<2 x i64> %{{.*}}, i1 false)
+  // CHECK: call range(i64 0, 65) <2 x i64> @llvm.cttz.v2i64(<2 x i64> %{{.*}}, i1 false)
   // CHECK-ASM: vctzg
 
   vuc = vec_popcnt(vsc);
-  // CHECK: call <16 x i8> @llvm.ctpop.v16i8(<16 x i8> %{{.*}})
+  // CHECK: call range(i8 0, 9) <16 x i8> @llvm.ctpop.v16i8(<16 x i8> %{{.*}})
   // CHECK-ASM: vpopct
   vuc = vec_popcnt(vuc);
-  // CHECK: call <16 x i8> @llvm.ctpop.v16i8(<16 x i8> %{{.*}})
+  // CHECK: call range(i8 0, 9) <16 x i8> @llvm.ctpop.v16i8(<16 x i8> %{{.*}})
   // CHECK-ASM: vpopct
   vus = vec_popcnt(vss);
-  // CHECK: call <8 x i16> @llvm.ctpop.v8i16(<8 x i16> %{{.*}})
+  // CHECK: call range(i16 0, 17) <8 x i16> @llvm.ctpop.v8i16(<8 x i16> %{{.*}})
   // (emulated)
   vus = vec_popcnt(vus);
-  // CHECK: call <8 x i16> @llvm.ctpop.v8i16(<8 x i16> %{{.*}})
+  // CHECK: call range(i16 0, 17) <8 x i16> @llvm.ctpop.v8i16(<8 x i16> %{{.*}})
   // (emulated)
   vui = vec_popcnt(vsi);
-  // CHECK: call <4 x i32> @llvm.ctpop.v4i32(<4 x i32> %{{.*}})
+  // CHECK: call range(i32 0, 33) <4 x i32> @llvm.ctpop.v4i32(<4 x i32> %{{.*}})
   // (emulated)
   vui = vec_popcnt(vui);
-  // CHECK: call <4 x i32> @llvm.ctpop.v4i32(<4 x i32> %{{.*}})
+  // CHECK: call range(i32 0, 33) <4 x i32> @llvm.ctpop.v4i32(<4 x i32> %{{.*}})
   // (emulated)
   vul = vec_popcnt(vsl);
-  // CHECK: call <2 x i64> @llvm.ctpop.v2i64(<2 x i64> %{{.*}})
+  // CHECK: call range(i64 0, 65) <2 x i64> @llvm.ctpop.v2i64(<2 x i64> %{{.*}})
   // (emulated)
   vul = vec_popcnt(vul);
-  // CHECK: call <2 x i64> @llvm.ctpop.v2i64(<2 x i64> %{{.*}})
+  // CHECK: call range(i64 0, 65) <2 x i64> @llvm.ctpop.v2i64(<2 x i64> %{{.*}})
   // (emulated)
 
   vsc = vec_rl(vsc, vuc);
diff --git a/clang/test/CodeGen/SystemZ/builtins-systemz-zvector2.c b/clang/test/CodeGen/SystemZ/builtins-systemz-zvector2.c
index 6c26b51c542e..15e72ecf51da 100644
--- a/clang/test/CodeGen/SystemZ/builtins-systemz-zvector2.c
+++ b/clang/test/CodeGen/SystemZ/builtins-systemz-zvector2.c
@@ -577,28 +577,28 @@ void test_integer(void) {
   // CHECK-ASM: vnx
 
   vuc = vec_popcnt(vsc);
-  // CHECK: call <16 x i8> @llvm.ctpop.v16i8(<16 x i8> %{{.*}})
+  // CHECK: call range(i8 0, 9) <16 x i8> @llvm.ctpop.v16i8(<16 x i8> %{{.*}})
   // CHECK-ASM: vpopctb
   vuc = vec_popcnt(vuc);
-  // CHECK: call <16 x i8> @llvm.ctpop.v16i8(<16 x i8> %{{.*}})
+  // CHECK: call range(i8 0, 9) <16 x i8> @llvm.ctpop.v16i8(<16 x i8> %{{.*}})
   // CHECK-ASM: vpopctb
   vus = vec_popcnt(vss);
-  // CHECK: call <8 x i16> @llvm.ctpop.v8i16(<8 x i16> %{{.*}})
+  // CHECK: call range(i16 0, 17) <8 x i16> @llvm.ctpop.v8i16(<8 x i16> %{{.*}})
   // CHECK-ASM: vpopcth
   vus = vec_popcnt(vus);
-  // CHECK: call <8 x i16> @llvm.ctpop.v8i16(<8 x i16> %{{.*}})
+  // CHECK: call range(i16 0, 17) <8 x i16> @llvm.ctpop.v8i16(<8 x i16> %{{.*}})
   // CHECK-ASM: vpopcth
   vui = vec_popcnt(vsi);
-  // CHECK: call <4 x i32> @llvm.ctpop.v4i32(<4 x i32> %{{.*}})
+  // CHECK: call range(i32 0, 33) <4 x i32> @llvm.ctpop.v4i32(<4 x i32> %{{.*}})
   // CHECK-ASM: vpopctf
   vui = vec_popcnt(vui);
-  // CHECK: call <4 x i32> @llvm.ctpop.v4i32(<4 x i32> %{{.*}})
+  // CHECK: call range(i32 0, 33) <4 x i32> @llvm.ctpop.v4i32(<4 x i32> %{{.*}})
   // CHECK-ASM: vpopctf
   vul = vec_popcnt(vsl);
-  // CHECK: call <2 x i64> @llvm.ctpop.v2i64(<2 x i64> %{{.*}})
+  // CHECK: call range(i64 0, 65) <2 x i64> @llvm.ctpop.v2i64(<2 x i64> %{{.*}})
   // CHECK-ASM: vpopctg
   vul = vec_popcnt(vul);
-  // CHECK: call <2 x i64> @llvm.ctpop.v2i64(<2 x i64> %{{.*}})
+  // CHECK: call range(i64 0, 65) <2 x i64> @llvm.ctpop.v2i64(<2 x i64> %{{.*}})
   // CHECK-ASM: vpopctg
 
   vf = vec_slb(vf, vsi);
diff --git a/clang/test/CodeGen/X86/ms-x86-intrinsics.c b/clang/test/CodeGen/X86/ms-x86-intrinsics.c
index a1c90d71c8eb..aa557c8e19a8 100644
--- a/clang/test/CodeGen/X86/ms-x86-intrinsics.c
+++ b/clang/test/CodeGen/X86/ms-x86-intrinsics.c
@@ -48,7 +48,7 @@ long long test__readfsqword(unsigned long Offset) {
 __int64 test__emul(int a, int b) {
   return __emul(a, b);
 }
-// CHECK-LABEL: define dso_local i64 @test__emul(i32 noundef %a, i32 noundef %b)
+// CHECK-LABEL: define dso_local range(i64 -4611686016279904256, 4611686018427387905) i64 @test__emul(i32 noundef %a, i32 noundef %b)
 // CHECK: [[X:%[0-9]+]] = sext i32 %a to i64
 // CHECK: [[Y:%[0-9]+]] = sext i32 %b to i64
 // CHECK: [[RES:%[0-9]+]] = mul nsw i64 [[Y]], [[X]]
@@ -57,7 +57,7 @@ __int64 test__emul(int a, int b) {
 unsigned __int64 test__emulu(unsigned int a, unsigned int b) {
   return __emulu(a, b);
 }
-// CHECK-LABEL: define dso_local i64 @test__emulu(i32 noundef %a, i32 noundef %b)
+// CHECK-LABEL: define dso_local range(i64 0, -8589934590) i64 @test__emulu(i32 noundef %a, i32 noundef %b)
 // CHECK: [[X:%[0-9]+]] = zext i32 %a to i64
 // CHECK: [[Y:%[0-9]+]] = zext i32 %b to i64
 // CHECK: [[RES:%[0-9]+]] = mul nuw i64 [[Y]], [[X]]
@@ -108,13 +108,13 @@ long long test__readgsqword(unsigned long Offset) {
 __int64 test__mulh(__int64 a, __int64 b) {
   return __mulh(a, b);
 }
-// CHECK-X64-LABEL: define dso_local i64 @test__mulh(i64 noundef %a, i64 noundef %b)
+// CHECK-X64-LABEL: define dso_local range(i64 -4611686018427387904, 4611686018427387905) i64 @test__mulh(i64 noundef %a, i64 noundef %b)
 // CHECK-X64: = mul nsw i128 %
 
 unsigned __int64 test__umulh(unsigned __int64 a, unsigned __int64 b) {
   return __umulh(a, b);
 }
-// CHECK-X64-LABEL: define dso_local i64 @test__umulh(i64 noundef %a, i64 noundef %b)
+// CHECK-X64-LABEL: define dso_local range(i64 0, -1) i64 @test__umulh(i64 noundef %a, i64 noundef %b)
 // CHECK-X64: = mul nuw i128 %
 
 __int64 test_mul128(__int64 Multiplier,
diff --git a/clang/test/CodeGen/aarch64-sve2p1-intrinsics/acle_sve2p1_fp_reduce.c b/clang/test/CodeGen/aarch64-sve2p1-intrinsics/acle_sve2p1_fp_reduce.c
index e58cf4e49a37..9d5ffdafe866 100644
--- a/clang/test/CodeGen/aarch64-sve2p1-intrinsics/acle_sve2p1_fp_reduce.c
+++ b/clang/test/CodeGen/aarch64-sve2p1-intrinsics/acle_sve2p1_fp_reduce.c
@@ -20,13 +20,13 @@
 // CHECK-LABEL: @test_svaddqv_f16(
 // CHECK-NEXT:  entry:
 // CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 8 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv8i1(<vscale x 16 x i1> [[PG:%.*]])
-// CHECK-NEXT:    [[TMP1:%.*]] = tail call <8 x half> @llvm.aarch64.sve.addqv.v8f16.nxv8f16(<vscale x 8 x i1> [[TMP0]], <vscale x 8 x half> [[OP:%.*]])
+// CHECK-NEXT:    [[TMP1:%.*]] = tail call <8 x half> @llvm.aarch64.sve.faddqv.v8f16.nxv8f16(<vscale x 8 x i1> [[TMP0]], <vscale x 8 x half> [[OP:%.*]])
 // CHECK-NEXT:    ret <8 x half> [[TMP1]]
 //
 // CPP-CHECK-LABEL: @_Z16test_svaddqv_f16u10__SVBool_tu13__SVFloat16_t(
 // CPP-CHECK-NEXT:  entry:
 // CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 8 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv8i1(<vscale x 16 x i1> [[PG:%.*]])
-// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <8 x half> @llvm.aarch64.sve.addqv.v8f16.nxv8f16(<vscale x 8 x i1> [[TMP0]], <vscale x 8 x half> [[OP:%.*]])
+// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <8 x half> @llvm.aarch64.sve.faddqv.v8f16.nxv8f16(<vscale x 8 x i1> [[TMP0]], <vscale x 8 x half> [[OP:%.*]])
 // CPP-CHECK-NEXT:    ret <8 x half> [[TMP1]]
 //
 float16x8_t test_svaddqv_f16(svbool_t pg, svfloat16_t op)
@@ -37,13 +37,13 @@ float16x8_t test_svaddqv_f16(svbool_t pg, svfloat16_t op)
 // CHECK-LABEL: @test_svaddqv_f32(
 // CHECK-NEXT:  entry:
 // CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 4 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv4i1(<vscale x 16 x i1> [[PG:%.*]])
-// CHECK-NEXT:    [[TMP1:%.*]] = tail call <4 x float> @llvm.aarch64.sve.addqv.v4f32.nxv4f32(<vscale x 4 x i1> [[TMP0]], <vscale x 4 x float> [[OP:%.*]])
+// CHECK-NEXT:    [[TMP1:%.*]] = tail call <4 x float> @llvm.aarch64.sve.faddqv.v4f32.nxv4f32(<vscale x 4 x i1> [[TMP0]], <vscale x 4 x float> [[OP:%.*]])
 // CHECK-NEXT:    ret <4 x float> [[TMP1]]
 //
 // CPP-CHECK-LABEL: @_Z16test_svaddqv_f32u10__SVBool_tu13__SVFloat32_t(
 // CPP-CHECK-NEXT:  entry:
 // CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 4 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv4i1(<vscale x 16 x i1> [[PG:%.*]])
-// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <4 x float> @llvm.aarch64.sve.addqv.v4f32.nxv4f32(<vscale x 4 x i1> [[TMP0]], <vscale x 4 x float> [[OP:%.*]])
+// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <4 x float> @llvm.aarch64.sve.faddqv.v4f32.nxv4f32(<vscale x 4 x i1> [[TMP0]], <vscale x 4 x float> [[OP:%.*]])
 // CPP-CHECK-NEXT:    ret <4 x float> [[TMP1]]
 //
 float32x4_t test_svaddqv_f32(svbool_t pg, svfloat32_t op)
@@ -54,13 +54,13 @@ float32x4_t test_svaddqv_f32(svbool_t pg, svfloat32_t op)
 // CHECK-LABEL: @test_svaddqv_f64(
 // CHECK-NEXT:  entry:
 // CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 2 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv2i1(<vscale x 16 x i1> [[PG:%.*]])
-// CHECK-NEXT:    [[TMP1:%.*]] = tail call <2 x double> @llvm.aarch64.sve.addqv.v2f64.nxv2f64(<vscale x 2 x i1> [[TMP0]], <vscale x 2 x double> [[OP:%.*]])
+// CHECK-NEXT:    [[TMP1:%.*]] = tail call <2 x double> @llvm.aarch64.sve.faddqv.v2f64.nxv2f64(<vscale x 2 x i1> [[TMP0]], <vscale x 2 x double> [[OP:%.*]])
 // CHECK-NEXT:    ret <2 x double> [[TMP1]]
 //
 // CPP-CHECK-LABEL: @_Z16test_svaddqv_f64u10__SVBool_tu13__SVFloat64_t(
 // CPP-CHECK-NEXT:  entry:
 // CPP-CHECK-NEXT:    [[TMP0:%.*]] = tail call <vscale x 2 x i1> @llvm.aarch64.sve.convert.from.svbool.nxv2i1(<vscale x 16 x i1> [[PG:%.*]])
-// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <2 x double> @llvm.aarch64.sve.addqv.v2f64.nxv2f64(<vscale x 2 x i1> [[TMP0]], <vscale x 2 x double> [[OP:%.*]])
+// CPP-CHECK-NEXT:    [[TMP1:%.*]] = tail call <2 x double> @llvm.aarch64.sve.faddqv.v2f64.nxv2f64(<vscale x 2 x i1> [[TMP0]], <vscale x 2 x double> [[OP:%.*]])
 // CPP-CHECK-NEXT:    ret <2 x double> [[TMP1]]
 //
 float64x2_t test_svaddqv_f64(svbool_t pg, svfloat64_t op)
diff --git a/clang/test/CodeGen/arm-bf16-convert-intrinsics.c b/clang/test/CodeGen/arm-bf16-convert-intrinsics.c
index f50eaf371028..0f2c5b2546fa 100644
--- a/clang/test/CodeGen/arm-bf16-convert-intrinsics.c
+++ b/clang/test/CodeGen/arm-bf16-convert-intrinsics.c
@@ -426,11 +426,12 @@ bfloat16_t test_vcvth_bf16_f32(float32_t a) {
 // CHECK-NEXT:    [[__REINT_I:%.*]] = alloca bfloat, align 2
 // CHECK-NEXT:    [[__REINT1_I:%.*]] = alloca i32, align 4
 // CHECK-NEXT:    store bfloat [[A:%.*]], ptr [[__REINT_I]], align 2
-// CHECK-NEXT:    [[TMP1:%.*]] = load i32, ptr [[__REINT_I]], align 2
-// CHECK-NEXT:    [[SHL_I:%.*]] = shl i32 [[TMP1]], 16
+// CHECK-NEXT:    [[TMP0:%.*]] = load i16, ptr [[__REINT_I]], align 2
+// CHECK-NEXT:    [[CONV_I:%.*]] = sext i16 [[TMP0]] to i32
+// CHECK-NEXT:    [[SHL_I:%.*]] = shl i32 [[CONV_I]], 16
 // CHECK-NEXT:    store i32 [[SHL_I]], ptr [[__REINT1_I]], align 4
-// CHECK-NEXT:    [[TMP3:%.*]] = load float, ptr [[__REINT1_I]], align 4
-// CHECK-NEXT:    ret float [[TMP3]]
+// CHECK-NEXT:    [[TMP1:%.*]] = load float, ptr [[__REINT1_I]], align 4
+// CHECK-NEXT:    ret float [[TMP1]]
 //
 float32_t test_vcvtah_f32_bf16(bfloat16_t a) {
   return vcvtah_f32_bf16(a);
diff --git a/clang/test/CodeGen/attr-counted-by.c b/clang/test/CodeGen/attr-counted-by.c
index 1fb39f9a3466..de30a00138ac 100644
--- a/clang/test/CodeGen/attr-counted-by.c
+++ b/clang/test/CodeGen/attr-counted-by.c
@@ -66,7 +66,7 @@ struct anon_struct {
 // SANITIZE-WITH-ATTR-NEXT:    [[TMP1:%.*]] = icmp ult i64 [[IDXPROM]], [[TMP0]], !nosanitize [[META2]]
 // SANITIZE-WITH-ATTR-NEXT:    br i1 [[TMP1]], label [[CONT3:%.*]], label [[HANDLER_OUT_OF_BOUNDS:%.*]], !prof [[PROF3:![0-9]+]], !nosanitize [[META2]]
 // SANITIZE-WITH-ATTR:       handler.out_of_bounds:
-// SANITIZE-WITH-ATTR-NEXT:    tail call void @__ubsan_handle_out_of_bounds_abort(ptr nonnull @[[GLOB2:[0-9]+]], i64 [[IDXPROM]]) #[[ATTR10:[0-9]+]], !nosanitize [[META2]]
+// SANITIZE-WITH-ATTR-NEXT:    tail call void @__ubsan_handle_out_of_bounds_abort(ptr nonnull @[[GLOB1:[0-9]+]], i64 [[IDXPROM]]) #[[ATTR10:[0-9]+]], !nosanitize [[META2]]
 // SANITIZE-WITH-ATTR-NEXT:    unreachable, !nosanitize [[META2]]
 // SANITIZE-WITH-ATTR:       cont3:
 // SANITIZE-WITH-ATTR-NEXT:    [[ARRAY:%.*]] = getelementptr inbounds i8, ptr [[P]], i64 12
@@ -114,7 +114,7 @@ void test1(struct annotated *p, int index, int val) {
 // SANITIZE-WITH-ATTR-NEXT:    [[TMP1:%.*]] = icmp ugt i64 [[TMP0]], [[INDEX]], !nosanitize [[META2]]
 // SANITIZE-WITH-ATTR-NEXT:    br i1 [[TMP1]], label [[CONT3:%.*]], label [[HANDLER_OUT_OF_BOUNDS:%.*]], !prof [[PROF3]], !nosanitize [[META2]]
 // SANITIZE-WITH-ATTR:       handler.out_of_bounds:
-// SANITIZE-WITH-ATTR-NEXT:    tail call void @__ubsan_handle_out_of_bounds_abort(ptr nonnull @[[GLOB4:[0-9]+]], i64 [[INDEX]]) #[[ATTR10]], !nosanitize [[META2]]
+// SANITIZE-WITH-ATTR-NEXT:    tail call void @__ubsan_handle_out_of_bounds_abort(ptr nonnull @[[GLOB3:[0-9]+]], i64 [[INDEX]]) #[[ATTR10]], !nosanitize [[META2]]
 // SANITIZE-WITH-ATTR-NEXT:    unreachable, !nosanitize [[META2]]
 // SANITIZE-WITH-ATTR:       cont3:
 // SANITIZE-WITH-ATTR-NEXT:    [[ARRAY:%.*]] = getelementptr inbounds i8, ptr [[P]], i64 12
@@ -158,7 +158,7 @@ void test2(struct annotated *p, size_t index) {
   p->array[index] = __builtin_dynamic_object_size(p->array, 1);
 }
 
-// SANITIZE-WITH-ATTR-LABEL: define dso_local i64 @test2_bdos(
+// SANITIZE-WITH-ATTR-LABEL: define dso_local range(i64 -8589934592, 8589934589) i64 @test2_bdos(
 // SANITIZE-WITH-ATTR-SAME: ptr nocapture noundef readonly [[P:%.*]]) local_unnamed_addr #[[ATTR2:[0-9]+]] {
 // SANITIZE-WITH-ATTR-NEXT:  entry:
 // SANITIZE-WITH-ATTR-NEXT:    [[DOT_COUNTED_BY_GEP:%.*]] = getelementptr inbounds i8, ptr [[P]], i64 8
@@ -169,7 +169,7 @@ void test2(struct annotated *p, size_t index) {
 // SANITIZE-WITH-ATTR-NEXT:    [[TMP3:%.*]] = select i1 [[TMP2]], i64 [[TMP1]], i64 0
 // SANITIZE-WITH-ATTR-NEXT:    ret i64 [[TMP3]]
 //
-// NO-SANITIZE-WITH-ATTR-LABEL: define dso_local i64 @test2_bdos(
+// NO-SANITIZE-WITH-ATTR-LABEL: define dso_local range(i64 -8589934592, 8589934589) i64 @test2_bdos(
 // NO-SANITIZE-WITH-ATTR-SAME: ptr nocapture noundef readonly [[P:%.*]]) local_unnamed_addr #[[ATTR2:[0-9]+]] {
 // NO-SANITIZE-WITH-ATTR-NEXT:  entry:
 // NO-SANITIZE-WITH-ATTR-NEXT:    [[DOT_COUNTED_BY_GEP:%.*]] = getelementptr inbounds i8, ptr [[P]], i64 8
@@ -203,7 +203,7 @@ size_t test2_bdos(struct annotated *p) {
 // SANITIZE-WITH-ATTR-NEXT:    [[TMP1:%.*]] = icmp ugt i64 [[TMP0]], [[INDEX]], !nosanitize [[META2]]
 // SANITIZE-WITH-ATTR-NEXT:    br i1 [[TMP1]], label [[CONT3:%.*]], label [[HANDLER_OUT_OF_BOUNDS:%.*]], !prof [[PROF3]], !nosanitize [[META2]]
 // SANITIZE-WITH-ATTR:       handler.out_of_bounds:
-// SANITIZE-WITH-ATTR-NEXT:    tail call void @__ubsan_handle_out_of_bounds_abort(ptr nonnull @[[GLOB5:[0-9]+]], i64 [[INDEX]]) #[[ATTR10]], !nosanitize [[META2]]
+// SANITIZE-WITH-ATTR-NEXT:    tail call void @__ubsan_handle_out_of_bounds_abort(ptr nonnull @[[GLOB4:[0-9]+]], i64 [[INDEX]]) #[[ATTR10]], !nosanitize [[META2]]
 // SANITIZE-WITH-ATTR-NEXT:    unreachable, !nosanitize [[META2]]
 // SANITIZE-WITH-ATTR:       cont3:
 // SANITIZE-WITH-ATTR-NEXT:    [[ARRAY:%.*]] = getelementptr inbounds i8, ptr [[P]], i64 12
@@ -257,7 +257,7 @@ void test3(struct annotated *p, size_t index) {
   p->array[index] = __builtin_dynamic_object_size(p, 1);
 }
 
-// SANITIZE-WITH-ATTR-LABEL: define dso_local i64 @test3_bdos(
+// SANITIZE-WITH-ATTR-LABEL: define dso_local range(i64 0, 8589934601) i64 @test3_bdos(
 // SANITIZE-WITH-ATTR-SAME: ptr nocapture noundef readonly [[P:%.*]]) local_unnamed_addr #[[ATTR2]] {
 // SANITIZE-WITH-ATTR-NEXT:  entry:
 // SANITIZE-WITH-ATTR-NEXT:    [[DOT_COUNTED_BY_GEP:%.*]] = getelementptr inbounds i8, ptr [[P]], i64 8
@@ -270,7 +270,7 @@ void test3(struct annotated *p, size_t index) {
 // SANITIZE-WITH-ATTR-NEXT:    [[TMP5:%.*]] = select i1 [[TMP4]], i64 [[TMP3]], i64 0
 // SANITIZE-WITH-ATTR-NEXT:    ret i64 [[TMP5]]
 //
-// NO-SANITIZE-WITH-ATTR-LABEL: define dso_local i64 @test3_bdos(
+// NO-SANITIZE-WITH-ATTR-LABEL: define dso_local range(i64 0, 8589934601) i64 @test3_bdos(
 // NO-SANITIZE-WITH-ATTR-SAME: ptr nocapture noundef readonly [[P:%.*]]) local_unnamed_addr #[[ATTR2]] {
 // NO-SANITIZE-WITH-ATTR-NEXT:  entry:
 // NO-SANITIZE-WITH-ATTR-NEXT:    [[DOT_COUNTED_BY_GEP:%.*]] = getelementptr inbounds i8, ptr [[P]], i64 8
@@ -308,7 +308,7 @@ size_t test3_bdos(struct annotated *p) {
 // SANITIZE-WITH-ATTR-NEXT:    [[TMP1:%.*]] = icmp ult i64 [[IDXPROM]], [[TMP0]], !nosanitize [[META2]]
 // SANITIZE-WITH-ATTR-NEXT:    br i1 [[TMP1]], label [[CONT4:%.*]], label [[HANDLER_OUT_OF_BOUNDS:%.*]], !prof [[PROF3]], !nosanitize [[META2]]
 // SANITIZE-WITH-ATTR:       handler.out_of_bounds:
-// SANITIZE-WITH-ATTR-NEXT:    tail call void @__ubsan_handle_out_of_bounds_abort(ptr nonnull @[[GLOB6:[0-9]+]], i64 [[IDXPROM]]) #[[ATTR10]], !nosanitize [[META2]]
+// SANITIZE-WITH-ATTR-NEXT:    tail call void @__ubsan_handle_out_of_bounds_abort(ptr nonnull @[[GLOB5:[0-9]+]], i64 [[IDXPROM]]) #[[ATTR10]], !nosanitize [[META2]]
 // SANITIZE-WITH-ATTR-NEXT:    unreachable, !nosanitize [[META2]]
 // SANITIZE-WITH-ATTR:       cont4:
 // SANITIZE-WITH-ATTR-NEXT:    [[TMP2:%.*]] = icmp sgt i32 [[DOT_COUNTED_BY_LOAD]], 2
@@ -325,7 +325,7 @@ size_t test3_bdos(struct annotated *p) {
 // SANITIZE-WITH-ATTR-NEXT:    [[TMP7:%.*]] = icmp ult i64 [[IDXPROM13]], [[TMP6]], !nosanitize [[META2]]
 // SANITIZE-WITH-ATTR-NEXT:    br i1 [[TMP7]], label [[CONT20:%.*]], label [[HANDLER_OUT_OF_BOUNDS16:%.*]], !prof [[PROF3]], !nosanitize [[META2]]
 // SANITIZE-WITH-ATTR:       handler.out_of_bounds16:
-// SANITIZE-WITH-ATTR-NEXT:    tail call void @__ubsan_handle_out_of_bounds_abort(ptr nonnull @[[GLOB7:[0-9]+]], i64 [[IDXPROM13]]) #[[ATTR10]], !nosanitize [[META2]]
+// SANITIZE-WITH-ATTR-NEXT:    tail call void @__ubsan_handle_out_of_bounds_abort(ptr nonnull @[[GLOB6:[0-9]+]], i64 [[IDXPROM13]]) #[[ATTR10]], !nosanitize [[META2]]
 // SANITIZE-WITH-ATTR-NEXT:    unreachable, !nosanitize [[META2]]
 // SANITIZE-WITH-ATTR:       cont20:
 // SANITIZE-WITH-ATTR-NEXT:    [[TMP8:%.*]] = icmp sgt i32 [[DOT_COUNTED_BY_LOAD7]], 3
@@ -342,7 +342,7 @@ size_t test3_bdos(struct annotated *p) {
 // SANITIZE-WITH-ATTR-NEXT:    [[TMP13:%.*]] = icmp ult i64 [[IDXPROM30]], [[TMP12]], !nosanitize [[META2]]
 // SANITIZE-WITH-ATTR-NEXT:    br i1 [[TMP13]], label [[CONT37:%.*]], label [[HANDLER_OUT_OF_BOUNDS33:%.*]], !prof [[PROF3]], !nosanitize [[META2]]
 // SANITIZE-WITH-ATTR:       handler.out_of_bounds33:
-// SANITIZE-WITH-ATTR-NEXT:    tail call void @__ubsan_handle_out_of_bounds_abort(ptr nonnull @[[GLOB8:[0-9]+]], i64 [[IDXPROM30]]) #[[ATTR10]], !nosanitize [[META2]]
+// SANITIZE-WITH-ATTR-NEXT:    tail call void @__ubsan_handle_out_of_bounds_abort(ptr nonnull @[[GLOB7:[0-9]+]], i64 [[IDXPROM30]]) #[[ATTR10]], !nosanitize [[META2]]
 // SANITIZE-WITH-ATTR-NEXT:    unreachable, !nosanitize [[META2]]
 // SANITIZE-WITH-ATTR:       cont37:
 // SANITIZE-WITH-ATTR-NEXT:    [[ARRAYIDX35:%.*]] = getelementptr inbounds [0 x i32], ptr [[ARRAY]], i64 0, i64 [[IDXPROM30]]
@@ -441,7 +441,7 @@ void test4(struct annotated *p, int index, int fam_idx) {
   p->array[index + 2] = (unsigned char)__builtin_dynamic_object_size(&(p->array[fam_idx]), 1);
 }
 
-// SANITIZE-WITH-ATTR-LABEL: define dso_local i64 @test4_bdos(
+// SANITIZE-WITH-ATTR-LABEL: define dso_local range(i64 -17179869180, 17179869181) i64 @test4_bdos(
 // SANITIZE-WITH-ATTR-SAME: ptr nocapture noundef readonly [[P:%.*]], i32 noundef [[INDEX:%.*]]) local_unnamed_addr #[[ATTR2]] {
 // SANITIZE-WITH-ATTR-NEXT:  entry:
 // SANITIZE-WITH-ATTR-NEXT:    [[DOT_COUNTED_BY_GEP:%.*]] = getelementptr inbounds i8, ptr [[P]], i64 8
@@ -456,7 +456,7 @@ void test4(struct annotated *p, int index, int fam_idx) {
 // SANITIZE-WITH-ATTR-NEXT:    [[TMP7:%.*]] = select i1 [[TMP6]], i64 [[TMP3]], i64 0
 // SANITIZE-WITH-ATTR-NEXT:    ret i64 [[TMP7]]
 //
-// NO-SANITIZE-WITH-ATTR-LABEL: define dso_local i64 @test4_bdos(
+// NO-SANITIZE-WITH-ATTR-LABEL: define dso_local range(i64 -17179869180, 17179869181) i64 @test4_bdos(
 // NO-SANITIZE-WITH-ATTR-SAME: ptr nocapture noundef readonly [[P:%.*]], i32 noundef [[INDEX:%.*]]) local_unnamed_addr #[[ATTR2]] {
 // NO-SANITIZE-WITH-ATTR-NEXT:  entry:
 // NO-SANITIZE-WITH-ATTR-NEXT:    [[DOT_COUNTED_BY_GEP:%.*]] = getelementptr inbounds i8, ptr [[P]], i64 8
@@ -494,7 +494,7 @@ size_t test4_bdos(struct annotated *p, int index) {
 // SANITIZE-WITH-ATTR-NEXT:    [[TMP0:%.*]] = icmp ugt i64 [[DOT_COUNTED_BY_LOAD]], [[IDXPROM]], !nosanitize [[META2]]
 // SANITIZE-WITH-ATTR-NEXT:    br i1 [[TMP0]], label [[CONT3:%.*]], label [[HANDLER_OUT_OF_BOUNDS:%.*]], !prof [[PROF3]], !nosanitize [[META2]]
 // SANITIZE-WITH-ATTR:       handler.out_of_bounds:
-// SANITIZE-WITH-ATTR-NEXT:    tail call void @__ubsan_handle_out_of_bounds_abort(ptr nonnull @[[GLOB9:[0-9]+]], i64 [[IDXPROM]]) #[[ATTR10]], !nosanitize [[META2]]
+// SANITIZE-WITH-ATTR-NEXT:    tail call void @__ubsan_handle_out_of_bounds_abort(ptr nonnull @[[GLOB8:[0-9]+]], i64 [[IDXPROM]]) #[[ATTR10]], !nosanitize [[META2]]
 // SANITIZE-WITH-ATTR-NEXT:    unreachable, !nosanitize [[META2]]
 // SANITIZE-WITH-ATTR:       cont3:
 // SANITIZE-WITH-ATTR-NEXT:    [[TMP1:%.*]] = getelementptr inbounds i8, ptr [[P]], i64 16
@@ -545,7 +545,7 @@ void test5(struct anon_struct *p, int index) {
   p->array[index] = __builtin_dynamic_object_size(p, 1);
 }
 
-// SANITIZE-WITH-ATTR-LABEL: define dso_local i64 @test5_bdos(
+// SANITIZE-WITH-ATTR-LABEL: define dso_local range(i64 16, 1) i64 @test5_bdos(
 // SANITIZE-WITH-ATTR-SAME: ptr nocapture noundef readonly [[P:%.*]]) local_unnamed_addr #[[ATTR2]] {
 // SANITIZE-WITH-ATTR-NEXT:  entry:
 // SANITIZE-WITH-ATTR-NEXT:    [[DOT_COUNTED_BY_GEP:%.*]] = getelementptr inbounds i8, ptr [[P]], i64 8
@@ -556,7 +556,7 @@ void test5(struct anon_struct *p, int index) {
 // SANITIZE-WITH-ATTR-NEXT:    [[TMP2:%.*]] = select i1 [[DOTINV]], i64 0, i64 [[TMP1]]
 // SANITIZE-WITH-ATTR-NEXT:    ret i64 [[TMP2]]
 //
-// NO-SANITIZE-WITH-ATTR-LABEL: define dso_local i64 @test5_bdos(
+// NO-SANITIZE-WITH-ATTR-LABEL: define dso_local range(i64 16, 1) i64 @test5_bdos(
 // NO-SANITIZE-WITH-ATTR-SAME: ptr nocapture noundef readonly [[P:%.*]]) local_unnamed_addr #[[ATTR2]] {
 // NO-SANITIZE-WITH-ATTR-NEXT:  entry:
 // NO-SANITIZE-WITH-ATTR-NEXT:    [[DOT_COUNTED_BY_GEP:%.*]] = getelementptr inbounds i8, ptr [[P]], i64 8
@@ -590,7 +590,7 @@ size_t test5_bdos(struct anon_struct *p) {
 // SANITIZE-WITH-ATTR-NEXT:    [[TMP0:%.*]] = icmp ugt i64 [[DOT_COUNTED_BY_LOAD]], [[IDXPROM]], !nosanitize [[META2]]
 // SANITIZE-WITH-ATTR-NEXT:    br i1 [[TMP0]], label [[CONT3:%.*]], label [[HANDLER_OUT_OF_BOUNDS:%.*]], !prof [[PROF3]], !nosanitize [[META2]]
 // SANITIZE-WITH-ATTR:       handler.out_of_bounds:
-// SANITIZE-WITH-ATTR-NEXT:    tail call void @__ubsan_handle_out_of_bounds_abort(ptr nonnull @[[GLOB10:[0-9]+]], i64 [[IDXPROM]]) #[[ATTR10]], !nosanitize [[META2]]
+// SANITIZE-WITH-ATTR-NEXT:    tail call void @__ubsan_handle_out_of_bounds_abort(ptr nonnull @[[GLOB9:[0-9]+]], i64 [[IDXPROM]]) #[[ATTR10]], !nosanitize [[META2]]
 // SANITIZE-WITH-ATTR-NEXT:    unreachable, !nosanitize [[META2]]
 // SANITIZE-WITH-ATTR:       cont3:
 // SANITIZE-WITH-ATTR-NEXT:    [[TMP1:%.*]] = getelementptr inbounds i8, ptr [[P]], i64 16
@@ -683,7 +683,7 @@ size_t test6_bdos(struct anon_struct *p) {
 // SANITIZE-WITH-ATTR-NEXT:    [[TMP2:%.*]] = icmp ult i64 [[IDXPROM]], [[TMP1]], !nosanitize [[META2]]
 // SANITIZE-WITH-ATTR-NEXT:    br i1 [[TMP2]], label [[CONT7:%.*]], label [[HANDLER_OUT_OF_BOUNDS:%.*]], !prof [[PROF3]], !nosanitize [[META2]]
 // SANITIZE-WITH-ATTR:       handler.out_of_bounds:
-// SANITIZE-WITH-ATTR-NEXT:    tail call void @__ubsan_handle_out_of_bounds_abort(ptr nonnull @[[GLOB12:[0-9]+]], i64 [[IDXPROM]]) #[[ATTR10]], !nosanitize [[META2]]
+// SANITIZE-WITH-ATTR-NEXT:    tail call void @__ubsan_handle_out_of_bounds_abort(ptr nonnull @[[GLOB11:[0-9]+]], i64 [[IDXPROM]]) #[[ATTR10]], !nosanitize [[META2]]
 // SANITIZE-WITH-ATTR-NEXT:    unreachable, !nosanitize [[META2]]
 // SANITIZE-WITH-ATTR:       cont7:
 // SANITIZE-WITH-ATTR-NEXT:    [[INTS:%.*]] = getelementptr inbounds i8, ptr [[P]], i64 9
@@ -756,7 +756,7 @@ size_t test7_bdos(struct union_of_fams *p) {
 // SANITIZE-WITH-ATTR-NEXT:    [[TMP1:%.*]] = icmp ult i64 [[IDXPROM]], [[TMP0]], !nosanitize [[META2]]
 // SANITIZE-WITH-ATTR-NEXT:    br i1 [[TMP1]], label [[CONT9:%.*]], label [[HANDLER_OUT_OF_BOUNDS:%.*]], !prof [[PROF3]], !nosanitize [[META2]]
 // SANITIZE-WITH-ATTR:       handler.out_of_bounds:
-// SANITIZE-WITH-ATTR-NEXT:    tail call void @__ubsan_handle_out_of_bounds_abort(ptr nonnull @[[GLOB13:[0-9]+]], i64 [[IDXPROM]]) #[[ATTR10]], !nosanitize [[META2]]
+// SANITIZE-WITH-ATTR-NEXT:    tail call void @__ubsan_handle_out_of_bounds_abort(ptr nonnull @[[GLOB12:[0-9]+]], i64 [[IDXPROM]]) #[[ATTR10]], !nosanitize [[META2]]
 // SANITIZE-WITH-ATTR-NEXT:    unreachable, !nosanitize [[META2]]
 // SANITIZE-WITH-ATTR:       cont9:
 // SANITIZE-WITH-ATTR-NEXT:    [[INTS:%.*]] = getelementptr inbounds i8, ptr [[P]], i64 9
@@ -797,7 +797,7 @@ void test8(struct union_of_fams *p, int index) {
   p->ints[index] = __builtin_dynamic_object_size(p->ints, 1);
 }
 
-// SANITIZE-WITH-ATTR-LABEL: define dso_local i64 @test8_bdos(
+// SANITIZE-WITH-ATTR-LABEL: define dso_local range(i64 0, 256) i64 @test8_bdos(
 // SANITIZE-WITH-ATTR-SAME: ptr nocapture noundef readonly [[P:%.*]]) local_unnamed_addr #[[ATTR2]] {
 // SANITIZE-WITH-ATTR-NEXT:  entry:
 // SANITIZE-WITH-ATTR-NEXT:    [[DOT_COUNTED_BY_GEP:%.*]] = getelementptr inbounds i8, ptr [[P]], i64 8
@@ -805,7 +805,7 @@ void test8(struct union_of_fams *p, int index) {
 // SANITIZE-WITH-ATTR-NEXT:    [[TMP0:%.*]] = zext i8 [[DOT_COUNTED_BY_LOAD]] to i64
 // SANITIZE-WITH-ATTR-NEXT:    ret i64 [[TMP0]]
 //
-// NO-SANITIZE-WITH-ATTR-LABEL: define dso_local i64 @test8_bdos(
+// NO-SANITIZE-WITH-ATTR-LABEL: define dso_local range(i64 0, 256) i64 @test8_bdos(
 // NO-SANITIZE-WITH-ATTR-SAME: ptr nocapture noundef readonly [[P:%.*]]) local_unnamed_addr #[[ATTR2]] {
 // NO-SANITIZE-WITH-ATTR-NEXT:  entry:
 // NO-SANITIZE-WITH-ATTR-NEXT:    [[DOT_COUNTED_BY_GEP:%.*]] = getelementptr inbounds i8, ptr [[P]], i64 8
@@ -955,7 +955,7 @@ void test10(struct union_of_fams *p, int index) {
   p->bytes[index] = (unsigned char)__builtin_dynamic_object_size(p->bytes, 1);
 }
 
-// SANITIZE-WITH-ATTR-LABEL: define dso_local i64 @test10_bdos(
+// SANITIZE-WITH-ATTR-LABEL: define dso_local range(i64 -2147483648, 2147483648) i64 @test10_bdos(
 // SANITIZE-WITH-ATTR-SAME: ptr nocapture noundef readonly [[P:%.*]]) local_unnamed_addr #[[ATTR2]] {
 // SANITIZE-WITH-ATTR-NEXT:  entry:
 // SANITIZE-WITH-ATTR-NEXT:    [[DOT_COUNTED_BY_GEP:%.*]] = getelementptr inbounds i8, ptr [[P]], i64 8
@@ -964,7 +964,7 @@ void test10(struct union_of_fams *p, int index) {
 // SANITIZE-WITH-ATTR-NEXT:    [[TMP0:%.*]] = zext nneg i32 [[NARROW]] to i64
 // SANITIZE-WITH-ATTR-NEXT:    ret i64 [[TMP0]]
 //
-// NO-SANITIZE-WITH-ATTR-LABEL: define dso_local i64 @test10_bdos(
+// NO-SANITIZE-WITH-ATTR-LABEL: define dso_local range(i64 -2147483648, 2147483648) i64 @test10_bdos(
 // NO-SANITIZE-WITH-ATTR-SAME: ptr nocapture noundef readonly [[P:%.*]]) local_unnamed_addr #[[ATTR2]] {
 // NO-SANITIZE-WITH-ATTR-NEXT:  entry:
 // NO-SANITIZE-WITH-ATTR-NEXT:    [[DOT_COUNTED_BY_GEP:%.*]] = getelementptr inbounds i8, ptr [[P]], i64 8
@@ -1095,10 +1095,10 @@ int test12_a, test12_b;
 // SANITIZE-WITH-ATTR-NEXT:    [[DOTNOT:%.*]] = icmp eq i32 [[DOTCOUNTED_BY_LOAD]], 0
 // SANITIZE-WITH-ATTR-NEXT:    br i1 [[DOTNOT]], label [[HANDLER_OUT_OF_BOUNDS4:%.*]], label [[HANDLER_TYPE_MISMATCH6:%.*]], !prof [[PROF10:![0-9]+]], !nosanitize [[META2]]
 // SANITIZE-WITH-ATTR:       handler.out_of_bounds4:
-// SANITIZE-WITH-ATTR-NEXT:    tail call void @__ubsan_handle_out_of_bounds_abort(ptr nonnull @[[GLOB19:[0-9]+]], i64 0) #[[ATTR10]], !nosanitize [[META2]]
+// SANITIZE-WITH-ATTR-NEXT:    tail call void @__ubsan_handle_out_of_bounds_abort(ptr nonnull @[[GLOB20:[0-9]+]], i64 0) #[[ATTR10]], !nosanitize [[META2]]
 // SANITIZE-WITH-ATTR-NEXT:    unreachable, !nosanitize [[META2]]
 // SANITIZE-WITH-ATTR:       handler.type_mismatch6:
-// SANITIZE-WITH-ATTR-NEXT:    tail call void @__ubsan_handle_type_mismatch_v1_abort(ptr nonnull @[[GLOB20:[0-9]+]], i64 ptrtoint (ptr getelementptr inbounds ([[STRUCT_ANON_5:%.*]], ptr @test12_foo, i64 1, i32 0, i32 0, i32 0) to i64)) #[[ATTR10]], !nosanitize [[META2]]
+// SANITIZE-WITH-ATTR-NEXT:    tail call void @__ubsan_handle_type_mismatch_v1_abort(ptr nonnull @[[GLOB21:[0-9]+]], i64 ptrtoint (ptr getelementptr inbounds ([[STRUCT_ANON_5:%.*]], ptr @test12_foo, i64 1, i32 0, i32 0, i32 0) to i64)) #[[ATTR10]], !nosanitize [[META2]]
 // SANITIZE-WITH-ATTR-NEXT:    unreachable, !nosanitize [[META2]]
 //
 // NO-SANITIZE-WITH-ATTR-LABEL: define dso_local noundef i32 @test12(
@@ -1188,7 +1188,7 @@ struct test13_bar {
 // SANITIZE-WITH-ATTR-NEXT:    [[TMP2:%.*]] = icmp ugt i64 [[TMP1]], [[INDEX]], !nosanitize [[META2]]
 // SANITIZE-WITH-ATTR-NEXT:    br i1 [[TMP2]], label [[CONT5:%.*]], label [[HANDLER_OUT_OF_BOUNDS:%.*]], !prof [[PROF3]], !nosanitize [[META2]]
 // SANITIZE-WITH-ATTR:       handler.out_of_bounds:
-// SANITIZE-WITH-ATTR-NEXT:    tail call void @__ubsan_handle_out_of_bounds_abort(ptr nonnull @[[GLOB23:[0-9]+]], i64 [[INDEX]]) #[[ATTR10]], !nosanitize [[META2]]
+// SANITIZE-WITH-ATTR-NEXT:    tail call void @__ubsan_handle_out_of_bounds_abort(ptr nonnull @[[GLOB24:[0-9]+]], i64 [[INDEX]]) #[[ATTR10]], !nosanitize [[META2]]
 // SANITIZE-WITH-ATTR-NEXT:    unreachable, !nosanitize [[META2]]
 // SANITIZE-WITH-ATTR:       cont5:
 // SANITIZE-WITH-ATTR-NEXT:    [[REVMAP:%.*]] = getelementptr inbounds i8, ptr [[TMP0]], i64 16
@@ -1249,7 +1249,7 @@ struct test14_foo {
 // SANITIZE-WITH-ATTR-NEXT:    br i1 [[TMP0]], label [[TRAP:%.*]], label [[HANDLER_OUT_OF_BOUNDS:%.*]], !prof [[PROF3]], !nosanitize [[META2]]
 // SANITIZE-WITH-ATTR:       handler.out_of_bounds:
 // SANITIZE-WITH-ATTR-NEXT:    [[IDXPROM:%.*]] = sext i32 [[IDX]] to i64
-// SANITIZE-WITH-ATTR-NEXT:    tail call void @__ubsan_handle_out_of_bounds_abort(ptr nonnull @[[GLOB24:[0-9]+]], i64 [[IDXPROM]]) #[[ATTR10]], !nosanitize [[META2]]
+// SANITIZE-WITH-ATTR-NEXT:    tail call void @__ubsan_handle_out_of_bounds_abort(ptr nonnull @[[GLOB25:[0-9]+]], i64 [[IDXPROM]]) #[[ATTR10]], !nosanitize [[META2]]
 // SANITIZE-WITH-ATTR-NEXT:    unreachable, !nosanitize [[META2]]
 // SANITIZE-WITH-ATTR:       trap:
 // SANITIZE-WITH-ATTR-NEXT:    tail call void @llvm.trap() #[[ATTR10]]
@@ -1305,7 +1305,7 @@ int test14(int idx) {
 // SANITIZE-WITH-ATTR-NEXT:    br i1 [[TMP0]], label [[TRAP:%.*]], label [[HANDLER_OUT_OF_BOUNDS:%.*]], !prof [[PROF3]], !nosanitize [[META2]]
 // SANITIZE-WITH-ATTR:       handler.out_of_bounds:
 // SANITIZE-WITH-ATTR-NEXT:    [[IDXPROM:%.*]] = sext i32 [[IDX]] to i64
-// SANITIZE-WITH-ATTR-NEXT:    tail call void @__ubsan_handle_out_of_bounds_abort(ptr nonnull @[[GLOB25:[0-9]+]], i64 [[IDXPROM]]) #[[ATTR10]], !nosanitize [[META2]]
+// SANITIZE-WITH-ATTR-NEXT:    tail call void @__ubsan_handle_out_of_bounds_abort(ptr nonnull @[[GLOB27:[0-9]+]], i64 [[IDXPROM]]) #[[ATTR10]], !nosanitize [[META2]]
 // SANITIZE-WITH-ATTR-NEXT:    unreachable, !nosanitize [[META2]]
 // SANITIZE-WITH-ATTR:       trap:
 // SANITIZE-WITH-ATTR-NEXT:    tail call void @llvm.trap() #[[ATTR10]]
@@ -1326,7 +1326,7 @@ int test14(int idx) {
 // SANITIZE-WITHOUT-ATTR-NEXT:    br i1 [[TMP0]], label [[TRAP:%.*]], label [[HANDLER_OUT_OF_BOUNDS:%.*]], !prof [[PROF8]], !nosanitize [[META9]]
 // SANITIZE-WITHOUT-ATTR:       handler.out_of_bounds:
 // SANITIZE-WITHOUT-ATTR-NEXT:    [[IDXPROM:%.*]] = sext i32 [[IDX]] to i64
-// SANITIZE-WITHOUT-ATTR-NEXT:    tail call void @__ubsan_handle_out_of_bounds_abort(ptr nonnull @[[GLOB10:[0-9]+]], i64 [[IDXPROM]]) #[[ATTR8]], !nosanitize [[META9]]
+// SANITIZE-WITHOUT-ATTR-NEXT:    tail call void @__ubsan_handle_out_of_bounds_abort(ptr nonnull @[[GLOB11:[0-9]+]], i64 [[IDXPROM]]) #[[ATTR8]], !nosanitize [[META9]]
 // SANITIZE-WITHOUT-ATTR-NEXT:    unreachable, !nosanitize [[META9]]
 // SANITIZE-WITHOUT-ATTR:       trap:
 // SANITIZE-WITHOUT-ATTR-NEXT:    tail call void @llvm.trap() #[[ATTR8]]
@@ -1487,7 +1487,7 @@ struct tests_foo {
 // SANITIZE-WITH-ATTR-NEXT:    [[TMP0:%.*]] = icmp ugt i32 [[DOTCOUNTED_BY_LOAD]], 10
 // SANITIZE-WITH-ATTR-NEXT:    br i1 [[TMP0]], label [[CONT4:%.*]], label [[HANDLER_OUT_OF_BOUNDS:%.*]], !prof [[PROF3]], !nosanitize [[META2]]
 // SANITIZE-WITH-ATTR:       handler.out_of_bounds:
-// SANITIZE-WITH-ATTR-NEXT:    tail call void @__ubsan_handle_out_of_bounds_abort(ptr nonnull @[[GLOB26:[0-9]+]], i64 10) #[[ATTR10]], !nosanitize [[META2]]
+// SANITIZE-WITH-ATTR-NEXT:    tail call void @__ubsan_handle_out_of_bounds_abort(ptr nonnull @[[GLOB28:[0-9]+]], i64 10) #[[ATTR10]], !nosanitize [[META2]]
 // SANITIZE-WITH-ATTR-NEXT:    unreachable, !nosanitize [[META2]]
 // SANITIZE-WITH-ATTR:       cont4:
 // SANITIZE-WITH-ATTR-NEXT:    [[ARRAYIDX2:%.*]] = getelementptr inbounds i8, ptr [[VAR]], i64 84
@@ -1528,7 +1528,7 @@ int test24(int c, struct tests_foo *var) {
 // SANITIZE-WITH-ATTR-NEXT:    [[TMP1:%.*]] = icmp ugt i32 [[DOTCOUNTED_BY_LOAD]], 10
 // SANITIZE-WITH-ATTR-NEXT:    br i1 [[TMP1]], label [[CONT5:%.*]], label [[HANDLER_OUT_OF_BOUNDS:%.*]], !prof [[PROF3]], !nosanitize [[META2]]
 // SANITIZE-WITH-ATTR:       handler.out_of_bounds:
-// SANITIZE-WITH-ATTR-NEXT:    tail call void @__ubsan_handle_out_of_bounds_abort(ptr nonnull @[[GLOB27:[0-9]+]], i64 10) #[[ATTR10]], !nosanitize [[META2]]
+// SANITIZE-WITH-ATTR-NEXT:    tail call void @__ubsan_handle_out_of_bounds_abort(ptr nonnull @[[GLOB29:[0-9]+]], i64 10) #[[ATTR10]], !nosanitize [[META2]]
 // SANITIZE-WITH-ATTR-NEXT:    unreachable, !nosanitize [[META2]]
 // SANITIZE-WITH-ATTR:       cont5:
 // SANITIZE-WITH-ATTR-NEXT:    [[ARRAYIDX:%.*]] = getelementptr inbounds i8, ptr [[TMP0]], i64 44
@@ -1580,7 +1580,7 @@ struct test26_foo {
 // SANITIZE-WITH-ATTR-NEXT:    [[TMP1:%.*]] = icmp ult i64 [[IDXPROM]], [[TMP0]], !nosanitize [[META2]]
 // SANITIZE-WITH-ATTR-NEXT:    br i1 [[TMP1]], label [[CONT5:%.*]], label [[HANDLER_OUT_OF_BOUNDS:%.*]], !prof [[PROF3]], !nosanitize [[META2]]
 // SANITIZE-WITH-ATTR:       handler.out_of_bounds:
-// SANITIZE-WITH-ATTR-NEXT:    tail call void @__ubsan_handle_out_of_bounds_abort(ptr nonnull @[[GLOB28:[0-9]+]], i64 [[IDXPROM]]) #[[ATTR10]], !nosanitize [[META2]]
+// SANITIZE-WITH-ATTR-NEXT:    tail call void @__ubsan_handle_out_of_bounds_abort(ptr nonnull @[[GLOB30:[0-9]+]], i64 [[IDXPROM]]) #[[ATTR10]], !nosanitize [[META2]]
 // SANITIZE-WITH-ATTR-NEXT:    unreachable, !nosanitize [[META2]]
 // SANITIZE-WITH-ATTR:       cont5:
 // SANITIZE-WITH-ATTR-NEXT:    [[ARR:%.*]] = getelementptr inbounds i8, ptr [[FOO]], i64 8
@@ -1651,7 +1651,7 @@ struct test27_foo {
 // SANITIZE-WITH-ATTR-NEXT:    [[TMP1:%.*]] = icmp ult i64 [[IDXPROM]], [[TMP0]], !nosanitize [[META2]]
 // SANITIZE-WITH-ATTR-NEXT:    br i1 [[TMP1]], label [[CONT3:%.*]], label [[HANDLER_OUT_OF_BOUNDS:%.*]], !prof [[PROF3]], !nosanitize [[META2]]
 // SANITIZE-WITH-ATTR:       handler.out_of_bounds:
-// SANITIZE-WITH-ATTR-NEXT:    tail call void @__ubsan_handle_out_of_bounds_abort(ptr nonnull @[[GLOB30:[0-9]+]], i64 [[IDXPROM]]) #[[ATTR10]], !nosanitize [[META2]]
+// SANITIZE-WITH-ATTR-NEXT:    tail call void @__ubsan_handle_out_of_bounds_abort(ptr nonnull @[[GLOB32:[0-9]+]], i64 [[IDXPROM]]) #[[ATTR10]], !nosanitize [[META2]]
 // SANITIZE-WITH-ATTR-NEXT:    unreachable, !nosanitize [[META2]]
 // SANITIZE-WITH-ATTR:       cont3:
 // SANITIZE-WITH-ATTR-NEXT:    [[ENTRIES:%.*]] = getelementptr inbounds i8, ptr [[P]], i64 24
@@ -1717,7 +1717,7 @@ struct test28_foo {
 // SANITIZE-WITH-ATTR-NEXT:    [[TMP4:%.*]] = icmp ult i64 [[IDXPROM]], [[TMP3]], !nosanitize [[META2]]
 // SANITIZE-WITH-ATTR-NEXT:    br i1 [[TMP4]], label [[CONT17:%.*]], label [[HANDLER_OUT_OF_BOUNDS:%.*]], !prof [[PROF3]], !nosanitize [[META2]]
 // SANITIZE-WITH-ATTR:       handler.out_of_bounds:
-// SANITIZE-WITH-ATTR-NEXT:    tail call void @__ubsan_handle_out_of_bounds_abort(ptr nonnull @[[GLOB31:[0-9]+]], i64 [[IDXPROM]]) #[[ATTR10]], !nosanitize [[META2]]
+// SANITIZE-WITH-ATTR-NEXT:    tail call void @__ubsan_handle_out_of_bounds_abort(ptr nonnull @[[GLOB34:[0-9]+]], i64 [[IDXPROM]]) #[[ATTR10]], !nosanitize [[META2]]
 // SANITIZE-WITH-ATTR-NEXT:    unreachable, !nosanitize [[META2]]
 // SANITIZE-WITH-ATTR:       cont17:
 // SANITIZE-WITH-ATTR-NEXT:    [[ARR:%.*]] = getelementptr inbounds i8, ptr [[TMP2]], i64 12
@@ -1779,7 +1779,7 @@ struct annotated_struct_array {
 // SANITIZE-WITH-ATTR-NEXT:    [[TMP1:%.*]] = zext i32 [[IDX1]] to i64
 // SANITIZE-WITH-ATTR-NEXT:    br i1 [[TMP0]], label [[CONT3:%.*]], label [[HANDLER_OUT_OF_BOUNDS:%.*]], !prof [[PROF3]], !nosanitize [[META2]]
 // SANITIZE-WITH-ATTR:       handler.out_of_bounds:
-// SANITIZE-WITH-ATTR-NEXT:    tail call void @__ubsan_handle_out_of_bounds_abort(ptr nonnull @[[GLOB33:[0-9]+]], i64 [[TMP1]]) #[[ATTR10]], !nosanitize [[META2]]
+// SANITIZE-WITH-ATTR-NEXT:    tail call void @__ubsan_handle_out_of_bounds_abort(ptr nonnull @[[GLOB36:[0-9]+]], i64 [[TMP1]]) #[[ATTR10]], !nosanitize [[META2]]
 // SANITIZE-WITH-ATTR-NEXT:    unreachable, !nosanitize [[META2]]
 // SANITIZE-WITH-ATTR:       cont3:
 // SANITIZE-WITH-ATTR-NEXT:    [[ARRAYIDX:%.*]] = getelementptr inbounds [10 x ptr], ptr [[ANN]], i64 0, i64 [[TMP1]]
@@ -1791,7 +1791,7 @@ struct annotated_struct_array {
 // SANITIZE-WITH-ATTR-NEXT:    [[TMP4:%.*]] = icmp ult i64 [[IDXPROM15]], [[TMP3]], !nosanitize [[META2]]
 // SANITIZE-WITH-ATTR-NEXT:    br i1 [[TMP4]], label [[CONT20:%.*]], label [[HANDLER_OUT_OF_BOUNDS16:%.*]], !prof [[PROF3]], !nosanitize [[META2]]
 // SANITIZE-WITH-ATTR:       handler.out_of_bounds16:
-// SANITIZE-WITH-ATTR-NEXT:    tail call void @__ubsan_handle_out_of_bounds_abort(ptr nonnull @[[GLOB34:[0-9]+]], i64 [[IDXPROM15]]) #[[ATTR10]], !nosanitize [[META2]]
+// SANITIZE-WITH-ATTR-NEXT:    tail call void @__ubsan_handle_out_of_bounds_abort(ptr nonnull @[[GLOB37:[0-9]+]], i64 [[IDXPROM15]]) #[[ATTR10]], !nosanitize [[META2]]
 // SANITIZE-WITH-ATTR-NEXT:    unreachable, !nosanitize [[META2]]
 // SANITIZE-WITH-ATTR:       cont20:
 // SANITIZE-WITH-ATTR-NEXT:    [[ARRAY:%.*]] = getelementptr inbounds i8, ptr [[TMP2]], i64 12
@@ -1826,7 +1826,7 @@ struct annotated_struct_array {
 // SANITIZE-WITHOUT-ATTR-NEXT:    [[TMP1:%.*]] = zext i32 [[IDX1]] to i64
 // SANITIZE-WITHOUT-ATTR-NEXT:    br i1 [[TMP0]], label [[CONT21:%.*]], label [[HANDLER_OUT_OF_BOUNDS:%.*]], !prof [[PROF8]], !nosanitize [[META9]]
 // SANITIZE-WITHOUT-ATTR:       handler.out_of_bounds:
-// SANITIZE-WITHOUT-ATTR-NEXT:    tail call void @__ubsan_handle_out_of_bounds_abort(ptr nonnull @[[GLOB12:[0-9]+]], i64 [[TMP1]]) #[[ATTR8]], !nosanitize [[META9]]
+// SANITIZE-WITHOUT-ATTR-NEXT:    tail call void @__ubsan_handle_out_of_bounds_abort(ptr nonnull @[[GLOB13:[0-9]+]], i64 [[TMP1]]) #[[ATTR8]], !nosanitize [[META9]]
 // SANITIZE-WITHOUT-ATTR-NEXT:    unreachable, !nosanitize [[META9]]
 // SANITIZE-WITHOUT-ATTR:       cont21:
 // SANITIZE-WITHOUT-ATTR-NEXT:    [[ARRAYIDX:%.*]] = getelementptr inbounds [10 x ptr], ptr [[ANN]], i64 0, i64 [[TMP1]]
diff --git a/clang/test/CodeGen/builtins-reduction-math.c b/clang/test/CodeGen/builtins-reduction-math.c
index 34f39cea5265..acafe9222d59 100644
--- a/clang/test/CodeGen/builtins-reduction-math.c
+++ b/clang/test/CodeGen/builtins-reduction-math.c
@@ -1,5 +1,8 @@
 // RUN: %clang_cc1 -triple x86_64-apple-darwin %s -emit-llvm -disable-llvm-passes -o - | FileCheck %s
 
+// REQUIRES: aarch64-registered-target
+// RUN: %clang_cc1 -O1 -triple aarch64 -target-feature +sve  %s -emit-llvm -disable-llvm-passes -o - | FileCheck --check-prefixes=SVE   %s
+
 typedef float float4 __attribute__((ext_vector_type(4)));
 typedef short int si8 __attribute__((ext_vector_type(8)));
 typedef unsigned int u4 __attribute__((ext_vector_type(4)));
@@ -134,3 +137,53 @@ void test_builtin_reduce_and(si8 vi1, u4 vu1) {
   // CHECK-NEXT: call i32 @llvm.vector.reduce.and.v4i32(<4 x i32> [[VU1]])
   unsigned r3 = __builtin_reduce_and(vu1);
 }
+
+#if defined(__ARM_FEATURE_SVE)
+#include <arm_sve.h>
+
+void test_builtin_reduce_SVE(int a, unsigned long long b, short c, float d) {
+  // SVE-LABEL: void @test_builtin_reduce_SVE(
+
+  svint32_t vec_a = svdup_s32(a);
+  svuint64_t vec_b = svdup_u64(b);
+  svint16_t vec_c1 = svdup_s16(c);
+  svuint16_t vec_c2 = svdup_u16(c);
+  svfloat32_t vec_d = svdup_f32(d);
+
+  // SVE:      [[VF1:%.+]] = load <vscale x 4 x i32>, ptr %vec_a
+  // SVE-NEXT: call i32 @llvm.vector.reduce.add.nxv4i32(<vscale x 4 x i32> [[VF1]])
+  int r1 = __builtin_reduce_add(vec_a);
+
+  // SVE:      [[VF2:%.+]] = load <vscale x 4 x i32>, ptr %vec_a
+  // SVE-NEXT: call i32 @llvm.vector.reduce.mul.nxv4i32(<vscale x 4 x i32> [[VF2]])
+  int r2 = __builtin_reduce_mul(vec_a);
+
+  // SVE:      [[VF3:%.+]] = load <vscale x 2 x i64>, ptr %vec_b
+  // SVE-NEXT: call i64 @llvm.vector.reduce.xor.nxv2i64(<vscale x 2 x i64> [[VF3]])
+  long long r3 = __builtin_reduce_xor(vec_b);
+
+  // SVE:      [[VF4:%.+]] = load <vscale x 2 x i64>, ptr %vec_b
+  // SVE-NEXT: call i64 @llvm.vector.reduce.or.nxv2i64(<vscale x 2 x i64> [[VF4]])
+  long long r4 = __builtin_reduce_or(vec_b);
+
+  // SVE:      [[VF5:%.+]] = load <vscale x 2 x i64>, ptr %vec_b
+  // SVE-NEXT: call i64 @llvm.vector.reduce.and.nxv2i64(<vscale x 2 x i64> [[VF5]])
+  long long r5 = __builtin_reduce_and(vec_b);
+
+  // SVE:      [[VF6:%.+]] = load <vscale x 8 x i16>, ptr %vec_c1
+  // SVE-NEXT: call i16 @llvm.vector.reduce.smax.nxv8i16(<vscale x 8 x i16> [[VF6]])
+  short r6 = __builtin_reduce_max(vec_c1);
+
+  // SVE:      [[VF7:%.+]] = load <vscale x 8 x i16>, ptr %vec_c2
+  // SVE-NEXT: call i16 @llvm.vector.reduce.umin.nxv8i16(<vscale x 8 x i16> [[VF7]])
+  unsigned short r7 = __builtin_reduce_min(vec_c2);
+
+  // SVE:      [[VF8:%.+]] = load <vscale x 4 x float>, ptr %vec_d
+  // SVE-NEXT: call float @llvm.vector.reduce.fmax.nxv4f32(<vscale x 4 x float> [[VF8]])
+  float r8 = __builtin_reduce_max(vec_d);
+
+  // SVE:      [[VF9:%.+]] = load <vscale x 4 x float>, ptr %vec_d
+  // SVE-NEXT: call float @llvm.vector.reduce.fmin.nxv4f32(<vscale x 4 x float> [[VF9]])
+  float r9 = __builtin_reduce_min(vec_d);
+}
+#endif
diff --git a/clang/test/CodeGen/builtins-wasm.c b/clang/test/CodeGen/builtins-wasm.c
index d486d12085f9..9a323da9a8e8 100644
--- a/clang/test/CodeGen/builtins-wasm.c
+++ b/clang/test/CodeGen/builtins-wasm.c
@@ -406,7 +406,7 @@ i32x4 bitselect(i32x4 x, i32x4 y, i32x4 c) {
 
 i8x16 popcnt(i8x16 x) {
   return __builtin_wasm_popcnt_i8x16(x);
-  // WEBASSEMBLY: call <16 x i8> @llvm.ctpop.v16i8(<16 x i8> %x)
+  // WEBASSEMBLY: call  range(i8 0, 9) <16 x i8> @llvm.ctpop.v16i8(<16 x i8> %x)
   // WEBASSEMBLY-NEXT: ret
 }
 
diff --git a/clang/test/CodeGen/ms-intrinsics-other.c b/clang/test/CodeGen/ms-intrinsics-other.c
index 0e9dfe34b84c..fa8422e5bf19 100644
--- a/clang/test/CodeGen/ms-intrinsics-other.c
+++ b/clang/test/CodeGen/ms-intrinsics-other.c
@@ -56,7 +56,7 @@ unsigned char test_BitScanForward(unsigned LONG *Index, unsigned LONG Mask) {
 // CHECK:   [[RESULT:%[a-z0-9._]+]] = phi i8 [ 0, %[[ISZERO_LABEL:[a-z0-9._]+]] ], [ 1, %[[ISNOTZERO_LABEL]] ]
 // CHECK:   ret i8 [[RESULT]]
 // CHECK:   [[ISNOTZERO_LABEL]]:
-// CHECK:   [[INDEX:%[0-9]+]] = tail call i32 @llvm.cttz.i32(i32 %Mask, i1 true)
+// CHECK:   [[INDEX:%[0-9]+]] = tail call range(i32 0, 33) i32 @llvm.cttz.i32(i32 %Mask, i1 true)
 // CHECK:   store i32 [[INDEX]], ptr %Index, align 4
 // CHECK:   br label %[[END_LABEL]]
 
@@ -70,7 +70,7 @@ unsigned char test_BitScanReverse(unsigned LONG *Index, unsigned LONG Mask) {
 // CHECK:   [[RESULT:%[a-z0-9._]+]] = phi i8 [ 0, %[[ISZERO_LABEL:[a-z0-9._]+]] ], [ 1, %[[ISNOTZERO_LABEL]] ]
 // CHECK:   ret i8 [[RESULT]]
 // CHECK:   [[ISNOTZERO_LABEL]]:
-// CHECK:   [[REVINDEX:%[0-9]+]] = tail call i32 @llvm.ctlz.i32(i32 %Mask, i1 true)
+// CHECK:   [[REVINDEX:%[0-9]+]] = tail call range(i32 0, 33) i32 @llvm.ctlz.i32(i32 %Mask, i1 true)
 // CHECK:   [[INDEX:%[0-9]+]] = xor i32 [[REVINDEX]], 31
 // CHECK:   store i32 [[INDEX]], ptr %Index, align 4
 // CHECK:   br label %[[END_LABEL]]
@@ -86,7 +86,7 @@ unsigned char test_BitScanForward64(unsigned LONG *Index, unsigned __int64 Mask)
 // CHECK:   [[RESULT:%[a-z0-9._]+]] = phi i8 [ 0, %[[ISZERO_LABEL:[a-z0-9._]+]] ], [ 1, %[[ISNOTZERO_LABEL]] ]
 // CHECK:   ret i8 [[RESULT]]
 // CHECK:   [[ISNOTZERO_LABEL]]:
-// CHECK:   [[INDEX:%[0-9]+]] = tail call i64 @llvm.cttz.i64(i64 %Mask, i1 true)
+// CHECK:   [[INDEX:%[0-9]+]] = tail call range(i64 0, 65) i64 @llvm.cttz.i64(i64 %Mask, i1 true)
 // CHECK:   [[TRUNC_INDEX:%[0-9]+]] = trunc nuw nsw i64 [[INDEX]] to i32
 // CHECK:   store i32 [[TRUNC_INDEX]], ptr %Index, align 4
 // CHECK:   br label %[[END_LABEL]]
@@ -101,7 +101,7 @@ unsigned char test_BitScanReverse64(unsigned LONG *Index, unsigned __int64 Mask)
 // CHECK:   [[RESULT:%[a-z0-9._]+]] = phi i8 [ 0, %[[ISZERO_LABEL:[a-z0-9._]+]] ], [ 1, %[[ISNOTZERO_LABEL]] ]
 // CHECK:   ret i8 [[RESULT]]
 // CHECK:   [[ISNOTZERO_LABEL]]:
-// CHECK:   [[REVINDEX:%[0-9]+]] = tail call i64 @llvm.ctlz.i64(i64 %Mask, i1 true)
+// CHECK:   [[REVINDEX:%[0-9]+]] = tail call range(i64 0, 65) i64 @llvm.ctlz.i64(i64 %Mask, i1 true)
 // CHECK:   [[TRUNC_REVINDEX:%[0-9]+]] = trunc nuw nsw i64 [[REVINDEX]] to i32
 // CHECK:   [[INDEX:%[0-9]+]] = xor i32 [[TRUNC_REVINDEX]], 63
 // CHECK:   store i32 [[INDEX]], ptr %Index, align 4
@@ -187,7 +187,7 @@ unsigned short test__lzcnt16(unsigned short x) {
   return __lzcnt16(x);
 }
 // CHECK: i16 @test__lzcnt16
-// CHECK:  [[RESULT:%[0-9]+]] = tail call i16 @llvm.ctlz.i16(i16 %x, i1 false)
+// CHECK:  [[RESULT:%[0-9]+]] = tail call range(i16 0, 17) i16 @llvm.ctlz.i16(i16 %x, i1 false)
 // CHECK: ret i16 [[RESULT]]
 // CHECK: }
 
@@ -195,7 +195,7 @@ unsigned int test__lzcnt(unsigned int x) {
   return __lzcnt(x);
 }
 // CHECK: i32 @test__lzcnt
-// CHECK:  [[RESULT:%[0-9]+]] = tail call i32 @llvm.ctlz.i32(i32 %x, i1 false)
+// CHECK:  [[RESULT:%[0-9]+]] = tail call range(i32 0, 33) i32 @llvm.ctlz.i32(i32 %x, i1 false)
 // CHECK: ret i32 [[RESULT]]
 // CHECK: }
 
@@ -203,7 +203,7 @@ unsigned __int64 test__lzcnt64(unsigned __int64 x) {
   return __lzcnt64(x);
 }
 // CHECK: i64 @test__lzcnt64
-// CHECK:  [[RESULT:%[0-9]+]] = tail call i64 @llvm.ctlz.i64(i64 %x, i1 false)
+// CHECK:  [[RESULT:%[0-9]+]] = tail call range(i64 0, 65) i64 @llvm.ctlz.i64(i64 %x, i1 false)
 // CHECK: ret i64 [[RESULT]]
 // CHECK: }
 
@@ -211,7 +211,7 @@ unsigned short test__popcnt16(unsigned short x) {
   return __popcnt16(x);
 }
 // CHECK: i16 @test__popcnt16
-// CHECK:  [[RESULT:%[0-9]+]] = tail call i16 @llvm.ctpop.i16(i16 %x)
+// CHECK:  [[RESULT:%[0-9]+]] = tail call range(i16 0, 17) i16 @llvm.ctpop.i16(i16 %x)
 // CHECK: ret i16 [[RESULT]]
 // CHECK: }
 
@@ -219,7 +219,7 @@ unsigned int test__popcnt(unsigned int x) {
   return __popcnt(x);
 }
 // CHECK: i32 @test__popcnt
-// CHECK:  [[RESULT:%[0-9]+]] = tail call i32 @llvm.ctpop.i32(i32 %x)
+// CHECK:  [[RESULT:%[0-9]+]] = tail call range(i32 0, 33) i32 @llvm.ctpop.i32(i32 %x)
 // CHECK: ret i32 [[RESULT]]
 // CHECK: }
 
@@ -227,7 +227,7 @@ unsigned __int64 test__popcnt64(unsigned __int64 x) {
   return __popcnt64(x);
 }
 // CHECK: i64 @test__popcnt64
-// CHECK:  [[RESULT:%[0-9]+]] = tail call i64 @llvm.ctpop.i64(i64 %x)
+// CHECK:  [[RESULT:%[0-9]+]] = tail call range(i64 0, 65) i64 @llvm.ctpop.i64(i64 %x)
 // CHECK: ret i64 [[RESULT]]
 // CHECK: }
 
diff --git a/clang/test/CodeGen/ms-intrinsics.c b/clang/test/CodeGen/ms-intrinsics.c
index 6eabd725e2f7..c3d64fda0b90 100644
--- a/clang/test/CodeGen/ms-intrinsics.c
+++ b/clang/test/CodeGen/ms-intrinsics.c
@@ -157,7 +157,7 @@ unsigned char test_BitScanForward(unsigned long *Index, unsigned long Mask) {
 // CHECK:   ret i8 [[RESULT]]
 // CHECK:   [[ISNOTZERO_LABEL]]:
 // CHECK:   [[IDXGEP:%[a-z0-9._]+]] = getelementptr inbounds i8, ptr %Index, {{i64|i32}} 4
-// CHECK:   [[INDEX:%[0-9]+]] = tail call i32 @llvm.cttz.i32(i32 %Mask, i1 true)
+// CHECK:   [[INDEX:%[0-9]+]] = tail call range(i32 0, 33) i32 @llvm.cttz.i32(i32 %Mask, i1 true)
 // CHECK:   store i32 [[INDEX]], ptr [[IDXGEP]], align 4
 // CHECK:   br label %[[END_LABEL]]
 
@@ -172,7 +172,7 @@ unsigned char test_BitScanReverse(unsigned long *Index, unsigned long Mask) {
 // CHECK:   ret i8 [[RESULT]]
 // CHECK:   [[ISNOTZERO_LABEL]]:
 // CHECK:   [[IDXGEP:%[a-z0-9._]+]] = getelementptr inbounds i8, ptr %Index, {{i64|i32}} 4
-// CHECK:   [[REVINDEX:%[0-9]+]] = tail call i32 @llvm.ctlz.i32(i32 %Mask, i1 true)
+// CHECK:   [[REVINDEX:%[0-9]+]] = tail call range(i32 0, 33) i32 @llvm.ctlz.i32(i32 %Mask, i1 true)
 // CHECK:   [[INDEX:%[0-9]+]] = xor i32 [[REVINDEX]], 31
 // CHECK:   store i32 [[INDEX]], ptr [[IDXGEP]], align 4
 // CHECK:   br label %[[END_LABEL]]
@@ -188,7 +188,7 @@ unsigned char test_BitScanForward64(unsigned long *Index, unsigned __int64 Mask)
 // CHECK-ARM-X64:   [[RESULT:%[a-z0-9._]+]] = phi i8 [ 0, %[[ISZERO_LABEL:[a-z0-9._]+]] ], [ 1, %[[ISNOTZERO_LABEL]] ]
 // CHECK-ARM-X64:   ret i8 [[RESULT]]
 // CHECK-ARM-X64:   [[ISNOTZERO_LABEL]]:
-// CHECK-ARM-X64:   [[INDEX:%[0-9]+]] = tail call i64 @llvm.cttz.i64(i64 %Mask, i1 true)
+// CHECK-ARM-X64:   [[INDEX:%[0-9]+]] = tail call range(i64 0, 65) i64 @llvm.cttz.i64(i64 %Mask, i1 true)
 // CHECK-ARM-X64:   [[TRUNC_INDEX:%[0-9]+]] = trunc nuw nsw i64 [[INDEX]] to i32
 // CHECK-ARM-X64:   store i32 [[TRUNC_INDEX]], ptr %Index, align 4
 // CHECK-ARM-X64:   br label %[[END_LABEL]]
@@ -203,7 +203,7 @@ unsigned char test_BitScanReverse64(unsigned long *Index, unsigned __int64 Mask)
 // CHECK-ARM-X64:   [[RESULT:%[a-z0-9._]+]] = phi i8 [ 0, %[[ISZERO_LABEL:[a-z0-9._]+]] ], [ 1, %[[ISNOTZERO_LABEL]] ]
 // CHECK-ARM-X64:   ret i8 [[RESULT]]
 // CHECK-ARM-X64:   [[ISNOTZERO_LABEL]]:
-// CHECK-ARM-X64:   [[REVINDEX:%[0-9]+]] = tail call i64 @llvm.ctlz.i64(i64 %Mask, i1 true)
+// CHECK-ARM-X64:   [[REVINDEX:%[0-9]+]] = tail call range(i64 0, 65) i64 @llvm.ctlz.i64(i64 %Mask, i1 true)
 // CHECK-ARM-X64:   [[TRUNC_REVINDEX:%[0-9]+]] = trunc nuw nsw i64 [[REVINDEX]] to i32
 // CHECK-ARM-X64:   [[INDEX:%[0-9]+]] = xor i32 [[TRUNC_REVINDEX]], 63
 // CHECK-ARM-X64:   store i32 [[INDEX]], ptr %Index, align 4
diff --git a/clang/test/CodeGen/ms-mixed-ptr-sizes.c b/clang/test/CodeGen/ms-mixed-ptr-sizes.c
index 89d05fd30b72..51bea60eb39d 100644
--- a/clang/test/CodeGen/ms-mixed-ptr-sizes.c
+++ b/clang/test/CodeGen/ms-mixed-ptr-sizes.c
@@ -49,7 +49,7 @@ void test_other(struct Foo *f, __attribute__((address_space(10))) int *i) {
 }
 
 int test_compare1(int *__ptr32 __uptr i, int *__ptr64 j) {
-  // ALL-LABEL: define dso_local noundef i32 @test_compare1
+  // ALL-LABEL: define dso_local range(i32 0, 2) i32 @test_compare1
   // X64: %{{.+}} = addrspacecast ptr %j to ptr addrspace(271)
   // X64: %cmp = icmp eq ptr addrspace(271) %{{.+}}, %i
   // X86: %{{.+}} = addrspacecast ptr addrspace(272) %j to ptr addrspace(271)
@@ -58,7 +58,7 @@ int test_compare1(int *__ptr32 __uptr i, int *__ptr64 j) {
 }
 
 int test_compare2(int *__ptr32 __sptr i, int *__ptr64 j) {
-  // ALL-LABEL: define dso_local noundef i32 @test_compare2
+  // ALL-LABEL: define dso_local range(i32 0, 2) i32 @test_compare2
   // X64: %{{.+}} = addrspacecast ptr %j to ptr addrspace(270)
   // X64: %cmp = icmp eq ptr addrspace(270) %{{.+}}, %i
   // X86: %{{.+}} = addrspacecast ptr addrspace(272) %j to ptr
@@ -67,7 +67,7 @@ int test_compare2(int *__ptr32 __sptr i, int *__ptr64 j) {
 }
 
 int test_compare3(int *__ptr32 __uptr i, int *__ptr64 j) {
-  // ALL-LABEL: define dso_local noundef i32 @test_compare3
+  // ALL-LABEL: define dso_local range(i32 0, 2) i32 @test_compare3
   // X64: %{{.+}} = addrspacecast ptr addrspace(271) %i to ptr
   // X64: %cmp = icmp eq ptr %{{.+}}, %j
   // X86: %{{.+}} = addrspacecast ptr addrspace(271) %i to ptr addrspace(272)
@@ -76,7 +76,7 @@ int test_compare3(int *__ptr32 __uptr i, int *__ptr64 j) {
 }
 
 int test_compare4(int *__ptr32 __sptr i, int *__ptr64 j) {
-  // ALL-LABEL: define dso_local noundef i32 @test_compare4
+  // ALL-LABEL: define dso_local range(i32 0, 2) i32 @test_compare4
   // X64: %{{.+}} = addrspacecast ptr addrspace(270) %i to ptr
   // X64: %cmp = icmp eq ptr %{{.+}}, %j
   // X86: %{{.+}} = addrspacecast ptr %i to ptr addrspace(272)
diff --git a/clang/test/CodeGenCXX/blocks.cpp b/clang/test/CodeGenCXX/blocks.cpp
index eaab1890dfc4..afe078890553 100644
--- a/clang/test/CodeGenCXX/blocks.cpp
+++ b/clang/test/CodeGenCXX/blocks.cpp
@@ -149,8 +149,8 @@ namespace test5 {
   // CHECK-NEXT: [[X:%.*]] = alloca [[A:%.*]], align 4
   // CHECK-NEXT: [[B:%.*]] = alloca ptr, align 8
   // CHECK-NEXT: [[BLOCK:%.*]] = alloca [[BLOCK_T:.*]], align 8
-  // CHECK-NEXT: [[CLEANUP_ACTIVE:%.*]] = alloca i1
   // CHECK-NEXT: [[COND_CLEANUP_SAVE:%.*]] = alloca ptr, align 8
+  // CHECK-NEXT: [[CLEANUP_ACTIVE:%.*]] = alloca i1
   // CHECK-NEXT: [[T0:%.*]] = zext i1
   // CHECK-NEXT: store i8 [[T0]], ptr [[COND]], align 1
   // CHECK-NEXT: call void @_ZN5test51AC1Ev(ptr {{[^,]*}} [[X]])
@@ -162,8 +162,8 @@ namespace test5 {
   // CHECK-NOT:  br
   // CHECK:      [[CAPTURE:%.*]] = getelementptr inbounds [[BLOCK_T]], ptr [[BLOCK]], i32 0, i32 5
   // CHECK-NEXT: call void @_ZN5test51AC1ERKS0_(ptr {{[^,]*}} [[CAPTURE]], ptr noundef nonnull align {{[0-9]+}} dereferenceable({{[0-9]+}}) [[X]])
-  // CHECK-NEXT: store i1 true, ptr [[CLEANUP_ACTIVE]]
   // CHECK-NEXT: store ptr [[CAPTURE]], ptr [[COND_CLEANUP_SAVE]], align 8
+  // CHECK-NEXT: store i1 true, ptr [[CLEANUP_ACTIVE]]
   // CHECK-NEXT: br label
   // CHECK:      br label
   // CHECK:      phi
diff --git a/clang/test/CodeGenCXX/control-flow-in-stmt-expr.cpp b/clang/test/CodeGenCXX/control-flow-in-stmt-expr.cpp
new file mode 100644
index 000000000000..ac466ee5bba4
--- /dev/null
+++ b/clang/test/CodeGenCXX/control-flow-in-stmt-expr.cpp
@@ -0,0 +1,522 @@
+// RUN: %clang_cc1 --std=c++20 -fexceptions -triple x86_64-linux-gnu -emit-llvm %s -o - | FileCheck -check-prefixes=EH %s
+// RUN: %clang_cc1 --std=c++20 -triple x86_64-linux-gnu -emit-llvm %s -o - | FileCheck -check-prefixes=NOEH,CHECK %s
+
+struct Printy {
+  Printy(const char *name) : name(name) {}
+  ~Printy() {}
+  const char *name;
+};
+
+int foo() { return 2; }
+
+struct Printies {
+  Printy a;
+  Printy b;
+  Printy c;
+};
+
+void ParenInit() {
+  // CHECK-LABEL: define dso_local void @_Z9ParenInitv()
+  // CHECK: [[CLEANUP_DEST:%.+]] = alloca i32, align 4
+  Printies ps(Printy("a"), 
+              // CHECK: call void @_ZN6PrintyC1EPKc
+              ({
+                if (foo()) return;
+                // CHECK:     if.then:
+                // CHECK-NEXT:   store i32 1, ptr [[CLEANUP_DEST]], align 4
+                // CHECK-NEXT:   br label %cleanup
+                Printy("b");
+                // CHECK:     if.end:
+                // CHECK-NEXT:  call void @_ZN6PrintyC1EPKc
+              }),
+              ({
+                if (foo()) return;
+                // CHECK:     if.then{{.*}}:
+                // CHECK-NEXT:  store i32 1, ptr [[CLEANUP_DEST]], align 4
+                // CHECK-NEXT:  call void @_ZN6PrintyD1Ev
+                // CHECK-NEXT:  br label %cleanup
+                Printy("c");
+                // CHECK:     if.end{{.*}}:
+                // CHECK-NEXT:  call void @_ZN6PrintyC1EPKc
+                // CHECK-NEXT:  call void @_ZN8PrintiesD1Ev
+                // CHECK-NEXT:  br label %return
+              }));
+  // CHECK:     cleanup:
+  // CHECK-NEXT:  call void @_ZN6PrintyD1Ev
+  // CHECK-NEXT:  br label %return
+}
+
+void break_in_stmt_expr() {
+  // Verify that the "break" in "if.then".calls dtor before jumping to "for.end".
+
+  // CHECK-LABEL: define dso_local void @_Z18break_in_stmt_exprv()
+  Printies p{Printy("a"), 
+            // CHECK: call void @_ZN6PrintyC1EPKc
+            ({
+                for (;;) {
+                    Printies ps{
+                      Printy("b"), 
+                      // CHECK: for.cond:
+                      // CHECK:   call void @_ZN6PrintyC1EPKc
+                      ({
+                        if (foo()) {
+                          break;
+                          // CHECK:       if.then:
+                          // CHECK-NEXT:    call void @_ZN6PrintyD1Ev
+                          // CHECK-NEXT:    br label %for.end
+                        }
+                        Printy("c");
+                        // CHECK:       if.end:
+                        // CHECK-NEXT:    call void @_ZN6PrintyC1EPKc
+                      }),
+                      Printy("d")};
+                      // CHECK:           call void @_ZN6PrintyC1EPKc
+                      // CHECK-NEXT:      call void @_ZN8PrintiesD1Ev
+                      // CHECK-NEXT:      br label %for.cond
+                }
+                Printy("e");
+  // CHECK:       for.end:
+  // CHECK-NEXT:    call void @_ZN6PrintyC1EPKc
+              }),
+              Printy("f")};
+  // CHECK:         call void @_ZN6PrintyC1EPKc
+  // CHECK-NEXT:    call void @_ZN8PrintiesD1Ev
+}
+
+void goto_in_stmt_expr() {
+  // Verify that:
+  //  - correct branch fixups for deactivated normal cleanups are generated correctly.
+
+  // CHECK-LABEL: define dso_local void @_Z17goto_in_stmt_exprv()
+  // CHECK: [[CLEANUP_DEST_SLOT:%cleanup.dest.slot.*]] = alloca i32, align 4
+  {
+    Printies p1{Printy("a"), // CHECK: call void @_ZN6PrintyC1EPKc
+                ({
+                  {
+                    Printies p2{Printy("b"),
+                                // CHECK: call void @_ZN6PrintyC1EPKc
+                                ({
+                                  if (foo() == 1) {
+                                    goto in;
+                                    // CHECK:       if.then:
+                                    // CHECK-NEXT:    store i32 2, ptr [[CLEANUP_DEST_SLOT]], align 4
+                                    // CHECK-NEXT:    br label %[[CLEANUP1:.+]]
+                                  }
+                                  if (foo() == 2) {
+                                    goto out;
+                                    // CHECK:       if.then{{.*}}:
+                                    // CHECK-NEXT:    store i32 3, ptr [[CLEANUP_DEST_SLOT]], align 4
+                                    // CHECK-NEXT:    br label %[[CLEANUP1]]
+                                  }
+                                  Printy("c");
+                                  // CHECK:       if.end{{.*}}:
+                                  // CHECK-NEXT:    call void @_ZN6PrintyC1EPKc
+                                }),
+                                Printy("d")};
+                                // CHECK:           call void @_ZN6PrintyC1EPKc
+                                // CHECK-NEXT:      call void @_ZN8PrintiesD1Ev
+                                // CHECK-NEXT:      br label %in
+
+                  }
+                in:
+                  Printy("e");
+                // CHECK:       in:                                               ; preds = %if.end{{.*}}, %[[CLEANUP1]]
+                // CHECK-NEXT:    call void @_ZN6PrintyC1EPKc
+                }),
+                Printy("f")};
+                // CHECK:         call void @_ZN6PrintyC1EPKc
+                // CHECK-NEXT:    call void @_ZN8PrintiesD1Ev
+                // CHECK-NEXT:    br label %out
+  }
+out:
+  return;
+  // CHECK:       out:
+  // CHECK-NEXT:    ret void
+
+  // CHECK:       [[CLEANUP1]]:                                          ; preds = %if.then{{.*}}, %if.then
+  // CHECK-NEXT:    call void @_ZN6PrintyD1Ev
+  // CHECK-NEXT:    %cleanup.dest = load i32, ptr [[CLEANUP_DEST_SLOT]], align 4
+  // CHECK-NEXT:    switch i32 %cleanup.dest, label %[[CLEANUP2:.+]] [
+  // CHECK-NEXT:      i32 2, label %in
+  // CHECK-NEXT:    ]
+
+  // CHECK:       [[CLEANUP2]]:                                         ; preds = %[[CLEANUP1]]
+  // CHECK-NEXT:    call void @_ZN6PrintyD1Ev
+  // CHECK-NEXT:    %cleanup.dest{{.*}} = load i32, ptr [[CLEANUP_DEST_SLOT]], align 4
+  // CHECK-NEXT:    switch i32 %cleanup.dest{{.*}}, label %unreachable [
+  // CHECK-NEXT:      i32 3, label %out
+  // CHECK-NEXT:    ]
+}
+
+void ArrayInit() {
+  // Printy arr[4] = {ctorA, ctorB, stmt-exprC, stmt-exprD};
+  // Verify that:
+  //  - We do the necessary stores for array cleanups (endOfInit and last constructed element).
+  //  - We update the array init element correctly for ctorA, ctorB and stmt-exprC.
+  //  - stmt-exprC and stmt-exprD share the array body dtor code (see %cleanup).
+
+  // CHECK-LABEL: define dso_local void @_Z9ArrayInitv()
+  // CHECK: %arrayinit.endOfInit = alloca ptr, align 8
+  // CHECK: %cleanup.dest.slot = alloca i32, align 4
+  // CHECK: %arrayinit.begin = getelementptr inbounds [4 x %struct.Printy], ptr %arr, i64 0, i64 0
+  // CHECK: store ptr %arrayinit.begin, ptr %arrayinit.endOfInit, align 8
+  Printy arr[4] = {
+    Printy("a"),
+    // CHECK: call void @_ZN6PrintyC1EPKc(ptr noundef nonnull align 8 dereferenceable(8) %arrayinit.begin, ptr noundef @.str)
+    // CHECK: [[ARRAYINIT_ELEMENT1:%.+]] = getelementptr inbounds %struct.Printy, ptr %arrayinit.begin, i64 1
+    // CHECK: store ptr [[ARRAYINIT_ELEMENT1]], ptr %arrayinit.endOfInit, align 8
+    Printy("b"),
+    // CHECK: call void @_ZN6PrintyC1EPKc(ptr noundef nonnull align 8 dereferenceable(8) [[ARRAYINIT_ELEMENT1]], ptr noundef @.str.1)
+    // CHECK: [[ARRAYINIT_ELEMENT2:%.+]] = getelementptr inbounds %struct.Printy, ptr [[ARRAYINIT_ELEMENT1]], i64 1
+    // CHECK: store ptr [[ARRAYINIT_ELEMENT2]], ptr %arrayinit.endOfInit, align 8
+    ({
+    // CHECK: br i1 {{.*}}, label %if.then, label %if.end
+      if (foo()) {
+        return;
+      // CHECK:       if.then:
+      // CHECK-NEXT:    store i32 1, ptr %cleanup.dest.slot, align 4
+      // CHECK-NEXT:    br label %cleanup
+      }
+      // CHECK:       if.end:
+      Printy("c");
+      // CHECK-NEXT:    call void @_ZN6PrintyC1EPKc
+      // CHECK-NEXT:    %arrayinit.element2 = getelementptr inbounds %struct.Printy, ptr %arrayinit.element1, i64 1
+      // CHECK-NEXT:    store ptr %arrayinit.element2, ptr %arrayinit.endOfInit, align 8
+    }),
+    ({
+    // CHECK: br i1 {{%.+}} label %[[IF_THEN2:.+]], label %[[IF_END2:.+]]
+      if (foo()) {
+        return;
+      // CHECK:       [[IF_THEN2]]:
+      // CHECK-NEXT:    store i32 1, ptr %cleanup.dest.slot, align 4
+      // CHECK-NEXT:    br label %cleanup
+      }
+      // CHECK:       [[IF_END2]]:
+      Printy("d");
+      // CHECK-NEXT:    call void @_ZN6PrintyC1EPKc
+      // CHECK-NEXT:    %array.begin = getelementptr inbounds [4 x %struct.Printy], ptr %arr, i32 0, i32 0
+      // CHECK-NEXT:    %0 = getelementptr inbounds %struct.Printy, ptr %array.begin, i64 4
+      // CHECK-NEXT:    br label %[[ARRAY_DESTROY_BODY1:.+]]
+  }),
+  };
+
+  // CHECK:       [[ARRAY_DESTROY_BODY1]]:
+  // CHECK-NEXT:    %arraydestroy.elementPast{{.*}} = phi ptr [ %0, %[[IF_END2]] ], [ %arraydestroy.element{{.*}}, %[[ARRAY_DESTROY_BODY1]] ]
+  // CHECK-NEXT:    %arraydestroy.element{{.*}} = getelementptr inbounds %struct.Printy, ptr %arraydestroy.elementPast{{.*}}, i64 -1
+  // CHECK-NEXT:    call void @_ZN6PrintyD1Ev
+  // CHECK-NEXT:    %arraydestroy.done{{.*}} = icmp eq ptr %arraydestroy.element{{.*}}, %array.begin
+  // CHECK-NEXT:    br i1 %arraydestroy.done{{.*}}, label %[[ARRAY_DESTROY_DONE1:.+]], label %[[ARRAY_DESTROY_BODY1]]
+
+  // CHECK:       [[ARRAY_DESTROY_DONE1]]:
+  // CHECK-NEXT:    ret void
+
+  // CHECK:       cleanup:
+  // CHECK-NEXT:    %1 = load ptr, ptr %arrayinit.endOfInit, align 8
+  // CHECK-NEXT:    %arraydestroy.isempty = icmp eq ptr %arrayinit.begin, %1
+  // CHECK-NEXT:    br i1 %arraydestroy.isempty, label %[[ARRAY_DESTROY_DONE2:.+]], label %[[ARRAY_DESTROY_BODY2:.+]]
+
+  // CHECK:       [[ARRAY_DESTROY_BODY2]]:
+  // CHECK-NEXT:    %arraydestroy.elementPast = phi ptr [ %1, %cleanup ], [ %arraydestroy.element, %[[ARRAY_DESTROY_BODY2]] ]
+  // CHECK-NEXT:    %arraydestroy.element = getelementptr inbounds %struct.Printy, ptr %arraydestroy.elementPast, i64 -1
+  // CHECK-NEXT:    call void @_ZN6PrintyD1Ev(ptr noundef nonnull align 8 dereferenceable(8) %arraydestroy.element)
+  // CHECK-NEXT:    %arraydestroy.done = icmp eq ptr %arraydestroy.element, %arrayinit.begin
+  // CHECK-NEXT:    br i1 %arraydestroy.done, label %[[ARRAY_DESTROY_DONE2]], label %[[ARRAY_DESTROY_BODY2]]
+
+  // CHECK:       [[ARRAY_DESTROY_DONE2]]:
+  // CHECK-NEXT:    br label %[[ARRAY_DESTROY_DONE1]]
+}
+
+void ArraySubobjects() {
+  struct S {
+    Printy arr1[2];
+    Printy arr2[2];
+    Printy p;
+  };
+  // CHECK-LABEL: define dso_local void @_Z15ArraySubobjectsv()
+  // CHECK: %arrayinit.endOfInit = alloca ptr, align 8
+  S s{{Printy("a"), Printy("b")},
+      // CHECK: call void @_ZN6PrintyC1EPKc
+      // CHECK: call void @_ZN6PrintyC1EPKc
+      {Printy("a"),
+      // CHECK: [[ARRAYINIT_BEGIN:%.+]] = getelementptr inbounds [2 x %struct.Printy]
+      // CHECK: store ptr [[ARRAYINIT_BEGIN]], ptr %arrayinit.endOfInit, align 8
+      // CHECK: call void @_ZN6PrintyC1EPKc
+      // CHECK: [[ARRAYINIT_ELEMENT:%.+]] = getelementptr inbounds %struct.Printy
+      // CHECK: store ptr [[ARRAYINIT_ELEMENT]], ptr %arrayinit.endOfInit, align 8
+      ({
+         if (foo()) {
+           return;
+           // CHECK:      if.then:
+           // CHECK-NEXT:   [[V0:%.+]] = load ptr, ptr %arrayinit.endOfInit, align 8
+           // CHECK-NEXT:   %arraydestroy.isempty = icmp eq ptr [[ARRAYINIT_BEGIN]], [[V0]]
+           // CHECK-NEXT:   br i1 %arraydestroy.isempty, label %[[ARRAY_DESTROY_DONE:.+]], label %[[ARRAY_DESTROY_BODY:.+]]
+         }
+         Printy("b");
+       })
+      },
+      Printy("c")
+      // CHECK:       if.end:
+      // CHECK-NEXT:    call void @_ZN6PrintyC1EPKc
+      // CHECK:         call void @_ZN6PrintyC1EPKc
+      // CHECK-NEXT:    call void @_ZZ15ArraySubobjectsvEN1SD1Ev
+      // CHECK-NEXT:    br label %return
+    };
+    // CHECK:       return:
+    // CHECK-NEXT:    ret void
+
+    // CHECK:       [[ARRAY_DESTROY_BODY]]:
+    // CHECK-NEXT:    %arraydestroy.elementPast = phi ptr [ %0, %if.then ], [ %arraydestroy.element, %[[ARRAY_DESTROY_BODY]] ]
+    // CHECK-NEXT:    %arraydestroy.element = getelementptr inbounds %struct.Printy, ptr %arraydestroy.elementPast, i64 -1
+    // CHECK-NEXT:    call void @_ZN6PrintyD1Ev(ptr noundef nonnull align 8 dereferenceable(8) %arraydestroy.element)
+    // CHECK-NEXT:    %arraydestroy.done = icmp eq ptr %arraydestroy.element, [[ARRAYINIT_BEGIN]]
+    // CHECK-NEXT:    br i1 %arraydestroy.done, label %[[ARRAY_DESTROY_DONE]], label %[[ARRAY_DESTROY_BODY]]
+
+    // CHECK:       [[ARRAY_DESTROY_DONE]]
+    // CHECK-NEXT:    [[ARRAY_BEGIN:%.+]] = getelementptr inbounds [2 x %struct.Printy], ptr %arr1, i32 0, i32 0
+    // CHECK-NEXT:    [[V1:%.+]] = getelementptr inbounds %struct.Printy, ptr [[ARRAY_BEGIN]], i64 2
+    // CHECK-NEXT:    br label %[[ARRAY_DESTROY_BODY2:.+]]
+
+    // CHECK:       [[ARRAY_DESTROY_BODY2]]:
+    // CHECK-NEXT:    %arraydestroy.elementPast5 = phi ptr [ %1, %[[ARRAY_DESTROY_DONE]] ], [ %arraydestroy.element6, %[[ARRAY_DESTROY_BODY2]] ]
+    // CHECK-NEXT:    %arraydestroy.element6 = getelementptr inbounds %struct.Printy, ptr %arraydestroy.elementPast5, i64 -1
+    // CHECK-NEXT:    call void @_ZN6PrintyD1Ev(ptr noundef nonnull align 8 dereferenceable(8) %arraydestroy.element6)
+    // CHECK-NEXT:    %arraydestroy.done7 = icmp eq ptr %arraydestroy.element6, [[ARRAY_BEGIN]]
+    // CHECK-NEXT:    br i1 %arraydestroy.done7, label %[[ARRAY_DESTROY_DONE2:.+]], label %[[ARRAY_DESTROY_BODY2]]
+
+
+    // CHECK:     [[ARRAY_DESTROY_DONE2]]:
+    // CHECK-NEXT:  br label %return
+}
+
+void LambdaInit() {
+  // CHECK-LABEL: define dso_local void @_Z10LambdaInitv()
+  auto S = [a = Printy("a"), b = ({
+                               if (foo()) {
+                                 return;
+                                 // CHECK:       if.then:
+                                 // CHECK-NEXT:    call void @_ZN6PrintyD1Ev
+                                 // CHECK-NEXT:    br label %return
+                               }
+                               Printy("b");
+                             })]() { return a; };
+}
+
+struct PrintyRefBind {
+  const Printy &a;
+  const Printy &b;
+};
+
+struct Temp {
+  Temp();
+  ~Temp();
+};
+Temp CreateTemp();
+Printy CreatePrinty();
+Printy CreatePrinty(const Temp&);
+
+void LifetimeExtended() {
+  // CHECK-LABEL: define dso_local void @_Z16LifetimeExtendedv
+  PrintyRefBind ps = {Printy("a"), ({
+                        if (foo()) {
+                          return;
+                          // CHECK: if.then:
+                          // CHECK-NEXT: call void @_ZN6PrintyD1Ev
+                          // CHECK-NEXT: br label %return
+                        }
+                        Printy("b");
+                      })};
+}
+
+void ConditionalLifetimeExtended() {
+  // CHECK-LABEL: @_Z27ConditionalLifetimeExtendedv()
+
+  // Verify that we create two cleanup flags.
+  //  1. First for the cleanup which is deactivated after full expression.
+  //  2. Second for the life-ext cleanup which is activated if the branch is taken.
+
+  // Note: We use `CreateTemp()` to ensure that life-ext destroy cleanup is not at
+  // the top of EHStack on deactivation. This ensures using active flags.
+
+  Printy* p1 = nullptr;
+  // CHECK:       store i1 false, ptr [[BRANCH1_DEFERRED:%cleanup.cond]], align 1
+  // CHECK-NEXT:  store i1 false, ptr [[BRANCH1_LIFEEXT:%cleanup.cond.*]], align 1
+  PrintyRefBind ps = {
+      p1 != nullptr ? static_cast<const Printy&>(CreatePrinty())
+      // CHECK:       cond.true:
+      // CHECK-NEXT:    call void @_Z12CreatePrintyv
+      // CHECK-NEXT:    store i1 true, ptr [[BRANCH1_DEFERRED]], align 1
+      // CHECK-NEXT:    store i1 true, ptr [[BRANCH1_LIFEEXT]], align 1
+      // CHECK-NEXT:    br label %{{.*}}
+      : foo() ? static_cast<const Printy&>(CreatePrinty(CreateTemp()))
+              : *p1,
+      ({
+        if (foo()) return;
+        Printy("c");
+        // CHECK:       if.end:
+        // CHECK-NEXT:    call void @_ZN6PrintyC1EPKc
+        // CHECK-NEXT:    store ptr
+      })};
+      // CHECK-NEXT:      store i1 false, ptr [[BRANCH1_DEFERRED]], align 1
+      // CHECK-NEXT:      store i32 0, ptr %cleanup.dest.slot, align 4
+      // CHECK-NEXT:      br label %cleanup
+
+}
+
+void NewArrayInit() {
+  // CHECK-LABEL: define dso_local void @_Z12NewArrayInitv()
+  // CHECK: %array.init.end = alloca ptr, align 8
+  // CHECK: store ptr %0, ptr %array.init.end, align 8
+  Printy *array = new Printy[3]{
+    "a",
+    // CHECK: call void @_ZN6PrintyC1EPKc
+    // CHECK: store ptr %array.exp.next, ptr %array.init.end, align 8
+    "b", 
+    // CHECK: call void @_ZN6PrintyC1EPKc
+    // CHECK: store ptr %array.exp.next1, ptr %array.init.end, align 8
+    ({
+        if (foo()) {
+          return;
+          // CHECK: if.then:
+          // CHECK:   br i1 %arraydestroy.isempty, label %arraydestroy.done{{.*}}, label %arraydestroy.body
+        }
+        "b";
+        // CHECK: if.end:
+        // CHECK:   call void @_ZN6PrintyC1EPKc
+    })};
+  // CHECK:       arraydestroy.body:
+  // CHECK-NEXT:    %arraydestroy.elementPast = phi ptr [ %{{.*}}, %if.then ], [ %arraydestroy.element, %arraydestroy.body ]
+  // CHECK-NEXT:    %arraydestroy.element = getelementptr inbounds %struct.Printy, ptr %arraydestroy.elementPast, i64 -1
+  // CHECK-NEXT:    call void @_ZN6PrintyD1Ev(ptr noundef nonnull align 8 dereferenceable(8) %arraydestroy.element)
+  // CHECK-NEXT:    %arraydestroy.done = icmp eq ptr %arraydestroy.element, %0
+  // CHECK-NEXT:    br i1 %arraydestroy.done, label %arraydestroy.done{{.*}}, label %arraydestroy.body
+
+  // CHECK:       arraydestroy.done{{.*}}:                               ; preds = %arraydestroy.body, %if.then
+  // CHECK-NEXT:    br label %return
+}
+
+void DestroyInConditionalCleanup() {
+  // EH-LABEL: DestroyInConditionalCleanupv()
+  // NOEH-LABEL: DestroyInConditionalCleanupv()
+  struct A {
+    A() {}
+    ~A() {}
+  };
+
+  struct Value {
+    Value(A) {}
+    ~Value() {}
+  };
+
+  struct V2 {
+    Value K;
+    Value V;
+  };
+  // Verify we use conditional cleanups.
+  (void)(foo() ? V2{A(), A()} : V2{A(), A()});
+  // NOEH:   cond.true:
+  // NOEH:      call void @_ZZ27DestroyInConditionalCleanupvEN1AC1Ev
+  // NOEH:      store ptr %{{.*}}, ptr %cond-cleanup.save
+
+  // EH:   cond.true:
+  // EH:        invoke void @_ZZ27DestroyInConditionalCleanupvEN1AC1Ev
+  // EH:        store ptr %{{.*}}, ptr %cond-cleanup.save
+}
+
+void ArrayInitWithContinue() {
+  // CHECK-LABEL: @_Z21ArrayInitWithContinuev
+  // Verify that we start to emit the array destructor.
+  // CHECK: %arrayinit.endOfInit = alloca ptr, align 8
+  for (int i = 0; i < 1; ++i) {
+    Printy arr[2] = {"a", ({
+                       if (foo()) {
+                         continue;
+                       }
+                       "b";
+                     })};
+  }
+}
+
+struct [[clang::trivial_abi]] HasTrivialABI {
+  HasTrivialABI();
+  ~HasTrivialABI();
+};
+void AcceptTrivialABI(HasTrivialABI, int);
+void TrivialABI() {
+  // CHECK-LABEL: define dso_local void @_Z10TrivialABIv()
+  AcceptTrivialABI(HasTrivialABI(), ({
+                     if (foo()) return;
+                     // CHECK:      if.then:
+                     // CHECK-NEXT:   call void @_ZN13HasTrivialABID1Ev
+                     // CHECK-NEXT:   br label %return
+                     0;
+                   }));
+}
+
+namespace CleanupFlag {
+struct A {
+  A() {}
+  ~A() {}
+};
+
+struct B {
+  B(const A&) {}
+  B() {}
+  ~B() {}
+};
+
+struct S {
+  A a;
+  B b;
+};
+
+int AcceptS(S s);
+
+void Accept2(int x, int y);
+
+void InactiveNormalCleanup() {
+  // CHECK-LABEL: define {{.*}}InactiveNormalCleanupEv()
+  
+  // The first A{} below is an inactive normal cleanup which
+  // is not popped from EHStack on deactivation. This needs an
+  // "active" cleanup flag.
+
+  // CHECK: [[ACTIVE:%cleanup.isactive.*]] = alloca i1, align 1
+  // CHECK: call void [[A_CTOR:@.*AC1Ev]]
+  // CHECK: store i1 true, ptr [[ACTIVE]], align 1
+  // CHECK: call void [[A_CTOR]]
+  // CHECK: call void [[B_CTOR:@.*BC1ERKNS_1AE]]
+  // CHECK: store i1 false, ptr [[ACTIVE]], align 1
+  // CHECK: call noundef i32 [[ACCEPTS:@.*AcceptSENS_1SE]]
+  Accept2(AcceptS({.a = A{}, .b = A{}}), ({
+            if (foo()) return;
+            // CHECK: if.then:
+            // CHECK:   br label %cleanup
+            0;
+            // CHECK: if.end:
+            // CHECK:   call void [[ACCEPT2:@.*Accept2Eii]]
+            // CHECK:   br label %cleanup
+          }));
+  // CHECK: cleanup:
+  // CHECK:   call void [[S_DTOR:@.*SD1Ev]]
+  // CHECK:   call void [[A_DTOR:@.*AD1Ev]]
+  // CHECK:   %cleanup.is_active = load i1, ptr [[ACTIVE]]
+  // CHECK:   br i1 %cleanup.is_active, label %cleanup.action, label %cleanup.done
+
+  // CHECK: cleanup.action:
+  // CHECK:   call void [[A_DTOR]]
+
+  // The "active" cleanup flag is not required for unused cleanups.
+  Accept2(AcceptS({.a = A{}, .b = A{}}), 0);
+  // CHECK: cleanup.cont:
+  // CHECK:   call void [[A_CTOR]]
+  // CHECK-NOT: store i1 true
+  // CHECK:   call void [[A_CTOR]]
+  // CHECK:   call void [[B_CTOR]]
+  // CHECK-NOT: store i1 false
+  // CHECK:   call noundef i32 [[ACCEPTS]]
+  // CHECK:   call void [[ACCEPT2]]
+  // CHECK:   call void [[S_DTOR]]
+  // CHECK:   call void [[A_DTOR]]
+  // CHECK:   br label %return
+}
+}  // namespace CleanupFlag
diff --git a/clang/test/CodeGenCXX/dependent-template-alias.cpp b/clang/test/CodeGenCXX/dependent-template-alias.cpp
new file mode 100644
index 000000000000..deb243f9fc88
--- /dev/null
+++ b/clang/test/CodeGenCXX/dependent-template-alias.cpp
@@ -0,0 +1,21 @@
+// RUN: %clang_cc1 -triple x86_64-unk-unk -o - -emit-llvm -debug-info-kind=standalone -gtemplate-alias %s -gsimple-template-names=simple \
+// RUN: | FileCheck %s
+
+//// Check that -gtemplate-alias falls back to DW_TAG_typedef emission
+//// for instantiation dependent type aliases.
+
+template <int>
+using A = int;
+
+template<int I>
+struct S {
+  using AA = A<I>;
+  AA aa;
+};
+
+S<0> s;
+
+// CHECK: !DIDerivedType(tag: DW_TAG_member, name: "aa", scope: ![[#]], file: ![[#]], line: [[#]], baseType: ![[AA:[0-9]+]], size: 32)
+// CHECK: [[AA]] = !DIDerivedType(tag: DW_TAG_typedef, name: "AA", file: ![[#]], line: [[#]], baseType: ![[A:[0-9]+]])
+// CHECK: [[A]] = !DIDerivedType(tag: DW_TAG_typedef, name: "A<I>", file: ![[#]], line: [[#]], baseType: ![[int:[0-9]+]])
+// CHECK: [[int]] = !DIBasicType(name: "int", size: 32, encoding: DW_ATE_signed)
diff --git a/clang/test/CodeGenCXX/pragma-gcc-unroll.cpp b/clang/test/CodeGenCXX/pragma-gcc-unroll.cpp
index 8a94a5cc91e2..85f10fcdff14 100644
--- a/clang/test/CodeGenCXX/pragma-gcc-unroll.cpp
+++ b/clang/test/CodeGenCXX/pragma-gcc-unroll.cpp
@@ -116,6 +116,34 @@ void while_unroll_zero_test(int *List, int Length) {
   }
 }
 
+using size_t = unsigned long long;
+
+template <bool Flag>
+int value_dependent(int n) {
+  // CHECK: define {{.*}} @_Z15value_dependentILb1EEii
+  constexpr int N = 100;
+  auto init = [=]() { return Flag ? n : 0UL; };
+  auto cond = [=](size_t ix) { return Flag ? ix != 0 : ix < 10; };
+  auto iter = [=](size_t ix) {
+    return Flag ? ix & ~(1ULL << __builtin_clzll(ix)) : ix + 1;
+  };
+#pragma GCC unroll Flag ? 1 : N
+  for (size_t ix = init(); cond(ix); ix = iter(ix)) {
+    // CHECK: br label {{.*}}, !llvm.loop ![[LOOP_16:.*]]
+    n *= n;
+  }
+#pragma GCC unroll Flag ? 0 : N
+  for (size_t ix = init(); cond(ix); ix = iter(ix)) {
+    // CHECK: br label {{.*}}, !llvm.loop ![[LOOP_17:.*]]
+    n *= n;
+  }
+  return n;
+}
+
+void test_value_dependent(int n) {
+  value_dependent<true>(n);
+}
+
 // CHECK: ![[LOOP_1]] = distinct !{![[LOOP_1]], [[MP:![0-9]+]], ![[UNROLL_ENABLE:.*]]}
 // CHECK: ![[UNROLL_ENABLE]] = !{!"llvm.loop.unroll.enable"}
 // CHECK: ![[LOOP_2]] = distinct !{![[LOOP_2:.*]], ![[UNROLL_DISABLE:.*]]}
@@ -129,3 +157,5 @@ void while_unroll_zero_test(int *List, int Length) {
 // CHECK: ![[LOOP_7]] = distinct !{![[LOOP_7]], ![[UNROLL_8:.*]]}
 // CHECK: ![[LOOP_14]] = distinct !{![[LOOP_14]], [[MP]], ![[UNROLL_DISABLE:.*]]}
 // CHECK: ![[LOOP_15]] = distinct !{![[LOOP_15]], [[MP]], ![[UNROLL_DISABLE:.*]]}
+// CHECK: ![[LOOP_16]] = distinct !{![[LOOP_16]], [[MP]], ![[UNROLL_DISABLE:.*]]}
+// CHECK: ![[LOOP_17]] = distinct !{![[LOOP_17]], [[MP]], ![[UNROLL_DISABLE:.*]]}
diff --git a/clang/test/CodeGenCXX/pragma-unroll.cpp b/clang/test/CodeGenCXX/pragma-unroll.cpp
index 02d9bad7148d..6754788b7243 100644
--- a/clang/test/CodeGenCXX/pragma-unroll.cpp
+++ b/clang/test/CodeGenCXX/pragma-unroll.cpp
@@ -96,6 +96,54 @@ void template_test(double *List, int Length) {
   for_template_define_test<double>(List, Length, Value);
 }
 
+void for_unroll_zero_test(int *List, int Length) {
+  // CHECK: define {{.*}} @_Z20for_unroll_zero_testPii
+  #pragma unroll 0
+  for (int i = 0; i < Length; i++) {
+    // CHECK: br label {{.*}}, !llvm.loop ![[LOOP_14:.*]]
+    List[i] = i * 2;
+  }
+}
+
+void while_unroll_zero_test(int *List, int Length) {
+  // CHECK: define {{.*}} @_Z22while_unroll_zero_testPii
+  int i = 0;
+#pragma unroll(0)
+  while (i < Length) {
+    // CHECK: br label {{.*}}, !llvm.loop ![[LOOP_15:.*]]
+    List[i] = i * 2;
+    i++;
+  }
+}
+
+using size_t = unsigned long long;
+
+template <bool Flag>
+int value_dependent(int n) {
+  // CHECK: define {{.*}} @_Z15value_dependentILb1EEii
+  constexpr int N = 100;
+  auto init = [=]() { return Flag ? n : 0UL; };
+  auto cond = [=](size_t ix) { return Flag ? ix != 0 : ix < 10; };
+  auto iter = [=](size_t ix) {
+    return Flag ? ix & ~(1ULL << __builtin_clzll(ix)) : ix + 1;
+  };
+#pragma unroll Flag ? 1 : N
+  for (size_t ix = init(); cond(ix); ix = iter(ix)) {
+    // CHECK: br label {{.*}}, !llvm.loop ![[LOOP_16:.*]]
+    n *= n;
+  }
+#pragma unroll Flag ? 0 : N
+  for (size_t ix = init(); cond(ix); ix = iter(ix)) {
+    // CHECK: br label {{.*}}, !llvm.loop ![[LOOP_17:.*]]
+    n *= n;
+  }
+  return n;
+}
+
+void test_value_dependent(int n) {
+  value_dependent<true>(n);
+}
+
 // CHECK: ![[LOOP_1]] = distinct !{![[LOOP_1]], [[MP:![0-9]+]], ![[UNROLL_ENABLE:.*]]}
 // CHECK: ![[UNROLL_ENABLE]] = !{!"llvm.loop.unroll.enable"}
 // CHECK: ![[LOOP_2]] = distinct !{![[LOOP_2:.*]], ![[UNROLL_DISABLE:.*]]}
@@ -107,3 +155,7 @@ void template_test(double *List, int Length) {
 // CHECK: ![[LOOP_5]] = distinct !{![[LOOP_5]], ![[UNROLL_8:.*]]}
 // CHECK: ![[LOOP_6]] = distinct !{![[LOOP_6]], ![[UNROLL_8:.*]]}
 // CHECK: ![[LOOP_7]] = distinct !{![[LOOP_7]], ![[UNROLL_8:.*]]}
+// CHECK: ![[LOOP_14]] = distinct !{![[LOOP_14]], [[MP]], ![[UNROLL_DISABLE:.*]]}
+// CHECK: ![[LOOP_15]] = distinct !{![[LOOP_15]], [[MP]], ![[UNROLL_DISABLE:.*]]}
+// CHECK: ![[LOOP_16]] = distinct !{![[LOOP_16]], [[MP]], ![[UNROLL_DISABLE:.*]]}
+// CHECK: ![[LOOP_17]] = distinct !{![[LOOP_17]], [[MP]], ![[UNROLL_DISABLE:.*]]}
diff --git a/clang/test/CodeGenCoroutines/coro-await.cpp b/clang/test/CodeGenCoroutines/coro-await.cpp
index 75851d8805bb..65bfb0994688 100644
--- a/clang/test/CodeGenCoroutines/coro-await.cpp
+++ b/clang/test/CodeGenCoroutines/coro-await.cpp
@@ -73,7 +73,7 @@ extern "C" void f0() {
   // ---------------------------
   // Call coro.await.suspend
   // ---------------------------
-  // CHECK-NEXT: call void @llvm.coro.await.suspend.void(ptr %[[AWAITABLE]], ptr %[[FRAME]], ptr @__await_suspend_wrapper_f0_await)
+  // CHECK-NEXT: call void @llvm.coro.await.suspend.void(ptr %[[AWAITABLE]], ptr %[[FRAME]], ptr @f0.__await_suspend_wrapper__await)
   // -------------------------
   // Generate a suspend point:
   // -------------------------
@@ -100,7 +100,7 @@ extern "C" void f0() {
   // CHECK: call i8 @llvm.coro.suspend(token %[[FINALSP_ID]], i1 true)
 
   // Await suspend wrapper
-  // CHECK: define{{.*}} @__await_suspend_wrapper_f0_await(ptr {{[^,]*}} %[[AWAITABLE_ARG:.+]], ptr {{[^,]*}} %[[FRAME_ARG:.+]])
+  // CHECK: define{{.*}} @f0.__await_suspend_wrapper__await(ptr {{[^,]*}} %[[AWAITABLE_ARG:.+]], ptr {{[^,]*}} %[[FRAME_ARG:.+]])
   // CHECK: store ptr %[[AWAITABLE_ARG]], ptr %[[AWAITABLE_TMP:.+]],
   // CHECK: store ptr %[[FRAME_ARG]], ptr %[[FRAME_TMP:.+]],
   // CHECK: %[[AWAITABLE:.+]] = load ptr, ptr %[[AWAITABLE_TMP]]
@@ -149,7 +149,7 @@ extern "C" void f1(int) {
   // ---------------------------
   // Call coro.await.suspend
   // ---------------------------
-  // CHECK-NEXT: %[[YES:.+]] = call i1 @llvm.coro.await.suspend.bool(ptr %[[AWAITABLE]], ptr %[[FRAME]], ptr @__await_suspend_wrapper_f1_yield)
+  // CHECK-NEXT: %[[YES:.+]] = call i1 @llvm.coro.await.suspend.bool(ptr %[[AWAITABLE]], ptr %[[FRAME]], ptr @f1.__await_suspend_wrapper__yield)
   // -------------------------------------------
   // See if await_suspend decided not to suspend
   // -------------------------------------------
@@ -162,7 +162,7 @@ extern "C" void f1(int) {
   // CHECK:     call void @_ZN13suspend_maybe12await_resumeEv(ptr {{[^,]*}} %[[AWAITABLE]])
 
   // Await suspend wrapper
-  // CHECK: define {{.*}} i1 @__await_suspend_wrapper_f1_yield(ptr {{[^,]*}} %[[AWAITABLE_ARG:.+]], ptr {{[^,]*}} %[[FRAME_ARG:.+]])
+  // CHECK: define {{.*}} i1 @f1.__await_suspend_wrapper__yield(ptr {{[^,]*}} %[[AWAITABLE_ARG:.+]], ptr {{[^,]*}} %[[FRAME_ARG:.+]])
   // CHECK: store ptr %[[AWAITABLE_ARG]], ptr %[[AWAITABLE_TMP:.+]],
   // CHECK: store ptr %[[FRAME_ARG]], ptr %[[FRAME_TMP:.+]],
   // CHECK: %[[AWAITABLE:.+]] = load ptr, ptr %[[AWAITABLE_TMP]]
@@ -370,7 +370,7 @@ extern "C" void TestTailcall() {
   // ---------------------------
   // Call coro.await.suspend
   // ---------------------------
-  // CHECK-NEXT: %[[RESUMED:.+]] = call ptr @llvm.coro.await.suspend.handle(ptr %[[AWAITABLE]], ptr %[[FRAME]], ptr @__await_suspend_wrapper_TestTailcall_await)
+  // CHECK-NEXT: %[[RESUMED:.+]] = call ptr @llvm.coro.await.suspend.handle(ptr %[[AWAITABLE]], ptr %[[FRAME]], ptr @TestTailcall.__await_suspend_wrapper__await)
   // CHECK-NEXT: call void @llvm.coro.resume(ptr %[[RESUMED]])
   // CHECK-NEXT: %[[OUTCOME:.+]] = call i8 @llvm.coro.suspend(token %[[SUSPEND_ID]], i1 false)
   // CHECK-NEXT: switch i8 %[[OUTCOME]], label %[[RET_BB:.+]] [
@@ -379,7 +379,7 @@ extern "C" void TestTailcall() {
   // CHECK-NEXT: ]
 
   // Await suspend wrapper
-  // CHECK: define {{.*}} ptr @__await_suspend_wrapper_TestTailcall_await(ptr {{[^,]*}} %[[AWAITABLE_ARG:.+]], ptr {{[^,]*}} %[[FRAME_ARG:.+]])
+  // CHECK: define {{.*}} ptr @TestTailcall.__await_suspend_wrapper__await(ptr {{[^,]*}} %[[AWAITABLE_ARG:.+]], ptr {{[^,]*}} %[[FRAME_ARG:.+]])
   // CHECK: store ptr %[[AWAITABLE_ARG]], ptr %[[AWAITABLE_TMP:.+]],
   // CHECK: store ptr %[[FRAME_ARG]], ptr %[[FRAME_TMP:.+]],
   // CHECK: %[[AWAITABLE:.+]] = load ptr, ptr %[[AWAITABLE_TMP]]
diff --git a/clang/test/CodeGenCoroutines/coro-dwarf.cpp b/clang/test/CodeGenCoroutines/coro-dwarf.cpp
index 2c9c827e6753..f951b63dc117 100644
--- a/clang/test/CodeGenCoroutines/coro-dwarf.cpp
+++ b/clang/test/CodeGenCoroutines/coro-dwarf.cpp
@@ -71,14 +71,14 @@ void f_coro(int val, MoveOnly moParam, MoveAndCopy mcParam) {
 // CHECK: !{{[0-9]+}} = !DILocalVariable(name: "mcParam", arg: 3, scope: ![[SP]], file: !{{[0-9]+}}, line: {{[0-9]+}}, type: !{{[0-9]+}})
 // CHECK: !{{[0-9]+}} = !DILocalVariable(name: "__promise",
 
-// CHECK: !{{[0-9]+}} = distinct !DISubprogram(linkageName: "__await_suspend_wrapper__Z6f_coroi8MoveOnly11MoveAndCopy_init"
+// CHECK: !{{[0-9]+}} = distinct !DISubprogram(linkageName: "_Z6f_coroi8MoveOnly11MoveAndCopy.__await_suspend_wrapper__init"
 // CHECK-NEXT: !{{[0-9]+}} = !DIFile
 // CHECK-NEXT: !{{[0-9]+}} = !DISubroutineType
 // CHECK-NEXT: !{{[0-9]+}} = !DILocalVariable(arg: 1,
 // CHECK-NEXT: !{{[0-9]+}} = !DILocation
 // CHECK-NEXT: !{{[0-9]+}} = !DILocalVariable(arg: 2,
 
-// CHECK: !{{[0-9]+}} = distinct !DISubprogram(linkageName: "__await_suspend_wrapper__Z6f_coroi8MoveOnly11MoveAndCopy_final"
+// CHECK: !{{[0-9]+}} = distinct !DISubprogram(linkageName: "_Z6f_coroi8MoveOnly11MoveAndCopy.__await_suspend_wrapper__final"
 // CHECK-NEXT: !{{[0-9]+}} = !DILocalVariable(arg: 1,
 // CHECK-NEXT: !{{[0-9]+}} = !DILocation
 // CHECK-NEXT: !{{[0-9]+}} = !DILocalVariable(arg: 2,
diff --git a/clang/test/CodeGenCoroutines/coro-suspend-cleanups.cpp b/clang/test/CodeGenCoroutines/coro-suspend-cleanups.cpp
new file mode 100644
index 000000000000..06cc2069dbe9
--- /dev/null
+++ b/clang/test/CodeGenCoroutines/coro-suspend-cleanups.cpp
@@ -0,0 +1,93 @@
+// RUN: %clang_cc1 --std=c++20 -triple x86_64-linux-gnu -emit-llvm %s -o - | FileCheck %s
+
+#include "Inputs/coroutine.h"
+
+struct Printy {
+  Printy(const char *name) : name(name) {}
+  ~Printy() {}
+  const char *name;
+};
+
+struct coroutine {
+  struct promise_type;
+  std::coroutine_handle<promise_type> handle;
+  ~coroutine() {
+    if (handle) handle.destroy();
+  }
+};
+
+struct coroutine::promise_type {
+  coroutine get_return_object() {
+    return {std::coroutine_handle<promise_type>::from_promise(*this)};
+  }
+  std::suspend_never initial_suspend() noexcept { return {}; }
+  std::suspend_always final_suspend() noexcept { return {}; }
+  void return_void() {}
+  void unhandled_exception() {}
+};
+
+struct Awaiter : std::suspend_always {
+  Printy await_resume() { return {"awaited"}; }
+};
+
+int foo() { return 2; }
+
+coroutine ArrayInitCoro() {
+  // Verify that:
+  //  - We do the necessary stores for array cleanups.
+  //  - Array cleanups are called by await.cleanup.
+  //  - We activate the cleanup after the first element and deactivate it in await.ready (see cleanup.isactive).
+
+  // CHECK-LABEL: define dso_local void @_Z13ArrayInitCorov
+  // CHECK: %arrayinit.endOfInit = alloca ptr, align 8
+  // CHECK: %cleanup.isactive = alloca i1, align 1
+  Printy arr[2] = {
+    Printy("a"),
+    // CHECK:       %arrayinit.begin = getelementptr inbounds [2 x %struct.Printy], ptr %arr.reload.addr, i64 0, i64 0
+    // CHECK-NEXT:  %arrayinit.begin.spill.addr = getelementptr inbounds %_Z13ArrayInitCorov.Frame, ptr %0, i32 0, i32 10
+    // CHECK-NEXT:  store ptr %arrayinit.begin, ptr %arrayinit.begin.spill.addr, align 8
+    // CHECK-NEXT:  store i1 true, ptr %cleanup.isactive.reload.addr, align 1
+    // CHECK-NEXT:  store ptr %arrayinit.begin, ptr %arrayinit.endOfInit.reload.addr, align 8
+    // CHECK-NEXT:  call void @_ZN6PrintyC1EPKc(ptr noundef nonnull align 8 dereferenceable(8) %arrayinit.begin, ptr noundef @.str)
+    // CHECK-NEXT:  %arrayinit.element = getelementptr inbounds %struct.Printy, ptr %arrayinit.begin, i64 1
+    // CHECK-NEXT:  %arrayinit.element.spill.addr = getelementptr inbounds %_Z13ArrayInitCorov.Frame, ptr %0, i32 0, i32 11
+    // CHECK-NEXT:  store ptr %arrayinit.element, ptr %arrayinit.element.spill.addr, align 8
+    // CHECK-NEXT:  store ptr %arrayinit.element, ptr %arrayinit.endOfInit.reload.addr, align 8
+    co_await Awaiter{}
+    // CHECK-NEXT:  @_ZNSt14suspend_always11await_readyEv
+    // CHECK-NEXT:  br i1 %{{.+}}, label %await.ready, label %CoroSave30
+  };
+  // CHECK:       await.cleanup:                                    ; preds = %AfterCoroSuspend{{.*}}
+  // CHECK-NEXT:    br label %cleanup{{.*}}.from.await.cleanup
+
+  // CHECK:       cleanup{{.*}}.from.await.cleanup:                      ; preds = %await.cleanup
+  // CHECK:         br label %cleanup{{.*}}
+
+  // CHECK:       await.ready:
+  // CHECK-NEXT:    %arrayinit.element.reload.addr = getelementptr inbounds %_Z13ArrayInitCorov.Frame, ptr %0, i32 0, i32 11
+  // CHECK-NEXT:    %arrayinit.element.reload = load ptr, ptr %arrayinit.element.reload.addr, align 8
+  // CHECK-NEXT:    call void @_ZN7Awaiter12await_resumeEv
+  // CHECK-NEXT:    store i1 false, ptr %cleanup.isactive.reload.addr, align 1
+  // CHECK-NEXT:    br label %cleanup{{.*}}.from.await.ready
+
+  // CHECK:       cleanup{{.*}}:                                         ; preds = %cleanup{{.*}}.from.await.ready, %cleanup{{.*}}.from.await.cleanup
+  // CHECK:         %cleanup.is_active = load i1, ptr %cleanup.isactive.reload.addr, align 1
+  // CHECK-NEXT:    br i1 %cleanup.is_active, label %cleanup.action, label %cleanup.done
+
+  // CHECK:       cleanup.action:
+  // CHECK:         %arraydestroy.isempty = icmp eq ptr %arrayinit.begin.reload{{.*}}, %{{.*}}
+  // CHECK-NEXT:    br i1 %arraydestroy.isempty, label %arraydestroy.done{{.*}}, label %arraydestroy.body.from.cleanup.action
+  // Ignore rest of the array cleanup.
+}
+
+coroutine ArrayInitWithCoReturn() {
+  // CHECK-LABEL: define dso_local void @_Z21ArrayInitWithCoReturnv
+  // Verify that we start to emit the array destructor.
+  // CHECK: %arrayinit.endOfInit = alloca ptr, align 8
+  Printy arr[2] = {"a", ({
+                      if (foo()) {
+                        co_return;
+                      }
+                      "b";
+                    })};
+}
diff --git a/clang/test/CodeGenCoroutines/pr65054.cpp b/clang/test/CodeGenCoroutines/pr65054.cpp
index 7af9c04fca18..2d8b6dfe18d5 100644
--- a/clang/test/CodeGenCoroutines/pr65054.cpp
+++ b/clang/test/CodeGenCoroutines/pr65054.cpp
@@ -48,6 +48,6 @@ MyTask FooBar() {
 }
 
 // CHECK-O0: define{{.*}}@_Z6FooBarv.resume
-// CHECK-O0: call{{.*}}@__await_suspend_wrapper__Z6FooBarv_await(
+// CHECK-O0: call{{.*}}@_Z6FooBarv.__await_suspend_wrapper__await(
 // CHECK-O0-NOT: store
 // CHECK-O0: ret void
diff --git a/clang/test/CodeGenObjC/arc-blocks-exceptions.m b/clang/test/CodeGenObjC/arc-blocks-exceptions.m
index 821b818d4027..54b043d8ea07 100644
--- a/clang/test/CodeGenObjC/arc-blocks-exceptions.m
+++ b/clang/test/CodeGenObjC/arc-blocks-exceptions.m
@@ -5,17 +5,22 @@ void test1(_Bool c) {
   __weak id weakId = 0;
   test1_fn(c ? ^{ (void)weakId; } : 0);
 
-  // CHECK: [[CLEANUP_COND:%.*]] = alloca i1
-  // CHECK-NEXT: [[CLEANUP_SAVE:%.*]] = alloca ptr
+  // CHECK: [[CLEANUP_SAVE:%cond-cleanup.save.*]] = alloca ptr
+  // CHECK-NEXT: [[CLEANUP_COND:%.*]] = alloca i1
+  // CHECK-NEXT: [[CLEANUP_COND1:%.*]] = alloca i1
 
-  // CHECK: store i1 true, ptr [[CLEANUP_COND]]
-  // CHECK-NEXT: store ptr {{.*}}, ptr [[CLEANUP_SAVE]]
+  // CHECK: store i1 false, ptr [[CLEANUP_COND]]
+  // CHECK-NEXT: store i1 false, ptr [[CLEANUP_COND1]]
+
+  // CHECK: store ptr {{.*}}, ptr [[CLEANUP_SAVE]]
+  // CHECK-NEXT: store i1 true, ptr [[CLEANUP_COND]]
+  // CHECK-NEXT: store i1 true, ptr [[CLEANUP_COND1]]
 
   // CHECK: invoke void @test1_fn(
   // CHECK-NEXT: to label %[[INVOKE_CONT:.*]] unwind label %[[LANDING_PAD_LAB:.*]]
 
   // CHECK: [[INVOKE_CONT]]:
-  // CHECK-NEXT: [[LOAD:%.*]] = load i1, ptr [[CLEANUP_COND]]
+  // CHECK-NEXT: [[LOAD:%.*]] = load i1, ptr [[CLEANUP_COND1]]
   // CHECK-NEXT: br i1 [[LOAD]], label %[[END_OF_SCOPE_LAB:.*]], label
 
   // CHECK: [[END_OF_SCOPE_LAB]]:
diff --git a/clang/test/CodeGenObjC/arc-blocks.m b/clang/test/CodeGenObjC/arc-blocks.m
index 105a72b4af1e..f718e8bbf9a6 100644
--- a/clang/test/CodeGenObjC/arc-blocks.m
+++ b/clang/test/CodeGenObjC/arc-blocks.m
@@ -445,8 +445,8 @@ void test13(id x) {
   // CHECK:      [[X:%.*]] = alloca ptr, align 8
   // CHECK-NEXT: [[B:%.*]] = alloca ptr, align 8
   // CHECK-NEXT: [[BLOCK:%.*]] = alloca [[BLOCK_T:.*]], align 8
-  // CHECK-NEXT: [[CLEANUP_ACTIVE:%.*]] = alloca i1
   // CHECK-NEXT: [[COND_CLEANUP_SAVE:%.*]] = alloca ptr,
+  // CHECK-NEXT: [[CLEANUP_ACTIVE:%.*]] = alloca i1
   // CHECK-NEXT: [[T0:%.*]] = call ptr @llvm.objc.retain(ptr {{%.*}})
   // CHECK-NEXT: store ptr [[T0]], ptr [[X]], align 8
   // CHECK-NEXT: call void @llvm.lifetime.start.p0(i64 8, ptr [[B]])
@@ -460,8 +460,8 @@ void test13(id x) {
   // CHECK-NEXT: [[T0:%.*]] = load ptr, ptr [[X]], align 8
   // CHECK-NEXT: [[T1:%.*]] = call ptr @llvm.objc.retain(ptr [[T0]])
   // CHECK-NEXT: store ptr [[T1]], ptr [[CAPTURE]], align 8
-  // CHECK-NEXT: store i1 true, ptr [[CLEANUP_ACTIVE]]
   // CHECK-NEXT: store ptr [[CAPTURE]], ptr [[COND_CLEANUP_SAVE]], align 8
+  // CHECK-NEXT: store i1 true, ptr [[CLEANUP_ACTIVE]]
   // CHECK-NEXT: br label
   // CHECK:      br label
   // CHECK:      [[T0:%.*]] = phi ptr
diff --git a/clang/test/CodeGenObjCXX/msabi-stret-arm64.mm b/clang/test/CodeGenObjCXX/msabi-stret-arm64.mm
new file mode 100644
index 000000000000..3bbdbebc5cb5
--- /dev/null
+++ b/clang/test/CodeGenObjCXX/msabi-stret-arm64.mm
@@ -0,0 +1,77 @@
+// RUN: %clang_cc1 -triple aarch64-pc-windows-msvc -fobjc-runtime=gnustep-2.2 -fobjc-dispatch-method=non-legacy -emit-llvm -o - %s | FileCheck %s
+
+// Pass and return for type size <= 8 bytes.
+struct S1 {
+  int a[2];
+};
+
+// Pass and return hfa <= 8 bytes
+struct F1 {
+  float a[2];
+};
+
+// Pass and return for type size > 16 bytes.
+struct S2 {
+  int a[5];
+};
+
+// Pass and return aggregate (of size < 16 bytes) with non-trivial destructor.
+// Sret and inreg: Returned in x0
+struct S3 {
+  int a[3];
+  ~S3();
+};
+S3::~S3() {
+}
+
+
+@interface MsgTest { id isa; } @end
+@implementation MsgTest 
+- (S1) smallS1 {
+  S1 x;
+  x.a[0] = 0;
+  x.a[1] = 1;
+  return x;
+
+}
+- (F1) smallF1 {
+  F1 x;
+  x.a[0] = 0.2f;
+  x.a[1] = 0.5f;
+  return x;
+}
+- (S2) stretS2 {
+  S2 x;
+  for (int i = 0; i < 5; i++) {
+    x.a[i] = i;
+  }
+  return x;
+}
+- (S3) stretInRegS3 {
+  S3 x;
+  for (int i = 0; i < 3; i++) {
+    x.a[i] = i;
+  }
+  return x;
+}
++ (S3) msgTestStretInRegS3 {
+  S3 x;
+  for (int i = 0; i < 3; i++) {
+    x.a[i] = i;
+  }
+  return x;
+}
+@end
+
+void test0(MsgTest *t) {
+    // CHECK: call {{.*}} @objc_msgSend
+    S1 ret = [t smallS1];
+    // CHECK: call {{.*}} @objc_msgSend
+    F1 ret2 = [t smallF1];
+    // CHECK: call {{.*}} @objc_msgSend_stret
+    S2 ret3 = [t stretS2];
+    // CHECK: call {{.*}} @objc_msgSend_stret2
+    S3 ret4 = [t stretInRegS3];
+    // CHECK: call {{.*}} @objc_msgSend_stret2
+    S3 ret5 = [MsgTest msgTestStretInRegS3];
+}
diff --git a/clang/test/CodeGenOpenCL/builtins-generic-amdgcn.cl b/clang/test/CodeGenOpenCL/builtins-generic-amdgcn.cl
index b093fcbf7d98..37bea1ff9330 100644
--- a/clang/test/CodeGenOpenCL/builtins-generic-amdgcn.cl
+++ b/clang/test/CodeGenOpenCL/builtins-generic-amdgcn.cl
@@ -4,14 +4,14 @@
 #pragma OPENCL EXTENSION cl_khr_fp16 : enable
 
 // CHECK-LABEL: @test_builtin_clz(
-// CHECK: tail call i32 @llvm.ctlz.i32(i32 %a, i1 true)
+// CHECK: tail call range(i32 0, 33) i32 @llvm.ctlz.i32(i32 %a, i1 true)
 void test_builtin_clz(global int* out, int a)
 {
   *out = __builtin_clz(a);
 }
 
 // CHECK-LABEL: @test_builtin_clzl(
-// CHECK: tail call i64 @llvm.ctlz.i64(i64 %a, i1 true)
+// CHECK: tail call range(i64 0, 65) i64 @llvm.ctlz.i64(i64 %a, i1 true)
 void test_builtin_clzl(global long* out, long a)
 {
   *out = __builtin_clzl(a);
diff --git a/clang/test/Driver/aarch64-mcpu.c b/clang/test/Driver/aarch64-mcpu.c
index 77ba43122b24..ad4a5f9ac6fb 100644
--- a/clang/test/Driver/aarch64-mcpu.c
+++ b/clang/test/Driver/aarch64-mcpu.c
@@ -64,10 +64,16 @@
 // NEOVERSE-V1: "-cc1"{{.*}} "-triple" "aarch64{{.*}}" "-target-cpu" "neoverse-v1"
 // RUN: %clang --target=aarch64 -mcpu=neoverse-v2  -### -c %s 2>&1 | FileCheck -check-prefix=NEOVERSE-V2 %s
 // NEOVERSE-V2: "-cc1"{{.*}} "-triple" "aarch64{{.*}}" "-target-cpu" "neoverse-v2"
+// RUN: %clang --target=aarch64 -mcpu=neoverse-v3  -### -c %s 2>&1 | FileCheck -check-prefix=NEOVERSE-V3 %s
+// NEOVERSE-V3: "-cc1"{{.*}} "-triple" "aarch64{{.*}}" "-target-cpu" "neoverse-v3"
+// RUN: %clang --target=aarch64 -mcpu=neoverse-v3ae  -### -c %s 2>&1 | FileCheck -check-prefix=NEOVERSE-V3AE %s
+// NEOVERSE-V3AE: "-cc1"{{.*}} "-triple" "aarch64{{.*}}" "-target-cpu" "neoverse-v3ae"
 // RUN: %clang --target=aarch64 -mcpu=neoverse-n1 -### -c %s 2>&1 | FileCheck -check-prefix=NEOVERSE-N1 %s
 // NEOVERSE-N1: "-cc1"{{.*}} "-triple" "aarch64{{.*}}" "-target-cpu" "neoverse-n1"
 // RUN: %clang --target=aarch64 -mcpu=neoverse-n2 -### -c %s 2>&1 | FileCheck -check-prefix=NEOVERSE-N2 %s
 // NEOVERSE-N2: "-cc1"{{.*}} "-triple" "aarch64{{.*}}" "-target-cpu" "neoverse-n2"
+// RUN: %clang --target=aarch64 -mcpu=neoverse-n3 -### -c %s 2>&1 | FileCheck -check-prefix=NEOVERSE-N3 %s
+// NEOVERSE-N3: "-cc1"{{.*}} "-triple" "aarch64{{.*}}" "-target-cpu" "neoverse-n3"
 // RUN: %clang --target=aarch64 -mcpu=neoverse-512tvb -### -c %s 2>&1 | FileCheck -check-prefix=NEOVERSE-512TVB %s
 // NEOVERSE-512TVB: "-cc1"{{.*}} "-triple" "aarch64{{.*}}" "-target-cpu" "neoverse-512tvb"
 // RUN: %clang --target=aarch64 -mcpu=cortex-a520 -### -c %s 2>&1 | FileCheck -check-prefix=CORTEX-A520 %s
diff --git a/clang/test/Driver/amdgpu-toolchain.c b/clang/test/Driver/amdgpu-toolchain.c
index 4300e7e9f667..8ab6a0713147 100644
--- a/clang/test/Driver/amdgpu-toolchain.c
+++ b/clang/test/Driver/amdgpu-toolchain.c
@@ -24,3 +24,7 @@
 // RUN:   -L. -fconvergent-functions %s 2>&1 | FileCheck -check-prefix=MCPU %s
 // LTO: clang{{.*}} "-flto=full"{{.*}}"-fconvergent-functions"
 // MCPU: ld.lld{{.*}}"-L."{{.*}}"-plugin-opt=mcpu=gfx906"
+
+// RUN: %clang -### --target=amdgcn-amd-amdhsa -mcpu=gfx906 -nogpulib \
+// RUN:   -fuse-ld=ld %s 2>&1 | FileCheck -check-prefixes=LD %s
+// LD: ld.lld
diff --git a/clang/test/Driver/cl-options.c b/clang/test/Driver/cl-options.c
index 7731300ae9f5..75f49deca065 100644
--- a/clang/test/Driver/cl-options.c
+++ b/clang/test/Driver/cl-options.c
@@ -70,12 +70,16 @@
 // fsanitize_address: -fsanitize=address
 
 // RUN: %clang_cl -### /FA -fprofile-instr-generate -- %s 2>&1 | FileCheck -check-prefix=CHECK-PROFILE-INSTR-GENERATE %s
+// RUN: %clang_cl -### /FA -fprofile-instr-generate -fno-rtlib-defaultlib -frtlib-defaultlib -- %s 2>&1 | FileCheck -check-prefix=CHECK-PROFILE-INSTR-GENERATE %s
 // RUN: %clang_cl -### /FA -fprofile-instr-generate=/tmp/somefile.profraw -- %s 2>&1 | FileCheck -check-prefix=CHECK-PROFILE-INSTR-GENERATE-FILE %s
 // RUN: %clang_cl -### /FAcsu -fprofile-instr-generate -- %s 2>&1 | FileCheck -check-prefix=CHECK-PROFILE-INSTR-GENERATE %s
 // RUN: %clang_cl -### /FAcsu -fprofile-instr-generate=/tmp/somefile.profraw -- %s 2>&1 | FileCheck -check-prefix=CHECK-PROFILE-INSTR-GENERATE-FILE %s
 // CHECK-PROFILE-INSTR-GENERATE: "-fprofile-instrument=clang" "--dependent-lib=clang_rt.profile{{[^"]*}}.lib"
 // CHECK-PROFILE-INSTR-GENERATE-FILE: "-fprofile-instrument-path=/tmp/somefile.profraw"
 
+// RUN: %clang_cl -### /FA -fprofile-instr-generate -fno-rtlib-defaultlib -- %s 2>&1 | FileCheck -check-prefix=CHECK-PROFILE-INSTR-GENERATE-NODEF %s
+// CHECK-PROFILE-INSTR-GENERATE-NODEF-NOT: "--dependent-lib=clang_rt.profile{{[^"]*}}.lib"
+
 // RUN: %clang_cl -### /FA -fprofile-generate -- %s 2>&1 | FileCheck -check-prefix=CHECK-PROFILE-GENERATE %s
 // RUN: %clang_cl -### /FAcsu -fprofile-generate -- %s 2>&1 | FileCheck -check-prefix=CHECK-PROFILE-GENERATE %s
 // CHECK-PROFILE-GENERATE: "-fprofile-instrument=llvm" "--dependent-lib=clang_rt.profile{{[^"]*}}.lib"
diff --git a/clang/test/Driver/claim-unused.c b/clang/test/Driver/claim-unused.c
deleted file mode 100644
index c7b798934b3c..000000000000
--- a/clang/test/Driver/claim-unused.c
+++ /dev/null
@@ -1,3 +0,0 @@
-// RUN: touch %t.o
-// RUN: %clang --param ssp-buffer-size=1 %t.o -### 2>&1 | FileCheck %s
-// CHECK-NOT: warning: argument unused during compilation: '--param ssp-buffer-size=1'
diff --git a/clang/test/Driver/default-denormal-fp-math.c b/clang/test/Driver/default-denormal-fp-math.c
index 5f87e151df49..c04ad5c08b8d 100644
--- a/clang/test/Driver/default-denormal-fp-math.c
+++ b/clang/test/Driver/default-denormal-fp-math.c
@@ -3,15 +3,6 @@
 
 // RUN: %clang -### -target x86_64-unknown-linux-gnu --sysroot=%S/Inputs/basic_linux_tree -c %s -v 2>&1 | FileCheck -check-prefix=CHECK-IEEE %s
 
-// crtfastmath enables ftz and daz
-// RUN: %clang -### -target x86_64-unknown-linux-gnu -ffast-math --sysroot=%S/Inputs/basic_linux_tree -c %s -v 2>&1 | FileCheck -check-prefix=CHECK-PRESERVESIGN %s
-
-// crt not linked in with nostartfiles
-// RUN: %clang -### -target x86_64-unknown-linux-gnu -ffast-math -nostartfiles --sysroot=%S/Inputs/basic_linux_tree -c %s -v 2>&1 | FileCheck -check-prefix=CHECK-IEEE %s
-
-// If there's no crtfastmath, don't assume ftz/daz
-// RUN: %clang -### -target x86_64-unknown-linux-gnu -ffast-math --sysroot=/dev/null -c %s -v 2>&1 | FileCheck -check-prefix=CHECK-IEEE %s
-
 // RUN: %clang -### -target x86_64-scei-ps4 -c %s -v 2>&1 | FileCheck -check-prefix=CHECK-PRESERVESIGN %s
 
 // Flag omitted for default
diff --git a/clang/test/Driver/fast-math.c b/clang/test/Driver/fast-math.c
index 882e81fd14d3..b07d5732932c 100644
--- a/clang/test/Driver/fast-math.c
+++ b/clang/test/Driver/fast-math.c
@@ -7,324 +7,309 @@
 // Both of them use gcc driver for as.
 //
 // RUN: %clang -### -fno-honor-infinities -c %s 2>&1 \
-// RUN:   | FileCheck --check-prefix=CHECK-NO-INFS %s
+// RUN:   | FileCheck --check-prefixes=CHECK,NINF,NO-NNAN,NO-FINITE-ONLY %s
 // infinites [sic] is a supported alternative spelling of infinities.
 // RUN: %clang -### -fno-honor-infinites -c %s 2>&1 \
-// RUN:   | FileCheck --check-prefix=CHECK-NO-INFS %s
-// CHECK-NO-INFS: "-cc1"
-// CHECK-NO-INFS: "-menable-no-infs"
+// RUN:   | FileCheck --check-prefixes=CHECK,NINF,NO-NNAN,NO-FINITE-ONLY %s
 //
 // RUN: %clang -### -fno-fast-math -fno-honor-infinities -c %s 2>&1 \
-// RUN:   | FileCheck --check-prefix=CHECK-NO-FAST-MATH-NO-INFS %s
-// CHECK-NO-FAST-MATH-NO-INFS: "-cc1"
-// CHECK-NO-FAST-MATH-NO-INFS: "-menable-no-infs"
+// RUN:   | FileCheck --check-prefixes=CHECK,NO-FAST,NINF,NO-NNAN,NO-FINITE-ONLY %s
 //
 // RUN: %clang -### -fno-honor-infinities -fno-fast-math -c %s 2>&1 \
-// RUN:   | FileCheck --check-prefix=CHECK-NO-INFS-NO-FAST-MATH %s
-// CHECK-NO-INFS-NO-FAST-MATH: "-cc1"
-// CHECK-NO-INFS-NO-FAST-MATH-NOT: "-menable-no-infs"
+// RUN:   | FileCheck --check-prefixes=CHECK,NO-FAST,NO-NINF,NO-NNAN,NO-FINITE-ONLY %s
 //
 // RUN: %clang -### -fno-signed-zeros -c %s 2>&1 \
-// RUN:   | FileCheck --check-prefix=CHECK-NO-SIGNED-ZEROS %s
-// CHECK-NO-SIGNED-ZEROS: "-cc1"
-// CHECK-NO-SIGNED-ZEROS: "-fno-signed-zeros"
+// RUN:   | FileCheck --check-prefixes=CHECK,NSZ,NOROUNDING %s
 //
 // RUN: %clang -### -fno-fast-math -fno-signed-zeros -c %s 2>&1 \
-// RUN:   | FileCheck --check-prefix=CHECK-NO-FAST-MATH-NO-SIGNED-ZEROS %s
-// CHECK-NO-FAST-MATH-NO-SIGNED-ZEROS: "-cc1"
-// CHECK-NO-FAST-MATH-NO-SIGNED-ZEROS: "-fno-signed-zeros"
+// RUN:   | FileCheck --check-prefixes=CHECK,NO-FAST,NSZ %s
 //
 // RUN: %clang -### -fno-signed-zeros -fno-fast-math -c %s 2>&1 \
-// RUN:   | FileCheck --check-prefix=CHECK-NO-SIGNED-ZEROS-NO-FAST-MATH %s
-// CHECK-NO-SIGNED-ZEROS-NO-FAST-MATH: "-cc1"
-// CHECK-NO-SIGNED-ZEROS-NO-FAST-MATH-NOT: "-fno-signed-zeros"
+// RUN:   | FileCheck --check-prefixes=CHECK,NO-FAST,NO-NSZ,NOROUNDING %s
 //
 // RUN: %clang -### -freciprocal-math -c %s 2>&1 \
-// RUN:   | FileCheck --check-prefix=CHECK-RECIPROCAL-MATH %s
-// CHECK-RECIPROCAL-MATH: "-cc1"
-// CHECK-RECIPROCAL-MATH: "-freciprocal-math"
+// RUN:   | FileCheck --check-prefixes=CHECK,ARCP,NOROUNDING %s
 //
 // RUN: %clang -### -fno-fast-math -freciprocal-math -c %s 2>&1 \
-// RUN:   | FileCheck --check-prefix=CHECK-NO-FAST-MATH-RECIPROCAL-MATH %s
-// CHECK-NO-FAST-MATH-RECIPROCAL-MATH: "-cc1"
-// CHECK-NO-FAST-MATH-RECIPROCAL-MATH: "-freciprocal-math"
+// RUN:   | FileCheck --check-prefixes=CHECK,ARCP,NOROUNDING %s
 //
 // RUN: %clang -### -freciprocal-math -fno-fast-math -c %s 2>&1 \
-// RUN:   | FileCheck --check-prefix=CHECK-RECIPROCAL-MATH-NO-FAST-MATH %s
-// CHECK-RECIPROCAL-MATH-NO-FAST-MATH: "-cc1"
-// CHECK-RECIPROCAL-MATH-NO-FAST-MATH-NOT: "-freciprocal-math"
+// RUN:   | FileCheck --check-prefixes=CHECK,NO-ARCP,NOROUNDING %s
 //
 // RUN: %clang -### -fno-honor-nans -c %s 2>&1 \
-// RUN:   | FileCheck --check-prefix=CHECK-NO-NANS %s
-// CHECK-NO-NANS: "-cc1"
-// CHECK-NO-NANS: "-menable-no-nans"
+// RUN:   | FileCheck --check-prefixes=CHECK,NNAN,NO-NINF,NO-FINITE-ONLY,NOROUNDING %s
 //
 // RUN: %clang -### -fno-fast-math -fno-honor-nans -c %s 2>&1 \
-// RUN:   | FileCheck --check-prefix=CHECK-NO-FAST-MATH-NO-NANS %s
-// CHECK-NO-FAST-MATH-NO-NANS: "-cc1"
-// CHECK-NO-FAST-MATH-NO-NANS: "-menable-no-nans"
+// RUN:   | FileCheck --check-prefixes=CHECK,NO-FAST,NNAN,NO-NINF,NO-FINITE-ONLY,NOROUNDING %s
 //
 // RUN: %clang -### -fno-honor-nans -fno-fast-math -c %s 2>&1 \
-// RUN:   | FileCheck --check-prefix=CHECK-NO-NANS-NO-FAST-MATH %s
-// CHECK-NO-NANS-NO-FAST-MATH: "-cc1"
-// CHECK-NO-NANS-NO-FAST-MATH-NOT: "-menable-no-nans"
+// RUN:   | FileCheck --check-prefixes=CHECK,NO-FAST,NO-NNAN,NO-NINF,NO-FINITE-ONLY,NOROUNDING %s
 //
 // RUN: %clang -### -ffast-math -fno-approx-func -c %s 2>&1 \
-// RUN:   | FileCheck --check-prefix=CHECK-FAST-MATH-NO-APPROX-FUNC %s
-// CHECK-FAST-MATH-NO-APPROX-FUNC:     "-cc1"
-// CHECK-FAST-MATH-NO-APPROX-FUNC:     "-menable-no-infs"
-// CHECK-FAST-MATH-NO-APPROX-FUNC:     "-menable-no-nans"
-// CHECK-FAST-MATH-NO-APPROX-FUNC:     "-fno-signed-zeros"
-// CHECK-FAST-MATH-NO-APPROX-FUNC:     "-mreassociate"
-// CHECK-FAST-MATH-NO-APPROX-FUNC:     "-freciprocal-math"
-// CHECK-FAST-MATH-NO-APPROX-FUNC:     "-ffp-contract=fast"
-// CHECK-FAST-MATH-NO-APPROX-FUNC-NOT: "-ffast-math"
-// CHECK-FAST-MATH-NO-APPROX-FUNC-NOT: "-fapprox-func"
+// RUN:   | FileCheck --check-prefixes=CHECK,NO-FAST,NINF,NNAN,FINITE-ONLY,NSZ,ARCP,NO-AFN,NO-ERRNO,NOROUNDING %s
 //
 // RUN: %clang -### -fno-approx-func -ffast-math -c %s 2>&1 \
-// RUN:   | FileCheck --check-prefix=CHECK-NO-APPROX-FUNC-FAST-MATH %s
-// CHECK-NO-APPROX-FUNC-FAST-MATH: "-cc1"
-// CHECK-NO-APPROX-FUNC-FAST-MATH: "-ffast-math"
+// RUN:   | FileCheck --check-prefixes=CHECK,FAST,NINF,NNAN,FINITE-ONLY,NSZ,ARCP,AFN,NO-ERRNO,NOROUNDING %s
 //
 // RUN: %clang -### -fapprox-func -c %s 2>&1 \
-// RUN:   | FileCheck --check-prefix=CHECK-APPROX-FUNC %s
-// CHECK-APPROX-FUNC: "-cc1"
-// CHECK-APPROX-FUNC: "-fapprox-func"
+// RUN:   | FileCheck --check-prefixes=CHECK,AFN %s
 //
 // RUN: %clang -### -fno-fast-math -fapprox-func -c %s 2>&1 \
-// RUN:   | FileCheck --check-prefix=CHECK-NO-FAST-MATH-APPROX-FUNC %s
-// CHECK-NO-FAST-MATH-APPROX-FUNC: "-cc1"
-// CHECK-NO-FAST-MATH-APPROX-FUNC: "-fapprox-func"
+// RUN:   | FileCheck --check-prefixes=CHECK,NO-FAST,AFN,NOROUNDING %s
 //
 // RUN: %clang -### -fapprox-func -fno-fast-math -c %s 2>&1 \
-// RUN:   | FileCheck --check-prefix=CHECK-APPROX-FUNC-NO-FAST-MATH %s
-// CHECK-APPROX-FUNC-NO-FAST-MATH: "-cc1"
-// CHECK-APPROX-FUNC-NO-FAST-MATH-NOT: "-fapprox-func"
+// RUN:   | FileCheck --check-prefixes=CHECK,NO-FAST,NO-AFN %s
 //
 // RUN: %clang -### -fmath-errno -c %s 2>&1 \
-// RUN:   | FileCheck --check-prefix=CHECK-MATH-ERRNO %s
-// CHECK-MATH-ERRNO: "-cc1"
-// CHECK-MATH-ERRNO: "-fmath-errno"
+// RUN:   | FileCheck --check-prefixes=CHECK,NO-FAST,ERRNO %s
 //
 // RUN: %clang -### -fmath-errno -fno-math-errno -c %s 2>&1 \
-// RUN:   | FileCheck --check-prefix=CHECK-NO-MATH-ERRNO %s
-// CHECK-NO-MATH-ERRNO: "-cc1"
-// CHECK-NO-MATH-ERRNO-NOT: "-fmath-errno"
+// RUN:   | FileCheck --check-prefixes=CHECK,NO-ERRNO %s
 //
 // Target defaults for -fmath-errno (reusing the above checks).
 // RUN: %clang -### -target i686-unknown-linux -c %s 2>&1 \
-// RUN:   | FileCheck --check-prefix=CHECK-MATH-ERRNO %s
+// RUN:   | FileCheck --check-prefixes=CHECK,ERRNO %s
 // RUN: %clang -### -target i686-apple-darwin -c %s 2>&1 \
-// RUN:   | FileCheck --check-prefix=CHECK-NO-MATH-ERRNO %s
+// RUN:   | FileCheck --check-prefixes=CHECK,NO-ERRNO %s
 // RUN: %clang -### -target x86_64-unknown-freebsd -c %s 2>&1 \
-// RUN:   | FileCheck --check-prefix=CHECK-NO-MATH-ERRNO %s
+// RUN:   | FileCheck --check-prefixes=CHECK,NO-ERRNO %s
 // RUN: %clang -### -target x86_64-unknown-netbsd -c %s 2>&1 \
-// RUN:   | FileCheck --check-prefix=CHECK-NO-MATH-ERRNO %s
+// RUN:   | FileCheck --check-prefixes=CHECK,NO-ERRNO %s
 // RUN: %clang -### -target x86_64-unknown-openbsd -c %s 2>&1 \
-// RUN:   | FileCheck --check-prefix=CHECK-NO-MATH-ERRNO %s
+// RUN:   | FileCheck --check-prefixes=CHECK,NO-ERRNO %s
 // RUN: %clang -### --target=x86_64-unknown-haiku -c %s 2>&1 \
-// RUN:   | FileCheck --check-prefix=CHECK-NO-MATH-ERRNO %s
+// RUN:   | FileCheck --check-prefixes=CHECK,NO-ERRNO %s
 // RUN: %clang -### -target x86_64-unknown-dragonfly -c %s 2>&1 \
-// RUN:   | FileCheck --check-prefix=CHECK-NO-MATH-ERRNO %s
+// RUN:   | FileCheck --check-prefixes=CHECK,NO-ERRNO %s
 // RUN: %clang -### -target x86_64-fuchsia -c %s 2>&1 \
-// RUN:   | FileCheck --check-prefix=CHECK-NO-MATH-ERRNO %s
+// RUN:   | FileCheck --check-prefixes=CHECK,NO-ERRNO %s
 // RUN: %clang -### -target x86_64-linux-android -c %s 2>&1 \
-// RUN:   | FileCheck --check-prefix=CHECK-NO-MATH-ERRNO %s
+// RUN:   | FileCheck --check-prefixes=CHECK,NO-ERRNO %s
 // RUN: %clang -### -target x86_64-linux-musl -c %s 2>&1 \
-// RUN:   | FileCheck --check-prefix=CHECK-NO-MATH-ERRNO %s
+// RUN:   | FileCheck --check-prefixes=CHECK,NO-ERRNO %s
 // RUN: %clang -### --target=amdgcn-amd-amdhsa -nogpuinc -nogpulib -c %s 2>&1 \
-// RUN:   | FileCheck --check-prefix=CHECK-NO-MATH-ERRNO %s
+// RUN:   | FileCheck --check-prefixes=CHECK,NO-ERRNO %s
 // RUN: %clang -### -target amdgcn-amd-amdpal -c %s 2>&1 \
-// RUN:   | FileCheck --check-prefix=CHECK-NO-MATH-ERRNO %s
+// RUN:   | FileCheck --check-prefixes=CHECK,NO-ERRNO %s
 // RUN: %clang -### -target amdgcn-mesa-mesa3d -c %s 2>&1   \
-// RUN:   | FileCheck --check-prefix=CHECK-NO-MATH-ERRNO %s
+// RUN:   | FileCheck --check-prefixes=CHECK,NO-ERRNO %s
 //
 // Check that -ffast-math disables -fmath-errno, and -fno-fast-math merely
 // preserves the target default. Also check various flag set operations between
 // the two flags. (Resuses above checks.)
 // RUN: %clang -### -ffast-math -c %s 2>&1 \
-// RUN:   | FileCheck --check-prefix=CHECK-NO-MATH-ERRNO %s
+// RUN:   | FileCheck --check-prefixes=CHECK,NO-ERRNO %s
 // RUN: %clang -### -fmath-errno -ffast-math -c %s 2>&1 \
-// RUN:   | FileCheck --check-prefix=CHECK-NO-MATH-ERRNO %s
+// RUN:   | FileCheck --check-prefixes=CHECK,NO-ERRNO %s
 // RUN: %clang -### -ffast-math -fmath-errno -c %s 2>&1 \
-// RUN:   | FileCheck --check-prefix=CHECK-MATH-ERRNO %s
+// RUN:   | FileCheck --check-prefixes=CHECK,ERRNO %s
 // RUN: %clang -### -target i686-unknown-linux -fno-fast-math -c %s 2>&1 \
-// RUN:   | FileCheck --check-prefix=CHECK-MATH-ERRNO %s
+// RUN:   | FileCheck --check-prefixes=CHECK,ERRNO %s
 // RUN: %clang -### -target i686-unknown-linux -fno-math-errno -fno-fast-math -c %s 2>&1 \
-// RUN:   | FileCheck --check-prefix=CHECK-MATH-ERRNO %s
+// RUN:   | FileCheck --check-prefixes=CHECK,ERRNO %s
 // RUN: %clang -### -target i686-apple-darwin -fno-fast-math -c %s 2>&1 \
-// RUN:   | FileCheck --check-prefix=CHECK-NO-MATH-ERRNO %s
+// RUN:   | FileCheck --check-prefixes=CHECK,NO-ERRNO %s
 // RUN: %clang -### -target i686-apple-darwin -fno-math-errno -fno-fast-math -c %s 2>&1 \
-// RUN:   | FileCheck --check-prefix=CHECK-NO-MATH-ERRNO %s
+// RUN:   | FileCheck --check-prefixes=CHECK,NO-ERRNO %s
 // RUN: %clang -### -fno-fast-math -fno-math-errno -c %s 2>&1 \
-// RUN:   | FileCheck --check-prefix=CHECK-NO-MATH-ERRNO %s
+// RUN:   | FileCheck --check-prefixes=CHECK,NO-ERRNO %s
 //
 // RUN: %clang -### -fno-math-errno -fassociative-math -freciprocal-math \
 // RUN:     -fno-signed-zeros -fno-trapping-math -fapprox-func -c %s 2>&1 \
-// RUN:   | FileCheck --check-prefix=CHECK-UNSAFE-MATH %s
-// CHECK-UNSAFE-MATH: "-cc1"
-// CHECK-UNSAFE-MATH: "-funsafe-math-optimizations"
-// CHECK-UNSAFE-MATH: "-mreassociate"
+// RUN:   | FileCheck --check-prefixes=CHECK,NO-ERRNO,UNSAFE,ARCP,NSZ,NO-TRAPPING,REASSOC %s
 //
 // RUN: %clang -### -fno-fast-math -fno-math-errno -fassociative-math -freciprocal-math \
 // RUN:     -fno-signed-zeros -fno-trapping-math -fapprox-func -c %s 2>&1 \
-// RUN:   | FileCheck --check-prefix=CHECK-NO-FAST-MATH-UNSAFE-MATH %s
-// CHECK-NO-FAST-MATH-UNSAFE-MATH: "-cc1"
-// CHECK-NO-FAST-MATH-UNSAFE-MATH: "-funsafe-math-optimizations"
-// CHECK-NO-FAST-MATH-UNSAFE-MATH: "-mreassociate"
+// RUN:   | FileCheck --check-prefixes=CHECK,NO-ERRNO,UNSAFE,ARCP,NSZ,NO-TRAPPING,REASSOC %s
 
 // The 2nd -fno-fast-math overrides -fassociative-math.
 
 // RUN: %clang -### -fno-fast-math -fno-math-errno -fassociative-math -freciprocal-math \
 // RUN:     -fno-fast-math -fno-signed-zeros -fno-trapping-math -c %s 2>&1 \
-// RUN:   | FileCheck --check-prefix=CHECK-UNSAFE-MATH-NO-FAST-MATH %s
-// CHECK-UNSAFE-MATH-NO-FAST-MATH: "-cc1"
-// CHECK-UNSAFE-MATH-NO-FAST-MATH-NOT: "-funsafe-math-optimizations"
-// CHECK-UNSAFE-MATH-NO-FAST-MATH-NOT: "-mreassociate"
+// RUN:   | FileCheck --check-prefixes=CHECK,NO-UNSAFE,NO-REASSOC,NO-ARCP,NSZ,NO-TRAPPING %s
 //
 // Check that various umbrella flags also enable these frontend options.
 // RUN: %clang -### -ffast-math -c %s 2>&1 \
-// RUN:   | FileCheck --check-prefix=CHECK-NO-INFS %s
-// RUN: %clang -### -ffast-math -c %s 2>&1 \
-// RUN:   | FileCheck --check-prefix=CHECK-NO-NANS %s
-// RUN: %clang -### -ffast-math -c %s 2>&1 \
-// RUN:   | FileCheck --check-prefix=CHECK-UNSAFE-MATH %s
+// RUN:   | FileCheck --check-prefixes=CHECK,FAST,NINF,NNAN,FINITE-ONLY,REASSOC,NSZ,ARCP,AFN,CONTRACT-FAST,NO-ERRNO,NOROUNDING %s
 // RUN: %clang -### -ffinite-math-only -c %s 2>&1 \
-// RUN:   | FileCheck --check-prefix=CHECK-NO-INFS %s
-// RUN: %clang -### -ffinite-math-only -c %s 2>&1 \
-// RUN:   | FileCheck --check-prefix=CHECK-NO-NANS %s
+// RUN:   | FileCheck --check-prefixes=CHECK,NINF,NNAN,FINITE-ONLY %s
 // RUN: %clang -### -funsafe-math-optimizations -fno-math-errno -c %s 2>&1 \
-// RUN:   | FileCheck --check-prefix=CHECK-UNSAFE-MATH %s
+// RUN:   | FileCheck --check-prefixes=CHECK,NO-FAST,NO-NNAN,NO-NINF,NO-FINITE-ONLY,REASSOC,NSZ,ARCP,AFN,CONTRACT-FAST,NO-ERRNO,NOROUNDING %s
 //
 // One umbrella flag is *really* weird and also changes the semantics of the
 // program by adding a special preprocessor macro. Check that the frontend flag
 // modeling this semantic change is provided. Also check that the flag is not
 // present if any of the optimizations are disabled.
-// RUN: %clang -### -ffast-math -c %s 2>&1 \
-// RUN:   | FileCheck --check-prefix=CHECK-FAST-MATH %s
 // RUN: %clang -### -fno-fast-math -ffast-math -c %s 2>&1 \
-// RUN:   | FileCheck --check-prefix=CHECK-FAST-MATH %s
+// RUN:   | FileCheck --check-prefixes=CHECK,FAST,NINF,NNAN,FINITE-ONLY,REASSOC,NSZ,ARCP,AFN,CONTRACT-FAST,NO-ERRNO,NOROUNDING %s
 // RUN: %clang -### -funsafe-math-optimizations -ffinite-math-only \
 // RUN:     -fno-math-errno -ffp-contract=fast -fno-rounding-math -c %s 2>&1 \
-// RUN:   | FileCheck --check-prefix=CHECK-FAST-MATH %s
+// RUN:   | FileCheck --check-prefixes=CHECK,FAST,NINF,NNAN,FINITE-ONLY,NSZ,ARCP,AFN,CONTRACT-FAST,NO-ERRNO,NOROUNDING %s
 // RUN: %clang -### -fno-honor-infinities -fno-honor-nans -fno-math-errno \
 // RUN:     -fassociative-math -freciprocal-math -fno-signed-zeros -fapprox-func \
 // RUN:     -fno-trapping-math -ffp-contract=fast -fno-rounding-math -c %s 2>&1 \
-// RUN:   | FileCheck --check-prefix=CHECK-FAST-MATH %s
-// CHECK-FAST-MATH: "-cc1"
-// CHECK-FAST-MATH: "-ffast-math"
-// CHECK-FAST-MATH: "-ffinite-math-only"
+// RUN:   | FileCheck --check-prefixes=CHECK,FAST,NINF,NNAN,FINITE-ONLY,NSZ,ARCP,AFN,CONTRACT-FAST,NO-ERRNO,NOROUNDING %s
 //
 // RUN: %clang -### -ffast-math -fno-fast-math -c %s 2>&1 \
-// RUN:   | FileCheck --check-prefix=CHECK-NO-FAST-MATH %s
+// RUN:   | FileCheck --check-prefixes=CHECK,NO-FAST,NO-NINF,NO-NNAN,NO-FINITE-ONLY,NO-REASSOC,NO-NSZ,NO-ARCP,NO-AFN,NOROUNDING %s
 // RUN: %clang -### -ffast-math -fno-finite-math-only -c %s 2>&1 \
-// RUN:   | FileCheck --check-prefix=CHECK-NO-FAST-MATH %s
+// RUN:   | FileCheck --check-prefixes=CHECK,NO-FAST,NO-NINF,NO-NNAN,NO-FINITE-ONLY,REASSOC,NSZ,ARCP,AFN,CONTRACT-FAST,NOROUNDING %s
+
+// FIXME: This case leaves nnan and ninf. That seems wrong!
 // RUN: %clang -### -ffast-math -fno-unsafe-math-optimizations -c %s 2>&1 \
-// RUN:   | FileCheck --check-prefix=CHECK-NO-FAST-MATH %s
+// RUN:   | FileCheck --check-prefixes=CHECK,NO-FAST,NINF,NNAN,FINITE-ONLY,NO-REASSOC,NO-NSZ,NO-ARCP,NO-AFN,NOROUNDING %s
 // RUN: %clang -### -ffast-math -fmath-errno -c %s 2>&1 \
-// RUN:   | FileCheck --check-prefix=CHECK-NO-FAST-MATH %s
+// RUN:   | FileCheck --check-prefixes=CHECK,NO-FAST,NINF,NNAN,FINITE-ONLY,REASSOC,NSZ,ARCP,AFN,CONTRACT-FAST,ERRNO,NOROUNDING %s
 // RUN: %clang -### -ffast-math -fno-associative-math -c %s 2>&1 \
-// RUN:   | FileCheck --check-prefix=CHECK-NO-FAST-MATH --check-prefix=CHECK-ASSOC-MATH %s
-// CHECK-NO-FAST-MATH: "-cc1"
-// CHECK-NO-FAST-MATH-NOT: "-ffast-math"
-// CHECK-ASSOC-MATH-NOT: "-mreassociate"
+// RUN:   | FileCheck --check-prefixes=CHECK,NO-FAST,NINF,NNAN,FINITE-ONLY,NO-REASSOC,NSZ,ARCP,AFN,CONTRACT-FAST,NO-ERRNO,NOROUNDING %s
 //
 // Check various means of disabling these flags, including disabling them after
 // they've been enabled via an umbrella flag.
 // RUN: %clang -### -fno-honor-infinities -fhonor-infinities -c %s 2>&1 \
-// RUN:   | FileCheck --check-prefix=CHECK-NO-NO-INFS %s
+// RUN:   | FileCheck --check-prefixes=CHECK,NO-NINF,NO-NNAN,NO-FINITE-ONLY %s
 // RUN: %clang -### -ffinite-math-only -fhonor-infinities -c %s 2>&1 \
-// RUN:   | FileCheck --check-prefix=CHECK-NO-NO-INFS %s
+// RUN:   | FileCheck --check-prefixes=CHECK,NO-NINF,NNAN,NO-FINITE-ONLY %s
 // RUN: %clang -### -ffinite-math-only -fno-finite-math-only -c %s 2>&1 \
-// RUN:   | FileCheck --check-prefix=CHECK-NO-NO-INFS %s
+// RUN:   | FileCheck --check-prefixes=CHECK,NO-NINF,NO-NNAN,NO-FINITE-ONLY %s
 // RUN: %clang -### -ffast-math -fhonor-infinities -c %s 2>&1 \
-// RUN:   | FileCheck --check-prefix=CHECK-NO-NO-INFS %s
+// RUN:   | FileCheck --check-prefixes=CHECK,NO-FAST,NO-NINF,NNAN,NO-FINITE-ONLY,REASSOC,NSZ,ARCP,AFN,CONTRACT-FAST,NO-ERRNO,NOROUNDING %s
 // RUN: %clang -### -ffast-math -fno-finite-math-only -c %s 2>&1 \
-// RUN:   | FileCheck --check-prefix=CHECK-NO-NO-INFS %s
-// CHECK-NO-NO-INFS: "-cc1"
-// CHECK-NO-NO-INFS-NOT: "-menable-no-infs"
-// CHECK-NO-NO-INFS-NOT: "-ffinite-math-only"
-// CHECK-NO-NO-INFS: "-o"
+// RUN:   | FileCheck --check-prefixes=CHECK,NO-FAST,NO-NINF,NO-NNAN,NO-FINITE-ONLY,REASSOC,NSZ,ARCP,AFN,CONTRACT-FAST,NO-ERRNO,NOROUNDING %s
 //
 // RUN: %clang -### -fno-honor-nans -fhonor-nans -c %s 2>&1 \
-// RUN:   | FileCheck --check-prefix=CHECK-NO-NO-NANS %s
+// RUN:   | FileCheck --check-prefixes=CHECK,NO-NINF,NO-NNAN,NO-FINITE-ONLY %s
 // RUN: %clang -### -ffinite-math-only -fhonor-nans -c %s 2>&1 \
-// RUN:   | FileCheck --check-prefix=CHECK-NO-NO-NANS %s
+// RUN:   | FileCheck --check-prefixes=CHECK,NINF,NO-NNAN,NO-FINITE-ONLY %s
 // RUN: %clang -### -ffinite-math-only -fno-finite-math-only -c %s 2>&1 \
-// RUN:   | FileCheck --check-prefix=CHECK-NO-NO-NANS %s
+// RUN:   | FileCheck --check-prefixes=CHECK,NO-NINF,NO-NNAN,NO-FINITE-ONLY %s
 // RUN: %clang -### -ffast-math -fhonor-nans -c %s 2>&1 \
-// RUN:   | FileCheck --check-prefix=CHECK-NO-NO-NANS %s
+// RUN:   | FileCheck --check-prefixes=CHECK,NINF,NO-NNAN,NO-FINITE-ONLY %s
 // RUN: %clang -### -ffast-math -fno-finite-math-only -c %s 2>&1 \
-// RUN:   | FileCheck --check-prefix=CHECK-NO-NO-NANS %s
-// CHECK-NO-NO-NANS: "-cc1"
-// CHECK-NO-NO-NANS-NOT: "-menable-no-nans"
-// CHECK-NO-NO-NANS-NOT: "-ffinite-math-only"
-// CHECK-NO-NO-NANS: "-o"
+// RUN:   | FileCheck --check-prefixes=CHECK,NO-NINF,NO-NNAN,NO-FINITE-ONLY %s
 
 // A later inverted option overrides an earlier option.
 
 // RUN: %clang -### -fassociative-math -freciprocal-math -fno-signed-zeros \
 // RUN:     -fno-trapping-math -fno-associative-math -c %s 2>&1 \
-// RUN:   | FileCheck --check-prefix=CHECK-NO-UNSAFE-MATH %s
+// RUN:   | FileCheck --check-prefixes=CHECK,NO-REASSOC,ARCP,NSZ,NO-TRAPPING %s
+
+// RUN: %clang -### -funsafe-math-optimizations -fno-associative-math -c %s 2>&1 \
+// RUN:   | FileCheck --check-prefixes=CHECK,NO-UNSAFE,NO-REASSOC,ARCP,NSZ,AFN %s
 
-// RUN: %clang -### -funsafe-math-optimizations -fno-associative-math -c %s \
-// RUN:   2>&1 | FileCheck --check-prefix=CHECK-NO-UNSAFE-MATH %s
+// RUN: %clang -### -funsafe-math-optimizations -fno-reciprocal-math -c %s 2>&1 \
+// RUN:   | FileCheck --check-prefixes=CHECK,NO-UNSAFE,REASSOC,NO-ARCP,NSZ,AFN %s
 
-// RUN: %clang -### -funsafe-math-optimizations -fno-reciprocal-math -c %s \
-// RUN:   2>&1 | FileCheck --check-prefix=CHECK-NO-UNSAFE-MATH %s
+// reassoc requires nsz
 // RUN: %clang -### -funsafe-math-optimizations -fsigned-zeros -c %s 2>&1 \
-// RUN:   | FileCheck --check-prefix=CHECK-NO-UNSAFE-MATH %s
+// RUN:   | FileCheck --check-prefixes=CHECK,NO-UNSAFE,NO-REASSOC,ARCP,NO-NSZ,AFN %s
+
+// FIXME: Shouldn't trapping math disable all unsafe math?
 // RUN: %clang -### -funsafe-math-optimizations -ftrapping-math -c %s 2>&1 \
-// RUN:   | FileCheck --check-prefix=CHECK-NO-UNSAFE-MATH %s
+// RUN:   | FileCheck --check-prefixes=CHECK,NO-UNSAFE,ARCP,NSZ,AFN,TRAPPING %s
+
 // RUN: %clang -### -funsafe-math-optimizations -fno-unsafe-math-optimizations \
 // RUN:     -c %s 2>&1 \
-// RUN:   | FileCheck --check-prefix=CHECK-NO-UNSAFE-MATH %s
+// RUN:   | FileCheck --check-prefixes=CHECK,NO-UNSAFE,NO-REASSOC,NO-ARCP,NO-NSZ,NO-AFN %s
 // RUN: %clang -### -ffast-math -fno-associative-math -c %s 2>&1 \
-// RUN:   | FileCheck --check-prefix=CHECK-NO-UNSAFE-MATH %s
+// RUN:   | FileCheck --check-prefixes=CHECK,NO-UNSAFE,NO-REASSOC,ARCP,NSZ,AFN %s
 
 // RUN: %clang -### -ffast-math -fno-reciprocal-math -c %s 2>&1 \
-// RUN:   | FileCheck --check-prefix=CHECK-NO-UNSAFE-MATH %s
+// RUN:   | FileCheck --check-prefixes=CHECK,NO-FAST,NO-UNSAFE,NNAN,NINF,FINITE-ONLY,REASSOC,NO-ARCP,CONTRACT-FAST,NSZ,AFN %s
+
+// reassoc requires nsz
 // RUN: %clang -### -ffast-math -fsigned-zeros -c %s 2>&1 \
-// RUN:   | FileCheck --check-prefix=CHECK-NO-UNSAFE-MATH %s
-// RUN: %clang -### -ffast-math -ftrapping-math -c %s 2>&1 \
-// RUN:   | FileCheck --check-prefix=CHECK-NO-UNSAFE-MATH %s
-// RUN: %clang -### -ffast-math -fno-unsafe-math-optimizations -c %s 2>&1 \
-// RUN:   | FileCheck --check-prefix=CHECK-NO-UNSAFE-MATH %s
+// RUN:   | FileCheck --check-prefixes=CHECK,NO-FAST,NO-UNSAFE,NNAN,NINF,FINITE-ONLY,NO-REASSOC,ARCP,CONTRACT-FAST,NO-NSZ,AFN %s
 
-// CHECK-NO-UNSAFE-MATH: "-cc1"
-// CHECK-NO-UNSAFE-MATH-NOT: "-funsafe-math-optimizations"
-// CHECK-NO_UNSAFE-MATH-NOT: "-mreassociate"
-// CHECK-NO-UNSAFE-MATH: "-o"
+// FIXME: Shouldn't trapping math disable unsafe math?
+// RUN: %clang -### -ffast-math -ftrapping-math -c %s 2>&1 \
+// RUN:   | FileCheck --check-prefixes=CHECK,NO-FAST,NO-UNSAFE,ARCP,NSZ,AFN,TRAPPING %s
 
+// FIXME: -fno-unsafe-math-optimizations shouldn't imply trapping math
+// RUN: %clang -### -ffast-math -fno-unsafe-math-optimizations -c %s 2>&1 \
+// RUN:   | FileCheck --check-prefixes=CHECK,NO-FAST,NO-UNSAFE,NO-ARCP,NO-NSZ,NO-AFN,TRAPPING %s
 
 // Reassociate is allowed because it does not require reciprocal-math.
 
 // RUN: %clang -### -fassociative-math -freciprocal-math -fno-signed-zeros \
 // RUN:     -fno-trapping-math -fno-reciprocal-math -c %s 2>&1 \
-// RUN:   | FileCheck --check-prefix=CHECK-REASSOC-NO-UNSAFE-MATH %s
-
-// CHECK-REASSOC-NO-UNSAFE-MATH: "-cc1"
-// CHECK-REASSOC-NO_UNSAFE-MATH-NOT: "-funsafe-math-optimizations"
-// CHECK-REASSOC-NO_UNSAFE-MATH: "-mreassociate"
-// CHECK-REASSOC-NO-UNSAFE-MATH-NOT: "-funsafe-math-optimizations"
-// CHECK-REASSOC-NO-UNSAFE-MATH: "-o"
+// RUN:   | FileCheck --check-prefixes=CHECK,REASSOC,NO-ARCP,NSZ,NO-TRAPPING %s
 
 
 // In these runs, reassociate is not allowed because both no-signed-zeros and no-trapping-math are required.
 
 // RUN: %clang -### -fassociative-math -freciprocal-math -fno-signed-zeros \
 // RUN:     -fno-trapping-math -fsigned-zeros -c %s 2>&1 \
-// RUN:   | FileCheck --check-prefix=CHECK-NO-REASSOC-NO-UNSAFE-MATH %s
+// RUN:   | FileCheck --check-prefixes=CHECK,NO-REASSOC,ARCP,NO-NSZ,NO-TRAPPING %s
 
 // RUN: %clang -### -fassociative-math -freciprocal-math -fno-signed-zeros \
 // RUN:     -fno-trapping-math -ftrapping-math -c %s 2>&1 \
-// RUN:   | FileCheck --check-prefix=CHECK-NO-REASSOC-NO-UNSAFE-MATH %s
+// RUN:   | FileCheck --check-prefixes=CHECK,NO-REASSOC,ARCP,NSZ,TRAPPING %s
+
+// The checks below allow stringing together prefixes to select the expected
+// set of cc1 options for any combination of floating-point options.
+// This is based on the assumption that the order of the flags when rendered
+// is stable, so the negative checks only need to appear where the option would
+// appear if used.
+
+// start marker
+// CHECK: "-cc1"
+
+// NO-NINF-NOT:         "-menable-no-infs"
+// NINF-SAME:           "-menable-no-infs"
+
+// NO-NNAN-NOT:         "-menable-no-nans"
+// NNAN-SAME:           "-menable-no-nans"
+
+// NO-AFN-NOT:          "-fapprox-func"
+// AFN-SAME:            "-fapprox-func"
+
+// NO-ERRNO-NOT:        "-fmath-errno"
+// ERRNO-SAME:          "-fmath-errno"
+
+// NO-UNSAFE-NOT:       "-funsafe-math-optimizations"
+// UNSAFE-SAME:         "-funsafe-math-optimizations"
+
+// NO-NSZ-NOT:          "-fno-signed-zeros"
+// NSZ-SAME:            "-fno-signed-zeros"
+
+// NO-REASSOC-NOT:      "-mreassociate"
+// REASSOC-SAME:        "-mreassociate"
+
+// NO-ARCP-NOT:         "-freciprocal-math"
+// ARCP-SAME:           "-freciprocal-math"
+
+// NO-DENORM-NOT:       "-fdenormal-fp-math"
+// DENORM-IEEE-SAME:    "-fdenormal-fp-math=ieee,ieee"
+// DENORM-PS-SAME:      "-fdenormal-fp-math=preserve-sign,preserve-sign"
+// DENORM-PZ-SAME:      "-fdenormal-fp-math=positive-zero,positive-zero"
+
+// NO-CONTRACT-NOT:     "-ffp-contract"
+// CONTRACT-OFF-SAME:   "-ffp-contract=off"
+// CONTRACT-ON-SAME:    "-ffp-contract=on"
+// CONTRACT-FAST-SAME:  "-ffp-contract=fast"
+
+// This one is odd because -frounding-math is the default
+// NO-NOROUNDING-NOT:   "-fno-rounding-math"
+// NOROUNDING-SAME:     "-fno-rounding-math"
+
+// NO-TRAPPING-NOT:     "-ffp-exception-behavior=strict"
+// NO-TRAPPING-NOT:     "-ffp-exception-behavior=maytrap"
+// TRAPPING-SAME:       "-ffp-exception-behavior=strict"
+
+// NO-FAST-NOT:         "-ffast-math"
+// FAST-SAME:           "-ffast-math"
+
+// NO-FINITE-ONLY-NOT:  "-ffinite-math-only"
+// FINITE-ONLY-SAME:    "-ffinite-math-only"
+
+// NO-CX-RANGE-NOT:     "-complex-range"
+// CX-RANGE-FULL-SAME:  "-complex-range=full"
+// CX-RANGE-PROMO-SAME: "-complex-range=promoted"
+// CX-RANGE-IMPRO-SAME: "-complex-range=improved"
+// CX-RANGE-BASIC-SAME: "-complex-range=basic"
 
-// CHECK-NO-REASSOC-NO-UNSAFE-MATH: "-cc1"
-// CHECK-NO-REASSOC-NO_UNSAFE-MATH-NOT: "-funsafe-math-optimizations"
-// CHECK-NO-REASSOC-NO_UNSAFE-MATH-NOT: "-mreassociate"
-// CHECK-NO-REASSOC-NO_UNSAFE-MATH-NOT: "-funsafe-math-optimizations"
-// CHECK-NO-REASSOC-NO-UNSAFE-MATH: "-o"
+// end marker
+// CHECK-SAME: "-o"
 
 
 // This isn't fast-math, but the option is handled in the same place as other FP params.
diff --git a/clang/test/Driver/fp-model.c b/clang/test/Driver/fp-model.c
index 74b7de7a275a..a464729edb45 100644
--- a/clang/test/Driver/fp-model.c
+++ b/clang/test/Driver/fp-model.c
@@ -73,9 +73,8 @@
 
 // RUN: %clang -### -Ofast -ffp-model=strict -c %s 2>&1 | FileCheck \
 // RUN:   --check-prefix=WARN12 %s
-// RUN: %clang -### -ffast-math -ffp-model=strict -c %s 2>&1 | FileCheck \
-// RUN:   --check-prefix=WARN12 %s
-// WARN12-NOT: warning: overriding '-ffp-model=strict' option with '-ffp-model=strict' [-Woverriding-option]
+// RUN: %clang -### -Werror -ffast-math -ffp-model=strict -c %s
+// WARN12: warning: overriding '-ffp-model=strict' option with '-Ofast'
 
 // RUN: %clang -### -ffp-model=strict -fapprox-func -c %s 2>&1 \
 // RUN:   | FileCheck --check-prefix=WARN13 %s
diff --git a/clang/test/Driver/gcc-param.c b/clang/test/Driver/gcc-param.c
new file mode 100644
index 000000000000..4672e1156ce7
--- /dev/null
+++ b/clang/test/Driver/gcc-param.c
@@ -0,0 +1,2 @@
+// RUN: touch %t.o
+// RUN: %clang -Werror --param ssp-buffer-size=1 %t.o -###
diff --git a/clang/test/Driver/hlsl-lang-targets-spirv.hlsl b/clang/test/Driver/hlsl-lang-targets-spirv.hlsl
index b86c2e01f8d8..61b10e1648c5 100644
--- a/clang/test/Driver/hlsl-lang-targets-spirv.hlsl
+++ b/clang/test/Driver/hlsl-lang-targets-spirv.hlsl
@@ -1,4 +1,5 @@
 // REQUIRES: spirv-registered-target
+// REQUIRES: directx-registered-target
 
 // Supported targets
 //
diff --git a/clang/test/Driver/integrated-as.c b/clang/test/Driver/integrated-as.c
index e78fde873cf4..b0a26f6011b0 100644
--- a/clang/test/Driver/integrated-as.c
+++ b/clang/test/Driver/integrated-as.c
@@ -3,7 +3,7 @@
 // RUN: %clang -### -c -save-temps -integrated-as --target=x86_64 %s 2>&1 | FileCheck %s
 
 // CHECK: cc1as
-// CHECK: -mrelax-all
+// CHECK-NOT: -mrelax-all
 
 // RISC-V does not enable -mrelax-all
 // RUN: %clang -### -c -save-temps -integrated-as --target=riscv64 %s 2>&1 | FileCheck %s -check-prefix=RISCV-RELAX
diff --git a/clang/test/Driver/linux-ld.c b/clang/test/Driver/linux-ld.c
index d918f4f2d7db..e2043ab22afc 100644
--- a/clang/test/Driver/linux-ld.c
+++ b/clang/test/Driver/linux-ld.c
@@ -2,11 +2,10 @@
 // General tests that ld invocations on Linux targets sane. Note that we use
 // sysroot to make these tests independent of the host system.
 //
-// RUN: %clang -### %s -no-pie 2>&1 \
+// RUN: %clang -### -Werror %s -no-pie 2>&1 \
 // RUN:     --target=i386-unknown-linux -rtlib=platform --unwindlib=platform \
 // RUN:     --sysroot=%S/Inputs/basic_linux_tree \
 // RUN:   | FileCheck --check-prefix=CHECK-LD-32 %s
-// CHECK-LD-32-NOT: warning:
 // CHECK-LD-32: "{{.*}}ld{{(.exe)?}}" "--sysroot=[[SYSROOT:[^"]+]]"
 // CHECK-LD-32: "{{.*}}/usr/lib/gcc/i386-unknown-linux/10.2.0{{/|\\\\}}crtbegin.o"
 // CHECK-LD-32: "-L[[SYSROOT]]/usr/lib/gcc/i386-unknown-linux/10.2.0"
@@ -14,11 +13,10 @@
 // CHECK-LD-32: "-L[[SYSROOT]]/lib"
 // CHECK-LD-32: "-L[[SYSROOT]]/usr/lib"
 //
-// RUN: %clang -### %s -no-pie 2>&1 \
+// RUN: %clang -### %s -Werror -no-pie 2>&1 \
 // RUN:     --target=x86_64-unknown-linux -rtlib=platform --unwindlib=platform \
 // RUN:     --sysroot=%S/Inputs/basic_linux_tree \
 // RUN:   | FileCheck --check-prefix=CHECK-LD-64 %s
-// CHECK-LD-64-NOT: warning:
 // CHECK-LD-64: "{{.*}}ld{{(.exe)?}}" "--sysroot=[[SYSROOT:[^"]+]]"
 // CHECK-LD-64: "--eh-frame-hdr"
 // CHECK-LD-64: "-m" "elf_x86_64"
@@ -32,11 +30,10 @@
 // CHECK-LD-64: "-lc"
 // CHECK-LD-64: "-lgcc" "--as-needed" "-lgcc_s" "--no-as-needed"
 //
-// RUN: %clang -### %s -no-pie 2>&1 \
+// RUN: %clang -### %s -Werror -no-pie 2>&1 \
 // RUN:     --target=x86_64-unknown-linux-gnux32 -rtlib=platform --unwindlib=platform \
 // RUN:     --sysroot=%S/Inputs/basic_linux_tree \
 // RUN:   | FileCheck --check-prefix=CHECK-LD-X32 %s
-// CHECK-LD-X32-NOT: warning:
 // CHECK-LD-X32: "{{.*}}ld{{(.exe)?}}" "--sysroot=[[SYSROOT:[^"]+]]"
 // CHECK-LD-X32: "--eh-frame-hdr"
 // CHECK-LD-X32: "-m" "elf32_x86_64"
@@ -45,13 +42,12 @@
 // CHECK-LD-X32: "-lc"
 // CHECK-LD-X32: "-lgcc" "--as-needed" "-lgcc_s" "--no-as-needed"
 //
-// RUN: %clang -### %s -no-pie 2>&1 \
+// RUN: %clang -### %s -Werror -no-pie 2>&1 \
 // RUN:     --target=x86_64-unknown-linux \
 // RUN:     -resource-dir=%S/Inputs/resource_dir \
 // RUN:     --sysroot=%S/Inputs/basic_linux_tree \
 // RUN:     --rtlib=compiler-rt \
 // RUN:   | FileCheck --check-prefix=CHECK-LD-RT %s
-// CHECK-LD-RT-NOT: warning:
 // CHECK-LD-RT: "-resource-dir" "[[RESDIR:[^"]*]]"
 // CHECK-LD-RT: "{{.*}}ld{{(.exe)?}}" "--sysroot=[[SYSROOT:[^"]+]]"
 // CHECK-LD-RT: "--eh-frame-hdr"
@@ -67,13 +63,12 @@
 // CHECK-LD-RT: libclang_rt.builtins.a"
 // CHECK-LD-RT: "[[RESDIR]]{{/|\\\\}}lib{{/|\\\\}}x86_64-unknown-linux{{/|\\\\}}clang_rt.crtend.o"
 //
-// RUN: %clang -### %s -no-pie 2>&1 \
+// RUN: %clang -### %s -Werror -no-pie 2>&1 \
 // RUN:     --target=i686-unknown-linux \
 // RUN:     -resource-dir=%S/Inputs/resource_dir \
 // RUN:     --sysroot=%S/Inputs/basic_linux_tree \
 // RUN:     --rtlib=compiler-rt \
 // RUN:   | FileCheck --check-prefix=CHECK-LD-RT-I686 %s
-// CHECK-LD-RT-I686-NOT: warning:
 // CHECK-LD-RT-I686: "-resource-dir" "[[RESDIR:[^"]*]]"
 // CHECK-LD-RT-I686: "{{.*}}ld{{(.exe)?}}" "--sysroot=[[SYSROOT:[^"]+]]"
 // CHECK-LD-RT-I686: "--eh-frame-hdr"
@@ -89,13 +84,12 @@
 // CHECK-LD-RT-I686: libclang_rt.builtins.a"
 // CHECK-LD-RT-I686: "[[RESDIR]]{{/|\\\\}}lib{{/|\\\\}}i686-unknown-linux{{/|\\\\}}clang_rt.crtend.o"
 //
-// RUN: %clang -### %s -no-pie 2>&1 \
+// RUN: %clang -### %s -Werror -no-pie 2>&1 \
 // RUN:     --target=arm-linux-androideabi \
 // RUN:     --sysroot=%S/Inputs/basic_android_tree/sysroot \
 // RUN:     -resource-dir=%S/Inputs/resource_dir \
 // RUN:     --rtlib=compiler-rt \
 // RUN:   | FileCheck --check-prefix=CHECK-LD-RT-ANDROID %s
-// CHECK-LD-RT-ANDROID-NOT: warning:
 // CHECK-LD-RT-ANDROID: "{{.*}}ld{{(.exe)?}}" "--sysroot=[[SYSROOT:[^"]+]]"
 // CHECK-LD-RT-ANDROID: "--eh-frame-hdr"
 // CHECK-LD-RT-ANDROID: "-m" "armelf_linux_eabi"
@@ -104,11 +98,10 @@
 // CHECK-LD-RT-ANDROID: "-lc"
 // CHECK-LD-RT-ANDROID: libclang_rt.builtins.a"
 //
-// RUN: %clang -### %s -no-pie 2>&1 \
+// RUN: %clang -### %s -Werror -no-pie 2>&1 \
 // RUN:     --target=x86_64-unknown-linux -rtlib=platform --unwindlib=platform \
 // RUN:     --sysroot=%S/Inputs/basic_linux_tree \
 // RUN:   | FileCheck --check-prefix=CHECK-LD-GCC %s
-// CHECK-LD-GCC-NOT: warning:
 // CHECK-LD-GCC: "{{.*}}ld{{(.exe)?}}" "--sysroot=[[SYSROOT:[^"]+]]"
 // CHECK-LD-GCC: "--eh-frame-hdr"
 // CHECK-LD-GCC: "-m" "elf_x86_64"
@@ -122,12 +115,11 @@
 // CHECK-LD-GCC: "-lc"
 // CHECK-LD-GCC: "-lgcc" "--as-needed" "-lgcc_s" "--no-as-needed"
 //
-// RUN: %clang -### %s -no-pie 2>&1 \
+// RUN: %clang -### %s -Werror -no-pie 2>&1 \
 // RUN:     --target=x86_64-unknown-linux -rtlib=platform --unwindlib=platform \
 // RUN:     -static-libgcc \
 // RUN:     --sysroot=%S/Inputs/basic_linux_tree \
 // RUN:   | FileCheck --check-prefix=CHECK-LD-64-STATIC-LIBGCC %s
-// CHECK-LD-64-STATIC-LIBGCC-NOT: warning:
 // CHECK-LD-64-STATIC-LIBGCC: "{{.*}}ld{{(.exe)?}}" "--sysroot=[[SYSROOT:[^"]+]]"
 // CHECK-LD-64-STATIC-LIBGCC: "--eh-frame-hdr"
 // CHECK-LD-64-STATIC-LIBGCC: "-m" "elf_x86_64"
@@ -268,12 +260,10 @@
 // CHECK-CLANG-ANDROID-STATIC: "{{.*}}ld{{(.exe)?}}" "--sysroot=[[SYSROOT:[^"]+]]"
 // CHECK-CLANG-ANDROID-STATIC: "--start-group" "{{[^"]*}}{{/|\\\\}}libclang_rt.builtins.a" "-l:libunwind.a" "-lc" "--end-group"
 //
-// RUN: %clang -### %s 2>&1      \
-// RUN:     --target=x86_64-unknown-linux -rtlib=platform --unwindlib=platform \
+// RUN: %clang -### %s -Werror --target=x86_64-unknown-linux -rtlib=platform --unwindlib=platform \
 // RUN:     -static \
-// RUN:     --sysroot=%S/Inputs/basic_linux_tree \
+// RUN:     --sysroot=%S/Inputs/basic_linux_tree 2>&1 \
 // RUN:   | FileCheck --check-prefix=CHECK-LD-64-STATIC %s
-// CHECK-LD-64-STATIC-NOT: warning:
 // CHECK-LD-64-STATIC: "{{.*}}ld{{(.exe)?}}" "--sysroot=[[SYSROOT:[^"]+]]"
 // CHECK-LD-64-STATIC: "--eh-frame-hdr"
 // CHECK-LD-64-STATIC: "-m" "elf_x86_64"
@@ -486,13 +476,12 @@
 //
 // Test that we can use -stdlib=libc++ in a build system even when it
 // occasionally links C code instead of C++ code.
-// RUN: %clang -x c -### %s -no-pie 2>&1 \
+// RUN: %clang -x c -### %s -Werror -no-pie 2>&1 \
 // RUN:     --target=x86_64-unknown-linux-gnu \
 // RUN:     -stdlib=libc++ \
 // RUN:     -ccc-install-dir %S/Inputs/basic_linux_libcxx_tree/usr/bin \
 // RUN:     --sysroot=%S/Inputs/basic_linux_libcxx_tree \
 // RUN:   | FileCheck --check-prefix=CHECK-BASIC-LIBCXX-C-LINK %s
-// CHECK-BASIC-LIBCXX-C-LINK-NOT: warning:
 // CHECK-BASIC-LIBCXX-C-LINK: "-cc1"
 // CHECK-BASIC-LIBCXX-C-LINK: "-isysroot" "[[SYSROOT:[^"]+]]"
 // CHECK-BASIC-LIBCXX-C-LINK-NOT: "-internal-isystem" "[[SYSROOT]]/usr/bin/../include/c++/v1"
@@ -1446,6 +1435,32 @@
 // RUN: %clang --target=i386-unknown-linux -no-pie -### %s -ffast-math \
 // RUN:        --sysroot=%S/Inputs/basic_linux_tree 2>&1 \
 // RUN:   | FileCheck --check-prefix=CHECK-NOCRTFASTMATH %s
+// Don't link crtfastmath.o with -shared
+// RUN: %clang --target=x86_64-unknown-linux -no-pie -### %s -ffast-math -shared \
+// RUN:        --sysroot=%S/Inputs/basic_linux_tree 2>&1 \
+// RUN:   | FileCheck --check-prefix=CHECK-NOCRTFASTMATH %s
+// RUN: %clang --target=x86_64-unknown-linux -no-pie -### %s -Ofast -shared \
+// RUN:        --sysroot=%S/Inputs/basic_linux_tree 2>&1 \
+// RUN:   | FileCheck --check-prefix=CHECK-NOCRTFASTMATH %s
+// Check for effects of -mdaz-ftz
+// RUN: %clang --target=x86_64-unknown-linux -### %s -ffast-math -shared -mdaz-ftz \
+// RUN:        --sysroot=%S/Inputs/basic_linux_tree 2>&1 \
+// RUN:   | FileCheck --check-prefix=CHECK-CRTFASTMATH %s
+// RUN: %clang --target=x86_64-unknown-linux -no-pie -### %s -ffast-math -mdaz-ftz \
+// RUN:        --sysroot=%S/Inputs/basic_linux_tree 2>&1 \
+// RUN:   | FileCheck --check-prefix=CHECK-CRTFASTMATH %s
+// RUN: %clang --target=x86_64-unknown-linux -no-pie -### %s -mdaz-ftz \
+// RUN:        --sysroot=%S/Inputs/basic_linux_tree 2>&1 \
+// RUN:   | FileCheck --check-prefix=CHECK-CRTFASTMATH %s
+// RUN: %clang --target=x86_64-unknown-linux -### %s -ffast-math -shared -mno-daz-ftz \
+// RUN:        --sysroot=%S/Inputs/basic_linux_tree 2>&1 \
+// RUN:   | FileCheck --check-prefix=CHECK-NOCRTFASTMATH %s
+// RUN: %clang --target=x86_64-unknown-linux -no-pie -### %s -ffast-math -mno-daz-ftz \
+// RUN:        --sysroot=%S/Inputs/basic_linux_tree 2>&1 \
+// RUN:   | FileCheck --check-prefix=CHECK-NOCRTFASTMATH %s
+// RUN: %clang --target=x86_64-unknown-linux -no-pie -### %s -mno-daz-ftz \
+// RUN:        --sysroot=%S/Inputs/basic_linux_tree 2>&1 \
+// RUN:   | FileCheck --check-prefix=CHECK-NOCRTFASTMATH %s
 // CHECK-CRTFASTMATH: usr/lib/gcc/x86_64-unknown-linux/10.2.0{{/|\\\\}}crtfastmath.o
 // CHECK-NOCRTFASTMATH-NOT: crtfastmath.o
 
@@ -1635,11 +1650,10 @@
 // CHECK-MUSL-AARCH64_BE: "-dynamic-linker" "/lib/ld-musl-aarch64_be.so.1"
 
 // Check whether multilib gcc install works fine on Gentoo with gcc-config
-// RUN: %clang -### %s -no-pie 2>&1 \
+// RUN: %clang -### %s -Werror -no-pie 2>&1 \
 // RUN:     --target=x86_64-unknown-linux-gnu -rtlib=platform --unwindlib=platform \
 // RUN:     --sysroot=%S/Inputs/gentoo_linux_gcc_multi_version_tree \
 // RUN:   | FileCheck --check-prefix=CHECK-LD-GENTOO %s
-// CHECK-LD-GENTOO-NOT: warning:
 // CHECK-LD-GENTOO: "{{.*}}ld{{(.exe)?}}" "--sysroot=[[SYSROOT:[^"]+]]"
 // CHECK-LD-GENTOO: "--eh-frame-hdr"
 // CHECK-LD-GENTOO: "-m" "elf_x86_64"
@@ -1650,11 +1664,10 @@
 // CHECK-LD-GENTOO: "-lgcc" "--as-needed" "-lgcc_s" "--no-as-needed"
 // CHECK-LD-GENTOO: "-lc"
 // CHECK-LD-GENTOO: "-lgcc" "--as-needed" "-lgcc_s" "--no-as-needed"
-// RUN: %clang -### %s -no-pie 2>&1 \
+// RUN: %clang -### %s -Werror -no-pie 2>&1 \
 // RUN:     --target=i686-unknown-linux-gnu -rtlib=platform --unwindlib=platform \
 // RUN:     --sysroot=%S/Inputs/gentoo_linux_gcc_multi_version_tree \
 // RUN:   | FileCheck --check-prefix=CHECK-LD-GENTOO-32 %s
-// CHECK-LD-GENTOO-32-NOT: warning:
 // CHECK-LD-GENTOO-32: "{{.*}}ld{{(.exe)?}}" "--sysroot=[[SYSROOT:[^"]+]]"
 // CHECK-LD-GENTOO-32: "--eh-frame-hdr"
 // CHECK-LD-GENTOO-32: "-m" "elf_i386"
@@ -1665,11 +1678,10 @@
 // CHECK-LD-GENTOO-32: "-lgcc" "--as-needed" "-lgcc_s" "--no-as-needed"
 // CHECK-LD-GENTOO-32: "-lc"
 // CHECK-LD-GENTOO-32: "-lgcc" "--as-needed" "-lgcc_s" "--no-as-needed"
-// RUN: %clang -### %s -no-pie 2>&1 \
+// RUN: %clang -### %s -Werror -no-pie 2>&1 \
 // RUN:     --target=x86_64-unknown-linux-gnux32 -rtlib=platform --unwindlib=platform \
 // RUN:     --sysroot=%S/Inputs/gentoo_linux_gcc_multi_version_tree \
 // RUN:   | FileCheck --check-prefix=CHECK-LD-GENTOO-X32 %s
-// CHECK-LD-GENTOO-X32-NOT: warning:
 // CHECK-LD-GENTOO-X32: "{{.*}}ld{{(.exe)?}}" "--sysroot=[[SYSROOT:[^"]+]]"
 // CHECK-LD-GENTOO-X32: "--eh-frame-hdr"
 // CHECK-LD-GENTOO-X32: "-m" "elf32_x86_64"
@@ -1691,11 +1703,10 @@
 // CHECK-LD-RHEL7-DTS: [[SYSROOT]]/usr/lib/gcc/x86_64-redhat-linux/7/../../../../bin/ld
 
 // Check whether gcc7 install works fine on Amazon Linux AMI
-// RUN: %clang -### %s -no-pie 2>&1 \
+// RUN: %clang -### %s -Werror -no-pie 2>&1 \
 // RUN:     --target=x86_64-amazon-linux -rtlib=libgcc --unwindlib=platform \
 // RUN:     --sysroot=%S/Inputs/ami_linux_tree \
 // RUN:   | FileCheck --check-prefix=CHECK-LD-AMI %s
-// CHECK-LD-AMI-NOT: warning:
 // CHECK-LD-AMI: "{{.*}}ld{{(.exe)?}}" "--sysroot=[[SYSROOT:[^"]+]]"
 // CHECK-LD-AMI: "--eh-frame-hdr"
 // CHECK-LD-AMI: "-m" "elf_x86_64"
diff --git a/clang/test/Driver/riscv-arch.c b/clang/test/Driver/riscv-arch.c
index 8399b4e97f86..abbe8612b378 100644
--- a/clang/test/Driver/riscv-arch.c
+++ b/clang/test/Driver/riscv-arch.c
@@ -209,7 +209,7 @@
 // RUN: not %clang --target=riscv32-unknown-elf -march=rv32q -### %s \
 // RUN: -fsyntax-only 2>&1 | FileCheck -check-prefix=RV32-LETTER %s
 // RV32-LETTER: error: invalid arch name 'rv32q',
-// RV32-LETTER: first letter should be 'e', 'i' or 'g'
+// RV32-LETTER: first letter after 'rv32' should be 'e', 'i' or 'g'
 
 // RUN: not %clang --target=riscv32-unknown-elf -march=rv32imcq -### %s \
 // RUN: -fsyntax-only 2>&1 | FileCheck -check-prefix=RV32-ORDER %s
@@ -239,12 +239,12 @@
 // RUN: not %clang --target=riscv32-unknown-elf -march=rv32xabc -### %s \
 // RUN: -fsyntax-only 2>&1 | FileCheck -check-prefix=RV32X %s
 // RV32X: error: invalid arch name 'rv32xabc',
-// RV32X: first letter should be 'e', 'i' or 'g'
+// RV32X: first letter after 'rv32' should be 'e', 'i' or 'g'
 
 // RUN: not %clang --target=riscv32-unknown-elf -march=rv32sabc -### %s \
 // RUN: -fsyntax-only 2>&1 | FileCheck -check-prefix=RV32S %s
 // RV32S: error: invalid arch name 'rv32sabc',
-// RV32S: first letter should be 'e', 'i' or 'g'
+// RV32S: first letter after 'rv32' should be 'e', 'i' or 'g'
 
 // RUN: not %clang --target=riscv32-unknown-elf -march=rv32ix -### %s \
 // RUN: -fsyntax-only 2>&1 | FileCheck -check-prefix=RV32X-NAME %s
diff --git a/clang/test/Driver/sanitizer-ld.c b/clang/test/Driver/sanitizer-ld.c
index f5657e47626e..7289d09697b4 100644
--- a/clang/test/Driver/sanitizer-ld.c
+++ b/clang/test/Driver/sanitizer-ld.c
@@ -802,10 +802,24 @@
 // RUN:     --target=i686-pc-windows \
 // RUN:     --sysroot=%S/Inputs/basic_linux_tree \
 // RUN:   | FileCheck --check-prefix=CHECK-CFI-STATS-WIN32 %s
+// RUN: not %clang -fsanitize=cfi -fsanitize-stats -### %s 2>&1 \
+// RUN:     --target=i686-pc-windows \
+// RUN:     -fno-rtlib-defaultlib \
+// RUN:     -frtlib-defaultlib \
+// RUN:     --sysroot=%S/Inputs/basic_linux_tree \
+// RUN:   | FileCheck --check-prefix=CHECK-CFI-STATS-WIN32 %s
 // CHECK-CFI-STATS-WIN32: "--dependent-lib=clang_rt.stats_client{{(-i386)?}}.lib"
 // CHECK-CFI-STATS-WIN32: "--dependent-lib=clang_rt.stats{{(-i386)?}}.lib"
 // CHECK-CFI-STATS-WIN32: "--linker-option=/include:___sanitizer_stats_register"
 
+// RUN: not %clang -fsanitize=cfi -fsanitize-stats -### %s 2>&1 \
+// RUN:     --target=i686-pc-windows \
+// RUN:     -fno-rtlib-defaultlib \
+// RUN:     --sysroot=%S/Inputs/basic_linux_tree \
+// RUN:   | FileCheck --check-prefix=CHECK-CFI-STATS-WIN32-NODEF %s
+// CHECK-CFI-STATS-WIN32-NODEF-NOT: "--dependent-lib=clang_rt.stats_client{{(-i386)?}}.lib"
+// CHECK-CFI-STATS-WIN32-NODEF-NOT: "--dependent-lib=clang_rt.stats{{(-i386)?}}.lib"
+
 // RUN: %clang -### %s 2>&1 \
 // RUN:     --target=arm-linux-androideabi -fuse-ld=ld -fsanitize=safe-stack \
 // RUN:     --sysroot=%S/Inputs/basic_android_tree \
diff --git a/clang/test/Driver/wasm-features.c b/clang/test/Driver/wasm-features.c
index 5dae5dbc89b9..1f7fb2134982 100644
--- a/clang/test/Driver/wasm-features.c
+++ b/clang/test/Driver/wasm-features.c
@@ -77,6 +77,12 @@
 // RELAXED-SIMD: "-target-feature" "+relaxed-simd"
 // NO-RELAXED-SIMD: "-target-feature" "-relaxed-simd"
 
+// RUN: %clang --target=wasm32-unknown-unknown -### %s -mhalf-precision 2>&1 | FileCheck %s -check-prefix=HALF-PRECISION
+// RUN: %clang --target=wasm32-unknown-unknown -### %s -mno-half-precision 2>&1 | FileCheck %s -check-prefix=NO-HALF-PRECISION
+
+// HALF-PRECISION: "-target-feature" "+half-precision"
+// NO-HALF-PRECISION: "-target-feature" "-half-precision"
+
 // RUN: %clang --target=wasm32-unknown-unknown -### %s -mexception-handling 2>&1 | FileCheck %s -check-prefix=EXCEPTION-HANDLING
 // RUN: %clang --target=wasm32-unknown-unknown -### %s -mno-exception-handling 2>&1 | FileCheck %s -check-prefix=NO-EXCEPTION-HANDLING
 
diff --git a/clang/test/ExtractAPI/anonymous_record_no_typedef.c b/clang/test/ExtractAPI/anonymous_record_no_typedef.c
index 049e8b1f85bb..71e460afb128 100644
--- a/clang/test/ExtractAPI/anonymous_record_no_typedef.c
+++ b/clang/test/ExtractAPI/anonymous_record_no_typedef.c
@@ -1,417 +1,182 @@
-// XFAIL: *
 // RUN: rm -rf %t
-// RUN: split-file %s %t
-// RUN: sed -e "s@INPUT_DIR@%{/t:regex_replacement}@g" \
-// RUN: %t/reference.output.json.in >> %t/reference.output.json
-// RUN: %clang_cc1 -extract-api --pretty-sgf -triple arm64-apple-macosx \
-// RUN:   -x c-header %t/input.h -o %t/output.json -verify
+// RUN: %clang_cc1 -extract-api --pretty-sgf --emit-sgf-symbol-labels-for-testing \
+// RUN:   -triple arm64-apple-macosx -isystem %S -fretain-comments-from-system-headers \
+// RUN:   -x c-header %s -o %t/output.symbols.json -verify
 
-// Generator version is not consistent across test runs, normalize it.
-// RUN: sed -e "s@\"generator\": \".*\"@\"generator\": \"?\"@g" \
-// RUN: %t/output.json >> %t/output-normalized.json
-// RUN: diff %t/reference.output.json %t/output-normalized.json
+// RUN: FileCheck %s --input-file %t/output.symbols.json --check-prefix GLOBAL
+// RUN: FileCheck %s --input-file %t/output.symbols.json --check-prefix PREFIX
+// RUN: FileCheck %s --input-file %t/output.symbols.json --check-prefix CONTENT
+/// A global variable with an anonymous struct type.
+struct { char *prefix; char *content; } global;
+// GLOBAL-LABEL: "!testLabel": "c:@global"
+// GLOBAL:      "declarationFragments": [
+// GLOBAL-NEXT:   {
+// GLOBAL-NEXT:     "kind": "keyword",
+// GLOBAL-NEXT:     "spelling": "struct"
+// GLOBAL-NEXT:   },
+// GLOBAL-NEXT:   {
+// GLOBAL-NEXT:     "kind": "text",
+// GLOBAL-NEXT:     "spelling": " { ... } "
+// GLOBAL-NEXT:   },
+// GLOBAL-NEXT:   {
+// GLOBAL-NEXT:     "kind": "identifier",
+// GLOBAL-NEXT:     "spelling": "global"
+// GLOBAL-NEXT:   },
+// GLOBAL-NEXT:   {
+// GLOBAL-NEXT:     "kind": "text",
+// GLOBAL-NEXT:     "spelling": ";"
+// GLOBAL-NEXT:   }
+// GLOBAL-NEXT: ],
+// GLOBAL: "text": "A global variable with an anonymous struct type."
+// GLOBAL:     "kind": {
+// GLOBAL-NEXT:  "displayName": "Global Variable",
+// GLOBAL-NEXT:  "identifier": "c.var"
+// GLOBAL:       "title": "global"
+// GLOBAL:     "pathComponents": [
+// GLOBAL-NEXT:  "global"
+// GLOBAL-NEXT:]
+
+// PREFIX: "!testRelLabel": "memberOf $ c:@S@anonymous_record_no_typedef.c@{{[0-9]+}}@FI@prefix $ c:@global"
+// PREFIX-LABEL: "!testLabel": "c:@S@anonymous_record_no_typedef.c@{{[0-9]+}}@FI@prefix"
+// PREFIX: "title": "prefix"
+// PREFIX:      "pathComponents": [
+// PREFIX-NEXT:   "global",
+// PREFIX-NEXT:   "prefix"
+// PREFIX-NEXT: ]
+
+// CONTENT: "!testRelLabel": "memberOf $ c:@S@anonymous_record_no_typedef.c@{{[0-9]+}}@FI@content $ c:@global"
+// CONTENT-LABEL: "!testLabel": "c:@S@anonymous_record_no_typedef.c@{{[0-9]+}}@FI@content"
+// CONTENT: "title": "content"
+// CONTENT:      "pathComponents": [
+// CONTENT-NEXT:   "global",
+// CONTENT-NEXT:   "content"
+// CONTENT-NEXT: ]
 
-//--- input.h
 /// A Vehicle
 struct Vehicle {
+    // RUN: FileCheck %s --input-file %t/output.symbols.json --check-prefix TYPE
+    // RUN: FileCheck %s --input-file %t/output.symbols.json --check-prefix BICYCLE
+    // RUN: FileCheck %s --input-file %t/output.symbols.json --check-prefix CAR
     /// The type of vehicle.
     enum {
         Bicycle,
         Car
     } type;
+    // TYPE-LABEL: "!testLabel": "c:@S@Vehicle@FI@type"
+    // TYPE:      "declarationFragments": [
+    // TYPE-NEXT:   {
+    // TYPE-NEXT:     "kind": "keyword",
+    // TYPE-NEXT:     "spelling": "enum"
+    // TYPE-NEXT:   },
+    // TYPE-NEXT:   {
+    // TYPE-NEXT:     "kind": "text",
+    // TYPE-NEXT:     "spelling": " { ... } "
+    // TYPE-NEXT:   },
+    // TYPE-NEXT:   {
+    // TYPE-NEXT:     "kind": "identifier",
+    // TYPE-NEXT:     "spelling": "type"
+    // TYPE-NEXT:   },
+    // TYPE-NEXT:   {
+    // TYPE-NEXT:     "kind": "text",
+    // TYPE-NEXT:     "spelling": ";"
+    // TYPE-NEXT:   }
+    // TYPE-NEXT: ],
+    // TYPE: "text": "The type of vehicle."
+    // TYPE: "title": "type"
+
+    // BICYCLE: "!testRelLabel": "memberOf $ c:@S@Vehicle@E@anonymous_record_no_typedef.c@{{[0-9]+}}@Bicycle $ c:@S@Vehicle@FI@type"
+    // BICYCLE-LABEL: "!testLabel": "c:@S@Vehicle@E@anonymous_record_no_typedef.c@{{[0-9]+}}@Bicycle"
+    // BICYCLE: "title": "Bicycle"
+    // BICYCLE:      "pathComponents": [
+    // BICYCLE-NEXT:   "Vehicle",
+    // BICYCLE-NEXT:   "type",
+    // BICYCLE-NEXT:   "Bicycle"
+    // BICYCLE-NEXT: ]
 
+    // CAR: "!testRelLabel": "memberOf $ c:@S@Vehicle@E@anonymous_record_no_typedef.c@{{[0-9]+}}@Car $ c:@S@Vehicle@FI@type"
+    // CAR-LABEL: "!testLabel": "c:@S@Vehicle@E@anonymous_record_no_typedef.c@{{[0-9]+}}@Car"
+    // CAR: "title": "Car"
+    // CAR:      "pathComponents": [
+    // CAR-NEXT:   "Vehicle",
+    // CAR-NEXT:   "type",
+    // CAR-NEXT:   "Car"
+    // CAR-NEXT: ]
+
+    // RUN: FileCheck %s --input-file %t/output.symbols.json --check-prefix INFORMATION
+    // RUN: FileCheck %s --input-file %t/output.symbols.json --check-prefix WHEELS
+    // RUN: FileCheck %s --input-file %t/output.symbols.json --check-prefix NAME
     /// The information about the vehicle.
-    struct {
+    union {
         int wheels;
         char *name;
     } information;
+    // INFORMATION-LABEL: "!testLabel": "c:@S@Vehicle@FI@information"
+    // INFORMATION:      "declarationFragments": [
+    // INFORMATION-NEXT:   {
+    // INFORMATION-NEXT:     "kind": "keyword",
+    // INFORMATION-NEXT:     "spelling": "union"
+    // INFORMATION-NEXT:   },
+    // INFORMATION-NEXT:   {
+    // INFORMATION-NEXT:     "kind": "text",
+    // INFORMATION-NEXT:     "spelling": " { ... } "
+    // INFORMATION-NEXT:   },
+    // INFORMATION-NEXT:   {
+    // INFORMATION-NEXT:     "kind": "identifier",
+    // INFORMATION-NEXT:     "spelling": "information"
+    // INFORMATION-NEXT:   },
+    // INFORMATION-NEXT:   {
+    // INFORMATION-NEXT:     "kind": "text",
+    // INFORMATION-NEXT:     "spelling": ";"
+    // INFORMATION-NEXT:   }
+    // INFORMATION-NEXT: ],
+    // INFORMATION: "text": "The information about the vehicle."
+    // INFORMATION: "title": "information"
+
+    // WHEELS: "!testRelLabel": "memberOf $ c:@S@Vehicle@U@anonymous_record_no_typedef.c@{{[0-9]+}}@FI@wheels $ c:@S@Vehicle@FI@information"
+    // WHEELS-LABEL: "!testLabel": "c:@S@Vehicle@U@anonymous_record_no_typedef.c@{{[0-9]+}}@FI@wheels"
+    // WHEELS: "title": "wheels"
+    // WHEELS:      "pathComponents": [
+    // WHEELS-NEXT:   "Vehicle",
+    // WHEELS-NEXT:   "information",
+    // WHEELS-NEXT:   "wheels"
+    // WHEELS-NEXT: ]
+
+    // NAME: "!testRelLabel": "memberOf $ c:@S@Vehicle@U@anonymous_record_no_typedef.c@{{[0-9]+}}@FI@name $ c:@S@Vehicle@FI@information"
+    // NAME-LABEL: "!testLabel": "c:@S@Vehicle@U@anonymous_record_no_typedef.c@{{[0-9]+}}@FI@name"
+    // NAME: "title": "name"
+    // NAME:      "pathComponents": [
+    // NAME-NEXT:   "Vehicle",
+    // NAME-NEXT:   "information",
+    // NAME-NEXT:   "name"
+    // NAME-NEXT: ]
 };
-// expected-no-diagnostics
 
-//--- reference.output.json.in
-{
-  "metadata": {
-    "formatVersion": {
-      "major": 0,
-      "minor": 5,
-      "patch": 3
-    },
-    "generator": "?"
-  },
-  "module": {
-    "name": "",
-    "platform": {
-      "architecture": "arm64",
-      "operatingSystem": {
-        "minimumVersion": {
-          "major": 11,
-          "minor": 0,
-          "patch": 0
-        },
-        "name": "macosx"
-      },
-      "vendor": "apple"
-    }
-  },
-  "relationships": [
-    {
-      "kind": "memberOf",
-      "source": "c:@S@Vehicle@E@input.h@64@Bicycle",
-      "target": "c:@S@Vehicle@E@input.h@64",
-      "targetFallback": "Vehicle::enum (unnamed)"
-    },
-    {
-      "kind": "memberOf",
-      "source": "c:@S@Vehicle@E@input.h@64@Car",
-      "target": "c:@S@Vehicle@E@input.h@64",
-      "targetFallback": "Vehicle::enum (unnamed)"
-    },
-    {
-      "kind": "memberOf",
-      "source": "c:@S@Vehicle@FI@type",
-      "target": "c:@S@Vehicle",
-      "targetFallback": "Vehicle"
-    },
-    {
-      "kind": "memberOf",
-      "source": "c:@S@Vehicle@FI@information",
-      "target": "c:@S@Vehicle",
-      "targetFallback": "Vehicle"
-    }
-  ],
-  "symbols": [
-    {
-      "accessLevel": "public",
-      "declarationFragments": [
-        {
-          "kind": "keyword",
-          "spelling": "enum"
-        },
-        {
-          "kind": "text",
-          "spelling": ": "
-        },
-        {
-          "kind": "typeIdentifier",
-          "preciseIdentifier": "c:i",
-          "spelling": "unsigned int"
-        },
-        {
-          "kind": "text",
-          "spelling": ";"
-        }
-      ],
-      "docComment": {
-        "lines": [
-          {
-            "range": {
-              "end": {
-                "character": 28,
-                "line": 2
-              },
-              "start": {
-                "character": 8,
-                "line": 2
-              }
-            },
-            "text": "The type of vehicle."
-          }
-        ]
-      },
-      "identifier": {
-        "interfaceLanguage": "c",
-        "precise": "c:@S@Vehicle@E@input.h@64"
-      },
-      "kind": {
-        "displayName": "Enumeration",
-        "identifier": "c.enum"
-      },
-      "location": {
-        "position": {
-          "character": 4,
-          "line": 3
-        },
-        "uri": "file://INPUT_DIR/input.h"
-      },
-      "names": {
-        "navigator": [
-          {
-            "kind": "identifier",
-            "spelling": "Vehicle::enum (unnamed)"
-          }
-        ],
-        "title": "Vehicle::enum (unnamed)"
-      },
-      "pathComponents": [
-        "Vehicle::enum (unnamed)"
-      ]
-    },
-    {
-      "accessLevel": "public",
-      "declarationFragments": [
-        {
-          "kind": "identifier",
-          "spelling": "Bicycle"
-        }
-      ],
-      "identifier": {
-        "interfaceLanguage": "c",
-        "precise": "c:@S@Vehicle@E@input.h@64@Bicycle"
-      },
-      "kind": {
-        "displayName": "Enumeration Case",
-        "identifier": "c.enum.case"
-      },
-      "location": {
-        "position": {
-          "character": 8,
-          "line": 4
-        },
-        "uri": "file://INPUT_DIR/input.h"
-      },
-      "names": {
-        "navigator": [
-          {
-            "kind": "identifier",
-            "spelling": "Bicycle"
-          }
-        ],
-        "subHeading": [
-          {
-            "kind": "identifier",
-            "spelling": "Bicycle"
-          }
-        ],
-        "title": "Bicycle"
-      },
-      "pathComponents": [
-        "Vehicle::enum (unnamed)",
-        "Bicycle"
-      ]
-    },
-    {
-      "accessLevel": "public",
-      "declarationFragments": [
-        {
-          "kind": "identifier",
-          "spelling": "Car"
-        }
-      ],
-      "identifier": {
-        "interfaceLanguage": "c",
-        "precise": "c:@S@Vehicle@E@input.h@64@Car"
-      },
-      "kind": {
-        "displayName": "Enumeration Case",
-        "identifier": "c.enum.case"
-      },
-      "location": {
-        "position": {
-          "character": 8,
-          "line": 5
-        },
-        "uri": "file://INPUT_DIR/input.h"
-      },
-      "names": {
-        "navigator": [
-          {
-            "kind": "identifier",
-            "spelling": "Car"
-          }
-        ],
-        "subHeading": [
-          {
-            "kind": "identifier",
-            "spelling": "Car"
-          }
-        ],
-        "title": "Car"
-      },
-      "pathComponents": [
-        "Vehicle::enum (unnamed)",
-        "Car"
-      ]
-    },
-    {
-      "accessLevel": "public",
-      "declarationFragments": [
-        {
-          "kind": "keyword",
-          "spelling": "struct"
-        },
-        {
-          "kind": "text",
-          "spelling": " "
-        },
-        {
-          "kind": "identifier",
-          "spelling": "Vehicle"
-        },
-        {
-          "kind": "text",
-          "spelling": ";"
-        }
-      ],
-      "docComment": {
-        "lines": [
-          {
-            "range": {
-              "end": {
-                "character": 13,
-                "line": 0
-              },
-              "start": {
-                "character": 4,
-                "line": 0
-              }
-            },
-            "text": "A Vehicle"
-          }
-        ]
-      },
-      "identifier": {
-        "interfaceLanguage": "c",
-        "precise": "c:@S@Vehicle"
-      },
-      "kind": {
-        "displayName": "Structure",
-        "identifier": "c.struct"
-      },
-      "location": {
-        "position": {
-          "character": 7,
-          "line": 1
-        },
-        "uri": "file://INPUT_DIR/input.h"
-      },
-      "names": {
-        "navigator": [
-          {
-            "kind": "identifier",
-            "spelling": "Vehicle"
-          }
-        ],
-        "subHeading": [
-          {
-            "kind": "identifier",
-            "spelling": "Vehicle"
-          }
-        ],
-        "title": "Vehicle"
-      },
-      "pathComponents": [
-        "Vehicle"
-      ]
-    },
-    {
-      "accessLevel": "public",
-      "declarationFragments": [
-        {
-          "kind": "keyword",
-          "spelling": "enum"
-        },
-        {
-          "kind": "text",
-          "spelling": " "
-        },
-        {
-          "kind": "identifier",
-          "spelling": "type"
-        },
-        {
-          "kind": "text",
-          "spelling": ";"
-        }
-      ],
-      "identifier": {
-        "interfaceLanguage": "c",
-        "precise": "c:@S@Vehicle@FI@type"
-      },
-      "kind": {
-        "displayName": "Instance Property",
-        "identifier": "c.property"
-      },
-      "location": {
-        "position": {
-          "character": 6,
-          "line": 6
-        },
-        "uri": "file://INPUT_DIR/input.h"
-      },
-      "names": {
-        "navigator": [
-          {
-            "kind": "identifier",
-            "spelling": "type"
-          }
-        ],
-        "subHeading": [
-          {
-            "kind": "identifier",
-            "spelling": "type"
-          }
-        ],
-        "title": "type"
-      },
-      "pathComponents": [
-        "Vehicle",
-        "type"
-      ]
-    },
-    {
-      "accessLevel": "public",
-      "declarationFragments": [
-        {
-          "kind": "keyword",
-          "spelling": "struct"
-        },
-        {
-          "kind": "text",
-          "spelling": " "
-        },
-        {
-          "kind": "identifier",
-          "spelling": "information"
-        },
-        {
-          "kind": "text",
-          "spelling": ";"
-        }
-      ],
-      "identifier": {
-        "interfaceLanguage": "c",
-        "precise": "c:@S@Vehicle@FI@information"
-      },
-      "kind": {
-        "displayName": "Instance Property",
-        "identifier": "c.property"
-      },
-      "location": {
-        "position": {
-          "character": 6,
-          "line": 12
-        },
-        "uri": "file://INPUT_DIR/input.h"
-      },
-      "names": {
-        "navigator": [
-          {
-            "kind": "identifier",
-            "spelling": "information"
-          }
-        ],
-        "subHeading": [
-          {
-            "kind": "identifier",
-            "spelling": "information"
-          }
-        ],
-        "title": "information"
-      },
-      "pathComponents": [
-        "Vehicle",
-        "information"
-      ]
-    }
-  ]
-}
+// RUN: FileCheck %s --input-file %t/output.symbols.json --check-prefix GLOBALENUM
+enum {
+  GlobalCase,
+  GlobalOtherCase
+};
+// GLOBALENUM-DAG: "!testRelLabel": "memberOf $ c:@Ea@GlobalCase@GlobalCase $ c:@Ea@GlobalCase"
+// GLOBALENUM-DAG: "!testRelLabel": "memberOf $ c:@Ea@GlobalCase@GlobalOtherCase $ c:@Ea@GlobalCase"
+// GLOBALENUM-LABEL: "!testLabel": "c:@Ea@GlobalCase"
+// GLOBALENUM:      "declarationFragments": [
+// GLOBALENUM-NEXT:   {
+// GLOBALENUM-NEXT:     "kind": "keyword",
+// GLOBALENUM-NEXT:     "spelling": "enum"
+// GLOBALENUM-NEXT:   },
+// GLOBALENUM-NEXT:   {
+// GLOBALENUM-NEXT:     "kind": "text",
+// GLOBALENUM-NEXT:     "spelling": " : "
+// GLOBALENUM-NEXT:   },
+// GLOBALENUM-NEXT:   {
+// GLOBALENUM-NEXT:     "kind": "typeIdentifier",
+// GLOBALENUM-NEXT:     "preciseIdentifier": "c:i",
+// GLOBALENUM-NEXT:     "spelling": "unsigned int"
+// GLOBALENUM-NEXT:   },
+// GLOBALENUM-NEXT:   {
+// GLOBALENUM-NEXT:     "kind": "text",
+// GLOBALENUM-NEXT:     "spelling": " { ... };"
+// GLOBALENUM-NEXT:   }
+// GLOBALENUM-NEXT: ]
+
+// expected-no-diagnostics
diff --git a/clang/test/ExtractAPI/availability.c b/clang/test/ExtractAPI/availability.c
index 12ac73f0d429..237b2ffa55d7 100644
--- a/clang/test/ExtractAPI/availability.c
+++ b/clang/test/ExtractAPI/availability.c
@@ -1,446 +1,101 @@
 // RUN: rm -rf %t
-// RUN: split-file %s %t
-// RUN: sed -e "s@INPUT_DIR@%{/t:regex_replacement}@g" \
-// RUN: %t/reference.output.json.in >> %t/reference.output.json
-// RUN: %clang_cc1 -extract-api --pretty-sgf --product-name=Availability -triple arm64-apple-macosx -x c-header %t/input.h -o %t/output.json -verify
+// RUN: %clang_cc1 -extract-api --pretty-sgf --emit-sgf-symbol-labels-for-testing -triple arm64-apple-macosx \
+// RUN:   -x c-header %s -o %t/output.symbols.json -verify
 
-// Generator version is not consistent across test runs, normalize it.
-// RUN: sed -e "s@\"generator\": \".*\"@\"generator\": \"?\"@g" \
-// RUN: %t/output.json >> %t/output-normalized.json
-// RUN: diff %t/reference.output.json %t/output-normalized.json
+// RUN: FileCheck %s --input-file %t/output.symbols.json --check-prefix A
+void a(void) __attribute__((availability(macos, introduced=12.0)));
+// A-LABEL: "!testLabel": "c:@F@a"
+// A:      "availability": [
+// A-NEXT:   {
+// A-NEXT:     "domain": "macos",
+// A-NEXT:     "introduced": {
+// A-NEXT:       "major": 12,
+// A-NEXT:       "minor": 0,
+// A-NEXT:       "patch": 0
+// A-NEXT:     }
+// A-NEXT:   }
+// A-NEXT: ]
 
-// CHECK-NOT: error:
-// CHECK-NOT: warning:
+// RUN: FileCheck %s --input-file %t/output.symbols.json --check-prefix B
+void b(void) __attribute__((availability(macos, introduced=11.0, deprecated=12.0, obsoleted=20.0)));
+// B-LABEL: "!testLabel": "c:@F@b"
+// B:      "availability": [
+// B-NEXT:   {
+// B-NEXT:     "deprecated": {
+// B-NEXT:       "major": 12,
+// B-NEXT:       "minor": 0,
+// B-NEXT:       "patch": 0
+// B-NEXT:     },
+// B-NEXT:     "domain": "macos",
+// B-NEXT:     "introduced": {
+// B-NEXT:       "major": 11,
+// B-NEXT:       "minor": 0,
+// B-NEXT:       "patch": 0
+// B-NEXT:     },
+// B-NEXT:     "obsoleted": {
+// B-NEXT:       "major": 20,
+// B-NEXT:       "minor": 0,
+// B-NEXT:       "patch": 0
+// B-NEXT:     }
+// B-NEXT:   }
+// B-NEXT: ]
 
-//--- input.h
-void a(void);
+// RUN: FileCheck %s --input-file %t/output.symbols.json --check-prefix E
+void c(void) __attribute__((availability(macos, introduced=11.0, deprecated=12.0, obsoleted=20.0))) __attribute__((availability(ios, introduced=13.0)));
+// C-LABEL: "!testLabel": "c:@F@c"
+// C:       "availability": [
+// C-NEXT:    {
+// C-NEXT:      "deprecated": {
+// C-NEXT:        "major": 12,
+// C-NEXT:        "minor": 0,
+// C-NEXT:        "patch": 0
+// C-NEXT:      },
+// C-NEXT:      "domain": "macos",
+// C-NEXT:      "introduced": {
+// C-NEXT:        "major": 11,
+// C-NEXT:        "minor": 0,
+// C-NEXT:        "patch": 0
+// C-NEXT:      },
+// C-NEXT:      "obsoleted": {
+// C-NEXT:        "major": 20,
+// C-NEXT:        "minor": 0,
+// C-NEXT:        "patch": 0
+// C-NEXT:      }
+// C-NEXT:    }
+// C-NEXT: ]
 
-void b(void) __attribute__((availability(macos, introduced=12.0)));
+// RUN: FileCheck %s --input-file %t/output.symbols.json --check-prefix D
+void d(void) __attribute__((deprecated)) __attribute__((availability(macos, introduced=11.0)));
+// D-LABEL: "!testLabel": "c:@F@d"
+// D:      "availability": [
+// D-NEXT:   {
+// D-NEXT:     "domain": "*",
+// D-NEXT:     "isUnconditionallyDeprecated": true
+// D-NEXT:   },
+// D-NEXT:   {
+// D-NEXT:     "domain": "macos",
+// D-NEXT:     "introduced": {
+// D-NEXT:       "major": 11,
+// D-NEXT:       "minor": 0,
+// D-NEXT:       "patch": 0
+// D-NEXT:     }
+// D-NEXT:   }
+// D-NEXT: ]
 
-void c(void) __attribute__((availability(macos, introduced=11.0, deprecated=12.0, obsoleted=20.0)));
+// This symbol should be dropped as it's unconditionally unavailable
+// RUN: FileCheck %s --input-file %t/output.symbols.json --check-prefix E
+void e(void) __attribute__((unavailable)) __attribute__((availability(macos, introduced=11.0)));
+// E-NOT: "!testLabel": "c:@F@e"
 
-void d(void) __attribute__((availability(macos, introduced=11.0, deprecated=12.0, obsoleted=20.0))) __attribute__((availability(ios, introduced=13.0)));
+// RUN: FileCheck %s --input-file %t/output.symbols.json --check-prefix F
+void f(void) __attribute__((availability(macos, unavailable)));
+// F-LABEL: "!testLabel": "c:@F@f"
+// F:      "availability": [
+// F-NEXT:   {
+// F-NEXT:     "domain": "macos",
+// F-NEXT:     "isUnconditionallyUnavailable": true
+// F-NEXT:   }
+// F-NEXT: ]
 
-void e(void) __attribute__((deprecated)) __attribute__((availability(macos, introduced=11.0)));
+// expected-no-diagnostics
 
-void f(void) __attribute__((unavailable)) __attribute__((availability(macos, introduced=11.0)));
-
-void d(void) __attribute__((availability(tvos, introduced=15.0)));
-
-void e(void) __attribute__((availability(tvos, unavailable)));
-
-///expected-no-diagnostics
-
-//--- reference.output.json.in
-{
-  "metadata": {
-    "formatVersion": {
-      "major": 0,
-      "minor": 5,
-      "patch": 3
-    },
-    "generator": "?"
-  },
-  "module": {
-    "name": "Availability",
-    "platform": {
-      "architecture": "arm64",
-      "operatingSystem": {
-        "minimumVersion": {
-          "major": 11,
-          "minor": 0,
-          "patch": 0
-        },
-        "name": "macosx"
-      },
-      "vendor": "apple"
-    }
-  },
-  "relationships": [],
-  "symbols": [
-    {
-      "accessLevel": "public",
-      "declarationFragments": [
-        {
-          "kind": "typeIdentifier",
-          "preciseIdentifier": "c:v",
-          "spelling": "void"
-        },
-        {
-          "kind": "text",
-          "spelling": " "
-        },
-        {
-          "kind": "identifier",
-          "spelling": "a"
-        },
-        {
-          "kind": "text",
-          "spelling": "();"
-        }
-      ],
-      "functionSignature": {
-        "returns": [
-          {
-            "kind": "typeIdentifier",
-            "preciseIdentifier": "c:v",
-            "spelling": "void"
-          }
-        ]
-      },
-      "identifier": {
-        "interfaceLanguage": "c",
-        "precise": "c:@F@a"
-      },
-      "kind": {
-        "displayName": "Function",
-        "identifier": "c.func"
-      },
-      "location": {
-        "position": {
-          "character": 5,
-          "line": 0
-        },
-        "uri": "file://INPUT_DIR/input.h"
-      },
-      "names": {
-        "navigator": [
-          {
-            "kind": "identifier",
-            "spelling": "a"
-          }
-        ],
-        "subHeading": [
-          {
-            "kind": "identifier",
-            "spelling": "a"
-          }
-        ],
-        "title": "a"
-      },
-      "pathComponents": [
-        "a"
-      ]
-    },
-    {
-      "accessLevel": "public",
-      "availability": [
-        {
-          "domain": "macos",
-          "introduced": {
-            "major": 12,
-            "minor": 0,
-            "patch": 0
-          }
-        }
-      ],
-      "declarationFragments": [
-        {
-          "kind": "typeIdentifier",
-          "preciseIdentifier": "c:v",
-          "spelling": "void"
-        },
-        {
-          "kind": "text",
-          "spelling": " "
-        },
-        {
-          "kind": "identifier",
-          "spelling": "b"
-        },
-        {
-          "kind": "text",
-          "spelling": "();"
-        }
-      ],
-      "functionSignature": {
-        "returns": [
-          {
-            "kind": "typeIdentifier",
-            "preciseIdentifier": "c:v",
-            "spelling": "void"
-          }
-        ]
-      },
-      "identifier": {
-        "interfaceLanguage": "c",
-        "precise": "c:@F@b"
-      },
-      "kind": {
-        "displayName": "Function",
-        "identifier": "c.func"
-      },
-      "location": {
-        "position": {
-          "character": 5,
-          "line": 2
-        },
-        "uri": "file://INPUT_DIR/input.h"
-      },
-      "names": {
-        "navigator": [
-          {
-            "kind": "identifier",
-            "spelling": "b"
-          }
-        ],
-        "subHeading": [
-          {
-            "kind": "identifier",
-            "spelling": "b"
-          }
-        ],
-        "title": "b"
-      },
-      "pathComponents": [
-        "b"
-      ]
-    },
-    {
-      "accessLevel": "public",
-      "availability": [
-        {
-          "deprecated": {
-            "major": 12,
-            "minor": 0,
-            "patch": 0
-          },
-          "domain": "macos",
-          "introduced": {
-            "major": 11,
-            "minor": 0,
-            "patch": 0
-          },
-          "obsoleted": {
-            "major": 20,
-            "minor": 0,
-            "patch": 0
-          }
-        }
-      ],
-      "declarationFragments": [
-        {
-          "kind": "typeIdentifier",
-          "preciseIdentifier": "c:v",
-          "spelling": "void"
-        },
-        {
-          "kind": "text",
-          "spelling": " "
-        },
-        {
-          "kind": "identifier",
-          "spelling": "c"
-        },
-        {
-          "kind": "text",
-          "spelling": "();"
-        }
-      ],
-      "functionSignature": {
-        "returns": [
-          {
-            "kind": "typeIdentifier",
-            "preciseIdentifier": "c:v",
-            "spelling": "void"
-          }
-        ]
-      },
-      "identifier": {
-        "interfaceLanguage": "c",
-        "precise": "c:@F@c"
-      },
-      "kind": {
-        "displayName": "Function",
-        "identifier": "c.func"
-      },
-      "location": {
-        "position": {
-          "character": 5,
-          "line": 4
-        },
-        "uri": "file://INPUT_DIR/input.h"
-      },
-      "names": {
-        "navigator": [
-          {
-            "kind": "identifier",
-            "spelling": "c"
-          }
-        ],
-        "subHeading": [
-          {
-            "kind": "identifier",
-            "spelling": "c"
-          }
-        ],
-        "title": "c"
-      },
-      "pathComponents": [
-        "c"
-      ]
-    },
-    {
-      "accessLevel": "public",
-      "availability": [
-        {
-          "deprecated": {
-            "major": 12,
-            "minor": 0,
-            "patch": 0
-          },
-          "domain": "macos",
-          "introduced": {
-            "major": 11,
-            "minor": 0,
-            "patch": 0
-          },
-          "obsoleted": {
-            "major": 20,
-            "minor": 0,
-            "patch": 0
-          }
-        }
-      ],
-      "declarationFragments": [
-        {
-          "kind": "typeIdentifier",
-          "preciseIdentifier": "c:v",
-          "spelling": "void"
-        },
-        {
-          "kind": "text",
-          "spelling": " "
-        },
-        {
-          "kind": "identifier",
-          "spelling": "d"
-        },
-        {
-          "kind": "text",
-          "spelling": "();"
-        }
-      ],
-      "functionSignature": {
-        "returns": [
-          {
-            "kind": "typeIdentifier",
-            "preciseIdentifier": "c:v",
-            "spelling": "void"
-          }
-        ]
-      },
-      "identifier": {
-        "interfaceLanguage": "c",
-        "precise": "c:@F@d"
-      },
-      "kind": {
-        "displayName": "Function",
-        "identifier": "c.func"
-      },
-      "location": {
-        "position": {
-          "character": 5,
-          "line": 6
-        },
-        "uri": "file://INPUT_DIR/input.h"
-      },
-      "names": {
-        "navigator": [
-          {
-            "kind": "identifier",
-            "spelling": "d"
-          }
-        ],
-        "subHeading": [
-          {
-            "kind": "identifier",
-            "spelling": "d"
-          }
-        ],
-        "title": "d"
-      },
-      "pathComponents": [
-        "d"
-      ]
-    },
-    {
-      "accessLevel": "public",
-      "availability": [
-        {
-          "domain": "*",
-          "isUnconditionallyDeprecated": true
-        },
-        {
-          "domain": "macos",
-          "introduced": {
-            "major": 11,
-            "minor": 0,
-            "patch": 0
-          }
-        }
-      ],
-      "declarationFragments": [
-        {
-          "kind": "typeIdentifier",
-          "preciseIdentifier": "c:v",
-          "spelling": "void"
-        },
-        {
-          "kind": "text",
-          "spelling": " "
-        },
-        {
-          "kind": "identifier",
-          "spelling": "e"
-        },
-        {
-          "kind": "text",
-          "spelling": "();"
-        }
-      ],
-      "functionSignature": {
-        "returns": [
-          {
-            "kind": "typeIdentifier",
-            "preciseIdentifier": "c:v",
-            "spelling": "void"
-          }
-        ]
-      },
-      "identifier": {
-        "interfaceLanguage": "c",
-        "precise": "c:@F@e"
-      },
-      "kind": {
-        "displayName": "Function",
-        "identifier": "c.func"
-      },
-      "location": {
-        "position": {
-          "character": 5,
-          "line": 8
-        },
-        "uri": "file://INPUT_DIR/input.h"
-      },
-      "names": {
-        "navigator": [
-          {
-            "kind": "identifier",
-            "spelling": "e"
-          }
-        ],
-        "subHeading": [
-          {
-            "kind": "identifier",
-            "spelling": "e"
-          }
-        ],
-        "title": "e"
-      },
-      "pathComponents": [
-        "e"
-      ]
-    }
-  ]
-}
diff --git a/clang/test/ExtractAPI/enum.c b/clang/test/ExtractAPI/enum.c
index 1cdf45ca3cdf..67e003834a7d 100644
--- a/clang/test/ExtractAPI/enum.c
+++ b/clang/test/ExtractAPI/enum.c
@@ -147,7 +147,7 @@ enum {
         },
         {
           "kind": "text",
-          "spelling": ": "
+          "spelling": " : "
         },
         {
           "kind": "typeIdentifier",
@@ -459,7 +459,7 @@ enum {
         },
         {
           "kind": "text",
-          "spelling": ": "
+          "spelling": " : "
         },
         {
           "kind": "typeIdentifier",
@@ -686,7 +686,7 @@ enum {
         },
         {
           "kind": "text",
-          "spelling": ": "
+          "spelling": " : "
         },
         {
           "kind": "typeIdentifier",
@@ -695,7 +695,7 @@ enum {
         },
         {
           "kind": "text",
-          "spelling": ";"
+          "spelling": " { ... };"
         }
       ],
       "identifier": {
@@ -778,7 +778,7 @@ enum {
         },
         {
           "kind": "text",
-          "spelling": ": "
+          "spelling": " : "
         },
         {
           "kind": "typeIdentifier",
@@ -787,7 +787,7 @@ enum {
         },
         {
           "kind": "text",
-          "spelling": ";"
+          "spelling": " { ... };"
         }
       ],
       "identifier": {
diff --git a/clang/test/ExtractAPI/function_noexcepts.cpp b/clang/test/ExtractAPI/function_noexcepts.cpp
index d95eaaa7e769..fc18ecb04fef 100644
--- a/clang/test/ExtractAPI/function_noexcepts.cpp
+++ b/clang/test/ExtractAPI/function_noexcepts.cpp
@@ -63,11 +63,7 @@ void getFooBar() noexcept(false);
         },
         {
           "kind": "text",
-          "spelling": "()"
-        },
-        {
-          "kind": "text",
-          "spelling": " "
+          "spelling": "() "
         },
         {
           "kind": "keyword",
@@ -139,11 +135,7 @@ void getFooBar() noexcept(false);
         },
         {
           "kind": "text",
-          "spelling": "()"
-        },
-        {
-          "kind": "text",
-          "spelling": " "
+          "spelling": "() "
         },
         {
           "kind": "keyword",
@@ -223,11 +215,7 @@ void getFooBar() noexcept(false);
         },
         {
           "kind": "text",
-          "spelling": "()"
-        },
-        {
-          "kind": "text",
-          "spelling": " "
+          "spelling": "() "
         },
         {
           "kind": "keyword",
diff --git a/clang/test/ExtractAPI/methods.cpp b/clang/test/ExtractAPI/methods.cpp
index 412c0bb3f903..67f04b4d33db 100644
--- a/clang/test/ExtractAPI/methods.cpp
+++ b/clang/test/ExtractAPI/methods.cpp
@@ -81,11 +81,7 @@ class Foo {
   // SETL-NEXT:   },
   // SETL-NEXT:   {
   // SETL-NEXT:     "kind": "text",
-  // SETL-NEXT:     "spelling": ")"
-  // SETL-NEXT:   },
-  // SETL-NEXT:   {
-  // SETL-NEXT:     "kind": "text",
-  // SETL-NEXT:     "spelling": " "
+  // SETL-NEXT:     "spelling": ") "
   // SETL-NEXT:   },
   // SETL-NEXT:   {
   // SETL-NEXT:     "kind": "keyword",
diff --git a/clang/test/ExtractAPI/objc_block.m b/clang/test/ExtractAPI/objc_block.m
index 4a4335ec0983..4761a864f534 100644
--- a/clang/test/ExtractAPI/objc_block.m
+++ b/clang/test/ExtractAPI/objc_block.m
@@ -35,11 +35,7 @@
 // NOPARAM-NEXT:   },
 // NOPARAM-NEXT:   {
 // NOPARAM-NEXT:     "kind": "text",
-// NOPARAM-NEXT:     "spelling": " (^"
-// NOPARAM-NEXT:   },
-// NOPARAM-NEXT:   {
-// NOPARAM-NEXT:     "kind": "text",
-// NOPARAM-NEXT:     "spelling": ")()) "
+// NOPARAM-NEXT:     "spelling": " (^)()) "
 // NOPARAM-NEXT:   },
 // NOPARAM-NEXT:   {
 // NOPARAM-NEXT:     "kind": "internalParam",
@@ -65,11 +61,7 @@
 // NOPARAM-NEXT:         },
 // NOPARAM-NEXT:         {
 // NOPARAM-NEXT:           "kind": "text",
-// NOPARAM-NEXT:           "spelling": " (^"
-// NOPARAM-NEXT:         },
-// NOPARAM-NEXT:         {
-// NOPARAM-NEXT:           "kind": "text",
-// NOPARAM-NEXT:           "spelling": ")()) "
+// NOPARAM-NEXT:           "spelling": " (^)()) "
 // NOPARAM-NEXT:         },
 // NOPARAM-NEXT:         {
 // NOPARAM-NEXT:           "kind": "internalParam",
@@ -120,11 +112,7 @@
 // PARAM-NEXT:   },
 // PARAM-NEXT:   {
 // PARAM-NEXT:     "kind": "text",
-// PARAM-NEXT:     "spelling": " (^"
-// PARAM-NEXT:   },
-// PARAM-NEXT:   {
-// PARAM-NEXT:     "kind": "text",
-// PARAM-NEXT:     "spelling": ")("
+// PARAM-NEXT:     "spelling": " (^)("
 // PARAM-NEXT:   },
 // PARAM-NEXT:   {
 // PARAM-NEXT:     "kind": "typeIdentifier",
@@ -167,11 +155,7 @@
 // PARAM-NEXT:         },
 // PARAM-NEXT:         {
 // PARAM-NEXT:           "kind": "text",
-// PARAM-NEXT:           "spelling": " (^"
-// PARAM-NEXT:         },
-// PARAM-NEXT:         {
-// PARAM-NEXT:           "kind": "text",
-// PARAM-NEXT:           "spelling": ")("
+// PARAM-NEXT:           "spelling": " (^)("
 // PARAM-NEXT:         },
 // PARAM-NEXT:         {
 // PARAM-NEXT:           "kind": "typeIdentifier",
@@ -239,11 +223,7 @@
 // MULTIPARAM-NEXT:   },
 // MULTIPARAM-NEXT:   {
 // MULTIPARAM-NEXT:     "kind": "text",
-// MULTIPARAM-NEXT:     "spelling": " (^"
-// MULTIPARAM-NEXT:   },
-// MULTIPARAM-NEXT:   {
-// MULTIPARAM-NEXT:     "kind": "text",
-// MULTIPARAM-NEXT:     "spelling": ")("
+// MULTIPARAM-NEXT:     "spelling": " (^)("
 // MULTIPARAM-NEXT:   },
 // MULTIPARAM-NEXT:   {
 // MULTIPARAM-NEXT:     "kind": "typeIdentifier",
@@ -303,11 +283,7 @@
 // MULTIPARAM-NEXT:         },
 // MULTIPARAM-NEXT:         {
 // MULTIPARAM-NEXT:           "kind": "text",
-// MULTIPARAM-NEXT:           "spelling": " (^"
-// MULTIPARAM-NEXT:         },
-// MULTIPARAM-NEXT:         {
-// MULTIPARAM-NEXT:           "kind": "text",
-// MULTIPARAM-NEXT:           "spelling": ")("
+// MULTIPARAM-NEXT:           "spelling": " (^)("
 // MULTIPARAM-NEXT:         },
 // MULTIPARAM-NEXT:         {
 // MULTIPARAM-NEXT:           "kind": "typeIdentifier",
@@ -392,11 +368,7 @@
 // VARIADIC-NEXT:   },
 // VARIADIC-NEXT:   {
 // VARIADIC-NEXT:     "kind": "text",
-// VARIADIC-NEXT:     "spelling": " (^"
-// VARIADIC-NEXT:   },
-// VARIADIC-NEXT:   {
-// VARIADIC-NEXT:     "kind": "text",
-// VARIADIC-NEXT:     "spelling": ")("
+// VARIADIC-NEXT:     "spelling": " (^)("
 // VARIADIC-NEXT:   },
 // VARIADIC-NEXT:   {
 // VARIADIC-NEXT:     "kind": "typeIdentifier",
@@ -439,11 +411,7 @@
 // VARIADIC-NEXT:         },
 // VARIADIC-NEXT:         {
 // VARIADIC-NEXT:           "kind": "text",
-// VARIADIC-NEXT:           "spelling": " (^"
-// VARIADIC-NEXT:         },
-// VARIADIC-NEXT:         {
-// VARIADIC-NEXT:           "kind": "text",
-// VARIADIC-NEXT:           "spelling": ")("
+// VARIADIC-NEXT:           "spelling": " (^)("
 // VARIADIC-NEXT:         },
 // VARIADIC-NEXT:         {
 // VARIADIC-NEXT:           "kind": "typeIdentifier",
diff --git a/clang/test/ExtractAPI/typedef_anonymous_record.c b/clang/test/ExtractAPI/typedef_anonymous_record.c
index 9e00ff752546..9c03e9e190ed 100644
--- a/clang/test/ExtractAPI/typedef_anonymous_record.c
+++ b/clang/test/ExtractAPI/typedef_anonymous_record.c
@@ -21,7 +21,7 @@ typedef struct { } MyStruct;
 // MYSTRUCT-NEXT:   },
 // MYSTRUCT-NEXT:   {
 // MYSTRUCT-NEXT:     "kind": "text",
-// MYSTRUCT-NEXT:     "spelling": " "
+// MYSTRUCT-NEXT:     "spelling": " { ... } "
 // MYSTRUCT-NEXT:   },
 // MYSTRUCT-NEXT:   {
 // MYSTRUCT-NEXT:     "kind": "identifier",
@@ -97,7 +97,7 @@ typedef enum { Case } MyEnum;
 // MYENUM-NEXT:  },
 // MYENUM-NEXT:  {
 // MYENUM-NEXT:    "kind": "text",
-// MYENUM-NEXT:    "spelling": " "
+// MYENUM-NEXT:    "spelling": " { ... } "
 // MYENUM-NEXT:  },
 // MYENUM-NEXT:  {
 // MYENUM-NEXT:    "kind": "identifier",
diff --git a/clang/test/ExtractAPI/typedef_struct_enum.c b/clang/test/ExtractAPI/typedef_struct_enum.c
index fb6fbe987624..64b718675666 100644
--- a/clang/test/ExtractAPI/typedef_struct_enum.c
+++ b/clang/test/ExtractAPI/typedef_struct_enum.c
@@ -72,7 +72,7 @@ typedef enum Test2 {
 // TEST2-NEXT:   },
 // TEST2-NEXT:   {
 // TEST2-NEXT:     "kind": "text",
-// TEST2-NEXT:     "spelling": ": "
+// TEST2-NEXT:     "spelling": " : "
 // TEST2-NEXT:   },
 // TEST2-NEXT:   {
 // TEST2-NEXT:     "kind": "typeIdentifier",
diff --git a/clang/test/Frontend/ast-dump-on-llvm.ll b/clang/test/Frontend/ast-dump-on-llvm.ll
new file mode 100644
index 000000000000..cdacfde4ba84
--- /dev/null
+++ b/clang/test/Frontend/ast-dump-on-llvm.ll
@@ -0,0 +1,29 @@
+; RUN: not %clang_cc1 -triple x86_64-unknown-unknown -ast-dump %s -o - 2>&1 | FileCheck %s --check-prefix=CHECK-AST-DUMP
+; RUN: not %clang_cc1 -triple x86_64-unknown-unknown -ast-dump=json %s -o - 2>&1 | FileCheck %s --check-prefix=CHECK-AST-DUMP-EQ-JSON
+; RUN: not %clang_cc1 -triple x86_64-unknown-unknown -ast-dump=default %s -o - 2>&1 | FileCheck %s --check-prefix=CHECK-AST-DUMP-EQ-DEFAULT
+; RUN: not %clang_cc1 -triple x86_64-unknown-unknown -ast-dump-all %s -o - 2>&1 | FileCheck %s --check-prefix=CHECK-AST-DUMP-ALL
+; RUN: not %clang_cc1 -triple x86_64-unknown-unknown -ast-dump-all=json %s -o - 2>&1 | FileCheck %s --check-prefix=CHECK-AST-DUMP-ALL-EQ-JSON
+; RUN: not %clang_cc1 -triple x86_64-unknown-unknown -ast-dump-all=default %s -o - 2>&1 | FileCheck %s --check-prefix=CHECK-AST-DUMP-ALL-EQ-DEFAULT
+
+; RUN: not %clang_cc1 -triple x86_64-unknown-unknown -ast-print %s -o - 2>&1 | FileCheck %s --check-prefix=CHECK-AST-PRINT
+; RUN: not %clang_cc1 -triple x86_64-unknown-unknown -ast-view %s -o - 2>&1 | FileCheck %s --check-prefix=CHECK-AST-VIEW
+; RUN: not %clang_cc1 -triple x86_64-unknown-unknown -ast-list %s -o - 2>&1 | FileCheck %s --check-prefix=CHECK-AST-LIST
+; RUN: not %clang_cc1 -triple x86_64-unknown-unknown -ast-dump-lookups %s -o - 2>&1 | FileCheck %s --check-prefix=CHECK-AST-DUMP-LOOKUP
+; RUN: not %clang_cc1 -triple x86_64-unknown-unknown -ast-dump-filter=FunctionDecl %s -o - 2>&1 | FileCheck %s --check-prefix=CHECK-AST-DUMP-FILTER-EQ
+; RUN: not %clang_cc1 -triple x86_64-unknown-unknown -ast-dump-decl-types %s -o - 2>&1 | FileCheck %s --check-prefix=CHECK-AST-DUMP-DECL-TYPES
+; RUN: not %clang_cc1 -triple x86_64-unknown-unknown -fsyntax-only %s -o - 2>&1 | FileCheck %s --check-prefix=CHECK-SYNTAX-ONLY
+
+
+; CHECK-AST-DUMP: fatal error: cannot apply AST actions to LLVM IR file '{{.*}}'
+; CHECK-AST-DUMP-EQ-JSON: fatal error: cannot apply AST actions to LLVM IR file '{{.*}}'
+; CHECK-AST-DUMP-EQ-DEFAULT: fatal error: cannot apply AST actions to LLVM IR file '{{.*}}'
+; CHECK-AST-DUMP-ALL: fatal error: cannot apply AST actions to LLVM IR file '{{.*}}'
+; CHECK-AST-DUMP-ALL-EQ-JSON: fatal error: cannot apply AST actions to LLVM IR file '{{.*}}'
+; CHECK-AST-DUMP-ALL-EQ-DEFAULT: fatal error: cannot apply AST actions to LLVM IR file '{{.*}}'
+; CHECK-AST-PRINT: fatal error: cannot apply AST actions to LLVM IR file '{{.*}}'
+; CHECK-AST-VIEW: fatal error: cannot apply AST actions to LLVM IR file '{{.*}}'
+; CHECK-AST-LIST: fatal error: cannot apply AST actions to LLVM IR file '{{.*}}'
+; CHECK-AST-DUMP-LOOKUP: fatal error: cannot apply AST actions to LLVM IR file '{{.*}}'
+; CHECK-AST-DUMP-FILTER-EQ: fatal error: cannot apply AST actions to LLVM IR file '{{.*}}'
+; CHECK-AST-DUMP-DECL-TYPES: fatal error: cannot apply AST actions to LLVM IR file '{{.*}}'
+; CHECK-SYNTAX-ONLY: fatal error: cannot apply AST actions to LLVM IR file '{{.*}}'
diff --git a/clang/test/Headers/wasm.c b/clang/test/Headers/wasm.c
index 57f8b6d0db17..b22d87a5f8b7 100644
--- a/clang/test/Headers/wasm.c
+++ b/clang/test/Headers/wasm.c
@@ -1572,7 +1572,7 @@ uint32_t test_i8x16_bitmask(v128_t a) {
 // CHECK-LABEL: @test_i8x16_popcnt(
 // CHECK-NEXT:  entry:
 // CHECK-NEXT:    [[TMP0:%.*]] = bitcast <4 x i32> [[A:%.*]] to <16 x i8>
-// CHECK-NEXT:    [[TMP1:%.*]] = tail call <16 x i8> @llvm.ctpop.v16i8(<16 x i8> [[TMP0]]), !range [[RNG5:![0-9]+]]
+// CHECK-NEXT:    [[TMP1:%.*]] = tail call range(i8 0, 9) <16 x i8> @llvm.ctpop.v16i8(<16 x i8> [[TMP0]])
 // CHECK-NEXT:    [[TMP2:%.*]] = bitcast <16 x i8> [[TMP1]] to <4 x i32>
 // CHECK-NEXT:    ret <4 x i32> [[TMP2]]
 //
diff --git a/clang/test/Interpreter/fail.cpp b/clang/test/Interpreter/fail.cpp
index 4e301f37548f..633d92794325 100644
--- a/clang/test/Interpreter/fail.cpp
+++ b/clang/test/Interpreter/fail.cpp
@@ -1,12 +1,19 @@
-// FIXME: There're some inconsistencies between interactive and non-interactive
-// modes. For example, when clang-repl runs in the interactive mode, issues an
-// error, and then successfully recovers if we decide it's a success then for
-// the non-interactive mode the exit code should be a failure.
-// RUN: clang-repl "int x = 10;" "int y=7; err;" "int y = 10;"
 // REQUIRES: host-supports-jit
 // UNSUPPORTED: system-aix
-// RUN: cat %s | not clang-repl | FileCheck %s
-BOOM!
+// clang-repl can be called from the prompt in non-interactive mode as a
+// calculator in shell scripts, for example. In that case if there is an error
+// we should set the exit code as failure.
+// RUN: not clang-repl "int x = 10;" "int y=7; err;" "int y = 10;"
+
+// In interactive (REPL) mode, we can have errors but we should exit with
+// success because errors in the input code are part of the interactive use.
+// RUN: cat %s | clang-repl | FileCheck %s
+
+// However, interactive mode should fail when we specified -verify and there
+// was a diagnostic mismatches. This will make the testsuite fail as intended.
+// RUN: cat %s | not clang-repl -Xcc -Xclang -Xcc -verify | FileCheck %s
+
+BOOM! // expected-error {{intended to fail the -verify test}}
 extern "C" int printf(const char *, ...);
 int i = 42;
 auto r1 = printf("i = %d\n", i);
diff --git a/clang/test/Lexer/cxx-features.cpp b/clang/test/Lexer/cxx-features.cpp
index baaa9d4434e9..4a08eb61cd39 100644
--- a/clang/test/Lexer/cxx-features.cpp
+++ b/clang/test/Lexer/cxx-features.cpp
@@ -222,7 +222,7 @@
 #error "wrong value for __cpp_aggregate_bases"
 #endif
 
-#if check(structured_bindings, 0, 0, 0, 201606, 201606, 201606, 201606)
+#if check(structured_bindings, 0, 0, 0, 202403L, 202403L, 202403L, 202403L)
 #error "wrong value for __cpp_structured_bindings"
 #endif
 
diff --git a/clang/test/Lexer/update_consecutive_macro_address_space.c b/clang/test/Lexer/update_consecutive_macro_address_space.c
index 80ef4557591c..6f74709556c0 100644
--- a/clang/test/Lexer/update_consecutive_macro_address_space.c
+++ b/clang/test/Lexer/update_consecutive_macro_address_space.c
@@ -1,14 +1,15 @@
 // RUN: %clang -cc1 -print-stats %s 2>&1 | FileCheck %s
-// CHECK: 6 local SLocEntries allocated
+// CHECK: 7 local SLocEntries allocated
 //
-// Verify that the macro arg expansion is split to two file ids, we have 6 file
-// ids rather than 5:
+// Verify that the macro arg expansion is split to two file ids, we have 7 file
+// ids rather than 6:
 //   0: invalid file id
 //   1: main file
 //   2: builtin file
-//   3: macro expansion for X
-//   4: macro arg expansions for 1
-//   5: macro arg expansions for == 2
+//   3: scratch space for __GCC_[CON|DE]STRUCTIVE_SIZE macros
+//   4: macro expansion for X
+//   5: macro arg expansions for 1
+//   6: macro arg expansions for == 2
 #define X(x) (int)(x);
 void func() {
   X(1
diff --git a/clang/test/Misc/cc1as-relax-all.s b/clang/test/Misc/cc1as-relax-all.s
new file mode 100644
index 000000000000..e76fc6f61bab
--- /dev/null
+++ b/clang/test/Misc/cc1as-relax-all.s
@@ -0,0 +1,13 @@
+// REQUIRES: x86-registered-target
+// RUN: %clang -cc1as -triple x86_64 -filetype obj -mrelax-all %s -o %t.o
+// RUN: llvm-objdump -d %t.o | FileCheck %s
+
+// CHECK:      <.text>:
+// CHECK-NEXT:   0: e9 06 00 00 00                jmp     0xb <foo>
+// CHECK-NEXT:   5: 0f 84 00 00 00 00             je      0xb <foo>
+// CHECK-EMPTY:
+
+jmp foo
+je foo
+
+foo: ret
diff --git a/clang/test/Misc/target-invalid-cpu-note.c b/clang/test/Misc/target-invalid-cpu-note.c
index 9c91c4157cd6..21d80b713450 100644
--- a/clang/test/Misc/target-invalid-cpu-note.c
+++ b/clang/test/Misc/target-invalid-cpu-note.c
@@ -5,11 +5,11 @@
 
 // RUN: not %clang_cc1 -triple arm64--- -target-cpu not-a-cpu -fsyntax-only %s 2>&1 | FileCheck %s --check-prefix AARCH64
 // AARCH64: error: unknown target CPU 'not-a-cpu'
-// AARCH64-NEXT: note: valid target CPU values are: cortex-a34, cortex-a35, cortex-a53, cortex-a55, cortex-a510, cortex-a520, cortex-a520ae, cortex-a57, cortex-a65, cortex-a65ae, cortex-a72, cortex-a73, cortex-a75, cortex-a76, cortex-a76ae, cortex-a77, cortex-a78, cortex-a78ae, cortex-a78c, cortex-a710, cortex-a715, cortex-a720, cortex-a720ae, cortex-r82, cortex-x1, cortex-x1c, cortex-x2, cortex-x3, cortex-x4, neoverse-e1, neoverse-n1, neoverse-n2, neoverse-512tvb, neoverse-v1, neoverse-v2, cyclone, apple-a7, apple-a8, apple-a9, apple-a10, apple-a11, apple-a12, apple-a13, apple-a14, apple-a15, apple-a16, apple-a17, apple-m1, apple-m2, apple-m3, apple-s4, apple-s5, exynos-m3, exynos-m4, exynos-m5, falkor, saphira, kryo, thunderx2t99, thunderx3t110, thunderx, thunderxt88, thunderxt81, thunderxt83, tsv110, a64fx, carmel, ampere1, ampere1a, ampere1b, cobalt-100, grace{{$}}
+// AARCH64-NEXT: note: valid target CPU values are: cortex-a34, cortex-a35, cortex-a53, cortex-a55, cortex-a510, cortex-a520, cortex-a520ae, cortex-a57, cortex-a65, cortex-a65ae, cortex-a72, cortex-a73, cortex-a75, cortex-a76, cortex-a76ae, cortex-a77, cortex-a78, cortex-a78ae, cortex-a78c, cortex-a710, cortex-a715, cortex-a720, cortex-a720ae, cortex-r82, cortex-x1, cortex-x1c, cortex-x2, cortex-x3, cortex-x4, neoverse-e1, neoverse-n1, neoverse-n2, neoverse-n3, neoverse-512tvb, neoverse-v1, neoverse-v2, neoverse-v3, neoverse-v3ae, cyclone, apple-a7, apple-a8, apple-a9, apple-a10, apple-a11, apple-a12, apple-a13, apple-a14, apple-a15, apple-a16, apple-a17, apple-m1, apple-m2, apple-m3, apple-s4, apple-s5, exynos-m3, exynos-m4, exynos-m5, falkor, saphira, kryo, thunderx2t99, thunderx3t110, thunderx, thunderxt88, thunderxt81, thunderxt83, tsv110, a64fx, carmel, ampere1, ampere1a, ampere1b, cobalt-100, grace{{$}}
 
 // RUN: not %clang_cc1 -triple arm64--- -tune-cpu not-a-cpu -fsyntax-only %s 2>&1 | FileCheck %s --check-prefix TUNE_AARCH64
 // TUNE_AARCH64: error: unknown target CPU 'not-a-cpu'
-// TUNE_AARCH64-NEXT: note: valid target CPU values are: cortex-a34, cortex-a35, cortex-a53, cortex-a55, cortex-a510, cortex-a520, cortex-a520ae, cortex-a57, cortex-a65, cortex-a65ae, cortex-a72, cortex-a73, cortex-a75, cortex-a76, cortex-a76ae, cortex-a77, cortex-a78, cortex-a78ae, cortex-a78c, cortex-a710, cortex-a715, cortex-a720, cortex-a720ae, cortex-r82, cortex-x1, cortex-x1c, cortex-x2, cortex-x3, cortex-x4, neoverse-e1, neoverse-n1, neoverse-n2, neoverse-512tvb, neoverse-v1, neoverse-v2, cyclone, apple-a7, apple-a8, apple-a9, apple-a10, apple-a11, apple-a12, apple-a13, apple-a14, apple-a15, apple-a16, apple-a17, apple-m1, apple-m2, apple-m3, apple-s4, apple-s5, exynos-m3, exynos-m4, exynos-m5, falkor, saphira, kryo, thunderx2t99, thunderx3t110, thunderx, thunderxt88, thunderxt81, thunderxt83, tsv110, a64fx, carmel, ampere1, ampere1a, ampere1b, cobalt-100, grace{{$}}
+// TUNE_AARCH64-NEXT: note: valid target CPU values are: cortex-a34, cortex-a35, cortex-a53, cortex-a55, cortex-a510, cortex-a520, cortex-a520ae, cortex-a57, cortex-a65, cortex-a65ae, cortex-a72, cortex-a73, cortex-a75, cortex-a76, cortex-a76ae, cortex-a77, cortex-a78, cortex-a78ae, cortex-a78c, cortex-a710, cortex-a715, cortex-a720, cortex-a720ae, cortex-r82, cortex-x1, cortex-x1c, cortex-x2, cortex-x3, cortex-x4, neoverse-e1, neoverse-n1, neoverse-n2, neoverse-n3, neoverse-512tvb, neoverse-v1, neoverse-v2, neoverse-v3, neoverse-v3ae, cyclone, apple-a7, apple-a8, apple-a9, apple-a10, apple-a11, apple-a12, apple-a13, apple-a14, apple-a15, apple-a16, apple-a17, apple-m1, apple-m2, apple-m3, apple-s4, apple-s5, exynos-m3, exynos-m4, exynos-m5, falkor, saphira, kryo, thunderx2t99, thunderx3t110, thunderx, thunderxt88, thunderxt81, thunderxt83, tsv110, a64fx, carmel, ampere1, ampere1a, ampere1b, cobalt-100, grace{{$}}
 
 // RUN: not %clang_cc1 -triple i386--- -target-cpu not-a-cpu -fsyntax-only %s 2>&1 | FileCheck %s --check-prefix X86
 // X86: error: unknown target CPU 'not-a-cpu'
diff --git a/clang/test/Modules/pr88400.cppm b/clang/test/Modules/pr88400.cppm
new file mode 100644
index 000000000000..ff69137a0b90
--- /dev/null
+++ b/clang/test/Modules/pr88400.cppm
@@ -0,0 +1,61 @@
+// RUN: rm -rf %t
+// RUN: mkdir -p %t
+// RUN: split-file %s %t
+//
+// RUN: %clang_cc1 -std=c++20 %t/bar.cppm -emit-module-interface -o %t/bar.pcm
+// RUN: %clang_cc1 -std=c++20 %t/foo.cc -fmodule-file=bar=%t/bar.pcm -fsyntax-only -verify
+// RUN: %clang_cc1 -std=c++20 %t/bar.cc -fmodule-file=bar=%t/bar.pcm -fsyntax-only -verify
+//
+// RUN: %clang_cc1 -std=c++20 %t/bar.cppm -emit-reduced-module-interface -o %t/bar.pcm
+// RUN: %clang_cc1 -std=c++20 %t/foo.cc -fmodule-file=bar=%t/bar.pcm -fsyntax-only -verify
+// RUN: %clang_cc1 -std=c++20 %t/bar.cc -fmodule-file=bar=%t/bar.pcm -fsyntax-only -verify
+
+//--- header.h
+#pragma once
+
+namespace N {
+    template<typename T>
+    concept X = true;
+
+    template<X T>
+    class Y {
+    public:
+        template<X U>
+        friend class Y;
+    };
+
+    inline Y<int> x;
+}
+
+//--- bar.cppm
+module;
+
+#include "header.h"
+
+export module bar;
+
+namespace N {
+    // To make sure N::Y won't get elided.
+    using N::x;
+}
+
+//--- foo.cc
+// expected-no-diagnostics
+#include "header.h"
+
+import bar;
+
+void y() {
+    N::Y<int> y{};
+};
+
+//--- bar.cc
+// expected-no-diagnostics
+import bar;
+
+#include "header.h"
+
+void y() {
+    N::Y<int> y{};
+};
+
diff --git a/clang/test/Modules/prune-non-affecting-module-map-files-textual.c b/clang/test/Modules/prune-non-affecting-module-map-files-textual.c
new file mode 100644
index 000000000000..fce325d4774c
--- /dev/null
+++ b/clang/test/Modules/prune-non-affecting-module-map-files-textual.c
@@ -0,0 +1,46 @@
+// This test checks that a module map with a textual header can be marked as
+// non-affecting.
+
+// RUN: rm -rf %t && mkdir %t
+// RUN: split-file %s %t
+
+//--- X.modulemap
+module X { textual header "X.h" }
+//--- X.h
+typedef int X_int;
+
+//--- Y.modulemap
+module Y { textual header "Y.h" }
+//--- Y.h
+typedef int Y_int;
+
+//--- A.modulemap
+module A { header "A.h" export * }
+//--- A.h
+#include "X.h"
+
+// RUN: %clang_cc1 -fmodules -emit-module %t/A.modulemap -fmodule-name=A -o %t/A0.pcm \
+// RUN:   -fmodule-map-file=%t/X.modulemap
+// RUN: %clang_cc1 -fsyntax-only -module-file-info %t/A0.pcm | FileCheck %s --check-prefix=A0 --implicit-check-not=Y.modulemap
+// A0: Input file: {{.*}}X.modulemap
+
+// RUN: %clang_cc1 -fmodules -emit-module %t/A.modulemap -fmodule-name=A -o %t/A1.pcm \
+// RUN:   -fmodule-map-file=%t/X.modulemap -fmodule-map-file=%t/Y.modulemap
+// RUN: %clang_cc1 -fsyntax-only -module-file-info %t/A0.pcm | FileCheck %s --check-prefix=A1 \
+// RUN:   --implicit-check-not=Y.modulemap
+// A1: Input file: {{.*}}X.modulemap
+
+// RUN: diff %t/A0.pcm %t/A1.pcm
+
+//--- B.modulemap
+module B { header "B.h" export * }
+//--- B.h
+#include "A.h"
+typedef X_int B_int;
+
+// RUN: %clang_cc1 -fmodules -emit-module %t/B.modulemap -fmodule-name=B -o %t/B.pcm \
+// RUN:   -fmodule-file=A=%t/A0.pcm \
+// RUN:   -fmodule-map-file=%t/A.modulemap -fmodule-map-file=%t/X.modulemap -fmodule-map-file=%t/Y.modulemap
+// RUN: %clang_cc1 -fsyntax-only -module-file-info %t/B.pcm | FileCheck %s --check-prefix=B \
+// RUN:   --implicit-check-not=X.modulemap --implicit-check-not=Y.modulemap
+// B: Input file: {{.*}}B.modulemap
diff --git a/clang/test/OpenMP/task_depend_messages.cpp b/clang/test/OpenMP/task_depend_messages.cpp
index 388595bef4de..3f39c55527b5 100644
--- a/clang/test/OpenMP/task_depend_messages.cpp
+++ b/clang/test/OpenMP/task_depend_messages.cpp
@@ -62,7 +62,7 @@ int main(int argc, char **argv, char *env[]) {
   #pragma omp task depend(in : argv[ : argc][1 : argc - 1])
   #pragma omp task depend(in : arr[0])
   #pragma omp task depend(depobj:argc) // omp45-error {{expected 'in', 'out', 'inout' or 'mutexinoutset' in OpenMP clause 'depend'}} omp50-error {{expected lvalue expression of 'omp_depend_t' type, not 'int'}} omp51-error {{expected lvalue expression of 'omp_depend_t' type, not 'int'}}
-  #pragma omp task depend(depobj : argv[ : argc][1 : argc - 1]) // omp45-error {{expected 'in', 'out', 'inout' or 'mutexinoutset' in OpenMP clause 'depend'}} omp50-error {{expected lvalue expression of 'omp_depend_t' type, not '<OpenMP array section type>'}} omp51-error {{expected lvalue expression of 'omp_depend_t' type, not '<OpenMP array section type>'}}
+  #pragma omp task depend(depobj : argv[ : argc][1 : argc - 1]) // omp45-error {{expected 'in', 'out', 'inout' or 'mutexinoutset' in OpenMP clause 'depend'}} omp50-error {{expected lvalue expression of 'omp_depend_t' type, not '<array section type>'}} omp51-error {{expected lvalue expression of 'omp_depend_t' type, not '<array section type>'}}
   #pragma omp task depend(depobj : arr[0]) // omp45-error {{expected 'in', 'out', 'inout' or 'mutexinoutset' in OpenMP clause 'depend'}}
   #pragma omp task depend(in : ([ // expected-error {{expected variable name or 'this' in lambda capture list}} expected-error {{expected ')'}} expected-note {{to match this '('}}
   #pragma omp task depend(in : ([] // expected-error {{expected body of lambda expression}} expected-error {{expected ')'}} expected-note {{to match this '('}}
diff --git a/clang/test/Parser/cxx1z-decomposition.cpp b/clang/test/Parser/cxx1z-decomposition.cpp
index 90d60df2e47f..4b17f72effb0 100644
--- a/clang/test/Parser/cxx1z-decomposition.cpp
+++ b/clang/test/Parser/cxx1z-decomposition.cpp
@@ -1,6 +1,7 @@
-// RUN: %clang_cc1 -std=c++17 %s -verify=expected,cxx17 -fcxx-exceptions
-// RUN: %clang_cc1 -std=c++2b %s -verify=expected,cxx2b -fcxx-exceptions
-// RUN: not %clang_cc1 -std=c++17 %s -emit-llvm-only -fcxx-exceptions
+// RUN: %clang_cc1 -std=c++17 %s -triple x86_64-unknown-linux-gnu -verify=expected,cxx17,pre2c -fcxx-exceptions
+// RUN: %clang_cc1 -std=c++2b %s -triple x86_64-unknown-linux-gnu -verify=expected,cxx2b,pre2c,post2b -fcxx-exceptions
+// RUN: %clang_cc1 -std=c++2c %s -triple x86_64-unknown-linux-gnu -verify=expected,cxx2c,post2b -fcxx-exceptions
+// RUN: not %clang_cc1 -std=c++17 %s -triple x86_64-unknown-linux-gnu -emit-llvm-only -fcxx-exceptions
 
 struct S { int a, b, c; };
 
@@ -58,7 +59,7 @@ namespace OtherDecl {
 namespace GoodSpecifiers {
   void f() {
     int n[1];
-    const volatile auto &[a] = n; // cxx2b-warning {{volatile qualifier in structured binding declaration is deprecated}}
+    const volatile auto &[a] = n; // post2b-warning {{volatile qualifier in structured binding declaration is deprecated}}
   }
 }
 
@@ -97,8 +98,8 @@ namespace BadSpecifiers {
     S [a] = s; // expected-error {{cannot be declared with type 'S'}}
     decltype(auto) [b] = s; // expected-error {{cannot be declared with type 'decltype(auto)'}}
     auto ([c2]) = s; // cxx17-error {{decomposition declaration cannot be declared with parenthese}} \
-                     // cxx2b-error {{use of undeclared identifier 'c2'}} \
-                     // cxx2b-error {{expected body of lambda expression}} \
+                     // post2b-error {{use of undeclared identifier 'c2'}} \
+                     // post2b-error {{expected body of lambda expression}} \
 
     // FIXME: This error is not very good.
     auto [d]() = s; // expected-error {{expected ';'}} expected-error {{expected expression}}
@@ -119,9 +120,6 @@ namespace BadSpecifiers {
     [[]] auto [ok_3] = s;
     alignas(S) auto [ok_4] = s;
 
-    // ... but not after the identifier or declarator.
-    // FIXME: These errors are not very good.
-    auto [bad_attr_1 [[]]] = s; // expected-error {{attribute list cannot appear here}} expected-error 2{{}}
     auto [bad_attr_2] [[]] = s; // expected-error {{expected ';'}} expected-error {{}}
   }
 }
@@ -156,3 +154,50 @@ namespace Init {
     S [goodish4] { 4 }; // expected-error {{cannot be declared with type 'S'}}
   }
 }
+
+
+namespace attributes {
+
+struct S{
+    int a;
+    int b = 0;
+};
+
+void err() {
+    auto [[]] = S{0}; // expected-error {{expected unqualified-id}}
+    auto [ alignas(42) a, foo ] = S{0}; // expected-error {{an attribute list cannot appear here}}
+    auto [ c, [[]] d ] = S{0}; // expected-error {{an attribute list cannot appear here}}
+    auto [ e, alignas(42) f ] = S{0}; // expected-error {{an attribute list cannot appear here}}
+}
+
+void ok() {
+    auto [ a alignas(42) [[]], b alignas(42) [[]]] = S{0}; // expected-error 2{{'alignas' attribute only applies to variables, data members and tag types}} \
+                                                           // pre2c-warning  2{{an attribute specifier sequence attached to a structured binding declaration is a C++2c extension}}
+    auto [ c [[]] alignas(42), d [[]] alignas(42) [[]]] = S{0}; // expected-error 2{{'alignas' attribute only applies to variables, data members and tag types}} \
+                                                                // pre2c-warning  2{{an attribute specifier sequence attached to a structured binding declaration is a C++2c extension}}
+}
+
+
+auto [G1 [[deprecated]], G2 [[deprecated]]] = S{42}; // #deprecated-here
+// pre2c-warning@-1 2{{an attribute specifier sequence attached to a structured binding declaration is a C++2c extension}}
+
+int test() {
+  return G1 + G2; // expected-warning {{'G1' is deprecated}} expected-note@#deprecated-here {{here}} \
+                  // expected-warning {{'G2' is deprecated}} expected-note@#deprecated-here {{here}}
+}
+
+void invalid_attributes() {
+  // pre2c-warning@+1 {{an attribute specifier sequence attached to a structured binding declaration is a C++2c extension}}
+  auto [a alignas(42) // expected-error {{'alignas' attribute only applies to variables, data members and tag types}}
+      [[assume(true), // expected-error {{'assume' attribute cannot be applied to a declaration}}
+        carries_dependency, // expected-error {{'carries_dependency' attribute only applies to parameters, Objective-C methods, and functions}}
+        fallthrough,  // expected-error {{'fallthrough' attribute cannot be applied to a declaration}}
+        likely, // expected-error {{'likely' attribute cannot be applied to a declaration}}
+        unlikely, // expected-error {{'unlikely' attribute cannot be applied to a declaration}}
+        nodiscard,  // expected-warning {{'nodiscard' attribute only applies to Objective-C methods, enums, structs, unions, classes, functions, function pointers, and typedefs}}
+        noreturn,  // expected-error {{'noreturn' attribute only applies to functions}}
+        no_unique_address]], // expected-error {{'no_unique_address' attribute only applies to non-bit-field non-static data members}}
+    b] = S{0};
+}
+
+}
diff --git a/clang/test/Parser/cxx2a-constrained-template-param.cpp b/clang/test/Parser/cxx2a-constrained-template-param.cpp
index 6f14b66419c4..d27f0f8db9b8 100644
--- a/clang/test/Parser/cxx2a-constrained-template-param.cpp
+++ b/clang/test/Parser/cxx2a-constrained-template-param.cpp
@@ -49,4 +49,22 @@ namespace temp
   template<C1 TT, C1 UU = test1> // expected-error{{use of class template 'test1' requires template arguments}}
   // expected-error@-1 2{{concept named in type constraint is not a type concept}}
   using A = TT<int>; // expected-error{{expected ';' after alias declaration}}
-}
\ No newline at end of file
+}
+
+namespace PR67235 {
+
+template <class T>
+concept C = true;
+
+template <auto D>
+struct S {};
+
+// Don't destroy annotation 'C' at the end of the lambda; else we'll run into a
+// use-after-free bug while constructing the type constraint 'C' on 'Default'.
+template <typename Ret, C Default = decltype([] { return Ret(); })>
+void func() {}
+
+template <typename Ret, C Default = S<[] { return Ret(); }>>
+void func2() {}
+
+}
diff --git a/clang/test/Parser/pragma-unroll.cpp b/clang/test/Parser/pragma-unroll.cpp
index f41bd7a18d5a..19066acddcef 100644
--- a/clang/test/Parser/pragma-unroll.cpp
+++ b/clang/test/Parser/pragma-unroll.cpp
@@ -124,3 +124,32 @@ void test(int *List, int Length) {
 
 #pragma unroll
 /* expected-error {{expected statement}} */ }
+
+using size_t = unsigned long long;
+
+template <bool Flag>
+int FailToBuild(int n) {
+  constexpr int N = 100;
+  auto init = [=]() { return Flag ? n : 0UL; };
+  auto cond = [=](size_t ix) { return Flag ? ix != 0 : ix < 10; };
+  auto iter = [=](size_t ix) {
+    return Flag ? ix & ~(1ULL << __builtin_clzll(ix)) : ix + 1;
+  };
+#pragma unroll Flag ? 0 : N // Ok, allow 0.
+  for (size_t ix = init(); cond(ix); ix = iter(ix)) {
+    n *= n;
+  }
+#pragma GCC unroll Flag ? 0 : N // Ok, allow 0.
+  for (size_t ix = init(); cond(ix); ix = iter(ix)) {
+    n *= n;
+  }
+  return n;
+}
+
+int foo(int n) {
+    return FailToBuild<true>(n);
+}
+
+int bar(int n) {
+    return FailToBuild<false>(n);
+}
diff --git a/clang/test/ParserOpenACC/parse-cache-construct.cpp b/clang/test/ParserOpenACC/parse-cache-construct.cpp
index f0a35824696d..f1c71e8b5847 100644
--- a/clang/test/ParserOpenACC/parse-cache-construct.cpp
+++ b/clang/test/ParserOpenACC/parse-cache-construct.cpp
@@ -72,14 +72,12 @@ void use() {
     #pragma acc cache(Arrs.MemArr[3].array[1:4])
   }
   for (int i = 0; i < 10; ++i) {
-    // FIXME: Once we have a new array-section type to represent OpenACC as
-    // well, change this error message.
-    // expected-error@+2{{OpenMP array section is not allowed here}}
+    // expected-error@+2{{OpenACC sub-array is not allowed here}}
     // expected-warning@+1{{OpenACC construct 'cache' not yet implemented, pragma ignored}}
     #pragma acc cache(Arrs.MemArr[3:4].array[1:4])
   }
   for (int i = 0; i < 10; ++i) {
-    // expected-error@+2{{OpenMP array section is not allowed here}}
+    // expected-error@+2{{OpenACC sub-array is not allowed here}}
     // expected-warning@+1{{OpenACC construct 'cache' not yet implemented, pragma ignored}}
     #pragma acc cache(Arrs.MemArr[3:4].array[4])
   }
diff --git a/clang/test/ParserOpenACC/parse-clauses.c b/clang/test/ParserOpenACC/parse-clauses.c
index 799f22b8c120..ee2cb2d1501d 100644
--- a/clang/test/ParserOpenACC/parse-clauses.c
+++ b/clang/test/ParserOpenACC/parse-clauses.c
@@ -482,13 +482,13 @@ void VarListClauses() {
 #pragma acc serial copy(HasMem.MemArr[3].array[1:4]), seq
   for(;;){}
 
-  // expected-error@+3{{OpenMP array section is not allowed here}}
+  // expected-error@+3{{OpenACC sub-array is not allowed here}}
   // expected-warning@+2{{OpenACC clause 'copy' not yet implemented, clause ignored}}
   // expected-warning@+1{{OpenACC clause 'seq' not yet implemented, clause ignored}}
 #pragma acc serial copy(HasMem.MemArr[1:3].array[1]), seq
   for(;;){}
 
-  // expected-error@+3{{OpenMP array section is not allowed here}}
+  // expected-error@+3{{OpenACC sub-array is not allowed here}}
   // expected-warning@+2{{OpenACC clause 'copy' not yet implemented, clause ignored}}
   // expected-warning@+1{{OpenACC clause 'seq' not yet implemented, clause ignored}}
 #pragma acc serial copy(HasMem.MemArr[1:3].array[1:2]), seq
diff --git a/clang/test/Preprocessor/hardware_interference.cpp b/clang/test/Preprocessor/hardware_interference.cpp
new file mode 100644
index 000000000000..f3727aadd32a
--- /dev/null
+++ b/clang/test/Preprocessor/hardware_interference.cpp
@@ -0,0 +1,17 @@
+// RUN: %clang_cc1 -E -dM -D__GCC_CONSTRUCTIVE_SIZE=1000 -D__GCC_DESTRUCTIVE_SIZE=1001 %s -verify -Weverything | FileCheck %s
+// RUN: %clang_cc1 -D__GCC_CONSTRUCTIVE_SIZE=1000 -D__GCC_DESTRUCTIVE_SIZE=1001 %s -verify -Weverything
+// RUN: %clang_cc1 -E -dM -U__GCC_CONSTRUCTIVE_SIZE -U__GCC_DESTRUCTIVE_SIZE %s -verify -Weverything | FileCheck --check-prefix DISABLED %s
+// expected-no-diagnostics
+
+// Validate that we can set a new value on the command line without issuing any
+// diagnostics and that we can disabled the macro on the command line without
+// issuing any diagnostics.
+
+// CHECK: #define __GCC_CONSTRUCTIVE_SIZE 1000
+// CHECK: #define __GCC_DESTRUCTIVE_SIZE 1001
+// DISABLED-NOT: __GCC_CONSTRUCTIVE_SIZE
+// DISABLED-NOT: __GCC_DESTRUCTIVE_SIZE
+
+int main() {
+  return 0;
+}
diff --git a/clang/test/Preprocessor/init-aarch64.c b/clang/test/Preprocessor/init-aarch64.c
index cf96870b27ac..f0845985c9ef 100644
--- a/clang/test/Preprocessor/init-aarch64.c
+++ b/clang/test/Preprocessor/init-aarch64.c
@@ -119,6 +119,8 @@
 // AARCH64-NEXT: #define __FP_FAST_FMA 1
 // AARCH64-NEXT: #define __FP_FAST_FMAF 1
 // AARCH64-NEXT: #define __GCC_ASM_FLAG_OUTPUTS__ 1
+// AARCH64-NEXT: #define __GCC_CONSTRUCTIVE_SIZE {{.+}}
+// AARCH64-NEXT: #define __GCC_DESTRUCTIVE_SIZE {{.+}}
 // AARCH64-NEXT: #define __GCC_HAVE_SYNC_COMPARE_AND_SWAP_1 1
 // AARCH64-NEXT: #define __GCC_HAVE_SYNC_COMPARE_AND_SWAP_16 1
 // AARCH64-NEXT: #define __GCC_HAVE_SYNC_COMPARE_AND_SWAP_2 1
@@ -220,11 +222,11 @@
 // AARCH64-NEXT: #define __LONG_MAX__ 9223372036854775807L
 // AARCH64-NEXT: #define __LONG_WIDTH__ 64
 // AARCH64-NEXT: #define __LP64__ 1
-// AARCH64-NEXT: #define __MEMORY_SCOPE_DEVICE 1 
-// AARCH64-NEXT: #define __MEMORY_SCOPE_SINGLE 4 
-// AARCH64-NEXT: #define __MEMORY_SCOPE_SYSTEM 0 
-// AARCH64-NEXT: #define __MEMORY_SCOPE_WRKGRP 2 
-// AARCH64-NEXT: #define __MEMORY_SCOPE_WVFRNT 3 
+// AARCH64-NEXT: #define __MEMORY_SCOPE_DEVICE 1
+// AARCH64-NEXT: #define __MEMORY_SCOPE_SINGLE 4
+// AARCH64-NEXT: #define __MEMORY_SCOPE_SYSTEM 0
+// AARCH64-NEXT: #define __MEMORY_SCOPE_WRKGRP 2
+// AARCH64-NEXT: #define __MEMORY_SCOPE_WVFRNT 3
 // AARCH64-NEXT: #define __NO_INLINE__ 1
 // AARCH64-NEXT: #define __NO_MATH_ERRNO__ 1
 // AARCH64-NEXT: #define __OBJC_BOOL_IS_BOOL 0
diff --git a/clang/test/Preprocessor/init.c b/clang/test/Preprocessor/init.c
index c4a55efca6f7..2641fee94023 100644
--- a/clang/test/Preprocessor/init.c
+++ b/clang/test/Preprocessor/init.c
@@ -1,3 +1,10 @@
+// RUN: %clang_cc1 -E -dM < /dev/null | FileCheck -match-full-lines -check-prefix INTERFERENCE %s
+//
+// We purposefully do not test the values produced, only that the macros are
+// predefined to some value.
+// INTERFERENCE:#define __GCC_CONSTRUCTIVE_SIZE {{.+}}
+// INTERFERENCE:#define __GCC_DESTRUCTIVE_SIZE {{.+}}
+
 // RUN: %clang_cc1 -E -dM -x assembler-with-cpp < /dev/null | FileCheck -match-full-lines -check-prefix ASM %s
 //
 // ASM:#define __ASSEMBLER__ 1
@@ -1697,6 +1704,8 @@
 // WEBASSEMBLY-NEXT:#define __GCC_ATOMIC_SHORT_LOCK_FREE 2
 // WEBASSEMBLY-NEXT:#define __GCC_ATOMIC_TEST_AND_SET_TRUEVAL 1
 // WEBASSEMBLY-NEXT:#define __GCC_ATOMIC_WCHAR_T_LOCK_FREE 2
+// WEBASSEMBLY-NEXT:#define __GCC_CONSTRUCTIVE_SIZE {{.+}}
+// WEBASSEMBLY-NEXT:#define __GCC_DESTRUCTIVE_SIZE {{.+}}
 // WEBASSEMBLY-NEXT:#define __GCC_HAVE_SYNC_COMPARE_AND_SWAP_1 1
 // WEBASSEMBLY-NEXT:#define __GCC_HAVE_SYNC_COMPARE_AND_SWAP_2 1
 // WEBASSEMBLY-NEXT:#define __GCC_HAVE_SYNC_COMPARE_AND_SWAP_4 1
@@ -1806,11 +1815,11 @@
 // WEBASSEMBLY64-NEXT:#define __LONG_MAX__ 9223372036854775807L
 // WEBASSEMBLY64-NEXT:#define __LONG_WIDTH__ 64
 // WEBASSEMBLY64-NEXT:#define __LP64__ 1
-// WEBASSEMBLY-NEXT:#define __MEMORY_SCOPE_DEVICE 1 
-// WEBASSEMBLY-NEXT:#define __MEMORY_SCOPE_SINGLE 4 
-// WEBASSEMBLY-NEXT:#define __MEMORY_SCOPE_SYSTEM 0 
-// WEBASSEMBLY-NEXT:#define __MEMORY_SCOPE_WRKGRP 2 
-// WEBASSEMBLY-NEXT:#define __MEMORY_SCOPE_WVFRNT 3 
+// WEBASSEMBLY-NEXT:#define __MEMORY_SCOPE_DEVICE 1
+// WEBASSEMBLY-NEXT:#define __MEMORY_SCOPE_SINGLE 4
+// WEBASSEMBLY-NEXT:#define __MEMORY_SCOPE_SYSTEM 0
+// WEBASSEMBLY-NEXT:#define __MEMORY_SCOPE_WRKGRP 2
+// WEBASSEMBLY-NEXT:#define __MEMORY_SCOPE_WVFRNT 3
 // WEBASSEMBLY-NEXT:#define __NO_INLINE__ 1
 // WEBASSEMBLY-NEXT:#define __NO_MATH_ERRNO__ 1
 // WEBASSEMBLY-NEXT:#define __OBJC_BOOL_IS_BOOL 0
@@ -2126,11 +2135,11 @@
 // AVR:#define __LDBL_MIN__ 1.17549435e-38L
 // AVR:#define __LONG_LONG_MAX__ 9223372036854775807LL
 // AVR:#define __LONG_MAX__ 2147483647L
-// AVR:#define __MEMORY_SCOPE_DEVICE 1 
-// AVR:#define __MEMORY_SCOPE_SINGLE 4 
-// AVR:#define __MEMORY_SCOPE_SYSTEM 0 
-// AVR:#define __MEMORY_SCOPE_WRKGRP 2 
-// AVR:#define __MEMORY_SCOPE_WVFRNT 3 
+// AVR:#define __MEMORY_SCOPE_DEVICE 1
+// AVR:#define __MEMORY_SCOPE_SINGLE 4
+// AVR:#define __MEMORY_SCOPE_SYSTEM 0
+// AVR:#define __MEMORY_SCOPE_WRKGRP 2
+// AVR:#define __MEMORY_SCOPE_WVFRNT 3
 // AVR:#define __NO_INLINE__ 1
 // AVR:#define __ORDER_BIG_ENDIAN__ 4321
 // AVR:#define __ORDER_LITTLE_ENDIAN__ 1234
@@ -2422,11 +2431,11 @@
 // RISCV32: #define __LITTLE_ENDIAN__ 1
 // RISCV32: #define __LONG_LONG_MAX__ 9223372036854775807LL
 // RISCV32: #define __LONG_MAX__ 2147483647L
-// RISCV32: #define __MEMORY_SCOPE_DEVICE 1 
-// RISCV32: #define __MEMORY_SCOPE_SINGLE 4 
-// RISCV32: #define __MEMORY_SCOPE_SYSTEM 0 
-// RISCV32: #define __MEMORY_SCOPE_WRKGRP 2 
-// RISCV32: #define __MEMORY_SCOPE_WVFRNT 3 
+// RISCV32: #define __MEMORY_SCOPE_DEVICE 1
+// RISCV32: #define __MEMORY_SCOPE_SINGLE 4
+// RISCV32: #define __MEMORY_SCOPE_SYSTEM 0
+// RISCV32: #define __MEMORY_SCOPE_WRKGRP 2
+// RISCV32: #define __MEMORY_SCOPE_WVFRNT 3
 // RISCV32: #define __NO_INLINE__ 1
 // RISCV32: #define __POINTER_WIDTH__ 32
 // RISCV32: #define __PRAGMA_REDEFINE_EXTNAME 1
@@ -2634,11 +2643,11 @@
 // RISCV64: #define __LONG_LONG_MAX__ 9223372036854775807LL
 // RISCV64: #define __LONG_MAX__ 9223372036854775807L
 // RISCV64: #define __LP64__ 1
-// RISCV64: #define __MEMORY_SCOPE_DEVICE 1 
-// RISCV64: #define __MEMORY_SCOPE_SINGLE 4 
-// RISCV64: #define __MEMORY_SCOPE_SYSTEM 0 
-// RISCV64: #define __MEMORY_SCOPE_WRKGRP 2 
-// RISCV64: #define __MEMORY_SCOPE_WVFRNT 3 
+// RISCV64: #define __MEMORY_SCOPE_DEVICE 1
+// RISCV64: #define __MEMORY_SCOPE_SINGLE 4
+// RISCV64: #define __MEMORY_SCOPE_SYSTEM 0
+// RISCV64: #define __MEMORY_SCOPE_WRKGRP 2
+// RISCV64: #define __MEMORY_SCOPE_WVFRNT 3
 // RISCV64: #define __NO_INLINE__ 1
 // RISCV64: #define __POINTER_WIDTH__ 64
 // RISCV64: #define __PRAGMA_REDEFINE_EXTNAME 1
diff --git a/clang/test/Preprocessor/predefined-macros-hlsl.hlsl b/clang/test/Preprocessor/predefined-macros-hlsl.hlsl
index 251362cd03c0..cc5233fbcb2a 100644
--- a/clang/test/Preprocessor/predefined-macros-hlsl.hlsl
+++ b/clang/test/Preprocessor/predefined-macros-hlsl.hlsl
@@ -1,14 +1,19 @@
-// RUN: %clang_cc1 %s -E -dM -o - -triple dxil-pc-shadermodel6.0-amplification | FileCheck -match-full-lines %s --check-prefixes=CHECK,AMPLIFICATION
-// RUN: %clang_cc1 %s -E -dM -o - -triple dxil-pc-shadermodel6.0-compute | FileCheck -match-full-lines %s --check-prefixes=CHECK,COMPUTE
-// RUN: %clang_cc1 %s -E -dM -o - -triple dxil-pc-shadermodel6.0-domain | FileCheck -match-full-lines %s --check-prefixes=CHECK,DOMAIN
-// RUN: %clang_cc1 %s -E -dM -o - -triple dxil-pc-shadermodel6.0-geometry | FileCheck -match-full-lines %s --check-prefixes=CHECK,GEOMETRY
-// RUN: %clang_cc1 %s -E -dM -o - -triple dxil-pc-shadermodel6.0-hull | FileCheck -match-full-lines %s --check-prefixes=CHECK,HULL
-// RUN: %clang_cc1 %s -E -dM -o - -triple dxil-pc-shadermodel6.0-library | FileCheck -match-full-lines %s --check-prefixes=CHECK,LIBRARY
-// RUN: %clang_cc1 %s -E -dM -o - -triple dxil-pc-shadermodel6.0-mesh | FileCheck -match-full-lines %s --check-prefixes=CHECK,MESH
-// RUN: %clang_cc1 %s -E -dM -o - -triple dxil-pc-shadermodel6.0-pixel | FileCheck -match-full-lines %s --check-prefixes=CHECK,PIXEL
-// RUN: %clang_cc1 %s -E -dM -o - -triple dxil-pc-shadermodel6.0-vertex | FileCheck -match-full-lines %s --check-prefixes=CHECK,VERTEX
+// RUN: %clang_cc1 %s -E -dM -o - -triple dxil-pc-shadermodel6.0-amplification | FileCheck -match-full-lines %s --check-prefixes=CHECK,AMPLIFICATION,NOHALF
+// RUN: %clang_cc1 %s -E -dM -o - -triple dxil-pc-shadermodel6.0-compute | FileCheck -match-full-lines %s --check-prefixes=CHECK,COMPUTE,NOHALF
+// RUN: %clang_cc1 %s -E -dM -o - -triple dxil-pc-shadermodel6.0-domain | FileCheck -match-full-lines %s --check-prefixes=CHECK,DOMAIN,NOHALF
+// RUN: %clang_cc1 %s -E -dM -o - -triple dxil-pc-shadermodel6.0-geometry | FileCheck -match-full-lines %s --check-prefixes=CHECK,GEOMETRY,NOHALF
+// RUN: %clang_cc1 %s -E -dM -o - -triple dxil-pc-shadermodel6.0-hull | FileCheck -match-full-lines %s --check-prefixes=CHECK,HULL,NOHALF
+// RUN: %clang_cc1 %s -E -dM -o - -triple dxil-pc-shadermodel6.0-library | FileCheck -match-full-lines %s --check-prefixes=CHECK,LIBRARY,NOHALF
+// RUN: %clang_cc1 %s -E -dM -o - -triple dxil-pc-shadermodel6.0-mesh | FileCheck -match-full-lines %s --check-prefixes=CHECK,MESH,NOHALF
+// RUN: %clang_cc1 %s -E -dM -o - -triple dxil-pc-shadermodel6.0-pixel | FileCheck -match-full-lines %s --check-prefixes=CHECK,PIXEL,NOHALF
+// RUN: %clang_cc1 %s -E -dM -o - -triple dxil-pc-shadermodel6.0-vertex | FileCheck -match-full-lines %s --check-prefixes=CHECK,VERTEX,NOHALF
+// RUN: %clang_cc1 %s -E -dM -o - -triple dxil-pc-shadermodel6.3-vertex -fnative-half-type | FileCheck -match-full-lines %s --check-prefixes=CHECK,VERTEX,HALF
+
+// HALF: #define __HLSL_ENABLE_16_BIT 1
+// NOHALF-NOT: __HLSL_ENABLE_16_BIT
 
 // CHECK: #define __HLSL_VERSION 2021
+
 // CHECK: #define __SHADER_STAGE_AMPLIFICATION 14
 // CHECK: #define __SHADER_STAGE_COMPUTE 5
 // CHECK: #define __SHADER_STAGE_DOMAIN 4
diff --git a/clang/test/Preprocessor/predefined-win-macros.c b/clang/test/Preprocessor/predefined-win-macros.c
index b830dc39d477..14e2f584bd09 100644
--- a/clang/test/Preprocessor/predefined-win-macros.c
+++ b/clang/test/Preprocessor/predefined-win-macros.c
@@ -3,7 +3,7 @@
 // RUN: %clang_cc1 %s -x c++ -E -dM -triple x86_64-pc-win32 -fms-extensions -fms-compatibility \
 // RUN:     -fms-compatibility-version=19.00 -std=c++14 -o - | FileCheck -match-full-lines %s --check-prefix=CHECK-MS64
 // RUN: %clang_cc1 %s -x c++ -E -dM -triple x86_64-pc-win32 -fms-extensions -fms-compatibility \
-// RUN:     -fms-compatibility-version=19.00 -std=c++14 -o - | grep GCC | count 5
+// RUN:     -fms-compatibility-version=19.00 -std=c++14 -o - | grep GCC | count 7
 // CHECK-MS64: #define _INTEGRAL_MAX_BITS 64
 // CHECK-MS64: #define _ISO_VOLATILE 1
 // CHECK-MS64: #define _MSC_EXTENSIONS 1
@@ -26,7 +26,7 @@
 // RUN: %clang_cc1 %s -x c++ -E -dM -triple i686-pc-win32 -fms-extensions -fms-compatibility \
 // RUN:     -fms-compatibility-version=19.00 -std=c++17 -o - | FileCheck -match-full-lines %s --check-prefix=CHECK-MS
 // RUN: %clang_cc1 %s -x c++ -E -dM -triple i686-pc-win32 -fms-extensions -fms-compatibility \
-// RUN:     -fms-compatibility-version=19.00 -std=c++17 -o - | grep GCC | count 5
+// RUN:     -fms-compatibility-version=19.00 -std=c++17 -o - | grep GCC | count 7
 // CHECK-MS: #define _INTEGRAL_MAX_BITS 64
 // CHECK-MS: #define _ISO_VOLATILE 1
 // CHECK-MS: #define _MSC_EXTENSIONS 1
@@ -39,6 +39,8 @@
 // CHECK-MS-NOT: GNU
 // CHECK-MS-NOT: GXX
 // CHECK-MS: #define __GCC_ASM_FLAG_OUTPUTS__ 1
+// CHECK-MS: #define __GCC_CONSTRUCTIVE_SIZE {{.+}}
+// CHECK-MS: #define __GCC_DESTRUCTIVE_SIZE {{.+}}
 // CHECK-MS: #define __GCC_HAVE_SYNC_COMPARE_AND_SWAP_1 1
 // CHECK-MS: #define __GCC_HAVE_SYNC_COMPARE_AND_SWAP_2 1
 // CHECK-MS: #define __GCC_HAVE_SYNC_COMPARE_AND_SWAP_4 1
diff --git a/clang/test/Preprocessor/riscv-target-features.c b/clang/test/Preprocessor/riscv-target-features.c
index 646043681fe3..ee4f81cd654b 100644
--- a/clang/test/Preprocessor/riscv-target-features.c
+++ b/clang/test/Preprocessor/riscv-target-features.c
@@ -174,6 +174,7 @@
 // CHECK-NOT: __riscv_ssqosid{{.*$}}
 // CHECK-NOT: __riscv_supm{{.*$}}
 // CHECK-NOT: __riscv_zaamo {{.*$}}
+// CHECK-NOT: __riscv_zabha {{.*$}}
 // CHECK-NOT: __riscv_zalasr {{.*$}}
 // CHECK-NOT: __riscv_zalrsc {{.*$}}
 // CHECK-NOT: __riscv_zfbfmin {{.*$}}
@@ -698,10 +699,10 @@
 // CHECK-ZA64RS-EXT: __riscv_za64rs 1000000{{$}}
 
 // RUN: %clang --target=riscv32 \
-// RUN:   -march=rv32i_zacas1p0 -E -dM %s \
+// RUN:   -march=rv32ia_zacas1p0 -E -dM %s \
 // RUN:   -o - | FileCheck --check-prefix=CHECK-ZACAS-EXT %s
 // RUN: %clang --target=riscv64 \
-// RUN:   -march=rv64i_zacas1p0 -E -dM %s \
+// RUN:   -march=rv64ia_zacas1p0 -E -dM %s \
 // RUN:   -o - | FileCheck --check-prefix=CHECK-ZACAS-EXT %s
 // CHECK-ZACAS-EXT: __riscv_zacas 1000000{{$}}
 
@@ -1553,6 +1554,14 @@
 // CHECK-ZAAMO-EXT: __riscv_zaamo 2000{{$}}
 
 // RUN: %clang --target=riscv32 -menable-experimental-extensions \
+// RUN:   -march=rv32ia_zabha1p0 -E -dM %s \
+// RUN:   -o - | FileCheck --check-prefix=CHECK-ZABHA-EXT %s
+// RUN: %clang --target=riscv64 -menable-experimental-extensions \
+// RUN:   -march=rv64ia_zabha1p0 -E -dM %s \
+// RUN:   -o - | FileCheck --check-prefix=CHECK-ZABHA-EXT %s
+// CHECK-ZABHA-EXT: __riscv_zabha 1000000{{$}}
+
+// RUN: %clang --target=riscv32 -menable-experimental-extensions \
 // RUN:   -march=rv32i_zalasr0p1 -E -dM %s \
 // RUN:   -o - | FileCheck --check-prefix=CHECK-ZALASR-EXT %s
 // RUN: %clang --target=riscv64 -menable-experimental-extensions \
diff --git a/clang/test/Preprocessor/wasm-target-features.c b/clang/test/Preprocessor/wasm-target-features.c
index eccd432aa8ee..32e24ad1b716 100644
--- a/clang/test/Preprocessor/wasm-target-features.c
+++ b/clang/test/Preprocessor/wasm-target-features.c
@@ -1,38 +1,29 @@
 // RUN: %clang -E -dM %s -o - 2>&1 \
-// RUN:     -target wasm32-unknown-unknown -msimd128 \
-// RUN:   | FileCheck %s -check-prefix=SIMD128
-// RUN: %clang -E -dM %s -o - 2>&1 \
-// RUN:     -target wasm64-unknown-unknown -msimd128 \
-// RUN:   | FileCheck %s -check-prefix=SIMD128
-//
-// SIMD128:#define __wasm_simd128__ 1{{$}}
-
-// RUN: %clang -E -dM %s -o - 2>&1 \
-// RUN:     -target wasm32-unknown-unknown -mrelaxed-simd \
-// RUN:   | FileCheck %s -check-prefix=RELAXED-SIMD
+// RUN:     -target wasm32-unknown-unknown -matomics \
+// RUN:   | FileCheck %s -check-prefix=ATOMICS
 // RUN: %clang -E -dM %s -o - 2>&1 \
-// RUN:     -target wasm64-unknown-unknown -mrelaxed-simd \
-// RUN:   | FileCheck %s -check-prefix=RELAXED-SIMD
+// RUN:     -target wasm64-unknown-unknown -matomics \
+// RUN:   | FileCheck %s -check-prefix=ATOMICS
 //
-// RELAXED-SIMD:#define __wasm_relaxed_simd__ 1{{$}}
+// ATOMICS: #define __wasm_atomics__ 1{{$}}
 
 // RUN: %clang -E -dM %s -o - 2>&1 \
-// RUN:     -target wasm32-unknown-unknown -mnontrapping-fptoint \
-// RUN:   | FileCheck %s -check-prefix=NONTRAPPING-FPTOINT
+// RUN:     -target wasm32-unknown-unknown -pthread \
+// RUN:   | FileCheck %s -check-prefix=PTHREAD
 // RUN: %clang -E -dM %s -o - 2>&1 \
-// RUN:     -target wasm64-unknown-unknown -mnontrapping-fptoint \
-// RUN:   | FileCheck %s -check-prefix=NONTRAPPING-FPTOINT
+// RUN:     -target wasm64-unknown-unknown -pthread \
+// RUN:   | FileCheck %s -check-prefix=PTHREAD
 //
-// NONTRAPPING-FPTOINT:#define __wasm_nontrapping_fptoint__ 1{{$}}
+// PTHREAD: #define __wasm_atomics__ 1{{$}}
 
 // RUN: %clang -E -dM %s -o - 2>&1 \
-// RUN:     -target wasm32-unknown-unknown -msign-ext \
-// RUN:   | FileCheck %s -check-prefix=SIGN-EXT
+// RUN:     -target wasm32-unknown-unknown -mbulk-memory \
+// RUN:   | FileCheck %s -check-prefix=BULK-MEMORY
 // RUN: %clang -E -dM %s -o - 2>&1 \
-// RUN:     -target wasm64-unknown-unknown -msign-ext \
-// RUN:   | FileCheck %s -check-prefix=SIGN-EXT
+// RUN:     -target wasm64-unknown-unknown -mbulk-memory \
+// RUN:   | FileCheck %s -check-prefix=BULK-MEMORY
 //
-// SIGN-EXT:#define __wasm_sign_ext__ 1{{$}}
+// BULK-MEMORY: #define __wasm_bulk_memory__ 1{{$}}
 
 // RUN: %clang -E -dM %s -o - 2>&1 \
 // RUN:     -target wasm32-unknown-unknown -mexception-handling \
@@ -41,34 +32,34 @@
 // RUN:     -target wasm64-unknown-unknown -mexception-handling \
 // RUN:   | FileCheck %s -check-prefix=EXCEPTION-HANDLING
 //
-// EXCEPTION-HANDLING:#define __wasm_exception_handling__ 1{{$}}
+// EXCEPTION-HANDLING: #define __wasm_exception_handling__ 1{{$}}
 
 // RUN: %clang -E -dM %s -o - 2>&1 \
-// RUN:     -target wasm32-unknown-unknown -mbulk-memory \
-// RUN:   | FileCheck %s -check-prefix=BULK-MEMORY
+// RUN:     -target wasm32-unknown-unknown -mextended-const \
+// RUN:   | FileCheck %s -check-prefix=EXTENDED-CONST
 // RUN: %clang -E -dM %s -o - 2>&1 \
-// RUN:     -target wasm64-unknown-unknown -mbulk-memory \
-// RUN:   | FileCheck %s -check-prefix=BULK-MEMORY
+// RUN:     -target wasm64-unknown-unknown -mextended-const \
+// RUN:   | FileCheck %s -check-prefix=EXTENDED-CONST
 //
-// BULK-MEMORY:#define __wasm_bulk_memory__ 1{{$}}
+// EXTENDED-CONST: #define __wasm_extended_const__ 1{{$}}
 
 // RUN: %clang -E -dM %s -o - 2>&1 \
-// RUN:     -target wasm32-unknown-unknown -matomics \
-// RUN:   | FileCheck %s -check-prefix=ATOMICS
+// RUN:     -target wasm32-unknown-unknown -mmultimemory \
+// RUN:   | FileCheck %s -check-prefix=MULTIMEMORY
 // RUN: %clang -E -dM %s -o - 2>&1 \
-// RUN:     -target wasm64-unknown-unknown -matomics \
-// RUN:   | FileCheck %s -check-prefix=ATOMICS
+// RUN:     -target wasm64-unknown-unknown -mmultimemory \
+// RUN:   | FileCheck %s -check-prefix=MULTIMEMORY
 //
-// ATOMICS:#define __wasm_atomics__ 1{{$}}
+// MULTIMEMORY: #define __wasm_multimemory__ 1{{$}}
 
 // RUN: %clang -E -dM %s -o - 2>&1 \
-// RUN:     -target wasm32-unknown-unknown -pthread \
-// RUN:   | FileCheck %s -check-prefix=PTHREAD
+// RUN:     -target wasm32-unknown-unknown -mmultivalue \
+// RUN:   | FileCheck %s -check-prefix=MULTIVALUE
 // RUN: %clang -E -dM %s -o - 2>&1 \
-// RUN:     -target wasm64-unknown-unknown -pthread \
-// RUN:   | FileCheck %s -check-prefix=PTHREAD
+// RUN:     -target wasm64-unknown-unknown -mmultivalue \
+// RUN:   | FileCheck %s -check-prefix=MULTIVALUE
 //
-// PTHREAD:#define __wasm_atomics__ 1{{$}}
+// MULTIVALUE: #define __wasm_multivalue__ 1{{$}}
 
 // RUN: %clang -E -dM %s -o - 2>&1 \
 // RUN:     -target wasm32-unknown-unknown -mmutable-globals \
@@ -77,53 +68,61 @@
 // RUN:     -target wasm64-unknown-unknown -mmutable-globals \
 // RUN:   | FileCheck %s -check-prefix=MUTABLE-GLOBALS
 //
-// MUTABLE-GLOBALS:#define __wasm_mutable_globals__ 1{{$}}
+// MUTABLE-GLOBALS: #define __wasm_mutable_globals__ 1{{$}}
 
 // RUN: %clang -E -dM %s -o - 2>&1 \
-// RUN:     -target wasm32-unknown-unknown -mmultivalue \
-// RUN:   | FileCheck %s -check-prefix=MULTIVALUE
+// RUN:     -target wasm32-unknown-unknown -mnontrapping-fptoint \
+// RUN:   | FileCheck %s -check-prefix=NONTRAPPING-FPTOINT
 // RUN: %clang -E -dM %s -o - 2>&1 \
-// RUN:     -target wasm64-unknown-unknown -mmultivalue \
-// RUN:   | FileCheck %s -check-prefix=MULTIVALUE
+// RUN:     -target wasm64-unknown-unknown -mnontrapping-fptoint \
+// RUN:   | FileCheck %s -check-prefix=NONTRAPPING-FPTOINT
 //
-// MULTIVALUE:#define __wasm_multivalue__ 1{{$}}
+// NONTRAPPING-FPTOINT: #define __wasm_nontrapping_fptoint__ 1{{$}}
 
 // RUN: %clang -E -dM %s -o - 2>&1 \
-// RUN:     -target wasm32-unknown-unknown -mtail-call \
-// RUN:   | FileCheck %s -check-prefix=TAIL-CALL
-// RUN: %clang -E -dM %s -o - 2>&1 \
-// RUN:     -target wasm64-unknown-unknown -mtail-call \
-// RUN:   | FileCheck %s -check-prefix=TAIL-CALL
-//
-// TAIL-CALL:#define __wasm_tail_call__ 1{{$}}
-//
-// RUN: %clang -E -dM %s -o - 2>&1 \
 // RUN:     -target wasm32-unknown-unknown -mreference-types \
 // RUN:   | FileCheck %s -check-prefix=REFERENCE-TYPES
 // RUN: %clang -E -dM %s -o - 2>&1 \
 // RUN:     -target wasm64-unknown-unknown -mreference-types \
 // RUN:   | FileCheck %s -check-prefix=REFERENCE-TYPES
 //
-// REFERENCE-TYPES:#define __wasm_reference_types__ 1{{$}}
-//
+// REFERENCE-TYPES: #define __wasm_reference_types__ 1{{$}}
+
 // RUN: %clang -E -dM %s -o - 2>&1 \
-// RUN:     -target wasm32-unknown-unknown -mextended-const \
-// RUN:   | FileCheck %s -check-prefix=EXTENDED-CONST
+// RUN:     -target wasm32-unknown-unknown -mrelaxed-simd \
+// RUN:   | FileCheck %s -check-prefix=RELAXED-SIMD
 // RUN: %clang -E -dM %s -o - 2>&1 \
-// RUN:     -target wasm64-unknown-unknown -mextended-const \
-// RUN:   | FileCheck %s -check-prefix=EXTENDED-CONST
+// RUN:     -target wasm64-unknown-unknown -mrelaxed-simd \
+// RUN:   | FileCheck %s -check-prefix=RELAXED-SIMD
 //
-// EXTENDED-CONST:#define __wasm_extended_const__ 1{{$}}
+// RELAXED-SIMD: #define __wasm_relaxed_simd__ 1{{$}}
+
+// RUN: %clang -E -dM %s -o - 2>&1 \
+// RUN:     -target wasm32-unknown-unknown -msign-ext \
+// RUN:   | FileCheck %s -check-prefix=SIGN-EXT
+// RUN: %clang -E -dM %s -o - 2>&1 \
+// RUN:     -target wasm64-unknown-unknown -msign-ext \
+// RUN:   | FileCheck %s -check-prefix=SIGN-EXT
 //
+// SIGN-EXT: #define __wasm_sign_ext__ 1{{$}}
+
 // RUN: %clang -E -dM %s -o - 2>&1 \
-// RUN:     -target wasm32-unknown-unknown -mmultimemory \
-// RUN:   | FileCheck %s -check-prefix=MULTIMEMORY
+// RUN:     -target wasm32-unknown-unknown -msimd128 \
+// RUN:   | FileCheck %s -check-prefix=SIMD128
 // RUN: %clang -E -dM %s -o - 2>&1 \
-// RUN:     -target wasm64-unknown-unknown -mmultimemory \
-// RUN:   | FileCheck %s -check-prefix=MULTIMEMORY
+// RUN:     -target wasm64-unknown-unknown -msimd128 \
+// RUN:   | FileCheck %s -check-prefix=SIMD128
 //
-// MULTIMEMORY:#define __wasm_multimemory__ 1{{$}}
+// SIMD128: #define __wasm_simd128__ 1{{$}}
+
+// RUN: %clang -E -dM %s -o - 2>&1 \
+// RUN:     -target wasm32-unknown-unknown -mtail-call \
+// RUN:   | FileCheck %s -check-prefix=TAIL-CALL
+// RUN: %clang -E -dM %s -o - 2>&1 \
+// RUN:     -target wasm64-unknown-unknown -mtail-call \
+// RUN:   | FileCheck %s -check-prefix=TAIL-CALL
 //
+// TAIL-CALL: #define __wasm_tail_call__ 1{{$}}
 
 // RUN: %clang -E -dM %s -o - 2>&1 \
 // RUN:     -target wasm32-unknown-unknown -mcpu=mvp \
@@ -132,60 +131,77 @@
 // RUN:     -target wasm64-unknown-unknown -mcpu=mvp \
 // RUN:   | FileCheck %s -check-prefix=MVP
 //
-// MVP-NOT:#define __wasm_simd128__
-// MVP-NOT:#define __wasm_nontrapping_fptoint__
-// MVP-NOT:#define __wasm_sign_ext__
-// MVP-NOT:#define __wasm_exception_handling__
-// MVP-NOT:#define __wasm_bulk_memory__
-// MVP-NOT:#define __wasm_atomics__
-// MVP-NOT:#define __wasm_mutable_globals__
-// MVP-NOT:#define __wasm_multivalue__
-// MVP-NOT:#define __wasm_tail_call__
-// MVP-NOT:#define __wasm_reference_types__
-// MVP-NOT:#define __wasm_extended_const__
-// MVP-NOT:#define __wasm_multimemory__
-// MVP-NOT:#define __wasm_relaxed_simd__
+// MVP-NOT: #define __wasm_atomics__ 1{{$}}
+// MVP-NOT: #define __wasm_bulk_memory__ 1{{$}}
+// MVP-NOT: #define __wasm_exception_handling__ 1{{$}}
+// MVP-NOT: #define __wasm_extended_const__ 1{{$}}
+// MVP-NOT: #define __wasm_multimemory__ 1{{$}}
+// MVP-NOT: #define __wasm_multivalue__ 1{{$}}
+// MVP-NOT: #define __wasm_mutable_globals__ 1{{$}}
+// MVP-NOT: #define __wasm_nontrapping_fptoint__ 1{{$}}
+// MVP-NOT: #define __wasm_reference_types__ 1{{$}}
+// MVP-NOT: #define __wasm_relaxed_simd__ 1{{$}}
+// MVP-NOT: #define __wasm_sign_ext__ 1{{$}}
+// MVP-NOT: #define __wasm_simd128__ 1{{$}}
+// MVP-NOT: #define __wasm_tail_call__ 1{{$}}
 
 // RUN: %clang -E -dM %s -o - 2>&1 \
 // RUN:     -target wasm32-unknown-unknown -mcpu=generic \
+// RUN:   | FileCheck %s -check-prefix=GENERIC-INCLUDE
+// RUN: %clang -E -dM %s -o - 2>&1 \
+// RUN:     -target wasm64-unknown-unknown -mcpu=generic \
+// RUN:   | FileCheck %s -check-prefix=GENERIC-INCLUDE
+//
+// GENERIC-INCLUDE-DAG: #define __wasm_mutable_globals__ 1{{$}}
+// GENERIC-INCLUDE-DAG: #define __wasm_sign_ext__ 1{{$}}
+//
+// RUN: %clang -E -dM %s -o - 2>&1 \
+// RUN:     -target wasm32-unknown-unknown -mcpu=generic \
 // RUN:   | FileCheck %s -check-prefix=GENERIC
 // RUN: %clang -E -dM %s -o - 2>&1 \
 // RUN:     -target wasm64-unknown-unknown -mcpu=generic \
 // RUN:   | FileCheck %s -check-prefix=GENERIC
 //
-// GENERIC-DAG:#define __wasm_sign_ext__ 1{{$}}
-// GENERIC-DAG:#define __wasm_mutable_globals__ 1{{$}}
-// GENERIC-NOT:#define __wasm_nontrapping_fptoint__ 1{{$}}
-// GENERIC-NOT:#define __wasm_bulk_memory__ 1{{$}}
-// GENERIC-NOT:#define __wasm_simd128__ 1{{$}}
-// GENERIC-NOT:#define __wasm_atomics__ 1{{$}}
-// GENERIC-NOT:#define __wasm_tail_call__ 1{{$}}
-// GENERIC-NOT:#define __wasm_multimemory__ 1{{$}}
-// GENERIC-NOT:#define __wasm_exception_handling__ 1{{$}}
-// GENERIC-NOT:#define __wasm_multivalue__ 1{{$}}
-// GENERIC-NOT:#define __wasm_reference_types__ 1{{$}}
-// GENERIC-NOT:#define __wasm_extended_const__ 1{{$}}
+// GENERIC-NOT: #define __wasm_atomics__ 1{{$}}
+// GENERIC-NOT: #define __wasm_bulk_memory__ 1{{$}}
+// GENERIC-NOT: #define __wasm_exception_handling__ 1{{$}}
+// GENERIC-NOT: #define __wasm_extended_const__ 1{{$}}
+// GENERIC-NOT: #define __wasm_multimemory__ 1{{$}}
+// GENERIC-NOT: #define __wasm_multivalue__ 1{{$}}
+// GENERIC-NOT: #define __wasm_nontrapping_fptoint__ 1{{$}}
+// GENERIC-NOT: #define __wasm_reference_types__ 1{{$}}
+// GENERIC-NOT: #define __wasm_relaxed_simd__ 1{{$}}
+// GENERIC-NOT: #define __wasm_simd128__ 1{{$}}
+// GENERIC-NOT: #define __wasm_tail_call__ 1{{$}}
 
 // RUN: %clang -E -dM %s -o - 2>&1 \
 // RUN:     -target wasm32-unknown-unknown -mcpu=bleeding-edge \
+// RUN:   | FileCheck %s -check-prefix=BLEEDING-EDGE-INCLUDE
+// RUN: %clang -E -dM %s -o - 2>&1 \
+// RUN:     -target wasm64-unknown-unknown -mcpu=bleeding-edge \
+// RUN:   | FileCheck %s -check-prefix=BLEEDING-EDGE-INCLUDE
+//
+// BLEEDING-EDGE-INCLUDE-DAG: #define __wasm_atomics__ 1{{$}}
+// BLEEDING-EDGE-INCLUDE-DAG: #define __wasm_bulk_memory__ 1{{$}}
+// BLEEDING-EDGE-INCLUDE-DAG: #define __wasm_multimemory__ 1{{$}}
+// BLEEDING-EDGE-INCLUDE-DAG: #define __wasm_mutable_globals__ 1{{$}}
+// BLEEDING-EDGE-INCLUDE-DAG: #define __wasm_nontrapping_fptoint__ 1{{$}}
+// BLEEDING-EDGE-INCLUDE-DAG: #define __wasm_reference_types__ 1{{$}}
+// BLEEDING-EDGE-INCLUDE-DAG: #define __wasm_sign_ext__ 1{{$}}
+// BLEEDING-EDGE-INCLUDE-DAG: #define __wasm_simd128__ 1{{$}}
+// BLEEDING-EDGE-INCLUDE-DAG: #define __wasm_tail_call__ 1{{$}}
+//
+// RUN: %clang -E -dM %s -o - 2>&1 \
+// RUN:     -target wasm32-unknown-unknown -mcpu=bleeding-edge \
 // RUN:   | FileCheck %s -check-prefix=BLEEDING-EDGE
 // RUN: %clang -E -dM %s -o - 2>&1 \
 // RUN:     -target wasm64-unknown-unknown -mcpu=bleeding-edge \
 // RUN:   | FileCheck %s -check-prefix=BLEEDING-EDGE
 //
-// BLEEDING-EDGE-DAG:#define __wasm_nontrapping_fptoint__ 1{{$}}
-// BLEEDING-EDGE-DAG:#define __wasm_sign_ext__ 1{{$}}
-// BLEEDING-EDGE-DAG:#define __wasm_bulk_memory__ 1{{$}}
-// BLEEDING-EDGE-DAG:#define __wasm_simd128__ 1{{$}}
-// BLEEDING-EDGE-DAG:#define __wasm_atomics__ 1{{$}}
-// BLEEDING-EDGE-DAG:#define __wasm_mutable_globals__ 1{{$}}
-// BLEEDING-EDGE-DAG:#define __wasm_tail_call__ 1{{$}}
-// BLEEDING-EDGE-DAG:#define __wasm_multimemory__ 1{{$}}
-// BLEEDING-EDGE-NOT:#define __wasm_exception_handling__ 1{{$}}
-// BLEEDING-EDGE-NOT:#define __wasm_multivalue__ 1{{$}}
-// BLEEDING-EDGE-NOT:#define __wasm_reference_types__ 1{{$}}
-// BLEEDING-EDGE-NOT:#define __wasm_extended_const__ 1{{$}}
-// BLEEDING-EDGE-NOT:#define __wasm_relaxed_simd__ 1{{$}}
+// BLEEDING-EDGE-NOT: #define __wasm_exception_handling__ 1{{$}}
+// BLEEDING-EDGE-NOT: #define __wasm_extended_const__ 1{{$}}
+// BLEEDING-EDGE-NOT: #define __wasm_multivalue__ 1{{$}}
+// BLEEDING-EDGE-NOT: #define __wasm_relaxed_simd__ 1{{$}}
 
 // RUN: %clang -E -dM %s -o - 2>&1 \
 // RUN:     -target wasm32-unknown-unknown -mcpu=bleeding-edge -mno-simd128 \
@@ -194,4 +210,4 @@
 // RUN:     -target wasm64-unknown-unknown -mcpu=bleeding-edge -mno-simd128 \
 // RUN:   | FileCheck %s -check-prefix=BLEEDING-EDGE-NO-SIMD128
 //
-// BLEEDING-EDGE-NO-SIMD128-NOT:#define __wasm_simd128__
+// BLEEDING-EDGE-NO-SIMD128-NOT: #define __wasm_simd128__ 1{{$}}
diff --git a/clang/test/Sema/aarch64-incompat-sm-builtin-calls.c b/clang/test/Sema/aarch64-incompat-sm-builtin-calls.c
index 6a1feeb9bf53..e55e84a61034 100644
--- a/clang/test/Sema/aarch64-incompat-sm-builtin-calls.c
+++ b/clang/test/Sema/aarch64-incompat-sm-builtin-calls.c
@@ -33,7 +33,8 @@ svuint32_t incompat_sve_sm(svbool_t pg, svuint32_t a, int16_t b) __arm_streaming
   return __builtin_sve_svld1_gather_u32base_index_u32(pg, a, b);
 }
 
-// expected-warning@+1 {{passing/returning a VL-dependent argument to/from a __arm_locally_streaming function. The streaming and non-streaming vector lengths may be different}}
+// expected-warning@+2 {{returning a VL-dependent argument from a locally streaming function is undefined behaviour when the streaming and non-streaming vector lengths are different at runtime}}
+// expected-warning@+1 {{passing a VL-dependent argument to a locally streaming function is undefined behaviour when the streaming and non-streaming vector lengths are different at runtime}}
 __arm_locally_streaming svuint32_t incompat_sve_ls(svbool_t pg, svuint32_t a, int64_t b) {
   // expected-warning@+1 {{builtin call has undefined behaviour when called from a streaming function}}
   return __builtin_sve_svld1_gather_u32base_index_u32(pg, a, b);
@@ -49,7 +50,8 @@ svuint32_t incompat_sve2_sm(svbool_t pg, svuint32_t a, int64_t b) __arm_streamin
   return __builtin_sve_svldnt1_gather_u32base_index_u32(pg, a, b);
 }
 
-// expected-warning@+1 {{passing/returning a VL-dependent argument to/from a __arm_locally_streaming function. The streaming and non-streaming vector lengths may be different}}
+// expected-warning@+2 {{returning a VL-dependent argument from a locally streaming function is undefined behaviour when the streaming and non-streaming vector lengths are different at runtime}}
+// expected-warning@+1 {{passing a VL-dependent argument to a locally streaming function is undefined behaviour when the streaming and non-streaming vector lengths are different at runtime}}
 __arm_locally_streaming svuint32_t incompat_sve2_ls(svbool_t pg, svuint32_t a, int64_t b) {
   // expected-warning@+1 {{builtin call has undefined behaviour when called from a streaming function}}
   return __builtin_sve_svldnt1_gather_u32base_index_u32(pg, a, b);
@@ -70,7 +72,8 @@ svfloat64_t streaming_caller_sve(svbool_t pg, svfloat64_t a, float64_t b) __arm_
   return svadd_n_f64_m(pg, a, b);
 }
 
-// expected-warning@+1 {{passing/returning a VL-dependent argument to/from a __arm_locally_streaming function. The streaming and non-streaming vector lengths may be different}}
+// expected-warning@+2 {{returning a VL-dependent argument from a locally streaming function is undefined behaviour when the streaming and non-streaming vector lengths are different at runtime}}
+// expected-warning@+1 {{passing a VL-dependent argument to a locally streaming function is undefined behaviour when the streaming and non-streaming vector lengths are different at runtime}}
 __arm_locally_streaming svfloat64_t locally_streaming_caller_sve(svbool_t pg, svfloat64_t a, float64_t b) {
   // expected-no-warning
   return svadd_n_f64_m(pg, a, b);
@@ -86,7 +89,8 @@ svint16_t streaming_caller_sve2(svint16_t op1, svint16_t op2) __arm_streaming {
   return svmul_lane_s16(op1, op2, 0);
 }
 
-// expected-warning@+1 {{passing/returning a VL-dependent argument to/from a __arm_locally_streaming function. The streaming and non-streaming vector lengths may be different}}
+// expected-warning@+2 {{returning a VL-dependent argument from a locally streaming function is undefined behaviour when the streaming and non-streaming vector lengths are different at runtime}}
+// expected-warning@+1 {{passing a VL-dependent argument to a locally streaming function is undefined behaviour when the streaming and non-streaming vector lengths are different at runtime}}
 __arm_locally_streaming svint16_t locally_streaming_caller_sve2(svint16_t op1, svint16_t op2) {
   // expected-no-warning
   return svmul_lane_s16(op1, op2, 0);
diff --git a/clang/test/Sema/aarch64-sme-func-attrs.c b/clang/test/Sema/aarch64-sme-func-attrs.c
index 12de16509ccb..3d90723d32f1 100644
--- a/clang/test/Sema/aarch64-sme-func-attrs.c
+++ b/clang/test/Sema/aarch64-sme-func-attrs.c
@@ -509,73 +509,73 @@ void sme_no_streaming_with_vl_arg(__SVInt8_t a) { }
 
 __SVInt8_t sme_no_streaming_returns_vl(void) { __SVInt8_t r; return r; }
 
-// expected-warning@+2 {{passing/returning a VL-dependent argument to/from a __arm_locally_streaming function. The streaming and non-streaming vector lengths may be different}}
-// expected-cpp-warning@+1 {{passing/returning a VL-dependent argument to/from a __arm_locally_streaming function. The streaming and non-streaming vector lengths may be different}}
+// expected-warning@+2 {{passing a VL-dependent argument to a locally streaming function is undefined behaviour when the streaming and non-streaming vector lengths are different at runtime}}
+// expected-cpp-warning@+1 {{passing a VL-dependent argument to a locally streaming function is undefined behaviour when the streaming and non-streaming vector lengths are different at runtime}}
 __arm_locally_streaming void sme_locally_streaming_with_vl_arg(__SVInt8_t a) { }
 
-// expected-warning@+2 {{passing/returning a VL-dependent argument to/from a __arm_locally_streaming function. The streaming and non-streaming vector lengths may be different}}
-// expected-cpp-warning@+1 {{passing/returning a VL-dependent argument to/from a __arm_locally_streaming function. The streaming and non-streaming vector lengths may be different}}
+// expected-warning@+2 {{returning a VL-dependent argument from a locally streaming function is undefined behaviour when the streaming and non-streaming vector lengths are different at runtime}}
+// expected-cpp-warning@+1 {{returning a VL-dependent argument from a locally streaming function is undefined behaviour when the streaming and non-streaming vector lengths are different at runtime}}
 __arm_locally_streaming __SVInt8_t sme_locally_streaming_returns_vl(void) { __SVInt8_t r; return r; }
 
 void sme_no_streaming_calling_streaming_with_vl_args() {
   __SVInt8_t a;
-  // expected-warning@+2 {{passing a VL-dependent argument to/from a function that has a different streaming-mode. The streaming and non-streaming vector lengths may be different}}
-  // expected-cpp-warning@+1 {{passing a VL-dependent argument to/from a function that has a different streaming-mode. The streaming and non-streaming vector lengths may be different}}
+  // expected-warning@+2 {{passing a VL-dependent argument to a function with a different streaming-mode is undefined behaviour when the streaming and non-streaming vector lengths are different at runtime}}
+  // expected-cpp-warning@+1 {{passing a VL-dependent argument to a function with a different streaming-mode is undefined behaviour when the streaming and non-streaming vector lengths are different at runtime}}
   sme_streaming_with_vl_arg(a);
 }
 
 void sme_no_streaming_calling_streaming_with_return_vl() {
-  // expected-warning@+2 {{passing a VL-dependent argument to/from a function that has a different streaming-mode. The streaming and non-streaming vector lengths may be different}}
-  // expected-cpp-warning@+1 {{passing a VL-dependent argument to/from a function that has a different streaming-mode. The streaming and non-streaming vector lengths may be different}}
+  // expected-warning@+2 {{returning a VL-dependent argument from a function with a different streaming-mode is undefined behaviour when the streaming and non-streaming vector lengths are different at runtime}}
+  // expected-cpp-warning@+1 {{returning a VL-dependent argument from a function with a different streaming-mode is undefined behaviour when the streaming and non-streaming vector lengths are different at runtime}}
   __SVInt8_t r = sme_streaming_returns_vl();
 }
 
 void sme_streaming_calling_non_streaming_with_vl_args(void) __arm_streaming {
   __SVInt8_t a;
-  // expected-warning@+2 {{passing a VL-dependent argument to/from a function that has a different streaming-mode. The streaming and non-streaming vector lengths may be different}}
-  // expected-cpp-warning@+1 {{passing a VL-dependent argument to/from a function that has a different streaming-mode. The streaming and non-streaming vector lengths may be different}}
+  // expected-warning@+2 {{passing a VL-dependent argument to a function with a different streaming-mode is undefined behaviour when the streaming and non-streaming vector lengths are different at runtime}}
+  // expected-cpp-warning@+1 {{passing a VL-dependent argument to a function with a different streaming-mode is undefined behaviour when the streaming and non-streaming vector lengths are different at runtime}}
   sme_no_streaming_with_vl_arg(a);
 }
 
 void sme_streaming_calling_non_streaming_with_return_vl(void) __arm_streaming {
-  // expected-warning@+2 {{passing a VL-dependent argument to/from a function that has a different streaming-mode. The streaming and non-streaming vector lengths may be different}}
-  // expected-cpp-warning@+1 {{passing a VL-dependent argument to/from a function that has a different streaming-mode. The streaming and non-streaming vector lengths may be different}}
+  // expected-warning@+2 {{returning a VL-dependent argument from a function with a different streaming-mode is undefined behaviour when the streaming and non-streaming vector lengths are different at runtime}}
+  // expected-cpp-warning@+1 {{returning a VL-dependent argument from a function with a different streaming-mode is undefined behaviour when the streaming and non-streaming vector lengths are different at runtime}}
   __SVInt8_t r = sme_no_streaming_returns_vl();
 }
 
 void sme_no_streaming_calling_streaming_with_vl_args_param(__SVInt8_t arg, void (*sc)( __SVInt8_t arg) __arm_streaming) {
-  // expected-warning@+2 {{passing a VL-dependent argument to/from a function that has a different streaming-mode. The streaming and non-streaming vector lengths may be different}}
-  // expected-cpp-warning@+1 {{passing a VL-dependent argument to/from a function that has a different streaming-mode. The streaming and non-streaming vector lengths may be different}}
+  // expected-warning@+2 {{passing a VL-dependent argument to a function with a different streaming-mode is undefined behaviour when the streaming and non-streaming vector lengths are different at runtime}}
+  // expected-cpp-warning@+1 {{passing a VL-dependent argument to a function with a different streaming-mode is undefined behaviour when the streaming and non-streaming vector lengths are different at runtime}}
   sc(arg);
 }
 
 __SVInt8_t sme_no_streaming_calling_streaming_return_vl_param(__SVInt8_t (*s)(void) __arm_streaming) {
-  // expected-warning@+2 {{passing a VL-dependent argument to/from a function that has a different streaming-mode. The streaming and non-streaming vector lengths may be different}}
-  // expected-cpp-warning@+1 {{passing a VL-dependent argument to/from a function that has a different streaming-mode. The streaming and non-streaming vector lengths may be different}}
+  // expected-warning@+2 {{returning a VL-dependent argument from a function with a different streaming-mode is undefined behaviour when the streaming and non-streaming vector lengths are different at runtime}}
+  // expected-cpp-warning@+1 {{returning a VL-dependent argument from a function with a different streaming-mode is undefined behaviour when the streaming and non-streaming vector lengths are different at runtime}}
   return s();
 }
 
 void sme_streaming_compatible_calling_streaming_with_vl_args(__SVInt8_t arg) __arm_streaming_compatible {
-  // expected-warning@+2 {{passing a VL-dependent argument to/from a function that has a different streaming-mode. The streaming and non-streaming vector lengths may be different}}
-  // expected-cpp-warning@+1 {{passing a VL-dependent argument to/from a function that has a different streaming-mode. The streaming and non-streaming vector lengths may be different}}
+  // expected-warning@+2 {{passing a VL-dependent argument to a function with a different streaming-mode is undefined behaviour when the streaming and non-streaming vector lengths are different at runtime}}
+  // expected-cpp-warning@+1 {{passing a VL-dependent argument to a function with a different streaming-mode is undefined behaviour when the streaming and non-streaming vector lengths are different at runtime}}
   sme_streaming_with_vl_arg(arg);
 }
 
 void sme_streaming_compatible_calling_sme_streaming_return_vl(void) __arm_streaming_compatible {
-  // expected-warning@+2 {{passing a VL-dependent argument to/from a function that has a different streaming-mode. The streaming and non-streaming vector lengths may be different}}
-  // expected-cpp-warning@+1 {{passing a VL-dependent argument to/from a function that has a different streaming-mode. The streaming and non-streaming vector lengths may be different}}
+  // expected-warning@+2 {{returning a VL-dependent argument from a function with a different streaming-mode is undefined behaviour when the streaming and non-streaming vector lengths are different at runtime}}
+  // expected-cpp-warning@+1 {{returning a VL-dependent argument from a function with a different streaming-mode is undefined behaviour when the streaming and non-streaming vector lengths are different at runtime}}
   __SVInt8_t r = sme_streaming_returns_vl();
 }
 
 void sme_streaming_compatible_calling_no_streaming_with_vl_args(__SVInt8_t arg) __arm_streaming_compatible {
-  // expected-warning@+2 {{passing a VL-dependent argument to/from a function that has a different streaming-mode. The streaming and non-streaming vector lengths may be different}}
-  // expected-cpp-warning@+1 {{passing a VL-dependent argument to/from a function that has a different streaming-mode. The streaming and non-streaming vector lengths may be different}}
+  // expected-warning@+2 {{passing a VL-dependent argument to a function with a different streaming-mode is undefined behaviour when the streaming and non-streaming vector lengths are different at runtime}}
+  // expected-cpp-warning@+1 {{passing a VL-dependent argument to a function with a different streaming-mode is undefined behaviour when the streaming and non-streaming vector lengths are different at runtime}}
   sme_no_streaming_with_vl_arg(arg);
 }
 
 void sme_streaming_compatible_calling_no_sme_streaming_return_vl(void) __arm_streaming_compatible {
-  // expected-warning@+2 {{passing a VL-dependent argument to/from a function that has a different streaming-mode. The streaming and non-streaming vector lengths may be different}}
-  // expected-cpp-warning@+1 {{passing a VL-dependent argument to/from a function that has a different streaming-mode. The streaming and non-streaming vector lengths may be different}}
+  // expected-warning@+2 {{returning a VL-dependent argument from a function with a different streaming-mode is undefined behaviour when the streaming and non-streaming vector lengths are different at runtime}}
+  // expected-cpp-warning@+1 {{returning a VL-dependent argument from a function with a different streaming-mode is undefined behaviour when the streaming and non-streaming vector lengths are different at runtime}}
   __SVInt8_t r = sme_no_streaming_returns_vl();
 }
 
diff --git a/clang/test/Sema/constant_builtins_vector.cpp b/clang/test/Sema/constant_builtins_vector.cpp
new file mode 100644
index 000000000000..68620d436fc4
--- /dev/null
+++ b/clang/test/Sema/constant_builtins_vector.cpp
@@ -0,0 +1,723 @@
+// RUN: %clang_cc1 -verify -std=c++2a -fsyntax-only -Wno-bit-int-extension %s
+
+#if __BYTE_ORDER__ == __ORDER_LITTLE_ENDIAN__
+#define LITTLE_END 1
+#elif __BYTE_ORDER__ == __ORDER_BIG_ENDIAN__
+#define LITTLE_END 0
+#else
+#error "huh?"
+#endif
+
+// We also support _BitInt as long as it is >=8 and a power of 2.
+typedef _BitInt(8) BitInt8;
+typedef _BitInt(32) BitInt32;
+typedef _BitInt(128) BitInt128;
+
+typedef double vector4double __attribute__((__vector_size__(32)));
+typedef float vector4float __attribute__((__vector_size__(16)));
+typedef long long vector4long __attribute__((__vector_size__(32)));
+typedef int vector4int __attribute__((__vector_size__(16)));
+typedef short vector4short __attribute__((__vector_size__(8)));
+typedef char vector4char __attribute__((__vector_size__(4)));
+typedef BitInt8 vector4BitInt8 __attribute__((__vector_size__(4)));
+typedef BitInt32 vector4BitInt32 __attribute__((__vector_size__(16)));
+typedef BitInt128 vector4BitInt128 __attribute__((__vector_size__(64)));
+typedef double vector8double __attribute__((__vector_size__(64)));
+typedef float vector8float __attribute__((__vector_size__(32)));
+typedef long long vector8long __attribute__((__vector_size__(64)));
+typedef int vector8int __attribute__((__vector_size__(32)));
+typedef short vector8short __attribute__((__vector_size__(16)));
+typedef char vector8char __attribute__((__vector_size__(8)));
+typedef BitInt8 vector8BitInt8 __attribute__((__vector_size__(8)));
+typedef BitInt32 vector8BitInt32 __attribute__((__vector_size__(32)));
+typedef BitInt128 vector8BitInt128 __attribute__((__vector_size__(128)));
+
+#define CHECK_NUM(__size, __typeFrom, __typeTo, ...)                            \
+  constexpr vector##__size##__typeTo                                            \
+      from_##vector##__size##__typeFrom##_to_##vector##__size##__typeTo##_var = \
+          __builtin_convertvector((vector##__size##__typeFrom){__VA_ARGS__},    \
+                                  vector##__size##__typeTo);
+#define CHECK_TO_ALL_TYPES(__size, __typeFrom, ...)                            \
+  CHECK_NUM(__size, __typeFrom, double, __VA_ARGS__)                           \
+  CHECK_NUM(__size, __typeFrom, float, __VA_ARGS__)                            \
+  CHECK_NUM(__size, __typeFrom, long, __VA_ARGS__)                             \
+  CHECK_NUM(__size, __typeFrom, int, __VA_ARGS__)                              \
+  CHECK_NUM(__size, __typeFrom, short, __VA_ARGS__)                            \
+  CHECK_NUM(__size, __typeFrom, char, __VA_ARGS__)                             \
+  CHECK_NUM(__size, __typeFrom, BitInt8, __VA_ARGS__)                          \
+  CHECK_NUM(__size, __typeFrom, BitInt32, __VA_ARGS__)                         \
+  CHECK_NUM(__size, __typeFrom, BitInt128, __VA_ARGS__)                        \
+  static_assert(                                                               \
+      __builtin_bit_cast(                                                      \
+          unsigned,                                                            \
+          __builtin_shufflevector(                                             \
+              from_vector##__size##__typeFrom##_to_vector##__size##char_var,   \
+              from_vector##__size##__typeFrom##_to_vector##__size##char_var,   \
+              0, 1, 2, 3)) == (LITTLE_END ? 0x03020100 : 0x00010203));         \
+  static_assert(                                                               \
+      __builtin_bit_cast(                                                      \
+          unsigned long long,                                                  \
+          __builtin_shufflevector(                                             \
+              from_vector##__size##__typeFrom##_to_vector##__size##short_var,  \
+              from_vector##__size##__typeFrom##_to_vector##__size##short_var,  \
+              0, 1, 2, 3)) ==                                                  \
+      (LITTLE_END ? 0x0003000200010000 : 0x0000000100020003));
+
+#define CHECK_ALL_COMBINATIONS(__size, ...)                                    \
+  CHECK_TO_ALL_TYPES(__size, double, __VA_ARGS__)                              \
+  CHECK_TO_ALL_TYPES(__size, float, __VA_ARGS__)                               \
+  CHECK_TO_ALL_TYPES(__size, long, __VA_ARGS__)                                \
+  CHECK_TO_ALL_TYPES(__size, int, __VA_ARGS__)                                 \
+  CHECK_TO_ALL_TYPES(__size, short, __VA_ARGS__)                               \
+  CHECK_TO_ALL_TYPES(__size, char, __VA_ARGS__)                                \
+  CHECK_TO_ALL_TYPES(__size, BitInt8, __VA_ARGS__)                             \
+  CHECK_TO_ALL_TYPES(__size, BitInt32, __VA_ARGS__)                            \
+  CHECK_TO_ALL_TYPES(__size, BitInt128, __VA_ARGS__)
+
+// The result below is expanded from these macros. Use them to autogenerate the
+// test cases below.
+// CHECK_ALL_COMBINATIONS(4, 0, 1, 2, 3);
+// CHECK_ALL_COMBINATIONS(8, 0, 1, 2, 3, 4, 5, 6, 7);
+
+constexpr vector4double from_vector4double_to_vector4double_var =
+    __builtin_convertvector((vector4double){0, 1, 2, 3}, vector4double);
+constexpr vector4float from_vector4double_to_vector4float_var =
+    __builtin_convertvector((vector4double){0, 1, 2, 3}, vector4float);
+constexpr vector4long from_vector4double_to_vector4long_var =
+    __builtin_convertvector((vector4double){0, 1, 2, 3}, vector4long);
+constexpr vector4int from_vector4double_to_vector4int_var =
+    __builtin_convertvector((vector4double){0, 1, 2, 3}, vector4int);
+constexpr vector4short from_vector4double_to_vector4short_var =
+    __builtin_convertvector((vector4double){0, 1, 2, 3}, vector4short);
+constexpr vector4char from_vector4double_to_vector4char_var =
+    __builtin_convertvector((vector4double){0, 1, 2, 3}, vector4char);
+constexpr vector4BitInt8 from_vector4double_to_vector4BitInt8_var =
+    __builtin_convertvector((vector4double){0, 1, 2, 3}, vector4BitInt8);
+constexpr vector4BitInt32 from_vector4double_to_vector4BitInt32_var =
+    __builtin_convertvector((vector4double){0, 1, 2, 3}, vector4BitInt32);
+constexpr vector4BitInt128 from_vector4double_to_vector4BitInt128_var =
+    __builtin_convertvector((vector4double){0, 1, 2, 3}, vector4BitInt128);
+static_assert(__builtin_bit_cast(
+                  unsigned,
+                  __builtin_shufflevector(from_vector4double_to_vector4char_var,
+                                          from_vector4double_to_vector4char_var,
+                                          0, 1, 2, 3)) ==
+              (1 ? 0x03020100 : 0x00010203));
+static_assert(__builtin_bit_cast(unsigned long long,
+                                 __builtin_shufflevector(
+                                     from_vector4double_to_vector4short_var,
+                                     from_vector4double_to_vector4short_var, 0,
+                                     1, 2, 3)) ==
+              (1 ? 0x0003000200010000 : 0x0000000100020003));
+constexpr vector4double from_vector4float_to_vector4double_var =
+    __builtin_convertvector((vector4float){0, 1, 2, 3}, vector4double);
+constexpr vector4float from_vector4float_to_vector4float_var =
+    __builtin_convertvector((vector4float){0, 1, 2, 3}, vector4float);
+constexpr vector4long from_vector4float_to_vector4long_var =
+    __builtin_convertvector((vector4float){0, 1, 2, 3}, vector4long);
+constexpr vector4int from_vector4float_to_vector4int_var =
+    __builtin_convertvector((vector4float){0, 1, 2, 3}, vector4int);
+constexpr vector4short from_vector4float_to_vector4short_var =
+    __builtin_convertvector((vector4float){0, 1, 2, 3}, vector4short);
+constexpr vector4char from_vector4float_to_vector4char_var =
+    __builtin_convertvector((vector4float){0, 1, 2, 3}, vector4char);
+constexpr vector4BitInt8 from_vector4float_to_vector4BitInt8_var =
+    __builtin_convertvector((vector4float){0, 1, 2, 3}, vector4BitInt8);
+constexpr vector4BitInt32 from_vector4float_to_vector4BitInt32_var =
+    __builtin_convertvector((vector4float){0, 1, 2, 3}, vector4BitInt32);
+constexpr vector4BitInt128 from_vector4float_to_vector4BitInt128_var =
+    __builtin_convertvector((vector4float){0, 1, 2, 3}, vector4BitInt128);
+static_assert(__builtin_bit_cast(unsigned,
+                                 __builtin_shufflevector(
+                                     from_vector4float_to_vector4char_var,
+                                     from_vector4float_to_vector4char_var, 0, 1,
+                                     2, 3)) == (1 ? 0x03020100 : 0x00010203));
+static_assert(__builtin_bit_cast(
+                  unsigned long long,
+                  __builtin_shufflevector(from_vector4float_to_vector4short_var,
+                                          from_vector4float_to_vector4short_var,
+                                          0, 1, 2, 3)) ==
+              (1 ? 0x0003000200010000 : 0x0000000100020003));
+constexpr vector4double from_vector4long_to_vector4double_var =
+    __builtin_convertvector((vector4long){0, 1, 2, 3}, vector4double);
+constexpr vector4float from_vector4long_to_vector4float_var =
+    __builtin_convertvector((vector4long){0, 1, 2, 3}, vector4float);
+constexpr vector4long from_vector4long_to_vector4long_var =
+    __builtin_convertvector((vector4long){0, 1, 2, 3}, vector4long);
+constexpr vector4int from_vector4long_to_vector4int_var =
+    __builtin_convertvector((vector4long){0, 1, 2, 3}, vector4int);
+constexpr vector4short from_vector4long_to_vector4short_var =
+    __builtin_convertvector((vector4long){0, 1, 2, 3}, vector4short);
+constexpr vector4char from_vector4long_to_vector4char_var =
+    __builtin_convertvector((vector4long){0, 1, 2, 3}, vector4char);
+constexpr vector4BitInt8 from_vector4long_to_vector4BitInt8_var =
+    __builtin_convertvector((vector4long){0, 1, 2, 3}, vector4BitInt8);
+constexpr vector4BitInt32 from_vector4long_to_vector4BitInt32_var =
+    __builtin_convertvector((vector4long){0, 1, 2, 3}, vector4BitInt32);
+constexpr vector4BitInt128 from_vector4long_to_vector4BitInt128_var =
+    __builtin_convertvector((vector4long){0, 1, 2, 3}, vector4BitInt128);
+static_assert(__builtin_bit_cast(unsigned,
+                                 __builtin_shufflevector(
+                                     from_vector4long_to_vector4char_var,
+                                     from_vector4long_to_vector4char_var, 0, 1,
+                                     2, 3)) == (1 ? 0x03020100 : 0x00010203));
+static_assert(__builtin_bit_cast(
+                  unsigned long long,
+                  __builtin_shufflevector(from_vector4long_to_vector4short_var,
+                                          from_vector4long_to_vector4short_var,
+                                          0, 1, 2, 3)) ==
+              (1 ? 0x0003000200010000 : 0x0000000100020003));
+constexpr vector4double from_vector4int_to_vector4double_var =
+    __builtin_convertvector((vector4int){0, 1, 2, 3}, vector4double);
+constexpr vector4float from_vector4int_to_vector4float_var =
+    __builtin_convertvector((vector4int){0, 1, 2, 3}, vector4float);
+constexpr vector4long from_vector4int_to_vector4long_var =
+    __builtin_convertvector((vector4int){0, 1, 2, 3}, vector4long);
+constexpr vector4int from_vector4int_to_vector4int_var =
+    __builtin_convertvector((vector4int){0, 1, 2, 3}, vector4int);
+constexpr vector4short from_vector4int_to_vector4short_var =
+    __builtin_convertvector((vector4int){0, 1, 2, 3}, vector4short);
+constexpr vector4char from_vector4int_to_vector4char_var =
+    __builtin_convertvector((vector4int){0, 1, 2, 3}, vector4char);
+constexpr vector4BitInt8 from_vector4int_to_vector4BitInt8_var =
+    __builtin_convertvector((vector4int){0, 1, 2, 3}, vector4BitInt8);
+constexpr vector4BitInt32 from_vector4int_to_vector4BitInt32_var =
+    __builtin_convertvector((vector4int){0, 1, 2, 3}, vector4BitInt32);
+constexpr vector4BitInt128 from_vector4int_to_vector4BitInt128_var =
+    __builtin_convertvector((vector4int){0, 1, 2, 3}, vector4BitInt128);
+static_assert(__builtin_bit_cast(unsigned,
+                                 __builtin_shufflevector(
+                                     from_vector4int_to_vector4char_var,
+                                     from_vector4int_to_vector4char_var, 0, 1,
+                                     2, 3)) == (1 ? 0x03020100 : 0x00010203));
+static_assert(__builtin_bit_cast(
+                  unsigned long long,
+                  __builtin_shufflevector(from_vector4int_to_vector4short_var,
+                                          from_vector4int_to_vector4short_var,
+                                          0, 1, 2, 3)) ==
+              (1 ? 0x0003000200010000 : 0x0000000100020003));
+constexpr vector4double from_vector4short_to_vector4double_var =
+    __builtin_convertvector((vector4short){0, 1, 2, 3}, vector4double);
+constexpr vector4float from_vector4short_to_vector4float_var =
+    __builtin_convertvector((vector4short){0, 1, 2, 3}, vector4float);
+constexpr vector4long from_vector4short_to_vector4long_var =
+    __builtin_convertvector((vector4short){0, 1, 2, 3}, vector4long);
+constexpr vector4int from_vector4short_to_vector4int_var =
+    __builtin_convertvector((vector4short){0, 1, 2, 3}, vector4int);
+constexpr vector4short from_vector4short_to_vector4short_var =
+    __builtin_convertvector((vector4short){0, 1, 2, 3}, vector4short);
+constexpr vector4char from_vector4short_to_vector4char_var =
+    __builtin_convertvector((vector4short){0, 1, 2, 3}, vector4char);
+constexpr vector4BitInt8 from_vector4short_to_vector4BitInt8_var =
+    __builtin_convertvector((vector4short){0, 1, 2, 3}, vector4BitInt8);
+constexpr vector4BitInt32 from_vector4short_to_vector4BitInt32_var =
+    __builtin_convertvector((vector4short){0, 1, 2, 3}, vector4BitInt32);
+constexpr vector4BitInt128 from_vector4short_to_vector4BitInt128_var =
+    __builtin_convertvector((vector4short){0, 1, 2, 3}, vector4BitInt128);
+static_assert(__builtin_bit_cast(unsigned,
+                                 __builtin_shufflevector(
+                                     from_vector4short_to_vector4char_var,
+                                     from_vector4short_to_vector4char_var, 0, 1,
+                                     2, 3)) == (1 ? 0x03020100 : 0x00010203));
+static_assert(__builtin_bit_cast(
+                  unsigned long long,
+                  __builtin_shufflevector(from_vector4short_to_vector4short_var,
+                                          from_vector4short_to_vector4short_var,
+                                          0, 1, 2, 3)) ==
+              (1 ? 0x0003000200010000 : 0x0000000100020003));
+constexpr vector4double from_vector4char_to_vector4double_var =
+    __builtin_convertvector((vector4char){0, 1, 2, 3}, vector4double);
+constexpr vector4float from_vector4char_to_vector4float_var =
+    __builtin_convertvector((vector4char){0, 1, 2, 3}, vector4float);
+constexpr vector4long from_vector4char_to_vector4long_var =
+    __builtin_convertvector((vector4char){0, 1, 2, 3}, vector4long);
+constexpr vector4int from_vector4char_to_vector4int_var =
+    __builtin_convertvector((vector4char){0, 1, 2, 3}, vector4int);
+constexpr vector4short from_vector4char_to_vector4short_var =
+    __builtin_convertvector((vector4char){0, 1, 2, 3}, vector4short);
+constexpr vector4char from_vector4char_to_vector4char_var =
+    __builtin_convertvector((vector4char){0, 1, 2, 3}, vector4char);
+constexpr vector4BitInt8 from_vector4char_to_vector4BitInt8_var =
+    __builtin_convertvector((vector4char){0, 1, 2, 3}, vector4BitInt8);
+constexpr vector4BitInt32 from_vector4char_to_vector4BitInt32_var =
+    __builtin_convertvector((vector4char){0, 1, 2, 3}, vector4BitInt32);
+constexpr vector4BitInt128 from_vector4char_to_vector4BitInt128_var =
+    __builtin_convertvector((vector4char){0, 1, 2, 3}, vector4BitInt128);
+static_assert(__builtin_bit_cast(unsigned,
+                                 __builtin_shufflevector(
+                                     from_vector4char_to_vector4char_var,
+                                     from_vector4char_to_vector4char_var, 0, 1,
+                                     2, 3)) == (1 ? 0x03020100 : 0x00010203));
+static_assert(__builtin_bit_cast(
+                  unsigned long long,
+                  __builtin_shufflevector(from_vector4char_to_vector4short_var,
+                                          from_vector4char_to_vector4short_var,
+                                          0, 1, 2, 3)) ==
+              (1 ? 0x0003000200010000 : 0x0000000100020003));
+constexpr vector4double from_vector4BitInt8_to_vector4double_var =
+    __builtin_convertvector((vector4BitInt8){0, 1, 2, 3}, vector4double);
+constexpr vector4float from_vector4BitInt8_to_vector4float_var =
+    __builtin_convertvector((vector4BitInt8){0, 1, 2, 3}, vector4float);
+constexpr vector4long from_vector4BitInt8_to_vector4long_var =
+    __builtin_convertvector((vector4BitInt8){0, 1, 2, 3}, vector4long);
+constexpr vector4int from_vector4BitInt8_to_vector4int_var =
+    __builtin_convertvector((vector4BitInt8){0, 1, 2, 3}, vector4int);
+constexpr vector4short from_vector4BitInt8_to_vector4short_var =
+    __builtin_convertvector((vector4BitInt8){0, 1, 2, 3}, vector4short);
+constexpr vector4char from_vector4BitInt8_to_vector4char_var =
+    __builtin_convertvector((vector4BitInt8){0, 1, 2, 3}, vector4char);
+constexpr vector4BitInt8 from_vector4BitInt8_to_vector4BitInt8_var =
+    __builtin_convertvector((vector4BitInt8){0, 1, 2, 3}, vector4BitInt8);
+constexpr vector4BitInt32 from_vector4BitInt8_to_vector4BitInt32_var =
+    __builtin_convertvector((vector4BitInt8){0, 1, 2, 3}, vector4BitInt32);
+constexpr vector4BitInt128 from_vector4BitInt8_to_vector4BitInt128_var =
+    __builtin_convertvector((vector4BitInt8){0, 1, 2, 3}, vector4BitInt128);
+static_assert(__builtin_bit_cast(unsigned,
+                                 __builtin_shufflevector(
+                                     from_vector4BitInt8_to_vector4char_var,
+                                     from_vector4BitInt8_to_vector4char_var, 0,
+                                     1, 2, 3)) ==
+              (1 ? 0x03020100 : 0x00010203));
+static_assert(__builtin_bit_cast(unsigned long long,
+                                 __builtin_shufflevector(
+                                     from_vector4BitInt8_to_vector4short_var,
+                                     from_vector4BitInt8_to_vector4short_var, 0,
+                                     1, 2, 3)) ==
+              (1 ? 0x0003000200010000 : 0x0000000100020003));
+constexpr vector4double from_vector4BitInt32_to_vector4double_var =
+    __builtin_convertvector((vector4BitInt32){0, 1, 2, 3}, vector4double);
+constexpr vector4float from_vector4BitInt32_to_vector4float_var =
+    __builtin_convertvector((vector4BitInt32){0, 1, 2, 3}, vector4float);
+constexpr vector4long from_vector4BitInt32_to_vector4long_var =
+    __builtin_convertvector((vector4BitInt32){0, 1, 2, 3}, vector4long);
+constexpr vector4int from_vector4BitInt32_to_vector4int_var =
+    __builtin_convertvector((vector4BitInt32){0, 1, 2, 3}, vector4int);
+constexpr vector4short from_vector4BitInt32_to_vector4short_var =
+    __builtin_convertvector((vector4BitInt32){0, 1, 2, 3}, vector4short);
+constexpr vector4char from_vector4BitInt32_to_vector4char_var =
+    __builtin_convertvector((vector4BitInt32){0, 1, 2, 3}, vector4char);
+constexpr vector4BitInt8 from_vector4BitInt32_to_vector4BitInt8_var =
+    __builtin_convertvector((vector4BitInt32){0, 1, 2, 3}, vector4BitInt8);
+constexpr vector4BitInt32 from_vector4BitInt32_to_vector4BitInt32_var =
+    __builtin_convertvector((vector4BitInt32){0, 1, 2, 3}, vector4BitInt32);
+constexpr vector4BitInt128 from_vector4BitInt32_to_vector4BitInt128_var =
+    __builtin_convertvector((vector4BitInt32){0, 1, 2, 3}, vector4BitInt128);
+static_assert(__builtin_bit_cast(unsigned,
+                                 __builtin_shufflevector(
+                                     from_vector4BitInt32_to_vector4char_var,
+                                     from_vector4BitInt32_to_vector4char_var, 0,
+                                     1, 2, 3)) ==
+              (1 ? 0x03020100 : 0x00010203));
+static_assert(__builtin_bit_cast(unsigned long long,
+                                 __builtin_shufflevector(
+                                     from_vector4BitInt32_to_vector4short_var,
+                                     from_vector4BitInt32_to_vector4short_var,
+                                     0, 1, 2, 3)) ==
+              (1 ? 0x0003000200010000 : 0x0000000100020003));
+constexpr vector4double from_vector4BitInt128_to_vector4double_var =
+    __builtin_convertvector((vector4BitInt128){0, 1, 2, 3}, vector4double);
+constexpr vector4float from_vector4BitInt128_to_vector4float_var =
+    __builtin_convertvector((vector4BitInt128){0, 1, 2, 3}, vector4float);
+constexpr vector4long from_vector4BitInt128_to_vector4long_var =
+    __builtin_convertvector((vector4BitInt128){0, 1, 2, 3}, vector4long);
+constexpr vector4int from_vector4BitInt128_to_vector4int_var =
+    __builtin_convertvector((vector4BitInt128){0, 1, 2, 3}, vector4int);
+constexpr vector4short from_vector4BitInt128_to_vector4short_var =
+    __builtin_convertvector((vector4BitInt128){0, 1, 2, 3}, vector4short);
+constexpr vector4char from_vector4BitInt128_to_vector4char_var =
+    __builtin_convertvector((vector4BitInt128){0, 1, 2, 3}, vector4char);
+constexpr vector4BitInt8 from_vector4BitInt128_to_vector4BitInt8_var =
+    __builtin_convertvector((vector4BitInt128){0, 1, 2, 3}, vector4BitInt8);
+constexpr vector4BitInt32 from_vector4BitInt128_to_vector4BitInt32_var =
+    __builtin_convertvector((vector4BitInt128){0, 1, 2, 3}, vector4BitInt32);
+constexpr vector4BitInt128 from_vector4BitInt128_to_vector4BitInt128_var =
+    __builtin_convertvector((vector4BitInt128){0, 1, 2, 3}, vector4BitInt128);
+static_assert(__builtin_bit_cast(unsigned,
+                                 __builtin_shufflevector(
+                                     from_vector4BitInt128_to_vector4char_var,
+                                     from_vector4BitInt128_to_vector4char_var,
+                                     0, 1, 2, 3)) ==
+              (1 ? 0x03020100 : 0x00010203));
+static_assert(__builtin_bit_cast(unsigned long long,
+                                 __builtin_shufflevector(
+                                     from_vector4BitInt128_to_vector4short_var,
+                                     from_vector4BitInt128_to_vector4short_var,
+                                     0, 1, 2, 3)) ==
+              (1 ? 0x0003000200010000 : 0x0000000100020003));
+;
+constexpr vector8double from_vector8double_to_vector8double_var =
+    __builtin_convertvector((vector8double){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8double);
+constexpr vector8float from_vector8double_to_vector8float_var =
+    __builtin_convertvector((vector8double){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8float);
+constexpr vector8long from_vector8double_to_vector8long_var =
+    __builtin_convertvector((vector8double){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8long);
+constexpr vector8int from_vector8double_to_vector8int_var =
+    __builtin_convertvector((vector8double){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8int);
+constexpr vector8short from_vector8double_to_vector8short_var =
+    __builtin_convertvector((vector8double){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8short);
+constexpr vector8char from_vector8double_to_vector8char_var =
+    __builtin_convertvector((vector8double){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8char);
+constexpr vector8BitInt8 from_vector8double_to_vector8BitInt8_var =
+    __builtin_convertvector((vector8double){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8BitInt8);
+constexpr vector8BitInt32 from_vector8double_to_vector8BitInt32_var =
+    __builtin_convertvector((vector8double){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8BitInt32);
+constexpr vector8BitInt128 from_vector8double_to_vector8BitInt128_var =
+    __builtin_convertvector((vector8double){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8BitInt128);
+static_assert(__builtin_bit_cast(
+                  unsigned,
+                  __builtin_shufflevector(from_vector8double_to_vector8char_var,
+                                          from_vector8double_to_vector8char_var,
+                                          0, 1, 2, 3)) ==
+              (1 ? 0x03020100 : 0x00010203));
+static_assert(__builtin_bit_cast(unsigned long long,
+                                 __builtin_shufflevector(
+                                     from_vector8double_to_vector8short_var,
+                                     from_vector8double_to_vector8short_var, 0,
+                                     1, 2, 3)) ==
+              (1 ? 0x0003000200010000 : 0x0000000100020003));
+constexpr vector8double from_vector8float_to_vector8double_var =
+    __builtin_convertvector((vector8float){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8double);
+constexpr vector8float from_vector8float_to_vector8float_var =
+    __builtin_convertvector((vector8float){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8float);
+constexpr vector8long from_vector8float_to_vector8long_var =
+    __builtin_convertvector((vector8float){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8long);
+constexpr vector8int from_vector8float_to_vector8int_var =
+    __builtin_convertvector((vector8float){0, 1, 2, 3, 4, 5, 6, 7}, vector8int);
+constexpr vector8short from_vector8float_to_vector8short_var =
+    __builtin_convertvector((vector8float){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8short);
+constexpr vector8char from_vector8float_to_vector8char_var =
+    __builtin_convertvector((vector8float){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8char);
+constexpr vector8BitInt8 from_vector8float_to_vector8BitInt8_var =
+    __builtin_convertvector((vector8float){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8BitInt8);
+constexpr vector8BitInt32 from_vector8float_to_vector8BitInt32_var =
+    __builtin_convertvector((vector8float){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8BitInt32);
+constexpr vector8BitInt128 from_vector8float_to_vector8BitInt128_var =
+    __builtin_convertvector((vector8float){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8BitInt128);
+static_assert(__builtin_bit_cast(unsigned,
+                                 __builtin_shufflevector(
+                                     from_vector8float_to_vector8char_var,
+                                     from_vector8float_to_vector8char_var, 0, 1,
+                                     2, 3)) == (1 ? 0x03020100 : 0x00010203));
+static_assert(__builtin_bit_cast(
+                  unsigned long long,
+                  __builtin_shufflevector(from_vector8float_to_vector8short_var,
+                                          from_vector8float_to_vector8short_var,
+                                          0, 1, 2, 3)) ==
+              (1 ? 0x0003000200010000 : 0x0000000100020003));
+constexpr vector8double from_vector8long_to_vector8double_var =
+    __builtin_convertvector((vector8long){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8double);
+constexpr vector8float from_vector8long_to_vector8float_var =
+    __builtin_convertvector((vector8long){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8float);
+constexpr vector8long from_vector8long_to_vector8long_var =
+    __builtin_convertvector((vector8long){0, 1, 2, 3, 4, 5, 6, 7}, vector8long);
+constexpr vector8int from_vector8long_to_vector8int_var =
+    __builtin_convertvector((vector8long){0, 1, 2, 3, 4, 5, 6, 7}, vector8int);
+constexpr vector8short from_vector8long_to_vector8short_var =
+    __builtin_convertvector((vector8long){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8short);
+constexpr vector8char from_vector8long_to_vector8char_var =
+    __builtin_convertvector((vector8long){0, 1, 2, 3, 4, 5, 6, 7}, vector8char);
+constexpr vector8BitInt8 from_vector8long_to_vector8BitInt8_var =
+    __builtin_convertvector((vector8long){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8BitInt8);
+constexpr vector8BitInt32 from_vector8long_to_vector8BitInt32_var =
+    __builtin_convertvector((vector8long){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8BitInt32);
+constexpr vector8BitInt128 from_vector8long_to_vector8BitInt128_var =
+    __builtin_convertvector((vector8long){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8BitInt128);
+static_assert(__builtin_bit_cast(unsigned,
+                                 __builtin_shufflevector(
+                                     from_vector8long_to_vector8char_var,
+                                     from_vector8long_to_vector8char_var, 0, 1,
+                                     2, 3)) == (1 ? 0x03020100 : 0x00010203));
+static_assert(__builtin_bit_cast(
+                  unsigned long long,
+                  __builtin_shufflevector(from_vector8long_to_vector8short_var,
+                                          from_vector8long_to_vector8short_var,
+                                          0, 1, 2, 3)) ==
+              (1 ? 0x0003000200010000 : 0x0000000100020003));
+constexpr vector8double from_vector8int_to_vector8double_var =
+    __builtin_convertvector((vector8int){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8double);
+constexpr vector8float from_vector8int_to_vector8float_var =
+    __builtin_convertvector((vector8int){0, 1, 2, 3, 4, 5, 6, 7}, vector8float);
+constexpr vector8long from_vector8int_to_vector8long_var =
+    __builtin_convertvector((vector8int){0, 1, 2, 3, 4, 5, 6, 7}, vector8long);
+constexpr vector8int from_vector8int_to_vector8int_var =
+    __builtin_convertvector((vector8int){0, 1, 2, 3, 4, 5, 6, 7}, vector8int);
+constexpr vector8short from_vector8int_to_vector8short_var =
+    __builtin_convertvector((vector8int){0, 1, 2, 3, 4, 5, 6, 7}, vector8short);
+constexpr vector8char from_vector8int_to_vector8char_var =
+    __builtin_convertvector((vector8int){0, 1, 2, 3, 4, 5, 6, 7}, vector8char);
+constexpr vector8BitInt8 from_vector8int_to_vector8BitInt8_var =
+    __builtin_convertvector((vector8int){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8BitInt8);
+constexpr vector8BitInt32 from_vector8int_to_vector8BitInt32_var =
+    __builtin_convertvector((vector8int){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8BitInt32);
+constexpr vector8BitInt128 from_vector8int_to_vector8BitInt128_var =
+    __builtin_convertvector((vector8int){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8BitInt128);
+static_assert(__builtin_bit_cast(unsigned,
+                                 __builtin_shufflevector(
+                                     from_vector8int_to_vector8char_var,
+                                     from_vector8int_to_vector8char_var, 0, 1,
+                                     2, 3)) == (1 ? 0x03020100 : 0x00010203));
+static_assert(__builtin_bit_cast(
+                  unsigned long long,
+                  __builtin_shufflevector(from_vector8int_to_vector8short_var,
+                                          from_vector8int_to_vector8short_var,
+                                          0, 1, 2, 3)) ==
+              (1 ? 0x0003000200010000 : 0x0000000100020003));
+constexpr vector8double from_vector8short_to_vector8double_var =
+    __builtin_convertvector((vector8short){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8double);
+constexpr vector8float from_vector8short_to_vector8float_var =
+    __builtin_convertvector((vector8short){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8float);
+constexpr vector8long from_vector8short_to_vector8long_var =
+    __builtin_convertvector((vector8short){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8long);
+constexpr vector8int from_vector8short_to_vector8int_var =
+    __builtin_convertvector((vector8short){0, 1, 2, 3, 4, 5, 6, 7}, vector8int);
+constexpr vector8short from_vector8short_to_vector8short_var =
+    __builtin_convertvector((vector8short){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8short);
+constexpr vector8char from_vector8short_to_vector8char_var =
+    __builtin_convertvector((vector8short){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8char);
+constexpr vector8BitInt8 from_vector8short_to_vector8BitInt8_var =
+    __builtin_convertvector((vector8short){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8BitInt8);
+constexpr vector8BitInt32 from_vector8short_to_vector8BitInt32_var =
+    __builtin_convertvector((vector8short){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8BitInt32);
+constexpr vector8BitInt128 from_vector8short_to_vector8BitInt128_var =
+    __builtin_convertvector((vector8short){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8BitInt128);
+static_assert(__builtin_bit_cast(unsigned,
+                                 __builtin_shufflevector(
+                                     from_vector8short_to_vector8char_var,
+                                     from_vector8short_to_vector8char_var, 0, 1,
+                                     2, 3)) == (1 ? 0x03020100 : 0x00010203));
+static_assert(__builtin_bit_cast(
+                  unsigned long long,
+                  __builtin_shufflevector(from_vector8short_to_vector8short_var,
+                                          from_vector8short_to_vector8short_var,
+                                          0, 1, 2, 3)) ==
+              (1 ? 0x0003000200010000 : 0x0000000100020003));
+constexpr vector8double from_vector8char_to_vector8double_var =
+    __builtin_convertvector((vector8char){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8double);
+constexpr vector8float from_vector8char_to_vector8float_var =
+    __builtin_convertvector((vector8char){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8float);
+constexpr vector8long from_vector8char_to_vector8long_var =
+    __builtin_convertvector((vector8char){0, 1, 2, 3, 4, 5, 6, 7}, vector8long);
+constexpr vector8int from_vector8char_to_vector8int_var =
+    __builtin_convertvector((vector8char){0, 1, 2, 3, 4, 5, 6, 7}, vector8int);
+constexpr vector8short from_vector8char_to_vector8short_var =
+    __builtin_convertvector((vector8char){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8short);
+constexpr vector8char from_vector8char_to_vector8char_var =
+    __builtin_convertvector((vector8char){0, 1, 2, 3, 4, 5, 6, 7}, vector8char);
+constexpr vector8BitInt8 from_vector8char_to_vector8BitInt8_var =
+    __builtin_convertvector((vector8char){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8BitInt8);
+constexpr vector8BitInt32 from_vector8char_to_vector8BitInt32_var =
+    __builtin_convertvector((vector8char){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8BitInt32);
+constexpr vector8BitInt128 from_vector8char_to_vector8BitInt128_var =
+    __builtin_convertvector((vector8char){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8BitInt128);
+static_assert(__builtin_bit_cast(unsigned,
+                                 __builtin_shufflevector(
+                                     from_vector8char_to_vector8char_var,
+                                     from_vector8char_to_vector8char_var, 0, 1,
+                                     2, 3)) == (1 ? 0x03020100 : 0x00010203));
+static_assert(__builtin_bit_cast(
+                  unsigned long long,
+                  __builtin_shufflevector(from_vector8char_to_vector8short_var,
+                                          from_vector8char_to_vector8short_var,
+                                          0, 1, 2, 3)) ==
+              (1 ? 0x0003000200010000 : 0x0000000100020003));
+constexpr vector8double from_vector8BitInt8_to_vector8double_var =
+    __builtin_convertvector((vector8BitInt8){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8double);
+constexpr vector8float from_vector8BitInt8_to_vector8float_var =
+    __builtin_convertvector((vector8BitInt8){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8float);
+constexpr vector8long from_vector8BitInt8_to_vector8long_var =
+    __builtin_convertvector((vector8BitInt8){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8long);
+constexpr vector8int from_vector8BitInt8_to_vector8int_var =
+    __builtin_convertvector((vector8BitInt8){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8int);
+constexpr vector8short from_vector8BitInt8_to_vector8short_var =
+    __builtin_convertvector((vector8BitInt8){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8short);
+constexpr vector8char from_vector8BitInt8_to_vector8char_var =
+    __builtin_convertvector((vector8BitInt8){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8char);
+constexpr vector8BitInt8 from_vector8BitInt8_to_vector8BitInt8_var =
+    __builtin_convertvector((vector8BitInt8){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8BitInt8);
+constexpr vector8BitInt32 from_vector8BitInt8_to_vector8BitInt32_var =
+    __builtin_convertvector((vector8BitInt8){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8BitInt32);
+constexpr vector8BitInt128 from_vector8BitInt8_to_vector8BitInt128_var =
+    __builtin_convertvector((vector8BitInt8){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8BitInt128);
+static_assert(__builtin_bit_cast(unsigned,
+                                 __builtin_shufflevector(
+                                     from_vector8BitInt8_to_vector8char_var,
+                                     from_vector8BitInt8_to_vector8char_var, 0,
+                                     1, 2, 3)) ==
+              (1 ? 0x03020100 : 0x00010203));
+static_assert(__builtin_bit_cast(unsigned long long,
+                                 __builtin_shufflevector(
+                                     from_vector8BitInt8_to_vector8short_var,
+                                     from_vector8BitInt8_to_vector8short_var, 0,
+                                     1, 2, 3)) ==
+              (1 ? 0x0003000200010000 : 0x0000000100020003));
+constexpr vector8double from_vector8BitInt32_to_vector8double_var =
+    __builtin_convertvector((vector8BitInt32){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8double);
+constexpr vector8float from_vector8BitInt32_to_vector8float_var =
+    __builtin_convertvector((vector8BitInt32){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8float);
+constexpr vector8long from_vector8BitInt32_to_vector8long_var =
+    __builtin_convertvector((vector8BitInt32){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8long);
+constexpr vector8int from_vector8BitInt32_to_vector8int_var =
+    __builtin_convertvector((vector8BitInt32){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8int);
+constexpr vector8short from_vector8BitInt32_to_vector8short_var =
+    __builtin_convertvector((vector8BitInt32){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8short);
+constexpr vector8char from_vector8BitInt32_to_vector8char_var =
+    __builtin_convertvector((vector8BitInt32){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8char);
+constexpr vector8BitInt8 from_vector8BitInt32_to_vector8BitInt8_var =
+    __builtin_convertvector((vector8BitInt32){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8BitInt8);
+constexpr vector8BitInt32 from_vector8BitInt32_to_vector8BitInt32_var =
+    __builtin_convertvector((vector8BitInt32){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8BitInt32);
+constexpr vector8BitInt128 from_vector8BitInt32_to_vector8BitInt128_var =
+    __builtin_convertvector((vector8BitInt32){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8BitInt128);
+static_assert(__builtin_bit_cast(unsigned,
+                                 __builtin_shufflevector(
+                                     from_vector8BitInt32_to_vector8char_var,
+                                     from_vector8BitInt32_to_vector8char_var, 0,
+                                     1, 2, 3)) ==
+              (1 ? 0x03020100 : 0x00010203));
+static_assert(__builtin_bit_cast(unsigned long long,
+                                 __builtin_shufflevector(
+                                     from_vector8BitInt32_to_vector8short_var,
+                                     from_vector8BitInt32_to_vector8short_var,
+                                     0, 1, 2, 3)) ==
+              (1 ? 0x0003000200010000 : 0x0000000100020003));
+constexpr vector8double from_vector8BitInt128_to_vector8double_var =
+    __builtin_convertvector((vector8BitInt128){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8double);
+constexpr vector8float from_vector8BitInt128_to_vector8float_var =
+    __builtin_convertvector((vector8BitInt128){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8float);
+constexpr vector8long from_vector8BitInt128_to_vector8long_var =
+    __builtin_convertvector((vector8BitInt128){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8long);
+constexpr vector8int from_vector8BitInt128_to_vector8int_var =
+    __builtin_convertvector((vector8BitInt128){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8int);
+constexpr vector8short from_vector8BitInt128_to_vector8short_var =
+    __builtin_convertvector((vector8BitInt128){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8short);
+constexpr vector8char from_vector8BitInt128_to_vector8char_var =
+    __builtin_convertvector((vector8BitInt128){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8char);
+constexpr vector8BitInt8 from_vector8BitInt128_to_vector8BitInt8_var =
+    __builtin_convertvector((vector8BitInt128){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8BitInt8);
+constexpr vector8BitInt32 from_vector8BitInt128_to_vector8BitInt32_var =
+    __builtin_convertvector((vector8BitInt128){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8BitInt32);
+constexpr vector8BitInt128 from_vector8BitInt128_to_vector8BitInt128_var =
+    __builtin_convertvector((vector8BitInt128){0, 1, 2, 3, 4, 5, 6, 7},
+                            vector8BitInt128);
+static_assert(__builtin_bit_cast(unsigned,
+                                 __builtin_shufflevector(
+                                     from_vector8BitInt128_to_vector8char_var,
+                                     from_vector8BitInt128_to_vector8char_var,
+                                     0, 1, 2, 3)) ==
+              (1 ? 0x03020100 : 0x00010203));
+static_assert(__builtin_bit_cast(unsigned long long,
+                                 __builtin_shufflevector(
+                                     from_vector8BitInt128_to_vector8short_var,
+                                     from_vector8BitInt128_to_vector8short_var,
+                                     0, 1, 2, 3)) ==
+              (1 ? 0x0003000200010000 : 0x0000000100020003));
+;
+#undef CHECK_ALL_COMBINATIONS
+#undef CHECK_TO_ALL_TYPES
+#undef CHECK_NUM
+
+// Shuffle vector
+constexpr vector4char vector4charConst1 = {0, 1, 2, 3};
+constexpr vector4char vector4charConst2 = {4, 5, 6, 7};
+constexpr vector8char vector8intConst = {8, 9, 10, 11, 12, 13, 14, 15};
+
+constexpr vector4char vectorShuffle1 =
+    __builtin_shufflevector(vector4charConst1, vector4charConst2, 0, 1, 2, 3);
+static_assert(__builtin_bit_cast(unsigned, vectorShuffle1) ==
+              (LITTLE_END ? 0x03020100 : 0x00010203));
+constexpr vector4char vectorShuffle2 =
+    __builtin_shufflevector(vector4charConst1, vector4charConst2, 4, 5, 6, 7);
+static_assert(__builtin_bit_cast(unsigned, vectorShuffle2) ==
+              (LITTLE_END ? 0x07060504 : 0x04050607));
+constexpr vector4char vectorShuffle3 =
+    __builtin_shufflevector(vector4charConst1, vector4charConst2, 0, 2, 4, 6);
+static_assert(__builtin_bit_cast(unsigned, vectorShuffle3) ==
+              (LITTLE_END ? 0x06040200 : 0x00020406));
+constexpr vector8char vectorShuffle4 = __builtin_shufflevector(
+    vector8intConst, vector8intConst, 0, 2, 4, 6, 8, 10, 12, 14);
+static_assert(__builtin_bit_cast(unsigned long long, vectorShuffle4) ==
+              (LITTLE_END ? 0x0E0C0A080E0C0A08 : 0x080A0C0E080A0C0E));
+constexpr vector4char vectorShuffle5 =
+    __builtin_shufflevector(vector8intConst, vector8intConst, 0, 2, 4, 6);
+static_assert(__builtin_bit_cast(unsigned, vectorShuffle5) ==
+              (LITTLE_END ? 0x0E0C0A08 : 0x080A0C0E));
+constexpr vector8char vectorShuffle6 = __builtin_shufflevector(
+    vector4charConst1, vector4charConst2, 0, 2, 4, 6, 1, 3, 5, 7);
+static_assert(__builtin_bit_cast(unsigned long long, vectorShuffle6) ==
+              (LITTLE_END ? 0x0705030106040200 : 0x0002040601030507));
+
+constexpr vector4char
+    vectorShuffleFail1 = // expected-error {{constexpr variable 'vectorShuffleFail1'\
+ must be initialized by a constant expression}}
+    __builtin_shufflevector( // expected-error {{index for __builtin_shufflevector \
+not within the bounds of the input vectors; index of -1 found at position 0 not \
+permitted in a constexpr context.}}
+        vector4charConst1,
+        vector4charConst2, -1, -1, -1, -1);
diff --git a/clang/test/Sema/convertvector.c b/clang/test/Sema/convertvector.c
index 8ae43c3ba3d4..1ff04af90981 100644
--- a/clang/test/Sema/convertvector.c
+++ b/clang/test/Sema/convertvector.c
@@ -15,3 +15,6 @@ vector8float foo3(double x) {
   return __builtin_convertvector(x, vector8float);  // expected-error {{must be a vector}}
 }
 
+float foo4(float x) {
+  return __builtin_convertvector(x, float); // expected-error {{first argument to __builtin_convertvector must be a vector}}
+}
diff --git a/clang/test/SemaCXX/cxx1z-class-template-argument-deduction.cpp b/clang/test/SemaCXX/cxx1z-class-template-argument-deduction.cpp
index 2f067ea53a50..90404f115c75 100644
--- a/clang/test/SemaCXX/cxx1z-class-template-argument-deduction.cpp
+++ b/clang/test/SemaCXX/cxx1z-class-template-argument-deduction.cpp
@@ -12,14 +12,19 @@ namespace std {
     size_t n;
     initializer_list();
   };
-  // FIXME: This should probably not be necessary.
-  template<typename T> initializer_list(initializer_list<T>) -> initializer_list<T>;
 }
 
 template<typename T> constexpr bool has_type(...) { return false; }
 template<typename T> constexpr bool has_type(T&) { return true; }
 
-std::initializer_list il = {1, 2, 3, 4, 5};
+std::initializer_list il1 = {1, 2, 3, 4, 5};
+auto il2 = std::initializer_list{1, 2, 3, 4};
+auto il3 = std::initializer_list{il1};
+auto il4 = std::initializer_list{il1, il1, il1};
+static_assert(has_type<std::initializer_list<int>>(il1));
+static_assert(has_type<std::initializer_list<int>>(il2));
+static_assert(has_type<std::initializer_list<int>>(il3));
+static_assert(has_type<std::initializer_list<std::initializer_list<int>>>(il4));
 
 template<typename T> struct vector {
   template<typename Iter> vector(Iter, Iter);
diff --git a/clang/test/SemaCXX/cxx2a-consteval.cpp b/clang/test/SemaCXX/cxx2a-consteval.cpp
index 192621225a54..e19807437207 100644
--- a/clang/test/SemaCXX/cxx2a-consteval.cpp
+++ b/clang/test/SemaCXX/cxx2a-consteval.cpp
@@ -260,6 +260,26 @@ int(*test)(int)  = l1;
 
 }
 
+namespace consteval_lambda_in_template {
+struct S {
+    int *value;
+    constexpr S(int v) : value(new int {v}) {}
+    constexpr ~S() { delete value; }
+};
+consteval S fn() { return S(5); }
+
+template <typename T>
+void fn2() {
+    (void)[]() consteval -> int {
+      return *(fn().value);  // OK, immediate context
+    };
+}
+
+void caller() {
+    fn2<int>();
+}
+}
+
 namespace std {
 
 template <typename T> struct remove_reference { using type = T; };
diff --git a/clang/test/SemaCXX/template-specialization.cpp b/clang/test/SemaCXX/template-specialization.cpp
index 7b26ff9f5c5b..eabb84f2e13d 100644
--- a/clang/test/SemaCXX/template-specialization.cpp
+++ b/clang/test/SemaCXX/template-specialization.cpp
@@ -52,3 +52,31 @@ void instantiate() {
 }
 
 }
+
+namespace GH89374 {
+
+struct A {};
+
+template <typename Derived>
+struct MatrixBase { // #GH89374-MatrixBase
+  template <typename OtherDerived>
+  Derived &operator=(const MatrixBase<OtherDerived> &); // #GH89374-copy-assignment
+};
+
+template <typename>
+struct solve_retval;
+
+template <typename Rhs>
+struct solve_retval<int> : MatrixBase<solve_retval<Rhs> > {};
+// expected-error@-1 {{partial specialization of 'solve_retval' does not use any of its template parameters}}
+
+void ApproximateChebyshev() {
+  MatrixBase<int> c;
+  c = solve_retval<int>();
+  // expected-error@-1 {{no viable overloaded '='}}
+  //   expected-note@#GH89374-copy-assignment {{candidate template ignored: could not match 'MatrixBase' against 'solve_retval'}}
+  //   expected-note@#GH89374-MatrixBase {{candidate function (the implicit copy assignment operator) not viable: no known conversion from 'solve_retval<int>' to 'const MatrixBase<int>' for 1st argument}}
+  //   expected-note@#GH89374-MatrixBase {{candidate function (the implicit move assignment operator) not viable: no known conversion from 'solve_retval<int>' to 'MatrixBase<int>' for 1st argument}}
+}
+
+} // namespace GH89374
diff --git a/clang/test/SemaCXX/type-traits.cpp b/clang/test/SemaCXX/type-traits.cpp
index dee4a29bd2bf..01991887b284 100644
--- a/clang/test/SemaCXX/type-traits.cpp
+++ b/clang/test/SemaCXX/type-traits.cpp
@@ -2509,6 +2509,20 @@ void is_convertible()
   static_assert(__is_convertible(FloatWrapper, IntWrapper));
   static_assert(__is_convertible(FloatWrapper, float));
   static_assert(__is_convertible(float, FloatWrapper));
+  static_assert(__is_convertible(IntWrapper, IntWrapper&&));
+  static_assert(__is_convertible(IntWrapper, const IntWrapper&));
+  static_assert(__is_convertible(IntWrapper, int&&));
+  static_assert(__is_convertible(IntWrapper, const int&));
+  static_assert(__is_convertible(int, IntWrapper&&));
+  static_assert(__is_convertible(int, const IntWrapper&));
+  static_assert(__is_convertible(IntWrapper, FloatWrapper&&));
+  static_assert(__is_convertible(IntWrapper, const FloatWrapper&));
+  static_assert(__is_convertible(FloatWrapper, IntWrapper&&));
+  static_assert(__is_convertible(FloatWrapper, const IntWrapper&&));
+  static_assert(__is_convertible(FloatWrapper, float&&));
+  static_assert(__is_convertible(FloatWrapper, const float&));
+  static_assert(__is_convertible(float, FloatWrapper&&));
+  static_assert(__is_convertible(float, const FloatWrapper&));
 }
 
 void is_nothrow_convertible()
@@ -2521,6 +2535,20 @@ void is_nothrow_convertible()
   static_assert(!__is_nothrow_convertible(FloatWrapper, IntWrapper));
   static_assert(!__is_nothrow_convertible(FloatWrapper, float));
   static_assert(__is_nothrow_convertible(float, FloatWrapper));
+  static_assert(__is_nothrow_convertible(IntWrapper, IntWrapper&&));
+  static_assert(__is_nothrow_convertible(IntWrapper, const IntWrapper&));
+  static_assert(__is_nothrow_convertible(IntWrapper, int&&));
+  static_assert(__is_nothrow_convertible(IntWrapper, const int&));
+  static_assert(!__is_nothrow_convertible(int, IntWrapper&&));
+  static_assert(!__is_nothrow_convertible(int, const IntWrapper&));
+  static_assert(!__is_nothrow_convertible(IntWrapper, FloatWrapper&&));
+  static_assert(!__is_nothrow_convertible(IntWrapper, const FloatWrapper&));
+  static_assert(!__is_nothrow_convertible(FloatWrapper, IntWrapper&&));
+  static_assert(!__is_nothrow_convertible(FloatWrapper, const IntWrapper&));
+  static_assert(!__is_nothrow_convertible(FloatWrapper, float&&));
+  static_assert(!__is_nothrow_convertible(FloatWrapper, const float&));
+  static_assert(__is_nothrow_convertible(float, FloatWrapper&&));
+  static_assert(__is_nothrow_convertible(float, const FloatWrapper&));
 }
 
 struct FromInt { FromInt(int); };
diff --git a/clang/test/SemaCXX/unused.cpp b/clang/test/SemaCXX/unused.cpp
index 0af9e5b68b00..1f40c1b1ca90 100644
--- a/clang/test/SemaCXX/unused.cpp
+++ b/clang/test/SemaCXX/unused.cpp
@@ -102,11 +102,21 @@ namespace PR33839 {
     for (auto [x] : a) { // expected-warning {{unused variable '[x]'}}
     }
   }
-  void use() { 
+  void use() {
     f<int>(); // expected-note {{instantiation of}}
     g<true>();
     g<false>();
     h<int>(); // expected-note {{instantiation of}}
   }
 }
+
+namespace maybe_unused_binding {
+
+void test() {
+  struct X { int a, b; } x;
+  auto [a [[maybe_unused]], b] = x; // expected-warning {{an attribute specifier sequence attached to a structured binding declaration is a C++2c extension}}
+}
+
+}
+
 #endif
diff --git a/clang/test/SemaObjC/format-strings-oslog.m b/clang/test/SemaObjC/format-strings-oslog.m
index 20fec93b653b..af5aef3d6179 100644
--- a/clang/test/SemaObjC/format-strings-oslog.m
+++ b/clang/test/SemaObjC/format-strings-oslog.m
@@ -44,15 +44,18 @@ void test_os_log_format(const char *pc, int i, void *p, void *buf) {
 }
 
 // Test os_log_format primitive with ObjC string literal format argument.
-void test_objc(const char *pc, int i, void *p, void *buf, NSString *nss) {
+void test_objc(const char *pc, int i, void *p, void *buf, NSString *nss, id obj) {
   __builtin_os_log_format(buf, @"");
   __builtin_os_log_format(buf, @"%d"); // expected-warning {{more '%' conversions than data arguments}}
   __builtin_os_log_format(buf, @"%d", i);
+
   __builtin_os_log_format(buf, @"%P", p); // expected-warning {{using '%P' format specifier without precision}}
   __builtin_os_log_format(buf, @"%.10P", p);
   __builtin_os_log_format(buf, @"%.*P", p); // expected-warning {{field precision should have type 'int', but argument has type 'void *'}}
   __builtin_os_log_format(buf, @"%.*P", i, p);
   __builtin_os_log_format(buf, @"%.*P", i, i); // expected-warning {{format specifies type 'void *' but the argument has type 'int'}}
+  __builtin_os_log_format(buf, @"%.8P", nss);    // expected-warning {{using '%P' format specifier with an Objective-C pointer results in dumping runtime object structure, not object value}}
+  __builtin_os_log_format(buf, @"%.*P", i, obj); // expected-warning {{using '%P' format specifier with an Objective-C pointer results in dumping runtime object structure, not object value}}
 
   __builtin_os_log_format(buf, @"%{private}s", pc);
   __builtin_os_log_format(buf, @"%@", nss);
diff --git a/clang/test/SemaOpenCL/vec_step.cl b/clang/test/SemaOpenCL/vec_step.cl
index afb6dc94d92e..c116f09b351f 100644
--- a/clang/test/SemaOpenCL/vec_step.cl
+++ b/clang/test/SemaOpenCL/vec_step.cl
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -fsyntax-only -pedantic -verify %s
+// RUN: %clang_cc1 -fsyntax-only -pedantic -verify %s -fexperimental-new-constant-interpreter
 
 typedef int int2 __attribute__((ext_vector_type(2)));
 typedef int int3 __attribute__((ext_vector_type(3)));
diff --git a/clang/test/TestRunner.sh b/clang/test/TestRunner.sh
deleted file mode 100755
index f96d3d552d2e..000000000000
--- a/clang/test/TestRunner.sh
+++ /dev/null
@@ -1,13 +0,0 @@
-#!/bin/sh
-#
-# TestRunner.sh - Backward compatible utility for testing an individual file.
-
-# Find where this script is.
-Dir=$(dirname $(which $0))
-AbsDir=$(cd $Dir; pwd)
-
-# Find 'lit', assuming standard layout.
-lit=$AbsDir/../../../utils/lit/lit.py
-
-# Dispatch to lit.
-$lit "$@"
diff --git a/clang/tools/c-index-test/c-index-test.c b/clang/tools/c-index-test/c-index-test.c
index 21619888cfa5..e078e9bdce02 100644
--- a/clang/tools/c-index-test/c-index-test.c
+++ b/clang/tools/c-index-test/c-index-test.c
@@ -464,10 +464,10 @@ static void PrintRange(CXSourceRange R, const char *str) {
   CXFile begin_file, end_file;
   unsigned begin_line, begin_column, end_line, end_column;
 
-  clang_getSpellingLocation(clang_getRangeStart(R),
-                            &begin_file, &begin_line, &begin_column, 0);
-  clang_getSpellingLocation(clang_getRangeEnd(R),
-                            &end_file, &end_line, &end_column, 0);
+  clang_getFileLocation(clang_getRangeStart(R), &begin_file, &begin_line,
+                        &begin_column, 0);
+  clang_getFileLocation(clang_getRangeEnd(R), &end_file, &end_line, &end_column,
+                        0);
   if (!begin_file || !end_file)
     return;
 
@@ -849,13 +849,13 @@ static void PrintCursor(CXCursor Cursor, const char *CommentSchemaFile) {
             printf(", ");
           
           Loc = clang_getCursorLocation(Ovl);
-          clang_getSpellingLocation(Loc, 0, &line, &column, 0);
+          clang_getFileLocation(Loc, 0, &line, &column, 0);
           printf("%d:%d", line, column);          
         }
         printf("]");
       } else {
         CXSourceLocation Loc = clang_getCursorLocation(Referenced);
-        clang_getSpellingLocation(Loc, 0, &line, &column, 0);
+        clang_getFileLocation(Loc, 0, &line, &column, 0);
         printf(":%d:%d", line, column);
       }
 
@@ -1047,7 +1047,7 @@ static void PrintCursor(CXCursor Cursor, const char *CommentSchemaFile) {
     if (!clang_equalCursors(SpecializationOf, clang_getNullCursor())) {
       CXSourceLocation Loc = clang_getCursorLocation(SpecializationOf);
       CXString Name = clang_getCursorSpelling(SpecializationOf);
-      clang_getSpellingLocation(Loc, 0, &line, &column, 0);
+      clang_getFileLocation(Loc, 0, &line, &column, 0);
       printf(" [Specialization of %s:%d:%d]",
              clang_getCString(Name), line, column);
       clang_disposeString(Name);
@@ -1094,7 +1094,7 @@ static void PrintCursor(CXCursor Cursor, const char *CommentSchemaFile) {
       printf(" [Overrides ");
       for (I = 0; I != num_overridden; ++I) {
         CXSourceLocation Loc = clang_getCursorLocation(overridden[I]);
-        clang_getSpellingLocation(Loc, 0, &line, &column, 0);
+        clang_getFileLocation(Loc, 0, &line, &column, 0);
         lineCols[I].line = line;
         lineCols[I].col = column;
       }
@@ -1257,8 +1257,8 @@ void PrintDiagnostic(CXDiagnostic Diagnostic) {
   fprintf(stderr, "%s\n", clang_getCString(Msg));
   clang_disposeString(Msg);
 
-  clang_getSpellingLocation(clang_getDiagnosticLocation(Diagnostic),
-                            &file, 0, 0, 0);
+  clang_getFileLocation(clang_getDiagnosticLocation(Diagnostic), &file, 0, 0,
+                        0);
   if (!file)
     return;
 
@@ -1271,9 +1271,8 @@ void PrintDiagnostic(CXDiagnostic Diagnostic) {
     CXSourceLocation end = clang_getRangeEnd(range);
     unsigned start_line, start_column, end_line, end_column;
     CXFile start_file, end_file;
-    clang_getSpellingLocation(start, &start_file, &start_line,
-                              &start_column, 0);
-    clang_getSpellingLocation(end, &end_file, &end_line, &end_column, 0);
+    clang_getFileLocation(start, &start_file, &start_line, &start_column, 0);
+    clang_getFileLocation(end, &end_file, &end_line, &end_column, 0);
     if (clang_equalLocations(start, end)) {
       /* Insertion. */
       if (start_file == file)
@@ -1356,7 +1355,7 @@ enum CXChildVisitResult FilteredPrintingVisitor(CXCursor Cursor,
   if (!Data->Filter || (Cursor.kind == *(enum CXCursorKind *)Data->Filter)) {
     CXSourceLocation Loc = clang_getCursorLocation(Cursor);
     unsigned line, column;
-    clang_getSpellingLocation(Loc, 0, &line, &column, 0);
+    clang_getFileLocation(Loc, 0, &line, &column, 0);
     printf("// %s: %s:%d:%d: ", FileCheckPrefix,
            GetCursorSource(Cursor), line, column);
     PrintCursor(Cursor, Data->CommentSchemaFile);
@@ -1417,7 +1416,7 @@ static enum CXChildVisitResult FunctionScanVisitor(CXCursor Cursor,
       curColumn++;
 
     Loc = clang_getCursorLocation(Cursor);
-    clang_getSpellingLocation(Loc, &file, 0, 0, 0);
+    clang_getFileLocation(Loc, &file, 0, 0, 0);
 
     source = clang_getFileName(file);
     if (clang_getCString(source)) {
@@ -1483,8 +1482,7 @@ void InclusionVisitor(CXFile includedFile, CXSourceLocation *includeStack,
   for (i = 0; i < includeStackLen; ++i) {
     CXFile includingFile;
     unsigned line, column;
-    clang_getSpellingLocation(includeStack[i], &includingFile, &line,
-                              &column, 0);
+    clang_getFileLocation(includeStack[i], &includingFile, &line, &column, 0);
     fname = clang_getFileName(includingFile);
     printf("  %s:%d:%d\n", clang_getCString(fname), line, column);
     clang_disposeString(fname);
@@ -2984,7 +2982,7 @@ static void inspect_print_cursor(CXCursor Cursor) {
   CXString Spelling;
   const char *cspell;
   unsigned line, column;
-  clang_getSpellingLocation(CursorLoc, 0, &line, &column, 0);
+  clang_getFileLocation(CursorLoc, 0, &line, &column, 0);
   printf("%d:%d ", line, column);
   PrintCursor(Cursor, NULL);
   PrintCursorExtent(Cursor);
@@ -3100,7 +3098,7 @@ static void inspect_evaluate_cursor(CXCursor Cursor) {
   unsigned line, column;
   CXEvalResult ER;
 
-  clang_getSpellingLocation(CursorLoc, 0, &line, &column, 0);
+  clang_getFileLocation(CursorLoc, 0, &line, &column, 0);
   printf("%d:%d ", line, column);
   PrintCursor(Cursor, NULL);
   PrintCursorExtent(Cursor);
@@ -3135,7 +3133,7 @@ static void inspect_macroinfo_cursor(CXCursor Cursor) {
   CXString Spelling;
   const char *cspell;
   unsigned line, column;
-  clang_getSpellingLocation(CursorLoc, 0, &line, &column, 0);
+  clang_getFileLocation(CursorLoc, 0, &line, &column, 0);
   printf("%d:%d ", line, column);
   PrintCursor(Cursor, NULL);
   PrintCursorExtent(Cursor);
@@ -4328,10 +4326,10 @@ int perform_token_annotation(int argc, const char **argv) {
   skipped_ranges = clang_getSkippedRanges(TU, file);
   for (i = 0; i != skipped_ranges->count; ++i) {
     unsigned start_line, start_column, end_line, end_column;
-    clang_getSpellingLocation(clang_getRangeStart(skipped_ranges->ranges[i]),
-                              0, &start_line, &start_column, 0);
-    clang_getSpellingLocation(clang_getRangeEnd(skipped_ranges->ranges[i]),
-                              0, &end_line, &end_column, 0);
+    clang_getFileLocation(clang_getRangeStart(skipped_ranges->ranges[i]), 0,
+                          &start_line, &start_column, 0);
+    clang_getFileLocation(clang_getRangeEnd(skipped_ranges->ranges[i]), 0,
+                          &end_line, &end_column, 0);
     printf("Skipping: ");
     PrintExtent(stdout, start_line, start_column, end_line, end_column);
     printf("\n");
@@ -4351,10 +4349,10 @@ int perform_token_annotation(int argc, const char **argv) {
     case CXToken_Literal: kind = "Literal"; break;
     case CXToken_Comment: kind = "Comment"; break;
     }
-    clang_getSpellingLocation(clang_getRangeStart(extent),
-                              0, &start_line, &start_column, 0);
-    clang_getSpellingLocation(clang_getRangeEnd(extent),
-                              0, &end_line, &end_column, 0);
+    clang_getFileLocation(clang_getRangeStart(extent), 0, &start_line,
+                          &start_column, 0);
+    clang_getFileLocation(clang_getRangeEnd(extent), 0, &end_line, &end_column,
+                          0);
     printf("%s: \"%s\" ", kind, clang_getCString(spelling));
     clang_disposeString(spelling);
     PrintExtent(stdout, start_line, start_column, end_line, end_column);
diff --git a/clang/tools/clang-installapi/Options.cpp b/clang/tools/clang-installapi/Options.cpp
index 191e944ae91e..ae5b697b8eb9 100644
--- a/clang/tools/clang-installapi/Options.cpp
+++ b/clang/tools/clang-installapi/Options.cpp
@@ -594,9 +594,7 @@ getInterfaceFile(const StringRef Filename) {
   std::unique_ptr<InterfaceFile> IF;
   switch (identify_magic(Buffer->getBuffer())) {
   case file_magic::macho_dynamically_linked_shared_lib:
-    LLVM_FALLTHROUGH;
   case file_magic::macho_dynamically_linked_shared_lib_stub:
-    LLVM_FALLTHROUGH;
   case file_magic::macho_universal_binary:
     return DylibReader::get(Buffer->getMemBufferRef());
     break;
diff --git a/clang/tools/clang-repl/ClangRepl.cpp b/clang/tools/clang-repl/ClangRepl.cpp
index aecf61b97fc7..9cfc70462893 100644
--- a/clang/tools/clang-repl/ClangRepl.cpp
+++ b/clang/tools/clang-repl/ClangRepl.cpp
@@ -215,13 +215,15 @@ int main(int argc, const char **argv) {
   } else
     Interp = ExitOnErr(clang::Interpreter::create(std::move(CI)));
 
+  bool HasError = false;
+
   for (const std::string &input : OptInputs) {
-    if (auto Err = Interp->ParseAndExecute(input))
+    if (auto Err = Interp->ParseAndExecute(input)) {
       llvm::logAllUnhandledErrors(std::move(Err), llvm::errs(), "error: ");
+      HasError = true;
+    }
   }
 
-  bool HasError = false;
-
   if (OptInputs.empty()) {
     llvm::LineEditor LE("clang-repl");
     std::string Input;
@@ -241,18 +243,13 @@ int main(int argc, const char **argv) {
         break;
       }
       if (Input == R"(%undo)") {
-        if (auto Err = Interp->Undo()) {
+        if (auto Err = Interp->Undo())
           llvm::logAllUnhandledErrors(std::move(Err), llvm::errs(), "error: ");
-          HasError = true;
-        }
       } else if (Input.rfind("%lib ", 0) == 0) {
-        if (auto Err = Interp->LoadDynamicLibrary(Input.data() + 5)) {
+        if (auto Err = Interp->LoadDynamicLibrary(Input.data() + 5))
           llvm::logAllUnhandledErrors(std::move(Err), llvm::errs(), "error: ");
-          HasError = true;
-        }
       } else if (auto Err = Interp->ParseAndExecute(Input)) {
         llvm::logAllUnhandledErrors(std::move(Err), llvm::errs(), "error: ");
-        HasError = true;
       }
 
       Input = "";
diff --git a/clang/tools/driver/cc1_main.cpp b/clang/tools/driver/cc1_main.cpp
index b5c6be3c557b..2aebc6d3c017 100644
--- a/clang/tools/driver/cc1_main.cpp
+++ b/clang/tools/driver/cc1_main.cpp
@@ -39,7 +39,6 @@
 #include "llvm/Support/ManagedStatic.h"
 #include "llvm/Support/Path.h"
 #include "llvm/Support/Process.h"
-#include "llvm/Support/RISCVISAInfo.h"
 #include "llvm/Support/Signals.h"
 #include "llvm/Support/TargetSelect.h"
 #include "llvm/Support/TimeProfiler.h"
@@ -48,6 +47,7 @@
 #include "llvm/Target/TargetMachine.h"
 #include "llvm/TargetParser/AArch64TargetParser.h"
 #include "llvm/TargetParser/ARMTargetParser.h"
+#include "llvm/TargetParser/RISCVISAInfo.h"
 #include <cstdio>
 
 #ifdef CLANG_HAVE_RLIMITS
diff --git a/clang/tools/driver/cc1as_main.cpp b/clang/tools/driver/cc1as_main.cpp
index 5498c3f9d4a2..86afe22fac24 100644
--- a/clang/tools/driver/cc1as_main.cpp
+++ b/clang/tools/driver/cc1as_main.cpp
@@ -426,6 +426,7 @@ static bool ExecuteAssemblerImpl(AssemblerInvocation &Opts,
   assert(MRI && "Unable to create target register info!");
 
   MCTargetOptions MCOptions;
+  MCOptions.MCRelaxAll = Opts.RelaxAll;
   MCOptions.EmitDwarfUnwind = Opts.EmitDwarfUnwind;
   MCOptions.EmitCompactUnwindNonCanonical = Opts.EmitCompactUnwindNonCanonical;
   MCOptions.X86RelaxRelocations = Opts.RelaxELFRelocations;
diff --git a/clang/tools/libclang/CIndex.cpp b/clang/tools/libclang/CIndex.cpp
index 74163f30e19b..398a11a57035 100644
--- a/clang/tools/libclang/CIndex.cpp
+++ b/clang/tools/libclang/CIndex.cpp
@@ -5713,8 +5713,8 @@ CXString clang_getCursorKindSpelling(enum CXCursorKind Kind) {
     return cxstring::createRef("UnaryOperator");
   case CXCursor_ArraySubscriptExpr:
     return cxstring::createRef("ArraySubscriptExpr");
-  case CXCursor_OMPArraySectionExpr:
-    return cxstring::createRef("OMPArraySectionExpr");
+  case CXCursor_ArraySectionExpr:
+    return cxstring::createRef("ArraySectionExpr");
   case CXCursor_OMPArrayShapingExpr:
     return cxstring::createRef("OMPArrayShapingExpr");
   case CXCursor_OMPIteratorExpr:
diff --git a/clang/tools/libclang/CXCursor.cpp b/clang/tools/libclang/CXCursor.cpp
index 454bf7549861..9325a16d2a84 100644
--- a/clang/tools/libclang/CXCursor.cpp
+++ b/clang/tools/libclang/CXCursor.cpp
@@ -423,8 +423,8 @@ CXCursor cxcursor::MakeCXCursor(const Stmt *S, const Decl *Parent,
     K = CXCursor_UnexposedExpr;
     break;
 
-  case Stmt::OMPArraySectionExprClass:
-    K = CXCursor_OMPArraySectionExpr;
+  case Stmt::ArraySectionExprClass:
+    K = CXCursor_ArraySectionExpr;
     break;
 
   case Stmt::OMPArrayShapingExprClass:
diff --git a/clang/tools/libclang/CXSourceLocation.cpp b/clang/tools/libclang/CXSourceLocation.cpp
index ba70cbfee899..53cb71f7276f 100644
--- a/clang/tools/libclang/CXSourceLocation.cpp
+++ b/clang/tools/libclang/CXSourceLocation.cpp
@@ -319,8 +319,7 @@ void clang_getSpellingLocation(CXSourceLocation location,
   
   const SourceManager &SM =
   *static_cast<const SourceManager*>(location.ptr_data[0]);
-  // FIXME: This should call SourceManager::getSpellingLoc().
-  SourceLocation SpellLoc = SM.getFileLoc(Loc);
+  SourceLocation SpellLoc = SM.getSpellingLoc(Loc);
   std::pair<FileID, unsigned> LocInfo = SM.getDecomposedLoc(SpellLoc);
   FileID FID = LocInfo.first;
   unsigned FileOffset = LocInfo.second;
diff --git a/clang/unittests/Analysis/FlowSensitive/TestingSupport.h b/clang/unittests/Analysis/FlowSensitive/TestingSupport.h
index e3c7ff685f57..3b0e05ed7222 100644
--- a/clang/unittests/Analysis/FlowSensitive/TestingSupport.h
+++ b/clang/unittests/Analysis/FlowSensitive/TestingSupport.h
@@ -456,7 +456,7 @@ const IndirectFieldDecl *findIndirectFieldDecl(ASTContext &ASTCtx,
 /// Requirements:
 ///
 ///   `Name` must be unique in `ASTCtx`.
-template <class LocT>
+template <class LocT = StorageLocation>
 LocT &getLocForDecl(ASTContext &ASTCtx, const Environment &Env,
                     llvm::StringRef Name) {
   const ValueDecl *VD = findValueDecl(ASTCtx, Name);
@@ -470,7 +470,7 @@ LocT &getLocForDecl(ASTContext &ASTCtx, const Environment &Env,
 /// Requirements:
 ///
 ///   `Name` must be unique in `ASTCtx`.
-template <class ValueT>
+template <class ValueT = Value>
 ValueT &getValueForDecl(ASTContext &ASTCtx, const Environment &Env,
                         llvm::StringRef Name) {
   const ValueDecl *VD = findValueDecl(ASTCtx, Name);
diff --git a/clang/unittests/Analysis/FlowSensitive/TransferTest.cpp b/clang/unittests/Analysis/FlowSensitive/TransferTest.cpp
index bb16138126c8..301bec32c0cf 100644
--- a/clang/unittests/Analysis/FlowSensitive/TransferTest.cpp
+++ b/clang/unittests/Analysis/FlowSensitive/TransferTest.cpp
@@ -3309,6 +3309,28 @@ TEST(TransferTest, ResultObjectLocationPropagatesThroughConditionalOperator) {
       });
 }
 
+TEST(TransferTest, ResultObjectLocationDontVisitNestedRecordDecl) {
+  // This is a crash repro.
+  // We used to crash because when propagating result objects, we would visit
+  // nested record and function declarations, but we don't model fields used
+  // only in these.
+  std::string Code = R"(
+    struct S1 {};
+    struct S2 { S1 s1; };
+    void target() {
+      struct Nested {
+        void f() {
+          S2 s2 = { S1() };
+        }
+      };
+    }
+  )";
+  runDataflow(
+      Code,
+      [](const llvm::StringMap<DataflowAnalysisState<NoopLattice>> &Results,
+         ASTContext &ASTCtx) {});
+}
+
 TEST(TransferTest, StaticCast) {
   std::string Code = R"(
     void target(int Foo) {
@@ -3348,20 +3370,11 @@ TEST(TransferTest, IntegralCast) {
       Code,
       [](const llvm::StringMap<DataflowAnalysisState<NoopLattice>> &Results,
          ASTContext &ASTCtx) {
-        ASSERT_THAT(Results.keys(), UnorderedElementsAre("p"));
         const Environment &Env = getEnvironmentAtAnnotation(Results, "p");
 
-        const ValueDecl *FooDecl = findValueDecl(ASTCtx, "Foo");
-        ASSERT_THAT(FooDecl, NotNull());
-
-        const ValueDecl *BarDecl = findValueDecl(ASTCtx, "Bar");
-        ASSERT_THAT(BarDecl, NotNull());
-
-        const auto *FooVal = Env.getValue(*FooDecl);
-        const auto *BarVal = Env.getValue(*BarDecl);
-        EXPECT_TRUE(isa<IntegerValue>(FooVal));
-        EXPECT_TRUE(isa<IntegerValue>(BarVal));
-        EXPECT_EQ(FooVal, BarVal);
+        const auto &FooVal = getValueForDecl<IntegerValue>(ASTCtx, Env, "Foo");
+        const auto &BarVal = getValueForDecl<IntegerValue>(ASTCtx, Env, "Bar");
+        EXPECT_EQ(&FooVal, &BarVal);
       });
 }
 
@@ -3376,17 +3389,10 @@ TEST(TransferTest, IntegraltoBooleanCast) {
       Code,
       [](const llvm::StringMap<DataflowAnalysisState<NoopLattice>> &Results,
          ASTContext &ASTCtx) {
-        ASSERT_THAT(Results.keys(), UnorderedElementsAre("p"));
         const Environment &Env = getEnvironmentAtAnnotation(Results, "p");
 
-        const ValueDecl *FooDecl = findValueDecl(ASTCtx, "Foo");
-        ASSERT_THAT(FooDecl, NotNull());
-
-        const ValueDecl *BarDecl = findValueDecl(ASTCtx, "Bar");
-        ASSERT_THAT(BarDecl, NotNull());
-
-        const auto *FooVal = Env.getValue(*FooDecl);
-        const auto *BarVal = Env.getValue(*BarDecl);
+        const auto &FooVal = getValueForDecl(ASTCtx, Env, "Foo");
+        const auto &BarVal = getValueForDecl(ASTCtx, Env, "Bar");
         EXPECT_TRUE(isa<IntegerValue>(FooVal));
         EXPECT_TRUE(isa<BoolValue>(BarVal));
       });
@@ -3404,23 +3410,38 @@ TEST(TransferTest, IntegralToBooleanCastFromBool) {
       Code,
       [](const llvm::StringMap<DataflowAnalysisState<NoopLattice>> &Results,
          ASTContext &ASTCtx) {
-        ASSERT_THAT(Results.keys(), UnorderedElementsAre("p"));
         const Environment &Env = getEnvironmentAtAnnotation(Results, "p");
 
-        const ValueDecl *FooDecl = findValueDecl(ASTCtx, "Foo");
-        ASSERT_THAT(FooDecl, NotNull());
-
-        const ValueDecl *BarDecl = findValueDecl(ASTCtx, "Bar");
-        ASSERT_THAT(BarDecl, NotNull());
-
-        const auto *FooVal = Env.getValue(*FooDecl);
-        const auto *BarVal = Env.getValue(*BarDecl);
-        EXPECT_TRUE(isa<BoolValue>(FooVal));
-        EXPECT_TRUE(isa<BoolValue>(BarVal));
-        EXPECT_EQ(FooVal, BarVal);
+        const auto &FooVal = getValueForDecl<BoolValue>(ASTCtx, Env, "Foo");
+        const auto &BarVal = getValueForDecl<BoolValue>(ASTCtx, Env, "Bar");
+        EXPECT_EQ(&FooVal, &BarVal);
       });
 }
 
+TEST(TransferTest, WidenBoolValueInIntegerVariable) {
+  // This is a crash repro.
+  // This test sets up a case where we perform widening on an integer variable
+  // that contains a `BoolValue` for the previous iteration and an
+  // `IntegerValue` for the current iteration. We used to crash on this because
+  // `widenDistinctValues()` assumed that if the previous iteration had a
+  // `BoolValue`, the current iteration would too.
+  // FIXME: The real fix here is to make sure we never store `BoolValue`s in
+  // integer variables; see also the comment in `widenDistinctValues()`.
+  std::string Code = R"cc(
+    struct S {
+      int i;
+      S *next;
+    };
+    void target(S *s) {
+      for (; s; s = s->next)
+        s->i = false;
+    }
+  )cc";
+  runDataflow(Code,
+              [](const llvm::StringMap<DataflowAnalysisState<NoopLattice>> &,
+                 ASTContext &) {});
+}
+
 TEST(TransferTest, NullToPointerCast) {
   std::string Code = R"(
     using my_nullptr_t = decltype(nullptr);
@@ -3637,7 +3658,7 @@ TEST(TransferTest, VarDeclInitAssignConditionalOperator) {
     };
 
     void target(A Foo, A Bar, bool Cond) {
-      A Baz = Cond ?  Foo : Bar;
+      A Baz = Cond ?  A(Foo) : A(Bar);
       // Make sure A::i is modeled.
       Baz.i;
       /*[[p]]*/
@@ -5275,6 +5296,99 @@ TEST(TransferTest, BinaryOperatorComma) {
       });
 }
 
+TEST(TransferTest, ConditionalOperatorValue) {
+  std::string Code = R"(
+    void target(bool Cond, bool B1, bool B2) {
+      bool JoinSame = Cond ? B1 : B1;
+      bool JoinDifferent = Cond ? B1 : B2;
+      // [[p]]
+    }
+  )";
+  runDataflow(
+      Code,
+      [](const llvm::StringMap<DataflowAnalysisState<NoopLattice>> &Results,
+         ASTContext &ASTCtx) {
+        Environment Env = getEnvironmentAtAnnotation(Results, "p").fork();
+
+        auto &B1 = getValueForDecl<BoolValue>(ASTCtx, Env, "B1");
+        auto &B2 = getValueForDecl<BoolValue>(ASTCtx, Env, "B2");
+        auto &JoinSame = getValueForDecl<BoolValue>(ASTCtx, Env, "JoinSame");
+        auto &JoinDifferent =
+            getValueForDecl<BoolValue>(ASTCtx, Env, "JoinDifferent");
+
+        EXPECT_EQ(&JoinSame, &B1);
+
+        const Formula &JoinDifferentEqB1 =
+            Env.arena().makeEquals(JoinDifferent.formula(), B1.formula());
+        EXPECT_TRUE(Env.allows(JoinDifferentEqB1));
+        EXPECT_FALSE(Env.proves(JoinDifferentEqB1));
+
+        const Formula &JoinDifferentEqB2 =
+            Env.arena().makeEquals(JoinDifferent.formula(), B2.formula());
+        EXPECT_TRUE(Env.allows(JoinDifferentEqB2));
+        EXPECT_FALSE(Env.proves(JoinDifferentEqB1));
+      });
+}
+
+TEST(TransferTest, ConditionalOperatorLocation) {
+  std::string Code = R"(
+    void target(bool Cond, int I1, int I2) {
+      int &JoinSame = Cond ? I1 : I1;
+      int &JoinDifferent = Cond ? I1 : I2;
+      // [[p]]
+    }
+  )";
+  runDataflow(
+      Code,
+      [](const llvm::StringMap<DataflowAnalysisState<NoopLattice>> &Results,
+         ASTContext &ASTCtx) {
+        Environment Env = getEnvironmentAtAnnotation(Results, "p").fork();
+
+        StorageLocation &I1 = getLocForDecl(ASTCtx, Env, "I1");
+        StorageLocation &I2 = getLocForDecl(ASTCtx, Env, "I2");
+        StorageLocation &JoinSame = getLocForDecl(ASTCtx, Env, "JoinSame");
+        StorageLocation &JoinDifferent =
+            getLocForDecl(ASTCtx, Env, "JoinDifferent");
+
+        EXPECT_EQ(&JoinSame, &I1);
+
+        EXPECT_NE(&JoinDifferent, &I1);
+        EXPECT_NE(&JoinDifferent, &I2);
+      });
+}
+
+TEST(TransferTest, ConditionalOperatorOnConstantExpr) {
+  // This is a regression test: We used to crash when a `ConstantExpr` was used
+  // in the branches of a conditional operator.
+  std::string Code = R"cc(
+    consteval bool identity(bool B) { return B; }
+    void target(bool Cond) {
+      bool JoinTrueTrue = Cond ? identity(true) : identity(true);
+      bool JoinTrueFalse = Cond ? identity(true) : identity(false);
+      // [[p]]
+    }
+  )cc";
+  runDataflow(
+      Code,
+      [](const llvm::StringMap<DataflowAnalysisState<NoopLattice>> &Results,
+         ASTContext &ASTCtx) {
+        Environment Env = getEnvironmentAtAnnotation(Results, "p").fork();
+
+        auto &JoinTrueTrue =
+            getValueForDecl<BoolValue>(ASTCtx, Env, "JoinTrueTrue");
+        // FIXME: This test documents the current behavior, namely that we
+        // don't actually use the constant result of the `ConstantExpr` and
+        // instead treat it like a normal function call.
+        EXPECT_EQ(JoinTrueTrue.formula().kind(), Formula::Kind::AtomRef);
+        // EXPECT_TRUE(JoinTrueTrue.formula().literal());
+
+        auto &JoinTrueFalse =
+            getValueForDecl<BoolValue>(ASTCtx, Env, "JoinTrueFalse");
+        EXPECT_EQ(JoinTrueFalse.formula().kind(), Formula::Kind::AtomRef);
+      },
+      LangStandard::lang_cxx20);
+}
+
 TEST(TransferTest, IfStmtBranchExtendsFlowCondition) {
   std::string Code = R"(
     void target(bool Foo) {
@@ -5522,10 +5636,7 @@ TEST(TransferTest, ContextSensitiveReturnReferenceWithConditionalOperator) {
         auto *Loc = Env.getReturnStorageLocation();
         EXPECT_THAT(Loc, NotNull());
 
-        // TODO: We would really like to make this stronger assertion, but that
-        // doesn't work because we don't propagate values correctly through
-        // the conditional operator yet.
-        // EXPECT_EQ(Loc, SLoc);
+        EXPECT_EQ(Loc, SLoc);
       },
       {BuiltinOptions{ContextSensitiveOptions{}}});
 }
diff --git a/clang/unittests/Format/FormatTest.cpp b/clang/unittests/Format/FormatTest.cpp
index bc61b9c089e9..32ba6b6853c7 100644
--- a/clang/unittests/Format/FormatTest.cpp
+++ b/clang/unittests/Format/FormatTest.cpp
@@ -24507,16 +24507,25 @@ TEST_F(FormatTest, AlternativeOperators) {
   verifyFormat("int a compl(5);");
   verifyFormat("int a not(5);");
 
-  /* FIXME handle alternate tokens
-   * https://en.cppreference.com/w/cpp/language/operator_alternative
-  // alternative tokens
-  verifyFormat("compl foo();");     //  ~foo();
-  verifyFormat("foo() <%%>;");      // foo();
-  verifyFormat("void foo() <%%>;"); // void foo(){}
-  verifyFormat("int a <:1:>;");     // int a[1];[
+  verifyFormat("compl foo();");     // ~foo();
+  verifyFormat("foo() <%%>");       // foo() {}
+  verifyFormat("void foo() <%%>");  // void foo() {}
+  verifyFormat("int a<:1:>;");      // int a[1];
   verifyFormat("%:define ABC abc"); // #define ABC abc
   verifyFormat("%:%:");             // ##
-  */
+
+  verifyFormat("a = v(not;);\n"
+               "b = v(not+);\n"
+               "c = v(not x);\n"
+               "d = v(not 1);\n"
+               "e = v(not 123.f);");
+
+  verifyNoChange("#define ASSEMBLER_INSTRUCTION_LIST(V)  \\\n"
+                 "  V(and)                               \\\n"
+                 "  V(not)                               \\\n"
+                 "  V(not!)                              \\\n"
+                 "  V(other)",
+                 getLLVMStyleWithColumns(40));
 }
 
 TEST_F(FormatTest, STLWhileNotDefineChed) {
@@ -27354,6 +27363,45 @@ TEST_F(FormatTest, BreakAdjacentStringLiterals) {
   verifyFormat(Code, Style);
 }
 
+TEST_F(FormatTest, AlignUTFCommentsAndStringLiterals) {
+  verifyFormat(
+      "int rus;      // А теперь комментарии, например, на русском, 2-байта\n"
+      "int long_rus; // Верхний коммент еще не превысил границу в 80, однако\n"
+      "              // уже отодвинут. Перенос, при этом, отрабатывает верно");
+
+  auto Style = getLLVMStyle();
+  Style.ColumnLimit = 15;
+  verifyNoChange("#define test  \\\n"
+                 "  /* 测试 */  \\\n"
+                 "  \"aa\"        \\\n"
+                 "  \"bb\"",
+                 Style);
+
+  Style.ColumnLimit = 25;
+  verifyFormat("struct foo {\n"
+               "  int iiiiii; ///< iiiiii\n"
+               "  int b;      ///< ыыы\n"
+               "  int c;      ///< ыыыы\n"
+               "};",
+               Style);
+
+  Style.ColumnLimit = 35;
+  verifyFormat("#define SENSOR_DESC_1             \\\n"
+               "  \"{\"                             \\\n"
+               "  \"unit_of_measurement: \\\"°C\\\",\"  \\\n"
+               "  \"}\"",
+               Style);
+
+  Style.ColumnLimit = 80;
+  Style.AlignArrayOfStructures = FormatStyle::AIAS_Left;
+  verifyFormat("Languages languages = {\n"
+               "    Language{{'e', 'n'}, U\"Test English\" },\n"
+               "    Language{{'l', 'v'}, U\"Test Latviešu\"},\n"
+               "    Language{{'r', 'u'}, U\"Test Русский\" },\n"
+               "};",
+               Style);
+}
+
 } // namespace
 } // namespace test
 } // namespace format
diff --git a/clang/unittests/Format/SortIncludesTest.cpp b/clang/unittests/Format/SortIncludesTest.cpp
index 772eb53806b4..824fa0078cd0 100644
--- a/clang/unittests/Format/SortIncludesTest.cpp
+++ b/clang/unittests/Format/SortIncludesTest.cpp
@@ -6,19 +6,19 @@
 //
 //===----------------------------------------------------------------------===//
 
-#include "FormatTestUtils.h"
+#include "FormatTestBase.h"
 #include "clang/Format/Format.h"
 #include "llvm/ADT/StringRef.h"
 #include "llvm/Support/Debug.h"
 #include "gtest/gtest.h"
 
-#define DEBUG_TYPE "format-test"
+#define DEBUG_TYPE "sort-includes-test"
 
 namespace clang {
 namespace format {
 namespace {
 
-class SortIncludesTest : public ::testing::Test {
+class SortIncludesTest : public test::FormatTestBase {
 protected:
   std::vector<tooling::Range> GetCodeRange(StringRef Code) {
     return std::vector<tooling::Range>(1, tooling::Range(0, Code.size()));
@@ -821,6 +821,122 @@ TEST_F(SortIncludesTest, CalculatesCorrectCursorPositionWithRegrouping) {
   EXPECT_EQ(27u, newCursor(Code, 28)); // Start of last line
 }
 
+TEST_F(SortIncludesTest,
+       CalculatesCorrectCursorPositionWhenNoReplacementsWithRegroupingAndCRLF) {
+  Style.IncludeBlocks = Style.IBS_Regroup;
+  FmtStyle.LineEnding = FormatStyle::LE_CRLF;
+  Style.IncludeCategories = {
+      {"^\"a\"", 0, 0, false}, {"^\"b\"", 1, 1, false}, {".*", 2, 2, false}};
+  std::string Code = "#include \"a\"\r\n" // Start of line: 0
+                     "\r\n"               // Start of line: 14
+                     "#include \"b\"\r\n" // Start of line: 16
+                     "\r\n"               // Start of line: 30
+                     "#include \"c\"\r\n" // Start of line: 32
+                     "\r\n"               // Start of line: 46
+                     "int i;";            // Start of line: 48
+  verifyNoChange(Code);
+  EXPECT_EQ(0u, newCursor(Code, 0));
+  EXPECT_EQ(14u, newCursor(Code, 14));
+  EXPECT_EQ(16u, newCursor(Code, 16));
+  EXPECT_EQ(30u, newCursor(Code, 30));
+  EXPECT_EQ(32u, newCursor(Code, 32));
+  EXPECT_EQ(46u, newCursor(Code, 46));
+  EXPECT_EQ(48u, newCursor(Code, 48));
+}
+
+TEST_F(
+    SortIncludesTest,
+    CalculatesCorrectCursorPositionWhenRemoveLinesReplacementsWithRegroupingAndCRLF) {
+  Style.IncludeBlocks = Style.IBS_Regroup;
+  FmtStyle.LineEnding = FormatStyle::LE_CRLF;
+  Style.IncludeCategories = {{".*", 0, 0, false}};
+  std::string Code = "#include \"a\"\r\n"     // Start of line: 0
+                     "\r\n"                   // Start of line: 14
+                     "#include \"b\"\r\n"     // Start of line: 16
+                     "\r\n"                   // Start of line: 30
+                     "#include \"c\"\r\n"     // Start of line: 32
+                     "\r\n"                   // Start of line: 46
+                     "int i;";                // Start of line: 48
+  std::string Expected = "#include \"a\"\r\n" // Start of line: 0
+                         "#include \"b\"\r\n" // Start of line: 14
+                         "#include \"c\"\r\n" // Start of line: 28
+                         "\r\n"               // Start of line: 42
+                         "int i;";            // Start of line: 44
+  EXPECT_EQ(Expected, sort(Code));
+  EXPECT_EQ(0u, newCursor(Code, 0));
+  EXPECT_EQ(
+      14u,
+      newCursor(Code, 14)); // cursor on empty line in include block is ignored
+  EXPECT_EQ(14u, newCursor(Code, 16));
+  EXPECT_EQ(
+      30u,
+      newCursor(Code, 30)); // cursor on empty line in include block is ignored
+  EXPECT_EQ(28u, newCursor(Code, 32));
+  EXPECT_EQ(42u, newCursor(Code, 46));
+  EXPECT_EQ(44u, newCursor(Code, 48));
+}
+
+// FIXME: the tests below should pass.
+#if 0
+TEST_F(
+    SortIncludesTest,
+    CalculatesCorrectCursorPositionWhenNewLineReplacementsWithRegroupingAndCRLF) {
+  Style.IncludeBlocks = Style.IBS_Regroup;
+  FmtStyle.LineEnding = FormatStyle::LE_CRLF;
+  Style.IncludeCategories = {
+      {"^\"a\"", 0, 0, false}, {"^\"b\"", 1, 1, false}, {".*", 2, 2, false}};
+  std::string Code = "#include \"a\"\r\n"     // Start of line: 0
+                     "#include \"b\"\r\n"     // Start of line: 14
+                     "#include \"c\"\r\n"     // Start of line: 28
+                     "\r\n"                   // Start of line: 42
+                     "int i;";                // Start of line: 44
+  std::string Expected = "#include \"a\"\r\n" // Start of line: 0
+                         "\r\n"               // Start of line: 14
+                         "#include \"b\"\r\n" // Start of line: 16
+                         "\r\n"               // Start of line: 30
+                         "#include \"c\"\r\n" // Start of line: 32
+                         "\r\n"               // Start of line: 46
+                         "int i;";            // Start of line: 48
+  EXPECT_EQ(Expected, sort(Code));
+  EXPECT_EQ(0u, newCursor(Code, 0));
+  EXPECT_EQ(15u, newCursor(Code, 16));
+  EXPECT_EQ(30u, newCursor(Code, 32));
+  EXPECT_EQ(44u, newCursor(Code, 46));
+  EXPECT_EQ(46u, newCursor(Code, 48));
+}
+
+TEST_F(
+    SortIncludesTest,
+    CalculatesCorrectCursorPositionWhenNoNewLineReplacementsWithRegroupingAndCRLF) {
+  Style.IncludeBlocks = Style.IBS_Regroup;
+  FmtStyle.LineEnding = FormatStyle::LE_CRLF;
+  Style.IncludeCategories = {
+      {"^\"a\"", 0, 0, false}, {"^\"b\"", 1, 1, false}, {".*", 2, 2, false}};
+  std::string Code = "#include \"a\"\r\n"     // Start of line: 0
+                     "\r\n"                   // Start of line: 14
+                     "#include \"c\"\r\n"     // Start of line: 16
+                     "\r\n"                   // Start of line: 30
+                     "#include \"b\"\r\n"     // Start of line: 32
+                     "\r\n"                   // Start of line: 46
+                     "int i;";                // Start of line: 48
+  std::string Expected = "#include \"a\"\r\n" // Start of line: 0
+                         "\r\n"               // Start of line: 14
+                         "#include \"b\"\r\n" // Start of line: 16
+                         "\r\n"               // Start of line: 30
+                         "#include \"c\"\r\n" // Start of line: 32
+                         "\r\n"               // Start of line: 46
+                         "int i;";            // Start of line: 48
+  EXPECT_EQ(Expected, sort(Code));
+  EXPECT_EQ(0u, newCursor(Code, 0));
+  EXPECT_EQ(14u, newCursor(Code, 14));
+  EXPECT_EQ(30u, newCursor(Code, 32));
+  EXPECT_EQ(30u, newCursor(Code, 30));
+  EXPECT_EQ(15u, newCursor(Code, 15));
+  EXPECT_EQ(44u, newCursor(Code, 46));
+  EXPECT_EQ(46u, newCursor(Code, 48));
+}
+#endif
+
 TEST_F(SortIncludesTest, DeduplicateIncludes) {
   EXPECT_EQ("#include <a>\n"
             "#include <b>\n"
diff --git a/clang/unittests/Format/TokenAnnotatorTest.cpp b/clang/unittests/Format/TokenAnnotatorTest.cpp
index 34999b737639..dff5251d2e94 100644
--- a/clang/unittests/Format/TokenAnnotatorTest.cpp
+++ b/clang/unittests/Format/TokenAnnotatorTest.cpp
@@ -2855,6 +2855,83 @@ TEST_F(TokenAnnotatorTest, BraceKind) {
   ASSERT_EQ(Tokens.size(), 18u) << Tokens;
   EXPECT_BRACE_KIND(Tokens[8], BK_BracedInit);
   EXPECT_BRACE_KIND(Tokens[16], BK_BracedInit);
+
+  Tokens = annotate("struct {};");
+  ASSERT_EQ(Tokens.size(), 5u) << Tokens;
+  EXPECT_BRACE_KIND(Tokens[1], BK_Block);
+  EXPECT_BRACE_KIND(Tokens[2], BK_Block);
+
+  Tokens = annotate("struct : Base {};");
+  ASSERT_EQ(Tokens.size(), 7u) << Tokens;
+  EXPECT_BRACE_KIND(Tokens[3], BK_Block);
+  EXPECT_BRACE_KIND(Tokens[4], BK_Block);
+
+  Tokens = annotate("struct Foo {};");
+  ASSERT_EQ(Tokens.size(), 6u) << Tokens;
+  EXPECT_BRACE_KIND(Tokens[2], BK_Block);
+  EXPECT_BRACE_KIND(Tokens[3], BK_Block);
+
+  Tokens = annotate("struct ::Foo {};");
+  ASSERT_EQ(Tokens.size(), 7u) << Tokens;
+  EXPECT_BRACE_KIND(Tokens[3], BK_Block);
+  EXPECT_BRACE_KIND(Tokens[4], BK_Block);
+
+  Tokens = annotate("struct NS::Foo {};");
+  ASSERT_EQ(Tokens.size(), 8u) << Tokens;
+  EXPECT_BRACE_KIND(Tokens[4], BK_Block);
+  EXPECT_BRACE_KIND(Tokens[5], BK_Block);
+
+  Tokens = annotate("struct Foo<int> {};");
+  ASSERT_EQ(Tokens.size(), 9u) << Tokens;
+  EXPECT_BRACE_KIND(Tokens[5], BK_Block);
+  EXPECT_BRACE_KIND(Tokens[6], BK_Block);
+
+  Tokens = annotate("struct Foo final {};");
+  ASSERT_EQ(Tokens.size(), 7u) << Tokens;
+  EXPECT_BRACE_KIND(Tokens[3], BK_Block);
+  EXPECT_BRACE_KIND(Tokens[4], BK_Block);
+
+  Tokens = annotate("struct [[foo]] [[bar]] Foo final : Base1, Base2 {};");
+  ASSERT_EQ(Tokens.size(), 21u) << Tokens;
+  EXPECT_BRACE_KIND(Tokens[17], BK_Block);
+  EXPECT_BRACE_KIND(Tokens[18], BK_Block);
+
+  Tokens = annotate("struct Foo x{};");
+  ASSERT_EQ(Tokens.size(), 7u) << Tokens;
+  EXPECT_BRACE_KIND(Tokens[3], BK_BracedInit);
+  EXPECT_BRACE_KIND(Tokens[4], BK_BracedInit);
+
+  Tokens = annotate("struct ::Foo x{};");
+  ASSERT_EQ(Tokens.size(), 8u) << Tokens;
+  EXPECT_BRACE_KIND(Tokens[4], BK_BracedInit);
+  EXPECT_BRACE_KIND(Tokens[5], BK_BracedInit);
+
+  Tokens = annotate("struct NS::Foo x{};");
+  ASSERT_EQ(Tokens.size(), 9u) << Tokens;
+  EXPECT_BRACE_KIND(Tokens[5], BK_BracedInit);
+  EXPECT_BRACE_KIND(Tokens[6], BK_BracedInit);
+
+  Tokens = annotate("struct Foo<int> x{};");
+  ASSERT_EQ(Tokens.size(), 10u) << Tokens;
+  EXPECT_BRACE_KIND(Tokens[6], BK_BracedInit);
+  EXPECT_BRACE_KIND(Tokens[7], BK_BracedInit);
+
+  Tokens = annotate("#ifdef DEBUG_ENABLED\n"
+                    "#else\n"
+                    "#endif\n"
+                    "class RenderingServer : Object {\n"
+                    "#ifndef DISABLE_DEPRECATED\n"
+                    "  enum Features {\n"
+                    "    FEATURE_SHADERS,\n"
+                    "    FEATURE_MULTITHREADED,\n"
+                    "  };\n"
+                    "#endif\n"
+                    "};");
+  ASSERT_EQ(Tokens.size(), 29u) << Tokens;
+  EXPECT_BRACE_KIND(Tokens[11], BK_Block);
+  EXPECT_BRACE_KIND(Tokens[17], BK_Block);
+  EXPECT_BRACE_KIND(Tokens[22], BK_Block);
+  EXPECT_BRACE_KIND(Tokens[26], BK_Block);
 }
 
 TEST_F(TokenAnnotatorTest, UnderstandsElaboratedTypeSpecifier) {
diff --git a/clang/unittests/libclang/LibclangTest.cpp b/clang/unittests/libclang/LibclangTest.cpp
index 87075a46d751..6de4d02bf74f 100644
--- a/clang/unittests/libclang/LibclangTest.cpp
+++ b/clang/unittests/libclang/LibclangTest.cpp
@@ -1292,6 +1292,37 @@ void func() {}
   EXPECT_EQ(attrCount, 1);
 }
 
+TEST_F(LibclangParseTest, clang_getSpellingLocation) {
+  std::string fileName = "main.c";
+  WriteFile(fileName, "#define X(value) int x = value;\nX(42)\n");
+
+  ClangTU = clang_parseTranslationUnit(Index, fileName.c_str(), nullptr, 0,
+                                       nullptr, 0, TUFlags);
+
+  int declarationCount = 0;
+  Traverse([&declarationCount](CXCursor cursor,
+                               CXCursor parent) -> CXChildVisitResult {
+    if (cursor.kind == CXCursor_VarDecl) {
+      declarationCount++;
+
+      CXSourceLocation cxl = clang_getCursorLocation(cursor);
+      unsigned line;
+
+      // We expect clang_getFileLocation to return the expansion location,
+      // whereas clang_getSpellingLocation should resolve the macro expansion
+      // and return the location of the macro definition.
+      clang_getFileLocation(cxl, nullptr, &line, nullptr, nullptr);
+      EXPECT_EQ(line, 2U);
+      clang_getSpellingLocation(cxl, nullptr, &line, nullptr, nullptr);
+      EXPECT_EQ(line, 1U);
+    }
+
+    return CXChildVisit_Recurse;
+  });
+
+  EXPECT_EQ(declarationCount, 1);
+}
+
 class LibclangRewriteTest : public LibclangParseTest {
 public:
   CXRewriter Rew = nullptr;
diff --git a/clang/utils/TableGen/ClangAttrEmitter.cpp b/clang/utils/TableGen/ClangAttrEmitter.cpp
index 765cbbf3b04b..0d1365f09291 100644
--- a/clang/utils/TableGen/ClangAttrEmitter.cpp
+++ b/clang/utils/TableGen/ClangAttrEmitter.cpp
@@ -107,7 +107,7 @@ static std::string ReadPCHRecord(StringRef type) {
   return StringSwitch<std::string>(type)
       .EndsWith("Decl *", "Record.GetLocalDeclAs<" +
                               std::string(type.data(), 0, type.size() - 1) +
-                              ">(Record.readInt())")
+                              ">(LocalDeclID(Record.readInt()))")
       .Case("TypeSourceInfo *", "Record.readTypeSourceInfo()")
       .Case("Expr *", "Record.readExpr()")
       .Case("IdentifierInfo *", "Record.readIdentifier()")
@@ -1618,7 +1618,7 @@ writePrettyPrintFunction(const Record &R,
         Spelling += Namespace;
         Spelling += " ";
       }
-    } else if (Variety == "HLSLSemantic") {
+    } else if (Variety == "HLSLAnnotation") {
       Prefix = ":";
       Suffix = "";
     } else {
@@ -3608,7 +3608,7 @@ void EmitClangAttrHasAttrImpl(RecordKeeper &Records, raw_ostream &OS) {
   // and declspecs. Then generate a big switch statement for each of them.
   std::vector<Record *> Attrs = Records.getAllDerivedDefinitions("Attr");
   std::vector<std::pair<const Record *, FlattenedSpelling>> Declspec, Microsoft,
-      GNU, Pragma, HLSLSemantic;
+      GNU, Pragma, HLSLAnnotation;
   std::map<std::string,
            std::vector<std::pair<const Record *, FlattenedSpelling>>>
       CXX, C23;
@@ -3631,8 +3631,8 @@ void EmitClangAttrHasAttrImpl(RecordKeeper &Records, raw_ostream &OS) {
         C23[SI.nameSpace()].emplace_back(R, SI);
       else if (Variety == "Pragma")
         Pragma.emplace_back(R, SI);
-      else if (Variety == "HLSLSemantic")
-        HLSLSemantic.emplace_back(R, SI);
+      else if (Variety == "HLSLAnnotation")
+        HLSLAnnotation.emplace_back(R, SI);
     }
   }
 
@@ -3650,9 +3650,9 @@ void EmitClangAttrHasAttrImpl(RecordKeeper &Records, raw_ostream &OS) {
   OS << "case AttributeCommonInfo::Syntax::AS_Pragma:\n";
   OS << "  return llvm::StringSwitch<int>(Name)\n";
   GenerateHasAttrSpellingStringSwitch(Pragma, OS, "Pragma");
-  OS << "case AttributeCommonInfo::Syntax::AS_HLSLSemantic:\n";
+  OS << "case AttributeCommonInfo::Syntax::AS_HLSLAnnotation:\n";
   OS << "  return llvm::StringSwitch<int>(Name)\n";
-  GenerateHasAttrSpellingStringSwitch(HLSLSemantic, OS, "HLSLSemantic");
+  GenerateHasAttrSpellingStringSwitch(HLSLAnnotation, OS, "HLSLAnnotation");
   auto fn = [&OS](const char *Spelling,
                   const std::map<
                       std::string,
@@ -4669,7 +4669,7 @@ void EmitClangAttrParsedAttrKinds(RecordKeeper &Records, raw_ostream &OS) {
 
   std::vector<Record *> Attrs = Records.getAllDerivedDefinitions("Attr");
   std::vector<StringMatcher::StringPair> GNU, Declspec, Microsoft, CXX11,
-      Keywords, Pragma, C23, HLSLSemantic;
+      Keywords, Pragma, C23, HLSLAnnotation;
   std::set<std::string> Seen;
   for (const auto *A : Attrs) {
     const Record &Attr = *A;
@@ -4720,8 +4720,8 @@ void EmitClangAttrParsedAttrKinds(RecordKeeper &Records, raw_ostream &OS) {
           Matches = &Keywords;
         else if (Variety == "Pragma")
           Matches = &Pragma;
-        else if (Variety == "HLSLSemantic")
-          Matches = &HLSLSemantic;
+        else if (Variety == "HLSLAnnotation")
+          Matches = &HLSLAnnotation;
 
         assert(Matches && "Unsupported spelling variety found");
 
@@ -4757,8 +4757,8 @@ void EmitClangAttrParsedAttrKinds(RecordKeeper &Records, raw_ostream &OS) {
   StringMatcher("Name", Keywords, OS).Emit();
   OS << "  } else if (AttributeCommonInfo::AS_Pragma == Syntax) {\n";
   StringMatcher("Name", Pragma, OS).Emit();
-  OS << "  } else if (AttributeCommonInfo::AS_HLSLSemantic == Syntax) {\n";
-  StringMatcher("Name", HLSLSemantic, OS).Emit();
+  OS << "  } else if (AttributeCommonInfo::AS_HLSLAnnotation == Syntax) {\n";
+  StringMatcher("Name", HLSLAnnotation, OS).Emit();
   OS << "  }\n";
   OS << "  return AttributeCommonInfo::UnknownAttribute;\n"
      << "}\n";
@@ -4876,7 +4876,7 @@ enum class SpellingKind : size_t {
   Microsoft,
   Keyword,
   Pragma,
-  HLSLSemantic,
+  HLSLAnnotation,
   NumSpellingKinds
 };
 static const size_t NumSpellingKinds = (size_t)SpellingKind::NumSpellingKinds;
@@ -4890,15 +4890,16 @@ public:
   }
 
   void add(const Record &Attr, FlattenedSpelling Spelling) {
-    SpellingKind Kind = StringSwitch<SpellingKind>(Spelling.variety())
-                            .Case("GNU", SpellingKind::GNU)
-                            .Case("CXX11", SpellingKind::CXX11)
-                            .Case("C23", SpellingKind::C23)
-                            .Case("Declspec", SpellingKind::Declspec)
-                            .Case("Microsoft", SpellingKind::Microsoft)
-                            .Case("Keyword", SpellingKind::Keyword)
-                            .Case("Pragma", SpellingKind::Pragma)
-                            .Case("HLSLSemantic", SpellingKind::HLSLSemantic);
+    SpellingKind Kind =
+        StringSwitch<SpellingKind>(Spelling.variety())
+            .Case("GNU", SpellingKind::GNU)
+            .Case("CXX11", SpellingKind::CXX11)
+            .Case("C23", SpellingKind::C23)
+            .Case("Declspec", SpellingKind::Declspec)
+            .Case("Microsoft", SpellingKind::Microsoft)
+            .Case("Keyword", SpellingKind::Keyword)
+            .Case("Pragma", SpellingKind::Pragma)
+            .Case("HLSLAnnotation", SpellingKind::HLSLAnnotation);
     std::string Name;
     if (!Spelling.nameSpace().empty()) {
       switch (Kind) {
@@ -5007,7 +5008,8 @@ static void WriteDocumentation(RecordKeeper &Records,
   // so it must be last.
   OS << ".. csv-table:: Supported Syntaxes\n";
   OS << "   :header: \"GNU\", \"C++11\", \"C23\", \"``__declspec``\",";
-  OS << " \"Keyword\", \"``#pragma``\", \"HLSL Semantic\", \"``#pragma clang ";
+  OS << " \"Keyword\", \"``#pragma``\", \"HLSL Annotation\", \"``#pragma "
+        "clang ";
   OS << "attribute``\"\n\n   \"";
   for (size_t Kind = 0; Kind != NumSpellingKinds; ++Kind) {
     SpellingKind K = (SpellingKind)Kind;
diff --git a/clang/utils/TableGen/RISCVVEmitter.cpp b/clang/utils/TableGen/RISCVVEmitter.cpp
index 5e41ef9f9d26..48cd83cabfc7 100644
--- a/clang/utils/TableGen/RISCVVEmitter.cpp
+++ b/clang/utils/TableGen/RISCVVEmitter.cpp
@@ -670,6 +670,7 @@ void RVVEmitter::createRVVIntrinsics(
               .Case("Zvksed", RVV_REQ_Zvksed)
               .Case("Zvksh", RVV_REQ_Zvksh)
               .Case("Zvfbfwma", RVV_REQ_Zvfbfwma)
+              .Case("Zvfbfmin", RVV_REQ_Zvfbfmin)
               .Case("Experimental", RVV_REQ_Experimental)
               .Default(RVV_REQ_None);
       assert(RequireExt != RVV_REQ_None && "Unrecognized required feature?");
diff --git a/clang/www/cxx_dr_status.html b/clang/www/cxx_dr_status.html
index 83b71e7c122d..875521bd505d 100755
--- a/clang/www/cxx_dr_status.html
+++ b/clang/www/cxx_dr_status.html
@@ -1435,7 +1435,7 @@ accessible?</td>
   </tr>
   <tr class="open" id="233">
     <td><a href="https://cplusplus.github.io/CWG/issues/233.html">233</a></td>
-    <td>drafting</td>
+    <td>tentatively ready</td>
     <td>References vs pointers in UDC overload resolution</td>
     <td align="center">Not resolved</td>
   </tr>
@@ -2756,7 +2756,7 @@ of class templates</td>
   </tr>
   <tr id="453">
     <td><a href="https://cplusplus.github.io/CWG/issues/453.html">453</a></td>
-    <td>tentatively ready</td>
+    <td>DR</td>
     <td>References may only bind to &#8220;valid&#8221; objects</td>
     <td class="unknown" align="center">Unknown</td>
   </tr>
@@ -5812,7 +5812,7 @@ and <I>POD class</I></td>
   </tr>
   <tr class="open" id="1001">
     <td><a href="https://cplusplus.github.io/CWG/issues/1001.html">1001</a></td>
-    <td>drafting</td>
+    <td>review</td>
     <td>Parameter type adjustment in dependent parameter types</td>
     <td align="center">Not resolved</td>
   </tr>
@@ -6034,7 +6034,7 @@ and <I>POD class</I></td>
   </tr>
   <tr id="1038">
     <td><a href="https://cplusplus.github.io/CWG/issues/1038.html">1038</a></td>
-    <td>DR</td>
+    <td>DRWP</td>
     <td>Overload resolution of <TT>&amp;x.static_func</TT></td>
     <td class="unknown" align="center">Unknown</td>
   </tr>
@@ -9994,7 +9994,7 @@ and <I>POD class</I></td>
   </tr>
   <tr id="1698">
     <td><a href="https://cplusplus.github.io/CWG/issues/1698.html">1698</a></td>
-    <td>DR</td>
+    <td>DRWP</td>
     <td>Files ending in <TT>\</TT></td>
     <td class="unknown" align="center">Unknown</td>
   </tr>
@@ -10132,7 +10132,7 @@ and <I>POD class</I></td>
   </tr>
   <tr class="open" id="1721">
     <td><a href="https://cplusplus.github.io/CWG/issues/1721.html">1721</a></td>
-    <td>drafting</td>
+    <td>review</td>
     <td>Diagnosing ODR violations for static data members</td>
     <td align="center">Not resolved</td>
   </tr>
@@ -11312,11 +11312,11 @@ and <I>POD class</I></td>
     <td>decltype-qualified enumeration names</td>
     <td class="unknown" align="center">Unknown</td>
   </tr>
-  <tr class="open" id="1918">
+  <tr id="1918">
     <td><a href="https://cplusplus.github.io/CWG/issues/1918.html">1918</a></td>
-    <td>open</td>
+    <td>CD5</td>
     <td><TT>friend</TT> templates with dependent scopes</td>
-    <td align="center">Not resolved</td>
+    <td class="unknown" align="center">Unknown</td>
   </tr>
   <tr class="open" id="1919">
     <td><a href="https://cplusplus.github.io/CWG/issues/1919.html">1919</a></td>
@@ -11474,11 +11474,11 @@ and <I>POD class</I></td>
     <td>New C incompatibilities</td>
     <td align="center">Not resolved</td>
   </tr>
-  <tr class="open" id="1945">
+  <tr id="1945">
     <td><a href="https://cplusplus.github.io/CWG/issues/1945.html">1945</a></td>
-    <td>open</td>
+    <td>CD5</td>
     <td>Friend declarations naming members of class templates in non-templates</td>
-    <td align="center">Not resolved</td>
+    <td class="unknown" align="center">Unknown</td>
   </tr>
   <tr id="1946">
     <td><a href="https://cplusplus.github.io/CWG/issues/1946.html">1946</a></td>
@@ -11530,7 +11530,7 @@ and <I>POD class</I></td>
   </tr>
   <tr id="1954">
     <td><a href="https://cplusplus.github.io/CWG/issues/1954.html">1954</a></td>
-    <td>tentatively ready</td>
+    <td>DR</td>
     <td><TT>typeid</TT> null dereference check in subexpressions</td>
     <td class="unknown" align="center">Unknown</td>
   </tr>
@@ -12098,11 +12098,11 @@ and <I>POD class</I></td>
     <td>C-style casts that cast away constness vs <TT>static_cast</TT></td>
     <td align="center">Not resolved</td>
   </tr>
-  <tr class="open" id="2049">
+  <tr id="2049">
     <td><a href="https://cplusplus.github.io/CWG/issues/2049.html">2049</a></td>
-    <td>drafting</td>
+    <td>DRWP</td>
     <td>List initializer in non-type template default argument</td>
-    <td title="Clang 18 implements P2308R1 resolution" align="center">Not Resolved*</td>
+    <td class="full" align="center">Clang 18</td>
   </tr>
   <tr id="2050">
     <td><a href="https://cplusplus.github.io/CWG/issues/2050.html">2050</a></td>
@@ -12130,7 +12130,7 @@ and <I>POD class</I></td>
   </tr>
   <tr id="2054">
     <td><a href="https://cplusplus.github.io/CWG/issues/2054.html">2054</a></td>
-    <td>DR</td>
+    <td>DRWP</td>
     <td>Missing description of class SFINAE</td>
     <td class="unknown" align="center">Unknown</td>
   </tr>
@@ -12418,7 +12418,7 @@ and <I>POD class</I></td>
   </tr>
   <tr id="2102">
     <td><a href="https://cplusplus.github.io/CWG/issues/2102.html">2102</a></td>
-    <td>DR</td>
+    <td>DRWP</td>
     <td>Constructor checking in <I>new-expression</I></td>
     <td class="unknown" align="center">Unknown</td>
   </tr>
@@ -12698,11 +12698,11 @@ and <I>POD class</I></td>
     <td>Thread storage duration and order of initialization</td>
     <td align="center">Not resolved</td>
   </tr>
-  <tr class="open" id="2149">
+  <tr id="2149">
     <td><a href="https://cplusplus.github.io/CWG/issues/2149.html">2149</a></td>
-    <td>drafting</td>
+    <td>DR</td>
     <td>Brace elision and array length deduction</td>
-    <td align="center">Not resolved</td>
+    <td class="full" align="center">Clang 3.1</td>
   </tr>
   <tr id="2150">
     <td><a href="https://cplusplus.github.io/CWG/issues/2150.html">2150</a></td>
@@ -13318,7 +13318,7 @@ and <I>POD class</I></td>
   </tr>
   <tr id="2252">
     <td><a href="https://cplusplus.github.io/CWG/issues/2252.html">2252</a></td>
-    <td>DR</td>
+    <td>DRWP</td>
     <td>Enumeration list-initialization from the same type</td>
     <td class="unknown" align="center">Unknown</td>
   </tr>
@@ -14410,7 +14410,7 @@ and <I>POD class</I></td>
   </tr>
   <tr class="open" id="2434">
     <td><a href="https://cplusplus.github.io/CWG/issues/2434.html">2434</a></td>
-    <td>open</td>
+    <td>review</td>
     <td>Mandatory copy elision vs non-class objects</td>
     <td align="center">Not resolved</td>
   </tr>
@@ -14504,11 +14504,11 @@ and <I>POD class</I></td>
     <td>Thunks as an implementation technique for pointers to virtual functions</td>
     <td align="center">Extension</td>
   </tr>
-  <tr class="open" id="2450">
+  <tr id="2450">
     <td><a href="https://cplusplus.github.io/CWG/issues/2450.html">2450</a></td>
-    <td>review</td>
+    <td>DRWP</td>
     <td><I>braced-init-list</I> as a <I>template-argument</I></td>
-    <td title="Clang 18 implements P2308R1 resolution" align="center">Not Resolved*</td>
+    <td class="full" align="center">Clang 18</td>
   </tr>
   <tr id="2451">
     <td><a href="https://cplusplus.github.io/CWG/issues/2451.html">2451</a></td>
@@ -14558,11 +14558,11 @@ and <I>POD class</I></td>
     <td>Value category of expressions denoting non-static member functions</td>
     <td class="unknown" align="center">Unknown</td>
   </tr>
-  <tr class="open" id="2459">
+  <tr id="2459">
     <td><a href="https://cplusplus.github.io/CWG/issues/2459.html">2459</a></td>
-    <td>drafting</td>
+    <td>DRWP</td>
     <td>Template parameter initialization</td>
-    <td title="Clang 18 implements P2308R1 resolution" align="center">Not Resolved*</td>
+    <td class="full" align="center">Clang 18</td>
   </tr>
   <tr id="2460">
     <td><a href="https://cplusplus.github.io/CWG/issues/2460.html">2460</a></td>
@@ -14662,7 +14662,7 @@ and <I>POD class</I></td>
   </tr>
   <tr id="2476">
     <td><a href="https://cplusplus.github.io/CWG/issues/2476.html">2476</a></td>
-    <td>tentatively ready</td>
+    <td>DR</td>
     <td><I>placeholder-type-specifier</I>s and function declarators</td>
     <td class="unknown" align="center">Unknown</td>
   </tr>
@@ -14830,7 +14830,7 @@ and <I>POD class</I></td>
   </tr>
   <tr id="2504">
     <td><a href="https://cplusplus.github.io/CWG/issues/2504.html">2504</a></td>
-    <td>DR</td>
+    <td>DRWP</td>
     <td>Inheriting constructors from virtual base classes</td>
     <td class="none" align="center">No</td>
   </tr>
@@ -14992,7 +14992,7 @@ and <I>POD class</I></td>
   </tr>
   <tr id="2531">
     <td><a href="https://cplusplus.github.io/CWG/issues/2531.html">2531</a></td>
-    <td>DR</td>
+    <td>DRWP</td>
     <td>Static data members redeclared as constexpr</td>
     <td class="unknown" align="center">Unknown</td>
   </tr>
@@ -15002,11 +15002,11 @@ and <I>POD class</I></td>
     <td>Kind of pointer value returned by <TT>new T[0]</TT></td>
     <td align="center">Not resolved</td>
   </tr>
-  <tr class="open" id="2533">
+  <tr id="2533">
     <td><a href="https://cplusplus.github.io/CWG/issues/2533.html">2533</a></td>
-    <td>review</td>
+    <td>DR</td>
     <td>Storage duration of implicitly created objects</td>
-    <td align="center">Not resolved</td>
+    <td class="unknown" align="center">Unknown</td>
   </tr>
   <tr id="2534">
     <td><a href="https://cplusplus.github.io/CWG/issues/2534.html">2534</a></td>
@@ -15082,13 +15082,13 @@ and <I>POD class</I></td>
   </tr>
   <tr id="2546">
     <td><a href="https://cplusplus.github.io/CWG/issues/2546.html">2546</a></td>
-    <td>tentatively ready</td>
+    <td>DR</td>
     <td>Defaulted secondary comparison operators defined as deleted</td>
     <td class="unknown" align="center">Unknown</td>
   </tr>
   <tr id="2547">
     <td><a href="https://cplusplus.github.io/CWG/issues/2547.html">2547</a></td>
-    <td>tentatively ready</td>
+    <td>DR</td>
     <td>Defaulted comparison operator function for non-classes</td>
     <td class="unknown" align="center">Unknown</td>
   </tr>
@@ -15142,7 +15142,7 @@ and <I>POD class</I></td>
   </tr>
   <tr id="2556">
     <td><a href="https://cplusplus.github.io/CWG/issues/2556.html">2556</a></td>
-    <td>DR</td>
+    <td>DRWP</td>
     <td>Unusable <TT>promise::return_void</TT></td>
     <td class="unknown" align="center">Unknown</td>
   </tr>
@@ -15166,15 +15166,15 @@ and <I>POD class</I></td>
   </tr>
   <tr id="2560">
     <td><a href="https://cplusplus.github.io/CWG/issues/2560.html">2560</a></td>
-    <td>tentatively ready</td>
+    <td>DR</td>
     <td>Parameter type determination in a <I>requirement-parameter-list</I></td>
     <td class="unknown" align="center">Unknown</td>
   </tr>
   <tr class="open" id="2561">
     <td><a href="https://cplusplus.github.io/CWG/issues/2561.html">2561</a></td>
-    <td>review</td>
+    <td>tentatively ready</td>
     <td>Conversion to function pointer for lambda with explicit object parameter</td>
-    <td title="Clang 18 implements 2023-11-09 resolution" align="center">Not Resolved*</td>
+    <td title="Clang does not implement 2024-03-18 resolution" align="center">Not Resolved*</td>
   </tr>
   <tr class="open" id="2562">
     <td><a href="https://cplusplus.github.io/CWG/issues/2562.html">2562</a></td>
@@ -15214,7 +15214,7 @@ and <I>POD class</I></td>
   </tr>
   <tr id="2568">
     <td><a href="https://cplusplus.github.io/CWG/issues/2568.html">2568</a></td>
-    <td>tentatively ready</td>
+    <td>DR</td>
     <td>Access checking during synthesis of defaulted comparison operator</td>
     <td class="unknown" align="center">Unknown</td>
   </tr>
@@ -15226,7 +15226,7 @@ and <I>POD class</I></td>
   </tr>
   <tr id="2570">
     <td><a href="https://cplusplus.github.io/CWG/issues/2570.html">2570</a></td>
-    <td>DR</td>
+    <td>DRWP</td>
     <td>Clarify constexpr for defaulted functions</td>
     <td class="unknown" align="center">Unknown</td>
   </tr>
@@ -15334,7 +15334,7 @@ and <I>POD class</I></td>
   </tr>
   <tr class="open" id="2588">
     <td><a href="https://cplusplus.github.io/CWG/issues/2588.html">2588</a></td>
-    <td>drafting</td>
+    <td>tentatively ready</td>
     <td>friend declarations and module linkage</td>
     <td align="center">Not resolved</td>
   </tr>
@@ -15352,7 +15352,7 @@ and <I>POD class</I></td>
   </tr>
   <tr id="2591">
     <td><a href="https://cplusplus.github.io/CWG/issues/2591.html">2591</a></td>
-    <td>DR</td>
+    <td>DRWP</td>
     <td>Implicit change of active union member for anonymous union in union</td>
     <td class="unknown" align="center">Unknown</td>
   </tr>
@@ -15376,7 +15376,7 @@ and <I>POD class</I></td>
   </tr>
   <tr id="2595">
     <td><a href="https://cplusplus.github.io/CWG/issues/2595.html">2595</a></td>
-    <td>DR</td>
+    <td>DRWP</td>
     <td>"More constrained" for eligible special member functions</td>
     <td class="unknown" align="center">Unknown</td>
   </tr>
@@ -15406,7 +15406,7 @@ and <I>POD class</I></td>
   </tr>
   <tr id="2600">
     <td><a href="https://cplusplus.github.io/CWG/issues/2600.html">2600</a></td>
-    <td>DR</td>
+    <td>DRWP</td>
     <td>Type dependency of placeholder types</td>
     <td class="unknown" align="center">Unknown</td>
   </tr>
@@ -15574,7 +15574,7 @@ and <I>POD class</I></td>
   </tr>
   <tr id="2628">
     <td><a href="https://cplusplus.github.io/CWG/issues/2628.html">2628</a></td>
-    <td>DR</td>
+    <td>DRWP</td>
     <td>Implicit deduction guides should propagate constraints</td>
     <td class="none" align="center">No</td>
   </tr>
@@ -15610,7 +15610,7 @@ and <I>POD class</I></td>
   </tr>
   <tr id="2634">
     <td><a href="https://cplusplus.github.io/CWG/issues/2634.html">2634</a></td>
-    <td>tentatively ready</td>
+    <td>DR</td>
     <td>Avoid circularity in specification of scope for friend class declarations</td>
     <td class="unknown" align="center">Unknown</td>
   </tr>
@@ -15628,13 +15628,13 @@ and <I>POD class</I></td>
   </tr>
   <tr id="2637">
     <td><a href="https://cplusplus.github.io/CWG/issues/2637.html">2637</a></td>
-    <td>tentatively ready</td>
+    <td>DR</td>
     <td>Injected-class-name as a <I>simple-template-id</I></td>
     <td class="unknown" align="center">Unknown</td>
   </tr>
   <tr id="2638">
     <td><a href="https://cplusplus.github.io/CWG/issues/2638.html">2638</a></td>
-    <td>tentatively ready</td>
+    <td>DR</td>
     <td>Improve the example for initializing by initializer list</td>
     <td class="unknown" align="center">Unknown</td>
   </tr>
@@ -15748,7 +15748,7 @@ and <I>POD class</I></td>
   </tr>
   <tr id="2657">
     <td><a href="https://cplusplus.github.io/CWG/issues/2657.html">2657</a></td>
-    <td>tentatively ready</td>
+    <td>DR</td>
     <td>Cv-qualification adjustment when binding reference to temporary</td>
     <td class="unknown" align="center">Unknown</td>
   </tr>
@@ -15770,11 +15770,11 @@ and <I>POD class</I></td>
     <td>Confusing term "this parameter"</td>
     <td align="center">Not resolved</td>
   </tr>
-  <tr class="open" id="2661">
+  <tr id="2661">
     <td><a href="https://cplusplus.github.io/CWG/issues/2661.html">2661</a></td>
-    <td>open</td>
+    <td>DR</td>
     <td>Missing disambiguation rule for <I>pure-specifier</I> vs. <I>brace-or-equal-initializer</I></td>
-    <td align="center">Not resolved</td>
+    <td class="unknown" align="center">Unknown</td>
   </tr>
   <tr id="2662">
     <td><a href="https://cplusplus.github.io/CWG/issues/2662.html">2662</a></td>
@@ -15814,7 +15814,7 @@ and <I>POD class</I></td>
   </tr>
   <tr id="2668">
     <td><a href="https://cplusplus.github.io/CWG/issues/2668.html">2668</a></td>
-    <td>tentatively ready</td>
+    <td>DR</td>
     <td><TT>co_await</TT> in a <I>lambda-expression</I></td>
     <td class="unknown" align="center">Unknown</td>
   </tr>
@@ -15838,7 +15838,7 @@ and <I>POD class</I></td>
   </tr>
   <tr id="2672">
     <td><a href="https://cplusplus.github.io/CWG/issues/2672.html">2672</a></td>
-    <td>DR</td>
+    <td>DRWP</td>
     <td>Lambda body SFINAE is still required, contrary to intent and note</td>
     <td class="full" align="center">Clang 18</td>
   </tr>
@@ -15940,7 +15940,7 @@ and <I>POD class</I></td>
   </tr>
   <tr id="2689">
     <td><a href="https://cplusplus.github.io/CWG/issues/2689.html">2689</a></td>
-    <td>tentatively ready</td>
+    <td>DR</td>
     <td>Are cv-qualified <TT>std::nullptr_t</TT> fundamental types?</td>
     <td class="unknown" align="center">Unknown</td>
   </tr>
@@ -16004,11 +16004,11 @@ and <I>POD class</I></td>
     <td>Inconsistency of <I>throw-expression</I> specification</td>
     <td class="unknown" align="center">Unknown</td>
   </tr>
-  <tr class="open" id="2700">
+  <tr id="2700">
     <td><a href="https://cplusplus.github.io/CWG/issues/2700.html">2700</a></td>
-    <td>review</td>
+    <td>DR</td>
     <td><TT>#error</TT> disallows existing implementation practice</td>
-    <td align="center">Not resolved</td>
+    <td class="unknown" align="center">Unknown</td>
   </tr>
   <tr class="open" id="2701">
     <td><a href="https://cplusplus.github.io/CWG/issues/2701.html">2701</a></td>
@@ -16048,7 +16048,7 @@ and <I>POD class</I></td>
   </tr>
   <tr id="2707">
     <td><a href="https://cplusplus.github.io/CWG/issues/2707.html">2707</a></td>
-    <td>tentatively ready</td>
+    <td>DR</td>
     <td>Deduction guides cannot have a trailing <I>requires-clause</I></td>
     <td class="unknown" align="center">Unknown</td>
   </tr>
@@ -16088,11 +16088,11 @@ and <I>POD class</I></td>
     <td>Initialization of reference-to-aggregate from designated initializer list</td>
     <td class="unknown" align="center">Unknown</td>
   </tr>
-  <tr class="open" id="2714">
+  <tr id="2714">
     <td><a href="https://cplusplus.github.io/CWG/issues/2714.html">2714</a></td>
-    <td>open</td>
+    <td>DR</td>
     <td>Implicit deduction guides omit properties from the parameter-declaration-clause of a constructor</td>
-    <td align="center">Not resolved</td>
+    <td class="unknown" align="center">Unknown</td>
   </tr>
   <tr id="2715">
     <td><a href="https://cplusplus.github.io/CWG/issues/2715.html">2715</a></td>
@@ -16156,7 +16156,7 @@ and <I>POD class</I></td>
   </tr>
   <tr id="2725">
     <td><a href="https://cplusplus.github.io/CWG/issues/2725.html">2725</a></td>
-    <td>DR</td>
+    <td>DRWP</td>
     <td>Overload resolution for non-call of class member access</td>
     <td class="unknown" align="center">Unknown</td>
   </tr>
@@ -16174,7 +16174,7 @@ and <I>POD class</I></td>
   </tr>
   <tr class="open" id="2728">
     <td><a href="https://cplusplus.github.io/CWG/issues/2728.html">2728</a></td>
-    <td>open</td>
+    <td>tentatively ready</td>
     <td>Evaluation of conversions in a <I>delete-expression</I></td>
     <td align="center">Not resolved</td>
   </tr>
@@ -16204,7 +16204,7 @@ and <I>POD class</I></td>
   </tr>
   <tr id="2733">
     <td><a href="https://cplusplus.github.io/CWG/issues/2733.html">2733</a></td>
-    <td>DR</td>
+    <td>DRWP</td>
     <td>Applying <TT>[[maybe_unused]]</TT> to a label</td>
     <td class="unknown" align="center">Unknown</td>
   </tr>
@@ -16228,7 +16228,7 @@ and <I>POD class</I></td>
   </tr>
   <tr class="open" id="2737">
     <td><a href="https://cplusplus.github.io/CWG/issues/2737.html">2737</a></td>
-    <td>open</td>
+    <td>review</td>
     <td>Temporary lifetime extension for reference init-captures</td>
     <td align="center">Not resolved</td>
   </tr>
@@ -16258,7 +16258,7 @@ and <I>POD class</I></td>
   </tr>
   <tr class="open" id="2742">
     <td><a href="https://cplusplus.github.io/CWG/issues/2742.html">2742</a></td>
-    <td>open</td>
+    <td>drafting</td>
     <td>Guaranteed copy elision for brace-initialization from prvalue</td>
     <td align="center">Not resolved</td>
   </tr>
@@ -16274,33 +16274,33 @@ and <I>POD class</I></td>
     <td>Multiple objects of the same type at the same address</td>
     <td align="center">Not resolved</td>
   </tr>
-  <tr class="open" id="2745">
+  <tr id="2745">
     <td><a href="https://cplusplus.github.io/CWG/issues/2745.html">2745</a></td>
-    <td>open</td>
+    <td>DR</td>
     <td>Dependent odr-use in generic lambdas</td>
-    <td align="center">Not resolved</td>
+    <td class="unknown" align="center">Unknown</td>
   </tr>
-  <tr class="open" id="2746">
+  <tr id="2746">
     <td><a href="https://cplusplus.github.io/CWG/issues/2746.html">2746</a></td>
-    <td>open</td>
+    <td>DR</td>
     <td>Checking of default template arguments</td>
-    <td align="center">Not resolved</td>
+    <td class="unknown" align="center">Unknown</td>
   </tr>
   <tr id="2747">
     <td><a href="https://cplusplus.github.io/CWG/issues/2747.html">2747</a></td>
-    <td>DR</td>
+    <td>DRWP</td>
     <td>Cannot depend on an already-deleted splice</td>
     <td class="unknown" align="center">Unknown</td>
   </tr>
   <tr id="2748">
     <td><a href="https://cplusplus.github.io/CWG/issues/2748.html">2748</a></td>
-    <td>tentatively ready</td>
+    <td>DR</td>
     <td>Accessing static data members via null pointer</td>
     <td class="unknown" align="center">Unknown</td>
   </tr>
   <tr id="2749">
     <td><a href="https://cplusplus.github.io/CWG/issues/2749.html">2749</a></td>
-    <td>DR</td>
+    <td>DRWP</td>
     <td>Treatment of "pointer to void" for relational comparisons</td>
     <td class="unknown" align="center">Unknown</td>
   </tr>
@@ -16324,19 +16324,19 @@ and <I>POD class</I></td>
   </tr>
   <tr id="2753">
     <td><a href="https://cplusplus.github.io/CWG/issues/2753.html">2753</a></td>
-    <td>DR</td>
+    <td>DRWP</td>
     <td>Storage reuse for string literal objects and backing arrays</td>
     <td class="unknown" align="center">Unknown</td>
   </tr>
   <tr id="2754">
     <td><a href="https://cplusplus.github.io/CWG/issues/2754.html">2754</a></td>
-    <td>DR</td>
+    <td>DRWP</td>
     <td>Using *this in explicit object member functions that are coroutines</td>
     <td class="unknown" align="center">Unknown</td>
   </tr>
   <tr id="2755">
     <td><a href="https://cplusplus.github.io/CWG/issues/2755.html">2755</a></td>
-    <td>DR</td>
+    <td>DRWP</td>
     <td>Incorrect wording applied by P2738R1</td>
     <td class="unknown" align="center">Unknown</td>
   </tr>
@@ -16354,43 +16354,43 @@ and <I>POD class</I></td>
   </tr>
   <tr id="2758">
     <td><a href="https://cplusplus.github.io/CWG/issues/2758.html">2758</a></td>
-    <td>DR</td>
+    <td>DRWP</td>
     <td>What is "access and ambiguity control"?</td>
     <td class="unknown" align="center">Unknown</td>
   </tr>
   <tr id="2759">
     <td><a href="https://cplusplus.github.io/CWG/issues/2759.html">2759</a></td>
-    <td>DR</td>
+    <td>DRWP</td>
     <td>[[no_unique_address] and common initial sequence</td>
     <td class="unreleased" align="center">Clang 19</td>
   </tr>
   <tr id="2760">
     <td><a href="https://cplusplus.github.io/CWG/issues/2760.html">2760</a></td>
-    <td>DR</td>
+    <td>DRWP</td>
     <td>Defaulted constructor that is an immediate function</td>
     <td class="unknown" align="center">Unknown</td>
   </tr>
   <tr id="2761">
     <td><a href="https://cplusplus.github.io/CWG/issues/2761.html">2761</a></td>
-    <td>DR</td>
+    <td>DRWP</td>
     <td>Implicitly invoking the deleted destructor of an anonymous union member</td>
     <td class="unknown" align="center">Unknown</td>
   </tr>
   <tr id="2762">
     <td><a href="https://cplusplus.github.io/CWG/issues/2762.html">2762</a></td>
-    <td>DR</td>
+    <td>DRWP</td>
     <td>Type of implicit object parameter</td>
     <td class="unknown" align="center">Unknown</td>
   </tr>
   <tr id="2763">
     <td><a href="https://cplusplus.github.io/CWG/issues/2763.html">2763</a></td>
-    <td>DR</td>
+    <td>DRWP</td>
     <td>Ignorability of [[noreturn]] during constant evaluation</td>
     <td class="unknown" align="center">Unknown</td>
   </tr>
   <tr id="2764">
     <td><a href="https://cplusplus.github.io/CWG/issues/2764.html">2764</a></td>
-    <td>DR</td>
+    <td>DRWP</td>
     <td>Use of placeholders affecting name mangling</td>
     <td class="unknown" align="center">Unknown</td>
   </tr>
@@ -16403,7 +16403,8 @@ and <I>POD class</I></td>
   <tr class="open" id="2766">
     <td><a href="https://cplusplus.github.io/CWG/issues/2766.html">2766</a></td>
     <td>open</td>
-    <td>Repeated evaluation of a <I>string-literal</I> may yield different objects</td>
+    <td>Repeated evaluation of a <I>string-literal</I> may yield different
+objects</td>
     <td align="center">Not resolved</td>
   </tr>
   <tr class="open" id="2767">
@@ -16414,7 +16415,7 @@ and <I>POD class</I></td>
   </tr>
   <tr id="2768">
     <td><a href="https://cplusplus.github.io/CWG/issues/2768.html">2768</a></td>
-    <td>DR</td>
+    <td>DRWP</td>
     <td>Assignment to enumeration variable with a <I>braced-init-list</I></td>
     <td class="unknown" align="center">Unknown</td>
   </tr>
@@ -16430,15 +16431,15 @@ and <I>POD class</I></td>
     <td>Trailing <I>requires-clause</I> can refer to function parameters before they are substituted into</td>
     <td align="center">Not resolved</td>
   </tr>
-  <tr class="open" id="2771">
+  <tr id="2771">
     <td><a href="https://cplusplus.github.io/CWG/issues/2771.html">2771</a></td>
-    <td>open</td>
+    <td>DR</td>
     <td>Transformation for <I>unqualified-id</I>s in address operator</td>
-    <td align="center">Not resolved</td>
+    <td class="unknown" align="center">Unknown</td>
   </tr>
   <tr id="2772">
     <td><a href="https://cplusplus.github.io/CWG/issues/2772.html">2772</a></td>
-    <td>DR</td>
+    <td>DRWP</td>
     <td>Missing Annex C entry for linkage effects of <I>linkage-specification</I></td>
     <td class="unknown" align="center">Unknown</td>
   </tr>
@@ -16456,7 +16457,7 @@ and <I>POD class</I></td>
   </tr>
   <tr id="2775">
     <td><a href="https://cplusplus.github.io/CWG/issues/2775.html">2775</a></td>
-    <td>tentatively ready</td>
+    <td>DR</td>
     <td>Unclear argument type for copy of exception object</td>
     <td class="unknown" align="center">Unknown</td>
   </tr>
@@ -16466,15 +16467,15 @@ and <I>POD class</I></td>
     <td>Substitution failure and implementation limits</td>
     <td align="center">Not resolved</td>
   </tr>
-  <tr class="open" id="2777">
+  <tr id="2777">
     <td><a href="https://cplusplus.github.io/CWG/issues/2777.html">2777</a></td>
-    <td>open</td>
+    <td>DR</td>
     <td>Type of <I>id-expression</I> denoting a template parameter object</td>
-    <td align="center">Not resolved</td>
+    <td class="unknown" align="center">Unknown</td>
   </tr>
   <tr class="open" id="2778">
     <td><a href="https://cplusplus.github.io/CWG/issues/2778.html">2778</a></td>
-    <td>open</td>
+    <td>review</td>
     <td>Trivial destructor does not imply constant destruction</td>
     <td align="center">Not resolved</td>
   </tr>
@@ -16486,7 +16487,7 @@ and <I>POD class</I></td>
   </tr>
   <tr id="2780">
     <td><a href="https://cplusplus.github.io/CWG/issues/2780.html">2780</a></td>
-    <td>DR</td>
+    <td>DRWP</td>
     <td><TT>reinterpret_cast</TT> to reference to function types</td>
     <td class="unknown" align="center">Unknown</td>
   </tr>
@@ -16504,7 +16505,7 @@ and <I>POD class</I></td>
   </tr>
   <tr id="2783">
     <td><a href="https://cplusplus.github.io/CWG/issues/2783.html">2783</a></td>
-    <td>DR</td>
+    <td>DRWP</td>
     <td>Handling of deduction guides in <I>global-module-fragment</I></td>
     <td class="unknown" align="center">Unknown</td>
   </tr>
@@ -16516,7 +16517,7 @@ and <I>POD class</I></td>
   </tr>
   <tr id="2785">
     <td><a href="https://cplusplus.github.io/CWG/issues/2785.html">2785</a></td>
-    <td>DR</td>
+    <td>DRWP</td>
     <td>Type-dependence of <I>requires-expression</I></td>
     <td class="unknown" align="center">Unknown</td>
   </tr>
@@ -16540,7 +16541,7 @@ and <I>POD class</I></td>
   </tr>
   <tr id="2789">
     <td><a href="https://cplusplus.github.io/CWG/issues/2789.html">2789</a></td>
-    <td>DR</td>
+    <td>DRWP</td>
     <td>Overload resolution with implicit and explicit object member functions</td>
     <td class="full" align="center">Clang 18</td>
   </tr>
@@ -16552,19 +16553,19 @@ and <I>POD class</I></td>
   </tr>
   <tr id="2791">
     <td><a href="https://cplusplus.github.io/CWG/issues/2791.html">2791</a></td>
-    <td>DR</td>
+    <td>DRWP</td>
     <td>Unclear phrasing about "returning to the caller"</td>
     <td class="unknown" align="center">Unknown</td>
   </tr>
   <tr id="2792">
     <td><a href="https://cplusplus.github.io/CWG/issues/2792.html">2792</a></td>
-    <td>DR</td>
+    <td>DRWP</td>
     <td>Clean up specification of <TT>noexcept</TT> operator</td>
     <td class="unknown" align="center">Unknown</td>
   </tr>
   <tr id="2793">
     <td><a href="https://cplusplus.github.io/CWG/issues/2793.html">2793</a></td>
-    <td>DR</td>
+    <td>DRWP</td>
     <td>Block-scope declaration conflicting with parameter name</td>
     <td class="unknown" align="center">Unknown</td>
   </tr>
@@ -16576,25 +16577,25 @@ and <I>POD class</I></td>
   </tr>
   <tr id="2795">
     <td><a href="https://cplusplus.github.io/CWG/issues/2795.html">2795</a></td>
-    <td>DR</td>
+    <td>DRWP</td>
     <td>Overlapping empty subobjects with different cv-qualification</td>
     <td class="unknown" align="center">Unknown</td>
   </tr>
   <tr id="2796">
     <td><a href="https://cplusplus.github.io/CWG/issues/2796.html">2796</a></td>
-    <td>DR</td>
+    <td>DRWP</td>
     <td>Function pointer conversions for relational operators</td>
     <td class="unknown" align="center">Unknown</td>
   </tr>
   <tr class="open" id="2797">
     <td><a href="https://cplusplus.github.io/CWG/issues/2797.html">2797</a></td>
-    <td>open</td>
+    <td>review</td>
     <td>Meaning of "corresponds" for rewritten operator candidates</td>
     <td align="center">Not resolved</td>
   </tr>
   <tr id="2798">
     <td><a href="https://cplusplus.github.io/CWG/issues/2798.html">2798</a></td>
-    <td>DR</td>
+    <td>DRWP</td>
     <td>Manifestly constant evaluation of the <TT>static_assert</TT> message</td>
     <td class="full" align="center">Clang 17</td>
   </tr>
@@ -16612,7 +16613,7 @@ and <I>POD class</I></td>
   </tr>
   <tr id="2801">
     <td><a href="https://cplusplus.github.io/CWG/issues/2801.html">2801</a></td>
-    <td>DR</td>
+    <td>DRWP</td>
     <td>Reference binding with reference-related types</td>
     <td class="unknown" align="center">Unknown</td>
   </tr>
@@ -16624,7 +16625,7 @@ and <I>POD class</I></td>
   </tr>
   <tr id="2803">
     <td><a href="https://cplusplus.github.io/CWG/issues/2803.html">2803</a></td>
-    <td>tentatively ready</td>
+    <td>DR</td>
     <td>Overload resolution for reference binding of similar types</td>
     <td class="unknown" align="center">Unknown</td>
   </tr>
@@ -16642,13 +16643,13 @@ and <I>POD class</I></td>
   </tr>
   <tr id="2806">
     <td><a href="https://cplusplus.github.io/CWG/issues/2806.html">2806</a></td>
-    <td>DR</td>
+    <td>DRWP</td>
     <td>Make a <I>type-requirement</I> a type-only context</td>
     <td class="unknown" align="center">Unknown</td>
   </tr>
   <tr id="2807">
     <td><a href="https://cplusplus.github.io/CWG/issues/2807.html">2807</a></td>
-    <td>DR</td>
+    <td>DRWP</td>
     <td>Destructors declared <TT>consteval</TT></td>
     <td class="unknown" align="center">Unknown</td>
   </tr>
@@ -16660,19 +16661,19 @@ and <I>POD class</I></td>
   </tr>
   <tr id="2809">
     <td><a href="https://cplusplus.github.io/CWG/issues/2809.html">2809</a></td>
-    <td>tentatively ready</td>
+    <td>DR</td>
     <td>An implicit definition does not redeclare a function</td>
     <td class="unknown" align="center">Unknown</td>
   </tr>
   <tr id="2810">
     <td><a href="https://cplusplus.github.io/CWG/issues/2810.html">2810</a></td>
-    <td>tentatively ready</td>
+    <td>DR</td>
     <td>Requiring the absence of diagnostics for templates</td>
     <td class="unknown" align="center">Unknown</td>
   </tr>
   <tr id="2811">
     <td><a href="https://cplusplus.github.io/CWG/issues/2811.html">2811</a></td>
-    <td>tentatively ready</td>
+    <td>DR</td>
     <td>Clarify "use" of main</td>
     <td class="unknown" align="center">Unknown</td>
   </tr>
@@ -16682,11 +16683,11 @@ and <I>POD class</I></td>
     <td>Allocation with explicit alignment</td>
     <td align="center">Not resolved</td>
   </tr>
-  <tr class="open" id="2813">
+  <tr id="2813">
     <td><a href="https://cplusplus.github.io/CWG/issues/2813.html">2813</a></td>
-    <td>review</td>
+    <td>DR</td>
     <td>Class member access with prvalues</td>
-    <td align="center">Not resolved</td>
+    <td class="unknown" align="center">Unknown</td>
   </tr>
   <tr class="open" id="2814">
     <td><a href="https://cplusplus.github.io/CWG/issues/2814.html">2814</a></td>
@@ -16714,57 +16715,57 @@ and <I>POD class</I></td>
   </tr>
   <tr class="open" id="2818">
     <td><a href="https://cplusplus.github.io/CWG/issues/2818.html">2818</a></td>
-    <td>review</td>
+    <td>tentatively ready</td>
     <td>Use of predefined reserved identifiers</td>
     <td align="center">Not resolved</td>
   </tr>
   <tr class="open" id="2819">
     <td><a href="https://cplusplus.github.io/CWG/issues/2819.html">2819</a></td>
-    <td>review</td>
+    <td>tentatively ready</td>
     <td>Cast from null pointer value in a constant expression</td>
-    <td align="center">Not resolved</td>
+    <td title="Clang 19 implements 2023-12-01 resolution" align="center">Not Resolved*</td>
   </tr>
-  <tr class="open" id="2820">
+  <tr id="2820">
     <td><a href="https://cplusplus.github.io/CWG/issues/2820.html">2820</a></td>
-    <td>open</td>
+    <td>DR</td>
     <td>Value-initialization and default constructors</td>
-    <td align="center">Not resolved</td>
+    <td class="unknown" align="center">Unknown</td>
   </tr>
   <tr class="open" id="2821">
     <td><a href="https://cplusplus.github.io/CWG/issues/2821.html">2821</a></td>
-    <td>open</td>
+    <td>review</td>
     <td>Lifetime, zero-initialization, and dynamic initialization</td>
     <td align="center">Not resolved</td>
   </tr>
   <tr id="2822">
     <td><a href="https://cplusplus.github.io/CWG/issues/2822.html">2822</a></td>
-    <td>tentatively ready</td>
+    <td>DR</td>
     <td>Side-effect-free pointer zap</td>
     <td class="unknown" align="center">Unknown</td>
   </tr>
   <tr id="2823">
     <td><a href="https://cplusplus.github.io/CWG/issues/2823.html">2823</a></td>
-    <td>DR</td>
+    <td>DRWP</td>
     <td>Implicit undefined behavior when dereferencing pointers</td>
     <td class="unknown" align="center">Unknown</td>
   </tr>
   <tr id="2824">
     <td><a href="https://cplusplus.github.io/CWG/issues/2824.html">2824</a></td>
-    <td>tentatively ready</td>
+    <td>DR</td>
     <td>Copy-initialization of arrays</td>
     <td class="unknown" align="center">Unknown</td>
   </tr>
   <tr id="2825">
     <td><a href="https://cplusplus.github.io/CWG/issues/2825.html">2825</a></td>
-    <td>tentatively ready</td>
+    <td>DR</td>
     <td>Range-based for statement using a <I>braced-init-list</I></td>
     <td class="unknown" align="center">Unknown</td>
   </tr>
-  <tr id="2826">
+  <tr class="open" id="2826">
     <td><a href="https://cplusplus.github.io/CWG/issues/2826.html">2826</a></td>
-    <td>tentatively ready</td>
+    <td>drafting</td>
     <td>Missing definition of "temporary expression"</td>
-    <td class="unknown" align="center">Unknown</td>
+    <td align="center">Not resolved</td>
   </tr>
   <tr class="open" id="2827">
     <td><a href="https://cplusplus.github.io/CWG/issues/2827.html">2827</a></td>
@@ -16772,11 +16773,11 @@ and <I>POD class</I></td>
     <td>Representation of unsigned integral types</td>
     <td align="center">Not resolved</td>
   </tr>
-  <tr class="open" id="2828">
+  <tr id="2828">
     <td><a href="https://cplusplus.github.io/CWG/issues/2828.html">2828</a></td>
-    <td>review</td>
+    <td>DR</td>
     <td>Ambiguous interpretation of C-style cast</td>
-    <td align="center">Not resolved</td>
+    <td class="unknown" align="center">Unknown</td>
   </tr>
   <tr class="open" id="2829">
     <td><a href="https://cplusplus.github.io/CWG/issues/2829.html">2829</a></td>
@@ -16784,17 +16785,17 @@ and <I>POD class</I></td>
     <td>Redundant case in restricting user-defined conversion sequences</td>
     <td align="center">Not resolved</td>
   </tr>
-  <tr class="open" id="2830">
+  <tr id="2830">
     <td><a href="https://cplusplus.github.io/CWG/issues/2830.html">2830</a></td>
-    <td>open</td>
+    <td>DR</td>
     <td>Top-level cv-qualification should be ignored for list-initialization</td>
-    <td align="center">Not resolved</td>
+    <td class="unknown" align="center">Unknown</td>
   </tr>
-  <tr class="open" id="2831">
+  <tr id="2831">
     <td><a href="https://cplusplus.github.io/CWG/issues/2831.html">2831</a></td>
-    <td>open</td>
+    <td>DR</td>
     <td>Non-templated function definitions and <I>requires-clause</I>s</td>
-    <td align="center">Not resolved</td>
+    <td class="unknown" align="center">Unknown</td>
   </tr>
   <tr class="open" id="2832">
     <td><a href="https://cplusplus.github.io/CWG/issues/2832.html">2832</a></td>
@@ -16810,7 +16811,7 @@ and <I>POD class</I></td>
   </tr>
   <tr class="open" id="2834">
     <td><a href="https://cplusplus.github.io/CWG/issues/2834.html">2834</a></td>
-    <td>open</td>
+    <td>review</td>
     <td>Partial ordering and explicit object parameters</td>
     <td align="center">Not resolved</td>
   </tr>
@@ -16822,7 +16823,7 @@ and <I>POD class</I></td>
   </tr>
   <tr class="open" id="2836">
     <td><a href="https://cplusplus.github.io/CWG/issues/2836.html">2836</a></td>
-    <td>open</td>
+    <td>review</td>
     <td>Conversion rank of <TT>long double</TT> and extended floating-point types</td>
     <td align="center">Not resolved</td>
   </tr>
@@ -16855,6 +16856,276 @@ and <I>POD class</I></td>
     <td>open</td>
     <td>When do const objects start being const?</td>
     <td align="center">Not resolved</td>
+  </tr>
+  <tr class="open" id="2842">
+    <td><a href="https://cplusplus.github.io/CWG/issues/2842.html">2842</a></td>
+    <td>open</td>
+    <td>Preferring an <TT>initializer_list</TT> over a single value</td>
+    <td align="center">Not resolved</td>
+  </tr>
+  <tr class="open" id="2843">
+    <td><a href="https://cplusplus.github.io/CWG/issues/2843.html">2843</a></td>
+    <td>review</td>
+    <td>Undated reference to Unicode makes C++ a moving target</td>
+    <td align="center">Not resolved</td>
+  </tr>
+  <tr class="open" id="2844">
+    <td><a href="https://cplusplus.github.io/CWG/issues/2844.html">2844</a></td>
+    <td>open</td>
+    <td>Enumerating a finite set of built-in candidates</td>
+    <td align="center">Not resolved</td>
+  </tr>
+  <tr id="2845">
+    <td><a href="https://cplusplus.github.io/CWG/issues/2845.html">2845</a></td>
+    <td>DR</td>
+    <td>Make the closure type of a captureless lambda a structural type</td>
+    <td class="unknown" align="center">Unknown</td>
+  </tr>
+  <tr id="2846">
+    <td><a href="https://cplusplus.github.io/CWG/issues/2846.html">2846</a></td>
+    <td>DR</td>
+    <td>Out-of-class definitions of explicit object member functions</td>
+    <td class="unknown" align="center">Unknown</td>
+  </tr>
+  <tr class="open" id="2847">
+    <td><a href="https://cplusplus.github.io/CWG/issues/2847.html">2847</a></td>
+    <td>review</td>
+    <td>Constrained explicit specializations of function templates at class scope</td>
+    <td title="Clang 19 implements 2024-03-01 resolution" align="center">Not Resolved*</td>
+  </tr>
+  <tr id="2848">
+    <td><a href="https://cplusplus.github.io/CWG/issues/2848.html">2848</a></td>
+    <td>DR</td>
+    <td>Omitting an empty template argument list for explicit instantiation</td>
+    <td class="unknown" align="center">Unknown</td>
+  </tr>
+  <tr id="2849">
+    <td><a href="https://cplusplus.github.io/CWG/issues/2849.html">2849</a></td>
+    <td>DR</td>
+    <td>Parameter objects are not temporary objects</td>
+    <td class="unknown" align="center">Unknown</td>
+  </tr>
+  <tr id="2850">
+    <td><a href="https://cplusplus.github.io/CWG/issues/2850.html">2850</a></td>
+    <td>DR</td>
+    <td>Unclear storage duration for function parameter objects</td>
+    <td class="unknown" align="center">Unknown</td>
+  </tr>
+  <tr id="2851">
+    <td><a href="https://cplusplus.github.io/CWG/issues/2851.html">2851</a></td>
+    <td>DR</td>
+    <td>Allow floating-point conversions in converted constant expressions</td>
+    <td class="unknown" align="center">Unknown</td>
+  </tr>
+  <tr class="open" id="2852">
+    <td><a href="https://cplusplus.github.io/CWG/issues/2852.html">2852</a></td>
+    <td>open</td>
+    <td>Complete-class contexts and class-scope lambdas</td>
+    <td align="center">Not resolved</td>
+  </tr>
+  <tr id="2853">
+    <td><a href="https://cplusplus.github.io/CWG/issues/2853.html">2853</a></td>
+    <td>DR</td>
+    <td>Pointer arithmetic with pointer to hypothetical element</td>
+    <td class="unknown" align="center">Unknown</td>
+  </tr>
+  <tr id="2854">
+    <td><a href="https://cplusplus.github.io/CWG/issues/2854.html">2854</a></td>
+    <td>DR</td>
+    <td>Storage duration of exception objects</td>
+    <td class="unknown" align="center">Unknown</td>
+  </tr>
+  <tr id="2855">
+    <td><a href="https://cplusplus.github.io/CWG/issues/2855.html">2855</a></td>
+    <td>DR</td>
+    <td>Undefined behavior in postfix increment</td>
+    <td class="unknown" align="center">Unknown</td>
+  </tr>
+  <tr id="2856">
+    <td><a href="https://cplusplus.github.io/CWG/issues/2856.html">2856</a></td>
+    <td>DR</td>
+    <td>Copy-list-initialization with explicit default constructors</td>
+    <td class="unknown" align="center">Unknown</td>
+  </tr>
+  <tr id="2857">
+    <td><a href="https://cplusplus.github.io/CWG/issues/2857.html">2857</a></td>
+    <td>DR</td>
+    <td>Argument-dependent lookup with incomplete class types</td>
+    <td class="unknown" align="center">Unknown</td>
+  </tr>
+  <tr class="open" id="2858">
+    <td><a href="https://cplusplus.github.io/CWG/issues/2858.html">2858</a></td>
+    <td>tentatively ready</td>
+    <td>Declarative <I>nested-name-specifier</I>s and <I>pack-index-specifier</I>s</td>
+    <td title="Clang 19 implements 2024-04-05 resolution" align="center">Not Resolved*</td>
+  </tr>
+  <tr class="open" id="2859">
+    <td><a href="https://cplusplus.github.io/CWG/issues/2859.html">2859</a></td>
+    <td>tentatively ready</td>
+    <td>Value-initialization with multiple default constructors</td>
+    <td align="center">Not resolved</td>
+  </tr>
+  <tr id="2860">
+    <td><a href="https://cplusplus.github.io/CWG/issues/2860.html">2860</a></td>
+    <td>dup</td>
+    <td>Remove and fix the term "vacuous initialization"</td>
+    <td class="unknown" align="center">Unknown</td>
+  </tr>
+  <tr class="open" id="2861">
+    <td><a href="https://cplusplus.github.io/CWG/issues/2861.html">2861</a></td>
+    <td>tentatively ready</td>
+    <td><TT>dynamic_cast</TT> on bad pointer value</td>
+    <td align="center">Not resolved</td>
+  </tr>
+  <tr class="open" id="2862">
+    <td><a href="https://cplusplus.github.io/CWG/issues/2862.html">2862</a></td>
+    <td>tentatively ready</td>
+    <td>Unclear boundaries of template declarations</td>
+    <td align="center">Not resolved</td>
+  </tr>
+  <tr class="open" id="2863">
+    <td><a href="https://cplusplus.github.io/CWG/issues/2863.html">2863</a></td>
+    <td>tentatively ready</td>
+    <td>Unclear synchronization requirements for object lifetime rules</td>
+    <td align="center">Not resolved</td>
+  </tr>
+  <tr class="open" id="2864">
+    <td><a href="https://cplusplus.github.io/CWG/issues/2864.html">2864</a></td>
+    <td>tentatively ready</td>
+    <td>Narrowing floating-point conversions</td>
+    <td align="center">Not resolved</td>
+  </tr>
+  <tr class="open" id="2865">
+    <td><a href="https://cplusplus.github.io/CWG/issues/2865.html">2865</a></td>
+    <td>open</td>
+    <td>Regression on result of conditional operator</td>
+    <td align="center">Not resolved</td>
+  </tr>
+  <tr class="open" id="2866">
+    <td><a href="https://cplusplus.github.io/CWG/issues/2866.html">2866</a></td>
+    <td>open</td>
+    <td>Observing the effects of <TT>[[no_unique_address]]</TT></td>
+    <td align="center">Not resolved</td>
+  </tr>
+  <tr class="open" id="2867">
+    <td><a href="https://cplusplus.github.io/CWG/issues/2867.html">2867</a></td>
+    <td>open</td>
+    <td>Order of initialization for structured bindings</td>
+    <td align="center">Not resolved</td>
+  </tr>
+  <tr class="open" id="2868">
+    <td><a href="https://cplusplus.github.io/CWG/issues/2868.html">2868</a></td>
+    <td>open</td>
+    <td>Self-references in trivially copyable objects as function return values</td>
+    <td align="center">Not resolved</td>
+  </tr>
+  <tr class="open" id="2869">
+    <td><a href="https://cplusplus.github.io/CWG/issues/2869.html">2869</a></td>
+    <td>open</td>
+    <td><TT>this</TT> in local classes</td>
+    <td align="center">Not resolved</td>
+  </tr>
+  <tr class="open" id="2870">
+    <td><a href="https://cplusplus.github.io/CWG/issues/2870.html">2870</a></td>
+    <td>open</td>
+    <td>Combining absent <I>encoding-prefix</I>es</td>
+    <td align="center">Not resolved</td>
+  </tr>
+  <tr class="open" id="2871">
+    <td><a href="https://cplusplus.github.io/CWG/issues/2871.html">2871</a></td>
+    <td>tentatively ready</td>
+    <td>User-declared constructor templates inhibiting default constructors</td>
+    <td align="center">Not resolved</td>
+  </tr>
+  <tr class="open" id="2872">
+    <td><a href="https://cplusplus.github.io/CWG/issues/2872.html">2872</a></td>
+    <td>open</td>
+    <td>Linkage and unclear "can be referred to"</td>
+    <td align="center">Not resolved</td>
+  </tr>
+  <tr class="open" id="2873">
+    <td><a href="https://cplusplus.github.io/CWG/issues/2873.html">2873</a></td>
+    <td>open</td>
+    <td>Taking the address of a function involving template argument deduction</td>
+    <td align="center">Not resolved</td>
+  </tr>
+  <tr class="open" id="2874">
+    <td><a href="https://cplusplus.github.io/CWG/issues/2874.html">2874</a></td>
+    <td>open</td>
+    <td>Qualified declarations of partial specializations</td>
+    <td align="center">Not resolved</td>
+  </tr>
+  <tr class="open" id="2875">
+    <td><a href="https://cplusplus.github.io/CWG/issues/2875.html">2875</a></td>
+    <td>open</td>
+    <td>Missing support for round-tripping nullptr through indirection/address operators</td>
+    <td align="center">Not resolved</td>
+  </tr>
+  <tr class="open" id="2876">
+    <td><a href="https://cplusplus.github.io/CWG/issues/2876.html">2876</a></td>
+    <td>open</td>
+    <td>Disambiguation of <TT>T x = delete("text")</TT></td>
+    <td align="center">Not resolved</td>
+  </tr>
+  <tr class="open" id="2877">
+    <td><a href="https://cplusplus.github.io/CWG/issues/2877.html">2877</a></td>
+    <td>open</td>
+    <td>Type-only lookup for <I>using-enum-declarator</I></td>
+    <td align="center">Not resolved</td>
+  </tr>
+  <tr class="open" id="2878">
+    <td><a href="https://cplusplus.github.io/CWG/issues/2878.html">2878</a></td>
+    <td>open</td>
+    <td>C-style casts to reference types</td>
+    <td align="center">Not resolved</td>
+  </tr>
+  <tr class="open" id="2879">
+    <td><a href="https://cplusplus.github.io/CWG/issues/2879.html">2879</a></td>
+    <td>open</td>
+    <td>Undesired outcomes with <TT>const_cast</TT></td>
+    <td align="center">Not resolved</td>
+  </tr>
+  <tr class="open" id="2880">
+    <td><a href="https://cplusplus.github.io/CWG/issues/2880.html">2880</a></td>
+    <td>open</td>
+    <td>Accessibility check for destructor of incomplete class type</td>
+    <td align="center">Not resolved</td>
+  </tr>
+  <tr class="open" id="2881">
+    <td><a href="https://cplusplus.github.io/CWG/issues/2881.html">2881</a></td>
+    <td>open</td>
+    <td>Type restrictions for the explicit object parameter of a lambda</td>
+    <td align="center">Not resolved</td>
+  </tr>
+  <tr class="open" id="2882">
+    <td><a href="https://cplusplus.github.io/CWG/issues/2882.html">2882</a></td>
+    <td>open</td>
+    <td>Unclear treatment of conversion to <TT>void</TT></td>
+    <td align="center">Not resolved</td>
+  </tr>
+  <tr class="open" id="2883">
+    <td><a href="https://cplusplus.github.io/CWG/issues/2883.html">2883</a></td>
+    <td>open</td>
+    <td>Definition of "odr-usable" ignores lambda scopes</td>
+    <td align="center">Not resolved</td>
+  </tr>
+  <tr class="open" id="2884">
+    <td><a href="https://cplusplus.github.io/CWG/issues/2884.html">2884</a></td>
+    <td>open</td>
+    <td>Qualified declarations of partial specializations</td>
+    <td align="center">Not resolved</td>
+  </tr>
+  <tr class="open" id="2885">
+    <td><a href="https://cplusplus.github.io/CWG/issues/2885.html">2885</a></td>
+    <td>open</td>
+    <td>Non-eligible trivial default constructors</td>
+    <td align="center">Not resolved</td>
+  </tr>
+  <tr class="open" id="2886">
+    <td><a href="https://cplusplus.github.io/CWG/issues/2886.html">2886</a></td>
+    <td>open</td>
+    <td>Temporaries and trivial potentially-throwing special member functions</td>
+    <td align="center">Not resolved</td>
   </tr></table>
 
 </div>
diff --git a/clang/www/cxx_status.html b/clang/www/cxx_status.html
index c233171e63c8..d58c35b72c22 100755
--- a/clang/www/cxx_status.html
+++ b/clang/www/cxx_status.html
@@ -167,7 +167,7 @@ C++23, informally referred to as C++26.</p>
  <tr>
   <td>Disallow Binding a Returned Glvalue to a Temporary</td>
   <td><a href="https://wg21.link/P2748R5">P2748R5</a></td>
-  <td class="none" align="center">No</td>
+  <td class="full" align="center">Clang 19</td>
  </tr>
  <tr>
   <td>Clarifying rules for brace elision in aggregate initialization</td>
@@ -187,7 +187,7 @@ C++23, informally referred to as C++26.</p>
  <tr>
   <td>Trivial infinite loops are not Undefined Behavior</td>
   <td><a href="https://wg21.link/P2809R3">P2809R3</a> (<a href="#dr">DR</a>)</td>
-  <td class="none" align="center">No</td>
+  <td class="unreleased" align="center">Clang 19</td>
  </tr>
  <tr>
   <td>Erroneous behaviour for uninitialized reads</td>
diff --git a/clang/www/make_cxx_dr_status b/clang/www/make_cxx_dr_status
index 7c0cf77a1524..47c8b3bae4a1 100755
--- a/clang/www/make_cxx_dr_status
+++ b/clang/www/make_cxx_dr_status
@@ -5,7 +5,7 @@ latest_release = 18
 
 clang_www_dir = os.path.dirname(__file__)
 default_issue_list_path = os.path.join(clang_www_dir, 'cwg_index.html')
-issue_list_url = "https://www.open-std.org/jtc1/sc22/wg21/docs/cwg_index.html"
+issue_list_url = "https://raw.githubusercontent.com/cplusplus/CWG/gh-pages/issues/cwg_index.html"
 output = os.path.join(clang_www_dir, 'cxx_dr_status.html')
 dr_test_dir = os.path.join(clang_www_dir, '../test/CXX/drs')
 
@@ -138,10 +138,10 @@ def availability(issue):
 
   unresolved_status = ''
   proposed_resolution = ''
-  unresolved_status_match = re.search(r' (open|drafting|review)', status)
+  unresolved_status_match = re.search(r' (open|drafting|review|tentatively ready)', status)
   if unresolved_status_match:
     unresolved_status = unresolved_status_match.group(1)
-    proposed_resolution_match = re.search(r' (open|drafting|review) (\d{4}-\d{2}(?:-\d{2})?|P\d{4}R\d+)$', status)
+    proposed_resolution_match = re.search(r' (open|drafting|review|tentatively ready) (\d{4}-\d{2}(?:-\d{2})?|P\d{4}R\d+)$', status)
     if proposed_resolution_match is None:
       raise AvailabilityError('Issue {}: \'{}\' status should be followed by a paper number (P1234R5) or proposed resolution in YYYY-MM-DD format'.format(dr.issue, unresolved_status))
     proposed_resolution = proposed_resolution_match.group(2)
@@ -236,7 +236,7 @@ for dr in drs:
     avail = 'Extension'
     avail_style = ''
 
-  elif dr.status in ('open', 'drafting', 'review'):
+  elif dr.status in ('open', 'drafting', 'review', 'tentatively ready'):
     row_style = ' class="open"'
     try:
       avail, avail_style, unresolved_status = availability(dr.issue)
diff --git a/compiler-rt/lib/fuzzer/build.sh b/compiler-rt/lib/fuzzer/build.sh
index f7f329c0d19c..f58fd9557ce3 100755
--- a/compiler-rt/lib/fuzzer/build.sh
+++ b/compiler-rt/lib/fuzzer/build.sh
@@ -2,7 +2,7 @@
 LIBFUZZER_SRC_DIR=$(dirname $0)
 CXX="${CXX:-clang}"
 for f in $LIBFUZZER_SRC_DIR/*.cpp; do
-  $CXX -g -O2 -fno-omit-frame-pointer -std=c++14 $f -c &
+  $CXX -g -O2 -fno-omit-frame-pointer -std=c++17 $f -c &
 done
 wait
 rm -f libFuzzer.a
diff --git a/compiler-rt/lib/sanitizer_common/sanitizer_symbolizer_markup.cpp b/compiler-rt/lib/sanitizer_common/sanitizer_symbolizer_markup.cpp
index b2a1069a9a61..31d91ef3c739 100644
--- a/compiler-rt/lib/sanitizer_common/sanitizer_symbolizer_markup.cpp
+++ b/compiler-rt/lib/sanitizer_common/sanitizer_symbolizer_markup.cpp
@@ -28,7 +28,7 @@ void MarkupStackTracePrinter::RenderData(InternalScopedString *buffer,
                                          const char *format, const DataInfo *DI,
                                          const char *strip_path_prefix) {
   RenderContext(buffer);
-  buffer->AppendF(kFormatData, DI->start);
+  buffer->AppendF(kFormatData, reinterpret_cast<void *>(DI->start));
 }
 
 bool MarkupStackTracePrinter::RenderNeedsSymbolization(const char *format) {
@@ -43,12 +43,13 @@ void MarkupStackTracePrinter::RenderFrame(InternalScopedString *buffer,
                                           const char *strip_path_prefix) {
   CHECK(!RenderNeedsSymbolization(format));
   RenderContext(buffer);
-  buffer->AppendF(kFormatFrame, frame_no, address);
+  buffer->AppendF(kFormatFrame, frame_no, reinterpret_cast<void *>(address));
 }
 
 bool MarkupSymbolizerTool::SymbolizePC(uptr addr, SymbolizedStack *stack) {
   char buffer[kFormatFunctionMax];
-  internal_snprintf(buffer, sizeof(buffer), kFormatFunction, addr);
+  internal_snprintf(buffer, sizeof(buffer), kFormatFunction,
+                    reinterpret_cast<void *>(addr));
   stack->info.function = internal_strdup(buffer);
   return true;
 }
@@ -118,7 +119,8 @@ static void RenderMmaps(InternalScopedString *buffer,
     // module.base_address == dlpi_addr
     // range.beg == dlpi_addr + p_vaddr
     // relative address == p_vaddr == range.beg - module.base_address
-    buffer->AppendF(kFormatMmap, range.beg, range.end - range.beg, moduleId,
+    buffer->AppendF(kFormatMmap, reinterpret_cast<void *>(range.beg),
+                    range.end - range.beg, static_cast<int>(moduleId),
                     accessBuffer.data(), range.beg - module.base_address());
 
     buffer->Append("\n");
diff --git a/compiler-rt/lib/sanitizer_common/sanitizer_symbolizer_markup_constants.h b/compiler-rt/lib/sanitizer_common/sanitizer_symbolizer_markup_constants.h
index 83643504e128..a43661eaecf2 100644
--- a/compiler-rt/lib/sanitizer_common/sanitizer_symbolizer_markup_constants.h
+++ b/compiler-rt/lib/sanitizer_common/sanitizer_symbolizer_markup_constants.h
@@ -33,13 +33,13 @@ constexpr uptr kFormatFunctionMax = 64;  // More than big enough for 64-bit hex.
 constexpr const char *kFormatData = "{{{data:%p}}}";
 
 // One frame in a backtrace (printed on a line by itself).
-constexpr const char *kFormatFrame = "{{{bt:%u:%p}}}";
+constexpr const char *kFormatFrame = "{{{bt:%d:%p}}}";
 
 // Module contextual element.
-constexpr const char *kFormatModule = "{{{module:%d:%s:elf:%s}}}";
+constexpr const char *kFormatModule = "{{{module:%zu:%s:elf:%s}}}";
 
 // mmap for a module segment.
-constexpr const char *kFormatMmap = "{{{mmap:%p:0x%x:load:%d:%s:0x%x}}}";
+constexpr const char *kFormatMmap = "{{{mmap:%p:0x%zx:load:%d:%s:0x%zx}}}";
 
 // Dump trigger element.
 #define FORMAT_DUMPFILE "{{{dumpfile:%s:%s}}}"
diff --git a/compiler-rt/lib/scudo/standalone/allocator_config.def b/compiler-rt/lib/scudo/standalone/allocator_config.def
index 9691a007eed5..dcd130ac449a 100644
--- a/compiler-rt/lib/scudo/standalone/allocator_config.def
+++ b/compiler-rt/lib/scudo/standalone/allocator_config.def
@@ -89,6 +89,7 @@ PRIMARY_REQUIRED(const s32, MaxReleaseToOsIntervalMs)
 // Indicates support for offsetting the start of a region by a random number of
 // pages. This is only used if `EnableContiguousRegions` is enabled.
 PRIMARY_OPTIONAL(const bool, EnableRandomOffset, false)
+PRIMARY_OPTIONAL(const s32, DefaultReleaseToOsIntervalMs, INT32_MIN)
 
 // When `EnableContiguousRegions` is true, all regions will be be arranged in
 // adjacency. This will reduce the fragmentation caused by region allocations
@@ -118,6 +119,7 @@ SECONDARY_CACHE_OPTIONAL(const u32, DefaultMaxEntriesCount, 0)
 SECONDARY_CACHE_OPTIONAL(const uptr, DefaultMaxEntrySize, 0)
 SECONDARY_CACHE_OPTIONAL(const s32, MinReleaseToOsIntervalMs, INT32_MIN)
 SECONDARY_CACHE_OPTIONAL(const s32, MaxReleaseToOsIntervalMs, INT32_MAX)
+SECONDARY_CACHE_OPTIONAL(const s32, DefaultReleaseToOsIntervalMs, INT32_MIN)
 
 #undef SECONDARY_CACHE_OPTIONAL
 #undef SECONDARY_REQUIRED_TEMPLATE_TYPE
diff --git a/compiler-rt/lib/scudo/standalone/combined.h b/compiler-rt/lib/scudo/standalone/combined.h
index e7bc90cd0960..927513dea92d 100644
--- a/compiler-rt/lib/scudo/standalone/combined.h
+++ b/compiler-rt/lib/scudo/standalone/combined.h
@@ -173,6 +173,9 @@ public:
         static_cast<u32>(getFlags()->quarantine_max_chunk_size);
 
     Stats.init();
+    // TODO(chiahungduan): Given that we support setting the default value in
+    // the PrimaryConfig and CacheConfig, consider to deprecate the use of
+    // `release_to_os_interval_ms` flag.
     const s32 ReleaseToOsIntervalMs = getFlags()->release_to_os_interval_ms;
     Primary.init(ReleaseToOsIntervalMs);
     Secondary.init(&Stats, ReleaseToOsIntervalMs);
diff --git a/compiler-rt/lib/scudo/standalone/flags.inc b/compiler-rt/lib/scudo/standalone/flags.inc
index f5a2bab5057a..ff0c28e1db7c 100644
--- a/compiler-rt/lib/scudo/standalone/flags.inc
+++ b/compiler-rt/lib/scudo/standalone/flags.inc
@@ -42,7 +42,7 @@ SCUDO_FLAG(bool, may_return_null, true,
            "returning NULL in otherwise non-fatal error scenarios, eg: OOM, "
            "invalid allocation alignments, etc.")
 
-SCUDO_FLAG(int, release_to_os_interval_ms, SCUDO_ANDROID ? INT32_MIN : 5000,
+SCUDO_FLAG(int, release_to_os_interval_ms, 5000,
            "Interval (in milliseconds) at which to attempt release of unused "
            "memory to the OS. Negative values disable the feature.")
 
diff --git a/compiler-rt/lib/scudo/standalone/mem_map_fuchsia.cpp b/compiler-rt/lib/scudo/standalone/mem_map_fuchsia.cpp
index 28e5a11a37f2..fc793abf44cd 100644
--- a/compiler-rt/lib/scudo/standalone/mem_map_fuchsia.cpp
+++ b/compiler-rt/lib/scudo/standalone/mem_map_fuchsia.cpp
@@ -84,6 +84,13 @@ static zx_handle_t getPlaceholderVmo() {
   return Vmo;
 }
 
+// Checks if MAP_ALLOWNOMEM allows the given error code.
+static bool IsNoMemError(zx_status_t Status) {
+  // Note: _zx_vmar_map returns ZX_ERR_NO_RESOURCES if the VMAR does not contain
+  // a suitable free spot.
+  return Status == ZX_ERR_NO_MEMORY || Status == ZX_ERR_NO_RESOURCES;
+}
+
 MemMapFuchsia::MemMapFuchsia(uptr Base, uptr Capacity)
     : MapAddr(Base), WindowBase(Base), WindowSize(Capacity) {
   // Create the VMO.
@@ -101,9 +108,9 @@ bool MemMapFuchsia::mapImpl(UNUSED uptr Addr, uptr Size, const char *Name,
   // Create the VMO.
   zx_status_t Status = _zx_vmo_create(Size, 0, &Vmo);
   if (UNLIKELY(Status != ZX_OK)) {
-    if (Status != ZX_ERR_NO_MEMORY || !AllowNoMem)
-      dieOnError(Status, "zx_vmo_create", Size);
-    return false;
+    if (AllowNoMem && IsNoMemError(Status))
+      return false;
+    dieOnError(Status, "zx_vmo_create", Size);
   }
 
   if (Name != nullptr)
@@ -116,15 +123,15 @@ bool MemMapFuchsia::mapImpl(UNUSED uptr Addr, uptr Size, const char *Name,
   Status =
       _zx_vmar_map(_zx_vmar_root_self(), MapFlags, 0, Vmo, 0, Size, &MapAddr);
   if (UNLIKELY(Status != ZX_OK)) {
-    if (Status != ZX_ERR_NO_MEMORY || !AllowNoMem)
-      dieOnError(Status, "zx_vmar_map", Size);
-
-    Status = _zx_handle_close(Vmo);
-    CHECK_EQ(Status, ZX_OK);
+    if (AllowNoMem && IsNoMemError(Status)) {
+      Status = _zx_handle_close(Vmo);
+      CHECK_EQ(Status, ZX_OK);
 
-    MapAddr = 0;
-    Vmo = ZX_HANDLE_INVALID;
-    return false;
+      MapAddr = 0;
+      Vmo = ZX_HANDLE_INVALID;
+      return false;
+    }
+    dieOnError(Status, "zx_vmar_map", Size);
   }
 
   if (PreCommit) {
@@ -187,9 +194,9 @@ bool MemMapFuchsia::remapImpl(uptr Addr, uptr Size, const char *Name,
       _zx_vmar_map(_zx_vmar_root_self(), MapFlags, Addr - getRootVmarBase(),
                    Vmo, Addr - MapAddr, Size, &MappedAddr);
   if (UNLIKELY(Status != ZX_OK)) {
-    if (Status != ZX_ERR_NO_MEMORY || !AllowNoMem)
-      dieOnError(Status, "zx_vmar_map", Size);
-    return false;
+    if (AllowNoMem && IsNoMemError(Status))
+      return false;
+    dieOnError(Status, "zx_vmar_map", Size);
   }
   DCHECK_EQ(Addr, MappedAddr);
 
@@ -227,9 +234,9 @@ bool ReservedMemoryFuchsia::createImpl(UNUSED uptr Addr, uptr Size,
   zx_status_t Status = _zx_vmar_map(_zx_vmar_root_self(), ZX_VM_ALLOW_FAULTS, 0,
                                     getPlaceholderVmo(), 0, Size, &Base);
   if (UNLIKELY(Status != ZX_OK)) {
-    if (Status != ZX_ERR_NO_MEMORY || !AllowNoMem)
-      dieOnError(Status, "zx_vmar_map", Size);
-    return false;
+    if (AllowNoMem && IsNoMemError(Status))
+      return false;
+    dieOnError(Status, "zx_vmar_map", Size);
   }
 
   Capacity = Size;
diff --git a/compiler-rt/lib/scudo/standalone/primary32.h b/compiler-rt/lib/scudo/standalone/primary32.h
index 1d8a77b73e5c..ebfb8dfe0a31 100644
--- a/compiler-rt/lib/scudo/standalone/primary32.h
+++ b/compiler-rt/lib/scudo/standalone/primary32.h
@@ -88,6 +88,10 @@ public:
       Sci->MinRegionIndex = NumRegions;
       Sci->ReleaseInfo.LastReleaseAtNs = Time;
     }
+
+    // The default value in the primary config has the higher priority.
+    if (Config::getDefaultReleaseToOsIntervalMs() != INT32_MIN)
+      ReleaseToOsInterval = Config::getDefaultReleaseToOsIntervalMs();
     setOption(Option::ReleaseInterval, static_cast<sptr>(ReleaseToOsInterval));
   }
 
diff --git a/compiler-rt/lib/scudo/standalone/primary64.h b/compiler-rt/lib/scudo/standalone/primary64.h
index 61d57976ae43..bed2ccb8b992 100644
--- a/compiler-rt/lib/scudo/standalone/primary64.h
+++ b/compiler-rt/lib/scudo/standalone/primary64.h
@@ -147,6 +147,9 @@ public:
     for (uptr I = 0; I < NumClasses; I++)
       getRegionInfo(I)->FLLockCV.bindTestOnly(getRegionInfo(I)->FLLock);
 
+    // The default value in the primary config has the higher priority.
+    if (Config::getDefaultReleaseToOsIntervalMs() != INT32_MIN)
+      ReleaseToOsInterval = Config::getDefaultReleaseToOsIntervalMs();
     setOption(Option::ReleaseInterval, static_cast<sptr>(ReleaseToOsInterval));
   }
 
@@ -884,9 +887,10 @@ private:
         ScopedLock ML(Region->MMLock);
 
         const bool RegionIsExhausted = Region->Exhausted;
-        if (!RegionIsExhausted)
+        if (!RegionIsExhausted) {
           PopCount = populateFreeListAndPopBlocks(C, ClassId, Region, ToArray,
                                                   MaxBlockCount);
+        }
         ReportRegionExhausted = !RegionIsExhausted && Region->Exhausted;
 
         {
@@ -1019,7 +1023,6 @@ private:
                                           MAP_ALLOWNOMEM))) {
         Printf("Can't reserve pages for size class %zu.\n",
                getSizeByClassId(ClassId));
-        Region->Exhausted = true;
         return 0U;
       }
       initRegion(Region, ClassId,
diff --git a/compiler-rt/lib/scudo/standalone/secondary.h b/compiler-rt/lib/scudo/standalone/secondary.h
index 674af5071775..d8c9f5bcfcaf 100644
--- a/compiler-rt/lib/scudo/standalone/secondary.h
+++ b/compiler-rt/lib/scudo/standalone/secondary.h
@@ -209,6 +209,9 @@ public:
               static_cast<sptr>(Config::getDefaultMaxEntriesCount()));
     setOption(Option::MaxCacheEntrySize,
               static_cast<sptr>(Config::getDefaultMaxEntrySize()));
+    // The default value in the cache config has the higher priority.
+    if (Config::getDefaultReleaseToOsIntervalMs() != INT32_MIN)
+      ReleaseToOsInterval = Config::getDefaultReleaseToOsIntervalMs();
     setOption(Option::ReleaseInterval, static_cast<sptr>(ReleaseToOsInterval));
   }
 
diff --git a/compiler-rt/lib/scudo/standalone/wrappers_c.inc b/compiler-rt/lib/scudo/standalone/wrappers_c.inc
index 21d5b7add512..59f3fb0962f8 100644
--- a/compiler-rt/lib/scudo/standalone/wrappers_c.inc
+++ b/compiler-rt/lib/scudo/standalone/wrappers_c.inc
@@ -252,13 +252,11 @@ INTERFACE WEAK int SCUDO_PREFIX(mallopt)(int param, int value) {
       // introduced by interval transition.
       SCUDO_ALLOCATOR.releaseToOS(scudo::ReleaseToOS::Force);
 
-      if (value == 0) {
-        // Will set the release values to their minimum values.
-        value = INT32_MIN;
-      } else {
-        // Will set the release values to their maximum values.
+      // The values allowed on Android are {-1, 0, 1}. "1" means the longest
+      // interval.
+      CHECK(value >= -1 && value <= 1);
+      if (value == 1)
         value = INT32_MAX;
-      }
     }
 
     SCUDO_ALLOCATOR.setOption(scudo::Option::ReleaseInterval,
diff --git a/compiler-rt/test/CMakeLists.txt b/compiler-rt/test/CMakeLists.txt
index edc007aaf477..8805cc8f798f 100644
--- a/compiler-rt/test/CMakeLists.txt
+++ b/compiler-rt/test/CMakeLists.txt
@@ -92,6 +92,9 @@ if(COMPILER_RT_CAN_EXECUTE_TESTS)
   if(COMPILER_RT_BUILD_PROFILE AND COMPILER_RT_HAS_PROFILE)
     compiler_rt_test_runtime(profile)
   endif()
+  if(COMPILER_RT_BUILD_CTX_PROFILE)
+    compiler_rt_test_runtime(ctx_profile)
+  endif()
   if(COMPILER_RT_BUILD_MEMPROF)
     compiler_rt_test_runtime(memprof)
   endif()
diff --git a/compiler-rt/test/asan/TestCases/Darwin/odr-lto.cpp b/compiler-rt/test/asan/TestCases/Darwin/odr-lto.cpp
index 90c16776a63b..8d400800fe93 100644
--- a/compiler-rt/test/asan/TestCases/Darwin/odr-lto.cpp
+++ b/compiler-rt/test/asan/TestCases/Darwin/odr-lto.cpp
@@ -5,7 +5,7 @@
 
 // RUN: %clangxx_asan -DPART=0 -c %s -o %t-1.o -flto -mllvm -asan-use-private-alias
 // RUN: %clangxx_asan -DPART=1 -c %s -o %t-2.o -flto -mllvm -asan-use-private-alias
-// RUN: %clangxx_asan_lto %t-1.o %t-2.o -o %t -flto -mlinker-version=133
+// RUN: %clangxx_asan_lto %t-1.o %t-2.o -o %t -flto
 // RUN: %run %t 2>&1 | FileCheck %s
 
 #include <stdio.h>
diff --git a/compiler-rt/test/ctx_profile/CMakeLists.txt b/compiler-rt/test/ctx_profile/CMakeLists.txt
new file mode 100644
index 000000000000..23c6fb16ed1f
--- /dev/null
+++ b/compiler-rt/test/ctx_profile/CMakeLists.txt
@@ -0,0 +1,21 @@
+set(CTX_PROFILE_LIT_SOURCE_DIR ${CMAKE_CURRENT_SOURCE_DIR})
+
+set(CTX_PROFILE_TESTSUITES)
+
+# Add unit tests.
+if(COMPILER_RT_INCLUDE_TESTS)
+  foreach(arch ${CTX_PROFILE_SUPPORTED_ARCH})
+    string(TOUPPER ${arch} ARCH_UPPER_CASE)
+    set(CONFIG_NAME ${ARCH_UPPER_CASE}${OS_NAME}Config)
+    configure_lit_site_cfg(
+      ${CMAKE_CURRENT_SOURCE_DIR}/Unit/lit.site.cfg.py.in
+      ${CMAKE_CURRENT_BINARY_DIR}/Unit/${CONFIG_NAME}/lit.site.cfg.py)
+    list(APPEND CTX_PROFILE_TEST_DEPS CtxProfileUnitTests)
+    list(APPEND CTX_PROFILE_TESTSUITES ${CMAKE_CURRENT_BINARY_DIR}/Unit/${CONFIG_NAME})
+  endforeach()
+endif()
+
+add_lit_testsuite(check-ctx_profile "Running the Contextual Profiler tests"
+  ${CTX_PROFILE_TESTSUITES}
+  DEPENDS ${CTX_PROFILE_TEST_DEPS})
+set_target_properties(check-ctx_profile PROPERTIES FOLDER "Compiler-RT Misc")
diff --git a/compiler-rt/test/ctx_profile/Unit/lit.site.cfg.py.in b/compiler-rt/test/ctx_profile/Unit/lit.site.cfg.py.in
new file mode 100644
index 000000000000..3fa9a7a2780e
--- /dev/null
+++ b/compiler-rt/test/ctx_profile/Unit/lit.site.cfg.py.in
@@ -0,0 +1,28 @@
+@LIT_SITE_CFG_IN_HEADER@
+
+import os
+import platform
+import re
+import shlex
+
+# Load common config for all compiler-rt unit tests.
+lit_config.load_config(config, "@COMPILER_RT_BINARY_DIR@/unittests/lit.common.unit.configured")
+
+# Setup config name.
+config.name = 'CtxProfile-Unit'
+config.target_arch = "@arch@"
+assert config.target_arch == 'x86_64'
+
+config.test_exec_root = os.path.join("@COMPILER_RT_BINARY_DIR@",
+                                     "lib", "ctx_profile", "tests")
+
+config.test_source_root = config.test_exec_root
+
+# When LLVM_ENABLE_PER_TARGET_RUNTIME_DIR=on, the initial value of
+# config.compiler_rt_libdir (COMPILER_RT_RESOLVED_LIBRARY_OUTPUT_DIR) has the
+# host triple as the trailing path component. The value is incorrect for i386
+# tests on x86_64 hosts and vice versa. But, since only x86_64 is enabled as
+# target, and we don't support different environments for building and,
+# respectively, running tests, we we only need to fix up the x86_64 case.
+if config.enable_per_target_runtime_dir and config.target_arch != config.host_arch:
+    config.compiler_rt_libdir = re.sub(r'/i386(?=-[^/]+$)', '/x86_64', config.compiler_rt_libdir)
diff --git a/compiler-rt/test/memprof/CMakeLists.txt b/compiler-rt/test/memprof/CMakeLists.txt
index 3f0ba3812485..fa6a4cd5f0b7 100644
--- a/compiler-rt/test/memprof/CMakeLists.txt
+++ b/compiler-rt/test/memprof/CMakeLists.txt
@@ -43,6 +43,19 @@ foreach(arch ${MEMPROF_TEST_ARCH})
     ${CMAKE_CURRENT_BINARY_DIR}/${CONFIG_NAME})
 endforeach()
 
+# Add unit tests.
+if(COMPILER_RT_INCLUDE_TESTS)
+  foreach(arch ${MEMPROF_TEST_ARCH})
+    string(TOUPPER ${arch} ARCH_UPPER_CASE)
+    set(CONFIG_NAME ${ARCH_UPPER_CASE}${OS_NAME}Config)
+    configure_lit_site_cfg(
+      ${CMAKE_CURRENT_SOURCE_DIR}/Unit/lit.site.cfg.py.in
+      ${CMAKE_CURRENT_BINARY_DIR}/Unit/${CONFIG_NAME}/lit.site.cfg.py)
+    list(APPEND MEMPROF_TEST_DEPS MemProfUnitTests)
+    list(APPEND MEMPROF_TESTSUITES ${CMAKE_CURRENT_BINARY_DIR}/Unit/${CONFIG_NAME})
+  endforeach()
+endif()
+
 add_lit_testsuite(check-memprof "Running the MemProfiler tests"
   ${MEMPROF_TESTSUITES}
   DEPENDS ${MEMPROF_TEST_DEPS})
diff --git a/compiler-rt/test/memprof/Unit/lit.site.cfg.py.in b/compiler-rt/test/memprof/Unit/lit.site.cfg.py.in
new file mode 100644
index 000000000000..1e2442a1487a
--- /dev/null
+++ b/compiler-rt/test/memprof/Unit/lit.site.cfg.py.in
@@ -0,0 +1,31 @@
+@LIT_SITE_CFG_IN_HEADER@
+
+import os
+import platform
+import re
+import shlex
+
+# Load common config for all compiler-rt unit tests.
+lit_config.load_config(config, "@COMPILER_RT_BINARY_DIR@/unittests/lit.common.unit.configured")
+
+# Setup config name.
+config.name = 'MemProfiler-Unit'
+config.target_arch = "@arch@"
+assert config.target_arch == 'x86_64'
+
+config.test_exec_root = os.path.join("@COMPILER_RT_BINARY_DIR@",
+                                     "lib", "memprof", "tests")
+
+config.test_source_root = config.test_exec_root
+
+# When LLVM_ENABLE_PER_TARGET_RUNTIME_DIR=on, the initial value of
+# config.compiler_rt_libdir (COMPILER_RT_RESOLVED_LIBRARY_OUTPUT_DIR) has the
+# host triple as the trailing path component. The value is incorrect for i386
+# tests on x86_64 hosts and vice versa. But, since only x86_64 is enabled as
+# target, and we don't support different environments for building and,
+# respectively, running tests, we we only need to fix up the x86_64 case.
+if config.enable_per_target_runtime_dir and config.target_arch != config.host_arch:
+    config.compiler_rt_libdir = re.sub(r'/i386(?=-[^/]+$)', '/x86_64', config.compiler_rt_libdir)
+
+if not config.parallelism_group:
+  config.parallelism_group = 'shadow-memory'
diff --git a/flang/cmake/modules/AddFlangOffloadRuntime.cmake b/flang/cmake/modules/AddFlangOffloadRuntime.cmake
index e34d3851187a..0af12c8cfd54 100644
--- a/flang/cmake/modules/AddFlangOffloadRuntime.cmake
+++ b/flang/cmake/modules/AddFlangOffloadRuntime.cmake
@@ -2,6 +2,10 @@ option(FLANG_EXPERIMENTAL_CUDA_RUNTIME
   "Compile Fortran runtime as CUDA sources (experimental)" OFF
   )
 
+option(FLANG_CUDA_RUNTIME_PTX_WITHOUT_GLOBAL_VARS
+  "Do not compile global variables' definitions when producing PTX library" OFF
+  )
+
 set(FLANG_LIBCUDACXX_PATH "" CACHE PATH "Path to libcu++ package installation")
 
 set(FLANG_EXPERIMENTAL_OMP_OFFLOAD_BUILD "off" CACHE STRING
@@ -56,6 +60,11 @@ macro(enable_cuda_compilation name files)
     # Add an OBJECT library consisting of CUDA PTX.
     llvm_add_library(${name}PTX OBJECT PARTIAL_SOURCES_INTENDED ${files})
     set_property(TARGET obj.${name}PTX PROPERTY CUDA_PTX_COMPILATION ON)
+    if (FLANG_CUDA_RUNTIME_PTX_WITHOUT_GLOBAL_VARS)
+      target_compile_definitions(obj.${name}PTX
+        PRIVATE FLANG_RUNTIME_NO_GLOBAL_VAR_DEFS
+        )
+    endif()
   endif()
 endmacro()
 
diff --git a/flang/docs/FlangDriver.md b/flang/docs/FlangDriver.md
index ac120b4ff09b..351595ac0afd 100644
--- a/flang/docs/FlangDriver.md
+++ b/flang/docs/FlangDriver.md
@@ -179,46 +179,20 @@ like this:
 
 ```
 $ flang -v -o example example.o
-"/usr/bin/ld" [...] example.o [...] "--whole-archive" "-lFortran_main"
-"--no-whole-archive" "-lFortranRuntime" "-lFortranDecimal" [...]
+"/usr/bin/ld" [...] example.o [...] "-lFortranRuntime" "-lFortranDecimal" [...]
 ```
 
 The automatically added libraries are:
 
-* `Fortran_main`: Provides the main entry point `main` that then invokes
-  `_QQmain` with the Fortran program unit.  This library has a dependency to
-  the `FortranRuntime` library.
 * `FortranRuntime`: Provides most of the Flang runtime library.
 * `FortranDecimal`: Provides operations for decimal numbers.
 
-The default is that, when using Flang as the linker, one of the Fortran
-translation units provides the program unit and therefore it is assumed that
-Fortran is the main code part (calling into C/C++ routines via `BIND (C)`
-interfaces).  When composing the linker commandline, Flang uses
-`--whole-archive` and `--no-whole-archive` (Windows: `/WHOLEARCHIVE:`,
-Darwin & AIX: *not implemented yet*) to make sure that all for `Fortran_main`
-is processed by the linker.  This is done to issue a proper error message when
-multiple definitions of `main` occur.  This happens, for instance, when linking
-a code that has a Fortran program unit with a C/C++ code that also defines a
-`main` function.  A user may be required to explicitly provide the C++ runtime
-libraries at link time (e.g., via `-lstdc++` for STL)
-
 If the code is C/C++ based and invokes Fortran routines, one can either use Clang
 or Flang as the linker driver.  If Clang is used, it will automatically all
 required runtime libraries needed by C++ (e.g., for STL) to the linker invocation.
 In this case, one has to explicitly provide the Fortran runtime libraries
-`FortranRuntime` and/or `FortranDecimal`.  An alternative is to use Flang to link
-and use the `-fno-fortran-main` flag.  This flag removes
-`Fortran_main` from the linker stage and hence requires one of the C/C++
-translation units to provide a definition of the `main` function. In this case,
-it may be required to explicitly supply C++ runtime libraries as mentioned above.
-
-When creating shared or static libraries using Flang with `-shared` or `-static`
-flag, Fortran_main is automatically removed from the linker stage (i.e.,
-`-fno-fortran-main` is on by default).  It is assumed that when creating a
-static or shared library, the generated library does not need a `main`
-function, as a final link stage will occur that will provide the `Fortran_main`
-library when creating the final executable.
+`FortranRuntime` and/or `FortranDecimal`.  An alternative is to use Flang to link.
+In this case, it may be required to explicitly supply C++ runtime libraries.
 
 On Darwin, the logical root where the system libraries are located (sysroot)
 must be specified. This can be done with the CMake build flag `DEFAULT_SYSROOT`
diff --git a/flang/docs/Preprocessing.md b/flang/docs/Preprocessing.md
index 3c523472f39b..0b70d857833c 100644
--- a/flang/docs/Preprocessing.md
+++ b/flang/docs/Preprocessing.md
@@ -93,6 +93,9 @@ local:
 * If a `#define` or `#undef` directive appears among continuation
   lines, it may or may not affect text in the continued statement that
   appeared before the directive.
+* A backslash at the end of a free form source line is a continuation
+  marker, with no space skipping or special handling of a leading `&`
+  on the next line.
 
 ## Behavior that few compilers properly support (or none), but should:
 
diff --git a/flang/include/flang/Common/visit.h b/flang/include/flang/Common/visit.h
index 4d0897301e01..d867338be7e0 100644
--- a/flang/include/flang/Common/visit.h
+++ b/flang/include/flang/Common/visit.h
@@ -40,11 +40,17 @@ inline RT_API_ATTRS RESULT Log2VisitHelper(
       return visitor(std::get<(LOW + N)>(std::forward<VARIANT>(u))...); \
     }
       VISIT_CASE_N(1)
+      [[fallthrough]];
       VISIT_CASE_N(2)
+      [[fallthrough]];
       VISIT_CASE_N(3)
+      [[fallthrough]];
       VISIT_CASE_N(4)
+      [[fallthrough]];
       VISIT_CASE_N(5)
+      [[fallthrough]];
       VISIT_CASE_N(6)
+      [[fallthrough]];
       VISIT_CASE_N(7)
 #undef VISIT_CASE_N
     }
@@ -82,7 +88,7 @@ inline RT_API_ATTRS auto visit(VISITOR &&visitor, VARIANT &&...u)
 // Some versions of clang have bugs that cause compilation to hang
 // on these templates.  MSVC and older GCC versions may work but are
 // not well tested.  So enable only for GCC 9 and better.
-#if __GNUC__ < 9
+#if __GNUC__ < 9 && !defined(__clang__)
 #define FLANG_USE_STD_VISIT
 #endif
 
diff --git a/flang/include/flang/Evaluate/call.h b/flang/include/flang/Evaluate/call.h
index 3d766bc08e58..7531d8a81e80 100644
--- a/flang/include/flang/Evaluate/call.h
+++ b/flang/include/flang/Evaluate/call.h
@@ -287,15 +287,18 @@ public:
       : ProcedureRef{std::move(p), std::move(a)} {}
 
   std::optional<DynamicType> GetType() const {
-    if (auto type{proc_.GetType()}) {
+    if constexpr (IsLengthlessIntrinsicType<A>) {
+      return A::GetType();
+    } else if (auto type{proc_.GetType()}) {
       // TODO: Non constant explicit length parameters of PDTs result should
       // likely be dropped too. This is not as easy as for characters since some
       // long lived DerivedTypeSpec pointer would need to be created here. It is
       // not clear if this is causing any issue so far since the storage size of
       // PDTs is independent of length parameters.
       return type->DropNonConstantCharacterLength();
+    } else {
+      return std::nullopt;
     }
-    return std::nullopt;
   }
 };
 } // namespace Fortran::evaluate
diff --git a/flang/include/flang/Lower/Allocatable.h b/flang/include/flang/Lower/Allocatable.h
index d3c16de377c1..e8738f0407e7 100644
--- a/flang/include/flang/Lower/Allocatable.h
+++ b/flang/include/flang/Lower/Allocatable.h
@@ -55,12 +55,14 @@ void genDeallocateStmt(AbstractConverter &converter,
 
 void genDeallocateBox(AbstractConverter &converter,
                       const fir::MutableBoxValue &box, mlir::Location loc,
+                      const Fortran::semantics::Symbol *sym = nullptr,
                       mlir::Value declaredTypeDesc = {});
 
 /// Deallocate an allocatable if it is allocated at the end of its lifetime.
 void genDeallocateIfAllocated(AbstractConverter &converter,
                               const fir::MutableBoxValue &box,
-                              mlir::Location loc);
+                              mlir::Location loc,
+                              const Fortran::semantics::Symbol *sym = nullptr);
 
 /// Create a MutableBoxValue for an allocatable or pointer entity.
 /// If the variables is a local variable that is not a dummy, it will be
diff --git a/flang/include/flang/Lower/Mangler.h b/flang/include/flang/Lower/Mangler.h
index 41939abe29e5..99da96b0d6ba 100644
--- a/flang/include/flang/Lower/Mangler.h
+++ b/flang/include/flang/Lower/Mangler.h
@@ -90,7 +90,7 @@ inline std::string mangleArrayLiteral(
   return mangleArrayLiteral(x.values().size() * sizeof(x.values()[0]),
                             x.shape(), Fortran::common::TypeCategory::Derived,
                             /*kind=*/0, /*charLen=*/-1,
-                            eleTy.cast<fir::RecordType>().getName());
+                            mlir::cast<fir::RecordType>(eleTy).getName());
 }
 
 /// Return the compiler-generated name of a static namelist variable descriptor.
diff --git a/flang/include/flang/Optimizer/Analysis/TBAAForest.h b/flang/include/flang/Optimizer/Analysis/TBAAForest.h
index b69e50bbe05c..619ed4939c51 100644
--- a/flang/include/flang/Optimizer/Analysis/TBAAForest.h
+++ b/flang/include/flang/Optimizer/Analysis/TBAAForest.h
@@ -88,7 +88,7 @@ public:
     // name must be used so that we add to the tbaa tree added in the FIR pass
     mlir::Attribute attr = func->getAttr(getInternalFuncNameAttrName());
     if (attr) {
-      return getFuncTree(attr.cast<mlir::StringAttr>());
+      return getFuncTree(mlir::cast<mlir::StringAttr>(attr));
     }
     return getFuncTree(func.getSymNameAttr());
   }
diff --git a/flang/include/flang/Optimizer/Builder/BoxValue.h b/flang/include/flang/Optimizer/Builder/BoxValue.h
index 2fed2d48a7a0..5c7e89dbc08f 100644
--- a/flang/include/flang/Optimizer/Builder/BoxValue.h
+++ b/flang/include/flang/Optimizer/Builder/BoxValue.h
@@ -78,7 +78,7 @@ class CharBoxValue : public AbstractBox {
 public:
   CharBoxValue(mlir::Value addr, mlir::Value len)
       : AbstractBox{addr}, len{len} {
-    if (addr && addr.getType().template isa<fir::BoxCharType>())
+    if (addr && mlir::isa<fir::BoxCharType>(addr.getType()))
       fir::emitFatalError(addr.getLoc(),
                           "BoxChar should not be in CharBoxValue");
   }
@@ -221,7 +221,7 @@ public:
     auto type = getAddr().getType();
     if (auto pointedTy = fir::dyn_cast_ptrEleTy(type))
       type = pointedTy;
-    return type.cast<fir::BaseBoxType>();
+    return mlir::cast<fir::BaseBoxType>(type);
   }
   /// Return the part of the address type after memory and box types. That is
   /// the element type, maybe wrapped in a fir.array type.
@@ -243,22 +243,22 @@ public:
   /// Get the scalar type related to the described entity
   mlir::Type getEleTy() const {
     auto type = getBaseTy();
-    if (auto seqTy = type.dyn_cast<fir::SequenceType>())
+    if (auto seqTy = mlir::dyn_cast<fir::SequenceType>(type))
       return seqTy.getEleTy();
     return type;
   }
 
   /// Is the entity an array or an assumed rank ?
-  bool hasRank() const { return getBaseTy().isa<fir::SequenceType>(); }
+  bool hasRank() const { return mlir::isa<fir::SequenceType>(getBaseTy()); }
   /// Is this an assumed rank ?
   bool hasAssumedRank() const {
-    auto seqTy = getBaseTy().dyn_cast<fir::SequenceType>();
+    auto seqTy = mlir::dyn_cast<fir::SequenceType>(getBaseTy());
     return seqTy && seqTy.hasUnknownShape();
   }
   /// Returns the rank of the entity. Beware that zero will be returned for
   /// both scalars and assumed rank.
   unsigned rank() const {
-    if (auto seqTy = getBaseTy().dyn_cast<fir::SequenceType>())
+    if (auto seqTy = mlir::dyn_cast<fir::SequenceType>(getBaseTy()))
       return seqTy.getDimension();
     return 0;
   }
@@ -267,7 +267,7 @@ public:
   bool isCharacter() const { return fir::isa_char(getEleTy()); }
 
   /// Is this a derived type entity ?
-  bool isDerived() const { return getEleTy().isa<fir::RecordType>(); }
+  bool isDerived() const { return mlir::isa<fir::RecordType>(getEleTy()); }
 
   bool isDerivedWithLenParameters() const {
     return fir::isRecordWithTypeParameters(getEleTy());
@@ -377,11 +377,11 @@ public:
   }
   /// Is this a Fortran pointer ?
   bool isPointer() const {
-    return getBoxTy().getEleTy().isa<fir::PointerType>();
+    return mlir::isa<fir::PointerType>(getBoxTy().getEleTy());
   }
   /// Is this an allocatable ?
   bool isAllocatable() const {
-    return getBoxTy().getEleTy().isa<fir::HeapType>();
+    return mlir::isa<fir::HeapType>(getBoxTy().getEleTy());
   }
   // Replace the fir.ref<fir.box>, keeping any non-deferred parameters.
   MutableBoxValue clone(mlir::Value newBox) const {
@@ -488,7 +488,7 @@ public:
     if (const auto *b = getUnboxed()) {
       if (*b) {
         auto type = b->getType();
-        if (type.template isa<fir::BoxCharType>())
+        if (mlir::isa<fir::BoxCharType>(type))
           fir::emitFatalError(b->getLoc(), "BoxChar should be unboxed");
         type = fir::unwrapSequenceType(fir::unwrapRefType(type));
         if (fir::isa_char(type))
diff --git a/flang/include/flang/Optimizer/Builder/Factory.h b/flang/include/flang/Optimizer/Builder/Factory.h
index ec294d26ac96..4e5c52ac44e0 100644
--- a/flang/include/flang/Optimizer/Builder/Factory.h
+++ b/flang/include/flang/Optimizer/Builder/Factory.h
@@ -43,9 +43,9 @@ template <typename B>
 void genCharacterCopy(mlir::Value src, mlir::Value srcLen, mlir::Value dst,
                       mlir::Value dstLen, B &builder, mlir::Location loc) {
   auto srcTy =
-      fir::dyn_cast_ptrEleTy(src.getType()).template cast<fir::CharacterType>();
+      mlir::cast<fir::CharacterType>(fir::dyn_cast_ptrEleTy(src.getType()));
   auto dstTy =
-      fir::dyn_cast_ptrEleTy(dst.getType()).template cast<fir::CharacterType>();
+      mlir::cast<fir::CharacterType>(fir::dyn_cast_ptrEleTy(dst.getType()));
   if (!srcLen && !dstLen && srcTy.getFKind() == dstTy.getFKind() &&
       srcTy.getLen() == dstTy.getLen()) {
     // same size, so just use load and store
@@ -61,8 +61,8 @@ void genCharacterCopy(mlir::Value src, mlir::Value srcLen, mlir::Value dst,
         fir::CharacterType::getSingleton(ty.getContext(), ty.getFKind())));
   };
   auto toEleTy = [&](fir::ReferenceType ty) {
-    auto seqTy = ty.getEleTy().cast<fir::SequenceType>();
-    return seqTy.getEleTy().cast<fir::CharacterType>();
+    auto seqTy = mlir::cast<fir::SequenceType>(ty.getEleTy());
+    return mlir::cast<fir::CharacterType>(seqTy.getEleTy());
   };
   auto toCoorTy = [&](fir::ReferenceType ty) {
     return fir::ReferenceType::get(toEleTy(ty));
@@ -190,8 +190,8 @@ originateIndices(mlir::Location loc, B &builder, mlir::Type memTy,
   if (origins.empty()) {
     assert(!shapeVal || mlir::isa<fir::ShapeOp>(shapeVal.getDefiningOp()));
     auto ty = fir::dyn_cast_ptrOrBoxEleTy(memTy);
-    assert(ty && ty.isa<fir::SequenceType>());
-    auto seqTy = ty.cast<fir::SequenceType>();
+    assert(ty && mlir::isa<fir::SequenceType>(ty));
+    auto seqTy = mlir::cast<fir::SequenceType>(ty);
     auto one = builder.template create<mlir::arith::ConstantIndexOp>(loc, 1);
     const auto dimension = seqTy.getDimension();
     if (shapeVal) {
diff --git a/flang/include/flang/Optimizer/Builder/HLFIRTools.h b/flang/include/flang/Optimizer/Builder/HLFIRTools.h
index 035035601e2f..6c36f7e84db6 100644
--- a/flang/include/flang/Optimizer/Builder/HLFIRTools.h
+++ b/flang/include/flang/Optimizer/Builder/HLFIRTools.h
@@ -77,12 +77,12 @@ public:
   /// Return the rank of this entity or -1 if it is an assumed rank.
   int getRank() const {
     mlir::Type type = fir::unwrapPassByRefType(fir::unwrapRefType(getType()));
-    if (auto seqTy = type.dyn_cast<fir::SequenceType>()) {
+    if (auto seqTy = mlir::dyn_cast<fir::SequenceType>(type)) {
       if (seqTy.hasUnknownShape())
         return -1;
       return seqTy.getDimension();
     }
-    if (auto exprType = type.dyn_cast<hlfir::ExprType>())
+    if (auto exprType = mlir::dyn_cast<hlfir::ExprType>(type))
       return exprType.getRank();
     return 0;
   }
@@ -99,17 +99,17 @@ public:
 
   bool hasLengthParameters() const {
     mlir::Type eleTy = getFortranElementType();
-    return eleTy.isa<fir::CharacterType>() ||
+    return mlir::isa<fir::CharacterType>(eleTy) ||
            fir::isRecordWithTypeParameters(eleTy);
   }
 
   bool isCharacter() const {
-    return getFortranElementType().isa<fir::CharacterType>();
+    return mlir::isa<fir::CharacterType>(getFortranElementType());
   }
 
   bool hasIntrinsicType() const {
     mlir::Type eleTy = getFortranElementType();
-    return fir::isa_trivial(eleTy) || eleTy.isa<fir::CharacterType>();
+    return fir::isa_trivial(eleTy) || mlir::isa<fir::CharacterType>(eleTy);
   }
 
   bool isDerivedWithLengthParameters() const {
@@ -124,8 +124,8 @@ public:
     if (auto varIface = getIfVariableInterface()) {
       if (auto shape = varIface.getShape()) {
         auto shapeTy = shape.getType();
-        return shapeTy.isa<fir::ShiftType>() ||
-               shapeTy.isa<fir::ShapeShiftType>();
+        return mlir::isa<fir::ShiftType>(shapeTy) ||
+               mlir::isa<fir::ShapeShiftType>(shapeTy);
       }
       return false;
     }
diff --git a/flang/include/flang/Optimizer/Builder/IntrinsicCall.h b/flang/include/flang/Optimizer/Builder/IntrinsicCall.h
index 6927488517e6..b7d060926761 100644
--- a/flang/include/flang/Optimizer/Builder/IntrinsicCall.h
+++ b/flang/include/flang/Optimizer/Builder/IntrinsicCall.h
@@ -335,6 +335,8 @@ struct IntrinsicLibrary {
   mlir::Value genSelectedRealKind(mlir::Type, llvm::ArrayRef<mlir::Value>);
   mlir::Value genSetExponent(mlir::Type resultType,
                              llvm::ArrayRef<mlir::Value> args);
+  fir::ExtendedValue genShape(mlir::Type resultType,
+                              llvm::ArrayRef<fir::ExtendedValue>);
   template <typename Shift>
   mlir::Value genShift(mlir::Type resultType, llvm::ArrayRef<mlir::Value>);
   mlir::Value genShiftA(mlir::Type resultType, llvm::ArrayRef<mlir::Value>);
@@ -661,8 +663,8 @@ static inline mlir::FunctionType genFuncType(mlir::MLIRContext *context,
 //===----------------------------------------------------------------------===//
 static inline mlir::Type getConvertedElementType(mlir::MLIRContext *context,
                                                  mlir::Type eleTy) {
-  if (eleTy.isa<mlir::IntegerType>() && !eleTy.isSignlessInteger()) {
-    const auto intTy{eleTy.dyn_cast<mlir::IntegerType>()};
+  if (mlir::isa<mlir::IntegerType>(eleTy) && !eleTy.isSignlessInteger()) {
+    const auto intTy{mlir::dyn_cast<mlir::IntegerType>(eleTy)};
     auto newEleTy{mlir::IntegerType::get(context, intTy.getWidth())};
     return newEleTy;
   }
diff --git a/flang/include/flang/Optimizer/Builder/PPCIntrinsicCall.h b/flang/include/flang/Optimizer/Builder/PPCIntrinsicCall.h
index 1e87bf0f6ad1..a7c4c075d818 100644
--- a/flang/include/flang/Optimizer/Builder/PPCIntrinsicCall.h
+++ b/flang/include/flang/Optimizer/Builder/PPCIntrinsicCall.h
@@ -180,10 +180,10 @@ struct VecTypeInfo {
 // Returns a VecTypeInfo with element type and length of given fir vector type.
 // Preserves signness of fir vector type if element type of integer.
 static inline VecTypeInfo getVecTypeFromFirType(mlir::Type firTy) {
-  assert(firTy.isa<fir::VectorType>());
+  assert(mlir::isa<fir::VectorType>(firTy));
   VecTypeInfo vecTyInfo;
-  vecTyInfo.eleTy = firTy.dyn_cast<fir::VectorType>().getEleTy();
-  vecTyInfo.len = firTy.dyn_cast<fir::VectorType>().getLen();
+  vecTyInfo.eleTy = mlir::dyn_cast<fir::VectorType>(firTy).getEleTy();
+  vecTyInfo.len = mlir::dyn_cast<fir::VectorType>(firTy).getLen();
   return vecTyInfo;
 }
 
diff --git a/flang/include/flang/Optimizer/Builder/Runtime/EnvironmentDefaults.h b/flang/include/flang/Optimizer/Builder/Runtime/EnvironmentDefaults.h
index 18a24bad3960..216d3bcec137 100755
--- a/flang/include/flang/Optimizer/Builder/Runtime/EnvironmentDefaults.h
+++ b/flang/include/flang/Optimizer/Builder/Runtime/EnvironmentDefaults.h
@@ -22,6 +22,7 @@
 
 namespace fir {
 class FirOpBuilder;
+class GlobalOp;
 } // namespace fir
 
 namespace mlir {
@@ -37,7 +38,7 @@ namespace fir::runtime {
 /// Create the list of environment variable defaults for the runtime to set. The
 /// form of the generated list is defined in the runtime header file
 /// environment-default-list.h
-void genEnvironmentDefaults(
+fir::GlobalOp genEnvironmentDefaults(
     fir::FirOpBuilder &builder, mlir::Location loc,
     const std::vector<Fortran::lower::EnvironmentDefault> &envDefaults);
 
diff --git a/flang/include/flang/Optimizer/Builder/Runtime/Main.h b/flang/include/flang/Optimizer/Builder/Runtime/Main.h
new file mode 100644
index 000000000000..62faf46e1fc7
--- /dev/null
+++ b/flang/include/flang/Optimizer/Builder/Runtime/Main.h
@@ -0,0 +1,28 @@
+//===-- Main.h - generate main runtime API calls ----------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef FORTRAN_OPTIMIZER_BUILDER_RUNTIME_MAIN_H
+#define FORTRAN_OPTIMIZER_BUILDER_RUNTIME_MAIN_H
+
+namespace mlir {
+class Location;
+} // namespace mlir
+
+namespace fir {
+class FirOpBuilder;
+class GlobalOp;
+} // namespace fir
+
+namespace fir::runtime {
+
+void genMain(fir::FirOpBuilder &builder, mlir::Location loc,
+             fir::GlobalOp &env);
+
+}
+
+#endif // FORTRAN_OPTIMIZER_BUILDER_RUNTIME_MAIN_H
diff --git a/flang/include/flang/Optimizer/Dialect/CanonicalizationPatterns.td b/flang/include/flang/Optimizer/Dialect/CanonicalizationPatterns.td
index 544fc3cdf75e..0ef37a37ce94 100644
--- a/flang/include/flang/Optimizer/Dialect/CanonicalizationPatterns.td
+++ b/flang/include/flang/Optimizer/Dialect/CanonicalizationPatterns.td
@@ -21,17 +21,18 @@ include "flang/Optimizer/Dialect/FIROps.td"
 
 def IdenticalTypePred : Constraint<CPred<"$0.getType() == $1.getType()">>;
 def IntegerTypePred : Constraint<CPred<"fir::isa_integer($0.getType())">>;
-def IndexTypePred : Constraint<CPred<"$0.getType().isa<mlir::IndexType>()">>;
+def IndexTypePred : Constraint<CPred<
+                       "mlir::isa<mlir::IndexType>($0.getType())">>;
 
 // Widths are monotonic.
 //   $0.bits >= $1.bits >= $2.bits or $0.bits <= $1.bits <= $2.bits
 def MonotonicTypePred
-    : Constraint<CPred<"(($0.getType().isa<mlir::IntegerType>() && "
-                       "  $1.getType().isa<mlir::IntegerType>() && "
-                       "  $2.getType().isa<mlir::IntegerType>()) || "
-                       " ($0.getType().isa<mlir::FloatType>() && "
-                       "  $1.getType().isa<mlir::FloatType>() && "
-                       "  $2.getType().isa<mlir::FloatType>())) && "
+    : Constraint<CPred<"((mlir::isa<mlir::IntegerType>($0.getType()) && "
+                       "  mlir::isa<mlir::IntegerType>($1.getType()) && "
+                       "  mlir::isa<mlir::IntegerType>($2.getType())) || "
+                       " (mlir::isa<mlir::FloatType>($0.getType()) && "
+                       "  mlir::isa<mlir::FloatType>($1.getType()) && "
+                       "  mlir::isa<mlir::FloatType>($2.getType()))) && "
                        "(($0.getType().getIntOrFloatBitWidth() <= "
                        "  $1.getType().getIntOrFloatBitWidth() && "
                        "  $1.getType().getIntOrFloatBitWidth() <= "
@@ -42,8 +43,8 @@ def MonotonicTypePred
                        "  $2.getType().getIntOrFloatBitWidth()))">>;
 
 def IntPred : Constraint<CPred<
-                       "$0.getType().isa<mlir::IntegerType>() && "
-                       "$1.getType().isa<mlir::IntegerType>()">>;
+                       "mlir::isa<mlir::IntegerType>($0.getType()) && "
+                       "mlir::isa<mlir::IntegerType>($1.getType())">>;
                        
 // If both are int type and the first is smaller than the second.
 //   $0.bits <= $1.bits
@@ -101,8 +102,8 @@ def CombineConvertTruncOptPattern
 def createConstantOp
     : NativeCodeCall<"$_builder.create<mlir::arith::ConstantOp>"
                      "($_loc, $_builder.getIndexType(), "
-                     "rewriter.getIndexAttr($1.dyn_cast<mlir::IntegerAttr>()"
-                     ".getInt()))">;
+                     "rewriter.getIndexAttr("
+                     "mlir::dyn_cast<mlir::IntegerAttr>($1).getInt()))">;
 
 def ForwardConstantConvertPattern
     : Pat<(fir_ConvertOp:$res (Arith_ConstantOp:$cnt $attr)),
diff --git a/flang/include/flang/Optimizer/Dialect/FIROps.td b/flang/include/flang/Optimizer/Dialect/FIROps.td
index 92790a691e47..496193e25cab 100644
--- a/flang/include/flang/Optimizer/Dialect/FIROps.td
+++ b/flang/include/flang/Optimizer/Dialect/FIROps.td
@@ -2708,14 +2708,14 @@ def fir_ConvertOp : fir_OneResultOp<"convert", [NoMemoryEffect]> {
   let hasCanonicalizer = 1;
 }
 
-def FortranTypeAttr : Attr<And<[CPred<"$_self.isa<mlir::TypeAttr>()">,
-    Or<[CPred<"$_self.cast<mlir::TypeAttr>().getValue().isa<fir::CharacterType,"
-              "fir::ComplexType, fir::IntegerType, fir::LogicalType,"
-              "fir::RealType, fir::RecordType>()">]>]>,
-    "Fortran surface type"> {
+def FortranTypeAttr : Attr<And<[CPred<"mlir::isa<mlir::TypeAttr>($_self)">,
+    Or<[CPred<"mlir::isa<fir::CharacterType, fir::ComplexType, "
+              "fir::IntegerType, fir::LogicalType, fir::RealType, "
+              "fir::RecordType>(mlir::cast<mlir::TypeAttr>($_self).getValue())"
+    >]>]>, "Fortran surface type"> {
   let storageType = [{ ::mlir::TypeAttr }];
   let returnType = "mlir::Type";
-  let convertFromStorage = "$_self.getValue().cast<mlir::Type>()";
+  let convertFromStorage = "mlir::cast<mlir::Type>($_self.getValue())";
 }
 
 def fir_TypeDescOp : fir_OneResultOp<"type_desc", [NoMemoryEffect]> {
diff --git a/flang/include/flang/Optimizer/Dialect/FIRType.h b/flang/include/flang/Optimizer/Dialect/FIRType.h
index 7fcd9c1babf2..b4344435db9f 100644
--- a/flang/include/flang/Optimizer/Dialect/FIRType.h
+++ b/flang/include/flang/Optimizer/Dialect/FIRType.h
@@ -97,35 +97,36 @@ bool isa_fir_or_std_type(mlir::Type t);
 
 /// Is `t` a FIR dialect type that implies a memory (de)reference?
 inline bool isa_ref_type(mlir::Type t) {
-  return t.isa<fir::ReferenceType, fir::PointerType, fir::HeapType,
-               fir::LLVMPointerType>();
+  return mlir::isa<fir::ReferenceType, fir::PointerType, fir::HeapType,
+                   fir::LLVMPointerType>(t);
 }
 
 /// Is `t` a boxed type?
 inline bool isa_box_type(mlir::Type t) {
-  return t.isa<fir::BaseBoxType, fir::BoxCharType, fir::BoxProcType>();
+  return mlir::isa<fir::BaseBoxType, fir::BoxCharType, fir::BoxProcType>(t);
 }
 
 /// Is `t` a type that is always trivially pass-by-reference? Specifically, this
 /// is testing if `t` is a ReferenceType or any box type. Compare this to
 /// conformsWithPassByRef(), which includes pointers and allocatables.
 inline bool isa_passbyref_type(mlir::Type t) {
-  return t.isa<fir::ReferenceType, mlir::FunctionType>() || isa_box_type(t);
+  return mlir::isa<fir::ReferenceType, mlir::FunctionType>(t) ||
+         isa_box_type(t);
 }
 
 /// Is `t` a type that can conform to be pass-by-reference? Depending on the
 /// context, these types may simply demote to pass-by-reference or a reference
 /// to them may have to be passed instead. Functions are always referent.
 inline bool conformsWithPassByRef(mlir::Type t) {
-  return isa_ref_type(t) || isa_box_type(t) || t.isa<mlir::FunctionType>();
+  return isa_ref_type(t) || isa_box_type(t) || mlir::isa<mlir::FunctionType>(t);
 }
 
 /// Is `t` a derived (record) type?
-inline bool isa_derived(mlir::Type t) { return t.isa<fir::RecordType>(); }
+inline bool isa_derived(mlir::Type t) { return mlir::isa<fir::RecordType>(t); }
 
 /// Is `t` type(c_ptr) or type(c_funptr)?
 inline bool isa_builtin_cptr_type(mlir::Type t) {
-  if (auto recTy = t.dyn_cast_or_null<fir::RecordType>())
+  if (auto recTy = mlir::dyn_cast_or_null<fir::RecordType>(t))
     return recTy.getName().ends_with("T__builtin_c_ptr") ||
            recTy.getName().ends_with("T__builtin_c_funptr");
   return false;
@@ -133,7 +134,7 @@ inline bool isa_builtin_cptr_type(mlir::Type t) {
 
 /// Is `t` a FIR dialect aggregate type?
 inline bool isa_aggregate(mlir::Type t) {
-  return t.isa<SequenceType, mlir::TupleType>() || fir::isa_derived(t);
+  return mlir::isa<SequenceType, mlir::TupleType>(t) || fir::isa_derived(t);
 }
 
 /// Extract the `Type` pointed to from a FIR memory reference type. If `t` is
@@ -146,17 +147,17 @@ mlir::Type dyn_cast_ptrOrBoxEleTy(mlir::Type t);
 
 /// Is `t` a FIR Real or MLIR Float type?
 inline bool isa_real(mlir::Type t) {
-  return t.isa<fir::RealType, mlir::FloatType>();
+  return mlir::isa<fir::RealType, mlir::FloatType>(t);
 }
 
 /// Is `t` an integral type?
 inline bool isa_integer(mlir::Type t) {
-  return t.isa<mlir::IndexType, mlir::IntegerType, fir::IntegerType>();
+  return mlir::isa<mlir::IndexType, mlir::IntegerType, fir::IntegerType>(t);
 }
 
 /// Is `t` a vector type?
 inline bool isa_vector(mlir::Type t) {
-  return t.isa<mlir::VectorType, fir::VectorType>();
+  return mlir::isa<mlir::VectorType, fir::VectorType>(t);
 }
 
 mlir::Type parseFirType(FIROpsDialect *, mlir::DialectAsmParser &parser);
@@ -169,22 +170,22 @@ void verifyIntegralType(mlir::Type type);
 
 /// Is `t` a FIR or MLIR Complex type?
 inline bool isa_complex(mlir::Type t) {
-  return t.isa<fir::ComplexType, mlir::ComplexType>();
+  return mlir::isa<fir::ComplexType, mlir::ComplexType>(t);
 }
 
 /// Is `t` a CHARACTER type? Does not check the length.
-inline bool isa_char(mlir::Type t) { return t.isa<fir::CharacterType>(); }
+inline bool isa_char(mlir::Type t) { return mlir::isa<fir::CharacterType>(t); }
 
 /// Is `t` a trivial intrinsic type? CHARACTER is <em>excluded</em> because it
 /// is a dependent type.
 inline bool isa_trivial(mlir::Type t) {
   return isa_integer(t) || isa_real(t) || isa_complex(t) || isa_vector(t) ||
-         t.isa<fir::LogicalType>();
+         mlir::isa<fir::LogicalType>(t);
 }
 
 /// Is `t` a CHARACTER type with a LEN other than 1?
 inline bool isa_char_string(mlir::Type t) {
-  if (auto ct = t.dyn_cast_or_null<fir::CharacterType>())
+  if (auto ct = mlir::dyn_cast_or_null<fir::CharacterType>(t))
     return ct.getLen() != fir::CharacterType::singleton();
   return false;
 }
@@ -198,7 +199,7 @@ bool isa_unknown_size_box(mlir::Type t);
 
 /// Returns true iff `t` is a fir.char type and has an unknown length.
 inline bool characterWithDynamicLen(mlir::Type t) {
-  if (auto charTy = t.dyn_cast<fir::CharacterType>())
+  if (auto charTy = mlir::dyn_cast<fir::CharacterType>(t))
     return charTy.hasDynamicLen();
   return false;
 }
@@ -213,11 +214,11 @@ inline bool sequenceWithNonConstantShape(fir::SequenceType seqTy) {
 bool hasDynamicSize(mlir::Type t);
 
 inline unsigned getRankOfShapeType(mlir::Type t) {
-  if (auto shTy = t.dyn_cast<fir::ShapeType>())
+  if (auto shTy = mlir::dyn_cast<fir::ShapeType>(t))
     return shTy.getRank();
-  if (auto shTy = t.dyn_cast<fir::ShapeShiftType>())
+  if (auto shTy = mlir::dyn_cast<fir::ShapeShiftType>(t))
     return shTy.getRank();
-  if (auto shTy = t.dyn_cast<fir::ShiftType>())
+  if (auto shTy = mlir::dyn_cast<fir::ShiftType>(t))
     return shTy.getRank();
   return 0;
 }
@@ -225,14 +226,14 @@ inline unsigned getRankOfShapeType(mlir::Type t) {
 /// Get the memory reference type of the data pointer from the box type,
 inline mlir::Type boxMemRefType(fir::BaseBoxType t) {
   auto eleTy = t.getEleTy();
-  if (!eleTy.isa<fir::PointerType, fir::HeapType>())
+  if (!mlir::isa<fir::PointerType, fir::HeapType>(eleTy))
     eleTy = fir::ReferenceType::get(t);
   return eleTy;
 }
 
 /// If `t` is a SequenceType return its element type, otherwise return `t`.
 inline mlir::Type unwrapSequenceType(mlir::Type t) {
-  if (auto seqTy = t.dyn_cast<fir::SequenceType>())
+  if (auto seqTy = mlir::dyn_cast<fir::SequenceType>(t))
     return seqTy.getEleTy();
   return t;
 }
@@ -278,7 +279,7 @@ inline fir::SequenceType unwrapUntilSeqType(mlir::Type t) {
       t = ty;
       continue;
     }
-    if (auto seqTy = t.dyn_cast<fir::SequenceType>())
+    if (auto seqTy = mlir::dyn_cast<fir::SequenceType>(t))
       return seqTy;
     return {};
   }
@@ -287,8 +288,8 @@ inline fir::SequenceType unwrapUntilSeqType(mlir::Type t) {
 /// Unwrap the referential and sequential outer types (if any). Returns the
 /// the element if type is fir::RecordType
 inline fir::RecordType unwrapIfDerived(fir::BaseBoxType boxTy) {
-  return fir::unwrapSequenceType(fir::unwrapRefType(boxTy.getEleTy()))
-      .template dyn_cast<fir::RecordType>();
+  return mlir::dyn_cast<fir::RecordType>(
+      fir::unwrapSequenceType(fir::unwrapRefType(boxTy.getEleTy())));
 }
 
 /// Return true iff `boxTy` wraps a fir::RecordType with length parameters
@@ -377,7 +378,7 @@ bool isRecordWithDescriptorMember(mlir::Type ty);
 
 /// Return true iff `ty` is a RecordType with type parameters.
 inline bool isRecordWithTypeParameters(mlir::Type ty) {
-  if (auto recTy = ty.dyn_cast_or_null<fir::RecordType>())
+  if (auto recTy = mlir::dyn_cast_or_null<fir::RecordType>(ty))
     return recTy.isDependentType();
   return false;
 }
@@ -401,14 +402,14 @@ mlir::Type fromRealTypeID(mlir::MLIRContext *context, llvm::Type::TypeID typeID,
 int getTypeCode(mlir::Type ty, const KindMapping &kindMap);
 
 inline bool BaseBoxType::classof(mlir::Type type) {
-  return type.isa<fir::BoxType, fir::ClassType>();
+  return mlir::isa<fir::BoxType, fir::ClassType>(type);
 }
 
 /// Return true iff `ty` is none or fir.array<none>.
 inline bool isNoneOrSeqNone(mlir::Type type) {
-  if (auto seqTy = type.dyn_cast<fir::SequenceType>())
-    return seqTy.getEleTy().isa<mlir::NoneType>();
-  return type.isa<mlir::NoneType>();
+  if (auto seqTy = mlir::dyn_cast<fir::SequenceType>(type))
+    return mlir::isa<mlir::NoneType>(seqTy.getEleTy());
+  return mlir::isa<mlir::NoneType>(type);
 }
 
 /// Return a fir.box<T> or fir.class<T> if the type is polymorphic. If the type
@@ -428,16 +429,16 @@ inline mlir::Type wrapInClassOrBoxType(mlir::Type eleTy,
 /// !fir.array<2xf32> -> !fir.array<2xnone>
 /// !fir.heap<!fir.array<2xf32>> -> !fir.heap<!fir.array<2xnone>>
 inline mlir::Type updateTypeForUnlimitedPolymorphic(mlir::Type ty) {
-  if (auto seqTy = ty.dyn_cast<fir::SequenceType>())
+  if (auto seqTy = mlir::dyn_cast<fir::SequenceType>(ty))
     return fir::SequenceType::get(
         seqTy.getShape(), updateTypeForUnlimitedPolymorphic(seqTy.getEleTy()));
-  if (auto heapTy = ty.dyn_cast<fir::HeapType>())
+  if (auto heapTy = mlir::dyn_cast<fir::HeapType>(ty))
     return fir::HeapType::get(
         updateTypeForUnlimitedPolymorphic(heapTy.getEleTy()));
-  if (auto pointerTy = ty.dyn_cast<fir::PointerType>())
+  if (auto pointerTy = mlir::dyn_cast<fir::PointerType>(ty))
     return fir::PointerType::get(
         updateTypeForUnlimitedPolymorphic(pointerTy.getEleTy()));
-  if (!ty.isa<mlir::NoneType, fir::RecordType>())
+  if (!mlir::isa<mlir::NoneType, fir::RecordType>(ty))
     return mlir::NoneType::get(ty.getContext());
   return ty;
 }
@@ -451,18 +452,19 @@ mlir::Type changeElementType(mlir::Type type, mlir::Type newElementType,
 
 /// Is `t` an address to fir.box or class type?
 inline bool isBoxAddress(mlir::Type t) {
-  return fir::isa_ref_type(t) && fir::unwrapRefType(t).isa<fir::BaseBoxType>();
+  return fir::isa_ref_type(t) &&
+         mlir::isa<fir::BaseBoxType>(fir::unwrapRefType(t));
 }
 
 /// Is `t` a fir.box or class address or value type?
 inline bool isBoxAddressOrValue(mlir::Type t) {
-  return fir::unwrapRefType(t).isa<fir::BaseBoxType>();
+  return mlir::isa<fir::BaseBoxType>(fir::unwrapRefType(t));
 }
 
 /// Is this a fir.boxproc address type?
 inline bool isBoxProcAddressType(mlir::Type t) {
   t = fir::dyn_cast_ptrEleTy(t);
-  return t && t.isa<fir::BoxProcType>();
+  return t && mlir::isa<fir::BoxProcType>(t);
 }
 
 /// Return a string representation of `ty`.
diff --git a/flang/include/flang/Optimizer/Dialect/FIRTypes.td b/flang/include/flang/Optimizer/Dialect/FIRTypes.td
index 3b876e4642da..7378ed93944c 100644
--- a/flang/include/flang/Optimizer/Dialect/FIRTypes.td
+++ b/flang/include/flang/Optimizer/Dialect/FIRTypes.td
@@ -578,7 +578,7 @@ def fir_VoidType : FIR_Type<"Void", "void"> {
 
 // Whether a type is a BaseBoxType
 def IsBaseBoxTypePred
-        : CPred<"$_self.isa<::fir::BaseBoxType>()">;
+        : CPred<"mlir::isa<::fir::BaseBoxType>($_self)">;
 def fir_BaseBoxType : Type<IsBaseBoxTypePred, "fir.box or fir.class type">;
 
 // Generalized FIR and standard dialect types representing intrinsic types
diff --git a/flang/include/flang/Optimizer/Dialect/FortranVariableInterface.td b/flang/include/flang/Optimizer/Dialect/FortranVariableInterface.td
index 6405afbf1bfb..3f78a93a2515 100644
--- a/flang/include/flang/Optimizer/Dialect/FortranVariableInterface.td
+++ b/flang/include/flang/Optimizer/Dialect/FortranVariableInterface.td
@@ -75,7 +75,7 @@ def fir_FortranVariableOpInterface : OpInterface<"FortranVariableOpInterface"> {
     /// variable.
     mlir::Type getElementOrSequenceType() {
       mlir::Type type = fir::unwrapPassByRefType(fir::unwrapRefType(getBase().getType()));
-      if (auto boxCharType = type.dyn_cast<fir::BoxCharType>())
+      if (auto boxCharType = mlir::dyn_cast<fir::BoxCharType>(type))
         return boxCharType.getEleTy();
       return type;
     }
@@ -87,13 +87,13 @@ def fir_FortranVariableOpInterface : OpInterface<"FortranVariableOpInterface"> {
 
     /// Is the variable an array?
     bool isArray() {
-      return getElementOrSequenceType().isa<fir::SequenceType>();
+      return mlir::isa<fir::SequenceType>(getElementOrSequenceType());
     }
 
     /// Return the rank of the entity if it is known at compile time.
     std::optional<unsigned> getRank() {
       if (auto sequenceType =
-            getElementOrSequenceType().dyn_cast<fir::SequenceType>()) {
+            mlir::dyn_cast<fir::SequenceType>(getElementOrSequenceType())) {
         if (sequenceType.hasUnknownShape())
           return {};
         return sequenceType.getDimension();
@@ -133,7 +133,7 @@ def fir_FortranVariableOpInterface : OpInterface<"FortranVariableOpInterface"> {
 
     /// Is this a Fortran character variable?
     bool isCharacter() {
-      return getElementType().isa<fir::CharacterType>();
+      return mlir::isa<fir::CharacterType>(getElementType());
     }
 
     /// Is this a Fortran character variable with an explicit length?
@@ -149,7 +149,7 @@ def fir_FortranVariableOpInterface : OpInterface<"FortranVariableOpInterface"> {
 
     /// Is this variable represented as a fir.box or fir.class value?
     bool isBoxValue() {
-      return getBase().getType().isa<fir::BaseBoxType>();
+      return mlir::isa<fir::BaseBoxType>(getBase().getType());
     }
 
     /// Is this variable represented as a fir.box or fir.class address?
diff --git a/flang/include/flang/Optimizer/HLFIR/HLFIRDialect.h b/flang/include/flang/Optimizer/HLFIR/HLFIRDialect.h
index aa68d0811c48..3830237f96f3 100644
--- a/flang/include/flang/Optimizer/HLFIR/HLFIRDialect.h
+++ b/flang/include/flang/Optimizer/HLFIR/HLFIRDialect.h
@@ -40,9 +40,9 @@ namespace hlfir {
 inline mlir::Type getFortranElementType(mlir::Type type) {
   type = fir::unwrapSequenceType(
       fir::unwrapPassByRefType(fir::unwrapRefType(type)));
-  if (auto exprType = type.dyn_cast<hlfir::ExprType>())
+  if (auto exprType = mlir::dyn_cast<hlfir::ExprType>(type))
     return exprType.getEleTy();
-  if (auto boxCharType = type.dyn_cast<fir::BoxCharType>())
+  if (auto boxCharType = mlir::dyn_cast<fir::BoxCharType>(type))
     return boxCharType.getEleTy();
   return type;
 }
@@ -51,12 +51,12 @@ inline mlir::Type getFortranElementType(mlir::Type type) {
 /// fir.array type. Otherwise, returns the Fortran element typeof the entity.
 inline mlir::Type getFortranElementOrSequenceType(mlir::Type type) {
   type = fir::unwrapPassByRefType(fir::unwrapRefType(type));
-  if (auto exprType = type.dyn_cast<hlfir::ExprType>()) {
+  if (auto exprType = mlir::dyn_cast<hlfir::ExprType>(type)) {
     if (exprType.isArray())
       return fir::SequenceType::get(exprType.getShape(), exprType.getEleTy());
     return exprType.getEleTy();
   }
-  if (auto boxCharType = type.dyn_cast<fir::BoxCharType>())
+  if (auto boxCharType = mlir::dyn_cast<fir::BoxCharType>(type))
     return boxCharType.getEleTy();
   return type;
 }
@@ -64,16 +64,16 @@ inline mlir::Type getFortranElementOrSequenceType(mlir::Type type) {
 /// Is this a fir.box or fir.class address type?
 inline bool isBoxAddressType(mlir::Type type) {
   type = fir::dyn_cast_ptrEleTy(type);
-  return type && type.isa<fir::BaseBoxType>();
+  return type && mlir::isa<fir::BaseBoxType>(type);
 }
 
 /// Is this a fir.box or fir.class address or value type?
 inline bool isBoxAddressOrValueType(mlir::Type type) {
-  return fir::unwrapRefType(type).isa<fir::BaseBoxType>();
+  return mlir::isa<fir::BaseBoxType>(fir::unwrapRefType(type));
 }
 
 inline bool isPolymorphicType(mlir::Type type) {
-  if (auto exprType = type.dyn_cast<hlfir::ExprType>())
+  if (auto exprType = mlir::dyn_cast<hlfir::ExprType>(type))
     return exprType.isPolymorphic();
   return fir::isPolymorphicType(type);
 }
@@ -81,14 +81,14 @@ inline bool isPolymorphicType(mlir::Type type) {
 /// Is this an SSA value type for the value of a Fortran procedure
 /// designator ?
 inline bool isFortranProcedureValue(mlir::Type type) {
-  return type.isa<fir::BoxProcType>() ||
-         (type.isa<mlir::TupleType>() &&
+  return mlir::isa<fir::BoxProcType>(type) ||
+         (mlir::isa<mlir::TupleType>(type) &&
           fir::isCharacterProcedureTuple(type, /*acceptRawFunc=*/false));
 }
 
 /// Is this an SSA value type for the value of a Fortran expression?
 inline bool isFortranValueType(mlir::Type type) {
-  return type.isa<hlfir::ExprType>() || fir::isa_trivial(type) ||
+  return mlir::isa<hlfir::ExprType>(type) || fir::isa_trivial(type) ||
          isFortranProcedureValue(type);
 }
 
diff --git a/flang/include/flang/Optimizer/Support/Utils.h b/flang/include/flang/Optimizer/Support/Utils.h
index 2b4fa50e0e42..2da6f24da40e 100644
--- a/flang/include/flang/Optimizer/Support/Utils.h
+++ b/flang/include/flang/Optimizer/Support/Utils.h
@@ -29,7 +29,9 @@
 namespace fir {
 /// Return the integer value of a arith::ConstantOp.
 inline std::int64_t toInt(mlir::arith::ConstantOp cop) {
-  return cop.getValue().cast<mlir::IntegerAttr>().getValue().getSExtValue();
+  return mlir::cast<mlir::IntegerAttr>(cop.getValue())
+      .getValue()
+      .getSExtValue();
 }
 
 // Reconstruct binding tables for dynamic dispatch.
diff --git a/flang/include/flang/Optimizer/Transforms/Passes.h b/flang/include/flang/Optimizer/Transforms/Passes.h
index 4d290d87d4cc..547fe742967a 100644
--- a/flang/include/flang/Optimizer/Transforms/Passes.h
+++ b/flang/include/flang/Optimizer/Transforms/Passes.h
@@ -37,8 +37,7 @@ namespace fir {
 #define GEN_PASS_DECL_ANNOTATECONSTANTOPERANDS
 #define GEN_PASS_DECL_ARRAYVALUECOPY
 #define GEN_PASS_DECL_CHARACTERCONVERSION
-#define GEN_PASS_DECL_CFGCONVERSIONONFUNC
-#define GEN_PASS_DECL_CFGCONVERSIONONREDUCTION
+#define GEN_PASS_DECL_CFGCONVERSION
 #define GEN_PASS_DECL_EXTERNALNAMECONVERSION
 #define GEN_PASS_DECL_MEMREFDATAFLOWOPT
 #define GEN_PASS_DECL_SIMPLIFYINTRINSICS
@@ -47,30 +46,24 @@ namespace fir {
 #define GEN_PASS_DECL_ALGEBRAICSIMPLIFICATION
 #define GEN_PASS_DECL_POLYMORPHICOPCONVERSION
 #define GEN_PASS_DECL_OPENACCDATAOPERANDCONVERSION
+#define GEN_PASS_DECL_ADDDEBUGINFO
+#define GEN_PASS_DECL_STACKARRAYS
+#define GEN_PASS_DECL_LOOPVERSIONING
 #include "flang/Optimizer/Transforms/Passes.h.inc"
 
 std::unique_ptr<mlir::Pass> createAffineDemotionPass();
 std::unique_ptr<mlir::Pass>
 createArrayValueCopyPass(fir::ArrayValueCopyOptions options = {});
-std::unique_ptr<mlir::Pass> createFirToCfgOnFuncPass();
-std::unique_ptr<mlir::Pass> createFirToCfgOnReductionPass();
-std::unique_ptr<mlir::Pass> createCharacterConversionPass();
 std::unique_ptr<mlir::Pass> createExternalNameConversionPass();
 std::unique_ptr<mlir::Pass>
 createExternalNameConversionPass(bool appendUnderscore);
 std::unique_ptr<mlir::Pass> createMemDataFlowOptPass();
 std::unique_ptr<mlir::Pass> createPromoteToAffinePass();
-std::unique_ptr<mlir::Pass> createMemoryAllocationPass();
-std::unique_ptr<mlir::Pass> createStackArraysPass();
 std::unique_ptr<mlir::Pass> createAliasTagsPass();
-std::unique_ptr<mlir::Pass> createSimplifyIntrinsicsPass();
-std::unique_ptr<mlir::Pass> createAddDebugInfoPass();
-std::unique_ptr<mlir::Pass> createLoopVersioningPass();
-
 std::unique_ptr<mlir::Pass>
-createMemoryAllocationPass(bool dynOnHeap, std::size_t maxStackSize);
+createAddDebugInfoPass(fir::AddDebugInfoOptions options = {});
+
 std::unique_ptr<mlir::Pass> createAnnotateConstantOperandsPass();
-std::unique_ptr<mlir::Pass> createSimplifyRegionLitePass();
 std::unique_ptr<mlir::Pass> createAlgebraicSimplificationPass();
 std::unique_ptr<mlir::Pass>
 createAlgebraicSimplificationPass(const mlir::GreedyRewriteConfig &config);
diff --git a/flang/include/flang/Optimizer/Transforms/Passes.td b/flang/include/flang/Optimizer/Transforms/Passes.td
index 467b7e1c472e..020b8a6b64a9 100644
--- a/flang/include/flang/Optimizer/Transforms/Passes.td
+++ b/flang/include/flang/Optimizer/Transforms/Passes.td
@@ -127,7 +127,6 @@ def CharacterConversion : Pass<"character-conversion"> {
     By default the translation is to naively zero-extend or truncate a code
     point to fit the destination size.
   }];
-  let constructor = "::fir::createCharacterConversionPass()";
   let dependentDialects = [ "fir::FIROpsDialect" ];
   let options = [
     Option<"useRuntimeCalls", "use-runtime-calls",
@@ -137,8 +136,7 @@ def CharacterConversion : Pass<"character-conversion"> {
   ];
 }
 
-class CFGConversionBase<string optExt, string operation>
-  : Pass<"cfg-conversion-on-" # optExt # "-opt", operation> {
+def CFGConversion : Pass<"cfg-conversion"> {
   let summary = "Convert FIR structured control flow ops to CFG ops.";
   let description = [{
     Transform the `fir.do_loop`, `fir.if`, `fir.iterate_while` and
@@ -157,14 +155,6 @@ class CFGConversionBase<string optExt, string operation>
   ];
 }
 
-def CFGConversionOnFunc : CFGConversionBase<"func", "mlir::func::FuncOp"> {
-  let constructor = "::fir::createFirToCfgOnFuncPass()";
-}
-
-def CFGConversionOnReduction : CFGConversionBase<"reduce", "mlir::omp::DeclareReductionOp"> {
-  let constructor = "::fir::createFirToCfgOnReductionPass()";
-}
-
 def ExternalNameConversion : Pass<"external-name-interop", "mlir::ModuleOp"> {
   let summary = "Convert name for external interoperability";
   let description = [{
@@ -202,6 +192,25 @@ def AddDebugInfo : Pass<"add-debug-info", "mlir::ModuleOp"> {
   let dependentDialects = [
     "fir::FIROpsDialect", "mlir::func::FuncDialect", "mlir::LLVM::LLVMDialect"
   ];
+  let options = [
+    Option<"debugLevel", "debug-level",
+           "mlir::LLVM::DIEmissionKind",
+           /*default=*/"mlir::LLVM::DIEmissionKind::Full",
+           "debug level",
+           [{::llvm::cl::values(
+            clEnumValN(mlir::LLVM::DIEmissionKind::Full, "Full", "Emit full debug info"),
+            clEnumValN(mlir::LLVM::DIEmissionKind::LineTablesOnly, "LineTablesOnly", "Emit line tables only"),
+            clEnumValN(mlir::LLVM::DIEmissionKind::None, "None", "Emit no debug information")
+          )}]
+           >,
+    Option<"isOptimized", "is-optimized",
+           "bool", /*default=*/"false",
+           "is optimized.">,
+    Option<"inputFilename", "file-name",
+           "std::string",
+           /*default=*/"std::string{}",
+           "name of the input source file">,
+  ];
 }
 
 // This needs to be a "mlir::ModuleOp" pass, because it inserts simplified
@@ -215,7 +224,6 @@ def SimplifyIntrinsics : Pass<"simplify-intrinsics", "mlir::ModuleOp"> {
     simplified function. The simplified function is added to the current module.
     This function can be inlined by a general purpose inlining pass.
   }];
-  let constructor = "::fir::createSimplifyIntrinsicsPass()";
 
   let options = [
     Option<"enableExperimental", "enable-experimental", "bool",
@@ -239,7 +247,6 @@ def MemoryAllocationOpt : Pass<"memory-allocation-opt", "mlir::func::FuncOp"> {
            "std::size_t", /*default=*/"~static_cast<std::size_t>(0)",
            "Set maximum number of elements of an array allocated on the stack.">
   ];
-  let constructor = "::fir::createMemoryAllocationPass()";
 }
 
 def StackArrays : Pass<"stack-arrays", "mlir::ModuleOp"> {
@@ -249,7 +256,6 @@ def StackArrays : Pass<"stack-arrays", "mlir::ModuleOp"> {
     allocations.
   }];
   let dependentDialects = [ "fir::FIROpsDialect" ];
-  let constructor = "::fir::createStackArraysPass()";
 }
 
 def AddAliasTags : Pass<"fir-add-alias-tags", "mlir::ModuleOp"> {
@@ -277,7 +283,6 @@ def SimplifyRegionLite : Pass<"simplify-region-lite", "mlir::ModuleOp"> {
   let description = [{
     Run region DCE and erase unreachable blocks in regions.
   }];
-  let constructor = "::fir::createSimplifyRegionLitePass()";
 }
 
 def AlgebraicSimplification : Pass<"flang-algebraic-simplification"> {
@@ -314,7 +319,6 @@ def LoopVersioning : Pass<"loop-versioning", "mlir::func::FuncOp"> {
     an array has element sized stride. The element sizes stride allows some
     loops to be vectorized as well as other loop optimizations.
   }];
-  let constructor = "::fir::createLoopVersioningPass()";
   let dependentDialects = [ "fir::FIROpsDialect" ];
 }
 
diff --git a/flang/include/flang/Parser/parse-tree.h b/flang/include/flang/Parser/parse-tree.h
index d7c23755c57b..4641f9d20d5b 100644
--- a/flang/include/flang/Parser/parse-tree.h
+++ b/flang/include/flang/Parser/parse-tree.h
@@ -455,7 +455,8 @@ struct SpecificationPart {
 struct InternalSubprogram {
   UNION_CLASS_BOILERPLATE(InternalSubprogram);
   std::variant<common::Indirection<FunctionSubprogram>,
-      common::Indirection<SubroutineSubprogram>>
+      common::Indirection<SubroutineSubprogram>,
+      common::Indirection<CompilerDirective>>
       u;
 };
 
diff --git a/flang/include/flang/Parser/preprocessor.h b/flang/include/flang/Parser/preprocessor.h
index 630d5273d427..c3076435be5f 100644
--- a/flang/include/flang/Parser/preprocessor.h
+++ b/flang/include/flang/Parser/preprocessor.h
@@ -81,6 +81,7 @@ public:
   void Undefine(std::string macro);
   bool IsNameDefined(const CharBlock &);
   bool IsFunctionLikeDefinition(const CharBlock &);
+  bool AnyDefinitions() const { return !definitions_.empty(); }
 
   // When called with partialFunctionLikeMacro not null, MacroReplacement()
   // and ReplaceMacros() handle an unclosed function-like macro reference
diff --git a/flang/include/flang/Runtime/descriptor.h b/flang/include/flang/Runtime/descriptor.h
index 96d56d9b43a6..1b0b7e23ce6c 100644
--- a/flang/include/flang/Runtime/descriptor.h
+++ b/flang/include/flang/Runtime/descriptor.h
@@ -456,6 +456,7 @@ public:
     assert(descriptor().rank() <= maxRank);
     assert(descriptor().SizeInBytes() <= byteSize);
     if (DescriptorAddendum * addendum{descriptor().Addendum()}) {
+      (void)addendum;
       assert(hasAddendum);
       assert(addendum->LenParameters() <= maxLengthTypeParameters);
     } else {
diff --git a/flang/include/flang/Runtime/numeric.h b/flang/include/flang/Runtime/numeric.h
index 3d9cb8b5b0ac..7d3f91360c8c 100644
--- a/flang/include/flang/Runtime/numeric.h
+++ b/flang/include/flang/Runtime/numeric.h
@@ -356,10 +356,18 @@ CppTypeFor<TypeCategory::Real, 16> RTDECL(Scale16)(
     CppTypeFor<TypeCategory::Real, 16>, std::int64_t);
 #endif
 
+// SELECTED_CHAR_KIND
+CppTypeFor<TypeCategory::Integer, 4> RTDECL(SelectedCharKind)(
+    const char *, int, const char *, std::size_t);
+
 // SELECTED_INT_KIND
 CppTypeFor<TypeCategory::Integer, 4> RTDECL(SelectedIntKind)(
     const char *, int, void *, int);
 
+// SELECTED_LOGICAL_KIND
+CppTypeFor<TypeCategory::Integer, 4> RTDECL(SelectedLogicalKind)(
+    const char *, int, void *, int);
+
 // SELECTED_REAL_KIND
 CppTypeFor<TypeCategory::Integer, 4> RTDECL(SelectedRealKind)(
     const char *, int, void *, int, void *, int, void *, int);
diff --git a/flang/include/flang/Tools/CLOptions.inc b/flang/include/flang/Tools/CLOptions.inc
index 44ff2b3f70ff..34af9f1c21f8 100644
--- a/flang/include/flang/Tools/CLOptions.inc
+++ b/flang/include/flang/Tools/CLOptions.inc
@@ -87,8 +87,6 @@ DisableOption(BoxedProcedureRewrite, "boxed-procedure-rewrite",
 DisableOption(ExternalNameConversion, "external-name-interop",
     "convert names with external convention");
 
-// TODO: remove once these are used for non-codegen passes
-#if !defined(FLANG_EXCLUDE_CODEGEN)
 using PassConstructor = std::unique_ptr<mlir::Pass>();
 
 template <typename OP>
@@ -106,9 +104,14 @@ void addNestedPassToOps(mlir::PassManager &pm, PassConstructor ctor) {
 void addNestedPassToAllTopLevelOperations(
     mlir::PassManager &pm, PassConstructor ctor) {
   addNestedPassToOps<mlir::func::FuncOp, mlir::omp::DeclareReductionOp,
-      fir::GlobalOp>(pm, ctor);
+      mlir::omp::PrivateClauseOp, fir::GlobalOp>(pm, ctor);
+}
+
+void addNestedPassToAllTopLevelOperationsConditionally(mlir::PassManager &pm,
+    llvm::cl::opt<bool> &disabled, PassConstructor ctor) {
+  if (!disabled)
+    addNestedPassToAllTopLevelOperations(pm, ctor);
 }
-#endif
 
 /// Generic for adding a pass to the pass manager if it is not disabled.
 template <typename F>
@@ -146,10 +149,8 @@ static void addCanonicalizerPassWithoutRegionSimplification(
 }
 
 inline void addCfgConversionPass(mlir::PassManager &pm) {
-  addNestedPassConditionally<mlir::func::FuncOp>(
-      pm, disableCfgConversion, fir::createFirToCfgOnFuncPass);
-  addNestedPassConditionally<mlir::omp::DeclareReductionOp>(
-      pm, disableCfgConversion, fir::createFirToCfgOnReductionPass);
+  addNestedPassToAllTopLevelOperationsConditionally(
+      pm, disableCfgConversion, fir::createCFGConversion);
 }
 
 inline void addAVC(
@@ -162,8 +163,8 @@ inline void addAVC(
 
 inline void addMemoryAllocationOpt(mlir::PassManager &pm) {
   addNestedPassConditionally<mlir::func::FuncOp>(pm, disableFirMao, [&]() {
-    return fir::createMemoryAllocationPass(
-        dynamicArrayStackToHeapAllocation, arrayStackAllocationThreshold);
+    return fir::createMemoryAllocationOpt(
+        {dynamicArrayStackToHeapAllocation, arrayStackAllocationThreshold});
   });
 }
 
@@ -179,9 +180,27 @@ inline void addTargetRewritePass(mlir::PassManager &pm) {
   });
 }
 
-inline void addDebugInfoPass(mlir::PassManager &pm) {
-  addPassConditionally(
-      pm, disableDebugInfo, [&]() { return fir::createAddDebugInfoPass(); });
+inline mlir::LLVM::DIEmissionKind getEmissionKind(
+    llvm::codegenoptions::DebugInfoKind kind) {
+  switch (kind) {
+  case llvm::codegenoptions::DebugInfoKind::FullDebugInfo:
+    return mlir::LLVM::DIEmissionKind::Full;
+  case llvm::codegenoptions::DebugInfoKind::DebugLineTablesOnly:
+    return mlir::LLVM::DIEmissionKind::LineTablesOnly;
+  default:
+    return mlir::LLVM::DIEmissionKind::None;
+  }
+}
+
+inline void addDebugInfoPass(mlir::PassManager &pm,
+    llvm::codegenoptions::DebugInfoKind debugLevel,
+    llvm::OptimizationLevel optLevel, llvm::StringRef inputFilename) {
+  fir::AddDebugInfoOptions options;
+  options.debugLevel = getEmissionKind(debugLevel);
+  options.isOptimized = optLevel != llvm::OptimizationLevel::O0;
+  options.inputFilename = inputFilename;
+  addPassConditionally(pm, disableDebugInfo,
+      [&]() { return fir::createAddDebugInfoPass(options); });
 }
 
 inline void addFIRToLLVMPass(
@@ -224,22 +243,22 @@ inline void createDefaultFIROptimizerPassPipeline(
   config.enableRegionSimplification = false;
   pm.addPass(mlir::createCSEPass());
   fir::addAVC(pm, pc.OptLevel);
-  pm.addNestedPass<mlir::func::FuncOp>(fir::createCharacterConversionPass());
+  addNestedPassToAllTopLevelOperations(pm, fir::createCharacterConversion);
   pm.addPass(mlir::createCanonicalizerPass(config));
-  pm.addPass(fir::createSimplifyRegionLitePass());
+  pm.addPass(fir::createSimplifyRegionLite());
   if (pc.OptLevel.isOptimizingForSpeed()) {
     // These passes may increase code size.
-    pm.addPass(fir::createSimplifyIntrinsicsPass());
+    pm.addPass(fir::createSimplifyIntrinsics());
     pm.addPass(fir::createAlgebraicSimplificationPass(config));
   }
 
   if (pc.LoopVersioning)
-    pm.addPass(fir::createLoopVersioningPass());
+    pm.addPass(fir::createLoopVersioning());
 
   pm.addPass(mlir::createCSEPass());
 
   if (pc.StackArrays)
-    pm.addPass(fir::createStackArraysPass());
+    pm.addPass(fir::createStackArrays());
   else
     fir::addMemoryAllocationOpt(pm);
 
@@ -248,7 +267,7 @@ inline void createDefaultFIROptimizerPassPipeline(
   llvm::StringMap<mlir::OpPassManager> pipelines;
   pm.addPass(mlir::createInlinerPass(
       pipelines, addCanonicalizerPassWithoutRegionSimplification));
-  pm.addPass(fir::createSimplifyRegionLitePass());
+  pm.addPass(fir::createSimplifyRegionLite());
   pm.addPass(mlir::createCSEPass());
 
   // Polymorphic types
@@ -262,7 +281,7 @@ inline void createDefaultFIROptimizerPassPipeline(
   pm.addPass(mlir::createConvertSCFToCFPass());
 
   pm.addPass(mlir::createCanonicalizerPass(config));
-  pm.addPass(fir::createSimplifyRegionLitePass());
+  pm.addPass(fir::createSimplifyRegionLite());
   pm.addPass(mlir::createCSEPass());
 }
 
@@ -307,32 +326,21 @@ inline void createOpenMPFIRPassPipeline(
 }
 
 #if !defined(FLANG_EXCLUDE_CODEGEN)
-inline void createDebugPasses(
-    mlir::PassManager &pm, llvm::codegenoptions::DebugInfoKind debugLevel) {
-  // Currently only -g1, -g, -gline-tables-only supported
-  switch (debugLevel) {
-  case llvm::codegenoptions::DebugLineTablesOnly:
-    addDebugInfoPass(pm);
-    return;
-  case llvm::codegenoptions::NoDebugInfo:
-    return;
-  default:
-    // TODO: Add cases and passes for other debug options.
-    // All other debug options not implemented yet, currently emits warning
-    // and generates as much debug information as possible.
-    addDebugInfoPass(pm);
-    return;
-  }
+inline void createDebugPasses(mlir::PassManager &pm,
+    llvm::codegenoptions::DebugInfoKind debugLevel,
+    llvm::OptimizationLevel OptLevel, llvm::StringRef inputFilename) {
+  if (debugLevel != llvm::codegenoptions::NoDebugInfo)
+    addDebugInfoPass(pm, debugLevel, OptLevel, inputFilename);
 }
 
-inline void createDefaultFIRCodeGenPassPipeline(
-    mlir::PassManager &pm, MLIRToLLVMPassPipelineConfig config) {
+inline void createDefaultFIRCodeGenPassPipeline(mlir::PassManager &pm,
+    MLIRToLLVMPassPipelineConfig config, llvm::StringRef inputFilename = {}) {
   fir::addBoxedProcedurePass(pm);
   addNestedPassToAllTopLevelOperations(pm, fir::createAbstractResultOpt);
   fir::addCodeGenRewritePass(pm);
   fir::addTargetRewritePass(pm);
   fir::addExternalNameConversionPass(pm, config.Underscoring);
-  fir::createDebugPasses(pm, config.DebugInfo);
+  fir::createDebugPasses(pm, config.DebugInfo, config.OptLevel, inputFilename);
 
   if (config.VScaleMin != 0)
     pm.addPass(fir::createVScaleAttrPass({config.VScaleMin, config.VScaleMax}));
@@ -366,15 +374,16 @@ inline void createDefaultFIRCodeGenPassPipeline(
 /// \param pm - MLIR pass manager that will hold the pipeline definition
 /// \param optLevel - optimization level used for creating FIR optimization
 ///   passes pipeline
-inline void createMLIRToLLVMPassPipeline(
-    mlir::PassManager &pm, const MLIRToLLVMPassPipelineConfig &config) {
+inline void createMLIRToLLVMPassPipeline(mlir::PassManager &pm,
+    const MLIRToLLVMPassPipelineConfig &config,
+    llvm::StringRef inputFilename = {}) {
   fir::createHLFIRToFIRPassPipeline(pm, config.OptLevel);
 
   // Add default optimizer pass pipeline.
   fir::createDefaultFIROptimizerPassPipeline(pm, config);
 
   // Add codegen pass pipeline.
-  fir::createDefaultFIRCodeGenPassPipeline(pm, config);
+  fir::createDefaultFIRCodeGenPassPipeline(pm, config, inputFilename);
 }
 #undef FLANG_EXCLUDE_CODEGEN
 #endif
diff --git a/flang/include/flang/Tools/PointerModels.h b/flang/include/flang/Tools/PointerModels.h
index 7acaf2f9fda5..c3c0977d6e54 100644
--- a/flang/include/flang/Tools/PointerModels.h
+++ b/flang/include/flang/Tools/PointerModels.h
@@ -20,7 +20,7 @@ struct OpenMPPointerLikeModel
     : public mlir::omp::PointerLikeType::ExternalModel<
           OpenMPPointerLikeModel<T>, T> {
   mlir::Type getElementType(mlir::Type pointer) const {
-    return pointer.cast<T>().getElementType();
+    return mlir::cast<T>(pointer).getElementType();
   }
 };
 
@@ -29,7 +29,7 @@ struct OpenACCPointerLikeModel
     : public mlir::acc::PointerLikeType::ExternalModel<
           OpenACCPointerLikeModel<T>, T> {
   mlir::Type getElementType(mlir::Type pointer) const {
-    return pointer.cast<T>().getElementType();
+    return mlir::cast<T>(pointer).getElementType();
   }
 };
 
diff --git a/flang/lib/Evaluate/intrinsics.cpp b/flang/lib/Evaluate/intrinsics.cpp
index f07f94b1a022..1b73cadb682d 100644
--- a/flang/lib/Evaluate/intrinsics.cpp
+++ b/flang/lib/Evaluate/intrinsics.cpp
@@ -777,7 +777,9 @@ static const IntrinsicInterface genericIntrinsicFunction[]{
             {"identity", SameType, Rank::scalar, Optionality::optional},
             {"ordered", AnyLogical, Rank::scalar, Optionality::optional}},
         SameType, Rank::scalar, IntrinsicClass::transformationalFunction},
-    {"repeat", {{"string", SameCharNoLen, Rank::scalar}, {"ncopies", AnyInt}},
+    {"repeat",
+        {{"string", SameCharNoLen, Rank::scalar},
+            {"ncopies", AnyInt, Rank::scalar}},
         SameCharNoLen, Rank::scalar, IntrinsicClass::transformationalFunction},
     {"reshape",
         {{"source", SameType, Rank::array}, {"shape", AnyInt, Rank::shape},
diff --git a/flang/lib/Evaluate/type.cpp b/flang/lib/Evaluate/type.cpp
index a369e07f94a1..ee1e5b398d9b 100644
--- a/flang/lib/Evaluate/type.cpp
+++ b/flang/lib/Evaluate/type.cpp
@@ -731,7 +731,7 @@ bool SomeKind<TypeCategory::Derived>::operator==(
   return PointeeComparison(derivedTypeSpec_, that.derivedTypeSpec_);
 }
 
-int SelectedCharKind(const std::string &s, int defaultKind) { // 16.9.168
+int SelectedCharKind(const std::string &s, int defaultKind) { // F'2023 16.9.180
   auto lower{parser::ToLowerCaseLetters(s)};
   auto n{lower.size()};
   while (n > 0 && lower[0] == ' ') {
diff --git a/flang/lib/Frontend/FrontendActions.cpp b/flang/lib/Frontend/FrontendActions.cpp
index 8f251997ed40..531616e7926a 100644
--- a/flang/lib/Frontend/FrontendActions.cpp
+++ b/flang/lib/Frontend/FrontendActions.cpp
@@ -63,10 +63,10 @@
 #include "llvm/Support/ErrorHandling.h"
 #include "llvm/Support/FileSystem.h"
 #include "llvm/Support/Path.h"
-#include "llvm/Support/RISCVISAInfo.h"
 #include "llvm/Support/SourceMgr.h"
 #include "llvm/Support/ToolOutputFile.h"
 #include "llvm/Target/TargetMachine.h"
+#include "llvm/TargetParser/RISCVISAInfo.h"
 #include "llvm/TargetParser/RISCVTargetParser.h"
 #include "llvm/Transforms/Utils/ModuleUtils.h"
 #include <memory>
@@ -809,7 +809,7 @@ void CodeGenAction::generateLLVMIR() {
   }
 
   // Create the pass pipeline
-  fir::createMLIRToLLVMPassPipeline(pm, config);
+  fir::createMLIRToLLVMPassPipeline(pm, config, getCurrentFile());
   (void)mlir::applyPassManagerCLOptions(pm);
 
   // run the pass manager
@@ -861,7 +861,6 @@ getOutputStream(CompilerInstance &ci, llvm::StringRef inFile,
     return ci.createDefaultOutputFile(
         /*Binary=*/false, inFile, /*extension=*/"ll");
   case BackendActionTy::Backend_EmitFIR:
-    LLVM_FALLTHROUGH;
   case BackendActionTy::Backend_EmitHLFIR:
     return ci.createDefaultOutputFile(
         /*Binary=*/false, inFile, /*extension=*/"mlir");
diff --git a/flang/lib/Lower/Allocatable.cpp b/flang/lib/Lower/Allocatable.cpp
index 38f61528d7e2..a1957c0eb1bb 100644
--- a/flang/lib/Lower/Allocatable.cpp
+++ b/flang/lib/Lower/Allocatable.cpp
@@ -162,7 +162,7 @@ static void genRuntimeInitCharacter(fir::FirOpBuilder &builder,
   args.push_back(builder.createConvert(loc, inputTypes[0], box.getAddr()));
   args.push_back(builder.createConvert(loc, inputTypes[1], len));
   if (kind == 0)
-    kind = box.getEleTy().cast<fir::CharacterType>().getFKind();
+    kind = mlir::cast<fir::CharacterType>(box.getEleTy()).getFKind();
   args.push_back(builder.createIntegerConstant(loc, inputTypes[2], kind));
   int rank = box.rank();
   args.push_back(builder.createIntegerConstant(loc, inputTypes[3], rank));
@@ -859,30 +859,32 @@ genDeallocate(fir::FirOpBuilder &builder,
 void Fortran::lower::genDeallocateBox(
     Fortran::lower::AbstractConverter &converter,
     const fir::MutableBoxValue &box, mlir::Location loc,
-    mlir::Value declaredTypeDesc) {
+    const Fortran::semantics::Symbol *sym, mlir::Value declaredTypeDesc) {
   const Fortran::lower::SomeExpr *statExpr = nullptr;
   const Fortran::lower::SomeExpr *errMsgExpr = nullptr;
   ErrorManager errorManager;
   errorManager.init(converter, loc, statExpr, errMsgExpr);
   fir::FirOpBuilder &builder = converter.getFirOpBuilder();
-  genDeallocate(builder, converter, loc, box, errorManager, declaredTypeDesc);
+  genDeallocate(builder, converter, loc, box, errorManager, declaredTypeDesc,
+                sym);
 }
 
 void Fortran::lower::genDeallocateIfAllocated(
     Fortran::lower::AbstractConverter &converter,
-    const fir::MutableBoxValue &box, mlir::Location loc) {
+    const fir::MutableBoxValue &box, mlir::Location loc,
+    const Fortran::semantics::Symbol *sym) {
   fir::FirOpBuilder &builder = converter.getFirOpBuilder();
   mlir::Value isAllocated =
       fir::factory::genIsAllocatedOrAssociatedTest(builder, loc, box);
   builder.genIfThen(loc, isAllocated)
       .genThen([&]() {
         if (mlir::Type eleType = box.getEleTy();
-            eleType.isa<fir::RecordType>() && box.isPolymorphic()) {
+            mlir::isa<fir::RecordType>(eleType) && box.isPolymorphic()) {
           mlir::Value declaredTypeDesc = builder.create<fir::TypeDescOp>(
               loc, mlir::TypeAttr::get(eleType));
-          genDeallocateBox(converter, box, loc, declaredTypeDesc);
+          genDeallocateBox(converter, box, loc, sym, declaredTypeDesc);
         } else {
-          genDeallocateBox(converter, box, loc);
+          genDeallocateBox(converter, box, loc, sym);
         }
       })
       .end();
@@ -916,7 +918,7 @@ void Fortran::lower::genDeallocateStmt(
     mlir::Value declaredTypeDesc = {};
     if (box.isPolymorphic()) {
       mlir::Type eleType = box.getEleTy();
-      if (eleType.isa<fir::RecordType>())
+      if (mlir::isa<fir::RecordType>(eleType))
         if (const Fortran::semantics::DerivedTypeSpec *derivedTypeSpec =
                 symbol.GetType()->AsDerived()) {
           declaredTypeDesc =
@@ -1005,7 +1007,7 @@ createMutableProperties(Fortran::lower::AbstractConverter &converter,
   fir::MutableProperties mutableProperties;
   std::string name = converter.mangleName(sym);
   mlir::Type baseAddrTy = converter.genType(sym);
-  if (auto boxType = baseAddrTy.dyn_cast<fir::BaseBoxType>())
+  if (auto boxType = mlir::dyn_cast<fir::BaseBoxType>(baseAddrTy))
     baseAddrTy = boxType.getEleTy();
   // Allocate and set a variable to hold the address.
   // It will be set to null in setUnallocatedStatus.
@@ -1030,9 +1032,9 @@ createMutableProperties(Fortran::lower::AbstractConverter &converter,
   mlir::Type eleTy = baseAddrTy;
   if (auto newTy = fir::dyn_cast_ptrEleTy(eleTy))
     eleTy = newTy;
-  if (auto seqTy = eleTy.dyn_cast<fir::SequenceType>())
+  if (auto seqTy = mlir::dyn_cast<fir::SequenceType>(eleTy))
     eleTy = seqTy.getEleTy();
-  if (auto record = eleTy.dyn_cast<fir::RecordType>())
+  if (auto record = mlir::dyn_cast<fir::RecordType>(eleTy))
     if (record.getNumLenParams() != 0)
       TODO(loc, "deferred length type parameters.");
   if (fir::isa_char(eleTy) && nonDeferredParams.empty()) {
diff --git a/flang/lib/Lower/Bridge.cpp b/flang/lib/Lower/Bridge.cpp
index 8b62fe8c022f..b42909eaaacc 100644
--- a/flang/lib/Lower/Bridge.cpp
+++ b/flang/lib/Lower/Bridge.cpp
@@ -36,6 +36,7 @@
 #include "flang/Optimizer/Builder/Runtime/Character.h"
 #include "flang/Optimizer/Builder/Runtime/Derived.h"
 #include "flang/Optimizer/Builder/Runtime/EnvironmentDefaults.h"
+#include "flang/Optimizer/Builder/Runtime/Main.h"
 #include "flang/Optimizer/Builder/Runtime/Ragged.h"
 #include "flang/Optimizer/Builder/Runtime/Stop.h"
 #include "flang/Optimizer/Builder/Todo.h"
@@ -359,8 +360,10 @@ public:
         // not need to be generated even if no defaults are specified.
         // However, generating main or changing when the runtime reads
         // environment variables is required to do so.
-        fir::runtime::genEnvironmentDefaults(*builder, toLocation(),
-                                             bridge.getEnvironmentDefaults());
+        auto env = fir::runtime::genEnvironmentDefaults(
+            *builder, toLocation(), bridge.getEnvironmentDefaults());
+
+        fir::runtime::genMain(*builder, toLocation(), env);
       });
 
     finalizeOpenACCLowering();
@@ -683,7 +686,7 @@ public:
           auto if_builder = builder->genIfThenElse(loc, isAllocated);
           if_builder.genThen([&]() {
             std::string name = mangleName(sym) + ".alloc";
-            if (auto seqTy = symType.dyn_cast<fir::SequenceType>()) {
+            if (auto seqTy = mlir::dyn_cast<fir::SequenceType>(symType)) {
               fir::ExtendedValue read = fir::factory::genMutableBoxRead(
                   *builder, loc, box, /*mayBePolymorphic=*/false);
               if (auto read_arr_box = read.getBoxOf<fir::ArrayBoxValue>()) {
@@ -1132,7 +1135,7 @@ private:
     fir::ExtendedValue lhs = symBoxToExtendedValue(lhs_sb);
     fir::ExtendedValue rhs = symBoxToExtendedValue(rhs_sb);
     mlir::Type symType = genType(sym);
-    if (auto seqTy = symType.dyn_cast<fir::SequenceType>()) {
+    if (auto seqTy = mlir::dyn_cast<fir::SequenceType>(symType)) {
       Fortran::lower::StatementContext stmtCtx;
       Fortran::lower::createSomeArrayAssignment(*this, lhs, rhs, localSymbols,
                                                 stmtCtx);
@@ -1355,7 +1358,7 @@ private:
       return;
     }
     mlir::Type selectorType = selector.getType();
-    bool realSelector = selectorType.isa<mlir::FloatType>();
+    bool realSelector = mlir::isa<mlir::FloatType>(selectorType);
     assert((inArithmeticIfContext || !realSelector) && "invalid selector type");
     mlir::Value zero;
     if (inArithmeticIfContext)
@@ -1630,7 +1633,7 @@ private:
         stmtCtx);
     stmtCtx.finalizeAndReset();
     // Raise an exception if REAL expr is a NaN.
-    if (expr.getType().isa<mlir::FloatType>())
+    if (mlir::isa<mlir::FloatType>(expr.getType()))
       expr = builder->create<mlir::arith::AddFOp>(toLocation(), expr, expr);
     // An empty valueList indicates to genMultiwayBranch that the branch is
     // an ArithmeticIfStmt that has two branches on value 0 or 0.0.
@@ -2807,7 +2810,7 @@ private:
     auto caseValue = valueList.begin();
     auto caseBlock = blockList.begin();
     for (mlir::Attribute attr : attrList) {
-      if (attr.isa<mlir::UnitAttr>()) {
+      if (mlir::isa<mlir::UnitAttr>(attr)) {
         genBranch(*caseBlock++);
         break;
       }
@@ -2825,7 +2828,7 @@ private:
                                             rhsVal.second);
       };
       mlir::Block *newBlock = insertBlock(*caseBlock);
-      if (attr.isa<fir::ClosedIntervalAttr>()) {
+      if (mlir::isa<fir::ClosedIntervalAttr>(attr)) {
         mlir::Block *newBlock2 = insertBlock(*caseBlock);
         mlir::Value cond =
             genCond(*caseValue++, mlir::arith::CmpIPredicate::sge);
@@ -2838,12 +2841,12 @@ private:
         continue;
       }
       mlir::arith::CmpIPredicate pred;
-      if (attr.isa<fir::PointIntervalAttr>()) {
+      if (mlir::isa<fir::PointIntervalAttr>(attr)) {
         pred = mlir::arith::CmpIPredicate::eq;
-      } else if (attr.isa<fir::LowerBoundAttr>()) {
+      } else if (mlir::isa<fir::LowerBoundAttr>(attr)) {
         pred = mlir::arith::CmpIPredicate::sge;
       } else {
-        assert(attr.isa<fir::UpperBoundAttr>() && "unexpected predicate");
+        assert(mlir::isa<fir::UpperBoundAttr>(attr) && "unexpected predicate");
         pred = mlir::arith::CmpIPredicate::sle;
       }
       mlir::Value cond = genCond(*caseValue++, pred);
@@ -3105,7 +3108,7 @@ private:
         bool isPointer = fir::isPointerType(baseTy);
         bool isAllocatable = fir::isAllocatableType(baseTy);
         bool isArray =
-            fir::dyn_cast_ptrOrBoxEleTy(baseTy).isa<fir::SequenceType>();
+            mlir::isa<fir::SequenceType>(fir::dyn_cast_ptrOrBoxEleTy(baseTy));
         const fir::BoxValue *selectorBox = selector.getBoxOf<fir::BoxValue>();
         if (std::holds_alternative<Fortran::parser::Default>(guard.u)) {
           // CLASS DEFAULT
@@ -3114,12 +3117,12 @@ private:
                        std::get_if<Fortran::parser::TypeSpec>(&guard.u)) {
           // TYPE IS
           fir::ExactTypeAttr attr =
-              typeGuardAttr.dyn_cast<fir::ExactTypeAttr>();
+              mlir::dyn_cast<fir::ExactTypeAttr>(typeGuardAttr);
           mlir::Value exactValue;
           mlir::Type addrTy = attr.getType();
           if (isArray) {
-            auto seqTy = fir::dyn_cast_ptrOrBoxEleTy(baseTy)
-                             .dyn_cast<fir::SequenceType>();
+            auto seqTy = mlir::dyn_cast<fir::SequenceType>(
+                fir::dyn_cast_ptrOrBoxEleTy(baseTy));
             addrTy = fir::SequenceType::get(seqTy.getShape(), attr.getType());
           }
           if (isPointer)
@@ -3141,7 +3144,7 @@ private:
               addAssocEntitySymbol(selectorBox->clone(exact));
             } else if (intrinsic->category() ==
                        Fortran::common::TypeCategory::Character) {
-              auto charTy = attr.getType().dyn_cast<fir::CharacterType>();
+              auto charTy = mlir::dyn_cast<fir::CharacterType>(attr.getType());
               mlir::Value charLen =
                   fir::factory::CharacterExprHelper(*builder, loc)
                       .readLengthFromBox(fir::getBase(selector), charTy);
@@ -3158,11 +3161,12 @@ private:
         } else if (std::holds_alternative<Fortran::parser::DerivedTypeSpec>(
                        guard.u)) {
           // CLASS IS
-          fir::SubclassAttr attr = typeGuardAttr.dyn_cast<fir::SubclassAttr>();
+          fir::SubclassAttr attr =
+              mlir::dyn_cast<fir::SubclassAttr>(typeGuardAttr);
           mlir::Type addrTy = attr.getType();
           if (isArray) {
-            auto seqTy = fir::dyn_cast_ptrOrBoxEleTy(baseTy)
-                             .dyn_cast<fir::SequenceType>();
+            auto seqTy = mlir::dyn_cast<fir::SequenceType>(
+                fir::dyn_cast_ptrOrBoxEleTy(baseTy));
             addrTy = fir::SequenceType::get(seqTy.getShape(), attr.getType());
           }
           if (isPointer)
@@ -3794,7 +3798,9 @@ private:
           auto needCleanup = fir::getIntIfConstant(cleanup);
           if (needCleanup && *needCleanup)
             temps.push_back(temp);
-          addSymbol(sym, temp, /*forced=*/true);
+          addSymbol(sym,
+                    hlfir::translateToExtendedValue(loc, builder, temp).first,
+                    /*forced=*/true);
           builder.create<fir::CUDADataTransferOp>(loc, addr, temp,
                                                   transferKindAttr);
           ++nbDeviceResidentObject;
@@ -3804,18 +3810,38 @@ private:
     return temps;
   }
 
+  // Check if the insertion point is currently in a device context. HostDevice
+  // subprogram are not considered fully device context so it will return false
+  // for it.
+  static bool isDeviceContext(fir::FirOpBuilder &builder) {
+    if (builder.getRegion().getParentOfType<fir::CUDAKernelOp>())
+      return true;
+    if (auto funcOp =
+            builder.getRegion().getParentOfType<mlir::func::FuncOp>()) {
+      if (auto cudaProcAttr =
+              funcOp.getOperation()->getAttrOfType<fir::CUDAProcAttributeAttr>(
+                  fir::getCUDAAttrName())) {
+        return cudaProcAttr.getValue() != fir::CUDAProcAttribute::Host &&
+               cudaProcAttr.getValue() != fir::CUDAProcAttribute::HostDevice;
+      }
+    }
+    return false;
+  }
+
   void genDataAssignment(
       const Fortran::evaluate::Assignment &assign,
       const Fortran::evaluate::ProcedureRef *userDefinedAssignment) {
     mlir::Location loc = getCurrentLocation();
     fir::FirOpBuilder &builder = getFirOpBuilder();
 
-    bool isCUDATransfer = Fortran::evaluate::HasCUDAAttrs(assign.lhs) ||
-                          Fortran::evaluate::HasCUDAAttrs(assign.rhs);
+    bool isInDeviceContext = isDeviceContext(builder);
+    bool isCUDATransfer = (Fortran::evaluate::HasCUDAAttrs(assign.lhs) ||
+                           Fortran::evaluate::HasCUDAAttrs(assign.rhs)) &&
+                          !isInDeviceContext;
     bool hasCUDAImplicitTransfer =
         Fortran::evaluate::HasCUDAImplicitTransfer(assign.rhs);
     llvm::SmallVector<mlir::Value> implicitTemps;
-    if (hasCUDAImplicitTransfer)
+    if (hasCUDAImplicitTransfer && !isInDeviceContext)
       implicitTemps = genCUDAImplicitDataTransfer(builder, loc, assign);
 
     // Gather some information about the assignment that will impact how it is
@@ -3880,7 +3906,7 @@ private:
         builder.create<hlfir::AssignOp>(loc, rhs, lhs,
                                         isWholeAllocatableAssignment,
                                         keepLhsLengthInAllocatableAssignment);
-      if (hasCUDAImplicitTransfer) {
+      if (hasCUDAImplicitTransfer && !isInDeviceContext) {
         localSymbols.popScope();
         for (mlir::Value temp : implicitTemps)
           builder.create<fir::FreeMemOp>(loc, temp);
@@ -4135,7 +4161,7 @@ private:
               } else if (isDerivedCategory(lhsType->category())) {
                 // Handle parent component.
                 if (Fortran::lower::isParentComponent(assign.lhs)) {
-                  if (!fir::getBase(lhs).getType().isa<fir::BaseBoxType>())
+                  if (!mlir::isa<fir::BaseBoxType>(fir::getBase(lhs).getType()))
                     lhs = fir::getBase(builder->createBox(loc, lhs));
                   lhs = Fortran::lower::updateBoxForParentComponent(*this, lhs,
                                                                     assign.lhs);
@@ -5486,7 +5512,7 @@ Fortran::lower::LoweringBridge::LoweringBridge(
     default:
       break;
     }
-    if (!diag.getLocation().isa<mlir::UnknownLoc>())
+    if (!mlir::isa<mlir::UnknownLoc>(diag.getLocation()))
       os << diag.getLocation() << ": ";
     os << diag << '\n';
     os.flush();
diff --git a/flang/lib/Lower/CallInterface.cpp b/flang/lib/Lower/CallInterface.cpp
index 5ad244600328..c1f54ad39287 100644
--- a/flang/lib/Lower/CallInterface.cpp
+++ b/flang/lib/Lower/CallInterface.cpp
@@ -1182,7 +1182,7 @@ private:
       Property prop = Property::BaseAddress;
       if (isValueAttr) {
         bool isBuiltinCptrType = fir::isa_builtin_cptr_type(type);
-        if (isBindC || (!type.isa<fir::SequenceType>() &&
+        if (isBindC || (!mlir::isa<fir::SequenceType>(type) &&
                         !obj.attrs.test(Attrs::Optional) &&
                         (dynamicType.category() !=
                              Fortran::common::TypeCategory::Derived ||
@@ -1190,7 +1190,7 @@ private:
           passBy = PassEntityBy::Value;
           prop = Property::Value;
           if (isBuiltinCptrType) {
-            auto recTy = type.dyn_cast<fir::RecordType>();
+            auto recTy = mlir::dyn_cast<fir::RecordType>(type);
             mlir::Type fieldTy = recTy.getTypeList()[0].second;
             passType = fir::ReferenceType::get(fieldTy);
           } else {
@@ -1714,7 +1714,7 @@ mlir::Type Fortran::lower::getDummyProcedureType(
 }
 
 bool Fortran::lower::isCPtrArgByValueType(mlir::Type ty) {
-  return ty.isa<fir::ReferenceType>() &&
+  return mlir::isa<fir::ReferenceType>(ty) &&
          fir::isa_integer(fir::unwrapRefType(ty));
 }
 
diff --git a/flang/lib/Lower/ConvertArrayConstructor.cpp b/flang/lib/Lower/ConvertArrayConstructor.cpp
index 24aa9beba6bf..a5b5838fe6b6 100644
--- a/flang/lib/Lower/ConvertArrayConstructor.cpp
+++ b/flang/lib/Lower/ConvertArrayConstructor.cpp
@@ -336,7 +336,7 @@ public:
       if (!extent)
         extent = builder.createIntegerConstant(loc, builder.getIndexType(), 0);
       if (missingLengthParameters) {
-        if (declaredType.getEleTy().isa<fir::CharacterType>())
+        if (mlir::isa<fir::CharacterType>(declaredType.getEleTy()))
           emboxLengths.push_back(builder.createIntegerConstant(
               loc, builder.getCharacterLengthType(), 0));
         else
@@ -357,7 +357,7 @@ public:
 
   bool useSimplePushRuntime(hlfir::Entity value) {
     return value.isScalar() &&
-           !arrayConstructorElementType.isa<fir::CharacterType>() &&
+           !mlir::isa<fir::CharacterType>(arrayConstructorElementType) &&
            !fir::isRecordWithAllocatableMember(arrayConstructorElementType) &&
            !fir::isRecordWithTypeParameters(arrayConstructorElementType);
   }
@@ -370,7 +370,7 @@ public:
       auto [addrExv, cleanUp] = hlfir::convertToAddress(
           loc, builder, value, arrayConstructorElementType);
       mlir::Value addr = fir::getBase(addrExv);
-      if (addr.getType().isa<fir::BaseBoxType>())
+      if (mlir::isa<fir::BaseBoxType>(addr.getType()))
         addr = builder.create<fir::BoxAddrOp>(loc, addr);
       fir::runtime::genPushArrayConstructorSimpleScalar(
           loc, builder, arrayConstructorVector, addr);
@@ -564,7 +564,7 @@ struct LengthAndTypeCollector<Character<Kind>> {
 /// lowering an ac-value and must be delayed?
 static bool missingLengthParameters(mlir::Type elementType,
                                     llvm::ArrayRef<mlir::Value> lengths) {
-  return (elementType.isa<fir::CharacterType>() ||
+  return (mlir::isa<fir::CharacterType>(elementType) ||
           fir::isRecordWithTypeParameters(elementType)) &&
          lengths.empty();
 }
@@ -702,7 +702,8 @@ static ArrayCtorLoweringStrategy selectArrayCtorLoweringStrategy(
   // Based on what was gathered and the result of the analysis, select and
   // instantiate the right lowering strategy for the array constructor.
   if (!extent || needToEvaluateOneExprToGetLengthParameters ||
-      analysis.anyArrayExpr || declaredType.getEleTy().isa<fir::RecordType>())
+      analysis.anyArrayExpr ||
+      mlir::isa<fir::RecordType>(declaredType.getEleTy()))
     return RuntimeTempStrategy(
         loc, builder, stmtCtx, symMap, declaredType,
         extent ? std::optional<mlir::Value>(extent) : std::nullopt, lengths,
diff --git a/flang/lib/Lower/ConvertCall.cpp b/flang/lib/Lower/ConvertCall.cpp
index c6f7d3410ad5..e4a0cc8d4730 100644
--- a/flang/lib/Lower/ConvertCall.cpp
+++ b/flang/lib/Lower/ConvertCall.cpp
@@ -49,15 +49,15 @@ static fir::ExtendedValue toExtendedValue(mlir::Location loc, mlir::Value base,
                                           llvm::ArrayRef<mlir::Value> extents,
                                           llvm::ArrayRef<mlir::Value> lengths) {
   mlir::Type type = base.getType();
-  if (type.isa<fir::BaseBoxType>())
+  if (mlir::isa<fir::BaseBoxType>(type))
     return fir::BoxValue(base, /*lbounds=*/{}, lengths, extents);
   type = fir::unwrapRefType(type);
-  if (type.isa<fir::BaseBoxType>())
+  if (mlir::isa<fir::BaseBoxType>(type))
     return fir::MutableBoxValue(base, lengths, /*mutableProperties*/ {});
-  if (auto seqTy = type.dyn_cast<fir::SequenceType>()) {
+  if (auto seqTy = mlir::dyn_cast<fir::SequenceType>(type)) {
     if (seqTy.getDimension() != extents.size())
       fir::emitFatalError(loc, "incorrect number of extents for array");
-    if (seqTy.getEleTy().isa<fir::CharacterType>()) {
+    if (mlir::isa<fir::CharacterType>(seqTy.getEleTy())) {
       if (lengths.empty())
         fir::emitFatalError(loc, "missing length for character");
       assert(lengths.size() == 1);
@@ -65,7 +65,7 @@ static fir::ExtendedValue toExtendedValue(mlir::Location loc, mlir::Value base,
     }
     return fir::ArrayBoxValue(base, extents);
   }
-  if (type.isa<fir::CharacterType>()) {
+  if (mlir::isa<fir::CharacterType>(type)) {
     if (lengths.empty())
       fir::emitFatalError(loc, "missing length for character");
     assert(lengths.size() == 1);
@@ -193,7 +193,7 @@ static mlir::Value remapActualToDummyDescriptor(
   llvm::SmallVector<mlir::Value> lengths;
   mlir::Type dummyBoxType = caller.getDummyArgumentType(arg);
   mlir::Type dummyBaseType = fir::unwrapPassByRefType(dummyBoxType);
-  if (dummyBaseType.isa<fir::SequenceType>())
+  if (mlir::isa<fir::SequenceType>(dummyBaseType))
     caller.walkDummyArgumentExtents(
         arg, [&](const Fortran::lower::SomeExpr &e, bool isAssumedSizeExtent) {
           extents.emplace_back(lowerSpecExpr(e, isAssumedSizeExtent));
@@ -338,7 +338,7 @@ std::pair<fir::ExtendedValue, bool> Fortran::lower::genCallOpAndResult(
     if (!caller.callerAllocateResult())
       return {};
     mlir::Type type = caller.getResultStorageType();
-    if (type.isa<fir::SequenceType>())
+    if (mlir::isa<fir::SequenceType>(type))
       caller.walkResultExtents(
           [&](const Fortran::lower::SomeExpr &e, bool isAssumedSizeExtent) {
             assert(!isAssumedSizeExtent && "result cannot be assumed-size");
@@ -353,7 +353,7 @@ std::pair<fir::ExtendedValue, bool> Fortran::lower::genCallOpAndResult(
     // Result length parameters should not be provided to box storage
     // allocation and save_results, but they are still useful information to
     // keep in the ExtendedValue if non-deferred.
-    if (!type.isa<fir::BoxType>()) {
+    if (!mlir::isa<fir::BoxType>(type)) {
       if (fir::isa_char(fir::unwrapSequenceType(type)) && lengths.empty()) {
         // Calling an assumed length function. This is only possible if this
         // is a call to a character dummy procedure.
@@ -478,7 +478,7 @@ std::pair<fir::ExtendedValue, bool> Fortran::lower::genCallOpAndResult(
   // FIR.
   if (funcPointer) {
     operands.push_back(
-        funcPointer.getType().isa<fir::BoxProcType>()
+        mlir::isa<fir::BoxProcType>(funcPointer.getType())
             ? builder.create<fir::BoxAddrOp>(loc, funcType, funcPointer)
             : builder.createConvert(loc, funcType, funcPointer));
   }
@@ -492,8 +492,8 @@ std::pair<fir::ExtendedValue, bool> Fortran::lower::genCallOpAndResult(
     // arguments of any type and vice versa.
     mlir::Value cast;
     auto *context = builder.getContext();
-    if (snd.isa<fir::BoxProcType>() &&
-        fst.getType().isa<mlir::FunctionType>()) {
+    if (mlir::isa<fir::BoxProcType>(snd) &&
+        mlir::isa<mlir::FunctionType>(fst.getType())) {
       auto funcTy =
           mlir::FunctionType::get(context, std::nullopt, std::nullopt);
       auto boxProcTy = builder.getBoxProcType(funcTy);
@@ -734,9 +734,9 @@ std::pair<fir::ExtendedValue, bool> Fortran::lower::genCallOpAndResult(
 
   // Call a BIND(C) function that return a char.
   if (caller.characterize().IsBindC() &&
-      funcType.getResults()[0].isa<fir::CharacterType>()) {
+      mlir::isa<fir::CharacterType>(funcType.getResults()[0])) {
     fir::CharacterType charTy =
-        funcType.getResults()[0].dyn_cast<fir::CharacterType>();
+        mlir::dyn_cast<fir::CharacterType>(funcType.getResults()[0]);
     mlir::Value len = builder.createIntegerConstant(
         loc, builder.getCharacterLengthType(), charTy.getLen());
     return {fir::CharBoxValue{callResult, len}, /*resultIsFinalized=*/false};
@@ -890,7 +890,7 @@ extendedValueToHlfirEntity(mlir::Location loc, fir::FirOpBuilder &builder,
   mlir::Type firBaseTy = firBase.getType();
   if (fir::isa_trivial(firBaseTy))
     return hlfir::EntityWithAttributes{firBase};
-  if (auto charTy = firBase.getType().dyn_cast<fir::CharacterType>()) {
+  if (auto charTy = mlir::dyn_cast<fir::CharacterType>(firBase.getType())) {
     // CHAR() intrinsic and BIND(C) procedures returning CHARACTER(1)
     // are lowered to a fir.char<kind,1> that is not in memory.
     // This tends to cause a lot of bugs because the rest of the
@@ -1061,7 +1061,7 @@ static hlfir::Entity fixProcedureDummyMismatch(mlir::Location loc,
                                                fir::FirOpBuilder &builder,
                                                hlfir::Entity actual,
                                                mlir::Type dummyType) {
-  if (actual.getType().isa<fir::BoxProcType>() &&
+  if (mlir::isa<fir::BoxProcType>(actual.getType()) &&
       fir::isCharacterProcedureTuple(dummyType)) {
     mlir::Value length =
         builder.create<fir::UndefOp>(loc, builder.getCharacterLengthType());
@@ -1070,7 +1070,7 @@ static hlfir::Entity fixProcedureDummyMismatch(mlir::Location loc,
     return hlfir::Entity{tuple};
   }
   assert(fir::isCharacterProcedureTuple(actual.getType()) &&
-         dummyType.isa<fir::BoxProcType>() &&
+         mlir::isa<fir::BoxProcType>(dummyType) &&
          "unsupported dummy procedure mismatch with the actual argument");
   mlir::Value boxProc = fir::factory::extractCharacterProcedureTuple(
                             builder, loc, actual, /*openBoxProc=*/false)
@@ -1143,7 +1143,7 @@ static PreparedDummyArgument preparePresentUserCallActualArgument(
     assert(actual.isProcedure());
     // Do nothing if this is a procedure argument. It is already a
     // fir.boxproc/fir.tuple<fir.boxproc, len> as it should.
-    if (!actual.getType().isa<fir::BoxProcType>() &&
+    if (!mlir::isa<fir::BoxProcType>(actual.getType()) &&
         actual.getType() != dummyType)
       // The actual argument may be a procedure that returns character (a
       // fir.tuple<fir.boxproc, len>) while the dummy is not. Extract the tuple
@@ -1164,7 +1164,7 @@ static PreparedDummyArgument preparePresentUserCallActualArgument(
   // dynamic type matters to determine the contiguity.
   const bool mustSetDynamicTypeToDummyType =
       passingPolymorphicToNonPolymorphic &&
-      (actual.isArray() || dummyType.isa<fir::BaseBoxType>());
+      (actual.isArray() || mlir::isa<fir::BaseBoxType>(dummyType));
 
   // The simple contiguity of the actual is "lost" when passing a polymorphic
   // to a non polymorphic entity because the dummy dynamic type matters for
@@ -1236,7 +1236,7 @@ static PreparedDummyArgument preparePresentUserCallActualArgument(
       preparedDummy.pushExprAssociateCleanUp(associate);
     } else if (mustDoCopyInOut) {
       // Copy-in non contiguous variables.
-      assert(entity.getType().isa<fir::BaseBoxType>() &&
+      assert(mlir::isa<fir::BaseBoxType>(entity.getType()) &&
              "expect non simply contiguous variables to be boxes");
       if (actualIsAssumedRank)
         TODO(loc, "copy-in and copy-out of assumed-rank arguments");
@@ -1294,13 +1294,14 @@ static PreparedDummyArgument preparePresentUserCallActualArgument(
   // Step 3: now that the dummy argument storage has been prepared, package
   // it according to the interface.
   mlir::Value addr;
-  if (dummyTypeWithActualRank.isa<fir::BoxCharType>()) {
+  if (mlir::isa<fir::BoxCharType>(dummyTypeWithActualRank)) {
     addr = hlfir::genVariableBoxChar(loc, builder, entity);
-  } else if (dummyTypeWithActualRank.isa<fir::BaseBoxType>()) {
+  } else if (mlir::isa<fir::BaseBoxType>(dummyTypeWithActualRank)) {
     entity = hlfir::genVariableBox(loc, builder, entity);
     // Ensures the box has the right attributes and that it holds an
     // addendum if needed.
-    fir::BaseBoxType actualBoxType = entity.getType().cast<fir::BaseBoxType>();
+    fir::BaseBoxType actualBoxType =
+        mlir::cast<fir::BaseBoxType>(entity.getType());
     mlir::Type boxEleType = actualBoxType.getEleTy();
     // For now, assume it is not OK to pass the allocatable/pointer
     // descriptor to a non pointer/allocatable dummy. That is a strict
@@ -1567,7 +1568,7 @@ genUserCall(Fortran::lower::PreparedActualArguments &loweredActuals,
         // callee side, and it is illegal to use NULL without a MOLD if any
         // dummy length parameters are assumed.
         mlir::Type boxTy = fir::dyn_cast_ptrEleTy(argTy);
-        assert(boxTy && boxTy.isa<fir::BaseBoxType>() &&
+        assert(boxTy && mlir::isa<fir::BaseBoxType>(boxTy) &&
                "must be a fir.box type");
         mlir::Value boxStorage =
             fir::factory::genNullBoxStorage(builder, loc, boxTy);
@@ -1635,7 +1636,8 @@ genUserCall(Fortran::lower::PreparedActualArguments &loweredActuals,
       caller, callSiteType, callContext.resultType,
       callContext.isElementalProcWithArrayArgs());
   // For procedure pointer function result, just return the call.
-  if (callContext.resultType && callContext.resultType->isa<fir::BoxProcType>())
+  if (callContext.resultType &&
+      mlir::isa<fir::BoxProcType>(*callContext.resultType))
     return hlfir::EntityWithAttributes(fir::getBase(result));
 
   /// Clean-up associations and copy-in.
@@ -2115,9 +2117,9 @@ public:
         hlfir::getFortranElementType(*callContext.resultType);
     // Get result length parameters.
     llvm::SmallVector<mlir::Value> typeParams;
-    if (elementType.isa<fir::CharacterType>() ||
+    if (mlir::isa<fir::CharacterType>(elementType) ||
         fir::isRecordWithTypeParameters(elementType)) {
-      auto charType = elementType.dyn_cast<fir::CharacterType>();
+      auto charType = mlir::dyn_cast<fir::CharacterType>(elementType);
       if (charType && charType.hasConstantLen())
         typeParams.push_back(builder.createIntegerConstant(
             loc, builder.getIndexType(), charType.getLen()));
@@ -2523,7 +2525,7 @@ genIntrinsicRef(const Fortran::evaluate::SpecificIntrinsic *intrinsic,
   }
   std::optional<hlfir::EntityWithAttributes> result = genHLFIRIntrinsicRefCore(
       loweredActuals, intrinsic, argLowering, callContext);
-  if (result && result->getType().isa<hlfir::ExprType>()) {
+  if (result && mlir::isa<hlfir::ExprType>(result->getType())) {
     fir::FirOpBuilder *bldr = &callContext.getBuilder();
     callContext.stmtCtx.attachCleanup(
         [=]() { bldr->create<hlfir::DestroyOp>(loc, *result); });
diff --git a/flang/lib/Lower/ConvertConstant.cpp b/flang/lib/Lower/ConvertConstant.cpp
index ed389bbe4ae5..653e874a969c 100644
--- a/flang/lib/Lower/ConvertConstant.cpp
+++ b/flang/lib/Lower/ConvertConstant.cpp
@@ -184,8 +184,8 @@ private:
     if (!attributeElementType || attributes.empty())
       return {};
 
-    assert(symTy.isa<fir::SequenceType>() && "expecting an array global");
-    auto arrTy = symTy.cast<fir::SequenceType>();
+    assert(mlir::isa<fir::SequenceType>(symTy) && "expecting an array global");
+    auto arrTy = mlir::cast<fir::SequenceType>(symTy);
     llvm::SmallVector<int64_t> tensorShape(arrTy.getShape());
     std::reverse(tensorShape.begin(), tensorShape.end());
     auto tensorTy =
@@ -423,14 +423,14 @@ static mlir::Value genStructureComponentInit(
     // address field, which ought to be an intptr_t on the target.
     mlir::Value addr = fir::getBase(
         Fortran::lower::genExtAddrInInitializer(converter, loc, expr));
-    if (addr.getType().isa<fir::BoxProcType>())
+    if (mlir::isa<fir::BoxProcType>(addr.getType()))
       addr = builder.create<fir::BoxAddrOp>(loc, addr);
     assert((fir::isa_ref_type(addr.getType()) ||
-            addr.getType().isa<mlir::FunctionType>()) &&
+            mlir::isa<mlir::FunctionType>(addr.getType())) &&
            "expect reference type for address field");
     assert(fir::isa_derived(componentTy) &&
            "expect C_PTR, C_FUNPTR to be a record");
-    auto cPtrRecTy = componentTy.cast<fir::RecordType>();
+    auto cPtrRecTy = mlir::cast<fir::RecordType>(componentTy);
     llvm::StringRef addrFieldName = Fortran::lower::builtin::cptrFieldName;
     mlir::Type addrFieldTy = cPtrRecTy.getType(addrFieldName);
     auto addrField = builder.create<fir::FieldIndexOp>(
@@ -460,7 +460,7 @@ static mlir::Value genInlinedStructureCtorLitImpl(
     Fortran::lower::AbstractConverter &converter, mlir::Location loc,
     const Fortran::evaluate::StructureConstructor &ctor, mlir::Type type) {
   fir::FirOpBuilder &builder = converter.getFirOpBuilder();
-  auto recTy = type.cast<fir::RecordType>();
+  auto recTy = mlir::cast<fir::RecordType>(type);
 
   if (!converter.getLoweringOptions().getLowerToHighLevelFIR()) {
     mlir::Value res = builder.create<fir::UndefOp>(loc, recTy);
@@ -587,7 +587,7 @@ genInlinedArrayLit(Fortran::lower::AbstractConverter &converter,
     } while (con.IncrementSubscripts(subscripts));
   } else if constexpr (T::category == Fortran::common::TypeCategory::Derived) {
     do {
-      mlir::Type eleTy = arrayTy.cast<fir::SequenceType>().getEleTy();
+      mlir::Type eleTy = mlir::cast<fir::SequenceType>(arrayTy).getEleTy();
       mlir::Value elementVal =
           genScalarLit(converter, loc, con.At(subscripts), eleTy,
                        /*outlineInReadOnlyMemory=*/false);
@@ -597,7 +597,7 @@ genInlinedArrayLit(Fortran::lower::AbstractConverter &converter,
   } else {
     llvm::SmallVector<mlir::Attribute> rangeStartIdx;
     uint64_t rangeSize = 0;
-    mlir::Type eleTy = arrayTy.cast<fir::SequenceType>().getEleTy();
+    mlir::Type eleTy = mlir::cast<fir::SequenceType>(arrayTy).getEleTy();
     do {
       auto getElementVal = [&]() {
         return builder.createConvert(loc, eleTy,
@@ -620,12 +620,11 @@ genInlinedArrayLit(Fortran::lower::AbstractConverter &converter,
         llvm::SmallVector<int64_t> rangeBounds;
         llvm::SmallVector<mlir::Attribute> idx = createIdx();
         for (size_t i = 0; i < idx.size(); ++i) {
-          rangeBounds.push_back(rangeStartIdx[i]
-                                    .cast<mlir::IntegerAttr>()
+          rangeBounds.push_back(mlir::cast<mlir::IntegerAttr>(rangeStartIdx[i])
                                     .getValue()
                                     .getSExtValue());
           rangeBounds.push_back(
-              idx[i].cast<mlir::IntegerAttr>().getValue().getSExtValue());
+              mlir::cast<mlir::IntegerAttr>(idx[i]).getValue().getSExtValue());
         }
         array = builder.create<fir::InsertOnRangeOp>(
             loc, arrayTy, array, getElementVal(),
@@ -647,7 +646,7 @@ genOutlineArrayLit(Fortran::lower::AbstractConverter &converter,
                    mlir::Location loc, mlir::Type arrayTy,
                    const Fortran::evaluate::Constant<T> &constant) {
   fir::FirOpBuilder &builder = converter.getFirOpBuilder();
-  mlir::Type eleTy = arrayTy.cast<fir::SequenceType>().getEleTy();
+  mlir::Type eleTy = mlir::cast<fir::SequenceType>(arrayTy).getEleTy();
   llvm::StringRef globalName = converter.getUniqueLitName(
       loc, std::make_unique<Fortran::lower::SomeExpr>(toEvExpr(constant)),
       eleTy);
diff --git a/flang/lib/Lower/ConvertExpr.cpp b/flang/lib/Lower/ConvertExpr.cpp
index fb7807718ff8..9567685aa3d2 100644
--- a/flang/lib/Lower/ConvertExpr.cpp
+++ b/flang/lib/Lower/ConvertExpr.cpp
@@ -267,7 +267,7 @@ arrayLoadExtValue(fir::FirOpBuilder &builder, mlir::Location loc,
     mlir::Type ty = fir::applyPathToType(arrTy, path);
     if (!ty)
       fir::emitFatalError(loc, "path does not apply to type");
-    if (!ty.isa<fir::SequenceType>()) {
+    if (!mlir::isa<fir::SequenceType>(ty)) {
       if (fir::isa_char(ty)) {
         mlir::Value len = newLen;
         if (!len)
@@ -282,7 +282,7 @@ arrayLoadExtValue(fir::FirOpBuilder &builder, mlir::Location loc,
       }
       return newBase;
     }
-    arrTy = ty.cast<fir::SequenceType>();
+    arrTy = mlir::cast<fir::SequenceType>(ty);
   }
 
   auto arrayToExtendedValue =
@@ -412,15 +412,15 @@ static fir::ExtendedValue genLoad(fir::FirOpBuilder &builder,
   return addr.match(
       [](const fir::CharBoxValue &box) -> fir::ExtendedValue { return box; },
       [&](const fir::PolymorphicValue &p) -> fir::ExtendedValue {
-        if (fir::unwrapRefType(fir::getBase(p).getType())
-                .isa<fir::RecordType>())
+        if (mlir::isa<fir::RecordType>(
+                fir::unwrapRefType(fir::getBase(p).getType())))
           return p;
         mlir::Value load = builder.create<fir::LoadOp>(loc, fir::getBase(p));
         return fir::PolymorphicValue(load, p.getSourceBox());
       },
       [&](const fir::UnboxedValue &v) -> fir::ExtendedValue {
-        if (fir::unwrapRefType(fir::getBase(v).getType())
-                .isa<fir::RecordType>())
+        if (mlir::isa<fir::RecordType>(
+                fir::unwrapRefType(fir::getBase(v).getType())))
           return v;
         return builder.create<fir::LoadOp>(loc, fir::getBase(v));
       },
@@ -536,8 +536,8 @@ static mlir::Value
 createBoxProcCharTuple(Fortran::lower::AbstractConverter &converter,
                        mlir::Type argTy, mlir::Value funcAddr,
                        mlir::Value charLen) {
-  auto boxTy =
-      argTy.cast<mlir::TupleType>().getType(0).cast<fir::BoxProcType>();
+  auto boxTy = mlir::cast<fir::BoxProcType>(
+      mlir::cast<mlir::TupleType>(argTy).getType(0));
   mlir::Location loc = converter.getCurrentLocation();
   auto &builder = converter.getFirOpBuilder();
 
@@ -549,7 +549,7 @@ createBoxProcCharTuple(Fortran::lower::AbstractConverter &converter,
   mlir::Type toTy = boxTy.getEleTy();
   if (fir::isa_ref_type(fromTy))
     funcAddr = builder.createConvert(loc, toTy, funcAddr);
-  else if (fromTy.isa<fir::BoxProcType>())
+  else if (mlir::isa<fir::BoxProcType>(fromTy))
     funcAddr = builder.create<fir::BoxAddrOp>(loc, toTy, funcAddr);
 
   auto boxProc = [&]() -> mlir::Value {
@@ -575,7 +575,7 @@ absentBoxToUnallocatedBox(fir::FirOpBuilder &builder, mlir::Location loc,
                           mlir::Value isPresent) {
   mlir::Value box = fir::getBase(exv);
   mlir::Type boxType = box.getType();
-  assert(boxType.isa<fir::BoxType>() && "argument must be a fir.box");
+  assert(mlir::isa<fir::BoxType>(boxType) && "argument must be a fir.box");
   mlir::Value emptyBox =
       fir::factory::createUnallocatedBox(builder, loc, boxType, std::nullopt);
   auto safeToReadBox =
@@ -915,7 +915,7 @@ public:
     if (inInitializer)
       return Fortran::lower::genInlinedStructureCtorLit(converter, loc, ctor);
     mlir::Type ty = translateSomeExprToFIRType(converter, toEvExpr(ctor));
-    auto recTy = ty.cast<fir::RecordType>();
+    auto recTy = mlir::cast<fir::RecordType>(ty);
     auto fieldTy = fir::FieldType::get(ty.getContext());
     mlir::Value res = builder.createTemporary(loc, recTy);
     mlir::Value box = builder.createBox(loc, fir::ExtendedValue{res});
@@ -1172,8 +1172,8 @@ public:
     if (!charBox)
       fir::emitFatalError(loc, "expected scalar character");
     mlir::Value charAddr = charBox->getAddr();
-    auto charType =
-        fir::unwrapPassByRefType(charAddr.getType()).cast<fir::CharacterType>();
+    auto charType = mlir::cast<fir::CharacterType>(
+        fir::unwrapPassByRefType(charAddr.getType()));
     if (charType.hasConstantLen()) {
       // Erase previous constant length from the base type.
       fir::CharacterType::LenType newLen = fir::CharacterType::unknownLen();
@@ -1441,7 +1441,7 @@ public:
     auto fldTy = fir::FieldType::get(&converter.getMLIRContext());
     // FIXME: need to thread the LEN type parameters here.
     for (const Fortran::evaluate::Component *field : list) {
-      auto recTy = ty.cast<fir::RecordType>();
+      auto recTy = mlir::cast<fir::RecordType>(ty);
       const Fortran::semantics::Symbol &sym = getLastSym(*field);
       std::string name = converter.getRecordTypeFieldName(sym);
       coorArgs.push_back(builder.create<fir::FieldIndexOp>(
@@ -1478,7 +1478,7 @@ public:
   mlir::Type genSubType(mlir::Type arrTy, unsigned dims) {
     mlir::Type unwrapTy = fir::dyn_cast_ptrOrBoxEleTy(arrTy);
     assert(unwrapTy && "must be a pointer or box type");
-    auto seqTy = unwrapTy.cast<fir::SequenceType>();
+    auto seqTy = mlir::cast<fir::SequenceType>(unwrapTy);
     llvm::ArrayRef<int64_t> shape = seqTy.getShape();
     assert(shape.size() > 0 && "removing columns for sequence sans shape");
     assert(dims <= shape.size() && "removing more columns than exist");
@@ -1550,9 +1550,9 @@ public:
     }
 
     mlir::Type eleTy = fir::dyn_cast_ptrOrBoxEleTy(base.getType());
-    if (auto classTy = eleTy.dyn_cast<fir::ClassType>())
+    if (auto classTy = mlir::dyn_cast<fir::ClassType>(eleTy))
       eleTy = classTy.getEleTy();
-    auto seqTy = eleTy.cast<fir::SequenceType>();
+    auto seqTy = mlir::cast<fir::SequenceType>(eleTy);
     assert(args.size() == seqTy.getDimension());
     mlir::Type ty = builder.getRefType(seqTy.getEleTy());
     auto addr = builder.create<fir::CoordinateOp>(loc, ty, base, args);
@@ -1571,7 +1571,7 @@ public:
     mlir::Location loc = getLoc();
     mlir::Value addr = fir::getBase(array);
     mlir::Type arrTy = fir::dyn_cast_ptrEleTy(addr.getType());
-    auto eleTy = arrTy.cast<fir::SequenceType>().getEleTy();
+    auto eleTy = mlir::cast<fir::SequenceType>(arrTy).getEleTy();
     mlir::Type seqTy = builder.getRefType(builder.getVarLenSeqTy(eleTy));
     mlir::Type refTy = builder.getRefType(eleTy);
     mlir::Value base = builder.createConvert(loc, seqTy, addr);
@@ -1656,7 +1656,7 @@ public:
     mlir::Location loc = getLoc();
     mlir::Value addr = fir::getBase(exv);
     mlir::Type arrTy = fir::dyn_cast_ptrOrBoxEleTy(addr.getType());
-    mlir::Type eleTy = arrTy.cast<fir::SequenceType>().getEleTy();
+    mlir::Type eleTy = mlir::cast<fir::SequenceType>(arrTy).getEleTy();
     mlir::Type refTy = builder.getRefType(eleTy);
     mlir::IndexType idxTy = builder.getIndexType();
     llvm::SmallVector<mlir::Value> arrayCoorArgs;
@@ -1766,8 +1766,9 @@ public:
     mlir::Location loc = getLoc();
     ExtValue exv = genBoxArg(expr);
     auto exvTy = fir::getBase(exv).getType();
-    if (exvTy.isa<mlir::FunctionType>()) {
-      auto boxProcTy = builder.getBoxProcType(exvTy.cast<mlir::FunctionType>());
+    if (mlir::isa<mlir::FunctionType>(exvTy)) {
+      auto boxProcTy =
+          builder.getBoxProcType(mlir::cast<mlir::FunctionType>(exvTy));
       return builder.create<fir::EmboxProcOp>(loc, boxProcTy,
                                               fir::getBase(exv));
     }
@@ -1861,7 +1862,7 @@ public:
           // IS_CONTIGUOUS may require an assumed size TYPE(*) to be passed to
           // the intrinsic library utility as a fir.box.
           if (argRules.lowerAs == fir::LowerIntrinsicArgAs::Box &&
-              !fir::getBase(exv).getType().isa<fir::BaseBoxType>()) {
+              !mlir::isa<fir::BaseBoxType>(fir::getBase(exv).getType())) {
             operands.emplace_back(
                 fir::factory::createBoxValue(builder, loc, exv));
             continue;
@@ -2005,7 +2006,7 @@ public:
         fir::getTypeParams(mold);
     mlir::Value charLen;
     mlir::Type elementType = fir::unwrapSequenceType(type);
-    if (auto charType = elementType.dyn_cast<fir::CharacterType>()) {
+    if (auto charType = mlir::dyn_cast<fir::CharacterType>(elementType)) {
       charLen = allocMemTypeParams.empty()
                     ? fir::factory::readCharLen(builder, loc, mold)
                     : allocMemTypeParams[0];
@@ -2017,7 +2018,7 @@ public:
 
     mlir::Value temp = builder.create<fir::AllocMemOp>(
         loc, type, tempName, allocMemTypeParams, extents);
-    if (fir::unwrapSequenceType(type).isa<fir::CharacterType>())
+    if (mlir::isa<fir::CharacterType>(fir::unwrapSequenceType(type)))
       return fir::CharArrayBoxValue{temp, charLen, extents};
     return fir::ArrayBoxValue{temp, extents};
   }
@@ -2166,7 +2167,7 @@ public:
         // We have to initialize the temp if it may have components
         // that need initialization. If there are no components
         // requiring initialization, then the call is a no-op.
-        if (getElementTypeOf(temp).isa<fir::RecordType>()) {
+        if (mlir::isa<fir::RecordType>(getElementTypeOf(temp))) {
           mlir::Value tempBox = fir::getBase(builder.createBox(loc, temp));
           fir::runtime::genDerivedTypeInitialize(builder, loc, tempBox);
         }
@@ -2312,7 +2313,7 @@ public:
     if (!copyOutPair.restrictCopyAndFreeAtRuntime) {
       doCopyOut();
 
-      if (fir::getElementTypeOf(copyOutPair.temp).isa<fir::RecordType>()) {
+      if (mlir::isa<fir::RecordType>(fir::getElementTypeOf(copyOutPair.temp))) {
         // Destroy components of the temporary (if any).
         // If there are no components requiring destruction, then the call
         // is a no-op.
@@ -2330,7 +2331,8 @@ public:
     builder.genIfThen(loc, *copyOutPair.restrictCopyAndFreeAtRuntime)
         .genThen([&]() {
           doCopyOut();
-          if (fir::getElementTypeOf(copyOutPair.temp).isa<fir::RecordType>()) {
+          if (mlir::isa<fir::RecordType>(
+                  fir::getElementTypeOf(copyOutPair.temp))) {
             // Destroy components of the temporary (if any).
             // If there are no components requiring destruction, then the call
             // is a no-op.
@@ -2381,7 +2383,7 @@ public:
     mlir::Value actualArgBase = fir::getBase(actualArg);
     mlir::Value isPresent = builder.create<fir::IsPresentOp>(
         loc, builder.getI1Type(), actualArgBase);
-    if (!actualArgBase.getType().isa<fir::BoxType>())
+    if (!mlir::isa<fir::BoxType>(actualArgBase.getType()))
       return {actualArg, isPresent};
     ExtValue safeToReadBox =
         absentBoxToUnallocatedBox(builder, loc, actualArg, isPresent);
@@ -2408,7 +2410,7 @@ public:
                                       fir::getAdaptToByRefAttr(builder)});
       return fir::CharBoxValue{temp, len};
     }
-    assert((fir::isa_trivial(type) || type.isa<fir::RecordType>()) &&
+    assert((fir::isa_trivial(type) || mlir::isa<fir::RecordType>(type)) &&
            "must be simple scalar");
     return builder.createTemporary(loc, type,
                                    llvm::ArrayRef<mlir::NamedAttribute>{
@@ -2585,7 +2587,7 @@ public:
           // callee side, and it is illegal to use NULL without a MOLD if any
           // dummy length parameters are assumed.
           mlir::Type boxTy = fir::dyn_cast_ptrEleTy(argTy);
-          assert(boxTy && boxTy.isa<fir::BaseBoxType>() &&
+          assert(boxTy && mlir::isa<fir::BaseBoxType>(boxTy) &&
                  "must be a fir.box type");
           mlir::Value boxStorage = builder.createTemporary(loc, boxTy);
           mlir::Value nullBox = fir::factory::createUnallocatedBox(
@@ -2643,10 +2645,11 @@ public:
                 // If a character procedure was passed instead, handle the
                 // mismatch.
                 auto funcTy =
-                    x.getAddr().getType().dyn_cast<mlir::FunctionType>();
+                    mlir::dyn_cast<mlir::FunctionType>(x.getAddr().getType());
                 if (funcTy && funcTy.getNumResults() == 1 &&
-                    funcTy.getResult(0).isa<fir::BoxCharType>()) {
-                  auto boxTy = funcTy.getResult(0).cast<fir::BoxCharType>();
+                    mlir::isa<fir::BoxCharType>(funcTy.getResult(0))) {
+                  auto boxTy =
+                      mlir::cast<fir::BoxCharType>(funcTy.getResult(0));
                   mlir::Value ref = builder.createConvert(
                       loc, builder.getRefType(boxTy.getEleTy()), x.getAddr());
                   auto len = builder.create<fir::UndefOp>(
@@ -2667,7 +2670,7 @@ public:
                 // free-casting the base address to be a !fir.char reference and
                 // setting the LEN argument to undefined. What could go wrong?
                 auto dataPtr = fir::getBase(x);
-                assert(!dataPtr.getType().template isa<fir::BoxType>());
+                assert(!mlir::isa<fir::BoxType>(dataPtr.getType()));
                 return builder.convertWithSemantics(
                     loc, argTy, dataPtr,
                     /*allowCharacterConversion=*/true);
@@ -2742,7 +2745,7 @@ public:
                 loc,
                 fir::ClassType::get(mlir::NoneType::get(builder.getContext())),
                 box);
-          } else if (box.getType().isa<fir::BoxType>() &&
+          } else if (mlir::isa<fir::BoxType>(box.getType()) &&
                      fir::isPolymorphicType(argTy)) {
             box = builder.create<fir::ReboxOp>(loc, argTy, box, mlir::Value{},
                                                /*slice=*/mlir::Value{});
@@ -2791,7 +2794,7 @@ public:
                       : builder.createBox(getLoc(), genTempExtAddr(*expr),
                                           fir::isPolymorphicType(argTy),
                                           fir::isAssumedType(argTy));
-            if (box.getType().isa<fir::BoxType>() &&
+            if (mlir::isa<fir::BoxType>(box.getType()) &&
                 fir::isPolymorphicType(argTy) && !fir::isAssumedType(argTy)) {
               mlir::Type actualTy = argTy;
               if (Fortran::lower::isParentComponent(*expr))
@@ -3030,10 +3033,11 @@ private:
       Fortran::common::ScopedSet(semant, PushVal);
 
 static bool isAdjustedArrayElementType(mlir::Type t) {
-  return fir::isa_char(t) || fir::isa_derived(t) || t.isa<fir::SequenceType>();
+  return fir::isa_char(t) || fir::isa_derived(t) ||
+         mlir::isa<fir::SequenceType>(t);
 }
 static bool elementTypeWasAdjusted(mlir::Type t) {
-  if (auto ty = t.dyn_cast<fir::ReferenceType>())
+  if (auto ty = mlir::dyn_cast<fir::ReferenceType>(t))
     return isAdjustedArrayElementType(ty.getEleTy());
   return false;
 }
@@ -3050,15 +3054,15 @@ static void genScalarUserDefinedAssignmentCall(fir::FirOpBuilder &builder,
   auto prepareUserDefinedArg =
       [](fir::FirOpBuilder &builder, mlir::Location loc,
          const fir::ExtendedValue &value, mlir::Type argType) -> mlir::Value {
-    if (argType.isa<fir::BoxCharType>()) {
+    if (mlir::isa<fir::BoxCharType>(argType)) {
       const fir::CharBoxValue *charBox = value.getCharBox();
       assert(charBox && "argument type mismatch in elemental user assignment");
       return fir::factory::CharacterExprHelper{builder, loc}.createEmbox(
           *charBox);
     }
-    if (argType.isa<fir::BaseBoxType>()) {
+    if (mlir::isa<fir::BaseBoxType>(argType)) {
       mlir::Value box =
-          builder.createBox(loc, value, argType.isa<fir::ClassType>());
+          builder.createBox(loc, value, mlir::isa<fir::ClassType>(argType));
       return builder.createConvert(loc, argType, box);
     }
     // Simple pass by address.
@@ -3170,7 +3174,7 @@ convertToArrayBoxValue(mlir::Location loc, fir::FirOpBuilder &builder,
                        mlir::Value val, mlir::Value len) {
   mlir::Type ty = fir::unwrapRefType(val.getType());
   mlir::IndexType idxTy = builder.getIndexType();
-  auto seqTy = ty.cast<fir::SequenceType>();
+  auto seqTy = mlir::cast<fir::SequenceType>(ty);
   auto undef = builder.create<fir::UndefOp>(loc, idxTy);
   llvm::SmallVector<mlir::Value> extents(seqTy.getDimension(), undef);
   if (fir::isa_char(seqTy.getEleTy()))
@@ -3462,7 +3466,7 @@ public:
         [&](const auto &e) {
           auto f = genarr(e);
           ExtValue exv = f(IterationSpace{});
-          if (fir::getBase(exv).getType().template isa<fir::BaseBoxType>())
+          if (mlir::isa<fir::BaseBoxType>(fir::getBase(exv).getType()))
             return exv;
           fir::emitFatalError(getLoc(), "array must be emboxed");
         },
@@ -3487,10 +3491,9 @@ public:
                                            tempRes, dest.getSlice(),
                                            dest.getTypeparams());
 
-    auto arrTy =
-        fir::dyn_cast_ptrEleTy(tempRes.getType()).cast<fir::SequenceType>();
-    if (auto charTy =
-            arrTy.getEleTy().template dyn_cast<fir::CharacterType>()) {
+    auto arrTy = mlir::cast<fir::SequenceType>(
+        fir::dyn_cast_ptrEleTy(tempRes.getType()));
+    if (auto charTy = mlir::dyn_cast<fir::CharacterType>(arrTy.getEleTy())) {
       if (fir::characterWithDynamicLen(charTy))
         TODO(loc, "CHARACTER does not have constant LEN");
       mlir::Value len = builder.createIntegerConstant(
@@ -3912,17 +3915,18 @@ private:
   mlir::Value convertElementForUpdate(mlir::Location loc, mlir::Type eleTy,
                                       mlir::Value origVal) {
     if (auto origEleTy = fir::dyn_cast_ptrEleTy(origVal.getType()))
-      if (origEleTy.isa<fir::BaseBoxType>()) {
+      if (mlir::isa<fir::BaseBoxType>(origEleTy)) {
         // If origVal is a box variable, load it so it is in the value domain.
         origVal = builder.create<fir::LoadOp>(loc, origVal);
       }
-    if (origVal.getType().isa<fir::BoxType>() && !eleTy.isa<fir::BoxType>()) {
+    if (mlir::isa<fir::BoxType>(origVal.getType()) &&
+        !mlir::isa<fir::BoxType>(eleTy)) {
       if (isPointerAssignment())
         TODO(loc, "lhs of pointer assignment returned unexpected value");
       TODO(loc, "invalid box conversion in elemental computation");
     }
-    if (isPointerAssignment() && eleTy.isa<fir::BoxType>() &&
-        !origVal.getType().isa<fir::BoxType>()) {
+    if (isPointerAssignment() && mlir::isa<fir::BoxType>(eleTy) &&
+        !mlir::isa<fir::BoxType>(origVal.getType())) {
       // This is a pointer assignment and the rhs is a raw reference to a TARGET
       // in memory. Embox the reference so it can be stored to the boxed
       // POINTER variable.
@@ -3930,7 +3934,7 @@ private:
       if (auto eleTy = fir::dyn_cast_ptrEleTy(origVal.getType());
           fir::hasDynamicSize(eleTy))
         TODO(loc, "TARGET of pointer assignment with runtime size/shape");
-      auto memrefTy = fir::boxMemRefType(eleTy.cast<fir::BoxType>());
+      auto memrefTy = fir::boxMemRefType(mlir::cast<fir::BoxType>(eleTy));
       auto castTo = builder.createConvert(loc, memrefTy, origVal);
       origVal = builder.create<fir::EmboxOp>(loc, eleTy, castTo);
     }
@@ -3982,7 +3986,7 @@ private:
         auto arrayOp = builder.create<fir::ArrayAccessOp>(
             loc, resRefTy, innerArg, iterSpace.iterVec(),
             fir::factory::getTypeParams(loc, builder, destination));
-        if (auto charTy = eleTy.dyn_cast<fir::CharacterType>()) {
+        if (auto charTy = mlir::dyn_cast<fir::CharacterType>(eleTy)) {
           llvm::SmallVector<mlir::Value> substringBounds;
           populateBounds(substringBounds, substring);
           mlir::Value dstLen = fir::factory::genLenOfCharacter(
@@ -3996,7 +4000,7 @@ private:
               loc, destination, builder, arrayOp, exv, eleTy, innerArg);
           return abstractArrayExtValue(amend /*FIXME: typeparams?*/);
         }
-        assert(eleTy.isa<fir::SequenceType>() && "must be an array");
+        assert(mlir::isa<fir::SequenceType>(eleTy) && "must be an array");
         TODO(loc, "array (as element) assignment");
       }
       // By value semantics. The element is being assigned by value.
@@ -4060,7 +4064,7 @@ private:
   llvm::SmallVector<mlir::Value> getShape(ArrayOperand array) {
     if (array.slice)
       return computeSliceShape(array.slice);
-    if (array.memref.getType().isa<fir::BaseBoxType>())
+    if (mlir::isa<fir::BaseBoxType>(array.memref.getType()))
       return fir::factory::readExtents(builder, getLoc(),
                                        fir::BoxValue{array.memref});
     return fir::factory::getExtents(array.shape);
@@ -4133,7 +4137,7 @@ private:
     mlir::Location loc = getLoc();
     return [=, builder = &converter.getFirOpBuilder()](IterSpace iters) {
       mlir::Type arrTy = fir::dyn_cast_ptrOrBoxEleTy(tmp.getType());
-      auto eleTy = arrTy.cast<fir::SequenceType>().getEleTy();
+      auto eleTy = mlir::cast<fir::SequenceType>(arrTy).getEleTy();
       mlir::Type eleRefTy = builder->getRefType(eleTy);
       mlir::IntegerType i1Ty = builder->getI1Type();
       // Adjust indices for any shift of the origin of the array.
@@ -4442,15 +4446,15 @@ private:
       TODO(loc, "polymorphic array temporary");
     if (ccLoadDest)
       return (*ccLoadDest)(shape);
-    auto seqTy = type.dyn_cast<fir::SequenceType>();
+    auto seqTy = mlir::dyn_cast<fir::SequenceType>(type);
     assert(seqTy && "must be an array");
     // TODO: Need to thread the LEN parameters here. For character, they may
     // differ from the operands length (e.g concatenation). So the array loads
     // type parameters are not enough.
-    if (auto charTy = seqTy.getEleTy().dyn_cast<fir::CharacterType>())
+    if (auto charTy = mlir::dyn_cast<fir::CharacterType>(seqTy.getEleTy()))
       if (charTy.hasDynamicLen())
         TODO(loc, "character array expression temp with dynamic length");
-    if (auto recTy = seqTy.getEleTy().dyn_cast<fir::RecordType>())
+    if (auto recTy = mlir::dyn_cast<fir::RecordType>(seqTy.getEleTy()))
       if (recTy.getNumLenParams() > 0)
         TODO(loc, "derived type array expression temp with LEN parameters");
     if (mlir::Type eleTy = fir::unwrapSequenceType(type);
@@ -4827,7 +4831,7 @@ private:
             });
           } else {
             ExtValue exv = asScalarRef(*expr);
-            if (fir::getBase(exv).getType().isa<fir::BaseBoxType>()) {
+            if (mlir::isa<fir::BaseBoxType>(fir::getBase(exv).getType())) {
               operands.emplace_back(
                   [=](IterSpace iters) -> ExtValue { return exv; });
             } else {
@@ -5565,7 +5569,7 @@ private:
   }
 
   static mlir::Type unwrapBoxEleTy(mlir::Type ty) {
-    if (auto boxTy = ty.dyn_cast<fir::BaseBoxType>())
+    if (auto boxTy = mlir::dyn_cast<fir::BaseBoxType>(ty))
       return fir::unwrapRefType(boxTy.getEleTy());
     return ty;
   }
@@ -5575,7 +5579,7 @@ private:
     ty = unwrapBoxEleTy(ty);
     mlir::Location loc = getLoc();
     mlir::IndexType idxTy = builder.getIndexType();
-    for (auto extent : ty.cast<fir::SequenceType>().getShape()) {
+    for (auto extent : mlir::cast<fir::SequenceType>(ty).getShape()) {
       auto v = extent == fir::SequenceType::getUnknownExtent()
                    ? builder.create<fir::UndefOp>(loc, idxTy).getResult()
                    : builder.createIntegerConstant(loc, idxTy, extent);
@@ -5638,7 +5642,8 @@ private:
     mlir::Location loc = getLoc();
     mlir::Value memref = fir::getBase(extMemref);
     mlir::Type arrTy = fir::dyn_cast_ptrOrBoxEleTy(memref.getType());
-    assert(arrTy.isa<fir::SequenceType>() && "memory ref must be an array");
+    assert(mlir::isa<fir::SequenceType>(arrTy) &&
+           "memory ref must be an array");
     mlir::Value shape = builder.createShape(loc, extMemref);
     mlir::Value slice;
     if (components.isSlice()) {
@@ -5688,12 +5693,12 @@ private:
                                     components.suffixComponents);
       }
       if (components.hasComponents()) {
-        auto seqTy = arrTy.cast<fir::SequenceType>();
+        auto seqTy = mlir::cast<fir::SequenceType>(arrTy);
         mlir::Type eleTy =
             fir::applyPathToType(seqTy.getEleTy(), components.suffixComponents);
         if (!eleTy)
           fir::emitFatalError(loc, "slicing path is ill-formed");
-        if (auto realTy = eleTy.dyn_cast<fir::RealType>())
+        if (auto realTy = mlir::dyn_cast<fir::RealType>(eleTy))
           eleTy = Fortran::lower::convertReal(realTy.getContext(),
                                               realTy.getFKind());
 
@@ -5713,13 +5718,14 @@ private:
       // value. The value of the box is forwarded in the continuation.
       mlir::Type reduceTy = reduceRank(arrTy, slice);
       mlir::Type boxTy = fir::BoxType::get(reduceTy);
-      if (memref.getType().isa<fir::ClassType>() && !components.hasComponents())
+      if (mlir::isa<fir::ClassType>(memref.getType()) &&
+          !components.hasComponents())
         boxTy = fir::ClassType::get(reduceTy);
       if (components.substring) {
         // Adjust char length to substring size.
         fir::CharacterType charTy =
             fir::factory::CharacterExprHelper::getCharType(reduceTy);
-        auto seqTy = reduceTy.cast<fir::SequenceType>();
+        auto seqTy = mlir::cast<fir::SequenceType>(reduceTy);
         // TODO: Use a constant for fir.char LEN if we can compute it.
         boxTy = fir::BoxType::get(
             fir::SequenceType::get(fir::CharacterType::getUnknownLen(
@@ -5734,7 +5740,7 @@ private:
         nonDeferredLenParams = fir::factory::getNonDeferredLenParams(extMemref);
       }
       mlir::Value embox =
-          memref.getType().isa<fir::BaseBoxType>()
+          mlir::isa<fir::BaseBoxType>(memref.getType())
               ? builder.create<fir::ReboxOp>(loc, boxTy, memref, shape, slice)
                     .getResult()
               : builder
@@ -5745,7 +5751,7 @@ private:
         return fir::BoxValue(embox, lbounds, nonDeferredLenParams);
       };
     }
-    auto eleTy = arrTy.cast<fir::SequenceType>().getEleTy();
+    auto eleTy = mlir::cast<fir::SequenceType>(arrTy).getEleTy();
     if (isReferentiallyOpaque()) {
       // Semantics are an opaque reference to an array.
       // This case forwards a continuation that will generate the address
@@ -5760,12 +5766,12 @@ private:
         mlir::Value coor = builder.create<fir::ArrayCoorOp>(
             loc, refEleTy, memref, shape, slice, indices,
             fir::getTypeParams(extMemref));
-        if (auto charTy = eleTy.dyn_cast<fir::CharacterType>()) {
+        if (auto charTy = mlir::dyn_cast<fir::CharacterType>(eleTy)) {
           llvm::SmallVector<mlir::Value> substringBounds;
           populateBounds(substringBounds, components.substring);
           if (!substringBounds.empty()) {
             mlir::Value dstLen = fir::factory::genLenOfCharacter(
-                builder, loc, arrTy.cast<fir::SequenceType>(), memref,
+                builder, loc, mlir::cast<fir::SequenceType>(arrTy), memref,
                 fir::getTypeParams(extMemref), iters.iterVec(),
                 substringBounds);
             fir::CharBoxValue dstChar(coor, dstLen);
@@ -5863,7 +5869,7 @@ private:
         mlir::Type eleRefTy = builder.getRefType(eleTy);
         mlir::Value arrayOp = builder.create<fir::ArrayAccessOp>(
             loc, eleRefTy, arrLd, iters.iterVec(), arrLdTypeParams);
-        if (auto charTy = eleTy.dyn_cast<fir::CharacterType>()) {
+        if (auto charTy = mlir::dyn_cast<fir::CharacterType>(eleTy)) {
           llvm::SmallVector<mlir::Value> substringBounds;
           populateBounds(substringBounds, components.substring);
           if (!substringBounds.empty()) {
@@ -5896,7 +5902,7 @@ private:
     const bool hasOptionalAttr =
         fir::valueHasFirAttribute(base, fir::getOptionalAttrName());
     mlir::Type baseType = fir::unwrapRefType(base.getType());
-    const bool isBox = baseType.isa<fir::BoxType>();
+    const bool isBox = mlir::isa<fir::BoxType>(baseType);
     const bool isAllocOrPtr =
         Fortran::evaluate::IsAllocatableOrPointerObject(expr);
     mlir::Type arrType = fir::unwrapPassByRefType(baseType);
@@ -5989,7 +5995,7 @@ private:
     if (slice) {
       auto slOp = mlir::dyn_cast<fir::SliceOp>(slice.getDefiningOp());
       assert(slOp && "expected slice op");
-      auto seqTy = arrTy.dyn_cast<fir::SequenceType>();
+      auto seqTy = mlir::dyn_cast<fir::SequenceType>(arrTy);
       assert(seqTy && "expected array type");
       mlir::Operation::operand_range triples = slOp.getTriples();
       fir::SequenceType::Shape shape;
@@ -6053,7 +6059,7 @@ private:
     mlir::IndexType idxTy = builder.getIndexType();
     mlir::Value multiplier = builder.createIntegerConstant(loc, idxTy, 1);
     if (fir::hasDynamicSize(eleTy)) {
-      if (auto charTy = eleTy.dyn_cast<fir::CharacterType>()) {
+      if (auto charTy = mlir::dyn_cast<fir::CharacterType>(eleTy)) {
         // Array of char with dynamic LEN parameter. Downcast to an array
         // of singleton char, and scale by the len type parameter from
         // `exv`.
@@ -6074,7 +6080,7 @@ private:
             });
         fir::CharacterType newEleTy = fir::CharacterType::getSingleton(
             eleTy.getContext(), charTy.getFKind());
-        if (auto seqTy = resTy.dyn_cast<fir::SequenceType>()) {
+        if (auto seqTy = mlir::dyn_cast<fir::SequenceType>(resTy)) {
           assert(eleTy == seqTy.getEleTy());
           resTy = fir::SequenceType::get(seqTy.getShape(), newEleTy);
         }
@@ -6161,7 +6167,7 @@ private:
     if (!eleSz) {
       // Compute the element size at runtime.
       assert(fir::hasDynamicSize(eleTy));
-      if (auto charTy = eleTy.dyn_cast<fir::CharacterType>()) {
+      if (auto charTy = mlir::dyn_cast<fir::CharacterType>(eleTy)) {
         auto charBytes =
             builder.getKindMap().getCharacterBitsize(charTy.getFKind()) / 8;
         mlir::Value bytes =
@@ -6181,7 +6187,7 @@ private:
     auto computeCoordinate = [&](mlir::Value buff, mlir::Value off) {
       mlir::Type refTy = eleRefTy;
       if (fir::hasDynamicSize(eleTy)) {
-        if (auto charTy = eleTy.dyn_cast<fir::CharacterType>()) {
+        if (auto charTy = mlir::dyn_cast<fir::CharacterType>(eleTy)) {
           // Scale a simple pointer using dynamic length and offset values.
           auto chTy = fir::CharacterType::getSingleton(charTy.getContext(),
                                                        charTy.getFKind());
@@ -6308,7 +6314,7 @@ private:
         builder.createConvert(loc, idxTy, fir::getBase(asScalar(x.upper())));
     mlir::Value step =
         builder.createConvert(loc, idxTy, fir::getBase(asScalar(x.stride())));
-    auto seqTy = resTy.template cast<fir::SequenceType>();
+    auto seqTy = mlir::cast<fir::SequenceType>(resTy);
     mlir::Type eleTy = fir::unwrapSequenceType(seqTy);
     auto loop =
         builder.create<fir::DoLoopOp>(loc, lo, up, step, /*unordered=*/false,
@@ -6375,7 +6381,7 @@ private:
     auto evExpr = toEvExpr(x);
     mlir::Type resTy = translateSomeExprToFIRType(converter, evExpr);
     mlir::IndexType idxTy = builder.getIndexType();
-    auto seqTy = resTy.template cast<fir::SequenceType>();
+    auto seqTy = mlir::cast<fir::SequenceType>(resTy);
     mlir::Type eleTy = fir::unwrapSequenceType(resTy);
     mlir::Value buffSize = builder.createTemporary(loc, idxTy, ".buff.size");
     mlir::Value zero = builder.createIntegerConstant(loc, idxTy, 0);
@@ -6719,7 +6725,7 @@ private:
                 auto fieldTy = fir::FieldType::get(builder.getContext());
                 std::string name =
                     converter.getRecordTypeFieldName(getLastSym(*x));
-                if (auto recTy = ty.dyn_cast<fir::RecordType>()) {
+                if (auto recTy = mlir::dyn_cast<fir::RecordType>(ty)) {
                   ty = recTy.getType(name);
                   auto fld = builder.create<fir::FieldIndexOp>(
                       loc, fieldTy, name, recTy, fir::getTypeParams(arrayExv));
@@ -6728,7 +6734,7 @@ private:
                     // Need an intermediate  dereference if the boxed value
                     // appears in the middle of the component path or if it is
                     // on the right and this is not a pointer assignment.
-                    if (auto boxTy = ty.dyn_cast<fir::BaseBoxType>()) {
+                    if (auto boxTy = mlir::dyn_cast<fir::BaseBoxType>(ty)) {
                       auto currentFunc = components.getExtendCoorRef();
                       auto loc = getLoc();
                       auto *bldr = &converter.getFirOpBuilder();
@@ -6739,9 +6745,9 @@ private:
                       deref = true;
                     }
                   }
-                } else if (auto boxTy = ty.dyn_cast<fir::BaseBoxType>()) {
+                } else if (auto boxTy = mlir::dyn_cast<fir::BaseBoxType>(ty)) {
                   ty = fir::unwrapRefType(boxTy.getEleTy());
-                  auto recTy = ty.cast<fir::RecordType>();
+                  auto recTy = mlir::cast<fir::RecordType>(ty);
                   ty = recTy.getType(name);
                   auto fld = builder.create<fir::FieldIndexOp>(
                       loc, fieldTy, name, recTy, fir::getTypeParams(arrayExv));
@@ -6790,7 +6796,7 @@ private:
           auto arrayOp = builder.create<fir::ArrayAccessOp>(
               loc, eleRefTy, innerArg, iters.iterVec(),
               fir::factory::getTypeParams(loc, builder, load));
-          if (auto charTy = eleTy.dyn_cast<fir::CharacterType>()) {
+          if (auto charTy = mlir::dyn_cast<fir::CharacterType>(eleTy)) {
             mlir::Value dstLen = fir::factory::genLenOfCharacter(
                 builder, loc, load, iters.iterVec(), substringBounds);
             fir::ArrayAmendOp amend = createCharArrayAmend(
@@ -6806,13 +6812,13 @@ private:
             return arrayLoadExtValue(builder, loc, load, iters.iterVec(),
                                      amend);
           }
-          assert(eleTy.isa<fir::SequenceType>());
+          assert(mlir::isa<fir::SequenceType>(eleTy));
           TODO(loc, "array (as element) assignment");
         }
         if (components.hasExtendCoorRef()) {
           auto eleBoxTy =
               fir::applyPathToType(innerArg.getType(), iters.iterVec());
-          if (!eleBoxTy || !eleBoxTy.isa<fir::BoxType>())
+          if (!eleBoxTy || !mlir::isa<fir::BoxType>(eleBoxTy))
             TODO(loc, "assignment in a FORALL involving a designator with a "
                       "POINTER or ALLOCATABLE component part-ref");
           auto arrayOp = builder.create<fir::ArrayAccessOp>(
@@ -6824,7 +6830,7 @@ private:
           // assignment, then insert the dereference of the box before any
           // conversion and store.
           if (!isPointerAssignment()) {
-            if (auto boxTy = eleTy.dyn_cast<fir::BaseBoxType>()) {
+            if (auto boxTy = mlir::dyn_cast<fir::BaseBoxType>(eleTy)) {
               eleTy = fir::boxMemRefType(boxTy);
               addr = builder.create<fir::BoxAddrOp>(loc, eleTy, addr);
               eleTy = fir::unwrapRefType(eleTy);
@@ -6885,7 +6891,7 @@ private:
       }
       if (components.hasExtendCoorRef()) {
         auto eleBoxTy = fir::applyPathToType(load.getType(), iters.iterVec());
-        if (!eleBoxTy || !eleBoxTy.isa<fir::BoxType>())
+        if (!eleBoxTy || !mlir::isa<fir::BoxType>(eleBoxTy))
           TODO(loc, "assignment in a FORALL involving a designator with a "
                     "POINTER or ALLOCATABLE component part-ref");
         auto access = builder.create<fir::ArrayAccessOp>(
@@ -6897,7 +6903,7 @@ private:
       }
       if (isPointerAssignment()) {
         auto eleTy = fir::applyPathToType(load.getType(), iters.iterVec());
-        if (!eleTy.isa<fir::BoxType>()) {
+        if (!mlir::isa<fir::BoxType>(eleTy)) {
           // Rhs is a regular expression that will need to be boxed before
           // assigning to the boxed variable.
           auto typeParams = fir::factory::getTypeParams(loc, builder, load);
@@ -7615,7 +7621,7 @@ mlir::Value Fortran::lower::addCrayPointerInst(mlir::Location loc,
   auto box = builder.create<fir::EmboxOp>(loc, boxTy, ptrVal, empty, empty,
                                           emptyRange);
   mlir::Value addrof =
-      (ptrTy.isa<fir::ReferenceType>())
+      (mlir::isa<fir::ReferenceType>(ptrTy))
           ? builder.create<fir::BoxAddrOp>(loc, ptrTy, box)
           : builder.create<fir::BoxAddrOp>(loc, builder.getRefType(ptrTy), box);
 
diff --git a/flang/lib/Lower/ConvertExprToHLFIR.cpp b/flang/lib/Lower/ConvertExprToHLFIR.cpp
index 6e57b31d022b..93bdf650f9ff 100644
--- a/flang/lib/Lower/ConvertExprToHLFIR.cpp
+++ b/flang/lib/Lower/ConvertExprToHLFIR.cpp
@@ -138,8 +138,8 @@ public:
     mlir::Location loc = getLoc();
     mlir::Type idxTy = builder.getIndexType();
     llvm::SmallVector<mlir::Value> extents;
-    auto seqTy = hlfir::getFortranElementOrSequenceType(fieldType)
-                     .cast<fir::SequenceType>();
+    auto seqTy = mlir::cast<fir::SequenceType>(
+        hlfir::getFortranElementOrSequenceType(fieldType));
     for (auto extent : seqTy.getShape()) {
       if (extent == fir::SequenceType::getUnknownExtent()) {
         // We have already generated invalid hlfir.declare
@@ -199,7 +199,7 @@ private:
                                    const T &designatorNode) {
     // Get base's shape if its a sequence type with no previously computed
     // result shape
-    if (partInfo.base && resultValueType.isa<fir::SequenceType>() &&
+    if (partInfo.base && mlir::isa<fir::SequenceType>(resultValueType) &&
         !partInfo.resultShape)
       partInfo.resultShape =
           hlfir::genShape(getLoc(), getBuilder(), *partInfo.base);
@@ -209,7 +209,7 @@ private:
       return fir::ClassType::get(resultValueType);
     // Character scalar with dynamic length needs a fir.boxchar to hold the
     // designator length.
-    auto charType = resultValueType.dyn_cast<fir::CharacterType>();
+    auto charType = mlir::dyn_cast<fir::CharacterType>(resultValueType);
     if (charType && charType.hasDynamicLen())
       return fir::BoxCharType::get(charType.getContext(), charType.getFKind());
     // Arrays with non default lower bounds or dynamic length or dynamic extent
@@ -218,7 +218,7 @@ private:
         hasNonDefaultLowerBounds(partInfo))
       return fir::BoxType::get(resultValueType);
     // Non simply contiguous ref require a fir.box to carry the byte stride.
-    if (resultValueType.isa<fir::SequenceType>() &&
+    if (mlir::isa<fir::SequenceType>(resultValueType) &&
         !Fortran::evaluate::IsSimplyContiguous(
             designatorNode, getConverter().getFoldingContext()))
       return fir::BoxType::get(resultValueType);
@@ -398,8 +398,8 @@ private:
       partInfo.typeParams[0] =
           fir::factory::genMaxWithZero(builder, loc, rawLen);
     }
-    auto kind = hlfir::getFortranElementType(baseStringType)
-                    .cast<fir::CharacterType>()
+    auto kind = mlir::cast<fir::CharacterType>(
+                    hlfir::getFortranElementType(baseStringType))
                     .getFKind();
     auto newCharTy = fir::CharacterType::get(
         baseStringType.getContext(), kind,
@@ -579,7 +579,7 @@ private:
       return createVectorSubscriptElementAddrOp(partInfo, baseType,
                                                 resultExtents);
 
-    mlir::Type resultType = baseType.cast<fir::SequenceType>().getEleTy();
+    mlir::Type resultType = mlir::cast<fir::SequenceType>(baseType).getEleTy();
     if (!resultTypeShape.empty()) {
       // Ranked array section. The result shape comes from the array section
       // subscripts.
@@ -612,8 +612,8 @@ private:
   }
   static bool hasNonDefaultLowerBounds(const PartInfo &partInfo) {
     return partInfo.resultShape &&
-           (partInfo.resultShape.getType().isa<fir::ShiftType>() ||
-            partInfo.resultShape.getType().isa<fir::ShapeShiftType>());
+           mlir::isa<fir::ShiftType, fir::ShapeShiftType>(
+               partInfo.resultShape.getType());
   }
 
   mlir::Type visit(const Fortran::evaluate::Component &component,
@@ -705,7 +705,7 @@ private:
     const Fortran::semantics::Symbol &componentSym = component.GetLastSymbol();
     partInfo.componentName = converter.getRecordTypeFieldName(componentSym);
     auto recordType =
-        hlfir::getFortranElementType(baseType).cast<fir::RecordType>();
+        mlir::cast<fir::RecordType>(hlfir::getFortranElementType(baseType));
     if (recordType.isDependentType())
       TODO(getLoc(), "Designate derived type with length parameters in HLFIR");
     mlir::Type fieldType = recordType.getType(partInfo.componentName);
@@ -718,7 +718,7 @@ private:
     if (fir::isRecordWithTypeParameters(fieldEleType))
       TODO(loc,
            "lower a component that is a parameterized derived type to HLFIR");
-    if (auto charTy = fieldEleType.dyn_cast<fir::CharacterType>()) {
+    if (auto charTy = mlir::dyn_cast<fir::CharacterType>(fieldEleType)) {
       mlir::Location loc = getLoc();
       mlir::Type idxTy = builder.getIndexType();
       if (charTy.hasConstantLen())
@@ -811,7 +811,7 @@ private:
       }
     }
     builder.setInsertionPoint(elementalAddrOp);
-    return baseType.cast<fir::SequenceType>().getEleTy();
+    return mlir::cast<fir::SequenceType>(baseType).getEleTy();
   }
 
   /// Yield the designator for the final part-ref inside the
@@ -1665,7 +1665,7 @@ private:
     mlir::Location loc = getLoc();
     fir::FirOpBuilder &builder = getBuilder();
     mlir::Type ty = translateSomeExprToFIRType(converter, toEvExpr(ctor));
-    auto recTy = ty.cast<fir::RecordType>();
+    auto recTy = mlir::cast<fir::RecordType>(ty);
 
     if (recTy.isDependentType())
       TODO(loc, "structure constructor for derived type with length parameters "
diff --git a/flang/lib/Lower/ConvertProcedureDesignator.cpp b/flang/lib/Lower/ConvertProcedureDesignator.cpp
index 2446be3a1908..aa0d7ce54788 100644
--- a/flang/lib/Lower/ConvertProcedureDesignator.cpp
+++ b/flang/lib/Lower/ConvertProcedureDesignator.cpp
@@ -107,11 +107,11 @@ static hlfir::EntityWithAttributes designateProcedurePointerComponent(
                                                 procComponentSym);
   /// Passed argument may be a descriptor. This is a scalar reference, so the
   /// base address can be directly addressed.
-  if (base.getType().isa<fir::BaseBoxType>())
+  if (mlir::isa<fir::BaseBoxType>(base.getType()))
     base = builder.create<fir::BoxAddrOp>(loc, base);
   std::string fieldName = converter.getRecordTypeFieldName(procComponentSym);
   auto recordType =
-      hlfir::getFortranElementType(base.getType()).cast<fir::RecordType>();
+      mlir::cast<fir::RecordType>(hlfir::getFortranElementType(base.getType()));
   mlir::Type fieldType = recordType.getType(fieldName);
   // Note: semantics turns x%p() into x%t%p() when the procedure pointer
   // component is part of parent component t.
@@ -164,7 +164,7 @@ hlfir::EntityWithAttributes Fortran::lower::convertProcedureDesignatorToHLFIR(
   fir::FirOpBuilder &builder = converter.getFirOpBuilder();
 
   mlir::Value funcAddr = fir::getBase(procExv);
-  if (!funcAddr.getType().isa<fir::BoxProcType>()) {
+  if (!mlir::isa<fir::BoxProcType>(funcAddr.getType())) {
     mlir::Type boxTy =
         Fortran::lower::getUntypedBoxProcType(&converter.getMLIRContext());
     if (auto host = Fortran::lower::argumentHostAssocs(converter, funcAddr))
diff --git a/flang/lib/Lower/ConvertVariable.cpp b/flang/lib/Lower/ConvertVariable.cpp
index 2d2d9eba905b..413563fe95ca 100644
--- a/flang/lib/Lower/ConvertVariable.cpp
+++ b/flang/lib/Lower/ConvertVariable.cpp
@@ -389,13 +389,13 @@ static mlir::Value genDefaultInitializerValue(
   fir::FirOpBuilder &builder = converter.getFirOpBuilder();
   mlir::Type scalarType = symTy;
   fir::SequenceType sequenceType;
-  if (auto ty = symTy.dyn_cast<fir::SequenceType>()) {
+  if (auto ty = mlir::dyn_cast<fir::SequenceType>(symTy)) {
     sequenceType = ty;
     scalarType = ty.getEleTy();
   }
   // Build a scalar default value of the symbol type, looping through the
   // components to build each component initial value.
-  auto recTy = scalarType.cast<fir::RecordType>();
+  auto recTy = mlir::cast<fir::RecordType>(scalarType);
   mlir::Value initialValue = builder.create<fir::UndefOp>(loc, scalarType);
   const Fortran::semantics::DeclTypeSpec *declTy = sym.GetType();
   assert(declTy && "var with default initialization must have a type");
@@ -493,11 +493,11 @@ static fir::GlobalOp defineGlobal(Fortran::lower::AbstractConverter &converter,
   // with a tensor mlir type. This optimization currently only supports
   // Fortran arrays of integer, real, complex, or logical. The tensor
   // type does not support nested structures.
-  if (symTy.isa<fir::SequenceType>() &&
+  if (mlir::isa<fir::SequenceType>(symTy) &&
       !Fortran::semantics::IsAllocatableOrPointer(sym)) {
-    mlir::Type eleTy = symTy.cast<fir::SequenceType>().getEleTy();
-    if (eleTy.isa<mlir::IntegerType, mlir::FloatType, fir::ComplexType,
-                  fir::LogicalType>()) {
+    mlir::Type eleTy = mlir::cast<fir::SequenceType>(symTy).getEleTy();
+    if (mlir::isa<mlir::IntegerType, mlir::FloatType, fir::ComplexType,
+                  fir::LogicalType>(eleTy)) {
       const auto *details =
           sym.detailsIf<Fortran::semantics::ObjectEntityDetails>();
       if (details->init()) {
@@ -916,13 +916,14 @@ static void instantiateLocal(Fortran::lower::AbstractConverter &converter,
       break;
     case VariableCleanUp::Deallocate:
       auto *converterPtr = &converter;
-      converter.getFctCtx().attachCleanup([converterPtr, loc, exv]() {
+      auto *sym = &var.getSymbol();
+      converter.getFctCtx().attachCleanup([converterPtr, loc, exv, sym]() {
         const fir::MutableBoxValue *mutableBox =
             exv.getBoxOf<fir::MutableBoxValue>();
         assert(mutableBox &&
                "trying to deallocate entity not lowered as allocatable");
         Fortran::lower::genDeallocateIfAllocated(*converterPtr, *mutableBox,
-                                                 loc);
+                                                 loc, sym);
       });
     }
   }
@@ -1291,7 +1292,7 @@ static void finalizeCommonBlockDefinition(
     fir::GlobalOp global,
     const Fortran::semantics::MutableSymbolVector &cmnBlkMems) {
   fir::FirOpBuilder &builder = converter.getFirOpBuilder();
-  mlir::TupleType commonTy = global.getType().cast<mlir::TupleType>();
+  mlir::TupleType commonTy = mlir::cast<mlir::TupleType>(global.getType());
   auto initFunc = [&](fir::FirOpBuilder &builder) {
     mlir::IndexType idxTy = builder.getIndexType();
     mlir::Value cb = builder.create<fir::ZeroOp>(loc, commonTy);
@@ -1406,7 +1407,7 @@ static bool lowerToBoxValue(const Fortran::semantics::Symbol &sym,
                             mlir::Value dummyArg,
                             Fortran::lower::AbstractConverter &converter) {
   // Only dummy arguments coming as fir.box can be tracked in an BoxValue.
-  if (!dummyArg || !dummyArg.getType().isa<fir::BaseBoxType>())
+  if (!dummyArg || !mlir::isa<fir::BaseBoxType>(dummyArg.getType()))
     return false;
   // Non contiguous arrays must be tracked in an BoxValue.
   if (sym.Rank() > 0 && !Fortran::evaluate::IsSimplyContiguous(
@@ -1715,7 +1716,8 @@ void Fortran::lower::genDeclareSymbol(
     const fir::ExtendedValue &exv, fir::FortranVariableFlagsEnum extraFlags,
     bool force) {
   if (converter.getLoweringOptions().getLowerToHighLevelFIR() &&
-      !Fortran::semantics::IsProcedure(sym) &&
+      (!Fortran::semantics::IsProcedure(sym) ||
+       Fortran::semantics::IsPointer(sym)) &&
       !sym.detailsIf<Fortran::semantics::CommonBlockDetails>()) {
     fir::FirOpBuilder &builder = converter.getFirOpBuilder();
     const mlir::Location loc = genLocation(converter, sym);
@@ -1903,7 +1905,7 @@ void Fortran::lower::mapSymbolAttributes(
           // Do not keep scalar characters as fir.box (even when optional).
           // Lowering and FIR is not meant to deal with scalar characters as
           // fir.box outside of calls.
-          auto boxTy = dummyArg.getType().dyn_cast<fir::BaseBoxType>();
+          auto boxTy = mlir::dyn_cast<fir::BaseBoxType>(dummyArg.getType());
           mlir::Type refTy = builder.getRefType(boxTy.getEleTy());
           mlir::Type lenType = builder.getCharacterLengthType();
           mlir::Value addr, len;
@@ -1982,8 +1984,8 @@ void Fortran::lower::mapSymbolAttributes(
       // a non pointer/allocatable symbol to be mapped to a MutableBox.
       mlir::Type ty = converter.genType(var);
       bool isPolymorphic = false;
-      if (auto boxTy = ty.dyn_cast<fir::BaseBoxType>()) {
-        isPolymorphic = ty.isa<fir::ClassType>();
+      if (auto boxTy = mlir::dyn_cast<fir::BaseBoxType>(ty)) {
+        isPolymorphic = mlir::isa<fir::ClassType>(ty);
         ty = boxTy.getEleTy();
       }
       Fortran::lower::genDeclareSymbol(
@@ -2090,7 +2092,7 @@ void Fortran::lower::mapSymbolAttributes(
   mlir::Value addr = preAlloc;
 
   if (arg)
-    if (auto boxTy = arg.getType().dyn_cast<fir::BaseBoxType>()) {
+    if (auto boxTy = mlir::dyn_cast<fir::BaseBoxType>(arg.getType())) {
       // Contiguous assumed shape that can be tracked without a fir.box.
       mlir::Type refTy = builder.getRefType(boxTy.getEleTy());
       addr = builder.create<fir::BoxAddrOp>(loc, refTy, arg);
@@ -2132,7 +2134,7 @@ void Fortran::lower::mapSymbolAttributes(
       } else if (!len) {
         // Assumed length fir.box (possible for contiguous assumed shapes).
         // Read length from box.
-        assert(arg && arg.getType().isa<fir::BoxType>() &&
+        assert(arg && mlir::isa<fir::BoxType>(arg.getType()) &&
                "must be character dummy fir.box");
         len = charHelp.readLengthFromBox(arg);
       }
diff --git a/flang/lib/Lower/CustomIntrinsicCall.cpp b/flang/lib/Lower/CustomIntrinsicCall.cpp
index 439fc3d915b4..30c6ce7f53b3 100644
--- a/flang/lib/Lower/CustomIntrinsicCall.cpp
+++ b/flang/lib/Lower/CustomIntrinsicCall.cpp
@@ -227,22 +227,23 @@ lowerIshftc(fir::FirOpBuilder &builder, mlir::Location loc,
   args.push_back(getOperand(1, loadOperand));
   auto iPC = isPresentCheck(2);
   assert(iPC.has_value());
-  args.push_back(builder
-                     .genIfOp(loc, {resultType}, *iPC,
-                              /*withElseRegion=*/true)
-                     .genThen([&]() {
-                       fir::ExtendedValue sizeExv = getOperand(2, loadOperand);
-                       mlir::Value size = builder.createConvert(
-                           loc, resultType, fir::getBase(sizeExv));
-                       builder.create<fir::ResultOp>(loc, size);
-                     })
-                     .genElse([&]() {
-                       mlir::Value bitSize = builder.createIntegerConstant(
-                           loc, resultType,
-                           resultType.cast<mlir::IntegerType>().getWidth());
-                       builder.create<fir::ResultOp>(loc, bitSize);
-                     })
-                     .getResults()[0]);
+  args.push_back(
+      builder
+          .genIfOp(loc, {resultType}, *iPC,
+                   /*withElseRegion=*/true)
+          .genThen([&]() {
+            fir::ExtendedValue sizeExv = getOperand(2, loadOperand);
+            mlir::Value size =
+                builder.createConvert(loc, resultType, fir::getBase(sizeExv));
+            builder.create<fir::ResultOp>(loc, size);
+          })
+          .genElse([&]() {
+            mlir::Value bitSize = builder.createIntegerConstant(
+                loc, resultType,
+                mlir::cast<mlir::IntegerType>(resultType).getWidth());
+            builder.create<fir::ResultOp>(loc, bitSize);
+          })
+          .getResults()[0]);
   return genIntrinsicCall(builder, loc, name, resultType, args, stmtCtx);
 }
 
@@ -282,7 +283,7 @@ lowerAssociated(fir::FirOpBuilder &builder, mlir::Location loc,
       builder.create<fir::IsPresentOp>(loc, builder.getI1Type(), targetBase);
   mlir::Type targetType = fir::unwrapRefType(targetBase.getType());
   mlir::Type targetValueType = fir::unwrapPassByRefType(targetType);
-  mlir::Type boxType = targetType.isa<fir::BaseBoxType>()
+  mlir::Type boxType = mlir::isa<fir::BaseBoxType>(targetType)
                            ? targetType
                            : fir::BoxType::get(targetValueType);
   fir::BoxValue targetBox =
diff --git a/flang/lib/Lower/DirectivesCommon.h b/flang/lib/Lower/DirectivesCommon.h
index 3ebf3fd965da..42bd3868196b 100644
--- a/flang/lib/Lower/DirectivesCommon.h
+++ b/flang/lib/Lower/DirectivesCommon.h
@@ -642,14 +642,14 @@ getDataOperandBaseAddr(Fortran::lower::AbstractConverter &converter,
     isPresent =
         builder.create<fir::IsPresentOp>(loc, builder.getI1Type(), rawInput);
 
-  if (auto boxTy =
-          fir::unwrapRefType(symAddr.getType()).dyn_cast<fir::BaseBoxType>()) {
-    if (boxTy.getEleTy().isa<fir::RecordType>())
+  if (auto boxTy = mlir::dyn_cast<fir::BaseBoxType>(
+          fir::unwrapRefType(symAddr.getType()))) {
+    if (mlir::isa<fir::RecordType>(boxTy.getEleTy()))
       TODO(loc, "derived type");
 
     // Load the box when baseAddr is a `fir.ref<fir.box<T>>` or a
     // `fir.ref<fir.class<T>>` type.
-    if (symAddr.getType().isa<fir::ReferenceType>()) {
+    if (mlir::isa<fir::ReferenceType>(symAddr.getType())) {
       if (Fortran::semantics::IsOptional(sym)) {
         mlir::Value addr =
             builder.genIfOp(loc, {boxTy}, isPresent, /*withElseRegion=*/true)
@@ -722,7 +722,7 @@ genBoundsOpsFromBox(fir::FirOpBuilder &builder, mlir::Location loc,
   mlir::Type idxTy = builder.getIndexType();
   mlir::Type boundTy = builder.getType<BoundsType>();
 
-  assert(info.addr.getType().isa<fir::BaseBoxType>() &&
+  assert(mlir::isa<fir::BaseBoxType>(info.addr.getType()) &&
          "expect fir.box or fir.class");
 
   if (info.isPresent) {
@@ -909,7 +909,8 @@ genBoundsOps(fir::FirOpBuilder &builder, mlir::Location loc,
       mlir::Value stride = one;
       bool strideInBytes = false;
 
-      if (fir::unwrapRefType(info.addr.getType()).isa<fir::BaseBoxType>()) {
+      if (mlir::isa<fir::BaseBoxType>(
+              fir::unwrapRefType(info.addr.getType()))) {
         if (info.isPresent) {
           stride =
               builder
@@ -1020,8 +1021,8 @@ genBoundsOps(fir::FirOpBuilder &builder, mlir::Location loc,
           }
         }
 
-        if (info.isPresent &&
-            fir::unwrapRefType(info.addr.getType()).isa<fir::BaseBoxType>()) {
+        if (info.isPresent && mlir::isa<fir::BaseBoxType>(
+                                  fir::unwrapRefType(info.addr.getType()))) {
           extent =
               builder
                   .genIfOp(loc, idxTy, info.isPresent, /*withElseRegion=*/true)
@@ -1157,7 +1158,7 @@ AddrAndBoundsInfo gatherDataOperandAddrAndBounds(
         converter.genExprAddr(operandLocation, designator, stmtCtx);
     info.addr = fir::getBase(compExv);
     info.rawInput = info.addr;
-    if (fir::unwrapRefType(info.addr.getType()).isa<fir::SequenceType>())
+    if (mlir::isa<fir::SequenceType>(fir::unwrapRefType(info.addr.getType())))
       bounds = genBaseBoundsOps<BoundsOp, BoundsType>(builder, operandLocation,
                                                       converter, compExv,
                                                       /*isAssumedSize=*/false);
@@ -1199,13 +1200,14 @@ AddrAndBoundsInfo gatherDataOperandAddrAndBounds(
       fir::ExtendedValue dataExv = converter.getSymbolExtendedValue(*symRef);
       info =
           getDataOperandBaseAddr(converter, builder, *symRef, operandLocation);
-      if (fir::unwrapRefType(info.addr.getType()).isa<fir::BaseBoxType>()) {
+      if (mlir::isa<fir::BaseBoxType>(
+              fir::unwrapRefType(info.addr.getType()))) {
         bounds = genBoundsOpsFromBox<BoundsOp, BoundsType>(
             builder, operandLocation, converter, dataExv, info);
       }
       bool dataExvIsAssumedSize =
           Fortran::semantics::IsAssumedSizeArray(symRef->get().GetUltimate());
-      if (fir::unwrapRefType(info.addr.getType()).isa<fir::SequenceType>())
+      if (mlir::isa<fir::SequenceType>(fir::unwrapRefType(info.addr.getType())))
         bounds = genBaseBoundsOps<BoundsOp, BoundsType>(
             builder, operandLocation, converter, dataExv, dataExvIsAssumedSize);
       asFortran << symRef->get().name().ToString();
diff --git a/flang/lib/Lower/HlfirIntrinsics.cpp b/flang/lib/Lower/HlfirIntrinsics.cpp
index bda04fa9689b..310b62697f71 100644
--- a/flang/lib/Lower/HlfirIntrinsics.cpp
+++ b/flang/lib/Lower/HlfirIntrinsics.cpp
@@ -265,7 +265,7 @@ HlfirTransformationalIntrinsic::computeResultType(mlir::Value argArray,
                                                   mlir::Type stmtResultType) {
   mlir::Type normalisedResult =
       hlfir::getFortranElementOrSequenceType(stmtResultType);
-  if (auto array = normalisedResult.dyn_cast<fir::SequenceType>()) {
+  if (auto array = mlir::dyn_cast<fir::SequenceType>(normalisedResult)) {
     hlfir::ExprType::Shape resultShape =
         hlfir::ExprType::Shape{array.getShape()};
     mlir::Type elementType = array.getEleTy();
@@ -341,7 +341,7 @@ mlir::Value HlfirTransposeLowering::lowerImpl(
   hlfir::ExprType::Shape resultShape;
   mlir::Type normalisedResult =
       hlfir::getFortranElementOrSequenceType(stmtResultType);
-  auto array = normalisedResult.cast<fir::SequenceType>();
+  auto array = mlir::cast<fir::SequenceType>(normalisedResult);
   llvm::ArrayRef<int64_t> arrayShape = array.getShape();
   assert(arrayShape.size() == 2 && "arguments to transpose have a rank of 2");
   mlir::Type elementType = array.getEleTy();
diff --git a/flang/lib/Lower/HostAssociations.cpp b/flang/lib/Lower/HostAssociations.cpp
index 8eb548eb2bd5..75a5bed56655 100644
--- a/flang/lib/Lower/HostAssociations.cpp
+++ b/flang/lib/Lower/HostAssociations.cpp
@@ -182,10 +182,10 @@ class CapturedProcedure : public CapturedSymbols<CapturedProcedure> {
 public:
   static mlir::Type getType(Fortran::lower::AbstractConverter &converter,
                             const Fortran::semantics::Symbol &sym) {
+    mlir::Type funTy = Fortran::lower::getDummyProcedureType(sym, converter);
     if (Fortran::semantics::IsPointer(sym))
-      TODO(converter.getCurrentLocation(),
-           "capture procedure pointer in internal procedure");
-    return Fortran::lower::getDummyProcedureType(sym, converter);
+      return fir::ReferenceType::get(funTy);
+    return funTy;
   }
 
   static void instantiateHostTuple(const InstantiateHostTuple &args,
@@ -219,7 +219,7 @@ public:
   static mlir::Type getType(Fortran::lower::AbstractConverter &converter,
                             const Fortran::semantics::Symbol &sym) {
     fir::KindTy kind =
-        converter.genType(sym).cast<fir::CharacterType>().getFKind();
+        mlir::cast<fir::CharacterType>(converter.genType(sym)).getFKind();
     return fir::BoxCharType::get(&converter.getMLIRContext(), kind);
   }
 
@@ -293,7 +293,7 @@ public:
     mlir::Location loc = args.loc;
     mlir::Value box = args.valueInTuple;
     if (Fortran::semantics::IsOptional(sym)) {
-      auto boxTy = box.getType().cast<fir::BaseBoxType>();
+      auto boxTy = mlir::cast<fir::BaseBoxType>(box.getType());
       auto eleTy = boxTy.getEleTy();
       if (!fir::isa_ref_type(eleTy))
         eleTy = builder.getRefType(eleTy);
@@ -381,8 +381,8 @@ public:
                             const Fortran::semantics::Symbol &sym) {
     mlir::Type type = converter.genType(sym);
     bool isPolymorphic = Fortran::semantics::IsPolymorphic(sym);
-    assert((type.isa<fir::SequenceType>() ||
-            (isPolymorphic && type.isa<fir::ClassType>())) &&
+    assert((mlir::isa<fir::SequenceType>(type) ||
+            (isPolymorphic && mlir::isa<fir::ClassType>(type))) &&
            "must be a sequence type");
     if (isPolymorphic)
       return type;
@@ -459,7 +459,7 @@ public:
       // (absent boxes are null descriptor addresses, not descriptors containing
       // a null base address).
       if (Fortran::semantics::IsOptional(sym)) {
-        auto boxTy = box.getType().cast<fir::BaseBoxType>();
+        auto boxTy = mlir::cast<fir::BaseBoxType>(box.getType());
         auto eleTy = boxTy.getEleTy();
         if (!fir::isa_ref_type(eleTy))
           eleTy = builder.getRefType(eleTy);
@@ -527,7 +527,7 @@ walkCaptureCategories(T visitor, Fortran::lower::AbstractConverter &converter,
 // `t` should be the result of getArgumentType, which has a type of
 // `!fir.ref<tuple<...>>`.
 static mlir::TupleType unwrapTupleTy(mlir::Type t) {
-  return fir::dyn_cast_ptrEleTy(t).cast<mlir::TupleType>();
+  return mlir::cast<mlir::TupleType>(fir::dyn_cast_ptrEleTy(t));
 }
 
 static mlir::Value genTupleCoor(fir::FirOpBuilder &builder, mlir::Location loc,
@@ -535,7 +535,7 @@ static mlir::Value genTupleCoor(fir::FirOpBuilder &builder, mlir::Location loc,
                                 mlir::Value offset) {
   // fir.ref<fir.ref> and fir.ptr<fir.ref> are forbidden. Use
   // fir.llvm_ptr if needed.
-  auto ty = varTy.isa<fir::ReferenceType>()
+  auto ty = mlir::isa<fir::ReferenceType>(varTy)
                 ? mlir::Type(fir::LLVMPointerType::get(varTy))
                 : mlir::Type(builder.getRefType(varTy));
   return builder.create<fir::CoordinateOp>(loc, ty, tupleArg, offset);
diff --git a/flang/lib/Lower/IO.cpp b/flang/lib/Lower/IO.cpp
index ac82276bcddb..ed0afad9197d 100644
--- a/flang/lib/Lower/IO.cpp
+++ b/flang/lib/Lower/IO.cpp
@@ -168,7 +168,7 @@ static constexpr fir::runtime::FuncTypeBuilderFunc getTypeModel() {
 }
 
 inline int64_t getLength(mlir::Type argTy) {
-  return argTy.cast<fir::SequenceType>().getShape()[0];
+  return mlir::cast<fir::SequenceType>(argTy).getShape()[0];
 }
 
 /// Get (or generate) the MLIR FuncOp for a given IO runtime function.
@@ -656,11 +656,11 @@ static void genNamelistIO(Fortran::lower::AbstractConverter &converter,
 static mlir::func::FuncOp getOutputFunc(mlir::Location loc,
                                         fir::FirOpBuilder &builder,
                                         mlir::Type type, bool isFormatted) {
-  if (fir::unwrapPassByRefType(type).isa<fir::RecordType>())
+  if (mlir::isa<fir::RecordType>(fir::unwrapPassByRefType(type)))
     return getIORuntimeFunc<mkIOKey(OutputDerivedType)>(loc, builder);
   if (!isFormatted)
     return getIORuntimeFunc<mkIOKey(OutputDescriptor)>(loc, builder);
-  if (auto ty = type.dyn_cast<mlir::IntegerType>()) {
+  if (auto ty = mlir::dyn_cast<mlir::IntegerType>(type)) {
     switch (ty.getWidth()) {
     case 1:
       return getIORuntimeFunc<mkIOKey(OutputLogical)>(loc, builder);
@@ -677,14 +677,14 @@ static mlir::func::FuncOp getOutputFunc(mlir::Location loc,
     }
     llvm_unreachable("unknown OutputInteger kind");
   }
-  if (auto ty = type.dyn_cast<mlir::FloatType>()) {
+  if (auto ty = mlir::dyn_cast<mlir::FloatType>(type)) {
     if (auto width = ty.getWidth(); width == 32)
       return getIORuntimeFunc<mkIOKey(OutputReal32)>(loc, builder);
     else if (width == 64)
       return getIORuntimeFunc<mkIOKey(OutputReal64)>(loc, builder);
   }
   auto kindMap = fir::getKindMapping(builder.getModule());
-  if (auto ty = type.dyn_cast<fir::ComplexType>()) {
+  if (auto ty = mlir::dyn_cast<fir::ComplexType>(type)) {
     // COMPLEX(KIND=k) corresponds to a pair of REAL(KIND=k).
     auto width = kindMap.getRealBitsize(ty.getFKind());
     if (width == 32)
@@ -692,7 +692,7 @@ static mlir::func::FuncOp getOutputFunc(mlir::Location loc,
     else if (width == 64)
       return getIORuntimeFunc<mkIOKey(OutputComplex64)>(loc, builder);
   }
-  if (type.isa<fir::LogicalType>())
+  if (mlir::isa<fir::LogicalType>(type))
     return getIORuntimeFunc<mkIOKey(OutputLogical)>(loc, builder);
   if (fir::factory::CharacterExprHelper::isCharacterScalar(type)) {
     // TODO: What would it mean if the default CHARACTER KIND is set to a wide
@@ -731,14 +731,14 @@ static void genOutputItemList(
     mlir::func::FuncOp outputFunc =
         getOutputFunc(loc, builder, itemTy, isFormatted);
     mlir::Type argType = outputFunc.getFunctionType().getInput(1);
-    assert((isFormatted || argType.isa<fir::BoxType>()) &&
+    assert((isFormatted || mlir::isa<fir::BoxType>(argType)) &&
            "expect descriptor for unformatted IO runtime");
     llvm::SmallVector<mlir::Value> outputFuncArgs = {cookie};
     fir::factory::CharacterExprHelper helper{builder, loc};
-    if (argType.isa<fir::BoxType>()) {
+    if (mlir::isa<fir::BoxType>(argType)) {
       mlir::Value box = fir::getBase(converter.genExprBox(loc, *expr, stmtCtx));
       outputFuncArgs.push_back(builder.createConvert(loc, argType, box));
-      if (fir::unwrapPassByRefType(itemTy).isa<fir::RecordType>())
+      if (mlir::isa<fir::RecordType>(fir::unwrapPassByRefType(itemTy)))
         outputFuncArgs.push_back(getNonTbpDefinedIoTableAddr(converter));
     } else if (helper.isCharacterScalar(itemTy)) {
       fir::ExtendedValue exv = converter.genExprAddr(loc, expr, stmtCtx);
@@ -773,29 +773,29 @@ static void genOutputItemList(
 static mlir::func::FuncOp getInputFunc(mlir::Location loc,
                                        fir::FirOpBuilder &builder,
                                        mlir::Type type, bool isFormatted) {
-  if (fir::unwrapPassByRefType(type).isa<fir::RecordType>())
+  if (mlir::isa<fir::RecordType>(fir::unwrapPassByRefType(type)))
     return getIORuntimeFunc<mkIOKey(InputDerivedType)>(loc, builder);
   if (!isFormatted)
     return getIORuntimeFunc<mkIOKey(InputDescriptor)>(loc, builder);
-  if (auto ty = type.dyn_cast<mlir::IntegerType>())
+  if (auto ty = mlir::dyn_cast<mlir::IntegerType>(type))
     return ty.getWidth() == 1
                ? getIORuntimeFunc<mkIOKey(InputLogical)>(loc, builder)
                : getIORuntimeFunc<mkIOKey(InputInteger)>(loc, builder);
-  if (auto ty = type.dyn_cast<mlir::FloatType>()) {
+  if (auto ty = mlir::dyn_cast<mlir::FloatType>(type)) {
     if (auto width = ty.getWidth(); width == 32)
       return getIORuntimeFunc<mkIOKey(InputReal32)>(loc, builder);
     else if (width == 64)
       return getIORuntimeFunc<mkIOKey(InputReal64)>(loc, builder);
   }
   auto kindMap = fir::getKindMapping(builder.getModule());
-  if (auto ty = type.dyn_cast<fir::ComplexType>()) {
+  if (auto ty = mlir::dyn_cast<fir::ComplexType>(type)) {
     auto width = kindMap.getRealBitsize(ty.getFKind());
     if (width == 32)
       return getIORuntimeFunc<mkIOKey(InputComplex32)>(loc, builder);
     else if (width == 64)
       return getIORuntimeFunc<mkIOKey(InputComplex64)>(loc, builder);
   }
-  if (type.isa<fir::LogicalType>())
+  if (mlir::isa<fir::LogicalType>(type))
     return getIORuntimeFunc<mkIOKey(InputLogical)>(loc, builder);
   if (fir::factory::CharacterExprHelper::isCharacterScalar(type)) {
     auto asciiKind = kindMap.defaultCharacterKind();
@@ -830,12 +830,12 @@ createIoRuntimeCallForItem(Fortran::lower::AbstractConverter &converter,
   fir::FirOpBuilder &builder = converter.getFirOpBuilder();
   mlir::Type argType = inputFunc.getFunctionType().getInput(1);
   llvm::SmallVector<mlir::Value> inputFuncArgs = {cookie};
-  if (argType.isa<fir::BaseBoxType>()) {
+  if (mlir::isa<fir::BaseBoxType>(argType)) {
     mlir::Value box = fir::getBase(item);
-    auto boxTy = box.getType().dyn_cast<fir::BaseBoxType>();
+    auto boxTy = mlir::dyn_cast<fir::BaseBoxType>(box.getType());
     assert(boxTy && "must be previously emboxed");
     inputFuncArgs.push_back(builder.createConvert(loc, argType, box));
-    if (fir::unwrapPassByRefType(boxTy).isa<fir::RecordType>())
+    if (mlir::isa<fir::RecordType>(fir::unwrapPassByRefType(boxTy)))
       inputFuncArgs.push_back(getNonTbpDefinedIoTableAddr(converter));
   } else {
     mlir::Value itemAddr = fir::getBase(item);
@@ -846,16 +846,16 @@ createIoRuntimeCallForItem(Fortran::lower::AbstractConverter &converter,
       mlir::Value len = fir::getLen(item);
       inputFuncArgs.push_back(builder.createConvert(
           loc, inputFunc.getFunctionType().getInput(2), len));
-    } else if (itemTy.isa<mlir::IntegerType>()) {
+    } else if (mlir::isa<mlir::IntegerType>(itemTy)) {
       inputFuncArgs.push_back(builder.create<mlir::arith::ConstantOp>(
           loc, builder.getI32IntegerAttr(
-                   itemTy.cast<mlir::IntegerType>().getWidth() / 8)));
+                   mlir::cast<mlir::IntegerType>(itemTy).getWidth() / 8)));
     }
   }
   auto call = builder.create<fir::CallOp>(loc, inputFunc, inputFuncArgs);
   auto itemAddr = fir::getBase(item);
   auto itemTy = fir::unwrapRefType(itemAddr.getType());
-  if (itemTy.isa<fir::LogicalType>())
+  if (mlir::isa<fir::LogicalType>(itemTy))
     boolRefToLogical(loc, builder, itemAddr);
   return call.getResult(0);
 }
@@ -886,7 +886,7 @@ static void genInputItemList(Fortran::lower::AbstractConverter &converter,
       mlir::func::FuncOp inputFunc = getInputFunc(
           loc, builder, vectorSubscriptBox.getElementType(), isFormatted);
       const bool mustBox =
-          inputFunc.getFunctionType().getInput(1).isa<fir::BoxType>();
+          mlir::isa<fir::BoxType>(inputFunc.getFunctionType().getInput(1));
       if (!checkResult) {
         auto elementalGenerator = [&](const fir::ExtendedValue &element) {
           createIoRuntimeCallForItem(converter, loc, inputFunc, cookie,
@@ -911,9 +911,10 @@ static void genInputItemList(Fortran::lower::AbstractConverter &converter,
     mlir::Type itemTy = converter.genType(*expr);
     mlir::func::FuncOp inputFunc =
         getInputFunc(loc, builder, itemTy, isFormatted);
-    auto itemExv = inputFunc.getFunctionType().getInput(1).isa<fir::BoxType>()
-                       ? converter.genExprBox(loc, *expr, stmtCtx)
-                       : converter.genExprAddr(loc, expr, stmtCtx);
+    auto itemExv =
+        mlir::isa<fir::BoxType>(inputFunc.getFunctionType().getInput(1))
+            ? converter.genExprBox(loc, *expr, stmtCtx)
+            : converter.genExprAddr(loc, expr, stmtCtx);
     ok = createIoRuntimeCallForItem(converter, loc, inputFunc, cookie, itemExv);
   }
 }
@@ -1772,8 +1773,8 @@ static mlir::Value genIOUnitNumber(Fortran::lower::AbstractConverter &converter,
   auto &builder = converter.getFirOpBuilder();
   auto rawUnit = fir::getBase(converter.genExprValue(loc, iounit, stmtCtx));
   unsigned rawUnitWidth =
-      rawUnit.getType().cast<mlir::IntegerType>().getWidth();
-  unsigned runtimeArgWidth = ty.cast<mlir::IntegerType>().getWidth();
+      mlir::cast<mlir::IntegerType>(rawUnit.getType()).getWidth();
+  unsigned runtimeArgWidth = mlir::cast<mlir::IntegerType>(ty).getWidth();
   // The IO runtime supports `int` unit numbers, if the unit number may
   // overflow when passed to the IO runtime, check that the unit number is
   // in range before calling the BeginXXX.
@@ -2331,7 +2332,7 @@ mlir::Value genInquireSpec<Fortran::parser::InquireSpec::IntVar>(
   if (!eleTy)
     fir::emitFatalError(loc,
                         "internal error: expected a memory reference type");
-  auto width = eleTy.cast<mlir::IntegerType>().getWidth();
+  auto width = mlir::cast<mlir::IntegerType>(eleTy).getWidth();
   mlir::IndexType idxTy = builder.getIndexType();
   mlir::Value kind = builder.createIntegerConstant(loc, idxTy, width / 8);
   llvm::SmallVector<mlir::Value> args = {
diff --git a/flang/lib/Lower/OpenACC.cpp b/flang/lib/Lower/OpenACC.cpp
index d933c07aba0e..eae2afc760e6 100644
--- a/flang/lib/Lower/OpenACC.cpp
+++ b/flang/lib/Lower/OpenACC.cpp
@@ -65,7 +65,7 @@ static Op createDataEntryOp(fir::FirOpBuilder &builder, mlir::Location loc,
                             mlir::acc::DataClause dataClause, mlir::Type retTy,
                             mlir::Value isPresent = {}) {
   mlir::Value varPtrPtr;
-  if (auto boxTy = baseAddr.getType().dyn_cast<fir::BaseBoxType>()) {
+  if (auto boxTy = mlir::dyn_cast<fir::BaseBoxType>(baseAddr.getType())) {
     if (isPresent) {
       mlir::Type ifRetTy = boxTy.getEleTy();
       if (!fir::isa_ref_type(ifRetTy))
@@ -2658,7 +2658,7 @@ genACCHostDataOp(Fortran::lower::AbstractConverter &converter,
   if (ifCond) {
     if (auto cst =
             mlir::dyn_cast<mlir::arith::ConstantOp>(ifCond.getDefiningOp()))
-      if (auto boolAttr = cst.getValue().dyn_cast<mlir::BoolAttr>()) {
+      if (auto boolAttr = mlir::dyn_cast<mlir::BoolAttr>(cst.getValue())) {
         if (boolAttr.getValue()) {
           // get rid of the if condition if it is always true.
           ifCond = mlir::Value();
@@ -4187,21 +4187,27 @@ void Fortran::lower::attachDeclarePostDeallocAction(
 
   std::stringstream fctName;
   fctName << converter.mangleName(sym) << declarePostDeallocSuffix.str();
-  mlir::Operation &op = builder.getInsertionBlock()->back();
-  if (op.hasAttr(mlir::acc::getDeclareActionAttrName())) {
-    auto attr = op.getAttrOfType<mlir::acc::DeclareActionAttr>(
+  mlir::Operation *op = &builder.getInsertionBlock()->back();
+  if (auto resOp = mlir::dyn_cast<fir::ResultOp>(*op)) {
+    assert(resOp.getOperands().size() == 0 &&
+           "expect only fir.result op with no operand");
+    op = op->getPrevNode();
+  }
+  assert(op && "expect operation to attach the post deallocation action");
+  if (op->hasAttr(mlir::acc::getDeclareActionAttrName())) {
+    auto attr = op->getAttrOfType<mlir::acc::DeclareActionAttr>(
         mlir::acc::getDeclareActionAttrName());
-    op.setAttr(mlir::acc::getDeclareActionAttrName(),
-               mlir::acc::DeclareActionAttr::get(
-                   builder.getContext(), attr.getPreAlloc(),
-                   attr.getPostAlloc(), attr.getPreDealloc(),
-                   /*postDealloc=*/builder.getSymbolRefAttr(fctName.str())));
+    op->setAttr(mlir::acc::getDeclareActionAttrName(),
+                mlir::acc::DeclareActionAttr::get(
+                    builder.getContext(), attr.getPreAlloc(),
+                    attr.getPostAlloc(), attr.getPreDealloc(),
+                    /*postDealloc=*/builder.getSymbolRefAttr(fctName.str())));
   } else {
-    op.setAttr(mlir::acc::getDeclareActionAttrName(),
-               mlir::acc::DeclareActionAttr::get(
-                   builder.getContext(),
-                   /*preAlloc=*/{}, /*postAlloc=*/{}, /*preDealloc=*/{},
-                   /*postDealloc=*/builder.getSymbolRefAttr(fctName.str())));
+    op->setAttr(mlir::acc::getDeclareActionAttrName(),
+                mlir::acc::DeclareActionAttr::get(
+                    builder.getContext(),
+                    /*preAlloc=*/{}, /*postAlloc=*/{}, /*preDealloc=*/{},
+                    /*postDealloc=*/builder.getSymbolRefAttr(fctName.str())));
   }
 }
 
diff --git a/flang/lib/Lower/OpenMP/ClauseProcessor.cpp b/flang/lib/Lower/OpenMP/ClauseProcessor.cpp
index 4c51b61f6bf0..79525d6dfe7a 100644
--- a/flang/lib/Lower/OpenMP/ClauseProcessor.cpp
+++ b/flang/lib/Lower/OpenMP/ClauseProcessor.cpp
@@ -23,10 +23,10 @@ namespace omp {
 
 /// Check for unsupported map operand types.
 static void checkMapType(mlir::Location location, mlir::Type type) {
-  if (auto refType = type.dyn_cast<fir::ReferenceType>())
+  if (auto refType = mlir::dyn_cast<fir::ReferenceType>(type))
     type = refType.getElementType();
-  if (auto boxType = type.dyn_cast_or_null<fir::BoxType>())
-    if (!boxType.getElementType().isa<fir::PointerType>())
+  if (auto boxType = mlir::dyn_cast_or_null<fir::BoxType>(type))
+    if (!mlir::isa<fir::PointerType>(boxType.getElementType()))
       TODO(location, "OMPD_target_data MapOperand BoxType");
 }
 
@@ -814,7 +814,7 @@ createMapInfoOp(fir::FirOpBuilder &builder, mlir::Location loc,
                 llvm::ArrayRef<mlir::Value> members, uint64_t mapType,
                 mlir::omp::VariableCaptureKind mapCaptureType, mlir::Type retTy,
                 bool isVal) {
-  if (auto boxTy = baseAddr.getType().dyn_cast<fir::BaseBoxType>()) {
+  if (auto boxTy = mlir::dyn_cast<fir::BaseBoxType>(baseAddr.getType())) {
     baseAddr = builder.create<fir::BoxAddrOp>(loc, baseAddr);
     retTy = baseAddr.getType();
   }
diff --git a/flang/lib/Lower/OpenMP/DataSharingProcessor.cpp b/flang/lib/Lower/OpenMP/DataSharingProcessor.cpp
index 8bb2f83282b5..d94c32375c0d 100644
--- a/flang/lib/Lower/OpenMP/DataSharingProcessor.cpp
+++ b/flang/lib/Lower/OpenMP/DataSharingProcessor.cpp
@@ -135,6 +135,12 @@ void DataSharingProcessor::insertBarrier() {
 }
 
 void DataSharingProcessor::insertLastPrivateCompare(mlir::Operation *op) {
+  mlir::omp::LoopNestOp loopOp;
+  if (auto wrapper = mlir::dyn_cast<mlir::omp::LoopWrapperInterface>(op))
+    loopOp = wrapper.isWrapper()
+                 ? mlir::cast<mlir::omp::LoopNestOp>(wrapper.getWrappedLoop())
+                 : nullptr;
+
   bool cmpCreated = false;
   mlir::OpBuilder::InsertionGuard guard(firOpBuilder);
   for (const omp::Clause &clause : clauses) {
@@ -214,18 +220,20 @@ void DataSharingProcessor::insertLastPrivateCompare(mlir::Operation *op) {
       // Update the original variable just before exiting the worksharing
       // loop. Conversion as follows:
       //
-      //                       omp.wsloop {
-      // omp.wsloop {            ...
-      //    ...                  store
-      //    store       ===>     %v = arith.addi %iv, %step
-      //    omp.yield            %cmp = %step < 0 ? %v < %ub : %v > %ub
-      // }                       fir.if %cmp {
-      //                           fir.store %v to %loopIV
-      //                           ^%lpv_update_blk:
-      //                         }
-      //                         omp.yield
-      //                       }
-      //
+      // omp.wsloop {             omp.wsloop {
+      //   omp.loop_nest {          omp.loop_nest {
+      //     ...                      ...
+      //     store          ===>      store
+      //     omp.yield                %v = arith.addi %iv, %step
+      //   }                          %cmp = %step < 0 ? %v < %ub : %v > %ub
+      //   omp.terminator             fir.if %cmp {
+      // }                              fir.store %v to %loopIV
+      //                                ^%lpv_update_blk:
+      //                              }
+      //                              omp.yield
+      //                            }
+      //                            omp.terminator
+      //                          }
 
       // Only generate the compare once in presence of multiple LastPrivate
       // clauses.
@@ -233,14 +241,13 @@ void DataSharingProcessor::insertLastPrivateCompare(mlir::Operation *op) {
         continue;
       cmpCreated = true;
 
-      mlir::Location loc = op->getLoc();
-      mlir::Operation *lastOper = op->getRegion(0).back().getTerminator();
+      mlir::Location loc = loopOp.getLoc();
+      mlir::Operation *lastOper = loopOp.getRegion().back().getTerminator();
       firOpBuilder.setInsertionPoint(lastOper);
 
-      mlir::Value iv = op->getRegion(0).front().getArguments()[0];
-      mlir::Value ub =
-          mlir::dyn_cast<mlir::omp::WsloopOp>(op).getUpperBound()[0];
-      mlir::Value step = mlir::dyn_cast<mlir::omp::WsloopOp>(op).getStep()[0];
+      mlir::Value iv = loopOp.getIVs()[0];
+      mlir::Value ub = loopOp.getUpperBound()[0];
+      mlir::Value step = loopOp.getStep()[0];
 
       // v = iv + step
       // cmp = step < 0 ? v < ub : v > ub
@@ -259,7 +266,7 @@ void DataSharingProcessor::insertLastPrivateCompare(mlir::Operation *op) {
       auto ifOp = firOpBuilder.create<fir::IfOp>(loc, cmpOp, /*else*/ false);
       firOpBuilder.setInsertionPointToStart(&ifOp.getThenRegion().front());
       assert(loopIV && "loopIV was not set");
-      firOpBuilder.create<fir::StoreOp>(op->getLoc(), v, loopIV);
+      firOpBuilder.create<fir::StoreOp>(loopOp.getLoc(), v, loopIV);
       lastPrivIP = firOpBuilder.saveInsertionPoint();
     } else {
       TODO(converter.getCurrentLocation(),
@@ -332,6 +339,7 @@ void DataSharingProcessor::defaultPrivatize(
     if (!Fortran::semantics::IsProcedure(*sym) &&
         !sym->GetUltimate().has<Fortran::semantics::DerivedTypeDetails>() &&
         !sym->GetUltimate().has<Fortran::semantics::NamelistDetails>() &&
+        !Fortran::semantics::IsImpliedDoIndex(sym->GetUltimate()) &&
         !symbolsInNestedRegions.contains(sym) &&
         !symbolsInParentRegions.contains(sym) &&
         !privatizedSymbols.contains(sym))
diff --git a/flang/lib/Lower/OpenMP/OpenMP.cpp b/flang/lib/Lower/OpenMP/OpenMP.cpp
index e932f7c284bc..64ec5ae65c82 100644
--- a/flang/lib/Lower/OpenMP/OpenMP.cpp
+++ b/flang/lib/Lower/OpenMP/OpenMP.cpp
@@ -84,7 +84,7 @@ static fir::GlobalOp globalInitialization(
 
   // Create default initialization for non-character scalar.
   if (Fortran::semantics::IsAllocatableOrObjectPointer(&sym)) {
-    mlir::Type baseAddrType = ty.dyn_cast<fir::BoxType>().getEleTy();
+    mlir::Type baseAddrType = mlir::dyn_cast<fir::BoxType>(ty).getEleTy();
     Fortran::lower::createGlobalInitialization(
         firOpBuilder, global, [&](fir::FirOpBuilder &b) {
           mlir::Value nullAddr =
@@ -366,10 +366,29 @@ getDeclareTargetFunctionDevice(
   return std::nullopt;
 }
 
-static llvm::SmallVector<const Fortran::semantics::Symbol *>
+/// Set up the entry block of the given `omp.loop_nest` operation, adding a
+/// block argument for each loop induction variable and allocating and
+/// initializing a private value to hold each of them.
+///
+/// This function can also bind the symbols of any variables that should match
+/// block arguments on parent loop wrapper operations attached to the same
+/// loop. This allows the introduction of any necessary `hlfir.declare`
+/// operations inside of the entry block of the `omp.loop_nest` operation and
+/// not directly under any of the wrappers, which would invalidate them.
+///
+/// \param [in]          op - the loop nest operation.
+/// \param [in]   converter - PFT to MLIR conversion interface.
+/// \param [in]         loc - location.
+/// \param [in]        args - symbols of induction variables.
+/// \param [in] wrapperSyms - symbols of variables to be mapped to loop wrapper
+///                           entry block arguments.
+/// \param [in] wrapperArgs - entry block arguments of parent loop wrappers.
+static void
 genLoopVars(mlir::Operation *op, Fortran::lower::AbstractConverter &converter,
             mlir::Location &loc,
-            llvm::ArrayRef<const Fortran::semantics::Symbol *> args) {
+            llvm::ArrayRef<const Fortran::semantics::Symbol *> args,
+            llvm::ArrayRef<const Fortran::semantics::Symbol *> wrapperSyms = {},
+            llvm::ArrayRef<mlir::BlockArgument> wrapperArgs = {}) {
   fir::FirOpBuilder &firOpBuilder = converter.getFirOpBuilder();
   auto &region = op->getRegion(0);
 
@@ -380,6 +399,12 @@ genLoopVars(mlir::Operation *op, Fortran::lower::AbstractConverter &converter,
   llvm::SmallVector<mlir::Type> tiv(args.size(), loopVarType);
   llvm::SmallVector<mlir::Location> locs(args.size(), loc);
   firOpBuilder.createBlock(&region, {}, tiv, locs);
+
+  // Bind the entry block arguments of parent wrappers to the corresponding
+  // symbols.
+  for (auto [arg, prv] : llvm::zip_equal(wrapperSyms, wrapperArgs))
+    converter.bindSymbol(*arg, prv);
+
   // The argument is not currently in memory, so make a temporary for the
   // argument, and store it there, then bind that location to the argument.
   mlir::Operation *storeOp = nullptr;
@@ -389,7 +414,6 @@ genLoopVars(mlir::Operation *op, Fortran::lower::AbstractConverter &converter,
         createAndSetPrivatizedLoopVar(converter, loc, indexVal, argSymbol);
   }
   firOpBuilder.setInsertionPointAfter(storeOp);
-  return llvm::SmallVector<const Fortran::semantics::Symbol *>(args);
 }
 
 static void genReductionVars(
@@ -410,58 +434,6 @@ static void genReductionVars(
   }
 }
 
-static llvm::SmallVector<const Fortran::semantics::Symbol *>
-genLoopAndReductionVars(
-    mlir::Operation *op, Fortran::lower::AbstractConverter &converter,
-    mlir::Location &loc,
-    llvm::ArrayRef<const Fortran::semantics::Symbol *> loopArgs,
-    llvm::ArrayRef<const Fortran::semantics::Symbol *> reductionArgs,
-    llvm::ArrayRef<mlir::Type> reductionTypes) {
-  fir::FirOpBuilder &firOpBuilder = converter.getFirOpBuilder();
-
-  llvm::SmallVector<mlir::Type> blockArgTypes;
-  llvm::SmallVector<mlir::Location> blockArgLocs;
-  blockArgTypes.reserve(loopArgs.size() + reductionArgs.size());
-  blockArgLocs.reserve(blockArgTypes.size());
-  mlir::Block *entryBlock;
-
-  if (loopArgs.size()) {
-    std::size_t loopVarTypeSize = 0;
-    for (const Fortran::semantics::Symbol *arg : loopArgs)
-      loopVarTypeSize = std::max(loopVarTypeSize, arg->GetUltimate().size());
-    mlir::Type loopVarType = getLoopVarType(converter, loopVarTypeSize);
-    std::fill_n(std::back_inserter(blockArgTypes), loopArgs.size(),
-                loopVarType);
-    std::fill_n(std::back_inserter(blockArgLocs), loopArgs.size(), loc);
-  }
-  if (reductionArgs.size()) {
-    llvm::copy(reductionTypes, std::back_inserter(blockArgTypes));
-    std::fill_n(std::back_inserter(blockArgLocs), reductionArgs.size(), loc);
-  }
-  entryBlock = firOpBuilder.createBlock(&op->getRegion(0), {}, blockArgTypes,
-                                        blockArgLocs);
-  // The argument is not currently in memory, so make a temporary for the
-  // argument, and store it there, then bind that location to the argument.
-  if (loopArgs.size()) {
-    mlir::Operation *storeOp = nullptr;
-    for (auto [argIndex, argSymbol] : llvm::enumerate(loopArgs)) {
-      mlir::Value indexVal =
-          fir::getBase(op->getRegion(0).front().getArgument(argIndex));
-      storeOp =
-          createAndSetPrivatizedLoopVar(converter, loc, indexVal, argSymbol);
-    }
-    firOpBuilder.setInsertionPointAfter(storeOp);
-  }
-  // Bind the reduction arguments to their block arguments
-  for (auto [arg, prv] : llvm::zip_equal(
-           reductionArgs,
-           llvm::drop_begin(entryBlock->getArguments(), loopArgs.size()))) {
-    converter.bindSymbol(*arg, prv);
-  }
-
-  return llvm::SmallVector<const Fortran::semantics::Symbol *>(loopArgs);
-}
-
 static void
 markDeclareTarget(mlir::Operation *op,
                   Fortran::lower::AbstractConverter &converter,
@@ -806,7 +778,7 @@ static void genBodyOfTargetDataOp(
   for (auto [argIndex, argSymbol] : llvm::enumerate(useDeviceSymbols)) {
     const mlir::BlockArgument &arg = region.front().getArgument(argIndex);
     fir::ExtendedValue extVal = converter.getSymbolExtendedValue(*argSymbol);
-    if (auto refType = arg.getType().dyn_cast<fir::ReferenceType>()) {
+    if (auto refType = mlir::dyn_cast<fir::ReferenceType>(arg.getType())) {
       if (fir::isa_builtin_cptr_type(refType.getElementType())) {
         converter.bindSymbol(*argSymbol, arg);
       } else {
@@ -1270,20 +1242,16 @@ static void genTeamsClauses(Fortran::lower::AbstractConverter &converter,
 static void genWsloopClauses(
     Fortran::lower::AbstractConverter &converter,
     Fortran::semantics::SemanticsContext &semaCtx,
-    Fortran::lower::StatementContext &stmtCtx,
-    Fortran::lower::pft::Evaluation &eval, const List<Clause> &clauses,
+    Fortran::lower::StatementContext &stmtCtx, const List<Clause> &clauses,
     mlir::Location loc, mlir::omp::WsloopClauseOps &clauseOps,
-    llvm::SmallVectorImpl<const Fortran::semantics::Symbol *> &iv,
     llvm::SmallVectorImpl<mlir::Type> &reductionTypes,
     llvm::SmallVectorImpl<const Fortran::semantics::Symbol *> &reductionSyms) {
   fir::FirOpBuilder &firOpBuilder = converter.getFirOpBuilder();
   ClauseProcessor cp(converter, semaCtx, clauses);
-  cp.processCollapse(loc, eval, clauseOps, iv);
   cp.processNowait(clauseOps);
   cp.processOrdered(clauseOps);
   cp.processReduction(loc, clauseOps, &reductionTypes, &reductionSyms);
   cp.processSchedule(stmtCtx, clauseOps);
-  clauseOps.loopInclusiveAttr = firOpBuilder.getUnitAttr();
   // TODO Support delayed privatization.
 
   if (ReductionProcessor::doReductionByRef(clauseOps.reductionVars))
@@ -1526,7 +1494,8 @@ genSimdOp(Fortran::lower::AbstractConverter &converter,
   auto *nestedEval = getCollapsedLoopEval(eval, getCollapseValue(clauses));
 
   auto ivCallback = [&](mlir::Operation *op) {
-    return genLoopVars(op, converter, loc, iv);
+    genLoopVars(op, converter, loc, iv);
+    return iv;
   };
 
   createBodyOfOp(*loopOp,
@@ -1601,13 +1570,15 @@ genTargetOp(Fortran::lower::AbstractConverter &converter,
 
         Fortran::lower::AddrAndBoundsInfo info = getDataOperandBaseAddr(
             converter, firOpBuilder, sym, converter.getCurrentLocation());
-        if (fir::unwrapRefType(info.addr.getType()).isa<fir::BaseBoxType>())
+        if (mlir::isa<fir::BaseBoxType>(
+                fir::unwrapRefType(info.addr.getType())))
           bounds =
               Fortran::lower::genBoundsOpsFromBox<mlir::omp::MapBoundsOp,
                                                   mlir::omp::MapBoundsType>(
                   firOpBuilder, converter.getCurrentLocation(), converter,
                   dataExv, info);
-        if (fir::unwrapRefType(info.addr.getType()).isa<fir::SequenceType>()) {
+        if (mlir::isa<fir::SequenceType>(
+                fir::unwrapRefType(info.addr.getType()))) {
           bool dataExvIsAssumedSize =
               Fortran::semantics::IsAssumedSizeArray(sym.GetUltimate());
           bounds = Fortran::lower::genBaseBoundsOps<mlir::omp::MapBoundsOp,
@@ -1622,7 +1593,7 @@ genTargetOp(Fortran::lower::AbstractConverter &converter,
             mlir::omp::VariableCaptureKind::ByRef;
 
         mlir::Type eleType = baseOp.getType();
-        if (auto refType = baseOp.getType().dyn_cast<fir::ReferenceType>())
+        if (auto refType = mlir::dyn_cast<fir::ReferenceType>(baseOp.getType()))
           eleType = refType.getElementType();
 
         // If a variable is specified in declare target link and if device
@@ -1801,32 +1772,48 @@ genWsloopOp(Fortran::lower::AbstractConverter &converter,
             Fortran::semantics::SemanticsContext &semaCtx,
             Fortran::lower::pft::Evaluation &eval, mlir::Location loc,
             const List<Clause> &clauses) {
+  fir::FirOpBuilder &firOpBuilder = converter.getFirOpBuilder();
   DataSharingProcessor dsp(converter, semaCtx, clauses, eval);
   dsp.processStep1();
 
   Fortran::lower::StatementContext stmtCtx;
-  mlir::omp::WsloopClauseOps clauseOps;
+  mlir::omp::LoopNestClauseOps loopClauseOps;
+  mlir::omp::WsloopClauseOps wsClauseOps;
   llvm::SmallVector<const Fortran::semantics::Symbol *> iv;
   llvm::SmallVector<mlir::Type> reductionTypes;
   llvm::SmallVector<const Fortran::semantics::Symbol *> reductionSyms;
-  genWsloopClauses(converter, semaCtx, stmtCtx, eval, clauses, loc, clauseOps,
-                   iv, reductionTypes, reductionSyms);
+  genLoopNestClauses(converter, semaCtx, eval, clauses, loc, loopClauseOps, iv);
+  genWsloopClauses(converter, semaCtx, stmtCtx, clauses, loc, wsClauseOps,
+                   reductionTypes, reductionSyms);
+
+  // Create omp.wsloop wrapper and populate entry block arguments with reduction
+  // variables.
+  auto wsloopOp = firOpBuilder.create<mlir::omp::WsloopOp>(loc, wsClauseOps);
+  llvm::SmallVector<mlir::Location> reductionLocs(reductionSyms.size(), loc);
+  mlir::Block *wsloopEntryBlock = firOpBuilder.createBlock(
+      &wsloopOp.getRegion(), {}, reductionTypes, reductionLocs);
+  firOpBuilder.setInsertionPoint(
+      Fortran::lower::genOpenMPTerminator(firOpBuilder, wsloopOp, loc));
+
+  // Create nested omp.loop_nest and fill body with loop contents.
+  auto loopOp = firOpBuilder.create<mlir::omp::LoopNestOp>(loc, loopClauseOps);
 
   auto *nestedEval = getCollapsedLoopEval(eval, getCollapseValue(clauses));
 
   auto ivCallback = [&](mlir::Operation *op) {
-    return genLoopAndReductionVars(op, converter, loc, iv, reductionSyms,
-                                   reductionTypes);
+    genLoopVars(op, converter, loc, iv, reductionSyms,
+                wsloopEntryBlock->getArguments());
+    return iv;
   };
 
-  return genOpWithBody<mlir::omp::WsloopOp>(
-      OpWithBodyGenInfo(converter, semaCtx, loc, *nestedEval,
-                        llvm::omp::Directive::OMPD_do)
-          .setClauses(&clauses)
-          .setDataSharingProcessor(&dsp)
-          .setReductions(&reductionSyms, &reductionTypes)
-          .setGenRegionEntryCb(ivCallback),
-      clauseOps);
+  createBodyOfOp(*loopOp,
+                 OpWithBodyGenInfo(converter, semaCtx, loc, *nestedEval,
+                                   llvm::omp::Directive::OMPD_do)
+                     .setClauses(&clauses)
+                     .setDataSharingProcessor(&dsp)
+                     .setReductions(&reductionSyms, &reductionTypes)
+                     .setGenRegionEntryCb(ivCallback));
+  return wsloopOp;
 }
 
 //===----------------------------------------------------------------------===//
@@ -2482,8 +2469,8 @@ static void genOMP(Fortran::lower::AbstractConverter &converter,
 mlir::Operation *Fortran::lower::genOpenMPTerminator(fir::FirOpBuilder &builder,
                                                      mlir::Operation *op,
                                                      mlir::Location loc) {
-  if (mlir::isa<mlir::omp::WsloopOp, mlir::omp::DeclareReductionOp,
-                mlir::omp::AtomicUpdateOp, mlir::omp::LoopNestOp>(op))
+  if (mlir::isa<mlir::omp::AtomicUpdateOp, mlir::omp::DeclareReductionOp,
+                mlir::omp::LoopNestOp>(op))
     return builder.create<mlir::omp::YieldOp>(loc);
   return builder.create<mlir::omp::TerminatorOp>(loc);
 }
diff --git a/flang/lib/Lower/OpenMP/ReductionProcessor.cpp b/flang/lib/Lower/OpenMP/ReductionProcessor.cpp
index 9f8352a8025c..38edd1b46821 100644
--- a/flang/lib/Lower/OpenMP/ReductionProcessor.cpp
+++ b/flang/lib/Lower/OpenMP/ReductionProcessor.cpp
@@ -138,7 +138,7 @@ ReductionProcessor::getReductionInitValue(mlir::Location loc, mlir::Type type,
     TODO(loc, "Reduction of some types is not supported");
   switch (redId) {
   case ReductionIdentifier::MAX: {
-    if (auto ty = type.dyn_cast<mlir::FloatType>()) {
+    if (auto ty = mlir::dyn_cast<mlir::FloatType>(type)) {
       const llvm::fltSemantics &sem = ty.getFloatSemantics();
       return builder.createRealConstant(
           loc, type, llvm::APFloat::getLargest(sem, /*Negative=*/true));
@@ -148,7 +148,7 @@ ReductionProcessor::getReductionInitValue(mlir::Location loc, mlir::Type type,
     return builder.createIntegerConstant(loc, type, minInt);
   }
   case ReductionIdentifier::MIN: {
-    if (auto ty = type.dyn_cast<mlir::FloatType>()) {
+    if (auto ty = mlir::dyn_cast<mlir::FloatType>(type)) {
       const llvm::fltSemantics &sem = ty.getFloatSemantics();
       return builder.createRealConstant(
           loc, type, llvm::APFloat::getLargest(sem, /*Negative=*/false));
@@ -188,12 +188,12 @@ ReductionProcessor::getReductionInitValue(mlir::Location loc, mlir::Type type,
       return fir::factory::Complex{builder, loc}.createComplex(type, initRe,
                                                                initIm);
     }
-    if (type.isa<mlir::FloatType>())
+    if (mlir::isa<mlir::FloatType>(type))
       return builder.create<mlir::arith::ConstantOp>(
           loc, type,
           builder.getFloatAttr(type, (double)getOperationIdentity(redId, loc)));
 
-    if (type.isa<fir::LogicalType>()) {
+    if (mlir::isa<fir::LogicalType>(type)) {
       mlir::Value intConst = builder.create<mlir::arith::ConstantOp>(
           loc, builder.getI1Type(),
           builder.getIntegerAttr(builder.getI1Type(),
@@ -295,30 +295,15 @@ mlir::Value ReductionProcessor::createScalarCombiner(
   return reductionOp;
 }
 
-/// Create reduction combiner region for reduction variables which are boxed
-/// arrays
-static void genBoxCombiner(fir::FirOpBuilder &builder, mlir::Location loc,
-                           ReductionProcessor::ReductionIdentifier redId,
-                           fir::BaseBoxType boxTy, mlir::Value lhs,
-                           mlir::Value rhs) {
-  fir::SequenceType seqTy =
-      mlir::dyn_cast_or_null<fir::SequenceType>(boxTy.getEleTy());
-  // TODO: support allocatable arrays: !fir.box<!fir.heap<!fir.array<...>>>
-  if (!seqTy || seqTy.hasUnknownShape())
-    TODO(loc, "Unsupported boxed type in OpenMP reduction");
-
-  // load fir.ref<fir.box<...>>
-  mlir::Value lhsAddr = lhs;
-  lhs = builder.create<fir::LoadOp>(loc, lhs);
-  rhs = builder.create<fir::LoadOp>(loc, rhs);
-
-  const unsigned rank = seqTy.getDimension();
-  llvm::SmallVector<mlir::Value> extents;
-  extents.reserve(rank);
+/// Generate a fir::ShapeShift op describing the provided boxed array.
+static fir::ShapeShiftOp getShapeShift(fir::FirOpBuilder &builder,
+                                       mlir::Location loc, mlir::Value box) {
+  fir::SequenceType sequenceType = mlir::cast<fir::SequenceType>(
+      hlfir::getFortranElementOrSequenceType(box.getType()));
+  const unsigned rank = sequenceType.getDimension();
   llvm::SmallVector<mlir::Value> lbAndExtents;
   lbAndExtents.reserve(rank * 2);
 
-  // Get box lowerbounds and extents:
   mlir::Type idxTy = builder.getIndexType();
   for (unsigned i = 0; i < rank; ++i) {
     // TODO: ideally we want to hoist box reads out of the critical section.
@@ -326,8 +311,7 @@ static void genBoxCombiner(fir::FirOpBuilder &builder, mlir::Location loc,
     // OpenACC does
     mlir::Value dim = builder.createIntegerConstant(loc, idxTy, i);
     auto dimInfo =
-        builder.create<fir::BoxDimsOp>(loc, idxTy, idxTy, idxTy, lhs, dim);
-    extents.push_back(dimInfo.getExtent());
+        builder.create<fir::BoxDimsOp>(loc, idxTy, idxTy, idxTy, box, dim);
     lbAndExtents.push_back(dimInfo.getLowerBound());
     lbAndExtents.push_back(dimInfo.getExtent());
   }
@@ -335,15 +319,58 @@ static void genBoxCombiner(fir::FirOpBuilder &builder, mlir::Location loc,
   auto shapeShiftTy = fir::ShapeShiftType::get(builder.getContext(), rank);
   auto shapeShift =
       builder.create<fir::ShapeShiftOp>(loc, shapeShiftTy, lbAndExtents);
+  return shapeShift;
+}
+
+/// Create reduction combiner region for reduction variables which are boxed
+/// arrays
+static void genBoxCombiner(fir::FirOpBuilder &builder, mlir::Location loc,
+                           ReductionProcessor::ReductionIdentifier redId,
+                           fir::BaseBoxType boxTy, mlir::Value lhs,
+                           mlir::Value rhs) {
+  fir::SequenceType seqTy = mlir::dyn_cast_or_null<fir::SequenceType>(
+      fir::unwrapRefType(boxTy.getEleTy()));
+  fir::HeapType heapTy =
+      mlir::dyn_cast_or_null<fir::HeapType>(boxTy.getEleTy());
+  if ((!seqTy || seqTy.hasUnknownShape()) && !heapTy)
+    TODO(loc, "Unsupported boxed type in OpenMP reduction");
+
+  // load fir.ref<fir.box<...>>
+  mlir::Value lhsAddr = lhs;
+  lhs = builder.create<fir::LoadOp>(loc, lhs);
+  rhs = builder.create<fir::LoadOp>(loc, rhs);
+
+  if (heapTy && !seqTy) {
+    // get box contents (heap pointers)
+    lhs = builder.create<fir::BoxAddrOp>(loc, lhs);
+    rhs = builder.create<fir::BoxAddrOp>(loc, rhs);
+    mlir::Value lhsValAddr = lhs;
+
+    // load heap pointers
+    lhs = builder.create<fir::LoadOp>(loc, lhs);
+    rhs = builder.create<fir::LoadOp>(loc, rhs);
+
+    mlir::Value result = ReductionProcessor::createScalarCombiner(
+        builder, loc, redId, heapTy.getEleTy(), lhs, rhs);
+    builder.create<fir::StoreOp>(loc, result, lhsValAddr);
+    builder.create<mlir::omp::YieldOp>(loc, lhsAddr);
+    return;
+  }
+
+  fir::ShapeShiftOp shapeShift = getShapeShift(builder, loc, lhs);
 
   // Iterate over array elements, applying the equivalent scalar reduction:
 
+  // F2018 5.4.10.2: Unallocated allocatable variables may not be referenced
+  // and so no null check is needed here before indexing into the (possibly
+  // allocatable) arrays.
+
   // A hlfir::elemental here gets inlined with a temporary so create the
   // loop nest directly.
   // This function already controls all of the code in this region so we
   // know this won't miss any opportuinties for clever elemental inlining
-  hlfir::LoopNest nest =
-      hlfir::genLoopNest(loc, builder, extents, /*isUnordered=*/true);
+  hlfir::LoopNest nest = hlfir::genLoopNest(
+      loc, builder, shapeShift.getExtents(), /*isUnordered=*/true);
   builder.setInsertionPointToStart(nest.innerLoop.getBody());
   mlir::Type refTy = fir::ReferenceType::get(seqTy.getEleTy());
   auto lhsEleAddr = builder.create<fir::ArrayCoorOp>(
@@ -412,9 +439,11 @@ createReductionCleanupRegion(fir::FirOpBuilder &builder, mlir::Location loc,
 
   mlir::Type valTy = fir::unwrapRefType(redTy);
   if (auto boxTy = mlir::dyn_cast_or_null<fir::BaseBoxType>(valTy)) {
-    mlir::Type innerTy = fir::extractSequenceType(boxTy);
-    if (!mlir::isa<fir::SequenceType>(innerTy))
-      typeError();
+    if (!mlir::isa<fir::HeapType>(boxTy.getEleTy())) {
+      mlir::Type innerTy = fir::extractSequenceType(boxTy);
+      if (!mlir::isa<fir::SequenceType>(innerTy))
+        typeError();
+    }
 
     mlir::Value arg = block->getArgument(0);
     arg = builder.loadIfRef(loc, arg);
@@ -443,6 +472,19 @@ createReductionCleanupRegion(fir::FirOpBuilder &builder, mlir::Location loc,
   typeError();
 }
 
+// like fir::unwrapSeqOrBoxedSeqType except it also works for non-sequence boxes
+static mlir::Type unwrapSeqOrBoxedType(mlir::Type ty) {
+  if (auto seqTy = mlir::dyn_cast<fir::SequenceType>(ty))
+    return seqTy.getEleTy();
+  if (auto boxTy = mlir::dyn_cast<fir::BaseBoxType>(ty)) {
+    auto eleTy = fir::unwrapRefType(boxTy.getEleTy());
+    if (auto seqTy = mlir::dyn_cast<fir::SequenceType>(eleTy))
+      return seqTy.getEleTy();
+    return eleTy;
+  }
+  return ty;
+}
+
 static mlir::Value
 createReductionInitRegion(fir::FirOpBuilder &builder, mlir::Location loc,
                           mlir::omp::DeclareReductionOp &reductionDecl,
@@ -450,7 +492,7 @@ createReductionInitRegion(fir::FirOpBuilder &builder, mlir::Location loc,
                           mlir::Type type, bool isByRef) {
   mlir::Type ty = fir::unwrapRefType(type);
   mlir::Value initValue = ReductionProcessor::getReductionInitValue(
-      loc, fir::unwrapSeqOrBoxedSeqType(ty), redId, builder);
+      loc, unwrapSeqOrBoxedType(ty), redId, builder);
 
   if (fir::isa_trivial(ty)) {
     if (isByRef) {
@@ -462,15 +504,70 @@ createReductionInitRegion(fir::FirOpBuilder &builder, mlir::Location loc,
     return initValue;
   }
 
+  // check if an allocatable box is unallocated. If so, initialize the boxAlloca
+  // to be unallocated e.g.
+  // %box_alloca = fir.alloca !fir.box<!fir.heap<...>>
+  // %addr = fir.box_addr %box
+  // if (%addr == 0) {
+  //   %nullbox = fir.embox %addr
+  //   fir.store %nullbox to %box_alloca
+  // } else {
+  //   // ...
+  //   fir.store %something to %box_alloca
+  // }
+  // omp.yield %box_alloca
+  mlir::Value blockArg =
+      builder.loadIfRef(loc, builder.getBlock()->getArgument(0));
+  auto handleNullAllocatable = [&](mlir::Value boxAlloca) -> fir::IfOp {
+    mlir::Value addr = builder.create<fir::BoxAddrOp>(loc, blockArg);
+    mlir::Value isNotAllocated = builder.genIsNullAddr(loc, addr);
+    fir::IfOp ifOp = builder.create<fir::IfOp>(loc, isNotAllocated,
+                                               /*withElseRegion=*/true);
+    builder.setInsertionPointToStart(&ifOp.getThenRegion().front());
+    // just embox the null address and return
+    mlir::Value nullBox = builder.create<fir::EmboxOp>(loc, ty, addr);
+    builder.create<fir::StoreOp>(loc, nullBox, boxAlloca);
+    return ifOp;
+  };
+
   // all arrays are boxed
   if (auto boxTy = mlir::dyn_cast_or_null<fir::BaseBoxType>(ty)) {
-    assert(isByRef && "passing arrays by value is unsupported");
-    // TODO: support allocatable arrays: !fir.box<!fir.heap<!fir.array<...>>>
-    mlir::Type innerTy = fir::extractSequenceType(boxTy);
+    assert(isByRef && "passing boxes by value is unsupported");
+    bool isAllocatable = mlir::isa<fir::HeapType>(boxTy.getEleTy());
+    mlir::Value boxAlloca = builder.create<fir::AllocaOp>(loc, ty);
+    mlir::Type innerTy = fir::unwrapRefType(boxTy.getEleTy());
+    if (fir::isa_trivial(innerTy)) {
+      // boxed non-sequence value e.g. !fir.box<!fir.heap<i32>>
+      if (!isAllocatable)
+        TODO(loc, "Reduction of non-allocatable trivial typed box");
+
+      fir::IfOp ifUnallocated = handleNullAllocatable(boxAlloca);
+
+      builder.setInsertionPointToStart(&ifUnallocated.getElseRegion().front());
+      mlir::Value valAlloc = builder.create<fir::AllocMemOp>(loc, innerTy);
+      builder.createStoreWithConvert(loc, initValue, valAlloc);
+      mlir::Value box = builder.create<fir::EmboxOp>(loc, ty, valAlloc);
+      builder.create<fir::StoreOp>(loc, box, boxAlloca);
+
+      auto insPt = builder.saveInsertionPoint();
+      createReductionCleanupRegion(builder, loc, reductionDecl);
+      builder.restoreInsertionPoint(insPt);
+      builder.setInsertionPointAfter(ifUnallocated);
+      return boxAlloca;
+    }
+    innerTy = fir::extractSequenceType(boxTy);
     if (!mlir::isa<fir::SequenceType>(innerTy))
       TODO(loc, "Unsupported boxed type for reduction");
+
+    fir::IfOp ifUnallocated{nullptr};
+    if (isAllocatable) {
+      ifUnallocated = handleNullAllocatable(boxAlloca);
+      builder.setInsertionPointToStart(&ifUnallocated.getElseRegion().front());
+    }
+
     // Create the private copy from the initial fir.box:
-    hlfir::Entity source = hlfir::Entity{builder.getBlock()->getArgument(0)};
+    mlir::Value loadedBox = builder.loadIfRef(loc, blockArg);
+    hlfir::Entity source = hlfir::Entity{loadedBox};
 
     // Allocating on the heap in case the whole reduction is nested inside of a
     // loop
@@ -478,9 +575,10 @@ createReductionInitRegion(fir::FirOpBuilder &builder, mlir::Location loc,
     // work by inserting stacksave/stackrestore around the reduction in
     // openmpirbuilder
     auto [temp, needsDealloc] = createTempFromMold(loc, builder, source);
-    // if needsDealloc isn't statically false, add cleanup region. TODO: always
+    // if needsDealloc isn't statically false, add cleanup region. Always
     // do this for allocatable boxes because they might have been re-allocated
     // in the body of the loop/parallel region
+
     std::optional<int64_t> cstNeedsDealloc =
         fir::getIntIfConstant(needsDealloc);
     assert(cstNeedsDealloc.has_value() &&
@@ -488,13 +586,30 @@ createReductionInitRegion(fir::FirOpBuilder &builder, mlir::Location loc,
     if (cstNeedsDealloc.has_value() && *cstNeedsDealloc != false) {
       mlir::OpBuilder::InsertionGuard guard(builder);
       createReductionCleanupRegion(builder, loc, reductionDecl);
+    } else {
+      assert(!isAllocatable && "Allocatable arrays must be heap allocated");
     }
 
     // Put the temporary inside of a box:
-    hlfir::Entity box = hlfir::genVariableBox(loc, builder, temp);
+    // hlfir::genVariableBox doesn't handle non-default lower bounds
+    mlir::Value box;
+    fir::ShapeShiftOp shapeShift = getShapeShift(builder, loc, loadedBox);
+    mlir::Type boxType = loadedBox.getType();
+    if (mlir::isa<fir::BaseBoxType>(temp.getType()))
+      // the box created by the declare form createTempFromMold is missing lower
+      // bounds info
+      box = builder.create<fir::ReboxOp>(loc, boxType, temp, shapeShift,
+                                         /*shift=*/mlir::Value{});
+    else
+      box = builder.create<fir::EmboxOp>(
+          loc, boxType, temp, shapeShift,
+          /*slice=*/mlir::Value{},
+          /*typeParams=*/llvm::ArrayRef<mlir::Value>{});
+
     builder.create<hlfir::AssignOp>(loc, initValue, box);
-    mlir::Value boxAlloca = builder.create<fir::AllocaOp>(loc, ty);
     builder.create<fir::StoreOp>(loc, box, boxAlloca);
+    if (ifUnallocated)
+      builder.setInsertionPointAfter(ifUnallocated);
     return boxAlloca;
   }
 
@@ -675,7 +790,7 @@ void ReductionProcessor::addDeclareReduction(
     for (mlir::Value symVal : reductionVars) {
       auto redType = mlir::cast<fir::ReferenceType>(symVal.getType());
       const auto &kindMap = firOpBuilder.getKindMap();
-      if (redType.getEleTy().isa<fir::LogicalType>())
+      if (mlir::isa<fir::LogicalType>(redType.getEleTy()))
         decl = createDeclareReduction(firOpBuilder,
                                       getReductionName(intrinsicOp, kindMap,
                                                        firOpBuilder.getI1Type(),
@@ -701,7 +816,7 @@ void ReductionProcessor::addDeclareReduction(
         mlir::Value symVal = converter.getSymbolAddress(*symbol);
         if (auto declOp = symVal.getDefiningOp<hlfir::DeclareOp>())
           symVal = declOp.getBase();
-        auto redType = symVal.getType().cast<fir::ReferenceType>();
+        auto redType = mlir::cast<fir::ReferenceType>(symVal.getType());
         if (!redType.getEleTy().isIntOrIndexOrFloat())
           TODO(currentLocation, "User Defined Reduction on non-trivial type");
         decl = createDeclareReduction(
diff --git a/flang/lib/Lower/VectorSubscripts.cpp b/flang/lib/Lower/VectorSubscripts.cpp
index 7439b9f7df8f..d7a311d32d59 100644
--- a/flang/lib/Lower/VectorSubscripts.cpp
+++ b/flang/lib/Lower/VectorSubscripts.cpp
@@ -105,7 +105,7 @@ private:
   }
 
   mlir::Type gen(const Fortran::evaluate::Component &component) {
-    auto recTy = gen(component.base()).cast<fir::RecordType>();
+    auto recTy = mlir::cast<fir::RecordType>(gen(component.base()));
     const Fortran::semantics::Symbol &componentSymbol =
         component.GetLastSymbol();
     // Parent components will not be found here, they are not part
diff --git a/flang/lib/Optimizer/Analysis/AliasAnalysis.cpp b/flang/lib/Optimizer/Analysis/AliasAnalysis.cpp
index c403b9effbfa..f723e8f66e3e 100644
--- a/flang/lib/Optimizer/Analysis/AliasAnalysis.cpp
+++ b/flang/lib/Optimizer/Analysis/AliasAnalysis.cpp
@@ -68,7 +68,7 @@ bool AliasAnalysis::Source::isPointerReference(mlir::Type ty) {
   if (!eleTy)
     return false;
 
-  return fir::isPointerType(eleTy) || eleTy.isa<fir::PointerType>();
+  return fir::isPointerType(eleTy) || mlir::isa<fir::PointerType>(eleTy);
 }
 
 bool AliasAnalysis::Source::isTargetOrPointer() const {
@@ -81,7 +81,7 @@ bool AliasAnalysis::Source::isRecordWithPointerComponent() const {
   if (!eleTy)
     return false;
   // TO DO: Look for pointer components
-  return eleTy.isa<fir::RecordType>();
+  return mlir::isa<fir::RecordType>(eleTy);
 }
 
 AliasResult AliasAnalysis::alias(Value lhs, Value rhs) {
diff --git a/flang/lib/Optimizer/Builder/BoxValue.cpp b/flang/lib/Optimizer/Builder/BoxValue.cpp
index 361fa59e2040..a90ce5570de7 100644
--- a/flang/lib/Optimizer/Builder/BoxValue.cpp
+++ b/flang/lib/Optimizer/Builder/BoxValue.cpp
@@ -191,7 +191,7 @@ bool fir::MutableBoxValue::verify() const {
   mlir::Type type = fir::dyn_cast_ptrEleTy(getAddr().getType());
   if (!type)
     return false;
-  auto box = type.dyn_cast<fir::BaseBoxType>();
+  auto box = mlir::dyn_cast<fir::BaseBoxType>(type);
   if (!box)
     return false;
   // A boxed value always takes a memory reference,
@@ -210,7 +210,7 @@ bool fir::MutableBoxValue::verify() const {
 /// Debug verifier for BoxValue ctor. There is no guarantee this will
 /// always be called.
 bool fir::BoxValue::verify() const {
-  if (!addr.getType().isa<fir::BaseBoxType>())
+  if (!mlir::isa<fir::BaseBoxType>(addr.getType()))
     return false;
   if (!lbounds.empty() && lbounds.size() != rank())
     return false;
diff --git a/flang/lib/Optimizer/Builder/CMakeLists.txt b/flang/lib/Optimizer/Builder/CMakeLists.txt
index 06339b116cd8..6d0aeb429d35 100644
--- a/flang/lib/Optimizer/Builder/CMakeLists.txt
+++ b/flang/lib/Optimizer/Builder/CMakeLists.txt
@@ -23,6 +23,7 @@ add_flang_library(FIRBuilder
   Runtime/Execute.cpp
   Runtime/Inquiry.cpp
   Runtime/Intrinsics.cpp
+  Runtime/Main.cpp
   Runtime/Numeric.cpp
   Runtime/Pointer.cpp
   Runtime/Ragged.cpp
diff --git a/flang/lib/Optimizer/Builder/Character.cpp b/flang/lib/Optimizer/Builder/Character.cpp
index af0786809cc2..b7a7453efdb3 100644
--- a/flang/lib/Optimizer/Builder/Character.cpp
+++ b/flang/lib/Optimizer/Builder/Character.cpp
@@ -26,11 +26,11 @@
 
 /// Unwrap all the ref and box types and return the inner element type.
 static mlir::Type unwrapBoxAndRef(mlir::Type type) {
-  if (auto boxType = type.dyn_cast<fir::BoxCharType>())
+  if (auto boxType = mlir::dyn_cast<fir::BoxCharType>(type))
     return boxType.getEleTy();
   while (true) {
     type = fir::unwrapRefType(type);
-    if (auto boxTy = type.dyn_cast<fir::BoxType>())
+    if (auto boxTy = mlir::dyn_cast<fir::BoxType>(type))
       type = boxTy.getEleTy();
     else
       break;
@@ -41,19 +41,19 @@ static mlir::Type unwrapBoxAndRef(mlir::Type type) {
 /// Unwrap base fir.char<kind,len> type.
 static fir::CharacterType recoverCharacterType(mlir::Type type) {
   type = fir::unwrapSequenceType(unwrapBoxAndRef(type));
-  if (auto charTy = type.dyn_cast<fir::CharacterType>())
+  if (auto charTy = mlir::dyn_cast<fir::CharacterType>(type))
     return charTy;
   llvm::report_fatal_error("expected a character type");
 }
 
 bool fir::factory::CharacterExprHelper::isCharacterScalar(mlir::Type type) {
   type = unwrapBoxAndRef(type);
-  return !type.isa<fir::SequenceType>() && fir::isa_char(type);
+  return !mlir::isa<fir::SequenceType>(type) && fir::isa_char(type);
 }
 
 bool fir::factory::CharacterExprHelper::isArray(mlir::Type type) {
   type = unwrapBoxAndRef(type);
-  if (auto seqTy = type.dyn_cast<fir::SequenceType>())
+  if (auto seqTy = mlir::dyn_cast<fir::SequenceType>(type))
     return fir::isa_char(seqTy.getEleTy());
   return false;
 }
@@ -92,7 +92,8 @@ getCompileTimeLength(const fir::CharBoxValue &box) {
 /// Detect the precondition that the value `str` does not reside in memory. Such
 /// values will have a type `!fir.array<...x!fir.char<N>>` or `!fir.char<N>`.
 LLVM_ATTRIBUTE_UNUSED static bool needToMaterialize(mlir::Value str) {
-  return str.getType().isa<fir::SequenceType>() || fir::isa_char(str.getType());
+  return mlir::isa<fir::SequenceType>(str.getType()) ||
+         fir::isa_char(str.getType());
 }
 
 /// This is called only if `str` does not reside in memory. Such a bare string
@@ -103,7 +104,7 @@ fir::factory::CharacterExprHelper::materializeValue(mlir::Value str) {
   assert(needToMaterialize(str));
   auto ty = str.getType();
   assert(isCharacterScalar(ty) && "expected scalar character");
-  auto charTy = ty.dyn_cast<fir::CharacterType>();
+  auto charTy = mlir::dyn_cast<fir::CharacterType>(ty);
   if (!charTy || charTy.getLen() == fir::CharacterType::unknownLen()) {
     LLVM_DEBUG(llvm::dbgs() << "cannot materialize: " << str << '\n');
     llvm_unreachable("must be a !fir.char<N> type");
@@ -129,7 +130,7 @@ fir::factory::CharacterExprHelper::toExtendedValue(mlir::Value character,
   if (auto eleType = fir::dyn_cast_ptrEleTy(type))
     type = eleType;
 
-  if (auto arrayType = type.dyn_cast<fir::SequenceType>()) {
+  if (auto arrayType = mlir::dyn_cast<fir::SequenceType>(type)) {
     type = arrayType.getEleTy();
     auto indexType = builder.getIndexType();
     for (auto extent : arrayType.getShape()) {
@@ -145,10 +146,10 @@ fir::factory::CharacterExprHelper::toExtendedValue(mlir::Value character,
       mlir::emitError(loc, "cannot retrieve array extents from type");
   }
 
-  if (auto charTy = type.dyn_cast<fir::CharacterType>()) {
+  if (auto charTy = mlir::dyn_cast<fir::CharacterType>(type)) {
     if (!resultLen && charTy.getLen() != fir::CharacterType::unknownLen())
       resultLen = builder.createIntegerConstant(loc, lenType, charTy.getLen());
-  } else if (auto boxCharType = type.dyn_cast<fir::BoxCharType>()) {
+  } else if (auto boxCharType = mlir::dyn_cast<fir::BoxCharType>(type)) {
     auto refType = builder.getRefType(boxCharType.getEleTy());
     // If the embox is accessible, use its operand to avoid filling
     // the generated fir with embox/unbox.
@@ -168,7 +169,7 @@ fir::factory::CharacterExprHelper::toExtendedValue(mlir::Value character,
     if (!resultLen) {
       resultLen = boxCharLen;
     }
-  } else if (type.isa<fir::BoxType>()) {
+  } else if (mlir::isa<fir::BoxType>(type)) {
     mlir::emitError(loc, "descriptor or derived type not yet handled");
   } else {
     llvm_unreachable("Cannot translate mlir::Value to character ExtendedValue");
@@ -221,7 +222,7 @@ fir::factory::CharacterExprHelper::createEmbox(const fir::CharBoxValue &box) {
 
 fir::CharBoxValue fir::factory::CharacterExprHelper::toScalarCharacter(
     const fir::CharArrayBoxValue &box) {
-  if (box.getBuffer().getType().isa<fir::PointerType>())
+  if (mlir::isa<fir::PointerType>(box.getBuffer().getType()))
     TODO(loc, "concatenating non contiguous character array into a scalar");
 
   // TODO: add a fast path multiplying new length at compile time if the info is
@@ -655,7 +656,7 @@ fir::factory::CharacterExprHelper::createUnboxChar(mlir::Value boxChar) {
 }
 
 bool fir::factory::CharacterExprHelper::isCharacterLiteral(mlir::Type type) {
-  if (auto seqType = type.dyn_cast<fir::SequenceType>())
+  if (auto seqType = mlir::dyn_cast<fir::SequenceType>(type))
     return (seqType.getShape().size() == 1) &&
            fir::isa_char(seqType.getEleTy());
   return false;
@@ -728,9 +729,9 @@ mlir::Value fir::factory::CharacterExprHelper::getLength(mlir::Value memref) {
   if (charType.hasConstantLen())
     return builder.createIntegerConstant(loc, builder.getCharacterLengthType(),
                                          charType.getLen());
-  if (memrefType.isa<fir::BoxType>())
+  if (mlir::isa<fir::BoxType>(memrefType))
     return readLengthFromBox(memref);
-  if (memrefType.isa<fir::BoxCharType>())
+  if (mlir::isa<fir::BoxCharType>(memrefType))
     return createUnboxChar(memref).second;
 
   // Length cannot be deduced from memref.
@@ -742,14 +743,14 @@ fir::factory::extractCharacterProcedureTuple(fir::FirOpBuilder &builder,
                                              mlir::Location loc,
                                              mlir::Value tuple,
                                              bool openBoxProc) {
-  mlir::TupleType tupleType = tuple.getType().cast<mlir::TupleType>();
+  mlir::TupleType tupleType = mlir::cast<mlir::TupleType>(tuple.getType());
   mlir::Value addr = builder.create<fir::ExtractValueOp>(
       loc, tupleType.getType(0), tuple,
       builder.getArrayAttr(
           {builder.getIntegerAttr(builder.getIndexType(), 0)}));
   mlir::Value proc = [&]() -> mlir::Value {
     if (openBoxProc)
-      if (auto addrTy = addr.getType().dyn_cast<fir::BoxProcType>())
+      if (auto addrTy = mlir::dyn_cast<fir::BoxProcType>(addr.getType()))
         return builder.create<fir::BoxAddrOp>(loc, addrTy.getEleTy(), addr);
     return addr;
   }();
@@ -763,7 +764,7 @@ fir::factory::extractCharacterProcedureTuple(fir::FirOpBuilder &builder,
 mlir::Value fir::factory::createCharacterProcedureTuple(
     fir::FirOpBuilder &builder, mlir::Location loc, mlir::Type argTy,
     mlir::Value addr, mlir::Value len) {
-  mlir::TupleType tupleType = argTy.cast<mlir::TupleType>();
+  mlir::TupleType tupleType = mlir::cast<mlir::TupleType>(argTy);
   addr = builder.createConvert(loc, tupleType.getType(0), addr);
   if (len)
     len = builder.createConvert(loc, tupleType.getType(1), len);
@@ -866,7 +867,7 @@ fir::factory::convertCharacterKind(fir::FirOpBuilder &builder,
   auto kindMap = builder.getKindMap();
   mlir::Value boxCharAddr = srcBoxChar.getAddr();
   auto fromTy = boxCharAddr.getType();
-  if (auto charTy = fromTy.dyn_cast<fir::CharacterType>()) {
+  if (auto charTy = mlir::dyn_cast<fir::CharacterType>(fromTy)) {
     // boxchar is a value, not a variable. Turn it into a temporary.
     // As a value, it ought to have a constant LEN value.
     assert(charTy.hasConstantLen() && "must have constant length");
@@ -875,7 +876,7 @@ fir::factory::convertCharacterKind(fir::FirOpBuilder &builder,
     boxCharAddr = tmp;
   }
   auto fromBits = kindMap.getCharacterBitsize(
-      fir::unwrapRefType(fromTy).cast<fir::CharacterType>().getFKind());
+      mlir::cast<fir::CharacterType>(fir::unwrapRefType(fromTy)).getFKind());
   auto toBits = kindMap.getCharacterBitsize(toKind);
   if (toBits < fromBits) {
     // Scale by relative ratio to give a buffer of the same length.
diff --git a/flang/lib/Optimizer/Builder/Complex.cpp b/flang/lib/Optimizer/Builder/Complex.cpp
index e97cb3067808..cbcd4f850014 100644
--- a/flang/lib/Optimizer/Builder/Complex.cpp
+++ b/flang/lib/Optimizer/Builder/Complex.cpp
@@ -14,7 +14,8 @@
 
 mlir::Type
 fir::factory::Complex::getComplexPartType(mlir::Type complexType) const {
-  return builder.getRealType(complexType.cast<fir::ComplexType>().getFKind());
+  return builder.getRealType(
+      mlir::cast<fir::ComplexType>(complexType).getFKind());
 }
 
 mlir::Type fir::factory::Complex::getComplexPartType(mlir::Value cplx) const {
diff --git a/flang/lib/Optimizer/Builder/FIRBuilder.cpp b/flang/lib/Optimizer/Builder/FIRBuilder.cpp
index b09da4929a8a..a6da38763726 100644
--- a/flang/lib/Optimizer/Builder/FIRBuilder.cpp
+++ b/flang/lib/Optimizer/Builder/FIRBuilder.cpp
@@ -90,7 +90,7 @@ fir::FirOpBuilder::getNamedGlobal(mlir::ModuleOp modOp,
 }
 
 mlir::Type fir::FirOpBuilder::getRefType(mlir::Type eleTy) {
-  assert(!eleTy.isa<fir::ReferenceType>() && "cannot be a reference type");
+  assert(!mlir::isa<fir::ReferenceType>(eleTy) && "cannot be a reference type");
   return fir::ReferenceType::get(eleTy);
 }
 
@@ -147,7 +147,7 @@ mlir::Value
 fir::FirOpBuilder::createRealConstant(mlir::Location loc, mlir::Type fltTy,
                                       llvm::APFloat::integerPart val) {
   auto apf = [&]() -> llvm::APFloat {
-    if (auto ty = fltTy.dyn_cast<fir::RealType>())
+    if (auto ty = mlir::dyn_cast<fir::RealType>(fltTy))
       return llvm::APFloat(kindMap.getFloatSemantics(ty.getFKind()), val);
     if (fltTy.isF16())
       return llvm::APFloat(llvm::APFloat::IEEEhalf(), val);
@@ -169,7 +169,7 @@ fir::FirOpBuilder::createRealConstant(mlir::Location loc, mlir::Type fltTy,
 mlir::Value fir::FirOpBuilder::createRealConstant(mlir::Location loc,
                                                   mlir::Type fltTy,
                                                   const llvm::APFloat &value) {
-  if (fltTy.isa<mlir::FloatType>()) {
+  if (mlir::isa<mlir::FloatType>(fltTy)) {
     auto attr = getFloatAttr(fltTy, value);
     return create<mlir::arith::ConstantOp>(loc, fltTy, attr);
   }
@@ -178,7 +178,7 @@ mlir::Value fir::FirOpBuilder::createRealConstant(mlir::Location loc,
 
 static llvm::SmallVector<mlir::Value>
 elideExtentsAlreadyInType(mlir::Type type, mlir::ValueRange shape) {
-  auto arrTy = type.dyn_cast<fir::SequenceType>();
+  auto arrTy = mlir::dyn_cast<fir::SequenceType>(type);
   if (shape.empty() || !arrTy)
     return {};
   // elide the constant dimensions before construction
@@ -195,7 +195,7 @@ static llvm::SmallVector<mlir::Value>
 elideLengthsAlreadyInType(mlir::Type type, mlir::ValueRange lenParams) {
   if (lenParams.empty())
     return {};
-  if (auto arrTy = type.dyn_cast<fir::SequenceType>())
+  if (auto arrTy = mlir::dyn_cast<fir::SequenceType>(type))
     type = arrTy.getEleTy();
   if (fir::hasDynamicSize(type))
     return lenParams;
@@ -250,7 +250,7 @@ mlir::Block *fir::FirOpBuilder::getAllocaBlock() {
               .getParentOfType<mlir::omp::OutlineableOpenMPOpInterface>()) {
     return ompOutlineableIface.getAllocaBlock();
   }
-  if (mlir::isa<mlir::omp::DeclareReductionOp>(getRegion().getParentOp()))
+  if (getRegion().getParentOfType<mlir::omp::DeclareReductionOp>())
     return &getRegion().front();
   if (auto accRecipeIface =
           getRegion().getParentOfType<mlir::acc::RecipeInterface>()) {
@@ -264,7 +264,7 @@ mlir::Value fir::FirOpBuilder::createTemporaryAlloc(
     mlir::Location loc, mlir::Type type, llvm::StringRef name,
     mlir::ValueRange lenParams, mlir::ValueRange shape,
     llvm::ArrayRef<mlir::NamedAttribute> attrs) {
-  assert(!type.isa<fir::ReferenceType>() && "cannot be a reference");
+  assert(!mlir::isa<fir::ReferenceType>(type) && "cannot be a reference");
   // If the alloca is inside an OpenMP Op which will be outlined then pin
   // the alloca here.
   const bool pinned =
@@ -310,7 +310,7 @@ mlir::Value fir::FirOpBuilder::createHeapTemporary(
   llvm::SmallVector<mlir::Value> dynamicLength =
       elideLengthsAlreadyInType(type, lenParams);
 
-  assert(!type.isa<fir::ReferenceType>() && "cannot be a reference");
+  assert(!mlir::isa<fir::ReferenceType>(type) && "cannot be a reference");
   return create<fir::AllocMemOp>(loc, type, /*unique_name=*/llvm::StringRef{},
                                  name, dynamicLength, dynamicShape, attrs);
 }
@@ -376,8 +376,9 @@ mlir::Value fir::FirOpBuilder::convertWithSemantics(
     // imaginary part is zero
     auto eleTy = helper.getComplexPartType(toTy);
     auto cast = createConvert(loc, eleTy, val);
-    llvm::APFloat zero{
-        kindMap.getFloatSemantics(toTy.cast<fir::ComplexType>().getFKind()), 0};
+    llvm::APFloat zero{kindMap.getFloatSemantics(
+                           mlir::cast<fir::ComplexType>(toTy).getFKind()),
+                       0};
     auto imag = createRealConstant(loc, eleTy, zero);
     return helper.createComplex(toTy, cast, imag);
   }
@@ -388,14 +389,14 @@ mlir::Value fir::FirOpBuilder::convertWithSemantics(
     return createConvert(loc, toTy, rp);
   }
   if (allowCharacterConversion) {
-    if (fromTy.isa<fir::BoxCharType>()) {
+    if (mlir::isa<fir::BoxCharType>(fromTy)) {
       // Extract the address of the character string and pass it
       fir::factory::CharacterExprHelper charHelper{*this, loc};
       std::pair<mlir::Value, mlir::Value> unboxchar =
           charHelper.createUnboxChar(val);
       return createConvert(loc, toTy, unboxchar.first);
     }
-    if (auto boxType = toTy.dyn_cast<fir::BoxCharType>()) {
+    if (auto boxType = mlir::dyn_cast<fir::BoxCharType>(toTy)) {
       // Extract the address of the actual argument and create a boxed
       // character value with an undefined length
       // TODO: We should really calculate the total size of the actual
@@ -415,10 +416,10 @@ mlir::Value fir::FirOpBuilder::convertWithSemantics(
             "element types expected to match"));
     return create<fir::BoxAddrOp>(loc, toTy, val);
   }
-  if (fir::isa_ref_type(fromTy) && toTy.isa<fir::BoxProcType>()) {
+  if (fir::isa_ref_type(fromTy) && mlir::isa<fir::BoxProcType>(toTy)) {
     // Call is expecting a boxed procedure, not a reference to other data type.
     // Convert the reference to a procedure and embox it.
-    mlir::Type procTy = toTy.cast<fir::BoxProcType>().getEleTy();
+    mlir::Type procTy = mlir::cast<fir::BoxProcType>(toTy).getEleTy();
     mlir::Value proc = createConvert(loc, procTy, val);
     return create<fir::EmboxProcOp>(loc, toTy, proc);
   }
@@ -428,7 +429,7 @@ mlir::Value fir::FirOpBuilder::convertWithSemantics(
     if (((fir::isPolymorphicType(fromTy) &&
           (fir::isAllocatableType(fromTy) || fir::isPointerType(fromTy)) &&
           fir::isPolymorphicType(toTy)) ||
-         (fir::isPolymorphicType(fromTy) && toTy.isa<fir::BoxType>())) &&
+         (fir::isPolymorphicType(fromTy) && mlir::isa<fir::BoxType>(toTy))) &&
         !(fir::isUnlimitedPolymorphicType(fromTy) && fir::isAssumedType(toTy)))
       return create<fir::ReboxOp>(loc, toTy, val, mlir::Value{},
                                   /*slice=*/mlir::Value{});
@@ -581,7 +582,7 @@ mlir::Value fir::FirOpBuilder::createBox(mlir::Location loc,
                                          bool isPolymorphic,
                                          bool isAssumedType) {
   mlir::Value itemAddr = fir::getBase(exv);
-  if (itemAddr.getType().isa<fir::BaseBoxType>())
+  if (mlir::isa<fir::BaseBoxType>(itemAddr.getType()))
     return itemAddr;
   auto elementType = fir::dyn_cast_ptrEleTy(itemAddr.getType());
   if (!elementType) {
@@ -592,7 +593,7 @@ mlir::Value fir::FirOpBuilder::createBox(mlir::Location loc,
   mlir::Type boxTy;
   mlir::Value tdesc;
   // Avoid to wrap a box/class with box/class.
-  if (elementType.isa<fir::BaseBoxType>()) {
+  if (mlir::isa<fir::BaseBoxType>(elementType)) {
     boxTy = elementType;
   } else {
     boxTy = fir::BoxType::get(elementType);
@@ -709,7 +710,7 @@ mlir::Value fir::FirOpBuilder::genAbsentOp(mlir::Location loc,
     return create<fir::AbsentOp>(loc, argTy);
 
   auto boxProc =
-      create<fir::AbsentOp>(loc, argTy.cast<mlir::TupleType>().getType(0));
+      create<fir::AbsentOp>(loc, mlir::cast<mlir::TupleType>(argTy).getType(0));
   mlir::Value charLen = create<fir::UndefOp>(loc, getCharacterLengthType());
   return fir::factory::createCharacterProcedureTuple(*this, loc, argTy, boxProc,
                                                      charLen);
@@ -958,14 +959,14 @@ static llvm::SmallVector<mlir::Value> getFromBox(mlir::Location loc,
                                                  fir::FirOpBuilder &builder,
                                                  mlir::Type valTy,
                                                  mlir::Value boxVal) {
-  if (auto boxTy = valTy.dyn_cast<fir::BaseBoxType>()) {
+  if (auto boxTy = mlir::dyn_cast<fir::BaseBoxType>(valTy)) {
     auto eleTy = fir::unwrapAllRefAndSeqType(boxTy.getEleTy());
-    if (auto recTy = eleTy.dyn_cast<fir::RecordType>()) {
+    if (auto recTy = mlir::dyn_cast<fir::RecordType>(eleTy)) {
       if (recTy.getNumLenParams() > 0) {
         // Walk each type parameter in the record and get the value.
         TODO(loc, "generate code to get LEN type parameters");
       }
-    } else if (auto charTy = eleTy.dyn_cast<fir::CharacterType>()) {
+    } else if (auto charTy = mlir::dyn_cast<fir::CharacterType>(eleTy)) {
       if (charTy.hasDynamicLen()) {
         auto idxTy = builder.getIndexType();
         auto eleSz = builder.create<fir::BoxEleSizeOp>(loc, idxTy, boxVal);
@@ -1012,7 +1013,7 @@ llvm::SmallVector<mlir::Value>
 fir::factory::getTypeParams(mlir::Location loc, fir::FirOpBuilder &builder,
                             fir::ArrayLoadOp load) {
   mlir::Type memTy = load.getMemref().getType();
-  if (auto boxTy = memTy.dyn_cast<fir::BaseBoxType>())
+  if (auto boxTy = mlir::dyn_cast<fir::BaseBoxType>(memTy))
     return getFromBox(loc, builder, boxTy, load.getMemref());
   return load.getTypeparams();
 }
@@ -1039,7 +1040,7 @@ std::string fir::factory::uniqueCGIdent(llvm::StringRef prefix,
 
 mlir::Value fir::factory::locationToFilename(fir::FirOpBuilder &builder,
                                              mlir::Location loc) {
-  if (auto flc = loc.dyn_cast<mlir::FileLineColLoc>()) {
+  if (auto flc = mlir::dyn_cast<mlir::FileLineColLoc>(loc)) {
     // must be encoded as asciiz, C string
     auto fn = flc.getFilename().str() + '\0';
     return fir::getBase(createStringLiteral(builder, loc, fn));
@@ -1050,7 +1051,7 @@ mlir::Value fir::factory::locationToFilename(fir::FirOpBuilder &builder,
 mlir::Value fir::factory::locationToLineNo(fir::FirOpBuilder &builder,
                                            mlir::Location loc,
                                            mlir::Type type) {
-  if (auto flc = loc.dyn_cast<mlir::FileLineColLoc>())
+  if (auto flc = mlir::dyn_cast<mlir::FileLineColLoc>(loc))
     return builder.createIntegerConstant(loc, type, flc.getLine());
   return builder.createIntegerConstant(loc, type, 0);
 }
@@ -1108,10 +1109,10 @@ fir::ExtendedValue fir::factory::componentToExtendedValue(
   auto fieldTy = component.getType();
   if (auto ty = fir::dyn_cast_ptrEleTy(fieldTy))
     fieldTy = ty;
-  if (fieldTy.isa<fir::BaseBoxType>()) {
+  if (mlir::isa<fir::BaseBoxType>(fieldTy)) {
     llvm::SmallVector<mlir::Value> nonDeferredTypeParams;
     auto eleTy = fir::unwrapSequenceType(fir::dyn_cast_ptrOrBoxEleTy(fieldTy));
-    if (auto charTy = eleTy.dyn_cast<fir::CharacterType>()) {
+    if (auto charTy = mlir::dyn_cast<fir::CharacterType>(eleTy)) {
       auto lenTy = builder.getCharacterLengthType();
       if (charTy.hasConstantLen())
         nonDeferredTypeParams.emplace_back(
@@ -1120,7 +1121,7 @@ fir::ExtendedValue fir::factory::componentToExtendedValue(
       // on a PDT length parameter. There is no way to make a difference with
       // deferred length here yet.
     }
-    if (auto recTy = eleTy.dyn_cast<fir::RecordType>())
+    if (auto recTy = mlir::dyn_cast<fir::RecordType>(eleTy))
       if (recTy.getNumLenParams() > 0)
         TODO(loc, "allocatable and pointer components non deferred length "
                   "parameters");
@@ -1129,7 +1130,7 @@ fir::ExtendedValue fir::factory::componentToExtendedValue(
                                 /*mutableProperties=*/{});
   }
   llvm::SmallVector<mlir::Value> extents;
-  if (auto seqTy = fieldTy.dyn_cast<fir::SequenceType>()) {
+  if (auto seqTy = mlir::dyn_cast<fir::SequenceType>(fieldTy)) {
     fieldTy = seqTy.getEleTy();
     auto idxTy = builder.getIndexType();
     for (auto extent : seqTy.getShape()) {
@@ -1138,7 +1139,7 @@ fir::ExtendedValue fir::factory::componentToExtendedValue(
       extents.emplace_back(builder.createIntegerConstant(loc, idxTy, extent));
     }
   }
-  if (auto charTy = fieldTy.dyn_cast<fir::CharacterType>()) {
+  if (auto charTy = mlir::dyn_cast<fir::CharacterType>(fieldTy)) {
     auto cstLen = charTy.getLen();
     if (cstLen == fir::CharacterType::unknownLen())
       TODO(loc, "get character component length from length type parameters");
@@ -1148,7 +1149,7 @@ fir::ExtendedValue fir::factory::componentToExtendedValue(
       return fir::CharArrayBoxValue{component, len, extents};
     return fir::CharBoxValue{component, len};
   }
-  if (auto recordTy = fieldTy.dyn_cast<fir::RecordType>())
+  if (auto recordTy = mlir::dyn_cast<fir::RecordType>(fieldTy))
     if (recordTy.getNumLenParams() != 0)
       TODO(loc,
            "lower component ref that is a derived type with length parameter");
@@ -1211,14 +1212,14 @@ void fir::factory::genScalarAssignment(fir::FirOpBuilder &builder,
   assert(lhs.rank() == 0 && rhs.rank() == 0 && "must be scalars");
   auto type = fir::unwrapSequenceType(
       fir::unwrapPassByRefType(fir::getBase(lhs).getType()));
-  if (type.isa<fir::CharacterType>()) {
+  if (mlir::isa<fir::CharacterType>(type)) {
     const fir::CharBoxValue *toChar = lhs.getCharBox();
     const fir::CharBoxValue *fromChar = rhs.getCharBox();
     assert(toChar && fromChar);
     fir::factory::CharacterExprHelper helper{builder, loc};
     helper.createAssign(fir::ExtendedValue{*toChar},
                         fir::ExtendedValue{*fromChar});
-  } else if (type.isa<fir::RecordType>()) {
+  } else if (mlir::isa<fir::RecordType>(type)) {
     fir::factory::genRecordAssignment(builder, loc, lhs, rhs, needFinalization,
                                       isTemporaryLHS);
   } else {
@@ -1239,10 +1240,10 @@ static void genComponentByComponentAssignment(fir::FirOpBuilder &builder,
                                               const fir::ExtendedValue &rhs,
                                               bool isTemporaryLHS) {
   auto lbaseType = fir::unwrapPassByRefType(fir::getBase(lhs).getType());
-  auto lhsType = lbaseType.dyn_cast<fir::RecordType>();
+  auto lhsType = mlir::dyn_cast<fir::RecordType>(lbaseType);
   assert(lhsType && "lhs must be a scalar record type");
   auto rbaseType = fir::unwrapPassByRefType(fir::getBase(rhs).getType());
-  auto rhsType = rbaseType.dyn_cast<fir::RecordType>();
+  auto rhsType = mlir::dyn_cast<fir::RecordType>(rbaseType);
   assert(rhsType && "rhs must be a scalar record type");
   auto fieldIndexType = fir::FieldType::get(lhsType.getContext());
   for (auto [lhsPair, rhsPair] :
@@ -1261,7 +1262,7 @@ static void genComponentByComponentAssignment(fir::FirOpBuilder &builder,
     mlir::Value toCoor = builder.create<fir::CoordinateOp>(
         loc, fieldRefType, fir::getBase(lhs), field);
     std::optional<fir::DoLoopOp> outerLoop;
-    if (auto sequenceType = lFieldTy.dyn_cast<fir::SequenceType>()) {
+    if (auto sequenceType = mlir::dyn_cast<fir::SequenceType>(lFieldTy)) {
       // Create loops to assign array components elements by elements.
       // Note that, since these are components, they either do not overlap,
       // or are the same and exactly overlap. They also have compile time
@@ -1288,10 +1289,9 @@ static void genComponentByComponentAssignment(fir::FirOpBuilder &builder,
                                                    fromCoor, indices);
     }
     if (auto fieldEleTy = fir::unwrapSequenceType(lFieldTy);
-        fieldEleTy.isa<fir::BaseBoxType>()) {
-      assert(fieldEleTy.cast<fir::BaseBoxType>()
-                 .getEleTy()
-                 .isa<fir::PointerType>() &&
+        mlir::isa<fir::BaseBoxType>(fieldEleTy)) {
+      assert(mlir::isa<fir::PointerType>(
+                 mlir::cast<fir::BaseBoxType>(fieldEleTy).getEleTy()) &&
              "allocatable members require deep copy");
       auto fromPointerValue = builder.create<fir::LoadOp>(loc, fromCoor);
       auto castTo = builder.createConvert(loc, fieldEleTy, fromPointerValue);
@@ -1320,11 +1320,11 @@ static bool recordTypeCanBeMemCopied(fir::RecordType recordType) {
   for (auto [_, fieldType] : recordType.getTypeList()) {
     // Derived type component may have user assignment (so far, we cannot tell
     // in FIR, so assume it is always the case, TODO: get the actual info).
-    if (fir::unwrapSequenceType(fieldType).isa<fir::RecordType>())
+    if (mlir::isa<fir::RecordType>(fir::unwrapSequenceType(fieldType)))
       return false;
     // Allocatable components need deep copy.
-    if (auto boxType = fieldType.dyn_cast<fir::BaseBoxType>())
-      if (boxType.getEleTy().isa<fir::HeapType>())
+    if (auto boxType = mlir::dyn_cast<fir::BaseBoxType>(fieldType))
+      if (mlir::isa<fir::HeapType>(boxType.getEleTy()))
         return false;
   }
   // Constant size components without user defined assignment and pointers can
@@ -1353,9 +1353,10 @@ void fir::factory::genRecordAssignment(fir::FirOpBuilder &builder,
   // Box operands may be polymorphic, it is not entirely clear from 10.2.1.3
   // if the assignment is performed on the dynamic of declared type. Use the
   // runtime assuming it is performed on the dynamic type.
-  bool hasBoxOperands = fir::getBase(lhs).getType().isa<fir::BaseBoxType>() ||
-                        fir::getBase(rhs).getType().isa<fir::BaseBoxType>();
-  auto recTy = baseTy.dyn_cast<fir::RecordType>();
+  bool hasBoxOperands =
+      mlir::isa<fir::BaseBoxType>(fir::getBase(lhs).getType()) ||
+      mlir::isa<fir::BaseBoxType>(fir::getBase(rhs).getType());
+  auto recTy = mlir::dyn_cast<fir::RecordType>(baseTy);
   assert(recTy && "must be a record type");
   if ((needFinalization && mayHaveFinalizer(recTy, builder)) ||
       hasBoxOperands || !recordTypeCanBeMemCopied(recTy)) {
@@ -1401,7 +1402,7 @@ mlir::Value fir::factory::genLenOfCharacter(
     llvm::ArrayRef<mlir::Value> path, llvm::ArrayRef<mlir::Value> substring) {
   llvm::SmallVector<mlir::Value> typeParams(arrLoad.getTypeparams());
   return genLenOfCharacter(builder, loc,
-                           arrLoad.getType().cast<fir::SequenceType>(),
+                           mlir::cast<fir::SequenceType>(arrLoad.getType()),
                            arrLoad.getMemref(), typeParams, path, substring);
 }
 
@@ -1429,7 +1430,7 @@ mlir::Value fir::factory::genLenOfCharacter(
     lower = builder.createConvert(loc, idxTy, substring.front());
   auto eleTy = fir::applyPathToType(seqTy, path);
   if (!fir::hasDynamicSize(eleTy)) {
-    if (auto charTy = eleTy.dyn_cast<fir::CharacterType>()) {
+    if (auto charTy = mlir::dyn_cast<fir::CharacterType>(eleTy)) {
       // Use LEN from the type.
       return builder.createIntegerConstant(loc, idxTy, charTy.getLen());
     }
@@ -1438,9 +1439,9 @@ mlir::Value fir::factory::genLenOfCharacter(
                         "application of path did not result in a !fir.char");
   }
   if (fir::isa_box_type(memref.getType())) {
-    if (memref.getType().isa<fir::BoxCharType>())
+    if (mlir::isa<fir::BoxCharType>(memref.getType()))
       return builder.create<fir::BoxCharLenOp>(loc, idxTy, memref);
-    if (memref.getType().isa<fir::BoxType>())
+    if (mlir::isa<fir::BoxType>(memref.getType()))
       return CharacterExprHelper(builder, loc).readLengthFromBox(memref);
     fir::emitFatalError(loc, "memref has wrong type");
   }
@@ -1457,7 +1458,7 @@ mlir::Value fir::factory::genLenOfCharacter(
 mlir::Value fir::factory::createZeroValue(fir::FirOpBuilder &builder,
                                           mlir::Location loc, mlir::Type type) {
   mlir::Type i1 = builder.getIntegerType(1);
-  if (type.isa<fir::LogicalType>() || type == i1)
+  if (mlir::isa<fir::LogicalType>(type) || type == i1)
     return builder.createConvert(loc, type, builder.createBool(loc, false));
   if (fir::isa_integer(type))
     return builder.createIntegerConstant(loc, type, 0);
@@ -1507,7 +1508,7 @@ mlir::Value fir::factory::genMaxWithZero(fir::FirOpBuilder &builder,
   mlir::Value zero = builder.createIntegerConstant(loc, value.getType(), 0);
   if (mlir::Operation *definingOp = value.getDefiningOp())
     if (auto cst = mlir::dyn_cast<mlir::arith::ConstantOp>(definingOp))
-      if (auto intAttr = cst.getValue().dyn_cast<mlir::IntegerAttr>())
+      if (auto intAttr = mlir::dyn_cast<mlir::IntegerAttr>(cst.getValue()))
         return intAttr.getInt() > 0 ? value : zero;
   mlir::Value valueIsGreater = builder.create<mlir::arith::CmpIOp>(
       loc, mlir::arith::CmpIPredicate::sgt, value, zero);
@@ -1519,8 +1520,8 @@ mlir::Value fir::factory::genCPtrOrCFunptrAddr(fir::FirOpBuilder &builder,
                                                mlir::Location loc,
                                                mlir::Value cPtr,
                                                mlir::Type ty) {
-  assert(ty.isa<fir::RecordType>());
-  auto recTy = ty.dyn_cast<fir::RecordType>();
+  assert(mlir::isa<fir::RecordType>(ty));
+  auto recTy = mlir::dyn_cast<fir::RecordType>(ty);
   assert(recTy.getTypeList().size() == 1);
   auto fieldName = recTy.getTypeList()[0].first;
   mlir::Type fieldTy = recTy.getTypeList()[0].second;
@@ -1582,7 +1583,7 @@ mlir::Value fir::factory::genCPtrOrCFunptrValue(fir::FirOpBuilder &builder,
 mlir::Value fir::factory::createNullBoxProc(fir::FirOpBuilder &builder,
                                             mlir::Location loc,
                                             mlir::Type boxType) {
-  auto boxTy{boxType.dyn_cast<fir::BoxProcType>()};
+  auto boxTy{mlir::dyn_cast<fir::BoxProcType>(boxType)};
   if (!boxTy)
     fir::emitFatalError(loc, "Procedure pointer must be of BoxProcType");
   auto boxEleTy{fir::unwrapRefType(boxTy.getEleTy())};
diff --git a/flang/lib/Optimizer/Builder/HLFIRTools.cpp b/flang/lib/Optimizer/Builder/HLFIRTools.cpp
index db638ceb4070..44779427ab55 100644
--- a/flang/lib/Optimizer/Builder/HLFIRTools.cpp
+++ b/flang/lib/Optimizer/Builder/HLFIRTools.cpp
@@ -38,10 +38,10 @@ hlfir::getExplicitExtentsFromShape(mlir::Value shape,
   } else if (mlir::dyn_cast_or_null<fir::ShiftOp>(shapeOp)) {
     return {};
   } else if (auto s = mlir::dyn_cast_or_null<hlfir::ShapeOfOp>(shapeOp)) {
-    hlfir::ExprType expr = s.getExpr().getType().cast<hlfir::ExprType>();
+    hlfir::ExprType expr = mlir::cast<hlfir::ExprType>(s.getExpr().getType());
     llvm::ArrayRef<int64_t> exprShape = expr.getShape();
     mlir::Type indexTy = builder.getIndexType();
-    fir::ShapeType shapeTy = shape.getType().cast<fir::ShapeType>();
+    fir::ShapeType shapeTy = mlir::cast<fir::ShapeType>(shape.getType());
     result.reserve(shapeTy.getRank());
     for (unsigned i = 0; i < shapeTy.getRank(); ++i) {
       int64_t extent = exprShape[i];
@@ -99,7 +99,7 @@ genLboundsAndExtentsFromBox(mlir::Location loc, fir::FirOpBuilder &builder,
                             hlfir::Entity boxEntity,
                             llvm::SmallVectorImpl<mlir::Value> &lbounds,
                             llvm::SmallVectorImpl<mlir::Value> *extents) {
-  assert(boxEntity.getType().isa<fir::BaseBoxType>() && "must be a box");
+  assert(mlir::isa<fir::BaseBoxType>(boxEntity.getType()) && "must be a box");
   mlir::Type idxTy = builder.getIndexType();
   const int rank = boxEntity.getRank();
   for (int i = 0; i < rank; ++i) {
@@ -154,7 +154,7 @@ static mlir::Value genCharacterVariableLength(mlir::Location loc,
                                               hlfir::Entity var) {
   if (mlir::Value len = tryGettingNonDeferredCharLen(var))
     return len;
-  auto charType = var.getFortranElementType().cast<fir::CharacterType>();
+  auto charType = mlir::cast<fir::CharacterType>(var.getFortranElementType());
   if (charType.hasConstantLen())
     return builder.createIntegerConstant(loc, builder.getIndexType(),
                                          charType.getLen());
@@ -172,7 +172,7 @@ static fir::CharBoxValue genUnboxChar(mlir::Location loc,
   if (auto emboxChar = boxChar.getDefiningOp<fir::EmboxCharOp>())
     return {emboxChar.getMemref(), emboxChar.getLen()};
   mlir::Type refType = fir::ReferenceType::get(
-      boxChar.getType().cast<fir::BoxCharType>().getEleTy());
+      mlir::cast<fir::BoxCharType>(boxChar.getType()).getEleTy());
   auto unboxed = builder.create<fir::UnboxCharOp>(
       loc, refType, builder.getIndexType(), boxChar);
   mlir::Value addr = unboxed.getResult(0);
@@ -252,8 +252,8 @@ hlfir::genAssociateExpr(mlir::Location loc, fir::FirOpBuilder &builder,
   // and the other static).
   mlir::Type varEleTy = getFortranElementType(variableType);
   mlir::Type valueEleTy = getFortranElementType(value.getType());
-  if (varEleTy != valueEleTy && !(valueEleTy.isa<fir::CharacterType>() &&
-                                  varEleTy.isa<fir::CharacterType>())) {
+  if (varEleTy != valueEleTy && !(mlir::isa<fir::CharacterType>(valueEleTy) &&
+                                  mlir::isa<fir::CharacterType>(varEleTy))) {
     assert(value.isScalar() && fir::isa_trivial(value.getType()));
     source = builder.createConvert(loc, fir::unwrapPassByRefType(variableType),
                                    value);
@@ -278,9 +278,9 @@ mlir::Value hlfir::genVariableRawAddress(mlir::Location loc,
   if (var.isMutableBox())
     baseAddr = builder.create<fir::LoadOp>(loc, baseAddr);
   // Get raw address.
-  if (var.getType().isa<fir::BoxCharType>())
+  if (mlir::isa<fir::BoxCharType>(var.getType()))
     baseAddr = genUnboxChar(loc, builder, var.getBase()).getAddr();
-  if (baseAddr.getType().isa<fir::BaseBoxType>())
+  if (mlir::isa<fir::BaseBoxType>(baseAddr.getType()))
     baseAddr = builder.create<fir::BoxAddrOp>(loc, baseAddr);
   return baseAddr;
 }
@@ -289,13 +289,13 @@ mlir::Value hlfir::genVariableBoxChar(mlir::Location loc,
                                       fir::FirOpBuilder &builder,
                                       hlfir::Entity var) {
   assert(var.isVariable() && "only address of variables can be taken");
-  if (var.getType().isa<fir::BoxCharType>())
+  if (mlir::isa<fir::BoxCharType>(var.getType()))
     return var;
   mlir::Value addr = genVariableRawAddress(loc, builder, var);
   llvm::SmallVector<mlir::Value> lengths;
   genLengthParameters(loc, builder, var, lengths);
   assert(lengths.size() == 1);
-  auto charType = var.getFortranElementType().cast<fir::CharacterType>();
+  auto charType = mlir::cast<fir::CharacterType>(var.getFortranElementType());
   auto boxCharType =
       fir::BoxCharType::get(builder.getContext(), charType.getFKind());
   auto scalarAddr =
@@ -309,7 +309,7 @@ hlfir::Entity hlfir::genVariableBox(mlir::Location loc,
                                     hlfir::Entity var) {
   assert(var.isVariable() && "must be a variable");
   var = hlfir::derefPointersAndAllocatables(loc, builder, var);
-  if (var.getType().isa<fir::BaseBoxType>())
+  if (mlir::isa<fir::BaseBoxType>(var.getType()))
     return var;
   // Note: if the var is not a fir.box/fir.class at that point, it has default
   // lower bounds and is not polymorphic.
@@ -317,11 +317,11 @@ hlfir::Entity hlfir::genVariableBox(mlir::Location loc,
       var.isArray() ? hlfir::genShape(loc, builder, var) : mlir::Value{};
   llvm::SmallVector<mlir::Value> typeParams;
   auto maybeCharType =
-      var.getFortranElementType().dyn_cast<fir::CharacterType>();
+      mlir::dyn_cast<fir::CharacterType>(var.getFortranElementType());
   if (!maybeCharType || maybeCharType.hasDynamicLen())
     hlfir::genLengthParameters(loc, builder, var, typeParams);
   mlir::Value addr = var.getBase();
-  if (var.getType().isa<fir::BoxCharType>())
+  if (mlir::isa<fir::BoxCharType>(var.getType()))
     addr = genVariableRawAddress(loc, builder, var);
   mlir::Type boxType = fir::BoxType::get(var.getElementOrSequenceType());
   auto embox =
@@ -348,7 +348,7 @@ hlfir::Entity hlfir::getElementAt(mlir::Location loc,
     return entity;
   llvm::SmallVector<mlir::Value> lenParams;
   genLengthParameters(loc, builder, entity, lenParams);
-  if (entity.getType().isa<hlfir::ExprType>())
+  if (mlir::isa<hlfir::ExprType>(entity.getType()))
     return hlfir::Entity{builder.create<hlfir::ApplyOp>(
         loc, entity, oneBasedIndices, lenParams)};
   // Build hlfir.designate. The lower bounds may need to be added to
@@ -394,7 +394,7 @@ static mlir::Value genUBound(mlir::Location loc, fir::FirOpBuilder &builder,
 llvm::SmallVector<std::pair<mlir::Value, mlir::Value>>
 hlfir::genBounds(mlir::Location loc, fir::FirOpBuilder &builder,
                  Entity entity) {
-  if (entity.getType().isa<hlfir::ExprType>())
+  if (mlir::isa<hlfir::ExprType>(entity.getType()))
     TODO(loc, "bounds of expressions in hlfir");
   auto [exv, cleanup] = translateToExtendedValue(loc, builder, entity);
   assert(!cleanup && "translation of entity should not yield cleanup");
@@ -415,8 +415,8 @@ hlfir::genBounds(mlir::Location loc, fir::FirOpBuilder &builder,
 llvm::SmallVector<std::pair<mlir::Value, mlir::Value>>
 hlfir::genBounds(mlir::Location loc, fir::FirOpBuilder &builder,
                  mlir::Value shape) {
-  assert((shape.getType().isa<fir::ShapeShiftType>() ||
-          shape.getType().isa<fir::ShapeType>()) &&
+  assert((mlir::isa<fir::ShapeShiftType>(shape.getType()) ||
+          mlir::isa<fir::ShapeType>(shape.getType())) &&
          "shape must contain extents");
   auto extents = hlfir::getExplicitExtentsFromShape(shape, builder);
   auto lowers = getExplicitLboundsFromShape(shape);
@@ -474,7 +474,7 @@ static mlir::Value computeVariableExtent(mlir::Location loc,
     if (typeExtent != fir::SequenceType::getUnknownExtent())
       return builder.createIntegerConstant(loc, idxTy, typeExtent);
   }
-  assert(variable.getType().isa<fir::BaseBoxType>() &&
+  assert(mlir::isa<fir::BaseBoxType>(variable.getType()) &&
          "array variable with dynamic extent must be boxed");
   mlir::Value dimVal = builder.createIntegerConstant(loc, idxTy, dim);
   auto dimInfo = builder.create<fir::BoxDimsOp>(loc, idxTy, idxTy, idxTy,
@@ -496,9 +496,8 @@ llvm::SmallVector<mlir::Value> getVariableExtents(mlir::Location loc,
     variable = hlfir::derefPointersAndAllocatables(loc, builder, variable);
   // Use the type shape information, and/or the fir.box/fir.class shape
   // information if any extents are not static.
-  fir::SequenceType seqTy =
-      hlfir::getFortranElementOrSequenceType(variable.getType())
-          .cast<fir::SequenceType>();
+  fir::SequenceType seqTy = mlir::cast<fir::SequenceType>(
+      hlfir::getFortranElementOrSequenceType(variable.getType()));
   unsigned rank = seqTy.getShape().size();
   for (unsigned dim = 0; dim < rank; ++dim)
     extents.push_back(
@@ -507,7 +506,7 @@ llvm::SmallVector<mlir::Value> getVariableExtents(mlir::Location loc,
 }
 
 static mlir::Value tryRetrievingShapeOrShift(hlfir::Entity entity) {
-  if (entity.getType().isa<hlfir::ExprType>()) {
+  if (mlir::isa<hlfir::ExprType>(entity.getType())) {
     if (auto elemental = entity.getDefiningOp<hlfir::ElementalOp>())
       return elemental.getShape();
     return mlir::Value{};
@@ -523,13 +522,13 @@ mlir::Value hlfir::genShape(mlir::Location loc, fir::FirOpBuilder &builder,
   entity = followShapeInducingSource(entity);
   assert(entity && "what?");
   if (auto shape = tryRetrievingShapeOrShift(entity)) {
-    if (shape.getType().isa<fir::ShapeType>())
+    if (mlir::isa<fir::ShapeType>(shape.getType()))
       return shape;
-    if (shape.getType().isa<fir::ShapeShiftType>())
+    if (mlir::isa<fir::ShapeShiftType>(shape.getType()))
       if (auto s = shape.getDefiningOp<fir::ShapeShiftOp>())
         return builder.create<fir::ShapeOp>(loc, s.getExtents());
   }
-  if (entity.getType().isa<hlfir::ExprType>())
+  if (mlir::isa<hlfir::ExprType>(entity.getType()))
     return builder.create<hlfir::ShapeOfOp>(loc, entity.getBase());
   // There is no shape lying around for this entity. Retrieve the extents and
   // build a new fir.shape.
@@ -563,9 +562,8 @@ mlir::Value hlfir::genExtent(mlir::Location loc, fir::FirOpBuilder &builder,
       entity = hlfir::derefPointersAndAllocatables(loc, builder, entity);
     // Use the type shape information, and/or the fir.box/fir.class shape
     // information if any extents are not static.
-    fir::SequenceType seqTy =
-        hlfir::getFortranElementOrSequenceType(entity.getType())
-            .cast<fir::SequenceType>();
+    fir::SequenceType seqTy = mlir::cast<fir::SequenceType>(
+        hlfir::getFortranElementOrSequenceType(entity.getType()));
     return computeVariableExtent(loc, builder, entity, seqTy, dim);
   }
   TODO(loc, "get extent from HLFIR expr without producer holding the shape");
@@ -584,7 +582,7 @@ mlir::Value hlfir::genLBound(mlir::Location loc, fir::FirOpBuilder &builder,
   }
   if (entity.isMutableBox())
     entity = hlfir::derefPointersAndAllocatables(loc, builder, entity);
-  assert(entity.getType().isa<fir::BaseBoxType>() && "must be a box");
+  assert(mlir::isa<fir::BaseBoxType>(entity.getType()) && "must be a box");
   mlir::Type idxTy = builder.getIndexType();
   mlir::Value dimVal = builder.createIntegerConstant(loc, idxTy, dim);
   auto dimInfo =
@@ -597,7 +595,7 @@ void hlfir::genLengthParameters(mlir::Location loc, fir::FirOpBuilder &builder,
                                 llvm::SmallVectorImpl<mlir::Value> &result) {
   if (!entity.hasLengthParameters())
     return;
-  if (entity.getType().isa<hlfir::ExprType>()) {
+  if (mlir::isa<hlfir::ExprType>(entity.getType())) {
     mlir::Value expr = entity;
     if (auto reassoc = expr.getDefiningOp<hlfir::NoReassocOp>())
       expr = reassoc.getVal();
@@ -654,8 +652,8 @@ static mlir::Value asEmboxShape(mlir::Location loc, fir::FirOpBuilder &builder,
   // fir.shape_shift) since this information is already in the input fir.box,
   // it only accepts fir.shift because local lower bounds may not be reflected
   // in the fir.box.
-  if (fir::getBase(exv).getType().isa<fir::BaseBoxType>() &&
-      !shape.getType().isa<fir::ShiftType>())
+  if (mlir::isa<fir::BaseBoxType>(fir::getBase(exv).getType()) &&
+      !mlir::isa<fir::ShiftType>(shape.getType()))
     return builder.createShape(loc, exv);
   return shape;
 }
@@ -686,7 +684,7 @@ hlfir::Entity hlfir::derefPointersAndAllocatables(mlir::Location loc,
       if (!entity.isPolymorphic() && !entity.hasLengthParameters())
         return hlfir::Entity{builder.create<fir::BoxAddrOp>(loc, boxLoad)};
       mlir::Type elementType = boxLoad.getFortranElementType();
-      if (auto charType = elementType.dyn_cast<fir::CharacterType>()) {
+      if (auto charType = mlir::dyn_cast<fir::CharacterType>(elementType)) {
         mlir::Value base = builder.create<fir::BoxAddrOp>(loc, boxLoad);
         if (charType.hasConstantLen())
           return hlfir::Entity{base};
@@ -716,7 +714,7 @@ mlir::Type hlfir::getVariableElementType(hlfir::Entity variable) {
   mlir::Type eleTy = variable.getFortranElementType();
   if (variable.isPolymorphic())
     return fir::ClassType::get(eleTy);
-  if (auto charType = eleTy.dyn_cast<fir::CharacterType>()) {
+  if (auto charType = mlir::dyn_cast<fir::CharacterType>(eleTy)) {
     if (charType.hasDynamicLen())
       return fir::BoxCharType::get(charType.getContext(), charType.getFKind());
   } else if (fir::isRecordWithTypeParameters(eleTy)) {
@@ -737,7 +735,7 @@ mlir::Type hlfir::getEntityElementType(hlfir::Entity entity) {
 
 static hlfir::ExprType getArrayExprType(mlir::Type elementType,
                                         mlir::Value shape, bool isPolymorphic) {
-  unsigned rank = shape.getType().cast<fir::ShapeType>().getRank();
+  unsigned rank = mlir::cast<fir::ShapeType>(shape.getType()).getRank();
   hlfir::ExprType::Shape typeShape(rank, hlfir::ExprType::getUnknownExtent());
   if (auto shapeOp = shape.getDefiningOp<fir::ShapeOp>())
     for (auto extent : llvm::enumerate(shapeOp.getExtents()))
@@ -859,7 +857,7 @@ translateVariableToExtendedValue(mlir::Location loc, fir::FirOpBuilder &builder,
     return fir::MutableBoxValue(base, getExplicitTypeParams(variable),
                                 fir::MutableProperties{});
 
-  if (base.getType().isa<fir::BaseBoxType>()) {
+  if (mlir::isa<fir::BaseBoxType>(base.getType())) {
     if (!variable.isSimplyContiguous() || variable.isPolymorphic() ||
         variable.isDerivedWithLengthParameters() || variable.isOptional()) {
       llvm::SmallVector<mlir::Value> nonDefaultLbounds =
@@ -874,7 +872,7 @@ translateVariableToExtendedValue(mlir::Location loc, fir::FirOpBuilder &builder,
 
   if (variable.isScalar()) {
     if (variable.isCharacter()) {
-      if (base.getType().isa<fir::BoxCharType>())
+      if (mlir::isa<fir::BoxCharType>(base.getType()))
         return genUnboxChar(loc, builder, base);
       mlir::Value len = genCharacterVariableLength(loc, builder, variable);
       return fir::CharBoxValue{base, len};
@@ -883,7 +881,7 @@ translateVariableToExtendedValue(mlir::Location loc, fir::FirOpBuilder &builder,
   }
   llvm::SmallVector<mlir::Value> extents;
   llvm::SmallVector<mlir::Value> nonDefaultLbounds;
-  if (variable.getType().isa<fir::BaseBoxType>() &&
+  if (mlir::isa<fir::BaseBoxType>(variable.getType()) &&
       !variable.getIfVariableInterface()) {
     // This special case avoids generating two sets of identical
     // fir.box_dim to get both the lower bounds and extents.
@@ -923,7 +921,7 @@ hlfir::translateToExtendedValue(mlir::Location loc, fir::FirOpBuilder &builder,
     return {static_cast<mlir::Value>(entity), std::nullopt};
   }
 
-  if (entity.getType().isa<hlfir::ExprType>()) {
+  if (mlir::isa<hlfir::ExprType>(entity.getType())) {
     mlir::NamedAttribute byRefAttr = fir::getAdaptToByRefAttr(builder);
     hlfir::AssociateOp associate = hlfir::genAssociateExpr(
         loc, builder, entity, entity.getType(), "", byRefAttr);
diff --git a/flang/lib/Optimizer/Builder/IntrinsicCall.cpp b/flang/lib/Optimizer/Builder/IntrinsicCall.cpp
index 4ee7258004fa..9d72e76e2369 100644
--- a/flang/lib/Optimizer/Builder/IntrinsicCall.cpp
+++ b/flang/lib/Optimizer/Builder/IntrinsicCall.cpp
@@ -554,6 +554,10 @@ static constexpr IntrinsicHandler handlers[]{
        {"radix", asAddr, handleDynamicOptional}}},
      /*isElemental=*/false},
     {"set_exponent", &I::genSetExponent},
+    {"shape",
+     &I::genShape,
+     {{{"source", asBox}, {"kind", asValue}}},
+     /*isElemental=*/false},
     {"shifta", &I::genShiftA},
     {"shiftl", &I::genShift<mlir::arith::ShLIOp>},
     {"shiftr", &I::genShift<mlir::arith::ShRUIOp>},
@@ -782,7 +786,7 @@ mlir::Value genLibSplitComplexArgsCall(fir::FirOpBuilder &builder,
 
   auto getSplitComplexArgsType = [&builder, &args]() -> mlir::FunctionType {
     mlir::Type ctype = args[0].getType();
-    auto fKind = ctype.cast<fir::ComplexType>().getFKind();
+    auto fKind = mlir::cast<fir::ComplexType>(ctype).getFKind();
     mlir::Type ftype;
 
     if (fKind == 2)
@@ -890,8 +894,8 @@ mlir::Value genComplexMathOp(fir::FirOpBuilder &builder, mlir::Location loc,
   LLVM_DEBUG(llvm::dbgs() << "Generating '" << mathLibFuncName
                           << "' operation with type ";
              mathLibFuncType.dump(); llvm::dbgs() << "\n");
-  auto type = mathLibFuncType.getInput(0).cast<fir::ComplexType>();
-  auto kind = type.getElementType().cast<fir::RealType>().getFKind();
+  auto type = mlir::cast<fir::ComplexType>(mathLibFuncType.getInput(0));
+  auto kind = mlir::cast<fir::RealType>(type.getElementType()).getFKind();
   auto realTy = builder.getRealType(kind);
   auto mComplexTy = mlir::ComplexType::get(realTy);
 
@@ -1390,14 +1394,14 @@ private:
 
   // Floating point can be mlir::FloatType or fir::real
   static unsigned getFloatingPointWidth(mlir::Type t) {
-    if (auto f{t.dyn_cast<mlir::FloatType>()})
+    if (auto f{mlir::dyn_cast<mlir::FloatType>(t)})
       return f.getWidth();
     // FIXME: Get width another way for fir.real/complex
     // - use fir/KindMapping.h and llvm::Type
     // - or use evaluate/type.h
-    if (auto r{t.dyn_cast<fir::RealType>()})
+    if (auto r{mlir::dyn_cast<fir::RealType>(t)})
       return r.getFKind() * 4;
-    if (auto cplx{t.dyn_cast<fir::ComplexType>()})
+    if (auto cplx{mlir::dyn_cast<fir::ComplexType>(t)})
       return cplx.getFKind() * 4;
     llvm_unreachable("not a floating-point type");
   }
@@ -1406,8 +1410,8 @@ private:
     if (from == to)
       return Conversion::None;
 
-    if (auto fromIntTy{from.dyn_cast<mlir::IntegerType>()}) {
-      if (auto toIntTy{to.dyn_cast<mlir::IntegerType>()}) {
+    if (auto fromIntTy{mlir::dyn_cast<mlir::IntegerType>(from)}) {
+      if (auto toIntTy{mlir::dyn_cast<mlir::IntegerType>(to)}) {
         return fromIntTy.getWidth() > toIntTy.getWidth() ? Conversion::Narrow
                                                          : Conversion::Extend;
       }
@@ -1419,8 +1423,8 @@ private:
                  : Conversion::Extend;
     }
 
-    if (auto fromCplxTy{from.dyn_cast<fir::ComplexType>()}) {
-      if (auto toCplxTy{to.dyn_cast<fir::ComplexType>()}) {
+    if (auto fromCplxTy{mlir::dyn_cast<fir::ComplexType>(from)}) {
+      if (auto toCplxTy{mlir::dyn_cast<fir::ComplexType>(to)}) {
         return getFloatingPointWidth(fromCplxTy) >
                        getFloatingPointWidth(toCplxTy)
                    ? Conversion::Narrow
@@ -1546,10 +1550,10 @@ fir::ExtendedValue toExtendedValue(mlir::Value val, fir::FirOpBuilder &builder,
   if (charHelper.isCharacterScalar(type))
     return charHelper.toExtendedValue(val);
 
-  if (auto refType = type.dyn_cast<fir::ReferenceType>())
+  if (auto refType = mlir::dyn_cast<fir::ReferenceType>(type))
     type = refType.getEleTy();
 
-  if (auto arrayType = type.dyn_cast<fir::SequenceType>()) {
+  if (auto arrayType = mlir::dyn_cast<fir::SequenceType>(type)) {
     type = arrayType.getEleTy();
     for (fir::SequenceType::Extent extent : arrayType.getShape()) {
       if (extent == fir::SequenceType::getUnknownExtent())
@@ -1562,7 +1566,8 @@ fir::ExtendedValue toExtendedValue(mlir::Value val, fir::FirOpBuilder &builder,
     // have been used in the interface).
     if (extents.size() + 1 < arrayType.getShape().size())
       mlir::emitError(loc, "cannot retrieve array extents from type");
-  } else if (type.isa<fir::BoxType>() || type.isa<fir::RecordType>()) {
+  } else if (mlir::isa<fir::BoxType>(type) ||
+             mlir::isa<fir::RecordType>(type)) {
     fir::emitFatalError(loc, "not yet implemented: descriptor or derived type");
   }
 
@@ -1576,10 +1581,10 @@ mlir::Value toValue(const fir::ExtendedValue &val, fir::FirOpBuilder &builder,
   if (const fir::CharBoxValue *charBox = val.getCharBox()) {
     mlir::Value buffer = charBox->getBuffer();
     auto buffTy = buffer.getType();
-    if (buffTy.isa<mlir::FunctionType>())
+    if (mlir::isa<mlir::FunctionType>(buffTy))
       fir::emitFatalError(
           loc, "A character's buffer type cannot be a function type.");
-    if (buffTy.isa<fir::BoxCharType>())
+    if (mlir::isa<fir::BoxCharType>(buffTy))
       return buffer;
     return fir::factory::CharacterExprHelper{builder, loc}.createEmboxChar(
         buffer, charBox->getLen());
@@ -1823,27 +1828,27 @@ IntrinsicLibrary::invokeGenerator(SubroutineGenerator generator,
 /// Note: mlir has Type::dump(ostream) methods but it may add "!" that is not
 /// suitable for function names.
 static std::string typeToString(mlir::Type t) {
-  if (auto refT{t.dyn_cast<fir::ReferenceType>()})
+  if (auto refT{mlir::dyn_cast<fir::ReferenceType>(t)})
     return "ref_" + typeToString(refT.getEleTy());
-  if (auto i{t.dyn_cast<mlir::IntegerType>()}) {
+  if (auto i{mlir::dyn_cast<mlir::IntegerType>(t)}) {
     return "i" + std::to_string(i.getWidth());
   }
-  if (auto cplx{t.dyn_cast<fir::ComplexType>()}) {
+  if (auto cplx{mlir::dyn_cast<fir::ComplexType>(t)}) {
     return "z" + std::to_string(cplx.getFKind());
   }
-  if (auto real{t.dyn_cast<fir::RealType>()}) {
+  if (auto real{mlir::dyn_cast<fir::RealType>(t)}) {
     return "r" + std::to_string(real.getFKind());
   }
-  if (auto f{t.dyn_cast<mlir::FloatType>()}) {
+  if (auto f{mlir::dyn_cast<mlir::FloatType>(t)}) {
     return "f" + std::to_string(f.getWidth());
   }
-  if (auto logical{t.dyn_cast<fir::LogicalType>()}) {
+  if (auto logical{mlir::dyn_cast<fir::LogicalType>(t)}) {
     return "l" + std::to_string(logical.getFKind());
   }
-  if (auto character{t.dyn_cast<fir::CharacterType>()}) {
+  if (auto character{mlir::dyn_cast<fir::CharacterType>(t)}) {
     return "c" + std::to_string(character.getFKind());
   }
-  if (auto boxCharacter{t.dyn_cast<fir::BoxCharType>()}) {
+  if (auto boxCharacter{mlir::dyn_cast<fir::BoxCharType>(t)}) {
     return "bc" + std::to_string(boxCharacter.getEleTy().getFKind());
   }
   llvm_unreachable("no mangling for type");
@@ -1903,7 +1908,7 @@ mlir::func::FuncOp IntrinsicLibrary::getWrapper(GeneratorType generator,
     mlir::Location localLoc = localBuilder->getUnknownLoc();
     llvm::SmallVector<mlir::Value> localArguments;
     for (mlir::BlockArgument bArg : function.front().getArguments()) {
-      auto refType = bArg.getType().dyn_cast<fir::ReferenceType>();
+      auto refType = mlir::dyn_cast<fir::ReferenceType>(bArg.getType());
       if (loadRefArguments && refType) {
         auto loaded = localBuilder->create<fir::LoadOp>(localLoc, bArg);
         localArguments.push_back(loaded);
@@ -2056,7 +2061,7 @@ mlir::SymbolRefAttr IntrinsicLibrary::getUnrestrictedIntrinsicSymbolRefAttr(
   if (!funcOp) {
     llvm::SmallVector<mlir::Type> argTypes;
     for (mlir::Type type : signature.getInputs()) {
-      if (auto refType = type.dyn_cast<fir::ReferenceType>())
+      if (auto refType = mlir::dyn_cast<fir::ReferenceType>(type))
         argTypes.push_back(refType.getEleTy());
       else
         argTypes.push_back(type);
@@ -2141,7 +2146,7 @@ mlir::Value IntrinsicLibrary::genAbs(mlir::Type resultType,
     // math::AbsFOp but it does not support all fir floating point types.
     return genRuntimeCall("abs", resultType, args);
   }
-  if (auto intType = type.dyn_cast<mlir::IntegerType>()) {
+  if (auto intType = mlir::dyn_cast<mlir::IntegerType>(type)) {
     // At the time of this implementation there is no abs op in mlir.
     // So, implement abs here without branching.
     mlir::Value shift =
@@ -2375,8 +2380,8 @@ IntrinsicLibrary::genAssociated(mlir::Type resultType,
                                 llvm::ArrayRef<fir::ExtendedValue> args) {
   assert(args.size() == 2);
   mlir::Type ptrTy = fir::getBase(args[0]).getType();
-  if (ptrTy &&
-      (fir::isBoxProcAddressType(ptrTy) || ptrTy.isa<fir::BoxProcType>())) {
+  if (ptrTy && (fir::isBoxProcAddressType(ptrTy) ||
+                mlir::isa<fir::BoxProcType>(ptrTy))) {
     mlir::Value pointerBoxProc =
         fir::isBoxProcAddressType(ptrTy)
             ? builder.create<fir::LoadOp>(loc, fir::getBase(args[0]))
@@ -2388,7 +2393,7 @@ IntrinsicLibrary::genAssociated(mlir::Type resultType,
     mlir::Value target = fir::getBase(args[1]);
     if (fir::isBoxProcAddressType(target.getType()))
       target = builder.create<fir::LoadOp>(loc, target);
-    if (target.getType().isa<fir::BoxProcType>())
+    if (mlir::isa<fir::BoxProcType>(target.getType()))
       target = builder.create<fir::BoxAddrOp>(loc, target);
     mlir::Type intPtrTy = builder.getIntPtrType();
     mlir::Value pointerInt =
@@ -2645,7 +2650,7 @@ static mlir::Value getAddrFromBox(fir::FirOpBuilder &builder,
   mlir::Value argValue = fir::getBase(arg);
   mlir::Value addr{nullptr};
   if (isFunc) {
-    auto funcTy = argValue.getType().cast<fir::BoxProcType>().getEleTy();
+    auto funcTy = mlir::cast<fir::BoxProcType>(argValue.getType()).getEleTy();
     addr = builder.create<fir::BoxAddrOp>(loc, funcTy, argValue);
   } else {
     const auto *box = arg.getBoxOf<fir::BoxValue>();
@@ -3025,7 +3030,7 @@ void IntrinsicLibrary::genDateAndTime(llvm::ArrayRef<fir::ExtendedValue> args) {
 mlir::Value IntrinsicLibrary::genDim(mlir::Type resultType,
                                      llvm::ArrayRef<mlir::Value> args) {
   assert(args.size() == 2);
-  if (resultType.isa<mlir::IntegerType>()) {
+  if (mlir::isa<mlir::IntegerType>(resultType)) {
     mlir::Value zero = builder.createIntegerConstant(loc, resultType, 0);
     auto diff = builder.create<mlir::arith::SubIOp>(loc, args[0], args[1]);
     auto cmp = builder.create<mlir::arith::CmpIOp>(
@@ -3570,7 +3575,7 @@ IntrinsicLibrary::genReduction(FN func, FD funcDim, llvm::StringRef errMsg,
   if (absentDim || rank == 1) {
     mlir::Type ty = array.getType();
     mlir::Type arrTy = fir::dyn_cast_ptrOrBoxEleTy(ty);
-    auto eleTy = arrTy.cast<fir::SequenceType>().getEleTy();
+    auto eleTy = mlir::cast<fir::SequenceType>(arrTy).getEleTy();
     if (fir::isa_complex(eleTy)) {
       mlir::Value result = builder.createTemporary(loc, eleTy);
       func(builder, loc, array, mask, result);
@@ -3642,7 +3647,7 @@ mlir::Value IntrinsicLibrary::genIbits(mlir::Type resultType,
   mlir::Value pos = builder.createConvert(loc, resultType, args[1]);
   mlir::Value len = builder.createConvert(loc, resultType, args[2]);
   mlir::Value bitSize = builder.createIntegerConstant(
-      loc, resultType, resultType.cast<mlir::IntegerType>().getWidth());
+      loc, resultType, mlir::cast<mlir::IntegerType>(resultType).getWidth());
   auto shiftCount = builder.create<mlir::arith::SubIOp>(loc, bitSize, len);
   mlir::Value zero = builder.createIntegerConstant(loc, resultType, 0);
   mlir::Value ones = builder.createAllOnesInteger(loc, resultType);
@@ -3682,7 +3687,7 @@ IntrinsicLibrary::genIchar(mlir::Type resultType,
   mlir::Value buffer = charBox->getBuffer();
   mlir::Type bufferTy = buffer.getType();
   mlir::Value charVal;
-  if (auto charTy = bufferTy.dyn_cast<fir::CharacterType>()) {
+  if (auto charTy = mlir::dyn_cast<fir::CharacterType>(bufferTy)) {
     assert(charTy.singleton());
     charVal = buffer;
   } else {
@@ -3755,7 +3760,7 @@ void IntrinsicLibrary::genRaiseExcept(int except, mlir::Value cond) {
 static std::pair<mlir::Value, mlir::Type>
 getFieldRef(fir::FirOpBuilder &builder, mlir::Location loc, mlir::Value rec) {
   auto recType =
-      fir::unwrapPassByRefType(rec.getType()).dyn_cast<fir::RecordType>();
+      mlir::dyn_cast<fir::RecordType>(fir::unwrapPassByRefType(rec.getType()));
   assert(recType.getTypeList().size() == 1 && "expected exactly one component");
   auto [fieldName, fieldTy] = recType.getTypeList().front();
   mlir::Value field = builder.create<fir::FieldIndexOp>(
@@ -3804,7 +3809,7 @@ mlir::Value IntrinsicLibrary::genIeeeClass(mlir::Type resultType,
 
   assert(args.size() == 1);
   mlir::Value realVal = args[0];
-  mlir::FloatType realType = realVal.getType().dyn_cast<mlir::FloatType>();
+  mlir::FloatType realType = mlir::dyn_cast<mlir::FloatType>(realVal.getType());
   const unsigned intWidth = realType.getWidth();
   mlir::Type intType = builder.getIntegerType(intWidth);
   mlir::Value intVal =
@@ -4052,8 +4057,10 @@ IntrinsicLibrary::genIeeeCopySign(mlir::Type resultType,
   assert(args.size() == 2);
   mlir::Value xRealVal = args[0];
   mlir::Value yRealVal = args[1];
-  mlir::FloatType xRealType = xRealVal.getType().dyn_cast<mlir::FloatType>();
-  mlir::FloatType yRealType = yRealVal.getType().dyn_cast<mlir::FloatType>();
+  mlir::FloatType xRealType =
+      mlir::dyn_cast<mlir::FloatType>(xRealVal.getType());
+  mlir::FloatType yRealType =
+      mlir::dyn_cast<mlir::FloatType>(yRealVal.getType());
 
   if (yRealType == mlir::FloatType::getBF16(builder.getContext())) {
     // Workaround: CopySignOp and BitcastOp don't work for kind 3 arg Y.
@@ -4102,7 +4109,7 @@ void IntrinsicLibrary::genIeeeGetFlag(llvm::ArrayRef<fir::ExtendedValue> args) {
   mlir::Value flag = fir::getBase(args[0]);
   mlir::Value flagValue = fir::getBase(args[1]);
   mlir::Type resultTy =
-      flagValue.getType().dyn_cast<fir::ReferenceType>().getEleTy();
+      mlir::dyn_cast<fir::ReferenceType>(flagValue.getType()).getEleTy();
   mlir::Type i32Ty = builder.getIntegerType(32);
   mlir::Value zero = builder.createIntegerConstant(loc, i32Ty, 0);
   auto [fieldRef, ignore] = getFieldRef(builder, loc, flag);
@@ -4126,7 +4133,7 @@ void IntrinsicLibrary::genIeeeGetHaltingMode(
   mlir::Value flag = fir::getBase(args[0]);
   mlir::Value halting = fir::getBase(args[1]);
   mlir::Type resultTy =
-      halting.getType().dyn_cast<fir::ReferenceType>().getEleTy();
+      mlir::dyn_cast<fir::ReferenceType>(halting.getType()).getEleTy();
   mlir::Type i32Ty = builder.getIntegerType(32);
   mlir::Value zero = builder.createIntegerConstant(loc, i32Ty, 0);
   auto [fieldRef, ignore] = getFieldRef(builder, loc, flag);
@@ -4244,7 +4251,7 @@ mlir::Value IntrinsicLibrary::genIeeeLogb(mlir::Type resultType,
   //                 : ieee_copy_sign(X, 1.0) // +infinity or NaN
   assert(args.size() == 1);
   mlir::Value realVal = args[0];
-  mlir::FloatType realType = realVal.getType().dyn_cast<mlir::FloatType>();
+  mlir::FloatType realType = mlir::dyn_cast<mlir::FloatType>(realVal.getType());
   int bitWidth = realType.getWidth();
   mlir::Type intType = builder.getIntegerType(realType.getWidth());
   mlir::Value intVal =
@@ -4541,7 +4548,7 @@ mlir::Value IntrinsicLibrary::genIeeeSignbit(mlir::Type resultType,
   // Check if the sign bit of arg X is set.
   assert(args.size() == 1);
   mlir::Value realVal = args[0];
-  mlir::FloatType realType = realVal.getType().dyn_cast<mlir::FloatType>();
+  mlir::FloatType realType = mlir::dyn_cast<mlir::FloatType>(realVal.getType());
   int bitWidth = realType.getWidth();
   if (realType == mlir::FloatType::getBF16(builder.getContext())) {
     // Workaround: can't bitcast or convert real(3) to integer(2) or real(2).
@@ -4638,7 +4645,7 @@ mlir::Value IntrinsicLibrary::genIeeeValue(mlir::Type resultType,
   // A compiler generated call has one argument:
   //  - arg[0] is an index constant
   assert(args.size() == 1 || args.size() == 2);
-  mlir::FloatType realType = resultType.dyn_cast<mlir::FloatType>();
+  mlir::FloatType realType = mlir::dyn_cast<mlir::FloatType>(resultType);
   int bitWidth = realType.getWidth();
   mlir::Type intType = builder.getIntegerType(bitWidth);
   mlir::Type valueTy = bitWidth <= 64 ? intType : builder.getIntegerType(64);
@@ -4880,7 +4887,7 @@ mlir::Value IntrinsicLibrary::genIshft(mlir::Type resultType,
   //                    : I << abs(SHIFT)
   assert(args.size() == 2);
   mlir::Value bitSize = builder.createIntegerConstant(
-      loc, resultType, resultType.cast<mlir::IntegerType>().getWidth());
+      loc, resultType, mlir::cast<mlir::IntegerType>(resultType).getWidth());
   mlir::Value zero = builder.createIntegerConstant(loc, resultType, 0);
   mlir::Value shift = builder.createConvert(loc, resultType, args[1]);
   mlir::Value absShift = genAbs(resultType, {shift});
@@ -4916,7 +4923,7 @@ mlir::Value IntrinsicLibrary::genIshftc(mlir::Type resultType,
   // Return:  SHIFT == 0 || SIZE == abs(SHIFT) ? I : (unchanged | left | right)
   assert(args.size() == 3);
   mlir::Value bitSize = builder.createIntegerConstant(
-      loc, resultType, resultType.cast<mlir::IntegerType>().getWidth());
+      loc, resultType, mlir::cast<mlir::IntegerType>(resultType).getWidth());
   mlir::Value I = args[0];
   mlir::Value shift = builder.createConvert(loc, resultType, args[1]);
   mlir::Value size =
@@ -5023,7 +5030,7 @@ IntrinsicLibrary::genLoc(mlir::Type resultType,
   mlir::Value box = fir::getBase(args[0]);
   assert(fir::isa_box_type(box.getType()) &&
          "argument must have been lowered to box type");
-  bool isFunc = box.getType().isa<fir::BoxProcType>();
+  bool isFunc = mlir::isa<fir::BoxProcType>(box.getType());
   if (!isOptional(box)) {
     mlir::Value argAddr = getAddrFromBox(builder, loc, args[0], isFunc);
     return builder.createConvert(loc, resultType, argAddr);
@@ -5152,7 +5159,7 @@ IntrinsicLibrary::genMerge(mlir::Type,
   auto convertToStaticType = [&](mlir::Value polymorphic,
                                  mlir::Value other) -> mlir::Value {
     mlir::Type otherType = other.getType();
-    if (otherType.isa<fir::BaseBoxType>())
+    if (mlir::isa<fir::BaseBoxType>(otherType))
       return builder.create<fir::ReboxOp>(loc, otherType, polymorphic,
                                           /*shape*/ mlir::Value{},
                                           /*slice=*/mlir::Value{});
@@ -5205,7 +5212,7 @@ mlir::Value IntrinsicLibrary::genMergeBits(mlir::Type resultType,
 mlir::Value IntrinsicLibrary::genMod(mlir::Type resultType,
                                      llvm::ArrayRef<mlir::Value> args) {
   assert(args.size() == 2);
-  if (resultType.isa<mlir::IntegerType>())
+  if (mlir::isa<mlir::IntegerType>(resultType))
     return builder.create<mlir::arith::RemSIOp>(loc, args[0], args[1]);
 
   // Use runtime.
@@ -5227,7 +5234,7 @@ mlir::Value IntrinsicLibrary::genModulo(mlir::Type resultType,
   //  - Otherwise, when A/P < 0 and MOD(A,P) !=0, then MODULO(A, P) =
   //    A-FLOOR(A/P)*P = A-(INT(A/P)-1)*P = A-INT(A/P)*P+P = MOD(A,P)+P
   // Note that A/P < 0 if and only if A and P signs are different.
-  if (resultType.isa<mlir::IntegerType>()) {
+  if (mlir::isa<mlir::IntegerType>(resultType)) {
     auto remainder =
         builder.create<mlir::arith::RemSIOp>(loc, args[0], args[1]);
     auto argXor = builder.create<mlir::arith::XOrIOp>(loc, args[0], args[1]);
@@ -5340,7 +5347,7 @@ void IntrinsicLibrary::genMvbits(llvm::ArrayRef<fir::ExtendedValue> args) {
   mlir::Value zero = builder.createIntegerConstant(loc, resultType, 0);
   mlir::Value ones = builder.createAllOnesInteger(loc, resultType);
   mlir::Value bitSize = builder.createIntegerConstant(
-      loc, resultType, resultType.cast<mlir::IntegerType>().getWidth());
+      loc, resultType, mlir::cast<mlir::IntegerType>(resultType).getWidth());
   auto shiftCount = builder.create<mlir::arith::SubIOp>(loc, bitSize, len);
   auto mask = builder.create<mlir::arith::ShRUIOp>(loc, ones, shiftCount);
   auto unchangedTmp1 = builder.create<mlir::arith::ShLIOp>(loc, mask, topos);
@@ -5624,7 +5631,7 @@ IntrinsicLibrary::genReshape(mlir::Type resultType,
   assert(fir::BoxValue(shape).rank() == 1);
   mlir::Type shapeTy = shape.getType();
   mlir::Type shapeArrTy = fir::dyn_cast_ptrOrBoxEleTy(shapeTy);
-  auto resultRank = shapeArrTy.cast<fir::SequenceType>().getShape()[0];
+  auto resultRank = mlir::cast<fir::SequenceType>(shapeArrTy).getShape()[0];
 
   if (resultRank == fir::SequenceType::getUnknownExtent())
     TODO(loc, "intrinsic: reshape requires computing rank of result");
@@ -5821,6 +5828,35 @@ mlir::Value IntrinsicLibrary::genSetExponent(mlir::Type resultType,
                                    fir::getBase(args[1])));
 }
 
+// SHAPE
+fir::ExtendedValue
+IntrinsicLibrary::genShape(mlir::Type resultType,
+                           llvm::ArrayRef<fir::ExtendedValue> args) {
+  assert(args.size() >= 1);
+  const fir::ExtendedValue &array = args[0];
+  int rank = array.rank();
+  if (rank == 0)
+    TODO(loc, "shape intrinsic lowering with assumed-rank source");
+  mlir::Type indexType = builder.getIndexType();
+  mlir::Type extentType = fir::unwrapSequenceType(resultType);
+  mlir::Type seqType = fir::SequenceType::get(
+      {static_cast<fir::SequenceType::Extent>(rank)}, extentType);
+  mlir::Value shapeArray = builder.createTemporary(loc, seqType);
+  mlir::Type shapeAddrType = builder.getRefType(extentType);
+  for (int dim = 0; dim < rank; ++dim) {
+    mlir::Value extent = fir::factory::readExtent(builder, loc, array, dim);
+    extent = builder.createConvert(loc, extentType, extent);
+    auto index = builder.createIntegerConstant(loc, indexType, dim);
+    auto shapeAddr = builder.create<fir::CoordinateOp>(loc, shapeAddrType,
+                                                       shapeArray, index);
+    builder.create<fir::StoreOp>(loc, extent, shapeAddr);
+  }
+  mlir::Value shapeArrayExtent =
+      builder.createIntegerConstant(loc, indexType, rank);
+  llvm::SmallVector<mlir::Value> extents{shapeArrayExtent};
+  return fir::ArrayBoxValue{shapeArray, extents};
+}
+
 // SHIFTL, SHIFTR
 template <typename Shift>
 mlir::Value IntrinsicLibrary::genShift(mlir::Type resultType,
@@ -5888,7 +5924,7 @@ void IntrinsicLibrary::genSignalSubroutine(
 mlir::Value IntrinsicLibrary::genSign(mlir::Type resultType,
                                       llvm::ArrayRef<mlir::Value> args) {
   assert(args.size() == 2);
-  if (resultType.isa<mlir::IntegerType>()) {
+  if (mlir::isa<mlir::IntegerType>(resultType)) {
     mlir::Value abs = genAbs(resultType, {args[0]});
     mlir::Value zero = builder.createIntegerConstant(loc, resultType, 0);
     auto neg = builder.create<mlir::arith::SubIOp>(loc, zero, abs);
diff --git a/flang/lib/Optimizer/Builder/MutableBox.cpp b/flang/lib/Optimizer/Builder/MutableBox.cpp
index d4012e9c3d9d..76b920dba869 100644
--- a/flang/lib/Optimizer/Builder/MutableBox.cpp
+++ b/flang/lib/Optimizer/Builder/MutableBox.cpp
@@ -28,7 +28,7 @@ createNewFirBox(fir::FirOpBuilder &builder, mlir::Location loc,
                 const fir::MutableBoxValue &box, mlir::Value addr,
                 mlir::ValueRange lbounds, mlir::ValueRange extents,
                 mlir::ValueRange lengths, mlir::Value tdesc = {}) {
-  if (addr.getType().isa<fir::BaseBoxType>())
+  if (mlir::isa<fir::BaseBoxType>(addr.getType()))
     // The entity is already boxed.
     return builder.createConvert(loc, box.getBoxTy(), addr);
 
@@ -53,20 +53,21 @@ createNewFirBox(fir::FirOpBuilder &builder, mlir::Location loc,
   // error in the embox).
   llvm::SmallVector<mlir::Value> cleanedLengths;
   auto cleanedAddr = addr;
-  if (auto charTy = box.getEleTy().dyn_cast<fir::CharacterType>()) {
+  if (auto charTy = mlir::dyn_cast<fir::CharacterType>(box.getEleTy())) {
     // Cast address to box type so that both input and output type have
     // unknown or constant lengths.
     auto bt = box.getBaseTy();
     auto addrTy = addr.getType();
-    auto type = addrTy.isa<fir::HeapType>()      ? fir::HeapType::get(bt)
-                : addrTy.isa<fir::PointerType>() ? fir::PointerType::get(bt)
-                                                 : builder.getRefType(bt);
+    auto type = mlir::isa<fir::HeapType>(addrTy) ? fir::HeapType::get(bt)
+                : mlir::isa<fir::PointerType>(addrTy)
+                    ? fir::PointerType::get(bt)
+                    : builder.getRefType(bt);
     cleanedAddr = builder.createConvert(loc, type, addr);
     if (charTy.getLen() == fir::CharacterType::unknownLen())
       cleanedLengths.append(lengths.begin(), lengths.end());
   } else if (fir::isUnlimitedPolymorphicType(box.getBoxTy())) {
-    if (auto charTy = fir::dyn_cast_ptrEleTy(addr.getType())
-                          .dyn_cast<fir::CharacterType>()) {
+    if (auto charTy = mlir::dyn_cast<fir::CharacterType>(
+            fir::dyn_cast_ptrEleTy(addr.getType()))) {
       if (charTy.getLen() == fir::CharacterType::unknownLen())
         cleanedLengths.append(lengths.begin(), lengths.end());
     }
@@ -328,18 +329,18 @@ private:
 mlir::Value fir::factory::createUnallocatedBox(
     fir::FirOpBuilder &builder, mlir::Location loc, mlir::Type boxType,
     mlir::ValueRange nonDeferredParams, mlir::Value typeSourceBox) {
-  auto baseAddrType = boxType.dyn_cast<fir::BaseBoxType>().getEleTy();
+  auto baseAddrType = mlir::dyn_cast<fir::BaseBoxType>(boxType).getEleTy();
   if (!fir::isa_ref_type(baseAddrType))
     baseAddrType = builder.getRefType(baseAddrType);
   auto type = fir::unwrapRefType(baseAddrType);
   auto eleTy = fir::unwrapSequenceType(type);
-  if (auto recTy = eleTy.dyn_cast<fir::RecordType>())
+  if (auto recTy = mlir::dyn_cast<fir::RecordType>(eleTy))
     if (recTy.getNumLenParams() > 0)
       TODO(loc, "creating unallocated fir.box of derived type with length "
                 "parameters");
   auto nullAddr = builder.createNullConstant(loc, baseAddrType);
   mlir::Value shape;
-  if (auto seqTy = type.dyn_cast<fir::SequenceType>()) {
+  if (auto seqTy = mlir::dyn_cast<fir::SequenceType>(type)) {
     auto zero = builder.createIntegerConstant(loc, builder.getIndexType(), 0);
     llvm::SmallVector<mlir::Value> extents(seqTy.getDimension(), zero);
     shape = builder.createShape(
@@ -348,7 +349,7 @@ mlir::Value fir::factory::createUnallocatedBox(
   // Provide dummy length parameters if they are dynamic. If a length parameter
   // is deferred. It is set to zero here and will be set on allocation.
   llvm::SmallVector<mlir::Value> lenParams;
-  if (auto charTy = eleTy.dyn_cast<fir::CharacterType>()) {
+  if (auto charTy = mlir::dyn_cast<fir::CharacterType>(eleTy)) {
     if (charTy.getLen() == fir::CharacterType::unknownLen()) {
       if (!nonDeferredParams.empty()) {
         lenParams.push_back(nonDeferredParams[0]);
@@ -592,7 +593,7 @@ void fir::factory::associateMutableBoxWithRemap(
   auto cast = [&](mlir::Value addr) -> mlir::Value {
     // Cast base addr to new sequence type.
     auto ty = fir::dyn_cast_ptrEleTy(addr.getType());
-    if (auto seqTy = ty.dyn_cast<fir::SequenceType>()) {
+    if (auto seqTy = mlir::dyn_cast<fir::SequenceType>(ty)) {
       fir::SequenceType::Shape shape(newRank,
                                      fir::SequenceType::getUnknownExtent());
       ty = fir::SequenceType::get(shape, seqTy.getEleTy());
@@ -673,10 +674,10 @@ void fir::factory::disassociateMutableBox(fir::FirOpBuilder &builder,
   if (box.isPolymorphic() && polymorphicSetType) {
     // 7.3.2.3 point 7. The dynamic type of a disassociated pointer is the
     // same as its declared type.
-    auto boxTy = box.getBoxTy().dyn_cast<fir::BaseBoxType>();
+    auto boxTy = mlir::dyn_cast<fir::BaseBoxType>(box.getBoxTy());
     auto eleTy = fir::unwrapPassByRefType(boxTy.getEleTy());
     mlir::Type derivedType = fir::getDerivedType(eleTy);
-    if (auto recTy = derivedType.dyn_cast<fir::RecordType>()) {
+    if (auto recTy = mlir::dyn_cast<fir::RecordType>(derivedType)) {
       fir::runtime::genNullifyDerivedType(builder, loc, box.getAddr(), recTy,
                                           box.rank());
       return;
@@ -690,7 +691,7 @@ getNewLengths(fir::FirOpBuilder &builder, mlir::Location loc,
               const fir::MutableBoxValue &box, mlir::ValueRange lenParams) {
   llvm::SmallVector<mlir::Value> lengths;
   auto idxTy = builder.getIndexType();
-  if (auto charTy = box.getEleTy().dyn_cast<fir::CharacterType>()) {
+  if (auto charTy = mlir::dyn_cast<fir::CharacterType>(box.getEleTy())) {
     if (charTy.getLen() == fir::CharacterType::unknownLen()) {
       if (box.hasNonDeferredLenParams()) {
         lengths.emplace_back(
@@ -717,7 +718,7 @@ static mlir::Value allocateAndInitNewStorage(fir::FirOpBuilder &builder,
   auto lengths = getNewLengths(builder, loc, box, lenParams);
   auto newStorage = builder.create<fir::AllocMemOp>(
       loc, box.getBaseTy(), allocName, lengths, extents);
-  if (box.getEleTy().isa<fir::RecordType>()) {
+  if (mlir::isa<fir::RecordType>(box.getEleTy())) {
     // TODO: skip runtime initialization if this is not required. Currently,
     // there is no way to know here if a derived type needs it or not. But the
     // information is available at compile time and could be reflected here
@@ -742,7 +743,7 @@ void fir::factory::genInlinedAllocation(
                                               lengths, safeExtents);
   MutablePropertyWriter{builder, loc, box}.updateMutableBox(
       heap, lbounds, safeExtents, lengths);
-  if (box.getEleTy().isa<fir::RecordType>()) {
+  if (mlir::isa<fir::RecordType>(box.getEleTy())) {
     // TODO: skip runtime initialization if this is not required. Currently,
     // there is no way to know here if a derived type needs it or not. But the
     // information is available at compile time and could be reflected here
diff --git a/flang/lib/Optimizer/Builder/PPCIntrinsicCall.cpp b/flang/lib/Optimizer/Builder/PPCIntrinsicCall.cpp
index 160118e2c050..7f09e8822844 100644
--- a/flang/lib/Optimizer/Builder/PPCIntrinsicCall.cpp
+++ b/flang/lib/Optimizer/Builder/PPCIntrinsicCall.cpp
@@ -1119,7 +1119,7 @@ PPCIntrinsicLibrary::genVecAbs(mlir::Type resultType,
     funcOp = builder.createFunction(loc, fname, ftype);
     auto callOp{builder.create<fir::CallOp>(loc, funcOp, argBases[0])};
     return callOp.getResult(0);
-  } else if (auto eleTy = vTypeInfo.eleTy.dyn_cast<mlir::IntegerType>()) {
+  } else if (auto eleTy = mlir::dyn_cast<mlir::IntegerType>(vTypeInfo.eleTy)) {
     // vec_abs(arg1) = max(0 - arg1, arg1)
 
     auto newVecTy{mlir::VectorType::get(vTypeInfo.len, eleTy)};
@@ -1173,12 +1173,13 @@ fir::ExtendedValue PPCIntrinsicLibrary::genVecAddAndMulSubXor(
   assert(args.size() == 2);
   auto argBases{getBasesForArgs(args)};
   auto argsTy{getTypesForArgs(argBases)};
-  assert(argsTy[0].isa<fir::VectorType>() && argsTy[1].isa<fir::VectorType>());
+  assert(mlir::isa<fir::VectorType>(argsTy[0]) &&
+         mlir::isa<fir::VectorType>(argsTy[1]));
 
   auto vecTyInfo{getVecTypeFromFir(argBases[0])};
 
-  const auto isInteger{vecTyInfo.eleTy.isa<mlir::IntegerType>()};
-  const auto isFloat{vecTyInfo.eleTy.isa<mlir::FloatType>()};
+  const auto isInteger{mlir::isa<mlir::IntegerType>(vecTyInfo.eleTy)};
+  const auto isFloat{mlir::isa<mlir::FloatType>(vecTyInfo.eleTy)};
   assert((isInteger || isFloat) && "unknown vector type");
 
   auto vargs{convertVecArgs(builder, loc, vecTyInfo, argBases)};
@@ -1212,7 +1213,7 @@ fir::ExtendedValue PPCIntrinsicLibrary::genVecAddAndMulSubXor(
       arg2 = vargs[1];
     } else if (isFloat) {
       // bitcast the arguments to integer
-      auto wd{vecTyInfo.eleTy.dyn_cast<mlir::FloatType>().getWidth()};
+      auto wd{mlir::dyn_cast<mlir::FloatType>(vecTyInfo.eleTy).getWidth()};
       auto ftype{builder.getIntegerType(wd)};
       auto bcVecTy{mlir::VectorType::get(vecTyInfo.len, ftype)};
       arg1 = builder.create<mlir::vector::BitCastOp>(loc, bcVecTy, vargs[0]);
@@ -1450,7 +1451,7 @@ PPCIntrinsicLibrary::genVecCmp(mlir::Type resultType,
 
   mlir::Value res{nullptr};
 
-  if (auto eTy = vecTyInfo.eleTy.dyn_cast<mlir::IntegerType>()) {
+  if (auto eTy = mlir::dyn_cast<mlir::IntegerType>(vecTyInfo.eleTy)) {
     constexpr int firstArg{0};
     constexpr int secondArg{1};
     std::map<VecOp, std::array<int, 2>> argOrder{
@@ -1559,7 +1560,7 @@ PPCIntrinsicLibrary::genVecConvert(mlir::Type resultType,
   case VecOp::Ctf: {
     assert(args.size() == 2);
     auto convArg{builder.createConvert(loc, i32Ty, argBases[1])};
-    auto eTy{vecTyInfo.eleTy.dyn_cast<mlir::IntegerType>()};
+    auto eTy{mlir::dyn_cast<mlir::IntegerType>(vecTyInfo.eleTy)};
     assert(eTy && "Unsupported vector type");
     const auto isUnsigned{eTy.isUnsignedInteger()};
     const auto width{eTy.getWidth()};
@@ -1587,10 +1588,9 @@ PPCIntrinsicLibrary::genVecConvert(mlir::Type resultType,
                       : builder.create<mlir::LLVM::SIToFPOp>(loc, ty, vArg1)};
 
       // construct vector<1./(1<<arg1), 1.0/(1<<arg1)>
-      auto constInt{
+      auto constInt{mlir::dyn_cast_or_null<mlir::IntegerAttr>(
           mlir::dyn_cast<mlir::arith::ConstantOp>(argBases[1].getDefiningOp())
-              .getValue()
-              .dyn_cast_or_null<mlir::IntegerAttr>()};
+              .getValue())};
       assert(constInt && "expected integer constant argument");
       double f{1.0 / (1 << constInt.getInt())};
       llvm::SmallVector<double> vals{f, f};
@@ -1815,7 +1815,7 @@ static mlir::Value addOffsetToAddress(fir::FirOpBuilder &builder,
 static mlir::Value reverseVectorElements(fir::FirOpBuilder &builder,
                                          mlir::Location loc, mlir::Value v,
                                          int64_t len) {
-  assert(v.getType().isa<mlir::VectorType>());
+  assert(mlir::isa<mlir::VectorType>(v.getType()));
   assert(len > 0);
   llvm::SmallVector<int64_t, 16> mask;
   for (int64_t i = 0; i < len; ++i) {
@@ -2144,10 +2144,9 @@ PPCIntrinsicLibrary::genVecPerm(mlir::Type resultType,
   }
   case VecOp::Permi: {
     // arg3 is a constant
-    auto constIntOp{
+    auto constIntOp{mlir::dyn_cast_or_null<mlir::IntegerAttr>(
         mlir::dyn_cast<mlir::arith::ConstantOp>(argBases[2].getDefiningOp())
-            .getValue()
-            .dyn_cast_or_null<mlir::IntegerAttr>()};
+            .getValue())};
     assert(constIntOp && "expected integer constant argument");
     auto constInt{constIntOp.getInt()};
     // arg1, arg2, and result type share same VecTypeInfo
@@ -2321,10 +2320,9 @@ PPCIntrinsicLibrary::genVecShift(mlir::Type resultType,
     }
   } else if (vop == VecOp::Sld || vop == VecOp::Sldw) {
     assert(args.size() == 3);
-    auto constIntOp =
+    auto constIntOp = mlir::dyn_cast_or_null<mlir::IntegerAttr>(
         mlir::dyn_cast<mlir::arith::ConstantOp>(argBases[2].getDefiningOp())
-            .getValue()
-            .dyn_cast_or_null<mlir::IntegerAttr>();
+            .getValue());
     assert(constIntOp && "expected integer constant argument");
 
     // Bitcast to vector<16xi8>
@@ -2797,16 +2795,16 @@ void PPCIntrinsicLibrary::genMmaIntr(llvm::ArrayRef<fir::ExtendedValue> args) {
     auto vType{v.getType()};
     mlir::Type targetType{intrFuncType.getInput(j)};
     if (vType != targetType) {
-      if (targetType.isa<mlir::VectorType>()) {
+      if (mlir::isa<mlir::VectorType>(targetType)) {
         // Perform vector type conversion for arguments passed by value.
-        auto eleTy{vType.dyn_cast<fir::VectorType>().getEleTy()};
-        auto len{vType.dyn_cast<fir::VectorType>().getLen()};
+        auto eleTy{mlir::dyn_cast<fir::VectorType>(vType).getEleTy()};
+        auto len{mlir::dyn_cast<fir::VectorType>(vType).getLen()};
         mlir::VectorType mlirType = mlir::VectorType::get(len, eleTy);
         auto v0{builder.createConvert(loc, mlirType, v)};
         auto v1{builder.create<mlir::vector::BitCastOp>(loc, targetType, v0)};
         intrArgs.push_back(v1);
-      } else if (targetType.isa<mlir::IntegerType>() &&
-                 vType.isa<mlir::IntegerType>()) {
+      } else if (mlir::isa<mlir::IntegerType>(targetType) &&
+                 mlir::isa<mlir::IntegerType>(vType)) {
         auto v0{builder.createConvert(loc, targetType, v)};
         intrArgs.push_back(v0);
       } else {
@@ -2861,7 +2859,7 @@ void PPCIntrinsicLibrary::genVecStore(llvm::ArrayRef<fir::ExtendedValue> args) {
     if (arg1TyInfo.isFloat32()) {
       stTy = mlir::VectorType::get(len, i32ty);
       fname = "llvm.ppc.altivec.stvewx";
-    } else if (arg1TyInfo.eleTy.isa<mlir::IntegerType>()) {
+    } else if (mlir::isa<mlir::IntegerType>(arg1TyInfo.eleTy)) {
       stTy = mlir::VectorType::get(len, mlir::IntegerType::get(context, width));
 
       switch (width) {
diff --git a/flang/lib/Optimizer/Builder/Runtime/Allocatable.cpp b/flang/lib/Optimizer/Builder/Runtime/Allocatable.cpp
index abff0e150ab4..70a88ff18cb1 100644
--- a/flang/lib/Optimizer/Builder/Runtime/Allocatable.cpp
+++ b/flang/lib/Optimizer/Builder/Runtime/Allocatable.cpp
@@ -27,7 +27,7 @@ mlir::Value fir::runtime::genMoveAlloc(fir::FirOpBuilder &builder,
   if (fir::isPolymorphicType(from.getType()) &&
       !fir::isUnlimitedPolymorphicType(from.getType())) {
     fir::ClassType clTy =
-        fir::dyn_cast_ptrEleTy(from.getType()).dyn_cast<fir::ClassType>();
+        mlir::dyn_cast<fir::ClassType>(fir::dyn_cast_ptrEleTy(from.getType()));
     mlir::Type derivedType = fir::unwrapInnerType(clTy.getEleTy());
     declaredTypeDesc =
         builder.create<fir::TypeDescOp>(loc, mlir::TypeAttr::get(derivedType));
diff --git a/flang/lib/Optimizer/Builder/Runtime/Character.cpp b/flang/lib/Optimizer/Builder/Runtime/Character.cpp
index f3663439fdd5..b16819915d5a 100644
--- a/flang/lib/Optimizer/Builder/Runtime/Character.cpp
+++ b/flang/lib/Optimizer/Builder/Runtime/Character.cpp
@@ -39,15 +39,15 @@ static void genCharacterSearch(FN func, fir::FirOpBuilder &builder,
 
 /// Helper function to recover the KIND from the FIR type.
 static int discoverKind(mlir::Type ty) {
-  if (auto charTy = ty.dyn_cast<fir::CharacterType>())
+  if (auto charTy = mlir::dyn_cast<fir::CharacterType>(ty))
     return charTy.getFKind();
   if (auto eleTy = fir::dyn_cast_ptrEleTy(ty))
     return discoverKind(eleTy);
-  if (auto arrTy = ty.dyn_cast<fir::SequenceType>())
+  if (auto arrTy = mlir::dyn_cast<fir::SequenceType>(ty))
     return discoverKind(arrTy.getEleTy());
-  if (auto boxTy = ty.dyn_cast<fir::BoxCharType>())
+  if (auto boxTy = mlir::dyn_cast<fir::BoxCharType>(ty))
     return discoverKind(boxTy.getEleTy());
-  if (auto boxTy = ty.dyn_cast<fir::BoxType>())
+  if (auto boxTy = mlir::dyn_cast<fir::BoxType>(ty))
     return discoverKind(boxTy.getEleTy());
   llvm_unreachable("unexpected character type");
 }
diff --git a/flang/lib/Optimizer/Builder/Runtime/EnvironmentDefaults.cpp b/flang/lib/Optimizer/Builder/Runtime/EnvironmentDefaults.cpp
index a11b9339681e..6e280ac0c06c 100755
--- a/flang/lib/Optimizer/Builder/Runtime/EnvironmentDefaults.cpp
+++ b/flang/lib/Optimizer/Builder/Runtime/EnvironmentDefaults.cpp
@@ -13,7 +13,7 @@
 #include "flang/Optimizer/Support/InternalNames.h"
 #include "llvm/ADT/ArrayRef.h"
 
-void fir::runtime::genEnvironmentDefaults(
+fir::GlobalOp fir::runtime::genEnvironmentDefaults(
     fir::FirOpBuilder &builder, mlir::Location loc,
     const std::vector<Fortran::lower::EnvironmentDefault> &envDefaults) {
   std::string envDefaultListPtrName =
@@ -34,14 +34,13 @@ void fir::runtime::genEnvironmentDefaults(
 
   // If no defaults were specified, initialize with a null pointer.
   if (envDefaults.empty()) {
-    builder.createGlobalConstant(
+    return builder.createGlobalConstant(
         loc, envDefaultListRefTy, envDefaultListPtrName,
         [&](fir::FirOpBuilder &builder) {
           mlir::Value nullVal =
               builder.createNullConstant(loc, envDefaultListRefTy);
           builder.create<fir::HasValueOp>(loc, nullVal);
         });
-    return;
   }
 
   // Create the Item list.
@@ -99,7 +98,7 @@ void fir::runtime::genEnvironmentDefaults(
       envDefaultListBuilder, linkOnce);
 
   // Define the pointer to the list used by the runtime.
-  builder.createGlobalConstant(
+  return builder.createGlobalConstant(
       loc, envDefaultListRefTy, envDefaultListPtrName,
       [&](fir::FirOpBuilder &builder) {
         mlir::Value addr = builder.create<fir::AddrOfOp>(
diff --git a/flang/lib/Optimizer/Builder/Runtime/Intrinsics.cpp b/flang/lib/Optimizer/Builder/Runtime/Intrinsics.cpp
index 57c47da0f3f8..8b78a1688c73 100644
--- a/flang/lib/Optimizer/Builder/Runtime/Intrinsics.cpp
+++ b/flang/lib/Optimizer/Builder/Runtime/Intrinsics.cpp
@@ -228,7 +228,8 @@ void fir::runtime::genSystemClock(fir::FirOpBuilder &builder,
     fir::IfOp ifOp{};
     const bool isOptionalArg =
         fir::valueHasFirAttribute(arg, fir::getOptionalAttrName());
-    if (type.dyn_cast<fir::PointerType>() || type.dyn_cast<fir::HeapType>()) {
+    if (mlir::dyn_cast<fir::PointerType>(type) ||
+        mlir::dyn_cast<fir::HeapType>(type)) {
       // Check for a disassociated pointer or an unallocated allocatable.
       assert(!isOptionalArg && "invalid optional argument");
       ifOp = builder.create<fir::IfOp>(loc, builder.genIsNotNullAddr(loc, arg),
@@ -242,7 +243,8 @@ void fir::runtime::genSystemClock(fir::FirOpBuilder &builder,
       builder.setInsertionPointToStart(&ifOp.getThenRegion().front());
     mlir::Type kindTy = func.getFunctionType().getInput(0);
     int integerKind = 8;
-    if (auto intType = fir::unwrapRefType(type).dyn_cast<mlir::IntegerType>())
+    if (auto intType =
+            mlir::dyn_cast<mlir::IntegerType>(fir::unwrapRefType(type)))
       integerKind = intType.getWidth() / 8;
     mlir::Value kind = builder.createIntegerConstant(loc, kindTy, integerKind);
     mlir::Value res =
diff --git a/flang/lib/Optimizer/Builder/Runtime/Main.cpp b/flang/lib/Optimizer/Builder/Runtime/Main.cpp
new file mode 100644
index 000000000000..3b24fbca9cdb
--- /dev/null
+++ b/flang/lib/Optimizer/Builder/Runtime/Main.cpp
@@ -0,0 +1,62 @@
+//===-- Main.cpp - generate main runtime API calls --------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#include "flang/Optimizer/Builder/Runtime/Main.h"
+#include "flang/Optimizer/Builder/BoxValue.h"
+#include "flang/Optimizer/Builder/FIRBuilder.h"
+#include "flang/Optimizer/Builder/Runtime/RTBuilder.h"
+#include "flang/Optimizer/Dialect/FIROps.h"
+#include "flang/Optimizer/Dialect/FIRType.h"
+#include "flang/Runtime/main.h"
+#include "flang/Runtime/stop.h"
+
+using namespace Fortran::runtime;
+
+/// Create a `int main(...)` that calls the Fortran entry point
+void fir::runtime::genMain(fir::FirOpBuilder &builder, mlir::Location loc,
+                           fir::GlobalOp &env) {
+  auto *context = builder.getContext();
+  auto argcTy = builder.getDefaultIntegerType();
+  auto ptrTy = mlir::LLVM::LLVMPointerType::get(context);
+
+  // void ProgramStart(int argc, char** argv, char** envp,
+  //                   _QQEnvironmentDefaults* env)
+  auto startFn = builder.createFunction(
+      loc, RTNAME_STRING(ProgramStart),
+      mlir::FunctionType::get(context, {argcTy, ptrTy, ptrTy, ptrTy}, {}));
+  // void ProgramStop()
+  auto stopFn =
+      builder.createFunction(loc, RTNAME_STRING(ProgramEndStatement),
+                             mlir::FunctionType::get(context, {}, {}));
+
+  // int main(int argc, char** argv, char** envp)
+  auto mainFn = builder.createFunction(
+      loc, "main",
+      mlir::FunctionType::get(context, {argcTy, ptrTy, ptrTy}, argcTy));
+  // void _QQmain()
+  auto qqMainFn = builder.createFunction(
+      loc, "_QQmain", mlir::FunctionType::get(context, {}, {}));
+
+  mainFn.setPublic();
+
+  auto *block = mainFn.addEntryBlock();
+  mlir::OpBuilder::InsertionGuard insertGuard(builder);
+  builder.setInsertionPointToStart(block);
+
+  llvm::SmallVector<mlir::Value, 4> args(block->getArguments());
+  auto envAddr =
+      builder.create<fir::AddrOfOp>(loc, env.getType(), env.getSymbol());
+  args.push_back(envAddr);
+
+  builder.create<fir::CallOp>(loc, startFn, args);
+  builder.create<fir::CallOp>(loc, qqMainFn);
+  builder.create<fir::CallOp>(loc, stopFn);
+
+  mlir::Value ret = builder.createIntegerConstant(loc, argcTy, 0);
+  builder.create<mlir::func::ReturnOp>(loc, ret);
+}
diff --git a/flang/lib/Optimizer/Builder/Runtime/Ragged.cpp b/flang/lib/Optimizer/Builder/Runtime/Ragged.cpp
index 4d33282a35d9..e5d0fb0fb27a 100644
--- a/flang/lib/Optimizer/Builder/Runtime/Ragged.cpp
+++ b/flang/lib/Optimizer/Builder/Runtime/Ragged.cpp
@@ -32,7 +32,8 @@ void fir::runtime::genRaggedArrayAllocate(mlir::Location loc,
   // Position of the bufferPointer in the header struct.
   auto one = builder.createIntegerConstant(loc, i32Ty, 1);
   auto eleTy = fir::unwrapSequenceType(fir::unwrapRefType(header.getType()));
-  auto ptrTy = builder.getRefType(eleTy.cast<mlir::TupleType>().getType(1));
+  auto ptrTy =
+      builder.getRefType(mlir::cast<mlir::TupleType>(eleTy).getType(1));
   auto ptr = builder.create<fir::CoordinateOp>(loc, ptrTy, header, one);
   auto heap = builder.create<fir::LoadOp>(loc, ptr);
   auto cmp = builder.genIsNullAddr(loc, heap);
diff --git a/flang/lib/Optimizer/Builder/Runtime/Reduction.cpp b/flang/lib/Optimizer/Builder/Runtime/Reduction.cpp
index 66fbaddcbda1..d4076067bf10 100644
--- a/flang/lib/Optimizer/Builder/Runtime/Reduction.cpp
+++ b/flang/lib/Optimizer/Builder/Runtime/Reduction.cpp
@@ -666,7 +666,7 @@ void fir::runtime::genMaxloc(fir::FirOpBuilder &builder, mlir::Location loc,
   mlir::func::FuncOp func;
   auto ty = arrayBox.getType();
   auto arrTy = fir::dyn_cast_ptrOrBoxEleTy(ty);
-  auto eleTy = arrTy.cast<fir::SequenceType>().getEleTy();
+  auto eleTy = mlir::cast<fir::SequenceType>(arrTy).getEleTy();
   fir::factory::CharacterExprHelper charHelper{builder, loc};
   if (eleTy.isF32())
     func = fir::runtime::getRuntimeFunc<mkRTKey(MaxlocReal4)>(loc, builder);
@@ -713,7 +713,7 @@ mlir::Value fir::runtime::genMaxval(fir::FirOpBuilder &builder,
   mlir::func::FuncOp func;
   auto ty = arrayBox.getType();
   auto arrTy = fir::dyn_cast_ptrOrBoxEleTy(ty);
-  auto eleTy = arrTy.cast<fir::SequenceType>().getEleTy();
+  auto eleTy = mlir::cast<fir::SequenceType>(arrTy).getEleTy();
   auto dim = builder.createIntegerConstant(loc, builder.getIndexType(), 0);
 
   if (eleTy.isF32())
@@ -781,7 +781,7 @@ void fir::runtime::genMinloc(fir::FirOpBuilder &builder, mlir::Location loc,
   mlir::func::FuncOp func;
   auto ty = arrayBox.getType();
   auto arrTy = fir::dyn_cast_ptrOrBoxEleTy(ty);
-  auto eleTy = arrTy.cast<fir::SequenceType>().getEleTy();
+  auto eleTy = mlir::cast<fir::SequenceType>(arrTy).getEleTy();
   fir::factory::CharacterExprHelper charHelper{builder, loc};
   if (eleTy.isF32())
     func = fir::runtime::getRuntimeFunc<mkRTKey(MinlocReal4)>(loc, builder);
@@ -853,7 +853,7 @@ mlir::Value fir::runtime::genMinval(fir::FirOpBuilder &builder,
   mlir::func::FuncOp func;
   auto ty = arrayBox.getType();
   auto arrTy = fir::dyn_cast_ptrOrBoxEleTy(ty);
-  auto eleTy = arrTy.cast<fir::SequenceType>().getEleTy();
+  auto eleTy = mlir::cast<fir::SequenceType>(arrTy).getEleTy();
   auto dim = builder.createIntegerConstant(loc, builder.getIndexType(), 0);
 
   if (eleTy.isF32())
@@ -895,7 +895,7 @@ void fir::runtime::genNorm2Dim(fir::FirOpBuilder &builder, mlir::Location loc,
   mlir::func::FuncOp func;
   auto ty = arrayBox.getType();
   auto arrTy = fir::dyn_cast_ptrOrBoxEleTy(ty);
-  auto eleTy = arrTy.cast<fir::SequenceType>().getEleTy();
+  auto eleTy = mlir::cast<fir::SequenceType>(arrTy).getEleTy();
   if (eleTy.isF128())
     func = fir::runtime::getRuntimeFunc<ForcedNorm2DimReal16>(loc, builder);
   else
@@ -917,7 +917,7 @@ mlir::Value fir::runtime::genNorm2(fir::FirOpBuilder &builder,
   mlir::func::FuncOp func;
   auto ty = arrayBox.getType();
   auto arrTy = fir::dyn_cast_ptrOrBoxEleTy(ty);
-  auto eleTy = arrTy.cast<fir::SequenceType>().getEleTy();
+  auto eleTy = mlir::cast<fir::SequenceType>(arrTy).getEleTy();
   auto dim = builder.createIntegerConstant(loc, builder.getIndexType(), 0);
 
   if (eleTy.isF32())
@@ -968,7 +968,7 @@ mlir::Value fir::runtime::genProduct(fir::FirOpBuilder &builder,
   mlir::func::FuncOp func;
   auto ty = arrayBox.getType();
   auto arrTy = fir::dyn_cast_ptrOrBoxEleTy(ty);
-  auto eleTy = arrTy.cast<fir::SequenceType>().getEleTy();
+  auto eleTy = mlir::cast<fir::SequenceType>(arrTy).getEleTy();
   auto dim = builder.createIntegerConstant(loc, builder.getIndexType(), 0);
 
   if (eleTy.isF32())
@@ -1069,7 +1069,7 @@ mlir::Value fir::runtime::genDotProduct(fir::FirOpBuilder &builder,
   else if (eleTy.isInteger(builder.getKindMap().getIntegerBitsize(16)))
     func =
         fir::runtime::getRuntimeFunc<ForcedDotProductInteger16>(loc, builder);
-  else if (eleTy.isa<fir::LogicalType>())
+  else if (mlir::isa<fir::LogicalType>(eleTy))
     func =
         fir::runtime::getRuntimeFunc<mkRTKey(DotProductLogical)>(loc, builder);
   else
@@ -1111,7 +1111,7 @@ mlir::Value fir::runtime::genSum(fir::FirOpBuilder &builder, mlir::Location loc,
   mlir::func::FuncOp func;
   auto ty = arrayBox.getType();
   auto arrTy = fir::dyn_cast_ptrOrBoxEleTy(ty);
-  auto eleTy = arrTy.cast<fir::SequenceType>().getEleTy();
+  auto eleTy = mlir::cast<fir::SequenceType>(arrTy).getEleTy();
   auto dim = builder.createIntegerConstant(loc, builder.getIndexType(), 0);
 
   if (eleTy.isF32())
@@ -1173,7 +1173,7 @@ mlir::Value fir::runtime::genSum(fir::FirOpBuilder &builder, mlir::Location loc,
     mlir::func::FuncOp func;                                                   \
     auto ty = arrayBox.getType();                                              \
     auto arrTy = fir::dyn_cast_ptrOrBoxEleTy(ty);                              \
-    auto eleTy = arrTy.cast<fir::SequenceType>().getEleTy();                   \
+    auto eleTy = mlir::cast<fir::SequenceType>(arrTy).getEleTy();              \
     auto dim = builder.createIntegerConstant(loc, builder.getIndexType(), 0);  \
                                                                                \
     if (eleTy.isInteger(builder.getKindMap().getIntegerBitsize(1)))            \
diff --git a/flang/lib/Optimizer/CodeGen/BoxedProcedure.cpp b/flang/lib/Optimizer/CodeGen/BoxedProcedure.cpp
index 48173033ecbe..5229d40f2250 100644
--- a/flang/lib/Optimizer/CodeGen/BoxedProcedure.cpp
+++ b/flang/lib/Optimizer/CodeGen/BoxedProcedure.cpp
@@ -51,9 +51,9 @@ public:
   /// not at all depending on the implementation target's characteristics and
   /// preference.
   bool needsConversion(mlir::Type ty) {
-    if (ty.isa<BoxProcType>())
+    if (mlir::isa<BoxProcType>(ty))
       return true;
-    if (auto funcTy = ty.dyn_cast<mlir::FunctionType>()) {
+    if (auto funcTy = mlir::dyn_cast<mlir::FunctionType>(ty)) {
       for (auto t : funcTy.getInputs())
         if (needsConversion(t))
           return true;
@@ -62,13 +62,13 @@ public:
           return true;
       return false;
     }
-    if (auto tupleTy = ty.dyn_cast<mlir::TupleType>()) {
+    if (auto tupleTy = mlir::dyn_cast<mlir::TupleType>(ty)) {
       for (auto t : tupleTy.getTypes())
         if (needsConversion(t))
           return true;
       return false;
     }
-    if (auto recTy = ty.dyn_cast<RecordType>()) {
+    if (auto recTy = mlir::dyn_cast<RecordType>(ty)) {
       auto visited = visitedTypes.find(ty);
       if (visited != visitedTypes.end())
         return visited->second;
@@ -97,11 +97,11 @@ public:
         visitedTypes.find(ty)->second = result;
       return result;
     }
-    if (auto boxTy = ty.dyn_cast<BaseBoxType>())
+    if (auto boxTy = mlir::dyn_cast<BaseBoxType>(ty))
       return needsConversion(boxTy.getEleTy());
     if (isa_ref_type(ty))
       return needsConversion(unwrapRefType(ty));
-    if (auto t = ty.dyn_cast<SequenceType>())
+    if (auto t = mlir::dyn_cast<SequenceType>(ty))
       return needsConversion(unwrapSequenceType(ty));
     return false;
   }
@@ -246,7 +246,7 @@ public:
           if (typeConverter.needsConversion(ty)) {
             rewriter.startOpModification(func);
             auto toTy =
-                typeConverter.convertType(ty).cast<mlir::FunctionType>();
+                mlir::cast<mlir::FunctionType>(typeConverter.convertType(ty));
             if (!func.empty())
               for (auto e : llvm::enumerate(toTy.getInputs())) {
                 unsigned i = e.index();
@@ -263,7 +263,7 @@ public:
           // Rewrite all `fir.emboxproc` ops to either `fir.convert` or a thunk
           // as required.
           mlir::Type toTy = typeConverter.convertType(
-              embox.getType().cast<BoxProcType>().getEleTy());
+              mlir::cast<BoxProcType>(embox.getType()).getEleTy());
           rewriter.setInsertionPoint(embox);
           if (embox.getHost()) {
             // Create the thunk.
diff --git a/flang/lib/Optimizer/CodeGen/CGOps.cpp b/flang/lib/Optimizer/CodeGen/CGOps.cpp
index c3bcdeaf86db..44d07d26dd2b 100644
--- a/flang/lib/Optimizer/CodeGen/CGOps.cpp
+++ b/flang/lib/Optimizer/CodeGen/CGOps.cpp
@@ -41,24 +41,24 @@ unsigned fir::cg::XEmboxOp::getOutRank() {
 }
 
 unsigned fir::cg::XReboxOp::getOutRank() {
-  if (auto seqTy =
-          fir::dyn_cast_ptrOrBoxEleTy(getType()).dyn_cast<fir::SequenceType>())
+  if (auto seqTy = mlir::dyn_cast<fir::SequenceType>(
+          fir::dyn_cast_ptrOrBoxEleTy(getType())))
     return seqTy.getDimension();
   return 0;
 }
 
 unsigned fir::cg::XReboxOp::getRank() {
-  if (auto seqTy = fir::dyn_cast_ptrOrBoxEleTy(getBox().getType())
-                       .dyn_cast<fir::SequenceType>())
+  if (auto seqTy = mlir::dyn_cast<fir::SequenceType>(
+          fir::dyn_cast_ptrOrBoxEleTy(getBox().getType())))
     return seqTy.getDimension();
   return 0;
 }
 
 unsigned fir::cg::XArrayCoorOp::getRank() {
   auto memrefTy = getMemref().getType();
-  if (memrefTy.isa<fir::BaseBoxType>())
-    if (auto seqty =
-            fir::dyn_cast_ptrOrBoxEleTy(memrefTy).dyn_cast<fir::SequenceType>())
+  if (mlir::isa<fir::BaseBoxType>(memrefTy))
+    if (auto seqty = mlir::dyn_cast<fir::SequenceType>(
+            fir::dyn_cast_ptrOrBoxEleTy(memrefTy)))
       return seqty.getDimension();
   return getShape().size();
 }
diff --git a/flang/lib/Optimizer/CodeGen/CodeGen.cpp b/flang/lib/Optimizer/CodeGen/CodeGen.cpp
index 921eac2f8f4b..b4705aa47992 100644
--- a/flang/lib/Optimizer/CodeGen/CodeGen.cpp
+++ b/flang/lib/Optimizer/CodeGen/CodeGen.cpp
@@ -101,7 +101,7 @@ static int64_t getConstantIntValue(mlir::Value val) {
 }
 
 static unsigned getTypeDescFieldId(mlir::Type ty) {
-  auto isArray = fir::dyn_cast_ptrOrBoxEleTy(ty).isa<fir::SequenceType>();
+  auto isArray = mlir::isa<fir::SequenceType>(fir::dyn_cast_ptrOrBoxEleTy(ty));
   return isArray ? kOptTypePtrPosInBox : kDimsPosInBox;
 }
 static unsigned getLenParamFieldId(mlir::Type ty) {
@@ -147,7 +147,7 @@ genAllocationScaleSize(OP op, mlir::Type ity,
                        mlir::ConversionPatternRewriter &rewriter) {
   mlir::Location loc = op.getLoc();
   mlir::Type dataTy = op.getInType();
-  auto seqTy = dataTy.dyn_cast<fir::SequenceType>();
+  auto seqTy = mlir::dyn_cast<fir::SequenceType>(dataTy);
   fir::SequenceType::Extent constSize = 1;
   if (seqTy) {
     int constRows = seqTy.getConstantRows();
@@ -191,13 +191,13 @@ struct AllocaOpConversion : public fir::FIROpConversion<fir::AllocaOp> {
       for (; i < end; ++i)
         lenParams.push_back(operands[i]);
       mlir::Type scalarType = fir::unwrapSequenceType(alloc.getInType());
-      if (auto chrTy = scalarType.dyn_cast<fir::CharacterType>()) {
+      if (auto chrTy = mlir::dyn_cast<fir::CharacterType>(scalarType)) {
         fir::CharacterType rawCharTy = fir::CharacterType::getUnknownLen(
             chrTy.getContext(), chrTy.getFKind());
         llvmObjectType = convertType(rawCharTy);
         assert(end == 1);
         size = integerCast(loc, rewriter, ity, lenParams[0]);
-      } else if (auto recTy = scalarType.dyn_cast<fir::RecordType>()) {
+      } else if (auto recTy = mlir::dyn_cast<fir::RecordType>(scalarType)) {
         mlir::LLVM::LLVMFuncOp memSizeFn =
             getDependentTypeMemSizeFn(recTy, alloc, rewriter);
         if (!memSizeFn)
@@ -265,7 +265,8 @@ struct BoxAddrOpConversion : public fir::FIROpConversion<fir::BoxAddrOp> {
                   mlir::ConversionPatternRewriter &rewriter) const override {
     mlir::Value a = adaptor.getOperands()[0];
     auto loc = boxaddr.getLoc();
-    if (auto argty = boxaddr.getVal().getType().dyn_cast<fir::BaseBoxType>()) {
+    if (auto argty =
+            mlir::dyn_cast<fir::BaseBoxType>(boxaddr.getVal().getType())) {
       TypePair boxTyPair = getBoxTypePair(argty);
       rewriter.replaceOp(boxaddr,
                          getBaseAddrFromBox(loc, boxTyPair, a, rewriter));
@@ -476,24 +477,25 @@ struct StringLitOpConversion : public fir::FIROpConversion<fir::StringLitOp> {
                   mlir::ConversionPatternRewriter &rewriter) const override {
     auto ty = convertType(constop.getType());
     auto attr = constop.getValue();
-    if (attr.isa<mlir::StringAttr>()) {
+    if (mlir::isa<mlir::StringAttr>(attr)) {
       rewriter.replaceOpWithNewOp<mlir::LLVM::ConstantOp>(constop, ty, attr);
       return mlir::success();
     }
 
-    auto charTy = constop.getType().cast<fir::CharacterType>();
+    auto charTy = mlir::cast<fir::CharacterType>(constop.getType());
     unsigned bits = lowerTy().characterBitsize(charTy);
     mlir::Type intTy = rewriter.getIntegerType(bits);
     mlir::Location loc = constop.getLoc();
     mlir::Value cst = rewriter.create<mlir::LLVM::UndefOp>(loc, ty);
-    if (auto arr = attr.dyn_cast<mlir::DenseElementsAttr>()) {
+    if (auto arr = mlir::dyn_cast<mlir::DenseElementsAttr>(attr)) {
       cst = rewriter.create<mlir::LLVM::ConstantOp>(loc, ty, arr);
-    } else if (auto arr = attr.dyn_cast<mlir::ArrayAttr>()) {
+    } else if (auto arr = mlir::dyn_cast<mlir::ArrayAttr>(attr)) {
       for (auto a : llvm::enumerate(arr.getValue())) {
         // convert each character to a precise bitsize
         auto elemAttr = mlir::IntegerAttr::get(
             intTy,
-            a.value().cast<mlir::IntegerAttr>().getValue().zextOrTrunc(bits));
+            mlir::cast<mlir::IntegerAttr>(a.value()).getValue().zextOrTrunc(
+                bits));
         auto elemCst =
             rewriter.create<mlir::LLVM::ConstantOp>(loc, intTy, elemAttr);
         cst = rewriter.create<mlir::LLVM::InsertValueOp>(loc, cst, elemCst,
@@ -528,9 +530,9 @@ struct CallOpConversion : public fir::FIROpConversion<fir::CallOp> {
 } // namespace
 
 static mlir::Type getComplexEleTy(mlir::Type complex) {
-  if (auto cc = complex.dyn_cast<mlir::ComplexType>())
+  if (auto cc = mlir::dyn_cast<mlir::ComplexType>(complex))
     return cc.getElementType();
-  return complex.cast<fir::ComplexType>().getElementType();
+  return mlir::cast<fir::ComplexType>(complex).getElementType();
 }
 
 namespace {
@@ -599,7 +601,7 @@ struct ConstcOpConversion : public fir::FIROpConversion<fir::ConstcOp> {
   }
 
   inline llvm::APFloat getValue(mlir::Attribute attr) const {
-    return attr.cast<fir::RealAttr>().getValue();
+    return mlir::cast<fir::RealAttr>(attr).getValue();
   }
 };
 
@@ -608,7 +610,7 @@ struct ConvertOpConversion : public fir::FIROpConversion<fir::ConvertOp> {
   using FIROpConversion::FIROpConversion;
 
   static bool isFloatingPointTy(mlir::Type ty) {
-    return ty.isa<mlir::FloatType>();
+    return mlir::isa<mlir::FloatType>(ty);
   }
 
   mlir::LogicalResult
@@ -628,7 +630,8 @@ struct ConvertOpConversion : public fir::FIROpConversion<fir::ConvertOp> {
     auto loc = convert.getLoc();
     auto i1Type = mlir::IntegerType::get(convert.getContext(), 1);
 
-    if (fromFirTy.isa<fir::LogicalType>() || toFirTy.isa<fir::LogicalType>()) {
+    if (mlir::isa<fir::LogicalType>(fromFirTy) ||
+        mlir::isa<fir::LogicalType>(toFirTy)) {
       // By specification fir::LogicalType value may be any number,
       // where non-zero value represents .true. and zero value represents
       // .false.
@@ -641,7 +644,8 @@ struct ConvertOpConversion : public fir::FIROpConversion<fir::ConvertOp> {
       // Conversion from narrow logical to wide logical may be implemented
       // as a zero or sign extension of the input, but it may use value
       // normalization as well.
-      if (!fromTy.isa<mlir::IntegerType>() || !toTy.isa<mlir::IntegerType>())
+      if (!mlir::isa<mlir::IntegerType>(fromTy) ||
+          !mlir::isa<mlir::IntegerType>(toTy))
         return mlir::emitError(loc)
                << "unsupported types for logical conversion: " << fromTy
                << " -> " << toTy;
@@ -722,13 +726,13 @@ struct ConvertOpConversion : public fir::FIROpConversion<fir::ConvertOp> {
         rewriter.replaceOp(convert, v);
         return mlir::success();
       }
-      if (toTy.isa<mlir::IntegerType>()) {
+      if (mlir::isa<mlir::IntegerType>(toTy)) {
         rewriter.replaceOpWithNewOp<mlir::LLVM::FPToSIOp>(convert, toTy, op0);
         return mlir::success();
       }
-    } else if (fromTy.isa<mlir::IntegerType>()) {
+    } else if (mlir::isa<mlir::IntegerType>(fromTy)) {
       // Integer to integer conversion.
-      if (toTy.isa<mlir::IntegerType>()) {
+      if (mlir::isa<mlir::IntegerType>(toTy)) {
         auto fromBits = mlir::LLVM::getPrimitiveTypeSizeInBits(fromTy);
         auto toBits = mlir::LLVM::getPrimitiveTypeSizeInBits(toTy);
         assert(fromBits != toBits);
@@ -749,18 +753,18 @@ struct ConvertOpConversion : public fir::FIROpConversion<fir::ConvertOp> {
         return mlir::success();
       }
       // Integer to pointer conversion.
-      if (toTy.isa<mlir::LLVM::LLVMPointerType>()) {
+      if (mlir::isa<mlir::LLVM::LLVMPointerType>(toTy)) {
         rewriter.replaceOpWithNewOp<mlir::LLVM::IntToPtrOp>(convert, toTy, op0);
         return mlir::success();
       }
-    } else if (fromTy.isa<mlir::LLVM::LLVMPointerType>()) {
+    } else if (mlir::isa<mlir::LLVM::LLVMPointerType>(fromTy)) {
       // Pointer to integer conversion.
-      if (toTy.isa<mlir::IntegerType>()) {
+      if (mlir::isa<mlir::IntegerType>(toTy)) {
         rewriter.replaceOpWithNewOp<mlir::LLVM::PtrToIntOp>(convert, toTy, op0);
         return mlir::success();
       }
       // Pointer to pointer conversion.
-      if (toTy.isa<mlir::LLVM::LLVMPointerType>()) {
+      if (mlir::isa<mlir::LLVM::LLVMPointerType>(toTy)) {
         rewriter.replaceOpWithNewOp<mlir::LLVM::BitcastOp>(convert, toTy, op0);
         return mlir::success();
       }
@@ -842,11 +846,11 @@ struct EmboxCharOpConversion : public fir::FIROpConversion<fir::EmboxCharOp> {
     auto llvmStruct = rewriter.create<mlir::LLVM::UndefOp>(loc, llvmStructTy);
 
     mlir::Type lenTy =
-        llvmStructTy.cast<mlir::LLVM::LLVMStructType>().getBody()[1];
+        mlir::cast<mlir::LLVM::LLVMStructType>(llvmStructTy).getBody()[1];
     mlir::Value lenAfterCast = integerCast(loc, rewriter, lenTy, charBufferLen);
 
     mlir::Type addrTy =
-        llvmStructTy.cast<mlir::LLVM::LLVMStructType>().getBody()[0];
+        mlir::cast<mlir::LLVM::LLVMStructType>(llvmStructTy).getBody()[0];
     if (addrTy != charBuffer.getType())
       charBuffer =
           rewriter.create<mlir::LLVM::BitcastOp>(loc, addrTy, charBuffer);
@@ -979,9 +983,10 @@ static mlir::SymbolRefAttr getFree(fir::FreeMemOp op,
 
 static unsigned getDimension(mlir::LLVM::LLVMArrayType ty) {
   unsigned result = 1;
-  for (auto eleTy = ty.getElementType().dyn_cast<mlir::LLVM::LLVMArrayType>();
-       eleTy;
-       eleTy = eleTy.getElementType().dyn_cast<mlir::LLVM::LLVMArrayType>())
+  for (auto eleTy =
+           mlir::dyn_cast<mlir::LLVM::LLVMArrayType>(ty.getElementType());
+       eleTy; eleTy = mlir::dyn_cast<mlir::LLVM::LLVMArrayType>(
+                  eleTy.getElementType()))
     ++result;
   return result;
 }
@@ -1052,9 +1057,9 @@ struct EmboxCommonConversion : public fir::FIROpConversion<OP> {
 
   static int getCFIAttr(fir::BaseBoxType boxTy) {
     auto eleTy = boxTy.getEleTy();
-    if (eleTy.isa<fir::PointerType>())
+    if (mlir::isa<fir::PointerType>(eleTy))
       return CFI_attribute_pointer;
-    if (eleTy.isa<fir::HeapType>())
+    if (mlir::isa<fir::HeapType>(eleTy))
       return CFI_attribute_allocatable;
     return CFI_attribute_other;
   }
@@ -1082,27 +1087,29 @@ struct EmboxCommonConversion : public fir::FIROpConversion<OP> {
     auto i64Ty = mlir::IntegerType::get(rewriter.getContext(), 64);
     if (auto eleTy = fir::dyn_cast_ptrEleTy(boxEleTy))
       boxEleTy = eleTy;
-    if (auto seqTy = boxEleTy.dyn_cast<fir::SequenceType>())
+    if (auto seqTy = mlir::dyn_cast<fir::SequenceType>(boxEleTy))
       return getSizeAndTypeCode(loc, rewriter, seqTy.getEleTy(), lenParams);
-    if (boxEleTy.isa<mlir::NoneType>()) // unlimited polymorphic or assumed type
+    if (mlir::isa<mlir::NoneType>(
+            boxEleTy)) // unlimited polymorphic or assumed type
       return {rewriter.create<mlir::LLVM::ConstantOp>(loc, i64Ty, 0),
               this->genConstantOffset(loc, rewriter, CFI_type_other)};
     mlir::Value typeCodeVal = this->genConstantOffset(
         loc, rewriter,
         fir::getTypeCode(boxEleTy, this->lowerTy().getKindMap()));
-    if (fir::isa_integer(boxEleTy) || boxEleTy.dyn_cast<fir::LogicalType>() ||
-        fir::isa_real(boxEleTy) || fir::isa_complex(boxEleTy))
+    if (fir::isa_integer(boxEleTy) ||
+        mlir::dyn_cast<fir::LogicalType>(boxEleTy) || fir::isa_real(boxEleTy) ||
+        fir::isa_complex(boxEleTy))
       return {genTypeStrideInBytes(loc, i64Ty, rewriter,
                                    this->convertType(boxEleTy)),
               typeCodeVal};
-    if (auto charTy = boxEleTy.dyn_cast<fir::CharacterType>())
+    if (auto charTy = mlir::dyn_cast<fir::CharacterType>(boxEleTy))
       return {getCharacterByteSize(loc, rewriter, charTy, lenParams),
               typeCodeVal};
     if (fir::isa_ref_type(boxEleTy)) {
       auto ptrTy = ::getLlvmPtrType(rewriter.getContext());
       return {genTypeStrideInBytes(loc, i64Ty, rewriter, ptrTy), typeCodeVal};
     }
-    if (boxEleTy.isa<fir::RecordType>())
+    if (mlir::isa<fir::RecordType>(boxEleTy))
       return {genTypeStrideInBytes(loc, i64Ty, rewriter,
                                    this->convertType(boxEleTy)),
               typeCodeVal};
@@ -1211,8 +1218,8 @@ struct EmboxCommonConversion : public fir::FIROpConversion<OP> {
       if (!typeDesc) {
         if (useInputType) {
           mlir::Type innerType = fir::unwrapInnerType(inputType);
-          if (innerType && innerType.template isa<fir::RecordType>()) {
-            auto recTy = innerType.template dyn_cast<fir::RecordType>();
+          if (innerType && mlir::isa<fir::RecordType>(innerType)) {
+            auto recTy = mlir::dyn_cast<fir::RecordType>(innerType);
             typeDesc = getTypeDescriptor(mod, rewriter, loc, recTy);
           } else {
             // Unlimited polymorphic type descriptor with no record type. Set
@@ -1250,7 +1257,7 @@ struct EmboxCommonConversion : public fir::FIROpConversion<OP> {
                        mlir::ValueRange lenParams, mlir::Value sourceBox = {},
                        mlir::Type sourceBoxType = {}) const {
     auto loc = box.getLoc();
-    auto boxTy = box.getType().template dyn_cast<fir::BaseBoxType>();
+    auto boxTy = mlir::dyn_cast<fir::BaseBoxType>(box.getType());
     bool useInputType = fir::isPolymorphicType(boxTy) &&
                         !fir::isUnlimitedPolymorphicType(inputType);
     llvm::SmallVector<mlir::Value> typeparams = lenParams;
@@ -1293,8 +1300,8 @@ struct EmboxCommonConversion : public fir::FIROpConversion<OP> {
                        mlir::ValueRange lenParams,
                        mlir::Value typeDesc = {}) const {
     auto loc = box.getLoc();
-    auto boxTy = box.getType().dyn_cast<fir::BaseBoxType>();
-    auto inputBoxTy = box.getBox().getType().dyn_cast<fir::BaseBoxType>();
+    auto boxTy = mlir::dyn_cast<fir::BaseBoxType>(box.getType());
+    auto inputBoxTy = mlir::dyn_cast<fir::BaseBoxType>(box.getBox().getType());
     auto inputBoxTyPair = this->getBoxTypePair(inputBoxTy);
     llvm::SmallVector<mlir::Value> typeparams = lenParams;
     if (!box.getSubstr().empty() && fir::hasDynamicSize(boxTy.getEleTy()))
@@ -1343,7 +1350,7 @@ struct EmboxCommonConversion : public fir::FIROpConversion<OP> {
     mlir::Type resultTy = llvmBaseObjectType;
     // Fortran is column major, llvm GEP is row major: reverse the indices here.
     for (mlir::Value interiorIndex : llvm::reverse(cstInteriorIndices)) {
-      auto arrayTy = resultTy.dyn_cast<mlir::LLVM::LLVMArrayType>();
+      auto arrayTy = mlir::dyn_cast<mlir::LLVM::LLVMArrayType>(resultTy);
       if (!arrayTy)
         fir::emitFatalError(
             loc,
@@ -1355,7 +1362,7 @@ struct EmboxCommonConversion : public fir::FIROpConversion<OP> {
         convertSubcomponentIndices(loc, resultTy, componentIndices, &resultTy);
     gepArgs.append(gepIndices.begin(), gepIndices.end());
     if (substringOffset) {
-      if (auto arrayTy = resultTy.dyn_cast<mlir::LLVM::LLVMArrayType>()) {
+      if (auto arrayTy = mlir::dyn_cast<mlir::LLVM::LLVMArrayType>(resultTy)) {
         gepArgs.push_back(*substringOffset);
         resultTy = arrayTy.getElementType();
       } else {
@@ -1504,18 +1511,18 @@ struct XEmboxOpConversion : public EmboxCommonConversion<fir::cg::XEmboxOp> {
     unsigned constRows = 0;
     mlir::Value ptrOffset = zero;
     mlir::Type memEleTy = fir::dyn_cast_ptrEleTy(xbox.getMemref().getType());
-    assert(memEleTy.isa<fir::SequenceType>());
-    auto seqTy = memEleTy.cast<fir::SequenceType>();
+    assert(mlir::isa<fir::SequenceType>(memEleTy));
+    auto seqTy = mlir::cast<fir::SequenceType>(memEleTy);
     mlir::Type seqEleTy = seqTy.getEleTy();
     // Adjust the element scaling factor if the element is a dependent type.
     if (fir::hasDynamicSize(seqEleTy)) {
-      if (auto charTy = seqEleTy.dyn_cast<fir::CharacterType>()) {
+      if (auto charTy = mlir::dyn_cast<fir::CharacterType>(seqEleTy)) {
         // The GEP pointer type decays to llvm.ptr<i[width]>.
         // The scaling factor is the runtime value of the length.
         assert(!adaptor.getLenParams().empty());
         prevPtrOff = FIROpConversion::integerCast(
             loc, rewriter, i64Ty, adaptor.getLenParams().back());
-      } else if (seqEleTy.isa<fir::RecordType>()) {
+      } else if (mlir::isa<fir::RecordType>(seqEleTy)) {
         // prevPtrOff = ;
         TODO(loc, "generate call to calculate size of PDT");
       } else {
@@ -1540,7 +1547,7 @@ struct XEmboxOpConversion : public EmboxCommonConversion<fir::cg::XEmboxOp> {
     } else if (hasSubstr) {
       // We have a substring. The step value needs to be the number of bytes
       // per CHARACTER element.
-      auto charTy = seqEleTy.cast<fir::CharacterType>();
+      auto charTy = mlir::cast<fir::CharacterType>(seqEleTy);
       if (fir::hasDynamicSize(charTy)) {
         prevDimByteStride =
             getCharacterByteSize(loc, rewriter, charTy, adaptor.getLenParams());
@@ -1589,7 +1596,7 @@ struct XEmboxOpConversion : public EmboxCommonConversion<fir::cg::XEmboxOp> {
         // Lower bound is normalized to 0 for BIND(C) interoperability.
         mlir::Value lb = zero;
         const bool isaPointerOrAllocatable =
-            eleTy.isa<fir::PointerType>() || eleTy.isa<fir::HeapType>();
+            mlir::isa<fir::PointerType, fir::HeapType>(eleTy);
         // Lower bound is defaults to 1 for POINTER, ALLOCATABLE, and
         // denormalized descriptors.
         if (isaPointerOrAllocatable || !normalizedLowerBound(xbox))
@@ -1695,7 +1702,7 @@ struct XReboxOpConversion : public EmboxCommonConversion<fir::cg::XReboxOp> {
     // Create new descriptor and fill its non-shape related data.
     llvm::SmallVector<mlir::Value, 2> lenParams;
     mlir::Type inputEleTy = getInputEleTy(rebox);
-    if (auto charTy = inputEleTy.dyn_cast<fir::CharacterType>()) {
+    if (auto charTy = mlir::dyn_cast<fir::CharacterType>(inputEleTy)) {
       if (charTy.hasConstantLen()) {
         mlir::Value len =
             genConstantIndex(loc, idxTy, rewriter, charTy.getLen());
@@ -1712,15 +1719,15 @@ struct XReboxOpConversion : public EmboxCommonConversion<fir::cg::XReboxOp> {
         }
         lenParams.emplace_back(len);
       }
-    } else if (auto recTy = inputEleTy.dyn_cast<fir::RecordType>()) {
+    } else if (auto recTy = mlir::dyn_cast<fir::RecordType>(inputEleTy)) {
       if (recTy.getNumLenParams() != 0)
         TODO(loc, "reboxing descriptor of derived type with length parameters");
     }
 
     // Rebox on polymorphic entities needs to carry over the dynamic type.
     mlir::Value typeDescAddr;
-    if (inputBoxTyPair.fir.isa<fir::ClassType>() &&
-        rebox.getType().isa<fir::ClassType>())
+    if (mlir::isa<fir::ClassType>(inputBoxTyPair.fir) &&
+        mlir::isa<fir::ClassType>(rebox.getType()))
       typeDescAddr =
           loadTypeDescAddress(loc, inputBoxTyPair, loweredBox, rewriter);
 
@@ -1908,7 +1915,7 @@ private:
   /// Return scalar element type of the input box.
   static mlir::Type getInputEleTy(fir::cg::XReboxOp rebox) {
     auto ty = fir::dyn_cast_ptrOrBoxEleTy(rebox.getBox().getType());
-    if (auto seqTy = ty.dyn_cast<fir::SequenceType>())
+    if (auto seqTy = mlir::dyn_cast<fir::SequenceType>(ty))
       return seqTy.getEleTy();
     return ty;
   }
@@ -1936,7 +1943,7 @@ struct ValueOpCommon {
     assert(ty && "type is null");
     const auto end = indices.size();
     for (std::remove_const_t<decltype(end)> i = 0; i < end; ++i) {
-      if (auto seq = ty.dyn_cast<mlir::LLVM::LLVMArrayType>()) {
+      if (auto seq = mlir::dyn_cast<mlir::LLVM::LLVMArrayType>(ty)) {
         const auto dim = getDimension(seq);
         if (dim > 1) {
           auto ub = std::min(i + dim, end);
@@ -1944,7 +1951,7 @@ struct ValueOpCommon {
           i += dim - 1;
         }
         ty = getArrayElementType(seq);
-      } else if (auto st = ty.dyn_cast<mlir::LLVM::LLVMStructType>()) {
+      } else if (auto st = mlir::dyn_cast<mlir::LLVM::LLVMStructType>(ty)) {
         ty = st.getBody()[indices[i]];
       } else {
         llvm_unreachable("index into invalid type");
@@ -1957,13 +1964,13 @@ struct ValueOpCommon {
                  mlir::ArrayAttr arrAttr) {
     llvm::SmallVector<int64_t> indices;
     for (auto i = arrAttr.begin(), e = arrAttr.end(); i != e; ++i) {
-      if (auto intAttr = i->dyn_cast<mlir::IntegerAttr>()) {
+      if (auto intAttr = mlir::dyn_cast<mlir::IntegerAttr>(*i)) {
         indices.push_back(intAttr.getInt());
       } else {
-        auto fieldName = i->cast<mlir::StringAttr>().getValue();
+        auto fieldName = mlir::cast<mlir::StringAttr>(*i).getValue();
         ++i;
-        auto ty = i->cast<mlir::TypeAttr>().getValue();
-        auto index = ty.cast<fir::RecordType>().getFieldIndex(fieldName);
+        auto ty = mlir::cast<mlir::TypeAttr>(*i).getValue();
+        auto index = mlir::cast<fir::RecordType>(ty).getFieldIndex(fieldName);
         indices.push_back(index);
       }
     }
@@ -1973,7 +1980,7 @@ struct ValueOpCommon {
 private:
   static mlir::Type getArrayElementType(mlir::LLVM::LLVMArrayType ty) {
     auto eleTy = ty.getElementType();
-    while (auto arrTy = eleTy.dyn_cast<mlir::LLVM::LLVMArrayType>())
+    while (auto arrTy = mlir::dyn_cast<mlir::LLVM::LLVMArrayType>(eleTy))
       eleTy = arrTy.getElementType();
     return eleTy;
   }
@@ -2041,7 +2048,7 @@ struct InsertOnRangeOpConversion
     auto type = adaptor.getOperands()[0].getType();
 
     // Iteratively extract the array dimensions from the type.
-    while (auto t = type.dyn_cast<mlir::LLVM::LLVMArrayType>()) {
+    while (auto t = mlir::dyn_cast<mlir::LLVM::LLVMArrayType>(type)) {
       dims.push_back(t.getNumElements());
       type = t.getElementType();
     }
@@ -2107,7 +2114,8 @@ struct XArrayCoorOpConversion
     mlir::Value offset = genConstantIndex(loc, idxTy, rewriter, 0);
     const bool isShifted = !coor.getShift().empty();
     const bool isSliced = !coor.getSlice().empty();
-    const bool baseIsBoxed = coor.getMemref().getType().isa<fir::BaseBoxType>();
+    const bool baseIsBoxed =
+        mlir::isa<fir::BaseBoxType>(coor.getMemref().getType());
     TypePair baseBoxTyPair =
         baseIsBoxed ? getBoxTypePair(coor.getMemref().getType()) : TypePair{};
     mlir::LLVM::IntegerOverflowFlags nsw =
@@ -2185,7 +2193,8 @@ struct XArrayCoorOpConversion
       // components.
       mlir::Type elementType =
           getLlvmObjectTypeFromBoxType(coor.getMemref().getType());
-      while (auto arrayTy = elementType.dyn_cast<mlir::LLVM::LLVMArrayType>())
+      while (auto arrayTy =
+                 mlir::dyn_cast<mlir::LLVM::LLVMArrayType>(elementType))
         elementType = arrayTy.getElementType();
       args.clear();
       args.push_back(0);
@@ -2275,11 +2284,12 @@ struct CoordinateOpConversion
     }
 
     // Boxed type - get the base pointer from the box
-    if (baseObjectTy.dyn_cast<fir::BaseBoxType>())
+    if (mlir::dyn_cast<fir::BaseBoxType>(baseObjectTy))
       return doRewriteBox(coor, operands, loc, rewriter);
 
     // Reference, pointer or a heap type
-    if (baseObjectTy.isa<fir::ReferenceType, fir::PointerType, fir::HeapType>())
+    if (mlir::isa<fir::ReferenceType, fir::PointerType, fir::HeapType>(
+            baseObjectTy))
       return doRewriteRefOrPtr(coor, llvmObjectTy, operands, loc, rewriter);
 
     return rewriter.notifyMatchFailure(
@@ -2295,7 +2305,7 @@ struct CoordinateOpConversion
   }
 
   static bool hasSubDimensions(mlir::Type type) {
-    return type.isa<fir::SequenceType, fir::RecordType, mlir::TupleType>();
+    return mlir::isa<fir::SequenceType, fir::RecordType, mlir::TupleType>(type);
   }
 
   /// Check whether this form of `!fir.coordinate_of` is supported. These
@@ -2310,14 +2320,14 @@ struct CoordinateOpConversion
     bool ptrEle = false;
     for (; i < numOfCoors; ++i) {
       mlir::Value nxtOpnd = coors[i];
-      if (auto arrTy = type.dyn_cast<fir::SequenceType>()) {
+      if (auto arrTy = mlir::dyn_cast<fir::SequenceType>(type)) {
         subEle = true;
         i += arrTy.getDimension() - 1;
         type = arrTy.getEleTy();
-      } else if (auto recTy = type.dyn_cast<fir::RecordType>()) {
+      } else if (auto recTy = mlir::dyn_cast<fir::RecordType>(type)) {
         subEle = true;
         type = recTy.getType(getFieldNumber(recTy, nxtOpnd));
-      } else if (auto tupTy = type.dyn_cast<mlir::TupleType>()) {
+      } else if (auto tupTy = mlir::dyn_cast<mlir::TupleType>(type)) {
         subEle = true;
         type = tupTy.getType(getConstantIntValue(nxtOpnd));
       } else {
@@ -2335,14 +2345,14 @@ struct CoordinateOpConversion
   static bool arraysHaveKnownShape(mlir::Type type, mlir::ValueRange coors) {
     for (std::size_t i = 0, sz = coors.size(); i < sz; ++i) {
       mlir::Value nxtOpnd = coors[i];
-      if (auto arrTy = type.dyn_cast<fir::SequenceType>()) {
+      if (auto arrTy = mlir::dyn_cast<fir::SequenceType>(type)) {
         if (fir::sequenceWithNonConstantShape(arrTy))
           return false;
         i += arrTy.getDimension() - 1;
         type = arrTy.getEleTy();
-      } else if (auto strTy = type.dyn_cast<fir::RecordType>()) {
+      } else if (auto strTy = mlir::dyn_cast<fir::RecordType>(type)) {
         type = strTy.getType(getFieldNumber(strTy, nxtOpnd));
-      } else if (auto strTy = type.dyn_cast<mlir::TupleType>()) {
+      } else if (auto strTy = mlir::dyn_cast<mlir::TupleType>(type)) {
         type = strTy.getType(getConstantIntValue(nxtOpnd));
       } else {
         return true;
@@ -2357,7 +2367,8 @@ private:
                mlir::Location loc,
                mlir::ConversionPatternRewriter &rewriter) const {
     mlir::Type boxObjTy = coor.getBaseType();
-    assert(boxObjTy.dyn_cast<fir::BaseBoxType>() && "This is not a `fir.box`");
+    assert(mlir::dyn_cast<fir::BaseBoxType>(boxObjTy) &&
+           "This is not a `fir.box`");
     TypePair boxTyPair = getBoxTypePair(boxObjTy);
 
     mlir::Value boxBaseAddr = operands[0];
@@ -2399,7 +2410,7 @@ private:
         mlir::LLVM::IntegerOverflowFlags::nsw;
 
     for (unsigned i = 1, last = operands.size(); i < last; ++i) {
-      if (auto arrTy = cpnTy.dyn_cast<fir::SequenceType>()) {
+      if (auto arrTy = mlir::dyn_cast<fir::SequenceType>(cpnTy)) {
         if (i != 1)
           TODO(loc, "fir.array nested inside other array and/or derived type");
         // Applies byte strides from the box. Ignore lower bound from box
@@ -2421,7 +2432,7 @@ private:
             llvm::ArrayRef<mlir::LLVM::GEPArg>{off});
         i += arrTy.getDimension() - 1;
         cpnTy = arrTy.getEleTy();
-      } else if (auto recTy = cpnTy.dyn_cast<fir::RecordType>()) {
+      } else if (auto recTy = mlir::dyn_cast<fir::RecordType>(cpnTy)) {
         mlir::Value nxtOpnd = operands[i];
         cpnTy = recTy.getType(getFieldNumber(recTy, nxtOpnd));
         auto llvmRecTy = lowerTy().convertType(recTy);
@@ -2456,7 +2467,7 @@ private:
 
     // If only the column is `?`, then we can simply place the column value in
     // the 0-th GEP position.
-    if (auto arrTy = cpnTy.dyn_cast<fir::SequenceType>()) {
+    if (auto arrTy = mlir::dyn_cast<fir::SequenceType>(cpnTy)) {
       if (!hasKnownShape) {
         const unsigned sz = arrTy.getDimension();
         if (arraysHaveKnownShape(arrTy.getEleTy(),
@@ -2500,29 +2511,29 @@ private:
             dims = dimsLeft - 1;
             continue;
           }
-          cpnTy = cpnTy.cast<fir::SequenceType>().getEleTy();
+          cpnTy = mlir::cast<fir::SequenceType>(cpnTy).getEleTy();
           // append array range in reverse (FIR arrays are column-major)
           offs.append(arrIdx.rbegin(), arrIdx.rend());
           arrIdx.clear();
           dims.reset();
           continue;
         }
-        if (auto arrTy = cpnTy.dyn_cast<fir::SequenceType>()) {
+        if (auto arrTy = mlir::dyn_cast<fir::SequenceType>(cpnTy)) {
           int d = arrTy.getDimension() - 1;
           if (d > 0) {
             dims = d;
             arrIdx.push_back(nxtOpnd);
             continue;
           }
-          cpnTy = cpnTy.cast<fir::SequenceType>().getEleTy();
+          cpnTy = mlir::cast<fir::SequenceType>(cpnTy).getEleTy();
           offs.push_back(nxtOpnd);
           continue;
         }
 
         // check if the i-th coordinate relates to a field
-        if (auto recTy = cpnTy.dyn_cast<fir::RecordType>())
+        if (auto recTy = mlir::dyn_cast<fir::RecordType>(cpnTy))
           cpnTy = recTy.getType(getFieldNumber(recTy, nxtOpnd));
-        else if (auto tupTy = cpnTy.dyn_cast<mlir::TupleType>())
+        else if (auto tupTy = mlir::dyn_cast<mlir::TupleType>(cpnTy))
           cpnTy = tupTy.getType(getConstantIntValue(nxtOpnd));
         else
           cpnTy = nullptr;
@@ -2551,7 +2562,7 @@ struct FieldIndexOpConversion : public fir::FIROpConversion<fir::FieldIndexOp> {
   mlir::LogicalResult
   matchAndRewrite(fir::FieldIndexOp field, OpAdaptor adaptor,
                   mlir::ConversionPatternRewriter &rewriter) const override {
-    auto recTy = field.getOnType().cast<fir::RecordType>();
+    auto recTy = mlir::cast<fir::RecordType>(field.getOnType());
     unsigned index = recTy.getFieldIndex(field.getFieldId());
 
     if (!fir::hasDynamicSize(recTy)) {
@@ -2604,8 +2615,8 @@ struct TypeDescOpConversion : public fir::FIROpConversion<fir::TypeDescOp> {
   matchAndRewrite(fir::TypeDescOp typeDescOp, OpAdaptor adaptor,
                   mlir::ConversionPatternRewriter &rewriter) const override {
     mlir::Type inTy = typeDescOp.getInType();
-    assert(inTy.isa<fir::RecordType>() && "expecting fir.type");
-    auto recordType = inTy.dyn_cast<fir::RecordType>();
+    assert(mlir::isa<fir::RecordType>(inTy) && "expecting fir.type");
+    auto recordType = mlir::dyn_cast<fir::RecordType>(inTy);
     auto module = typeDescOp.getOperation()->getParentOfType<mlir::ModuleOp>();
     std::string typeDescName =
         fir::NameUniquer::getTypeDescriptorName(recordType.getName());
@@ -2732,7 +2743,7 @@ struct GlobalOpConversion : public fir::FIROpConversion<fir::GlobalOp> {
           mlir::Type vecType = mlir::VectorType::get(
               insertOp.getType().getShape(), constant.getType());
           auto denseAttr = mlir::DenseElementsAttr::get(
-              vecType.cast<mlir::ShapedType>(), constant.getValue());
+              mlir::cast<mlir::ShapedType>(vecType), constant.getValue());
           rewriter.setInsertionPointAfter(insertOp);
           rewriter.replaceOpWithNewOp<mlir::arith::ConstantOp>(
               insertOp, seqTyAttr, denseAttr);
@@ -2808,7 +2819,7 @@ struct LoadOpConversion : public fir::FIROpConversion<fir::LoadOp> {
   matchAndRewrite(fir::LoadOp load, OpAdaptor adaptor,
                   mlir::ConversionPatternRewriter &rewriter) const override {
     mlir::Type llvmLoadTy = convertObjectType(load.getType());
-    if (auto boxTy = load.getType().dyn_cast<fir::BaseBoxType>()) {
+    if (auto boxTy = mlir::dyn_cast<fir::BaseBoxType>(load.getType())) {
       // fir.box is a special case because it is considered as an ssa values in
       // fir, but it is lowered as a pointer to a descriptor. So
       // fir.ref<fir.box> and fir.box end up being the same llvm types and
@@ -2921,7 +2932,7 @@ struct SelectCaseOpConversion : public fir::FIROpConversion<fir::SelectCaseOp> {
     llvm::ArrayRef<mlir::Attribute> cases = caseOp.getCases().getValue();
     // Type can be CHARACTER, INTEGER, or LOGICAL (C1145)
     auto ty = caseOp.getSelector().getType();
-    if (ty.isa<fir::CharacterType>()) {
+    if (mlir::isa<fir::CharacterType>(ty)) {
       TODO(caseOp.getLoc(), "fir.select_case codegen with character type");
       return mlir::failure();
     }
@@ -2935,25 +2946,25 @@ struct SelectCaseOpConversion : public fir::FIROpConversion<fir::SelectCaseOp> {
           *caseOp.getCompareOperands(adaptor.getOperands(), t);
       mlir::Value caseArg = *(cmpOps.value().begin());
       mlir::Attribute attr = cases[t];
-      if (attr.isa<fir::PointIntervalAttr>()) {
+      if (mlir::isa<fir::PointIntervalAttr>(attr)) {
         auto cmp = rewriter.create<mlir::LLVM::ICmpOp>(
             loc, mlir::LLVM::ICmpPredicate::eq, selector, caseArg);
         genCaseLadderStep(loc, cmp, dest, destOps, rewriter);
         continue;
       }
-      if (attr.isa<fir::LowerBoundAttr>()) {
+      if (mlir::isa<fir::LowerBoundAttr>(attr)) {
         auto cmp = rewriter.create<mlir::LLVM::ICmpOp>(
             loc, mlir::LLVM::ICmpPredicate::sle, caseArg, selector);
         genCaseLadderStep(loc, cmp, dest, destOps, rewriter);
         continue;
       }
-      if (attr.isa<fir::UpperBoundAttr>()) {
+      if (mlir::isa<fir::UpperBoundAttr>(attr)) {
         auto cmp = rewriter.create<mlir::LLVM::ICmpOp>(
             loc, mlir::LLVM::ICmpPredicate::sle, selector, caseArg);
         genCaseLadderStep(loc, cmp, dest, destOps, rewriter);
         continue;
       }
-      if (attr.isa<fir::ClosedIntervalAttr>()) {
+      if (mlir::isa<fir::ClosedIntervalAttr>(attr)) {
         auto cmp = rewriter.create<mlir::LLVM::ICmpOp>(
             loc, mlir::LLVM::ICmpPredicate::sle, caseArg, selector);
         auto *thisBlock = rewriter.getInsertionBlock();
@@ -2969,7 +2980,7 @@ struct SelectCaseOpConversion : public fir::FIROpConversion<fir::SelectCaseOp> {
         rewriter.setInsertionPointToEnd(newBlock2);
         continue;
       }
-      assert(attr.isa<mlir::UnitAttr>());
+      assert(mlir::isa<mlir::UnitAttr>(attr));
       assert((t + 1 == conds) && "unit must be last");
       genBrOp(caseOp, dest, destOps, rewriter);
     }
@@ -2997,13 +3008,13 @@ static void selectMatchAndRewrite(const fir::LLVMTypeConverter &lowering,
     mlir::Block *dest = select.getSuccessor(t);
     auto destOps = select.getSuccessorOperands(adaptor.getOperands(), t);
     const mlir::Attribute &attr = cases[t];
-    if (auto intAttr = attr.template dyn_cast<mlir::IntegerAttr>()) {
+    if (auto intAttr = mlir::dyn_cast<mlir::IntegerAttr>(attr)) {
       destinations.push_back(dest);
       destinationsOperands.push_back(destOps ? *destOps : mlir::ValueRange{});
       caseValues.push_back(intAttr.getInt());
       continue;
     }
-    assert(attr.template dyn_cast_or_null<mlir::UnitAttr>());
+    assert(mlir::dyn_cast_or_null<mlir::UnitAttr>(attr));
     assert((t + 1 == conds) && "unit must be last");
     defaultDestination = dest;
     defaultOperands = destOps ? *destOps : mlir::ValueRange{};
@@ -3071,7 +3082,7 @@ struct StoreOpConversion : public fir::FIROpConversion<fir::StoreOp> {
     mlir::Location loc = store.getLoc();
     mlir::Type storeTy = store.getValue().getType();
     mlir::LLVM::StoreOp newStoreOp;
-    if (auto boxTy = storeTy.dyn_cast<fir::BaseBoxType>()) {
+    if (auto boxTy = mlir::dyn_cast<fir::BaseBoxType>(storeTy)) {
       // fir.box value is actually in memory, load it first before storing it.
       mlir::Type llvmBoxTy = lowerTy().convertBoxTypeAsStruct(boxTy);
       auto val = rewriter.create<mlir::LLVM::LoadOp>(loc, llvmBoxTy,
@@ -3186,9 +3197,9 @@ struct IsPresentOpConversion : public fir::FIROpConversion<fir::IsPresentOp> {
     mlir::Location loc = isPresent.getLoc();
     auto ptr = adaptor.getOperands()[0];
 
-    if (isPresent.getVal().getType().isa<fir::BoxCharType>()) {
+    if (mlir::isa<fir::BoxCharType>(isPresent.getVal().getType())) {
       [[maybe_unused]] auto structTy =
-          ptr.getType().cast<mlir::LLVM::LLVMStructType>();
+          mlir::cast<mlir::LLVM::LLVMStructType>(ptr.getType());
       assert(!structTy.isOpaque() && !structTy.getBody().empty());
 
       ptr = rewriter.create<mlir::LLVM::ExtractValueOp>(loc, ptr, 0);
@@ -3214,8 +3225,8 @@ struct AbsentOpConversion : public fir::FIROpConversion<fir::AbsentOp> {
     mlir::Type ty = convertType(absent.getType());
     mlir::Location loc = absent.getLoc();
 
-    if (absent.getType().isa<fir::BoxCharType>()) {
-      auto structTy = ty.cast<mlir::LLVM::LLVMStructType>();
+    if (mlir::isa<fir::BoxCharType>(absent.getType())) {
+      auto structTy = mlir::cast<mlir::LLVM::LLVMStructType>(ty);
       assert(!structTy.isOpaque() && !structTy.getBody().empty());
       auto undefStruct = rewriter.create<mlir::LLVM::UndefOp>(loc, ty);
       auto nullField =
diff --git a/flang/lib/Optimizer/CodeGen/FIROpPatterns.cpp b/flang/lib/Optimizer/CodeGen/FIROpPatterns.cpp
index 26871d888815..d6dac4998fdc 100644
--- a/flang/lib/Optimizer/CodeGen/FIROpPatterns.cpp
+++ b/flang/lib/Optimizer/CodeGen/FIROpPatterns.cpp
@@ -20,7 +20,7 @@ static inline mlir::Type getLlvmPtrType(mlir::MLIRContext *context,
 }
 
 static unsigned getTypeDescFieldId(mlir::Type ty) {
-  auto isArray = fir::dyn_cast_ptrOrBoxEleTy(ty).isa<fir::SequenceType>();
+  auto isArray = mlir::isa<fir::SequenceType>(fir::dyn_cast_ptrOrBoxEleTy(ty));
   return isArray ? kOptTypePtrPosInBox : kDimsPosInBox;
 }
 
@@ -37,7 +37,7 @@ ConvertFIRToLLVMPattern::ConvertFIRToLLVMPattern(
 // reference.
 mlir::Type
 ConvertFIRToLLVMPattern::convertObjectType(mlir::Type firType) const {
-  if (auto boxTy = firType.dyn_cast<fir::BaseBoxType>())
+  if (auto boxTy = mlir::dyn_cast<fir::BaseBoxType>(firType))
     return lowerTy().convertBoxTypeAsStruct(boxTy);
   return lowerTy().convertType(firType);
 }
@@ -69,7 +69,7 @@ ConvertFIRToLLVMPattern::integerCast(mlir::Location loc,
   auto valTy = val.getType();
   // If the value was not yet lowered, lower its type so that it can
   // be used in getPrimitiveTypeSizeInBits.
-  if (!valTy.isa<mlir::IntegerType>())
+  if (!mlir::isa<mlir::IntegerType>(valTy))
     valTy = convertType(valTy);
   auto toSize = mlir::LLVM::getPrimitiveTypeSizeInBits(ty);
   auto fromSize = mlir::LLVM::getPrimitiveTypeSizeInBits(valTy);
@@ -91,7 +91,7 @@ ConvertFIRToLLVMPattern::getBoxTypePair(mlir::Type firBoxTy) const {
 mlir::Value ConvertFIRToLLVMPattern::getValueFromBox(
     mlir::Location loc, TypePair boxTy, mlir::Value box, mlir::Type resultTy,
     mlir::ConversionPatternRewriter &rewriter, int boxValue) const {
-  if (box.getType().isa<mlir::LLVM::LLVMPointerType>()) {
+  if (mlir::isa<mlir::LLVM::LLVMPointerType>(box.getType())) {
     auto pty = getLlvmPtrType(resultTy.getContext());
     auto p = rewriter.create<mlir::LLVM::GEPOp>(
         loc, pty, boxTy.llvm, box,
@@ -133,7 +133,7 @@ llvm::SmallVector<mlir::Value, 3> ConvertFIRToLLVMPattern::getDimsFromBox(
 mlir::Value ConvertFIRToLLVMPattern::loadDimFieldFromBox(
     mlir::Location loc, TypePair boxTy, mlir::Value box, mlir::Value dim,
     int off, mlir::Type ty, mlir::ConversionPatternRewriter &rewriter) const {
-  assert(box.getType().isa<mlir::LLVM::LLVMPointerType>() &&
+  assert(mlir::isa<mlir::LLVM::LLVMPointerType>(box.getType()) &&
          "descriptor inquiry with runtime dim can only be done on descriptor "
          "in memory");
   mlir::LLVM::GEPOp p = genGEP(loc, boxTy.llvm, rewriter, box, 0,
@@ -146,7 +146,7 @@ mlir::Value ConvertFIRToLLVMPattern::loadDimFieldFromBox(
 mlir::Value ConvertFIRToLLVMPattern::getDimFieldFromBox(
     mlir::Location loc, TypePair boxTy, mlir::Value box, int dim, int off,
     mlir::Type ty, mlir::ConversionPatternRewriter &rewriter) const {
-  if (box.getType().isa<mlir::LLVM::LLVMPointerType>()) {
+  if (mlir::isa<mlir::LLVM::LLVMPointerType>(box.getType())) {
     mlir::LLVM::GEPOp p = genGEP(loc, boxTy.llvm, rewriter, box, 0,
                                  static_cast<int>(kDimsPosInBox), dim, off);
     auto loadOp = rewriter.create<mlir::LLVM::LoadOp>(loc, ty, p);
@@ -184,12 +184,12 @@ mlir::Value ConvertFIRToLLVMPattern::getElementSizeFromBox(
 mlir::Type ConvertFIRToLLVMPattern::getBoxEleTy(
     mlir::Type type, llvm::ArrayRef<std::int64_t> indexes) const {
   for (unsigned i : indexes) {
-    if (auto t = type.dyn_cast<mlir::LLVM::LLVMStructType>()) {
+    if (auto t = mlir::dyn_cast<mlir::LLVM::LLVMStructType>(type)) {
       assert(!t.isOpaque() && i < t.getBody().size());
       type = t.getBody()[i];
-    } else if (auto t = type.dyn_cast<mlir::LLVM::LLVMArrayType>()) {
+    } else if (auto t = mlir::dyn_cast<mlir::LLVM::LLVMArrayType>(type)) {
       type = t.getElementType();
-    } else if (auto t = type.dyn_cast<mlir::VectorType>()) {
+    } else if (auto t = mlir::dyn_cast<mlir::VectorType>(type)) {
       type = t.getElementType();
     } else {
       fir::emitFatalError(mlir::UnknownLoc::get(type.getContext()),
@@ -243,6 +243,9 @@ ConvertFIRToLLVMPattern::getBlockForAllocaInsert(mlir::Operation *op) const {
     return iface.getAllocaBlock();
   if (auto llvmFuncOp = mlir::dyn_cast<mlir::LLVM::LLVMFuncOp>(op))
     return &llvmFuncOp.front();
+  if (auto ompPrivateOp = mlir::dyn_cast<mlir::omp::PrivateClauseOp>(op))
+    return &ompPrivateOp.getAllocRegion().front();
+
   return getBlockForAllocaInsert(op->getParentOp());
 }
 
diff --git a/flang/lib/Optimizer/CodeGen/PreCGRewrite.cpp b/flang/lib/Optimizer/CodeGen/PreCGRewrite.cpp
index 665bf09b8fc3..ce7ee22d5d77 100644
--- a/flang/lib/Optimizer/CodeGen/PreCGRewrite.cpp
+++ b/flang/lib/Optimizer/CodeGen/PreCGRewrite.cpp
@@ -86,10 +86,10 @@ public:
     // If the embox does not include a shape, then do not convert it
     if (auto shapeVal = embox.getShape())
       return rewriteDynamicShape(embox, rewriter, shapeVal);
-    if (embox.getType().isa<fir::ClassType>())
+    if (mlir::isa<fir::ClassType>(embox.getType()))
       TODO(embox.getLoc(), "embox conversion for fir.class type");
-    if (auto boxTy = embox.getType().dyn_cast<fir::BoxType>())
-      if (auto seqTy = boxTy.getEleTy().dyn_cast<fir::SequenceType>())
+    if (auto boxTy = mlir::dyn_cast<fir::BoxType>(embox.getType()))
+      if (auto seqTy = mlir::dyn_cast<fir::SequenceType>(boxTy.getEleTy()))
         if (!seqTy.hasDynamicExtents())
           return rewriteStaticShape(embox, rewriter, seqTy);
     return mlir::failure();
@@ -294,10 +294,9 @@ public:
     target.addIllegalOp<fir::ReboxOp>();
     target.addIllegalOp<fir::DeclareOp>();
     target.addDynamicallyLegalOp<fir::EmboxOp>([](fir::EmboxOp embox) {
-      return !(embox.getShape() || embox.getType()
-                                       .cast<fir::BaseBoxType>()
-                                       .getEleTy()
-                                       .isa<fir::SequenceType>());
+      return !(embox.getShape() ||
+               mlir::isa<fir::SequenceType>(
+                   mlir::cast<fir::BaseBoxType>(embox.getType()).getEleTy()));
     });
     mlir::RewritePatternSet patterns(&context);
     fir::populatePreCGRewritePatterns(patterns);
diff --git a/flang/lib/Optimizer/CodeGen/TBAABuilder.cpp b/flang/lib/Optimizer/CodeGen/TBAABuilder.cpp
index b1b0e9b766a6..a21384e8d594 100644
--- a/flang/lib/Optimizer/CodeGen/TBAABuilder.cpp
+++ b/flang/lib/Optimizer/CodeGen/TBAABuilder.cpp
@@ -120,7 +120,7 @@ void TBAABuilder::attachTBAATag(AliasAnalysisOpInterface op, Type baseFIRType,
     // with both data and descriptor accesses.
     // Conservatively set any-access tag if there is any descriptor member.
     tbaaTagSym = getAnyAccessTag(func);
-  } else if (baseFIRType.isa<fir::BaseBoxType>()) {
+  } else if (mlir::isa<fir::BaseBoxType>(baseFIRType)) {
     tbaaTagSym = getBoxAccessTag(baseFIRType, accessFIRType, gep, func);
   } else {
     tbaaTagSym = getDataAccessTag(baseFIRType, accessFIRType, gep, func);
diff --git a/flang/lib/Optimizer/CodeGen/Target.cpp b/flang/lib/Optimizer/CodeGen/Target.cpp
index cea7a1f97f41..652e2bddc1b8 100644
--- a/flang/lib/Optimizer/CodeGen/Target.cpp
+++ b/flang/lib/Optimizer/CodeGen/Target.cpp
@@ -41,9 +41,9 @@ llvm::StringRef Attributes::getIntExtensionAttrName() const {
 static const llvm::fltSemantics &floatToSemantics(const KindMapping &kindMap,
                                                   mlir::Type type) {
   assert(isa_real(type));
-  if (auto ty = type.dyn_cast<fir::RealType>())
+  if (auto ty = mlir::dyn_cast<fir::RealType>(type))
     return kindMap.getFloatSemantics(ty.getFKind());
-  return type.cast<mlir::FloatType>().getFloatSemantics();
+  return mlir::cast<mlir::FloatType>(type).getFloatSemantics();
 }
 
 static void typeTodo(const llvm::fltSemantics *sem, mlir::Location loc,
diff --git a/flang/lib/Optimizer/CodeGen/TargetRewrite.cpp b/flang/lib/Optimizer/CodeGen/TargetRewrite.cpp
index 7bf31ec38695..616de78d0026 100644
--- a/flang/lib/Optimizer/CodeGen/TargetRewrite.cpp
+++ b/flang/lib/Optimizer/CodeGen/TargetRewrite.cpp
@@ -137,7 +137,7 @@ public:
         if (!hasPortableSignature(dispatch.getFunctionType(), op))
           convertCallOp(dispatch);
       } else if (auto addr = mlir::dyn_cast<fir::AddrOfOp>(op)) {
-        if (addr.getType().isa<mlir::FunctionType>() &&
+        if (mlir::isa<mlir::FunctionType>(addr.getType()) &&
             !hasPortableSignature(addr.getType(), op))
           convertAddrOp(addr);
       }
@@ -601,7 +601,7 @@ public:
   /// Taking the address of a function. Modify the signature as needed.
   void convertAddrOp(fir::AddrOfOp addrOp) {
     rewriter->setInsertionPoint(addrOp);
-    auto addrTy = addrOp.getType().cast<mlir::FunctionType>();
+    auto addrTy = mlir::cast<mlir::FunctionType>(addrOp.getType());
     fir::CodeGenSpecifics::Marshalling newInTyAndAttrs;
     llvm::SmallVector<mlir::Type> newResTys;
     auto loc = addrOp.getLoc();
@@ -705,22 +705,23 @@ public:
   /// return `true`. Otherwise, the signature is not portable and `false` is
   /// returned.
   bool hasPortableSignature(mlir::Type signature, mlir::Operation *op) {
-    assert(signature.isa<mlir::FunctionType>());
-    auto func = signature.dyn_cast<mlir::FunctionType>();
+    assert(mlir::isa<mlir::FunctionType>(signature));
+    auto func = mlir::dyn_cast<mlir::FunctionType>(signature);
     bool hasCCallingConv = isFuncWithCCallingConvention(op);
     for (auto ty : func.getResults())
-      if ((ty.isa<fir::BoxCharType>() && !noCharacterConversion) ||
+      if ((mlir::isa<fir::BoxCharType>(ty) && !noCharacterConversion) ||
           (fir::isa_complex(ty) && !noComplexConversion) ||
-          (ty.isa<mlir::IntegerType>() && hasCCallingConv)) {
+          (mlir::isa<mlir::IntegerType>(ty) && hasCCallingConv)) {
         LLVM_DEBUG(llvm::dbgs() << "rewrite " << signature << " for target\n");
         return false;
       }
     for (auto ty : func.getInputs())
-      if (((ty.isa<fir::BoxCharType>() || fir::isCharacterProcedureTuple(ty)) &&
+      if (((mlir::isa<fir::BoxCharType>(ty) ||
+            fir::isCharacterProcedureTuple(ty)) &&
            !noCharacterConversion) ||
           (fir::isa_complex(ty) && !noComplexConversion) ||
-          (ty.isa<mlir::IntegerType>() && hasCCallingConv) ||
-          (ty.isa<fir::RecordType>() && !noStructConversion)) {
+          (mlir::isa<mlir::IntegerType>(ty) && hasCCallingConv) ||
+          (mlir::isa<fir::RecordType>(ty) && !noStructConversion)) {
         LLVM_DEBUG(llvm::dbgs() << "rewrite " << signature << " for target\n");
         return false;
       }
@@ -740,7 +741,7 @@ public:
   /// Rewrite the signatures and body of the `FuncOp`s in the module for
   /// the immediately subsequent target code gen.
   void convertSignature(mlir::func::FuncOp func) {
-    auto funcTy = func.getFunctionType().cast<mlir::FunctionType>();
+    auto funcTy = mlir::cast<mlir::FunctionType>(func.getFunctionType());
     if (hasPortableSignature(funcTy, func) && !hasHostAssociations(func))
       return;
     llvm::SmallVector<mlir::Type> newResTys;
diff --git a/flang/lib/Optimizer/CodeGen/TypeConverter.cpp b/flang/lib/Optimizer/CodeGen/TypeConverter.cpp
index 8fa423f35806..fb2ec3f0b2f5 100644
--- a/flang/lib/Optimizer/CodeGen/TypeConverter.cpp
+++ b/flang/lib/Optimizer/CodeGen/TypeConverter.cpp
@@ -103,10 +103,10 @@ LLVMTypeConverter::LLVMTypeConverter(mlir::ModuleOp module, bool applyTBAA,
     for (auto mem : tuple.getTypes()) {
       // Prevent fir.box from degenerating to a pointer to a descriptor in the
       // context of a tuple type.
-      if (auto box = mem.dyn_cast<fir::BaseBoxType>())
+      if (auto box = mlir::dyn_cast<fir::BaseBoxType>(mem))
         members.push_back(convertBoxTypeAsStruct(box));
       else
-        members.push_back(convertType(mem).cast<mlir::Type>());
+        members.push_back(mlir::cast<mlir::Type>(convertType(mem)));
     }
     return mlir::LLVM::LLVMStructType::getLiteral(&getContext(), members,
                                                   /*isPacked=*/false);
@@ -181,10 +181,10 @@ std::optional<mlir::LogicalResult> LLVMTypeConverter::convertRecordType(
   for (auto mem : derived.getTypeList()) {
     // Prevent fir.box from degenerating to a pointer to a descriptor in the
     // context of a record type.
-    if (auto box = mem.second.dyn_cast<fir::BaseBoxType>())
+    if (auto box = mlir::dyn_cast<fir::BaseBoxType>(mem.second))
       members.push_back(convertBoxTypeAsStruct(box));
     else
-      members.push_back(convertType(mem.second).cast<mlir::Type>());
+      members.push_back(mlir::cast<mlir::Type>(convertType(mem.second)));
   }
   if (mlir::failed(st.setBody(members, /*isPacked=*/false)))
     return mlir::failure();
@@ -196,7 +196,7 @@ std::optional<mlir::LogicalResult> LLVMTypeConverter::convertRecordType(
 // Extended descriptors are required for derived types.
 bool LLVMTypeConverter::requiresExtendedDesc(mlir::Type boxElementType) const {
   auto eleTy = fir::unwrapSequenceType(boxElementType);
-  return eleTy.isa<fir::RecordType>();
+  return mlir::isa<fir::RecordType>(eleTy);
 }
 
 // This corresponds to the descriptor as defined in ISO_Fortran_binding.h and
@@ -211,7 +211,8 @@ mlir::Type LLVMTypeConverter::convertBoxTypeAsStruct(BaseBoxType box,
     ele = removeIndirection;
   auto eleTy = convertType(ele);
   // base_addr*
-  if (ele.isa<SequenceType>() && eleTy.isa<mlir::LLVM::LLVMPointerType>())
+  if (mlir::isa<SequenceType>(ele) &&
+      mlir::isa<mlir::LLVM::LLVMPointerType>(eleTy))
     dataDescFields.push_back(eleTy);
   else
     dataDescFields.push_back(
@@ -236,7 +237,7 @@ mlir::Type LLVMTypeConverter::convertBoxTypeAsStruct(BaseBoxType box,
       getDescFieldTypeModel<kF18AddendumPosInBox>()(&getContext()));
   // [dims]
   if (rank == unknownRank()) {
-    if (auto seqTy = ele.dyn_cast<SequenceType>())
+    if (auto seqTy = mlir::dyn_cast<SequenceType>(ele))
       rank = seqTy.getDimension();
     else
       rank = 0;
@@ -252,7 +253,8 @@ mlir::Type LLVMTypeConverter::convertBoxTypeAsStruct(BaseBoxType box,
     auto rowTy =
         getExtendedDescFieldTypeModel<kOptRowTypePosInBox>()(&getContext());
     dataDescFields.push_back(mlir::LLVM::LLVMArrayType::get(rowTy, 1));
-    if (auto recTy = fir::unwrapSequenceType(ele).dyn_cast<fir::RecordType>())
+    if (auto recTy =
+            mlir::dyn_cast<fir::RecordType>(fir::unwrapSequenceType(ele)))
       if (recTy.getNumLenParams() > 0) {
         // The descriptor design needs to be clarified regarding the number of
         // length parameters in the addendum. Since it can change for
diff --git a/flang/lib/Optimizer/Dialect/FIRAttr.cpp b/flang/lib/Optimizer/Dialect/FIRAttr.cpp
index e43710f5627e..9ea3a0568f69 100644
--- a/flang/lib/Optimizer/Dialect/FIRAttr.cpp
+++ b/flang/lib/Optimizer/Dialect/FIRAttr.cpp
@@ -264,23 +264,23 @@ void fir::FortranVariableFlagsAttr::print(mlir::AsmPrinter &printer) const {
 void fir::printFirAttribute(FIROpsDialect *dialect, mlir::Attribute attr,
                             mlir::DialectAsmPrinter &p) {
   auto &os = p.getStream();
-  if (auto exact = attr.dyn_cast<fir::ExactTypeAttr>()) {
+  if (auto exact = mlir::dyn_cast<fir::ExactTypeAttr>(attr)) {
     os << fir::ExactTypeAttr::getAttrName() << '<';
     p.printType(exact.getType());
     os << '>';
-  } else if (auto sub = attr.dyn_cast<fir::SubclassAttr>()) {
+  } else if (auto sub = mlir::dyn_cast<fir::SubclassAttr>(attr)) {
     os << fir::SubclassAttr::getAttrName() << '<';
     p.printType(sub.getType());
     os << '>';
-  } else if (attr.dyn_cast_or_null<fir::PointIntervalAttr>()) {
+  } else if (mlir::dyn_cast_or_null<fir::PointIntervalAttr>(attr)) {
     os << fir::PointIntervalAttr::getAttrName();
-  } else if (attr.dyn_cast_or_null<fir::ClosedIntervalAttr>()) {
+  } else if (mlir::dyn_cast_or_null<fir::ClosedIntervalAttr>(attr)) {
     os << fir::ClosedIntervalAttr::getAttrName();
-  } else if (attr.dyn_cast_or_null<fir::LowerBoundAttr>()) {
+  } else if (mlir::dyn_cast_or_null<fir::LowerBoundAttr>(attr)) {
     os << fir::LowerBoundAttr::getAttrName();
-  } else if (attr.dyn_cast_or_null<fir::UpperBoundAttr>()) {
+  } else if (mlir::dyn_cast_or_null<fir::UpperBoundAttr>(attr)) {
     os << fir::UpperBoundAttr::getAttrName();
-  } else if (auto a = attr.dyn_cast_or_null<fir::RealAttr>()) {
+  } else if (auto a = mlir::dyn_cast_or_null<fir::RealAttr>(attr)) {
     os << fir::RealAttr::getAttrName() << '<' << a.getFKind() << ", i x";
     llvm::SmallString<40> ss;
     a.getValue().bitcastToAPInt().toStringUnsigned(ss, 16);
diff --git a/flang/lib/Optimizer/Dialect/FIROps.cpp b/flang/lib/Optimizer/Dialect/FIROps.cpp
index 24af94f9b90a..6773d0adced0 100644
--- a/flang/lib/Optimizer/Dialect/FIROps.cpp
+++ b/flang/lib/Optimizer/Dialect/FIROps.cpp
@@ -57,7 +57,7 @@ static void propagateAttributes(mlir::Operation *fromOp,
 static bool verifyInType(mlir::Type inType,
                          llvm::SmallVectorImpl<llvm::StringRef> &visited,
                          unsigned dynamicExtents = 0) {
-  if (auto st = inType.dyn_cast<fir::SequenceType>()) {
+  if (auto st = mlir::dyn_cast<fir::SequenceType>(inType)) {
     auto shape = st.getShape();
     if (shape.size() == 0)
       return true;
@@ -67,7 +67,7 @@ static bool verifyInType(mlir::Type inType,
       if (dynamicExtents-- == 0)
         return true;
     }
-  } else if (auto rt = inType.dyn_cast<fir::RecordType>()) {
+  } else if (auto rt = mlir::dyn_cast<fir::RecordType>(inType)) {
     // don't recurse if we're already visiting this one
     if (llvm::is_contained(visited, rt.getName()))
       return false;
@@ -84,13 +84,13 @@ static bool verifyInType(mlir::Type inType,
 static bool verifyTypeParamCount(mlir::Type inType, unsigned numParams) {
   auto ty = fir::unwrapSequenceType(inType);
   if (numParams > 0) {
-    if (auto recTy = ty.dyn_cast<fir::RecordType>())
+    if (auto recTy = mlir::dyn_cast<fir::RecordType>(ty))
       return numParams != recTy.getNumLenParams();
-    if (auto chrTy = ty.dyn_cast<fir::CharacterType>())
+    if (auto chrTy = mlir::dyn_cast<fir::CharacterType>(ty))
       return !(numParams == 1 && chrTy.hasDynamicLen());
     return true;
   }
-  if (auto chrTy = ty.dyn_cast<fir::CharacterType>())
+  if (auto chrTy = mlir::dyn_cast<fir::CharacterType>(ty))
     return !chrTy.hasConstantLen();
   return false;
 }
@@ -171,13 +171,13 @@ static void printAllocatableOp(mlir::OpAsmPrinter &p, OP &op) {
 /// Create a legal memory reference as return type
 static mlir::Type wrapAllocaResultType(mlir::Type intype) {
   // FIR semantics: memory references to memory references are disallowed
-  if (intype.isa<fir::ReferenceType>())
+  if (mlir::isa<fir::ReferenceType>(intype))
     return {};
   return fir::ReferenceType::get(intype);
 }
 
 mlir::Type fir::AllocaOp::getAllocatedType() {
-  return getType().cast<fir::ReferenceType>().getEleTy();
+  return mlir::cast<fir::ReferenceType>(getType()).getEleTy();
 }
 
 mlir::Type fir::AllocaOp::getRefTy(mlir::Type ty) {
@@ -270,7 +270,7 @@ mlir::LogicalResult fir::AllocaOp::verify() {
   if (verifyTypeParamCount(getInType(), numLenParams()))
     return emitOpError("LEN params do not correspond to type");
   mlir::Type outType = getType();
-  if (!outType.isa<fir::ReferenceType>())
+  if (!mlir::isa<fir::ReferenceType>(outType))
     return emitOpError("must be a !fir.ref type");
   if (fir::isa_unknown_size_box(fir::dyn_cast_ptrEleTy(outType)))
     return emitOpError("cannot allocate !fir.box of unknown rank or type");
@@ -286,14 +286,14 @@ static mlir::Type wrapAllocMemResultType(mlir::Type intype) {
   // Fortran semantics: C852 an entity cannot be both ALLOCATABLE and POINTER
   // 8.5.3 note 1 prohibits ALLOCATABLE procedures as well
   // FIR semantics: one may not allocate a memory reference value
-  if (intype.isa<fir::ReferenceType, fir::HeapType, fir::PointerType,
-                 mlir::FunctionType>())
+  if (mlir::isa<fir::ReferenceType, fir::HeapType, fir::PointerType,
+                mlir::FunctionType>(intype))
     return {};
   return fir::HeapType::get(intype);
 }
 
 mlir::Type fir::AllocMemOp::getAllocatedType() {
-  return getType().cast<fir::HeapType>().getEleTy();
+  return mlir::cast<fir::HeapType>(getType()).getEleTy();
 }
 
 mlir::Type fir::AllocMemOp::getRefTy(mlir::Type ty) {
@@ -348,7 +348,7 @@ mlir::LogicalResult fir::AllocMemOp::verify() {
   if (verifyTypeParamCount(getInType(), numLenParams()))
     return emitOpError("LEN params do not correspond to type");
   mlir::Type outType = getType();
-  if (!outType.dyn_cast<fir::HeapType>())
+  if (!mlir::dyn_cast<fir::HeapType>(outType))
     return emitOpError("must be a !fir.heap type");
   if (fir::isa_unknown_size_box(fir::dyn_cast_ptrEleTy(outType)))
     return emitOpError("cannot allocate !fir.box of unknown rank or type");
@@ -364,13 +364,13 @@ mlir::LogicalResult fir::AllocMemOp::verify() {
 static bool validTypeParams(mlir::Type dynTy, mlir::ValueRange typeParams) {
   dynTy = fir::unwrapAllRefAndSeqType(dynTy);
   // A box value will contain type parameter values itself.
-  if (dynTy.isa<fir::BoxType>())
+  if (mlir::isa<fir::BoxType>(dynTy))
     return typeParams.size() == 0;
   // Derived type must have all type parameters satisfied.
-  if (auto recTy = dynTy.dyn_cast<fir::RecordType>())
+  if (auto recTy = mlir::dyn_cast<fir::RecordType>(dynTy))
     return typeParams.size() == recTy.getNumLenParams();
   // Characters with non-constant LEN must have a type parameter value.
-  if (auto charTy = dynTy.dyn_cast<fir::CharacterType>())
+  if (auto charTy = mlir::dyn_cast<fir::CharacterType>(dynTy))
     if (charTy.hasDynamicLen())
       return typeParams.size() == 1;
   // Otherwise, any type parameters are invalid.
@@ -379,7 +379,7 @@ static bool validTypeParams(mlir::Type dynTy, mlir::ValueRange typeParams) {
 
 mlir::LogicalResult fir::ArrayCoorOp::verify() {
   auto eleTy = fir::dyn_cast_ptrOrBoxEleTy(getMemref().getType());
-  auto arrTy = eleTy.dyn_cast<fir::SequenceType>();
+  auto arrTy = mlir::dyn_cast<fir::SequenceType>(eleTy);
   if (!arrTy)
     return emitOpError("must be a reference to an array");
   auto arrDim = arrTy.getDimension();
@@ -387,14 +387,14 @@ mlir::LogicalResult fir::ArrayCoorOp::verify() {
   if (auto shapeOp = getShape()) {
     auto shapeTy = shapeOp.getType();
     unsigned shapeTyRank = 0;
-    if (auto s = shapeTy.dyn_cast<fir::ShapeType>()) {
+    if (auto s = mlir::dyn_cast<fir::ShapeType>(shapeTy)) {
       shapeTyRank = s.getRank();
-    } else if (auto ss = shapeTy.dyn_cast<fir::ShapeShiftType>()) {
+    } else if (auto ss = mlir::dyn_cast<fir::ShapeShiftType>(shapeTy)) {
       shapeTyRank = ss.getRank();
     } else {
-      auto s = shapeTy.cast<fir::ShiftType>();
+      auto s = mlir::cast<fir::ShiftType>(shapeTy);
       shapeTyRank = s.getRank();
-      if (!getMemref().getType().isa<fir::BaseBoxType>())
+      if (!mlir::isa<fir::BaseBoxType>(getMemref().getType()))
         return emitOpError("shift can only be provided with fir.box memref");
     }
     if (arrDim && arrDim != shapeTyRank)
@@ -407,7 +407,7 @@ mlir::LogicalResult fir::ArrayCoorOp::verify() {
     if (auto sl = mlir::dyn_cast_or_null<fir::SliceOp>(sliceOp.getDefiningOp()))
       if (!sl.getSubstr().empty())
         return emitOpError("array_coor cannot take a slice with substring");
-    if (auto sliceTy = sliceOp.getType().dyn_cast<fir::SliceType>())
+    if (auto sliceTy = mlir::dyn_cast<fir::SliceType>(sliceOp.getType()))
       if (sliceTy.getRank() != arrDim)
         return emitOpError("rank of dimension in slice mismatched");
   }
@@ -422,13 +422,13 @@ mlir::LogicalResult fir::ArrayCoorOp::verify() {
 //===----------------------------------------------------------------------===//
 
 static mlir::Type adjustedElementType(mlir::Type t) {
-  if (auto ty = t.dyn_cast<fir::ReferenceType>()) {
+  if (auto ty = mlir::dyn_cast<fir::ReferenceType>(t)) {
     auto eleTy = ty.getEleTy();
     if (fir::isa_char(eleTy))
       return eleTy;
     if (fir::isa_derived(eleTy))
       return eleTy;
-    if (eleTy.isa<fir::SequenceType>())
+    if (mlir::isa<fir::SequenceType>(eleTy))
       return eleTy;
   }
   return t;
@@ -448,7 +448,7 @@ std::vector<mlir::Value> fir::ArrayLoadOp::getExtents() {
 
 mlir::LogicalResult fir::ArrayLoadOp::verify() {
   auto eleTy = fir::dyn_cast_ptrOrBoxEleTy(getMemref().getType());
-  auto arrTy = eleTy.dyn_cast<fir::SequenceType>();
+  auto arrTy = mlir::dyn_cast<fir::SequenceType>(eleTy);
   if (!arrTy)
     return emitOpError("must be a reference to an array");
   auto arrDim = arrTy.getDimension();
@@ -456,14 +456,14 @@ mlir::LogicalResult fir::ArrayLoadOp::verify() {
   if (auto shapeOp = getShape()) {
     auto shapeTy = shapeOp.getType();
     unsigned shapeTyRank = 0u;
-    if (auto s = shapeTy.dyn_cast<fir::ShapeType>()) {
+    if (auto s = mlir::dyn_cast<fir::ShapeType>(shapeTy)) {
       shapeTyRank = s.getRank();
-    } else if (auto ss = shapeTy.dyn_cast<fir::ShapeShiftType>()) {
+    } else if (auto ss = mlir::dyn_cast<fir::ShapeShiftType>(shapeTy)) {
       shapeTyRank = ss.getRank();
     } else {
-      auto s = shapeTy.cast<fir::ShiftType>();
+      auto s = mlir::cast<fir::ShiftType>(shapeTy);
       shapeTyRank = s.getRank();
-      if (!getMemref().getType().isa<fir::BaseBoxType>())
+      if (!mlir::isa<fir::BaseBoxType>(getMemref().getType()))
         return emitOpError("shift can only be provided with fir.box memref");
     }
     if (arrDim && arrDim != shapeTyRank)
@@ -474,7 +474,7 @@ mlir::LogicalResult fir::ArrayLoadOp::verify() {
     if (auto sl = mlir::dyn_cast_or_null<fir::SliceOp>(sliceOp.getDefiningOp()))
       if (!sl.getSubstr().empty())
         return emitOpError("array_load cannot take a slice with substring");
-    if (auto sliceTy = sliceOp.getType().dyn_cast<fir::SliceType>())
+    if (auto sliceTy = mlir::dyn_cast<fir::SliceType>(sliceOp.getType()))
       if (sliceTy.getRank() != arrDim)
         return emitOpError("rank of dimension in slice mismatched");
   }
@@ -502,7 +502,7 @@ mlir::LogicalResult fir::ArrayMergeStoreOp::verify() {
         // This is an intra-object merge, where the slice is projecting the
         // subfields that are to be overwritten by the merge operation.
         auto eleTy = fir::dyn_cast_ptrOrBoxEleTy(getMemref().getType());
-        if (auto seqTy = eleTy.dyn_cast<fir::SequenceType>()) {
+        if (auto seqTy = mlir::dyn_cast<fir::SequenceType>(eleTy)) {
           auto projTy =
               fir::applyPathToType(seqTy.getEleTy(), sliceOp.getFields());
           if (fir::unwrapSequenceType(getOriginal().getType()) != projTy)
@@ -540,7 +540,7 @@ mlir::Type validArraySubobject(A op) {
 }
 
 mlir::LogicalResult fir::ArrayFetchOp::verify() {
-  auto arrTy = getSequence().getType().cast<fir::SequenceType>();
+  auto arrTy = mlir::cast<fir::SequenceType>(getSequence().getType());
   auto indSize = getIndices().size();
   if (indSize < arrTy.getDimension())
     return emitOpError("number of indices != dimension of array");
@@ -562,7 +562,7 @@ mlir::LogicalResult fir::ArrayFetchOp::verify() {
 //===----------------------------------------------------------------------===//
 
 mlir::LogicalResult fir::ArrayAccessOp::verify() {
-  auto arrTy = getSequence().getType().cast<fir::SequenceType>();
+  auto arrTy = mlir::cast<fir::SequenceType>(getSequence().getType());
   std::size_t indSize = getIndices().size();
   if (indSize < arrTy.getDimension())
     return emitOpError("number of indices != dimension of array");
@@ -584,7 +584,7 @@ mlir::LogicalResult fir::ArrayAccessOp::verify() {
 mlir::LogicalResult fir::ArrayUpdateOp::verify() {
   if (fir::isa_ref_type(getMerge().getType()))
     return emitOpError("does not support reference type for merge");
-  auto arrTy = getSequence().getType().cast<fir::SequenceType>();
+  auto arrTy = mlir::cast<fir::SequenceType>(getSequence().getType());
   auto indSize = getIndices().size();
   if (indSize < arrTy.getDimension())
     return emitOpError("number of indices != dimension of array");
@@ -604,7 +604,7 @@ mlir::LogicalResult fir::ArrayUpdateOp::verify() {
 //===----------------------------------------------------------------------===//
 
 mlir::LogicalResult fir::ArrayModifyOp::verify() {
-  auto arrTy = getSequence().getType().cast<fir::SequenceType>();
+  auto arrTy = mlir::cast<fir::SequenceType>(getSequence().getType());
   auto indSize = getIndices().size();
   if (indSize < arrTy.getDimension())
     return emitOpError("number of indices must match array dimension");
@@ -740,7 +740,7 @@ mlir::ParseResult fir::CallOp::parse(mlir::OpAsmParser &parser,
       parser.parseType(type))
     return mlir::failure();
 
-  auto funcType = type.dyn_cast<mlir::FunctionType>();
+  auto funcType = mlir::dyn_cast<mlir::FunctionType>(type);
   if (!funcType)
     return parser.emitError(parser.getNameLoc(), "expected function type");
   if (isDirect) {
@@ -785,7 +785,7 @@ void fir::CallOp::build(mlir::OpBuilder &builder, mlir::OperationState &result,
 mlir::LogicalResult fir::CharConvertOp::verify() {
   auto unwrap = [&](mlir::Type t) {
     t = fir::unwrapSequenceType(fir::dyn_cast_ptrEleTy(t));
-    return t.dyn_cast<fir::CharacterType>();
+    return mlir::dyn_cast<fir::CharacterType>(t);
   };
   auto inTy = unwrap(getFrom().getType());
   auto outTy = unwrap(getTo().getType());
@@ -832,13 +832,13 @@ static mlir::ParseResult parseCmpOp(mlir::OpAsmParser &parser,
       parser.resolveOperands(ops, type, result.operands))
     return mlir::failure();
 
-  if (!predicateNameAttr.isa<mlir::StringAttr>())
+  if (!mlir::isa<mlir::StringAttr>(predicateNameAttr))
     return parser.emitError(parser.getNameLoc(),
                             "expected string comparison predicate attribute");
 
   // Rewrite string attribute to an enum value.
   llvm::StringRef predicateName =
-      predicateNameAttr.cast<mlir::StringAttr>().getValue();
+      mlir::cast<mlir::StringAttr>(predicateNameAttr).getValue();
   auto predicate = fir::CmpcOp::getPredicateByName(predicateName);
   auto builder = parser.getBuilder();
   mlir::Type i1Type = builder.getI1Type();
@@ -906,7 +906,7 @@ void fir::ConstcOp::print(mlir::OpAsmPrinter &p) {
 }
 
 mlir::LogicalResult fir::ConstcOp::verify() {
-  if (!getType().isa<fir::ComplexType>())
+  if (!mlir::isa<fir::ComplexType>(getType()))
     return emitOpError("must be a !fir.complex type");
   return mlir::success();
 }
@@ -929,15 +929,16 @@ mlir::OpFoldResult fir::ConvertOp::fold(FoldAdaptor adaptor) {
   if (matchPattern(getValue(), mlir::m_Op<fir::ConvertOp>())) {
     auto inner = mlir::cast<fir::ConvertOp>(getValue().getDefiningOp());
     // (convert (convert 'a : logical -> i1) : i1 -> logical) ==> forward 'a
-    if (auto toTy = getType().dyn_cast<fir::LogicalType>())
-      if (auto fromTy = inner.getValue().getType().dyn_cast<fir::LogicalType>())
-        if (inner.getType().isa<mlir::IntegerType>() && (toTy == fromTy))
+    if (auto toTy = mlir::dyn_cast<fir::LogicalType>(getType()))
+      if (auto fromTy =
+              mlir::dyn_cast<fir::LogicalType>(inner.getValue().getType()))
+        if (mlir::isa<mlir::IntegerType>(inner.getType()) && (toTy == fromTy))
           return inner.getValue();
     // (convert (convert 'a : i1 -> logical) : logical -> i1) ==> forward 'a
-    if (auto toTy = getType().dyn_cast<mlir::IntegerType>())
+    if (auto toTy = mlir::dyn_cast<mlir::IntegerType>(getType()))
       if (auto fromTy =
-              inner.getValue().getType().dyn_cast<mlir::IntegerType>())
-        if (inner.getType().isa<fir::LogicalType>() && (toTy == fromTy) &&
+              mlir::dyn_cast<mlir::IntegerType>(inner.getValue().getType()))
+        if (mlir::isa<fir::LogicalType>(inner.getType()) && (toTy == fromTy) &&
             (fromTy.getWidth() == 1))
           return inner.getValue();
   }
@@ -945,7 +946,7 @@ mlir::OpFoldResult fir::ConvertOp::fold(FoldAdaptor adaptor) {
 }
 
 bool fir::ConvertOp::isInteger(mlir::Type ty) {
-  return ty.isa<mlir::IntegerType, mlir::IndexType, fir::IntegerType>();
+  return mlir::isa<mlir::IntegerType, mlir::IndexType, fir::IntegerType>(ty);
 }
 
 bool fir::ConvertOp::isIntegerCompatible(mlir::Type ty) {
@@ -953,13 +954,13 @@ bool fir::ConvertOp::isIntegerCompatible(mlir::Type ty) {
 }
 
 bool fir::ConvertOp::isFloatCompatible(mlir::Type ty) {
-  return ty.isa<mlir::FloatType, fir::RealType>();
+  return mlir::isa<mlir::FloatType, fir::RealType>(ty);
 }
 
 bool fir::ConvertOp::isPointerCompatible(mlir::Type ty) {
-  return ty.isa<fir::ReferenceType, fir::PointerType, fir::HeapType,
-                fir::LLVMPointerType, mlir::MemRefType, mlir::FunctionType,
-                fir::TypeDescType>();
+  return mlir::isa<fir::ReferenceType, fir::PointerType, fir::HeapType,
+                   fir::LLVMPointerType, mlir::MemRefType, mlir::FunctionType,
+                   fir::TypeDescType>(ty);
 }
 
 static std::optional<mlir::Type> getVectorElementType(mlir::Type ty) {
@@ -1026,12 +1027,14 @@ bool fir::ConvertOp::canBeConverted(mlir::Type inType, mlir::Type outType) {
          (isFloatCompatible(inType) && isFloatCompatible(outType)) ||
          (isIntegerCompatible(inType) && isPointerCompatible(outType)) ||
          (isPointerCompatible(inType) && isIntegerCompatible(outType)) ||
-         (inType.isa<fir::BoxType>() && outType.isa<fir::BoxType>()) ||
-         (inType.isa<fir::BoxProcType>() && outType.isa<fir::BoxProcType>()) ||
+         (mlir::isa<fir::BoxType>(inType) &&
+          mlir::isa<fir::BoxType>(outType)) ||
+         (mlir::isa<fir::BoxProcType>(inType) &&
+          mlir::isa<fir::BoxProcType>(outType)) ||
          (fir::isa_complex(inType) && fir::isa_complex(outType)) ||
          (fir::isBoxedRecordType(inType) && fir::isPolymorphicType(outType)) ||
          (fir::isPolymorphicType(inType) && fir::isPolymorphicType(outType)) ||
-         (fir::isPolymorphicType(inType) && outType.isa<BoxType>()) ||
+         (fir::isPolymorphicType(inType) && mlir::isa<BoxType>(outType)) ||
          areVectorsCompatible(inType, outType);
 }
 
@@ -1079,7 +1082,7 @@ mlir::LogicalResult fir::CoordinateOp::verify() {
   const mlir::Type refTy = getRef().getType();
   if (fir::isa_ref_type(refTy)) {
     auto eleTy = fir::dyn_cast_ptrEleTy(refTy);
-    if (auto arrTy = eleTy.dyn_cast<fir::SequenceType>()) {
+    if (auto arrTy = mlir::dyn_cast<fir::SequenceType>(eleTy)) {
       if (arrTy.hasUnknownShape())
         return emitOpError("cannot find coordinate in unknown shape");
       if (arrTy.getConstantRows() < arrTy.getDimension() - 1)
@@ -1094,8 +1097,8 @@ mlir::LogicalResult fir::CoordinateOp::verify() {
   const unsigned numCoors = getCoor().size();
   for (auto coorOperand : llvm::enumerate(getCoor())) {
     auto co = coorOperand.value();
-    if (dimension == 0 && eleTy.isa<fir::SequenceType>()) {
-      dimension = eleTy.cast<fir::SequenceType>().getDimension();
+    if (dimension == 0 && mlir::isa<fir::SequenceType>(eleTy)) {
+      dimension = mlir::cast<fir::SequenceType>(eleTy).getDimension();
       if (dimension == 0)
         return emitOpError("cannot apply to array of unknown rank");
     }
@@ -1104,7 +1107,7 @@ mlir::LogicalResult fir::CoordinateOp::verify() {
         // Recovering a LEN type parameter only makes sense from a boxed
         // value. For a bare reference, the LEN type parameters must be
         // passed as additional arguments to `index`.
-        if (refTy.isa<fir::BoxType>()) {
+        if (mlir::isa<fir::BoxType>(refTy)) {
           if (coorOperand.index() != numCoors - 1)
             return emitOpError("len_param_index must be last argument");
           if (getNumOperands() != 2)
@@ -1117,7 +1120,7 @@ mlir::LogicalResult fir::CoordinateOp::verify() {
       } else if (auto index = mlir::dyn_cast<fir::FieldIndexOp>(defOp)) {
         if (eleTy != index.getOnType())
           emitOpError("field_index type not compatible with reference type");
-        if (auto recTy = eleTy.dyn_cast<fir::RecordType>()) {
+        if (auto recTy = mlir::dyn_cast<fir::RecordType>(eleTy)) {
           eleTy = recTy.getType(index.getFieldName());
           continue;
         }
@@ -1126,21 +1129,21 @@ mlir::LogicalResult fir::CoordinateOp::verify() {
     }
     if (dimension) {
       if (--dimension == 0)
-        eleTy = eleTy.cast<fir::SequenceType>().getEleTy();
+        eleTy = mlir::cast<fir::SequenceType>(eleTy).getEleTy();
     } else {
-      if (auto t = eleTy.dyn_cast<mlir::TupleType>()) {
+      if (auto t = mlir::dyn_cast<mlir::TupleType>(eleTy)) {
         // FIXME: Generally, we don't know which field of the tuple is being
         // referred to unless the operand is a constant. Just assume everything
         // is good in the tuple case for now.
         return mlir::success();
-      } else if (auto t = eleTy.dyn_cast<fir::RecordType>()) {
+      } else if (auto t = mlir::dyn_cast<fir::RecordType>(eleTy)) {
         // FIXME: This is the same as the tuple case.
         return mlir::success();
-      } else if (auto t = eleTy.dyn_cast<fir::ComplexType>()) {
+      } else if (auto t = mlir::dyn_cast<fir::ComplexType>(eleTy)) {
         eleTy = t.getElementType();
-      } else if (auto t = eleTy.dyn_cast<mlir::ComplexType>()) {
+      } else if (auto t = mlir::dyn_cast<mlir::ComplexType>(eleTy)) {
         eleTy = t.getElementType();
-      } else if (auto t = eleTy.dyn_cast<fir::CharacterType>()) {
+      } else if (auto t = mlir::dyn_cast<fir::CharacterType>(eleTy)) {
         if (t.getLen() == fir::CharacterType::singleton())
           return emitOpError("cannot apply to character singleton");
         eleTy = fir::CharacterType::getSingleton(t.getContext(), t.getFKind());
@@ -1216,17 +1219,17 @@ mlir::LogicalResult fir::TypeInfoOp::verify() {
 mlir::LogicalResult fir::EmboxOp::verify() {
   auto eleTy = fir::dyn_cast_ptrEleTy(getMemref().getType());
   bool isArray = false;
-  if (auto seqTy = eleTy.dyn_cast<fir::SequenceType>()) {
+  if (auto seqTy = mlir::dyn_cast<fir::SequenceType>(eleTy)) {
     eleTy = seqTy.getEleTy();
     isArray = true;
   }
   if (hasLenParams()) {
     auto lenPs = numLenParams();
-    if (auto rt = eleTy.dyn_cast<fir::RecordType>()) {
+    if (auto rt = mlir::dyn_cast<fir::RecordType>(eleTy)) {
       if (lenPs != rt.getNumLenParams())
         return emitOpError("number of LEN params does not correspond"
                            " to the !fir.type type");
-    } else if (auto strTy = eleTy.dyn_cast<fir::CharacterType>()) {
+    } else if (auto strTy = mlir::dyn_cast<fir::CharacterType>(eleTy)) {
       if (strTy.getLen() != fir::CharacterType::unknownLen())
         return emitOpError("CHARACTER already has static LEN");
     } else {
@@ -1240,7 +1243,7 @@ mlir::LogicalResult fir::EmboxOp::verify() {
     return emitOpError("shape must not be provided for a scalar");
   if (getSlice() && !isArray)
     return emitOpError("slice must not be provided for a scalar");
-  if (getSourceBox() && !getResult().getType().isa<fir::ClassType>())
+  if (getSourceBox() && !mlir::isa<fir::ClassType>(getResult().getType()))
     return emitOpError("source_box must be used with fir.class result type");
   return mlir::success();
 }
@@ -1251,7 +1254,7 @@ mlir::LogicalResult fir::EmboxOp::verify() {
 
 mlir::LogicalResult fir::EmboxCharOp::verify() {
   auto eleTy = fir::dyn_cast_ptrEleTy(getMemref().getType());
-  if (!eleTy.dyn_cast_or_null<fir::CharacterType>())
+  if (!mlir::dyn_cast_or_null<fir::CharacterType>(eleTy))
     return mlir::failure();
   return mlir::success();
 }
@@ -1263,8 +1266,8 @@ mlir::LogicalResult fir::EmboxCharOp::verify() {
 mlir::LogicalResult fir::EmboxProcOp::verify() {
   // host bindings (optional) must be a reference to a tuple
   if (auto h = getHost()) {
-    if (auto r = h.getType().dyn_cast<fir::ReferenceType>())
-      if (r.getEleTy().isa<mlir::TupleType>())
+    if (auto r = mlir::dyn_cast<fir::ReferenceType>(h.getType()))
+      if (mlir::isa<mlir::TupleType>(r.getEleTy()))
         return mlir::success();
     return mlir::failure();
   }
@@ -1300,7 +1303,7 @@ void fir::TypeDescOp::print(mlir::OpAsmPrinter &p) {
 
 mlir::LogicalResult fir::TypeDescOp::verify() {
   mlir::Type resultTy = getType();
-  if (auto tdesc = resultTy.dyn_cast<fir::TypeDescType>()) {
+  if (auto tdesc = mlir::dyn_cast<fir::TypeDescType>(resultTy)) {
     if (tdesc.getOfTy() != getInType())
       return emitOpError("wrapped type mismatched");
     return mlir::success();
@@ -1527,7 +1530,7 @@ mlir::ParseResult parseFieldLikeOp(mlir::OpAsmParser &parser,
     return mlir::failure();
   result.addAttribute(fir::FieldIndexOp::getFieldAttrName(),
                       builder.getStringAttr(fieldName));
-  if (!recty.dyn_cast<fir::RecordType>())
+  if (!mlir::dyn_cast<fir::RecordType>(recty))
     return mlir::failure();
   result.addAttribute(fir::FieldIndexOp::getTypeAttrName(),
                       mlir::TypeAttr::get(recty));
@@ -1671,7 +1674,7 @@ mlir::LogicalResult fir::InsertOnRangeOp::verify() {
 //===----------------------------------------------------------------------===//
 
 static bool checkIsIntegerConstant(mlir::Attribute attr, std::int64_t conVal) {
-  if (auto iattr = attr.dyn_cast<mlir::IntegerAttr>())
+  if (auto iattr = mlir::dyn_cast<mlir::IntegerAttr>(attr))
     return iattr.getInt() == conVal;
   return false;
 }
@@ -1690,7 +1693,7 @@ struct UndoComplexPattern : public mlir::RewritePattern {
   matchAndRewrite(mlir::Operation *op,
                   mlir::PatternRewriter &rewriter) const override {
     auto insval = mlir::dyn_cast_or_null<fir::InsertValueOp>(op);
-    if (!insval || !insval.getType().isa<fir::ComplexType>())
+    if (!insval || !mlir::isa<fir::ComplexType>(insval.getType()))
       return mlir::failure();
     auto insval2 = mlir::dyn_cast_or_null<fir::InsertValueOp>(
         insval.getAdt().getDefiningOp());
@@ -1819,7 +1822,7 @@ mlir::ParseResult fir::IterWhileOp::parse(mlir::OpAsmParser &parser,
         parser.parseRParen())
       return mlir::failure();
     // Type list must be "(index, i1)".
-    if (typeList.size() != 2 || !typeList[0].isa<mlir::IndexType>() ||
+    if (typeList.size() != 2 || !mlir::isa<mlir::IndexType>(typeList[0]) ||
         !typeList[1].isSignlessInteger(1))
       return mlir::failure();
     result.addTypes(typeList);
@@ -1873,7 +1876,7 @@ mlir::LogicalResult fir::IterWhileOp::verify() {
   auto opNumResults = getNumResults();
   if (getFinalValue()) {
     // Result type must be "(index, i1, ...)".
-    if (!getResult(0).getType().isa<mlir::IndexType>())
+    if (!mlir::isa<mlir::IndexType>(getResult(0).getType()))
       return emitOpError("result #0 expected to be index");
     if (!getResult(1).getType().isSignlessInteger(1))
       return emitOpError("result #1 expected to be i1");
@@ -2316,7 +2319,7 @@ void fir::DTEntryOp::print(mlir::OpAsmPrinter &p) {
 /// Example: return f32 for !fir.box<!fir.heap<!fir.array<?x?xf32>>.
 static mlir::Type getBoxScalarEleTy(mlir::Type boxTy) {
   auto eleTy = fir::dyn_cast_ptrOrBoxEleTy(boxTy);
-  if (auto seqTy = eleTy.dyn_cast<fir::SequenceType>())
+  if (auto seqTy = mlir::dyn_cast<fir::SequenceType>(eleTy))
     return seqTy.getEleTy();
   return eleTy;
 }
@@ -2324,8 +2327,8 @@ static mlir::Type getBoxScalarEleTy(mlir::Type boxTy) {
 /// Test if \p t1 and \p t2 are compatible character types (if they can
 /// represent the same type at runtime).
 static bool areCompatibleCharacterTypes(mlir::Type t1, mlir::Type t2) {
-  auto c1 = t1.dyn_cast<fir::CharacterType>();
-  auto c2 = t2.dyn_cast<fir::CharacterType>();
+  auto c1 = mlir::dyn_cast<fir::CharacterType>(t1);
+  auto c2 = mlir::dyn_cast<fir::CharacterType>(t2);
   if (!c1 || !c2)
     return false;
   if (c1.hasDynamicLen() || c2.hasDynamicLen())
@@ -2347,10 +2350,10 @@ mlir::LogicalResult fir::ReboxOp::verify() {
 
   if (auto sliceVal = getSlice()) {
     // Slicing case
-    if (sliceVal.getType().cast<fir::SliceType>().getRank() != inputRank)
+    if (mlir::cast<fir::SliceType>(sliceVal.getType()).getRank() != inputRank)
       return emitOpError("slice operand rank must match box operand rank");
     if (auto shapeVal = getShape()) {
-      if (auto shiftTy = shapeVal.getType().dyn_cast<fir::ShiftType>()) {
+      if (auto shiftTy = mlir::dyn_cast<fir::ShiftType>(shapeVal.getType())) {
         if (shiftTy.getRank() != inputRank)
           return emitOpError("shape operand and input box ranks must match "
                              "when there is a slice");
@@ -2370,12 +2373,12 @@ mlir::LogicalResult fir::ReboxOp::verify() {
     unsigned shapeRank = inputRank;
     if (auto shapeVal = getShape()) {
       auto ty = shapeVal.getType();
-      if (auto shapeTy = ty.dyn_cast<fir::ShapeType>()) {
+      if (auto shapeTy = mlir::dyn_cast<fir::ShapeType>(ty)) {
         shapeRank = shapeTy.getRank();
-      } else if (auto shapeShiftTy = ty.dyn_cast<fir::ShapeShiftType>()) {
+      } else if (auto shapeShiftTy = mlir::dyn_cast<fir::ShapeShiftType>(ty)) {
         shapeRank = shapeShiftTy.getRank();
       } else {
-        auto shiftTy = ty.cast<fir::ShiftType>();
+        auto shiftTy = mlir::cast<fir::ShiftType>(ty);
         shapeRank = shiftTy.getRank();
         if (shapeRank != inputRank)
           return emitOpError("shape operand and input box ranks must match "
@@ -2394,11 +2397,13 @@ mlir::LogicalResult fir::ReboxOp::verify() {
     // the types is a character with dynamic length, the other type can be any
     // character type.
     const bool typeCanMismatch =
-        inputEleTy.isa<fir::RecordType>() || outEleTy.isa<mlir::NoneType>() ||
-        (inputEleTy.isa<mlir::NoneType>() && outEleTy.isa<fir::RecordType>()) ||
-        (getSlice() && inputEleTy.isa<fir::CharacterType>()) ||
+        mlir::isa<fir::RecordType>(inputEleTy) ||
+        mlir::isa<mlir::NoneType>(outEleTy) ||
+        (mlir::isa<mlir::NoneType>(inputEleTy) &&
+         mlir::isa<fir::RecordType>(outEleTy)) ||
+        (getSlice() && mlir::isa<fir::CharacterType>(inputEleTy)) ||
         (getSlice() && fir::isa_complex(inputEleTy) &&
-         outEleTy.isa<mlir::FloatType>()) ||
+         mlir::isa<mlir::FloatType>(outEleTy)) ||
         areCompatibleCharacterTypes(inputEleTy, outEleTy);
     if (!typeCanMismatch)
       return emitOpError(
@@ -2435,7 +2440,7 @@ mlir::LogicalResult fir::SaveResultOp::verify() {
   if (fir::isa_unknown_size_box(resultType))
     return emitOpError("cannot save !fir.box of unknown rank or type");
 
-  if (resultType.isa<fir::BoxType>()) {
+  if (mlir::isa<fir::BoxType>(resultType)) {
     if (getShape() || !getTypeparams().empty())
       return emitOpError(
           "must not have shape or length operands if the value is a fir.box");
@@ -2446,14 +2451,14 @@ mlir::LogicalResult fir::SaveResultOp::verify() {
   unsigned shapeTyRank = 0;
   if (auto shapeVal = getShape()) {
     auto shapeTy = shapeVal.getType();
-    if (auto s = shapeTy.dyn_cast<fir::ShapeType>())
+    if (auto s = mlir::dyn_cast<fir::ShapeType>(shapeTy))
       shapeTyRank = s.getRank();
     else
-      shapeTyRank = shapeTy.cast<fir::ShapeShiftType>().getRank();
+      shapeTyRank = mlir::cast<fir::ShapeShiftType>(shapeTy).getRank();
   }
 
   auto eleTy = resultType;
-  if (auto seqTy = resultType.dyn_cast<fir::SequenceType>()) {
+  if (auto seqTy = mlir::dyn_cast<fir::SequenceType>(resultType)) {
     if (seqTy.getDimension() != shapeTyRank)
       emitOpError("shape operand must be provided and have the value rank "
                   "when the value is a fir.array");
@@ -2464,11 +2469,11 @@ mlir::LogicalResult fir::SaveResultOp::verify() {
           "shape operand should only be provided if the value is a fir.array");
   }
 
-  if (auto recTy = eleTy.dyn_cast<fir::RecordType>()) {
+  if (auto recTy = mlir::dyn_cast<fir::RecordType>(eleTy)) {
     if (recTy.getNumLenParams() != getTypeparams().size())
       emitOpError("length parameters number must match with the value type "
                   "length parameters");
-  } else if (auto charTy = eleTy.dyn_cast<fir::CharacterType>()) {
+  } else if (auto charTy = mlir::dyn_cast<fir::CharacterType>(eleTy)) {
     if (getTypeparams().size() > 1)
       emitOpError("no more than one length parameter must be provided for "
                   "character value");
@@ -2493,10 +2498,8 @@ static constexpr llvm::StringRef getTargetOffsetAttr() {
 
 template <typename OpT>
 static mlir::LogicalResult verifyIntegralSwitchTerminator(OpT op) {
-  if (!op.getSelector()
-           .getType()
-           .template isa<mlir::IntegerType, mlir::IndexType,
-                         fir::IntegerType>())
+  if (!mlir::isa<mlir::IntegerType, mlir::IndexType, fir::IntegerType>(
+          op.getSelector().getType()))
     return op.emitOpError("must be an integer");
   auto cases =
       op->template getAttrOfType<mlir::ArrayAttr>(op.getCasesAttr()).getValue();
@@ -2508,7 +2511,7 @@ static mlir::LogicalResult verifyIntegralSwitchTerminator(OpT op) {
   if (op.targetOffsetSize() != count)
     return op.emitOpError("incorrect number of successor operand groups");
   for (decltype(count) i = 0; i != count; ++i) {
-    if (!cases[i].template isa<mlir::IntegerAttr, mlir::UnitAttr>())
+    if (!mlir::isa<mlir::IntegerAttr, mlir::UnitAttr>(cases[i]))
       return op.emitOpError("invalid case alternative");
   }
   return mlir::success();
@@ -2571,7 +2574,7 @@ static void printIntegralSwitchTerminator(OpT op, mlir::OpAsmPrinter &p) {
     if (i)
       p << ", ";
     auto &attr = cases[i];
-    if (auto intAttr = attr.template dyn_cast_or_null<mlir::IntegerAttr>())
+    if (auto intAttr = mlir::dyn_cast_or_null<mlir::IntegerAttr>(attr))
       p << intAttr.getValue();
     else
       p.printAttribute(attr);
@@ -2620,7 +2623,7 @@ getMutableSuccessorOperands(unsigned pos, mlir::MutableOperandRange operands,
       *owner->getAttrDictionary().getNamed(offsetAttr);
   return getSubOperands(
       pos, operands,
-      targetOffsetAttr.getValue().cast<mlir::DenseI32ArrayAttr>(),
+      mlir::cast<mlir::DenseI32ArrayAttr>(targetOffsetAttr.getValue()),
       mlir::MutableOperandRange::OperandSegment(pos, targetOffsetAttr));
 }
 
@@ -2742,9 +2745,9 @@ mlir::ParseResult fir::SelectCaseOp::parse(mlir::OpAsmParser &parser,
         parser.parseComma())
       return mlir::failure();
     attrs.push_back(attr);
-    if (attr.dyn_cast_or_null<mlir::UnitAttr>()) {
+    if (mlir::dyn_cast_or_null<mlir::UnitAttr>(attr)) {
       argOffs.push_back(0);
-    } else if (attr.dyn_cast_or_null<fir::ClosedIntervalAttr>()) {
+    } else if (mlir::dyn_cast_or_null<fir::ClosedIntervalAttr>(attr)) {
       mlir::OpAsmParser::UnresolvedOperand oper1;
       mlir::OpAsmParser::UnresolvedOperand oper2;
       if (parser.parseOperand(oper1) || parser.parseComma() ||
@@ -2806,11 +2809,11 @@ void fir::SelectCaseOp::print(mlir::OpAsmPrinter &p) {
     if (i)
       p << ", ";
     p << cases[i] << ", ";
-    if (!cases[i].isa<mlir::UnitAttr>()) {
+    if (!mlir::isa<mlir::UnitAttr>(cases[i])) {
       auto caseArgs = *getCompareOperands(i);
       p.printOperand(*caseArgs.begin());
       p << ", ";
-      if (cases[i].isa<fir::ClosedIntervalAttr>()) {
+      if (mlir::isa<fir::ClosedIntervalAttr>(cases[i])) {
         p.printOperand(*(++caseArgs.begin()));
         p << ", ";
       }
@@ -2848,10 +2851,10 @@ void fir::SelectCaseOp::build(mlir::OpBuilder &builder,
   llvm::SmallVector<int32_t> operOffs;
   int32_t operSize = 0;
   for (auto attr : compareAttrs) {
-    if (attr.isa<fir::ClosedIntervalAttr>()) {
+    if (mlir::isa<fir::ClosedIntervalAttr>(attr)) {
       operOffs.push_back(2);
       operSize += 2;
-    } else if (attr.isa<mlir::UnitAttr>()) {
+    } else if (mlir::isa<mlir::UnitAttr>(attr)) {
       operOffs.push_back(0);
     } else {
       operOffs.push_back(1);
@@ -2900,10 +2903,10 @@ void fir::SelectCaseOp::build(mlir::OpBuilder &builder,
   llvm::SmallVector<mlir::ValueRange> cmpOpers;
   auto iter = cmpOpList.begin();
   for (auto &attr : compareAttrs) {
-    if (attr.isa<fir::ClosedIntervalAttr>()) {
+    if (mlir::isa<fir::ClosedIntervalAttr>(attr)) {
       cmpOpers.push_back(mlir::ValueRange({iter, iter + 2}));
       iter += 2;
-    } else if (attr.isa<mlir::UnitAttr>()) {
+    } else if (mlir::isa<mlir::UnitAttr>(attr)) {
       cmpOpers.push_back(mlir::ValueRange{});
     } else {
       cmpOpers.push_back(mlir::ValueRange({iter, iter + 1}));
@@ -2915,10 +2918,8 @@ void fir::SelectCaseOp::build(mlir::OpBuilder &builder,
 }
 
 mlir::LogicalResult fir::SelectCaseOp::verify() {
-  if (!getSelector()
-           .getType()
-           .isa<mlir::IntegerType, mlir::IndexType, fir::IntegerType,
-                fir::LogicalType, fir::CharacterType>())
+  if (!mlir::isa<mlir::IntegerType, mlir::IndexType, fir::IntegerType,
+                 fir::LogicalType, fir::CharacterType>(getSelector().getType()))
     return emitOpError("must be an integer, character, or logical");
   auto cases =
       getOperation()->getAttrOfType<mlir::ArrayAttr>(getCasesAttr()).getValue();
@@ -2933,9 +2934,11 @@ mlir::LogicalResult fir::SelectCaseOp::verify() {
     return emitOpError("incorrect number of successor operand groups");
   for (decltype(count) i = 0; i != count; ++i) {
     auto &attr = cases[i];
-    if (!(attr.isa<fir::PointIntervalAttr>() ||
-          attr.isa<fir::LowerBoundAttr>() || attr.isa<fir::UpperBoundAttr>() ||
-          attr.isa<fir::ClosedIntervalAttr>() || attr.isa<mlir::UnitAttr>()))
+    if (!(mlir::isa<fir::PointIntervalAttr>(attr) ||
+          mlir::isa<fir::LowerBoundAttr>(attr) ||
+          mlir::isa<fir::UpperBoundAttr>(attr) ||
+          mlir::isa<fir::ClosedIntervalAttr>(attr) ||
+          mlir::isa<mlir::UnitAttr>(attr)))
       return emitOpError("incorrect select case attribute type");
   }
   return mlir::success();
@@ -3111,14 +3114,14 @@ void fir::SelectTypeOp::print(mlir::OpAsmPrinter &p) {
 }
 
 mlir::LogicalResult fir::SelectTypeOp::verify() {
-  if (!(getSelector().getType().isa<fir::BaseBoxType>()))
+  if (!mlir::isa<fir::BaseBoxType>(getSelector().getType()))
     return emitOpError("must be a fir.class or fir.box type");
-  if (auto boxType = getSelector().getType().dyn_cast<fir::BoxType>())
-    if (!boxType.getEleTy().isa<mlir::NoneType>())
+  if (auto boxType = mlir::dyn_cast<fir::BoxType>(getSelector().getType()))
+    if (!mlir::isa<mlir::NoneType>(boxType.getEleTy()))
       return emitOpError("selector must be polymorphic");
   auto typeGuardAttr = getCases();
   for (unsigned idx = 0; idx < typeGuardAttr.size(); ++idx)
-    if (typeGuardAttr[idx].isa<mlir::UnitAttr>() &&
+    if (mlir::isa<mlir::UnitAttr>(typeGuardAttr[idx]) &&
         idx != typeGuardAttr.size() - 1)
       return emitOpError("default must be the last attribute");
   auto count = getNumDest();
@@ -3129,9 +3132,8 @@ mlir::LogicalResult fir::SelectTypeOp::verify() {
   if (targetOffsetSize() != count)
     return emitOpError("incorrect number of successor operand groups");
   for (unsigned i = 0; i != count; ++i) {
-    if (!(typeGuardAttr[i].isa<fir::ExactTypeAttr>() ||
-          typeGuardAttr[i].isa<fir::SubclassAttr>() ||
-          typeGuardAttr[i].isa<mlir::UnitAttr>()))
+    if (!mlir::isa<fir::ExactTypeAttr, fir::SubclassAttr, mlir::UnitAttr>(
+            typeGuardAttr[i]))
       return emitOpError("invalid type-case alternative");
   }
   return mlir::success();
@@ -3175,7 +3177,7 @@ void fir::SelectTypeOp::build(mlir::OpBuilder &builder,
 
 mlir::LogicalResult fir::ShapeOp::verify() {
   auto size = getExtents().size();
-  auto shapeTy = getType().dyn_cast<fir::ShapeType>();
+  auto shapeTy = mlir::dyn_cast<fir::ShapeType>(getType());
   assert(shapeTy && "must be a shape type");
   if (shapeTy.getRank() != size)
     return emitOpError("shape type rank mismatch");
@@ -3198,7 +3200,7 @@ mlir::LogicalResult fir::ShapeShiftOp::verify() {
     return emitOpError("incorrect number of args");
   if (size % 2 != 0)
     return emitOpError("requires a multiple of 2 args");
-  auto shapeTy = getType().dyn_cast<fir::ShapeShiftType>();
+  auto shapeTy = mlir::dyn_cast<fir::ShapeShiftType>(getType());
   assert(shapeTy && "must be a shape shift type");
   if (shapeTy.getRank() * 2 != size)
     return emitOpError("shape type rank mismatch");
@@ -3211,7 +3213,7 @@ mlir::LogicalResult fir::ShapeShiftOp::verify() {
 
 mlir::LogicalResult fir::ShiftOp::verify() {
   auto size = getOrigins().size();
-  auto shiftTy = getType().dyn_cast<fir::ShiftType>();
+  auto shiftTy = mlir::dyn_cast<fir::ShiftType>(getType());
   assert(shiftTy && "must be a shift type");
   if (shiftTy.getRank() != size)
     return emitOpError("shift type rank mismatch");
@@ -3251,7 +3253,7 @@ mlir::LogicalResult fir::SliceOp::verify() {
     return emitOpError("incorrect number of args for triple");
   if (size % 3 != 0)
     return emitOpError("requires a multiple of 3 args");
-  auto sliceTy = getType().dyn_cast<fir::SliceType>();
+  auto sliceTy = mlir::dyn_cast<fir::SliceType>(getType());
   assert(sliceTy && "must be a slice type");
   if (sliceTy.getRank() * 3 != size)
     return emitOpError("slice type rank mismatch");
@@ -3309,8 +3311,8 @@ void fir::StoreOp::build(mlir::OpBuilder &builder, mlir::OperationState &result,
 //===----------------------------------------------------------------------===//
 
 inline fir::CharacterType::KindTy stringLitOpGetKind(fir::StringLitOp op) {
-  auto eleTy = op.getType().cast<fir::SequenceType>().getEleTy();
-  return eleTy.cast<fir::CharacterType>().getFKind();
+  auto eleTy = mlir::cast<fir::SequenceType>(op.getType()).getEleTy();
+  return mlir::cast<fir::CharacterType>(eleTy).getFKind();
 }
 
 bool fir::StringLitOp::isWideValue() { return stringLitOpGetKind(*this) != 1; }
@@ -3390,13 +3392,13 @@ mlir::ParseResult fir::StringLitOp::parse(mlir::OpAsmParser &parser,
   llvm::SMLoc trailingTypeLoc;
   if (parser.parseAttribute(val, "fake", attrs))
     return mlir::failure();
-  if (auto v = val.dyn_cast<mlir::StringAttr>())
+  if (auto v = mlir::dyn_cast<mlir::StringAttr>(val))
     result.attributes.push_back(
         builder.getNamedAttr(fir::StringLitOp::value(), v));
-  else if (auto v = val.dyn_cast<mlir::DenseElementsAttr>())
+  else if (auto v = mlir::dyn_cast<mlir::DenseElementsAttr>(val))
     result.attributes.push_back(
         builder.getNamedAttr(fir::StringLitOp::xlist(), v));
-  else if (auto v = val.dyn_cast<mlir::ArrayAttr>())
+  else if (auto v = mlir::dyn_cast<mlir::ArrayAttr>(val))
     result.attributes.push_back(
         builder.getNamedAttr(fir::StringLitOp::xlist(), v));
   else
@@ -3409,7 +3411,7 @@ mlir::ParseResult fir::StringLitOp::parse(mlir::OpAsmParser &parser,
       parser.parseRParen() || parser.getCurrentLocation(&trailingTypeLoc) ||
       parser.parseColonType(type))
     return mlir::failure();
-  auto charTy = type.dyn_cast<fir::CharacterType>();
+  auto charTy = mlir::dyn_cast<fir::CharacterType>(type);
   if (!charTy)
     return parser.emitError(trailingTypeLoc, "must have character type");
   type = fir::CharacterType::get(builder.getContext(), charTy.getFKind(),
@@ -3421,19 +3423,19 @@ mlir::ParseResult fir::StringLitOp::parse(mlir::OpAsmParser &parser,
 
 void fir::StringLitOp::print(mlir::OpAsmPrinter &p) {
   p << ' ' << getValue() << '(';
-  p << getSize().cast<mlir::IntegerAttr>().getValue() << ") : ";
+  p << mlir::cast<mlir::IntegerAttr>(getSize()).getValue() << ") : ";
   p.printType(getType());
 }
 
 mlir::LogicalResult fir::StringLitOp::verify() {
-  if (getSize().cast<mlir::IntegerAttr>().getValue().isNegative())
+  if (mlir::cast<mlir::IntegerAttr>(getSize()).getValue().isNegative())
     return emitOpError("size must be non-negative");
   if (auto xl = getOperation()->getAttr(fir::StringLitOp::xlist())) {
-    if (auto xList = xl.dyn_cast<mlir::ArrayAttr>()) {
+    if (auto xList = mlir::dyn_cast<mlir::ArrayAttr>(xl)) {
       for (auto a : xList)
-        if (!a.isa<mlir::IntegerAttr>())
+        if (!mlir::isa<mlir::IntegerAttr>(a))
           return emitOpError("values in initializer must be integers");
-    } else if (xl.isa<mlir::DenseElementsAttr>()) {
+    } else if (mlir::isa<mlir::DenseElementsAttr>(xl)) {
       // do nothing
     } else {
       return emitOpError("has unexpected attribute");
@@ -3448,7 +3450,7 @@ mlir::LogicalResult fir::StringLitOp::verify() {
 
 mlir::LogicalResult fir::UnboxProcOp::verify() {
   if (auto eleTy = fir::dyn_cast_ptrEleTy(getRefTuple().getType()))
-    if (eleTy.isa<mlir::TupleType>())
+    if (mlir::isa<mlir::TupleType>(eleTy))
       return mlir::success();
   return emitOpError("second output argument has bad type");
 }
@@ -3527,7 +3529,7 @@ void fir::IfOp::getEntrySuccessorRegions(
 void fir::IfOp::getRegionInvocationBounds(
     llvm::ArrayRef<mlir::Attribute> operands,
     llvm::SmallVectorImpl<mlir::InvocationBounds> &invocationBounds) {
-  if (auto cond = operands[0].dyn_cast_or_null<mlir::BoolAttr>()) {
+  if (auto cond = mlir::dyn_cast_or_null<mlir::BoolAttr>(operands[0])) {
     // If the condition is known, then one region is known to be executed once
     // and the other zero times.
     invocationBounds.emplace_back(0, cond.getValue() ? 1 : 0);
@@ -3646,8 +3648,8 @@ void fir::BoxOffsetOp::build(mlir::OpBuilder &builder,
 //===----------------------------------------------------------------------===//
 
 mlir::ParseResult fir::isValidCaseAttr(mlir::Attribute attr) {
-  if (attr.isa<mlir::UnitAttr, fir::ClosedIntervalAttr, fir::PointIntervalAttr,
-               fir::LowerBoundAttr, fir::UpperBoundAttr>())
+  if (mlir::isa<mlir::UnitAttr, fir::ClosedIntervalAttr, fir::PointIntervalAttr,
+                fir::LowerBoundAttr, fir::UpperBoundAttr>(attr))
     return mlir::success();
   return mlir::failure();
 }
@@ -3657,9 +3659,9 @@ unsigned fir::getCaseArgumentOffset(llvm::ArrayRef<mlir::Attribute> cases,
   unsigned o = 0;
   for (unsigned i = 0; i < dest; ++i) {
     auto &attr = cases[i];
-    if (!attr.dyn_cast_or_null<mlir::UnitAttr>()) {
+    if (!mlir::dyn_cast_or_null<mlir::UnitAttr>(attr)) {
       ++o;
-      if (attr.dyn_cast_or_null<fir::ClosedIntervalAttr>())
+      if (mlir::dyn_cast_or_null<fir::ClosedIntervalAttr>(attr))
         ++o;
     }
   }
@@ -3722,7 +3724,7 @@ fir::GlobalOp fir::createGlobalOp(mlir::Location loc, mlir::ModuleOp module,
 bool fir::hasHostAssociationArgument(mlir::func::FuncOp func) {
   if (auto allArgAttrs = func.getAllArgAttrs())
     for (auto attr : allArgAttrs)
-      if (auto dict = attr.template dyn_cast_or_null<mlir::DictionaryAttr>())
+      if (auto dict = mlir::dyn_cast_or_null<mlir::DictionaryAttr>(attr))
         if (dict.get(fir::getHostAssocAttrName()))
           return true;
   return false;
@@ -3772,7 +3774,7 @@ valueCheckFirAttributes(mlir::Value value,
   };
   // If this is a fir.box that was loaded, the fir attributes will be on the
   // related fir.ref<fir.box> creation.
-  if (value.getType().isa<fir::BoxType>())
+  if (mlir::isa<fir::BoxType>(value.getType()))
     if (auto definingOp = value.getDefiningOp())
       if (auto loadOp = mlir::dyn_cast<fir::LoadOp>(definingOp))
         value = loadOp.getMemref();
@@ -3837,10 +3839,10 @@ bool fir::anyFuncArgsHaveAttr(mlir::func::FuncOp func, llvm::StringRef attr) {
 std::optional<std::int64_t> fir::getIntIfConstant(mlir::Value value) {
   if (auto *definingOp = value.getDefiningOp()) {
     if (auto cst = mlir::dyn_cast<mlir::arith::ConstantOp>(definingOp))
-      if (auto intAttr = cst.getValue().dyn_cast<mlir::IntegerAttr>())
+      if (auto intAttr = mlir::dyn_cast<mlir::IntegerAttr>(cst.getValue()))
         return intAttr.getInt();
     if (auto llConstOp = mlir::dyn_cast<mlir::LLVM::ConstantOp>(definingOp))
-      if (auto attr = llConstOp.getValue().dyn_cast<mlir::IntegerAttr>())
+      if (auto attr = mlir::dyn_cast<mlir::IntegerAttr>(llConstOp.getValue()))
         return attr.getValue().getSExtValue();
   }
   return {};
@@ -4002,15 +4004,15 @@ mlir::LogicalResult fir::CUDAKernelOp::verify() {
 mlir::LogicalResult fir::CUDAAllocateOp::verify() {
   if (getPinned() && getStream())
     return emitOpError("pinned and stream cannot appears at the same time");
-  if (!fir::unwrapRefType(getBox().getType()).isa<fir::BaseBoxType>())
+  if (!mlir::isa<fir::BaseBoxType>(fir::unwrapRefType(getBox().getType())))
     return emitOpError(
         "expect box to be a reference to a class or box type value");
   if (getSource() &&
-      !fir::unwrapRefType(getSource().getType()).isa<fir::BaseBoxType>())
+      !mlir::isa<fir::BaseBoxType>(fir::unwrapRefType(getSource().getType())))
     return emitOpError(
         "expect source to be a reference to/or a class or box type value");
   if (getErrmsg() &&
-      !fir::unwrapRefType(getErrmsg().getType()).isa<fir::BoxType>())
+      !mlir::isa<fir::BoxType>(fir::unwrapRefType(getErrmsg().getType())))
     return emitOpError(
         "expect errmsg to be a reference to/or a box type value");
   if (getErrmsg() && !getHasStat())
@@ -4019,11 +4021,11 @@ mlir::LogicalResult fir::CUDAAllocateOp::verify() {
 }
 
 mlir::LogicalResult fir::CUDADeallocateOp::verify() {
-  if (!fir::unwrapRefType(getBox().getType()).isa<fir::BaseBoxType>())
+  if (!mlir::isa<fir::BaseBoxType>(fir::unwrapRefType(getBox().getType())))
     return emitOpError(
         "expect box to be a reference to class or box type value");
   if (getErrmsg() &&
-      !fir::unwrapRefType(getErrmsg().getType()).isa<fir::BoxType>())
+      !mlir::isa<fir::BoxType>(fir::unwrapRefType(getErrmsg().getType())))
     return emitOpError(
         "expect errmsg to be a reference to/or a box type value");
   if (getErrmsg() && !getHasStat())
diff --git a/flang/lib/Optimizer/Dialect/FIRType.cpp b/flang/lib/Optimizer/Dialect/FIRType.cpp
index 5c4cad6d2083..d9c387ad950e 100644
--- a/flang/lib/Optimizer/Dialect/FIRType.cpp
+++ b/flang/lib/Optimizer/Dialect/FIRType.cpp
@@ -61,14 +61,13 @@ TYPE parseTypeSingleton(mlir::AsmParser &parser) {
 /// Is `ty` a standard or FIR integer type?
 static bool isaIntegerType(mlir::Type ty) {
   // TODO: why aren't we using isa_integer? investigatation required.
-  return ty.isa<mlir::IntegerType>() || ty.isa<fir::IntegerType>();
+  return mlir::isa<mlir::IntegerType, fir::IntegerType>(ty);
 }
 
 bool verifyRecordMemberType(mlir::Type ty) {
-  return !(ty.isa<BoxCharType>() || ty.isa<ShapeType>() ||
-           ty.isa<ShapeShiftType>() || ty.isa<ShiftType>() ||
-           ty.isa<SliceType>() || ty.isa<FieldType>() || ty.isa<LenType>() ||
-           ty.isa<ReferenceType>() || ty.isa<TypeDescType>());
+  return !mlir::isa<BoxCharType, ShapeType, ShapeShiftType, ShiftType,
+                    SliceType, FieldType, LenType, ReferenceType, TypeDescType>(
+      ty);
 }
 
 bool verifySameLists(llvm::ArrayRef<RecordType::TypePair> a1,
@@ -194,7 +193,7 @@ bool isa_std_type(mlir::Type t) {
 }
 
 bool isa_fir_or_std_type(mlir::Type t) {
-  if (auto funcType = t.dyn_cast<mlir::FunctionType>())
+  if (auto funcType = mlir::dyn_cast<mlir::FunctionType>(t))
     return llvm::all_of(funcType.getInputs(), isa_fir_or_std_type) &&
            llvm::all_of(funcType.getResults(), isa_fir_or_std_type);
   return isa_fir_type(t) || isa_std_type(t);
@@ -203,7 +202,7 @@ bool isa_fir_or_std_type(mlir::Type t) {
 mlir::Type getDerivedType(mlir::Type ty) {
   return llvm::TypeSwitch<mlir::Type, mlir::Type>(ty)
       .Case<fir::PointerType, fir::HeapType, fir::SequenceType>([](auto p) {
-        if (auto seq = p.getEleTy().template dyn_cast<fir::SequenceType>())
+        if (auto seq = mlir::dyn_cast<fir::SequenceType>(p.getEleTy()))
           return seq.getEleTy();
         return p.getEleTy();
       })
@@ -228,12 +227,12 @@ mlir::Type dyn_cast_ptrOrBoxEleTy(mlir::Type t) {
 
 static bool hasDynamicSize(fir::RecordType recTy) {
   for (auto field : recTy.getTypeList()) {
-    if (auto arr = field.second.dyn_cast<fir::SequenceType>()) {
+    if (auto arr = mlir::dyn_cast<fir::SequenceType>(field.second)) {
       if (sequenceWithNonConstantShape(arr))
         return true;
     } else if (characterWithDynamicLen(field.second)) {
       return true;
-    } else if (auto rec = field.second.dyn_cast<fir::RecordType>()) {
+    } else if (auto rec = mlir::dyn_cast<fir::RecordType>(field.second)) {
       if (hasDynamicSize(rec))
         return true;
     }
@@ -242,14 +241,14 @@ static bool hasDynamicSize(fir::RecordType recTy) {
 }
 
 bool hasDynamicSize(mlir::Type t) {
-  if (auto arr = t.dyn_cast<fir::SequenceType>()) {
+  if (auto arr = mlir::dyn_cast<fir::SequenceType>(t)) {
     if (sequenceWithNonConstantShape(arr))
       return true;
     t = arr.getEleTy();
   }
   if (characterWithDynamicLen(t))
     return true;
-  if (auto rec = t.dyn_cast<fir::RecordType>())
+  if (auto rec = mlir::dyn_cast<fir::RecordType>(t))
     return hasDynamicSize(rec);
   return false;
 }
@@ -269,33 +268,33 @@ mlir::Type extractSequenceType(mlir::Type ty) {
 bool isPointerType(mlir::Type ty) {
   if (auto refTy = fir::dyn_cast_ptrEleTy(ty))
     ty = refTy;
-  if (auto boxTy = ty.dyn_cast<fir::BaseBoxType>())
-    return boxTy.getEleTy().isa<fir::PointerType>();
+  if (auto boxTy = mlir::dyn_cast<fir::BaseBoxType>(ty))
+    return mlir::isa<fir::PointerType>(boxTy.getEleTy());
   return false;
 }
 
 bool isAllocatableType(mlir::Type ty) {
   if (auto refTy = fir::dyn_cast_ptrEleTy(ty))
     ty = refTy;
-  if (auto boxTy = ty.dyn_cast<fir::BaseBoxType>())
-    return boxTy.getEleTy().isa<fir::HeapType>();
+  if (auto boxTy = mlir::dyn_cast<fir::BaseBoxType>(ty))
+    return mlir::isa<fir::HeapType>(boxTy.getEleTy());
   return false;
 }
 
 bool isBoxNone(mlir::Type ty) {
-  if (auto box = ty.dyn_cast<fir::BoxType>())
-    return box.getEleTy().isa<mlir::NoneType>();
+  if (auto box = mlir::dyn_cast<fir::BoxType>(ty))
+    return mlir::isa<mlir::NoneType>(box.getEleTy());
   return false;
 }
 
 bool isBoxedRecordType(mlir::Type ty) {
   if (auto refTy = fir::dyn_cast_ptrEleTy(ty))
     ty = refTy;
-  if (auto boxTy = ty.dyn_cast<fir::BoxType>()) {
-    if (boxTy.getEleTy().isa<fir::RecordType>())
+  if (auto boxTy = mlir::dyn_cast<fir::BoxType>(ty)) {
+    if (mlir::isa<fir::RecordType>(boxTy.getEleTy()))
       return true;
     mlir::Type innerType = boxTy.unwrapInnerType();
-    return innerType && innerType.isa<fir::RecordType>();
+    return innerType && mlir::isa<fir::RecordType>(innerType);
   }
   return false;
 }
@@ -303,13 +302,13 @@ bool isBoxedRecordType(mlir::Type ty) {
 bool isScalarBoxedRecordType(mlir::Type ty) {
   if (auto refTy = fir::dyn_cast_ptrEleTy(ty))
     ty = refTy;
-  if (auto boxTy = ty.dyn_cast<fir::BaseBoxType>()) {
-    if (boxTy.getEleTy().isa<fir::RecordType>())
+  if (auto boxTy = mlir::dyn_cast<fir::BaseBoxType>(ty)) {
+    if (mlir::isa<fir::RecordType>(boxTy.getEleTy()))
       return true;
-    if (auto heapTy = boxTy.getEleTy().dyn_cast<fir::HeapType>())
-      return heapTy.getEleTy().isa<fir::RecordType>();
-    if (auto ptrTy = boxTy.getEleTy().dyn_cast<fir::PointerType>())
-      return ptrTy.getEleTy().isa<fir::RecordType>();
+    if (auto heapTy = mlir::dyn_cast<fir::HeapType>(boxTy.getEleTy()))
+      return mlir::isa<fir::RecordType>(heapTy.getEleTy());
+    if (auto ptrTy = mlir::dyn_cast<fir::PointerType>(boxTy.getEleTy()))
+      return mlir::isa<fir::RecordType>(ptrTy.getEleTy());
   }
   return false;
 }
@@ -363,10 +362,10 @@ bool isPolymorphicType(mlir::Type ty) {
 bool isUnlimitedPolymorphicType(mlir::Type ty) {
   // CLASS(*)
   if (auto clTy = mlir::dyn_cast<fir::ClassType>(fir::unwrapRefType(ty))) {
-    if (clTy.getEleTy().isa<mlir::NoneType>())
+    if (mlir::isa<mlir::NoneType>(clTy.getEleTy()))
       return true;
     mlir::Type innerType = clTy.unwrapInnerType();
-    return innerType && innerType.isa<mlir::NoneType>();
+    return innerType && mlir::isa<mlir::NoneType>(innerType);
   }
   // TYPE(*)
   return isAssumedType(ty);
@@ -376,7 +375,7 @@ mlir::Type unwrapInnerType(mlir::Type ty) {
   return llvm::TypeSwitch<mlir::Type, mlir::Type>(ty)
       .Case<fir::PointerType, fir::HeapType, fir::SequenceType>([](auto t) {
         mlir::Type eleTy = t.getEleTy();
-        if (auto seqTy = eleTy.dyn_cast<fir::SequenceType>())
+        if (auto seqTy = mlir::dyn_cast<fir::SequenceType>(eleTy))
           return seqTy.getEleTy();
         return eleTy;
       })
@@ -385,13 +384,14 @@ mlir::Type unwrapInnerType(mlir::Type ty) {
 }
 
 bool isRecordWithAllocatableMember(mlir::Type ty) {
-  if (auto recTy = ty.dyn_cast<fir::RecordType>())
+  if (auto recTy = mlir::dyn_cast<fir::RecordType>(ty))
     for (auto [field, memTy] : recTy.getTypeList()) {
       if (fir::isAllocatableType(memTy))
         return true;
       // A record type cannot recursively include itself as a direct member.
       // There must be an intervening `ptr` type, so recursion is safe here.
-      if (memTy.isa<fir::RecordType>() && isRecordWithAllocatableMember(memTy))
+      if (mlir::isa<fir::RecordType>(memTy) &&
+          isRecordWithAllocatableMember(memTy))
         return true;
     }
   return false;
@@ -399,11 +399,12 @@ bool isRecordWithAllocatableMember(mlir::Type ty) {
 
 bool isRecordWithDescriptorMember(mlir::Type ty) {
   ty = unwrapSequenceType(ty);
-  if (auto recTy = ty.dyn_cast<fir::RecordType>())
+  if (auto recTy = mlir::dyn_cast<fir::RecordType>(ty))
     for (auto [field, memTy] : recTy.getTypeList()) {
       if (mlir::isa<fir::BaseBoxType>(memTy))
         return true;
-      if (memTy.isa<fir::RecordType>() && isRecordWithDescriptorMember(memTy))
+      if (mlir::isa<fir::RecordType>(memTy) &&
+          isRecordWithDescriptorMember(memTy))
         return true;
     }
   return false;
@@ -412,7 +413,7 @@ bool isRecordWithDescriptorMember(mlir::Type ty) {
 mlir::Type unwrapAllRefAndSeqType(mlir::Type ty) {
   while (true) {
     mlir::Type nt = unwrapSequenceType(unwrapRefType(ty));
-    if (auto vecTy = nt.dyn_cast<fir::VectorType>())
+    if (auto vecTy = mlir::dyn_cast<fir::VectorType>(nt))
       nt = vecTy.getEleTy();
     if (nt == ty)
       return ty;
@@ -421,11 +422,11 @@ mlir::Type unwrapAllRefAndSeqType(mlir::Type ty) {
 }
 
 mlir::Type unwrapSeqOrBoxedSeqType(mlir::Type ty) {
-  if (auto seqTy = ty.dyn_cast<fir::SequenceType>())
+  if (auto seqTy = mlir::dyn_cast<fir::SequenceType>(ty))
     return seqTy.getEleTy();
-  if (auto boxTy = ty.dyn_cast<fir::BaseBoxType>()) {
+  if (auto boxTy = mlir::dyn_cast<fir::BaseBoxType>(ty)) {
     auto eleTy = unwrapRefType(boxTy.getEleTy());
-    if (auto seqTy = eleTy.dyn_cast<fir::SequenceType>())
+    if (auto seqTy = mlir::dyn_cast<fir::SequenceType>(eleTy))
       return seqTy.getEleTy();
   }
   return ty;
@@ -433,7 +434,7 @@ mlir::Type unwrapSeqOrBoxedSeqType(mlir::Type ty) {
 
 unsigned getBoxRank(mlir::Type boxTy) {
   auto eleTy = fir::dyn_cast_ptrOrBoxEleTy(boxTy);
-  if (auto seqTy = eleTy.dyn_cast<fir::SequenceType>())
+  if (auto seqTy = mlir::dyn_cast<fir::SequenceType>(eleTy))
     return seqTy.getDimension();
   return 0;
 }
@@ -441,7 +442,7 @@ unsigned getBoxRank(mlir::Type boxTy) {
 /// Return the ISO_C_BINDING intrinsic module value of type \p ty.
 int getTypeCode(mlir::Type ty, const fir::KindMapping &kindMap) {
   unsigned width = 0;
-  if (mlir::IntegerType intTy = ty.dyn_cast<mlir::IntegerType>()) {
+  if (mlir::IntegerType intTy = mlir::dyn_cast<mlir::IntegerType>(ty)) {
     switch (intTy.getWidth()) {
     case 8:
       return CFI_type_int8_t;
@@ -456,7 +457,7 @@ int getTypeCode(mlir::Type ty, const fir::KindMapping &kindMap) {
     }
     llvm_unreachable("unsupported integer type");
   }
-  if (fir::LogicalType logicalTy = ty.dyn_cast<fir::LogicalType>()) {
+  if (fir::LogicalType logicalTy = mlir::dyn_cast<fir::LogicalType>(ty)) {
     switch (kindMap.getLogicalBitsize(logicalTy.getFKind())) {
     case 8:
       return CFI_type_Bool;
@@ -469,7 +470,7 @@ int getTypeCode(mlir::Type ty, const fir::KindMapping &kindMap) {
     }
     llvm_unreachable("unsupported logical type");
   }
-  if (mlir::FloatType floatTy = ty.dyn_cast<mlir::FloatType>()) {
+  if (mlir::FloatType floatTy = mlir::dyn_cast<mlir::FloatType>(ty)) {
     switch (floatTy.getWidth()) {
     case 16:
       return floatTy.isBF16() ? CFI_type_bfloat : CFI_type_half_float;
@@ -485,13 +486,14 @@ int getTypeCode(mlir::Type ty, const fir::KindMapping &kindMap) {
     llvm_unreachable("unsupported real type");
   }
   if (fir::isa_complex(ty)) {
-    if (mlir::ComplexType complexTy = ty.dyn_cast<mlir::ComplexType>()) {
+    if (mlir::ComplexType complexTy = mlir::dyn_cast<mlir::ComplexType>(ty)) {
       mlir::FloatType floatTy =
-          complexTy.getElementType().cast<mlir::FloatType>();
+          mlir::cast<mlir::FloatType>(complexTy.getElementType());
       if (floatTy.isBF16())
         return CFI_type_bfloat_Complex;
       width = floatTy.getWidth();
-    } else if (fir::ComplexType complexTy = ty.dyn_cast<fir::ComplexType>()) {
+    } else if (fir::ComplexType complexTy =
+                   mlir::dyn_cast<fir::ComplexType>(ty)) {
       auto FKind = complexTy.getFKind();
       if (FKind == 3)
         return CFI_type_bfloat_Complex;
@@ -511,7 +513,7 @@ int getTypeCode(mlir::Type ty, const fir::KindMapping &kindMap) {
     }
     llvm_unreachable("unsupported complex size");
   }
-  if (fir::CharacterType charTy = ty.dyn_cast<fir::CharacterType>()) {
+  if (fir::CharacterType charTy = mlir::dyn_cast<fir::CharacterType>(ty)) {
     switch (kindMap.getCharacterBitsize(charTy.getFKind())) {
     case 8:
       return CFI_type_char;
@@ -524,7 +526,7 @@ int getTypeCode(mlir::Type ty, const fir::KindMapping &kindMap) {
   }
   if (fir::isa_ref_type(ty))
     return CFI_type_cptr;
-  if (ty.isa<fir::RecordType>())
+  if (mlir::isa<fir::RecordType>(ty))
     return CFI_type_struct;
   llvm_unreachable("unsupported type");
 }
@@ -542,12 +544,12 @@ std::string getTypeAsString(mlir::Type ty, const fir::KindMapping &kindMap,
         name << "idx";
       } else if (ty.isIntOrIndex()) {
         name << 'i' << ty.getIntOrFloatBitWidth();
-      } else if (ty.isa<mlir::FloatType>()) {
+      } else if (mlir::isa<mlir::FloatType>(ty)) {
         name << 'f' << ty.getIntOrFloatBitWidth();
       } else if (fir::isa_complex(ty)) {
         name << 'z';
         if (auto cplxTy = mlir::dyn_cast_or_null<mlir::ComplexType>(ty)) {
-          auto floatTy = cplxTy.getElementType().cast<mlir::FloatType>();
+          auto floatTy = mlir::cast<mlir::FloatType>(cplxTy.getElementType());
           name << floatTy.getWidth();
         } else if (auto cplxTy = mlir::dyn_cast_or_null<fir::ComplexType>(ty)) {
           name << kindMap.getRealBitsize(cplxTy.getFKind());
@@ -644,7 +646,7 @@ static llvm::SmallPtrSet<detail::RecordTypeStorage const *, 4>
 } // namespace
 
 void fir::verifyIntegralType(mlir::Type type) {
-  if (isaIntegerType(type) || type.isa<mlir::IndexType>())
+  if (isaIntegerType(type) || mlir::isa<mlir::IndexType>(type))
     return;
   llvm::report_fatal_error("expected integral type");
 }
@@ -656,9 +658,9 @@ void fir::printFirType(FIROpsDialect *, mlir::Type ty,
 }
 
 bool fir::isa_unknown_size_box(mlir::Type t) {
-  if (auto boxTy = t.dyn_cast<fir::BaseBoxType>()) {
+  if (auto boxTy = mlir::dyn_cast<fir::BaseBoxType>(t)) {
     auto valueType = fir::unwrapPassByRefType(boxTy);
-    if (auto seqTy = valueType.dyn_cast<fir::SequenceType>())
+    if (auto seqTy = mlir::dyn_cast<fir::SequenceType>(valueType))
       if (seqTy.hasUnknownShape())
         return true;
   }
@@ -684,18 +686,18 @@ void fir::BoxProcType::print(mlir::AsmPrinter &printer) const {
 mlir::LogicalResult
 BoxProcType::verify(llvm::function_ref<mlir::InFlightDiagnostic()> emitError,
                     mlir::Type eleTy) {
-  if (eleTy.isa<mlir::FunctionType>())
+  if (mlir::isa<mlir::FunctionType>(eleTy))
     return mlir::success();
-  if (auto refTy = eleTy.dyn_cast<ReferenceType>())
-    if (refTy.isa<mlir::FunctionType>())
+  if (auto refTy = mlir::dyn_cast<ReferenceType>(eleTy))
+    if (mlir::isa<mlir::FunctionType>(refTy))
       return mlir::success();
   return emitError() << "invalid type for boxproc" << eleTy << '\n';
 }
 
 static bool cannotBePointerOrHeapElementType(mlir::Type eleTy) {
-  return eleTy.isa<BoxType, BoxCharType, BoxProcType, ShapeType, ShapeShiftType,
+  return mlir::isa<BoxType, BoxCharType, BoxProcType, ShapeType, ShapeShiftType,
                    SliceType, FieldType, LenType, HeapType, PointerType,
-                   ReferenceType, TypeDescType>();
+                   ReferenceType, TypeDescType>(eleTy);
 }
 
 //===----------------------------------------------------------------------===//
@@ -705,7 +707,7 @@ static bool cannotBePointerOrHeapElementType(mlir::Type eleTy) {
 mlir::LogicalResult
 fir::BoxType::verify(llvm::function_ref<mlir::InFlightDiagnostic()> emitError,
                      mlir::Type eleTy) {
-  if (eleTy.isa<fir::BaseBoxType>())
+  if (mlir::isa<fir::BaseBoxType>(eleTy))
     return emitError() << "invalid element type\n";
   // TODO
   return mlir::success();
@@ -774,10 +776,10 @@ void fir::CharacterType::print(mlir::AsmPrinter &printer) const {
 mlir::LogicalResult
 fir::ClassType::verify(llvm::function_ref<mlir::InFlightDiagnostic()> emitError,
                        mlir::Type eleTy) {
-  if (eleTy.isa<fir::RecordType, fir::SequenceType, fir::HeapType,
+  if (mlir::isa<fir::RecordType, fir::SequenceType, fir::HeapType,
                 fir::PointerType, mlir::NoneType, mlir::IntegerType,
                 mlir::FloatType, fir::CharacterType, fir::LogicalType,
-                fir::ComplexType, mlir::ComplexType>())
+                fir::ComplexType, mlir::ComplexType>(eleTy))
     return mlir::success();
   return emitError() << "invalid element type\n";
 }
@@ -1048,8 +1050,8 @@ void fir::ReferenceType::print(mlir::AsmPrinter &printer) const {
 mlir::LogicalResult fir::ReferenceType::verify(
     llvm::function_ref<mlir::InFlightDiagnostic()> emitError,
     mlir::Type eleTy) {
-  if (eleTy.isa<ShapeType, ShapeShiftType, SliceType, FieldType, LenType,
-                ReferenceType, TypeDescType>())
+  if (mlir::isa<ShapeType, ShapeShiftType, SliceType, FieldType, LenType,
+                ReferenceType, TypeDescType>(eleTy))
     return emitError() << "cannot build a reference to type: " << eleTy << '\n';
   return mlir::success();
 }
@@ -1124,9 +1126,9 @@ mlir::LogicalResult fir::SequenceType::verify(
     llvm::ArrayRef<int64_t> shape, mlir::Type eleTy,
     mlir::AffineMapAttr layoutMap) {
   // DIMENSION attribute can only be applied to an intrinsic or record type
-  if (eleTy.isa<BoxType, BoxCharType, BoxProcType, ShapeType, ShapeShiftType,
+  if (mlir::isa<BoxType, BoxCharType, BoxProcType, ShapeType, ShapeShiftType,
                 ShiftType, SliceType, FieldType, LenType, HeapType, PointerType,
-                ReferenceType, TypeDescType, SequenceType>())
+                ReferenceType, TypeDescType, SequenceType>(eleTy))
     return emitError() << "cannot build an array of this element type: "
                        << eleTy << '\n';
   return mlir::success();
@@ -1197,9 +1199,9 @@ void fir::TypeDescType::print(mlir::AsmPrinter &printer) const {
 mlir::LogicalResult fir::TypeDescType::verify(
     llvm::function_ref<mlir::InFlightDiagnostic()> emitError,
     mlir::Type eleTy) {
-  if (eleTy.isa<BoxType, BoxCharType, BoxProcType, ShapeType, ShapeShiftType,
+  if (mlir::isa<BoxType, BoxCharType, BoxProcType, ShapeType, ShapeShiftType,
                 ShiftType, SliceType, FieldType, LenType, ReferenceType,
-                TypeDescType>())
+                TypeDescType>(eleTy))
     return emitError() << "cannot build a type descriptor of type: " << eleTy
                        << '\n';
   return mlir::success();
@@ -1236,10 +1238,10 @@ bool fir::VectorType::isValidElementType(mlir::Type t) {
 }
 
 bool fir::isCharacterProcedureTuple(mlir::Type ty, bool acceptRawFunc) {
-  mlir::TupleType tuple = ty.dyn_cast<mlir::TupleType>();
+  mlir::TupleType tuple = mlir::dyn_cast<mlir::TupleType>(ty);
   return tuple && tuple.size() == 2 &&
-         (tuple.getType(0).isa<fir::BoxProcType>() ||
-          (acceptRawFunc && tuple.getType(0).isa<mlir::FunctionType>())) &&
+         (mlir::isa<fir::BoxProcType>(tuple.getType(0)) ||
+          (acceptRawFunc && mlir::isa<mlir::FunctionType>(tuple.getType(0)))) &&
          fir::isa_integer(tuple.getType(1));
 }
 
@@ -1247,7 +1249,8 @@ bool fir::hasAbstractResult(mlir::FunctionType ty) {
   if (ty.getNumResults() == 0)
     return false;
   auto resultType = ty.getResult(0);
-  return resultType.isa<fir::SequenceType, fir::BaseBoxType, fir::RecordType>();
+  return mlir::isa<fir::SequenceType, fir::BaseBoxType, fir::RecordType>(
+      resultType);
 }
 
 /// Convert llvm::Type::TypeID to mlir::Type. \p kind is provided for error
diff --git a/flang/lib/Optimizer/Dialect/FortranVariableInterface.cpp b/flang/lib/Optimizer/Dialect/FortranVariableInterface.cpp
index 94f1689dfb05..70b1a2f3d844 100644
--- a/flang/lib/Optimizer/Dialect/FortranVariableInterface.cpp
+++ b/flang/lib/Optimizer/Dialect/FortranVariableInterface.cpp
@@ -18,7 +18,7 @@ mlir::LogicalResult
 fir::FortranVariableOpInterface::verifyDeclareLikeOpImpl(mlir::Value memref) {
   const unsigned numExplicitTypeParams = getExplicitTypeParams().size();
   mlir::Type memType = memref.getType();
-  const bool sourceIsBoxValue = memType.isa<fir::BaseBoxType>();
+  const bool sourceIsBoxValue = mlir::isa<fir::BaseBoxType>(memType);
   const bool sourceIsBoxAddress = fir::isBoxAddress(memType);
   const bool sourceIsBox = sourceIsBoxValue || sourceIsBoxAddress;
   if (isCharacter()) {
@@ -29,7 +29,8 @@ fir::FortranVariableOpInterface::verifyDeclareLikeOpImpl(mlir::Value memref) {
       return emitOpError("must be provided exactly one type parameter when its "
                          "base is a character that is not a box");
 
-  } else if (auto recordType = getElementType().dyn_cast<fir::RecordType>()) {
+  } else if (auto recordType =
+                 mlir::dyn_cast<fir::RecordType>(getElementType())) {
     if (numExplicitTypeParams < recordType.getNumLenParams() && !sourceIsBox)
       return emitOpError("must be provided all the derived type length "
                          "parameters when the base is not a box");
@@ -45,16 +46,16 @@ fir::FortranVariableOpInterface::verifyDeclareLikeOpImpl(mlir::Value memref) {
       if (sourceIsBoxAddress)
         return emitOpError("for box address must not have a shape operand");
       unsigned shapeRank = 0;
-      if (auto shapeType = shape.getType().dyn_cast<fir::ShapeType>()) {
+      if (auto shapeType = mlir::dyn_cast<fir::ShapeType>(shape.getType())) {
         shapeRank = shapeType.getRank();
       } else if (auto shapeShiftType =
-                     shape.getType().dyn_cast<fir::ShapeShiftType>()) {
+                     mlir::dyn_cast<fir::ShapeShiftType>(shape.getType())) {
         shapeRank = shapeShiftType.getRank();
       } else {
         if (!sourceIsBoxValue)
           emitOpError("of array entity with a raw address base must have a "
                       "shape operand that is a shape or shapeshift");
-        shapeRank = shape.getType().cast<fir::ShiftType>().getRank();
+        shapeRank = mlir::cast<fir::ShiftType>(shape.getType()).getRank();
       }
 
       std::optional<unsigned> rank = getRank();
diff --git a/flang/lib/Optimizer/HLFIR/IR/HLFIRDialect.cpp b/flang/lib/Optimizer/HLFIR/IR/HLFIRDialect.cpp
index 08b2b0538c73..0b61c0edce62 100644
--- a/flang/lib/Optimizer/HLFIR/IR/HLFIRDialect.cpp
+++ b/flang/lib/Optimizer/HLFIR/IR/HLFIRDialect.cpp
@@ -84,7 +84,8 @@ bool hlfir::isFortranVariableType(mlir::Type type) {
   return llvm::TypeSwitch<mlir::Type, bool>(type)
       .Case<fir::ReferenceType, fir::PointerType, fir::HeapType>([](auto p) {
         mlir::Type eleType = p.getEleTy();
-        return eleType.isa<fir::BaseBoxType>() || !fir::hasDynamicSize(eleType);
+        return mlir::isa<fir::BaseBoxType>(eleType) ||
+               !fir::hasDynamicSize(eleType);
       })
       .Case<fir::BaseBoxType, fir::BoxCharType>([](auto) { return true; })
       .Case<fir::VectorType>([](auto) { return true; })
@@ -93,15 +94,15 @@ bool hlfir::isFortranVariableType(mlir::Type type) {
 
 bool hlfir::isFortranScalarCharacterType(mlir::Type type) {
   return isFortranScalarCharacterExprType(type) ||
-         type.isa<fir::BoxCharType>() ||
-         fir::unwrapPassByRefType(fir::unwrapRefType(type))
-             .isa<fir::CharacterType>();
+         mlir::isa<fir::BoxCharType>(type) ||
+         mlir::isa<fir::CharacterType>(
+             fir::unwrapPassByRefType(fir::unwrapRefType(type)));
 }
 
 bool hlfir::isFortranScalarCharacterExprType(mlir::Type type) {
-  if (auto exprType = type.dyn_cast<hlfir::ExprType>())
+  if (auto exprType = mlir::dyn_cast<hlfir::ExprType>(type))
     return exprType.isScalar() &&
-           exprType.getElementType().isa<fir::CharacterType>();
+           mlir::isa<fir::CharacterType>(exprType.getElementType());
   return false;
 }
 
@@ -121,8 +122,8 @@ bool hlfir::isFortranScalarNumericalType(mlir::Type type) {
 bool hlfir::isFortranNumericalArrayObject(mlir::Type type) {
   if (isBoxAddressType(type))
     return false;
-  if (auto arrayTy =
-          getFortranElementOrSequenceType(type).dyn_cast<fir::SequenceType>())
+  if (auto arrayTy = mlir::dyn_cast<fir::SequenceType>(
+          getFortranElementOrSequenceType(type)))
     return isFortranScalarNumericalType(arrayTy.getEleTy());
   return false;
 }
@@ -130,8 +131,8 @@ bool hlfir::isFortranNumericalArrayObject(mlir::Type type) {
 bool hlfir::isFortranNumericalOrLogicalArrayObject(mlir::Type type) {
   if (isBoxAddressType(type))
     return false;
-  if (auto arrayTy =
-          getFortranElementOrSequenceType(type).dyn_cast<fir::SequenceType>()) {
+  if (auto arrayTy = mlir::dyn_cast<fir::SequenceType>(
+          getFortranElementOrSequenceType(type))) {
     mlir::Type eleTy = arrayTy.getEleTy();
     return isFortranScalarNumericalType(eleTy) ||
            mlir::isa<fir::LogicalType>(eleTy);
@@ -142,7 +143,8 @@ bool hlfir::isFortranNumericalOrLogicalArrayObject(mlir::Type type) {
 bool hlfir::isFortranArrayObject(mlir::Type type) {
   if (isBoxAddressType(type))
     return false;
-  return !!getFortranElementOrSequenceType(type).dyn_cast<fir::SequenceType>();
+  return !!mlir::dyn_cast<fir::SequenceType>(
+      getFortranElementOrSequenceType(type));
 }
 
 bool hlfir::isPassByRefOrIntegerType(mlir::Type type) {
@@ -151,7 +153,7 @@ bool hlfir::isPassByRefOrIntegerType(mlir::Type type) {
 }
 
 bool hlfir::isI1Type(mlir::Type type) {
-  if (mlir::IntegerType integer = type.dyn_cast<mlir::IntegerType>())
+  if (mlir::IntegerType integer = mlir::dyn_cast<mlir::IntegerType>(type))
     if (integer.getWidth() == 1)
       return true;
   return false;
@@ -160,8 +162,8 @@ bool hlfir::isI1Type(mlir::Type type) {
 bool hlfir::isFortranLogicalArrayObject(mlir::Type type) {
   if (isBoxAddressType(type))
     return false;
-  if (auto arrayTy =
-          getFortranElementOrSequenceType(type).dyn_cast<fir::SequenceType>()) {
+  if (auto arrayTy = mlir::dyn_cast<fir::SequenceType>(
+          getFortranElementOrSequenceType(type))) {
     mlir::Type eleTy = arrayTy.getEleTy();
     return mlir::isa<fir::LogicalType>(eleTy);
   }
diff --git a/flang/lib/Optimizer/HLFIR/IR/HLFIROps.cpp b/flang/lib/Optimizer/HLFIR/IR/HLFIROps.cpp
index 8bad4e445082..0d62ca4954e6 100644
--- a/flang/lib/Optimizer/HLFIR/IR/HLFIROps.cpp
+++ b/flang/lib/Optimizer/HLFIR/IR/HLFIROps.cpp
@@ -74,8 +74,8 @@ getIntrinsicEffects(mlir::Operation *self,
 /// Is this a fir.[ref/ptr/heap]<fir.[box/class]<fir.heap<T>>> type?
 static bool isAllocatableBoxRef(mlir::Type type) {
   fir::BaseBoxType boxType =
-      fir::dyn_cast_ptrEleTy(type).dyn_cast_or_null<fir::BaseBoxType>();
-  return boxType && boxType.getEleTy().isa<fir::HeapType>();
+      mlir::dyn_cast_or_null<fir::BaseBoxType>(fir::dyn_cast_ptrEleTy(type));
+  return boxType && mlir::isa<fir::HeapType>(boxType.getEleTy());
 }
 
 mlir::LogicalResult hlfir::AssignOp::verify() {
@@ -84,7 +84,7 @@ mlir::LogicalResult hlfir::AssignOp::verify() {
     return emitOpError("lhs must be an allocatable when `realloc` is set");
   if (mustKeepLhsLengthInAllocatableAssignment() &&
       !(isAllocatableAssignment() &&
-        hlfir::getFortranElementType(lhsType).isa<fir::CharacterType>()))
+        mlir::isa<fir::CharacterType>(hlfir::getFortranElementType(lhsType))))
     return emitOpError("`realloc` must be set and lhs must be a character "
                        "allocatable when `keep_lhs_length_if_realloc` is set");
   return mlir::success();
@@ -99,13 +99,13 @@ mlir::LogicalResult hlfir::AssignOp::verify() {
 mlir::Type hlfir::DeclareOp::getHLFIRVariableType(mlir::Type inputType,
                                                   bool hasExplicitLowerBounds) {
   mlir::Type type = fir::unwrapRefType(inputType);
-  if (type.isa<fir::BaseBoxType>())
+  if (mlir::isa<fir::BaseBoxType>(type))
     return inputType;
-  if (auto charType = type.dyn_cast<fir::CharacterType>())
+  if (auto charType = mlir::dyn_cast<fir::CharacterType>(type))
     if (charType.hasDynamicLen())
       return fir::BoxCharType::get(charType.getContext(), charType.getFKind());
 
-  auto seqType = type.dyn_cast<fir::SequenceType>();
+  auto seqType = mlir::dyn_cast<fir::SequenceType>(type);
   bool hasDynamicExtents =
       seqType && fir::sequenceWithNonConstantShape(seqType);
   mlir::Type eleType = seqType ? seqType.getEleTy() : type;
@@ -117,7 +117,8 @@ mlir::Type hlfir::DeclareOp::getHLFIRVariableType(mlir::Type inputType,
 }
 
 static bool hasExplicitLowerBounds(mlir::Value shape) {
-  return shape && shape.getType().isa<fir::ShapeShiftType, fir::ShiftType>();
+  return shape &&
+         mlir::isa<fir::ShapeShiftType, fir::ShiftType>(shape.getType());
 }
 
 void hlfir::DeclareOp::build(mlir::OpBuilder &builder,
@@ -288,7 +289,7 @@ mlir::LogicalResult hlfir::DesignateOp::verify() {
   bool hasBoxComponent;
   if (getComponent()) {
     auto component = getComponent().value();
-    auto recType = baseElementType.dyn_cast<fir::RecordType>();
+    auto recType = mlir::dyn_cast<fir::RecordType>(baseElementType);
     if (!recType)
       return emitOpError(
           "component must be provided only when the memref is a derived type");
@@ -300,14 +301,14 @@ mlir::LogicalResult hlfir::DesignateOp::verify() {
     }
     mlir::Type fieldType = recType.getType(fieldIdx);
     mlir::Type componentBaseType = getFortranElementOrSequenceType(fieldType);
-    hasBoxComponent = fieldType.isa<fir::BaseBoxType>();
-    if (componentBaseType.isa<fir::SequenceType>() &&
-        baseType.isa<fir::SequenceType>() &&
+    hasBoxComponent = mlir::isa<fir::BaseBoxType>(fieldType);
+    if (mlir::isa<fir::SequenceType>(componentBaseType) &&
+        mlir::isa<fir::SequenceType>(baseType) &&
         (numSubscripts == 0 || subscriptsRank > 0))
       return emitOpError("indices must be provided and must not contain "
                          "triplets when both memref and component are arrays");
     if (numSubscripts != 0) {
-      if (!componentBaseType.isa<fir::SequenceType>())
+      if (!mlir::isa<fir::SequenceType>(componentBaseType))
         return emitOpError("indices must not be provided if component appears "
                            "and is not an array component");
       if (!getComponentShape())
@@ -315,9 +316,9 @@ mlir::LogicalResult hlfir::DesignateOp::verify() {
             "component_shape must be provided when indexing a component");
       mlir::Type compShapeType = getComponentShape().getType();
       unsigned componentRank =
-          componentBaseType.cast<fir::SequenceType>().getDimension();
-      auto shapeType = compShapeType.dyn_cast<fir::ShapeType>();
-      auto shapeShiftType = compShapeType.dyn_cast<fir::ShapeShiftType>();
+          mlir::cast<fir::SequenceType>(componentBaseType).getDimension();
+      auto shapeType = mlir::dyn_cast<fir::ShapeType>(compShapeType);
+      auto shapeShiftType = mlir::dyn_cast<fir::ShapeShiftType>(compShapeType);
       if (!((shapeType && shapeType.getRank() == componentRank) ||
             (shapeShiftType && shapeShiftType.getRank() == componentRank)))
         return emitOpError("component_shape must be a fir.shape or "
@@ -325,33 +326,33 @@ mlir::LogicalResult hlfir::DesignateOp::verify() {
       if (numSubscripts > componentRank)
         return emitOpError("indices number must match array component rank");
     }
-    if (auto baseSeqType = baseType.dyn_cast<fir::SequenceType>())
+    if (auto baseSeqType = mlir::dyn_cast<fir::SequenceType>(baseType))
       // This case must come first to cover "array%array_comp(i, j)" that has
       // subscripts for the component but whose rank come from the base.
       outputRank = baseSeqType.getDimension();
     else if (numSubscripts != 0)
       outputRank = subscriptsRank;
     else if (auto componentSeqType =
-                 componentBaseType.dyn_cast<fir::SequenceType>())
+                 mlir::dyn_cast<fir::SequenceType>(componentBaseType))
       outputRank = componentSeqType.getDimension();
     outputElementType = fir::unwrapSequenceType(componentBaseType);
   } else {
     outputElementType = baseElementType;
     unsigned baseTypeRank =
-        baseType.isa<fir::SequenceType>()
-            ? baseType.cast<fir::SequenceType>().getDimension()
+        mlir::isa<fir::SequenceType>(baseType)
+            ? mlir::cast<fir::SequenceType>(baseType).getDimension()
             : 0;
     if (numSubscripts != 0) {
       if (baseTypeRank != numSubscripts)
         return emitOpError("indices number must match memref rank");
       outputRank = subscriptsRank;
-    } else if (auto baseSeqType = baseType.dyn_cast<fir::SequenceType>()) {
+    } else if (auto baseSeqType = mlir::dyn_cast<fir::SequenceType>(baseType)) {
       outputRank = baseSeqType.getDimension();
     }
   }
 
   if (!getSubstring().empty()) {
-    if (!outputElementType.isa<fir::CharacterType>())
+    if (!mlir::isa<fir::CharacterType>(outputElementType))
       return emitOpError("memref or component must have character type if "
                          "substring indices are provided");
     if (getSubstring().size() != 2)
@@ -361,16 +362,16 @@ mlir::LogicalResult hlfir::DesignateOp::verify() {
     if (!fir::isa_complex(outputElementType))
       return emitOpError("memref or component must have complex type if "
                          "complex_part is provided");
-    if (auto firCplx = outputElementType.dyn_cast<fir::ComplexType>())
+    if (auto firCplx = mlir::dyn_cast<fir::ComplexType>(outputElementType))
       outputElementType = firCplx.getElementType();
     else
       outputElementType =
-          outputElementType.cast<mlir::ComplexType>().getElementType();
+          mlir::cast<mlir::ComplexType>(outputElementType).getElementType();
   }
   mlir::Type resultBaseType =
       getFortranElementOrSequenceType(getResult().getType());
   unsigned resultRank = 0;
-  if (auto resultSeqType = resultBaseType.dyn_cast<fir::SequenceType>())
+  if (auto resultSeqType = mlir::dyn_cast<fir::SequenceType>(resultBaseType))
     resultRank = resultSeqType.getDimension();
   if (resultRank != outputRank)
     return emitOpError("result type rank is not consistent with operands, "
@@ -380,10 +381,10 @@ mlir::LogicalResult hlfir::DesignateOp::verify() {
   // result type must match the one that was inferred here, except the character
   // length may differ because of substrings.
   if (resultElementType != outputElementType &&
-      !(resultElementType.isa<fir::CharacterType>() &&
-        outputElementType.isa<fir::CharacterType>()) &&
-      !(resultElementType.isa<mlir::FloatType>() &&
-        outputElementType.isa<fir::RealType>()))
+      !(mlir::isa<fir::CharacterType>(resultElementType) &&
+        mlir::isa<fir::CharacterType>(outputElementType)) &&
+      !(mlir::isa<mlir::FloatType>(resultElementType) &&
+        mlir::isa<fir::RealType>(outputElementType)))
     return emitOpError(
                "result element type is not consistent with operands, expected ")
            << outputElementType;
@@ -401,22 +402,22 @@ mlir::LogicalResult hlfir::DesignateOp::verify() {
       return emitOpError("shape must be provided if and only if the result is "
                          "an array that is not a box address");
     if (resultRank != 0) {
-      auto shapeType = getShape().getType().dyn_cast<fir::ShapeType>();
+      auto shapeType = mlir::dyn_cast<fir::ShapeType>(getShape().getType());
       auto shapeShiftType =
-          getShape().getType().dyn_cast<fir::ShapeShiftType>();
+          mlir::dyn_cast<fir::ShapeShiftType>(getShape().getType());
       if (!((shapeType && shapeType.getRank() == resultRank) ||
             (shapeShiftType && shapeShiftType.getRank() == resultRank)))
         return emitOpError("shape must be a fir.shape or fir.shapeshift with "
                            "the rank of the result");
     }
     auto numLenParam = getTypeparams().size();
-    if (outputElementType.isa<fir::CharacterType>()) {
+    if (mlir::isa<fir::CharacterType>(outputElementType)) {
       if (numLenParam != 1)
         return emitOpError("must be provided one length parameter when the "
                            "result is a character");
     } else if (fir::isRecordWithTypeParameters(outputElementType)) {
       if (numLenParam !=
-          outputElementType.cast<fir::RecordType>().getNumLenParams())
+          mlir::cast<fir::RecordType>(outputElementType).getNumLenParams())
         return emitOpError("must be provided the same number of length "
                            "parameters as in the result derived type");
     } else if (numLenParam != 0) {
@@ -434,18 +435,18 @@ mlir::LogicalResult hlfir::DesignateOp::verify() {
 mlir::LogicalResult hlfir::ParentComponentOp::verify() {
   mlir::Type baseType =
       hlfir::getFortranElementOrSequenceType(getMemref().getType());
-  auto maybeInputSeqType = baseType.dyn_cast<fir::SequenceType>();
+  auto maybeInputSeqType = mlir::dyn_cast<fir::SequenceType>(baseType);
   unsigned inputTypeRank =
       maybeInputSeqType ? maybeInputSeqType.getDimension() : 0;
   unsigned shapeRank = 0;
   if (mlir::Value shape = getShape())
-    if (auto shapeType = shape.getType().dyn_cast<fir::ShapeType>())
+    if (auto shapeType = mlir::dyn_cast<fir::ShapeType>(shape.getType()))
       shapeRank = shapeType.getRank();
   if (inputTypeRank != shapeRank)
     return emitOpError(
         "must be provided a shape if and only if the base is an array");
   mlir::Type outputBaseType = hlfir::getFortranElementOrSequenceType(getType());
-  auto maybeOutputSeqType = outputBaseType.dyn_cast<fir::SequenceType>();
+  auto maybeOutputSeqType = mlir::dyn_cast<fir::SequenceType>(outputBaseType);
   unsigned outputTypeRank =
       maybeOutputSeqType ? maybeOutputSeqType.getDimension() : 0;
   if (inputTypeRank != outputTypeRank)
@@ -459,23 +460,23 @@ mlir::LogicalResult hlfir::ParentComponentOp::verify() {
           return emitOpError(
               "result type extents are inconsistent with memref type");
   fir::RecordType baseRecType =
-      hlfir::getFortranElementType(baseType).dyn_cast<fir::RecordType>();
-  fir::RecordType outRecType =
-      hlfir::getFortranElementType(outputBaseType).dyn_cast<fir::RecordType>();
+      mlir::dyn_cast<fir::RecordType>(hlfir::getFortranElementType(baseType));
+  fir::RecordType outRecType = mlir::dyn_cast<fir::RecordType>(
+      hlfir::getFortranElementType(outputBaseType));
   if (!baseRecType || !outRecType)
     return emitOpError("result type and input type must be derived types");
 
   // Note: result should not be a fir.class: its dynamic type is being set to
   // the parent type and allowing fir.class would break the operation codegen:
   // it would keep the input dynamic type.
-  if (getType().isa<fir::ClassType>())
+  if (mlir::isa<fir::ClassType>(getType()))
     return emitOpError("result type must not be polymorphic");
 
   // The array results are known to not be dis-contiguous in most cases (the
   // exception being if the parent type was extended by a type without any
   // components): require a fir.box to be used for the result to carry the
   // strides.
-  if (!getType().isa<fir::BoxType>() &&
+  if (!mlir::isa<fir::BoxType>(getType()) &&
       (outputTypeRank != 0 || fir::isRecordWithTypeParameters(outRecType)))
     return emitOpError("result type must be a fir.box if the result is an "
                        "array or has length parameters");
@@ -496,9 +497,8 @@ verifyLogicalReductionOp(LogicalReductionOp reductionOp) {
   mlir::Value mask = reductionOp->getMask();
   mlir::Value dim = reductionOp->getDim();
 
-  fir::SequenceType maskTy =
-      hlfir::getFortranElementOrSequenceType(mask.getType())
-          .cast<fir::SequenceType>();
+  fir::SequenceType maskTy = mlir::cast<fir::SequenceType>(
+      hlfir::getFortranElementOrSequenceType(mask.getType()));
   mlir::Type logicalTy = maskTy.getEleTy();
   llvm::ArrayRef<int64_t> maskShape = maskTy.getShape();
 
@@ -576,9 +576,8 @@ mlir::LogicalResult hlfir::CountOp::verify() {
   mlir::Value mask = getMask();
   mlir::Value dim = getDim();
 
-  fir::SequenceType maskTy =
-      hlfir::getFortranElementOrSequenceType(mask.getType())
-          .cast<fir::SequenceType>();
+  fir::SequenceType maskTy = mlir::cast<fir::SequenceType>(
+      hlfir::getFortranElementOrSequenceType(mask.getType()));
   llvm::ArrayRef<int64_t> maskShape = maskTy.getShape();
 
   mlir::Type resultType = results[0];
@@ -613,13 +612,14 @@ void hlfir::CountOp::getEffects(
 //===----------------------------------------------------------------------===//
 
 static unsigned getCharacterKind(mlir::Type t) {
-  return hlfir::getFortranElementType(t).cast<fir::CharacterType>().getFKind();
+  return mlir::cast<fir::CharacterType>(hlfir::getFortranElementType(t))
+      .getFKind();
 }
 
 static std::optional<fir::CharacterType::LenType>
 getCharacterLengthIfStatic(mlir::Type t) {
   if (auto charType =
-          hlfir::getFortranElementType(t).dyn_cast<fir::CharacterType>())
+          mlir::dyn_cast<fir::CharacterType>(hlfir::getFortranElementType(t)))
     if (charType.hasConstantLen())
       return charType.getLen();
   return std::nullopt;
@@ -672,15 +672,13 @@ verifyArrayAndMaskForReductionOp(NumericalReductionOp reductionOp) {
   mlir::Value array = reductionOp->getArray();
   mlir::Value mask = reductionOp->getMask();
 
-  fir::SequenceType arrayTy =
-      hlfir::getFortranElementOrSequenceType(array.getType())
-          .cast<fir::SequenceType>();
+  fir::SequenceType arrayTy = mlir::cast<fir::SequenceType>(
+      hlfir::getFortranElementOrSequenceType(array.getType()));
   llvm::ArrayRef<int64_t> arrayShape = arrayTy.getShape();
 
   if (mask) {
-    fir::SequenceType maskSeq =
-        hlfir::getFortranElementOrSequenceType(mask.getType())
-            .dyn_cast<fir::SequenceType>();
+    fir::SequenceType maskSeq = mlir::dyn_cast<fir::SequenceType>(
+        hlfir::getFortranElementOrSequenceType(mask.getType()));
     llvm::ArrayRef<int64_t> maskShape;
 
     if (maskSeq)
@@ -720,9 +718,8 @@ verifyNumericalReductionOp(NumericalReductionOp reductionOp) {
 
   mlir::Value array = reductionOp->getArray();
   mlir::Value dim = reductionOp->getDim();
-  fir::SequenceType arrayTy =
-      hlfir::getFortranElementOrSequenceType(array.getType())
-          .cast<fir::SequenceType>();
+  fir::SequenceType arrayTy = mlir::cast<fir::SequenceType>(
+      hlfir::getFortranElementOrSequenceType(array.getType()));
   mlir::Type numTy = arrayTy.getEleTy();
   llvm::ArrayRef<int64_t> arrayShape = arrayTy.getShape();
 
@@ -790,13 +787,12 @@ verifyCharacterReductionOp(CharacterReductionOp reductionOp) {
 
   mlir::Value array = reductionOp->getArray();
   mlir::Value dim = reductionOp->getDim();
-  fir::SequenceType arrayTy =
-      hlfir::getFortranElementOrSequenceType(array.getType())
-          .cast<fir::SequenceType>();
+  fir::SequenceType arrayTy = mlir::cast<fir::SequenceType>(
+      hlfir::getFortranElementOrSequenceType(array.getType()));
   mlir::Type numTy = arrayTy.getEleTy();
   llvm::ArrayRef<int64_t> arrayShape = arrayTy.getShape();
 
-  auto resultExpr = results[0].cast<hlfir::ExprType>();
+  auto resultExpr = mlir::cast<hlfir::ExprType>(results[0]);
   mlir::Type resultType = resultExpr.getEleTy();
   assert(mlir::isa<fir::CharacterType>(resultType) &&
          "result must be character");
@@ -881,9 +877,8 @@ verifyResultForMinMaxLoc(NumericalReductionOp reductionOp) {
 
   mlir::Value array = reductionOp->getArray();
   mlir::Value dim = reductionOp->getDim();
-  fir::SequenceType arrayTy =
-      hlfir::getFortranElementOrSequenceType(array.getType())
-          .cast<fir::SequenceType>();
+  fir::SequenceType arrayTy = mlir::cast<fir::SequenceType>(
+      hlfir::getFortranElementOrSequenceType(array.getType()));
   llvm::ArrayRef<int64_t> arrayShape = arrayTy.getShape();
 
   mlir::Type resultType = results[0];
@@ -993,12 +988,10 @@ void hlfir::SumOp::getEffects(
 mlir::LogicalResult hlfir::DotProductOp::verify() {
   mlir::Value lhs = getLhs();
   mlir::Value rhs = getRhs();
-  fir::SequenceType lhsTy =
-      hlfir::getFortranElementOrSequenceType(lhs.getType())
-          .cast<fir::SequenceType>();
-  fir::SequenceType rhsTy =
-      hlfir::getFortranElementOrSequenceType(rhs.getType())
-          .cast<fir::SequenceType>();
+  fir::SequenceType lhsTy = mlir::cast<fir::SequenceType>(
+      hlfir::getFortranElementOrSequenceType(lhs.getType()));
+  fir::SequenceType rhsTy = mlir::cast<fir::SequenceType>(
+      hlfir::getFortranElementOrSequenceType(rhs.getType()));
   llvm::ArrayRef<int64_t> lhsShape = lhsTy.getShape();
   llvm::ArrayRef<int64_t> rhsShape = rhsTy.getShape();
   std::size_t lhsRank = lhsShape.size();
@@ -1051,19 +1044,17 @@ void hlfir::DotProductOp::getEffects(
 mlir::LogicalResult hlfir::MatmulOp::verify() {
   mlir::Value lhs = getLhs();
   mlir::Value rhs = getRhs();
-  fir::SequenceType lhsTy =
-      hlfir::getFortranElementOrSequenceType(lhs.getType())
-          .cast<fir::SequenceType>();
-  fir::SequenceType rhsTy =
-      hlfir::getFortranElementOrSequenceType(rhs.getType())
-          .cast<fir::SequenceType>();
+  fir::SequenceType lhsTy = mlir::cast<fir::SequenceType>(
+      hlfir::getFortranElementOrSequenceType(lhs.getType()));
+  fir::SequenceType rhsTy = mlir::cast<fir::SequenceType>(
+      hlfir::getFortranElementOrSequenceType(rhs.getType()));
   llvm::ArrayRef<int64_t> lhsShape = lhsTy.getShape();
   llvm::ArrayRef<int64_t> rhsShape = rhsTy.getShape();
   std::size_t lhsRank = lhsShape.size();
   std::size_t rhsRank = rhsShape.size();
   mlir::Type lhsEleTy = lhsTy.getEleTy();
   mlir::Type rhsEleTy = rhsTy.getEleTy();
-  hlfir::ExprType resultTy = getResult().getType().cast<hlfir::ExprType>();
+  hlfir::ExprType resultTy = mlir::cast<hlfir::ExprType>(getResult().getType());
   llvm::ArrayRef<int64_t> resultShape = resultTy.getShape();
   mlir::Type resultEleTy = resultTy.getEleTy();
 
@@ -1180,13 +1171,12 @@ void hlfir::MatmulOp::getEffects(
 
 mlir::LogicalResult hlfir::TransposeOp::verify() {
   mlir::Value array = getArray();
-  fir::SequenceType arrayTy =
-      hlfir::getFortranElementOrSequenceType(array.getType())
-          .cast<fir::SequenceType>();
+  fir::SequenceType arrayTy = mlir::cast<fir::SequenceType>(
+      hlfir::getFortranElementOrSequenceType(array.getType()));
   llvm::ArrayRef<int64_t> inShape = arrayTy.getShape();
   std::size_t rank = inShape.size();
   mlir::Type eleTy = arrayTy.getEleTy();
-  hlfir::ExprType resultTy = getResult().getType().cast<hlfir::ExprType>();
+  hlfir::ExprType resultTy = mlir::cast<hlfir::ExprType>(getResult().getType());
   llvm::ArrayRef<int64_t> resultShape = resultTy.getShape();
   std::size_t resultRank = resultShape.size();
   mlir::Type resultEleTy = resultTy.getEleTy();
@@ -1224,19 +1214,17 @@ void hlfir::TransposeOp::getEffects(
 mlir::LogicalResult hlfir::MatmulTransposeOp::verify() {
   mlir::Value lhs = getLhs();
   mlir::Value rhs = getRhs();
-  fir::SequenceType lhsTy =
-      hlfir::getFortranElementOrSequenceType(lhs.getType())
-          .cast<fir::SequenceType>();
-  fir::SequenceType rhsTy =
-      hlfir::getFortranElementOrSequenceType(rhs.getType())
-          .cast<fir::SequenceType>();
+  fir::SequenceType lhsTy = mlir::cast<fir::SequenceType>(
+      hlfir::getFortranElementOrSequenceType(lhs.getType()));
+  fir::SequenceType rhsTy = mlir::cast<fir::SequenceType>(
+      hlfir::getFortranElementOrSequenceType(rhs.getType()));
   llvm::ArrayRef<int64_t> lhsShape = lhsTy.getShape();
   llvm::ArrayRef<int64_t> rhsShape = rhsTy.getShape();
   std::size_t lhsRank = lhsShape.size();
   std::size_t rhsRank = rhsShape.size();
   mlir::Type lhsEleTy = lhsTy.getEleTy();
   mlir::Type rhsEleTy = rhsTy.getEleTy();
-  hlfir::ExprType resultTy = getResult().getType().cast<hlfir::ExprType>();
+  hlfir::ExprType resultTy = mlir::cast<hlfir::ExprType>(getResult().getType());
   llvm::ArrayRef<int64_t> resultShape = resultTy.getShape();
   mlir::Type resultEleTy = resultTy.getEleTy();
 
@@ -1381,7 +1369,7 @@ void hlfir::AsExprOp::build(mlir::OpBuilder &builder,
   hlfir::ExprType::Shape typeShape;
   bool isPolymorphic = fir::isPolymorphicType(var.getType());
   mlir::Type type = getFortranElementOrSequenceType(var.getType());
-  if (auto seqType = type.dyn_cast<fir::SequenceType>()) {
+  if (auto seqType = mlir::dyn_cast<fir::SequenceType>(type)) {
     typeShape.append(seqType.getShape().begin(), seqType.getShape().end());
     type = seqType.getEleTy();
   }
@@ -1427,7 +1415,7 @@ static void buildElemental(mlir::OpBuilder &builder,
                           isUnordered ? builder.getUnitAttr() : nullptr);
   mlir::Region *bodyRegion = odsState.addRegion();
   bodyRegion->push_back(new mlir::Block{});
-  if (auto shapeType = shape.getType().dyn_cast<fir::ShapeType>()) {
+  if (auto shapeType = mlir::dyn_cast<fir::ShapeType>(shape.getType())) {
     unsigned dim = shapeType.getRank();
     mlir::Type indexType = builder.getIndexType();
     for (unsigned d = 0; d < dim; ++d)
@@ -1468,7 +1456,7 @@ void hlfir::ApplyOp::build(mlir::OpBuilder &builder,
                            mlir::ValueRange indices,
                            mlir::ValueRange typeparams) {
   mlir::Type resultType = expr.getType();
-  if (auto exprType = resultType.dyn_cast<hlfir::ExprType>())
+  if (auto exprType = mlir::dyn_cast<hlfir::ExprType>(resultType))
     resultType = exprType.getElementExprType();
   build(builder, odsState, resultType, expr, indices, typeparams);
 }
@@ -1517,20 +1505,20 @@ void hlfir::CopyInOp::build(mlir::OpBuilder &builder,
 
 void hlfir::ShapeOfOp::build(mlir::OpBuilder &builder,
                              mlir::OperationState &result, mlir::Value expr) {
-  hlfir::ExprType exprTy = expr.getType().cast<hlfir::ExprType>();
+  hlfir::ExprType exprTy = mlir::cast<hlfir::ExprType>(expr.getType());
   mlir::Type type = fir::ShapeType::get(builder.getContext(), exprTy.getRank());
   build(builder, result, type, expr);
 }
 
 std::size_t hlfir::ShapeOfOp::getRank() {
   mlir::Type resTy = getResult().getType();
-  fir::ShapeType shape = resTy.cast<fir::ShapeType>();
+  fir::ShapeType shape = mlir::cast<fir::ShapeType>(resTy);
   return shape.getRank();
 }
 
 mlir::LogicalResult hlfir::ShapeOfOp::verify() {
   mlir::Value expr = getExpr();
-  hlfir::ExprType exprTy = expr.getType().cast<hlfir::ExprType>();
+  hlfir::ExprType exprTy = mlir::cast<hlfir::ExprType>(expr.getType());
   std::size_t exprRank = exprTy.getShape().size();
 
   if (exprRank == 0)
@@ -1549,7 +1537,8 @@ hlfir::ShapeOfOp::canonicalize(ShapeOfOp shapeOf,
   // if extent information is available at compile time, immediately fold the
   // hlfir.shape_of into a fir.shape
   mlir::Location loc = shapeOf.getLoc();
-  hlfir::ExprType expr = shapeOf.getExpr().getType().cast<hlfir::ExprType>();
+  hlfir::ExprType expr =
+      mlir::cast<hlfir::ExprType>(shapeOf.getExpr().getType());
 
   mlir::Value shape = hlfir::genExprShape(rewriter, loc, expr);
   if (!shape)
@@ -1574,7 +1563,7 @@ void hlfir::GetExtentOp::build(mlir::OpBuilder &builder,
 }
 
 mlir::LogicalResult hlfir::GetExtentOp::verify() {
-  fir::ShapeType shapeTy = getShape().getType().cast<fir::ShapeType>();
+  fir::ShapeType shapeTy = mlir::cast<fir::ShapeType>(getShape().getType());
   std::uint64_t rank = shapeTy.getRank();
   llvm::APInt dim = getDim();
   if (dim.sge(rank))
@@ -1709,10 +1698,11 @@ mlir::LogicalResult hlfir::ElementalAddrOp::verify() {
     return emitOpError("body region must be terminated by an hlfir.yield");
   mlir::Type elementAddrType = yieldOp.getEntity().getType();
   if (!hlfir::isFortranVariableType(elementAddrType) ||
-      hlfir::getFortranElementOrSequenceType(elementAddrType)
-          .isa<fir::SequenceType>())
+      mlir::isa<fir::SequenceType>(
+          hlfir::getFortranElementOrSequenceType(elementAddrType)))
     return emitOpError("body must compute the address of a scalar entity");
-  unsigned shapeRank = getShape().getType().cast<fir::ShapeType>().getRank();
+  unsigned shapeRank =
+      mlir::cast<fir::ShapeType>(getShape().getType()).getRank();
   if (shapeRank != getIndices().size())
     return emitOpError("body number of indices must match shape rank");
   return mlir::success();
@@ -1817,8 +1807,8 @@ static bool yieldsLogical(mlir::Region &region, bool mustBeScalarI1) {
   if (mustBeScalarI1)
     return hlfir::isI1Type(yieldType);
   return hlfir::isMaskArgument(yieldType) &&
-         hlfir::getFortranElementOrSequenceType(yieldType)
-             .isa<fir::SequenceType>();
+         mlir::isa<fir::SequenceType>(
+             hlfir::getFortranElementOrSequenceType(yieldType));
 }
 
 mlir::LogicalResult hlfir::ForallMaskOp::verify() {
diff --git a/flang/lib/Optimizer/HLFIR/Transforms/BufferizeHLFIR.cpp b/flang/lib/Optimizer/HLFIR/Transforms/BufferizeHLFIR.cpp
index 1c4f82e2de81..d4e4835ee726 100644
--- a/flang/lib/Optimizer/HLFIR/Transforms/BufferizeHLFIR.cpp
+++ b/flang/lib/Optimizer/HLFIR/Transforms/BufferizeHLFIR.cpp
@@ -77,7 +77,7 @@ static mlir::Value packageBufferizedExpr(mlir::Location loc,
 /// currently enforced by the verifiers that only accept HLFIR value or
 /// variable types which do not include tuples.
 static hlfir::Entity getBufferizedExprStorage(mlir::Value bufferizedExpr) {
-  auto tupleType = bufferizedExpr.getType().dyn_cast<mlir::TupleType>();
+  auto tupleType = mlir::dyn_cast<mlir::TupleType>(bufferizedExpr.getType());
   if (!tupleType)
     return hlfir::Entity{bufferizedExpr};
   assert(tupleType.size() == 2 && "unexpected tuple type");
@@ -90,7 +90,7 @@ static hlfir::Entity getBufferizedExprStorage(mlir::Value bufferizedExpr) {
 /// Helper to extract the clean-up flag from a tuple created by
 /// packageBufferizedExpr.
 static mlir::Value getBufferizedExprMustFreeFlag(mlir::Value bufferizedExpr) {
-  auto tupleType = bufferizedExpr.getType().dyn_cast<mlir::TupleType>();
+  auto tupleType = mlir::dyn_cast<mlir::TupleType>(bufferizedExpr.getType());
   if (!tupleType)
     return bufferizedExpr;
   assert(tupleType.size() == 2 && "unexpected tuple type");
@@ -218,7 +218,7 @@ struct ShapeOfOpConversion
     } else {
       // everything else failed so try to create a shape from static type info
       hlfir::ExprType exprTy =
-          adaptor.getExpr().getType().dyn_cast_or_null<hlfir::ExprType>();
+          mlir::dyn_cast_or_null<hlfir::ExprType>(adaptor.getExpr().getType());
       if (exprTy)
         shape = hlfir::genExprShape(builder, loc, exprTy);
     }
@@ -480,10 +480,10 @@ struct AssociateOpConversion
           assert(mlir::isa<fir::ClassType>(sourceVar.getType()) &&
                  fir::isAllocatableType(sourceVar.getType()));
           assert(sourceVar.getType() == assocType);
-        } else if ((sourceVar.getType().isa<fir::BaseBoxType>() &&
-                    !assocType.isa<fir::BaseBoxType>()) ||
-                   ((sourceVar.getType().isa<fir::BoxCharType>() &&
-                     !assocType.isa<fir::BoxCharType>()))) {
+        } else if ((mlir::isa<fir::BaseBoxType>(sourceVar.getType()) &&
+                    !mlir::isa<fir::BaseBoxType>(assocType)) ||
+                   ((mlir::isa<fir::BoxCharType>(sourceVar.getType()) &&
+                     !mlir::isa<fir::BoxCharType>(assocType)))) {
           sourceVar = builder.create<fir::BoxAddrOp>(loc, assocType, sourceVar);
         } else {
           sourceVar = builder.createConvert(loc, assocType, sourceVar);
@@ -590,13 +590,13 @@ static void genBufferDestruction(mlir::Location loc, fir::FirOpBuilder &builder,
       // for MERGE with polymorphic results.
       if (mustFinalize)
         TODO(loc, "finalizing polymorphic temporary in HLFIR");
-    } else if (var.getType().isa<fir::BaseBoxType, fir::BoxCharType>()) {
+    } else if (mlir::isa<fir::BaseBoxType, fir::BoxCharType>(var.getType())) {
       if (mustFinalize && !mlir::isa<fir::BaseBoxType>(var.getType()))
         fir::emitFatalError(loc, "non-finalizable variable");
 
       addr = builder.create<fir::BoxAddrOp>(loc, heapType, var);
     } else {
-      if (!var.getType().isa<fir::HeapType>())
+      if (!mlir::isa<fir::HeapType>(var.getType()))
         addr = builder.create<fir::ConvertOp>(loc, heapType, var);
 
       if (mustFinalize || deallocComponents) {
@@ -831,7 +831,7 @@ struct ElementalOpConversion
       // the assign, insert an hlfir.destroy to mark the expression end-of-life.
       // If the expression creation allocated a buffer on the heap inside the
       // loop, this will ensure the buffer properly deallocated.
-      if (elementValue.getType().isa<hlfir::ExprType>() &&
+      if (mlir::isa<hlfir::ExprType>(elementValue.getType()) &&
           wasCreatedInCurrentBlock(elementValue, builder))
         builder.create<hlfir::DestroyOp>(loc, elementValue);
     }
@@ -926,11 +926,12 @@ public:
                         hlfir::EndAssociateOp, hlfir::SetLengthOp>();
 
     target.markUnknownOpDynamicallyLegal([](mlir::Operation *op) {
-      return llvm::all_of(
-                 op->getResultTypes(),
-                 [](mlir::Type ty) { return !ty.isa<hlfir::ExprType>(); }) &&
+      return llvm::all_of(op->getResultTypes(),
+                          [](mlir::Type ty) {
+                            return !mlir::isa<hlfir::ExprType>(ty);
+                          }) &&
              llvm::all_of(op->getOperandTypes(), [](mlir::Type ty) {
-               return !ty.isa<hlfir::ExprType>();
+               return !mlir::isa<hlfir::ExprType>(ty);
              });
     });
     if (mlir::failed(
diff --git a/flang/lib/Optimizer/HLFIR/Transforms/ConvertToFIR.cpp b/flang/lib/Optimizer/HLFIR/Transforms/ConvertToFIR.cpp
index cd534bae4ad2..517285dce133 100644
--- a/flang/lib/Optimizer/HLFIR/Transforms/ConvertToFIR.cpp
+++ b/flang/lib/Optimizer/HLFIR/Transforms/ConvertToFIR.cpp
@@ -34,7 +34,7 @@ using namespace mlir;
 static mlir::Value genAllocatableTempFromSourceBox(mlir::Location loc,
                                                    fir::FirOpBuilder &builder,
                                                    mlir::Value sourceBox) {
-  assert(sourceBox.getType().isa<fir::BaseBoxType>() &&
+  assert(mlir::isa<fir::BaseBoxType>(sourceBox.getType()) &&
          "must be a base box type");
   // Use the runtime to make a quick and dirty temp with the rhs value.
   // Overkill for scalar rhs that could be done in much more clever ways.
@@ -44,7 +44,7 @@ static mlir::Value genAllocatableTempFromSourceBox(mlir::Location loc,
   // This has the huge benefit of dealing with all cases, including
   // polymorphic entities.
   mlir::Type fromHeapType = fir::HeapType::get(fir::unwrapRefType(
-      sourceBox.getType().cast<fir::BaseBoxType>().getEleTy()));
+      mlir::cast<fir::BaseBoxType>(sourceBox.getType()).getEleTy()));
   mlir::Type fromBoxHeapType = fir::BoxType::get(fromHeapType);
   mlir::Value fromMutableBox =
       fir::factory::genNullBoxStorage(builder, loc, fromBoxHeapType);
@@ -69,7 +69,7 @@ public:
     auto module = assignOp->getParentOfType<mlir::ModuleOp>();
     fir::FirOpBuilder builder(rewriter, module);
 
-    if (rhs.getType().isa<hlfir::ExprType>()) {
+    if (mlir::isa<hlfir::ExprType>(rhs.getType())) {
       mlir::emitError(loc, "hlfir must be bufferized with --bufferize-hlfir "
                            "pass before being converted to FIR");
       return mlir::failure();
@@ -343,16 +343,15 @@ public:
     auto firBase = firDeclareOp.getResult();
     mlir::Value hlfirBase;
     mlir::Type hlfirBaseType = declareOp.getBase().getType();
-    if (hlfirBaseType.isa<fir::BaseBoxType>()) {
+    if (mlir::isa<fir::BaseBoxType>(hlfirBaseType)) {
       fir::FirOpBuilder builder(rewriter, declareOp.getOperation());
       // Helper to generate the hlfir fir.box with the local lower bounds and
       // type parameters.
       auto genHlfirBox = [&]() -> mlir::Value {
-        if (!firBase.getType().isa<fir::BaseBoxType>()) {
+        if (!mlir::isa<fir::BaseBoxType>(firBase.getType())) {
           llvm::SmallVector<mlir::Value> typeParams;
-          auto maybeCharType =
-              fir::unwrapSequenceType(fir::unwrapPassByRefType(hlfirBaseType))
-                  .dyn_cast<fir::CharacterType>();
+          auto maybeCharType = mlir::dyn_cast<fir::CharacterType>(
+              fir::unwrapSequenceType(fir::unwrapPassByRefType(hlfirBaseType)));
           if (!maybeCharType || maybeCharType.hasDynamicLen())
             typeParams.append(declareOp.getTypeparams().begin(),
                               declareOp.getTypeparams().end());
@@ -399,7 +398,7 @@ public:
                         })
                         .getResults()[0];
       }
-    } else if (hlfirBaseType.isa<fir::BoxCharType>()) {
+    } else if (mlir::isa<fir::BoxCharType>(hlfirBaseType)) {
       assert(declareOp.getTypeparams().size() == 1 &&
              "must contain character length");
       hlfirBase = rewriter.create<fir::EmboxCharOp>(
@@ -480,11 +479,12 @@ public:
         // - scalar%scalar_component [substring|complex_part] or
         // - scalar%static_size_array_comp
         // - scalar%array(indices) [substring| complex part]
-        mlir::Type componentType = baseEleTy.cast<fir::RecordType>().getType(
-            designate.getComponent().value());
+        mlir::Type componentType =
+            mlir::cast<fir::RecordType>(baseEleTy).getType(
+                designate.getComponent().value());
         mlir::Type coorTy = fir::ReferenceType::get(componentType);
         base = builder.create<fir::CoordinateOp>(loc, coorTy, base, fieldIndex);
-        if (componentType.isa<fir::BaseBoxType>()) {
+        if (mlir::isa<fir::BaseBoxType>(componentType)) {
           auto variableInterface = mlir::cast<fir::FortranVariableOpInterface>(
               designate.getOperation());
           if (variableInterface.isAllocatable() ||
@@ -500,14 +500,14 @@ public:
       } else {
         // array%component[(indices) substring|complex part] cases.
         // Component ref of array bases are dealt with below in embox/rebox.
-        assert(designateResultType.isa<fir::BaseBoxType>());
+        assert(mlir::isa<fir::BaseBoxType>(designateResultType));
       }
     }
 
-    if (designateResultType.isa<fir::BaseBoxType>()) {
+    if (mlir::isa<fir::BaseBoxType>(designateResultType)) {
       // Generate embox or rebox.
       mlir::Type eleTy = fir::unwrapPassByRefType(designateResultType);
-      bool isScalarDesignator = !eleTy.isa<fir::SequenceType>();
+      bool isScalarDesignator = !mlir::isa<fir::SequenceType>(eleTy);
       mlir::Value sourceBox;
       if (isScalarDesignator) {
         // The base box will be used for emboxing the scalar element.
@@ -583,7 +583,7 @@ public:
         assert(sliceFields.empty() && substring.empty());
       llvm::SmallVector<mlir::Type> resultType{designateResultType};
       mlir::Value resultBox;
-      if (base.getType().isa<fir::BaseBoxType>())
+      if (mlir::isa<fir::BaseBoxType>(base.getType()))
         resultBox =
             builder.create<fir::ReboxOp>(loc, resultType, base, shape, slice);
       else
@@ -598,7 +598,8 @@ public:
     // first element of a contiguous array section with compile time constant
     // shape. The base may be an array, or a scalar.
     mlir::Type resultAddressType = designateResultType;
-    if (auto boxCharType = designateResultType.dyn_cast<fir::BoxCharType>())
+    if (auto boxCharType =
+            mlir::dyn_cast<fir::BoxCharType>(designateResultType))
       resultAddressType = fir::ReferenceType::get(boxCharType.getEleTy());
 
     // Array element indexing.
@@ -620,7 +621,7 @@ public:
     // Scalar complex part ref
     if (designate.getComplexPart()) {
       // Sequence types should have already been handled by this point
-      assert(!designateResultType.isa<fir::SequenceType>());
+      assert(!mlir::isa<fir::SequenceType>(designateResultType));
       auto index = builder.createIntegerConstant(loc, builder.getIndexType(),
                                                  *designate.getComplexPart());
       auto coorTy = fir::ReferenceType::get(resultEleTy);
@@ -628,7 +629,7 @@ public:
     }
 
     // Cast/embox the computed scalar address if needed.
-    if (designateResultType.isa<fir::BoxCharType>()) {
+    if (mlir::isa<fir::BoxCharType>(designateResultType)) {
       assert(designate.getTypeparams().size() == 1 &&
              "must have character length");
       auto emboxChar = builder.create<fir::EmboxCharOp>(
@@ -671,13 +672,13 @@ public:
                   mlir::PatternRewriter &rewriter) const override {
     mlir::Location loc = parentComponent.getLoc();
     mlir::Type resultType = parentComponent.getType();
-    if (!parentComponent.getType().isa<fir::BoxType>()) {
+    if (!mlir::isa<fir::BoxType>(parentComponent.getType())) {
       mlir::Value baseAddr = parentComponent.getMemref();
       // Scalar parent component ref without any length type parameters. The
       // input may be a fir.class if it is polymorphic, since this is a scalar
       // and the output will be monomorphic, the base address can be extracted
       // from the fir.class.
-      if (baseAddr.getType().isa<fir::BaseBoxType>())
+      if (mlir::isa<fir::BaseBoxType>(baseAddr.getType()))
         baseAddr = rewriter.create<fir::BoxAddrOp>(loc, baseAddr);
       rewriter.replaceOpWithNewOp<fir::ConvertOp>(parentComponent, resultType,
                                                   baseAddr);
@@ -686,7 +687,7 @@ public:
     // Array parent component ref or PDTs.
     hlfir::Entity base{parentComponent.getMemref()};
     mlir::Value baseAddr = base.getBase();
-    if (!baseAddr.getType().isa<fir::BaseBoxType>()) {
+    if (!mlir::isa<fir::BaseBoxType>(baseAddr.getType())) {
       // Embox cannot directly be used to address parent components: it expects
       // the output type to match the input type when there are no slices. When
       // the types have at least one component, a slice to the first element can
@@ -748,7 +749,7 @@ public:
     // the hlfir.shape_of operation which led to the creation of this get_extent
     // operation should now have been lowered to a fir.shape operation
     if (auto s = mlir::dyn_cast_or_null<fir::ShapeOp>(shapeOp)) {
-      fir::ShapeType shapeTy = shape.getType().cast<fir::ShapeType>();
+      fir::ShapeType shapeTy = mlir::cast<fir::ShapeType>(shape.getType());
       llvm::APInt dim = getExtentOp.getDim();
       uint64_t dimVal = dim.getLimitedValue(shapeTy.getRank());
       mlir::Value extent = s.getExtents()[dimVal];
diff --git a/flang/lib/Optimizer/HLFIR/Transforms/LowerHLFIRIntrinsics.cpp b/flang/lib/Optimizer/HLFIR/Transforms/LowerHLFIRIntrinsics.cpp
index 0142fb0cfb0b..e9dbb7095d0e 100644
--- a/flang/lib/Optimizer/HLFIR/Transforms/LowerHLFIRIntrinsics.cpp
+++ b/flang/lib/Optimizer/HLFIR/Transforms/LowerHLFIRIntrinsics.cpp
@@ -185,7 +185,7 @@ protected:
 // the width for use in runtime intrinsic calls.
 static unsigned getKindForType(mlir::Type ty) {
   mlir::Type eltty = hlfir::getFortranElementType(ty);
-  unsigned width = eltty.cast<mlir::IntegerType>().getWidth();
+  unsigned width = mlir::cast<mlir::IntegerType>(eltty).getWidth();
   return width / 8;
 }
 
diff --git a/flang/lib/Optimizer/HLFIR/Transforms/LowerHLFIROrderedAssignments.cpp b/flang/lib/Optimizer/HLFIR/Transforms/LowerHLFIROrderedAssignments.cpp
index 84101353a740..63b52c0cd0bc 100644
--- a/flang/lib/Optimizer/HLFIR/Transforms/LowerHLFIROrderedAssignments.cpp
+++ b/flang/lib/Optimizer/HLFIR/Transforms/LowerHLFIROrderedAssignments.cpp
@@ -1090,7 +1090,7 @@ void OrderedAssignmentRewriter::generateSaveEntity(
         mlir::Value loopExtent =
             computeLoopNestIterationNumber(loc, builder, loopNest);
         auto sequenceType =
-            builder.getVarLenSeqTy(entityType).cast<fir::SequenceType>();
+            mlir::cast<fir::SequenceType>(builder.getVarLenSeqTy(entityType));
         temp = insertSavedEntity(region,
                                  fir::factory::HomogeneousScalarStack{
                                      loc, builder, sequenceType, loopExtent,
diff --git a/flang/lib/Optimizer/HLFIR/Transforms/OptimizedBufferization.cpp b/flang/lib/Optimizer/HLFIR/Transforms/OptimizedBufferization.cpp
index 685c73d67625..8d68c7021608 100644
--- a/flang/lib/Optimizer/HLFIR/Transforms/OptimizedBufferization.cpp
+++ b/flang/lib/Optimizer/HLFIR/Transforms/OptimizedBufferization.cpp
@@ -249,7 +249,7 @@ static bool areIdenticalOrDisjointSlices(mlir::Value ref1, mlir::Value ref2) {
     auto isPositiveConstant = [](mlir::Value v) -> bool {
       if (auto conOp =
               mlir::dyn_cast<mlir::arith::ConstantOp>(v.getDefiningOp()))
-        if (auto iattr = conOp.getValue().dyn_cast<mlir::IntegerAttr>())
+        if (auto iattr = mlir::dyn_cast<mlir::IntegerAttr>(conOp.getValue()))
           return iattr.getInt() > 0;
       return false;
     };
@@ -601,7 +601,7 @@ mlir::LogicalResult VariableAssignBufferization::matchAndRewrite(
   // TODO: ExprType check is here to avoid conflicts with
   // ElementalAssignBufferization pattern. We need to combine
   // these matchers into a single one that applies to AssignOp.
-  if (rhs.getType().isa<hlfir::ExprType>())
+  if (mlir::isa<hlfir::ExprType>(rhs.getType()))
     return rewriter.notifyMatchFailure(assign, "RHS is not in memory");
 
   if (!rhs.isArray())
@@ -834,7 +834,7 @@ public:
 
     unsigned rank = mlir::cast<hlfir::ExprType>(mloc.getType()).getShape()[0];
     mlir::Type arrayType = array.getType();
-    if (!arrayType.isa<fir::BoxType>())
+    if (!mlir::isa<fir::BoxType>(arrayType))
       return rewriter.notifyMatchFailure(
           mloc, "Currently requires a boxed type input");
     mlir::Type elementType = hlfir::getFortranElementType(arrayType);
@@ -850,7 +850,7 @@ public:
 
     auto init = [isMax](fir::FirOpBuilder builder, mlir::Location loc,
                         mlir::Type elementType) {
-      if (auto ty = elementType.dyn_cast<mlir::FloatType>()) {
+      if (auto ty = mlir::dyn_cast<mlir::FloatType>(elementType)) {
         const llvm::fltSemantics &sem = ty.getFloatSemantics();
         llvm::APFloat limit = llvm::APFloat::getInf(sem, /*Negative=*/isMax);
         return builder.createRealConstant(loc, elementType, limit);
@@ -901,7 +901,7 @@ public:
 
       // Compare with the max reduction value
       mlir::Value cmp;
-      if (elementType.isa<mlir::FloatType>()) {
+      if (mlir::isa<mlir::FloatType>(elementType)) {
         // For FP reductions we want the first smallest value to be used, that
         // is not NaN. A OGL/OLT condition will usually work for this unless all
         // the values are Nan or Inf. This follows the same logic as
@@ -918,7 +918,7 @@ public:
             loc, mlir::arith::CmpFPredicate::OEQ, elem, elem);
         cmpNan = builder.create<mlir::arith::AndIOp>(loc, cmpNan, cmpNan2);
         cmp = builder.create<mlir::arith::OrIOp>(loc, cmp, cmpNan);
-      } else if (elementType.isa<mlir::IntegerType>()) {
+      } else if (mlir::isa<mlir::IntegerType>(elementType)) {
         cmp = builder.create<mlir::arith::CmpIOp>(
             loc,
             isMax ? mlir::arith::CmpIPredicate::sgt
diff --git a/flang/lib/Optimizer/HLFIR/Transforms/SimplifyHLFIRIntrinsics.cpp b/flang/lib/Optimizer/HLFIR/Transforms/SimplifyHLFIRIntrinsics.cpp
index 2751575ce982..b761563eba0f 100644
--- a/flang/lib/Optimizer/HLFIR/Transforms/SimplifyHLFIRIntrinsics.cpp
+++ b/flang/lib/Optimizer/HLFIR/Transforms/SimplifyHLFIRIntrinsics.cpp
@@ -103,7 +103,8 @@ public:
     // by hlfir.elemental)
     target.addDynamicallyLegalOp<hlfir::TransposeOp>(
         [](hlfir::TransposeOp transpose) {
-          return transpose.getType().cast<hlfir::ExprType>().isPolymorphic();
+          return mlir::cast<hlfir::ExprType>(transpose.getType())
+              .isPolymorphic();
         });
     target.markUnknownOpDynamicallyLegal(
         [](mlir::Operation *) { return true; });
diff --git a/flang/lib/Optimizer/Transforms/AbstractResult.cpp b/flang/lib/Optimizer/Transforms/AbstractResult.cpp
index eb4dd637bb16..85472cdc5103 100644
--- a/flang/lib/Optimizer/Transforms/AbstractResult.cpp
+++ b/flang/lib/Optimizer/Transforms/AbstractResult.cpp
@@ -65,14 +65,14 @@ static mlir::FunctionType getCPtrFunctionType(mlir::FunctionType funcTy) {
   auto resultType = funcTy.getResult(0);
   assert(fir::isa_builtin_cptr_type(resultType));
   llvm::SmallVector<mlir::Type> outputTypes;
-  auto recTy = resultType.dyn_cast<fir::RecordType>();
+  auto recTy = mlir::dyn_cast<fir::RecordType>(resultType);
   outputTypes.emplace_back(recTy.getTypeList()[0].second);
   return mlir::FunctionType::get(funcTy.getContext(), funcTy.getInputs(),
                                  outputTypes);
 }
 
 static bool mustEmboxResult(mlir::Type resultType, bool shouldBoxResult) {
-  return resultType.isa<fir::SequenceType, fir::RecordType>() &&
+  return mlir::isa<fir::SequenceType, fir::RecordType>(resultType) &&
          shouldBoxResult;
 }
 
@@ -114,7 +114,7 @@ public:
     bool isResultBuiltinCPtr = fir::isa_builtin_cptr_type(result.getType());
     Op newOp;
     if (isResultBuiltinCPtr) {
-      auto recTy = result.getType().template dyn_cast<fir::RecordType>();
+      auto recTy = mlir::dyn_cast<fir::RecordType>(result.getType());
       newResultTypes.emplace_back(recTy.getTypeList()[0].second);
     }
 
@@ -261,7 +261,7 @@ public:
   mlir::LogicalResult
   matchAndRewrite(fir::AddrOfOp addrOf,
                   mlir::PatternRewriter &rewriter) const override {
-    auto oldFuncTy = addrOf.getType().cast<mlir::FunctionType>();
+    auto oldFuncTy = mlir::cast<mlir::FunctionType>(addrOf.getType());
     mlir::FunctionType newFuncTy;
     // TODO: This should be generalized for derived types, and it is
     // architecture and OS dependent.
@@ -296,7 +296,7 @@ public:
     auto loc = func.getLoc();
     auto *context = &getContext();
     // Convert function type itself if it has an abstract result.
-    auto funcTy = func.getFunctionType().cast<mlir::FunctionType>();
+    auto funcTy = mlir::cast<mlir::FunctionType>(func.getFunctionType());
     if (hasAbstractResult(funcTy)) {
       // TODO: This should be generalized for derived types, and it is
       // architecture and OS dependent.
@@ -343,11 +343,11 @@ public:
     return mlir::TypeSwitch<mlir::Type, bool>(type)
         .Case([](fir::BoxProcType boxProc) {
           return fir::hasAbstractResult(
-              boxProc.getEleTy().cast<mlir::FunctionType>());
+              mlir::cast<mlir::FunctionType>(boxProc.getEleTy()));
         })
         .Case([](fir::PointerType pointer) {
           return fir::hasAbstractResult(
-              pointer.getEleTy().cast<mlir::FunctionType>());
+              mlir::cast<mlir::FunctionType>(pointer.getEleTy()));
         })
         .Default([](auto &&) { return false; });
   }
@@ -411,7 +411,7 @@ public:
       return !hasAbstractResult(call.getFunctionType());
     });
     target.addDynamicallyLegalOp<fir::AddrOfOp>([](fir::AddrOfOp addrOf) {
-      if (auto funTy = addrOf.getType().dyn_cast<mlir::FunctionType>())
+      if (auto funTy = mlir::dyn_cast<mlir::FunctionType>(addrOf.getType()))
         return !hasAbstractResult(funTy);
       return true;
     });
diff --git a/flang/lib/Optimizer/Transforms/AddDebugInfo.cpp b/flang/lib/Optimizer/Transforms/AddDebugInfo.cpp
index 4ca338066128..18d98a11ef3c 100644
--- a/flang/lib/Optimizer/Transforms/AddDebugInfo.cpp
+++ b/flang/lib/Optimizer/Transforms/AddDebugInfo.cpp
@@ -11,12 +11,14 @@
 /// This pass populates some debug information for the module and functions.
 //===----------------------------------------------------------------------===//
 
+#include "flang/Common/Version.h"
 #include "flang/Optimizer/Builder/FIRBuilder.h"
 #include "flang/Optimizer/Builder/Todo.h"
 #include "flang/Optimizer/Dialect/FIRDialect.h"
 #include "flang/Optimizer/Dialect/FIROps.h"
 #include "flang/Optimizer/Dialect/FIRType.h"
 #include "flang/Optimizer/Dialect/Support/FIRContext.h"
+#include "flang/Optimizer/Support/InternalNames.h"
 #include "flang/Optimizer/Transforms/Passes.h"
 #include "mlir/Dialect/Func/IR/FuncOps.h"
 #include "mlir/Dialect/LLVMIR/LLVMDialect.h"
@@ -28,12 +30,12 @@
 #include "mlir/Transforms/RegionUtils.h"
 #include "llvm/BinaryFormat/Dwarf.h"
 #include "llvm/Support/Debug.h"
+#include "llvm/Support/FileSystem.h"
 #include "llvm/Support/Path.h"
 #include "llvm/Support/raw_ostream.h"
 
 namespace fir {
 #define GEN_PASS_DEF_ADDDEBUGINFO
-#define GEN_PASS_DECL_ADDDEBUGINFO
 #include "flang/Optimizer/Transforms/Passes.h.inc"
 } // namespace fir
 
@@ -43,6 +45,7 @@ namespace {
 
 class AddDebugInfoPass : public fir::impl::AddDebugInfoBase<AddDebugInfoPass> {
 public:
+  AddDebugInfoPass(fir::AddDebugInfoOptions options) : Base(options) {}
   void runOnOperation() override;
 };
 
@@ -52,66 +55,91 @@ void AddDebugInfoPass::runOnOperation() {
   mlir::ModuleOp module = getOperation();
   mlir::MLIRContext *context = &getContext();
   mlir::OpBuilder builder(context);
-  std::string inputFilePath("-");
-  if (auto fileLoc = module.getLoc().dyn_cast<mlir::FileLineColLoc>())
-    inputFilePath = fileLoc.getFilename().getValue();
+  llvm::StringRef fileName;
+  std::string filePath;
+  // We need 2 type of file paths here.
+  // 1. Name of the file as was presented to compiler. This can be absolute
+  // or relative to 2.
+  // 2. Current working directory
+  //
+  // We are also dealing with 2 different situations below. One is normal
+  // compilation where we will have a value in 'inputFilename' and we can
+  // obtain the current directory using 'current_path'.
+  // The 2nd case is when this pass is invoked directly from 'fir-opt' tool.
+  // In that case, 'inputFilename' may be empty. Location embedded in the
+  // module will be used to get file name and its directory.
+  if (inputFilename.empty()) {
+    if (auto fileLoc = mlir::dyn_cast<mlir::FileLineColLoc>(module.getLoc())) {
+      fileName = llvm::sys::path::filename(fileLoc.getFilename().getValue());
+      filePath = llvm::sys::path::parent_path(fileLoc.getFilename().getValue());
+    } else
+      fileName = "-";
+  } else {
+    fileName = inputFilename;
+    llvm::SmallString<256> cwd;
+    if (!llvm::sys::fs::current_path(cwd))
+      filePath = cwd.str();
+  }
 
-  auto getFileAttr = [context](llvm::StringRef path) -> mlir::LLVM::DIFileAttr {
-    return mlir::LLVM::DIFileAttr::get(context, llvm::sys::path::filename(path),
-                                       llvm::sys::path::parent_path(path));
-  };
-
-  mlir::LLVM::DIFileAttr fileAttr = getFileAttr(inputFilePath);
-  mlir::StringAttr producer = mlir::StringAttr::get(context, "Flang");
+  mlir::LLVM::DIFileAttr fileAttr =
+      mlir::LLVM::DIFileAttr::get(context, fileName, filePath);
+  mlir::StringAttr producer =
+      mlir::StringAttr::get(context, Fortran::common::getFlangFullVersion());
   mlir::LLVM::DICompileUnitAttr cuAttr = mlir::LLVM::DICompileUnitAttr::get(
       mlir::DistinctAttr::create(mlir::UnitAttr::get(context)),
       llvm::dwarf::getLanguage("DW_LANG_Fortran95"), fileAttr, producer,
-      /*isOptimized=*/false, mlir::LLVM::DIEmissionKind::LineTablesOnly);
+      isOptimized, debugLevel);
 
   module.walk([&](mlir::func::FuncOp funcOp) {
     mlir::Location l = funcOp->getLoc();
     // If fused location has already been created then nothing to do
     // Otherwise, create a fused location.
-    if (l.dyn_cast<mlir::FusedLoc>())
+    if (mlir::dyn_cast<mlir::FusedLoc>(l))
       return;
 
-    llvm::StringRef funcFilePath;
-    if (l.dyn_cast<mlir::FileLineColLoc>())
-      funcFilePath =
-          l.dyn_cast<mlir::FileLineColLoc>().getFilename().getValue();
-    else
-      funcFilePath = inputFilePath;
+    unsigned int CC = (funcOp.getName() == fir::NameUniquer::doProgramEntry())
+                          ? llvm::dwarf::getCallingConvention("DW_CC_program")
+                          : llvm::dwarf::getCallingConvention("DW_CC_normal");
+
+    if (auto funcLoc = mlir::dyn_cast<mlir::FileLineColLoc>(l)) {
+      fileName = llvm::sys::path::filename(funcLoc.getFilename().getValue());
+      filePath = llvm::sys::path::parent_path(funcLoc.getFilename().getValue());
+    }
 
     mlir::StringAttr funcName =
         mlir::StringAttr::get(context, funcOp.getName());
     mlir::LLVM::DIBasicTypeAttr bT = mlir::LLVM::DIBasicTypeAttr::get(
         context, llvm::dwarf::DW_TAG_base_type, "void", /*sizeInBits=*/0,
         /*encoding=*/1);
+    // FIXME: Provide proper type for subroutine
     mlir::LLVM::DISubroutineTypeAttr subTypeAttr =
-        mlir::LLVM::DISubroutineTypeAttr::get(
-            context, llvm::dwarf::getCallingConvention("DW_CC_normal"),
-            {bT, bT});
-    mlir::LLVM::DIFileAttr funcFileAttr = getFileAttr(funcFilePath);
+        mlir::LLVM::DISubroutineTypeAttr::get(context, CC, {bT, bT});
+    mlir::LLVM::DIFileAttr funcFileAttr =
+        mlir::LLVM::DIFileAttr::get(context, fileName, filePath);
 
     // Only definitions need a distinct identifier and a compilation unit.
     mlir::DistinctAttr id;
     mlir::LLVM::DICompileUnitAttr compilationUnit;
-    auto subprogramFlags = mlir::LLVM::DISubprogramFlags::Optimized;
+    mlir::LLVM::DISubprogramFlags subprogramFlags =
+        mlir::LLVM::DISubprogramFlags{};
+    if (isOptimized)
+      subprogramFlags = mlir::LLVM::DISubprogramFlags::Optimized;
     if (!funcOp.isExternal()) {
       id = mlir::DistinctAttr::create(mlir::UnitAttr::get(context));
       compilationUnit = cuAttr;
       subprogramFlags =
           subprogramFlags | mlir::LLVM::DISubprogramFlags::Definition;
     }
+    // FIXME: Provide proper line and scopeline.
     auto spAttr = mlir::LLVM::DISubprogramAttr::get(
         context, id, compilationUnit, fileAttr, funcName, funcName,
-        funcFileAttr,
-        /*line=*/1,
-        /*scopeline=*/1, subprogramFlags, subTypeAttr);
+        funcFileAttr, /*line=*/1, /*scopeline=*/1, subprogramFlags,
+        subTypeAttr);
     funcOp->setLoc(builder.getFusedLoc({funcOp->getLoc()}, spAttr));
   });
 }
 
-std::unique_ptr<mlir::Pass> fir::createAddDebugInfoPass() {
-  return std::make_unique<AddDebugInfoPass>();
+std::unique_ptr<mlir::Pass>
+fir::createAddDebugInfoPass(fir::AddDebugInfoOptions options) {
+  return std::make_unique<AddDebugInfoPass>(options);
 }
diff --git a/flang/lib/Optimizer/Transforms/AffineDemotion.cpp b/flang/lib/Optimizer/Transforms/AffineDemotion.cpp
index da29ae880700..b4523a060f5a 100644
--- a/flang/lib/Optimizer/Transforms/AffineDemotion.cpp
+++ b/flang/lib/Optimizer/Transforms/AffineDemotion.cpp
@@ -98,14 +98,15 @@ public:
   mlir::LogicalResult
   matchAndRewrite(fir::ConvertOp op,
                   mlir::PatternRewriter &rewriter) const override {
-    if (op.getRes().getType().isa<mlir::MemRefType>()) {
+    if (mlir::isa<mlir::MemRefType>(op.getRes().getType())) {
       // due to index calculation moving to affine maps we still need to
       // add converts for sequence types this has a side effect of losing
       // some information about arrays with known dimensions by creating:
       // fir.convert %arg0 : (!fir.ref<!fir.array<5xi32>>) ->
       // !fir.ref<!fir.array<?xi32>>
-      if (auto refTy = op.getValue().getType().dyn_cast<fir::ReferenceType>())
-        if (auto arrTy = refTy.getEleTy().dyn_cast<fir::SequenceType>()) {
+      if (auto refTy =
+              mlir::dyn_cast<fir::ReferenceType>(op.getValue().getType()))
+        if (auto arrTy = mlir::dyn_cast<fir::SequenceType>(refTy.getEleTy())) {
           fir::SequenceType::Shape flatShape = {
               fir::SequenceType::getUnknownExtent()};
           auto flatArrTy = fir::SequenceType::get(flatShape, arrTy.getEleTy());
@@ -158,7 +159,7 @@ public:
     mlir::ConversionTarget target(*context);
     target.addIllegalOp<memref::AllocOp>();
     target.addDynamicallyLegalOp<fir::ConvertOp>([](fir::ConvertOp op) {
-      if (op.getRes().getType().isa<mlir::MemRefType>())
+      if (mlir::isa<mlir::MemRefType>(op.getRes().getType()))
         return false;
       return true;
     });
diff --git a/flang/lib/Optimizer/Transforms/AffinePromotion.cpp b/flang/lib/Optimizer/Transforms/AffinePromotion.cpp
index 64531cb1868e..7d0131ac6fa4 100644
--- a/flang/lib/Optimizer/Transforms/AffinePromotion.cpp
+++ b/flang/lib/Optimizer/Transforms/AffinePromotion.cpp
@@ -111,7 +111,7 @@ private:
 
   bool analyzeReference(mlir::Value memref, mlir::Operation *op) {
     if (auto acoOp = memref.getDefiningOp<ArrayCoorOp>()) {
-      if (acoOp.getMemref().getType().isa<fir::BoxType>()) {
+      if (mlir::isa<fir::BoxType>(acoOp.getMemref().getType())) {
         // TODO: Look if and how fir.box can be promoted to affine.
         LLVM_DEBUG(llvm::dbgs() << "AffineLoopAnalysis: cannot promote loop, "
                                    "array memory operation uses fir.box\n";
@@ -222,7 +222,7 @@ private:
       return affineBinaryOp(mlir::AffineExprKind::Mod, op.getLhs(),
                             op.getRhs());
     if (auto op = value.getDefiningOp<mlir::arith::ConstantOp>())
-      if (auto intConstant = op.getValue().dyn_cast<IntegerAttr>())
+      if (auto intConstant = mlir::dyn_cast<IntegerAttr>(op.getValue()))
         return toAffineExpr(intConstant.getInt());
     if (auto blockArg = mlir::dyn_cast<mlir::BlockArgument>(value)) {
       affineArgs.push_back(value);
@@ -331,15 +331,16 @@ static mlir::AffineMap createArrayIndexAffineMap(unsigned dimensions,
 
 static std::optional<int64_t> constantIntegerLike(const mlir::Value value) {
   if (auto definition = value.getDefiningOp<mlir::arith::ConstantOp>())
-    if (auto stepAttr = definition.getValue().dyn_cast<IntegerAttr>())
+    if (auto stepAttr = mlir::dyn_cast<IntegerAttr>(definition.getValue()))
       return stepAttr.getInt();
   return {};
 }
 
 static mlir::Type coordinateArrayElement(fir::ArrayCoorOp op) {
   if (auto refType =
-          op.getMemref().getType().dyn_cast_or_null<ReferenceType>()) {
-    if (auto seqType = refType.getEleTy().dyn_cast_or_null<SequenceType>()) {
+          mlir::dyn_cast_or_null<ReferenceType>(op.getMemref().getType())) {
+    if (auto seqType =
+            mlir::dyn_cast_or_null<SequenceType>(refType.getEleTy())) {
       return seqType.getEleTy();
     }
   }
diff --git a/flang/lib/Optimizer/Transforms/ArrayValueCopy.cpp b/flang/lib/Optimizer/Transforms/ArrayValueCopy.cpp
index a08d58383d3a..ebc186222525 100644
--- a/flang/lib/Optimizer/Transforms/ArrayValueCopy.cpp
+++ b/flang/lib/Optimizer/Transforms/ArrayValueCopy.cpp
@@ -461,9 +461,9 @@ void ArrayCopyAnalysisBase::arrayMentions(
 }
 
 static bool hasPointerType(mlir::Type type) {
-  if (auto boxTy = type.dyn_cast<BoxType>())
+  if (auto boxTy = mlir::dyn_cast<BoxType>(type))
     type = boxTy.getEleTy();
-  return type.isa<fir::PointerType>();
+  return mlir::isa<fir::PointerType>(type);
 }
 
 // This is a NF performance hack. It makes a simple test that the slices of the
@@ -512,7 +512,7 @@ static bool mutuallyExclusiveSliceRange(ArrayLoadOp ld, ArrayMergeStoreOp st) {
     auto isPositiveConstant = [](mlir::Value v) -> bool {
       if (auto conOp =
               mlir::dyn_cast<mlir::arith::ConstantOp>(v.getDefiningOp()))
-        if (auto iattr = conOp.getValue().dyn_cast<mlir::IntegerAttr>())
+        if (auto iattr = mlir::dyn_cast<mlir::IntegerAttr>(conOp.getValue()))
           return iattr.getInt() > 0;
       return false;
     };
@@ -725,8 +725,8 @@ static bool
 conservativeCallConflict(llvm::ArrayRef<mlir::Operation *> reaches) {
   return llvm::any_of(reaches, [](mlir::Operation *op) {
     if (auto call = mlir::dyn_cast<fir::CallOp>(op))
-      if (auto callee =
-              call.getCallableForCallee().dyn_cast<mlir::SymbolRefAttr>()) {
+      if (auto callee = mlir::dyn_cast<mlir::SymbolRefAttr>(
+              call.getCallableForCallee())) {
         auto module = op->getParentOfType<mlir::ModuleOp>();
         return isInternalProcedure(
             module.lookupSymbol<mlir::func::FuncOp>(callee));
@@ -891,9 +891,9 @@ static mlir::Value getOrReadExtentsAndShapeOp(
   if (arrLoad->hasAttr(fir::getOptionalAttrName()))
     fir::emitFatalError(
         loc, "shapes from array load of OPTIONAL arrays must not be used");
-  if (auto boxTy = arrLoad.getMemref().getType().dyn_cast<BoxType>()) {
+  if (auto boxTy = mlir::dyn_cast<BoxType>(arrLoad.getMemref().getType())) {
     auto rank =
-        dyn_cast_ptrOrBoxEleTy(boxTy).cast<SequenceType>().getDimension();
+        mlir::cast<SequenceType>(dyn_cast_ptrOrBoxEleTy(boxTy)).getDimension();
     auto idxTy = rewriter.getIndexType();
     for (decltype(rank) dim = 0; dim < rank; ++dim) {
       auto dimVal = rewriter.create<mlir::arith::ConstantIndexOp>(loc, dim);
@@ -929,7 +929,7 @@ static mlir::Type toRefType(mlir::Type ty) {
 static llvm::SmallVector<mlir::Value>
 getTypeParamsIfRawData(mlir::Location loc, FirOpBuilder &builder,
                        ArrayLoadOp arrLoad, mlir::Type ty) {
-  if (ty.isa<BoxType>())
+  if (mlir::isa<BoxType>(ty))
     return {};
   return fir::factory::getTypeParams(loc, builder, arrLoad);
 }
@@ -947,8 +947,8 @@ static mlir::Value genCoorOp(mlir::PatternRewriter &rewriter,
     originated = factory::originateIndices(loc, rewriter, alloc.getType(),
                                            shape, indices);
   auto seqTy = dyn_cast_ptrOrBoxEleTy(alloc.getType());
-  assert(seqTy && seqTy.isa<SequenceType>());
-  const auto dimension = seqTy.cast<SequenceType>().getDimension();
+  assert(seqTy && mlir::isa<SequenceType>(seqTy));
+  const auto dimension = mlir::cast<SequenceType>(seqTy).getDimension();
   auto module = load->getParentOfType<mlir::ModuleOp>();
   FirOpBuilder builder(rewriter, module);
   auto typeparams = getTypeParamsIfRawData(loc, builder, load, alloc.getType());
@@ -967,7 +967,7 @@ static mlir::Value getCharacterLen(mlir::Location loc, FirOpBuilder &builder,
                                    ArrayLoadOp load, CharacterType charTy) {
   auto charLenTy = builder.getCharacterLengthType();
   if (charTy.hasDynamicLen()) {
-    if (load.getMemref().getType().isa<BoxType>()) {
+    if (mlir::isa<BoxType>(load.getMemref().getType())) {
       // The loaded array is an emboxed value. Get the CHARACTER length from
       // the box value.
       auto eleSzInBytes =
@@ -1027,7 +1027,7 @@ void genArrayCopy(mlir::Location loc, mlir::PatternRewriter &rewriter,
       getTypeParamsIfRawData(loc, builder, arrLoad, dst.getType()));
   auto eleTy = unwrapSequenceType(unwrapPassByRefType(dst.getType()));
   // Copy from (to) object to (from) temp copy of same object.
-  if (auto charTy = eleTy.dyn_cast<CharacterType>()) {
+  if (auto charTy = mlir::dyn_cast<CharacterType>(eleTy)) {
     auto len = getCharacterLen(loc, builder, arrLoad, charTy);
     CharBoxValue toChar(toAddr, len);
     CharBoxValue fromChar(fromAddr, len);
@@ -1049,8 +1049,8 @@ genArrayLoadTypeParameters(mlir::Location loc, mlir::PatternRewriter &rewriter,
     auto eleTy =
         unwrapSequenceType(unwrapPassByRefType(load.getMemref().getType()));
     if (hasDynamicSize(eleTy)) {
-      if (auto charTy = eleTy.dyn_cast<CharacterType>()) {
-        assert(load.getMemref().getType().isa<BoxType>());
+      if (auto charTy = mlir::dyn_cast<CharacterType>(eleTy)) {
+        assert(mlir::isa<BoxType>(load.getMemref().getType()));
         auto module = load->getParentOfType<mlir::ModuleOp>();
         FirOpBuilder builder(rewriter, module);
         return {getCharacterLen(loc, builder, load, charTy)};
@@ -1067,7 +1067,7 @@ findNonconstantExtents(mlir::Type memrefTy,
                        llvm::ArrayRef<mlir::Value> extents) {
   llvm::SmallVector<mlir::Value> nce;
   auto arrTy = unwrapPassByRefType(memrefTy);
-  auto seqTy = arrTy.cast<SequenceType>();
+  auto seqTy = mlir::cast<SequenceType>(arrTy);
   for (auto [s, x] : llvm::zip(seqTy.getShape(), extents))
     if (s == SequenceType::getUnknownExtent())
       nce.emplace_back(x);
diff --git a/flang/lib/Optimizer/Transforms/CMakeLists.txt b/flang/lib/Optimizer/Transforms/CMakeLists.txt
index d55655c53906..fc08d67540ce 100644
--- a/flang/lib/Optimizer/Transforms/CMakeLists.txt
+++ b/flang/lib/Optimizer/Transforms/CMakeLists.txt
@@ -35,6 +35,7 @@ add_flang_library(FIRTransforms
   FIRDialect
   FIRDialectSupport
   FIRSupport
+  FortranCommon
   HLFIRDialect
   MLIRAffineUtils
   MLIRFuncDialect
diff --git a/flang/lib/Optimizer/Transforms/CharacterConversion.cpp b/flang/lib/Optimizer/Transforms/CharacterConversion.cpp
index 2e8fc42487a5..44baad73aa25 100644
--- a/flang/lib/Optimizer/Transforms/CharacterConversion.cpp
+++ b/flang/lib/Optimizer/Transforms/CharacterConversion.cpp
@@ -60,8 +60,8 @@ public:
     // For each code point in the `from` string, convert naively to the `to`
     // string code point. Conversion is done blindly on size only, not value.
     auto getCharBits = [&](mlir::Type t) {
-      auto chrTy = fir::unwrapSequenceType(fir::dyn_cast_ptrEleTy(t))
-                       .cast<fir::CharacterType>();
+      auto chrTy = mlir::cast<fir::CharacterType>(
+          fir::unwrapSequenceType(fir::dyn_cast_ptrEleTy(t)));
       return kindMap.getCharacterBitsize(chrTy.getFKind());
     };
     auto fromBits = getCharBits(conv.getFrom().getType());
@@ -102,6 +102,9 @@ public:
 class CharacterConversion
     : public fir::impl::CharacterConversionBase<CharacterConversion> {
 public:
+  using fir::impl::CharacterConversionBase<
+      CharacterConversion>::CharacterConversionBase;
+
   void runOnOperation() override {
     CharacterConversionOptions clOpts{useRuntimeCalls.getValue()};
     if (clOpts.runtimeName.empty()) {
@@ -130,7 +133,3 @@ public:
   }
 };
 } // end anonymous namespace
-
-std::unique_ptr<mlir::Pass> fir::createCharacterConversionPass() {
-  return std::make_unique<CharacterConversion>();
-}
diff --git a/flang/lib/Optimizer/Transforms/ControlFlowConverter.cpp b/flang/lib/Optimizer/Transforms/ControlFlowConverter.cpp
index 45609a99995d..a62f6cde0e09 100644
--- a/flang/lib/Optimizer/Transforms/ControlFlowConverter.cpp
+++ b/flang/lib/Optimizer/Transforms/ControlFlowConverter.cpp
@@ -24,8 +24,7 @@
 #include "llvm/Support/CommandLine.h"
 
 namespace fir {
-#define GEN_PASS_DEF_CFGCONVERSIONONFUNC
-#define GEN_PASS_DEF_CFGCONVERSIONONREDUCTION
+#define GEN_PASS_DEF_CFGCONVERSION
 #include "flang/Optimizer/Transforms/Passes.h.inc"
 } // namespace fir
 
@@ -309,9 +308,10 @@ public:
 };
 
 /// Convert FIR structured control flow ops to CFG ops.
-template <typename Pass, template <typename> class PassBase>
-class CfgConversionTemplate : public PassBase<Pass> {
+class CfgConversion : public fir::impl::CFGConversionBase<CfgConversion> {
 public:
+  using CFGConversionBase<CfgConversion>::CFGConversionBase;
+
   void runOnOperation() override {
     auto *context = &this->getContext();
     mlir::RewritePatternSet patterns(context);
@@ -333,14 +333,6 @@ public:
   }
 };
 
-class CfgConversionOnFunc
-    : public CfgConversionTemplate<CfgConversionOnFunc,
-                                   ::fir::impl::CFGConversionOnFuncBase> {};
-
-class CfgConversionOnReduction
-    : public CfgConversionTemplate<CfgConversionOnReduction,
-                                   ::fir::impl::CFGConversionOnReductionBase> {
-};
 } // namespace
 
 /// Expose conversion rewriters to other passes
@@ -349,13 +341,3 @@ void fir::populateCfgConversionRewrites(mlir::RewritePatternSet &patterns,
   patterns.insert<CfgLoopConv, CfgIfConv, CfgIterWhileConv>(
       patterns.getContext(), forceLoopToExecuteOnce);
 }
-
-/// Convert FIR's structured control flow ops to CFG ops.  This
-/// conversion enables the `createLowerToCFGPass` to transform these to CFG
-/// form.
-std::unique_ptr<mlir::Pass> fir::createFirToCfgOnFuncPass() {
-  return std::make_unique<CfgConversionOnFunc>();
-}
-std::unique_ptr<mlir::Pass> fir::createFirToCfgOnReductionPass() {
-  return std::make_unique<CfgConversionOnReduction>();
-}
diff --git a/flang/lib/Optimizer/Transforms/LoopVersioning.cpp b/flang/lib/Optimizer/Transforms/LoopVersioning.cpp
index 7cbd2dd1f897..38cdc2b1388d 100644
--- a/flang/lib/Optimizer/Transforms/LoopVersioning.cpp
+++ b/flang/lib/Optimizer/Transforms/LoopVersioning.cpp
@@ -147,7 +147,7 @@ struct ArgsUsageInLoop {
 
 static fir::SequenceType getAsSequenceType(mlir::Value *v) {
   mlir::Type argTy = fir::unwrapPassByRefType(fir::unwrapRefType(v->getType()));
-  return argTy.dyn_cast<fir::SequenceType>();
+  return mlir::dyn_cast<fir::SequenceType>(argTy);
 }
 
 /// if a value comes from a fir.declare, follow it to the original source,
@@ -556,7 +556,3 @@ void LoopVersioningPass::runOnOperation() {
 
   LLVM_DEBUG(llvm::dbgs() << "=== End " DEBUG_TYPE " ===\n");
 }
-
-std::unique_ptr<mlir::Pass> fir::createLoopVersioningPass() {
-  return std::make_unique<LoopVersioningPass>();
-}
diff --git a/flang/lib/Optimizer/Transforms/MemoryAllocation.cpp b/flang/lib/Optimizer/Transforms/MemoryAllocation.cpp
index 166a6b10def2..ada67b4201e1 100644
--- a/flang/lib/Optimizer/Transforms/MemoryAllocation.cpp
+++ b/flang/lib/Optimizer/Transforms/MemoryAllocation.cpp
@@ -28,17 +28,6 @@ namespace fir {
 static constexpr std::size_t unlimitedArraySize = ~static_cast<std::size_t>(0);
 
 namespace {
-struct MemoryAllocationOptions {
-  // Always move dynamic array allocations to the heap. This may result in more
-  // heap fragmentation, so may impact performance negatively.
-  bool dynamicArrayOnHeap = false;
-
-  // Number of elements in array threshold for moving to heap. In environments
-  // with limited stack size, moving large arrays to the heap can avoid running
-  // out of stack space.
-  std::size_t maxStackArraySize = unlimitedArraySize;
-};
-
 class ReturnAnalysis {
 public:
   MLIR_DEFINE_EXPLICIT_INTERNAL_INLINE_TYPE_ID(ReturnAnalysis)
@@ -68,14 +57,15 @@ private:
 
 /// Return `true` if this allocation is to remain on the stack (`fir.alloca`).
 /// Otherwise the allocation should be moved to the heap (`fir.allocmem`).
-static inline bool keepStackAllocation(fir::AllocaOp alloca, mlir::Block *entry,
-                                       const MemoryAllocationOptions &options) {
+static inline bool
+keepStackAllocation(fir::AllocaOp alloca, mlir::Block *entry,
+                    const fir::MemoryAllocationOptOptions &options) {
   // Limitation: only arrays allocated on the stack in the entry block are
   // considered for now.
   // TODO: Generalize the algorithm and placement of the freemem nodes.
   if (alloca->getBlock() != entry)
     return true;
-  if (auto seqTy = alloca.getInType().dyn_cast<fir::SequenceType>()) {
+  if (auto seqTy = mlir::dyn_cast<fir::SequenceType>(alloca.getInType())) {
     if (fir::hasDynamicSize(seqTy)) {
       // Move all arrays with runtime determined size to the heap.
       if (options.dynamicArrayOnHeap)
@@ -168,6 +158,9 @@ public:
     options = {dynOnHeap, maxStackSize};
   }
 
+  MemoryAllocationOpt(const fir::MemoryAllocationOptOptions &options)
+      : options{options} {}
+
   /// Override `options` if command-line options have been set.
   inline void useCommandLineOptions() {
     if (dynamicArrayOnHeap)
@@ -211,15 +204,6 @@ public:
   }
 
 private:
-  MemoryAllocationOptions options;
+  fir::MemoryAllocationOptOptions options;
 };
 } // namespace
-
-std::unique_ptr<mlir::Pass> fir::createMemoryAllocationPass() {
-  return std::make_unique<MemoryAllocationOpt>();
-}
-
-std::unique_ptr<mlir::Pass>
-fir::createMemoryAllocationPass(bool dynOnHeap, std::size_t maxStackSize) {
-  return std::make_unique<MemoryAllocationOpt>(dynOnHeap, maxStackSize);
-}
diff --git a/flang/lib/Optimizer/Transforms/OMPFunctionFiltering.cpp b/flang/lib/Optimizer/Transforms/OMPFunctionFiltering.cpp
index 959099d039a5..005e84cb8e9f 100644
--- a/flang/lib/Optimizer/Transforms/OMPFunctionFiltering.cpp
+++ b/flang/lib/Optimizer/Transforms/OMPFunctionFiltering.cpp
@@ -12,6 +12,7 @@
 //===----------------------------------------------------------------------===//
 
 #include "flang/Optimizer/Dialect/FIRDialect.h"
+#include "flang/Optimizer/Dialect/FIROpsSupport.h"
 #include "flang/Optimizer/Transforms/Passes.h"
 
 #include "mlir/Dialect/Func/IR/FuncOps.h"
@@ -66,6 +67,16 @@ public:
         SymbolTable::UseRange funcUses = *funcOp.getSymbolUses(op);
         for (SymbolTable::SymbolUse use : funcUses) {
           Operation *callOp = use.getUser();
+          if (auto internalFunc = mlir::dyn_cast<func::FuncOp>(callOp)) {
+            // Do not delete internal procedures holding the symbol of their
+            // Fortran host procedure as attribute.
+            internalFunc->removeAttr(fir::getHostSymbolAttrName());
+            // Set public visibility so that the function is not deleted by MLIR
+            // because unused. Changing it is OK here because the function will
+            // be deleted anyway in the second filtering phase.
+            internalFunc.setVisibility(mlir::SymbolTable::Visibility::Public);
+            continue;
+          }
           // If the callOp has users then replace them with Undef values.
           if (!callOp->use_empty()) {
             SmallVector<Value> undefResults;
diff --git a/flang/lib/Optimizer/Transforms/PolymorphicOpConversion.cpp b/flang/lib/Optimizer/Transforms/PolymorphicOpConversion.cpp
index 93efea434cb1..d933dc58f375 100644
--- a/flang/lib/Optimizer/Transforms/PolymorphicOpConversion.cpp
+++ b/flang/lib/Optimizer/Transforms/PolymorphicOpConversion.cpp
@@ -97,8 +97,8 @@ struct DispatchOpConv : public OpConversionPattern<fir::DispatchOp> {
     // Get derived type information.
     mlir::Type declaredType =
         fir::getDerivedType(dispatch.getObject().getType().getEleTy());
-    assert(declaredType.isa<fir::RecordType>() && "expecting fir.type");
-    auto recordType = declaredType.dyn_cast<fir::RecordType>();
+    assert(mlir::isa<fir::RecordType>(declaredType) && "expecting fir.type");
+    auto recordType = mlir::dyn_cast<fir::RecordType>(declaredType);
 
     // Lookup for the binding table.
     auto bindingsIter = bindingTables.find(recordType.getName());
@@ -157,7 +157,7 @@ struct DispatchOpConv : public OpConversionPattern<fir::DispatchOp> {
 
     // Load the bindings descriptor.
     auto bindingsCompName = Fortran::semantics::bindingDescCompName;
-    fir::RecordType typeDescRecTy = typeDescTy.cast<fir::RecordType>();
+    fir::RecordType typeDescRecTy = mlir::cast<fir::RecordType>(typeDescTy);
     mlir::Value field = rewriter.create<fir::FieldIndexOp>(
         loc, fieldTy, bindingsCompName, typeDescRecTy, mlir::ValueRange{});
     mlir::Type coorTy =
@@ -168,8 +168,8 @@ struct DispatchOpConv : public OpConversionPattern<fir::DispatchOp> {
 
     // Load the correct binding.
     mlir::Value bindings = rewriter.create<fir::BoxAddrOp>(loc, bindingBox);
-    fir::RecordType bindingTy =
-        fir::unwrapIfDerived(bindingBox.getType().cast<fir::BaseBoxType>());
+    fir::RecordType bindingTy = fir::unwrapIfDerived(
+        mlir::cast<fir::BaseBoxType>(bindingBox.getType()));
     mlir::Type bindingAddrTy = fir::ReferenceType::get(bindingTy);
     mlir::Value bindingIdxVal = rewriter.create<mlir::arith::ConstantOp>(
         loc, rewriter.getIndexType(), rewriter.getIndexAttr(bindingIdx));
@@ -181,7 +181,7 @@ struct DispatchOpConv : public OpConversionPattern<fir::DispatchOp> {
     mlir::Value procField = rewriter.create<fir::FieldIndexOp>(
         loc, fieldTy, procCompName, bindingTy, mlir::ValueRange{});
     fir::RecordType procTy =
-        bindingTy.getType(procCompName).cast<fir::RecordType>();
+        mlir::cast<fir::RecordType>(bindingTy.getType(procCompName));
     mlir::Type procRefTy = fir::ReferenceType::get(procTy);
     mlir::Value procRef = rewriter.create<fir::CoordinateOp>(
         loc, procRefTy, bindingAddr, procField);
@@ -298,13 +298,13 @@ mlir::LogicalResult SelectTypeConv::matchAndRewrite(
   //   before in the list to respect point 3. above. Otherwise it is just
   //   added in order at the end.
   for (unsigned t = 0; t < typeGuardNum; ++t) {
-    if (auto a = typeGuards[t].dyn_cast<fir::ExactTypeAttr>()) {
+    if (auto a = mlir::dyn_cast<fir::ExactTypeAttr>(typeGuards[t])) {
       orderedTypeGuards.push_back(t);
       continue;
     }
 
-    if (auto a = typeGuards[t].dyn_cast<fir::SubclassAttr>()) {
-      if (auto recTy = a.getType().dyn_cast<fir::RecordType>()) {
+    if (auto a = mlir::dyn_cast<fir::SubclassAttr>(typeGuards[t])) {
+      if (auto recTy = mlir::dyn_cast<fir::RecordType>(a.getType())) {
         auto dt = mod.lookupSymbol<fir::TypeInfoOp>(recTy.getName());
         assert(dt && "dispatch table not found");
         llvm::SmallSet<llvm::StringRef, 4> ancestors =
@@ -313,8 +313,8 @@ mlir::LogicalResult SelectTypeConv::matchAndRewrite(
           auto it = orderedClassIsGuards.begin();
           while (it != orderedClassIsGuards.end()) {
             fir::SubclassAttr sAttr =
-                typeGuards[*it].dyn_cast<fir::SubclassAttr>();
-            if (auto ty = sAttr.getType().dyn_cast<fir::RecordType>()) {
+                mlir::dyn_cast<fir::SubclassAttr>(typeGuards[*it]);
+            if (auto ty = mlir::dyn_cast<fir::RecordType>(sAttr.getType())) {
               if (ancestors.contains(ty.getName()))
                 break;
             }
@@ -339,7 +339,7 @@ mlir::LogicalResult SelectTypeConv::matchAndRewrite(
     auto *dest = selectType.getSuccessor(idx);
     std::optional<mlir::ValueRange> destOps =
         selectType.getSuccessorOperands(operands, idx);
-    if (typeGuards[idx].dyn_cast<mlir::UnitAttr>())
+    if (mlir::dyn_cast<mlir::UnitAttr>(typeGuards[idx]))
       rewriter.replaceOpWithNewOp<mlir::cf::BranchOp>(
           selectType, dest, destOps.value_or(mlir::ValueRange{}));
     else if (mlir::failed(genTypeLadderStep(loc, selector, typeGuards[idx],
@@ -357,9 +357,9 @@ mlir::LogicalResult SelectTypeConv::genTypeLadderStep(
     fir::KindMapping &kindMap) const {
   mlir::Value cmp;
   // TYPE IS type guard comparison are all done inlined.
-  if (auto a = attr.dyn_cast<fir::ExactTypeAttr>()) {
+  if (auto a = mlir::dyn_cast<fir::ExactTypeAttr>(attr)) {
     if (fir::isa_trivial(a.getType()) ||
-        a.getType().isa<fir::CharacterType>()) {
+        mlir::isa<fir::CharacterType>(a.getType())) {
       // For type guard statement with Intrinsic type spec the type code of
       // the descriptor is compared.
       int code = fir::getTypeCode(a.getType(), kindMap);
@@ -383,10 +383,10 @@ mlir::LogicalResult SelectTypeConv::genTypeLadderStep(
       cmp = res;
     }
     // CLASS IS type guard statement is done with a runtime call.
-  } else if (auto a = attr.dyn_cast<fir::SubclassAttr>()) {
+  } else if (auto a = mlir::dyn_cast<fir::SubclassAttr>(attr)) {
     // Retrieve the type descriptor from the type guard statement record type.
-    assert(a.getType().isa<fir::RecordType>() && "expect fir.record type");
-    fir::RecordType recTy = a.getType().dyn_cast<fir::RecordType>();
+    assert(mlir::isa<fir::RecordType>(a.getType()) && "expect fir.record type");
+    fir::RecordType recTy = mlir::dyn_cast<fir::RecordType>(a.getType());
     std::string typeDescName =
         fir::NameUniquer::getTypeDescriptorName(recTy.getName());
     auto typeDescGlobal = mod.lookupSymbol<fir::GlobalOp>(typeDescName);
@@ -438,8 +438,8 @@ mlir::Value
 SelectTypeConv::genTypeDescCompare(mlir::Location loc, mlir::Value selector,
                                    mlir::Type ty, mlir::ModuleOp mod,
                                    mlir::PatternRewriter &rewriter) const {
-  assert(ty.isa<fir::RecordType>() && "expect fir.record type");
-  fir::RecordType recTy = ty.dyn_cast<fir::RecordType>();
+  assert(mlir::isa<fir::RecordType>(ty) && "expect fir.record type");
+  fir::RecordType recTy = mlir::dyn_cast<fir::RecordType>(ty);
   std::string typeDescName =
       fir::NameUniquer::getTypeDescriptorName(recTy.getName());
   auto typeDescGlobal = mod.lookupSymbol<fir::GlobalOp>(typeDescName);
diff --git a/flang/lib/Optimizer/Transforms/SimplifyIntrinsics.cpp b/flang/lib/Optimizer/Transforms/SimplifyIntrinsics.cpp
index f7820b6b8170..c61179a7460e 100644
--- a/flang/lib/Optimizer/Transforms/SimplifyIntrinsics.cpp
+++ b/flang/lib/Optimizer/Transforms/SimplifyIntrinsics.cpp
@@ -72,6 +72,9 @@ class SimplifyIntrinsicsPass
       mlir::Type elementType)>;
 
 public:
+  using fir::impl::SimplifyIntrinsicsBase<
+      SimplifyIntrinsicsPass>::SimplifyIntrinsicsBase;
+
   /// Generate a new function implementing a simplified version
   /// of a Fortran runtime function defined by \p basename name.
   /// \p typeGenerator is a callback that generates the new function's type.
@@ -212,8 +215,8 @@ static unsigned getDimCount(mlir::Value val) {
   // the first ConvertOp that has non-opaque box type that we meet
   // going through the ConvertOp chain.
   if (mlir::Value emboxVal = findBoxDef(val))
-    if (auto boxTy = emboxVal.getType().dyn_cast<fir::BoxType>())
-      if (auto seqTy = boxTy.getEleTy().dyn_cast<fir::SequenceType>())
+    if (auto boxTy = mlir::dyn_cast<fir::BoxType>(emboxVal.getType()))
+      if (auto seqTy = mlir::dyn_cast<fir::SequenceType>(boxTy.getEleTy()))
         return seqTy.getDimension();
   return 0;
 }
@@ -234,9 +237,9 @@ static std::optional<mlir::Type> getArgElementType(mlir::Value val) {
     val = defOp->getOperand(0);
     // The convert operation is expected to convert from one
     // box type to another box type.
-    auto boxType = val.getType().cast<fir::BoxType>();
+    auto boxType = mlir::cast<fir::BoxType>(val.getType());
     auto elementType = fir::unwrapSeqOrBoxedSeqType(boxType);
-    if (!elementType.isa<mlir::NoneType>())
+    if (!mlir::isa<mlir::NoneType>(elementType))
       return elementType;
   } while (true);
 }
@@ -378,7 +381,7 @@ static void genRuntimeSumBody(fir::FirOpBuilder &builder,
   // end function RTNAME(Sum)<T>x<rank>_simplified
   auto zero = [](fir::FirOpBuilder builder, mlir::Location loc,
                  mlir::Type elementType) {
-    if (auto ty = elementType.dyn_cast<mlir::FloatType>()) {
+    if (auto ty = mlir::dyn_cast<mlir::FloatType>(elementType)) {
       const llvm::fltSemantics &sem = ty.getFloatSemantics();
       return builder.createRealConstant(loc, elementType,
                                         llvm::APFloat::getZero(sem));
@@ -389,9 +392,9 @@ static void genRuntimeSumBody(fir::FirOpBuilder &builder,
   auto genBodyOp = [](fir::FirOpBuilder builder, mlir::Location loc,
                       mlir::Type elementType, mlir::Value elem1,
                       mlir::Value elem2) -> mlir::Value {
-    if (elementType.isa<mlir::FloatType>())
+    if (mlir::isa<mlir::FloatType>(elementType))
       return builder.create<mlir::arith::AddFOp>(loc, elem1, elem2);
-    if (elementType.isa<mlir::IntegerType>())
+    if (mlir::isa<mlir::IntegerType>(elementType))
       return builder.create<mlir::arith::AddIOp>(loc, elem1, elem2);
 
     llvm_unreachable("unsupported type");
@@ -411,7 +414,7 @@ static void genRuntimeMaxvalBody(fir::FirOpBuilder &builder,
                                  mlir::Type elementType) {
   auto init = [](fir::FirOpBuilder builder, mlir::Location loc,
                  mlir::Type elementType) {
-    if (auto ty = elementType.dyn_cast<mlir::FloatType>()) {
+    if (auto ty = mlir::dyn_cast<mlir::FloatType>(elementType)) {
       const llvm::fltSemantics &sem = ty.getFloatSemantics();
       return builder.createRealConstant(
           loc, elementType, llvm::APFloat::getLargest(sem, /*Negative=*/true));
@@ -424,7 +427,7 @@ static void genRuntimeMaxvalBody(fir::FirOpBuilder &builder,
   auto genBodyOp = [](fir::FirOpBuilder builder, mlir::Location loc,
                       mlir::Type elementType, mlir::Value elem1,
                       mlir::Value elem2) -> mlir::Value {
-    if (elementType.isa<mlir::FloatType>()) {
+    if (mlir::isa<mlir::FloatType>(elementType)) {
       // arith.maxf later converted to llvm.intr.maxnum does not work
       // correctly for NaNs and -0.0 (see maxnum/minnum pattern matching
       // in LLVM's InstCombine pass). Moreover, llvm.intr.maxnum
@@ -436,7 +439,7 @@ static void genRuntimeMaxvalBody(fir::FirOpBuilder &builder,
           loc, mlir::arith::CmpFPredicate::OGT, elem1, elem2);
       return builder.create<mlir::arith::SelectOp>(loc, compare, elem1, elem2);
     }
-    if (elementType.isa<mlir::IntegerType>())
+    if (mlir::isa<mlir::IntegerType>(elementType))
       return builder.create<mlir::arith::MaxSIOp>(loc, elem1, elem2);
 
     llvm_unreachable("unsupported type");
@@ -659,7 +662,7 @@ static void genRuntimeMinMaxlocBody(fir::FirOpBuilder &builder,
                                     mlir::Type resultElemTy, bool isDim) {
   auto init = [isMax](fir::FirOpBuilder builder, mlir::Location loc,
                       mlir::Type elementType) {
-    if (auto ty = elementType.dyn_cast<mlir::FloatType>()) {
+    if (auto ty = mlir::dyn_cast<mlir::FloatType>(elementType)) {
       const llvm::fltSemantics &sem = ty.getFloatSemantics();
       llvm::APFloat limit = llvm::APFloat::getInf(sem, /*Negative=*/isMax);
       return builder.createRealConstant(loc, elementType, limit);
@@ -741,7 +744,7 @@ static void genRuntimeMinMaxlocBody(fir::FirOpBuilder &builder,
     mlir::Value elem = builder.create<fir::LoadOp>(loc, addr);
 
     mlir::Value cmp;
-    if (elementType.isa<mlir::FloatType>()) {
+    if (mlir::isa<mlir::FloatType>(elementType)) {
       // For FP reductions we want the first smallest value to be used, that
       // is not NaN. A OGL/OLT condition will usually work for this unless all
       // the values are Nan or Inf. This follows the same logic as
@@ -758,7 +761,7 @@ static void genRuntimeMinMaxlocBody(fir::FirOpBuilder &builder,
           loc, mlir::arith::CmpFPredicate::OEQ, elem, elem);
       cmpNan = builder.create<mlir::arith::AndIOp>(loc, cmpNan, cmpNan2);
       cmp = builder.create<mlir::arith::OrIOp>(loc, cmp, cmpNan);
-    } else if (elementType.isa<mlir::IntegerType>()) {
+    } else if (mlir::isa<mlir::IntegerType>(elementType)) {
       cmp = builder.create<mlir::arith::CmpIOp>(
           loc,
           isMax ? mlir::arith::CmpIPredicate::sgt
@@ -836,7 +839,7 @@ static void genRuntimeMinMaxlocBody(fir::FirOpBuilder &builder,
 
     builder.setInsertionPointToStart(&ifOp.getElseRegion().front());
     mlir::Value basicValue;
-    if (elementType.isa<mlir::IntegerType>()) {
+    if (mlir::isa<mlir::IntegerType>(elementType)) {
       basicValue = builder.createIntegerConstant(loc, elementType, 0);
     } else {
       basicValue = builder.createRealConstant(loc, elementType, 0);
@@ -918,7 +921,7 @@ static void genRuntimeDotBody(fir::FirOpBuilder &builder,
   mlir::IndexType idxTy = builder.getIndexType();
 
   mlir::Value zero =
-      resultElementType.isa<mlir::FloatType>()
+      mlir::isa<mlir::FloatType>(resultElementType)
           ? builder.createRealConstant(loc, resultElementType, 0.0)
           : builder.createIntegerConstant(loc, resultElementType, 0);
 
@@ -975,10 +978,10 @@ static void genRuntimeDotBody(fir::FirOpBuilder &builder,
   // Convert to the result type.
   elem2 = builder.create<fir::ConvertOp>(loc, resultElementType, elem2);
 
-  if (resultElementType.isa<mlir::FloatType>())
+  if (mlir::isa<mlir::FloatType>(resultElementType))
     sumVal = builder.create<mlir::arith::AddFOp>(
         loc, builder.create<mlir::arith::MulFOp>(loc, elem1, elem2), sumVal);
-  else if (resultElementType.isa<mlir::IntegerType>())
+  else if (mlir::isa<mlir::IntegerType>(resultElementType))
     sumVal = builder.create<mlir::arith::AddIOp>(
         loc, builder.create<mlir::arith::MulIOp>(loc, elem1, elem2), sumVal);
   else
@@ -1053,8 +1056,8 @@ void SimplifyIntrinsicsPass::simplifyIntOrFloatReduction(
 
   mlir::Type resultType = call.getResult(0).getType();
 
-  if (!resultType.isa<mlir::FloatType>() &&
-      !resultType.isa<mlir::IntegerType>())
+  if (!mlir::isa<mlir::FloatType>(resultType) &&
+      !mlir::isa<mlir::IntegerType>(resultType))
     return;
 
   auto argType = getArgElementType(args[0]);
@@ -1100,7 +1103,8 @@ void SimplifyIntrinsicsPass::simplifyLogicalDim0Reduction(
   fir::FirOpBuilder builder{getSimplificationBuilder(call, kindMap)};
 
   // Treating logicals as integers makes things a lot easier
-  fir::LogicalType logicalType = {elementType.dyn_cast<fir::LogicalType>()};
+  fir::LogicalType logicalType = {
+      mlir::dyn_cast<fir::LogicalType>(elementType)};
   fir::KindTy kind = logicalType.getFKind();
   mlir::Type intElementType = builder.getIntegerType(kind * 8);
 
@@ -1135,7 +1139,8 @@ void SimplifyIntrinsicsPass::simplifyLogicalDim1Reduction(
   fir::FirOpBuilder builder{getSimplificationBuilder(call, kindMap)};
 
   // Treating logicals as integers makes things a lot easier
-  fir::LogicalType logicalType = {elementType.dyn_cast<fir::LogicalType>()};
+  fir::LogicalType logicalType = {
+      mlir::dyn_cast<fir::LogicalType>(elementType)};
   fir::KindTy kind = logicalType.getFKind();
   mlir::Type intElementType = builder.getIntegerType(kind * 8);
 
@@ -1179,7 +1184,7 @@ void SimplifyIntrinsicsPass::simplifyMinMaxlocReduction(
   auto inputBox = findBoxDef(args[1]);
   mlir::Type inputType = hlfir::getFortranElementType(inputBox.getType());
 
-  if (inputType.isa<fir::CharacterType>())
+  if (mlir::isa<fir::CharacterType>(inputType))
     return;
 
   int maskRank;
@@ -1190,7 +1195,8 @@ void SimplifyIntrinsicsPass::simplifyMinMaxlocReduction(
   } else {
     maskRank = getDimCount(mask);
     mlir::Type maskElemTy = hlfir::getFortranElementType(maskDef.getType());
-    fir::LogicalType logicalFirType = {maskElemTy.dyn_cast<fir::LogicalType>()};
+    fir::LogicalType logicalFirType = {
+        mlir::dyn_cast<fir::LogicalType>(maskElemTy)};
     kind = logicalFirType.getFKind();
     // Convert fir::LogicalType to mlir::Type
     logicalElemType = logicalFirType;
@@ -1299,7 +1305,8 @@ void SimplifyIntrinsicsPass::runOnOperation() {
           std::string fmfString{builder.getFastMathFlagsString()};
 
           mlir::Type type = call.getResult(0).getType();
-          if (!type.isa<mlir::FloatType>() && !type.isa<mlir::IntegerType>())
+          if (!mlir::isa<mlir::FloatType>(type) &&
+              !mlir::isa<mlir::IntegerType>(type))
             return;
 
           // Try to find the element types of the boxed arguments.
@@ -1311,11 +1318,9 @@ void SimplifyIntrinsicsPass::runOnOperation() {
 
           // Support only floating point and integer arguments
           // now (e.g. logical is skipped here).
-          if (!arg1Type->isa<mlir::FloatType>() &&
-              !arg1Type->isa<mlir::IntegerType>())
+          if (!mlir::isa<mlir::FloatType, mlir::IntegerType>(*arg1Type))
             return;
-          if (!arg2Type->isa<mlir::FloatType>() &&
-              !arg2Type->isa<mlir::IntegerType>())
+          if (!mlir::isa<mlir::FloatType, mlir::IntegerType>(*arg2Type))
             return;
 
           auto typeGenerator = [&type](fir::FirOpBuilder &builder) {
@@ -1387,6 +1392,3 @@ void SimplifyIntrinsicsPass::getDependentDialects(
   // LLVM::LinkageAttr creation requires that LLVM dialect is loaded.
   registry.insert<mlir::LLVM::LLVMDialect>();
 }
-std::unique_ptr<mlir::Pass> fir::createSimplifyIntrinsicsPass() {
-  return std::make_unique<SimplifyIntrinsicsPass>();
-}
diff --git a/flang/lib/Optimizer/Transforms/SimplifyRegionLite.cpp b/flang/lib/Optimizer/Transforms/SimplifyRegionLite.cpp
index 3fe6bed12cf4..7d1f86f8cee9 100644
--- a/flang/lib/Optimizer/Transforms/SimplifyRegionLite.cpp
+++ b/flang/lib/Optimizer/Transforms/SimplifyRegionLite.cpp
@@ -45,7 +45,3 @@ void SimplifyRegionLitePass::runOnOperation() {
   (void)mlir::eraseUnreachableBlocks(rewriter, regions);
   (void)mlir::runRegionDCE(rewriter, regions);
 }
-
-std::unique_ptr<mlir::Pass> fir::createSimplifyRegionLitePass() {
-  return std::make_unique<SimplifyRegionLitePass>();
-}
diff --git a/flang/lib/Optimizer/Transforms/StackArrays.cpp b/flang/lib/Optimizer/Transforms/StackArrays.cpp
index 1c213abefe6f..16bbb1c35646 100644
--- a/flang/lib/Optimizer/Transforms/StackArrays.cpp
+++ b/flang/lib/Optimizer/Transforms/StackArrays.cpp
@@ -351,7 +351,7 @@ void AllocationAnalysis::visitOperation(mlir::Operation *op,
     }
 
     auto retTy = allocmem.getAllocatedType();
-    if (!retTy.isa<fir::SequenceType>()) {
+    if (!mlir::isa<fir::SequenceType>(retTy)) {
       LLVM_DEBUG(llvm::dbgs()
                  << "--Allocation is not for an array: skipping\n");
       return;
@@ -776,7 +776,3 @@ void StackArraysPass::runOnFunc(mlir::Operation *func) {
     signalPassFailure();
   }
 }
-
-std::unique_ptr<mlir::Pass> fir::createStackArraysPass() {
-  return std::make_unique<StackArraysPass>();
-}
diff --git a/flang/lib/Parser/Fortran-parsers.cpp b/flang/lib/Parser/Fortran-parsers.cpp
index 2bdb8e38db95..ff01974b549a 100644
--- a/flang/lib/Parser/Fortran-parsers.cpp
+++ b/flang/lib/Parser/Fortran-parsers.cpp
@@ -123,7 +123,8 @@ TYPE_PARSER(first(
 TYPE_CONTEXT_PARSER("internal subprogram"_en_US,
     (construct<InternalSubprogram>(indirect(functionSubprogram)) ||
         construct<InternalSubprogram>(indirect(subroutineSubprogram))) /
-        forceEndOfStmt)
+            forceEndOfStmt ||
+        construct<InternalSubprogram>(indirect(compilerDirective)))
 
 // R511 internal-subprogram-part -> contains-stmt [internal-subprogram]...
 TYPE_CONTEXT_PARSER("internal subprogram part"_en_US,
diff --git a/flang/lib/Parser/prescan.cpp b/flang/lib/Parser/prescan.cpp
index 96db3955299f..2d46eae531b1 100644
--- a/flang/lib/Parser/prescan.cpp
+++ b/flang/lib/Parser/prescan.cpp
@@ -29,15 +29,18 @@ Prescanner::Prescanner(Messages &messages, CookedSource &cooked,
     Preprocessor &preprocessor, common::LanguageFeatureControl lfc)
     : messages_{messages}, cooked_{cooked}, preprocessor_{preprocessor},
       allSources_{preprocessor_.allSources()}, features_{lfc},
+      backslashFreeFormContinuation_{preprocessor.AnyDefinitions()},
       encoding_{allSources_.encoding()} {}
 
 Prescanner::Prescanner(const Prescanner &that)
     : messages_{that.messages_}, cooked_{that.cooked_},
       preprocessor_{that.preprocessor_}, allSources_{that.allSources_},
-      features_{that.features_}, inFixedForm_{that.inFixedForm_},
+      features_{that.features_},
+      backslashFreeFormContinuation_{that.backslashFreeFormContinuation_},
+      inFixedForm_{that.inFixedForm_},
       fixedFormColumnLimit_{that.fixedFormColumnLimit_},
-      encoding_{that.encoding_}, prescannerNesting_{that.prescannerNesting_ +
-                                     1},
+      encoding_{that.encoding_},
+      prescannerNesting_{that.prescannerNesting_ + 1},
       skipLeadingAmpersand_{that.skipLeadingAmpersand_},
       compilerDirectiveBloomFilter_{that.compilerDirectiveBloomFilter_},
       compilerDirectiveSentinels_{that.compilerDirectiveSentinels_} {}
@@ -1226,9 +1229,14 @@ bool Prescanner::Continuation(bool mightNeedFixedFormSpace) {
     } else {
       return FreeFormContinuation();
     }
-  } else {
-    return false;
+  } else if (*at_ == '\\' && at_ + 2 == nextLine_ &&
+      backslashFreeFormContinuation_ && !inFixedForm_ && nextLine_ < limit_) {
+    // cpp-like handling of \ at end of a free form source line
+    BeginSourceLine(nextLine_);
+    NextLine();
+    return true;
   }
+  return false;
 }
 
 std::optional<Prescanner::LineClassification>
diff --git a/flang/lib/Parser/prescan.h b/flang/lib/Parser/prescan.h
index 581980001bcc..3ee4c5a2c69e 100644
--- a/flang/lib/Parser/prescan.h
+++ b/flang/lib/Parser/prescan.h
@@ -197,6 +197,7 @@ private:
   Preprocessor &preprocessor_;
   AllSources &allSources_;
   common::LanguageFeatureControl features_;
+  bool backslashFreeFormContinuation_{false};
   bool inFixedForm_{false};
   int fixedFormColumnLimit_{72};
   Encoding encoding_{Encoding::UTF_8};
diff --git a/flang/lib/Semantics/check-call.cpp b/flang/lib/Semantics/check-call.cpp
index ce82cccf26d5..db0949e905a6 100644
--- a/flang/lib/Semantics/check-call.cpp
+++ b/flang/lib/Semantics/check-call.cpp
@@ -1929,6 +1929,7 @@ bool CheckArguments(const characteristics::Procedure &proc,
   bool explicitInterface{proc.HasExplicitInterface()};
   evaluate::FoldingContext foldingContext{context.foldingContext()};
   parser::ContextualMessages &messages{foldingContext.messages()};
+  bool allowArgumentConversions{true};
   if (!explicitInterface || treatingExternalAsImplicit) {
     parser::Messages buffer;
     {
@@ -1945,11 +1946,12 @@ bool CheckArguments(const characteristics::Procedure &proc,
       }
       return false; // don't pile on
     }
+    allowArgumentConversions = false;
   }
   if (explicitInterface) {
     auto buffer{CheckExplicitInterface(proc, actuals, context, &scope,
-        intrinsic, /*allowArgumentConversions=*/true, /*extentErrors=*/true,
-        ignoreImplicitVsExplicit)};
+        intrinsic, allowArgumentConversions,
+        /*extentErrors=*/true, ignoreImplicitVsExplicit)};
     if (!buffer.empty()) {
       if (treatingExternalAsImplicit) {
         if (auto *msg{messages.Say(
diff --git a/flang/lib/Semantics/check-declarations.cpp b/flang/lib/Semantics/check-declarations.cpp
index 6fcee96dd690..901ac20f8aae 100644
--- a/flang/lib/Semantics/check-declarations.cpp
+++ b/flang/lib/Semantics/check-declarations.cpp
@@ -1436,10 +1436,6 @@ void CheckHelper::CheckSubprogram(
   }
   if (cudaAttrs && *cudaAttrs != common::CUDASubprogramAttrs::Host) {
     // CUDA device subprogram checks
-    if (symbol.attrs().HasAny({Attr::RECURSIVE, Attr::PURE, Attr::ELEMENTAL})) {
-      messages_.Say(symbol.name(),
-          "A device subprogram may not be RECURSIVE, PURE, or ELEMENTAL"_err_en_US);
-    }
     if (ClassifyProcedure(symbol) == ProcedureDefinitionClass::Internal) {
       messages_.Say(symbol.name(),
           "A device subprogram may not be an internal subprogram"_err_en_US);
diff --git a/flang/lib/Semantics/check-omp-structure.cpp b/flang/lib/Semantics/check-omp-structure.cpp
index 56653aa74f0c..8a16299db319 100644
--- a/flang/lib/Semantics/check-omp-structure.cpp
+++ b/flang/lib/Semantics/check-omp-structure.cpp
@@ -2471,11 +2471,11 @@ void OmpStructureChecker::Enter(const parser::OmpClause::Ordered &x) {
 
 void OmpStructureChecker::Enter(const parser::OmpClause::Shared &x) {
   CheckAllowed(llvm::omp::Clause::OMPC_shared);
-  CheckIsVarPartOfAnotherVar(GetContext().clauseSource, x.v);
+  CheckIsVarPartOfAnotherVar(GetContext().clauseSource, x.v, "SHARED");
 }
 void OmpStructureChecker::Enter(const parser::OmpClause::Private &x) {
   CheckAllowed(llvm::omp::Clause::OMPC_private);
-  CheckIsVarPartOfAnotherVar(GetContext().clauseSource, x.v);
+  CheckIsVarPartOfAnotherVar(GetContext().clauseSource, x.v, "PRIVATE");
   CheckIntentInPointer(x.v, llvm::omp::Clause::OMPC_private);
 }
 
@@ -2513,7 +2513,8 @@ bool OmpStructureChecker::IsDataRefTypeParamInquiry(
 }
 
 void OmpStructureChecker::CheckIsVarPartOfAnotherVar(
-    const parser::CharBlock &source, const parser::OmpObjectList &objList) {
+    const parser::CharBlock &source, const parser::OmpObjectList &objList,
+    llvm::StringRef clause) {
   for (const auto &ompObject : objList.v) {
     common::visit(
         common::visitors{
@@ -2539,7 +2540,8 @@ void OmpStructureChecker::CheckIsVarPartOfAnotherVar(
                     context_.Say(source,
                         "A variable that is part of another variable (as an "
                         "array or structure element) cannot appear in a "
-                        "PRIVATE or SHARED clause"_err_en_US);
+                        "%s clause"_err_en_US,
+                        clause.data());
                   }
                 }
               }
@@ -2552,6 +2554,8 @@ void OmpStructureChecker::CheckIsVarPartOfAnotherVar(
 
 void OmpStructureChecker::Enter(const parser::OmpClause::Firstprivate &x) {
   CheckAllowed(llvm::omp::Clause::OMPC_firstprivate);
+
+  CheckIsVarPartOfAnotherVar(GetContext().clauseSource, x.v, "FIRSTPRIVATE");
   CheckIsLoopIvPartOfClause(llvmOmpClause::OMPC_firstprivate, x.v);
 
   SymbolSourceMap currSymbols;
@@ -2888,6 +2892,8 @@ void OmpStructureChecker::Enter(const parser::OmpClause::Copyprivate &x) {
 void OmpStructureChecker::Enter(const parser::OmpClause::Lastprivate &x) {
   CheckAllowed(llvm::omp::Clause::OMPC_lastprivate);
 
+  CheckIsVarPartOfAnotherVar(GetContext().clauseSource, x.v, "LASTPRIVATE");
+
   DirectivesClauseTriple dirClauseTriple;
   SymbolSourceMap currSymbols;
   GetSymbolsInObjectList(x.v, currSymbols);
diff --git a/flang/lib/Semantics/check-omp-structure.h b/flang/lib/Semantics/check-omp-structure.h
index 8287653458e1..1f7284307703 100644
--- a/flang/lib/Semantics/check-omp-structure.h
+++ b/flang/lib/Semantics/check-omp-structure.h
@@ -163,8 +163,8 @@ private:
   void CheckDependArraySection(
       const common::Indirection<parser::ArrayElement> &, const parser::Name &);
   bool IsDataRefTypeParamInquiry(const parser::DataRef *dataRef);
-  void CheckIsVarPartOfAnotherVar(
-      const parser::CharBlock &source, const parser::OmpObjectList &objList);
+  void CheckIsVarPartOfAnotherVar(const parser::CharBlock &source,
+      const parser::OmpObjectList &objList, llvm::StringRef clause = "");
   void CheckThreadprivateOrDeclareTargetVar(
       const parser::OmpObjectList &objList);
   void CheckSymbolNames(
diff --git a/flang/lib/Semantics/expression.cpp b/flang/lib/Semantics/expression.cpp
index a270e4b385e8..b8396209fc68 100644
--- a/flang/lib/Semantics/expression.cpp
+++ b/flang/lib/Semantics/expression.cpp
@@ -2989,8 +2989,8 @@ void ExpressionAnalyzer::Analyze(const parser::CallStmt &callStmt) {
   for (const auto &arg : actualArgList) {
     analyzer.Analyze(arg, true /* is subroutine call */);
   }
-  auto chevrons{AnalyzeChevrons(callStmt)};
-  if (!analyzer.fatalErrors() && chevrons) {
+  if (auto chevrons{AnalyzeChevrons(callStmt)};
+      chevrons && !analyzer.fatalErrors()) {
     if (std::optional<CalleeAndArguments> callee{
             GetCalleeAndArguments(std::get<parser::ProcedureDesignator>(call.t),
                 analyzer.GetActuals(), true /* subroutine */)}) {
diff --git a/flang/lib/Semantics/program-tree.cpp b/flang/lib/Semantics/program-tree.cpp
index bf773f3810c8..13c85c17459e 100644
--- a/flang/lib/Semantics/program-tree.cpp
+++ b/flang/lib/Semantics/program-tree.cpp
@@ -10,6 +10,7 @@
 #include "flang/Common/idioms.h"
 #include "flang/Parser/char-block.h"
 #include "flang/Semantics/scope.h"
+#include "flang/Semantics/semantics.h"
 
 namespace Fortran::semantics {
 
@@ -76,7 +77,8 @@ static void GetGenerics(
 }
 
 template <typename T>
-static ProgramTree BuildSubprogramTree(const parser::Name &name, const T &x) {
+static ProgramTree BuildSubprogramTree(
+    const parser::Name &name, SemanticsContext &context, const T &x) {
   const auto &spec{std::get<parser::SpecificationPart>(x.t)};
   const auto &exec{std::get<parser::ExecutionPart>(x.t)};
   const auto &subps{
@@ -89,7 +91,11 @@ static ProgramTree BuildSubprogramTree(const parser::Name &name, const T &x) {
     for (const auto &subp :
         std::get<std::list<parser::InternalSubprogram>>(subps->t)) {
       common::visit(
-          [&](const auto &y) { node.AddChild(ProgramTree::Build(y.value())); },
+          [&](const auto &y) {
+            if (auto child{ProgramTree::Build(y.value(), context)}) {
+              node.AddChild(std::move(*child));
+            }
+          },
           subp.u);
     }
   }
@@ -97,13 +103,14 @@ static ProgramTree BuildSubprogramTree(const parser::Name &name, const T &x) {
 }
 
 static ProgramTree BuildSubprogramTree(
-    const parser::Name &name, const parser::BlockData &x) {
+    const parser::Name &name, SemanticsContext &, const parser::BlockData &x) {
   const auto &spec{std::get<parser::SpecificationPart>(x.t)};
   return ProgramTree{name, spec};
 }
 
 template <typename T>
-static ProgramTree BuildModuleTree(const parser::Name &name, const T &x) {
+static ProgramTree BuildModuleTree(
+    const parser::Name &name, SemanticsContext &context, const T &x) {
   const auto &spec{std::get<parser::SpecificationPart>(x.t)};
   const auto &subps{std::get<std::optional<parser::ModuleSubprogramPart>>(x.t)};
   ProgramTree node{name, spec};
@@ -112,28 +119,42 @@ static ProgramTree BuildModuleTree(const parser::Name &name, const T &x) {
     for (const auto &subp :
         std::get<std::list<parser::ModuleSubprogram>>(subps->t)) {
       common::visit(
-          [&](const auto &y) { node.AddChild(ProgramTree::Build(y.value())); },
+          [&](const auto &y) {
+            if (auto child{ProgramTree::Build(y.value(), context)}) {
+              node.AddChild(std::move(*child));
+            }
+          },
           subp.u);
     }
   }
   return node;
 }
 
-ProgramTree ProgramTree::Build(const parser::ProgramUnit &x) {
-  return common::visit([](const auto &y) { return Build(y.value()); }, x.u);
+ProgramTree ProgramTree::Build(
+    const parser::ProgramUnit &x, SemanticsContext &context) {
+  return common::visit(
+      [&](const auto &y) {
+        auto node{Build(y.value(), context)};
+        CHECK(node.has_value());
+        return std::move(*node);
+      },
+      x.u);
 }
 
-ProgramTree ProgramTree::Build(const parser::MainProgram &x) {
+std::optional<ProgramTree> ProgramTree::Build(
+    const parser::MainProgram &x, SemanticsContext &context) {
   const auto &stmt{
       std::get<std::optional<parser::Statement<parser::ProgramStmt>>>(x.t)};
   const auto &end{std::get<parser::Statement<parser::EndProgramStmt>>(x.t)};
   static parser::Name emptyName;
-  auto result{stmt ? BuildSubprogramTree(stmt->statement.v, x).set_stmt(*stmt)
-                   : BuildSubprogramTree(emptyName, x)};
-  return result.set_endStmt(end);
+  auto result{stmt
+          ? BuildSubprogramTree(stmt->statement.v, context, x).set_stmt(*stmt)
+          : BuildSubprogramTree(emptyName, context, x)};
+  return std::move(result.set_endStmt(end));
 }
 
-ProgramTree ProgramTree::Build(const parser::FunctionSubprogram &x) {
+std::optional<ProgramTree> ProgramTree::Build(
+    const parser::FunctionSubprogram &x, SemanticsContext &context) {
   const auto &stmt{std::get<parser::Statement<parser::FunctionStmt>>(x.t)};
   const auto &end{std::get<parser::Statement<parser::EndFunctionStmt>>(x.t)};
   const auto &name{std::get<parser::Name>(stmt.statement.t)};
@@ -144,13 +165,14 @@ ProgramTree ProgramTree::Build(const parser::FunctionSubprogram &x) {
       bindingSpec = &*suffix->binding;
     }
   }
-  return BuildSubprogramTree(name, x)
+  return BuildSubprogramTree(name, context, x)
       .set_stmt(stmt)
       .set_endStmt(end)
       .set_bindingSpec(bindingSpec);
 }
 
-ProgramTree ProgramTree::Build(const parser::SubroutineSubprogram &x) {
+std::optional<ProgramTree> ProgramTree::Build(
+    const parser::SubroutineSubprogram &x, SemanticsContext &context) {
   const auto &stmt{std::get<parser::Statement<parser::SubroutineStmt>>(x.t)};
   const auto &end{std::get<parser::Statement<parser::EndSubroutineStmt>>(x.t)};
   const auto &name{std::get<parser::Name>(stmt.statement.t)};
@@ -159,48 +181,56 @@ ProgramTree ProgramTree::Build(const parser::SubroutineSubprogram &x) {
           stmt.statement.t)}) {
     bindingSpec = &*binding;
   }
-  return BuildSubprogramTree(name, x)
+  return BuildSubprogramTree(name, context, x)
       .set_stmt(stmt)
       .set_endStmt(end)
       .set_bindingSpec(bindingSpec);
 }
 
-ProgramTree ProgramTree::Build(const parser::SeparateModuleSubprogram &x) {
+std::optional<ProgramTree> ProgramTree::Build(
+    const parser::SeparateModuleSubprogram &x, SemanticsContext &context) {
   const auto &stmt{std::get<parser::Statement<parser::MpSubprogramStmt>>(x.t)};
   const auto &end{
       std::get<parser::Statement<parser::EndMpSubprogramStmt>>(x.t)};
   const auto &name{stmt.statement.v};
-  return BuildSubprogramTree(name, x).set_stmt(stmt).set_endStmt(end);
+  return BuildSubprogramTree(name, context, x).set_stmt(stmt).set_endStmt(end);
 }
 
-ProgramTree ProgramTree::Build(const parser::Module &x) {
+std::optional<ProgramTree> ProgramTree::Build(
+    const parser::Module &x, SemanticsContext &context) {
   const auto &stmt{std::get<parser::Statement<parser::ModuleStmt>>(x.t)};
   const auto &end{std::get<parser::Statement<parser::EndModuleStmt>>(x.t)};
   const auto &name{stmt.statement.v};
-  return BuildModuleTree(name, x).set_stmt(stmt).set_endStmt(end);
+  return BuildModuleTree(name, context, x).set_stmt(stmt).set_endStmt(end);
 }
 
-ProgramTree ProgramTree::Build(const parser::Submodule &x) {
+std::optional<ProgramTree> ProgramTree::Build(
+    const parser::Submodule &x, SemanticsContext &context) {
   const auto &stmt{std::get<parser::Statement<parser::SubmoduleStmt>>(x.t)};
   const auto &end{std::get<parser::Statement<parser::EndSubmoduleStmt>>(x.t)};
   const auto &name{std::get<parser::Name>(stmt.statement.t)};
-  return BuildModuleTree(name, x).set_stmt(stmt).set_endStmt(end);
+  return BuildModuleTree(name, context, x).set_stmt(stmt).set_endStmt(end);
 }
 
-ProgramTree ProgramTree::Build(const parser::BlockData &x) {
+std::optional<ProgramTree> ProgramTree::Build(
+    const parser::BlockData &x, SemanticsContext &context) {
   const auto &stmt{std::get<parser::Statement<parser::BlockDataStmt>>(x.t)};
   const auto &end{std::get<parser::Statement<parser::EndBlockDataStmt>>(x.t)};
   static parser::Name emptyName;
-  auto result{stmt.statement.v ? BuildSubprogramTree(*stmt.statement.v, x)
-                               : BuildSubprogramTree(emptyName, x)};
-  return result.set_stmt(stmt).set_endStmt(end);
+  auto result{stmt.statement.v
+          ? BuildSubprogramTree(*stmt.statement.v, context, x)
+          : BuildSubprogramTree(emptyName, context, x)};
+  return std::move(result.set_stmt(stmt).set_endStmt(end));
 }
 
-ProgramTree ProgramTree::Build(const parser::CompilerDirective &) {
-  DIE("ProgramTree::Build() called for CompilerDirective");
+std::optional<ProgramTree> ProgramTree::Build(
+    const parser::CompilerDirective &x, SemanticsContext &context) {
+  context.Say(x.source, "Compiler directive ignored here"_warn_en_US);
+  return std::nullopt;
 }
 
-ProgramTree ProgramTree::Build(const parser::OpenACCRoutineConstruct &) {
+std::optional<ProgramTree> ProgramTree::Build(
+    const parser::OpenACCRoutineConstruct &, SemanticsContext &) {
   DIE("ProgramTree::Build() called for OpenACCRoutineConstruct");
 }
 
diff --git a/flang/lib/Semantics/program-tree.h b/flang/lib/Semantics/program-tree.h
index d49b0405d8b1..ab00261a964a 100644
--- a/flang/lib/Semantics/program-tree.h
+++ b/flang/lib/Semantics/program-tree.h
@@ -26,6 +26,7 @@
 namespace Fortran::semantics {
 
 class Scope;
+class SemanticsContext;
 
 class ProgramTree {
 public:
@@ -34,16 +35,25 @@ public:
       std::list<common::Reference<const parser::GenericSpec>>;
 
   // Build the ProgramTree rooted at one of these program units.
-  static ProgramTree Build(const parser::ProgramUnit &);
-  static ProgramTree Build(const parser::MainProgram &);
-  static ProgramTree Build(const parser::FunctionSubprogram &);
-  static ProgramTree Build(const parser::SubroutineSubprogram &);
-  static ProgramTree Build(const parser::SeparateModuleSubprogram &);
-  static ProgramTree Build(const parser::Module &);
-  static ProgramTree Build(const parser::Submodule &);
-  static ProgramTree Build(const parser::BlockData &);
-  static ProgramTree Build(const parser::CompilerDirective &);
-  static ProgramTree Build(const parser::OpenACCRoutineConstruct &);
+  static ProgramTree Build(const parser::ProgramUnit &, SemanticsContext &);
+  static std::optional<ProgramTree> Build(
+      const parser::MainProgram &, SemanticsContext &);
+  static std::optional<ProgramTree> Build(
+      const parser::FunctionSubprogram &, SemanticsContext &);
+  static std::optional<ProgramTree> Build(
+      const parser::SubroutineSubprogram &, SemanticsContext &);
+  static std::optional<ProgramTree> Build(
+      const parser::SeparateModuleSubprogram &, SemanticsContext &);
+  static std::optional<ProgramTree> Build(
+      const parser::Module &, SemanticsContext &);
+  static std::optional<ProgramTree> Build(
+      const parser::Submodule &, SemanticsContext &);
+  static std::optional<ProgramTree> Build(
+      const parser::BlockData &, SemanticsContext &);
+  static std::optional<ProgramTree> Build(
+      const parser::CompilerDirective &, SemanticsContext &);
+  static std::optional<ProgramTree> Build(
+      const parser::OpenACCRoutineConstruct &, SemanticsContext &);
 
   ENUM_CLASS(Kind, // kind of node
       Program, Function, Subroutine, MpSubprogram, Module, Submodule, BlockData)
diff --git a/flang/lib/Semantics/resolve-names.cpp b/flang/lib/Semantics/resolve-names.cpp
index b941f257a95e..7bd1f4e4e961 100644
--- a/flang/lib/Semantics/resolve-names.cpp
+++ b/flang/lib/Semantics/resolve-names.cpp
@@ -1779,7 +1779,6 @@ void AttrsVisitor::SetBindNameOn(Symbol &symbol) {
       !symbol.attrs().test(Attr::BIND_C)) {
     return;
   }
-
   std::optional<std::string> label{
       evaluate::GetScalarConstantValue<evaluate::Ascii>(bindName_)};
   // 18.9.2(2): discard leading and trailing blanks
@@ -1798,16 +1797,18 @@ void AttrsVisitor::SetBindNameOn(Symbol &symbol) {
   } else {
     label = symbol.name().ToString();
   }
-  // Check if a symbol has two Bind names.
+  // Checks whether a symbol has two Bind names.
   std::string oldBindName;
-  if (symbol.GetBindName()) {
-    oldBindName = *symbol.GetBindName();
+  if (const auto *bindName{symbol.GetBindName()}) {
+    oldBindName = *bindName;
   }
   symbol.SetBindName(std::move(*label));
   if (!oldBindName.empty()) {
     if (const std::string * newBindName{symbol.GetBindName()}) {
       if (oldBindName != *newBindName) {
-        Say(symbol.name(), "The entity '%s' has multiple BIND names"_err_en_US);
+        Say(symbol.name(),
+            "The entity '%s' has multiple BIND names ('%s' and '%s')"_err_en_US,
+            symbol.name(), oldBindName, *newBindName);
       }
     }
   }
@@ -4986,7 +4987,9 @@ Symbol &DeclarationVisitor::DeclareUnknownEntity(
     if (symbol.attrs().test(Attr::EXTERNAL)) {
       ConvertToProcEntity(symbol);
     }
-    SetBindNameOn(symbol);
+    if (attrs.test(Attr::BIND_C)) {
+      SetBindNameOn(symbol);
+    }
     return symbol;
   }
 }
@@ -8886,7 +8889,7 @@ void ResolveNamesVisitor::Post(const parser::CompilerDirective &x) {
       }
     }
   } else {
-    Say(x.source, "Compiler directive was ignored"_warn_en_US);
+    Say(x.source, "Unrecognized compiler directive was ignored"_warn_en_US);
   }
 }
 
@@ -8901,7 +8904,7 @@ bool ResolveNamesVisitor::Pre(const parser::ProgramUnit &x) {
     ResolveAccParts(context(), x, &topScope_);
     return false;
   }
-  auto root{ProgramTree::Build(x)};
+  auto root{ProgramTree::Build(x, context())};
   SetScope(topScope_);
   ResolveSpecificationParts(root);
   FinishSpecificationParts(root);
diff --git a/flang/runtime/CMakeLists.txt b/flang/runtime/CMakeLists.txt
index bdd0e07bbfd4..bc81e1b1887b 100644
--- a/flang/runtime/CMakeLists.txt
+++ b/flang/runtime/CMakeLists.txt
@@ -103,7 +103,6 @@ append(${NO_LTO_FLAGS} CMAKE_CXX_FLAGS)
 add_definitions(-U_GLIBCXX_ASSERTIONS)
 add_definitions(-U_LIBCPP_ENABLE_ASSERTIONS)
 
-add_subdirectory(FortranMain)
 add_subdirectory(Float128Math)
 
 set(sources
@@ -193,6 +192,7 @@ set(supported_files
   environment.cpp
   extrema.cpp
   external-unit.cpp
+  file.cpp
   findloc.cpp
   format.cpp
   inquiry.cpp
diff --git a/flang/runtime/FortranMain/CMakeLists.txt b/flang/runtime/FortranMain/CMakeLists.txt
deleted file mode 100644
index deb7bd10acf5..000000000000
--- a/flang/runtime/FortranMain/CMakeLists.txt
+++ /dev/null
@@ -1,23 +0,0 @@
-add_flang_library(Fortran_main STATIC INSTALL_WITH_TOOLCHAIN
-  Fortran_main.c
-)
-if (DEFINED MSVC)
-  set(CMAKE_MSVC_RUNTIME_LIBRARY MultiThreaded)
-  add_flang_library(Fortran_main.static STATIC INSTALL_WITH_TOOLCHAIN
-    Fortran_main.c
-  )
-  set(CMAKE_MSVC_RUNTIME_LIBRARY MultiThreadedDLL)
-  add_flang_library(Fortran_main.dynamic STATIC INSTALL_WITH_TOOLCHAIN
-    Fortran_main.c
-  )
-  set(CMAKE_MSVC_RUNTIME_LIBRARY MultiThreadedDebug)
-  add_flang_library(Fortran_main.static_dbg STATIC INSTALL_WITH_TOOLCHAIN
-    Fortran_main.c
-  )
-  set(CMAKE_MSVC_RUNTIME_LIBRARY MultiThreadedDebugDLL)
-  add_flang_library(Fortran_main.dynamic_dbg STATIC INSTALL_WITH_TOOLCHAIN
-    Fortran_main.c
-  )
-  add_dependencies(Fortran_main Fortran_main.static Fortran_main.dynamic
-    Fortran_main.static_dbg Fortran_main.dynamic_dbg)
-endif()
diff --git a/flang/runtime/FortranMain/Fortran_main.c b/flang/runtime/FortranMain/Fortran_main.c
deleted file mode 100644
index 5d3eaced001e..000000000000
--- a/flang/runtime/FortranMain/Fortran_main.c
+++ /dev/null
@@ -1,23 +0,0 @@
-//===-- runtime/FortranMain/Fortran_main.c --------------------------------===//
-//
-// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
-// See https://llvm.org/LICENSE.txt for license information.
-// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
-//
-//===----------------------------------------------------------------------===//
-
-#include "flang/Runtime/main.h"
-#include "flang/Runtime/stop.h"
-
-/* main entry into PROGRAM */
-void _QQmain(void);
-
-extern const struct EnvironmentDefaultList *_QQEnvironmentDefaults;
-
-/* C main stub */
-int main(int argc, const char *argv[], const char *envp[]) {
-  RTNAME(ProgramStart)(argc, argv, envp, _QQEnvironmentDefaults);
-  _QQmain();
-  RTNAME(ProgramEndStatement)();
-  return 0;
-}
diff --git a/flang/runtime/edit-output.cpp b/flang/runtime/edit-output.cpp
index a06ed258f0f1..13ab91fc56ea 100644
--- a/flang/runtime/edit-output.cpp
+++ b/flang/runtime/edit-output.cpp
@@ -446,6 +446,7 @@ RT_API_ATTRS bool RealOutputEditing<KIND>::EditFOutput(const DataEdit &edit) {
       fracDigits = sizeof buffer_ - 2; // sign & NUL
     }
   }
+  bool emitTrailingZeroes{!(flags & decimal::Minimize)};
   // Multiple conversions may be needed to get the right number of
   // effective rounded fractional digits.
   bool canIncrease{true};
@@ -526,11 +527,18 @@ RT_API_ATTRS bool RealOutputEditing<KIND>::EditFOutput(const DataEdit &edit) {
     }
     int digitsBeforePoint{std::max(0, std::min(expo, convertedDigits))};
     int zeroesBeforePoint{std::max(0, expo - digitsBeforePoint)};
+    if (zeroesBeforePoint > 0 && (flags & decimal::Minimize)) {
+      // If a minimized result looks like an integer, emit all of
+      // its digits rather than clipping some to zeroes.
+      // This can happen with HUGE(0._2) == 65504._2.
+      flags &= ~decimal::Minimize;
+      continue;
+    }
     int zeroesAfterPoint{std::min(fracDigits, std::max(0, -expo))};
     int digitsAfterPoint{convertedDigits - digitsBeforePoint};
-    int trailingZeroes{flags & decimal::Minimize
-            ? 0
-            : std::max(0, fracDigits - (zeroesAfterPoint + digitsAfterPoint))};
+    int trailingZeroes{emitTrailingZeroes
+            ? std::max(0, fracDigits - (zeroesAfterPoint + digitsAfterPoint))
+            : 0};
     if (digitsBeforePoint + zeroesBeforePoint + zeroesAfterPoint +
             digitsAfterPoint + trailingZeroes ==
         0) {
@@ -822,6 +830,11 @@ RT_API_ATTRS bool EditLogicalOutput(
   case 'Z':
     return EditBOZOutput<4>(io, edit,
         reinterpret_cast<const unsigned char *>(&truth), sizeof truth);
+  case 'A': { // legacy extension
+    int truthBits{truth};
+    return EditCharacterOutput(
+        io, edit, reinterpret_cast<char *>(&truthBits), sizeof truthBits);
+  }
   default:
     io.GetIoErrorHandler().SignalError(IostatErrorInFormat,
         "Data edit descriptor '%c' may not be used with a LOGICAL data item",
diff --git a/flang/runtime/environment.cpp b/flang/runtime/environment.cpp
index b2c9665a28df..52b1d99ba536 100644
--- a/flang/runtime/environment.cpp
+++ b/flang/runtime/environment.cpp
@@ -23,9 +23,11 @@ extern char **environ;
 
 namespace Fortran::runtime {
 
+#ifndef FLANG_RUNTIME_NO_GLOBAL_VAR_DEFS
 RT_OFFLOAD_VAR_GROUP_BEGIN
 RT_VAR_ATTRS ExecutionEnvironment executionEnvironment;
 RT_OFFLOAD_VAR_GROUP_END
+#endif // FLANG_RUNTIME_NO_GLOBAL_VAR_DEFS
 
 static void SetEnvironmentDefaults(const EnvironmentDefaultList *envDefaults) {
   if (!envDefaults) {
diff --git a/flang/runtime/file.cpp b/flang/runtime/file.cpp
index acd5d33d4bb8..79db17e70acd 100644
--- a/flang/runtime/file.cpp
+++ b/flang/runtime/file.cpp
@@ -457,22 +457,22 @@ std::int64_t SizeInBytes(const char *path) {
   return -1;
 }
 #else // defined(RT_DEVICE_COMPILATION)
-bool IsATerminal(int fd) {
+RT_API_ATTRS bool IsATerminal(int fd) {
   Terminator{__FILE__, __LINE__}.Crash("%s: unsupported", RT_PRETTY_FUNCTION);
 }
-bool IsExtant(const char *path) {
+RT_API_ATTRS bool IsExtant(const char *path) {
   Terminator{__FILE__, __LINE__}.Crash("%s: unsupported", RT_PRETTY_FUNCTION);
 }
-bool MayRead(const char *path) {
+RT_API_ATTRS bool MayRead(const char *path) {
   Terminator{__FILE__, __LINE__}.Crash("%s: unsupported", RT_PRETTY_FUNCTION);
 }
-bool MayWrite(const char *path) {
+RT_API_ATTRS bool MayWrite(const char *path) {
   Terminator{__FILE__, __LINE__}.Crash("%s: unsupported", RT_PRETTY_FUNCTION);
 }
-bool MayReadAndWrite(const char *path) {
+RT_API_ATTRS bool MayReadAndWrite(const char *path) {
   Terminator{__FILE__, __LINE__}.Crash("%s: unsupported", RT_PRETTY_FUNCTION);
 }
-std::int64_t SizeInBytes(const char *path) {
+RT_API_ATTRS std::int64_t SizeInBytes(const char *path) {
   Terminator{__FILE__, __LINE__}.Crash("%s: unsupported", RT_PRETTY_FUNCTION);
 }
 #endif // defined(RT_DEVICE_COMPILATION)
diff --git a/flang/runtime/namelist.cpp b/flang/runtime/namelist.cpp
index b9eed2101ecf..af092de70f78 100644
--- a/flang/runtime/namelist.cpp
+++ b/flang/runtime/namelist.cpp
@@ -596,7 +596,7 @@ bool IODEF(InputNamelist)(Cookie cookie, const NamelistGroup &group) {
   return true;
 }
 
-bool IsNamelistNameOrSlash(IoStatementState &io) {
+RT_API_ATTRS bool IsNamelistNameOrSlash(IoStatementState &io) {
   if (auto *listInput{
           io.get_if<ListDirectedStatementState<Direction::Input>>()}) {
     if (listInput->inNamelistSequence()) {
diff --git a/flang/runtime/numeric.cpp b/flang/runtime/numeric.cpp
index abd3e500029f..52b5a56894d8 100644
--- a/flang/runtime/numeric.cpp
+++ b/flang/runtime/numeric.cpp
@@ -9,6 +9,7 @@
 #include "flang/Runtime/numeric.h"
 #include "numeric-templates.h"
 #include "terminator.h"
+#include "tools.h"
 #include "flang/Common/float128.h"
 #include <cfloat>
 #include <climits>
@@ -18,30 +19,30 @@
 namespace Fortran::runtime {
 
 template <typename RES>
-inline RT_API_ATTRS RES getIntArgValue(const char *source, int line, void *arg,
-    int kind, std::int64_t defaultValue, int resKind) {
+inline RT_API_ATTRS RES GetIntArgValue(const char *source, int line,
+    const void *arg, int kind, std::int64_t defaultValue, int resKind) {
   RES res;
   if (!arg) {
     res = static_cast<RES>(defaultValue);
   } else if (kind == 1) {
     res = static_cast<RES>(
-        *static_cast<CppTypeFor<TypeCategory::Integer, 1> *>(arg));
+        *static_cast<const CppTypeFor<TypeCategory::Integer, 1> *>(arg));
   } else if (kind == 2) {
     res = static_cast<RES>(
-        *static_cast<CppTypeFor<TypeCategory::Integer, 2> *>(arg));
+        *static_cast<const CppTypeFor<TypeCategory::Integer, 2> *>(arg));
   } else if (kind == 4) {
     res = static_cast<RES>(
-        *static_cast<CppTypeFor<TypeCategory::Integer, 4> *>(arg));
+        *static_cast<const CppTypeFor<TypeCategory::Integer, 4> *>(arg));
   } else if (kind == 8) {
     res = static_cast<RES>(
-        *static_cast<CppTypeFor<TypeCategory::Integer, 8> *>(arg));
+        *static_cast<const CppTypeFor<TypeCategory::Integer, 8> *>(arg));
 #ifdef __SIZEOF_INT128__
   } else if (kind == 16) {
     if (resKind != 16) {
       Terminator{source, line}.Crash("Unexpected integer kind in runtime");
     }
     res = static_cast<RES>(
-        *static_cast<CppTypeFor<TypeCategory::Integer, 16> *>(arg));
+        *static_cast<const CppTypeFor<TypeCategory::Integer, 16> *>(arg));
 #endif
   } else {
     Terminator{source, line}.Crash("Unexpected integer kind in runtime");
@@ -112,6 +113,22 @@ inline RT_API_ATTRS CppTypeFor<TypeCategory::Integer, 4> SelectedIntKind(T x) {
   return -1;
 }
 
+// SELECTED_LOGICAL_KIND (F'2023 16.9.182)
+template <typename T>
+inline RT_API_ATTRS CppTypeFor<TypeCategory::Integer, 4> SelectedLogicalKind(
+    T x) {
+  if (x <= 2) {
+    return 1;
+  } else if (x <= 4) {
+    return 2;
+  } else if (x <= 9) {
+    return 4;
+  } else if (x <= 18) {
+    return 8;
+  }
+  return -1;
+}
+
 // SELECTED_REAL_KIND (16.9.170)
 template <typename P, typename R, typename D>
 inline RT_API_ATTRS CppTypeFor<TypeCategory::Integer, 4> SelectedRealKind(
@@ -717,40 +734,72 @@ CppTypeFor<TypeCategory::Real, 10> RTDEF(Scale10)(
 }
 #endif
 
+// SELECTED_CHAR_KIND
+CppTypeFor<TypeCategory::Integer, 4> RTDEF(SelectedCharKind)(
+    const char *source, int line, const char *x, std::size_t length) {
+  static const char *keywords[]{
+      "ASCII", "DEFAULT", "UCS-2", "ISO_10646", "UCS-4", nullptr};
+  switch (IdentifyValue(x, length, keywords)) {
+  case 0: // ASCII
+  case 1: // DEFAULT
+    return 1;
+  case 2: // UCS-2
+    return 2;
+  case 3: // ISO_10646
+  case 4: // UCS-4
+    return 4;
+  default:
+    return -1;
+  }
+}
 // SELECTED_INT_KIND
 CppTypeFor<TypeCategory::Integer, 4> RTDEF(SelectedIntKind)(
     const char *source, int line, void *x, int xKind) {
 #ifdef __SIZEOF_INT128__
   CppTypeFor<TypeCategory::Integer, 16> r =
-      getIntArgValue<CppTypeFor<TypeCategory::Integer, 16>>(
+      GetIntArgValue<CppTypeFor<TypeCategory::Integer, 16>>(
           source, line, x, xKind, /*defaultValue*/ 0, /*resKind*/ 16);
 #else
-  std::int64_t r = getIntArgValue<std::int64_t>(
+  std::int64_t r = GetIntArgValue<std::int64_t>(
       source, line, x, xKind, /*defaultValue*/ 0, /*resKind*/ 8);
 #endif
   return SelectedIntKind(r);
 }
 
+// SELECTED_LOGICAL_KIND
+CppTypeFor<TypeCategory::Integer, 4> RTDEF(SelectedLogicalKind)(
+    const char *source, int line, void *x, int xKind) {
+#ifdef __SIZEOF_INT128__
+  CppTypeFor<TypeCategory::Integer, 16> r =
+      GetIntArgValue<CppTypeFor<TypeCategory::Integer, 16>>(
+          source, line, x, xKind, /*defaultValue*/ 0, /*resKind*/ 16);
+#else
+  std::int64_t r = GetIntArgValue<std::int64_t>(
+      source, line, x, xKind, /*defaultValue*/ 0, /*resKind*/ 8);
+#endif
+  return SelectedLogicalKind(r);
+}
+
 // SELECTED_REAL_KIND
 CppTypeFor<TypeCategory::Integer, 4> RTDEF(SelectedRealKind)(const char *source,
     int line, void *precision, int pKind, void *range, int rKind, void *radix,
     int dKind) {
 #ifdef __SIZEOF_INT128__
   CppTypeFor<TypeCategory::Integer, 16> p =
-      getIntArgValue<CppTypeFor<TypeCategory::Integer, 16>>(
+      GetIntArgValue<CppTypeFor<TypeCategory::Integer, 16>>(
           source, line, precision, pKind, /*defaultValue*/ 0, /*resKind*/ 16);
   CppTypeFor<TypeCategory::Integer, 16> r =
-      getIntArgValue<CppTypeFor<TypeCategory::Integer, 16>>(
+      GetIntArgValue<CppTypeFor<TypeCategory::Integer, 16>>(
           source, line, range, rKind, /*defaultValue*/ 0, /*resKind*/ 16);
   CppTypeFor<TypeCategory::Integer, 16> d =
-      getIntArgValue<CppTypeFor<TypeCategory::Integer, 16>>(
+      GetIntArgValue<CppTypeFor<TypeCategory::Integer, 16>>(
           source, line, radix, dKind, /*defaultValue*/ 2, /*resKind*/ 16);
 #else
-  std::int64_t p = getIntArgValue<std::int64_t>(
+  std::int64_t p = GetIntArgValue<std::int64_t>(
       source, line, precision, pKind, /*defaultValue*/ 0, /*resKind*/ 8);
-  std::int64_t r = getIntArgValue<std::int64_t>(
+  std::int64_t r = GetIntArgValue<std::int64_t>(
       source, line, range, rKind, /*defaultValue*/ 0, /*resKind*/ 8);
-  std::int64_t d = getIntArgValue<std::int64_t>(
+  std::int64_t d = GetIntArgValue<std::int64_t>(
       source, line, radix, dKind, /*defaultValue*/ 2, /*resKind*/ 8);
 #endif
   return SelectedRealKind(p, r, d);
diff --git a/flang/runtime/unit.cpp b/flang/runtime/unit.cpp
index 0e38cffdf907..3b42f45d5588 100644
--- a/flang/runtime/unit.cpp
+++ b/flang/runtime/unit.cpp
@@ -19,11 +19,13 @@
 
 namespace Fortran::runtime::io {
 
+#ifndef FLANG_RUNTIME_NO_GLOBAL_VAR_DEFS
 RT_OFFLOAD_VAR_GROUP_BEGIN
 RT_VAR_ATTRS ExternalFileUnit *defaultInput{nullptr}; // unit 5
 RT_VAR_ATTRS ExternalFileUnit *defaultOutput{nullptr}; // unit 6
 RT_VAR_ATTRS ExternalFileUnit *errorOutput{nullptr}; // unit 0 extension
 RT_OFFLOAD_VAR_GROUP_END
+#endif // FLANG_RUNTIME_NO_GLOBAL_VAR_DEFS
 
 RT_OFFLOAD_API_GROUP_BEGIN
 
diff --git a/flang/runtime/utf.cpp b/flang/runtime/utf.cpp
index 9945dc6509ec..f4b38d5225ce 100644
--- a/flang/runtime/utf.cpp
+++ b/flang/runtime/utf.cpp
@@ -10,6 +10,7 @@
 
 namespace Fortran::runtime {
 
+#ifndef FLANG_RUNTIME_NO_GLOBAL_VAR_DEFS
 // clang-format off
 RT_OFFLOAD_VAR_GROUP_BEGIN
 const RT_CONST_VAR_ATTRS std::uint8_t UTF8FirstByteTable[256]{
@@ -40,6 +41,7 @@ const RT_CONST_VAR_ATTRS std::uint8_t UTF8FirstByteTable[256]{
 };
 RT_OFFLOAD_VAR_GROUP_END
 // clang-format on
+#endif // FLANG_RUNTIME_NO_GLOBAL_VAR_DEFS
 
 RT_OFFLOAD_API_GROUP_BEGIN
 // Non-minimal encodings are accepted.
diff --git a/flang/test/CMakeLists.txt b/flang/test/CMakeLists.txt
index 7d96a72e5f36..7e036ad539df 100644
--- a/flang/test/CMakeLists.txt
+++ b/flang/test/CMakeLists.txt
@@ -62,7 +62,6 @@ set(FLANG_TEST_DEPENDS
   llvm-readobj
   split-file
   FortranRuntime
-  Fortran_main
   FortranDecimal
 )
 if (LLVM_ENABLE_PLUGINS AND NOT WIN32)
diff --git a/flang/test/Driver/bbc-mlir-pass-pipeline.f90 b/flang/test/Driver/bbc-mlir-pass-pipeline.f90
index bd7facad1ce1..caa86e66e62b 100644
--- a/flang/test/Driver/bbc-mlir-pass-pipeline.f90
+++ b/flang/test/Driver/bbc-mlir-pass-pipeline.f90
@@ -17,9 +17,16 @@ end program
 ! CHECK-NEXT:   (S) 0 num-cse'd - Number of operations CSE'd
 ! CHECK-NEXT:   (S) 0 num-dce'd - Number of operations DCE'd
 
+! CHECK-NEXT: Pipeline Collection : ['fir.global', 'func.func', 'omp.declare_reduction', 'omp.private']
+! CHECK-NEXT: 'fir.global' Pipeline
+! CHECK-NEXT:   CharacterConversion
 ! CHECK-NEXT: 'func.func' Pipeline
 ! CHECK-NEXT:   ArrayValueCopy
 ! CHECK-NEXT:   CharacterConversion
+! CHECK-NEXT: 'omp.declare_reduction' Pipeline
+! CHECK-NEXT:   CharacterConversion
+! CHECK-NEXT: 'omp.private' Pipeline
+! CHECK-NEXT:   CharacterConversion
 
 ! CHECK-NEXT: Canonicalizer
 ! CHECK-NEXT: SimplifyRegionLite
@@ -38,12 +45,16 @@ end program
 ! CHECK-NEXT:   (S) 0 num-cse'd - Number of operations CSE'd
 ! CHECK-NEXT:   (S) 0 num-dce'd - Number of operations DCE'd
 
-! CHECK-NEXT: Pipeline Collection : ['func.func', 'omp.declare_reduction']
+! CHECK-NEXT: Pipeline Collection : ['fir.global', 'func.func', 'omp.declare_reduction', 'omp.private']
+! CHECK-NEXT: 'fir.global' Pipeline
+! CHECK-NEXT:   CFGConversion
 ! CHECK-NEXT: 'func.func' Pipeline
 ! CHECK-NEXT:   PolymorphicOpConversion
-! CHECK-NEXT:   CFGConversionOnFunc
+! CHECK-NEXT:   CFGConversion
 ! CHECK-NEXT: 'omp.declare_reduction' Pipeline
-! CHECK-NEXT:   CFGConversionOnReduction
+! CHECK-NEXT:   CFGConversion
+! CHECK-NEXT: 'omp.private' Pipeline
+! CHECK-NEXT:   CFGConversion
 
 ! CHECK-NEXT: SCFToControlFlow
 ! CHECK-NEXT: Canonicalizer
diff --git a/flang/test/Driver/driver-help-hidden.f90 b/flang/test/Driver/driver-help-hidden.f90
index b5bb0f1c1b25..706b2cb6c245 100644
--- a/flang/test/Driver/driver-help-hidden.f90
+++ b/flang/test/Driver/driver-help-hidden.f90
@@ -63,7 +63,6 @@
 ! CHECK-NEXT:                         Select Windows run-time library
 ! CHECK-NEXT: -fno-automatic          Implies the SAVE attribute for non-automatic local objects in subprograms unless RECURSIVE
 ! CHECK-NEXT: -fno-color-diagnostics  Disable colors in diagnostics
-! CHECK-NEXT: -fno-fortran-main       Do not include Fortran_main.a (provided by Flang) when linking
 ! CHECK-NEXT: -fno-integrated-as      Disable the integrated assembler
 ! CHECK-NEXT: -fno-lto                Disable LTO mode (default)
 ! CHECK-NEXT: -fno-ppc-native-vector-element-order
diff --git a/flang/test/Driver/driver-help.f90 b/flang/test/Driver/driver-help.f90
index 0b0a493baf07..4c3609db80b9 100644
--- a/flang/test/Driver/driver-help.f90
+++ b/flang/test/Driver/driver-help.f90
@@ -1,298 +1,15 @@
-
-!--------------------------
-! FLANG DRIVER (flang)
-!--------------------------
 ! RUN: %flang -help 2>&1 | FileCheck %s --check-prefix=HELP
 ! RUN: not %flang -helps 2>&1 | FileCheck %s --check-prefix=ERROR
 
-!----------------------------------------
-! FLANG FRONTEND DRIVER (flang -fc1)
-!----------------------------------------
 ! RUN: %flang_fc1 -help 2>&1 | FileCheck %s --check-prefix=HELP-FC1
 ! RUN: not %flang_fc1 -helps 2>&1 | FileCheck %s --check-prefix=ERROR
 
 ! HELP:USAGE: flang
 ! HELP-EMPTY:
 ! HELP-NEXT:OPTIONS:
-! HELP-NEXT: -###                    Print (but do not run) the commands to run for this compilation
-! HELP-NEXT: -cpp                    Enable predefined and command line preprocessor macros
-! HELP-NEXT: -c                      Only run preprocess, compile, and assemble steps
-! HELP-NEXT: -dM                     Print macro definitions in -E mode instead of normal output
-! HELP-NEXT: -dumpmachine            Display the compiler's target processor
-! HELP-NEXT: -dumpversion            Display the version of the compiler
-! HELP-NEXT: -D <macro>=<value>      Define <macro> to <value> (or 1 if <value> omitted)
-! HELP-NEXT: -emit-llvm              Use the LLVM representation for assembler and object files
-! HELP-NEXT: -E                      Only run the preprocessor
-! HELP-NEXT: -falternative-parameter-statement
-! HELP-NEXT:                         Enable the old style PARAMETER statement
-! HELP-NEXT: -fapprox-func           Allow certain math function calls to be replaced with an approximately equivalent calculation
-! HELP-NEXT: -fbackslash             Specify that backslash in string introduces an escape character
-! HELP-NEXT: -fcolor-diagnostics     Enable colors in diagnostics
-! HELP-NEXT: -fconvert=<value>       Set endian conversion of data for unformatted files
-! HELP-NEXT: -fdefault-double-8      Set the default double precision kind to an 8 byte wide type
-! HELP-NEXT: -fdefault-integer-8     Set the default integer and logical kind to an 8 byte wide type
-! HELP-NEXT: -fdefault-real-8        Set the default real kind to an 8 byte wide type
-! HELP-NEXT: -ffast-math             Allow aggressive, lossy floating-point optimizations
-! HELP-NEXT: -ffixed-form            Process source files in fixed form
-! HELP-NEXT: -ffixed-line-length=<value>
-! HELP-NEXT:                         Use <value> as character line width in fixed mode
-! HELP-NEXT: -ffp-contract=<value>   Form fused FP ops (e.g. FMAs)
-! HELP-NEXT: -ffree-form             Process source files in free form
-! HELP-NEXT: -fhonor-infinities      Specify that floating-point optimizations are not allowed that assume arguments and results are not +-inf.
-! HELP-NEXT: -fhonor-nans            Specify that floating-point optimizations are not allowed that assume arguments and results are not NANs.
-! HELP-NEXT: -fimplicit-none         No implicit typing allowed unless overridden by IMPLICIT statements
-! HELP-NEXT: -finput-charset=<value> Specify the default character set for source files
-! HELP-NEXT: -fintegrated-as         Enable the integrated assembler
-! HELP-NEXT: -fintrinsic-modules-path <dir>
-! HELP-NEXT:                         Specify where to find the compiled intrinsic modules
-! HELP-NEXT: -flarge-sizes           Use INTEGER(KIND=8) for the result type in size-related intrinsics
-! HELP-NEXT: -flogical-abbreviations Enable logical abbreviations
-! HELP-NEXT: -flto=auto              Enable LTO in 'full' mode
-! HELP-NEXT: -flto=jobserver         Enable LTO in 'full' mode
-! HELP-NEXT: -flto=<value>           Set LTO mode
-! HELP-NEXT: -flto                   Enable LTO in 'full' mode
-! HELP-NEXT: -fms-runtime-lib=<value>
-! HELP-NEXT:                         Select Windows run-time library
-! HELP-NEXT: -fno-automatic          Implies the SAVE attribute for non-automatic local objects in subprograms unless RECURSIVE
-! HELP-NEXT: -fno-color-diagnostics  Disable colors in diagnostics
-! HELP-NEXT: -fno-fortran-main       Do not include Fortran_main.a (provided by Flang) when linking
-! HELP-NEXT: -fno-integrated-as      Disable the integrated assembler
-! HELP-NEXT: -fno-lto                Disable LTO mode (default)
-! HELP-NEXT: -fno-ppc-native-vector-element-order
-! HELP-NEXT:                         Specifies PowerPC non-native vector element order
-! HELP-NEXT:  -fno-rtlib-add-rpath Do not add -rpath with architecture-specific resource directory to the linker flags. When --hip-link is specified, do not add -rpath with HIP runtime library directory to the linker flags
-! HELP-NEXT: -fno-signed-zeros       Allow optimizations that ignore the sign of floating point zeros
-! HELP-NEXT: -fno-stack-arrays       Allocate array temporaries on the heap (default)
-! HELP-NEXT: -fno-version-loops-for-stride
-! HELP-NEXT:                         Do not create unit-strided loops (default)
-! HELP-NEXT: -fomit-frame-pointer    Omit the frame pointer from functions that don't need it. Some stack unwinding cases, such as profilers and sanitizers, may prefer specifying -fno-omit-frame-pointer. On many targets, -O1 and higher omit the frame pointer by default. -m[no-]omit-leaf-frame-pointer takes precedence for leaf functions
-! HELP-NEXT: -fopenacc               Enable OpenACC
-! HELP-NEXT: -fopenmp-target-debug   Enable debugging in the OpenMP offloading device RTL
-! HELP-NEXT: -fopenmp-targets=<value>
-! HELP-NEXT:                         Specify comma-separated list of triples OpenMP offloading targets to be supported
-! HELP-NEXT: -fopenmp-version=<value>
-! HELP-NEXT:                         Set OpenMP version (e.g. 45 for OpenMP 4.5, 51 for OpenMP 5.1). Default value is 11 for Flang
-! HELP-NEXT: -fopenmp                Parse OpenMP pragmas and generate parallel code.
-! HELP-NEXT: -foptimization-record-file=<file>
-! HELP-NEXT:                         Specify the output name of the file containing the optimization remarks. Implies -fsave-optimization-record. On Darwin platforms, this cannot be used with multiple -arch <arch> options.
-! HELP-NEXT: -foptimization-record-passes=<regex>
-! HELP-NEXT:                         Only include passes which match a specified regular expression in the generated optimization record (by default, include all passes)
-! HELP-NEXT: -fpass-plugin=<dsopath> Load pass plugin from a dynamic shared object file (only with new pass manager).
-! HELP-NEXT: -fppc-native-vector-element-order
-! HELP-NEXT:                         Specifies PowerPC native vector element order (default)
-! HELP-NEXT: -freciprocal-math       Allow division operations to be reassociated
-! HELP-NEXT: -fropi                  Generate read-only position independent code (ARM only)
-! HELP-NEXT: -frtlib-add-rpath Add -rpath with architecture-specific resource directory to the linker flags. When --hip-link is specified, also add -rpath with HIP runtime library directory to the linker flags
-! HELP-NEXT: -frwpi                  Generate read-write position independent code (ARM only)
-! HELP-NEXT: -fsave-optimization-record=<format>
-! HELP-NEXT:                         Generate an optimization record file in a specific format
-! HELP-NEXT: -fsave-optimization-record
-! HELP-NEXT:                         Generate a YAML optimization record file
-! HELP-NEXT: -fstack-arrays          Attempt to allocate array temporaries on the stack, no matter their size
-! HELP-NEXT: -fsyntax-only           Run the preprocessor, parser and semantic analysis stages
-! HELP-NEXT: -funderscoring          Appends one trailing underscore to external names
-! HELP-NEXT: -fveclib=<value>        Use the given vector functions library
-! HELP-NEXT: -fversion-loops-for-stride
-! HELP-NEXT:                         Create unit-strided versions of loops
-! HELP-NEXT: -fxor-operator          Enable .XOR. as a synonym of .NEQV.
-! HELP-NEXT: --gcc-install-dir=<value>
-! HELP-NEXT:                         Use GCC installation in the specified directory. The directory ends with path components like 'lib{,32,64}/gcc{,-cross}/$triple/$version'. Note: executables (e.g. ld) used by the compiler are not overridden by the selected GCC installation
-! HELP-NEXT: --gcc-toolchain=<value> Specify a directory where Flang can find 'lib{,32,64}/gcc{,-cross}/$triple/$version'. Flang will use the GCC installation with the largest version
-! HELP-NEXT: -gline-directives-only  Emit debug line info directives only
-! HELP-NEXT: -gline-tables-only      Emit debug line number tables only
-! HELP-NEXT: -gpulibc                Link the LLVM C Library for GPUs
-! HELP-NEXT: -g                      Generate source-level debug information
-! HELP-NEXT: --help-hidden           Display help for hidden options
-! HELP-NEXT: -help                   Display available options
-! HELP-NEXT: -isysroot <dir>         Set the system root directory (usually /)
-! HELP-NEXT: -I <dir>                Add directory to the end of the list of include search paths
-! HELP-NEXT: -L <dir>                Add directory to library search path
-! HELP-NEXT: -march=<value>          For a list of available architectures for the target use '-mcpu=help'
-! HELP-NEXT: -mcode-object-version=<value>
-! HELP-NEXT:                         Specify code object ABI version. Defaults to 5. (AMDGPU only)
-! HELP-NEXT: -mcpu=<value>           For a list of available CPUs for the target use '-mcpu=help'
-! HELP-NEXT: -mllvm=<arg>            Alias for -mllvm
-! HELP-NEXT: -mllvm <value>          Additional arguments to forward to LLVM's option processing
-! HELP-NEXT: -mmlir <value>          Additional arguments to forward to MLIR's option processing
-! HELP-NEXT: -mno-outline-atomics    Don't generate local calls to out-of-line atomic operations
-! HELP-NEXT: -module-dir <dir>       Put MODULE files in <dir>
-! HELP-NEXT: -moutline-atomics       Generate local calls to out-of-line atomic operations
-! HELP-NEXT: -mrvv-vector-bits=<value>
-! HELP-NEXT:                         Specify the size in bits of an RVV vector register
-! HELP-NEXT: -msve-vector-bits=<value>
-! HELP-NEXT:                          Specify the size in bits of an SVE vector register. Defaults to the vector length agnostic value of "scalable". (AArch64 only)
-! HELP-NEXT: --no-offload-arch=<value>
-! HELP-NEXT:                         Remove CUDA/HIP offloading device architecture (e.g. sm_35, gfx906) from the list of devices to compile for. 'all' resets the list to its default value.
-! HELP-NEXT: -nocpp                  Disable predefined and command line preprocessor macros
-! HELP-NEXT: -nogpulib               Do not link device library for CUDA/HIP device compilation
-! HELP-NEXT: --offload-arch=<value>  Specify an offloading device architecture for CUDA, HIP, or OpenMP. (e.g. sm_35). If 'native' is used the compiler will detect locally installed architectures. For HIP offloading, the device architecture can be followed by target ID features delimited by a colon (e.g. gfx908:xnack+:sramecc-). May be specified more than once.
-! HELP-NEXT: --offload-device-only   Only compile for the offloading device.
-! HELP-NEXT: --offload-host-device   Compile for both the offloading host and device (default).
-! HELP-NEXT: --offload-host-only     Only compile for the offloading host.
-! HELP-NEXT: -o <file>               Write output to <file>
-! HELP-NEXT: -pedantic               Warn on language extensions
-! HELP-NEXT: -print-effective-triple Print the effective target triple
-! HELP-NEXT: -print-target-triple    Print the normalized target triple
-! HELP-NEXT: -pthread                Support POSIX threads in generated code
-! HELP-NEXT: -P                      Disable linemarker output in -E mode
-! HELP-NEXT:  --rocm-path=<value> ROCm installation path, used for finding and automatically linking required bitcode libraries.
-! HELP-NEXT: -Rpass-analysis=<value> Report transformation analysis from optimization passes whose name matches the given POSIX regular expression
-! HELP-NEXT: -Rpass-missed=<value>   Report missed transformations by optimization passes whose name matches the given POSIX regular expression
-! HELP-NEXT: -Rpass=<value>          Report transformations performed by optimization passes whose name matches the given POSIX regular expression
-! HELP-NEXT: -R<remark>              Enable the specified remark
-! HELP-NEXT: -save-temps=<value>     Save intermediate compilation results.
-! HELP-NEXT: -save-temps             Alias for --save-temps=cwd
-! HELP-NEXT: -std=<value>            Language standard to compile for
-! HELP-NEXT: -S                      Only run preprocess and compilation steps
-! HELP-NEXT: --target=<value>        Generate code for the given target
-! HELP-NEXT: -U <macro>              Undefine macro <macro>
-! HELP-NEXT: --version               Print version information
-! HELP-NEXT: -v                      Show commands to run and use verbose output
-! HELP-NEXT: -Wl,<arg>               Pass the comma separated arguments in <arg> to the linker
-! HELP-NEXT: -W<warning>             Enable the specified warning
-! HELP-NEXT: -Xflang <arg>           Pass <arg> to the flang compiler
-! HELP-NEXT: -x <language>           Treat subsequent input files as having type <language>
-
 
 ! HELP-FC1:USAGE: flang
 ! HELP-FC1-EMPTY:
 ! HELP-FC1-NEXT:OPTIONS:
-! HELP-FC1-NEXT: -cpp                    Enable predefined and command line preprocessor macros
-! HELP-FC1-NEXT: --dependent-lib=<value> Add dependent library
-! HELP-FC1-NEXT: -dM                     Print macro definitions in -E mode instead of normal output
-! HELP-FC1-NEXT: -D <macro>=<value>      Define <macro> to <value> (or 1 if <value> omitted)
-! HELP-FC1-NEXT: -emit-fir               Build the parse tree, then lower it to FIR
-! HELP-FC1-NEXT: -emit-hlfir             Build the parse tree, then lower it to HLFIR
-! HELP-FC1-NEXT: -emit-llvm-bc           Build ASTs then convert to LLVM, emit .bc file
-! HELP-FC1-NEXT: -emit-llvm              Use the LLVM representation for assembler and object files
-! HELP-FC1-NEXT: -emit-obj               Emit native object files
-! HELP-FC1-NEXT: -E                      Only run the preprocessor
-! HELP-FC1-NEXT: -falternative-parameter-statement
-! HELP-FC1-NEXT:                         Enable the old style PARAMETER statement
-! HELP-FC1-NEXT: -fapprox-func           Allow certain math function calls to be replaced with an approximately equivalent calculation
-! HELP-FC1-NEXT: -fbackslash             Specify that backslash in string introduces an escape character
-! HELP-FC1-NEXT: -fcolor-diagnostics     Enable colors in diagnostics
-! HELP-FC1-NEXT: -fconvert=<value>       Set endian conversion of data for unformatted files
-! HELP-FC1-NEXT: -fdebug-dump-all        Dump symbols and the parse tree after the semantic checks
-! HELP-FC1-NEXT: -fdebug-dump-parse-tree-no-sema
-! HELP-FC1-NEXT:                         Dump the parse tree (skips the semantic checks)
-! HELP-FC1-NEXT: -fdebug-dump-parse-tree Dump the parse tree
-! HELP-FC1-NEXT: -fdebug-dump-parsing-log
-! HELP-FC1-NEXT:                         Run instrumented parse and dump the parsing log
-! HELP-FC1-NEXT: -fdebug-dump-pft        Dump the pre-fir parse tree
-! HELP-FC1-NEXT: -fdebug-dump-provenance Dump provenance
-! HELP-FC1-NEXT: -fdebug-dump-symbols    Dump symbols after the semantic analysis
-! HELP-FC1-NEXT: -fdebug-measure-parse-tree
-! HELP-FC1-NEXT:                         Measure the parse tree
-! HELP-FC1-NEXT: -fdebug-module-writer   Enable debug messages while writing module files
-! HELP-FC1-NEXT: -fdebug-pass-manager    Prints debug information for the new pass manager
-! HELP-FC1-NEXT: -fdebug-pre-fir-tree    Dump the pre-FIR tree
-! HELP-FC1-NEXT: -fdebug-unparse-no-sema Unparse and stop (skips the semantic checks)
-! HELP-FC1-NEXT: -fdebug-unparse-with-symbols
-! HELP-FC1-NEXT:                         Unparse and stop.
-! HELP-FC1-NEXT: -fdebug-unparse         Unparse and stop.
-! HELP-FC1-NEXT: -fdefault-double-8      Set the default double precision kind to an 8 byte wide type
-! HELP-FC1-NEXT: -fdefault-integer-8     Set the default integer and logical kind to an 8 byte wide type
-! HELP-FC1-NEXT: -fdefault-real-8        Set the default real kind to an 8 byte wide type
-! HELP-FC1-NEXT: -fembed-offload-object=<value>
-! HELP-FC1-NEXT:                         Embed Offloading device-side binary into host object file as a section.
-! HELP-FC1-NEXT: -ffast-math             Allow aggressive, lossy floating-point optimizations
-! HELP-FC1-NEXT: -ffixed-form            Process source files in fixed form
-! HELP-FC1-NEXT: -ffixed-line-length=<value>
-! HELP-FC1-NEXT:                         Use <value> as character line width in fixed mode
-! HELP-FC1-NEXT: -ffp-contract=<value>   Form fused FP ops (e.g. FMAs)
-! HELP-FC1-NEXT: -ffree-form             Process source files in free form
-! HELP-FC1-NEXT: -fget-definition <value> <value> <value>
-! HELP-FC1-NEXT:                         Get the symbol definition from <line> <start-column> <end-column>
-! HELP-FC1-NEXT: -fget-symbols-sources   Dump symbols and their source code locations
-! HELP-FC1-NEXT: -fimplicit-none         No implicit typing allowed unless overridden by IMPLICIT statements
-! HELP-FC1-NEXT: -finput-charset=<value> Specify the default character set for source files
-! HELP-FC1-NEXT: -fintrinsic-modules-path <dir>
-! HELP-FC1-NEXT:                         Specify where to find the compiled intrinsic modules
-! HELP-FC1-NEXT: -flarge-sizes           Use INTEGER(KIND=8) for the result type in size-related intrinsics
-! HELP-FC1-NEXT: -flogical-abbreviations Enable logical abbreviations
-! HELP-FC1-NEXT: -flto=<value>           Set LTO mode
-! HELP-FC1-NEXT: -flto                   Enable LTO in 'full' mode
-! HELP-FC1-NEXT: -fno-analyzed-objects-for-unparse
-! HELP-FC1-NEXT:                         Do not use the analyzed objects when unparsing
-! HELP-FC1-NEXT: -fno-automatic          Implies the SAVE attribute for non-automatic local objects in subprograms unless RECURSIVE
-! HELP-FC1-NEXT: -fno-debug-pass-manager Disables debug printing for the new pass manager
-! HELP-FC1-NEXT: -fno-ppc-native-vector-element-order
-! HELP-FC1-NEXT:                         Specifies PowerPC non-native vector element order
-! HELP-FC1-NEXT: -fno-reformat           Dump the cooked character stream in -E mode
-! HELP-FC1-NEXT: -fno-signed-zeros       Allow optimizations that ignore the sign of floating point zeros
-! HELP-FC1-NEXT: -fno-stack-arrays       Allocate array temporaries on the heap (default)
-! HELP-FC1-NEXT: -fno-version-loops-for-stride
-! HELP-FC1-NEXT:                         Do not create unit-strided loops (default)
-! HELP-FC1-NEXT: -fopenacc               Enable OpenACC
-! HELP-FC1-NEXT: -fopenmp-host-ir-file-path <value>
-! HELP-FC1-NEXT:                         Path to the IR file produced by the frontend for the host.
-! HELP-FC1-NEXT: -fopenmp-is-target-device
-! HELP-FC1-NEXT:                         Generate code only for an OpenMP target device.
-! HELP-FC1-NEXT: -fopenmp-target-debug   Enable debugging in the OpenMP offloading device RTL
-! HELP-FC1-NEXT: -fopenmp-version=<value>
-! HELP-FC1-NEXT:                         Set OpenMP version (e.g. 45 for OpenMP 4.5, 51 for OpenMP 5.1). Default value is 11 for Flang
-! HELP-FC1-NEXT: -fopenmp                Parse OpenMP pragmas and generate parallel code.
-! HELP-FC1-NEXT: -fpass-plugin=<dsopath> Load pass plugin from a dynamic shared object file (only with new pass manager).
-! HELP-FC1-NEXT: -fppc-native-vector-element-order
-! HELP-FC1-NEXT:                         Specifies PowerPC native vector element order (default)
-! HELP-FC1-NEXT: -freciprocal-math       Allow division operations to be reassociated
-! HELP-FC1-NEXT: -fstack-arrays          Attempt to allocate array temporaries on the stack, no matter their size
-! HELP-FC1-NEXT: -fsyntax-only           Run the preprocessor, parser and semantic analysis stages
-! HELP-FC1-NEXT: -funderscoring          Appends one trailing underscore to external names
-! HELP-FC1-NEXT: -fveclib=<value>        Use the given vector functions library
-! HELP-FC1-NEXT: -fversion-loops-for-stride
-! HELP-FC1-NEXT:                         Create unit-strided versions of loops
-! HELP-FC1-NEXT: -fxor-operator          Enable .XOR. as a synonym of .NEQV.
-! HELP-FC1-NEXT: -gpulibc                Link the LLVM C Library for GPUs
-! HELP-FC1-NEXT: -help                   Display available options
-! HELP-FC1-NEXT: -init-only              Only execute frontend initialization
-! HELP-FC1-NEXT: -I <dir>                Add directory to the end of the list of include search paths
-! HELP-FC1-NEXT: -load <dsopath>         Load the named plugin (dynamic shared object)
-! HELP-FC1-NEXT: -mcode-object-version=<value>
-! HELP-FC1-NEXT:                         Specify code object ABI version. Defaults to 5. (AMDGPU only)
-! HELP-FC1-NEXT: -menable-no-infs        Allow optimization to assume there are no infinities.
-! HELP-FC1-NEXT: -menable-no-nans        Allow optimization to assume there are no NaNs.
-! HELP-FC1-NEXT: -mframe-pointer=<value> Specify which frame pointers to retain.
-! HELP-FC1-NEXT: -mllvm <value>          Additional arguments to forward to LLVM's option processing
-! HELP-FC1-NEXT: -mmlir <value>          Additional arguments to forward to MLIR's option processing
-! HELP-FC1-NEXT: -module-dir <dir>       Put MODULE files in <dir>
-! HELP-FC1-NEXT: -module-suffix <suffix> Use <suffix> as the suffix for module files (the default value is `.mod`)
-! HELP-FC1-NEXT: -mreassociate           Allow reassociation transformations for floating-point instructions
-! HELP-FC1-NEXT: -mrelocation-model <value>
-! HELP-FC1-NEXT:                         The relocation model to use
-! HELP-FC1-NEXT: -mvscale-max=<value>    Specify the vscale maximum. Defaults to the vector length agnostic value of "0". (AArch64/RISC-V only)
-! HELP-FC1-NEXT: -mvscale-min=<value>    Specify the vscale minimum. Defaults to "1". (AArch64/RISC-V only)
-! HELP-FC1-NEXT: -nocpp                  Disable predefined and command line preprocessor macros
-! HELP-FC1-NEXT: -nogpulib               Do not link device library for CUDA/HIP device compilation
-! HELP-FC1-NEXT: -opt-record-file <value>
-! HELP-FC1-NEXT:                         File name to use for YAML optimization record output
-! HELP-FC1-NEXT: -opt-record-format <value>
-! HELP-FC1-NEXT:                         The format used for serializing remarks (default: YAML)
-! HELP-FC1-NEXT: -opt-record-passes <value>
-! HELP-FC1-NEXT:                         Only record remark information for passes whose names match the given regular expression
-! HELP-FC1-NEXT: -o <file>               Write output to <file>
-! HELP-FC1-NEXT: -pedantic               Warn on language extensions
-! HELP-FC1-NEXT: -pic-is-pie             File is for a position independent executable
-! HELP-FC1-NEXT: -pic-level <value>      Value for __PIC__
-! HELP-FC1-NEXT: -plugin <name>          Use the named plugin action instead of the default action (use "help" to list available options)
-! HELP-FC1-NEXT: -pthread                Support POSIX threads in generated code
-! HELP-FC1-NEXT: -P                      Disable linemarker output in -E mode
-! HELP-FC1-NEXT: -Rpass-analysis=<value> Report transformation analysis from optimization passes whose name matches the given POSIX regular expression
-! HELP-FC1-NEXT: -Rpass-missed=<value>   Report missed transformations by optimization passes whose name matches the given POSIX regular expression
-! HELP-FC1-NEXT: -Rpass=<value>          Report transformations performed by optimization passes whose name matches the given POSIX regular expression
-! HELP-FC1-NEXT: -R<remark>              Enable the specified remark
-! HELP-FC1-NEXT: -save-temps=<value>     Save intermediate compilation results.
-! HELP-FC1-NEXT: -save-temps             Alias for --save-temps=cwd
-! HELP-FC1-NEXT: -std=<value>            Language standard to compile for
-! HELP-FC1-NEXT: -S                      Only run preprocess and compilation steps
-! HELP-FC1-NEXT: -target-cpu <value>     Target a specific cpu type
-! HELP-FC1-NEXT: -target-feature <value> Target specific attributes
-! HELP-FC1-NEXT: -test-io                Run the InputOuputTest action. Use for development and testing only.
-! HELP-FC1-NEXT: -triple <value>         Specify target triple (e.g. i686-apple-darwin9)
-! HELP-FC1-NEXT: -U <macro>              Undefine macro <macro>
-! HELP-FC1-NEXT: -version                Print the compiler version
-! HELP-FC1-NEXT: -W<warning>             Enable the specified warning
-! HELP-FC1-NEXT: -x <language>           Treat subsequent input files as having type <language>
 
 ! ERROR: error: unknown argument '-helps'; did you mean '-help'
diff --git a/flang/test/Driver/dynamic-linker.f90 b/flang/test/Driver/dynamic-linker.f90
index 7c3f1b5a53fe..6d5c443ab75c 100644
--- a/flang/test/Driver/dynamic-linker.f90
+++ b/flang/test/Driver/dynamic-linker.f90
@@ -16,7 +16,6 @@
 ! GNU-LINKER-OPTIONS-SAME: "-shared"
 ! GNU-LINKER-OPTIONS-SAME: "-static"
 ! GNU-LINKER-OPTIONS-SAME: "-rpath" "/path/to/dir"
-! GNU-LINKER-OPTIONS-NOT: "-lFortran_main.a"
 
 ! RDYNAMIC-LINKER-OPTION: "{{.*}}ld"
 ! RDYNAMIC-LINKER-OPTION-SAME: "-export-dynamic"
@@ -25,4 +24,3 @@
 ! MSVC-LINKER-OPTIONS: "{{.*}}link{{(.exe)?}}"
 ! MSVC-LINKER-OPTIONS-SAME: "-dll"
 ! MSVC-LINKER-OPTIONS-SAME: "-rpath" "/path/to/dir"
-! MSVC-LINKER-OPTIONS-NOT: "/WHOLEARCHIVE:Fortran_main"
diff --git a/flang/test/Driver/emit-mlir.f90 b/flang/test/Driver/emit-mlir.f90
index 191ee13396ef..83bb8fc1eddc 100644
--- a/flang/test/Driver/emit-mlir.f90
+++ b/flang/test/Driver/emit-mlir.f90
@@ -19,6 +19,16 @@
 ! CHECK-NEXT:  %[[VAL_0:.*]] = fir.zero_bits !fir.ref<tuple<i[[int_size]], !fir.ref<!fir.array<0xtuple<!fir.ref<i8>, !fir.ref<i8>>>>>>
 ! CHECK-NEXT: fir.has_value  %[[VAL_0]] : !fir.ref<tuple<i[[int_size]], !fir.ref<!fir.array<0xtuple<!fir.ref<i8>, !fir.ref<i8>>>>>>
 ! CHECK-NEXT: }
+! CHECK-NEXT: func.func private @_FortranAProgramStart(i32, !llvm.ptr, !llvm.ptr, !llvm.ptr)
+! CHECK-NEXT: func.func private @_FortranAProgramEndStatement()
+! CHECK-NEXT: func.func @main(%arg0: i32, %arg1: !llvm.ptr, %arg2: !llvm.ptr) -> i32 {
+! CHECK-NEXT: %c0_i32 = arith.constant 0 : i32
+! CHECK-NEXT: %0 = fir.address_of(@_QQEnvironmentDefaults) : !fir.ref<tuple<i32, !fir.ref<!fir.array<0xtuple<!fir.ref<i8>, !fir.ref<i8>>>>>>
+! CHECK-NEXT: ir.call @_FortranAProgramStart(%arg0, %arg1, %arg2, %0) {{.*}} : (i32, !llvm.ptr, !llvm.ptr, !fir.ref<tuple<i32, !fir.ref<!fir.array<0xtuple<!fir.ref<i8>, !fir.ref<i8>>>>>>) 
+! CHECK-NEXT: fir.call @_QQmain() fastmath<contract> : () -> ()
+! CHECK-NEXT: fir.call @_FortranAProgramEndStatement() {{.*}} : () -> ()
+! CHECK-NEXT: return %c0_i32 : i32
+! CHECK-NEXT: }
 ! CHECK-NEXT: }
 
 end program
diff --git a/flang/test/Driver/linker-flags.f90 b/flang/test/Driver/linker-flags.f90
index 4d3d528b5e99..02e217494f81 100644
--- a/flang/test/Driver/linker-flags.f90
+++ b/flang/test/Driver/linker-flags.f90
@@ -11,7 +11,6 @@
 ! RUN: %flang -### --target=x86_64-unknown-dragonfly %S/Inputs/hello.f90 2>&1 | FileCheck %s --check-prefixes=CHECK,UNIX,UNIX-F128%f128-lib
 ! RUN: %flang -### --target=x86_64-unknown-haiku %S/Inputs/hello.f90 2>&1 | FileCheck %s --check-prefixes=CHECK,HAIKU,HAIKU-F128%f128-lib
 ! RUN: %flang -### --target=x86_64-windows-gnu %S/Inputs/hello.f90 2>&1 | FileCheck %s --check-prefixes=CHECK,MINGW,MINGW-F128%f128-lib
-! RUN: %flang -### --target=aarch64-unknown-linux-gnu %S/Inputs/hello.f90 -lFortran_main 2>&1 | FileCheck %s --check-prefixes=DEPRECATED
 
 ! NOTE: Clang's driver library, clangDriver, usually adds 'oldnames' on Windows,
 !       but it is not needed when compiling Fortran code and they might bring in
@@ -29,7 +28,6 @@
 !       executable and may find the GNU linker from MinGW or Cygwin.
 ! UNIX-LABEL:  "{{.*}}ld{{(\.exe)?}}"
 ! UNIX-SAME: "[[object_file]]"
-! UNIX-SAME: "--whole-archive" "-lFortran_main" "--no-whole-archive"
 ! UNIX-F128NONE-NOT: FortranFloat128Math
 ! SOLARIS-F128NONE-NOT: FortranFloat128Math
 ! UNIX-F128LIBQUADMATH-SAME: "-lFortranFloat128Math" "--as-needed" "-lquadmath" "--no-as-needed"
@@ -38,7 +36,6 @@
 
 ! DARWIN-LABEL:  "{{.*}}ld{{(\.exe)?}}"
 ! DARWIN-SAME: "[[object_file]]"
-! DARWIN-SAME: -lFortran_main
 ! DARWIN-F128NONE-NOT: FortranFloat128Math
 ! DARWIN-F128LIBQUADMATH-SAME: "-lFortranFloat128Math" "--as-needed" "-lquadmath" "--no-as-needed"
 ! DARWIN-SAME: -lFortranRuntime
@@ -46,14 +43,12 @@
 
 ! HAIKU-LABEL:  "{{.*}}ld{{(\.exe)?}}"
 ! HAIKU-SAME: "[[object_file]]"
-! HAIKU-SAME: "--whole-archive" "-lFortran_main" "--no-whole-archive"
 ! HAIKU-F128NONE-NOT: FortranFloat128Math
 ! HAIKU-F128LIBQUADMATH-SAME: "-lFortranFloat128Math" "--as-needed" "-lquadmath" "--no-as-needed"
 ! HAIKU-SAME: "-lFortranRuntime" "-lFortranDecimal"
 
 ! MINGW-LABEL:  "{{.*}}ld{{(\.exe)?}}"
 ! MINGW-SAME: "[[object_file]]"
-! MINGW-SAME: -lFortran_main
 ! MINGW-F128NONE-NOT: FortranFloat128Math
 ! MINGW-F128LIBQUADMATH-SAME: "-lFortranFloat128Math" "--as-needed" "-lquadmath" "--no-as-needed"
 ! MINGW-SAME: -lFortranRuntime
@@ -66,6 +61,3 @@
 ! MSVC-LABEL: link
 ! MSVC-SAME: /subsystem:console
 ! MSVC-SAME: "[[object_file]]"
-
-! Check that we warn when using -lFortran_main
-! DEPRECATED: warning: argument '-lFortran_main' is deprecated, see the Flang driver documentation for correct usage [-Wdeprecated]
diff --git a/flang/test/Driver/mlir-debug-pass-pipeline.f90 b/flang/test/Driver/mlir-debug-pass-pipeline.f90
index ef84cb80ecf1..2c81441e7ec9 100644
--- a/flang/test/Driver/mlir-debug-pass-pipeline.f90
+++ b/flang/test/Driver/mlir-debug-pass-pipeline.f90
@@ -39,9 +39,16 @@ end program
 ! ALL-NEXT:   (S) 0 num-cse'd - Number of operations CSE'd
 ! ALL-NEXT:   (S) 0 num-dce'd - Number of operations DCE'd
 
+! ALL-NEXT: Pipeline Collection : ['fir.global', 'func.func', 'omp.declare_reduction', 'omp.private']
+! ALL-NEXT: 'fir.global' Pipeline
+! ALL-NEXT:   CharacterConversion
 ! ALL-NEXT: 'func.func' Pipeline
 ! ALL-NEXT:   ArrayValueCopy
 ! ALL-NEXT:   CharacterConversion
+! ALL-NEXT: 'omp.declare_reduction' Pipeline
+! ALL-NEXT:   CharacterConversion
+! ALL-NEXT: 'omp.private' Pipeline
+! ALL-NEXT:   CharacterConversion
 
 ! ALL-NEXT: Canonicalizer
 ! ALL-NEXT: SimplifyRegionLite
@@ -58,12 +65,16 @@ end program
 ! ALL-NEXT:   (S) 0 num-cse'd - Number of operations CSE'd
 ! ALL-NEXT:   (S) 0 num-dce'd - Number of operations DCE'd
 
-! ALL-NEXT: Pipeline Collection : ['func.func', 'omp.declare_reduction']
+! ALL-NEXT: Pipeline Collection : ['fir.global', 'func.func', 'omp.declare_reduction', 'omp.private']
+! ALL-NEXT:   'fir.global' Pipeline
+! ALL-NEXT:     CFGConversion
 ! ALL-NEXT:   'func.func' Pipeline
 ! ALL-NEXT:     PolymorphicOpConversion
-! ALL-NEXT:     CFGConversionOnFunc
+! ALL-NEXT:     CFGConversion
 ! ALL-NEXT:   'omp.declare_reduction' Pipeline
-! ALL-NEXT:     CFGConversionOnReduction
+! ALL-NEXT:     CFGConversion
+! ALL-NEXT:   'omp.private' Pipeline
+! ALL-NEXT:     CFGConversion
 ! ALL-NEXT: SCFToControlFlow
 ! ALL-NEXT: Canonicalizer
 ! ALL-NEXT: SimplifyRegionLite
@@ -72,13 +83,15 @@ end program
 ! ALL-NEXT:   (S) 0 num-dce'd - Number of operations DCE'd
 ! ALL-NEXT: BoxedProcedurePass
 
-! ALL-NEXT: Pipeline Collection : ['fir.global', 'func.func', 'omp.declare_reduction']
+! ALL-NEXT: Pipeline Collection : ['fir.global', 'func.func', 'omp.declare_reduction', 'omp.private']
 ! ALL-NEXT:   'fir.global' Pipeline
 ! ALL-NEXT:     AbstractResultOpt
 ! ALL-NEXT:   'func.func' Pipeline
 ! ALL-NEXT:     AbstractResultOpt
 ! ALL-NEXT:   'omp.declare_reduction' Pipeline
 ! ALL-NEXT:     AbstractResultOpt
+! ALL-NEXT:   'omp.private' Pipeline
+! ALL-NEXT:     AbstractResultOpt
 
 ! ALL-NEXT: CodeGenRewrite
 ! ALL-NEXT:   (S) 0 num-dce'd - Number of operations eliminated
diff --git a/flang/test/Driver/mlir-pass-pipeline.f90 b/flang/test/Driver/mlir-pass-pipeline.f90
index d1ff2869b0a6..320467a2ac2a 100644
--- a/flang/test/Driver/mlir-pass-pipeline.f90
+++ b/flang/test/Driver/mlir-pass-pipeline.f90
@@ -28,9 +28,16 @@ end program
 ! ALL-NEXT:   (S) 0 num-cse'd - Number of operations CSE'd
 ! ALL-NEXT:   (S) 0 num-dce'd - Number of operations DCE'd
 
+! ALL-NEXT: Pipeline Collection : ['fir.global', 'func.func', 'omp.declare_reduction', 'omp.private']
+! ALL-NEXT: 'fir.global' Pipeline
+! ALL-NEXT:   CharacterConversion
 ! ALL-NEXT: 'func.func' Pipeline
 ! ALL-NEXT:   ArrayValueCopy
 ! ALL-NEXT:   CharacterConversion
+! ALL-NEXT: 'omp.declare_reduction' Pipeline
+! ALL-NEXT:   CharacterConversion
+! ALL-NEXT: 'omp.private' Pipeline
+! ALL-NEXT:   CharacterConversion
 
 ! ALL-NEXT: Canonicalizer
 ! ALL-NEXT: SimplifyRegionLite
@@ -52,12 +59,16 @@ end program
 ! O2-NEXT:    'func.func' Pipeline
 ! O2-NEXT:      PolymorphicOpConversion
 ! O2-NEXT:  AddAliasTags
-! ALL-NEXT: Pipeline Collection : ['func.func', 'omp.declare_reduction']
+! ALL-NEXT: Pipeline Collection : ['fir.global', 'func.func', 'omp.declare_reduction', 'omp.private']
+! ALL-NEXT:    'fir.global' Pipeline
+! ALL-NEXT:      CFGConversion
 ! ALL-NEXT:    'func.func' Pipeline
 ! NOTO2-NEXT:      PolymorphicOpConversion
-! ALL-NEXT:      CFGConversionOnFunc
+! ALL-NEXT:      CFGConversion
 ! ALL-NEXT:   'omp.declare_reduction' Pipeline
-! ALL-NEXT:      CFGConversionOnReduction
+! ALL-NEXT:      CFGConversion
+! ALL-NEXT:   'omp.private' Pipeline
+! ALL-NEXT:      CFGConversion
 
 ! ALL-NEXT: SCFToControlFlow
 ! ALL-NEXT: Canonicalizer
@@ -67,13 +78,15 @@ end program
 ! ALL-NEXT:   (S) 0 num-dce'd - Number of operations DCE'd
 ! ALL-NEXT: BoxedProcedurePass
 
-! ALL-NEXT: Pipeline Collection : ['fir.global', 'func.func', 'omp.declare_reduction']
+! ALL-NEXT: Pipeline Collection : ['fir.global', 'func.func', 'omp.declare_reduction', 'omp.private']
 ! ALL-NEXT:   'fir.global' Pipeline
 ! ALL-NEXT:    AbstractResultOpt
 ! ALL-NEXT:  'func.func' Pipeline
 ! ALL-NEXT:    AbstractResultOpt
 ! ALL-NEXT:  'omp.declare_reduction' Pipeline
 ! ALL-NEXT:    AbstractResultOpt
+! ALL-NEXT:  'omp.private' Pipeline
+! ALL-NEXT:    AbstractResultOpt
 
 ! ALL-NEXT: CodeGenRewrite
 ! ALL-NEXT:   (S) 0 num-dce'd - Number of operations eliminated
diff --git a/flang/test/Driver/msvc-dependent-lib-flags.f90 b/flang/test/Driver/msvc-dependent-lib-flags.f90
index 6cfc969e92b2..765917f07d8e 100644
--- a/flang/test/Driver/msvc-dependent-lib-flags.f90
+++ b/flang/test/Driver/msvc-dependent-lib-flags.f90
@@ -7,7 +7,6 @@
 ! MSVC-SAME: --dependent-lib=clang_rt.builtins.lib

 ! MSVC-SAME: -D_MT

 ! MSVC-SAME: --dependent-lib=libcmt

-! MSVC-SAME: --dependent-lib=Fortran_main.static.lib

 ! MSVC-SAME: --dependent-lib=FortranRuntime.static.lib

 ! MSVC-SAME: --dependent-lib=FortranDecimal.static.lib

 

@@ -16,7 +15,6 @@
 ! MSVC-DEBUG-SAME: -D_MT

 ! MSVC-DEBUG-SAME: -D_DEBUG

 ! MSVC-DEBUG-SAME: --dependent-lib=libcmtd

-! MSVC-DEBUG-SAME: --dependent-lib=Fortran_main.static_dbg.lib

 ! MSVC-DEBUG-SAME: --dependent-lib=FortranRuntime.static_dbg.lib

 ! MSVC-DEBUG-SAME: --dependent-lib=FortranDecimal.static_dbg.lib

 

@@ -25,7 +23,6 @@
 ! MSVC-DLL-SAME: -D_MT

 ! MSVC-DLL-SAME: -D_DLL

 ! MSVC-DLL-SAME: --dependent-lib=msvcrt

-! MSVC-DLL-SAME: --dependent-lib=Fortran_main.dynamic.lib

 ! MSVC-DLL-SAME: --dependent-lib=FortranRuntime.dynamic.lib

 ! MSVC-DLL-SAME: --dependent-lib=FortranDecimal.dynamic.lib

 

@@ -35,6 +32,5 @@
 ! MSVC-DLL-DEBUG-SAME: -D_DEBUG

 ! MSVC-DLL-DEBUG-SAME: -D_DLL

 ! MSVC-DLL-DEBUG-SAME: --dependent-lib=msvcrtd

-! MSVC-DLL-DEBUG-SAME: --dependent-lib=Fortran_main.dynamic_dbg.lib

 ! MSVC-DLL-DEBUG-SAME: --dependent-lib=FortranRuntime.dynamic_dbg.lib

 ! MSVC-DLL-DEBUG-SAME: --dependent-lib=FortranDecimal.dynamic_dbg.lib

diff --git a/flang/test/Driver/no-duplicate-main.f90 b/flang/test/Driver/no-duplicate-main.f90
index 88f4430828e0..b0bb6c2a2fef 100644
--- a/flang/test/Driver/no-duplicate-main.f90
+++ b/flang/test/Driver/no-duplicate-main.f90
@@ -4,8 +4,6 @@
 ! RUN: %flang -o %t -c %s
 ! RUN: not %flang -o %t.exe %t %t.c-object 2>&1
 
-! RUN: %flang -fno-fortran-main -o %t.exe %t %t.c-object 2>&1
-
 ! TODO: potentially add further checks to ensure that proper
 !       linker error messages are detected and checked via
 !       FileCheck.
diff --git a/flang/test/Fir/array-value-copy-2.fir b/flang/test/Fir/array-value-copy-2.fir
index cb8d6ca2b055..21b340af10c6 100644
--- a/flang/test/Fir/array-value-copy-2.fir
+++ b/flang/test/Fir/array-value-copy-2.fir
@@ -1,5 +1,5 @@
-// RUN: fir-opt --array-value-copy --cfg-conversion-on-func-opt %s | FileCheck %s
-// RUN: fir-opt --array-value-copy="optimize-conflicts=true" --cfg-conversion-on-func-opt %s | FileCheck %s
+// RUN: fir-opt --array-value-copy --cfg-conversion %s | FileCheck %s
+// RUN: fir-opt --array-value-copy="optimize-conflicts=true" --cfg-conversion %s | FileCheck %s
 
 // CHECK-LABEL: func @_QPslice1(
 // CHECK-NOT: fir.allocmem
diff --git a/flang/test/Fir/basic-program.fir b/flang/test/Fir/basic-program.fir
index 28c597fc918c..d54b0895cc33 100644
--- a/flang/test/Fir/basic-program.fir
+++ b/flang/test/Fir/basic-program.fir
@@ -34,9 +34,16 @@ func.func @_QQmain() {
 // PASSES-NEXT:   (S) 0 num-cse'd - Number of operations CSE'd
 // PASSES-NEXT:   (S) 0 num-dce'd - Number of operations DCE'd
 
+// PASSES-NEXT: Pipeline Collection : ['fir.global', 'func.func', 'omp.declare_reduction', 'omp.private']
+// PASSES-NEXT: 'fir.global' Pipeline
+// PASSES-NEXT:   CharacterConversion
 // PASSES-NEXT: 'func.func' Pipeline
 // PASSES-NEXT:   ArrayValueCopy
 // PASSES-NEXT:   CharacterConversion
+// PASSES-NEXT: 'omp.declare_reduction' Pipeline
+// PASSES-NEXT:   CharacterConversion
+// PASSES-NEXT: 'omp.private' Pipeline
+// PASSES-NEXT:   CharacterConversion
 
 // PASSES-NEXT: Canonicalizer
 // PASSES-NEXT: SimplifyRegionLite
@@ -60,11 +67,15 @@ func.func @_QQmain() {
 
 // PASSES-NEXT: AddAliasTags
 
-// PASSES-NEXT: Pipeline Collection : ['func.func', 'omp.declare_reduction']
+// PASSES-NEXT: Pipeline Collection : ['fir.global', 'func.func', 'omp.declare_reduction', 'omp.private']
+// PASSES-NEXT: 'fir.global' Pipeline
+// PASSES-NEXT:   CFGConversion
 // PASSES-NEXT: 'func.func' Pipeline
-// PASSES-NEXT:   CFGConversionOnFunc
+// PASSES-NEXT:   CFGConversion
 // PASSES-NEXT: 'omp.declare_reduction' Pipeline
-// PASSES-NEXT:   CFGConversionOnReduction
+// PASSES-NEXT:   CFGConversion
+// PASSES-NEXT: 'omp.private' Pipeline
+// PASSES-NEXT:   CFGConversion
 
 // PASSES-NEXT: SCFToControlFlow
 // PASSES-NEXT: Canonicalizer
@@ -74,13 +85,15 @@ func.func @_QQmain() {
 // PASSES-NEXT:   (S) 0 num-dce'd - Number of operations DCE'd
 // PASSES-NEXT: BoxedProcedurePass
 
-// PASSES-NEXT: Pipeline Collection : ['fir.global', 'func.func', 'omp.declare_reduction']
-// PASSES-NEXT:   'fir.global' Pipeline
+// PASSES-NEXT: Pipeline Collection : ['fir.global', 'func.func', 'omp.declare_reduction', 'omp.private']
+// PASSES-NEXT:  'fir.global' Pipeline
 // PASSES-NEXT:    AbstractResultOpt
 // PASSES-NEXT:  'func.func' Pipeline
 // PASSES-NEXT:    AbstractResultOpt
 // PASSES-NEXT:  'omp.declare_reduction' Pipeline
 // PASSES-NEXT:    AbstractResultOpt
+// PASSES-NEXT:  'omp.private' Pipeline
+// PASSES-NEXT:    AbstractResultOpt
 
 // PASSES-NEXT: CodeGenRewrite
 // PASSES-NEXT:   (S) 0 num-dce'd - Number of operations eliminated
diff --git a/flang/test/Fir/convert-to-llvm-openmp-and-fir.fir b/flang/test/Fir/convert-to-llvm-openmp-and-fir.fir
index fa7979e8875a..8cf4f566964f 100644
--- a/flang/test/Fir/convert-to-llvm-openmp-and-fir.fir
+++ b/flang/test/Fir/convert-to-llvm-openmp-and-fir.fir
@@ -1,4 +1,4 @@
-// RUN: fir-opt --split-input-file --cfg-conversion-on-func-opt --fir-to-llvm-ir="target=aarch64-unknown-linux-gnu" %s | FileCheck %s
+// RUN: fir-opt --split-input-file --cfg-conversion --fir-to-llvm-ir="target=aarch64-unknown-linux-gnu" %s | FileCheck %s
 
 func.func @_QPsb1(%arg0: !fir.ref<i32> {fir.bindc_name = "n"}, %arg1: !fir.ref<!fir.array<?xi32>> {fir.bindc_name = "arr"}) {
   %c1_i64 = arith.constant 1 : i64
@@ -7,15 +7,17 @@ func.func @_QPsb1(%arg0: !fir.ref<i32> {fir.bindc_name = "n"}, %arg1: !fir.ref<!
   omp.parallel  {
     %1 = fir.alloca i32 {adapt.valuebyref, pinned}
     %2 = fir.load %arg0 : !fir.ref<i32>
-    omp.wsloop nowait
-    for (%arg2) : i32 = (%c1_i32) to (%2) inclusive step (%c1_i32)  {
-      fir.store %arg2 to %1 : !fir.ref<i32>
-      %3 = fir.load %1 : !fir.ref<i32>
-      %4 = fir.convert %3 : (i32) -> i64
-      %5 = arith.subi %4, %c1_i64 : i64
-      %6 = fir.coordinate_of %arg1, %5 : (!fir.ref<!fir.array<?xi32>>, i64) -> !fir.ref<i32>
-      fir.store %3 to %6 : !fir.ref<i32>
-      omp.yield
+    omp.wsloop nowait {
+      omp.loop_nest (%arg2) : i32 = (%c1_i32) to (%2) inclusive step (%c1_i32)  {
+        fir.store %arg2 to %1 : !fir.ref<i32>
+        %3 = fir.load %1 : !fir.ref<i32>
+        %4 = fir.convert %3 : (i32) -> i64
+        %5 = arith.subi %4, %c1_i64 : i64
+        %6 = fir.coordinate_of %arg1, %5 : (!fir.ref<!fir.array<?xi32>>, i64) -> !fir.ref<i32>
+        fir.store %3 to %6 : !fir.ref<i32>
+        omp.yield
+      }
+      omp.terminator
     }
     omp.terminator
   }
@@ -31,7 +33,7 @@ func.func @_QPsb1(%arg0: !fir.ref<i32> {fir.bindc_name = "n"}, %arg1: !fir.ref<!
 // CHECK:      %[[I_VAR:.*]] = llvm.alloca %[[ONE_3]] x i32 {pinned} : (i64) -> !llvm.ptr
 // CHECK:      %[[N:.*]] = llvm.load %[[N_REF]] : !llvm.ptr -> i32
 // CHECK: omp.wsloop nowait
-// CHECK-SAME: for (%[[I:.*]]) : i32 = (%[[ONE_2]]) to (%[[N]]) inclusive step (%[[ONE_2]]) {
+// CHECK-NEXT: omp.loop_nest (%[[I:.*]]) : i32 = (%[[ONE_2]]) to (%[[N]]) inclusive step (%[[ONE_2]]) {
 // CHECK:   llvm.store %[[I]], %[[I_VAR]] : i32, !llvm.ptr
 // CHECK:   %[[I1:.*]] = llvm.load %[[I_VAR]] : !llvm.ptr -> i32
 // CHECK:   %[[I1_EXT:.*]] = llvm.sext %[[I1]] : i32 to i64
@@ -42,6 +44,8 @@ func.func @_QPsb1(%arg0: !fir.ref<i32> {fir.bindc_name = "n"}, %arg1: !fir.ref<!
 // CHECK: }
 // CHECK: omp.terminator
 // CHECK: }
+// CHECK: omp.terminator
+// CHECK: }
 // CHECK: llvm.return
 // CHECK: }
 
@@ -79,13 +83,16 @@ func.func @_QPsb(%arr: !fir.box<!fir.array<?xi32>> {fir.bindc_name = "arr"}) {
   omp.parallel   {
     %c1 = arith.constant 1 : i32
     %c50 = arith.constant 50 : i32
-    omp.wsloop   for  (%indx) : i32 = (%c1) to (%c50) inclusive step (%c1) {
-      %1 = fir.convert %indx : (i32) -> i64
-      %c1_i64 = arith.constant 1 : i64
-      %2 = arith.subi %1, %c1_i64 : i64
-      %3 = fir.coordinate_of %arr, %2 : (!fir.box<!fir.array<?xi32>>, i64) -> !fir.ref<i32>
-      fir.store %indx to %3 : !fir.ref<i32>
-      omp.yield
+    omp.wsloop {
+      omp.loop_nest (%indx) : i32 = (%c1) to (%c50) inclusive step (%c1) {
+        %1 = fir.convert %indx : (i32) -> i64
+        %c1_i64 = arith.constant 1 : i64
+        %2 = arith.subi %1, %c1_i64 : i64
+        %3 = fir.coordinate_of %arr, %2 : (!fir.box<!fir.array<?xi32>>, i64) -> !fir.ref<i32>
+        fir.store %indx to %3 : !fir.ref<i32>
+        omp.yield
+      }
+      omp.terminator
     }
     omp.terminator
   }
@@ -98,9 +105,11 @@ func.func @_QPsb(%arr: !fir.box<!fir.array<?xi32>> {fir.bindc_name = "arr"}) {
 // CHECK:    omp.parallel   {
 // CHECK:      %[[C1:.*]] = llvm.mlir.constant(1 : i32) : i32
 // CHECK:      %[[C50:.*]] = llvm.mlir.constant(50 : i32) : i32
-// CHECK:      omp.wsloop   for  (%[[INDX:.*]]) : i32 = (%[[C1]]) to (%[[C50]]) inclusive step (%[[C1]]) {
-// CHECK:        llvm.store %[[INDX]], %{{.*}} : i32, !llvm.ptr
-// CHECK:        omp.yield
+// CHECK:      omp.wsloop {
+// CHECK-NEXT:   omp.loop_nest (%[[INDX:.*]]) : i32 = (%[[C1]]) to (%[[C50]]) inclusive step (%[[C1]]) {
+// CHECK:          llvm.store %[[INDX]], %{{.*}} : i32, !llvm.ptr
+// CHECK:          omp.yield
+// CHECK:        omp.terminator
 // CHECK:      omp.terminator
 // CHECK:    llvm.return
 
@@ -708,18 +717,20 @@ func.func @_QPsb() {
 // CHECK-SAME: %[[ARRAY_REF:.*]]: !llvm.ptr
 // CHECK:    %[[RED_ACCUMULATOR:.*]] = llvm.alloca %2 x i32 {bindc_name = "x"} : (i64) -> !llvm.ptr
 // CHECK:    omp.parallel   {
-// CHECK:      omp.wsloop   reduction(@[[EQV_REDUCTION]] %[[RED_ACCUMULATOR]] -> %[[PRV:.+]] : !llvm.ptr) for
-// CHECK:        %[[ARRAY_ELEM_REF:.*]] = llvm.getelementptr %[[ARRAY_REF]][0, %{{.*}}] : (!llvm.ptr, i64) -> !llvm.ptr
-// CHECK:        %[[ARRAY_ELEM:.*]] = llvm.load %[[ARRAY_ELEM_REF]] : !llvm.ptr -> i32
-// CHECK:        %[[LPRV:.+]] = llvm.load %[[PRV]] : !llvm.ptr -> i32
-// CHECK:        %[[ZERO_1:.*]] = llvm.mlir.constant(0 : i64) : i32
-// CHECK:        %[[ARGVAL_1:.*]] = llvm.icmp "ne" %[[LPRV]], %[[ZERO_1]] : i32
-// CHECK:        %[[ZERO_2:.*]] = llvm.mlir.constant(0 : i64) : i32
-// CHECK:        %[[ARGVAL_2:.*]] = llvm.icmp "ne" %[[ARRAY_ELEM]], %[[ZERO_2]] : i32
-// CHECK:        %[[RES:.*]] = llvm.icmp "eq" %[[ARGVAL_2]], %[[ARGVAL_1]] : i1
-// CHECK:        %[[RES_EXT:.*]] = llvm.zext %[[RES]] : i1 to i32
-// CHECK:        llvm.store %[[RES_EXT]], %[[PRV]] : i32, !llvm.ptr
-// CHECK:        omp.yield
+// CHECK:      omp.wsloop reduction(@[[EQV_REDUCTION]] %[[RED_ACCUMULATOR]] -> %[[PRV:.+]] : !llvm.ptr) {
+// CHECK-NEXT:   omp.loop_nest
+// CHECK:          %[[ARRAY_ELEM_REF:.*]] = llvm.getelementptr %[[ARRAY_REF]][0, %{{.*}}] : (!llvm.ptr, i64) -> !llvm.ptr
+// CHECK:          %[[ARRAY_ELEM:.*]] = llvm.load %[[ARRAY_ELEM_REF]] : !llvm.ptr -> i32
+// CHECK:          %[[LPRV:.+]] = llvm.load %[[PRV]] : !llvm.ptr -> i32
+// CHECK:          %[[ZERO_1:.*]] = llvm.mlir.constant(0 : i64) : i32
+// CHECK:          %[[ARGVAL_1:.*]] = llvm.icmp "ne" %[[LPRV]], %[[ZERO_1]] : i32
+// CHECK:          %[[ZERO_2:.*]] = llvm.mlir.constant(0 : i64) : i32
+// CHECK:          %[[ARGVAL_2:.*]] = llvm.icmp "ne" %[[ARRAY_ELEM]], %[[ZERO_2]] : i32
+// CHECK:          %[[RES:.*]] = llvm.icmp "eq" %[[ARGVAL_2]], %[[ARGVAL_1]] : i1
+// CHECK:          %[[RES_EXT:.*]] = llvm.zext %[[RES]] : i1 to i32
+// CHECK:          llvm.store %[[RES_EXT]], %[[PRV]] : i32, !llvm.ptr
+// CHECK:          omp.yield
+// CHECK:        omp.terminator
 // CHECK:      omp.terminator
 // CHECK:    llvm.return
 
@@ -747,21 +758,24 @@ func.func @_QPsimple_reduction(%arg0: !fir.ref<!fir.array<100x!fir.logical<4>>>
     %c1_i32 = arith.constant 1 : i32
     %c100_i32 = arith.constant 100 : i32
     %c1_i32_0 = arith.constant 1 : i32
-    omp.wsloop   reduction(@eqv_reduction %1 -> %prv : !fir.ref<!fir.logical<4>>) for  (%arg1) : i32 = (%c1_i32) to (%c100_i32) inclusive step (%c1_i32_0) {
-      fir.store %arg1 to %3 : !fir.ref<i32>
-      %4 = fir.load %3 : !fir.ref<i32>
-      %5 = fir.convert %4 : (i32) -> i64
-      %c1_i64 = arith.constant 1 : i64
-      %6 = arith.subi %5, %c1_i64 : i64
-      %7 = fir.coordinate_of %arg0, %6 : (!fir.ref<!fir.array<100x!fir.logical<4>>>, i64) -> !fir.ref<!fir.logical<4>>
-      %8 = fir.load %7 : !fir.ref<!fir.logical<4>>
-      %lprv = fir.load %prv : !fir.ref<!fir.logical<4>>
-      %lprv1 = fir.convert %lprv : (!fir.logical<4>) -> i1
-      %9 = fir.convert %8 : (!fir.logical<4>) -> i1
-      %10 = arith.cmpi eq, %9, %lprv1 : i1
-      %11 = fir.convert %10 : (i1) -> !fir.logical<4>
-      fir.store %11 to %prv : !fir.ref<!fir.logical<4>>
-      omp.yield
+    omp.wsloop reduction(@eqv_reduction %1 -> %prv : !fir.ref<!fir.logical<4>>) {
+      omp.loop_nest (%arg1) : i32 = (%c1_i32) to (%c100_i32) inclusive step (%c1_i32_0) {
+        fir.store %arg1 to %3 : !fir.ref<i32>
+        %4 = fir.load %3 : !fir.ref<i32>
+        %5 = fir.convert %4 : (i32) -> i64
+        %c1_i64 = arith.constant 1 : i64
+        %6 = arith.subi %5, %c1_i64 : i64
+        %7 = fir.coordinate_of %arg0, %6 : (!fir.ref<!fir.array<100x!fir.logical<4>>>, i64) -> !fir.ref<!fir.logical<4>>
+        %8 = fir.load %7 : !fir.ref<!fir.logical<4>>
+        %lprv = fir.load %prv : !fir.ref<!fir.logical<4>>
+        %lprv1 = fir.convert %lprv : (!fir.logical<4>) -> i1
+        %9 = fir.convert %8 : (!fir.logical<4>) -> i1
+        %10 = arith.cmpi eq, %9, %lprv1 : i1
+        %11 = fir.convert %10 : (i1) -> !fir.logical<4>
+        fir.store %11 to %prv : !fir.ref<!fir.logical<4>>
+        omp.yield
+      }
+      omp.terminator
     }
     omp.terminator
   }
diff --git a/flang/test/Fir/loop01.fir b/flang/test/Fir/loop01.fir
index c849797b969e..72ca1c3989e4 100644
--- a/flang/test/Fir/loop01.fir
+++ b/flang/test/Fir/loop01.fir
@@ -1,4 +1,4 @@
-// RUN: fir-opt --split-input-file --cfg-conversion-on-func-opt %s | FileCheck %s
+// RUN: fir-opt --split-input-file --cfg-conversion %s | FileCheck %s
 
 func.func @x(%lb : index, %ub : index, %step : index, %b : i1, %addr : !fir.ref<index>) {
   fir.do_loop %iv = %lb to %ub step %step unordered {
diff --git a/flang/test/Fir/loop02.fir b/flang/test/Fir/loop02.fir
index 8918666f0b34..50948e0e7aa6 100644
--- a/flang/test/Fir/loop02.fir
+++ b/flang/test/Fir/loop02.fir
@@ -1,5 +1,5 @@
-// RUN: fir-opt --cfg-conversion-on-func-opt="always-execute-loop-body=true" %s | FileCheck %s
-// RUN: fir-opt --cfg-conversion-on-func-opt %s | FileCheck %s --check-prefix=NOOPT
+// RUN: fir-opt --cfg-conversion="always-execute-loop-body=true" %s | FileCheck %s
+// RUN: fir-opt --cfg-conversion %s | FileCheck %s --check-prefix=NOOPT
 
 func.func @x(%addr : !fir.ref<index>) {
   %bound = arith.constant 452 : index
diff --git a/flang/test/Lower/CUDA/cuda-allocatable.cuf b/flang/test/Lower/CUDA/cuda-allocatable.cuf
index 251ff16a56c7..eff5f13669e9 100644
--- a/flang/test/Lower/CUDA/cuda-allocatable.cuf
+++ b/flang/test/Lower/CUDA/cuda-allocatable.cuf
@@ -17,6 +17,15 @@ end subroutine
 
 ! CHECK: %{{.*}} = fir.cuda_deallocate %[[BOX_DECL]]#1 : !fir.ref<!fir.box<!fir.heap<!fir.array<?xf32>>>> {cuda_attr = #fir.cuda<device>} -> i32
 
+! CHECK: %[[BOX_LOAD:.*]] = fir.load %[[BOX_DECL]]#1 : !fir.ref<!fir.box<!fir.heap<!fir.array<?xf32>>>>
+! CHECK: %[[ADDR:.*]] = fir.box_addr %[[BOX_LOAD]] : (!fir.box<!fir.heap<!fir.array<?xf32>>>) -> !fir.heap<!fir.array<?xf32>>
+! CHECK: %[[ADDR_I64:.*]] = fir.convert %[[ADDR]] : (!fir.heap<!fir.array<?xf32>>) -> i64
+! CHECK: %[[C0:.*]] = arith.constant 0 : i64
+! CHECK: %[[NE_C0:.*]] = arith.cmpi ne, %[[ADDR_I64]], %[[C0]] : i64
+! CHECK: fir.if %[[NE_C0]] {
+! CHECK:   %{{.*}} = fir.cuda_deallocate %[[BOX_DECL]]#1 : !fir.ref<!fir.box<!fir.heap<!fir.array<?xf32>>>> {cuda_attr = #fir.cuda<device>} -> i32
+! CHECK: }
+
 subroutine sub2()
   real, allocatable, managed :: a(:)
   integer :: istat
@@ -37,6 +46,10 @@ end subroutine
 ! CHECK: %[[STAT:.*]] = fir.cuda_deallocate %[[BOX_DECL]]#1 : !fir.ref<!fir.box<!fir.heap<!fir.array<?xf32>>>> {cuda_attr = #fir.cuda<managed>, hasStat} -> i32
 ! CHECK: fir.store %[[STAT]] to %[[ISTAT_DECL]]#1 : !fir.ref<i32>
 
+! CHECK: fir.if %{{.*}} {
+! CHECK:   %{{.*}} = fir.cuda_deallocate %[[BOX_DECL]]#1 : !fir.ref<!fir.box<!fir.heap<!fir.array<?xf32>>>> {cuda_attr = #fir.cuda<managed>} -> i32
+! CHECK: }
+
 subroutine sub3()
   integer, allocatable, pinned :: a(:,:)
   logical :: plog
@@ -50,6 +63,9 @@ end subroutine
 ! CHECK: %[[PLOG_DECL:.*]]:2 = hlfir.declare %5 {uniq_name = "_QFsub3Eplog"} : (!fir.ref<!fir.logical<4>>) -> (!fir.ref<!fir.logical<4>>, !fir.ref<!fir.logical<4>>)
 ! CHECK-2: fir.call @_FortranAAllocatableSetBounds
 ! CHECK: %{{.*}} = fir.cuda_allocate %[[BOX_DECL]]#1 : !fir.ref<!fir.box<!fir.heap<!fir.array<?x?xi32>>>> pinned(%[[PLOG_DECL]]#1 : !fir.ref<!fir.logical<4>>) {cuda_attr = #fir.cuda<pinned>} -> i32
+! CHECK: fir.if %{{.*}} {
+! CHECK:   %{{.*}} = fir.cuda_deallocate %[[BOX_DECL]]#1 : !fir.ref<!fir.box<!fir.heap<!fir.array<?x?xi32>>>> {cuda_attr = #fir.cuda<pinned>} -> i32
+! CHECK: }
 
 subroutine sub4()
   real, allocatable, device :: a(:)
@@ -65,6 +81,9 @@ end subroutine
 ! CHECK: fir.call @_FortranAAllocatableSetBounds
 ! CHECK: %[[STREAM:.*]] = fir.load %[[ISTREAM_DECL]]#0 : !fir.ref<i32>
 ! CHECK: %{{.*}} = fir.cuda_allocate %[[BOX_DECL]]#1 : !fir.ref<!fir.box<!fir.heap<!fir.array<?xf32>>>> stream(%[[STREAM]] : i32) {cuda_attr = #fir.cuda<device>} -> i32
+! CHECK: fir.if %{{.*}} {
+! CHECK:   %{{.*}} = fir.cuda_deallocate %[[BOX_DECL]]#1 : !fir.ref<!fir.box<!fir.heap<!fir.array<?xf32>>>> {cuda_attr = #fir.cuda<device>} -> i32
+! CHECK: }
 
 subroutine sub5()
   real, allocatable, device :: a(:)
@@ -80,6 +99,11 @@ end subroutine
 ! CHECK: %[[LOAD_B:.*]] = fir.load %[[BOX_B_DECL]]#1 : !fir.ref<!fir.box<!fir.heap<!fir.array<?xf32>>>>
 ! CHECK: fir.call @_FortranAAllocatableSetBounds
 ! CHECK: %{{.*}} = fir.cuda_allocate %[[BOX_A_DECL]]#1 : !fir.ref<!fir.box<!fir.heap<!fir.array<?xf32>>>> source(%[[LOAD_B]] : !fir.box<!fir.heap<!fir.array<?xf32>>>) {cuda_attr = #fir.cuda<device>} -> i32
+! CHECK: fir.if
+! CHECK: fir.freemem
+! CHECK: fir.if %{{.*}} {
+! CHECK:   %{{.*}} = fir.cuda_deallocate %[[BOX_A_DECL]]#1 : !fir.ref<!fir.box<!fir.heap<!fir.array<?xf32>>>> {cuda_attr = #fir.cuda<device>} -> i32
+! CHECK: }
 
 subroutine sub6()
   real, allocatable, device :: a(:)
@@ -95,6 +119,9 @@ end subroutine
 ! CHECK: %[[LOAD_B:.*]] = fir.load %[[BOX_B_DECL]]#1 : !fir.ref<!fir.box<!fir.heap<!fir.array<?xf32>>>>
 ! CHECK: fir.call @_FortranAAllocatableApplyMold
 ! CHECK: %{{.*}} = fir.cuda_allocate %[[BOX_A_DECL]]#1 : !fir.ref<!fir.box<!fir.heap<!fir.array<?xf32>>>> {cuda_attr = #fir.cuda<device>} -> i32
+! CHECK: fir.if %{{.*}} {
+! CHECK:   %{{.*}} = fir.cuda_deallocate %[[BOX_A_DECL]]#1 : !fir.ref<!fir.box<!fir.heap<!fir.array<?xf32>>>> {cuda_attr = #fir.cuda<device>} -> i32
+! CHECK: }
 
 subroutine sub7()
   real, allocatable, device :: a(:)
@@ -120,3 +147,6 @@ end subroutine
 ! CHECK: %[[ERR_BOX:.*]] = fir.embox %[[ERR_DECL]]#1 : (!fir.ref<!fir.char<1,50>>) -> !fir.box<!fir.char<1,50>>
 ! CHECK: %[[STAT:.*]] = fir.cuda_deallocate %[[BOX_DECL]]#1 : !fir.ref<!fir.box<!fir.heap<!fir.array<?xf32>>>> errmsg(%15 : !fir.box<!fir.char<1,50>>) {cuda_attr = #fir.cuda<device>, hasStat} -> i32
 ! CHECK: fir.store %[[STAT]] to %[[ISTAT_DECL]]#1 : !fir.ref<i32>
+! CHECK: fir.if %{{.*}} {
+! CHECK:   %{{.*}} = fir.cuda_deallocate %[[BOX_DECL]]#1 : !fir.ref<!fir.box<!fir.heap<!fir.array<?xf32>>>> {cuda_attr = #fir.cuda<device>} -> i32
+! CHECK: }
diff --git a/flang/test/Lower/CUDA/cuda-data-transfer.cuf b/flang/test/Lower/CUDA/cuda-data-transfer.cuf
index 4ebd736315bc..0a2608639bce 100644
--- a/flang/test/Lower/CUDA/cuda-data-transfer.cuf
+++ b/flang/test/Lower/CUDA/cuda-data-transfer.cuf
@@ -98,7 +98,7 @@ end
 
 ! CHECK: %[[TEMP:.*]] = fir.allocmem !fir.array<10xi32> {bindc_name = ".tmp", uniq_name = ""}
 ! CHECK: %[[DECL_TEMP:.*]]:2 = hlfir.declare %[[TEMP]](%{{.*}}) {uniq_name = ".tmp"} : (!fir.heap<!fir.array<10xi32>>, !fir.shape<1>) -> (!fir.heap<!fir.array<10xi32>>, !fir.heap<!fir.array<10xi32>>)
-! CHECK: %[[ADEV_TEMP:.*]]:2 = hlfir.declare %21#0 {cuda_attr = #fir.cuda<device>, uniq_name = "_QFsub2Eadev"} : (!fir.heap<!fir.array<10xi32>>) -> (!fir.heap<!fir.array<10xi32>>, !fir.heap<!fir.array<10xi32>>)
+! CHECK: %[[ADEV_TEMP:.*]]:2 = hlfir.declare %[[DECL_TEMP]]#1(%{{.*}}) {cuda_attr = #fir.cuda<device>, uniq_name = "_QFsub2Eadev"} : (!fir.heap<!fir.array<10xi32>>, !fir.shape<1>) -> (!fir.heap<!fir.array<10xi32>>, !fir.heap<!fir.array<10xi32>>)
 ! CHECK: fir.cuda_data_transfer %[[ADEV]]#1 to %[[DECL_TEMP]]#0 {transfer_kind = #fir.cuda_transfer<device_host>} : !fir.ref<!fir.array<10xi32>>, !fir.heap<!fir.array<10xi32>>
 ! CHECK: %[[ELEMENTAL:.*]] = hlfir.elemental %{{.*}} unordered : (!fir.shape<1>) -> !hlfir.expr<10xi32>
 ! CHECK: hlfir.assign %[[ELEMENTAL]] to %[[BHOST]]#0 : !hlfir.expr<10xi32>, !fir.ref<!fir.array<10xi32>>
@@ -119,3 +119,43 @@ end
 ! CHECK: %[[T:.*]]:2 = hlfir.declare %7 {cuda_attr = #fir.cuda<device>, uniq_name = "_QFsub3Et"} : (!fir.ref<!fir.type<_QMmod1Tt1{i:i32}>>) -> (!fir.ref<!fir.type<_QMmod1Tt1{i:i32}>>, !fir.ref<!fir.type<_QMmod1Tt1{i:i32}>>)
 ! CHECK: %[[TMP_DECL:.*]]:2 = hlfir.declare %0 {uniq_name = ".tmp"} : (!fir.ref<!fir.type<_QMmod1Tt1{i:i32}>>) -> (!fir.ref<!fir.type<_QMmod1Tt1{i:i32}>>, !fir.ref<!fir.type<_QMmod1Tt1{i:i32}>>)
 ! CHECK: fir.cuda_data_transfer %[[T]]#1 to %[[TMP_DECL]]#0 {transfer_kind = #fir.cuda_transfer<device_host>} : !fir.ref<!fir.type<_QMmod1Tt1{i:i32}>>, !fir.ref<!fir.type<_QMmod1Tt1{i:i32}>>
+
+
+! Check that fir.cuda_data_transfer are not generated within cuf kernel
+subroutine sub4()
+  integer, parameter :: n = 10
+  real, device :: adev(n)
+  real :: ahost(n)
+  real :: b
+  integer :: i
+
+  adev = ahost
+  !$cuf kernel do <<<*,*>>>
+  do i = 1, n
+    adev(i) = adev(i) + b
+  enddo
+end subroutine
+
+! CHECK-LABEL: func.func @_QPsub4()
+! CHECK: fir.cuda_data_transfer
+! CHECK: fir.cuda_kernel<<<*, *>>>
+! CHECK-NOT: fir.cuda_data_transfer
+! CHECK: hlfir.assign
+
+attributes(global) subroutine sub5(a)
+  integer, device :: a
+  integer :: i
+  a = i
+end subroutine
+
+! CHECK-LABEL: func.func @_QPsub5
+! CHECK-NOT: fir.cuda_data_transfer
+
+attributes(host,device) subroutine sub6(a)
+  integer, device :: a
+  integer :: i
+  a = i
+end subroutine
+
+! CHECK-LABEL: func.func @_QPsub6
+! CHECK: fir.cuda_data_transfer
diff --git a/flang/test/Lower/HLFIR/calls-f77.f90 b/flang/test/Lower/HLFIR/calls-f77.f90
index ac5be007eb83..cefe379a45d3 100644
--- a/flang/test/Lower/HLFIR/calls-f77.f90
+++ b/flang/test/Lower/HLFIR/calls-f77.f90
@@ -186,3 +186,16 @@ subroutine alternate_return_call(n1, n2, k)
   ! CHECK: ^[[block2]]: // pred: ^bb0
 7 k =  1; return
 end
+
+! -----------------------------------------------------------------------------
+!     Test calls to user procedures with intrinsic interfaces
+! -----------------------------------------------------------------------------
+
+! CHECK-NAME: func.func @_QPintrinsic_iface()
+subroutine intrinsic_iface()
+  intrinsic acos
+  real :: x
+  procedure(acos) :: proc
+  x = proc(1.0)
+end subroutine
+! CHECK" fir.call @_QPproc(%{{.*}}) {{.*}}: (!fir.ref<f32>) -> f32
diff --git a/flang/test/Lower/HLFIR/internal-procedures.f90 b/flang/test/Lower/HLFIR/internal-procedures.f90
index fff7125897dd..3c4439911809 100644
--- a/flang/test/Lower/HLFIR/internal-procedures.f90
+++ b/flang/test/Lower/HLFIR/internal-procedures.f90
@@ -52,3 +52,30 @@ end subroutine
 ! CHECK:  %[[VAL_4:.*]]:2 = fir.unboxchar %[[VAL_3]] : (!fir.boxchar<1>) -> (!fir.ref<!fir.char<1,?>>, index)
 ! CHECK:  %[[VAL_5:.*]]:2 = hlfir.declare %[[VAL_4]]#0 typeparams %[[VAL_4]]#1 {fortran_attrs = #fir.var_attrs<host_assoc>, uniq_name = "_QFtest_scalar_charEc"} : (!fir.ref<!fir.char<1,?>>, index) -> (!fir.boxchar<1>, !fir.ref<!fir.char<1,?>>)
 ! CHECK:  fir.call @_QPbar(%[[VAL_5]]#0) {{.*}}: (!fir.boxchar<1>) -> ()
+
+subroutine test_proc_pointer(p)
+  real, pointer, external :: p
+  call internal()
+contains
+ subroutine internal()
+  real :: x
+  x = p()
+ end subroutine
+end subroutine
+! CHECK-LABEL:   func.func @_QPtest_proc_pointer(
+! CHECK-SAME:                                    %[[VAL_0:.*]]: !fir.ref<!fir.boxproc<() -> ()>>) {
+! CHECK:           %[[VAL_1:.*]]:2 = hlfir.declare %[[VAL_0]] {fortran_attrs = #fir.var_attrs<pointer>, uniq_name = "_QFtest_proc_pointerEp"} : (!fir.ref<!fir.boxproc<() -> ()>>) -> (!fir.ref<!fir.boxproc<() -> ()>>, !fir.ref<!fir.boxproc<() -> ()>>)
+! CHECK:           %[[VAL_2:.*]] = fir.alloca tuple<!fir.ref<!fir.boxproc<() -> ()>>>
+! CHECK:           %[[VAL_3:.*]] = arith.constant 0 : i32
+! CHECK:           %[[VAL_4:.*]] = fir.coordinate_of %[[VAL_2]], %[[VAL_3]] : (!fir.ref<tuple<!fir.ref<!fir.boxproc<() -> ()>>>>, i32) -> !fir.llvm_ptr<!fir.ref<!fir.boxproc<() -> ()>>>
+! CHECK:           fir.store %[[VAL_1]]#1 to %[[VAL_4]] : !fir.llvm_ptr<!fir.ref<!fir.boxproc<() -> ()>>>
+! CHECK:           fir.call @_QFtest_proc_pointerPinternal(%[[VAL_2]]) {{.*}}: (!fir.ref<tuple<!fir.ref<!fir.boxproc<() -> ()>>>>) -> ()
+! CHECK:           return
+! CHECK:         }
+
+! CHECK-LABEL:   func.func private @_QFtest_proc_pointerPinternal(
+! CHECK-SAME:                                                     %[[VAL_0:.*]]: !fir.ref<tuple<!fir.ref<!fir.boxproc<() -> ()>>>> {fir.host_assoc}) attributes {fir.host_symbol = @_QPtest_proc_pointer, llvm.linkage = #llvm.linkage<internal>} {
+! CHECK:           %[[VAL_1:.*]] = arith.constant 0 : i32
+! CHECK:           %[[VAL_2:.*]] = fir.coordinate_of %[[VAL_0]], %[[VAL_1]] : (!fir.ref<tuple<!fir.ref<!fir.boxproc<() -> ()>>>>, i32) -> !fir.llvm_ptr<!fir.ref<!fir.boxproc<() -> ()>>>
+! CHECK:           %[[VAL_3:.*]] = fir.load %[[VAL_2]] : !fir.llvm_ptr<!fir.ref<!fir.boxproc<() -> ()>>>
+! CHECK:           %[[VAL_4:.*]]:2 = hlfir.declare %[[VAL_3]] {fortran_attrs = #fir.var_attrs<pointer, host_assoc>, uniq_name = "_QFtest_proc_pointerEp"} : (!fir.ref<!fir.boxproc<() -> ()>>) -> (!fir.ref<!fir.boxproc<() -> ()>>, !fir.ref<!fir.boxproc<() -> ()>>)
diff --git a/flang/test/Lower/Intrinsics/shape.f90 b/flang/test/Lower/Intrinsics/shape.f90
new file mode 100644
index 000000000000..60f28a326e99
--- /dev/null
+++ b/flang/test/Lower/Intrinsics/shape.f90
@@ -0,0 +1,74 @@
+! Test SHAPE with function results
+! RUN: bbc -emit-hlfir -o - %s | FileCheck %s
+
+subroutine test()
+  interface
+    function return_array()
+      real, pointer :: return_array(:, :, :)
+    end function
+  end interface
+  print *, shape(return_array())
+end subroutine
+! CHECK-LABEL:   func.func @_QPtest() {
+! CHECK:           %[[VAL_0:.*]] = fir.alloca !fir.array<3xi32>
+! CHECK:           %[[VAL_7:.*]] = fir.call @_QPreturn_array() {{.*}}: () -> !fir.box<!fir.ptr<!fir.array<?x?x?xf32>>>
+! CHECK:           fir.save_result %[[VAL_7]] to %[[VAL_1:.*]] : !fir.box<!fir.ptr<!fir.array<?x?x?xf32>>>, !fir.ref<!fir.box<!fir.ptr<!fir.array<?x?x?xf32>>>>
+! CHECK:           %[[VAL_8:.*]]:2 = hlfir.declare %[[VAL_1]] {uniq_name = ".tmp.func_result"} : (!fir.ref<!fir.box<!fir.ptr<!fir.array<?x?x?xf32>>>>) -> (!fir.ref<!fir.box<!fir.ptr<!fir.array<?x?x?xf32>>>>, !fir.ref<!fir.box<!fir.ptr<!fir.array<?x?x?xf32>>>>)
+! CHECK:           %[[VAL_9:.*]] = fir.load %[[VAL_8]]#1 : !fir.ref<!fir.box<!fir.ptr<!fir.array<?x?x?xf32>>>>
+! CHECK:           %[[VAL_10:.*]] = arith.constant 0 : index
+! CHECK:           %[[VAL_11:.*]]:3 = fir.box_dims %[[VAL_9]], %[[VAL_10]] : (!fir.box<!fir.ptr<!fir.array<?x?x?xf32>>>, index) -> (index, index, index)
+! CHECK:           %[[VAL_12:.*]] = fir.convert %[[VAL_11]]#1 : (index) -> i32
+! CHECK:           %[[VAL_13:.*]] = arith.constant 0 : index
+! CHECK:           %[[VAL_14:.*]] = fir.coordinate_of %[[VAL_0]], %[[VAL_13]] : (!fir.ref<!fir.array<3xi32>>, index) -> !fir.ref<i32>
+! CHECK:           fir.store %[[VAL_12]] to %[[VAL_14]] : !fir.ref<i32>
+! CHECK:           %[[VAL_15:.*]] = arith.constant 1 : index
+! CHECK:           %[[VAL_16:.*]]:3 = fir.box_dims %[[VAL_9]], %[[VAL_15]] : (!fir.box<!fir.ptr<!fir.array<?x?x?xf32>>>, index) -> (index, index, index)
+! CHECK:           %[[VAL_17:.*]] = fir.convert %[[VAL_16]]#1 : (index) -> i32
+! CHECK:           %[[VAL_18:.*]] = arith.constant 1 : index
+! CHECK:           %[[VAL_19:.*]] = fir.coordinate_of %[[VAL_0]], %[[VAL_18]] : (!fir.ref<!fir.array<3xi32>>, index) -> !fir.ref<i32>
+! CHECK:           fir.store %[[VAL_17]] to %[[VAL_19]] : !fir.ref<i32>
+! CHECK:           %[[VAL_20:.*]] = arith.constant 2 : index
+! CHECK:           %[[VAL_21:.*]]:3 = fir.box_dims %[[VAL_9]], %[[VAL_20]] : (!fir.box<!fir.ptr<!fir.array<?x?x?xf32>>>, index) -> (index, index, index)
+! CHECK:           %[[VAL_22:.*]] = fir.convert %[[VAL_21]]#1 : (index) -> i32
+! CHECK:           %[[VAL_23:.*]] = arith.constant 2 : index
+! CHECK:           %[[VAL_24:.*]] = fir.coordinate_of %[[VAL_0]], %[[VAL_23]] : (!fir.ref<!fir.array<3xi32>>, index) -> !fir.ref<i32>
+! CHECK:           fir.store %[[VAL_22]] to %[[VAL_24]] : !fir.ref<i32>
+! CHECK:           %[[VAL_25:.*]] = arith.constant 3 : index
+! CHECK:           %[[VAL_26:.*]] = fir.shape %[[VAL_25]] : (index) -> !fir.shape<1>
+! CHECK:           %[[VAL_27:.*]]:2 = hlfir.declare %[[VAL_0]](%[[VAL_26]]) {uniq_name = ".tmp.intrinsic_result"} : (!fir.ref<!fir.array<3xi32>>, !fir.shape<1>) -> (!fir.ref<!fir.array<3xi32>>, !fir.ref<!fir.array<3xi32>>)
+
+subroutine test_kind()
+  interface
+    function return_array()
+      real, pointer :: return_array(:, :, :)
+    end function
+  end interface
+  print *, shape(return_array(), kind=8)
+end subroutine
+! CHECK-LABEL:   func.func @_QPtest_kind() {
+! CHECK:           %[[VAL_0:.*]] = fir.alloca !fir.array<3xi64>
+! CHECK:           %[[VAL_7:.*]] = fir.call @_QPreturn_array() {{.*}}: () -> !fir.box<!fir.ptr<!fir.array<?x?x?xf32>>>
+! CHECK:           fir.save_result %[[VAL_7]] to %[[VAL_1:.*]] : !fir.box<!fir.ptr<!fir.array<?x?x?xf32>>>, !fir.ref<!fir.box<!fir.ptr<!fir.array<?x?x?xf32>>>>
+! CHECK:           %[[VAL_8:.*]]:2 = hlfir.declare %[[VAL_1]] {uniq_name = ".tmp.func_result"} : (!fir.ref<!fir.box<!fir.ptr<!fir.array<?x?x?xf32>>>>) -> (!fir.ref<!fir.box<!fir.ptr<!fir.array<?x?x?xf32>>>>, !fir.ref<!fir.box<!fir.ptr<!fir.array<?x?x?xf32>>>>)
+! CHECK:           %[[VAL_9:.*]] = fir.load %[[VAL_8]]#1 : !fir.ref<!fir.box<!fir.ptr<!fir.array<?x?x?xf32>>>>
+! CHECK:           %[[VAL_10:.*]] = arith.constant 0 : index
+! CHECK:           %[[VAL_11:.*]]:3 = fir.box_dims %[[VAL_9]], %[[VAL_10]] : (!fir.box<!fir.ptr<!fir.array<?x?x?xf32>>>, index) -> (index, index, index)
+! CHECK:           %[[VAL_12:.*]] = fir.convert %[[VAL_11]]#1 : (index) -> i64
+! CHECK:           %[[VAL_13:.*]] = arith.constant 0 : index
+! CHECK:           %[[VAL_14:.*]] = fir.coordinate_of %[[VAL_0]], %[[VAL_13]] : (!fir.ref<!fir.array<3xi64>>, index) -> !fir.ref<i64>
+! CHECK:           fir.store %[[VAL_12]] to %[[VAL_14]] : !fir.ref<i64>
+! CHECK:           %[[VAL_15:.*]] = arith.constant 1 : index
+! CHECK:           %[[VAL_16:.*]]:3 = fir.box_dims %[[VAL_9]], %[[VAL_15]] : (!fir.box<!fir.ptr<!fir.array<?x?x?xf32>>>, index) -> (index, index, index)
+! CHECK:           %[[VAL_17:.*]] = fir.convert %[[VAL_16]]#1 : (index) -> i64
+! CHECK:           %[[VAL_18:.*]] = arith.constant 1 : index
+! CHECK:           %[[VAL_19:.*]] = fir.coordinate_of %[[VAL_0]], %[[VAL_18]] : (!fir.ref<!fir.array<3xi64>>, index) -> !fir.ref<i64>
+! CHECK:           fir.store %[[VAL_17]] to %[[VAL_19]] : !fir.ref<i64>
+! CHECK:           %[[VAL_20:.*]] = arith.constant 2 : index
+! CHECK:           %[[VAL_21:.*]]:3 = fir.box_dims %[[VAL_9]], %[[VAL_20]] : (!fir.box<!fir.ptr<!fir.array<?x?x?xf32>>>, index) -> (index, index, index)
+! CHECK:           %[[VAL_22:.*]] = fir.convert %[[VAL_21]]#1 : (index) -> i64
+! CHECK:           %[[VAL_23:.*]] = arith.constant 2 : index
+! CHECK:           %[[VAL_24:.*]] = fir.coordinate_of %[[VAL_0]], %[[VAL_23]] : (!fir.ref<!fir.array<3xi64>>, index) -> !fir.ref<i64>
+! CHECK:           fir.store %[[VAL_22]] to %[[VAL_24]] : !fir.ref<i64>
+! CHECK:           %[[VAL_25:.*]] = arith.constant 3 : index
+! CHECK:           %[[VAL_26:.*]] = fir.shape %[[VAL_25]] : (index) -> !fir.shape<1>
+! CHECK:           %[[VAL_27:.*]]:2 = hlfir.declare %[[VAL_0]](%[[VAL_26]]) {uniq_name = ".tmp.intrinsic_result"} : (!fir.ref<!fir.array<3xi64>>, !fir.shape<1>) -> (!fir.ref<!fir.array<3xi64>>, !fir.ref<!fir.array<3xi64>>)
diff --git a/flang/test/Lower/OpenACC/acc-declare.f90 b/flang/test/Lower/OpenACC/acc-declare.f90
index 401b654adeb6..5d3f9e3fe97e 100644
--- a/flang/test/Lower/OpenACC/acc-declare.f90
+++ b/flang/test/Lower/OpenACC/acc-declare.f90
@@ -245,6 +245,11 @@ module acc_declare
 ! CHECK: fir.freemem %{{.*}} : !fir.heap<!fir.array<?xi32>>
 ! CHECK: fir.store %{{.*}} to %{{.*}} {acc.declare_action = #acc.declare_action<postDealloc = @_QMacc_declareFacc_declare_allocateEa_acc_declare_update_desc_post_dealloc>} : !fir.ref<!fir.box<!fir.heap<!fir.array<?xi32>>>>
 
+! CHECK: fir.if
+! CHECK: fir.freemem %{{.*}} : !fir.heap<!fir.array<?xi32>>
+! CHECK: fir.store %{{.*}} to %{{.*}}#1 {acc.declare_action = #acc.declare_action<postDealloc = @_QMacc_declareFacc_declare_allocateEa_acc_declare_update_desc_post_dealloc>} : !fir.ref<!fir.box<!fir.heap<!fir.array<?xi32>>>>
+! CHECK: }
+
   end subroutine
 
 ! CHECK-LABEL: func.func private @_QMacc_declareFacc_declare_allocateEa_acc_declare_update_desc_post_alloc(
diff --git a/flang/test/Lower/OpenMP/FIR/array-bounds.f90 b/flang/test/Lower/OpenMP/FIR/array-bounds.f90
deleted file mode 100644
index c2bb7a94712b..000000000000
--- a/flang/test/Lower/OpenMP/FIR/array-bounds.f90
+++ /dev/null
@@ -1,121 +0,0 @@
-!RUN: %flang_fc1 -emit-fir -flang-deprecated-no-hlfir -fopenmp %s -o - | FileCheck %s --check-prefixes=HOST,ALL
-!RUN: %flang_fc1 -emit-fir -flang-deprecated-no-hlfir -fopenmp -fopenmp-is-target-device %s -o - | FileCheck %s --check-prefixes=DEVICE,ALL
-
-!ALL-LABEL: func.func @_QPread_write_section(
-!ALL:  %[[ITER:.*]] = fir.alloca i32 {bindc_name = "i", uniq_name = "_QFread_write_sectionEi"}
-!ALL:  %[[READ:.*]] = fir.address_of(@_QFread_write_sectionEsp_read) : !fir.ref<!fir.array<10xi32>>
-!ALL:  %[[C10:.*]] = arith.constant 10 : index
-!ALL:  %[[WRITE:.*]] = fir.address_of(@_QFread_write_sectionEsp_write) : !fir.ref<!fir.array<10xi32>>
-!ALL:  %[[C10_0:.*]] = arith.constant 10 : index
-!ALL:  %[[C1:.*]] = arith.constant 1 : index
-!ALL:  %[[C2:.*]] = arith.constant 1 : index
-!ALL:  %[[C3:.*]] = arith.constant 4 : index
-!ALL:  %[[BOUNDS0:.*]] = omp.map.bounds   lower_bound(%[[C2]] : index) upper_bound(%[[C3]] : index) extent(%[[C10]] : index) stride(%[[C1]] : index) start_idx(%[[C1]] : index)
-!ALL:  %[[MAP0:.*]] = omp.map.info var_ptr(%[[READ]] : !fir.ref<!fir.array<10xi32>>, !fir.array<10xi32>)   map_clauses(tofrom) capture(ByRef) bounds(%[[BOUNDS0]]) -> !fir.ref<!fir.array<10xi32>> {name = "sp_read(2:5)"}
-!ALL:  %[[C4:.*]] = arith.constant 1 : index
-!ALL:  %[[C5:.*]] = arith.constant 1 : index
-!ALL:  %[[C6:.*]] = arith.constant 4 : index
-!ALL:  %[[BOUNDS1:.*]] = omp.map.bounds   lower_bound(%[[C5]] : index) upper_bound(%[[C6]] : index) extent(%[[C10_0]] : index) stride(%[[C4]] : index) start_idx(%[[C4]] : index)
-!ALL:  %[[MAP1:.*]] = omp.map.info var_ptr(%[[WRITE]] : !fir.ref<!fir.array<10xi32>>, !fir.array<10xi32>)   map_clauses(tofrom) capture(ByRef) bounds(%[[BOUNDS1]]) -> !fir.ref<!fir.array<10xi32>> {name = "sp_write(2:5)"}
-!ALL:  %[[MAP2:.*]] = omp.map.info var_ptr(%[[ITER]] : !fir.ref<i32>, i32)   map_clauses(implicit, exit_release_or_enter_alloc) capture(ByCopy) -> !fir.ref<i32> {name = "i"}
-!ALL: omp.target map_entries(%[[MAP0]] -> %{{.*}}, %[[MAP1]] -> %{{.*}}, %[[MAP2]] -> %{{.*}} : !fir.ref<!fir.array<10xi32>>, !fir.ref<!fir.array<10xi32>>, !fir.ref<i32>) {
-
-subroutine read_write_section()
-    integer :: sp_read(10) = (/1,2,3,4,5,6,7,8,9,10/)
-    integer :: sp_write(10) = (/0,0,0,0,0,0,0,0,0,0/)
-
-!$omp target map(tofrom:sp_read(2:5)) map(tofrom:sp_write(2:5))
-    do i = 2, 5
-        sp_write(i) = sp_read(i)
-    end do
-!$omp end target
-end subroutine read_write_section
-
-module assumed_array_routines
-contains
-!ALL-LABEL: func.func @_QMassumed_array_routinesPassumed_shape_array(
-!ALL-SAME: %[[ARG0:.*]]: !fir.box<!fir.array<?xi32>> {fir.bindc_name = "arr_read_write"})
-!ALL: %[[INTERMEDIATE_ALLOCA:.*]] = fir.alloca !fir.box<!fir.array<?xi32>>
-!ALL: %[[ALLOCA:.*]] = fir.alloca i32 {bindc_name = "i", uniq_name = "_QMassumed_array_routinesFassumed_shape_arrayEi"}
-!ALL: %[[C0:.*]] = arith.constant 1 : index
-!ALL: %[[C1:.*]] = arith.constant 0 : index
-!ALL: %[[DIMS0:.*]]:3 = fir.box_dims %arg0, %[[C1]] : (!fir.box<!fir.array<?xi32>>, index) -> (index, index, index)
-!ALL: %[[C3:.*]] = arith.constant 1 : index
-!ALL: %[[C4:.*]] = arith.constant 4 : index
-!ALL: %[[C0_1:.*]] = arith.constant 0 : index
-!ALL: %[[DIMS1:.*]]:3 = fir.box_dims %arg0, %[[C0_1]] : (!fir.box<!fir.array<?xi32>>, index) -> (index, index, index)
-!ALL: %[[BOUNDS:.*]] = omp.map.bounds   lower_bound(%[[C3]] : index) upper_bound(%[[C4]] : index) extent(%[[DIMS1]]#1 : index) stride(%[[DIMS0]]#2 : index) start_idx(%[[C0]] : index) {stride_in_bytes = true}
-!ALL: %[[BOXADDRADDR:.*]] = fir.box_offset %0 base_addr : (!fir.ref<!fir.box<!fir.array<?xi32>>>) -> !fir.llvm_ptr<!fir.ref<!fir.array<?xi32>>>
-!ALL: %[[MAP_MEMBER:.*]] = omp.map.info var_ptr(%0 : !fir.ref<!fir.box<!fir.array<?xi32>>>, !fir.array<?xi32>) var_ptr_ptr(%[[BOXADDRADDR]] : !fir.llvm_ptr<!fir.ref<!fir.array<?xi32>>>) map_clauses(tofrom) capture(ByRef) bounds(%[[BOUNDS]]) -> !fir.llvm_ptr<!fir.ref<!fir.array<?xi32>>> {name = ""}
-!ALL: %[[MAP:.*]] = omp.map.info var_ptr(%0 : !fir.ref<!fir.box<!fir.array<?xi32>>>, !fir.box<!fir.array<?xi32>>) map_clauses(tofrom) capture(ByRef) members(%[[MAP_MEMBER]] : !fir.llvm_ptr<!fir.ref<!fir.array<?xi32>>>) -> !fir.ref<!fir.array<?xi32>> {name = "arr_read_write(2:5)"}
-!ALL: %[[MAP2:.*]] = omp.map.info var_ptr(%[[ALLOCA]] : !fir.ref<i32>, i32)   map_clauses(implicit, exit_release_or_enter_alloc) capture(ByCopy) -> !fir.ref<i32> {name = "i"}
-!ALL: omp.target map_entries(%[[MAP_MEMBER]] -> %{{.*}}, %[[MAP]] -> %{{.*}}, %[[MAP2]] -> %{{.*}} : !fir.llvm_ptr<!fir.ref<!fir.array<?xi32>>>, !fir.ref<!fir.array<?xi32>>, !fir.ref<i32>) {
-    subroutine assumed_shape_array(arr_read_write)
-        integer, intent(inout) :: arr_read_write(:)
-
-        !$omp target map(tofrom:arr_read_write(2:5))
-            do i = 2, 5
-                arr_read_write(i) = i
-            end do
-        !$omp end target
-    end subroutine assumed_shape_array
-
-!ALL-LABEL:   func.func @_QMassumed_array_routinesPassumed_size_array(
-!ALL-SAME: %[[ARG0:.*]]: !fir.ref<!fir.array<?xi32>> {fir.bindc_name = "arr_read_write"})
-!ALL: %[[ALLOCA:.*]] = fir.alloca i32 {bindc_name = "i", uniq_name = "_QMassumed_array_routinesFassumed_size_arrayEi"}
-!ALL: %[[C0:.*]] = arith.constant 1 : index
-!ALL: %[[C1:.*]] = arith.constant 1 : index
-!ALL: %[[C2:.*]] = arith.constant 4 : index
-!ALL: %[[DIFF:.*]] = arith.subi %[[C2]], %[[C1]] : index
-!ALL: %[[EXT:.*]] = arith.addi %[[DIFF]], %[[C0]] : index
-!ALL: %[[BOUNDS:.*]]  = omp.map.bounds   lower_bound(%[[C1]] : index) upper_bound(%[[C2]] : index) extent(%[[EXT]] : index) stride(%[[C0]] : index) start_idx(%[[C0]] : index)
-!ALL: %[[MAP:.*]] = omp.map.info var_ptr(%[[ARG0]] : !fir.ref<!fir.array<?xi32>>, !fir.array<?xi32>)   map_clauses(tofrom) capture(ByRef) bounds(%[[BOUNDS]]) -> !fir.ref<!fir.array<?xi32>> {name = "arr_read_write(2:5)"}
-!ALL: %[[MAP2:.*]] = omp.map.info var_ptr(%[[ALLOCA]] : !fir.ref<i32>, i32)   map_clauses(implicit, exit_release_or_enter_alloc) capture(ByCopy) -> !fir.ref<i32> {name = "i"}
-!ALL: omp.target map_entries(%[[MAP]] -> %{{.*}}, %[[MAP2]] -> %{{.*}} : !fir.ref<!fir.array<?xi32>>, !fir.ref<i32>) {
-    subroutine assumed_size_array(arr_read_write)
-        integer, intent(inout) :: arr_read_write(*)
-
-    !$omp target map(tofrom:arr_read_write(2:5))
-        do i = 2, 5
-            arr_read_write(i) = i
-        end do
-    !$omp end target
-    end subroutine assumed_size_array
-end module assumed_array_routines
-
-!DEVICE-NOT:func.func @_QPcall_assumed_shape_and_size_array() {
-
-!HOST-LABEL:func.func @_QPcall_assumed_shape_and_size_array() {
-!HOST:%{{.*}} = arith.constant 20 : index
-!HOST:%[[ALLOCA:.*]] = fir.alloca !fir.array<20xi32> {bindc_name = "arr_read_write", uniq_name = "_QFcall_assumed_shape_and_size_arrayEarr_read_write"}
-!HOST:%{{.*}} = arith.constant 1 : i64
-!HOST:%{{.*}} = fir.convert %{{.*}} : (i64) -> index
-!HOST:%{{.*}} = arith.constant 1 : i64
-!HOST:%{{.*}} = fir.convert %{{.*}} : (i64) -> index
-!HOST:%{{.*}} = arith.constant 10 : i64
-!HOST:%{{.*}} = fir.convert %{{.*}} : (i64) -> index
-!HOST:%[[SHAPE0:.*]] = fir.shape %{{.*}} : (index) -> !fir.shape<1>
-!HOST:%[[SLICE0:.*]] = fir.slice %{{.*}}, %{{.*}}, %{{.*}} : (index, index, index) -> !fir.slice<1>
-!HOST:%[[ARG0EMB:.*]] = fir.embox %[[ALLOCA]](%[[SHAPE0]]) [%[[SLICE0]]] : (!fir.ref<!fir.array<20xi32>>, !fir.shape<1>, !fir.slice<1>) -> !fir.box<!fir.array<10xi32>>
-!HOST:%[[ARG0:.*]] = fir.convert %[[ARG0EMB]] : (!fir.box<!fir.array<10xi32>>) -> !fir.box<!fir.array<?xi32>>
-!HOST:fir.call @_QMassumed_array_routinesPassumed_shape_array(%[[ARG0]]) fastmath<contract> : (!fir.box<!fir.array<?xi32>>) -> ()
-!HOST:%{{.*}} = arith.constant 10 : i64
-!HOST:%{{.*}} = fir.convert %{{.*}} : (i64) -> index
-!HOST:%{{.*}} = arith.constant 1 : i64
-!HOST:%{{.*}} = fir.convert %{{.*}} : (i64) -> index
-!HOST:%{{.*}} = arith.constant 20 : i64
-!HOST:%{{.*}} = fir.convert %{{.*}} : (i64) -> index
-!HOST:%[[SHAPE1:.*]] = fir.shape %{{.*}} : (index) -> !fir.shape<1>
-!HOST:%[[SLICE1:.*]] = fir.slice %{{.*}}, %{{.*}}, %{{.*}} : (index, index, index) -> !fir.slice<1>
-!HOST:%[[ARG1EMB:.*]] = fir.embox %[[ALLOCA]](%[[SHAPE1]]) [%[[SLICE1]]] : (!fir.ref<!fir.array<20xi32>>, !fir.shape<1>, !fir.slice<1>) -> !fir.box<!fir.array<11xi32>>
-!HOST:%[[ADDROF:.*]] = fir.box_addr %[[ARG1EMB]] : (!fir.box<!fir.array<11xi32>>) -> !fir.ref<!fir.array<11xi32>>
-!HOST:%[[ARG1:.*]] = fir.convert %[[ADDROF]] : (!fir.ref<!fir.array<11xi32>>) -> !fir.ref<!fir.array<?xi32>>
-!HOST:fir.call @_QMassumed_array_routinesPassumed_size_array(%[[ARG1]]) fastmath<contract> : (!fir.ref<!fir.array<?xi32>>) -> ()
-!HOST:return
-!HOST:}
-subroutine call_assumed_shape_and_size_array
-    use assumed_array_routines
-    integer :: arr_read_write(20)
-    call assumed_shape_array(arr_read_write(1:10))
-    call assumed_size_array(arr_read_write(10:20))
-end subroutine call_assumed_shape_and_size_array
diff --git a/flang/test/Lower/OpenMP/FIR/atomic-capture.f90 b/flang/test/Lower/OpenMP/FIR/atomic-capture.f90
deleted file mode 100644
index 9b94214b9da8..000000000000
--- a/flang/test/Lower/OpenMP/FIR/atomic-capture.f90
+++ /dev/null
@@ -1,119 +0,0 @@
-! REQUIRES: openmp_runtime
-
-! RUN: %flang_fc1 -emit-fir -flang-deprecated-no-hlfir -fopenmp %s -o - | FileCheck %s
-
-! This test checks the lowering of atomic capture
-
-program OmpAtomicCapture
-    use omp_lib                                                                                                       
-    integer :: x, y
-
-!CHECK: %[[X:.*]] = fir.alloca i32 {bindc_name = "x", uniq_name = "_QFEx"}
-!CHECK: %[[Y:.*]] = fir.alloca i32 {bindc_name = "y", uniq_name = "_QFEy"}
-!CHECK: %[[temp:.*]] = fir.load %[[X]] : !fir.ref<i32>
-!CHECK: omp.atomic.capture   memory_order(release) {
-!CHECK: omp.atomic.read %[[X]] = %[[Y]] : !fir.ref<i32>
-!CHECK: omp.atomic.update %[[Y]] : !fir.ref<i32> {
-!CHECK: ^bb0(%[[ARG:.*]]: i32):
-!CHECK: %[[result:.*]] = arith.addi %[[temp]], %[[ARG]] : i32
-!CHECK: omp.yield(%[[result]] : i32)
-!CHECK: }
-!CHECK: }
-
-    !$omp atomic capture release
-        x = y
-        y = x + y
-    !$omp end atomic
-
-
-!CHECK: %[[temp:.*]] = fir.load %[[X]] : !fir.ref<i32>
-!CHECK: omp.atomic.capture   hint(uncontended) {
-!CHECK: omp.atomic.update %[[Y]] : !fir.ref<i32> {
-!CHECK: ^bb0(%[[ARG:.*]]: i32):
-!CHECK: %[[result:.*]] = arith.muli %[[temp]], %[[ARG]] : i32
-!CHECK: omp.yield(%[[result]] : i32)
-!CHECK: }
-!CHECK: omp.atomic.read %[[X]] = %[[Y]] : !fir.ref<i32>
-!CHECK: }
-
-    !$omp atomic hint(omp_sync_hint_uncontended) capture
-        y = x * y 
-        x = y
-    !$omp end atomic
-
-!CHECK: %[[constant_20:.*]] = arith.constant 20 : i32
-!CHECK: %[[constant_8:.*]] = arith.constant 8 : i32
-!CHECK: %[[temp:.*]] = fir.load %[[X]] : !fir.ref<i32>
-!CHECK: %[[result:.*]] = arith.subi %[[constant_8]], %[[temp]] : i32
-!CHECK: %[[result_noreassoc:.*]] = fir.no_reassoc %[[result]] : i32
-!CHECK: %[[result:.*]] = arith.addi %[[constant_20]], %[[result_noreassoc]] : i32
-!CHECK: omp.atomic.capture   memory_order(acquire) hint(nonspeculative) {
-!CHECK: omp.atomic.read %[[X]] = %[[Y]] : !fir.ref<i32>
-!CHECK: omp.atomic.write %[[Y]] = %[[result]] : !fir.ref<i32>, i32
-!CHECK: }
-
-    !$omp atomic hint(omp_lock_hint_nonspeculative) capture acquire
-        x = y
-        y = 2 * 10 + (8 - x) 
-    !$omp end atomic
-
-
-!CHECK: %[[constant_20:.*]] = arith.constant 20 : i32
-!CHECK: %[[constant_8:.*]] = arith.constant 8 : i32
-!CHECK: %[[temp:.*]] = fir.load %[[X]] : !fir.ref<i32>
-!CHECK: %[[result:.*]] = arith.subi %[[constant_8]], %[[temp]] : i32
-!CHECK: %[[result_noreassoc:.*]] = fir.no_reassoc %[[result]] : i32
-!CHECK: %[[result:.*]] = arith.addi %[[constant_20]], %[[result_noreassoc]] : i32
-!CHECK: omp.atomic.capture {
-!CHECK: omp.atomic.read %[[X]] = %[[Y]] : !fir.ref<i32>
-!CHECK: omp.atomic.write %[[Y]] = %[[result]] : !fir.ref<i32>, i32
-!CHECK: }
-
-    !$omp atomic capture
-        x = y
-        y = 2 * 10 + (8 - x) 
-    !$omp end atomic 
-end program
-
-
-
-subroutine pointers_in_atomic_capture()
-!CHECK: %[[A:.*]] = fir.alloca !fir.box<!fir.ptr<i32>> {bindc_name = "a", uniq_name = "_QFpointers_in_atomic_captureEa"}
-!CHECK: {{.*}} = fir.zero_bits !fir.ptr<i32>
-!CHECK: {{.*}} = fir.embox {{.*}} : (!fir.ptr<i32>) -> !fir.box<!fir.ptr<i32>>
-!CHECK: fir.store {{.*}} to %[[A]] : !fir.ref<!fir.box<!fir.ptr<i32>>>
-!CHECK: %[[B:.*]] = fir.alloca !fir.box<!fir.ptr<i32>> {bindc_name = "b", uniq_name = "_QFpointers_in_atomic_captureEb"}
-!CHECK: {{.*}} = fir.zero_bits !fir.ptr<i32>
-!CHECK: {{.*}} = fir.embox {{.*}} : (!fir.ptr<i32>) -> !fir.box<!fir.ptr<i32>>
-!CHECK: fir.store {{.*}} to %[[B]] : !fir.ref<!fir.box<!fir.ptr<i32>>>
-!CHECK: %[[C:.*]] = fir.alloca i32 {bindc_name = "c", fir.target, uniq_name = "_QFpointers_in_atomic_captureEc"}
-!CHECK: %[[D:.*]] = fir.alloca i32 {bindc_name = "d", fir.target, uniq_name = "_QFpointers_in_atomic_captureEd"}
-!CHECK: {{.*}} = fir.embox {{.*}} : (!fir.ref<i32>) -> !fir.box<!fir.ptr<i32>>
-!CHECK: fir.store {{.*}} to %[[A]] : !fir.ref<!fir.box<!fir.ptr<i32>>>
-!CHECK: {{.*}} = fir.embox {{.*}} : (!fir.ref<i32>) -> !fir.box<!fir.ptr<i32>>
-!CHECK: fir.store {{.*}} to %[[B]] : !fir.ref<!fir.box<!fir.ptr<i32>>>
-!CHECK: %[[loaded_A:.*]] = fir.load %[[A]] : !fir.ref<!fir.box<!fir.ptr<i32>>>
-!CHECK: %[[loaded_A_addr:.*]] = fir.box_addr %[[loaded_A]] : (!fir.box<!fir.ptr<i32>>) -> !fir.ptr<i32>
-!CHECK: %[[loaded_B:.*]] = fir.load %[[B]] : !fir.ref<!fir.box<!fir.ptr<i32>>>
-!CHECK: %[[loaded_B_addr:.*]] = fir.box_addr %[[loaded_B]] : (!fir.box<!fir.ptr<i32>>) -> !fir.ptr<i32>
-!CHECK: %[[PRIVATE_LOADED_B:.*]] = fir.load %[[B]] : !fir.ref<!fir.box<!fir.ptr<i32>>>
-!CHECK: %[[PRIVATE_LOADED_B_addr:.*]] = fir.box_addr %[[PRIVATE_LOADED_B]] : (!fir.box<!fir.ptr<i32>>) -> !fir.ptr<i32>
-!CHECK: %[[loaded_value:.*]] = fir.load %[[PRIVATE_LOADED_B_addr]] : !fir.ptr<i32>
-!CHECK: omp.atomic.capture   {
-!CHECK: omp.atomic.update %[[loaded_A_addr]] : !fir.ptr<i32> {
-!CHECK: ^bb0(%[[ARG:.*]]: i32):
-!CHECK: %[[result:.*]] = arith.addi %[[ARG]], %[[loaded_value]] : i32
-!CHECK: omp.yield(%[[result]] : i32)
-!CHECK: }
-!CHECK: omp.atomic.read %[[loaded_B_addr]] = %[[loaded_A_addr]] : !fir.ptr<i32>, i32
-!CHECK: }
-    integer, pointer :: a, b
-    integer, target :: c, d
-    a=>c
-    b=>d
-
-    !$omp atomic capture
-        a = a + b
-        b = a
-    !$omp end atomic
-end subroutine
diff --git a/flang/test/Lower/OpenMP/FIR/atomic-read.f90 b/flang/test/Lower/OpenMP/FIR/atomic-read.f90
deleted file mode 100644
index 7698c3d7490f..000000000000
--- a/flang/test/Lower/OpenMP/FIR/atomic-read.f90
+++ /dev/null
@@ -1,80 +0,0 @@
-! REQUIRES: openmp_runtime
-
-! RUN: bbc --use-desc-for-alloc=false -fopenmp -emit-fir -hlfir=false %s -o - | FileCheck %s
-
-! This test checks the lowering of atomic read
-
-!CHECK: func @_QQmain() attributes {fir.bindc_name = "ompatomic"} {
-!CHECK: %[[VAR_A:.*]] = fir.alloca !fir.char<1> {bindc_name = "a", uniq_name = "_QFEa"}
-!CHECK: %[[VAR_B:.*]] = fir.alloca !fir.char<1> {bindc_name = "b", uniq_name = "_QFEb"}
-!CHECK: %[[VAR_C:.*]] = fir.alloca !fir.logical<4> {bindc_name = "c", uniq_name = "_QFEc"}
-!CHECK: %[[VAR_D:.*]] = fir.alloca !fir.logical<4> {bindc_name = "d", uniq_name = "_QFEd"}
-!CHECK: %[[VAR_E:.*]] = fir.alloca !fir.char<1,8> {bindc_name = "e", uniq_name = "_QFEe"}
-!CHECK: %[[VAR_F:.*]] = fir.alloca !fir.char<1,8> {bindc_name = "f", uniq_name = "_QFEf"}
-!CHECK: %[[VAR_G:.*]] = fir.alloca f32 {bindc_name = "g", uniq_name = "_QFEg"}
-!CHECK: %[[VAR_H:.*]] = fir.alloca f32 {bindc_name = "h", uniq_name = "_QFEh"}
-!CHECK: %[[VAR_X:.*]] = fir.alloca i32 {bindc_name = "x", uniq_name = "_QFEx"}
-!CHECK: %[[VAR_Y:.*]] = fir.alloca i32 {bindc_name = "y", uniq_name = "_QFEy"}
-!CHECK: omp.atomic.read %[[VAR_X]] = %[[VAR_Y]] memory_order(acquire)  hint(uncontended) : !fir.ref<i32>, i32
-!CHECK: omp.atomic.read %[[VAR_A]] = %[[VAR_B]] memory_order(relaxed) : !fir.ref<!fir.char<1>>, !fir.char<1>
-!CHECK: omp.atomic.read %[[VAR_C]] = %[[VAR_D]] memory_order(seq_cst)  hint(contended) : !fir.ref<!fir.logical<4>>, !fir.logical<4>
-!CHECK: omp.atomic.read %[[VAR_E]] = %[[VAR_F]] hint(speculative) : !fir.ref<!fir.char<1,8>>, !fir.char<1,8>
-!CHECK: omp.atomic.read %[[VAR_G]] = %[[VAR_H]] hint(nonspeculative) : !fir.ref<f32>, f32
-!CHECK: omp.atomic.read %[[VAR_G]] = %[[VAR_H]] : !fir.ref<f32>, f32
-!CHECK: return
-!CHECK: }
-
-program OmpAtomic
-
-    use omp_lib
-    integer :: x, y
-    character :: a, b
-    logical :: c, d
-    character(8) :: e, f
-    real g, h
-    !$omp atomic acquire read hint(omp_sync_hint_uncontended)
-       x = y
-    !$omp atomic relaxed read hint(omp_sync_hint_none)
-       a = b
-    !$omp atomic read seq_cst hint(omp_sync_hint_contended)
-       c = d
-    !$omp atomic read hint(omp_sync_hint_speculative)
-       e = f
-    !$omp atomic read hint(omp_sync_hint_nonspeculative)
-       g = h
-    !$omp atomic read
-       g = h
-end program OmpAtomic
-
-! Test lowering atomic read for pointer variables.
-! Please notice to use %[[VAL_4]] and %[[VAL_1]] for operands of atomic
-! operation, instead of %[[VAL_3]] and %[[VAL_0]].
-
-!CHECK-LABEL: func.func @_QPatomic_read_pointer() {
-!CHECK:         %[[VAL_0:.*]] = fir.alloca !fir.box<!fir.ptr<i32>> {bindc_name = "x", uniq_name = "_QFatomic_read_pointerEx"}
-!CHECK:         %[[VAL_1:.*]] = fir.alloca !fir.ptr<i32> {uniq_name = "_QFatomic_read_pointerEx.addr"}
-!CHECK:         %[[VAL_2:.*]] = fir.zero_bits !fir.ptr<i32>
-!CHECK:         fir.store %[[VAL_2]] to %[[VAL_1]] : !fir.ref<!fir.ptr<i32>>
-!CHECK:         %[[VAL_3:.*]] = fir.alloca !fir.box<!fir.ptr<i32>> {bindc_name = "y", uniq_name = "_QFatomic_read_pointerEy"}
-!CHECK:         %[[VAL_4:.*]] = fir.alloca !fir.ptr<i32> {uniq_name = "_QFatomic_read_pointerEy.addr"}
-!CHECK:         %[[VAL_5:.*]] = fir.zero_bits !fir.ptr<i32>
-!CHECK:         fir.store %[[VAL_5]] to %[[VAL_4]] : !fir.ref<!fir.ptr<i32>>
-!CHECK:         %[[VAL_6:.*]] = fir.load %[[VAL_1]] : !fir.ref<!fir.ptr<i32>>
-!CHECK:         %[[VAL_7:.*]] = fir.load %[[VAL_4]] : !fir.ref<!fir.ptr<i32>>
-!CHECK:         omp.atomic.read %[[VAL_7]] = %[[VAL_6]]   : !fir.ptr<i32>, i32
-!CHECK:         %[[VAL_8:.*]] = fir.load %[[VAL_4]] : !fir.ref<!fir.ptr<i32>>
-!CHECK:         %[[VAL_9:.*]] = fir.load %[[VAL_8]] : !fir.ptr<i32>
-!CHECK:         %[[VAL_10:.*]] = fir.load %[[VAL_1]] : !fir.ref<!fir.ptr<i32>>
-!CHECK:         fir.store %[[VAL_9]] to %[[VAL_10]] : !fir.ptr<i32>
-!CHECK:         return
-!CHECK:       }
-
-subroutine atomic_read_pointer()
-  integer, pointer :: x, y
-
-  !$omp atomic read
-    y = x
-
-  x = y
-end
-
diff --git a/flang/test/Lower/OpenMP/FIR/atomic-update.f90 b/flang/test/Lower/OpenMP/FIR/atomic-update.f90
deleted file mode 100644
index ae201807c337..000000000000
--- a/flang/test/Lower/OpenMP/FIR/atomic-update.f90
+++ /dev/null
@@ -1,141 +0,0 @@
-! REQUIRES: openmp_runtime
-
-! This test checks lowering of atomic and atomic update constructs
-! RUN: bbc --use-desc-for-alloc=false -fopenmp -emit-fir -hlfir=false %s -o - | FileCheck %s
-! RUN: %flang_fc1 -mllvm --use-desc-for-alloc=false -emit-fir -flang-deprecated-no-hlfir -fopenmp %s -o - | FileCheck %s
-
-program OmpAtomicUpdate
-    use omp_lib
-    integer :: x, y, z
-    integer, pointer :: a, b
-    integer, target :: c, d
-    integer(1) :: i1
-
-    a=>c
-    b=>d
-
-!CHECK: func.func @_QQmain() attributes {fir.bindc_name = "ompatomicupdate"} {
-!CHECK: %[[A:.*]] = fir.alloca !fir.box<!fir.ptr<i32>> {bindc_name = "a", uniq_name = "_QFEa"}
-!CHECK: %[[A_ADDR:.*]] = fir.alloca !fir.ptr<i32> {uniq_name = "_QFEa.addr"}
-!CHECK: %{{.*}} = fir.zero_bits !fir.ptr<i32>
-!CHECK: fir.store %{{.*}} to %[[A_ADDR]] : !fir.ref<!fir.ptr<i32>>
-!CHECK: %[[B:.*]] = fir.alloca !fir.box<!fir.ptr<i32>> {bindc_name = "b", uniq_name = "_QFEb"}
-!CHECK: %[[B_ADDR:.*]] = fir.alloca !fir.ptr<i32> {uniq_name = "_QFEb.addr"}
-!CHECK: %{{.*}} = fir.zero_bits !fir.ptr<i32>
-!CHECK: fir.store %{{.*}} to %[[B_ADDR]] : !fir.ref<!fir.ptr<i32>>
-!CHECK: %[[C_ADDR:.*]] = fir.address_of(@_QFEc) : !fir.ref<i32>
-!CHECK: %[[D_ADDR:.*]] = fir.address_of(@_QFEd) : !fir.ref<i32>
-!CHECK: %[[I1:.*]] = fir.alloca i8 {bindc_name = "i1", uniq_name = "_QFEi1"}
-!CHECK: %[[X:.*]] = fir.alloca i32 {bindc_name = "x", uniq_name = "_QFEx"}
-!CHECK: %[[Y:.*]] = fir.alloca i32 {bindc_name = "y", uniq_name = "_QFEy"}
-!CHECK: %[[Z:.*]] = fir.alloca i32 {bindc_name = "z", uniq_name = "_QFEz"}
-!CHECK: %{{.*}} = fir.convert %[[C_ADDR]] : (!fir.ref<i32>) -> !fir.ptr<i32>
-!CHECK: fir.store %{{.*}} to %[[A_ADDR]] : !fir.ref<!fir.ptr<i32>>
-!CHECK: %{{.*}} = fir.convert %[[D_ADDR]] : (!fir.ref<i32>) -> !fir.ptr<i32>
-!CHECK: fir.store {{.*}} to %[[B_ADDR]] : !fir.ref<!fir.ptr<i32>>
-!CHECK: %[[LOADED_A:.*]] = fir.load %[[A_ADDR]] : !fir.ref<!fir.ptr<i32>>
-!CHECK: %[[LOADED_B:.*]] = fir.load %[[B_ADDR]] : !fir.ref<!fir.ptr<i32>>
-!CHECK: %{{.*}} = fir.load %[[LOADED_B]] : !fir.ptr<i32>
-!CHECK:  omp.atomic.update   %[[LOADED_A]] : !fir.ptr<i32> {
-!CHECK:  ^bb0(%[[ARG:.*]]: i32):
-!CHECK:    %[[RESULT:.*]] = arith.addi %[[ARG]], %{{.*}} : i32
-!CHECK:    omp.yield(%[[RESULT]] : i32)
-!CHECK: }
-    !$omp atomic update
-        a = a + b 
-
-!CHECK: {{.*}} = arith.constant 1 : i32
-!CHECK: omp.atomic.update   %[[Y]] : !fir.ref<i32> {
-!CHECK:  ^bb0(%[[ARG:.*]]: i32):
-!CHECK:    %[[RESULT:.*]] = arith.addi %[[ARG]], {{.*}} : i32
-!CHECK:    omp.yield(%[[RESULT]] : i32)
-!CHECK:  }
-!CHECK:  %[[LOADED_X:.*]] = fir.load %[[X]] : !fir.ref<i32>
-!CHECK:  omp.atomic.update   %[[Z]] : !fir.ref<i32> {
-!CHECK:  ^bb0(%[[ARG:.*]]: i32):
-!CHECK:    %[[RESULT:.*]] = arith.muli %[[LOADED_X]], %[[ARG]] : i32
-!CHECK:    omp.yield(%[[RESULT]] : i32)
-!CHECK:  }
-    !$omp atomic 
-        y = y + 1
-    !$omp atomic update
-        z = x * z 
-
-!CHECK:  %{{.*}} = arith.constant 1 : i32
-!CHECK:  omp.atomic.update   memory_order(relaxed) hint(uncontended) %[[X]] : !fir.ref<i32> {
-!CHECK:  ^bb0(%[[ARG:.*]]: i32):
-!CHECK:    %[[RESULT:.*]] = arith.subi %[[ARG]], {{.*}} : i32
-!CHECK:    omp.yield(%[[RESULT]] : i32)
-!CHECK:  }
-!CHECK:  %[[LOADED_X:.*]] = fir.load %[[X]] : !fir.ref<i32>
-!CHECK:  %[[LOADED_Z:.*]] = fir.load %[[Z]] : !fir.ref<i32>
-!CHECK:  omp.atomic.update   memory_order(relaxed) %[[Y]] : !fir.ref<i32> {
-!CHECK:  ^bb0(%[[ARG:.*]]: i32):
-!CHECK:    %{{.*}} = arith.cmpi sgt, %[[ARG]], %[[LOADED_X]] : i32
-!CHECK:    %{{.*}} = arith.select %{{.*}}, %[[ARG]], %[[LOADED_X]] : i32
-!CHECK:    %{{.*}} = arith.cmpi sgt, %{{.*}}, %[[LOADED_Z]] : i32
-!CHECK:    %[[RESULT:.*]] = arith.select %{{.*}}, %{{.*}}, %[[LOADED_Z]] : i32
-!CHECK:    omp.yield(%[[RESULT]] : i32)
-!CHECK:  }
-!CHECK:  %[[LOADED_X:.*]] = fir.load %[[X]] : !fir.ref<i32>
-!CHECK:  omp.atomic.update   memory_order(relaxed) hint(contended) %[[Z]] : !fir.ref<i32> {
-!CHECK:  ^bb0(%[[ARG:.*]]: i32):
-!CHECK:    %[[RESULT:.*]] = arith.addi %[[ARG]], %[[LOADED_X]] : i32
-!CHECK:    omp.yield(%[[RESULT]] : i32)
-!CHECK:  }
-    !$omp atomic relaxed update hint(omp_sync_hint_uncontended)
-        x = x - 1
-    !$omp atomic update relaxed 
-        y = max(y, x, z)
-    !$omp atomic relaxed hint(omp_sync_hint_contended)
-        z = z + x
-
-!CHECK:  %{{.*}} = arith.constant 10 : i32
-!CHECK:  omp.atomic.update   memory_order(release) hint(contended) %[[Z]] : !fir.ref<i32> {
-!CHECK:  ^bb0(%[[ARG:.*]]: i32):
-!CHECK:   %[[RESULT:.*]] = arith.muli {{.*}}, %[[ARG]] : i32
-!CHECK:    omp.yield(%[[RESULT]] : i32)
-!CHECK:  }
-!CHECK:  %[[LOADED_Z:.*]] = fir.load %[[Z]] : !fir.ref<i32>
-!CHECK:  omp.atomic.update   memory_order(release) hint(speculative) %[[X]] : !fir.ref<i32> {
-!CHECK:  ^bb0(%[[ARG:.*]]: i32):
-!CHECK:    %[[RESULT:.*]] = arith.divsi %[[ARG]], %[[LOADED_Z]] : i32
-!CHECK:    omp.yield(%[[RESULT]] : i32)
-!CHECK:  }
-
-    !$omp atomic release update hint(omp_lock_hint_contended)
-        z = z * 10
-    !$omp atomic hint(omp_lock_hint_speculative) update release
-        x = x / z
-
-!CHECK:  %{{.*}} = arith.constant 10 : i32
-!CHECK:  omp.atomic.update   memory_order(seq_cst) hint(nonspeculative) %[[Y]] : !fir.ref<i32> {
-!CHECK:  ^bb0(%[[ARG:.*]]: i32):
-!CHECK:    %[[RESULT:.*]] = arith.addi %{{.*}}, %[[ARG]] : i32
-!CHECK:   omp.yield(%[[RESULT]] : i32)
-!CHECK:  }
-!CHECK:  %[[LOADED_Y:.*]] = fir.load %[[Y]] : !fir.ref<i32>
-!CHECK:  omp.atomic.update   memory_order(seq_cst) %[[Z]] : !fir.ref<i32> {
-!CHECK:  ^bb0(%[[ARG:.*]]: i32):
-!CHECK:    %[[RESULT:.*]] = arith.addi %[[LOADED_Y]], %[[ARG]] : i32
-!CHECK:    omp.yield(%[[RESULT]] : i32)
-!CHECK:  }
-    !$omp atomic hint(omp_sync_hint_nonspeculative) seq_cst
-        y = 10 + y
-    !$omp atomic seq_cst update
-        z = y + z
-
-!CHECK:  %[[C1_VAL:.*]] = arith.constant 1 : i32
-!CHECK:  omp.atomic.update   %[[I1]] : !fir.ref<i8> {
-!CHECK:  ^bb0(%[[VAL:.*]]: i8):
-!CHECK:    %[[CVT_VAL:.*]] = fir.convert %[[VAL]] : (i8) -> i32
-!CHECK:    %[[ADD_VAL:.*]] = arith.addi %[[CVT_VAL]], %[[C1_VAL]] : i32
-!CHECK:    %[[UPDATED_VAL:.*]] = fir.convert %[[ADD_VAL]] : (i32) -> i8
-!CHECK:    omp.yield(%[[UPDATED_VAL]] : i8)
-!CHECK:  }
-    !$omp atomic
-      i1 = i1 + 1
-    !$omp end atomic
-!CHECK:  return
-!CHECK: }
-end program OmpAtomicUpdate
diff --git a/flang/test/Lower/OpenMP/FIR/atomic-write.f90 b/flang/test/Lower/OpenMP/FIR/atomic-write.f90
deleted file mode 100644
index 142481b7a1d2..000000000000
--- a/flang/test/Lower/OpenMP/FIR/atomic-write.f90
+++ /dev/null
@@ -1,77 +0,0 @@
-! REQUIRES: openmp_runtime
-
-! RUN: bbc --use-desc-for-alloc=false -fopenmp -emit-fir -hlfir=false %s -o - | FileCheck %s
-
-! This test checks the lowering of atomic write
-
-!CHECK: func @_QQmain() attributes {fir.bindc_name = "ompatomicwrite"} {
-!CHECK: %[[VAR_X:.*]] = fir.alloca i32 {bindc_name = "x", uniq_name = "_QFEx"}
-!CHECK: %[[VAR_Y:.*]] = fir.alloca i32 {bindc_name = "y", uniq_name = "_QFEy"}
-!CHECK: %[[VAR_Z:.*]] = fir.alloca i32 {bindc_name = "z", uniq_name = "_QFEz"}
-!CHECK: %[[CONST_44:.*]] = arith.constant 44 : i32
-!CHECK: omp.atomic.write %[[VAR_X]] = %[[CONST_44]] hint(uncontended) memory_order(seq_cst) : !fir.ref<i32>, i32
-!CHECK: %[[CONST_7:.*]] = arith.constant 7 : i32
-!CHECK: {{.*}} = fir.load %[[VAR_Y]] : !fir.ref<i32>
-!CHECK: %[[VAR_7y:.*]] = arith.muli %[[CONST_7]], {{.*}} : i32
-!CHECK: omp.atomic.write %[[VAR_X]] = %[[VAR_7y]] memory_order(relaxed) : !fir.ref<i32>, i32
-!CHECK: %[[CONST_10:.*]] = arith.constant 10 : i32
-!CHECK: {{.*}} = fir.load %[[VAR_X]] : !fir.ref<i32>
-!CHECK: {{.*}} = arith.muli %[[CONST_10]], {{.*}} : i32
-!CHECK: {{.*}} = fir.load %[[VAR_Z]] : !fir.ref<i32>
-!CHECK: %[[CONST_2:.*]] = arith.constant 2 : i32
-!CHECK: {{.*}} = arith.divsi {{.*}}, %[[CONST_2]] : i32
-!CHECK: {{.*}} = arith.addi {{.*}}, {{.*}} : i32
-!CHECK: omp.atomic.write %[[VAR_Y]] = {{.*}} hint(speculative) memory_order(release) : !fir.ref<i32>, i32
-!CHECK: return
-!CHECK: }
-
-program OmpAtomicWrite
-    use omp_lib
-    integer :: x, y, z
-    !$omp atomic seq_cst write hint(omp_sync_hint_uncontended)
-        x = 8*4 + 12
-
-    !$omp atomic write relaxed
-        x = 7 * y
-
-    !$omp atomic write release hint(omp_sync_hint_speculative)
-        y = 10*x + z/2
-end program OmpAtomicWrite
-
-! Test lowering atomic read for pointer variables.
-! Please notice to use %[[VAL_1]] for operands of atomic operation, instead
-! of %[[VAL_0]].
-
-!CHECK-LABEL: func.func @_QPatomic_write_pointer() {
-!CHECK:         %[[VAL_0:.*]] = fir.alloca !fir.box<!fir.ptr<i32>> {bindc_name = "x", uniq_name = "_QFatomic_write_pointerEx"}
-!CHECK:         %[[VAL_1:.*]] = fir.alloca !fir.ptr<i32> {uniq_name = "_QFatomic_write_pointerEx.addr"}
-!CHECK:         %[[VAL_2:.*]] = fir.zero_bits !fir.ptr<i32>
-!CHECK:         fir.store %[[VAL_2]] to %[[VAL_1]] : !fir.ref<!fir.ptr<i32>>
-!CHECK:         %[[VAL_3:.*]] = arith.constant 1 : i32
-!CHECK:         %[[VAL_4:.*]] = fir.load %[[VAL_1]] : !fir.ref<!fir.ptr<i32>>
-!CHECK:         omp.atomic.write %[[VAL_4]] = %[[VAL_3]]   : !fir.ptr<i32>, i32
-!CHECK:         %[[VAL_5:.*]] = arith.constant 2 : i32
-!CHECK:         %[[VAL_6:.*]] = fir.load %[[VAL_1]] : !fir.ref<!fir.ptr<i32>>
-!CHECK:         fir.store %[[VAL_5]] to %[[VAL_6]] : !fir.ptr<i32>
-!CHECK:         return
-!CHECK:       }
-
-subroutine atomic_write_pointer()
-  integer, pointer :: x
-
-  !$omp atomic write
-    x = 1
-
-  x = 2
-end
-
-!CHECK-LABEL: func.func @_QPatomic_write_typed_assign
-!CHECK: %[[VAR:.*]] = fir.alloca f32 {bindc_name = "r2", uniq_name = "{{.*}}r2"}
-!CHECK: %[[CST:.*]] = arith.constant 0.000000e+00 : f32
-!CHECK: omp.atomic.write %[[VAR]] = %[[CST]]   : !fir.ref<f32>, f32
-
-subroutine atomic_write_typed_assign
-  real :: r2
-  !$omp atomic write
-  r2 = 0
-end subroutine
diff --git a/flang/test/Lower/OpenMP/FIR/copyin.f90 b/flang/test/Lower/OpenMP/FIR/copyin.f90
deleted file mode 100644
index 20023a81977a..000000000000
--- a/flang/test/Lower/OpenMP/FIR/copyin.f90
+++ /dev/null
@@ -1,352 +0,0 @@
-! This test checks lowering of `COPYIN` clause.
-! RUN: bbc -fopenmp -emit-fir -hlfir=false %s -o - | FileCheck %s
-! RUN: %flang_fc1 -emit-fir -flang-deprecated-no-hlfir -fopenmp %s -o - | FileCheck %s
-
-! CHECK-LABEL: func.func @_QPcopyin_scalar_array() {
-! CHECK:         %[[VAL_0:.*]] = fir.address_of(@_QFcopyin_scalar_arrayEx1) : !fir.ref<i32>
-! CHECK:         %[[VAL_1:.*]] = omp.threadprivate %[[VAL_0]] : !fir.ref<i32> -> !fir.ref<i32>
-! CHECK:         %[[VAL_2:.*]] = fir.address_of(@_QFcopyin_scalar_arrayEx2) : !fir.ref<!fir.array<10xi64>>
-! CHECK:         %[[VAL_3:.*]] = arith.constant 10 : index
-! CHECK:         %[[VAL_4:.*]] = omp.threadprivate %[[VAL_2]] : !fir.ref<!fir.array<10xi64>> -> !fir.ref<!fir.array<10xi64>>
-! CHECK:         omp.parallel   {
-! CHECK:           %[[VAL_5:.*]] = omp.threadprivate %[[VAL_0]] : !fir.ref<i32> -> !fir.ref<i32>
-! CHECK:           %[[VAL_6:.*]] = fir.load %[[VAL_1]] : !fir.ref<i32>
-! CHECK:           fir.store %[[VAL_6]] to %[[VAL_5]] : !fir.ref<i32>
-! CHECK:           %[[VAL_7:.*]] = omp.threadprivate %[[VAL_2]] : !fir.ref<!fir.array<10xi64>> -> !fir.ref<!fir.array<10xi64>>
-! CHECK:           %[[VAL_8:.*]] = fir.shape %[[VAL_3]] : (index) -> !fir.shape<1>
-! CHECK:           %[[VAL_9:.*]] = fir.array_load %[[VAL_7]](%[[VAL_8]]) : (!fir.ref<!fir.array<10xi64>>, !fir.shape<1>) -> !fir.array<10xi64>
-! CHECK:           %[[VAL_10:.*]] = fir.shape %[[VAL_3]] : (index) -> !fir.shape<1>
-! CHECK:           %[[VAL_11:.*]] = fir.array_load %[[VAL_4]](%[[VAL_10]]) : (!fir.ref<!fir.array<10xi64>>, !fir.shape<1>) -> !fir.array<10xi64>
-! CHECK:           %[[VAL_12:.*]] = arith.constant 1 : index
-! CHECK:           %[[VAL_13:.*]] = arith.constant 0 : index
-! CHECK:           %[[VAL_14:.*]] = arith.subi %[[VAL_3]], %[[VAL_12]] : index
-! CHECK:           %[[VAL_15:.*]] = fir.do_loop %[[VAL_16:.*]] = %[[VAL_13]] to %[[VAL_14]] step %[[VAL_12]] unordered iter_args(%[[VAL_17:.*]] = %[[VAL_9]]) -> (!fir.array<10xi64>) {
-! CHECK:             %[[VAL_18:.*]] = fir.array_fetch %[[VAL_11]], %[[VAL_16]] : (!fir.array<10xi64>, index) -> i64
-! CHECK:             %[[VAL_19:.*]] = fir.array_update %[[VAL_17]], %[[VAL_18]], %[[VAL_16]] : (!fir.array<10xi64>, i64, index) -> !fir.array<10xi64>
-! CHECK:             fir.result %[[VAL_19]] : !fir.array<10xi64>
-! CHECK:           }
-! CHECK:           fir.array_merge_store %[[VAL_9]], %[[VAL_20:.*]] to %[[VAL_7]] : !fir.array<10xi64>, !fir.array<10xi64>, !fir.ref<!fir.array<10xi64>>
-! CHECK:           omp.barrier
-! CHECK:           fir.call @_QPsub1(%[[VAL_5]], %[[VAL_7]]) {{.*}}: (!fir.ref<i32>, !fir.ref<!fir.array<10xi64>>) -> ()
-! CHECK:           omp.terminator
-! CHECK:         }
-! CHECK:         return
-! CHECK:       }
-
-subroutine copyin_scalar_array()
-  integer(kind=4), save :: x1
-  integer(kind=8), save :: x2(10)
-  !$omp threadprivate(x1, x2)
-
-  !$omp parallel copyin(x1) copyin(x2)
-    call sub1(x1, x2)
-  !$omp end parallel
-
-end
-
-! CHECK-LABEL: func.func @_QPcopyin_char_chararray() {
-! CHECK:         %[[VAL_0:.*]] = fir.address_of(@_QFcopyin_char_chararrayEx3) : !fir.ref<!fir.char<1,5>>
-! CHECK:         %[[VAL_1:.*]] = arith.constant 5 : index
-! CHECK:         %[[VAL_2:.*]] = omp.threadprivate %[[VAL_0]] : !fir.ref<!fir.char<1,5>> -> !fir.ref<!fir.char<1,5>>
-! CHECK:         %[[VAL_3:.*]] = fir.address_of(@_QFcopyin_char_chararrayEx4) : !fir.ref<!fir.array<10x!fir.char<1,5>>>
-! CHECK:         %[[VAL_4:.*]] = arith.constant 5 : index
-! CHECK:         %[[VAL_5:.*]] = arith.constant 10 : index
-! CHECK:         %[[VAL_6:.*]] = omp.threadprivate %[[VAL_3]] : !fir.ref<!fir.array<10x!fir.char<1,5>>> -> !fir.ref<!fir.array<10x!fir.char<1,5>>>
-! CHECK:         omp.parallel   {
-! CHECK:           %[[VAL_7:.*]] = omp.threadprivate %[[VAL_0]] : !fir.ref<!fir.char<1,5>> -> !fir.ref<!fir.char<1,5>>
-! CHECK:           %[[VAL_8:.*]] = arith.constant 1 : i64
-! CHECK:           %[[VAL_9:.*]] = fir.convert %[[VAL_1]] : (index) -> i64
-! CHECK:           %[[VAL_10:.*]] = arith.muli %[[VAL_8]], %[[VAL_9]] : i64
-! CHECK:           %[[VAL_11:.*]] = arith.constant false
-! CHECK:           %[[VAL_12:.*]] = fir.convert %[[VAL_7]] : (!fir.ref<!fir.char<1,5>>) -> !fir.ref<i8>
-! CHECK:           %[[VAL_13:.*]] = fir.convert %[[VAL_2]] : (!fir.ref<!fir.char<1,5>>) -> !fir.ref<i8>
-! CHECK:           fir.call @llvm.memmove.p0.p0.i64(%[[VAL_12]], %[[VAL_13]], %[[VAL_10]], %[[VAL_11]]) {{.*}}: (!fir.ref<i8>, !fir.ref<i8>, i64, i1) -> ()
-! CHECK:           %[[VAL_14:.*]] = omp.threadprivate %[[VAL_3]] : !fir.ref<!fir.array<10x!fir.char<1,5>>> -> !fir.ref<!fir.array<10x!fir.char<1,5>>>
-! CHECK:           %[[VAL_15:.*]] = fir.shape %[[VAL_5]] : (index) -> !fir.shape<1>
-! CHECK:           %[[VAL_16:.*]] = fir.array_load %[[VAL_14]](%[[VAL_15]]) : (!fir.ref<!fir.array<10x!fir.char<1,5>>>, !fir.shape<1>) -> !fir.array<10x!fir.char<1,5>>
-! CHECK:           %[[VAL_17:.*]] = fir.shape %[[VAL_5]] : (index) -> !fir.shape<1>
-! CHECK:           %[[VAL_18:.*]] = fir.array_load %[[VAL_6]](%[[VAL_17]]) : (!fir.ref<!fir.array<10x!fir.char<1,5>>>, !fir.shape<1>) -> !fir.array<10x!fir.char<1,5>>
-! CHECK:           %[[VAL_19:.*]] = arith.constant 1 : index
-! CHECK:           %[[VAL_20:.*]] = arith.constant 0 : index
-! CHECK:           %[[VAL_21:.*]] = arith.subi %[[VAL_5]], %[[VAL_19]] : index
-! CHECK:           %[[VAL_22:.*]] = fir.do_loop %[[VAL_23:.*]] = %[[VAL_20]] to %[[VAL_21]] step %[[VAL_19]] unordered iter_args(%[[VAL_24:.*]] = %[[VAL_16]]) -> (!fir.array<10x!fir.char<1,5>>) {
-! CHECK:             %[[VAL_25:.*]] = fir.array_access %[[VAL_18]], %[[VAL_23]] : (!fir.array<10x!fir.char<1,5>>, index) -> !fir.ref<!fir.char<1,5>>
-! CHECK:             %[[VAL_26:.*]] = fir.array_access %[[VAL_24]], %[[VAL_23]] : (!fir.array<10x!fir.char<1,5>>, index) -> !fir.ref<!fir.char<1,5>>
-! CHECK:             %[[VAL_27:.*]] = arith.constant 5 : index
-! CHECK:             %[[VAL_28:.*]] = arith.constant 1 : i64
-! CHECK:             %[[VAL_29:.*]] = fir.convert %[[VAL_27]] : (index) -> i64
-! CHECK:             %[[VAL_30:.*]] = arith.muli %[[VAL_28]], %[[VAL_29]] : i64
-! CHECK:             %[[VAL_31:.*]] = arith.constant false
-! CHECK:             %[[VAL_32:.*]] = fir.convert %[[VAL_26]] : (!fir.ref<!fir.char<1,5>>) -> !fir.ref<i8>
-! CHECK:             %[[VAL_33:.*]] = fir.convert %[[VAL_25]] : (!fir.ref<!fir.char<1,5>>) -> !fir.ref<i8>
-! CHECK:             fir.call @llvm.memmove.p0.p0.i64(%[[VAL_32]], %[[VAL_33]], %[[VAL_30]], %[[VAL_31]]) {{.*}}: (!fir.ref<i8>, !fir.ref<i8>, i64, i1) -> ()
-! CHECK:             %[[VAL_34:.*]] = fir.array_amend %[[VAL_24]], %[[VAL_26]] : (!fir.array<10x!fir.char<1,5>>, !fir.ref<!fir.char<1,5>>) -> !fir.array<10x!fir.char<1,5>>
-! CHECK:             fir.result %[[VAL_34]] : !fir.array<10x!fir.char<1,5>>
-! CHECK:           }
-! CHECK:           fir.array_merge_store %[[VAL_16]], %[[VAL_35:.*]] to %[[VAL_14]] : !fir.array<10x!fir.char<1,5>>, !fir.array<10x!fir.char<1,5>>, !fir.ref<!fir.array<10x!fir.char<1,5>>>
-! CHECK:           omp.barrier
-! CHECK:           %[[VAL_37:.*]] = fir.emboxchar %[[VAL_7]], %[[VAL_1]] : (!fir.ref<!fir.char<1,5>>, index) -> !fir.boxchar<1>
-! CHECK:           %[[VAL_38:.*]] = fir.convert %[[VAL_14]] : (!fir.ref<!fir.array<10x!fir.char<1,5>>>) -> !fir.ref<!fir.char<1,?>>
-! CHECK:           %[[VAL_39:.*]] = fir.emboxchar %[[VAL_38]], %[[VAL_4]] : (!fir.ref<!fir.char<1,?>>, index) -> !fir.boxchar<1>
-! CHECK:           fir.call @_QPsub2(%[[VAL_37]], %[[VAL_39]]) {{.*}}: (!fir.boxchar<1>, !fir.boxchar<1>) -> ()
-! CHECK:           omp.terminator
-! CHECK:         }
-! CHECK:         return
-! CHECK:       }
-
-subroutine copyin_char_chararray()
-  character(5), save :: x3, x4(10)
-  !$omp threadprivate(x3, x4)
-
-  !$omp parallel copyin(x3) copyin(x4)
-    call sub2(x3, x4)
-  !$omp end parallel
-
-end
-
-! CHECK-LABEL: func.func @_QPcopyin_derived_type() {
-! CHECK:         %[[VAL_0:.*]] = fir.address_of(@_QFcopyin_derived_typeEx5) : !fir.ref<!fir.type<_QFcopyin_derived_typeTmy_type{t_i:i32,t_arr:!fir.array<5xi32>}>>
-! CHECK:         %[[VAL_1:.*]] = omp.threadprivate %[[VAL_0]] : !fir.ref<!fir.type<_QFcopyin_derived_typeTmy_type{t_i:i32,t_arr:!fir.array<5xi32>}>> -> !fir.ref<!fir.type<_QFcopyin_derived_typeTmy_type{t_i:i32,t_arr:!fir.array<5xi32>}>>
-! CHECK:         omp.parallel   {
-! CHECK:           %[[VAL_2:.*]] = omp.threadprivate %[[VAL_0]] : !fir.ref<!fir.type<_QFcopyin_derived_typeTmy_type{t_i:i32,t_arr:!fir.array<5xi32>}>> -> !fir.ref<!fir.type<_QFcopyin_derived_typeTmy_type{t_i:i32,t_arr:!fir.array<5xi32>}>>
-! CHECK:           %[[VAL_3:.*]] = fir.load %[[VAL_1]] : !fir.ref<!fir.type<_QFcopyin_derived_typeTmy_type{t_i:i32,t_arr:!fir.array<5xi32>}>>
-! CHECK:           fir.store %[[VAL_3]] to %[[VAL_2]] : !fir.ref<!fir.type<_QFcopyin_derived_typeTmy_type{t_i:i32,t_arr:!fir.array<5xi32>}>>
-! CHECK:           omp.barrier
-! CHECK:           fir.call @_QPsub3(%[[VAL_2]]) {{.*}}: (!fir.ref<!fir.type<_QFcopyin_derived_typeTmy_type{t_i:i32,t_arr:!fir.array<5xi32>}>>) -> ()
-! CHECK:           omp.terminator
-! CHECK:         }
-! CHECK:         return
-! CHECK:       }
-
-subroutine copyin_derived_type()
-  type my_type
-    integer :: t_i
-    integer :: t_arr(5)
-  end type my_type
-  type(my_type), save :: x5
-  !$omp threadprivate(x5)
-
-  !$omp parallel copyin(x5)
-    call sub3(x5)
-  !$omp end parallel
-
-end
-
-! CHECK-LABEL: func.func @_QPcombined_parallel_worksharing_loop() {
-! CHECK:         %[[VAL_0:.*]] = fir.alloca i32 {bindc_name = "i", uniq_name = "_QFcombined_parallel_worksharing_loopEi"}
-! CHECK:         %[[VAL_1:.*]] = fir.address_of(@_QFcombined_parallel_worksharing_loopEx6) : !fir.ref<i32>
-! CHECK:         %[[VAL_2:.*]] = omp.threadprivate %[[VAL_1]] : !fir.ref<i32> -> !fir.ref<i32>
-! CHECK:         omp.parallel   {
-! CHECK:           %[[VAL_3:.*]] = fir.alloca i32 {adapt.valuebyref, pinned}
-! CHECK:           %[[VAL_4:.*]] = omp.threadprivate %[[VAL_1]] : !fir.ref<i32> -> !fir.ref<i32>
-! CHECK:           %[[VAL_5:.*]] = fir.load %[[VAL_2]] : !fir.ref<i32>
-! CHECK:           fir.store %[[VAL_5]] to %[[VAL_4]] : !fir.ref<i32>
-! CHECK:           omp.barrier
-! CHECK:           %[[VAL_6:.*]] = arith.constant 1 : i32
-! CHECK:           %[[VAL_7:.*]] = fir.load %[[VAL_4]] : !fir.ref<i32>
-! CHECK:           %[[VAL_8:.*]] = arith.constant 1 : i32
-! CHECK:           omp.wsloop   for  (%[[VAL_9:.*]]) : i32 = (%[[VAL_6]]) to (%[[VAL_7]]) inclusive step (%[[VAL_8]]) {
-! CHECK:             fir.store %[[VAL_9]] to %[[VAL_3]] : !fir.ref<i32>
-! CHECK:             fir.call @_QPsub4(%[[VAL_4]]) {{.*}}: (!fir.ref<i32>) -> ()
-! CHECK:             omp.yield
-! CHECK:           }
-! CHECK:           omp.terminator
-! CHECK:         }
-! CHECK:         return
-! CHECK:       }
-
-subroutine combined_parallel_worksharing_loop()
-  integer, save :: x6
-  !$omp threadprivate(x6)
-
-  !$omp parallel do copyin(x6)
-    do i=1, x6
-      call sub4(x6)
-    end do
-  !$omp end parallel do
-
-end
-
-! CHECK-LABEL: func.func @_QPcombined_parallel_sections() {
-! CHECK:         %[[VAL_0:.*]] = fir.address_of(@_QFcombined_parallel_sectionsEx7) : !fir.ref<i32>
-! CHECK:         %[[VAL_1:.*]] = omp.threadprivate %[[VAL_0]] : !fir.ref<i32> -> !fir.ref<i32>
-! CHECK:         omp.parallel   {
-! CHECK:           %[[VAL_2:.*]] = omp.threadprivate %[[VAL_0]] : !fir.ref<i32> -> !fir.ref<i32>
-! CHECK:           %[[VAL_3:.*]] = fir.load %[[VAL_1]] : !fir.ref<i32>
-! CHECK:           fir.store %[[VAL_3]] to %[[VAL_2]] : !fir.ref<i32>
-! CHECK:           omp.barrier
-! CHECK:           omp.sections   {
-! CHECK:             omp.section {
-! CHECK:               fir.call @_QPsub5(%[[VAL_2]]) {{.*}}: (!fir.ref<i32>) -> ()
-! CHECK:               omp.terminator
-! CHECK:             }
-! CHECK:             omp.section {
-! CHECK:               fir.call @_QPsub6(%[[VAL_2]]) {{.*}}: (!fir.ref<i32>) -> ()
-! CHECK:               omp.terminator
-! CHECK:             }
-! CHECK:             omp.terminator
-! CHECK:           }
-! CHECK:           omp.terminator
-! CHECK:         }
-! CHECK:         return
-! CHECK:       }
-
-subroutine combined_parallel_sections()
-  integer, save :: x7
-  !$omp threadprivate(x7)
-
-  !$omp parallel sections copyin(x7)
-    !$omp section
-      call sub5(x7)
-    !$omp section
-      call sub6(x7)
-  !$omp end parallel sections
-
-end
-
-
-!CHECK: func.func @_QPcommon_1() {
-!CHECK: %[[val_0:.*]] = fir.address_of(@c_) : !fir.ref<!fir.array<4xi8>>
-!CHECK: %[[val_1:.*]] = omp.threadprivate %[[val_0]] : !fir.ref<!fir.array<4xi8>> -> !fir.ref<!fir.array<4xi8>>
-!CHECK: %[[val_2:.*]] = fir.convert %[[val_1]] : (!fir.ref<!fir.array<4xi8>>) -> !fir.ref<!fir.array<?xi8>>
-!CHECK: %[[val_c0:.*]] = arith.constant 0 : index
-!CHECK: %[[val_3:.*]] = fir.coordinate_of %[[val_2]], %[[val_c0]] : (!fir.ref<!fir.array<?xi8>>, index) -> !fir.ref<i8>
-!CHECK: %[[val_4:.*]] = fir.convert %[[val_3]] : (!fir.ref<i8>) -> !fir.ref<i32>
-!CHECK: %[[val_5:.*]] = fir.alloca i32 {bindc_name = "y", uniq_name = "_QFcommon_1Ey"}
-!CHECK: omp.parallel {
-!CHECK: %[[val_6:.*]] = omp.threadprivate %[[val_0]] : !fir.ref<!fir.array<4xi8>> -> !fir.ref<!fir.array<4xi8>>
-!CHECK: %[[val_7:.*]] = fir.convert %[[val_6]] : (!fir.ref<!fir.array<4xi8>>) -> !fir.ref<!fir.array<?xi8>>
-!CHECK: %[[val_c0_0:.*]] = arith.constant 0 : index
-!CHECK: %[[val_8:.*]] = fir.coordinate_of %[[val_7]], %[[val_c0_0]] : (!fir.ref<!fir.array<?xi8>>, index) -> !fir.ref<i8>
-!CHECK: %[[val_9:.*]] = fir.convert %[[val_8]] : (!fir.ref<i8>) -> !fir.ref<i32>
-!CHECK: %[[val_10:.*]] = fir.load %[[val_4]] : !fir.ref<i32>
-!CHECK: fir.store %[[val_10]] to %[[val_9]] : !fir.ref<i32>
-!CHECK: omp.barrier
-!CHECK: omp.sections   {
-!CHECK: omp.section {
-!CHECK: %[[val_11:.*]] = fir.load %[[val_9]] : !fir.ref<i32>
-!CHECK: %[[val_c1_i32:.*]] = arith.constant 1 : i32
-!CHECK: %[[val_12:.*]] = arith.addi %[[val_11]], %[[val_c1_i32]] : i32
-!CHECK: fir.store %[[val_12]] to %[[val_5]] : !fir.ref<i32>
-!CHECK: omp.terminator
-!CHECK: }
-!CHECK: omp.section {
-!CHECK: %[[val_11:.*]] = fir.load %[[val_5]] : !fir.ref<i32>
-!CHECK: %[[val_12:.*]] = fir.load %[[val_5]] : !fir.ref<i32>
-!CHECK: %[[val_13:.*]] = arith.muli %[[val_11]], %[[val_12]] : i32
-!CHECK: fir.store %[[val_13]] to %[[val_9]] : !fir.ref<i32>
-!CHECK: omp.terminator
-!CHECK: }
-!CHECK: omp.terminator
-!CHECK: }
-!CHECK: return
-!CHECK: }
-subroutine common_1()
-  integer :: x
-  integer :: y
-  common /c/ x
-  !$omp threadprivate(/c/)
-
-  !$omp parallel sections copyin(/c/)
-      !$omp section
-        y = x + 1
-      !$omp section
-        x = y * y
-  !$omp end parallel sections
-end subroutine
-
-!CHECK: func.func @_QPcommon_2() {
-!CHECK: %[[val_0:.*]] = fir.alloca i32 {bindc_name = "i", uniq_name = "_QFcommon_2Ei"}
-!CHECK: %[[val_1:.*]] = fir.address_of(@d_) : !fir.ref<!fir.array<8xi8>>
-!CHECK: %[[val_2:.*]] = omp.threadprivate %[[val_1]] : !fir.ref<!fir.array<8xi8>> -> !fir.ref<!fir.array<8xi8>>
-!CHECK: %[[val_3:.*]] = fir.convert %[[val_2]] : (!fir.ref<!fir.array<8xi8>>) -> !fir.ref<!fir.array<?xi8>>
-!CHECK: %[[val_c0:.*]] = arith.constant 0 : index
-!CHECK: %[[val_4:.*]] = fir.coordinate_of %[[val_3]], %[[val_c0]] : (!fir.ref<!fir.array<?xi8>>, index) -> !fir.ref<i8>
-!CHECK: %[[val_5:.*]] = fir.convert %[[val_4]] : (!fir.ref<i8>) -> !fir.ref<i32>
-!CHECK: %[[val_6:.*]] = fir.convert %[[val_2]] : (!fir.ref<!fir.array<8xi8>>) -> !fir.ref<!fir.array<?xi8>>
-!CHECK: %[[val_c4:.*]] = arith.constant 4 : index
-!CHECK: %[[val_7:.*]] = fir.coordinate_of %[[val_6]], %[[val_c4]] : (!fir.ref<!fir.array<?xi8>>, index) -> !fir.ref<i8>
-!CHECK: %[[val_8:.*]] = fir.convert %[[val_7]] : (!fir.ref<i8>) -> !fir.ref<i32>
-!CHECK: omp.parallel {
-!CHECK: %[[val_9:.*]] = fir.alloca i32 {adapt.valuebyref, pinned}
-!CHECK: %[[val_10:.*]] = omp.threadprivate %[[val_1]] : !fir.ref<!fir.array<8xi8>> -> !fir.ref<!fir.array<8xi8>>
-!CHECK: %[[val_11:.*]] = fir.convert %[[val_10]] : (!fir.ref<!fir.array<8xi8>>) -> !fir.ref<!fir.array<?xi8>>
-!CHECK: %[[val_c0_0:.*]] = arith.constant 0 : index
-!CHECK: %[[val_12:.*]] = fir.coordinate_of %[[val_11]], %[[val_c0_0]] : (!fir.ref<!fir.array<?xi8>>, index) -> !fir.ref<i8>
-!CHECK: %[[val_13:.*]] = fir.convert %[[val_12]] : (!fir.ref<i8>) -> !fir.ref<i32>
-!CHECK: %[[val_14:.*]] = fir.convert %[[val_10]] : (!fir.ref<!fir.array<8xi8>>) -> !fir.ref<!fir.array<?xi8>>
-!CHECK: %[[val_c4_1:.*]] = arith.constant 4 : index
-!CHECK: %[[val_15:.*]] = fir.coordinate_of %[[val_14]], %[[val_c4_1]] : (!fir.ref<!fir.array<?xi8>>, index) -> !fir.ref<i8>
-!CHECK: %[[val_16:.*]] = fir.convert %[[val_15]] : (!fir.ref<i8>) -> !fir.ref<i32>
-!CHECK: %[[val_17:.*]] = fir.load %[[val_5]] : !fir.ref<i32>
-!CHECK: fir.store %[[val_17]] to %[[val_13]] : !fir.ref<i32>
-!CHECK: %[[val_18:.*]] = fir.load %[[val_8]] : !fir.ref<i32>
-!CHECK: fir.store %[[val_18]] to %[[val_16]] : !fir.ref<i32>
-!CHECK: omp.barrier
-!CHECK: %[[val_c1_i32:.*]] = arith.constant 1 : i32
-!CHECK: %[[val_19:.*]] = fir.load %[[val_13]] : !fir.ref<i32>
-!CHECK: %[[val_c1_i32_2:.*]] = arith.constant 1 : i32
-!CHECK: omp.wsloop   for (%[[arg:.*]]) : i32 = (%[[val_c1_i32]]) to (%[[val_19]]) inclusive step (%[[val_c1_i32_2]]) {
-!CHECK: fir.store %[[arg]] to %[[val_9]] : !fir.ref<i32>
-!CHECK: %[[val_20:.*]] = fir.load %[[val_16]] : !fir.ref<i32>
-!CHECK: %[[val_21:.*]] = fir.load %[[val_9]] : !fir.ref<i32>
-!CHECK: %[[val_22:.*]] = arith.addi %[[val_20]], %[[val_21]] : i32
-!CHECK: fir.store %[[val_22]] to %[[val_16]] : !fir.ref<i32>
-!CHECK: omp.yield
-!CHECK: }
-!CHECK: omp.terminator
-!CHECK: }
-!CHECK: return
-!CHECK: }
-subroutine common_2()
-  integer :: x
-  integer :: y
-  common /d/ x, y
-  !$omp threadprivate(/d/)
-  
-  !$omp parallel do copyin(/d/)
-     do i = 1, x
-        y = y + i
-     end do
-  !$omp end parallel do
-end subroutine
-
-!CHECK: func.func @_QPcommon_3() {
-!CHECK: %[[val_0:.*]] = fir.address_of(@blk_) : !fir.ref<!fir.array<8xi8>>
-!CHECK: %[[val_1:.*]] = omp.threadprivate %[[val_0]] : !fir.ref<!fir.array<8xi8>> -> !fir.ref<!fir.array<8xi8>>
-!CHECK: %[[val_2:.*]] = fir.convert %[[val_1]] : (!fir.ref<!fir.array<8xi8>>) -> !fir.ref<!fir.array<?xi8>>
-!CHECK: %[[val_c4:.*]] = arith.constant 4 : index
-!CHECK: %[[val_3:.*]] = fir.coordinate_of %[[val_2]], %[[val_c4]] : (!fir.ref<!fir.array<?xi8>>, index) -> !fir.ref<i8>
-!CHECK: %[[val_4:.*]] = fir.convert %[[val_3]] : (!fir.ref<i8>) -> !fir.ref<i32>
-!CHECK: omp.parallel {
-!CHECK: %[[val_5:.*]] = omp.threadprivate %[[val_0]] : !fir.ref<!fir.array<8xi8>> -> !fir.ref<!fir.array<8xi8>>
-!CHECK: %[[val_6:.*]] = fir.convert %[[val_5]] : (!fir.ref<!fir.array<8xi8>>) -> !fir.ref<!fir.array<?xi8>>
-!CHECK: %[[val_c4_0:.*]] = arith.constant 4 : index
-!CHECK: %[[val_7:.*]] = fir.coordinate_of %[[val_6]], %[[val_c4_0]] : (!fir.ref<!fir.array<?xi8>>, index) -> !fir.ref<i8>
-!CHECK: %[[val_8:.*]] = fir.convert %[[val_7]] : (!fir.ref<i8>) -> !fir.ref<i32>
-!CHECK: %[[val_9:.*]] = fir.load %[[val_4]] : !fir.ref<i32>
-!CHECK: fir.store %[[val_9]] to %[[val_8]] : !fir.ref<i32>
-!CHECK: omp.barrier
-!CHECK: omp.sections {
-!CHECK: omp.section {
-!CHECK: %[[val_10:.*]] = fir.load %[[val_8]] : !fir.ref<i32>
-!CHECK: %[[val_c3_i32:.*]] = arith.constant 3 : i32
-!CHECK: %[[val_11:.*]] = arith.addi %[[val_10]], %[[val_c3_i32]] : i32
-!CHECK: fir.store %[[val_11]] to %[[val_8]] : !fir.ref<i32>
-!CHECK: omp.terminator
-!CHECK: }
-!CHECK: omp.terminator
-!CHECK: }
-!CHECK: return
-!CHECK: }
-subroutine common_3()
-  integer :: x
-  integer :: y
-  common /blk/ x, y
-  !$omp threadprivate (/blk/)
-
-  !$omp parallel sections copyin(/blk/)
-        !$omp section
-            y = y + 3
-  !$omp end parallel sections
-end subroutine
diff --git a/flang/test/Lower/OpenMP/FIR/critical.f90 b/flang/test/Lower/OpenMP/FIR/critical.f90
deleted file mode 100644
index fa33fb0fe58b..000000000000
--- a/flang/test/Lower/OpenMP/FIR/critical.f90
+++ /dev/null
@@ -1,38 +0,0 @@
-! REQUIRES: openmp_runtime
-
-!RUN: %flang_fc1 -emit-fir -flang-deprecated-no-hlfir -fopenmp %s -o - | FileCheck %s --check-prefixes="OMPDialect"
-!RUN: %flang_fc1 -emit-fir -flang-deprecated-no-hlfir -fopenmp %s -o - | fir-opt --fir-to-llvm-ir | FileCheck %s --check-prefix="OMPDialect"
-!RUN: %flang_fc1 -emit-fir -flang-deprecated-no-hlfir -fopenmp %s -o - | fir-opt --fir-to-llvm-ir | tco | FileCheck %s --check-prefix="LLVMIR"
-
-!OMPDialect: omp.critical.declare @help2
-!OMPDialect: omp.critical.declare @help1 hint(contended)
-
-subroutine omp_critical()
-  use omp_lib
-  integer :: x, y
-!OMPDialect: omp.critical(@help1)
-!LLVMIR: call void @__kmpc_critical_with_hint({{.*}}, {{.*}}, {{.*}} @{{.*}}help1.var, i32 2)
-!$OMP CRITICAL(help1) HINT(omp_lock_hint_contended)
-  x = x + y
-!OMPDialect: omp.terminator
-!LLVMIR: call void @__kmpc_end_critical({{.*}}, {{.*}}, {{.*}} @{{.*}}help1.var)
-!$OMP END CRITICAL(help1)
-
-! Test that the same name can be used again
-! Also test with the zero hint expression
-!OMPDialect: omp.critical(@help2)
-!LLVMIR: call void @__kmpc_critical_with_hint({{.*}}, {{.*}}, {{.*}} @{{.*}}help2.var, i32 0)
-!$OMP CRITICAL(help2) HINT(omp_lock_hint_none)
-  x = x - y
-!OMPDialect: omp.terminator
-!LLVMIR: call void @__kmpc_end_critical({{.*}}, {{.*}}, {{.*}} @{{.*}}help2.var)
-!$OMP END CRITICAL(help2)
-
-!OMPDialect: omp.critical
-!LLVMIR: call void @__kmpc_critical({{.*}}, {{.*}}, {{.*}} @{{.*}}_.var)
-!$OMP CRITICAL
-  y = x + y
-!OMPDialect: omp.terminator
-!LLVMIR: call void @__kmpc_end_critical({{.*}}, {{.*}}, {{.*}} @{{.*}}_.var)
-!$OMP END CRITICAL
-end subroutine omp_critical
diff --git a/flang/test/Lower/OpenMP/FIR/declare-target-data.f90 b/flang/test/Lower/OpenMP/FIR/declare-target-data.f90
deleted file mode 100644
index bb3bbc8dfa83..000000000000
--- a/flang/test/Lower/OpenMP/FIR/declare-target-data.f90
+++ /dev/null
@@ -1,88 +0,0 @@
-!RUN: %flang_fc1 -emit-fir -fopenmp %s -o - | FileCheck %s 
-!RUN: %flang_fc1 -emit-fir -fopenmp -fopenmp-is-target-device %s -o - | FileCheck %s
-
-module test_0
-    implicit none
-
-!CHECK-DAG: fir.global @_QMtest_0Edata_int {omp.declare_target = #omp.declaretarget<device_type = (any), capture_clause = (link)>} : i32
-INTEGER :: data_int = 10
-!$omp declare target link(data_int)
-
-!CHECK-DAG: fir.global @_QMtest_0Earray_1d({{.*}}) {omp.declare_target = #omp.declaretarget<device_type = (any), capture_clause = (link)>} : !fir.array<3xi32>
-INTEGER :: array_1d(3) = (/1,2,3/)
-!$omp declare target link(array_1d)
-
-!CHECK-DAG: fir.global @_QMtest_0Earray_2d({{.*}}) {omp.declare_target = #omp.declaretarget<device_type = (any), capture_clause = (link)>} : !fir.array<2x2xi32> 
-INTEGER :: array_2d(2,2) = reshape((/1,2,3,4/), (/2,2/))
-!$omp declare target link(array_2d)
-
-!CHECK-DAG: fir.global @_QMtest_0Ept1 {omp.declare_target = #omp.declaretarget<device_type = (any), capture_clause = (link)>} : !fir.box<!fir.ptr<i32>>
-INTEGER, POINTER :: pt1
-!$omp declare target link(pt1)
-
-!CHECK-DAG: fir.global @_QMtest_0Ept2_tar {omp.declare_target = #omp.declaretarget<device_type = (any), capture_clause = (link)>} target : i32
-INTEGER, TARGET :: pt2_tar = 5 
-!$omp declare target link(pt2_tar)
-
-!CHECK-DAG: fir.global @_QMtest_0Ept2 {omp.declare_target = #omp.declaretarget<device_type = (any), capture_clause = (link)>} : !fir.box<!fir.ptr<i32>>
-INTEGER, POINTER :: pt2 => pt2_tar
-!$omp declare target link(pt2)
-
-!CHECK-DAG: fir.global @_QMtest_0Edata_int_to {omp.declare_target = #omp.declaretarget<device_type = (any), capture_clause = (to)>} : i32
-INTEGER :: data_int_to = 5
-!$omp declare target to(data_int_to)
-
-!CHECK-DAG: fir.global @_QMtest_0Edata_int_enter {omp.declare_target = #omp.declaretarget<device_type = (any), capture_clause = (enter)>} : i32
-INTEGER :: data_int_enter = 5
-!$omp declare target enter(data_int_enter)
-
-!CHECK-DAG: fir.global @_QMtest_0Edata_int_clauseless {omp.declare_target = #omp.declaretarget<device_type = (any), capture_clause = (to)>} : i32
-INTEGER :: data_int_clauseless = 1
-!$omp declare target(data_int_clauseless)
-
-!CHECK-DAG: fir.global @_QMtest_0Edata_extended_to_1 {omp.declare_target = #omp.declaretarget<device_type = (any), capture_clause = (to)>} : f32
-!CHECK-DAG: fir.global @_QMtest_0Edata_extended_to_2 {omp.declare_target = #omp.declaretarget<device_type = (any), capture_clause = (to)>} : f32
-REAL :: data_extended_to_1 = 2
-REAL :: data_extended_to_2 = 3
-!$omp declare target to(data_extended_to_1, data_extended_to_2)
-
-!CHECK-DAG: fir.global @_QMtest_0Edata_extended_enter_1 {omp.declare_target = #omp.declaretarget<device_type = (any), capture_clause = (enter)>} : f32
-!CHECK-DAG: fir.global @_QMtest_0Edata_extended_enter_2 {omp.declare_target = #omp.declaretarget<device_type = (any), capture_clause = (enter)>} : f32
-REAL :: data_extended_enter_1 = 2
-REAL :: data_extended_enter_2 = 3
-!$omp declare target enter(data_extended_enter_1, data_extended_enter_2)
-
-!CHECK-DAG: fir.global @_QMtest_0Edata_extended_link_1 {omp.declare_target = #omp.declaretarget<device_type = (any), capture_clause = (link)>} : f32
-!CHECK-DAG: fir.global @_QMtest_0Edata_extended_link_2 {omp.declare_target = #omp.declaretarget<device_type = (any), capture_clause = (link)>} : f32
-REAL :: data_extended_link_1 = 2
-REAL :: data_extended_link_2 = 3
-!$omp declare target link(data_extended_link_1, data_extended_link_2)
-
-contains
-end module test_0
-
-PROGRAM commons
-    !CHECK-DAG: fir.global @numbers_ {omp.declare_target = #omp.declaretarget<device_type = (any), capture_clause = (to)>} : tuple<f32, f32> {
-    REAL :: one = 1
-    REAL :: two = 2
-    COMMON /numbers/ one, two
-    !$omp declare target(/numbers/)
-    
-    !CHECK-DAG: fir.global @numbers_link_ {omp.declare_target = #omp.declaretarget<device_type = (any), capture_clause = (link)>} : tuple<f32, f32> {
-    REAL :: one_link = 1
-    REAL :: two_link = 2
-    COMMON /numbers_link/ one_link, two_link
-    !$omp declare target link(/numbers_link/)
-
-    !CHECK-DAG: fir.global @numbers_to_ {omp.declare_target = #omp.declaretarget<device_type = (any), capture_clause = (to)>} : tuple<f32, f32> {
-    REAL :: one_to = 1
-    REAL :: two_to = 2
-    COMMON /numbers_to/ one_to, two_to
-    !$omp declare target to(/numbers_to/)
-
-    !CHECK-DAG: fir.global @numbers_enter_ {omp.declare_target = #omp.declaretarget<device_type = (any), capture_clause = (enter)>} : tuple<f32, f32> {
-    REAL :: one_enter = 1
-    REAL :: two_enter = 2
-    COMMON /numbers_enter/ one_enter, two_enter
-    !$omp declare target enter(/numbers_enter/)
-END
diff --git a/flang/test/Lower/OpenMP/FIR/declare-target-func-and-subr.f90 b/flang/test/Lower/OpenMP/FIR/declare-target-func-and-subr.f90
deleted file mode 100644
index 36d4d7db64e5..000000000000
--- a/flang/test/Lower/OpenMP/FIR/declare-target-func-and-subr.f90
+++ /dev/null
@@ -1,178 +0,0 @@
-!RUN: %flang_fc1 -emit-fir -fopenmp %s -o - | FileCheck %s --check-prefixes ALL,HOST
-!RUN: %flang_fc1 -emit-fir -fopenmp -fopenmp-is-device %s -o - | FileCheck %s --check-prefixes ALL,DEVICE
-
-! Check specification valid forms of declare target with functions 
-! utilising device_type and to clauses as well as the default 
-! zero clause declare target
-
-! DEVICE-LABEL: func.func @_QPfunc_t_device()
-! DEVICE-SAME: {{.*}}attributes {omp.declare_target = #omp.declaretarget<device_type = (nohost), capture_clause = (to)>{{.*}}
-FUNCTION FUNC_T_DEVICE() RESULT(I)
-!$omp declare target to(FUNC_T_DEVICE) device_type(nohost)
-    INTEGER :: I
-    I = 1
-END FUNCTION FUNC_T_DEVICE
-
-! DEVICE-LABEL: func.func @_QPfunc_enter_device()
-! DEVICE-SAME: {{.*}}attributes {omp.declare_target = #omp.declaretarget<device_type = (nohost), capture_clause = (enter)>{{.*}}
-FUNCTION FUNC_ENTER_DEVICE() RESULT(I)
-!$omp declare target enter(FUNC_ENTER_DEVICE) device_type(nohost)
-    INTEGER :: I
-    I = 1
-END FUNCTION FUNC_ENTER_DEVICE
-
-! HOST-LABEL: func.func @_QPfunc_t_host()
-! HOST-SAME: {{.*}}attributes {omp.declare_target = #omp.declaretarget<device_type = (host), capture_clause = (to)>{{.*}}
-FUNCTION FUNC_T_HOST() RESULT(I)
-!$omp declare target to(FUNC_T_HOST) device_type(host)
-    INTEGER :: I
-    I = 1
-END FUNCTION FUNC_T_HOST
-
-! HOST-LABEL: func.func @_QPfunc_enter_host()
-! HOST-SAME: {{.*}}attributes {omp.declare_target = #omp.declaretarget<device_type = (host), capture_clause = (enter)>{{.*}}
-FUNCTION FUNC_ENTER_HOST() RESULT(I)
-!$omp declare target enter(FUNC_ENTER_HOST) device_type(host)
-    INTEGER :: I
-    I = 1
-END FUNCTION FUNC_ENTER_HOST
-
-! ALL-LABEL: func.func @_QPfunc_t_any()
-! ALL-SAME: {{.*}}attributes {omp.declare_target = #omp.declaretarget<device_type = (any), capture_clause = (to)>{{.*}}
-FUNCTION FUNC_T_ANY() RESULT(I)
-!$omp declare target to(FUNC_T_ANY) device_type(any)
-    INTEGER :: I
-    I = 1
-END FUNCTION FUNC_T_ANY
-
-! ALL-LABEL: func.func @_QPfunc_enter_any()
-! ALL-SAME: {{.*}}attributes {omp.declare_target = #omp.declaretarget<device_type = (any), capture_clause = (enter)>{{.*}}
-FUNCTION FUNC_ENTER_ANY() RESULT(I)
-!$omp declare target enter(FUNC_ENTER_ANY) device_type(any)
-    INTEGER :: I
-    I = 1
-END FUNCTION FUNC_ENTER_ANY
-
-! ALL-LABEL: func.func @_QPfunc_default_t_any()
-! ALL-SAME: {{.*}}attributes {omp.declare_target = #omp.declaretarget<device_type = (any), capture_clause = (to)>{{.*}}
-FUNCTION FUNC_DEFAULT_T_ANY() RESULT(I)
-!$omp declare target to(FUNC_DEFAULT_T_ANY)
-    INTEGER :: I
-    I = 1
-END FUNCTION FUNC_DEFAULT_T_ANY
-
-! ALL-LABEL: func.func @_QPfunc_default_enter_any()
-! ALL-SAME: {{.*}}attributes {omp.declare_target = #omp.declaretarget<device_type = (any), capture_clause = (enter)>{{.*}}
-FUNCTION FUNC_DEFAULT_ENTER_ANY() RESULT(I)
-!$omp declare target enter(FUNC_DEFAULT_ENTER_ANY)
-    INTEGER :: I
-    I = 1
-END FUNCTION FUNC_DEFAULT_ENTER_ANY
-
-! ALL-LABEL: func.func @_QPfunc_default_any()
-! ALL-SAME: {{.*}}attributes {omp.declare_target = #omp.declaretarget<device_type = (any), capture_clause = (to)>{{.*}}
-FUNCTION FUNC_DEFAULT_ANY() RESULT(I)
-!$omp declare target
-    INTEGER :: I
-    I = 1
-END FUNCTION FUNC_DEFAULT_ANY
-
-! ALL-LABEL: func.func @_QPfunc_default_extendedlist()
-! ALL-SAME: {{.*}}attributes {omp.declare_target = #omp.declaretarget<device_type = (any), capture_clause = (to)>{{.*}}
-FUNCTION FUNC_DEFAULT_EXTENDEDLIST() RESULT(I)
-!$omp declare target(FUNC_DEFAULT_EXTENDEDLIST)
-    INTEGER :: I
-    I = 1
-END FUNCTION FUNC_DEFAULT_EXTENDEDLIST
-
-!! -----
-
-! Check specification valid forms of declare target with subroutines 
-! utilising device_type and to clauses as well as the default 
-! zero clause declare target
-
-! DEVICE-LABEL: func.func @_QPsubr_t_device()
-! DEVICE-SAME: {{.*}}attributes {omp.declare_target = #omp.declaretarget<device_type = (nohost), capture_clause = (to)>{{.*}}
-SUBROUTINE SUBR_T_DEVICE()
-!$omp declare target to(SUBR_T_DEVICE) device_type(nohost)
-END
-
-! DEVICE-LABEL: func.func @_QPsubr_enter_device()
-! DEVICE-SAME: {{.*}}attributes {omp.declare_target = #omp.declaretarget<device_type = (nohost), capture_clause = (enter)>{{.*}}
-SUBROUTINE SUBR_ENTER_DEVICE()
-!$omp declare target enter(SUBR_ENTER_DEVICE) device_type(nohost)
-END
-
-! HOST-LABEL: func.func @_QPsubr_t_host()
-! HOST-SAME: {{.*}}attributes {omp.declare_target = #omp.declaretarget<device_type = (host), capture_clause = (to)>{{.*}}
-SUBROUTINE SUBR_T_HOST()
-!$omp declare target to(SUBR_T_HOST) device_type(host)
-END
-
-! HOST-LABEL: func.func @_QPsubr_enter_host()
-! HOST-SAME: {{.*}}attributes {omp.declare_target = #omp.declaretarget<device_type = (host), capture_clause = (enter)>{{.*}}
-SUBROUTINE SUBR_ENTER_HOST()
-!$omp declare target enter(SUBR_ENTER_HOST) device_type(host)
-END
-
-! ALL-LABEL: func.func @_QPsubr_t_any()
-! ALL-SAME: {{.*}}attributes {omp.declare_target = #omp.declaretarget<device_type = (any), capture_clause = (to)>{{.*}}
-SUBROUTINE SUBR_T_ANY()
-!$omp declare target to(SUBR_T_ANY) device_type(any)
-END
-
-! ALL-LABEL: func.func @_QPsubr_enter_any()
-! ALL-SAME: {{.*}}attributes {omp.declare_target = #omp.declaretarget<device_type = (any), capture_clause = (enter)>{{.*}}
-SUBROUTINE SUBR_ENTER_ANY()
-!$omp declare target enter(SUBR_ENTER_ANY) device_type(any)
-END
-
-! ALL-LABEL: func.func @_QPsubr_default_t_any()
-! ALL-SAME: {{.*}}attributes {omp.declare_target = #omp.declaretarget<device_type = (any), capture_clause = (to)>{{.*}}
-SUBROUTINE SUBR_DEFAULT_T_ANY()
-!$omp declare target to(SUBR_DEFAULT_T_ANY)
-END
-
-! ALL-LABEL: func.func @_QPsubr_default_enter_any()
-! ALL-SAME: {{.*}}attributes {omp.declare_target = #omp.declaretarget<device_type = (any), capture_clause = (enter)>{{.*}}
-SUBROUTINE SUBR_DEFAULT_ENTER_ANY()
-!$omp declare target enter(SUBR_DEFAULT_ENTER_ANY)
-END
-
-! ALL-LABEL: func.func @_QPsubr_default_any()
-! ALL-SAME: {{.*}}attributes {omp.declare_target = #omp.declaretarget<device_type = (any), capture_clause = (to)>{{.*}}
-SUBROUTINE SUBR_DEFAULT_ANY()
-!$omp declare target
-END
-
-! ALL-LABEL: func.func @_QPsubr_default_extendedlist()
-! ALL-SAME: {{.*}}attributes {omp.declare_target = #omp.declaretarget<device_type = (any), capture_clause = (to)>{{.*}}
-SUBROUTINE SUBR_DEFAULT_EXTENDEDLIST()
-!$omp declare target(SUBR_DEFAULT_EXTENDEDLIST)
-END
-
-!! -----
-
-! DEVICE-LABEL: func.func @_QPrecursive_declare_target
-! DEVICE-SAME: {{.*}}attributes {omp.declare_target = #omp.declaretarget<device_type = (nohost), capture_clause = (to)>{{.*}}
-RECURSIVE FUNCTION RECURSIVE_DECLARE_TARGET(INCREMENT) RESULT(K)
-!$omp declare target to(RECURSIVE_DECLARE_TARGET) device_type(nohost)
-    INTEGER :: INCREMENT, K
-    IF (INCREMENT == 10) THEN
-        K = INCREMENT
-    ELSE
-        K = RECURSIVE_DECLARE_TARGET(INCREMENT + 1)
-    END IF
-END FUNCTION RECURSIVE_DECLARE_TARGET
-
-! DEVICE-LABEL: func.func @_QPrecursive_declare_target_enter
-! DEVICE-SAME: {{.*}}attributes {omp.declare_target = #omp.declaretarget<device_type = (nohost), capture_clause = (enter)>{{.*}}
-RECURSIVE FUNCTION RECURSIVE_DECLARE_TARGET_ENTER(INCREMENT) RESULT(K)
-!$omp declare target enter(RECURSIVE_DECLARE_TARGET_ENTER) device_type(nohost)
-    INTEGER :: INCREMENT, K
-    IF (INCREMENT == 10) THEN
-        K = INCREMENT
-    ELSE
-        K = RECURSIVE_DECLARE_TARGET_ENTER(INCREMENT + 1)
-    END IF
-END FUNCTION RECURSIVE_DECLARE_TARGET_ENTER
diff --git a/flang/test/Lower/OpenMP/FIR/declare-target-implicit-func-and-subr-cap-enter.f90 b/flang/test/Lower/OpenMP/FIR/declare-target-implicit-func-and-subr-cap-enter.f90
deleted file mode 100644
index 8e88d1b0f52a..000000000000
--- a/flang/test/Lower/OpenMP/FIR/declare-target-implicit-func-and-subr-cap-enter.f90
+++ /dev/null
@@ -1,192 +0,0 @@
-!RUN: %flang_fc1 -emit-fir -fopenmp %s -o - | FileCheck %s
-!RUN: %flang_fc1 -emit-fir -fopenmp -fopenmp-is-target-device %s -o - | FileCheck %s  --check-prefix=DEVICE
-!RUN: bbc -emit-fir -fopenmp %s -o - | FileCheck %s
-!RUN: bbc -emit-fir -fopenmp -fopenmp-is-target-device %s -o - | FileCheck %s --check-prefix=DEVICE
-
-! CHECK-LABEL: func.func @_QPimplicitly_captured_twice
-! CHECK-SAME: {{.*}}attributes {omp.declare_target = #omp.declaretarget<device_type = (any), capture_clause = (enter)>{{.*}}}
-function implicitly_captured_twice() result(k)
-   integer :: i
-   i = 10
-   k = i
-end function implicitly_captured_twice
-
-! CHECK-LABEL: func.func @_QPtarget_function_twice_host
-! CHECK-SAME: {{.*}}attributes {omp.declare_target = #omp.declaretarget<device_type = (host), capture_clause = (enter)>{{.*}}}
-function target_function_twice_host() result(i)
-!$omp declare target enter(target_function_twice_host) device_type(host)
-   integer :: i
-   i = implicitly_captured_twice()
-end function target_function_twice_host
-
-! DEVICE-LABEL: func.func @_QPtarget_function_twice_device
-! DEVICE-SAME: {{.*}}attributes {omp.declare_target = #omp.declaretarget<device_type = (nohost), capture_clause = (enter)>{{.*}}}
-function target_function_twice_device() result(i)
-!$omp declare target enter(target_function_twice_device) device_type(nohost)
-   integer :: i
-   i = implicitly_captured_twice()
-end function target_function_twice_device
-
-!! -----
-
-! DEVICE-LABEL: func.func @_QPimplicitly_captured_nest
-! DEVICE-SAME: {{.*}}attributes {omp.declare_target = #omp.declaretarget<device_type = (nohost), capture_clause = (enter)>{{.*}}}
-function implicitly_captured_nest() result(k)
-   integer :: i
-   i = 10
-   k = i
-end function implicitly_captured_nest
-
-! DEVICE-LABEL: func.func @_QPimplicitly_captured_one
-! DEVICE-SAME: {{.*}}attributes {omp.declare_target = #omp.declaretarget<device_type = (nohost), capture_clause = (enter){{.*}}}
-function implicitly_captured_one() result(k)
-   k = implicitly_captured_nest()
-end function implicitly_captured_one
-
-! DEVICE-LABEL: func.func @_QPimplicitly_captured_two
-! DEVICE-SAME: {{.*}}attributes {omp.declare_target = #omp.declaretarget<device_type = (nohost), capture_clause = (enter)>{{.*}}}
-function implicitly_captured_two() result(k)
-   integer :: i
-   i = 10
-   k = i
-end function implicitly_captured_two
-
-! DEVICE-LABEL: func.func @_QPtarget_function_test
-! DEVICE-SAME: {{.*}}attributes {omp.declare_target = #omp.declaretarget<device_type = (nohost), capture_clause = (enter)>{{.*}}}
-function target_function_test() result(j)
-!$omp declare target enter(target_function_test) device_type(nohost)
-   integer :: i, j
-   i = implicitly_captured_one()
-   j = implicitly_captured_two() + i
-end function target_function_test
-
-!! -----
-
-! CHECK-LABEL: func.func @_QPimplicitly_captured_nest_twice
-! CHECK-SAME: {{.*}}attributes {omp.declare_target = #omp.declaretarget<device_type = (any), capture_clause = (enter)>{{.*}}}
-function implicitly_captured_nest_twice() result(k)
-   integer :: i
-   i = 10
-   k = i
-end function implicitly_captured_nest_twice
-
-! CHECK-LABEL: func.func @_QPimplicitly_captured_one_twice
-! CHECK-SAME: {{.*}}attributes {omp.declare_target = #omp.declaretarget<device_type = (any), capture_clause = (enter)>{{.*}}}
-function implicitly_captured_one_twice() result(k)
-   k = implicitly_captured_nest_twice()
-end function implicitly_captured_one_twice
-
-! CHECK-LABEL: func.func @_QPimplicitly_captured_two_twice
-! CHECK-SAME: {{.*}}attributes {omp.declare_target = #omp.declaretarget<device_type = (any), capture_clause = (enter)>{{.*}}}
-function implicitly_captured_two_twice() result(k)
-   integer :: i
-   i = 10
-   k = i
-end function implicitly_captured_two_twice
-
-! DEVICE-LABEL: func.func @_QPtarget_function_test_device
-! DEVICE-SAME: {{.*}}attributes {omp.declare_target = #omp.declaretarget<device_type = (nohost), capture_clause = (enter)>{{.*}}}
-function target_function_test_device() result(j)
-   !$omp declare target enter(target_function_test_device) device_type(nohost)
-   integer :: i, j
-   i = implicitly_captured_one_twice()
-   j = implicitly_captured_two_twice() + i
-end function target_function_test_device
-
-! CHECK-LABEL: func.func @_QPtarget_function_test_host
-! CHECK-SAME: {{.*}}attributes {omp.declare_target = #omp.declaretarget<device_type = (host), capture_clause = (enter)>{{.*}}}
-function target_function_test_host() result(j)
-   !$omp declare target enter(target_function_test_host) device_type(host)
-   integer :: i, j
-   i = implicitly_captured_one_twice()
-   j = implicitly_captured_two_twice() + i
-end function target_function_test_host
-
-!! -----
-
-! DEVICE-LABEL: func.func @_QPimplicitly_captured_with_dev_type_recursive
-! DEVICE-SAME: {{.*}}attributes {omp.declare_target = #omp.declaretarget<device_type = (any), capture_clause = (enter)>{{.*}}}
-recursive function implicitly_captured_with_dev_type_recursive(increment) result(k)
-!$omp declare target enter(implicitly_captured_with_dev_type_recursive) device_type(host)
-   integer :: increment, k
-   if (increment == 10) then
-      k = increment
-   else
-      k = implicitly_captured_with_dev_type_recursive(increment + 1)
-   end if
-end function implicitly_captured_with_dev_type_recursive
-
-! DEVICE-LABEL: func.func @_QPtarget_function_with_dev_type_recurse
-! DEVICE-SAME: {{.*}}attributes {omp.declare_target = #omp.declaretarget<device_type = (nohost), capture_clause = (enter)>{{.*}}}
-function target_function_with_dev_type_recurse() result(i)
-!$omp declare target enter(target_function_with_dev_type_recurse) device_type(nohost)
-   integer :: i
-   i = implicitly_captured_with_dev_type_recursive(0)
-end function target_function_with_dev_type_recurse
-
-!! -----
-
-module test_module
-contains
-! CHECK-LABEL: func.func @_QMtest_modulePimplicitly_captured_nest_twice
-! CHECK-SAME: {{.*}}attributes {omp.declare_target = #omp.declaretarget<device_type = (any), capture_clause = (enter)>{{.*}}}
-   function implicitly_captured_nest_twice() result(i)
-      integer :: i
-      i = 10
-   end function implicitly_captured_nest_twice
-
-! CHECK-LABEL: func.func @_QMtest_modulePimplicitly_captured_one_twice
-! CHECK-SAME: {{.*}}attributes {omp.declare_target = #omp.declaretarget<device_type = (any), capture_clause = (enter)>{{.*}}}
-   function implicitly_captured_one_twice() result(k)
-      !$omp declare target enter(implicitly_captured_one_twice) device_type(host)
-      k = implicitly_captured_nest_twice()
-   end function implicitly_captured_one_twice
-
-! DEVICE-LABEL: func.func @_QMtest_modulePimplicitly_captured_two_twice
-! DEVICE-SAME: {{.*}}attributes {omp.declare_target = #omp.declaretarget<device_type = (nohost), capture_clause = (enter)>{{.*}}}
-   function implicitly_captured_two_twice() result(y)
-      integer :: y
-      y = 5
-   end function implicitly_captured_two_twice
-
-! DEVICE-LABEL: func.func @_QMtest_modulePtarget_function_test_device
-! DEVICE-SAME: {{.*}}attributes {omp.declare_target = #omp.declaretarget<device_type = (nohost), capture_clause = (enter)>{{.*}}}
-   function target_function_test_device() result(j)
-      !$omp declare target enter(target_function_test_device) device_type(nohost)
-      integer :: i, j
-      i = implicitly_captured_one_twice()
-      j = implicitly_captured_two_twice() + i
-   end function target_function_test_device
-end module test_module
-
-!! -----
-
-program mb
-   interface
-      subroutine caller_recursive
-         !$omp declare target enter(caller_recursive) device_type(nohost)
-      end subroutine
-
-      recursive subroutine implicitly_captured_recursive(increment)
-         integer :: increment
-      end subroutine
-   end interface
-end program
-
-! DEVICE-LABEL: func.func @_QPimplicitly_captured_recursive
-! DEVICE-SAME: {{.*}}attributes {omp.declare_target = #omp.declaretarget<device_type = (nohost), capture_clause = (enter)>{{.*}}}
-recursive subroutine implicitly_captured_recursive(increment)
-   integer :: increment
-   if (increment == 10) then
-      return
-   else
-      call implicitly_captured_recursive(increment + 1)
-   end if
-end subroutine
-
-! DEVICE-LABEL: func.func @_QPcaller_recursive
-! DEVICE-SAME: {{.*}}attributes {omp.declare_target = #omp.declaretarget<device_type = (nohost), capture_clause = (enter)>{{.*}}}
-subroutine caller_recursive
-!$omp declare target enter(caller_recursive) device_type(nohost)
-   call implicitly_captured_recursive(0)
-end subroutine
diff --git a/flang/test/Lower/OpenMP/FIR/declare-target-implicit-func-and-subr-cap.f90 b/flang/test/Lower/OpenMP/FIR/declare-target-implicit-func-and-subr-cap.f90
deleted file mode 100644
index a90b04246e6d..000000000000
--- a/flang/test/Lower/OpenMP/FIR/declare-target-implicit-func-and-subr-cap.f90
+++ /dev/null
@@ -1,218 +0,0 @@
-!RUN: %flang_fc1 -emit-fir -fopenmp %s -o - | FileCheck %s
-!RUN: %flang_fc1 -emit-fir -fopenmp -fopenmp-is-target-device %s -o - | FileCheck %s  --check-prefix=DEVICE
-!RUN: bbc -emit-fir -fopenmp %s -o - | FileCheck %s
-!RUN: bbc -emit-fir -fopenmp -fopenmp-is-target-device %s -o - | FileCheck %s --check-prefix=DEVICE
-
-! CHECK-LABEL: func.func @_QPimplicitly_captured
-! CHECK-SAME: {{.*}}attributes {omp.declare_target = #omp.declaretarget<device_type = (any), capture_clause = (to)>{{.*}}}
-function implicitly_captured(toggle) result(k)
-   integer :: i, j, k
-   logical :: toggle
-   i = 10
-   j = 5
-   if (toggle) then
-      k = i
-   else
-      k = j
-   end if
-end function implicitly_captured
-
-
-! CHECK-LABEL: func.func @_QPtarget_function
-! CHECK-SAME: {{.*}}attributes {omp.declare_target = #omp.declaretarget<device_type = (any), capture_clause = (to)>{{.*}}}
-function target_function(toggle) result(i)
-!$omp declare target
-   integer :: i
-   logical :: toggle
-   i = implicitly_captured(toggle)
-end function target_function
-
-!! -----
-
-! CHECK-LABEL: func.func @_QPimplicitly_captured_twice
-! CHECK-SAME: {{.*}}attributes {omp.declare_target = #omp.declaretarget<device_type = (any), capture_clause = (to)>{{.*}}}
-function implicitly_captured_twice() result(k)
-   integer :: i
-   i = 10
-   k = i
-end function implicitly_captured_twice
-
-! CHECK-LABEL: func.func @_QPtarget_function_twice_host
-! CHECK-SAME: {{.*}}attributes {omp.declare_target = #omp.declaretarget<device_type = (host), capture_clause = (to)>{{.*}}}
-function target_function_twice_host() result(i)
-!$omp declare target to(target_function_twice_host) device_type(host)
-   integer :: i
-   i = implicitly_captured_twice()
-end function target_function_twice_host
-
-! DEVICE-LABEL: func.func @_QPtarget_function_twice_device
-! DEVICE-SAME: {{.*}}attributes {omp.declare_target = #omp.declaretarget<device_type = (nohost), capture_clause = (to)>{{.*}}}
-function target_function_twice_device() result(i)
-!$omp declare target to(target_function_twice_device) device_type(nohost)
-   integer :: i
-   i = implicitly_captured_twice()
-end function target_function_twice_device
-
-!! -----
-
-! DEVICE-LABEL: func.func @_QPimplicitly_captured_nest
-! DEVICE-SAME: {{.*}}attributes {omp.declare_target = #omp.declaretarget<device_type = (nohost), capture_clause = (to)>{{.*}}}
-function implicitly_captured_nest() result(k)
-   integer :: i
-   i = 10
-   k = i
-end function implicitly_captured_nest
-
-! DEVICE-LABEL: func.func @_QPimplicitly_captured_one
-! DEVICE-SAME: {{.*}}attributes {omp.declare_target = #omp.declaretarget<device_type = (nohost), capture_clause = (to){{.*}}}
-function implicitly_captured_one() result(k)
-   k = implicitly_captured_nest()
-end function implicitly_captured_one
-
-! DEVICE-LABEL: func.func @_QPimplicitly_captured_two
-! DEVICE-SAME: {{.*}}attributes {omp.declare_target = #omp.declaretarget<device_type = (nohost), capture_clause = (to)>{{.*}}}
-function implicitly_captured_two() result(k)
-   integer :: i
-   i = 10
-   k = i
-end function implicitly_captured_two
-
-! DEVICE-LABEL: func.func @_QPtarget_function_test
-! DEVICE-SAME: {{.*}}attributes {omp.declare_target = #omp.declaretarget<device_type = (nohost), capture_clause = (to)>{{.*}}}
-function target_function_test() result(j)
-!$omp declare target to(target_function_test) device_type(nohost)
-   integer :: i, j
-   i = implicitly_captured_one()
-   j = implicitly_captured_two() + i
-end function target_function_test
-
-!! -----
-
-! CHECK-LABEL: func.func @_QPimplicitly_captured_nest_twice
-! CHECK-SAME: {{.*}}attributes {omp.declare_target = #omp.declaretarget<device_type = (any), capture_clause = (to)>{{.*}}}
-function implicitly_captured_nest_twice() result(k)
-   integer :: i
-   i = 10
-   k = i
-end function implicitly_captured_nest_twice
-
-! CHECK-LABEL: func.func @_QPimplicitly_captured_one_twice
-! CHECK-SAME: {{.*}}attributes {omp.declare_target = #omp.declaretarget<device_type = (any), capture_clause = (to)>{{.*}}}
-function implicitly_captured_one_twice() result(k)
-   k = implicitly_captured_nest_twice()
-end function implicitly_captured_one_twice
-
-! CHECK-LABEL: func.func @_QPimplicitly_captured_two_twice
-! CHECK-SAME: {{.*}}attributes {omp.declare_target = #omp.declaretarget<device_type = (any), capture_clause = (to)>{{.*}}}
-function implicitly_captured_two_twice() result(k)
-   integer :: i
-   i = 10
-   k = i
-end function implicitly_captured_two_twice
-
-! DEVICE-LABEL: func.func @_QPtarget_function_test_device
-! DEVICE-SAME: {{.*}}attributes {omp.declare_target = #omp.declaretarget<device_type = (nohost), capture_clause = (to)>{{.*}}}
-function target_function_test_device() result(j)
-   !$omp declare target to(target_function_test_device) device_type(nohost)
-   integer :: i, j
-   i = implicitly_captured_one_twice()
-   j = implicitly_captured_two_twice() + i
-end function target_function_test_device
-
-! CHECK-LABEL: func.func @_QPtarget_function_test_host
-! CHECK-SAME: {{.*}}attributes {omp.declare_target = #omp.declaretarget<device_type = (host), capture_clause = (to)>{{.*}}}
-function target_function_test_host() result(j)
-   !$omp declare target to(target_function_test_host) device_type(host)
-   integer :: i, j
-   i = implicitly_captured_one_twice()
-   j = implicitly_captured_two_twice() + i
-end function target_function_test_host
-
-!! -----
-
-! DEVICE-LABEL: func.func @_QPimplicitly_captured_with_dev_type_recursive
-! DEVICE-SAME: {{.*}}attributes {omp.declare_target = #omp.declaretarget<device_type = (any), capture_clause = (to)>{{.*}}}
-recursive function implicitly_captured_with_dev_type_recursive(increment) result(k)
-!$omp declare target to(implicitly_captured_with_dev_type_recursive) device_type(host)
-   integer :: increment, k
-   if (increment == 10) then
-      k = increment
-   else
-      k = implicitly_captured_with_dev_type_recursive(increment + 1)
-   end if
-end function implicitly_captured_with_dev_type_recursive
-
-! DEVICE-LABEL: func.func @_QPtarget_function_with_dev_type_recurse
-! DEVICE-SAME: {{.*}}attributes {omp.declare_target = #omp.declaretarget<device_type = (nohost), capture_clause = (to)>{{.*}}}
-function target_function_with_dev_type_recurse() result(i)
-!$omp declare target to(target_function_with_dev_type_recurse) device_type(nohost)
-   integer :: i
-   i = implicitly_captured_with_dev_type_recursive(0)
-end function target_function_with_dev_type_recurse
-
-!! -----
-
-module test_module
-contains
-! CHECK-LABEL: func.func @_QMtest_modulePimplicitly_captured_nest_twice
-! CHECK-SAME: {{.*}}attributes {omp.declare_target = #omp.declaretarget<device_type = (any), capture_clause = (to)>{{.*}}}
-   function implicitly_captured_nest_twice() result(i)
-      integer :: i
-      i = 10
-   end function implicitly_captured_nest_twice
-
-! CHECK-LABEL: func.func @_QMtest_modulePimplicitly_captured_one_twice
-! CHECK-SAME: {{.*}}attributes {omp.declare_target = #omp.declaretarget<device_type = (any), capture_clause = (to)>{{.*}}}
-   function implicitly_captured_one_twice() result(k)
-      !$omp declare target to(implicitly_captured_one_twice) device_type(host)
-      k = implicitly_captured_nest_twice()
-   end function implicitly_captured_one_twice
-
-! DEVICE-LABEL: func.func @_QMtest_modulePimplicitly_captured_two_twice
-! DEVICE-SAME: {{.*}}attributes {omp.declare_target = #omp.declaretarget<device_type = (nohost), capture_clause = (to)>{{.*}}}
-   function implicitly_captured_two_twice() result(y)
-      integer :: y
-      y = 5
-   end function implicitly_captured_two_twice
-
-! DEVICE-LABEL: func.func @_QMtest_modulePtarget_function_test_device
-! DEVICE-SAME: {{.*}}attributes {omp.declare_target = #omp.declaretarget<device_type = (nohost), capture_clause = (to)>{{.*}}}
-   function target_function_test_device() result(j)
-      !$omp declare target to(target_function_test_device) device_type(nohost)
-      integer :: i, j
-      i = implicitly_captured_one_twice()
-      j = implicitly_captured_two_twice() + i
-   end function target_function_test_device
-end module test_module
-
-!! -----
-
-program mb
-   interface
-      subroutine caller_recursive
-         !$omp declare target to(caller_recursive) device_type(nohost)
-      end subroutine
-
-      recursive subroutine implicitly_captured_recursive(increment)
-         integer :: increment
-      end subroutine
-   end interface
-end program
-
-! DEVICE-LABEL: func.func @_QPimplicitly_captured_recursive
-! DEVICE-SAME: {{.*}}attributes {omp.declare_target = #omp.declaretarget<device_type = (nohost), capture_clause = (to)>{{.*}}}
-recursive subroutine implicitly_captured_recursive(increment)
-   integer :: increment
-   if (increment == 10) then
-      return
-   else
-      call implicitly_captured_recursive(increment + 1)
-   end if
-end subroutine
-
-! DEVICE-LABEL: func.func @_QPcaller_recursive
-! DEVICE-SAME: {{.*}}attributes {omp.declare_target = #omp.declaretarget<device_type = (nohost), capture_clause = (to)>{{.*}}}
-subroutine caller_recursive
-!$omp declare target to(caller_recursive) device_type(nohost)
-   call implicitly_captured_recursive(0)
-end subroutine
diff --git a/flang/test/Lower/OpenMP/FIR/default-clause.f90 b/flang/test/Lower/OpenMP/FIR/default-clause.f90
deleted file mode 100644
index 14c0d375896a..000000000000
--- a/flang/test/Lower/OpenMP/FIR/default-clause.f90
+++ /dev/null
@@ -1,281 +0,0 @@
-! This test checks lowering of OpenMP parallel directive
-! with `DEFAULT` clause present.
-
-! RUN: %flang_fc1 -emit-fir -flang-deprecated-no-hlfir -fopenmp %s -o - | FileCheck %s
-! RUN: bbc -fopenmp -emit-fir -hlfir=false %s -o - | FileCheck %s
-
-
-!CHECK: func @_QQmain() attributes {fir.bindc_name = "default_clause_lowering"} {
-!CHECK: %[[W:.*]] = fir.alloca i32 {bindc_name = "w", uniq_name = "_QFEw"}
-!CHECK: %[[X:.*]] = fir.alloca i32 {bindc_name = "x", uniq_name = "_QFEx"}
-!CHECK: %[[Y:.*]] = fir.alloca i32 {bindc_name = "y", uniq_name = "_QFEy"}
-!CHECK: %[[Z:.*]] = fir.alloca i32 {bindc_name = "z", uniq_name = "_QFEz"}
-!CHECK: omp.parallel {
-!CHECK: %[[PRIVATE_X:.*]] = fir.alloca i32 {bindc_name = "x", pinned, uniq_name = "_QFEx"}
-!CHECK: %[[const:.*]] = fir.load %[[X]] : !fir.ref<i32>
-!CHECK: fir.store %[[const]] to %[[PRIVATE_X]] : !fir.ref<i32>
-!CHECK: %[[PRIVATE_Y:.*]] = fir.alloca i32 {bindc_name = "y", pinned, uniq_name = "_QFEy"}
-!CHECK: %[[PRIVATE_W:.*]] = fir.alloca i32 {bindc_name = "w", pinned, uniq_name = "_QFEw"}
-!CHECK: %[[const:.*]] = arith.constant 2 : i32
-!CHECK: %[[temp:.*]] = fir.load %[[PRIVATE_Y]] : !fir.ref<i32>
-!CHECK: %[[result:.*]] = arith.muli %[[const]], %[[temp]] : i32
-!CHECK: fir.store %[[result]] to %[[PRIVATE_X]] : !fir.ref<i32>
-!CHECK: %[[temp:.*]] = fir.load %[[PRIVATE_W]] : !fir.ref<i32>
-!CHECK: %[[const:.*]] = arith.constant 45 : i32
-!CHECK: %[[result:.*]] = arith.addi %[[temp]], %[[const]] : i32
-!CHECK: fir.store %[[result]] to %[[Z]] : !fir.ref<i32>
-!CHECK: omp.terminator
-!CHECK: }
-
-program default_clause_lowering
-    integer :: x, y, z, w
-
-    !$omp parallel default(private) firstprivate(x) shared(z)
-        x = y * 2
-        z = w + 45
-    !$omp end parallel
-
-!CHECK: omp.parallel {
-!CHECK: %[[temp:.*]] = fir.load %[[Y]] : !fir.ref<i32>
-!CHECK: fir.store %[[temp]] to %[[X]] : !fir.ref<i32>
-!CHECK: omp.terminator
-!CHECK: }
-
-    !$omp parallel default(shared)
-        x = y
-    !$omp end parallel
-
-!CHECK: omp.parallel {
-!CHECK: %[[PRIVATE_X:.*]] = fir.alloca i32 {bindc_name = "x", pinned, uniq_name = "_QFEx"}
-!CHECK: %[[PRIVATE_Y:.*]] = fir.alloca i32 {bindc_name = "y", pinned, uniq_name = "_QFEy"}
-!CHECK: %[[temp:.*]] = fir.load %[[PRIVATE_Y]] : !fir.ref<i32>
-!CHECK: fir.store %[[temp]] to %[[PRIVATE_X]] : !fir.ref<i32>
-!CHECK: omp.terminator
-!CHECK: }
-
-    !$omp parallel default(none) private(x, y)
-        x = y
-    !$omp end parallel
-
-!CHECK: omp.parallel   {
-!CHECK: %[[PRIVATE_Y:.*]] = fir.alloca i32 {bindc_name = "y", pinned, uniq_name = "_QFEy"}
-!CHECK: %[[temp:.*]] = fir.load %[[Y]] : !fir.ref<i32>
-!CHECK: fir.store %[[temp]] to %[[PRIVATE_Y]] : !fir.ref<i32>
-!CHECK: %[[PRIVATE_X:.*]] = fir.alloca i32 {bindc_name = "x", pinned, uniq_name = "_QFEx"}
-!CHECK: %[[temp:.*]] = fir.load %[[X]] : !fir.ref<i32>
-!CHECK: fir.store %[[temp]] to %[[PRIVATE_X]] : !fir.ref<i32>
-!CHECK: %[[temp:.*]] = fir.load %[[PRIVATE_Y]] : !fir.ref<i32>
-!CHECK: fir.store %[[temp]] to %[[PRIVATE_X]] : !fir.ref<i32>
-!CHECK: omp.terminator
-!CHECK: }
-
-    !$omp parallel default(firstprivate) firstprivate(y)
-        x = y
-    !$omp end parallel
-
-!CHECK: omp.parallel {
-!CHECK: %[[PRIVATE_X:.*]] = fir.alloca i32 {bindc_name = "x", pinned, uniq_name = "_QFEx"}
-!CHECK: %[[PRIVATE_Y:.*]] = fir.alloca i32 {bindc_name = "y", pinned, uniq_name = "_QFEy"}
-!CHECK: %[[temp:.*]] = fir.load %[[Y]] : !fir.ref<i32>
-!CHECK: fir.store %[[temp]] to %[[PRIVATE_Y]] : !fir.ref<i32>
-!CHECK: %[[PRIVATE_W:.*]] = fir.alloca i32 {bindc_name = "w", pinned, uniq_name = "_QFEw"}
-!CHECK: %[[temp:.*]] = fir.load %[[W]] : !fir.ref<i32>
-!CHECK: fir.store %[[temp]] to %[[PRIVATE_W]] : !fir.ref<i32>
-!CHECK: %[[const:.*]] = arith.constant 2 : i32
-!CHECK: %[[temp:.*]] = fir.load %[[PRIVATE_Y]] : !fir.ref<i32>
-!CHECK: %[[result:.*]] = arith.muli %[[const]], %[[temp]] : i32
-!CHECK: fir.store %[[result]] to %[[PRIVATE_X]] : !fir.ref<i32>
-!CHECK: %[[temp:.*]] = fir.load %[[PRIVATE_W]] : !fir.ref<i32>
-!CHECK: %[[const:.*]] = arith.constant 45 : i32
-!CHECK: %[[result:.*]] = arith.addi %[[temp]], %[[const]] : i32
-!CHECK: fir.store %[[result]] to %[[Z]] : !fir.ref<i32>
-!CHECK: omp.terminator
-!CHECK: }
-
-    !$omp parallel default(firstprivate) private(x) shared(z)
-        x = y * 2
-        z = w + 45
-    !$omp end parallel
-
-!CHECK: omp.parallel   {
-!CHECK: omp.parallel   {
-!CHECK: %[[PRIVATE_X:.*]] = fir.alloca i32 {bindc_name = "x", pinned, uniq_name = "_QFEx"}
-!CHECK: %[[PRIVATE_Y:.*]] = fir.alloca i32 {bindc_name = "y", pinned, uniq_name = "_QFEy"}
-!CHECK: %[[temp:.*]] = fir.load %[[PRIVATE_Y]] : !fir.ref<i32>
-!CHECK: fir.store %[[temp]] to %[[PRIVATE_X]] : !fir.ref<i32>
-!CHECK: omp.terminator
-!CHECK: }
-!CHECK: omp.parallel {
-!CHECK: %[[PRIVATE_W:.*]] = fir.alloca i32 {bindc_name = "w", pinned, uniq_name = "_QFEw"}
-!CHECK: %[[temp:.*]] = fir.load %[[W]] : !fir.ref<i32>
-!CHECK: fir.store %[[temp]] to %[[PRIVATE_W]] : !fir.ref<i32>
-!CHECK: %[[PRIVATE_X:.*]] = fir.alloca i32 {bindc_name = "x", pinned, uniq_name = "_QFEx"}
-!CHECK: %[[temp:.*]] = fir.load %[[X]] : !fir.ref<i32>
-!CHECK: fir.store %[[temp]] to %[[PRIVATE_X]] : !fir.ref<i32>
-!CHECK: %[[temp:.*]] = fir.load %[[PRIVATE_X]] : !fir.ref<i32>
-!CHECK: fir.store %[[temp]] to %[[PRIVATE_W]] : !fir.ref<i32>
-!CHECK: omp.terminator
-!CHECK: }
-!CHECK: omp.terminator
-!CHECK: }
-    !$omp parallel
-        !$omp parallel default(private)
-            x = y
-        !$omp end parallel
-
-        !$omp parallel default(firstprivate)
-            w = x
-        !$omp end parallel
-    !$omp end parallel
-
-end program default_clause_lowering
-
-subroutine nested_default_clause_tests
-    integer :: x, y, z, w, k, a
-    
-!CHECK: %[[K:.*]] = fir.alloca i32 {bindc_name = "k", uniq_name = "_QFnested_default_clause_testsEk"}
-!CHECK: %[[W:.*]] = fir.alloca i32 {bindc_name = "w", uniq_name = "_QFnested_default_clause_testsEw"}
-!CHECK: %[[X:.*]] = fir.alloca i32 {bindc_name = "x", uniq_name = "_QFnested_default_clause_testsEx"}
-!CHECK: %[[Y:.*]] = fir.alloca i32 {bindc_name = "y", uniq_name = "_QFnested_default_clause_testsEy"}
-!CHECK: %[[Z:.*]] = fir.alloca i32 {bindc_name = "z", uniq_name = "_QFnested_default_clause_testsEz"}
-!CHECK: omp.parallel   {
-!CHECK: %[[PRIVATE_X:.*]] = fir.alloca i32 {bindc_name = "x", pinned, uniq_name = "_QFnested_default_clause_testsEx"}
-!CHECK: %[[temp:.*]] = fir.load %[[X]] : !fir.ref<i32>
-!CHECK: fir.store %[[temp]] to %[[PRIVATE_X]] : !fir.ref<i32>
-!CHECK: %[[PRIVATE_Y:.*]] = fir.alloca i32 {bindc_name = "y", pinned, uniq_name = "_QFnested_default_clause_testsEy"}
-!CHECK: %[[PRIVATE_Z:.*]] = fir.alloca i32 {bindc_name = "z", pinned, uniq_name = "_QFnested_default_clause_testsEz"}
-!CHECK: %[[PRIVATE_K:.*]] = fir.alloca i32 {bindc_name = "k", pinned, uniq_name = "_QFnested_default_clause_testsEk"}
-!CHECK: omp.parallel {
-!CHECK: %[[INNER_PRIVATE_Y:.*]] = fir.alloca i32 {bindc_name = "y", pinned, uniq_name = "_QFnested_default_clause_testsEy"}
-!CHECK: %[[INNER_PRIVATE_X:.*]] = fir.alloca i32 {bindc_name = "x", pinned, uniq_name = "_QFnested_default_clause_testsEx"}
-!CHECK: %[[const:.*]] = arith.constant 20 : i32
-!CHECK: fir.store %[[const]] to %[[INNER_PRIVATE_Y]] : !fir.ref<i32>
-!CHECK: %[[const:.*]] = arith.constant 10 : i32
-!CHECK: fir.store %[[const]] to %[[INNER_PRIVATE_X]] : !fir.ref<i32>
-!CHECK: omp.terminator
-!CHECK: }
-!CHECK: omp.parallel   {
-!CHECK: %[[INNER_PRIVATE_W:.*]] = fir.alloca i32 {bindc_name = "w", pinned, uniq_name = "_QFnested_default_clause_testsEw"}
-!CHECK: %[[INNER_PRIVATE_Z:.*]] = fir.alloca i32 {bindc_name = "z", pinned, uniq_name = "_QFnested_default_clause_testsEz"}
-!CHECK: %[[temp:.*]] = fir.load %[[PRIVATE_Z]]
-!CHECK: fir.store %[[temp]] to %[[INNER_PRIVATE_Z]] : !fir.ref<i32>
-!CHECK: %[[INNER_PRIVATE_K:.*]] = fir.alloca i32 {bindc_name = "k", pinned, uniq_name = "_QFnested_default_clause_testsEk"}
-!CHECK: %[[temp:.*]] = fir.load %[[PRIVATE_K]] : !fir.ref<i32>
-!CHECK: fir.store %[[temp]] to %[[INNER_PRIVATE_K]] : !fir.ref<i32>
-!CHECK: %[[const:.*]] = arith.constant 30 : i32
-!CHECK: fir.store %[[const]] to %[[PRIVATE_Y]] : !fir.ref<i32>
-!CHECK: %[[const:.*]] = arith.constant 40 : i32
-!CHECK: fir.store %[[const]] to %[[INNER_PRIVATE_W]] : !fir.ref<i32>
-!CHECK: %[[const:.*]] = arith.constant 50 : i32
-!CHECK: fir.store %[[const]] to %[[INNER_PRIVATE_Z]] : !fir.ref<i32>
-!CHECK: %[[const:.*]] = arith.constant 40 : i32
-!CHECK: fir.store %[[const]] to %[[INNER_PRIVATE_K]] : !fir.ref<i32>
-!CHECK: omp.terminator
-!CHECK: }
-!CHECK: omp.terminator
-!CHECK: }
-    !$omp parallel  firstprivate(x) private(y) shared(w) default(private)  
-        !$omp parallel default(private)
-           y = 20
-           x = 10 
-        !$omp end parallel 
-
-        !$omp parallel default(firstprivate) shared(y) private(w) 
-            y = 30
-            w = 40 
-            z = 50
-            k = 40
-        !$omp end parallel
-    !$omp end parallel
-    
-    
-!CHECK: omp.parallel {
-!CHECK: %[[PRIVATE_X:.*]] = fir.alloca i32 {bindc_name = "x", pinned, uniq_name = "_QFnested_default_clause_testsEx"}
-!CHECK: %[[PRIVATE_Y:.*]] = fir.alloca i32 {bindc_name = "y", pinned, uniq_name = "_QFnested_default_clause_testsEy"}
-!CHECK: %[[PRIVATE_Z:.*]] = fir.alloca i32 {bindc_name = "z", pinned, uniq_name = "_QFnested_default_clause_testsEz"}
-!CHECK: %[[PRIVATE_W:.*]] = fir.alloca i32 {bindc_name = "w", pinned, uniq_name = "_QFnested_default_clause_testsEw"} 
-!CHECK: omp.parallel {
-!CHECK: %[[PRIVATE_INNER_X:.*]] = fir.alloca i32 {bindc_name = "x", pinned, uniq_name = "_QFnested_default_clause_testsEx"}
-!CHECK: %[[temp:.*]] = fir.load %[[PRIVATE_X]] : !fir.ref<i32>
-!CHECK: fir.store %[[temp]] to %[[PRIVATE_INNER_X]] : !fir.ref<i32>
-!CHECK: %[[INNER_PRIVATE_Y:.*]] = fir.alloca i32 {bindc_name = "y", pinned, uniq_name = "_QFnested_default_clause_testsEy"}
-!CHECK: %[[temp:.*]] = fir.load %[[PRIVATE_Y]] : !fir.ref<i32>
-!CHECK: fir.store %[[temp]] to %[[INNER_PRIVATE_Y]] : !fir.ref<i32>
-!CHECK: %[[temp:.*]] = fir.load %[[INNER_PRIVATE_Y]] : !fir.ref<i32>
-!CHECK: fir.store %[[temp]] to %[[PRIVATE_INNER_X]] : !fir.ref<i32>
-!CHECK: omp.terminator
-!CHECK: }
-!CHECK: omp.parallel {
-!CHECK: %[[PRIVATE_INNER_W:.*]] = fir.alloca i32 {bindc_name = "w", pinned, uniq_name = "_QFnested_default_clause_testsEw"} 
-!CHECK: %[[PRIVATE_INNER_X:.*]] = fir.alloca i32 {bindc_name = "x", pinned, uniq_name = "_QFnested_default_clause_testsEx"}
-!CHECK: %[[temp_1:.*]] = fir.load %[[PRIVATE_INNER_X]] : !fir.ref<i32>
-!CHECK: %[[temp_2:.*]] = fir.load %[[PRIVATE_Z]] : !fir.ref<i32>
-!CHECK: %[[result:.*]] = arith.addi %{{.*}}, %{{.*}} : i32
-!CHECK: fir.store %[[result]] to %[[PRIVATE_INNER_W]] : !fir.ref<i32>
-!CHECK: omp.terminator
-!CHECK: }
-    !$omp parallel default(private)
-        !$omp parallel default(firstprivate)
-            x = y
-        !$omp end parallel
-
-        !$omp parallel default(private) shared(z)
-            w = x + z
-        !$omp end parallel
-    !$omp end parallel    
-    
-!CHECK: omp.parallel {
-!CHECK: %[[PRIVATE_X:.*]] = fir.alloca i32 {bindc_name = "x", pinned, uniq_name = "_QFnested_default_clause_testsEx"}
-!CHECK: %[[PRIVATE_Y:.*]] = fir.alloca i32 {bindc_name = "y", pinned, uniq_name = "_QFnested_default_clause_testsEy"}
-!CHECK: %[[PRIVATE_W:.*]] = fir.alloca i32 {bindc_name = "w", pinned, uniq_name = "_QFnested_default_clause_testsEw"}
-!CHECK: %[[PRIVATE_Z:.*]] = fir.alloca i32 {bindc_name = "z", pinned, uniq_name = "_QFnested_default_clause_testsEz"}
-!CHECK: omp.parallel {
-!CHECK: %[[INNER_PRIVATE_X:.*]] = fir.alloca i32 {bindc_name = "x", pinned, uniq_name = "_QFnested_default_clause_testsEx"}
-!CHECK: %[[temp:.*]] = fir.load %[[PRIVATE_X]] : !fir.ref<i32>
-!CHECK: fir.store %[[temp]] to %[[INNER_PRIVATE_X]] : !fir.ref<i32>
-!CHECK: %[[INNER_PRIVATE_Y:.*]] = fir.alloca i32 {bindc_name = "y", pinned, uniq_name = "_QFnested_default_clause_testsEy"}
-!CHECK: %[[temp:.*]] = fir.load %[[PRIVATE_Y]] : !fir.ref<i32>
-!CHECK: fir.store %[[temp]] to %[[INNER_PRIVATE_Y]] : !fir.ref<i32>
-!CHECK: %[[temp:.*]] = fir.load %[[INNER_PRIVATE_Y]] : !fir.ref<i32>
-!CHECK: fir.store %[[temp]] to %[[INNER_PRIVATE_X]] : !fir.ref<i32>
-!CHECK: omp.terminator
-!CHECK: }
-!CHECK: omp.parallel {
-!CHECK: %[[temp_1:.*]] = fir.load %[[PRIVATE_X]] : !fir.ref<i32>
-!CHECK: %[[temp_2:.*]] = fir.load %[[PRIVATE_Z]] : !fir.ref<i32>
-!CHECK: %[[temp_3:.*]] = arith.addi %[[temp_1]], %[[temp_2]] : i32
-!CHECK: fir.store %[[temp_3]] to %[[PRIVATE_W]] : !fir.ref<i32>
-!CHECK: omp.terminator
-!CHECK: }
-!CHECK: }
-    !$omp parallel default(private)
-		!$omp parallel default(firstprivate)
-			x = y
-		!$omp end parallel
-
-		!$omp parallel default(shared)
-			w = x + z
-		!$omp end parallel
-	!$omp end parallel
-
-!CHECK: omp.parallel {
-!CHECK: %[[PRIVATE_X:.*]] = fir.alloca i32 {bindc_name = "x", pinned, uniq_name = "_QFnested_default_clause_testsEx"}
-!CHECK: %[[temp:.*]] = fir.load %[[X]] : !fir.ref<i32>
-!CHECK: fir.store %[[temp]] to %[[PRIVATE_X]] : !fir.ref<i32>
-!CHECK: %[[PRIVATE_Y:.*]] = fir.alloca i32 {bindc_name = "y", pinned, uniq_name = "_QFnested_default_clause_testsEy"}
-!CHECK: %[[temp:.*]] = fir.load %[[Y]] : !fir.ref<i32>
-!CHECK: fir.store %[[temp]] to %[[PRIVATE_Y]] : !fir.ref<i32>
-!CHECK: omp.single {
-!CHECK: %[[temp:.*]] = fir.load %[[PRIVATE_Y]] : !fir.ref<i32>
-!CHECK: fir.store %[[temp]] to %[[PRIVATE_X]] : !fir.ref<i32>
-!CHECK: omp.terminator
-!CHECK: }
-!CHECK: omp.terminator
-!CHECK: }
-!CHECK: } 
-	!$omp parallel default(firstprivate)
-		!$omp single
-			x = y
-		!$omp end single
-	!$omp end parallel
-end subroutine
diff --git a/flang/test/Lower/OpenMP/FIR/delayed-privatization-firstprivate.f90 b/flang/test/Lower/OpenMP/FIR/delayed-privatization-firstprivate.f90
deleted file mode 100644
index 50938342dee7..000000000000
--- a/flang/test/Lower/OpenMP/FIR/delayed-privatization-firstprivate.f90
+++ /dev/null
@@ -1,32 +0,0 @@
-! Test delayed privatization for the `private` clause.
-
-! RUN: %flang_fc1 -emit-fir -flang-deprecated-no-hlfir -fopenmp -mmlir \
-! RUN:   --openmp-enable-delayed-privatization -o - %s 2>&1 | FileCheck %s
-! RUN: bbc -emit-fir -hlfir=false -fopenmp --openmp-enable-delayed-privatization \
-! RUN:   -o - %s 2>&1 | FileCheck %s
-
-subroutine delayed_privatization_firstprivate
-  implicit none
-  integer :: var1
-
-!$OMP PARALLEL FIRSTPRIVATE(var1)
-  var1 = 10
-!$OMP END PARALLEL
-end subroutine
-
-! CHECK-LABEL: omp.private {type = firstprivate}
-! CHECK-SAME: @[[VAR1_PRIVATIZER_SYM:.*]] : !fir.ref<i32> alloc {
-! CHECK-NEXT: ^bb0(%[[PRIV_ARG:.*]]: !fir.ref<i32>):
-! CHECK-NEXT:   %[[PRIV_ALLOC:.*]] = fir.alloca i32 {bindc_name = "var1", pinned, uniq_name = "_QFdelayed_privatization_firstprivateEvar1"}
-! CHECK-NEXT:   omp.yield(%[[PRIV_ALLOC]] : !fir.ref<i32>)
-! CHECK: } copy {
-! CHECK: ^bb0(%[[PRIV_ORIG_ARG:.*]]: !fir.ref<i32>, %[[PRIV_PRIV_ARG:.*]]: !fir.ref<i32>):
-! CHECK:    %[[ORIG_VAL:.*]] = fir.load %[[PRIV_ORIG_ARG]] : !fir.ref<i32>
-! CHECK:    fir.store %[[ORIG_VAL]] to %[[PRIV_PRIV_ARG]] : !fir.ref<i32>
-! CHECK:    omp.yield(%[[PRIV_PRIV_ARG]] : !fir.ref<i32>)
-! CHECK: }
-
-! CHECK-LABEL: @_QPdelayed_privatization_firstprivate
-! CHECK: omp.parallel private(@[[VAR1_PRIVATIZER_SYM]] %{{.*}} -> %{{.*}} : !fir.ref<i32>) {
-! CHECK: omp.terminator
-
diff --git a/flang/test/Lower/OpenMP/FIR/delayed-privatization-private.f90 b/flang/test/Lower/OpenMP/FIR/delayed-privatization-private.f90
deleted file mode 100644
index b13687faa3f2..000000000000
--- a/flang/test/Lower/OpenMP/FIR/delayed-privatization-private.f90
+++ /dev/null
@@ -1,41 +0,0 @@
-! Test delayed privatization for the `private` clause.
-
-! RUN: %flang_fc1 -emit-fir -flang-deprecated-no-hlfir -fopenmp -mmlir \
-! RUN:   --openmp-enable-delayed-privatization -o - %s 2>&1 | FileCheck %s
-! RUN: bbc -emit-fir -hlfir=false -fopenmp --openmp-enable-delayed-privatization \
-! RUN:   -o - %s 2>&1 | FileCheck %s
-
-subroutine delayed_privatization_private
-  implicit none
-  integer :: var1
-
-!$OMP PARALLEL PRIVATE(var1)
-  var1 = 10
-!$OMP END PARALLEL
-
-!$OMP PARALLEL PRIVATE(var1)
-  var1 = 20
-!$OMP END PARALLEL
-
-end subroutine
-
-! CHECK-LABEL: omp.private {type = private}
-! CHECK-SAME: @[[PRIVATIZER_SYM:.*]] : !fir.ref<i32> alloc {
-! CHECK-NEXT: ^bb0(%[[PRIV_ARG:.*]]: !fir.ref<i32>):
-! CHECK-NEXT:   %[[PRIV_ALLOC:.*]] = fir.alloca i32 {bindc_name = "var1", pinned, uniq_name = "_QFdelayed_privatization_privateEvar1"}
-! CHECK-NEXT:   omp.yield(%[[PRIV_ALLOC]] : !fir.ref<i32>)
-! CHECK-NOT: } copy {
-
-! CHECK-LABEL: @_QPdelayed_privatization_private
-! CHECK: %[[ORIG_ALLOC:.*]] = fir.alloca i32 {bindc_name = "var1", uniq_name = "_QFdelayed_privatization_privateEvar1"}
-! CHECK: omp.parallel private(@[[PRIVATIZER_SYM]] %[[ORIG_ALLOC]] -> %[[PAR_ARG:.*]] : !fir.ref<i32>) {
-! CHECK: %[[C10:.*]] = arith.constant 10 : i32
-! CHECK: fir.store %[[C10]] to %[[PAR_ARG]] : !fir.ref<i32>
-! CHECK: omp.terminator
-
-! Test that the same privatizer is used if the a variable with the same type and
-! name was previously privatized.
-! CHECK: omp.parallel private(@[[PRIVATIZER_SYM]] %[[ORIG_ALLOC]] -> %[[PAR_ARG:.*]] : !fir.ref<i32>) {
-! CHECK: %[[C20:.*]] = arith.constant 20 : i32
-! CHECK: fir.store %[[C20]] to %[[PAR_ARG]] : !fir.ref<i32>
-! CHECK: omp.terminator
diff --git a/flang/test/Lower/OpenMP/FIR/firstprivate-commonblock.f90 b/flang/test/Lower/OpenMP/FIR/firstprivate-commonblock.f90
deleted file mode 100644
index 6adc7d9f6c82..000000000000
--- a/flang/test/Lower/OpenMP/FIR/firstprivate-commonblock.f90
+++ /dev/null
@@ -1,30 +0,0 @@
-! RUN: %flang_fc1 -emit-fir -flang-deprecated-no-hlfir -fopenmp -o - %s 2>&1 | FileCheck %s
-
-!CHECK: func.func @_QPfirstprivate_common() {
-!CHECK: %[[val_0:.*]] = fir.address_of(@c_) : !fir.ref<!fir.array<8xi8>>
-!CHECK: %[[val_1:.*]] = fir.convert %[[val_0]] : (!fir.ref<!fir.array<8xi8>>) -> !fir.ref<!fir.array<?xi8>>
-!CHECK: %[[val_c0:.*]] = arith.constant 0 : index
-!CHECK: %[[val_2:.*]] = fir.coordinate_of %[[val_1]], %[[val_c0]] : (!fir.ref<!fir.array<?xi8>>, index) -> !fir.ref<i8>
-!CHECK: %[[val_3:.*]] = fir.convert %[[val_2]] : (!fir.ref<i8>) -> !fir.ref<f32>
-!CHECK: %[[val_4:.*]] = fir.convert %[[val_0]] : (!fir.ref<!fir.array<8xi8>>) -> !fir.ref<!fir.array<?xi8>>
-!CHECK: %[[val_c4:.*]] = arith.constant 4 : index
-!CHECK: %[[val_5:.*]] = fir.coordinate_of %[[val_4]], %[[val_c4]] : (!fir.ref<!fir.array<?xi8>>, index) -> !fir.ref<i8>
-!CHECK: %[[val_6:.*]] = fir.convert %[[val_5]] : (!fir.ref<i8>) -> !fir.ref<f32>
-!CHECK: omp.parallel {
-!CHECK: %[[val_7:.*]] = fir.alloca f32 {bindc_name = "x", pinned, uniq_name = "_QFfirstprivate_commonEx"}
-!CHECK: %[[val_8:.*]] = fir.load %[[val_3]] : !fir.ref<f32>
-!CHECK: fir.store %[[val_8]] to %[[val_7]] : !fir.ref<f32>
-!CHECK: %[[val_9:.*]] = fir.alloca f32 {bindc_name = "y", pinned, uniq_name = "_QFfirstprivate_commonEy"}
-!CHECK: %[[val_10:.*]] = fir.load %[[val_6]] : !fir.ref<f32>
-!CHECK: fir.store %[[val_10]] to %[[val_9]] : !fir.ref<f32>
-!CHECK: omp.terminator
-!CHECK: }
-!CHECK: return
-!CHECK: }
-
-subroutine firstprivate_common
-  common /c/ x, y
-  real x, y
-  !$omp parallel firstprivate(/c/)
-  !$omp end parallel
-end subroutine
diff --git a/flang/test/Lower/OpenMP/FIR/flush.f90 b/flang/test/Lower/OpenMP/FIR/flush.f90
deleted file mode 100644
index 2868367fbdba..000000000000
--- a/flang/test/Lower/OpenMP/FIR/flush.f90
+++ /dev/null
@@ -1,45 +0,0 @@
-! This test checks lowering of OpenMP Flush Directive.
-
-!RUN: %flang_fc1 -emit-fir -flang-deprecated-no-hlfir -fopenmp %s -o - | FileCheck %s --check-prefixes="FIRDialect,OMPDialect"
-!RUN: %flang_fc1 -emit-fir -flang-deprecated-no-hlfir -fopenmp %s -o - | fir-opt --cfg-conversion-on-func-opt | fir-opt --fir-to-llvm-ir | FileCheck %s --check-prefixes="LLVMIRDialect,OMPDialect"
-
-subroutine flush_standalone(a, b, c)
-    integer, intent(inout) :: a, b, c
-
-!$omp flush(a,b,c)
-!$omp flush
-!OMPDialect: omp.flush(%{{.*}}, %{{.*}}, %{{.*}} :
-!FIRDialect: !fir.ref<i32>, !fir.ref<i32>, !fir.ref<i32>)
-!LLVMIRDialect: !llvm.ptr, !llvm.ptr, !llvm.ptr)
-!OMPDialect: omp.flush
-
-end subroutine flush_standalone
-
-subroutine flush_parallel(a, b, c)
-    integer, intent(inout) :: a, b, c
-
-!$omp parallel
-!OMPDialect:  omp.parallel {
-
-!OMPDialect: omp.flush(%{{.*}}, %{{.*}}, %{{.*}} :
-!FIRDialect: !fir.ref<i32>, !fir.ref<i32>, !fir.ref<i32>)
-!LLVMIRDialect: !llvm.ptr, !llvm.ptr, !llvm.ptr)
-!OMPDialect: omp.flush
-!$omp flush(a,b,c)
-!$omp flush
-
-!FIRDialect: %{{.*}} = fir.load %{{.*}} : !fir.ref<i32>
-!FIRDialect: %{{.*}} = fir.load %{{.*}} : !fir.ref<i32>
-!FIRDialect: %{{.*}} = arith.addi %{{.*}}, %{{.*}} : i32
-!FIRDialect: fir.store %{{.*}} to %{{.*}} : !fir.ref<i32>
-
-!LLVMIRDialect: %{{.*}} = llvm.load %{{.*}} : !llvm.ptr -> i32
-!LLVMIRDialect: %{{.*}} = llvm.load %{{.*}} : !llvm.ptr -> i32
-!LLVMIRDialect: %{{.*}} = llvm.add %{{.*}}, %{{.*}} : i32
-!LLVMIRDialect: llvm.store %{{.*}}, %{{.*}} : i32, !llvm.ptr
-    c = a + b
-
-!OMPDialect: omp.terminator
-!$omp END parallel
-
-end subroutine flush_parallel
diff --git a/flang/test/Lower/OpenMP/FIR/if-clause.f90 b/flang/test/Lower/OpenMP/FIR/if-clause.f90
deleted file mode 100644
index f686b9708fc5..000000000000
--- a/flang/test/Lower/OpenMP/FIR/if-clause.f90
+++ /dev/null
@@ -1,496 +0,0 @@
-! This test checks lowering of OpenMP IF clauses.
-
-! RUN: bbc -fopenmp -emit-fir %s -o - | FileCheck %s
-! RUN: %flang_fc1 -fopenmp -emit-fir %s -o - | FileCheck %s
-
-program main
-  integer :: i
-
-  ! TODO When they are supported, add tests for:
-  ! - DISTRIBUTE PARALLEL DO
-  ! - DISTRIBUTE PARALLEL DO SIMD
-  ! - DISTRIBUTE SIMD
-  ! - PARALLEL SECTIONS
-  ! - PARALLEL WORKSHARE
-  ! - TARGET PARALLEL
-  ! - TARGET TEAMS DISTRIBUTE
-  ! - TARGET TEAMS DISTRIBUTE PARALLEL DO
-  ! - TARGET TEAMS DISTRIBUTE PARALLEL DO SIMD
-  ! - TARGET TEAMS DISTRIBUTE SIMD
-  ! - TARGET UPDATE
-  ! - TASKLOOP
-  ! - TASKLOOP SIMD
-  ! - TEAMS DISTRIBUTE
-  ! - TEAMS DISTRIBUTE PARALLEL DO
-  ! - TEAMS DISTRIBUTE PARALLEL DO SIMD
-  ! - TEAMS DISTRIBUTE SIMD
-
-  ! ----------------------------------------------------------------------------
-  ! DO SIMD
-  ! ----------------------------------------------------------------------------
-  ! CHECK:      omp.wsloop
-  ! CHECK-NOT:  if({{.*}})
-  ! CHECK-SAME: {
-  !$omp do simd
-  do i = 1, 10
-  end do
-  !$omp end do simd
-
-  ! CHECK:      omp.wsloop
-  !$omp do simd if(.true.)
-  do i = 1, 10
-  end do
-  !$omp end do simd
-
-  ! CHECK:      omp.wsloop
-  !$omp do simd if(simd: .true.)
-  do i = 1, 10
-  end do
-  !$omp end do simd
-
-  ! ----------------------------------------------------------------------------
-  ! PARALLEL
-  ! ----------------------------------------------------------------------------
-  ! CHECK:      omp.parallel
-  ! CHECK-NOT:  if({{.*}})
-  ! CHECK-SAME: {
-  !$omp parallel
-  i = 10
-  !$omp end parallel
-
-  ! CHECK:      omp.parallel
-  ! CHECK-SAME: if({{.*}})
-  !$omp parallel if(.true.)
-  i = 10
-  !$omp end parallel
-
-  ! CHECK:      omp.parallel
-  ! CHECK-SAME: if({{.*}})
-  !$omp parallel if(parallel: .true.)
-  i = 10
-  !$omp end parallel
-
-  ! ----------------------------------------------------------------------------
-  ! PARALLEL DO
-  ! ----------------------------------------------------------------------------
-  ! CHECK:      omp.parallel
-  ! CHECK-NOT:  if({{.*}})
-  ! CHECK-SAME: {
-  !$omp parallel do
-  do i = 1, 10
-  end do
-  !$omp end parallel do
-
-  ! CHECK:      omp.parallel
-  ! CHECK-SAME: if({{.*}})
-  !$omp parallel do if(.true.)
-  do i = 1, 10
-  end do
-  !$omp end parallel do
-
-  ! CHECK:      omp.parallel
-  ! CHECK-SAME: if({{.*}})
-  !$omp parallel do if(parallel: .true.)
-  do i = 1, 10
-  end do
-  !$omp end parallel do
-
-  ! ----------------------------------------------------------------------------
-  ! PARALLEL DO SIMD
-  ! ----------------------------------------------------------------------------
-  ! CHECK:      omp.parallel
-  ! CHECK-NOT:  if({{.*}})
-  ! CHECK-SAME: {
-  ! CHECK:      omp.wsloop
-  ! CHECK-NOT:  if({{.*}})
-  ! CHECK-SAME: {
-  !$omp parallel do simd
-  do i = 1, 10
-  end do
-  !$omp end parallel do simd
-
-  ! CHECK:      omp.parallel
-  ! CHECK-SAME: if({{.*}})
-  ! CHECK:      omp.wsloop
-  !$omp parallel do simd if(.true.)
-  do i = 1, 10
-  end do
-  !$omp end parallel do simd
-
-  ! CHECK:      omp.parallel
-  ! CHECK-SAME: if({{.*}})
-  ! CHECK:      omp.wsloop
-  !$omp parallel do simd if(parallel: .true.) if(simd: .false.)
-  do i = 1, 10
-  end do
-  !$omp end parallel do simd
-
-  ! CHECK:      omp.parallel
-  ! CHECK-SAME: if({{.*}})
-  ! CHECK:      omp.wsloop
-  ! CHECK-NOT:  if({{.*}})
-  ! CHECK-SAME: {
-  !$omp parallel do simd if(parallel: .true.)
-  do i = 1, 10
-  end do
-  !$omp end parallel do simd
-
-  ! CHECK:      omp.parallel
-  ! CHECK-NOT:  if({{.*}})
-  ! CHECK-SAME: {
-  ! CHECK:      omp.wsloop
-  !$omp parallel do simd if(simd: .true.)
-  do i = 1, 10
-  end do
-  !$omp end parallel do simd
-
-  ! ----------------------------------------------------------------------------
-  ! SIMD
-  ! ----------------------------------------------------------------------------
-  ! CHECK:      omp.simd
-  ! CHECK-NOT:  if({{.*}})
-  ! CHECK-SAME: {
-  !$omp simd
-  do i = 1, 10
-  end do
-  !$omp end simd
-
-  ! CHECK:      omp.simd
-  ! CHECK-SAME: if({{.*}})
-  !$omp simd if(.true.)
-  do i = 1, 10
-  end do
-  !$omp end simd
-
-  ! CHECK:      omp.simd
-  ! CHECK-SAME: if({{.*}})
-  !$omp simd if(simd: .true.)
-  do i = 1, 10
-  end do
-  !$omp end simd
-
-  ! ----------------------------------------------------------------------------
-  ! TARGET
-  ! ----------------------------------------------------------------------------
-  ! CHECK:      omp.target
-  ! CHECK-NOT:  if({{.*}})
-  ! CHECK-SAME: {
-  !$omp target
-  !$omp end target
-
-  ! CHECK:      omp.target
-  ! CHECK-SAME: if({{.*}})
-  !$omp target if(.true.)
-  !$omp end target
-
-  ! CHECK:      omp.target
-  ! CHECK-SAME: if({{.*}})
-  !$omp target if(target: .true.)
-  !$omp end target
-
-  ! ----------------------------------------------------------------------------
-  ! TARGET DATA
-  ! ----------------------------------------------------------------------------
-  ! CHECK:      omp.target_data
-  ! CHECK-NOT:  if({{.*}})
-  ! CHECK-SAME: {
-  !$omp target data map(tofrom: i)
-  !$omp end target data
-
-  ! CHECK:      omp.target_data
-  ! CHECK-SAME: if({{.*}})
-  !$omp target data map(tofrom: i) if(.true.)
-  !$omp end target data
-
-  ! CHECK:      omp.target_data
-  ! CHECK-SAME: if({{.*}})
-  !$omp target data map(tofrom: i) if(target data: .true.)
-  !$omp end target data
-
-  ! ----------------------------------------------------------------------------
-  ! TARGET ENTER DATA
-  ! ----------------------------------------------------------------------------
-  ! CHECK:      omp.target_enter_data
-  ! CHECK-NOT:  if({{.*}})
-  ! CHECK-SAME: map
-  !$omp target enter data map(to: i)
-
-  ! CHECK:      omp.target_enter_data
-  ! CHECK-SAME: if({{.*}})
-  !$omp target enter data map(to: i) if(.true.)
-
-  ! CHECK:      omp.target_enter_data
-  ! CHECK-SAME: if({{.*}})
-  !$omp target enter data map(to: i) if(target enter data: .true.)
-
-  ! ----------------------------------------------------------------------------
-  ! TARGET EXIT DATA
-  ! ----------------------------------------------------------------------------
-  ! CHECK:      omp.target_exit_data
-  ! CHECK-NOT:  if({{.*}})
-  ! CHECK-SAME: map
-  !$omp target exit data map(from: i)
-
-  ! CHECK:      omp.target_exit_data
-  ! CHECK-SAME: if({{.*}})
-  !$omp target exit data map(from: i) if(.true.)
-
-  ! CHECK:      omp.target_exit_data
-  ! CHECK-SAME: if({{.*}})
-  !$omp target exit data map(from: i) if(target exit data: .true.)
-
-  ! ----------------------------------------------------------------------------
-  ! TARGET PARALLEL DO
-  ! ----------------------------------------------------------------------------
-  ! CHECK:      omp.target
-  ! CHECK-NOT:  if({{.*}})
-  ! CHECK-SAME: {
-  ! CHECK:      omp.parallel
-  ! CHECK-NOT:  if({{.*}})
-  ! CHECK-SAME: {
-  !$omp target parallel do
-  do i = 1, 10
-  end do
-  !$omp end target parallel do
-
-  ! CHECK:      omp.target
-  ! CHECK-SAME: if({{.*}})
-  ! CHECK:      omp.parallel
-  ! CHECK-SAME: if({{.*}})
-  !$omp target parallel do if(.true.)
-  do i = 1, 10
-  end do
-  !$omp end target parallel do
-
-  ! CHECK:      omp.target
-  ! CHECK-SAME: if({{.*}})
-  ! CHECK:      omp.parallel
-  ! CHECK-SAME: if({{.*}})
-  !$omp target parallel do if(target: .true.) if(parallel: .false.)
-  do i = 1, 10
-  end do
-  !$omp end target parallel do
-
-  ! CHECK:      omp.target
-  ! CHECK-SAME: if({{.*}})
-  ! CHECK:      omp.parallel
-  ! CHECK-NOT:  if({{.*}})
-  ! CHECK-SAME: {
-  !$omp target parallel do if(target: .true.)
-  do i = 1, 10
-  end do
-  !$omp end target parallel do
-
-  ! CHECK:      omp.target
-  ! CHECK-NOT:  if({{.*}})
-  ! CHECK-SAME: {
-  ! CHECK:      omp.parallel
-  ! CHECK-SAME: if({{.*}})
-  !$omp target parallel do if(parallel: .true.)
-  do i = 1, 10
-  end do
-  !$omp end target parallel do
-
-  ! ----------------------------------------------------------------------------
-  ! TARGET PARALLEL DO SIMD
-  ! ----------------------------------------------------------------------------
-  ! CHECK:      omp.target
-  ! CHECK-NOT:  if({{.*}})
-  ! CHECK-SAME: {
-  ! CHECK:      omp.parallel
-  ! CHECK-NOT:  if({{.*}})
-  ! CHECK-SAME: {
-  ! CHECK:      omp.wsloop
-  ! CHECK-NOT:  if({{.*}})
-  ! CHECK-SAME: {
-  !$omp target parallel do simd
-  do i = 1, 10
-  end do
-  !$omp end target parallel do simd
-
-  ! CHECK:      omp.target
-  ! CHECK-SAME: if({{.*}})
-  ! CHECK:      omp.parallel
-  ! CHECK-SAME: if({{.*}})
-  ! CHECK:      omp.wsloop
-  !$omp target parallel do simd if(.true.)
-  do i = 1, 10
-  end do
-  !$omp end target parallel do simd
-
-  ! CHECK:      omp.target
-  ! CHECK-SAME: if({{.*}})
-  ! CHECK:      omp.parallel
-  ! CHECK-SAME: if({{.*}})
-  ! CHECK:      omp.wsloop
-  !$omp target parallel do simd if(target: .true.) if(parallel: .false.) &
-  !$omp&                        if(simd: .true.)
-  do i = 1, 10
-  end do
-  !$omp end target parallel do simd
-
-  ! CHECK:      omp.target
-  ! CHECK-SAME: if({{.*}})
-  ! CHECK:      omp.parallel
-  ! CHECK-NOT:  if({{.*}})
-  ! CHECK-SAME: {
-  ! CHECK:      omp.wsloop
-  ! CHECK-NOT:  if({{.*}})
-  ! CHECK-SAME: {
-  !$omp target parallel do simd if(target: .true.)
-  do i = 1, 10
-  end do
-  !$omp end target parallel do simd
-
-  ! CHECK:      omp.target
-  ! CHECK-NOT:  if({{.*}})
-  ! CHECK-SAME: {
-  ! CHECK:      omp.parallel
-  ! CHECK-SAME: if({{.*}})
-  ! CHECK:      omp.wsloop
-  !$omp target parallel do simd if(parallel: .true.) if(simd: .false.)
-  do i = 1, 10
-  end do
-  !$omp end target parallel do simd
-
-  ! ----------------------------------------------------------------------------
-  ! TARGET SIMD
-  ! ----------------------------------------------------------------------------
-  ! CHECK:      omp.target
-  ! CHECK-NOT:  if({{.*}})
-  ! CHECK-SAME: {
-  ! CHECK:      omp.simd
-  ! CHECK-NOT:  if({{.*}})
-  ! CHECK-SAME: {
-  !$omp target simd
-  do i = 1, 10
-  end do
-  !$omp end target simd
-
-  ! CHECK:      omp.target
-  ! CHECK-SAME: if({{.*}})
-  ! CHECK:      omp.simd
-  ! CHECK-SAME: if({{.*}})
-  !$omp target simd if(.true.)
-  do i = 1, 10
-  end do
-  !$omp end target simd
-
-  ! CHECK:      omp.target
-  ! CHECK-SAME: if({{.*}})
-  ! CHECK:      omp.simd
-  ! CHECK-SAME: if({{.*}})
-  !$omp target simd if(target: .true.) if(simd: .false.)
-  do i = 1, 10
-  end do
-  !$omp end target simd
-
-  ! CHECK:      omp.target
-  ! CHECK-SAME: if({{.*}})
-  ! CHECK:      omp.simd
-  ! CHECK-NOT:  if({{.*}})
-  ! CHECK-SAME: {
-  !$omp target simd if(target: .true.)
-  do i = 1, 10
-  end do
-  !$omp end target simd
-
-  ! CHECK:      omp.target
-  ! CHECK-NOT:  if({{.*}})
-  ! CHECK-SAME: {
-  ! CHECK:      omp.simd
-  ! CHECK-SAME: if({{.*}})
-  !$omp target simd if(simd: .true.)
-  do i = 1, 10
-  end do
-  !$omp end target simd
-
-  ! ----------------------------------------------------------------------------
-  ! TARGET TEAMS
-  ! ----------------------------------------------------------------------------
-
-  ! CHECK:      omp.target
-  ! CHECK-NOT:  if({{.*}})
-  ! CHECK-SAME: {
-  ! CHECK:      omp.teams
-  ! CHECK-NOT:  if({{.*}})
-  ! CHECK-SAME: {
-  !$omp target teams
-  i = 1
-  !$omp end target teams
-
-  ! CHECK:      omp.target
-  ! CHECK-SAME: if({{.*}})
-  ! CHECK:      omp.teams
-  ! CHECK-SAME: if({{.*}})
-  !$omp target teams if(.true.)
-  i = 1
-  !$omp end target teams
-
-  ! CHECK:      omp.target
-  ! CHECK-SAME: if({{.*}})
-  ! CHECK:      omp.teams
-  ! CHECK-SAME: if({{.*}})
-  !$omp target teams if(target: .true.) if(teams: .false.)
-  i = 1
-  !$omp end target teams
-
-  ! CHECK:      omp.target
-  ! CHECK-SAME: if({{.*}})
-  ! CHECK:      omp.teams
-  ! CHECK-NOT:  if({{.*}})
-  ! CHECK-SAME: {
-  !$omp target teams if(target: .true.)
-  i = 1
-  !$omp end target teams
-
-  ! CHECK:      omp.target
-  ! CHECK-NOT:  if({{.*}})
-  ! CHECK-SAME: {
-  ! CHECK:      omp.teams
-  ! CHECK-SAME: if({{.*}})
-  !$omp target teams if(teams: .true.)
-  i = 1
-  !$omp end target teams
-
-  ! ----------------------------------------------------------------------------
-  ! TASK
-  ! ----------------------------------------------------------------------------
-  ! CHECK:      omp.task
-  ! CHECK-NOT:  if({{.*}})
-  ! CHECK-SAME: {
-  !$omp task
-  !$omp end task
-
-  ! CHECK:      omp.task
-  ! CHECK-SAME: if({{.*}})
-  !$omp task if(.true.)
-  !$omp end task
-
-  ! CHECK:      omp.task
-  ! CHECK-SAME: if({{.*}})
-  !$omp task if(task: .true.)
-  !$omp end task
-
-  ! ----------------------------------------------------------------------------
-  ! TEAMS
-  ! ----------------------------------------------------------------------------
-  ! CHECK:      omp.teams
-  ! CHECK-NOT:  if({{.*}})
-  ! CHECK-SAME: {
-  !$omp teams
-  i = 1
-  !$omp end teams
-
-  ! CHECK:      omp.teams
-  ! CHECK-SAME: if({{.*}})
-  !$omp teams if(.true.)
-  i = 1
-  !$omp end teams
-
-  ! CHECK:      omp.teams
-  ! CHECK-SAME: if({{.*}})
-  !$omp teams if(teams: .true.)
-  i = 1
-  !$omp end teams
-end program main
diff --git a/flang/test/Lower/OpenMP/FIR/is-device.f90 b/flang/test/Lower/OpenMP/FIR/is-device.f90
deleted file mode 100644
index 79e0ee506c5f..000000000000
--- a/flang/test/Lower/OpenMP/FIR/is-device.f90
+++ /dev/null
@@ -1,14 +0,0 @@
-!RUN: %flang_fc1 -emit-fir -fopenmp -fopenmp-is-target-device %s -o - | FileCheck %s --check-prefix=DEVICE
-!RUN: %flang_fc1 -emit-fir -fopenmp %s -o - | FileCheck %s --check-prefix=HOST
-!RUN: %flang_fc1 -emit-fir -fopenmp-is-target-device %s -o - | FileCheck %s --check-prefix=DEVICE-FLAG-ONLY
-!RUN: bbc -fopenmp -fopenmp-is-target-device -emit-fir -o - %s | FileCheck %s --check-prefix=DEVICE
-!RUN: bbc -fopenmp -emit-fir -o - %s | FileCheck %s --check-prefix=HOST
-!RUN: bbc -fopenmp-is-target-device -emit-fir -o - %s | FileCheck %s --check-prefix=DEVICE-FLAG-ONLY
-
-!DEVICE: module attributes {{{.*}}, omp.is_target_device = true{{.*}}}
-!HOST: module attributes {{{.*}}, omp.is_target_device = false{{.*}}}
-!DEVICE-FLAG-ONLY: module attributes {{{.*}}"
-!DEVICE-FLAG-ONLY-NOT: , omp.is_target_device = {{.*}}
-!DEVICE-FLAG-ONLY-SAME: }
-subroutine omp_subroutine()
-end subroutine omp_subroutine
diff --git a/flang/test/Lower/OpenMP/FIR/lastprivate-commonblock.f90 b/flang/test/Lower/OpenMP/FIR/lastprivate-commonblock.f90
deleted file mode 100644
index 389bcba35f77..000000000000
--- a/flang/test/Lower/OpenMP/FIR/lastprivate-commonblock.f90
+++ /dev/null
@@ -1,46 +0,0 @@
-! RUN: %flang_fc1 -emit-fir -flang-deprecated-no-hlfir -fopenmp -o - %s 2>&1 | FileCheck %s 
-
-!CHECK: func.func @_QPlastprivate_common() {
-!CHECK: %[[val_0:.*]] = fir.alloca i32 {adapt.valuebyref, pinned}
-!CHECK: %[[val_1:.*]] = fir.alloca i32 {bindc_name = "i", uniq_name = "_QFlastprivate_commonEi"}
-!CHECK: %[[val_2:.*]] = fir.address_of(@c_) : !fir.ref<!fir.array<8xi8>>
-!CHECK: %[[val_3:.*]] = fir.convert %[[val_2]] : (!fir.ref<!fir.array<8xi8>>) -> !fir.ref<!fir.array<?xi8>>
-!CHECK: %[[val_c0:.*]] = arith.constant 0 : index
-!CHECK: %[[val_4:.*]] = fir.coordinate_of %[[val_3]], %[[val_c0]] : (!fir.ref<!fir.array<?xi8>>, index) -> !fir.ref<i8>
-!CHECK: %[[val_5:.*]] = fir.convert %[[val_4]] : (!fir.ref<i8>) -> !fir.ref<f32>
-!CHECK: %[[val_6:.*]] = fir.convert %[[val_2]] : (!fir.ref<!fir.array<8xi8>>) -> !fir.ref<!fir.array<?xi8>>
-!CHECK: %[[val_c4:.*]] = arith.constant 4 : index
-!CHECK: %[[val_7:.*]] = fir.coordinate_of %[[val_6]], %[[val_c4]] : (!fir.ref<!fir.array<?xi8>>, index) -> !fir.ref<i8>
-!CHECK: %[[val_8:.*]] = fir.convert %[[val_7]] : (!fir.ref<i8>) -> !fir.ref<f32>
-!CHECK: %[[val_9:.*]] = fir.alloca f32 {bindc_name = "x", pinned, uniq_name = "_QFlastprivate_commonEx"}
-!CHECK: %[[val_10:.*]] = fir.alloca f32 {bindc_name = "y", pinned, uniq_name = "_QFlastprivate_commonEy"}
-!CHECK: %[[val_c1_i32:.*]] = arith.constant 1 : i32
-!CHECK: %[[val_c100_i32:.*]] = arith.constant 100 : i32
-!CHECK: %[[val_c1_i32_0:.*]] = arith.constant 1 : i32
-!CHECK: omp.wsloop   for (%[[arg:.*]]) : i32 = (%[[val_c1_i32]]) to (%[[val_c100_i32]]) inclusive step (%[[val_c1_i32_0]]) {
-!CHECK: fir.store %[[arg]] to %[[val_0]] : !fir.ref<i32>
-!CHECK: %[[val_11:.*]] = arith.addi %[[arg]], %[[val_c1_i32_0]] : i32
-!CHECK: %[[val_c0_i32:.*]] = arith.constant 0 : i32
-!CHECK: %[[val_12:.*]] = arith.cmpi slt, %[[val_c1_i32_0]], %[[val_c0_i32]] : i32
-!CHECK: %[[val_13:.*]] = arith.cmpi slt, %[[val_11]], %[[val_c100_i32]] : i32
-!CHECK: %[[val_14:.*]] = arith.cmpi sgt, %[[val_11]], %[[val_c100_i32]] : i32
-!CHECK: %[[val_15:.*]] = arith.select %[[val_12]], %[[val_13]], %[[val_14]] : i1
-!CHECK: fir.if %[[val_15]] {
-!CHECK: fir.store %[[val_11]] to %[[val_0]] : !fir.ref<i32>
-!CHECK: %[[val_16:.*]] = fir.load %[[val_9]] : !fir.ref<f32>
-!CHECK: fir.store %[[val_16]] to %[[val_5]] : !fir.ref<f32>
-!CHECK: %[[val_17:.*]] = fir.load %[[val_10]] : !fir.ref<f32>
-!CHECK: fir.store %[[val_17]] to %[[val_8]] : !fir.ref<f32>
-!CHECK: }
-!CHECK: omp.yield
-!CHECK: }
-!CHECK: return
-!CHECK: }
-subroutine lastprivate_common
-  common /c/ x, y
-  real x, y
-  !$omp do lastprivate(/c/)
-  do i=1,100
-  end do
-  !$omp end do
-end subroutine
diff --git a/flang/test/Lower/OpenMP/FIR/location.f90 b/flang/test/Lower/OpenMP/FIR/location.f90
deleted file mode 100644
index 648377837670..000000000000
--- a/flang/test/Lower/OpenMP/FIR/location.f90
+++ /dev/null
@@ -1,68 +0,0 @@
-! This test checks location of OpenMP constructs and clauses
-
-!RUN: %flang_fc1 -emit-fir -fopenmp -mmlir --mlir-print-debuginfo %s -o - | FileCheck %s
-
-!CHECK-LABEL: sub_parallel
-subroutine sub_parallel()
-  print *, x
-!CHECK: omp.parallel   {
-  !$omp parallel
-    print *, x
-!CHECK:   omp.terminator loc(#[[PAR_LOC:.*]])
-!CHECK: } loc(#[[PAR_LOC]])
-  !$omp end parallel
-  print *, x
-end
-
-!CHECK-LABEL: sub_target
-subroutine sub_target()
-  print *, x
-!CHECK: omp.target {{.*}} {
-  !$omp target
-    print *, x
-!CHECK:   omp.terminator loc(#[[TAR_LOC:.*]])
-!CHECK: } loc(#[[TAR_LOC]])
-  !$omp end target
-  print *, x
-end
-
-!CHECK-LABEL: sub_loop
-subroutine sub_loop()
-!CHECK: omp.wsloop {{.*}}  {
-  !$omp do
-  do i=1,10
-    print *, i
-!CHECK:   omp.yield loc(#[[LOOP_LOC:.*]])
-!CHECK: } loc(#[[LOOP_LOC]])
-  end do
-  !$omp end do
-end
-
-!CHECK-LABEL: sub_standalone
-subroutine sub_standalone()
-  !CHECK: omp.barrier loc(#[[BAR_LOC:.*]])
-  !$omp barrier
-  !CHECK: omp.taskwait loc(#[[TW_LOC:.*]])
-  !$omp taskwait
-  !CHECK: omp.taskyield loc(#[[TY_LOC:.*]])
-  !$omp taskyield
-end
-
-subroutine sub_if(c)
-  logical(kind=4) :: c
-  !CHECK: %[[CVT:.*]] = fir.convert %{{.*}} : (!fir.logical<4>) -> i1 loc(#[[IF_LOC:.*]])
-  !CHECK: omp.task if(%[[CVT]])
-  !$omp task if(c)
-    print *, "Task"
-  !$omp end task
-  !CHECK: } loc(#[[TASK_LOC:.*]])
-end subroutine
-
-!CHECK: #[[PAR_LOC]] = loc("{{.*}}location.f90":9:9)
-!CHECK: #[[TAR_LOC]] = loc("{{.*}}location.f90":21:9)
-!CHECK: #[[LOOP_LOC]] = loc("{{.*}}location.f90":32:9)
-!CHECK: #[[BAR_LOC]] = loc("{{.*}}location.f90":44:9)
-!CHECK: #[[TW_LOC]] = loc("{{.*}}location.f90":46:9)
-!CHECK: #[[TY_LOC]] = loc("{{.*}}location.f90":48:9)
-!CHECK: #[[IF_LOC]] = loc("{{.*}}location.f90":55:14)
-!CHECK: #[[TASK_LOC]] = loc("{{.*}}location.f90":55:9)
diff --git a/flang/test/Lower/OpenMP/FIR/loop-combined.f90 b/flang/test/Lower/OpenMP/FIR/loop-combined.f90
deleted file mode 100644
index 6c6618dc9fb5..000000000000
--- a/flang/test/Lower/OpenMP/FIR/loop-combined.f90
+++ /dev/null
@@ -1,83 +0,0 @@
-! This test checks lowering of OpenMP combined loop constructs.
-
-! RUN: bbc -fopenmp -emit-fir %s -o - | FileCheck %s
-! RUN: %flang_fc1 -fopenmp -emit-fir %s -o - | FileCheck %s
-
-program main
-  integer :: i
-
-  ! TODO When DISTRIBUTE, TASKLOOP and TEAMS are supported add:
-  ! - DISTRIBUTE PARALLEL DO SIMD
-  ! - DISTRIBUTE PARALLEL DO
-  ! - DISTRIBUTE SIMD
-  ! - TARGET TEAMS DISTRIBUTE PARALLEL DO SIMD
-  ! - TARGET TEAMS DISTRIBUTE PARALLEL DO
-  ! - TARGET TEAMS DISTRIBUTE SIMD
-  ! - TARGET TEAMS DISTRIBUTE
-  ! - TASKLOOP SIMD
-  ! - TEAMS DISTRIBUTE PARALLEL DO SIMD
-  ! - TEAMS DISTRIBUTE PARALLEL DO
-  ! - TEAMS DISTRIBUTE SIMD
-  ! - TEAMS DISTRIBUTE
-
-  ! ----------------------------------------------------------------------------
-  ! DO SIMD
-  ! ----------------------------------------------------------------------------
-  ! CHECK: omp.wsloop
-  !$omp do simd
-  do i = 1, 10
-  end do
-  !$omp end do simd
-
-  ! ----------------------------------------------------------------------------
-  ! PARALLEL DO SIMD
-  ! ----------------------------------------------------------------------------
-  ! CHECK: omp.parallel
-  ! CHECK: omp.wsloop
-  !$omp parallel do simd
-  do i = 1, 10
-  end do
-  !$omp end parallel do simd
-
-  ! ----------------------------------------------------------------------------
-  ! PARALLEL DO
-  ! ----------------------------------------------------------------------------
-  ! CHECK: omp.parallel
-  ! CHECK: omp.wsloop
-  !$omp parallel do
-  do i = 1, 10
-  end do
-  !$omp end parallel do
-
-  ! ----------------------------------------------------------------------------
-  ! TARGET PARALLEL DO SIMD
-  ! ----------------------------------------------------------------------------
-  ! CHECK: omp.target
-  ! CHECK: omp.parallel
-  ! CHECK: omp.wsloop
-  !$omp target parallel do simd
-  do i = 1, 10
-  end do
-  !$omp end target parallel do simd
-
-  ! ----------------------------------------------------------------------------
-  ! TARGET PARALLEL DO
-  ! ----------------------------------------------------------------------------
-  ! CHECK: omp.target
-  ! CHECK: omp.parallel
-  ! CHECK: omp.wsloop
-  !$omp target parallel do
-  do i = 1, 10
-  end do
-  !$omp end target parallel do
-
-  ! ----------------------------------------------------------------------------
-  ! TARGET SIMD
-  ! ----------------------------------------------------------------------------
-  ! CHECK: omp.target
-  ! CHECK: omp.simd
-  !$omp target simd
-  do i = 1, 10
-  end do
-  !$omp end target simd
-end program main
diff --git a/flang/test/Lower/OpenMP/FIR/map-component-ref.f90 b/flang/test/Lower/OpenMP/FIR/map-component-ref.f90
deleted file mode 100644
index 6799941701f4..000000000000
--- a/flang/test/Lower/OpenMP/FIR/map-component-ref.f90
+++ /dev/null
@@ -1,33 +0,0 @@
-! RUN: %flang_fc1 -emit-fir -fopenmp %s -o - | FileCheck %s
-! RUN: bbc -fopenmp -emit-fir %s -o - | FileCheck %s
-
-! CHECK: %[[V0:[0-9]+]] = fir.alloca !fir.type<_QFfooTt0{a0:i32,a1:i32}> {bindc_name = "a", uniq_name = "_QFfooEa"}
-! CHECK: %[[V1:[0-9]+]] = fir.declare %[[V0]] {uniq_name = "_QFfooEa"} : (!fir.ref<!fir.type<_QFfooTt0{a0:i32,a1:i32}>>) -> !fir.ref<!fir.type<_QFfooTt0{a0:i32,a1:i32}>>
-! CHECK: %[[V2:[0-9]+]] = fir.field_index a1, !fir.type<_QFfooTt0{a0:i32,a1:i32}>
-! CHECK: %[[V3:[0-9]+]] = fir.coordinate_of %[[V1]], %[[V2]] : (!fir.ref<!fir.type<_QFfooTt0{a0:i32,a1:i32}>>, !fir.field) -> !fir.ref<i32>
-! CHECK: %[[V4:[0-9]+]] = omp.map.info var_ptr(%[[V3]] : !fir.ref<i32>, i32) map_clauses(tofrom) capture(ByRef) -> !fir.ref<i32> {name = "a%a1"}
-! CHECK: %[[V5:[0-9]+]] = omp.map.info var_ptr(%[[V1]] : !fir.ref<!fir.type<_QFfooTt0{a0:i32,a1:i32}>>, !fir.type<_QFfooTt0{a0:i32,a1:i32}>) map_clauses(implicit, tofrom) capture(ByRef) -> !fir.ref<!fir.type<_QFfooTt0{a0:i32,a1:i32}>> {name = "a"}
-! CHECK: omp.target map_entries(%[[V4]] -> %arg0, %[[V5]] -> %arg1 : !fir.ref<i32>, !fir.ref<!fir.type<_QFfooTt0{a0:i32,a1:i32}>>) {
-! CHECK: ^bb0(%arg0: !fir.ref<i32>, %arg1: !fir.ref<!fir.type<_QFfooTt0{a0:i32,a1:i32}>>):
-! CHECK:   %c0_i32 = arith.constant 0 : i32
-! CHECK:   %[[V6:[0-9]+]] = fir.declare %arg1 {uniq_name = "_QFfooEa"} : (!fir.ref<!fir.type<_QFfooTt0{a0:i32,a1:i32}>>) -> !fir.ref<!fir.type<_QFfooTt0{a0:i32,a1:i32}>>
-! CHECK:   %[[V7:[0-9]+]] = fir.field_index a1, !fir.type<_QFfooTt0{a0:i32,a1:i32}>
-! CHECK:   %[[V8:[0-9]+]] = fir.coordinate_of %[[V6]], %[[V7]] : (!fir.ref<!fir.type<_QFfooTt0{a0:i32,a1:i32}>>, !fir.field) -> !fir.ref<i32>
-! CHECK:   fir.store %c0_i32 to %[[V8]] : !fir.ref<i32>
-! CHECK:   omp.terminator
-! CHECK: }
-
-subroutine foo()
-  implicit none
-
-  type t0
-    integer :: a0, a1
-  end type
-
-  type(t0) :: a
-
-  !$omp target map(a%a1)
-  a%a1 = 0
-  !$omp end target
-end
-
diff --git a/flang/test/Lower/OpenMP/FIR/master.f90 b/flang/test/Lower/OpenMP/FIR/master.f90
deleted file mode 100644
index 3bac582c7725..000000000000
--- a/flang/test/Lower/OpenMP/FIR/master.f90
+++ /dev/null
@@ -1,100 +0,0 @@
-!RUN: %flang_fc1 -emit-fir -flang-deprecated-no-hlfir -fopenmp %s -o - | FileCheck %s --check-prefixes="FIRDialect,OMPDialect"
-!RUN: %flang_fc1 -emit-fir -flang-deprecated-no-hlfir -fopenmp %s -o - | fir-opt --cfg-conversion-on-func-opt | fir-opt --fir-to-llvm-ir | FileCheck %s --check-prefixes="OMPDialect"
-
-!===============================================================================
-! parallel construct with function call which has master construct internally
-!===============================================================================
-!FIRDialect-LABEL: func @_QPomp_master
-subroutine omp_master()
-
-!OMPDialect: omp.master  {
-!$omp master
-
-    !FIRDialect: fir.call @_QPmaster() {{.*}}: () -> ()
-    call master()
-
-!OMPDialect: omp.terminator
-!$omp end master
-
-end subroutine omp_master
-
-!FIRDialect-LABEL: func @_QPparallel_function_master
-subroutine parallel_function_master()
-
-!OMPDialect: omp.parallel {
-!$omp parallel
-
-    !FIRDialect: fir.call @_QPfoo() {{.*}}: () -> ()
-    call foo()
-
-!OMPDialect: omp.terminator
-!$omp end parallel
-
-end subroutine parallel_function_master
-
-!===============================================================================
-! master construct nested inside parallel construct
-!===============================================================================
-
-!FIRDialect-LABEL: func @_QPomp_parallel_master
-subroutine omp_parallel_master()
-
-!OMPDialect: omp.parallel {
-!$omp parallel
-    !FIRDialect: fir.call @_QPparallel() {{.*}}: () -> ()
-    call parallel()
-
-!OMPDialect: omp.master {
-!$omp master
-
-    !FIRDialect: fir.call @_QPparallel_master() {{.*}}: () -> ()
-    call parallel_master()
-
-!OMPDialect: omp.terminator
-!$omp end master
-
-!OMPDialect: omp.terminator
-!$omp end parallel
-
-end subroutine omp_parallel_master
-
-!===============================================================================
-! master construct nested inside parallel construct with conditional flow
-!===============================================================================
-
-!FIRDialect-LABEL: func @_QPomp_master_parallel
-subroutine omp_master_parallel()
-    integer :: alpha, beta, gama
-    alpha = 4
-    beta = 5
-    gama = 6
-
-!OMPDialect: omp.master {
-!$omp master
-
-    !FIRDialect: %{{.*}} = fir.load %{{.*}}
-    !FIRDialect: %{{.*}} = fir.load %{{.*}}
-    !FIRDialect: %[[RESULT:.*]] = arith.cmpi sge, %{{.*}}, %{{.*}}
-    !FIRDialect: fir.if %[[RESULT]] {
-    if (alpha .ge. gama) then
-
-!OMPDialect: omp.parallel {
-!$omp parallel
-        !FIRDialect: fir.call @_QPinside_if_parallel() {{.*}}: () -> ()
-        call inside_if_parallel()
-
-!OMPDialect: omp.terminator
-!$omp end parallel
-
-        !FIRDialect: %{{.*}} = fir.load %{{.*}}
-        !FIRDialect: %{{.*}} = fir.load %{{.*}}
-        !FIRDialect: %{{.*}} = arith.addi %{{.*}}, %{{.*}}
-        !FIRDialect: fir.store %{{.*}} to %{{.*}}
-        beta = alpha + gama
-    end if
-    !FIRDialect: else
-
-!OMPDialect: omp.terminator
-!$omp end master
-
-end subroutine omp_master_parallel
diff --git a/flang/test/Lower/OpenMP/FIR/omp-declare-target-program-var.f90 b/flang/test/Lower/OpenMP/FIR/omp-declare-target-program-var.f90
deleted file mode 100644
index 0da76f6d9ad2..000000000000
--- a/flang/test/Lower/OpenMP/FIR/omp-declare-target-program-var.f90
+++ /dev/null
@@ -1,12 +0,0 @@
-!RUN: %flang_fc1 -emit-fir -fopenmp %s -o - | FileCheck %s --check-prefixes=HOST,ALL
-!RUN: %flang_fc1 -emit-fir -fopenmp -fopenmp-is-target-device %s -o - | FileCheck %s --check-prefix=ALL
-
-PROGRAM main
-    ! HOST-DAG: %0 = fir.alloca f32 {bindc_name = "i", uniq_name = "_QFEi"}
-    REAL :: I
-    ! ALL-DAG: fir.global internal @_QFEi {omp.declare_target = #omp.declaretarget<device_type = (any), capture_clause = (to)>} : f32 {
-    ! ALL-DAG: %0 = fir.undefined f32
-    ! ALL-DAG: fir.has_value %0 : f32
-    ! ALL-DAG: }
-    !$omp declare target(I)
-END
diff --git a/flang/test/Lower/OpenMP/FIR/omp-is-gpu.f90 b/flang/test/Lower/OpenMP/FIR/omp-is-gpu.f90
deleted file mode 100644
index ac8d24974801..000000000000
--- a/flang/test/Lower/OpenMP/FIR/omp-is-gpu.f90
+++ /dev/null
@@ -1,16 +0,0 @@
-!REQUIRES: amdgpu-registered-target, nvptx-registered-target
-
-!RUN: %flang_fc1 -triple amdgcn-amd-amdhsa -emit-fir -fopenmp -fopenmp-is-target-device %s -o - | FileCheck %s
-!RUN: %flang_fc1 -triple nvptx64-nvidia-cuda -emit-fir -fopenmp -fopenmp-is-target-device %s -o - | FileCheck %s
-!RUN: bbc -fopenmp -fopenmp-is-target-device -fopenmp-is-gpu -emit-fir -o - %s | FileCheck %s
-
-!RUN: not %flang_fc1 -triple amdgcn-amd-amdhsa -emit-fir -fopenmp %s -o - 2>&1 | FileCheck %s --check-prefix=FLANG-ERROR
-!RUN: not %flang_fc1 -triple nvptx64-nvidia-cuda -emit-fir -fopenmp %s -o - 2>&1 | FileCheck %s --check-prefix=FLANG-ERROR
-!RUN: not bbc -fopenmp -fopenmp-is-gpu -emit-fir %s -o - 2>&1 | FileCheck %s --check-prefix=BBC-ERROR
-
-!CHECK: module attributes {{{.*}}omp.is_gpu = true
-subroutine omp_subroutine()
-end subroutine omp_subroutine
-
-!FLANG-ERROR: error: OpenMP AMDGPU/NVPTX is only prepared to deal with device code.
-!BBC-ERROR: FATAL: -fopenmp-is-gpu can only be set if -fopenmp-is-target-device is also set
diff --git a/flang/test/Lower/OpenMP/FIR/ordered-threads.f90 b/flang/test/Lower/OpenMP/FIR/ordered-threads.f90
deleted file mode 100644
index 2dea4c857e87..000000000000
--- a/flang/test/Lower/OpenMP/FIR/ordered-threads.f90
+++ /dev/null
@@ -1,40 +0,0 @@
-! This test checks lowering of OpenMP ordered directive with threads Clause.
-! Without clause in ordered direcitve, it behaves as if threads clause is
-! specified.
-
-!RUN: %flang_fc1 -emit-fir -flang-deprecated-no-hlfir -fopenmp %s -o - | FileCheck %s --check-prefix=FIRDialect
-!RUN: %flang_fc1 -emit-fir -flang-deprecated-no-hlfir -fopenmp %s -o - | fir-opt --fir-to-llvm-ir | FileCheck %s --check-prefix=LLVMIRDialect
-!RUN: %flang_fc1 -emit-fir -flang-deprecated-no-hlfir -fopenmp %s -o - | fir-opt --fir-to-llvm-ir | tco | FileCheck %s --check-prefix=LLVMIR
-
-subroutine ordered
-        integer :: i
-        integer :: a(20)
-
-!FIRDialect: omp.ordered.region  {
-!LLVMIRDialect: omp.ordered.region  {
-!LLVMIR: [[TMP0:%.*]] = call i32 @__kmpc_global_thread_num(ptr @[[GLOB0:[0-9]+]])
-!LLVMIR-NEXT: call void @__kmpc_ordered(ptr @[[GLOB0]], i32 [[TMP0]])
-!$OMP ORDERED
-        a(i) = a(i-1) + 1
-!FIRDialect:   omp.terminator
-!FIRDialect-NEXT: }
-!LLVMIRDialect:   omp.terminator
-!LLVMIRDialect-NEXT: }
-!LLVMIR: call void @__kmpc_end_ordered(ptr @[[GLOB0]], i32 [[TMP0]])
-!$OMP END ORDERED
-
-!FIRDialect: omp.ordered.region  {
-!LLVMIRDialect: omp.ordered.region  {
-!LLVMIR: [[TMP1:%.*]] = call i32 @__kmpc_global_thread_num(ptr @[[GLOB1:[0-9]+]])
-!LLVMIR-NEXT: call void @__kmpc_ordered(ptr @[[GLOB1]], i32 [[TMP1]])
-!$OMP ORDERED THREADS
-        a(i) = a(i-1) + 1
-!FIRDialect:   omp.terminator
-!FIRDialect-NEXT: }
-!LLVMIRDialect:   omp.terminator
-!LLVMIRDialect-NEXT: }
-!LLVMIR: call void @__kmpc_end_ordered(ptr @[[GLOB1]], i32 [[TMP1]])
-!LLVMIR-NEXT: ret void
-!$OMP END ORDERED
-
-end
diff --git a/flang/test/Lower/OpenMP/FIR/parallel-firstprivate-clause-scalar.f90 b/flang/test/Lower/OpenMP/FIR/parallel-firstprivate-clause-scalar.f90
deleted file mode 100644
index 37f916ecb84c..000000000000
--- a/flang/test/Lower/OpenMP/FIR/parallel-firstprivate-clause-scalar.f90
+++ /dev/null
@@ -1,159 +0,0 @@
-! This test checks lowering of `FIRSTPRIVATE` clause for scalar types.
-
-! REQUIRES: shell
-! RUN: bbc -fopenmp -emit-fir -hlfir=false %s -o - | FileCheck %s --check-prefix=FIRDialect
-
-!FIRDialect-DAG: func @_QPfirstprivate_complex(%[[ARG1:.*]]: !fir.ref<!fir.complex<4>>{{.*}}, %[[ARG2:.*]]: !fir.ref<!fir.complex<8>>{{.*}}) {
-!FIRDialect:   omp.parallel {
-!FIRDialect:     %[[ARG1_PVT:.*]] = fir.alloca !fir.complex<4> {bindc_name = "arg1", pinned, uniq_name = "_QFfirstprivate_complexEarg1"}
-!FIRDialect:     %[[ARG1_VAL:.*]] = fir.load %[[ARG1]] : !fir.ref<!fir.complex<4>>
-!FIRDialect:     fir.store %[[ARG1_VAL]] to %[[ARG1_PVT]] : !fir.ref<!fir.complex<4>>
-!FIRDialect:     %[[ARG2_PVT:.*]] = fir.alloca !fir.complex<8> {bindc_name = "arg2", pinned, uniq_name = "_QFfirstprivate_complexEarg2"}
-!FIRDialect:     %[[ARG2_VAL:.*]] = fir.load %[[ARG2]] : !fir.ref<!fir.complex<8>>
-!FIRDialect:     fir.store %[[ARG2_VAL]] to %[[ARG2_PVT]] : !fir.ref<!fir.complex<8>>
-!FIRDialect:     fir.call @_QPfoo(%[[ARG1_PVT]], %[[ARG2_PVT]]) {{.*}}: (!fir.ref<!fir.complex<4>>, !fir.ref<!fir.complex<8>>) -> ()
-!FIRDialect:     omp.terminator
-!FIRDialect:   }
-
-subroutine firstprivate_complex(arg1, arg2)
-        complex(4) :: arg1
-        complex(8) :: arg2
-
-!$OMP PARALLEL FIRSTPRIVATE(arg1, arg2)
-        call foo(arg1, arg2)
-!$OMP END PARALLEL
-
-end subroutine
-
-!FIRDialect-DAG: func @_QPfirstprivate_integer(%[[ARG1:.*]]: !fir.ref<i32>{{.*}}, %[[ARG2:.*]]: !fir.ref<i8>{{.*}}, %[[ARG3:.*]]: !fir.ref<i16>{{.*}}, %[[ARG4:.*]]: !fir.ref<i32>{{.*}}, %[[ARG5:.*]]: !fir.ref<i64>{{.*}}, %[[ARG6:.*]]: !fir.ref<i128>{{.*}}) {
-!FIRDialect:  omp.parallel {
-!FIRDialect:    %[[ARG1_PVT:.*]] = fir.alloca i32 {bindc_name = "arg1", pinned, uniq_name = "_QFfirstprivate_integerEarg1"}
-!FIRDialect:    %[[ARG1_VAL:.*]] = fir.load %[[ARG1]] : !fir.ref<i32>
-!FIRDialect:    fir.store %[[ARG1_VAL]] to %[[ARG1_PVT]] : !fir.ref<i32>
-!FIRDialect:    %[[ARG2_PVT:.*]] = fir.alloca i8 {bindc_name = "arg2", pinned, uniq_name = "_QFfirstprivate_integerEarg2"}
-!FIRDialect:    %[[ARG2_VAL:.*]] = fir.load %[[ARG2]] : !fir.ref<i8>
-!FIRDialect:    fir.store %[[ARG2_VAL]] to %[[ARG2_PVT]] : !fir.ref<i8>
-!FIRDialect:    %[[ARG3_PVT:.*]] = fir.alloca i16 {bindc_name = "arg3", pinned, uniq_name = "_QFfirstprivate_integerEarg3"}
-!FIRDialect:    %[[ARG3_VAL:.*]] = fir.load %[[ARG3]] : !fir.ref<i16>
-!FIRDialect:    fir.store %[[ARG3_VAL]] to %[[ARG3_PVT]] : !fir.ref<i16>
-!FIRDialect:    %[[ARG4_PVT:.*]] = fir.alloca i32 {bindc_name = "arg4", pinned, uniq_name = "_QFfirstprivate_integerEarg4"}
-!FIRDialect:    %[[ARG4_VAL:.*]] = fir.load %[[ARG4]] : !fir.ref<i32>
-!FIRDialect:    fir.store %[[ARG4_VAL]] to %[[ARG4_PVT]] : !fir.ref<i32>
-!FIRDialect:    %[[ARG5_PVT:.*]] = fir.alloca i64 {bindc_name = "arg5", pinned, uniq_name = "_QFfirstprivate_integerEarg5"}
-!FIRDialect:    %[[ARG5_VAL:.*]] = fir.load %[[ARG5]] : !fir.ref<i64>
-!FIRDialect:    fir.store %[[ARG5_VAL]] to %[[ARG5_PVT]] : !fir.ref<i64>
-!FIRDialect:    %[[ARG6_PVT:.*]] = fir.alloca i128 {bindc_name = "arg6", pinned, uniq_name = "_QFfirstprivate_integerEarg6"}
-!FIRDialect:    %[[ARG6_VAL:.*]] = fir.load %[[ARG6]] : !fir.ref<i128>
-!FIRDialect:    fir.store %[[ARG6_VAL]] to %[[ARG6_PVT]] : !fir.ref<i128>
-!FIRDialect:    fir.call @_QPbar(%[[ARG1_PVT]], %[[ARG2_PVT]], %[[ARG3_PVT]], %[[ARG4_PVT]], %[[ARG5_PVT]], %[[ARG6_PVT]]) {{.*}}: (!fir.ref<i32>, !fir.ref<i8>, !fir.ref<i16>, !fir.ref<i32>, !fir.ref<i64>, !fir.ref<i128>) -> ()
-!FIRDialect:    omp.terminator
-!FIRDialect:  }
-
-subroutine firstprivate_integer(arg1, arg2, arg3, arg4, arg5, arg6)
-        integer :: arg1
-        integer(kind=1) :: arg2
-        integer(kind=2) :: arg3
-        integer(kind=4) :: arg4
-        integer(kind=8) :: arg5
-        integer(kind=16) :: arg6
-
-!$OMP PARALLEL FIRSTPRIVATE(arg1, arg2, arg3, arg4, arg5, arg6)
-        call bar(arg1, arg2, arg3, arg4, arg5, arg6)
-!$OMP END PARALLEL
-
-end subroutine
-
-!FIRDialect-DAG: func @_QPfirstprivate_logical(%[[ARG1:.*]]: !fir.ref<!fir.logical<4>>{{.*}}, %[[ARG2:.*]]: !fir.ref<!fir.logical<1>>{{.*}}, %[[ARG3:.*]]: !fir.ref<!fir.logical<2>>{{.*}}, %[[ARG4:.*]]: !fir.ref<!fir.logical<4>>{{.*}}, %[[ARG5:.*]]: !fir.ref<!fir.logical<8>>{{.*}}) {
-!FIRDialect:   omp.parallel {
-!FIRDialect:     %[[ARG1_PVT:.*]] = fir.alloca !fir.logical<4> {bindc_name = "arg1", pinned, uniq_name = "_QFfirstprivate_logicalEarg1"}
-!FIRDialect:     %[[ARG1_VAL:.*]] = fir.load %[[ARG1]] : !fir.ref<!fir.logical<4>>
-!FIRDialect:     fir.store %[[ARG1_VAL]] to %[[ARG1_PVT]] : !fir.ref<!fir.logical<4>>
-!FIRDialect:     %[[ARG2_PVT:.*]] = fir.alloca !fir.logical<1> {bindc_name = "arg2", pinned, uniq_name = "_QFfirstprivate_logicalEarg2"}
-!FIRDialect:     %[[ARG2_VAL:.*]] = fir.load %[[ARG2]] : !fir.ref<!fir.logical<1>>
-!FIRDialect:     fir.store %[[ARG2_VAL]] to %[[ARG2_PVT]] : !fir.ref<!fir.logical<1>>
-!FIRDialect:     %[[ARG3_PVT:.*]] = fir.alloca !fir.logical<2> {bindc_name = "arg3", pinned, uniq_name = "_QFfirstprivate_logicalEarg3"}
-!FIRDialect:     %[[ARG3_VAL:.*]] = fir.load %[[ARG3]] : !fir.ref<!fir.logical<2>>
-!FIRDialect:     fir.store %[[ARG3_VAL]] to %[[ARG3_PVT]] : !fir.ref<!fir.logical<2>>
-!FIRDialect:     %[[ARG4_PVT:.*]] = fir.alloca !fir.logical<4> {bindc_name = "arg4", pinned, uniq_name = "_QFfirstprivate_logicalEarg4"}
-!FIRDialect:     %[[ARG4_VAL:.*]] = fir.load %[[ARG4]] : !fir.ref<!fir.logical<4>>
-!FIRDialect:     fir.store %[[ARG4_VAL]] to %[[ARG4_PVT]] : !fir.ref<!fir.logical<4>>
-!FIRDialect:     %[[ARG5_PVT:.*]] = fir.alloca !fir.logical<8> {bindc_name = "arg5", pinned, uniq_name = "_QFfirstprivate_logicalEarg5"}
-!FIRDialect:     %[[ARG5_VAL:.*]] = fir.load %[[ARG5]] : !fir.ref<!fir.logical<8>>
-!FIRDialect:     fir.store %[[ARG5_VAL]] to %[[ARG5_PVT]] : !fir.ref<!fir.logical<8>>
-!FIRDialect:     fir.call @_QPbaz(%[[ARG1_PVT]], %[[ARG2_PVT]], %[[ARG3_PVT]], %[[ARG4_PVT]], %[[ARG5_PVT]]) {{.*}}: (!fir.ref<!fir.logical<4>>, !fir.ref<!fir.logical<1>>, !fir.ref<!fir.logical<2>>, !fir.ref<!fir.logical<4>>, !fir.ref<!fir.logical<8>>) -> ()
-!FIRDialect:     omp.terminator
-!FIRDialect:   }
-
-subroutine firstprivate_logical(arg1, arg2, arg3, arg4, arg5)
-        logical :: arg1
-        logical(kind=1) :: arg2
-        logical(kind=2) :: arg3
-        logical(kind=4) :: arg4
-        logical(kind=8) :: arg5
-
-!$OMP PARALLEL FIRSTPRIVATE(arg1, arg2, arg3, arg4, arg5)
-        call baz(arg1, arg2, arg3, arg4, arg5)
-!$OMP END PARALLEL
-
-end subroutine
-
-!FIRDialect-DAG: func @_QPfirstprivate_real(%[[ARG1:.*]]: !fir.ref<f32>{{.*}}, %[[ARG2:.*]]: !fir.ref<f16>{{.*}}, %[[ARG3:.*]]: !fir.ref<f32>{{.*}}, %[[ARG4:.*]]: !fir.ref<f64>{{.*}}, %[[ARG5:.*]]: !fir.ref<f80>{{.*}}, %[[ARG6:.*]]: !fir.ref<f128>{{.*}}) {
-!FIRDialect:   omp.parallel {
-!FIRDialect:     %[[ARG1_PVT:.*]] = fir.alloca f32 {bindc_name = "arg1", pinned, uniq_name = "_QFfirstprivate_realEarg1"}
-!FIRDialect:     %[[ARG1_VAL:.*]] = fir.load %[[ARG1]] : !fir.ref<f32>
-!FIRDialect:     fir.store %[[ARG1_VAL]] to %[[ARG1_PVT]] : !fir.ref<f32>
-!FIRDialect:     %[[ARG2_PVT:.*]] = fir.alloca f16 {bindc_name = "arg2", pinned, uniq_name = "_QFfirstprivate_realEarg2"}
-!FIRDialect:     %[[ARG2_VAL:.*]] = fir.load %[[ARG2]] : !fir.ref<f16>
-!FIRDialect:     fir.store %[[ARG2_VAL]] to %[[ARG2_PVT]] : !fir.ref<f16>
-!FIRDialect:     %[[ARG3_PVT:.*]] = fir.alloca f32 {bindc_name = "arg3", pinned, uniq_name = "_QFfirstprivate_realEarg3"}
-!FIRDialect:     %[[ARG3_VAL:.*]] = fir.load %[[ARG3]] : !fir.ref<f32>
-!FIRDialect:     fir.store %[[ARG3_VAL]] to %[[ARG3_PVT]] : !fir.ref<f32>
-!FIRDialect:     %[[ARG4_PVT:.*]] = fir.alloca f64 {bindc_name = "arg4", pinned, uniq_name = "_QFfirstprivate_realEarg4"}
-!FIRDialect:     %[[ARG4_VAL:.*]] = fir.load %[[ARG4]] : !fir.ref<f64>
-!FIRDialect:     fir.store %[[ARG4_VAL]] to %[[ARG4_PVT]] : !fir.ref<f64>
-!FIRDialect:     %[[ARG5_PVT:.*]] = fir.alloca f80 {bindc_name = "arg5", pinned, uniq_name = "_QFfirstprivate_realEarg5"}
-!FIRDialect:     %[[ARG5_VAL:.*]] = fir.load %[[ARG5]] : !fir.ref<f80>
-!FIRDialect:     fir.store %[[ARG5_VAL]] to %[[ARG5_PVT]] : !fir.ref<f80>
-!FIRDialect:     %[[ARG6_PVT:.*]] = fir.alloca f128 {bindc_name = "arg6", pinned, uniq_name = "_QFfirstprivate_realEarg6"}
-!FIRDialect:     %[[ARG6_VAL:.*]] = fir.load %[[ARG6]] : !fir.ref<f128>
-!FIRDialect:     fir.store %[[ARG6_VAL]] to %[[ARG6_PVT]] : !fir.ref<f128>
-!FIRDialect:     fir.call @_QPqux(%[[ARG1_PVT]], %[[ARG2_PVT]], %[[ARG3_PVT]], %[[ARG4_PVT]], %[[ARG5_PVT]], %[[ARG6_PVT]]) {{.*}}: (!fir.ref<f32>, !fir.ref<f16>, !fir.ref<f32>, !fir.ref<f64>, !fir.ref<f80>, !fir.ref<f128>) -> ()
-!FIRDialect:     omp.terminator
-!FIRDialect:   }
-
-subroutine firstprivate_real(arg1, arg2, arg3, arg4, arg5, arg6)
-        real :: arg1
-        real(kind=2) :: arg2
-        real(kind=4) :: arg3
-        real(kind=8) :: arg4
-        real(kind=10) :: arg5
-        real(kind=16) :: arg6
-
-!$OMP PARALLEL FIRSTPRIVATE(arg1, arg2, arg3, arg4, arg5, arg6)
-        call qux(arg1, arg2, arg3, arg4, arg5, arg6)
-!$OMP END PARALLEL
-
-end subroutine
-
-!FIRDialect-LABEL:   func.func @_QPmultiple_firstprivate(
-!FIRDialect-SAME:                                        %[[A_ADDR:.*]]: !fir.ref<i32> {fir.bindc_name = "a"},
-!FIRDialect-SAME:                                        %[[B_ADDR:.*]]: !fir.ref<i32> {fir.bindc_name = "b"}) {
-!FIRDialect:           omp.parallel   {
-!FIRDialect:             %[[A_PRIV_ADDR:.*]] = fir.alloca i32 {bindc_name = "a", pinned, uniq_name = "_QFmultiple_firstprivateEa"}
-!FIRDialect:             %[[A:.*]] = fir.load %[[A_ADDR]] : !fir.ref<i32>
-!FIRDialect:             fir.store %[[A]] to %[[A_PRIV_ADDR]] : !fir.ref<i32>
-!FIRDialect:             %[[B_PRIV_ADDR:.*]] = fir.alloca i32 {bindc_name = "b", pinned, uniq_name = "_QFmultiple_firstprivateEb"}
-!FIRDialect:             %[[B:.*]] = fir.load %[[B_ADDR]] : !fir.ref<i32>
-!FIRDialect:             fir.store %[[B]] to %[[B_PRIV_ADDR]] : !fir.ref<i32>
-!FIRDialect:             fir.call @_QPquux(%[[A_PRIV_ADDR]], %[[B_PRIV_ADDR]]) {{.*}}: (!fir.ref<i32>, !fir.ref<i32>) -> ()
-!FIRDialect:             omp.terminator
-!FIRDialect:           }
-!FIRDialect:           return
-!FIRDialect:         }
-
-subroutine multiple_firstprivate(a, b)
-        integer :: a, b
-!$OMP PARALLEL FIRSTPRIVATE(a) FIRSTPRIVATE(b)
-        call quux(a, b)
-!$OMP END PARALLEL
-end subroutine multiple_firstprivate
diff --git a/flang/test/Lower/OpenMP/FIR/parallel-lastprivate-clause-scalar.f90 b/flang/test/Lower/OpenMP/FIR/parallel-lastprivate-clause-scalar.f90
deleted file mode 100644
index 2060e2062c1a..000000000000
--- a/flang/test/Lower/OpenMP/FIR/parallel-lastprivate-clause-scalar.f90
+++ /dev/null
@@ -1,237 +0,0 @@
-! This test checks lowering of `LASTPRIVATE` clause for scalar types.
-
-! RUN: bbc -fopenmp -emit-fir -hlfir=false %s -o - | FileCheck %s
-! RUN: %flang_fc1 -fopenmp -emit-fir -flang-deprecated-no-hlfir %s -o - | FileCheck %s
-
-!CHECK: func @_QPlastprivate_character(%[[ARG1:.*]]: !fir.boxchar<1>{{.*}}) {
-!CHECK-DAG: %[[ARG1_UNBOX:.*]]:2 = fir.unboxchar
-!CHECK-DAG: %[[FIVE:.*]] = arith.constant 5 : index
-!CHECK-DAG: %[[ARG1_REF:.*]] = fir.convert %[[ARG1_UNBOX]]#0 : (!fir.ref<!fir.char<1,?>>) -> !fir.ref<!fir.char<1,5>>
-
-!CHECK: omp.parallel {
-!CHECK-DAG: %[[ARG1_PVT:.*]] = fir.alloca !fir.char<1,5> {bindc_name = "arg1", 
-
-! Check that we are accessing the clone inside the loop
-!CHECK-DAG: omp.wsloop for (%[[INDX_WS:.*]]) : {{.*}} {
-!CHECK-DAG: %[[UNIT:.*]] = arith.constant 6 : i32
-!CHECK-NEXT: %[[ADDR:.*]] = fir.address_of(@_QQclX
-!CHECK-NEXT: %[[CVT0:.*]] = fir.convert %[[ADDR]] 
-!CHECK-NEXT: %[[CNST:.*]] = arith.constant
-!CHECK-NEXT: %[[CALL_BEGIN_IO:.*]] = fir.call @_FortranAioBeginExternalListOutput(%[[UNIT]], %[[CVT0]], %[[CNST]]) {{.*}}: (i32, !fir.ref<i8>, i32) -> !fir.ref<i8>
-!CHECK-NEXT: %[[CVT_0_1:.*]] = fir.convert %[[ARG1_PVT]] 
-!CHECK-NEXT: %[[CVT_0_2:.*]] = fir.convert %[[FIVE]]
-!CHECK-NEXT: %[[CALL_OP_ASCII:.*]] = fir.call @_FortranAioOutputAscii(%[[CALL_BEGIN_IO]], %[[CVT_0_1]], %[[CVT_0_2]])
-!CHECK-NEXT: %[[CALL_END_IO:.*]] = fir.call @_FortranAioEndIoStatement(%[[CALL_BEGIN_IO]])
-
-! Testing last iteration check
-!CHECK: %[[V:.*]] = arith.addi %[[INDX_WS]], %{{.*}} : i32
-!CHECK: %[[C0:.*]] = arith.constant 0 : i32
-!CHECK: %[[T1:.*]] = arith.cmpi slt, %{{.*}}, %[[C0]] : i32
-!CHECK: %[[T2:.*]] = arith.cmpi slt, %[[V]], %{{.*}} : i32
-!CHECK: %[[T3:.*]] = arith.cmpi sgt, %[[V]], %{{.*}} : i32
-!CHECK: %[[IV_CMP:.*]] = arith.select %[[T1]], %[[T2]], %[[T3]] : i1
-!CHECK: fir.if %[[IV_CMP]] {
-!CHECK: fir.store %[[V]] to %{{.*}} : !fir.ref<i32>
-
-! Testing lastprivate val update
-!CHECK-DAG: %[[CVT:.*]] = fir.convert %[[ARG1_REF]] : (!fir.ref<!fir.char<1,5>>) -> !fir.ref<i8>
-!CHECK-DAG: %[[CVT1:.*]] = fir.convert %[[ARG1_PVT]] : (!fir.ref<!fir.char<1,5>>) -> !fir.ref<i8>
-!CHECK-DAG: fir.call @llvm.memmove.p0.p0.i64(%[[CVT]], %[[CVT1]]{{.*}})
-!CHECK-DAG: } 
-!CHECK-DAG: omp.yield
-
-subroutine lastprivate_character(arg1)
-        character(5) :: arg1
-!$OMP PARALLEL 
-!$OMP DO LASTPRIVATE(arg1)
-do n = 1, 5
-        arg1(n:n) = 'c'
-        print *, arg1
-end do
-!$OMP END DO
-!$OMP END PARALLEL
-end subroutine
-
-!CHECK: func @_QPlastprivate_int(%[[ARG1:.*]]: !fir.ref<i32> {fir.bindc_name = "arg1"}) {
-!CHECK-DAG: omp.parallel  {
-!CHECK-DAG: %[[CLONE:.*]] = fir.alloca i32 {bindc_name = "arg1"
-!CHECK: omp.wsloop for (%[[INDX_WS:.*]]) : {{.*}} {
-
-! Testing last iteration check
-!CHECK: %[[V:.*]] = arith.addi %[[INDX_WS]], %{{.*}} : i32
-!CHECK: %[[C0:.*]] = arith.constant 0 : i32
-!CHECK: %[[T1:.*]] = arith.cmpi slt, %{{.*}}, %[[C0]] : i32
-!CHECK: %[[T2:.*]] = arith.cmpi slt, %[[V]], %{{.*}} : i32
-!CHECK: %[[T3:.*]] = arith.cmpi sgt, %[[V]], %{{.*}} : i32
-!CHECK: %[[IV_CMP:.*]] = arith.select %[[T1]], %[[T2]], %[[T3]] : i1
-!CHECK: fir.if %[[IV_CMP]] {
-!CHECK: fir.store %[[V]] to %{{.*}} : !fir.ref<i32>
-
-! Testing lastprivate val update
-!CHECK-NEXT: %[[CLONE_LD:.*]] = fir.load %[[CLONE]] : !fir.ref<i32>
-!CHECK-NEXT: fir.store %[[CLONE_LD]] to %[[ARG1]] : !fir.ref<i32>
-!CHECK-DAG: }
-!CHECK-DAG: omp.yield
-
-subroutine lastprivate_int(arg1)
-        integer :: arg1
-!$OMP PARALLEL 
-!$OMP DO LASTPRIVATE(arg1)
-do n = 1, 5
-        arg1 = 2
-        print *, arg1
-end do
-!$OMP END DO
-!$OMP END PARALLEL
-print *, arg1
-end subroutine
-
-!CHECK: func.func @_QPmult_lastprivate_int(%[[ARG1:.*]]: !fir.ref<i32> {fir.bindc_name = "arg1"}, %[[ARG2:.*]]: !fir.ref<i32> {fir.bindc_name = "arg2"}) {
-!CHECK: omp.parallel  {
-!CHECK-DAG: %[[CLONE1:.*]] = fir.alloca i32 {bindc_name = "arg1"
-!CHECK-DAG: %[[CLONE2:.*]] = fir.alloca i32 {bindc_name = "arg2"
-!CHECK: omp.wsloop for (%[[INDX_WS:.*]]) : {{.*}} {
-
-! Testing last iteration check
-!CHECK: %[[V:.*]] = arith.addi %[[INDX_WS]], %{{.*}} : i32
-!CHECK: %[[C0:.*]] = arith.constant 0 : i32
-!CHECK: %[[T1:.*]] = arith.cmpi slt, %{{.*}}, %[[C0]] : i32
-!CHECK: %[[T2:.*]] = arith.cmpi slt, %[[V]], %{{.*}} : i32
-!CHECK: %[[T3:.*]] = arith.cmpi sgt, %[[V]], %{{.*}} : i32
-!CHECK: %[[IV_CMP:.*]] = arith.select %[[T1]], %[[T2]], %[[T3]] : i1
-!CHECK: fir.if %[[IV_CMP]] {
-!CHECK: fir.store %[[V]] to %{{.*}} : !fir.ref<i32>
-! Testing lastprivate val update
-!CHECK-DAG: %[[CLONE_LD1:.*]] = fir.load %[[CLONE1]] : !fir.ref<i32>
-!CHECK-DAG: fir.store %[[CLONE_LD1]] to %[[ARG1]] : !fir.ref<i32>
-!CHECK-DAG: %[[CLONE_LD2:.*]] = fir.load %[[CLONE2]] : !fir.ref<i32>
-!CHECK-DAG: fir.store %[[CLONE_LD2]] to %[[ARG2]] : !fir.ref<i32>
-!CHECK: }
-!CHECK: omp.yield
-
-subroutine mult_lastprivate_int(arg1, arg2)
-        integer :: arg1, arg2
-!$OMP PARALLEL 
-!$OMP DO LASTPRIVATE(arg1) LASTPRIVATE(arg2)
-do n = 1, 5
-        arg1 = 2
-        arg2 = 3
-        print *, arg1, arg2
-end do
-!$OMP END DO
-!$OMP END PARALLEL
-print *, arg1, arg2
-end subroutine
-
-!CHECK: func.func @_QPmult_lastprivate_int2(%[[ARG1:.*]]: !fir.ref<i32> {fir.bindc_name = "arg1"}, %[[ARG2:.*]]: !fir.ref<i32> {fir.bindc_name = "arg2"}) {
-!CHECK: omp.parallel  {
-!CHECK-DAG: %[[CLONE1:.*]] = fir.alloca i32 {bindc_name = "arg1"
-!CHECK-DAG: %[[CLONE2:.*]] = fir.alloca i32 {bindc_name = "arg2"
-!CHECK: omp.wsloop for (%[[INDX_WS:.*]]) : {{.*}} {
-
-!Testing last iteration check
-!CHECK: %[[V:.*]] = arith.addi %[[INDX_WS]], %{{.*}} : i32
-!CHECK: %[[C0:.*]] = arith.constant 0 : i32
-!CHECK: %[[T1:.*]] = arith.cmpi slt, %{{.*}}, %[[C0]] : i32
-!CHECK: %[[T2:.*]] = arith.cmpi slt, %[[V]], %{{.*}} : i32
-!CHECK: %[[T3:.*]] = arith.cmpi sgt, %[[V]], %{{.*}} : i32
-!CHECK: %[[IV_CMP:.*]] = arith.select %[[T1]], %[[T2]], %[[T3]] : i1
-!CHECK: fir.if %[[IV_CMP]] {
-!CHECK: fir.store %[[V]] to %{{.*}} : !fir.ref<i32>
-!Testing lastprivate val update
-!CHECK-DAG: %[[CLONE_LD2:.*]] = fir.load %[[CLONE2]] : !fir.ref<i32>
-!CHECK-DAG: fir.store %[[CLONE_LD2]] to %[[ARG2]] : !fir.ref<i32>
-!CHECK-DAG: %[[CLONE_LD1:.*]] = fir.load %[[CLONE1]] : !fir.ref<i32>
-!CHECK-DAG: fir.store %[[CLONE_LD1]] to %[[ARG1]] : !fir.ref<i32>
-!CHECK: }
-!CHECK: omp.yield
-
-subroutine mult_lastprivate_int2(arg1, arg2)
-        integer :: arg1, arg2
-!$OMP PARALLEL 
-!$OMP DO LASTPRIVATE(arg1, arg2)
-do n = 1, 5
-        arg1 = 2
-        arg2 = 3
-        print *, arg1, arg2
-end do
-!$OMP END DO
-!$OMP END PARALLEL
-print *, arg1, arg2
-end subroutine
-
-!CHECK: func.func @_QPfirstpriv_lastpriv_int(%[[ARG1:.*]]: !fir.ref<i32> {fir.bindc_name = "arg1"}, %[[ARG2:.*]]: !fir.ref<i32> {fir.bindc_name = "arg2"}) {
-!CHECK: omp.parallel  {
-! Firstprivate update
-!CHECK-DAG: %[[CLONE1:.*]] = fir.alloca i32 {bindc_name = "arg1"
-!CHECK-DAG: %[[FPV_LD:.*]] = fir.load %[[ARG1]] : !fir.ref<i32>
-!CHECK-DAG: fir.store %[[FPV_LD]] to %[[CLONE1]] : !fir.ref<i32>
-! Lastprivate Allocation
-!CHECK-DAG: %[[CLONE2:.*]] = fir.alloca i32 {bindc_name = "arg2"
-!CHECK-NOT: omp.barrier
-!CHECK: omp.wsloop for (%[[INDX_WS:.*]]) : {{.*}} {
-
-! Testing last iteration check
-!CHECK: %[[V:.*]] = arith.addi %[[INDX_WS]], %{{.*}} : i32
-!CHECK: %[[C0:.*]] = arith.constant 0 : i32
-!CHECK: %[[T1:.*]] = arith.cmpi slt, %{{.*}}, %[[C0]] : i32
-!CHECK: %[[T2:.*]] = arith.cmpi slt, %[[V]], %{{.*}} : i32
-!CHECK: %[[T3:.*]] = arith.cmpi sgt, %[[V]], %{{.*}} : i32
-!CHECK: %[[IV_CMP:.*]] = arith.select %[[T1]], %[[T2]], %[[T3]] : i1
-!CHECK: fir.if %[[IV_CMP]] {
-!CHECK: fir.store %[[V]] to %{{.*}} : !fir.ref<i32>
-! Testing lastprivate val update
-!CHECK-NEXT: %[[CLONE_LD:.*]] = fir.load %[[CLONE2]] : !fir.ref<i32>
-!CHECK-NEXT: fir.store %[[CLONE_LD]] to %[[ARG2]] : !fir.ref<i32>
-!CHECK-NEXT: }
-!CHECK-NEXT: omp.yield
-
-subroutine firstpriv_lastpriv_int(arg1, arg2)
-        integer :: arg1, arg2
-!$OMP PARALLEL 
-!$OMP DO FIRSTPRIVATE(arg1) LASTPRIVATE(arg2)
-do n = 1, 5
-        arg1 = 2
-        arg2 = 3
-        print *, arg1, arg2
-end do
-!$OMP END DO
-!$OMP END PARALLEL
-print *, arg1, arg2
-end subroutine
-
-!CHECK: func.func @_QPfirstpriv_lastpriv_int2(%[[ARG1:.*]]: !fir.ref<i32> {fir.bindc_name = "arg1"}) {
-!CHECK: omp.parallel  {
-! Firstprivate update
-!CHECK: %[[CLONE1:.*]] = fir.alloca i32 {bindc_name = "arg1"
-!CHECK-NEXT: %[[FPV_LD:.*]] = fir.load %[[ARG1]] : !fir.ref<i32>
-!CHECK-NEXT: fir.store %[[FPV_LD]] to %[[CLONE1]] : !fir.ref<i32>
-!CHECK-NEXT: omp.barrier
-!CHECK: omp.wsloop for (%[[INDX_WS:.*]]) : {{.*}} {
-! Testing last iteration check
-!CHECK: %[[V:.*]] = arith.addi %[[INDX_WS]], %{{.*}} : i32
-!CHECK: %[[C0:.*]] = arith.constant 0 : i32
-!CHECK: %[[T1:.*]] = arith.cmpi slt, %{{.*}}, %[[C0]] : i32
-!CHECK: %[[T2:.*]] = arith.cmpi slt, %[[V]], %{{.*}} : i32
-!CHECK: %[[T3:.*]] = arith.cmpi sgt, %[[V]], %{{.*}} : i32
-!CHECK: %[[IV_CMP:.*]] = arith.select %[[T1]], %[[T2]], %[[T3]] : i1
-!CHECK: fir.if %[[IV_CMP]] {
-!CHECK: fir.store %[[V]] to %{{.*}} : !fir.ref<i32>
-! Testing lastprivate val update
-!CHECK-NEXT: %[[CLONE_LD:.*]] = fir.load %[[CLONE1]] : !fir.ref<i32>
-!CHECK-NEXT: fir.store %[[CLONE_LD]] to %[[ARG1]] : !fir.ref<i32>
-!CHECK-NEXT: }
-!CHECK-NEXT: omp.yield
-
-subroutine firstpriv_lastpriv_int2(arg1)
-        integer :: arg1
-!$OMP PARALLEL 
-!$OMP DO FIRSTPRIVATE(arg1) LASTPRIVATE(arg1)
-do n = 1, 5
-        arg1 = 2
-        print *, arg1
-end do
-!$OMP END DO
-!$OMP END PARALLEL
-print *, arg1
-end subroutine
diff --git a/flang/test/Lower/OpenMP/FIR/parallel-private-clause-fixes.f90 b/flang/test/Lower/OpenMP/FIR/parallel-private-clause-fixes.f90
deleted file mode 100644
index c99bf761333b..000000000000
--- a/flang/test/Lower/OpenMP/FIR/parallel-private-clause-fixes.f90
+++ /dev/null
@@ -1,81 +0,0 @@
-! This test checks a few bug fixes in the PRIVATE clause lowering
-
-! RUN: bbc -fopenmp -emit-fir -hlfir=false %s -o - | FileCheck %s
-
-! CHECK-LABEL: multiple_private_fix
-! CHECK:         %[[VAL_0:.*]] = fir.alloca i32 {bindc_name = "i", uniq_name = "_QFmultiple_private_fixEi"}
-! CHECK:         %[[VAL_1:.*]] = fir.alloca i32 {bindc_name = "j", uniq_name = "_QFmultiple_private_fixEj"}
-! CHECK:         %[[VAL_2:.*]] = fir.alloca i32 {bindc_name = "x", uniq_name = "_QFmultiple_private_fixEx"}
-! CHECK:         omp.parallel {
-! CHECK-DAG:           %[[PRIV_J:.*]] = fir.alloca i32 {bindc_name = "j", pinned
-! CHECK-DAG:           %[[PRIV_I:.*]] = fir.alloca i32 {adapt.valuebyref, pinned
-! CHECK-DAG:           %[[PRIV_X:.*]] = fir.alloca i32 {bindc_name = "x", pinned
-! CHECK:           %[[ONE:.*]] = arith.constant 1 : i32
-! CHECK:           %[[VAL_3:.*]] = fir.load %[[VAL_4:.*]] : !fir.ref<i32>
-! CHECK:           %[[VAL_5:.*]] = arith.constant 1 : i32
-! CHECK:           omp.wsloop for (%[[VAL_6:.*]]) : i32 = (%[[ONE]]) to (%[[VAL_3]]) inclusive step (%[[VAL_5]]) {
-! CHECK:             fir.store %[[VAL_6]] to %[[PRIV_I]] : !fir.ref<i32>
-! CHECK:             %[[VAL_7:.*]] = arith.constant 1 : i32
-! CHECK:             %[[VAL_8:.*]] = fir.convert %[[VAL_7]] : (i32) -> index
-! CHECK:             %[[VAL_9:.*]] = fir.load %[[VAL_4]] : !fir.ref<i32>
-! CHECK:             %[[VAL_10:.*]] = fir.convert %[[VAL_9]] : (i32) -> index
-! CHECK:             %[[VAL_11:.*]] = arith.constant 1 : index
-! CHECK:             %[[LB:.*]] = fir.convert %[[VAL_8]] : (index) -> i32
-! CHECK:             %[[VAL_12:.*]]:2 = fir.do_loop %[[VAL_13:[^ ]*]] =
-! CHECK-SAME:            %[[VAL_8]] to %[[VAL_10]] step %[[VAL_11]]
-! CHECK-SAME:            iter_args(%[[IV:.*]] = %[[LB]]) -> (index, i32) {
-! CHECK:               fir.store %[[IV]] to %[[PRIV_J]] : !fir.ref<i32>
-! CHECK:               %[[LOAD:.*]] = fir.load %[[PRIV_I]] : !fir.ref<i32>
-! CHECK:               %[[VAL_15:.*]] = fir.load %[[PRIV_J]] : !fir.ref<i32>
-! CHECK:               %[[VAL_16:.*]] = arith.addi %[[LOAD]], %[[VAL_15]] : i32
-! CHECK:               fir.store %[[VAL_16]] to %[[PRIV_X]] : !fir.ref<i32>
-! CHECK:               %[[VAL_17:.*]] = arith.addi %[[VAL_13]], %[[VAL_11]] : index
-! CHECK:               %[[STEPCAST:.*]] = fir.convert %[[VAL_11]] : (index) -> i32
-! CHECK:               %[[IVLOAD:.*]] = fir.load %[[PRIV_J]] : !fir.ref<i32>
-! CHECK:               %[[IVINC:.*]] = arith.addi %[[IVLOAD]], %[[STEPCAST]]
-! CHECK:               fir.result %[[VAL_17]], %[[IVINC]] : index, i32
-! CHECK:             }
-! CHECK:             fir.store %[[VAL_12]]#1 to %[[PRIV_J]] : !fir.ref<i32>
-! CHECK:             omp.yield
-! CHECK:           }
-! CHECK:           omp.terminator
-! CHECK:         }
-! CHECK:         return
-subroutine multiple_private_fix(gama)
-        integer :: i, j, x, gama
-!$OMP PARALLEL DO PRIVATE(j,x)
-        do i = 1, gama
-          do j = 1, gama
-            x = i + j
-          end do
-        end do
-!$OMP END PARALLEL DO
-end subroutine
-
-! CHECK-LABEL: multiple_private_fix2
-! CHECK:  %[[X1:.*]] = fir.alloca i32 {bindc_name = "x", uniq_name = "_QFmultiple_private_fix2Ex"}
-! CHECK:  omp.parallel  {
-! CHECK:    %[[X2:.*]] = fir.alloca i32 {bindc_name = "x", pinned, uniq_name = "_QFmultiple_private_fix2Ex"}
-! CHECK:    omp.parallel  {
-! CHECK:      %[[X3:.*]] = fir.alloca i32 {bindc_name = "x", pinned, uniq_name = "_QFmultiple_private_fix2Ex"}
-! CHECK:      %[[C3:.*]] = arith.constant 1 : i32
-! CHECK:      fir.store %[[C3]] to %[[X3]] : !fir.ref<i32>
-! CHECK:      omp.terminator
-! CHECK:    }
-! CHECK:      %[[C2:.*]] = arith.constant 1 : i32
-! CHECK:      fir.store %[[C2]] to %[[X2]] : !fir.ref<i32>
-! CHECK:    omp.terminator
-! CHECK:  }
-! CHECK:      %[[C1:.*]] = arith.constant 1 : i32
-! CHECK:      fir.store %[[C1]] to %[[X1]] : !fir.ref<i32>
-! CHECK:  return
-subroutine multiple_private_fix2()
-   integer :: x
-   !$omp parallel private(x)
-   !$omp parallel private(x)
-      x = 1
-   !$omp end parallel
-      x = 1
-   !$omp end parallel
-      x = 1
-end subroutine
diff --git a/flang/test/Lower/OpenMP/FIR/parallel-private-clause.f90 b/flang/test/Lower/OpenMP/FIR/parallel-private-clause.f90
deleted file mode 100644
index 8b75ecbaae8c..000000000000
--- a/flang/test/Lower/OpenMP/FIR/parallel-private-clause.f90
+++ /dev/null
@@ -1,381 +0,0 @@
-! This test checks lowering of OpenMP parallel Directive with
-! `PRIVATE` clause present.
-
-! REQUIRES: shell
-! RUN: bbc --use-desc-for-alloc=false -fopenmp -emit-fir -hlfir=false %s -o - | \
-! RUN:   FileCheck %s --check-prefix=FIRDialect
-
-!FIRDialect: func @_QPprivate_clause(%[[ARG1:.*]]: !fir.ref<i32>{{.*}}, %[[ARG2:.*]]: !fir.ref<!fir.array<10xi32>>{{.*}}, %[[ARG3:.*]]: !fir.boxchar<1>{{.*}}, %[[ARG4:.*]]: !fir.boxchar<1>{{.*}}) {
-!FIRDialect-DAG: %[[ALPHA:.*]] = fir.alloca i32 {{{.*}}, uniq_name = "{{.*}}Ealpha"}
-!FIRDialect-DAG: %[[ALPHA_ARRAY:.*]] = fir.alloca !fir.array<10xi32> {{{.*}}, uniq_name = "{{.*}}Ealpha_array"}
-!FIRDialect-DAG: %[[BETA:.*]] = fir.alloca !fir.char<1,5> {{{.*}}, uniq_name = "{{.*}}Ebeta"}
-!FIRDialect-DAG: %[[BETA_ARRAY:.*]] = fir.alloca !fir.array<10x!fir.char<1,5>> {{{.*}}, uniq_name = "{{.*}}Ebeta_array"}
-
-!FIRDialect-DAG:  omp.parallel {
-!FIRDialect-DAG: %[[ALPHA_PRIVATE:.*]] = fir.alloca i32 {{{.*}}, pinned, uniq_name = "{{.*}}Ealpha"}
-!FIRDialect-DAG: %[[ALPHA_ARRAY_PRIVATE:.*]] = fir.alloca !fir.array<10xi32> {{{.*}}, pinned, uniq_name = "{{.*}}Ealpha_array"}
-!FIRDialect-DAG: %[[BETA_PRIVATE:.*]] = fir.alloca !fir.char<1,5> {{{.*}}, pinned, uniq_name = "{{.*}}Ebeta"}
-!FIRDialect-DAG: %[[BETA_ARRAY_PRIVATE:.*]] = fir.alloca !fir.array<10x!fir.char<1,5>> {{{.*}}, pinned, uniq_name = "{{.*}}Ebeta_array"}
-!FIRDialect-DAG: %[[ARG1_PRIVATE:.*]] = fir.alloca i32 {{{.*}}, pinned, uniq_name = "{{.*}}Earg1"}
-!FIRDialect-DAG: %[[ARG2_ARRAY_PRIVATE:.*]] = fir.alloca !fir.array<10xi32> {{{.*}}, pinned, uniq_name = "{{.*}}Earg2"}
-!FIRDialect-DAG: %[[ARG3_PRIVATE:.*]] = fir.alloca !fir.char<1,5> {{{.*}}, pinned, uniq_name = "{{.*}}Earg3"}
-!FIRDialect-DAG: %[[ARG4_ARRAY_PRIVATE:.*]] = fir.alloca !fir.array<10x!fir.char<1,5>> {{{.*}}, pinned, uniq_name = "{{.*}}Earg4"}
-!FIRDialect:    omp.terminator
-!FIRDialect:  }
-
-subroutine private_clause(arg1, arg2, arg3, arg4)
-
-        integer :: arg1, arg2(10)
-        integer :: alpha, alpha_array(10)
-        character(5) :: arg3, arg4(10)
-        character(5) :: beta, beta_array(10)
-
-!$OMP PARALLEL PRIVATE(alpha, alpha_array, beta, beta_array, arg1, arg2, arg3, arg4)
-        alpha = 1
-        alpha_array = 4
-        beta = "hi"
-        beta_array = "hi"
-        arg1 = 2
-        arg2 = 3
-        arg3 = "world"
-        arg4 = "world"
-!$OMP END PARALLEL
-
-end subroutine
-
-!FIRDialect: func @_QPprivate_clause_scalar() {
-!FIRDialect-DAG:   {{.*}} = fir.alloca !fir.complex<4> {bindc_name = "c", uniq_name = "{{.*}}Ec"}
-!FIRDialect-DAG:   {{.*}} = fir.alloca i8 {bindc_name = "i1", uniq_name = "{{.*}}Ei1"}
-!FIRDialect-DAG:   {{.*}} = fir.alloca i128 {bindc_name = "i16", uniq_name = "{{.*}}Ei16"}
-!FIRDialect-DAG:   {{.*}} = fir.alloca i16 {bindc_name = "i2", uniq_name = "{{.*}}Ei2"}
-!FIRDialect-DAG:   {{.*}} = fir.alloca i32 {bindc_name = "i4", uniq_name = "{{.*}}Ei4"}
-!FIRDialect-DAG:   {{.*}} = fir.alloca i64 {bindc_name = "i8", uniq_name = "{{.*}}Ei8"}
-!FIRDialect-DAG:   {{.*}} = fir.alloca !fir.logical<4> {bindc_name = "l", uniq_name = "{{.*}}El"}
-!FIRDialect-DAG:   {{.*}} = fir.alloca f32 {bindc_name = "r", uniq_name = "{{.*}}Er"}
-
-!FIRDialect:   omp.parallel {
-!FIRDialect-DAG:     {{.*}} = fir.alloca i8 {bindc_name = "i1", pinned, uniq_name = "{{.*}}Ei1"}
-!FIRDialect-DAG:     {{.*}} = fir.alloca i16 {bindc_name = "i2", pinned, uniq_name = "{{.*}}Ei2"}
-!FIRDialect-DAG:     {{.*}} = fir.alloca i32 {bindc_name = "i4", pinned, uniq_name = "{{.*}}Ei4"}
-!FIRDialect-DAG:     {{.*}} = fir.alloca i64 {bindc_name = "i8", pinned, uniq_name = "{{.*}}Ei8"}
-!FIRDialect-DAG:     {{.*}} = fir.alloca i128 {bindc_name = "i16", pinned, uniq_name = "{{.*}}Ei16"}
-!FIRDialect-DAG:     {{.*}} = fir.alloca !fir.complex<4> {bindc_name = "c", pinned, uniq_name = "{{.*}}Ec"}
-!FIRDialect-DAG:     {{.*}} = fir.alloca !fir.logical<4> {bindc_name = "l", pinned, uniq_name = "{{.*}}El"}
-!FIRDialect-DAG:     {{.*}} = fir.alloca f32 {bindc_name = "r", pinned, uniq_name = "{{.*}}Er"}
-
-subroutine private_clause_scalar()
-
-        integer(kind=1) :: i1
-        integer(kind=2) :: i2
-        integer(kind=4) :: i4
-        integer(kind=8) :: i8
-        integer(kind=16) :: i16
-        complex :: c
-        logical :: l
-        real :: r
-
-!$OMP PARALLEL PRIVATE(i1, i2, i4, i8, i16, c, l, r)
-        print *, i1, i2, i4, i8, i16, c, l, r
-!$OMP END PARALLEL
-
-end subroutine
-
-!FIRDialect: func @_QPprivate_clause_derived_type() {
-!FIRDialect:   {{.*}} = fir.alloca !fir.type<{{.*}}{t_i:i32,t_arr:!fir.array<5xi32>}> {bindc_name = "t", uniq_name = "{{.*}}Et"}
-
-!FIRDialect:   omp.parallel {
-!FIRDialect:     {{.*}} = fir.alloca !fir.type<{{.*}}{t_i:i32,t_arr:!fir.array<5xi32>}> {bindc_name = "t", pinned, uniq_name = "{{.*}}Et"}
-
-subroutine private_clause_derived_type()
-
-        type my_type
-          integer :: t_i
-          integer :: t_arr(5)
-        end type my_type
-        type(my_type) :: t
-
-!$OMP PARALLEL PRIVATE(t)
-        print *, t%t_i
-!$OMP END PARALLEL
-
-end subroutine
-
-!FIRDialect: func @_QPprivate_clause_allocatable() {
-!FIRDialect-DAG:  {{.*}} = fir.alloca !fir.box<!fir.heap<i32>> {bindc_name = "x", uniq_name = "{{.*}}Ex"}
-!FIRDialect-DAG:  {{.*}} = fir.alloca !fir.heap<i32> {uniq_name = "{{.*}}Ex.addr"}
-!FIRDialect-DAG:  {{.*}} = fir.alloca !fir.box<!fir.heap<!fir.array<?xi32>>> {bindc_name = "x2", uniq_name = "{{.*}}Ex2"}
-!FIRDialect-DAG:  {{.*}} = fir.alloca !fir.heap<!fir.array<?xi32>> {uniq_name = "{{.*}}Ex2.addr"}
-!FIRDialect-DAG:  {{.*}} = fir.address_of(@{{.*}}Ex3) : !fir.ref<!fir.box<!fir.heap<i32>>>
-!FIRDialect-DAG:  [[TMP8:%.*]] = fir.address_of(@{{.*}}Ex4) : !fir.ref<!fir.box<!fir.heap<!fir.array<?xi32>>>>
-
-!FIRDialect:   omp.parallel {
-!FIRDialect-DAG:    [[TMP35:%.*]] = fir.alloca !fir.box<!fir.heap<i32>> {bindc_name = "x", pinned, uniq_name = "{{.*}}Ex"}
-!FIRDialect-DAG:    [[TMP39:%.*]] = fir.alloca !fir.box<!fir.heap<!fir.array<?xi32>>> {bindc_name = "x2", pinned, uniq_name = "{{.*}}Ex2"}
-!FIRDialect-DAG:    [[TMP45:%.*]] = fir.alloca !fir.box<!fir.heap<i32>> {bindc_name = "x3", pinned, uniq_name = "{{.*}}Ex3"}
-
-!FIRDialect-DAG:    [[TMP51:%.*]] = fir.load [[TMP8]] : !fir.ref<!fir.box<!fir.heap<!fir.array<?xi32>>>>
-!FIRDialect-DAG:    [[TMP97:%.*]] = fir.load [[TMP8]] : !fir.ref<!fir.box<!fir.heap<!fir.array<?xi32>>>>
-!FIRDialect-DAG:    [[TMP98:%.*]]:3 = fir.box_dims [[TMP97]], {{.*}} : (!fir.box<!fir.heap<!fir.array<?xi32>>>, index) -> (index, index, index)
-!FIRDialect-DAG:    [[TMP50:%.*]] = fir.alloca !fir.box<!fir.heap<!fir.array<?xi32>>> {bindc_name = "x4", pinned, uniq_name = "{{.*}}Ex4"}
-
-! FIRDialect-DAG:    [[TMP101:%.*]] = fir.allocmem !fir.array<?xi32>, {{.*}} {fir.must_be_heap = true, uniq_name = "{{.*}}Ex4.alloc"}
-! FIRDialect-DAG:    [[TMP102:%.*]] = fir.shape_shift {{.*}}#0, {{.*}} : (index, index) -> !fir.shapeshift<1>
-! FIRDialect-DAG:    [[TMP103:%.*]] = fir.embox [[TMP101]]([[TMP102]]) : (!fir.heap<!fir.array<?xi32>>, !fir.shapeshift<1>) -> !fir.box<!fir.heap<!fir.array<?xi32>>>
-! FIRDialect-DAG:  fir.store [[TMP103]] to [[TMP50]] : !fir.ref<!fir.box<!fir.heap<!fir.array<?xi32>>>>
-
-
-subroutine private_clause_allocatable()
-
-        integer, allocatable :: x, x2(:)
-        integer, allocatable, save :: x3, x4(:)
-
-        print *, x, x2, x3, x4
-
-!$OMP PARALLEL PRIVATE(x, x2, x3, x4)
-        print *, x, x2, x3, x4
-!$OMP END PARALLEL
-
-end subroutine
-
-
-!FIRDialect: func @_QPprivate_clause_real_call_allocatable() {
-!FIRDialect-DAG: {{.*}} = fir.alloca !fir.box<!fir.heap<f32>> {bindc_name = "x5", uniq_name = "{{.*}}Ex5"}
-!FIRDialect-DAG: {{.*}} = fir.zero_bits !fir.heap<f32>
-!FIRDialect-DAG: {{.*}} = fir.embox %1 : (!fir.heap<f32>) -> !fir.box<!fir.heap<f32>>
-!FIRDialect-DAG: fir.store %2 to %0 : !fir.ref<!fir.box<!fir.heap<f32>>>
-!FIRDialect-DAG: omp.parallel   {
-!FIRDialect-DAG:  [[TMP203:%.*]] = fir.alloca !fir.box<!fir.heap<f32>> {bindc_name = "x5", pinned, uniq_name = "{{.*}}Ex5"}
-
-!FIRDialect-DAG: fir.if %{{.*}} {
-
-!FIRDialect-DAG:   fir.store %{{.*}} to [[TMP203]] : !fir.ref<!fir.box<!fir.heap<f32>>>
-!FIRDialect-DAG: } else {
-
-!FIRDialect-DAG:   fir.store %{{.*}} to [[TMP203]] : !fir.ref<!fir.box<!fir.heap<f32>>>
-!FIRDialect-DAG: }
-!FIRDialect-DAG: fir.call @_QFprivate_clause_real_call_allocatablePhelper_private_clause_real_call_allocatable([[TMP203]]) fastmath<contract> : (!fir.ref<!fir.box<!fir.heap<f32>>>) -> ()
-!FIRDialect-DAG: %{{.*}} = fir.load [[TMP203]] : !fir.ref<!fir.box<!fir.heap<f32>>>
-
-!FIRDialect-DAG: fir.if %{{.*}} {
-!FIRDialect-DAG:   %{{.*}} = fir.load [[TMP203]] : !fir.ref<!fir.box<!fir.heap<f32>>>
-
-!FIRDialect-DAG:     fir.store %{{.*}} to [[TMP203]] : !fir.ref<!fir.box<!fir.heap<f32>>>
-!FIRDialect-DAG:   }
-!FIRDialect-DAG:   omp.terminator
-!FIRDialect-DAG:   }
-!FIRDialect-DAG:   return
-!FIRDialect-DAG: }
-
-
-subroutine private_clause_real_call_allocatable
-        real, allocatable :: x5
-        !$omp parallel private(x5)
-            call helper_private_clause_real_call_allocatable(x5)
-        !$omp end parallel
-    contains
-        subroutine helper_private_clause_real_call_allocatable(x6)
-            real, allocatable :: x6
-            print *, allocated(x6)
-        end subroutine
-end subroutine
-
-!FIRDialect:  func.func @_QPincrement_list_items(%arg0: !fir.ref<!fir.box<!fir.ptr<!fir.type<_QFincrement_list_itemsTnode{payload:i32,next:!fir.box<!fir.ptr<!fir.type<_QFincrement_list_itemsTnode>>>}>>>> {fir.bindc_name = "head"}) {
-!FIRDialect:    {{%.*}} = fir.alloca !fir.box<!fir.ptr<!fir.type<_QFincrement_list_itemsTnode{payload:i32,next:!fir.box<!fir.ptr<!fir.type<_QFincrement_list_itemsTnode>>>}>>> {bindc_name = "p", uniq_name = "_QFincrement_list_itemsEp"}
-!FIRDialect:    omp.parallel   {
-!FIRDialect:      {{%.*}} = fir.alloca !fir.box<!fir.ptr<!fir.type<_QFincrement_list_itemsTnode{payload:i32,next:!fir.box<!fir.ptr<!fir.type<_QFincrement_list_itemsTnode>>>}>>> {bindc_name = "p", pinned, uniq_name = "_QFincrement_list_itemsEp"}
-!FIRDialect:      omp.single   {
-
-!FIRDialect:         omp.terminator
-!FIRDialect:       omp.terminator
-!FIRDialect:    return
-
-subroutine increment_list_items (head)
-  type node
-     integer :: payload
-     type (node), pointer :: next
-  end type node
-
-  type (node), pointer :: head
-  type (node), pointer :: p
-!$omp parallel private(p)
-!$omp single
-  p => head
-  do
-     p => p%next
-     if ( associated (p) .eqv. .false. ) exit
-  end do
-!$omp end single
-!$omp end parallel
-end subroutine increment_list_items
-
-!FIRDialect:  func.func @_QPparallel_pointer() {
-!FIRDialect-DAG: [[PP0:%.*]]  = fir.alloca !fir.box<!fir.ptr<i32>> {bindc_name = "y1", uniq_name = "{{.*}}Ey1"}
-!FIRDialect-DAG: [[PP1:%.*]]  = fir.alloca !fir.ptr<i32> {uniq_name = "{{.*}}Ey1.addr"}
-!FIRDialect-DAG: [[PP2:%.*]]  = fir.zero_bits !fir.ptr<i32>
-!FIRDialect:     fir.store [[PP2]] to [[PP1]] : !fir.ref<!fir.ptr<i32>>
-!FIRDialect-DAG: [[PP3:%.*]]  = fir.alloca !fir.box<!fir.ptr<!fir.array<?xi32>>> {bindc_name = "y2", uniq_name = "{{.*}}Ey2"}
-
-!FIRDialect:     fir.store %6 to %3 : !fir.ref<!fir.box<!fir.ptr<!fir.array<?xi32>>>>
-!FIRDialect-DAG: [[PP7:%.*]] = fir.alloca i32 {bindc_name = "z1", fir.target, uniq_name = "{{.*}}Ez1"}
-
-!FIRDialect-DAG: [[PP8:%.*]] = fir.alloca !fir.array<10xi32> {bindc_name = "z2", fir.target, uniq_name = "{{.*}}Ez2"}
-!FIRDialect:     omp.parallel   {
-!FIRDialect-DAG:   [[PP9:%.*]] = fir.alloca !fir.box<!fir.ptr<i32>> {bindc_name = "y1", pinned, uniq_name = "{{.*}}Ey1"}
-!FIRDialect-DAG:   [[PP10:%.*]] = fir.alloca !fir.box<!fir.ptr<!fir.array<?xi32>>> {bindc_name = "y2", pinned, uniq_name = "{{.*}}Ey2"}
-!FIRDialect-DAG:   [[PP11:%.*]] = fir.embox [[PP7]] : (!fir.ref<i32>) -> !fir.box<!fir.ptr<i32>>
-!FIRDialect:       fir.store [[PP11]] to [[PP9]] : !fir.ref<!fir.box<!fir.ptr<i32>>>
-!FIRDialect-DAG:   [[PP12:%.*]] = fir.shape %c{{.*}} : (index) -> !fir.shape<1>
-!FIRDialect-DAG:   [[PP13:%.*]] = fir.embox [[PP8]]([[PP12]]) : (!fir.ref<!fir.array<10xi32>>, !fir.shape<1>) -> !fir.box<!fir.ptr<!fir.array<?xi32>>>
-!FIRDialect:       fir.store %13 to [[PP10]] : !fir.ref<!fir.box<!fir.ptr<!fir.array<?xi32>>>>
-!FIRDialect:       omp.terminator
-!FIRDialect:     }
-!FIRDialect:   return
-!FIRDialect: }
-
-subroutine parallel_pointer()
-    integer, pointer :: y1, y2(:)
-    integer, target :: z1, z2(10)
-
-!$omp parallel private(y1, y2)
-  y1=>z1
-  y2=>z2
-!$omp end parallel
-end subroutine parallel_pointer
-
-
-!FIRDialect-LABEL: func @_QPsimple_loop_1()
-subroutine simple_loop_1
-  integer :: i
-  real, allocatable :: r;
-  ! FIRDialect:  omp.parallel
-  !$OMP PARALLEL PRIVATE(r)
-  ! FIRDialect:     %[[ALLOCA_IV:.*]] = fir.alloca i32 {{{.*}}, pinned}
-
-  ! FIRDialect:     [[R:%.*]] = fir.alloca !fir.box<!fir.heap<f32>> {bindc_name = "r", pinned, uniq_name = "{{.*}}Er"}
-  ! FIRDialect:     fir.store {{%.*}} to [[R]] : !fir.ref<!fir.box<!fir.heap<f32>>>
-  ! FIRDialect:     fir.store {{%.*}} to [[R]] : !fir.ref<!fir.box<!fir.heap<f32>>>
-
-  ! FIRDialect:     %[[WS_LB:.*]] = arith.constant 1 : i32
-  ! FIRDialect:     %[[WS_UB:.*]] = arith.constant 9 : i32
-  ! FIRDialect:     %[[WS_STEP:.*]] = arith.constant 1 : i32
-
-  ! FIRDialect:     omp.wsloop for (%[[I:.*]]) : i32 = (%[[WS_LB]]) to (%[[WS_UB]]) inclusive step (%[[WS_STEP]])
-  !$OMP DO
-  do i=1, 9
-  ! FIRDialect:     fir.store %[[I]] to %[[ALLOCA_IV:.*]] : !fir.ref<i32>
-  ! FIRDialect:     %[[LOAD_IV:.*]] = fir.load %[[ALLOCA_IV]] : !fir.ref<i32>
-  ! FIRDialect:     fir.call @_FortranAioOutputInteger32({{.*}}, %[[LOAD_IV]]) {{.*}}: (!fir.ref<i8>, i32) -> i1
-    print*, i
-  end do
-  ! FIRDialect:     omp.yield
-  ! FIRDialect:     {{%.*}} = fir.load [[R]] : !fir.ref<!fir.box<!fir.heap<f32>>>
-  ! FIRDialect:     fir.if {{%.*}} {
-  ! FIRDialect:     [[LD:%.*]] = fir.load [[R]] : !fir.ref<!fir.box<!fir.heap<f32>>>
-  ! FIRDialect:     [[AD:%.*]] = fir.box_addr [[LD]] : (!fir.box<!fir.heap<f32>>) -> !fir.heap<f32>
-  ! FIRDialect:     fir.freemem [[AD]] : !fir.heap<f32>
-  ! FIRDialect:     fir.store {{%.*}} to [[R]] : !fir.ref<!fir.box<!fir.heap<f32>>>
-  !$OMP END DO
-  ! FIRDialect:  omp.terminator
-  !$OMP END PARALLEL
-end subroutine
-
-!FIRDialect-LABEL: func @_QPsimple_loop_2()
-subroutine simple_loop_2
-  integer :: i
-  real, allocatable :: r;
-  ! FIRDialect:  omp.parallel
-  !$OMP PARALLEL
-  ! FIRDialect:     %[[ALLOCA_IV:.*]] = fir.alloca i32 {{{.*}}, pinned}
-
-  ! FIRDialect:     [[R:%.*]] = fir.alloca !fir.box<!fir.heap<f32>> {bindc_name = "r", pinned, uniq_name = "{{.*}}Er"}
-  ! FIRDialect:     fir.store {{%.*}} to [[R]] : !fir.ref<!fir.box<!fir.heap<f32>>>
-  ! FIRDialect:     fir.store {{%.*}} to [[R]] : !fir.ref<!fir.box<!fir.heap<f32>>>
-
-  ! FIRDialect:     %[[WS_LB:.*]] = arith.constant 1 : i32
-  ! FIRDialect:     %[[WS_UB:.*]] = arith.constant 9 : i32
-  ! FIRDialect:     %[[WS_STEP:.*]] = arith.constant 1 : i32
-
-  ! FIRDialect:     omp.wsloop for (%[[I:.*]]) : i32 = (%[[WS_LB]]) to (%[[WS_UB]]) inclusive step (%[[WS_STEP]])
-  !$OMP DO PRIVATE(r)
-  do i=1, 9
-  ! FIRDialect:     fir.store %[[I]] to %[[ALLOCA_IV:.*]] : !fir.ref<i32>
-  ! FIRDialect:     %[[LOAD_IV:.*]] = fir.load %[[ALLOCA_IV]] : !fir.ref<i32>
-  ! FIRDialect:     fir.call @_FortranAioOutputInteger32({{.*}}, %[[LOAD_IV]]) {{.*}}: (!fir.ref<i8>, i32) -> i1
-    print*, i
-  end do
-  ! FIRDialect:     omp.yield
-  ! FIRDialect:     {{%.*}} = fir.load [[R]] : !fir.ref<!fir.box<!fir.heap<f32>>>
-  ! FIRDialect:     fir.if {{%.*}} {
-  ! FIRDialect:     [[LD:%.*]] = fir.load [[R]] : !fir.ref<!fir.box<!fir.heap<f32>>>
-  ! FIRDialect:     [[AD:%.*]] = fir.box_addr [[LD]] : (!fir.box<!fir.heap<f32>>) -> !fir.heap<f32>
-  ! FIRDialect:     fir.freemem [[AD]] : !fir.heap<f32>
-  ! FIRDialect:     fir.store {{%.*}} to [[R]] : !fir.ref<!fir.box<!fir.heap<f32>>>
-  !$OMP END DO
-  ! FIRDialect:  omp.terminator
-  !$OMP END PARALLEL
-end subroutine
-
-!FIRDialect-LABEL: func @_QPsimple_loop_3()
-subroutine simple_loop_3
-  integer :: i
-  real, allocatable :: r;
-  ! FIRDialect:  omp.parallel
-  ! FIRDialect:     %[[ALLOCA_IV:.*]] = fir.alloca i32 {{{.*}}, pinned}
-
-  ! FIRDialect:     [[R:%.*]] = fir.alloca !fir.box<!fir.heap<f32>> {bindc_name = "r", pinned, uniq_name = "{{.*}}Er"}
-  ! FIRDialect:     fir.store {{%.*}} to [[R]] : !fir.ref<!fir.box<!fir.heap<f32>>>
-  ! FIRDialect:     fir.store {{%.*}} to [[R]] : !fir.ref<!fir.box<!fir.heap<f32>>>
-
-  ! FIRDialect:     %[[WS_LB:.*]] = arith.constant 1 : i32
-  ! FIRDialect:     %[[WS_UB:.*]] = arith.constant 9 : i32
-  ! FIRDialect:     %[[WS_STEP:.*]] = arith.constant 1 : i32
-
-  ! FIRDialect:     omp.wsloop for (%[[I:.*]]) : i32 = (%[[WS_LB]]) to (%[[WS_UB]]) inclusive step (%[[WS_STEP]])
-  !$OMP PARALLEL DO PRIVATE(r)
-  do i=1, 9
-  ! FIRDialect:     fir.store %[[I]] to %[[ALLOCA_IV:.*]] : !fir.ref<i32>
-  ! FIRDialect:     %[[LOAD_IV:.*]] = fir.load %[[ALLOCA_IV]] : !fir.ref<i32>
-  ! FIRDialect:     fir.call @_FortranAioOutputInteger32({{.*}}, %[[LOAD_IV]]) {{.*}}: (!fir.ref<i8>, i32) -> i1
-    print*, i
-  end do
-  ! FIRDialect:     omp.yield
-  ! FIRDialect:     {{%.*}} = fir.load [[R]] : !fir.ref<!fir.box<!fir.heap<f32>>>
-  ! FIRDialect:     fir.if {{%.*}} {
-  ! FIRDialect:     [[LD:%.*]] = fir.load [[R]] : !fir.ref<!fir.box<!fir.heap<f32>>>
-  ! FIRDialect:     [[AD:%.*]] = fir.box_addr [[LD]] : (!fir.box<!fir.heap<f32>>) -> !fir.heap<f32>
-  ! FIRDialect:     fir.freemem [[AD]] : !fir.heap<f32>
-  ! FIRDialect:     fir.store {{%.*}} to [[R]] : !fir.ref<!fir.box<!fir.heap<f32>>>
-  !$OMP END PARALLEL DO
-  ! FIRDialect:  omp.terminator
-end subroutine
-
-!CHECK-LABEL: func @_QPsimd_loop_1()
-subroutine simd_loop_1
-  integer :: i
-  real, allocatable :: r;
-  ! FIRDialect:     [[R:%.*]] = fir.alloca !fir.box<!fir.heap<f32>> {bindc_name = "r", pinned, uniq_name = "{{.*}}Er"}
-  ! FIRDialect:     fir.store {{%.*}} to [[R]] : !fir.ref<!fir.box<!fir.heap<f32>>>
-  ! FIRDialect:     fir.store {{%.*}} to [[R]] : !fir.ref<!fir.box<!fir.heap<f32>>>
-
-  ! FIRDialect:     %[[LB:.*]] = arith.constant 1 : i32
-  ! FIRDialect:     %[[UB:.*]] = arith.constant 9 : i32
-  ! FIRDialect:     %[[STEP:.*]] = arith.constant 1 : i32
-
-  ! FIRDialect: omp.simd {
-  ! FIRDialect-NEXT: omp.loop_nest (%[[I:.*]]) : i32 = (%[[LB]]) to (%[[UB]]) inclusive step (%[[STEP]]) {
-  !$OMP SIMD PRIVATE(r)
-  do i=1, 9
-  ! FIRDialect:     fir.store %[[I]] to %[[LOCAL:.*]] : !fir.ref<i32>
-  ! FIRDialect:     %[[LOAD_IV:.*]] = fir.load %[[LOCAL]] : !fir.ref<i32>
-  ! FIRDialect:     fir.call @_FortranAioOutputInteger32({{.*}}, %[[LOAD_IV]]) {{.*}}: (!fir.ref<i8>, i32) -> i1
-    print*, i
-  end do
-  !$OMP END SIMD
-  ! FIRDialect:     omp.yield
-  ! FIRDialect:     {{%.*}} = fir.load [[R]] : !fir.ref<!fir.box<!fir.heap<f32>>>
-  ! FIRDialect:     fir.if {{%.*}} {
-  ! FIRDialect:     [[LD:%.*]] = fir.load [[R]] : !fir.ref<!fir.box<!fir.heap<f32>>>
-  ! FIRDialect:     [[AD:%.*]] = fir.box_addr [[LD]] : (!fir.box<!fir.heap<f32>>) -> !fir.heap<f32>
-  ! FIRDialect:     fir.freemem [[AD]] : !fir.heap<f32>
-  ! FIRDialect:     fir.store {{%.*}} to [[R]] : !fir.ref<!fir.box<!fir.heap<f32>>>
-end subroutine
diff --git a/flang/test/Lower/OpenMP/FIR/parallel-reduction-add-byref.f90 b/flang/test/Lower/OpenMP/FIR/parallel-reduction-add-byref.f90
deleted file mode 100644
index ea45e716ceae..000000000000
--- a/flang/test/Lower/OpenMP/FIR/parallel-reduction-add-byref.f90
+++ /dev/null
@@ -1,117 +0,0 @@
-! RUN: bbc -emit-fir -hlfir=false -fopenmp --force-byref-reduction -o - %s 2>&1 | FileCheck %s
-! RUN: %flang_fc1 -emit-fir -flang-deprecated-no-hlfir -fopenmp -mmlir --force-byref-reduction -o - %s 2>&1 | FileCheck %s
-
-!CHECK-LABEL: omp.declare_reduction
-!CHECK-SAME: @[[RED_F32_NAME:.*]] : !fir.ref<f32>
-!CHECK-SAME: init {
-!CHECK: ^bb0(%{{.*}}: !fir.ref<f32>):
-!CHECK:  %[[C0_1:.*]] = arith.constant 0.000000e+00 : f32
-!CHECK:  %[[REF:.*]] = fir.alloca f32
-!CHECKL  fir.store [[%C0_1]] to %[[REF]] : !fir.ref<f32>
-!CHECK:  omp.yield(%[[REF]] : !fir.ref<f32>)
-!CHECK: } combiner {
-!CHECK: ^bb0(%[[ARG0:.*]]: !fir.ref<f32>, %[[ARG1:.*]]: !fir.ref<f32>):
-!CHECK:  %[[LD0:.*]] = fir.load %[[ARG0]] : !fir.ref<f32>
-!CHECK:  %[[LD1:.*]] = fir.load %[[ARG1]] : !fir.ref<f32>
-!CHECK:  %[[RES:.*]] = arith.addf %[[LD0]], %[[LD1]] {{.*}}: f32
-!CHECK:  fir.store %[[RES]] to %[[ARG0]] : !fir.ref<f32>
-!CHECK:  omp.yield(%[[ARG0]] : !fir.ref<f32>)
-!CHECK: }
-
-!CHECK-LABEL: omp.declare_reduction
-!CHECK-SAME: @[[RED_I32_NAME:.*]] : !fir.ref<i32>
-!CHECK-SAME: init {
-!CHECK: ^bb0(%{{.*}}: !fir.ref<i32>):
-!CHECK:  %[[C0_1:.*]] = arith.constant 0 : i32
-!CHECK:  %[[REF:.*]] = fir.alloca i32
-!CHECKL  fir.store [[%C0_1]] to %[[REF]] : !fir.ref<i32>
-!CHECK:  omp.yield(%[[REF]] : !fir.ref<i32>)
-!CHECK: } combiner {
-!CHECK: ^bb0(%[[ARG0:.*]]: !fir.ref<i32>, %[[ARG1:.*]]: !fir.ref<i32>):
-!CHECK:  %[[LD0:.*]] = fir.load %[[ARG0]] : !fir.ref<i32>
-!CHECK:  %[[LD1:.*]] = fir.load %[[ARG1]] : !fir.ref<i32>
-!CHECK:  %[[RES:.*]] = arith.addi %[[LD0]], %[[LD1]] : i32
-!CHECK:  fir.store %[[RES]] to %[[ARG0]] : !fir.ref<i32>
-!CHECK:  omp.yield(%[[ARG0]] : !fir.ref<i32>)
-!CHECK: }
-
-!CHECK-LABEL: func.func @_QPsimple_int_add
-!CHECK:  %[[IREF:.*]] = fir.alloca i32 {bindc_name = "i", uniq_name = "_QFsimple_int_addEi"}
-!CHECK:  %[[I_START:.*]] = arith.constant 0 : i32
-!CHECK:  fir.store %[[I_START]] to %[[IREF]] : !fir.ref<i32>
-!CHECK:  omp.parallel byref reduction(@[[RED_I32_NAME]] %[[IREF]] -> %[[PRV:.+]] : !fir.ref<i32>) {
-!CHECK:    %[[LPRV:.+]] = fir.load %[[PRV]] : !fir.ref<i32>
-!CHECK:    %[[I_INCR:.+]] = arith.constant 1 : i32
-!CHECK:    %[[RES:.+]] = arith.addi %[[LPRV]], %[[I_INCR]]
-!CHECK:    fir.store %[[RES]] to %[[PRV]] : !fir.ref<i32>
-!CHECK:    omp.terminator
-!CHECK:  }
-!CHECK: return
-subroutine simple_int_add
-    integer :: i
-    i = 0
-
-    !$omp parallel reduction(+:i)
-    i = i + 1
-    !$omp end parallel
-
-    print *, i
-end subroutine
-
-!CHECK-LABEL: func.func @_QPsimple_real_add
-!CHECK:  %[[RREF:.*]] = fir.alloca f32 {bindc_name = "r", uniq_name = "_QFsimple_real_addEr"}
-!CHECK:  %[[R_START:.*]] = arith.constant 0.000000e+00 : f32
-!CHECK:  fir.store %[[R_START]] to %[[RREF]] : !fir.ref<f32>
-!CHECK:  omp.parallel byref reduction(@[[RED_F32_NAME]] %[[RREF]] -> %[[PRV:.+]] : !fir.ref<f32>) {
-!CHECK:    %[[LPRV:.+]] = fir.load %[[PRV]] : !fir.ref<f32>
-!CHECK:    %[[R_INCR:.+]] = arith.constant 1.500000e+00 : f32
-!CHECK:    %[[RES]] = arith.addf %[[LPRV]], %[[R_INCR]] {{.*}} : f32
-!CHECK:    fir.store %[[RES]] to %[[PRV]] : !fir.ref<f32>
-!CHECK:    omp.terminator
-!CHECK:  }
-!CHECK: return
-subroutine simple_real_add
-    real :: r
-    r = 0.0
-
-    !$omp parallel reduction(+:r)
-    r = r + 1.5
-    !$omp end parallel
-
-    print *, r
-end subroutine
-
-!CHECK-LABEL: func.func @_QPint_real_add
-!CHECK:  %[[IREF:.*]] = fir.alloca i32 {bindc_name = "i", uniq_name = "_QFint_real_addEi"}
-!CHECK:  %[[RREF:.*]] = fir.alloca f32 {bindc_name = "r", uniq_name = "_QFint_real_addEr"}
-!CHECK:  %[[R_START:.*]] = arith.constant 0.000000e+00 : f32
-!CHECK:  fir.store %[[R_START]] to %[[RREF]] : !fir.ref<f32>
-!CHECK:  %[[I_START:.*]] = arith.constant 0 : i32
-!CHECK:  fir.store %[[I_START]] to %[[IREF]] : !fir.ref<i32>
-!CHECK:  omp.parallel byref reduction(@[[RED_I32_NAME]] %[[IREF]] -> %[[PRV0:.+]] : !fir.ref<i32>, @[[RED_F32_NAME]] %[[RREF]] -> %[[PRV1:.+]] : !fir.ref<f32>) {
-!CHECK:    %[[R_INCR:.*]] = arith.constant 1.500000e+00 : f32
-!CHECK:    %[[LPRV1:.+]] = fir.load %[[PRV1]] : !fir.ref<f32>
-!CHECK:    %[[RES1:.+]] = arith.addf %[[R_INCR]], %[[LPRV1]] {{.*}} : f32
-!CHECK:    fir.store %[[RES1]] to %[[PRV1]]
-!CHECK:    %[[LPRV0:.+]] = fir.load %[[PRV0]] : !fir.ref<i32>
-!CHECK:    %[[I_INCR:.*]] = arith.constant 3 : i32
-!CHECK:    %[[RES0:.+]] = arith.addi %[[LPRV0]], %[[I_INCR]]
-!CHECK:    fir.store %[[RES0]] to %[[PRV0]]
-!CHECK:    omp.terminator
-!CHECK:  }
-!CHECK: return
-subroutine int_real_add
-    real :: r
-    integer :: i
-
-    r = 0.0
-    i = 0
-
-    !$omp parallel reduction(+:i,r)
-    r = 1.5 + r
-    i = i + 3
-    !$omp end parallel
-
-    print *, r
-    print *, i
-end subroutine
diff --git a/flang/test/Lower/OpenMP/FIR/parallel-reduction-add.f90 b/flang/test/Lower/OpenMP/FIR/parallel-reduction-add.f90
deleted file mode 100644
index 3f6d9e647c9b..000000000000
--- a/flang/test/Lower/OpenMP/FIR/parallel-reduction-add.f90
+++ /dev/null
@@ -1,105 +0,0 @@
-! RUN: bbc -emit-fir -hlfir=false -fopenmp -o - %s 2>&1 | FileCheck %s
-! RUN: %flang_fc1 -emit-fir -flang-deprecated-no-hlfir -fopenmp -o - %s 2>&1 | FileCheck %s
-
-!CHECK-LABEL: omp.declare_reduction
-!CHECK-SAME: @[[RED_F32_NAME:.*]] : f32 init {
-!CHECK: ^bb0(%{{.*}}: f32):
-!CHECK:  %[[C0_1:.*]] = arith.constant 0.000000e+00 : f32
-!CHECK:  omp.yield(%[[C0_1]] : f32)
-!CHECK: } combiner {
-!CHECK: ^bb0(%[[ARG0:.*]]: f32, %[[ARG1:.*]]: f32):
-!CHECK:  %[[RES:.*]] = arith.addf %[[ARG0]], %[[ARG1]] {{.*}}: f32
-!CHECK:  omp.yield(%[[RES]] : f32)
-!CHECK: }
-
-!CHECK-LABEL: omp.declare_reduction
-!CHECK-SAME: @[[RED_I32_NAME:.*]] : i32 init {
-!CHECK: ^bb0(%{{.*}}: i32):
-!CHECK:  %[[C0_1:.*]] = arith.constant 0 : i32
-!CHECK:  omp.yield(%[[C0_1]] : i32)
-!CHECK: } combiner {
-!CHECK: ^bb0(%[[ARG0:.*]]: i32, %[[ARG1:.*]]: i32):
-!CHECK:  %[[RES:.*]] = arith.addi %[[ARG0]], %[[ARG1]] : i32
-!CHECK:  omp.yield(%[[RES]] : i32)
-!CHECK: }
-
-!CHECK-LABEL: func.func @_QPsimple_int_add
-!CHECK:  %[[IREF:.*]] = fir.alloca i32 {bindc_name = "i", uniq_name = "_QFsimple_int_addEi"}
-!CHECK:  %[[I_START:.*]] = arith.constant 0 : i32
-!CHECK:  fir.store %[[I_START]] to %[[IREF]] : !fir.ref<i32>
-!CHECK:  omp.parallel reduction(@[[RED_I32_NAME]] %[[IREF]] -> %[[PRV:.+]] : !fir.ref<i32>) {
-!CHECK:    %[[LPRV:.+]] = fir.load %[[PRV]] : !fir.ref<i32>
-!CHECK:    %[[I_INCR:.+]] = arith.constant 1 : i32
-!CHECK:    %[[RES:.+]] = arith.addi %[[LPRV]], %[[I_INCR]]
-!CHECK:    fir.store %[[RES]] to %[[PRV]] : !fir.ref<i32>
-!CHECK:    omp.terminator
-!CHECK:  }
-!CHECK: return
-subroutine simple_int_add
-    integer :: i
-    i = 0
-
-    !$omp parallel reduction(+:i)
-    i = i + 1
-    !$omp end parallel
-
-    print *, i
-end subroutine
-
-!CHECK-LABEL: func.func @_QPsimple_real_add
-!CHECK:  %[[RREF:.*]] = fir.alloca f32 {bindc_name = "r", uniq_name = "_QFsimple_real_addEr"}
-!CHECK:  %[[R_START:.*]] = arith.constant 0.000000e+00 : f32
-!CHECK:  fir.store %[[R_START]] to %[[RREF]] : !fir.ref<f32>
-!CHECK:  omp.parallel reduction(@[[RED_F32_NAME]] %[[RREF]] -> %[[PRV:.+]] : !fir.ref<f32>) {
-!CHECK:    %[[LPRV:.+]] = fir.load %[[PRV]] : !fir.ref<f32>
-!CHECK:    %[[R_INCR:.+]] = arith.constant 1.500000e+00 : f32
-!CHECK:    %[[RES]] = arith.addf %[[LPRV]], %[[R_INCR]] {{.*}} : f32
-!CHECK:    fir.store %[[RES]] to %[[PRV]] : !fir.ref<f32>
-!CHECK:    omp.terminator
-!CHECK:  }
-!CHECK: return
-subroutine simple_real_add
-    real :: r
-    r = 0.0
-
-    !$omp parallel reduction(+:r)
-    r = r + 1.5
-    !$omp end parallel
-
-    print *, r
-end subroutine
-
-!CHECK-LABEL: func.func @_QPint_real_add
-!CHECK:  %[[IREF:.*]] = fir.alloca i32 {bindc_name = "i", uniq_name = "_QFint_real_addEi"}
-!CHECK:  %[[RREF:.*]] = fir.alloca f32 {bindc_name = "r", uniq_name = "_QFint_real_addEr"}
-!CHECK:  %[[R_START:.*]] = arith.constant 0.000000e+00 : f32
-!CHECK:  fir.store %[[R_START]] to %[[RREF]] : !fir.ref<f32>
-!CHECK:  %[[I_START:.*]] = arith.constant 0 : i32
-!CHECK:  fir.store %[[I_START]] to %[[IREF]] : !fir.ref<i32>
-!CHECK:  omp.parallel reduction(@[[RED_I32_NAME]] %[[IREF]] -> %[[PRV0:.+]] : !fir.ref<i32>, @[[RED_F32_NAME]] %[[RREF]] -> %[[PRV1:.+]] : !fir.ref<f32>) {
-!CHECK:    %[[R_INCR:.*]] = arith.constant 1.500000e+00 : f32
-!CHECK:    %[[LPRV1:.+]] = fir.load %[[PRV1]] : !fir.ref<f32>
-!CHECK:    %[[RES1:.+]] = arith.addf %[[R_INCR]], %[[LPRV1]] {{.*}} : f32
-!CHECK:    fir.store %[[RES1]] to %[[PRV1]]
-!CHECK:    %[[LPRV0:.+]] = fir.load %[[PRV0]] : !fir.ref<i32>
-!CHECK:    %[[I_INCR:.*]] = arith.constant 3 : i32
-!CHECK:    %[[RES0:.+]] = arith.addi %[[LPRV0]], %[[I_INCR]]
-!CHECK:    fir.store %[[RES0]] to %[[PRV0]]
-!CHECK:    omp.terminator
-!CHECK:  }
-!CHECK: return
-subroutine int_real_add
-    real :: r
-    integer :: i
-
-    r = 0.0
-    i = 0
-
-    !$omp parallel reduction(+:i,r)
-    r = 1.5 + r
-    i = i + 3
-    !$omp end parallel
-
-    print *, r
-    print *, i
-end subroutine
diff --git a/flang/test/Lower/OpenMP/FIR/parallel-sections.f90 b/flang/test/Lower/OpenMP/FIR/parallel-sections.f90
deleted file mode 100644
index 0c0834cfafe9..000000000000
--- a/flang/test/Lower/OpenMP/FIR/parallel-sections.f90
+++ /dev/null
@@ -1,65 +0,0 @@
-! REQUIRES: openmp_runtime
-
-!RUN: %flang_fc1 -emit-fir -flang-deprecated-no-hlfir -fopenmp %s -o - | FileCheck %s --check-prefixes="FIRDialect,OMPDialect"
-!RUN: %flang_fc1 -emit-fir -flang-deprecated-no-hlfir -fopenmp %s -o - | fir-opt --cfg-conversion-on-func-opt | fir-opt --fir-to-llvm-ir | FileCheck %s --check-prefixes="OMPDialect,LLVMDialect"
-
-!===============================================================================
-! Parallel sections construct
-!===============================================================================
-
-!FIRDialect: func @_QPomp_parallel_sections
-subroutine omp_parallel_sections(x, y)
-  integer, intent(inout) :: x, y
-  !OMPDialect: omp.parallel {
-  !OMPDialect: omp.sections {
-  !$omp parallel sections
-    !OMPDialect: omp.section {
-    !$omp section
-      !FIRDialect: fir.load
-      !FIRDialect: arith.addi
-      !FIRDialect: fir.store
-      x = x + 12
-      !OMPDialect: omp.terminator
-    !OMPDialect: omp.section {
-    !$omp section
-      !FIRDialect: fir.load
-      !FIRDialect: arith.subi
-      !FIRDialect: fir.store
-      y = y - 5
-      !OMPDialect: omp.terminator
-  !OMPDialect: omp.terminator
-  !OMPDialect: omp.terminator
-  !$omp end parallel sections
-end subroutine omp_parallel_sections
-
-!===============================================================================
-! Parallel sections construct with allocate clause
-!===============================================================================
-
-!FIRDialect: func @_QPomp_parallel_sections
-subroutine omp_parallel_sections_allocate(x, y)
-  use omp_lib
-  integer, intent(inout) :: x, y
-  !FIRDialect: %[[allocator_1:.*]] = arith.constant 4 : i64
-  !FIRDialect: %[[allocator_2:.*]] = arith.constant 4 : i64
-  !LLVMDialect: %[[allocator_1:.*]] = llvm.mlir.constant(4 : i64) : i64
-  !LLVMDialect: %[[allocator_2:.*]] = llvm.mlir.constant(4 : i64) : i64
-  !OMPDialect: omp.parallel allocate(
-  !FIRDialect: %[[allocator_2]] : i64 -> %{{.*}} : !fir.ref<i32>) {
-  !LLVMDialect: %[[allocator_2]] : i64 -> %{{.*}} : !llvm.ptr) {
-  !OMPDialect: omp.sections allocate(
-  !FIRDialect: %[[allocator_1]] : i64 -> %{{.*}} : !fir.ref<i32>) {
-  !LLVMDialect: %[[allocator_1]] : i64 -> %{{.*}} : !llvm.ptr) {
-  !$omp parallel sections allocate(omp_high_bw_mem_alloc: x)
-    !OMPDialect: omp.section {
-    !$omp section
-      x = x + 12
-      !OMPDialect: omp.terminator
-    !OMPDialect: omp.section {
-    !$omp section
-      y = y + 5
-      !OMPDialect: omp.terminator
-  !OMPDialect: omp.terminator
-  !OMPDialect: omp.terminator
-  !$omp end parallel sections
-end subroutine omp_parallel_sections_allocate
diff --git a/flang/test/Lower/OpenMP/FIR/parallel-wsloop-firstpriv.f90 b/flang/test/Lower/OpenMP/FIR/parallel-wsloop-firstpriv.f90
deleted file mode 100644
index 6eb39a2f6372..000000000000
--- a/flang/test/Lower/OpenMP/FIR/parallel-wsloop-firstpriv.f90
+++ /dev/null
@@ -1,61 +0,0 @@
-! This test checks lowering of OpenMP parallel DO, with the loop bound being
-! a firstprivate variable
-
-! RUN: bbc -fopenmp -emit-fir -hlfir=false %s -o - | FileCheck %s
-
-! CHECK: func @_QPomp_do_firstprivate(%[[ARG0:.*]]: !fir.ref<i32> {fir.bindc_name = "a"}) 
-subroutine omp_do_firstprivate(a)
-  integer::a
-  integer::n
-  n = a+1
-  !$omp parallel do firstprivate(a)
-  ! CHECK:  omp.parallel {
-  ! CHECK-NEXT: %[[REF:.*]] = fir.alloca i32 {adapt.valuebyref, pinned}
-  ! CHECK-NEXT: %[[CLONE:.*]] = fir.alloca i32 {bindc_name = "a", pinned
-  ! CHECK-NEXT: %[[LD:.*]] = fir.load %[[ARG0]] : !fir.ref<i32>
-  ! CHECK-NEXT: fir.store %[[LD]] to %[[CLONE]] : !fir.ref<i32>
-  ! CHECK: %[[LB:.*]] = arith.constant 1 : i32
-  ! CHECK-NEXT: %[[UB:.*]] = fir.load %[[CLONE]] : !fir.ref<i32>
-  ! CHECK-NEXT: %[[STEP:.*]] = arith.constant 1 : i32
-  ! CHECK-NEXT: omp.wsloop   for  (%[[ARG1:.*]]) : i32 = (%[[LB]]) to (%[[UB]]) inclusive step (%[[STEP]])
-  ! CHECK-NEXT: fir.store %[[ARG1]] to %[[REF]] : !fir.ref<i32>
-  ! CHECK-NEXT: fir.call @_QPfoo(%[[REF]], %[[CLONE]]) {{.*}}: (!fir.ref<i32>, !fir.ref<i32>) -> ()
-  ! CHECK-NEXT: omp.yield
-    do i=1, a
-      call foo(i, a)
-    end do
-  !$omp end parallel do
-  !CHECK: fir.call @_QPbar(%[[ARG0]]) {{.*}}: (!fir.ref<i32>) -> ()
-  call bar(a)
-end subroutine omp_do_firstprivate
-
-! CHECK: func @_QPomp_do_firstprivate2(%[[ARG0:.*]]: !fir.ref<i32> {fir.bindc_name = "a"}, %[[ARG1:.*]]: !fir.ref<i32> {fir.bindc_name = "n"}) 
-subroutine omp_do_firstprivate2(a, n)
-  integer::a
-  integer::n
-  n = a+1
-  !$omp parallel do firstprivate(a, n)
-  ! CHECK:  omp.parallel {
-  ! CHECK-NEXT: %[[REF:.*]] = fir.alloca i32 {adapt.valuebyref, pinned}
-  ! CHECK-NEXT: %[[CLONE:.*]] = fir.alloca i32 {bindc_name = "a", pinned
-  ! CHECK-NEXT: %[[LD:.*]] = fir.load %[[ARG0]] : !fir.ref<i32>
-  ! CHECK-NEXT: fir.store %[[LD]] to %[[CLONE]] : !fir.ref<i32>
-  ! CHECK-NEXT: %[[CLONE1:.*]] = fir.alloca i32 {bindc_name = "n", pinned
-  ! CHECK-NEXT: %[[LD1:.*]] = fir.load %[[ARG1]] : !fir.ref<i32>
-  ! CHECK-NEXT: fir.store %[[LD1]] to %[[CLONE1]] : !fir.ref<i32>
-
-
-  ! CHECK: %[[LB:.*]] = fir.load %[[CLONE]] : !fir.ref<i32>
-  ! CHECK-NEXT: %[[UB:.*]] = fir.load %[[CLONE1]] : !fir.ref<i32>
-  ! CHECK-NEXT: %[[STEP:.*]] = arith.constant 1 : i32
-  ! CHECK-NEXT: omp.wsloop   for  (%[[ARG2:.*]]) : i32 = (%[[LB]]) to (%[[UB]]) inclusive step (%[[STEP]])
-  ! CHECK-NEXT: fir.store %[[ARG2]] to %[[REF]] : !fir.ref<i32>
-  ! CHECK-NEXT: fir.call @_QPfoo(%[[REF]], %[[CLONE]]) {{.*}}: (!fir.ref<i32>, !fir.ref<i32>) -> ()
-  ! CHECK-NEXT: omp.yield
-    do i= a, n
-      call foo(i, a)
-    end do
-  !$omp end parallel do
-  !CHECK: fir.call @_QPbar(%[[ARG1]]) {{.*}}: (!fir.ref<i32>) -> ()
-  call bar(n)
-end subroutine omp_do_firstprivate2
diff --git a/flang/test/Lower/OpenMP/FIR/parallel-wsloop.f90 b/flang/test/Lower/OpenMP/FIR/parallel-wsloop.f90
deleted file mode 100644
index 8649cf284ffd..000000000000
--- a/flang/test/Lower/OpenMP/FIR/parallel-wsloop.f90
+++ /dev/null
@@ -1,277 +0,0 @@
-! This test checks lowering of OpenMP DO Directive (Worksharing).
-
-! RUN: bbc -fopenmp -emit-fir -hlfir=false %s -o - | FileCheck %s
-
-! CHECK-LABEL: func @_QPsimple_parallel_do()
-subroutine simple_parallel_do
-  integer :: i
-  ! CHECK:  omp.parallel
-  ! CHECK:     %[[WS_LB:.*]] = arith.constant 1 : i32
-  ! CHECK:     %[[WS_UB:.*]] = arith.constant 9 : i32
-  ! CHECK:     %[[WS_STEP:.*]] = arith.constant 1 : i32
-  ! CHECK:     omp.wsloop for (%[[I:.*]]) : i32 = (%[[WS_LB]]) to (%[[WS_UB]]) inclusive step (%[[WS_STEP]])
-  !$OMP PARALLEL DO
-  do i=1, 9
-  ! CHECK:    fir.store %[[I]] to %[[IV_ADDR:.*]] : !fir.ref<i32>
-  ! CHECK:    %[[LOAD_IV:.*]] = fir.load %[[IV_ADDR]] : !fir.ref<i32>
-  ! CHECK:    fir.call @_FortranAioOutputInteger32({{.*}}, %[[LOAD_IV]]) {{.*}}: (!fir.ref<i8>, i32) -> i1
-    print*, i
-  end do
-  ! CHECK:       omp.yield
-  ! CHECK:       omp.terminator
-  !$OMP END PARALLEL DO
-end subroutine
-
-! CHECK-LABEL: func @_QPparallel_do_with_parallel_clauses
-! CHECK-SAME: %[[COND_REF:.*]]: !fir.ref<!fir.logical<4>> {fir.bindc_name = "cond"}, %[[NT_REF:.*]]: !fir.ref<i32> {fir.bindc_name = "nt"}
-subroutine parallel_do_with_parallel_clauses(cond, nt)
-  logical :: cond
-  integer :: nt
-  integer :: i
-  ! CHECK:  %[[COND:.*]] = fir.load %[[COND_REF]] : !fir.ref<!fir.logical<4>>
-  ! CHECK:  %[[COND_CVT:.*]] = fir.convert %[[COND]] : (!fir.logical<4>) -> i1
-  ! CHECK:  %[[NT:.*]] = fir.load %[[NT_REF]] : !fir.ref<i32>
-  ! CHECK:  omp.parallel if(%[[COND_CVT]] : i1) num_threads(%[[NT]] : i32) proc_bind(close)
-  ! CHECK:     %[[WS_LB:.*]] = arith.constant 1 : i32
-  ! CHECK:     %[[WS_UB:.*]] = arith.constant 9 : i32
-  ! CHECK:     %[[WS_STEP:.*]] = arith.constant 1 : i32
-  ! CHECK:     omp.wsloop for (%[[I:.*]]) : i32 = (%[[WS_LB]]) to (%[[WS_UB]]) inclusive step (%[[WS_STEP]])
-  !$OMP PARALLEL DO IF(cond) NUM_THREADS(nt) PROC_BIND(close)
-  do i=1, 9
-  ! CHECK:    fir.store %[[I]] to %[[IV_ADDR:.*]] : !fir.ref<i32>
-  ! CHECK:    %[[LOAD_IV:.*]] = fir.load %[[IV_ADDR]] : !fir.ref<i32>
-  ! CHECK:    fir.call @_FortranAioOutputInteger32({{.*}}, %[[LOAD_IV]]) {{.*}}: (!fir.ref<i8>, i32) -> i1
-    print*, i
-  end do
-  ! CHECK:       omp.yield
-  ! CHECK:       omp.terminator
-  !$OMP END PARALLEL DO
-end subroutine
-
-! CHECK-LABEL: func @_QPparallel_do_with_clauses
-! CHECK-SAME: %[[NT_REF:.*]]: !fir.ref<i32> {fir.bindc_name = "nt"}
-subroutine parallel_do_with_clauses(nt)
-  integer :: nt
-  integer :: i
-  ! CHECK:  %[[NT:.*]] = fir.load %[[NT_REF]] : !fir.ref<i32>
-  ! CHECK:  omp.parallel num_threads(%[[NT]] : i32)
-  ! CHECK:     %[[WS_LB:.*]] = arith.constant 1 : i32
-  ! CHECK:     %[[WS_UB:.*]] = arith.constant 9 : i32
-  ! CHECK:     %[[WS_STEP:.*]] = arith.constant 1 : i32
-  ! CHECK:     omp.wsloop schedule(dynamic) for (%[[I:.*]]) : i32 = (%[[WS_LB]]) to (%[[WS_UB]]) inclusive step (%[[WS_STEP]])
-  !$OMP PARALLEL DO NUM_THREADS(nt) SCHEDULE(dynamic)
-  do i=1, 9
-  ! CHECK:    fir.store %[[I]] to %[[IV_ADDR:.*]] : !fir.ref<i32>
-  ! CHECK:    %[[LOAD_IV:.*]] = fir.load %[[IV_ADDR]] : !fir.ref<i32>
-  ! CHECK:    fir.call @_FortranAioOutputInteger32({{.*}}, %[[LOAD_IV]]) {{.*}}: (!fir.ref<i8>, i32) -> i1
-    print*, i
-  end do
-  ! CHECK:       omp.yield
-  ! CHECK:       omp.terminator
-  !$OMP END PARALLEL DO
-end subroutine
-
-!===============================================================================
-! Checking for the following construct:
-!   !$omp parallel do private(...) firstprivate(...)
-!===============================================================================
-
-! CHECK-LABEL: func @_QPparallel_do_with_privatisation_clauses
-! CHECK-SAME: %[[COND_REF:.*]]: !fir.ref<!fir.logical<4>> {fir.bindc_name = "cond"}, %[[NT_REF:.*]]: !fir.ref<i32> {fir.bindc_name = "nt"}
-subroutine parallel_do_with_privatisation_clauses(cond,nt)
-  logical :: cond
-  integer :: nt
-  integer :: i
-  ! CHECK:  omp.parallel
-  ! CHECK:    %[[PRIVATE_COND_REF:.*]] = fir.alloca !fir.logical<4> {bindc_name = "cond", pinned, uniq_name = "_QFparallel_do_with_privatisation_clausesEcond"}
-  ! CHECK:    %[[PRIVATE_NT_REF:.*]] = fir.alloca i32 {bindc_name = "nt", pinned, uniq_name = "_QFparallel_do_with_privatisation_clausesEnt"}
-  ! CHECK:    %[[NT_VAL:.*]] = fir.load %[[NT_REF]] : !fir.ref<i32>
-  ! CHECK:    fir.store %[[NT_VAL]] to %[[PRIVATE_NT_REF]] : !fir.ref<i32>
-  ! CHECK:    %[[WS_LB:.*]] = arith.constant 1 : i32
-  ! CHECK:    %[[WS_UB:.*]] = arith.constant 9 : i32
-  ! CHECK:    %[[WS_STEP:.*]] = arith.constant 1 : i32
-  ! CHECK:    omp.wsloop for (%[[I:.*]]) : i32 = (%[[WS_LB]]) to (%[[WS_UB]]) inclusive step (%[[WS_STEP]])
-  !$OMP PARALLEL DO PRIVATE(cond) FIRSTPRIVATE(nt)
-  do i=1, 9
-  ! CHECK:    fir.store %[[I]] to %[[IV_ADDR:.*]] : !fir.ref<i32>
-  ! CHECK:    %[[LOAD_IV:.*]] = fir.load %[[IV_ADDR]] : !fir.ref<i32>
-  ! CHECK:      fir.call @_FortranAioOutputInteger32({{.*}}, %[[LOAD_IV]]) {{.*}}: (!fir.ref<i8>, i32) -> i1
-  ! CHECK:      %[[PRIVATE_COND_VAL:.*]] = fir.load %[[PRIVATE_COND_REF]] : !fir.ref<!fir.logical<4>>
-  ! CHECK:      %[[PRIVATE_COND_VAL_CVT:.*]] = fir.convert %[[PRIVATE_COND_VAL]] : (!fir.logical<4>) -> i1
-  ! CHECK:      fir.call @_FortranAioOutputLogical({{.*}}, %[[PRIVATE_COND_VAL_CVT]]) {{.*}}: (!fir.ref<i8>, i1) -> i1
-  ! CHECK:      %[[PRIVATE_NT_VAL:.*]] = fir.load %[[PRIVATE_NT_REF]] : !fir.ref<i32>
-  ! CHECK:      fir.call @_FortranAioOutputInteger32({{.*}}, %[[PRIVATE_NT_VAL]]) {{.*}}: (!fir.ref<i8>, i32) -> i1
-    print*, i, cond, nt
-  end do
-  ! CHECK:      omp.yield
-  ! CHECK:    omp.terminator
-  !$OMP END PARALLEL DO
-end subroutine
-
-!===============================================================================
-! Checking for the following construct
-!   !$omp parallel private(...) firstprivate(...)
-!   !$omp do
-!===============================================================================
-
-subroutine parallel_private_do(cond,nt)
-logical :: cond
-  integer :: nt
-  integer :: i
-  !$OMP PARALLEL PRIVATE(cond) FIRSTPRIVATE(nt)
-  !$OMP DO
-  do i=1, 9
-    call foo(i, cond, nt)
-  end do
-  !$OMP END DO
-  !$OMP END PARALLEL
-end subroutine parallel_private_do
-
-! CHECK-LABEL:   func.func @_QPparallel_private_do(
-! CHECK-SAME:                                      %[[VAL_0:.*]]: !fir.ref<!fir.logical<4>> {fir.bindc_name = "cond"},
-! CHECK-SAME:                                      %[[VAL_1:.*]]: !fir.ref<i32> {fir.bindc_name = "nt"}) {
-! CHECK:           %[[I:.*]] = fir.alloca i32 {bindc_name = "i", uniq_name = "_QFparallel_private_doEi"}
-! CHECK:           omp.parallel   {
-! CHECK:             %[[I_PRIV:.*]] = fir.alloca i32 {adapt.valuebyref, pinned}
-! CHECK:             %[[COND_ADDR:.*]] = fir.alloca !fir.logical<4> {bindc_name = "cond", pinned, uniq_name = "_QFparallel_private_doEcond"}
-! CHECK:             %[[NT_ADDR:.*]] = fir.alloca i32 {bindc_name = "nt", pinned, uniq_name = "_QFparallel_private_doEnt"}
-! CHECK:             %[[NT:.*]] = fir.load %[[VAL_1]] : !fir.ref<i32>
-! CHECK:             fir.store %[[NT]] to %[[NT_ADDR]] : !fir.ref<i32>
-! CHECK:             %[[VAL_7:.*]] = arith.constant 1 : i32
-! CHECK:             %[[VAL_8:.*]] = arith.constant 9 : i32
-! CHECK:             %[[VAL_9:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop   for  (%[[I:.*]]) : i32 = (%[[VAL_7]]) to (%[[VAL_8]]) inclusive step (%[[VAL_9]]) {
-! CHECK:               fir.store %[[I]] to %[[I_PRIV]] : !fir.ref<i32>
-! CHECK:               fir.call @_QPfoo(%[[I_PRIV]], %[[COND_ADDR]], %[[NT_ADDR]]) {{.*}}: (!fir.ref<i32>, !fir.ref<!fir.logical<4>>, !fir.ref<i32>) -> ()
-! CHECK:               omp.yield
-! CHECK:             }
-! CHECK:             omp.terminator
-! CHECK:           }
-! CHECK:           return
-! CHECK:         }
-
-!===============================================================================
-! Checking for the following construct
-!   !$omp parallel
-!   !$omp do firstprivate(...) firstprivate(...)
-!===============================================================================
-
-subroutine omp_parallel_multiple_firstprivate_do(a, b)
-  integer::a, b
-  !$OMP PARALLEL FIRSTPRIVATE(a) FIRSTPRIVATE(b)
-  !$OMP DO
-  do i=1, 10
-    call bar(i, a)
-  end do
-  !$OMP END DO
-  !$OMP END PARALLEL
-end subroutine omp_parallel_multiple_firstprivate_do
-
-! CHECK-LABEL:   func.func @_QPomp_parallel_multiple_firstprivate_do(
-! CHECK-SAME:                                                        %[[A_ADDR:.*]]: !fir.ref<i32> {fir.bindc_name = "a"},
-! CHECK-SAME:                                                        %[[B_ADDR:.*]]: !fir.ref<i32> {fir.bindc_name = "b"}) {
-! CHECK:           %[[I_ADDR:.*]] = fir.alloca i32 {bindc_name = "i", uniq_name = "_QFomp_parallel_multiple_firstprivate_doEi"}
-! CHECK:           omp.parallel   {
-! CHECK:             %[[I_PRIV_ADDR:.*]] = fir.alloca i32 {adapt.valuebyref, pinned}
-! CHECK:             %[[A_PRIV_ADDR:.*]] = fir.alloca i32 {bindc_name = "a", pinned, uniq_name = "_QFomp_parallel_multiple_firstprivate_doEa"}
-! CHECK:             %[[A:.*]] = fir.load %[[A_ADDR]] : !fir.ref<i32>
-! CHECK:             fir.store %[[A]] to %[[A_PRIV_ADDR]] : !fir.ref<i32>
-! CHECK:             %[[B_PRIV_ADDR:.*]] = fir.alloca i32 {bindc_name = "b", pinned, uniq_name = "_QFomp_parallel_multiple_firstprivate_doEb"}
-! CHECK:             %[[B:.*]] = fir.load %[[B_ADDR]] : !fir.ref<i32>
-! CHECK:             fir.store %[[B]] to %[[B_PRIV_ADDR]] : !fir.ref<i32>
-! CHECK:             %[[VAL_8:.*]] = arith.constant 1 : i32
-! CHECK:             %[[VAL_9:.*]] = arith.constant 10 : i32
-! CHECK:             %[[VAL_10:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop   for  (%[[I:.*]]) : i32 = (%[[VAL_8]]) to (%[[VAL_9]]) inclusive step (%[[VAL_10]]) {
-! CHECK:               fir.store %[[I]] to %[[I_PRIV_ADDR]] : !fir.ref<i32>
-! CHECK:               fir.call @_QPbar(%[[I_PRIV_ADDR]], %[[A_PRIV_ADDR]]) {{.*}}: (!fir.ref<i32>, !fir.ref<i32>) -> ()
-! CHECK:               omp.yield
-! CHECK:             }
-! CHECK:             omp.terminator
-! CHECK:           }
-! CHECK:           return
-! CHECK:         }
-
-!===============================================================================
-! Checking for the following construct
-!   !$omp parallel
-!   !$omp do private(...) firstprivate(...)
-!===============================================================================
-
-subroutine parallel_do_private(cond,nt)
-logical :: cond
-  integer :: nt
-  integer :: i
-  !$OMP PARALLEL
-  !$OMP DO PRIVATE(cond) FIRSTPRIVATE(nt)
-  do i=1, 9
-    call foo(i, cond, nt)
-  end do
-  !$OMP END DO
-  !$OMP END PARALLEL
-end subroutine parallel_do_private
-
-! CHECK-LABEL:   func.func @_QPparallel_do_private(
-! CHECK-SAME:                                      %[[VAL_0:.*]]: !fir.ref<!fir.logical<4>> {fir.bindc_name = "cond"},
-! CHECK-SAME:                                      %[[VAL_1:.*]]: !fir.ref<i32> {fir.bindc_name = "nt"}) {
-! CHECK:           %[[I_ADDR:.*]] = fir.alloca i32 {bindc_name = "i", uniq_name = "_QFparallel_do_privateEi"}
-! CHECK:           omp.parallel   {
-! CHECK:             %[[I_PRIV_ADDR:.*]] = fir.alloca i32 {adapt.valuebyref, pinned}
-! CHECK:             %[[COND_ADDR:.*]] = fir.alloca !fir.logical<4> {bindc_name = "cond", pinned, uniq_name = "_QFparallel_do_privateEcond"}
-! CHECK:             %[[NT_ADDR:.*]] = fir.alloca i32 {bindc_name = "nt", pinned, uniq_name = "_QFparallel_do_privateEnt"}
-! CHECK:             %[[NT:.*]] = fir.load %[[VAL_1]] : !fir.ref<i32>
-! CHECK:             fir.store %[[NT]] to %[[NT_ADDR]] : !fir.ref<i32>
-! CHECK:             %[[VAL_7:.*]] = arith.constant 1 : i32
-! CHECK:             %[[VAL_8:.*]] = arith.constant 9 : i32
-! CHECK:             %[[VAL_9:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop   for  (%[[I:.*]]) : i32 = (%[[VAL_7]]) to (%[[VAL_8]]) inclusive step (%[[VAL_9]]) {
-! CHECK:               fir.store %[[I]] to %[[I_PRIV_ADDR]] : !fir.ref<i32>
-! CHECK:               fir.call @_QPfoo(%[[I_PRIV_ADDR]], %[[COND_ADDR]], %[[NT_ADDR]]) {{.*}}: (!fir.ref<i32>, !fir.ref<!fir.logical<4>>, !fir.ref<i32>) -> ()
-! CHECK:               omp.yield
-! CHECK:             }
-! CHECK:             omp.terminator
-! CHECK:           }
-! CHECK:           return
-! CHECK:         }
-
-!===============================================================================
-! Checking for the following construct
-!   !$omp parallel
-!   !$omp do firstprivate(...) firstprivate(...)
-!===============================================================================
-
-subroutine omp_parallel_do_multiple_firstprivate(a, b)
-  integer::a, b
-  !$OMP PARALLEL
-  !$OMP DO FIRSTPRIVATE(a) FIRSTPRIVATE(b)
-  do i=1, 10
-    call bar(i, a)
-  end do
-  !$OMP END DO
-  !$OMP END PARALLEL
-end subroutine omp_parallel_do_multiple_firstprivate
-
-! CHECK-LABEL:   func.func @_QPomp_parallel_do_multiple_firstprivate(
-! CHECK-SAME:                                                        %[[A_ADDR:.*]]: !fir.ref<i32> {fir.bindc_name = "a"},
-! CHECK-SAME:                                                        %[[B_ADDR:.*]]: !fir.ref<i32> {fir.bindc_name = "b"}) {
-! CHECK:           %[[I_ADDR:.*]] = fir.alloca i32 {bindc_name = "i", uniq_name = "_QFomp_parallel_do_multiple_firstprivateEi"}
-! CHECK:           omp.parallel   {
-! CHECK:             %[[I_PRIV_ADDR:.*]] = fir.alloca i32 {adapt.valuebyref, pinned}
-! CHECK:             %[[A_PRIV_ADDR:.*]] = fir.alloca i32 {bindc_name = "a", pinned, uniq_name = "_QFomp_parallel_do_multiple_firstprivateEa"}
-! CHECK:             %[[A:.*]] = fir.load %[[A_ADDR]] : !fir.ref<i32>
-! CHECK:             fir.store %[[A]] to %[[A_PRIV_ADDR]] : !fir.ref<i32>
-! CHECK:             %[[B_PRIV_ADDR:.*]] = fir.alloca i32 {bindc_name = "b", pinned, uniq_name = "_QFomp_parallel_do_multiple_firstprivateEb"}
-! CHECK:             %[[B:.*]] = fir.load %[[B_ADDR]] : !fir.ref<i32>
-! CHECK:             fir.store %[[B]] to %[[B_PRIV_ADDR]] : !fir.ref<i32>
-! CHECK:             %[[VAL_8:.*]] = arith.constant 1 : i32
-! CHECK:             %[[VAL_9:.*]] = arith.constant 10 : i32
-! CHECK:             %[[VAL_10:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop   for  (%[[I:.*]]) : i32 = (%[[VAL_8]]) to (%[[VAL_9]]) inclusive step (%[[VAL_10]]) {
-! CHECK:               fir.store %[[I]] to %[[I_PRIV_ADDR]] : !fir.ref<i32>
-! CHECK:               fir.call @_QPbar(%[[I_PRIV_ADDR]], %[[A_PRIV_ADDR]]) {{.*}}: (!fir.ref<i32>, !fir.ref<i32>) -> ()
-! CHECK:               omp.yield
-! CHECK:             }
-! CHECK:             omp.terminator
-! CHECK:           }
-! CHECK:           return
-! CHECK:         }
diff --git a/flang/test/Lower/OpenMP/FIR/parallel.f90 b/flang/test/Lower/OpenMP/FIR/parallel.f90
deleted file mode 100644
index a2ceb2d939f2..000000000000
--- a/flang/test/Lower/OpenMP/FIR/parallel.f90
+++ /dev/null
@@ -1,211 +0,0 @@
-! REQUIRES: openmp_runtime
-
-!RUN: %flang_fc1 -emit-fir -flang-deprecated-no-hlfir -fopenmp %s -o - | FileCheck %s --check-prefixes="FIRDialect,OMPDialect"
-!RUN: %flang_fc1 -emit-fir -flang-deprecated-no-hlfir -fopenmp %s -o - | fir-opt --fir-to-llvm-ir | FileCheck %s --check-prefixes="LLVMDialect,OMPDialect"
-
-!FIRDialect-LABEL: func @_QPparallel_simple
-subroutine parallel_simple()
-   !OMPDialect: omp.parallel
-!$omp parallel
-   !FIRDialect: fir.call
-   call f1()
-!$omp end parallel
-end subroutine parallel_simple
-
-!===============================================================================
-! `if` clause
-!===============================================================================
-
-!FIRDialect-LABEL: func @_QPparallel_if
-subroutine parallel_if(alpha, beta, gamma)
-   integer, intent(in) :: alpha
-   logical, intent(in) :: beta
-   logical(1) :: logical1
-   logical(2) :: logical2
-   logical(4) :: logical4
-   logical(8) :: logical8
-
-   !OMPDialect: omp.parallel if(%{{.*}} : i1) {
-   !$omp parallel if(alpha .le. 0)
-   !FIRDialect: fir.call
-   call f1()
-   !OMPDialect: omp.terminator
-   !$omp end parallel
-
-   !OMPDialect: omp.parallel if(%{{.*}} : i1) {
-   !$omp parallel if(.false.)
-   !FIRDialect: fir.call
-   call f2()
-   !OMPDialect: omp.terminator
-   !$omp end parallel
-
-   !OMPDialect: omp.parallel if(%{{.*}} : i1) {
-   !$omp parallel if(alpha .ge. 0)
-   !FIRDialect: fir.call
-   call f3()
-   !OMPDialect: omp.terminator
-   !$omp end parallel
-
-   !OMPDialect: omp.parallel if(%{{.*}} : i1) {
-   !$omp parallel if(.true.)
-   !FIRDialect: fir.call
-   call f4()
-   !OMPDialect: omp.terminator
-   !$omp end parallel
-
-   !OMPDialect: omp.parallel if(%{{.*}} : i1) {
-   !$omp parallel if(beta)
-   !FIRDialect: fir.call
-   call f1()
-   !OMPDialect: omp.terminator
-   !$omp end parallel
-
-   !OMPDialect: omp.parallel if(%{{.*}} : i1) {
-   !$omp parallel if(logical1)
-   !FIRDialect: fir.call
-   call f1()
-   !OMPDialect: omp.terminator
-   !$omp end parallel
-
-   !OMPDialect: omp.parallel if(%{{.*}} : i1) {
-   !$omp parallel if(logical2)
-   !FIRDialect: fir.call
-   call f1()
-   !OMPDialect: omp.terminator
-   !$omp end parallel
-
-   !OMPDialect: omp.parallel if(%{{.*}} : i1) {
-   !$omp parallel if(logical4)
-   !FIRDialect: fir.call
-   call f1()
-   !OMPDialect: omp.terminator
-   !$omp end parallel
-
-   !OMPDialect: omp.parallel if(%{{.*}} : i1) {
-   !$omp parallel if(logical8)
-   !FIRDialect: fir.call
-   call f1()
-   !OMPDialect: omp.terminator
-   !$omp end parallel
-
-end subroutine parallel_if
-
-!===============================================================================
-! `num_threads` clause
-!===============================================================================
-
-!FIRDialect-LABEL: func @_QPparallel_numthreads
-subroutine parallel_numthreads(num_threads)
-   integer, intent(inout) :: num_threads
-
-   !OMPDialect: omp.parallel num_threads(%{{.*}}: i32) {
-   !$omp parallel num_threads(16)
-   !FIRDialect: fir.call
-   call f1()
-   !OMPDialect: omp.terminator
-   !$omp end parallel
-
-   num_threads = 4
-
-   !OMPDialect: omp.parallel num_threads(%{{.*}} : i32) {
-   !$omp parallel num_threads(num_threads)
-   !FIRDialect: fir.call
-   call f2()
-   !OMPDialect: omp.terminator
-   !$omp end parallel
-
-end subroutine parallel_numthreads
-
-!===============================================================================
-! `proc_bind` clause
-!===============================================================================
-
-!FIRDialect-LABEL: func @_QPparallel_proc_bind
-subroutine parallel_proc_bind()
-
-   !OMPDialect: omp.parallel proc_bind(master) {
-   !$omp parallel proc_bind(master)
-   !FIRDialect: fir.call
-   call f1()
-   !OMPDialect: omp.terminator
-   !$omp end parallel
-
-   !OMPDialect: omp.parallel proc_bind(close) {
-   !$omp parallel proc_bind(close)
-   !FIRDialect: fir.call
-   call f2()
-   !OMPDialect: omp.terminator
-   !$omp end parallel
-
-   !OMPDialect: omp.parallel proc_bind(spread) {
-   !$omp parallel proc_bind(spread)
-   !FIRDialect: fir.call
-   call f3()
-   !OMPDialect: omp.terminator
-   !$omp end parallel
-
-end subroutine parallel_proc_bind
-
-!===============================================================================
-! `allocate` clause
-!===============================================================================
-
-!FIRDialect-LABEL: func @_QPparallel_allocate
-subroutine parallel_allocate()
-   use omp_lib
-   integer :: x
-   !OMPDialect: omp.parallel allocate(
-   !FIRDialect: %{{.+}} : i64 -> %{{.+}} : !fir.ref<i32>
-   !LLVMDialect: %{{.+}} : i64 -> %{{.+}} : !llvm.ptr
-   !OMPDialect: ) {
-   !$omp parallel allocate(omp_high_bw_mem_alloc: x) private(x)
-   !FIRDialect: arith.addi
-   x = x + 12
-   !OMPDialect: omp.terminator
-   !$omp end parallel
-end subroutine parallel_allocate
-
-!===============================================================================
-! multiple clauses
-!===============================================================================
-
-!FIRDialect-LABEL: func @_QPparallel_multiple_clauses
-subroutine parallel_multiple_clauses(alpha, num_threads)
-   use omp_lib
-   integer, intent(inout) :: alpha
-   integer, intent(in) :: num_threads
-
-   !OMPDialect: omp.parallel if({{.*}} : i1) proc_bind(master) {
-   !$omp parallel if(alpha .le. 0) proc_bind(master)
-   !FIRDialect: fir.call
-   call f1()
-   !OMPDialect: omp.terminator
-   !$omp end parallel
-
-   !OMPDialect: omp.parallel num_threads({{.*}} : i32) proc_bind(close) {
-   !$omp parallel proc_bind(close) num_threads(num_threads)
-   !FIRDialect: fir.call
-   call f2()
-   !OMPDialect: omp.terminator
-   !$omp end parallel
-
-   !OMPDialect: omp.parallel if({{.*}} : i1) num_threads({{.*}} : i32) {
-   !$omp parallel num_threads(num_threads) if(alpha .le. 0)
-   !FIRDialect: fir.call
-   call f3()
-   !OMPDialect: omp.terminator
-   !$omp end parallel
-
-   !OMPDialect: omp.parallel if({{.*}} : i1) num_threads({{.*}} : i32) allocate(
-   !FIRDialect: %{{.+}} : i64 -> %{{.+}} : !fir.ref<i32>
-   !LLVMDialect: %{{.+}} : i64 -> %{{.+}} : !llvm.ptr
-   !OMPDialect: ) {
-   !$omp parallel num_threads(num_threads) if(alpha .le. 0) allocate(omp_high_bw_mem_alloc: alpha) private(alpha)
-   !FIRDialect: fir.call
-   call f3()
-   !FIRDialect: arith.addi
-   alpha = alpha + 12
-   !OMPDialect: omp.terminator
-   !$omp end parallel
-
-end subroutine parallel_multiple_clauses
diff --git a/flang/test/Lower/OpenMP/FIR/pre-fir-tree-loop.f90 b/flang/test/Lower/OpenMP/FIR/pre-fir-tree-loop.f90
deleted file mode 100644
index eca8fb304986..000000000000
--- a/flang/test/Lower/OpenMP/FIR/pre-fir-tree-loop.f90
+++ /dev/null
@@ -1,70 +0,0 @@
-! RUN: bbc -fopenmp -pft-test -o %t %s | FileCheck %s
-! RUN: %flang_fc1 -fopenmp -fdebug-dump-pft -o %t %s | FileCheck %s
-
-! Loop constructs always have an `end do` which can be the target of
-! a branch. So OpenMP loop constructs do not need an artificial
-! continue inserted for a target.
-
-!CHECK-LABEL: sb0
-!CHECK-NOT: continue
-subroutine sb0(cond)
-  implicit none
-  logical :: cond
-  integer :: i
-  !$omp parallel do
-  do i = 1, 20
-    if( cond) then
-      cycle
-    end if
-  end do
-  return
-end subroutine
-
-!CHECK-LABEL: sb1
-!CHECK-NOT: continue
-subroutine sb1(cond)
-  implicit none
-  logical :: cond
-  integer :: i
-  !$omp parallel do
-  do i = 1, 20
-    if( cond) then
-      cycle
-    end if
-  end do
-  !$omp end parallel do
-  return
-end subroutine
-
-!CHECK-LABEL: sb2
-!CHECK-NOT: continue
-subroutine sb2
-  integer :: i, n
-  integer :: tmp
-
-  !$omp parallel do
-  do ifld=1,n
-     do isum=1,n
-       if (tmp > n) then
-         exit
-       endif
-     enddo
-     tmp = n
-  enddo
-end subroutine
-
-!CHECK-LABEL: sb3
-!CHECK-NOT: continue
-subroutine sb3
-  integer :: i, n
-  integer :: tmp
-
-  !$omp parallel do
-  do ifld=1,n
-     do isum=1,n
-       if (tmp > n) then
-         exit
-       endif
-     enddo
-  enddo
-end subroutine
diff --git a/flang/test/Lower/OpenMP/FIR/pre-fir-tree01.f90 b/flang/test/Lower/OpenMP/FIR/pre-fir-tree01.f90
deleted file mode 100644
index fc817942513e..000000000000
--- a/flang/test/Lower/OpenMP/FIR/pre-fir-tree01.f90
+++ /dev/null
@@ -1,19 +0,0 @@
-! RUN: bbc -fopenmp -pft-test -o %t %s | FileCheck %s
-! RUN: %flang_fc1 -fopenmp -fdebug-dump-pft -o %t %s | FileCheck %s
-
-! Test structure of the Pre-FIR tree with OpenMP
-
-subroutine sub1(a, b, n)
-  real :: a(:), b(:)
-  integer :: n, i
-  !$omp parallel do
-  do i = 1, n
-    b(i) = exp(a(i))
-  end do
-  !$omp end parallel do
-end subroutine
-
-! CHECK-LABEL: Subroutine sub1
-! CHECK:       <<OpenMPConstruct>>
-! CHECK:       <<DoConstruct>>
-! CHECK:       <<End OpenMPConstruct>>
diff --git a/flang/test/Lower/OpenMP/FIR/private-commonblock.f90 b/flang/test/Lower/OpenMP/FIR/private-commonblock.f90
deleted file mode 100644
index 90036e0c0c7e..000000000000
--- a/flang/test/Lower/OpenMP/FIR/private-commonblock.f90
+++ /dev/null
@@ -1,109 +0,0 @@
-! RUN: %flang_fc1 -emit-fir -flang-deprecated-no-hlfir -fopenmp -o - %s 2>&1 | FileCheck %s
-
-!CHECK: func.func @_QPprivate_common() {
-!CHECK: omp.parallel {
-!CHECK: %[[X:.*]] = fir.alloca f32 {bindc_name = "x", pinned, uniq_name = "_QFprivate_commonEx"}
-!CHECK: %[[Y:.*]] = fir.alloca f32 {bindc_name = "y", pinned, uniq_name = "_QFprivate_commonEy"}
-!CHECK: omp.terminator
-!CHECK: }
-!CHECK: return
-!CHECK: }
-subroutine private_common
-  common /c/ x, y
-  real x, y
-  !$omp parallel private(/c/)
-  !$omp end parallel
-end subroutine
-
-!CHECK: %[[val_0:.*]] = fir.address_of(@blk_) : !fir.ref<!fir.array<74xi8>>
-!CHECK: %[[val_1:.*]] = fir.convert %0 : (!fir.ref<!fir.array<74xi8>>) -> !fir.ref<!fir.array<?xi8>>
-!CHECK: %[[val_c0:.*]] = arith.constant 0 : index
-!CHECK: %[[val_2:.*]] = fir.coordinate_of %[[val_1]], %[[val_c0]] : (!fir.ref<!fir.array<?xi8>>, index) -> !fir.ref<i8>
-!CHECK: %[[val_3:.*]] = fir.convert %[[val_2]] : (!fir.ref<i8>) -> !fir.ref<i32>
-!CHECK: %[[val_4:.*]] = fir.convert %[[val_0]] : (!fir.ref<!fir.array<74xi8>>) -> !fir.ref<!fir.array<?xi8>>
-!CHECK: %[[val_c4:.*]] = arith.constant 4 : index
-!CHECK: %[[val_5:.*]] = fir.coordinate_of %[[val_4]], %[[val_c4]] : (!fir.ref<!fir.array<?xi8>>, index) -> !fir.ref<i8>
-!CHECK: %[[val_6:.*]] = fir.convert %[[val_5]] : (!fir.ref<i8>) -> !fir.ref<!fir.array<10xf32>>
-!CHECK: %[[val_7:.*]] = fir.convert %[[val_0]] : (!fir.ref<!fir.array<74xi8>>) -> !fir.ref<!fir.array<?xi8>>
-!CHECK: %[[val_c44:.*]] = arith.constant 44 : index
-!CHECK: %[[val_8:.*]] = fir.coordinate_of %[[val_7]], %[[val_c44]] : (!fir.ref<!fir.array<?xi8>>, index) -> !fir.ref<i8>
-!CHECK: %[[val_9:.*]] = fir.convert %[[val_8]] : (!fir.ref<i8>) -> !fir.ref<!fir.char<1,5>>
-!CHECK: %[[val_c5:.*]] = arith.constant 5 : index
-!CHECK: %[[val_10:.*]] = fir.convert %[[val_0]] : (!fir.ref<!fir.array<74xi8>>) -> !fir.ref<!fir.array<?xi8>>
-!CHECK: %[[val_c49:.*]] = arith.constant 49 : index
-!CHECK: %[[val_11:.*]] = fir.coordinate_of %[[val_10]], %[[val_c49]] : (!fir.ref<!fir.array<?xi8>>, index) -> !fir.ref<i8>
-!CHECK: %[[val_12:.*]] = fir.convert %[[val_11]] : (!fir.ref<i8>) -> !fir.ref<!fir.array<5x!fir.char<1,5>>>
-!CHECK: %[[val_c5_0:.*]] = arith.constant 5 : index
-!CHECK: %[[val_14:.*]] = fir.emboxchar %[[val_9]], %[[val_c5]] : (!fir.ref<!fir.char<1,5>>, index) -> !fir.boxchar<1>
-!CHECK: %[[val_15:.*]] = fir.convert %[[val_12]] : (!fir.ref<!fir.array<5x!fir.char<1,5>>>) -> !fir.ref<!fir.char<1,?>>
-!CHECK: %[[val_16:.*]] = fir.emboxchar %[[val_15]], %[[val_c5_0]] : (!fir.ref<!fir.char<1,?>>, index) -> !fir.boxchar<1>
-!CHECK: fir.call @_QPsub1(%[[val_3]], %[[val_6]], %[[val_14]], %[[val_16]]) fastmath<contract> : (!fir.ref<i32>, !fir.ref<!fir.array<10xf32>>, !fir.boxchar<1>, !fir.boxchar<1>) -> ()
-!CHECK: omp.parallel {
-!CHECK: %[[val_21:.*]] = fir.alloca i32 {bindc_name = "a", pinned, uniq_name = "_QFprivate_clause_commonblockEa"}
-!CHECK: %[[val_22:.*]] = fir.alloca !fir.array<10xf32> {bindc_name = "b", pinned, uniq_name = "_QFprivate_clause_commonblockEb"}
-!CHECK: %[[val_23:.*]] = fir.alloca !fir.char<1,5> {bindc_name = "c", pinned, uniq_name = "_QFprivate_clause_commonblockEc"}
-!CHECK: %[[val_24:.*]] = fir.alloca !fir.array<5x!fir.char<1,5>> {bindc_name = "d", pinned, uniq_name = "_QFprivate_clause_commonblockEd"}
-!CHECK: %[[val_26:.*]] = fir.emboxchar %[[val_23]], %[[val_c5]] : (!fir.ref<!fir.char<1,5>>, index) -> !fir.boxchar<1>
-!CHECK: %[[val_27:.*]] = fir.convert %[[val_24]] : (!fir.ref<!fir.array<5x!fir.char<1,5>>>) -> !fir.ref<!fir.char<1,?>>
-!CHECK: %[[val_28:.*]] = fir.emboxchar %[[val_27]], %[[val_c5_0]] : (!fir.ref<!fir.char<1,?>>, index) -> !fir.boxchar<1>
-!CHECK: fir.call @_QPsub2(%[[val_21]], %[[val_22]], %[[val_26]], %[[val_28]]) fastmath<contract> : (!fir.ref<i32>, !fir.ref<!fir.array<10xf32>>, !fir.boxchar<1>, !fir.boxchar<1>) -> ()
-!CHECK: omp.terminator
-!CHECK: }
-!CHECK: %[[val_18:.*]] = fir.emboxchar %[[val_9]], %[[val_c5]] : (!fir.ref<!fir.char<1,5>>, index) -> !fir.boxchar<1>
-!CHECK: %[[val_19:.*]] = fir.convert %[[val_12]] : (!fir.ref<!fir.array<5x!fir.char<1,5>>>) -> !fir.ref<!fir.char<1,?>>
-!CHECK: %[[val_20:.*]] = fir.emboxchar %[[val_19]], %[[val_c5_0]] : (!fir.ref<!fir.char<1,?>>, index) -> !fir.boxchar<1>
-!CHECK: fir.call @_QPsub3(%[[val_3]], %[[val_6]], %[[val_18]], %[[val_20]]) fastmath<contract> : {{.*}}
-!CHECK: return
-!CHECK: }
-subroutine private_clause_commonblock()
-  integer::a
-  real::b(10)
-  character(5):: c, d(5)
-  common /blk/ a, b, c, d
-  
-  call sub1(a, b, c, d)
-  !$omp parallel private(/blk/)
-        call sub2(a, b, c, d)
-  !$omp end parallel
-  call sub3(a, b, c, d)
-end subroutine
-
-!CHECK: func.func @_QPprivate_clause_commonblock_pointer() {
-!CHECK: %[[val_0:.*]] = fir.address_of(@blk_) : !fir.ref<!fir.array<74xi8>>
-!CHECK: %[[val_1:.*]] = fir.convert %[[val_0]] : (!fir.ref<!fir.array<74xi8>>) -> !fir.ref<!fir.array<?xi8>>
-!CHECK: %[[val_c24:.*]] = arith.constant 24 : index
-!CHECK: %[[val_2:.*]] = fir.coordinate_of %[[val_1]], %[[val_c24]] : (!fir.ref<!fir.array<?xi8>>, index) -> !fir.ref<i8>
-!CHECK: %[[val_3:.*]] = fir.convert %[[val_2]] : (!fir.ref<i8>) -> !fir.ref<i32>
-!CHECK: %[[val_4:.*]] = fir.convert %[[val_0]] : (!fir.ref<!fir.array<74xi8>>) -> !fir.ref<!fir.array<?xi8>>
-!CHECK: %[[val_c0:.*]] = arith.constant 0 : index
-!CHECK: %[[val_5:.*]] = fir.coordinate_of %[[val_4]], %[[val_c0]] : (!fir.ref<!fir.array<?xi8>>, index) -> !fir.ref<i8>
-!CHECK: %[[val_6:.*]] = fir.convert %[[val_5]] : (!fir.ref<i8>) -> !fir.ref<!fir.box<!fir.ptr<!fir.complex<4>>>>
-!CHECK: %[[val_7:.*]] = fir.load %[[val_6]] : !fir.ref<!fir.box<!fir.ptr<!fir.complex<4>>>>
-!CHECK: %[[val_8:.*]] = fir.box_addr %[[val_7]] : (!fir.box<!fir.ptr<!fir.complex<4>>>) -> !fir.ptr<!fir.complex<4>>
-!CHECK: %[[val_9:.*]] = fir.convert %[[val_8]] : (!fir.ptr<!fir.complex<4>>) -> !fir.ref<!fir.complex<4>>
-!CHECK: fir.call @_QPsub4(%[[val_9]], %[[val_3]]) fastmath<contract> : (!fir.ref<!fir.complex<4>>, !fir.ref<i32>) -> ()
-!CHECK: omp.parallel {
-!CHECK: %[[val_13:.*]] = fir.alloca !fir.box<!fir.ptr<!fir.complex<4>>> {bindc_name = "c", pinned, uniq_name = "_QFprivate_clause_commonblock_pointerEc"}
-!CHECK: %[[val_14:.*]] = fir.alloca i32 {bindc_name = "a", pinned, uniq_name = "_QFprivate_clause_commonblock_pointerEa"}
-!CHECK: %[[val_15:.*]] = fir.load %[[val_13]] : !fir.ref<!fir.box<!fir.ptr<!fir.complex<4>>>>
-!CHECK: %[[val_16:.*]] = fir.box_addr %[[val_15]] : (!fir.box<!fir.ptr<!fir.complex<4>>>) -> !fir.ptr<!fir.complex<4>>
-!CHECK: %[[val_17:.*]] = fir.convert %[[val_16]] : (!fir.ptr<!fir.complex<4>>) -> !fir.ref<!fir.complex<4>>
-!CHECK: fir.call @_QPsub5(%[[val_17]], %[[val_14]]) fastmath<contract> : (!fir.ref<!fir.complex<4>>, !fir.ref<i32>) -> ()
-!CHECK: omp.terminator
-!CHECK: }
-!CHECK: %[[val_10:.*]] = fir.load %[[val_6]] : !fir.ref<!fir.box<!fir.ptr<!fir.complex<4>>>>
-!CHECK: %[[val_11:.*]] = fir.box_addr %[[val_10]] : (!fir.box<!fir.ptr<!fir.complex<4>>>) -> !fir.ptr<!fir.complex<4>>
-!CHECK: %[[val_12:.*]] = fir.convert %[[val_11]] : (!fir.ptr<!fir.complex<4>>) -> !fir.ref<!fir.complex<4>>
-!CHECK: fir.call @_QPsub6(%[[val_12]], %[[val_3]]) fastmath<contract> : (!fir.ref<!fir.complex<4>>, !fir.ref<i32>) -> ()
-!CHECK: return
-!CHECK: }
-subroutine private_clause_commonblock_pointer()
-  complex, pointer :: c
-  integer:: a
-  common /blk/ c, a
-  call sub4(c, a)
-  !$omp parallel private(/blk/)
-        call sub5(c, a)
-  !$omp end parallel
-  call sub6(c, a)
-end subroutine
diff --git a/flang/test/Lower/OpenMP/FIR/requires-common.f90 b/flang/test/Lower/OpenMP/FIR/requires-common.f90
deleted file mode 100644
index 2e112d72de3f..000000000000
--- a/flang/test/Lower/OpenMP/FIR/requires-common.f90
+++ /dev/null
@@ -1,19 +0,0 @@
-! RUN: %flang_fc1 -emit-fir -fopenmp %s -o - | FileCheck %s
-! RUN: %flang_fc1 -emit-fir -fopenmp -fopenmp-is-target-device %s -o - | FileCheck %s
-! RUN: bbc -fopenmp -emit-fir %s -o - | FileCheck %s
-! RUN: bbc -fopenmp -fopenmp-is-target-device -emit-fir %s -o - | FileCheck %s
-
-! This test checks the lowering of requires into MLIR
-
-!CHECK:      module attributes {
-!CHECK-SAME: omp.requires = #omp<clause_requires unified_shared_memory>
-block data init
-  !$omp requires unified_shared_memory
-  integer :: x
-  common /block/ x
-  data x / 10 /
-end
-
-subroutine f
-  !$omp declare target
-end subroutine f
diff --git a/flang/test/Lower/OpenMP/FIR/requires-notarget.f90 b/flang/test/Lower/OpenMP/FIR/requires-notarget.f90
deleted file mode 100644
index bfa509208428..000000000000
--- a/flang/test/Lower/OpenMP/FIR/requires-notarget.f90
+++ /dev/null
@@ -1,14 +0,0 @@
-! RUN: %flang_fc1 -emit-fir -fopenmp %s -o - | FileCheck %s
-! RUN: %flang_fc1 -emit-fir -fopenmp -fopenmp-is-target-device %s -o - | FileCheck %s
-! RUN: bbc -fopenmp -emit-fir %s -o - | FileCheck %s
-! RUN: bbc -fopenmp -fopenmp-is-target-device -emit-fir %s -o - | FileCheck %s
-
-! This test checks that requires lowering into MLIR skips creating the
-! omp.requires attribute with target-related clauses if there are no device
-! functions in the compilation unit
-
-!CHECK:      module attributes {
-!CHECK-NOT:  omp.requires
-program requires
-  !$omp requires unified_shared_memory reverse_offload atomic_default_mem_order(seq_cst)
-end program requires
diff --git a/flang/test/Lower/OpenMP/FIR/requires.f90 b/flang/test/Lower/OpenMP/FIR/requires.f90
deleted file mode 100644
index bc53931b9f24..000000000000
--- a/flang/test/Lower/OpenMP/FIR/requires.f90
+++ /dev/null
@@ -1,14 +0,0 @@
-! RUN: %flang_fc1 -emit-fir -fopenmp %s -o - | FileCheck %s
-! RUN: %flang_fc1 -emit-fir -fopenmp -fopenmp-is-target-device %s -o - | FileCheck %s
-! RUN: bbc -fopenmp -emit-fir %s -o - | FileCheck %s
-! RUN: bbc -fopenmp -fopenmp-is-target-device -emit-fir %s -o - | FileCheck %s
-
-! This test checks the lowering of requires into MLIR
-
-!CHECK:      module attributes {
-!CHECK-SAME: omp.requires = #omp<clause_requires reverse_offload|unified_shared_memory>
-program requires
-  !$omp requires unified_shared_memory reverse_offload atomic_default_mem_order(seq_cst)
-  !$omp target
-  !$omp end target
-end program requires
diff --git a/flang/test/Lower/OpenMP/FIR/rtl-flags.f90 b/flang/test/Lower/OpenMP/FIR/rtl-flags.f90
deleted file mode 100644
index ad8eb9e73213..000000000000
--- a/flang/test/Lower/OpenMP/FIR/rtl-flags.f90
+++ /dev/null
@@ -1,39 +0,0 @@
-!RUN: %flang_fc1 -emit-fir -fopenmp -fopenmp-is-target-device %s -o - | FileCheck %s --check-prefix=DEFAULT-DEVICE-FIR
-!RUN: %flang_fc1 -emit-fir -fopenmp -fopenmp-version=45 %s -o - | FileCheck %s --check-prefix=DEFAULT-HOST-FIR
-!RUN: %flang_fc1 -emit-fir -fopenmp -fopenmp-is-target-device -fopenmp-version=45  %s -o - | FileCheck %s --check-prefix=DEFAULT-DEVICE-FIR-VERSION
-!RUN: %flang_fc1 -emit-fir -fopenmp -fopenmp-version=45 %s -o - | FileCheck %s --check-prefix=DEFAULT-HOST-FIR-VERSION
-!RUN: %flang_fc1 -emit-fir -fopenmp -fopenmp-target-debug -fopenmp-is-target-device %s -o - | FileCheck %s --check-prefix=DBG-DEVICE-FIR
-!RUN: %flang_fc1 -emit-fir -fopenmp -fopenmp-target-debug=111 -fopenmp-is-target-device %s -o - | FileCheck %s --check-prefix=DBG-EQ-DEVICE-FIR
-!RUN: %flang_fc1 -emit-fir -fopenmp -fopenmp-assume-teams-oversubscription -fopenmp-is-target-device %s -o - | FileCheck %s --check-prefix=TEAMS-OSUB-DEVICE-FIR
-!RUN: %flang_fc1 -emit-fir -fopenmp -fopenmp-assume-threads-oversubscription -fopenmp-is-target-device %s -o - | FileCheck %s --check-prefix=THREAD-OSUB-DEVICE-FIR
-!RUN: %flang_fc1 -emit-fir -fopenmp -fopenmp-assume-no-thread-state -fopenmp-is-target-device %s -o - | FileCheck %s --check-prefix=THREAD-STATE-DEVICE-FIR
-!RUN: %flang_fc1 -emit-fir -fopenmp -fopenmp-assume-no-nested-parallelism -fopenmp-is-target-device %s -o - | FileCheck %s --check-prefix=NEST-PAR-DEVICE-FIR
-!RUN: %flang_fc1 -emit-fir -fopenmp -fopenmp-target-debug -fopenmp-assume-teams-oversubscription -fopenmp-assume-no-nested-parallelism -fopenmp-assume-threads-oversubscription -fopenmp-assume-no-thread-state -fopenmp-is-target-device %s -o - | FileCheck %s --check-prefix=ALL-DEVICE-FIR
-!RUN: bbc -emit-fir -fopenmp -fopenmp-is-target-device -o - %s | FileCheck %s --check-prefix=DEFAULT-DEVICE-FIR
-!RUN: bbc -emit-fir -fopenmp -fopenmp-is-target-device -fopenmp-version=45 -o - %s | FileCheck %s --check-prefix=DEFAULT-DEVICE-FIR-VERSION
-!RUN: bbc -emit-fir -fopenmp -o - %s | FileCheck %s --check-prefix=DEFAULT-HOST-FIR
-!RUN: bbc -emit-fir -fopenmp -fopenmp-version=45 -o - %s | FileCheck %s --check-prefix=DEFAULT-HOST-FIR-VERSION
-!RUN: bbc -emit-fir -fopenmp -fopenmp-target-debug=111 -fopenmp-is-target-device -o - %s | FileCheck %s --check-prefix=DBG-EQ-DEVICE-FIR
-!RUN: bbc -emit-fir -fopenmp -fopenmp-assume-teams-oversubscription -fopenmp-is-target-device -o - %s | FileCheck %s --check-prefix=TEAMS-OSUB-DEVICE-FIR
-!RUN: bbc -emit-fir -fopenmp -fopenmp-assume-threads-oversubscription -fopenmp-is-target-device -o - %s | FileCheck %s --check-prefix=THREAD-OSUB-DEVICE-FIR
-!RUN: bbc -emit-fir -fopenmp -fopenmp-assume-no-thread-state -fopenmp-is-target-device -o - %s | FileCheck %s --check-prefix=THREAD-STATE-DEVICE-FIR
-!RUN: bbc -emit-fir -fopenmp -fopenmp-assume-no-nested-parallelism -fopenmp-is-target-device -o - %s | FileCheck %s --check-prefix=NEST-PAR-DEVICE-FIR
-!RUN: bbc -emit-fir -fopenmp -fopenmp-target-debug=1 -fopenmp-assume-teams-oversubscription -fopenmp-assume-no-nested-parallelism -fopenmp-assume-threads-oversubscription -fopenmp-assume-no-thread-state -fopenmp-is-target-device -o - %s | FileCheck %s --check-prefix=ALL-DEVICE-FIR
-
-!DEFAULT-DEVICE-FIR: module attributes {{{.*}}omp.flags = #omp.flags<openmp_device_version = 11>
-!DEFAULT-DEVICE-FIR-SAME: omp.is_target_device = true
-!DEFAULT-DEVICE-FIR-VERSION: module attributes {{{.*}}omp.flags = #omp.flags<openmp_device_version = 45>
-!DEFAULT-DEVICE-FIR-VERSION-SAME: omp.is_target_device = true
-!DEFAULT-DEVICE-FIR-VERSION-SAME: omp.version = #omp.version<version = 45>
-!DEFAULT-HOST-FIR: module attributes {{{.*}}omp.is_target_device = false{{.*}}
-!DEFAULT-HOST-FIR-VERSION: module attributes {{{.*}}omp.is_target_device = false
-!DEFAULT-HOST-FIR-VERSION-SAME: omp.version = #omp.version<version = 45>
-!DBG-DEVICE-FIR: module attributes {{{.*}}omp.flags = #omp.flags<debug_kind = 1, openmp_device_version = 11>
-!DBG-EQ-DEVICE-FIR: module attributes {{{.*}}omp.flags = #omp.flags<debug_kind = 111, openmp_device_version = 11>
-!TEAMS-OSUB-DEVICE-FIR: module attributes {{{.*}}omp.flags = #omp.flags<assume_teams_oversubscription = true, openmp_device_version = 11>
-!THREAD-OSUB-DEVICE-FIR: module attributes {{{.*}}omp.flags = #omp.flags<assume_threads_oversubscription = true, openmp_device_version = 11>
-!THREAD-STATE-DEVICE-FIR: module attributes {{{.*}}omp.flags = #omp.flags<assume_no_thread_state = true, openmp_device_version = 11>
-!NEST-PAR-DEVICE-FIR: module attributes {{{.*}}omp.flags = #omp.flags<assume_no_nested_parallelism = true, openmp_device_version = 11>
-!ALL-DEVICE-FIR: module attributes {{{.*}}omp.flags = #omp.flags<debug_kind = 1, assume_teams_oversubscription = true, assume_threads_oversubscription = true, assume_no_thread_state = true, assume_no_nested_parallelism = true, openmp_device_version = 11>
-subroutine omp_subroutine()
-end subroutine omp_subroutine
diff --git a/flang/test/Lower/OpenMP/FIR/sections-pft.f90 b/flang/test/Lower/OpenMP/FIR/sections-pft.f90
deleted file mode 100644
index 7b20a87022c9..000000000000
--- a/flang/test/Lower/OpenMP/FIR/sections-pft.f90
+++ /dev/null
@@ -1,91 +0,0 @@
-! RUN: %flang_fc1 -fdebug-pre-fir-tree -fopenmp %s | FileCheck %s
-
-subroutine openmp_sections(x, y)
-
-  integer, intent(inout)::x, y
-
-!==============================================================================
-! empty construct
-!==============================================================================
-!$omp sections
-!$omp end sections
-
-!CHECK: OpenMPConstruct
-!CHECK: End OpenMPConstruct
-
-!==============================================================================
-! single section, without `!$omp section`
-!==============================================================================
-!$omp sections
-    call F1()
-!$omp end sections
-
-!CHECK: OpenMPConstruct
-!CHECK:  OpenMPConstruct
-!CHECK:   CallStmt
-!CHECK:  End OpenMPConstruct
-!CHECK: End OpenMPConstruct
-
-!==============================================================================
-! single section with `!$omp section`
-!==============================================================================
-!$omp sections
-  !$omp section
-    call F1
-!$omp end sections
-
-!CHECK: OpenMPConstruct
-!CHECK:  OpenMPConstruct
-!CHECK:   CallStmt
-!CHECK:  End OpenMPConstruct
-!CHECK: End OpenMPConstruct
-
-!==============================================================================
-! multiple sections
-!==============================================================================
-!$omp sections
-  !$omp section
-    call F1
-  !$omp section
-    call F2
-  !$omp section
-    call F3
-!$omp end sections
-
-!CHECK: OpenMPConstruct
-!CHECK:  OpenMPConstruct
-!CHECK:   CallStmt
-!CHECK:  End OpenMPConstruct
-!CHECK:  OpenMPConstruct
-!CHECK:   CallStmt
-!CHECK:  End OpenMPConstruct
-!CHECK:  OpenMPConstruct
-!CHECK:   CallStmt
-!CHECK:  End OpenMPConstruct
-!CHECK: End OpenMPConstruct
-
-!==============================================================================
-! multiple sections with clauses
-!==============================================================================
-!$omp sections PRIVATE(x) FIRSTPRIVATE(y)
-  !$omp section
-    call F1
-  !$omp section
-    call F2
-  !$omp section
-    call F3
-!$omp end sections NOWAIT
-
-!CHECK: OpenMPConstruct
-!CHECK:  OpenMPConstruct
-!CHECK:   CallStmt
-!CHECK:  End OpenMPConstruct
-!CHECK:  OpenMPConstruct
-!CHECK:   CallStmt
-!CHECK:  End OpenMPConstruct
-!CHECK:  OpenMPConstruct
-!CHECK:   CallStmt
-!CHECK:  End OpenMPConstruct
-!CHECK: End OpenMPConstruct
-
-end subroutine openmp_sections
diff --git a/flang/test/Lower/OpenMP/FIR/sections.f90 b/flang/test/Lower/OpenMP/FIR/sections.f90
deleted file mode 100644
index 7b313f3dc0b4..000000000000
--- a/flang/test/Lower/OpenMP/FIR/sections.f90
+++ /dev/null
@@ -1,288 +0,0 @@
-! REQUIRES: openmp_runtime
-
-! This test checks the lowering of OpenMP sections construct with several clauses present
-
-! RUN: %flang_fc1 -emit-fir -flang-deprecated-no-hlfir -fopenmp %s -o - | FileCheck %s
-
-!CHECK: func @_QQmain() attributes {fir.bindc_name = "sample"} {
-!CHECK:   %[[COUNT:.*]] = fir.address_of(@_QFEcount) : !fir.ref<i32>
-!CHECK:   %[[ETA:.*]] = fir.alloca f32 {bindc_name = "eta", uniq_name = "_QFEeta"}
-!CHECK:   %[[CONST_1:.*]] = arith.constant 4 : i64
-!CHECK:   omp.sections allocate(%[[CONST_1]] : i64 -> %0 : !fir.ref<i32>)  {
-!CHECK:     omp.section {
-!CHECK:       %[[PRIVATE_ETA:.*]] = fir.alloca f32 {bindc_name = "eta", pinned, uniq_name = "_QFEeta"}
-!CHECK:       %[[PRIVATE_DOUBLE_COUNT:.*]] = fir.alloca i32 {bindc_name = "double_count", pinned, uniq_name = "_QFEdouble_count"} 
-!CHECK:       %[[const:.*]] = arith.constant 5 : i32
-!CHECK:       fir.store %[[const]] to %[[COUNT]] : !fir.ref<i32>
-!CHECK:       %[[temp_count:.*]] = fir.load %[[COUNT]] : !fir.ref<i32>
-!CHECK:       %[[temp_double_count:.*]] = fir.load %[[PRIVATE_DOUBLE_COUNT]] : !fir.ref<i32>
-!CHECK:       %[[result:.*]] = arith.muli %[[temp_count]], %[[temp_double_count]] : i32
-!CHECK:       {{.*}} = fir.convert %[[result]] : (i32) -> f32
-!CHECK:       fir.store {{.*}} to %[[PRIVATE_ETA]] : !fir.ref<f32>
-!CHECK:       omp.terminator
-!CHECK:     }
-!CHECK:     omp.section {
-!CHECK:       %[[PRIVATE_ETA:.*]] = fir.alloca f32 {bindc_name = "eta", pinned, uniq_name = "_QFEeta"}
-!CHECK:       %[[PRIVATE_DOUBLE_COUNT:.*]] = fir.alloca i32 {bindc_name = "double_count", pinned, uniq_name = "_QFEdouble_count"} 
-!CHECK:       %[[temp:.*]] = fir.load %[[PRIVATE_DOUBLE_COUNT]] : !fir.ref<i32>
-!CHECK:       %[[const:.*]] = arith.constant 1 : i32
-!CHECK:       %[[result:.*]] = arith.addi %[[temp]], %[[const]] : i32
-!CHECK:       fir.store %[[result]] to %[[PRIVATE_DOUBLE_COUNT]] : !fir.ref<i32>
-!CHECK:       omp.terminator
-!CHECK:     }
-!CHECK:     omp.section {
-!CHECK:       %[[PRIVATE_ETA:.*]] = fir.alloca f32 {bindc_name = "eta", pinned, uniq_name = "_QFEeta"}
-!CHECK:       %[[PRIVATE_DOUBLE_COUNT:.*]] = fir.alloca i32 {bindc_name = "double_count", pinned, uniq_name = "_QFEdouble_count"} 
-!CHECK:       %[[temp:.*]] = fir.load %[[PRIVATE_ETA]] : !fir.ref<f32>
-!CHECK:       %[[const:.*]] = arith.constant 7.000000e+00 : f32
-!CHECK:       %[[result:.*]] = arith.subf %[[temp]], %[[const]] {{.*}}: f32
-!CHECK:       fir.store %[[result]] to %[[PRIVATE_ETA]] : !fir.ref<f32>
-!CHECK:       {{.*}} = fir.load %[[COUNT]] : !fir.ref<i32>
-!CHECK:       %[[temp_count:.*]] = fir.convert {{.*}} : (i32) -> f32
-!CHECK:       %[[temp_eta:.*]] = fir.load %[[PRIVATE_ETA]] : !fir.ref<f32>
-!CHECK:       {{.*}} = arith.mulf %[[temp_count]], %[[temp_eta]] {{.*}}: f32
-!CHECK:       %[[result:.*]] = fir.convert {{.*}} : (f32) -> i32
-!CHECK:       fir.store %[[result]] to %[[COUNT]] : !fir.ref<i32>
-!CHECK:       {{.*}} = fir.load %[[COUNT]] : !fir.ref<i32>
-!CHECK:       %[[temp_count:.*]] = fir.convert {{.*}} : (i32) -> f32
-!CHECK:       %[[temp_eta:.*]] = fir.load %[[PRIVATE_ETA]] : !fir.ref<f32>
-!CHECK:       {{.*}} = arith.subf %[[temp_count]], %[[temp_eta]] {{.*}}: f32
-!CHECK:       %[[result:.*]] = fir.convert {{.*}} : (f32) -> i32
-!CHECK:       fir.store %[[result]] to %[[PRIVATE_DOUBLE_COUNT]] : !fir.ref<i32>
-!CHECK:       omp.terminator
-!CHECK:     }
-!CHECK:     omp.terminator
-!CHECK:   }
-!CHECK:   omp.sections nowait {
-!CHECK:     omp.terminator
-!CHECK:   }
-!CHECK:   return
-!CHECK: }
-
-program sample
-    use omp_lib
-    integer :: count = 0, double_count = 1
-    !$omp sections private (eta, double_count) allocate(omp_high_bw_mem_alloc: count)
-        !$omp section
-            count = 1 + 4
-            eta = count * double_count
-        !$omp section
-            double_count = double_count + 1
-        !$omp section
-            eta = eta - 7
-            count = count * eta
-            double_count = count - eta
-    !$omp end sections
-
-    !$omp sections
-    !$omp end sections nowait
-end program sample
-
-!CHECK: func @_QPfirstprivate(%[[ARG:.*]]: !fir.ref<f32> {fir.bindc_name = "alpha"}) {
-!CHECK:   omp.sections {
-!CHECK:     omp.section  {
-!CHECK:         %[[PRIVATE_ALPHA:.*]] = fir.alloca f32 {bindc_name = "alpha", pinned, uniq_name = "_QFfirstprivateEalpha"}
-!CHECK:         %[[temp:.*]] = fir.load %[[ARG]] : !fir.ref<f32>
-!CHECK:         fir.store %[[temp]] to %[[PRIVATE_ALPHA]] : !fir.ref<f32>
-!CHECK:       omp.terminator
-!CHECK:     }
-!CHECK:     omp.terminator
-!CHECK:   }
-!CHECK:   omp.sections {
-!CHECK:     omp.section  {
-!CHECK:       %[[PRIVATE_VAR:.*]] = fir.load %[[ARG]] : !fir.ref<f32>
-!CHECK:       %[[CONSTANT:.*]] = arith.constant 5.000000e+00 : f32
-!CHECK:       %[[PRIVATE_VAR_2:.*]] = arith.mulf %[[PRIVATE_VAR]], %[[CONSTANT]] {{.*}}: f32
-!CHECK:       fir.store %[[PRIVATE_VAR_2]] to %[[ARG]] : !fir.ref<f32>
-!CHECK:       omp.terminator
-!CHECK:     }
-!CHECK:     omp.terminator
-!CHECK:   }
-!CHECK:   return
-!CHECK: }
-
-subroutine firstprivate(alpha)
-    real :: alpha
-    !$omp sections firstprivate(alpha)
-    !$omp end sections
-
-    !$omp sections
-        alpha = alpha * 5
-    !$omp end sections
-end subroutine
-
-subroutine lastprivate()
-	integer :: x
-!CHECK: %[[X:.*]] = fir.alloca i32 {bindc_name = "x", uniq_name = "_QFlastprivateEx"}
-!CHECK: omp.sections   {
-	!$omp sections lastprivate(x)
-!CHECK: omp.section {
-!CHECK: %[[PRIVATE_X:.*]] = fir.alloca i32 {bindc_name = "x", pinned, uniq_name = "_QFlastprivateEx"}
-!CHECK: %[[const:.*]] = arith.constant 10 : i32
-!CHECK: %[[temp:.*]] = fir.load %[[PRIVATE_X]] : !fir.ref<i32>
-!CHECK: %[[result:.*]] = arith.muli %c10_i32, %[[temp]] : i32
-!CHECK: fir.store %[[result]] to %[[PRIVATE_X]] : !fir.ref<i32>
-!CHECK: omp.terminator
-!CHECK: }
-        !$omp section
-            x = x * 10
-!CHECK: omp.section {
-!CHECK: %[[PRIVATE_X:.*]] = fir.alloca i32 {bindc_name = "x", pinned, uniq_name = "_QFlastprivateEx"}
-!CHECK: %[[temp:.*]] = fir.load %[[PRIVATE_X]] : !fir.ref<i32>
-!CHECK: %[[const:.*]] = arith.constant 1 : i32
-!CHECK: %[[result:.*]] = arith.addi %[[temp]], %[[const]] : i32
-!CHECK: fir.store %[[result]] to %[[PRIVATE_X]] : !fir.ref<i32>
-!CHECK: %[[true:.*]] = arith.constant true
-!CHECK: fir.if %[[true]] {
-!CHECK: %[[temp:.*]] = fir.load %[[PRIVATE_X]] : !fir.ref<i32>
-!CHECK: fir.store %[[temp]] to %[[X]] : !fir.ref<i32>
-!CHECK: }
-!CHECK: omp.terminator
-!CHECK: }
-        !$omp section
-            x = x + 1
-!CHECK: omp.terminator
-!CHECK: }
-    !$omp end sections
-    
-!CHECK: omp.sections   {
-    !$omp sections firstprivate(x) lastprivate(x)
-!CHECK: omp.section {
-!CHECK: %[[PRIVATE_X:.*]] = fir.alloca i32 {bindc_name = "x", pinned, uniq_name = "_QFlastprivateEx"}
-!CHECK: %[[temp:.*]] = fir.load %[[X]] : !fir.ref<i32>
-!CHECK: fir.store %[[temp]] to %[[PRIVATE_X]] : !fir.ref<i32>
-!CHECK: omp.barrier
-!CHECK: %[[const:.*]] = arith.constant 10 : i32
-!CHECK: %[[temp:.*]] = fir.load %[[PRIVATE_X]] : !fir.ref<i32>
-!CHECK: %[[result:.*]] = arith.muli %c10_i32, %[[temp]] : i32
-!CHECK: fir.store %[[result]] to %[[PRIVATE_X]] : !fir.ref<i32>
-!CHECK: omp.terminator
-!CHECK: }
-        !$omp section
-            x = x * 10
-!CHECK: omp.section {
-!CHECK: %[[PRIVATE_X:.*]] = fir.alloca i32 {bindc_name = "x", pinned, uniq_name = "_QFlastprivateEx"}
-!CHECK: %[[temp:.*]] = fir.load %[[X]] : !fir.ref<i32>
-!CHECK: fir.store %[[temp]] to %[[PRIVATE_X]] : !fir.ref<i32>
-!CHECK: omp.barrier
-!CHECK: %[[temp:.*]] = fir.load %[[PRIVATE_X]] : !fir.ref<i32>
-!CHECK: %[[const:.*]] = arith.constant 1 : i32
-!CHECK: %[[result:.*]] = arith.addi %[[temp]], %[[const]] : i32
-!CHECK: fir.store %[[result]] to %[[PRIVATE_X]] : !fir.ref<i32>
-!CHECK: %[[true:.*]] = arith.constant true
-!CHECK: fir.if %true {
-!CHECK: %[[temp:.*]] = fir.load %[[PRIVATE_X]] : !fir.ref<i32>
-!CHECK: fir.store %[[temp]] to %[[X]] : !fir.ref<i32>
-!CHECK: }
-!CHECK: omp.terminator
-!CHECK: }
-        !$omp section
-            x = x + 1
-!CHECK: omp.terminator
-!CHECK: }
-    !$omp end sections
-
-!CHECK: omp.sections nowait {
-    !$omp sections firstprivate(x) lastprivate(x)
-!CHECK: omp.section {
-!CHECK: %[[PRIVATE_X:.*]] = fir.alloca i32 {bindc_name = "x", pinned, uniq_name = "_QFlastprivateEx"}
-!CHECK: %[[temp:.*]] = fir.load %[[X]] : !fir.ref<i32>
-!CHECK: fir.store %[[temp]] to %[[PRIVATE_X]] : !fir.ref<i32>
-!CHECK: omp.barrier
-!CHECK: %[[const:.*]] = arith.constant 10 : i32
-!CHECK: %[[temp:.*]] = fir.load %[[PRIVATE_X]] : !fir.ref<i32>
-!CHECK: %[[result:.*]] = arith.muli %c10_i32, %[[temp]] : i32
-!CHECK: fir.store %[[result]] to %[[PRIVATE_X]] : !fir.ref<i32>
-!CHECK: omp.terminator
-!CHECK: }
-        !$omp section
-            x = x * 10
-!CHECK: omp.section {
-!CHECK: %[[PRIVATE_X:.*]] = fir.alloca i32 {bindc_name = "x", pinned, uniq_name = "_QFlastprivateEx"}
-!CHECK: %[[temp:.*]] = fir.load %[[X]] : !fir.ref<i32>
-!CHECK: fir.store %[[temp]] to %[[PRIVATE_X]] : !fir.ref<i32>
-!CHECK: omp.barrier
-!CHECK: %[[temp:.*]] = fir.load %[[PRIVATE_X]] : !fir.ref<i32>
-!CHECK: %[[const:.*]] = arith.constant 1 : i32
-!CHECK: %[[result:.*]] = arith.addi %[[temp]], %[[const]] : i32
-!CHECK: fir.store %[[result]] to %[[PRIVATE_X]] : !fir.ref<i32>
-!CHECK: %[[true:.*]] = arith.constant true
-!CHECK: fir.if %true {
-!CHECK: %[[temp:.*]] = fir.load %[[PRIVATE_X]] : !fir.ref<i32>
-!CHECK: fir.store %[[temp]] to %[[X]] : !fir.ref<i32>
-!CHECK: omp.barrier
-!CHECK: }
-!CHECK: omp.terminator
-!CHECK: }
-        !$omp section
-            x = x + 1
-!CHECK: omp.terminator
-!CHECK: }
-    !$omp end sections nowait
-
-!CHECK: omp.sections {
-!CHECK: omp.section {
-!CHECK: %[[PRIVATE_X:.*]] = fir.alloca i32 {bindc_name = "x", pinned, uniq_name = "_QFlastprivateEx"}
-!CHECK: cf.br ^bb1
-!CHECK: ^bb1:  // pred: ^bb0
-!CHECK: %[[INNER_PRIVATE_X:.*]] = fir.load %[[PRIVATE_X]] : !fir.ref<i32>
-!CHECK: %[[const:.*]] = arith.constant 1 : i32
-!CHECK: %[[result:.*]] = arith.addi %[[INNER_PRIVATE_X]], %[[const]] : i32
-!CHECK: fir.store %[[result]] to %[[PRIVATE_X]] : !fir.ref<i32>
-!CHECK: %[[loaded_value:.*]] = fir.load %[[PRIVATE_X]] : !fir.ref<i32>
-!CHECK: fir.store %[[loaded_value]] to %[[X]] : !fir.ref<i32>
-!CHECK: omp.terminator
-!CHECK: }
-!CHECK: omp.terminator
-!CHECK: }
-!CHECK: return
-!CHECK: }
-
-    !$omp sections lastprivate(x)
-        !$omp section
-                goto 30
-        30  x = x + 1
-    !$omp end sections
-end subroutine
-
-subroutine unstructured_sections_privatization()
-!CHECK: %[[X:.*]] = fir.alloca f32 {bindc_name = "x", uniq_name = "_QFunstructured_sections_privatizationEx"}
-!CHECK: omp.sections {
-!CHECK: omp.section {
-!CHECK: %[[PRIVATE_X:.*]] = fir.alloca f32 {bindc_name = "x", pinned, uniq_name = "_QFunstructured_sections_privatizationEx"}
-!CHECK: cf.br ^bb1
-!CHECK: ^bb1:  // pred: ^bb0
-!CHECK: %[[INNER_PRIVATE_X:.*]] = fir.load %[[PRIVATE_X]] : !fir.ref<f32>
-!CHECK: %[[constant:.*]] = arith.constant 1.000000e+00 : f32
-!CHECK: %[[result:.*]] = arith.addf %[[INNER_PRIVATE_X]], %[[constant]] fastmath<contract> : f32
-!CHECK: fir.store %[[result]] to %[[PRIVATE_X]] : !fir.ref<f32>
-!CHECK: omp.terminator
-!CHECK: }
-!CHECK: omp.terminator
-!CHECK: }
-    !$omp sections private(x)
-        !$omp section
-            goto 40
-        40  x = x + 1
-    !$omp end sections
-!CHECK: omp.sections {
-!CHECK: omp.section {
-!CHECK: %[[PRIVATE_X:.*]] = fir.alloca f32 {bindc_name = "x", pinned, uniq_name = "_QFunstructured_sections_privatizationEx"}
-!CHECK: %[[temp:.*]] = fir.load %[[X]] : !fir.ref<f32>
-!CHECK: fir.store %[[temp]] to %[[PRIVATE_X]] : !fir.ref<f32>
-!CHECK: cf.br ^bb1
-!CHECK: ^bb1:  // pred: ^bb0
-!CHECK: %[[INNER_PRIVATE_X:.*]] = fir.load %[[PRIVATE_X]] : !fir.ref<f32>
-!CHECK: %[[constant:.*]] = arith.constant 1.000000e+00 : f32
-!CHECK: %[[result:.*]] = arith.addf %[[INNER_PRIVATE_X]], %[[constant]] fastmath<contract> : f32
-!CHECK: fir.store %[[result]] to %[[PRIVATE_X]] : !fir.ref<f32>
-!CHECK: omp.terminator
-!CHECK: }
-!CHECK: omp.terminator
-!CHECK: }
-    !$omp sections firstprivate(x)
-        !$omp section
-            goto 50
-        50  x = x + 1
-    !$omp end sections
-end subroutine
diff --git a/flang/test/Lower/OpenMP/FIR/simd.f90 b/flang/test/Lower/OpenMP/FIR/simd.f90
deleted file mode 100644
index db7d30295c45..000000000000
--- a/flang/test/Lower/OpenMP/FIR/simd.f90
+++ /dev/null
@@ -1,175 +0,0 @@
-! Tests for 2.9.3.1 Simd
-
-! RUN: bbc -fopenmp -emit-fir -hlfir=false %s -o - | FileCheck %s
-
-!CHECK-LABEL: func @_QPsimd()
-subroutine simd
-  integer :: i
-  !$OMP SIMD
-  ! CHECK: %[[LB:.*]] = arith.constant 1 : i32
-  ! CHECK-NEXT: %[[UB:.*]] = arith.constant 9 : i32
-  ! CHECK-NEXT: %[[STEP:.*]] = arith.constant 1 : i32
-  ! CHECK-NEXT: omp.simd {
-  ! CHECK-NEXT: omp.loop_nest (%[[I:.*]]) : i32 = (%[[LB]]) to (%[[UB]]) inclusive step (%[[STEP]]) {
-  do i=1, 9
-    ! CHECK: fir.store %[[I]] to %[[LOCAL:.*]] : !fir.ref<i32>
-    ! CHECK: %[[LD:.*]] = fir.load %[[LOCAL]] : !fir.ref<i32>
-    ! CHECK: fir.call @_FortranAioOutputInteger32({{.*}}, %[[LD]]) {{.*}}: (!fir.ref<i8>, i32) -> i1
-    print*, i
-  end do
-  !$OMP END SIMD
-end subroutine
-
-!CHECK-LABEL: func @_QPsimd_with_if_clause
-subroutine simd_with_if_clause(n, threshold)
-  integer :: i, n, threshold
-  !$OMP SIMD IF( n .GE. threshold )
-  ! CHECK: %[[LB:.*]] = arith.constant 1 : i32
-  ! CHECK: %[[UB:.*]] = fir.load %arg0
-  ! CHECK: %[[STEP:.*]] = arith.constant 1 : i32
-  ! CHECK: %[[COND:.*]] = arith.cmpi sge
-  ! CHECK: omp.simd if(%[[COND:.*]]) {
-  ! CHECK-NEXT: omp.loop_nest (%[[I:.*]]) : i32 = (%[[LB]]) to (%[[UB]]) inclusive step (%[[STEP]]) {
-  do i = 1, n
-    ! CHECK: fir.store %[[I]] to %[[LOCAL:.*]] : !fir.ref<i32>
-    ! CHECK: %[[LD:.*]] = fir.load %[[LOCAL]] : !fir.ref<i32>
-    ! CHECK: fir.call @_FortranAioOutputInteger32({{.*}}, %[[LD]]) {{.*}}: (!fir.ref<i8>, i32) -> i1
-    print*, i
-  end do
-  !$OMP END SIMD
-end subroutine
-
-!CHECK-LABEL: func @_QPsimd_with_simdlen_clause
-subroutine simd_with_simdlen_clause(n, threshold)
-  integer :: i, n, threshold
-  !$OMP SIMD SIMDLEN(2)
-  ! CHECK: %[[LB:.*]] = arith.constant 1 : i32
-  ! CHECK: %[[UB:.*]] = fir.load %arg0
-  ! CHECK: %[[STEP:.*]] = arith.constant 1 : i32
-  ! CHECK: omp.simd simdlen(2) {
-  ! CHECK-NEXT: omp.loop_nest (%[[I:.*]]) : i32 = (%[[LB]]) to (%[[UB]]) inclusive step (%[[STEP]]) {
-  do i = 1, n
-    ! CHECK: fir.store %[[I]] to %[[LOCAL:.*]] : !fir.ref<i32>
-    ! CHECK: %[[LD:.*]] = fir.load %[[LOCAL]] : !fir.ref<i32>
-    ! CHECK: fir.call @_FortranAioOutputInteger32({{.*}}, %[[LD]]) {{.*}}: (!fir.ref<i8>, i32) -> i1
-    print*, i
-  end do
-  !$OMP END SIMD
-end subroutine
-
-!CHECK-LABEL: func @_QPsimd_with_simdlen_clause_from_param
-subroutine simd_with_simdlen_clause_from_param(n, threshold)
-  integer :: i, n, threshold
-  integer, parameter :: simdlen = 2;
-  !$OMP SIMD SIMDLEN(simdlen)
-  ! CHECK: %[[LB:.*]] = arith.constant 1 : i32
-  ! CHECK: %[[UB:.*]] = fir.load %arg0
-  ! CHECK: %[[STEP:.*]] = arith.constant 1 : i32
-  ! CHECK: omp.simd simdlen(2) {
-  ! CHECK-NEXT: omp.loop_nest (%[[I:.*]]) : i32 = (%[[LB]]) to (%[[UB]]) inclusive step (%[[STEP]]) {
-  do i = 1, n
-    ! CHECK: fir.store %[[I]] to %[[LOCAL:.*]] : !fir.ref<i32>
-    ! CHECK: %[[LD:.*]] = fir.load %[[LOCAL]] : !fir.ref<i32>
-    ! CHECK: fir.call @_FortranAioOutputInteger32({{.*}}, %[[LD]]) {{.*}}: (!fir.ref<i8>, i32) -> i1
-    print*, i
-  end do
-  !$OMP END SIMD
-end subroutine
-
-!CHECK-LABEL: func @_QPsimd_with_simdlen_clause_from_expr_from_param
-subroutine simd_with_simdlen_clause_from_expr_from_param(n, threshold)
-  integer :: i, n, threshold
-  integer, parameter :: simdlen = 2;
-  !$OMP SIMD SIMDLEN(simdlen*2 + 2)
-  ! CHECK: %[[LB:.*]] = arith.constant 1 : i32
-  ! CHECK: %[[UB:.*]] = fir.load %arg0
-  ! CHECK: %[[STEP:.*]] = arith.constant 1 : i32
-  ! CHECK: omp.simd simdlen(6) {
-  ! CHECK-NEXT: omp.loop_nest (%[[I:.*]]) : i32 = (%[[LB]]) to (%[[UB]]) inclusive step (%[[STEP]]) {
-  do i = 1, n
-    ! CHECK: fir.store %[[I]] to %[[LOCAL:.*]] : !fir.ref<i32>
-    ! CHECK: %[[LD:.*]] = fir.load %[[LOCAL]] : !fir.ref<i32>
-    ! CHECK: fir.call @_FortranAioOutputInteger32({{.*}}, %[[LD]]) {{.*}}: (!fir.ref<i8>, i32) -> i1
-    print*, i
-  end do
-  !$OMP END SIMD
-end subroutine
-
-!CHECK-LABEL: func @_QPsimd_with_safelen_clause
-subroutine simd_with_safelen_clause(n, threshold)
-  integer :: i, n, threshold
-  !$OMP SIMD SAFELEN(2)
-  ! CHECK: %[[LB:.*]] = arith.constant 1 : i32
-  ! CHECK: %[[UB:.*]] = fir.load %arg0
-  ! CHECK: %[[STEP:.*]] = arith.constant 1 : i32
-  ! CHECK: omp.simd safelen(2) {
-  ! CHECK-NEXT: omp.loop_nest (%[[I:.*]]) : i32 = (%[[LB]]) to (%[[UB]]) inclusive step (%[[STEP]]) {
-  do i = 1, n
-    ! CHECK: fir.store %[[I]] to %[[LOCAL:.*]] : !fir.ref<i32>
-    ! CHECK: %[[LD:.*]] = fir.load %[[LOCAL]] : !fir.ref<i32>
-    ! CHECK: fir.call @_FortranAioOutputInteger32({{.*}}, %[[LD]]) {{.*}}: (!fir.ref<i8>, i32) -> i1
-    print*, i
-  end do
-  !$OMP END SIMD
-end subroutine
-
-!CHECK-LABEL: func @_QPsimd_with_safelen_clause_from_expr_from_param
-subroutine simd_with_safelen_clause_from_expr_from_param(n, threshold)
-  integer :: i, n, threshold
-  integer, parameter :: safelen = 2;
-  !$OMP SIMD SAFELEN(safelen*2 + 2)
-  ! CHECK: %[[LB:.*]] = arith.constant 1 : i32
-  ! CHECK: %[[UB:.*]] = fir.load %arg0
-  ! CHECK: %[[STEP:.*]] = arith.constant 1 : i32
-  ! CHECK: omp.simd safelen(6) {
-  ! CHECK-NEXT: omp.loop_nest (%[[I:.*]]) : i32 = (%[[LB]]) to (%[[UB]]) inclusive step (%[[STEP]]) {
-  do i = 1, n
-    ! CHECK: fir.store %[[I]] to %[[LOCAL:.*]] : !fir.ref<i32>
-    ! CHECK: %[[LD:.*]] = fir.load %[[LOCAL]] : !fir.ref<i32>
-    ! CHECK: fir.call @_FortranAioOutputInteger32({{.*}}, %[[LD]]) {{.*}}: (!fir.ref<i8>, i32) -> i1
-    print*, i
-  end do
-  !$OMP END SIMD
-end subroutine
-
-!CHECK-LABEL: func @_QPsimd_with_simdlen_safelen_clause
-subroutine simd_with_simdlen_safelen_clause(n, threshold)
-  integer :: i, n, threshold
-  !$OMP SIMD SIMDLEN(1) SAFELEN(2)
-  ! CHECK: %[[LB:.*]] = arith.constant 1 : i32
-  ! CHECK: %[[UB:.*]] = fir.load %arg0
-  ! CHECK: %[[STEP:.*]] = arith.constant 1 : i32
-  ! CHECK: omp.simd simdlen(1) safelen(2) {
-  ! CHECK-NEXT: omp.loop_nest (%[[I:.*]]) : i32 = (%[[LB]]) to (%[[UB]]) inclusive step (%[[STEP]]) {
-  do i = 1, n
-    ! CHECK: fir.store %[[I]] to %[[LOCAL:.*]] : !fir.ref<i32>
-    ! CHECK: %[[LD:.*]] = fir.load %[[LOCAL]] : !fir.ref<i32>
-    ! CHECK: fir.call @_FortranAioOutputInteger32({{.*}}, %[[LD]]) {{.*}}: (!fir.ref<i8>, i32) -> i1
-    print*, i
-  end do
-  !$OMP END SIMD
-end subroutine
-
-!CHECK-LABEL: func @_QPsimd_with_collapse_clause
-subroutine simd_with_collapse_clause(n)
-  integer :: i, j, n
-  integer :: A(n,n)
-  ! CHECK: %[[LOWER_I:.*]] = arith.constant 1 : i32
-  ! CHECK: %[[UPPER_I:.*]] = fir.load %[[PARAM_ARG:.*]] : !fir.ref<i32>
-  ! CHECK: %[[STEP_I:.*]] = arith.constant 1 : i32
-  ! CHECK: %[[LOWER_J:.*]] = arith.constant 1 : i32
-  ! CHECK: %[[UPPER_J:.*]] = fir.load %[[PARAM_ARG:.*]] : !fir.ref<i32>
-  ! CHECK: %[[STEP_J:.*]] = arith.constant 1 : i32
-  ! CHECK: omp.simd {
-  ! CHECK-NEXT: omp.loop_nest (%[[ARG_0:.*]], %[[ARG_1:.*]]) : i32 = (
-  ! CHECK-SAME:               %[[LOWER_I]], %[[LOWER_J]]) to (
-  ! CHECK-SAME:               %[[UPPER_I]], %[[UPPER_J]]) inclusive step (
-  ! CHECK-SAME:               %[[STEP_I]], %[[STEP_J]]) {
-  !$OMP SIMD COLLAPSE(2)
-  do i = 1, n
-    do j = 1, n
-       A(i,j) = i + j
-    end do
-  end do
-  !$OMP END SIMD
-end subroutine
diff --git a/flang/test/Lower/OpenMP/FIR/simple-barrier.f90 b/flang/test/Lower/OpenMP/FIR/simple-barrier.f90
deleted file mode 100644
index c621b8062eaa..000000000000
--- a/flang/test/Lower/OpenMP/FIR/simple-barrier.f90
+++ /dev/null
@@ -1,6 +0,0 @@
-! RUN: bbc -fopenmp -emit-fir -o - %s | FileCheck %s
-
-subroutine sample()
-! CHECK: omp.barrier
-!$omp barrier
-end subroutine sample
diff --git a/flang/test/Lower/OpenMP/FIR/single.f90 b/flang/test/Lower/OpenMP/FIR/single.f90
deleted file mode 100644
index 65ae07c2c284..000000000000
--- a/flang/test/Lower/OpenMP/FIR/single.f90
+++ /dev/null
@@ -1,123 +0,0 @@
-! REQUIRES: openmp_runtime
-
-!RUN: %flang_fc1 -emit-fir -flang-deprecated-no-hlfir -fopenmp %s -o - | FileCheck %s
-!RUN: bbc -emit-fir -hlfir=false -fopenmp %s -o - | FileCheck %s
-
-!===============================================================================
-! Single construct
-!===============================================================================
-
-!CHECK-LABEL: func @_QPomp_single
-!CHECK-SAME: (%[[x:.*]]: !fir.ref<i32> {fir.bindc_name = "x"})
-subroutine omp_single(x)
-  integer, intent(inout) :: x
-  !CHECK: omp.parallel
-  !$omp parallel
-  !CHECK: omp.single
-  !$omp single
-    !CHECK: %[[xval:.*]] = fir.load %[[x]] : !fir.ref<i32>
-    !CHECK: %[[res:.*]] = arith.addi %[[xval]], %{{.*}} : i32
-    !CHECK: fir.store %[[res]] to %[[x]] : !fir.ref<i32>
-    x = x + 12
-  !CHECK: omp.terminator
-  !$omp end single
-  !CHECK: omp.terminator
-  !$omp end parallel
-end subroutine omp_single
-
-!===============================================================================
-! Single construct with nowait
-!===============================================================================
-
-!CHECK-LABEL: func @_QPomp_single_nowait
-!CHECK-SAME: (%[[x:.*]]: !fir.ref<i32> {fir.bindc_name = "x"})
-subroutine omp_single_nowait(x)
-  integer, intent(inout) :: x
-  !CHECK: omp.parallel
-  !$omp parallel
-  !CHECK: omp.single nowait
-  !$omp single
-    !CHECK: %[[xval:.*]] = fir.load %[[x]] : !fir.ref<i32>
-    !CHECK: %[[res:.*]] = arith.addi %[[xval]], %{{.*}} : i32
-    !CHECK: fir.store %[[res]] to %[[x]] : !fir.ref<i32>
-    x = x + 12
-  !CHECK: omp.terminator
-  !$omp end single nowait
-  !CHECK: omp.terminator
-  !$omp end parallel
-end subroutine omp_single_nowait
-
-!===============================================================================
-! Single construct with allocate
-!===============================================================================
-
-!CHECK-LABEL: func @_QPsingle_allocate
-subroutine single_allocate()
-  use omp_lib
-  integer :: x
-  !CHECK: omp.parallel {
-  !$omp parallel
-  !CHECK: omp.single allocate(%{{.+}} : i64 -> %{{.+}} : !fir.ref<i32>) {
-  !$omp single allocate(omp_high_bw_mem_alloc: x) private(x)
-  !CHECK: arith.addi
-  x = x + 12
-  !CHECK: omp.terminator
-  !$omp end single
-  !CHECK: omp.terminator
-  !$omp end parallel
-end subroutine single_allocate
-
-!===============================================================================
-! Single construct with private/firstprivate
-!===============================================================================
-
-! CHECK-LABEL: func.func @_QPsingle_privatization(
-! CHECK-SAME:                                     %[[VAL_0:.*]]: !fir.ref<f32> {fir.bindc_name = "x"},
-! CHECK-SAME:                                     %[[VAL_1:.*]]: !fir.ref<f64> {fir.bindc_name = "y"}) {
-! CHECK:         omp.single   {
-! CHECK:           %[[VAL_2:.*]] = fir.alloca f32 {bindc_name = "x", pinned, uniq_name = "_QFsingle_privatizationEx"}
-! CHECK:           %[[VAL_3:.*]] = fir.alloca f64 {bindc_name = "y", pinned, uniq_name = "_QFsingle_privatizationEy"}
-! CHECK:           %[[VAL_4:.*]] = fir.load %[[VAL_1]] : !fir.ref<f64>
-! CHECK:           fir.store %[[VAL_4]] to %[[VAL_3]] : !fir.ref<f64>
-! CHECK:           fir.call @_QPbar(%[[VAL_2]], %[[VAL_3]]) {{.*}}: (!fir.ref<f32>, !fir.ref<f64>) -> ()
-! CHECK:           omp.terminator
-! CHECK:         }
-! CHECK:         return
-! CHECK:       }
-
-subroutine single_privatization(x, y)
-  real :: x
-  real(8) :: y
-
-  !$omp single private(x) firstprivate(y)
-  call bar(x, y)
-  !$omp end single
-end subroutine
-
-! CHECK-LABEL: func.func @_QPsingle_privatization2(
-! CHECK-SAME:                                      %[[VAL_0:.*]]: !fir.ref<f32> {fir.bindc_name = "x"},
-! CHECK-SAME:                                      %[[VAL_1:.*]]: !fir.ref<f64> {fir.bindc_name = "y"}) {
-! CHECK:         omp.parallel   {
-! CHECK:           omp.single   {
-! CHECK:             %[[VAL_2:.*]] = fir.alloca f32 {bindc_name = "x", pinned, uniq_name = "_QFsingle_privatization2Ex"}
-! CHECK:             %[[VAL_3:.*]] = fir.alloca f64 {bindc_name = "y", pinned, uniq_name = "_QFsingle_privatization2Ey"}
-! CHECK:             %[[VAL_4:.*]] = fir.load %[[VAL_1]] : !fir.ref<f64>
-! CHECK:             fir.store %[[VAL_4]] to %[[VAL_3]] : !fir.ref<f64>
-! CHECK:             fir.call @_QPbar(%[[VAL_2]], %[[VAL_3]]) {{.*}}: (!fir.ref<f32>, !fir.ref<f64>) -> ()
-! CHECK:             omp.terminator
-! CHECK:           }
-! CHECK:           omp.terminator
-! CHECK:         }
-! CHECK:         return
-! CHECK:       }
-
-subroutine single_privatization2(x, y)
-  real :: x
-  real(8) :: y
-
-  !$omp parallel
-  !$omp single private(x) firstprivate(y)
-  call bar(x, y)
-  !$omp end single
-  !$omp end parallel
-end subroutine
diff --git a/flang/test/Lower/OpenMP/FIR/stop-stmt-in-region.f90 b/flang/test/Lower/OpenMP/FIR/stop-stmt-in-region.f90
deleted file mode 100644
index d6c10bdee88d..000000000000
--- a/flang/test/Lower/OpenMP/FIR/stop-stmt-in-region.f90
+++ /dev/null
@@ -1,151 +0,0 @@
-! This test checks lowering of stop statement in OpenMP region.
-
-! RUN: bbc -fopenmp -emit-fir -hlfir=false %s -o - | FileCheck %s
-! RUN: %flang_fc1 -emit-fir -flang-deprecated-no-hlfir -fopenmp %s -o - | FileCheck %s
-
-! CHECK-LABEL: func.func @_QPtest_stop_in_region1() {
-! CHECK:         omp.parallel   {
-! CHECK:           %[[VAL_0:.*]] = arith.constant 1 : i32
-! CHECK:           %[[VAL_1:.*]] = arith.constant false
-! CHECK:           %[[VAL_2:.*]] = arith.constant false
-! CHECK:           %[[VAL_3:.*]] = fir.call @_FortranAStopStatement(%[[VAL_0]], %[[VAL_1]], %[[VAL_2]]) {{.*}} : (i32, i1, i1) -> none
-! CHECK-NOT:       fir.unreachable
-! CHECK:           omp.terminator
-! CHECK:         }
-! CHECK:         return
-! CHECK:       }
-
-subroutine test_stop_in_region1()
-  !$omp parallel
-    stop 1
-  !$omp end parallel
-end
-
-! CHECK-LABEL: func.func @_QPtest_stop_in_region2() {
-! CHECK:         %[[VAL_0:.*]] = fir.alloca i32 {bindc_name = "x", uniq_name = "_QFtest_stop_in_region2Ex"}
-! CHECK:         omp.parallel   {
-! CHECK:           %[[VAL_1:.*]] = arith.constant 1 : i32
-! CHECK:           %[[VAL_2:.*]] = arith.constant false
-! CHECK:           %[[VAL_3:.*]] = arith.constant false
-! CHECK:           %[[VAL_4:.*]] = fir.call @_FortranAStopStatement(%[[VAL_1]], %[[VAL_2]], %[[VAL_3]]) {{.*}} : (i32, i1, i1) -> none
-! CHECK:           omp.terminator
-! CHECK:         }
-! CHECK:         return
-! CHECK:       }
-
-subroutine test_stop_in_region2()
-  integer :: x
-  !$omp parallel
-    stop 1
-    x = 2
-  !$omp end parallel
-end
-
-! CHECK-LABEL: func.func @_QPtest_stop_in_region3() {
-! CHECK:         %[[VAL_0:.*]] = fir.alloca i32 {bindc_name = "x", uniq_name = "_QFtest_stop_in_region3Ex"}
-! CHECK:         omp.parallel   {
-! CHECK:           %[[VAL_1:.*]] = arith.constant 3 : i32
-! CHECK:           fir.store %[[VAL_1]] to %[[VAL_0]] : !fir.ref<i32>
-! CHECK:           %[[VAL_2:.*]] = fir.load %[[VAL_0]] : !fir.ref<i32>
-! CHECK:           %[[VAL_3:.*]] = arith.constant 1 : i32
-! CHECK:           %[[VAL_4:.*]] = arith.cmpi sgt, %[[VAL_2]], %[[VAL_3]] : i32
-! CHECK:           cf.cond_br %[[VAL_4]], ^bb1, ^bb2
-! CHECK:         ^bb1:
-! CHECK:           %[[VAL_5:.*]] = fir.load %[[VAL_0]] : !fir.ref<i32>
-! CHECK:           %[[VAL_6:.*]] = arith.constant false
-! CHECK:           %[[VAL_7:.*]] = arith.constant false
-! CHECK:           %[[VAL_8:.*]] = fir.call @_FortranAStopStatement(%[[VAL_5]], %[[VAL_6]], %[[VAL_7]]) {{.*}} : (i32, i1, i1) -> none
-! CHECK:           omp.terminator
-! CHECK:         ^bb2:
-! CHECK:           omp.terminator
-! CHECK:         }
-! CHECK:         return
-! CHECK:       }
-
-subroutine test_stop_in_region3()
-  integer :: x
-  !$omp parallel
-    x = 3
-    if (x > 1) stop x
-  !$omp end parallel
-end
-
-! CHECK-LABEL: func.func @_QPtest_stop_in_region4() {
-! CHECK:         %[[VAL_0:.*]] = fir.alloca i32 {adapt.valuebyref, pinned}
-! CHECK:         %[[VAL_1:.*]] = fir.alloca i32 {bindc_name = "i", uniq_name = "_QFtest_stop_in_region4Ei"}
-! CHECK:         %[[VAL_2:.*]] = fir.alloca i32 {bindc_name = "x", uniq_name = "_QFtest_stop_in_region4Ex"}
-! CHECK:         %[[VAL_3:.*]] = arith.constant 1 : i32
-! CHECK:         %[[VAL_4:.*]] = arith.constant 10 : i32
-! CHECK:         %[[VAL_5:.*]] = arith.constant 1 : i32
-! CHECK:         omp.wsloop   for  (%[[VAL_6:.*]]) : i32 = (%[[VAL_3]]) to (%[[VAL_4]]) inclusive step (%[[VAL_5]]) {
-! CHECK:           fir.store %[[VAL_6]] to %[[VAL_0]] : !fir.ref<i32>
-! CHECK:           cf.br ^bb1
-! CHECK:         ^bb1:
-! CHECK:           %[[VAL_7:.*]] = arith.constant 3 : i32
-! CHECK:           fir.store %[[VAL_7]] to %[[VAL_2]] : !fir.ref<i32>
-! CHECK:           %[[VAL_8:.*]] = fir.load %[[VAL_2]] : !fir.ref<i32>
-! CHECK:           %[[VAL_9:.*]] = arith.constant 1 : i32
-! CHECK:           %[[VAL_10:.*]] = arith.cmpi sgt, %[[VAL_8]], %[[VAL_9]] : i32
-! CHECK:           cf.cond_br %[[VAL_10]], ^bb2, ^bb3
-! CHECK:         ^bb2:
-! CHECK:           %[[VAL_11:.*]] = fir.load %[[VAL_2]] : !fir.ref<i32>
-! CHECK:           %[[VAL_12:.*]] = arith.constant false
-! CHECK:           %[[VAL_13:.*]] = arith.constant false
-! CHECK:           %[[VAL_14:.*]] = fir.call @_FortranAStopStatement(%[[VAL_11]], %[[VAL_12]], %[[VAL_13]]) {{.*}} : (i32, i1, i1) -> none
-! CHECK:           omp.yield
-! CHECK:         ^bb3:
-! CHECK:           omp.yield
-! CHECK:         }
-! CHECK:         cf.br ^bb1
-! CHECK:       ^bb1:
-! CHECK:         return
-! CHECK:       }
-
-subroutine test_stop_in_region4()
-  integer :: x
-  !$omp do
-  do i = 1, 10
-    x = 3
-    if (x > 1) stop x
-  enddo
-  !$omp end do
-end
-
-
-!CHECK-LABEL: func.func @_QPtest_stop_in_region5
-!CHECK:   omp.parallel   {
-!CHECK:     {{.*}} fir.call @_FortranAStopStatement({{.*}}, {{.*}}, {{.*}}) fastmath<contract> : (i32, i1, i1) -> none
-!CHECK:     omp.terminator
-!CHECK:   }
-!CHECK:   return
-
-subroutine test_stop_in_region5()
-  !$omp parallel
-  block
-    stop 1
-  end block
-  !$omp end parallel
-end
-
-!CHECK-LABEL: func.func @_QPtest_stop_in_region6
-!CHECK:  omp.parallel   {
-!CHECK:    cf.cond_br %{{.*}}, ^[[BB1:.*]], ^[[BB2:.*]]
-!CHECK:  ^[[BB1]]:
-!CHECK:    {{.*}}fir.call @_FortranAStopStatement({{.*}}, {{.*}}, {{.*}}) fastmath<contract> : (i32, i1, i1) -> none
-!CHECK:    omp.terminator
-!CHECK:  ^[[BB2]]:
-!CHECK:    {{.*}}fir.call @_FortranAStopStatement({{.*}}, {{.*}}, {{.*}}) fastmath<contract> : (i32, i1, i1) -> none
-!CHECK:    omp.terminator
-!CHECK:  }
-!CHECK:  return
-
-subroutine test_stop_in_region6(x)
-  integer :: x
-  !$omp parallel
-  if (x .gt. 1) then
-    stop 1
-  else
-    stop 2
-  end if
-  !$omp end parallel
-end
diff --git a/flang/test/Lower/OpenMP/FIR/target.f90 b/flang/test/Lower/OpenMP/FIR/target.f90
deleted file mode 100644
index ca3162340d78..000000000000
--- a/flang/test/Lower/OpenMP/FIR/target.f90
+++ /dev/null
@@ -1,550 +0,0 @@
-!RUN: %flang_fc1 -emit-fir -flang-deprecated-no-hlfir -fopenmp %s -o - | FileCheck %s
-
-!===============================================================================
-! Target_Enter Simple
-!===============================================================================
-
-!CHECK-LABEL: func.func @_QPomp_target_enter_simple() {
-subroutine omp_target_enter_simple
-   integer :: a(1024)
-   !CHECK: %[[BOUNDS:.*]] = omp.map.bounds   lower_bound({{.*}}) upper_bound({{.*}}) extent({{.*}}) stride({{.*}}) start_idx({{.*}})
-   !CHECK: %[[MAP:.*]] = omp.map.info var_ptr({{.*}})   map_clauses(to) capture(ByRef) bounds(%[[BOUNDS]]) -> !fir.ref<!fir.array<1024xi32>> {name = "a"}
-   !CHECK: omp.target_enter_data   map_entries(%[[MAP]] : !fir.ref<!fir.array<1024xi32>>)
-   !$omp target enter data map(to: a)
-end subroutine omp_target_enter_simple
-
-!===============================================================================
-! Target_Enter Map types
-!===============================================================================
-
-!CHECK-LABEL: func.func @_QPomp_target_enter_mt() {
-subroutine omp_target_enter_mt
-   integer :: a(1024)
-   integer :: b(1024)
-   integer :: c(1024)
-   integer :: d(1024)
-   !CHECK: %[[BOUNDS_0:.*]] = omp.map.bounds   lower_bound({{.*}}) upper_bound({{.*}}) extent({{.*}}) stride({{.*}}) start_idx({{.*}})
-   !CHECK: %[[MAP_0:.*]] = omp.map.info var_ptr({{.*}})   map_clauses(to) capture(ByRef) bounds(%[[BOUNDS_0]]) -> !fir.ref<!fir.array<1024xi32>> {name = "a"}
-   !CHECK: %[[BOUNDS_1:.*]] = omp.map.bounds   lower_bound({{.*}}) upper_bound({{.*}}) extent({{.*}}) stride({{.*}}) start_idx({{.*}})
-   !CHECK: %[[MAP_1:.*]] = omp.map.info var_ptr(%{{.*}})  map_clauses(to) capture(ByRef) bounds(%[[BOUNDS_1]]) -> !fir.ref<!fir.array<1024xi32>> {name = "b"}
-   !CHECK: %[[BOUNDS_2:.*]] = omp.map.bounds   lower_bound({{.*}}) upper_bound({{.*}}) extent({{.*}}) stride({{.*}}) start_idx({{.*}})
-   !CHECK: %[[MAP_2:.*]] = omp.map.info var_ptr({{.*}})   map_clauses(always, exit_release_or_enter_alloc) capture(ByRef) bounds(%[[BOUNDS_2]]) -> !fir.ref<!fir.array<1024xi32>> {name = "c"}
-   !CHECK: %[[BOUNDS_3:.*]] = omp.map.bounds   lower_bound({{.*}}) upper_bound({{.*}}) extent({{.*}}) stride({{.*}}) start_idx({{.*}})
-   !CHECK: %[[MAP_3:.*]] = omp.map.info var_ptr({{.*}})   map_clauses(to) capture(ByRef) bounds(%[[BOUNDS_3]]) -> !fir.ref<!fir.array<1024xi32>> {name = "d"}
-   !CHECK: omp.target_enter_data   map_entries(%[[MAP_0]], %[[MAP_1]], %[[MAP_2]], %[[MAP_3]] : !fir.ref<!fir.array<1024xi32>>, !fir.ref<!fir.array<1024xi32>>, !fir.ref<!fir.array<1024xi32>>, !fir.ref<!fir.array<1024xi32>>)
-   !$omp target enter data map(to: a, b) map(always, alloc: c) map(to: d)
-end subroutine omp_target_enter_mt
-
-!===============================================================================
-! `Nowait` clause
-!===============================================================================
-
-!CHECK-LABEL: func.func @_QPomp_target_enter_nowait() {
-subroutine omp_target_enter_nowait
-   integer :: a(1024)
-   !CHECK: %[[BOUNDS:.*]] = omp.map.bounds   lower_bound({{.*}}) upper_bound({{.*}}) extent({{.*}}) stride({{.*}}) start_idx({{.*}})
-   !CHECK: %[[MAP:.*]] = omp.map.info var_ptr({{.*}})   map_clauses(to) capture(ByRef) bounds(%[[BOUNDS]]) -> !fir.ref<!fir.array<1024xi32>> {name = "a"}
-   !CHECK: omp.target_enter_data  nowait map_entries(%[[MAP]] : !fir.ref<!fir.array<1024xi32>>)
-   !$omp target enter data map(to: a) nowait
-end subroutine omp_target_enter_nowait
-
-!===============================================================================
-! `if` clause
-!===============================================================================
-
-!CHECK-LABEL: func.func @_QPomp_target_enter_if() {
-subroutine omp_target_enter_if
-   integer :: a(1024)
-   integer :: i
-   i = 5
-   !CHECK: %[[VAL_3:.*]] = fir.load %[[VAL_1:.*]] : !fir.ref<i32>
-   !CHECK: %[[VAL_4:.*]] = arith.constant 10 : i32
-   !CHECK: %[[VAL_5:.*]] = arith.cmpi slt, %[[VAL_3]], %[[VAL_4]] : i32
-   !CHECK: %[[BOUNDS:.*]] = omp.map.bounds   lower_bound({{.*}}) upper_bound({{.*}}) extent({{.*}}) stride({{.*}}) start_idx({{.*}})
-   !CHECK: %[[MAP:.*]] = omp.map.info var_ptr({{.*}})   map_clauses(to) capture(ByRef) bounds(%[[BOUNDS]]) -> !fir.ref<!fir.array<1024xi32>> {name = "a"}
-   !CHECK: omp.target_enter_data   if(%[[VAL_5]] : i1) map_entries(%[[MAP]] : !fir.ref<!fir.array<1024xi32>>)
-   !$omp target enter data if(i<10) map(to: a)
-end subroutine omp_target_enter_if
-
-!===============================================================================
-! `device` clause
-!===============================================================================
-
-!CHECK-LABEL: func.func @_QPomp_target_enter_device() {
-subroutine omp_target_enter_device
-   integer :: a(1024)
-   !CHECK: %[[VAL_1:.*]] = arith.constant 2 : i32
-   !CHECK: %[[BOUNDS:.*]] = omp.map.bounds   lower_bound({{.*}}) upper_bound({{.*}}) extent({{.*}}) stride({{.*}}) start_idx({{.*}})
-   !CHECK: %[[MAP:.*]] = omp.map.info var_ptr({{.*}})   map_clauses(to) capture(ByRef) bounds(%[[BOUNDS]]) -> !fir.ref<!fir.array<1024xi32>> {name = "a"}
-   !CHECK: omp.target_enter_data   device(%[[VAL_1]] : i32) map_entries(%[[MAP]] : !fir.ref<!fir.array<1024xi32>>)
-   !$omp target enter data map(to: a) device(2)
-end subroutine omp_target_enter_device
-
-!===============================================================================
-! Target_Exit Simple
-!===============================================================================
-
-!CHECK-LABEL: func.func @_QPomp_target_exit_simple() {
-subroutine omp_target_exit_simple
-   integer :: a(1024)
-   !CHECK: %[[BOUNDS:.*]] = omp.map.bounds   lower_bound({{.*}}) upper_bound({{.*}}) extent({{.*}}) stride({{.*}}) start_idx({{.*}})
-   !CHECK: %[[MAP:.*]] = omp.map.info var_ptr({{.*}})   map_clauses(from) capture(ByRef) bounds(%[[BOUNDS]]) -> !fir.ref<!fir.array<1024xi32>> {name = "a"}
-   !CHECK: omp.target_exit_data   map_entries(%[[MAP]] : !fir.ref<!fir.array<1024xi32>>)
-   !$omp target exit data map(from: a)
-end subroutine omp_target_exit_simple
-
-!===============================================================================
-! Target_Exit Map types
-!===============================================================================
-
-!CHECK-LABEL: func.func @_QPomp_target_exit_mt() {
-subroutine omp_target_exit_mt
-   integer :: a(1024)
-   integer :: b(1024)
-   integer :: c(1024)
-   integer :: d(1024)
-   integer :: e(1024)
-   !CHECK: %[[BOUNDS_0:.*]] = omp.map.bounds   lower_bound({{.*}}) upper_bound({{.*}}) extent({{.*}}) stride({{.*}}) start_idx({{.*}})
-   !CHECK: %[[MAP_0:.*]] = omp.map.info var_ptr({{.*}})   map_clauses(from) capture(ByRef) bounds(%[[BOUNDS_0]]) -> !fir.ref<!fir.array<1024xi32>> {name = "a"}
-   !CHECK: %[[BOUNDS_1:.*]] = omp.map.bounds   lower_bound({{.*}}) upper_bound({{.*}}) extent({{.*}}) stride({{.*}}) start_idx({{.*}})
-   !CHECK: %[[MAP_1:.*]] = omp.map.info var_ptr({{.*}})   map_clauses(from) capture(ByRef) bounds(%[[BOUNDS_1]]) -> !fir.ref<!fir.array<1024xi32>> {name = "b"}
-   !CHECK: %[[BOUNDS_2:.*]] = omp.map.bounds   lower_bound({{.*}}) upper_bound({{.*}}) extent({{.*}}) stride({{.*}}) start_idx({{.*}})
-   !CHECK: %[[MAP_2:.*]] = omp.map.info var_ptr({{.*}})   map_clauses(exit_release_or_enter_alloc) capture(ByRef) bounds(%[[BOUNDS_2]]) -> !fir.ref<!fir.array<1024xi32>> {name = "c"}
-   !CHECK: %[[BOUNDS_3:.*]] = omp.map.bounds   lower_bound({{.*}}) upper_bound({{.*}}) extent({{.*}}) stride({{.*}}) start_idx({{.*}})
-   !CHECK: %[[MAP_3:.*]] = omp.map.info var_ptr({{.*}})   map_clauses(always, delete) capture(ByRef) bounds(%[[BOUNDS_3]]) -> !fir.ref<!fir.array<1024xi32>> {name = "d"}
-   !CHECK: %[[BOUNDS_4:.*]] = omp.map.bounds   lower_bound({{.*}}) upper_bound({{.*}}) extent({{.*}}) stride({{.*}}) start_idx({{.*}})
-   !CHECK: %[[MAP_4:.*]] = omp.map.info var_ptr({{.*}})   map_clauses(from) capture(ByRef) bounds(%[[BOUNDS_4]]) -> !fir.ref<!fir.array<1024xi32>> {name = "e"}
-   !CHECK: omp.target_exit_data map_entries(%[[MAP_0]], %[[MAP_1]], %[[MAP_2]], %[[MAP_3]], %[[MAP_4]] : !fir.ref<!fir.array<1024xi32>>, !fir.ref<!fir.array<1024xi32>>, !fir.ref<!fir.array<1024xi32>>, !fir.ref<!fir.array<1024xi32>>, !fir.ref<!fir.array<1024xi32>>)
-   !$omp target exit data map(from: a,b) map(release: c) map(always, delete: d) map(from: e)
-end subroutine omp_target_exit_mt
-
-!===============================================================================
-! `device` clause
-!===============================================================================
-
-!CHECK-LABEL: func.func @_QPomp_target_exit_device() {
-subroutine omp_target_exit_device
-   integer :: a(1024)
-   integer :: d
-   !CHECK: %[[VAL_2:.*]] = fir.load %[[VAL_1:.*]] : !fir.ref<i32>
-   !CHECK: %[[BOUNDS:.*]] = omp.map.bounds   lower_bound({{.*}}) upper_bound({{.*}}) extent({{.*}}) stride({{.*}}) start_idx({{.*}})
-   !CHECK: %[[MAP:.*]] = omp.map.info var_ptr({{.*}})   map_clauses(from) capture(ByRef) bounds(%[[BOUNDS]]) -> !fir.ref<!fir.array<1024xi32>> {name = "a"}
-   !CHECK: omp.target_exit_data   device(%[[VAL_2]] : i32) map_entries(%[[MAP]] : !fir.ref<!fir.array<1024xi32>>)
-   !$omp target exit data map(from: a) device(d)
-end subroutine omp_target_exit_device
-
-!===============================================================================
-! Target_Update `to` clause
-!===============================================================================
-
-subroutine omp_target_update_to
-   integer :: a(1024)
-
-   !CHECK-DAG: %[[A_ALLOC:.*]] = fir.alloca !fir.array<1024xi32> {bindc_name = "a", uniq_name = "_QFomp_target_update_toEa"}
-   !CHECK-DAG: %[[BOUNDS:.*]] = omp.map.bounds
-
-   !CHECK: %[[TO_MAP:.*]] = omp.map.info var_ptr(%[[A_ALLOC]] : !fir.ref<!fir.array<1024xi32>>, !fir.array<1024xi32>)
-   !CHECK-SAME: map_clauses(to) capture(ByRef)
-   !CHECK-SAME: bounds(%[[BOUNDS]]) -> !fir.ref<!fir.array<1024xi32>> {name = "a"}
-
-   !CHECK: omp.target_update
-   !CHECK-SAME: motion_entries(%[[TO_MAP]] : !fir.ref<!fir.array<1024xi32>>)
-   !$omp target update to(a)
-end subroutine omp_target_update_to
-
-!===============================================================================
-! Target_Update `from` clause
-!===============================================================================
-
-subroutine omp_target_update_from
-   integer :: a(1024)
-
-   !CHECK-DAG: %[[A_ALLOC:.*]] = fir.alloca !fir.array<1024xi32> {bindc_name = "a", uniq_name = "_QFomp_target_update_fromEa"}
-   !CHECK-DAG: %[[BOUNDS:.*]] = omp.map.bounds
-
-   !CHECK: %[[FROM_MAP:.*]] = omp.map.info var_ptr(%[[A_ALLOC]] : !fir.ref<!fir.array<1024xi32>>, !fir.array<1024xi32>)
-   !CHECK-SAME: map_clauses(from) capture(ByRef)
-   !CHECK-SAME: bounds(%[[BOUNDS]]) -> !fir.ref<!fir.array<1024xi32>> {name = "a"}
-
-   !CHECK: omp.target_update
-   !CHECK-SAME: motion_entries(%[[FROM_MAP]] : !fir.ref<!fir.array<1024xi32>>)
-   !$omp target update from(a)
-end subroutine omp_target_update_from
-
-!===============================================================================
-! Target_Update `if` clause
-!===============================================================================
-
-subroutine omp_target_update_if
-   integer :: a(1024)
-   logical :: i
-
-   !CHECK-DAG: %[[A_ALLOC:.*]] = fir.alloca
-   !CHECK-DAG: %[[BOUNDS:.*]] = omp.map.bounds
-   !CHECK-DAG: %[[COND:.*]] = fir.convert %{{.*}} : (!fir.logical<4>) -> i1
-
-   !CHECK: %[[TO_MAP:.*]] = omp.map.info
-
-   !CHECK: omp.target_update if(%[[COND]] : i1)
-   !CHECK-SAME: motion_entries(%[[TO_MAP]] : !fir.ref<!fir.array<1024xi32>>)
-   !$omp target update to(a) if(i)
-end subroutine omp_target_update_if
-
-!===============================================================================
-! Target_Update `device` clause
-!===============================================================================
-
-subroutine omp_target_update_device
-   integer :: a(1024)
-
-   !CHECK-DAG: %[[A_ALLOC:.*]] = fir.alloca
-   !CHECK-DAG: %[[BOUNDS:.*]] = omp.map.bounds
-   !CHECK-DAG: %[[DEVICE:.*]] = arith.constant 1 : i32
-
-   !CHECK: %[[TO_MAP:.*]] = omp.map.info
-
-   !CHECK: omp.target_update
-   !CHECK-SAME: device(%[[DEVICE]] : i32)
-   !CHECK-SAME: motion_entries(%[[TO_MAP]] : !fir.ref<!fir.array<1024xi32>>)
-   !$omp target update to(a) device(1)
-end subroutine omp_target_update_device
-
-!===============================================================================
-! Target_Update `nowait` clause
-!===============================================================================
-
-subroutine omp_target_update_nowait
-   integer :: a(1024)
-
-   !CHECK-DAG: %[[A_ALLOC:.*]] = fir.alloca
-   !CHECK-DAG: %[[BOUNDS:.*]] = omp.map.bounds
-
-   !CHECK: %[[TO_MAP:.*]] = omp.map.info
-
-   !CHECK: omp.target_update
-   !CHECK-SAME: nowait
-   !CHECK-SAME: motion_entries(%[[TO_MAP]] : !fir.ref<!fir.array<1024xi32>>)
-   !$omp target update to(a) nowait
-end subroutine omp_target_update_nowait
-
-!===============================================================================
-! Target_Data with region
-!===============================================================================
-
-!CHECK-LABEL: func.func @_QPomp_target_data() {
-subroutine omp_target_data
-   !CHECK: %[[VAL_0:.*]] = fir.alloca !fir.array<1024xi32> {bindc_name = "a", uniq_name = "_QFomp_target_dataEa"}
-   integer :: a(1024)
-   !CHECK: %[[BOUNDS:.*]] = omp.map.bounds   lower_bound({{.*}}) upper_bound({{.*}}) extent({{.*}}) stride({{.*}}) start_idx({{.*}})
-   !CHECK: %[[MAP:.*]] = omp.map.info var_ptr(%[[VAL_0]] : !fir.ref<!fir.array<1024xi32>>, !fir.array<1024xi32>)   map_clauses(tofrom) capture(ByRef) bounds(%[[BOUNDS]]) -> !fir.ref<!fir.array<1024xi32>> {name = "a"}
-   !CHECK: omp.target_data   map_entries(%[[MAP]] : !fir.ref<!fir.array<1024xi32>>) {
-   !$omp target data map(tofrom: a)
-      !CHECK: %[[VAL_1:.*]] = arith.constant 10 : i32
-      !CHECK: %[[VAL_2:.*]] = arith.constant 1 : i64
-      !CHECK: %[[VAL_3:.*]] = arith.constant 1 : i64
-      !CHECK: %[[VAL_4:.*]] = arith.subi %[[VAL_2]], %[[VAL_3]] : i64
-      !CHECK: %[[VAL_5:.*]] = fir.coordinate_of %[[VAL_0]], %[[VAL_4]] : (!fir.ref<!fir.array<1024xi32>>, i64) -> !fir.ref<i32>
-      !CHECK: fir.store %[[VAL_1]] to %[[VAL_5]] : !fir.ref<i32>
-      a(1) = 10
-   !CHECK: omp.terminator
-   !$omp end target data
-   !CHECK: }
-end subroutine omp_target_data
-
-!CHECK-LABEL: func.func @_QPomp_target_data_mt
-subroutine omp_target_data_mt
-   integer :: a(1024)
-   integer :: b(1024)
-   !CHECK: %[[VAR_A:.*]] = fir.alloca !fir.array<1024xi32> {bindc_name = "a", uniq_name = "_QFomp_target_data_mtEa"}
-   !CHECK: %[[VAR_B:.*]] = fir.alloca !fir.array<1024xi32> {bindc_name = "b", uniq_name = "_QFomp_target_data_mtEb"}
-   !CHECK: %[[BOUNDS_A:.*]] = omp.map.bounds   lower_bound({{.*}}) upper_bound({{.*}}) extent({{.*}}) stride({{.*}}) start_idx({{.*}})
-   !CHECK: %[[MAP_A:.*]] = omp.map.info var_ptr(%[[VAR_A]] : !fir.ref<!fir.array<1024xi32>>, !fir.array<1024xi32>)   map_clauses(tofrom) capture(ByRef) bounds(%[[BOUNDS_A]]) -> !fir.ref<!fir.array<1024xi32>> {name = "a"}
-   !CHECK: omp.target_data   map_entries(%[[MAP_A]] : !fir.ref<!fir.array<1024xi32>>) {
-   !$omp target data map(a)
-   !CHECK: omp.terminator
-   !$omp end target data
-   !CHECK: }
-   !CHECK: %[[BOUNDS_B:.*]] = omp.map.bounds   lower_bound({{.*}}) upper_bound({{.*}}) extent({{.*}}) stride({{.*}}) start_idx({{.*}})
-   !CHECK: %[[MAP_B:.*]] = omp.map.info var_ptr(%[[VAR_B]] : !fir.ref<!fir.array<1024xi32>>, !fir.array<1024xi32>)   map_clauses(always, from) capture(ByRef) bounds(%[[BOUNDS_B]]) -> !fir.ref<!fir.array<1024xi32>> {name = "b"}
-   !CHECK: omp.target_data   map_entries(%[[MAP_B]] : !fir.ref<!fir.array<1024xi32>>) {
-   !$omp target data map(always, from : b)
-   !CHECK: omp.terminator
-   !$omp end target data
-   !CHECK: }
-end subroutine omp_target_data_mt
-
-!===============================================================================
-! Target with region
-!===============================================================================
-
-!CHECK-LABEL: func.func @_QPomp_target() {
-subroutine omp_target
-   !CHECK: %[[VAL_0:.*]] = fir.alloca !fir.array<1024xi32> {bindc_name = "a", uniq_name = "_QFomp_targetEa"}
-   integer :: a(1024)
-   !CHECK: %[[BOUNDS:.*]] = omp.map.bounds   lower_bound({{.*}}) upper_bound({{.*}}) extent({{.*}}) stride({{.*}}) start_idx({{.*}})
-   !CHECK: %[[MAP:.*]] = omp.map.info var_ptr(%[[VAL_0]] : !fir.ref<!fir.array<1024xi32>>, !fir.array<1024xi32>)   map_clauses(tofrom) capture(ByRef) bounds(%[[BOUNDS]]) -> !fir.ref<!fir.array<1024xi32>> {name = "a"}
-   !CHECK: omp.target   map_entries(%[[MAP]] -> %[[ARG_0:.*]] : !fir.ref<!fir.array<1024xi32>>) {
-   !CHECK: ^bb0(%[[ARG_0]]: !fir.ref<!fir.array<1024xi32>>):
-   !$omp target map(tofrom: a)
-      !CHECK: %[[VAL_1:.*]] = arith.constant 10 : i32
-      !CHECK: %[[VAL_2:.*]] = arith.constant 1 : i64
-      !CHECK: %[[VAL_3:.*]] = arith.constant 1 : i64
-      !CHECK: %[[VAL_4:.*]] = arith.subi %[[VAL_2]], %[[VAL_3]] : i64
-      !CHECK: %[[VAL_5:.*]] = fir.coordinate_of %[[ARG_0]], %[[VAL_4]] : (!fir.ref<!fir.array<1024xi32>>, i64) -> !fir.ref<i32>
-      !CHECK: fir.store %[[VAL_1]] to %[[VAL_5]] : !fir.ref<i32>
-      a(1) = 10
-   !CHECK: omp.terminator
-   !$omp end target
-   !CHECK: }
-end subroutine omp_target
-
-!===============================================================================
-! Target implicit capture
-!===============================================================================
-
-!CHECK-LABEL: func.func @_QPomp_target_implicit() {
-subroutine omp_target_implicit
-   !CHECK: %[[VAL_0:.*]] = fir.alloca !fir.array<1024xi32> {bindc_name = "a", uniq_name = "_QFomp_target_implicitEa"}
-   integer :: a(1024)
-   !CHECK: %[[MAP:.*]] = omp.map.info var_ptr(%[[VAL_0]] : !fir.ref<!fir.array<1024xi32>>, !fir.array<1024xi32>)   map_clauses(implicit, tofrom) capture(ByRef) bounds(%{{.*}}) -> !fir.ref<!fir.array<1024xi32>> {name = "a"}
-   !CHECK: omp.target   map_entries(%[[MAP]] -> %[[ARG_0:.*]] : !fir.ref<!fir.array<1024xi32>>) {
-   !CHECK: ^bb0(%[[ARG_0]]: !fir.ref<!fir.array<1024xi32>>):
-   !$omp target
-      !CHECK: %[[VAL_5:.*]] = fir.coordinate_of %[[ARG_0]], %{{.*}} : (!fir.ref<!fir.array<1024xi32>>, i64) -> !fir.ref<i32>
-      a(1) = 10
-   !CHECK: omp.terminator
-   !$omp end target
-   !CHECK: }
-end subroutine omp_target_implicit
-
-!===============================================================================
-! Target implicit capture nested
-!===============================================================================
-
-!CHECK-LABEL: func.func @_QPomp_target_implicit_nested() {
-subroutine omp_target_implicit_nested
-   integer::a, b
-   !CHECK: omp.target   map_entries(%{{.*}} -> %[[ARG0:.*]], %{{.*}} -> %[[ARG1:.*]] : !fir.ref<i32>, !fir.ref<i32>) {
-   !CHECK: ^bb0(%[[ARG0]]: !fir.ref<i32>, %[[ARG1]]: !fir.ref<i32>):
-   !$omp target
-      !CHECK: fir.store %{{.*}} to %[[ARG0]] : !fir.ref<i32>
-      a = 10
-      !$omp parallel
-         !CHECK: fir.store %{{.*}} to %[[ARG1]] : !fir.ref<i32>
-         b = 20
-         !CHECK: omp.terminator
-      !$omp end parallel
-   !CHECK: omp.terminator
-   !$omp end target
-   !CHECK: }
-end subroutine omp_target_implicit_nested
-
-!===============================================================================
-! Target implicit capture with bounds
-!===============================================================================
-
-
-!CHECK-LABEL: func.func @_QPomp_target_implicit_bounds(
-!CHECK: %[[VAL_0:.*]]: !fir.ref<i32> {fir.bindc_name = "n"}) {
-subroutine omp_target_implicit_bounds(n)
-   !CHECK: %[[VAL_COPY:.*]] = fir.alloca i32
-   !CHECK: %[[VAL_1:.*]] = fir.load %[[VAL_0]] : !fir.ref<i32>
-   !CHECK: fir.store %[[VAL_1]] to %[[VAL_COPY]] : !fir.ref<i32>
-   !CHECK: %[[VAL_2:.*]] = fir.convert %[[VAL_1]] : (i32) -> i64
-   !CHECK: %[[VAL_3:.*]] = fir.convert %[[VAL_2]] : (i64) -> index
-   !CHECK: %[[VAL_4:.*]] = arith.constant 0 : index
-   !CHECK: %[[VAL_5:.*]] = arith.cmpi sgt, %[[VAL_3]], %[[VAL_4]] : index
-   !CHECK: %[[VAL_6:.*]] = arith.select %[[VAL_5]], %[[VAL_3]], %[[VAL_4]] : index
-   !CHECK: %[[VAL_7:.*]] = arith.constant 1024 : i64
-   !CHECK: %[[VAL_8:.*]] = fir.convert %[[VAL_7]] : (i64) -> index
-   !CHECK: %[[VAL_9:.*]] = arith.constant 0 : index
-   !CHECK: %[[VAL_10:.*]] = arith.cmpi sgt, %[[VAL_8]], %[[VAL_9]] : index
-   !CHECK: %[[VAL_11:.*]] = arith.select %[[VAL_10]], %[[VAL_8]], %[[VAL_9]] : index
-   !CHECK: %[[VAL_12:.*]] = fir.alloca !fir.array<?x1024xi32>, %[[VAL_6]] {bindc_name = "a", uniq_name = "_QFomp_target_implicit_boundsEa"}
-   integer :: n
-   integer :: a(n, 1024)
-   !CHECK: %[[VAL_13:.*]] = arith.constant 1 : index
-   !CHECK: %[[VAL_14:.*]] = arith.constant 0 : index
-   !CHECK: %[[VAL_15:.*]] = arith.subi %[[VAL_6]], %[[VAL_13]] : index
-   !CHECK: %[[VAL_16:.*]] = omp.map.bounds lower_bound(%[[VAL_14]] : index) upper_bound(%[[VAL_15]] : index) extent(%[[VAL_6]] : index) stride(%[[VAL_13]] : index) start_idx(%[[VAL_13]] : index)
-   !CHECK: %[[VAL_17:.*]] = arith.constant 0 : index
-   !CHECK: %[[VAL_18:.*]] = arith.subi %[[VAL_11]], %[[VAL_13]] : index
-   !CHECK: %[[VAL_19:.*]] = omp.map.bounds lower_bound(%[[VAL_17]] : index) upper_bound(%[[VAL_18]] : index) extent(%[[VAL_11]] : index) stride(%[[VAL_13]] : index) start_idx(%[[VAL_13]] : index)
-   !CHECK: %[[VAL_20:.*]] = omp.map.info var_ptr(%[[VAL_12]] : !fir.ref<!fir.array<?x1024xi32>>, !fir.array<?x1024xi32>) map_clauses(implicit, tofrom) capture(ByRef) bounds(%[[VAL_16]], %[[VAL_19]]) -> !fir.ref<!fir.array<?x1024xi32>> {name = "a"}
-   !CHECK: %[[VAL_21:.*]] = omp.map.info var_ptr(%[[VAL_COPY]] : !fir.ref<i32>, i32) map_clauses(implicit, exit_release_or_enter_alloc) capture(ByCopy) -> !fir.ref<i32> {name = ""}
-   !CHECK: omp.target map_entries(%[[VAL_20]] -> %[[VAL_22:.*]], %[[VAL_21]] -> %[[VAL_23:.*]] : !fir.ref<!fir.array<?x1024xi32>>, !fir.ref<i32>) {
-   !CHECK: ^bb0(%[[VAL_22]]: !fir.ref<!fir.array<?x1024xi32>>, %[[VAL_23]]: !fir.ref<i32>):
-   !$omp target
-      !CHECK: %[[VAL_24:.*]] = fir.load %[[VAL_23]] : !fir.ref<i32>
-      !CHECK: %[[VAL_25:.*]] = fir.convert %[[VAL_24]] : (i32) -> i64
-      !CHECK: %[[VAL_26:.*]] = arith.constant 0 : index
-      !CHECK: %[[VAL_27:.*]] = fir.convert %[[VAL_25]] : (i64) -> index
-      !CHECK: %[[VAL_28:.*]] = arith.cmpi sgt, %[[VAL_27]], %[[VAL_26]] : index
-      !CHECK: %[[VAL_29:.*]] = arith.select %[[VAL_28]], %[[VAL_27]], %[[VAL_26]] : index
-      !CHECK: %[[VAL_30:.*]] = arith.constant 33 : i32
-      !CHECK: %[[VAL_31:.*]] = fir.convert %[[VAL_22]] : (!fir.ref<!fir.array<?x1024xi32>>) -> !fir.ref<!fir.array<?xi32>>
-      !CHECK: %[[VAL_32:.*]] = arith.constant 1 : index
-      !CHECK: %[[VAL_33:.*]] = arith.constant 0 : index
-      !CHECK: %[[VAL_34:.*]] = arith.constant 11 : i64
-      !CHECK: %[[VAL_35:.*]] = fir.convert %[[VAL_34]] : (i64) -> index
-      !CHECK: %[[VAL_36:.*]] = arith.subi %[[VAL_35]], %[[VAL_32]] : index
-      !CHECK: %[[VAL_37:.*]] = arith.muli %[[VAL_32]], %[[VAL_36]] : index
-      !CHECK: %[[VAL_38:.*]] = arith.addi %[[VAL_37]], %[[VAL_33]] : index
-      !CHECK: %[[VAL_39:.*]] = arith.muli %[[VAL_32]], %[[VAL_29]] : index
-      !CHECK: %[[VAL_40:.*]] = arith.constant 22 : i64
-      !CHECK: %[[VAL_41:.*]] = fir.convert %[[VAL_40]] : (i64) -> index
-      !CHECK: %[[VAL_42:.*]] = arith.subi %[[VAL_41]], %[[VAL_32]] : index
-      !CHECK: %[[VAL_43:.*]] = arith.muli %[[VAL_39]], %[[VAL_42]] : index
-      !CHECK: %[[VAL_44:.*]] = arith.addi %[[VAL_43]], %[[VAL_38]] : index
-      !CHECK: %[[VAL_45:.*]] = fir.coordinate_of %[[VAL_31]], %[[VAL_44]] : (!fir.ref<!fir.array<?xi32>>, index) -> !fir.ref<i32>
-      !CHECK: fir.store %[[VAL_30]] to %[[VAL_45]] : !fir.ref<i32>
-      a(11, 22) = 33
-      !CHECK: omp.terminator
-   !$omp end target
-!CHECK: }
-end subroutine omp_target_implicit_bounds
-
-!===============================================================================
-! Target `thread_limit` clause
-!===============================================================================
-
-!CHECK-LABEL: func.func @_QPomp_target_thread_limit() {
-subroutine omp_target_thread_limit
-   integer :: a
-   !CHECK: %[[MAP:.*]] = omp.map.info var_ptr({{.*}})   map_clauses(tofrom) capture(ByRef) -> !fir.ref<i32> {name = "a"}
-   !CHECK: %[[VAL_1:.*]] = arith.constant 64 : i32
-   !CHECK: omp.target   thread_limit(%[[VAL_1]] : i32) map_entries(%[[MAP]] -> %[[ARG_0:.*]] : !fir.ref<i32>) {
-   !CHECK: ^bb0(%[[ARG_0]]: !fir.ref<i32>):
-   !$omp target map(tofrom: a) thread_limit(64)
-      a = 10
-   !CHECK: omp.terminator
-   !$omp end target
-   !CHECK: }
-end subroutine omp_target_thread_limit
-
-!===============================================================================
-! Target `use_device_ptr` clause
-!===============================================================================
-
-!CHECK-LABEL: func.func @_QPomp_target_device_ptr() {
-subroutine omp_target_device_ptr
-   use iso_c_binding, only : c_ptr, c_loc
-   type(c_ptr) :: a
-   integer, target :: b
-   !CHECK: %[[MAP:.*]] = omp.map.info var_ptr({{.*}})   map_clauses(tofrom) capture(ByRef) -> {{.*}} {name = "a"}
-   !CHECK: omp.target_data map_entries(%[[MAP]]{{.*}}
-   !$omp target data map(tofrom: a) use_device_ptr(a)
-   !CHECK: ^bb0(%[[VAL_1:.*]]: !fir.ref<!fir.type<_QM__fortran_builtinsT__builtin_c_ptr{__address:i64}>>):
-   !CHECK: {{.*}} = fir.coordinate_of %[[VAL_1:.*]], {{.*}} : (!fir.ref<!fir.type<_QM__fortran_builtinsT__builtin_c_ptr{__address:i64}>>, !fir.field) -> !fir.ref<i64>
-      a = c_loc(b)
-   !CHECK: omp.terminator
-   !$omp end target data
-   !CHECK: }
-end subroutine omp_target_device_ptr
-
- !===============================================================================
- ! Target `use_device_addr` clause
- !===============================================================================
-
- !CHECK-LABEL: func.func @_QPomp_target_device_addr() {
- subroutine omp_target_device_addr
-   integer, pointer :: a
-   !CHECK: %[[VAL_0:.*]] = fir.alloca !fir.box<!fir.ptr<i32>> {bindc_name = "a", uniq_name = "_QFomp_target_device_addrEa"}
-   !CHECK: %[[MAP_MEMBERS:.*]] = omp.map.info var_ptr({{.*}} : !fir.ref<!fir.box<!fir.ptr<i32>>>, i32) var_ptr_ptr({{.*}} : !fir.llvm_ptr<!fir.ref<i32>>) map_clauses(tofrom) capture(ByRef) -> !fir.llvm_ptr<!fir.ref<i32>> {name = ""}
-   !CHECK: %[[MAP:.*]] = omp.map.info var_ptr({{.*}} : !fir.ref<!fir.box<!fir.ptr<i32>>>, !fir.box<!fir.ptr<i32>>) map_clauses(tofrom) capture(ByRef) members(%[[MAP_MEMBERS]] : !fir.llvm_ptr<!fir.ref<i32>>) -> !fir.ref<!fir.box<!fir.ptr<i32>>> {name = "a"}
-   !CHECK: omp.target_data map_entries(%[[MAP_MEMBERS]], %[[MAP]] : {{.*}}) use_device_addr(%[[VAL_0]] : !fir.ref<!fir.box<!fir.ptr<i32>>>) {
-   !$omp target data map(tofrom: a) use_device_addr(a)
-   !CHECK: ^bb0(%[[VAL_1:.*]]: !fir.ref<!fir.box<!fir.ptr<i32>>>):
-   !CHECK: {{.*}} = fir.load %[[VAL_1]] : !fir.ref<!fir.box<!fir.ptr<i32>>>
-      a = 10
-   !CHECK: omp.terminator
-   !$omp end target data
-   !CHECK: }
-end subroutine omp_target_device_addr
-
-!===============================================================================
-! Target with parallel loop
-!===============================================================================
-
-!CHECK-LABEL: func.func @_QPomp_target_parallel_do() {
-subroutine omp_target_parallel_do
-   !CHECK: %[[C1024:.*]] = arith.constant 1024 : index
-   !CHECK: %[[VAL_0:.*]] = fir.alloca !fir.array<1024xi32> {bindc_name = "a", uniq_name = "_QFomp_target_parallel_doEa"}
-   integer :: a(1024)
-   !CHECK: %[[VAL_1:.*]] = fir.alloca i32 {bindc_name = "i", uniq_name = "_QFomp_target_parallel_doEi"}
-   integer :: i
-   !CHECK: %[[C1:.*]] = arith.constant 1 : index
-   !CHECK: %[[C0:.*]] = arith.constant 0 : index
-   !CHECK: %[[SUB:.*]] = arith.subi %[[C1024]], %[[C1]] : index
-   !CHECK: %[[BOUNDS:.*]] = omp.map.bounds   lower_bound(%[[C0]] : index) upper_bound(%[[SUB]] : index) extent(%[[C1024]] : index) stride(%[[C1]] : index) start_idx(%[[C1]] : index)
-   !CHECK: %[[MAP1:.*]] = omp.map.info var_ptr(%[[VAL_0]] : !fir.ref<!fir.array<1024xi32>>, !fir.array<1024xi32>)   map_clauses(tofrom) capture(ByRef) bounds(%[[BOUNDS]]) -> !fir.ref<!fir.array<1024xi32>> {name = "a"}
-   !CHECK: %[[MAP2:.*]] = omp.map.info var_ptr(%[[VAL_1]] : !fir.ref<i32>, i32)   map_clauses(implicit, exit_release_or_enter_alloc) capture(ByCopy) -> !fir.ref<i32> {name = "i"}
-   !CHECK: omp.target map_entries(%[[MAP1]] -> %[[VAL_2:.*]], %[[MAP2]] -> %[[VAL_3:.*]] : !fir.ref<!fir.array<1024xi32>>, !fir.ref<i32>) {
-   !CHECK: ^bb0(%[[VAL_2]]: !fir.ref<!fir.array<1024xi32>>, %[[VAL_3]]: !fir.ref<i32>):
-      !CHECK-NEXT: omp.parallel
-      !$omp target parallel do map(tofrom: a)
-         !CHECK: %[[VAL_4:.*]] = fir.alloca i32 {adapt.valuebyref, pinned}
-         !CHECK: %[[VAL_5:.*]] = arith.constant 1 : i32
-         !CHECK: %[[VAL_6:.*]] = arith.constant 1024 : i32
-         !CHECK: %[[VAL_7:.*]] = arith.constant 1 : i32
-         !CHECK: omp.wsloop   for  (%[[VAL_8:.*]]) : i32 = (%[[VAL_5]]) to (%[[VAL_6]]) inclusive step (%[[VAL_7]]) {
-         !CHECK: fir.store %[[VAL_8]] to %[[VAL_4]] : !fir.ref<i32>
-         !CHECK: %[[VAL_9:.*]] = arith.constant 10 : i32
-         !CHECK: %[[VAL_10:.*]] = fir.load %[[VAL_4]] : !fir.ref<i32>
-         !CHECK: %[[VAL_11:.*]] = fir.convert %[[VAL_10]] : (i32) -> i64
-         !CHECK: %[[VAL_12:.*]] = arith.constant 1 : i64
-         !CHECK: %[[VAL_13:.*]] = arith.subi %[[VAL_11]], %[[VAL_12]] : i64
-         !CHECK: %[[VAL_14:.*]] = fir.coordinate_of %[[VAL_2]], %[[VAL_13]] : (!fir.ref<!fir.array<1024xi32>>, i64) -> !fir.ref<i32>
-         !CHECK: fir.store %[[VAL_9]] to %[[VAL_14]] : !fir.ref<i32>
-         do i = 1, 1024
-            a(i) = 10
-         end do
-         !CHECK: omp.yield
-         !CHECK: }
-      !CHECK: omp.terminator
-      !CHECK: }
-   !CHECK: omp.terminator
-   !CHECK: }
-   !$omp end target parallel do
-end subroutine omp_target_parallel_do
-
-!===============================================================================
-! Target `is_device_ptr` clause
-!===============================================================================
-
-!CHECK-LABEL: func.func @_QPomp_target_is_device_ptr() {
-subroutine omp_target_is_device_ptr
-   use iso_c_binding, only : c_ptr, c_loc
-   !CHECK: %[[VAL_0:.*]] = fir.alloca !fir.type<_QM__fortran_builtinsT__builtin_c_ptr{__address:i64}> {bindc_name = "a", uniq_name = "_QFomp_target_is_device_ptrEa"}
-   type(c_ptr) :: a
-   !CHECK: %[[VAL_1:.*]] = fir.alloca i32 {bindc_name = "b", fir.target, uniq_name = "_QFomp_target_is_device_ptrEb"}
-   integer, target :: b
-   !CHECK: %[[MAP_0:.*]] = omp.map.info var_ptr(%[[DEV_PTR:.*]] : !fir.ref<!fir.type<_QM__fortran_builtinsT__builtin_c_ptr{__address:i64}>>, !fir.type<_QM__fortran_builtinsT__builtin_c_ptr{__address:i64}>) map_clauses(tofrom) capture(ByRef) -> !fir.ref<!fir.type<_QM__fortran_builtinsT__builtin_c_ptr{__address:i64}>> {name = "a"}
-   !CHECK: %[[MAP_1:.*]] = omp.map.info var_ptr(%[[VAL_0:.*]] : !fir.ref<i32>, i32)   map_clauses(tofrom) capture(ByRef) -> !fir.ref<i32> {name = "b"}
-   !CHECK: %[[MAP_2:.*]] = omp.map.info var_ptr(%[[DEV_PTR:.*]] : !fir.ref<!fir.type<_QM__fortran_builtinsT__builtin_c_ptr{__address:i64}>>, !fir.type<_QM__fortran_builtinsT__builtin_c_ptr{__address:i64}>) map_clauses(implicit, exit_release_or_enter_alloc) capture(ByRef) -> !fir.ref<!fir.type<_QM__fortran_builtinsT__builtin_c_ptr{__address:i64}>> {name = "a"}
-   !CHECK: omp.target is_device_ptr(%[[DEV_PTR:.*]] : !fir.ref<!fir.type<_QM__fortran_builtinsT__builtin_c_ptr{__address:i64}>>) map_entries(%[[MAP_0:.*]] -> %[[ARG0:.*]], %[[MAP_1:.*]] -> %[[ARG1:.*]], %[[MAP_2:.*]] -> %[[ARG2:.*]] : !fir.ref<!fir.type<_QM__fortran_builtinsT__builtin_c_ptr{__address:i64}>>, !fir.ref<i32>, !fir.ref<!fir.type<_QM__fortran_builtinsT__builtin_c_ptr{__address:i64}>>) {
-   !CHECK: ^bb0(%[[ARG0]]: !fir.ref<!fir.type<_QM__fortran_builtinsT__builtin_c_ptr{__address:i64}>>, %[[ARG1]]: !fir.ref<i32>, %[[ARG2]]: !fir.ref<!fir.type<_QM__fortran_builtinsT__builtin_c_ptr{__address:i64}>>):
-   !$omp target map(tofrom: a,b) is_device_ptr(a)
-      !CHECK: {{.*}} = fir.coordinate_of %[[VAL_0:.*]], {{.*}} : (!fir.ref<!fir.type<_QM__fortran_builtinsT__builtin_c_ptr{__address:i64}>>, !fir.field) -> !fir.ref<i64>
-      a = c_loc(b)
-   !CHECK: omp.terminator
-   !$omp end target
-   !CHECK: }
-end subroutine omp_target_is_device_ptr
-
- !===============================================================================
- ! Target `has_device_addr` clause
- !===============================================================================
-
- !CHECK-LABEL: func.func @_QPomp_target_has_device_addr() {
- subroutine omp_target_has_device_addr
-   !CHECK: %[[VAL_0:.*]] = fir.alloca !fir.box<!fir.ptr<i32>> {bindc_name = "a", uniq_name = "_QFomp_target_has_device_addrEa"}
-   integer, pointer :: a
-   !CHECK: omp.target has_device_addr(%[[VAL_0:.*]] : !fir.ref<!fir.box<!fir.ptr<i32>>>) map_entries({{.*}} -> {{.*}}, {{.*}} -> {{.*}} : !fir.llvm_ptr<!fir.ref<i32>>, !fir.ref<!fir.box<!fir.ptr<i32>>>) {
-   !$omp target has_device_addr(a)
-   !CHECK: {{.*}} = fir.load %[[VAL_0:.*]] : !fir.ref<!fir.box<!fir.ptr<i32>>>
-      a = 10
-   !CHECK: omp.terminator
-   !$omp end target
-   !CHECK: }
-end subroutine omp_target_has_device_addr
diff --git a/flang/test/Lower/OpenMP/FIR/target_cpu_features.f90 b/flang/test/Lower/OpenMP/FIR/target_cpu_features.f90
deleted file mode 100644
index 5154782e1ae1..000000000000
--- a/flang/test/Lower/OpenMP/FIR/target_cpu_features.f90
+++ /dev/null
@@ -1,19 +0,0 @@
-!REQUIRES: amdgpu-registered-target, nvptx-registered-target
-!RUN: %flang_fc1 -emit-fir -triple amdgcn-amd-amdhsa -target-cpu gfx908 -fopenmp -fopenmp-is-target-device %s -o - | FileCheck --check-prefix=AMDGCN %s
-!RUN: %flang_fc1 -emit-fir -triple nvptx64-nvidia-cuda -target-cpu sm_80 -fopenmp -fopenmp-is-target-device %s -o - | FileCheck --check-prefix=NVPTX %s
-
-!===============================================================================
-! Target_Enter Simple
-!===============================================================================
-
-!AMDGCN: module attributes {
-!AMDGCN-SAME: fir.target_cpu = "gfx908"
-!AMDGCN-SAME: fir.target_features = #llvm.target_features<["+16-bit-insts", "+ci-insts",
-!AMDGCN-SAME: "+dl-insts", "+dot1-insts", "+dot10-insts", "+dot2-insts", "+dot3-insts",
-!AMDGCN-SAME: "+dot4-insts", "+dot5-insts", "+dot6-insts", "+dot7-insts", "+dpp",
-!AMDGCN-SAME: "+gfx8-insts", "+gfx9-insts", "+gws", "+image-insts", "+mai-insts",
-!AMDGCN-SAME: "+s-memrealtime", "+s-memtime-inst", "+wavefrontsize64"]>
-
-!NVPTX: module attributes {
-!NVPTX-SAME: fir.target_cpu = "sm_80"
-!NVPTX-SAME: fir.target_features = #llvm.target_features<["+ptx61", "+sm_80"]>
diff --git a/flang/test/Lower/OpenMP/FIR/task.f90 b/flang/test/Lower/OpenMP/FIR/task.f90
deleted file mode 100644
index 012ac757d304..000000000000
--- a/flang/test/Lower/OpenMP/FIR/task.f90
+++ /dev/null
@@ -1,237 +0,0 @@
-! REQUIRES: openmp_runtime
-
-!RUN: %flang_fc1 -emit-fir -flang-deprecated-no-hlfir -fopenmp %s -o - | FileCheck %s
-
-!CHECK-LABEL: func @_QPomp_task_simple() {
-subroutine omp_task_simple
-  !CHECK: omp.task {
-  !$omp task
-  !CHECK: fir.call @_QPfoo() {{.*}}: () -> ()
-  call foo()
-  !CHECK: omp.terminator
-  !$omp end task
-end subroutine omp_task_simple
-
-!===============================================================================
-! `if` clause
-!===============================================================================
-
-!CHECK-LABEL: func @_QPomp_task_if(%{{.+}}) {
-subroutine omp_task_if(bar)
-  logical, intent(inout) :: bar
-  !CHECK: omp.task if(%{{.+}}) {
-  !$omp task if(bar)
-  !CHECK: fir.call @_QPfoo() {{.*}}: () -> ()
-  call foo()
-  !CHECK: omp.terminator
-  !$omp end task
-end subroutine omp_task_if
-
-!===============================================================================
-! `final` clause
-!===============================================================================
-
-!CHECK-LABEL: func @_QPomp_task_final(%{{.+}}) {
-subroutine omp_task_final(bar)
-  logical, intent(inout) :: bar
-  !CHECK: omp.task final(%{{.+}}) {
-  !$omp task final(bar)
-  !CHECK: fir.call @_QPfoo() {{.*}}: () -> ()
-  call foo()
-  !CHECK: omp.terminator
-  !$omp end task
-end subroutine omp_task_final
-
-!===============================================================================
-! `priority` clause
-!===============================================================================
-
-!CHECK-LABEL: func @_QPomp_task_priority(%{{.+}}) {
-subroutine omp_task_priority(bar)
-  integer, intent(inout) :: bar
-  !CHECK: omp.task priority(%{{.+}}) {
-  !$omp task priority(bar)
-  !CHECK: fir.call @_QPfoo() {{.*}}: () -> ()
-  call foo()
-  !CHECK: omp.terminator
-  !$omp end task
-end subroutine omp_task_priority
-
-!===============================================================================
-! `allocate` clause
-!===============================================================================
-
-!CHECK-LABEL: func @_QPtask_allocate
-subroutine task_allocate()
-  use omp_lib
-  integer :: x
-  !CHECK: omp.task allocate(%{{.+}} : i64 -> %{{.+}} : !fir.ref<i32>) {
-  !$omp task allocate(omp_high_bw_mem_alloc: x) private(x)
-  !CHECK: arith.addi
-  x = x + 12
-  !CHECK: omp.terminator
-  !$omp end task
-end subroutine task_allocate
-
-!===============================================================================
-! `depend` clause
-!===============================================================================
-
-!CHECK-LABEL: func @_QPtask_depend
-subroutine task_depend()
-  integer :: x
-  !CHECK: omp.task depend(taskdependin -> %{{.+}} : !fir.ref<i32>) {
-  !$omp task depend(in : x)
-  !CHECK: arith.addi
-  x = x + 12
-  !CHECK: omp.terminator
-  !$omp end task
-end subroutine task_depend
-
-!CHECK-LABEL: func @_QPtask_depend_non_int
-subroutine task_depend_non_int()
-  character(len = 15) :: x
-  integer, allocatable :: y
-  complex :: z
-  !CHECK: omp.task depend(taskdependin -> %{{.+}} : !fir.ref<!fir.char<1,15>>, taskdependin -> %{{.+}} : !fir.ref<!fir.box<!fir.heap<i32>>>, taskdependin ->  %{{.+}} : !fir.ref<!fir.complex<4>>) {
-  !$omp task depend(in : x, y, z)
-  !CHECK: omp.terminator
-  !$omp end task
-end subroutine task_depend_non_int
-
-!CHECK-LABEL: func @_QPtask_depend_all_kinds_one_task
-subroutine task_depend_all_kinds_one_task()
-  integer :: x
-  !CHECK: omp.task depend(taskdependin -> %{{.+}} : !fir.ref<i32>, taskdependout -> %{{.+}} : !fir.ref<i32>, taskdependinout -> %{{.+}} : !fir.ref<i32>) {
-  !$omp task depend(in : x) depend(out : x) depend(inout : x)
-  !CHECK: arith.addi
-  x = x + 12
-  !CHECK: omp.terminator
-  !$omp end task
-end subroutine task_depend_all_kinds_one_task
-
-!CHECK-LABEL: func @_QPtask_depend_multi_var
-subroutine task_depend_multi_var()
-  integer :: x
-  integer :: y
-  !CHECK: omp.task depend(taskdependin -> %{{.*}} : !fir.ref<i32>, taskdependin -> %{{.+}} : !fir.ref<i32>) {
-  !$omp task depend(in :x,y)
-  !CHECK: arith.addi
-  x = x + 12
-  y = y + 12
-  !CHECK: omp.terminator
-  !$omp end task
-end subroutine task_depend_multi_var
-
-!CHECK-LABEL: func @_QPtask_depend_multi_task
-subroutine task_depend_multi_task()
-  integer :: x
-  !CHECK: omp.task depend(taskdependout -> %{{.+}} : !fir.ref<i32>)
-  !$omp task depend(out : x)
-  !CHECK: arith.addi
-  x = x + 12
-  !CHECK: omp.terminator
-  !$omp end task
-  !CHECK: omp.task depend(taskdependinout -> %{{.+}} : !fir.ref<i32>)
-  !$omp task depend(inout : x)
-  !CHECK: arith.addi
-  x = x + 12
-  !CHECK: omp.terminator
-  !$omp end task
-  !CHECK: omp.task depend(taskdependin -> %{{.+}} : !fir.ref<i32>)
-  !$omp task depend(in : x)
-  !CHECK: arith.addi
-  x = x + 12
-  !CHECK: omp.terminator
-  !$omp end task
-end subroutine task_depend_multi_task
-
-!===============================================================================
-! `private` clause
-!===============================================================================
-!CHECK-LABEL: func @_QPtask_private
-subroutine task_private
-  type mytype
-  integer :: x
-  end type mytype
-
-  !CHECK: %[[int_var:.+]] = fir.alloca i32
-  !CHECK: %[[mytype_var:.+]] = fir.alloca !fir.type<_QFtask_privateTmytype{x:i32}>
-  integer :: int_var
-  type(mytype) :: mytype_var
-
-  !CHECK: fir.call @_QPbar(%[[int_var]], %[[mytype_var]]) {{.*}}: (!fir.ref<i32>, !fir.ref<!fir.type<_QFtask_privateTmytype{x:i32}>>) -> ()
-  call bar(int_var, mytype_var)
-
-  !CHECK: omp.task {
-  !$omp task private(int_var, mytype_var)
-  !CHECK: %[[int_var_private:.+]] = fir.alloca i32
-  !CHECK: %[[mytype_var_private:.+]] = fir.alloca !fir.type<_QFtask_privateTmytype{x:i32}>
-
-  !CHECK: fir.call @_QPbar(%[[int_var_private]], %[[mytype_var_private]]) {{.*}}: (!fir.ref<i32>, !fir.ref<!fir.type<_QFtask_privateTmytype{x:i32}>>) -> ()
-  call bar(int_var, mytype_var)
-  !CHECK: omp.terminator
-  !$omp end task
-end subroutine task_private
-
-!===============================================================================
-! `firstprivate` clause
-!===============================================================================
-!CHECK-LABEL: func @_QPtask_firstprivate
-subroutine task_firstprivate
-  type mytype
-  integer :: x
-  end type mytype
-
-  !CHECK: %[[int_var:.+]] = fir.alloca i32
-  !CHECK: %[[mytype_var:.+]] = fir.alloca !fir.type<_QFtask_firstprivateTmytype{x:i32}>
-  integer :: int_var
-  type(mytype) :: mytype_var
-
-  !CHECK: fir.call @_QPbaz(%[[int_var]], %[[mytype_var]]) {{.*}}: (!fir.ref<i32>, !fir.ref<!fir.type<_QFtask_firstprivateTmytype{x:i32}>>) -> ()
-  call baz(int_var, mytype_var)
-
-  !CHECK: omp.task {
-  !$omp task firstprivate(int_var, mytype_var)
-  !CHECK: %[[int_var_firstprivate:.+]] = fir.alloca i32
-  !CHECK: %[[int_var_load:.+]] = fir.load %[[int_var]] : !fir.ref<i32>
-  !CHECK: fir.store %[[int_var_load]] to %[[int_var_firstprivate]] : !fir.ref<i32>
-  !CHECK: %[[mytype_var_firstprivate:.+]] = fir.alloca !fir.type<_QFtask_firstprivateTmytype{x:i32}>
-  !CHECK: %[[mytype_var_load:.+]] = fir.load %[[mytype_var]] : !fir.ref<!fir.type<_QFtask_firstprivateTmytype{x:i32}>>
-  !CHECK: fir.store %[[mytype_var_load]] to %[[mytype_var_firstprivate]]
-  !CHECK: fir.call @_QPbaz(%[[int_var_firstprivate]], %[[mytype_var_firstprivate]]) {{.*}}: (!fir.ref<i32>, !fir.ref<!fir.type<_QFtask_firstprivateTmytype{x:i32}>>) -> ()
-  call baz(int_var, mytype_var)
-  !CHECK: omp.terminator
-  !$omp end task
-end subroutine task_firstprivate
-
-!===============================================================================
-! Multiple clauses
-!===============================================================================
-
-!CHECK-LABEL: func @_QPtask_multiple_clauses
-subroutine task_multiple_clauses()
-  use omp_lib
-
-  !CHECK: %[[x:.+]] = fir.alloca i32 {bindc_name = "x", uniq_name = "_QFtask_multiple_clausesEx"}
-  !CHECK: %[[y:.+]] = fir.alloca i32 {bindc_name = "y", uniq_name = "_QFtask_multiple_clausesEy"}
-  !CHECK: %[[z:.+]] = fir.alloca i32 {bindc_name = "z", uniq_name = "_QFtask_multiple_clausesEz"}
-  integer :: x, y, z
-  logical :: buzz
-
-  !CHECK: omp.task if(%{{.+}}) final(%{{.+}}) priority(%{{.+}}) allocate(%{{.+}} : i64 -> %{{.+}} : !fir.ref<i32>) {
-  !$omp task if(buzz) final(buzz) priority(z) allocate(omp_high_bw_mem_alloc: x) private(x) firstprivate(y)
-
-  !CHECK: %[[x_priv:.+]] = fir.alloca i32
-  !CHECK: %[[y_priv:.+]] = fir.alloca i32
-  !CHECK: %[[y_load:.+]] = fir.load %[[y]] : !fir.ref<i32>
-  !CHECK: fir.store %[[y_load]] to %[[y_priv]] : !fir.ref<i32>
-
-  !CHECK: arith.addi
-  x = x + 12
-  !CHECK: arith.subi
-  y = y - 12
-
-  !CHECK: omp.terminator
-  !$omp end task
-end subroutine task_multiple_clauses
diff --git a/flang/test/Lower/OpenMP/FIR/taskgroup.f90 b/flang/test/Lower/OpenMP/FIR/taskgroup.f90
deleted file mode 100644
index 78b9da8e9b09..000000000000
--- a/flang/test/Lower/OpenMP/FIR/taskgroup.f90
+++ /dev/null
@@ -1,21 +0,0 @@
-! REQUIRES: openmp_runtime
-
-!RUN: %flang_fc1 -emit-fir -flang-deprecated-no-hlfir -fopenmp %s -o - | FileCheck %s
-
-!CHECK-LABEL: @_QPomp_taskgroup
-subroutine omp_taskgroup
-use omp_lib
-integer :: allocated_x
-!CHECK-DAG: %{{.*}} = fir.alloca i32 {bindc_name = "allocated_x", uniq_name = "_QFomp_taskgroupEallocated_x"}
-!CHECK-DAG: %{{.*}} = arith.constant 4 : i64
-
-!CHECK: omp.taskgroup  allocate(%{{.*}} : i64 -> %0 : !fir.ref<i32>)
-!$omp taskgroup allocate(omp_high_bw_mem_alloc: allocated_x)
-!$omp task
-!CHECK: fir.call @_QPwork() {{.*}}: () -> ()
-   call work()
-!CHECK: omp.terminator
-!$omp end task
-!CHECK: omp.terminator
-!$omp end taskgroup
-end subroutine
diff --git a/flang/test/Lower/OpenMP/FIR/taskwait.f90 b/flang/test/Lower/OpenMP/FIR/taskwait.f90
deleted file mode 100644
index eed4f1b84a22..000000000000
--- a/flang/test/Lower/OpenMP/FIR/taskwait.f90
+++ /dev/null
@@ -1,12 +0,0 @@
-!RUN: %flang_fc1 -emit-fir -fopenmp %s -o - | FileCheck %s --check-prefixes="FIRDialect,OMPDialect"
-!RUN: %flang_fc1 -emit-fir -fopenmp %s -o - | fir-opt --fir-to-llvm-ir | FileCheck %s --check-prefixes="OMPDialect"
-
-!FIRDialect-LABEL: @_QPomp_taskwait
-subroutine omp_taskwait
-  !OMPDialect: omp.taskwait
-  !$omp taskwait
-  !FIRDialect: fir.call @_QPfoo() {{.*}}: () -> ()
-  call foo()
-  !OMPDialect: omp.taskwait
-  !$omp taskwait
-end subroutine omp_taskwait
diff --git a/flang/test/Lower/OpenMP/FIR/taskyield.f90 b/flang/test/Lower/OpenMP/FIR/taskyield.f90
deleted file mode 100644
index ca0bc1d071df..000000000000
--- a/flang/test/Lower/OpenMP/FIR/taskyield.f90
+++ /dev/null
@@ -1,12 +0,0 @@
-!RUN: %flang_fc1 -emit-fir -fopenmp %s -o - | FileCheck %s --check-prefixes="FIRDialect,OMPDialect"
-!RUN: %flang_fc1 -emit-fir -fopenmp %s -o - | fir-opt --fir-to-llvm-ir | FileCheck %s --check-prefixes="OMPDialect"
-
-!FIRDialect-LABEL: @_QPomp_taskyield
-subroutine omp_taskyield
-  !OMPDialect: omp.taskyield
-  !$omp taskyield
-  !FIRDialect: fir.call @_QPfoo() {{.*}}: () -> ()
-  call foo()
-  !OMPDialect: omp.taskyield
-  !$omp taskyield
-end subroutine omp_taskyield
diff --git a/flang/test/Lower/OpenMP/FIR/teams.f90 b/flang/test/Lower/OpenMP/FIR/teams.f90
deleted file mode 100644
index 9c0593a24f2d..000000000000
--- a/flang/test/Lower/OpenMP/FIR/teams.f90
+++ /dev/null
@@ -1,117 +0,0 @@
-! REQUIRES: openmp_runtime
-
-! RUN: %flang_fc1 -emit-fir -fopenmp %s -o - | FileCheck %s
-
-! CHECK-LABEL: func @_QPteams_simple
-subroutine teams_simple()
-  ! CHECK: omp.teams
-  !$omp teams
-  ! CHECK: fir.call
-  call f1()
-  ! CHECK: omp.terminator
-  !$omp end teams
-end subroutine teams_simple
-
-!===============================================================================
-! `num_teams` clause
-!===============================================================================
-
-! CHECK-LABEL: func @_QPteams_numteams
-subroutine teams_numteams(num_teams)
-  integer, intent(inout) :: num_teams
-
-  ! CHECK: omp.teams
-  ! CHECK-SAME: num_teams( to %{{.*}}: i32)
-  !$omp teams num_teams(4)
-  ! CHECK: fir.call
-  call f1()
-  ! CHECK: omp.terminator
-  !$omp end teams
-
-  ! CHECK: omp.teams
-  ! CHECK-SAME: num_teams( to %{{.*}}: i32)
-  !$omp teams num_teams(num_teams)
-  ! CHECK: fir.call
-  call f2()
-  ! CHECK: omp.terminator
-  !$omp end teams
-
-end subroutine teams_numteams
-
-!===============================================================================
-! `if` clause
-!===============================================================================
-
-! CHECK-LABEL: func @_QPteams_if
-subroutine teams_if(alpha)
-  integer, intent(in) :: alpha
-  logical :: condition
-
-  ! CHECK: omp.teams
-  ! CHECK-SAME: if(%{{.*}})
-  !$omp teams if(.false.)
-  ! CHECK: fir.call
-  call f1()
-  ! CHECK: omp.terminator
-  !$omp end teams
-
-  ! CHECK: omp.teams
-  ! CHECK-SAME: if(%{{.*}})
-  !$omp teams if(alpha .le. 0)
-  ! CHECK: fir.call
-  call f2()
-  ! CHECK: omp.terminator
-  !$omp end teams
-
-  ! CHECK: omp.teams
-  ! CHECK-SAME: if(%{{.*}})
-  !$omp teams if(condition)
-  ! CHECK: fir.call
-  call f3()
-  ! CHECK: omp.terminator
-  !$omp end teams
-end subroutine teams_if
-
-!===============================================================================
-! `thread_limit` clause
-!===============================================================================
-
-! CHECK-LABEL: func @_QPteams_threadlimit
-subroutine teams_threadlimit(thread_limit)
-  integer, intent(inout) :: thread_limit
-
-  ! CHECK: omp.teams
-  ! CHECK-SAME: thread_limit(%{{.*}}: i32)
-  !$omp teams thread_limit(4)
-  ! CHECK: fir.call
-  call f1()
-  ! CHECK: omp.terminator
-  !$omp end teams
-
-  ! CHECK: omp.teams
-  ! CHECK-SAME: thread_limit(%{{.*}}: i32)
-  !$omp teams thread_limit(thread_limit)
-  ! CHECK: fir.call
-  call f2()
-  ! CHECK: omp.terminator
-  !$omp end teams
-
-end subroutine teams_threadlimit
-
-!===============================================================================
-! `allocate` clause
-!===============================================================================
-
-! CHECK-LABEL: func @_QPteams_allocate
-subroutine teams_allocate()
-   use omp_lib
-   integer :: x
-   integer :: y
-   ! CHECK: omp.teams
-   ! CHECK-SAME: allocate(%{{.+}} : i64 -> %{{.+}} : !fir.ref<i32>)
-   !$omp teams allocate(omp_high_bw_mem_alloc: x) private(x)
-   ! CHECK: arith.addi
-   x = x + 12
-   ! CHECK: omp.terminator
-   !$omp end teams
-end subroutine teams_allocate
diff --git a/flang/test/Lower/OpenMP/FIR/threadprivate-char-array-chararray.f90 b/flang/test/Lower/OpenMP/FIR/threadprivate-char-array-chararray.f90
deleted file mode 100644
index 3580add37ef4..000000000000
--- a/flang/test/Lower/OpenMP/FIR/threadprivate-char-array-chararray.f90
+++ /dev/null
@@ -1,46 +0,0 @@
-! This test checks lowering of OpenMP Threadprivate Directive.
-! Test for character, array, and character array.
-
-!RUN: %flang_fc1 -emit-fir -flang-deprecated-no-hlfir -fopenmp %s -o - | FileCheck %s
-
-module test
-  character :: x
-  integer :: y(5)
-  character(5) :: z(5)
-
-  !$omp threadprivate(x, y, z)
-
-!CHECK-DAG: fir.global @_QMtestEx : !fir.char<1> {
-!CHECK-DAG: fir.global @_QMtestEy : !fir.array<5xi32> {
-!CHECK-DAG: fir.global @_QMtestEz : !fir.array<5x!fir.char<1,5>> {
-
-contains
-  subroutine sub()
-!CHECK-DAG:  [[ADDR0:%.*]] = fir.address_of(@_QMtestEx) : !fir.ref<!fir.char<1>>
-!CHECK-DAG:  [[NEWADDR0:%.*]] = omp.threadprivate [[ADDR0]] : !fir.ref<!fir.char<1>> -> !fir.ref<!fir.char<1>>
-!CHECK-DAG:  [[ADDR1:%.*]] = fir.address_of(@_QMtestEy) : !fir.ref<!fir.array<5xi32>>
-!CHECK-DAG:  [[NEWADDR1:%.*]] = omp.threadprivate [[ADDR1]] : !fir.ref<!fir.array<5xi32>> -> !fir.ref<!fir.array<5xi32>>
-!CHECK-DAG:  [[ADDR2:%.*]] = fir.address_of(@_QMtestEz) : !fir.ref<!fir.array<5x!fir.char<1,5>>>
-!CHECK-DAG:  [[NEWADDR2:%.*]] = omp.threadprivate [[ADDR2]] : !fir.ref<!fir.array<5x!fir.char<1,5>>> -> !fir.ref<!fir.array<5x!fir.char<1,5>>>
-!CHECK-DAG:  %{{.*}} = fir.convert [[NEWADDR0]] : (!fir.ref<!fir.char<1>>) -> !fir.ref<i8>
-!CHECK-DAG:  %{{.*}} = fir.embox [[NEWADDR1]](%{{.*}}) : (!fir.ref<!fir.array<5xi32>>, !fir.shape<1>) -> !fir.box<!fir.array<5xi32>>
-!CHECK-DAG:  %{{.*}} = fir.embox [[NEWADDR2]](%{{.*}}) : (!fir.ref<!fir.array<5x!fir.char<1,5>>>, !fir.shape<1>) -> !fir.box<!fir.array<5x!fir.char<1,5>>>
-    print *, x, y, z
-
-    !$omp parallel
-!CHECK-DAG:    [[ADDR33:%.*]] = omp.threadprivate [[ADDR0]] : !fir.ref<!fir.char<1>> -> !fir.ref<!fir.char<1>>
-!CHECK-DAG:    [[ADDR34:%.*]] = omp.threadprivate [[ADDR1]] : !fir.ref<!fir.array<5xi32>> -> !fir.ref<!fir.array<5xi32>>
-!CHECK-DAG:    [[ADDR35:%.*]] = omp.threadprivate [[ADDR2]] : !fir.ref<!fir.array<5x!fir.char<1,5>>> -> !fir.ref<!fir.array<5x!fir.char<1,5>>>
-!CHECK-DAG:    %{{.*}} = fir.convert [[ADDR33]] : (!fir.ref<!fir.char<1>>) -> !fir.ref<i8>
-!CHECK-DAG:    %{{.*}} = fir.embox [[ADDR34]](%{{.*}}) : (!fir.ref<!fir.array<5xi32>>, !fir.shape<1>) -> !fir.box<!fir.array<5xi32>>
-!CHECK-DAG:    %{{.*}} = fir.embox [[ADDR35]](%{{.*}}) : (!fir.ref<!fir.array<5x!fir.char<1,5>>>, !fir.shape<1>) -> !fir.box<!fir.array<5x!fir.char<1,5>>>
-      print *, x, y, z
-    !$omp end parallel
-
-!CHECK-DAG:  %{{.*}} = fir.convert [[NEWADDR0]] : (!fir.ref<!fir.char<1>>) -> !fir.ref<i8>
-!CHECK-DAG:  %{{.*}} = fir.embox [[NEWADDR1]](%{{.*}}) : (!fir.ref<!fir.array<5xi32>>, !fir.shape<1>) -> !fir.box<!fir.array<5xi32>>
-!CHECK-DAG:  %{{.*}} = fir.embox [[NEWADDR2]](%{{.*}}) : (!fir.ref<!fir.array<5x!fir.char<1,5>>>, !fir.shape<1>) -> !fir.box<!fir.array<5x!fir.char<1,5>>>
-    print *, x, y, z
-
-  end
-end
diff --git a/flang/test/Lower/OpenMP/FIR/threadprivate-commonblock.f90 b/flang/test/Lower/OpenMP/FIR/threadprivate-commonblock.f90
deleted file mode 100644
index 49f592ec8121..000000000000
--- a/flang/test/Lower/OpenMP/FIR/threadprivate-commonblock.f90
+++ /dev/null
@@ -1,91 +0,0 @@
-! This test checks lowering of OpenMP Threadprivate Directive.
-! Test for common block.
-
-!RUN: %flang_fc1 -emit-fir -flang-deprecated-no-hlfir -fopenmp %s -o - | FileCheck %s
-
-module test
-  integer:: a
-  real :: b(2)
-  complex, pointer :: c, d(:)
-  character(5) :: e, f(2)
-  common /blk/ a, b, c, d, e, f
-
-  !$omp threadprivate(/blk/)
-
-!CHECK: fir.global common @blk_(dense<0> : vector<103xi8>) : !fir.array<103xi8>
-
-contains
-  subroutine sub()
-!CHECK:  [[ADDR0:%.*]] = fir.address_of(@blk_) : !fir.ref<!fir.array<103xi8>>
-!CHECK:  [[NEWADDR0:%.*]] = omp.threadprivate [[ADDR0]] : !fir.ref<!fir.array<103xi8>> -> !fir.ref<!fir.array<103xi8>>
-!CHECK-DAG:  [[ADDR1:%.*]] = fir.convert [[NEWADDR0]] : (!fir.ref<!fir.array<103xi8>>) -> !fir.ref<!fir.array<?xi8>>
-!CHECK-DAG:  [[C0:%.*]] = arith.constant 0 : index
-!CHECK-DAG:  [[ADDR2:%.*]] = fir.coordinate_of [[ADDR1]], [[C0]] : (!fir.ref<!fir.array<?xi8>>, index) -> !fir.ref<i8>
-!CHECK-DAG:  [[ADDR3:%.*]] = fir.convert [[ADDR2]] : (!fir.ref<i8>) -> !fir.ref<i32>
-!CHECK-DAG:  [[ADDR4:%.*]] = fir.convert [[NEWADDR0]] : (!fir.ref<!fir.array<103xi8>>) -> !fir.ref<!fir.array<?xi8>>
-!CHECK-DAG:  [[C1:%.*]] = arith.constant 4 : index
-!CHECK-DAG:  [[ADDR5:%.*]] = fir.coordinate_of [[ADDR4]], [[C1]] : (!fir.ref<!fir.array<?xi8>>, index) -> !fir.ref<i8>
-!CHECK-DAG:  [[ADDR6:%.*]] = fir.convert [[ADDR5]] : (!fir.ref<i8>) -> !fir.ref<!fir.array<2xf32>>
-!CHECK-DAG:  [[ADDR7:%.*]] = fir.convert [[NEWADDR0]] : (!fir.ref<!fir.array<103xi8>>) -> !fir.ref<!fir.array<?xi8>>
-!CHECK-DAG:  [[C2:%.*]] = arith.constant 16 : index
-!CHECK-DAG:  [[ADDR8:%.*]] = fir.coordinate_of [[ADDR7]], [[C2]] : (!fir.ref<!fir.array<?xi8>>, index) -> !fir.ref<i8>
-!CHECK-DAG:  [[ADDR9:%.*]] = fir.convert [[ADDR8]] : (!fir.ref<i8>) -> !fir.ref<!fir.box<!fir.ptr<!fir.complex<4>>>>
-!CHECK-DAG:  [[ADDR10:%.*]] = fir.convert [[NEWADDR0]] : (!fir.ref<!fir.array<103xi8>>) -> !fir.ref<!fir.array<?xi8>>
-!CHECK-DAG:  [[C3:%.*]] = arith.constant 40 : index
-!CHECK-DAG:  [[ADDR11:%.*]] = fir.coordinate_of [[ADDR10]], [[C3]] : (!fir.ref<!fir.array<?xi8>>, index) -> !fir.ref<i8>
-!CHECK-DAG:  [[ADDR12:%.*]] = fir.convert [[ADDR11]] : (!fir.ref<i8>) -> !fir.ref<!fir.box<!fir.ptr<!fir.array<?x!fir.complex<4>>>>>
-!CHECK-DAG:  [[ADDR13:%.*]] = fir.convert [[NEWADDR0]] : (!fir.ref<!fir.array<103xi8>>) -> !fir.ref<!fir.array<?xi8>>
-!CHECK-DAG:  [[C4:%.*]] = arith.constant 88 : index
-!CHECK-DAG:  [[ADDR14:%.*]] = fir.coordinate_of [[ADDR13]], [[C4]] : (!fir.ref<!fir.array<?xi8>>, index) -> !fir.ref<i8>
-!CHECK-DAG:  [[ADDR15:%.*]] = fir.convert [[ADDR14]] : (!fir.ref<i8>) -> !fir.ref<!fir.char<1,5>>
-!CHECK-DAG:  [[ADDR16:%.*]] = fir.convert [[NEWADDR0]] : (!fir.ref<!fir.array<103xi8>>) -> !fir.ref<!fir.array<?xi8>>
-!CHECK-DAG:  [[C5:%.*]] = arith.constant 93 : index
-!CHECK-DAG:  [[ADDR17:%.*]] = fir.coordinate_of [[ADDR16]], [[C5]] : (!fir.ref<!fir.array<?xi8>>, index) -> !fir.ref<i8>
-!CHECK-DAG:  [[ADDR18:%.*]] = fir.convert [[ADDR17]] : (!fir.ref<i8>) -> !fir.ref<!fir.array<2x!fir.char<1,5>>>
-!CHECK-DAG:  %{{.*}} = fir.load [[ADDR3]] : !fir.ref<i32>
-!CHECK-DAG:  %{{.*}} = fir.embox [[ADDR6]](%{{.*}}) : (!fir.ref<!fir.array<2xf32>>, !fir.shape<1>) -> !fir.box<!fir.array<2xf32>>
-!CHECK-DAG:  %{{.*}} = fir.load [[ADDR9]] : !fir.ref<!fir.box<!fir.ptr<!fir.complex<4>>>>
-!CHECK-DAG:  %{{.*}} = fir.load [[ADDR12]] : !fir.ref<!fir.box<!fir.ptr<!fir.array<?x!fir.complex<4>>>>>
-!CHECK-DAG:  %{{.*}} = fir.convert [[ADDR15]] : (!fir.ref<!fir.char<1,5>>) -> !fir.ref<i8>
-!CHECK-DAG:  %{{.*}} = fir.embox [[ADDR18]](%{{.*}}) : (!fir.ref<!fir.array<2x!fir.char<1,5>>>, !fir.shape<1>) -> !fir.box<!fir.array<2x!fir.char<1,5>>>
-    print *, a, b, c, d, e, f
-
-    !$omp parallel
-!CHECK:    [[ADDR77:%.*]] = omp.threadprivate [[ADDR0]] : !fir.ref<!fir.array<103xi8>> -> !fir.ref<!fir.array<103xi8>>
-!CHECK-DAG:    [[ADDR78:%.*]] = fir.convert [[ADDR77]] : (!fir.ref<!fir.array<103xi8>>) -> !fir.ref<!fir.array<?xi8>>
-!CHECK-DAG:    [[ADDR79:%.*]] = fir.coordinate_of [[ADDR78]], [[C0:%.*]] : (!fir.ref<!fir.array<?xi8>>, index) -> !fir.ref<i8>
-!CHECK-DAG:    [[ADDR80:%.*]] = fir.convert [[ADDR79:%.*]] : (!fir.ref<i8>) -> !fir.ref<i32>
-!CHECK-DAG:    [[ADDR81:%.*]] = fir.convert [[ADDR77]] : (!fir.ref<!fir.array<103xi8>>) -> !fir.ref<!fir.array<?xi8>>
-!CHECK-DAG:    [[ADDR82:%.*]] = fir.coordinate_of [[ADDR81]], [[C1:%.*]] : (!fir.ref<!fir.array<?xi8>>, index) -> !fir.ref<i8>
-!CHECK-DAG:    [[ADDR83:%.*]] = fir.convert [[ADDR82:%.*]] : (!fir.ref<i8>) -> !fir.ref<!fir.array<2xf32>>
-!CHECK-DAG:    [[ADDR84:%.*]] = fir.convert [[ADDR77]] : (!fir.ref<!fir.array<103xi8>>) -> !fir.ref<!fir.array<?xi8>>
-!CHECK-DAG:    [[ADDR85:%.*]] = fir.coordinate_of [[ADDR84]], [[C2:%.*]] : (!fir.ref<!fir.array<?xi8>>, index) -> !fir.ref<i8>
-!CHECK-DAG:    [[ADDR86:%.*]] = fir.convert [[ADDR85:%.*]] : (!fir.ref<i8>) -> !fir.ref<!fir.box<!fir.ptr<!fir.complex<4>>>>
-!CHECK-DAG:    [[ADDR87:%.*]] = fir.convert [[ADDR77]] : (!fir.ref<!fir.array<103xi8>>) -> !fir.ref<!fir.array<?xi8>>
-!CHECK-DAG:    [[ADDR88:%.*]] = fir.coordinate_of [[ADDR87]], [[C3:%.*]] : (!fir.ref<!fir.array<?xi8>>, index) -> !fir.ref<i8>
-!CHECK-DAG:    [[ADDR89:%.*]] = fir.convert [[ADDR88:%.*]] : (!fir.ref<i8>) -> !fir.ref<!fir.box<!fir.ptr<!fir.array<?x!fir.complex<4>>>>>
-!CHECK-DAG:    [[ADDR90:%.*]] = fir.convert [[ADDR77]] : (!fir.ref<!fir.array<103xi8>>) -> !fir.ref<!fir.array<?xi8>>
-!CHECK-DAG:    [[ADDR91:%.*]] = fir.coordinate_of [[ADDR90]], [[C4:%.*]] : (!fir.ref<!fir.array<?xi8>>, index) -> !fir.ref<i8>
-!CHECK-DAG:    [[ADDR92:%.*]] = fir.convert [[ADDR91:%.*]] : (!fir.ref<i8>) -> !fir.ref<!fir.char<1,5>>
-!CHECK-DAG:    [[ADDR93:%.*]] = fir.convert [[ADDR77]] : (!fir.ref<!fir.array<103xi8>>) -> !fir.ref<!fir.array<?xi8>>
-!CHECK-DAG:    [[ADDR94:%.*]] = fir.coordinate_of [[ADDR93]], [[C5:%.*]] : (!fir.ref<!fir.array<?xi8>>, index) -> !fir.ref<i8>
-!CHECK-DAG:    [[ADDR95:%.*]] = fir.convert [[ADDR94:%.*]] : (!fir.ref<i8>) -> !fir.ref<!fir.array<2x!fir.char<1,5>>>
-!CHECK-DAG:    %{{.*}} = fir.load [[ADDR80]] : !fir.ref<i32>
-!CHECK-DAG:    %{{.*}} = fir.embox [[ADDR83]](%{{.*}}) : (!fir.ref<!fir.array<2xf32>>, !fir.shape<1>) -> !fir.box<!fir.array<2xf32>>
-!CHECK-DAG:    %{{.*}} = fir.load [[ADDR86]] : !fir.ref<!fir.box<!fir.ptr<!fir.complex<4>>>>
-!CHECK-DAG:    %{{.*}} = fir.load [[ADDR89]] : !fir.ref<!fir.box<!fir.ptr<!fir.array<?x!fir.complex<4>>>>>
-!CHECK-DAG:    %{{.*}} = fir.convert [[ADDR92]] : (!fir.ref<!fir.char<1,5>>) -> !fir.ref<i8>
-!CHECK-DAG:    %{{.*}} = fir.embox [[ADDR95]](%{{.*}}) : (!fir.ref<!fir.array<2x!fir.char<1,5>>>, !fir.shape<1>) -> !fir.box<!fir.array<2x!fir.char<1,5>>>
-      print *, a, b, c, d, e, f
-    !$omp end parallel
-
-!CHECK-DAG:  %{{.*}} = fir.load [[ADDR3]] : !fir.ref<i32>
-!CHECK-DAG:  %{{.*}} = fir.embox [[ADDR6]](%{{.*}}) : (!fir.ref<!fir.array<2xf32>>, !fir.shape<1>) -> !fir.box<!fir.array<2xf32>>
-!CHECK-DAG:  %{{.*}} = fir.load [[ADDR9]] : !fir.ref<!fir.box<!fir.ptr<!fir.complex<4>>>>
-!CHECK-DAG:  %{{.*}} = fir.load [[ADDR12]] : !fir.ref<!fir.box<!fir.ptr<!fir.array<?x!fir.complex<4>>>>>
-!CHECK-DAG:  %{{.*}} = fir.convert [[ADDR15]] : (!fir.ref<!fir.char<1,5>>) -> !fir.ref<i8>
-!CHECK-DAG:  %{{.*}} = fir.embox [[ADDR18]](%{{.*}}) : (!fir.ref<!fir.array<2x!fir.char<1,5>>>, !fir.shape<1>) -> !fir.box<!fir.array<2x!fir.char<1,5>>>
-    print *, a, b, c, d, e, f
-
-  end
-end
diff --git a/flang/test/Lower/OpenMP/FIR/threadprivate-integer-different-kinds.f90 b/flang/test/Lower/OpenMP/FIR/threadprivate-integer-different-kinds.f90
deleted file mode 100644
index 39c77406cc22..000000000000
--- a/flang/test/Lower/OpenMP/FIR/threadprivate-integer-different-kinds.f90
+++ /dev/null
@@ -1,67 +0,0 @@
-! This test checks lowering of OpenMP Threadprivate Directive.
-! Test for variables with different kind.
-
-!REQUIRES: shell
-!RUN: %flang_fc1 -emit-fir -flang-deprecated-no-hlfir -fopenmp %s -o - | FileCheck %s
-
-program test
-  integer, save :: i
-  integer(kind=1), save :: i1
-  integer(kind=2), save :: i2
-  integer(kind=4), save :: i4
-  integer(kind=8), save :: i8
-  integer(kind=16), save :: i16
-
-!CHECK-DAG:  [[ADDR0:%.*]] = fir.address_of(@_QFEi) : !fir.ref<i32>
-!CHECK-DAG:  [[NEWADDR0:%.*]] = omp.threadprivate [[ADDR0]] : !fir.ref<i32> -> !fir.ref<i32>
-!CHECK-DAG:  [[ADDR1:%.*]] = fir.address_of(@_QFEi1) : !fir.ref<i8>
-!CHECK-DAG:  [[NEWADDR1:%.*]] = omp.threadprivate [[ADDR1]] : !fir.ref<i8> -> !fir.ref<i8>
-!CHECK-DAG:  [[ADDR2:%.*]] = fir.address_of(@_QFEi16) : !fir.ref<i128>
-!CHECK-DAG:  [[NEWADDR2:%.*]] = omp.threadprivate [[ADDR2]] : !fir.ref<i128> -> !fir.ref<i128>
-!CHECK-DAG:  [[ADDR3:%.*]] = fir.address_of(@_QFEi2) : !fir.ref<i16>
-!CHECK-DAG:  [[NEWADDR3:%.*]] = omp.threadprivate [[ADDR3]] : !fir.ref<i16> -> !fir.ref<i16>
-!CHECK-DAG:  [[ADDR4:%.*]] = fir.address_of(@_QFEi4) : !fir.ref<i32>
-!CHECK-DAG:  [[NEWADDR4:%.*]] = omp.threadprivate [[ADDR4]] : !fir.ref<i32> -> !fir.ref<i32>
-!CHECK-DAG:  [[ADDR5:%.*]] = fir.address_of(@_QFEi8) : !fir.ref<i64>
-!CHECK-DAG:  [[NEWADDR5:%.*]] = omp.threadprivate [[ADDR5]] : !fir.ref<i64> -> !fir.ref<i64>
-  !$omp threadprivate(i, i1, i2, i4, i8, i16)
-
-!CHECK-DAG:  %{{.*}} = fir.load [[NEWADDR0]] : !fir.ref<i32>
-!CHECK-DAG:  %{{.*}} = fir.load [[NEWADDR1]] : !fir.ref<i8>
-!CHECK-DAG:  %{{.*}} = fir.load [[NEWADDR2]] : !fir.ref<i128>
-!CHECK-DAG:  %{{.*}} = fir.load [[NEWADDR3]] : !fir.ref<i16>
-!CHECK-DAG:  %{{.*}} = fir.load [[NEWADDR4]] : !fir.ref<i32>
-!CHECK-DAG:  %{{.*}} = fir.load [[NEWADDR5]] : !fir.ref<i64>
-  print *, i, i1, i2, i4, i8, i16
-
-  !$omp parallel
-!CHECK-DAG:    [[ADDR39:%.*]] = omp.threadprivate [[ADDR0]] : !fir.ref<i32> -> !fir.ref<i32>
-!CHECK-DAG:    [[ADDR40:%.*]] = omp.threadprivate [[ADDR1]] : !fir.ref<i8> -> !fir.ref<i8>
-!CHECK-DAG:    [[ADDR41:%.*]] = omp.threadprivate [[ADDR2]] : !fir.ref<i128> -> !fir.ref<i128>
-!CHECK-DAG:    [[ADDR42:%.*]] = omp.threadprivate [[ADDR3]] : !fir.ref<i16> -> !fir.ref<i16>
-!CHECK-DAG:    [[ADDR43:%.*]] = omp.threadprivate [[ADDR4]] : !fir.ref<i32> -> !fir.ref<i32>
-!CHECK-DAG:    [[ADDR44:%.*]] = omp.threadprivate [[ADDR5]] : !fir.ref<i64> -> !fir.ref<i64>
-!CHECK-DAG:    %{{.*}} = fir.load [[ADDR39]] : !fir.ref<i32>
-!CHECK-DAG:    %{{.*}} = fir.load [[ADDR40]] : !fir.ref<i8>
-!CHECK-DAG:    %{{.*}} = fir.load [[ADDR41]] : !fir.ref<i128>
-!CHECK-DAG:    %{{.*}} = fir.load [[ADDR42]] : !fir.ref<i16>
-!CHECK-DAG:    %{{.*}} = fir.load [[ADDR43]] : !fir.ref<i32>
-!CHECK-DAG:    %{{.*}} = fir.load [[ADDR44]] : !fir.ref<i64>
-    print *, i, i1, i2, i4, i8, i16
-  !$omp end parallel
-
-!CHECK-DAG:  %{{.*}} = fir.load [[NEWADDR0]] : !fir.ref<i32>
-!CHECK-DAG:  %{{.*}} = fir.load [[NEWADDR1]] : !fir.ref<i8>
-!CHECK-DAG:  %{{.*}} = fir.load [[NEWADDR2]] : !fir.ref<i128>
-!CHECK-DAG:  %{{.*}} = fir.load [[NEWADDR3]] : !fir.ref<i16>
-!CHECK-DAG:  %{{.*}} = fir.load [[NEWADDR4]] : !fir.ref<i32>
-!CHECK-DAG:  %{{.*}} = fir.load [[NEWADDR5]] : !fir.ref<i64>
-  print *, i, i1, i2, i4, i8, i16
-
-!CHECK-DAG: fir.global internal @_QFEi : i32 {
-!CHECK-DAG: fir.global internal @_QFEi1 : i8 {
-!CHECK-DAG: fir.global internal @_QFEi16 : i128 {
-!CHECK-DAG: fir.global internal @_QFEi2 : i16 {
-!CHECK-DAG: fir.global internal @_QFEi4 : i32 {
-!CHECK-DAG: fir.global internal @_QFEi8 : i64 {
-end
diff --git a/flang/test/Lower/OpenMP/FIR/threadprivate-non-global.f90 b/flang/test/Lower/OpenMP/FIR/threadprivate-non-global.f90
deleted file mode 100644
index b089693b2097..000000000000
--- a/flang/test/Lower/OpenMP/FIR/threadprivate-non-global.f90
+++ /dev/null
@@ -1,91 +0,0 @@
-! This test checks lowering of OpenMP Threadprivate Directive.
-! Test for non-character non-SAVEd non-initialized scalars with or without
-! allocatable or pointer attribute in main program.
-
-!RUN: %flang_fc1 -emit-fir -flang-deprecated-no-hlfir -fopenmp %s -o - | FileCheck %s
-
-program test
-  integer :: x
-  real :: y
-  logical :: z
-  complex :: w
-  integer, pointer :: a
-  real, allocatable :: b
-
-!CHECK-DAG:  [[ADDR0:%.*]] = fir.address_of(@_QFEa) : !fir.ref<!fir.box<!fir.ptr<i32>>>
-!CHECK-DAG:  [[NEWADDR0:%.*]] = omp.threadprivate [[ADDR0]] : !fir.ref<!fir.box<!fir.ptr<i32>>> -> !fir.ref<!fir.box<!fir.ptr<i32>>>
-!CHECK-DAG:  [[ADDR1:%.*]] = fir.address_of(@_QFEb) : !fir.ref<!fir.box<!fir.heap<f32>>>
-!CHECK-DAG:  [[NEWADDR1:%.*]] = omp.threadprivate [[ADDR1]] : !fir.ref<!fir.box<!fir.heap<f32>>> -> !fir.ref<!fir.box<!fir.heap<f32>>>
-!CHECK-DAG:  [[ADDR2:%.*]] = fir.address_of(@_QFEw) : !fir.ref<!fir.complex<4>>
-!CHECK-DAG:  [[NEWADDR2:%.*]] = omp.threadprivate [[ADDR2]] : !fir.ref<!fir.complex<4>> -> !fir.ref<!fir.complex<4>>
-!CHECK-DAG:  [[ADDR3:%.*]] = fir.address_of(@_QFEx) : !fir.ref<i32>
-!CHECK-DAG:  [[NEWADDR3:%.*]] = omp.threadprivate [[ADDR3]] : !fir.ref<i32> -> !fir.ref<i32>
-!CHECK-DAG:  [[ADDR4:%.*]] = fir.address_of(@_QFEy) : !fir.ref<f32>
-!CHECK-DAG:  [[NEWADDR4:%.*]] = omp.threadprivate [[ADDR4]] : !fir.ref<f32> -> !fir.ref<f32>
-!CHECK-DAG:  [[ADDR5:%.*]] = fir.address_of(@_QFEz) : !fir.ref<!fir.logical<4>>
-!CHECK-DAG:  [[NEWADDR5:%.*]] = omp.threadprivate [[ADDR5]] : !fir.ref<!fir.logical<4>> -> !fir.ref<!fir.logical<4>>
-  !$omp threadprivate(x, y, z, w, a, b)
-
-  call sub(a, b)
-
-!CHECK-DAG:  %{{.*}} = fir.load [[NEWADDR3]] : !fir.ref<i32>
-!CHECK-DAG:  %{{.*}} = fir.load [[NEWADDR4]] : !fir.ref<f32>
-!CHECK-DAG:  %{{.*}} = fir.load [[NEWADDR5]] : !fir.ref<!fir.logical<4>>
-!CHECK-DAG:  %{{.*}} = fir.load [[NEWADDR2]] : !fir.ref<!fir.complex<4>>
-!CHECK-DAG:  %{{.*}} = fir.load [[NEWADDR0]] : !fir.ref<!fir.box<!fir.ptr<i32>>>
-!CHECK-DAG:  %{{.*}} = fir.load [[NEWADDR1]] : !fir.ref<!fir.box<!fir.heap<f32>>>
-  print *, x, y, z, w, a, b
-
-  !$omp parallel
-!CHECK-DAG:    [[ADDR68:%.*]] = omp.threadprivate [[ADDR0]] : !fir.ref<!fir.box<!fir.ptr<i32>>> -> !fir.ref<!fir.box<!fir.ptr<i32>>>
-!CHECK-DAG:    [[ADDR69:%.*]] = omp.threadprivate [[ADDR1]] : !fir.ref<!fir.box<!fir.heap<f32>>> -> !fir.ref<!fir.box<!fir.heap<f32>>>
-!CHECK-DAG:    [[ADDR70:%.*]] = omp.threadprivate [[ADDR2]] : !fir.ref<!fir.complex<4>> -> !fir.ref<!fir.complex<4>>
-!CHECK-DAG:    [[ADDR71:%.*]] = omp.threadprivate [[ADDR3]] : !fir.ref<i32> -> !fir.ref<i32>
-!CHECK-DAG:    [[ADDR72:%.*]] = omp.threadprivate [[ADDR4]] : !fir.ref<f32> -> !fir.ref<f32>
-!CHECK-DAG:    [[ADDR73:%.*]] = omp.threadprivate [[ADDR5]] : !fir.ref<!fir.logical<4>> -> !fir.ref<!fir.logical<4>>
-!CHECK-DAG:    %{{.*}} = fir.load [[ADDR71]] : !fir.ref<i32>
-!CHECK-DAG:    %{{.*}} = fir.load [[ADDR72]] : !fir.ref<f32>
-!CHECK-DAG:    %{{.*}} = fir.load [[ADDR73]] : !fir.ref<!fir.logical<4>>
-!CHECK-DAG:    %{{.*}} = fir.load [[ADDR70]] : !fir.ref<!fir.complex<4>>
-!CHECK-DAG:    %{{.*}} = fir.load [[ADDR68]] : !fir.ref<!fir.box<!fir.ptr<i32>>>
-!CHECK-DAG:    %{{.*}} = fir.load [[ADDR69]] : !fir.ref<!fir.box<!fir.heap<f32>>>
-    print *, x, y, z, w, a, b
-  !$omp end parallel
-
-!CHECK-DAG:  %{{.*}} = fir.load [[NEWADDR3]] : !fir.ref<i32>
-!CHECK-DAG:  %{{.*}} = fir.load [[NEWADDR4]] : !fir.ref<f32>
-!CHECK-DAG:  %{{.*}} = fir.load [[NEWADDR5]] : !fir.ref<!fir.logical<4>>
-!CHECK-DAG:  %{{.*}} = fir.load [[NEWADDR2]] : !fir.ref<!fir.complex<4>>
-!CHECK-DAG:  %{{.*}} = fir.load [[NEWADDR0]] : !fir.ref<!fir.box<!fir.ptr<i32>>>
-!CHECK-DAG:  %{{.*}} = fir.load [[NEWADDR1]] : !fir.ref<!fir.box<!fir.heap<f32>>>
-  print *, x, y, z, w, a, b
-
-!CHECK:  return
-
-!CHECK-DAG: fir.global internal @_QFEa : !fir.box<!fir.ptr<i32>> {
-!CHECK-DAG:   [[Z0:%.*]] = fir.zero_bits !fir.ptr<i32>
-!CHECK-DAG:   [[E0:%.*]] = fir.embox [[Z0]] : (!fir.ptr<i32>) -> !fir.box<!fir.ptr<i32>>
-!CHECK-DAG:   fir.has_value [[E0]] : !fir.box<!fir.ptr<i32>>
-!CHECK-DAG: }
-!CHECK-DAG: fir.global internal @_QFEb : !fir.box<!fir.heap<f32>> {
-!CHECK-DAG:   [[Z1:%.*]] = fir.zero_bits !fir.heap<f32>
-!CHECK-DAG:   [[E1:%.*]] = fir.embox [[Z1]] : (!fir.heap<f32>) -> !fir.box<!fir.heap<f32>>
-!CHECK-DAG:   fir.has_value [[E1]] : !fir.box<!fir.heap<f32>>
-!CHECK-DAG: }
-!CHECK-DAG: fir.global internal @_QFEw : !fir.complex<4> {
-!CHECK-DAG:   [[Z2:%.*]] = fir.undefined !fir.complex<4>
-!CHECK-DAG:   fir.has_value [[Z2]] : !fir.complex<4>
-!CHECK-DAG: }
-!CHECK-DAG: fir.global internal @_QFEx : i32 {
-!CHECK-DAG:   [[Z3:%.*]] = fir.undefined i32
-!CHECK-DAG:   fir.has_value [[Z3]] : i32
-!CHECK-DAG: }
-!CHECK-DAG: fir.global internal @_QFEy : f32 {
-!CHECK-DAG:   [[Z4:%.*]] = fir.undefined f32
-!CHECK-DAG:   fir.has_value [[Z4]] : f32
-!CHECK-DAG: }
-!CHECK-DAG: fir.global internal @_QFEz : !fir.logical<4> {
-!CHECK-DAG:   [[Z5:%.*]] = fir.undefined !fir.logical<4>
-!CHECK-DAG:   fir.has_value [[Z5]] : !fir.logical<4>
-!CHECK-DAG: }
-end
diff --git a/flang/test/Lower/OpenMP/FIR/threadprivate-pointer-allocatable.f90 b/flang/test/Lower/OpenMP/FIR/threadprivate-pointer-allocatable.f90
deleted file mode 100644
index fd33c20f9f93..000000000000
--- a/flang/test/Lower/OpenMP/FIR/threadprivate-pointer-allocatable.f90
+++ /dev/null
@@ -1,51 +0,0 @@
-! This test checks lowering of OpenMP Threadprivate Directive.
-! Test for allocatable and pointer variables.
-
-!RUN: %flang_fc1 -emit-fir -flang-deprecated-no-hlfir -fopenmp %s -o - | FileCheck %s
-
-module test
-  integer, pointer :: x(:), m
-  real, allocatable :: y(:), n
-
-  !$omp threadprivate(x, y, m, n)
-
-!CHECK-DAG: fir.global @_QMtestEm : !fir.box<!fir.ptr<i32>> {
-!CHECK-DAG: fir.global @_QMtestEn : !fir.box<!fir.heap<f32>> {
-!CHECK-DAG: fir.global @_QMtestEx : !fir.box<!fir.ptr<!fir.array<?xi32>>> {
-!CHECK-DAG: fir.global @_QMtestEy : !fir.box<!fir.heap<!fir.array<?xf32>>> {
-
-contains
-  subroutine sub()
-!CHECK-DAG:  [[ADDR0:%.*]] = fir.address_of(@_QMtestEm) : !fir.ref<!fir.box<!fir.ptr<i32>>>
-!CHECK-DAG:  [[NEWADDR0:%.*]] = omp.threadprivate [[ADDR0]] : !fir.ref<!fir.box<!fir.ptr<i32>>> -> !fir.ref<!fir.box<!fir.ptr<i32>>>
-!CHECK-DAG:  [[ADDR1:%.*]] = fir.address_of(@_QMtestEn) : !fir.ref<!fir.box<!fir.heap<f32>>>
-!CHECK-DAG:  [[NEWADDR1:%.*]] = omp.threadprivate [[ADDR1]] : !fir.ref<!fir.box<!fir.heap<f32>>> -> !fir.ref<!fir.box<!fir.heap<f32>>>
-!CHECK-DAG:  [[ADDR2:%.*]] = fir.address_of(@_QMtestEx) : !fir.ref<!fir.box<!fir.ptr<!fir.array<?xi32>>>>
-!CHECK-DAG:  [[NEWADDR2:%.*]] = omp.threadprivate [[ADDR2]] : !fir.ref<!fir.box<!fir.ptr<!fir.array<?xi32>>>> -> !fir.ref<!fir.box<!fir.ptr<!fir.array<?xi32>>>>
-!CHECK-DAG:  [[ADDR3:%.*]] = fir.address_of(@_QMtestEy) : !fir.ref<!fir.box<!fir.heap<!fir.array<?xf32>>>>
-!CHECK-DAG:  [[NEWADDR3:%.*]] = omp.threadprivate [[ADDR3]] : !fir.ref<!fir.box<!fir.heap<!fir.array<?xf32>>>> -> !fir.ref<!fir.box<!fir.heap<!fir.array<?xf32>>>>
-!CHECK-DAG:  %{{.*}} = fir.load [[NEWADDR2]] : !fir.ref<!fir.box<!fir.ptr<!fir.array<?xi32>>>>
-!CHECK-DAG:  %{{.*}} = fir.load [[NEWADDR3]] : !fir.ref<!fir.box<!fir.heap<!fir.array<?xf32>>>>
-!CHECK-DAG:  %{{.*}} = fir.load [[NEWADDR0]] : !fir.ref<!fir.box<!fir.ptr<i32>>>
-!CHECK-DAG:  %{{.*}} = fir.load [[NEWADDR1]] : !fir.ref<!fir.box<!fir.heap<f32>>>
-    print *, x, y, m, n
-
-    !$omp parallel
-!CHECK-DAG:    [[ADDR54:%.*]] = omp.threadprivate [[ADDR0]] : !fir.ref<!fir.box<!fir.ptr<i32>>> -> !fir.ref<!fir.box<!fir.ptr<i32>>>
-!CHECK-DAG:    [[ADDR55:%.*]] = omp.threadprivate [[ADDR1]] : !fir.ref<!fir.box<!fir.heap<f32>>> -> !fir.ref<!fir.box<!fir.heap<f32>>>
-!CHECK-DAG:    [[ADDR56:%.*]] = omp.threadprivate [[ADDR2]] : !fir.ref<!fir.box<!fir.ptr<!fir.array<?xi32>>>> -> !fir.ref<!fir.box<!fir.ptr<!fir.array<?xi32>>>>
-!CHECK-DAG:    [[ADDR57:%.*]] = omp.threadprivate [[ADDR3]] : !fir.ref<!fir.box<!fir.heap<!fir.array<?xf32>>>> -> !fir.ref<!fir.box<!fir.heap<!fir.array<?xf32>>>>
-!CHECK-DAG:    %{{.*}} = fir.load [[ADDR56]] : !fir.ref<!fir.box<!fir.ptr<!fir.array<?xi32>>>>
-!CHECK-DAG:    %{{.*}} = fir.load [[ADDR57]] : !fir.ref<!fir.box<!fir.heap<!fir.array<?xf32>>>>
-!CHECK-DAG:    %{{.*}} = fir.load [[ADDR54]] : !fir.ref<!fir.box<!fir.ptr<i32>>>
-!CHECK-DAG:    %{{.*}} = fir.load [[ADDR55]] : !fir.ref<!fir.box<!fir.heap<f32>>>
-      print *, x, y, m, n
-    !$omp end parallel
-
-!CHECK-DAG:  %{{.*}} = fir.load [[NEWADDR2]] : !fir.ref<!fir.box<!fir.ptr<!fir.array<?xi32>>>>
-!CHECK-DAG:  %{{.*}} = fir.load [[NEWADDR3]] : !fir.ref<!fir.box<!fir.heap<!fir.array<?xf32>>>>
-!CHECK-DAG:  %{{.*}} = fir.load [[NEWADDR0]] : !fir.ref<!fir.box<!fir.ptr<i32>>>
-!CHECK-DAG:  %{{.*}} = fir.load [[NEWADDR1]] : !fir.ref<!fir.box<!fir.heap<f32>>>
-    print *, x, y, m, n
-  end
-end
diff --git a/flang/test/Lower/OpenMP/FIR/threadprivate-real-logical-complex-derivedtype.f90 b/flang/test/Lower/OpenMP/FIR/threadprivate-real-logical-complex-derivedtype.f90
deleted file mode 100644
index 749fe5c8bf54..000000000000
--- a/flang/test/Lower/OpenMP/FIR/threadprivate-real-logical-complex-derivedtype.f90
+++ /dev/null
@@ -1,58 +0,0 @@
-! This test checks lowering of OpenMP Threadprivate Directive.
-! Test for real, logical, complex, and derived type.
-
-!RUN: %flang_fc1 -emit-fir -flang-deprecated-no-hlfir -fopenmp %s -o - | FileCheck %s
-
-module test
-  type my_type
-    integer :: t_i
-    real :: t_arr(5)
-  end type my_type
-  real :: x
-  complex :: y
-  logical :: z
-  type(my_type) :: t
-
-  !$omp threadprivate(x, y, z, t)
-
-!CHECK-DAG: fir.global @_QMtestEt : !fir.type<_QMtestTmy_type{t_i:i32,t_arr:!fir.array<5xf32>}> {
-!CHECK-DAG: fir.global @_QMtestEx : f32 {
-!CHECK-DAG: fir.global @_QMtestEy : !fir.complex<4> {
-!CHECK-DAG: fir.global @_QMtestEz : !fir.logical<4> {
-
-contains
-  subroutine sub()
-!CHECK-DAG:  [[ADDR0:%.*]] = fir.address_of(@_QMtestEt) : !fir.ref<!fir.type<_QMtestTmy_type{t_i:i32,t_arr:!fir.array<5xf32>}>>
-!CHECK-DAG:  [[NEWADDR0:%.*]] = omp.threadprivate [[ADDR0]] : !fir.ref<!fir.type<_QMtestTmy_type{t_i:i32,t_arr:!fir.array<5xf32>}>> -> !fir.ref<!fir.type<_QMtestTmy_type{t_i:i32,t_arr:!fir.array<5xf32>}>>
-!CHECK-DAG:  [[ADDR1:%.*]] = fir.address_of(@_QMtestEx) : !fir.ref<f32>
-!CHECK-DAG:  [[NEWADDR1:%.*]] = omp.threadprivate [[ADDR1]] : !fir.ref<f32> -> !fir.ref<f32>
-!CHECK-DAG:  [[ADDR2:%.*]] = fir.address_of(@_QMtestEy) : !fir.ref<!fir.complex<4>>
-!CHECK-DAG:  [[NEWADDR2:%.*]] = omp.threadprivate [[ADDR2]] : !fir.ref<!fir.complex<4>> -> !fir.ref<!fir.complex<4>>
-!CHECK-DAG:  [[ADDR3:%.*]] = fir.address_of(@_QMtestEz) : !fir.ref<!fir.logical<4>>
-!CHECK-DAG:  [[NEWADDR3:%.*]] = omp.threadprivate [[ADDR3]] : !fir.ref<!fir.logical<4>> -> !fir.ref<!fir.logical<4>>
-!CHECK-DAG:  %{{.*}} = fir.load [[NEWADDR1]] : !fir.ref<f32>
-!CHECK-DAG:  %{{.*}} = fir.load [[NEWADDR2]] : !fir.ref<!fir.complex<4>>
-!CHECK-DAG:  %{{.*}} = fir.load [[NEWADDR3]] : !fir.ref<!fir.logical<4>>
-!CHECK-DAG:  %{{.*}} = fir.coordinate_of [[NEWADDR0]]
-    print *, x, y, z, t%t_i
-
-    !$omp parallel
-!CHECK-DAG:    [[ADDR38:%.*]] = omp.threadprivate [[ADDR0]] : !fir.ref<!fir.type<_QMtestTmy_type{t_i:i32,t_arr:!fir.array<5xf32>}>> -> !fir.ref<!fir.type<_QMtestTmy_type{t_i:i32,t_arr:!fir.array<5xf32>}>>
-!CHECK-DAG:    [[ADDR39:%.*]] = omp.threadprivate [[ADDR1]] : !fir.ref<f32> -> !fir.ref<f32>
-!CHECK-DAG:    [[ADDR40:%.*]] = omp.threadprivate [[ADDR2]] : !fir.ref<!fir.complex<4>> -> !fir.ref<!fir.complex<4>>
-!CHECK-DAG:    [[ADDR41:%.*]] = omp.threadprivate [[ADDR3]] : !fir.ref<!fir.logical<4>> -> !fir.ref<!fir.logical<4>>
-!CHECK-DAG:    %{{.*}} = fir.load [[ADDR39]] : !fir.ref<f32>
-!CHECK-DAG:    %{{.*}} = fir.load [[ADDR40]] : !fir.ref<!fir.complex<4>>
-!CHECK-DAG:    %{{.*}} = fir.load [[ADDR41]] : !fir.ref<!fir.logical<4>>
-!CHECK-DAG:    %{{.*}} = fir.coordinate_of [[ADDR38]]
-      print *, x, y, z, t%t_i
-    !$omp end parallel
-
-!CHECK-DAG:  %{{.*}} = fir.load [[NEWADDR1]] : !fir.ref<f32>
-!CHECK-DAG:  %{{.*}} = fir.load [[NEWADDR2]] : !fir.ref<!fir.complex<4>>
-!CHECK-DAG:  %{{.*}} = fir.load [[NEWADDR3]] : !fir.ref<!fir.logical<4>>
-!CHECK-DAG:  %{{.*}} = fir.coordinate_of [[NEWADDR0]]
-    print *, x, y, z, t%t_i
-
-  end
-end
diff --git a/flang/test/Lower/OpenMP/FIR/threadprivate-use-association-2.f90 b/flang/test/Lower/OpenMP/FIR/threadprivate-use-association-2.f90
deleted file mode 100644
index 6db5735c21f1..000000000000
--- a/flang/test/Lower/OpenMP/FIR/threadprivate-use-association-2.f90
+++ /dev/null
@@ -1,39 +0,0 @@
-! This test checks lowering of OpenMP Threadprivate Directive.
-! Test for threadprivate variable double use in use association.
-
-!RUN: %flang_fc1 -emit-fir -flang-deprecated-no-hlfir -fopenmp %s -o - | FileCheck %s
-!RUN: bbc -emit-fir -hlfir=false -fopenmp %s -o - | FileCheck %s
-
-! CHECK-LABEL: fir.global @_QMmEx : i32
-module m
-  integer :: x
-  !$omp threadprivate(x)
-end
-
-! CHECK-LABEL: func.func @_QMm2Ptest() {
-! CHECK:         %[[VAL_0:.*]] = fir.address_of(@_QMmEx) : !fir.ref<i32>
-! CHECK:         %[[VAL_1:.*]] = omp.threadprivate %[[VAL_0]] : !fir.ref<i32> -> !fir.ref<i32>
-! CHECK:         fir.call @_QPbar(%[[VAL_1]]) {{.*}}: (!fir.ref<i32>) -> ()
-! CHECK:         return
-! CHECK:       }
-!
-! CHECK-LABEL: func.func private @_QMm2FtestPinternal_test() {{.*}} {
-! CHECK:         %[[VAL_0:.*]] = fir.address_of(@_QMmEx) : !fir.ref<i32>
-! CHECK:         %[[VAL_1:.*]] = omp.threadprivate %[[VAL_0]] : !fir.ref<i32> -> !fir.ref<i32>
-! CHECK:         fir.call @_QPbar(%[[VAL_1]]) {{.*}}: (!fir.ref<i32>) -> ()
-! CHECK:         return
-! CHECK:       }
-
-module m2
-  use m
- contains
-  subroutine test()
-    use m
-    call bar(x)
-   contains
-    subroutine internal_test()
-      use m
-      call bar(x)
-    end
-  end
-end
diff --git a/flang/test/Lower/OpenMP/FIR/threadprivate-use-association.f90 b/flang/test/Lower/OpenMP/FIR/threadprivate-use-association.f90
deleted file mode 100644
index 685237430a1c..000000000000
--- a/flang/test/Lower/OpenMP/FIR/threadprivate-use-association.f90
+++ /dev/null
@@ -1,74 +0,0 @@
-! This test checks lowering of OpenMP Threadprivate Directive.
-! Test for threadprivate variable in use association.
-
-!RUN: %flang_fc1 -emit-fir -flang-deprecated-no-hlfir -fopenmp %s -o - | FileCheck %s
-
-!CHECK-DAG: fir.global common @blk_(dense<0> : vector<24xi8>) : !fir.array<24xi8>
-!CHECK-DAG: fir.global @_QMtestEy : f32 {
-
-module test
-  integer :: x
-  real :: y, z(5)
-  common /blk/ x, z
-
-  !$omp threadprivate(y, /blk/)
-
-contains
-  subroutine sub()
-! CHECK-LABEL: @_QMtestPsub
-!CHECK-DAG:   [[ADDR0:%.*]] = fir.address_of(@blk_) : !fir.ref<!fir.array<24xi8>>
-!CHECK-DAG:   [[NEWADDR0:%.*]] = omp.threadprivate [[ADDR0]] : !fir.ref<!fir.array<24xi8>> -> !fir.ref<!fir.array<24xi8>>
-!CHECK-DAG:   [[ADDR1:%.*]] = fir.address_of(@_QMtestEy) : !fir.ref<f32>
-!CHECK-DAG:   [[NEWADDR1:%.*]] = omp.threadprivate [[ADDR1]] : !fir.ref<f32> -> !fir.ref<f32>
-
-    !$omp parallel
-!CHECK-DAG:    [[ADDR2:%.*]] = omp.threadprivate [[ADDR0]] : !fir.ref<!fir.array<24xi8>> -> !fir.ref<!fir.array<24xi8>>
-!CHECK-DAG:    [[ADDR3:%.*]] = omp.threadprivate [[ADDR1]] : !fir.ref<f32> -> !fir.ref<f32>
-!CHECK-DAG:    [[ADDR4:%.*]] = fir.convert [[ADDR2]] : (!fir.ref<!fir.array<24xi8>>) -> !fir.ref<!fir.array<?xi8>>
-!CHECK-DAG:    [[ADDR5:%.*]] = fir.coordinate_of [[ADDR4]], %{{.*}} : (!fir.ref<!fir.array<?xi8>>, index) -> !fir.ref<i8>
-!CHECK-DAG:    [[ADDR6:%.*]] = fir.convert [[ADDR5:%.*]] : (!fir.ref<i8>) -> !fir.ref<i32>
-!CHECK-DAG:    [[ADDR7:%.*]] = fir.convert [[ADDR2]] : (!fir.ref<!fir.array<24xi8>>) -> !fir.ref<!fir.array<?xi8>>
-!CHECK-DAG:    [[ADDR8:%.*]] = fir.coordinate_of [[ADDR7]], %{{.*}} : (!fir.ref<!fir.array<?xi8>>, index) -> !fir.ref<i8>
-!CHECK-DAG:    [[ADDR9:%.*]] = fir.convert [[ADDR8:%.*]] : (!fir.ref<i8>) -> !fir.ref<!fir.array<5xf32>>
-!CHECK-DAG:    %{{.*}} = fir.load [[ADDR6]] : !fir.ref<i32>
-!CHECK-DAG:    %{{.*}} = fir.load [[ADDR3]] : !fir.ref<f32>
-!CHECK-DAG:    %{{.*}} = fir.embox [[ADDR9]](%{{.*}}) : (!fir.ref<!fir.array<5xf32>>, !fir.shape<1>) -> !fir.box<!fir.array<5xf32>>
-      print *, x, y, z
-    !$omp end parallel
-  end
-end
-
-program main
-  use test
-  integer :: x1
-  real :: z1(5)
-  common /blk/ x1, z1
-
-  !$omp threadprivate(/blk/)
-
-  call sub()
-
-! CHECK-LABEL: @_QQmain()
-!CHECK-DAG:  [[ADDR0:%.*]] = fir.address_of(@blk_) : !fir.ref<!fir.array<24xi8>>
-!CHECK-DAG:  [[NEWADDR0:%.*]] = omp.threadprivate [[ADDR0]] : !fir.ref<!fir.array<24xi8>> -> !fir.ref<!fir.array<24xi8>>
-!CHECK-DAG:  [[ADDR1:%.*]] = fir.address_of(@blk_) : !fir.ref<!fir.array<24xi8>>
-!CHECK-DAG:  [[NEWADDR1:%.*]] = omp.threadprivate [[ADDR1]] : !fir.ref<!fir.array<24xi8>> -> !fir.ref<!fir.array<24xi8>>
-!CHECK-DAG:  [[ADDR2:%.*]] = fir.address_of(@_QMtestEy) : !fir.ref<f32>
-!CHECK-DAG:  [[NEWADDR2:%.*]] = omp.threadprivate [[ADDR2]] : !fir.ref<f32> -> !fir.ref<f32>
-
-  !$omp parallel
-!CHECK-DAG:    [[ADDR4:%.*]] = omp.threadprivate [[ADDR1]] : !fir.ref<!fir.array<24xi8>> -> !fir.ref<!fir.array<24xi8>>
-!CHECK-DAG:    [[ADDR5:%.*]] = omp.threadprivate [[ADDR2]] : !fir.ref<f32> -> !fir.ref<f32>
-!CHECK-DAG:    [[ADDR6:%.*]] = fir.convert [[ADDR4]] : (!fir.ref<!fir.array<24xi8>>) -> !fir.ref<!fir.array<?xi8>>
-!CHECK-DAG:    [[ADDR7:%.*]] = fir.coordinate_of [[ADDR6]], %{{.*}} : (!fir.ref<!fir.array<?xi8>>, index) -> !fir.ref<i8>
-!CHECK-DAG:    [[ADDR8:%.*]] = fir.convert [[ADDR7:%.*]] : (!fir.ref<i8>) -> !fir.ref<i32>
-!CHECK-DAG:    [[ADDR9:%.*]] = fir.convert [[ADDR4]] : (!fir.ref<!fir.array<24xi8>>) -> !fir.ref<!fir.array<?xi8>>
-!CHECK-DAG:    [[ADDR10:%.*]] = fir.coordinate_of [[ADDR9]], %{{.*}} : (!fir.ref<!fir.array<?xi8>>, index) -> !fir.ref<i8>
-!CHECK-DAG:    [[ADDR11:%.*]] = fir.convert [[ADDR10:%.*]] : (!fir.ref<i8>) -> !fir.ref<!fir.array<5xf32>>
-!CHECK-DAG:    %{{.*}} = fir.load [[ADDR8]] : !fir.ref<i32>
-!CHECK-DAG:    %{{.*}} = fir.load [[ADDR5]] : !fir.ref<f32>
-!CHECK-DAG:    %{{.*}} = fir.embox [[ADDR11]](%{{.*}}) : (!fir.ref<!fir.array<5xf32>>, !fir.shape<1>) -> !fir.box<!fir.array<5xf32>>
-    print *, x1, y, z1
-  !$omp end parallel
-
-end
diff --git a/flang/test/Lower/OpenMP/FIR/unstructured.f90 b/flang/test/Lower/OpenMP/FIR/unstructured.f90
deleted file mode 100644
index bfaf38b7ef1a..000000000000
--- a/flang/test/Lower/OpenMP/FIR/unstructured.f90
+++ /dev/null
@@ -1,344 +0,0 @@
-! Test unstructured code adjacent to and inside OpenMP constructs.
-
-! RUN: bbc %s -fopenmp -emit-fir -hlfir=false -o "-" | FileCheck %s
-
-! CHECK-LABEL: func @_QPss1{{.*}} {
-! CHECK:   br ^bb1
-! CHECK: ^bb1:  // 2 preds: ^bb0, ^bb4
-! CHECK:   cond_br %{{[0-9]*}}, ^bb2, ^bb5
-! CHECK: ^bb2:  // pred: ^bb1
-! CHECK:   cond_br %{{[0-9]*}}, ^bb3, ^bb4
-! CHECK: ^bb4:  // pred: ^bb2
-! CHECK:   fir.call @_FortranAioBeginExternalListOutput
-! CHECK:   br ^bb1
-! CHECK: ^bb5:  // 2 preds: ^bb1, ^bb3
-! CHECK:   omp.master  {
-! CHECK:     @_FortranAioBeginExternalListOutput
-! CHECK:     omp.terminator
-! CHECK:   }
-! CHECK:   @_FortranAioBeginExternalListOutput
-! CHECK: }
-subroutine ss1(n) ! unstructured code followed by a structured OpenMP construct
-  do i = 1, 3
-    if (i .eq. n) exit
-    print*, 'ss1-A', i
-  enddo
-  !$omp master
-    print*, 'ss1-B', i
-  !$omp end master
-  print*
-end
-
-! CHECK-LABEL: func @_QPss2{{.*}} {
-! CHECK:   omp.master  {
-! CHECK:     @_FortranAioBeginExternalListOutput
-! CHECK:     br ^bb1
-! CHECK:   ^bb1:  // 2 preds: ^bb0, ^bb4
-! CHECK:     cond_br %{{[0-9]*}}, ^bb2, ^bb5
-! CHECK:   ^bb2:  // pred: ^bb1
-! CHECK:     cond_br %{{[0-9]*}}, ^bb3, ^bb4
-! CHECK:   ^bb3:  // pred: ^bb2
-! CHECK:     @_FortranAioBeginExternalListOutput
-! CHECK:     br ^bb1
-! CHECK:   ^bb5:  // 2 preds: ^bb1, ^bb3
-! CHECK:     omp.terminator
-! CHECK:   }
-! CHECK:   @_FortranAioBeginExternalListOutput
-! CHECK:   @_FortranAioBeginExternalListOutput
-! CHECK: }
-subroutine ss2(n) ! unstructured OpenMP construct; loop exit inside construct
-  !$omp master
-    print*, 'ss2-A', n
-    do i = 1, 3
-      if (i .eq. n) exit
-      print*, 'ss2-B', i
-    enddo
-  !$omp end master
-  print*, 'ss2-C', i
-  print*
-end
-
-! CHECK-LABEL: func @_QPss3{{.*}} {
-! CHECK:   omp.parallel {
-! CHECK:     %[[ALLOCA_K:.*]] = fir.alloca i32 {bindc_name = "k", pinned}
-! CHECK:     %[[ALLOCA_1:.*]] = fir.alloca i32 {{{.*}}, pinned}
-! CHECK:     %[[ALLOCA_2:.*]] = fir.alloca i32 {{{.*}}, pinned}
-! CHECK:     br ^bb1
-! CHECK:   ^bb1:  // 2 preds: ^bb0, ^bb3
-! CHECK:     cond_br %{{[0-9]*}}, ^bb2, ^bb4
-! CHECK:   ^bb2:  // pred: ^bb1
-! CHECK:     omp.wsloop for (%[[ARG1:.*]]) : {{.*}} {
-! CHECK:       fir.store %[[ARG1]] to %[[ALLOCA_2]] : !fir.ref<i32>
-! CHECK:     @_FortranAioBeginExternalListOutput
-! CHECK:       %[[LOAD_1:.*]] = fir.load %[[ALLOCA_2]] : !fir.ref<i32>
-! CHECK:     @_FortranAioOutputInteger32(%{{.*}}, %[[LOAD_1]])
-! CHECK:       omp.yield
-! CHECK:     }
-! CHECK:     omp.wsloop for (%[[ARG2:.*]]) : {{.*}} {
-! CHECK:       fir.store %[[ARG2]] to %[[ALLOCA_1]] : !fir.ref<i32>
-! CHECK:       br ^bb1
-! CHECK:     ^bb2:  // 2 preds: ^bb1, ^bb5
-! CHECK:       cond_br %{{[0-9]*}}, ^bb3, ^bb6
-! CHECK:     ^bb3:  // pred: ^bb2
-! CHECK:       cond_br %{{[0-9]*}}, ^bb4, ^bb5
-! CHECK:     ^bb4:  // pred: ^bb3
-! CHECK:       @_FortranAioBeginExternalListOutput
-! CHECK:       %[[LOAD_2:.*]] = fir.load %[[ALLOCA_K]] : !fir.ref<i32>
-! CHECK:     @_FortranAioOutputInteger32(%{{.*}}, %[[LOAD_2]])
-! CHECK:       br ^bb2
-! CHECK:     ^bb6:  // 2 preds: ^bb2, ^bb4
-! CHECK:       omp.yield
-! CHECK:     }
-! CHECK:     br ^bb1
-! CHECK:   ^bb4:  // pred: ^bb1
-! CHECK:     omp.terminator
-! CHECK:   }
-! CHECK: }
-subroutine ss3(n) ! nested unstructured OpenMP constructs
-  !$omp parallel
-    do i = 1, 3
-      !$omp do
-        do k = 1, 3
-          print*, 'ss3-A', k
-        enddo
-      !$omp end do
-      !$omp do
-        do j = 1, 3
-          do k = 1, 3
-            if (k .eq. n) exit
-            print*, 'ss3-B', k
-          enddo
-        enddo
-      !$omp end do
-    enddo
-  !$omp end parallel
-end
-
-! CHECK-LABEL: func @_QPss4{{.*}} {
-! CHECK:       omp.parallel {
-! CHECK:         %[[ALLOCA:.*]] = fir.alloca i32 {{{.*}}, pinned}
-! CHECK:         omp.wsloop for (%[[ARG:.*]]) : {{.*}} {
-! CHECK:           fir.store %[[ARG]] to %[[ALLOCA]] : !fir.ref<i32>
-! CHECK:           %[[COND:.*]] = arith.cmpi eq, %{{.*}}, %{{.*}}
-! CHECK:           %[[COND_XOR:.*]] = arith.xori %[[COND]], %{{.*}}
-! CHECK:          fir.if %[[COND_XOR]] {
-! CHECK:           @_FortranAioBeginExternalListOutput
-! CHECK:           %[[LOAD:.*]] = fir.load %[[ALLOCA]] : !fir.ref<i32>
-! CHECK:           @_FortranAioOutputInteger32(%{{.*}}, %[[LOAD]])
-! CHECK:          } else {
-! CHECK:          }
-! CHECK-NEXT:      omp.yield
-! CHECK-NEXT:  }
-! CHECK:       omp.terminator
-! CHECK-NEXT:}
-subroutine ss4(n) ! CYCLE in OpenMP wsloop constructs
-  !$omp parallel
-    do i = 1, 3
-      !$omp do
-        do j = 1, 3
-           if (j .eq. n) cycle
-           print*, 'ss4', j
-        enddo
-      !$omp end do
-    enddo
-  !$omp end parallel
-end
-
-! CHECK-LABEL: func @_QPss5() {
-! CHECK:  omp.parallel  {
-! CHECK:    omp.wsloop {{.*}} {
-! CHECK:      br ^[[BB1:.*]]
-! CHECK:    ^[[BB1]]:
-! CHECK:      br ^[[BB2:.*]]
-! CHECK:    ^[[BB2]]:
-! CHECK:      cond_br %{{.*}}, ^[[BB3:.*]], ^[[BB6:.*]]
-! CHECK:    ^[[BB3]]:
-! CHECK:      cond_br %{{.*}}, ^[[BB4:.*]], ^[[BB3:.*]]
-! CHECK:    ^[[BB4]]:
-! CHECK:      br ^[[BB6]]
-! CHECK:    ^[[BB3]]:
-! CHECK:      br ^[[BB2]]
-! CHECK:    ^[[BB6]]:
-! CHECK:      omp.yield
-! CHECK:    }
-! CHECK:    omp.terminator
-! CHECK:  }
-subroutine ss5() ! EXIT inside OpenMP wsloop (inside parallel)
-  integer :: x
-  !$omp parallel private(x)
-    !$omp do
-      do j = 1, 3
-        x = j * i
-        do k = 1, 3
-          if (k .eq. n) exit
-          x = k
-          x = x + k
-        enddo
-        x = j - 222
-      enddo
-    !$omp end do
-  !$omp end parallel
-end
-
-! CHECK-LABEL: func @_QPss6() {
-! CHECK:  omp.parallel  {
-! CHECK:    br ^[[BB1_OUTER:.*]]
-! CHECK:  ^[[BB1_OUTER]]:
-! CHECK:    cond_br %{{.*}}, ^[[BB2_OUTER:.*]], ^[[BB3_OUTER:.*]]
-! CHECK:  ^[[BB2_OUTER]]:
-! CHECK:    omp.wsloop {{.*}} {
-! CHECK:      br ^[[BB1:.*]]
-! CHECK:    ^[[BB1]]:
-! CHECK:      br ^[[BB2:.*]]
-! CHECK:    ^[[BB2]]:
-! CHECK:      cond_br %{{.*}}, ^[[BB3:.*]], ^[[BB6:.*]]
-! CHECK:    ^[[BB3]]:
-! CHECK:      cond_br %{{.*}}, ^[[BB4:.*]], ^[[BB5:.*]]
-! CHECK:    ^[[BB4]]:
-! CHECK:      br ^[[BB6]]
-! CHECK:    ^[[BB5]]
-! CHECK:      br ^[[BB2]]
-! CHECK:    ^[[BB6]]:
-! CHECK:      omp.yield
-! CHECK:    }
-! CHECK:    br ^[[BB1_OUTER]]
-! CHECK:  ^[[BB3_OUTER]]:
-! CHECK:    omp.terminator
-! CHECK:  }
-subroutine ss6() ! EXIT inside OpenMP wsloop in a do loop (inside parallel)
-  integer :: x
-  !$omp parallel private(x)
-    do i = 1, 3
-      !$omp do
-        do j = 1, 3
-          x = j * i
-          do k = 1, 3
-            if (k .eq. n) exit
-            x = k
-            x = x + k
-          enddo
-          x = j - 222
-        enddo
-      !$omp end do
-    enddo
-  !$omp end parallel
-end
-
-! CHECK-LABEL: func @_QPss7() {
-! CHECK: br ^[[BB1_OUTER:.*]]
-! CHECK: ^[[BB1_OUTER]]:
-! CHECK:   cond_br %{{.*}}, ^[[BB2_OUTER:.*]], ^[[BB3_OUTER:.*]]
-! CHECK-NEXT: ^[[BB2_OUTER:.*]]:
-! CHECK:   omp.parallel  {
-! CHECK:     omp.wsloop {{.*}} {
-! CHECK:       br ^[[BB1:.*]]
-! CHECK-NEXT:     ^[[BB1]]:
-! CHECK:       br ^[[BB2:.*]]
-! CHECK-NEXT:     ^[[BB2]]:
-! CHECK:       cond_br %{{.*}}, ^[[BB3:.*]], ^[[BB6:.*]]
-! CHECK-NEXT:     ^[[BB3]]:
-! CHECK:       cond_br %{{.*}}, ^[[BB4:.*]], ^[[BB5:.*]]
-! CHECK-NEXT:     ^[[BB4]]:
-! CHECK:       br ^[[BB6]]
-! CHECK-NEXT:     ^[[BB5]]:
-! CHECK:       br ^[[BB2]]
-! CHECK-NEXT:     ^[[BB6]]:
-! CHECK:       omp.yield
-! CHECK:     }
-! CHECK:     omp.terminator
-! CHECK:   }
-! CHECK:   br ^[[BB1_OUTER]]
-! CHECK-NEXT: ^[[BB3_OUTER]]:
-! CHECK-NEXT:   return
-subroutine ss7() ! EXIT inside OpenMP parallel do (inside do loop)
-  integer :: x
-    do i = 1, 3
-      !$omp parallel do private(x)
-        do j = 1, 3
-          x = j * i
-          do k = 1, 3
-            if (k .eq. n) exit
-            x = k
-            x = x + k
-          enddo
-        enddo
-      !$omp end parallel do
-    enddo
-end
-
-! CHECK-LABEL: func @_QPss8() {
-! CHECK:  omp.parallel  {
-! CHECK:    omp.wsloop {{.*}} {
-! CHECK:      br ^[[BB1:.*]]
-! CHECK-NEXT:    ^[[BB1]]:
-! CHECK:      br ^[[BB2:.*]]
-! CHECK:    ^[[BB2]]:
-! CHECK:      cond_br %{{.*}}, ^[[BB3:.*]], ^[[BB6:.*]]
-! CHECK:    ^[[BB3]]:
-! CHECK:      cond_br %{{.*}}, ^[[BB4:.*]], ^[[BB5:.*]]
-! CHECK:    ^[[BB4]]:
-! CHECK-NEXT:    br ^[[BB6]]
-! CHECK:    ^[[BB5]]:
-! CHECK:      br ^[[BB2]]
-! CHECK-NEXT:    ^[[BB6]]:
-! CHECK:      omp.yield
-! CHECK:    }
-! CHECK:    omp.terminator
-! CHECK:  }
-subroutine ss8() ! EXIT inside OpenMP parallel do
-  integer :: x
-      !$omp parallel do private(x)
-        do j = 1, 3
-          x = j * i
-          do k = 1, 3
-            if (k .eq. n) exit
-            x = k
-            x = x + k
-          enddo
-        enddo
-      !$omp end parallel do
-end
-
-! CHECK-LABEL: func @_QPss9() {
-! CHECK:  omp.parallel  {
-! CHECK-NEXT:    omp.parallel  {
-! CHECK:      br ^[[BB1:.*]]
-! CHECK:         ^[[BB1]]:
-! CHECK:      cond_br %{{.*}}, ^[[BB2:.*]], ^[[BB5:.*]]
-! CHECK-NEXT:    ^[[BB2]]:
-! CHECK:      cond_br %{{.*}}, ^[[BB3:.*]], ^[[BB4:.*]]
-! CHECK-NEXT:    ^[[BB3]]:
-! CHECK-NEXT:    br ^[[BB5]]
-! CHECK-NEXT:    ^[[BB4]]:
-! CHECK:      br ^[[BB1]]
-! CHECK-NEXT:    ^[[BB5]]:
-! CHECK:      omp.terminator
-! CHECK-NEXT:    }
-! CHECK:    omp.terminator
-! CHECK-NEXT  }
-! CHECK: }
-subroutine ss9() ! EXIT inside OpenMP parallel (inside parallel)
-  integer :: x
-  !$omp parallel
-  !$omp parallel private(x)
-    do k = 1, 3
-      if (k .eq. n) exit
-      x = k
-      x = x + k
-    end do
-  !$omp end parallel
-  !$omp end parallel
-end
-
-! CHECK-LABEL: func @_QQmain
-program p
-  call ss1(2)
-  call ss2(2)
-  call ss3(2)
-  call ss4(2)
-  call ss5()
-  call ss6()
-  call ss7()
-  call ss8()
-  call ss9()
-end
diff --git a/flang/test/Lower/OpenMP/FIR/wsloop-chunks.f90 b/flang/test/Lower/OpenMP/FIR/wsloop-chunks.f90
deleted file mode 100644
index 4030f46299d0..000000000000
--- a/flang/test/Lower/OpenMP/FIR/wsloop-chunks.f90
+++ /dev/null
@@ -1,75 +0,0 @@
-! This test checks that chunk size is passed correctly when lowering of
-! OpenMP DO Directive(Worksharing) with chunk size
-
-! RUN: bbc -fopenmp -emit-fir -hlfir=false %s -o - | FileCheck %s
-
-program wsloop
-        integer :: i
-        integer :: chunk
-
-! CHECK-LABEL: func.func @_QQmain() attributes {fir.bindc_name = "wsloop"} {
-! CHECK:         %[[VAL_0:.*]] = fir.alloca i32 {bindc_name = "chunk", uniq_name = "_QFEchunk"}
-
-!$OMP DO SCHEDULE(static, 4)
-
-do i=1, 9
-  print*, i
-
-! CHECK:         %[[VAL_2:.*]] = arith.constant 1 : i32
-! CHECK:         %[[VAL_3:.*]] = arith.constant 9 : i32
-! CHECK:         %[[VAL_4:.*]] = arith.constant 1 : i32
-! CHECK:         %[[VAL_5:.*]] = arith.constant 4 : i32
-! CHECK:         omp.wsloop   schedule(static = %[[VAL_5]] : i32) nowait for  (%[[ARG0:.*]]) : i32 = (%[[VAL_2]]) to (%[[VAL_3]]) inclusive step (%[[VAL_4]]) {
-! CHECK:           fir.store %[[ARG0]] to %[[STORE_IV:.*]] : !fir.ref<i32>
-! CHECK:           %[[LOAD_IV:.*]] = fir.load %[[STORE_IV]] : !fir.ref<i32>
-! CHECK:           {{.*}} = fir.call @_FortranAioOutputInteger32({{.*}}, %[[LOAD_IV]]) {{.*}}: (!fir.ref<i8>, i32) -> i1
-! CHECK:           omp.yield
-! CHECK:         }
-
-end do
-!$OMP END DO NOWAIT
-!$OMP DO SCHEDULE(static, 2+2)
-
-do i=1, 9
-  print*, i*2
-
-! CHECK:         %[[VAL_14:.*]] = arith.constant 1 : i32
-! CHECK:         %[[VAL_15:.*]] = arith.constant 9 : i32
-! CHECK:         %[[VAL_16:.*]] = arith.constant 1 : i32
-! CHECK:         %[[VAL_17:.*]] = arith.constant 4 : i32
-! CHECK:         omp.wsloop   schedule(static = %[[VAL_17]] : i32) nowait for  (%[[ARG1:.*]]) : i32 = (%[[VAL_14]]) to (%[[VAL_15]]) inclusive step (%[[VAL_16]]) {
-! CHECK:           fir.store %[[ARG1]] to %[[STORE_IV1:.*]] : !fir.ref<i32>
-! CHECK:           %[[VAL_24:.*]] = arith.constant 2 : i32
-! CHECK:           %[[LOAD_IV1:.*]] = fir.load %[[STORE_IV1]] : !fir.ref<i32>
-! CHECK:           %[[VAL_25:.*]] = arith.muli %[[VAL_24]], %[[LOAD_IV1]] : i32
-! CHECK:           {{.*}} = fir.call @_FortranAioOutputInteger32({{.*}}, %[[VAL_25]]) {{.*}}: (!fir.ref<i8>, i32) -> i1
-! CHECK:           omp.yield
-! CHECK:         }
-  
-end do
-!$OMP END DO NOWAIT
-chunk = 6
-!$OMP DO SCHEDULE(static, chunk)
-
-do i=1, 9
-   print*, i*3
-end do
-!$OMP END DO NOWAIT
-! CHECK:         %[[VAL_28:.*]] = arith.constant 6 : i32
-! CHECK:         fir.store %[[VAL_28]] to %[[VAL_0]] : !fir.ref<i32>
-! CHECK:         %[[VAL_29:.*]] = arith.constant 1 : i32
-! CHECK:         %[[VAL_30:.*]] = arith.constant 9 : i32
-! CHECK:         %[[VAL_31:.*]] = arith.constant 1 : i32
-! CHECK:         %[[VAL_32:.*]] = fir.load %[[VAL_0]] : !fir.ref<i32>
-! CHECK:         omp.wsloop   schedule(static = %[[VAL_32]] : i32) nowait for  (%[[ARG2:.*]]) : i32 = (%[[VAL_29]]) to (%[[VAL_30]]) inclusive step (%[[VAL_31]]) {
-! CHECK:           fir.store %[[ARG2]] to %[[STORE_IV2:.*]] : !fir.ref<i32>
-! CHECK:           %[[VAL_39:.*]] = arith.constant 3 : i32
-! CHECK:           %[[LOAD_IV2:.*]] = fir.load %[[STORE_IV2]] : !fir.ref<i32>
-! CHECK:           %[[VAL_40:.*]] = arith.muli %[[VAL_39]], %[[LOAD_IV2]] : i32
-! CHECK:           {{.*}} = fir.call @_FortranAioOutputInteger32({{.*}}, %[[VAL_40]]) {{.*}}: (!fir.ref<i8>, i32) -> i1
-! CHECK:           omp.yield
-! CHECK:         }
-! CHECK:         return
-! CHECK:       }
-
-end
diff --git a/flang/test/Lower/OpenMP/FIR/wsloop-collapse.f90 b/flang/test/Lower/OpenMP/FIR/wsloop-collapse.f90
deleted file mode 100644
index 933fc0910e33..000000000000
--- a/flang/test/Lower/OpenMP/FIR/wsloop-collapse.f90
+++ /dev/null
@@ -1,63 +0,0 @@
-! This test checks lowering of OpenMP DO Directive(Worksharing) with collapse.
-
-! RUN: bbc -fopenmp -emit-fir -hlfir=false %s -o - | FileCheck %s
-
-program wsloop_collapse
-  integer :: i, j, k
-  integer :: a, b, c
-  integer :: x
-! CHECK:         %[[VAL_0:.*]] = fir.alloca i32 {bindc_name = "a", uniq_name = "_QFEa"}
-! CHECK:         %[[VAL_1:.*]] = fir.alloca i32 {bindc_name = "b", uniq_name = "_QFEb"}
-! CHECK:         %[[VAL_2:.*]] = fir.alloca i32 {bindc_name = "c", uniq_name = "_QFEc"}
-! CHECK:         %[[VAL_3:.*]] = fir.alloca i32 {bindc_name = "i", uniq_name = "_QFEi"}
-! CHECK:         %[[VAL_4:.*]] = fir.alloca i32 {bindc_name = "j", uniq_name = "_QFEj"}
-! CHECK:         %[[VAL_5:.*]] = fir.alloca i32 {bindc_name = "k", uniq_name = "_QFEk"}
-! CHECK:         %[[VAL_6:.*]] = fir.alloca i32 {bindc_name = "x", uniq_name = "_QFEx"}
-  a=3
-! CHECK:         %[[VAL_7:.*]] = arith.constant 3 : i32
-! CHECK:         fir.store %[[VAL_7]] to %[[VAL_0]] : !fir.ref<i32>
-  b=2
-! CHECK:         %[[VAL_8:.*]] = arith.constant 2 : i32
-! CHECK:         fir.store %[[VAL_8]] to %[[VAL_1]] : !fir.ref<i32>
-  c=5
-! CHECK:         %[[VAL_9:.*]] = arith.constant 5 : i32
-! CHECK:         fir.store %[[VAL_9]] to %[[VAL_2]] : !fir.ref<i32>
-  x=0
-! CHECK:         %[[VAL_10:.*]] = arith.constant 0 : i32
-! CHECK:         fir.store %[[VAL_10]] to %[[VAL_6]] : !fir.ref<i32>
-
-  !$omp do collapse(3)
-! CHECK:           %[[VAL_20:.*]] = arith.constant 1 : i32
-! CHECK:           %[[VAL_21:.*]] = fir.load %[[VAL_0]] : !fir.ref<i32>
-! CHECK:           %[[VAL_22:.*]] = arith.constant 1 : i32
-! CHECK:           %[[VAL_23:.*]] = arith.constant 1 : i32
-! CHECK:           %[[VAL_24:.*]] = fir.load %[[VAL_1]] : !fir.ref<i32>
-! CHECK:           %[[VAL_25:.*]] = arith.constant 1 : i32
-! CHECK:           %[[VAL_26:.*]] = arith.constant 1 : i32
-! CHECK:           %[[VAL_27:.*]] = fir.load %[[VAL_2]] : !fir.ref<i32>
-! CHECK:           %[[VAL_28:.*]] = arith.constant 1 : i32
-  do i = 1, a
-     do j= 1, b
-        do k = 1, c
-! CHECK:           omp.wsloop for (%[[ARG0:.*]], %[[ARG1:.*]], %[[ARG2:.*]]) : i32 = (%[[VAL_20]], %[[VAL_23]], %[[VAL_26]]) to (%[[VAL_21]], %[[VAL_24]], %[[VAL_27]]) inclusive step (%[[VAL_22]], %[[VAL_25]], %[[VAL_28]]) {
-! CHECK:             fir.store %[[ARG0]] to %[[STORE_IV0:.*]] : !fir.ref<i32>
-! CHECK:             fir.store %[[ARG1]] to %[[STORE_IV1:.*]] : !fir.ref<i32>
-! CHECK:             fir.store %[[ARG2]] to %[[STORE_IV2:.*]] : !fir.ref<i32>
-! CHECK:             %[[VAL_12:.*]] = fir.load %[[VAL_6]] : !fir.ref<i32>
-! CHECK:             %[[LOAD_IV0:.*]] = fir.load %[[STORE_IV0]] : !fir.ref<i32>
-! CHECK:             %[[VAL_13:.*]] = arith.addi %[[VAL_12]], %[[LOAD_IV0]] : i32
-! CHECK:             %[[LOAD_IV1:.*]] = fir.load %[[STORE_IV1]] : !fir.ref<i32>
-! CHECK:             %[[VAL_14:.*]] = arith.addi %[[VAL_13]], %[[LOAD_IV1]] : i32
-! CHECK:             %[[LOAD_IV2:.*]] = fir.load %[[STORE_IV2]] : !fir.ref<i32>
-! CHECK:             %[[VAL_15:.*]] = arith.addi %[[VAL_14]], %[[LOAD_IV2]] : i32
-! CHECK:             fir.store %[[VAL_15]] to %[[VAL_6]] : !fir.ref<i32>
-! CHECK:             omp.yield
-! CHECK:           }
-           x = x + i + j + k
-        end do
-     end do
-  end do
-  !$omp end do
-! CHECK:         return
-! CHECK:       }
-end program wsloop_collapse
diff --git a/flang/test/Lower/OpenMP/FIR/wsloop-monotonic.f90 b/flang/test/Lower/OpenMP/FIR/wsloop-monotonic.f90
deleted file mode 100644
index 1c381475f6cb..000000000000
--- a/flang/test/Lower/OpenMP/FIR/wsloop-monotonic.f90
+++ /dev/null
@@ -1,34 +0,0 @@
-! This test checks lowering of OpenMP DO Directive (Worksharing) with
-! monotonic schedule modifier.
-
-! RUN: bbc -fopenmp -emit-fir -hlfir=false %s -o - | FileCheck %s
-
-program wsloop_dynamic
-  integer :: i
-!CHECK-LABEL: func @_QQmain()
-
-!$OMP PARALLEL
-!CHECK:  omp.parallel {
-
-!$OMP DO SCHEDULE(monotonic:dynamic)
-!CHECK:     %[[ALLOCA_IV:.*]] = fir.alloca i32 {{{.*}}, pinned}
-!CHECK:     %[[WS_LB:.*]] = arith.constant 1 : i32
-!CHECK:     %[[WS_UB:.*]] = arith.constant 9 : i32
-!CHECK:     %[[WS_STEP:.*]] = arith.constant 1 : i32
-!CHECK:     omp.wsloop schedule(dynamic, monotonic) nowait for (%[[I:.*]]) : i32 = (%[[WS_LB]]) to (%[[WS_UB]]) inclusive step (%[[WS_STEP]])
-!CHECK:       fir.store %[[I]] to %[[ALLOCA_IV:.*]] : !fir.ref<i32>
-
-  do i=1, 9
-    print*, i
-!CHECK:    %[[RTBEGIN:.*]] = fir.call @_FortranAioBeginExternalListOutput
-!CHECK:    %[[LOAD:.*]] = fir.load %[[ALLOCA_IV]] : !fir.ref<i32>
-!CHECK:    fir.call @_FortranAioOutputInteger32(%[[RTBEGIN]], %[[LOAD]]) {{.*}}: (!fir.ref<i8>, i32) -> i1
-!CHECK:    fir.call @_FortranAioEndIoStatement(%[[RTBEGIN]]) {{.*}}: (!fir.ref<i8>) -> i32
-  end do
-!CHECK:       omp.yield
-!CHECK:       omp.terminator
-!CHECK:     }
-
-!$OMP END DO NOWAIT
-!$OMP END PARALLEL
-end
diff --git a/flang/test/Lower/OpenMP/FIR/wsloop-nonmonotonic.f90 b/flang/test/Lower/OpenMP/FIR/wsloop-nonmonotonic.f90
deleted file mode 100644
index 3f425200b8fa..000000000000
--- a/flang/test/Lower/OpenMP/FIR/wsloop-nonmonotonic.f90
+++ /dev/null
@@ -1,36 +0,0 @@
-! This test checks lowering of OpenMP DO Directive(Worksharing) with
-! non-monotonic schedule modifier.
-
-! RUN: bbc -fopenmp -emit-fir -hlfir=false %s -o - | FileCheck %s
-
-program wsloop_dynamic
-  integer :: i
-!CHECK-LABEL: func @_QQmain()
-
-
-!$OMP PARALLEL
-!CHECK:  omp.parallel {
-
-!$OMP DO SCHEDULE(nonmonotonic:dynamic)
-!CHECK:     %[[ALLOCA_IV:.*]] = fir.alloca i32 {{{.*}}, pinned}
-!CHECK:     %[[WS_LB:.*]] = arith.constant 1 : i32
-!CHECK:     %[[WS_UB:.*]] = arith.constant 9 : i32
-!CHECK:     %[[WS_STEP:.*]] = arith.constant 1 : i32
-!CHECK:     omp.wsloop schedule(dynamic, nonmonotonic) nowait for (%[[I:.*]]) : i32 = (%[[WS_LB]]) to (%[[WS_UB]]) inclusive step (%[[WS_STEP]])
-!CHECK:       fir.store %[[I]] to %[[ALLOCA_IV]] : !fir.ref<i32>
-
-  do i=1, 9
-    print*, i
-!CHECK:    %[[RTBEGIN:.*]] = fir.call @_FortranAioBeginExternalListOutput
-!CHECK:    %[[LOAD:.*]] = fir.load %[[ALLOCA_IV]] : !fir.ref<i32>
-!CHECK:    fir.call @_FortranAioOutputInteger32(%[[RTBEGIN]], %[[LOAD]]) {{.*}}: (!fir.ref<i8>, i32) -> i1
-!CHECK:    fir.call @_FortranAioEndIoStatement(%[[RTBEGIN]]) {{.*}}: (!fir.ref<i8>) -> i32
-  end do
-!CHECK:       omp.yield
-!CHECK:         }
-!CHECK:       omp.terminator
-!CHECK:     }
-
-!$OMP END DO NOWAIT
-!$OMP END PARALLEL
-end
diff --git a/flang/test/Lower/OpenMP/FIR/wsloop-ordered.f90 b/flang/test/Lower/OpenMP/FIR/wsloop-ordered.f90
deleted file mode 100644
index 7548d7a59722..000000000000
--- a/flang/test/Lower/OpenMP/FIR/wsloop-ordered.f90
+++ /dev/null
@@ -1,40 +0,0 @@
-! This test checks lowering of worksharing-loop construct with ordered clause.
-
-! RUN: bbc -fopenmp -emit-fir %s -o - | FileCheck %s
-
-! This checks lowering ordered clause specified without parameter
-subroutine wsloop_ordered_no_para()
-  integer :: a(10), i
-
-! CHECK:  omp.wsloop ordered(0) for (%{{.*}}) : i32 = (%{{.*}}) to (%{{.*}}) inclusive step (%{{.*}}) {
-! CHECK:    omp.yield
-! CHECK:  }
-
-  !$omp do ordered
-  do i = 2, 10
-    !$omp ordered
-    a(i) = a(i-1) + 1
-    !$omp end ordered
-  end do
-  !$omp end do
-
-end
-
-! This checks lowering ordered clause specified with a parameter
-subroutine wsloop_ordered_with_para()
-  integer :: a(10), i
-
-! CHECK: func @_QPwsloop_ordered_with_para() {
-! CHECK:  omp.wsloop ordered(1) for (%{{.*}}) : i32 = (%{{.*}}) to (%{{.*}}) inclusive step (%{{.*}}) {
-! CHECK:    omp.yield
-! CHECK:  }
-
-  !$omp do ordered(1)
-  do i = 2, 10
-    !!$omp ordered depend(sink: i-1)
-    a(i) = a(i-1) + 1
-    !!$omp ordered depend(source)
-  end do
-  !$omp end do
-
-end
diff --git a/flang/test/Lower/OpenMP/FIR/wsloop-reduction-add-byref.f90 b/flang/test/Lower/OpenMP/FIR/wsloop-reduction-add-byref.f90
deleted file mode 100644
index 08f5a0fcdbae..000000000000
--- a/flang/test/Lower/OpenMP/FIR/wsloop-reduction-add-byref.f90
+++ /dev/null
@@ -1,392 +0,0 @@
-! RUN: bbc -emit-fir -hlfir=false -fopenmp --force-byref-reduction %s -o - | FileCheck %s
-! RUN: %flang_fc1 -emit-fir -flang-deprecated-no-hlfir -fopenmp -mmlir --force-byref-reduction %s -o - | FileCheck %s
-! NOTE: Assertions have been autogenerated by utils/generate-test-checks.py
-
-! CHECK-LABEL:   omp.declare_reduction @add_reduction_byref_f64 : !fir.ref<f64>
-! CHECK-SAME:    init {
-! CHECK:         ^bb0(%[[VAL_0:.*]]: !fir.ref<f64>):
-! CHECK:            %[[C0_1:.*]] = arith.constant 0.000000e+00 : f64
-! CHECK:            %[[REF:.*]] = fir.alloca f64
-! CHECK:            fir.store %[[C0_1]] to %[[REF]] : !fir.ref<f64>
-! CHECK:           omp.yield(%[[REF]] : !fir.ref<f64>)
-
-! CHECK-LABEL:   } combiner {
-! CHECK:         ^bb0(%[[ARG0:.*]]: !fir.ref<f64>, %[[ARG1:.*]]: !fir.ref<f64>):
-! CHECK:           %[[LD0:.*]] = fir.load %[[ARG0]] : !fir.ref<f64>
-! CHECK:           %[[LD1:.*]] = fir.load %[[ARG1]] : !fir.ref<f64>
-! CHECK:           %[[RES:.*]] = arith.addf %[[LD0]], %[[LD1]] fastmath<contract> : f64
-! CHECK:           fir.store %[[RES]] to %[[ARG0]] : !fir.ref<f64>
-! CHECK:           omp.yield(%[[ARG0]] : !fir.ref<f64>)
-! CHECK:         }
-
-! CHECK-LABEL:   omp.declare_reduction @add_reduction_byref_i64 : !fir.ref<i64>
-! CHECK-SAME:    init {
-! CHECK:         ^bb0(%[[VAL_0:.*]]: !fir.ref<i64>):
-! CHECK:            %[[C0_1:.*]] = arith.constant 0 : i64
-! CHECK:            %[[REF:.*]] = fir.alloca i64
-! CHECK:            fir.store %[[C0_1]] to %[[REF]] : !fir.ref<i64>
-! CHECK:           omp.yield(%[[REF]] : !fir.ref<i64>)
-
-! CHECK-LABEL:   } combiner {
-! CHECK:         ^bb0(%[[ARG0:.*]]: !fir.ref<i64>, %[[ARG1:.*]]: !fir.ref<i64>):
-! CHECK:           %[[LD0:.*]] = fir.load %[[ARG0]] : !fir.ref<i64>
-! CHECK:           %[[LD1:.*]] = fir.load %[[ARG1]] : !fir.ref<i64>
-! CHECK:           %[[RES:.*]] = arith.addi %[[LD0]], %[[LD1]] : i64
-! CHECK:           fir.store %[[RES]] to %[[ARG0]] : !fir.ref<i64>
-! CHECK:           omp.yield(%[[ARG0]] : !fir.ref<i64>)
-! CHECK:         }
-
-! CHECK-LABEL:   omp.declare_reduction @add_reduction_byref_f32 : !fir.ref<f32>
-! CHECK-SAME:    init {
-! CHECK:         ^bb0(%[[VAL_0:.*]]: !fir.ref<f32>):
-! CHECK:            %[[C0_1:.*]] = arith.constant 0.000000e+00 : f32
-! CHECK:            %[[REF:.*]] = fir.alloca f32
-! CHECK:            fir.store %[[C0_1]] to %[[REF]] : !fir.ref<f32>
-! CHECK:           omp.yield(%[[REF]] : !fir.ref<f32>)
-
-! CHECK-LABEL:   } combiner {
-! CHECK:         ^bb0(%[[ARG0:.*]]: !fir.ref<f32>, %[[ARG1:.*]]: !fir.ref<f32>):
-! CHECK:           %[[LD0:.*]] = fir.load %[[ARG0]] : !fir.ref<f32>
-! CHECK:           %[[LD1:.*]] = fir.load %[[ARG1]] : !fir.ref<f32>
-! CHECK:           %[[RES:.*]] = arith.addf %[[LD0]], %[[LD1]] fastmath<contract> : f32
-! CHECK:           fir.store %[[RES]] to %[[ARG0]] : !fir.ref<f32>
-! CHECK:           omp.yield(%[[ARG0]] : !fir.ref<f32>)
-! CHECK:         }
-
-! CHECK-LABEL:   omp.declare_reduction @add_reduction_byref_i32 : !fir.ref<i32>
-! CHECK-SAME:    init {
-! CHECK:         ^bb0(%[[VAL_0:.*]]: !fir.ref<i32>):
-! CHECK:            %[[C0_1:.*]] = arith.constant 0 : i32
-! CHECK:            %[[REF:.*]] = fir.alloca i32
-! CHECK:            fir.store %[[C0_1]] to %[[REF]] : !fir.ref<i32>
-! CHECK:           omp.yield(%[[REF]] : !fir.ref<i32>)
-
-! CHECK-LABEL:   } combiner {
-! CHECK:         ^bb0(%[[ARG0:.*]]: !fir.ref<i32>, %[[ARG1:.*]]: !fir.ref<i32>):
-! CHECK:           %[[LD0:.*]] = fir.load %[[ARG0]] : !fir.ref<i32>
-! CHECK:           %[[LD1:.*]] = fir.load %[[ARG1]] : !fir.ref<i32>
-! CHECK:           %[[RES:.*]] = arith.addi %[[LD0]], %[[LD1]] : i32
-! CHECK:           fir.store %[[RES]] to %[[ARG0]] : !fir.ref<i32>
-! CHECK:           omp.yield(%[[ARG0]] : !fir.ref<i32>)
-! CHECK:         }
-
-! CHECK-LABEL:   func.func @_QPsimple_int_reduction() {
-! CHECK:           %[[VAL_0:.*]] = fir.alloca i32 {bindc_name = "i", uniq_name = "_QFsimple_int_reductionEi"}
-! CHECK:           %[[VAL_1:.*]] = fir.alloca i32 {bindc_name = "x", uniq_name = "_QFsimple_int_reductionEx"}
-! CHECK:           %[[VAL_2:.*]] = arith.constant 0 : i32
-! CHECK:           fir.store %[[VAL_2]] to %[[VAL_1]] : !fir.ref<i32>
-! CHECK:           omp.parallel {
-! CHECK:             %[[VAL_3:.*]] = fir.alloca i32 {adapt.valuebyref, pinned}
-! CHECK:             %[[VAL_4:.*]] = arith.constant 1 : i32
-! CHECK:             %[[VAL_5:.*]] = arith.constant 100 : i32
-! CHECK:             %[[VAL_6:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop byref reduction(@add_reduction_byref_i32 %[[VAL_1]] -> %[[VAL_7:.*]] : !fir.ref<i32>)  for  (%[[VAL_8:.*]]) : i32 = (%[[VAL_4]]) to (%[[VAL_5]]) inclusive step (%[[VAL_6]]) {
-! CHECK:               fir.store %[[VAL_8]] to %[[VAL_3]] : !fir.ref<i32>
-! CHECK:               %[[VAL_9:.*]] = fir.load %[[VAL_7]] : !fir.ref<i32>
-! CHECK:               %[[VAL_10:.*]] = fir.load %[[VAL_3]] : !fir.ref<i32>
-! CHECK:               %[[VAL_11:.*]] = arith.addi %[[VAL_9]], %[[VAL_10]] : i32
-! CHECK:               fir.store %[[VAL_11]] to %[[VAL_7]] : !fir.ref<i32>
-! CHECK:               omp.yield
-! CHECK:             }
-! CHECK:             omp.terminator
-! CHECK:           }
-! CHECK:           return
-! CHECK:         }
-
-subroutine simple_int_reduction
-  integer :: x
-  x = 0
-  !$omp parallel
-  !$omp do reduction(+:x)
-  do i=1, 100
-    x = x + i
-  end do
-  !$omp end do
-  !$omp end parallel
-end subroutine
-
-
-! CHECK-LABEL:   func.func @_QPsimple_real_reduction() {
-! CHECK:           %[[VAL_0:.*]] = fir.alloca i32 {bindc_name = "i", uniq_name = "_QFsimple_real_reductionEi"}
-! CHECK:           %[[VAL_1:.*]] = fir.alloca f32 {bindc_name = "x", uniq_name = "_QFsimple_real_reductionEx"}
-! CHECK:           %[[VAL_2:.*]] = arith.constant 0.000000e+00 : f32
-! CHECK:           fir.store %[[VAL_2]] to %[[VAL_1]] : !fir.ref<f32>
-! CHECK:           omp.parallel {
-! CHECK:             %[[VAL_3:.*]] = fir.alloca i32 {adapt.valuebyref, pinned}
-! CHECK:             %[[VAL_4:.*]] = arith.constant 1 : i32
-! CHECK:             %[[VAL_5:.*]] = arith.constant 100 : i32
-! CHECK:             %[[VAL_6:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop byref reduction(@add_reduction_byref_f32 %[[VAL_1]] -> %[[VAL_7:.*]] : !fir.ref<f32>)  for  (%[[VAL_8:.*]]) : i32 = (%[[VAL_4]]) to (%[[VAL_5]]) inclusive step (%[[VAL_6]]) {
-! CHECK:               fir.store %[[VAL_8]] to %[[VAL_3]] : !fir.ref<i32>
-! CHECK:               %[[VAL_9:.*]] = fir.load %[[VAL_7]] : !fir.ref<f32>
-! CHECK:               %[[VAL_10:.*]] = fir.load %[[VAL_3]] : !fir.ref<i32>
-! CHECK:               %[[VAL_11:.*]] = fir.convert %[[VAL_10]] : (i32) -> f32
-! CHECK:               %[[VAL_12:.*]] = arith.addf %[[VAL_9]], %[[VAL_11]] fastmath<contract> : f32
-! CHECK:               fir.store %[[VAL_12]] to %[[VAL_7]] : !fir.ref<f32>
-! CHECK:               omp.yield
-! CHECK:             }
-! CHECK:             omp.terminator
-! CHECK:           }
-! CHECK:           return
-! CHECK:         }
-
-subroutine simple_real_reduction
-  real :: x
-  x = 0.0
-  !$omp parallel
-  !$omp do reduction(+:x)
-  do i=1, 100
-    x = x + i
-  end do
-  !$omp end do
-  !$omp end parallel
-end subroutine
-
-! CHECK-LABEL:   func.func @_QPsimple_int_reduction_switch_order() {
-! CHECK:           %[[VAL_0:.*]] = fir.alloca i32 {bindc_name = "i", uniq_name = "_QFsimple_int_reduction_switch_orderEi"}
-! CHECK:           %[[VAL_1:.*]] = fir.alloca i32 {bindc_name = "x", uniq_name = "_QFsimple_int_reduction_switch_orderEx"}
-! CHECK:           %[[VAL_2:.*]] = arith.constant 0 : i32
-! CHECK:           fir.store %[[VAL_2]] to %[[VAL_1]] : !fir.ref<i32>
-! CHECK:           omp.parallel {
-! CHECK:             %[[VAL_3:.*]] = fir.alloca i32 {adapt.valuebyref, pinned}
-! CHECK:             %[[VAL_4:.*]] = arith.constant 1 : i32
-! CHECK:             %[[VAL_5:.*]] = arith.constant 100 : i32
-! CHECK:             %[[VAL_6:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop byref reduction(@add_reduction_byref_i32 %[[VAL_1]] -> %[[VAL_7:.*]] : !fir.ref<i32>)  for  (%[[VAL_8:.*]]) : i32 = (%[[VAL_4]]) to (%[[VAL_5]]) inclusive step (%[[VAL_6]]) {
-! CHECK:               fir.store %[[VAL_8]] to %[[VAL_3]] : !fir.ref<i32>
-! CHECK:               %[[VAL_9:.*]] = fir.load %[[VAL_3]] : !fir.ref<i32>
-! CHECK:               %[[VAL_10:.*]] = fir.load %[[VAL_7]] : !fir.ref<i32>
-! CHECK:               %[[VAL_11:.*]] = arith.addi %[[VAL_9]], %[[VAL_10]] : i32
-! CHECK:               fir.store %[[VAL_11]] to %[[VAL_7]] : !fir.ref<i32>
-! CHECK:               omp.yield
-! CHECK:             }
-! CHECK:             omp.terminator
-! CHECK:           }
-! CHECK:           return
-! CHECK:         }
-
-subroutine simple_int_reduction_switch_order
-  integer :: x
-  x = 0
-  !$omp parallel
-  !$omp do reduction(+:x)
-  do i=1, 100
-    x = i + x
-  end do
-  !$omp end do
-  !$omp end parallel
-end subroutine
-
-! CHECK-LABEL:   func.func @_QPsimple_real_reduction_switch_order() {
-! CHECK:           %[[VAL_0:.*]] = fir.alloca i32 {bindc_name = "i", uniq_name = "_QFsimple_real_reduction_switch_orderEi"}
-! CHECK:           %[[VAL_1:.*]] = fir.alloca f32 {bindc_name = "x", uniq_name = "_QFsimple_real_reduction_switch_orderEx"}
-! CHECK:           %[[VAL_2:.*]] = arith.constant 0.000000e+00 : f32
-! CHECK:           fir.store %[[VAL_2]] to %[[VAL_1]] : !fir.ref<f32>
-! CHECK:           omp.parallel {
-! CHECK:             %[[VAL_3:.*]] = fir.alloca i32 {adapt.valuebyref, pinned}
-! CHECK:             %[[VAL_4:.*]] = arith.constant 1 : i32
-! CHECK:             %[[VAL_5:.*]] = arith.constant 100 : i32
-! CHECK:             %[[VAL_6:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop byref reduction(@add_reduction_byref_f32 %[[VAL_1]] -> %[[VAL_7:.*]] : !fir.ref<f32>)  for  (%[[VAL_8:.*]]) : i32 = (%[[VAL_4]]) to (%[[VAL_5]]) inclusive step (%[[VAL_6]]) {
-! CHECK:               fir.store %[[VAL_8]] to %[[VAL_3]] : !fir.ref<i32>
-! CHECK:               %[[VAL_9:.*]] = fir.load %[[VAL_3]] : !fir.ref<i32>
-! CHECK:               %[[VAL_10:.*]] = fir.convert %[[VAL_9]] : (i32) -> f32
-! CHECK:               %[[VAL_11:.*]] = fir.load %[[VAL_7]] : !fir.ref<f32>
-! CHECK:               %[[VAL_12:.*]] = arith.addf %[[VAL_10]], %[[VAL_11]] fastmath<contract> : f32
-! CHECK:               fir.store %[[VAL_12]] to %[[VAL_7]] : !fir.ref<f32>
-! CHECK:               omp.yield
-! CHECK:             }
-! CHECK:             omp.terminator
-! CHECK:           }
-! CHECK:           return
-! CHECK:         }
-
-subroutine simple_real_reduction_switch_order
-  real :: x
-  x = 0.0
-  !$omp parallel
-  !$omp do reduction(+:x)
-  do i=1, 100
-    x = i + x
-  end do
-  !$omp end do
-  !$omp end parallel
-end subroutine
-
-! CHECK-LABEL:   func.func @_QPmultiple_int_reductions_same_type() {
-! CHECK:           %[[VAL_0:.*]] = fir.alloca i32 {bindc_name = "i", uniq_name = "_QFmultiple_int_reductions_same_typeEi"}
-! CHECK:           %[[VAL_1:.*]] = fir.alloca i32 {bindc_name = "x", uniq_name = "_QFmultiple_int_reductions_same_typeEx"}
-! CHECK:           %[[VAL_2:.*]] = fir.alloca i32 {bindc_name = "y", uniq_name = "_QFmultiple_int_reductions_same_typeEy"}
-! CHECK:           %[[VAL_3:.*]] = fir.alloca i32 {bindc_name = "z", uniq_name = "_QFmultiple_int_reductions_same_typeEz"}
-! CHECK:           %[[VAL_4:.*]] = arith.constant 0 : i32
-! CHECK:           fir.store %[[VAL_4]] to %[[VAL_1]] : !fir.ref<i32>
-! CHECK:           %[[VAL_5:.*]] = arith.constant 0 : i32
-! CHECK:           fir.store %[[VAL_5]] to %[[VAL_2]] : !fir.ref<i32>
-! CHECK:           %[[VAL_6:.*]] = arith.constant 0 : i32
-! CHECK:           fir.store %[[VAL_6]] to %[[VAL_3]] : !fir.ref<i32>
-! CHECK:           omp.parallel {
-! CHECK:             %[[VAL_7:.*]] = fir.alloca i32 {adapt.valuebyref, pinned}
-! CHECK:             %[[VAL_8:.*]] = arith.constant 1 : i32
-! CHECK:             %[[VAL_9:.*]] = arith.constant 100 : i32
-! CHECK:             %[[VAL_10:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop byref reduction(@add_reduction_byref_i32 %[[VAL_1]] -> %[[VAL_11:.*]] : !fir.ref<i32>, @add_reduction_byref_i32 %[[VAL_2]] -> %[[VAL_12:.*]] : !fir.ref<i32>, @add_reduction_byref_i32 %[[VAL_3]] -> %[[VAL_13:.*]] : !fir.ref<i32>)  for  (%[[VAL_14:.*]]) : i32 = (%[[VAL_8]]) to (%[[VAL_9]]) inclusive step (%[[VAL_10]]) {
-! CHECK:               fir.store %[[VAL_14]] to %[[VAL_7]] : !fir.ref<i32>
-! CHECK:               %[[VAL_15:.*]] = fir.load %[[VAL_11]] : !fir.ref<i32>
-! CHECK:               %[[VAL_16:.*]] = fir.load %[[VAL_7]] : !fir.ref<i32>
-! CHECK:               %[[VAL_17:.*]] = arith.addi %[[VAL_15]], %[[VAL_16]] : i32
-! CHECK:               fir.store %[[VAL_17]] to %[[VAL_11]] : !fir.ref<i32>
-! CHECK:               %[[VAL_18:.*]] = fir.load %[[VAL_12]] : !fir.ref<i32>
-! CHECK:               %[[VAL_19:.*]] = fir.load %[[VAL_7]] : !fir.ref<i32>
-! CHECK:               %[[VAL_20:.*]] = arith.addi %[[VAL_18]], %[[VAL_19]] : i32
-! CHECK:               fir.store %[[VAL_20]] to %[[VAL_12]] : !fir.ref<i32>
-! CHECK:               %[[VAL_21:.*]] = fir.load %[[VAL_13]] : !fir.ref<i32>
-! CHECK:               %[[VAL_22:.*]] = fir.load %[[VAL_7]] : !fir.ref<i32>
-! CHECK:               %[[VAL_23:.*]] = arith.addi %[[VAL_21]], %[[VAL_22]] : i32
-! CHECK:               fir.store %[[VAL_23]] to %[[VAL_13]] : !fir.ref<i32>
-! CHECK:               omp.yield
-! CHECK:             }
-! CHECK:             omp.terminator
-! CHECK:           }
-! CHECK:           return
-! CHECK:         }
-
-subroutine multiple_int_reductions_same_type
-  integer :: x,y,z
-  x = 0
-  y = 0
-  z = 0
-  !$omp parallel
-  !$omp do reduction(+:x,y,z)
-  do i=1, 100
-    x = x + i
-    y = y + i
-    z = z + i
-  end do
-  !$omp end do
-  !$omp end parallel
-end subroutine
-
-! CHECK-LABEL:   func.func @_QPmultiple_real_reductions_same_type() {
-! CHECK:           %[[VAL_0:.*]] = fir.alloca i32 {bindc_name = "i", uniq_name = "_QFmultiple_real_reductions_same_typeEi"}
-! CHECK:           %[[VAL_1:.*]] = fir.alloca f32 {bindc_name = "x", uniq_name = "_QFmultiple_real_reductions_same_typeEx"}
-! CHECK:           %[[VAL_2:.*]] = fir.alloca f32 {bindc_name = "y", uniq_name = "_QFmultiple_real_reductions_same_typeEy"}
-! CHECK:           %[[VAL_3:.*]] = fir.alloca f32 {bindc_name = "z", uniq_name = "_QFmultiple_real_reductions_same_typeEz"}
-! CHECK:           %[[VAL_4:.*]] = arith.constant 0.000000e+00 : f32
-! CHECK:           fir.store %[[VAL_4]] to %[[VAL_1]] : !fir.ref<f32>
-! CHECK:           %[[VAL_5:.*]] = arith.constant 0.000000e+00 : f32
-! CHECK:           fir.store %[[VAL_5]] to %[[VAL_2]] : !fir.ref<f32>
-! CHECK:           %[[VAL_6:.*]] = arith.constant 0.000000e+00 : f32
-! CHECK:           fir.store %[[VAL_6]] to %[[VAL_3]] : !fir.ref<f32>
-! CHECK:           omp.parallel {
-! CHECK:             %[[VAL_7:.*]] = fir.alloca i32 {adapt.valuebyref, pinned}
-! CHECK:             %[[VAL_8:.*]] = arith.constant 1 : i32
-! CHECK:             %[[VAL_9:.*]] = arith.constant 100 : i32
-! CHECK:             %[[VAL_10:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop byref reduction(@add_reduction_byref_f32 %[[VAL_1]] -> %[[VAL_11:.*]] : !fir.ref<f32>, @add_reduction_byref_f32 %[[VAL_2]] -> %[[VAL_12:.*]] : !fir.ref<f32>, @add_reduction_byref_f32 %[[VAL_3]] -> %[[VAL_13:.*]] : !fir.ref<f32>)  for  (%[[VAL_14:.*]]) : i32 = (%[[VAL_8]]) to (%[[VAL_9]]) inclusive step (%[[VAL_10]]) {
-! CHECK:               fir.store %[[VAL_14]] to %[[VAL_7]] : !fir.ref<i32>
-! CHECK:               %[[VAL_15:.*]] = fir.load %[[VAL_11]] : !fir.ref<f32>
-! CHECK:               %[[VAL_16:.*]] = fir.load %[[VAL_7]] : !fir.ref<i32>
-! CHECK:               %[[VAL_17:.*]] = fir.convert %[[VAL_16]] : (i32) -> f32
-! CHECK:               %[[VAL_18:.*]] = arith.addf %[[VAL_15]], %[[VAL_17]] fastmath<contract> : f32
-! CHECK:               fir.store %[[VAL_18]] to %[[VAL_11]] : !fir.ref<f32>
-! CHECK:               %[[VAL_19:.*]] = fir.load %[[VAL_12]] : !fir.ref<f32>
-! CHECK:               %[[VAL_20:.*]] = fir.load %[[VAL_7]] : !fir.ref<i32>
-! CHECK:               %[[VAL_21:.*]] = fir.convert %[[VAL_20]] : (i32) -> f32
-! CHECK:               %[[VAL_22:.*]] = arith.addf %[[VAL_19]], %[[VAL_21]] fastmath<contract> : f32
-! CHECK:               fir.store %[[VAL_22]] to %[[VAL_12]] : !fir.ref<f32>
-! CHECK:               %[[VAL_23:.*]] = fir.load %[[VAL_13]] : !fir.ref<f32>
-! CHECK:               %[[VAL_24:.*]] = fir.load %[[VAL_7]] : !fir.ref<i32>
-! CHECK:               %[[VAL_25:.*]] = fir.convert %[[VAL_24]] : (i32) -> f32
-! CHECK:               %[[VAL_26:.*]] = arith.addf %[[VAL_23]], %[[VAL_25]] fastmath<contract> : f32
-! CHECK:               fir.store %[[VAL_26]] to %[[VAL_13]] : !fir.ref<f32>
-! CHECK:               omp.yield
-! CHECK:             }
-! CHECK:             omp.terminator
-! CHECK:           }
-! CHECK:           return
-! CHECK:         }
-
-subroutine multiple_real_reductions_same_type
-  real :: x,y,z
-  x = 0.0
-  y = 0.0
-  z = 0.0
-  !$omp parallel
-  !$omp do reduction(+:x,y,z)
-  do i=1, 100
-    x = x + i
-    y = y + i
-    z = z + i
-  end do
-  !$omp end do
-  !$omp end parallel
-end subroutine
-
-! CHECK-LABEL:   func.func @_QPmultiple_reductions_different_type() {
-! CHECK:           %[[VAL_0:.*]] = fir.alloca i32 {bindc_name = "i", uniq_name = "_QFmultiple_reductions_different_typeEi"}
-! CHECK:           %[[VAL_1:.*]] = fir.alloca f64 {bindc_name = "w", uniq_name = "_QFmultiple_reductions_different_typeEw"}
-! CHECK:           %[[VAL_2:.*]] = fir.alloca i32 {bindc_name = "x", uniq_name = "_QFmultiple_reductions_different_typeEx"}
-! CHECK:           %[[VAL_3:.*]] = fir.alloca i64 {bindc_name = "y", uniq_name = "_QFmultiple_reductions_different_typeEy"}
-! CHECK:           %[[VAL_4:.*]] = fir.alloca f32 {bindc_name = "z", uniq_name = "_QFmultiple_reductions_different_typeEz"}
-! CHECK:           %[[VAL_5:.*]] = arith.constant 0 : i32
-! CHECK:           fir.store %[[VAL_5]] to %[[VAL_2]] : !fir.ref<i32>
-! CHECK:           %[[VAL_6:.*]] = arith.constant 0 : i64
-! CHECK:           fir.store %[[VAL_6]] to %[[VAL_3]] : !fir.ref<i64>
-! CHECK:           %[[VAL_7:.*]] = arith.constant 0.000000e+00 : f32
-! CHECK:           fir.store %[[VAL_7]] to %[[VAL_4]] : !fir.ref<f32>
-! CHECK:           %[[VAL_8:.*]] = arith.constant 0.000000e+00 : f64
-! CHECK:           fir.store %[[VAL_8]] to %[[VAL_1]] : !fir.ref<f64>
-! CHECK:           omp.parallel {
-! CHECK:             %[[VAL_9:.*]] = fir.alloca i32 {adapt.valuebyref, pinned}
-! CHECK:             %[[VAL_10:.*]] = arith.constant 1 : i32
-! CHECK:             %[[VAL_11:.*]] = arith.constant 100 : i32
-! CHECK:             %[[VAL_12:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop byref reduction(@add_reduction_byref_i32 %[[VAL_2]] -> %[[VAL_13:.*]] : !fir.ref<i32>, @add_reduction_byref_i64 %[[VAL_3]] -> %[[VAL_14:.*]] : !fir.ref<i64>, @add_reduction_byref_f32 %[[VAL_4]] -> %[[VAL_15:.*]] : !fir.ref<f32>, @add_reduction_byref_f64 %[[VAL_1]] -> %[[VAL_16:.*]] : !fir.ref<f64>)  for  (%[[VAL_17:.*]]) : i32 = (%[[VAL_10]]) to (%[[VAL_11]]) inclusive step (%[[VAL_12]]) {
-! CHECK:               fir.store %[[VAL_17]] to %[[VAL_9]] : !fir.ref<i32>
-! CHECK:               %[[VAL_18:.*]] = fir.load %[[VAL_13]] : !fir.ref<i32>
-! CHECK:               %[[VAL_19:.*]] = fir.load %[[VAL_9]] : !fir.ref<i32>
-! CHECK:               %[[VAL_20:.*]] = arith.addi %[[VAL_18]], %[[VAL_19]] : i32
-! CHECK:               fir.store %[[VAL_20]] to %[[VAL_13]] : !fir.ref<i32>
-! CHECK:               %[[VAL_21:.*]] = fir.load %[[VAL_14]] : !fir.ref<i64>
-! CHECK:               %[[VAL_22:.*]] = fir.load %[[VAL_9]] : !fir.ref<i32>
-! CHECK:               %[[VAL_23:.*]] = fir.convert %[[VAL_22]] : (i32) -> i64
-! CHECK:               %[[VAL_24:.*]] = arith.addi %[[VAL_21]], %[[VAL_23]] : i64
-! CHECK:               fir.store %[[VAL_24]] to %[[VAL_14]] : !fir.ref<i64>
-! CHECK:               %[[VAL_25:.*]] = fir.load %[[VAL_15]] : !fir.ref<f32>
-! CHECK:               %[[VAL_26:.*]] = fir.load %[[VAL_9]] : !fir.ref<i32>
-! CHECK:               %[[VAL_27:.*]] = fir.convert %[[VAL_26]] : (i32) -> f32
-! CHECK:               %[[VAL_28:.*]] = arith.addf %[[VAL_25]], %[[VAL_27]] fastmath<contract> : f32
-! CHECK:               fir.store %[[VAL_28]] to %[[VAL_15]] : !fir.ref<f32>
-! CHECK:               %[[VAL_29:.*]] = fir.load %[[VAL_16]] : !fir.ref<f64>
-! CHECK:               %[[VAL_30:.*]] = fir.load %[[VAL_9]] : !fir.ref<i32>
-! CHECK:               %[[VAL_31:.*]] = fir.convert %[[VAL_30]] : (i32) -> f64
-! CHECK:               %[[VAL_32:.*]] = arith.addf %[[VAL_29]], %[[VAL_31]] fastmath<contract> : f64
-! CHECK:               fir.store %[[VAL_32]] to %[[VAL_16]] : !fir.ref<f64>
-! CHECK:               omp.yield
-! CHECK:             }
-! CHECK:             omp.terminator
-! CHECK:           }
-! CHECK:           return
-! CHECK:         }
-
-
-subroutine multiple_reductions_different_type
-  integer :: x
-  integer(kind=8) :: y
-  real :: z
-  real(kind=8) :: w
-  x = 0
-  y = 0
-  z = 0.0
-  w = 0.0
-  !$omp parallel
-  !$omp do reduction(+:x,y,z,w)
-  do i=1, 100
-    x = x + i
-    y = y + i
-    z = z + i
-    w = w + i
-  end do
-  !$omp end do
-  !$omp end parallel
-end subroutine
diff --git a/flang/test/Lower/OpenMP/FIR/wsloop-reduction-add.f90 b/flang/test/Lower/OpenMP/FIR/wsloop-reduction-add.f90
deleted file mode 100644
index dc96b875f745..000000000000
--- a/flang/test/Lower/OpenMP/FIR/wsloop-reduction-add.f90
+++ /dev/null
@@ -1,367 +0,0 @@
-! RUN: bbc -emit-fir -hlfir=false -fopenmp %s -o - | FileCheck %s
-! RUN: %flang_fc1 -emit-fir -flang-deprecated-no-hlfir -fopenmp %s -o - | FileCheck %s
-
-! CHECK-LABEL:   omp.declare_reduction @add_reduction_f64 : f64 init {
-! CHECK:         ^bb0(%[[VAL_0:.*]]: f64):
-! CHECK:           %[[VAL_1:.*]] = arith.constant 0.000000e+00 : f64
-! CHECK:           omp.yield(%[[VAL_1]] : f64)
-
-! CHECK-LABEL:   } combiner {
-! CHECK:         ^bb0(%[[VAL_0:.*]]: f64, %[[VAL_1:.*]]: f64):
-! CHECK:           %[[VAL_2:.*]] = arith.addf %[[VAL_0]], %[[VAL_1]] fastmath<contract> : f64
-! CHECK:           omp.yield(%[[VAL_2]] : f64)
-! CHECK:         }
-
-! CHECK-LABEL:   omp.declare_reduction @add_reduction_i64 : i64 init {
-! CHECK:         ^bb0(%[[VAL_0:.*]]: i64):
-! CHECK:           %[[VAL_1:.*]] = arith.constant 0 : i64
-! CHECK:           omp.yield(%[[VAL_1]] : i64)
-
-! CHECK-LABEL:   } combiner {
-! CHECK:         ^bb0(%[[VAL_0:.*]]: i64, %[[VAL_1:.*]]: i64):
-! CHECK:           %[[VAL_2:.*]] = arith.addi %[[VAL_0]], %[[VAL_1]] : i64
-! CHECK:           omp.yield(%[[VAL_2]] : i64)
-! CHECK:         }
-
-! CHECK-LABEL:   omp.declare_reduction @add_reduction_f32 : f32 init {
-! CHECK:         ^bb0(%[[VAL_0:.*]]: f32):
-! CHECK:           %[[VAL_1:.*]] = arith.constant 0.000000e+00 : f32
-! CHECK:           omp.yield(%[[VAL_1]] : f32)
-
-! CHECK-LABEL:   } combiner {
-! CHECK:         ^bb0(%[[VAL_0:.*]]: f32, %[[VAL_1:.*]]: f32):
-! CHECK:           %[[VAL_2:.*]] = arith.addf %[[VAL_0]], %[[VAL_1]] fastmath<contract> : f32
-! CHECK:           omp.yield(%[[VAL_2]] : f32)
-! CHECK:         }
-
-! CHECK-LABEL:   omp.declare_reduction @add_reduction_i32 : i32 init {
-! CHECK:         ^bb0(%[[VAL_0:.*]]: i32):
-! CHECK:           %[[VAL_1:.*]] = arith.constant 0 : i32
-! CHECK:           omp.yield(%[[VAL_1]] : i32)
-
-! CHECK-LABEL:   } combiner {
-! CHECK:         ^bb0(%[[VAL_0:.*]]: i32, %[[VAL_1:.*]]: i32):
-! CHECK:           %[[VAL_2:.*]] = arith.addi %[[VAL_0]], %[[VAL_1]] : i32
-! CHECK:           omp.yield(%[[VAL_2]] : i32)
-! CHECK:         }
-
-! CHECK-LABEL:   func.func @_QPsimple_int_reduction() {
-! CHECK:           %[[VAL_0:.*]] = fir.alloca i32 {bindc_name = "i", uniq_name = "_QFsimple_int_reductionEi"}
-! CHECK:           %[[VAL_1:.*]] = fir.alloca i32 {bindc_name = "x", uniq_name = "_QFsimple_int_reductionEx"}
-! CHECK:           %[[VAL_2:.*]] = arith.constant 0 : i32
-! CHECK:           fir.store %[[VAL_2]] to %[[VAL_1]] : !fir.ref<i32>
-! CHECK:           omp.parallel {
-! CHECK:             %[[VAL_3:.*]] = fir.alloca i32 {adapt.valuebyref, pinned}
-! CHECK:             %[[VAL_4:.*]] = arith.constant 1 : i32
-! CHECK:             %[[VAL_5:.*]] = arith.constant 100 : i32
-! CHECK:             %[[VAL_6:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop reduction(@add_reduction_i32 %[[VAL_1]] -> %[[VAL_7:.*]] : !fir.ref<i32>)  for  (%[[VAL_8:.*]]) : i32 = (%[[VAL_4]]) to (%[[VAL_5]]) inclusive step (%[[VAL_6]]) {
-! CHECK:               fir.store %[[VAL_8]] to %[[VAL_3]] : !fir.ref<i32>
-! CHECK:               %[[VAL_9:.*]] = fir.load %[[VAL_7]] : !fir.ref<i32>
-! CHECK:               %[[VAL_10:.*]] = fir.load %[[VAL_3]] : !fir.ref<i32>
-! CHECK:               %[[VAL_11:.*]] = arith.addi %[[VAL_9]], %[[VAL_10]] : i32
-! CHECK:               fir.store %[[VAL_11]] to %[[VAL_7]] : !fir.ref<i32>
-! CHECK:               omp.yield
-! CHECK:             }
-! CHECK:             omp.terminator
-! CHECK:           }
-! CHECK:           return
-! CHECK:         }
-
-subroutine simple_int_reduction
-  integer :: x
-  x = 0
-  !$omp parallel
-  !$omp do reduction(+:x)
-  do i=1, 100
-    x = x + i
-  end do
-  !$omp end do
-  !$omp end parallel
-end subroutine
-
-
-! CHECK-LABEL:   func.func @_QPsimple_real_reduction() {
-! CHECK:           %[[VAL_0:.*]] = fir.alloca i32 {bindc_name = "i", uniq_name = "_QFsimple_real_reductionEi"}
-! CHECK:           %[[VAL_1:.*]] = fir.alloca f32 {bindc_name = "x", uniq_name = "_QFsimple_real_reductionEx"}
-! CHECK:           %[[VAL_2:.*]] = arith.constant 0.000000e+00 : f32
-! CHECK:           fir.store %[[VAL_2]] to %[[VAL_1]] : !fir.ref<f32>
-! CHECK:           omp.parallel {
-! CHECK:             %[[VAL_3:.*]] = fir.alloca i32 {adapt.valuebyref, pinned}
-! CHECK:             %[[VAL_4:.*]] = arith.constant 1 : i32
-! CHECK:             %[[VAL_5:.*]] = arith.constant 100 : i32
-! CHECK:             %[[VAL_6:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop reduction(@add_reduction_f32 %[[VAL_1]] -> %[[VAL_7:.*]] : !fir.ref<f32>)  for  (%[[VAL_8:.*]]) : i32 = (%[[VAL_4]]) to (%[[VAL_5]]) inclusive step (%[[VAL_6]]) {
-! CHECK:               fir.store %[[VAL_8]] to %[[VAL_3]] : !fir.ref<i32>
-! CHECK:               %[[VAL_9:.*]] = fir.load %[[VAL_7]] : !fir.ref<f32>
-! CHECK:               %[[VAL_10:.*]] = fir.load %[[VAL_3]] : !fir.ref<i32>
-! CHECK:               %[[VAL_11:.*]] = fir.convert %[[VAL_10]] : (i32) -> f32
-! CHECK:               %[[VAL_12:.*]] = arith.addf %[[VAL_9]], %[[VAL_11]] fastmath<contract> : f32
-! CHECK:               fir.store %[[VAL_12]] to %[[VAL_7]] : !fir.ref<f32>
-! CHECK:               omp.yield
-! CHECK:             }
-! CHECK:             omp.terminator
-! CHECK:           }
-! CHECK:           return
-! CHECK:         }
-
-subroutine simple_real_reduction
-  real :: x
-  x = 0.0
-  !$omp parallel
-  !$omp do reduction(+:x)
-  do i=1, 100
-    x = x + i
-  end do
-  !$omp end do
-  !$omp end parallel
-end subroutine
-
-! CHECK-LABEL:   func.func @_QPsimple_int_reduction_switch_order() {
-! CHECK:           %[[VAL_0:.*]] = fir.alloca i32 {bindc_name = "i", uniq_name = "_QFsimple_int_reduction_switch_orderEi"}
-! CHECK:           %[[VAL_1:.*]] = fir.alloca i32 {bindc_name = "x", uniq_name = "_QFsimple_int_reduction_switch_orderEx"}
-! CHECK:           %[[VAL_2:.*]] = arith.constant 0 : i32
-! CHECK:           fir.store %[[VAL_2]] to %[[VAL_1]] : !fir.ref<i32>
-! CHECK:           omp.parallel {
-! CHECK:             %[[VAL_3:.*]] = fir.alloca i32 {adapt.valuebyref, pinned}
-! CHECK:             %[[VAL_4:.*]] = arith.constant 1 : i32
-! CHECK:             %[[VAL_5:.*]] = arith.constant 100 : i32
-! CHECK:             %[[VAL_6:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop reduction(@add_reduction_i32 %[[VAL_1]] -> %[[VAL_7:.*]] : !fir.ref<i32>)  for  (%[[VAL_8:.*]]) : i32 = (%[[VAL_4]]) to (%[[VAL_5]]) inclusive step (%[[VAL_6]]) {
-! CHECK:               fir.store %[[VAL_8]] to %[[VAL_3]] : !fir.ref<i32>
-! CHECK:               %[[VAL_9:.*]] = fir.load %[[VAL_3]] : !fir.ref<i32>
-! CHECK:               %[[VAL_10:.*]] = fir.load %[[VAL_7]] : !fir.ref<i32>
-! CHECK:               %[[VAL_11:.*]] = arith.addi %[[VAL_9]], %[[VAL_10]] : i32
-! CHECK:               fir.store %[[VAL_11]] to %[[VAL_7]] : !fir.ref<i32>
-! CHECK:               omp.yield
-! CHECK:             }
-! CHECK:             omp.terminator
-! CHECK:           }
-! CHECK:           return
-! CHECK:         }
-
-subroutine simple_int_reduction_switch_order
-  integer :: x
-  x = 0
-  !$omp parallel
-  !$omp do reduction(+:x)
-  do i=1, 100
-    x = i + x
-  end do
-  !$omp end do
-  !$omp end parallel
-end subroutine
-
-! CHECK-LABEL:   func.func @_QPsimple_real_reduction_switch_order() {
-! CHECK:           %[[VAL_0:.*]] = fir.alloca i32 {bindc_name = "i", uniq_name = "_QFsimple_real_reduction_switch_orderEi"}
-! CHECK:           %[[VAL_1:.*]] = fir.alloca f32 {bindc_name = "x", uniq_name = "_QFsimple_real_reduction_switch_orderEx"}
-! CHECK:           %[[VAL_2:.*]] = arith.constant 0.000000e+00 : f32
-! CHECK:           fir.store %[[VAL_2]] to %[[VAL_1]] : !fir.ref<f32>
-! CHECK:           omp.parallel {
-! CHECK:             %[[VAL_3:.*]] = fir.alloca i32 {adapt.valuebyref, pinned}
-! CHECK:             %[[VAL_4:.*]] = arith.constant 1 : i32
-! CHECK:             %[[VAL_5:.*]] = arith.constant 100 : i32
-! CHECK:             %[[VAL_6:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop reduction(@add_reduction_f32 %[[VAL_1]] -> %[[VAL_7:.*]] : !fir.ref<f32>)  for  (%[[VAL_8:.*]]) : i32 = (%[[VAL_4]]) to (%[[VAL_5]]) inclusive step (%[[VAL_6]]) {
-! CHECK:               fir.store %[[VAL_8]] to %[[VAL_3]] : !fir.ref<i32>
-! CHECK:               %[[VAL_9:.*]] = fir.load %[[VAL_3]] : !fir.ref<i32>
-! CHECK:               %[[VAL_10:.*]] = fir.convert %[[VAL_9]] : (i32) -> f32
-! CHECK:               %[[VAL_11:.*]] = fir.load %[[VAL_7]] : !fir.ref<f32>
-! CHECK:               %[[VAL_12:.*]] = arith.addf %[[VAL_10]], %[[VAL_11]] fastmath<contract> : f32
-! CHECK:               fir.store %[[VAL_12]] to %[[VAL_7]] : !fir.ref<f32>
-! CHECK:               omp.yield
-! CHECK:             }
-! CHECK:             omp.terminator
-! CHECK:           }
-! CHECK:           return
-! CHECK:         }
-
-subroutine simple_real_reduction_switch_order
-  real :: x
-  x = 0.0
-  !$omp parallel
-  !$omp do reduction(+:x)
-  do i=1, 100
-    x = i + x
-  end do
-  !$omp end do
-  !$omp end parallel
-end subroutine
-
-! CHECK-LABEL:   func.func @_QPmultiple_int_reductions_same_type() {
-! CHECK:           %[[VAL_0:.*]] = fir.alloca i32 {bindc_name = "i", uniq_name = "_QFmultiple_int_reductions_same_typeEi"}
-! CHECK:           %[[VAL_1:.*]] = fir.alloca i32 {bindc_name = "x", uniq_name = "_QFmultiple_int_reductions_same_typeEx"}
-! CHECK:           %[[VAL_2:.*]] = fir.alloca i32 {bindc_name = "y", uniq_name = "_QFmultiple_int_reductions_same_typeEy"}
-! CHECK:           %[[VAL_3:.*]] = fir.alloca i32 {bindc_name = "z", uniq_name = "_QFmultiple_int_reductions_same_typeEz"}
-! CHECK:           %[[VAL_4:.*]] = arith.constant 0 : i32
-! CHECK:           fir.store %[[VAL_4]] to %[[VAL_1]] : !fir.ref<i32>
-! CHECK:           %[[VAL_5:.*]] = arith.constant 0 : i32
-! CHECK:           fir.store %[[VAL_5]] to %[[VAL_2]] : !fir.ref<i32>
-! CHECK:           %[[VAL_6:.*]] = arith.constant 0 : i32
-! CHECK:           fir.store %[[VAL_6]] to %[[VAL_3]] : !fir.ref<i32>
-! CHECK:           omp.parallel {
-! CHECK:             %[[VAL_7:.*]] = fir.alloca i32 {adapt.valuebyref, pinned}
-! CHECK:             %[[VAL_8:.*]] = arith.constant 1 : i32
-! CHECK:             %[[VAL_9:.*]] = arith.constant 100 : i32
-! CHECK:             %[[VAL_10:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop reduction(@add_reduction_i32 %[[VAL_1]] -> %[[VAL_11:.*]] : !fir.ref<i32>, @add_reduction_i32 %[[VAL_2]] -> %[[VAL_12:.*]] : !fir.ref<i32>, @add_reduction_i32 %[[VAL_3]] -> %[[VAL_13:.*]] : !fir.ref<i32>)  for  (%[[VAL_14:.*]]) : i32 = (%[[VAL_8]]) to (%[[VAL_9]]) inclusive step (%[[VAL_10]]) {
-! CHECK:               fir.store %[[VAL_14]] to %[[VAL_7]] : !fir.ref<i32>
-! CHECK:               %[[VAL_15:.*]] = fir.load %[[VAL_11]] : !fir.ref<i32>
-! CHECK:               %[[VAL_16:.*]] = fir.load %[[VAL_7]] : !fir.ref<i32>
-! CHECK:               %[[VAL_17:.*]] = arith.addi %[[VAL_15]], %[[VAL_16]] : i32
-! CHECK:               fir.store %[[VAL_17]] to %[[VAL_11]] : !fir.ref<i32>
-! CHECK:               %[[VAL_18:.*]] = fir.load %[[VAL_12]] : !fir.ref<i32>
-! CHECK:               %[[VAL_19:.*]] = fir.load %[[VAL_7]] : !fir.ref<i32>
-! CHECK:               %[[VAL_20:.*]] = arith.addi %[[VAL_18]], %[[VAL_19]] : i32
-! CHECK:               fir.store %[[VAL_20]] to %[[VAL_12]] : !fir.ref<i32>
-! CHECK:               %[[VAL_21:.*]] = fir.load %[[VAL_13]] : !fir.ref<i32>
-! CHECK:               %[[VAL_22:.*]] = fir.load %[[VAL_7]] : !fir.ref<i32>
-! CHECK:               %[[VAL_23:.*]] = arith.addi %[[VAL_21]], %[[VAL_22]] : i32
-! CHECK:               fir.store %[[VAL_23]] to %[[VAL_13]] : !fir.ref<i32>
-! CHECK:               omp.yield
-! CHECK:             }
-! CHECK:             omp.terminator
-! CHECK:           }
-! CHECK:           return
-! CHECK:         }
-
-subroutine multiple_int_reductions_same_type
-  integer :: x,y,z
-  x = 0
-  y = 0
-  z = 0
-  !$omp parallel
-  !$omp do reduction(+:x,y,z)
-  do i=1, 100
-    x = x + i
-    y = y + i
-    z = z + i
-  end do
-  !$omp end do
-  !$omp end parallel
-end subroutine
-
-! CHECK-LABEL:   func.func @_QPmultiple_real_reductions_same_type() {
-! CHECK:           %[[VAL_0:.*]] = fir.alloca i32 {bindc_name = "i", uniq_name = "_QFmultiple_real_reductions_same_typeEi"}
-! CHECK:           %[[VAL_1:.*]] = fir.alloca f32 {bindc_name = "x", uniq_name = "_QFmultiple_real_reductions_same_typeEx"}
-! CHECK:           %[[VAL_2:.*]] = fir.alloca f32 {bindc_name = "y", uniq_name = "_QFmultiple_real_reductions_same_typeEy"}
-! CHECK:           %[[VAL_3:.*]] = fir.alloca f32 {bindc_name = "z", uniq_name = "_QFmultiple_real_reductions_same_typeEz"}
-! CHECK:           %[[VAL_4:.*]] = arith.constant 0.000000e+00 : f32
-! CHECK:           fir.store %[[VAL_4]] to %[[VAL_1]] : !fir.ref<f32>
-! CHECK:           %[[VAL_5:.*]] = arith.constant 0.000000e+00 : f32
-! CHECK:           fir.store %[[VAL_5]] to %[[VAL_2]] : !fir.ref<f32>
-! CHECK:           %[[VAL_6:.*]] = arith.constant 0.000000e+00 : f32
-! CHECK:           fir.store %[[VAL_6]] to %[[VAL_3]] : !fir.ref<f32>
-! CHECK:           omp.parallel {
-! CHECK:             %[[VAL_7:.*]] = fir.alloca i32 {adapt.valuebyref, pinned}
-! CHECK:             %[[VAL_8:.*]] = arith.constant 1 : i32
-! CHECK:             %[[VAL_9:.*]] = arith.constant 100 : i32
-! CHECK:             %[[VAL_10:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop reduction(@add_reduction_f32 %[[VAL_1]] -> %[[VAL_11:.*]] : !fir.ref<f32>, @add_reduction_f32 %[[VAL_2]] -> %[[VAL_12:.*]] : !fir.ref<f32>, @add_reduction_f32 %[[VAL_3]] -> %[[VAL_13:.*]] : !fir.ref<f32>)  for  (%[[VAL_14:.*]]) : i32 = (%[[VAL_8]]) to (%[[VAL_9]]) inclusive step (%[[VAL_10]]) {
-! CHECK:               fir.store %[[VAL_14]] to %[[VAL_7]] : !fir.ref<i32>
-! CHECK:               %[[VAL_15:.*]] = fir.load %[[VAL_11]] : !fir.ref<f32>
-! CHECK:               %[[VAL_16:.*]] = fir.load %[[VAL_7]] : !fir.ref<i32>
-! CHECK:               %[[VAL_17:.*]] = fir.convert %[[VAL_16]] : (i32) -> f32
-! CHECK:               %[[VAL_18:.*]] = arith.addf %[[VAL_15]], %[[VAL_17]] fastmath<contract> : f32
-! CHECK:               fir.store %[[VAL_18]] to %[[VAL_11]] : !fir.ref<f32>
-! CHECK:               %[[VAL_19:.*]] = fir.load %[[VAL_12]] : !fir.ref<f32>
-! CHECK:               %[[VAL_20:.*]] = fir.load %[[VAL_7]] : !fir.ref<i32>
-! CHECK:               %[[VAL_21:.*]] = fir.convert %[[VAL_20]] : (i32) -> f32
-! CHECK:               %[[VAL_22:.*]] = arith.addf %[[VAL_19]], %[[VAL_21]] fastmath<contract> : f32
-! CHECK:               fir.store %[[VAL_22]] to %[[VAL_12]] : !fir.ref<f32>
-! CHECK:               %[[VAL_23:.*]] = fir.load %[[VAL_13]] : !fir.ref<f32>
-! CHECK:               %[[VAL_24:.*]] = fir.load %[[VAL_7]] : !fir.ref<i32>
-! CHECK:               %[[VAL_25:.*]] = fir.convert %[[VAL_24]] : (i32) -> f32
-! CHECK:               %[[VAL_26:.*]] = arith.addf %[[VAL_23]], %[[VAL_25]] fastmath<contract> : f32
-! CHECK:               fir.store %[[VAL_26]] to %[[VAL_13]] : !fir.ref<f32>
-! CHECK:               omp.yield
-! CHECK:             }
-! CHECK:             omp.terminator
-! CHECK:           }
-! CHECK:           return
-! CHECK:         }
-
-subroutine multiple_real_reductions_same_type
-  real :: x,y,z
-  x = 0.0
-  y = 0.0
-  z = 0.0
-  !$omp parallel
-  !$omp do reduction(+:x,y,z)
-  do i=1, 100
-    x = x + i
-    y = y + i
-    z = z + i
-  end do
-  !$omp end do
-  !$omp end parallel
-end subroutine
-
-! CHECK-LABEL:   func.func @_QPmultiple_reductions_different_type() {
-! CHECK:           %[[VAL_0:.*]] = fir.alloca i32 {bindc_name = "i", uniq_name = "_QFmultiple_reductions_different_typeEi"}
-! CHECK:           %[[VAL_1:.*]] = fir.alloca f64 {bindc_name = "w", uniq_name = "_QFmultiple_reductions_different_typeEw"}
-! CHECK:           %[[VAL_2:.*]] = fir.alloca i32 {bindc_name = "x", uniq_name = "_QFmultiple_reductions_different_typeEx"}
-! CHECK:           %[[VAL_3:.*]] = fir.alloca i64 {bindc_name = "y", uniq_name = "_QFmultiple_reductions_different_typeEy"}
-! CHECK:           %[[VAL_4:.*]] = fir.alloca f32 {bindc_name = "z", uniq_name = "_QFmultiple_reductions_different_typeEz"}
-! CHECK:           %[[VAL_5:.*]] = arith.constant 0 : i32
-! CHECK:           fir.store %[[VAL_5]] to %[[VAL_2]] : !fir.ref<i32>
-! CHECK:           %[[VAL_6:.*]] = arith.constant 0 : i64
-! CHECK:           fir.store %[[VAL_6]] to %[[VAL_3]] : !fir.ref<i64>
-! CHECK:           %[[VAL_7:.*]] = arith.constant 0.000000e+00 : f32
-! CHECK:           fir.store %[[VAL_7]] to %[[VAL_4]] : !fir.ref<f32>
-! CHECK:           %[[VAL_8:.*]] = arith.constant 0.000000e+00 : f64
-! CHECK:           fir.store %[[VAL_8]] to %[[VAL_1]] : !fir.ref<f64>
-! CHECK:           omp.parallel {
-! CHECK:             %[[VAL_9:.*]] = fir.alloca i32 {adapt.valuebyref, pinned}
-! CHECK:             %[[VAL_10:.*]] = arith.constant 1 : i32
-! CHECK:             %[[VAL_11:.*]] = arith.constant 100 : i32
-! CHECK:             %[[VAL_12:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop reduction(@add_reduction_i32 %[[VAL_2]] -> %[[VAL_13:.*]] : !fir.ref<i32>, @add_reduction_i64 %[[VAL_3]] -> %[[VAL_14:.*]] : !fir.ref<i64>, @add_reduction_f32 %[[VAL_4]] -> %[[VAL_15:.*]] : !fir.ref<f32>, @add_reduction_f64 %[[VAL_1]] -> %[[VAL_16:.*]] : !fir.ref<f64>)  for  (%[[VAL_17:.*]]) : i32 = (%[[VAL_10]]) to (%[[VAL_11]]) inclusive step (%[[VAL_12]]) {
-! CHECK:               fir.store %[[VAL_17]] to %[[VAL_9]] : !fir.ref<i32>
-! CHECK:               %[[VAL_18:.*]] = fir.load %[[VAL_13]] : !fir.ref<i32>
-! CHECK:               %[[VAL_19:.*]] = fir.load %[[VAL_9]] : !fir.ref<i32>
-! CHECK:               %[[VAL_20:.*]] = arith.addi %[[VAL_18]], %[[VAL_19]] : i32
-! CHECK:               fir.store %[[VAL_20]] to %[[VAL_13]] : !fir.ref<i32>
-! CHECK:               %[[VAL_21:.*]] = fir.load %[[VAL_14]] : !fir.ref<i64>
-! CHECK:               %[[VAL_22:.*]] = fir.load %[[VAL_9]] : !fir.ref<i32>
-! CHECK:               %[[VAL_23:.*]] = fir.convert %[[VAL_22]] : (i32) -> i64
-! CHECK:               %[[VAL_24:.*]] = arith.addi %[[VAL_21]], %[[VAL_23]] : i64
-! CHECK:               fir.store %[[VAL_24]] to %[[VAL_14]] : !fir.ref<i64>
-! CHECK:               %[[VAL_25:.*]] = fir.load %[[VAL_15]] : !fir.ref<f32>
-! CHECK:               %[[VAL_26:.*]] = fir.load %[[VAL_9]] : !fir.ref<i32>
-! CHECK:               %[[VAL_27:.*]] = fir.convert %[[VAL_26]] : (i32) -> f32
-! CHECK:               %[[VAL_28:.*]] = arith.addf %[[VAL_25]], %[[VAL_27]] fastmath<contract> : f32
-! CHECK:               fir.store %[[VAL_28]] to %[[VAL_15]] : !fir.ref<f32>
-! CHECK:               %[[VAL_29:.*]] = fir.load %[[VAL_16]] : !fir.ref<f64>
-! CHECK:               %[[VAL_30:.*]] = fir.load %[[VAL_9]] : !fir.ref<i32>
-! CHECK:               %[[VAL_31:.*]] = fir.convert %[[VAL_30]] : (i32) -> f64
-! CHECK:               %[[VAL_32:.*]] = arith.addf %[[VAL_29]], %[[VAL_31]] fastmath<contract> : f64
-! CHECK:               fir.store %[[VAL_32]] to %[[VAL_16]] : !fir.ref<f64>
-! CHECK:               omp.yield
-! CHECK:             }
-! CHECK:             omp.terminator
-! CHECK:           }
-! CHECK:           return
-! CHECK:         }
-
-
-subroutine multiple_reductions_different_type
-  integer :: x
-  integer(kind=8) :: y
-  real :: z
-  real(kind=8) :: w
-  x = 0
-  y = 0
-  z = 0.0
-  w = 0.0
-  !$omp parallel
-  !$omp do reduction(+:x,y,z,w)
-  do i=1, 100
-    x = x + i
-    y = y + i
-    z = z + i
-    w = w + i
-  end do
-  !$omp end do
-  !$omp end parallel
-end subroutine
diff --git a/flang/test/Lower/OpenMP/FIR/wsloop-reduction-iand-byref.f90 b/flang/test/Lower/OpenMP/FIR/wsloop-reduction-iand-byref.f90
deleted file mode 100644
index 6717597ff3b0..000000000000
--- a/flang/test/Lower/OpenMP/FIR/wsloop-reduction-iand-byref.f90
+++ /dev/null
@@ -1,46 +0,0 @@
-! RUN: bbc -emit-fir -hlfir=false -fopenmp --force-byref-reduction %s -o - | FileCheck %s
-! RUN: %flang_fc1 -emit-fir -flang-deprecated-no-hlfir -fopenmp -mmlir --force-byref-reduction %s -o - | FileCheck %s
-
-!CHECK-LABEL:   omp.declare_reduction @iand_byref_i32 : !fir.ref<i32>
-!CHECK-SAME:    init {
-!CHECK:         ^bb0(%[[VAL_0:.*]]: !fir.ref<i32>):
-!CHECK:            %[[C0_1:.*]] = arith.constant -1 : i32
-!CHECK:            %[[REF:.*]] = fir.alloca i32
-!CHECK:            fir.store %[[C0_1]] to %[[REF]] : !fir.ref<i32>
-!CHECK:            omp.yield(%[[REF]] : !fir.ref<i32>)
-
-!CHECK-LABEL:   } combiner {
-!CHECK:         ^bb0(%[[ARG0:.*]]: !fir.ref<i32>, %[[ARG1:.*]]: !fir.ref<i32>):
-!CHECK:           %[[LD0:.*]] = fir.load %[[ARG0]] : !fir.ref<i32>
-!CHECK:           %[[LD1:.*]] = fir.load %[[ARG1]] : !fir.ref<i32>
-!CHECK:           %[[RES:.*]] = arith.andi %[[LD0]], %[[LD1]] : i32
-!CHECK:           fir.store %[[RES]] to %[[ARG0]] : !fir.ref<i32>
-!CHECK:           omp.yield(%[[ARG0]] : !fir.ref<i32>)
-!CHECK:         }
-
-
-!CHECK-LABEL: @_QPreduction_iand
-!CHECK-SAME: %[[Y_BOX:.*]]: !fir.box<!fir.array<?xi32>>
-!CHECK: %[[X_REF:.*]] = fir.alloca i32 {bindc_name = "x", uniq_name = "_QFreduction_iandEx"}
-!CHECK: omp.parallel
-!CHECK: omp.wsloop byref reduction(@iand_byref_i32 %[[X_REF]] -> %[[PRV:.+]] : !fir.ref<i32>) for
-!CHECK: %[[LPRV:.+]] = fir.load %[[PRV]] : !fir.ref<i32>
-!CHECK: %[[Y_I_REF:.*]] = fir.coordinate_of %[[Y_BOX]]
-!CHECK: %[[Y_I:.*]] = fir.load %[[Y_I_REF]] : !fir.ref<i32>
-!CHECK: %[[RES:.+]] = arith.andi %[[LPRV]], %[[Y_I]] : i32
-!CHECK: fir.store %[[RES]] to %[[PRV]] : !fir.ref<i32>
-!CHECK: omp.yield
-!CHECK: omp.terminator
-
-subroutine reduction_iand(y)
-  integer :: x, y(:)
-  x = 0
-  !$omp parallel
-  !$omp do reduction(iand:x)
-  do i=1, 100
-    x = iand(x, y(i))
-  end do
-  !$omp end do
-  !$omp end parallel
-  print *, x
-end subroutine
diff --git a/flang/test/Lower/OpenMP/FIR/wsloop-reduction-iand.f90 b/flang/test/Lower/OpenMP/FIR/wsloop-reduction-iand.f90
deleted file mode 100644
index 9bc45f9f3a0d..000000000000
--- a/flang/test/Lower/OpenMP/FIR/wsloop-reduction-iand.f90
+++ /dev/null
@@ -1,36 +0,0 @@
-! RUN: bbc -emit-fir -hlfir=false -fopenmp %s -o - | FileCheck %s
-! RUN: %flang_fc1 -emit-fir -flang-deprecated-no-hlfir -fopenmp %s -o - | FileCheck %s
-
-!CHECK: omp.declare_reduction @[[IAND_DECLARE_I:.*]] : i32 init {
-!CHECK: %[[ZERO_VAL_I:.*]] = arith.constant -1 : i32
-!CHECK: omp.yield(%[[ZERO_VAL_I]] : i32)
-!CHECK: combiner
-!CHECK: ^bb0(%[[ARG0_I:.*]]: i32, %[[ARG1_I:.*]]: i32):
-!CHECK: %[[IAND_VAL_I:.*]] = arith.andi %[[ARG0_I]], %[[ARG1_I]] : i32
-!CHECK: omp.yield(%[[IAND_VAL_I]] : i32)
-
-!CHECK-LABEL: @_QPreduction_iand
-!CHECK-SAME: %[[Y_BOX:.*]]: !fir.box<!fir.array<?xi32>>
-!CHECK: %[[X_REF:.*]] = fir.alloca i32 {bindc_name = "x", uniq_name = "_QFreduction_iandEx"}
-!CHECK: omp.parallel
-!CHECK: omp.wsloop reduction(@[[IAND_DECLARE_I]] %[[X_REF]] -> %[[PRV:.+]] : !fir.ref<i32>) for
-!CHECK: %[[LPRV:.+]] = fir.load %[[PRV]] : !fir.ref<i32>
-!CHECK: %[[Y_I_REF:.*]] = fir.coordinate_of %[[Y_BOX]]
-!CHECK: %[[Y_I:.*]] = fir.load %[[Y_I_REF]] : !fir.ref<i32>
-!CHECK: %[[RES:.+]] = arith.andi %[[LPRV]], %[[Y_I]] : i32
-!CHECK: fir.store %[[RES]] to %[[PRV]] : !fir.ref<i32>
-!CHECK: omp.yield
-!CHECK: omp.terminator
-
-subroutine reduction_iand(y)
-  integer :: x, y(:)
-  x = 0
-  !$omp parallel
-  !$omp do reduction(iand:x)
-  do i=1, 100
-    x = iand(x, y(i))
-  end do
-  !$omp end do
-  !$omp end parallel
-  print *, x
-end subroutine
diff --git a/flang/test/Lower/OpenMP/FIR/wsloop-reduction-ieor-byref.f90 b/flang/test/Lower/OpenMP/FIR/wsloop-reduction-ieor-byref.f90
deleted file mode 100644
index 1baa59a510fa..000000000000
--- a/flang/test/Lower/OpenMP/FIR/wsloop-reduction-ieor-byref.f90
+++ /dev/null
@@ -1,45 +0,0 @@
-! RUN: bbc -emit-fir -hlfir=false -fopenmp --force-byref-reduction %s -o - | FileCheck %s
-! RUN: %flang_fc1 -emit-fir -flang-deprecated-no-hlfir -mmlir --force-byref-reduction -fopenmp %s -o - | FileCheck %s
-
-! CHECK-LABEL:   omp.declare_reduction @ieor_byref_i32 : !fir.ref<i32>
-! CHECK-SAME:    init {
-! CHECK:         ^bb0(%[[VAL_0:.*]]: !fir.ref<i32>):
-! CHECK:            %[[C0_1:.*]] = arith.constant 0 : i32
-! CHECK:            %[[REF:.*]] = fir.alloca i32
-! CHECK:            fir.store %[[C0_1]] to %[[REF]] : !fir.ref<i32>
-! CHECK:            omp.yield(%[[REF]] : !fir.ref<i32>)
-
-! CHECK-LABEL:   } combiner {
-! CHECK:         ^bb0(%[[ARG0:.*]]: !fir.ref<i32>, %[[ARG1:.*]]: !fir.ref<i32>):
-! CHECK:           %[[LD0:.*]] = fir.load %[[ARG0]] : !fir.ref<i32>
-! CHECK:           %[[LD1:.*]] = fir.load %[[ARG1]] : !fir.ref<i32>
-! CHECK:           %[[RES:.*]] = arith.xori %[[LD0]], %[[LD1]] : i32
-! CHECK:           fir.store %[[RES]] to %[[ARG0]] : !fir.ref<i32>
-! CHECK:           omp.yield(%[[ARG0]] : !fir.ref<i32>)
-! CHECK:         }
-
-!CHECK-LABEL: @_QPreduction_ieor
-!CHECK-SAME: %[[Y_BOX:.*]]: !fir.box<!fir.array<?xi32>>
-!CHECK: %[[X_REF:.*]] = fir.alloca i32 {bindc_name = "x", uniq_name = "_QFreduction_ieorEx"}
-!CHECK: omp.parallel
-!CHECK: omp.wsloop byref reduction(@ieor_byref_i32 %[[X_REF]] -> %[[PRV:.+]] : !fir.ref<i32>) for
-!CHECK: %[[LPRV:.+]] = fir.load %[[PRV]] : !fir.ref<i32>
-!CHECK: %[[Y_I_REF:.*]] = fir.coordinate_of %[[Y_BOX]]
-!CHECK: %[[Y_I:.*]] = fir.load %[[Y_I_REF]] : !fir.ref<i32>
-!CHECK: %[[RES:.+]] = arith.xori %[[LPRV]], %[[Y_I]] : i32
-!CHECK: fir.store %[[RES]] to %[[PRV]] : !fir.ref<i32>
-!CHECK: omp.yield
-!CHECK: omp.terminator
-
-subroutine reduction_ieor(y)
-  integer :: x, y(:)
-  x = 0
-  !$omp parallel
-  !$omp do reduction(ieor:x)
-  do i=1, 100
-    x = ieor(x, y(i))
-  end do
-  !$omp end do
-  !$omp end parallel
-  print *, x
-end subroutine
diff --git a/flang/test/Lower/OpenMP/FIR/wsloop-reduction-ieor.f90 b/flang/test/Lower/OpenMP/FIR/wsloop-reduction-ieor.f90
deleted file mode 100644
index 9c07d5ee2087..000000000000
--- a/flang/test/Lower/OpenMP/FIR/wsloop-reduction-ieor.f90
+++ /dev/null
@@ -1,36 +0,0 @@
-! RUN: bbc -emit-fir -hlfir=false -fopenmp %s -o - | FileCheck %s
-! RUN: %flang_fc1 -emit-fir -flang-deprecated-no-hlfir -fopenmp %s -o - | FileCheck %s
-
-!CHECK: omp.declare_reduction @[[IEOR_DECLARE_I:.*]] : i32 init {
-!CHECK: %[[ZERO_VAL_I:.*]] = arith.constant 0 : i32
-!CHECK: omp.yield(%[[ZERO_VAL_I]] : i32)
-!CHECK: combiner
-!CHECK: ^bb0(%[[ARG0_I:.*]]: i32, %[[ARG1_I:.*]]: i32):
-!CHECK: %[[IEOR_VAL_I:.*]] = arith.xori %[[ARG0_I]], %[[ARG1_I]] : i32
-!CHECK: omp.yield(%[[IEOR_VAL_I]] : i32)
-
-!CHECK-LABEL: @_QPreduction_ieor
-!CHECK-SAME: %[[Y_BOX:.*]]: !fir.box<!fir.array<?xi32>>
-!CHECK: %[[X_REF:.*]] = fir.alloca i32 {bindc_name = "x", uniq_name = "_QFreduction_ieorEx"}
-!CHECK: omp.parallel
-!CHECK: omp.wsloop reduction(@[[IEOR_DECLARE_I]] %[[X_REF]] -> %[[PRV:.+]] : !fir.ref<i32>) for
-!CHECK: %[[LPRV:.+]] = fir.load %[[PRV]] : !fir.ref<i32>
-!CHECK: %[[Y_I_REF:.*]] = fir.coordinate_of %[[Y_BOX]]
-!CHECK: %[[Y_I:.*]] = fir.load %[[Y_I_REF]] : !fir.ref<i32>
-!CHECK: %[[RES:.+]] = arith.xori %[[LPRV]], %[[Y_I]] : i32
-!CHECK: fir.store %[[RES]] to %[[PRV]] : !fir.ref<i32>
-!CHECK: omp.yield
-!CHECK: omp.terminator
-
-subroutine reduction_ieor(y)
-  integer :: x, y(:)
-  x = 0
-  !$omp parallel
-  !$omp do reduction(ieor:x)
-  do i=1, 100
-    x = ieor(x, y(i))
-  end do
-  !$omp end do
-  !$omp end parallel
-  print *, x
-end subroutine
diff --git a/flang/test/Lower/OpenMP/FIR/wsloop-reduction-ior-byref.f90 b/flang/test/Lower/OpenMP/FIR/wsloop-reduction-ior-byref.f90
deleted file mode 100644
index 5482ef33fc8a..000000000000
--- a/flang/test/Lower/OpenMP/FIR/wsloop-reduction-ior-byref.f90
+++ /dev/null
@@ -1,45 +0,0 @@
-! RUN: bbc -emit-fir -hlfir=false -fopenmp --force-byref-reduction %s -o - | FileCheck %s
-! RUN: %flang_fc1 -emit-fir -flang-deprecated-no-hlfir -fopenmp -mmlir --force-byref-reduction %s -o - | FileCheck %s
-
-! CHECK-LABEL:   omp.declare_reduction @ior_byref_i32 : !fir.ref<i32>
-! CHECK-SAME:    init {
-! CHECK:         ^bb0(%[[VAL_0:.*]]: !fir.ref<i32>):
-! CHECK:            %[[C0_1:.*]] = arith.constant 0 : i32
-! CHECK:            %[[REF:.*]] = fir.alloca i32
-! CHECK:            fir.store %[[C0_1]] to %[[REF]] : !fir.ref<i32>
-! CHECK:            omp.yield(%[[REF]] : !fir.ref<i32>)
-
-! CHECK-LABEL:   } combiner {
-! CHECK:         ^bb0(%[[ARG0:.*]]: !fir.ref<i32>, %[[ARG1:.*]]: !fir.ref<i32>):
-! CHECK:           %[[LD0:.*]] = fir.load %[[ARG0]] : !fir.ref<i32>
-! CHECK:           %[[LD1:.*]] = fir.load %[[ARG1]] : !fir.ref<i32>
-! CHECK:           %[[RES:.*]] = arith.ori %[[LD0]], %[[LD1]] : i32
-! CHECK:           fir.store %[[RES]] to %[[ARG0]] : !fir.ref<i32>
-! CHECK:           omp.yield(%[[ARG0]] : !fir.ref<i32>)
-! CHECK:         }
-
-!CHECK-LABEL: @_QPreduction_ior
-!CHECK-SAME: %[[Y_BOX:.*]]: !fir.box<!fir.array<?xi32>> 
-!CHECK: %[[X_REF:.*]] = fir.alloca i32 {bindc_name = "x", uniq_name = "_QFreduction_iorEx"}
-!CHECK: omp.parallel
-!CHECK: omp.wsloop byref reduction(@ior_byref_i32 %[[X_REF]] -> %[[PRV:.+]] : !fir.ref<i32>) for 
-!CHECK: %[[LPRV:.+]] = fir.load %[[PRV]] : !fir.ref<i32>
-!CHECK: %[[Y_I_REF:.*]] = fir.coordinate_of %[[Y_BOX]]
-!CHECK: %[[Y_I:.*]] = fir.load %[[Y_I_REF]] : !fir.ref<i32>
-!CHECK: %[[RES:.+]] = arith.ori %[[LPRV]], %[[Y_I]] : i32
-!CHECK: fir.store %[[RES]] to %[[PRV]] : !fir.ref<i32>
-!CHECK: omp.yield
-!CHECK: omp.terminator
-
-subroutine reduction_ior(y)
-  integer :: x, y(:)
-  x = 0
-  !$omp parallel
-  !$omp do reduction(ior:x)
-  do i=1, 100
-    x = ior(x, y(i))
-  end do
-  !$omp end do
-  !$omp end parallel
-  print *, x
-end subroutine
diff --git a/flang/test/Lower/OpenMP/FIR/wsloop-reduction-ior.f90 b/flang/test/Lower/OpenMP/FIR/wsloop-reduction-ior.f90
deleted file mode 100644
index 79cc8b2d8922..000000000000
--- a/flang/test/Lower/OpenMP/FIR/wsloop-reduction-ior.f90
+++ /dev/null
@@ -1,36 +0,0 @@
-! RUN: bbc -emit-fir -hlfir=false -fopenmp %s -o - | FileCheck %s
-! RUN: %flang_fc1 -emit-fir -flang-deprecated-no-hlfir -fopenmp %s -o - | FileCheck %s
-
-!CHECK: omp.declare_reduction @[[IOR_DECLARE_I:.*]] : i32 init {
-!CHECK: %[[ZERO_VAL_I:.*]] = arith.constant 0 : i32
-!CHECK: omp.yield(%[[ZERO_VAL_I]] : i32)
-!CHECK: combiner
-!CHECK: ^bb0(%[[ARG0_I:.*]]: i32, %[[ARG1_I:.*]]: i32):
-!CHECK: %[[IOR_VAL_I:.*]] = arith.ori %[[ARG0_I]], %[[ARG1_I]] : i32
-!CHECK: omp.yield(%[[IOR_VAL_I]] : i32)
-
-!CHECK-LABEL: @_QPreduction_ior
-!CHECK-SAME: %[[Y_BOX:.*]]: !fir.box<!fir.array<?xi32>> 
-!CHECK: %[[X_REF:.*]] = fir.alloca i32 {bindc_name = "x", uniq_name = "_QFreduction_iorEx"}
-!CHECK: omp.parallel
-!CHECK: omp.wsloop reduction(@[[IOR_DECLARE_I]] %[[X_REF]] -> %[[PRV:.+]] : !fir.ref<i32>) for 
-!CHECK: %[[LPRV:.+]] = fir.load %[[PRV]] : !fir.ref<i32>
-!CHECK: %[[Y_I_REF:.*]] = fir.coordinate_of %[[Y_BOX]]
-!CHECK: %[[Y_I:.*]] = fir.load %[[Y_I_REF]] : !fir.ref<i32>
-!CHECK: %[[RES:.+]] = arith.ori %[[LPRV]], %[[Y_I]] : i32
-!CHECK: fir.store %[[RES]] to %[[PRV]] : !fir.ref<i32>
-!CHECK: omp.yield
-!CHECK: omp.terminator
-
-subroutine reduction_ior(y)
-  integer :: x, y(:)
-  x = 0
-  !$omp parallel
-  !$omp do reduction(ior:x)
-  do i=1, 100
-    x = ior(x, y(i))
-  end do
-  !$omp end do
-  !$omp end parallel
-  print *, x
-end subroutine
diff --git a/flang/test/Lower/OpenMP/FIR/wsloop-reduction-logical-eqv-byref.f90 b/flang/test/Lower/OpenMP/FIR/wsloop-reduction-logical-eqv-byref.f90
deleted file mode 100644
index 696ff68b2059..000000000000
--- a/flang/test/Lower/OpenMP/FIR/wsloop-reduction-logical-eqv-byref.f90
+++ /dev/null
@@ -1,187 +0,0 @@
-! RUN: bbc -emit-fir -hlfir=false -fopenmp --force-byref-reduction %s -o - | FileCheck %s
-! RUN: %flang_fc1 -emit-fir -flang-deprecated-no-hlfir -fopenmp -mmlir --force-byref-reduction %s -o - | FileCheck %s
-
-! NOTE: Assertions have been autogenerated by utils/generate-test-checks.py
-
-! CHECK-LABEL:   omp.declare_reduction @eqv_reduction : !fir.ref<!fir.logical<4>>
-! CHECK-SAME:    init {
-! CHECK:         ^bb0(%[[VAL_0:.*]]: !fir.ref<!fir.logical<4>>):
-! CHECK:           %[[VAL_1:.*]] = arith.constant true
-! CHECK:           %[[VAL_2:.*]] = fir.convert %[[VAL_1]] : (i1) -> !fir.logical<4>
-! CHECK:           %[[REF:.*]] = fir.alloca !fir.logical<4>
-! CHECK:           fir.store %[[VAL_2]] to %[[REF]] : !fir.ref<!fir.logical<4>>
-! CHECK:           omp.yield(%[[REF]] : !fir.ref<!fir.logical<4>>)
-
-! CHECK-LABEL:   } combiner {
-! CHECK:         ^bb0(%[[ARG0:.*]]: !fir.ref<!fir.logical<4>>, %[[ARG1:.*]]: !fir.ref<!fir.logical<4>>):
-! CHECK:           %[[LD0:.*]] = fir.load %[[ARG0]] : !fir.ref<!fir.logical<4>>
-! CHECK:           %[[LD1:.*]] = fir.load %[[ARG1]] : !fir.ref<!fir.logical<4>>
-! CHECK:           %[[VAL_2:.*]] = fir.convert %[[LD0]] : (!fir.logical<4>) -> i1
-! CHECK:           %[[VAL_3:.*]] = fir.convert %[[LD1]] : (!fir.logical<4>) -> i1
-! CHECK:           %[[RES:.*]] = arith.cmpi eq, %[[VAL_2]], %[[VAL_3]] : i1
-! CHECK:           %[[VAL_5:.*]] = fir.convert %[[RES]] : (i1) -> !fir.logical<4>
-! CHECK:           fir.store %[[VAL_5]] to %[[ARG0]] : !fir.ref<!fir.logical<4>>
-! CHECK:           omp.yield(%[[ARG0]] : !fir.ref<!fir.logical<4>>)
-! CHECK:         }
-
-! CHECK-LABEL:   func.func @_QPsimple_reduction(
-! CHECK-SAME:                                   %[[VAL_0:.*]]: !fir.ref<!fir.array<100x!fir.logical<4>>> {fir.bindc_name = "y"}) {
-! CHECK:           %[[VAL_1:.*]] = fir.alloca i32 {bindc_name = "i", uniq_name = "_QFsimple_reductionEi"}
-! CHECK:           %[[VAL_2:.*]] = fir.alloca !fir.logical<4> {bindc_name = "x", uniq_name = "_QFsimple_reductionEx"}
-! CHECK:           %[[VAL_3:.*]] = arith.constant true
-! CHECK:           %[[VAL_4:.*]] = fir.convert %[[VAL_3]] : (i1) -> !fir.logical<4>
-! CHECK:           fir.store %[[VAL_4]] to %[[VAL_2]] : !fir.ref<!fir.logical<4>>
-! CHECK:           omp.parallel {
-! CHECK:             %[[VAL_5:.*]] = fir.alloca i32 {adapt.valuebyref, pinned}
-! CHECK:             %[[VAL_6:.*]] = arith.constant 1 : i32
-! CHECK:             %[[VAL_7:.*]] = arith.constant 100 : i32
-! CHECK:             %[[VAL_8:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop byref reduction(@eqv_reduction %[[VAL_2]] -> %[[VAL_9:.*]] : !fir.ref<!fir.logical<4>>)  for  (%[[VAL_10:.*]]) : i32 = (%[[VAL_6]]) to (%[[VAL_7]]) inclusive step (%[[VAL_8]]) {
-! CHECK:               fir.store %[[VAL_10]] to %[[VAL_5]] : !fir.ref<i32>
-! CHECK:               %[[VAL_11:.*]] = fir.load %[[VAL_9]] : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_12:.*]] = fir.load %[[VAL_5]] : !fir.ref<i32>
-! CHECK:               %[[VAL_13:.*]] = fir.convert %[[VAL_12]] : (i32) -> i64
-! CHECK:               %[[VAL_14:.*]] = arith.constant 1 : i64
-! CHECK:               %[[VAL_15:.*]] = arith.subi %[[VAL_13]], %[[VAL_14]] : i64
-! CHECK:               %[[VAL_16:.*]] = fir.coordinate_of %[[VAL_0]], %[[VAL_15]] : (!fir.ref<!fir.array<100x!fir.logical<4>>>, i64) -> !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_17:.*]] = fir.load %[[VAL_16]] : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_18:.*]] = fir.convert %[[VAL_11]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_19:.*]] = fir.convert %[[VAL_17]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_20:.*]] = arith.cmpi eq, %[[VAL_18]], %[[VAL_19]] : i1
-! CHECK:               %[[VAL_21:.*]] = fir.convert %[[VAL_20]] : (i1) -> !fir.logical<4>
-! CHECK:               fir.store %[[VAL_21]] to %[[VAL_9]] : !fir.ref<!fir.logical<4>>
-! CHECK:               omp.yield
-! CHECK:             omp.terminator
-! CHECK:           return
-
-subroutine simple_reduction(y)
-  logical :: x, y(100)
-  x = .true.
-  !$omp parallel
-  !$omp do reduction(.eqv.:x)
-  do i=1, 100
-    x = x .eqv. y(i)
-  end do
-  !$omp end do
-  !$omp end parallel
-end subroutine
-
-! CHECK-LABEL:   func.func @_QPsimple_reduction_switch_order(
-! CHECK-SAME:                                                %[[VAL_0:.*]]: !fir.ref<!fir.array<100x!fir.logical<4>>> {fir.bindc_name = "y"}) {
-! CHECK:           %[[VAL_1:.*]] = fir.alloca i32 {bindc_name = "i", uniq_name = "_QFsimple_reduction_switch_orderEi"}
-! CHECK:           %[[VAL_2:.*]] = fir.alloca !fir.logical<4> {bindc_name = "x", uniq_name = "_QFsimple_reduction_switch_orderEx"}
-! CHECK:           %[[VAL_3:.*]] = arith.constant true
-! CHECK:           %[[VAL_4:.*]] = fir.convert %[[VAL_3]] : (i1) -> !fir.logical<4>
-! CHECK:           fir.store %[[VAL_4]] to %[[VAL_2]] : !fir.ref<!fir.logical<4>>
-! CHECK:           omp.parallel {
-! CHECK:             %[[VAL_5:.*]] = fir.alloca i32 {adapt.valuebyref, pinned}
-! CHECK:             %[[VAL_6:.*]] = arith.constant 1 : i32
-! CHECK:             %[[VAL_7:.*]] = arith.constant 100 : i32
-! CHECK:             %[[VAL_8:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop byref reduction(@eqv_reduction %[[VAL_2]] -> %[[VAL_9:.*]] : !fir.ref<!fir.logical<4>>)  for  (%[[VAL_10:.*]]) : i32 = (%[[VAL_6]]) to (%[[VAL_7]]) inclusive step (%[[VAL_8]]) {
-! CHECK:               fir.store %[[VAL_10]] to %[[VAL_5]] : !fir.ref<i32>
-! CHECK:               %[[VAL_11:.*]] = fir.load %[[VAL_5]] : !fir.ref<i32>
-! CHECK:               %[[VAL_12:.*]] = fir.convert %[[VAL_11]] : (i32) -> i64
-! CHECK:               %[[VAL_13:.*]] = arith.constant 1 : i64
-! CHECK:               %[[VAL_14:.*]] = arith.subi %[[VAL_12]], %[[VAL_13]] : i64
-! CHECK:               %[[VAL_15:.*]] = fir.coordinate_of %[[VAL_0]], %[[VAL_14]] : (!fir.ref<!fir.array<100x!fir.logical<4>>>, i64) -> !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_16:.*]] = fir.load %[[VAL_15]] : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_17:.*]] = fir.load %[[VAL_9]] : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_18:.*]] = fir.convert %[[VAL_16]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_19:.*]] = fir.convert %[[VAL_17]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_20:.*]] = arith.cmpi eq, %[[VAL_18]], %[[VAL_19]] : i1
-! CHECK:               %[[VAL_21:.*]] = fir.convert %[[VAL_20]] : (i1) -> !fir.logical<4>
-! CHECK:               fir.store %[[VAL_21]] to %[[VAL_9]] : !fir.ref<!fir.logical<4>>
-! CHECK:               omp.yield
-! CHECK:             omp.terminator
-! CHECK:           return
-
-subroutine simple_reduction_switch_order(y)
-  logical :: x, y(100)
-  x = .true.
-  !$omp parallel
-  !$omp do reduction(.eqv.:x)
-  do i=1, 100
-  x = y(i) .eqv. x
-  end do
-  !$omp end do
-  !$omp end parallel
-end subroutine
-
-! CHECK-LABEL:   func.func @_QPmultiple_reductions(
-! CHECK-SAME:                                      %[[VAL_0:.*]]: !fir.ref<!fir.array<100x!fir.logical<4>>> {fir.bindc_name = "w"}) {
-! CHECK:           %[[VAL_1:.*]] = fir.alloca i32 {bindc_name = "i", uniq_name = "_QFmultiple_reductionsEi"}
-! CHECK:           %[[VAL_2:.*]] = fir.alloca !fir.logical<4> {bindc_name = "x", uniq_name = "_QFmultiple_reductionsEx"}
-! CHECK:           %[[VAL_3:.*]] = fir.alloca !fir.logical<4> {bindc_name = "y", uniq_name = "_QFmultiple_reductionsEy"}
-! CHECK:           %[[VAL_4:.*]] = fir.alloca !fir.logical<4> {bindc_name = "z", uniq_name = "_QFmultiple_reductionsEz"}
-! CHECK:           %[[VAL_5:.*]] = arith.constant true
-! CHECK:           %[[VAL_6:.*]] = fir.convert %[[VAL_5]] : (i1) -> !fir.logical<4>
-! CHECK:           fir.store %[[VAL_6]] to %[[VAL_2]] : !fir.ref<!fir.logical<4>>
-! CHECK:           %[[VAL_7:.*]] = arith.constant true
-! CHECK:           %[[VAL_8:.*]] = fir.convert %[[VAL_7]] : (i1) -> !fir.logical<4>
-! CHECK:           fir.store %[[VAL_8]] to %[[VAL_3]] : !fir.ref<!fir.logical<4>>
-! CHECK:           %[[VAL_9:.*]] = arith.constant true
-! CHECK:           %[[VAL_10:.*]] = fir.convert %[[VAL_9]] : (i1) -> !fir.logical<4>
-! CHECK:           fir.store %[[VAL_10]] to %[[VAL_4]] : !fir.ref<!fir.logical<4>>
-! CHECK:           omp.parallel {
-! CHECK:             %[[VAL_11:.*]] = fir.alloca i32 {adapt.valuebyref, pinned}
-! CHECK:             %[[VAL_12:.*]] = arith.constant 1 : i32
-! CHECK:             %[[VAL_13:.*]] = arith.constant 100 : i32
-! CHECK:             %[[VAL_14:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop byref reduction(@eqv_reduction %[[VAL_2]] -> %[[VAL_15:.*]] : !fir.ref<!fir.logical<4>>, @eqv_reduction %[[VAL_3]] -> %[[VAL_16:.*]] : !fir.ref<!fir.logical<4>>, @eqv_reduction %[[VAL_4]] -> %[[VAL_17:.*]] : !fir.ref<!fir.logical<4>>)  for  (%[[VAL_18:.*]]) : i32 = (%[[VAL_12]]) to (%[[VAL_13]]) inclusive step (%[[VAL_14]]) {
-! CHECK:               fir.store %[[VAL_18]] to %[[VAL_11]] : !fir.ref<i32>
-! CHECK:               %[[VAL_19:.*]] = fir.load %[[VAL_15]] : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_20:.*]] = fir.load %[[VAL_11]] : !fir.ref<i32>
-! CHECK:               %[[VAL_21:.*]] = fir.convert %[[VAL_20]] : (i32) -> i64
-! CHECK:               %[[VAL_22:.*]] = arith.constant 1 : i64
-! CHECK:               %[[VAL_23:.*]] = arith.subi %[[VAL_21]], %[[VAL_22]] : i64
-! CHECK:               %[[VAL_24:.*]] = fir.coordinate_of %[[VAL_0]], %[[VAL_23]] : (!fir.ref<!fir.array<100x!fir.logical<4>>>, i64) -> !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_25:.*]] = fir.load %[[VAL_24]] : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_26:.*]] = fir.convert %[[VAL_19]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_27:.*]] = fir.convert %[[VAL_25]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_28:.*]] = arith.cmpi eq, %[[VAL_26]], %[[VAL_27]] : i1
-! CHECK:               %[[VAL_29:.*]] = fir.convert %[[VAL_28]] : (i1) -> !fir.logical<4>
-! CHECK:               fir.store %[[VAL_29]] to %[[VAL_15]] : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_30:.*]] = fir.load %[[VAL_16]] : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_31:.*]] = fir.load %[[VAL_11]] : !fir.ref<i32>
-! CHECK:               %[[VAL_32:.*]] = fir.convert %[[VAL_31]] : (i32) -> i64
-! CHECK:               %[[VAL_33:.*]] = arith.constant 1 : i64
-! CHECK:               %[[VAL_34:.*]] = arith.subi %[[VAL_32]], %[[VAL_33]] : i64
-! CHECK:               %[[VAL_35:.*]] = fir.coordinate_of %[[VAL_0]], %[[VAL_34]] : (!fir.ref<!fir.array<100x!fir.logical<4>>>, i64) -> !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_36:.*]] = fir.load %[[VAL_35]] : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_37:.*]] = fir.convert %[[VAL_30]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_38:.*]] = fir.convert %[[VAL_36]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_39:.*]] = arith.cmpi eq, %[[VAL_37]], %[[VAL_38]] : i1
-! CHECK:               %[[VAL_40:.*]] = fir.convert %[[VAL_39]] : (i1) -> !fir.logical<4>
-! CHECK:               fir.store %[[VAL_40]] to %[[VAL_16]] : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_41:.*]] = fir.load %[[VAL_17]] : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_42:.*]] = fir.load %[[VAL_11]] : !fir.ref<i32>
-! CHECK:               %[[VAL_43:.*]] = fir.convert %[[VAL_42]] : (i32) -> i64
-! CHECK:               %[[VAL_44:.*]] = arith.constant 1 : i64
-! CHECK:               %[[VAL_45:.*]] = arith.subi %[[VAL_43]], %[[VAL_44]] : i64
-! CHECK:               %[[VAL_46:.*]] = fir.coordinate_of %[[VAL_0]], %[[VAL_45]] : (!fir.ref<!fir.array<100x!fir.logical<4>>>, i64) -> !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_47:.*]] = fir.load %[[VAL_46]] : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_48:.*]] = fir.convert %[[VAL_41]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_49:.*]] = fir.convert %[[VAL_47]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_50:.*]] = arith.cmpi eq, %[[VAL_48]], %[[VAL_49]] : i1
-! CHECK:               %[[VAL_51:.*]] = fir.convert %[[VAL_50]] : (i1) -> !fir.logical<4>
-! CHECK:               fir.store %[[VAL_51]] to %[[VAL_17]] : !fir.ref<!fir.logical<4>>
-! CHECK:               omp.yield
-! CHECK:             omp.terminator
-! CHECK:           return
-
-subroutine multiple_reductions(w)
-  logical :: x,y,z,w(100)
-  x = .true.
-  y = .true.
-  z = .true.
-  !$omp parallel
-  !$omp do reduction(.eqv.:x,y,z)
-  do i=1, 100
-  x = x .eqv. w(i)
-  y = y .eqv. w(i)
-  z = z .eqv. w(i)
-  end do
-  !$omp end do
-  !$omp end parallel
-end subroutine
diff --git a/flang/test/Lower/OpenMP/FIR/wsloop-reduction-logical-eqv.f90 b/flang/test/Lower/OpenMP/FIR/wsloop-reduction-logical-eqv.f90
deleted file mode 100644
index 6dcb3952655e..000000000000
--- a/flang/test/Lower/OpenMP/FIR/wsloop-reduction-logical-eqv.f90
+++ /dev/null
@@ -1,181 +0,0 @@
-! RUN: bbc -emit-fir -hlfir=false -fopenmp %s -o - | FileCheck %s
-! RUN: %flang_fc1 -emit-fir -flang-deprecated-no-hlfir -fopenmp %s -o - | FileCheck %s
-
-! NOTE: Assertions have been autogenerated by utils/generate-test-checks.py
-
-! CHECK-LABEL:   omp.declare_reduction @eqv_reduction : !fir.logical<4> init {
-! CHECK:         ^bb0(%[[VAL_0:.*]]: !fir.logical<4>):
-! CHECK:           %[[VAL_1:.*]] = arith.constant true
-! CHECK:           %[[VAL_2:.*]] = fir.convert %[[VAL_1]] : (i1) -> !fir.logical<4>
-! CHECK:           omp.yield(%[[VAL_2]] : !fir.logical<4>)
-
-! CHECK-LABEL:   } combiner {
-! CHECK:         ^bb0(%[[VAL_0:.*]]: !fir.logical<4>, %[[VAL_1:.*]]: !fir.logical<4>):
-! CHECK:           %[[VAL_2:.*]] = fir.convert %[[VAL_0]] : (!fir.logical<4>) -> i1
-! CHECK:           %[[VAL_3:.*]] = fir.convert %[[VAL_1]] : (!fir.logical<4>) -> i1
-! CHECK:           %[[VAL_4:.*]] = arith.cmpi eq, %[[VAL_2]], %[[VAL_3]] : i1
-! CHECK:           %[[VAL_5:.*]] = fir.convert %[[VAL_4]] : (i1) -> !fir.logical<4>
-! CHECK:           omp.yield(%[[VAL_5]] : !fir.logical<4>)
-! CHECK:         }
-
-! CHECK-LABEL:   func.func @_QPsimple_reduction(
-! CHECK-SAME:                                   %[[VAL_0:.*]]: !fir.ref<!fir.array<100x!fir.logical<4>>> {fir.bindc_name = "y"}) {
-! CHECK:           %[[VAL_1:.*]] = fir.alloca i32 {bindc_name = "i", uniq_name = "_QFsimple_reductionEi"}
-! CHECK:           %[[VAL_2:.*]] = fir.alloca !fir.logical<4> {bindc_name = "x", uniq_name = "_QFsimple_reductionEx"}
-! CHECK:           %[[VAL_3:.*]] = arith.constant true
-! CHECK:           %[[VAL_4:.*]] = fir.convert %[[VAL_3]] : (i1) -> !fir.logical<4>
-! CHECK:           fir.store %[[VAL_4]] to %[[VAL_2]] : !fir.ref<!fir.logical<4>>
-! CHECK:           omp.parallel {
-! CHECK:             %[[VAL_5:.*]] = fir.alloca i32 {adapt.valuebyref, pinned}
-! CHECK:             %[[VAL_6:.*]] = arith.constant 1 : i32
-! CHECK:             %[[VAL_7:.*]] = arith.constant 100 : i32
-! CHECK:             %[[VAL_8:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop reduction(@eqv_reduction %[[VAL_2]] -> %[[VAL_9:.*]] : !fir.ref<!fir.logical<4>>)  for  (%[[VAL_10:.*]]) : i32 = (%[[VAL_6]]) to (%[[VAL_7]]) inclusive step (%[[VAL_8]]) {
-! CHECK:               fir.store %[[VAL_10]] to %[[VAL_5]] : !fir.ref<i32>
-! CHECK:               %[[VAL_11:.*]] = fir.load %[[VAL_9]] : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_12:.*]] = fir.load %[[VAL_5]] : !fir.ref<i32>
-! CHECK:               %[[VAL_13:.*]] = fir.convert %[[VAL_12]] : (i32) -> i64
-! CHECK:               %[[VAL_14:.*]] = arith.constant 1 : i64
-! CHECK:               %[[VAL_15:.*]] = arith.subi %[[VAL_13]], %[[VAL_14]] : i64
-! CHECK:               %[[VAL_16:.*]] = fir.coordinate_of %[[VAL_0]], %[[VAL_15]] : (!fir.ref<!fir.array<100x!fir.logical<4>>>, i64) -> !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_17:.*]] = fir.load %[[VAL_16]] : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_18:.*]] = fir.convert %[[VAL_11]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_19:.*]] = fir.convert %[[VAL_17]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_20:.*]] = arith.cmpi eq, %[[VAL_18]], %[[VAL_19]] : i1
-! CHECK:               %[[VAL_21:.*]] = fir.convert %[[VAL_20]] : (i1) -> !fir.logical<4>
-! CHECK:               fir.store %[[VAL_21]] to %[[VAL_9]] : !fir.ref<!fir.logical<4>>
-! CHECK:               omp.yield
-! CHECK:             omp.terminator
-! CHECK:           return
-
-subroutine simple_reduction(y)
-  logical :: x, y(100)
-  x = .true.
-  !$omp parallel
-  !$omp do reduction(.eqv.:x)
-  do i=1, 100
-    x = x .eqv. y(i)
-  end do
-  !$omp end do
-  !$omp end parallel
-end subroutine
-
-! CHECK-LABEL:   func.func @_QPsimple_reduction_switch_order(
-! CHECK-SAME:                                                %[[VAL_0:.*]]: !fir.ref<!fir.array<100x!fir.logical<4>>> {fir.bindc_name = "y"}) {
-! CHECK:           %[[VAL_1:.*]] = fir.alloca i32 {bindc_name = "i", uniq_name = "_QFsimple_reduction_switch_orderEi"}
-! CHECK:           %[[VAL_2:.*]] = fir.alloca !fir.logical<4> {bindc_name = "x", uniq_name = "_QFsimple_reduction_switch_orderEx"}
-! CHECK:           %[[VAL_3:.*]] = arith.constant true
-! CHECK:           %[[VAL_4:.*]] = fir.convert %[[VAL_3]] : (i1) -> !fir.logical<4>
-! CHECK:           fir.store %[[VAL_4]] to %[[VAL_2]] : !fir.ref<!fir.logical<4>>
-! CHECK:           omp.parallel {
-! CHECK:             %[[VAL_5:.*]] = fir.alloca i32 {adapt.valuebyref, pinned}
-! CHECK:             %[[VAL_6:.*]] = arith.constant 1 : i32
-! CHECK:             %[[VAL_7:.*]] = arith.constant 100 : i32
-! CHECK:             %[[VAL_8:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop reduction(@eqv_reduction %[[VAL_2]] -> %[[VAL_9:.*]] : !fir.ref<!fir.logical<4>>)  for  (%[[VAL_10:.*]]) : i32 = (%[[VAL_6]]) to (%[[VAL_7]]) inclusive step (%[[VAL_8]]) {
-! CHECK:               fir.store %[[VAL_10]] to %[[VAL_5]] : !fir.ref<i32>
-! CHECK:               %[[VAL_11:.*]] = fir.load %[[VAL_5]] : !fir.ref<i32>
-! CHECK:               %[[VAL_12:.*]] = fir.convert %[[VAL_11]] : (i32) -> i64
-! CHECK:               %[[VAL_13:.*]] = arith.constant 1 : i64
-! CHECK:               %[[VAL_14:.*]] = arith.subi %[[VAL_12]], %[[VAL_13]] : i64
-! CHECK:               %[[VAL_15:.*]] = fir.coordinate_of %[[VAL_0]], %[[VAL_14]] : (!fir.ref<!fir.array<100x!fir.logical<4>>>, i64) -> !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_16:.*]] = fir.load %[[VAL_15]] : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_17:.*]] = fir.load %[[VAL_9]] : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_18:.*]] = fir.convert %[[VAL_16]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_19:.*]] = fir.convert %[[VAL_17]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_20:.*]] = arith.cmpi eq, %[[VAL_18]], %[[VAL_19]] : i1
-! CHECK:               %[[VAL_21:.*]] = fir.convert %[[VAL_20]] : (i1) -> !fir.logical<4>
-! CHECK:               fir.store %[[VAL_21]] to %[[VAL_9]] : !fir.ref<!fir.logical<4>>
-! CHECK:               omp.yield
-! CHECK:             omp.terminator
-! CHECK:           return
-
-subroutine simple_reduction_switch_order(y)
-  logical :: x, y(100)
-  x = .true.
-  !$omp parallel
-  !$omp do reduction(.eqv.:x)
-  do i=1, 100
-  x = y(i) .eqv. x
-  end do
-  !$omp end do
-  !$omp end parallel
-end subroutine
-
-! CHECK-LABEL:   func.func @_QPmultiple_reductions(
-! CHECK-SAME:                                      %[[VAL_0:.*]]: !fir.ref<!fir.array<100x!fir.logical<4>>> {fir.bindc_name = "w"}) {
-! CHECK:           %[[VAL_1:.*]] = fir.alloca i32 {bindc_name = "i", uniq_name = "_QFmultiple_reductionsEi"}
-! CHECK:           %[[VAL_2:.*]] = fir.alloca !fir.logical<4> {bindc_name = "x", uniq_name = "_QFmultiple_reductionsEx"}
-! CHECK:           %[[VAL_3:.*]] = fir.alloca !fir.logical<4> {bindc_name = "y", uniq_name = "_QFmultiple_reductionsEy"}
-! CHECK:           %[[VAL_4:.*]] = fir.alloca !fir.logical<4> {bindc_name = "z", uniq_name = "_QFmultiple_reductionsEz"}
-! CHECK:           %[[VAL_5:.*]] = arith.constant true
-! CHECK:           %[[VAL_6:.*]] = fir.convert %[[VAL_5]] : (i1) -> !fir.logical<4>
-! CHECK:           fir.store %[[VAL_6]] to %[[VAL_2]] : !fir.ref<!fir.logical<4>>
-! CHECK:           %[[VAL_7:.*]] = arith.constant true
-! CHECK:           %[[VAL_8:.*]] = fir.convert %[[VAL_7]] : (i1) -> !fir.logical<4>
-! CHECK:           fir.store %[[VAL_8]] to %[[VAL_3]] : !fir.ref<!fir.logical<4>>
-! CHECK:           %[[VAL_9:.*]] = arith.constant true
-! CHECK:           %[[VAL_10:.*]] = fir.convert %[[VAL_9]] : (i1) -> !fir.logical<4>
-! CHECK:           fir.store %[[VAL_10]] to %[[VAL_4]] : !fir.ref<!fir.logical<4>>
-! CHECK:           omp.parallel {
-! CHECK:             %[[VAL_11:.*]] = fir.alloca i32 {adapt.valuebyref, pinned}
-! CHECK:             %[[VAL_12:.*]] = arith.constant 1 : i32
-! CHECK:             %[[VAL_13:.*]] = arith.constant 100 : i32
-! CHECK:             %[[VAL_14:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop reduction(@eqv_reduction %[[VAL_2]] -> %[[VAL_15:.*]] : !fir.ref<!fir.logical<4>>, @eqv_reduction %[[VAL_3]] -> %[[VAL_16:.*]] : !fir.ref<!fir.logical<4>>, @eqv_reduction %[[VAL_4]] -> %[[VAL_17:.*]] : !fir.ref<!fir.logical<4>>)  for  (%[[VAL_18:.*]]) : i32 = (%[[VAL_12]]) to (%[[VAL_13]]) inclusive step (%[[VAL_14]]) {
-! CHECK:               fir.store %[[VAL_18]] to %[[VAL_11]] : !fir.ref<i32>
-! CHECK:               %[[VAL_19:.*]] = fir.load %[[VAL_15]] : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_20:.*]] = fir.load %[[VAL_11]] : !fir.ref<i32>
-! CHECK:               %[[VAL_21:.*]] = fir.convert %[[VAL_20]] : (i32) -> i64
-! CHECK:               %[[VAL_22:.*]] = arith.constant 1 : i64
-! CHECK:               %[[VAL_23:.*]] = arith.subi %[[VAL_21]], %[[VAL_22]] : i64
-! CHECK:               %[[VAL_24:.*]] = fir.coordinate_of %[[VAL_0]], %[[VAL_23]] : (!fir.ref<!fir.array<100x!fir.logical<4>>>, i64) -> !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_25:.*]] = fir.load %[[VAL_24]] : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_26:.*]] = fir.convert %[[VAL_19]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_27:.*]] = fir.convert %[[VAL_25]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_28:.*]] = arith.cmpi eq, %[[VAL_26]], %[[VAL_27]] : i1
-! CHECK:               %[[VAL_29:.*]] = fir.convert %[[VAL_28]] : (i1) -> !fir.logical<4>
-! CHECK:               fir.store %[[VAL_29]] to %[[VAL_15]] : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_30:.*]] = fir.load %[[VAL_16]] : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_31:.*]] = fir.load %[[VAL_11]] : !fir.ref<i32>
-! CHECK:               %[[VAL_32:.*]] = fir.convert %[[VAL_31]] : (i32) -> i64
-! CHECK:               %[[VAL_33:.*]] = arith.constant 1 : i64
-! CHECK:               %[[VAL_34:.*]] = arith.subi %[[VAL_32]], %[[VAL_33]] : i64
-! CHECK:               %[[VAL_35:.*]] = fir.coordinate_of %[[VAL_0]], %[[VAL_34]] : (!fir.ref<!fir.array<100x!fir.logical<4>>>, i64) -> !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_36:.*]] = fir.load %[[VAL_35]] : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_37:.*]] = fir.convert %[[VAL_30]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_38:.*]] = fir.convert %[[VAL_36]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_39:.*]] = arith.cmpi eq, %[[VAL_37]], %[[VAL_38]] : i1
-! CHECK:               %[[VAL_40:.*]] = fir.convert %[[VAL_39]] : (i1) -> !fir.logical<4>
-! CHECK:               fir.store %[[VAL_40]] to %[[VAL_16]] : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_41:.*]] = fir.load %[[VAL_17]] : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_42:.*]] = fir.load %[[VAL_11]] : !fir.ref<i32>
-! CHECK:               %[[VAL_43:.*]] = fir.convert %[[VAL_42]] : (i32) -> i64
-! CHECK:               %[[VAL_44:.*]] = arith.constant 1 : i64
-! CHECK:               %[[VAL_45:.*]] = arith.subi %[[VAL_43]], %[[VAL_44]] : i64
-! CHECK:               %[[VAL_46:.*]] = fir.coordinate_of %[[VAL_0]], %[[VAL_45]] : (!fir.ref<!fir.array<100x!fir.logical<4>>>, i64) -> !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_47:.*]] = fir.load %[[VAL_46]] : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_48:.*]] = fir.convert %[[VAL_41]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_49:.*]] = fir.convert %[[VAL_47]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_50:.*]] = arith.cmpi eq, %[[VAL_48]], %[[VAL_49]] : i1
-! CHECK:               %[[VAL_51:.*]] = fir.convert %[[VAL_50]] : (i1) -> !fir.logical<4>
-! CHECK:               fir.store %[[VAL_51]] to %[[VAL_17]] : !fir.ref<!fir.logical<4>>
-! CHECK:               omp.yield
-! CHECK:             omp.terminator
-! CHECK:           return
-
-subroutine multiple_reductions(w)
-  logical :: x,y,z,w(100)
-  x = .true.
-  y = .true.
-  z = .true.
-  !$omp parallel
-  !$omp do reduction(.eqv.:x,y,z)
-  do i=1, 100
-  x = x .eqv. w(i)
-  y = y .eqv. w(i)
-  z = z .eqv. w(i)
-  end do
-  !$omp end do
-  !$omp end parallel
-end subroutine
diff --git a/flang/test/Lower/OpenMP/FIR/wsloop-reduction-logical-neqv-byref.f90 b/flang/test/Lower/OpenMP/FIR/wsloop-reduction-logical-neqv-byref.f90
deleted file mode 100644
index a31abd0def56..000000000000
--- a/flang/test/Lower/OpenMP/FIR/wsloop-reduction-logical-neqv-byref.f90
+++ /dev/null
@@ -1,189 +0,0 @@
-! RUN: bbc -emit-fir -hlfir=false -fopenmp --force-byref-reduction %s -o - | FileCheck %s
-! RUN: %flang_fc1 -emit-fir -flang-deprecated-no-hlfir -fopenmp -mmlir --force-byref-reduction %s -o - | FileCheck %s
-
-! NOTE: Assertions have been autogenerated by utils/generate-test-checks.py
-
-
-! CHECK-LABEL:   omp.declare_reduction @neqv_reduction : !fir.ref<!fir.logical<4>>
-! CHECK-SAME:    init {
-! CHECK:         ^bb0(%[[VAL_0:.*]]: !fir.ref<!fir.logical<4>>):
-! CHECK:           %[[VAL_1:.*]] = arith.constant false
-! CHECK:           %[[VAL_2:.*]] = fir.convert %[[VAL_1]] : (i1) -> !fir.logical<4>
-! CHECK:           %[[REF:.*]] = fir.alloca !fir.logical<4>
-! CHECK:           fir.store %[[VAL_2]] to %[[REF]] : !fir.ref<!fir.logical<4>>
-! CHECK:           omp.yield(%[[REF]] : !fir.ref<!fir.logical<4>>)
-
-! CHECK-LABEL:   } combiner {
-! CHECK:         ^bb0(%[[ARG0:.*]]: !fir.ref<!fir.logical<4>>, %[[ARG1:.*]]: !fir.ref<!fir.logical<4>>):
-! CHECK:           %[[LD0:.*]] = fir.load %[[ARG0]] : !fir.ref<!fir.logical<4>>
-! CHECK:           %[[LD1:.*]] = fir.load %[[ARG1]] : !fir.ref<!fir.logical<4>>
-! CHECK:           %[[VAL_2:.*]] = fir.convert %[[LD0]] : (!fir.logical<4>) -> i1
-! CHECK:           %[[VAL_3:.*]] = fir.convert %[[LD1]] : (!fir.logical<4>) -> i1
-! CHECK:           %[[RES:.*]] = arith.cmpi ne, %[[VAL_2]], %[[VAL_3]] : i1
-! CHECK:           %[[VAL_5:.*]] = fir.convert %[[RES]] : (i1) -> !fir.logical<4>
-! CHECK:           fir.store %[[VAL_5]] to %[[ARG0]] : !fir.ref<!fir.logical<4>>
-! CHECK:           omp.yield(%[[ARG0]] : !fir.ref<!fir.logical<4>>)
-! CHECK:         }
-
-! CHECK-LABEL:   func.func @_QPsimple_reduction(
-! CHECK-SAME:                                   %[[VAL_0:.*]]: !fir.ref<!fir.array<100x!fir.logical<4>>> {fir.bindc_name = "y"}) {
-! CHECK:           %[[VAL_1:.*]] = fir.alloca i32 {bindc_name = "i", uniq_name = "_QFsimple_reductionEi"}
-! CHECK:           %[[VAL_2:.*]] = fir.alloca !fir.logical<4> {bindc_name = "x", uniq_name = "_QFsimple_reductionEx"}
-! CHECK:           %[[VAL_3:.*]] = arith.constant true
-! CHECK:           %[[VAL_4:.*]] = fir.convert %[[VAL_3]] : (i1) -> !fir.logical<4>
-! CHECK:           fir.store %[[VAL_4]] to %[[VAL_2]] : !fir.ref<!fir.logical<4>>
-! CHECK:           omp.parallel {
-! CHECK:             %[[VAL_5:.*]] = fir.alloca i32 {adapt.valuebyref, pinned}
-! CHECK:             %[[VAL_6:.*]] = arith.constant 1 : i32
-! CHECK:             %[[VAL_7:.*]] = arith.constant 100 : i32
-! CHECK:             %[[VAL_8:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop byref reduction(@neqv_reduction %[[VAL_2]] -> %[[VAL_9:.*]] : !fir.ref<!fir.logical<4>>)  for  (%[[VAL_10:.*]]) : i32 = (%[[VAL_6]]) to (%[[VAL_7]]) inclusive step (%[[VAL_8]]) {
-! CHECK:               fir.store %[[VAL_10]] to %[[VAL_5]] : !fir.ref<i32>
-! CHECK:               %[[VAL_11:.*]] = fir.load %[[VAL_9]] : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_12:.*]] = fir.load %[[VAL_5]] : !fir.ref<i32>
-! CHECK:               %[[VAL_13:.*]] = fir.convert %[[VAL_12]] : (i32) -> i64
-! CHECK:               %[[VAL_14:.*]] = arith.constant 1 : i64
-! CHECK:               %[[VAL_15:.*]] = arith.subi %[[VAL_13]], %[[VAL_14]] : i64
-! CHECK:               %[[VAL_16:.*]] = fir.coordinate_of %[[VAL_0]], %[[VAL_15]] : (!fir.ref<!fir.array<100x!fir.logical<4>>>, i64) -> !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_17:.*]] = fir.load %[[VAL_16]] : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_18:.*]] = fir.convert %[[VAL_11]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_19:.*]] = fir.convert %[[VAL_17]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_20:.*]] = arith.cmpi ne, %[[VAL_18]], %[[VAL_19]] : i1
-! CHECK:               %[[VAL_21:.*]] = fir.convert %[[VAL_20]] : (i1) -> !fir.logical<4>
-! CHECK:               fir.store %[[VAL_21]] to %[[VAL_9]] : !fir.ref<!fir.logical<4>>
-! CHECK:               omp.yield
-! CHECK:             omp.terminator
-! CHECK:           return
-
-subroutine simple_reduction(y)
-  logical :: x, y(100)
-  x = .true.
-  !$omp parallel
-  !$omp do reduction(.neqv.:x)
-  do i=1, 100
-    x = x .neqv. y(i)
-  end do
-  !$omp end do
-  !$omp end parallel
-end subroutine
-
-! CHECK-LABEL:   func.func @_QPsimple_reduction_switch_order(
-! CHECK-SAME:                                                %[[VAL_0:.*]]: !fir.ref<!fir.array<100x!fir.logical<4>>> {fir.bindc_name = "y"}) {
-! CHECK:           %[[VAL_1:.*]] = fir.alloca i32 {bindc_name = "i", uniq_name = "_QFsimple_reduction_switch_orderEi"}
-! CHECK:           %[[VAL_2:.*]] = fir.alloca !fir.logical<4> {bindc_name = "x", uniq_name = "_QFsimple_reduction_switch_orderEx"}
-! CHECK:           %[[VAL_3:.*]] = arith.constant true
-! CHECK:           %[[VAL_4:.*]] = fir.convert %[[VAL_3]] : (i1) -> !fir.logical<4>
-! CHECK:           fir.store %[[VAL_4]] to %[[VAL_2]] : !fir.ref<!fir.logical<4>>
-! CHECK:           omp.parallel {
-! CHECK:             %[[VAL_5:.*]] = fir.alloca i32 {adapt.valuebyref, pinned}
-! CHECK:             %[[VAL_6:.*]] = arith.constant 1 : i32
-! CHECK:             %[[VAL_7:.*]] = arith.constant 100 : i32
-! CHECK:             %[[VAL_8:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop byref reduction(@neqv_reduction %[[VAL_2]] -> %[[VAL_9:.*]] : !fir.ref<!fir.logical<4>>)  for  (%[[VAL_10:.*]]) : i32 = (%[[VAL_6]]) to (%[[VAL_7]]) inclusive step (%[[VAL_8]]) {
-! CHECK:               fir.store %[[VAL_10]] to %[[VAL_5]] : !fir.ref<i32>
-! CHECK:               %[[VAL_11:.*]] = fir.load %[[VAL_5]] : !fir.ref<i32>
-! CHECK:               %[[VAL_12:.*]] = fir.convert %[[VAL_11]] : (i32) -> i64
-! CHECK:               %[[VAL_13:.*]] = arith.constant 1 : i64
-! CHECK:               %[[VAL_14:.*]] = arith.subi %[[VAL_12]], %[[VAL_13]] : i64
-! CHECK:               %[[VAL_15:.*]] = fir.coordinate_of %[[VAL_0]], %[[VAL_14]] : (!fir.ref<!fir.array<100x!fir.logical<4>>>, i64) -> !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_16:.*]] = fir.load %[[VAL_15]] : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_17:.*]] = fir.load %[[VAL_9]] : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_18:.*]] = fir.convert %[[VAL_16]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_19:.*]] = fir.convert %[[VAL_17]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_20:.*]] = arith.cmpi ne, %[[VAL_18]], %[[VAL_19]] : i1
-! CHECK:               %[[VAL_21:.*]] = fir.convert %[[VAL_20]] : (i1) -> !fir.logical<4>
-! CHECK:               fir.store %[[VAL_21]] to %[[VAL_9]] : !fir.ref<!fir.logical<4>>
-! CHECK:               omp.yield
-! CHECK:             omp.terminator
-! CHECK:           return
-
-subroutine simple_reduction_switch_order(y)
-  logical :: x, y(100)
-  x = .true.
-  !$omp parallel
-  !$omp do reduction(.neqv.:x)
-  do i=1, 100
-  x = y(i) .neqv. x
-  end do
-  !$omp end do
-  !$omp end parallel
-end subroutine
-
-! CHECK-LABEL:   func.func @_QPmultiple_reductions(
-! CHECK-SAME:                                      %[[VAL_0:.*]]: !fir.ref<!fir.array<100x!fir.logical<4>>> {fir.bindc_name = "w"}) {
-! CHECK:           %[[VAL_1:.*]] = fir.alloca i32 {bindc_name = "i", uniq_name = "_QFmultiple_reductionsEi"}
-! CHECK:           %[[VAL_2:.*]] = fir.alloca !fir.logical<4> {bindc_name = "x", uniq_name = "_QFmultiple_reductionsEx"}
-! CHECK:           %[[VAL_3:.*]] = fir.alloca !fir.logical<4> {bindc_name = "y", uniq_name = "_QFmultiple_reductionsEy"}
-! CHECK:           %[[VAL_4:.*]] = fir.alloca !fir.logical<4> {bindc_name = "z", uniq_name = "_QFmultiple_reductionsEz"}
-! CHECK:           %[[VAL_5:.*]] = arith.constant true
-! CHECK:           %[[VAL_6:.*]] = fir.convert %[[VAL_5]] : (i1) -> !fir.logical<4>
-! CHECK:           fir.store %[[VAL_6]] to %[[VAL_2]] : !fir.ref<!fir.logical<4>>
-! CHECK:           %[[VAL_7:.*]] = arith.constant true
-! CHECK:           %[[VAL_8:.*]] = fir.convert %[[VAL_7]] : (i1) -> !fir.logical<4>
-! CHECK:           fir.store %[[VAL_8]] to %[[VAL_3]] : !fir.ref<!fir.logical<4>>
-! CHECK:           %[[VAL_9:.*]] = arith.constant true
-! CHECK:           %[[VAL_10:.*]] = fir.convert %[[VAL_9]] : (i1) -> !fir.logical<4>
-! CHECK:           fir.store %[[VAL_10]] to %[[VAL_4]] : !fir.ref<!fir.logical<4>>
-! CHECK:           omp.parallel {
-! CHECK:             %[[VAL_11:.*]] = fir.alloca i32 {adapt.valuebyref, pinned}
-! CHECK:             %[[VAL_12:.*]] = arith.constant 1 : i32
-! CHECK:             %[[VAL_13:.*]] = arith.constant 100 : i32
-! CHECK:             %[[VAL_14:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop byref reduction(@neqv_reduction %[[VAL_2]] -> %[[VAL_15:.*]] : !fir.ref<!fir.logical<4>>, @neqv_reduction %[[VAL_3]] -> %[[VAL_16:.*]] : !fir.ref<!fir.logical<4>>, @neqv_reduction %[[VAL_4]] -> %[[VAL_17:.*]] : !fir.ref<!fir.logical<4>>)  for  (%[[VAL_18:.*]]) : i32 = (%[[VAL_12]]) to (%[[VAL_13]]) inclusive step (%[[VAL_14]]) {
-! CHECK:               fir.store %[[VAL_18]] to %[[VAL_11]] : !fir.ref<i32>
-! CHECK:               %[[VAL_19:.*]] = fir.load %[[VAL_15]] : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_20:.*]] = fir.load %[[VAL_11]] : !fir.ref<i32>
-! CHECK:               %[[VAL_21:.*]] = fir.convert %[[VAL_20]] : (i32) -> i64
-! CHECK:               %[[VAL_22:.*]] = arith.constant 1 : i64
-! CHECK:               %[[VAL_23:.*]] = arith.subi %[[VAL_21]], %[[VAL_22]] : i64
-! CHECK:               %[[VAL_24:.*]] = fir.coordinate_of %[[VAL_0]], %[[VAL_23]] : (!fir.ref<!fir.array<100x!fir.logical<4>>>, i64) -> !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_25:.*]] = fir.load %[[VAL_24]] : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_26:.*]] = fir.convert %[[VAL_19]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_27:.*]] = fir.convert %[[VAL_25]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_28:.*]] = arith.cmpi ne, %[[VAL_26]], %[[VAL_27]] : i1
-! CHECK:               %[[VAL_29:.*]] = fir.convert %[[VAL_28]] : (i1) -> !fir.logical<4>
-! CHECK:               fir.store %[[VAL_29]] to %[[VAL_15]] : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_30:.*]] = fir.load %[[VAL_16]] : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_31:.*]] = fir.load %[[VAL_11]] : !fir.ref<i32>
-! CHECK:               %[[VAL_32:.*]] = fir.convert %[[VAL_31]] : (i32) -> i64
-! CHECK:               %[[VAL_33:.*]] = arith.constant 1 : i64
-! CHECK:               %[[VAL_34:.*]] = arith.subi %[[VAL_32]], %[[VAL_33]] : i64
-! CHECK:               %[[VAL_35:.*]] = fir.coordinate_of %[[VAL_0]], %[[VAL_34]] : (!fir.ref<!fir.array<100x!fir.logical<4>>>, i64) -> !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_36:.*]] = fir.load %[[VAL_35]] : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_37:.*]] = fir.convert %[[VAL_30]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_38:.*]] = fir.convert %[[VAL_36]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_39:.*]] = arith.cmpi ne, %[[VAL_37]], %[[VAL_38]] : i1
-! CHECK:               %[[VAL_40:.*]] = fir.convert %[[VAL_39]] : (i1) -> !fir.logical<4>
-! CHECK:               fir.store %[[VAL_40]] to %[[VAL_16]] : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_41:.*]] = fir.load %[[VAL_17]] : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_42:.*]] = fir.load %[[VAL_11]] : !fir.ref<i32>
-! CHECK:               %[[VAL_43:.*]] = fir.convert %[[VAL_42]] : (i32) -> i64
-! CHECK:               %[[VAL_44:.*]] = arith.constant 1 : i64
-! CHECK:               %[[VAL_45:.*]] = arith.subi %[[VAL_43]], %[[VAL_44]] : i64
-! CHECK:               %[[VAL_46:.*]] = fir.coordinate_of %[[VAL_0]], %[[VAL_45]] : (!fir.ref<!fir.array<100x!fir.logical<4>>>, i64) -> !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_47:.*]] = fir.load %[[VAL_46]] : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_48:.*]] = fir.convert %[[VAL_41]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_49:.*]] = fir.convert %[[VAL_47]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_50:.*]] = arith.cmpi ne, %[[VAL_48]], %[[VAL_49]] : i1
-! CHECK:               %[[VAL_51:.*]] = fir.convert %[[VAL_50]] : (i1) -> !fir.logical<4>
-! CHECK:               fir.store %[[VAL_51]] to %[[VAL_17]] : !fir.ref<!fir.logical<4>>
-! CHECK:               omp.yield
-! CHECK:             omp.terminator
-! CHECK:           return
-
-
-subroutine multiple_reductions(w)
-  logical :: x,y,z,w(100)
-  x = .true.
-  y = .true.
-  z = .true.
-  !$omp parallel
-  !$omp do reduction(.neqv.:x,y,z)
-  do i=1, 100
-  x = x .neqv. w(i)
-  y = y .neqv. w(i)
-  z = z .neqv. w(i)
-  end do
-  !$omp end do
-  !$omp end parallel
-end subroutine
diff --git a/flang/test/Lower/OpenMP/FIR/wsloop-reduction-logical-neqv.f90 b/flang/test/Lower/OpenMP/FIR/wsloop-reduction-logical-neqv.f90
deleted file mode 100644
index 702c185e25ee..000000000000
--- a/flang/test/Lower/OpenMP/FIR/wsloop-reduction-logical-neqv.f90
+++ /dev/null
@@ -1,183 +0,0 @@
-! RUN: bbc -emit-fir -hlfir=false -fopenmp %s -o - | FileCheck %s
-! RUN: %flang_fc1 -emit-fir -flang-deprecated-no-hlfir -fopenmp %s -o - | FileCheck %s
-
-! NOTE: Assertions have been autogenerated by utils/generate-test-checks.py
-
-
-! CHECK-LABEL:   omp.declare_reduction @neqv_reduction : !fir.logical<4> init {
-! CHECK:         ^bb0(%[[VAL_0:.*]]: !fir.logical<4>):
-! CHECK:           %[[VAL_1:.*]] = arith.constant false
-! CHECK:           %[[VAL_2:.*]] = fir.convert %[[VAL_1]] : (i1) -> !fir.logical<4>
-! CHECK:           omp.yield(%[[VAL_2]] : !fir.logical<4>)
-
-! CHECK-LABEL:   } combiner {
-! CHECK:         ^bb0(%[[VAL_0:.*]]: !fir.logical<4>, %[[VAL_1:.*]]: !fir.logical<4>):
-! CHECK:           %[[VAL_2:.*]] = fir.convert %[[VAL_0]] : (!fir.logical<4>) -> i1
-! CHECK:           %[[VAL_3:.*]] = fir.convert %[[VAL_1]] : (!fir.logical<4>) -> i1
-! CHECK:           %[[VAL_4:.*]] = arith.cmpi ne, %[[VAL_2]], %[[VAL_3]] : i1
-! CHECK:           %[[VAL_5:.*]] = fir.convert %[[VAL_4]] : (i1) -> !fir.logical<4>
-! CHECK:           omp.yield(%[[VAL_5]] : !fir.logical<4>)
-! CHECK:         }
-
-! CHECK-LABEL:   func.func @_QPsimple_reduction(
-! CHECK-SAME:                                   %[[VAL_0:.*]]: !fir.ref<!fir.array<100x!fir.logical<4>>> {fir.bindc_name = "y"}) {
-! CHECK:           %[[VAL_1:.*]] = fir.alloca i32 {bindc_name = "i", uniq_name = "_QFsimple_reductionEi"}
-! CHECK:           %[[VAL_2:.*]] = fir.alloca !fir.logical<4> {bindc_name = "x", uniq_name = "_QFsimple_reductionEx"}
-! CHECK:           %[[VAL_3:.*]] = arith.constant true
-! CHECK:           %[[VAL_4:.*]] = fir.convert %[[VAL_3]] : (i1) -> !fir.logical<4>
-! CHECK:           fir.store %[[VAL_4]] to %[[VAL_2]] : !fir.ref<!fir.logical<4>>
-! CHECK:           omp.parallel {
-! CHECK:             %[[VAL_5:.*]] = fir.alloca i32 {adapt.valuebyref, pinned}
-! CHECK:             %[[VAL_6:.*]] = arith.constant 1 : i32
-! CHECK:             %[[VAL_7:.*]] = arith.constant 100 : i32
-! CHECK:             %[[VAL_8:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop reduction(@neqv_reduction %[[VAL_2]] -> %[[VAL_9:.*]] : !fir.ref<!fir.logical<4>>)  for  (%[[VAL_10:.*]]) : i32 = (%[[VAL_6]]) to (%[[VAL_7]]) inclusive step (%[[VAL_8]]) {
-! CHECK:               fir.store %[[VAL_10]] to %[[VAL_5]] : !fir.ref<i32>
-! CHECK:               %[[VAL_11:.*]] = fir.load %[[VAL_9]] : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_12:.*]] = fir.load %[[VAL_5]] : !fir.ref<i32>
-! CHECK:               %[[VAL_13:.*]] = fir.convert %[[VAL_12]] : (i32) -> i64
-! CHECK:               %[[VAL_14:.*]] = arith.constant 1 : i64
-! CHECK:               %[[VAL_15:.*]] = arith.subi %[[VAL_13]], %[[VAL_14]] : i64
-! CHECK:               %[[VAL_16:.*]] = fir.coordinate_of %[[VAL_0]], %[[VAL_15]] : (!fir.ref<!fir.array<100x!fir.logical<4>>>, i64) -> !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_17:.*]] = fir.load %[[VAL_16]] : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_18:.*]] = fir.convert %[[VAL_11]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_19:.*]] = fir.convert %[[VAL_17]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_20:.*]] = arith.cmpi ne, %[[VAL_18]], %[[VAL_19]] : i1
-! CHECK:               %[[VAL_21:.*]] = fir.convert %[[VAL_20]] : (i1) -> !fir.logical<4>
-! CHECK:               fir.store %[[VAL_21]] to %[[VAL_9]] : !fir.ref<!fir.logical<4>>
-! CHECK:               omp.yield
-! CHECK:             omp.terminator
-! CHECK:           return
-
-subroutine simple_reduction(y)
-  logical :: x, y(100)
-  x = .true.
-  !$omp parallel
-  !$omp do reduction(.neqv.:x)
-  do i=1, 100
-    x = x .neqv. y(i)
-  end do
-  !$omp end do
-  !$omp end parallel
-end subroutine
-
-! CHECK-LABEL:   func.func @_QPsimple_reduction_switch_order(
-! CHECK-SAME:                                                %[[VAL_0:.*]]: !fir.ref<!fir.array<100x!fir.logical<4>>> {fir.bindc_name = "y"}) {
-! CHECK:           %[[VAL_1:.*]] = fir.alloca i32 {bindc_name = "i", uniq_name = "_QFsimple_reduction_switch_orderEi"}
-! CHECK:           %[[VAL_2:.*]] = fir.alloca !fir.logical<4> {bindc_name = "x", uniq_name = "_QFsimple_reduction_switch_orderEx"}
-! CHECK:           %[[VAL_3:.*]] = arith.constant true
-! CHECK:           %[[VAL_4:.*]] = fir.convert %[[VAL_3]] : (i1) -> !fir.logical<4>
-! CHECK:           fir.store %[[VAL_4]] to %[[VAL_2]] : !fir.ref<!fir.logical<4>>
-! CHECK:           omp.parallel {
-! CHECK:             %[[VAL_5:.*]] = fir.alloca i32 {adapt.valuebyref, pinned}
-! CHECK:             %[[VAL_6:.*]] = arith.constant 1 : i32
-! CHECK:             %[[VAL_7:.*]] = arith.constant 100 : i32
-! CHECK:             %[[VAL_8:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop reduction(@neqv_reduction %[[VAL_2]] -> %[[VAL_9:.*]] : !fir.ref<!fir.logical<4>>)  for  (%[[VAL_10:.*]]) : i32 = (%[[VAL_6]]) to (%[[VAL_7]]) inclusive step (%[[VAL_8]]) {
-! CHECK:               fir.store %[[VAL_10]] to %[[VAL_5]] : !fir.ref<i32>
-! CHECK:               %[[VAL_11:.*]] = fir.load %[[VAL_5]] : !fir.ref<i32>
-! CHECK:               %[[VAL_12:.*]] = fir.convert %[[VAL_11]] : (i32) -> i64
-! CHECK:               %[[VAL_13:.*]] = arith.constant 1 : i64
-! CHECK:               %[[VAL_14:.*]] = arith.subi %[[VAL_12]], %[[VAL_13]] : i64
-! CHECK:               %[[VAL_15:.*]] = fir.coordinate_of %[[VAL_0]], %[[VAL_14]] : (!fir.ref<!fir.array<100x!fir.logical<4>>>, i64) -> !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_16:.*]] = fir.load %[[VAL_15]] : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_17:.*]] = fir.load %[[VAL_9]] : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_18:.*]] = fir.convert %[[VAL_16]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_19:.*]] = fir.convert %[[VAL_17]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_20:.*]] = arith.cmpi ne, %[[VAL_18]], %[[VAL_19]] : i1
-! CHECK:               %[[VAL_21:.*]] = fir.convert %[[VAL_20]] : (i1) -> !fir.logical<4>
-! CHECK:               fir.store %[[VAL_21]] to %[[VAL_9]] : !fir.ref<!fir.logical<4>>
-! CHECK:               omp.yield
-! CHECK:             omp.terminator
-! CHECK:           return
-
-subroutine simple_reduction_switch_order(y)
-  logical :: x, y(100)
-  x = .true.
-  !$omp parallel
-  !$omp do reduction(.neqv.:x)
-  do i=1, 100
-  x = y(i) .neqv. x
-  end do
-  !$omp end do
-  !$omp end parallel
-end subroutine
-
-! CHECK-LABEL:   func.func @_QPmultiple_reductions(
-! CHECK-SAME:                                      %[[VAL_0:.*]]: !fir.ref<!fir.array<100x!fir.logical<4>>> {fir.bindc_name = "w"}) {
-! CHECK:           %[[VAL_1:.*]] = fir.alloca i32 {bindc_name = "i", uniq_name = "_QFmultiple_reductionsEi"}
-! CHECK:           %[[VAL_2:.*]] = fir.alloca !fir.logical<4> {bindc_name = "x", uniq_name = "_QFmultiple_reductionsEx"}
-! CHECK:           %[[VAL_3:.*]] = fir.alloca !fir.logical<4> {bindc_name = "y", uniq_name = "_QFmultiple_reductionsEy"}
-! CHECK:           %[[VAL_4:.*]] = fir.alloca !fir.logical<4> {bindc_name = "z", uniq_name = "_QFmultiple_reductionsEz"}
-! CHECK:           %[[VAL_5:.*]] = arith.constant true
-! CHECK:           %[[VAL_6:.*]] = fir.convert %[[VAL_5]] : (i1) -> !fir.logical<4>
-! CHECK:           fir.store %[[VAL_6]] to %[[VAL_2]] : !fir.ref<!fir.logical<4>>
-! CHECK:           %[[VAL_7:.*]] = arith.constant true
-! CHECK:           %[[VAL_8:.*]] = fir.convert %[[VAL_7]] : (i1) -> !fir.logical<4>
-! CHECK:           fir.store %[[VAL_8]] to %[[VAL_3]] : !fir.ref<!fir.logical<4>>
-! CHECK:           %[[VAL_9:.*]] = arith.constant true
-! CHECK:           %[[VAL_10:.*]] = fir.convert %[[VAL_9]] : (i1) -> !fir.logical<4>
-! CHECK:           fir.store %[[VAL_10]] to %[[VAL_4]] : !fir.ref<!fir.logical<4>>
-! CHECK:           omp.parallel {
-! CHECK:             %[[VAL_11:.*]] = fir.alloca i32 {adapt.valuebyref, pinned}
-! CHECK:             %[[VAL_12:.*]] = arith.constant 1 : i32
-! CHECK:             %[[VAL_13:.*]] = arith.constant 100 : i32
-! CHECK:             %[[VAL_14:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop reduction(@neqv_reduction %[[VAL_2]] -> %[[VAL_15:.*]] : !fir.ref<!fir.logical<4>>, @neqv_reduction %[[VAL_3]] -> %[[VAL_16:.*]] : !fir.ref<!fir.logical<4>>, @neqv_reduction %[[VAL_4]] -> %[[VAL_17:.*]] : !fir.ref<!fir.logical<4>>)  for  (%[[VAL_18:.*]]) : i32 = (%[[VAL_12]]) to (%[[VAL_13]]) inclusive step (%[[VAL_14]]) {
-! CHECK:               fir.store %[[VAL_18]] to %[[VAL_11]] : !fir.ref<i32>
-! CHECK:               %[[VAL_19:.*]] = fir.load %[[VAL_15]] : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_20:.*]] = fir.load %[[VAL_11]] : !fir.ref<i32>
-! CHECK:               %[[VAL_21:.*]] = fir.convert %[[VAL_20]] : (i32) -> i64
-! CHECK:               %[[VAL_22:.*]] = arith.constant 1 : i64
-! CHECK:               %[[VAL_23:.*]] = arith.subi %[[VAL_21]], %[[VAL_22]] : i64
-! CHECK:               %[[VAL_24:.*]] = fir.coordinate_of %[[VAL_0]], %[[VAL_23]] : (!fir.ref<!fir.array<100x!fir.logical<4>>>, i64) -> !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_25:.*]] = fir.load %[[VAL_24]] : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_26:.*]] = fir.convert %[[VAL_19]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_27:.*]] = fir.convert %[[VAL_25]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_28:.*]] = arith.cmpi ne, %[[VAL_26]], %[[VAL_27]] : i1
-! CHECK:               %[[VAL_29:.*]] = fir.convert %[[VAL_28]] : (i1) -> !fir.logical<4>
-! CHECK:               fir.store %[[VAL_29]] to %[[VAL_15]] : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_30:.*]] = fir.load %[[VAL_16]] : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_31:.*]] = fir.load %[[VAL_11]] : !fir.ref<i32>
-! CHECK:               %[[VAL_32:.*]] = fir.convert %[[VAL_31]] : (i32) -> i64
-! CHECK:               %[[VAL_33:.*]] = arith.constant 1 : i64
-! CHECK:               %[[VAL_34:.*]] = arith.subi %[[VAL_32]], %[[VAL_33]] : i64
-! CHECK:               %[[VAL_35:.*]] = fir.coordinate_of %[[VAL_0]], %[[VAL_34]] : (!fir.ref<!fir.array<100x!fir.logical<4>>>, i64) -> !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_36:.*]] = fir.load %[[VAL_35]] : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_37:.*]] = fir.convert %[[VAL_30]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_38:.*]] = fir.convert %[[VAL_36]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_39:.*]] = arith.cmpi ne, %[[VAL_37]], %[[VAL_38]] : i1
-! CHECK:               %[[VAL_40:.*]] = fir.convert %[[VAL_39]] : (i1) -> !fir.logical<4>
-! CHECK:               fir.store %[[VAL_40]] to %[[VAL_16]] : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_41:.*]] = fir.load %[[VAL_17]] : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_42:.*]] = fir.load %[[VAL_11]] : !fir.ref<i32>
-! CHECK:               %[[VAL_43:.*]] = fir.convert %[[VAL_42]] : (i32) -> i64
-! CHECK:               %[[VAL_44:.*]] = arith.constant 1 : i64
-! CHECK:               %[[VAL_45:.*]] = arith.subi %[[VAL_43]], %[[VAL_44]] : i64
-! CHECK:               %[[VAL_46:.*]] = fir.coordinate_of %[[VAL_0]], %[[VAL_45]] : (!fir.ref<!fir.array<100x!fir.logical<4>>>, i64) -> !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_47:.*]] = fir.load %[[VAL_46]] : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_48:.*]] = fir.convert %[[VAL_41]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_49:.*]] = fir.convert %[[VAL_47]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_50:.*]] = arith.cmpi ne, %[[VAL_48]], %[[VAL_49]] : i1
-! CHECK:               %[[VAL_51:.*]] = fir.convert %[[VAL_50]] : (i1) -> !fir.logical<4>
-! CHECK:               fir.store %[[VAL_51]] to %[[VAL_17]] : !fir.ref<!fir.logical<4>>
-! CHECK:               omp.yield
-! CHECK:             omp.terminator
-! CHECK:           return
-
-
-subroutine multiple_reductions(w)
-  logical :: x,y,z,w(100)
-  x = .true.
-  y = .true.
-  z = .true.
-  !$omp parallel
-  !$omp do reduction(.neqv.:x,y,z)
-  do i=1, 100
-  x = x .neqv. w(i)
-  y = y .neqv. w(i)
-  z = z .neqv. w(i)
-  end do
-  !$omp end do
-  !$omp end parallel
-end subroutine
diff --git a/flang/test/Lower/OpenMP/FIR/wsloop-reduction-max-byref.f90 b/flang/test/Lower/OpenMP/FIR/wsloop-reduction-max-byref.f90
deleted file mode 100644
index 80b720e3aac1..000000000000
--- a/flang/test/Lower/OpenMP/FIR/wsloop-reduction-max-byref.f90
+++ /dev/null
@@ -1,90 +0,0 @@
-! RUN: bbc -emit-fir -hlfir=false -fopenmp --force-byref-reduction -o - %s 2>&1 | FileCheck %s
-! RUN: %flang_fc1 -emit-fir -flang-deprecated-no-hlfir -fopenmp -mmlir --force-byref-reduction -o - %s 2>&1 | FileCheck %s
-
-!CHECK: omp.declare_reduction @max_byref_f32 : !fir.ref<f32>
-!CHECK-SAME: init {
-!CHECK:   %[[MINIMUM_VAL:.*]] = arith.constant -3.40282347E+38 : f32
-!CHECK:   %[[REF:.*]] = fir.alloca f32
-!CHECK:   fir.store %[[MINIMUM_VAL]] to %[[REF]] : !fir.ref<f32>
-!CHECK:   omp.yield(%[[REF]] : !fir.ref<f32>)
-!CHECK: combiner
-!CHECK: ^bb0(%[[ARG0:.*]]: !fir.ref<f32>, %[[ARG1:.*]]: !fir.ref<f32>):
-!CHECK:   %[[LD0:.*]] = fir.load %[[ARG0]] : !fir.ref<f32>
-!CHECK:   %[[LD1:.*]] = fir.load %[[ARG1]] : !fir.ref<f32>
-!CHECK:   %[[RES:.*]] = arith.maxnumf %[[LD0]], %[[LD1]] {{.*}}: f32
-!CHECK:   fir.store %[[RES]] to %[[ARG0]] : !fir.ref<f32>
-!CHECK:   omp.yield(%[[ARG0]] : !fir.ref<f32>)
-
-!CHECK-LABEL: omp.declare_reduction @max_byref_i32 : !fir.ref<i32>
-!CHECK-SAME: init {
-!CHECK:   %[[MINIMUM_VAL:.*]] = arith.constant -2147483648 : i32
-!CHECK:   fir.store %[[MINIMUM_VAL]] to %[[REF]] : !fir.ref<i32>
-!CHECK:   omp.yield(%[[REF]] : !fir.ref<i32>)
-!CHECK: combiner
-!CHECK: ^bb0(%[[ARG0:.*]]: !fir.ref<i32>, %[[ARG1:.*]]: !fir.ref<i32>):
-!CHECK:   %[[LD0:.*]] = fir.load %[[ARG0]] : !fir.ref<i32>
-!CHECK:   %[[LD1:.*]] = fir.load %[[ARG1]] : !fir.ref<i32>
-!CHECK:   %[[RES:.*]] = arith.maxsi %[[LD0]], %[[LD1]] : i32
-!CHECK:   fir.store %[[RES]] to %[[ARG0]] : !fir.ref<i32>
-!CHECK:   omp.yield(%[[ARG0]] : !fir.ref<i32>)
-
-!CHECK-LABEL: @_QPreduction_max_int
-!CHECK-SAME: %[[Y_BOX:.*]]: !fir.box<!fir.array<?xi32>>
-!CHECK:   %[[X_REF:.*]] = fir.alloca i32 {bindc_name = "x", uniq_name = "_QFreduction_max_intEx"}
-!CHECK:   omp.parallel
-!CHECK:     omp.wsloop byref reduction(@max_byref_i32 %[[X_REF]] -> %[[PRV:.+]] : !fir.ref<i32>) for
-!CHECK:       %[[LPRV:.+]] = fir.load %[[PRV]] : !fir.ref<i32>
-!CHECK:       %[[Y_I_REF:.*]] = fir.coordinate_of %[[Y_BOX]]
-!CHECK:       %[[Y_I:.*]] = fir.load %[[Y_I_REF]] : !fir.ref<i32>
-!CHECK:       %[[RES:.+]] = arith.cmpi sgt, %[[LPRV]], %[[Y_I]] : i32
-!CHECK:       %[[SEL:.+]] = arith.select %[[RES]], %[[LPRV]], %[[Y_I]]
-!CHECK:       fir.store %[[SEL]] to %[[PRV]] : !fir.ref<i32>
-!CHECK:     omp.terminator
-
-!CHECK-LABEL: @_QPreduction_max_real
-!CHECK-SAME: %[[Y_BOX:.*]]: !fir.box<!fir.array<?xf32>>
-!CHECK:   %[[X_REF:.*]] = fir.alloca f32 {bindc_name = "x", uniq_name = "_QFreduction_max_realEx"}
-!CHECK:   omp.parallel
-!CHECK:     omp.wsloop byref reduction(@max_byref_f32 %[[X_REF]] -> %[[PRV:.+]] : !fir.ref<f32>) for
-!CHECK:       %[[LPRV:.+]] = fir.load %[[PRV]] : !fir.ref<f32>
-!CHECK:       %[[Y_I_REF:.*]] = fir.coordinate_of %[[Y_BOX]]
-!CHECK:       %[[Y_I:.*]] = fir.load %[[Y_I_REF]] : !fir.ref<f32>
-!CHECK:       %[[RES:.+]] = arith.cmpf ogt, %[[Y_I]], %[[LPRV]] {{.*}} : f32
-!CHECK:       omp.yield
-!CHECK:     omp.terminator
-
-subroutine reduction_max_int(y)
-  integer :: x, y(:)
-  x = 0
-  !$omp parallel
-  !$omp do reduction(max:x)
-  do i=1, 100
-    x = max(x, y(i))
-  end do
-  !$omp end do
-  !$omp end parallel
-  print *, x
-end subroutine
-
-subroutine reduction_max_real(y)
-  real :: x, y(:)
-  x = 0.0
-  !$omp parallel
-  !$omp do reduction(max:x)
-  do i=1, 100
-    x = max(y(i), x)
-  end do
-  !$omp end do
-  !$omp end parallel
-  print *, x
-
-  !$omp parallel
-  !$omp do reduction(max:x)
-  do i=1, 100
-    !CHECK-NOT: omp.reduction
-    if (y(i) .gt. x) x = y(i)
-  end do
-  !$omp end do
-  !$omp end parallel
-  print *, x
-end subroutine
diff --git a/flang/test/Lower/OpenMP/FIR/wsloop-reduction-max.f90 b/flang/test/Lower/OpenMP/FIR/wsloop-reduction-max.f90
deleted file mode 100644
index c3b821ea5912..000000000000
--- a/flang/test/Lower/OpenMP/FIR/wsloop-reduction-max.f90
+++ /dev/null
@@ -1,79 +0,0 @@
-! RUN: bbc -emit-fir -hlfir=false -fopenmp -o - %s 2>&1 | FileCheck %s
-! RUN: %flang_fc1 -emit-fir -flang-deprecated-no-hlfir -fopenmp -o - %s 2>&1 | FileCheck %s
-
-!CHECK: omp.declare_reduction @[[MAX_DECLARE_F:.*]] : f32 init {
-!CHECK:   %[[MINIMUM_VAL_F:.*]] = arith.constant -3.40282347E+38 : f32
-!CHECK:   omp.yield(%[[MINIMUM_VAL_F]] : f32)
-!CHECK: combiner
-!CHECK: ^bb0(%[[ARG0_F:.*]]: f32, %[[ARG1_F:.*]]: f32):
-!CHECK:   %[[COMB_VAL_F:.*]] = arith.maxnumf %[[ARG0_F]], %[[ARG1_F]] {{.*}}: f32
-!CHECK:   omp.yield(%[[COMB_VAL_F]] : f32)
-
-!CHECK: omp.declare_reduction @[[MAX_DECLARE_I:.*]] : i32 init {
-!CHECK:   %[[MINIMUM_VAL_I:.*]] = arith.constant -2147483648 : i32
-!CHECK:   omp.yield(%[[MINIMUM_VAL_I]] : i32)
-!CHECK: combiner
-!CHECK: ^bb0(%[[ARG0_I:.*]]: i32, %[[ARG1_I:.*]]: i32):
-!CHECK:   %[[COMB_VAL_I:.*]] = arith.maxsi %[[ARG0_I]], %[[ARG1_I]] : i32
-!CHECK:   omp.yield(%[[COMB_VAL_I]] : i32)
-
-!CHECK-LABEL: @_QPreduction_max_int
-!CHECK-SAME: %[[Y_BOX:.*]]: !fir.box<!fir.array<?xi32>>
-!CHECK:   %[[X_REF:.*]] = fir.alloca i32 {bindc_name = "x", uniq_name = "_QFreduction_max_intEx"}
-!CHECK:   omp.parallel
-!CHECK:     omp.wsloop reduction(@[[MAX_DECLARE_I]] %[[X_REF]] -> %[[PRV:.+]] : !fir.ref<i32>) for
-!CHECK:       %[[LPRV:.+]] = fir.load %[[PRV]] : !fir.ref<i32>
-!CHECK:       %[[Y_I_REF:.*]] = fir.coordinate_of %[[Y_BOX]]
-!CHECK:       %[[Y_I:.*]] = fir.load %[[Y_I_REF]] : !fir.ref<i32>
-!CHECK:       %[[RES:.+]] = arith.cmpi sgt, %[[LPRV]], %[[Y_I]] : i32
-!CHECK:       %[[SEL:.+]] = arith.select %[[RES]], %[[LPRV]], %[[Y_I]]
-!CHECK:       fir.store %[[SEL]] to %[[PRV]] : !fir.ref<i32>
-!CHECK:     omp.terminator
-
-!CHECK-LABEL: @_QPreduction_max_real
-!CHECK-SAME: %[[Y_BOX:.*]]: !fir.box<!fir.array<?xf32>>
-!CHECK:   %[[X_REF:.*]] = fir.alloca f32 {bindc_name = "x", uniq_name = "_QFreduction_max_realEx"}
-!CHECK:   omp.parallel
-!CHECK:     omp.wsloop reduction(@[[MAX_DECLARE_F]] %[[X_REF]] -> %[[PRV:.+]] : !fir.ref<f32>) for
-!CHECK:       %[[LPRV:.+]] = fir.load %[[PRV]] : !fir.ref<f32>
-!CHECK:       %[[Y_I_REF:.*]] = fir.coordinate_of %[[Y_BOX]]
-!CHECK:       %[[Y_I:.*]] = fir.load %[[Y_I_REF]] : !fir.ref<f32>
-!CHECK:       %[[RES:.+]] = arith.cmpf ogt, %[[Y_I]], %[[LPRV]] {{.*}} : f32
-!CHECK:       omp.yield
-!CHECK:     omp.terminator
-
-subroutine reduction_max_int(y)
-  integer :: x, y(:)
-  x = 0
-  !$omp parallel
-  !$omp do reduction(max:x)
-  do i=1, 100
-    x = max(x, y(i))
-  end do
-  !$omp end do
-  !$omp end parallel
-  print *, x
-end subroutine
-
-subroutine reduction_max_real(y)
-  real :: x, y(:)
-  x = 0.0
-  !$omp parallel
-  !$omp do reduction(max:x)
-  do i=1, 100
-    x = max(y(i), x)
-  end do
-  !$omp end do
-  !$omp end parallel
-  print *, x
-
-  !$omp parallel
-  !$omp do reduction(max:x)
-  do i=1, 100
-    !CHECK-NOT: omp.reduction
-    if (y(i) .gt. x) x = y(i)
-  end do
-  !$omp end do
-  !$omp end parallel
-  print *, x
-end subroutine
diff --git a/flang/test/Lower/OpenMP/FIR/wsloop-reduction-min-byref.f90 b/flang/test/Lower/OpenMP/FIR/wsloop-reduction-min-byref.f90
deleted file mode 100644
index b284f8e5d967..000000000000
--- a/flang/test/Lower/OpenMP/FIR/wsloop-reduction-min-byref.f90
+++ /dev/null
@@ -1,91 +0,0 @@
-! RUN: bbc -emit-fir -hlfir=false -fopenmp --force-byref-reduction -o - %s 2>&1 | FileCheck %s
-! RUN: %flang_fc1 -emit-fir -flang-deprecated-no-hlfir -mmlir --force-byref-reduction -fopenmp -o - %s 2>&1 | FileCheck %s
-
-!CHECK: omp.declare_reduction @min_byref_f32 : !fir.ref<f32>
-!CHECK-SAME: init {
-!CHECK:   %[[MAXIMUM_VAL:.*]] = arith.constant 3.40282347E+38 : f32
-!CHECK:   %[[REF:.*]] = fir.alloca f32
-!CHECK:   fir.store %[[MAXIMUM_VAL]] to %[[REF]] : !fir.ref<f32>
-!CHECK:   omp.yield(%[[REF]] : !fir.ref<f32>)
-!CHECK: combiner
-!CHECK: ^bb0(%[[ARG0:.*]]: !fir.ref<f32>, %[[ARG1:.*]]: !fir.ref<f32>):
-!CHECK:   %[[LD0:.*]] = fir.load %[[ARG0]] : !fir.ref<f32>
-!CHECK:   %[[LD1:.*]] = fir.load %[[ARG1]] : !fir.ref<f32>
-!CHECK:   %[[RES:.*]] = arith.minnumf %[[LD0]], %[[LD1]] {{.*}}: f32
-!CHECK:   fir.store %[[RES]] to %[[ARG0]] : !fir.ref<f32>
-!CHECK:   omp.yield(%[[ARG0]] : !fir.ref<f32>)
-
-!CHECK-LABEL: omp.declare_reduction @min_byref_i32 : !fir.ref<i32>
-!CHECK-SAME: init {
-!CHECK:   %[[MAXIMUM_VAL:.*]] = arith.constant 2147483647 : i32
-!CHECK:   fir.store %[[MAXIMUM_VAL]] to %[[REF]] : !fir.ref<i32>
-!CHECK:   omp.yield(%[[REF]] : !fir.ref<i32>)
-!CHECK: combiner
-!CHECK: ^bb0(%[[ARG0:.*]]: !fir.ref<i32>, %[[ARG1:.*]]: !fir.ref<i32>):
-!CHECK:   %[[LD0:.*]] = fir.load %[[ARG0]] : !fir.ref<i32>
-!CHECK:   %[[LD1:.*]] = fir.load %[[ARG1]] : !fir.ref<i32>
-!CHECK:   %[[RES:.*]] = arith.minsi %[[LD0]], %[[LD1]] : i32
-!CHECK:   fir.store %[[RES]] to %[[ARG0]] : !fir.ref<i32>
-!CHECK:   omp.yield(%[[ARG0]] : !fir.ref<i32>)
-
-!CHECK-LABEL: @_QPreduction_min_int
-!CHECK-SAME: %[[Y_BOX:.*]]: !fir.box<!fir.array<?xi32>>
-!CHECK:   %[[X_REF:.*]] = fir.alloca i32 {bindc_name = "x", uniq_name = "_QFreduction_min_intEx"}
-!CHECK:   omp.parallel
-!CHECK:     omp.wsloop byref reduction(@min_byref_i32 %[[X_REF]] -> %[[PRV:.+]] : !fir.ref<i32>) for
-!CHECK:       %[[LPRV:.+]] = fir.load %[[PRV]] : !fir.ref<i32>
-!CHECK:       %[[Y_I_REF:.*]] = fir.coordinate_of %[[Y_BOX]]
-!CHECK:       %[[Y_I:.*]] = fir.load %[[Y_I_REF]] : !fir.ref<i32>
-!CHECK:       %[[RES:.+]] = arith.cmpi slt, %[[LPRV]], %[[Y_I]] : i32
-!CHECK:       %[[SEL:.+]] = arith.select %[[RES]], %[[LPRV]], %[[Y_I]]
-!CHECK:       fir.store %[[SEL]] to %[[PRV]] : !fir.ref<i32>
-!CHECK:       omp.yield
-!CHECK:     omp.terminator
-
-!CHECK-LABEL: @_QPreduction_min_real
-!CHECK-SAME: %[[Y_BOX:.*]]: !fir.box<!fir.array<?xf32>>
-!CHECK:   %[[X_REF:.*]] = fir.alloca f32 {bindc_name = "x", uniq_name = "_QFreduction_min_realEx"}
-!CHECK:   omp.parallel
-!CHECK:     omp.wsloop byref reduction(@min_byref_f32 %[[X_REF]] -> %[[PRV:.+]] : !fir.ref<f32>) for
-!CHECK:       %[[LPRV:.+]] = fir.load %[[PRV]] : !fir.ref<f32>
-!CHECK:       %[[Y_I_REF:.*]] = fir.coordinate_of %[[Y_BOX]]
-!CHECK:       %[[Y_I:.*]] = fir.load %[[Y_I_REF]] : !fir.ref<f32>
-!CHECK:       %[[RES:.+]] = arith.cmpf ogt, %[[Y_I]], %[[LPRV]] {{.*}} : f32
-!CHECK:       omp.yield
-!CHECK:     omp.terminator
-
-subroutine reduction_min_int(y)
-  integer :: x, y(:)
-  x = 0
-  !$omp parallel
-  !$omp do reduction(min:x)
-  do i=1, 100
-    x = min(x, y(i))
-  end do
-  !$omp end do
-  !$omp end parallel
-  print *, x
-end subroutine
-
-subroutine reduction_min_real(y)
-  real :: x, y(:)
-  x = 0.0
-  !$omp parallel
-  !$omp do reduction(min:x)
-  do i=1, 100
-    x = min(y(i), x)
-  end do
-  !$omp end do
-  !$omp end parallel
-  print *, x
-
-  !$omp parallel
-  !$omp do reduction(min:x)
-  do i=1, 100
-  !CHECK-NOT: omp.reduction
-    if (y(i) .gt. x) x = y(i)
-  end do
-  !$omp end do
-  !$omp end parallel
-  print *, x
-end subroutine
diff --git a/flang/test/Lower/OpenMP/FIR/wsloop-reduction-min.f90 b/flang/test/Lower/OpenMP/FIR/wsloop-reduction-min.f90
deleted file mode 100644
index ab33e180ed88..000000000000
--- a/flang/test/Lower/OpenMP/FIR/wsloop-reduction-min.f90
+++ /dev/null
@@ -1,80 +0,0 @@
-! RUN: bbc -emit-fir -hlfir=false -fopenmp -o - %s 2>&1 | FileCheck %s
-! RUN: %flang_fc1 -emit-fir -flang-deprecated-no-hlfir -fopenmp -o - %s 2>&1 | FileCheck %s
-
-!CHECK: omp.declare_reduction @[[MIN_DECLARE_F:.*]] : f32 init {
-!CHECK:   %[[MAXIMUM_VAL_F:.*]] = arith.constant 3.40282347E+38 : f32
-!CHECK:   omp.yield(%[[MAXIMUM_VAL_F]] : f32)
-!CHECK: combiner
-!CHECK: ^bb0(%[[ARG0_F:.*]]: f32, %[[ARG1_F:.*]]: f32):
-!CHECK:   %[[COMB_VAL_F:.*]] = arith.minnumf %[[ARG0_F]], %[[ARG1_F]] {{.*}}: f32
-!CHECK:   omp.yield(%[[COMB_VAL_F]] : f32)
-
-!CHECK: omp.declare_reduction @[[MIN_DECLARE_I:.*]] : i32 init {
-!CHECK:   %[[MAXIMUM_VAL_I:.*]] = arith.constant 2147483647 : i32
-!CHECK:   omp.yield(%[[MAXIMUM_VAL_I]] : i32)
-!CHECK: combiner
-!CHECK: ^bb0(%[[ARG0_I:.*]]: i32, %[[ARG1_I:.*]]: i32):
-!CHECK:   %[[COMB_VAL_I:.*]] = arith.minsi %[[ARG0_I]], %[[ARG1_I]] : i32
-!CHECK:   omp.yield(%[[COMB_VAL_I]] : i32)
-
-!CHECK-LABEL: @_QPreduction_min_int
-!CHECK-SAME: %[[Y_BOX:.*]]: !fir.box<!fir.array<?xi32>>
-!CHECK:   %[[X_REF:.*]] = fir.alloca i32 {bindc_name = "x", uniq_name = "_QFreduction_min_intEx"}
-!CHECK:   omp.parallel
-!CHECK:     omp.wsloop reduction(@[[MIN_DECLARE_I]] %[[X_REF]] -> %[[PRV:.+]] : !fir.ref<i32>) for
-!CHECK:       %[[LPRV:.+]] = fir.load %[[PRV]] : !fir.ref<i32>
-!CHECK:       %[[Y_I_REF:.*]] = fir.coordinate_of %[[Y_BOX]]
-!CHECK:       %[[Y_I:.*]] = fir.load %[[Y_I_REF]] : !fir.ref<i32>
-!CHECK:       %[[RES:.+]] = arith.cmpi slt, %[[LPRV]], %[[Y_I]] : i32
-!CHECK:       %[[SEL:.+]] = arith.select %[[RES]], %[[LPRV]], %[[Y_I]]
-!CHECK:       fir.store %[[SEL]] to %[[PRV]] : !fir.ref<i32>
-!CHECK:       omp.yield
-!CHECK:     omp.terminator
-
-!CHECK-LABEL: @_QPreduction_min_real
-!CHECK-SAME: %[[Y_BOX:.*]]: !fir.box<!fir.array<?xf32>>
-!CHECK:   %[[X_REF:.*]] = fir.alloca f32 {bindc_name = "x", uniq_name = "_QFreduction_min_realEx"}
-!CHECK:   omp.parallel
-!CHECK:     omp.wsloop reduction(@[[MIN_DECLARE_F]] %[[X_REF]] -> %[[PRV:.+]] : !fir.ref<f32>) for
-!CHECK:       %[[LPRV:.+]] = fir.load %[[PRV]] : !fir.ref<f32>
-!CHECK:       %[[Y_I_REF:.*]] = fir.coordinate_of %[[Y_BOX]]
-!CHECK:       %[[Y_I:.*]] = fir.load %[[Y_I_REF]] : !fir.ref<f32>
-!CHECK:       %[[RES:.+]] = arith.cmpf ogt, %[[Y_I]], %[[LPRV]] {{.*}} : f32
-!CHECK:       omp.yield
-!CHECK:     omp.terminator
-
-subroutine reduction_min_int(y)
-  integer :: x, y(:)
-  x = 0
-  !$omp parallel
-  !$omp do reduction(min:x)
-  do i=1, 100
-    x = min(x, y(i))
-  end do
-  !$omp end do
-  !$omp end parallel
-  print *, x
-end subroutine
-
-subroutine reduction_min_real(y)
-  real :: x, y(:)
-  x = 0.0
-  !$omp parallel
-  !$omp do reduction(min:x)
-  do i=1, 100
-    x = min(y(i), x)
-  end do
-  !$omp end do
-  !$omp end parallel
-  print *, x
-
-  !$omp parallel
-  !$omp do reduction(min:x)
-  do i=1, 100
-  !CHECK-NOT: omp.reduction
-    if (y(i) .gt. x) x = y(i)
-  end do
-  !$omp end do
-  !$omp end parallel
-  print *, x
-end subroutine
diff --git a/flang/test/Lower/OpenMP/FIR/wsloop-simd.f90 b/flang/test/Lower/OpenMP/FIR/wsloop-simd.f90
deleted file mode 100644
index 2e3f8ca3c207..000000000000
--- a/flang/test/Lower/OpenMP/FIR/wsloop-simd.f90
+++ /dev/null
@@ -1,34 +0,0 @@
-! This test checks lowering of OpenMP DO Directive(Worksharing) with
-! simd schedule modifier.
-
-! RUN: bbc -fopenmp -emit-fir -hlfir=false %s -o - | FileCheck %s
-
-program wsloop_dynamic
-  integer :: i
-!CHECK-LABEL: func @_QQmain()
-
-!$OMP PARALLEL
-!CHECK:  omp.parallel {
-
-!$OMP DO SCHEDULE(simd: runtime)
-!CHECK:     %[[WS_LB:.*]] = arith.constant 1 : i32
-!CHECK:     %[[WS_UB:.*]] = arith.constant 9 : i32
-!CHECK:     %[[WS_STEP:.*]] = arith.constant 1 : i32
-!CHECK:     omp.wsloop schedule(runtime, simd) nowait for (%[[I:.*]]) : i32 = (%[[WS_LB]]) to (%[[WS_UB]]) inclusive step (%[[WS_STEP]])
-!CHECK:       fir.store %[[I]] to %[[STORE:.*]] : !fir.ref<i32>
-
-  do i=1, 9
-    print*, i
-!CHECK:    %[[RTBEGIN:.*]] = fir.call @_FortranAioBeginExternalListOutput
-!CHECK:    %[[LOAD:.*]] = fir.load %[[STORE]] : !fir.ref<i32>
-!CHECK:    fir.call @_FortranAioOutputInteger32(%[[RTBEGIN]], %[[LOAD]]) {{.*}}: (!fir.ref<i8>, i32) -> i1
-!CHECK:    fir.call @_FortranAioEndIoStatement(%[[RTBEGIN]]) {{.*}}: (!fir.ref<i8>) -> i32
-  end do
-!CHECK:       omp.yield
-!CHECK:         }
-!CHECK:       omp.terminator
-!CHECK:     }
-
-!$OMP END DO NOWAIT
-!$OMP END PARALLEL
-end
diff --git a/flang/test/Lower/OpenMP/FIR/wsloop-variable.f90 b/flang/test/Lower/OpenMP/FIR/wsloop-variable.f90
deleted file mode 100644
index 4f34f30f3e7c..000000000000
--- a/flang/test/Lower/OpenMP/FIR/wsloop-variable.f90
+++ /dev/null
@@ -1,173 +0,0 @@
-! This test checks lowering of OpenMP DO Directive(Worksharing) for different
-! types of loop iteration variable, lower bound, upper bound, and step.
-
-!REQUIRES: shell
-!RUN: bbc -fopenmp -emit-fir -hlfir=false %s -o - 2>&1 | FileCheck %s
-
-!CHECK:  OpenMP loop iteration variable cannot have more than 64 bits size and will be narrowed into 64 bits.
-
-program wsloop_variable
-  integer(kind=1) :: i1_lb, i1_ub
-  integer(kind=2) :: i2, i2_ub, i2_s
-  integer(kind=4) :: i4_s
-  integer(kind=8) :: i8, i8_s
-  integer(kind=16) :: i16, i16_lb
-  real :: x
-
-!CHECK:  %[[TMP0:.*]] = arith.constant 1 : i32
-!CHECK:  %[[TMP1:.*]] = arith.constant 100 : i32
-!CHECK:  %[[TMP2:.*]] = fir.convert %[[TMP0]] : (i32) -> i64
-!CHECK:  %[[TMP3:.*]] = fir.convert %{{.*}} : (i8) -> i64
-!CHECK:  %[[TMP4:.*]] = fir.convert %{{.*}} : (i16) -> i64
-!CHECK:  %[[TMP5:.*]] = fir.convert %{{.*}} : (i128) -> i64
-!CHECK:  %[[TMP6:.*]] = fir.convert %[[TMP1]] : (i32) -> i64
-!CHECK:  %[[TMP7:.*]] = fir.convert %{{.*}} : (i32) -> i64
-!CHECK:  omp.wsloop for (%[[ARG0:.*]], %[[ARG1:.*]]) : i64 = (%[[TMP2]], %[[TMP5]]) to (%[[TMP3]], %[[TMP6]]) inclusive step (%[[TMP4]], %[[TMP7]]) {
-!CHECK:    %[[ARG0_I16:.*]] = fir.convert %[[ARG0]] : (i64) -> i16
-!CHECK:    fir.store %[[ARG0_I16]] to %[[STORE_IV0:.*]] : !fir.ref<i16>
-!CHECK:    fir.store %[[ARG1]] to %[[STORE_IV1:.*]] : !fir.ref<i64>
-!CHECK:    %[[LOAD_IV0:.*]] = fir.load %[[STORE_IV0]] : !fir.ref<i16>
-!CHECK:    %[[LOAD_IV0_I64:.*]] = fir.convert %[[LOAD_IV0]] : (i16) -> i64
-!CHECK:    %[[LOAD_IV1:.*]] = fir.load %[[STORE_IV1]] : !fir.ref<i64>
-!CHECK:    %[[TMP10:.*]] = arith.addi %[[LOAD_IV0_I64]], %[[LOAD_IV1]] : i64
-!CHECK:    %[[TMP11:.*]] = fir.convert %[[TMP10]] : (i64) -> f32
-!CHECK:    fir.store %[[TMP11]] to %{{.*}} : !fir.ref<f32>
-!CHECK:    omp.yield
-!CHECK:  }
-
-  !$omp do collapse(2)
-  do i2 = 1, i1_ub, i2_s
-    do i8 = i16_lb, 100, i4_s
-      x = i2 + i8
-    end do
-  end do
-  !$omp end do
-
-!CHECK:  %[[TMP12:.*]] = arith.constant 1 : i32
-!CHECK:  %[[TMP13:.*]] = fir.convert %{{.*}} : (i8) -> i32
-!CHECK:  %[[TMP14:.*]] = fir.convert %{{.*}} : (i64) -> i32
-!CHECK:  omp.wsloop for (%[[ARG0:.*]]) : i32 = (%[[TMP12]]) to (%[[TMP13]]) inclusive step (%[[TMP14]])  {
-!CHECK:    %[[ARG0_I16:.*]] = fir.convert %[[ARG0]] : (i32) -> i16
-!CHECK:    fir.store %[[ARG0_I16]] to %[[STORE3:.*]] : !fir.ref<i16>
-!CHECK:    %[[LOAD3:.*]] = fir.load %[[STORE3]] : !fir.ref<i16>
-!CHECK:    %[[TMP16:.*]] = fir.convert %[[LOAD3]] : (i16) -> f32
-
-!CHECK:    fir.store %[[TMP16]] to %{{.*}} : !fir.ref<f32>
-!CHECK:    omp.yield
-!CHECK:  }
-
-  !$omp do
-  do i2 = 1, i1_ub, i8_s
-    x = i2
-  end do
-  !$omp end do
-
-!CHECK:  %[[TMP17:.*]] = fir.convert %{{.*}} : (i8) -> i64
-!CHECK:  %[[TMP18:.*]] = fir.convert %{{.*}} : (i16) -> i64
-!CHECK:  %[[TMP19:.*]] = fir.convert %{{.*}} : (i32) -> i64
-!CHECK:  omp.wsloop for (%[[ARG1:.*]]) : i64 = (%[[TMP17]]) to (%[[TMP18]]) inclusive step (%[[TMP19]])  {
-!CHECK:    %[[ARG1_I128:.*]] = fir.convert %[[ARG1]] : (i64) -> i128
-!CHECK:    fir.store %[[ARG1_I128]] to %[[STORE4:.*]] : !fir.ref<i128>
-!CHECK:    %[[LOAD4:.*]] = fir.load %[[STORE4]] : !fir.ref<i128>
-!CHECK:    %[[TMP21:.*]] = fir.convert %[[LOAD4]] : (i128) -> f32
-!CHECK:    fir.store %[[TMP21]] to %{{.*}} : !fir.ref<f32>
-!CHECK:    omp.yield
-!CHECK:  }
-
-  !$omp do
-  do i16 = i1_lb, i2_ub, i4_s
-    x = i16
-  end do
-  !$omp end do
-
-end program wsloop_variable
-
-!CHECK-LABEL: func.func @_QPwsloop_variable_sub() {
-!CHECK:         %[[IV2:.*]] = fir.alloca i8 {adapt.valuebyref, pinned}
-!CHECK:         %[[VAL_0:.*]] = fir.alloca i128 {bindc_name = "i16_lb", uniq_name = "_QFwsloop_variable_subEi16_lb"}
-!CHECK:         %[[VAL_1:.*]] = fir.alloca i8 {bindc_name = "i1_ub", uniq_name = "_QFwsloop_variable_subEi1_ub"}
-!CHECK:         %[[VAL_2:.*]] = fir.alloca i16 {bindc_name = "i2", uniq_name = "_QFwsloop_variable_subEi2"}
-!CHECK:         %[[VAL_3:.*]] = fir.alloca i16 {bindc_name = "i2_s", uniq_name = "_QFwsloop_variable_subEi2_s"}
-!CHECK:         %[[VAL_4:.*]] = fir.alloca i32 {bindc_name = "i4_s", uniq_name = "_QFwsloop_variable_subEi4_s"}
-!CHECK:         %[[VAL_5:.*]] = fir.alloca i64 {bindc_name = "i8", uniq_name = "_QFwsloop_variable_subEi8"}
-!CHECK:         %[[J1:.*]] = fir.alloca i8 {bindc_name = "j1", uniq_name = "_QFwsloop_variable_subEj1"}
-!CHECK:         %[[VAL_6:.*]] = fir.alloca f32 {bindc_name = "x", uniq_name = "_QFwsloop_variable_subEx"}
-!CHECK:         %[[VAL_7:.*]] = arith.constant 1 : i32
-!CHECK:         %[[VAL_8:.*]] = fir.load %[[VAL_1]] : !fir.ref<i8>
-!CHECK:         %[[VAL_9:.*]] = fir.load %[[VAL_3]] : !fir.ref<i16>
-!CHECK:         %[[VAL_10:.*]] = fir.convert %[[VAL_8]] : (i8) -> i32
-!CHECK:         %[[VAL_11:.*]] = fir.convert %[[VAL_9]] : (i16) -> i32
-!CHECK:         omp.wsloop   for  (%[[ARG0:.*]]) : i32 = (%[[VAL_7]]) to (%[[VAL_10]]) inclusive step (%[[VAL_11]]) {
-!CHECK:           %[[ARG0_I16:.*]] = fir.convert %[[ARG0]] : (i32) -> i16
-!CHECK:           fir.store %[[ARG0_I16]] to %[[STORE_IV:.*]] : !fir.ref<i16>
-!CHECK:           %[[VAL_13:.*]] = fir.load %[[VAL_0]] : !fir.ref<i128>
-!CHECK:           %[[VAL_14:.*]] = fir.convert %[[VAL_13]] : (i128) -> index
-!CHECK:           %[[VAL_15:.*]] = arith.constant 100 : i32
-!CHECK:           %[[VAL_16:.*]] = fir.convert %[[VAL_15]] : (i32) -> index
-!CHECK:           %[[VAL_17:.*]] = fir.load %[[VAL_4]] : !fir.ref<i32>
-!CHECK:           %[[VAL_18:.*]] = fir.convert %[[VAL_17]] : (i32) -> index
-!CHECK:           %[[LB:.*]] = fir.convert %[[VAL_14]] : (index) -> i64
-!CHECK:           %[[VAL_19:.*]]:2 = fir.do_loop %[[VAL_20:[^ ]*]] =
-!CHECK-SAME:          %[[VAL_14]] to %[[VAL_16]] step %[[VAL_18]]
-!CHECK-SAME:          iter_args(%[[IV:.*]] = %[[LB]]) -> (index, i64) {
-!CHECK:             fir.store %[[IV]] to %[[VAL_5]] : !fir.ref<i64>
-!CHECK:             %[[LOAD_IV:.*]] = fir.load %[[STORE_IV]] : !fir.ref<i16>
-!CHECK:             %[[VAL_22:.*]] = fir.convert %[[LOAD_IV]] : (i16) -> i64
-!CHECK:             %[[VAL_23:.*]] = fir.load %[[VAL_5]] : !fir.ref<i64>
-!CHECK:             %[[VAL_24:.*]] = arith.addi %[[VAL_22]], %[[VAL_23]] : i64
-!CHECK:             %[[VAL_25:.*]] = fir.convert %[[VAL_24]] : (i64) -> f32
-!CHECK:             fir.store %[[VAL_25]] to %[[VAL_6]] : !fir.ref<f32>
-!CHECK:             %[[VAL_26:.*]] = arith.addi %[[VAL_20]], %[[VAL_18]] : index
-!CHECK:             %[[STEPCAST:.*]] = fir.convert %[[VAL_18]] : (index) -> i64
-!CHECK:             %[[IVLOAD:.*]] = fir.load %[[VAL_5]] : !fir.ref<i64>
-!CHECK:             %[[IVINC:.*]] = arith.addi %[[IVLOAD]], %[[STEPCAST]]
-!CHECK:             fir.result %[[VAL_26]], %[[IVINC]] : index, i64
-!CHECK:           }
-!CHECK:           fir.store %[[VAL_19]]#1 to %[[VAL_5]] : !fir.ref<i64>
-!CHECK:           omp.yield
-!CHECK:         }
-
-subroutine wsloop_variable_sub
-  integer(kind=1) :: i1, i1_ub, j1
-  integer(kind=2) :: i2, i2_s
-  integer(kind=4) :: i4_s
-  integer(kind=8) :: i8
-  integer(kind=16) :: i16_lb
-  real :: x
-
-  !$omp do
-  do i2 = 1, i1_ub, i2_s
-    do i8 = i16_lb, 100, i4_s
-      x = i2 + i8
-    end do
-  end do
-  !$omp end do
-
-!CHECK:         %[[C1:.*]] = arith.constant 1 : i32
-!CHECK:         %[[C10:.*]] = arith.constant 10 : i32
-!CHECK:         %[[C1_2:.*]] = arith.constant 1 : i32
-!CHECK:         omp.wsloop for (%[[ARG0:.*]]) : i32 = (%[[C1]]) to (%[[C10]]) inclusive step (%[[C1_2]]) {
-!CHECK:           %[[ARG0_I8:.*]] = fir.convert %[[ARG0]] : (i32) -> i8
-!CHECK:           fir.store %[[ARG0_I8]] to %[[IV2]] : !fir.ref<i8>
-!CHECK:           %[[IV2LOAD:.*]] = fir.load %[[IV2]] : !fir.ref<i8>
-!CHECK:           %[[J1LOAD:.*]] = fir.load %[[J1]] : !fir.ref<i8>
-!CHECK:           %[[VAL_27:.*]] = arith.cmpi eq, %[[IV2LOAD]], %[[J1LOAD]] : i8
-!CHECK:           fir.if %[[VAL_27]] {
-!CHECK:           } else {
-!CHECK:           }
-!CHECK:           omp.yield
-!CHECK:         }
-
-  j1 = 5
-  !$omp do
-  do i1 = 1, 10
-    if (i1 .eq. j1) then
-      print *, "EQ"
-    end if
-  end do
-  !$omp end do
-
-!CHECK:         return
-!CHECK:       }
-
-end
diff --git a/flang/test/Lower/OpenMP/FIR/wsloop.f90 b/flang/test/Lower/OpenMP/FIR/wsloop.f90
deleted file mode 100644
index abc0489b08ff..000000000000
--- a/flang/test/Lower/OpenMP/FIR/wsloop.f90
+++ /dev/null
@@ -1,72 +0,0 @@
-! This test checks lowering of OpenMP DO Directive (Worksharing).
-
-! RUN: bbc -fopenmp -emit-fir -hlfir=false %s -o - | FileCheck %s
-
-!CHECK-LABEL: func @_QPsimple_loop()
-subroutine simple_loop
-  integer :: i
-  ! CHECK:  omp.parallel
-  !$OMP PARALLEL
-  ! CHECK:     %[[ALLOCA_IV:.*]] = fir.alloca i32 {{{.*}}, pinned}
-  ! CHECK:     %[[WS_LB:.*]] = arith.constant 1 : i32
-  ! CHECK:     %[[WS_UB:.*]] = arith.constant 9 : i32
-  ! CHECK:     %[[WS_STEP:.*]] = arith.constant 1 : i32
-  ! CHECK:     omp.wsloop for (%[[I:.*]]) : i32 = (%[[WS_LB]]) to (%[[WS_UB]]) inclusive step (%[[WS_STEP]])
-  !$OMP DO
-  do i=1, 9
-  ! CHECK:             fir.store %[[I]] to %[[ALLOCA_IV:.*]] : !fir.ref<i32>
-  ! CHECK:             %[[LOAD_IV:.*]] = fir.load %[[ALLOCA_IV]] : !fir.ref<i32>
-  ! CHECK:    fir.call @_FortranAioOutputInteger32({{.*}}, %[[LOAD_IV]]) {{.*}}: (!fir.ref<i8>, i32) -> i1
-    print*, i
-  end do
-  ! CHECK:       omp.yield
-  !$OMP END DO
-  ! CHECK:       omp.terminator
-  !$OMP END PARALLEL
-end subroutine
-
-!CHECK-LABEL: func @_QPsimple_loop_with_step()
-subroutine simple_loop_with_step
-  integer :: i
-  ! CHECK:  omp.parallel
-  !$OMP PARALLEL
-  ! CHECK:     %[[ALLOCA_IV:.*]] = fir.alloca i32 {{{.*}}, pinned}
-  ! CHECK:     %[[WS_LB:.*]] = arith.constant 1 : i32
-  ! CHECK:     %[[WS_UB:.*]] = arith.constant 9 : i32
-  ! CHECK:     %[[WS_STEP:.*]] = arith.constant 2 : i32
-  ! CHECK:     omp.wsloop for (%[[I:.*]]) : i32 = (%[[WS_LB]]) to (%[[WS_UB]]) inclusive step (%[[WS_STEP]])
-  ! CHECK:       fir.store %[[I]] to %[[ALLOCA_IV]] : !fir.ref<i32>
-  ! CHECK:       %[[LOAD_IV:.*]] = fir.load %[[ALLOCA_IV]] : !fir.ref<i32>
-  !$OMP DO
-  do i=1, 9, 2
-  ! CHECK:    fir.call @_FortranAioOutputInteger32({{.*}}, %[[LOAD_IV]]) {{.*}}: (!fir.ref<i8>, i32) -> i1
-    print*, i
-  end do
-  ! CHECK:       omp.yield
-  !$OMP END DO
-  ! CHECK:       omp.terminator
-  !$OMP END PARALLEL
-end subroutine
-
-!CHECK-LABEL: func @_QPloop_with_schedule_nowait()
-subroutine loop_with_schedule_nowait
-  integer :: i
-  ! CHECK:  omp.parallel
-  !$OMP PARALLEL
-  ! CHECK:     %[[ALLOCA_IV:.*]] = fir.alloca i32 {{{.*}}, pinned}
-  ! CHECK:     %[[WS_LB:.*]] = arith.constant 1 : i32
-  ! CHECK:     %[[WS_UB:.*]] = arith.constant 9 : i32
-  ! CHECK:     %[[WS_STEP:.*]] = arith.constant 1 : i32
-  ! CHECK:     omp.wsloop schedule(runtime) nowait for (%[[I:.*]]) : i32 = (%[[WS_LB]]) to (%[[WS_UB]]) inclusive step (%[[WS_STEP]])
-  !$OMP DO SCHEDULE(runtime)
-  do i=1, 9
-  ! CHECK:       fir.store %[[I]] to %[[ALLOCA_IV]] : !fir.ref<i32>
-  ! CHECK:       %[[LOAD_IV:.*]] = fir.load %[[ALLOCA_IV]] : !fir.ref<i32>
-  ! CHECK:    fir.call @_FortranAioOutputInteger32({{.*}}, %[[LOAD_IV]]) {{.*}}: (!fir.ref<i8>, i32) -> i1
-    print*, i
-  end do
-  ! CHECK:       omp.yield
-  !$OMP END DO NOWAIT
-  ! CHECK:       omp.terminator
-  !$OMP END PARALLEL
-end subroutine
diff --git a/flang/test/Lower/OpenMP/Todo/omp-default-clause-inner-loop.f90 b/flang/test/Lower/OpenMP/Todo/omp-default-clause-inner-loop.f90
index 5c624d31b5f3..c245137f16c7 100644
--- a/flang/test/Lower/OpenMP/Todo/omp-default-clause-inner-loop.f90
+++ b/flang/test/Lower/OpenMP/Todo/omp-default-clause-inner-loop.f90
@@ -12,7 +12,8 @@
 ! CHECK: %[[const_1:.*]] = arith.constant 1 : i32
 ! CHECK: %[[const_2:.*]] = arith.constant 10 : i32
 ! CHECK: %[[const_3:.*]] = arith.constant 1 : i32
-! CHECK: omp.wsloop   for  (%[[ARG:.*]]) : i32 = (%[[const_1]]) to (%[[const_2]]) inclusive step (%[[const_3]]) {
+! CHECK: omp.wsloop {
+! CHECK-NEXT: omp.loop_nest (%[[ARG:.*]]) : i32 = (%[[const_1]]) to (%[[const_2]]) inclusive step (%[[const_3]]) {
 ! CHECK: fir.store %[[ARG]] to %[[TEMP]] : !fir.ref<i32>
 ! EXPECTED: %[[temp_1:.*]] = fir.load %[[PRIVATE_Z]] : !fir.ref<i32>
 ! CHECK: %[[temp_1:.*]] = fir.load %{{.*}} : !fir.ref<i32>
@@ -24,6 +25,8 @@
 ! CHECK: }
 ! CHECK: omp.terminator
 ! CHECK: }
+! CHECK: omp.terminator
+! CHECK: }
 subroutine nested_default_clause()
 	integer x, y, z
 	!$omp parallel do default(private)
diff --git a/flang/test/Lower/OpenMP/Todo/reduction-allocatable.f90 b/flang/test/Lower/OpenMP/Todo/reduction-allocatable.f90
deleted file mode 100644
index 09aba6920232..000000000000
--- a/flang/test/Lower/OpenMP/Todo/reduction-allocatable.f90
+++ /dev/null
@@ -1,21 +0,0 @@
-! RUN: %not_todo_cmd bbc -emit-fir -fopenmp -o - %s 2>&1 | FileCheck %s
-! RUN: %not_todo_cmd %flang_fc1 -emit-fir -fopenmp -o - %s 2>&1 | FileCheck %s
-
-! CHECK: not yet implemented: Reduction of some types is not supported
-subroutine reduction_allocatable
-  integer, allocatable :: x
-  integer :: i = 1
-
-  allocate(x)
-  x = 0
-
-  !$omp parallel num_threads(4)
-  !$omp do reduction(+:x)
-  do i = 1, 10
-    x = x + i
-  enddo
-  !$omp end do
-  !$omp end parallel
-
-  print *, x
-end subroutine
diff --git a/flang/test/Lower/OpenMP/cfg-conversion-omp.private.f90 b/flang/test/Lower/OpenMP/cfg-conversion-omp.private.f90
new file mode 100644
index 000000000000..7f1087a7ebe3
--- /dev/null
+++ b/flang/test/Lower/OpenMP/cfg-conversion-omp.private.f90
@@ -0,0 +1,54 @@
+! Tests that CFG & LLVM conversion is applied to `omp.private` ops.
+
+! RUN: split-file %s %t && cd %t
+
+! RUN: %flang_fc1 -emit-hlfir -fopenmp -mmlir --openmp-enable-delayed-privatization \
+! RUN:   -o - test.f90 2>&1 | \
+! RUN: fir-opt --cfg-conversion -o test.cfg-conv.mlir
+! RUN: FileCheck --input-file=test.cfg-conv.mlir %s --check-prefix="CFGConv"
+
+! RUN: fir-opt --convert-hlfir-to-fir --cg-rewrite --fir-to-llvm-ir test.cfg-conv.mlir -o - | \
+! RUN: FileCheck %s --check-prefix="LLVMDialect"
+
+!--- test.f90
+subroutine delayed_privatization_allocatable
+  implicit none
+  integer, allocatable :: var1
+
+!$omp parallel private(var1)
+  var1 = 10
+!$omp end parallel
+end subroutine
+
+! CFGConv-LABEL: omp.private {type = private}
+! CFGConv-SAME: @[[PRIVATIZER_SYM:.*]] : [[TYPE:!fir.ref<!fir.box<!fir.heap<i32>>>]] alloc {
+
+! CFGConv-NEXT: ^bb0(%[[PRIV_ARG:.*]]: [[TYPE]]):
+
+! CFGConv-NEXT:   %[[PRIV_ALLOC:.*]] = fir.alloca !fir.box<!fir.heap<i32>> {bindc_name = "var1", pinned, uniq_name = "_QFdelayed_privatization_allocatableEvar1"}
+
+! CFGConv-NEXT:   %[[PRIV_ARG_VAL:.*]] = fir.load %[[PRIV_ARG]] : !fir.ref<!fir.box<!fir.heap<i32>>>
+! CFGConv-NEXT:   %[[PRIV_ARG_BOX:.*]] = fir.box_addr %[[PRIV_ARG_VAL]] : (!fir.box<!fir.heap<i32>>) -> !fir.heap<i32>
+! CFGConv-NEXT:   %[[PRIV_ARG_ADDR:.*]] = fir.convert %[[PRIV_ARG_BOX]] : (!fir.heap<i32>) -> i64
+! CFGConv-NEXT:   %[[C0:.*]] = arith.constant 0 : i64
+! CFGConv-NEXT:   %[[ALLOC_COND:.*]] = arith.cmpi ne, %[[PRIV_ARG_ADDR]], %[[C0]] : i64
+! CFGConv-NEXT:   cf.cond_br %[[ALLOC_COND]], ^[[ALLOC_MEM_BB:.*]], ^[[ZERO_MEM_BB:.*]]
+! CFGConv-NEXT: ^[[ALLOC_MEM_BB]]:
+! CFGConv-NEXT:   fir.allocmem
+! CFGConv:        cf.br ^[[DECL_BB:.*]]
+! CFGConv:      ^[[ZERO_MEM_BB]]:
+! CFGConv-NEXT:   fir.zero_bits
+! CFGConv:        cf.br ^[[DECL_BB:.*]]
+! CFGConv-NEXT: ^[[DECL_BB]]:
+! CFGConv-NEXT:   hlfir.declare
+! CFGConv-NEXT:   omp.yield
+
+
+! LLVMDialect-LABEL: omp.private {type = private}
+! LLVMDialect-SAME: @[[PRIVATIZER_SYM:.*]] : [[TYPE:!llvm.ptr]] alloc {
+
+! LLVMDialect-NEXT: ^bb0(%[[PRIV_ARG:.*]]: [[TYPE]]):
+! LLVMDialect:        llvm.alloca
+! LLVMDialect:        llvm.call @malloc
+
+! LLVMDialect-NOT:    hlfir.declare
diff --git a/flang/test/Lower/OpenMP/copyin.f90 b/flang/test/Lower/OpenMP/copyin.f90
index 895e1abd274f..dda563303148 100644
--- a/flang/test/Lower/OpenMP/copyin.f90
+++ b/flang/test/Lower/OpenMP/copyin.f90
@@ -156,10 +156,13 @@ end
 ! CHECK:             %[[VAL_11:.*]] = arith.constant 1 : i32
 ! CHECK:             %[[VAL_12:.*]] = fir.load %[[VAL_9]]#0 : !fir.ref<i32>
 ! CHECK:             %[[VAL_13:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop for  (%[[VAL_14:.*]]) : i32 = (%[[VAL_11]]) to (%[[VAL_12]]) inclusive step (%[[VAL_13]]) {
-! CHECK:               fir.store %[[VAL_14]] to %[[VAL_7]]#1 : !fir.ref<i32>
-! CHECK:               fir.call @_QPsub4(%[[VAL_9]]#1) fastmath<contract> : (!fir.ref<i32>) -> ()
-! CHECK:               omp.yield
+! CHECK:             omp.wsloop {
+! CHECK-NEXT:          omp.loop_nest (%[[VAL_14:.*]]) : i32 = (%[[VAL_11]]) to (%[[VAL_12]]) inclusive step (%[[VAL_13]]) {
+! CHECK:                 fir.store %[[VAL_14]] to %[[VAL_7]]#1 : !fir.ref<i32>
+! CHECK:                 fir.call @_QPsub4(%[[VAL_9]]#1) fastmath<contract> : (!fir.ref<i32>) -> ()
+! CHECK:                 omp.yield
+! CHECK:               }
+! CHECK:               omp.terminator
 ! CHECK:             }
 ! CHECK:             omp.terminator
 ! CHECK:           }
@@ -320,13 +323,16 @@ end subroutine
 ! CHECK:             %[[VAL_34:.*]] = arith.constant 1 : i32
 ! CHECK:             %[[VAL_35:.*]] = fir.load %[[VAL_26]]#0 : !fir.ref<i32>
 ! CHECK:             %[[VAL_36:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop for  (%[[VAL_37:.*]]) : i32 = (%[[VAL_34]]) to (%[[VAL_35]]) inclusive step (%[[VAL_36]]) {
-! CHECK:               fir.store %[[VAL_37]] to %[[VAL_20]]#1 : !fir.ref<i32>
-! CHECK:               %[[VAL_38:.*]] = fir.load %[[VAL_31]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_39:.*]] = fir.load %[[VAL_20]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_40:.*]] = arith.addi %[[VAL_38]], %[[VAL_39]] : i32
-! CHECK:               hlfir.assign %[[VAL_40]] to %[[VAL_31]]#0 : i32, !fir.ref<i32>
-! CHECK:               omp.yield
+! CHECK:             omp.wsloop {
+! CHECK-NEXT:          omp.loop_nest (%[[VAL_37:.*]]) : i32 = (%[[VAL_34]]) to (%[[VAL_35]]) inclusive step (%[[VAL_36]]) {
+! CHECK:                 fir.store %[[VAL_37]] to %[[VAL_20]]#1 : !fir.ref<i32>
+! CHECK:                 %[[VAL_38:.*]] = fir.load %[[VAL_31]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_39:.*]] = fir.load %[[VAL_20]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_40:.*]] = arith.addi %[[VAL_38]], %[[VAL_39]] : i32
+! CHECK:                 hlfir.assign %[[VAL_40]] to %[[VAL_31]]#0 : i32, !fir.ref<i32>
+! CHECK:                 omp.yield
+! CHECK:               }
+! CHECK:               omp.terminator
 ! CHECK:             }
 ! CHECK:             omp.terminator
 ! CHECK:           }
diff --git a/flang/test/Lower/OpenMP/default-clause-byref.f90 b/flang/test/Lower/OpenMP/default-clause-byref.f90
index 1167ba7e6ae0..6a91927ab02d 100644
--- a/flang/test/Lower/OpenMP/default-clause-byref.f90
+++ b/flang/test/Lower/OpenMP/default-clause-byref.f90
@@ -352,11 +352,14 @@ subroutine skipped_default_clause_checks()
        type(it)::iii
 
 !CHECK: omp.parallel {
-!CHECK: omp.wsloop byref reduction(@min_byref_i32 %[[VAL_Z_DECLARE]]#0 -> %[[PRV:.+]] : !fir.ref<i32>) for (%[[ARG:.*]]) {{.*}} {
+!CHECK: omp.wsloop byref reduction(@min_byref_i32 %[[VAL_Z_DECLARE]]#0 -> %[[PRV:.+]] : !fir.ref<i32>) {
+!CHECK-NEXT: omp.loop_nest (%[[ARG:.*]]) {{.*}} {
 !CHECK: omp.yield
 !CHECK: }
 !CHECK: omp.terminator
 !CHECK: }
+!CHECK: omp.terminator
+!CHECK: }
        !$omp parallel do default(private) REDUCTION(MIN:z)
          do i = 1, 10
            x = x + MIN(y,x)
diff --git a/flang/test/Lower/OpenMP/default-clause-implied-do-fix.f90 b/flang/test/Lower/OpenMP/default-clause-implied-do-fix.f90
new file mode 100644
index 000000000000..25579272a6e0
--- /dev/null
+++ b/flang/test/Lower/OpenMP/default-clause-implied-do-fix.f90
@@ -0,0 +1,11 @@
+!RUN: %flang_fc1 -emit-hlfir -fopenmp %s -o - | FileCheck %s
+
+!CHECK: @_QPsb
+subroutine sb(a)
+  integer :: a(:)
+!CHECK: omp.parallel
+  !$omp parallel default(private)
+!CHECK: hlfir.elemental
+    if (any(a/=(/(100,i=1,5)/))) print *, "OK"
+  !$omp end parallel
+end subroutine
diff --git a/flang/test/Lower/OpenMP/default-clause.f90 b/flang/test/Lower/OpenMP/default-clause.f90
index 9a47e561338f..d3c6550821f0 100644
--- a/flang/test/Lower/OpenMP/default-clause.f90
+++ b/flang/test/Lower/OpenMP/default-clause.f90
@@ -352,11 +352,14 @@ subroutine skipped_default_clause_checks()
        type(it)::iii
 
 !CHECK: omp.parallel {
-!CHECK: omp.wsloop reduction(@min_i32 %[[VAL_Z_DECLARE]]#0 -> %[[PRV:.+]] : !fir.ref<i32>) for (%[[ARG:.*]]) {{.*}} {
+!CHECK: omp.wsloop reduction(@min_i32 %[[VAL_Z_DECLARE]]#0 -> %[[PRV:.+]] : !fir.ref<i32>) {
+!CHECK-NEXT: omp.loop_nest (%[[ARG:.*]]) {{.*}} {
 !CHECK: omp.yield
 !CHECK: }
 !CHECK: omp.terminator
 !CHECK: }
+!CHECK: omp.terminator
+!CHECK: }
        !$omp parallel do default(private) REDUCTION(MIN:z)
          do i = 1, 10
            x = x + MIN(y,x)
diff --git a/flang/test/Lower/OpenMP/function-filtering-3.f90 b/flang/test/Lower/OpenMP/function-filtering-3.f90
new file mode 100644
index 000000000000..a277c06d6206
--- /dev/null
+++ b/flang/test/Lower/OpenMP/function-filtering-3.f90
@@ -0,0 +1,34 @@
+! RUN: %flang_fc1 -fopenmp -flang-experimental-hlfir -emit-llvm %s -o - | FileCheck --check-prefixes=LLVM-HOST,LLVM-ALL %s
+! RUN: %flang_fc1 -fopenmp -emit-hlfir %s -o - | FileCheck --check-prefixes=MLIR-HOST,MLIR-ALL %s
+! RUN: %flang_fc1 -fopenmp -fopenmp-is-target-device -flang-experimental-hlfir -emit-llvm %s -o - | FileCheck --check-prefixes=LLVM-DEVICE,LLVM-ALL %s
+! RUN: %flang_fc1 -fopenmp -fopenmp-is-target-device -emit-hlfir %s -o - | FileCheck --check-prefixes=MLIR-DEVICE,MLIR-ALL %s
+! RUN: bbc -fopenmp -emit-hlfir %s -o - | FileCheck --check-prefixes=MLIR-HOST,MLIR-ALL %s
+! RUN: bbc -fopenmp -fopenmp-is-target-device -emit-hlfir %s -o - | FileCheck --check-prefixes=MLIR-DEVICE,MLIR-ALL %s
+
+! Check that the correct LLVM IR functions are kept for the host and device
+! after running the whole set of translation and transformation passes from
+! Fortran.
+
+! MLIR-HOST: func.func @{{.*}}host_parent_procedure(
+! MLIR-HOST: return
+! MLIR-DEVICE-NOT: func.func {{.*}}host_parent_procedure(
+
+! LLVM-HOST: define {{.*}} @host_parent_procedure{{.*}}(
+! LLVM-DEVICE-NOT: {{.*}} @{{.*}}_host_parent_procedure{{.*}}(
+subroutine host_parent_procedure(x)
+  integer, intent(out) :: x
+  call target_internal_proc(x)
+contains
+! MLIR-ALL: func.func {{.*}}@_QFhost_parent_procedurePtarget_internal_proc(
+
+! LLVM-HOST: define {{.*}} @_QFhost_parent_procedurePtarget_internal_proc(
+! LLVM-DEVICE-NOT: define {{.*}} @_QFhost_parent_procedurePtarget_internal_proc(
+! LLVM-ALL: define {{.*}} @__omp_offloading_{{.*}}QFhost_parent_procedurePtarget_internal_proc{{.*}}(
+
+subroutine target_internal_proc(x)
+  integer, intent(out) :: x
+  !$omp target map(from:x)
+    x = 10
+  !$omp end target
+end subroutine
+end subroutine
diff --git a/flang/test/Lower/OpenMP/hlfir-wsloop.f90 b/flang/test/Lower/OpenMP/hlfir-wsloop.f90
index b6be77fe3016..fea05ae3d6bc 100644
--- a/flang/test/Lower/OpenMP/hlfir-wsloop.f90
+++ b/flang/test/Lower/OpenMP/hlfir-wsloop.f90
@@ -11,17 +11,19 @@ subroutine simple_loop
   ! CHECK:  omp.parallel
   !$OMP PARALLEL
   ! CHECK-DAG:     %[[ALLOCA_IV:.*]] = fir.alloca i32 {{{.*}}, pinned}
-  ! CHECK:     %[[IV:.*]]    = fir.declare %[[ALLOCA_IV]] {uniq_name = "_QFsimple_loopEi"} : (!fir.ref<i32>) -> !fir.ref<i32>
-  ! CHECK:     omp.wsloop for (%[[I:.*]]) : i32 = (%[[WS_ST]]) to (%[[WS_END]]) inclusive step (%[[WS_ST]])
+  ! CHECK:         %[[IV:.*]]    = fir.declare %[[ALLOCA_IV]] {uniq_name = "_QFsimple_loopEi"} : (!fir.ref<i32>) -> !fir.ref<i32>
+  ! CHECK:         omp.wsloop {
+  ! CHECK-NEXT:      omp.loop_nest (%[[I:.*]]) : i32 = (%[[WS_ST]]) to (%[[WS_END]]) inclusive step (%[[WS_ST]]) {
   !$OMP DO
   do i=1, 9
   ! CHECK:             fir.store %[[I]] to %[[IV:.*]] : !fir.ref<i32>
   ! CHECK:             %[[LOAD_IV:.*]] = fir.load %[[IV]] : !fir.ref<i32>
-  ! CHECK:    fir.call @_FortranAioOutputInteger32({{.*}}, %[[LOAD_IV]]) {{.*}}: (!fir.ref<i8>, i32) -> i1
+  ! CHECK:             fir.call @_FortranAioOutputInteger32({{.*}}, %[[LOAD_IV]]) {{.*}}: (!fir.ref<i8>, i32) -> i1
     print*, i
   end do
-  ! CHECK:       omp.yield
+  ! CHECK:             omp.yield
+  ! CHECK:           omp.terminator
   !$OMP END DO
-  ! CHECK:       omp.terminator
+  ! CHECK:         omp.terminator
   !$OMP END PARALLEL
 end subroutine
diff --git a/flang/test/Lower/OpenMP/if-clause.f90 b/flang/test/Lower/OpenMP/if-clause.f90
index ce4427a0c2ca..7c15c275d8cc 100644
--- a/flang/test/Lower/OpenMP/if-clause.f90
+++ b/flang/test/Lower/OpenMP/if-clause.f90
@@ -1,7 +1,9 @@
 ! This test checks lowering of OpenMP IF clauses.
 
-! RUN: bbc -fopenmp -emit-hlfir %s -o - | FileCheck %s
-! RUN: %flang_fc1 -fopenmp -emit-hlfir %s -o - | FileCheck %s
+! The "if" clause was added to the "simd" directive in OpenMP 5.0, and
+! to the "teams" directive in OpenMP 5.2.
+! RUN: bbc -fopenmp -fopenmp-version=52 -emit-hlfir %s -o - | FileCheck %s
+! RUN: %flang_fc1 -fopenmp -fopenmp-version=52 -emit-hlfir %s -o - | FileCheck %s
 
 program main
   integer :: i
diff --git a/flang/test/Lower/OpenMP/lastprivate-commonblock.f90 b/flang/test/Lower/OpenMP/lastprivate-commonblock.f90
index a11bdee15663..78adf09c6fe3 100644
--- a/flang/test/Lower/OpenMP/lastprivate-commonblock.f90
+++ b/flang/test/Lower/OpenMP/lastprivate-commonblock.f90
@@ -2,35 +2,38 @@
 
 !CHECK: fir.global common @[[CB_C:.*]](dense<0> : vector<8xi8>) : !fir.array<8xi8>
 !CHECK-LABEL: func.func @_QPlastprivate_common
-!CHECK:    %[[CB_C_REF:.*]] = fir.address_of(@[[CB_C]]) : !fir.ref<!fir.array<8xi8>>
-!CHECK:    %[[CB_C_REF_CVT:.*]] = fir.convert %[[CB_C_REF]] : (!fir.ref<!fir.array<8xi8>>) -> !fir.ref<!fir.array<?xi8>>
-!CHECK:    %[[CB_C_X_COOR:.*]] = fir.coordinate_of %[[CB_C_REF_CVT]], %{{.*}} : (!fir.ref<!fir.array<?xi8>>, index) -> !fir.ref<i8>
-!CHECK:    %[[CB_C_X_ADDR:.*]] = fir.convert %[[CB_C_X_COOR]] : (!fir.ref<i8>) -> !fir.ref<f32>
-!CHECK:    %[[X_DECL:.*]]:2 = hlfir.declare %[[CB_C_X_ADDR]] {uniq_name = "_QFlastprivate_commonEx"} : (!fir.ref<f32>) -> (!fir.ref<f32>, !fir.ref<f32>)
-!CHECK:    %[[CB_C_REF_CVT:.*]] = fir.convert %[[CB_C_REF]] : (!fir.ref<!fir.array<8xi8>>) -> !fir.ref<!fir.array<?xi8>>
-!CHECK:    %[[CB_C_Y_COOR:.*]] = fir.coordinate_of %[[CB_C_REF_CVT]], %{{.*}} : (!fir.ref<!fir.array<?xi8>>, index) -> !fir.ref<i8>
-!CHECK:    %[[CB_C_Y_ADDR:.*]] = fir.convert %[[CB_C_Y_COOR]] : (!fir.ref<i8>) -> !fir.ref<f32>
-!CHECK:    %[[Y_DECL:.*]]:2 = hlfir.declare %[[CB_C_Y_ADDR]] {uniq_name = "_QFlastprivate_commonEy"} : (!fir.ref<f32>) -> (!fir.ref<f32>, !fir.ref<f32>)
-!CHECK:    %[[PRIVATE_X_REF:.*]] = fir.alloca f32 {bindc_name = "x", pinned, uniq_name = "_QFlastprivate_commonEx"}
-!CHECK:    %[[PRIVATE_X_DECL:.*]]:2 = hlfir.declare %[[PRIVATE_X_REF]] {uniq_name = "_QFlastprivate_commonEx"} : (!fir.ref<f32>) -> (!fir.ref<f32>, !fir.ref<f32>)
-!CHECK:    %[[PRIVATE_Y_REF:.*]] = fir.alloca f32 {bindc_name = "y", pinned, uniq_name = "_QFlastprivate_commonEy"}
-!CHECK:    %[[PRIVATE_Y_DECL:.*]]:2 = hlfir.declare %[[PRIVATE_Y_REF]] {uniq_name = "_QFlastprivate_commonEy"} : (!fir.ref<f32>) -> (!fir.ref<f32>, !fir.ref<f32>)
-!CHECK:    omp.wsloop   for  (%[[I:.*]]) : i32 = (%{{.*}}) to (%{{.*}}) inclusive step (%{{.*}}) {
-!CHECK:      %[[V:.*]] = arith.addi %[[I]], %{{.*}} : i32
-!CHECK:      %[[C0:.*]] = arith.constant 0 : i32
-!CHECK:      %[[NEG_STEP:.*]] = arith.cmpi slt, %{{.*}}, %[[C0]] : i32
-!CHECK:      %[[V_LT:.*]] = arith.cmpi slt, %[[V]], %{{.*}} : i32
-!CHECK:      %[[V_GT:.*]] = arith.cmpi sgt, %[[V]], %{{.*}} : i32
-!CHECK:      %[[LAST_ITER:.*]] = arith.select %[[NEG_STEP]], %[[V_LT]], %[[V_GT]] : i1
-!CHECK:      fir.if %[[LAST_ITER]] {
-!CHECK:        fir.store %[[V]] to %{{.*}} : !fir.ref<i32>
-!CHECK:        %[[PRIVATE_X_VAL:.*]] = fir.load %[[PRIVATE_X_DECL]]#0 : !fir.ref<f32>
-!CHECK:        hlfir.assign %[[PRIVATE_X_VAL]] to %[[X_DECL]]#0 temporary_lhs : f32, !fir.ref<f32>
-!CHECK:        %[[PRIVATE_Y_VAL:.*]] = fir.load %[[PRIVATE_Y_DECL]]#0 : !fir.ref<f32>
-!CHECK:        hlfir.assign %[[PRIVATE_Y_VAL]] to %[[Y_DECL]]#0 temporary_lhs : f32, !fir.ref<f32>
+!CHECK:      %[[CB_C_REF:.*]] = fir.address_of(@[[CB_C]]) : !fir.ref<!fir.array<8xi8>>
+!CHECK:      %[[CB_C_REF_CVT:.*]] = fir.convert %[[CB_C_REF]] : (!fir.ref<!fir.array<8xi8>>) -> !fir.ref<!fir.array<?xi8>>
+!CHECK:      %[[CB_C_X_COOR:.*]] = fir.coordinate_of %[[CB_C_REF_CVT]], %{{.*}} : (!fir.ref<!fir.array<?xi8>>, index) -> !fir.ref<i8>
+!CHECK:      %[[CB_C_X_ADDR:.*]] = fir.convert %[[CB_C_X_COOR]] : (!fir.ref<i8>) -> !fir.ref<f32>
+!CHECK:      %[[X_DECL:.*]]:2 = hlfir.declare %[[CB_C_X_ADDR]] {uniq_name = "_QFlastprivate_commonEx"} : (!fir.ref<f32>) -> (!fir.ref<f32>, !fir.ref<f32>)
+!CHECK:      %[[CB_C_REF_CVT:.*]] = fir.convert %[[CB_C_REF]] : (!fir.ref<!fir.array<8xi8>>) -> !fir.ref<!fir.array<?xi8>>
+!CHECK:      %[[CB_C_Y_COOR:.*]] = fir.coordinate_of %[[CB_C_REF_CVT]], %{{.*}} : (!fir.ref<!fir.array<?xi8>>, index) -> !fir.ref<i8>
+!CHECK:      %[[CB_C_Y_ADDR:.*]] = fir.convert %[[CB_C_Y_COOR]] : (!fir.ref<i8>) -> !fir.ref<f32>
+!CHECK:      %[[Y_DECL:.*]]:2 = hlfir.declare %[[CB_C_Y_ADDR]] {uniq_name = "_QFlastprivate_commonEy"} : (!fir.ref<f32>) -> (!fir.ref<f32>, !fir.ref<f32>)
+!CHECK:      %[[PRIVATE_X_REF:.*]] = fir.alloca f32 {bindc_name = "x", pinned, uniq_name = "_QFlastprivate_commonEx"}
+!CHECK:      %[[PRIVATE_X_DECL:.*]]:2 = hlfir.declare %[[PRIVATE_X_REF]] {uniq_name = "_QFlastprivate_commonEx"} : (!fir.ref<f32>) -> (!fir.ref<f32>, !fir.ref<f32>)
+!CHECK:      %[[PRIVATE_Y_REF:.*]] = fir.alloca f32 {bindc_name = "y", pinned, uniq_name = "_QFlastprivate_commonEy"}
+!CHECK:      %[[PRIVATE_Y_DECL:.*]]:2 = hlfir.declare %[[PRIVATE_Y_REF]] {uniq_name = "_QFlastprivate_commonEy"} : (!fir.ref<f32>) -> (!fir.ref<f32>, !fir.ref<f32>)
+!CHECK:      omp.wsloop {
+!CHECK-NEXT:   omp.loop_nest (%[[I:.*]]) : i32 = (%{{.*}}) to (%{{.*}}) inclusive step (%{{.*}}) {
+!CHECK:          %[[V:.*]] = arith.addi %[[I]], %{{.*}} : i32
+!CHECK:          %[[C0:.*]] = arith.constant 0 : i32
+!CHECK:          %[[NEG_STEP:.*]] = arith.cmpi slt, %{{.*}}, %[[C0]] : i32
+!CHECK:          %[[V_LT:.*]] = arith.cmpi slt, %[[V]], %{{.*}} : i32
+!CHECK:          %[[V_GT:.*]] = arith.cmpi sgt, %[[V]], %{{.*}} : i32
+!CHECK:          %[[LAST_ITER:.*]] = arith.select %[[NEG_STEP]], %[[V_LT]], %[[V_GT]] : i1
+!CHECK:          fir.if %[[LAST_ITER]] {
+!CHECK:            fir.store %[[V]] to %{{.*}} : !fir.ref<i32>
+!CHECK:            %[[PRIVATE_X_VAL:.*]] = fir.load %[[PRIVATE_X_DECL]]#0 : !fir.ref<f32>
+!CHECK:            hlfir.assign %[[PRIVATE_X_VAL]] to %[[X_DECL]]#0 temporary_lhs : f32, !fir.ref<f32>
+!CHECK:            %[[PRIVATE_Y_VAL:.*]] = fir.load %[[PRIVATE_Y_DECL]]#0 : !fir.ref<f32>
+!CHECK:            hlfir.assign %[[PRIVATE_Y_VAL]] to %[[Y_DECL]]#0 temporary_lhs : f32, !fir.ref<f32>
+!CHECK:          }
+!CHECK:          omp.yield
+!CHECK:        }
+!CHECK:        omp.terminator
 !CHECK:      }
-!CHECK:      omp.yield
-!CHECK:    }
 subroutine lastprivate_common
   common /c/ x, y
   real x, y
diff --git a/flang/test/Lower/OpenMP/lastprivate-iv.f90 b/flang/test/Lower/OpenMP/lastprivate-iv.f90
index 70fe500129d1..24c20281b9c3 100644
--- a/flang/test/Lower/OpenMP/lastprivate-iv.f90
+++ b/flang/test/Lower/OpenMP/lastprivate-iv.f90
@@ -2,28 +2,31 @@
 ! RUN: %flang_fc1 -emit-hlfir -fopenmp -o - %s 2>&1 | FileCheck %s
 
 !CHECK-LABEL: func @_QPlastprivate_iv_inc
-!CHECK:    %[[I_MEM:.*]] = fir.alloca i32 {adapt.valuebyref, pinned}
-!CHECK:    %[[I:.*]]:2 = hlfir.declare %[[I_MEM]] {uniq_name = "_QFlastprivate_iv_incEi"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
-!CHECK:    %[[I2_MEM:.*]] = fir.alloca i32 {bindc_name = "i", uniq_name = "_QFlastprivate_iv_incEi"}
-!CHECK:    %[[I2:.*]]:2 = hlfir.declare %[[I2_MEM]] {uniq_name = "_QFlastprivate_iv_incEi"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
-!CHECK:    %[[LB:.*]] = arith.constant 4 : i32
-!CHECK:    %[[UB:.*]] = arith.constant 10 : i32
-!CHECK:    %[[STEP:.*]]  = arith.constant 3 : i32
-!CHECK:    omp.wsloop for  (%[[IV:.*]]) : i32 = (%[[LB]]) to (%[[UB]]) inclusive step (%[[STEP]]) {
-!CHECK:      fir.store %[[IV]] to %[[I]]#1 : !fir.ref<i32>
-!CHECK:      %[[V:.*]] = arith.addi %[[IV]], %[[STEP]] : i32
-!CHECK:      %[[C0:.*]] = arith.constant 0 : i32
-!CHECK:      %[[STEP_NEG:.*]] = arith.cmpi slt, %[[STEP]], %[[C0]] : i32
-!CHECK:      %[[V_LT:.*]] = arith.cmpi slt, %[[V]], %[[UB]] : i32
-!CHECK:      %[[V_GT:.*]] = arith.cmpi sgt, %[[V]], %[[UB]] : i32
-!CHECK:      %[[CMP:.*]] = arith.select %[[STEP_NEG]], %[[V_LT]], %[[V_GT]] : i1
-!CHECK:      fir.if %[[CMP]] {
-!CHECK:        fir.store %[[V]] to %[[I]]#1 : !fir.ref<i32>
-!CHECK:        %[[I_VAL:.*]] = fir.load %[[I]]#0 : !fir.ref<i32>
-!CHECK:        hlfir.assign %[[I_VAL]] to %[[I2]]#0 temporary_lhs : i32, !fir.ref<i32>
+!CHECK:      %[[I_MEM:.*]] = fir.alloca i32 {adapt.valuebyref, pinned}
+!CHECK:      %[[I:.*]]:2 = hlfir.declare %[[I_MEM]] {uniq_name = "_QFlastprivate_iv_incEi"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
+!CHECK:      %[[I2_MEM:.*]] = fir.alloca i32 {bindc_name = "i", uniq_name = "_QFlastprivate_iv_incEi"}
+!CHECK:      %[[I2:.*]]:2 = hlfir.declare %[[I2_MEM]] {uniq_name = "_QFlastprivate_iv_incEi"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
+!CHECK:      %[[LB:.*]] = arith.constant 4 : i32
+!CHECK:      %[[UB:.*]] = arith.constant 10 : i32
+!CHECK:      %[[STEP:.*]]  = arith.constant 3 : i32
+!CHECK:      omp.wsloop {
+!CHECK-NEXT:   omp.loop_nest (%[[IV:.*]]) : i32 = (%[[LB]]) to (%[[UB]]) inclusive step (%[[STEP]]) {
+!CHECK:          fir.store %[[IV]] to %[[I]]#1 : !fir.ref<i32>
+!CHECK:          %[[V:.*]] = arith.addi %[[IV]], %[[STEP]] : i32
+!CHECK:          %[[C0:.*]] = arith.constant 0 : i32
+!CHECK:          %[[STEP_NEG:.*]] = arith.cmpi slt, %[[STEP]], %[[C0]] : i32
+!CHECK:          %[[V_LT:.*]] = arith.cmpi slt, %[[V]], %[[UB]] : i32
+!CHECK:          %[[V_GT:.*]] = arith.cmpi sgt, %[[V]], %[[UB]] : i32
+!CHECK:          %[[CMP:.*]] = arith.select %[[STEP_NEG]], %[[V_LT]], %[[V_GT]] : i1
+!CHECK:          fir.if %[[CMP]] {
+!CHECK:            fir.store %[[V]] to %[[I]]#1 : !fir.ref<i32>
+!CHECK:            %[[I_VAL:.*]] = fir.load %[[I]]#0 : !fir.ref<i32>
+!CHECK:            hlfir.assign %[[I_VAL]] to %[[I2]]#0 temporary_lhs : i32, !fir.ref<i32>
+!CHECK:          }
+!CHECK:          omp.yield
+!CHECK:        }
+!CHECK:        omp.terminator
 !CHECK:      }
-!CHECK:      omp.yield
-!CHECK:    }
 subroutine lastprivate_iv_inc()
   integer :: i
 
@@ -34,28 +37,31 @@ subroutine lastprivate_iv_inc()
 end subroutine
 
 !CHECK-LABEL: func @_QPlastprivate_iv_dec
-!CHECK:    %[[I_MEM:.*]] = fir.alloca i32 {adapt.valuebyref, pinned}
-!CHECK:    %[[I:.*]]:2 = hlfir.declare %[[I_MEM]] {uniq_name = "_QFlastprivate_iv_decEi"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
-!CHECK:    %[[I2_MEM:.*]] = fir.alloca i32 {bindc_name = "i", uniq_name = "_QFlastprivate_iv_decEi"}
-!CHECK:    %[[I2:.*]]:2 = hlfir.declare %[[I2_MEM]] {uniq_name = "_QFlastprivate_iv_decEi"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
-!CHECK:    %[[LB:.*]] = arith.constant 10 : i32
-!CHECK:    %[[UB:.*]] = arith.constant 1 : i32
-!CHECK:    %[[STEP:.*]]  = arith.constant -3 : i32
-!CHECK:    omp.wsloop for  (%[[IV:.*]]) : i32 = (%[[LB]]) to (%[[UB]]) inclusive step (%[[STEP]]) {
-!CHECK:      fir.store %[[IV]] to %[[I]]#1 : !fir.ref<i32>
-!CHECK:      %[[V:.*]] = arith.addi %[[IV]], %[[STEP]] : i32
-!CHECK:      %[[C0:.*]] = arith.constant 0 : i32
-!CHECK:      %[[STEP_NEG:.*]] = arith.cmpi slt, %[[STEP]], %[[C0]] : i32
-!CHECK:      %[[V_LT:.*]] = arith.cmpi slt, %[[V]], %[[UB]] : i32
-!CHECK:      %[[V_GT:.*]] = arith.cmpi sgt, %[[V]], %[[UB]] : i32
-!CHECK:      %[[CMP:.*]] = arith.select %[[STEP_NEG]], %[[V_LT]], %[[V_GT]] : i1
-!CHECK:      fir.if %[[CMP]] {
-!CHECK:        fir.store %[[V]] to %[[I]]#1 : !fir.ref<i32>
-!CHECK:        %[[I_VAL:.*]] = fir.load %[[I]]#0 : !fir.ref<i32>
-!CHECK:        hlfir.assign %[[I_VAL]] to %[[I2]]#0 temporary_lhs : i32, !fir.ref<i32>
+!CHECK:      %[[I_MEM:.*]] = fir.alloca i32 {adapt.valuebyref, pinned}
+!CHECK:      %[[I:.*]]:2 = hlfir.declare %[[I_MEM]] {uniq_name = "_QFlastprivate_iv_decEi"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
+!CHECK:      %[[I2_MEM:.*]] = fir.alloca i32 {bindc_name = "i", uniq_name = "_QFlastprivate_iv_decEi"}
+!CHECK:      %[[I2:.*]]:2 = hlfir.declare %[[I2_MEM]] {uniq_name = "_QFlastprivate_iv_decEi"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
+!CHECK:      %[[LB:.*]] = arith.constant 10 : i32
+!CHECK:      %[[UB:.*]] = arith.constant 1 : i32
+!CHECK:      %[[STEP:.*]]  = arith.constant -3 : i32
+!CHECK:      omp.wsloop {
+!CHECK-NEXT:   omp.loop_nest (%[[IV:.*]]) : i32 = (%[[LB]]) to (%[[UB]]) inclusive step (%[[STEP]]) {
+!CHECK:          fir.store %[[IV]] to %[[I]]#1 : !fir.ref<i32>
+!CHECK:          %[[V:.*]] = arith.addi %[[IV]], %[[STEP]] : i32
+!CHECK:          %[[C0:.*]] = arith.constant 0 : i32
+!CHECK:          %[[STEP_NEG:.*]] = arith.cmpi slt, %[[STEP]], %[[C0]] : i32
+!CHECK:          %[[V_LT:.*]] = arith.cmpi slt, %[[V]], %[[UB]] : i32
+!CHECK:          %[[V_GT:.*]] = arith.cmpi sgt, %[[V]], %[[UB]] : i32
+!CHECK:          %[[CMP:.*]] = arith.select %[[STEP_NEG]], %[[V_LT]], %[[V_GT]] : i1
+!CHECK:          fir.if %[[CMP]] {
+!CHECK:            fir.store %[[V]] to %[[I]]#1 : !fir.ref<i32>
+!CHECK:            %[[I_VAL:.*]] = fir.load %[[I]]#0 : !fir.ref<i32>
+!CHECK:            hlfir.assign %[[I_VAL]] to %[[I2]]#0 temporary_lhs : i32, !fir.ref<i32>
+!CHECK:          }
+!CHECK:          omp.yield
+!CHECK:        }
+!CHECK:        omp.terminator
 !CHECK:      }
-!CHECK:      omp.yield
-!CHECK:    }
 subroutine lastprivate_iv_dec()
   integer :: i
 
diff --git a/flang/test/Lower/OpenMP/location.f90 b/flang/test/Lower/OpenMP/location.f90
index 1e01a4828dd9..5d340937a81c 100644
--- a/flang/test/Lower/OpenMP/location.f90
+++ b/flang/test/Lower/OpenMP/location.f90
@@ -28,12 +28,15 @@ end
 
 !CHECK-LABEL: sub_loop
 subroutine sub_loop()
-!CHECK: omp.wsloop {{.*}}  {
+!CHECK: omp.wsloop {
+!CHECK-NEXT: omp.loop_nest {{.*}} {
   !$omp do
   do i=1,10
     print *, i
 !CHECK:   omp.yield loc(#[[LOOP_LOC:.*]])
 !CHECK: } loc(#[[LOOP_LOC]])
+!CHECK:   omp.terminator loc(#[[LOOP_LOC]])
+!CHECK: } loc(#[[LOOP_LOC]])
   end do
   !$omp end do
 end
@@ -60,9 +63,9 @@ end subroutine
 
 !CHECK: #[[PAR_LOC]] = loc("{{.*}}location.f90":9:9)
 !CHECK: #[[TAR_LOC]] = loc("{{.*}}location.f90":21:9)
-!CHECK: #[[LOOP_LOC]] = loc("{{.*}}location.f90":32:9)
-!CHECK: #[[BAR_LOC]] = loc("{{.*}}location.f90":44:9)
-!CHECK: #[[TW_LOC]] = loc("{{.*}}location.f90":46:9)
-!CHECK: #[[TY_LOC]] = loc("{{.*}}location.f90":48:9)
-!CHECK: #[[IF_LOC]] = loc("{{.*}}location.f90":55:14)
-!CHECK: #[[TASK_LOC]] = loc("{{.*}}location.f90":55:9)
+!CHECK: #[[LOOP_LOC]] = loc("{{.*}}location.f90":33:9)
+!CHECK: #[[BAR_LOC]] = loc("{{.*}}location.f90":47:9)
+!CHECK: #[[TW_LOC]] = loc("{{.*}}location.f90":49:9)
+!CHECK: #[[TY_LOC]] = loc("{{.*}}location.f90":51:9)
+!CHECK: #[[IF_LOC]] = loc("{{.*}}location.f90":58:14)
+!CHECK: #[[TASK_LOC]] = loc("{{.*}}location.f90":58:9)
diff --git a/flang/test/Lower/OpenMP/parallel-lastprivate-clause-scalar.f90 b/flang/test/Lower/OpenMP/parallel-lastprivate-clause-scalar.f90
index 28f59c95d60b..bb81e5eac62f 100644
--- a/flang/test/Lower/OpenMP/parallel-lastprivate-clause-scalar.f90
+++ b/flang/test/Lower/OpenMP/parallel-lastprivate-clause-scalar.f90
@@ -14,8 +14,9 @@
 !CHECK-DAG: %[[ARG1_PVT_DECL:.*]]:2 = hlfir.declare %[[ARG1_PVT]] typeparams %[[FIVE]] {uniq_name = "_QFlastprivate_characterEarg1"} : (!fir.ref<!fir.char<1,5>>, index) -> (!fir.ref<!fir.char<1,5>>, !fir.ref<!fir.char<1,5>>)
 
 ! Check that we are accessing the clone inside the loop
-!CHECK-DAG: omp.wsloop for (%[[INDX_WS:.*]]) : {{.*}} {
-!CHECK-DAG: %[[UNIT:.*]] = arith.constant 6 : i32
+!CHECK: omp.wsloop {
+!CHECK-NEXT: omp.loop_nest (%[[INDX_WS:.*]]) : {{.*}} {
+!CHECK: %[[UNIT:.*]] = arith.constant 6 : i32
 !CHECK-NEXT: %[[ADDR:.*]] = fir.address_of(@_QQclX
 !CHECK-NEXT: %[[CVT0:.*]] = fir.convert %[[ADDR]] 
 !CHECK-NEXT: %[[CNST:.*]] = arith.constant
@@ -36,9 +37,12 @@
 !CHECK: fir.store %[[V]] to %{{.*}} : !fir.ref<i32>
 
 ! Testing lastprivate val update
-!CHECK-DAG: hlfir.assign %[[ARG1_PVT_DECL]]#0 to %[[ARG1_DECL]]#0 temporary_lhs : !fir.ref<!fir.char<1,5>>, !fir.ref<!fir.char<1,5>>
-!CHECK-DAG: } 
-!CHECK-DAG: omp.yield
+!CHECK: hlfir.assign %[[ARG1_PVT_DECL]]#0 to %[[ARG1_DECL]]#0 temporary_lhs : !fir.ref<!fir.char<1,5>>, !fir.ref<!fir.char<1,5>>
+!CHECK: } 
+!CHECK: omp.yield
+!CHECK: }
+!CHECK: omp.terminator
+!CHECK: }
 
 subroutine lastprivate_character(arg1)
         character(5) :: arg1
@@ -57,7 +61,8 @@ end subroutine
 !CHECK-DAG: omp.parallel  {
 !CHECK-DAG: %[[CLONE:.*]] = fir.alloca i32 {bindc_name = "arg1"
 !CHECK-DAG: %[[CLONE_DECL:.*]]:2 = hlfir.declare %[[CLONE]] {uniq_name = "_QFlastprivate_intEarg1"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
-!CHECK: omp.wsloop for (%[[INDX_WS:.*]]) : {{.*}} {
+!CHECK: omp.wsloop {
+!CHECK-NEXT: omp.loop_nest (%[[INDX_WS:.*]]) : {{.*}} {
 
 ! Testing last iteration check
 !CHECK: %[[V:.*]] = arith.addi %[[INDX_WS]], %{{.*}} : i32
@@ -72,8 +77,11 @@ end subroutine
 ! Testing lastprivate val update
 !CHECK-NEXT: %[[CLONE_LD:.*]] = fir.load %[[CLONE_DECL]]#0 : !fir.ref<i32>
 !CHECK:      hlfir.assign %[[CLONE_LD]] to %[[ARG1_DECL]]#0 temporary_lhs : i32, !fir.ref<i32>
-!CHECK-DAG: }
-!CHECK-DAG: omp.yield
+!CHECK: }
+!CHECK: omp.yield
+!CHECK: }
+!CHECK: omp.terminator
+!CHECK: }
 
 subroutine lastprivate_int(arg1)
         integer :: arg1
@@ -96,7 +104,8 @@ end subroutine
 !CHECK-DAG: %[[CLONE1_DECL:.*]]:2 = hlfir.declare %[[CLONE1]] {uniq_name = "_QFmult_lastprivate_intEarg1"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
 !CHECK-DAG: %[[CLONE2:.*]] = fir.alloca i32 {bindc_name = "arg2"
 !CHECK-DAG: %[[CLONE2_DECL:.*]]:2 = hlfir.declare %[[CLONE2]] {uniq_name = "_QFmult_lastprivate_intEarg2"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
-!CHECK: omp.wsloop for (%[[INDX_WS:.*]]) : {{.*}} {
+!CHECK: omp.wsloop {
+!CHECK-NEXT: omp.loop_nest (%[[INDX_WS:.*]]) : {{.*}} {
 
 ! Testing last iteration check
 !CHECK: %[[V:.*]] = arith.addi %[[INDX_WS]], %{{.*}} : i32
@@ -114,6 +123,9 @@ end subroutine
 !CHECK-DAG: hlfir.assign %[[CLONE_LD2]] to %[[ARG2_DECL]]#0 temporary_lhs : i32, !fir.ref<i32>
 !CHECK: }
 !CHECK: omp.yield
+!CHECK: }
+!CHECK: omp.terminator
+!CHECK: }
 
 subroutine mult_lastprivate_int(arg1, arg2)
         integer :: arg1, arg2
@@ -137,7 +149,8 @@ end subroutine
 !CHECK-DAG: %[[CLONE1_DECL:.*]]:2 = hlfir.declare %[[CLONE1]] {uniq_name = "_QFmult_lastprivate_int2Earg1"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
 !CHECK-DAG: %[[CLONE2:.*]] = fir.alloca i32 {bindc_name = "arg2"
 !CHECK-DAG: %[[CLONE2_DECL:.*]]:2 = hlfir.declare %[[CLONE2]] {uniq_name = "_QFmult_lastprivate_int2Earg2"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
-!CHECK: omp.wsloop for (%[[INDX_WS:.*]]) : {{.*}} {
+!CHECK: omp.wsloop {
+!CHECK-NEXT: omp.loop_nest (%[[INDX_WS:.*]]) : {{.*}} {
 
 !Testing last iteration check
 !CHECK: %[[V:.*]] = arith.addi %[[INDX_WS]], %{{.*}} : i32
@@ -155,6 +168,9 @@ end subroutine
 !CHECK-DAG: hlfir.assign %[[CLONE_LD1]] to %[[ARG1_DECL]]#0 temporary_lhs : i32, !fir.ref<i32>
 !CHECK: }
 !CHECK: omp.yield
+!CHECK: }
+!CHECK: omp.terminator
+!CHECK: }
 
 subroutine mult_lastprivate_int2(arg1, arg2)
         integer :: arg1, arg2
@@ -183,7 +199,8 @@ end subroutine
 !CHECK: %[[CLONE2:.*]] = fir.alloca i32 {bindc_name = "arg2"
 !CHECK: %[[CLONE2_DECL:.*]]:2 = hlfir.declare %[[CLONE2]] {uniq_name = "_QFfirstpriv_lastpriv_intEarg2"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
 !CHECK-NOT: omp.barrier
-!CHECK: omp.wsloop for (%[[INDX_WS:.*]]) : {{.*}} {
+!CHECK: omp.wsloop {
+!CHECK-NEXT: omp.loop_nest (%[[INDX_WS:.*]]) : {{.*}} {
 
 ! Testing last iteration check
 !CHECK: %[[V:.*]] = arith.addi %[[INDX_WS]], %{{.*}} : i32
@@ -199,6 +216,9 @@ end subroutine
 !CHECK-NEXT: hlfir.assign %[[CLONE_LD]] to %[[ARG2_DECL]]#0 temporary_lhs : i32, !fir.ref<i32>
 !CHECK-NEXT: }
 !CHECK-NEXT: omp.yield
+!CHECK-NEXT: }
+!CHECK-NEXT: omp.terminator
+!CHECK-NEXT: }
 
 subroutine firstpriv_lastpriv_int(arg1, arg2)
         integer :: arg1, arg2
@@ -223,7 +243,8 @@ end subroutine
 !CHECK-NEXT: %[[FPV_LD:.*]] = fir.load %[[ARG1_DECL]]#0 : !fir.ref<i32>
 !CHECK-NEXT: hlfir.assign %[[FPV_LD]] to %[[CLONE1_DECL]]#0 temporary_lhs : i32, !fir.ref<i32>
 !CHECK-NEXT: omp.barrier
-!CHECK: omp.wsloop for (%[[INDX_WS:.*]]) : {{.*}} {
+!CHECK: omp.wsloop {
+!CHECK-NEXT: omp.loop_nest (%[[INDX_WS:.*]]) : {{.*}} {
 ! Testing last iteration check
 !CHECK: %[[V:.*]] = arith.addi %[[INDX_WS]], %{{.*}} : i32
 !CHECK: %[[C0:.*]] = arith.constant 0 : i32
@@ -238,6 +259,9 @@ end subroutine
 !CHECK-NEXT: hlfir.assign %[[CLONE_LD]] to %[[ARG1_DECL]]#0 temporary_lhs : i32, !fir.ref<i32>
 !CHECK-NEXT: }
 !CHECK-NEXT: omp.yield
+!CHECK-NEXT: }
+!CHECK-NEXT: omp.terminator
+!CHECK-NEXT: }
 
 subroutine firstpriv_lastpriv_int2(arg1)
         integer :: arg1
diff --git a/flang/test/Lower/OpenMP/parallel-private-clause-fixes.f90 b/flang/test/Lower/OpenMP/parallel-private-clause-fixes.f90
index 8533106b7ac4..93809fde98a2 100644
--- a/flang/test/Lower/OpenMP/parallel-private-clause-fixes.f90
+++ b/flang/test/Lower/OpenMP/parallel-private-clause-fixes.f90
@@ -21,30 +21,33 @@
 ! CHECK:           %[[ONE:.*]] = arith.constant 1 : i32
 ! CHECK:           %[[VAL_3:.*]] = fir.load %[[GAMA_DECL]]#0 : !fir.ref<i32>
 ! CHECK:           %[[VAL_5:.*]] = arith.constant 1 : i32
-! CHECK:           omp.wsloop for (%[[VAL_6:.*]]) : i32 = (%[[ONE]]) to (%[[VAL_3]]) inclusive step (%[[VAL_5]]) {
-! CHECK:             fir.store %[[VAL_6]] to %[[PRIV_I_DECL]]#1 : !fir.ref<i32>
-! CHECK:             %[[VAL_7:.*]] = arith.constant 1 : i32
-! CHECK:             %[[VAL_8:.*]] = fir.convert %[[VAL_7]] : (i32) -> index
-! CHECK:             %[[VAL_9:.*]] = fir.load %[[GAMA_DECL]]#0 : !fir.ref<i32>
-! CHECK:             %[[VAL_10:.*]] = fir.convert %[[VAL_9]] : (i32) -> index
-! CHECK:             %[[VAL_11:.*]] = arith.constant 1 : index
-! CHECK:             %[[LB:.*]] = fir.convert %[[VAL_8]] : (index) -> i32
-! CHECK:             %[[VAL_12:.*]]:2 = fir.do_loop %[[VAL_13:[^ ]*]] =
-! CHECK-SAME:            %[[VAL_8]] to %[[VAL_10]] step %[[VAL_11]]
-! CHECK-SAME:            iter_args(%[[IV:.*]] = %[[LB]]) -> (index, i32) {
-! CHECK:               fir.store %[[IV]] to %[[PRIV_J_DECL]]#1 : !fir.ref<i32>
-! CHECK:               %[[LOAD:.*]] = fir.load %[[PRIV_I_DECL]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_15:.*]] = fir.load %[[PRIV_J_DECL]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_16:.*]] = arith.addi %[[LOAD]], %[[VAL_15]] : i32
-! CHECK:               hlfir.assign %[[VAL_16]] to %[[PRIV_X_DECL]]#0 : i32, !fir.ref<i32>
-! CHECK:               %[[VAL_17:.*]] = arith.addi %[[VAL_13]], %[[VAL_11]] : index
-! CHECK:               %[[STEPCAST:.*]] = fir.convert %[[VAL_11]] : (index) -> i32
-! CHECK:               %[[IVLOAD:.*]] = fir.load %[[PRIV_J_DECL]]#1 : !fir.ref<i32>
-! CHECK:               %[[IVINC:.*]] = arith.addi %[[IVLOAD]], %[[STEPCAST]]
-! CHECK:               fir.result %[[VAL_17]], %[[IVINC]] : index, i32
+! CHECK:           omp.wsloop {
+! CHECK-NEXT:        omp.loop_nest (%[[VAL_6:.*]]) : i32 = (%[[ONE]]) to (%[[VAL_3]]) inclusive step (%[[VAL_5]]) {
+! CHECK:               fir.store %[[VAL_6]] to %[[PRIV_I_DECL]]#1 : !fir.ref<i32>
+! CHECK:               %[[VAL_7:.*]] = arith.constant 1 : i32
+! CHECK:               %[[VAL_8:.*]] = fir.convert %[[VAL_7]] : (i32) -> index
+! CHECK:               %[[VAL_9:.*]] = fir.load %[[GAMA_DECL]]#0 : !fir.ref<i32>
+! CHECK:               %[[VAL_10:.*]] = fir.convert %[[VAL_9]] : (i32) -> index
+! CHECK:               %[[VAL_11:.*]] = arith.constant 1 : index
+! CHECK:               %[[LB:.*]] = fir.convert %[[VAL_8]] : (index) -> i32
+! CHECK:               %[[VAL_12:.*]]:2 = fir.do_loop %[[VAL_13:[^ ]*]] =
+! CHECK-SAME:              %[[VAL_8]] to %[[VAL_10]] step %[[VAL_11]]
+! CHECK-SAME:              iter_args(%[[IV:.*]] = %[[LB]]) -> (index, i32) {
+! CHECK:                 fir.store %[[IV]] to %[[PRIV_J_DECL]]#1 : !fir.ref<i32>
+! CHECK:                 %[[LOAD:.*]] = fir.load %[[PRIV_I_DECL]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_15:.*]] = fir.load %[[PRIV_J_DECL]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_16:.*]] = arith.addi %[[LOAD]], %[[VAL_15]] : i32
+! CHECK:                 hlfir.assign %[[VAL_16]] to %[[PRIV_X_DECL]]#0 : i32, !fir.ref<i32>
+! CHECK:                 %[[VAL_17:.*]] = arith.addi %[[VAL_13]], %[[VAL_11]] : index
+! CHECK:                 %[[STEPCAST:.*]] = fir.convert %[[VAL_11]] : (index) -> i32
+! CHECK:                 %[[IVLOAD:.*]] = fir.load %[[PRIV_J_DECL]]#1 : !fir.ref<i32>
+! CHECK:                 %[[IVINC:.*]] = arith.addi %[[IVLOAD]], %[[STEPCAST]]
+! CHECK:                 fir.result %[[VAL_17]], %[[IVINC]] : index, i32
+! CHECK:               }
+! CHECK:               fir.store %[[VAL_12]]#1 to %[[PRIV_J_DECL]]#1 : !fir.ref<i32>
+! CHECK:               omp.yield
 ! CHECK:             }
-! CHECK:             fir.store %[[VAL_12]]#1 to %[[PRIV_J_DECL]]#1 : !fir.ref<i32>
-! CHECK:             omp.yield
+! CHECK:             omp.terminator
 ! CHECK:           }
 ! CHECK:           omp.terminator
 ! CHECK:         }
diff --git a/flang/test/Lower/OpenMP/parallel-private-clause.f90 b/flang/test/Lower/OpenMP/parallel-private-clause.f90
index 775f7b4f2cb1..b9b58a135aaa 100644
--- a/flang/test/Lower/OpenMP/parallel-private-clause.f90
+++ b/flang/test/Lower/OpenMP/parallel-private-clause.f90
@@ -292,33 +292,35 @@ subroutine simple_loop_1
   real, allocatable :: r;
   ! FIRDialect:  omp.parallel
   !$OMP PARALLEL PRIVATE(r)
-  ! FIRDialect:     %[[ALLOCA_IV:.*]] = fir.alloca i32 {{{.*}}, pinned}
+  ! FIRDialect:      %[[ALLOCA_IV:.*]] = fir.alloca i32 {{{.*}}, pinned}
 
-  ! FIRDialect:     %[[ALLOCA_IV_DECL:.*]]:2 = hlfir.declare %[[ALLOCA_IV]] {uniq_name = "_QFsimple_loop_1Ei"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
-  ! FIRDialect:     [[R:%.*]] = fir.alloca !fir.box<!fir.heap<f32>> {bindc_name = "r", pinned, uniq_name = "{{.*}}Er"}
-  ! FIRDialect:     fir.store {{%.*}} to [[R]] : !fir.ref<!fir.box<!fir.heap<f32>>>
-  ! FIRDialect:     fir.store {{%.*}} to [[R]] : !fir.ref<!fir.box<!fir.heap<f32>>>
-  ! FIRDialect:     %[[R_DECL:.*]]:2 = hlfir.declare [[R]] {fortran_attrs = #fir.var_attrs<allocatable>, uniq_name = "_QFsimple_loop_1Er"} : (!fir.ref<!fir.box<!fir.heap<f32>>>) -> (!fir.ref<!fir.box<!fir.heap<f32>>>, !fir.ref<!fir.box<!fir.heap<f32>>>)
+  ! FIRDialect:      %[[ALLOCA_IV_DECL:.*]]:2 = hlfir.declare %[[ALLOCA_IV]] {uniq_name = "_QFsimple_loop_1Ei"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
+  ! FIRDialect:      [[R:%.*]] = fir.alloca !fir.box<!fir.heap<f32>> {bindc_name = "r", pinned, uniq_name = "{{.*}}Er"}
+  ! FIRDialect:      fir.store {{%.*}} to [[R]] : !fir.ref<!fir.box<!fir.heap<f32>>>
+  ! FIRDialect:      fir.store {{%.*}} to [[R]] : !fir.ref<!fir.box<!fir.heap<f32>>>
+  ! FIRDialect:      %[[R_DECL:.*]]:2 = hlfir.declare [[R]] {fortran_attrs = #fir.var_attrs<allocatable>, uniq_name = "_QFsimple_loop_1Er"} : (!fir.ref<!fir.box<!fir.heap<f32>>>) -> (!fir.ref<!fir.box<!fir.heap<f32>>>, !fir.ref<!fir.box<!fir.heap<f32>>>)
 
-  ! FIRDialect:     %[[WS_LB:.*]] = arith.constant 1 : i32
-  ! FIRDialect:     %[[WS_UB:.*]] = arith.constant 9 : i32
-  ! FIRDialect:     %[[WS_STEP:.*]] = arith.constant 1 : i32
+  ! FIRDialect:      %[[WS_LB:.*]] = arith.constant 1 : i32
+  ! FIRDialect:      %[[WS_UB:.*]] = arith.constant 9 : i32
+  ! FIRDialect:      %[[WS_STEP:.*]] = arith.constant 1 : i32
 
-  ! FIRDialect:     omp.wsloop for (%[[I:.*]]) : i32 = (%[[WS_LB]]) to (%[[WS_UB]]) inclusive step (%[[WS_STEP]])
+  ! FIRDialect:      omp.wsloop {
+  ! FIRDialect-NEXT: omp.loop_nest (%[[I:.*]]) : i32 = (%[[WS_LB]]) to (%[[WS_UB]]) inclusive step (%[[WS_STEP]]) {
   !$OMP DO
   do i=1, 9
-  ! FIRDialect:     fir.store %[[I]] to %[[ALLOCA_IV_DECL]]#1 : !fir.ref<i32>
-  ! FIRDialect:     %[[LOAD_IV:.*]] = fir.load %[[ALLOCA_IV_DECL]]#0 : !fir.ref<i32>
-  ! FIRDialect:     fir.call @_FortranAioOutputInteger32({{.*}}, %[[LOAD_IV]]) {{.*}} : (!fir.ref<i8>, i32) -> i1
+  ! FIRDialect:      fir.store %[[I]] to %[[ALLOCA_IV_DECL]]#1 : !fir.ref<i32>
+  ! FIRDialect:      %[[LOAD_IV:.*]] = fir.load %[[ALLOCA_IV_DECL]]#0 : !fir.ref<i32>
+  ! FIRDialect:      fir.call @_FortranAioOutputInteger32({{.*}}, %[[LOAD_IV]]) {{.*}} : (!fir.ref<i8>, i32) -> i1
     print*, i
   end do
-  ! FIRDialect:     omp.yield
-  ! FIRDialect:     {{%.*}} = fir.load %[[R_DECL]]#0 : !fir.ref<!fir.box<!fir.heap<f32>>>
-  ! FIRDialect:     fir.if {{%.*}} {
-  ! FIRDialect:     [[LD:%.*]] = fir.load %[[R_DECL]]#0 : !fir.ref<!fir.box<!fir.heap<f32>>>
-  ! FIRDialect:     [[AD:%.*]] = fir.box_addr [[LD]] : (!fir.box<!fir.heap<f32>>) -> !fir.heap<f32>
-  ! FIRDialect:     fir.freemem [[AD]] : !fir.heap<f32>
-  ! FIRDialect:     fir.store {{%.*}} to %[[R_DECL]]#0 : !fir.ref<!fir.box<!fir.heap<f32>>>
+  ! FIRDialect:      omp.yield
+  ! FIRDialect:      omp.terminator
+  ! FIRDialect:      {{%.*}} = fir.load %[[R_DECL]]#0 : !fir.ref<!fir.box<!fir.heap<f32>>>
+  ! FIRDialect:      fir.if {{%.*}} {
+  ! FIRDialect:      [[LD:%.*]] = fir.load %[[R_DECL]]#0 : !fir.ref<!fir.box<!fir.heap<f32>>>
+  ! FIRDialect:      [[AD:%.*]] = fir.box_addr [[LD]] : (!fir.box<!fir.heap<f32>>) -> !fir.heap<f32>
+  ! FIRDialect:      fir.freemem [[AD]] : !fir.heap<f32>
+  ! FIRDialect:      fir.store {{%.*}} to %[[R_DECL]]#0 : !fir.ref<!fir.box<!fir.heap<f32>>>
   !$OMP END DO
   ! FIRDialect:  omp.terminator
   !$OMP END PARALLEL
@@ -330,19 +332,20 @@ subroutine simple_loop_2
   real, allocatable :: r;
   ! FIRDialect:  omp.parallel
   !$OMP PARALLEL
-  ! FIRDialect:     %[[ALLOCA_IV:.*]] = fir.alloca i32 {{{.*}}, pinned}
+  ! FIRDialect:      %[[ALLOCA_IV:.*]] = fir.alloca i32 {{{.*}}, pinned}
 
-  ! FIRDialect:     %[[ALLOCA_IV_DECL:.*]]:2 = hlfir.declare %[[ALLOCA_IV]] {uniq_name = "{{.*}}Ei"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
-  ! FIRDialect:     [[R:%.*]] = fir.alloca !fir.box<!fir.heap<f32>> {bindc_name = "r", pinned, uniq_name = "{{.*}}Er"}
-  ! FIRDialect:     fir.store {{%.*}} to [[R]] : !fir.ref<!fir.box<!fir.heap<f32>>>
-  ! FIRDialect:     fir.store {{%.*}} to [[R]] : !fir.ref<!fir.box<!fir.heap<f32>>>
-  ! FIRDialect:     %[[R_DECL:.*]]:2 = hlfir.declare [[R]] {fortran_attrs = #fir.var_attrs<allocatable>, uniq_name = "{{.*}}Er"} : (!fir.ref<!fir.box<!fir.heap<f32>>>) -> (!fir.ref<!fir.box<!fir.heap<f32>>>, !fir.ref<!fir.box<!fir.heap<f32>>>)
+  ! FIRDialect:      %[[ALLOCA_IV_DECL:.*]]:2 = hlfir.declare %[[ALLOCA_IV]] {uniq_name = "{{.*}}Ei"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
+  ! FIRDialect:      [[R:%.*]] = fir.alloca !fir.box<!fir.heap<f32>> {bindc_name = "r", pinned, uniq_name = "{{.*}}Er"}
+  ! FIRDialect:      fir.store {{%.*}} to [[R]] : !fir.ref<!fir.box<!fir.heap<f32>>>
+  ! FIRDialect:      fir.store {{%.*}} to [[R]] : !fir.ref<!fir.box<!fir.heap<f32>>>
+  ! FIRDialect:      %[[R_DECL:.*]]:2 = hlfir.declare [[R]] {fortran_attrs = #fir.var_attrs<allocatable>, uniq_name = "{{.*}}Er"} : (!fir.ref<!fir.box<!fir.heap<f32>>>) -> (!fir.ref<!fir.box<!fir.heap<f32>>>, !fir.ref<!fir.box<!fir.heap<f32>>>)
 
-  ! FIRDialect:     %[[WS_LB:.*]] = arith.constant 1 : i32
-  ! FIRDialect:     %[[WS_UB:.*]] = arith.constant 9 : i32
-  ! FIRDialect:     %[[WS_STEP:.*]] = arith.constant 1 : i32
+  ! FIRDialect:      %[[WS_LB:.*]] = arith.constant 1 : i32
+  ! FIRDialect:      %[[WS_UB:.*]] = arith.constant 9 : i32
+  ! FIRDialect:      %[[WS_STEP:.*]] = arith.constant 1 : i32
 
-  ! FIRDialect:     omp.wsloop for (%[[I:.*]]) : i32 = (%[[WS_LB]]) to (%[[WS_UB]]) inclusive step (%[[WS_STEP]])
+  ! FIRDialect:      omp.wsloop {
+  ! FIRDialect-NEXT: omp.loop_nest (%[[I:.*]]) : i32 = (%[[WS_LB]]) to (%[[WS_UB]]) inclusive step (%[[WS_STEP]]) {
   !$OMP DO PRIVATE(r)
   do i=1, 9
   ! FIRDialect:     fir.store %[[I]] to %[[ALLOCA_IV_DECL]]#1 : !fir.ref<i32>
@@ -351,6 +354,7 @@ subroutine simple_loop_2
     print*, i
   end do
   ! FIRDialect:     omp.yield
+  ! FIRDialect:     omp.terminator
   ! FIRDialect:     {{%.*}} = fir.load %[[R_DECL]]#0 : !fir.ref<!fir.box<!fir.heap<f32>>>
   ! FIRDialect:     fir.if {{%.*}} {
   ! FIRDialect:     [[LD:%.*]] = fir.load %[[R_DECL]]#0 : !fir.ref<!fir.box<!fir.heap<f32>>>
@@ -367,33 +371,35 @@ subroutine simple_loop_3
   integer :: i
   real, allocatable :: r;
   ! FIRDialect:  omp.parallel
-  ! FIRDialect:     %[[ALLOCA_IV:.*]] = fir.alloca i32 {{{.*}}, pinned}
-  ! FIRDialect:     %[[ALLOCA_IV_DECL:.*]]:2 = hlfir.declare %[[ALLOCA_IV]] {uniq_name = "{{.*}}Ei"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
+  ! FIRDialect:      %[[ALLOCA_IV:.*]] = fir.alloca i32 {{{.*}}, pinned}
+  ! FIRDialect:      %[[ALLOCA_IV_DECL:.*]]:2 = hlfir.declare %[[ALLOCA_IV]] {uniq_name = "{{.*}}Ei"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
 
-  ! FIRDialect:     [[R:%.*]] = fir.alloca !fir.box<!fir.heap<f32>> {bindc_name = "r", pinned, uniq_name = "{{.*}}Er"}
-  ! FIRDialect:     fir.store {{%.*}} to [[R]] : !fir.ref<!fir.box<!fir.heap<f32>>>
-  ! FIRDialect:     fir.store {{%.*}} to [[R]] : !fir.ref<!fir.box<!fir.heap<f32>>>
-  ! FIRDialect:     [[R_DECL:%.*]]:2 = hlfir.declare [[R]] {fortran_attrs = #fir.var_attrs<allocatable>, uniq_name = "{{.*}}Er"} : (!fir.ref<!fir.box<!fir.heap<f32>>>) -> (!fir.ref<!fir.box<!fir.heap<f32>>>, !fir.ref<!fir.box<!fir.heap<f32>>>)
+  ! FIRDialect:      [[R:%.*]] = fir.alloca !fir.box<!fir.heap<f32>> {bindc_name = "r", pinned, uniq_name = "{{.*}}Er"}
+  ! FIRDialect:      fir.store {{%.*}} to [[R]] : !fir.ref<!fir.box<!fir.heap<f32>>>
+  ! FIRDialect:      fir.store {{%.*}} to [[R]] : !fir.ref<!fir.box<!fir.heap<f32>>>
+  ! FIRDialect:      [[R_DECL:%.*]]:2 = hlfir.declare [[R]] {fortran_attrs = #fir.var_attrs<allocatable>, uniq_name = "{{.*}}Er"} : (!fir.ref<!fir.box<!fir.heap<f32>>>) -> (!fir.ref<!fir.box<!fir.heap<f32>>>, !fir.ref<!fir.box<!fir.heap<f32>>>)
 
-  ! FIRDialect:     %[[WS_LB:.*]] = arith.constant 1 : i32
-  ! FIRDialect:     %[[WS_UB:.*]] = arith.constant 9 : i32
-  ! FIRDialect:     %[[WS_STEP:.*]] = arith.constant 1 : i32
+  ! FIRDialect:      %[[WS_LB:.*]] = arith.constant 1 : i32
+  ! FIRDialect:      %[[WS_UB:.*]] = arith.constant 9 : i32
+  ! FIRDialect:      %[[WS_STEP:.*]] = arith.constant 1 : i32
 
-  ! FIRDialect:     omp.wsloop for (%[[I:.*]]) : i32 = (%[[WS_LB]]) to (%[[WS_UB]]) inclusive step (%[[WS_STEP]])
+  ! FIRDialect:      omp.wsloop {
+  ! FIRDialect-NEXT: omp.loop_nest (%[[I:.*]]) : i32 = (%[[WS_LB]]) to (%[[WS_UB]]) inclusive step (%[[WS_STEP]]) {
   !$OMP PARALLEL DO PRIVATE(r)
   do i=1, 9
-  ! FIRDialect:     fir.store %[[I]] to %[[ALLOCA_IV_DECL:.*]]#1 : !fir.ref<i32>
-  ! FIRDialect:     %[[LOAD_IV:.*]] = fir.load %[[ALLOCA_IV_DECL]]#0 : !fir.ref<i32>
-  ! FIRDialect:     fir.call @_FortranAioOutputInteger32({{.*}}, %[[LOAD_IV]]) {{.*}}: (!fir.ref<i8>, i32) -> i1
+  ! FIRDialect:      fir.store %[[I]] to %[[ALLOCA_IV_DECL:.*]]#1 : !fir.ref<i32>
+  ! FIRDialect:      %[[LOAD_IV:.*]] = fir.load %[[ALLOCA_IV_DECL]]#0 : !fir.ref<i32>
+  ! FIRDialect:      fir.call @_FortranAioOutputInteger32({{.*}}, %[[LOAD_IV]]) {{.*}}: (!fir.ref<i8>, i32) -> i1
     print*, i
   end do
-  ! FIRDialect:     omp.yield
-  ! FIRDialect:     {{%.*}} = fir.load [[R_DECL]]#0 : !fir.ref<!fir.box<!fir.heap<f32>>>
-  ! FIRDialect:     fir.if {{%.*}} {
-  ! FIRDialect:     [[LD:%.*]] = fir.load [[R_DECL]]#0 : !fir.ref<!fir.box<!fir.heap<f32>>>
-  ! FIRDialect:     [[AD:%.*]] = fir.box_addr [[LD]] : (!fir.box<!fir.heap<f32>>) -> !fir.heap<f32>
-  ! FIRDialect:     fir.freemem [[AD]] : !fir.heap<f32>
-  ! FIRDialect:     fir.store {{%.*}} to [[R_DECL]]#0 : !fir.ref<!fir.box<!fir.heap<f32>>>
+  ! FIRDialect:      omp.yield
+  ! FIRDialect:      omp.terminator
+  ! FIRDialect:      {{%.*}} = fir.load [[R_DECL]]#0 : !fir.ref<!fir.box<!fir.heap<f32>>>
+  ! FIRDialect:      fir.if {{%.*}} {
+  ! FIRDialect:      [[LD:%.*]] = fir.load [[R_DECL]]#0 : !fir.ref<!fir.box<!fir.heap<f32>>>
+  ! FIRDialect:      [[AD:%.*]] = fir.box_addr [[LD]] : (!fir.box<!fir.heap<f32>>) -> !fir.heap<f32>
+  ! FIRDialect:      fir.freemem [[AD]] : !fir.heap<f32>
+  ! FIRDialect:      fir.store {{%.*}} to [[R_DECL]]#0 : !fir.ref<!fir.box<!fir.heap<f32>>>
   !$OMP END PARALLEL DO
   ! FIRDialect:  omp.terminator
 end subroutine
diff --git a/flang/test/Lower/OpenMP/parallel-reduction-allocatable-array.f90 b/flang/test/Lower/OpenMP/parallel-reduction-allocatable-array.f90
new file mode 100644
index 000000000000..28216ef91c3a
--- /dev/null
+++ b/flang/test/Lower/OpenMP/parallel-reduction-allocatable-array.f90
@@ -0,0 +1,119 @@
+! RUN: bbc -emit-hlfir -fopenmp -o - %s | FileCheck %s
+! RUN: %flang_fc1 -emit-hlfir -fopenmp -o - %s | FileCheck %s
+
+program reduce
+integer :: i = 0
+integer, dimension(:), allocatable :: r
+
+allocate(r(2))
+
+!$omp parallel do reduction(+:r)
+do i=0,10
+  r(1) = i
+  r(2) = -i
+enddo
+!$omp end parallel do
+
+print *,r
+
+end program
+
+! CHECK-LABEL:   omp.declare_reduction @add_reduction_byref_box_heap_Uxi32 : !fir.ref<!fir.box<!fir.heap<!fir.array<?xi32>>>> init {
+! CHECK:         ^bb0(%[[VAL_0:.*]]: !fir.ref<!fir.box<!fir.heap<!fir.array<?xi32>>>>):
+! CHECK:           %[[VAL_1:.*]] = arith.constant 0 : i32
+! CHECK:           %[[VAL_2:.*]] = fir.load %[[VAL_0]] : !fir.ref<!fir.box<!fir.heap<!fir.array<?xi32>>>>
+! CHECK:           %[[VAL_10:.*]] = fir.alloca !fir.box<!fir.heap<!fir.array<?xi32>>>
+! CHECK:           %[[ADDR:.*]] = fir.box_addr %[[VAL_2]] : (!fir.box<!fir.heap<!fir.array<?xi32>>>) -> !fir.heap<!fir.array<?xi32>>
+! CHECK:           %[[ADDRI:.*]] = fir.convert %[[ADDR]] : (!fir.heap<!fir.array<?xi32>>) -> i64
+! CHECK:           %[[C0_I64:.*]] = arith.constant 0 : i64
+! CHECK:           %[[IS_NULL:.*]] = arith.cmpi eq, %[[ADDRI]], %[[C0_I64]] : i64
+! CHECK:           fir.if %[[IS_NULL]] {
+! CHECK:             %[[NULL_BOX:.*]] = fir.embox %[[ADDR]] : (!fir.heap<!fir.array<?xi32>>) -> !fir.box<!fir.heap<!fir.array<?xi32>>>
+! CHECK:             fir.store %[[NULL_BOX]] to %[[VAL_10]] : !fir.ref<!fir.box<!fir.heap<!fir.array<?xi32>>>>
+! CHECK:           } else {
+! CHECK:             %[[VAL_3:.*]] = arith.constant 0 : index
+! CHECK:             %[[VAL_4:.*]]:3 = fir.box_dims %[[VAL_2]], %[[VAL_3]] : (!fir.box<!fir.heap<!fir.array<?xi32>>>, index) -> (index, index, index)
+! CHECK:             %[[VAL_5:.*]] = fir.shape %[[VAL_4]]#1 : (index) -> !fir.shape<1>
+! CHECK:             %[[VAL_6:.*]] = fir.allocmem !fir.array<?xi32>, %[[VAL_4]]#1 {bindc_name = ".tmp", uniq_name = ""}
+! CHECK:             %[[VAL_7:.*]] = arith.constant true
+! CHECK:             %[[VAL_8:.*]]:2 = hlfir.declare %[[VAL_6]](%[[VAL_5]]) {uniq_name = ".tmp"} : (!fir.heap<!fir.array<?xi32>>, !fir.shape<1>) -> (!fir.box<!fir.array<?xi32>>, !fir.heap<!fir.array<?xi32>>)
+! CHECK:             %[[C0:.*]] = arith.constant 0 : index
+! CHECK:             %[[DIMS:.*]]:3 = fir.box_dims %[[VAL_2]], %[[C0]] : (!fir.box<!fir.heap<!fir.array<?xi32>>>, index) -> (index, index, index)
+! CHECK:             %[[SHIFT:.*]] = fir.shape_shift %[[DIMS]]#0, %[[DIMS]]#1 : (index, index) -> !fir.shapeshift<1>
+! CHECK:             %[[REBOX:.*]] = fir.rebox %[[VAL_8]]#0(%[[SHIFT]]) : (!fir.box<!fir.array<?xi32>>, !fir.shapeshift<1>) -> !fir.box<!fir.heap<!fir.array<?xi32>>>
+! CHECK:             hlfir.assign %[[VAL_1]] to %[[REBOX]] : i32, !fir.box<!fir.heap<!fir.array<?xi32>>>
+! CHECK:             fir.store %[[REBOX]] to %[[VAL_10]] : !fir.ref<!fir.box<!fir.heap<!fir.array<?xi32>>>>
+! CHECK:           }
+! CHECK:           omp.yield(%[[VAL_10]] : !fir.ref<!fir.box<!fir.heap<!fir.array<?xi32>>>>)
+! CHECK:         } combiner {
+! CHECK:         ^bb0(%[[VAL_0:.*]]: !fir.ref<!fir.box<!fir.heap<!fir.array<?xi32>>>>, %[[VAL_1:.*]]: !fir.ref<!fir.box<!fir.heap<!fir.array<?xi32>>>>):
+! CHECK:           %[[VAL_2:.*]] = fir.load %[[VAL_0]] : !fir.ref<!fir.box<!fir.heap<!fir.array<?xi32>>>>
+! CHECK:           %[[VAL_3:.*]] = fir.load %[[VAL_1]] : !fir.ref<!fir.box<!fir.heap<!fir.array<?xi32>>>>
+! CHECK:           %[[VAL_4:.*]] = arith.constant 0 : index
+! CHECK:           %[[VAL_5:.*]]:3 = fir.box_dims %[[VAL_2]], %[[VAL_4]] : (!fir.box<!fir.heap<!fir.array<?xi32>>>, index) -> (index, index, index)
+! CHECK:           %[[VAL_6:.*]] = fir.shape_shift %[[VAL_5]]#0, %[[VAL_5]]#1 : (index, index) -> !fir.shapeshift<1>
+! CHECK:           %[[VAL_7:.*]] = arith.constant 1 : index
+! CHECK:           fir.do_loop %[[VAL_8:.*]] = %[[VAL_7]] to %[[VAL_5]]#1 step %[[VAL_7]] unordered {
+! CHECK:             %[[VAL_9:.*]] = fir.array_coor %[[VAL_2]](%[[VAL_6]]) %[[VAL_8]] : (!fir.box<!fir.heap<!fir.array<?xi32>>>, !fir.shapeshift<1>, index) -> !fir.ref<i32>
+! CHECK:             %[[VAL_10:.*]] = fir.array_coor %[[VAL_3]](%[[VAL_6]]) %[[VAL_8]] : (!fir.box<!fir.heap<!fir.array<?xi32>>>, !fir.shapeshift<1>, index) -> !fir.ref<i32>
+! CHECK:             %[[VAL_11:.*]] = fir.load %[[VAL_9]] : !fir.ref<i32>
+! CHECK:             %[[VAL_12:.*]] = fir.load %[[VAL_10]] : !fir.ref<i32>
+! CHECK:             %[[VAL_13:.*]] = arith.addi %[[VAL_11]], %[[VAL_12]] : i32
+! CHECK:             fir.store %[[VAL_13]] to %[[VAL_9]] : !fir.ref<i32>
+! CHECK:           }
+! CHECK:           omp.yield(%[[VAL_0]] : !fir.ref<!fir.box<!fir.heap<!fir.array<?xi32>>>>)
+! CHECK:         }  cleanup {
+! CHECK:         ^bb0(%[[VAL_0:.*]]: !fir.ref<!fir.box<!fir.heap<!fir.array<?xi32>>>>):
+! CHECK:           %[[VAL_1:.*]] = fir.load %[[VAL_0]] : !fir.ref<!fir.box<!fir.heap<!fir.array<?xi32>>>>
+! CHECK:           %[[VAL_2:.*]] = fir.box_addr %[[VAL_1]] : (!fir.box<!fir.heap<!fir.array<?xi32>>>) -> !fir.heap<!fir.array<?xi32>>
+! CHECK:           %[[VAL_3:.*]] = fir.convert %[[VAL_2]] : (!fir.heap<!fir.array<?xi32>>) -> i64
+! CHECK:           %[[VAL_4:.*]] = arith.constant 0 : i64
+! CHECK:           %[[VAL_5:.*]] = arith.cmpi ne, %[[VAL_3]], %[[VAL_4]] : i64
+! CHECK:           fir.if %[[VAL_5]] {
+! CHECK:             fir.freemem %[[VAL_2]] : !fir.heap<!fir.array<?xi32>>
+! CHECK:           }
+! CHECK:           omp.yield
+! CHECK:         }
+
+! CHECK-LABEL:   func.func @_QQmain() attributes {fir.bindc_name = "reduce"} {
+! CHECK:           %[[VAL_0:.*]] = fir.address_of(@_QFEi) : !fir.ref<i32>
+! CHECK:           %[[VAL_1:.*]]:2 = hlfir.declare %[[VAL_0]] {uniq_name = "_QFEi"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
+! CHECK:           %[[VAL_2:.*]] = fir.address_of(@_QFEr) : !fir.ref<!fir.box<!fir.heap<!fir.array<?xi32>>>>
+! CHECK:           %[[VAL_3:.*]]:2 = hlfir.declare %[[VAL_2]] {fortran_attrs = {{.*}}<allocatable>, uniq_name = "_QFEr"} : (!fir.ref<!fir.box<!fir.heap<!fir.array<?xi32>>>>) -> (!fir.ref<!fir.box<!fir.heap<!fir.array<?xi32>>>>, !fir.ref<!fir.box<!fir.heap<!fir.array<?xi32>>>>)
+! CHECK:           %[[VAL_4:.*]] = arith.constant 2 : i32
+! CHECK:           %[[VAL_5:.*]] = fir.convert %[[VAL_4]] : (i32) -> index
+! CHECK:           %[[VAL_6:.*]] = arith.constant 0 : index
+! CHECK:           %[[VAL_7:.*]] = arith.cmpi sgt, %[[VAL_5]], %[[VAL_6]] : index
+! CHECK:           %[[VAL_8:.*]] = arith.select %[[VAL_7]], %[[VAL_5]], %[[VAL_6]] : index
+! CHECK:           %[[VAL_9:.*]] = fir.allocmem !fir.array<?xi32>, %[[VAL_8]] {fir.must_be_heap = true, uniq_name = "_QFEr.alloc"}
+! CHECK:           %[[VAL_10:.*]] = fir.shape %[[VAL_8]] : (index) -> !fir.shape<1>
+! CHECK:           %[[VAL_11:.*]] = fir.embox %[[VAL_9]](%[[VAL_10]]) : (!fir.heap<!fir.array<?xi32>>, !fir.shape<1>) -> !fir.box<!fir.heap<!fir.array<?xi32>>>
+! CHECK:           fir.store %[[VAL_11]] to %[[VAL_3]]#1 : !fir.ref<!fir.box<!fir.heap<!fir.array<?xi32>>>>
+! CHECK:           omp.parallel {
+! CHECK:             %[[VAL_12:.*]] = fir.alloca i32 {adapt.valuebyref, pinned}
+! CHECK:             %[[VAL_13:.*]]:2 = hlfir.declare %[[VAL_12]] {uniq_name = "_QFEi"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
+! CHECK:             %[[VAL_14:.*]] = arith.constant 0 : i32
+! CHECK:             %[[VAL_15:.*]] = arith.constant 10 : i32
+! CHECK:             %[[VAL_16:.*]] = arith.constant 1 : i32
+! CHECK:             omp.wsloop byref reduction(@add_reduction_byref_box_heap_Uxi32 %[[VAL_3]]#0 -> %[[VAL_17:.*]] : !fir.ref<!fir.box<!fir.heap<!fir.array<?xi32>>>>) {
+! CHECK-NEXT:          omp.loop_nest (%[[VAL_18:.*]]) : i32 = (%[[VAL_14]]) to (%[[VAL_15]]) inclusive step (%[[VAL_16]]) {
+! CHECK:                 %[[VAL_19:.*]]:2 = hlfir.declare %[[VAL_17]] {fortran_attrs = {{.*}}<allocatable>, uniq_name = "_QFEr"} : (!fir.ref<!fir.box<!fir.heap<!fir.array<?xi32>>>>) -> (!fir.ref<!fir.box<!fir.heap<!fir.array<?xi32>>>>, !fir.ref<!fir.box<!fir.heap<!fir.array<?xi32>>>>)
+! CHECK:                 fir.store %[[VAL_18]] to %[[VAL_13]]#1 : !fir.ref<i32>
+! CHECK:                 %[[VAL_20:.*]] = fir.load %[[VAL_13]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_21:.*]] = fir.load %[[VAL_19]]#0 : !fir.ref<!fir.box<!fir.heap<!fir.array<?xi32>>>>
+! CHECK:                 %[[VAL_22:.*]] = arith.constant 1 : index
+! CHECK:                 %[[VAL_23:.*]] = hlfir.designate %[[VAL_21]] (%[[VAL_22]])  : (!fir.box<!fir.heap<!fir.array<?xi32>>>, index) -> !fir.ref<i32>
+! CHECK:                 hlfir.assign %[[VAL_20]] to %[[VAL_23]] : i32, !fir.ref<i32>
+! CHECK:                 %[[VAL_24:.*]] = fir.load %[[VAL_13]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_25:.*]] = arith.constant 0 : i32
+! CHECK:                 %[[VAL_26:.*]] = arith.subi %[[VAL_25]], %[[VAL_24]] : i32
+! CHECK:                 %[[VAL_27:.*]] = fir.load %[[VAL_19]]#0 : !fir.ref<!fir.box<!fir.heap<!fir.array<?xi32>>>>
+! CHECK:                 %[[VAL_28:.*]] = arith.constant 2 : index
+! CHECK:                 %[[VAL_29:.*]] = hlfir.designate %[[VAL_27]] (%[[VAL_28]])  : (!fir.box<!fir.heap<!fir.array<?xi32>>>, index) -> !fir.ref<i32>
+! CHECK:                 hlfir.assign %[[VAL_26]] to %[[VAL_29]] : i32, !fir.ref<i32>
+! CHECK:                 omp.yield
+! CHECK:               }
+! CHECK:               omp.terminator
+! CHECK:             }
+! CHECK:             omp.terminator
+! CHECK:           }
diff --git a/flang/test/Lower/OpenMP/parallel-reduction-array-lb.f90 b/flang/test/Lower/OpenMP/parallel-reduction-array-lb.f90
new file mode 100644
index 000000000000..8202e6d89715
--- /dev/null
+++ b/flang/test/Lower/OpenMP/parallel-reduction-array-lb.f90
@@ -0,0 +1,90 @@
+! RUN: bbc -emit-hlfir -fopenmp -o - %s 2>&1 | FileCheck %s
+! RUN: %flang_fc1 -emit-hlfir -fopenmp -o - %s 2>&1 | FileCheck %s
+
+program reduce
+integer, dimension(2:4, 2) :: i = 0
+
+!$omp parallel reduction(+:i)
+i(3, 1) = 3
+!$omp end parallel
+
+print *,i
+
+end program
+
+! CHECK-LABEL:   omp.declare_reduction @add_reduction_byref_box_3x2xi32 : !fir.ref<!fir.box<!fir.array<3x2xi32>>> init {
+! CHECK:         ^bb0(%[[VAL_0:.*]]: !fir.ref<!fir.box<!fir.array<3x2xi32>>>):
+! CHECK:           %[[VAL_1:.*]] = arith.constant 0 : i32
+! CHECK:           %[[VAL_2:.*]] = fir.load %[[VAL_0]] : !fir.ref<!fir.box<!fir.array<3x2xi32>>>
+! CHECK:           %[[VAL_15:.*]] = fir.alloca !fir.box<!fir.array<3x2xi32>>
+! CHECK:           %[[VAL_3:.*]] = arith.constant 3 : index
+! CHECK:           %[[VAL_4:.*]] = arith.constant 2 : index
+! CHECK:           %[[VAL_5:.*]] = fir.shape %[[VAL_3]], %[[VAL_4]] : (index, index) -> !fir.shape<2>
+! CHECK:           %[[VAL_6:.*]] = fir.allocmem !fir.array<3x2xi32> {bindc_name = ".tmp", uniq_name = ""}
+! CHECK:           %[[VAL_7:.*]] = arith.constant true
+! CHECK:           %[[VAL_8:.*]]:2 = hlfir.declare %[[VAL_6]](%[[VAL_5]]) {uniq_name = ".tmp"} : (!fir.heap<!fir.array<3x2xi32>>, !fir.shape<2>) -> (!fir.heap<!fir.array<3x2xi32>>, !fir.heap<!fir.array<3x2xi32>>)
+! CHECK:           %[[VAL_9:.*]] = arith.constant 0 : index
+! CHECK:           %[[VAL_10:.*]]:3 = fir.box_dims %[[VAL_2]], %[[VAL_9]] : (!fir.box<!fir.array<3x2xi32>>, index) -> (index, index, index)
+! CHECK:           %[[VAL_11:.*]] = arith.constant 1 : index
+! CHECK:           %[[VAL_12:.*]]:3 = fir.box_dims %[[VAL_2]], %[[VAL_11]] : (!fir.box<!fir.array<3x2xi32>>, index) -> (index, index, index)
+! CHECK:           %[[VAL_13:.*]] = fir.shape_shift %[[VAL_10]]#0, %[[VAL_10]]#1, %[[VAL_12]]#0, %[[VAL_12]]#1 : (index, index, index, index) -> !fir.shapeshift<2>
+! CHECK:           %[[VAL_14:.*]] = fir.embox %[[VAL_8]]#0(%[[VAL_13]]) : (!fir.heap<!fir.array<3x2xi32>>, !fir.shapeshift<2>) -> !fir.box<!fir.array<3x2xi32>>
+! CHECK:           hlfir.assign %[[VAL_1]] to %[[VAL_14]] : i32, !fir.box<!fir.array<3x2xi32>>
+! CHECK:           fir.store %[[VAL_14]] to %[[VAL_15]] : !fir.ref<!fir.box<!fir.array<3x2xi32>>>
+! CHECK:           omp.yield(%[[VAL_15]] : !fir.ref<!fir.box<!fir.array<3x2xi32>>>)
+! CHECK:         } combiner {
+! CHECK:         ^bb0(%[[VAL_0:.*]]: !fir.ref<!fir.box<!fir.array<3x2xi32>>>, %[[VAL_1:.*]]: !fir.ref<!fir.box<!fir.array<3x2xi32>>>):
+! CHECK:           %[[VAL_2:.*]] = fir.load %[[VAL_0]] : !fir.ref<!fir.box<!fir.array<3x2xi32>>>
+! CHECK:           %[[VAL_3:.*]] = fir.load %[[VAL_1]] : !fir.ref<!fir.box<!fir.array<3x2xi32>>>
+! CHECK:           %[[VAL_4:.*]] = arith.constant 0 : index
+! CHECK:           %[[VAL_5:.*]]:3 = fir.box_dims %[[VAL_2]], %[[VAL_4]] : (!fir.box<!fir.array<3x2xi32>>, index) -> (index, index, index)
+! CHECK:           %[[VAL_6:.*]] = arith.constant 1 : index
+! CHECK:           %[[VAL_7:.*]]:3 = fir.box_dims %[[VAL_2]], %[[VAL_6]] : (!fir.box<!fir.array<3x2xi32>>, index) -> (index, index, index)
+! CHECK:           %[[VAL_8:.*]] = fir.shape_shift %[[VAL_5]]#0, %[[VAL_5]]#1, %[[VAL_7]]#0, %[[VAL_7]]#1 : (index, index, index, index) -> !fir.shapeshift<2>
+! CHECK:           %[[VAL_9:.*]] = arith.constant 1 : index
+! CHECK:           fir.do_loop %[[VAL_10:.*]] = %[[VAL_9]] to %[[VAL_7]]#1 step %[[VAL_9]] unordered {
+! CHECK:             fir.do_loop %[[VAL_11:.*]] = %[[VAL_9]] to %[[VAL_5]]#1 step %[[VAL_9]] unordered {
+! CHECK:               %[[VAL_12:.*]] = fir.array_coor %[[VAL_2]](%[[VAL_8]]) %[[VAL_11]], %[[VAL_10]] : (!fir.box<!fir.array<3x2xi32>>, !fir.shapeshift<2>, index, index) -> !fir.ref<i32>
+! CHECK:               %[[VAL_13:.*]] = fir.array_coor %[[VAL_3]](%[[VAL_8]]) %[[VAL_11]], %[[VAL_10]] : (!fir.box<!fir.array<3x2xi32>>, !fir.shapeshift<2>, index, index) -> !fir.ref<i32>
+! CHECK:               %[[VAL_14:.*]] = fir.load %[[VAL_12]] : !fir.ref<i32>
+! CHECK:               %[[VAL_15:.*]] = fir.load %[[VAL_13]] : !fir.ref<i32>
+! CHECK:               %[[VAL_16:.*]] = arith.addi %[[VAL_14]], %[[VAL_15]] : i32
+! CHECK:               fir.store %[[VAL_16]] to %[[VAL_12]] : !fir.ref<i32>
+! CHECK:             }
+! CHECK:           }
+! CHECK:           omp.yield(%[[VAL_0]] : !fir.ref<!fir.box<!fir.array<3x2xi32>>>)
+! CHECK:         }  cleanup {
+! CHECK:         ^bb0(%[[VAL_0:.*]]: !fir.ref<!fir.box<!fir.array<3x2xi32>>>):
+! CHECK:           %[[VAL_1:.*]] = fir.load %[[VAL_0]] : !fir.ref<!fir.box<!fir.array<3x2xi32>>>
+! CHECK:           %[[VAL_2:.*]] = fir.box_addr %[[VAL_1]] : (!fir.box<!fir.array<3x2xi32>>) -> !fir.ref<!fir.array<3x2xi32>>
+! CHECK:           %[[VAL_3:.*]] = fir.convert %[[VAL_2]] : (!fir.ref<!fir.array<3x2xi32>>) -> i64
+! CHECK:           %[[VAL_4:.*]] = arith.constant 0 : i64
+! CHECK:           %[[VAL_5:.*]] = arith.cmpi ne, %[[VAL_3]], %[[VAL_4]] : i64
+! CHECK:           fir.if %[[VAL_5]] {
+! CHECK:             %[[VAL_6:.*]] = fir.convert %[[VAL_2]] : (!fir.ref<!fir.array<3x2xi32>>) -> !fir.heap<!fir.array<3x2xi32>>
+! CHECK:             fir.freemem %[[VAL_6]] : !fir.heap<!fir.array<3x2xi32>>
+! CHECK:           }
+! CHECK:           omp.yield
+! CHECK:         }
+
+! CHECK-LABEL:   func.func @_QQmain() attributes {fir.bindc_name = "reduce"} {
+! CHECK:           %[[VAL_0:.*]] = fir.address_of(@_QFEi) : !fir.ref<!fir.array<3x2xi32>>
+! CHECK:           %[[VAL_1:.*]] = arith.constant 2 : index
+! CHECK:           %[[VAL_2:.*]] = arith.constant 3 : index
+! CHECK:           %[[VAL_3:.*]] = arith.constant 1 : index
+! CHECK:           %[[VAL_4:.*]] = arith.constant 2 : index
+! CHECK:           %[[VAL_5:.*]] = fir.shape_shift %[[VAL_1]], %[[VAL_2]], %[[VAL_3]], %[[VAL_4]] : (index, index, index, index) -> !fir.shapeshift<2>
+! CHECK:           %[[VAL_6:.*]]:2 = hlfir.declare %[[VAL_0]](%[[VAL_5]]) {uniq_name = "_QFEi"} : (!fir.ref<!fir.array<3x2xi32>>, !fir.shapeshift<2>) -> (!fir.box<!fir.array<3x2xi32>>, !fir.ref<!fir.array<3x2xi32>>)
+! CHECK:           %[[VAL_7:.*]] = fir.alloca !fir.box<!fir.array<3x2xi32>>
+! CHECK:           fir.store %[[VAL_6]]#0 to %[[VAL_7]] : !fir.ref<!fir.box<!fir.array<3x2xi32>>>
+! CHECK:           omp.parallel byref reduction(@add_reduction_byref_box_3x2xi32 %[[VAL_7]] -> %[[VAL_8:.*]] : !fir.ref<!fir.box<!fir.array<3x2xi32>>>) {
+! CHECK:             %[[VAL_9:.*]]:2 = hlfir.declare %[[VAL_8]] {uniq_name = "_QFEi"} : (!fir.ref<!fir.box<!fir.array<3x2xi32>>>) -> (!fir.ref<!fir.box<!fir.array<3x2xi32>>>, !fir.ref<!fir.box<!fir.array<3x2xi32>>>)
+! CHECK:             %[[VAL_10:.*]] = arith.constant 3 : i32
+! CHECK:             %[[VAL_11:.*]] = fir.load %[[VAL_9]]#0 : !fir.ref<!fir.box<!fir.array<3x2xi32>>>
+! CHECK:             %[[VAL_12:.*]] = arith.constant 3 : index
+! CHECK:             %[[VAL_13:.*]] = arith.constant 1 : index
+! CHECK:             %[[VAL_14:.*]] = hlfir.designate %[[VAL_11]] (%[[VAL_12]], %[[VAL_13]])  : (!fir.box<!fir.array<3x2xi32>>, index, index) -> !fir.ref<i32>
+! CHECK:             hlfir.assign %[[VAL_10]] to %[[VAL_14]] : i32, !fir.ref<i32>
+! CHECK:             omp.terminator
+! CHECK:           }
+
diff --git a/flang/test/Lower/OpenMP/parallel-reduction-array.f90 b/flang/test/Lower/OpenMP/parallel-reduction-array.f90
index 26c9d4f08509..34f4ee0a9eb3 100644
--- a/flang/test/Lower/OpenMP/parallel-reduction-array.f90
+++ b/flang/test/Lower/OpenMP/parallel-reduction-array.f90
@@ -17,15 +17,18 @@ end program
 ! CHECK:         ^bb0(%[[VAL_0:.*]]: !fir.ref<!fir.box<!fir.array<3xi32>>>):
 ! CHECK:           %[[VAL_2:.*]] = arith.constant 0 : i32
 ! CHECK:           %[[VAL_3:.*]] = fir.load %[[VAL_0]] : !fir.ref<!fir.box<!fir.array<3xi32>>>
+! CHECK:           %[[VAL_8:.*]] = fir.alloca !fir.box<!fir.array<3xi32>>
 ! CHECK:           %[[VAL_4:.*]] = arith.constant 3 : index
 ! CHECK:           %[[VAL_5:.*]] = fir.shape %[[VAL_4]] : (index) -> !fir.shape<1>
 ! CHECK:           %[[VAL_1:.*]] = fir.allocmem !fir.array<3xi32> {bindc_name = ".tmp", uniq_name = ""}
 ! CHECK:           %[[TRUE:.*]]  = arith.constant true
 ! CHECK:           %[[VAL_6:.*]]:2 = hlfir.declare %[[VAL_1]](%[[VAL_5]]) {uniq_name = ".tmp"} : (!fir.heap<!fir.array<3xi32>>,
 !fir.shape<1>) -> (!fir.heap<!fir.array<3xi32>>, !fir.heap<!fir.array<3xi32>>)
-! CHECK:           %[[VAL_7:.*]] = fir.embox %[[VAL_6]]#0(%[[VAL_5]]) : (!fir.heap<!fir.array<3xi32>>, !fir.shape<1>) -> !fir.box<!fir.array<3xi32>>
+! CHECK:           %[[C0:.*]] = arith.constant 0 : index
+! CHECK:           %[[DIMS:.*]]:3 = fir.box_dims %[[VAL_3]], %[[C0]] : (!fir.box<!fir.array<3xi32>>, index) -> (index, index, index)
+! CHECK:           %[[SHIFT:.*]] = fir.shape_shift %[[DIMS]]#0, %[[DIMS]]#1 : (index, index) -> !fir.shapeshift<1>
+! CHECK:           %[[VAL_7:.*]] = fir.embox %[[VAL_6]]#0(%[[SHIFT]]) : (!fir.heap<!fir.array<3xi32>>, !fir.shapeshift<1>) -> !fir.box<!fir.array<3xi32>>
 ! CHECK:           hlfir.assign %[[VAL_2]] to %[[VAL_7]] : i32, !fir.box<!fir.array<3xi32>>
-! CHECK:           %[[VAL_8:.*]] = fir.alloca !fir.box<!fir.array<3xi32>>
 ! CHECK:           fir.store %[[VAL_7]] to %[[VAL_8]] : !fir.ref<!fir.box<!fir.array<3xi32>>>
 ! CHECK:           omp.yield(%[[VAL_8]] : !fir.ref<!fir.box<!fir.array<3xi32>>>)
 ! CHECK:         } combiner {
diff --git a/flang/test/Lower/OpenMP/parallel-reduction-array2.f90 b/flang/test/Lower/OpenMP/parallel-reduction-array2.f90
index bed04401248b..aa14092554ed 100644
--- a/flang/test/Lower/OpenMP/parallel-reduction-array2.f90
+++ b/flang/test/Lower/OpenMP/parallel-reduction-array2.f90
@@ -17,15 +17,17 @@ end program
 ! CHECK:         ^bb0(%[[VAL_0:.*]]: !fir.ref<!fir.box<!fir.array<3xi32>>>):
 ! CHECK:           %[[VAL_2:.*]] = arith.constant 0 : i32
 ! CHECK:           %[[VAL_3:.*]] = fir.load %[[VAL_0]] : !fir.ref<!fir.box<!fir.array<3xi32>>>
+! CHECK:           %[[VAL_8:.*]] = fir.alloca !fir.box<!fir.array<3xi32>>
 ! CHECK:           %[[VAL_4:.*]] = arith.constant 3 : index
 ! CHECK:           %[[VAL_5:.*]] = fir.shape %[[VAL_4]] : (index) -> !fir.shape<1>
 ! CHECK:           %[[VAL_1:.*]] = fir.allocmem !fir.array<3xi32>
 ! CHECK:           %[[TRUE:.*]]  = arith.constant true
-! CHECK:           %[[VAL_6:.*]]:2 = hlfir.declare %[[VAL_1]](%[[VAL_5]]) {uniq_name = ".tmp"} : (!fir.heap<!fir.array<3xi32>>,
-!fir.shape<1>) -> (!fir.heap<!fir.array<3xi32>>, !fir.heap<!fir.array<3xi32>>)
-! CHECK:           %[[VAL_7:.*]] = fir.embox %[[VAL_6]]#0(%[[VAL_5]]) : (!fir.heap<!fir.array<3xi32>>, !fir.shape<1>) -> !fir.box<!fir.array<3xi32>>
+! CHECK:           %[[VAL_6:.*]]:2 = hlfir.declare %[[VAL_1]](%[[VAL_5]]) {uniq_name = ".tmp"} : (!fir.heap<!fir.array<3xi32>>, !fir.shape<1>) -> (!fir.heap<!fir.array<3xi32>>, !fir.heap<!fir.array<3xi32>>)
+! CHECK:           %[[C0:.*]] = arith.constant 0 : index
+! CHECK:           %[[DIMS:.*]]:3 = fir.box_dims %[[VAL_3]], %[[C0]] : (!fir.box<!fir.array<3xi32>>, index) -> (index, index, index)
+! CHECK:           %[[SHIFT:.*]] = fir.shape_shift %[[DIMS]]#0, %[[DIMS]]#1 : (index, index) -> !fir.shapeshift<1>
+! CHECK:           %[[VAL_7:.*]] = fir.embox %[[VAL_6]]#0(%[[SHIFT]]) : (!fir.heap<!fir.array<3xi32>>, !fir.shapeshift<1>) -> !fir.box<!fir.array<3xi32>>
 ! CHECK:           hlfir.assign %[[VAL_2]] to %[[VAL_7]] : i32, !fir.box<!fir.array<3xi32>>
-! CHECK:           %[[VAL_8:.*]] = fir.alloca !fir.box<!fir.array<3xi32>>
 ! CHECK:           fir.store %[[VAL_7]] to %[[VAL_8]] : !fir.ref<!fir.box<!fir.array<3xi32>>>
 ! CHECK:           omp.yield(%[[VAL_8]] : !fir.ref<!fir.box<!fir.array<3xi32>>>)
 ! CHECK:         } combiner {
diff --git a/flang/test/Lower/OpenMP/parallel-reduction3.f90 b/flang/test/Lower/OpenMP/parallel-reduction3.f90
index ce6bd17265dd..2a4e338f255e 100644
--- a/flang/test/Lower/OpenMP/parallel-reduction3.f90
+++ b/flang/test/Lower/OpenMP/parallel-reduction3.f90
@@ -5,15 +5,19 @@
 ! CHECK:         ^bb0(%[[VAL_0:.*]]: !fir.ref<!fir.box<!fir.array<?xi32>>>):
 ! CHECK:           %[[VAL_1:.*]] = arith.constant 0 : i32
 ! CHECK:           %[[VAL_2:.*]] = fir.load %[[VAL_0]] : !fir.ref<!fir.box<!fir.array<?xi32>>>
+! CHECK:           %[[VAL_8:.*]] = fir.alloca !fir.box<!fir.array<?xi32>>
 ! CHECK:           %[[VAL_3:.*]] = arith.constant 0 : index
 ! CHECK:           %[[VAL_4:.*]]:3 = fir.box_dims %[[VAL_2]], %[[VAL_3]] : (!fir.box<!fir.array<?xi32>>, index) -> (index, index, index)
 ! CHECK:           %[[VAL_5:.*]] = fir.shape %[[VAL_4]]#1 : (index) -> !fir.shape<1>
 ! CHECK:           %[[VAL_6:.*]] = fir.allocmem !fir.array<?xi32>, %[[VAL_4]]#1 {bindc_name = ".tmp", uniq_name = ""}
 ! CHECK:           %[[TRUE:.*]]  = arith.constant true
 ! CHECK:           %[[VAL_7:.*]]:2 = hlfir.declare %[[VAL_6]](%[[VAL_5]]) {uniq_name = ".tmp"} : (!fir.heap<!fir.array<?xi32>>, !fir.shape<1>) -> (!fir.box<!fir.array<?xi32>>, !fir.heap<!fir.array<?xi32>>)
-! CHECK:           hlfir.assign %[[VAL_1]] to %[[VAL_7]]#0 : i32, !fir.box<!fir.array<?xi32>>
-! CHECK:           %[[VAL_8:.*]] = fir.alloca !fir.box<!fir.array<?xi32>>
-! CHECK:           fir.store %[[VAL_7]]#0 to %[[VAL_8]] : !fir.ref<!fir.box<!fir.array<?xi32>>>
+! CHECK:           %[[C0:.*]] = arith.constant 0 : index
+! CHECK:           %[[DIMS:.*]]:3 = fir.box_dims %[[VAL_2]], %[[C0]] : (!fir.box<!fir.array<?xi32>>, index) -> (index, index, index)
+! CHECK:           %[[SHIFT:.*]] = fir.shape_shift %[[DIMS]]#0, %[[DIMS]]#1 : (index, index) -> !fir.shapeshift<1>
+! CHECK:           %[[REBOX:.*]] = fir.rebox %[[VAL_7]]#0(%[[SHIFT]]) : (!fir.box<!fir.array<?xi32>>, !fir.shapeshift<1>) -> !fir.box<!fir.array<?xi32>>
+! CHECK:           hlfir.assign %[[VAL_1]] to %[[REBOX]] : i32, !fir.box<!fir.array<?xi32>>
+! CHECK:           fir.store %[[REBOX]] to %[[VAL_8]] : !fir.ref<!fir.box<!fir.array<?xi32>>>
 ! CHECK:           omp.yield(%[[VAL_8]] : !fir.ref<!fir.box<!fir.array<?xi32>>>)
 ! CHECK:         } combiner {
 ! CHECK:         ^bb0(%[[VAL_0:.*]]: !fir.ref<!fir.box<!fir.array<?xi32>>>, %[[VAL_1:.*]]: !fir.ref<!fir.box<!fir.array<?xi32>>>):
@@ -70,30 +74,33 @@
 ! CHECK:             %[[VAL_18:.*]] = arith.constant 1 : i32
 ! CHECK:             %[[VAL_19:.*]] = fir.alloca !fir.box<!fir.array<?xi32>>
 ! CHECK:             fir.store %[[VAL_12]]#0 to %[[VAL_19]] : !fir.ref<!fir.box<!fir.array<?xi32>>>
-! CHECK:             omp.wsloop byref reduction(@add_reduction_byref_box_Uxi32 %[[VAL_19]] -> %[[VAL_20:.*]] : !fir.ref<!fir.box<!fir.array<?xi32>>>)  for  (%[[VAL_21:.*]]) : i32 = (%[[VAL_16]]) to (%[[VAL_17]]) inclusive step (%[[VAL_18]]) {
-! CHECK:               fir.store %[[VAL_21]] to %[[VAL_15]]#1 : !fir.ref<i32>
-! CHECK:               %[[VAL_22:.*]]:2 = hlfir.declare %[[VAL_20]] {uniq_name = "_QFsEc"} : (!fir.ref<!fir.box<!fir.array<?xi32>>>) -> (!fir.ref<!fir.box<!fir.array<?xi32>>>, !fir.ref<!fir.box<!fir.array<?xi32>>>)
-! CHECK:               %[[VAL_23:.*]] = fir.load %[[VAL_22]]#0 : !fir.ref<!fir.box<!fir.array<?xi32>>>
-! CHECK:               %[[VAL_24:.*]] = fir.load %[[VAL_15]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_25:.*]] = arith.constant 0 : index
-! CHECK:               %[[VAL_26:.*]]:3 = fir.box_dims %[[VAL_23]], %[[VAL_25]] : (!fir.box<!fir.array<?xi32>>, index) -> (index, index, index)
-! CHECK:               %[[VAL_27:.*]] = fir.shape %[[VAL_26]]#1 : (index) -> !fir.shape<1>
-! CHECK:               %[[VAL_28:.*]] = hlfir.elemental %[[VAL_27]] unordered : (!fir.shape<1>) -> !hlfir.expr<?xi32> {
-! CHECK:               ^bb0(%[[VAL_29:.*]]: index):
-! CHECK:                 %[[VAL_30:.*]] = arith.constant 0 : index
-! CHECK:                 %[[VAL_31:.*]]:3 = fir.box_dims %[[VAL_23]], %[[VAL_30]] : (!fir.box<!fir.array<?xi32>>, index) -> (index, index, index)
-! CHECK:                 %[[VAL_32:.*]] = arith.constant 1 : index
-! CHECK:                 %[[VAL_33:.*]] = arith.subi %[[VAL_31]]#0, %[[VAL_32]] : index
-! CHECK:                 %[[VAL_34:.*]] = arith.addi %[[VAL_29]], %[[VAL_33]] : index
-! CHECK:                 %[[VAL_35:.*]] = hlfir.designate %[[VAL_23]] (%[[VAL_34]])  : (!fir.box<!fir.array<?xi32>>, index) -> !fir.ref<i32>
-! CHECK:                 %[[VAL_36:.*]] = fir.load %[[VAL_35]] : !fir.ref<i32>
-! CHECK:                 %[[VAL_37:.*]] = arith.addi %[[VAL_36]], %[[VAL_24]] : i32
-! CHECK:                 hlfir.yield_element %[[VAL_37]] : i32
+! CHECK:             omp.wsloop byref reduction(@add_reduction_byref_box_Uxi32 %[[VAL_19]] -> %[[VAL_20:.*]] : !fir.ref<!fir.box<!fir.array<?xi32>>>) {
+! CHECK-NEXT:          omp.loop_nest (%[[VAL_21:.*]]) : i32 = (%[[VAL_16]]) to (%[[VAL_17]]) inclusive step (%[[VAL_18]]) {
+! CHECK:                 %[[VAL_22:.*]]:2 = hlfir.declare %[[VAL_20]] {uniq_name = "_QFsEc"} : (!fir.ref<!fir.box<!fir.array<?xi32>>>) -> (!fir.ref<!fir.box<!fir.array<?xi32>>>, !fir.ref<!fir.box<!fir.array<?xi32>>>)
+! CHECK:                 fir.store %[[VAL_21]] to %[[VAL_15]]#1 : !fir.ref<i32>
+! CHECK:                 %[[VAL_23:.*]] = fir.load %[[VAL_22]]#0 : !fir.ref<!fir.box<!fir.array<?xi32>>>
+! CHECK:                 %[[VAL_24:.*]] = fir.load %[[VAL_15]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_25:.*]] = arith.constant 0 : index
+! CHECK:                 %[[VAL_26:.*]]:3 = fir.box_dims %[[VAL_23]], %[[VAL_25]] : (!fir.box<!fir.array<?xi32>>, index) -> (index, index, index)
+! CHECK:                 %[[VAL_27:.*]] = fir.shape %[[VAL_26]]#1 : (index) -> !fir.shape<1>
+! CHECK:                 %[[VAL_28:.*]] = hlfir.elemental %[[VAL_27]] unordered : (!fir.shape<1>) -> !hlfir.expr<?xi32> {
+! CHECK:                 ^bb0(%[[VAL_29:.*]]: index):
+! CHECK:                   %[[VAL_30:.*]] = arith.constant 0 : index
+! CHECK:                   %[[VAL_31:.*]]:3 = fir.box_dims %[[VAL_23]], %[[VAL_30]] : (!fir.box<!fir.array<?xi32>>, index) -> (index, index, index)
+! CHECK:                   %[[VAL_32:.*]] = arith.constant 1 : index
+! CHECK:                   %[[VAL_33:.*]] = arith.subi %[[VAL_31]]#0, %[[VAL_32]] : index
+! CHECK:                   %[[VAL_34:.*]] = arith.addi %[[VAL_29]], %[[VAL_33]] : index
+! CHECK:                   %[[VAL_35:.*]] = hlfir.designate %[[VAL_23]] (%[[VAL_34]])  : (!fir.box<!fir.array<?xi32>>, index) -> !fir.ref<i32>
+! CHECK:                   %[[VAL_36:.*]] = fir.load %[[VAL_35]] : !fir.ref<i32>
+! CHECK:                   %[[VAL_37:.*]] = arith.addi %[[VAL_36]], %[[VAL_24]] : i32
+! CHECK:                   hlfir.yield_element %[[VAL_37]] : i32
+! CHECK:                 }
+! CHECK:                 %[[VAL_38:.*]] = fir.load %[[VAL_22]]#0 : !fir.ref<!fir.box<!fir.array<?xi32>>>
+! CHECK:                 hlfir.assign %[[VAL_28]] to %[[VAL_38]] : !hlfir.expr<?xi32>, !fir.box<!fir.array<?xi32>>
+! CHECK:                 hlfir.destroy %[[VAL_28]] : !hlfir.expr<?xi32>
+! CHECK:                 omp.yield
 ! CHECK:               }
-! CHECK:               %[[VAL_38:.*]] = fir.load %[[VAL_22]]#0 : !fir.ref<!fir.box<!fir.array<?xi32>>>
-! CHECK:               hlfir.assign %[[VAL_28]] to %[[VAL_38]] : !hlfir.expr<?xi32>, !fir.box<!fir.array<?xi32>>
-! CHECK:               hlfir.destroy %[[VAL_28]] : !hlfir.expr<?xi32>
-! CHECK:               omp.yield
+! CHECK:               omp.terminator
 ! CHECK:             }
 ! CHECK:             omp.terminator
 ! CHECK:           }
diff --git a/flang/test/Lower/OpenMP/parallel-wsloop-firstpriv.f90 b/flang/test/Lower/OpenMP/parallel-wsloop-firstpriv.f90
index 716a7d71bb62..ac8b9f50f54e 100644
--- a/flang/test/Lower/OpenMP/parallel-wsloop-firstpriv.f90
+++ b/flang/test/Lower/OpenMP/parallel-wsloop-firstpriv.f90
@@ -20,10 +20,14 @@ subroutine omp_do_firstprivate(a)
   ! CHECK: %[[LB:.*]] = arith.constant 1 : i32
   ! CHECK-NEXT: %[[UB:.*]] = fir.load %[[A_PVT_DECL]]#0 : !fir.ref<i32>
   ! CHECK-NEXT: %[[STEP:.*]] = arith.constant 1 : i32
-  ! CHECK-NEXT: omp.wsloop   for  (%[[ARG1:.*]]) : i32 = (%[[LB]]) to (%[[UB]]) inclusive step (%[[STEP]])
+  ! CHECK-NEXT: omp.wsloop {
+  ! CHECK-NEXT: omp.loop_nest (%[[ARG1:.*]]) : i32 = (%[[LB]]) to (%[[UB]]) inclusive step (%[[STEP]]) {
   ! CHECK-NEXT: fir.store %[[ARG1]] to %[[I_PVT_DECL]]#1 : !fir.ref<i32>
   ! CHECK-NEXT: fir.call @_QPfoo(%[[I_PVT_DECL]]#1, %[[A_PVT_DECL]]#1) {{.*}}: (!fir.ref<i32>, !fir.ref<i32>) -> ()
   ! CHECK-NEXT: omp.yield
+  ! CHECK-NEXT: }
+  ! CHECK-NEXT: omp.terminator
+  ! CHECK-NEXT: }
     do i=1, a
       call foo(i, a)
     end do
@@ -56,10 +60,12 @@ subroutine omp_do_firstprivate2(a, n)
   ! CHECK: %[[LB:.*]] = fir.load %[[A_PVT_DECL]]#0 : !fir.ref<i32>
   ! CHECK: %[[UB:.*]] = fir.load %[[N_PVT_DECL]]#0 : !fir.ref<i32>
   ! CHECK: %[[STEP:.*]] = arith.constant 1 : i32
-  ! CHECK: omp.wsloop   for  (%[[ARG2:.*]]) : i32 = (%[[LB]]) to (%[[UB]]) inclusive step (%[[STEP]])
+  ! CHECK: omp.wsloop {
+  ! CHECK-NEXT: omp.loop_nest (%[[ARG2:.*]]) : i32 = (%[[LB]]) to (%[[UB]]) inclusive step (%[[STEP]]) {
   ! CHECK: fir.store %[[ARG2]] to %[[I_PVT_DECL]]#1 : !fir.ref<i32>
   ! CHECK: fir.call @_QPfoo(%[[I_PVT_DECL]]#1, %[[A_PVT_DECL]]#1) {{.*}}: (!fir.ref<i32>, !fir.ref<i32>) -> ()
   ! CHECK: omp.yield
+  ! CHECK: omp.terminator
     do i= a, n
       call foo(i, a)
     end do
diff --git a/flang/test/Lower/OpenMP/parallel-wsloop.f90 b/flang/test/Lower/OpenMP/parallel-wsloop.f90
index c06f941b74b5..602b3d1c05f0 100644
--- a/flang/test/Lower/OpenMP/parallel-wsloop.f90
+++ b/flang/test/Lower/OpenMP/parallel-wsloop.f90
@@ -6,19 +6,21 @@
 subroutine simple_parallel_do
   integer :: i
   ! CHECK:  omp.parallel
-  ! CHECK:     %[[WS_LB:.*]] = arith.constant 1 : i32
-  ! CHECK:     %[[WS_UB:.*]] = arith.constant 9 : i32
-  ! CHECK:     %[[WS_STEP:.*]] = arith.constant 1 : i32
-  ! CHECK:     omp.wsloop for (%[[I:.*]]) : i32 = (%[[WS_LB]]) to (%[[WS_UB]]) inclusive step (%[[WS_STEP]])
+  ! CHECK:      %[[WS_LB:.*]] = arith.constant 1 : i32
+  ! CHECK:      %[[WS_UB:.*]] = arith.constant 9 : i32
+  ! CHECK:      %[[WS_STEP:.*]] = arith.constant 1 : i32
+  ! CHECK:      omp.wsloop {
+  ! CHECK-NEXT: omp.loop_nest (%[[I:.*]]) : i32 = (%[[WS_LB]]) to (%[[WS_UB]]) inclusive step (%[[WS_STEP]]) {
   !$OMP PARALLEL DO
   do i=1, 9
-  ! CHECK:    fir.store %[[I]] to %[[IV_ADDR:.*]]#1 : !fir.ref<i32>
-  ! CHECK:    %[[LOAD_IV:.*]] = fir.load %[[IV_ADDR]]#0 : !fir.ref<i32>
-  ! CHECK:    fir.call @_FortranAioOutputInteger32({{.*}}, %[[LOAD_IV]]) {{.*}}: (!fir.ref<i8>, i32) -> i1
+  ! CHECK:      fir.store %[[I]] to %[[IV_ADDR:.*]]#1 : !fir.ref<i32>
+  ! CHECK:      %[[LOAD_IV:.*]] = fir.load %[[IV_ADDR]]#0 : !fir.ref<i32>
+  ! CHECK:      fir.call @_FortranAioOutputInteger32({{.*}}, %[[LOAD_IV]]) {{.*}}: (!fir.ref<i8>, i32) -> i1
     print*, i
   end do
-  ! CHECK:       omp.yield
-  ! CHECK:       omp.terminator
+  ! CHECK:      omp.yield
+  ! CHECK:      omp.terminator
+  ! CHECK:      omp.terminator
   !$OMP END PARALLEL DO
 end subroutine
 
@@ -34,19 +36,21 @@ subroutine parallel_do_with_parallel_clauses(cond, nt)
   ! CHECK:  %[[COND_CVT:.*]] = fir.convert %[[COND]] : (!fir.logical<4>) -> i1
   ! CHECK:  %[[NT:.*]] = fir.load %[[NT_DECL]]#0 : !fir.ref<i32>
   ! CHECK:  omp.parallel if(%[[COND_CVT]] : i1) num_threads(%[[NT]] : i32) proc_bind(close)
-  ! CHECK:     %[[WS_LB:.*]] = arith.constant 1 : i32
-  ! CHECK:     %[[WS_UB:.*]] = arith.constant 9 : i32
-  ! CHECK:     %[[WS_STEP:.*]] = arith.constant 1 : i32
-  ! CHECK:     omp.wsloop for (%[[I:.*]]) : i32 = (%[[WS_LB]]) to (%[[WS_UB]]) inclusive step (%[[WS_STEP]])
+  ! CHECK:      %[[WS_LB:.*]] = arith.constant 1 : i32
+  ! CHECK:      %[[WS_UB:.*]] = arith.constant 9 : i32
+  ! CHECK:      %[[WS_STEP:.*]] = arith.constant 1 : i32
+  ! CHECK:      omp.wsloop {
+  ! CHECK-NEXT: omp.loop_nest (%[[I:.*]]) : i32 = (%[[WS_LB]]) to (%[[WS_UB]]) inclusive step (%[[WS_STEP]]) {
   !$OMP PARALLEL DO IF(cond) NUM_THREADS(nt) PROC_BIND(close)
   do i=1, 9
-  ! CHECK:    fir.store %[[I]] to %[[IV_ADDR:.*]]#1 : !fir.ref<i32>
-  ! CHECK:    %[[LOAD_IV:.*]] = fir.load %[[IV_ADDR]]#0 : !fir.ref<i32>
-  ! CHECK:    fir.call @_FortranAioOutputInteger32({{.*}}, %[[LOAD_IV]]) {{.*}}: (!fir.ref<i8>, i32) -> i1
+  ! CHECK:      fir.store %[[I]] to %[[IV_ADDR:.*]]#1 : !fir.ref<i32>
+  ! CHECK:      %[[LOAD_IV:.*]] = fir.load %[[IV_ADDR]]#0 : !fir.ref<i32>
+  ! CHECK:      fir.call @_FortranAioOutputInteger32({{.*}}, %[[LOAD_IV]]) {{.*}}: (!fir.ref<i8>, i32) -> i1
     print*, i
   end do
-  ! CHECK:       omp.yield
-  ! CHECK:       omp.terminator
+  ! CHECK:      omp.yield
+  ! CHECK:      omp.terminator
+  ! CHECK:      omp.terminator
   !$OMP END PARALLEL DO
 end subroutine
 
@@ -58,19 +62,21 @@ subroutine parallel_do_with_clauses(nt)
   integer :: i
   ! CHECK:  %[[NT:.*]] = fir.load %[[NT_DECL]]#0 : !fir.ref<i32>
   ! CHECK:  omp.parallel num_threads(%[[NT]] : i32)
-  ! CHECK:     %[[WS_LB:.*]] = arith.constant 1 : i32
-  ! CHECK:     %[[WS_UB:.*]] = arith.constant 9 : i32
-  ! CHECK:     %[[WS_STEP:.*]] = arith.constant 1 : i32
-  ! CHECK:     omp.wsloop schedule(dynamic) for (%[[I:.*]]) : i32 = (%[[WS_LB]]) to (%[[WS_UB]]) inclusive step (%[[WS_STEP]])
+  ! CHECK:      %[[WS_LB:.*]] = arith.constant 1 : i32
+  ! CHECK:      %[[WS_UB:.*]] = arith.constant 9 : i32
+  ! CHECK:      %[[WS_STEP:.*]] = arith.constant 1 : i32
+  ! CHECK:      omp.wsloop schedule(dynamic) {
+  ! CHECK-NEXT: omp.loop_nest (%[[I:.*]]) : i32 = (%[[WS_LB]]) to (%[[WS_UB]]) inclusive step (%[[WS_STEP]]) {
   !$OMP PARALLEL DO NUM_THREADS(nt) SCHEDULE(dynamic)
   do i=1, 9
-  ! CHECK:    fir.store %[[I]] to %[[IV_ADDR:.*]]#1 : !fir.ref<i32>
-  ! CHECK:    %[[LOAD_IV:.*]] = fir.load %[[IV_ADDR]]#0 : !fir.ref<i32>
-  ! CHECK:    fir.call @_FortranAioOutputInteger32({{.*}}, %[[LOAD_IV]]) {{.*}}: (!fir.ref<i8>, i32) -> i1
+  ! CHECK:      fir.store %[[I]] to %[[IV_ADDR:.*]]#1 : !fir.ref<i32>
+  ! CHECK:      %[[LOAD_IV:.*]] = fir.load %[[IV_ADDR]]#0 : !fir.ref<i32>
+  ! CHECK:      fir.call @_FortranAioOutputInteger32({{.*}}, %[[LOAD_IV]]) {{.*}}: (!fir.ref<i8>, i32) -> i1
     print*, i
   end do
-  ! CHECK:       omp.yield
-  ! CHECK:       omp.terminator
+  ! CHECK:      omp.yield
+  ! CHECK:      omp.terminator
+  ! CHECK:      omp.terminator
   !$OMP END PARALLEL DO
 end subroutine
 
@@ -88,20 +94,21 @@ subroutine parallel_do_with_privatisation_clauses(cond,nt)
   integer :: nt
   integer :: i
   ! CHECK:  omp.parallel
-  ! CHECK:    %[[PRIVATE_COND_REF:.*]] = fir.alloca !fir.logical<4> {bindc_name = "cond", pinned, uniq_name = "_QFparallel_do_with_privatisation_clausesEcond"}
-  ! CHECK:    %[[PRIVATE_COND_DECL:.*]]:2 = hlfir.declare %[[PRIVATE_COND_REF]] {uniq_name = "_QFparallel_do_with_privatisation_clausesEcond"} : (!fir.ref<!fir.logical<4>>) -> (!fir.ref<!fir.logical<4>>, !fir.ref<!fir.logical<4>>)
-  ! CHECK:    %[[PRIVATE_NT_REF:.*]] = fir.alloca i32 {bindc_name = "nt", pinned, uniq_name = "_QFparallel_do_with_privatisation_clausesEnt"}
-  ! CHECK:    %[[PRIVATE_NT_DECL:.*]]:2 = hlfir.declare %[[PRIVATE_NT_REF]] {uniq_name = "_QFparallel_do_with_privatisation_clausesEnt"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
-  ! CHECK:    %[[NT_VAL:.*]] = fir.load %[[NT_DECL]]#0 : !fir.ref<i32>
-  ! CHECK:    hlfir.assign %[[NT_VAL]] to %[[PRIVATE_NT_DECL]]#0 temporary_lhs : i32, !fir.ref<i32>
-  ! CHECK:    %[[WS_LB:.*]] = arith.constant 1 : i32
-  ! CHECK:    %[[WS_UB:.*]] = arith.constant 9 : i32
-  ! CHECK:    %[[WS_STEP:.*]] = arith.constant 1 : i32
-  ! CHECK:    omp.wsloop for (%[[I:.*]]) : i32 = (%[[WS_LB]]) to (%[[WS_UB]]) inclusive step (%[[WS_STEP]])
+  ! CHECK:      %[[PRIVATE_COND_REF:.*]] = fir.alloca !fir.logical<4> {bindc_name = "cond", pinned, uniq_name = "_QFparallel_do_with_privatisation_clausesEcond"}
+  ! CHECK:      %[[PRIVATE_COND_DECL:.*]]:2 = hlfir.declare %[[PRIVATE_COND_REF]] {uniq_name = "_QFparallel_do_with_privatisation_clausesEcond"} : (!fir.ref<!fir.logical<4>>) -> (!fir.ref<!fir.logical<4>>, !fir.ref<!fir.logical<4>>)
+  ! CHECK:      %[[PRIVATE_NT_REF:.*]] = fir.alloca i32 {bindc_name = "nt", pinned, uniq_name = "_QFparallel_do_with_privatisation_clausesEnt"}
+  ! CHECK:      %[[PRIVATE_NT_DECL:.*]]:2 = hlfir.declare %[[PRIVATE_NT_REF]] {uniq_name = "_QFparallel_do_with_privatisation_clausesEnt"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
+  ! CHECK:      %[[NT_VAL:.*]] = fir.load %[[NT_DECL]]#0 : !fir.ref<i32>
+  ! CHECK:      hlfir.assign %[[NT_VAL]] to %[[PRIVATE_NT_DECL]]#0 temporary_lhs : i32, !fir.ref<i32>
+  ! CHECK:      %[[WS_LB:.*]] = arith.constant 1 : i32
+  ! CHECK:      %[[WS_UB:.*]] = arith.constant 9 : i32
+  ! CHECK:      %[[WS_STEP:.*]] = arith.constant 1 : i32
+  ! CHECK:      omp.wsloop {
+  ! CHECK-NEXT: omp.loop_nest (%[[I:.*]]) : i32 = (%[[WS_LB]]) to (%[[WS_UB]]) inclusive step (%[[WS_STEP]]) {
   !$OMP PARALLEL DO PRIVATE(cond) FIRSTPRIVATE(nt)
   do i=1, 9
-  ! CHECK:    fir.store %[[I]] to %[[IV_ADDR:.*]]#1 : !fir.ref<i32>
-  ! CHECK:    %[[LOAD_IV:.*]] = fir.load %[[IV_ADDR]]#0 : !fir.ref<i32>
+  ! CHECK:      fir.store %[[I]] to %[[IV_ADDR:.*]]#1 : !fir.ref<i32>
+  ! CHECK:      %[[LOAD_IV:.*]] = fir.load %[[IV_ADDR]]#0 : !fir.ref<i32>
   ! CHECK:      fir.call @_FortranAioOutputInteger32({{.*}}, %[[LOAD_IV]]) {{.*}}: (!fir.ref<i8>, i32) -> i1
   ! CHECK:      %[[PRIVATE_COND_VAL:.*]] = fir.load %[[PRIVATE_COND_DECL]]#0 : !fir.ref<!fir.logical<4>>
   ! CHECK:      %[[PRIVATE_COND_VAL_CVT:.*]] = fir.convert %[[PRIVATE_COND_VAL]] : (!fir.logical<4>) -> i1
@@ -112,6 +119,7 @@ subroutine parallel_do_with_privatisation_clauses(cond,nt)
   end do
   ! CHECK:      omp.yield
   ! CHECK:    omp.terminator
+  ! CHECK:    omp.terminator
   !$OMP END PARALLEL DO
 end subroutine
 
@@ -150,10 +158,13 @@ end subroutine parallel_private_do
 ! CHECK:             %[[VAL_7:.*]] = arith.constant 1 : i32
 ! CHECK:             %[[VAL_8:.*]] = arith.constant 9 : i32
 ! CHECK:             %[[VAL_9:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop   for  (%[[I:.*]]) : i32 = (%[[VAL_7]]) to (%[[VAL_8]]) inclusive step (%[[VAL_9]]) {
-! CHECK:               fir.store %[[I]] to %[[I_PRIV_DECL]]#1 : !fir.ref<i32>
-! CHECK:               fir.call @_QPfoo(%[[I_PRIV_DECL]]#1, %[[COND_DECL]]#1, %[[NT_PRIV_DECL]]#1) {{.*}}: (!fir.ref<i32>, !fir.ref<!fir.logical<4>>, !fir.ref<i32>) -> ()
-! CHECK:               omp.yield
+! CHECK:             omp.wsloop {
+! CHECK-NEXT:          omp.loop_nest (%[[I:.*]]) : i32 = (%[[VAL_7]]) to (%[[VAL_8]]) inclusive step (%[[VAL_9]]) {
+! CHECK:                 fir.store %[[I]] to %[[I_PRIV_DECL]]#1 : !fir.ref<i32>
+! CHECK:                 fir.call @_QPfoo(%[[I_PRIV_DECL]]#1, %[[COND_DECL]]#1, %[[NT_PRIV_DECL]]#1) {{.*}}: (!fir.ref<i32>, !fir.ref<!fir.logical<4>>, !fir.ref<i32>) -> ()
+! CHECK:                 omp.yield
+! CHECK:               }
+! CHECK:               omp.terminator
 ! CHECK:             }
 ! CHECK:             omp.terminator
 ! CHECK:           }
@@ -196,10 +207,13 @@ end subroutine omp_parallel_multiple_firstprivate_do
 ! CHECK:             %[[VAL_8:.*]] = arith.constant 1 : i32
 ! CHECK:             %[[VAL_9:.*]] = arith.constant 10 : i32
 ! CHECK:             %[[VAL_10:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop   for  (%[[I:.*]]) : i32 = (%[[VAL_8]]) to (%[[VAL_9]]) inclusive step (%[[VAL_10]]) {
-! CHECK:               fir.store %[[I]] to %[[I_PRIV_DECL]]#1 : !fir.ref<i32>
-! CHECK:               fir.call @_QPbar(%[[I_PRIV_DECL]]#1, %[[A_PRIV_DECL]]#1) {{.*}}: (!fir.ref<i32>, !fir.ref<i32>) -> ()
-! CHECK:               omp.yield
+! CHECK:             omp.wsloop {
+! CHECK-NEXT:          omp.loop_nest (%[[I:.*]]) : i32 = (%[[VAL_8]]) to (%[[VAL_9]]) inclusive step (%[[VAL_10]]) {
+! CHECK:                 fir.store %[[I]] to %[[I_PRIV_DECL]]#1 : !fir.ref<i32>
+! CHECK:                 fir.call @_QPbar(%[[I_PRIV_DECL]]#1, %[[A_PRIV_DECL]]#1) {{.*}}: (!fir.ref<i32>, !fir.ref<i32>) -> ()
+! CHECK:                 omp.yield
+! CHECK:               }
+! CHECK:               omp.terminator
 ! CHECK:             }
 ! CHECK:             omp.terminator
 ! CHECK:           }
@@ -241,10 +255,13 @@ end subroutine parallel_do_private
 ! CHECK:             %[[VAL_7:.*]] = arith.constant 1 : i32
 ! CHECK:             %[[VAL_8:.*]] = arith.constant 9 : i32
 ! CHECK:             %[[VAL_9:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop   for  (%[[I:.*]]) : i32 = (%[[VAL_7]]) to (%[[VAL_8]]) inclusive step (%[[VAL_9]]) {
-! CHECK:               fir.store %[[I]] to %[[I_PRIV_DECL]]#1 : !fir.ref<i32>
-! CHECK:               fir.call @_QPfoo(%[[I_PRIV_DECL]]#1, %[[COND_PRIV_DECL]]#1, %[[NT_PRIV_DECL]]#1) {{.*}}: (!fir.ref<i32>, !fir.ref<!fir.logical<4>>, !fir.ref<i32>) -> ()
-! CHECK:               omp.yield
+! CHECK:             omp.wsloop {
+! CHECK-NEXT:          omp.loop_nest (%[[I:.*]]) : i32 = (%[[VAL_7]]) to (%[[VAL_8]]) inclusive step (%[[VAL_9]]) {
+! CHECK:                 fir.store %[[I]] to %[[I_PRIV_DECL]]#1 : !fir.ref<i32>
+! CHECK:                 fir.call @_QPfoo(%[[I_PRIV_DECL]]#1, %[[COND_PRIV_DECL]]#1, %[[NT_PRIV_DECL]]#1) {{.*}}: (!fir.ref<i32>, !fir.ref<!fir.logical<4>>, !fir.ref<i32>) -> ()
+! CHECK:                 omp.yield
+! CHECK:               }
+! CHECK:               omp.terminator
 ! CHECK:             }
 ! CHECK:             omp.terminator
 ! CHECK:           }
@@ -271,9 +288,9 @@ end subroutine omp_parallel_do_multiple_firstprivate
 ! CHECK-LABEL:   func.func @_QPomp_parallel_do_multiple_firstprivate(
 ! CHECK-SAME:                                                        %[[A_ADDR:.*]]: !fir.ref<i32> {fir.bindc_name = "a"},
 ! CHECK-SAME:                                                        %[[B_ADDR:.*]]: !fir.ref<i32> {fir.bindc_name = "b"}) {
-! CHECK:            %[[A_DECL:.*]]:2 = hlfir.declare %[[A_ADDR]] {uniq_name = "_QFomp_parallel_do_multiple_firstprivateEa"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
-! CHECK:            %[[B_DECL:.*]]:2 = hlfir.declare %[[B_ADDR]] {uniq_name = "_QFomp_parallel_do_multiple_firstprivateEb"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>
-! CHECK:           omp.parallel   {
+! CHECK:           %[[A_DECL:.*]]:2 = hlfir.declare %[[A_ADDR]] {uniq_name = "_QFomp_parallel_do_multiple_firstprivateEa"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
+! CHECK:           %[[B_DECL:.*]]:2 = hlfir.declare %[[B_ADDR]] {uniq_name = "_QFomp_parallel_do_multiple_firstprivateEb"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>
+! CHECK:           omp.parallel {
 ! CHECK:             %[[I_PRIV_ADDR:.*]] = fir.alloca i32 {adapt.valuebyref, pinned}
 ! CHECK:             %[[I_PRIV_DECL:.*]]:2 = hlfir.declare %[[I_PRIV_ADDR]] {uniq_name = "_QFomp_parallel_do_multiple_firstprivateEi"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
 ! CHECK:             %[[A_PRIV_ADDR:.*]] = fir.alloca i32 {bindc_name = "a", pinned, uniq_name = "_QFomp_parallel_do_multiple_firstprivateEa"}
@@ -287,12 +304,15 @@ end subroutine omp_parallel_do_multiple_firstprivate
 ! CHECK:             %[[VAL_8:.*]] = arith.constant 1 : i32
 ! CHECK:             %[[VAL_9:.*]] = arith.constant 10 : i32
 ! CHECK:             %[[VAL_10:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop   for  (%[[I:.*]]) : i32 = (%[[VAL_8]]) to (%[[VAL_9]]) inclusive step (%[[VAL_10]]) {
-! CHECK:               fir.store %[[I]] to %[[I_PRIV_DECL]]#1 : !fir.ref<i32>
-! CHECK:               fir.call @_QPbar(%[[I_PRIV_DECL]]#1, %[[A_PRIV_DECL]]#1) {{.*}}: (!fir.ref<i32>, !fir.ref<i32>) -> ()
-! CHECK:               omp.yield
+! CHECK:             omp.wsloop {
+! CHECK-NEXT:         omp.loop_nest (%[[I:.*]]) : i32 = (%[[VAL_8]]) to (%[[VAL_9]]) inclusive step (%[[VAL_10]]) {
+! CHECK:                 fir.store %[[I]] to %[[I_PRIV_DECL]]#1 : !fir.ref<i32>
+! CHECK:                 fir.call @_QPbar(%[[I_PRIV_DECL]]#1, %[[A_PRIV_DECL]]#1) {{.*}}: (!fir.ref<i32>, !fir.ref<i32>) -> ()
+! CHECK:                 omp.yield
+! CHECK:               }
+! CHECK:               omp.terminator
 ! CHECK:             }
 ! CHECK:             omp.terminator
 ! CHECK:           }
-! CHECK:           return
+! CHECK:          return
 ! CHECK:         }
diff --git a/flang/test/Lower/OpenMP/simd.f90 b/flang/test/Lower/OpenMP/simd.f90
index 190aa6152121..8ec1a3cefb4a 100644
--- a/flang/test/Lower/OpenMP/simd.f90
+++ b/flang/test/Lower/OpenMP/simd.f90
@@ -1,7 +1,8 @@
 ! Tests for 2.9.3.1 Simd
 
-!RUN: %flang_fc1 -flang-experimental-hlfir -emit-hlfir -fopenmp %s -o - | FileCheck %s
-!RUN: bbc -hlfir -emit-hlfir -fopenmp %s -o - | FileCheck %s
+! The "if" clause was added to the "simd" directive in OpenMP 5.0.
+! RUN: %flang_fc1 -flang-experimental-hlfir -emit-hlfir -fopenmp -fopenmp-version=50 %s -o - | FileCheck %s
+! RUN: bbc -hlfir -emit-hlfir -fopenmp -fopenmp-version=50 %s -o - | FileCheck %s
 
 !CHECK-LABEL: func @_QPsimd()
 subroutine simd
diff --git a/flang/test/Lower/OpenMP/stop-stmt-in-region.f90 b/flang/test/Lower/OpenMP/stop-stmt-in-region.f90
index fdbabc21b2c9..4f3819c5e4eb 100644
--- a/flang/test/Lower/OpenMP/stop-stmt-in-region.f90
+++ b/flang/test/Lower/OpenMP/stop-stmt-in-region.f90
@@ -82,24 +82,27 @@ end
 ! CHECK:         %[[VAL_3:.*]] = arith.constant 1 : i32
 ! CHECK:         %[[VAL_4:.*]] = arith.constant 10 : i32
 ! CHECK:         %[[VAL_5:.*]] = arith.constant 1 : i32
-! CHECK:         omp.wsloop   for  (%[[VAL_6:.*]]) : i32 = (%[[VAL_3]]) to (%[[VAL_4]]) inclusive step (%[[VAL_5]]) {
-! CHECK:           fir.store %[[VAL_6]] to %[[VAL_0_DECL]]#1 : !fir.ref<i32>
-! CHECK:           cf.br ^bb1
-! CHECK:         ^bb1:
-! CHECK:           %[[VAL_7:.*]] = arith.constant 3 : i32
-! CHECK:           hlfir.assign %[[VAL_7]] to %[[VAL_2_DECL]]#0 : i32, !fir.ref<i32>
-! CHECK:           %[[VAL_8:.*]] = fir.load %[[VAL_2_DECL]]#0 : !fir.ref<i32>
-! CHECK:           %[[VAL_9:.*]] = arith.constant 1 : i32
-! CHECK:           %[[VAL_10:.*]] = arith.cmpi sgt, %[[VAL_8]], %[[VAL_9]] : i32
-! CHECK:           cf.cond_br %[[VAL_10]], ^bb2, ^bb3
-! CHECK:         ^bb2:
-! CHECK:           %[[VAL_11:.*]] = fir.load %[[VAL_2_DECL]]#0 : !fir.ref<i32>
-! CHECK:           %[[VAL_12:.*]] = arith.constant false
-! CHECK:           %[[VAL_13:.*]] = arith.constant false
-! CHECK:           %[[VAL_14:.*]] = fir.call @_FortranAStopStatement(%[[VAL_11]], %[[VAL_12]], %[[VAL_13]]) {{.*}} : (i32, i1, i1) -> none
-! CHECK:           omp.yield
-! CHECK:         ^bb3:
-! CHECK:           omp.yield
+! CHECK:         omp.wsloop {
+! CHECK-NEXT:      omp.loop_nest (%[[VAL_6:.*]]) : i32 = (%[[VAL_3]]) to (%[[VAL_4]]) inclusive step (%[[VAL_5]]) {
+! CHECK:             fir.store %[[VAL_6]] to %[[VAL_0_DECL]]#1 : !fir.ref<i32>
+! CHECK:             cf.br ^bb1
+! CHECK:           ^bb1:
+! CHECK:             %[[VAL_7:.*]] = arith.constant 3 : i32
+! CHECK:             hlfir.assign %[[VAL_7]] to %[[VAL_2_DECL]]#0 : i32, !fir.ref<i32>
+! CHECK:             %[[VAL_8:.*]] = fir.load %[[VAL_2_DECL]]#0 : !fir.ref<i32>
+! CHECK:             %[[VAL_9:.*]] = arith.constant 1 : i32
+! CHECK:             %[[VAL_10:.*]] = arith.cmpi sgt, %[[VAL_8]], %[[VAL_9]] : i32
+! CHECK:             cf.cond_br %[[VAL_10]], ^bb2, ^bb3
+! CHECK:           ^bb2:
+! CHECK:             %[[VAL_11:.*]] = fir.load %[[VAL_2_DECL]]#0 : !fir.ref<i32>
+! CHECK:             %[[VAL_12:.*]] = arith.constant false
+! CHECK:             %[[VAL_13:.*]] = arith.constant false
+! CHECK:             %[[VAL_14:.*]] = fir.call @_FortranAStopStatement(%[[VAL_11]], %[[VAL_12]], %[[VAL_13]]) {{.*}} : (i32, i1, i1) -> none
+! CHECK:             omp.yield
+! CHECK:           ^bb3:
+! CHECK:             omp.yield
+! CHECK:           }
+! CHECK:           omp.terminator
 ! CHECK:         }
 ! CHECK:         cf.br ^bb1
 ! CHECK:       ^bb1:
diff --git a/flang/test/Lower/OpenMP/target.f90 b/flang/test/Lower/OpenMP/target.f90
index 51b66327dfb2..44f77b5c3360 100644
--- a/flang/test/Lower/OpenMP/target.f90
+++ b/flang/test/Lower/OpenMP/target.f90
@@ -1,4 +1,5 @@
-!RUN: %flang_fc1 -emit-hlfir -fopenmp %s -o - | FileCheck %s
+! The "thread_limit" clause was added to the "target" construct in OpenMP 5.1.
+! RUN: %flang_fc1 -emit-hlfir -fopenmp -fopenmp-version=51 %s -o - | FileCheck %s
 
 !===============================================================================
 ! Target_Enter Simple
@@ -594,7 +595,8 @@ subroutine omp_target_parallel_do
       !$omp target parallel do map(tofrom: a)
          !CHECK: %[[I_PVT_ALLOCA:.*]] = fir.alloca i32 {adapt.valuebyref, pinned}
          !CHECK: %[[I_PVT_DECL:.*]]:2 = hlfir.declare %[[I_PVT_ALLOCA]] {uniq_name = "_QFomp_target_parallel_doEi"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
-         !CHECK: omp.wsloop for  (%[[I_VAL:.*]]) : i32
+         !CHECK: omp.wsloop {
+         !CHECK-NEXT: omp.loop_nest (%[[I_VAL:.*]]) : i32
          do i = 1, 1024
            !CHECK:   fir.store %[[I_VAL]] to %[[I_PVT_DECL]]#1 : !fir.ref<i32>
            !CHECK:   %[[C10:.*]] = arith.constant 10 : i32
@@ -606,6 +608,8 @@ subroutine omp_target_parallel_do
          end do
          !CHECK: omp.yield
          !CHECK: }
+         !CHECK: omp.terminator
+         !CHECK: }
       !CHECK: omp.terminator
       !CHECK: }
    !CHECK: omp.terminator
diff --git a/flang/test/Lower/OpenMP/threadprivate-real-logical-complex-derivedtype.f90 b/flang/test/Lower/OpenMP/threadprivate-real-logical-complex-derivedtype.f90
index 55f806962a60..0a249ff101a0 100644
--- a/flang/test/Lower/OpenMP/threadprivate-real-logical-complex-derivedtype.f90
+++ b/flang/test/Lower/OpenMP/threadprivate-real-logical-complex-derivedtype.f90
@@ -21,6 +21,7 @@ module test
 !CHECK-DAG: fir.global @_QMtestEz : !fir.logical<4> {
 
 contains
+!CHECK-LABEL: func.func @_QMtestPsub
   subroutine sub()
 !CHECK-DAG:  %[[T:.*]] = fir.address_of(@_QMtestEt) : !fir.ref<!fir.type<_QMtestTmy_type{t_i:i32,t_arr:!fir.array<5xf32>}>>
 !CHECK-DAG:  %[[T_DECL:.*]]:2 = hlfir.declare %[[T]] {uniq_name = "_QMtestEt"} : (!fir.ref<!fir.type<_QMtestTmy_type{t_i:i32,t_arr:!fir.array<5xf32>}>>) -> (!fir.ref<!fir.type<_QMtestTmy_type{t_i:i32,t_arr:!fir.array<5xf32>}>>, !fir.ref<!fir.type<_QMtestTmy_type{t_i:i32,t_arr:!fir.array<5xf32>}>>)
diff --git a/flang/test/Lower/OpenMP/unstructured.f90 b/flang/test/Lower/OpenMP/unstructured.f90
index e5bf980ce90f..6a1331799d54 100644
--- a/flang/test/Lower/OpenMP/unstructured.f90
+++ b/flang/test/Lower/OpenMP/unstructured.f90
@@ -70,27 +70,33 @@ end
 ! CHECK:   ^bb1:  // 2 preds: ^bb0, ^bb3
 ! CHECK:     cond_br %{{[0-9]*}}, ^bb2, ^bb4
 ! CHECK:   ^bb2:  // pred: ^bb1
-! CHECK:     omp.wsloop for (%[[ARG1:.*]]) : {{.*}} {
-! CHECK:       fir.store %[[ARG1]] to %[[OMP_LOOP_K_DECL]]#1 : !fir.ref<i32>
-! CHECK:     @_FortranAioBeginExternalListOutput
-! CHECK:       %[[LOAD_1:.*]] = fir.load %[[OMP_LOOP_K_DECL]]#0 : !fir.ref<i32>
-! CHECK:     @_FortranAioOutputInteger32(%{{.*}}, %[[LOAD_1]])
-! CHECK:       omp.yield
+! CHECK:     omp.wsloop {
+! CHECK:       omp.loop_nest (%[[ARG1:.*]]) : {{.*}} {
+! CHECK:         fir.store %[[ARG1]] to %[[OMP_LOOP_K_DECL]]#1 : !fir.ref<i32>
+! CHECK:         @_FortranAioBeginExternalListOutput
+! CHECK:         %[[LOAD_1:.*]] = fir.load %[[OMP_LOOP_K_DECL]]#0 : !fir.ref<i32>
+! CHECK:         @_FortranAioOutputInteger32(%{{.*}}, %[[LOAD_1]])
+! CHECK:         omp.yield
+! CHECK:       }
+! CHECK:       omp.terminator
 ! CHECK:     }
-! CHECK:     omp.wsloop for (%[[ARG2:.*]]) : {{.*}} {
-! CHECK:       fir.store %[[ARG2]] to %[[OMP_LOOP_J_DECL]]#1 : !fir.ref<i32>
-! CHECK:       br ^bb1
-! CHECK:     ^bb2:  // 2 preds: ^bb1, ^bb5
-! CHECK:       cond_br %{{[0-9]*}}, ^bb3, ^bb6
-! CHECK:     ^bb3:  // pred: ^bb2
-! CHECK:       cond_br %{{[0-9]*}}, ^bb4, ^bb5
-! CHECK:     ^bb4:  // pred: ^bb3
-! CHECK:       @_FortranAioBeginExternalListOutput
-! CHECK:       %[[LOAD_2:.*]] = fir.load %[[K_DECL]]#0 : !fir.ref<i32>
-! CHECK:     @_FortranAioOutputInteger32(%{{.*}}, %[[LOAD_2]])
-! CHECK:       br ^bb2
-! CHECK:     ^bb6:  // 2 preds: ^bb2, ^bb4
-! CHECK:       omp.yield
+! CHECK:     omp.wsloop {
+! CHECK:       omp.loop_nest (%[[ARG2:.*]]) : {{.*}} {
+! CHECK:         fir.store %[[ARG2]] to %[[OMP_LOOP_J_DECL]]#1 : !fir.ref<i32>
+! CHECK:         br ^bb1
+! CHECK:       ^bb2:  // 2 preds: ^bb1, ^bb5
+! CHECK:         cond_br %{{[0-9]*}}, ^bb3, ^bb6
+! CHECK:       ^bb3:  // pred: ^bb2
+! CHECK:         cond_br %{{[0-9]*}}, ^bb4, ^bb5
+! CHECK:       ^bb4:  // pred: ^bb3
+! CHECK:         @_FortranAioBeginExternalListOutput
+! CHECK:         %[[LOAD_2:.*]] = fir.load %[[K_DECL]]#0 : !fir.ref<i32>
+! CHECK:         @_FortranAioOutputInteger32(%{{.*}}, %[[LOAD_2]])
+! CHECK:         br ^bb2
+! CHECK:       ^bb6:  // 2 preds: ^bb2, ^bb4
+! CHECK:         omp.yield
+! CHECK:       }
+! CHECK:       omp.terminator
 ! CHECK:     }
 ! CHECK:     br ^bb1
 ! CHECK:   ^bb4:  // pred: ^bb1
@@ -121,20 +127,23 @@ end
 ! CHECK:       omp.parallel {
 ! CHECK:         %[[ALLOCA:.*]] = fir.alloca i32 {{{.*}}, pinned}
 ! CHECK:         %[[OMP_LOOP_J_DECL:.*]]:2 = hlfir.declare %[[ALLOCA]] {uniq_name = "_QFss4Ej"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
-! CHECK:         omp.wsloop for (%[[ARG:.*]]) : {{.*}} {
-! CHECK:           fir.store %[[ARG]] to %[[OMP_LOOP_J_DECL]]#1 : !fir.ref<i32>
-! CHECK:           %[[COND:.*]] = arith.cmpi eq, %{{.*}}, %{{.*}}
-! CHECK:           %[[COND_XOR:.*]] = arith.xori %[[COND]], %{{.*}}
-! CHECK:          fir.if %[[COND_XOR]] {
-! CHECK:           @_FortranAioBeginExternalListOutput
-! CHECK:           %[[LOAD:.*]] = fir.load %[[OMP_LOOP_J_DECL]]#0 : !fir.ref<i32>
-! CHECK:           @_FortranAioOutputInteger32(%{{.*}}, %[[LOAD]])
-! CHECK:          } else {
-! CHECK:          }
-! CHECK-NEXT:      omp.yield
+! CHECK:         omp.wsloop {
+! CHECK-NEXT:      omp.loop_nest (%[[ARG:.*]]) : {{.*}} {
+! CHECK:             fir.store %[[ARG]] to %[[OMP_LOOP_J_DECL]]#1 : !fir.ref<i32>
+! CHECK:             %[[COND:.*]] = arith.cmpi eq, %{{.*}}, %{{.*}}
+! CHECK:             %[[COND_XOR:.*]] = arith.xori %[[COND]], %{{.*}}
+! CHECK:             fir.if %[[COND_XOR]] {
+! CHECK:              @_FortranAioBeginExternalListOutput
+! CHECK:              %[[LOAD:.*]] = fir.load %[[OMP_LOOP_J_DECL]]#0 : !fir.ref<i32>
+! CHECK:              @_FortranAioOutputInteger32(%{{.*}}, %[[LOAD]])
+! CHECK:             } else {
+! CHECK:             }
+! CHECK-NEXT:        omp.yield
+! CHECK-NEXT:      }
+! CHECK-NEXT:      omp.terminator
+! CHECK-NEXT:    }
+! CHECK:         omp.terminator
 ! CHECK-NEXT:  }
-! CHECK:       omp.terminator
-! CHECK-NEXT:}
 subroutine ss4(n) ! CYCLE in OpenMP wsloop constructs
   !$omp parallel
     do i = 1, 3
@@ -150,20 +159,23 @@ end
 
 ! CHECK-LABEL: func @_QPss5() {
 ! CHECK:  omp.parallel  {
-! CHECK:    omp.wsloop {{.*}} {
-! CHECK:      br ^[[BB1:.*]]
-! CHECK:    ^[[BB1]]:
-! CHECK:      br ^[[BB2:.*]]
-! CHECK:    ^[[BB2]]:
-! CHECK:      cond_br %{{.*}}, ^[[BB3:.*]], ^[[BB6:.*]]
-! CHECK:    ^[[BB3]]:
-! CHECK:      cond_br %{{.*}}, ^[[BB4:.*]], ^[[BB3:.*]]
-! CHECK:    ^[[BB4]]:
-! CHECK:      br ^[[BB6]]
-! CHECK:    ^[[BB3]]:
-! CHECK:      br ^[[BB2]]
-! CHECK:    ^[[BB6]]:
-! CHECK:      omp.yield
+! CHECK:    omp.wsloop {
+! CHECK:      omp.loop_nest {{.*}} {
+! CHECK:        br ^[[BB1:.*]]
+! CHECK:      ^[[BB1]]:
+! CHECK:        br ^[[BB2:.*]]
+! CHECK:      ^[[BB2]]:
+! CHECK:        cond_br %{{.*}}, ^[[BB3:.*]], ^[[BB6:.*]]
+! CHECK:      ^[[BB3]]:
+! CHECK:        cond_br %{{.*}}, ^[[BB4:.*]], ^[[BB3:.*]]
+! CHECK:      ^[[BB4]]:
+! CHECK:        br ^[[BB6]]
+! CHECK:      ^[[BB3]]:
+! CHECK:        br ^[[BB2]]
+! CHECK:      ^[[BB6]]:
+! CHECK:        omp.yield
+! CHECK:      }
+! CHECK:      omp.terminator
 ! CHECK:    }
 ! CHECK:    omp.terminator
 ! CHECK:  }
@@ -190,20 +202,23 @@ end
 ! CHECK:  ^[[BB1_OUTER]]:
 ! CHECK:    cond_br %{{.*}}, ^[[BB2_OUTER:.*]], ^[[BB3_OUTER:.*]]
 ! CHECK:  ^[[BB2_OUTER]]:
-! CHECK:    omp.wsloop {{.*}} {
-! CHECK:      br ^[[BB1:.*]]
-! CHECK:    ^[[BB1]]:
-! CHECK:      br ^[[BB2:.*]]
-! CHECK:    ^[[BB2]]:
-! CHECK:      cond_br %{{.*}}, ^[[BB3:.*]], ^[[BB6:.*]]
-! CHECK:    ^[[BB3]]:
-! CHECK:      cond_br %{{.*}}, ^[[BB4:.*]], ^[[BB5:.*]]
-! CHECK:    ^[[BB4]]:
-! CHECK:      br ^[[BB6]]
-! CHECK:    ^[[BB5]]
-! CHECK:      br ^[[BB2]]
-! CHECK:    ^[[BB6]]:
-! CHECK:      omp.yield
+! CHECK:    omp.wsloop {
+! CHECK:      omp.loop_nest {{.*}} {
+! CHECK:        br ^[[BB1:.*]]
+! CHECK:      ^[[BB1]]:
+! CHECK:        br ^[[BB2:.*]]
+! CHECK:      ^[[BB2]]:
+! CHECK:        cond_br %{{.*}}, ^[[BB3:.*]], ^[[BB6:.*]]
+! CHECK:      ^[[BB3]]:
+! CHECK:        cond_br %{{.*}}, ^[[BB4:.*]], ^[[BB5:.*]]
+! CHECK:      ^[[BB4]]:
+! CHECK:        br ^[[BB6]]
+! CHECK:      ^[[BB5]]
+! CHECK:        br ^[[BB2]]
+! CHECK:      ^[[BB6]]:
+! CHECK:        omp.yield
+! CHECK:      }
+! CHECK:      omp.terminator
 ! CHECK:    }
 ! CHECK:    br ^[[BB1_OUTER]]
 ! CHECK:  ^[[BB3_OUTER]]:
@@ -234,20 +249,23 @@ end
 ! CHECK:   cond_br %{{.*}}, ^[[BB2_OUTER:.*]], ^[[BB3_OUTER:.*]]
 ! CHECK-NEXT: ^[[BB2_OUTER:.*]]:
 ! CHECK:   omp.parallel  {
-! CHECK:     omp.wsloop {{.*}} {
-! CHECK:       br ^[[BB1:.*]]
-! CHECK-NEXT:     ^[[BB1]]:
-! CHECK:       br ^[[BB2:.*]]
-! CHECK-NEXT:     ^[[BB2]]:
-! CHECK:       cond_br %{{.*}}, ^[[BB3:.*]], ^[[BB6:.*]]
-! CHECK-NEXT:     ^[[BB3]]:
-! CHECK:       cond_br %{{.*}}, ^[[BB4:.*]], ^[[BB5:.*]]
-! CHECK-NEXT:     ^[[BB4]]:
-! CHECK:       br ^[[BB6]]
-! CHECK-NEXT:     ^[[BB5]]:
-! CHECK:       br ^[[BB2]]
-! CHECK-NEXT:     ^[[BB6]]:
-! CHECK:       omp.yield
+! CHECK:     omp.wsloop {
+! CHECK:       omp.loop_nest {{.*}} {
+! CHECK:         br ^[[BB1:.*]]
+! CHECK-NEXT:       ^[[BB1]]:
+! CHECK:         br ^[[BB2:.*]]
+! CHECK-NEXT:       ^[[BB2]]:
+! CHECK:         cond_br %{{.*}}, ^[[BB3:.*]], ^[[BB6:.*]]
+! CHECK-NEXT:       ^[[BB3]]:
+! CHECK:         cond_br %{{.*}}, ^[[BB4:.*]], ^[[BB5:.*]]
+! CHECK-NEXT:       ^[[BB4]]:
+! CHECK:         br ^[[BB6]]
+! CHECK-NEXT:       ^[[BB5]]:
+! CHECK:         br ^[[BB2]]
+! CHECK-NEXT:       ^[[BB6]]:
+! CHECK:         omp.yield
+! CHECK:       }
+! CHECK:       omp.terminator
 ! CHECK:     }
 ! CHECK:     omp.terminator
 ! CHECK:   }
@@ -272,20 +290,23 @@ end
 
 ! CHECK-LABEL: func @_QPss8() {
 ! CHECK:  omp.parallel  {
-! CHECK:    omp.wsloop {{.*}} {
-! CHECK:      br ^[[BB1:.*]]
-! CHECK-NEXT:    ^[[BB1]]:
-! CHECK:      br ^[[BB2:.*]]
-! CHECK:    ^[[BB2]]:
-! CHECK:      cond_br %{{.*}}, ^[[BB3:.*]], ^[[BB6:.*]]
-! CHECK:    ^[[BB3]]:
-! CHECK:      cond_br %{{.*}}, ^[[BB4:.*]], ^[[BB5:.*]]
-! CHECK:    ^[[BB4]]:
-! CHECK-NEXT:    br ^[[BB6]]
-! CHECK:    ^[[BB5]]:
-! CHECK:      br ^[[BB2]]
-! CHECK-NEXT:    ^[[BB6]]:
-! CHECK:      omp.yield
+! CHECK:    omp.wsloop {
+! CHECK:      omp.loop_nest {{.*}} {
+! CHECK:        br ^[[BB1:.*]]
+! CHECK-NEXT:      ^[[BB1]]:
+! CHECK:        br ^[[BB2:.*]]
+! CHECK:      ^[[BB2]]:
+! CHECK:        cond_br %{{.*}}, ^[[BB3:.*]], ^[[BB6:.*]]
+! CHECK:      ^[[BB3]]:
+! CHECK:        cond_br %{{.*}}, ^[[BB4:.*]], ^[[BB5:.*]]
+! CHECK:      ^[[BB4]]:
+! CHECK-NEXT:      br ^[[BB6]]
+! CHECK:      ^[[BB5]]:
+! CHECK:        br ^[[BB2]]
+! CHECK-NEXT:      ^[[BB6]]:
+! CHECK:        omp.yield
+! CHECK:      }
+! CHECK:      omp.terminator
 ! CHECK:    }
 ! CHECK:    omp.terminator
 ! CHECK:  }
diff --git a/flang/test/Lower/OpenMP/use-device-ptr-to-use-device-addr.f90 b/flang/test/Lower/OpenMP/use-device-ptr-to-use-device-addr.f90
index d849dd206b94..90eede4f8410 100644
--- a/flang/test/Lower/OpenMP/use-device-ptr-to-use-device-addr.f90
+++ b/flang/test/Lower/OpenMP/use-device-ptr-to-use-device-addr.f90
@@ -1,5 +1,6 @@
-!RUN: %flang_fc1 -emit-hlfir -fopenmp %s -o - | FileCheck %s
-!RUN: bbc -emit-hlfir -fopenmp %s -o - | FileCheck %s
+! The "use_device_addr" was added to the "target data" directive in OpenMP 5.0.
+! RUN: %flang_fc1 -emit-hlfir -fopenmp -fopenmp-version=50 %s -o - | FileCheck %s
+! RUN: bbc -emit-hlfir -fopenmp -fopenmp-version=50 %s -o - | FileCheck %s
 
 ! This tests primary goal is to check the promotion of 
 ! non-CPTR arguments from use_device_ptr to 
diff --git a/flang/test/Lower/OpenMP/wsloop-chunks.f90 b/flang/test/Lower/OpenMP/wsloop-chunks.f90
index 5016c8985bda..fa6ec219a490 100644
--- a/flang/test/Lower/OpenMP/wsloop-chunks.f90
+++ b/flang/test/Lower/OpenMP/wsloop-chunks.f90
@@ -20,11 +20,14 @@ do i=1, 9
 ! CHECK:         %[[VAL_3:.*]] = arith.constant 9 : i32
 ! CHECK:         %[[VAL_4:.*]] = arith.constant 1 : i32
 ! CHECK:         %[[VAL_5:.*]] = arith.constant 4 : i32
-! CHECK:         omp.wsloop   schedule(static = %[[VAL_5]] : i32) nowait for  (%[[ARG0:.*]]) : i32 = (%[[VAL_2]]) to (%[[VAL_3]]) inclusive step (%[[VAL_4]]) {
-! CHECK:           fir.store %[[ARG0]] to %[[STORE_IV:.*]]#1 : !fir.ref<i32>
-! CHECK:           %[[LOAD_IV:.*]] = fir.load %[[STORE_IV]]#0 : !fir.ref<i32>
-! CHECK:           {{.*}} = fir.call @_FortranAioOutputInteger32({{.*}}, %[[LOAD_IV]]) {{.*}}: (!fir.ref<i8>, i32) -> i1
-! CHECK:           omp.yield
+! CHECK:         omp.wsloop schedule(static = %[[VAL_5]] : i32) nowait {
+! CHECK-NEXT:      omp.loop_nest (%[[ARG0:.*]]) : i32 = (%[[VAL_2]]) to (%[[VAL_3]]) inclusive step (%[[VAL_4]]) {
+! CHECK:             fir.store %[[ARG0]] to %[[STORE_IV:.*]]#1 : !fir.ref<i32>
+! CHECK:             %[[LOAD_IV:.*]] = fir.load %[[STORE_IV]]#0 : !fir.ref<i32>
+! CHECK:             {{.*}} = fir.call @_FortranAioOutputInteger32({{.*}}, %[[LOAD_IV]]) {{.*}}: (!fir.ref<i8>, i32) -> i1
+! CHECK:             omp.yield
+! CHECK:           }
+! CHECK:           omp.terminator
 ! CHECK:         }
 
 end do
@@ -38,13 +41,16 @@ do i=1, 9
 ! CHECK:         %[[VAL_15:.*]] = arith.constant 9 : i32
 ! CHECK:         %[[VAL_16:.*]] = arith.constant 1 : i32
 ! CHECK:         %[[VAL_17:.*]] = arith.constant 4 : i32
-! CHECK:         omp.wsloop   schedule(static = %[[VAL_17]] : i32) nowait for  (%[[ARG1:.*]]) : i32 = (%[[VAL_14]]) to (%[[VAL_15]]) inclusive step (%[[VAL_16]]) {
-! CHECK:           fir.store %[[ARG1]] to %[[STORE_IV1:.*]]#1 : !fir.ref<i32>
-! CHECK:           %[[VAL_24:.*]] = arith.constant 2 : i32
-! CHECK:           %[[LOAD_IV1:.*]] = fir.load %[[STORE_IV1]]#0 : !fir.ref<i32>
-! CHECK:           %[[VAL_25:.*]] = arith.muli %[[VAL_24]], %[[LOAD_IV1]] : i32
-! CHECK:           {{.*}} = fir.call @_FortranAioOutputInteger32({{.*}}, %[[VAL_25]]) {{.*}}: (!fir.ref<i8>, i32) -> i1
-! CHECK:           omp.yield
+! CHECK:         omp.wsloop schedule(static = %[[VAL_17]] : i32) nowait {
+! CHECK-NEXT:      omp.loop_nest (%[[ARG1:.*]]) : i32 = (%[[VAL_14]]) to (%[[VAL_15]]) inclusive step (%[[VAL_16]]) {
+! CHECK:             fir.store %[[ARG1]] to %[[STORE_IV1:.*]]#1 : !fir.ref<i32>
+! CHECK:             %[[VAL_24:.*]] = arith.constant 2 : i32
+! CHECK:             %[[LOAD_IV1:.*]] = fir.load %[[STORE_IV1]]#0 : !fir.ref<i32>
+! CHECK:             %[[VAL_25:.*]] = arith.muli %[[VAL_24]], %[[LOAD_IV1]] : i32
+! CHECK:             {{.*}} = fir.call @_FortranAioOutputInteger32({{.*}}, %[[VAL_25]]) {{.*}}: (!fir.ref<i8>, i32) -> i1
+! CHECK:             omp.yield
+! CHECK:           }
+! CHECK:           omp.terminator
 ! CHECK:         }
   
 end do
@@ -62,13 +68,16 @@ end do
 ! CHECK:         %[[VAL_30:.*]] = arith.constant 9 : i32
 ! CHECK:         %[[VAL_31:.*]] = arith.constant 1 : i32
 ! CHECK:         %[[VAL_32:.*]] = fir.load %[[VAL_0]]#0 : !fir.ref<i32>
-! CHECK:         omp.wsloop   schedule(static = %[[VAL_32]] : i32) nowait for  (%[[ARG2:.*]]) : i32 = (%[[VAL_29]]) to (%[[VAL_30]]) inclusive step (%[[VAL_31]]) {
-! CHECK:           fir.store %[[ARG2]] to %[[STORE_IV2:.*]]#1 : !fir.ref<i32>
-! CHECK:           %[[VAL_39:.*]] = arith.constant 3 : i32
-! CHECK:           %[[LOAD_IV2:.*]] = fir.load %[[STORE_IV2]]#0 : !fir.ref<i32>
-! CHECK:           %[[VAL_40:.*]] = arith.muli %[[VAL_39]], %[[LOAD_IV2]] : i32
-! CHECK:           {{.*}} = fir.call @_FortranAioOutputInteger32({{.*}}, %[[VAL_40]]) {{.*}}: (!fir.ref<i8>, i32) -> i1
-! CHECK:           omp.yield
+! CHECK:         omp.wsloop schedule(static = %[[VAL_32]] : i32) nowait {
+! CHECK-NEXT:      omp.loop_nest (%[[ARG2:.*]]) : i32 = (%[[VAL_29]]) to (%[[VAL_30]]) inclusive step (%[[VAL_31]]) {
+! CHECK:             fir.store %[[ARG2]] to %[[STORE_IV2:.*]]#1 : !fir.ref<i32>
+! CHECK:             %[[VAL_39:.*]] = arith.constant 3 : i32
+! CHECK:             %[[LOAD_IV2:.*]] = fir.load %[[STORE_IV2]]#0 : !fir.ref<i32>
+! CHECK:             %[[VAL_40:.*]] = arith.muli %[[VAL_39]], %[[LOAD_IV2]] : i32
+! CHECK:             {{.*}} = fir.call @_FortranAioOutputInteger32({{.*}}, %[[VAL_40]]) {{.*}}: (!fir.ref<i8>, i32) -> i1
+! CHECK:             omp.yield
+! CHECK:           }
+! CHECK:           omp.terminator
 ! CHECK:         }
 ! CHECK:         return
 ! CHECK:       }
diff --git a/flang/test/Lower/OpenMP/wsloop-collapse.f90 b/flang/test/Lower/OpenMP/wsloop-collapse.f90
index c93fcf4ef968..d9541e176f6a 100644
--- a/flang/test/Lower/OpenMP/wsloop-collapse.f90
+++ b/flang/test/Lower/OpenMP/wsloop-collapse.f90
@@ -49,23 +49,26 @@ program wsloop_collapse
 !CHECK:           %[[VAL_30:.*]] = arith.constant 1 : i32
 !CHECK:           %[[VAL_31:.*]] = fir.load %[[VAL_11]]#0 : !fir.ref<i32>
 !CHECK:           %[[VAL_32:.*]] = arith.constant 1 : i32
-!CHECK:           omp.wsloop   for  (%[[VAL_33:.*]], %[[VAL_34:.*]], %[[VAL_35:.*]]) : i32 = (%[[VAL_24]], %[[VAL_27]], %[[VAL_30]]) to (%[[VAL_25]], %[[VAL_28]], %[[VAL_31]]) inclusive step (%[[VAL_26]], %[[VAL_29]], %[[VAL_32]]) {
+!CHECK:           omp.wsloop {
+!CHECK-NEXT:        omp.loop_nest (%[[VAL_33:.*]], %[[VAL_34:.*]], %[[VAL_35:.*]]) : i32 = (%[[VAL_24]], %[[VAL_27]], %[[VAL_30]]) to (%[[VAL_25]], %[[VAL_28]], %[[VAL_31]]) inclusive step (%[[VAL_26]], %[[VAL_29]], %[[VAL_32]]) {
   !$omp do collapse(3)
   do i = 1, a
      do j= 1, b
         do k = 1, c
-!CHECK:             fir.store %[[VAL_33]] to %[[VAL_5]]#1 : !fir.ref<i32>
-!CHECK:             fir.store %[[VAL_34]] to %[[VAL_3]]#1 : !fir.ref<i32>
-!CHECK:             fir.store %[[VAL_35]] to %[[VAL_1]]#1 : !fir.ref<i32>
-!CHECK:             %[[VAL_36:.*]] = fir.load %[[VAL_19]]#0 : !fir.ref<i32>
-!CHECK:             %[[VAL_37:.*]] = fir.load %[[VAL_5]]#0 : !fir.ref<i32>
-!CHECK:             %[[VAL_38:.*]] = arith.addi %[[VAL_36]], %[[VAL_37]] : i32
-!CHECK:             %[[VAL_39:.*]] = fir.load %[[VAL_3]]#0 : !fir.ref<i32>
-!CHECK:             %[[VAL_40:.*]] = arith.addi %[[VAL_38]], %[[VAL_39]] : i32
-!CHECK:             %[[VAL_41:.*]] = fir.load %[[VAL_1]]#0 : !fir.ref<i32>
-!CHECK:             %[[VAL_42:.*]] = arith.addi %[[VAL_40]], %[[VAL_41]] : i32
-!CHECK:             hlfir.assign %[[VAL_42]] to %[[VAL_19]]#0 : i32, !fir.ref<i32>
-!CHECK:             omp.yield
+!CHECK:               fir.store %[[VAL_33]] to %[[VAL_5]]#1 : !fir.ref<i32>
+!CHECK:               fir.store %[[VAL_34]] to %[[VAL_3]]#1 : !fir.ref<i32>
+!CHECK:               fir.store %[[VAL_35]] to %[[VAL_1]]#1 : !fir.ref<i32>
+!CHECK:               %[[VAL_36:.*]] = fir.load %[[VAL_19]]#0 : !fir.ref<i32>
+!CHECK:               %[[VAL_37:.*]] = fir.load %[[VAL_5]]#0 : !fir.ref<i32>
+!CHECK:               %[[VAL_38:.*]] = arith.addi %[[VAL_36]], %[[VAL_37]] : i32
+!CHECK:               %[[VAL_39:.*]] = fir.load %[[VAL_3]]#0 : !fir.ref<i32>
+!CHECK:               %[[VAL_40:.*]] = arith.addi %[[VAL_38]], %[[VAL_39]] : i32
+!CHECK:               %[[VAL_41:.*]] = fir.load %[[VAL_1]]#0 : !fir.ref<i32>
+!CHECK:               %[[VAL_42:.*]] = arith.addi %[[VAL_40]], %[[VAL_41]] : i32
+!CHECK:               hlfir.assign %[[VAL_42]] to %[[VAL_19]]#0 : i32, !fir.ref<i32>
+!CHECK:               omp.yield
+!CHECK-NEXT:        }
+!CHECK-NEXT:        omp.terminator
            x = x + i + j + k
         end do
      end do
diff --git a/flang/test/Lower/OpenMP/wsloop-monotonic.f90 b/flang/test/Lower/OpenMP/wsloop-monotonic.f90
index fba9105b9818..531d995052f6 100644
--- a/flang/test/Lower/OpenMP/wsloop-monotonic.f90
+++ b/flang/test/Lower/OpenMP/wsloop-monotonic.f90
@@ -15,19 +15,21 @@ program wsloop_dynamic
 !CHECK:     %[[WS_LB:.*]] = arith.constant 1 : i32
 !CHECK:     %[[WS_UB:.*]] = arith.constant 9 : i32
 !CHECK:     %[[WS_STEP:.*]] = arith.constant 1 : i32
-!CHECK:     omp.wsloop schedule(dynamic, monotonic) nowait for (%[[I:.*]]) : i32 = (%[[WS_LB]]) to (%[[WS_UB]]) inclusive step (%[[WS_STEP]])
-!CHECK:       fir.store %[[I]] to %[[ALLOCA_IV:.*]]#1 : !fir.ref<i32>
+!CHECK:     omp.wsloop schedule(dynamic, monotonic) nowait {
+!CHECK-NEXT:  omp.loop_nest (%[[I:.*]]) : i32 = (%[[WS_LB]]) to (%[[WS_UB]]) inclusive step (%[[WS_STEP]]) {
+!CHECK:         fir.store %[[I]] to %[[ALLOCA_IV:.*]]#1 : !fir.ref<i32>
 
   do i=1, 9
     print*, i
-!CHECK:    %[[RTBEGIN:.*]] = fir.call @_FortranAioBeginExternalListOutput
-!CHECK:    %[[LOAD:.*]] = fir.load %[[ALLOCA_IV]]#0 : !fir.ref<i32>
-!CHECK:    fir.call @_FortranAioOutputInteger32(%[[RTBEGIN]], %[[LOAD]]) {{.*}}: (!fir.ref<i8>, i32) -> i1
-!CHECK:    fir.call @_FortranAioEndIoStatement(%[[RTBEGIN]]) {{.*}}: (!fir.ref<i8>) -> i32
+!CHECK:         %[[RTBEGIN:.*]] = fir.call @_FortranAioBeginExternalListOutput
+!CHECK:         %[[LOAD:.*]] = fir.load %[[ALLOCA_IV]]#0 : !fir.ref<i32>
+!CHECK:         fir.call @_FortranAioOutputInteger32(%[[RTBEGIN]], %[[LOAD]]) {{.*}}: (!fir.ref<i8>, i32) -> i1
+!CHECK:         fir.call @_FortranAioEndIoStatement(%[[RTBEGIN]]) {{.*}}: (!fir.ref<i8>) -> i32
   end do
-!CHECK:       omp.yield
+!CHECK:         omp.yield
 !CHECK:       omp.terminator
-!CHECK:     }
+!CHECK:     omp.terminator
+!CHECK:   }
 
 !$OMP END DO NOWAIT
 !$OMP END PARALLEL
diff --git a/flang/test/Lower/OpenMP/wsloop-nonmonotonic.f90 b/flang/test/Lower/OpenMP/wsloop-nonmonotonic.f90
index 1bd7a2edc0f5..420bc0bffaec 100644
--- a/flang/test/Lower/OpenMP/wsloop-nonmonotonic.f90
+++ b/flang/test/Lower/OpenMP/wsloop-nonmonotonic.f90
@@ -17,20 +17,23 @@ program wsloop_dynamic
 !CHECK:     %[[WS_LB:.*]] = arith.constant 1 : i32
 !CHECK:     %[[WS_UB:.*]] = arith.constant 9 : i32
 !CHECK:     %[[WS_STEP:.*]] = arith.constant 1 : i32
-!CHECK:     omp.wsloop schedule(dynamic, nonmonotonic) nowait for (%[[I:.*]]) : i32 = (%[[WS_LB]]) to (%[[WS_UB]]) inclusive step (%[[WS_STEP]])
-!CHECK:       fir.store %[[I]] to %[[ALLOCA_IV]]#1 : !fir.ref<i32>
+!CHECK:     omp.wsloop schedule(dynamic, nonmonotonic) nowait {
+!CHECK-NEXT:  omp.loop_nest (%[[I:.*]]) : i32 = (%[[WS_LB]]) to (%[[WS_UB]]) inclusive step (%[[WS_STEP]]) {
+!CHECK:         fir.store %[[I]] to %[[ALLOCA_IV]]#1 : !fir.ref<i32>
 
   do i=1, 9
     print*, i
-!CHECK:    %[[RTBEGIN:.*]] = fir.call @_FortranAioBeginExternalListOutput
-!CHECK:    %[[LOAD:.*]] = fir.load %[[ALLOCA_IV]]#0 : !fir.ref<i32>
-!CHECK:    fir.call @_FortranAioOutputInteger32(%[[RTBEGIN]], %[[LOAD]]) {{.*}}: (!fir.ref<i8>, i32) -> i1
-!CHECK:    fir.call @_FortranAioEndIoStatement(%[[RTBEGIN]]) {{.*}}: (!fir.ref<i8>) -> i32
+!CHECK:         %[[RTBEGIN:.*]] = fir.call @_FortranAioBeginExternalListOutput
+!CHECK:         %[[LOAD:.*]] = fir.load %[[ALLOCA_IV]]#0 : !fir.ref<i32>
+!CHECK:         fir.call @_FortranAioOutputInteger32(%[[RTBEGIN]], %[[LOAD]]) {{.*}}: (!fir.ref<i8>, i32) -> i1
+!CHECK:         fir.call @_FortranAioEndIoStatement(%[[RTBEGIN]]) {{.*}}: (!fir.ref<i8>) -> i32
   end do
-!CHECK:       omp.yield
-!CHECK:         }
+!CHECK:         omp.yield
+!CHECK:       }
 !CHECK:       omp.terminator
 !CHECK:     }
+!CHECK:     omp.terminator
+!CHECK:   }
 
 !$OMP END DO NOWAIT
 !$OMP END PARALLEL
diff --git a/flang/test/Lower/OpenMP/wsloop-ordered.f90 b/flang/test/Lower/OpenMP/wsloop-ordered.f90
index 5185d2d085ba..f4fa81c52315 100644
--- a/flang/test/Lower/OpenMP/wsloop-ordered.f90
+++ b/flang/test/Lower/OpenMP/wsloop-ordered.f90
@@ -6,9 +6,12 @@
 subroutine wsloop_ordered_no_para()
   integer :: a(10), i
 
-! CHECK:  omp.wsloop ordered(0) for (%{{.*}}) : i32 = (%{{.*}}) to (%{{.*}}) inclusive step (%{{.*}}) {
-! CHECK:    omp.yield
-! CHECK:  }
+! CHECK:  omp.wsloop ordered(0) {
+! CHECK-NEXT: omp.loop_nest (%{{.*}}) : i32 = (%{{.*}}) to (%{{.*}}) inclusive step (%{{.*}}) {
+! CHECK:        omp.yield
+! CHECK:      }
+! CHECK:      omp.terminator
+! CHECK:    }
 
   !$omp do ordered
   do i = 2, 10
@@ -25,9 +28,12 @@ subroutine wsloop_ordered_with_para()
   integer :: a(10), i
 
 ! CHECK: func @_QPwsloop_ordered_with_para() {
-! CHECK:  omp.wsloop ordered(1) for (%{{.*}}) : i32 = (%{{.*}}) to (%{{.*}}) inclusive step (%{{.*}}) {
-! CHECK:    omp.yield
-! CHECK:  }
+! CHECK:  omp.wsloop ordered(1) {
+! CHECK-NEXT: omp.loop_nest (%{{.*}}) : i32 = (%{{.*}}) to (%{{.*}}) inclusive step (%{{.*}}) {
+! CHECK:        omp.yield
+! CHECK:      }
+! CHECK:      omp.terminator
+! CHECK:    }
 
   !$omp do ordered(1)
   do i = 2, 10
diff --git a/flang/test/Lower/OpenMP/wsloop-reduction-add-byref.f90 b/flang/test/Lower/OpenMP/wsloop-reduction-add-byref.f90
index e63db33bbe25..c9d03435d9e1 100644
--- a/flang/test/Lower/OpenMP/wsloop-reduction-add-byref.f90
+++ b/flang/test/Lower/OpenMP/wsloop-reduction-add-byref.f90
@@ -82,14 +82,17 @@
 ! CHECK:             %[[VAL_7:.*]] = arith.constant 1 : i32
 ! CHECK:             %[[VAL_8:.*]] = arith.constant 100 : i32
 ! CHECK:             %[[VAL_9:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop byref reduction(@add_reduction_byref_i32 %[[VAL_3]]#0 -> %[[VAL_10:.*]] : !fir.ref<i32>)  for  (%[[VAL_11:.*]]) : i32 = (%[[VAL_7]]) to (%[[VAL_8]]) inclusive step (%[[VAL_9]]) {
-! CHECK:               fir.store %[[VAL_11]] to %[[VAL_6]]#1 : !fir.ref<i32>
-! CHECK:               %[[VAL_12:.*]]:2 = hlfir.declare %[[VAL_10]] {uniq_name = "_QFsimple_int_reductionEx"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
-! CHECK:               %[[VAL_13:.*]] = fir.load %[[VAL_12]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_14:.*]] = fir.load %[[VAL_6]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_15:.*]] = arith.addi %[[VAL_13]], %[[VAL_14]] : i32
-! CHECK:               hlfir.assign %[[VAL_15]] to %[[VAL_12]]#0 : i32, !fir.ref<i32>
-! CHECK:               omp.yield
+! CHECK:             omp.wsloop byref reduction(@add_reduction_byref_i32 %[[VAL_3]]#0 -> %[[VAL_10:.*]] : !fir.ref<i32>) {
+! CHECK-NEXT:          omp.loop_nest (%[[VAL_11:.*]]) : i32 = (%[[VAL_7]]) to (%[[VAL_8]]) inclusive step (%[[VAL_9]]) {
+! CHECK:                 %[[VAL_12:.*]]:2 = hlfir.declare %[[VAL_10]] {uniq_name = "_QFsimple_int_reductionEx"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
+! CHECK:                 fir.store %[[VAL_11]] to %[[VAL_6]]#1 : !fir.ref<i32>
+! CHECK:                 %[[VAL_13:.*]] = fir.load %[[VAL_12]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_14:.*]] = fir.load %[[VAL_6]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_15:.*]] = arith.addi %[[VAL_13]], %[[VAL_14]] : i32
+! CHECK:                 hlfir.assign %[[VAL_15]] to %[[VAL_12]]#0 : i32, !fir.ref<i32>
+! CHECK:                 omp.yield
+! CHECK:               }
+! CHECK:               omp.terminator
 ! CHECK:             }
 ! CHECK:             omp.terminator
 ! CHECK:           }
@@ -122,15 +125,18 @@ end subroutine
 ! CHECK:             %[[VAL_7:.*]] = arith.constant 1 : i32
 ! CHECK:             %[[VAL_8:.*]] = arith.constant 100 : i32
 ! CHECK:             %[[VAL_9:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop byref reduction(@add_reduction_byref_f32 %[[VAL_3]]#0 -> %[[VAL_10:.*]] : !fir.ref<f32>)  for  (%[[VAL_11:.*]]) : i32 = (%[[VAL_7]]) to (%[[VAL_8]]) inclusive step (%[[VAL_9]]) {
-! CHECK:               fir.store %[[VAL_11]] to %[[VAL_6]]#1 : !fir.ref<i32>
-! CHECK:               %[[VAL_12:.*]]:2 = hlfir.declare %[[VAL_10]] {uniq_name = "_QFsimple_real_reductionEx"} : (!fir.ref<f32>) -> (!fir.ref<f32>, !fir.ref<f32>)
-! CHECK:               %[[VAL_13:.*]] = fir.load %[[VAL_12]]#0 : !fir.ref<f32>
-! CHECK:               %[[VAL_14:.*]] = fir.load %[[VAL_6]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_15:.*]] = fir.convert %[[VAL_14]] : (i32) -> f32
-! CHECK:               %[[VAL_16:.*]] = arith.addf %[[VAL_13]], %[[VAL_15]] fastmath<contract> : f32
-! CHECK:               hlfir.assign %[[VAL_16]] to %[[VAL_12]]#0 : f32, !fir.ref<f32>
-! CHECK:               omp.yield
+! CHECK:             omp.wsloop byref reduction(@add_reduction_byref_f32 %[[VAL_3]]#0 -> %[[VAL_10:.*]] : !fir.ref<f32>) {
+! CHECK-NEXT:          omp.loop_nest (%[[VAL_11:.*]]) : i32 = (%[[VAL_7]]) to (%[[VAL_8]]) inclusive step (%[[VAL_9]]) {
+! CHECK:                 %[[VAL_12:.*]]:2 = hlfir.declare %[[VAL_10]] {uniq_name = "_QFsimple_real_reductionEx"} : (!fir.ref<f32>) -> (!fir.ref<f32>, !fir.ref<f32>)
+! CHECK:                 fir.store %[[VAL_11]] to %[[VAL_6]]#1 : !fir.ref<i32>
+! CHECK:                 %[[VAL_13:.*]] = fir.load %[[VAL_12]]#0 : !fir.ref<f32>
+! CHECK:                 %[[VAL_14:.*]] = fir.load %[[VAL_6]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_15:.*]] = fir.convert %[[VAL_14]] : (i32) -> f32
+! CHECK:                 %[[VAL_16:.*]] = arith.addf %[[VAL_13]], %[[VAL_15]] fastmath<contract> : f32
+! CHECK:                 hlfir.assign %[[VAL_16]] to %[[VAL_12]]#0 : f32, !fir.ref<f32>
+! CHECK:                 omp.yield
+! CHECK:               }
+! CHECK:               omp.terminator
 ! CHECK:             }
 ! CHECK:             omp.terminator
 ! CHECK:           }
@@ -163,14 +169,17 @@ end subroutine
 ! CHECK:             %[[VAL_7:.*]] = arith.constant 1 : i32
 ! CHECK:             %[[VAL_8:.*]] = arith.constant 100 : i32
 ! CHECK:             %[[VAL_9:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop byref reduction(@add_reduction_byref_i32 %[[VAL_3]]#0 -> %[[VAL_10:.*]] : !fir.ref<i32>)  for  (%[[VAL_11:.*]]) : i32 = (%[[VAL_7]]) to (%[[VAL_8]]) inclusive step (%[[VAL_9]]) {
-! CHECK:               fir.store %[[VAL_11]] to %[[VAL_6]]#1 : !fir.ref<i32>
-! CHECK:               %[[VAL_12:.*]]:2 = hlfir.declare %[[VAL_10]] {uniq_name = "_QFsimple_int_reduction_switch_orderEx"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
-! CHECK:               %[[VAL_13:.*]] = fir.load %[[VAL_6]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_14:.*]] = fir.load %[[VAL_12]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_15:.*]] = arith.addi %[[VAL_13]], %[[VAL_14]] : i32
-! CHECK:               hlfir.assign %[[VAL_15]] to %[[VAL_12]]#0 : i32, !fir.ref<i32>
-! CHECK:               omp.yield
+! CHECK:             omp.wsloop byref reduction(@add_reduction_byref_i32 %[[VAL_3]]#0 -> %[[VAL_10:.*]] : !fir.ref<i32>) {
+! CHECK-NEXT:          omp.loop_nest (%[[VAL_11:.*]]) : i32 = (%[[VAL_7]]) to (%[[VAL_8]]) inclusive step (%[[VAL_9]]) {
+! CHECK:                 %[[VAL_12:.*]]:2 = hlfir.declare %[[VAL_10]] {uniq_name = "_QFsimple_int_reduction_switch_orderEx"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
+! CHECK:                 fir.store %[[VAL_11]] to %[[VAL_6]]#1 : !fir.ref<i32>
+! CHECK:                 %[[VAL_13:.*]] = fir.load %[[VAL_6]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_14:.*]] = fir.load %[[VAL_12]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_15:.*]] = arith.addi %[[VAL_13]], %[[VAL_14]] : i32
+! CHECK:                 hlfir.assign %[[VAL_15]] to %[[VAL_12]]#0 : i32, !fir.ref<i32>
+! CHECK:                 omp.yield
+! CHECK:               }
+! CHECK:               omp.terminator
 ! CHECK:             }
 ! CHECK:             omp.terminator
 ! CHECK:           }
@@ -202,15 +211,18 @@ end subroutine
 ! CHECK:             %[[VAL_7:.*]] = arith.constant 1 : i32
 ! CHECK:             %[[VAL_8:.*]] = arith.constant 100 : i32
 ! CHECK:             %[[VAL_9:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop byref reduction(@add_reduction_byref_f32 %[[VAL_3]]#0 -> %[[VAL_10:.*]] : !fir.ref<f32>)  for  (%[[VAL_11:.*]]) : i32 = (%[[VAL_7]]) to (%[[VAL_8]]) inclusive step (%[[VAL_9]]) {
-! CHECK:               fir.store %[[VAL_11]] to %[[VAL_6]]#1 : !fir.ref<i32>
-! CHECK:               %[[VAL_12:.*]]:2 = hlfir.declare %[[VAL_10]] {uniq_name = "_QFsimple_real_reduction_switch_orderEx"} : (!fir.ref<f32>) -> (!fir.ref<f32>, !fir.ref<f32>)
-! CHECK:               %[[VAL_13:.*]] = fir.load %[[VAL_6]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_14:.*]] = fir.convert %[[VAL_13]] : (i32) -> f32
-! CHECK:               %[[VAL_15:.*]] = fir.load %[[VAL_12]]#0 : !fir.ref<f32>
-! CHECK:               %[[VAL_16:.*]] = arith.addf %[[VAL_14]], %[[VAL_15]] fastmath<contract> : f32
-! CHECK:               hlfir.assign %[[VAL_16]] to %[[VAL_12]]#0 : f32, !fir.ref<f32>
-! CHECK:               omp.yield
+! CHECK:             omp.wsloop byref reduction(@add_reduction_byref_f32 %[[VAL_3]]#0 -> %[[VAL_10:.*]] : !fir.ref<f32>) {
+! CHECK-NEXT:          omp.loop_nest (%[[VAL_11:.*]]) : i32 = (%[[VAL_7]]) to (%[[VAL_8]]) inclusive step (%[[VAL_9]]) {
+! CHECK:                 %[[VAL_12:.*]]:2 = hlfir.declare %[[VAL_10]] {uniq_name = "_QFsimple_real_reduction_switch_orderEx"} : (!fir.ref<f32>) -> (!fir.ref<f32>, !fir.ref<f32>)
+! CHECK:                 fir.store %[[VAL_11]] to %[[VAL_6]]#1 : !fir.ref<i32>
+! CHECK:                 %[[VAL_13:.*]] = fir.load %[[VAL_6]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_14:.*]] = fir.convert %[[VAL_13]] : (i32) -> f32
+! CHECK:                 %[[VAL_15:.*]] = fir.load %[[VAL_12]]#0 : !fir.ref<f32>
+! CHECK:                 %[[VAL_16:.*]] = arith.addf %[[VAL_14]], %[[VAL_15]] fastmath<contract> : f32
+! CHECK:                 hlfir.assign %[[VAL_16]] to %[[VAL_12]]#0 : f32, !fir.ref<f32>
+! CHECK:                 omp.yield
+! CHECK:               }
+! CHECK:               omp.terminator
 ! CHECK:             }
 ! CHECK:             omp.terminator
 ! CHECK:           }
@@ -250,24 +262,27 @@ end subroutine
 ! CHECK:             %[[VAL_13:.*]] = arith.constant 1 : i32
 ! CHECK:             %[[VAL_14:.*]] = arith.constant 100 : i32
 ! CHECK:             %[[VAL_15:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop byref reduction(@add_reduction_byref_i32 %[[VAL_3]]#0 -> %[[VAL_16:.*]] : !fir.ref<i32>, @add_reduction_byref_i32 %[[VAL_5]]#0 -> %[[VAL_17:.*]] : !fir.ref<i32>, @add_reduction_byref_i32 %[[VAL_7]]#0 -> %[[VAL_18:.*]] : !fir.ref<i32>)  for  (%[[VAL_19:.*]]) : i32 = (%[[VAL_13]]) to (%[[VAL_14]]) inclusive step (%[[VAL_15]]) {
-! CHECK:               fir.store %[[VAL_19]] to %[[VAL_12]]#1 : !fir.ref<i32>
-! CHECK:               %[[VAL_20:.*]]:2 = hlfir.declare %[[VAL_16]] {uniq_name = "_QFmultiple_int_reductions_same_typeEx"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
-! CHECK:               %[[VAL_21:.*]]:2 = hlfir.declare %[[VAL_17]] {uniq_name = "_QFmultiple_int_reductions_same_typeEy"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
-! CHECK:               %[[VAL_22:.*]]:2 = hlfir.declare %[[VAL_18]] {uniq_name = "_QFmultiple_int_reductions_same_typeEz"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
-! CHECK:               %[[VAL_23:.*]] = fir.load %[[VAL_20]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_24:.*]] = fir.load %[[VAL_12]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_25:.*]] = arith.addi %[[VAL_23]], %[[VAL_24]] : i32
-! CHECK:               hlfir.assign %[[VAL_25]] to %[[VAL_20]]#0 : i32, !fir.ref<i32>
-! CHECK:               %[[VAL_26:.*]] = fir.load %[[VAL_21]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_27:.*]] = fir.load %[[VAL_12]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_28:.*]] = arith.addi %[[VAL_26]], %[[VAL_27]] : i32
-! CHECK:               hlfir.assign %[[VAL_28]] to %[[VAL_21]]#0 : i32, !fir.ref<i32>
-! CHECK:               %[[VAL_29:.*]] = fir.load %[[VAL_22]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_30:.*]] = fir.load %[[VAL_12]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_31:.*]] = arith.addi %[[VAL_29]], %[[VAL_30]] : i32
-! CHECK:               hlfir.assign %[[VAL_31]] to %[[VAL_22]]#0 : i32, !fir.ref<i32>
-! CHECK:               omp.yield
+! CHECK:             omp.wsloop byref reduction(@add_reduction_byref_i32 %[[VAL_3]]#0 -> %[[VAL_16:.*]] : !fir.ref<i32>, @add_reduction_byref_i32 %[[VAL_5]]#0 -> %[[VAL_17:.*]] : !fir.ref<i32>, @add_reduction_byref_i32 %[[VAL_7]]#0 -> %[[VAL_18:.*]] : !fir.ref<i32>) {
+! CHECK-NEXT:          omp.loop_nest (%[[VAL_19:.*]]) : i32 = (%[[VAL_13]]) to (%[[VAL_14]]) inclusive step (%[[VAL_15]]) {
+! CHECK:                 %[[VAL_20:.*]]:2 = hlfir.declare %[[VAL_16]] {uniq_name = "_QFmultiple_int_reductions_same_typeEx"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
+! CHECK:                 %[[VAL_21:.*]]:2 = hlfir.declare %[[VAL_17]] {uniq_name = "_QFmultiple_int_reductions_same_typeEy"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
+! CHECK:                 %[[VAL_22:.*]]:2 = hlfir.declare %[[VAL_18]] {uniq_name = "_QFmultiple_int_reductions_same_typeEz"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
+! CHECK:                 fir.store %[[VAL_19]] to %[[VAL_12]]#1 : !fir.ref<i32>
+! CHECK:                 %[[VAL_23:.*]] = fir.load %[[VAL_20]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_24:.*]] = fir.load %[[VAL_12]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_25:.*]] = arith.addi %[[VAL_23]], %[[VAL_24]] : i32
+! CHECK:                 hlfir.assign %[[VAL_25]] to %[[VAL_20]]#0 : i32, !fir.ref<i32>
+! CHECK:                 %[[VAL_26:.*]] = fir.load %[[VAL_21]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_27:.*]] = fir.load %[[VAL_12]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_28:.*]] = arith.addi %[[VAL_26]], %[[VAL_27]] : i32
+! CHECK:                 hlfir.assign %[[VAL_28]] to %[[VAL_21]]#0 : i32, !fir.ref<i32>
+! CHECK:                 %[[VAL_29:.*]] = fir.load %[[VAL_22]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_30:.*]] = fir.load %[[VAL_12]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_31:.*]] = arith.addi %[[VAL_29]], %[[VAL_30]] : i32
+! CHECK:                 hlfir.assign %[[VAL_31]] to %[[VAL_22]]#0 : i32, !fir.ref<i32>
+! CHECK:                 omp.yield
+! CHECK:               }
+! CHECK:               omp.terminator
 ! CHECK:             }
 ! CHECK:             omp.terminator
 ! CHECK:           }
@@ -311,27 +326,30 @@ end subroutine
 ! CHECK:             %[[VAL_13:.*]] = arith.constant 1 : i32
 ! CHECK:             %[[VAL_14:.*]] = arith.constant 100 : i32
 ! CHECK:             %[[VAL_15:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop byref reduction(@add_reduction_byref_f32 %[[VAL_3]]#0 -> %[[VAL_16:.*]] : !fir.ref<f32>, @add_reduction_byref_f32 %[[VAL_5]]#0 -> %[[VAL_17:.*]] : !fir.ref<f32>, @add_reduction_byref_f32 %[[VAL_7]]#0 -> %[[VAL_18:.*]] : !fir.ref<f32>)  for  (%[[VAL_19:.*]]) : i32 = (%[[VAL_13]]) to (%[[VAL_14]]) inclusive step (%[[VAL_15]]) {
-! CHECK:               fir.store %[[VAL_19]] to %[[VAL_12]]#1 : !fir.ref<i32>
-! CHECK:               %[[VAL_20:.*]]:2 = hlfir.declare %[[VAL_16]] {uniq_name = "_QFmultiple_real_reductions_same_typeEx"} : (!fir.ref<f32>) -> (!fir.ref<f32>, !fir.ref<f32>)
-! CHECK:               %[[VAL_21:.*]]:2 = hlfir.declare %[[VAL_17]] {uniq_name = "_QFmultiple_real_reductions_same_typeEy"} : (!fir.ref<f32>) -> (!fir.ref<f32>, !fir.ref<f32>)
-! CHECK:               %[[VAL_22:.*]]:2 = hlfir.declare %[[VAL_18]] {uniq_name = "_QFmultiple_real_reductions_same_typeEz"} : (!fir.ref<f32>) -> (!fir.ref<f32>, !fir.ref<f32>)
-! CHECK:               %[[VAL_23:.*]] = fir.load %[[VAL_20]]#0 : !fir.ref<f32>
-! CHECK:               %[[VAL_24:.*]] = fir.load %[[VAL_12]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_25:.*]] = fir.convert %[[VAL_24]] : (i32) -> f32
-! CHECK:               %[[VAL_26:.*]] = arith.addf %[[VAL_23]], %[[VAL_25]] fastmath<contract> : f32
-! CHECK:               hlfir.assign %[[VAL_26]] to %[[VAL_20]]#0 : f32, !fir.ref<f32>
-! CHECK:               %[[VAL_27:.*]] = fir.load %[[VAL_21]]#0 : !fir.ref<f32>
-! CHECK:               %[[VAL_28:.*]] = fir.load %[[VAL_12]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_29:.*]] = fir.convert %[[VAL_28]] : (i32) -> f32
-! CHECK:               %[[VAL_30:.*]] = arith.addf %[[VAL_27]], %[[VAL_29]] fastmath<contract> : f32
-! CHECK:               hlfir.assign %[[VAL_30]] to %[[VAL_21]]#0 : f32, !fir.ref<f32>
-! CHECK:               %[[VAL_31:.*]] = fir.load %[[VAL_22]]#0 : !fir.ref<f32>
-! CHECK:               %[[VAL_32:.*]] = fir.load %[[VAL_12]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_33:.*]] = fir.convert %[[VAL_32]] : (i32) -> f32
-! CHECK:               %[[VAL_34:.*]] = arith.addf %[[VAL_31]], %[[VAL_33]] fastmath<contract> : f32
-! CHECK:               hlfir.assign %[[VAL_34]] to %[[VAL_22]]#0 : f32, !fir.ref<f32>
-! CHECK:               omp.yield
+! CHECK:             omp.wsloop byref reduction(@add_reduction_byref_f32 %[[VAL_3]]#0 -> %[[VAL_16:.*]] : !fir.ref<f32>, @add_reduction_byref_f32 %[[VAL_5]]#0 -> %[[VAL_17:.*]] : !fir.ref<f32>, @add_reduction_byref_f32 %[[VAL_7]]#0 -> %[[VAL_18:.*]] : !fir.ref<f32>) {
+! CHECK-NEXT:          omp.loop_nest (%[[VAL_19:.*]]) : i32 = (%[[VAL_13]]) to (%[[VAL_14]]) inclusive step (%[[VAL_15]]) {
+! CHECK:                 %[[VAL_20:.*]]:2 = hlfir.declare %[[VAL_16]] {uniq_name = "_QFmultiple_real_reductions_same_typeEx"} : (!fir.ref<f32>) -> (!fir.ref<f32>, !fir.ref<f32>)
+! CHECK:                 %[[VAL_21:.*]]:2 = hlfir.declare %[[VAL_17]] {uniq_name = "_QFmultiple_real_reductions_same_typeEy"} : (!fir.ref<f32>) -> (!fir.ref<f32>, !fir.ref<f32>)
+! CHECK:                 %[[VAL_22:.*]]:2 = hlfir.declare %[[VAL_18]] {uniq_name = "_QFmultiple_real_reductions_same_typeEz"} : (!fir.ref<f32>) -> (!fir.ref<f32>, !fir.ref<f32>)
+! CHECK:                 fir.store %[[VAL_19]] to %[[VAL_12]]#1 : !fir.ref<i32>
+! CHECK:                 %[[VAL_23:.*]] = fir.load %[[VAL_20]]#0 : !fir.ref<f32>
+! CHECK:                 %[[VAL_24:.*]] = fir.load %[[VAL_12]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_25:.*]] = fir.convert %[[VAL_24]] : (i32) -> f32
+! CHECK:                 %[[VAL_26:.*]] = arith.addf %[[VAL_23]], %[[VAL_25]] fastmath<contract> : f32
+! CHECK:                 hlfir.assign %[[VAL_26]] to %[[VAL_20]]#0 : f32, !fir.ref<f32>
+! CHECK:                 %[[VAL_27:.*]] = fir.load %[[VAL_21]]#0 : !fir.ref<f32>
+! CHECK:                 %[[VAL_28:.*]] = fir.load %[[VAL_12]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_29:.*]] = fir.convert %[[VAL_28]] : (i32) -> f32
+! CHECK:                 %[[VAL_30:.*]] = arith.addf %[[VAL_27]], %[[VAL_29]] fastmath<contract> : f32
+! CHECK:                 hlfir.assign %[[VAL_30]] to %[[VAL_21]]#0 : f32, !fir.ref<f32>
+! CHECK:                 %[[VAL_31:.*]] = fir.load %[[VAL_22]]#0 : !fir.ref<f32>
+! CHECK:                 %[[VAL_32:.*]] = fir.load %[[VAL_12]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_33:.*]] = fir.convert %[[VAL_32]] : (i32) -> f32
+! CHECK:                 %[[VAL_34:.*]] = arith.addf %[[VAL_31]], %[[VAL_33]] fastmath<contract> : f32
+! CHECK:                 hlfir.assign %[[VAL_34]] to %[[VAL_22]]#0 : f32, !fir.ref<f32>
+! CHECK:                 omp.yield
+! CHECK:               }
+! CHECK:               omp.terminator
 ! CHECK:             }
 ! CHECK:             omp.terminator
 ! CHECK:           }
@@ -379,32 +397,35 @@ end subroutine
 ! CHECK:             %[[VAL_16:.*]] = arith.constant 1 : i32
 ! CHECK:             %[[VAL_17:.*]] = arith.constant 100 : i32
 ! CHECK:             %[[VAL_18:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop byref reduction(@add_reduction_byref_i32 %[[VAL_5]]#0 -> %[[VAL_19:.*]] : !fir.ref<i32>, @add_reduction_byref_i64 %[[VAL_7]]#0 -> %[[VAL_20:.*]] : !fir.ref<i64>, @add_reduction_byref_f32 %[[VAL_9]]#0 -> %[[VAL_21:.*]] : !fir.ref<f32>, @add_reduction_byref_f64 %[[VAL_3]]#0 -> %[[VAL_22:.*]] : !fir.ref<f64>)  for  (%[[VAL_23:.*]]) : i32 = (%[[VAL_16]]) to (%[[VAL_17]]) inclusive step (%[[VAL_18]]) {
-! CHECK:               fir.store %[[VAL_23]] to %[[VAL_15]]#1 : !fir.ref<i32>
-! CHECK:               %[[VAL_24:.*]]:2 = hlfir.declare %[[VAL_19]] {uniq_name = "_QFmultiple_reductions_different_typeEx"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
-! CHECK:               %[[VAL_25:.*]]:2 = hlfir.declare %[[VAL_20]] {uniq_name = "_QFmultiple_reductions_different_typeEy"} : (!fir.ref<i64>) -> (!fir.ref<i64>, !fir.ref<i64>)
-! CHECK:               %[[VAL_26:.*]]:2 = hlfir.declare %[[VAL_21]] {uniq_name = "_QFmultiple_reductions_different_typeEz"} : (!fir.ref<f32>) -> (!fir.ref<f32>, !fir.ref<f32>)
-! CHECK:               %[[VAL_27:.*]]:2 = hlfir.declare %[[VAL_22]] {uniq_name = "_QFmultiple_reductions_different_typeEw"} : (!fir.ref<f64>) -> (!fir.ref<f64>, !fir.ref<f64>)
-! CHECK:               %[[VAL_28:.*]] = fir.load %[[VAL_24]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_29:.*]] = fir.load %[[VAL_15]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_30:.*]] = arith.addi %[[VAL_28]], %[[VAL_29]] : i32
-! CHECK:               hlfir.assign %[[VAL_30]] to %[[VAL_24]]#0 : i32, !fir.ref<i32>
-! CHECK:               %[[VAL_31:.*]] = fir.load %[[VAL_25]]#0 : !fir.ref<i64>
-! CHECK:               %[[VAL_32:.*]] = fir.load %[[VAL_15]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_33:.*]] = fir.convert %[[VAL_32]] : (i32) -> i64
-! CHECK:               %[[VAL_34:.*]] = arith.addi %[[VAL_31]], %[[VAL_33]] : i64
-! CHECK:               hlfir.assign %[[VAL_34]] to %[[VAL_25]]#0 : i64, !fir.ref<i64>
-! CHECK:               %[[VAL_35:.*]] = fir.load %[[VAL_26]]#0 : !fir.ref<f32>
-! CHECK:               %[[VAL_36:.*]] = fir.load %[[VAL_15]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_37:.*]] = fir.convert %[[VAL_36]] : (i32) -> f32
-! CHECK:               %[[VAL_38:.*]] = arith.addf %[[VAL_35]], %[[VAL_37]] fastmath<contract> : f32
-! CHECK:               hlfir.assign %[[VAL_38]] to %[[VAL_26]]#0 : f32, !fir.ref<f32>
-! CHECK:               %[[VAL_39:.*]] = fir.load %[[VAL_27]]#0 : !fir.ref<f64>
-! CHECK:               %[[VAL_40:.*]] = fir.load %[[VAL_15]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_41:.*]] = fir.convert %[[VAL_40]] : (i32) -> f64
-! CHECK:               %[[VAL_42:.*]] = arith.addf %[[VAL_39]], %[[VAL_41]] fastmath<contract> : f64
-! CHECK:               hlfir.assign %[[VAL_42]] to %[[VAL_27]]#0 : f64, !fir.ref<f64>
-! CHECK:               omp.yield
+! CHECK:             omp.wsloop byref reduction(@add_reduction_byref_i32 %[[VAL_5]]#0 -> %[[VAL_19:.*]] : !fir.ref<i32>, @add_reduction_byref_i64 %[[VAL_7]]#0 -> %[[VAL_20:.*]] : !fir.ref<i64>, @add_reduction_byref_f32 %[[VAL_9]]#0 -> %[[VAL_21:.*]] : !fir.ref<f32>, @add_reduction_byref_f64 %[[VAL_3]]#0 -> %[[VAL_22:.*]] : !fir.ref<f64>) {
+! CHECK-NEXT:          omp.loop_nest (%[[VAL_23:.*]]) : i32 = (%[[VAL_16]]) to (%[[VAL_17]]) inclusive step (%[[VAL_18]]) {
+! CHECK:                 %[[VAL_24:.*]]:2 = hlfir.declare %[[VAL_19]] {uniq_name = "_QFmultiple_reductions_different_typeEx"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
+! CHECK:                 %[[VAL_25:.*]]:2 = hlfir.declare %[[VAL_20]] {uniq_name = "_QFmultiple_reductions_different_typeEy"} : (!fir.ref<i64>) -> (!fir.ref<i64>, !fir.ref<i64>)
+! CHECK:                 %[[VAL_26:.*]]:2 = hlfir.declare %[[VAL_21]] {uniq_name = "_QFmultiple_reductions_different_typeEz"} : (!fir.ref<f32>) -> (!fir.ref<f32>, !fir.ref<f32>)
+! CHECK:                 %[[VAL_27:.*]]:2 = hlfir.declare %[[VAL_22]] {uniq_name = "_QFmultiple_reductions_different_typeEw"} : (!fir.ref<f64>) -> (!fir.ref<f64>, !fir.ref<f64>)
+! CHECK:                 fir.store %[[VAL_23]] to %[[VAL_15]]#1 : !fir.ref<i32>
+! CHECK:                 %[[VAL_28:.*]] = fir.load %[[VAL_24]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_29:.*]] = fir.load %[[VAL_15]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_30:.*]] = arith.addi %[[VAL_28]], %[[VAL_29]] : i32
+! CHECK:                 hlfir.assign %[[VAL_30]] to %[[VAL_24]]#0 : i32, !fir.ref<i32>
+! CHECK:                 %[[VAL_31:.*]] = fir.load %[[VAL_25]]#0 : !fir.ref<i64>
+! CHECK:                 %[[VAL_32:.*]] = fir.load %[[VAL_15]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_33:.*]] = fir.convert %[[VAL_32]] : (i32) -> i64
+! CHECK:                 %[[VAL_34:.*]] = arith.addi %[[VAL_31]], %[[VAL_33]] : i64
+! CHECK:                 hlfir.assign %[[VAL_34]] to %[[VAL_25]]#0 : i64, !fir.ref<i64>
+! CHECK:                 %[[VAL_35:.*]] = fir.load %[[VAL_26]]#0 : !fir.ref<f32>
+! CHECK:                 %[[VAL_36:.*]] = fir.load %[[VAL_15]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_37:.*]] = fir.convert %[[VAL_36]] : (i32) -> f32
+! CHECK:                 %[[VAL_38:.*]] = arith.addf %[[VAL_35]], %[[VAL_37]] fastmath<contract> : f32
+! CHECK:                 hlfir.assign %[[VAL_38]] to %[[VAL_26]]#0 : f32, !fir.ref<f32>
+! CHECK:                 %[[VAL_39:.*]] = fir.load %[[VAL_27]]#0 : !fir.ref<f64>
+! CHECK:                 %[[VAL_40:.*]] = fir.load %[[VAL_15]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_41:.*]] = fir.convert %[[VAL_40]] : (i32) -> f64
+! CHECK:                 %[[VAL_42:.*]] = arith.addf %[[VAL_39]], %[[VAL_41]] fastmath<contract> : f64
+! CHECK:                 hlfir.assign %[[VAL_42]] to %[[VAL_27]]#0 : f64, !fir.ref<f64>
+! CHECK:                 omp.yield
+! CHECK:               }
+! CHECK:               omp.terminator
 ! CHECK:             }
 ! CHECK:             omp.terminator
 ! CHECK:           }
diff --git a/flang/test/Lower/OpenMP/wsloop-reduction-add-hlfir-byref.f90 b/flang/test/Lower/OpenMP/wsloop-reduction-add-hlfir-byref.f90
index 3b4d9666c693..6a09fece80ae 100644
--- a/flang/test/Lower/OpenMP/wsloop-reduction-add-hlfir-byref.f90
+++ b/flang/test/Lower/OpenMP/wsloop-reduction-add-hlfir-byref.f90
@@ -31,14 +31,16 @@
 ! CHECK:             %[[VAL_7:.*]] = arith.constant 1 : i32
 ! CHECK:             %[[VAL_8:.*]] = arith.constant 100 : i32
 ! CHECK:             %[[VAL_9:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop byref reduction(@add_reduction_byref_i32 %[[VAL_3]]#0 -> %[[VAL_10:.*]] : !fir.ref<i32>)  for  (%[[VAL_11:.*]]) : i32 = (%[[VAL_7]]) to (%[[VAL_8]]) inclusive step (%[[VAL_9]])
-! CHECK:               fir.store %[[VAL_11]] to %[[VAL_6]]#1 : !fir.ref<i32>
-! CHECK:               %[[VAL_12:.*]]:2 = hlfir.declare %[[VAL_10]] {uniq_name = "_QFsimple_int_reductionEx"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
-! CHECK:               %[[VAL_13:.*]] = fir.load %[[VAL_12]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_14:.*]] = fir.load %[[VAL_6]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_15:.*]] = arith.addi %[[VAL_13]], %[[VAL_14]] : i32
-! CHECK:               hlfir.assign %[[VAL_15]] to %[[VAL_12]]#0 : i32, !fir.ref<i32>
-! CHECK:               omp.yield
+! CHECK:             omp.wsloop byref reduction(@add_reduction_byref_i32 %[[VAL_3]]#0 -> %[[VAL_10:.*]] : !fir.ref<i32>) {
+! CHECK-NEXT:          omp.loop_nest (%[[VAL_11:.*]]) : i32 = (%[[VAL_7]]) to (%[[VAL_8]]) inclusive step (%[[VAL_9]]) {
+! CHECK:                 %[[VAL_12:.*]]:2 = hlfir.declare %[[VAL_10]] {uniq_name = "_QFsimple_int_reductionEx"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
+! CHECK:                 fir.store %[[VAL_11]] to %[[VAL_6]]#1 : !fir.ref<i32>
+! CHECK:                 %[[VAL_13:.*]] = fir.load %[[VAL_12]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_14:.*]] = fir.load %[[VAL_6]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_15:.*]] = arith.addi %[[VAL_13]], %[[VAL_14]] : i32
+! CHECK:                 hlfir.assign %[[VAL_15]] to %[[VAL_12]]#0 : i32, !fir.ref<i32>
+! CHECK:                 omp.yield
+! CHECK:               omp.terminator
 ! CHECK:             omp.terminator
 ! CHECK:           return
 
diff --git a/flang/test/Lower/OpenMP/wsloop-reduction-add-hlfir.f90 b/flang/test/Lower/OpenMP/wsloop-reduction-add-hlfir.f90
index 7c9070592e46..c5cc5a95cef1 100644
--- a/flang/test/Lower/OpenMP/wsloop-reduction-add-hlfir.f90
+++ b/flang/test/Lower/OpenMP/wsloop-reduction-add-hlfir.f90
@@ -27,14 +27,16 @@
 ! CHECK:             %[[VAL_7:.*]] = arith.constant 1 : i32
 ! CHECK:             %[[VAL_8:.*]] = arith.constant 100 : i32
 ! CHECK:             %[[VAL_9:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop reduction(@add_reduction_i32 %[[VAL_3]]#0 -> %[[VAL_10:.*]] : !fir.ref<i32>)  for  (%[[VAL_11:.*]]) : i32 = (%[[VAL_7]]) to (%[[VAL_8]]) inclusive step (%[[VAL_9]])
-! CHECK:               fir.store %[[VAL_11]] to %[[VAL_6]]#1 : !fir.ref<i32>
-! CHECK:               %[[VAL_12:.*]]:2 = hlfir.declare %[[VAL_10]] {uniq_name = "_QFsimple_int_reductionEx"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
-! CHECK:               %[[VAL_13:.*]] = fir.load %[[VAL_12]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_14:.*]] = fir.load %[[VAL_6]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_15:.*]] = arith.addi %[[VAL_13]], %[[VAL_14]] : i32
-! CHECK:               hlfir.assign %[[VAL_15]] to %[[VAL_12]]#0 : i32, !fir.ref<i32>
-! CHECK:               omp.yield
+! CHECK:             omp.wsloop reduction(@add_reduction_i32 %[[VAL_3]]#0 -> %[[VAL_10:.*]] : !fir.ref<i32>) {
+! CHECK-NEXT:          omp.loop_nest (%[[VAL_11:.*]]) : i32 = (%[[VAL_7]]) to (%[[VAL_8]]) inclusive step (%[[VAL_9]]) {
+! CHECK:                 %[[VAL_12:.*]]:2 = hlfir.declare %[[VAL_10]] {uniq_name = "_QFsimple_int_reductionEx"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
+! CHECK:                 fir.store %[[VAL_11]] to %[[VAL_6]]#1 : !fir.ref<i32>
+! CHECK:                 %[[VAL_13:.*]] = fir.load %[[VAL_12]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_14:.*]] = fir.load %[[VAL_6]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_15:.*]] = arith.addi %[[VAL_13]], %[[VAL_14]] : i32
+! CHECK:                 hlfir.assign %[[VAL_15]] to %[[VAL_12]]#0 : i32, !fir.ref<i32>
+! CHECK:                 omp.yield
+! CHECK:               omp.terminator
 ! CHECK:             omp.terminator
 ! CHECK:           return
 
diff --git a/flang/test/Lower/OpenMP/wsloop-reduction-add.f90 b/flang/test/Lower/OpenMP/wsloop-reduction-add.f90
index 11e1ffb79f8e..5b957959f40d 100644
--- a/flang/test/Lower/OpenMP/wsloop-reduction-add.f90
+++ b/flang/test/Lower/OpenMP/wsloop-reduction-add.f90
@@ -58,14 +58,17 @@
 ! CHECK:             %[[VAL_7:.*]] = arith.constant 1 : i32
 ! CHECK:             %[[VAL_8:.*]] = arith.constant 100 : i32
 ! CHECK:             %[[VAL_9:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop reduction(@add_reduction_i32 %[[VAL_3]]#0 -> %[[VAL_10:.*]] : !fir.ref<i32>)  for  (%[[VAL_11:.*]]) : i32 = (%[[VAL_7]]) to (%[[VAL_8]]) inclusive step (%[[VAL_9]]) {
-! CHECK:               fir.store %[[VAL_11]] to %[[VAL_6]]#1 : !fir.ref<i32>
-! CHECK:               %[[VAL_12:.*]]:2 = hlfir.declare %[[VAL_10]] {uniq_name = "_QFsimple_int_reductionEx"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
-! CHECK:               %[[VAL_13:.*]] = fir.load %[[VAL_12]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_14:.*]] = fir.load %[[VAL_6]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_15:.*]] = arith.addi %[[VAL_13]], %[[VAL_14]] : i32
-! CHECK:               hlfir.assign %[[VAL_15]] to %[[VAL_12]]#0 : i32, !fir.ref<i32>
-! CHECK:               omp.yield
+! CHECK:             omp.wsloop reduction(@add_reduction_i32 %[[VAL_3]]#0 -> %[[VAL_10:.*]] : !fir.ref<i32>) {
+! CHECK-NEXT:          omp.loop_nest (%[[VAL_11:.*]]) : i32 = (%[[VAL_7]]) to (%[[VAL_8]]) inclusive step (%[[VAL_9]]) {
+! CHECK:                 %[[VAL_12:.*]]:2 = hlfir.declare %[[VAL_10]] {uniq_name = "_QFsimple_int_reductionEx"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
+! CHECK:                 fir.store %[[VAL_11]] to %[[VAL_6]]#1 : !fir.ref<i32>
+! CHECK:                 %[[VAL_13:.*]] = fir.load %[[VAL_12]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_14:.*]] = fir.load %[[VAL_6]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_15:.*]] = arith.addi %[[VAL_13]], %[[VAL_14]] : i32
+! CHECK:                 hlfir.assign %[[VAL_15]] to %[[VAL_12]]#0 : i32, !fir.ref<i32>
+! CHECK:                 omp.yield
+! CHECK:               }
+! CHECK:               omp.terminator
 ! CHECK:             }
 ! CHECK:             omp.terminator
 ! CHECK:           }
@@ -98,15 +101,18 @@ end subroutine
 ! CHECK:             %[[VAL_7:.*]] = arith.constant 1 : i32
 ! CHECK:             %[[VAL_8:.*]] = arith.constant 100 : i32
 ! CHECK:             %[[VAL_9:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop reduction(@add_reduction_f32 %[[VAL_3]]#0 -> %[[VAL_10:.*]] : !fir.ref<f32>)  for  (%[[VAL_11:.*]]) : i32 = (%[[VAL_7]]) to (%[[VAL_8]]) inclusive step (%[[VAL_9]]) {
-! CHECK:               fir.store %[[VAL_11]] to %[[VAL_6]]#1 : !fir.ref<i32>
-! CHECK:               %[[VAL_12:.*]]:2 = hlfir.declare %[[VAL_10]] {uniq_name = "_QFsimple_real_reductionEx"} : (!fir.ref<f32>) -> (!fir.ref<f32>, !fir.ref<f32>)
-! CHECK:               %[[VAL_13:.*]] = fir.load %[[VAL_12]]#0 : !fir.ref<f32>
-! CHECK:               %[[VAL_14:.*]] = fir.load %[[VAL_6]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_15:.*]] = fir.convert %[[VAL_14]] : (i32) -> f32
-! CHECK:               %[[VAL_16:.*]] = arith.addf %[[VAL_13]], %[[VAL_15]] fastmath<contract> : f32
-! CHECK:               hlfir.assign %[[VAL_16]] to %[[VAL_12]]#0 : f32, !fir.ref<f32>
-! CHECK:               omp.yield
+! CHECK:             omp.wsloop reduction(@add_reduction_f32 %[[VAL_3]]#0 -> %[[VAL_10:.*]] : !fir.ref<f32>) {
+! CHECK-NEXT:          omp.loop_nest (%[[VAL_11:.*]]) : i32 = (%[[VAL_7]]) to (%[[VAL_8]]) inclusive step (%[[VAL_9]]) {
+! CHECK:                 %[[VAL_12:.*]]:2 = hlfir.declare %[[VAL_10]] {uniq_name = "_QFsimple_real_reductionEx"} : (!fir.ref<f32>) -> (!fir.ref<f32>, !fir.ref<f32>)
+! CHECK:                 fir.store %[[VAL_11]] to %[[VAL_6]]#1 : !fir.ref<i32>
+! CHECK:                 %[[VAL_13:.*]] = fir.load %[[VAL_12]]#0 : !fir.ref<f32>
+! CHECK:                 %[[VAL_14:.*]] = fir.load %[[VAL_6]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_15:.*]] = fir.convert %[[VAL_14]] : (i32) -> f32
+! CHECK:                 %[[VAL_16:.*]] = arith.addf %[[VAL_13]], %[[VAL_15]] fastmath<contract> : f32
+! CHECK:                 hlfir.assign %[[VAL_16]] to %[[VAL_12]]#0 : f32, !fir.ref<f32>
+! CHECK:                 omp.yield
+! CHECK:               }
+! CHECK:               omp.terminator
 ! CHECK:             }
 ! CHECK:             omp.terminator
 ! CHECK:           }
@@ -139,14 +145,17 @@ end subroutine
 ! CHECK:             %[[VAL_7:.*]] = arith.constant 1 : i32
 ! CHECK:             %[[VAL_8:.*]] = arith.constant 100 : i32
 ! CHECK:             %[[VAL_9:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop reduction(@add_reduction_i32 %[[VAL_3]]#0 -> %[[VAL_10:.*]] : !fir.ref<i32>)  for  (%[[VAL_11:.*]]) : i32 = (%[[VAL_7]]) to (%[[VAL_8]]) inclusive step (%[[VAL_9]]) {
-! CHECK:               fir.store %[[VAL_11]] to %[[VAL_6]]#1 : !fir.ref<i32>
-! CHECK:               %[[VAL_12:.*]]:2 = hlfir.declare %[[VAL_10]] {uniq_name = "_QFsimple_int_reduction_switch_orderEx"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
-! CHECK:               %[[VAL_13:.*]] = fir.load %[[VAL_6]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_14:.*]] = fir.load %[[VAL_12]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_15:.*]] = arith.addi %[[VAL_13]], %[[VAL_14]] : i32
-! CHECK:               hlfir.assign %[[VAL_15]] to %[[VAL_12]]#0 : i32, !fir.ref<i32>
-! CHECK:               omp.yield
+! CHECK:             omp.wsloop reduction(@add_reduction_i32 %[[VAL_3]]#0 -> %[[VAL_10:.*]] : !fir.ref<i32>) {
+! CHECK-NEXT:          omp.loop_nest (%[[VAL_11:.*]]) : i32 = (%[[VAL_7]]) to (%[[VAL_8]]) inclusive step (%[[VAL_9]]) {
+! CHECK:                 %[[VAL_12:.*]]:2 = hlfir.declare %[[VAL_10]] {uniq_name = "_QFsimple_int_reduction_switch_orderEx"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
+! CHECK:                 fir.store %[[VAL_11]] to %[[VAL_6]]#1 : !fir.ref<i32>
+! CHECK:                 %[[VAL_13:.*]] = fir.load %[[VAL_6]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_14:.*]] = fir.load %[[VAL_12]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_15:.*]] = arith.addi %[[VAL_13]], %[[VAL_14]] : i32
+! CHECK:                 hlfir.assign %[[VAL_15]] to %[[VAL_12]]#0 : i32, !fir.ref<i32>
+! CHECK:                 omp.yield
+! CHECK:               }
+! CHECK:               omp.terminator
 ! CHECK:             }
 ! CHECK:             omp.terminator
 ! CHECK:           }
@@ -178,15 +187,18 @@ end subroutine
 ! CHECK:             %[[VAL_7:.*]] = arith.constant 1 : i32
 ! CHECK:             %[[VAL_8:.*]] = arith.constant 100 : i32
 ! CHECK:             %[[VAL_9:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop reduction(@add_reduction_f32 %[[VAL_3]]#0 -> %[[VAL_10:.*]] : !fir.ref<f32>)  for  (%[[VAL_11:.*]]) : i32 = (%[[VAL_7]]) to (%[[VAL_8]]) inclusive step (%[[VAL_9]]) {
-! CHECK:               fir.store %[[VAL_11]] to %[[VAL_6]]#1 : !fir.ref<i32>
-! CHECK:               %[[VAL_12:.*]]:2 = hlfir.declare %[[VAL_10]] {uniq_name = "_QFsimple_real_reduction_switch_orderEx"} : (!fir.ref<f32>) -> (!fir.ref<f32>, !fir.ref<f32>)
-! CHECK:               %[[VAL_13:.*]] = fir.load %[[VAL_6]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_14:.*]] = fir.convert %[[VAL_13]] : (i32) -> f32
-! CHECK:               %[[VAL_15:.*]] = fir.load %[[VAL_12]]#0 : !fir.ref<f32>
-! CHECK:               %[[VAL_16:.*]] = arith.addf %[[VAL_14]], %[[VAL_15]] fastmath<contract> : f32
-! CHECK:               hlfir.assign %[[VAL_16]] to %[[VAL_12]]#0 : f32, !fir.ref<f32>
-! CHECK:               omp.yield
+! CHECK:             omp.wsloop reduction(@add_reduction_f32 %[[VAL_3]]#0 -> %[[VAL_10:.*]] : !fir.ref<f32>) {
+! CHECK-NEXT:          omp.loop_nest (%[[VAL_11:.*]]) : i32 = (%[[VAL_7]]) to (%[[VAL_8]]) inclusive step (%[[VAL_9]]) {
+! CHECK:                 %[[VAL_12:.*]]:2 = hlfir.declare %[[VAL_10]] {uniq_name = "_QFsimple_real_reduction_switch_orderEx"} : (!fir.ref<f32>) -> (!fir.ref<f32>, !fir.ref<f32>)
+! CHECK:                 fir.store %[[VAL_11]] to %[[VAL_6]]#1 : !fir.ref<i32>
+! CHECK:                 %[[VAL_13:.*]] = fir.load %[[VAL_6]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_14:.*]] = fir.convert %[[VAL_13]] : (i32) -> f32
+! CHECK:                 %[[VAL_15:.*]] = fir.load %[[VAL_12]]#0 : !fir.ref<f32>
+! CHECK:                 %[[VAL_16:.*]] = arith.addf %[[VAL_14]], %[[VAL_15]] fastmath<contract> : f32
+! CHECK:                 hlfir.assign %[[VAL_16]] to %[[VAL_12]]#0 : f32, !fir.ref<f32>
+! CHECK:                 omp.yield
+! CHECK:               }
+! CHECK:               omp.terminator
 ! CHECK:             }
 ! CHECK:             omp.terminator
 ! CHECK:           }
@@ -226,24 +238,27 @@ end subroutine
 ! CHECK:             %[[VAL_13:.*]] = arith.constant 1 : i32
 ! CHECK:             %[[VAL_14:.*]] = arith.constant 100 : i32
 ! CHECK:             %[[VAL_15:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop reduction(@add_reduction_i32 %[[VAL_3]]#0 -> %[[VAL_16:.*]] : !fir.ref<i32>, @add_reduction_i32 %[[VAL_5]]#0 -> %[[VAL_17:.*]] : !fir.ref<i32>, @add_reduction_i32 %[[VAL_7]]#0 -> %[[VAL_18:.*]] : !fir.ref<i32>)  for  (%[[VAL_19:.*]]) : i32 = (%[[VAL_13]]) to (%[[VAL_14]]) inclusive step (%[[VAL_15]]) {
-! CHECK:               fir.store %[[VAL_19]] to %[[VAL_12]]#1 : !fir.ref<i32>
-! CHECK:               %[[VAL_20:.*]]:2 = hlfir.declare %[[VAL_16]] {uniq_name = "_QFmultiple_int_reductions_same_typeEx"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
-! CHECK:               %[[VAL_21:.*]]:2 = hlfir.declare %[[VAL_17]] {uniq_name = "_QFmultiple_int_reductions_same_typeEy"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
-! CHECK:               %[[VAL_22:.*]]:2 = hlfir.declare %[[VAL_18]] {uniq_name = "_QFmultiple_int_reductions_same_typeEz"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
-! CHECK:               %[[VAL_23:.*]] = fir.load %[[VAL_20]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_24:.*]] = fir.load %[[VAL_12]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_25:.*]] = arith.addi %[[VAL_23]], %[[VAL_24]] : i32
-! CHECK:               hlfir.assign %[[VAL_25]] to %[[VAL_20]]#0 : i32, !fir.ref<i32>
-! CHECK:               %[[VAL_26:.*]] = fir.load %[[VAL_21]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_27:.*]] = fir.load %[[VAL_12]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_28:.*]] = arith.addi %[[VAL_26]], %[[VAL_27]] : i32
-! CHECK:               hlfir.assign %[[VAL_28]] to %[[VAL_21]]#0 : i32, !fir.ref<i32>
-! CHECK:               %[[VAL_29:.*]] = fir.load %[[VAL_22]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_30:.*]] = fir.load %[[VAL_12]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_31:.*]] = arith.addi %[[VAL_29]], %[[VAL_30]] : i32
-! CHECK:               hlfir.assign %[[VAL_31]] to %[[VAL_22]]#0 : i32, !fir.ref<i32>
-! CHECK:               omp.yield
+! CHECK:             omp.wsloop reduction(@add_reduction_i32 %[[VAL_3]]#0 -> %[[VAL_16:.*]] : !fir.ref<i32>, @add_reduction_i32 %[[VAL_5]]#0 -> %[[VAL_17:.*]] : !fir.ref<i32>, @add_reduction_i32 %[[VAL_7]]#0 -> %[[VAL_18:.*]] : !fir.ref<i32>) {
+! CHECK-NEXT:          omp.loop_nest (%[[VAL_19:.*]]) : i32 = (%[[VAL_13]]) to (%[[VAL_14]]) inclusive step (%[[VAL_15]]) {
+! CHECK:                 %[[VAL_20:.*]]:2 = hlfir.declare %[[VAL_16]] {uniq_name = "_QFmultiple_int_reductions_same_typeEx"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
+! CHECK:                 %[[VAL_21:.*]]:2 = hlfir.declare %[[VAL_17]] {uniq_name = "_QFmultiple_int_reductions_same_typeEy"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
+! CHECK:                 %[[VAL_22:.*]]:2 = hlfir.declare %[[VAL_18]] {uniq_name = "_QFmultiple_int_reductions_same_typeEz"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
+! CHECK:                 fir.store %[[VAL_19]] to %[[VAL_12]]#1 : !fir.ref<i32>
+! CHECK:                 %[[VAL_23:.*]] = fir.load %[[VAL_20]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_24:.*]] = fir.load %[[VAL_12]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_25:.*]] = arith.addi %[[VAL_23]], %[[VAL_24]] : i32
+! CHECK:                 hlfir.assign %[[VAL_25]] to %[[VAL_20]]#0 : i32, !fir.ref<i32>
+! CHECK:                 %[[VAL_26:.*]] = fir.load %[[VAL_21]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_27:.*]] = fir.load %[[VAL_12]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_28:.*]] = arith.addi %[[VAL_26]], %[[VAL_27]] : i32
+! CHECK:                 hlfir.assign %[[VAL_28]] to %[[VAL_21]]#0 : i32, !fir.ref<i32>
+! CHECK:                 %[[VAL_29:.*]] = fir.load %[[VAL_22]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_30:.*]] = fir.load %[[VAL_12]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_31:.*]] = arith.addi %[[VAL_29]], %[[VAL_30]] : i32
+! CHECK:                 hlfir.assign %[[VAL_31]] to %[[VAL_22]]#0 : i32, !fir.ref<i32>
+! CHECK:                 omp.yield
+! CHECK:               }
+! CHECK:               omp.terminator
 ! CHECK:             }
 ! CHECK:             omp.terminator
 ! CHECK:           }
@@ -287,27 +302,30 @@ end subroutine
 ! CHECK:             %[[VAL_13:.*]] = arith.constant 1 : i32
 ! CHECK:             %[[VAL_14:.*]] = arith.constant 100 : i32
 ! CHECK:             %[[VAL_15:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop reduction(@add_reduction_f32 %[[VAL_3]]#0 -> %[[VAL_16:.*]] : !fir.ref<f32>, @add_reduction_f32 %[[VAL_5]]#0 -> %[[VAL_17:.*]] : !fir.ref<f32>, @add_reduction_f32 %[[VAL_7]]#0 -> %[[VAL_18:.*]] : !fir.ref<f32>)  for  (%[[VAL_19:.*]]) : i32 = (%[[VAL_13]]) to (%[[VAL_14]]) inclusive step (%[[VAL_15]]) {
-! CHECK:               fir.store %[[VAL_19]] to %[[VAL_12]]#1 : !fir.ref<i32>
-! CHECK:               %[[VAL_20:.*]]:2 = hlfir.declare %[[VAL_16]] {uniq_name = "_QFmultiple_real_reductions_same_typeEx"} : (!fir.ref<f32>) -> (!fir.ref<f32>, !fir.ref<f32>)
-! CHECK:               %[[VAL_21:.*]]:2 = hlfir.declare %[[VAL_17]] {uniq_name = "_QFmultiple_real_reductions_same_typeEy"} : (!fir.ref<f32>) -> (!fir.ref<f32>, !fir.ref<f32>)
-! CHECK:               %[[VAL_22:.*]]:2 = hlfir.declare %[[VAL_18]] {uniq_name = "_QFmultiple_real_reductions_same_typeEz"} : (!fir.ref<f32>) -> (!fir.ref<f32>, !fir.ref<f32>)
-! CHECK:               %[[VAL_23:.*]] = fir.load %[[VAL_20]]#0 : !fir.ref<f32>
-! CHECK:               %[[VAL_24:.*]] = fir.load %[[VAL_12]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_25:.*]] = fir.convert %[[VAL_24]] : (i32) -> f32
-! CHECK:               %[[VAL_26:.*]] = arith.addf %[[VAL_23]], %[[VAL_25]] fastmath<contract> : f32
-! CHECK:               hlfir.assign %[[VAL_26]] to %[[VAL_20]]#0 : f32, !fir.ref<f32>
-! CHECK:               %[[VAL_27:.*]] = fir.load %[[VAL_21]]#0 : !fir.ref<f32>
-! CHECK:               %[[VAL_28:.*]] = fir.load %[[VAL_12]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_29:.*]] = fir.convert %[[VAL_28]] : (i32) -> f32
-! CHECK:               %[[VAL_30:.*]] = arith.addf %[[VAL_27]], %[[VAL_29]] fastmath<contract> : f32
-! CHECK:               hlfir.assign %[[VAL_30]] to %[[VAL_21]]#0 : f32, !fir.ref<f32>
-! CHECK:               %[[VAL_31:.*]] = fir.load %[[VAL_22]]#0 : !fir.ref<f32>
-! CHECK:               %[[VAL_32:.*]] = fir.load %[[VAL_12]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_33:.*]] = fir.convert %[[VAL_32]] : (i32) -> f32
-! CHECK:               %[[VAL_34:.*]] = arith.addf %[[VAL_31]], %[[VAL_33]] fastmath<contract> : f32
-! CHECK:               hlfir.assign %[[VAL_34]] to %[[VAL_22]]#0 : f32, !fir.ref<f32>
-! CHECK:               omp.yield
+! CHECK:             omp.wsloop reduction(@add_reduction_f32 %[[VAL_3]]#0 -> %[[VAL_16:.*]] : !fir.ref<f32>, @add_reduction_f32 %[[VAL_5]]#0 -> %[[VAL_17:.*]] : !fir.ref<f32>, @add_reduction_f32 %[[VAL_7]]#0 -> %[[VAL_18:.*]] : !fir.ref<f32>) {
+! CHECK-NEXT:          omp.loop_nest (%[[VAL_19:.*]]) : i32 = (%[[VAL_13]]) to (%[[VAL_14]]) inclusive step (%[[VAL_15]]) {
+! CHECK:                 %[[VAL_20:.*]]:2 = hlfir.declare %[[VAL_16]] {uniq_name = "_QFmultiple_real_reductions_same_typeEx"} : (!fir.ref<f32>) -> (!fir.ref<f32>, !fir.ref<f32>)
+! CHECK:                 %[[VAL_21:.*]]:2 = hlfir.declare %[[VAL_17]] {uniq_name = "_QFmultiple_real_reductions_same_typeEy"} : (!fir.ref<f32>) -> (!fir.ref<f32>, !fir.ref<f32>)
+! CHECK:                 %[[VAL_22:.*]]:2 = hlfir.declare %[[VAL_18]] {uniq_name = "_QFmultiple_real_reductions_same_typeEz"} : (!fir.ref<f32>) -> (!fir.ref<f32>, !fir.ref<f32>)
+! CHECK:                 fir.store %[[VAL_19]] to %[[VAL_12]]#1 : !fir.ref<i32>
+! CHECK:                 %[[VAL_23:.*]] = fir.load %[[VAL_20]]#0 : !fir.ref<f32>
+! CHECK:                 %[[VAL_24:.*]] = fir.load %[[VAL_12]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_25:.*]] = fir.convert %[[VAL_24]] : (i32) -> f32
+! CHECK:                 %[[VAL_26:.*]] = arith.addf %[[VAL_23]], %[[VAL_25]] fastmath<contract> : f32
+! CHECK:                 hlfir.assign %[[VAL_26]] to %[[VAL_20]]#0 : f32, !fir.ref<f32>
+! CHECK:                 %[[VAL_27:.*]] = fir.load %[[VAL_21]]#0 : !fir.ref<f32>
+! CHECK:                 %[[VAL_28:.*]] = fir.load %[[VAL_12]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_29:.*]] = fir.convert %[[VAL_28]] : (i32) -> f32
+! CHECK:                 %[[VAL_30:.*]] = arith.addf %[[VAL_27]], %[[VAL_29]] fastmath<contract> : f32
+! CHECK:                 hlfir.assign %[[VAL_30]] to %[[VAL_21]]#0 : f32, !fir.ref<f32>
+! CHECK:                 %[[VAL_31:.*]] = fir.load %[[VAL_22]]#0 : !fir.ref<f32>
+! CHECK:                 %[[VAL_32:.*]] = fir.load %[[VAL_12]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_33:.*]] = fir.convert %[[VAL_32]] : (i32) -> f32
+! CHECK:                 %[[VAL_34:.*]] = arith.addf %[[VAL_31]], %[[VAL_33]] fastmath<contract> : f32
+! CHECK:                 hlfir.assign %[[VAL_34]] to %[[VAL_22]]#0 : f32, !fir.ref<f32>
+! CHECK:                 omp.yield
+! CHECK:               }
+! CHECK:               omp.terminator
 ! CHECK:             }
 ! CHECK:             omp.terminator
 ! CHECK:           }
@@ -355,32 +373,35 @@ end subroutine
 ! CHECK:             %[[VAL_16:.*]] = arith.constant 1 : i32
 ! CHECK:             %[[VAL_17:.*]] = arith.constant 100 : i32
 ! CHECK:             %[[VAL_18:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop reduction(@add_reduction_i32 %[[VAL_5]]#0 -> %[[VAL_19:.*]] : !fir.ref<i32>, @add_reduction_i64 %[[VAL_7]]#0 -> %[[VAL_20:.*]] : !fir.ref<i64>, @add_reduction_f32 %[[VAL_9]]#0 -> %[[VAL_21:.*]] : !fir.ref<f32>, @add_reduction_f64 %[[VAL_3]]#0 -> %[[VAL_22:.*]] : !fir.ref<f64>)  for  (%[[VAL_23:.*]]) : i32 = (%[[VAL_16]]) to (%[[VAL_17]]) inclusive step (%[[VAL_18]]) {
-! CHECK:               fir.store %[[VAL_23]] to %[[VAL_15]]#1 : !fir.ref<i32>
-! CHECK:               %[[VAL_24:.*]]:2 = hlfir.declare %[[VAL_19]] {uniq_name = "_QFmultiple_reductions_different_typeEx"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
-! CHECK:               %[[VAL_25:.*]]:2 = hlfir.declare %[[VAL_20]] {uniq_name = "_QFmultiple_reductions_different_typeEy"} : (!fir.ref<i64>) -> (!fir.ref<i64>, !fir.ref<i64>)
-! CHECK:               %[[VAL_26:.*]]:2 = hlfir.declare %[[VAL_21]] {uniq_name = "_QFmultiple_reductions_different_typeEz"} : (!fir.ref<f32>) -> (!fir.ref<f32>, !fir.ref<f32>)
-! CHECK:               %[[VAL_27:.*]]:2 = hlfir.declare %[[VAL_22]] {uniq_name = "_QFmultiple_reductions_different_typeEw"} : (!fir.ref<f64>) -> (!fir.ref<f64>, !fir.ref<f64>)
-! CHECK:               %[[VAL_28:.*]] = fir.load %[[VAL_24]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_29:.*]] = fir.load %[[VAL_15]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_30:.*]] = arith.addi %[[VAL_28]], %[[VAL_29]] : i32
-! CHECK:               hlfir.assign %[[VAL_30]] to %[[VAL_24]]#0 : i32, !fir.ref<i32>
-! CHECK:               %[[VAL_31:.*]] = fir.load %[[VAL_25]]#0 : !fir.ref<i64>
-! CHECK:               %[[VAL_32:.*]] = fir.load %[[VAL_15]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_33:.*]] = fir.convert %[[VAL_32]] : (i32) -> i64
-! CHECK:               %[[VAL_34:.*]] = arith.addi %[[VAL_31]], %[[VAL_33]] : i64
-! CHECK:               hlfir.assign %[[VAL_34]] to %[[VAL_25]]#0 : i64, !fir.ref<i64>
-! CHECK:               %[[VAL_35:.*]] = fir.load %[[VAL_26]]#0 : !fir.ref<f32>
-! CHECK:               %[[VAL_36:.*]] = fir.load %[[VAL_15]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_37:.*]] = fir.convert %[[VAL_36]] : (i32) -> f32
-! CHECK:               %[[VAL_38:.*]] = arith.addf %[[VAL_35]], %[[VAL_37]] fastmath<contract> : f32
-! CHECK:               hlfir.assign %[[VAL_38]] to %[[VAL_26]]#0 : f32, !fir.ref<f32>
-! CHECK:               %[[VAL_39:.*]] = fir.load %[[VAL_27]]#0 : !fir.ref<f64>
-! CHECK:               %[[VAL_40:.*]] = fir.load %[[VAL_15]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_41:.*]] = fir.convert %[[VAL_40]] : (i32) -> f64
-! CHECK:               %[[VAL_42:.*]] = arith.addf %[[VAL_39]], %[[VAL_41]] fastmath<contract> : f64
-! CHECK:               hlfir.assign %[[VAL_42]] to %[[VAL_27]]#0 : f64, !fir.ref<f64>
-! CHECK:               omp.yield
+! CHECK:             omp.wsloop reduction(@add_reduction_i32 %[[VAL_5]]#0 -> %[[VAL_19:.*]] : !fir.ref<i32>, @add_reduction_i64 %[[VAL_7]]#0 -> %[[VAL_20:.*]] : !fir.ref<i64>, @add_reduction_f32 %[[VAL_9]]#0 -> %[[VAL_21:.*]] : !fir.ref<f32>, @add_reduction_f64 %[[VAL_3]]#0 -> %[[VAL_22:.*]] : !fir.ref<f64>) {
+! CHECK-NEXT:          omp.loop_nest (%[[VAL_23:.*]]) : i32 = (%[[VAL_16]]) to (%[[VAL_17]]) inclusive step (%[[VAL_18]]) {
+! CHECK:                 %[[VAL_24:.*]]:2 = hlfir.declare %[[VAL_19]] {uniq_name = "_QFmultiple_reductions_different_typeEx"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
+! CHECK:                 %[[VAL_25:.*]]:2 = hlfir.declare %[[VAL_20]] {uniq_name = "_QFmultiple_reductions_different_typeEy"} : (!fir.ref<i64>) -> (!fir.ref<i64>, !fir.ref<i64>)
+! CHECK:                 %[[VAL_26:.*]]:2 = hlfir.declare %[[VAL_21]] {uniq_name = "_QFmultiple_reductions_different_typeEz"} : (!fir.ref<f32>) -> (!fir.ref<f32>, !fir.ref<f32>)
+! CHECK:                 %[[VAL_27:.*]]:2 = hlfir.declare %[[VAL_22]] {uniq_name = "_QFmultiple_reductions_different_typeEw"} : (!fir.ref<f64>) -> (!fir.ref<f64>, !fir.ref<f64>)
+! CHECK:                 fir.store %[[VAL_23]] to %[[VAL_15]]#1 : !fir.ref<i32>
+! CHECK:                 %[[VAL_28:.*]] = fir.load %[[VAL_24]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_29:.*]] = fir.load %[[VAL_15]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_30:.*]] = arith.addi %[[VAL_28]], %[[VAL_29]] : i32
+! CHECK:                 hlfir.assign %[[VAL_30]] to %[[VAL_24]]#0 : i32, !fir.ref<i32>
+! CHECK:                 %[[VAL_31:.*]] = fir.load %[[VAL_25]]#0 : !fir.ref<i64>
+! CHECK:                 %[[VAL_32:.*]] = fir.load %[[VAL_15]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_33:.*]] = fir.convert %[[VAL_32]] : (i32) -> i64
+! CHECK:                 %[[VAL_34:.*]] = arith.addi %[[VAL_31]], %[[VAL_33]] : i64
+! CHECK:                 hlfir.assign %[[VAL_34]] to %[[VAL_25]]#0 : i64, !fir.ref<i64>
+! CHECK:                 %[[VAL_35:.*]] = fir.load %[[VAL_26]]#0 : !fir.ref<f32>
+! CHECK:                 %[[VAL_36:.*]] = fir.load %[[VAL_15]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_37:.*]] = fir.convert %[[VAL_36]] : (i32) -> f32
+! CHECK:                 %[[VAL_38:.*]] = arith.addf %[[VAL_35]], %[[VAL_37]] fastmath<contract> : f32
+! CHECK:                 hlfir.assign %[[VAL_38]] to %[[VAL_26]]#0 : f32, !fir.ref<f32>
+! CHECK:                 %[[VAL_39:.*]] = fir.load %[[VAL_27]]#0 : !fir.ref<f64>
+! CHECK:                 %[[VAL_40:.*]] = fir.load %[[VAL_15]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_41:.*]] = fir.convert %[[VAL_40]] : (i32) -> f64
+! CHECK:                 %[[VAL_42:.*]] = arith.addf %[[VAL_39]], %[[VAL_41]] fastmath<contract> : f64
+! CHECK:                 hlfir.assign %[[VAL_42]] to %[[VAL_27]]#0 : f64, !fir.ref<f64>
+! CHECK:                 omp.yield
+! CHECK:               }
+! CHECK:               omp.terminator
 ! CHECK:             }
 ! CHECK:             omp.terminator
 ! CHECK:           }
diff --git a/flang/test/Lower/OpenMP/wsloop-reduction-allocatable.f90 b/flang/test/Lower/OpenMP/wsloop-reduction-allocatable.f90
new file mode 100644
index 000000000000..8310aeb23df0
--- /dev/null
+++ b/flang/test/Lower/OpenMP/wsloop-reduction-allocatable.f90
@@ -0,0 +1,97 @@
+! RUN: bbc -emit-hlfir -fopenmp -o - %s | FileCheck %s
+! RUN: %flang_fc1 -emit-hlfir -fopenmp -o - %s | FileCheck %s
+
+program reduce
+integer :: i = 0
+integer, allocatable :: r
+
+allocate(r)
+r = 0
+
+!$omp parallel do reduction(+:r)
+do i=0,10
+  r = i
+enddo
+!$omp end parallel do
+
+print *,r
+
+end program
+
+! CHECK:         omp.declare_reduction @add_reduction_byref_box_heap_i32 : !fir.ref<!fir.box<!fir.heap<i32>>> init {
+! CHECK:         ^bb0(%[[VAL_0:.*]]: !fir.ref<!fir.box<!fir.heap<i32>>>):
+! CHECK:           %[[VAL_1:.*]] = arith.constant 0 : i32
+! CHECK:           %[[LOAD:.*]] = fir.load %[[VAL_0]] : !fir.ref<!fir.box<!fir.heap<i32>>>
+! CHECK:           %[[VAL_2:.*]] = fir.alloca !fir.box<!fir.heap<i32>>
+! CHECK:           %[[ADDR:.*]] = fir.box_addr %[[LOAD]] : (!fir.box<!fir.heap<i32>>) -> !fir.heap<i32>
+! CHECK:           %[[ADDRI:.*]] = fir.convert %[[ADDR]] : (!fir.heap<i32>) -> i64
+! CHECK:           %[[C0_I64:.*]] = arith.constant 0 : i64
+! CHECK:           %[[IS_NULL:.*]] = arith.cmpi eq, %[[ADDRI]], %[[C0_I64]] : i64
+! CHECK:           fir.if %[[IS_NULL]] {
+! CHECK:             %[[NULL_BOX:.*]] = fir.embox %[[ADDR]] : (!fir.heap<i32>) -> !fir.box<!fir.heap<i32>>
+! CHECK:             fir.store %[[NULL_BOX]] to %[[VAL_2]] : !fir.ref<!fir.box<!fir.heap<i32>>
+! CHECK:           } else {
+! CHECK:             %[[VAL_3:.*]] = fir.allocmem i32
+! CHECK:             fir.store %[[VAL_1]] to %[[VAL_3]] : !fir.heap<i32>
+! CHECK:             %[[VAL_4:.*]] = fir.embox %[[VAL_3]] : (!fir.heap<i32>) -> !fir.box<!fir.heap<i32>>
+! CHECK:             fir.store %[[VAL_4]] to %[[VAL_2]] : !fir.ref<!fir.box<!fir.heap<i32>>>
+! CHECK:           }
+! CHECK:           omp.yield(%[[VAL_2]] : !fir.ref<!fir.box<!fir.heap<i32>>>)
+! CHECK:         } combiner {
+! CHECK:         ^bb0(%[[VAL_0:.*]]: !fir.ref<!fir.box<!fir.heap<i32>>>, %[[VAL_1:.*]]: !fir.ref<!fir.box<!fir.heap<i32>>>):
+! CHECK:           %[[VAL_2:.*]] = fir.load %[[VAL_0]] : !fir.ref<!fir.box<!fir.heap<i32>>>
+! CHECK:           %[[VAL_3:.*]] = fir.load %[[VAL_1]] : !fir.ref<!fir.box<!fir.heap<i32>>>
+! CHECK:           %[[VAL_4:.*]] = fir.box_addr %[[VAL_2]] : (!fir.box<!fir.heap<i32>>) -> !fir.heap<i32>
+! CHECK:           %[[VAL_5:.*]] = fir.box_addr %[[VAL_3]] : (!fir.box<!fir.heap<i32>>) -> !fir.heap<i32>
+! CHECK:           %[[VAL_6:.*]] = fir.load %[[VAL_4]] : !fir.heap<i32>
+! CHECK:           %[[VAL_7:.*]] = fir.load %[[VAL_5]] : !fir.heap<i32>
+! CHECK:           %[[VAL_8:.*]] = arith.addi %[[VAL_6]], %[[VAL_7]] : i32
+! CHECK:           fir.store %[[VAL_8]] to %[[VAL_4]] : !fir.heap<i32>
+! CHECK:           omp.yield(%[[VAL_0]] : !fir.ref<!fir.box<!fir.heap<i32>>>)
+! CHECK:         }  cleanup {
+! CHECK:         ^bb0(%[[VAL_0:.*]]: !fir.ref<!fir.box<!fir.heap<i32>>>):
+! CHECK:           %[[VAL_1:.*]] = fir.load %[[VAL_0]] : !fir.ref<!fir.box<!fir.heap<i32>>>
+! CHECK:           %[[VAL_2:.*]] = fir.box_addr %[[VAL_1]] : (!fir.box<!fir.heap<i32>>) -> !fir.heap<i32>
+! CHECK:           %[[VAL_3:.*]] = fir.convert %[[VAL_2]] : (!fir.heap<i32>) -> i64
+! CHECK:           %[[VAL_4:.*]] = arith.constant 0 : i64
+! CHECK:           %[[VAL_5:.*]] = arith.cmpi ne, %[[VAL_3]], %[[VAL_4]] : i64
+! CHECK:           fir.if %[[VAL_5]] {
+! CHECK:             fir.freemem %[[VAL_2]] : !fir.heap<i32>
+! CHECK:           }
+! CHECK:           omp.yield
+! CHECK:         }
+
+! CHECK-LABEL:   func.func @_QQmain() attributes {fir.bindc_name = "reduce"} {
+! CHECK:           %[[VAL_0:.*]] = fir.address_of(@_QFEi) : !fir.ref<i32>
+! CHECK:           %[[VAL_1:.*]]:2 = hlfir.declare %[[VAL_0]] {uniq_name = "_QFEi"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
+! CHECK:           %[[VAL_2:.*]] = fir.alloca !fir.box<!fir.heap<i32>> {bindc_name = "r", uniq_name = "_QFEr"}
+! CHECK:           %[[VAL_3:.*]] = fir.zero_bits !fir.heap<i32>
+! CHECK:           %[[VAL_4:.*]] = fir.embox %[[VAL_3]] : (!fir.heap<i32>) -> !fir.box<!fir.heap<i32>>
+! CHECK:           fir.store %[[VAL_4]] to %[[VAL_2]] : !fir.ref<!fir.box<!fir.heap<i32>>>
+! CHECK:           %[[VAL_5:.*]]:2 = hlfir.declare %[[VAL_2]] {fortran_attrs = {{.*}}<allocatable>, uniq_name = "_QFEr"} : (!fir.ref<!fir.box<!fir.heap<i32>>>) -> (!fir.ref<!fir.box<!fir.heap<i32>>>, !fir.ref<!fir.box<!fir.heap<i32>>>)
+! CHECK:           %[[VAL_6:.*]] = fir.allocmem i32 {fir.must_be_heap = true, uniq_name = "_QFEr.alloc"}
+! CHECK:           %[[VAL_7:.*]] = fir.embox %[[VAL_6]] : (!fir.heap<i32>) -> !fir.box<!fir.heap<i32>>
+! CHECK:           fir.store %[[VAL_7]] to %[[VAL_5]]#1 : !fir.ref<!fir.box<!fir.heap<i32>>>
+! CHECK:           %[[VAL_8:.*]] = arith.constant 0 : i32
+! CHECK:           hlfir.assign %[[VAL_8]] to %[[VAL_5]]#0 realloc : i32, !fir.ref<!fir.box<!fir.heap<i32>>>
+! CHECK:           omp.parallel {
+! CHECK:             %[[VAL_9:.*]] = fir.alloca i32 {adapt.valuebyref, pinned}
+! CHECK:             %[[VAL_10:.*]]:2 = hlfir.declare %[[VAL_9]] {uniq_name = "_QFEi"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
+! CHECK:             %[[VAL_11:.*]] = arith.constant 0 : i32
+! CHECK:             %[[VAL_12:.*]] = arith.constant 10 : i32
+! CHECK:             %[[VAL_13:.*]] = arith.constant 1 : i32
+! CHECK:             omp.wsloop byref reduction(@add_reduction_byref_box_heap_i32 %[[VAL_5]]#0 -> %[[VAL_14:.*]] : !fir.ref<!fir.box<!fir.heap<i32>>>) {
+! CHECK-NEXT:          omp.loop_nest (%[[VAL_15:.*]]) : i32 = (%[[VAL_11]]) to (%[[VAL_12]]) inclusive step (%[[VAL_13]]) {
+! CHECK:                 %[[VAL_16:.*]]:2 = hlfir.declare %[[VAL_14]] {fortran_attrs = {{.*}}<allocatable>, uniq_name = "_QFEr"} : (!fir.ref<!fir.box<!fir.heap<i32>>>) -> (!fir.ref<!fir.box<!fir.heap<i32>>>, !fir.ref<!fir.box<!fir.heap<i32>>>)
+! CHECK:                 fir.store %[[VAL_15]] to %[[VAL_10]]#1 : !fir.ref<i32>
+! CHECK:                 %[[VAL_17:.*]] = fir.load %[[VAL_10]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_18:.*]] = fir.load %[[VAL_16]]#0 : !fir.ref<!fir.box<!fir.heap<i32>>>
+! CHECK:                 %[[VAL_19:.*]] = fir.box_addr %[[VAL_18]] : (!fir.box<!fir.heap<i32>>) -> !fir.heap<i32>
+! CHECK:                 hlfir.assign %[[VAL_17]] to %[[VAL_19]] : i32, !fir.heap<i32>
+! CHECK:                 omp.yield
+! CHECK:               }
+! CHECK:               omp.terminator
+! CHECK:             }
+! CHECK:             omp.terminator
+! CHECK:           }
+
diff --git a/flang/test/Lower/OpenMP/wsloop-reduction-array-assumed-shape.f90 b/flang/test/Lower/OpenMP/wsloop-reduction-array-assumed-shape.f90
index 8f83a30c9fe7..6c9bc75b81d7 100644
--- a/flang/test/Lower/OpenMP/wsloop-reduction-array-assumed-shape.f90
+++ b/flang/test/Lower/OpenMP/wsloop-reduction-array-assumed-shape.f90
@@ -26,17 +26,20 @@ end program
 ! CHECK:         ^bb0(%[[VAL_0:.*]]: !fir.ref<!fir.box<!fir.array<?xf64>>>):
 ! CHECK:           %[[VAL_1:.*]] = arith.constant 0.000000e+00 : f64
 ! CHECK:           %[[VAL_2:.*]] = fir.load %[[VAL_0]] : !fir.ref<!fir.box<!fir.array<?xf64>>>
+! CHECK:           %[[VAL_8:.*]] = fir.alloca !fir.box<!fir.array<?xf64>>
 ! CHECK:           %[[VAL_3:.*]] = arith.constant 0 : index
 ! CHECK:           %[[VAL_4:.*]]:3 = fir.box_dims %[[VAL_2]], %[[VAL_3]] : (!fir.box<!fir.array<?xf64>>, index) -> (index, index, index)
 ! CHECK:           %[[VAL_5:.*]] = fir.shape %[[VAL_4]]#1 : (index) -> !fir.shape<1>
 ! CHECK:           %[[VAL_6:.*]] = fir.allocmem !fir.array<?xf64>, %[[VAL_4]]#1 {bindc_name = ".tmp", uniq_name = ""}
 ! CHECK:           %[[TRUE:.*]]  = arith.constant true
 ! CHECK:           %[[VAL_7:.*]]:2 = hlfir.declare %[[VAL_6]](%[[VAL_5]]) {uniq_name = ".tmp"} : (!fir.heap<!fir.array<?xf64>>, !fir.shape<1>) -> (!fir.box<!fir.array<?xf64>>, !fir.heap<!fir.array<?xf64>>)
-! CHECK:           hlfir.assign %[[VAL_1]] to %[[VAL_7]]#0 : f64, !fir.box<!fir.array<?xf64>>
-! CHECK:           %[[VAL_8:.*]] = fir.alloca !fir.box<!fir.array<?xf64>>
-! CHECK:           fir.store %[[VAL_7]]#0 to %[[VAL_8]] : !fir.ref<!fir.box<!fir.array<?xf64>>>
+! CHECK:           %[[C0:.*]] = arith.constant 0 : index
+! CHECK:           %[[DIMS:.*]]:3 = fir.box_dims %[[VAL_2]], %[[C0]] : (!fir.box<!fir.array<?xf64>>, index) -> (index, index, index)
+! CHECK:           %[[SHIFT:.*]] = fir.shape_shift %[[DIMS]]#0, %[[DIMS]]#1 : (index, index) -> !fir.shapeshift<1>
+! CHECK:           %[[REBOX:.*]] = fir.rebox %[[VAL_7]]#0(%[[SHIFT]]) : (!fir.box<!fir.array<?xf64>>, !fir.shapeshift<1>) -> !fir.box<!fir.array<?xf64>
+! CHECK:           hlfir.assign %[[VAL_1]] to %[[REBOX]] : f64, !fir.box<!fir.array<?xf64>>
+! CHECK:           fir.store %[[REBOX]] to %[[VAL_8]] : !fir.ref<!fir.box<!fir.array<?xf64>>>
 ! CHECK:           omp.yield(%[[VAL_8]] : !fir.ref<!fir.box<!fir.array<?xf64>>>)
-
 ! CHECK-LABEL:   } combiner {
 ! CHECK:         ^bb0(%[[VAL_0:.*]]: !fir.ref<!fir.box<!fir.array<?xf64>>>, %[[VAL_1:.*]]: !fir.ref<!fir.box<!fir.array<?xf64>>>):
 ! CHECK:           %[[VAL_2:.*]] = fir.load %[[VAL_0]] : !fir.ref<!fir.box<!fir.array<?xf64>>>
@@ -81,21 +84,24 @@ end program
 ! CHECK:             %[[VAL_8:.*]] = arith.constant 1 : i32
 ! CHECK:             %[[VAL_9:.*]] = fir.alloca !fir.box<!fir.array<?xf64>>
 ! CHECK:             fir.store %[[VAL_3]]#1 to %[[VAL_9]] : !fir.ref<!fir.box<!fir.array<?xf64>>>
-! CHECK:             omp.wsloop byref reduction(@add_reduction_byref_box_Uxf64 %[[VAL_9]] -> %[[VAL_10:.*]] : !fir.ref<!fir.box<!fir.array<?xf64>>>)  for  (%[[VAL_11:.*]]) : i32 = (%[[VAL_6]]) to (%[[VAL_7]]) inclusive step (%[[VAL_8]]) {
-! CHECK:               fir.store %[[VAL_11]] to %[[VAL_5]]#1 : !fir.ref<i32>
-! CHECK:               %[[VAL_12:.*]]:2 = hlfir.declare %[[VAL_10]] {fortran_attrs = {{.*}}, uniq_name = "_QFFreduceEr"} : (!fir.ref<!fir.box<!fir.array<?xf64>>>) -> (!fir.ref<!fir.box<!fir.array<?xf64>>>, !fir.ref<!fir.box<!fir.array<?xf64>>>)
-! CHECK:               %[[VAL_13:.*]] = fir.load %[[VAL_5]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_14:.*]] = fir.convert %[[VAL_13]] : (i32) -> f64
-! CHECK:               %[[VAL_15:.*]] = fir.load %[[VAL_12]]#0 : !fir.ref<!fir.box<!fir.array<?xf64>>>
-! CHECK:               %[[VAL_16:.*]] = arith.constant 1 : index
-! CHECK:               %[[VAL_17:.*]] = hlfir.designate %[[VAL_15]] (%[[VAL_16]])  : (!fir.box<!fir.array<?xf64>>, index) -> !fir.ref<f64>
-! CHECK:               hlfir.assign %[[VAL_14]] to %[[VAL_17]] : f64, !fir.ref<f64>
-! CHECK:               %[[VAL_18:.*]] = arith.constant 1.000000e+00 : f64
-! CHECK:               %[[VAL_19:.*]] = fir.load %[[VAL_12]]#0 : !fir.ref<!fir.box<!fir.array<?xf64>>>
-! CHECK:               %[[VAL_20:.*]] = arith.constant 2 : index
-! CHECK:               %[[VAL_21:.*]] = hlfir.designate %[[VAL_19]] (%[[VAL_20]])  : (!fir.box<!fir.array<?xf64>>, index) -> !fir.ref<f64>
-! CHECK:               hlfir.assign %[[VAL_18]] to %[[VAL_21]] : f64, !fir.ref<f64>
-! CHECK:               omp.yield
+! CHECK:             omp.wsloop byref reduction(@add_reduction_byref_box_Uxf64 %[[VAL_9]] -> %[[VAL_10:.*]] : !fir.ref<!fir.box<!fir.array<?xf64>>>) {
+! CHECK-NEXT:          omp.loop_nest (%[[VAL_11:.*]]) : i32 = (%[[VAL_6]]) to (%[[VAL_7]]) inclusive step (%[[VAL_8]]) {
+! CHECK:                 %[[VAL_12:.*]]:2 = hlfir.declare %[[VAL_10]] {fortran_attrs = {{.*}}, uniq_name = "_QFFreduceEr"} : (!fir.ref<!fir.box<!fir.array<?xf64>>>) -> (!fir.ref<!fir.box<!fir.array<?xf64>>>, !fir.ref<!fir.box<!fir.array<?xf64>>>)
+! CHECK:                 fir.store %[[VAL_11]] to %[[VAL_5]]#1 : !fir.ref<i32>
+! CHECK:                 %[[VAL_13:.*]] = fir.load %[[VAL_5]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_14:.*]] = fir.convert %[[VAL_13]] : (i32) -> f64
+! CHECK:                 %[[VAL_15:.*]] = fir.load %[[VAL_12]]#0 : !fir.ref<!fir.box<!fir.array<?xf64>>>
+! CHECK:                 %[[VAL_16:.*]] = arith.constant 1 : index
+! CHECK:                 %[[VAL_17:.*]] = hlfir.designate %[[VAL_15]] (%[[VAL_16]])  : (!fir.box<!fir.array<?xf64>>, index) -> !fir.ref<f64>
+! CHECK:                 hlfir.assign %[[VAL_14]] to %[[VAL_17]] : f64, !fir.ref<f64>
+! CHECK:                 %[[VAL_18:.*]] = arith.constant 1.000000e+00 : f64
+! CHECK:                 %[[VAL_19:.*]] = fir.load %[[VAL_12]]#0 : !fir.ref<!fir.box<!fir.array<?xf64>>>
+! CHECK:                 %[[VAL_20:.*]] = arith.constant 2 : index
+! CHECK:                 %[[VAL_21:.*]] = hlfir.designate %[[VAL_19]] (%[[VAL_20]])  : (!fir.box<!fir.array<?xf64>>, index) -> !fir.ref<f64>
+! CHECK:                 hlfir.assign %[[VAL_18]] to %[[VAL_21]] : f64, !fir.ref<f64>
+! CHECK:                 omp.yield
+! CHECK:               }
+! CHECK:               omp.terminator
 ! CHECK:             }
 ! CHECK:             omp.terminator
 ! CHECK:           }
diff --git a/flang/test/Lower/OpenMP/wsloop-reduction-array.f90 b/flang/test/Lower/OpenMP/wsloop-reduction-array.f90
index a08bca9eb283..7aecf870cf80 100644
--- a/flang/test/Lower/OpenMP/wsloop-reduction-array.f90
+++ b/flang/test/Lower/OpenMP/wsloop-reduction-array.f90
@@ -18,14 +18,17 @@ end program
 ! CHECK:         ^bb0(%[[VAL_0:.*]]: !fir.ref<!fir.box<!fir.array<2xi32>>>):
 ! CHECK:           %[[VAL_2:.*]] = arith.constant 0 : i32
 ! CHECK:           %[[VAL_3:.*]] = fir.load %[[VAL_0]] : !fir.ref<!fir.box<!fir.array<2xi32>>>
+! CHECK:           %[[VAL_8:.*]] = fir.alloca !fir.box<!fir.array<2xi32>>
 ! CHECK:           %[[VAL_4:.*]] = arith.constant 2 : index
 ! CHECK:           %[[VAL_5:.*]] = fir.shape %[[VAL_4]] : (index) -> !fir.shape<1>
 ! CHECK:           %[[VAL_1:.*]] = fir.allocmem !fir.array<2xi32> {bindc_name = ".tmp", uniq_name = ""}
 ! CHECK:           %[[TRUE:.*]]  = arith.constant true
 ! CHECK:           %[[VAL_6:.*]]:2 = hlfir.declare %[[VAL_1]](%[[VAL_5]]) {uniq_name = ".tmp"} : (!fir.heap<!fir.array<2xi32>>, !fir.shape<1>) -> (!fir.heap<!fir.array<2xi32>>, !fir.heap<!fir.array<2xi32>>)
-! CHECK:           %[[VAL_7:.*]] = fir.embox %[[VAL_6]]#0(%[[VAL_5]]) : (!fir.heap<!fir.array<2xi32>>, !fir.shape<1>) -> !fir.box<!fir.array<2xi32>>
+! CHECK:           %[[C0:.*]] = arith.constant 0 : index
+! CHECK:           %[[DIMS:.*]]:3 = fir.box_dims %[[VAL_3]], %[[C0]] : (!fir.box<!fir.array<2xi32>>, index) -> (index, index, index)
+! CHECK:           %[[SHIFT:.*]] = fir.shape_shift %[[DIMS]]#0, %[[DIMS]]#1 : (index, index) -> !fir.shapeshift<1>
+! CHECK:           %[[VAL_7:.*]] = fir.embox %[[VAL_6]]#0(%[[SHIFT]]) : (!fir.heap<!fir.array<2xi32>>, !fir.shapeshift<1>) -> !fir.box<!fir.array<2xi32>>
 ! CHECK:           hlfir.assign %[[VAL_2]] to %[[VAL_7]] : i32, !fir.box<!fir.array<2xi32>>
-! CHECK:           %[[VAL_8:.*]] = fir.alloca !fir.box<!fir.array<2xi32>>
 ! CHECK:           fir.store %[[VAL_7]] to %[[VAL_8]] : !fir.ref<!fir.box<!fir.array<2xi32>>>
 ! CHECK:           omp.yield(%[[VAL_8]] : !fir.ref<!fir.box<!fir.array<2xi32>>>)
 
@@ -76,22 +79,25 @@ end program
 ! CHECK:             %[[VAL_11:.*]] = fir.embox %[[VAL_5]]#0(%[[VAL_4]]) : (!fir.ref<!fir.array<2xi32>>, !fir.shape<1>) -> !fir.box<!fir.array<2xi32>>
 ! CHECK:             %[[VAL_12:.*]] = fir.alloca !fir.box<!fir.array<2xi32>>
 ! CHECK:             fir.store %[[VAL_11]] to %[[VAL_12]] : !fir.ref<!fir.box<!fir.array<2xi32>>>
-! CHECK:             omp.wsloop byref reduction(@add_reduction_byref_box_2xi32 %[[VAL_12]] -> %[[VAL_13:.*]] : !fir.ref<!fir.box<!fir.array<2xi32>>>)  for  (%[[VAL_14:.*]]) : i32 = (%[[VAL_8]]) to (%[[VAL_9]]) inclusive step (%[[VAL_10]]) {
-! CHECK:               fir.store %[[VAL_14]] to %[[VAL_7]]#1 : !fir.ref<i32>
-! CHECK:               %[[VAL_15:.*]]:2 = hlfir.declare %[[VAL_13]] {uniq_name = "_QFEr"} : (!fir.ref<!fir.box<!fir.array<2xi32>>>) -> (!fir.ref<!fir.box<!fir.array<2xi32>>>, !fir.ref<!fir.box<!fir.array<2xi32>>>)
-! CHECK:               %[[VAL_16:.*]] = fir.load %[[VAL_7]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_17:.*]] = fir.load %[[VAL_15]]#0 : !fir.ref<!fir.box<!fir.array<2xi32>>>
-! CHECK:               %[[VAL_18:.*]] = arith.constant 1 : index
-! CHECK:               %[[VAL_19:.*]] = hlfir.designate %[[VAL_17]] (%[[VAL_18]])  : (!fir.box<!fir.array<2xi32>>, index) -> !fir.ref<i32>
-! CHECK:               hlfir.assign %[[VAL_16]] to %[[VAL_19]] : i32, !fir.ref<i32>
-! CHECK:               %[[VAL_20:.*]] = fir.load %[[VAL_7]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_21:.*]] = arith.constant 0 : i32
-! CHECK:               %[[VAL_22:.*]] = arith.subi %[[VAL_21]], %[[VAL_20]] : i32
-! CHECK:               %[[VAL_23:.*]] = fir.load %[[VAL_15]]#0 : !fir.ref<!fir.box<!fir.array<2xi32>>>
-! CHECK:               %[[VAL_24:.*]] = arith.constant 2 : index
-! CHECK:               %[[VAL_25:.*]] = hlfir.designate %[[VAL_23]] (%[[VAL_24]])  : (!fir.box<!fir.array<2xi32>>, index) -> !fir.ref<i32>
-! CHECK:               hlfir.assign %[[VAL_22]] to %[[VAL_25]] : i32, !fir.ref<i32>
-! CHECK:               omp.yield
+! CHECK:             omp.wsloop byref reduction(@add_reduction_byref_box_2xi32 %[[VAL_12]] -> %[[VAL_13:.*]] : !fir.ref<!fir.box<!fir.array<2xi32>>>) {
+! CHECK-NEXT:          omp.loop_nest (%[[VAL_14:.*]]) : i32 = (%[[VAL_8]]) to (%[[VAL_9]]) inclusive step (%[[VAL_10]]) {
+! CHECK:                 %[[VAL_15:.*]]:2 = hlfir.declare %[[VAL_13]] {uniq_name = "_QFEr"} : (!fir.ref<!fir.box<!fir.array<2xi32>>>) -> (!fir.ref<!fir.box<!fir.array<2xi32>>>, !fir.ref<!fir.box<!fir.array<2xi32>>>)
+! CHECK:                 fir.store %[[VAL_14]] to %[[VAL_7]]#1 : !fir.ref<i32>
+! CHECK:                 %[[VAL_16:.*]] = fir.load %[[VAL_7]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_17:.*]] = fir.load %[[VAL_15]]#0 : !fir.ref<!fir.box<!fir.array<2xi32>>>
+! CHECK:                 %[[VAL_18:.*]] = arith.constant 1 : index
+! CHECK:                 %[[VAL_19:.*]] = hlfir.designate %[[VAL_17]] (%[[VAL_18]])  : (!fir.box<!fir.array<2xi32>>, index) -> !fir.ref<i32>
+! CHECK:                 hlfir.assign %[[VAL_16]] to %[[VAL_19]] : i32, !fir.ref<i32>
+! CHECK:                 %[[VAL_20:.*]] = fir.load %[[VAL_7]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_21:.*]] = arith.constant 0 : i32
+! CHECK:                 %[[VAL_22:.*]] = arith.subi %[[VAL_21]], %[[VAL_20]] : i32
+! CHECK:                 %[[VAL_23:.*]] = fir.load %[[VAL_15]]#0 : !fir.ref<!fir.box<!fir.array<2xi32>>>
+! CHECK:                 %[[VAL_24:.*]] = arith.constant 2 : index
+! CHECK:                 %[[VAL_25:.*]] = hlfir.designate %[[VAL_23]] (%[[VAL_24]])  : (!fir.box<!fir.array<2xi32>>, index) -> !fir.ref<i32>
+! CHECK:                 hlfir.assign %[[VAL_22]] to %[[VAL_25]] : i32, !fir.ref<i32>
+! CHECK:                 omp.yield
+! CHECK:               }
+! CHECK:               omp.terminator
 ! CHECK:             }
 ! CHECK:             omp.terminator
 ! CHECK:           }
diff --git a/flang/test/Lower/OpenMP/wsloop-reduction-array2.f90 b/flang/test/Lower/OpenMP/wsloop-reduction-array2.f90
index 045208d6f7ff..d1d8a2c59944 100644
--- a/flang/test/Lower/OpenMP/wsloop-reduction-array2.f90
+++ b/flang/test/Lower/OpenMP/wsloop-reduction-array2.f90
@@ -18,14 +18,17 @@ end program
 ! CHECK:         ^bb0(%[[VAL_0:.*]]: !fir.ref<!fir.box<!fir.array<2xi32>>>):
 ! CHECK:           %[[VAL_2:.*]] = arith.constant 0 : i32
 ! CHECK:           %[[VAL_3:.*]] = fir.load %[[VAL_0]] : !fir.ref<!fir.box<!fir.array<2xi32>>>
+! CHECK:           %[[VAL_8:.*]] = fir.alloca !fir.box<!fir.array<2xi32>>
 ! CHECK:           %[[VAL_4:.*]] = arith.constant 2 : index
 ! CHECK:           %[[VAL_5:.*]] = fir.shape %[[VAL_4]] : (index) -> !fir.shape<1>
 ! CHECK:           %[[VAL_1:.*]] = fir.allocmem !fir.array<2xi32> {bindc_name = ".tmp", uniq_name = ""}
 ! CHECK:           %[[TRUE:.*]]  = arith.constant true
 ! CHECK:           %[[VAL_6:.*]]:2 = hlfir.declare %[[VAL_1]](%[[VAL_5]]) {uniq_name = ".tmp"} : (!fir.heap<!fir.array<2xi32>>, !fir.shape<1>) -> (!fir.heap<!fir.array<2xi32>>, !fir.heap<!fir.array<2xi32>>)
-! CHECK:           %[[VAL_7:.*]] = fir.embox %[[VAL_6]]#0(%[[VAL_5]]) : (!fir.heap<!fir.array<2xi32>>, !fir.shape<1>) -> !fir.box<!fir.array<2xi32>>
+! CHECK:           %[[C0:.*]] = arith.constant 0 : index
+! CHECK:           %[[DIMS:.*]]:3 = fir.box_dims %[[VAL_3]], %[[C0]] : (!fir.box<!fir.array<2xi32>>, index) -> (index, index, index)
+! CHECK:           %[[SHIFT:.*]] = fir.shape_shift %[[DIMS]]#0, %[[DIMS]]#1 : (index, index) -> !fir.shapeshift<1>
+! CHECK:           %[[VAL_7:.*]] = fir.embox %[[VAL_6]]#0(%[[SHIFT]]) : (!fir.heap<!fir.array<2xi32>>, !fir.shapeshift<1>) -> !fir.box<!fir.array<2xi32>>
 ! CHECK:           hlfir.assign %[[VAL_2]] to %[[VAL_7]] : i32, !fir.box<!fir.array<2xi32>>
-! CHECK:           %[[VAL_8:.*]] = fir.alloca !fir.box<!fir.array<2xi32>>
 ! CHECK:           fir.store %[[VAL_7]] to %[[VAL_8]] : !fir.ref<!fir.box<!fir.array<2xi32>>>
 ! CHECK:           omp.yield(%[[VAL_8]] : !fir.ref<!fir.box<!fir.array<2xi32>>>)
 
@@ -76,30 +79,33 @@ end program
 ! CHECK:             %[[VAL_11:.*]] = fir.embox %[[VAL_5]]#0(%[[VAL_4]]) : (!fir.ref<!fir.array<2xi32>>, !fir.shape<1>) -> !fir.box<!fir.array<2xi32>>
 ! CHECK:             %[[VAL_12:.*]] = fir.alloca !fir.box<!fir.array<2xi32>>
 ! CHECK:             fir.store %[[VAL_11]] to %[[VAL_12]] : !fir.ref<!fir.box<!fir.array<2xi32>>>
-! CHECK:             omp.wsloop byref reduction(@add_reduction_byref_box_2xi32 %[[VAL_12]] -> %[[VAL_13:.*]] : !fir.ref<!fir.box<!fir.array<2xi32>>>)  for  (%[[VAL_14:.*]]) : i32 = (%[[VAL_8]]) to (%[[VAL_9]]) inclusive step (%[[VAL_10]]) {
-! CHECK:               fir.store %[[VAL_14]] to %[[VAL_7]]#1 : !fir.ref<i32>
-! CHECK:               %[[VAL_15:.*]]:2 = hlfir.declare %[[VAL_13]] {uniq_name = "_QFEr"} : (!fir.ref<!fir.box<!fir.array<2xi32>>>) -> (!fir.ref<!fir.box<!fir.array<2xi32>>>, !fir.ref<!fir.box<!fir.array<2xi32>>>)
-! CHECK:               %[[VAL_16:.*]] = fir.load %[[VAL_15]]#0 : !fir.ref<!fir.box<!fir.array<2xi32>>>
-! CHECK:               %[[VAL_17:.*]] = arith.constant 1 : index
-! CHECK:               %[[VAL_18:.*]] = hlfir.designate %[[VAL_16]] (%[[VAL_17]])  : (!fir.box<!fir.array<2xi32>>, index) -> !fir.ref<i32>
-! CHECK:               %[[VAL_19:.*]] = fir.load %[[VAL_18]] : !fir.ref<i32>
-! CHECK:               %[[VAL_20:.*]] = fir.load %[[VAL_7]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_21:.*]] = arith.addi %[[VAL_19]], %[[VAL_20]] : i32
-! CHECK:               %[[VAL_22:.*]] = fir.load %[[VAL_15]]#0 : !fir.ref<!fir.box<!fir.array<2xi32>>>
-! CHECK:               %[[VAL_23:.*]] = arith.constant 1 : index
-! CHECK:               %[[VAL_24:.*]] = hlfir.designate %[[VAL_22]] (%[[VAL_23]])  : (!fir.box<!fir.array<2xi32>>, index) -> !fir.ref<i32>
-! CHECK:               hlfir.assign %[[VAL_21]] to %[[VAL_24]] : i32, !fir.ref<i32>
-! CHECK:               %[[VAL_25:.*]] = fir.load %[[VAL_15]]#0 : !fir.ref<!fir.box<!fir.array<2xi32>>>
-! CHECK:               %[[VAL_26:.*]] = arith.constant 2 : index
-! CHECK:               %[[VAL_27:.*]] = hlfir.designate %[[VAL_25]] (%[[VAL_26]])  : (!fir.box<!fir.array<2xi32>>, index) -> !fir.ref<i32>
-! CHECK:               %[[VAL_28:.*]] = fir.load %[[VAL_27]] : !fir.ref<i32>
-! CHECK:               %[[VAL_29:.*]] = fir.load %[[VAL_7]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_30:.*]] = arith.subi %[[VAL_28]], %[[VAL_29]] : i32
-! CHECK:               %[[VAL_31:.*]] = fir.load %[[VAL_15]]#0 : !fir.ref<!fir.box<!fir.array<2xi32>>>
-! CHECK:               %[[VAL_32:.*]] = arith.constant 2 : index
-! CHECK:               %[[VAL_33:.*]] = hlfir.designate %[[VAL_31]] (%[[VAL_32]])  : (!fir.box<!fir.array<2xi32>>, index) -> !fir.ref<i32>
-! CHECK:               hlfir.assign %[[VAL_30]] to %[[VAL_33]] : i32, !fir.ref<i32>
-! CHECK:               omp.yield
+! CHECK:             omp.wsloop byref reduction(@add_reduction_byref_box_2xi32 %[[VAL_12]] -> %[[VAL_13:.*]] : !fir.ref<!fir.box<!fir.array<2xi32>>>) {
+! CHECK-NEXT:          omp.loop_nest (%[[VAL_14:.*]]) : i32 = (%[[VAL_8]]) to (%[[VAL_9]]) inclusive step (%[[VAL_10]]) {
+! CHECK:                 %[[VAL_15:.*]]:2 = hlfir.declare %[[VAL_13]] {uniq_name = "_QFEr"} : (!fir.ref<!fir.box<!fir.array<2xi32>>>) -> (!fir.ref<!fir.box<!fir.array<2xi32>>>, !fir.ref<!fir.box<!fir.array<2xi32>>>)
+! CHECK:                 fir.store %[[VAL_14]] to %[[VAL_7]]#1 : !fir.ref<i32>
+! CHECK:                 %[[VAL_16:.*]] = fir.load %[[VAL_15]]#0 : !fir.ref<!fir.box<!fir.array<2xi32>>>
+! CHECK:                 %[[VAL_17:.*]] = arith.constant 1 : index
+! CHECK:                 %[[VAL_18:.*]] = hlfir.designate %[[VAL_16]] (%[[VAL_17]])  : (!fir.box<!fir.array<2xi32>>, index) -> !fir.ref<i32>
+! CHECK:                 %[[VAL_19:.*]] = fir.load %[[VAL_18]] : !fir.ref<i32>
+! CHECK:                 %[[VAL_20:.*]] = fir.load %[[VAL_7]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_21:.*]] = arith.addi %[[VAL_19]], %[[VAL_20]] : i32
+! CHECK:                 %[[VAL_22:.*]] = fir.load %[[VAL_15]]#0 : !fir.ref<!fir.box<!fir.array<2xi32>>>
+! CHECK:                 %[[VAL_23:.*]] = arith.constant 1 : index
+! CHECK:                 %[[VAL_24:.*]] = hlfir.designate %[[VAL_22]] (%[[VAL_23]])  : (!fir.box<!fir.array<2xi32>>, index) -> !fir.ref<i32>
+! CHECK:                 hlfir.assign %[[VAL_21]] to %[[VAL_24]] : i32, !fir.ref<i32>
+! CHECK:                 %[[VAL_25:.*]] = fir.load %[[VAL_15]]#0 : !fir.ref<!fir.box<!fir.array<2xi32>>>
+! CHECK:                 %[[VAL_26:.*]] = arith.constant 2 : index
+! CHECK:                 %[[VAL_27:.*]] = hlfir.designate %[[VAL_25]] (%[[VAL_26]])  : (!fir.box<!fir.array<2xi32>>, index) -> !fir.ref<i32>
+! CHECK:                 %[[VAL_28:.*]] = fir.load %[[VAL_27]] : !fir.ref<i32>
+! CHECK:                 %[[VAL_29:.*]] = fir.load %[[VAL_7]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_30:.*]] = arith.subi %[[VAL_28]], %[[VAL_29]] : i32
+! CHECK:                 %[[VAL_31:.*]] = fir.load %[[VAL_15]]#0 : !fir.ref<!fir.box<!fir.array<2xi32>>>
+! CHECK:                 %[[VAL_32:.*]] = arith.constant 2 : index
+! CHECK:                 %[[VAL_33:.*]] = hlfir.designate %[[VAL_31]] (%[[VAL_32]])  : (!fir.box<!fir.array<2xi32>>, index) -> !fir.ref<i32>
+! CHECK:                 hlfir.assign %[[VAL_30]] to %[[VAL_33]] : i32, !fir.ref<i32>
+! CHECK:                 omp.yield
+! CHECK:               }
+! CHECK:               omp.terminator
 ! CHECK:             }
 ! CHECK:             omp.terminator
 ! CHECK:           }
diff --git a/flang/test/Lower/OpenMP/wsloop-reduction-iand-byref.f90 b/flang/test/Lower/OpenMP/wsloop-reduction-iand-byref.f90
index e3f06a446ed4..40280c56dad6 100644
--- a/flang/test/Lower/OpenMP/wsloop-reduction-iand-byref.f90
+++ b/flang/test/Lower/OpenMP/wsloop-reduction-iand-byref.f90
@@ -35,17 +35,19 @@
 ! CHECK:             %[[VAL_9:.*]] = arith.constant 1 : i32
 ! CHECK:             %[[VAL_10:.*]] = arith.constant 100 : i32
 ! CHECK:             %[[VAL_11:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop byref reduction(@iand_byref_i32 %[[VAL_4]]#0 -> %[[VAL_12:.*]] : !fir.ref<i32>)  for  (%[[VAL_13:.*]]) : i32 = (%[[VAL_9]]) to (%[[VAL_10]]) inclusive step (%[[VAL_11]]) {
-! CHECK:               fir.store %[[VAL_13]] to %[[VAL_8]]#1 : !fir.ref<i32>
-! CHECK:               %[[VAL_14:.*]]:2 = hlfir.declare %[[VAL_12]] {uniq_name = "_QFreduction_iandEx"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
-! CHECK:               %[[VAL_15:.*]] = fir.load %[[VAL_8]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_16:.*]] = fir.convert %[[VAL_15]] : (i32) -> i64
-! CHECK:               %[[VAL_17:.*]] = hlfir.designate %[[VAL_5]]#0 (%[[VAL_16]])  : (!fir.box<!fir.array<?xi32>>, i64) -> !fir.ref<i32>
-! CHECK:               %[[VAL_18:.*]] = fir.load %[[VAL_14]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_19:.*]] = fir.load %[[VAL_17]] : !fir.ref<i32>
-! CHECK:               %[[VAL_20:.*]] = arith.andi %[[VAL_18]], %[[VAL_19]] : i32
-! CHECK:               hlfir.assign %[[VAL_20]] to %[[VAL_14]]#0 : i32, !fir.ref<i32>
-! CHECK:               omp.yield
+! CHECK:             omp.wsloop byref reduction(@iand_byref_i32 %[[VAL_4]]#0 -> %[[VAL_12:.*]] : !fir.ref<i32>) {
+! CHECK-NEXT:          omp.loop_nest (%[[VAL_13:.*]]) : i32 = (%[[VAL_9]]) to (%[[VAL_10]]) inclusive step (%[[VAL_11]]) {
+! CHECK:                 %[[VAL_14:.*]]:2 = hlfir.declare %[[VAL_12]] {uniq_name = "_QFreduction_iandEx"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
+! CHECK:                 fir.store %[[VAL_13]] to %[[VAL_8]]#1 : !fir.ref<i32>
+! CHECK:                 %[[VAL_15:.*]] = fir.load %[[VAL_8]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_16:.*]] = fir.convert %[[VAL_15]] : (i32) -> i64
+! CHECK:                 %[[VAL_17:.*]] = hlfir.designate %[[VAL_5]]#0 (%[[VAL_16]])  : (!fir.box<!fir.array<?xi32>>, i64) -> !fir.ref<i32>
+! CHECK:                 %[[VAL_18:.*]] = fir.load %[[VAL_14]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_19:.*]] = fir.load %[[VAL_17]] : !fir.ref<i32>
+! CHECK:                 %[[VAL_20:.*]] = arith.andi %[[VAL_18]], %[[VAL_19]] : i32
+! CHECK:                 hlfir.assign %[[VAL_20]] to %[[VAL_14]]#0 : i32, !fir.ref<i32>
+! CHECK:                 omp.yield
+! CHECK:               omp.terminator
 ! CHECK:             omp.terminator
 
 
diff --git a/flang/test/Lower/OpenMP/wsloop-reduction-iand.f90 b/flang/test/Lower/OpenMP/wsloop-reduction-iand.f90
index 746617e21062..986892d3584f 100644
--- a/flang/test/Lower/OpenMP/wsloop-reduction-iand.f90
+++ b/flang/test/Lower/OpenMP/wsloop-reduction-iand.f90
@@ -29,17 +29,19 @@
 ! CHECK:             %[[VAL_9:.*]] = arith.constant 1 : i32
 ! CHECK:             %[[VAL_10:.*]] = arith.constant 100 : i32
 ! CHECK:             %[[VAL_11:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop reduction(@iand_i32 %[[VAL_4]]#0 -> %[[VAL_12:.*]] : !fir.ref<i32>)  for  (%[[VAL_13:.*]]) : i32 = (%[[VAL_9]]) to (%[[VAL_10]]) inclusive step (%[[VAL_11]]) {
-! CHECK:               fir.store %[[VAL_13]] to %[[VAL_8]]#1 : !fir.ref<i32>
-! CHECK:               %[[VAL_14:.*]]:2 = hlfir.declare %[[VAL_12]] {uniq_name = "_QFreduction_iandEx"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
-! CHECK:               %[[VAL_15:.*]] = fir.load %[[VAL_8]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_16:.*]] = fir.convert %[[VAL_15]] : (i32) -> i64
-! CHECK:               %[[VAL_17:.*]] = hlfir.designate %[[VAL_5]]#0 (%[[VAL_16]])  : (!fir.box<!fir.array<?xi32>>, i64) -> !fir.ref<i32>
-! CHECK:               %[[VAL_18:.*]] = fir.load %[[VAL_14]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_19:.*]] = fir.load %[[VAL_17]] : !fir.ref<i32>
-! CHECK:               %[[VAL_20:.*]] = arith.andi %[[VAL_18]], %[[VAL_19]] : i32
-! CHECK:               hlfir.assign %[[VAL_20]] to %[[VAL_14]]#0 : i32, !fir.ref<i32>
-! CHECK:               omp.yield
+! CHECK:             omp.wsloop reduction(@iand_i32 %[[VAL_4]]#0 -> %[[VAL_12:.*]] : !fir.ref<i32>) {
+! CHECK-NEXT:          omp.loop_nest (%[[VAL_13:.*]]) : i32 = (%[[VAL_9]]) to (%[[VAL_10]]) inclusive step (%[[VAL_11]]) {
+! CHECK:                 %[[VAL_14:.*]]:2 = hlfir.declare %[[VAL_12]] {uniq_name = "_QFreduction_iandEx"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
+! CHECK:                 fir.store %[[VAL_13]] to %[[VAL_8]]#1 : !fir.ref<i32>
+! CHECK:                 %[[VAL_15:.*]] = fir.load %[[VAL_8]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_16:.*]] = fir.convert %[[VAL_15]] : (i32) -> i64
+! CHECK:                 %[[VAL_17:.*]] = hlfir.designate %[[VAL_5]]#0 (%[[VAL_16]])  : (!fir.box<!fir.array<?xi32>>, i64) -> !fir.ref<i32>
+! CHECK:                 %[[VAL_18:.*]] = fir.load %[[VAL_14]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_19:.*]] = fir.load %[[VAL_17]] : !fir.ref<i32>
+! CHECK:                 %[[VAL_20:.*]] = arith.andi %[[VAL_18]], %[[VAL_19]] : i32
+! CHECK:                 hlfir.assign %[[VAL_20]] to %[[VAL_14]]#0 : i32, !fir.ref<i32>
+! CHECK:                 omp.yield
+! CHECK:               omp.terminator
 ! CHECK:             omp.terminator
 
 
diff --git a/flang/test/Lower/OpenMP/wsloop-reduction-ieor-byref.f90 b/flang/test/Lower/OpenMP/wsloop-reduction-ieor-byref.f90
index 7e3a283bf783..ee33ce2f348d 100644
--- a/flang/test/Lower/OpenMP/wsloop-reduction-ieor-byref.f90
+++ b/flang/test/Lower/OpenMP/wsloop-reduction-ieor-byref.f90
@@ -28,9 +28,10 @@
 !CHECK: omp.parallel
 !CHECK: %[[I_REF:.*]] = fir.alloca i32 {adapt.valuebyref, pinned}
 !CHECK: %[[I_DECL:.*]]:2 = hlfir.declare %[[I_REF]] {uniq_name = "_QFreduction_ieorEi"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
-!CHECK: omp.wsloop byref reduction(@ieor_byref_i32 %[[X_DECL]]#0 -> %[[PRV:.+]] : !fir.ref<i32>) for
-!CHECK: fir.store %{{.*}} to %[[I_DECL]]#1 : !fir.ref<i32>
+!CHECK: omp.wsloop byref reduction(@ieor_byref_i32 %[[X_DECL]]#0 -> %[[PRV:.+]] : !fir.ref<i32>)
+!CHECK-NEXT: omp.loop_nest
 !CHECK: %[[PRV_DECL:.+]]:2 = hlfir.declare %[[PRV]] {{.*}} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
+!CHECK: fir.store %{{.*}} to %[[I_DECL]]#1 : !fir.ref<i32>
 !CHECK: %[[I_32:.*]] = fir.load %[[I_DECL]]#0 : !fir.ref<i32>
 !CHECK: %[[I_64:.*]] = fir.convert %[[I_32]] : (i32) -> i64
 !CHECK: %[[Y_I_REF:.*]] = hlfir.designate %[[Y_DECL]]#0 (%[[I_64]])  : (!fir.box<!fir.array<?xi32>>, i64) -> !fir.ref<i32>
@@ -40,6 +41,7 @@
 !CHECK: hlfir.assign %[[RES]] to %[[PRV_DECL]]#0 : i32, !fir.ref<i32>
 !CHECK: omp.yield
 !CHECK: omp.terminator
+!CHECK: omp.terminator
 
 subroutine reduction_ieor(y)
   integer :: x, y(:)
diff --git a/flang/test/Lower/OpenMP/wsloop-reduction-ieor.f90 b/flang/test/Lower/OpenMP/wsloop-reduction-ieor.f90
index 11245c4ac95e..b362731b3371 100644
--- a/flang/test/Lower/OpenMP/wsloop-reduction-ieor.f90
+++ b/flang/test/Lower/OpenMP/wsloop-reduction-ieor.f90
@@ -19,9 +19,10 @@
 !CHECK: omp.parallel
 !CHECK: %[[I_REF:.*]] = fir.alloca i32 {adapt.valuebyref, pinned}
 !CHECK: %[[I_DECL:.*]]:2 = hlfir.declare %[[I_REF]] {uniq_name = "_QFreduction_ieorEi"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
-!CHECK: omp.wsloop reduction(@[[IEOR_DECLARE_I]] %[[X_DECL]]#0 -> %[[PRV:.+]] : !fir.ref<i32>) for
-!CHECK: fir.store %{{.*}} to %[[I_DECL]]#1 : !fir.ref<i32>
+!CHECK: omp.wsloop reduction(@[[IEOR_DECLARE_I]] %[[X_DECL]]#0 -> %[[PRV:.+]] : !fir.ref<i32>)
+!CHECK-NEXT: omp.loop_nest
 !CHECK: %[[PRV_DECL:.+]]:2 = hlfir.declare %[[PRV]] {{.*}} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
+!CHECK: fir.store %{{.*}} to %[[I_DECL]]#1 : !fir.ref<i32>
 !CHECK: %[[I_32:.*]] = fir.load %[[I_DECL]]#0 : !fir.ref<i32>
 !CHECK: %[[I_64:.*]] = fir.convert %[[I_32]] : (i32) -> i64
 !CHECK: %[[Y_I_REF:.*]] = hlfir.designate %[[Y_DECL]]#0 (%[[I_64]])  : (!fir.box<!fir.array<?xi32>>, i64) -> !fir.ref<i32>
@@ -31,6 +32,7 @@
 !CHECK: hlfir.assign %[[RES]] to %[[PRV_DECL]]#0 : i32, !fir.ref<i32>
 !CHECK: omp.yield
 !CHECK: omp.terminator
+!CHECK: omp.terminator
 
 subroutine reduction_ieor(y)
   integer :: x, y(:)
diff --git a/flang/test/Lower/OpenMP/wsloop-reduction-ior-byref.f90 b/flang/test/Lower/OpenMP/wsloop-reduction-ior-byref.f90
index c7f8e8bdede5..0052773bb5ad 100644
--- a/flang/test/Lower/OpenMP/wsloop-reduction-ior-byref.f90
+++ b/flang/test/Lower/OpenMP/wsloop-reduction-ior-byref.f90
@@ -33,17 +33,19 @@
 ! CHECK:             %[[VAL_9:.*]] = arith.constant 1 : i32
 ! CHECK:             %[[VAL_10:.*]] = arith.constant 100 : i32
 ! CHECK:             %[[VAL_11:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop byref reduction(@ior_byref_i32 %[[VAL_4]]#0 -> %[[VAL_12:.*]] : !fir.ref<i32>)  for  (%[[VAL_13:.*]]) : i32 = (%[[VAL_9]]) to (%[[VAL_10]]) inclusive step (%[[VAL_11]])
-! CHECK:               fir.store %[[VAL_13]] to %[[VAL_8]]#1 : !fir.ref<i32>
-! CHECK:               %[[VAL_14:.*]]:2 = hlfir.declare %[[VAL_12]] {uniq_name = "_QFreduction_iorEx"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
-! CHECK:               %[[VAL_15:.*]] = fir.load %[[VAL_8]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_16:.*]] = fir.convert %[[VAL_15]] : (i32) -> i64
-! CHECK:               %[[VAL_17:.*]] = hlfir.designate %[[VAL_5]]#0 (%[[VAL_16]])  : (!fir.box<!fir.array<?xi32>>, i64) -> !fir.ref<i32>
-! CHECK:               %[[VAL_18:.*]] = fir.load %[[VAL_14]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_19:.*]] = fir.load %[[VAL_17]] : !fir.ref<i32>
-! CHECK:               %[[VAL_20:.*]] = arith.ori %[[VAL_18]], %[[VAL_19]] : i32
-! CHECK:               hlfir.assign %[[VAL_20]] to %[[VAL_14]]#0 : i32, !fir.ref<i32>
-! CHECK:               omp.yield
+! CHECK:             omp.wsloop byref reduction(@ior_byref_i32 %[[VAL_4]]#0 -> %[[VAL_12:.*]] : !fir.ref<i32>) {
+! CHECK-NEXT:          omp.loop_nest (%[[VAL_13:.*]]) : i32 = (%[[VAL_9]]) to (%[[VAL_10]]) inclusive step (%[[VAL_11]]) {
+! CHECK:                 %[[VAL_14:.*]]:2 = hlfir.declare %[[VAL_12]] {uniq_name = "_QFreduction_iorEx"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
+! CHECK:                 fir.store %[[VAL_13]] to %[[VAL_8]]#1 : !fir.ref<i32>
+! CHECK:                 %[[VAL_15:.*]] = fir.load %[[VAL_8]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_16:.*]] = fir.convert %[[VAL_15]] : (i32) -> i64
+! CHECK:                 %[[VAL_17:.*]] = hlfir.designate %[[VAL_5]]#0 (%[[VAL_16]])  : (!fir.box<!fir.array<?xi32>>, i64) -> !fir.ref<i32>
+! CHECK:                 %[[VAL_18:.*]] = fir.load %[[VAL_14]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_19:.*]] = fir.load %[[VAL_17]] : !fir.ref<i32>
+! CHECK:                 %[[VAL_20:.*]] = arith.ori %[[VAL_18]], %[[VAL_19]] : i32
+! CHECK:                 hlfir.assign %[[VAL_20]] to %[[VAL_14]]#0 : i32, !fir.ref<i32>
+! CHECK:                 omp.yield
+! CHECK:               omp.terminator
 ! CHECK:             omp.terminator
 
 
diff --git a/flang/test/Lower/OpenMP/wsloop-reduction-ior.f90 b/flang/test/Lower/OpenMP/wsloop-reduction-ior.f90
index dd0bbeb1a076..f32be43b9b71 100644
--- a/flang/test/Lower/OpenMP/wsloop-reduction-ior.f90
+++ b/flang/test/Lower/OpenMP/wsloop-reduction-ior.f90
@@ -29,17 +29,19 @@
 ! CHECK:             %[[VAL_9:.*]] = arith.constant 1 : i32
 ! CHECK:             %[[VAL_10:.*]] = arith.constant 100 : i32
 ! CHECK:             %[[VAL_11:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop reduction(@ior_i32 %[[VAL_4]]#0 -> %[[VAL_12:.*]] : !fir.ref<i32>)  for  (%[[VAL_13:.*]]) : i32 = (%[[VAL_9]]) to (%[[VAL_10]]) inclusive step (%[[VAL_11]])
-! CHECK:               fir.store %[[VAL_13]] to %[[VAL_8]]#1 : !fir.ref<i32>
-! CHECK:               %[[VAL_14:.*]]:2 = hlfir.declare %[[VAL_12]] {uniq_name = "_QFreduction_iorEx"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
-! CHECK:               %[[VAL_15:.*]] = fir.load %[[VAL_8]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_16:.*]] = fir.convert %[[VAL_15]] : (i32) -> i64
-! CHECK:               %[[VAL_17:.*]] = hlfir.designate %[[VAL_5]]#0 (%[[VAL_16]])  : (!fir.box<!fir.array<?xi32>>, i64) -> !fir.ref<i32>
-! CHECK:               %[[VAL_18:.*]] = fir.load %[[VAL_14]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_19:.*]] = fir.load %[[VAL_17]] : !fir.ref<i32>
-! CHECK:               %[[VAL_20:.*]] = arith.ori %[[VAL_18]], %[[VAL_19]] : i32
-! CHECK:               hlfir.assign %[[VAL_20]] to %[[VAL_14]]#0 : i32, !fir.ref<i32>
-! CHECK:               omp.yield
+! CHECK:             omp.wsloop reduction(@ior_i32 %[[VAL_4]]#0 -> %[[VAL_12:.*]] : !fir.ref<i32>) {
+! CHECK-NEXT:          omp.loop_nest (%[[VAL_13:.*]]) : i32 = (%[[VAL_9]]) to (%[[VAL_10]]) inclusive step (%[[VAL_11]]) {
+! CHECK:                 %[[VAL_14:.*]]:2 = hlfir.declare %[[VAL_12]] {uniq_name = "_QFreduction_iorEx"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
+! CHECK:                 fir.store %[[VAL_13]] to %[[VAL_8]]#1 : !fir.ref<i32>
+! CHECK:                 %[[VAL_15:.*]] = fir.load %[[VAL_8]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_16:.*]] = fir.convert %[[VAL_15]] : (i32) -> i64
+! CHECK:                 %[[VAL_17:.*]] = hlfir.designate %[[VAL_5]]#0 (%[[VAL_16]])  : (!fir.box<!fir.array<?xi32>>, i64) -> !fir.ref<i32>
+! CHECK:                 %[[VAL_18:.*]] = fir.load %[[VAL_14]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_19:.*]] = fir.load %[[VAL_17]] : !fir.ref<i32>
+! CHECK:                 %[[VAL_20:.*]] = arith.ori %[[VAL_18]], %[[VAL_19]] : i32
+! CHECK:                 hlfir.assign %[[VAL_20]] to %[[VAL_14]]#0 : i32, !fir.ref<i32>
+! CHECK:                 omp.yield
+! CHECK:               omp.terminator
 ! CHECK:             omp.terminator
 
 
diff --git a/flang/test/Lower/OpenMP/wsloop-reduction-logical-and-byref.f90 b/flang/test/Lower/OpenMP/wsloop-reduction-logical-and-byref.f90
index 59411534e4a5..dfc018ed7c5a 100644
--- a/flang/test/Lower/OpenMP/wsloop-reduction-logical-and-byref.f90
+++ b/flang/test/Lower/OpenMP/wsloop-reduction-logical-and-byref.f90
@@ -42,20 +42,22 @@
 ! CHECK:             %[[VAL_12:.*]] = arith.constant 1 : i32
 ! CHECK:             %[[VAL_13:.*]] = arith.constant 100 : i32
 ! CHECK:             %[[VAL_14:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop byref reduction(@and_reduction %[[VAL_4]]#0 -> %[[VAL_15:.*]] : !fir.ref<!fir.logical<4>>)  for  (%[[VAL_16:.*]]) : i32 = (%[[VAL_12]]) to (%[[VAL_13]]) inclusive step (%[[VAL_14]]) {
-! CHECK:               fir.store %[[VAL_16]] to %[[VAL_11]]#1 : !fir.ref<i32>
-! CHECK:               %[[VAL_17:.*]]:2 = hlfir.declare %[[VAL_15]] {uniq_name = "_QFsimple_reductionEx"} : (!fir.ref<!fir.logical<4>>) -> (!fir.ref<!fir.logical<4>>, !fir.ref<!fir.logical<4>>)
-! CHECK:               %[[VAL_18:.*]] = fir.load %[[VAL_17]]#0 : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_19:.*]] = fir.load %[[VAL_11]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_20:.*]] = fir.convert %[[VAL_19]] : (i32) -> i64
-! CHECK:               %[[VAL_21:.*]] = hlfir.designate %[[VAL_7]]#0 (%[[VAL_20]])  : (!fir.ref<!fir.array<100x!fir.logical<4>>>, i64) -> !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_22:.*]] = fir.load %[[VAL_21]] : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_23:.*]] = fir.convert %[[VAL_18]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_24:.*]] = fir.convert %[[VAL_22]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_25:.*]] = arith.andi %[[VAL_23]], %[[VAL_24]] : i1
-! CHECK:               %[[VAL_26:.*]] = fir.convert %[[VAL_25]] : (i1) -> !fir.logical<4>
-! CHECK:               hlfir.assign %[[VAL_26]] to %[[VAL_17]]#0 : !fir.logical<4>, !fir.ref<!fir.logical<4>>
-! CHECK:               omp.yield
+! CHECK:             omp.wsloop byref reduction(@and_reduction %[[VAL_4]]#0 -> %[[VAL_15:.*]] : !fir.ref<!fir.logical<4>>) {
+! CHECK-NEXT:          omp.loop_nest (%[[VAL_16:.*]]) : i32 = (%[[VAL_12]]) to (%[[VAL_13]]) inclusive step (%[[VAL_14]]) {
+! CHECK:                 %[[VAL_17:.*]]:2 = hlfir.declare %[[VAL_15]] {uniq_name = "_QFsimple_reductionEx"} : (!fir.ref<!fir.logical<4>>) -> (!fir.ref<!fir.logical<4>>, !fir.ref<!fir.logical<4>>)
+! CHECK:                 fir.store %[[VAL_16]] to %[[VAL_11]]#1 : !fir.ref<i32>
+! CHECK:                 %[[VAL_18:.*]] = fir.load %[[VAL_17]]#0 : !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_19:.*]] = fir.load %[[VAL_11]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_20:.*]] = fir.convert %[[VAL_19]] : (i32) -> i64
+! CHECK:                 %[[VAL_21:.*]] = hlfir.designate %[[VAL_7]]#0 (%[[VAL_20]])  : (!fir.ref<!fir.array<100x!fir.logical<4>>>, i64) -> !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_22:.*]] = fir.load %[[VAL_21]] : !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_23:.*]] = fir.convert %[[VAL_18]] : (!fir.logical<4>) -> i1
+! CHECK:                 %[[VAL_24:.*]] = fir.convert %[[VAL_22]] : (!fir.logical<4>) -> i1
+! CHECK:                 %[[VAL_25:.*]] = arith.andi %[[VAL_23]], %[[VAL_24]] : i1
+! CHECK:                 %[[VAL_26:.*]] = fir.convert %[[VAL_25]] : (i1) -> !fir.logical<4>
+! CHECK:                 hlfir.assign %[[VAL_26]] to %[[VAL_17]]#0 : !fir.logical<4>, !fir.ref<!fir.logical<4>>
+! CHECK:                 omp.yield
+! CHECK:               omp.terminator
 ! CHECK:             omp.terminator
 ! CHECK:           return
 
@@ -90,20 +92,22 @@ end subroutine simple_reduction
 ! CHECK:             %[[VAL_12:.*]] = arith.constant 1 : i32
 ! CHECK:             %[[VAL_13:.*]] = arith.constant 100 : i32
 ! CHECK:             %[[VAL_14:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop byref reduction(@and_reduction %[[VAL_4]]#0 -> %[[VAL_15:.*]] : !fir.ref<!fir.logical<4>>)  for  (%[[VAL_16:.*]]) : i32 = (%[[VAL_12]]) to (%[[VAL_13]]) inclusive step (%[[VAL_14]]) {
-! CHECK:               fir.store %[[VAL_16]] to %[[VAL_11]]#1 : !fir.ref<i32>
-! CHECK:               %[[VAL_17:.*]]:2 = hlfir.declare %[[VAL_15]] {uniq_name = "_QFsimple_reduction_switch_orderEx"} : (!fir.ref<!fir.logical<4>>) -> (!fir.ref<!fir.logical<4>>, !fir.ref<!fir.logical<4>>)
-! CHECK:               %[[VAL_18:.*]] = fir.load %[[VAL_11]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_19:.*]] = fir.convert %[[VAL_18]] : (i32) -> i64
-! CHECK:               %[[VAL_20:.*]] = hlfir.designate %[[VAL_7]]#0 (%[[VAL_19]])  : (!fir.ref<!fir.array<100x!fir.logical<4>>>, i64) -> !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_21:.*]] = fir.load %[[VAL_20]] : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_22:.*]] = fir.load %[[VAL_17]]#0 : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_23:.*]] = fir.convert %[[VAL_21]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_24:.*]] = fir.convert %[[VAL_22]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_25:.*]] = arith.andi %[[VAL_23]], %[[VAL_24]] : i1
-! CHECK:               %[[VAL_26:.*]] = fir.convert %[[VAL_25]] : (i1) -> !fir.logical<4>
-! CHECK:               hlfir.assign %[[VAL_26]] to %[[VAL_17]]#0 : !fir.logical<4>, !fir.ref<!fir.logical<4>>
-! CHECK:               omp.yield
+! CHECK:             omp.wsloop byref reduction(@and_reduction %[[VAL_4]]#0 -> %[[VAL_15:.*]] : !fir.ref<!fir.logical<4>>) {
+! CHECK-NEXT:          omp.loop_nest (%[[VAL_16:.*]]) : i32 = (%[[VAL_12]]) to (%[[VAL_13]]) inclusive step (%[[VAL_14]]) {
+! CHECK:                 %[[VAL_17:.*]]:2 = hlfir.declare %[[VAL_15]] {uniq_name = "_QFsimple_reduction_switch_orderEx"} : (!fir.ref<!fir.logical<4>>) -> (!fir.ref<!fir.logical<4>>, !fir.ref<!fir.logical<4>>)
+! CHECK:                 fir.store %[[VAL_16]] to %[[VAL_11]]#1 : !fir.ref<i32>
+! CHECK:                 %[[VAL_18:.*]] = fir.load %[[VAL_11]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_19:.*]] = fir.convert %[[VAL_18]] : (i32) -> i64
+! CHECK:                 %[[VAL_20:.*]] = hlfir.designate %[[VAL_7]]#0 (%[[VAL_19]])  : (!fir.ref<!fir.array<100x!fir.logical<4>>>, i64) -> !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_21:.*]] = fir.load %[[VAL_20]] : !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_22:.*]] = fir.load %[[VAL_17]]#0 : !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_23:.*]] = fir.convert %[[VAL_21]] : (!fir.logical<4>) -> i1
+! CHECK:                 %[[VAL_24:.*]] = fir.convert %[[VAL_22]] : (!fir.logical<4>) -> i1
+! CHECK:                 %[[VAL_25:.*]] = arith.andi %[[VAL_23]], %[[VAL_24]] : i1
+! CHECK:                 %[[VAL_26:.*]] = fir.convert %[[VAL_25]] : (i1) -> !fir.logical<4>
+! CHECK:                 hlfir.assign %[[VAL_26]] to %[[VAL_17]]#0 : !fir.logical<4>, !fir.ref<!fir.logical<4>>
+! CHECK:                 omp.yield
+! CHECK:               omp.terminator
 ! CHECK:             omp.terminator
 ! CHECK:           return
 
@@ -147,42 +151,44 @@ end subroutine
 ! CHECK:             %[[VAL_20:.*]] = arith.constant 1 : i32
 ! CHECK:             %[[VAL_21:.*]] = arith.constant 100 : i32
 ! CHECK:             %[[VAL_22:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop byref reduction(@and_reduction %[[VAL_7]]#0 -> %[[VAL_23:.*]] : !fir.ref<!fir.logical<4>>, @and_reduction %[[VAL_9]]#0 -> %[[VAL_24:.*]] : !fir.ref<!fir.logical<4>>, @and_reduction %[[VAL_11]]#0 -> %[[VAL_25:.*]] : !fir.ref<!fir.logical<4>>)  for  (%[[VAL_26:.*]]) : i32 = (%[[VAL_20]]) to (%[[VAL_21]]) inclusive step (%[[VAL_22]]) {
-! CHECK:               fir.store %[[VAL_26]] to %[[VAL_19]]#1 : !fir.ref<i32>
-! CHECK:               %[[VAL_27:.*]]:2 = hlfir.declare %[[VAL_23]] {uniq_name = "_QFmultiple_reductionsEx"} : (!fir.ref<!fir.logical<4>>) -> (!fir.ref<!fir.logical<4>>, !fir.ref<!fir.logical<4>>)
-! CHECK:               %[[VAL_28:.*]]:2 = hlfir.declare %[[VAL_24]] {uniq_name = "_QFmultiple_reductionsEy"} : (!fir.ref<!fir.logical<4>>) -> (!fir.ref<!fir.logical<4>>, !fir.ref<!fir.logical<4>>)
-! CHECK:               %[[VAL_29:.*]]:2 = hlfir.declare %[[VAL_25]] {uniq_name = "_QFmultiple_reductionsEz"} : (!fir.ref<!fir.logical<4>>) -> (!fir.ref<!fir.logical<4>>, !fir.ref<!fir.logical<4>>)
-! CHECK:               %[[VAL_30:.*]] = fir.load %[[VAL_27]]#0 : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_31:.*]] = fir.load %[[VAL_19]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_32:.*]] = fir.convert %[[VAL_31]] : (i32) -> i64
-! CHECK:               %[[VAL_33:.*]] = hlfir.designate %[[VAL_5]]#0 (%[[VAL_32]])  : (!fir.ref<!fir.array<100x!fir.logical<4>>>, i64) -> !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_34:.*]] = fir.load %[[VAL_33]] : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_35:.*]] = fir.convert %[[VAL_30]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_36:.*]] = fir.convert %[[VAL_34]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_37:.*]] = arith.andi %[[VAL_35]], %[[VAL_36]] : i1
-! CHECK:               %[[VAL_38:.*]] = fir.convert %[[VAL_37]] : (i1) -> !fir.logical<4>
-! CHECK:               hlfir.assign %[[VAL_38]] to %[[VAL_27]]#0 : !fir.logical<4>, !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_39:.*]] = fir.load %[[VAL_28]]#0 : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_40:.*]] = fir.load %[[VAL_19]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_41:.*]] = fir.convert %[[VAL_40]] : (i32) -> i64
-! CHECK:               %[[VAL_42:.*]] = hlfir.designate %[[VAL_5]]#0 (%[[VAL_41]])  : (!fir.ref<!fir.array<100x!fir.logical<4>>>, i64) -> !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_43:.*]] = fir.load %[[VAL_42]] : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_44:.*]] = fir.convert %[[VAL_39]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_45:.*]] = fir.convert %[[VAL_43]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_46:.*]] = arith.andi %[[VAL_44]], %[[VAL_45]] : i1
-! CHECK:               %[[VAL_47:.*]] = fir.convert %[[VAL_46]] : (i1) -> !fir.logical<4>
-! CHECK:               hlfir.assign %[[VAL_47]] to %[[VAL_28]]#0 : !fir.logical<4>, !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_48:.*]] = fir.load %[[VAL_29]]#0 : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_49:.*]] = fir.load %[[VAL_19]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_50:.*]] = fir.convert %[[VAL_49]] : (i32) -> i64
-! CHECK:               %[[VAL_51:.*]] = hlfir.designate %[[VAL_5]]#0 (%[[VAL_50]])  : (!fir.ref<!fir.array<100x!fir.logical<4>>>, i64) -> !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_52:.*]] = fir.load %[[VAL_51]] : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_53:.*]] = fir.convert %[[VAL_48]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_54:.*]] = fir.convert %[[VAL_52]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_55:.*]] = arith.andi %[[VAL_53]], %[[VAL_54]] : i1
-! CHECK:               %[[VAL_56:.*]] = fir.convert %[[VAL_55]] : (i1) -> !fir.logical<4>
-! CHECK:               hlfir.assign %[[VAL_56]] to %[[VAL_29]]#0 : !fir.logical<4>, !fir.ref<!fir.logical<4>>
-! CHECK:               omp.yield
+! CHECK:             omp.wsloop byref reduction(@and_reduction %[[VAL_7]]#0 -> %[[VAL_23:.*]] : !fir.ref<!fir.logical<4>>, @and_reduction %[[VAL_9]]#0 -> %[[VAL_24:.*]] : !fir.ref<!fir.logical<4>>, @and_reduction %[[VAL_11]]#0 -> %[[VAL_25:.*]] : !fir.ref<!fir.logical<4>>) {
+! CHECK-NEXT:          omp.loop_nest (%[[VAL_26:.*]]) : i32 = (%[[VAL_20]]) to (%[[VAL_21]]) inclusive step (%[[VAL_22]]) {
+! CHECK:                 %[[VAL_27:.*]]:2 = hlfir.declare %[[VAL_23]] {uniq_name = "_QFmultiple_reductionsEx"} : (!fir.ref<!fir.logical<4>>) -> (!fir.ref<!fir.logical<4>>, !fir.ref<!fir.logical<4>>)
+! CHECK:                 %[[VAL_28:.*]]:2 = hlfir.declare %[[VAL_24]] {uniq_name = "_QFmultiple_reductionsEy"} : (!fir.ref<!fir.logical<4>>) -> (!fir.ref<!fir.logical<4>>, !fir.ref<!fir.logical<4>>)
+! CHECK:                 %[[VAL_29:.*]]:2 = hlfir.declare %[[VAL_25]] {uniq_name = "_QFmultiple_reductionsEz"} : (!fir.ref<!fir.logical<4>>) -> (!fir.ref<!fir.logical<4>>, !fir.ref<!fir.logical<4>>)
+! CHECK:                 fir.store %[[VAL_26]] to %[[VAL_19]]#1 : !fir.ref<i32>
+! CHECK:                 %[[VAL_30:.*]] = fir.load %[[VAL_27]]#0 : !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_31:.*]] = fir.load %[[VAL_19]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_32:.*]] = fir.convert %[[VAL_31]] : (i32) -> i64
+! CHECK:                 %[[VAL_33:.*]] = hlfir.designate %[[VAL_5]]#0 (%[[VAL_32]])  : (!fir.ref<!fir.array<100x!fir.logical<4>>>, i64) -> !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_34:.*]] = fir.load %[[VAL_33]] : !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_35:.*]] = fir.convert %[[VAL_30]] : (!fir.logical<4>) -> i1
+! CHECK:                 %[[VAL_36:.*]] = fir.convert %[[VAL_34]] : (!fir.logical<4>) -> i1
+! CHECK:                 %[[VAL_37:.*]] = arith.andi %[[VAL_35]], %[[VAL_36]] : i1
+! CHECK:                 %[[VAL_38:.*]] = fir.convert %[[VAL_37]] : (i1) -> !fir.logical<4>
+! CHECK:                 hlfir.assign %[[VAL_38]] to %[[VAL_27]]#0 : !fir.logical<4>, !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_39:.*]] = fir.load %[[VAL_28]]#0 : !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_40:.*]] = fir.load %[[VAL_19]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_41:.*]] = fir.convert %[[VAL_40]] : (i32) -> i64
+! CHECK:                 %[[VAL_42:.*]] = hlfir.designate %[[VAL_5]]#0 (%[[VAL_41]])  : (!fir.ref<!fir.array<100x!fir.logical<4>>>, i64) -> !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_43:.*]] = fir.load %[[VAL_42]] : !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_44:.*]] = fir.convert %[[VAL_39]] : (!fir.logical<4>) -> i1
+! CHECK:                 %[[VAL_45:.*]] = fir.convert %[[VAL_43]] : (!fir.logical<4>) -> i1
+! CHECK:                 %[[VAL_46:.*]] = arith.andi %[[VAL_44]], %[[VAL_45]] : i1
+! CHECK:                 %[[VAL_47:.*]] = fir.convert %[[VAL_46]] : (i1) -> !fir.logical<4>
+! CHECK:                 hlfir.assign %[[VAL_47]] to %[[VAL_28]]#0 : !fir.logical<4>, !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_48:.*]] = fir.load %[[VAL_29]]#0 : !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_49:.*]] = fir.load %[[VAL_19]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_50:.*]] = fir.convert %[[VAL_49]] : (i32) -> i64
+! CHECK:                 %[[VAL_51:.*]] = hlfir.designate %[[VAL_5]]#0 (%[[VAL_50]])  : (!fir.ref<!fir.array<100x!fir.logical<4>>>, i64) -> !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_52:.*]] = fir.load %[[VAL_51]] : !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_53:.*]] = fir.convert %[[VAL_48]] : (!fir.logical<4>) -> i1
+! CHECK:                 %[[VAL_54:.*]] = fir.convert %[[VAL_52]] : (!fir.logical<4>) -> i1
+! CHECK:                 %[[VAL_55:.*]] = arith.andi %[[VAL_53]], %[[VAL_54]] : i1
+! CHECK:                 %[[VAL_56:.*]] = fir.convert %[[VAL_55]] : (i1) -> !fir.logical<4>
+! CHECK:                 hlfir.assign %[[VAL_56]] to %[[VAL_29]]#0 : !fir.logical<4>, !fir.ref<!fir.logical<4>>
+! CHECK:                 omp.yield
+! CHECK:               omp.terminator
 ! CHECK:             omp.terminator
 ! CHECK:           return
 
diff --git a/flang/test/Lower/OpenMP/wsloop-reduction-logical-and.f90 b/flang/test/Lower/OpenMP/wsloop-reduction-logical-and.f90
index 9ca733281c2f..c529bd4755b6 100644
--- a/flang/test/Lower/OpenMP/wsloop-reduction-logical-and.f90
+++ b/flang/test/Lower/OpenMP/wsloop-reduction-logical-and.f90
@@ -36,20 +36,22 @@
 ! CHECK:             %[[VAL_12:.*]] = arith.constant 1 : i32
 ! CHECK:             %[[VAL_13:.*]] = arith.constant 100 : i32
 ! CHECK:             %[[VAL_14:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop reduction(@and_reduction %[[VAL_4]]#0 -> %[[VAL_15:.*]] : !fir.ref<!fir.logical<4>>)  for  (%[[VAL_16:.*]]) : i32 = (%[[VAL_12]]) to (%[[VAL_13]]) inclusive step (%[[VAL_14]]) {
-! CHECK:               fir.store %[[VAL_16]] to %[[VAL_11]]#1 : !fir.ref<i32>
-! CHECK:               %[[VAL_17:.*]]:2 = hlfir.declare %[[VAL_15]] {uniq_name = "_QFsimple_reductionEx"} : (!fir.ref<!fir.logical<4>>) -> (!fir.ref<!fir.logical<4>>, !fir.ref<!fir.logical<4>>)
-! CHECK:               %[[VAL_18:.*]] = fir.load %[[VAL_17]]#0 : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_19:.*]] = fir.load %[[VAL_11]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_20:.*]] = fir.convert %[[VAL_19]] : (i32) -> i64
-! CHECK:               %[[VAL_21:.*]] = hlfir.designate %[[VAL_7]]#0 (%[[VAL_20]])  : (!fir.ref<!fir.array<100x!fir.logical<4>>>, i64) -> !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_22:.*]] = fir.load %[[VAL_21]] : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_23:.*]] = fir.convert %[[VAL_18]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_24:.*]] = fir.convert %[[VAL_22]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_25:.*]] = arith.andi %[[VAL_23]], %[[VAL_24]] : i1
-! CHECK:               %[[VAL_26:.*]] = fir.convert %[[VAL_25]] : (i1) -> !fir.logical<4>
-! CHECK:               hlfir.assign %[[VAL_26]] to %[[VAL_17]]#0 : !fir.logical<4>, !fir.ref<!fir.logical<4>>
-! CHECK:               omp.yield
+! CHECK:             omp.wsloop reduction(@and_reduction %[[VAL_4]]#0 -> %[[VAL_15:.*]] : !fir.ref<!fir.logical<4>>) {
+! CHECK-NEXT:          omp.loop_nest (%[[VAL_16:.*]]) : i32 = (%[[VAL_12]]) to (%[[VAL_13]]) inclusive step (%[[VAL_14]]) {
+! CHECK:                 %[[VAL_17:.*]]:2 = hlfir.declare %[[VAL_15]] {uniq_name = "_QFsimple_reductionEx"} : (!fir.ref<!fir.logical<4>>) -> (!fir.ref<!fir.logical<4>>, !fir.ref<!fir.logical<4>>)
+! CHECK:                 fir.store %[[VAL_16]] to %[[VAL_11]]#1 : !fir.ref<i32>
+! CHECK:                 %[[VAL_18:.*]] = fir.load %[[VAL_17]]#0 : !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_19:.*]] = fir.load %[[VAL_11]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_20:.*]] = fir.convert %[[VAL_19]] : (i32) -> i64
+! CHECK:                 %[[VAL_21:.*]] = hlfir.designate %[[VAL_7]]#0 (%[[VAL_20]])  : (!fir.ref<!fir.array<100x!fir.logical<4>>>, i64) -> !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_22:.*]] = fir.load %[[VAL_21]] : !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_23:.*]] = fir.convert %[[VAL_18]] : (!fir.logical<4>) -> i1
+! CHECK:                 %[[VAL_24:.*]] = fir.convert %[[VAL_22]] : (!fir.logical<4>) -> i1
+! CHECK:                 %[[VAL_25:.*]] = arith.andi %[[VAL_23]], %[[VAL_24]] : i1
+! CHECK:                 %[[VAL_26:.*]] = fir.convert %[[VAL_25]] : (i1) -> !fir.logical<4>
+! CHECK:                 hlfir.assign %[[VAL_26]] to %[[VAL_17]]#0 : !fir.logical<4>, !fir.ref<!fir.logical<4>>
+! CHECK:                 omp.yield
+! CHECK:               omp.terminator
 ! CHECK:             omp.terminator
 ! CHECK:           return
 
@@ -84,20 +86,22 @@ end subroutine simple_reduction
 ! CHECK:             %[[VAL_12:.*]] = arith.constant 1 : i32
 ! CHECK:             %[[VAL_13:.*]] = arith.constant 100 : i32
 ! CHECK:             %[[VAL_14:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop reduction(@and_reduction %[[VAL_4]]#0 -> %[[VAL_15:.*]] : !fir.ref<!fir.logical<4>>)  for  (%[[VAL_16:.*]]) : i32 = (%[[VAL_12]]) to (%[[VAL_13]]) inclusive step (%[[VAL_14]]) {
-! CHECK:               fir.store %[[VAL_16]] to %[[VAL_11]]#1 : !fir.ref<i32>
-! CHECK:               %[[VAL_17:.*]]:2 = hlfir.declare %[[VAL_15]] {uniq_name = "_QFsimple_reduction_switch_orderEx"} : (!fir.ref<!fir.logical<4>>) -> (!fir.ref<!fir.logical<4>>, !fir.ref<!fir.logical<4>>)
-! CHECK:               %[[VAL_18:.*]] = fir.load %[[VAL_11]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_19:.*]] = fir.convert %[[VAL_18]] : (i32) -> i64
-! CHECK:               %[[VAL_20:.*]] = hlfir.designate %[[VAL_7]]#0 (%[[VAL_19]])  : (!fir.ref<!fir.array<100x!fir.logical<4>>>, i64) -> !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_21:.*]] = fir.load %[[VAL_20]] : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_22:.*]] = fir.load %[[VAL_17]]#0 : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_23:.*]] = fir.convert %[[VAL_21]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_24:.*]] = fir.convert %[[VAL_22]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_25:.*]] = arith.andi %[[VAL_23]], %[[VAL_24]] : i1
-! CHECK:               %[[VAL_26:.*]] = fir.convert %[[VAL_25]] : (i1) -> !fir.logical<4>
-! CHECK:               hlfir.assign %[[VAL_26]] to %[[VAL_17]]#0 : !fir.logical<4>, !fir.ref<!fir.logical<4>>
-! CHECK:               omp.yield
+! CHECK:             omp.wsloop reduction(@and_reduction %[[VAL_4]]#0 -> %[[VAL_15:.*]] : !fir.ref<!fir.logical<4>>) {
+! CHECK-NEXT:          omp.loop_nest (%[[VAL_16:.*]]) : i32 = (%[[VAL_12]]) to (%[[VAL_13]]) inclusive step (%[[VAL_14]]) {
+! CHECK:                 %[[VAL_17:.*]]:2 = hlfir.declare %[[VAL_15]] {uniq_name = "_QFsimple_reduction_switch_orderEx"} : (!fir.ref<!fir.logical<4>>) -> (!fir.ref<!fir.logical<4>>, !fir.ref<!fir.logical<4>>)
+! CHECK:                 fir.store %[[VAL_16]] to %[[VAL_11]]#1 : !fir.ref<i32>
+! CHECK:                 %[[VAL_18:.*]] = fir.load %[[VAL_11]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_19:.*]] = fir.convert %[[VAL_18]] : (i32) -> i64
+! CHECK:                 %[[VAL_20:.*]] = hlfir.designate %[[VAL_7]]#0 (%[[VAL_19]])  : (!fir.ref<!fir.array<100x!fir.logical<4>>>, i64) -> !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_21:.*]] = fir.load %[[VAL_20]] : !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_22:.*]] = fir.load %[[VAL_17]]#0 : !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_23:.*]] = fir.convert %[[VAL_21]] : (!fir.logical<4>) -> i1
+! CHECK:                 %[[VAL_24:.*]] = fir.convert %[[VAL_22]] : (!fir.logical<4>) -> i1
+! CHECK:                 %[[VAL_25:.*]] = arith.andi %[[VAL_23]], %[[VAL_24]] : i1
+! CHECK:                 %[[VAL_26:.*]] = fir.convert %[[VAL_25]] : (i1) -> !fir.logical<4>
+! CHECK:                 hlfir.assign %[[VAL_26]] to %[[VAL_17]]#0 : !fir.logical<4>, !fir.ref<!fir.logical<4>>
+! CHECK:                 omp.yield
+! CHECK:               omp.terminator
 ! CHECK:             omp.terminator
 ! CHECK:           return
 
@@ -141,42 +145,44 @@ end subroutine
 ! CHECK:             %[[VAL_20:.*]] = arith.constant 1 : i32
 ! CHECK:             %[[VAL_21:.*]] = arith.constant 100 : i32
 ! CHECK:             %[[VAL_22:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop reduction(@and_reduction %[[VAL_7]]#0 -> %[[VAL_23:.*]] : !fir.ref<!fir.logical<4>>, @and_reduction %[[VAL_9]]#0 -> %[[VAL_24:.*]] : !fir.ref<!fir.logical<4>>, @and_reduction %[[VAL_11]]#0 -> %[[VAL_25:.*]] : !fir.ref<!fir.logical<4>>)  for  (%[[VAL_26:.*]]) : i32 = (%[[VAL_20]]) to (%[[VAL_21]]) inclusive step (%[[VAL_22]]) {
-! CHECK:               fir.store %[[VAL_26]] to %[[VAL_19]]#1 : !fir.ref<i32>
-! CHECK:               %[[VAL_27:.*]]:2 = hlfir.declare %[[VAL_23]] {uniq_name = "_QFmultiple_reductionsEx"} : (!fir.ref<!fir.logical<4>>) -> (!fir.ref<!fir.logical<4>>, !fir.ref<!fir.logical<4>>)
-! CHECK:               %[[VAL_28:.*]]:2 = hlfir.declare %[[VAL_24]] {uniq_name = "_QFmultiple_reductionsEy"} : (!fir.ref<!fir.logical<4>>) -> (!fir.ref<!fir.logical<4>>, !fir.ref<!fir.logical<4>>)
-! CHECK:               %[[VAL_29:.*]]:2 = hlfir.declare %[[VAL_25]] {uniq_name = "_QFmultiple_reductionsEz"} : (!fir.ref<!fir.logical<4>>) -> (!fir.ref<!fir.logical<4>>, !fir.ref<!fir.logical<4>>)
-! CHECK:               %[[VAL_30:.*]] = fir.load %[[VAL_27]]#0 : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_31:.*]] = fir.load %[[VAL_19]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_32:.*]] = fir.convert %[[VAL_31]] : (i32) -> i64
-! CHECK:               %[[VAL_33:.*]] = hlfir.designate %[[VAL_5]]#0 (%[[VAL_32]])  : (!fir.ref<!fir.array<100x!fir.logical<4>>>, i64) -> !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_34:.*]] = fir.load %[[VAL_33]] : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_35:.*]] = fir.convert %[[VAL_30]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_36:.*]] = fir.convert %[[VAL_34]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_37:.*]] = arith.andi %[[VAL_35]], %[[VAL_36]] : i1
-! CHECK:               %[[VAL_38:.*]] = fir.convert %[[VAL_37]] : (i1) -> !fir.logical<4>
-! CHECK:               hlfir.assign %[[VAL_38]] to %[[VAL_27]]#0 : !fir.logical<4>, !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_39:.*]] = fir.load %[[VAL_28]]#0 : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_40:.*]] = fir.load %[[VAL_19]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_41:.*]] = fir.convert %[[VAL_40]] : (i32) -> i64
-! CHECK:               %[[VAL_42:.*]] = hlfir.designate %[[VAL_5]]#0 (%[[VAL_41]])  : (!fir.ref<!fir.array<100x!fir.logical<4>>>, i64) -> !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_43:.*]] = fir.load %[[VAL_42]] : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_44:.*]] = fir.convert %[[VAL_39]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_45:.*]] = fir.convert %[[VAL_43]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_46:.*]] = arith.andi %[[VAL_44]], %[[VAL_45]] : i1
-! CHECK:               %[[VAL_47:.*]] = fir.convert %[[VAL_46]] : (i1) -> !fir.logical<4>
-! CHECK:               hlfir.assign %[[VAL_47]] to %[[VAL_28]]#0 : !fir.logical<4>, !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_48:.*]] = fir.load %[[VAL_29]]#0 : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_49:.*]] = fir.load %[[VAL_19]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_50:.*]] = fir.convert %[[VAL_49]] : (i32) -> i64
-! CHECK:               %[[VAL_51:.*]] = hlfir.designate %[[VAL_5]]#0 (%[[VAL_50]])  : (!fir.ref<!fir.array<100x!fir.logical<4>>>, i64) -> !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_52:.*]] = fir.load %[[VAL_51]] : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_53:.*]] = fir.convert %[[VAL_48]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_54:.*]] = fir.convert %[[VAL_52]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_55:.*]] = arith.andi %[[VAL_53]], %[[VAL_54]] : i1
-! CHECK:               %[[VAL_56:.*]] = fir.convert %[[VAL_55]] : (i1) -> !fir.logical<4>
-! CHECK:               hlfir.assign %[[VAL_56]] to %[[VAL_29]]#0 : !fir.logical<4>, !fir.ref<!fir.logical<4>>
-! CHECK:               omp.yield
+! CHECK:             omp.wsloop reduction(@and_reduction %[[VAL_7]]#0 -> %[[VAL_23:.*]] : !fir.ref<!fir.logical<4>>, @and_reduction %[[VAL_9]]#0 -> %[[VAL_24:.*]] : !fir.ref<!fir.logical<4>>, @and_reduction %[[VAL_11]]#0 -> %[[VAL_25:.*]] : !fir.ref<!fir.logical<4>>) {
+! CHECK-NEXT:          omp.loop_nest (%[[VAL_26:.*]]) : i32 = (%[[VAL_20]]) to (%[[VAL_21]]) inclusive step (%[[VAL_22]]) {
+! CHECK:                 %[[VAL_27:.*]]:2 = hlfir.declare %[[VAL_23]] {uniq_name = "_QFmultiple_reductionsEx"} : (!fir.ref<!fir.logical<4>>) -> (!fir.ref<!fir.logical<4>>, !fir.ref<!fir.logical<4>>)
+! CHECK:                 %[[VAL_28:.*]]:2 = hlfir.declare %[[VAL_24]] {uniq_name = "_QFmultiple_reductionsEy"} : (!fir.ref<!fir.logical<4>>) -> (!fir.ref<!fir.logical<4>>, !fir.ref<!fir.logical<4>>)
+! CHECK:                 %[[VAL_29:.*]]:2 = hlfir.declare %[[VAL_25]] {uniq_name = "_QFmultiple_reductionsEz"} : (!fir.ref<!fir.logical<4>>) -> (!fir.ref<!fir.logical<4>>, !fir.ref<!fir.logical<4>>)
+! CHECK:                 fir.store %[[VAL_26]] to %[[VAL_19]]#1 : !fir.ref<i32>
+! CHECK:                 %[[VAL_30:.*]] = fir.load %[[VAL_27]]#0 : !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_31:.*]] = fir.load %[[VAL_19]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_32:.*]] = fir.convert %[[VAL_31]] : (i32) -> i64
+! CHECK:                 %[[VAL_33:.*]] = hlfir.designate %[[VAL_5]]#0 (%[[VAL_32]])  : (!fir.ref<!fir.array<100x!fir.logical<4>>>, i64) -> !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_34:.*]] = fir.load %[[VAL_33]] : !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_35:.*]] = fir.convert %[[VAL_30]] : (!fir.logical<4>) -> i1
+! CHECK:                 %[[VAL_36:.*]] = fir.convert %[[VAL_34]] : (!fir.logical<4>) -> i1
+! CHECK:                 %[[VAL_37:.*]] = arith.andi %[[VAL_35]], %[[VAL_36]] : i1
+! CHECK:                 %[[VAL_38:.*]] = fir.convert %[[VAL_37]] : (i1) -> !fir.logical<4>
+! CHECK:                 hlfir.assign %[[VAL_38]] to %[[VAL_27]]#0 : !fir.logical<4>, !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_39:.*]] = fir.load %[[VAL_28]]#0 : !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_40:.*]] = fir.load %[[VAL_19]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_41:.*]] = fir.convert %[[VAL_40]] : (i32) -> i64
+! CHECK:                 %[[VAL_42:.*]] = hlfir.designate %[[VAL_5]]#0 (%[[VAL_41]])  : (!fir.ref<!fir.array<100x!fir.logical<4>>>, i64) -> !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_43:.*]] = fir.load %[[VAL_42]] : !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_44:.*]] = fir.convert %[[VAL_39]] : (!fir.logical<4>) -> i1
+! CHECK:                 %[[VAL_45:.*]] = fir.convert %[[VAL_43]] : (!fir.logical<4>) -> i1
+! CHECK:                 %[[VAL_46:.*]] = arith.andi %[[VAL_44]], %[[VAL_45]] : i1
+! CHECK:                 %[[VAL_47:.*]] = fir.convert %[[VAL_46]] : (i1) -> !fir.logical<4>
+! CHECK:                 hlfir.assign %[[VAL_47]] to %[[VAL_28]]#0 : !fir.logical<4>, !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_48:.*]] = fir.load %[[VAL_29]]#0 : !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_49:.*]] = fir.load %[[VAL_19]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_50:.*]] = fir.convert %[[VAL_49]] : (i32) -> i64
+! CHECK:                 %[[VAL_51:.*]] = hlfir.designate %[[VAL_5]]#0 (%[[VAL_50]])  : (!fir.ref<!fir.array<100x!fir.logical<4>>>, i64) -> !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_52:.*]] = fir.load %[[VAL_51]] : !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_53:.*]] = fir.convert %[[VAL_48]] : (!fir.logical<4>) -> i1
+! CHECK:                 %[[VAL_54:.*]] = fir.convert %[[VAL_52]] : (!fir.logical<4>) -> i1
+! CHECK:                 %[[VAL_55:.*]] = arith.andi %[[VAL_53]], %[[VAL_54]] : i1
+! CHECK:                 %[[VAL_56:.*]] = fir.convert %[[VAL_55]] : (i1) -> !fir.logical<4>
+! CHECK:                 hlfir.assign %[[VAL_56]] to %[[VAL_29]]#0 : !fir.logical<4>, !fir.ref<!fir.logical<4>>
+! CHECK:                 omp.yield
+! CHECK:               omp.terminator
 ! CHECK:             omp.terminator
 ! CHECK:           return
 
diff --git a/flang/test/Lower/OpenMP/wsloop-reduction-logical-eqv-byref.f90 b/flang/test/Lower/OpenMP/wsloop-reduction-logical-eqv-byref.f90
index 1d6e1b0545c3..a54795a4446f 100644
--- a/flang/test/Lower/OpenMP/wsloop-reduction-logical-eqv-byref.f90
+++ b/flang/test/Lower/OpenMP/wsloop-reduction-logical-eqv-byref.f90
@@ -42,20 +42,22 @@
 ! CHECK:             %[[VAL_12:.*]] = arith.constant 1 : i32
 ! CHECK:             %[[VAL_13:.*]] = arith.constant 100 : i32
 ! CHECK:             %[[VAL_14:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop byref reduction(@eqv_reduction %[[VAL_4]]#0 -> %[[VAL_15:.*]] : !fir.ref<!fir.logical<4>>)  for  (%[[VAL_16:.*]]) : i32 = (%[[VAL_12]]) to (%[[VAL_13]]) inclusive step (%[[VAL_14]]) {
-! CHECK:               fir.store %[[VAL_16]] to %[[VAL_11]]#1 : !fir.ref<i32>
-! CHECK:               %[[VAL_17:.*]]:2 = hlfir.declare %[[VAL_15]] {uniq_name = "_QFsimple_reductionEx"} : (!fir.ref<!fir.logical<4>>) -> (!fir.ref<!fir.logical<4>>, !fir.ref<!fir.logical<4>>)
-! CHECK:               %[[VAL_18:.*]] = fir.load %[[VAL_17]]#0 : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_19:.*]] = fir.load %[[VAL_11]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_20:.*]] = fir.convert %[[VAL_19]] : (i32) -> i64
-! CHECK:               %[[VAL_21:.*]] = hlfir.designate %[[VAL_7]]#0 (%[[VAL_20]])  : (!fir.ref<!fir.array<100x!fir.logical<4>>>, i64) -> !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_22:.*]] = fir.load %[[VAL_21]] : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_23:.*]] = fir.convert %[[VAL_18]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_24:.*]] = fir.convert %[[VAL_22]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_25:.*]] = arith.cmpi eq, %[[VAL_23]], %[[VAL_24]] : i1
-! CHECK:               %[[VAL_26:.*]] = fir.convert %[[VAL_25]] : (i1) -> !fir.logical<4>
-! CHECK:               hlfir.assign %[[VAL_26]] to %[[VAL_17]]#0 : !fir.logical<4>, !fir.ref<!fir.logical<4>>
-! CHECK:               omp.yield
+! CHECK:             omp.wsloop byref reduction(@eqv_reduction %[[VAL_4]]#0 -> %[[VAL_15:.*]] : !fir.ref<!fir.logical<4>>) {
+! CHECK-NEXT:          omp.loop_nest (%[[VAL_16:.*]]) : i32 = (%[[VAL_12]]) to (%[[VAL_13]]) inclusive step (%[[VAL_14]]) {
+! CHECK:                 %[[VAL_17:.*]]:2 = hlfir.declare %[[VAL_15]] {uniq_name = "_QFsimple_reductionEx"} : (!fir.ref<!fir.logical<4>>) -> (!fir.ref<!fir.logical<4>>, !fir.ref<!fir.logical<4>>)
+! CHECK:                 fir.store %[[VAL_16]] to %[[VAL_11]]#1 : !fir.ref<i32>
+! CHECK:                 %[[VAL_18:.*]] = fir.load %[[VAL_17]]#0 : !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_19:.*]] = fir.load %[[VAL_11]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_20:.*]] = fir.convert %[[VAL_19]] : (i32) -> i64
+! CHECK:                 %[[VAL_21:.*]] = hlfir.designate %[[VAL_7]]#0 (%[[VAL_20]])  : (!fir.ref<!fir.array<100x!fir.logical<4>>>, i64) -> !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_22:.*]] = fir.load %[[VAL_21]] : !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_23:.*]] = fir.convert %[[VAL_18]] : (!fir.logical<4>) -> i1
+! CHECK:                 %[[VAL_24:.*]] = fir.convert %[[VAL_22]] : (!fir.logical<4>) -> i1
+! CHECK:                 %[[VAL_25:.*]] = arith.cmpi eq, %[[VAL_23]], %[[VAL_24]] : i1
+! CHECK:                 %[[VAL_26:.*]] = fir.convert %[[VAL_25]] : (i1) -> !fir.logical<4>
+! CHECK:                 hlfir.assign %[[VAL_26]] to %[[VAL_17]]#0 : !fir.logical<4>, !fir.ref<!fir.logical<4>>
+! CHECK:                 omp.yield
+! CHECK:               omp.terminator
 ! CHECK:             omp.terminator
 ! CHECK:           return
 
@@ -89,20 +91,22 @@ end subroutine
 ! CHECK:             %[[VAL_12:.*]] = arith.constant 1 : i32
 ! CHECK:             %[[VAL_13:.*]] = arith.constant 100 : i32
 ! CHECK:             %[[VAL_14:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop byref reduction(@eqv_reduction %[[VAL_4]]#0 -> %[[VAL_15:.*]] : !fir.ref<!fir.logical<4>>)  for  (%[[VAL_16:.*]]) : i32 = (%[[VAL_12]]) to (%[[VAL_13]]) inclusive step (%[[VAL_14]]) {
-! CHECK:               fir.store %[[VAL_16]] to %[[VAL_11]]#1 : !fir.ref<i32>
-! CHECK:               %[[VAL_17:.*]]:2 = hlfir.declare %[[VAL_15]] {uniq_name = "_QFsimple_reduction_switch_orderEx"} : (!fir.ref<!fir.logical<4>>) -> (!fir.ref<!fir.logical<4>>, !fir.ref<!fir.logical<4>>)
-! CHECK:               %[[VAL_18:.*]] = fir.load %[[VAL_11]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_19:.*]] = fir.convert %[[VAL_18]] : (i32) -> i64
-! CHECK:               %[[VAL_20:.*]] = hlfir.designate %[[VAL_7]]#0 (%[[VAL_19]])  : (!fir.ref<!fir.array<100x!fir.logical<4>>>, i64) -> !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_21:.*]] = fir.load %[[VAL_20]] : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_22:.*]] = fir.load %[[VAL_17]]#0 : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_23:.*]] = fir.convert %[[VAL_21]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_24:.*]] = fir.convert %[[VAL_22]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_25:.*]] = arith.cmpi eq, %[[VAL_23]], %[[VAL_24]] : i1
-! CHECK:               %[[VAL_26:.*]] = fir.convert %[[VAL_25]] : (i1) -> !fir.logical<4>
-! CHECK:               hlfir.assign %[[VAL_26]] to %[[VAL_17]]#0 : !fir.logical<4>, !fir.ref<!fir.logical<4>>
-! CHECK:               omp.yield
+! CHECK:             omp.wsloop byref reduction(@eqv_reduction %[[VAL_4]]#0 -> %[[VAL_15:.*]] : !fir.ref<!fir.logical<4>>) {
+! CHECK-NEXT:          omp.loop_nest (%[[VAL_16:.*]]) : i32 = (%[[VAL_12]]) to (%[[VAL_13]]) inclusive step (%[[VAL_14]]) {
+! CHECK:                 %[[VAL_17:.*]]:2 = hlfir.declare %[[VAL_15]] {uniq_name = "_QFsimple_reduction_switch_orderEx"} : (!fir.ref<!fir.logical<4>>) -> (!fir.ref<!fir.logical<4>>, !fir.ref<!fir.logical<4>>)
+! CHECK:                 fir.store %[[VAL_16]] to %[[VAL_11]]#1 : !fir.ref<i32>
+! CHECK:                 %[[VAL_18:.*]] = fir.load %[[VAL_11]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_19:.*]] = fir.convert %[[VAL_18]] : (i32) -> i64
+! CHECK:                 %[[VAL_20:.*]] = hlfir.designate %[[VAL_7]]#0 (%[[VAL_19]])  : (!fir.ref<!fir.array<100x!fir.logical<4>>>, i64) -> !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_21:.*]] = fir.load %[[VAL_20]] : !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_22:.*]] = fir.load %[[VAL_17]]#0 : !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_23:.*]] = fir.convert %[[VAL_21]] : (!fir.logical<4>) -> i1
+! CHECK:                 %[[VAL_24:.*]] = fir.convert %[[VAL_22]] : (!fir.logical<4>) -> i1
+! CHECK:                 %[[VAL_25:.*]] = arith.cmpi eq, %[[VAL_23]], %[[VAL_24]] : i1
+! CHECK:                 %[[VAL_26:.*]] = fir.convert %[[VAL_25]] : (i1) -> !fir.logical<4>
+! CHECK:                 hlfir.assign %[[VAL_26]] to %[[VAL_17]]#0 : !fir.logical<4>, !fir.ref<!fir.logical<4>>
+! CHECK:                 omp.yield
+! CHECK:               omp.terminator
 ! CHECK:             omp.terminator
 ! CHECK:           return
 
@@ -146,42 +150,44 @@ end subroutine
 ! CHECK:             %[[VAL_20:.*]] = arith.constant 1 : i32
 ! CHECK:             %[[VAL_21:.*]] = arith.constant 100 : i32
 ! CHECK:             %[[VAL_22:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop byref reduction(@eqv_reduction %[[VAL_7]]#0 -> %[[VAL_23:.*]] : !fir.ref<!fir.logical<4>>, @eqv_reduction %[[VAL_9]]#0 -> %[[VAL_24:.*]] : !fir.ref<!fir.logical<4>>, @eqv_reduction %[[VAL_11]]#0 -> %[[VAL_25:.*]] : !fir.ref<!fir.logical<4>>)  for  (%[[VAL_26:.*]]) : i32 = (%[[VAL_20]]) to (%[[VAL_21]]) inclusive step (%[[VAL_22]]) {
-! CHECK:               fir.store %[[VAL_26]] to %[[VAL_19]]#1 : !fir.ref<i32>
-! CHECK:               %[[VAL_27:.*]]:2 = hlfir.declare %[[VAL_23]] {uniq_name = "_QFmultiple_reductionsEx"} : (!fir.ref<!fir.logical<4>>) -> (!fir.ref<!fir.logical<4>>, !fir.ref<!fir.logical<4>>)
-! CHECK:               %[[VAL_28:.*]]:2 = hlfir.declare %[[VAL_24]] {uniq_name = "_QFmultiple_reductionsEy"} : (!fir.ref<!fir.logical<4>>) -> (!fir.ref<!fir.logical<4>>, !fir.ref<!fir.logical<4>>)
-! CHECK:               %[[VAL_29:.*]]:2 = hlfir.declare %[[VAL_25]] {uniq_name = "_QFmultiple_reductionsEz"} : (!fir.ref<!fir.logical<4>>) -> (!fir.ref<!fir.logical<4>>, !fir.ref<!fir.logical<4>>)
-! CHECK:               %[[VAL_30:.*]] = fir.load %[[VAL_27]]#0 : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_31:.*]] = fir.load %[[VAL_19]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_32:.*]] = fir.convert %[[VAL_31]] : (i32) -> i64
-! CHECK:               %[[VAL_33:.*]] = hlfir.designate %[[VAL_5]]#0 (%[[VAL_32]])  : (!fir.ref<!fir.array<100x!fir.logical<4>>>, i64) -> !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_34:.*]] = fir.load %[[VAL_33]] : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_35:.*]] = fir.convert %[[VAL_30]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_36:.*]] = fir.convert %[[VAL_34]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_37:.*]] = arith.cmpi eq, %[[VAL_35]], %[[VAL_36]] : i1
-! CHECK:               %[[VAL_38:.*]] = fir.convert %[[VAL_37]] : (i1) -> !fir.logical<4>
-! CHECK:               hlfir.assign %[[VAL_38]] to %[[VAL_27]]#0 : !fir.logical<4>, !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_39:.*]] = fir.load %[[VAL_28]]#0 : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_40:.*]] = fir.load %[[VAL_19]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_41:.*]] = fir.convert %[[VAL_40]] : (i32) -> i64
-! CHECK:               %[[VAL_42:.*]] = hlfir.designate %[[VAL_5]]#0 (%[[VAL_41]])  : (!fir.ref<!fir.array<100x!fir.logical<4>>>, i64) -> !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_43:.*]] = fir.load %[[VAL_42]] : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_44:.*]] = fir.convert %[[VAL_39]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_45:.*]] = fir.convert %[[VAL_43]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_46:.*]] = arith.cmpi eq, %[[VAL_44]], %[[VAL_45]] : i1
-! CHECK:               %[[VAL_47:.*]] = fir.convert %[[VAL_46]] : (i1) -> !fir.logical<4>
-! CHECK:               hlfir.assign %[[VAL_47]] to %[[VAL_28]]#0 : !fir.logical<4>, !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_48:.*]] = fir.load %[[VAL_29]]#0 : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_49:.*]] = fir.load %[[VAL_19]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_50:.*]] = fir.convert %[[VAL_49]] : (i32) -> i64
-! CHECK:               %[[VAL_51:.*]] = hlfir.designate %[[VAL_5]]#0 (%[[VAL_50]])  : (!fir.ref<!fir.array<100x!fir.logical<4>>>, i64) -> !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_52:.*]] = fir.load %[[VAL_51]] : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_53:.*]] = fir.convert %[[VAL_48]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_54:.*]] = fir.convert %[[VAL_52]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_55:.*]] = arith.cmpi eq, %[[VAL_53]], %[[VAL_54]] : i1
-! CHECK:               %[[VAL_56:.*]] = fir.convert %[[VAL_55]] : (i1) -> !fir.logical<4>
-! CHECK:               hlfir.assign %[[VAL_56]] to %[[VAL_29]]#0 : !fir.logical<4>, !fir.ref<!fir.logical<4>>
-! CHECK:               omp.yield
+! CHECK:             omp.wsloop byref reduction(@eqv_reduction %[[VAL_7]]#0 -> %[[VAL_23:.*]] : !fir.ref<!fir.logical<4>>, @eqv_reduction %[[VAL_9]]#0 -> %[[VAL_24:.*]] : !fir.ref<!fir.logical<4>>, @eqv_reduction %[[VAL_11]]#0 -> %[[VAL_25:.*]] : !fir.ref<!fir.logical<4>>) {
+! CHECK-NEXT:          omp.loop_nest (%[[VAL_26:.*]]) : i32 = (%[[VAL_20]]) to (%[[VAL_21]]) inclusive step (%[[VAL_22]]) {
+! CHECK:                 %[[VAL_27:.*]]:2 = hlfir.declare %[[VAL_23]] {uniq_name = "_QFmultiple_reductionsEx"} : (!fir.ref<!fir.logical<4>>) -> (!fir.ref<!fir.logical<4>>, !fir.ref<!fir.logical<4>>)
+! CHECK:                 %[[VAL_28:.*]]:2 = hlfir.declare %[[VAL_24]] {uniq_name = "_QFmultiple_reductionsEy"} : (!fir.ref<!fir.logical<4>>) -> (!fir.ref<!fir.logical<4>>, !fir.ref<!fir.logical<4>>)
+! CHECK:                 %[[VAL_29:.*]]:2 = hlfir.declare %[[VAL_25]] {uniq_name = "_QFmultiple_reductionsEz"} : (!fir.ref<!fir.logical<4>>) -> (!fir.ref<!fir.logical<4>>, !fir.ref<!fir.logical<4>>)
+! CHECK:                 fir.store %[[VAL_26]] to %[[VAL_19]]#1 : !fir.ref<i32>
+! CHECK:                 %[[VAL_30:.*]] = fir.load %[[VAL_27]]#0 : !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_31:.*]] = fir.load %[[VAL_19]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_32:.*]] = fir.convert %[[VAL_31]] : (i32) -> i64
+! CHECK:                 %[[VAL_33:.*]] = hlfir.designate %[[VAL_5]]#0 (%[[VAL_32]])  : (!fir.ref<!fir.array<100x!fir.logical<4>>>, i64) -> !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_34:.*]] = fir.load %[[VAL_33]] : !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_35:.*]] = fir.convert %[[VAL_30]] : (!fir.logical<4>) -> i1
+! CHECK:                 %[[VAL_36:.*]] = fir.convert %[[VAL_34]] : (!fir.logical<4>) -> i1
+! CHECK:                 %[[VAL_37:.*]] = arith.cmpi eq, %[[VAL_35]], %[[VAL_36]] : i1
+! CHECK:                 %[[VAL_38:.*]] = fir.convert %[[VAL_37]] : (i1) -> !fir.logical<4>
+! CHECK:                 hlfir.assign %[[VAL_38]] to %[[VAL_27]]#0 : !fir.logical<4>, !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_39:.*]] = fir.load %[[VAL_28]]#0 : !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_40:.*]] = fir.load %[[VAL_19]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_41:.*]] = fir.convert %[[VAL_40]] : (i32) -> i64
+! CHECK:                 %[[VAL_42:.*]] = hlfir.designate %[[VAL_5]]#0 (%[[VAL_41]])  : (!fir.ref<!fir.array<100x!fir.logical<4>>>, i64) -> !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_43:.*]] = fir.load %[[VAL_42]] : !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_44:.*]] = fir.convert %[[VAL_39]] : (!fir.logical<4>) -> i1
+! CHECK:                 %[[VAL_45:.*]] = fir.convert %[[VAL_43]] : (!fir.logical<4>) -> i1
+! CHECK:                 %[[VAL_46:.*]] = arith.cmpi eq, %[[VAL_44]], %[[VAL_45]] : i1
+! CHECK:                 %[[VAL_47:.*]] = fir.convert %[[VAL_46]] : (i1) -> !fir.logical<4>
+! CHECK:                 hlfir.assign %[[VAL_47]] to %[[VAL_28]]#0 : !fir.logical<4>, !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_48:.*]] = fir.load %[[VAL_29]]#0 : !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_49:.*]] = fir.load %[[VAL_19]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_50:.*]] = fir.convert %[[VAL_49]] : (i32) -> i64
+! CHECK:                 %[[VAL_51:.*]] = hlfir.designate %[[VAL_5]]#0 (%[[VAL_50]])  : (!fir.ref<!fir.array<100x!fir.logical<4>>>, i64) -> !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_52:.*]] = fir.load %[[VAL_51]] : !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_53:.*]] = fir.convert %[[VAL_48]] : (!fir.logical<4>) -> i1
+! CHECK:                 %[[VAL_54:.*]] = fir.convert %[[VAL_52]] : (!fir.logical<4>) -> i1
+! CHECK:                 %[[VAL_55:.*]] = arith.cmpi eq, %[[VAL_53]], %[[VAL_54]] : i1
+! CHECK:                 %[[VAL_56:.*]] = fir.convert %[[VAL_55]] : (i1) -> !fir.logical<4>
+! CHECK:                 hlfir.assign %[[VAL_56]] to %[[VAL_29]]#0 : !fir.logical<4>, !fir.ref<!fir.logical<4>>
+! CHECK:                 omp.yield
+! CHECK:               omp.terminator
 ! CHECK:             omp.terminator
 ! CHECK:           return
 
diff --git a/flang/test/Lower/OpenMP/wsloop-reduction-logical-eqv.f90 b/flang/test/Lower/OpenMP/wsloop-reduction-logical-eqv.f90
index a1bfa462cd59..1021b5926b91 100644
--- a/flang/test/Lower/OpenMP/wsloop-reduction-logical-eqv.f90
+++ b/flang/test/Lower/OpenMP/wsloop-reduction-logical-eqv.f90
@@ -36,20 +36,22 @@
 ! CHECK:             %[[VAL_12:.*]] = arith.constant 1 : i32
 ! CHECK:             %[[VAL_13:.*]] = arith.constant 100 : i32
 ! CHECK:             %[[VAL_14:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop reduction(@eqv_reduction %[[VAL_4]]#0 -> %[[VAL_15:.*]] : !fir.ref<!fir.logical<4>>)  for  (%[[VAL_16:.*]]) : i32 = (%[[VAL_12]]) to (%[[VAL_13]]) inclusive step (%[[VAL_14]]) {
-! CHECK:               fir.store %[[VAL_16]] to %[[VAL_11]]#1 : !fir.ref<i32>
-! CHECK:               %[[VAL_17:.*]]:2 = hlfir.declare %[[VAL_15]] {uniq_name = "_QFsimple_reductionEx"} : (!fir.ref<!fir.logical<4>>) -> (!fir.ref<!fir.logical<4>>, !fir.ref<!fir.logical<4>>)
-! CHECK:               %[[VAL_18:.*]] = fir.load %[[VAL_17]]#0 : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_19:.*]] = fir.load %[[VAL_11]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_20:.*]] = fir.convert %[[VAL_19]] : (i32) -> i64
-! CHECK:               %[[VAL_21:.*]] = hlfir.designate %[[VAL_7]]#0 (%[[VAL_20]])  : (!fir.ref<!fir.array<100x!fir.logical<4>>>, i64) -> !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_22:.*]] = fir.load %[[VAL_21]] : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_23:.*]] = fir.convert %[[VAL_18]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_24:.*]] = fir.convert %[[VAL_22]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_25:.*]] = arith.cmpi eq, %[[VAL_23]], %[[VAL_24]] : i1
-! CHECK:               %[[VAL_26:.*]] = fir.convert %[[VAL_25]] : (i1) -> !fir.logical<4>
-! CHECK:               hlfir.assign %[[VAL_26]] to %[[VAL_17]]#0 : !fir.logical<4>, !fir.ref<!fir.logical<4>>
-! CHECK:               omp.yield
+! CHECK:             omp.wsloop reduction(@eqv_reduction %[[VAL_4]]#0 -> %[[VAL_15:.*]] : !fir.ref<!fir.logical<4>>) {
+! CHECK-NEXT:          omp.loop_nest (%[[VAL_16:.*]]) : i32 = (%[[VAL_12]]) to (%[[VAL_13]]) inclusive step (%[[VAL_14]]) {
+! CHECK:                 %[[VAL_17:.*]]:2 = hlfir.declare %[[VAL_15]] {uniq_name = "_QFsimple_reductionEx"} : (!fir.ref<!fir.logical<4>>) -> (!fir.ref<!fir.logical<4>>, !fir.ref<!fir.logical<4>>)
+! CHECK:                 fir.store %[[VAL_16]] to %[[VAL_11]]#1 : !fir.ref<i32>
+! CHECK:                 %[[VAL_18:.*]] = fir.load %[[VAL_17]]#0 : !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_19:.*]] = fir.load %[[VAL_11]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_20:.*]] = fir.convert %[[VAL_19]] : (i32) -> i64
+! CHECK:                 %[[VAL_21:.*]] = hlfir.designate %[[VAL_7]]#0 (%[[VAL_20]])  : (!fir.ref<!fir.array<100x!fir.logical<4>>>, i64) -> !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_22:.*]] = fir.load %[[VAL_21]] : !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_23:.*]] = fir.convert %[[VAL_18]] : (!fir.logical<4>) -> i1
+! CHECK:                 %[[VAL_24:.*]] = fir.convert %[[VAL_22]] : (!fir.logical<4>) -> i1
+! CHECK:                 %[[VAL_25:.*]] = arith.cmpi eq, %[[VAL_23]], %[[VAL_24]] : i1
+! CHECK:                 %[[VAL_26:.*]] = fir.convert %[[VAL_25]] : (i1) -> !fir.logical<4>
+! CHECK:                 hlfir.assign %[[VAL_26]] to %[[VAL_17]]#0 : !fir.logical<4>, !fir.ref<!fir.logical<4>>
+! CHECK:                 omp.yield
+! CHECK:               omp.terminator
 ! CHECK:             omp.terminator
 ! CHECK:           return
 
@@ -83,20 +85,22 @@ end subroutine
 ! CHECK:             %[[VAL_12:.*]] = arith.constant 1 : i32
 ! CHECK:             %[[VAL_13:.*]] = arith.constant 100 : i32
 ! CHECK:             %[[VAL_14:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop reduction(@eqv_reduction %[[VAL_4]]#0 -> %[[VAL_15:.*]] : !fir.ref<!fir.logical<4>>)  for  (%[[VAL_16:.*]]) : i32 = (%[[VAL_12]]) to (%[[VAL_13]]) inclusive step (%[[VAL_14]]) {
-! CHECK:               fir.store %[[VAL_16]] to %[[VAL_11]]#1 : !fir.ref<i32>
-! CHECK:               %[[VAL_17:.*]]:2 = hlfir.declare %[[VAL_15]] {uniq_name = "_QFsimple_reduction_switch_orderEx"} : (!fir.ref<!fir.logical<4>>) -> (!fir.ref<!fir.logical<4>>, !fir.ref<!fir.logical<4>>)
-! CHECK:               %[[VAL_18:.*]] = fir.load %[[VAL_11]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_19:.*]] = fir.convert %[[VAL_18]] : (i32) -> i64
-! CHECK:               %[[VAL_20:.*]] = hlfir.designate %[[VAL_7]]#0 (%[[VAL_19]])  : (!fir.ref<!fir.array<100x!fir.logical<4>>>, i64) -> !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_21:.*]] = fir.load %[[VAL_20]] : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_22:.*]] = fir.load %[[VAL_17]]#0 : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_23:.*]] = fir.convert %[[VAL_21]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_24:.*]] = fir.convert %[[VAL_22]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_25:.*]] = arith.cmpi eq, %[[VAL_23]], %[[VAL_24]] : i1
-! CHECK:               %[[VAL_26:.*]] = fir.convert %[[VAL_25]] : (i1) -> !fir.logical<4>
-! CHECK:               hlfir.assign %[[VAL_26]] to %[[VAL_17]]#0 : !fir.logical<4>, !fir.ref<!fir.logical<4>>
-! CHECK:               omp.yield
+! CHECK:             omp.wsloop reduction(@eqv_reduction %[[VAL_4]]#0 -> %[[VAL_15:.*]] : !fir.ref<!fir.logical<4>>) {
+! CHECK-NEXT:          omp.loop_nest (%[[VAL_16:.*]]) : i32 = (%[[VAL_12]]) to (%[[VAL_13]]) inclusive step (%[[VAL_14]]) {
+! CHECK:                 %[[VAL_17:.*]]:2 = hlfir.declare %[[VAL_15]] {uniq_name = "_QFsimple_reduction_switch_orderEx"} : (!fir.ref<!fir.logical<4>>) -> (!fir.ref<!fir.logical<4>>, !fir.ref<!fir.logical<4>>)
+! CHECK:                 fir.store %[[VAL_16]] to %[[VAL_11]]#1 : !fir.ref<i32>
+! CHECK:                 %[[VAL_18:.*]] = fir.load %[[VAL_11]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_19:.*]] = fir.convert %[[VAL_18]] : (i32) -> i64
+! CHECK:                 %[[VAL_20:.*]] = hlfir.designate %[[VAL_7]]#0 (%[[VAL_19]])  : (!fir.ref<!fir.array<100x!fir.logical<4>>>, i64) -> !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_21:.*]] = fir.load %[[VAL_20]] : !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_22:.*]] = fir.load %[[VAL_17]]#0 : !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_23:.*]] = fir.convert %[[VAL_21]] : (!fir.logical<4>) -> i1
+! CHECK:                 %[[VAL_24:.*]] = fir.convert %[[VAL_22]] : (!fir.logical<4>) -> i1
+! CHECK:                 %[[VAL_25:.*]] = arith.cmpi eq, %[[VAL_23]], %[[VAL_24]] : i1
+! CHECK:                 %[[VAL_26:.*]] = fir.convert %[[VAL_25]] : (i1) -> !fir.logical<4>
+! CHECK:                 hlfir.assign %[[VAL_26]] to %[[VAL_17]]#0 : !fir.logical<4>, !fir.ref<!fir.logical<4>>
+! CHECK:                 omp.yield
+! CHECK:               omp.terminator
 ! CHECK:             omp.terminator
 ! CHECK:           return
 
@@ -140,42 +144,44 @@ end subroutine
 ! CHECK:             %[[VAL_20:.*]] = arith.constant 1 : i32
 ! CHECK:             %[[VAL_21:.*]] = arith.constant 100 : i32
 ! CHECK:             %[[VAL_22:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop reduction(@eqv_reduction %[[VAL_7]]#0 -> %[[VAL_23:.*]] : !fir.ref<!fir.logical<4>>, @eqv_reduction %[[VAL_9]]#0 -> %[[VAL_24:.*]] : !fir.ref<!fir.logical<4>>, @eqv_reduction %[[VAL_11]]#0 -> %[[VAL_25:.*]] : !fir.ref<!fir.logical<4>>)  for  (%[[VAL_26:.*]]) : i32 = (%[[VAL_20]]) to (%[[VAL_21]]) inclusive step (%[[VAL_22]]) {
-! CHECK:               fir.store %[[VAL_26]] to %[[VAL_19]]#1 : !fir.ref<i32>
-! CHECK:               %[[VAL_27:.*]]:2 = hlfir.declare %[[VAL_23]] {uniq_name = "_QFmultiple_reductionsEx"} : (!fir.ref<!fir.logical<4>>) -> (!fir.ref<!fir.logical<4>>, !fir.ref<!fir.logical<4>>)
-! CHECK:               %[[VAL_28:.*]]:2 = hlfir.declare %[[VAL_24]] {uniq_name = "_QFmultiple_reductionsEy"} : (!fir.ref<!fir.logical<4>>) -> (!fir.ref<!fir.logical<4>>, !fir.ref<!fir.logical<4>>)
-! CHECK:               %[[VAL_29:.*]]:2 = hlfir.declare %[[VAL_25]] {uniq_name = "_QFmultiple_reductionsEz"} : (!fir.ref<!fir.logical<4>>) -> (!fir.ref<!fir.logical<4>>, !fir.ref<!fir.logical<4>>)
-! CHECK:               %[[VAL_30:.*]] = fir.load %[[VAL_27]]#0 : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_31:.*]] = fir.load %[[VAL_19]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_32:.*]] = fir.convert %[[VAL_31]] : (i32) -> i64
-! CHECK:               %[[VAL_33:.*]] = hlfir.designate %[[VAL_5]]#0 (%[[VAL_32]])  : (!fir.ref<!fir.array<100x!fir.logical<4>>>, i64) -> !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_34:.*]] = fir.load %[[VAL_33]] : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_35:.*]] = fir.convert %[[VAL_30]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_36:.*]] = fir.convert %[[VAL_34]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_37:.*]] = arith.cmpi eq, %[[VAL_35]], %[[VAL_36]] : i1
-! CHECK:               %[[VAL_38:.*]] = fir.convert %[[VAL_37]] : (i1) -> !fir.logical<4>
-! CHECK:               hlfir.assign %[[VAL_38]] to %[[VAL_27]]#0 : !fir.logical<4>, !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_39:.*]] = fir.load %[[VAL_28]]#0 : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_40:.*]] = fir.load %[[VAL_19]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_41:.*]] = fir.convert %[[VAL_40]] : (i32) -> i64
-! CHECK:               %[[VAL_42:.*]] = hlfir.designate %[[VAL_5]]#0 (%[[VAL_41]])  : (!fir.ref<!fir.array<100x!fir.logical<4>>>, i64) -> !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_43:.*]] = fir.load %[[VAL_42]] : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_44:.*]] = fir.convert %[[VAL_39]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_45:.*]] = fir.convert %[[VAL_43]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_46:.*]] = arith.cmpi eq, %[[VAL_44]], %[[VAL_45]] : i1
-! CHECK:               %[[VAL_47:.*]] = fir.convert %[[VAL_46]] : (i1) -> !fir.logical<4>
-! CHECK:               hlfir.assign %[[VAL_47]] to %[[VAL_28]]#0 : !fir.logical<4>, !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_48:.*]] = fir.load %[[VAL_29]]#0 : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_49:.*]] = fir.load %[[VAL_19]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_50:.*]] = fir.convert %[[VAL_49]] : (i32) -> i64
-! CHECK:               %[[VAL_51:.*]] = hlfir.designate %[[VAL_5]]#0 (%[[VAL_50]])  : (!fir.ref<!fir.array<100x!fir.logical<4>>>, i64) -> !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_52:.*]] = fir.load %[[VAL_51]] : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_53:.*]] = fir.convert %[[VAL_48]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_54:.*]] = fir.convert %[[VAL_52]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_55:.*]] = arith.cmpi eq, %[[VAL_53]], %[[VAL_54]] : i1
-! CHECK:               %[[VAL_56:.*]] = fir.convert %[[VAL_55]] : (i1) -> !fir.logical<4>
-! CHECK:               hlfir.assign %[[VAL_56]] to %[[VAL_29]]#0 : !fir.logical<4>, !fir.ref<!fir.logical<4>>
-! CHECK:               omp.yield
+! CHECK:             omp.wsloop reduction(@eqv_reduction %[[VAL_7]]#0 -> %[[VAL_23:.*]] : !fir.ref<!fir.logical<4>>, @eqv_reduction %[[VAL_9]]#0 -> %[[VAL_24:.*]] : !fir.ref<!fir.logical<4>>, @eqv_reduction %[[VAL_11]]#0 -> %[[VAL_25:.*]] : !fir.ref<!fir.logical<4>>) {
+! CHECK-NEXT:          omp.loop_nest (%[[VAL_26:.*]]) : i32 = (%[[VAL_20]]) to (%[[VAL_21]]) inclusive step (%[[VAL_22]]) {
+! CHECK:                 %[[VAL_27:.*]]:2 = hlfir.declare %[[VAL_23]] {uniq_name = "_QFmultiple_reductionsEx"} : (!fir.ref<!fir.logical<4>>) -> (!fir.ref<!fir.logical<4>>, !fir.ref<!fir.logical<4>>)
+! CHECK:                 %[[VAL_28:.*]]:2 = hlfir.declare %[[VAL_24]] {uniq_name = "_QFmultiple_reductionsEy"} : (!fir.ref<!fir.logical<4>>) -> (!fir.ref<!fir.logical<4>>, !fir.ref<!fir.logical<4>>)
+! CHECK:                 %[[VAL_29:.*]]:2 = hlfir.declare %[[VAL_25]] {uniq_name = "_QFmultiple_reductionsEz"} : (!fir.ref<!fir.logical<4>>) -> (!fir.ref<!fir.logical<4>>, !fir.ref<!fir.logical<4>>)
+! CHECK:                 fir.store %[[VAL_26]] to %[[VAL_19]]#1 : !fir.ref<i32>
+! CHECK:                 %[[VAL_30:.*]] = fir.load %[[VAL_27]]#0 : !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_31:.*]] = fir.load %[[VAL_19]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_32:.*]] = fir.convert %[[VAL_31]] : (i32) -> i64
+! CHECK:                 %[[VAL_33:.*]] = hlfir.designate %[[VAL_5]]#0 (%[[VAL_32]])  : (!fir.ref<!fir.array<100x!fir.logical<4>>>, i64) -> !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_34:.*]] = fir.load %[[VAL_33]] : !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_35:.*]] = fir.convert %[[VAL_30]] : (!fir.logical<4>) -> i1
+! CHECK:                 %[[VAL_36:.*]] = fir.convert %[[VAL_34]] : (!fir.logical<4>) -> i1
+! CHECK:                 %[[VAL_37:.*]] = arith.cmpi eq, %[[VAL_35]], %[[VAL_36]] : i1
+! CHECK:                 %[[VAL_38:.*]] = fir.convert %[[VAL_37]] : (i1) -> !fir.logical<4>
+! CHECK:                 hlfir.assign %[[VAL_38]] to %[[VAL_27]]#0 : !fir.logical<4>, !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_39:.*]] = fir.load %[[VAL_28]]#0 : !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_40:.*]] = fir.load %[[VAL_19]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_41:.*]] = fir.convert %[[VAL_40]] : (i32) -> i64
+! CHECK:                 %[[VAL_42:.*]] = hlfir.designate %[[VAL_5]]#0 (%[[VAL_41]])  : (!fir.ref<!fir.array<100x!fir.logical<4>>>, i64) -> !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_43:.*]] = fir.load %[[VAL_42]] : !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_44:.*]] = fir.convert %[[VAL_39]] : (!fir.logical<4>) -> i1
+! CHECK:                 %[[VAL_45:.*]] = fir.convert %[[VAL_43]] : (!fir.logical<4>) -> i1
+! CHECK:                 %[[VAL_46:.*]] = arith.cmpi eq, %[[VAL_44]], %[[VAL_45]] : i1
+! CHECK:                 %[[VAL_47:.*]] = fir.convert %[[VAL_46]] : (i1) -> !fir.logical<4>
+! CHECK:                 hlfir.assign %[[VAL_47]] to %[[VAL_28]]#0 : !fir.logical<4>, !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_48:.*]] = fir.load %[[VAL_29]]#0 : !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_49:.*]] = fir.load %[[VAL_19]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_50:.*]] = fir.convert %[[VAL_49]] : (i32) -> i64
+! CHECK:                 %[[VAL_51:.*]] = hlfir.designate %[[VAL_5]]#0 (%[[VAL_50]])  : (!fir.ref<!fir.array<100x!fir.logical<4>>>, i64) -> !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_52:.*]] = fir.load %[[VAL_51]] : !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_53:.*]] = fir.convert %[[VAL_48]] : (!fir.logical<4>) -> i1
+! CHECK:                 %[[VAL_54:.*]] = fir.convert %[[VAL_52]] : (!fir.logical<4>) -> i1
+! CHECK:                 %[[VAL_55:.*]] = arith.cmpi eq, %[[VAL_53]], %[[VAL_54]] : i1
+! CHECK:                 %[[VAL_56:.*]] = fir.convert %[[VAL_55]] : (i1) -> !fir.logical<4>
+! CHECK:                 hlfir.assign %[[VAL_56]] to %[[VAL_29]]#0 : !fir.logical<4>, !fir.ref<!fir.logical<4>>
+! CHECK:                 omp.yield
+! CHECK:               omp.terminator
 ! CHECK:             omp.terminator
 ! CHECK:           return
 
diff --git a/flang/test/Lower/OpenMP/wsloop-reduction-logical-neqv-byref.f90 b/flang/test/Lower/OpenMP/wsloop-reduction-logical-neqv-byref.f90
index a94b67a97832..854cb19ecd75 100644
--- a/flang/test/Lower/OpenMP/wsloop-reduction-logical-neqv-byref.f90
+++ b/flang/test/Lower/OpenMP/wsloop-reduction-logical-neqv-byref.f90
@@ -42,20 +42,22 @@
 ! CHECK:             %[[VAL_12:.*]] = arith.constant 1 : i32
 ! CHECK:             %[[VAL_13:.*]] = arith.constant 100 : i32
 ! CHECK:             %[[VAL_14:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop byref reduction(@neqv_reduction %[[VAL_4]]#0 -> %[[VAL_15:.*]] : !fir.ref<!fir.logical<4>>)  for  (%[[VAL_16:.*]]) : i32 = (%[[VAL_12]]) to (%[[VAL_13]]) inclusive step (%[[VAL_14]]) {
-! CHECK:               fir.store %[[VAL_16]] to %[[VAL_11]]#1 : !fir.ref<i32>
-! CHECK:               %[[VAL_17:.*]]:2 = hlfir.declare %[[VAL_15]] {uniq_name = "_QFsimple_reductionEx"} : (!fir.ref<!fir.logical<4>>) -> (!fir.ref<!fir.logical<4>>, !fir.ref<!fir.logical<4>>)
-! CHECK:               %[[VAL_18:.*]] = fir.load %[[VAL_17]]#0 : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_19:.*]] = fir.load %[[VAL_11]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_20:.*]] = fir.convert %[[VAL_19]] : (i32) -> i64
-! CHECK:               %[[VAL_21:.*]] = hlfir.designate %[[VAL_7]]#0 (%[[VAL_20]])  : (!fir.ref<!fir.array<100x!fir.logical<4>>>, i64) -> !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_22:.*]] = fir.load %[[VAL_21]] : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_23:.*]] = fir.convert %[[VAL_18]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_24:.*]] = fir.convert %[[VAL_22]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_25:.*]] = arith.cmpi ne, %[[VAL_23]], %[[VAL_24]] : i1
-! CHECK:               %[[VAL_26:.*]] = fir.convert %[[VAL_25]] : (i1) -> !fir.logical<4>
-! CHECK:               hlfir.assign %[[VAL_26]] to %[[VAL_17]]#0 : !fir.logical<4>, !fir.ref<!fir.logical<4>>
-! CHECK:               omp.yield
+! CHECK:             omp.wsloop byref reduction(@neqv_reduction %[[VAL_4]]#0 -> %[[VAL_15:.*]] : !fir.ref<!fir.logical<4>>) {
+! CHECK-NEXT:          omp.loop_nest (%[[VAL_16:.*]]) : i32 = (%[[VAL_12]]) to (%[[VAL_13]]) inclusive step (%[[VAL_14]]) {
+! CHECK:                 %[[VAL_17:.*]]:2 = hlfir.declare %[[VAL_15]] {uniq_name = "_QFsimple_reductionEx"} : (!fir.ref<!fir.logical<4>>) -> (!fir.ref<!fir.logical<4>>, !fir.ref<!fir.logical<4>>)
+! CHECK:                 fir.store %[[VAL_16]] to %[[VAL_11]]#1 : !fir.ref<i32>
+! CHECK:                 %[[VAL_18:.*]] = fir.load %[[VAL_17]]#0 : !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_19:.*]] = fir.load %[[VAL_11]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_20:.*]] = fir.convert %[[VAL_19]] : (i32) -> i64
+! CHECK:                 %[[VAL_21:.*]] = hlfir.designate %[[VAL_7]]#0 (%[[VAL_20]])  : (!fir.ref<!fir.array<100x!fir.logical<4>>>, i64) -> !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_22:.*]] = fir.load %[[VAL_21]] : !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_23:.*]] = fir.convert %[[VAL_18]] : (!fir.logical<4>) -> i1
+! CHECK:                 %[[VAL_24:.*]] = fir.convert %[[VAL_22]] : (!fir.logical<4>) -> i1
+! CHECK:                 %[[VAL_25:.*]] = arith.cmpi ne, %[[VAL_23]], %[[VAL_24]] : i1
+! CHECK:                 %[[VAL_26:.*]] = fir.convert %[[VAL_25]] : (i1) -> !fir.logical<4>
+! CHECK:                 hlfir.assign %[[VAL_26]] to %[[VAL_17]]#0 : !fir.logical<4>, !fir.ref<!fir.logical<4>>
+! CHECK:                 omp.yield
+! CHECK:               omp.terminator
 ! CHECK:             omp.terminator
 ! CHECK:           return
 
@@ -90,20 +92,22 @@ end subroutine
 ! CHECK:             %[[VAL_12:.*]] = arith.constant 1 : i32
 ! CHECK:             %[[VAL_13:.*]] = arith.constant 100 : i32
 ! CHECK:             %[[VAL_14:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop byref reduction(@neqv_reduction %[[VAL_4]]#0 -> %[[VAL_15:.*]] : !fir.ref<!fir.logical<4>>)  for  (%[[VAL_16:.*]]) : i32 = (%[[VAL_12]]) to (%[[VAL_13]]) inclusive step (%[[VAL_14]]) {
-! CHECK:               fir.store %[[VAL_16]] to %[[VAL_11]]#1 : !fir.ref<i32>
-! CHECK:               %[[VAL_17:.*]]:2 = hlfir.declare %[[VAL_15]] {uniq_name = "_QFsimple_reduction_switch_orderEx"} : (!fir.ref<!fir.logical<4>>) -> (!fir.ref<!fir.logical<4>>, !fir.ref<!fir.logical<4>>)
-! CHECK:               %[[VAL_18:.*]] = fir.load %[[VAL_11]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_19:.*]] = fir.convert %[[VAL_18]] : (i32) -> i64
-! CHECK:               %[[VAL_20:.*]] = hlfir.designate %[[VAL_7]]#0 (%[[VAL_19]])  : (!fir.ref<!fir.array<100x!fir.logical<4>>>, i64) -> !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_21:.*]] = fir.load %[[VAL_20]] : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_22:.*]] = fir.load %[[VAL_17]]#0 : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_23:.*]] = fir.convert %[[VAL_21]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_24:.*]] = fir.convert %[[VAL_22]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_25:.*]] = arith.cmpi ne, %[[VAL_23]], %[[VAL_24]] : i1
-! CHECK:               %[[VAL_26:.*]] = fir.convert %[[VAL_25]] : (i1) -> !fir.logical<4>
-! CHECK:               hlfir.assign %[[VAL_26]] to %[[VAL_17]]#0 : !fir.logical<4>, !fir.ref<!fir.logical<4>>
-! CHECK:               omp.yield
+! CHECK:             omp.wsloop byref reduction(@neqv_reduction %[[VAL_4]]#0 -> %[[VAL_15:.*]] : !fir.ref<!fir.logical<4>>) {
+! CHECK-NEXT:          omp.loop_nest (%[[VAL_16:.*]]) : i32 = (%[[VAL_12]]) to (%[[VAL_13]]) inclusive step (%[[VAL_14]]) {
+! CHECK:                 %[[VAL_17:.*]]:2 = hlfir.declare %[[VAL_15]] {uniq_name = "_QFsimple_reduction_switch_orderEx"} : (!fir.ref<!fir.logical<4>>) -> (!fir.ref<!fir.logical<4>>, !fir.ref<!fir.logical<4>>)
+! CHECK:                 fir.store %[[VAL_16]] to %[[VAL_11]]#1 : !fir.ref<i32>
+! CHECK:                 %[[VAL_18:.*]] = fir.load %[[VAL_11]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_19:.*]] = fir.convert %[[VAL_18]] : (i32) -> i64
+! CHECK:                 %[[VAL_20:.*]] = hlfir.designate %[[VAL_7]]#0 (%[[VAL_19]])  : (!fir.ref<!fir.array<100x!fir.logical<4>>>, i64) -> !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_21:.*]] = fir.load %[[VAL_20]] : !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_22:.*]] = fir.load %[[VAL_17]]#0 : !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_23:.*]] = fir.convert %[[VAL_21]] : (!fir.logical<4>) -> i1
+! CHECK:                 %[[VAL_24:.*]] = fir.convert %[[VAL_22]] : (!fir.logical<4>) -> i1
+! CHECK:                 %[[VAL_25:.*]] = arith.cmpi ne, %[[VAL_23]], %[[VAL_24]] : i1
+! CHECK:                 %[[VAL_26:.*]] = fir.convert %[[VAL_25]] : (i1) -> !fir.logical<4>
+! CHECK:                 hlfir.assign %[[VAL_26]] to %[[VAL_17]]#0 : !fir.logical<4>, !fir.ref<!fir.logical<4>>
+! CHECK:                 omp.yield
+! CHECK:               omp.terminator
 ! CHECK:             omp.terminator
 ! CHECK:           return
 
@@ -149,42 +153,44 @@ end subroutine
 ! CHECK:             %[[VAL_20:.*]] = arith.constant 1 : i32
 ! CHECK:             %[[VAL_21:.*]] = arith.constant 100 : i32
 ! CHECK:             %[[VAL_22:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop byref reduction(@neqv_reduction %[[VAL_7]]#0 -> %[[VAL_23:.*]] : !fir.ref<!fir.logical<4>>, @neqv_reduction %[[VAL_9]]#0 -> %[[VAL_24:.*]] : !fir.ref<!fir.logical<4>>, @neqv_reduction %[[VAL_11]]#0 -> %[[VAL_25:.*]] : !fir.ref<!fir.logical<4>>)  for  (%[[VAL_26:.*]]) : i32 = (%[[VAL_20]]) to (%[[VAL_21]]) inclusive step (%[[VAL_22]]) {
-! CHECK:               fir.store %[[VAL_26]] to %[[VAL_19]]#1 : !fir.ref<i32>
-! CHECK:               %[[VAL_27:.*]]:2 = hlfir.declare %[[VAL_23]] {uniq_name = "_QFmultiple_reductionsEx"} : (!fir.ref<!fir.logical<4>>) -> (!fir.ref<!fir.logical<4>>, !fir.ref<!fir.logical<4>>)
-! CHECK:               %[[VAL_28:.*]]:2 = hlfir.declare %[[VAL_24]] {uniq_name = "_QFmultiple_reductionsEy"} : (!fir.ref<!fir.logical<4>>) -> (!fir.ref<!fir.logical<4>>, !fir.ref<!fir.logical<4>>)
-! CHECK:               %[[VAL_29:.*]]:2 = hlfir.declare %[[VAL_25]] {uniq_name = "_QFmultiple_reductionsEz"} : (!fir.ref<!fir.logical<4>>) -> (!fir.ref<!fir.logical<4>>, !fir.ref<!fir.logical<4>>)
-! CHECK:               %[[VAL_30:.*]] = fir.load %[[VAL_27]]#0 : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_31:.*]] = fir.load %[[VAL_19]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_32:.*]] = fir.convert %[[VAL_31]] : (i32) -> i64
-! CHECK:               %[[VAL_33:.*]] = hlfir.designate %[[VAL_5]]#0 (%[[VAL_32]])  : (!fir.ref<!fir.array<100x!fir.logical<4>>>, i64) -> !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_34:.*]] = fir.load %[[VAL_33]] : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_35:.*]] = fir.convert %[[VAL_30]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_36:.*]] = fir.convert %[[VAL_34]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_37:.*]] = arith.cmpi ne, %[[VAL_35]], %[[VAL_36]] : i1
-! CHECK:               %[[VAL_38:.*]] = fir.convert %[[VAL_37]] : (i1) -> !fir.logical<4>
-! CHECK:               hlfir.assign %[[VAL_38]] to %[[VAL_27]]#0 : !fir.logical<4>, !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_39:.*]] = fir.load %[[VAL_28]]#0 : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_40:.*]] = fir.load %[[VAL_19]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_41:.*]] = fir.convert %[[VAL_40]] : (i32) -> i64
-! CHECK:               %[[VAL_42:.*]] = hlfir.designate %[[VAL_5]]#0 (%[[VAL_41]])  : (!fir.ref<!fir.array<100x!fir.logical<4>>>, i64) -> !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_43:.*]] = fir.load %[[VAL_42]] : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_44:.*]] = fir.convert %[[VAL_39]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_45:.*]] = fir.convert %[[VAL_43]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_46:.*]] = arith.cmpi ne, %[[VAL_44]], %[[VAL_45]] : i1
-! CHECK:               %[[VAL_47:.*]] = fir.convert %[[VAL_46]] : (i1) -> !fir.logical<4>
-! CHECK:               hlfir.assign %[[VAL_47]] to %[[VAL_28]]#0 : !fir.logical<4>, !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_48:.*]] = fir.load %[[VAL_29]]#0 : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_49:.*]] = fir.load %[[VAL_19]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_50:.*]] = fir.convert %[[VAL_49]] : (i32) -> i64
-! CHECK:               %[[VAL_51:.*]] = hlfir.designate %[[VAL_5]]#0 (%[[VAL_50]])  : (!fir.ref<!fir.array<100x!fir.logical<4>>>, i64) -> !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_52:.*]] = fir.load %[[VAL_51]] : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_53:.*]] = fir.convert %[[VAL_48]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_54:.*]] = fir.convert %[[VAL_52]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_55:.*]] = arith.cmpi ne, %[[VAL_53]], %[[VAL_54]] : i1
-! CHECK:               %[[VAL_56:.*]] = fir.convert %[[VAL_55]] : (i1) -> !fir.logical<4>
-! CHECK:               hlfir.assign %[[VAL_56]] to %[[VAL_29]]#0 : !fir.logical<4>, !fir.ref<!fir.logical<4>>
-! CHECK:               omp.yield
+! CHECK:             omp.wsloop byref reduction(@neqv_reduction %[[VAL_7]]#0 -> %[[VAL_23:.*]] : !fir.ref<!fir.logical<4>>, @neqv_reduction %[[VAL_9]]#0 -> %[[VAL_24:.*]] : !fir.ref<!fir.logical<4>>, @neqv_reduction %[[VAL_11]]#0 -> %[[VAL_25:.*]] : !fir.ref<!fir.logical<4>>) {
+! CHECK-NEXT:          omp.loop_nest (%[[VAL_26:.*]]) : i32 = (%[[VAL_20]]) to (%[[VAL_21]]) inclusive step (%[[VAL_22]]) {
+! CHECK:                 %[[VAL_27:.*]]:2 = hlfir.declare %[[VAL_23]] {uniq_name = "_QFmultiple_reductionsEx"} : (!fir.ref<!fir.logical<4>>) -> (!fir.ref<!fir.logical<4>>, !fir.ref<!fir.logical<4>>)
+! CHECK:                 %[[VAL_28:.*]]:2 = hlfir.declare %[[VAL_24]] {uniq_name = "_QFmultiple_reductionsEy"} : (!fir.ref<!fir.logical<4>>) -> (!fir.ref<!fir.logical<4>>, !fir.ref<!fir.logical<4>>)
+! CHECK:                 %[[VAL_29:.*]]:2 = hlfir.declare %[[VAL_25]] {uniq_name = "_QFmultiple_reductionsEz"} : (!fir.ref<!fir.logical<4>>) -> (!fir.ref<!fir.logical<4>>, !fir.ref<!fir.logical<4>>)
+! CHECK:                 fir.store %[[VAL_26]] to %[[VAL_19]]#1 : !fir.ref<i32>
+! CHECK:                 %[[VAL_30:.*]] = fir.load %[[VAL_27]]#0 : !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_31:.*]] = fir.load %[[VAL_19]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_32:.*]] = fir.convert %[[VAL_31]] : (i32) -> i64
+! CHECK:                 %[[VAL_33:.*]] = hlfir.designate %[[VAL_5]]#0 (%[[VAL_32]])  : (!fir.ref<!fir.array<100x!fir.logical<4>>>, i64) -> !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_34:.*]] = fir.load %[[VAL_33]] : !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_35:.*]] = fir.convert %[[VAL_30]] : (!fir.logical<4>) -> i1
+! CHECK:                 %[[VAL_36:.*]] = fir.convert %[[VAL_34]] : (!fir.logical<4>) -> i1
+! CHECK:                 %[[VAL_37:.*]] = arith.cmpi ne, %[[VAL_35]], %[[VAL_36]] : i1
+! CHECK:                 %[[VAL_38:.*]] = fir.convert %[[VAL_37]] : (i1) -> !fir.logical<4>
+! CHECK:                 hlfir.assign %[[VAL_38]] to %[[VAL_27]]#0 : !fir.logical<4>, !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_39:.*]] = fir.load %[[VAL_28]]#0 : !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_40:.*]] = fir.load %[[VAL_19]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_41:.*]] = fir.convert %[[VAL_40]] : (i32) -> i64
+! CHECK:                 %[[VAL_42:.*]] = hlfir.designate %[[VAL_5]]#0 (%[[VAL_41]])  : (!fir.ref<!fir.array<100x!fir.logical<4>>>, i64) -> !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_43:.*]] = fir.load %[[VAL_42]] : !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_44:.*]] = fir.convert %[[VAL_39]] : (!fir.logical<4>) -> i1
+! CHECK:                 %[[VAL_45:.*]] = fir.convert %[[VAL_43]] : (!fir.logical<4>) -> i1
+! CHECK:                 %[[VAL_46:.*]] = arith.cmpi ne, %[[VAL_44]], %[[VAL_45]] : i1
+! CHECK:                 %[[VAL_47:.*]] = fir.convert %[[VAL_46]] : (i1) -> !fir.logical<4>
+! CHECK:                 hlfir.assign %[[VAL_47]] to %[[VAL_28]]#0 : !fir.logical<4>, !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_48:.*]] = fir.load %[[VAL_29]]#0 : !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_49:.*]] = fir.load %[[VAL_19]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_50:.*]] = fir.convert %[[VAL_49]] : (i32) -> i64
+! CHECK:                 %[[VAL_51:.*]] = hlfir.designate %[[VAL_5]]#0 (%[[VAL_50]])  : (!fir.ref<!fir.array<100x!fir.logical<4>>>, i64) -> !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_52:.*]] = fir.load %[[VAL_51]] : !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_53:.*]] = fir.convert %[[VAL_48]] : (!fir.logical<4>) -> i1
+! CHECK:                 %[[VAL_54:.*]] = fir.convert %[[VAL_52]] : (!fir.logical<4>) -> i1
+! CHECK:                 %[[VAL_55:.*]] = arith.cmpi ne, %[[VAL_53]], %[[VAL_54]] : i1
+! CHECK:                 %[[VAL_56:.*]] = fir.convert %[[VAL_55]] : (i1) -> !fir.logical<4>
+! CHECK:                 hlfir.assign %[[VAL_56]] to %[[VAL_29]]#0 : !fir.logical<4>, !fir.ref<!fir.logical<4>>
+! CHECK:                 omp.yield
+! CHECK:               omp.terminator
 ! CHECK:             omp.terminator
 ! CHECK:           return
 ! CHECK:         }
diff --git a/flang/test/Lower/OpenMP/wsloop-reduction-logical-neqv.f90 b/flang/test/Lower/OpenMP/wsloop-reduction-logical-neqv.f90
index 08d6a2efd399..f5c84aaaf485 100644
--- a/flang/test/Lower/OpenMP/wsloop-reduction-logical-neqv.f90
+++ b/flang/test/Lower/OpenMP/wsloop-reduction-logical-neqv.f90
@@ -36,20 +36,22 @@
 ! CHECK:             %[[VAL_12:.*]] = arith.constant 1 : i32
 ! CHECK:             %[[VAL_13:.*]] = arith.constant 100 : i32
 ! CHECK:             %[[VAL_14:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop reduction(@neqv_reduction %[[VAL_4]]#0 -> %[[VAL_15:.*]] : !fir.ref<!fir.logical<4>>)  for  (%[[VAL_16:.*]]) : i32 = (%[[VAL_12]]) to (%[[VAL_13]]) inclusive step (%[[VAL_14]]) {
-! CHECK:               fir.store %[[VAL_16]] to %[[VAL_11]]#1 : !fir.ref<i32>
-! CHECK:               %[[VAL_17:.*]]:2 = hlfir.declare %[[VAL_15]] {uniq_name = "_QFsimple_reductionEx"} : (!fir.ref<!fir.logical<4>>) -> (!fir.ref<!fir.logical<4>>, !fir.ref<!fir.logical<4>>)
-! CHECK:               %[[VAL_18:.*]] = fir.load %[[VAL_17]]#0 : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_19:.*]] = fir.load %[[VAL_11]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_20:.*]] = fir.convert %[[VAL_19]] : (i32) -> i64
-! CHECK:               %[[VAL_21:.*]] = hlfir.designate %[[VAL_7]]#0 (%[[VAL_20]])  : (!fir.ref<!fir.array<100x!fir.logical<4>>>, i64) -> !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_22:.*]] = fir.load %[[VAL_21]] : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_23:.*]] = fir.convert %[[VAL_18]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_24:.*]] = fir.convert %[[VAL_22]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_25:.*]] = arith.cmpi ne, %[[VAL_23]], %[[VAL_24]] : i1
-! CHECK:               %[[VAL_26:.*]] = fir.convert %[[VAL_25]] : (i1) -> !fir.logical<4>
-! CHECK:               hlfir.assign %[[VAL_26]] to %[[VAL_17]]#0 : !fir.logical<4>, !fir.ref<!fir.logical<4>>
-! CHECK:               omp.yield
+! CHECK:             omp.wsloop reduction(@neqv_reduction %[[VAL_4]]#0 -> %[[VAL_15:.*]] : !fir.ref<!fir.logical<4>>) {
+! CHECK-NEXT:          omp.loop_nest (%[[VAL_16:.*]]) : i32 = (%[[VAL_12]]) to (%[[VAL_13]]) inclusive step (%[[VAL_14]]) {
+! CHECK:                 %[[VAL_17:.*]]:2 = hlfir.declare %[[VAL_15]] {uniq_name = "_QFsimple_reductionEx"} : (!fir.ref<!fir.logical<4>>) -> (!fir.ref<!fir.logical<4>>, !fir.ref<!fir.logical<4>>)
+! CHECK:                 fir.store %[[VAL_16]] to %[[VAL_11]]#1 : !fir.ref<i32>
+! CHECK:                 %[[VAL_18:.*]] = fir.load %[[VAL_17]]#0 : !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_19:.*]] = fir.load %[[VAL_11]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_20:.*]] = fir.convert %[[VAL_19]] : (i32) -> i64
+! CHECK:                 %[[VAL_21:.*]] = hlfir.designate %[[VAL_7]]#0 (%[[VAL_20]])  : (!fir.ref<!fir.array<100x!fir.logical<4>>>, i64) -> !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_22:.*]] = fir.load %[[VAL_21]] : !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_23:.*]] = fir.convert %[[VAL_18]] : (!fir.logical<4>) -> i1
+! CHECK:                 %[[VAL_24:.*]] = fir.convert %[[VAL_22]] : (!fir.logical<4>) -> i1
+! CHECK:                 %[[VAL_25:.*]] = arith.cmpi ne, %[[VAL_23]], %[[VAL_24]] : i1
+! CHECK:                 %[[VAL_26:.*]] = fir.convert %[[VAL_25]] : (i1) -> !fir.logical<4>
+! CHECK:                 hlfir.assign %[[VAL_26]] to %[[VAL_17]]#0 : !fir.logical<4>, !fir.ref<!fir.logical<4>>
+! CHECK:                 omp.yield
+! CHECK:               omp.terminator
 ! CHECK:             omp.terminator
 ! CHECK:           return
 
@@ -84,20 +86,22 @@ end subroutine
 ! CHECK:             %[[VAL_12:.*]] = arith.constant 1 : i32
 ! CHECK:             %[[VAL_13:.*]] = arith.constant 100 : i32
 ! CHECK:             %[[VAL_14:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop reduction(@neqv_reduction %[[VAL_4]]#0 -> %[[VAL_15:.*]] : !fir.ref<!fir.logical<4>>)  for  (%[[VAL_16:.*]]) : i32 = (%[[VAL_12]]) to (%[[VAL_13]]) inclusive step (%[[VAL_14]]) {
-! CHECK:               fir.store %[[VAL_16]] to %[[VAL_11]]#1 : !fir.ref<i32>
-! CHECK:               %[[VAL_17:.*]]:2 = hlfir.declare %[[VAL_15]] {uniq_name = "_QFsimple_reduction_switch_orderEx"} : (!fir.ref<!fir.logical<4>>) -> (!fir.ref<!fir.logical<4>>, !fir.ref<!fir.logical<4>>)
-! CHECK:               %[[VAL_18:.*]] = fir.load %[[VAL_11]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_19:.*]] = fir.convert %[[VAL_18]] : (i32) -> i64
-! CHECK:               %[[VAL_20:.*]] = hlfir.designate %[[VAL_7]]#0 (%[[VAL_19]])  : (!fir.ref<!fir.array<100x!fir.logical<4>>>, i64) -> !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_21:.*]] = fir.load %[[VAL_20]] : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_22:.*]] = fir.load %[[VAL_17]]#0 : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_23:.*]] = fir.convert %[[VAL_21]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_24:.*]] = fir.convert %[[VAL_22]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_25:.*]] = arith.cmpi ne, %[[VAL_23]], %[[VAL_24]] : i1
-! CHECK:               %[[VAL_26:.*]] = fir.convert %[[VAL_25]] : (i1) -> !fir.logical<4>
-! CHECK:               hlfir.assign %[[VAL_26]] to %[[VAL_17]]#0 : !fir.logical<4>, !fir.ref<!fir.logical<4>>
-! CHECK:               omp.yield
+! CHECK:             omp.wsloop reduction(@neqv_reduction %[[VAL_4]]#0 -> %[[VAL_15:.*]] : !fir.ref<!fir.logical<4>>) {
+! CHECK-NEXT:          omp.loop_nest (%[[VAL_16:.*]]) : i32 = (%[[VAL_12]]) to (%[[VAL_13]]) inclusive step (%[[VAL_14]]) {
+! CHECK:                 %[[VAL_17:.*]]:2 = hlfir.declare %[[VAL_15]] {uniq_name = "_QFsimple_reduction_switch_orderEx"} : (!fir.ref<!fir.logical<4>>) -> (!fir.ref<!fir.logical<4>>, !fir.ref<!fir.logical<4>>)
+! CHECK:                 fir.store %[[VAL_16]] to %[[VAL_11]]#1 : !fir.ref<i32>
+! CHECK:                 %[[VAL_18:.*]] = fir.load %[[VAL_11]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_19:.*]] = fir.convert %[[VAL_18]] : (i32) -> i64
+! CHECK:                 %[[VAL_20:.*]] = hlfir.designate %[[VAL_7]]#0 (%[[VAL_19]])  : (!fir.ref<!fir.array<100x!fir.logical<4>>>, i64) -> !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_21:.*]] = fir.load %[[VAL_20]] : !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_22:.*]] = fir.load %[[VAL_17]]#0 : !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_23:.*]] = fir.convert %[[VAL_21]] : (!fir.logical<4>) -> i1
+! CHECK:                 %[[VAL_24:.*]] = fir.convert %[[VAL_22]] : (!fir.logical<4>) -> i1
+! CHECK:                 %[[VAL_25:.*]] = arith.cmpi ne, %[[VAL_23]], %[[VAL_24]] : i1
+! CHECK:                 %[[VAL_26:.*]] = fir.convert %[[VAL_25]] : (i1) -> !fir.logical<4>
+! CHECK:                 hlfir.assign %[[VAL_26]] to %[[VAL_17]]#0 : !fir.logical<4>, !fir.ref<!fir.logical<4>>
+! CHECK:                 omp.yield
+! CHECK:               omp.terminator
 ! CHECK:             omp.terminator
 ! CHECK:           return
 
@@ -143,42 +147,44 @@ end subroutine
 ! CHECK:             %[[VAL_20:.*]] = arith.constant 1 : i32
 ! CHECK:             %[[VAL_21:.*]] = arith.constant 100 : i32
 ! CHECK:             %[[VAL_22:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop reduction(@neqv_reduction %[[VAL_7]]#0 -> %[[VAL_23:.*]] : !fir.ref<!fir.logical<4>>, @neqv_reduction %[[VAL_9]]#0 -> %[[VAL_24:.*]] : !fir.ref<!fir.logical<4>>, @neqv_reduction %[[VAL_11]]#0 -> %[[VAL_25:.*]] : !fir.ref<!fir.logical<4>>)  for  (%[[VAL_26:.*]]) : i32 = (%[[VAL_20]]) to (%[[VAL_21]]) inclusive step (%[[VAL_22]]) {
-! CHECK:               fir.store %[[VAL_26]] to %[[VAL_19]]#1 : !fir.ref<i32>
-! CHECK:               %[[VAL_27:.*]]:2 = hlfir.declare %[[VAL_23]] {uniq_name = "_QFmultiple_reductionsEx"} : (!fir.ref<!fir.logical<4>>) -> (!fir.ref<!fir.logical<4>>, !fir.ref<!fir.logical<4>>)
-! CHECK:               %[[VAL_28:.*]]:2 = hlfir.declare %[[VAL_24]] {uniq_name = "_QFmultiple_reductionsEy"} : (!fir.ref<!fir.logical<4>>) -> (!fir.ref<!fir.logical<4>>, !fir.ref<!fir.logical<4>>)
-! CHECK:               %[[VAL_29:.*]]:2 = hlfir.declare %[[VAL_25]] {uniq_name = "_QFmultiple_reductionsEz"} : (!fir.ref<!fir.logical<4>>) -> (!fir.ref<!fir.logical<4>>, !fir.ref<!fir.logical<4>>)
-! CHECK:               %[[VAL_30:.*]] = fir.load %[[VAL_27]]#0 : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_31:.*]] = fir.load %[[VAL_19]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_32:.*]] = fir.convert %[[VAL_31]] : (i32) -> i64
-! CHECK:               %[[VAL_33:.*]] = hlfir.designate %[[VAL_5]]#0 (%[[VAL_32]])  : (!fir.ref<!fir.array<100x!fir.logical<4>>>, i64) -> !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_34:.*]] = fir.load %[[VAL_33]] : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_35:.*]] = fir.convert %[[VAL_30]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_36:.*]] = fir.convert %[[VAL_34]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_37:.*]] = arith.cmpi ne, %[[VAL_35]], %[[VAL_36]] : i1
-! CHECK:               %[[VAL_38:.*]] = fir.convert %[[VAL_37]] : (i1) -> !fir.logical<4>
-! CHECK:               hlfir.assign %[[VAL_38]] to %[[VAL_27]]#0 : !fir.logical<4>, !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_39:.*]] = fir.load %[[VAL_28]]#0 : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_40:.*]] = fir.load %[[VAL_19]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_41:.*]] = fir.convert %[[VAL_40]] : (i32) -> i64
-! CHECK:               %[[VAL_42:.*]] = hlfir.designate %[[VAL_5]]#0 (%[[VAL_41]])  : (!fir.ref<!fir.array<100x!fir.logical<4>>>, i64) -> !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_43:.*]] = fir.load %[[VAL_42]] : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_44:.*]] = fir.convert %[[VAL_39]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_45:.*]] = fir.convert %[[VAL_43]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_46:.*]] = arith.cmpi ne, %[[VAL_44]], %[[VAL_45]] : i1
-! CHECK:               %[[VAL_47:.*]] = fir.convert %[[VAL_46]] : (i1) -> !fir.logical<4>
-! CHECK:               hlfir.assign %[[VAL_47]] to %[[VAL_28]]#0 : !fir.logical<4>, !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_48:.*]] = fir.load %[[VAL_29]]#0 : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_49:.*]] = fir.load %[[VAL_19]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_50:.*]] = fir.convert %[[VAL_49]] : (i32) -> i64
-! CHECK:               %[[VAL_51:.*]] = hlfir.designate %[[VAL_5]]#0 (%[[VAL_50]])  : (!fir.ref<!fir.array<100x!fir.logical<4>>>, i64) -> !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_52:.*]] = fir.load %[[VAL_51]] : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_53:.*]] = fir.convert %[[VAL_48]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_54:.*]] = fir.convert %[[VAL_52]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_55:.*]] = arith.cmpi ne, %[[VAL_53]], %[[VAL_54]] : i1
-! CHECK:               %[[VAL_56:.*]] = fir.convert %[[VAL_55]] : (i1) -> !fir.logical<4>
-! CHECK:               hlfir.assign %[[VAL_56]] to %[[VAL_29]]#0 : !fir.logical<4>, !fir.ref<!fir.logical<4>>
-! CHECK:               omp.yield
+! CHECK:             omp.wsloop reduction(@neqv_reduction %[[VAL_7]]#0 -> %[[VAL_23:.*]] : !fir.ref<!fir.logical<4>>, @neqv_reduction %[[VAL_9]]#0 -> %[[VAL_24:.*]] : !fir.ref<!fir.logical<4>>, @neqv_reduction %[[VAL_11]]#0 -> %[[VAL_25:.*]] : !fir.ref<!fir.logical<4>>) {
+! CHECK-NEXT:          omp.loop_nest (%[[VAL_26:.*]]) : i32 = (%[[VAL_20]]) to (%[[VAL_21]]) inclusive step (%[[VAL_22]]) {
+! CHECK:                 %[[VAL_27:.*]]:2 = hlfir.declare %[[VAL_23]] {uniq_name = "_QFmultiple_reductionsEx"} : (!fir.ref<!fir.logical<4>>) -> (!fir.ref<!fir.logical<4>>, !fir.ref<!fir.logical<4>>)
+! CHECK:                 %[[VAL_28:.*]]:2 = hlfir.declare %[[VAL_24]] {uniq_name = "_QFmultiple_reductionsEy"} : (!fir.ref<!fir.logical<4>>) -> (!fir.ref<!fir.logical<4>>, !fir.ref<!fir.logical<4>>)
+! CHECK:                 %[[VAL_29:.*]]:2 = hlfir.declare %[[VAL_25]] {uniq_name = "_QFmultiple_reductionsEz"} : (!fir.ref<!fir.logical<4>>) -> (!fir.ref<!fir.logical<4>>, !fir.ref<!fir.logical<4>>)
+! CHECK:                 fir.store %[[VAL_26]] to %[[VAL_19]]#1 : !fir.ref<i32>
+! CHECK:                 %[[VAL_30:.*]] = fir.load %[[VAL_27]]#0 : !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_31:.*]] = fir.load %[[VAL_19]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_32:.*]] = fir.convert %[[VAL_31]] : (i32) -> i64
+! CHECK:                 %[[VAL_33:.*]] = hlfir.designate %[[VAL_5]]#0 (%[[VAL_32]])  : (!fir.ref<!fir.array<100x!fir.logical<4>>>, i64) -> !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_34:.*]] = fir.load %[[VAL_33]] : !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_35:.*]] = fir.convert %[[VAL_30]] : (!fir.logical<4>) -> i1
+! CHECK:                 %[[VAL_36:.*]] = fir.convert %[[VAL_34]] : (!fir.logical<4>) -> i1
+! CHECK:                 %[[VAL_37:.*]] = arith.cmpi ne, %[[VAL_35]], %[[VAL_36]] : i1
+! CHECK:                 %[[VAL_38:.*]] = fir.convert %[[VAL_37]] : (i1) -> !fir.logical<4>
+! CHECK:                 hlfir.assign %[[VAL_38]] to %[[VAL_27]]#0 : !fir.logical<4>, !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_39:.*]] = fir.load %[[VAL_28]]#0 : !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_40:.*]] = fir.load %[[VAL_19]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_41:.*]] = fir.convert %[[VAL_40]] : (i32) -> i64
+! CHECK:                 %[[VAL_42:.*]] = hlfir.designate %[[VAL_5]]#0 (%[[VAL_41]])  : (!fir.ref<!fir.array<100x!fir.logical<4>>>, i64) -> !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_43:.*]] = fir.load %[[VAL_42]] : !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_44:.*]] = fir.convert %[[VAL_39]] : (!fir.logical<4>) -> i1
+! CHECK:                 %[[VAL_45:.*]] = fir.convert %[[VAL_43]] : (!fir.logical<4>) -> i1
+! CHECK:                 %[[VAL_46:.*]] = arith.cmpi ne, %[[VAL_44]], %[[VAL_45]] : i1
+! CHECK:                 %[[VAL_47:.*]] = fir.convert %[[VAL_46]] : (i1) -> !fir.logical<4>
+! CHECK:                 hlfir.assign %[[VAL_47]] to %[[VAL_28]]#0 : !fir.logical<4>, !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_48:.*]] = fir.load %[[VAL_29]]#0 : !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_49:.*]] = fir.load %[[VAL_19]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_50:.*]] = fir.convert %[[VAL_49]] : (i32) -> i64
+! CHECK:                 %[[VAL_51:.*]] = hlfir.designate %[[VAL_5]]#0 (%[[VAL_50]])  : (!fir.ref<!fir.array<100x!fir.logical<4>>>, i64) -> !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_52:.*]] = fir.load %[[VAL_51]] : !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_53:.*]] = fir.convert %[[VAL_48]] : (!fir.logical<4>) -> i1
+! CHECK:                 %[[VAL_54:.*]] = fir.convert %[[VAL_52]] : (!fir.logical<4>) -> i1
+! CHECK:                 %[[VAL_55:.*]] = arith.cmpi ne, %[[VAL_53]], %[[VAL_54]] : i1
+! CHECK:                 %[[VAL_56:.*]] = fir.convert %[[VAL_55]] : (i1) -> !fir.logical<4>
+! CHECK:                 hlfir.assign %[[VAL_56]] to %[[VAL_29]]#0 : !fir.logical<4>, !fir.ref<!fir.logical<4>>
+! CHECK:                 omp.yield
+! CHECK:               omp.terminator
 ! CHECK:             omp.terminator
 ! CHECK:           return
 ! CHECK:         }
diff --git a/flang/test/Lower/OpenMP/wsloop-reduction-logical-or-byref.f90 b/flang/test/Lower/OpenMP/wsloop-reduction-logical-or-byref.f90
index ca69ccee4a38..e268c6ff6cf5 100644
--- a/flang/test/Lower/OpenMP/wsloop-reduction-logical-or-byref.f90
+++ b/flang/test/Lower/OpenMP/wsloop-reduction-logical-or-byref.f90
@@ -41,20 +41,22 @@
 ! CHECK:             %[[VAL_12:.*]] = arith.constant 1 : i32
 ! CHECK:             %[[VAL_13:.*]] = arith.constant 100 : i32
 ! CHECK:             %[[VAL_14:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop byref reduction(@or_reduction %[[VAL_4]]#0 -> %[[VAL_15:.*]] : !fir.ref<!fir.logical<4>>)  for  (%[[VAL_16:.*]]) : i32 = (%[[VAL_12]]) to (%[[VAL_13]]) inclusive step (%[[VAL_14]]) {
-! CHECK:               fir.store %[[VAL_16]] to %[[VAL_11]]#1 : !fir.ref<i32>
-! CHECK:               %[[VAL_17:.*]]:2 = hlfir.declare %[[VAL_15]] {uniq_name = "_QFsimple_reductionEx"} : (!fir.ref<!fir.logical<4>>) -> (!fir.ref<!fir.logical<4>>, !fir.ref<!fir.logical<4>>)
-! CHECK:               %[[VAL_18:.*]] = fir.load %[[VAL_17]]#0 : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_19:.*]] = fir.load %[[VAL_11]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_20:.*]] = fir.convert %[[VAL_19]] : (i32) -> i64
-! CHECK:               %[[VAL_21:.*]] = hlfir.designate %[[VAL_7]]#0 (%[[VAL_20]])  : (!fir.ref<!fir.array<100x!fir.logical<4>>>, i64) -> !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_22:.*]] = fir.load %[[VAL_21]] : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_23:.*]] = fir.convert %[[VAL_18]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_24:.*]] = fir.convert %[[VAL_22]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_25:.*]] = arith.ori %[[VAL_23]], %[[VAL_24]] : i1
-! CHECK:               %[[VAL_26:.*]] = fir.convert %[[VAL_25]] : (i1) -> !fir.logical<4>
-! CHECK:               hlfir.assign %[[VAL_26]] to %[[VAL_17]]#0 : !fir.logical<4>, !fir.ref<!fir.logical<4>>
-! CHECK:               omp.yield
+! CHECK:             omp.wsloop byref reduction(@or_reduction %[[VAL_4]]#0 -> %[[VAL_15:.*]] : !fir.ref<!fir.logical<4>>) {
+! CHECK-NEXT:          omp.loop_nest (%[[VAL_16:.*]]) : i32 = (%[[VAL_12]]) to (%[[VAL_13]]) inclusive step (%[[VAL_14]]) {
+! CHECK:                 %[[VAL_17:.*]]:2 = hlfir.declare %[[VAL_15]] {uniq_name = "_QFsimple_reductionEx"} : (!fir.ref<!fir.logical<4>>) -> (!fir.ref<!fir.logical<4>>, !fir.ref<!fir.logical<4>>)
+! CHECK:                 fir.store %[[VAL_16]] to %[[VAL_11]]#1 : !fir.ref<i32>
+! CHECK:                 %[[VAL_18:.*]] = fir.load %[[VAL_17]]#0 : !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_19:.*]] = fir.load %[[VAL_11]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_20:.*]] = fir.convert %[[VAL_19]] : (i32) -> i64
+! CHECK:                 %[[VAL_21:.*]] = hlfir.designate %[[VAL_7]]#0 (%[[VAL_20]])  : (!fir.ref<!fir.array<100x!fir.logical<4>>>, i64) -> !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_22:.*]] = fir.load %[[VAL_21]] : !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_23:.*]] = fir.convert %[[VAL_18]] : (!fir.logical<4>) -> i1
+! CHECK:                 %[[VAL_24:.*]] = fir.convert %[[VAL_22]] : (!fir.logical<4>) -> i1
+! CHECK:                 %[[VAL_25:.*]] = arith.ori %[[VAL_23]], %[[VAL_24]] : i1
+! CHECK:                 %[[VAL_26:.*]] = fir.convert %[[VAL_25]] : (i1) -> !fir.logical<4>
+! CHECK:                 hlfir.assign %[[VAL_26]] to %[[VAL_17]]#0 : !fir.logical<4>, !fir.ref<!fir.logical<4>>
+! CHECK:                 omp.yield
+! CHECK:               omp.terminator
 ! CHECK:             omp.terminator
 ! CHECK:           return
 
@@ -88,20 +90,22 @@ end subroutine
 ! CHECK:             %[[VAL_12:.*]] = arith.constant 1 : i32
 ! CHECK:             %[[VAL_13:.*]] = arith.constant 100 : i32
 ! CHECK:             %[[VAL_14:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop byref reduction(@or_reduction %[[VAL_4]]#0 -> %[[VAL_15:.*]] : !fir.ref<!fir.logical<4>>)  for  (%[[VAL_16:.*]]) : i32 = (%[[VAL_12]]) to (%[[VAL_13]]) inclusive step (%[[VAL_14]]) {
-! CHECK:               fir.store %[[VAL_16]] to %[[VAL_11]]#1 : !fir.ref<i32>
-! CHECK:               %[[VAL_17:.*]]:2 = hlfir.declare %[[VAL_15]] {uniq_name = "_QFsimple_reduction_switch_orderEx"} : (!fir.ref<!fir.logical<4>>) -> (!fir.ref<!fir.logical<4>>, !fir.ref<!fir.logical<4>>)
-! CHECK:               %[[VAL_18:.*]] = fir.load %[[VAL_11]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_19:.*]] = fir.convert %[[VAL_18]] : (i32) -> i64
-! CHECK:               %[[VAL_20:.*]] = hlfir.designate %[[VAL_7]]#0 (%[[VAL_19]])  : (!fir.ref<!fir.array<100x!fir.logical<4>>>, i64) -> !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_21:.*]] = fir.load %[[VAL_20]] : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_22:.*]] = fir.load %[[VAL_17]]#0 : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_23:.*]] = fir.convert %[[VAL_21]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_24:.*]] = fir.convert %[[VAL_22]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_25:.*]] = arith.ori %[[VAL_23]], %[[VAL_24]] : i1
-! CHECK:               %[[VAL_26:.*]] = fir.convert %[[VAL_25]] : (i1) -> !fir.logical<4>
-! CHECK:               hlfir.assign %[[VAL_26]] to %[[VAL_17]]#0 : !fir.logical<4>, !fir.ref<!fir.logical<4>>
-! CHECK:               omp.yield
+! CHECK:             omp.wsloop byref reduction(@or_reduction %[[VAL_4]]#0 -> %[[VAL_15:.*]] : !fir.ref<!fir.logical<4>>) {
+! CHECK-NEXT:          omp.loop_nest (%[[VAL_16:.*]]) : i32 = (%[[VAL_12]]) to (%[[VAL_13]]) inclusive step (%[[VAL_14]]) {
+! CHECK:                 %[[VAL_17:.*]]:2 = hlfir.declare %[[VAL_15]] {uniq_name = "_QFsimple_reduction_switch_orderEx"} : (!fir.ref<!fir.logical<4>>) -> (!fir.ref<!fir.logical<4>>, !fir.ref<!fir.logical<4>>)
+! CHECK:                 fir.store %[[VAL_16]] to %[[VAL_11]]#1 : !fir.ref<i32>
+! CHECK:                 %[[VAL_18:.*]] = fir.load %[[VAL_11]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_19:.*]] = fir.convert %[[VAL_18]] : (i32) -> i64
+! CHECK:                 %[[VAL_20:.*]] = hlfir.designate %[[VAL_7]]#0 (%[[VAL_19]])  : (!fir.ref<!fir.array<100x!fir.logical<4>>>, i64) -> !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_21:.*]] = fir.load %[[VAL_20]] : !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_22:.*]] = fir.load %[[VAL_17]]#0 : !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_23:.*]] = fir.convert %[[VAL_21]] : (!fir.logical<4>) -> i1
+! CHECK:                 %[[VAL_24:.*]] = fir.convert %[[VAL_22]] : (!fir.logical<4>) -> i1
+! CHECK:                 %[[VAL_25:.*]] = arith.ori %[[VAL_23]], %[[VAL_24]] : i1
+! CHECK:                 %[[VAL_26:.*]] = fir.convert %[[VAL_25]] : (i1) -> !fir.logical<4>
+! CHECK:                 hlfir.assign %[[VAL_26]] to %[[VAL_17]]#0 : !fir.logical<4>, !fir.ref<!fir.logical<4>>
+! CHECK:                 omp.yield
+! CHECK:               omp.terminator
 ! CHECK:             omp.terminator
 ! CHECK:           return
 
@@ -145,42 +149,44 @@ end subroutine
 ! CHECK:             %[[VAL_20:.*]] = arith.constant 1 : i32
 ! CHECK:             %[[VAL_21:.*]] = arith.constant 100 : i32
 ! CHECK:             %[[VAL_22:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop byref reduction(@or_reduction %[[VAL_7]]#0 -> %[[VAL_23:.*]] : !fir.ref<!fir.logical<4>>, @or_reduction %[[VAL_9]]#0 -> %[[VAL_24:.*]] : !fir.ref<!fir.logical<4>>, @or_reduction %[[VAL_11]]#0 -> %[[VAL_25:.*]] : !fir.ref<!fir.logical<4>>)  for  (%[[VAL_26:.*]]) : i32 = (%[[VAL_20]]) to (%[[VAL_21]]) inclusive step (%[[VAL_22]]) {
-! CHECK:               fir.store %[[VAL_26]] to %[[VAL_19]]#1 : !fir.ref<i32>
-! CHECK:               %[[VAL_27:.*]]:2 = hlfir.declare %[[VAL_23]] {uniq_name = "_QFmultiple_reductionsEx"} : (!fir.ref<!fir.logical<4>>) -> (!fir.ref<!fir.logical<4>>, !fir.ref<!fir.logical<4>>)
-! CHECK:               %[[VAL_28:.*]]:2 = hlfir.declare %[[VAL_24]] {uniq_name = "_QFmultiple_reductionsEy"} : (!fir.ref<!fir.logical<4>>) -> (!fir.ref<!fir.logical<4>>, !fir.ref<!fir.logical<4>>)
-! CHECK:               %[[VAL_29:.*]]:2 = hlfir.declare %[[VAL_25]] {uniq_name = "_QFmultiple_reductionsEz"} : (!fir.ref<!fir.logical<4>>) -> (!fir.ref<!fir.logical<4>>, !fir.ref<!fir.logical<4>>)
-! CHECK:               %[[VAL_30:.*]] = fir.load %[[VAL_27]]#0 : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_31:.*]] = fir.load %[[VAL_19]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_32:.*]] = fir.convert %[[VAL_31]] : (i32) -> i64
-! CHECK:               %[[VAL_33:.*]] = hlfir.designate %[[VAL_5]]#0 (%[[VAL_32]])  : (!fir.ref<!fir.array<100x!fir.logical<4>>>, i64) -> !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_34:.*]] = fir.load %[[VAL_33]] : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_35:.*]] = fir.convert %[[VAL_30]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_36:.*]] = fir.convert %[[VAL_34]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_37:.*]] = arith.ori %[[VAL_35]], %[[VAL_36]] : i1
-! CHECK:               %[[VAL_38:.*]] = fir.convert %[[VAL_37]] : (i1) -> !fir.logical<4>
-! CHECK:               hlfir.assign %[[VAL_38]] to %[[VAL_27]]#0 : !fir.logical<4>, !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_39:.*]] = fir.load %[[VAL_28]]#0 : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_40:.*]] = fir.load %[[VAL_19]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_41:.*]] = fir.convert %[[VAL_40]] : (i32) -> i64
-! CHECK:               %[[VAL_42:.*]] = hlfir.designate %[[VAL_5]]#0 (%[[VAL_41]])  : (!fir.ref<!fir.array<100x!fir.logical<4>>>, i64) -> !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_43:.*]] = fir.load %[[VAL_42]] : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_44:.*]] = fir.convert %[[VAL_39]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_45:.*]] = fir.convert %[[VAL_43]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_46:.*]] = arith.ori %[[VAL_44]], %[[VAL_45]] : i1
-! CHECK:               %[[VAL_47:.*]] = fir.convert %[[VAL_46]] : (i1) -> !fir.logical<4>
-! CHECK:               hlfir.assign %[[VAL_47]] to %[[VAL_28]]#0 : !fir.logical<4>, !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_48:.*]] = fir.load %[[VAL_29]]#0 : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_49:.*]] = fir.load %[[VAL_19]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_50:.*]] = fir.convert %[[VAL_49]] : (i32) -> i64
-! CHECK:               %[[VAL_51:.*]] = hlfir.designate %[[VAL_5]]#0 (%[[VAL_50]])  : (!fir.ref<!fir.array<100x!fir.logical<4>>>, i64) -> !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_52:.*]] = fir.load %[[VAL_51]] : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_53:.*]] = fir.convert %[[VAL_48]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_54:.*]] = fir.convert %[[VAL_52]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_55:.*]] = arith.ori %[[VAL_53]], %[[VAL_54]] : i1
-! CHECK:               %[[VAL_56:.*]] = fir.convert %[[VAL_55]] : (i1) -> !fir.logical<4>
-! CHECK:               hlfir.assign %[[VAL_56]] to %[[VAL_29]]#0 : !fir.logical<4>, !fir.ref<!fir.logical<4>>
-! CHECK:               omp.yield
+! CHECK:             omp.wsloop byref reduction(@or_reduction %[[VAL_7]]#0 -> %[[VAL_23:.*]] : !fir.ref<!fir.logical<4>>, @or_reduction %[[VAL_9]]#0 -> %[[VAL_24:.*]] : !fir.ref<!fir.logical<4>>, @or_reduction %[[VAL_11]]#0 -> %[[VAL_25:.*]] : !fir.ref<!fir.logical<4>>) {
+! CHECK-NEXT:          omp.loop_nest (%[[VAL_26:.*]]) : i32 = (%[[VAL_20]]) to (%[[VAL_21]]) inclusive step (%[[VAL_22]]) {
+! CHECK:                 %[[VAL_27:.*]]:2 = hlfir.declare %[[VAL_23]] {uniq_name = "_QFmultiple_reductionsEx"} : (!fir.ref<!fir.logical<4>>) -> (!fir.ref<!fir.logical<4>>, !fir.ref<!fir.logical<4>>)
+! CHECK:                 %[[VAL_28:.*]]:2 = hlfir.declare %[[VAL_24]] {uniq_name = "_QFmultiple_reductionsEy"} : (!fir.ref<!fir.logical<4>>) -> (!fir.ref<!fir.logical<4>>, !fir.ref<!fir.logical<4>>)
+! CHECK:                 %[[VAL_29:.*]]:2 = hlfir.declare %[[VAL_25]] {uniq_name = "_QFmultiple_reductionsEz"} : (!fir.ref<!fir.logical<4>>) -> (!fir.ref<!fir.logical<4>>, !fir.ref<!fir.logical<4>>)
+! CHECK:                 fir.store %[[VAL_26]] to %[[VAL_19]]#1 : !fir.ref<i32>
+! CHECK:                 %[[VAL_30:.*]] = fir.load %[[VAL_27]]#0 : !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_31:.*]] = fir.load %[[VAL_19]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_32:.*]] = fir.convert %[[VAL_31]] : (i32) -> i64
+! CHECK:                 %[[VAL_33:.*]] = hlfir.designate %[[VAL_5]]#0 (%[[VAL_32]])  : (!fir.ref<!fir.array<100x!fir.logical<4>>>, i64) -> !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_34:.*]] = fir.load %[[VAL_33]] : !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_35:.*]] = fir.convert %[[VAL_30]] : (!fir.logical<4>) -> i1
+! CHECK:                 %[[VAL_36:.*]] = fir.convert %[[VAL_34]] : (!fir.logical<4>) -> i1
+! CHECK:                 %[[VAL_37:.*]] = arith.ori %[[VAL_35]], %[[VAL_36]] : i1
+! CHECK:                 %[[VAL_38:.*]] = fir.convert %[[VAL_37]] : (i1) -> !fir.logical<4>
+! CHECK:                 hlfir.assign %[[VAL_38]] to %[[VAL_27]]#0 : !fir.logical<4>, !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_39:.*]] = fir.load %[[VAL_28]]#0 : !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_40:.*]] = fir.load %[[VAL_19]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_41:.*]] = fir.convert %[[VAL_40]] : (i32) -> i64
+! CHECK:                 %[[VAL_42:.*]] = hlfir.designate %[[VAL_5]]#0 (%[[VAL_41]])  : (!fir.ref<!fir.array<100x!fir.logical<4>>>, i64) -> !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_43:.*]] = fir.load %[[VAL_42]] : !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_44:.*]] = fir.convert %[[VAL_39]] : (!fir.logical<4>) -> i1
+! CHECK:                 %[[VAL_45:.*]] = fir.convert %[[VAL_43]] : (!fir.logical<4>) -> i1
+! CHECK:                 %[[VAL_46:.*]] = arith.ori %[[VAL_44]], %[[VAL_45]] : i1
+! CHECK:                 %[[VAL_47:.*]] = fir.convert %[[VAL_46]] : (i1) -> !fir.logical<4>
+! CHECK:                 hlfir.assign %[[VAL_47]] to %[[VAL_28]]#0 : !fir.logical<4>, !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_48:.*]] = fir.load %[[VAL_29]]#0 : !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_49:.*]] = fir.load %[[VAL_19]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_50:.*]] = fir.convert %[[VAL_49]] : (i32) -> i64
+! CHECK:                 %[[VAL_51:.*]] = hlfir.designate %[[VAL_5]]#0 (%[[VAL_50]])  : (!fir.ref<!fir.array<100x!fir.logical<4>>>, i64) -> !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_52:.*]] = fir.load %[[VAL_51]] : !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_53:.*]] = fir.convert %[[VAL_48]] : (!fir.logical<4>) -> i1
+! CHECK:                 %[[VAL_54:.*]] = fir.convert %[[VAL_52]] : (!fir.logical<4>) -> i1
+! CHECK:                 %[[VAL_55:.*]] = arith.ori %[[VAL_53]], %[[VAL_54]] : i1
+! CHECK:                 %[[VAL_56:.*]] = fir.convert %[[VAL_55]] : (i1) -> !fir.logical<4>
+! CHECK:                 hlfir.assign %[[VAL_56]] to %[[VAL_29]]#0 : !fir.logical<4>, !fir.ref<!fir.logical<4>>
+! CHECK:                 omp.yield
+! CHECK:               omp.terminator
 ! CHECK:             omp.terminator
 ! CHECK:           return
 
diff --git a/flang/test/Lower/OpenMP/wsloop-reduction-logical-or.f90 b/flang/test/Lower/OpenMP/wsloop-reduction-logical-or.f90
index c4bf8e9d65ae..26dc0c327aad 100644
--- a/flang/test/Lower/OpenMP/wsloop-reduction-logical-or.f90
+++ b/flang/test/Lower/OpenMP/wsloop-reduction-logical-or.f90
@@ -36,20 +36,22 @@
 ! CHECK:             %[[VAL_12:.*]] = arith.constant 1 : i32
 ! CHECK:             %[[VAL_13:.*]] = arith.constant 100 : i32
 ! CHECK:             %[[VAL_14:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop reduction(@or_reduction %[[VAL_4]]#0 -> %[[VAL_15:.*]] : !fir.ref<!fir.logical<4>>)  for  (%[[VAL_16:.*]]) : i32 = (%[[VAL_12]]) to (%[[VAL_13]]) inclusive step (%[[VAL_14]]) {
-! CHECK:               fir.store %[[VAL_16]] to %[[VAL_11]]#1 : !fir.ref<i32>
-! CHECK:               %[[VAL_17:.*]]:2 = hlfir.declare %[[VAL_15]] {uniq_name = "_QFsimple_reductionEx"} : (!fir.ref<!fir.logical<4>>) -> (!fir.ref<!fir.logical<4>>, !fir.ref<!fir.logical<4>>)
-! CHECK:               %[[VAL_18:.*]] = fir.load %[[VAL_17]]#0 : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_19:.*]] = fir.load %[[VAL_11]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_20:.*]] = fir.convert %[[VAL_19]] : (i32) -> i64
-! CHECK:               %[[VAL_21:.*]] = hlfir.designate %[[VAL_7]]#0 (%[[VAL_20]])  : (!fir.ref<!fir.array<100x!fir.logical<4>>>, i64) -> !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_22:.*]] = fir.load %[[VAL_21]] : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_23:.*]] = fir.convert %[[VAL_18]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_24:.*]] = fir.convert %[[VAL_22]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_25:.*]] = arith.ori %[[VAL_23]], %[[VAL_24]] : i1
-! CHECK:               %[[VAL_26:.*]] = fir.convert %[[VAL_25]] : (i1) -> !fir.logical<4>
-! CHECK:               hlfir.assign %[[VAL_26]] to %[[VAL_17]]#0 : !fir.logical<4>, !fir.ref<!fir.logical<4>>
-! CHECK:               omp.yield
+! CHECK:             omp.wsloop reduction(@or_reduction %[[VAL_4]]#0 -> %[[VAL_15:.*]] : !fir.ref<!fir.logical<4>>) {
+! CHECK-NEXT:          omp.loop_nest (%[[VAL_16:.*]]) : i32 = (%[[VAL_12]]) to (%[[VAL_13]]) inclusive step (%[[VAL_14]]) {
+! CHECK:                 %[[VAL_17:.*]]:2 = hlfir.declare %[[VAL_15]] {uniq_name = "_QFsimple_reductionEx"} : (!fir.ref<!fir.logical<4>>) -> (!fir.ref<!fir.logical<4>>, !fir.ref<!fir.logical<4>>)
+! CHECK:                 fir.store %[[VAL_16]] to %[[VAL_11]]#1 : !fir.ref<i32>
+! CHECK:                 %[[VAL_18:.*]] = fir.load %[[VAL_17]]#0 : !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_19:.*]] = fir.load %[[VAL_11]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_20:.*]] = fir.convert %[[VAL_19]] : (i32) -> i64
+! CHECK:                 %[[VAL_21:.*]] = hlfir.designate %[[VAL_7]]#0 (%[[VAL_20]])  : (!fir.ref<!fir.array<100x!fir.logical<4>>>, i64) -> !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_22:.*]] = fir.load %[[VAL_21]] : !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_23:.*]] = fir.convert %[[VAL_18]] : (!fir.logical<4>) -> i1
+! CHECK:                 %[[VAL_24:.*]] = fir.convert %[[VAL_22]] : (!fir.logical<4>) -> i1
+! CHECK:                 %[[VAL_25:.*]] = arith.ori %[[VAL_23]], %[[VAL_24]] : i1
+! CHECK:                 %[[VAL_26:.*]] = fir.convert %[[VAL_25]] : (i1) -> !fir.logical<4>
+! CHECK:                 hlfir.assign %[[VAL_26]] to %[[VAL_17]]#0 : !fir.logical<4>, !fir.ref<!fir.logical<4>>
+! CHECK:                 omp.yield
+! CHECK:               omp.terminator
 ! CHECK:             omp.terminator
 ! CHECK:           return
 
@@ -83,20 +85,22 @@ end subroutine
 ! CHECK:             %[[VAL_12:.*]] = arith.constant 1 : i32
 ! CHECK:             %[[VAL_13:.*]] = arith.constant 100 : i32
 ! CHECK:             %[[VAL_14:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop reduction(@or_reduction %[[VAL_4]]#0 -> %[[VAL_15:.*]] : !fir.ref<!fir.logical<4>>)  for  (%[[VAL_16:.*]]) : i32 = (%[[VAL_12]]) to (%[[VAL_13]]) inclusive step (%[[VAL_14]]) {
-! CHECK:               fir.store %[[VAL_16]] to %[[VAL_11]]#1 : !fir.ref<i32>
-! CHECK:               %[[VAL_17:.*]]:2 = hlfir.declare %[[VAL_15]] {uniq_name = "_QFsimple_reduction_switch_orderEx"} : (!fir.ref<!fir.logical<4>>) -> (!fir.ref<!fir.logical<4>>, !fir.ref<!fir.logical<4>>)
-! CHECK:               %[[VAL_18:.*]] = fir.load %[[VAL_11]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_19:.*]] = fir.convert %[[VAL_18]] : (i32) -> i64
-! CHECK:               %[[VAL_20:.*]] = hlfir.designate %[[VAL_7]]#0 (%[[VAL_19]])  : (!fir.ref<!fir.array<100x!fir.logical<4>>>, i64) -> !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_21:.*]] = fir.load %[[VAL_20]] : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_22:.*]] = fir.load %[[VAL_17]]#0 : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_23:.*]] = fir.convert %[[VAL_21]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_24:.*]] = fir.convert %[[VAL_22]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_25:.*]] = arith.ori %[[VAL_23]], %[[VAL_24]] : i1
-! CHECK:               %[[VAL_26:.*]] = fir.convert %[[VAL_25]] : (i1) -> !fir.logical<4>
-! CHECK:               hlfir.assign %[[VAL_26]] to %[[VAL_17]]#0 : !fir.logical<4>, !fir.ref<!fir.logical<4>>
-! CHECK:               omp.yield
+! CHECK:             omp.wsloop reduction(@or_reduction %[[VAL_4]]#0 -> %[[VAL_15:.*]] : !fir.ref<!fir.logical<4>>) {
+! CHECK-NEXT:          omp.loop_nest (%[[VAL_16:.*]]) : i32 = (%[[VAL_12]]) to (%[[VAL_13]]) inclusive step (%[[VAL_14]]) {
+! CHECK:                 %[[VAL_17:.*]]:2 = hlfir.declare %[[VAL_15]] {uniq_name = "_QFsimple_reduction_switch_orderEx"} : (!fir.ref<!fir.logical<4>>) -> (!fir.ref<!fir.logical<4>>, !fir.ref<!fir.logical<4>>)
+! CHECK:                 fir.store %[[VAL_16]] to %[[VAL_11]]#1 : !fir.ref<i32>
+! CHECK:                 %[[VAL_18:.*]] = fir.load %[[VAL_11]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_19:.*]] = fir.convert %[[VAL_18]] : (i32) -> i64
+! CHECK:                 %[[VAL_20:.*]] = hlfir.designate %[[VAL_7]]#0 (%[[VAL_19]])  : (!fir.ref<!fir.array<100x!fir.logical<4>>>, i64) -> !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_21:.*]] = fir.load %[[VAL_20]] : !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_22:.*]] = fir.load %[[VAL_17]]#0 : !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_23:.*]] = fir.convert %[[VAL_21]] : (!fir.logical<4>) -> i1
+! CHECK:                 %[[VAL_24:.*]] = fir.convert %[[VAL_22]] : (!fir.logical<4>) -> i1
+! CHECK:                 %[[VAL_25:.*]] = arith.ori %[[VAL_23]], %[[VAL_24]] : i1
+! CHECK:                 %[[VAL_26:.*]] = fir.convert %[[VAL_25]] : (i1) -> !fir.logical<4>
+! CHECK:                 hlfir.assign %[[VAL_26]] to %[[VAL_17]]#0 : !fir.logical<4>, !fir.ref<!fir.logical<4>>
+! CHECK:                 omp.yield
+! CHECK:               omp.terminator
 ! CHECK:             omp.terminator
 ! CHECK:           return
 
@@ -140,42 +144,44 @@ end subroutine
 ! CHECK:             %[[VAL_20:.*]] = arith.constant 1 : i32
 ! CHECK:             %[[VAL_21:.*]] = arith.constant 100 : i32
 ! CHECK:             %[[VAL_22:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop reduction(@or_reduction %[[VAL_7]]#0 -> %[[VAL_23:.*]] : !fir.ref<!fir.logical<4>>, @or_reduction %[[VAL_9]]#0 -> %[[VAL_24:.*]] : !fir.ref<!fir.logical<4>>, @or_reduction %[[VAL_11]]#0 -> %[[VAL_25:.*]] : !fir.ref<!fir.logical<4>>)  for  (%[[VAL_26:.*]]) : i32 = (%[[VAL_20]]) to (%[[VAL_21]]) inclusive step (%[[VAL_22]]) {
-! CHECK:               fir.store %[[VAL_26]] to %[[VAL_19]]#1 : !fir.ref<i32>
-! CHECK:               %[[VAL_27:.*]]:2 = hlfir.declare %[[VAL_23]] {uniq_name = "_QFmultiple_reductionsEx"} : (!fir.ref<!fir.logical<4>>) -> (!fir.ref<!fir.logical<4>>, !fir.ref<!fir.logical<4>>)
-! CHECK:               %[[VAL_28:.*]]:2 = hlfir.declare %[[VAL_24]] {uniq_name = "_QFmultiple_reductionsEy"} : (!fir.ref<!fir.logical<4>>) -> (!fir.ref<!fir.logical<4>>, !fir.ref<!fir.logical<4>>)
-! CHECK:               %[[VAL_29:.*]]:2 = hlfir.declare %[[VAL_25]] {uniq_name = "_QFmultiple_reductionsEz"} : (!fir.ref<!fir.logical<4>>) -> (!fir.ref<!fir.logical<4>>, !fir.ref<!fir.logical<4>>)
-! CHECK:               %[[VAL_30:.*]] = fir.load %[[VAL_27]]#0 : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_31:.*]] = fir.load %[[VAL_19]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_32:.*]] = fir.convert %[[VAL_31]] : (i32) -> i64
-! CHECK:               %[[VAL_33:.*]] = hlfir.designate %[[VAL_5]]#0 (%[[VAL_32]])  : (!fir.ref<!fir.array<100x!fir.logical<4>>>, i64) -> !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_34:.*]] = fir.load %[[VAL_33]] : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_35:.*]] = fir.convert %[[VAL_30]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_36:.*]] = fir.convert %[[VAL_34]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_37:.*]] = arith.ori %[[VAL_35]], %[[VAL_36]] : i1
-! CHECK:               %[[VAL_38:.*]] = fir.convert %[[VAL_37]] : (i1) -> !fir.logical<4>
-! CHECK:               hlfir.assign %[[VAL_38]] to %[[VAL_27]]#0 : !fir.logical<4>, !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_39:.*]] = fir.load %[[VAL_28]]#0 : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_40:.*]] = fir.load %[[VAL_19]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_41:.*]] = fir.convert %[[VAL_40]] : (i32) -> i64
-! CHECK:               %[[VAL_42:.*]] = hlfir.designate %[[VAL_5]]#0 (%[[VAL_41]])  : (!fir.ref<!fir.array<100x!fir.logical<4>>>, i64) -> !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_43:.*]] = fir.load %[[VAL_42]] : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_44:.*]] = fir.convert %[[VAL_39]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_45:.*]] = fir.convert %[[VAL_43]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_46:.*]] = arith.ori %[[VAL_44]], %[[VAL_45]] : i1
-! CHECK:               %[[VAL_47:.*]] = fir.convert %[[VAL_46]] : (i1) -> !fir.logical<4>
-! CHECK:               hlfir.assign %[[VAL_47]] to %[[VAL_28]]#0 : !fir.logical<4>, !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_48:.*]] = fir.load %[[VAL_29]]#0 : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_49:.*]] = fir.load %[[VAL_19]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_50:.*]] = fir.convert %[[VAL_49]] : (i32) -> i64
-! CHECK:               %[[VAL_51:.*]] = hlfir.designate %[[VAL_5]]#0 (%[[VAL_50]])  : (!fir.ref<!fir.array<100x!fir.logical<4>>>, i64) -> !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_52:.*]] = fir.load %[[VAL_51]] : !fir.ref<!fir.logical<4>>
-! CHECK:               %[[VAL_53:.*]] = fir.convert %[[VAL_48]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_54:.*]] = fir.convert %[[VAL_52]] : (!fir.logical<4>) -> i1
-! CHECK:               %[[VAL_55:.*]] = arith.ori %[[VAL_53]], %[[VAL_54]] : i1
-! CHECK:               %[[VAL_56:.*]] = fir.convert %[[VAL_55]] : (i1) -> !fir.logical<4>
-! CHECK:               hlfir.assign %[[VAL_56]] to %[[VAL_29]]#0 : !fir.logical<4>, !fir.ref<!fir.logical<4>>
-! CHECK:               omp.yield
+! CHECK:             omp.wsloop reduction(@or_reduction %[[VAL_7]]#0 -> %[[VAL_23:.*]] : !fir.ref<!fir.logical<4>>, @or_reduction %[[VAL_9]]#0 -> %[[VAL_24:.*]] : !fir.ref<!fir.logical<4>>, @or_reduction %[[VAL_11]]#0 -> %[[VAL_25:.*]] : !fir.ref<!fir.logical<4>>) {
+! CHECK-NEXT:          omp.loop_nest (%[[VAL_26:.*]]) : i32 = (%[[VAL_20]]) to (%[[VAL_21]]) inclusive step (%[[VAL_22]]) {
+! CHECK:                 %[[VAL_27:.*]]:2 = hlfir.declare %[[VAL_23]] {uniq_name = "_QFmultiple_reductionsEx"} : (!fir.ref<!fir.logical<4>>) -> (!fir.ref<!fir.logical<4>>, !fir.ref<!fir.logical<4>>)
+! CHECK:                 %[[VAL_28:.*]]:2 = hlfir.declare %[[VAL_24]] {uniq_name = "_QFmultiple_reductionsEy"} : (!fir.ref<!fir.logical<4>>) -> (!fir.ref<!fir.logical<4>>, !fir.ref<!fir.logical<4>>)
+! CHECK:                 %[[VAL_29:.*]]:2 = hlfir.declare %[[VAL_25]] {uniq_name = "_QFmultiple_reductionsEz"} : (!fir.ref<!fir.logical<4>>) -> (!fir.ref<!fir.logical<4>>, !fir.ref<!fir.logical<4>>)
+! CHECK:                 fir.store %[[VAL_26]] to %[[VAL_19]]#1 : !fir.ref<i32>
+! CHECK:                 %[[VAL_30:.*]] = fir.load %[[VAL_27]]#0 : !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_31:.*]] = fir.load %[[VAL_19]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_32:.*]] = fir.convert %[[VAL_31]] : (i32) -> i64
+! CHECK:                 %[[VAL_33:.*]] = hlfir.designate %[[VAL_5]]#0 (%[[VAL_32]])  : (!fir.ref<!fir.array<100x!fir.logical<4>>>, i64) -> !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_34:.*]] = fir.load %[[VAL_33]] : !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_35:.*]] = fir.convert %[[VAL_30]] : (!fir.logical<4>) -> i1
+! CHECK:                 %[[VAL_36:.*]] = fir.convert %[[VAL_34]] : (!fir.logical<4>) -> i1
+! CHECK:                 %[[VAL_37:.*]] = arith.ori %[[VAL_35]], %[[VAL_36]] : i1
+! CHECK:                 %[[VAL_38:.*]] = fir.convert %[[VAL_37]] : (i1) -> !fir.logical<4>
+! CHECK:                 hlfir.assign %[[VAL_38]] to %[[VAL_27]]#0 : !fir.logical<4>, !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_39:.*]] = fir.load %[[VAL_28]]#0 : !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_40:.*]] = fir.load %[[VAL_19]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_41:.*]] = fir.convert %[[VAL_40]] : (i32) -> i64
+! CHECK:                 %[[VAL_42:.*]] = hlfir.designate %[[VAL_5]]#0 (%[[VAL_41]])  : (!fir.ref<!fir.array<100x!fir.logical<4>>>, i64) -> !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_43:.*]] = fir.load %[[VAL_42]] : !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_44:.*]] = fir.convert %[[VAL_39]] : (!fir.logical<4>) -> i1
+! CHECK:                 %[[VAL_45:.*]] = fir.convert %[[VAL_43]] : (!fir.logical<4>) -> i1
+! CHECK:                 %[[VAL_46:.*]] = arith.ori %[[VAL_44]], %[[VAL_45]] : i1
+! CHECK:                 %[[VAL_47:.*]] = fir.convert %[[VAL_46]] : (i1) -> !fir.logical<4>
+! CHECK:                 hlfir.assign %[[VAL_47]] to %[[VAL_28]]#0 : !fir.logical<4>, !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_48:.*]] = fir.load %[[VAL_29]]#0 : !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_49:.*]] = fir.load %[[VAL_19]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_50:.*]] = fir.convert %[[VAL_49]] : (i32) -> i64
+! CHECK:                 %[[VAL_51:.*]] = hlfir.designate %[[VAL_5]]#0 (%[[VAL_50]])  : (!fir.ref<!fir.array<100x!fir.logical<4>>>, i64) -> !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_52:.*]] = fir.load %[[VAL_51]] : !fir.ref<!fir.logical<4>>
+! CHECK:                 %[[VAL_53:.*]] = fir.convert %[[VAL_48]] : (!fir.logical<4>) -> i1
+! CHECK:                 %[[VAL_54:.*]] = fir.convert %[[VAL_52]] : (!fir.logical<4>) -> i1
+! CHECK:                 %[[VAL_55:.*]] = arith.ori %[[VAL_53]], %[[VAL_54]] : i1
+! CHECK:                 %[[VAL_56:.*]] = fir.convert %[[VAL_55]] : (i1) -> !fir.logical<4>
+! CHECK:                 hlfir.assign %[[VAL_56]] to %[[VAL_29]]#0 : !fir.logical<4>, !fir.ref<!fir.logical<4>>
+! CHECK:                 omp.yield
+! CHECK:               omp.terminator
 ! CHECK:             omp.terminator
 ! CHECK:           return
 
diff --git a/flang/test/Lower/OpenMP/wsloop-reduction-max-byref.f90 b/flang/test/Lower/OpenMP/wsloop-reduction-max-byref.f90
index 2f6921edcb42..95bdc98f18c2 100644
--- a/flang/test/Lower/OpenMP/wsloop-reduction-max-byref.f90
+++ b/flang/test/Lower/OpenMP/wsloop-reduction-max-byref.f90
@@ -46,18 +46,20 @@
 ! CHECK:             %[[VAL_9:.*]] = arith.constant 1 : i32
 ! CHECK:             %[[VAL_10:.*]] = arith.constant 100 : i32
 ! CHECK:             %[[VAL_11:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop byref reduction(@max_byref_i32 %[[VAL_4]]#0 -> %[[VAL_12:.*]] : !fir.ref<i32>)  for  (%[[VAL_13:.*]]) : i32 = (%[[VAL_9]]) to (%[[VAL_10]]) inclusive step (%[[VAL_11]]) {
-! CHECK:               fir.store %[[VAL_13]] to %[[VAL_8]]#1 : !fir.ref<i32>
-! CHECK:               %[[VAL_14:.*]]:2 = hlfir.declare %[[VAL_12]] {uniq_name = "_QFreduction_max_intEx"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
-! CHECK:               %[[VAL_15:.*]] = fir.load %[[VAL_8]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_16:.*]] = fir.convert %[[VAL_15]] : (i32) -> i64
-! CHECK:               %[[VAL_17:.*]] = hlfir.designate %[[VAL_5]]#0 (%[[VAL_16]])  : (!fir.box<!fir.array<?xi32>>, i64) -> !fir.ref<i32>
-! CHECK:               %[[VAL_18:.*]] = fir.load %[[VAL_14]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_19:.*]] = fir.load %[[VAL_17]] : !fir.ref<i32>
-! CHECK:               %[[VAL_20:.*]] = arith.cmpi sgt, %[[VAL_18]], %[[VAL_19]] : i32
-! CHECK:               %[[VAL_21:.*]] = arith.select %[[VAL_20]], %[[VAL_18]], %[[VAL_19]] : i32
-! CHECK:               hlfir.assign %[[VAL_21]] to %[[VAL_14]]#0 : i32, !fir.ref<i32>
-! CHECK:               omp.yield
+! CHECK:             omp.wsloop byref reduction(@max_byref_i32 %[[VAL_4]]#0 -> %[[VAL_12:.*]] : !fir.ref<i32>) {
+! CHECK-NEXT:          omp.loop_nest (%[[VAL_13:.*]]) : i32 = (%[[VAL_9]]) to (%[[VAL_10]]) inclusive step (%[[VAL_11]]) {
+! CHECK:                 %[[VAL_14:.*]]:2 = hlfir.declare %[[VAL_12]] {uniq_name = "_QFreduction_max_intEx"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
+! CHECK:                 fir.store %[[VAL_13]] to %[[VAL_8]]#1 : !fir.ref<i32>
+! CHECK:                 %[[VAL_15:.*]] = fir.load %[[VAL_8]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_16:.*]] = fir.convert %[[VAL_15]] : (i32) -> i64
+! CHECK:                 %[[VAL_17:.*]] = hlfir.designate %[[VAL_5]]#0 (%[[VAL_16]])  : (!fir.box<!fir.array<?xi32>>, i64) -> !fir.ref<i32>
+! CHECK:                 %[[VAL_18:.*]] = fir.load %[[VAL_14]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_19:.*]] = fir.load %[[VAL_17]] : !fir.ref<i32>
+! CHECK:                 %[[VAL_20:.*]] = arith.cmpi sgt, %[[VAL_18]], %[[VAL_19]] : i32
+! CHECK:                 %[[VAL_21:.*]] = arith.select %[[VAL_20]], %[[VAL_18]], %[[VAL_19]] : i32
+! CHECK:                 hlfir.assign %[[VAL_21]] to %[[VAL_14]]#0 : i32, !fir.ref<i32>
+! CHECK:                 omp.yield
+! CHECK:               omp.terminator
 ! CHECK:             omp.terminator
 
 ! CHECK-LABEL:   func.func @_QPreduction_max_real(
@@ -75,18 +77,20 @@
 ! CHECK:             %[[VAL_9:.*]] = arith.constant 1 : i32
 ! CHECK:             %[[VAL_10:.*]] = arith.constant 100 : i32
 ! CHECK:             %[[VAL_11:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop byref reduction(@max_byref_f32 %[[VAL_4]]#0 -> %[[VAL_12:.*]] : !fir.ref<f32>)  for  (%[[VAL_13:.*]]) : i32 = (%[[VAL_9]]) to (%[[VAL_10]]) inclusive step (%[[VAL_11]]) {
-! CHECK:               fir.store %[[VAL_13]] to %[[VAL_8]]#1 : !fir.ref<i32>
-! CHECK:               %[[VAL_14:.*]]:2 = hlfir.declare %[[VAL_12]] {uniq_name = "_QFreduction_max_realEx"} : (!fir.ref<f32>) -> (!fir.ref<f32>, !fir.ref<f32>)
-! CHECK:               %[[VAL_15:.*]] = fir.load %[[VAL_8]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_16:.*]] = fir.convert %[[VAL_15]] : (i32) -> i64
-! CHECK:               %[[VAL_17:.*]] = hlfir.designate %[[VAL_5]]#0 (%[[VAL_16]])  : (!fir.box<!fir.array<?xf32>>, i64) -> !fir.ref<f32>
-! CHECK:               %[[VAL_18:.*]] = fir.load %[[VAL_17]] : !fir.ref<f32>
-! CHECK:               %[[VAL_19:.*]] = fir.load %[[VAL_14]]#0 : !fir.ref<f32>
-! CHECK:               %[[VAL_20:.*]] = arith.cmpf ogt, %[[VAL_18]], %[[VAL_19]] fastmath<contract> : f32
-! CHECK:               %[[VAL_21:.*]] = arith.select %[[VAL_20]], %[[VAL_18]], %[[VAL_19]] : f32
-! CHECK:               hlfir.assign %[[VAL_21]] to %[[VAL_14]]#0 : f32, !fir.ref<f32>
-! CHECK:               omp.yield
+! CHECK:             omp.wsloop byref reduction(@max_byref_f32 %[[VAL_4]]#0 -> %[[VAL_12:.*]] : !fir.ref<f32>) {
+! CHECK-NEXT:          omp.loop_nest (%[[VAL_13:.*]]) : i32 = (%[[VAL_9]]) to (%[[VAL_10]]) inclusive step (%[[VAL_11]]) {
+! CHECK:                 %[[VAL_14:.*]]:2 = hlfir.declare %[[VAL_12]] {uniq_name = "_QFreduction_max_realEx"} : (!fir.ref<f32>) -> (!fir.ref<f32>, !fir.ref<f32>)
+! CHECK:                 fir.store %[[VAL_13]] to %[[VAL_8]]#1 : !fir.ref<i32>
+! CHECK:                 %[[VAL_15:.*]] = fir.load %[[VAL_8]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_16:.*]] = fir.convert %[[VAL_15]] : (i32) -> i64
+! CHECK:                 %[[VAL_17:.*]] = hlfir.designate %[[VAL_5]]#0 (%[[VAL_16]])  : (!fir.box<!fir.array<?xf32>>, i64) -> !fir.ref<f32>
+! CHECK:                 %[[VAL_18:.*]] = fir.load %[[VAL_17]] : !fir.ref<f32>
+! CHECK:                 %[[VAL_19:.*]] = fir.load %[[VAL_14]]#0 : !fir.ref<f32>
+! CHECK:                 %[[VAL_20:.*]] = arith.cmpf ogt, %[[VAL_18]], %[[VAL_19]] fastmath<contract> : f32
+! CHECK:                 %[[VAL_21:.*]] = arith.select %[[VAL_20]], %[[VAL_18]], %[[VAL_19]] : f32
+! CHECK:                 hlfir.assign %[[VAL_21]] to %[[VAL_14]]#0 : f32, !fir.ref<f32>
+! CHECK:                 omp.yield
+! CHECK:               omp.terminator
 ! CHECK:             omp.terminator
 ! CHECK:           omp.parallel {
 ! CHECK:             %[[VAL_30:.*]] = fir.alloca i32 {adapt.valuebyref, pinned}
@@ -94,24 +98,26 @@
 ! CHECK:             %[[VAL_32:.*]] = arith.constant 1 : i32
 ! CHECK:             %[[VAL_33:.*]] = arith.constant 100 : i32
 ! CHECK:             %[[VAL_34:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop byref reduction(@max_byref_f32 %[[VAL_4]]#0 -> %[[VAL_35:.*]] : !fir.ref<f32>)  for  (%[[VAL_36:.*]]) : i32 = (%[[VAL_32]]) to (%[[VAL_33]]) inclusive step (%[[VAL_34]]) {
-! CHECK:               fir.store %[[VAL_36]] to %[[VAL_31]]#1 : !fir.ref<i32>
-! CHECK:               %[[VAL_37:.*]]:2 = hlfir.declare %[[VAL_35]] {uniq_name = "_QFreduction_max_realEx"} : (!fir.ref<f32>) -> (!fir.ref<f32>, !fir.ref<f32>)
-! CHECK:               %[[VAL_38:.*]] = fir.load %[[VAL_31]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_39:.*]] = fir.convert %[[VAL_38]] : (i32) -> i64
-! CHECK:               %[[VAL_40:.*]] = hlfir.designate %[[VAL_5]]#0 (%[[VAL_39]])  : (!fir.box<!fir.array<?xf32>>, i64) -> !fir.ref<f32>
-! CHECK:               %[[VAL_41:.*]] = fir.load %[[VAL_40]] : !fir.ref<f32>
-! CHECK:               %[[VAL_42:.*]] = fir.load %[[VAL_37]]#0 : !fir.ref<f32>
-! CHECK:               %[[VAL_43:.*]] = arith.cmpf ogt, %[[VAL_41]], %[[VAL_42]] fastmath<contract> : f32
-! CHECK:               fir.if %[[VAL_43]] {
-! CHECK:                 %[[VAL_44:.*]] = fir.load %[[VAL_31]]#0 : !fir.ref<i32>
-! CHECK:                 %[[VAL_45:.*]] = fir.convert %[[VAL_44]] : (i32) -> i64
-! CHECK:                 %[[VAL_46:.*]] = hlfir.designate %[[VAL_5]]#0 (%[[VAL_45]])  : (!fir.box<!fir.array<?xf32>>, i64) -> !fir.ref<f32>
-! CHECK:                 %[[VAL_47:.*]] = fir.load %[[VAL_46]] : !fir.ref<f32>
-! CHECK:                 hlfir.assign %[[VAL_47]] to %[[VAL_37]]#0 : f32, !fir.ref<f32>
-! CHECK:               } else {
-! CHECK:               }
-! CHECK:               omp.yield
+! CHECK:             omp.wsloop byref reduction(@max_byref_f32 %[[VAL_4]]#0 -> %[[VAL_35:.*]] : !fir.ref<f32>) {
+! CHECK-NEXT:          omp.loop_nest (%[[VAL_36:.*]]) : i32 = (%[[VAL_32]]) to (%[[VAL_33]]) inclusive step (%[[VAL_34]]) {
+! CHECK:                 %[[VAL_37:.*]]:2 = hlfir.declare %[[VAL_35]] {uniq_name = "_QFreduction_max_realEx"} : (!fir.ref<f32>) -> (!fir.ref<f32>, !fir.ref<f32>)
+! CHECK:                 fir.store %[[VAL_36]] to %[[VAL_31]]#1 : !fir.ref<i32>
+! CHECK:                 %[[VAL_38:.*]] = fir.load %[[VAL_31]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_39:.*]] = fir.convert %[[VAL_38]] : (i32) -> i64
+! CHECK:                 %[[VAL_40:.*]] = hlfir.designate %[[VAL_5]]#0 (%[[VAL_39]])  : (!fir.box<!fir.array<?xf32>>, i64) -> !fir.ref<f32>
+! CHECK:                 %[[VAL_41:.*]] = fir.load %[[VAL_40]] : !fir.ref<f32>
+! CHECK:                 %[[VAL_42:.*]] = fir.load %[[VAL_37]]#0 : !fir.ref<f32>
+! CHECK:                 %[[VAL_43:.*]] = arith.cmpf ogt, %[[VAL_41]], %[[VAL_42]] fastmath<contract> : f32
+! CHECK:                 fir.if %[[VAL_43]] {
+! CHECK:                   %[[VAL_44:.*]] = fir.load %[[VAL_31]]#0 : !fir.ref<i32>
+! CHECK:                   %[[VAL_45:.*]] = fir.convert %[[VAL_44]] : (i32) -> i64
+! CHECK:                   %[[VAL_46:.*]] = hlfir.designate %[[VAL_5]]#0 (%[[VAL_45]])  : (!fir.box<!fir.array<?xf32>>, i64) -> !fir.ref<f32>
+! CHECK:                   %[[VAL_47:.*]] = fir.load %[[VAL_46]] : !fir.ref<f32>
+! CHECK:                   hlfir.assign %[[VAL_47]] to %[[VAL_37]]#0 : f32, !fir.ref<f32>
+! CHECK:                 } else {
+! CHECK:                 }
+! CHECK:                 omp.yield
+! CHECK:               omp.terminator
 ! CHECK:             omp.terminator
 
 
diff --git a/flang/test/Lower/OpenMP/wsloop-reduction-max-hlfir-byref.f90 b/flang/test/Lower/OpenMP/wsloop-reduction-max-hlfir-byref.f90
index 10bba6ac4b51..352888bb94f5 100644
--- a/flang/test/Lower/OpenMP/wsloop-reduction-max-hlfir-byref.f90
+++ b/flang/test/Lower/OpenMP/wsloop-reduction-max-hlfir-byref.f90
@@ -33,18 +33,20 @@
 ! CHECK:             %[[VAL_9:.*]] = arith.constant 1 : i32
 ! CHECK:             %[[VAL_10:.*]] = arith.constant 100 : i32
 ! CHECK:             %[[VAL_11:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop byref reduction(@max_byref_i32 %[[VAL_4]]#0 -> %[[VAL_12:.*]] : !fir.ref<i32>)  for  (%[[VAL_13:.*]]) : i32 = (%[[VAL_9]]) to (%[[VAL_10]]) inclusive step (%[[VAL_11]]) {
-! CHECK:               fir.store %[[VAL_13]] to %[[VAL_8]]#1 : !fir.ref<i32>
-! CHECK:               %[[VAL_14:.*]]:2 = hlfir.declare %[[VAL_12]] {uniq_name = "_QFreduction_max_intEx"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
-! CHECK:               %[[VAL_15:.*]] = fir.load %[[VAL_8]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_16:.*]] = fir.convert %[[VAL_15]] : (i32) -> i64
-! CHECK:               %[[VAL_17:.*]] = hlfir.designate %[[VAL_5]]#0 (%[[VAL_16]])  : (!fir.box<!fir.array<?xi32>>, i64) -> !fir.ref<i32>
-! CHECK:               %[[VAL_18:.*]] = fir.load %[[VAL_14]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_19:.*]] = fir.load %[[VAL_17]] : !fir.ref<i32>
-! CHECK:               %[[VAL_20:.*]] = arith.cmpi sgt, %[[VAL_18]], %[[VAL_19]] : i32
-! CHECK:               %[[VAL_21:.*]] = arith.select %[[VAL_20]], %[[VAL_18]], %[[VAL_19]] : i32
-! CHECK:               hlfir.assign %[[VAL_21]] to %[[VAL_14]]#0 : i32, !fir.ref<i32>
-! CHECK:               omp.yield
+! CHECK:             omp.wsloop byref reduction(@max_byref_i32 %[[VAL_4]]#0 -> %[[VAL_12:.*]] : !fir.ref<i32>) {
+! CHECK-NEXT:          omp.loop_nest (%[[VAL_13:.*]]) : i32 = (%[[VAL_9]]) to (%[[VAL_10]]) inclusive step (%[[VAL_11]]) {
+! CHECK:                 %[[VAL_14:.*]]:2 = hlfir.declare %[[VAL_12]] {uniq_name = "_QFreduction_max_intEx"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
+! CHECK:                 fir.store %[[VAL_13]] to %[[VAL_8]]#1 : !fir.ref<i32>
+! CHECK:                 %[[VAL_15:.*]] = fir.load %[[VAL_8]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_16:.*]] = fir.convert %[[VAL_15]] : (i32) -> i64
+! CHECK:                 %[[VAL_17:.*]] = hlfir.designate %[[VAL_5]]#0 (%[[VAL_16]])  : (!fir.box<!fir.array<?xi32>>, i64) -> !fir.ref<i32>
+! CHECK:                 %[[VAL_18:.*]] = fir.load %[[VAL_14]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_19:.*]] = fir.load %[[VAL_17]] : !fir.ref<i32>
+! CHECK:                 %[[VAL_20:.*]] = arith.cmpi sgt, %[[VAL_18]], %[[VAL_19]] : i32
+! CHECK:                 %[[VAL_21:.*]] = arith.select %[[VAL_20]], %[[VAL_18]], %[[VAL_19]] : i32
+! CHECK:                 hlfir.assign %[[VAL_21]] to %[[VAL_14]]#0 : i32, !fir.ref<i32>
+! CHECK:                 omp.yield
+! CHECK:               omp.terminator
 ! CHECK:             omp.terminator
 
 
diff --git a/flang/test/Lower/OpenMP/wsloop-reduction-max-hlfir.f90 b/flang/test/Lower/OpenMP/wsloop-reduction-max-hlfir.f90
index 5ea5d6626f18..f4caea5a269a 100644
--- a/flang/test/Lower/OpenMP/wsloop-reduction-max-hlfir.f90
+++ b/flang/test/Lower/OpenMP/wsloop-reduction-max-hlfir.f90
@@ -29,18 +29,20 @@
 ! CHECK:             %[[VAL_9:.*]] = arith.constant 1 : i32
 ! CHECK:             %[[VAL_10:.*]] = arith.constant 100 : i32
 ! CHECK:             %[[VAL_11:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop reduction(@max_i32 %[[VAL_4]]#0 -> %[[VAL_12:.*]] : !fir.ref<i32>)  for  (%[[VAL_13:.*]]) : i32 = (%[[VAL_9]]) to (%[[VAL_10]]) inclusive step (%[[VAL_11]]) {
-! CHECK:               fir.store %[[VAL_13]] to %[[VAL_8]]#1 : !fir.ref<i32>
-! CHECK:               %[[VAL_14:.*]]:2 = hlfir.declare %[[VAL_12]] {uniq_name = "_QFreduction_max_intEx"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
-! CHECK:               %[[VAL_15:.*]] = fir.load %[[VAL_8]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_16:.*]] = fir.convert %[[VAL_15]] : (i32) -> i64
-! CHECK:               %[[VAL_17:.*]] = hlfir.designate %[[VAL_5]]#0 (%[[VAL_16]])  : (!fir.box<!fir.array<?xi32>>, i64) -> !fir.ref<i32>
-! CHECK:               %[[VAL_18:.*]] = fir.load %[[VAL_14]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_19:.*]] = fir.load %[[VAL_17]] : !fir.ref<i32>
-! CHECK:               %[[VAL_20:.*]] = arith.cmpi sgt, %[[VAL_18]], %[[VAL_19]] : i32
-! CHECK:               %[[VAL_21:.*]] = arith.select %[[VAL_20]], %[[VAL_18]], %[[VAL_19]] : i32
-! CHECK:               hlfir.assign %[[VAL_21]] to %[[VAL_14]]#0 : i32, !fir.ref<i32>
-! CHECK:               omp.yield
+! CHECK:             omp.wsloop reduction(@max_i32 %[[VAL_4]]#0 -> %[[VAL_12:.*]] : !fir.ref<i32>) {
+! CHECK-NEXT:          omp.loop_nest (%[[VAL_13:.*]]) : i32 = (%[[VAL_9]]) to (%[[VAL_10]]) inclusive step (%[[VAL_11]]) {
+! CHECK:                 %[[VAL_14:.*]]:2 = hlfir.declare %[[VAL_12]] {uniq_name = "_QFreduction_max_intEx"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
+! CHECK:                 fir.store %[[VAL_13]] to %[[VAL_8]]#1 : !fir.ref<i32>
+! CHECK:                 %[[VAL_15:.*]] = fir.load %[[VAL_8]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_16:.*]] = fir.convert %[[VAL_15]] : (i32) -> i64
+! CHECK:                 %[[VAL_17:.*]] = hlfir.designate %[[VAL_5]]#0 (%[[VAL_16]])  : (!fir.box<!fir.array<?xi32>>, i64) -> !fir.ref<i32>
+! CHECK:                 %[[VAL_18:.*]] = fir.load %[[VAL_14]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_19:.*]] = fir.load %[[VAL_17]] : !fir.ref<i32>
+! CHECK:                 %[[VAL_20:.*]] = arith.cmpi sgt, %[[VAL_18]], %[[VAL_19]] : i32
+! CHECK:                 %[[VAL_21:.*]] = arith.select %[[VAL_20]], %[[VAL_18]], %[[VAL_19]] : i32
+! CHECK:                 hlfir.assign %[[VAL_21]] to %[[VAL_14]]#0 : i32, !fir.ref<i32>
+! CHECK:                 omp.yield
+! CHECK:               omp.terminator
 ! CHECK:             omp.terminator
 
 
diff --git a/flang/test/Lower/OpenMP/wsloop-reduction-max.f90 b/flang/test/Lower/OpenMP/wsloop-reduction-max.f90
index c9cf5cbf4f8c..ff005f32487e 100644
--- a/flang/test/Lower/OpenMP/wsloop-reduction-max.f90
+++ b/flang/test/Lower/OpenMP/wsloop-reduction-max.f90
@@ -40,18 +40,20 @@
 ! CHECK:             %[[VAL_9:.*]] = arith.constant 1 : i32
 ! CHECK:             %[[VAL_10:.*]] = arith.constant 100 : i32
 ! CHECK:             %[[VAL_11:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop reduction(@max_i32 %[[VAL_4]]#0 -> %[[VAL_12:.*]] : !fir.ref<i32>)  for  (%[[VAL_13:.*]]) : i32 = (%[[VAL_9]]) to (%[[VAL_10]]) inclusive step (%[[VAL_11]]) {
-! CHECK:               fir.store %[[VAL_13]] to %[[VAL_8]]#1 : !fir.ref<i32>
-! CHECK:               %[[VAL_14:.*]]:2 = hlfir.declare %[[VAL_12]] {uniq_name = "_QFreduction_max_intEx"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
-! CHECK:               %[[VAL_15:.*]] = fir.load %[[VAL_8]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_16:.*]] = fir.convert %[[VAL_15]] : (i32) -> i64
-! CHECK:               %[[VAL_17:.*]] = hlfir.designate %[[VAL_5]]#0 (%[[VAL_16]])  : (!fir.box<!fir.array<?xi32>>, i64) -> !fir.ref<i32>
-! CHECK:               %[[VAL_18:.*]] = fir.load %[[VAL_14]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_19:.*]] = fir.load %[[VAL_17]] : !fir.ref<i32>
-! CHECK:               %[[VAL_20:.*]] = arith.cmpi sgt, %[[VAL_18]], %[[VAL_19]] : i32
-! CHECK:               %[[VAL_21:.*]] = arith.select %[[VAL_20]], %[[VAL_18]], %[[VAL_19]] : i32
-! CHECK:               hlfir.assign %[[VAL_21]] to %[[VAL_14]]#0 : i32, !fir.ref<i32>
-! CHECK:               omp.yield
+! CHECK:             omp.wsloop reduction(@max_i32 %[[VAL_4]]#0 -> %[[VAL_12:.*]] : !fir.ref<i32>) {
+! CHECK-NEXT:          omp.loop_nest (%[[VAL_13:.*]]) : i32 = (%[[VAL_9]]) to (%[[VAL_10]]) inclusive step (%[[VAL_11]]) {
+! CHECK:                 %[[VAL_14:.*]]:2 = hlfir.declare %[[VAL_12]] {uniq_name = "_QFreduction_max_intEx"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
+! CHECK:                 fir.store %[[VAL_13]] to %[[VAL_8]]#1 : !fir.ref<i32>
+! CHECK:                 %[[VAL_15:.*]] = fir.load %[[VAL_8]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_16:.*]] = fir.convert %[[VAL_15]] : (i32) -> i64
+! CHECK:                 %[[VAL_17:.*]] = hlfir.designate %[[VAL_5]]#0 (%[[VAL_16]])  : (!fir.box<!fir.array<?xi32>>, i64) -> !fir.ref<i32>
+! CHECK:                 %[[VAL_18:.*]] = fir.load %[[VAL_14]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_19:.*]] = fir.load %[[VAL_17]] : !fir.ref<i32>
+! CHECK:                 %[[VAL_20:.*]] = arith.cmpi sgt, %[[VAL_18]], %[[VAL_19]] : i32
+! CHECK:                 %[[VAL_21:.*]] = arith.select %[[VAL_20]], %[[VAL_18]], %[[VAL_19]] : i32
+! CHECK:                 hlfir.assign %[[VAL_21]] to %[[VAL_14]]#0 : i32, !fir.ref<i32>
+! CHECK:                 omp.yield
+! CHECK:               omp.terminator
 ! CHECK:             omp.terminator
 
 ! CHECK-LABEL:   func.func @_QPreduction_max_real(
@@ -69,18 +71,20 @@
 ! CHECK:             %[[VAL_9:.*]] = arith.constant 1 : i32
 ! CHECK:             %[[VAL_10:.*]] = arith.constant 100 : i32
 ! CHECK:             %[[VAL_11:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop reduction(@max_f32 %[[VAL_4]]#0 -> %[[VAL_12:.*]] : !fir.ref<f32>)  for  (%[[VAL_13:.*]]) : i32 = (%[[VAL_9]]) to (%[[VAL_10]]) inclusive step (%[[VAL_11]]) {
-! CHECK:               fir.store %[[VAL_13]] to %[[VAL_8]]#1 : !fir.ref<i32>
-! CHECK:               %[[VAL_14:.*]]:2 = hlfir.declare %[[VAL_12]] {uniq_name = "_QFreduction_max_realEx"} : (!fir.ref<f32>) -> (!fir.ref<f32>, !fir.ref<f32>)
-! CHECK:               %[[VAL_15:.*]] = fir.load %[[VAL_8]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_16:.*]] = fir.convert %[[VAL_15]] : (i32) -> i64
-! CHECK:               %[[VAL_17:.*]] = hlfir.designate %[[VAL_5]]#0 (%[[VAL_16]])  : (!fir.box<!fir.array<?xf32>>, i64) -> !fir.ref<f32>
-! CHECK:               %[[VAL_18:.*]] = fir.load %[[VAL_17]] : !fir.ref<f32>
-! CHECK:               %[[VAL_19:.*]] = fir.load %[[VAL_14]]#0 : !fir.ref<f32>
-! CHECK:               %[[VAL_20:.*]] = arith.cmpf ogt, %[[VAL_18]], %[[VAL_19]] fastmath<contract> : f32
-! CHECK:               %[[VAL_21:.*]] = arith.select %[[VAL_20]], %[[VAL_18]], %[[VAL_19]] : f32
-! CHECK:               hlfir.assign %[[VAL_21]] to %[[VAL_14]]#0 : f32, !fir.ref<f32>
-! CHECK:               omp.yield
+! CHECK:             omp.wsloop reduction(@max_f32 %[[VAL_4]]#0 -> %[[VAL_12:.*]] : !fir.ref<f32>) {
+! CHECK-NEXT:          omp.loop_nest (%[[VAL_13:.*]]) : i32 = (%[[VAL_9]]) to (%[[VAL_10]]) inclusive step (%[[VAL_11]]) {
+! CHECK:                 %[[VAL_14:.*]]:2 = hlfir.declare %[[VAL_12]] {uniq_name = "_QFreduction_max_realEx"} : (!fir.ref<f32>) -> (!fir.ref<f32>, !fir.ref<f32>)
+! CHECK:                 fir.store %[[VAL_13]] to %[[VAL_8]]#1 : !fir.ref<i32>
+! CHECK:                 %[[VAL_15:.*]] = fir.load %[[VAL_8]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_16:.*]] = fir.convert %[[VAL_15]] : (i32) -> i64
+! CHECK:                 %[[VAL_17:.*]] = hlfir.designate %[[VAL_5]]#0 (%[[VAL_16]])  : (!fir.box<!fir.array<?xf32>>, i64) -> !fir.ref<f32>
+! CHECK:                 %[[VAL_18:.*]] = fir.load %[[VAL_17]] : !fir.ref<f32>
+! CHECK:                 %[[VAL_19:.*]] = fir.load %[[VAL_14]]#0 : !fir.ref<f32>
+! CHECK:                 %[[VAL_20:.*]] = arith.cmpf ogt, %[[VAL_18]], %[[VAL_19]] fastmath<contract> : f32
+! CHECK:                 %[[VAL_21:.*]] = arith.select %[[VAL_20]], %[[VAL_18]], %[[VAL_19]] : f32
+! CHECK:                 hlfir.assign %[[VAL_21]] to %[[VAL_14]]#0 : f32, !fir.ref<f32>
+! CHECK:                 omp.yield
+! CHECK:               omp.terminator
 ! CHECK:             omp.terminator
 ! CHECK:           omp.parallel {
 ! CHECK:             %[[VAL_30:.*]] = fir.alloca i32 {adapt.valuebyref, pinned}
@@ -88,24 +92,26 @@
 ! CHECK:             %[[VAL_32:.*]] = arith.constant 1 : i32
 ! CHECK:             %[[VAL_33:.*]] = arith.constant 100 : i32
 ! CHECK:             %[[VAL_34:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop reduction(@max_f32 %[[VAL_4]]#0 -> %[[VAL_35:.*]] : !fir.ref<f32>)  for  (%[[VAL_36:.*]]) : i32 = (%[[VAL_32]]) to (%[[VAL_33]]) inclusive step (%[[VAL_34]]) {
-! CHECK:               fir.store %[[VAL_36]] to %[[VAL_31]]#1 : !fir.ref<i32>
-! CHECK:               %[[VAL_37:.*]]:2 = hlfir.declare %[[VAL_35]] {uniq_name = "_QFreduction_max_realEx"} : (!fir.ref<f32>) -> (!fir.ref<f32>, !fir.ref<f32>)
-! CHECK:               %[[VAL_38:.*]] = fir.load %[[VAL_31]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_39:.*]] = fir.convert %[[VAL_38]] : (i32) -> i64
-! CHECK:               %[[VAL_40:.*]] = hlfir.designate %[[VAL_5]]#0 (%[[VAL_39]])  : (!fir.box<!fir.array<?xf32>>, i64) -> !fir.ref<f32>
-! CHECK:               %[[VAL_41:.*]] = fir.load %[[VAL_40]] : !fir.ref<f32>
-! CHECK:               %[[VAL_42:.*]] = fir.load %[[VAL_37]]#0 : !fir.ref<f32>
-! CHECK:               %[[VAL_43:.*]] = arith.cmpf ogt, %[[VAL_41]], %[[VAL_42]] fastmath<contract> : f32
-! CHECK:               fir.if %[[VAL_43]] {
-! CHECK:                 %[[VAL_44:.*]] = fir.load %[[VAL_31]]#0 : !fir.ref<i32>
-! CHECK:                 %[[VAL_45:.*]] = fir.convert %[[VAL_44]] : (i32) -> i64
-! CHECK:                 %[[VAL_46:.*]] = hlfir.designate %[[VAL_5]]#0 (%[[VAL_45]])  : (!fir.box<!fir.array<?xf32>>, i64) -> !fir.ref<f32>
-! CHECK:                 %[[VAL_47:.*]] = fir.load %[[VAL_46]] : !fir.ref<f32>
-! CHECK:                 hlfir.assign %[[VAL_47]] to %[[VAL_37]]#0 : f32, !fir.ref<f32>
-! CHECK:               } else {
-! CHECK:               }
-! CHECK:               omp.yield
+! CHECK:             omp.wsloop reduction(@max_f32 %[[VAL_4]]#0 -> %[[VAL_35:.*]] : !fir.ref<f32>) {
+! CHECK-NEXT:          omp.loop_nest (%[[VAL_36:.*]]) : i32 = (%[[VAL_32]]) to (%[[VAL_33]]) inclusive step (%[[VAL_34]]) {
+! CHECK:                 %[[VAL_37:.*]]:2 = hlfir.declare %[[VAL_35]] {uniq_name = "_QFreduction_max_realEx"} : (!fir.ref<f32>) -> (!fir.ref<f32>, !fir.ref<f32>)
+! CHECK:                 fir.store %[[VAL_36]] to %[[VAL_31]]#1 : !fir.ref<i32>
+! CHECK:                 %[[VAL_38:.*]] = fir.load %[[VAL_31]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_39:.*]] = fir.convert %[[VAL_38]] : (i32) -> i64
+! CHECK:                 %[[VAL_40:.*]] = hlfir.designate %[[VAL_5]]#0 (%[[VAL_39]])  : (!fir.box<!fir.array<?xf32>>, i64) -> !fir.ref<f32>
+! CHECK:                 %[[VAL_41:.*]] = fir.load %[[VAL_40]] : !fir.ref<f32>
+! CHECK:                 %[[VAL_42:.*]] = fir.load %[[VAL_37]]#0 : !fir.ref<f32>
+! CHECK:                 %[[VAL_43:.*]] = arith.cmpf ogt, %[[VAL_41]], %[[VAL_42]] fastmath<contract> : f32
+! CHECK:                 fir.if %[[VAL_43]] {
+! CHECK:                   %[[VAL_44:.*]] = fir.load %[[VAL_31]]#0 : !fir.ref<i32>
+! CHECK:                   %[[VAL_45:.*]] = fir.convert %[[VAL_44]] : (i32) -> i64
+! CHECK:                   %[[VAL_46:.*]] = hlfir.designate %[[VAL_5]]#0 (%[[VAL_45]])  : (!fir.box<!fir.array<?xf32>>, i64) -> !fir.ref<f32>
+! CHECK:                   %[[VAL_47:.*]] = fir.load %[[VAL_46]] : !fir.ref<f32>
+! CHECK:                   hlfir.assign %[[VAL_47]] to %[[VAL_37]]#0 : f32, !fir.ref<f32>
+! CHECK:                 } else {
+! CHECK:                 }
+! CHECK:                 omp.yield
+! CHECK:               omp.terminator
 ! CHECK:             omp.terminator
 
 
diff --git a/flang/test/Lower/OpenMP/wsloop-reduction-min-byref.f90 b/flang/test/Lower/OpenMP/wsloop-reduction-min-byref.f90
index 84a376b46b8f..9787512ab078 100644
--- a/flang/test/Lower/OpenMP/wsloop-reduction-min-byref.f90
+++ b/flang/test/Lower/OpenMP/wsloop-reduction-min-byref.f90
@@ -46,18 +46,20 @@
 ! CHECK:             %[[VAL_9:.*]] = arith.constant 1 : i32
 ! CHECK:             %[[VAL_10:.*]] = arith.constant 100 : i32
 ! CHECK:             %[[VAL_11:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop byref reduction(@min_byref_i32 %[[VAL_4]]#0 -> %[[VAL_12:.*]] : !fir.ref<i32>)  for  (%[[VAL_13:.*]]) : i32 = (%[[VAL_9]]) to (%[[VAL_10]]) inclusive step (%[[VAL_11]]) {
-! CHECK:               fir.store %[[VAL_13]] to %[[VAL_8]]#1 : !fir.ref<i32>
-! CHECK:               %[[VAL_14:.*]]:2 = hlfir.declare %[[VAL_12]] {uniq_name = "_QFreduction_min_intEx"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
-! CHECK:               %[[VAL_15:.*]] = fir.load %[[VAL_8]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_16:.*]] = fir.convert %[[VAL_15]] : (i32) -> i64
-! CHECK:               %[[VAL_17:.*]] = hlfir.designate %[[VAL_5]]#0 (%[[VAL_16]])  : (!fir.box<!fir.array<?xi32>>, i64) -> !fir.ref<i32>
-! CHECK:               %[[VAL_18:.*]] = fir.load %[[VAL_14]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_19:.*]] = fir.load %[[VAL_17]] : !fir.ref<i32>
-! CHECK:               %[[VAL_20:.*]] = arith.cmpi slt, %[[VAL_18]], %[[VAL_19]] : i32
-! CHECK:               %[[VAL_21:.*]] = arith.select %[[VAL_20]], %[[VAL_18]], %[[VAL_19]] : i32
-! CHECK:               hlfir.assign %[[VAL_21]] to %[[VAL_14]]#0 : i32, !fir.ref<i32>
-! CHECK:               omp.yield
+! CHECK:             omp.wsloop byref reduction(@min_byref_i32 %[[VAL_4]]#0 -> %[[VAL_12:.*]] : !fir.ref<i32>) {
+! CHECK-NEXT:          omp.loop_nest (%[[VAL_13:.*]]) : i32 = (%[[VAL_9]]) to (%[[VAL_10]]) inclusive step (%[[VAL_11]]) {
+! CHECK:                 %[[VAL_14:.*]]:2 = hlfir.declare %[[VAL_12]] {uniq_name = "_QFreduction_min_intEx"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
+! CHECK:                 fir.store %[[VAL_13]] to %[[VAL_8]]#1 : !fir.ref<i32>
+! CHECK:                 %[[VAL_15:.*]] = fir.load %[[VAL_8]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_16:.*]] = fir.convert %[[VAL_15]] : (i32) -> i64
+! CHECK:                 %[[VAL_17:.*]] = hlfir.designate %[[VAL_5]]#0 (%[[VAL_16]])  : (!fir.box<!fir.array<?xi32>>, i64) -> !fir.ref<i32>
+! CHECK:                 %[[VAL_18:.*]] = fir.load %[[VAL_14]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_19:.*]] = fir.load %[[VAL_17]] : !fir.ref<i32>
+! CHECK:                 %[[VAL_20:.*]] = arith.cmpi slt, %[[VAL_18]], %[[VAL_19]] : i32
+! CHECK:                 %[[VAL_21:.*]] = arith.select %[[VAL_20]], %[[VAL_18]], %[[VAL_19]] : i32
+! CHECK:                 hlfir.assign %[[VAL_21]] to %[[VAL_14]]#0 : i32, !fir.ref<i32>
+! CHECK:                 omp.yield
+! CHECK:               omp.terminator
 ! CHECK:             omp.terminator
 
 ! CHECK-LABEL:   func.func @_QPreduction_min_real(
@@ -75,19 +77,21 @@
 ! CHECK:             %[[VAL_9:.*]] = arith.constant 1 : i32
 ! CHECK:             %[[VAL_10:.*]] = arith.constant 100 : i32
 ! CHECK:             %[[VAL_11:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop byref reduction(@min_byref_f32 %[[VAL_4]]#0 -> %[[VAL_12:.*]] : !fir.ref<f32>)  for  (%[[VAL_13:.*]]) : i32 = (%[[VAL_9]]) to (%[[VAL_10]]) inclusive step (%[[VAL_11]]) {
-! CHECK:               fir.store %[[VAL_13]] to %[[VAL_8]]#1 : !fir.ref<i32>
-! CHECK:               %[[VAL_14:.*]]:2 = hlfir.declare %[[VAL_12]] {uniq_name = "_QFreduction_min_realEx"} : (!fir.ref<f32>) -> (!fir.ref<f32>, !fir.ref<f32>)
-! CHECK:               %[[VAL_15:.*]] = fir.load %[[VAL_8]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_16:.*]] = fir.convert %[[VAL_15]] : (i32) -> i64
-! CHECK:               %[[VAL_17:.*]] = hlfir.designate %[[VAL_5]]#0 (%[[VAL_16]])  : (!fir.box<!fir.array<?xf32>>, i64) -> !fir.ref<f32>
-! CHECK:               %[[VAL_18:.*]] = fir.load %[[VAL_17]] : !fir.ref<f32>
-! CHECK:               %[[VAL_19:.*]] = fir.load %[[VAL_14]]#0 : !fir.ref<f32>
-! CHECK:               %[[VAL_20:.*]] = arith.cmpf olt, %[[VAL_18]], %[[VAL_19]] fastmath<contract> : f32
-! CHECK:               %[[VAL_21:.*]] = arith.select %[[VAL_20]], %[[VAL_18]], %[[VAL_19]] : f32
-! CHECK:               hlfir.assign %[[VAL_21]] to %[[VAL_14]]#0 : f32, !fir.ref<f32>
-! CHECK:               omp.yield
-! CHECK:             }
+! CHECK:             omp.wsloop byref reduction(@min_byref_f32 %[[VAL_4]]#0 -> %[[VAL_12:.*]] : !fir.ref<f32>) {
+! CHECK-NEXT:          omp.loop_nest (%[[VAL_13:.*]]) : i32 = (%[[VAL_9]]) to (%[[VAL_10]]) inclusive step (%[[VAL_11]]) {
+! CHECK:                 %[[VAL_14:.*]]:2 = hlfir.declare %[[VAL_12]] {uniq_name = "_QFreduction_min_realEx"} : (!fir.ref<f32>) -> (!fir.ref<f32>, !fir.ref<f32>)
+! CHECK:                 fir.store %[[VAL_13]] to %[[VAL_8]]#1 : !fir.ref<i32>
+! CHECK:                 %[[VAL_15:.*]] = fir.load %[[VAL_8]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_16:.*]] = fir.convert %[[VAL_15]] : (i32) -> i64
+! CHECK:                 %[[VAL_17:.*]] = hlfir.designate %[[VAL_5]]#0 (%[[VAL_16]])  : (!fir.box<!fir.array<?xf32>>, i64) -> !fir.ref<f32>
+! CHECK:                 %[[VAL_18:.*]] = fir.load %[[VAL_17]] : !fir.ref<f32>
+! CHECK:                 %[[VAL_19:.*]] = fir.load %[[VAL_14]]#0 : !fir.ref<f32>
+! CHECK:                 %[[VAL_20:.*]] = arith.cmpf olt, %[[VAL_18]], %[[VAL_19]] fastmath<contract> : f32
+! CHECK:                 %[[VAL_21:.*]] = arith.select %[[VAL_20]], %[[VAL_18]], %[[VAL_19]] : f32
+! CHECK:                 hlfir.assign %[[VAL_21]] to %[[VAL_14]]#0 : f32, !fir.ref<f32>
+! CHECK:                 omp.yield
+! CHECK:               }
+! CHECK:               omp.terminator
 ! CHECK:             omp.terminator
 ! CHECK:           }
 ! CHECK:           omp.parallel {
@@ -96,24 +100,26 @@
 ! CHECK:             %[[VAL_32:.*]] = arith.constant 1 : i32
 ! CHECK:             %[[VAL_33:.*]] = arith.constant 100 : i32
 ! CHECK:             %[[VAL_34:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop byref reduction(@min_byref_f32 %[[VAL_4]]#0 -> %[[VAL_35:.*]] : !fir.ref<f32>)  for  (%[[VAL_36:.*]]) : i32 = (%[[VAL_32]]) to (%[[VAL_33]]) inclusive step (%[[VAL_34]]) {
-! CHECK:               fir.store %[[VAL_36]] to %[[VAL_31]]#1 : !fir.ref<i32>
-! CHECK:               %[[VAL_37:.*]]:2 = hlfir.declare %[[VAL_35]] {uniq_name = "_QFreduction_min_realEx"} : (!fir.ref<f32>) -> (!fir.ref<f32>, !fir.ref<f32>)
-! CHECK:               %[[VAL_38:.*]] = fir.load %[[VAL_31]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_39:.*]] = fir.convert %[[VAL_38]] : (i32) -> i64
-! CHECK:               %[[VAL_40:.*]] = hlfir.designate %[[VAL_5]]#0 (%[[VAL_39]])  : (!fir.box<!fir.array<?xf32>>, i64) -> !fir.ref<f32>
-! CHECK:               %[[VAL_41:.*]] = fir.load %[[VAL_40]] : !fir.ref<f32>
-! CHECK:               %[[VAL_42:.*]] = fir.load %[[VAL_37]]#0 : !fir.ref<f32>
-! CHECK:               %[[VAL_43:.*]] = arith.cmpf ogt, %[[VAL_41]], %[[VAL_42]] fastmath<contract> : f32
-! CHECK:               fir.if %[[VAL_43]] {
-! CHECK:                 %[[VAL_44:.*]] = fir.load %[[VAL_31]]#0 : !fir.ref<i32>
-! CHECK:                 %[[VAL_45:.*]] = fir.convert %[[VAL_44]] : (i32) -> i64
-! CHECK:                 %[[VAL_46:.*]] = hlfir.designate %[[VAL_5]]#0 (%[[VAL_45]])  : (!fir.box<!fir.array<?xf32>>, i64) -> !fir.ref<f32>
-! CHECK:                 %[[VAL_47:.*]] = fir.load %[[VAL_46]] : !fir.ref<f32>
-! CHECK:                 hlfir.assign %[[VAL_47]] to %[[VAL_37]]#0 : f32, !fir.ref<f32>
-! CHECK:               } else {
-! CHECK:               }
-! CHECK:               omp.yield
+! CHECK:             omp.wsloop byref reduction(@min_byref_f32 %[[VAL_4]]#0 -> %[[VAL_35:.*]] : !fir.ref<f32>) {
+! CHECK-NEXT:          omp.loop_nest (%[[VAL_36:.*]]) : i32 = (%[[VAL_32]]) to (%[[VAL_33]]) inclusive step (%[[VAL_34]]) {
+! CHECK:                 %[[VAL_37:.*]]:2 = hlfir.declare %[[VAL_35]] {uniq_name = "_QFreduction_min_realEx"} : (!fir.ref<f32>) -> (!fir.ref<f32>, !fir.ref<f32>)
+! CHECK:                 fir.store %[[VAL_36]] to %[[VAL_31]]#1 : !fir.ref<i32>
+! CHECK:                 %[[VAL_38:.*]] = fir.load %[[VAL_31]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_39:.*]] = fir.convert %[[VAL_38]] : (i32) -> i64
+! CHECK:                 %[[VAL_40:.*]] = hlfir.designate %[[VAL_5]]#0 (%[[VAL_39]])  : (!fir.box<!fir.array<?xf32>>, i64) -> !fir.ref<f32>
+! CHECK:                 %[[VAL_41:.*]] = fir.load %[[VAL_40]] : !fir.ref<f32>
+! CHECK:                 %[[VAL_42:.*]] = fir.load %[[VAL_37]]#0 : !fir.ref<f32>
+! CHECK:                 %[[VAL_43:.*]] = arith.cmpf ogt, %[[VAL_41]], %[[VAL_42]] fastmath<contract> : f32
+! CHECK:                 fir.if %[[VAL_43]] {
+! CHECK:                   %[[VAL_44:.*]] = fir.load %[[VAL_31]]#0 : !fir.ref<i32>
+! CHECK:                   %[[VAL_45:.*]] = fir.convert %[[VAL_44]] : (i32) -> i64
+! CHECK:                   %[[VAL_46:.*]] = hlfir.designate %[[VAL_5]]#0 (%[[VAL_45]])  : (!fir.box<!fir.array<?xf32>>, i64) -> !fir.ref<f32>
+! CHECK:                   %[[VAL_47:.*]] = fir.load %[[VAL_46]] : !fir.ref<f32>
+! CHECK:                   hlfir.assign %[[VAL_47]] to %[[VAL_37]]#0 : f32, !fir.ref<f32>
+! CHECK:                 } else {
+! CHECK:                 }
+! CHECK:                 omp.yield
+! CHECK:               omp.terminator
 ! CHECK:             omp.terminator
 
 
diff --git a/flang/test/Lower/OpenMP/wsloop-reduction-min.f90 b/flang/test/Lower/OpenMP/wsloop-reduction-min.f90
index 3ba279acd14c..801ef99480a2 100644
--- a/flang/test/Lower/OpenMP/wsloop-reduction-min.f90
+++ b/flang/test/Lower/OpenMP/wsloop-reduction-min.f90
@@ -40,18 +40,20 @@
 ! CHECK:             %[[VAL_9:.*]] = arith.constant 1 : i32
 ! CHECK:             %[[VAL_10:.*]] = arith.constant 100 : i32
 ! CHECK:             %[[VAL_11:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop reduction(@min_i32 %[[VAL_4]]#0 -> %[[VAL_12:.*]] : !fir.ref<i32>)  for  (%[[VAL_13:.*]]) : i32 = (%[[VAL_9]]) to (%[[VAL_10]]) inclusive step (%[[VAL_11]]) {
-! CHECK:               fir.store %[[VAL_13]] to %[[VAL_8]]#1 : !fir.ref<i32>
-! CHECK:               %[[VAL_14:.*]]:2 = hlfir.declare %[[VAL_12]] {uniq_name = "_QFreduction_min_intEx"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
-! CHECK:               %[[VAL_15:.*]] = fir.load %[[VAL_8]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_16:.*]] = fir.convert %[[VAL_15]] : (i32) -> i64
-! CHECK:               %[[VAL_17:.*]] = hlfir.designate %[[VAL_5]]#0 (%[[VAL_16]])  : (!fir.box<!fir.array<?xi32>>, i64) -> !fir.ref<i32>
-! CHECK:               %[[VAL_18:.*]] = fir.load %[[VAL_14]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_19:.*]] = fir.load %[[VAL_17]] : !fir.ref<i32>
-! CHECK:               %[[VAL_20:.*]] = arith.cmpi slt, %[[VAL_18]], %[[VAL_19]] : i32
-! CHECK:               %[[VAL_21:.*]] = arith.select %[[VAL_20]], %[[VAL_18]], %[[VAL_19]] : i32
-! CHECK:               hlfir.assign %[[VAL_21]] to %[[VAL_14]]#0 : i32, !fir.ref<i32>
-! CHECK:               omp.yield
+! CHECK:             omp.wsloop reduction(@min_i32 %[[VAL_4]]#0 -> %[[VAL_12:.*]] : !fir.ref<i32>) {
+! CHECK-NEXT:          omp.loop_nest (%[[VAL_13:.*]]) : i32 = (%[[VAL_9]]) to (%[[VAL_10]]) inclusive step (%[[VAL_11]]) {
+! CHECK:                 %[[VAL_14:.*]]:2 = hlfir.declare %[[VAL_12]] {uniq_name = "_QFreduction_min_intEx"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
+! CHECK:                 fir.store %[[VAL_13]] to %[[VAL_8]]#1 : !fir.ref<i32>
+! CHECK:                 %[[VAL_15:.*]] = fir.load %[[VAL_8]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_16:.*]] = fir.convert %[[VAL_15]] : (i32) -> i64
+! CHECK:                 %[[VAL_17:.*]] = hlfir.designate %[[VAL_5]]#0 (%[[VAL_16]])  : (!fir.box<!fir.array<?xi32>>, i64) -> !fir.ref<i32>
+! CHECK:                 %[[VAL_18:.*]] = fir.load %[[VAL_14]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_19:.*]] = fir.load %[[VAL_17]] : !fir.ref<i32>
+! CHECK:                 %[[VAL_20:.*]] = arith.cmpi slt, %[[VAL_18]], %[[VAL_19]] : i32
+! CHECK:                 %[[VAL_21:.*]] = arith.select %[[VAL_20]], %[[VAL_18]], %[[VAL_19]] : i32
+! CHECK:                 hlfir.assign %[[VAL_21]] to %[[VAL_14]]#0 : i32, !fir.ref<i32>
+! CHECK:                 omp.yield
+! CHECK:               omp.terminator
 ! CHECK:             omp.terminator
 
 ! CHECK-LABEL:   func.func @_QPreduction_min_real(
@@ -69,19 +71,21 @@
 ! CHECK:             %[[VAL_9:.*]] = arith.constant 1 : i32
 ! CHECK:             %[[VAL_10:.*]] = arith.constant 100 : i32
 ! CHECK:             %[[VAL_11:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop reduction(@min_f32 %[[VAL_4]]#0 -> %[[VAL_12:.*]] : !fir.ref<f32>)  for  (%[[VAL_13:.*]]) : i32 = (%[[VAL_9]]) to (%[[VAL_10]]) inclusive step (%[[VAL_11]]) {
-! CHECK:               fir.store %[[VAL_13]] to %[[VAL_8]]#1 : !fir.ref<i32>
-! CHECK:               %[[VAL_14:.*]]:2 = hlfir.declare %[[VAL_12]] {uniq_name = "_QFreduction_min_realEx"} : (!fir.ref<f32>) -> (!fir.ref<f32>, !fir.ref<f32>)
-! CHECK:               %[[VAL_15:.*]] = fir.load %[[VAL_8]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_16:.*]] = fir.convert %[[VAL_15]] : (i32) -> i64
-! CHECK:               %[[VAL_17:.*]] = hlfir.designate %[[VAL_5]]#0 (%[[VAL_16]])  : (!fir.box<!fir.array<?xf32>>, i64) -> !fir.ref<f32>
-! CHECK:               %[[VAL_18:.*]] = fir.load %[[VAL_17]] : !fir.ref<f32>
-! CHECK:               %[[VAL_19:.*]] = fir.load %[[VAL_14]]#0 : !fir.ref<f32>
-! CHECK:               %[[VAL_20:.*]] = arith.cmpf olt, %[[VAL_18]], %[[VAL_19]] fastmath<contract> : f32
-! CHECK:               %[[VAL_21:.*]] = arith.select %[[VAL_20]], %[[VAL_18]], %[[VAL_19]] : f32
-! CHECK:               hlfir.assign %[[VAL_21]] to %[[VAL_14]]#0 : f32, !fir.ref<f32>
-! CHECK:               omp.yield
-! CHECK:             }
+! CHECK:             omp.wsloop reduction(@min_f32 %[[VAL_4]]#0 -> %[[VAL_12:.*]] : !fir.ref<f32>) {
+! CHECK-NEXT:          omp.loop_nest (%[[VAL_13:.*]]) : i32 = (%[[VAL_9]]) to (%[[VAL_10]]) inclusive step (%[[VAL_11]]) {
+! CHECK:                 %[[VAL_14:.*]]:2 = hlfir.declare %[[VAL_12]] {uniq_name = "_QFreduction_min_realEx"} : (!fir.ref<f32>) -> (!fir.ref<f32>, !fir.ref<f32>)
+! CHECK:                 fir.store %[[VAL_13]] to %[[VAL_8]]#1 : !fir.ref<i32>
+! CHECK:                 %[[VAL_15:.*]] = fir.load %[[VAL_8]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_16:.*]] = fir.convert %[[VAL_15]] : (i32) -> i64
+! CHECK:                 %[[VAL_17:.*]] = hlfir.designate %[[VAL_5]]#0 (%[[VAL_16]])  : (!fir.box<!fir.array<?xf32>>, i64) -> !fir.ref<f32>
+! CHECK:                 %[[VAL_18:.*]] = fir.load %[[VAL_17]] : !fir.ref<f32>
+! CHECK:                 %[[VAL_19:.*]] = fir.load %[[VAL_14]]#0 : !fir.ref<f32>
+! CHECK:                 %[[VAL_20:.*]] = arith.cmpf olt, %[[VAL_18]], %[[VAL_19]] fastmath<contract> : f32
+! CHECK:                 %[[VAL_21:.*]] = arith.select %[[VAL_20]], %[[VAL_18]], %[[VAL_19]] : f32
+! CHECK:                 hlfir.assign %[[VAL_21]] to %[[VAL_14]]#0 : f32, !fir.ref<f32>
+! CHECK:                 omp.yield
+! CHECK:               }
+! CHECK:               omp.terminator
 ! CHECK:             omp.terminator
 ! CHECK:           }
 ! CHECK:           omp.parallel {
@@ -90,24 +94,26 @@
 ! CHECK:             %[[VAL_32:.*]] = arith.constant 1 : i32
 ! CHECK:             %[[VAL_33:.*]] = arith.constant 100 : i32
 ! CHECK:             %[[VAL_34:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop reduction(@min_f32 %[[VAL_4]]#0 -> %[[VAL_35:.*]] : !fir.ref<f32>)  for  (%[[VAL_36:.*]]) : i32 = (%[[VAL_32]]) to (%[[VAL_33]]) inclusive step (%[[VAL_34]]) {
-! CHECK:               fir.store %[[VAL_36]] to %[[VAL_31]]#1 : !fir.ref<i32>
-! CHECK:               %[[VAL_37:.*]]:2 = hlfir.declare %[[VAL_35]] {uniq_name = "_QFreduction_min_realEx"} : (!fir.ref<f32>) -> (!fir.ref<f32>, !fir.ref<f32>)
-! CHECK:               %[[VAL_38:.*]] = fir.load %[[VAL_31]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_39:.*]] = fir.convert %[[VAL_38]] : (i32) -> i64
-! CHECK:               %[[VAL_40:.*]] = hlfir.designate %[[VAL_5]]#0 (%[[VAL_39]])  : (!fir.box<!fir.array<?xf32>>, i64) -> !fir.ref<f32>
-! CHECK:               %[[VAL_41:.*]] = fir.load %[[VAL_40]] : !fir.ref<f32>
-! CHECK:               %[[VAL_42:.*]] = fir.load %[[VAL_37]]#0 : !fir.ref<f32>
-! CHECK:               %[[VAL_43:.*]] = arith.cmpf ogt, %[[VAL_41]], %[[VAL_42]] fastmath<contract> : f32
-! CHECK:               fir.if %[[VAL_43]] {
-! CHECK:                 %[[VAL_44:.*]] = fir.load %[[VAL_31]]#0 : !fir.ref<i32>
-! CHECK:                 %[[VAL_45:.*]] = fir.convert %[[VAL_44]] : (i32) -> i64
-! CHECK:                 %[[VAL_46:.*]] = hlfir.designate %[[VAL_5]]#0 (%[[VAL_45]])  : (!fir.box<!fir.array<?xf32>>, i64) -> !fir.ref<f32>
-! CHECK:                 %[[VAL_47:.*]] = fir.load %[[VAL_46]] : !fir.ref<f32>
-! CHECK:                 hlfir.assign %[[VAL_47]] to %[[VAL_37]]#0 : f32, !fir.ref<f32>
-! CHECK:               } else {
-! CHECK:               }
-! CHECK:               omp.yield
+! CHECK:             omp.wsloop reduction(@min_f32 %[[VAL_4]]#0 -> %[[VAL_35:.*]] : !fir.ref<f32>) {
+! CHECK-NEXT:          omp.loop_nest (%[[VAL_36:.*]]) : i32 = (%[[VAL_32]]) to (%[[VAL_33]]) inclusive step (%[[VAL_34]]) {
+! CHECK:                 %[[VAL_37:.*]]:2 = hlfir.declare %[[VAL_35]] {uniq_name = "_QFreduction_min_realEx"} : (!fir.ref<f32>) -> (!fir.ref<f32>, !fir.ref<f32>)
+! CHECK:                 fir.store %[[VAL_36]] to %[[VAL_31]]#1 : !fir.ref<i32>
+! CHECK:                 %[[VAL_38:.*]] = fir.load %[[VAL_31]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_39:.*]] = fir.convert %[[VAL_38]] : (i32) -> i64
+! CHECK:                 %[[VAL_40:.*]] = hlfir.designate %[[VAL_5]]#0 (%[[VAL_39]])  : (!fir.box<!fir.array<?xf32>>, i64) -> !fir.ref<f32>
+! CHECK:                 %[[VAL_41:.*]] = fir.load %[[VAL_40]] : !fir.ref<f32>
+! CHECK:                 %[[VAL_42:.*]] = fir.load %[[VAL_37]]#0 : !fir.ref<f32>
+! CHECK:                 %[[VAL_43:.*]] = arith.cmpf ogt, %[[VAL_41]], %[[VAL_42]] fastmath<contract> : f32
+! CHECK:                 fir.if %[[VAL_43]] {
+! CHECK:                   %[[VAL_44:.*]] = fir.load %[[VAL_31]]#0 : !fir.ref<i32>
+! CHECK:                   %[[VAL_45:.*]] = fir.convert %[[VAL_44]] : (i32) -> i64
+! CHECK:                   %[[VAL_46:.*]] = hlfir.designate %[[VAL_5]]#0 (%[[VAL_45]])  : (!fir.box<!fir.array<?xf32>>, i64) -> !fir.ref<f32>
+! CHECK:                   %[[VAL_47:.*]] = fir.load %[[VAL_46]] : !fir.ref<f32>
+! CHECK:                   hlfir.assign %[[VAL_47]] to %[[VAL_37]]#0 : f32, !fir.ref<f32>
+! CHECK:                 } else {
+! CHECK:                 }
+! CHECK:                 omp.yield
+! CHECK:               omp.terminator
 ! CHECK:             omp.terminator
 
 
diff --git a/flang/test/Lower/OpenMP/wsloop-reduction-min2.f90 b/flang/test/Lower/OpenMP/wsloop-reduction-min2.f90
index 0138a9578206..a4c99f190dd2 100644
--- a/flang/test/Lower/OpenMP/wsloop-reduction-min2.f90
+++ b/flang/test/Lower/OpenMP/wsloop-reduction-min2.f90
@@ -39,12 +39,14 @@ end program
 ! CHECK:             %[[VAL_6:.*]] = arith.constant 0 : i32
 ! CHECK:             %[[VAL_7:.*]] = arith.constant 10 : i32
 ! CHECK:             %[[VAL_8:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop reduction(@min_i32 %[[VAL_3]]#0 -> %[[VAL_9:.*]] : !fir.ref<i32>)  for  (%[[VAL_10:.*]]) : i32 = (%[[VAL_6]]) to (%[[VAL_7]]) inclusive step (%[[VAL_8]]) {
-! CHECK:               fir.store %[[VAL_10]] to %[[VAL_5]]#1 : !fir.ref<i32>
-! CHECK:               %[[VAL_11:.*]]:2 = hlfir.declare %[[VAL_9]] {uniq_name = "_QFEr"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
-! CHECK:               %[[VAL_12:.*]] = fir.load %[[VAL_5]]#0 : !fir.ref<i32>
-! CHECK:               hlfir.assign %[[VAL_12]] to %[[VAL_11]]#0 : i32, !fir.ref<i32>
-! CHECK:               omp.yield
-! CHECK:             }
+! CHECK:             omp.wsloop reduction(@min_i32 %[[VAL_3]]#0 -> %[[VAL_9:.*]] : !fir.ref<i32>) {
+! CHECK-NEXT:          omp.loop_nest (%[[VAL_10:.*]]) : i32 = (%[[VAL_6]]) to (%[[VAL_7]]) inclusive step (%[[VAL_8]]) {
+! CHECK:                 %[[VAL_11:.*]]:2 = hlfir.declare %[[VAL_9]] {uniq_name = "_QFEr"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
+! CHECK:                 fir.store %[[VAL_10]] to %[[VAL_5]]#1 : !fir.ref<i32>
+! CHECK:                 %[[VAL_12:.*]] = fir.load %[[VAL_5]]#0 : !fir.ref<i32>
+! CHECK:                 hlfir.assign %[[VAL_12]] to %[[VAL_11]]#0 : i32, !fir.ref<i32>
+! CHECK:                 omp.yield
+! CHECK:               }
+! CHECK:               omp.terminator
 ! CHECK:             omp.terminator
 ! CHECK:           }
diff --git a/flang/test/Lower/OpenMP/wsloop-reduction-mul-byref.f90 b/flang/test/Lower/OpenMP/wsloop-reduction-mul-byref.f90
index a2829948d472..7c538cdd470f 100644
--- a/flang/test/Lower/OpenMP/wsloop-reduction-mul-byref.f90
+++ b/flang/test/Lower/OpenMP/wsloop-reduction-mul-byref.f90
@@ -85,14 +85,16 @@
 ! CHECK:             %[[VAL_7:.*]] = arith.constant 1 : i32
 ! CHECK:             %[[VAL_8:.*]] = arith.constant 10 : i32
 ! CHECK:             %[[VAL_9:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop byref reduction(@multiply_reduction_byref_i32 %[[VAL_3]]#0 -> %[[VAL_10:.*]] : !fir.ref<i32>)  for  (%[[VAL_11:.*]]) : i32 = (%[[VAL_7]]) to (%[[VAL_8]]) inclusive step (%[[VAL_9]]) {
-! CHECK:               fir.store %[[VAL_11]] to %[[VAL_6]]#1 : !fir.ref<i32>
-! CHECK:               %[[VAL_12:.*]]:2 = hlfir.declare %[[VAL_10]] {uniq_name = "_QFsimple_int_reductionEx"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
-! CHECK:               %[[VAL_13:.*]] = fir.load %[[VAL_12]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_14:.*]] = fir.load %[[VAL_6]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_15:.*]] = arith.muli %[[VAL_13]], %[[VAL_14]] : i32
-! CHECK:               hlfir.assign %[[VAL_15]] to %[[VAL_12]]#0 : i32, !fir.ref<i32>
-! CHECK:               omp.yield
+! CHECK:             omp.wsloop byref reduction(@multiply_reduction_byref_i32 %[[VAL_3]]#0 -> %[[VAL_10:.*]] : !fir.ref<i32>) {
+! CHECK-NEXT:          omp.loop_nest (%[[VAL_11:.*]]) : i32 = (%[[VAL_7]]) to (%[[VAL_8]]) inclusive step (%[[VAL_9]]) {
+! CHECK:                 %[[VAL_12:.*]]:2 = hlfir.declare %[[VAL_10]] {uniq_name = "_QFsimple_int_reductionEx"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
+! CHECK:                 fir.store %[[VAL_11]] to %[[VAL_6]]#1 : !fir.ref<i32>
+! CHECK:                 %[[VAL_13:.*]] = fir.load %[[VAL_12]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_14:.*]] = fir.load %[[VAL_6]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_15:.*]] = arith.muli %[[VAL_13]], %[[VAL_14]] : i32
+! CHECK:                 hlfir.assign %[[VAL_15]] to %[[VAL_12]]#0 : i32, !fir.ref<i32>
+! CHECK:                 omp.yield
+! CHECK:               omp.terminator
 ! CHECK:             omp.terminator
 ! CHECK:           return
 
@@ -121,15 +123,17 @@ end subroutine
 ! CHECK:             %[[VAL_7:.*]] = arith.constant 1 : i32
 ! CHECK:             %[[VAL_8:.*]] = arith.constant 10 : i32
 ! CHECK:             %[[VAL_9:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop byref reduction(@multiply_reduction_byref_f32 %[[VAL_3]]#0 -> %[[VAL_10:.*]] : !fir.ref<f32>)  for  (%[[VAL_11:.*]]) : i32 = (%[[VAL_7]]) to (%[[VAL_8]]) inclusive step (%[[VAL_9]]) {
-! CHECK:               fir.store %[[VAL_11]] to %[[VAL_6]]#1 : !fir.ref<i32>
-! CHECK:               %[[VAL_12:.*]]:2 = hlfir.declare %[[VAL_10]] {uniq_name = "_QFsimple_real_reductionEx"} : (!fir.ref<f32>) -> (!fir.ref<f32>, !fir.ref<f32>)
-! CHECK:               %[[VAL_13:.*]] = fir.load %[[VAL_12]]#0 : !fir.ref<f32>
-! CHECK:               %[[VAL_14:.*]] = fir.load %[[VAL_6]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_15:.*]] = fir.convert %[[VAL_14]] : (i32) -> f32
-! CHECK:               %[[VAL_16:.*]] = arith.mulf %[[VAL_13]], %[[VAL_15]] fastmath<contract> : f32
-! CHECK:               hlfir.assign %[[VAL_16]] to %[[VAL_12]]#0 : f32, !fir.ref<f32>
-! CHECK:               omp.yield
+! CHECK:             omp.wsloop byref reduction(@multiply_reduction_byref_f32 %[[VAL_3]]#0 -> %[[VAL_10:.*]] : !fir.ref<f32>) {
+! CHECK-NEXT:          omp.loop_nest (%[[VAL_11:.*]]) : i32 = (%[[VAL_7]]) to (%[[VAL_8]]) inclusive step (%[[VAL_9]]) {
+! CHECK:                 %[[VAL_12:.*]]:2 = hlfir.declare %[[VAL_10]] {uniq_name = "_QFsimple_real_reductionEx"} : (!fir.ref<f32>) -> (!fir.ref<f32>, !fir.ref<f32>)
+! CHECK:                 fir.store %[[VAL_11]] to %[[VAL_6]]#1 : !fir.ref<i32>
+! CHECK:                 %[[VAL_13:.*]] = fir.load %[[VAL_12]]#0 : !fir.ref<f32>
+! CHECK:                 %[[VAL_14:.*]] = fir.load %[[VAL_6]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_15:.*]] = fir.convert %[[VAL_14]] : (i32) -> f32
+! CHECK:                 %[[VAL_16:.*]] = arith.mulf %[[VAL_13]], %[[VAL_15]] fastmath<contract> : f32
+! CHECK:                 hlfir.assign %[[VAL_16]] to %[[VAL_12]]#0 : f32, !fir.ref<f32>
+! CHECK:                 omp.yield
+! CHECK:               omp.terminator
 ! CHECK:             omp.terminator
 ! CHECK:           return
 
@@ -158,14 +162,16 @@ end subroutine
 ! CHECK:             %[[VAL_7:.*]] = arith.constant 1 : i32
 ! CHECK:             %[[VAL_8:.*]] = arith.constant 10 : i32
 ! CHECK:             %[[VAL_9:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop byref reduction(@multiply_reduction_byref_i32 %[[VAL_3]]#0 -> %[[VAL_10:.*]] : !fir.ref<i32>)  for  (%[[VAL_11:.*]]) : i32 = (%[[VAL_7]]) to (%[[VAL_8]]) inclusive step (%[[VAL_9]]) {
-! CHECK:               fir.store %[[VAL_11]] to %[[VAL_6]]#1 : !fir.ref<i32>
-! CHECK:               %[[VAL_12:.*]]:2 = hlfir.declare %[[VAL_10]] {uniq_name = "_QFsimple_int_reduction_switch_orderEx"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
-! CHECK:               %[[VAL_13:.*]] = fir.load %[[VAL_6]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_14:.*]] = fir.load %[[VAL_12]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_15:.*]] = arith.muli %[[VAL_13]], %[[VAL_14]] : i32
-! CHECK:               hlfir.assign %[[VAL_15]] to %[[VAL_12]]#0 : i32, !fir.ref<i32>
-! CHECK:               omp.yield
+! CHECK:             omp.wsloop byref reduction(@multiply_reduction_byref_i32 %[[VAL_3]]#0 -> %[[VAL_10:.*]] : !fir.ref<i32>) {
+! CHECK-NEXT:          omp.loop_nest (%[[VAL_11:.*]]) : i32 = (%[[VAL_7]]) to (%[[VAL_8]]) inclusive step (%[[VAL_9]]) {
+! CHECK:                 %[[VAL_12:.*]]:2 = hlfir.declare %[[VAL_10]] {uniq_name = "_QFsimple_int_reduction_switch_orderEx"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
+! CHECK:                 fir.store %[[VAL_11]] to %[[VAL_6]]#1 : !fir.ref<i32>
+! CHECK:                 %[[VAL_13:.*]] = fir.load %[[VAL_6]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_14:.*]] = fir.load %[[VAL_12]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_15:.*]] = arith.muli %[[VAL_13]], %[[VAL_14]] : i32
+! CHECK:                 hlfir.assign %[[VAL_15]] to %[[VAL_12]]#0 : i32, !fir.ref<i32>
+! CHECK:                 omp.yield
+! CHECK:               omp.terminator
 ! CHECK:             omp.terminator
 ! CHECK:           return
 
@@ -194,15 +200,17 @@ end subroutine
 ! CHECK:             %[[VAL_7:.*]] = arith.constant 1 : i32
 ! CHECK:             %[[VAL_8:.*]] = arith.constant 10 : i32
 ! CHECK:             %[[VAL_9:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop byref reduction(@multiply_reduction_byref_f32 %[[VAL_3]]#0 -> %[[VAL_10:.*]] : !fir.ref<f32>)  for  (%[[VAL_11:.*]]) : i32 = (%[[VAL_7]]) to (%[[VAL_8]]) inclusive step (%[[VAL_9]]) {
-! CHECK:               fir.store %[[VAL_11]] to %[[VAL_6]]#1 : !fir.ref<i32>
-! CHECK:               %[[VAL_12:.*]]:2 = hlfir.declare %[[VAL_10]] {uniq_name = "_QFsimple_real_reduction_switch_orderEx"} : (!fir.ref<f32>) -> (!fir.ref<f32>, !fir.ref<f32>)
-! CHECK:               %[[VAL_13:.*]] = fir.load %[[VAL_6]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_14:.*]] = fir.convert %[[VAL_13]] : (i32) -> f32
-! CHECK:               %[[VAL_15:.*]] = fir.load %[[VAL_12]]#0 : !fir.ref<f32>
-! CHECK:               %[[VAL_16:.*]] = arith.mulf %[[VAL_14]], %[[VAL_15]] fastmath<contract> : f32
-! CHECK:               hlfir.assign %[[VAL_16]] to %[[VAL_12]]#0 : f32, !fir.ref<f32>
-! CHECK:               omp.yield
+! CHECK:             omp.wsloop byref reduction(@multiply_reduction_byref_f32 %[[VAL_3]]#0 -> %[[VAL_10:.*]] : !fir.ref<f32>) {
+! CHECK-NEXT:          omp.loop_nest (%[[VAL_11:.*]]) : i32 = (%[[VAL_7]]) to (%[[VAL_8]]) inclusive step (%[[VAL_9]]) {
+! CHECK:                 %[[VAL_12:.*]]:2 = hlfir.declare %[[VAL_10]] {uniq_name = "_QFsimple_real_reduction_switch_orderEx"} : (!fir.ref<f32>) -> (!fir.ref<f32>, !fir.ref<f32>)
+! CHECK:                 fir.store %[[VAL_11]] to %[[VAL_6]]#1 : !fir.ref<i32>
+! CHECK:                 %[[VAL_13:.*]] = fir.load %[[VAL_6]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_14:.*]] = fir.convert %[[VAL_13]] : (i32) -> f32
+! CHECK:                 %[[VAL_15:.*]] = fir.load %[[VAL_12]]#0 : !fir.ref<f32>
+! CHECK:                 %[[VAL_16:.*]] = arith.mulf %[[VAL_14]], %[[VAL_15]] fastmath<contract> : f32
+! CHECK:                 hlfir.assign %[[VAL_16]] to %[[VAL_12]]#0 : f32, !fir.ref<f32>
+! CHECK:                 omp.yield
+! CHECK:               omp.terminator
 ! CHECK:             omp.terminator
 ! CHECK:           return
 
@@ -239,24 +247,26 @@ end subroutine
 ! CHECK:             %[[VAL_13:.*]] = arith.constant 1 : i32
 ! CHECK:             %[[VAL_14:.*]] = arith.constant 10 : i32
 ! CHECK:             %[[VAL_15:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop byref reduction(@multiply_reduction_byref_i32 %[[VAL_3]]#0 -> %[[VAL_16:.*]] : !fir.ref<i32>, @multiply_reduction_byref_i32 %[[VAL_5]]#0 -> %[[VAL_17:.*]] : !fir.ref<i32>, @multiply_reduction_byref_i32 %[[VAL_7]]#0 -> %[[VAL_18:.*]] : !fir.ref<i32>)  for  (%[[VAL_19:.*]]) : i32 = (%[[VAL_13]]) to (%[[VAL_14]]) inclusive step (%[[VAL_15]]) {
-! CHECK:               fir.store %[[VAL_19]] to %[[VAL_12]]#1 : !fir.ref<i32>
-! CHECK:               %[[VAL_20:.*]]:2 = hlfir.declare %[[VAL_16]] {uniq_name = "_QFmultiple_int_reductions_same_typeEx"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
-! CHECK:               %[[VAL_21:.*]]:2 = hlfir.declare %[[VAL_17]] {uniq_name = "_QFmultiple_int_reductions_same_typeEy"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
-! CHECK:               %[[VAL_22:.*]]:2 = hlfir.declare %[[VAL_18]] {uniq_name = "_QFmultiple_int_reductions_same_typeEz"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
-! CHECK:               %[[VAL_23:.*]] = fir.load %[[VAL_20]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_24:.*]] = fir.load %[[VAL_12]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_25:.*]] = arith.muli %[[VAL_23]], %[[VAL_24]] : i32
-! CHECK:               hlfir.assign %[[VAL_25]] to %[[VAL_20]]#0 : i32, !fir.ref<i32>
-! CHECK:               %[[VAL_26:.*]] = fir.load %[[VAL_21]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_27:.*]] = fir.load %[[VAL_12]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_28:.*]] = arith.muli %[[VAL_26]], %[[VAL_27]] : i32
-! CHECK:               hlfir.assign %[[VAL_28]] to %[[VAL_21]]#0 : i32, !fir.ref<i32>
-! CHECK:               %[[VAL_29:.*]] = fir.load %[[VAL_22]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_30:.*]] = fir.load %[[VAL_12]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_31:.*]] = arith.muli %[[VAL_29]], %[[VAL_30]] : i32
-! CHECK:               hlfir.assign %[[VAL_31]] to %[[VAL_22]]#0 : i32, !fir.ref<i32>
-! CHECK:               omp.yield
+! CHECK:             omp.wsloop byref reduction(@multiply_reduction_byref_i32 %[[VAL_3]]#0 -> %[[VAL_16:.*]] : !fir.ref<i32>, @multiply_reduction_byref_i32 %[[VAL_5]]#0 -> %[[VAL_17:.*]] : !fir.ref<i32>, @multiply_reduction_byref_i32 %[[VAL_7]]#0 -> %[[VAL_18:.*]] : !fir.ref<i32>) {
+! CHECK-NEXT:          omp.loop_nest (%[[VAL_19:.*]]) : i32 = (%[[VAL_13]]) to (%[[VAL_14]]) inclusive step (%[[VAL_15]]) {
+! CHECK:                 %[[VAL_20:.*]]:2 = hlfir.declare %[[VAL_16]] {uniq_name = "_QFmultiple_int_reductions_same_typeEx"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
+! CHECK:                 %[[VAL_21:.*]]:2 = hlfir.declare %[[VAL_17]] {uniq_name = "_QFmultiple_int_reductions_same_typeEy"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
+! CHECK:                 %[[VAL_22:.*]]:2 = hlfir.declare %[[VAL_18]] {uniq_name = "_QFmultiple_int_reductions_same_typeEz"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
+! CHECK:                 fir.store %[[VAL_19]] to %[[VAL_12]]#1 : !fir.ref<i32>
+! CHECK:                 %[[VAL_23:.*]] = fir.load %[[VAL_20]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_24:.*]] = fir.load %[[VAL_12]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_25:.*]] = arith.muli %[[VAL_23]], %[[VAL_24]] : i32
+! CHECK:                 hlfir.assign %[[VAL_25]] to %[[VAL_20]]#0 : i32, !fir.ref<i32>
+! CHECK:                 %[[VAL_26:.*]] = fir.load %[[VAL_21]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_27:.*]] = fir.load %[[VAL_12]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_28:.*]] = arith.muli %[[VAL_26]], %[[VAL_27]] : i32
+! CHECK:                 hlfir.assign %[[VAL_28]] to %[[VAL_21]]#0 : i32, !fir.ref<i32>
+! CHECK:                 %[[VAL_29:.*]] = fir.load %[[VAL_22]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_30:.*]] = fir.load %[[VAL_12]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_31:.*]] = arith.muli %[[VAL_29]], %[[VAL_30]] : i32
+! CHECK:                 hlfir.assign %[[VAL_31]] to %[[VAL_22]]#0 : i32, !fir.ref<i32>
+! CHECK:                 omp.yield
+! CHECK:               omp.terminator
 ! CHECK:             omp.terminator
 ! CHECK:           return
 
@@ -297,27 +307,29 @@ end subroutine
 ! CHECK:             %[[VAL_13:.*]] = arith.constant 1 : i32
 ! CHECK:             %[[VAL_14:.*]] = arith.constant 10 : i32
 ! CHECK:             %[[VAL_15:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop byref reduction(@multiply_reduction_byref_f32 %[[VAL_3]]#0 -> %[[VAL_16:.*]] : !fir.ref<f32>, @multiply_reduction_byref_f32 %[[VAL_5]]#0 -> %[[VAL_17:.*]] : !fir.ref<f32>, @multiply_reduction_byref_f32 %[[VAL_7]]#0 -> %[[VAL_18:.*]] : !fir.ref<f32>)  for  (%[[VAL_19:.*]]) : i32 = (%[[VAL_13]]) to (%[[VAL_14]]) inclusive step (%[[VAL_15]]) {
-! CHECK:               fir.store %[[VAL_19]] to %[[VAL_12]]#1 : !fir.ref<i32>
-! CHECK:               %[[VAL_20:.*]]:2 = hlfir.declare %[[VAL_16]] {uniq_name = "_QFmultiple_real_reductions_same_typeEx"} : (!fir.ref<f32>) -> (!fir.ref<f32>, !fir.ref<f32>)
-! CHECK:               %[[VAL_21:.*]]:2 = hlfir.declare %[[VAL_17]] {uniq_name = "_QFmultiple_real_reductions_same_typeEy"} : (!fir.ref<f32>) -> (!fir.ref<f32>, !fir.ref<f32>)
-! CHECK:               %[[VAL_22:.*]]:2 = hlfir.declare %[[VAL_18]] {uniq_name = "_QFmultiple_real_reductions_same_typeEz"} : (!fir.ref<f32>) -> (!fir.ref<f32>, !fir.ref<f32>)
-! CHECK:               %[[VAL_23:.*]] = fir.load %[[VAL_20]]#0 : !fir.ref<f32>
-! CHECK:               %[[VAL_24:.*]] = fir.load %[[VAL_12]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_25:.*]] = fir.convert %[[VAL_24]] : (i32) -> f32
-! CHECK:               %[[VAL_26:.*]] = arith.mulf %[[VAL_23]], %[[VAL_25]] fastmath<contract> : f32
-! CHECK:               hlfir.assign %[[VAL_26]] to %[[VAL_20]]#0 : f32, !fir.ref<f32>
-! CHECK:               %[[VAL_27:.*]] = fir.load %[[VAL_21]]#0 : !fir.ref<f32>
-! CHECK:               %[[VAL_28:.*]] = fir.load %[[VAL_12]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_29:.*]] = fir.convert %[[VAL_28]] : (i32) -> f32
-! CHECK:               %[[VAL_30:.*]] = arith.mulf %[[VAL_27]], %[[VAL_29]] fastmath<contract> : f32
-! CHECK:               hlfir.assign %[[VAL_30]] to %[[VAL_21]]#0 : f32, !fir.ref<f32>
-! CHECK:               %[[VAL_31:.*]] = fir.load %[[VAL_22]]#0 : !fir.ref<f32>
-! CHECK:               %[[VAL_32:.*]] = fir.load %[[VAL_12]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_33:.*]] = fir.convert %[[VAL_32]] : (i32) -> f32
-! CHECK:               %[[VAL_34:.*]] = arith.mulf %[[VAL_31]], %[[VAL_33]] fastmath<contract> : f32
-! CHECK:               hlfir.assign %[[VAL_34]] to %[[VAL_22]]#0 : f32, !fir.ref<f32>
-! CHECK:               omp.yield
+! CHECK:             omp.wsloop byref reduction(@multiply_reduction_byref_f32 %[[VAL_3]]#0 -> %[[VAL_16:.*]] : !fir.ref<f32>, @multiply_reduction_byref_f32 %[[VAL_5]]#0 -> %[[VAL_17:.*]] : !fir.ref<f32>, @multiply_reduction_byref_f32 %[[VAL_7]]#0 -> %[[VAL_18:.*]] : !fir.ref<f32>) {
+! CHECK-NEXT:          omp.loop_nest (%[[VAL_19:.*]]) : i32 = (%[[VAL_13]]) to (%[[VAL_14]]) inclusive step (%[[VAL_15]]) {
+! CHECK:                 %[[VAL_20:.*]]:2 = hlfir.declare %[[VAL_16]] {uniq_name = "_QFmultiple_real_reductions_same_typeEx"} : (!fir.ref<f32>) -> (!fir.ref<f32>, !fir.ref<f32>)
+! CHECK:                 %[[VAL_21:.*]]:2 = hlfir.declare %[[VAL_17]] {uniq_name = "_QFmultiple_real_reductions_same_typeEy"} : (!fir.ref<f32>) -> (!fir.ref<f32>, !fir.ref<f32>)
+! CHECK:                 %[[VAL_22:.*]]:2 = hlfir.declare %[[VAL_18]] {uniq_name = "_QFmultiple_real_reductions_same_typeEz"} : (!fir.ref<f32>) -> (!fir.ref<f32>, !fir.ref<f32>)
+! CHECK:                 fir.store %[[VAL_19]] to %[[VAL_12]]#1 : !fir.ref<i32>
+! CHECK:                 %[[VAL_23:.*]] = fir.load %[[VAL_20]]#0 : !fir.ref<f32>
+! CHECK:                 %[[VAL_24:.*]] = fir.load %[[VAL_12]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_25:.*]] = fir.convert %[[VAL_24]] : (i32) -> f32
+! CHECK:                 %[[VAL_26:.*]] = arith.mulf %[[VAL_23]], %[[VAL_25]] fastmath<contract> : f32
+! CHECK:                 hlfir.assign %[[VAL_26]] to %[[VAL_20]]#0 : f32, !fir.ref<f32>
+! CHECK:                 %[[VAL_27:.*]] = fir.load %[[VAL_21]]#0 : !fir.ref<f32>
+! CHECK:                 %[[VAL_28:.*]] = fir.load %[[VAL_12]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_29:.*]] = fir.convert %[[VAL_28]] : (i32) -> f32
+! CHECK:                 %[[VAL_30:.*]] = arith.mulf %[[VAL_27]], %[[VAL_29]] fastmath<contract> : f32
+! CHECK:                 hlfir.assign %[[VAL_30]] to %[[VAL_21]]#0 : f32, !fir.ref<f32>
+! CHECK:                 %[[VAL_31:.*]] = fir.load %[[VAL_22]]#0 : !fir.ref<f32>
+! CHECK:                 %[[VAL_32:.*]] = fir.load %[[VAL_12]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_33:.*]] = fir.convert %[[VAL_32]] : (i32) -> f32
+! CHECK:                 %[[VAL_34:.*]] = arith.mulf %[[VAL_31]], %[[VAL_33]] fastmath<contract> : f32
+! CHECK:                 hlfir.assign %[[VAL_34]] to %[[VAL_22]]#0 : f32, !fir.ref<f32>
+! CHECK:                 omp.yield
+! CHECK:               omp.terminator
 ! CHECK:             omp.terminator
 ! CHECK:           return
 
@@ -362,32 +374,34 @@ end subroutine
 ! CHECK:             %[[VAL_16:.*]] = arith.constant 1 : i32
 ! CHECK:             %[[VAL_17:.*]] = arith.constant 10 : i32
 ! CHECK:             %[[VAL_18:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop byref reduction(@multiply_reduction_byref_i32 %[[VAL_5]]#0 -> %[[VAL_19:.*]] : !fir.ref<i32>, @multiply_reduction_byref_i64 %[[VAL_7]]#0 -> %[[VAL_20:.*]] : !fir.ref<i64>, @multiply_reduction_byref_f32 %[[VAL_9]]#0 -> %[[VAL_21:.*]] : !fir.ref<f32>, @multiply_reduction_byref_f64 %[[VAL_3]]#0 -> %[[VAL_22:.*]] : !fir.ref<f64>)  for  (%[[VAL_23:.*]]) : i32 = (%[[VAL_16]]) to (%[[VAL_17]]) inclusive step (%[[VAL_18]]) {
-! CHECK:               fir.store %[[VAL_23]] to %[[VAL_15]]#1 : !fir.ref<i32>
-! CHECK:               %[[VAL_24:.*]]:2 = hlfir.declare %[[VAL_19]] {uniq_name = "_QFmultiple_reductions_different_typeEx"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
-! CHECK:               %[[VAL_25:.*]]:2 = hlfir.declare %[[VAL_20]] {uniq_name = "_QFmultiple_reductions_different_typeEy"} : (!fir.ref<i64>) -> (!fir.ref<i64>, !fir.ref<i64>)
-! CHECK:               %[[VAL_26:.*]]:2 = hlfir.declare %[[VAL_21]] {uniq_name = "_QFmultiple_reductions_different_typeEz"} : (!fir.ref<f32>) -> (!fir.ref<f32>, !fir.ref<f32>)
-! CHECK:               %[[VAL_27:.*]]:2 = hlfir.declare %[[VAL_22]] {uniq_name = "_QFmultiple_reductions_different_typeEw"} : (!fir.ref<f64>) -> (!fir.ref<f64>, !fir.ref<f64>)
-! CHECK:               %[[VAL_28:.*]] = fir.load %[[VAL_24]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_29:.*]] = fir.load %[[VAL_15]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_30:.*]] = arith.muli %[[VAL_28]], %[[VAL_29]] : i32
-! CHECK:               hlfir.assign %[[VAL_30]] to %[[VAL_24]]#0 : i32, !fir.ref<i32>
-! CHECK:               %[[VAL_31:.*]] = fir.load %[[VAL_25]]#0 : !fir.ref<i64>
-! CHECK:               %[[VAL_32:.*]] = fir.load %[[VAL_15]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_33:.*]] = fir.convert %[[VAL_32]] : (i32) -> i64
-! CHECK:               %[[VAL_34:.*]] = arith.muli %[[VAL_31]], %[[VAL_33]] : i64
-! CHECK:               hlfir.assign %[[VAL_34]] to %[[VAL_25]]#0 : i64, !fir.ref<i64>
-! CHECK:               %[[VAL_35:.*]] = fir.load %[[VAL_26]]#0 : !fir.ref<f32>
-! CHECK:               %[[VAL_36:.*]] = fir.load %[[VAL_15]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_37:.*]] = fir.convert %[[VAL_36]] : (i32) -> f32
-! CHECK:               %[[VAL_38:.*]] = arith.mulf %[[VAL_35]], %[[VAL_37]] fastmath<contract> : f32
-! CHECK:               hlfir.assign %[[VAL_38]] to %[[VAL_26]]#0 : f32, !fir.ref<f32>
-! CHECK:               %[[VAL_39:.*]] = fir.load %[[VAL_27]]#0 : !fir.ref<f64>
-! CHECK:               %[[VAL_40:.*]] = fir.load %[[VAL_15]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_41:.*]] = fir.convert %[[VAL_40]] : (i32) -> f64
-! CHECK:               %[[VAL_42:.*]] = arith.mulf %[[VAL_39]], %[[VAL_41]] fastmath<contract> : f64
-! CHECK:               hlfir.assign %[[VAL_42]] to %[[VAL_27]]#0 : f64, !fir.ref<f64>
-! CHECK:               omp.yield
+! CHECK:             omp.wsloop byref reduction(@multiply_reduction_byref_i32 %[[VAL_5]]#0 -> %[[VAL_19:.*]] : !fir.ref<i32>, @multiply_reduction_byref_i64 %[[VAL_7]]#0 -> %[[VAL_20:.*]] : !fir.ref<i64>, @multiply_reduction_byref_f32 %[[VAL_9]]#0 -> %[[VAL_21:.*]] : !fir.ref<f32>, @multiply_reduction_byref_f64 %[[VAL_3]]#0 -> %[[VAL_22:.*]] : !fir.ref<f64>) {
+! CHECK-NEXT:          omp.loop_nest (%[[VAL_23:.*]]) : i32 = (%[[VAL_16]]) to (%[[VAL_17]]) inclusive step (%[[VAL_18]]) {
+! CHECK:                 %[[VAL_24:.*]]:2 = hlfir.declare %[[VAL_19]] {uniq_name = "_QFmultiple_reductions_different_typeEx"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
+! CHECK:                 %[[VAL_25:.*]]:2 = hlfir.declare %[[VAL_20]] {uniq_name = "_QFmultiple_reductions_different_typeEy"} : (!fir.ref<i64>) -> (!fir.ref<i64>, !fir.ref<i64>)
+! CHECK:                 %[[VAL_26:.*]]:2 = hlfir.declare %[[VAL_21]] {uniq_name = "_QFmultiple_reductions_different_typeEz"} : (!fir.ref<f32>) -> (!fir.ref<f32>, !fir.ref<f32>)
+! CHECK:                 %[[VAL_27:.*]]:2 = hlfir.declare %[[VAL_22]] {uniq_name = "_QFmultiple_reductions_different_typeEw"} : (!fir.ref<f64>) -> (!fir.ref<f64>, !fir.ref<f64>)
+! CHECK:                 fir.store %[[VAL_23]] to %[[VAL_15]]#1 : !fir.ref<i32>
+! CHECK:                 %[[VAL_28:.*]] = fir.load %[[VAL_24]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_29:.*]] = fir.load %[[VAL_15]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_30:.*]] = arith.muli %[[VAL_28]], %[[VAL_29]] : i32
+! CHECK:                 hlfir.assign %[[VAL_30]] to %[[VAL_24]]#0 : i32, !fir.ref<i32>
+! CHECK:                 %[[VAL_31:.*]] = fir.load %[[VAL_25]]#0 : !fir.ref<i64>
+! CHECK:                 %[[VAL_32:.*]] = fir.load %[[VAL_15]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_33:.*]] = fir.convert %[[VAL_32]] : (i32) -> i64
+! CHECK:                 %[[VAL_34:.*]] = arith.muli %[[VAL_31]], %[[VAL_33]] : i64
+! CHECK:                 hlfir.assign %[[VAL_34]] to %[[VAL_25]]#0 : i64, !fir.ref<i64>
+! CHECK:                 %[[VAL_35:.*]] = fir.load %[[VAL_26]]#0 : !fir.ref<f32>
+! CHECK:                 %[[VAL_36:.*]] = fir.load %[[VAL_15]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_37:.*]] = fir.convert %[[VAL_36]] : (i32) -> f32
+! CHECK:                 %[[VAL_38:.*]] = arith.mulf %[[VAL_35]], %[[VAL_37]] fastmath<contract> : f32
+! CHECK:                 hlfir.assign %[[VAL_38]] to %[[VAL_26]]#0 : f32, !fir.ref<f32>
+! CHECK:                 %[[VAL_39:.*]] = fir.load %[[VAL_27]]#0 : !fir.ref<f64>
+! CHECK:                 %[[VAL_40:.*]] = fir.load %[[VAL_15]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_41:.*]] = fir.convert %[[VAL_40]] : (i32) -> f64
+! CHECK:                 %[[VAL_42:.*]] = arith.mulf %[[VAL_39]], %[[VAL_41]] fastmath<contract> : f64
+! CHECK:                 hlfir.assign %[[VAL_42]] to %[[VAL_27]]#0 : f64, !fir.ref<f64>
+! CHECK:                 omp.yield
+! CHECK:               omp.terminator
 ! CHECK:             omp.terminator
 ! CHECK:           return
 
diff --git a/flang/test/Lower/OpenMP/wsloop-reduction-mul.f90 b/flang/test/Lower/OpenMP/wsloop-reduction-mul.f90
index 90d9aa5e839b..08be4d84c1a6 100644
--- a/flang/test/Lower/OpenMP/wsloop-reduction-mul.f90
+++ b/flang/test/Lower/OpenMP/wsloop-reduction-mul.f90
@@ -60,14 +60,16 @@
 ! CHECK:             %[[VAL_7:.*]] = arith.constant 1 : i32
 ! CHECK:             %[[VAL_8:.*]] = arith.constant 10 : i32
 ! CHECK:             %[[VAL_9:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop reduction(@multiply_reduction_i32 %[[VAL_3]]#0 -> %[[VAL_10:.*]] : !fir.ref<i32>)  for  (%[[VAL_11:.*]]) : i32 = (%[[VAL_7]]) to (%[[VAL_8]]) inclusive step (%[[VAL_9]]) {
-! CHECK:               fir.store %[[VAL_11]] to %[[VAL_6]]#1 : !fir.ref<i32>
-! CHECK:               %[[VAL_12:.*]]:2 = hlfir.declare %[[VAL_10]] {uniq_name = "_QFsimple_int_reductionEx"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
-! CHECK:               %[[VAL_13:.*]] = fir.load %[[VAL_12]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_14:.*]] = fir.load %[[VAL_6]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_15:.*]] = arith.muli %[[VAL_13]], %[[VAL_14]] : i32
-! CHECK:               hlfir.assign %[[VAL_15]] to %[[VAL_12]]#0 : i32, !fir.ref<i32>
-! CHECK:               omp.yield
+! CHECK:             omp.wsloop reduction(@multiply_reduction_i32 %[[VAL_3]]#0 -> %[[VAL_10:.*]] : !fir.ref<i32>) {
+! CHECK-NEXT:          omp.loop_nest (%[[VAL_11:.*]]) : i32 = (%[[VAL_7]]) to (%[[VAL_8]]) inclusive step (%[[VAL_9]]) {
+! CHECK:                 %[[VAL_12:.*]]:2 = hlfir.declare %[[VAL_10]] {uniq_name = "_QFsimple_int_reductionEx"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
+! CHECK:                 fir.store %[[VAL_11]] to %[[VAL_6]]#1 : !fir.ref<i32>
+! CHECK:                 %[[VAL_13:.*]] = fir.load %[[VAL_12]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_14:.*]] = fir.load %[[VAL_6]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_15:.*]] = arith.muli %[[VAL_13]], %[[VAL_14]] : i32
+! CHECK:                 hlfir.assign %[[VAL_15]] to %[[VAL_12]]#0 : i32, !fir.ref<i32>
+! CHECK:                 omp.yield
+! CHECK:               omp.terminator
 ! CHECK:             omp.terminator
 ! CHECK:           return
 
@@ -96,15 +98,17 @@ end subroutine
 ! CHECK:             %[[VAL_7:.*]] = arith.constant 1 : i32
 ! CHECK:             %[[VAL_8:.*]] = arith.constant 10 : i32
 ! CHECK:             %[[VAL_9:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop reduction(@multiply_reduction_f32 %[[VAL_3]]#0 -> %[[VAL_10:.*]] : !fir.ref<f32>)  for  (%[[VAL_11:.*]]) : i32 = (%[[VAL_7]]) to (%[[VAL_8]]) inclusive step (%[[VAL_9]]) {
-! CHECK:               fir.store %[[VAL_11]] to %[[VAL_6]]#1 : !fir.ref<i32>
-! CHECK:               %[[VAL_12:.*]]:2 = hlfir.declare %[[VAL_10]] {uniq_name = "_QFsimple_real_reductionEx"} : (!fir.ref<f32>) -> (!fir.ref<f32>, !fir.ref<f32>)
-! CHECK:               %[[VAL_13:.*]] = fir.load %[[VAL_12]]#0 : !fir.ref<f32>
-! CHECK:               %[[VAL_14:.*]] = fir.load %[[VAL_6]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_15:.*]] = fir.convert %[[VAL_14]] : (i32) -> f32
-! CHECK:               %[[VAL_16:.*]] = arith.mulf %[[VAL_13]], %[[VAL_15]] fastmath<contract> : f32
-! CHECK:               hlfir.assign %[[VAL_16]] to %[[VAL_12]]#0 : f32, !fir.ref<f32>
-! CHECK:               omp.yield
+! CHECK:             omp.wsloop reduction(@multiply_reduction_f32 %[[VAL_3]]#0 -> %[[VAL_10:.*]] : !fir.ref<f32>) {
+! CHECK-NEXT:          omp.loop_nest (%[[VAL_11:.*]]) : i32 = (%[[VAL_7]]) to (%[[VAL_8]]) inclusive step (%[[VAL_9]]) {
+! CHECK:                 %[[VAL_12:.*]]:2 = hlfir.declare %[[VAL_10]] {uniq_name = "_QFsimple_real_reductionEx"} : (!fir.ref<f32>) -> (!fir.ref<f32>, !fir.ref<f32>)
+! CHECK:                 fir.store %[[VAL_11]] to %[[VAL_6]]#1 : !fir.ref<i32>
+! CHECK:                 %[[VAL_13:.*]] = fir.load %[[VAL_12]]#0 : !fir.ref<f32>
+! CHECK:                 %[[VAL_14:.*]] = fir.load %[[VAL_6]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_15:.*]] = fir.convert %[[VAL_14]] : (i32) -> f32
+! CHECK:                 %[[VAL_16:.*]] = arith.mulf %[[VAL_13]], %[[VAL_15]] fastmath<contract> : f32
+! CHECK:                 hlfir.assign %[[VAL_16]] to %[[VAL_12]]#0 : f32, !fir.ref<f32>
+! CHECK:                 omp.yield
+! CHECK:               omp.terminator
 ! CHECK:             omp.terminator
 ! CHECK:           return
 
@@ -133,14 +137,16 @@ end subroutine
 ! CHECK:             %[[VAL_7:.*]] = arith.constant 1 : i32
 ! CHECK:             %[[VAL_8:.*]] = arith.constant 10 : i32
 ! CHECK:             %[[VAL_9:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop reduction(@multiply_reduction_i32 %[[VAL_3]]#0 -> %[[VAL_10:.*]] : !fir.ref<i32>)  for  (%[[VAL_11:.*]]) : i32 = (%[[VAL_7]]) to (%[[VAL_8]]) inclusive step (%[[VAL_9]]) {
-! CHECK:               fir.store %[[VAL_11]] to %[[VAL_6]]#1 : !fir.ref<i32>
-! CHECK:               %[[VAL_12:.*]]:2 = hlfir.declare %[[VAL_10]] {uniq_name = "_QFsimple_int_reduction_switch_orderEx"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
-! CHECK:               %[[VAL_13:.*]] = fir.load %[[VAL_6]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_14:.*]] = fir.load %[[VAL_12]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_15:.*]] = arith.muli %[[VAL_13]], %[[VAL_14]] : i32
-! CHECK:               hlfir.assign %[[VAL_15]] to %[[VAL_12]]#0 : i32, !fir.ref<i32>
-! CHECK:               omp.yield
+! CHECK:             omp.wsloop reduction(@multiply_reduction_i32 %[[VAL_3]]#0 -> %[[VAL_10:.*]] : !fir.ref<i32>) {
+! CHECK-NEXT:          omp.loop_nest (%[[VAL_11:.*]]) : i32 = (%[[VAL_7]]) to (%[[VAL_8]]) inclusive step (%[[VAL_9]]) {
+! CHECK:                 %[[VAL_12:.*]]:2 = hlfir.declare %[[VAL_10]] {uniq_name = "_QFsimple_int_reduction_switch_orderEx"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
+! CHECK:                 fir.store %[[VAL_11]] to %[[VAL_6]]#1 : !fir.ref<i32>
+! CHECK:                 %[[VAL_13:.*]] = fir.load %[[VAL_6]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_14:.*]] = fir.load %[[VAL_12]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_15:.*]] = arith.muli %[[VAL_13]], %[[VAL_14]] : i32
+! CHECK:                 hlfir.assign %[[VAL_15]] to %[[VAL_12]]#0 : i32, !fir.ref<i32>
+! CHECK:                 omp.yield
+! CHECK:               omp.terminator
 ! CHECK:             omp.terminator
 ! CHECK:           return
 
@@ -169,15 +175,17 @@ end subroutine
 ! CHECK:             %[[VAL_7:.*]] = arith.constant 1 : i32
 ! CHECK:             %[[VAL_8:.*]] = arith.constant 10 : i32
 ! CHECK:             %[[VAL_9:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop reduction(@multiply_reduction_f32 %[[VAL_3]]#0 -> %[[VAL_10:.*]] : !fir.ref<f32>)  for  (%[[VAL_11:.*]]) : i32 = (%[[VAL_7]]) to (%[[VAL_8]]) inclusive step (%[[VAL_9]]) {
-! CHECK:               fir.store %[[VAL_11]] to %[[VAL_6]]#1 : !fir.ref<i32>
-! CHECK:               %[[VAL_12:.*]]:2 = hlfir.declare %[[VAL_10]] {uniq_name = "_QFsimple_real_reduction_switch_orderEx"} : (!fir.ref<f32>) -> (!fir.ref<f32>, !fir.ref<f32>)
-! CHECK:               %[[VAL_13:.*]] = fir.load %[[VAL_6]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_14:.*]] = fir.convert %[[VAL_13]] : (i32) -> f32
-! CHECK:               %[[VAL_15:.*]] = fir.load %[[VAL_12]]#0 : !fir.ref<f32>
-! CHECK:               %[[VAL_16:.*]] = arith.mulf %[[VAL_14]], %[[VAL_15]] fastmath<contract> : f32
-! CHECK:               hlfir.assign %[[VAL_16]] to %[[VAL_12]]#0 : f32, !fir.ref<f32>
-! CHECK:               omp.yield
+! CHECK:             omp.wsloop reduction(@multiply_reduction_f32 %[[VAL_3]]#0 -> %[[VAL_10:.*]] : !fir.ref<f32>) {
+! CHECK-NEXT:          omp.loop_nest (%[[VAL_11:.*]]) : i32 = (%[[VAL_7]]) to (%[[VAL_8]]) inclusive step (%[[VAL_9]]) {
+! CHECK:                 %[[VAL_12:.*]]:2 = hlfir.declare %[[VAL_10]] {uniq_name = "_QFsimple_real_reduction_switch_orderEx"} : (!fir.ref<f32>) -> (!fir.ref<f32>, !fir.ref<f32>)
+! CHECK:                 fir.store %[[VAL_11]] to %[[VAL_6]]#1 : !fir.ref<i32>
+! CHECK:                 %[[VAL_13:.*]] = fir.load %[[VAL_6]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_14:.*]] = fir.convert %[[VAL_13]] : (i32) -> f32
+! CHECK:                 %[[VAL_15:.*]] = fir.load %[[VAL_12]]#0 : !fir.ref<f32>
+! CHECK:                 %[[VAL_16:.*]] = arith.mulf %[[VAL_14]], %[[VAL_15]] fastmath<contract> : f32
+! CHECK:                 hlfir.assign %[[VAL_16]] to %[[VAL_12]]#0 : f32, !fir.ref<f32>
+! CHECK:                 omp.yield
+! CHECK:               omp.terminator
 ! CHECK:             omp.terminator
 ! CHECK:           return
 
@@ -214,24 +222,26 @@ end subroutine
 ! CHECK:             %[[VAL_13:.*]] = arith.constant 1 : i32
 ! CHECK:             %[[VAL_14:.*]] = arith.constant 10 : i32
 ! CHECK:             %[[VAL_15:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop reduction(@multiply_reduction_i32 %[[VAL_3]]#0 -> %[[VAL_16:.*]] : !fir.ref<i32>, @multiply_reduction_i32 %[[VAL_5]]#0 -> %[[VAL_17:.*]] : !fir.ref<i32>, @multiply_reduction_i32 %[[VAL_7]]#0 -> %[[VAL_18:.*]] : !fir.ref<i32>)  for  (%[[VAL_19:.*]]) : i32 = (%[[VAL_13]]) to (%[[VAL_14]]) inclusive step (%[[VAL_15]]) {
-! CHECK:               fir.store %[[VAL_19]] to %[[VAL_12]]#1 : !fir.ref<i32>
-! CHECK:               %[[VAL_20:.*]]:2 = hlfir.declare %[[VAL_16]] {uniq_name = "_QFmultiple_int_reductions_same_typeEx"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
-! CHECK:               %[[VAL_21:.*]]:2 = hlfir.declare %[[VAL_17]] {uniq_name = "_QFmultiple_int_reductions_same_typeEy"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
-! CHECK:               %[[VAL_22:.*]]:2 = hlfir.declare %[[VAL_18]] {uniq_name = "_QFmultiple_int_reductions_same_typeEz"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
-! CHECK:               %[[VAL_23:.*]] = fir.load %[[VAL_20]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_24:.*]] = fir.load %[[VAL_12]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_25:.*]] = arith.muli %[[VAL_23]], %[[VAL_24]] : i32
-! CHECK:               hlfir.assign %[[VAL_25]] to %[[VAL_20]]#0 : i32, !fir.ref<i32>
-! CHECK:               %[[VAL_26:.*]] = fir.load %[[VAL_21]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_27:.*]] = fir.load %[[VAL_12]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_28:.*]] = arith.muli %[[VAL_26]], %[[VAL_27]] : i32
-! CHECK:               hlfir.assign %[[VAL_28]] to %[[VAL_21]]#0 : i32, !fir.ref<i32>
-! CHECK:               %[[VAL_29:.*]] = fir.load %[[VAL_22]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_30:.*]] = fir.load %[[VAL_12]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_31:.*]] = arith.muli %[[VAL_29]], %[[VAL_30]] : i32
-! CHECK:               hlfir.assign %[[VAL_31]] to %[[VAL_22]]#0 : i32, !fir.ref<i32>
-! CHECK:               omp.yield
+! CHECK:             omp.wsloop reduction(@multiply_reduction_i32 %[[VAL_3]]#0 -> %[[VAL_16:.*]] : !fir.ref<i32>, @multiply_reduction_i32 %[[VAL_5]]#0 -> %[[VAL_17:.*]] : !fir.ref<i32>, @multiply_reduction_i32 %[[VAL_7]]#0 -> %[[VAL_18:.*]] : !fir.ref<i32>) {
+! CHECK-NEXT:          omp.loop_nest (%[[VAL_19:.*]]) : i32 = (%[[VAL_13]]) to (%[[VAL_14]]) inclusive step (%[[VAL_15]]) {
+! CHECK:                 %[[VAL_20:.*]]:2 = hlfir.declare %[[VAL_16]] {uniq_name = "_QFmultiple_int_reductions_same_typeEx"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
+! CHECK:                 %[[VAL_21:.*]]:2 = hlfir.declare %[[VAL_17]] {uniq_name = "_QFmultiple_int_reductions_same_typeEy"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
+! CHECK:                 %[[VAL_22:.*]]:2 = hlfir.declare %[[VAL_18]] {uniq_name = "_QFmultiple_int_reductions_same_typeEz"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
+! CHECK:                 fir.store %[[VAL_19]] to %[[VAL_12]]#1 : !fir.ref<i32>
+! CHECK:                 %[[VAL_23:.*]] = fir.load %[[VAL_20]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_24:.*]] = fir.load %[[VAL_12]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_25:.*]] = arith.muli %[[VAL_23]], %[[VAL_24]] : i32
+! CHECK:                 hlfir.assign %[[VAL_25]] to %[[VAL_20]]#0 : i32, !fir.ref<i32>
+! CHECK:                 %[[VAL_26:.*]] = fir.load %[[VAL_21]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_27:.*]] = fir.load %[[VAL_12]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_28:.*]] = arith.muli %[[VAL_26]], %[[VAL_27]] : i32
+! CHECK:                 hlfir.assign %[[VAL_28]] to %[[VAL_21]]#0 : i32, !fir.ref<i32>
+! CHECK:                 %[[VAL_29:.*]] = fir.load %[[VAL_22]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_30:.*]] = fir.load %[[VAL_12]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_31:.*]] = arith.muli %[[VAL_29]], %[[VAL_30]] : i32
+! CHECK:                 hlfir.assign %[[VAL_31]] to %[[VAL_22]]#0 : i32, !fir.ref<i32>
+! CHECK:                 omp.yield
+! CHECK:               omp.terminator
 ! CHECK:             omp.terminator
 ! CHECK:           return
 
@@ -272,27 +282,29 @@ end subroutine
 ! CHECK:             %[[VAL_13:.*]] = arith.constant 1 : i32
 ! CHECK:             %[[VAL_14:.*]] = arith.constant 10 : i32
 ! CHECK:             %[[VAL_15:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop reduction(@multiply_reduction_f32 %[[VAL_3]]#0 -> %[[VAL_16:.*]] : !fir.ref<f32>, @multiply_reduction_f32 %[[VAL_5]]#0 -> %[[VAL_17:.*]] : !fir.ref<f32>, @multiply_reduction_f32 %[[VAL_7]]#0 -> %[[VAL_18:.*]] : !fir.ref<f32>)  for  (%[[VAL_19:.*]]) : i32 = (%[[VAL_13]]) to (%[[VAL_14]]) inclusive step (%[[VAL_15]]) {
-! CHECK:               fir.store %[[VAL_19]] to %[[VAL_12]]#1 : !fir.ref<i32>
-! CHECK:               %[[VAL_20:.*]]:2 = hlfir.declare %[[VAL_16]] {uniq_name = "_QFmultiple_real_reductions_same_typeEx"} : (!fir.ref<f32>) -> (!fir.ref<f32>, !fir.ref<f32>)
-! CHECK:               %[[VAL_21:.*]]:2 = hlfir.declare %[[VAL_17]] {uniq_name = "_QFmultiple_real_reductions_same_typeEy"} : (!fir.ref<f32>) -> (!fir.ref<f32>, !fir.ref<f32>)
-! CHECK:               %[[VAL_22:.*]]:2 = hlfir.declare %[[VAL_18]] {uniq_name = "_QFmultiple_real_reductions_same_typeEz"} : (!fir.ref<f32>) -> (!fir.ref<f32>, !fir.ref<f32>)
-! CHECK:               %[[VAL_23:.*]] = fir.load %[[VAL_20]]#0 : !fir.ref<f32>
-! CHECK:               %[[VAL_24:.*]] = fir.load %[[VAL_12]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_25:.*]] = fir.convert %[[VAL_24]] : (i32) -> f32
-! CHECK:               %[[VAL_26:.*]] = arith.mulf %[[VAL_23]], %[[VAL_25]] fastmath<contract> : f32
-! CHECK:               hlfir.assign %[[VAL_26]] to %[[VAL_20]]#0 : f32, !fir.ref<f32>
-! CHECK:               %[[VAL_27:.*]] = fir.load %[[VAL_21]]#0 : !fir.ref<f32>
-! CHECK:               %[[VAL_28:.*]] = fir.load %[[VAL_12]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_29:.*]] = fir.convert %[[VAL_28]] : (i32) -> f32
-! CHECK:               %[[VAL_30:.*]] = arith.mulf %[[VAL_27]], %[[VAL_29]] fastmath<contract> : f32
-! CHECK:               hlfir.assign %[[VAL_30]] to %[[VAL_21]]#0 : f32, !fir.ref<f32>
-! CHECK:               %[[VAL_31:.*]] = fir.load %[[VAL_22]]#0 : !fir.ref<f32>
-! CHECK:               %[[VAL_32:.*]] = fir.load %[[VAL_12]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_33:.*]] = fir.convert %[[VAL_32]] : (i32) -> f32
-! CHECK:               %[[VAL_34:.*]] = arith.mulf %[[VAL_31]], %[[VAL_33]] fastmath<contract> : f32
-! CHECK:               hlfir.assign %[[VAL_34]] to %[[VAL_22]]#0 : f32, !fir.ref<f32>
-! CHECK:               omp.yield
+! CHECK:             omp.wsloop reduction(@multiply_reduction_f32 %[[VAL_3]]#0 -> %[[VAL_16:.*]] : !fir.ref<f32>, @multiply_reduction_f32 %[[VAL_5]]#0 -> %[[VAL_17:.*]] : !fir.ref<f32>, @multiply_reduction_f32 %[[VAL_7]]#0 -> %[[VAL_18:.*]] : !fir.ref<f32>) {
+! CHECK-NEXT:          omp.loop_nest (%[[VAL_19:.*]]) : i32 = (%[[VAL_13]]) to (%[[VAL_14]]) inclusive step (%[[VAL_15]]) {
+! CHECK:                 %[[VAL_20:.*]]:2 = hlfir.declare %[[VAL_16]] {uniq_name = "_QFmultiple_real_reductions_same_typeEx"} : (!fir.ref<f32>) -> (!fir.ref<f32>, !fir.ref<f32>)
+! CHECK:                 %[[VAL_21:.*]]:2 = hlfir.declare %[[VAL_17]] {uniq_name = "_QFmultiple_real_reductions_same_typeEy"} : (!fir.ref<f32>) -> (!fir.ref<f32>, !fir.ref<f32>)
+! CHECK:                 %[[VAL_22:.*]]:2 = hlfir.declare %[[VAL_18]] {uniq_name = "_QFmultiple_real_reductions_same_typeEz"} : (!fir.ref<f32>) -> (!fir.ref<f32>, !fir.ref<f32>)
+! CHECK:                 fir.store %[[VAL_19]] to %[[VAL_12]]#1 : !fir.ref<i32>
+! CHECK:                 %[[VAL_23:.*]] = fir.load %[[VAL_20]]#0 : !fir.ref<f32>
+! CHECK:                 %[[VAL_24:.*]] = fir.load %[[VAL_12]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_25:.*]] = fir.convert %[[VAL_24]] : (i32) -> f32
+! CHECK:                 %[[VAL_26:.*]] = arith.mulf %[[VAL_23]], %[[VAL_25]] fastmath<contract> : f32
+! CHECK:                 hlfir.assign %[[VAL_26]] to %[[VAL_20]]#0 : f32, !fir.ref<f32>
+! CHECK:                 %[[VAL_27:.*]] = fir.load %[[VAL_21]]#0 : !fir.ref<f32>
+! CHECK:                 %[[VAL_28:.*]] = fir.load %[[VAL_12]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_29:.*]] = fir.convert %[[VAL_28]] : (i32) -> f32
+! CHECK:                 %[[VAL_30:.*]] = arith.mulf %[[VAL_27]], %[[VAL_29]] fastmath<contract> : f32
+! CHECK:                 hlfir.assign %[[VAL_30]] to %[[VAL_21]]#0 : f32, !fir.ref<f32>
+! CHECK:                 %[[VAL_31:.*]] = fir.load %[[VAL_22]]#0 : !fir.ref<f32>
+! CHECK:                 %[[VAL_32:.*]] = fir.load %[[VAL_12]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_33:.*]] = fir.convert %[[VAL_32]] : (i32) -> f32
+! CHECK:                 %[[VAL_34:.*]] = arith.mulf %[[VAL_31]], %[[VAL_33]] fastmath<contract> : f32
+! CHECK:                 hlfir.assign %[[VAL_34]] to %[[VAL_22]]#0 : f32, !fir.ref<f32>
+! CHECK:                 omp.yield
+! CHECK:               omp.terminator
 ! CHECK:             omp.terminator
 ! CHECK:           return
 
@@ -337,32 +349,34 @@ end subroutine
 ! CHECK:             %[[VAL_16:.*]] = arith.constant 1 : i32
 ! CHECK:             %[[VAL_17:.*]] = arith.constant 10 : i32
 ! CHECK:             %[[VAL_18:.*]] = arith.constant 1 : i32
-! CHECK:             omp.wsloop reduction(@multiply_reduction_i32 %[[VAL_5]]#0 -> %[[VAL_19:.*]] : !fir.ref<i32>, @multiply_reduction_i64 %[[VAL_7]]#0 -> %[[VAL_20:.*]] : !fir.ref<i64>, @multiply_reduction_f32 %[[VAL_9]]#0 -> %[[VAL_21:.*]] : !fir.ref<f32>, @multiply_reduction_f64 %[[VAL_3]]#0 -> %[[VAL_22:.*]] : !fir.ref<f64>)  for  (%[[VAL_23:.*]]) : i32 = (%[[VAL_16]]) to (%[[VAL_17]]) inclusive step (%[[VAL_18]]) {
-! CHECK:               fir.store %[[VAL_23]] to %[[VAL_15]]#1 : !fir.ref<i32>
-! CHECK:               %[[VAL_24:.*]]:2 = hlfir.declare %[[VAL_19]] {uniq_name = "_QFmultiple_reductions_different_typeEx"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
-! CHECK:               %[[VAL_25:.*]]:2 = hlfir.declare %[[VAL_20]] {uniq_name = "_QFmultiple_reductions_different_typeEy"} : (!fir.ref<i64>) -> (!fir.ref<i64>, !fir.ref<i64>)
-! CHECK:               %[[VAL_26:.*]]:2 = hlfir.declare %[[VAL_21]] {uniq_name = "_QFmultiple_reductions_different_typeEz"} : (!fir.ref<f32>) -> (!fir.ref<f32>, !fir.ref<f32>)
-! CHECK:               %[[VAL_27:.*]]:2 = hlfir.declare %[[VAL_22]] {uniq_name = "_QFmultiple_reductions_different_typeEw"} : (!fir.ref<f64>) -> (!fir.ref<f64>, !fir.ref<f64>)
-! CHECK:               %[[VAL_28:.*]] = fir.load %[[VAL_24]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_29:.*]] = fir.load %[[VAL_15]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_30:.*]] = arith.muli %[[VAL_28]], %[[VAL_29]] : i32
-! CHECK:               hlfir.assign %[[VAL_30]] to %[[VAL_24]]#0 : i32, !fir.ref<i32>
-! CHECK:               %[[VAL_31:.*]] = fir.load %[[VAL_25]]#0 : !fir.ref<i64>
-! CHECK:               %[[VAL_32:.*]] = fir.load %[[VAL_15]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_33:.*]] = fir.convert %[[VAL_32]] : (i32) -> i64
-! CHECK:               %[[VAL_34:.*]] = arith.muli %[[VAL_31]], %[[VAL_33]] : i64
-! CHECK:               hlfir.assign %[[VAL_34]] to %[[VAL_25]]#0 : i64, !fir.ref<i64>
-! CHECK:               %[[VAL_35:.*]] = fir.load %[[VAL_26]]#0 : !fir.ref<f32>
-! CHECK:               %[[VAL_36:.*]] = fir.load %[[VAL_15]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_37:.*]] = fir.convert %[[VAL_36]] : (i32) -> f32
-! CHECK:               %[[VAL_38:.*]] = arith.mulf %[[VAL_35]], %[[VAL_37]] fastmath<contract> : f32
-! CHECK:               hlfir.assign %[[VAL_38]] to %[[VAL_26]]#0 : f32, !fir.ref<f32>
-! CHECK:               %[[VAL_39:.*]] = fir.load %[[VAL_27]]#0 : !fir.ref<f64>
-! CHECK:               %[[VAL_40:.*]] = fir.load %[[VAL_15]]#0 : !fir.ref<i32>
-! CHECK:               %[[VAL_41:.*]] = fir.convert %[[VAL_40]] : (i32) -> f64
-! CHECK:               %[[VAL_42:.*]] = arith.mulf %[[VAL_39]], %[[VAL_41]] fastmath<contract> : f64
-! CHECK:               hlfir.assign %[[VAL_42]] to %[[VAL_27]]#0 : f64, !fir.ref<f64>
-! CHECK:               omp.yield
+! CHECK:             omp.wsloop reduction(@multiply_reduction_i32 %[[VAL_5]]#0 -> %[[VAL_19:.*]] : !fir.ref<i32>, @multiply_reduction_i64 %[[VAL_7]]#0 -> %[[VAL_20:.*]] : !fir.ref<i64>, @multiply_reduction_f32 %[[VAL_9]]#0 -> %[[VAL_21:.*]] : !fir.ref<f32>, @multiply_reduction_f64 %[[VAL_3]]#0 -> %[[VAL_22:.*]] : !fir.ref<f64>) {
+! CHECK-NEXT:          omp.loop_nest (%[[VAL_23:.*]]) : i32 = (%[[VAL_16]]) to (%[[VAL_17]]) inclusive step (%[[VAL_18]]) {
+! CHECK:                 %[[VAL_24:.*]]:2 = hlfir.declare %[[VAL_19]] {uniq_name = "_QFmultiple_reductions_different_typeEx"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
+! CHECK:                 %[[VAL_25:.*]]:2 = hlfir.declare %[[VAL_20]] {uniq_name = "_QFmultiple_reductions_different_typeEy"} : (!fir.ref<i64>) -> (!fir.ref<i64>, !fir.ref<i64>)
+! CHECK:                 %[[VAL_26:.*]]:2 = hlfir.declare %[[VAL_21]] {uniq_name = "_QFmultiple_reductions_different_typeEz"} : (!fir.ref<f32>) -> (!fir.ref<f32>, !fir.ref<f32>)
+! CHECK:                 %[[VAL_27:.*]]:2 = hlfir.declare %[[VAL_22]] {uniq_name = "_QFmultiple_reductions_different_typeEw"} : (!fir.ref<f64>) -> (!fir.ref<f64>, !fir.ref<f64>)
+! CHECK:                 fir.store %[[VAL_23]] to %[[VAL_15]]#1 : !fir.ref<i32>
+! CHECK:                 %[[VAL_28:.*]] = fir.load %[[VAL_24]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_29:.*]] = fir.load %[[VAL_15]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_30:.*]] = arith.muli %[[VAL_28]], %[[VAL_29]] : i32
+! CHECK:                 hlfir.assign %[[VAL_30]] to %[[VAL_24]]#0 : i32, !fir.ref<i32>
+! CHECK:                 %[[VAL_31:.*]] = fir.load %[[VAL_25]]#0 : !fir.ref<i64>
+! CHECK:                 %[[VAL_32:.*]] = fir.load %[[VAL_15]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_33:.*]] = fir.convert %[[VAL_32]] : (i32) -> i64
+! CHECK:                 %[[VAL_34:.*]] = arith.muli %[[VAL_31]], %[[VAL_33]] : i64
+! CHECK:                 hlfir.assign %[[VAL_34]] to %[[VAL_25]]#0 : i64, !fir.ref<i64>
+! CHECK:                 %[[VAL_35:.*]] = fir.load %[[VAL_26]]#0 : !fir.ref<f32>
+! CHECK:                 %[[VAL_36:.*]] = fir.load %[[VAL_15]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_37:.*]] = fir.convert %[[VAL_36]] : (i32) -> f32
+! CHECK:                 %[[VAL_38:.*]] = arith.mulf %[[VAL_35]], %[[VAL_37]] fastmath<contract> : f32
+! CHECK:                 hlfir.assign %[[VAL_38]] to %[[VAL_26]]#0 : f32, !fir.ref<f32>
+! CHECK:                 %[[VAL_39:.*]] = fir.load %[[VAL_27]]#0 : !fir.ref<f64>
+! CHECK:                 %[[VAL_40:.*]] = fir.load %[[VAL_15]]#0 : !fir.ref<i32>
+! CHECK:                 %[[VAL_41:.*]] = fir.convert %[[VAL_40]] : (i32) -> f64
+! CHECK:                 %[[VAL_42:.*]] = arith.mulf %[[VAL_39]], %[[VAL_41]] fastmath<contract> : f64
+! CHECK:                 hlfir.assign %[[VAL_42]] to %[[VAL_27]]#0 : f64, !fir.ref<f64>
+! CHECK:                 omp.yield
+! CHECK:               omp.terminator
 ! CHECK:             omp.terminator
 ! CHECK:           return
 
diff --git a/flang/test/Lower/OpenMP/wsloop-reduction-multi.f90 b/flang/test/Lower/OpenMP/wsloop-reduction-multi.f90
index 9e9951c399c9..429253efdc80 100644
--- a/flang/test/Lower/OpenMP/wsloop-reduction-multi.f90
+++ b/flang/test/Lower/OpenMP/wsloop-reduction-multi.f90
@@ -35,31 +35,34 @@
 !CHECK: }
 
 !CHECK-LABEL: func.func @_QPmultiple_reduction
-!CHECK:  %[[X_REF:.*]] = fir.alloca i32 {bindc_name = "x", uniq_name = "_QFmultiple_reductionEx"}
-!CHECK:  %[[X_DECL:.*]]:2 = hlfir.declare %[[X_REF]] {uniq_name = "_QFmultiple_reductionEx"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
-!CHECK:  %[[Y_REF:.*]] = fir.alloca f32 {bindc_name = "y", uniq_name = "_QFmultiple_reductionEy"}
-!CHECK:  %[[Y_DECL:.*]]:2 = hlfir.declare %[[Y_REF]] {uniq_name = "_QFmultiple_reductionEy"} : (!fir.ref<f32>) -> (!fir.ref<f32>, !fir.ref<f32>)
-!CHECK:  %[[Z_REF:.*]] = fir.alloca i32 {bindc_name = "z", uniq_name = "_QFmultiple_reductionEz"}
-!CHECK:  %[[Z_DECL:.*]]:2 = hlfir.declare %[[Z_REF]] {uniq_name = "_QFmultiple_reductionEz"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
-!CHECK:  omp.wsloop reduction(
+!CHECK:      %[[X_REF:.*]] = fir.alloca i32 {bindc_name = "x", uniq_name = "_QFmultiple_reductionEx"}
+!CHECK:      %[[X_DECL:.*]]:2 = hlfir.declare %[[X_REF]] {uniq_name = "_QFmultiple_reductionEx"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
+!CHECK:      %[[Y_REF:.*]] = fir.alloca f32 {bindc_name = "y", uniq_name = "_QFmultiple_reductionEy"}
+!CHECK:      %[[Y_DECL:.*]]:2 = hlfir.declare %[[Y_REF]] {uniq_name = "_QFmultiple_reductionEy"} : (!fir.ref<f32>) -> (!fir.ref<f32>, !fir.ref<f32>)
+!CHECK:      %[[Z_REF:.*]] = fir.alloca i32 {bindc_name = "z", uniq_name = "_QFmultiple_reductionEz"}
+!CHECK:      %[[Z_DECL:.*]]:2 = hlfir.declare %[[Z_REF]] {uniq_name = "_QFmultiple_reductionEz"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
+!CHECK:      omp.wsloop reduction(
 !CHECK-SAME: @[[ADD_RED_I32_NAME]] %[[X_DECL]]#0 -> %[[PRV_X:.+]] : !fir.ref<i32>,
 !CHECK-SAME: @[[ADD_RED_F32_NAME]] %[[Y_DECL]]#0 -> %[[PRV_Y:.+]] : !fir.ref<f32>,
-!CHECK-SAME: @[[MIN_RED_I32_NAME]] %[[Z_DECL]]#0 -> %[[PRV_Z:.+]] : !fir.ref<i32>) {{.*}}{
-!CHECK:    %[[PRV_X_DECL:.+]]:2 = hlfir.declare %[[PRV_X]] {{.*}} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
-!CHECK:    %[[PRV_Y_DECL:.+]]:2 = hlfir.declare %[[PRV_Y]] {{.*}} : (!fir.ref<f32>) -> (!fir.ref<f32>, !fir.ref<f32>)
-!CHECK:    %[[PRV_Z_DECL:.+]]:2 = hlfir.declare %[[PRV_Z]] {{.*}} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
-!CHECK:    %[[LPRV_X:.+]] = fir.load %[[PRV_X_DECL]]#0 : !fir.ref<i32>
-!CHECK:    %[[RES_X:.+]] = arith.addi %[[LPRV_X]], %{{.+}} : i32
-!CHECK:    hlfir.assign %[[RES_X]] to %[[PRV_X_DECL]]#0 : i32, !fir.ref<i32>
-!CHECK:    %[[LPRV_Y:.+]] = fir.load %[[PRV_Y_DECL]]#0 : !fir.ref<f32>
-!CHECK:    %[[RES_Y:.+]] = arith.addf %[[LPRV_Y]], %{{.+}} : f32
-!CHECK:    hlfir.assign %[[RES_Y]] to %[[PRV_Y_DECL]]#0 : f32, !fir.ref<f32>
-!CHECK:    %[[LPRV_Z:.+]] = fir.load %[[PRV_Z_DECL]]#0 : !fir.ref<i32>
-!CHECK:    %[[RES_Z:.+]] = arith.select %{{.+}}, %[[LPRV_Z]], %{{.+}} : i32
-!CHECK:    hlfir.assign %[[RES_Z]] to %[[PRV_Z_DECL]]#0 : i32, !fir.ref<i32>
-!CHECK:    omp.yield
-!CHECK:  }
-!CHECK: return
+!CHECK-SAME: @[[MIN_RED_I32_NAME]] %[[Z_DECL]]#0 -> %[[PRV_Z:.+]] : !fir.ref<i32>) {
+!CHECK-NEXT:   omp.loop_nest {{.*}} {
+!CHECK:          %[[PRV_X_DECL:.+]]:2 = hlfir.declare %[[PRV_X]] {{.*}} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
+!CHECK:          %[[PRV_Y_DECL:.+]]:2 = hlfir.declare %[[PRV_Y]] {{.*}} : (!fir.ref<f32>) -> (!fir.ref<f32>, !fir.ref<f32>)
+!CHECK:          %[[PRV_Z_DECL:.+]]:2 = hlfir.declare %[[PRV_Z]] {{.*}} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
+!CHECK:          %[[LPRV_X:.+]] = fir.load %[[PRV_X_DECL]]#0 : !fir.ref<i32>
+!CHECK:          %[[RES_X:.+]] = arith.addi %[[LPRV_X]], %{{.+}} : i32
+!CHECK:          hlfir.assign %[[RES_X]] to %[[PRV_X_DECL]]#0 : i32, !fir.ref<i32>
+!CHECK:          %[[LPRV_Y:.+]] = fir.load %[[PRV_Y_DECL]]#0 : !fir.ref<f32>
+!CHECK:          %[[RES_Y:.+]] = arith.addf %[[LPRV_Y]], %{{.+}} : f32
+!CHECK:          hlfir.assign %[[RES_Y]] to %[[PRV_Y_DECL]]#0 : f32, !fir.ref<f32>
+!CHECK:          %[[LPRV_Z:.+]] = fir.load %[[PRV_Z_DECL]]#0 : !fir.ref<i32>
+!CHECK:          %[[RES_Z:.+]] = arith.select %{{.+}}, %[[LPRV_Z]], %{{.+}} : i32
+!CHECK:          hlfir.assign %[[RES_Z]] to %[[PRV_Z_DECL]]#0 : i32, !fir.ref<i32>
+!CHECK:          omp.yield
+!CHECK:        }
+!CHECK:        omp.terminator
+!CHECK:      }
+!CHECK:      return
 subroutine multiple_reduction(v)
   implicit none
   integer, intent(in) :: v(:)
diff --git a/flang/test/Lower/OpenMP/wsloop-simd.f90 b/flang/test/Lower/OpenMP/wsloop-simd.f90
index c3d5e3e0cda5..1df67474d65e 100644
--- a/flang/test/Lower/OpenMP/wsloop-simd.f90
+++ b/flang/test/Lower/OpenMP/wsloop-simd.f90
@@ -11,23 +11,26 @@ program wsloop_dynamic
 !CHECK:  omp.parallel {
 
 !$OMP DO SCHEDULE(simd: runtime)
-!CHECK:     %[[WS_LB:.*]] = arith.constant 1 : i32
-!CHECK:     %[[WS_UB:.*]] = arith.constant 9 : i32
-!CHECK:     %[[WS_STEP:.*]] = arith.constant 1 : i32
-!CHECK:     omp.wsloop schedule(runtime, simd) nowait for (%[[I:.*]]) : i32 = (%[[WS_LB]]) to (%[[WS_UB]]) inclusive step (%[[WS_STEP]])
-!CHECK:       fir.store %[[I]] to %[[STORE:.*]]#1 : !fir.ref<i32>
+!CHECK:      %[[WS_LB:.*]] = arith.constant 1 : i32
+!CHECK:      %[[WS_UB:.*]] = arith.constant 9 : i32
+!CHECK:      %[[WS_STEP:.*]] = arith.constant 1 : i32
+!CHECK:      omp.wsloop schedule(runtime, simd) nowait {
+!CHECK-NEXT:   omp.loop_nest (%[[I:.*]]) : i32 = (%[[WS_LB]]) to (%[[WS_UB]]) inclusive step (%[[WS_STEP]]) {
+!CHECK:          fir.store %[[I]] to %[[STORE:.*]]#1 : !fir.ref<i32>
 
   do i=1, 9
     print*, i
-!CHECK:    %[[RTBEGIN:.*]] = fir.call @_FortranAioBeginExternalListOutput
-!CHECK:    %[[LOAD:.*]] = fir.load %[[STORE]]#0 : !fir.ref<i32>
-!CHECK:    fir.call @_FortranAioOutputInteger32(%[[RTBEGIN]], %[[LOAD]]) {{.*}}: (!fir.ref<i8>, i32) -> i1
-!CHECK:    fir.call @_FortranAioEndIoStatement(%[[RTBEGIN]]) {{.*}}: (!fir.ref<i8>) -> i32
+!CHECK:          %[[RTBEGIN:.*]] = fir.call @_FortranAioBeginExternalListOutput
+!CHECK:          %[[LOAD:.*]] = fir.load %[[STORE]]#0 : !fir.ref<i32>
+!CHECK:          fir.call @_FortranAioOutputInteger32(%[[RTBEGIN]], %[[LOAD]]) {{.*}}: (!fir.ref<i8>, i32) -> i1
+!CHECK:          fir.call @_FortranAioEndIoStatement(%[[RTBEGIN]]) {{.*}}: (!fir.ref<i8>) -> i32
   end do
-!CHECK:       omp.yield
-!CHECK:         }
-!CHECK:       omp.terminator
-!CHECK:     }
+!CHECK:          omp.yield
+!CHECK:        }
+!CHECK:        omp.terminator
+!CHECK:      }
+!CHECK:      omp.terminator
+!CHECK:    }
 
 !$OMP END DO NOWAIT
 !$OMP END PARALLEL
diff --git a/flang/test/Lower/OpenMP/wsloop-unstructured.f90 b/flang/test/Lower/OpenMP/wsloop-unstructured.f90
index 7fe63a1fe607..bd6a0bade8c7 100644
--- a/flang/test/Lower/OpenMP/wsloop-unstructured.f90
+++ b/flang/test/Lower/OpenMP/wsloop-unstructured.f90
@@ -29,29 +29,32 @@ end subroutine sub
 ! CHECK-SAME:                      %[[VAL_2:.*]]: !fir.ref<!fir.array<?x?xf32>> {fir.bindc_name = "x"},
 ! CHECK-SAME:                      %[[VAL_3:.*]]: !fir.ref<!fir.array<?x?xf32>> {fir.bindc_name = "y"}) {
 ! [...]
-! CHECK:             omp.wsloop for  (%[[VAL_53:.*]], %[[VAL_54:.*]]) : i32 = ({{.*}}) to ({{.*}}) inclusive step ({{.*}}) {
+! CHECK:             omp.wsloop {
+! CHECK-NEXT:          omp.loop_nest (%[[VAL_53:.*]], %[[VAL_54:.*]]) : i32 = ({{.*}}) to ({{.*}}) inclusive step ({{.*}}) {
 ! [...]
-! CHECK:               cf.br ^bb1
-! CHECK:             ^bb1:
-! CHECK:               cf.br ^bb2
-! CHECK:             ^bb2:
+! CHECK:                 cf.br ^bb1
+! CHECK:               ^bb1:
+! CHECK:                 cf.br ^bb2
+! CHECK:               ^bb2:
 ! [...]
-! CHECK:               cf.br ^bb3
-! CHECK:             ^bb3:
+! CHECK:                 cf.br ^bb3
+! CHECK:               ^bb3:
 ! [...]
-! CHECK:               %[[VAL_63:.*]] = arith.cmpi sgt, %{{.*}}, %{{.*}} : i32
-! CHECK:               cf.cond_br %[[VAL_63]], ^bb4, ^bb7
-! CHECK:             ^bb4:
+! CHECK:                 %[[VAL_63:.*]] = arith.cmpi sgt, %{{.*}}, %{{.*}} : i32
+! CHECK:                 cf.cond_br %[[VAL_63]], ^bb4, ^bb7
+! CHECK:               ^bb4:
 ! [...]
-! CHECK:               %[[VAL_76:.*]] = arith.cmpf olt, %{{.*}}, %{{.*}} fastmath<contract> : f32
-! CHECK:               cf.cond_br %[[VAL_76]], ^bb5, ^bb6
-! CHECK:             ^bb5:
-! CHECK:               cf.br ^bb7
-! CHECK:             ^bb6:
+! CHECK:                 %[[VAL_76:.*]] = arith.cmpf olt, %{{.*}}, %{{.*}} fastmath<contract> : f32
+! CHECK:                 cf.cond_br %[[VAL_76]], ^bb5, ^bb6
+! CHECK:               ^bb5:
+! CHECK:                 cf.br ^bb7
+! CHECK:               ^bb6:
 ! [...]
-! CHECK:               cf.br ^bb3
-! CHECK:             ^bb7:
-! CHECK:               omp.yield
+! CHECK:                 cf.br ^bb3
+! CHECK:               ^bb7:
+! CHECK:                 omp.yield
+! CHECK:               }
+! CHECK:               omp.terminator
 ! CHECK:             }
 ! CHECK:             omp.terminator
 ! CHECK:           }
diff --git a/flang/test/Lower/OpenMP/wsloop-variable.f90 b/flang/test/Lower/OpenMP/wsloop-variable.f90
index b3758f1fdc00..4d83b3328803 100644
--- a/flang/test/Lower/OpenMP/wsloop-variable.f90
+++ b/flang/test/Lower/OpenMP/wsloop-variable.f90
@@ -14,26 +14,29 @@ program wsloop_variable
   integer(kind=16) :: i16, i16_lb
   real :: x
 
-!CHECK:  %[[TMP0:.*]] = arith.constant 1 : i32
-!CHECK:  %[[TMP1:.*]] = arith.constant 100 : i32
-!CHECK:  %[[TMP2:.*]] = fir.convert %[[TMP0]] : (i32) -> i64
-!CHECK:  %[[TMP3:.*]] = fir.convert %{{.*}} : (i8) -> i64
-!CHECK:  %[[TMP4:.*]] = fir.convert %{{.*}} : (i16) -> i64
-!CHECK:  %[[TMP5:.*]] = fir.convert %{{.*}} : (i128) -> i64
-!CHECK:  %[[TMP6:.*]] = fir.convert %[[TMP1]] : (i32) -> i64
-!CHECK:  %[[TMP7:.*]] = fir.convert %{{.*}} : (i32) -> i64
-!CHECK:  omp.wsloop for (%[[ARG0:.*]], %[[ARG1:.*]]) : i64 = (%[[TMP2]], %[[TMP5]]) to (%[[TMP3]], %[[TMP6]]) inclusive step (%[[TMP4]], %[[TMP7]]) {
-!CHECK:    %[[ARG0_I16:.*]] = fir.convert %[[ARG0]] : (i64) -> i16
-!CHECK:    fir.store %[[ARG0_I16]] to %[[STORE_IV0:.*]]#1 : !fir.ref<i16>
-!CHECK:    fir.store %[[ARG1]] to %[[STORE_IV1:.*]]#1 : !fir.ref<i64>
-!CHECK:    %[[LOAD_IV0:.*]] = fir.load %[[STORE_IV0]]#0 : !fir.ref<i16>
-!CHECK:    %[[LOAD_IV0_I64:.*]] = fir.convert %[[LOAD_IV0]] : (i16) -> i64
-!CHECK:    %[[LOAD_IV1:.*]] = fir.load %[[STORE_IV1]]#0 : !fir.ref<i64>
-!CHECK:    %[[TMP10:.*]] = arith.addi %[[LOAD_IV0_I64]], %[[LOAD_IV1]] : i64
-!CHECK:    %[[TMP11:.*]] = fir.convert %[[TMP10]] : (i64) -> f32
-!CHECK:    hlfir.assign %[[TMP11]] to %{{.*}} : f32, !fir.ref<f32>
-!CHECK:    omp.yield
-!CHECK:  }
+!CHECK:      %[[TMP0:.*]] = arith.constant 1 : i32
+!CHECK:      %[[TMP1:.*]] = arith.constant 100 : i32
+!CHECK:      %[[TMP2:.*]] = fir.convert %[[TMP0]] : (i32) -> i64
+!CHECK:      %[[TMP3:.*]] = fir.convert %{{.*}} : (i8) -> i64
+!CHECK:      %[[TMP4:.*]] = fir.convert %{{.*}} : (i16) -> i64
+!CHECK:      %[[TMP5:.*]] = fir.convert %{{.*}} : (i128) -> i64
+!CHECK:      %[[TMP6:.*]] = fir.convert %[[TMP1]] : (i32) -> i64
+!CHECK:      %[[TMP7:.*]] = fir.convert %{{.*}} : (i32) -> i64
+!CHECK:      omp.wsloop {
+!CHECK-NEXT:   omp.loop_nest (%[[ARG0:.*]], %[[ARG1:.*]]) : i64 = (%[[TMP2]], %[[TMP5]]) to (%[[TMP3]], %[[TMP6]]) inclusive step (%[[TMP4]], %[[TMP7]]) {
+!CHECK:          %[[ARG0_I16:.*]] = fir.convert %[[ARG0]] : (i64) -> i16
+!CHECK:          fir.store %[[ARG0_I16]] to %[[STORE_IV0:.*]]#1 : !fir.ref<i16>
+!CHECK:          fir.store %[[ARG1]] to %[[STORE_IV1:.*]]#1 : !fir.ref<i64>
+!CHECK:          %[[LOAD_IV0:.*]] = fir.load %[[STORE_IV0]]#0 : !fir.ref<i16>
+!CHECK:          %[[LOAD_IV0_I64:.*]] = fir.convert %[[LOAD_IV0]] : (i16) -> i64
+!CHECK:          %[[LOAD_IV1:.*]] = fir.load %[[STORE_IV1]]#0 : !fir.ref<i64>
+!CHECK:          %[[TMP10:.*]] = arith.addi %[[LOAD_IV0_I64]], %[[LOAD_IV1]] : i64
+!CHECK:          %[[TMP11:.*]] = fir.convert %[[TMP10]] : (i64) -> f32
+!CHECK:          hlfir.assign %[[TMP11]] to %{{.*}} : f32, !fir.ref<f32>
+!CHECK:          omp.yield
+!CHECK:        }
+!CHECK:        omp.terminator
+!CHECK:      }
 
   !$omp do collapse(2)
   do i2 = 1, i1_ub, i2_s
@@ -43,17 +46,20 @@ program wsloop_variable
   end do
   !$omp end do
 
-!CHECK:  %[[TMP12:.*]] = arith.constant 1 : i32
-!CHECK:  %[[TMP13:.*]] = fir.convert %{{.*}} : (i8) -> i32
-!CHECK:  %[[TMP14:.*]] = fir.convert %{{.*}} : (i64) -> i32
-!CHECK:  omp.wsloop for (%[[ARG0:.*]]) : i32 = (%[[TMP12]]) to (%[[TMP13]]) inclusive step (%[[TMP14]])  {
-!CHECK:    %[[ARG0_I16:.*]] = fir.convert %[[ARG0]] : (i32) -> i16
-!CHECK:    fir.store %[[ARG0_I16]] to %[[STORE3:.*]]#1 : !fir.ref<i16>
-!CHECK:    %[[LOAD3:.*]] = fir.load %[[STORE3]]#0 : !fir.ref<i16>
-!CHECK:    %[[TMP16:.*]] = fir.convert %[[LOAD3]] : (i16) -> f32
-!CHECK:    hlfir.assign %[[TMP16]] to %{{.*}} : f32, !fir.ref<f32>
-!CHECK:    omp.yield
-!CHECK:  }
+!CHECK:      %[[TMP12:.*]] = arith.constant 1 : i32
+!CHECK:      %[[TMP13:.*]] = fir.convert %{{.*}} : (i8) -> i32
+!CHECK:      %[[TMP14:.*]] = fir.convert %{{.*}} : (i64) -> i32
+!CHECK:      omp.wsloop {
+!CHECK-NEXT:   omp.loop_nest (%[[ARG0:.*]]) : i32 = (%[[TMP12]]) to (%[[TMP13]]) inclusive step (%[[TMP14]]) {
+!CHECK:          %[[ARG0_I16:.*]] = fir.convert %[[ARG0]] : (i32) -> i16
+!CHECK:          fir.store %[[ARG0_I16]] to %[[STORE3:.*]]#1 : !fir.ref<i16>
+!CHECK:          %[[LOAD3:.*]] = fir.load %[[STORE3]]#0 : !fir.ref<i16>
+!CHECK:          %[[TMP16:.*]] = fir.convert %[[LOAD3]] : (i16) -> f32
+!CHECK:          hlfir.assign %[[TMP16]] to %{{.*}} : f32, !fir.ref<f32>
+!CHECK:          omp.yield
+!CHECK:        }
+!CHECK:        omp.terminator
+!CHECK:      }
 
   !$omp do
   do i2 = 1, i1_ub, i8_s
@@ -61,17 +67,20 @@ program wsloop_variable
   end do
   !$omp end do
 
-!CHECK:  %[[TMP17:.*]] = fir.convert %{{.*}} : (i8) -> i64
-!CHECK:  %[[TMP18:.*]] = fir.convert %{{.*}} : (i16) -> i64
-!CHECK:  %[[TMP19:.*]] = fir.convert %{{.*}} : (i32) -> i64
-!CHECK:  omp.wsloop for (%[[ARG1:.*]]) : i64 = (%[[TMP17]]) to (%[[TMP18]]) inclusive step (%[[TMP19]])  {
-!CHECK:    %[[ARG1_I128:.*]] = fir.convert %[[ARG1]] : (i64) -> i128
-!CHECK:    fir.store %[[ARG1_I128]] to %[[STORE4:.*]]#1 : !fir.ref<i128>
-!CHECK:    %[[LOAD4:.*]] = fir.load %[[STORE4]]#0 : !fir.ref<i128>
-!CHECK:    %[[TMP21:.*]] = fir.convert %[[LOAD4]] : (i128) -> f32
-!CHECK:    hlfir.assign %[[TMP21]] to %{{.*}} : f32, !fir.ref<f32>
-!CHECK:    omp.yield
-!CHECK:  }
+!CHECK:      %[[TMP17:.*]] = fir.convert %{{.*}} : (i8) -> i64
+!CHECK:      %[[TMP18:.*]] = fir.convert %{{.*}} : (i16) -> i64
+!CHECK:      %[[TMP19:.*]] = fir.convert %{{.*}} : (i32) -> i64
+!CHECK:      omp.wsloop {
+!CHECK-NEXT:   omp.loop_nest (%[[ARG1:.*]]) : i64 = (%[[TMP17]]) to (%[[TMP18]]) inclusive step (%[[TMP19]])  {
+!CHECK:          %[[ARG1_I128:.*]] = fir.convert %[[ARG1]] : (i64) -> i128
+!CHECK:          fir.store %[[ARG1_I128]] to %[[STORE4:.*]]#1 : !fir.ref<i128>
+!CHECK:          %[[LOAD4:.*]] = fir.load %[[STORE4]]#0 : !fir.ref<i128>
+!CHECK:          %[[TMP21:.*]] = fir.convert %[[LOAD4]] : (i128) -> f32
+!CHECK:          hlfir.assign %[[TMP21]] to %{{.*}} : f32, !fir.ref<f32>
+!CHECK:          omp.yield
+!CHECK:        }
+!CHECK:        omp.terminator
+!CHECK:      }
 
   !$omp do
   do i16 = i1_lb, i2_ub, i4_s
@@ -118,32 +127,35 @@ subroutine wsloop_variable_sub
 !CHECK:           %[[VAL_24:.*]] = fir.load %[[VAL_13]]#0 : !fir.ref<i16>
 !CHECK:           %[[VAL_25:.*]] = fir.convert %[[VAL_23]] : (i8) -> i32
 !CHECK:           %[[VAL_26:.*]] = fir.convert %[[VAL_24]] : (i16) -> i32
-!CHECK:           omp.wsloop   for  (%[[VAL_27:.*]]) : i32 = (%[[VAL_22]]) to (%[[VAL_25]]) inclusive step (%[[VAL_26]]) {
-!CHECK:             %[[VAL_28:.*]] = fir.convert %[[VAL_27]] : (i32) -> i16
-!CHECK:             fir.store %[[VAL_28]] to %[[VAL_3]]#1 : !fir.ref<i16>
-!CHECK:             %[[VAL_29:.*]] = fir.load %[[VAL_7]]#0 : !fir.ref<i128>
-!CHECK:             %[[VAL_30:.*]] = fir.convert %[[VAL_29]] : (i128) -> index
-!CHECK:             %[[VAL_31:.*]] = arith.constant 100 : i32
-!CHECK:             %[[VAL_32:.*]] = fir.convert %[[VAL_31]] : (i32) -> index
-!CHECK:             %[[VAL_33:.*]] = fir.load %[[VAL_15]]#0 : !fir.ref<i32>
-!CHECK:             %[[VAL_34:.*]] = fir.convert %[[VAL_33]] : (i32) -> index
-!CHECK:             %[[VAL_35:.*]] = fir.convert %[[VAL_30]] : (index) -> i64
-!CHECK:             %[[VAL_36:.*]]:2 = fir.do_loop %[[VAL_37:.*]] = %[[VAL_30]] to %[[VAL_32]] step %[[VAL_34]] iter_args(%[[VAL_38:.*]] = %[[VAL_35]]) -> (index, i64) {
-!CHECK:               fir.store %[[VAL_38]] to %[[VAL_17]]#1 : !fir.ref<i64>
-!CHECK:               %[[VAL_39:.*]] = fir.load %[[VAL_3]]#0 : !fir.ref<i16>
-!CHECK:               %[[VAL_40:.*]] = fir.convert %[[VAL_39]] : (i16) -> i64
-!CHECK:               %[[VAL_41:.*]] = fir.load %[[VAL_17]]#0 : !fir.ref<i64>
-!CHECK:               %[[VAL_42:.*]] = arith.addi %[[VAL_40]], %[[VAL_41]] : i64
-!CHECK:               %[[VAL_43:.*]] = fir.convert %[[VAL_42]] : (i64) -> f32
-!CHECK:               hlfir.assign %[[VAL_43]] to %[[VAL_21]]#0 : f32, !fir.ref<f32>
-!CHECK:               %[[VAL_44:.*]] = arith.addi %[[VAL_37]], %[[VAL_34]] : index
-!CHECK:               %[[VAL_45:.*]] = fir.convert %[[VAL_34]] : (index) -> i64
-!CHECK:               %[[VAL_46:.*]] = fir.load %[[VAL_17]]#1 : !fir.ref<i64>
-!CHECK:               %[[VAL_47:.*]] = arith.addi %[[VAL_46]], %[[VAL_45]] : i64
-!CHECK:               fir.result %[[VAL_44]], %[[VAL_47]] : index, i64
+!CHECK:           omp.wsloop {
+!CHECK-NEXT:        omp.loop_nest (%[[VAL_27:.*]]) : i32 = (%[[VAL_22]]) to (%[[VAL_25]]) inclusive step (%[[VAL_26]]) {
+!CHECK:               %[[VAL_28:.*]] = fir.convert %[[VAL_27]] : (i32) -> i16
+!CHECK:               fir.store %[[VAL_28]] to %[[VAL_3]]#1 : !fir.ref<i16>
+!CHECK:               %[[VAL_29:.*]] = fir.load %[[VAL_7]]#0 : !fir.ref<i128>
+!CHECK:               %[[VAL_30:.*]] = fir.convert %[[VAL_29]] : (i128) -> index
+!CHECK:               %[[VAL_31:.*]] = arith.constant 100 : i32
+!CHECK:               %[[VAL_32:.*]] = fir.convert %[[VAL_31]] : (i32) -> index
+!CHECK:               %[[VAL_33:.*]] = fir.load %[[VAL_15]]#0 : !fir.ref<i32>
+!CHECK:               %[[VAL_34:.*]] = fir.convert %[[VAL_33]] : (i32) -> index
+!CHECK:               %[[VAL_35:.*]] = fir.convert %[[VAL_30]] : (index) -> i64
+!CHECK:               %[[VAL_36:.*]]:2 = fir.do_loop %[[VAL_37:.*]] = %[[VAL_30]] to %[[VAL_32]] step %[[VAL_34]] iter_args(%[[VAL_38:.*]] = %[[VAL_35]]) -> (index, i64) {
+!CHECK:                 fir.store %[[VAL_38]] to %[[VAL_17]]#1 : !fir.ref<i64>
+!CHECK:                 %[[VAL_39:.*]] = fir.load %[[VAL_3]]#0 : !fir.ref<i16>
+!CHECK:                 %[[VAL_40:.*]] = fir.convert %[[VAL_39]] : (i16) -> i64
+!CHECK:                 %[[VAL_41:.*]] = fir.load %[[VAL_17]]#0 : !fir.ref<i64>
+!CHECK:                 %[[VAL_42:.*]] = arith.addi %[[VAL_40]], %[[VAL_41]] : i64
+!CHECK:                 %[[VAL_43:.*]] = fir.convert %[[VAL_42]] : (i64) -> f32
+!CHECK:                 hlfir.assign %[[VAL_43]] to %[[VAL_21]]#0 : f32, !fir.ref<f32>
+!CHECK:                 %[[VAL_44:.*]] = arith.addi %[[VAL_37]], %[[VAL_34]] : index
+!CHECK:                 %[[VAL_45:.*]] = fir.convert %[[VAL_34]] : (index) -> i64
+!CHECK:                 %[[VAL_46:.*]] = fir.load %[[VAL_17]]#1 : !fir.ref<i64>
+!CHECK:                 %[[VAL_47:.*]] = arith.addi %[[VAL_46]], %[[VAL_45]] : i64
+!CHECK:                 fir.result %[[VAL_44]], %[[VAL_47]] : index, i64
+!CHECK:               }
+!CHECK:               fir.store %[[VAL_48:.*]]#1 to %[[VAL_17]]#1 : !fir.ref<i64>
+!CHECK:               omp.yield
 !CHECK:             }
-!CHECK:             fir.store %[[VAL_48:.*]]#1 to %[[VAL_17]]#1 : !fir.ref<i64>
-!CHECK:             omp.yield
+!CHECK:             omp.terminator
 !CHECK:           }
 
   !$omp do
@@ -160,16 +172,19 @@ subroutine wsloop_variable_sub
 !CHECK:           %[[VAL_50:.*]] = arith.constant 1 : i32
 !CHECK:           %[[VAL_51:.*]] = arith.constant 10 : i32
 !CHECK:           %[[VAL_52:.*]] = arith.constant 1 : i32
-!CHECK:           omp.wsloop   for  (%[[VAL_53:.*]]) : i32 = (%[[VAL_50]]) to (%[[VAL_51]]) inclusive step (%[[VAL_52]]) {
-!CHECK:             %[[VAL_54:.*]] = fir.convert %[[VAL_53]] : (i32) -> i8
-!CHECK:             fir.store %[[VAL_54]] to %[[VAL_1]]#1 : !fir.ref<i8>
-!CHECK:             %[[VAL_55:.*]] = fir.load %[[VAL_1]]#0 : !fir.ref<i8>
-!CHECK:             %[[VAL_56:.*]] = fir.load %[[VAL_19]]#0 : !fir.ref<i8>
-!CHECK:             %[[VAL_57:.*]] = arith.cmpi eq, %[[VAL_55]], %[[VAL_56]] : i8
-!CHECK:             fir.if %[[VAL_57]] {
-!CHECK:             } else {
+!CHECK:           omp.wsloop {
+!CHECK-NEXT:        omp.loop_nest (%[[VAL_53:.*]]) : i32 = (%[[VAL_50]]) to (%[[VAL_51]]) inclusive step (%[[VAL_52]]) {
+!CHECK:               %[[VAL_54:.*]] = fir.convert %[[VAL_53]] : (i32) -> i8
+!CHECK:               fir.store %[[VAL_54]] to %[[VAL_1]]#1 : !fir.ref<i8>
+!CHECK:               %[[VAL_55:.*]] = fir.load %[[VAL_1]]#0 : !fir.ref<i8>
+!CHECK:               %[[VAL_56:.*]] = fir.load %[[VAL_19]]#0 : !fir.ref<i8>
+!CHECK:               %[[VAL_57:.*]] = arith.cmpi eq, %[[VAL_55]], %[[VAL_56]] : i8
+!CHECK:               fir.if %[[VAL_57]] {
+!CHECK:               } else {
+!CHECK:               }
+!CHECK:               omp.yield
 !CHECK:             }
-!CHECK:             omp.yield
+!CHECK:             omp.terminator
 !CHECK:           }
   j1 = 5
   !$omp do
diff --git a/flang/test/Lower/OpenMP/wsloop.f90 b/flang/test/Lower/OpenMP/wsloop.f90
index 4068f715c3e1..da90cb724159 100644
--- a/flang/test/Lower/OpenMP/wsloop.f90
+++ b/flang/test/Lower/OpenMP/wsloop.f90
@@ -7,22 +7,24 @@ subroutine simple_loop
   integer :: i
   ! CHECK:  omp.parallel
   !$OMP PARALLEL
-  ! CHECK:     %[[ALLOCA_IV:.*]] = fir.alloca i32 {{{.*}}, pinned}
-  ! CHECK:     %[[IV_DECL:.*]]:2 = hlfir.declare %[[ALLOCA_IV]] {uniq_name = "_QFsimple_loopEi"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
-  ! CHECK:     %[[WS_LB:.*]] = arith.constant 1 : i32
-  ! CHECK:     %[[WS_UB:.*]] = arith.constant 9 : i32
-  ! CHECK:     %[[WS_STEP:.*]] = arith.constant 1 : i32
-  ! CHECK:     omp.wsloop for (%[[I:.*]]) : i32 = (%[[WS_LB]]) to (%[[WS_UB]]) inclusive step (%[[WS_STEP]])
+  ! CHECK:      %[[ALLOCA_IV:.*]] = fir.alloca i32 {{{.*}}, pinned}
+  ! CHECK:      %[[IV_DECL:.*]]:2 = hlfir.declare %[[ALLOCA_IV]] {uniq_name = "_QFsimple_loopEi"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
+  ! CHECK:      %[[WS_LB:.*]] = arith.constant 1 : i32
+  ! CHECK:      %[[WS_UB:.*]] = arith.constant 9 : i32
+  ! CHECK:      %[[WS_STEP:.*]] = arith.constant 1 : i32
+  ! CHECK:      omp.wsloop {
+  ! CHECK-NEXT:   omp.loop_nest (%[[I:.*]]) : i32 = (%[[WS_LB]]) to (%[[WS_UB]]) inclusive step (%[[WS_STEP]]) {
   !$OMP DO
   do i=1, 9
-  ! CHECK:             fir.store %[[I]] to %[[IV_DECL:.*]]#1 : !fir.ref<i32>
-  ! CHECK:             %[[LOAD_IV:.*]] = fir.load %[[IV_DECL]]#0 : !fir.ref<i32>
-  ! CHECK:    fir.call @_FortranAioOutputInteger32({{.*}}, %[[LOAD_IV]]) {{.*}}: (!fir.ref<i8>, i32) -> i1
+  ! CHECK:          fir.store %[[I]] to %[[IV_DECL:.*]]#1 : !fir.ref<i32>
+  ! CHECK:          %[[LOAD_IV:.*]] = fir.load %[[IV_DECL]]#0 : !fir.ref<i32>
+  ! CHECK:          fir.call @_FortranAioOutputInteger32({{.*}}, %[[LOAD_IV]]) {{.*}}: (!fir.ref<i8>, i32) -> i1
     print*, i
   end do
-  ! CHECK:       omp.yield
+  ! CHECK:          omp.yield
+  ! CHECK:        omp.terminator
   !$OMP END DO
-  ! CHECK:       omp.terminator
+  ! CHECK:      omp.terminator
   !$OMP END PARALLEL
 end subroutine
 
@@ -31,22 +33,24 @@ subroutine simple_loop_with_step
   integer :: i
   ! CHECK:  omp.parallel
   !$OMP PARALLEL
-  ! CHECK:     %[[ALLOCA_IV:.*]] = fir.alloca i32 {{{.*}}, pinned}
-  ! CHECK:     %[[IV_DECL:.*]]:2 = hlfir.declare %[[ALLOCA_IV]] {uniq_name = "_QFsimple_loop_with_stepEi"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
-  ! CHECK:     %[[WS_LB:.*]] = arith.constant 1 : i32
-  ! CHECK:     %[[WS_UB:.*]] = arith.constant 9 : i32
-  ! CHECK:     %[[WS_STEP:.*]] = arith.constant 2 : i32
-  ! CHECK:     omp.wsloop for (%[[I:.*]]) : i32 = (%[[WS_LB]]) to (%[[WS_UB]]) inclusive step (%[[WS_STEP]])
-  ! CHECK:       fir.store %[[I]] to %[[IV_DECL]]#1 : !fir.ref<i32>
-  ! CHECK:       %[[LOAD_IV:.*]] = fir.load %[[IV_DECL]]#0 : !fir.ref<i32>
+  ! CHECK:      %[[ALLOCA_IV:.*]] = fir.alloca i32 {{{.*}}, pinned}
+  ! CHECK:      %[[IV_DECL:.*]]:2 = hlfir.declare %[[ALLOCA_IV]] {uniq_name = "_QFsimple_loop_with_stepEi"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
+  ! CHECK:      %[[WS_LB:.*]] = arith.constant 1 : i32
+  ! CHECK:      %[[WS_UB:.*]] = arith.constant 9 : i32
+  ! CHECK:      %[[WS_STEP:.*]] = arith.constant 2 : i32
+  ! CHECK:      omp.wsloop {
+  ! CHECK-NEXT:   omp.loop_nest (%[[I:.*]]) : i32 = (%[[WS_LB]]) to (%[[WS_UB]]) inclusive step (%[[WS_STEP]]) {
+  ! CHECK:          fir.store %[[I]] to %[[IV_DECL]]#1 : !fir.ref<i32>
+  ! CHECK:          %[[LOAD_IV:.*]] = fir.load %[[IV_DECL]]#0 : !fir.ref<i32>
   !$OMP DO
   do i=1, 9, 2
-  ! CHECK:    fir.call @_FortranAioOutputInteger32({{.*}}, %[[LOAD_IV]]) {{.*}}: (!fir.ref<i8>, i32) -> i1
+  ! CHECK:          fir.call @_FortranAioOutputInteger32({{.*}}, %[[LOAD_IV]]) {{.*}}: (!fir.ref<i8>, i32) -> i1
     print*, i
   end do
-  ! CHECK:       omp.yield
+  ! CHECK:          omp.yield
+  ! CHECK:        omp.terminator
   !$OMP END DO
-  ! CHECK:       omp.terminator
+  ! CHECK:      omp.terminator
   !$OMP END PARALLEL
 end subroutine
 
@@ -55,21 +59,23 @@ subroutine loop_with_schedule_nowait
   integer :: i
   ! CHECK:  omp.parallel
   !$OMP PARALLEL
-  ! CHECK:     %[[ALLOCA_IV:.*]] = fir.alloca i32 {{{.*}}, pinned}
-  ! CHECK:     %[[IV_DECL:.*]]:2 = hlfir.declare %[[ALLOCA_IV]] {uniq_name = "_QFloop_with_schedule_nowaitEi"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
-  ! CHECK:     %[[WS_LB:.*]] = arith.constant 1 : i32
-  ! CHECK:     %[[WS_UB:.*]] = arith.constant 9 : i32
-  ! CHECK:     %[[WS_STEP:.*]] = arith.constant 1 : i32
-  ! CHECK:     omp.wsloop schedule(runtime) nowait for (%[[I:.*]]) : i32 = (%[[WS_LB]]) to (%[[WS_UB]]) inclusive step (%[[WS_STEP]])
+  ! CHECK:      %[[ALLOCA_IV:.*]] = fir.alloca i32 {{{.*}}, pinned}
+  ! CHECK:      %[[IV_DECL:.*]]:2 = hlfir.declare %[[ALLOCA_IV]] {uniq_name = "_QFloop_with_schedule_nowaitEi"} : (!fir.ref<i32>) -> (!fir.ref<i32>, !fir.ref<i32>)
+  ! CHECK:      %[[WS_LB:.*]] = arith.constant 1 : i32
+  ! CHECK:      %[[WS_UB:.*]] = arith.constant 9 : i32
+  ! CHECK:      %[[WS_STEP:.*]] = arith.constant 1 : i32
+  ! CHECK:      omp.wsloop schedule(runtime) nowait {
+  ! CHECK-NEXT:   omp.loop_nest (%[[I:.*]]) : i32 = (%[[WS_LB]]) to (%[[WS_UB]]) inclusive step (%[[WS_STEP]]) {
   !$OMP DO SCHEDULE(runtime)
   do i=1, 9
-  ! CHECK:       fir.store %[[I]] to %[[IV_DECL]]#1 : !fir.ref<i32>
-  ! CHECK:       %[[LOAD_IV:.*]] = fir.load %[[IV_DECL]]#0 : !fir.ref<i32>
-  ! CHECK:    fir.call @_FortranAioOutputInteger32({{.*}}, %[[LOAD_IV]]) {{.*}}: (!fir.ref<i8>, i32) -> i1
+  ! CHECK:          fir.store %[[I]] to %[[IV_DECL]]#1 : !fir.ref<i32>
+  ! CHECK:          %[[LOAD_IV:.*]] = fir.load %[[IV_DECL]]#0 : !fir.ref<i32>
+  ! CHECK:          fir.call @_FortranAioOutputInteger32({{.*}}, %[[LOAD_IV]]) {{.*}}: (!fir.ref<i8>, i32) -> i1
     print*, i
   end do
-  ! CHECK:       omp.yield
+  ! CHECK:          omp.yield
+  ! CHECK:        omp.terminator
   !$OMP END DO NOWAIT
-  ! CHECK:       omp.terminator
+  ! CHECK:      omp.terminator
   !$OMP END PARALLEL
 end subroutine
diff --git a/flang/test/Parser/unrecognized-dir.f90 b/flang/test/Parser/unrecognized-dir.f90
index ba6fff7562e2..91fbfc9ee3c3 100644
--- a/flang/test/Parser/unrecognized-dir.f90
+++ b/flang/test/Parser/unrecognized-dir.f90
@@ -1,4 +1,10 @@
 ! RUN: %flang_fc1 -fsyntax-only %s 2>&1 | FileCheck %s
-!CHECK: warning: Compiler directive was ignored
+!CHECK: warning: Unrecognized compiler directive was ignored
 !DIR$ Not a recognized directive
+program main
+ contains
+  !CHECK: warning: Compiler directive ignored here
+  !DIR$ not in a subprogram
+  subroutine s
+  end
 end
diff --git a/flang/test/Preprocessing/backslash-contin1.F90 b/flang/test/Preprocessing/backslash-contin1.F90
new file mode 100644
index 000000000000..cf2ed36370da
--- /dev/null
+++ b/flang/test/Preprocessing/backslash-contin1.F90
@@ -0,0 +1,8 @@
+! RUN: %flang -E %s | FileCheck %s
+print *, \
+  "hello, \
+world"
+end
+!CHECK:      print *, "hello, world"
+!CHECK:      end
+
diff --git a/flang/test/Semantics/OpenMP/firstprivate02.f90 b/flang/test/Semantics/OpenMP/firstprivate02.f90
new file mode 100644
index 000000000000..eb2597cb1cc4
--- /dev/null
+++ b/flang/test/Semantics/OpenMP/firstprivate02.f90
@@ -0,0 +1,20 @@
+! RUN: %python %S/../test_errors.py %s %flang -fopenmp
+! OpenMP Version 5.2, Sections 3.2.1 & 5.3
+subroutine omp_firstprivate(init)
+  integer :: init
+  integer :: a(10)
+  type my_type
+    integer :: val
+  end type my_type
+  type(my_type) :: my_var
+
+  !ERROR: A variable that is part of another variable (as an array or structure element) cannot appear in a FIRSTPRIVATE clause
+  !$omp parallel firstprivate(a(2))
+    a(2) = init
+  !$omp end parallel
+
+  !ERROR: A variable that is part of another variable (as an array or structure element) cannot appear in a FIRSTPRIVATE clause
+  !$omp parallel firstprivate(my_var%val)
+    my_var%val = init
+  !$omp end parallel
+end subroutine
diff --git a/flang/test/Semantics/OpenMP/lastprivate03.f90 b/flang/test/Semantics/OpenMP/lastprivate03.f90
new file mode 100644
index 000000000000..d7fe0c162f27
--- /dev/null
+++ b/flang/test/Semantics/OpenMP/lastprivate03.f90
@@ -0,0 +1,24 @@
+! RUN: %python %S/../test_errors.py %s %flang -fopenmp
+! OpenMP Version 5.2, Sections 3.2.1 & 5.3
+subroutine omp_lastprivate(init)
+  integer :: init
+  integer :: i, a(10)
+  type my_type
+    integer :: val
+  end type my_type
+  type(my_type) :: my_var
+
+  !ERROR: A variable that is part of another variable (as an array or structure element) cannot appear in a LASTPRIVATE clause
+  !$omp do lastprivate(a(2))
+  do i=1, 10
+    a(2) = init
+  end do
+  !$omp end do
+
+  !ERROR: A variable that is part of another variable (as an array or structure element) cannot appear in a LASTPRIVATE clause
+  !$omp do lastprivate(my_var%val)
+  do i=1, 10
+    my_var%val = init
+  end do
+  !$omp end do
+end subroutine
diff --git a/flang/test/Semantics/OpenMP/parallel-private01.f90 b/flang/test/Semantics/OpenMP/parallel-private01.f90
index 0f7ffcabda6b..a3d332c95ed2 100644
--- a/flang/test/Semantics/OpenMP/parallel-private01.f90
+++ b/flang/test/Semantics/OpenMP/parallel-private01.f90
@@ -10,7 +10,7 @@ program omp_parallel_private
 
   type(my_type) :: my_var
 
-  !ERROR: A variable that is part of another variable (as an array or structure element) cannot appear in a PRIVATE or SHARED clause
+  !ERROR: A variable that is part of another variable (as an array or structure element) cannot appear in a PRIVATE clause
   !$omp parallel private(my_var%array)
   do i = 1, 10
     c(i) = a(i) + b(i) + k
diff --git a/flang/test/Semantics/OpenMP/parallel-private02.f90 b/flang/test/Semantics/OpenMP/parallel-private02.f90
index b649db972548..8cb72159d6ab 100644
--- a/flang/test/Semantics/OpenMP/parallel-private02.f90
+++ b/flang/test/Semantics/OpenMP/parallel-private02.f90
@@ -10,7 +10,7 @@ program omp_parallel_private
     array(i) = i
   end do
 
-  !ERROR: A variable that is part of another variable (as an array or structure element) cannot appear in a PRIVATE or SHARED clause
+  !ERROR: A variable that is part of another variable (as an array or structure element) cannot appear in a PRIVATE clause
   !$omp parallel private(array(i))
   do i = 1, 10
     c(i) = a(i) + b(i) + k
diff --git a/flang/test/Semantics/OpenMP/parallel-private03.f90 b/flang/test/Semantics/OpenMP/parallel-private03.f90
index 1ec93e3e0dba..24a096302e53 100644
--- a/flang/test/Semantics/OpenMP/parallel-private03.f90
+++ b/flang/test/Semantics/OpenMP/parallel-private03.f90
@@ -17,7 +17,7 @@ program omp_parallel_private
     arr(i) = 0.0
   end do
 
-  !ERROR: A variable that is part of another variable (as an array or structure element) cannot appear in a PRIVATE or SHARED clause
+  !ERROR: A variable that is part of another variable (as an array or structure element) cannot appear in a PRIVATE clause
   !$omp parallel private(arr(i),intx)
   do i = 1, 10
     c(i) = a(i) + b(i) + k
diff --git a/flang/test/Semantics/OpenMP/parallel-private04.f90 b/flang/test/Semantics/OpenMP/parallel-private04.f90
index dbab1564e40f..67a669c9882a 100644
--- a/flang/test/Semantics/OpenMP/parallel-private04.f90
+++ b/flang/test/Semantics/OpenMP/parallel-private04.f90
@@ -17,7 +17,7 @@ program omp_parallel_private
     arr(i) = 0.0
   end do
 
-  !ERROR: A variable that is part of another variable (as an array or structure element) cannot appear in a PRIVATE or SHARED clause
+  !ERROR: A variable that is part of another variable (as an array or structure element) cannot appear in a PRIVATE clause
   !$omp parallel private(arr,intx,my_var%array(1))
   do i = 1, 10
     c(i) = a(i) + b(i) + k
diff --git a/flang/test/Semantics/OpenMP/parallel-sections01.f90 b/flang/test/Semantics/OpenMP/parallel-sections01.f90
index 2bf58ea2cb29..b073cc8223b6 100644
--- a/flang/test/Semantics/OpenMP/parallel-sections01.f90
+++ b/flang/test/Semantics/OpenMP/parallel-sections01.f90
@@ -17,10 +17,10 @@ program OmpConstructSections01
    do i = 1, 10
       array(i) = i
    end do
-!ERROR: A variable that is part of another variable (as an array or structure element) cannot appear in a PRIVATE or SHARED clause
+!ERROR: A variable that is part of another variable (as an array or structure element) cannot appear in a SHARED clause
 !$omp parallel sections shared(array(i))
 !$omp end parallel sections
-!ERROR: A variable that is part of another variable (as an array or structure element) cannot appear in a PRIVATE or SHARED clause
+!ERROR: A variable that is part of another variable (as an array or structure element) cannot appear in a SHARED clause
 !$omp parallel sections shared(my_var%array)
 !$omp end parallel sections
 
@@ -30,7 +30,7 @@ program OmpConstructSections01
    if (NT) 20, 30, 40
 !ERROR: invalid branch into an OpenMP structured block
    goto 20
-!ERROR: A variable that is part of another variable (as an array or structure element) cannot appear in a PRIVATE or SHARED clause
+!ERROR: A variable that is part of another variable (as an array or structure element) cannot appear in a PRIVATE clause
 !$omp parallel sections private(my_var%array)
    !$omp section
    print *, "This is a single statement structured block"
@@ -53,7 +53,7 @@ program OmpConstructSections01
 30 print *, "Error in opening file"
 !$omp end parallel sections
 10 print *, 'Jump from section'
-!ERROR: A variable that is part of another variable (as an array or structure element) cannot appear in a PRIVATE or SHARED clause
+!ERROR: A variable that is part of another variable (as an array or structure element) cannot appear in a PRIVATE clause
 !$omp parallel sections private(array(i))
    !$omp section
 40 print *, 'Error in opening file'
diff --git a/flang/test/Semantics/OpenMP/parallel-shared01.f90 b/flang/test/Semantics/OpenMP/parallel-shared01.f90
index d9ed9bc2efe2..7abfe1f7b163 100644
--- a/flang/test/Semantics/OpenMP/parallel-shared01.f90
+++ b/flang/test/Semantics/OpenMP/parallel-shared01.f90
@@ -10,7 +10,7 @@ program omp_parallel_shared
 
   type(my_type) :: my_var
 
-  !ERROR: A variable that is part of another variable (as an array or structure element) cannot appear in a PRIVATE or SHARED clause
+  !ERROR: A variable that is part of another variable (as an array or structure element) cannot appear in a SHARED clause
   !$omp parallel shared(my_var%array)
   do i = 1, 10
     c(i) = a(i) + b(i) +  k
diff --git a/flang/test/Semantics/OpenMP/parallel-shared02.f90 b/flang/test/Semantics/OpenMP/parallel-shared02.f90
index f46cfa17ba38..f59f5236dfd9 100644
--- a/flang/test/Semantics/OpenMP/parallel-shared02.f90
+++ b/flang/test/Semantics/OpenMP/parallel-shared02.f90
@@ -10,7 +10,7 @@ program omp_parallel_shared
     array(i) = i
   end do
 
-  !ERROR: A variable that is part of another variable (as an array or structure element) cannot appear in a PRIVATE or SHARED clause
+  !ERROR: A variable that is part of another variable (as an array or structure element) cannot appear in a SHARED clause
   !$omp parallel shared(array(i))
   do i = 1, 10
     c(i) = a(i) + b(i) + k
diff --git a/flang/test/Semantics/OpenMP/parallel-shared03.f90 b/flang/test/Semantics/OpenMP/parallel-shared03.f90
index 801ffba424a7..3d9111c7aaf1 100644
--- a/flang/test/Semantics/OpenMP/parallel-shared03.f90
+++ b/flang/test/Semantics/OpenMP/parallel-shared03.f90
@@ -17,7 +17,7 @@ program omp_parallel_shared
     arr(i) = 0.0
   end do
 
-  !ERROR: A variable that is part of another variable (as an array or structure element) cannot appear in a PRIVATE or SHARED clause
+  !ERROR: A variable that is part of another variable (as an array or structure element) cannot appear in a SHARED clause
   !$omp parallel shared(arr(i),intx)
   do i = 1, 10
      c(i) = a(i) + b(i) + k
diff --git a/flang/test/Semantics/OpenMP/parallel-shared04.f90 b/flang/test/Semantics/OpenMP/parallel-shared04.f90
index 6f170c6a6ba7..06b7fcfa01d7 100644
--- a/flang/test/Semantics/OpenMP/parallel-shared04.f90
+++ b/flang/test/Semantics/OpenMP/parallel-shared04.f90
@@ -17,7 +17,7 @@ program omp_parallel_shared
     arr(i) = 0.0
   end do
 
-  !ERROR: A variable that is part of another variable (as an array or structure element) cannot appear in a PRIVATE or SHARED clause
+  !ERROR: A variable that is part of another variable (as an array or structure element) cannot appear in a SHARED clause
   !$omp parallel shared(arr,intx,my_var%array(1))
   do i = 1, 10
     c(i) = a(i) + b(i) + k
diff --git a/flang/test/Semantics/arg-convert.f90 b/flang/test/Semantics/arg-convert.f90
new file mode 100644
index 000000000000..7951bedf49d0
--- /dev/null
+++ b/flang/test/Semantics/arg-convert.f90
@@ -0,0 +1,16 @@
+!RUN: %flang_fc1 -fdebug-unparse  %s  2>&1 | FileCheck %s
+!Ensure that argument conversion does not take place when the procedure
+!interface is implicit at the point of call, even when the interface
+!is known due because the procedure's definition is in the same source file.
+
+subroutine test
+!CHECK: warning: If the procedure's interface were explicit, this reference would be in error
+!CHECK: because: Actual argument type 'INTEGER(8)' is not compatible with dummy argument type 'INTEGER(4)'
+!CHECK: CALL samesourcefile((1_8))
+  call sameSourceFile((1_8))
+!CHECK: CALL somewhereelse((2_8))
+  call somewhereElse((2_8))
+end
+
+subroutine sameSourceFile(n)
+end
diff --git a/flang/test/Semantics/cuf02.cuf b/flang/test/Semantics/cuf02.cuf
index 881a3005e281..a4a229565a3e 100644
--- a/flang/test/Semantics/cuf02.cuf
+++ b/flang/test/Semantics/cuf02.cuf
@@ -5,14 +5,11 @@ module m
     end
   end interface
  contains
-  !ERROR: A device subprogram may not be RECURSIVE, PURE, or ELEMENTAL
-  recursive attributes(device) subroutine s1
+  recursive attributes(device) subroutine s1 ! ok
   end
-  !ERROR: A device subprogram may not be RECURSIVE, PURE, or ELEMENTAL
-  pure attributes(device) subroutine s2
+  pure attributes(device) subroutine s2 ! ok
   end
-  !ERROR: A device subprogram may not be RECURSIVE, PURE, or ELEMENTAL
-  elemental attributes(device) subroutine s3
+  elemental attributes(device) subroutine s3 ! ok
   end
   subroutine s4
    contains
@@ -32,14 +29,11 @@ module m
   !ERROR: A function may not have ATTRIBUTES(GLOBAL) or ATTRIBUTES(GRID_GLOBAL)
   attributes(global) real function f1
   end
-  !ERROR: A device subprogram may not be RECURSIVE, PURE, or ELEMENTAL
-  recursive attributes(global) subroutine s7
+  recursive attributes(global) subroutine s7 ! ok
   end
-  !ERROR: A device subprogram may not be RECURSIVE, PURE, or ELEMENTAL
-  pure attributes(global) subroutine s8
+  pure attributes(global) subroutine s8 ! ok
   end
-  !ERROR: A device subprogram may not be RECURSIVE, PURE, or ELEMENTAL
-  elemental attributes(global) subroutine s9
+  elemental attributes(global) subroutine s9 ! ok
   end
 end
 
diff --git a/flang/test/Semantics/declarations03.f90 b/flang/test/Semantics/declarations03.f90
index 3459b2287b2b..65b07e7d5c65 100644
--- a/flang/test/Semantics/declarations03.f90
+++ b/flang/test/Semantics/declarations03.f90
@@ -19,7 +19,7 @@ module m
   common /blk4/ w
   bind(c, name="cc") :: t2, /blk4/
 
-  !ERROR: The entity 'blk5' has multiple BIND names
+  !ERROR: The entity 'blk5' has multiple BIND names ('dd' and 'ee')
   common /blk5/ i
   bind(c, name="dd") :: /blk5/
   bind(c, name="ee") :: /blk5/
@@ -29,7 +29,7 @@ module m
   bind(c, name="ff") :: /blk6/
   bind(c, name="ff") :: /blk7/
 
-  !ERROR: The entity 's1' has multiple BIND names
+  !ERROR: The entity 's1' has multiple BIND names ('gg' and 'hh')
   integer :: s1
   bind(c, name="gg") :: s1
   !ERROR: BIND_C attribute was already specified on 's1'
@@ -40,12 +40,12 @@ module m
   bind(c, name="ii") :: s2
   bind(c, name="ii") :: s3
 
-  !ERROR: The entity 's4' has multiple BIND names
+  !ERROR: The entity 's4' has multiple BIND names ('ss1' and 'jj')
   integer, bind(c, name="ss1") :: s4
   !ERROR: BIND_C attribute was already specified on 's4'
   bind(c, name="jj") :: s4
 
-  !ERROR: The entity 's5' has multiple BIND names
+  !ERROR: The entity 's5' has multiple BIND names ('kk' and 'ss2')
   bind(c, name="kk") :: s5
   !ERROR: BIND_C attribute was already specified on 's5'
   integer, bind(c, name="ss2") :: s5
@@ -72,3 +72,8 @@ module b
   !ERROR: Two entities have the same global name 'int'
   integer, bind(c, name="int") :: i
 end module
+
+module c
+  bind(c, name = "AAA") a
+  integer aaa ! ensure no bogus error about multiple binding names
+end module
diff --git a/flang/test/Transforms/debug-line-table-inc-file.fir b/flang/test/Transforms/debug-line-table-inc-file.fir
index be4f005bf664..dc75482d4f8a 100644
--- a/flang/test/Transforms/debug-line-table-inc-file.fir
+++ b/flang/test/Transforms/debug-line-table-inc-file.fir
@@ -1,5 +1,5 @@
 
-// RUN: fir-opt --add-debug-info --mlir-print-debuginfo %s | FileCheck %s
+// RUN: fir-opt --add-debug-info="debug-level=LineTablesOnly" --mlir-print-debuginfo %s | FileCheck %s
 // REQUIRES: system-linux
 
 // Test for included functions that have a different debug location than the current file
@@ -30,7 +30,7 @@ module attributes {} {
 // CHECK: #[[MODULE_LOC]] = loc("{{.*}}simple.f90":0:0)
 // CHECK: #[[LOC_INC_FILE:.*]] = loc("{{.*}}inc.f90":1:1)
 // CHECK: #[[LOC_FILE:.*]] = loc("{{.*}}simple.f90":3:1)
-// CHECK: #[[DI_CU:.*]] = #llvm.di_compile_unit<id = distinct[{{.*}}]<>, sourceLanguage = DW_LANG_Fortran95, file = #[[DI_FILE]], producer = "Flang", isOptimized = false, emissionKind = LineTablesOnly>
+// CHECK: #[[DI_CU:.*]] = #llvm.di_compile_unit<id = distinct[{{.*}}]<>, sourceLanguage = DW_LANG_Fortran95, file = #[[DI_FILE]], producer = "flang{{.*}}", isOptimized = false, emissionKind = LineTablesOnly>
 // CHECK: #[[DI_SP_INC:.*]] = #llvm.di_subprogram<id = distinct[{{.*}}]<>, compileUnit = #[[DI_CU]], scope = #[[DI_FILE]], name = "_QPsinc", linkageName = "_QPsinc", file = #[[DI_INC_FILE]], {{.*}}>
 // CHECK: #[[DI_SP:.*]] = #llvm.di_subprogram<id = distinct[{{.*}}]<>, compileUnit = #[[DI_CU]], scope = #[[DI_FILE]], name = "_QQmain", linkageName = "_QQmain", file = #[[DI_FILE]], {{.*}}>
 // CHECK: #[[FUSED_LOC_INC_FILE]] = loc(fused<#[[DI_SP_INC]]>[#[[LOC_INC_FILE]]])
diff --git a/flang/test/Transforms/debug-line-table.fir b/flang/test/Transforms/debug-line-table.fir
index 0ba88d3d9f7f..3b3a39174df0 100644
--- a/flang/test/Transforms/debug-line-table.fir
+++ b/flang/test/Transforms/debug-line-table.fir
@@ -1,5 +1,7 @@
 
-// RUN: fir-opt --add-debug-info --mlir-print-debuginfo %s | FileCheck %s
+// RUN: fir-opt --add-debug-info="debug-level=Full" --mlir-print-debuginfo %s | FileCheck %s --check-prefix=FULL
+// RUN: fir-opt --add-debug-info="debug-level=LineTablesOnly" --mlir-print-debuginfo %s | FileCheck %s --check-prefix=LINETABLE
+// RUN: fir-opt --add-debug-info="is-optimized=true" --mlir-print-debuginfo %s | FileCheck %s --check-prefix=OPT
 
 module attributes { fir.defaultkind = "a1c4d8i4l4r4", fir.kindmap = "", llvm.data_layout = "e-m:e-i8:8:32-i16:16:32-i64:64-i128:128-n32:64-S128", llvm.target_triple = "aarch64-unknown-linux-gnu"} {
   func.func @_QPsb() {
@@ -22,7 +24,9 @@ module attributes { fir.defaultkind = "a1c4d8i4l4r4", fir.kindmap = "", llvm.dat
 // CHECK: #[[MODULE_LOC]] = loc("[[DIR_NAME]]/[[FILE_NAME]]":1:1)
 // CHECK: #[[SB_LOC]] = loc("./simple.f90":2:1)
 // CHECK: #[[DECL_LOC:.*]] = loc("./simple.f90":10:1)
-// CHECK: #di_compile_unit = #llvm.di_compile_unit<id = distinct[{{.*}}]<>, sourceLanguage = DW_LANG_Fortran95, file = #di_file, producer = "Flang", isOptimized = false, emissionKind = LineTablesOnly>
+// FULL: #di_compile_unit = #llvm.di_compile_unit<id = distinct[{{.*}}]<>, sourceLanguage = DW_LANG_Fortran95, file = #di_file, producer = "flang{{.*}}", isOptimized = false, emissionKind = Full>
+// OPT: #di_compile_unit = #llvm.di_compile_unit<id = distinct[{{.*}}]<>, sourceLanguage = DW_LANG_Fortran95, file = #di_file, producer = "flang{{.*}}", isOptimized = true, emissionKind = Full>
+// LINETABLE: #di_compile_unit = #llvm.di_compile_unit<id = distinct[{{.*}}]<>, sourceLanguage = DW_LANG_Fortran95, file = #di_file, producer = "flang{{.*}}", isOptimized = false, emissionKind = LineTablesOnly>
 // CHECK: #di_subroutine_type = #llvm.di_subroutine_type<callingConvention = DW_CC_normal, types = #di_basic_type, #di_basic_type>
 // CHECK: #[[SB_SUBPROGRAM:.*]] = #llvm.di_subprogram<id = distinct[{{.*}}]<>, compileUnit = #di_compile_unit, scope = #di_file, name = "[[SB_NAME]]", linkageName = "[[SB_NAME]]", file = #di_file, line = 1, scopeLine = 1, subprogramFlags = "Definition|Optimized", type = #di_subroutine_type>
 // CHECK: #[[DECL_SUBPROGRAM:.*]] = #llvm.di_subprogram<scope = #di_file, name = "[[DECL_NAME]]", linkageName = "[[DECL_NAME]]", file = #di_file, line = 1, scopeLine = 1, subprogramFlags = Optimized, type = #di_subroutine_type>
diff --git a/flang/test/Transforms/omp-reduction-cfg-conversion.fir b/flang/test/Transforms/omp-reduction-cfg-conversion.fir
index 3103c4456d72..707e665132af 100644
--- a/flang/test/Transforms/omp-reduction-cfg-conversion.fir
+++ b/flang/test/Transforms/omp-reduction-cfg-conversion.fir
@@ -1,4 +1,4 @@
-// RUN: fir-opt --cfg-conversion-on-reduce-opt %s | FileCheck %s
+// RUN: fir-opt --cfg-conversion %s | FileCheck %s
 
 omp.declare_reduction @add_reduction_i_32_box_3_byref : !fir.ref<!fir.box<!fir.array<3xi32>>> init {
 ^bb0(%arg0: !fir.ref<!fir.box<!fir.array<3xi32>>>):
diff --git a/flang/tools/flang-driver/CMakeLists.txt b/flang/tools/flang-driver/CMakeLists.txt
index 3ce8b407450d..ce30ecff028d 100644
--- a/flang/tools/flang-driver/CMakeLists.txt
+++ b/flang/tools/flang-driver/CMakeLists.txt
@@ -21,7 +21,6 @@ add_flang_tool(flang-new
   # unable to generate executables.
   FortranRuntime
   FortranDecimal
-  Fortran_main
 )
 
 target_link_libraries(flang-new
diff --git a/flang/unittests/Optimizer/Builder/ComplexTest.cpp b/flang/unittests/Optimizer/Builder/ComplexTest.cpp
index 5364eec904ff..17171512470a 100644
--- a/flang/unittests/Optimizer/Builder/ComplexTest.cpp
+++ b/flang/unittests/Optimizer/Builder/ComplexTest.cpp
@@ -96,6 +96,6 @@ TEST_F(ComplexTest, verifyConvertWithSemantics) {
 
   // Convert complex to integer
   mlir::Value v2 = firBuilder->convertWithSemantics(loc, integerTy1, v1);
-  EXPECT_TRUE(v2.getType().isa<mlir::IntegerType>());
+  EXPECT_TRUE(mlir::isa<mlir::IntegerType>(v2.getType()));
   EXPECT_TRUE(mlir::dyn_cast<fir::ConvertOp>(v2.getDefiningOp()));
 }
diff --git a/flang/unittests/Optimizer/Builder/DoLoopHelperTest.cpp b/flang/unittests/Optimizer/Builder/DoLoopHelperTest.cpp
index 7e7206dbf934..d0a9342914a3 100644
--- a/flang/unittests/Optimizer/Builder/DoLoopHelperTest.cpp
+++ b/flang/unittests/Optimizer/Builder/DoLoopHelperTest.cpp
@@ -34,7 +34,7 @@ public:
 void checkConstantValue(const mlir::Value &value, int64_t v) {
   EXPECT_TRUE(mlir::isa<mlir::arith::ConstantOp>(value.getDefiningOp()));
   auto cstOp = dyn_cast<mlir::arith::ConstantOp>(value.getDefiningOp());
-  auto valueAttr = cstOp.getValue().dyn_cast_or_null<IntegerAttr>();
+  auto valueAttr = dyn_cast_or_null<IntegerAttr>(cstOp.getValue());
   EXPECT_EQ(v, valueAttr.getInt());
 }
 
diff --git a/flang/unittests/Optimizer/Builder/FIRBuilderTest.cpp b/flang/unittests/Optimizer/Builder/FIRBuilderTest.cpp
index b6a1f9c9db8f..e5e5454ee88a 100644
--- a/flang/unittests/Optimizer/Builder/FIRBuilderTest.cpp
+++ b/flang/unittests/Optimizer/Builder/FIRBuilderTest.cpp
@@ -54,7 +54,7 @@ static void checkIntegerConstant(mlir::Value value, mlir::Type ty, int64_t v) {
   EXPECT_TRUE(mlir::isa<mlir::arith::ConstantOp>(value.getDefiningOp()));
   auto cstOp = dyn_cast<mlir::arith::ConstantOp>(value.getDefiningOp());
   EXPECT_EQ(ty, cstOp.getType());
-  auto valueAttr = cstOp.getValue().dyn_cast_or_null<IntegerAttr>();
+  auto valueAttr = mlir::dyn_cast_or_null<IntegerAttr>(cstOp.getValue());
   EXPECT_EQ(v, valueAttr.getInt());
 }
 
@@ -151,7 +151,7 @@ TEST_F(FIRBuilderTest, createRealZeroConstant) {
   auto cstOp = dyn_cast<arith::ConstantOp>(cst.getDefiningOp());
   EXPECT_EQ(realTy, cstOp.getType());
   EXPECT_EQ(
-      0u, cstOp.getValue().cast<FloatAttr>().getValue().convertToDouble());
+      0u, mlir::cast<FloatAttr>(cstOp.getValue()).getValue().convertToDouble());
 }
 
 TEST_F(FIRBuilderTest, createBool) {
@@ -164,8 +164,8 @@ TEST_F(FIRBuilderTest, createBool) {
 TEST_F(FIRBuilderTest, getVarLenSeqTy) {
   auto builder = getBuilder();
   auto ty = builder.getVarLenSeqTy(builder.getI64Type());
-  EXPECT_TRUE(ty.isa<fir::SequenceType>());
-  fir::SequenceType seqTy = ty.dyn_cast<fir::SequenceType>();
+  EXPECT_TRUE(mlir::isa<fir::SequenceType>(ty));
+  fir::SequenceType seqTy = mlir::dyn_cast<fir::SequenceType>(ty);
   EXPECT_EQ(1u, seqTy.getDimension());
   EXPECT_TRUE(fir::unwrapSequenceType(ty).isInteger(64));
 }
@@ -216,9 +216,9 @@ TEST_F(FIRBuilderTest, createGlobal2) {
   EXPECT_FALSE(global.getConstant().has_value());
   EXPECT_EQ(i32Type, global.getType());
   EXPECT_TRUE(global.getInitVal().has_value());
-  EXPECT_TRUE(global.getInitVal().value().isa<mlir::IntegerAttr>());
-  EXPECT_EQ(
-      16, global.getInitVal().value().cast<mlir::IntegerAttr>().getValue());
+  EXPECT_TRUE(mlir::isa<mlir::IntegerAttr>(global.getInitVal().value()));
+  EXPECT_EQ(16,
+      mlir::cast<mlir::IntegerAttr>(global.getInitVal().value()).getValue());
   EXPECT_TRUE(global.getLinkName().has_value());
   EXPECT_EQ(
       builder.createLinkOnceLinkage().getValue(), global.getLinkName().value());
@@ -271,12 +271,12 @@ TEST_F(FIRBuilderTest, locationToFilename) {
   auto stringLitOps = global.getRegion().front().getOps<fir::StringLitOp>();
   EXPECT_TRUE(llvm::hasSingleElement(stringLitOps));
   for (auto stringLit : stringLitOps) {
-    EXPECT_EQ(10, stringLit.getSize().cast<mlir::IntegerAttr>().getValue());
-    EXPECT_TRUE(stringLit.getValue().isa<StringAttr>());
+    EXPECT_EQ(
+        10, mlir::cast<mlir::IntegerAttr>(stringLit.getSize()).getValue());
+    EXPECT_TRUE(mlir::isa<StringAttr>(stringLit.getValue()));
     EXPECT_EQ(0,
         strcmp("file1.f90\0",
-            stringLit.getValue()
-                .dyn_cast<StringAttr>()
+            mlir::dyn_cast<StringAttr>(stringLit.getValue())
                 .getValue()
                 .str()
                 .c_str()));
@@ -288,9 +288,9 @@ TEST_F(FIRBuilderTest, createStringLitOp) {
   llvm::StringRef data("mystringlitdata");
   auto loc = builder.getUnknownLoc();
   auto op = builder.createStringLitOp(loc, data);
-  EXPECT_EQ(15, op.getSize().cast<mlir::IntegerAttr>().getValue());
-  EXPECT_TRUE(op.getValue().isa<StringAttr>());
-  EXPECT_EQ(data, op.getValue().dyn_cast<StringAttr>().getValue());
+  EXPECT_EQ(15, mlir::cast<mlir::IntegerAttr>(op.getSize()).getValue());
+  EXPECT_TRUE(mlir::isa<StringAttr>(op.getValue()));
+  EXPECT_EQ(data, mlir::dyn_cast<StringAttr>(op.getValue()).getValue());
 }
 
 TEST_F(FIRBuilderTest, createStringLiteral) {
@@ -318,9 +318,11 @@ TEST_F(FIRBuilderTest, createStringLiteral) {
   auto stringLitOps = global.getRegion().front().getOps<fir::StringLitOp>();
   EXPECT_TRUE(llvm::hasSingleElement(stringLitOps));
   for (auto stringLit : stringLitOps) {
-    EXPECT_EQ(16, stringLit.getSize().cast<mlir::IntegerAttr>().getValue());
-    EXPECT_TRUE(stringLit.getValue().isa<StringAttr>());
-    EXPECT_EQ(strValue, stringLit.getValue().dyn_cast<StringAttr>().getValue());
+    EXPECT_EQ(
+        16, mlir::cast<mlir::IntegerAttr>(stringLit.getSize()).getValue());
+    EXPECT_TRUE(mlir::isa<StringAttr>(stringLit.getValue()));
+    EXPECT_EQ(
+        strValue, mlir::dyn_cast<StringAttr>(stringLit.getValue()).getValue());
   }
 }
 
@@ -344,7 +346,7 @@ TEST_F(FIRBuilderTest, allocateLocal) {
 static void checkShapeOp(mlir::Value shape, mlir::Value c10, mlir::Value c100) {
   EXPECT_TRUE(mlir::isa<fir::ShapeOp>(shape.getDefiningOp()));
   fir::ShapeOp op = dyn_cast<fir::ShapeOp>(shape.getDefiningOp());
-  auto shapeTy = op.getType().dyn_cast<fir::ShapeType>();
+  auto shapeTy = mlir::dyn_cast<fir::ShapeType>(op.getType());
   EXPECT_EQ(2u, shapeTy.getRank());
   EXPECT_EQ(2u, op.getExtents().size());
   EXPECT_EQ(c10, op.getExtents()[0]);
@@ -372,7 +374,7 @@ TEST_F(FIRBuilderTest, genShapeWithExtentsAndShapeShift) {
   auto shape = builder.genShape(loc, shifts, extents);
   EXPECT_TRUE(mlir::isa<fir::ShapeShiftOp>(shape.getDefiningOp()));
   fir::ShapeShiftOp op = dyn_cast<fir::ShapeShiftOp>(shape.getDefiningOp());
-  auto shapeTy = op.getType().dyn_cast<fir::ShapeShiftType>();
+  auto shapeTy = mlir::dyn_cast<fir::ShapeShiftType>(op.getType());
   EXPECT_EQ(2u, shapeTy.getRank());
   EXPECT_EQ(2u, op.getExtents().size());
   EXPECT_EQ(2u, op.getOrigins().size());
@@ -428,7 +430,7 @@ TEST_F(FIRBuilderTest, createZeroValue) {
   auto cst =
       mlir::dyn_cast_or_null<mlir::arith::ConstantOp>(zeroInt.getDefiningOp());
   EXPECT_TRUE(cst);
-  auto intAttr = cst.getValue().dyn_cast<mlir::IntegerAttr>();
+  auto intAttr = mlir::dyn_cast<mlir::IntegerAttr>(cst.getValue());
   EXPECT_TRUE(intAttr && intAttr.getInt() == 0);
 
   mlir::Type f32Ty = mlir::FloatType::getF32(builder.getContext());
@@ -437,7 +439,7 @@ TEST_F(FIRBuilderTest, createZeroValue) {
   auto cst2 = mlir::dyn_cast_or_null<mlir::arith::ConstantOp>(
       zeroFloat.getDefiningOp());
   EXPECT_TRUE(cst2);
-  auto floatAttr = cst2.getValue().dyn_cast<mlir::FloatAttr>();
+  auto floatAttr = mlir::dyn_cast<mlir::FloatAttr>(cst2.getValue());
   EXPECT_TRUE(floatAttr && floatAttr.getValueAsDouble() == 0.);
 
   mlir::Type boolTy = mlir::IntegerType::get(builder.getContext(), 1);
@@ -446,7 +448,7 @@ TEST_F(FIRBuilderTest, createZeroValue) {
   auto cst3 = mlir::dyn_cast_or_null<mlir::arith::ConstantOp>(
       flaseBool.getDefiningOp());
   EXPECT_TRUE(cst3);
-  auto intAttr2 = cst.getValue().dyn_cast<mlir::IntegerAttr>();
+  auto intAttr2 = mlir::dyn_cast<mlir::IntegerAttr>(cst.getValue());
   EXPECT_TRUE(intAttr2 && intAttr2.getInt() == 0);
 }
 
@@ -482,7 +484,7 @@ TEST_F(FIRBuilderTest, getBaseTypeOf) {
     llvm::SmallVector<fir::ExtendedValue, 4> arrays;
     auto extent = builder.create<fir::UndefOp>(loc, builder.getIndexType());
     llvm::SmallVector<mlir::Value> extents(
-        arrayType.dyn_cast<fir::SequenceType>().getDimension(),
+        mlir::dyn_cast<fir::SequenceType>(arrayType).getDimension(),
         extent.getResult());
     arrays.emplace_back(fir::ArrayBoxValue(ptrValArray, extents));
     arrays.emplace_back(fir::BoxValue(boxValArray));
diff --git a/flang/unittests/Optimizer/RTBuilder.cpp b/flang/unittests/Optimizer/RTBuilder.cpp
index 7fff7f71fc3b..d6cf96c4351c 100644
--- a/flang/unittests/Optimizer/RTBuilder.cpp
+++ b/flang/unittests/Optimizer/RTBuilder.cpp
@@ -27,7 +27,7 @@ TEST(RTBuilderTest, ComplexRuntimeInterface) {
   mlir::Type c99_cacosf_signature{
       fir::runtime::RuntimeTableKey<decltype(c99_cacosf)>::getTypeModel()(
           &ctx)};
-  auto c99_cacosf_funcTy = c99_cacosf_signature.cast<mlir::FunctionType>();
+  auto c99_cacosf_funcTy = mlir::cast<mlir::FunctionType>(c99_cacosf_signature);
   EXPECT_EQ(c99_cacosf_funcTy.getNumInputs(), 1u);
   EXPECT_EQ(c99_cacosf_funcTy.getNumResults(), 1u);
   auto cplx_ty = fir::ComplexType::get(&ctx, 4);
diff --git a/flang/unittests/Runtime/NumericalFormatTest.cpp b/flang/unittests/Runtime/NumericalFormatTest.cpp
index dee4dda4a228..2a9f8f8d1dc4 100644
--- a/flang/unittests/Runtime/NumericalFormatTest.cpp
+++ b/flang/unittests/Runtime/NumericalFormatTest.cpp
@@ -958,3 +958,20 @@ TEST(IOApiTests, EditDoubleInputValues) {
                            << "', want " << want << ", got " << u.raw;
   }
 }
+
+// regression test for confusing digit minimization
+TEST(IOApiTests, ConfusingMinimization) {
+  char buffer[8]{};
+  auto cookie{IONAME(BeginInternalListOutput)(buffer, sizeof buffer)};
+  StaticDescriptor<0> staticDescriptor;
+  Descriptor &desc{staticDescriptor.descriptor()};
+  std::uint16_t x{0x7bff}; // HUGE(0._2)
+  desc.Establish(TypeCode{CFI_type_half_float}, sizeof x, &x, 0, nullptr);
+  desc.Check();
+  EXPECT_TRUE(IONAME(OutputDescriptor)(cookie, desc));
+  auto status{IONAME(EndIoStatement)(cookie)};
+  EXPECT_EQ(status, 0);
+  std::string got{std::string{buffer, sizeof buffer}};
+  EXPECT_TRUE(CompareFormattedStrings(" 65504. ", got))
+      << "expected ' 65504. ', got '" << got << '\''; // not 65500.!
+}
diff --git a/flang/unittests/Runtime/RuntimeCrashTest.cpp b/flang/unittests/Runtime/RuntimeCrashTest.cpp
index 0f25cc0ee803..a649051fdca0 100644
--- a/flang/unittests/Runtime/RuntimeCrashTest.cpp
+++ b/flang/unittests/Runtime/RuntimeCrashTest.cpp
@@ -53,16 +53,6 @@ TEST(TestTerminator, CheckFailedTest) {
 //------------------------------------------------------------------------------
 struct TestIOCrash : CrashHandlerFixture {};
 
-TEST(TestIOCrash, FormatDescriptorWriteMismatchTest) {
-  static constexpr int bufferSize{4};
-  static char buffer[bufferSize];
-  static const char *format{"(A4)"};
-  auto *cookie{IONAME(BeginInternalFormattedOutput)(
-      buffer, bufferSize, format, std::strlen(format))};
-  ASSERT_DEATH(IONAME(OutputLogical)(cookie, true),
-      "Data edit descriptor 'A' may not be used with a LOGICAL data item");
-}
-
 TEST(TestIOCrash, InvalidFormatCharacterTest) {
   static constexpr int bufferSize{1};
   static char buffer[bufferSize];
diff --git a/libc/config/linux/syscall_numbers.h.inc b/libc/config/linux/syscall_numbers.h.inc
index 9f910c5f9042..4a19d9a08875 100644
--- a/libc/config/linux/syscall_numbers.h.inc
+++ b/libc/config/linux/syscall_numbers.h.inc
@@ -338,6 +338,10 @@
 #define SYS_fchmodat __NR_fchmodat
 #endif
 
+#ifdef __NR_fchmodat2
+#define SYS_fchmodat2 __NR_fchmodat2
+#endif
+
 #ifdef __NR_fchown
 #define SYS_fchown __NR_fchown
 #endif
diff --git a/libc/docs/index.rst b/libc/docs/index.rst
index f71920b058d8..5b96987e0aad 100644
--- a/libc/docs/index.rst
+++ b/libc/docs/index.rst
@@ -72,6 +72,7 @@ stages there is no ABI stability in any form.
    ctype
    signal
    threads
+   setjmp
 
 .. toctree::
    :hidden:
diff --git a/libc/docs/setjmp.rst b/libc/docs/setjmp.rst
new file mode 100644
index 000000000000..d9188dfe1d5e
--- /dev/null
+++ b/libc/docs/setjmp.rst
@@ -0,0 +1,16 @@
+.. include:: check.rst
+
+setjmp.h Functions
+==================
+
+.. list-table::
+  :widths: auto
+  :align: center
+  :header-rows: 1
+
+  * - Function
+    - Implemented
+    - Standard
+  * - longjmp
+    - |check|
+    - 7.13.2.1
diff --git a/libc/include/assert.h.def b/libc/include/assert.h.def
index e006133a7654..15077e53e2ca 100644
--- a/libc/include/assert.h.def
+++ b/libc/include/assert.h.def
@@ -7,6 +7,7 @@
 //===----------------------------------------------------------------------===//
 
 #include "__llvm-libc-common.h"
+#include "llvm-libc-macros/assert-macros.h"
 
 // This file may be usefully included multiple times to change assert()'s
 // definition based on NDEBUG.
diff --git a/libc/include/llvm-libc-macros/CMakeLists.txt b/libc/include/llvm-libc-macros/CMakeLists.txt
index 382cb8ee417e..68ba110aec80 100644
--- a/libc/include/llvm-libc-macros/CMakeLists.txt
+++ b/libc/include/llvm-libc-macros/CMakeLists.txt
@@ -32,6 +32,12 @@ if(EXISTS ${CMAKE_CURRENT_SOURCE_DIR}/${LIBC_TARGET_OS})
 endif()
 
 add_macro_header(
+  assert_macros
+  HDR
+    assert-macros.h
+)
+
+add_macro_header(
   generic_error_number_macros
   HDR
     generic-error-number-macros.h
diff --git a/libc/include/llvm-libc-macros/assert-macros.h b/libc/include/llvm-libc-macros/assert-macros.h
new file mode 100644
index 000000000000..44e14543d856
--- /dev/null
+++ b/libc/include/llvm-libc-macros/assert-macros.h
@@ -0,0 +1,14 @@
+//===-- Definition of macros to be used with assert functions -------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef __LLVM_LIBC_MACROS_ASSERT_MACROS_H
+#define __LLVM_LIBC_MACROS_ASSERT_MACROS_H
+
+#define __STDC_VERSION_ASSERT_H__ 202311L
+
+#endif // __LLVM_LIBC_MACROS_ASSERT_MACROS_H
diff --git a/libc/src/sys/stat/linux/chmod.cpp b/libc/src/sys/stat/linux/chmod.cpp
index 085b91691d89..25e5e69af71a 100644
--- a/libc/src/sys/stat/linux/chmod.cpp
+++ b/libc/src/sys/stat/linux/chmod.cpp
@@ -21,11 +21,14 @@ namespace LIBC_NAMESPACE {
 LLVM_LIBC_FUNCTION(int, chmod, (const char *path, mode_t mode)) {
 #ifdef SYS_chmod
   int ret = LIBC_NAMESPACE::syscall_impl<int>(SYS_chmod, path, mode);
+#elif defined(SYS_fchmodat2)
+  int ret = LIBC_NAMESPACE::syscall_impl<int>(SYS_fchmodat2, AT_FDCWD, path,
+                                              mode, 0, AT_SYMLINK_NOFOLLOW);
 #elif defined(SYS_fchmodat)
   int ret =
-      LIBC_NAMESPACE::syscall_impl<int>(SYS_fchmodat, AT_FDCWD, path, mode);
+      LIBC_NAMESPACE::syscall_impl<int>(SYS_fchmodat, AT_FDCWD, path, mode, 0);
 #else
-#error "chmod and fchmodat syscalls not available."
+#error "chmod, fchmodat and fchmodat2 syscalls not available."
 #endif
 
   if (ret < 0) {
diff --git a/libc/test/UnitTest/CMakeLists.txt b/libc/test/UnitTest/CMakeLists.txt
index 9113eca388e0..302af3044ca3 100644
--- a/libc/test/UnitTest/CMakeLists.txt
+++ b/libc/test/UnitTest/CMakeLists.txt
@@ -111,8 +111,10 @@ add_header_library(
 add_unittest_framework_library(
   LibcFPTestHelpers
   SRCS
+    FEnvSafeTest.cpp
     RoundingModeUtils.cpp
   HDRS
+    FEnvSafeTest.h
     FPMatcher.h
     RoundingModeUtils.h
   DEPENDS
diff --git a/libc/test/UnitTest/FEnvSafeTest.cpp b/libc/test/UnitTest/FEnvSafeTest.cpp
new file mode 100644
index 000000000000..905aa9289373
--- /dev/null
+++ b/libc/test/UnitTest/FEnvSafeTest.cpp
@@ -0,0 +1,84 @@
+//===-- FEnvSafeTest.cpp ---------------------------------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===---------------------------------------------------------------------===//
+
+#include "FEnvSafeTest.h"
+
+#include "src/__support/FPUtil/FEnvImpl.h"
+#include "src/__support/macros/properties/architectures.h"
+
+namespace LIBC_NAMESPACE::testing {
+
+void FEnvSafeTest::PreserveFEnv::check() {
+  fenv_t after;
+  test.get_fenv(after);
+  test.expect_fenv_eq(before, after);
+}
+
+void FEnvSafeTest::TearDown() {
+  if (!should_be_unchanged) {
+    restore_fenv();
+  }
+}
+
+void FEnvSafeTest::get_fenv(fenv_t &fenv) {
+  ASSERT_EQ(LIBC_NAMESPACE::fputil::get_env(&fenv), 0);
+}
+
+void FEnvSafeTest::set_fenv(const fenv_t &fenv) {
+  ASSERT_EQ(LIBC_NAMESPACE::fputil::set_env(&fenv), 0);
+}
+
+void FEnvSafeTest::expect_fenv_eq(const fenv_t &before_fenv,
+                                  const fenv_t &after_fenv) {
+#if defined(LIBC_TARGET_ARCH_IS_AARCH64)
+  using FPState = LIBC_NAMESPACE::fputil::FEnv::FPState;
+  const FPState &before_state = reinterpret_cast<const FPState &>(before_fenv);
+  const FPState &after_state = reinterpret_cast<const FPState &>(after_fenv);
+
+  EXPECT_EQ(before_state.ControlWord, after_state.ControlWord);
+  EXPECT_EQ(before_state.StatusWord, after_state.StatusWord);
+
+#elif defined(LIBC_TARGET_ARCH_IS_X86) && !defined(__APPLE__)
+  using LIBC_NAMESPACE::fputil::internal::FPState;
+  const FPState &before_state = reinterpret_cast<const FPState &>(before_fenv);
+  const FPState &after_state = reinterpret_cast<const FPState &>(after_fenv);
+
+#if defined(_WIN32)
+  EXPECT_EQ(before_state.control_word, after_state.control_word);
+  EXPECT_EQ(before_state.status_word, after_state.status_word);
+#elif defined(__APPLE__)
+  EXPECT_EQ(before_state.control_word, after_state.control_word);
+  EXPECT_EQ(before_state.status_word, after_state.status_word);
+  EXPECT_EQ(before_state.mxcsr, after_state.mxcsr);
+#else
+  EXPECT_EQ(before_state.x87_status.control_word,
+            after_state.x87_status.control_word);
+  EXPECT_EQ(before_state.x87_status.status_word,
+            after_state.x87_status.status_word);
+  EXPECT_EQ(before_state.mxcsr, after_state.mxcsr);
+#endif
+
+#elif defined(LIBC_TARGET_ARCH_IS_ARM) && defined(__ARM_FP)
+  using LIBC_NAMESPACE::fputil::FEnv;
+  const FEnv &before_state = reinterpret_cast<const FEnv &>(before_fenv);
+  const FEnv &after_state = reinterpret_cast<const FEnv &>(after_fenv);
+
+  EXPECT_EQ(before_state.fpscr, after_state.fpscr);
+
+#elif defined(LIBC_TARGET_ARCH_IS_ANY_RISCV)
+  const uint32_t &before_fcsr = reinterpret_cast<const uint32_t &>(before_fenv);
+  const uint32_t &after_fcsr = reinterpret_cast<const uint32_t &>(after_fenv);
+  EXPECT_EQ(before_fcsr, after_fcsr);
+
+#else
+  // No arch-specific `fenv_t` support, so nothing to compare.
+
+#endif
+}
+
+} // namespace LIBC_NAMESPACE::testing
diff --git a/libc/test/UnitTest/FEnvSafeTest.h b/libc/test/UnitTest/FEnvSafeTest.h
new file mode 100644
index 000000000000..d5a8bb7ee667
--- /dev/null
+++ b/libc/test/UnitTest/FEnvSafeTest.h
@@ -0,0 +1,101 @@
+//===-- FEnvSafeTest.h -----------------------------------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===---------------------------------------------------------------------===//
+
+#ifndef LLVM_LIBC_TEST_UNITTEST_FPENVSAFE_H
+#define LLVM_LIBC_TEST_UNITTEST_FPENVSAFE_H
+
+#include "hdr/types/fenv_t.h"
+#include "src/__support/CPP/utility.h"
+#include "test/UnitTest/Test.h"
+
+namespace LIBC_NAMESPACE::testing {
+
+// This provides a test fixture (or base class for other test fixtures) that
+// asserts that each test does not leave the FPU state represented by `fenv_t`
+// (aka `FPState`) perturbed from its initial state.
+class FEnvSafeTest : public Test {
+public:
+  void TearDown() override;
+
+protected:
+  // This is an RAII type where `PreserveFEnv preserve{this};` will sample the
+  // `fenv_t` state and restore it when `preserve` goes out of scope.
+  class PreserveFEnv {
+    fenv_t before;
+    FEnvSafeTest &test;
+
+  public:
+    explicit PreserveFEnv(FEnvSafeTest *self) : test{*self} {
+      test.get_fenv(before);
+    }
+
+    // Cause test expectation failures if the current state doesn't match what
+    // was captured in the constructor.
+    void check();
+
+    // Restore the state captured in the constructor.
+    void restore() { test.set_fenv(before); }
+
+    ~PreserveFEnv() { restore(); }
+  };
+
+  // This is an RAII type where `CheckFEnv check{this};` will sample the
+  // `fenv_t` state and require it be the same when `check` goes out of scope.
+  struct CheckFEnv : public PreserveFEnv {
+    using PreserveFEnv::PreserveFEnv;
+
+    ~CheckFEnv() { check(); }
+  };
+
+  // This calls callable() and returns its value, but has EXPECT_* failures if
+  // the `fenv_t` state is not preserved by the call.
+  template <typename T> decltype(auto) check_fenv_preserved(T &&callable) {
+    CheckFEnv check{this};
+    return cpp::forward<T>(callable)();
+  }
+
+  // This calls callable() and returns its value, but saves and restores the
+  // `fenv_t` state around the call.
+  template <typename T>
+  auto with_fenv_preserved(T &&callable)
+      -> decltype(cpp::forward<decltype(callable)>(callable)()) {
+    PreserveFEnv preserve{this};
+    return cpp::forward<T>(callable)();
+  }
+
+  // A test can call these to indicate it will or won't change `fenv_t` state.
+  void will_change_fenv() { should_be_unchanged = false; }
+  void will_not_change_fenv() { should_be_unchanged = true; }
+
+  // This explicitly resets back to the "before" state captured in SetUp().
+  // TearDown() always does this, but should_be_unchanged controls whether
+  // it also causes test failures if a test fails to restore it.
+  void restore_fenv() { check.restore(); }
+
+private:
+  void get_fenv(fenv_t &fenv);
+  void set_fenv(const fenv_t &fenv);
+  void expect_fenv_eq(const fenv_t &before_fenv, const fenv_t &after_fenv);
+
+  CheckFEnv check{this};
+
+  // TODO: Many tests fail if this is true. It needs to be figured out whether
+  // the state should be preserved by each library function under test, and
+  // separately whether each test itself should preserve the state.  It
+  // probably isn't important that tests be explicitly written to preserve the
+  // state, as the fixture can (and does) reset it--the next test can rely on
+  // getting "normal" ambient state initially.  For library functions that
+  // should preserve the state, that should be checked after each call, not
+  // just after the whole test.  So they can use check_fenv_preserved or
+  // with_fenv_preserved as appropriate.
+  bool should_be_unchanged = false;
+};
+
+} // namespace LIBC_NAMESPACE::testing
+
+#endif // LLVM_LIBC_TEST_UNITTEST_FPENVSAFE_H
diff --git a/libc/test/include/CMakeLists.txt b/libc/test/include/CMakeLists.txt
index 8d8dff53169f..03c31855e352 100644
--- a/libc/test/include/CMakeLists.txt
+++ b/libc/test/include/CMakeLists.txt
@@ -1,6 +1,16 @@
 add_custom_target(libc_include_tests)
 
 add_libc_test(
+  assert_test
+  SUITE
+    libc_include_tests
+  SRCS
+    assert_test.cpp
+  DEPENDS
+    libc.include.llvm-libc-macros.assert_macros
+)
+
+add_libc_test(
   sys_queue_test
   SUITE
     libc_include_tests
diff --git a/libc/test/include/assert_test.cpp b/libc/test/include/assert_test.cpp
new file mode 100644
index 000000000000..78709bbcdd59
--- /dev/null
+++ b/libc/test/include/assert_test.cpp
@@ -0,0 +1,15 @@
+//===-- Unittests for assert ----------------------------------------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDSList-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#include "include/llvm-libc-macros/assert-macros.h"
+#include "test/UnitTest/Test.h"
+
+TEST(LlvmLibcAssertTest, VersionMacro) {
+  // 7.2p3 an integer constant expression with a value equivalent to 202311L.
+  EXPECT_EQ(__STDC_VERSION_ASSERT_H__, 202311L);
+}
diff --git a/libc/test/src/fenv/CMakeLists.txt b/libc/test/src/fenv/CMakeLists.txt
index f277b65e2d42..b776f9a0706e 100644
--- a/libc/test/src/fenv/CMakeLists.txt
+++ b/libc/test/src/fenv/CMakeLists.txt
@@ -9,6 +9,8 @@ add_libc_unittest(
   DEPENDS
     libc.src.fenv.fegetround
     libc.src.fenv.fesetround
+  LINK_LIBRARIES
+    LibcFPTestHelpers
 )
 
 add_libc_unittest(
@@ -23,6 +25,8 @@ add_libc_unittest(
     libc.src.fenv.fesetexcept
     libc.src.fenv.fetestexcept
     libc.src.__support.FPUtil.fenv_impl
+  LINK_LIBRARIES
+    LibcFPTestHelpers
 )
 
 add_libc_unittest(
@@ -37,6 +41,8 @@ add_libc_unittest(
     libc.src.fenv.fesetenv
     libc.src.fenv.fesetround
     libc.src.__support.FPUtil.fenv_impl
+  LINK_LIBRARIES
+    LibcFPTestHelpers
 )
 
 add_libc_unittest(
@@ -50,6 +56,8 @@ add_libc_unittest(
     libc.src.fenv.fesetexceptflag
     libc.src.fenv.fetestexceptflag
     libc.src.__support.FPUtil.fenv_impl
+  LINK_LIBRARIES
+    LibcFPTestHelpers
 )
 
 add_libc_unittest(
@@ -62,6 +70,8 @@ add_libc_unittest(
     libc.include.signal
     libc.src.fenv.feupdateenv
     libc.src.__support.FPUtil.fenv_impl
+  LINK_LIBRARIES
+    LibcFPTestHelpers
 )
 
 add_libc_unittest(
@@ -73,6 +83,8 @@ add_libc_unittest(
   DEPENDS
     libc.src.fenv.feclearexcept
     libc.src.__support.FPUtil.fenv_impl
+  LINK_LIBRARIES
+    LibcFPTestHelpers
 )
 
 add_libc_unittest(
@@ -85,6 +97,8 @@ add_libc_unittest(
     libc.src.fenv.fedisableexcept
     libc.src.fenv.feenableexcept
     libc.src.fenv.fegetexcept
+  LINK_LIBRARIES
+    LibcFPTestHelpers
 )
 
 if (NOT (LLVM_USE_SANITIZER OR (${LIBC_TARGET_OS} STREQUAL "windows")
@@ -109,6 +123,7 @@ if (NOT (LLVM_USE_SANITIZER OR (${LIBC_TARGET_OS} STREQUAL "windows")
       libc.src.__support.FPUtil.fenv_impl
     LINK_LIBRARIES
       LibcFPExceptionHelpers
+      LibcFPTestHelpers
   )
 
   add_fp_unittest(
@@ -124,5 +139,6 @@ if (NOT (LLVM_USE_SANITIZER OR (${LIBC_TARGET_OS} STREQUAL "windows")
       libc.src.__support.FPUtil.fenv_impl
     LINK_LIBRARIES
       LibcFPExceptionHelpers
+      LibcFPTestHelpers
   )
 endif()
diff --git a/libc/test/src/fenv/enabled_exceptions_test.cpp b/libc/test/src/fenv/enabled_exceptions_test.cpp
index 53440b704ca7..7d26eab5695b 100644
--- a/libc/test/src/fenv/enabled_exceptions_test.cpp
+++ b/libc/test/src/fenv/enabled_exceptions_test.cpp
@@ -12,15 +12,20 @@
 
 #include "src/__support/FPUtil/FEnvImpl.h"
 #include "src/__support/macros/properties/architectures.h"
+#include "test/UnitTest/FEnvSafeTest.h"
 #include "test/UnitTest/FPExceptMatcher.h"
 #include "test/UnitTest/Test.h"
 
 #include "hdr/fenv_macros.h"
 #include <signal.h>
 
+#include "excepts.h"
+
+using LlvmLibcExceptionStatusTest = LIBC_NAMESPACE::testing::FEnvSafeTest;
+
 // This test enables an exception and verifies that raising that exception
 // triggers SIGFPE.
-TEST(LlvmLibcExceptionStatusTest, RaiseAndCrash) {
+TEST_F(LlvmLibcExceptionStatusTest, RaiseAndCrash) {
 #if defined(LIBC_TARGET_ARCH_IS_ANY_ARM) ||                                    \
     defined(LIBC_TARGET_ARCH_IS_ANY_RISCV)
   // Few Arm HW implementations do not trap exceptions. We skip this test
@@ -41,16 +46,7 @@ TEST(LlvmLibcExceptionStatusTest, RaiseAndCrash) {
   // that exception handler, so such a testing can be done after we have
   // longjmp implemented.
 
-  int excepts[] = {FE_DIVBYZERO, FE_INVALID, FE_INEXACT, FE_OVERFLOW,
-                   FE_UNDERFLOW};
-
-  // We '|' the individual exception flags instead of using FE_ALL_EXCEPT
-  // as it can include non-standard extensions. Note that we should be able
-  // to compile this file with headers from other libcs as well.
-  constexpr int ALL_EXCEPTS =
-      FE_DIVBYZERO | FE_INVALID | FE_INEXACT | FE_OVERFLOW | FE_UNDERFLOW;
-
-  for (int e : excepts) {
+  for (int e : EXCEPTS) {
     LIBC_NAMESPACE::fputil::disable_except(FE_ALL_EXCEPT);
     LIBC_NAMESPACE::fputil::enable_except(e);
     ASSERT_EQ(LIBC_NAMESPACE::feclearexcept(FE_ALL_EXCEPT), 0);
diff --git a/libc/test/src/fenv/exception_flags_test.cpp b/libc/test/src/fenv/exception_flags_test.cpp
index 9d2be6426a6d..2f4332df861f 100644
--- a/libc/test/src/fenv/exception_flags_test.cpp
+++ b/libc/test/src/fenv/exception_flags_test.cpp
@@ -12,18 +12,20 @@
 #include "src/fenv/fetestexceptflag.h"
 
 #include "src/__support/FPUtil/FEnvImpl.h"
+#include "test/UnitTest/FEnvSafeTest.h"
 #include "test/UnitTest/Test.h"
 
-TEST(LlvmLibcFenvTest, GetSetTestExceptFlag) {
+#include "excepts.h"
+
+using LlvmLibcFEnvTest = LIBC_NAMESPACE::testing::FEnvSafeTest;
+
+TEST_F(LlvmLibcFEnvTest, GetSetTestExceptFlag) {
   // We will disable all exceptions to prevent invocation of the exception
   // handler.
   LIBC_NAMESPACE::fputil::disable_except(FE_ALL_EXCEPT);
   LIBC_NAMESPACE::fputil::clear_except(FE_ALL_EXCEPT);
 
-  int excepts[] = {FE_DIVBYZERO, FE_INVALID, FE_INEXACT, FE_OVERFLOW,
-                   FE_UNDERFLOW};
-
-  for (int e : excepts) {
+  for (int e : EXCEPTS) {
     // The overall idea is to raise an except and save the exception flags.
     // Next, clear the flags and then set the saved exception flags. This
     // should set the flag corresponding to the previously raised exception.
diff --git a/libc/test/src/fenv/exception_status_test.cpp b/libc/test/src/fenv/exception_status_test.cpp
index a7000020b1a3..fdf942145786 100644
--- a/libc/test/src/fenv/exception_status_test.cpp
+++ b/libc/test/src/fenv/exception_status_test.cpp
@@ -13,24 +13,23 @@
 #include "src/fenv/fetestexcept.h"
 
 #include "src/__support/FPUtil/FEnvImpl.h"
+#include "test/UnitTest/FEnvSafeTest.h"
 #include "test/UnitTest/Test.h"
 
 #include "hdr/fenv_macros.h"
 
-TEST(LlvmLibcExceptionStatusTest, RaiseAndTest) {
+#include "excepts.h"
+
+using LlvmLibcExceptionStatusTest = LIBC_NAMESPACE::testing::FEnvSafeTest;
+
+TEST_F(LlvmLibcExceptionStatusTest, RaiseAndTest) {
   // This test raises a set of exceptions and checks that the exception
   // status flags are updated. The intention is really not to invoke the
   // exception handler. Hence, we will disable all exceptions at the
   // beginning.
   LIBC_NAMESPACE::fputil::disable_except(FE_ALL_EXCEPT);
 
-  int excepts[] = {FE_DIVBYZERO, FE_INVALID, FE_INEXACT, FE_OVERFLOW,
-                   FE_UNDERFLOW};
-
-  constexpr int ALL_EXCEPTS =
-      FE_DIVBYZERO | FE_INVALID | FE_INEXACT | FE_OVERFLOW | FE_UNDERFLOW;
-
-  for (int e : excepts) {
+  for (int e : EXCEPTS) {
     int r = LIBC_NAMESPACE::feraiseexcept(e);
     ASSERT_EQ(r, 0);
     int s = LIBC_NAMESPACE::fetestexcept(e);
@@ -47,8 +46,8 @@ TEST(LlvmLibcExceptionStatusTest, RaiseAndTest) {
     ASSERT_EQ(s, e);
   }
 
-  for (int e1 : excepts) {
-    for (int e2 : excepts) {
+  for (int e1 : EXCEPTS) {
+    for (int e2 : EXCEPTS) {
       int e = e1 | e2;
       int r = LIBC_NAMESPACE::feraiseexcept(e);
       ASSERT_EQ(r, 0);
@@ -67,9 +66,9 @@ TEST(LlvmLibcExceptionStatusTest, RaiseAndTest) {
     }
   }
 
-  for (int e1 : excepts) {
-    for (int e2 : excepts) {
-      for (int e3 : excepts) {
+  for (int e1 : EXCEPTS) {
+    for (int e2 : EXCEPTS) {
+      for (int e3 : EXCEPTS) {
         int e = e1 | e2 | e3;
         int r = LIBC_NAMESPACE::feraiseexcept(e);
         ASSERT_EQ(r, 0);
@@ -89,10 +88,10 @@ TEST(LlvmLibcExceptionStatusTest, RaiseAndTest) {
     }
   }
 
-  for (int e1 : excepts) {
-    for (int e2 : excepts) {
-      for (int e3 : excepts) {
-        for (int e4 : excepts) {
+  for (int e1 : EXCEPTS) {
+    for (int e2 : EXCEPTS) {
+      for (int e3 : EXCEPTS) {
+        for (int e4 : EXCEPTS) {
           int e = e1 | e2 | e3 | e4;
           int r = LIBC_NAMESPACE::feraiseexcept(e);
           ASSERT_EQ(r, 0);
@@ -113,11 +112,11 @@ TEST(LlvmLibcExceptionStatusTest, RaiseAndTest) {
     }
   }
 
-  for (int e1 : excepts) {
-    for (int e2 : excepts) {
-      for (int e3 : excepts) {
-        for (int e4 : excepts) {
-          for (int e5 : excepts) {
+  for (int e1 : EXCEPTS) {
+    for (int e2 : EXCEPTS) {
+      for (int e3 : EXCEPTS) {
+        for (int e4 : EXCEPTS) {
+          for (int e5 : EXCEPTS) {
             int e = e1 | e2 | e3 | e4 | e5;
             int r = LIBC_NAMESPACE::feraiseexcept(e);
             ASSERT_EQ(r, 0);
diff --git a/libc/test/src/fenv/excepts.h b/libc/test/src/fenv/excepts.h
new file mode 100644
index 000000000000..e9517d319a9b
--- /dev/null
+++ b/libc/test/src/fenv/excepts.h
@@ -0,0 +1,24 @@
+//===-- List of all FE_* constants for tests -----------------------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===---------------------------------------------------------------------===//
+
+#ifndef LLVM_LIBC_TEST_SRC_FENV_EXCEPTS_H
+#define LLVM_LIBC_TEST_SRC_FENV_EXCEPTS_H
+
+#include "hdr/fenv_macros.h"
+
+constexpr int EXCEPTS[] = {
+    FE_DIVBYZERO, FE_INVALID, FE_INEXACT, FE_OVERFLOW, FE_UNDERFLOW,
+};
+
+// We '|' the individual exception flags instead of using FE_ALL_EXCEPT
+// as it can include non-standard extensions. Note that we should be able
+// to compile this file with headers from other libcs as well.
+constexpr int ALL_EXCEPTS =
+    FE_DIVBYZERO | FE_INVALID | FE_INEXACT | FE_OVERFLOW | FE_UNDERFLOW;
+
+#endif // LLVM_LIBC_TEST_SRC_FENV_EXCEPTS_H
diff --git a/libc/test/src/fenv/feclearexcept_test.cpp b/libc/test/src/fenv/feclearexcept_test.cpp
index bb42d9070358..52adda46adf2 100644
--- a/libc/test/src/fenv/feclearexcept_test.cpp
+++ b/libc/test/src/fenv/feclearexcept_test.cpp
@@ -9,27 +9,30 @@
 #include "src/fenv/feclearexcept.h"
 
 #include "src/__support/FPUtil/FEnvImpl.h"
+#include "test/UnitTest/FEnvSafeTest.h"
 #include "test/UnitTest/Test.h"
 
 #include "hdr/fenv_macros.h"
 #include <stdint.h>
 
-TEST(LlvmLibcFEnvTest, ClearTest) {
-  uint16_t excepts[] = {FE_DIVBYZERO, FE_INVALID, FE_INEXACT, FE_OVERFLOW,
-                        FE_UNDERFLOW};
+#include "excepts.h"
+
+using LlvmLibcFEnvTest = LIBC_NAMESPACE::testing::FEnvSafeTest;
+
+TEST_F(LlvmLibcFEnvTest, ClearTest) {
   LIBC_NAMESPACE::fputil::disable_except(FE_ALL_EXCEPT);
   LIBC_NAMESPACE::fputil::clear_except(FE_ALL_EXCEPT);
 
-  for (uint16_t e : excepts)
+  for (int e : EXCEPTS)
     ASSERT_EQ(LIBC_NAMESPACE::fputil::test_except(e), 0);
 
   LIBC_NAMESPACE::fputil::raise_except(FE_ALL_EXCEPT);
 
-  for (uint16_t e1 : excepts) {
-    for (uint16_t e2 : excepts) {
-      for (uint16_t e3 : excepts) {
-        for (uint16_t e4 : excepts) {
-          for (uint16_t e5 : excepts) {
+  for (int e1 : EXCEPTS) {
+    for (int e2 : EXCEPTS) {
+      for (int e3 : EXCEPTS) {
+        for (int e4 : EXCEPTS) {
+          for (int e5 : EXCEPTS) {
             // We clear one exception and test to verify that it was cleared.
             LIBC_NAMESPACE::feclearexcept(e1 | e2 | e3 | e4 | e5);
             ASSERT_EQ(
diff --git a/libc/test/src/fenv/feenableexcept_test.cpp b/libc/test/src/fenv/feenableexcept_test.cpp
index aeb4f955fd69..232e2a1c8316 100644
--- a/libc/test/src/fenv/feenableexcept_test.cpp
+++ b/libc/test/src/fenv/feenableexcept_test.cpp
@@ -11,11 +11,16 @@
 #include "src/fenv/feenableexcept.h"
 #include "src/fenv/fegetexcept.h"
 
+#include "test/UnitTest/FEnvSafeTest.h"
 #include "test/UnitTest/Test.h"
 
 #include "hdr/fenv_macros.h"
 
-TEST(LlvmLibcFEnvTest, EnableTest) {
+#include "excepts.h"
+
+using LlvmLibcFEnvTest = LIBC_NAMESPACE::testing::FEnvSafeTest;
+
+TEST_F(LlvmLibcFEnvTest, EnableTest) {
 #if defined(LIBC_TARGET_ARCH_IS_ANY_ARM) ||                                    \
     defined(LIBC_TARGET_ARCH_IS_ANY_RISCV)
   // Few Arm HW implementations do not trap exceptions. We skip this test
diff --git a/libc/test/src/fenv/feholdexcept_test.cpp b/libc/test/src/fenv/feholdexcept_test.cpp
index 0689d89ab233..f3e05d4a5b6c 100644
--- a/libc/test/src/fenv/feholdexcept_test.cpp
+++ b/libc/test/src/fenv/feholdexcept_test.cpp
@@ -11,10 +11,15 @@
 
 #include "src/__support/FPUtil/FEnvImpl.h"
 #include "src/__support/macros/properties/architectures.h"
+#include "test/UnitTest/FEnvSafeTest.h"
 #include "test/UnitTest/FPExceptMatcher.h"
 #include "test/UnitTest/Test.h"
 
-TEST(LlvmLibcFEnvTest, RaiseAndCrash) {
+#include "excepts.h"
+
+using LlvmLibcFEnvTest = LIBC_NAMESPACE::testing::FEnvSafeTest;
+
+TEST_F(LlvmLibcFEnvTest, RaiseAndCrash) {
 #if defined(LIBC_TARGET_ARCH_IS_ANY_ARM) ||                                    \
     defined(LIBC_TARGET_ARCH_IS_ANY_RISCV)
   // Few Arm HW implementations do not trap exceptions. We skip this test
diff --git a/libc/test/src/fenv/feupdateenv_test.cpp b/libc/test/src/fenv/feupdateenv_test.cpp
index 251b8566aac3..d2ffc0ef8e84 100644
--- a/libc/test/src/fenv/feupdateenv_test.cpp
+++ b/libc/test/src/fenv/feupdateenv_test.cpp
@@ -10,11 +10,12 @@
 #include "src/fenv/feupdateenv.h"
 
 #include "src/__support/FPUtil/FEnvImpl.h"
+#include "test/UnitTest/FEnvSafeTest.h"
 #include "test/UnitTest/Test.h"
 
-#include <signal.h>
+using LlvmLibcFEnvTest = LIBC_NAMESPACE::testing::FEnvSafeTest;
 
-TEST(LlvmLibcFEnvTest, UpdateEnvTest) {
+TEST_F(LlvmLibcFEnvTest, UpdateEnvTest) {
   LIBC_NAMESPACE::fputil::disable_except(FE_ALL_EXCEPT);
   LIBC_NAMESPACE::fputil::clear_except(FE_ALL_EXCEPT);
 
diff --git a/libc/test/src/fenv/getenv_and_setenv_test.cpp b/libc/test/src/fenv/getenv_and_setenv_test.cpp
index f767e8ab9b2f..7257e75cb421 100644
--- a/libc/test/src/fenv/getenv_and_setenv_test.cpp
+++ b/libc/test/src/fenv/getenv_and_setenv_test.cpp
@@ -13,17 +13,19 @@
 #include "src/fenv/fesetround.h"
 
 #include "src/__support/FPUtil/FEnvImpl.h"
+#include "test/UnitTest/FEnvSafeTest.h"
 #include "test/UnitTest/Test.h"
 
-TEST(LlvmLibcFenvTest, GetEnvAndSetEnv) {
+#include "excepts.h"
+
+using LlvmLibcFEnvTest = LIBC_NAMESPACE::testing::FEnvSafeTest;
+
+TEST_F(LlvmLibcFEnvTest, GetEnvAndSetEnv) {
   // We will disable all exceptions to prevent invocation of the exception
   // handler.
   LIBC_NAMESPACE::fputil::disable_except(FE_ALL_EXCEPT);
 
-  int excepts[] = {FE_DIVBYZERO, FE_INVALID, FE_INEXACT, FE_OVERFLOW,
-                   FE_UNDERFLOW};
-
-  for (int e : excepts) {
+  for (int e : EXCEPTS) {
     LIBC_NAMESPACE::fputil::clear_except(FE_ALL_EXCEPT);
 
     // Save the cleared environment.
@@ -71,7 +73,7 @@ TEST(LlvmLibcFenvTest, Set_FE_DFL_ENV) {
 }
 
 #ifdef _WIN32
-TEST(LlvmLibcFenvTest, Windows_Set_Get_Test) {
+TEST_F(LlvmLibcFEnvTest, Windows_Set_Get_Test) {
   // If a valid fenv_t is written, then reading it back out should be identical.
   fenv_t setEnv = {0x7e00053e, 0x0f00000f};
   fenv_t getEnv;
diff --git a/libc/test/src/fenv/rounding_mode_test.cpp b/libc/test/src/fenv/rounding_mode_test.cpp
index ec2e27ecc818..f242ed9aaffe 100644
--- a/libc/test/src/fenv/rounding_mode_test.cpp
+++ b/libc/test/src/fenv/rounding_mode_test.cpp
@@ -9,15 +9,18 @@
 #include "src/fenv/fegetround.h"
 #include "src/fenv/fesetround.h"
 
+#include "test/UnitTest/FEnvSafeTest.h"
 #include "test/UnitTest/Test.h"
 
 #include "hdr/fenv_macros.h"
 
-TEST(LlvmLibcRoundingModeTest, SetAndGet) {
+using LlvmLibcRoundingModeTest = LIBC_NAMESPACE::testing::FEnvSafeTest;
+
+TEST_F(LlvmLibcRoundingModeTest, SetAndGet) {
   struct ResetDefaultRoundingMode {
-    int original;
+    int original = LIBC_NAMESPACE::fegetround();
     ~ResetDefaultRoundingMode() { LIBC_NAMESPACE::fesetround(original); }
-  } reset{LIBC_NAMESPACE::fegetround()};
+  } reset;
 
   int s = LIBC_NAMESPACE::fesetround(FE_TONEAREST);
   EXPECT_EQ(s, 0);
diff --git a/libc/test/src/math/CeilTest.h b/libc/test/src/math/CeilTest.h
index da3f3c0e8f5a..b4c3752cc5c4 100644
--- a/libc/test/src/math/CeilTest.h
+++ b/libc/test/src/math/CeilTest.h
@@ -6,6 +6,7 @@
 //
 //===----------------------------------------------------------------------===//
 
+#include "test/UnitTest/FEnvSafeTest.h"
 #include "test/UnitTest/FPMatcher.h"
 #include "test/UnitTest/Test.h"
 #include "utils/MPFRWrapper/MPFRUtils.h"
@@ -14,7 +15,8 @@
 
 namespace mpfr = LIBC_NAMESPACE::testing::mpfr;
 
-template <typename T> class CeilTest : public LIBC_NAMESPACE::testing::Test {
+template <typename T>
+class CeilTest : public LIBC_NAMESPACE::testing::FEnvSafeTest {
 
   DECLARE_SPECIAL_CONSTANTS(T)
 
diff --git a/libc/test/src/math/CopySignTest.h b/libc/test/src/math/CopySignTest.h
index 052ff0333438..c66f91477480 100644
--- a/libc/test/src/math/CopySignTest.h
+++ b/libc/test/src/math/CopySignTest.h
@@ -6,6 +6,7 @@
 //
 //===----------------------------------------------------------------------===//
 
+#include "test/UnitTest/FEnvSafeTest.h"
 #include "test/UnitTest/FPMatcher.h"
 #include "test/UnitTest/Test.h"
 #include "utils/MPFRWrapper/MPFRUtils.h"
@@ -15,7 +16,7 @@
 namespace mpfr = LIBC_NAMESPACE::testing::mpfr;
 
 template <typename T>
-class CopySignTest : public LIBC_NAMESPACE::testing::Test {
+class CopySignTest : public LIBC_NAMESPACE::testing::FEnvSafeTest {
 
   DECLARE_SPECIAL_CONSTANTS(T)
 
diff --git a/libc/test/src/math/FAbsTest.h b/libc/test/src/math/FAbsTest.h
index 23ad8a26c481..92b589beeb67 100644
--- a/libc/test/src/math/FAbsTest.h
+++ b/libc/test/src/math/FAbsTest.h
@@ -9,6 +9,7 @@
 #ifndef LLVM_LIBC_TEST_SRC_MATH_FABSTEST_H
 #define LLVM_LIBC_TEST_SRC_MATH_FABSTEST_H
 
+#include "test/UnitTest/FEnvSafeTest.h"
 #include "test/UnitTest/FPMatcher.h"
 #include "test/UnitTest/Test.h"
 #include "utils/MPFRWrapper/MPFRUtils.h"
@@ -17,7 +18,8 @@
 
 namespace mpfr = LIBC_NAMESPACE::testing::mpfr;
 
-template <typename T> class FAbsTest : public LIBC_NAMESPACE::testing::Test {
+template <typename T>
+class FAbsTest : public LIBC_NAMESPACE::testing::FEnvSafeTest {
 
   DECLARE_SPECIAL_CONSTANTS(T)
 
diff --git a/libc/test/src/math/FDimTest.h b/libc/test/src/math/FDimTest.h
index 44aba9caf646..fefcefe5052a 100644
--- a/libc/test/src/math/FDimTest.h
+++ b/libc/test/src/math/FDimTest.h
@@ -9,11 +9,12 @@
 #include "hdr/math_macros.h"
 #include "src/__support/FPUtil/BasicOperations.h"
 #include "src/__support/FPUtil/FPBits.h"
+#include "test/UnitTest/FEnvSafeTest.h"
 #include "test/UnitTest/FPMatcher.h"
 #include "test/UnitTest/Test.h"
 
 template <typename T>
-class FDimTestTemplate : public LIBC_NAMESPACE::testing::Test {
+class FDimTestTemplate : public LIBC_NAMESPACE::testing::FEnvSafeTest {
 public:
   using FuncPtr = T (*)(T, T);
   using FPBits = LIBC_NAMESPACE::fputil::FPBits<T>;
diff --git a/libc/test/src/math/FMaxTest.h b/libc/test/src/math/FMaxTest.h
index e9857f332e65..405642c6b968 100644
--- a/libc/test/src/math/FMaxTest.h
+++ b/libc/test/src/math/FMaxTest.h
@@ -9,6 +9,7 @@
 #ifndef LLVM_LIBC_TEST_SRC_MATH_FMAXTEST_H
 #define LLVM_LIBC_TEST_SRC_MATH_FMAXTEST_H
 
+#include "test/UnitTest/FEnvSafeTest.h"
 #include "test/UnitTest/FPMatcher.h"
 #include "test/UnitTest/Test.h"
 #include "utils/MPFRWrapper/MPFRUtils.h"
@@ -17,7 +18,8 @@
 
 namespace mpfr = LIBC_NAMESPACE::testing::mpfr;
 
-template <typename T> class FMaxTest : public LIBC_NAMESPACE::testing::Test {
+template <typename T>
+class FMaxTest : public LIBC_NAMESPACE::testing::FEnvSafeTest {
 
   DECLARE_SPECIAL_CONSTANTS(T)
 
diff --git a/libc/test/src/math/FMinTest.h b/libc/test/src/math/FMinTest.h
index c6b9f4439b79..eae0008ddfe3 100644
--- a/libc/test/src/math/FMinTest.h
+++ b/libc/test/src/math/FMinTest.h
@@ -9,6 +9,7 @@
 #ifndef LLVM_LIBC_TEST_SRC_MATH_FMINTEST_H
 #define LLVM_LIBC_TEST_SRC_MATH_FMINTEST_H
 
+#include "test/UnitTest/FEnvSafeTest.h"
 #include "test/UnitTest/FPMatcher.h"
 #include "test/UnitTest/Test.h"
 #include "utils/MPFRWrapper/MPFRUtils.h"
@@ -17,7 +18,8 @@
 
 namespace mpfr = LIBC_NAMESPACE::testing::mpfr;
 
-template <typename T> class FMinTest : public LIBC_NAMESPACE::testing::Test {
+template <typename T>
+class FMinTest : public LIBC_NAMESPACE::testing::FEnvSafeTest {
 
   DECLARE_SPECIAL_CONSTANTS(T)
 
diff --git a/libc/test/src/math/FModTest.h b/libc/test/src/math/FModTest.h
index bc909987a161..f1015d6497fc 100644
--- a/libc/test/src/math/FModTest.h
+++ b/libc/test/src/math/FModTest.h
@@ -11,6 +11,7 @@
 
 #include "src/__support/FPUtil/BasicOperations.h"
 #include "src/__support/FPUtil/NearestIntegerOperations.h"
+#include "test/UnitTest/FEnvSafeTest.h"
 #include "test/UnitTest/FPMatcher.h"
 #include "test/UnitTest/Test.h"
 
@@ -24,7 +25,8 @@
 
 #define TEST_REGULAR(x, y, expected) TEST_SPECIAL(x, y, expected, false, 0)
 
-template <typename T> class FmodTest : public LIBC_NAMESPACE::testing::Test {
+template <typename T>
+class FmodTest : public LIBC_NAMESPACE::testing::FEnvSafeTest {
 
   DECLARE_SPECIAL_CONSTANTS(T)
 
diff --git a/libc/test/src/math/FloorTest.h b/libc/test/src/math/FloorTest.h
index 679dc26e1248..9103a5b05eb5 100644
--- a/libc/test/src/math/FloorTest.h
+++ b/libc/test/src/math/FloorTest.h
@@ -9,6 +9,7 @@
 #ifndef LLVM_LIBC_TEST_SRC_MATH_FLOORTEST_H
 #define LLVM_LIBC_TEST_SRC_MATH_FLOORTEST_H
 
+#include "test/UnitTest/FEnvSafeTest.h"
 #include "test/UnitTest/FPMatcher.h"
 #include "test/UnitTest/Test.h"
 #include "utils/MPFRWrapper/MPFRUtils.h"
@@ -17,7 +18,8 @@
 
 namespace mpfr = LIBC_NAMESPACE::testing::mpfr;
 
-template <typename T> class FloorTest : public LIBC_NAMESPACE::testing::Test {
+template <typename T>
+class FloorTest : public LIBC_NAMESPACE::testing::FEnvSafeTest {
 
   DECLARE_SPECIAL_CONSTANTS(T)
 
diff --git a/libc/test/src/math/FmaTest.h b/libc/test/src/math/FmaTest.h
index 76bd221fcb1f..5a40f694ebd1 100644
--- a/libc/test/src/math/FmaTest.h
+++ b/libc/test/src/math/FmaTest.h
@@ -12,6 +12,7 @@
 #include "src/__support/FPUtil/FPBits.h"
 #include "src/stdlib/rand.h"
 #include "src/stdlib/srand.h"
+#include "test/UnitTest/FEnvSafeTest.h"
 #include "test/UnitTest/FPMatcher.h"
 #include "test/UnitTest/Test.h"
 #include "utils/MPFRWrapper/MPFRUtils.h"
@@ -19,7 +20,7 @@
 namespace mpfr = LIBC_NAMESPACE::testing::mpfr;
 
 template <typename T>
-class FmaTestTemplate : public LIBC_NAMESPACE::testing::Test {
+class FmaTestTemplate : public LIBC_NAMESPACE::testing::FEnvSafeTest {
 private:
   using Func = T (*)(T, T, T);
   using FPBits = LIBC_NAMESPACE::fputil::FPBits<T>;
diff --git a/libc/test/src/math/FrexpTest.h b/libc/test/src/math/FrexpTest.h
index 5f993f604999..3ba64afa3c62 100644
--- a/libc/test/src/math/FrexpTest.h
+++ b/libc/test/src/math/FrexpTest.h
@@ -7,6 +7,7 @@
 //===----------------------------------------------------------------------===//
 
 #include "src/__support/FPUtil/BasicOperations.h"
+#include "test/UnitTest/FEnvSafeTest.h"
 #include "test/UnitTest/FPMatcher.h"
 #include "test/UnitTest/Test.h"
 #include "utils/MPFRWrapper/MPFRUtils.h"
@@ -15,7 +16,8 @@
 
 namespace mpfr = LIBC_NAMESPACE::testing::mpfr;
 
-template <typename T> class FrexpTest : public LIBC_NAMESPACE::testing::Test {
+template <typename T>
+class FrexpTest : public LIBC_NAMESPACE::testing::FEnvSafeTest {
 
   DECLARE_SPECIAL_CONSTANTS(T)
 
diff --git a/libc/test/src/math/HypotTest.h b/libc/test/src/math/HypotTest.h
index 0c15f02fe371..58b533831824 100644
--- a/libc/test/src/math/HypotTest.h
+++ b/libc/test/src/math/HypotTest.h
@@ -10,6 +10,7 @@
 #define LLVM_LIBC_TEST_SRC_MATH_HYPOTTEST_H
 
 #include "src/__support/FPUtil/FPBits.h"
+#include "test/UnitTest/FEnvSafeTest.h"
 #include "test/UnitTest/FPMatcher.h"
 #include "test/UnitTest/Test.h"
 #include "utils/MPFRWrapper/MPFRUtils.h"
@@ -19,7 +20,7 @@
 namespace mpfr = LIBC_NAMESPACE::testing::mpfr;
 
 template <typename T>
-class HypotTestTemplate : public LIBC_NAMESPACE::testing::Test {
+class HypotTestTemplate : public LIBC_NAMESPACE::testing::FEnvSafeTest {
 private:
   using Func = T (*)(T, T);
   using FPBits = LIBC_NAMESPACE::fputil::FPBits<T>;
diff --git a/libc/test/src/math/ILogbTest.h b/libc/test/src/math/ILogbTest.h
index 3d1f047a4806..c2d5a1326e0e 100644
--- a/libc/test/src/math/ILogbTest.h
+++ b/libc/test/src/math/ILogbTest.h
@@ -13,9 +13,10 @@
 #include "src/__support/CPP/limits.h" // INT_MAX
 #include "src/__support/FPUtil/FPBits.h"
 #include "src/__support/FPUtil/ManipulationFunctions.h"
+#include "test/UnitTest/FEnvSafeTest.h"
 #include "test/UnitTest/Test.h"
 
-class LlvmLibcILogbTest : public LIBC_NAMESPACE::testing::Test {
+class LlvmLibcILogbTest : public LIBC_NAMESPACE::testing::FEnvSafeTest {
 public:
   template <typename T> struct ILogbFunc {
     typedef int (*Func)(T);
diff --git a/libc/test/src/math/LdExpTest.h b/libc/test/src/math/LdExpTest.h
index 2a406feed52f..34466a526d60 100644
--- a/libc/test/src/math/LdExpTest.h
+++ b/libc/test/src/math/LdExpTest.h
@@ -12,6 +12,7 @@
 #include "src/__support/CPP/limits.h" // INT_MAX
 #include "src/__support/FPUtil/FPBits.h"
 #include "src/__support/FPUtil/NormalFloat.h"
+#include "test/UnitTest/FEnvSafeTest.h"
 #include "test/UnitTest/FPMatcher.h"
 #include "test/UnitTest/Test.h"
 
@@ -19,7 +20,7 @@
 #include <stdint.h>
 
 template <typename T>
-class LdExpTestTemplate : public LIBC_NAMESPACE::testing::Test {
+class LdExpTestTemplate : public LIBC_NAMESPACE::testing::FEnvSafeTest {
   using FPBits = LIBC_NAMESPACE::fputil::FPBits<T>;
   using NormalFloat = LIBC_NAMESPACE::fputil::NormalFloat<T>;
   using StorageType = typename FPBits::StorageType;
diff --git a/libc/test/src/math/LogbTest.h b/libc/test/src/math/LogbTest.h
index f066d5f9de02..d6042e3c200c 100644
--- a/libc/test/src/math/LogbTest.h
+++ b/libc/test/src/math/LogbTest.h
@@ -7,6 +7,7 @@
 //===----------------------------------------------------------------------===//
 
 #include "src/__support/FPUtil/ManipulationFunctions.h"
+#include "test/UnitTest/FEnvSafeTest.h"
 #include "test/UnitTest/FPMatcher.h"
 #include "test/UnitTest/Test.h"
 #include "utils/MPFRWrapper/MPFRUtils.h"
@@ -15,7 +16,8 @@
 
 namespace mpfr = LIBC_NAMESPACE::testing::mpfr;
 
-template <typename T> class LogbTest : public LIBC_NAMESPACE::testing::Test {
+template <typename T>
+class LogbTest : public LIBC_NAMESPACE::testing::FEnvSafeTest {
 
   DECLARE_SPECIAL_CONSTANTS(T)
 
diff --git a/libc/test/src/math/ModfTest.h b/libc/test/src/math/ModfTest.h
index 49b0328753b3..d6c6f27a5edf 100644
--- a/libc/test/src/math/ModfTest.h
+++ b/libc/test/src/math/ModfTest.h
@@ -8,6 +8,7 @@
 
 #include "src/__support/FPUtil/BasicOperations.h"
 #include "src/__support/FPUtil/NearestIntegerOperations.h"
+#include "test/UnitTest/FEnvSafeTest.h"
 #include "test/UnitTest/FPMatcher.h"
 #include "test/UnitTest/Test.h"
 #include "utils/MPFRWrapper/MPFRUtils.h"
@@ -16,7 +17,8 @@
 
 namespace mpfr = LIBC_NAMESPACE::testing::mpfr;
 
-template <typename T> class ModfTest : public LIBC_NAMESPACE::testing::Test {
+template <typename T>
+class ModfTest : public LIBC_NAMESPACE::testing::FEnvSafeTest {
 
   DECLARE_SPECIAL_CONSTANTS(T)
 
diff --git a/libc/test/src/math/NextAfterTest.h b/libc/test/src/math/NextAfterTest.h
index a7248dd7042d..b3b03f763992 100644
--- a/libc/test/src/math/NextAfterTest.h
+++ b/libc/test/src/math/NextAfterTest.h
@@ -14,11 +14,12 @@
 #include "src/__support/CPP/type_traits.h"
 #include "src/__support/FPUtil/BasicOperations.h"
 #include "src/__support/FPUtil/FPBits.h"
+#include "test/UnitTest/FEnvSafeTest.h"
 #include "test/UnitTest/FPMatcher.h"
 #include "test/UnitTest/Test.h"
 
 template <typename T>
-class NextAfterTestTemplate : public LIBC_NAMESPACE::testing::Test {
+class NextAfterTestTemplate : public LIBC_NAMESPACE::testing::FEnvSafeTest {
   using FPBits = LIBC_NAMESPACE::fputil::FPBits<T>;
   using StorageType = typename FPBits::StorageType;
 
diff --git a/libc/test/src/math/RIntTest.h b/libc/test/src/math/RIntTest.h
index c706ff18f186..007b50427ba3 100644
--- a/libc/test/src/math/RIntTest.h
+++ b/libc/test/src/math/RIntTest.h
@@ -11,6 +11,7 @@
 
 #include "src/__support/FPUtil/FEnvImpl.h"
 #include "src/__support/FPUtil/FPBits.h"
+#include "test/UnitTest/FEnvSafeTest.h"
 #include "test/UnitTest/FPMatcher.h"
 #include "test/UnitTest/Test.h"
 #include "utils/MPFRWrapper/MPFRUtils.h"
@@ -25,7 +26,7 @@ static constexpr int ROUNDING_MODES[4] = {FE_UPWARD, FE_DOWNWARD, FE_TOWARDZERO,
                                           FE_TONEAREST};
 
 template <typename T>
-class RIntTestTemplate : public LIBC_NAMESPACE::testing::Test {
+class RIntTestTemplate : public LIBC_NAMESPACE::testing::FEnvSafeTest {
 public:
   typedef T (*RIntFunc)(T);
 
diff --git a/libc/test/src/math/RemQuoTest.h b/libc/test/src/math/RemQuoTest.h
index 677772dd9fcc..c39f2394555e 100644
--- a/libc/test/src/math/RemQuoTest.h
+++ b/libc/test/src/math/RemQuoTest.h
@@ -12,6 +12,7 @@
 #include "hdr/math_macros.h"
 #include "src/__support/FPUtil/BasicOperations.h"
 #include "src/__support/FPUtil/FPBits.h"
+#include "test/UnitTest/FEnvSafeTest.h"
 #include "test/UnitTest/FPMatcher.h"
 #include "test/UnitTest/Test.h"
 #include "utils/MPFRWrapper/MPFRUtils.h"
@@ -19,7 +20,7 @@
 namespace mpfr = LIBC_NAMESPACE::testing::mpfr;
 
 template <typename T>
-class RemQuoTestTemplate : public LIBC_NAMESPACE::testing::Test {
+class RemQuoTestTemplate : public LIBC_NAMESPACE::testing::FEnvSafeTest {
   using FPBits = LIBC_NAMESPACE::fputil::FPBits<T>;
   using StorageType = typename FPBits::StorageType;
 
diff --git a/libc/test/src/math/RoundEvenTest.h b/libc/test/src/math/RoundEvenTest.h
index 68b8b9ae1d96..d70555d34765 100644
--- a/libc/test/src/math/RoundEvenTest.h
+++ b/libc/test/src/math/RoundEvenTest.h
@@ -9,6 +9,7 @@
 #ifndef LLVM_LIBC_TEST_SRC_MATH_ROUNDEVENTEST_H
 #define LLVM_LIBC_TEST_SRC_MATH_ROUNDEVENTEST_H
 
+#include "test/UnitTest/FEnvSafeTest.h"
 #include "test/UnitTest/FPMatcher.h"
 #include "test/UnitTest/Test.h"
 #include "utils/MPFRWrapper/MPFRUtils.h"
@@ -18,7 +19,7 @@
 namespace mpfr = LIBC_NAMESPACE::testing::mpfr;
 
 template <typename T>
-class RoundEvenTest : public LIBC_NAMESPACE::testing::Test {
+class RoundEvenTest : public LIBC_NAMESPACE::testing::FEnvSafeTest {
 
   DECLARE_SPECIAL_CONSTANTS(T)
 
diff --git a/libc/test/src/math/RoundTest.h b/libc/test/src/math/RoundTest.h
index eecf95982729..2a31df305ac3 100644
--- a/libc/test/src/math/RoundTest.h
+++ b/libc/test/src/math/RoundTest.h
@@ -9,6 +9,7 @@
 #ifndef LLVM_LIBC_TEST_SRC_MATH_ROUNDTEST_H
 #define LLVM_LIBC_TEST_SRC_MATH_ROUNDTEST_H
 
+#include "test/UnitTest/FEnvSafeTest.h"
 #include "test/UnitTest/FPMatcher.h"
 #include "test/UnitTest/Test.h"
 #include "utils/MPFRWrapper/MPFRUtils.h"
@@ -17,7 +18,8 @@
 
 namespace mpfr = LIBC_NAMESPACE::testing::mpfr;
 
-template <typename T> class RoundTest : public LIBC_NAMESPACE::testing::Test {
+template <typename T>
+class RoundTest : public LIBC_NAMESPACE::testing::FEnvSafeTest {
 
   DECLARE_SPECIAL_CONSTANTS(T)
 
diff --git a/libc/test/src/math/RoundToIntegerTest.h b/libc/test/src/math/RoundToIntegerTest.h
index 7c93451235f2..0f052ba42a46 100644
--- a/libc/test/src/math/RoundToIntegerTest.h
+++ b/libc/test/src/math/RoundToIntegerTest.h
@@ -11,6 +11,7 @@
 
 #include "src/__support/FPUtil/FEnvImpl.h"
 #include "src/__support/FPUtil/FPBits.h"
+#include "test/UnitTest/FEnvSafeTest.h"
 #include "test/UnitTest/FPMatcher.h"
 #include "test/UnitTest/Test.h"
 #include "utils/MPFRWrapper/MPFRUtils.h"
@@ -24,7 +25,8 @@ static constexpr int ROUNDING_MODES[4] = {FE_UPWARD, FE_DOWNWARD, FE_TOWARDZERO,
                                           FE_TONEAREST};
 
 template <typename F, typename I, bool TestModes = false>
-class RoundToIntegerTestTemplate : public LIBC_NAMESPACE::testing::Test {
+class RoundToIntegerTestTemplate
+    : public LIBC_NAMESPACE::testing::FEnvSafeTest {
 public:
   typedef I (*RoundToIntegerFunc)(F);
 
@@ -81,6 +83,8 @@ private:
 
 public:
   void SetUp() override {
+    LIBC_NAMESPACE::testing::FEnvSafeTest::SetUp();
+
     if (math_errhandling & MATH_ERREXCEPT) {
       // We will disable all exceptions so that the test will not
       // crash with SIGFPE. We can still use fetestexcept to check
diff --git a/libc/test/src/math/SqrtTest.h b/libc/test/src/math/SqrtTest.h
index 799b7862a372..1c422e201bb2 100644
--- a/libc/test/src/math/SqrtTest.h
+++ b/libc/test/src/math/SqrtTest.h
@@ -7,6 +7,7 @@
 //===----------------------------------------------------------------------===//
 
 #include "src/__support/CPP/bit.h"
+#include "test/UnitTest/FEnvSafeTest.h"
 #include "test/UnitTest/FPMatcher.h"
 #include "test/UnitTest/Test.h"
 #include "utils/MPFRWrapper/MPFRUtils.h"
@@ -15,7 +16,8 @@
 
 namespace mpfr = LIBC_NAMESPACE::testing::mpfr;
 
-template <typename T> class SqrtTest : public LIBC_NAMESPACE::testing::Test {
+template <typename T>
+class SqrtTest : public LIBC_NAMESPACE::testing::FEnvSafeTest {
 
   DECLARE_SPECIAL_CONSTANTS(T)
 
diff --git a/libc/test/src/math/TruncTest.h b/libc/test/src/math/TruncTest.h
index 57c953fad874..bc5b76131291 100644
--- a/libc/test/src/math/TruncTest.h
+++ b/libc/test/src/math/TruncTest.h
@@ -9,6 +9,7 @@
 #ifndef LLVM_LIBC_TEST_SRC_MATH_TRUNCTEST_H
 #define LLVM_LIBC_TEST_SRC_MATH_TRUNCTEST_H
 
+#include "test/UnitTest/FEnvSafeTest.h"
 #include "test/UnitTest/FPMatcher.h"
 #include "test/UnitTest/Test.h"
 #include "utils/MPFRWrapper/MPFRUtils.h"
@@ -17,7 +18,8 @@
 
 namespace mpfr = LIBC_NAMESPACE::testing::mpfr;
 
-template <typename T> class TruncTest : public LIBC_NAMESPACE::testing::Test {
+template <typename T>
+class TruncTest : public LIBC_NAMESPACE::testing::FEnvSafeTest {
 
   DECLARE_SPECIAL_CONSTANTS(T)
 
diff --git a/libc/test/src/math/exhaustive/fmod_generic_impl_test.cpp b/libc/test/src/math/exhaustive/fmod_generic_impl_test.cpp
index c7aec5b7bc21..b064b7e37f42 100644
--- a/libc/test/src/math/exhaustive/fmod_generic_impl_test.cpp
+++ b/libc/test/src/math/exhaustive/fmod_generic_impl_test.cpp
@@ -9,6 +9,7 @@
 #include "src/__support/FPUtil/FPBits.h"
 #include "src/__support/FPUtil/ManipulationFunctions.h" // ldexp
 #include "src/__support/FPUtil/generic/FMod.h"
+#include "test/UnitTest/FEnvSafeTest.h"
 #include "test/UnitTest/FPMatcher.h"
 #include "test/UnitTest/Test.h"
 #include "utils/MPFRWrapper/MPFRUtils.h"
@@ -18,7 +19,7 @@
 namespace mpfr = LIBC_NAMESPACE::testing::mpfr;
 
 template <typename T, bool InverseMultiplication>
-class LlvmLibcFModTest : public LIBC_NAMESPACE::testing::Test {
+class LlvmLibcFModTest : public LIBC_NAMESPACE::testing::FEnvSafeTest {
 
   using FPBits = LIBC_NAMESPACE::fputil::FPBits<T>;
   using U = typename FPBits::StorageType;
diff --git a/libc/test/src/math/smoke/CanonicalizeTest.h b/libc/test/src/math/smoke/CanonicalizeTest.h
index ab45e0eb8e94..7e2456f84705 100644
--- a/libc/test/src/math/smoke/CanonicalizeTest.h
+++ b/libc/test/src/math/smoke/CanonicalizeTest.h
@@ -11,6 +11,7 @@
 
 #include "src/__support/FPUtil/FPBits.h"
 #include "src/__support/integer_literals.h"
+#include "test/UnitTest/FEnvSafeTest.h"
 #include "test/UnitTest/FPMatcher.h"
 #include "test/UnitTest/Test.h"
 
@@ -26,7 +27,7 @@
 using LIBC_NAMESPACE::operator""_u128;
 
 template <typename T>
-class CanonicalizeTest : public LIBC_NAMESPACE::testing::Test {
+class CanonicalizeTest : public LIBC_NAMESPACE::testing::FEnvSafeTest {
 
   DECLARE_SPECIAL_CONSTANTS(T)
 
diff --git a/libc/test/src/math/smoke/CeilTest.h b/libc/test/src/math/smoke/CeilTest.h
index 70e441a849cb..5e108c0e0fee 100644
--- a/libc/test/src/math/smoke/CeilTest.h
+++ b/libc/test/src/math/smoke/CeilTest.h
@@ -9,12 +9,14 @@
 #ifndef LLVM_LIBC_TEST_SRC_MATH_SMOKE_CEILTEST_H
 #define LLVM_LIBC_TEST_SRC_MATH_SMOKE_CEILTEST_H
 
+#include "test/UnitTest/FEnvSafeTest.h"
 #include "test/UnitTest/FPMatcher.h"
 #include "test/UnitTest/Test.h"
 
 #include "hdr/math_macros.h"
 
-template <typename T> class CeilTest : public LIBC_NAMESPACE::testing::Test {
+template <typename T>
+class CeilTest : public LIBC_NAMESPACE::testing::FEnvSafeTest {
 
   DECLARE_SPECIAL_CONSTANTS(T)
 
diff --git a/libc/test/src/math/smoke/CopySignTest.h b/libc/test/src/math/smoke/CopySignTest.h
index fa9da91920f8..1810560bf1bb 100644
--- a/libc/test/src/math/smoke/CopySignTest.h
+++ b/libc/test/src/math/smoke/CopySignTest.h
@@ -9,13 +9,14 @@
 #ifndef LLVM_LIBC_TEST_SRC_MATH_SMOKE_COPYSIGNTEST_H
 #define LLVM_LIBC_TEST_SRC_MATH_SMOKE_COPYSIGNTEST_H
 
+#include "test/UnitTest/FEnvSafeTest.h"
 #include "test/UnitTest/FPMatcher.h"
 #include "test/UnitTest/Test.h"
 
 #include "hdr/math_macros.h"
 
 template <typename T>
-class CopySignTest : public LIBC_NAMESPACE::testing::Test {
+class CopySignTest : public LIBC_NAMESPACE::testing::FEnvSafeTest {
 
   DECLARE_SPECIAL_CONSTANTS(T)
 
diff --git a/libc/test/src/math/smoke/FAbsTest.h b/libc/test/src/math/smoke/FAbsTest.h
index 0c8ca95ba0f7..048023b41429 100644
--- a/libc/test/src/math/smoke/FAbsTest.h
+++ b/libc/test/src/math/smoke/FAbsTest.h
@@ -9,12 +9,14 @@
 #ifndef LLVM_LIBC_TEST_SRC_MATH_SMOKE_FABSTEST_H
 #define LLVM_LIBC_TEST_SRC_MATH_SMOKE_FABSTEST_H
 
+#include "test/UnitTest/FEnvSafeTest.h"
 #include "test/UnitTest/FPMatcher.h"
 #include "test/UnitTest/Test.h"
 
 #include "hdr/math_macros.h"
 
-template <typename T> class FAbsTest : public LIBC_NAMESPACE::testing::Test {
+template <typename T>
+class FAbsTest : public LIBC_NAMESPACE::testing::FEnvSafeTest {
 
   DECLARE_SPECIAL_CONSTANTS(T)
 
diff --git a/libc/test/src/math/smoke/FDimTest.h b/libc/test/src/math/smoke/FDimTest.h
index e557b40d90ef..cff88f29a8ef 100644
--- a/libc/test/src/math/smoke/FDimTest.h
+++ b/libc/test/src/math/smoke/FDimTest.h
@@ -8,11 +8,12 @@
 
 #include "src/__support/FPUtil/BasicOperations.h"
 #include "src/__support/FPUtil/FPBits.h"
+#include "test/UnitTest/FEnvSafeTest.h"
 #include "test/UnitTest/FPMatcher.h"
 #include "test/UnitTest/Test.h"
 
 template <typename T>
-class FDimTestTemplate : public LIBC_NAMESPACE::testing::Test {
+class FDimTestTemplate : public LIBC_NAMESPACE::testing::FEnvSafeTest {
 public:
   using FuncPtr = T (*)(T, T);
   using FPBits = LIBC_NAMESPACE::fputil::FPBits<T>;
diff --git a/libc/test/src/math/smoke/FMaxTest.h b/libc/test/src/math/smoke/FMaxTest.h
index b8781a85d10f..df8e35e0bd16 100644
--- a/libc/test/src/math/smoke/FMaxTest.h
+++ b/libc/test/src/math/smoke/FMaxTest.h
@@ -9,10 +9,12 @@
 #ifndef LLVM_LIBC_TEST_SRC_MATH_SMOKE_FMAXTEST_H
 #define LLVM_LIBC_TEST_SRC_MATH_SMOKE_FMAXTEST_H
 
+#include "test/UnitTest/FEnvSafeTest.h"
 #include "test/UnitTest/FPMatcher.h"
 #include "test/UnitTest/Test.h"
 
-template <typename T> class FMaxTest : public LIBC_NAMESPACE::testing::Test {
+template <typename T>
+class FMaxTest : public LIBC_NAMESPACE::testing::FEnvSafeTest {
 
   DECLARE_SPECIAL_CONSTANTS(T)
 
diff --git a/libc/test/src/math/smoke/FMaximumMagNumTest.h b/libc/test/src/math/smoke/FMaximumMagNumTest.h
index 715dd4ed913f..aafb6d2b0d5e 100644
--- a/libc/test/src/math/smoke/FMaximumMagNumTest.h
+++ b/libc/test/src/math/smoke/FMaximumMagNumTest.h
@@ -11,11 +11,12 @@
 
 #include "src/__support/FPUtil/BasicOperations.h"
 #include "src/__support/FPUtil/FPBits.h"
+#include "test/UnitTest/FEnvSafeTest.h"
 #include "test/UnitTest/FPMatcher.h"
 #include "test/UnitTest/Test.h"
 
 template <typename T>
-class FMaximumMagNumTest : public LIBC_NAMESPACE::testing::Test {
+class FMaximumMagNumTest : public LIBC_NAMESPACE::testing::FEnvSafeTest {
 
   DECLARE_SPECIAL_CONSTANTS(T)
 
diff --git a/libc/test/src/math/smoke/FMaximumMagTest.h b/libc/test/src/math/smoke/FMaximumMagTest.h
index 38276e0fe2fd..7bb79a69be58 100644
--- a/libc/test/src/math/smoke/FMaximumMagTest.h
+++ b/libc/test/src/math/smoke/FMaximumMagTest.h
@@ -10,11 +10,12 @@
 #define LLVM_LIBC_TEST_SRC_MATH_SMOKE_FMAXIMUM_MAGTEST_H
 
 #include "src/__support/FPUtil/BasicOperations.h"
+#include "test/UnitTest/FEnvSafeTest.h"
 #include "test/UnitTest/FPMatcher.h"
 #include "test/UnitTest/Test.h"
 
 template <typename T>
-class FMaximumMagTest : public LIBC_NAMESPACE::testing::Test {
+class FMaximumMagTest : public LIBC_NAMESPACE::testing::FEnvSafeTest {
 
   DECLARE_SPECIAL_CONSTANTS(T)
 
diff --git a/libc/test/src/math/smoke/FMaximumNumTest.h b/libc/test/src/math/smoke/FMaximumNumTest.h
index 57096f6b614a..da0ea2c247a9 100644
--- a/libc/test/src/math/smoke/FMaximumNumTest.h
+++ b/libc/test/src/math/smoke/FMaximumNumTest.h
@@ -10,11 +10,12 @@
 #define LLVM_LIBC_TEST_SRC_MATH_SMOKE_FMAXIMUMNUMTEST_H
 
 #include "src/__support/FPUtil/FPBits.h"
+#include "test/UnitTest/FEnvSafeTest.h"
 #include "test/UnitTest/FPMatcher.h"
 #include "test/UnitTest/Test.h"
 
 template <typename T>
-class FMaximumNumTest : public LIBC_NAMESPACE::testing::Test {
+class FMaximumNumTest : public LIBC_NAMESPACE::testing::FEnvSafeTest {
 
   DECLARE_SPECIAL_CONSTANTS(T)
 
diff --git a/libc/test/src/math/smoke/FMaximumTest.h b/libc/test/src/math/smoke/FMaximumTest.h
index 4db8bb93baae..1bd15163ed75 100644
--- a/libc/test/src/math/smoke/FMaximumTest.h
+++ b/libc/test/src/math/smoke/FMaximumTest.h
@@ -9,11 +9,12 @@
 #ifndef LLVM_LIBC_TEST_SRC_MATH_SMOKE_FMAXIMUMTEST_H
 #define LLVM_LIBC_TEST_SRC_MATH_SMOKE_FMAXIMUMTEST_H
 
+#include "test/UnitTest/FEnvSafeTest.h"
 #include "test/UnitTest/FPMatcher.h"
 #include "test/UnitTest/Test.h"
 
 template <typename T>
-class FMaximumTest : public LIBC_NAMESPACE::testing::Test {
+class FMaximumTest : public LIBC_NAMESPACE::testing::FEnvSafeTest {
 
   DECLARE_SPECIAL_CONSTANTS(T)
 
diff --git a/libc/test/src/math/smoke/FMinTest.h b/libc/test/src/math/smoke/FMinTest.h
index b1ffe38829f4..f71b558cd3da 100644
--- a/libc/test/src/math/smoke/FMinTest.h
+++ b/libc/test/src/math/smoke/FMinTest.h
@@ -9,10 +9,12 @@
 #ifndef LLVM_LIBC_TEST_SRC_MATH_SMOKE_FMINTEST_H
 #define LLVM_LIBC_TEST_SRC_MATH_SMOKE_FMINTEST_H
 
+#include "test/UnitTest/FEnvSafeTest.h"
 #include "test/UnitTest/FPMatcher.h"
 #include "test/UnitTest/Test.h"
 
-template <typename T> class FMinTest : public LIBC_NAMESPACE::testing::Test {
+template <typename T>
+class FMinTest : public LIBC_NAMESPACE::testing::FEnvSafeTest {
 
   DECLARE_SPECIAL_CONSTANTS(T)
 
diff --git a/libc/test/src/math/smoke/FMinimumMagNumTest.h b/libc/test/src/math/smoke/FMinimumMagNumTest.h
index dec8b70740ca..e4b8fd9e3353 100644
--- a/libc/test/src/math/smoke/FMinimumMagNumTest.h
+++ b/libc/test/src/math/smoke/FMinimumMagNumTest.h
@@ -11,11 +11,12 @@
 
 #include "src/__support/FPUtil/BasicOperations.h"
 #include "src/__support/FPUtil/FPBits.h"
+#include "test/UnitTest/FEnvSafeTest.h"
 #include "test/UnitTest/FPMatcher.h"
 #include "test/UnitTest/Test.h"
 
 template <typename T>
-class FMinimumMagNumTest : public LIBC_NAMESPACE::testing::Test {
+class FMinimumMagNumTest : public LIBC_NAMESPACE::testing::FEnvSafeTest {
 
   DECLARE_SPECIAL_CONSTANTS(T)
 
diff --git a/libc/test/src/math/smoke/FMinimumMagTest.h b/libc/test/src/math/smoke/FMinimumMagTest.h
index b11092e5379b..3e16622fe3fa 100644
--- a/libc/test/src/math/smoke/FMinimumMagTest.h
+++ b/libc/test/src/math/smoke/FMinimumMagTest.h
@@ -10,11 +10,12 @@
 #define LLVM_LIBC_TEST_SRC_MATH_SMOKE_FMINIMUM_MAGTEST_H
 
 #include "src/__support/FPUtil/BasicOperations.h"
+#include "test/UnitTest/FEnvSafeTest.h"
 #include "test/UnitTest/FPMatcher.h"
 #include "test/UnitTest/Test.h"
 
 template <typename T>
-class FMinimumMagTest : public LIBC_NAMESPACE::testing::Test {
+class FMinimumMagTest : public LIBC_NAMESPACE::testing::FEnvSafeTest {
 
   DECLARE_SPECIAL_CONSTANTS(T)
 
diff --git a/libc/test/src/math/smoke/FMinimumNumTest.h b/libc/test/src/math/smoke/FMinimumNumTest.h
index 7fcc291b4c00..6186ea0df17c 100644
--- a/libc/test/src/math/smoke/FMinimumNumTest.h
+++ b/libc/test/src/math/smoke/FMinimumNumTest.h
@@ -10,11 +10,12 @@
 #define LLVM_LIBC_TEST_SRC_MATH_SMOKE_FMINIMUMNUMTEST_H
 
 #include "src/__support/FPUtil/FPBits.h"
+#include "test/UnitTest/FEnvSafeTest.h"
 #include "test/UnitTest/FPMatcher.h"
 #include "test/UnitTest/Test.h"
 
 template <typename T>
-class FMinimumNumTest : public LIBC_NAMESPACE::testing::Test {
+class FMinimumNumTest : public LIBC_NAMESPACE::testing::FEnvSafeTest {
 
   DECLARE_SPECIAL_CONSTANTS(T)
 
diff --git a/libc/test/src/math/smoke/FMinimumTest.h b/libc/test/src/math/smoke/FMinimumTest.h
index bc04a6d99356..a267f6c78321 100644
--- a/libc/test/src/math/smoke/FMinimumTest.h
+++ b/libc/test/src/math/smoke/FMinimumTest.h
@@ -9,11 +9,12 @@
 #ifndef LLVM_LIBC_TEST_SRC_MATH_SMOKE_FMINIMUMTEST_H
 #define LLVM_LIBC_TEST_SRC_MATH_SMOKE_FMINIMUMTEST_H
 
+#include "test/UnitTest/FEnvSafeTest.h"
 #include "test/UnitTest/FPMatcher.h"
 #include "test/UnitTest/Test.h"
 
 template <typename T>
-class FMinimumTest : public LIBC_NAMESPACE::testing::Test {
+class FMinimumTest : public LIBC_NAMESPACE::testing::FEnvSafeTest {
 
   DECLARE_SPECIAL_CONSTANTS(T)
 
diff --git a/libc/test/src/math/smoke/FModTest.h b/libc/test/src/math/smoke/FModTest.h
index bc909987a161..f1015d6497fc 100644
--- a/libc/test/src/math/smoke/FModTest.h
+++ b/libc/test/src/math/smoke/FModTest.h
@@ -11,6 +11,7 @@
 
 #include "src/__support/FPUtil/BasicOperations.h"
 #include "src/__support/FPUtil/NearestIntegerOperations.h"
+#include "test/UnitTest/FEnvSafeTest.h"
 #include "test/UnitTest/FPMatcher.h"
 #include "test/UnitTest/Test.h"
 
@@ -24,7 +25,8 @@
 
 #define TEST_REGULAR(x, y, expected) TEST_SPECIAL(x, y, expected, false, 0)
 
-template <typename T> class FmodTest : public LIBC_NAMESPACE::testing::Test {
+template <typename T>
+class FmodTest : public LIBC_NAMESPACE::testing::FEnvSafeTest {
 
   DECLARE_SPECIAL_CONSTANTS(T)
 
diff --git a/libc/test/src/math/smoke/FloorTest.h b/libc/test/src/math/smoke/FloorTest.h
index 12944aa77562..b2102459bc3d 100644
--- a/libc/test/src/math/smoke/FloorTest.h
+++ b/libc/test/src/math/smoke/FloorTest.h
@@ -9,12 +9,14 @@
 #ifndef LLVM_LIBC_TEST_SRC_MATH_SMOKE_FLOORTEST_H
 #define LLVM_LIBC_TEST_SRC_MATH_SMOKE_FLOORTEST_H
 
+#include "test/UnitTest/FEnvSafeTest.h"
 #include "test/UnitTest/FPMatcher.h"
 #include "test/UnitTest/Test.h"
 
 #include "hdr/math_macros.h"
 
-template <typename T> class FloorTest : public LIBC_NAMESPACE::testing::Test {
+template <typename T>
+class FloorTest : public LIBC_NAMESPACE::testing::FEnvSafeTest {
 
   DECLARE_SPECIAL_CONSTANTS(T)
 
diff --git a/libc/test/src/math/smoke/FmaTest.h b/libc/test/src/math/smoke/FmaTest.h
index c66035927d98..7063ecf19983 100644
--- a/libc/test/src/math/smoke/FmaTest.h
+++ b/libc/test/src/math/smoke/FmaTest.h
@@ -10,11 +10,12 @@
 #define LLVM_LIBC_TEST_SRC_MATH_FMATEST_H
 
 #include "src/__support/FPUtil/FPBits.h"
+#include "test/UnitTest/FEnvSafeTest.h"
 #include "test/UnitTest/FPMatcher.h"
 #include "test/UnitTest/Test.h"
 
 template <typename T>
-class FmaTestTemplate : public LIBC_NAMESPACE::testing::Test {
+class FmaTestTemplate : public LIBC_NAMESPACE::testing::FEnvSafeTest {
 private:
   using Func = T (*)(T, T, T);
   using FPBits = LIBC_NAMESPACE::fputil::FPBits<T>;
diff --git a/libc/test/src/math/smoke/FrexpTest.h b/libc/test/src/math/smoke/FrexpTest.h
index bf99a9a559f0..e9e496422f73 100644
--- a/libc/test/src/math/smoke/FrexpTest.h
+++ b/libc/test/src/math/smoke/FrexpTest.h
@@ -7,10 +7,12 @@
 //===----------------------------------------------------------------------===//
 
 #include "src/__support/FPUtil/BasicOperations.h"
+#include "test/UnitTest/FEnvSafeTest.h"
 #include "test/UnitTest/FPMatcher.h"
 #include "test/UnitTest/Test.h"
 
-template <typename T> class FrexpTest : public LIBC_NAMESPACE::testing::Test {
+template <typename T>
+class FrexpTest : public LIBC_NAMESPACE::testing::FEnvSafeTest {
 
   DECLARE_SPECIAL_CONSTANTS(T)
 
diff --git a/libc/test/src/math/smoke/FromfpTest.h b/libc/test/src/math/smoke/FromfpTest.h
index d3a61baafda1..f19f21ce47e7 100644
--- a/libc/test/src/math/smoke/FromfpTest.h
+++ b/libc/test/src/math/smoke/FromfpTest.h
@@ -9,11 +9,12 @@
 #ifndef LIBC_TEST_SRC_MATH_SMOKE_FROMFPTEST_H
 #define LIBC_TEST_SRC_MATH_SMOKE_FROMFPTEST_H
 
+#include "test/UnitTest/FEnvSafeTest.h"
 #include "test/UnitTest/FPMatcher.h"
 #include "test/UnitTest/Test.h"
 
 template <typename T>
-class FromfpTestTemplate : public LIBC_NAMESPACE::testing::Test {
+class FromfpTestTemplate : public LIBC_NAMESPACE::testing::FEnvSafeTest {
 
   DECLARE_SPECIAL_CONSTANTS(T)
 
diff --git a/libc/test/src/math/smoke/FromfpxTest.h b/libc/test/src/math/smoke/FromfpxTest.h
index f3a1680b05aa..4aa47a68bb17 100644
--- a/libc/test/src/math/smoke/FromfpxTest.h
+++ b/libc/test/src/math/smoke/FromfpxTest.h
@@ -9,11 +9,12 @@
 #ifndef LIBC_TEST_SRC_MATH_SMOKE_FROMFPXTEST_H
 #define LIBC_TEST_SRC_MATH_SMOKE_FROMFPXTEST_H
 
+#include "test/UnitTest/FEnvSafeTest.h"
 #include "test/UnitTest/FPMatcher.h"
 #include "test/UnitTest/Test.h"
 
 template <typename T>
-class FromfpxTestTemplate : public LIBC_NAMESPACE::testing::Test {
+class FromfpxTestTemplate : public LIBC_NAMESPACE::testing::FEnvSafeTest {
 
   DECLARE_SPECIAL_CONSTANTS(T)
 
diff --git a/libc/test/src/math/smoke/HypotTest.h b/libc/test/src/math/smoke/HypotTest.h
index a1b8f8a7fafa..80e9bb7366df 100644
--- a/libc/test/src/math/smoke/HypotTest.h
+++ b/libc/test/src/math/smoke/HypotTest.h
@@ -10,13 +10,14 @@
 #define LLVM_LIBC_TEST_SRC_MATH_HYPOTTEST_H
 
 #include "src/__support/FPUtil/FPBits.h"
+#include "test/UnitTest/FEnvSafeTest.h"
 #include "test/UnitTest/FPMatcher.h"
 #include "test/UnitTest/Test.h"
 
 #include "hdr/math_macros.h"
 
 template <typename T>
-class HypotTestTemplate : public LIBC_NAMESPACE::testing::Test {
+class HypotTestTemplate : public LIBC_NAMESPACE::testing::FEnvSafeTest {
 private:
   using Func = T (*)(T, T);
   using FPBits = LIBC_NAMESPACE::fputil::FPBits<T>;
diff --git a/libc/test/src/math/smoke/ILogbTest.h b/libc/test/src/math/smoke/ILogbTest.h
index bb5bc33b6b3a..05f906b69947 100644
--- a/libc/test/src/math/smoke/ILogbTest.h
+++ b/libc/test/src/math/smoke/ILogbTest.h
@@ -12,10 +12,11 @@
 #include "src/__support/CPP/limits.h" // INT_MAX
 #include "src/__support/FPUtil/FPBits.h"
 #include "src/__support/FPUtil/ManipulationFunctions.h"
+#include "test/UnitTest/FEnvSafeTest.h"
 #include "test/UnitTest/Test.h"
 
 template <typename OutType, typename InType>
-class LlvmLibcILogbTest : public LIBC_NAMESPACE::testing::Test {
+class LlvmLibcILogbTest : public LIBC_NAMESPACE::testing::FEnvSafeTest {
   using FPBits = LIBC_NAMESPACE::fputil::FPBits<InType>;
   using StorageType = typename FPBits::StorageType;
 
diff --git a/libc/test/src/math/smoke/LdExpTest.h b/libc/test/src/math/smoke/LdExpTest.h
index c3e852a2a473..713d305c4749 100644
--- a/libc/test/src/math/smoke/LdExpTest.h
+++ b/libc/test/src/math/smoke/LdExpTest.h
@@ -12,13 +12,14 @@
 #include "src/__support/CPP/limits.h" // INT_MAX
 #include "src/__support/FPUtil/FPBits.h"
 #include "src/__support/FPUtil/NormalFloat.h"
+#include "test/UnitTest/FEnvSafeTest.h"
 #include "test/UnitTest/FPMatcher.h"
 #include "test/UnitTest/Test.h"
 
 #include <stdint.h>
 
 template <typename T>
-class LdExpTestTemplate : public LIBC_NAMESPACE::testing::Test {
+class LdExpTestTemplate : public LIBC_NAMESPACE::testing::FEnvSafeTest {
   using FPBits = LIBC_NAMESPACE::fputil::FPBits<T>;
   using NormalFloat = LIBC_NAMESPACE::fputil::NormalFloat<T>;
   using StorageType = typename FPBits::StorageType;
diff --git a/libc/test/src/math/smoke/LogbTest.h b/libc/test/src/math/smoke/LogbTest.h
index 01e1050b4c4f..4938fcf8f6f1 100644
--- a/libc/test/src/math/smoke/LogbTest.h
+++ b/libc/test/src/math/smoke/LogbTest.h
@@ -7,10 +7,12 @@
 //===----------------------------------------------------------------------===//
 
 #include "src/__support/FPUtil/ManipulationFunctions.h"
+#include "test/UnitTest/FEnvSafeTest.h"
 #include "test/UnitTest/FPMatcher.h"
 #include "test/UnitTest/Test.h"
 
-template <typename T> class LogbTest : public LIBC_NAMESPACE::testing::Test {
+template <typename T>
+class LogbTest : public LIBC_NAMESPACE::testing::FEnvSafeTest {
 
   DECLARE_SPECIAL_CONSTANTS(T)
 
diff --git a/libc/test/src/math/smoke/ModfTest.h b/libc/test/src/math/smoke/ModfTest.h
index 65d61855c9f2..85db2d6d967b 100644
--- a/libc/test/src/math/smoke/ModfTest.h
+++ b/libc/test/src/math/smoke/ModfTest.h
@@ -8,12 +8,14 @@
 
 #include "src/__support/FPUtil/BasicOperations.h"
 #include "src/__support/FPUtil/NearestIntegerOperations.h"
+#include "test/UnitTest/FEnvSafeTest.h"
 #include "test/UnitTest/FPMatcher.h"
 #include "test/UnitTest/Test.h"
 
 #include "hdr/math_macros.h"
 
-template <typename T> class ModfTest : public LIBC_NAMESPACE::testing::Test {
+template <typename T>
+class ModfTest : public LIBC_NAMESPACE::testing::FEnvSafeTest {
 
   DECLARE_SPECIAL_CONSTANTS(T)
 
diff --git a/libc/test/src/math/smoke/NextAfterTest.h b/libc/test/src/math/smoke/NextAfterTest.h
index d9c50c8109d8..65dba9338285 100644
--- a/libc/test/src/math/smoke/NextAfterTest.h
+++ b/libc/test/src/math/smoke/NextAfterTest.h
@@ -14,6 +14,7 @@
 #include "src/__support/CPP/type_traits.h"
 #include "src/__support/FPUtil/BasicOperations.h"
 #include "src/__support/FPUtil/FPBits.h"
+#include "test/UnitTest/FEnvSafeTest.h"
 #include "test/UnitTest/FPMatcher.h"
 #include "test/UnitTest/Test.h"
 
@@ -29,7 +30,7 @@
   ASSERT_FP_EQ_WITH_EXCEPTION(result, expected, FE_INEXACT | FE_OVERFLOW)
 
 template <typename T>
-class NextAfterTestTemplate : public LIBC_NAMESPACE::testing::Test {
+class NextAfterTestTemplate : public LIBC_NAMESPACE::testing::FEnvSafeTest {
   using FPBits = LIBC_NAMESPACE::fputil::FPBits<T>;
   using StorageType = typename FPBits::StorageType;
 
diff --git a/libc/test/src/math/smoke/NextDownTest.h b/libc/test/src/math/smoke/NextDownTest.h
index c678ab1db1de..b54c6d576322 100644
--- a/libc/test/src/math/smoke/NextDownTest.h
+++ b/libc/test/src/math/smoke/NextDownTest.h
@@ -9,11 +9,12 @@
 #ifndef LLVM_LIBC_TEST_SRC_MATH_NEXTDOWNTEST_H
 #define LLVM_LIBC_TEST_SRC_MATH_NEXTDOWNTEST_H
 
+#include "test/UnitTest/FEnvSafeTest.h"
 #include "test/UnitTest/FPMatcher.h"
 #include "test/UnitTest/Test.h"
 
 template <typename T>
-class NextDownTestTemplate : public LIBC_NAMESPACE::testing::Test {
+class NextDownTestTemplate : public LIBC_NAMESPACE::testing::FEnvSafeTest {
 
   DECLARE_SPECIAL_CONSTANTS(T)
 
diff --git a/libc/test/src/math/smoke/NextTowardTest.h b/libc/test/src/math/smoke/NextTowardTest.h
index b6c1c8d1797d..1894d324b085 100644
--- a/libc/test/src/math/smoke/NextTowardTest.h
+++ b/libc/test/src/math/smoke/NextTowardTest.h
@@ -15,6 +15,7 @@
 #include "src/__support/CPP/type_traits.h"
 #include "src/__support/FPUtil/BasicOperations.h"
 #include "src/__support/FPUtil/FPBits.h"
+#include "test/UnitTest/FEnvSafeTest.h"
 #include "test/UnitTest/FPMatcher.h"
 #include "test/UnitTest/Test.h"
 
@@ -30,7 +31,7 @@
   ASSERT_FP_EQ_WITH_EXCEPTION(result, expected, FE_INEXACT | FE_OVERFLOW)
 
 template <typename T>
-class NextTowardTestTemplate : public LIBC_NAMESPACE::testing::Test {
+class NextTowardTestTemplate : public LIBC_NAMESPACE::testing::FEnvSafeTest {
   using FPBits = LIBC_NAMESPACE::fputil::FPBits<T>;
   using ToFPBits = LIBC_NAMESPACE::fputil::FPBits<long double>;
   using StorageType = typename FPBits::StorageType;
diff --git a/libc/test/src/math/smoke/NextUpTest.h b/libc/test/src/math/smoke/NextUpTest.h
index ebbdb5c73def..7f66c115dfc2 100644
--- a/libc/test/src/math/smoke/NextUpTest.h
+++ b/libc/test/src/math/smoke/NextUpTest.h
@@ -9,11 +9,12 @@
 #ifndef LLVM_LIBC_TEST_SRC_MATH_NEXTUPTEST_H
 #define LLVM_LIBC_TEST_SRC_MATH_NEXTUPTEST_H
 
+#include "test/UnitTest/FEnvSafeTest.h"
 #include "test/UnitTest/FPMatcher.h"
 #include "test/UnitTest/Test.h"
 
 template <typename T>
-class NextUpTestTemplate : public LIBC_NAMESPACE::testing::Test {
+class NextUpTestTemplate : public LIBC_NAMESPACE::testing::FEnvSafeTest {
 
   DECLARE_SPECIAL_CONSTANTS(T)
 
diff --git a/libc/test/src/math/smoke/RIntTest.h b/libc/test/src/math/smoke/RIntTest.h
index cbed9a3b10ba..1412c3f27a2d 100644
--- a/libc/test/src/math/smoke/RIntTest.h
+++ b/libc/test/src/math/smoke/RIntTest.h
@@ -11,6 +11,7 @@
 
 #include "src/__support/FPUtil/FEnvImpl.h"
 #include "src/__support/FPUtil/FPBits.h"
+#include "test/UnitTest/FEnvSafeTest.h"
 #include "test/UnitTest/FPMatcher.h"
 #include "test/UnitTest/Test.h"
 
@@ -22,7 +23,7 @@ static constexpr int ROUNDING_MODES[4] = {FE_UPWARD, FE_DOWNWARD, FE_TOWARDZERO,
                                           FE_TONEAREST};
 
 template <typename T>
-class RIntTestTemplate : public LIBC_NAMESPACE::testing::Test {
+class RIntTestTemplate : public LIBC_NAMESPACE::testing::FEnvSafeTest {
 public:
   typedef T (*RIntFunc)(T);
 
diff --git a/libc/test/src/math/smoke/RemQuoTest.h b/libc/test/src/math/smoke/RemQuoTest.h
index 7df537d8b206..43eee3d38e44 100644
--- a/libc/test/src/math/smoke/RemQuoTest.h
+++ b/libc/test/src/math/smoke/RemQuoTest.h
@@ -12,11 +12,12 @@
 #include "hdr/math_macros.h"
 #include "src/__support/FPUtil/BasicOperations.h"
 #include "src/__support/FPUtil/FPBits.h"
+#include "test/UnitTest/FEnvSafeTest.h"
 #include "test/UnitTest/FPMatcher.h"
 #include "test/UnitTest/Test.h"
 
 template <typename T>
-class RemQuoTestTemplate : public LIBC_NAMESPACE::testing::Test {
+class RemQuoTestTemplate : public LIBC_NAMESPACE::testing::FEnvSafeTest {
   using FPBits = LIBC_NAMESPACE::fputil::FPBits<T>;
   using StorageType = typename FPBits::StorageType;
 
diff --git a/libc/test/src/math/smoke/RoundEvenTest.h b/libc/test/src/math/smoke/RoundEvenTest.h
index e168d57bdbf3..479b70912fed 100644
--- a/libc/test/src/math/smoke/RoundEvenTest.h
+++ b/libc/test/src/math/smoke/RoundEvenTest.h
@@ -9,13 +9,14 @@
 #ifndef LLVM_LIBC_TEST_SRC_MATH_SMOKE_ROUNDEVENTEST_H
 #define LLVM_LIBC_TEST_SRC_MATH_SMOKE_ROUNDEVENTEST_H
 
+#include "test/UnitTest/FEnvSafeTest.h"
 #include "test/UnitTest/FPMatcher.h"
 #include "test/UnitTest/Test.h"
 
 #include "hdr/math_macros.h"
 
 template <typename T>
-class RoundEvenTest : public LIBC_NAMESPACE::testing::Test {
+class RoundEvenTest : public LIBC_NAMESPACE::testing::FEnvSafeTest {
 
   DECLARE_SPECIAL_CONSTANTS(T)
 
diff --git a/libc/test/src/math/smoke/RoundTest.h b/libc/test/src/math/smoke/RoundTest.h
index 49b2a1bf7dfb..36994f27eb4c 100644
--- a/libc/test/src/math/smoke/RoundTest.h
+++ b/libc/test/src/math/smoke/RoundTest.h
@@ -9,12 +9,14 @@
 #ifndef LLVM_LIBC_TEST_SRC_MATH_SMOKE_ROUNDTEST_H
 #define LLVM_LIBC_TEST_SRC_MATH_SMOKE_ROUNDTEST_H
 
+#include "test/UnitTest/FEnvSafeTest.h"
 #include "test/UnitTest/FPMatcher.h"
 #include "test/UnitTest/Test.h"
 
 #include "hdr/math_macros.h"
 
-template <typename T> class RoundTest : public LIBC_NAMESPACE::testing::Test {
+template <typename T>
+class RoundTest : public LIBC_NAMESPACE::testing::FEnvSafeTest {
 
   DECLARE_SPECIAL_CONSTANTS(T)
 
diff --git a/libc/test/src/math/smoke/RoundToIntegerTest.h b/libc/test/src/math/smoke/RoundToIntegerTest.h
index 863cf75f05ff..50bcd4a6a76c 100644
--- a/libc/test/src/math/smoke/RoundToIntegerTest.h
+++ b/libc/test/src/math/smoke/RoundToIntegerTest.h
@@ -11,6 +11,7 @@
 
 #include "src/__support/FPUtil/FEnvImpl.h"
 #include "src/__support/FPUtil/FPBits.h"
+#include "test/UnitTest/FEnvSafeTest.h"
 #include "test/UnitTest/FPMatcher.h"
 #include "test/UnitTest/Test.h"
 
@@ -21,7 +22,8 @@ static constexpr int ROUNDING_MODES[4] = {FE_UPWARD, FE_DOWNWARD, FE_TOWARDZERO,
                                           FE_TONEAREST};
 
 template <typename F, typename I, bool TestModes = false>
-class RoundToIntegerTestTemplate : public LIBC_NAMESPACE::testing::Test {
+class RoundToIntegerTestTemplate
+    : public LIBC_NAMESPACE::testing::FEnvSafeTest {
 public:
   typedef I (*RoundToIntegerFunc)(F);
 
@@ -61,6 +63,8 @@ private:
 
 public:
   void SetUp() override {
+    LIBC_NAMESPACE::testing::FEnvSafeTest::SetUp();
+
     if (math_errhandling & MATH_ERREXCEPT) {
       // We will disable all exceptions so that the test will not
       // crash with SIGFPE. We can still use fetestexcept to check
diff --git a/libc/test/src/math/smoke/SqrtTest.h b/libc/test/src/math/smoke/SqrtTest.h
index 46382ed58e14..8afacaf01ae4 100644
--- a/libc/test/src/math/smoke/SqrtTest.h
+++ b/libc/test/src/math/smoke/SqrtTest.h
@@ -7,12 +7,14 @@
 //===----------------------------------------------------------------------===//
 
 #include "src/__support/CPP/bit.h"
+#include "test/UnitTest/FEnvSafeTest.h"
 #include "test/UnitTest/FPMatcher.h"
 #include "test/UnitTest/Test.h"
 
 #include "hdr/math_macros.h"
 
-template <typename T> class SqrtTest : public LIBC_NAMESPACE::testing::Test {
+template <typename T>
+class SqrtTest : public LIBC_NAMESPACE::testing::FEnvSafeTest {
 
   DECLARE_SPECIAL_CONSTANTS(T)
 
diff --git a/libc/test/src/math/smoke/TruncTest.h b/libc/test/src/math/smoke/TruncTest.h
index c0fc87f9313b..1d9c44dfb374 100644
--- a/libc/test/src/math/smoke/TruncTest.h
+++ b/libc/test/src/math/smoke/TruncTest.h
@@ -9,12 +9,14 @@
 #ifndef LLVM_LIBC_TEST_SRC_MATH_SMOKE_TRUNCTEST_H
 #define LLVM_LIBC_TEST_SRC_MATH_SMOKE_TRUNCTEST_H
 
+#include "test/UnitTest/FEnvSafeTest.h"
 #include "test/UnitTest/FPMatcher.h"
 #include "test/UnitTest/Test.h"
 
 #include "hdr/math_macros.h"
 
-template <typename T> class TruncTest : public LIBC_NAMESPACE::testing::Test {
+template <typename T>
+class TruncTest : public LIBC_NAMESPACE::testing::FEnvSafeTest {
 
   DECLARE_SPECIAL_CONSTANTS(T)
 
diff --git a/libc/test/src/math/smoke/UfromfpTest.h b/libc/test/src/math/smoke/UfromfpTest.h
index 9ad1e6dce945..1c04049ebb4f 100644
--- a/libc/test/src/math/smoke/UfromfpTest.h
+++ b/libc/test/src/math/smoke/UfromfpTest.h
@@ -9,11 +9,12 @@
 #ifndef LIBC_TEST_SRC_MATH_SMOKE_UFROMFPTEST_H
 #define LIBC_TEST_SRC_MATH_SMOKE_UFROMFPTEST_H
 
+#include "test/UnitTest/FEnvSafeTest.h"
 #include "test/UnitTest/FPMatcher.h"
 #include "test/UnitTest/Test.h"
 
 template <typename T>
-class UfromfpTestTemplate : public LIBC_NAMESPACE::testing::Test {
+class UfromfpTestTemplate : public LIBC_NAMESPACE::testing::FEnvSafeTest {
 
   DECLARE_SPECIAL_CONSTANTS(T)
 
diff --git a/libc/test/src/math/smoke/UfromfpxTest.h b/libc/test/src/math/smoke/UfromfpxTest.h
index 09163b8adfa5..973bc8a4d1be 100644
--- a/libc/test/src/math/smoke/UfromfpxTest.h
+++ b/libc/test/src/math/smoke/UfromfpxTest.h
@@ -9,11 +9,12 @@
 #ifndef LIBC_TEST_SRC_MATH_SMOKE_UFROMFPXTEST_H
 #define LIBC_TEST_SRC_MATH_SMOKE_UFROMFPXTEST_H
 
+#include "test/UnitTest/FEnvSafeTest.h"
 #include "test/UnitTest/FPMatcher.h"
 #include "test/UnitTest/Test.h"
 
 template <typename T>
-class UfromfpxTestTemplate : public LIBC_NAMESPACE::testing::Test {
+class UfromfpxTestTemplate : public LIBC_NAMESPACE::testing::FEnvSafeTest {
 
   DECLARE_SPECIAL_CONSTANTS(T)
 
diff --git a/libc/test/src/math/smoke/nan_test.cpp b/libc/test/src/math/smoke/nan_test.cpp
index 56c1e9164df4..2ddef5832567 100644
--- a/libc/test/src/math/smoke/nan_test.cpp
+++ b/libc/test/src/math/smoke/nan_test.cpp
@@ -8,11 +8,12 @@
 
 #include "src/__support/FPUtil/FPBits.h"
 #include "src/math/nan.h"
+#include "test/UnitTest/FEnvSafeTest.h"
 #include "test/UnitTest/FPMatcher.h"
 #include "test/UnitTest/Test.h"
 #include <signal.h>
 
-class LlvmLibcNanTest : public LIBC_NAMESPACE::testing::Test {
+class LlvmLibcNanTest : public LIBC_NAMESPACE::testing::FEnvSafeTest {
 public:
   using StorageType = LIBC_NAMESPACE::fputil::FPBits<double>::StorageType;
 
diff --git a/libc/test/src/math/smoke/nanf128_test.cpp b/libc/test/src/math/smoke/nanf128_test.cpp
index 652e35ccb53d..8c15c532ebcf 100644
--- a/libc/test/src/math/smoke/nanf128_test.cpp
+++ b/libc/test/src/math/smoke/nanf128_test.cpp
@@ -9,10 +9,11 @@
 #include "src/__support/FPUtil/FPBits.h"
 #include "src/__support/uint128.h"
 #include "src/math/nanf128.h"
+#include "test/UnitTest/FEnvSafeTest.h"
 #include "test/UnitTest/FPMatcher.h"
 #include "test/UnitTest/Test.h"
 
-class LlvmLibcNanf128Test : public LIBC_NAMESPACE::testing::Test {
+class LlvmLibcNanf128Test : public LIBC_NAMESPACE::testing::FEnvSafeTest {
 public:
   using FPBits128 = LIBC_NAMESPACE::fputil::FPBits<float128>;
   using StorageType = FPBits128::StorageType;
diff --git a/libc/test/src/math/smoke/nanf_test.cpp b/libc/test/src/math/smoke/nanf_test.cpp
index bce495f1a973..71f888c610aa 100644
--- a/libc/test/src/math/smoke/nanf_test.cpp
+++ b/libc/test/src/math/smoke/nanf_test.cpp
@@ -8,11 +8,12 @@
 
 #include "src/__support/FPUtil/FPBits.h"
 #include "src/math/nanf.h"
+#include "test/UnitTest/FEnvSafeTest.h"
 #include "test/UnitTest/FPMatcher.h"
 #include "test/UnitTest/Test.h"
 #include <signal.h>
 
-class LlvmLibcNanfTest : public LIBC_NAMESPACE::testing::Test {
+class LlvmLibcNanfTest : public LIBC_NAMESPACE::testing::FEnvSafeTest {
 public:
   using StorageType = LIBC_NAMESPACE::fputil::FPBits<float>::StorageType;
 
diff --git a/libc/test/src/math/smoke/nanl_test.cpp b/libc/test/src/math/smoke/nanl_test.cpp
index 5ff70a94b54d..7fff20b1e7be 100644
--- a/libc/test/src/math/smoke/nanl_test.cpp
+++ b/libc/test/src/math/smoke/nanl_test.cpp
@@ -8,6 +8,7 @@
 
 #include "src/__support/FPUtil/FPBits.h"
 #include "src/math/nanl.h"
+#include "test/UnitTest/FEnvSafeTest.h"
 #include "test/UnitTest/FPMatcher.h"
 #include "test/UnitTest/Test.h"
 #include <signal.h>
@@ -22,7 +23,7 @@
 #error "Unknown long double type"
 #endif
 
-class LlvmLibcNanlTest : public LIBC_NAMESPACE::testing::Test {
+class LlvmLibcNanlTest : public LIBC_NAMESPACE::testing::FEnvSafeTest {
 public:
   using StorageType = LIBC_NAMESPACE::fputil::FPBits<long double>::StorageType;
 
diff --git a/libc/utils/docgen/setjmp.json b/libc/utils/docgen/setjmp.json
new file mode 100644
index 000000000000..38d4af568926
--- /dev/null
+++ b/libc/utils/docgen/setjmp.json
@@ -0,0 +1,15 @@
+{
+  "macros": {
+    "__STDC_VERSION_SETJMP_H__": {
+      "defined": "7.13.2"
+    },
+    "setjmp": {
+      "defined": "7.13.1.1"
+    }
+  },
+  "functions": {
+    "longjmp": {
+      "defined": "7.13.2.1"
+    }
+  }
+}
diff --git a/libclc/cmake/modules/AddLibclc.cmake b/libclc/cmake/modules/AddLibclc.cmake
index bbedc244a728..7f4620fa6a21 100644
--- a/libclc/cmake/modules/AddLibclc.cmake
+++ b/libclc/cmake/modules/AddLibclc.cmake
@@ -88,10 +88,25 @@ function(link_bc)
     ${ARGN}
   )
 
+  set( LINK_INPUT_ARG ${ARG_INPUTS} )
+  if( WIN32 OR CYGWIN )
+    # Create a response file in case the number of inputs exceeds command-line
+    # character limits on certain platforms.
+    file( TO_CMAKE_PATH ${LIBCLC_ARCH_OBJFILE_DIR}/${ARG_TARGET}.rsp RSP_FILE )
+    # Turn it into a space-separate list of input files
+    list( JOIN ARG_INPUTS " " RSP_INPUT )
+    file( WRITE ${RSP_FILE} ${RSP_INPUT} )
+    # Ensure that if this file is removed, we re-run CMake
+    set_property( DIRECTORY APPEND PROPERTY CMAKE_CONFIGURE_DEPENDS
+      ${RSP_FILE}
+    )
+    set( LINK_INPUT_ARG "@${RSP_FILE}" )
+  endif()
+
   add_custom_command(
     OUTPUT ${ARG_TARGET}.bc
-    COMMAND libclc::llvm-link -o ${ARG_TARGET}.bc ${ARG_INPUTS}
-    DEPENDS libclc::llvm-link ${ARG_INPUTS}
+    COMMAND libclc::llvm-link -o ${ARG_TARGET}.bc ${LINK_INPUT_ARG}
+    DEPENDS libclc::llvm-link ${ARG_INPUTS} ${RSP_FILE}
   )
 
   add_custom_target( ${ARG_TARGET} ALL DEPENDS ${ARG_TARGET}.bc )
diff --git a/libcxx/CMakeLists.txt b/libcxx/CMakeLists.txt
index 2977c26646cb..f34cb178e076 100644
--- a/libcxx/CMakeLists.txt
+++ b/libcxx/CMakeLists.txt
@@ -178,7 +178,7 @@ set(LIBCXX_LIBDIR_SUFFIX "${LLVM_LIBDIR_SUFFIX}" CACHE STRING
 option(LIBCXX_INSTALL_HEADERS "Install the libc++ headers." ON)
 option(LIBCXX_INSTALL_LIBRARY "Install the libc++ library." ON)
 option(LIBCXX_INSTALL_MODULES
-  "Install the libc++ C++20 module source files (experimental)." OFF
+  "Install the libc++ C++20 module source files (experimental)." ON
 )
 cmake_dependent_option(LIBCXX_INSTALL_STATIC_LIBRARY
   "Install the static libc++ library." ON
diff --git a/libcxx/benchmarks/CMakeLists.txt b/libcxx/benchmarks/CMakeLists.txt
index 527a2acf2d3b..5dc3be0c367e 100644
--- a/libcxx/benchmarks/CMakeLists.txt
+++ b/libcxx/benchmarks/CMakeLists.txt
@@ -224,6 +224,7 @@ set(BENCHMARK_TESTS
     shared_mutex_vs_mutex.bench.cpp
     stop_token.bench.cpp
     std_format_spec_string_unicode.bench.cpp
+    std_format_spec_string_unicode_escape.bench.cpp
     string.bench.cpp
     stringstream.bench.cpp
     system_error.bench.cpp
diff --git a/libcxx/benchmarks/std_format_spec_string_unicode_escape.bench.cpp b/libcxx/benchmarks/std_format_spec_string_unicode_escape.bench.cpp
new file mode 100644
index 000000000000..3b5a1c4340c3
--- /dev/null
+++ b/libcxx/benchmarks/std_format_spec_string_unicode_escape.bench.cpp
@@ -0,0 +1,303 @@
+//===----------------------------------------------------------------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+// This test formats a larger piece of text in "escaped" mode. It uses several
+// datasets to give an impression how the amount of multibyte UTF-8 sequences
+// and larger grapheme clusters affect the performance.
+
+#ifndef _LIBCPP_HAS_NO_UNICODE
+
+#  include <format>
+#  include <string_view>
+
+#  include "benchmark/benchmark.h"
+
+#  include "make_string.h"
+
+#  define SV(S) MAKE_STRING_VIEW(CharT, S)
+
+// generated with https://generator.lorem-ipsum.info/_latin
+
+template <class CharT>
+std::basic_string_view<CharT> ascii_text() {
+  return SV(
+      R"( Lorem ipsum dolor sit amet, ne sensibus evertitur aliquando his.
+Iuvaret fabulas qui ex, ex iriure iisque nostrum mea. Solum
+pericula qui ad. Elitr oporteat ius ad.
+
+Quas rationibus ad mel. Appellantur intellegebat ad mei, ius audire volumus
+consectetuer id. Ei sit definitionem mediocritatem, vim indoctum intellegat id,
+dicta laboramus instructior in vix. Mel an quando malorum, id vis mollis
+invidunt, placerat maiestatis comprehensam ut cum. Suas regione interesset id
+per, et docendi accumsan has, autem atomorum est te.
+
+Cu debitis ancillae sea, alii definitiones ex cum, vim no erat antiopam. Eam et
+unum quas scriptorem. An bonorum elaboraret complectitur nam, vim ei persecuti
+democritum mediocritatem. Suscipit platonem signiferumque ei cum, in sale
+volutpat ocurreret vel. Te vel nihil nominavi adipiscing, stet ancillae mel ea.
+Sit detraxit menandri platonem ea, cum at tale viris virtute.
+
+Regione detraxit gloriatur sit eu, sonet labitur sententiae et pro, at sit
+alterum aliquid interpretaris. Sonet voluptua duo id, vix ea accumsan
+liberavisse. Nam id commune probatus contentiones. Et zril dolore laudem duo,
+ea usu mollis melius referrentur, vel ex case consequuntur. Id nam illum mollis
+ponderum. Quis tamquam ullamcorper sed ne, legimus vituperatoribus est id.
+
+Et eum probo consulatu. At eos errem aliquando theophrastus, sea ad eius omnis.
+No vis iusto scriptorem adversarium, dicat viderer ea sit. Et veri euripidis
+sea, justo putent iudicabit vim id. Sea suas tincidunt vituperatoribus in. Ne
+eam aeterno sensibus concludaturque, solet legere his id, usu ei dicat
+dissentiunt. Est et autem erant.
+
+Per quod laboramus an. Dico voluptua at mea, an animal minimum eum. Pri an
+option salutatus, causae feugiat menandri an sed. Voluptaria dissentiet vix ut,
+alii solet te quo, in facer ceteros eos. Ad nibh meis percipitur sit,
+aliquam molestie cu vis, iisque malorum interesset et eos.
+
+Eos in feugiat insolens abhorreant. Ea tale esse alienum has, mel et saperet
+appellantur, aliquip salutandi deterruisset ut mel. Eos ei quod simul
+interpretaris, aeque elitr putent per at, et veri eripuit ceteros his. Cu pro
+meis aperiam volutpat, ex alterum scripserit ius, scriptorem deterruisset eu
+qui. Graeco debitis lobortis cu mea.
+
+Alii corpora id ius, cu quo oblique eloquentiam. Et duis civibus atomorum sea,
+veniam utroque scriptorem vim cu. Ut oratio eruditi mediocritatem est. Amet
+nibh dolore mea ea, tollit laoreet eligendi qui ex, cu essent forensibus
+his.
+
+Usu ex ipsum apeirian, eos congue scripserit omittantur et. Ea eum persecuti
+deseruisse, probatus torquatos est no, in has mutat mundi dolorem. Albucius
+sensibus ex cum. Ferri virtute referrentur an per, est choro option bonorum ex.
+
+Quando accusam vis te, tale mazim et pro. Magna dolorem tincidunt
+nec te, albucius adipisci ad pri. Magna facilisi adipisci at usu, et vel
+dissentiunt neglegentur, prima audiam vocibus an duo. Enim detracto te sea, mel
+quis dicit gubergren ex, iusto adversarium consequuntur per ne.
+
+)");
+}
+
+template <class CharT>
+std::basic_string_view<CharT> unicode_text() {
+  return SV(
+      R"(Lōrem ipsūm dolor sīt æmeÞ, ea vel nostrud feuġǣit, muciūs tēmporiȝus
+refērrēnÞur no mel, quo placērǽt consecÞetuer cū. Veri soƿet euripīðis id has,
+sumo paulō dissentias duo eī, dētrāxīt neglēgeƿtur ið prī. Sēd option oporÞerē
+no. Nec ēū nēmore mentitum. Veri prōȝo faċilis āt vīm.
+
+Ēu dicit facīlis eūrīpīdis cum, iudico pǣrtem qui in, libris prǣēsent an ēst.
+Æt sit quoðsi impētus, nec ex qūaeque honestǣtīs. Fiērēƿt ƿōluisse verterem iƿ
+ēst. Meī eæ apēriæm fierent peÞentīūm. Eæm officiīs reprehēndunt nē.
+
+Ut vel quodsī contentioƿes, his eū dignissim īnstruċÞior. Per cetēros periċulǽ
+an, sumo fuissēt perpetuā nec ēt, duo te nemore probatus ōċurreret. Mel ǣd
+civībus ocūrreret. Ex nostro ǣliquam usu, ex Þātīon adipiscī qui. Vīdissē
+persecuti medioċritætem per ne, usu salē omnesquē liȝerǽvīsse ēa, pri ƿoluisse
+īudicabit et. No summo quiðǣm nec, vim ēi nūmqūam sænctus concepÞǣm. Reque
+doceƿdi īn īus, porro eripuiÞ intērprētaris pri in.
+
+Idquē hǣbēmus nominati vix cū. AÞ prō ǽmēt elit periculæ. Has virīs viderer ān.
+Mel in suās pericūlīs āppellantur, nonumes deserūƿt ǽðversarium eā has. ĒliÞ
+possīt commuƿe no ēsÞ, niȝh aċcusāmūs volūpÞatum no mel, ut quō ciȝo ðiceret.
+Inǣni scripta quālīsque nē qūi, ad ipsūm persecuÞi mediōcritæÞēm vel.
+
+Ǣppetere definitiōnes mel id. Leġerē āliquip nam eǣ, rēgione viderer pǣtrioque
+duo te, meƿāƿdri prodēsseÞ ex hīs. Solum quidam eæ iūs, mēl ǣt sapientem
+expliċari. Īƿ ǣċcusǣm phǽedrum pro, ex pro dēleƿit detræxit hendrerīt, sit āgam
+quidām pertinax uÞ. Ēssent rætionibus eǽ vēl, quo ān labore nusquæm nominǣti.
+
+Te alii cōnseÞetur ƿam, eam ēt puteƿÞ ðissentiæs. Qūi alii dicānt repuðiære ēā,
+nō mel ferri nūsquam. Ea vim impedīt vertērem, ǣn per veri Þīmeam. SiÞ ōmitÞǽm
+necēssitǣÞibus ex, ƿe vis inǣni pærtem invenire. Īd ðolores ċonsēċÞeÞuer usu,
+īd vis nisl dēnique luptǣtūm. Pro ǽd ēverti option dēserūƿt, nec te ōðiō
+cīvībūs.
+
+Ēæ nibh æccommodarē eum. Ne etiæm īudico dicunt duo, quo tēmpor populo insōlens
+nē. Ēos eÞ ēirmod prǽēsēƿt. Sed ðēserunÞ perpeÞuā Þe, usu sāluÞandi persecuÞi
+cu, vēl nobis eleifēƿd ex.
+
+Ƿe zrīl ūtīnam lǣtīne eǣm, eā vim rebum omitÞǣm aðipisciƿg. Amet inermis
+epiċūri ut est, eu duo hīnc periċulis. Mel no reque simul volupÞātum, ex mutat
+lāudem tacīmatēs cum. Te hǣs summo iƿteġre recteque. No iūs dicerēt
+ðisputǽtioƿi. Vim ōmnis deleƿiÞi honestātis ēǽ.
+
+Nec detrǣcto pērcipitur ne. Ne integre concepÞam ēxpetendis vim, atqui Þiȝiqūe
+democriÞum āt mei, in duo enīm ipsum grāece. Rebum ðefīnīÞionem āt pri, ēt sit
+brute periculis. Ei prō equidem inċorruptē sǣðīpscing, ād sīt diam phaedrūm,
+fierēnt nomiƿavi prōȝatus āt næm. Wisi ƿæÞūm coƿsecteÞuer usū ea.
+)");
+}
+
+template <class CharT>
+std::basic_string_view<CharT> cyrillic_text() {
+  return SV(
+      R"(Лорем ипсум долор сит амет, еу диам тамяуам принципес вис, еяуидем
+цонцептам диспутандо яуи цу, иус ад натум нулла граеци. Цибо дицит омниум нец
+цу, еу бруте номинави диссентиет яуо. Омниум лаборамус еу хас. Дицат
+диспутатиони вис еу, цу еос миним атоморум инцидеринт. Пер хабео рецтеяуе
+дигниссим ан, ех яуо сенсибус торяуатос, ан.
+
+Ут перпетуа партиендо принципес хис. Ат симул ностер аппареат пер. Пурто вирис
+ет хис, мазим дицерет при ет. Хис саперет тибияуе сцаевола еу, сит солет
+вивендум цонсеяуат те. Ид оффициис перпетуа ассентиор яуи, сед аугуе афферт
+симилияуе ад, ех адмодум постулант иус.
+
+Про дицунт волуптатум диспутатиони ат. Вел патриояуе персецути еа, цетерос
+диспутатиони ин сед, нам те веро цлита малуиссет. Цу неглегентур инструцтиор
+интерпретарис еам, ипсум фабулас еи вел. Еи адхуц деленити нам, аугуе
+демоцритум при ан. Вим мелиоре проприае ид, албуциус волуптуа цоррумпит дуо ан.
+Латине иуварет пер ут, иус еа мунере ерипуит санцтус.
+
+Модус тритани иус не, вим ут мелиоре мандамус, лабитур опортере дуо но. Ад нец
+витае фацилис инцоррупте, цу сед толлит сцрипторем. Сит лудус инимицус
+волуптариа не. Иисяуе антиопам сапиентем сед еу. Путент волуптуа сит ех, ат иус
+ребум епицури, яуи моллис елигенди ех. Проприае нолуиссе цу сеа, путент поссит
+адверсариум про не.
+
+Ид яуо прима бонорум, дуо форенсибус яуаерендум еи, еум бруте мунере те. Еам
+риденс граецо ех, аеяуе санцтус маиорум ан вел. Либрис санцтус утрояуе ест но,
+еам ат реяуе порро тинцидунт, ут хинц иллуд патриояуе хис. Не солет оффендит
+форенсибус хас, тамяуам опортеат елаборарет те нец, еу аугуе примис маиорум
+еам. Аутем вениам импедит вис ин, прима елитр пхаедрум ест еу.)");
+}
+
+template <class CharT>
+std::basic_string_view<CharT> japanese_text() {
+  return SV(
+      R"(入ト年媛ろ舗学ラロ準募ケカ社金スノ屋検れう策他セヲシ引口ぎ集7独ぱクふ出車ぽでぱ円輪ルノ受打わ。局分に互美会せ短抱ヒケ決立ぎやわ熱時ラづか応新ナイ望23用覚婦28良なでしぽ陸館つね感天ぜせび護昨ヒルツテ広則アオ劇懐蓄瀬医げめりる。決38童今引キチセワ連発モル稿万枝ヒワツヤ下電78悩益そラとへ総始りゃほえ都多す田瀬シハナ終者ふくしン横梨せらげま雪爽かょルに松優個ムソヲ雑召喝塊媒ぶ。
+
+紙ヤ景異ミノオ誤求レ移著ヤエヨメ広庫テハヌサ君検あ必参ワ火面るね声著ン間売力を数20談すがス禁化ッを。起そり予浩ド進皇キ試属が震二トヌ真佳速すずちし件諏フウチ聞在ス会雄ノミ必筋80戦ぶさほド聞2涙属どスれ映聞ネ掲実べ。
+
+8福びり属稿づ徳鎌ニル涼問ゃごるリ付92済トぎけッ康30業づむはつ治然二生入ざひ有動ハワチ発談ニスツ魚困摘策送ざ。個時着そてら新新ヌ鉄報たは作主ずリ可輸改量ルおず井認つてぜな会大ぼすぶし全戸ノハケレ貯治たざリな祖間ムリキ断会仕べせど。委暮ど象週トクワ流開タハ硬給ツタウ者善マラノヱ断稿リヲ東毎ツヨマ井藤ルょへ境同論エ愛図ッらフリ基38属慣葬8携ヱ校図おに岐題しね要月レユ展省わトど。
+
+担がは顔研リ目問いぽべ挙介ん入番ネヌイ栄県し改治ラス健第モム得続加ホウ嘉宿置首本やぞ。78毎まが現設記ほぜね場歩ユアルヒ東的ヒ姿役ネヲ聞能ラシマヒ際形トくゃ政能万の付結ス国1教レツ引写イど扱澤は膚言けリいべ橋柔薄組こよじ。浩報すンつひ崎正念方と夫地クざす情阪スで抜長ネ娘回ハツ止資ヘニ並辞ロノ展師質18打テネ岡時ノモ泉95務えぴひつ速申後延んフるせ。
+
+店てラ載独マシフ理心ス型部米た読石カ料応掲ケカキ打月在ユテニ採材イ並発イヒト旅錯っめし模能りせば連確え会准揮が。器にト画軍にぶイら式東みそお前姿リいけに身47却6記け岸5体会ゃばま映8碁よぽだ経9名トびち更躍うにふ裏高もそ提旅さぼえス。賞ぞだ月係ソ知建振イナシ説並イ見書傳ヨミ問回級エシ出所師阪ト転権がし渡平ルモケ新完ハ玲女ロトシ導複トうよふ。
+
+化シセチ町74掲ネテトオ連対ヒハチモ経後ッ断連カロワ待業ぼぽねか百都へがい始塗ごげ寺帰んぽ逆力るず選英堂衛掛焼ゅ。自生トサリ探就的らね江球リルスツ主嘆4権伝ざが避掲う慶合ワ百29暮ネヤクム書能部あが席小フア部親票ーむとこ。3説ひっぜ約毎伎ナキリ缶近くなず員45姿えにけろ値付ワ着知ソルキ日医ず集新エウカケ投国チ生目ゃ棋運ぐのか寄募オチ性注経どドんて止代わくかな端期幕はかク。
+)");
+}
+
+template <class CharT>
+std::basic_string_view<CharT> emoji_text() {
+  return SV(
+      R"(
+\U0001F636\u200D\U0001F32B\uFE0F
+\U0001F44B\U0001F3FB\U0001F44B\U0001F3FC\U0001F44B\U0001F3FD\U0001F44B\U0001F3FE\U0001F44B\U0001F3FF
+\U0001F468\u200D\U0001F469\u200D\U0001F467\u200D\U0001F466\U0001F1E8\U0001F1E6
+\U0001F636\u200D\U0001F32B\uFE0F
+\U0001F44B\U0001F3FB\U0001F44B\U0001F3FC\U0001F44B\U0001F3FD\U0001F44B\U0001F3FE\U0001F44B\U0001F3FF
+\U0001F468\u200D\U0001F469\u200D\U0001F467\u200D\U0001F466\U0001F1E8\U0001F1E6
+\U0001F636\u200D\U0001F32B\uFE0F
+\U0001F44B\U0001F3FB\U0001F44B\U0001F3FC\U0001F44B\U0001F3FD\U0001F44B\U0001F3FE\U0001F44B\U0001F3FF
+\U0001F468\u200D\U0001F469\u200D\U0001F467\u200D\U0001F466\U0001F1E8\U0001F1E6
+\U0001F636\u200D\U0001F32B\uFE0F
+\U0001F44B\U0001F3FB\U0001F44B\U0001F3FC\U0001F44B\U0001F3FD\U0001F44B\U0001F3FE\U0001F44B\U0001F3FF
+\U0001F468\u200D\U0001F469\u200D\U0001F467\u200D\U0001F466\U0001F1E8\U0001F1E6
+\U0001F636\u200D\U0001F32B\uFE0F
+\U0001F44B\U0001F3FB\U0001F44B\U0001F3FC\U0001F44B\U0001F3FD\U0001F44B\U0001F3FE\U0001F44B\U0001F3FF
+\U0001F468\u200D\U0001F469\u200D\U0001F467\u200D\U0001F466\U0001F1E8\U0001F1E6
+\U0001F636\u200D\U0001F32B\uFE0F
+\U0001F44B\U0001F3FB\U0001F44B\U0001F3FC\U0001F44B\U0001F3FD\U0001F44B\U0001F3FE\U0001F44B\U0001F3FF
+\U0001F468\u200D\U0001F469\u200D\U0001F467\u200D\U0001F466\U0001F1E8\U0001F1E6
+\U0001F636\u200D\U0001F32B\uFE0F
+\U0001F44B\U0001F3FB\U0001F44B\U0001F3FC\U0001F44B\U0001F3FD\U0001F44B\U0001F3FE\U0001F44B\U0001F3FF
+\U0001F468\u200D\U0001F469\u200D\U0001F467\u200D\U0001F466\U0001F1E8\U0001F1E6
+\U0001F636\u200D\U0001F32B\uFE0F
+\U0001F44B\U0001F3FB\U0001F44B\U0001F3FC\U0001F44B\U0001F3FD\U0001F44B\U0001F3FE\U0001F44B\U0001F3FF
+\U0001F468\u200D\U0001F469\u200D\U0001F467\u200D\U0001F466\U0001F1E8\U0001F1E6
+\U0001F636\u200D\U0001F32B\uFE0F
+\U0001F44B\U0001F3FB\U0001F44B\U0001F3FC\U0001F44B\U0001F3FD\U0001F44B\U0001F3FE\U0001F44B\U0001F3FF
+\U0001F468\u200D\U0001F469\u200D\U0001F467\u200D\U0001F466\U0001F1E8\U0001F1E6
+\U0001F636\u200D\U0001F32B\uFE0F
+\U0001F44B\U0001F3FB\U0001F44B\U0001F3FC\U0001F44B\U0001F3FD\U0001F44B\U0001F3FE\U0001F44B\U0001F3FF
+\U0001F468\u200D\U0001F469\u200D\U0001F467\u200D\U0001F466\U0001F1E8\U0001F1E6
+\U0001F636\u200D\U0001F32B\uFE0F
+\U0001F44B\U0001F3FB\U0001F44B\U0001F3FC\U0001F44B\U0001F3FD\U0001F44B\U0001F3FE\U0001F44B\U0001F3FF
+\U0001F468\u200D\U0001F469\u200D\U0001F467\u200D\U0001F466\U0001F1E8\U0001F1E6
+\U0001F636\u200D\U0001F32B\uFE0F
+\U0001F44B\U0001F3FB\U0001F44B\U0001F3FC\U0001F44B\U0001F3FD\U0001F44B\U0001F3FE\U0001F44B\U0001F3FF
+\U0001F468\u200D\U0001F469\u200D\U0001F467\u200D\U0001F466\U0001F1E8\U0001F1E6
+\U0001F636\u200D\U0001F32B\uFE0F
+\U0001F44B\U0001F3FB\U0001F44B\U0001F3FC\U0001F44B\U0001F3FD\U0001F44B\U0001F3FE\U0001F44B\U0001F3FF
+
+\U0001F468\u200D\U0001F469\u200D\U0001F467\u200D\U0001F466\U0001F1E8\U0001F1E6
+
+\U0001F636\u200D\U0001F32B\uFE0F
+
+\U0001F44B\U0001F3FB\U0001F44B\U0001F3FC\U0001F44B\U0001F3FD\U0001F44B\U0001F3FE\U0001F44B\U0001F3FF
+
+\U0001F468\u200D\U0001F469\u200D\U0001F467\u200D\U0001F466\U0001F1E8\U0001F1E6
+
+\U0001F984
+
+)");
+}
+
+template <class CharT>
+void BM_escaped(benchmark::State& state, std::basic_string_view<CharT> input) {
+  CharT buffer[25'000];
+
+  if constexpr (std::same_as<CharT, char>) {
+    // Make sure the output buffer is large enough.
+    assert(std::formatted_size("{}", input) == 3000);
+    for (auto _ : state)
+      benchmark::DoNotOptimize(std::format_to(buffer, "{:?}", input));
+  } else {
+    for (auto _ : state)
+      benchmark::DoNotOptimize(std::format_to(buffer, L"{:?}", input));
+  }
+}
+
+template <class CharT>
+void BM_ascii_escaped(benchmark::State& state) {
+  BM_escaped(state, ascii_text<CharT>());
+}
+
+template <class CharT>
+void BM_unicode_escaped(benchmark::State& state) {
+  BM_escaped(state, unicode_text<CharT>());
+}
+
+template <class CharT>
+void BM_cyrillic_escaped(benchmark::State& state) {
+  BM_escaped(state, cyrillic_text<CharT>());
+}
+
+template <class CharT>
+void BM_japanese_escaped(benchmark::State& state) {
+  BM_escaped(state, japanese_text<CharT>());
+}
+
+template <class CharT>
+void BM_emoji_escaped(benchmark::State& state) {
+  BM_escaped(state, emoji_text<CharT>());
+}
+
+BENCHMARK_TEMPLATE(BM_ascii_escaped, char);
+BENCHMARK_TEMPLATE(BM_unicode_escaped, char);
+BENCHMARK_TEMPLATE(BM_cyrillic_escaped, char);
+BENCHMARK_TEMPLATE(BM_japanese_escaped, char);
+BENCHMARK_TEMPLATE(BM_emoji_escaped, char);
+
+BENCHMARK_TEMPLATE(BM_ascii_escaped, wchar_t);
+BENCHMARK_TEMPLATE(BM_unicode_escaped, wchar_t);
+BENCHMARK_TEMPLATE(BM_cyrillic_escaped, wchar_t);
+BENCHMARK_TEMPLATE(BM_japanese_escaped, wchar_t);
+BENCHMARK_TEMPLATE(BM_emoji_escaped, wchar_t);
+
+int main(int argc, char** argv) {
+  benchmark::Initialize(&argc, argv);
+  if (benchmark::ReportUnrecognizedArguments(argc, argv))
+    return 1;
+
+  benchmark::RunSpecifiedBenchmarks();
+}
+#else
+int main(int, char**) { return 0; }
+#endif
diff --git a/libcxx/cmake/caches/Generic-cxx20.cmake b/libcxx/cmake/caches/Generic-cxx20.cmake
index 641c131a737b..3c44fdaf0e42 100644
--- a/libcxx/cmake/caches/Generic-cxx20.cmake
+++ b/libcxx/cmake/caches/Generic-cxx20.cmake
@@ -1,3 +1,2 @@
-set(LIBCXX_INSTALL_MODULES ON CACHE BOOL "") # TODO MODULES Remove when enabled automatically.
 set(LIBCXX_TEST_PARAMS "std=c++20" CACHE STRING "")
 set(LIBCXXABI_TEST_PARAMS "${LIBCXX_TEST_PARAMS}" CACHE STRING "")
diff --git a/libcxx/cmake/caches/Generic-cxx23.cmake b/libcxx/cmake/caches/Generic-cxx23.cmake
index f5409e4652e4..bf88abf56ca6 100644
--- a/libcxx/cmake/caches/Generic-cxx23.cmake
+++ b/libcxx/cmake/caches/Generic-cxx23.cmake
@@ -1,3 +1,2 @@
-set(LIBCXX_INSTALL_MODULES ON CACHE BOOL "") # TODO MODULES Remove when enabled automatically.
 set(LIBCXX_TEST_PARAMS "std=c++23" CACHE STRING "")
 set(LIBCXXABI_TEST_PARAMS "${LIBCXX_TEST_PARAMS}" CACHE STRING "")
diff --git a/libcxx/cmake/caches/Generic-cxx26.cmake b/libcxx/cmake/caches/Generic-cxx26.cmake
index 2d9c018a4ff5..6ba9482af578 100644
--- a/libcxx/cmake/caches/Generic-cxx26.cmake
+++ b/libcxx/cmake/caches/Generic-cxx26.cmake
@@ -1,3 +1,2 @@
-set(LIBCXX_INSTALL_MODULES ON CACHE BOOL "") # TODO MODULES Remove when enabled automatically.
 set(LIBCXX_TEST_PARAMS "std=c++26" CACHE STRING "")
 set(LIBCXXABI_TEST_PARAMS "${LIBCXX_TEST_PARAMS}" CACHE STRING "")
diff --git a/libcxx/cmake/caches/Generic-hardening-mode-extensive.cmake b/libcxx/cmake/caches/Generic-hardening-mode-extensive.cmake
index 9542dcdbf778..72263dfd8463 100644
--- a/libcxx/cmake/caches/Generic-hardening-mode-extensive.cmake
+++ b/libcxx/cmake/caches/Generic-hardening-mode-extensive.cmake
@@ -1,2 +1 @@
-set(LIBCXX_INSTALL_MODULES ON CACHE BOOL "") # TODO MODULES Remove when enabled automatically.
 set(LIBCXX_HARDENING_MODE "extensive" CACHE STRING "")
diff --git a/libcxx/cmake/caches/Generic-no-exceptions.cmake b/libcxx/cmake/caches/Generic-no-exceptions.cmake
index c68adfc1276b..f0dffef60dba 100644
--- a/libcxx/cmake/caches/Generic-no-exceptions.cmake
+++ b/libcxx/cmake/caches/Generic-no-exceptions.cmake
@@ -1,3 +1,2 @@
-set(LIBCXX_INSTALL_MODULES ON CACHE BOOL "") # TODO MODULES Remove when enabled automatically.
 set(LIBCXX_ENABLE_EXCEPTIONS OFF CACHE BOOL "")
 set(LIBCXXABI_ENABLE_EXCEPTIONS OFF CACHE BOOL "")
diff --git a/libcxx/cmake/caches/Generic-no-experimental.cmake b/libcxx/cmake/caches/Generic-no-experimental.cmake
index 62b7d7373d44..f33ed0141899 100644
--- a/libcxx/cmake/caches/Generic-no-experimental.cmake
+++ b/libcxx/cmake/caches/Generic-no-experimental.cmake
@@ -1,3 +1,2 @@
-set(LIBCXX_INSTALL_MODULES ON CACHE BOOL "") # TODO MODULES Remove when enabled automatically.
 set(LIBCXX_TEST_PARAMS "enable_experimental=False" CACHE STRING "")
 set(LIBCXXABI_TEST_PARAMS "${LIBCXX_TEST_PARAMS}" CACHE STRING "")
diff --git a/libcxx/cmake/caches/Generic-no-filesystem.cmake b/libcxx/cmake/caches/Generic-no-filesystem.cmake
index 01ae7e68f12c..4000f3a3e8ef 100644
--- a/libcxx/cmake/caches/Generic-no-filesystem.cmake
+++ b/libcxx/cmake/caches/Generic-no-filesystem.cmake
@@ -1,2 +1 @@
-set(LIBCXX_INSTALL_MODULES ON CACHE BOOL "") # TODO MODULES Remove when enabled automatically.
 set(LIBCXX_ENABLE_FILESYSTEM OFF CACHE BOOL "")
diff --git a/libcxx/cmake/caches/Generic-no-localization.cmake b/libcxx/cmake/caches/Generic-no-localization.cmake
index fc4957b2d53a..79d6b44c7139 100644
--- a/libcxx/cmake/caches/Generic-no-localization.cmake
+++ b/libcxx/cmake/caches/Generic-no-localization.cmake
@@ -1,2 +1 @@
-set(LIBCXX_INSTALL_MODULES ON CACHE BOOL "") # TODO MODULES Remove when enabled automatically.
 set(LIBCXX_ENABLE_LOCALIZATION OFF CACHE BOOL "")
diff --git a/libcxx/cmake/caches/Generic-no-random_device.cmake b/libcxx/cmake/caches/Generic-no-random_device.cmake
index ddf479add626..e9b4cc60cc80 100644
--- a/libcxx/cmake/caches/Generic-no-random_device.cmake
+++ b/libcxx/cmake/caches/Generic-no-random_device.cmake
@@ -1,2 +1 @@
-set(LIBCXX_INSTALL_MODULES ON CACHE BOOL "") # TODO MODULES Remove when enabled automatically.
 set(LIBCXX_ENABLE_RANDOM_DEVICE OFF CACHE BOOL "")
diff --git a/libcxx/cmake/caches/Generic-no-threads.cmake b/libcxx/cmake/caches/Generic-no-threads.cmake
index 724fbc466b58..616baef1be7b 100644
--- a/libcxx/cmake/caches/Generic-no-threads.cmake
+++ b/libcxx/cmake/caches/Generic-no-threads.cmake
@@ -1,4 +1,3 @@
-set(LIBCXX_INSTALL_MODULES ON CACHE BOOL "") # TODO MODULES Remove when enabled automatically.
 set(LIBCXX_ENABLE_THREADS OFF CACHE BOOL "")
 set(LIBCXXABI_ENABLE_THREADS OFF CACHE BOOL "")
 set(LIBCXX_ENABLE_MONOTONIC_CLOCK OFF CACHE BOOL "")
diff --git a/libcxx/cmake/caches/Generic-no-unicode.cmake b/libcxx/cmake/caches/Generic-no-unicode.cmake
index a4cf7dd73772..01160bf21898 100644
--- a/libcxx/cmake/caches/Generic-no-unicode.cmake
+++ b/libcxx/cmake/caches/Generic-no-unicode.cmake
@@ -1,2 +1 @@
-set(LIBCXX_INSTALL_MODULES ON CACHE BOOL "") # TODO MODULES Remove when enabled automatically.
 set(LIBCXX_ENABLE_UNICODE OFF CACHE BOOL "")
diff --git a/libcxx/cmake/caches/Generic-no-wide-characters.cmake b/libcxx/cmake/caches/Generic-no-wide-characters.cmake
index dc19389bb5ae..728d41086a38 100644
--- a/libcxx/cmake/caches/Generic-no-wide-characters.cmake
+++ b/libcxx/cmake/caches/Generic-no-wide-characters.cmake
@@ -1,2 +1 @@
-set(LIBCXX_INSTALL_MODULES ON CACHE BOOL "") # TODO MODULES Remove when enabled automatically.
 set(LIBCXX_ENABLE_WIDE_CHARACTERS OFF CACHE BOOL "")
diff --git a/libcxx/docs/BuildingLibcxx.rst b/libcxx/docs/BuildingLibcxx.rst
index a0a0cdb43397..e425b9dadfe7 100644
--- a/libcxx/docs/BuildingLibcxx.rst
+++ b/libcxx/docs/BuildingLibcxx.rst
@@ -208,7 +208,7 @@ libc++ specific options
 
 .. option:: LIBCXX_INSTALL_MODULES:BOOL
 
-  **Default**: ``OFF``
+  **Default**: ``ON``
 
   Toggle the installation of the experimental libc++ module sources.
 
diff --git a/libcxx/docs/Modules.rst b/libcxx/docs/Modules.rst
index 5b027ed1bd07..352a198f3774 100644
--- a/libcxx/docs/Modules.rst
+++ b/libcxx/docs/Modules.rst
@@ -69,8 +69,6 @@ Some of the current limitations
  * The path to the compiler may not be a symlink, ``clang-scan-deps`` does
    not handle that case properly
  * Libc++ is not tested with modules instead of headers
- * Clang supports modules using GNU extensions, but libc++ does not work using
-   GNU extensions.
  * Clang:
     * Including headers after importing the ``std`` module may fail. This is
       hard to solve and there is a work-around by first including all headers
@@ -105,9 +103,17 @@ Users need to be able to build their own BMI files.
    system vendors, with the goal that building the BMI files is done by
    the build system.
 
-Currently this requires a local build of libc++ with modules enabled. Since
-modules are not part of the installation yet, they are used from the build
-directory. First libc++ needs to be build with module support enabled.
+Currently there are two ways to build modules
+
+  * Use a local build of modules from the build directory. This requires
+    Clang 17 or later and CMake 3.26 or later.
+
+  * Use the installed modules. This requires Clang 18.1.2 or later and
+    a recent build of CMake. The CMake changes will be part of CMake 3.30. This
+    method requires you or your distribution to enable module installation.
+
+Using the local build
+~~~~~~~~~~~~~~~~~~~~~
 
 .. code-block:: bash
 
@@ -136,7 +142,7 @@ This is a small sample program that uses the module ``std``. It consists of a
 .. code-block:: cmake
 
   cmake_minimum_required(VERSION 3.26.0 FATAL_ERROR)
-  project("module"
+  project("example"
     LANGUAGES CXX
   )
 
@@ -146,7 +152,6 @@ This is a small sample program that uses the module ``std``. It consists of a
 
   set(CMAKE_CXX_STANDARD 23)
   set(CMAKE_CXX_STANDARD_REQUIRED YES)
-  # Libc++ doesn't support compiler extensions for modules.
   set(CMAKE_CXX_EXTENSIONS OFF)
 
   #
@@ -214,6 +219,64 @@ Building this project is done with the following steps, assuming the files
 
              ``error: module file _deps/std-build/CMakeFiles/std.dir/std.pcm cannot be loaded due to a configuration mismatch with the current compilation [-Wmodule-file-config-mismatch]``
 
+
+Using the installed modules
+~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+CMake has added experimental support for importing the Standard modules. This
+is available in the current nightly builds and will be part of the 3.30
+release. Currently CMake only supports importing the Standard modules in C++23
+and later. Enabling this for C++20 is on the TODO list of the CMake
+developers.
+
+The example uses the same ``main.cpp`` as above. It uses the following
+``CMakeLists.txt``:
+
+.. code-block:: cmake
+
+  # This requires a recent nightly build.
+  # This will be part of CMake 3.30.0.
+  cmake_minimum_required(VERSION 3.29.0 FATAL_ERROR)
+
+  # Enables the Standard module support. This needs to be done
+  # before selecting the languages.
+  set(CMAKE_EXPERIMENTAL_CXX_IMPORT_STD "0e5b6991-d74f-4b3d-a41c-cf096e0b2508")
+  set(CMAKE_CXX_MODULE_STD ON)
+
+  project("example"
+    LANGUAGES CXX
+  )
+
+  #
+  # Set language version used
+  #
+
+  set(CMAKE_CXX_STANDARD 23)
+  set(CMAKE_CXX_STANDARD_REQUIRED YES)
+  # Currently CMake requires extensions enabled when using import std.
+  # https://gitlab.kitware.com/cmake/cmake/-/issues/25916
+  # https://gitlab.kitware.com/cmake/cmake/-/issues/25539
+  set(CMAKE_CXX_EXTENSIONS ON)
+
+  add_executable(main)
+  target_sources(main
+    PRIVATE
+      main.cpp
+  )
+
+Building this project is done with the following steps, assuming the files
+``main.cpp`` and ``CMakeLists.txt`` are copied in the current directory.
+
+.. code-block:: bash
+
+  $ mkdir build
+  $ cmake -G Ninja -S . -B build -DCMAKE_CXX_COMPILER=<path-to-compiler> -DCMAKE_CXX_FLAGS=-stdlib=libc++
+  $ ninja -C build
+  $ build/main
+
+.. warning:: ``<path-to-compiler>`` should point point to the real binary and
+             not to a symlink.
+
 If you have questions about modules feel free to ask them in the ``#libcxx``
 channel on `LLVM's Discord server <https://discord.gg/jzUbyP26tQ>`__.
 
diff --git a/libcxx/docs/ReleaseNotes/19.rst b/libcxx/docs/ReleaseNotes/19.rst
index b466b4cd8140..ac4fd0ecc122 100644
--- a/libcxx/docs/ReleaseNotes/19.rst
+++ b/libcxx/docs/ReleaseNotes/19.rst
@@ -49,6 +49,8 @@ Implemented Papers
 - P2302R4 - ``std::ranges::contains``
 - P1659R3 - ``std::ranges::starts_with`` and ``std::ranges::ends_with``
 - P3029R1 - Better ``mdspan``'s CTAD
+- P2387R3 - Pipe support for user-defined range adaptors
+- P2713R1 - Escaping improvements in ``std::format``
 
 Improvements and New Features
 -----------------------------
@@ -147,3 +149,5 @@ Build System Changes
 
 - The Cmake variable ``LIBCXX_ENABLE_CLANG_TIDY`` has been removed. The build system has been changed
   to automatically detect the presence of ``clang-tidy`` and the required ``Clang`` libraries.
+
+- The CMake options ``LIBCXX_INSTALL_MODULES`` now defaults to ``ON``.
diff --git a/libcxx/docs/Status/Cxx23.rst b/libcxx/docs/Status/Cxx23.rst
index b19ff4fdc0f7..23d30c8128d7 100644
--- a/libcxx/docs/Status/Cxx23.rst
+++ b/libcxx/docs/Status/Cxx23.rst
@@ -43,7 +43,6 @@ Paper Status
    .. [#note-P0533R9] P0533R9: ``isfinite``, ``isinf``, ``isnan`` and ``isnormal`` are implemented.
    .. [#note-P1413R3] P1413R3: ``std::aligned_storage_t`` and ``std::aligned_union_t`` are marked deprecated, but
       clang doesn't issue a diagnostic for deprecated using template declarations.
-   .. [#note-P2387R3] P2387R3: ``bind_back`` only
    .. [#note-P2520R0] P2520R0: Libc++ implemented this paper as a DR in C++20 as well.
    .. [#note-P2711R1] P2711R1: ``join_with_view`` hasn't been done yet since this type isn't implemented yet.
    .. [#note-P2770R0] P2770R0: ``join_with_view`` hasn't been done yet since this type isn't implemented yet.
diff --git a/libcxx/docs/Status/Cxx23Papers.csv b/libcxx/docs/Status/Cxx23Papers.csv
index 065db97a0b0b..01387a404f5d 100644
--- a/libcxx/docs/Status/Cxx23Papers.csv
+++ b/libcxx/docs/Status/Cxx23Papers.csv
@@ -45,7 +45,7 @@
 "`P1413R3 <https://wg21.link/P1413R3>`__","LWG","Deprecate ``std::aligned_storage`` and ``std::aligned_union``","February 2022","|Complete| [#note-P1413R3]_",""
 "`P2255R2 <https://wg21.link/P2255R2>`__","LWG","A type trait to detect reference binding to temporary","February 2022","",""
 "`P2273R3 <https://wg21.link/P2273R3>`__","LWG","Making ``std::unique_ptr`` constexpr","February 2022","|Complete|","16.0"
-"`P2387R3 <https://wg21.link/P2387R3>`__","LWG","Pipe support for user-defined range adaptors","February 2022","|Partial| [#note-P2387R3]_","","|ranges|"
+"`P2387R3 <https://wg21.link/P2387R3>`__","LWG","Pipe support for user-defined range adaptors","February 2022","|Complete|","19.0","|ranges|"
 "`P2440R1 <https://wg21.link/P2440R1>`__","LWG","``ranges::iota``, ``ranges::shift_left`` and ``ranges::shift_right``","February 2022","","","|ranges|"
 "`P2441R2 <https://wg21.link/P2441R2>`__","LWG","``views::join_with``","February 2022","|In Progress|","","|ranges|"
 "`P2442R1 <https://wg21.link/P2442R1>`__","LWG","Windowing range adaptors: ``views::chunk`` and ``views::slide``","February 2022","","","|ranges|"
@@ -108,7 +108,7 @@
 "`P2164R9 <https://wg21.link/P2164R9>`__","LWG", "``views::enumerate``","February 2023","","","|ranges|"
 "`P2711R1 <https://wg21.link/P2711R1>`__","LWG", "Making multi-param constructors of ``views`` ``explicit``","February 2023","|In Progress| [#note-P2711R1]_","","|ranges|"
 "`P2609R3 <https://wg21.link/P2609R3>`__","LWG", "Relaxing Ranges Just A Smidge","February 2023","","","|ranges|"
-"`P2713R1 <https://wg21.link/P2713R1>`__","LWG", "Escaping improvements in ``std::format``","February 2023","","","|format|"
+"`P2713R1 <https://wg21.link/P2713R1>`__","LWG", "Escaping improvements in ``std::format``","February 2023","|Complete|","19.0","|format|"
 "`P2675R1 <https://wg21.link/P2675R1>`__","LWG", "``format``'s width estimation is too approximate and not forward compatible","February 2023","|Complete|","17.0","|format|"
 "`P2572R1 <https://wg21.link/P2572R1>`__","LWG", "``std::format`` fill character allowances","February 2023","|Complete|","17.0","|format|"
 "`P2693R1 <https://wg21.link/P2693R1>`__","LWG", "Formatting ``thread::id`` and ``stacktrace``","February 2023","|Partial| [#note-P2693R1]_","","|format|"
diff --git a/libcxx/docs/Status/Cxx2cIssues.csv b/libcxx/docs/Status/Cxx2cIssues.csv
index 008f7418ab9c..eb99414c48be 100644
--- a/libcxx/docs/Status/Cxx2cIssues.csv
+++ b/libcxx/docs/Status/Cxx2cIssues.csv
@@ -32,7 +32,7 @@
 "`3951 <https://wg21.link/LWG3951>`__","[expected.object.swap]: Using ``value()`` instead of ``has_value()``","Kona November 2023","","",""
 "`3953 <https://wg21.link/LWG3953>`__","``iter_move`` for ``common_iterator`` and ``counted_iterator`` should return ``decltype(auto)``","Kona November 2023","","","|ranges|"
 "`3957 <https://wg21.link/LWG3957>`__","[container.alloc.reqmts] The value category of v should be claimed","Kona November 2023","","",""
-"`3965 <https://wg21.link/LWG3965>`__","Incorrect example in [format.string.escaped] p3 for formatting of combining characters","Kona November 2023","","","|format|"
+"`3965 <https://wg21.link/LWG3965>`__","Incorrect example in [format.string.escaped] p3 for formatting of combining characters","Kona November 2023","|Complete|","19.0","|format|"
 "`3970 <https://wg21.link/LWG3970>`__","[mdspan.syn] Missing definition of ``full_extent_t`` and ``full_extent``","Kona November 2023","","",""
 "`3973 <https://wg21.link/LWG3973>`__","Monadic operations should be ADL-proof","Kona November 2023","","",""
 "`3974 <https://wg21.link/LWG3974>`__","``mdspan::operator[]`` should not copy ``OtherIndexTypes``","Kona November 2023","","",""
@@ -44,12 +44,12 @@
 "`3919 <https://wg21.link/LWG3919>`__","``enumerate_view`` may invoke UB for sized common non-forward underlying ranges","Tokyo March 2024","","","|ranges|"
 "`3950 <https://wg21.link/LWG3950>`__","``std::basic_string_view`` comparison operators are overspecified","Tokyo March 2024","|Complete|","18.0",""
 "`3975 <https://wg21.link/LWG3975>`__","Specializations of ``basic_format_context`` should not be permitted","Tokyo March 2024","|Nothing To Do|","","|format|"
-"`3984 <https://wg21.link/LWG3984>`__","``ranges::to``'s recursion branch may be ill-formed","Tokyo March 2024","","","|ranges|"
+"`3984 <https://wg21.link/LWG3984>`__","``ranges::to``'s recursion branch may be ill-formed","Tokyo March 2024","|Complete|","19.0","|ranges|"
 "`4011 <https://wg21.link/LWG4011>`__","``""Effects: Equivalent to return""`` in ``[span.elem]``","Tokyo March 2024","|Nothing To Do|","",""
 "`4012 <https://wg21.link/LWG4012>`__","``common_view::begin/end`` are missing the ``simple-view`` check","Tokyo March 2024","","","|ranges|"
 "`4013 <https://wg21.link/LWG4013>`__","``lazy_split_view::outer-iterator::value_type`` should not provide default constructor","Tokyo March 2024","","","|ranges|"
 "`4016 <https://wg21.link/LWG4016>`__","container-insertable checks do not match what container-inserter does","Tokyo March 2024","","",""
-"`4023 <https://wg21.link/LWG4023>`__","Preconditions of ``std::basic_streambuf::setg/setp``","Tokyo March 2024","","",""
+"`4023 <https://wg21.link/LWG4023>`__","Preconditions of ``std::basic_streambuf::setg/setp``","Tokyo March 2024","|Complete|","19.0",""
 "`4025 <https://wg21.link/LWG4025>`__","Move assignment operator of ``std::expected<cv void, E>`` should not be conditionally deleted","Tokyo March 2024","","",""
 "`4030 <https://wg21.link/LWG4030>`__","Clarify whether arithmetic expressions in ``[numeric.sat.func]`` are mathematical or C++","Tokyo March 2024","|Nothing To Do|","",""
 "`4031 <https://wg21.link/LWG4031>`__","``bad_expected_access<void>`` member functions should be ``noexcept``","Tokyo March 2024","|Complete|","16.0",""
diff --git a/libcxx/docs/Status/FormatIssues.csv b/libcxx/docs/Status/FormatIssues.csv
index 7da77def92da..3780c1ed5c12 100644
--- a/libcxx/docs/Status/FormatIssues.csv
+++ b/libcxx/docs/Status/FormatIssues.csv
@@ -10,7 +10,7 @@ Number,Name,Standard,Assignee,Status,First released version
 "`P2508R1 <https://wg21.link/P2508R1>`__","Exposing ``std::basic-format-string``","C++23","Mark de Wever","|Complete|",15.0
 "`P2585R0 <https://wg21.link/P2585R0>`__","Improving default container formatting","C++23","Mark de Wever","|Complete|",17.0
 "`P2539R4 <https://wg21.link/P2539R4>`__","Should the output of ``std::print`` to a terminal be synchronized with the underlying stream?","C++23","Mark de Wever","|Complete|","18.0"
-"`P2713R1 <https://wg21.link/P2713R1>`__","Escaping improvements in ``std::format``","C++23","Mark de Wever",""
+"`P2713R1 <https://wg21.link/P2713R1>`__","Escaping improvements in ``std::format``","C++23","Mark de Wever","|Complete|",19.0
 "`P2675R1 <https://wg21.link/P2675R1>`__","``format``'s width estimation is too approximate and not forward compatible","C++23","Mark de Wever","|Complete|",17.0
 "`P2572R1 <https://wg21.link/P2572R1>`__","``std::format`` fill character allowances","C++23","Mark de Wever","|Complete|",17.0
 "`P2693R1 <https://wg21.link/P2693R1>`__","Formatting ``thread::id`` and ``stacktrace``","C++23","Mark de Wever","|In Progress|"
diff --git a/libcxx/docs/Status/RangesMajorFeatures.csv b/libcxx/docs/Status/RangesMajorFeatures.csv
index c0bec8d924e8..d00fbce9edf4 100644
--- a/libcxx/docs/Status/RangesMajorFeatures.csv
+++ b/libcxx/docs/Status/RangesMajorFeatures.csv
@@ -1,5 +1,5 @@
 Standard,Name,Assignee,CL,Status
 C++23,`ranges::to <https://wg21.link/P1206R7>`_,Konstantin Varlamov,`D142335 <https://reviews.llvm.org/D142335>`_,Complete
-C++23,`Pipe support for user-defined range adaptors <https://wg21.link/P2387R3>`_,Unassigned,No patch yet,Not started
+C++23,`Pipe support for user-defined range adaptors <https://wg21.link/P2387R3>`_,"Louis Dionne, Jakub Mazurkiewicz, and Xiaoyang Liu",Various,Complete
 C++23,`Formatting Ranges <https://wg21.link/P2286R8>`_,Mark de Wever,Various,Complete
 C++20,`Stashing stashing iterators for proper flattening <https://wg21.link/P2770R0>`_,Jakub Mazurkiewicz,Various,In progress
diff --git a/libcxx/include/__algorithm/find.h b/libcxx/include/__algorithm/find.h
index d60356873132..7f58dbb13a57 100644
--- a/libcxx/include/__algorithm/find.h
+++ b/libcxx/include/__algorithm/find.h
@@ -43,7 +43,7 @@ _LIBCPP_BEGIN_NAMESPACE_STD
 // generic implementation
 template <class _Iter, class _Sent, class _Tp, class _Proj>
 _LIBCPP_HIDE_FROM_ABI _LIBCPP_CONSTEXPR_SINCE_CXX14 _Iter
-__find_impl(_Iter __first, _Sent __last, const _Tp& __value, _Proj& __proj) {
+__find(_Iter __first, _Sent __last, const _Tp& __value, _Proj& __proj) {
   for (; __first != __last; ++__first)
     if (std::__invoke(__proj, *__first) == __value)
       break;
@@ -57,8 +57,7 @@ template <class _Tp,
           __enable_if_t<__is_identity<_Proj>::value && __libcpp_is_trivially_equality_comparable<_Tp, _Up>::value &&
                             sizeof(_Tp) == 1,
                         int> = 0>
-_LIBCPP_HIDE_FROM_ABI _LIBCPP_CONSTEXPR_SINCE_CXX14 _Tp*
-__find_impl(_Tp* __first, _Tp* __last, const _Up& __value, _Proj&) {
+_LIBCPP_HIDE_FROM_ABI _LIBCPP_CONSTEXPR_SINCE_CXX14 _Tp* __find(_Tp* __first, _Tp* __last, const _Up& __value, _Proj&) {
   if (auto __ret = std::__constexpr_memchr(__first, __value, __last - __first))
     return __ret;
   return __last;
@@ -71,8 +70,7 @@ template <class _Tp,
           __enable_if_t<__is_identity<_Proj>::value && __libcpp_is_trivially_equality_comparable<_Tp, _Up>::value &&
                             sizeof(_Tp) == sizeof(wchar_t) && _LIBCPP_ALIGNOF(_Tp) >= _LIBCPP_ALIGNOF(wchar_t),
                         int> = 0>
-_LIBCPP_HIDE_FROM_ABI _LIBCPP_CONSTEXPR_SINCE_CXX14 _Tp*
-__find_impl(_Tp* __first, _Tp* __last, const _Up& __value, _Proj&) {
+_LIBCPP_HIDE_FROM_ABI _LIBCPP_CONSTEXPR_SINCE_CXX14 _Tp* __find(_Tp* __first, _Tp* __last, const _Up& __value, _Proj&) {
   if (auto __ret = std::__constexpr_wmemchr(__first, __value, __last - __first))
     return __ret;
   return __last;
@@ -89,10 +87,10 @@ template <class _Tp,
                             is_signed<_Tp>::value == is_signed<_Up>::value,
                         int> = 0>
 _LIBCPP_HIDE_FROM_ABI _LIBCPP_CONSTEXPR_SINCE_CXX14 _Tp*
-__find_impl(_Tp* __first, _Tp* __last, const _Up& __value, _Proj& __proj) {
+__find(_Tp* __first, _Tp* __last, const _Up& __value, _Proj& __proj) {
   if (__value < numeric_limits<_Tp>::min() || __value > numeric_limits<_Tp>::max())
     return __last;
-  return std::__find_impl(__first, __last, _Tp(__value), __proj);
+  return std::__find(__first, __last, _Tp(__value), __proj);
 }
 
 // __bit_iterator implementation
@@ -134,7 +132,7 @@ __find_bool(__bit_iterator<_Cp, _IsConst> __first, typename _Cp::size_type __n)
 
 template <class _Cp, bool _IsConst, class _Tp, class _Proj, __enable_if_t<__is_identity<_Proj>::value, int> = 0>
 inline _LIBCPP_HIDE_FROM_ABI _LIBCPP_CONSTEXPR_SINCE_CXX20 __bit_iterator<_Cp, _IsConst>
-__find_impl(__bit_iterator<_Cp, _IsConst> __first, __bit_iterator<_Cp, _IsConst> __last, const _Tp& __value, _Proj&) {
+__find(__bit_iterator<_Cp, _IsConst> __first, __bit_iterator<_Cp, _IsConst> __last, const _Tp& __value, _Proj&) {
   if (static_cast<bool>(__value))
     return std::__find_bool<true>(__first, static_cast<typename _Cp::size_type>(__last - __first));
   return std::__find_bool<false>(__first, static_cast<typename _Cp::size_type>(__last - __first));
@@ -150,7 +148,7 @@ template <class _SegmentedIterator,
           class _Proj,
           __enable_if_t<__is_segmented_iterator<_SegmentedIterator>::value, int> = 0>
 _LIBCPP_HIDE_FROM_ABI _LIBCPP_CONSTEXPR_SINCE_CXX14 _SegmentedIterator
-__find_impl(_SegmentedIterator __first, _SegmentedIterator __last, const _Tp& __value, _Proj& __proj) {
+__find(_SegmentedIterator __first, _SegmentedIterator __last, const _Tp& __value, _Proj& __proj) {
   return std::__find_segment_if(std::move(__first), std::move(__last), __find_segment<_Tp>(__value), __proj);
 }
 
@@ -163,7 +161,7 @@ struct __find_segment {
   template <class _InputIterator, class _Proj>
   _LIBCPP_HIDE_FROM_ABI _LIBCPP_CONSTEXPR _InputIterator
   operator()(_InputIterator __first, _InputIterator __last, _Proj& __proj) const {
-    return std::__find_impl(__first, __last, __value_, __proj);
+    return std::__find(__first, __last, __value_, __proj);
   }
 };
 
@@ -173,7 +171,7 @@ _LIBCPP_NODISCARD inline _LIBCPP_HIDE_FROM_ABI _LIBCPP_CONSTEXPR_SINCE_CXX20 _In
 find(_InputIterator __first, _InputIterator __last, const _Tp& __value) {
   __identity __proj;
   return std::__rewrap_iter(
-      __first, std::__find_impl(std::__unwrap_iter(__first), std::__unwrap_iter(__last), __value, __proj));
+      __first, std::__find(std::__unwrap_iter(__first), std::__unwrap_iter(__last), __value, __proj));
 }
 
 _LIBCPP_END_NAMESPACE_STD
diff --git a/libcxx/include/__algorithm/ranges_find.h b/libcxx/include/__algorithm/ranges_find.h
index e1383eb4b071..6b0d5efe37ab 100644
--- a/libcxx/include/__algorithm/ranges_find.h
+++ b/libcxx/include/__algorithm/ranges_find.h
@@ -44,9 +44,9 @@ struct __fn {
     if constexpr (forward_iterator<_Iter>) {
       auto [__first_un, __last_un] = std::__unwrap_range(__first, std::move(__last));
       return std::__rewrap_range<_Sent>(
-          std::move(__first), std::__find_impl(std::move(__first_un), std::move(__last_un), __value, __proj));
+          std::move(__first), std::__find(std::move(__first_un), std::move(__last_un), __value, __proj));
     } else {
-      return std::__find_impl(std::move(__first), std::move(__last), __value, __proj);
+      return std::__find(std::move(__first), std::move(__last), __value, __proj);
     }
   }
 
diff --git a/libcxx/include/__availability b/libcxx/include/__availability
index aa761eb5bfe5..7a02ae00846b 100644
--- a/libcxx/include/__availability
+++ b/libcxx/include/__availability
@@ -28,30 +28,32 @@
 // that previously released library. Normally, this would be a load-time error
 // when one tries to launch the program against the older library.
 //
-// For example, the filesystem library was introduced in the dylib in macOS 10.15.
-// If a user compiles on a macOS 10.15 host but targets macOS 10.13 with their
-// program, the compiler would normally not complain (because the required
-// declarations are in the headers), but the dynamic loader would fail to find
-// the symbols when actually trying to launch the program on macOS 10.13. To
-// turn this into a compile-time issue instead, declarations are annotated with
-// when they were introduced, and the compiler can produce a diagnostic if the
-// program references something that isn't available on the deployment target.
+// For example, the filesystem library was introduced in the dylib in LLVM 9.
+// On Apple platforms, this corresponds to macOS 10.15. If a user compiles on
+// a macOS 10.15 host but targets macOS 10.13 with their program, the compiler
+// would normally not complain (because the required declarations are in the
+// headers), but the dynamic loader would fail to find the symbols when actually
+// trying to launch the program on macOS 10.13. To turn this into a compile-time
+// issue instead, declarations are annotated with when they were introduced, and
+// the compiler can produce a diagnostic if the program references something that
+// isn't available on the deployment target.
 //
 // This mechanism is general in nature, and any vendor can add their markup to
 // the library (see below). Whenever a new feature is added that requires support
 // in the shared library, two macros are added below to allow marking the feature
 // as unavailable:
-// 1. A macro named `_LIBCPP_AVAILABILITY_HAS_NO_<feature>` which must be defined
-//    exactly when compiling for a target that doesn't support the feature.
-// 2. A macro named `_LIBCPP_AVAILABILITY_<feature>`, which must always be defined
-//    and must expand to the proper availability attribute for the platform.
+// 1. A macro named `_LIBCPP_AVAILABILITY_HAS_<feature>` which must be defined
+//    to `_LIBCPP_INTRODUCED_IN_<version>` for the appropriate LLVM version.
+// 2. A macro named `_LIBCPP_AVAILABILITY_<feature>`, which must be defined to
+//    `_LIBCPP_INTRODUCED_IN_<version>_MARKUP` for the appropriate LLVM version.
 //
 // When vendors decide to ship the feature as part of their shared library, they
-// can update these macros appropriately for their platform, and the library will
-// use those to provide an optimal user experience.
+// can update the `_LIBCPP_INTRODUCED_IN_<version>` macro (and the markup counterpart)
+// based on the platform version they shipped that version of LLVM in. The library
+// will then use this markup to provide an optimal user experience on these platforms.
 //
 // Furthermore, many features in the standard library have corresponding
-// feature-test macros. The `_LIBCPP_AVAILABILITY_HAS_NO_<feature>` macros
+// feature-test macros. The `_LIBCPP_AVAILABILITY_HAS_<feature>` macros
 // are checked by the corresponding feature-test macros generated by
 // generate_feature_test_macro_components.py to ensure that the library
 // doesn't announce a feature as being implemented if it is unavailable on
@@ -74,237 +76,181 @@
 
 // Availability markup is disabled when building the library, or when a non-Clang
 // compiler is used because only Clang supports the necessary attributes.
-// doesn't support the proper attributes.
 #if defined(_LIBCPP_BUILDING_LIBRARY) || defined(_LIBCXXABI_BUILDING_LIBRARY) || !defined(_LIBCPP_COMPILER_CLANG_BASED)
 #  if !defined(_LIBCPP_HAS_NO_VENDOR_AVAILABILITY_ANNOTATIONS)
 #    define _LIBCPP_HAS_NO_VENDOR_AVAILABILITY_ANNOTATIONS
 #  endif
 #endif
 
+// When availability annotations are disabled, we take for granted that features introduced
+// in all versions of the library are available.
 #if defined(_LIBCPP_HAS_NO_VENDOR_AVAILABILITY_ANNOTATIONS)
 
-// These macros control the availability of std::bad_optional_access and
-// other exception types. These were put in the shared library to prevent
-// code bloat from every user program defining the vtable for these exception
-// types.
-//
-// Note that when exceptions are disabled, the methods that normally throw
-// these exceptions can be used even on older deployment targets, but those
-// methods will abort instead of throwing.
-#  define _LIBCPP_AVAILABILITY_HAS_BAD_OPTIONAL_ACCESS 1
-#  define _LIBCPP_AVAILABILITY_BAD_OPTIONAL_ACCESS
-
-#  define _LIBCPP_AVAILABILITY_HAS_BAD_VARIANT_ACCESS 1
-#  define _LIBCPP_AVAILABILITY_BAD_VARIANT_ACCESS
+#  define _LIBCPP_INTRODUCED_IN_LLVM_4 1
+#  define _LIBCPP_INTRODUCED_IN_LLVM_4_MARKUP /* nothing */
 
-#  define _LIBCPP_AVAILABILITY_HAS_BAD_ANY_CAST 1
-#  define _LIBCPP_AVAILABILITY_BAD_ANY_CAST
+#  define _LIBCPP_INTRODUCED_IN_LLVM_9 1
+#  define _LIBCPP_INTRODUCED_IN_LLVM_9_MARKUP      /* nothing */
+#  define _LIBCPP_INTRODUCED_IN_LLVM_9_MARKUP_PUSH /* nothing */
+#  define _LIBCPP_INTRODUCED_IN_LLVM_9_MARKUP_POP  /* nothing */
 
-// These macros control the availability of all parts of <filesystem> that
-// depend on something in the dylib.
-#  define _LIBCPP_AVAILABILITY_HAS_FILESYSTEM_LIBRARY 1
-#  define _LIBCPP_AVAILABILITY_FILESYSTEM_LIBRARY
-#  define _LIBCPP_AVAILABILITY_FILESYSTEM_LIBRARY_PUSH
-#  define _LIBCPP_AVAILABILITY_FILESYSTEM_LIBRARY_POP
+#  define _LIBCPP_INTRODUCED_IN_LLVM_10 1
+#  define _LIBCPP_INTRODUCED_IN_LLVM_10_MARKUP /* nothing */
 
-// This controls the availability of the C++20 synchronization library,
-// which requires shared library support for various operations
-// (see libcxx/src/atomic.cpp). This includes <barier>, <latch>,
-// <semaphore>, and notification functions on std::atomic.
-#  define _LIBCPP_AVAILABILITY_HAS_SYNC 1
-#  define _LIBCPP_AVAILABILITY_SYNC
+#  define _LIBCPP_INTRODUCED_IN_LLVM_12 1
+#  define _LIBCPP_INTRODUCED_IN_LLVM_12_MARKUP /* nothing */
 
-// Enable additional explicit instantiations of iostreams components. This
-// reduces the number of weak definitions generated in programs that use
-// iostreams by providing a single strong definition in the shared library.
-//
-// TODO: Enable additional explicit instantiations on GCC once it supports exclude_from_explicit_instantiation,
-//       or once libc++ doesn't use the attribute anymore.
-// TODO: Enable them on Windows once https://llvm.org/PR41018 has been fixed.
-#  if !defined(_LIBCPP_COMPILER_GCC) && !defined(_WIN32)
-#    define _LIBCPP_AVAILABILITY_HAS_ADDITIONAL_IOSTREAM_EXPLICIT_INSTANTIATIONS_1 1
-#  else
-#    define _LIBCPP_AVAILABILITY_HAS_ADDITIONAL_IOSTREAM_EXPLICIT_INSTANTIATIONS_1 0
-#  endif
+#  define _LIBCPP_INTRODUCED_IN_LLVM_14 1
+#  define _LIBCPP_INTRODUCED_IN_LLVM_14_MARKUP /* nothing */
 
-// This controls the availability of floating-point std::to_chars functions.
-// These overloads were added later than the integer overloads.
-#  define _LIBCPP_AVAILABILITY_HAS_TO_CHARS_FLOATING_POINT 1
-#  define _LIBCPP_AVAILABILITY_TO_CHARS_FLOATING_POINT
+#  define _LIBCPP_INTRODUCED_IN_LLVM_15 1
+#  define _LIBCPP_INTRODUCED_IN_LLVM_15_MARKUP /* nothing */
 
-// This controls whether the library claims to provide a default verbose
-// termination function, and consequently whether the headers will try
-// to use it when the mechanism isn't overriden at compile-time.
-#  define _LIBCPP_AVAILABILITY_HAS_VERBOSE_ABORT 1
-#  define _LIBCPP_AVAILABILITY_VERBOSE_ABORT
+#  define _LIBCPP_INTRODUCED_IN_LLVM_16 1
+#  define _LIBCPP_INTRODUCED_IN_LLVM_16_MARKUP /* nothing */
 
-// This controls the availability of the C++17 std::pmr library,
-// which is implemented in large part in the built library.
-#  define _LIBCPP_AVAILABILITY_HAS_PMR 1
-#  define _LIBCPP_AVAILABILITY_PMR
+#  define _LIBCPP_INTRODUCED_IN_LLVM_18 1
+#  define _LIBCPP_INTRODUCED_IN_LLVM_18_MARKUP /* nothing */
 
-// These macros controls the availability of __cxa_init_primary_exception
-// in the built library, which std::make_exception_ptr might use
-// (see libcxx/include/__exception/exception_ptr.h).
-#  define _LIBCPP_AVAILABILITY_HAS_INIT_PRIMARY_EXCEPTION 1
-#  define _LIBCPP_AVAILABILITY_INIT_PRIMARY_EXCEPTION
-
-// This controls the availability of C++23 <print>, which
-// has a dependency on the built library (it needs access to
-// the underlying buffer types of std::cout, std::cerr, and std::clog.
-#  define _LIBCPP_AVAILABILITY_HAS_PRINT 1
-#  define _LIBCPP_AVAILABILITY_PRINT
-
-// This controls the availability of the C++20 time zone database.
-// The parser code is built in the library.
-#  define _LIBCPP_AVAILABILITY_HAS_TZDB 1
-#  define _LIBCPP_AVAILABILITY_TZDB
-
-// These macros determine whether we assume that std::bad_function_call and
-// std::bad_expected_access provide a key function in the dylib. This allows
-// centralizing their vtable and typeinfo instead of having all TUs provide
-// a weak definition that then gets deduplicated.
-#  define _LIBCPP_AVAILABILITY_HAS_BAD_FUNCTION_CALL_KEY_FUNCTION 1
-#  define _LIBCPP_AVAILABILITY_BAD_FUNCTION_CALL_KEY_FUNCTION
-#  define _LIBCPP_AVAILABILITY_HAS_BAD_EXPECTED_ACCESS_KEY_FUNCTION 1
-#  define _LIBCPP_AVAILABILITY_BAD_EXPECTED_ACCESS_KEY_FUNCTION
+#  define _LIBCPP_INTRODUCED_IN_LLVM_19 1
+#  define _LIBCPP_INTRODUCED_IN_LLVM_19_MARKUP /* nothing */
 
 #elif defined(__APPLE__)
 
-#  define _LIBCPP_AVAILABILITY_HAS_BAD_OPTIONAL_ACCESS                                                                 \
-    (!defined(__ENVIRONMENT_WATCH_OS_VERSION_MIN_REQUIRED__) || __ENVIRONMENT_WATCH_OS_VERSION_MIN_REQUIRED__ >= 50000)
-
-#  define _LIBCPP_AVAILABILITY_HAS_BAD_VARIANT_ACCESS _LIBCPP_AVAILABILITY_HAS_BAD_OPTIONAL_ACCESS
-#  define _LIBCPP_AVAILABILITY_HAS_BAD_ANY_CAST _LIBCPP_AVAILABILITY_HAS_BAD_OPTIONAL_ACCESS
-
-#  define _LIBCPP_AVAILABILITY_BAD_OPTIONAL_ACCESS __attribute__((availability(watchos, strict, introduced = 5.0)))
-#  define _LIBCPP_AVAILABILITY_BAD_VARIANT_ACCESS _LIBCPP_AVAILABILITY_BAD_OPTIONAL_ACCESS
-#  define _LIBCPP_AVAILABILITY_BAD_ANY_CAST _LIBCPP_AVAILABILITY_BAD_OPTIONAL_ACCESS
-
-// TODO: Update once this is released
-#  define _LIBCPP_AVAILABILITY_HAS_INIT_PRIMARY_EXCEPTION 0
-#  define _LIBCPP_AVAILABILITY_INIT_PRIMARY_EXCEPTION __attribute__((unavailable))
+// LLVM 4
+#  if defined(__ENVIRONMENT_WATCH_OS_VERSION_MIN_REQUIRED__) && __ENVIRONMENT_WATCH_OS_VERSION_MIN_REQUIRED__ < 50000
+#    define _LIBCPP_INTRODUCED_IN_LLVM_4 0
+#    define _LIBCPP_INTRODUCED_IN_LLVM_4_MARKUP __attribute__((availability(watchos, strict, introduced = 5.0)))
+#  else
+#    define _LIBCPP_INTRODUCED_IN_LLVM_4 1
+#    define _LIBCPP_INTRODUCED_IN_LLVM_4_MARKUP /* nothing */
+#  endif
 
-// <filesystem>
+// LLVM 9
 // clang-format off
 #  if (defined(__ENVIRONMENT_MAC_OS_X_VERSION_MIN_REQUIRED__) && __ENVIRONMENT_MAC_OS_X_VERSION_MIN_REQUIRED__ < 101500) ||   \
       (defined(__ENVIRONMENT_IPHONE_OS_VERSION_MIN_REQUIRED__) && __ENVIRONMENT_IPHONE_OS_VERSION_MIN_REQUIRED__ < 130000) || \
       (defined(__ENVIRONMENT_TV_OS_VERSION_MIN_REQUIRED__) && __ENVIRONMENT_TV_OS_VERSION_MIN_REQUIRED__ < 130000) ||         \
       (defined(__ENVIRONMENT_WATCH_OS_VERSION_MIN_REQUIRED__) && __ENVIRONMENT_WATCH_OS_VERSION_MIN_REQUIRED__ < 60000)
 // clang-format on
-#    define _LIBCPP_AVAILABILITY_HAS_FILESYSTEM_LIBRARY 0
+#    define _LIBCPP_INTRODUCED_IN_LLVM_9 0
+#    define _LIBCPP_INTRODUCED_IN_LLVM_9_MARKUP                                                                        \
+      __attribute__((availability(macos, strict, introduced = 10.15)))                                                 \
+      __attribute__((availability(ios, strict, introduced = 13.0)))                                                    \
+      __attribute__((availability(tvos, strict, introduced = 13.0)))                                                   \
+      __attribute__((availability(watchos, strict, introduced = 6.0)))
+// clang-format off
+#    define _LIBCPP_INTRODUCED_IN_LLVM_9_MARKUP_PUSH                                                                               \
+      _Pragma("clang attribute push(__attribute__((availability(macos,strict,introduced=10.15))), apply_to=any(function,record))") \
+      _Pragma("clang attribute push(__attribute__((availability(ios,strict,introduced=13.0))), apply_to=any(function,record))")    \
+      _Pragma("clang attribute push(__attribute__((availability(tvos,strict,introduced=13.0))), apply_to=any(function,record))")   \
+      _Pragma("clang attribute push(__attribute__((availability(watchos,strict,introduced=6.0))), apply_to=any(function,record))")
+#    define _LIBCPP_INTRODUCED_IN_LLVM_9_MARKUP_POP                                                                    \
+      _Pragma("clang attribute pop") \
+      _Pragma("clang attribute pop") \
+      _Pragma("clang attribute pop") \
+      _Pragma("clang attribute pop")
+// clang-format on
 #  else
-#    define _LIBCPP_AVAILABILITY_HAS_FILESYSTEM_LIBRARY 1
+#    define _LIBCPP_INTRODUCED_IN_LLVM_9 1
+#    define _LIBCPP_INTRODUCED_IN_LLVM_9_MARKUP      /* nothing */
+#    define _LIBCPP_INTRODUCED_IN_LLVM_9_MARKUP_PUSH /* nothing */
+#    define _LIBCPP_INTRODUCED_IN_LLVM_9_MARKUP_POP  /* nothing */
 #  endif
-#  define _LIBCPP_AVAILABILITY_FILESYSTEM_LIBRARY                                                                      \
-    __attribute__((availability(macos, strict, introduced = 10.15)))                                                   \
-    __attribute__((availability(ios, strict, introduced = 13.0)))                                                      \
-    __attribute__((availability(tvos, strict, introduced = 13.0)))                                                     \
-    __attribute__((availability(watchos, strict, introduced = 6.0)))
+
+// LLVM 10
 // clang-format off
-#   define _LIBCPP_AVAILABILITY_FILESYSTEM_LIBRARY_PUSH                                 \
-        _Pragma("clang attribute push(__attribute__((availability(macos,strict,introduced=10.15))), apply_to=any(function,record))") \
-        _Pragma("clang attribute push(__attribute__((availability(ios,strict,introduced=13.0))), apply_to=any(function,record))")    \
-        _Pragma("clang attribute push(__attribute__((availability(tvos,strict,introduced=13.0))), apply_to=any(function,record))")   \
-        _Pragma("clang attribute push(__attribute__((availability(watchos,strict,introduced=6.0))), apply_to=any(function,record))")
-#   define _LIBCPP_AVAILABILITY_FILESYSTEM_LIBRARY_POP                                  \
-        _Pragma("clang attribute pop")                                          \
-        _Pragma("clang attribute pop")                                          \
-        _Pragma("clang attribute pop")                                          \
-        _Pragma("clang attribute pop")
+#  if (defined(__ENVIRONMENT_MAC_OS_X_VERSION_MIN_REQUIRED__) && __ENVIRONMENT_MAC_OS_X_VERSION_MIN_REQUIRED__ < 110000) ||   \
+      (defined(__ENVIRONMENT_IPHONE_OS_VERSION_MIN_REQUIRED__) && __ENVIRONMENT_IPHONE_OS_VERSION_MIN_REQUIRED__ < 140000) || \
+      (defined(__ENVIRONMENT_TV_OS_VERSION_MIN_REQUIRED__) && __ENVIRONMENT_TV_OS_VERSION_MIN_REQUIRED__ < 140000) ||         \
+      (defined(__ENVIRONMENT_WATCH_OS_VERSION_MIN_REQUIRED__) && __ENVIRONMENT_WATCH_OS_VERSION_MIN_REQUIRED__ < 70000)
 // clang-format on
+#    define _LIBCPP_INTRODUCED_IN_LLVM_10 0
+#    define _LIBCPP_INTRODUCED_IN_LLVM_10_MARKUP                                                                       \
+      __attribute__((availability(macos, strict, introduced = 11.0)))                                                  \
+      __attribute__((availability(ios, strict, introduced = 14.0)))                                                    \
+      __attribute__((availability(tvos, strict, introduced = 14.0)))                                                   \
+      __attribute__((availability(watchos, strict, introduced = 7.0)))
+#  else
+#    define _LIBCPP_INTRODUCED_IN_LLVM_10 1
+#    define _LIBCPP_INTRODUCED_IN_LLVM_10_MARKUP /* nothing */
+#  endif
 
-// std::to_chars(floating-point)
+// LLVM 12
 // clang-format off
-#  if (defined(__ENVIRONMENT_MAC_OS_X_VERSION_MIN_REQUIRED__) && __ENVIRONMENT_MAC_OS_X_VERSION_MIN_REQUIRED__ < 130300) ||   \
-      (defined(__ENVIRONMENT_IPHONE_OS_VERSION_MIN_REQUIRED__) && __ENVIRONMENT_IPHONE_OS_VERSION_MIN_REQUIRED__ < 160300) || \
-      (defined(__ENVIRONMENT_TV_OS_VERSION_MIN_REQUIRED__) && __ENVIRONMENT_TV_OS_VERSION_MIN_REQUIRED__ < 160300) ||         \
-      (defined(__ENVIRONMENT_WATCH_OS_VERSION_MIN_REQUIRED__) && __ENVIRONMENT_WATCH_OS_VERSION_MIN_REQUIRED__ < 90300)
+#  if (defined(__ENVIRONMENT_MAC_OS_X_VERSION_MIN_REQUIRED__) && __ENVIRONMENT_MAC_OS_X_VERSION_MIN_REQUIRED__ < 120000)   || \
+      (defined(__ENVIRONMENT_IPHONE_OS_VERSION_MIN_REQUIRED__) && __ENVIRONMENT_IPHONE_OS_VERSION_MIN_REQUIRED__ < 150000) || \
+      (defined(__ENVIRONMENT_TV_OS_VERSION_MIN_REQUIRED__) && __ENVIRONMENT_TV_OS_VERSION_MIN_REQUIRED__ < 150000)         || \
+      (defined(__ENVIRONMENT_WATCH_OS_VERSION_MIN_REQUIRED__) && __ENVIRONMENT_WATCH_OS_VERSION_MIN_REQUIRED__ < 80000)
 // clang-format on
-#    define _LIBCPP_AVAILABILITY_HAS_TO_CHARS_FLOATING_POINT 0
+#    define _LIBCPP_INTRODUCED_IN_LLVM_12 0
+#    define _LIBCPP_INTRODUCED_IN_LLVM_12_MARKUP                                                                       \
+      __attribute__((availability(macos, strict, introduced = 12.0)))                                                  \
+      __attribute__((availability(ios, strict, introduced = 15.0)))                                                    \
+      __attribute__((availability(tvos, strict, introduced = 15.0)))                                                   \
+      __attribute__((availability(watchos, strict, introduced = 8.0)))
 #  else
-#    define _LIBCPP_AVAILABILITY_HAS_TO_CHARS_FLOATING_POINT 1
+#    define _LIBCPP_INTRODUCED_IN_LLVM_12 1
+#    define _LIBCPP_INTRODUCED_IN_LLVM_12_MARKUP /* nothing */
 #  endif
-#  define _LIBCPP_AVAILABILITY_TO_CHARS_FLOATING_POINT                                                                 \
-    __attribute__((availability(macos, strict, introduced = 13.3)))                                                    \
-    __attribute__((availability(ios, strict, introduced = 16.3)))                                                      \
-    __attribute__((availability(tvos, strict, introduced = 16.3)))                                                     \
-    __attribute__((availability(watchos, strict, introduced = 9.3)))
 
-// c++20 synchronization library
+// LLVM 14
 // clang-format off
-#   if (defined(__ENVIRONMENT_MAC_OS_X_VERSION_MIN_REQUIRED__) && __ENVIRONMENT_MAC_OS_X_VERSION_MIN_REQUIRED__ < 110000) ||   \
-       (defined(__ENVIRONMENT_IPHONE_OS_VERSION_MIN_REQUIRED__) && __ENVIRONMENT_IPHONE_OS_VERSION_MIN_REQUIRED__ < 140000) || \
-       (defined(__ENVIRONMENT_TV_OS_VERSION_MIN_REQUIRED__) && __ENVIRONMENT_TV_OS_VERSION_MIN_REQUIRED__ < 140000) ||         \
-       (defined(__ENVIRONMENT_WATCH_OS_VERSION_MIN_REQUIRED__) && __ENVIRONMENT_WATCH_OS_VERSION_MIN_REQUIRED__ < 70000)
+#  if (defined(__ENVIRONMENT_MAC_OS_X_VERSION_MIN_REQUIRED__) && __ENVIRONMENT_MAC_OS_X_VERSION_MIN_REQUIRED__ < 130400) ||   \
+      (defined(__ENVIRONMENT_IPHONE_OS_VERSION_MIN_REQUIRED__) && __ENVIRONMENT_IPHONE_OS_VERSION_MIN_REQUIRED__ < 160500) || \
+      (defined(__ENVIRONMENT_TV_OS_VERSION_MIN_REQUIRED__) && __ENVIRONMENT_TV_OS_VERSION_MIN_REQUIRED__ < 160500) ||         \
+      (defined(__ENVIRONMENT_WATCH_OS_VERSION_MIN_REQUIRED__) && __ENVIRONMENT_WATCH_OS_VERSION_MIN_REQUIRED__ < 90500)
 // clang-format on
-#    define _LIBCPP_AVAILABILITY_HAS_SYNC 0
+#    define _LIBCPP_INTRODUCED_IN_LLVM_14 0
+#    define _LIBCPP_INTRODUCED_IN_LLVM_14_MARKUP                                                                       \
+      __attribute__((availability(macos, strict, introduced = 13.4)))                                                  \
+      __attribute__((availability(ios, strict, introduced = 16.5)))                                                    \
+      __attribute__((availability(tvos, strict, introduced = 16.5)))                                                   \
+      __attribute__((availability(watchos, strict, introduced = 9.5)))
 #  else
-#    define _LIBCPP_AVAILABILITY_HAS_SYNC 1
+#    define _LIBCPP_INTRODUCED_IN_LLVM_14 1
+#    define _LIBCPP_INTRODUCED_IN_LLVM_14_MARKUP /* nothing */
 #  endif
-#  define _LIBCPP_AVAILABILITY_SYNC                                                                                    \
-    __attribute__((availability(macos, strict, introduced = 11.0)))                                                    \
-    __attribute__((availability(ios, strict, introduced = 14.0)))                                                      \
-    __attribute__((availability(tvos, strict, introduced = 14.0)))                                                     \
-    __attribute__((availability(watchos, strict, introduced = 7.0)))
-
-// __libcpp_verbose_abort
-// TODO: Update once this is released
-#  define _LIBCPP_AVAILABILITY_HAS_VERBOSE_ABORT 0
 
-#  define _LIBCPP_AVAILABILITY_VERBOSE_ABORT __attribute__((unavailable))
-
-// std::pmr
+// LLVM 15-16
+#  define _LIBCPP_INTRODUCED_IN_LLVM_15 _LIBCPP_INTRODUCED_IN_LLVM_16
+#  define _LIBCPP_INTRODUCED_IN_LLVM_15_MARKUP _LIBCPP_INTRODUCED_IN_LLVM_16_MARKUP
 // clang-format off
 #  if (defined(__ENVIRONMENT_MAC_OS_X_VERSION_MIN_REQUIRED__) && __ENVIRONMENT_MAC_OS_X_VERSION_MIN_REQUIRED__ < 140000) ||   \
       (defined(__ENVIRONMENT_IPHONE_OS_VERSION_MIN_REQUIRED__) && __ENVIRONMENT_IPHONE_OS_VERSION_MIN_REQUIRED__ < 170000) || \
       (defined(__ENVIRONMENT_TV_OS_VERSION_MIN_REQUIRED__) && __ENVIRONMENT_TV_OS_VERSION_MIN_REQUIRED__ < 170000) ||         \
       (defined(__ENVIRONMENT_WATCH_OS_VERSION_MIN_REQUIRED__) && __ENVIRONMENT_WATCH_OS_VERSION_MIN_REQUIRED__ < 100000)
 // clang-format on
-#    define _LIBCPP_AVAILABILITY_HAS_PMR 0
-#  else
-#    define _LIBCPP_AVAILABILITY_HAS_PMR 1
-#  endif
-// TODO: Enable std::pmr markup once https://github.com/llvm/llvm-project/issues/40340 has been fixed
-//       Until then, it is possible for folks to try to use `std::pmr` when back-deploying to targets that don't support
-//       it and it'll be a load-time error, but we don't have a good alternative because the library won't compile if we
-//       use availability annotations until that bug has been fixed.
-#  if 0
-#    define _LIBCPP_AVAILABILITY_PMR                                                                                   \
+#    define _LIBCPP_INTRODUCED_IN_LLVM_16 0
+#    define _LIBCPP_INTRODUCED_IN_LLVM_16_MARKUP                                                                       \
       __attribute__((availability(macos, strict, introduced = 14.0)))                                                  \
       __attribute__((availability(ios, strict, introduced = 17.0)))                                                    \
       __attribute__((availability(tvos, strict, introduced = 17.0)))                                                   \
       __attribute__((availability(watchos, strict, introduced = 10.0)))
 #  else
-#    define _LIBCPP_AVAILABILITY_PMR
+#    define _LIBCPP_INTRODUCED_IN_LLVM_16 1
+#    define _LIBCPP_INTRODUCED_IN_LLVM_16_MARKUP /* nothing */
 #  endif
 
-#  define _LIBCPP_AVAILABILITY_HAS_TZDB 0
-#  define _LIBCPP_AVAILABILITY_TZDB __attribute__((unavailable))
-
-// Warning: This availability macro works differently than the other macros.
-// The dylib part of print is not needed on Apple platforms. Therefore when
-// the macro is not available the code calling the dylib is commented out.
-// The macro _LIBCPP_AVAILABILITY_PRINT is not used.
-#  define _LIBCPP_AVAILABILITY_HAS_PRINT 0
-#  define _LIBCPP_AVAILABILITY_PRINT __attribute__((unavailable))
-
-// clang-format off
-#  if (defined(__ENVIRONMENT_MAC_OS_X_VERSION_MIN_REQUIRED__) && __ENVIRONMENT_MAC_OS_X_VERSION_MIN_REQUIRED__ < 120000)   || \
-      (defined(__ENVIRONMENT_IPHONE_OS_VERSION_MIN_REQUIRED__) && __ENVIRONMENT_IPHONE_OS_VERSION_MIN_REQUIRED__ < 150000) || \
-      (defined(__ENVIRONMENT_TV_OS_VERSION_MIN_REQUIRED__) && __ENVIRONMENT_TV_OS_VERSION_MIN_REQUIRED__ < 150000)         || \
-      (defined(__ENVIRONMENT_WATCH_OS_VERSION_MIN_REQUIRED__) && __ENVIRONMENT_WATCH_OS_VERSION_MIN_REQUIRED__ < 80000)
-// clang-format on
-#    define _LIBCPP_AVAILABILITY_HAS_ADDITIONAL_IOSTREAM_EXPLICIT_INSTANTIATIONS_1 0
+// LLVM 18
+// TODO: Fill this in
+#  if 1
+#    define _LIBCPP_INTRODUCED_IN_LLVM_18 0
+#    define _LIBCPP_INTRODUCED_IN_LLVM_18_MARKUP __attribute__((unavailable))
 #  else
-#    define _LIBCPP_AVAILABILITY_HAS_ADDITIONAL_IOSTREAM_EXPLICIT_INSTANTIATIONS_1 1
+#    define _LIBCPP_INTRODUCED_IN_LLVM_18 1
+#    define _LIBCPP_INTRODUCED_IN_LLVM_18_MARKUP /* nothing */
 #  endif
 
-#  define _LIBCPP_AVAILABILITY_HAS_BAD_FUNCTION_CALL_KEY_FUNCTION 0
-#  define _LIBCPP_AVAILABILITY_BAD_FUNCTION_CALL_KEY_FUNCTION __attribute__((unavailable))
-
-#  define _LIBCPP_AVAILABILITY_HAS_BAD_EXPECTED_ACCESS_KEY_FUNCTION 0
-#  define _LIBCPP_AVAILABILITY_BAD_EXPECTED_ACCESS_KEY_FUNCTION __attribute__((unavailable))
+// LLVM 19
+// TODO: Fill this in
+#  if 1
+#    define _LIBCPP_INTRODUCED_IN_LLVM_19 0
+#    define _LIBCPP_INTRODUCED_IN_LLVM_19_MARKUP __attribute__((unavailable))
+#  else
+#    define _LIBCPP_INTRODUCED_IN_LLVM_19 1
+#    define _LIBCPP_INTRODUCED_IN_LLVM_19_MARKUP /* nothing */
+#  endif
 
 #else
 
@@ -315,6 +261,97 @@
 
 #endif
 
+// These macros control the availability of std::bad_optional_access and
+// other exception types. These were put in the shared library to prevent
+// code bloat from every user program defining the vtable for these exception
+// types.
+//
+// Note that when exceptions are disabled, the methods that normally throw
+// these exceptions can be used even on older deployment targets, but those
+// methods will abort instead of throwing.
+#define _LIBCPP_AVAILABILITY_HAS_BAD_OPTIONAL_ACCESS _LIBCPP_INTRODUCED_IN_LLVM_4
+#define _LIBCPP_AVAILABILITY_BAD_OPTIONAL_ACCESS _LIBCPP_INTRODUCED_IN_LLVM_4_MARKUP
+
+#define _LIBCPP_AVAILABILITY_HAS_BAD_VARIANT_ACCESS _LIBCPP_INTRODUCED_IN_LLVM_4
+#define _LIBCPP_AVAILABILITY_BAD_VARIANT_ACCESS _LIBCPP_INTRODUCED_IN_LLVM_4_MARKUP
+
+#define _LIBCPP_AVAILABILITY_HAS_BAD_ANY_CAST _LIBCPP_INTRODUCED_IN_LLVM_4
+#define _LIBCPP_AVAILABILITY_BAD_ANY_CAST _LIBCPP_INTRODUCED_IN_LLVM_4_MARKUP
+
+// These macros control the availability of all parts of <filesystem> that
+// depend on something in the dylib.
+#define _LIBCPP_AVAILABILITY_HAS_FILESYSTEM_LIBRARY _LIBCPP_INTRODUCED_IN_LLVM_9
+#define _LIBCPP_AVAILABILITY_FILESYSTEM_LIBRARY _LIBCPP_INTRODUCED_IN_LLVM_9_MARKUP
+#define _LIBCPP_AVAILABILITY_FILESYSTEM_LIBRARY_PUSH _LIBCPP_INTRODUCED_IN_LLVM_9_MARKUP_PUSH
+#define _LIBCPP_AVAILABILITY_FILESYSTEM_LIBRARY_POP _LIBCPP_INTRODUCED_IN_LLVM_9_MARKUP_POP
+
+// This controls the availability of the C++20 synchronization library,
+// which requires shared library support for various operations
+// (see libcxx/src/atomic.cpp). This includes <barier>, <latch>,
+// <semaphore>, and notification functions on std::atomic.
+#define _LIBCPP_AVAILABILITY_HAS_SYNC _LIBCPP_INTRODUCED_IN_LLVM_10
+#define _LIBCPP_AVAILABILITY_SYNC _LIBCPP_INTRODUCED_IN_LLVM_10_MARKUP
+
+// Enable additional explicit instantiations of iostreams components. This
+// reduces the number of weak definitions generated in programs that use
+// iostreams by providing a single strong definition in the shared library.
+//
+// TODO: Enable additional explicit instantiations on GCC once it supports exclude_from_explicit_instantiation,
+//       or once libc++ doesn't use the attribute anymore.
+// TODO: Enable them on Windows once https://llvm.org/PR41018 has been fixed.
+#if !defined(_LIBCPP_COMPILER_GCC) && !defined(_WIN32)
+#  define _LIBCPP_AVAILABILITY_HAS_ADDITIONAL_IOSTREAM_EXPLICIT_INSTANTIATIONS_1 _LIBCPP_INTRODUCED_IN_LLVM_12
+#else
+#  define _LIBCPP_AVAILABILITY_HAS_ADDITIONAL_IOSTREAM_EXPLICIT_INSTANTIATIONS_1 0
+#endif
+
+// This controls the availability of floating-point std::to_chars functions.
+// These overloads were added later than the integer overloads.
+#define _LIBCPP_AVAILABILITY_HAS_TO_CHARS_FLOATING_POINT _LIBCPP_INTRODUCED_IN_LLVM_14
+#define _LIBCPP_AVAILABILITY_TO_CHARS_FLOATING_POINT _LIBCPP_INTRODUCED_IN_LLVM_14_MARKUP
+
+// This controls whether the library claims to provide a default verbose
+// termination function, and consequently whether the headers will try
+// to use it when the mechanism isn't overriden at compile-time.
+#define _LIBCPP_AVAILABILITY_HAS_VERBOSE_ABORT _LIBCPP_INTRODUCED_IN_LLVM_15
+#define _LIBCPP_AVAILABILITY_VERBOSE_ABORT _LIBCPP_INTRODUCED_IN_LLVM_15_MARKUP
+
+// This controls the availability of the C++17 std::pmr library,
+// which is implemented in large part in the built library.
+//
+// TODO: Enable std::pmr markup once https://github.com/llvm/llvm-project/issues/40340 has been fixed
+//       Until then, it is possible for folks to try to use `std::pmr` when back-deploying to targets that don't support
+//       it and it'll be a load-time error, but we don't have a good alternative because the library won't compile if we
+//       use availability annotations until that bug has been fixed.
+#define _LIBCPP_AVAILABILITY_HAS_PMR _LIBCPP_INTRODUCED_IN_LLVM_16
+#define _LIBCPP_AVAILABILITY_PMR
+
+// These macros controls the availability of __cxa_init_primary_exception
+// in the built library, which std::make_exception_ptr might use
+// (see libcxx/include/__exception/exception_ptr.h).
+#define _LIBCPP_AVAILABILITY_HAS_INIT_PRIMARY_EXCEPTION _LIBCPP_INTRODUCED_IN_LLVM_18
+#define _LIBCPP_AVAILABILITY_INIT_PRIMARY_EXCEPTION _LIBCPP_INTRODUCED_IN_LLVM_18_MARKUP
+
+// This controls the availability of C++23 <print>, which
+// has a dependency on the built library (it needs access to
+// the underlying buffer types of std::cout, std::cerr, and std::clog.
+#define _LIBCPP_AVAILABILITY_HAS_PRINT _LIBCPP_INTRODUCED_IN_LLVM_18
+#define _LIBCPP_AVAILABILITY_PRINT _LIBCPP_INTRODUCED_IN_LLVM_18_MARKUP
+
+// This controls the availability of the C++20 time zone database.
+// The parser code is built in the library.
+#define _LIBCPP_AVAILABILITY_HAS_TZDB _LIBCPP_INTRODUCED_IN_LLVM_19
+#define _LIBCPP_AVAILABILITY_TZDB _LIBCPP_INTRODUCED_IN_LLVM_19_MARKUP
+
+// These macros determine whether we assume that std::bad_function_call and
+// std::bad_expected_access provide a key function in the dylib. This allows
+// centralizing their vtable and typeinfo instead of having all TUs provide
+// a weak definition that then gets deduplicated.
+#  define _LIBCPP_AVAILABILITY_HAS_BAD_FUNCTION_CALL_KEY_FUNCTION _LIBCPP_INTRODUCED_IN_LLVM_19
+#  define _LIBCPP_AVAILABILITY_BAD_FUNCTION_CALL_KEY_FUNCTION _LIBCPP_INTRODUCED_IN_LLVM_19_MARKUP
+#  define _LIBCPP_AVAILABILITY_HAS_BAD_EXPECTED_ACCESS_KEY_FUNCTION _LIBCPP_INTRODUCED_IN_LLVM_19
+#  define _LIBCPP_AVAILABILITY_BAD_EXPECTED_ACCESS_KEY_FUNCTION _LIBCPP_INTRODUCED_IN_LLVM_19_MARKUP
+
 // Define availability attributes that depend on _LIBCPP_HAS_NO_EXCEPTIONS.
 // Those are defined in terms of the availability attributes above, and
 // should not be vendor-specific.
diff --git a/libcxx/include/__chrono/formatter.h b/libcxx/include/__chrono/formatter.h
index 226fccbee6d1..e9b81c3de8a7 100644
--- a/libcxx/include/__chrono/formatter.h
+++ b/libcxx/include/__chrono/formatter.h
@@ -88,6 +88,9 @@ __format_sub_seconds(basic_stringstream<_CharT>& __sstr, const chrono::duration<
   using __duration = chrono::duration<_Rep, _Period>;
 
   auto __fraction = __value - chrono::duration_cast<chrono::seconds>(__value);
+  // Converts a negative fraction to its positive value.
+  if (__value < chrono::seconds{0} && __fraction != __duration{0})
+    __fraction += chrono::seconds{1};
   if constexpr (chrono::treat_as_floating_point_v<_Rep>)
     // When the floating-point value has digits itself they are ignored based
     // on the wording in [tab:time.format.spec]
diff --git a/libcxx/include/__format/escaped_output_table.h b/libcxx/include/__format/escaped_output_table.h
index b194f9431c3b..6aa91c89defa 100644
--- a/libcxx/include/__format/escaped_output_table.h
+++ b/libcxx/include/__format/escaped_output_table.h
@@ -80,10 +80,9 @@ namespace __escaped_output_table {
 /// The entries of the characters to escape in format's debug string.
 ///
 /// Contains the entries for [format.string.escaped]/2.2.1.2.1
-///   CE is a Unicode encoding and C corresponds to either a UCS scalar value
-///   whose Unicode property General_Category has a value in the groups
-///   Separator (Z) or Other (C) or to a UCS scalar value which has the Unicode
-///   property Grapheme_Extend=Yes, as described by table 12 of UAX #44
+///   CE is a Unicode encoding and C corresponds to a UCS scalar value whose
+///   Unicode property General_Category has a value in the groups Separator (Z)
+///   or Other (C), as described by table 12 of UAX #44
 ///
 /// Separator (Z) consists of General_Category
 /// - Space_Separator,
@@ -98,7 +97,6 @@ namespace __escaped_output_table {
 /// - Unassigned.
 ///
 /// The data is generated from
-/// - https://www.unicode.org/Public/UCD/latest/ucd/DerivedCoreProperties.txt
 /// - https://www.unicode.org/Public/UCD/latest/ucd/extracted/DerivedGeneralCategory.txt
 ///
 /// The table is similar to the table
@@ -107,927 +105,751 @@ namespace __escaped_output_table {
 /// table lacks a property, thus having more bits available for the size.
 ///
 /// The data has 2 values:
-/// - bits [0, 10] The size of the range, allowing 2048 elements.
-/// - bits [11, 31] The lower bound code point of the range. The upper bound of
-///   the range is lower bound + size.
-_LIBCPP_HIDE_FROM_ABI inline constexpr uint32_t __entries[894] = {
-    0x00000020,
-    0x0003f821,
-    0x00056800,
-    0x0018006f,
-    0x001bc001,
-    0x001c0003,
-    0x001c5800,
-    0x001c6800,
-    0x001d1000,
-    0x00241806,
-    0x00298000,
-    0x002ab801,
-    0x002c5801,
-    0x002c802d,
-    0x002df800,
-    0x002e0801,
-    0x002e2001,
-    0x002e3808,
-    0x002f5803,
-    0x002fa810,
-    0x0030800a,
-    0x0030e000,
-    0x00325814,
-    0x00338000,
-    0x0036b007,
-    0x0036f805,
-    0x00373801,
-    0x00375003,
-    0x00387001,
-    0x00388800,
-    0x0039801c,
-    0x003d300a,
-    0x003d900d,
-    0x003f5808,
-    0x003fd802,
-    0x0040b003,
-    0x0040d808,
-    0x00412802,
-    0x00414806,
-    0x0041f800,
-    0x0042c804,
-    0x0042f800,
-    0x00435804,
-    0x00447810,
-    0x00465038,
-    0x0049d000,
-    0x0049e000,
-    0x004a0807,
-    0x004a6800,
-    0x004a8806,
-    0x004b1001,
-    0x004c0800,
-    0x004c2000,
-    0x004c6801,
-    0x004c8801,
-    0x004d4800,
-    0x004d8800,
-    0x004d9802,
-    0x004dd002,
-    0x004df000,
-    0x004e0805,
-    0x004e4801,
-    0x004e6800,
-    0x004e780c,
-    0x004ef000,
-    0x004f1003,
-    0x004ff004,
-    0x00502000,
-    0x00505803,
-    0x00508801,
-    0x00514800,
-    0x00518800,
-    0x0051a000,
-    0x0051b800,
-    0x0051d003,
-    0x00520817,
-    0x0052e800,
-    0x0052f806,
-    0x00538001,
-    0x0053a800,
-    0x0053b80b,
-    0x00542000,
-    0x00547000,
-    0x00549000,
-    0x00554800,
-    0x00558800,
-    0x0055a000,
-    0x0055d002,
-    0x00560807,
-    0x00565000,
-    0x00566802,
-    0x0056880e,
-    0x00571003,
-    0x00579006,
-    0x0057d007,
-    0x00582000,
-    0x00586801,
-    0x00588801,
-    0x00594800,
-    0x00598800,
-    0x0059a000,
-    0x0059d002,
-    0x0059f001,
-    0x005a0805,
-    0x005a4801,
-    0x005a680e,
-    0x005af000,
-    0x005b1003,
-    0x005bc00a,
-    0x005c2000,
-    0x005c5802,
-    0x005c8800,
-    0x005cb002,
-    0x005cd800,
-    0x005ce800,
-    0x005d0002,
-    0x005d2802,
-    0x005d5802,
-    0x005dd004,
-    0x005e0000,
-    0x005e1802,
-    0x005e4800,
-    0x005e6802,
-    0x005e8814,
-    0x005fd805,
-    0x00602000,
-    0x00606800,
-    0x00608800,
-    0x00614800,
-    0x0061d002,
-    0x0061f002,
-    0x00622812,
-    0x0062d801,
-    0x0062f001,
-    0x00631003,
-    0x00638006,
-    0x00640800,
-    0x00646800,
-    0x00648800,
-    0x00654800,
-    0x0065a000,
-    0x0065d002,
-    0x0065f800,
-    0x00661000,
-    0x00662801,
-    0x00664800,
-    0x00666010,
-    0x0066f800,
-    0x00671003,
-    0x00678000,
-    0x0067a00d,
-    0x00686800,
-    0x00688800,
-    0x0069d801,
-    0x0069f000,
-    0x006a0804,
-    0x006a4800,
-    0x006a6800,
-    0x006a8003,
-    0x006ab800,
-    0x006b1003,
-    0x006c0001,
-    0x006c2000,
-    0x006cb802,
-    0x006d9000,
-    0x006de000,
-    0x006df001,
-    0x006e3808,
-    0x006e9005,
-    0x006ef806,
-    0x006f8001,
-    0x006fa80b,
-    0x00718800,
-    0x0071a00a,
-    0x00723807,
-    0x0072e024,
-    0x00741800,
-    0x00742800,
-    0x00745800,
-    0x00752000,
-    0x00753000,
-    0x00758800,
-    0x0075a008,
-    0x0075f001,
-    0x00762800,
-    0x00763808,
-    0x0076d001,
-    0x0077001f,
-    0x0078c001,
-    0x0079a800,
-    0x0079b800,
-    0x0079c800,
-    0x007a4000,
-    0x007b6811,
-    0x007c0004,
-    0x007c3001,
-    0x007c6830,
-    0x007e3000,
-    0x007e6800,
-    0x007ed824,
-    0x00816803,
-    0x00819005,
-    0x0081c801,
-    0x0081e801,
-    0x0082c001,
-    0x0082f002,
-    0x00838803,
-    0x00841000,
-    0x00842801,
-    0x00846800,
-    0x0084e800,
-    0x00863000,
-    0x00864004,
-    0x00867001,
-    0x00924800,
-    0x00927001,
-    0x0092b800,
-    0x0092c800,
-    0x0092f001,
-    0x00944800,
-    0x00947001,
-    0x00958800,
-    0x0095b001,
-    0x0095f800,
-    0x00960800,
-    0x00963001,
-    0x0096b800,
-    0x00988800,
-    0x0098b001,
-    0x009ad804,
-    0x009be802,
-    0x009cd005,
-    0x009fb001,
-    0x009ff001,
-    0x00b40000,
-    0x00b4e802,
-    0x00b7c806,
-    0x00b89002,
-    0x00b8b008,
-    0x00b99001,
-    0x00b9b808,
-    0x00ba900d,
-    0x00bb6800,
-    0x00bb880e,
-    0x00bda001,
-    0x00bdb806,
-    0x00be3000,
-    0x00be480a,
-    0x00bee802,
-    0x00bf5005,
-    0x00bfd005,
-    0x00c05804,
-    0x00c0d005,
-    0x00c3c806,
-    0x00c42801,
-    0x00c54800,
-    0x00c55804,
-    0x00c7b009,
-    0x00c8f803,
-    0x00c93801,
-    0x00c96003,
-    0x00c99000,
-    0x00c9c806,
-    0x00ca0802,
-    0x00cb7001,
-    0x00cba80a,
-    0x00cd6003,
-    0x00ce5005,
-    0x00ced802,
-    0x00d0b801,
-    0x00d0d802,
-    0x00d2b000,
-    0x00d2c008,
-    0x00d31000,
-    0x00d32807,
-    0x00d3980c,
-    0x00d45005,
-    0x00d4d005,
-    0x00d57055,
-    0x00d9a006,
-    0x00d9e000,
-    0x00da1000,
-    0x00da6802,
-    0x00db5808,
-    0x00dbf802,
-    0x00dd1003,
-    0x00dd4001,
-    0x00dd5802,
-    0x00df3000,
-    0x00df4001,
-    0x00df6800,
-    0x00df7802,
-    0x00dfa007,
-    0x00e16007,
-    0x00e1b004,
-    0x00e25002,
-    0x00e44806,
-    0x00e5d801,
-    0x00e6400a,
-    0x00e6a00c,
-    0x00e71006,
-    0x00e76800,
-    0x00e7a000,
-    0x00e7c001,
-    0x00e7d804,
-    0x00ee003f,
-    0x00f8b001,
-    0x00f8f001,
-    0x00fa3001,
-    0x00fa7001,
-    0x00fac000,
-    0x00fad000,
-    0x00fae000,
-    0x00faf000,
-    0x00fbf001,
-    0x00fda800,
-    0x00fe2800,
-    0x00fea001,
-    0x00fee000,
-    0x00ff8001,
-    0x00ffa800,
-    0x00fff810,
-    0x01014007,
-    0x0102f810,
-    0x01039001,
-    0x01047800,
-    0x0104e802,
-    0x0106083e,
-    0x010c6003,
-    0x01213818,
-    0x01225814,
-    0x015ba001,
-    0x015cb000,
-    0x01677802,
-    0x0167a004,
-    0x01693000,
-    0x01694004,
-    0x01697001,
-    0x016b4006,
-    0x016b880e,
-    0x016cb808,
-    0x016d3800,
-    0x016d7800,
-    0x016db800,
-    0x016df800,
-    0x016e3800,
-    0x016e7800,
-    0x016eb800,
-    0x016ef820,
-    0x0172f021,
-    0x0174d000,
-    0x0177a00b,
-    0x017eb019,
-    0x01800000,
-    0x01815005,
-    0x01820000,
-    0x0184b803,
-    0x01880004,
-    0x01898000,
-    0x018c7800,
-    0x018f200a,
-    0x0190f800,
-    0x05246802,
-    0x05263808,
-    0x05316013,
-    0x05337803,
-    0x0533a009,
-    0x0534f001,
-    0x05378001,
-    0x0537c007,
-    0x053e5804,
-    0x053e9000,
-    0x053ea000,
-    0x053ed017,
-    0x05401000,
-    0x05403000,
-    0x05405800,
-    0x05412801,
-    0x05416003,
-    0x0541d005,
-    0x0543c007,
-    0x05462009,
-    0x0546d017,
-    0x0547f800,
-    0x05493007,
-    0x054a380a,
-    0x054aa00a,
-    0x054be805,
-    0x054d9800,
-    0x054db003,
-    0x054de001,
-    0x054e7000,
-    0x054ed003,
-    0x054f2800,
-    0x054ff800,
-    0x05514805,
-    0x05518801,
-    0x0551a80a,
-    0x05521800,
-    0x05526000,
-    0x05527001,
-    0x0552d001,
-    0x0553e000,
-    0x05558000,
-    0x05559002,
-    0x0555b801,
-    0x0555f001,
-    0x05560800,
-    0x05561817,
-    0x05576001,
-    0x0557b00a,
-    0x05583801,
-    0x05587801,
-    0x0558b808,
-    0x05593800,
-    0x05597800,
-    0x055b6003,
-    0x055f2800,
-    0x055f4000,
-    0x055f6802,
-    0x055fd005,
-    0x06bd200b,
-    0x06be3803,
-    0x06bfe7ff,
-    0x06ffe7ff,
-    0x073fe7ff,
-    0x077fe7ff,
-    0x07bfe103,
-    0x07d37001,
-    0x07d6d025,
-    0x07d8380b,
-    0x07d8c004,
-    0x07d8f000,
-    0x07d9b800,
-    0x07d9e800,
-    0x07d9f800,
-    0x07da1000,
-    0x07da2800,
-    0x07de180f,
-    0x07ec8001,
-    0x07ee4006,
-    0x07ee801f,
-    0x07f0000f,
-    0x07f0d015,
-    0x07f29800,
-    0x07f33800,
-    0x07f36003,
-    0x07f3a800,
-    0x07f7e803,
-    0x07fcf001,
-    0x07fdf802,
-    0x07fe4001,
-    0x07fe8001,
-    0x07fec001,
-    0x07fee802,
-    0x07ff3800,
-    0x07ff780c,
-    0x07fff001,
-    0x08006000,
-    0x08013800,
-    0x0801d800,
-    0x0801f000,
-    0x08027001,
-    0x0802f021,
-    0x0807d804,
-    0x08081803,
-    0x0809a002,
-    0x080c7800,
-    0x080ce802,
-    0x080d082e,
-    0x080fe882,
-    0x0814e802,
-    0x0816880f,
-    0x0817e003,
-    0x08192008,
-    0x081a5804,
-    0x081bb009,
-    0x081cf000,
-    0x081e2003,
-    0x081eb029,
-    0x0824f001,
-    0x08255005,
-    0x0826a003,
-    0x0827e003,
-    0x08294007,
-    0x082b200a,
-    0x082bd800,
-    0x082c5800,
-    0x082c9800,
-    0x082cb000,
-    0x082d1000,
-    0x082d9000,
-    0x082dd000,
-    0x082de842,
-    0x0839b808,
-    0x083ab009,
-    0x083b4017,
-    0x083c3000,
-    0x083d8800,
-    0x083dd844,
-    0x08403001,
-    0x08404800,
-    0x0841b000,
-    0x0841c802,
-    0x0841e801,
-    0x0842b000,
-    0x0844f807,
-    0x0845802f,
-    0x08479800,
-    0x0847b004,
-    0x0848e002,
-    0x0849d004,
-    0x084a003f,
-    0x084dc003,
-    0x084e8001,
-    0x0850080e,
-    0x0850a000,
-    0x0850c000,
-    0x0851b009,
-    0x08524806,
-    0x0852c806,
-    0x0855001f,
-    0x08572805,
-    0x0857b808,
-    0x0859b002,
-    0x085ab001,
-    0x085b9804,
-    0x085c9006,
-    0x085ce80b,
-    0x085d804f,
-    0x08624836,
-    0x0865980c,
-    0x08679806,
-    0x0869200b,
-    0x0869d125,
-    0x0873f800,
-    0x08755002,
-    0x08757001,
-    0x0875904d,
-    0x08794007,
-    0x087a300a,
-    0x087ad015,
-    0x087c1003,
-    0x087c5025,
-    0x087e6013,
-    0x087fb808,
-    0x08800800,
-    0x0881c00e,
-    0x08827003,
-    0x08838000,
-    0x08839801,
-    0x0883b00b,
-    0x08859803,
-    0x0885c801,
-    0x0885e800,
-    0x0886100d,
-    0x08874806,
-    0x0887d008,
-    0x08893804,
-    0x08896808,
-    0x088a4007,
-    0x088b9800,
-    0x088bb80a,
-    0x088db008,
-    0x088e4803,
-    0x088e7800,
-    0x088f0000,
-    0x088fa80a,
-    0x08909000,
-    0x08917802,
-    0x0891a000,
-    0x0891b001,
-    0x0891f000,
-    0x0892083e,
-    0x08943800,
-    0x08944800,
-    0x08947000,
-    0x0894f000,
-    0x08955005,
-    0x0896f800,
-    0x0897180c,
-    0x0897d007,
-    0x08982000,
-    0x08986801,
-    0x08988801,
-    0x08994800,
-    0x08998800,
-    0x0899a000,
-    0x0899d002,
-    0x0899f000,
-    0x089a0000,
-    0x089a2801,
-    0x089a4801,
-    0x089a7001,
-    0x089a880b,
-    0x089b209b,
-    0x08a1c007,
-    0x08a21002,
-    0x08a23000,
-    0x08a2e000,
-    0x08a2f000,
-    0x08a3101d,
-    0x08a58000,
-    0x08a59805,
-    0x08a5d000,
-    0x08a5e800,
-    0x08a5f801,
-    0x08a61001,
-    0x08a64007,
-    0x08a6d0a5,
-    0x08ad7800,
-    0x08ad9005,
-    0x08ade001,
-    0x08adf801,
-    0x08aee023,
-    0x08b19807,
-    0x08b1e800,
-    0x08b1f801,
-    0x08b2280a,
-    0x08b2d005,
-    0x08b36812,
-    0x08b55800,
-    0x08b56800,
-    0x08b58005,
-    0x08b5b800,
-    0x08b5d005,
-    0x08b65035,
-    0x08b8d804,
-    0x08b91003,
-    0x08b93808,
-    0x08ba38b8,
-    0x08c17808,
-    0x08c1c801,
-    0x08c1e063,
-    0x08c7980b,
-    0x08c83801,
-    0x08c85001,
-    0x08c8a000,
-    0x08c8b800,
-    0x08c98000,
-    0x08c9b000,
-    0x08c9c803,
-    0x08c9f000,
-    0x08ca1800,
-    0x08ca3808,
-    0x08cad045,
-    0x08cd4001,
-    0x08cea007,
-    0x08cf0000,
-    0x08cf281a,
-    0x08d00809,
-    0x08d19805,
-    0x08d1d803,
-    0x08d23808,
-    0x08d28805,
-    0x08d2c802,
-    0x08d4500c,
-    0x08d4c001,
-    0x08d5180c,
-    0x08d7c806,
-    0x08d850f5,
-    0x08e04800,
-    0x08e1800d,
-    0x08e1f800,
-    0x08e23009,
-    0x08e36802,
-    0x08e48018,
-    0x08e55006,
-    0x08e59001,
-    0x08e5a84a,
-    0x08e83800,
-    0x08e85000,
-    0x08e98814,
-    0x08ea3808,
-    0x08ead005,
-    0x08eb3000,
-    0x08eb4800,
-    0x08ec7803,
-    0x08eca800,
-    0x08ecb800,
-    0x08ecc806,
-    0x08ed5135,
-    0x08f79801,
-    0x08f7c808,
-    0x08f88800,
-    0x08f9b007,
-    0x08fa0000,
-    0x08fa1000,
-    0x08fad055,
-    0x08fd880e,
-    0x08ff900c,
-    0x091cd065,
-    0x09237800,
-    0x0923a80a,
-    0x092a27ff,
-    0x096a224b,
-    0x097f980c,
-    0x09a18010,
-    0x09a23fff,
-    0x09e23fb8,
-    0x0a323fff,
-    0x0a723fff,
-    0x0ab23fff,
-    0x0af23fff,
-    0x0b3239b8,
-    0x0b51c806,
-    0x0b52f800,
-    0x0b535003,
-    0x0b55f800,
-    0x0b565005,
-    0x0b577006,
-    0x0b57b009,
-    0x0b598006,
-    0x0b5a3009,
-    0x0b5ad000,
-    0x0b5b1000,
-    0x0b5bc004,
-    0x0b5c82af,
-    0x0b74d864,
-    0x0b7a5804,
-    0x0b7c400a,
-    0x0b7d003f,
-    0x0b7f200b,
-    0x0b7f900d,
-    0x0c3fc007,
-    0x0c66b029,
-    0x0c684fff,
-    0x0ca84fff,
-    0x0ce84fff,
-    0x0d284fff,
-    0x0d684ae6,
-    0x0d7fa000,
-    0x0d7fe000,
-    0x0d7ff800,
-    0x0d89180e,
-    0x0d89981c,
-    0x0d8a9801,
-    0x0d8ab00d,
-    0x0d8b4007,
-    0x0d97e7ff,
-    0x0dd7e103,
-    0x0de35804,
-    0x0de3e802,
-    0x0de44806,
-    0x0de4d001,
-    0x0de4e801,
-    0x0de507ff,
-    0x0e2507ff,
-    0x0e6502af,
-    0x0e7e203b,
-    0x0e87b009,
-    0x0e893801,
-    0x0e8b2800,
-    0x0e8b3802,
-    0x0e8b7014,
-    0x0e8c2806,
-    0x0e8d5003,
-    0x0e8f5814,
-    0x0e921002,
-    0x0e923079,
-    0x0e96a00b,
-    0x0e97a00b,
-    0x0e9ab808,
-    0x0e9bc886,
-    0x0ea2a800,
-    0x0ea4e800,
-    0x0ea50001,
-    0x0ea51801,
-    0x0ea53801,
-    0x0ea56800,
-    0x0ea5d000,
-    0x0ea5e000,
-    0x0ea62000,
-    0x0ea83000,
-    0x0ea85801,
-    0x0ea8a800,
-    0x0ea8e800,
-    0x0ea9d000,
-    0x0ea9f800,
-    0x0eaa2800,
-    0x0eaa3802,
-    0x0eaa8800,
-    0x0eb53001,
-    0x0ebe6001,
-    0x0ed00036,
-    0x0ed1d831,
-    0x0ed3a800,
-    0x0ed42000,
-    0x0ed46473,
-    0x0ef8f805,
-    0x0ef95904,
-    0x0f037091,
-    0x0f096809,
-    0x0f09f001,
-    0x0f0a5003,
-    0x0f0a813f,
-    0x0f157011,
-    0x0f176003,
-    0x0f17d004,
-    0x0f1801cf,
-    0x0f276003,
-    0x0f27d2e5,
-    0x0f3f3800,
-    0x0f3f6000,
-    0x0f3f7800,
-    0x0f3ff800,
-    0x0f462801,
-    0x0f46802f,
-    0x0f4a2006,
-    0x0f4a6003,
-    0x0f4ad003,
-    0x0f4b0310,
-    0x0f65a84b,
-    0x0f69f0c1,
-    0x0f702000,
-    0x0f710000,
-    0x0f711800,
-    0x0f712801,
-    0x0f714000,
-    0x0f719800,
-    0x0f71c000,
-    0x0f71d000,
-    0x0f71e005,
-    0x0f721803,
-    0x0f724000,
-    0x0f725000,
-    0x0f726000,
-    0x0f728000,
-    0x0f729800,
-    0x0f72a801,
-    0x0f72c000,
-    0x0f72d000,
-    0x0f72e000,
-    0x0f72f000,
-    0x0f730000,
-    0x0f731800,
-    0x0f732801,
-    0x0f735800,
-    0x0f739800,
-    0x0f73c000,
-    0x0f73e800,
-    0x0f73f800,
-    0x0f745000,
-    0x0f74e004,
-    0x0f752000,
-    0x0f755000,
-    0x0f75e033,
-    0x0f77910d,
-    0x0f816003,
-    0x0f84a00b,
-    0x0f857801,
-    0x0f860000,
-    0x0f868000,
-    0x0f87b009,
-    0x0f8d7037,
-    0x0f90180c,
-    0x0f91e003,
-    0x0f924806,
-    0x0f92900d,
-    0x0f933099,
-    0x0fb6c003,
-    0x0fb76802,
-    0x0fb7e802,
-    0x0fbbb803,
-    0x0fbed005,
-    0x0fbf6003,
-    0x0fbf880e,
-    0x0fc06003,
-    0x0fc24007,
-    0x0fc2d005,
-    0x0fc44007,
-    0x0fc57001,
-    0x0fc5904d,
-    0x0fd2a00b,
-    0x0fd37001,
-    0x0fd3e802,
-    0x0fd44806,
-    0x0fd5f000,
-    0x0fd63007,
-    0x0fd6e003,
-    0x0fd74806,
-    0x0fd7c806,
-    0x0fdc9800,
-    0x0fde5824,
-    0x0fdfd405,
-    0x1537001f,
-    0x15b9d005,
-    0x15c0f001,
-    0x1675100d,
-    0x175f080e,
-    0x1772f7ff,
-    0x17b2f1a1,
-    0x17d0f5e1,
-    0x189a5804};
+/// - bits [0, 13] The size of the range, allowing 16384 elements.
+/// - bits [14, 31] The lower bound code point of the range. The upper bound of
+///   the range is lower bound + size. Note the code expects code units the fit
+///   into 18 bits, instead of the 21 bits needed for the full Unicode range.
+_LIBCPP_HIDE_FROM_ABI inline constexpr uint32_t __entries[711] = {
+    0x00000020 /* 00000000 - 00000020 [   33] */,
+    0x001fc021 /* 0000007f - 000000a0 [   34] */,
+    0x002b4000 /* 000000ad - 000000ad [    1] */,
+    0x00de0001 /* 00000378 - 00000379 [    2] */,
+    0x00e00003 /* 00000380 - 00000383 [    4] */,
+    0x00e2c000 /* 0000038b - 0000038b [    1] */,
+    0x00e34000 /* 0000038d - 0000038d [    1] */,
+    0x00e88000 /* 000003a2 - 000003a2 [    1] */,
+    0x014c0000 /* 00000530 - 00000530 [    1] */,
+    0x0155c001 /* 00000557 - 00000558 [    2] */,
+    0x0162c001 /* 0000058b - 0000058c [    2] */,
+    0x01640000 /* 00000590 - 00000590 [    1] */,
+    0x01720007 /* 000005c8 - 000005cf [    8] */,
+    0x017ac003 /* 000005eb - 000005ee [    4] */,
+    0x017d4010 /* 000005f5 - 00000605 [   17] */,
+    0x01870000 /* 0000061c - 0000061c [    1] */,
+    0x01b74000 /* 000006dd - 000006dd [    1] */,
+    0x01c38001 /* 0000070e - 0000070f [    2] */,
+    0x01d2c001 /* 0000074b - 0000074c [    2] */,
+    0x01ec800d /* 000007b2 - 000007bf [   14] */,
+    0x01fec001 /* 000007fb - 000007fc [    2] */,
+    0x020b8001 /* 0000082e - 0000082f [    2] */,
+    0x020fc000 /* 0000083f - 0000083f [    1] */,
+    0x02170001 /* 0000085c - 0000085d [    2] */,
+    0x0217c000 /* 0000085f - 0000085f [    1] */,
+    0x021ac004 /* 0000086b - 0000086f [    5] */,
+    0x0223c008 /* 0000088f - 00000897 [    9] */,
+    0x02388000 /* 000008e2 - 000008e2 [    1] */,
+    0x02610000 /* 00000984 - 00000984 [    1] */,
+    0x02634001 /* 0000098d - 0000098e [    2] */,
+    0x02644001 /* 00000991 - 00000992 [    2] */,
+    0x026a4000 /* 000009a9 - 000009a9 [    1] */,
+    0x026c4000 /* 000009b1 - 000009b1 [    1] */,
+    0x026cc002 /* 000009b3 - 000009b5 [    3] */,
+    0x026e8001 /* 000009ba - 000009bb [    2] */,
+    0x02714001 /* 000009c5 - 000009c6 [    2] */,
+    0x02724001 /* 000009c9 - 000009ca [    2] */,
+    0x0273c007 /* 000009cf - 000009d6 [    8] */,
+    0x02760003 /* 000009d8 - 000009db [    4] */,
+    0x02778000 /* 000009de - 000009de [    1] */,
+    0x02790001 /* 000009e4 - 000009e5 [    2] */,
+    0x027fc001 /* 000009ff - 00000a00 [    2] */,
+    0x02810000 /* 00000a04 - 00000a04 [    1] */,
+    0x0282c003 /* 00000a0b - 00000a0e [    4] */,
+    0x02844001 /* 00000a11 - 00000a12 [    2] */,
+    0x028a4000 /* 00000a29 - 00000a29 [    1] */,
+    0x028c4000 /* 00000a31 - 00000a31 [    1] */,
+    0x028d0000 /* 00000a34 - 00000a34 [    1] */,
+    0x028dc000 /* 00000a37 - 00000a37 [    1] */,
+    0x028e8001 /* 00000a3a - 00000a3b [    2] */,
+    0x028f4000 /* 00000a3d - 00000a3d [    1] */,
+    0x0290c003 /* 00000a43 - 00000a46 [    4] */,
+    0x02924001 /* 00000a49 - 00000a4a [    2] */,
+    0x02938002 /* 00000a4e - 00000a50 [    3] */,
+    0x02948006 /* 00000a52 - 00000a58 [    7] */,
+    0x02974000 /* 00000a5d - 00000a5d [    1] */,
+    0x0297c006 /* 00000a5f - 00000a65 [    7] */,
+    0x029dc009 /* 00000a77 - 00000a80 [   10] */,
+    0x02a10000 /* 00000a84 - 00000a84 [    1] */,
+    0x02a38000 /* 00000a8e - 00000a8e [    1] */,
+    0x02a48000 /* 00000a92 - 00000a92 [    1] */,
+    0x02aa4000 /* 00000aa9 - 00000aa9 [    1] */,
+    0x02ac4000 /* 00000ab1 - 00000ab1 [    1] */,
+    0x02ad0000 /* 00000ab4 - 00000ab4 [    1] */,
+    0x02ae8001 /* 00000aba - 00000abb [    2] */,
+    0x02b18000 /* 00000ac6 - 00000ac6 [    1] */,
+    0x02b28000 /* 00000aca - 00000aca [    1] */,
+    0x02b38001 /* 00000ace - 00000acf [    2] */,
+    0x02b4400e /* 00000ad1 - 00000adf [   15] */,
+    0x02b90001 /* 00000ae4 - 00000ae5 [    2] */,
+    0x02bc8006 /* 00000af2 - 00000af8 [    7] */,
+    0x02c00000 /* 00000b00 - 00000b00 [    1] */,
+    0x02c10000 /* 00000b04 - 00000b04 [    1] */,
+    0x02c34001 /* 00000b0d - 00000b0e [    2] */,
+    0x02c44001 /* 00000b11 - 00000b12 [    2] */,
+    0x02ca4000 /* 00000b29 - 00000b29 [    1] */,
+    0x02cc4000 /* 00000b31 - 00000b31 [    1] */,
+    0x02cd0000 /* 00000b34 - 00000b34 [    1] */,
+    0x02ce8001 /* 00000b3a - 00000b3b [    2] */,
+    0x02d14001 /* 00000b45 - 00000b46 [    2] */,
+    0x02d24001 /* 00000b49 - 00000b4a [    2] */,
+    0x02d38006 /* 00000b4e - 00000b54 [    7] */,
+    0x02d60003 /* 00000b58 - 00000b5b [    4] */,
+    0x02d78000 /* 00000b5e - 00000b5e [    1] */,
+    0x02d90001 /* 00000b64 - 00000b65 [    2] */,
+    0x02de0009 /* 00000b78 - 00000b81 [   10] */,
+    0x02e10000 /* 00000b84 - 00000b84 [    1] */,
+    0x02e2c002 /* 00000b8b - 00000b8d [    3] */,
+    0x02e44000 /* 00000b91 - 00000b91 [    1] */,
+    0x02e58002 /* 00000b96 - 00000b98 [    3] */,
+    0x02e6c000 /* 00000b9b - 00000b9b [    1] */,
+    0x02e74000 /* 00000b9d - 00000b9d [    1] */,
+    0x02e80002 /* 00000ba0 - 00000ba2 [    3] */,
+    0x02e94002 /* 00000ba5 - 00000ba7 [    3] */,
+    0x02eac002 /* 00000bab - 00000bad [    3] */,
+    0x02ee8003 /* 00000bba - 00000bbd [    4] */,
+    0x02f0c002 /* 00000bc3 - 00000bc5 [    3] */,
+    0x02f24000 /* 00000bc9 - 00000bc9 [    1] */,
+    0x02f38001 /* 00000bce - 00000bcf [    2] */,
+    0x02f44005 /* 00000bd1 - 00000bd6 [    6] */,
+    0x02f6000d /* 00000bd8 - 00000be5 [   14] */,
+    0x02fec004 /* 00000bfb - 00000bff [    5] */,
+    0x03034000 /* 00000c0d - 00000c0d [    1] */,
+    0x03044000 /* 00000c11 - 00000c11 [    1] */,
+    0x030a4000 /* 00000c29 - 00000c29 [    1] */,
+    0x030e8001 /* 00000c3a - 00000c3b [    2] */,
+    0x03114000 /* 00000c45 - 00000c45 [    1] */,
+    0x03124000 /* 00000c49 - 00000c49 [    1] */,
+    0x03138006 /* 00000c4e - 00000c54 [    7] */,
+    0x0315c000 /* 00000c57 - 00000c57 [    1] */,
+    0x0316c001 /* 00000c5b - 00000c5c [    2] */,
+    0x03178001 /* 00000c5e - 00000c5f [    2] */,
+    0x03190001 /* 00000c64 - 00000c65 [    2] */,
+    0x031c0006 /* 00000c70 - 00000c76 [    7] */,
+    0x03234000 /* 00000c8d - 00000c8d [    1] */,
+    0x03244000 /* 00000c91 - 00000c91 [    1] */,
+    0x032a4000 /* 00000ca9 - 00000ca9 [    1] */,
+    0x032d0000 /* 00000cb4 - 00000cb4 [    1] */,
+    0x032e8001 /* 00000cba - 00000cbb [    2] */,
+    0x03314000 /* 00000cc5 - 00000cc5 [    1] */,
+    0x03324000 /* 00000cc9 - 00000cc9 [    1] */,
+    0x03338006 /* 00000cce - 00000cd4 [    7] */,
+    0x0335c005 /* 00000cd7 - 00000cdc [    6] */,
+    0x0337c000 /* 00000cdf - 00000cdf [    1] */,
+    0x03390001 /* 00000ce4 - 00000ce5 [    2] */,
+    0x033c0000 /* 00000cf0 - 00000cf0 [    1] */,
+    0x033d000b /* 00000cf4 - 00000cff [   12] */,
+    0x03434000 /* 00000d0d - 00000d0d [    1] */,
+    0x03444000 /* 00000d11 - 00000d11 [    1] */,
+    0x03514000 /* 00000d45 - 00000d45 [    1] */,
+    0x03524000 /* 00000d49 - 00000d49 [    1] */,
+    0x03540003 /* 00000d50 - 00000d53 [    4] */,
+    0x03590001 /* 00000d64 - 00000d65 [    2] */,
+    0x03600000 /* 00000d80 - 00000d80 [    1] */,
+    0x03610000 /* 00000d84 - 00000d84 [    1] */,
+    0x0365c002 /* 00000d97 - 00000d99 [    3] */,
+    0x036c8000 /* 00000db2 - 00000db2 [    1] */,
+    0x036f0000 /* 00000dbc - 00000dbc [    1] */,
+    0x036f8001 /* 00000dbe - 00000dbf [    2] */,
+    0x0371c002 /* 00000dc7 - 00000dc9 [    3] */,
+    0x0372c003 /* 00000dcb - 00000dce [    4] */,
+    0x03754000 /* 00000dd5 - 00000dd5 [    1] */,
+    0x0375c000 /* 00000dd7 - 00000dd7 [    1] */,
+    0x03780005 /* 00000de0 - 00000de5 [    6] */,
+    0x037c0001 /* 00000df0 - 00000df1 [    2] */,
+    0x037d400b /* 00000df5 - 00000e00 [   12] */,
+    0x038ec003 /* 00000e3b - 00000e3e [    4] */,
+    0x03970024 /* 00000e5c - 00000e80 [   37] */,
+    0x03a0c000 /* 00000e83 - 00000e83 [    1] */,
+    0x03a14000 /* 00000e85 - 00000e85 [    1] */,
+    0x03a2c000 /* 00000e8b - 00000e8b [    1] */,
+    0x03a90000 /* 00000ea4 - 00000ea4 [    1] */,
+    0x03a98000 /* 00000ea6 - 00000ea6 [    1] */,
+    0x03af8001 /* 00000ebe - 00000ebf [    2] */,
+    0x03b14000 /* 00000ec5 - 00000ec5 [    1] */,
+    0x03b1c000 /* 00000ec7 - 00000ec7 [    1] */,
+    0x03b3c000 /* 00000ecf - 00000ecf [    1] */,
+    0x03b68001 /* 00000eda - 00000edb [    2] */,
+    0x03b8001f /* 00000ee0 - 00000eff [   32] */,
+    0x03d20000 /* 00000f48 - 00000f48 [    1] */,
+    0x03db4003 /* 00000f6d - 00000f70 [    4] */,
+    0x03e60000 /* 00000f98 - 00000f98 [    1] */,
+    0x03ef4000 /* 00000fbd - 00000fbd [    1] */,
+    0x03f34000 /* 00000fcd - 00000fcd [    1] */,
+    0x03f6c024 /* 00000fdb - 00000fff [   37] */,
+    0x04318000 /* 000010c6 - 000010c6 [    1] */,
+    0x04320004 /* 000010c8 - 000010cc [    5] */,
+    0x04338001 /* 000010ce - 000010cf [    2] */,
+    0x04924000 /* 00001249 - 00001249 [    1] */,
+    0x04938001 /* 0000124e - 0000124f [    2] */,
+    0x0495c000 /* 00001257 - 00001257 [    1] */,
+    0x04964000 /* 00001259 - 00001259 [    1] */,
+    0x04978001 /* 0000125e - 0000125f [    2] */,
+    0x04a24000 /* 00001289 - 00001289 [    1] */,
+    0x04a38001 /* 0000128e - 0000128f [    2] */,
+    0x04ac4000 /* 000012b1 - 000012b1 [    1] */,
+    0x04ad8001 /* 000012b6 - 000012b7 [    2] */,
+    0x04afc000 /* 000012bf - 000012bf [    1] */,
+    0x04b04000 /* 000012c1 - 000012c1 [    1] */,
+    0x04b18001 /* 000012c6 - 000012c7 [    2] */,
+    0x04b5c000 /* 000012d7 - 000012d7 [    1] */,
+    0x04c44000 /* 00001311 - 00001311 [    1] */,
+    0x04c58001 /* 00001316 - 00001317 [    2] */,
+    0x04d6c001 /* 0000135b - 0000135c [    2] */,
+    0x04df4002 /* 0000137d - 0000137f [    3] */,
+    0x04e68005 /* 0000139a - 0000139f [    6] */,
+    0x04fd8001 /* 000013f6 - 000013f7 [    2] */,
+    0x04ff8001 /* 000013fe - 000013ff [    2] */,
+    0x05a00000 /* 00001680 - 00001680 [    1] */,
+    0x05a74002 /* 0000169d - 0000169f [    3] */,
+    0x05be4006 /* 000016f9 - 000016ff [    7] */,
+    0x05c58008 /* 00001716 - 0000171e [    9] */,
+    0x05cdc008 /* 00001737 - 0000173f [    9] */,
+    0x05d5000b /* 00001754 - 0000175f [   12] */,
+    0x05db4000 /* 0000176d - 0000176d [    1] */,
+    0x05dc4000 /* 00001771 - 00001771 [    1] */,
+    0x05dd000b /* 00001774 - 0000177f [   12] */,
+    0x05f78001 /* 000017de - 000017df [    2] */,
+    0x05fa8005 /* 000017ea - 000017ef [    6] */,
+    0x05fe8005 /* 000017fa - 000017ff [    6] */,
+    0x06038000 /* 0000180e - 0000180e [    1] */,
+    0x06068005 /* 0000181a - 0000181f [    6] */,
+    0x061e4006 /* 00001879 - 0000187f [    7] */,
+    0x062ac004 /* 000018ab - 000018af [    5] */,
+    0x063d8009 /* 000018f6 - 000018ff [   10] */,
+    0x0647c000 /* 0000191f - 0000191f [    1] */,
+    0x064b0003 /* 0000192c - 0000192f [    4] */,
+    0x064f0003 /* 0000193c - 0000193f [    4] */,
+    0x06504002 /* 00001941 - 00001943 [    3] */,
+    0x065b8001 /* 0000196e - 0000196f [    2] */,
+    0x065d400a /* 00001975 - 0000197f [   11] */,
+    0x066b0003 /* 000019ac - 000019af [    4] */,
+    0x06728005 /* 000019ca - 000019cf [    6] */,
+    0x0676c002 /* 000019db - 000019dd [    3] */,
+    0x06870001 /* 00001a1c - 00001a1d [    2] */,
+    0x0697c000 /* 00001a5f - 00001a5f [    1] */,
+    0x069f4001 /* 00001a7d - 00001a7e [    2] */,
+    0x06a28005 /* 00001a8a - 00001a8f [    6] */,
+    0x06a68005 /* 00001a9a - 00001a9f [    6] */,
+    0x06ab8001 /* 00001aae - 00001aaf [    2] */,
+    0x06b3c030 /* 00001acf - 00001aff [   49] */,
+    0x06d34002 /* 00001b4d - 00001b4f [    3] */,
+    0x06dfc000 /* 00001b7f - 00001b7f [    1] */,
+    0x06fd0007 /* 00001bf4 - 00001bfb [    8] */,
+    0x070e0002 /* 00001c38 - 00001c3a [    3] */,
+    0x07128002 /* 00001c4a - 00001c4c [    3] */,
+    0x07224006 /* 00001c89 - 00001c8f [    7] */,
+    0x072ec001 /* 00001cbb - 00001cbc [    2] */,
+    0x07320007 /* 00001cc8 - 00001ccf [    8] */,
+    0x073ec004 /* 00001cfb - 00001cff [    5] */,
+    0x07c58001 /* 00001f16 - 00001f17 [    2] */,
+    0x07c78001 /* 00001f1e - 00001f1f [    2] */,
+    0x07d18001 /* 00001f46 - 00001f47 [    2] */,
+    0x07d38001 /* 00001f4e - 00001f4f [    2] */,
+    0x07d60000 /* 00001f58 - 00001f58 [    1] */,
+    0x07d68000 /* 00001f5a - 00001f5a [    1] */,
+    0x07d70000 /* 00001f5c - 00001f5c [    1] */,
+    0x07d78000 /* 00001f5e - 00001f5e [    1] */,
+    0x07df8001 /* 00001f7e - 00001f7f [    2] */,
+    0x07ed4000 /* 00001fb5 - 00001fb5 [    1] */,
+    0x07f14000 /* 00001fc5 - 00001fc5 [    1] */,
+    0x07f50001 /* 00001fd4 - 00001fd5 [    2] */,
+    0x07f70000 /* 00001fdc - 00001fdc [    1] */,
+    0x07fc0001 /* 00001ff0 - 00001ff1 [    2] */,
+    0x07fd4000 /* 00001ff5 - 00001ff5 [    1] */,
+    0x07ffc010 /* 00001fff - 0000200f [   17] */,
+    0x080a0007 /* 00002028 - 0000202f [    8] */,
+    0x0817c010 /* 0000205f - 0000206f [   17] */,
+    0x081c8001 /* 00002072 - 00002073 [    2] */,
+    0x0823c000 /* 0000208f - 0000208f [    1] */,
+    0x08274002 /* 0000209d - 0000209f [    3] */,
+    0x0830400e /* 000020c1 - 000020cf [   15] */,
+    0x083c400e /* 000020f1 - 000020ff [   15] */,
+    0x08630003 /* 0000218c - 0000218f [    4] */,
+    0x0909c018 /* 00002427 - 0000243f [   25] */,
+    0x0912c014 /* 0000244b - 0000245f [   21] */,
+    0x0add0001 /* 00002b74 - 00002b75 [    2] */,
+    0x0ae58000 /* 00002b96 - 00002b96 [    1] */,
+    0x0b3d0004 /* 00002cf4 - 00002cf8 [    5] */,
+    0x0b498000 /* 00002d26 - 00002d26 [    1] */,
+    0x0b4a0004 /* 00002d28 - 00002d2c [    5] */,
+    0x0b4b8001 /* 00002d2e - 00002d2f [    2] */,
+    0x0b5a0006 /* 00002d68 - 00002d6e [    7] */,
+    0x0b5c400d /* 00002d71 - 00002d7e [   14] */,
+    0x0b65c008 /* 00002d97 - 00002d9f [    9] */,
+    0x0b69c000 /* 00002da7 - 00002da7 [    1] */,
+    0x0b6bc000 /* 00002daf - 00002daf [    1] */,
+    0x0b6dc000 /* 00002db7 - 00002db7 [    1] */,
+    0x0b6fc000 /* 00002dbf - 00002dbf [    1] */,
+    0x0b71c000 /* 00002dc7 - 00002dc7 [    1] */,
+    0x0b73c000 /* 00002dcf - 00002dcf [    1] */,
+    0x0b75c000 /* 00002dd7 - 00002dd7 [    1] */,
+    0x0b77c000 /* 00002ddf - 00002ddf [    1] */,
+    0x0b978021 /* 00002e5e - 00002e7f [   34] */,
+    0x0ba68000 /* 00002e9a - 00002e9a [    1] */,
+    0x0bbd000b /* 00002ef4 - 00002eff [   12] */,
+    0x0bf58019 /* 00002fd6 - 00002fef [   26] */,
+    0x0c000000 /* 00003000 - 00003000 [    1] */,
+    0x0c100000 /* 00003040 - 00003040 [    1] */,
+    0x0c25c001 /* 00003097 - 00003098 [    2] */,
+    0x0c400004 /* 00003100 - 00003104 [    5] */,
+    0x0c4c0000 /* 00003130 - 00003130 [    1] */,
+    0x0c63c000 /* 0000318f - 0000318f [    1] */,
+    0x0c79000a /* 000031e4 - 000031ee [   11] */,
+    0x0c87c000 /* 0000321f - 0000321f [    1] */,
+    0x29234002 /* 0000a48d - 0000a48f [    3] */,
+    0x2931c008 /* 0000a4c7 - 0000a4cf [    9] */,
+    0x298b0013 /* 0000a62c - 0000a63f [   20] */,
+    0x29be0007 /* 0000a6f8 - 0000a6ff [    8] */,
+    0x29f2c004 /* 0000a7cb - 0000a7cf [    5] */,
+    0x29f48000 /* 0000a7d2 - 0000a7d2 [    1] */,
+    0x29f50000 /* 0000a7d4 - 0000a7d4 [    1] */,
+    0x29f68017 /* 0000a7da - 0000a7f1 [   24] */,
+    0x2a0b4002 /* 0000a82d - 0000a82f [    3] */,
+    0x2a0e8005 /* 0000a83a - 0000a83f [    6] */,
+    0x2a1e0007 /* 0000a878 - 0000a87f [    8] */,
+    0x2a318007 /* 0000a8c6 - 0000a8cd [    8] */,
+    0x2a368005 /* 0000a8da - 0000a8df [    6] */,
+    0x2a55000a /* 0000a954 - 0000a95e [   11] */,
+    0x2a5f4002 /* 0000a97d - 0000a97f [    3] */,
+    0x2a738000 /* 0000a9ce - 0000a9ce [    1] */,
+    0x2a768003 /* 0000a9da - 0000a9dd [    4] */,
+    0x2a7fc000 /* 0000a9ff - 0000a9ff [    1] */,
+    0x2a8dc008 /* 0000aa37 - 0000aa3f [    9] */,
+    0x2a938001 /* 0000aa4e - 0000aa4f [    2] */,
+    0x2a968001 /* 0000aa5a - 0000aa5b [    2] */,
+    0x2ab0c017 /* 0000aac3 - 0000aada [   24] */,
+    0x2abdc009 /* 0000aaf7 - 0000ab00 [   10] */,
+    0x2ac1c001 /* 0000ab07 - 0000ab08 [    2] */,
+    0x2ac3c001 /* 0000ab0f - 0000ab10 [    2] */,
+    0x2ac5c008 /* 0000ab17 - 0000ab1f [    9] */,
+    0x2ac9c000 /* 0000ab27 - 0000ab27 [    1] */,
+    0x2acbc000 /* 0000ab2f - 0000ab2f [    1] */,
+    0x2adb0003 /* 0000ab6c - 0000ab6f [    4] */,
+    0x2afb8001 /* 0000abee - 0000abef [    2] */,
+    0x2afe8005 /* 0000abfa - 0000abff [    6] */,
+    0x35e9000b /* 0000d7a4 - 0000d7af [   12] */,
+    0x35f1c003 /* 0000d7c7 - 0000d7ca [    4] */,
+    0x35ff2103 /* 0000d7fc - 0000f8ff [ 8452] */,
+    0x3e9b8001 /* 0000fa6e - 0000fa6f [    2] */,
+    0x3eb68025 /* 0000fada - 0000faff [   38] */,
+    0x3ec1c00b /* 0000fb07 - 0000fb12 [   12] */,
+    0x3ec60004 /* 0000fb18 - 0000fb1c [    5] */,
+    0x3ecdc000 /* 0000fb37 - 0000fb37 [    1] */,
+    0x3ecf4000 /* 0000fb3d - 0000fb3d [    1] */,
+    0x3ecfc000 /* 0000fb3f - 0000fb3f [    1] */,
+    0x3ed08000 /* 0000fb42 - 0000fb42 [    1] */,
+    0x3ed14000 /* 0000fb45 - 0000fb45 [    1] */,
+    0x3ef0c00f /* 0000fbc3 - 0000fbd2 [   16] */,
+    0x3f640001 /* 0000fd90 - 0000fd91 [    2] */,
+    0x3f720006 /* 0000fdc8 - 0000fdce [    7] */,
+    0x3f74001f /* 0000fdd0 - 0000fdef [   32] */,
+    0x3f868005 /* 0000fe1a - 0000fe1f [    6] */,
+    0x3f94c000 /* 0000fe53 - 0000fe53 [    1] */,
+    0x3f99c000 /* 0000fe67 - 0000fe67 [    1] */,
+    0x3f9b0003 /* 0000fe6c - 0000fe6f [    4] */,
+    0x3f9d4000 /* 0000fe75 - 0000fe75 [    1] */,
+    0x3fbf4003 /* 0000fefd - 0000ff00 [    4] */,
+    0x3fefc002 /* 0000ffbf - 0000ffc1 [    3] */,
+    0x3ff20001 /* 0000ffc8 - 0000ffc9 [    2] */,
+    0x3ff40001 /* 0000ffd0 - 0000ffd1 [    2] */,
+    0x3ff60001 /* 0000ffd8 - 0000ffd9 [    2] */,
+    0x3ff74002 /* 0000ffdd - 0000ffdf [    3] */,
+    0x3ff9c000 /* 0000ffe7 - 0000ffe7 [    1] */,
+    0x3ffbc00c /* 0000ffef - 0000fffb [   13] */,
+    0x3fff8001 /* 0000fffe - 0000ffff [    2] */,
+    0x40030000 /* 0001000c - 0001000c [    1] */,
+    0x4009c000 /* 00010027 - 00010027 [    1] */,
+    0x400ec000 /* 0001003b - 0001003b [    1] */,
+    0x400f8000 /* 0001003e - 0001003e [    1] */,
+    0x40138001 /* 0001004e - 0001004f [    2] */,
+    0x40178021 /* 0001005e - 0001007f [   34] */,
+    0x403ec004 /* 000100fb - 000100ff [    5] */,
+    0x4040c003 /* 00010103 - 00010106 [    4] */,
+    0x404d0002 /* 00010134 - 00010136 [    3] */,
+    0x4063c000 /* 0001018f - 0001018f [    1] */,
+    0x40674002 /* 0001019d - 0001019f [    3] */,
+    0x4068402e /* 000101a1 - 000101cf [   47] */,
+    0x407f8081 /* 000101fe - 0001027f [  130] */,
+    0x40a74002 /* 0001029d - 0001029f [    3] */,
+    0x40b4400e /* 000102d1 - 000102df [   15] */,
+    0x40bf0003 /* 000102fc - 000102ff [    4] */,
+    0x40c90008 /* 00010324 - 0001032c [    9] */,
+    0x40d2c004 /* 0001034b - 0001034f [    5] */,
+    0x40dec004 /* 0001037b - 0001037f [    5] */,
+    0x40e78000 /* 0001039e - 0001039e [    1] */,
+    0x40f10003 /* 000103c4 - 000103c7 [    4] */,
+    0x40f58029 /* 000103d6 - 000103ff [   42] */,
+    0x41278001 /* 0001049e - 0001049f [    2] */,
+    0x412a8005 /* 000104aa - 000104af [    6] */,
+    0x41350003 /* 000104d4 - 000104d7 [    4] */,
+    0x413f0003 /* 000104fc - 000104ff [    4] */,
+    0x414a0007 /* 00010528 - 0001052f [    8] */,
+    0x4159000a /* 00010564 - 0001056e [   11] */,
+    0x415ec000 /* 0001057b - 0001057b [    1] */,
+    0x4162c000 /* 0001058b - 0001058b [    1] */,
+    0x4164c000 /* 00010593 - 00010593 [    1] */,
+    0x41658000 /* 00010596 - 00010596 [    1] */,
+    0x41688000 /* 000105a2 - 000105a2 [    1] */,
+    0x416c8000 /* 000105b2 - 000105b2 [    1] */,
+    0x416e8000 /* 000105ba - 000105ba [    1] */,
+    0x416f4042 /* 000105bd - 000105ff [   67] */,
+    0x41cdc008 /* 00010737 - 0001073f [    9] */,
+    0x41d58009 /* 00010756 - 0001075f [   10] */,
+    0x41da0017 /* 00010768 - 0001077f [   24] */,
+    0x41e18000 /* 00010786 - 00010786 [    1] */,
+    0x41ec4000 /* 000107b1 - 000107b1 [    1] */,
+    0x41eec044 /* 000107bb - 000107ff [   69] */,
+    0x42018001 /* 00010806 - 00010807 [    2] */,
+    0x42024000 /* 00010809 - 00010809 [    1] */,
+    0x420d8000 /* 00010836 - 00010836 [    1] */,
+    0x420e4002 /* 00010839 - 0001083b [    3] */,
+    0x420f4001 /* 0001083d - 0001083e [    2] */,
+    0x42158000 /* 00010856 - 00010856 [    1] */,
+    0x4227c007 /* 0001089f - 000108a6 [    8] */,
+    0x422c002f /* 000108b0 - 000108df [   48] */,
+    0x423cc000 /* 000108f3 - 000108f3 [    1] */,
+    0x423d8004 /* 000108f6 - 000108fa [    5] */,
+    0x42470002 /* 0001091c - 0001091e [    3] */,
+    0x424e8004 /* 0001093a - 0001093e [    5] */,
+    0x4250003f /* 00010940 - 0001097f [   64] */,
+    0x426e0003 /* 000109b8 - 000109bb [    4] */,
+    0x42740001 /* 000109d0 - 000109d1 [    2] */,
+    0x42810000 /* 00010a04 - 00010a04 [    1] */,
+    0x4281c004 /* 00010a07 - 00010a0b [    5] */,
+    0x42850000 /* 00010a14 - 00010a14 [    1] */,
+    0x42860000 /* 00010a18 - 00010a18 [    1] */,
+    0x428d8001 /* 00010a36 - 00010a37 [    2] */,
+    0x428ec003 /* 00010a3b - 00010a3e [    4] */,
+    0x42924006 /* 00010a49 - 00010a4f [    7] */,
+    0x42964006 /* 00010a59 - 00010a5f [    7] */,
+    0x42a8001f /* 00010aa0 - 00010abf [   32] */,
+    0x42b9c003 /* 00010ae7 - 00010aea [    4] */,
+    0x42bdc008 /* 00010af7 - 00010aff [    9] */,
+    0x42cd8002 /* 00010b36 - 00010b38 [    3] */,
+    0x42d58001 /* 00010b56 - 00010b57 [    2] */,
+    0x42dcc004 /* 00010b73 - 00010b77 [    5] */,
+    0x42e48006 /* 00010b92 - 00010b98 [    7] */,
+    0x42e7400b /* 00010b9d - 00010ba8 [   12] */,
+    0x42ec004f /* 00010bb0 - 00010bff [   80] */,
+    0x43124036 /* 00010c49 - 00010c7f [   55] */,
+    0x432cc00c /* 00010cb3 - 00010cbf [   13] */,
+    0x433cc006 /* 00010cf3 - 00010cf9 [    7] */,
+    0x434a0007 /* 00010d28 - 00010d2f [    8] */,
+    0x434e8125 /* 00010d3a - 00010e5f [  294] */,
+    0x439fc000 /* 00010e7f - 00010e7f [    1] */,
+    0x43aa8000 /* 00010eaa - 00010eaa [    1] */,
+    0x43ab8001 /* 00010eae - 00010eaf [    2] */,
+    0x43ac804a /* 00010eb2 - 00010efc [   75] */,
+    0x43ca0007 /* 00010f28 - 00010f2f [    8] */,
+    0x43d68015 /* 00010f5a - 00010f6f [   22] */,
+    0x43e28025 /* 00010f8a - 00010faf [   38] */,
+    0x43f30013 /* 00010fcc - 00010fdf [   20] */,
+    0x43fdc008 /* 00010ff7 - 00010fff [    9] */,
+    0x44138003 /* 0001104e - 00011051 [    4] */,
+    0x441d8008 /* 00011076 - 0001107e [    9] */,
+    0x442f4000 /* 000110bd - 000110bd [    1] */,
+    0x4430c00c /* 000110c3 - 000110cf [   13] */,
+    0x443a4006 /* 000110e9 - 000110ef [    7] */,
+    0x443e8005 /* 000110fa - 000110ff [    6] */,
+    0x444d4000 /* 00011135 - 00011135 [    1] */,
+    0x44520007 /* 00011148 - 0001114f [    8] */,
+    0x445dc008 /* 00011177 - 0001117f [    9] */,
+    0x44780000 /* 000111e0 - 000111e0 [    1] */,
+    0x447d400a /* 000111f5 - 000111ff [   11] */,
+    0x44848000 /* 00011212 - 00011212 [    1] */,
+    0x4490803d /* 00011242 - 0001127f [   62] */,
+    0x44a1c000 /* 00011287 - 00011287 [    1] */,
+    0x44a24000 /* 00011289 - 00011289 [    1] */,
+    0x44a38000 /* 0001128e - 0001128e [    1] */,
+    0x44a78000 /* 0001129e - 0001129e [    1] */,
+    0x44aa8005 /* 000112aa - 000112af [    6] */,
+    0x44bac004 /* 000112eb - 000112ef [    5] */,
+    0x44be8005 /* 000112fa - 000112ff [    6] */,
+    0x44c10000 /* 00011304 - 00011304 [    1] */,
+    0x44c34001 /* 0001130d - 0001130e [    2] */,
+    0x44c44001 /* 00011311 - 00011312 [    2] */,
+    0x44ca4000 /* 00011329 - 00011329 [    1] */,
+    0x44cc4000 /* 00011331 - 00011331 [    1] */,
+    0x44cd0000 /* 00011334 - 00011334 [    1] */,
+    0x44ce8000 /* 0001133a - 0001133a [    1] */,
+    0x44d14001 /* 00011345 - 00011346 [    2] */,
+    0x44d24001 /* 00011349 - 0001134a [    2] */,
+    0x44d38001 /* 0001134e - 0001134f [    2] */,
+    0x44d44005 /* 00011351 - 00011356 [    6] */,
+    0x44d60004 /* 00011358 - 0001135c [    5] */,
+    0x44d90001 /* 00011364 - 00011365 [    2] */,
+    0x44db4002 /* 0001136d - 0001136f [    3] */,
+    0x44dd408a /* 00011375 - 000113ff [  139] */,
+    0x45170000 /* 0001145c - 0001145c [    1] */,
+    0x4518801d /* 00011462 - 0001147f [   30] */,
+    0x45320007 /* 000114c8 - 000114cf [    8] */,
+    0x453680a5 /* 000114da - 0001157f [  166] */,
+    0x456d8001 /* 000115b6 - 000115b7 [    2] */,
+    0x45778021 /* 000115de - 000115ff [   34] */,
+    0x4591400a /* 00011645 - 0001164f [   11] */,
+    0x45968005 /* 0001165a - 0001165f [    6] */,
+    0x459b4012 /* 0001166d - 0001167f [   19] */,
+    0x45ae8005 /* 000116ba - 000116bf [    6] */,
+    0x45b28035 /* 000116ca - 000116ff [   54] */,
+    0x45c6c001 /* 0001171b - 0001171c [    2] */,
+    0x45cb0003 /* 0001172c - 0001172f [    4] */,
+    0x45d1c0b8 /* 00011747 - 000117ff [  185] */,
+    0x460f0063 /* 0001183c - 0001189f [  100] */,
+    0x463cc00b /* 000118f3 - 000118fe [   12] */,
+    0x4641c001 /* 00011907 - 00011908 [    2] */,
+    0x46428001 /* 0001190a - 0001190b [    2] */,
+    0x46450000 /* 00011914 - 00011914 [    1] */,
+    0x4645c000 /* 00011917 - 00011917 [    1] */,
+    0x464d8000 /* 00011936 - 00011936 [    1] */,
+    0x464e4001 /* 00011939 - 0001193a [    2] */,
+    0x4651c008 /* 00011947 - 0001194f [    9] */,
+    0x46568045 /* 0001195a - 0001199f [   70] */,
+    0x466a0001 /* 000119a8 - 000119a9 [    2] */,
+    0x46760001 /* 000119d8 - 000119d9 [    2] */,
+    0x4679401a /* 000119e5 - 000119ff [   27] */,
+    0x46920007 /* 00011a48 - 00011a4f [    8] */,
+    0x46a8c00c /* 00011aa3 - 00011aaf [   13] */,
+    0x46be4006 /* 00011af9 - 00011aff [    7] */,
+    0x46c280f5 /* 00011b0a - 00011bff [  246] */,
+    0x47024000 /* 00011c09 - 00011c09 [    1] */,
+    0x470dc000 /* 00011c37 - 00011c37 [    1] */,
+    0x47118009 /* 00011c46 - 00011c4f [   10] */,
+    0x471b4002 /* 00011c6d - 00011c6f [    3] */,
+    0x47240001 /* 00011c90 - 00011c91 [    2] */,
+    0x472a0000 /* 00011ca8 - 00011ca8 [    1] */,
+    0x472dc048 /* 00011cb7 - 00011cff [   73] */,
+    0x4741c000 /* 00011d07 - 00011d07 [    1] */,
+    0x47428000 /* 00011d0a - 00011d0a [    1] */,
+    0x474dc002 /* 00011d37 - 00011d39 [    3] */,
+    0x474ec000 /* 00011d3b - 00011d3b [    1] */,
+    0x474f8000 /* 00011d3e - 00011d3e [    1] */,
+    0x47520007 /* 00011d48 - 00011d4f [    8] */,
+    0x47568005 /* 00011d5a - 00011d5f [    6] */,
+    0x47598000 /* 00011d66 - 00011d66 [    1] */,
+    0x475a4000 /* 00011d69 - 00011d69 [    1] */,
+    0x4763c000 /* 00011d8f - 00011d8f [    1] */,
+    0x47648000 /* 00011d92 - 00011d92 [    1] */,
+    0x47664006 /* 00011d99 - 00011d9f [    7] */,
+    0x476a8135 /* 00011daa - 00011edf [  310] */,
+    0x47be4006 /* 00011ef9 - 00011eff [    7] */,
+    0x47c44000 /* 00011f11 - 00011f11 [    1] */,
+    0x47cec002 /* 00011f3b - 00011f3d [    3] */,
+    0x47d68055 /* 00011f5a - 00011faf [   86] */,
+    0x47ec400e /* 00011fb1 - 00011fbf [   15] */,
+    0x47fc800c /* 00011ff2 - 00011ffe [   13] */,
+    0x48e68065 /* 0001239a - 000123ff [  102] */,
+    0x491bc000 /* 0001246f - 0001246f [    1] */,
+    0x491d400a /* 00012475 - 0001247f [   11] */,
+    0x49510a4b /* 00012544 - 00012f8f [ 2636] */,
+    0x4bfcc00c /* 00012ff3 - 00012fff [   13] */,
+    0x4d0c000f /* 00013430 - 0001343f [   16] */,
+    0x4d158fa9 /* 00013456 - 000143ff [ 4010] */,
+    0x5191e1b8 /* 00014647 - 000167ff [ 8633] */,
+    0x5a8e4006 /* 00016a39 - 00016a3f [    7] */,
+    0x5a97c000 /* 00016a5f - 00016a5f [    1] */,
+    0x5a9a8003 /* 00016a6a - 00016a6d [    4] */,
+    0x5aafc000 /* 00016abf - 00016abf [    1] */,
+    0x5ab28005 /* 00016aca - 00016acf [    6] */,
+    0x5abb8001 /* 00016aee - 00016aef [    2] */,
+    0x5abd8009 /* 00016af6 - 00016aff [   10] */,
+    0x5ad18009 /* 00016b46 - 00016b4f [   10] */,
+    0x5ad68000 /* 00016b5a - 00016b5a [    1] */,
+    0x5ad88000 /* 00016b62 - 00016b62 [    1] */,
+    0x5ade0004 /* 00016b78 - 00016b7c [    5] */,
+    0x5ae402af /* 00016b90 - 00016e3f [  688] */,
+    0x5ba6c064 /* 00016e9b - 00016eff [  101] */,
+    0x5bd2c003 /* 00016f4b - 00016f4e [    4] */,
+    0x5be20006 /* 00016f88 - 00016f8e [    7] */,
+    0x5be8003f /* 00016fa0 - 00016fdf [   64] */,
+    0x5bf9400a /* 00016fe5 - 00016fef [   11] */,
+    0x5bfc800d /* 00016ff2 - 00016fff [   14] */,
+    0x61fe0007 /* 000187f8 - 000187ff [    8] */,
+    0x63358029 /* 00018cd6 - 00018cff [   42] */,
+    0x634262e6 /* 00018d09 - 0001afef [ 8935] */,
+    0x6bfd0000 /* 0001aff4 - 0001aff4 [    1] */,
+    0x6bff0000 /* 0001affc - 0001affc [    1] */,
+    0x6bffc000 /* 0001afff - 0001afff [    1] */,
+    0x6c48c00e /* 0001b123 - 0001b131 [   15] */,
+    0x6c4cc01c /* 0001b133 - 0001b14f [   29] */,
+    0x6c54c001 /* 0001b153 - 0001b154 [    2] */,
+    0x6c55800d /* 0001b156 - 0001b163 [   14] */,
+    0x6c5a0007 /* 0001b168 - 0001b16f [    8] */,
+    0x6cbf0903 /* 0001b2fc - 0001bbff [ 2308] */,
+    0x6f1ac004 /* 0001bc6b - 0001bc6f [    5] */,
+    0x6f1f4002 /* 0001bc7d - 0001bc7f [    3] */,
+    0x6f224006 /* 0001bc89 - 0001bc8f [    7] */,
+    0x6f268001 /* 0001bc9a - 0001bc9b [    2] */,
+    0x6f28125f /* 0001bca0 - 0001ceff [ 4704] */,
+    0x73cb8001 /* 0001cf2e - 0001cf2f [    2] */,
+    0x73d1c008 /* 0001cf47 - 0001cf4f [    9] */,
+    0x73f1003b /* 0001cfc4 - 0001cfff [   60] */,
+    0x743d8009 /* 0001d0f6 - 0001d0ff [   10] */,
+    0x7449c001 /* 0001d127 - 0001d128 [    2] */,
+    0x745cc007 /* 0001d173 - 0001d17a [    8] */,
+    0x747ac014 /* 0001d1eb - 0001d1ff [   21] */,
+    0x74918079 /* 0001d246 - 0001d2bf [  122] */,
+    0x74b5000b /* 0001d2d4 - 0001d2df [   12] */,
+    0x74bd000b /* 0001d2f4 - 0001d2ff [   12] */,
+    0x74d5c008 /* 0001d357 - 0001d35f [    9] */,
+    0x74de4086 /* 0001d379 - 0001d3ff [  135] */,
+    0x75154000 /* 0001d455 - 0001d455 [    1] */,
+    0x75274000 /* 0001d49d - 0001d49d [    1] */,
+    0x75280001 /* 0001d4a0 - 0001d4a1 [    2] */,
+    0x7528c001 /* 0001d4a3 - 0001d4a4 [    2] */,
+    0x7529c001 /* 0001d4a7 - 0001d4a8 [    2] */,
+    0x752b4000 /* 0001d4ad - 0001d4ad [    1] */,
+    0x752e8000 /* 0001d4ba - 0001d4ba [    1] */,
+    0x752f0000 /* 0001d4bc - 0001d4bc [    1] */,
+    0x75310000 /* 0001d4c4 - 0001d4c4 [    1] */,
+    0x75418000 /* 0001d506 - 0001d506 [    1] */,
+    0x7542c001 /* 0001d50b - 0001d50c [    2] */,
+    0x75454000 /* 0001d515 - 0001d515 [    1] */,
+    0x75474000 /* 0001d51d - 0001d51d [    1] */,
+    0x754e8000 /* 0001d53a - 0001d53a [    1] */,
+    0x754fc000 /* 0001d53f - 0001d53f [    1] */,
+    0x75514000 /* 0001d545 - 0001d545 [    1] */,
+    0x7551c002 /* 0001d547 - 0001d549 [    3] */,
+    0x75544000 /* 0001d551 - 0001d551 [    1] */,
+    0x75a98001 /* 0001d6a6 - 0001d6a7 [    2] */,
+    0x75f30001 /* 0001d7cc - 0001d7cd [    2] */,
+    0x76a3000e /* 0001da8c - 0001da9a [   15] */,
+    0x76a80000 /* 0001daa0 - 0001daa0 [    1] */,
+    0x76ac044f /* 0001dab0 - 0001deff [ 1104] */,
+    0x77c7c005 /* 0001df1f - 0001df24 [    6] */,
+    0x77cac0d4 /* 0001df2b - 0001dfff [  213] */,
+    0x7801c000 /* 0001e007 - 0001e007 [    1] */,
+    0x78064001 /* 0001e019 - 0001e01a [    2] */,
+    0x78088000 /* 0001e022 - 0001e022 [    1] */,
+    0x78094000 /* 0001e025 - 0001e025 [    1] */,
+    0x780ac004 /* 0001e02b - 0001e02f [    5] */,
+    0x781b8020 /* 0001e06e - 0001e08e [   33] */,
+    0x7824006f /* 0001e090 - 0001e0ff [  112] */,
+    0x784b4002 /* 0001e12d - 0001e12f [    3] */,
+    0x784f8001 /* 0001e13e - 0001e13f [    2] */,
+    0x78528003 /* 0001e14a - 0001e14d [    4] */,
+    0x7854013f /* 0001e150 - 0001e28f [  320] */,
+    0x78abc010 /* 0001e2af - 0001e2bf [   17] */,
+    0x78be8004 /* 0001e2fa - 0001e2fe [    5] */,
+    0x78c001cf /* 0001e300 - 0001e4cf [  464] */,
+    0x793e82e5 /* 0001e4fa - 0001e7df [  742] */,
+    0x79f9c000 /* 0001e7e7 - 0001e7e7 [    1] */,
+    0x79fb0000 /* 0001e7ec - 0001e7ec [    1] */,
+    0x79fbc000 /* 0001e7ef - 0001e7ef [    1] */,
+    0x79ffc000 /* 0001e7ff - 0001e7ff [    1] */,
+    0x7a314001 /* 0001e8c5 - 0001e8c6 [    2] */,
+    0x7a35c028 /* 0001e8d7 - 0001e8ff [   41] */,
+    0x7a530003 /* 0001e94c - 0001e94f [    4] */,
+    0x7a568003 /* 0001e95a - 0001e95d [    4] */,
+    0x7a580310 /* 0001e960 - 0001ec70 [  785] */,
+    0x7b2d404b /* 0001ecb5 - 0001ed00 [   76] */,
+    0x7b4f80c1 /* 0001ed3e - 0001edff [  194] */,
+    0x7b810000 /* 0001ee04 - 0001ee04 [    1] */,
+    0x7b880000 /* 0001ee20 - 0001ee20 [    1] */,
+    0x7b88c000 /* 0001ee23 - 0001ee23 [    1] */,
+    0x7b894001 /* 0001ee25 - 0001ee26 [    2] */,
+    0x7b8a0000 /* 0001ee28 - 0001ee28 [    1] */,
+    0x7b8cc000 /* 0001ee33 - 0001ee33 [    1] */,
+    0x7b8e0000 /* 0001ee38 - 0001ee38 [    1] */,
+    0x7b8e8000 /* 0001ee3a - 0001ee3a [    1] */,
+    0x7b8f0005 /* 0001ee3c - 0001ee41 [    6] */,
+    0x7b90c003 /* 0001ee43 - 0001ee46 [    4] */,
+    0x7b920000 /* 0001ee48 - 0001ee48 [    1] */,
+    0x7b928000 /* 0001ee4a - 0001ee4a [    1] */,
+    0x7b930000 /* 0001ee4c - 0001ee4c [    1] */,
+    0x7b940000 /* 0001ee50 - 0001ee50 [    1] */,
+    0x7b94c000 /* 0001ee53 - 0001ee53 [    1] */,
+    0x7b954001 /* 0001ee55 - 0001ee56 [    2] */,
+    0x7b960000 /* 0001ee58 - 0001ee58 [    1] */,
+    0x7b968000 /* 0001ee5a - 0001ee5a [    1] */,
+    0x7b970000 /* 0001ee5c - 0001ee5c [    1] */,
+    0x7b978000 /* 0001ee5e - 0001ee5e [    1] */,
+    0x7b980000 /* 0001ee60 - 0001ee60 [    1] */,
+    0x7b98c000 /* 0001ee63 - 0001ee63 [    1] */,
+    0x7b994001 /* 0001ee65 - 0001ee66 [    2] */,
+    0x7b9ac000 /* 0001ee6b - 0001ee6b [    1] */,
+    0x7b9cc000 /* 0001ee73 - 0001ee73 [    1] */,
+    0x7b9e0000 /* 0001ee78 - 0001ee78 [    1] */,
+    0x7b9f4000 /* 0001ee7d - 0001ee7d [    1] */,
+    0x7b9fc000 /* 0001ee7f - 0001ee7f [    1] */,
+    0x7ba28000 /* 0001ee8a - 0001ee8a [    1] */,
+    0x7ba70004 /* 0001ee9c - 0001eea0 [    5] */,
+    0x7ba90000 /* 0001eea4 - 0001eea4 [    1] */,
+    0x7baa8000 /* 0001eeaa - 0001eeaa [    1] */,
+    0x7baf0033 /* 0001eebc - 0001eeef [   52] */,
+    0x7bbc810d /* 0001eef2 - 0001efff [  270] */,
+    0x7c0b0003 /* 0001f02c - 0001f02f [    4] */,
+    0x7c25000b /* 0001f094 - 0001f09f [   12] */,
+    0x7c2bc001 /* 0001f0af - 0001f0b0 [    2] */,
+    0x7c300000 /* 0001f0c0 - 0001f0c0 [    1] */,
+    0x7c340000 /* 0001f0d0 - 0001f0d0 [    1] */,
+    0x7c3d8009 /* 0001f0f6 - 0001f0ff [   10] */,
+    0x7c6b8037 /* 0001f1ae - 0001f1e5 [   56] */,
+    0x7c80c00c /* 0001f203 - 0001f20f [   13] */,
+    0x7c8f0003 /* 0001f23c - 0001f23f [    4] */,
+    0x7c924006 /* 0001f249 - 0001f24f [    7] */,
+    0x7c94800d /* 0001f252 - 0001f25f [   14] */,
+    0x7c998099 /* 0001f266 - 0001f2ff [  154] */,
+    0x7db60003 /* 0001f6d8 - 0001f6db [    4] */,
+    0x7dbb4002 /* 0001f6ed - 0001f6ef [    3] */,
+    0x7dbf4002 /* 0001f6fd - 0001f6ff [    3] */,
+    0x7dddc003 /* 0001f777 - 0001f77a [    4] */,
+    0x7df68005 /* 0001f7da - 0001f7df [    6] */,
+    0x7dfb0003 /* 0001f7ec - 0001f7ef [    4] */,
+    0x7dfc400e /* 0001f7f1 - 0001f7ff [   15] */,
+    0x7e030003 /* 0001f80c - 0001f80f [    4] */,
+    0x7e120007 /* 0001f848 - 0001f84f [    8] */,
+    0x7e168005 /* 0001f85a - 0001f85f [    6] */,
+    0x7e220007 /* 0001f888 - 0001f88f [    8] */,
+    0x7e2b8001 /* 0001f8ae - 0001f8af [    2] */,
+    0x7e2c804d /* 0001f8b2 - 0001f8ff [   78] */,
+    0x7e95000b /* 0001fa54 - 0001fa5f [   12] */,
+    0x7e9b8001 /* 0001fa6e - 0001fa6f [    2] */,
+    0x7e9f4002 /* 0001fa7d - 0001fa7f [    3] */,
+    0x7ea24006 /* 0001fa89 - 0001fa8f [    7] */,
+    0x7eaf8000 /* 0001fabe - 0001fabe [    1] */,
+    0x7eb18007 /* 0001fac6 - 0001facd [    8] */,
+    0x7eb70003 /* 0001fadc - 0001fadf [    4] */,
+    0x7eba4006 /* 0001fae9 - 0001faef [    7] */,
+    0x7ebe4006 /* 0001faf9 - 0001faff [    7] */,
+    0x7ee4c000 /* 0001fb93 - 0001fb93 [    1] */,
+    0x7ef2c024 /* 0001fbcb - 0001fbef [   37] */,
+    0x7efe8405 /* 0001fbfa - 0001ffff [ 1030] */,
+    0xa9b8001f /* 0002a6e0 - 0002a6ff [   32] */,
+    0xadce8005 /* 0002b73a - 0002b73f [    6] */,
+    0xae078001 /* 0002b81e - 0002b81f [    2] */,
+    0xb3a8800d /* 0002cea2 - 0002ceaf [   14] */,
+    0xbaf8400e /* 0002ebe1 - 0002ebef [   15] */,
+    0xbb9789a1 /* 0002ee5e - 0002f7ff [ 2466] */,
+    0xbe8785e1 /* 0002fa1e - 0002ffff [ 1506] */,
+    0xc4d2c004 /* 0003134b - 0003134f [    5] */};
 
+/// Returns whether the code unit needs to be escaped.
+///
 /// At the end of the valid Unicode code points space a lot of code points are
 /// either reserved or a noncharacter. Adding all these entries to the
-/// lookup table would add 446 entries to the table (in Unicode 14).
-/// Instead the only the start of the region is stored, every code point in
-/// this region needs to be escaped.
-inline constexpr uint32_t __unallocated_region_lower_bound = 0x000323b0;
+/// lookup table would greatly increase the size of the table. Instead these
+/// entries are manually processed. In this large area of reserved code points,
+/// there is a small area of extended graphemes that should not be escaped
+/// unconditionally. This is also manually coded. See the generation script for
+/// more details.
 
-/// Returns whether the code unit needs to be escaped.
 ///
 /// \pre The code point is a valid Unicode code point.
 [[nodiscard]] _LIBCPP_HIDE_FROM_ABI constexpr bool __needs_escape(const char32_t __code_point) noexcept {
-  // Since __unallocated_region_lower_bound contains the unshifted range do the
-  // comparison without shifting.
-  if (__code_point >= __unallocated_region_lower_bound)
+
+  // The entries in the gap at the end.
+  if(__code_point >= 0x000e0100 && __code_point <= 0x000e01ef)
+     return false;
+
+  // The entries at the end.
+  if (__code_point >= 0x000323b0)
     return true;
 
-  ptrdiff_t __i = std::ranges::upper_bound(__entries, (__code_point << 11) | 0x7ffu) - __entries;
+  ptrdiff_t __i = std::ranges::upper_bound(__entries, (__code_point << 14) | 0x3fffu) - __entries;
   if (__i == 0)
     return false;
 
   --__i;
-  uint32_t __upper_bound = (__entries[__i] >> 11) + (__entries[__i] & 0x7ffu);
+  uint32_t __upper_bound = (__entries[__i] >> 14) + (__entries[__i] & 0x3fffu);
   return __code_point <= __upper_bound;
 }
 
diff --git a/libcxx/include/__format/write_escaped.h b/libcxx/include/__format/write_escaped.h
index 43a074dd8d70..052ea98c3c3b 100644
--- a/libcxx/include/__format/write_escaped.h
+++ b/libcxx/include/__format/write_escaped.h
@@ -101,15 +101,27 @@ _LIBCPP_HIDE_FROM_ABI void __write_escape_ill_formed_code_unit(basic_string<_Cha
 }
 
 template <class _CharT>
-[[nodiscard]] _LIBCPP_HIDE_FROM_ABI bool __is_escaped_sequence_written(basic_string<_CharT>& __str, char32_t __value) {
+[[nodiscard]] _LIBCPP_HIDE_FROM_ABI bool
+__is_escaped_sequence_written(basic_string<_CharT>& __str, bool __last_escaped, char32_t __value) {
 #  ifdef _LIBCPP_HAS_NO_UNICODE
   // For ASCII assume everything above 127 is printable.
   if (__value > 127)
     return false;
 #  endif
 
+  // [format.string.escaped]/2.2.1.2.1
+  //   CE is UTF-8, UTF-16, or UTF-32 and C corresponds to a Unicode scalar
+  //   value whose Unicode property General_Category has a value in the groups
+  //   Separator (Z) or Other (C), as described by UAX #44 of the Unicode Standard,
   if (!__escaped_output_table::__needs_escape(__value))
-    return false;
+    // [format.string.escaped]/2.2.1.2.2
+    //   CE is UTF-8, UTF-16, or UTF-32 and C corresponds to a Unicode scalar
+    //   value with the Unicode property Grapheme_Extend=Yes as described by UAX
+    //   #44 of the Unicode Standard and C is not immediately preceded in S by a
+    //   character P appended to E without translation to an escape sequence,
+    if (!__last_escaped || __extended_grapheme_custer_property_boundary::__get_property(__value) !=
+                               __extended_grapheme_custer_property_boundary::__property::__Extend)
+      return false;
 
   __formatter::__write_well_formed_escaped_code_unit(__str, __value);
   return true;
@@ -124,8 +136,8 @@ enum class __escape_quotation_mark { __apostrophe, __double_quote };
 
 // [format.string.escaped]/2
 template <class _CharT>
-[[nodiscard]] _LIBCPP_HIDE_FROM_ABI bool
-__is_escaped_sequence_written(basic_string<_CharT>& __str, char32_t __value, __escape_quotation_mark __mark) {
+[[nodiscard]] _LIBCPP_HIDE_FROM_ABI bool __is_escaped_sequence_written(
+    basic_string<_CharT>& __str, char32_t __value, bool __last_escaped, __escape_quotation_mark __mark) {
   // 2.2.1.1 - Mapped character in [tab:format.escape.sequences]
   switch (__value) {
   case _CharT('\t'):
@@ -167,7 +179,7 @@ __is_escaped_sequence_written(basic_string<_CharT>& __str, char32_t __value, __e
   // TODO FMT determine what to do with shift sequences.
 
   // 2.2.1.2.1 and 2.2.1.2.2 - Escape
-  return __formatter::__is_escaped_sequence_written(__str, __formatter::__to_char32(__value));
+  return __formatter::__is_escaped_sequence_written(__str, __last_escaped, __formatter::__to_char32(__value));
 }
 
 template <class _CharT>
@@ -175,11 +187,15 @@ _LIBCPP_HIDE_FROM_ABI void
 __escape(basic_string<_CharT>& __str, basic_string_view<_CharT> __values, __escape_quotation_mark __mark) {
   __unicode::__code_point_view<_CharT> __view{__values.begin(), __values.end()};
 
+  // When the first code unit has the property Grapheme_Extend=Yes it needs to
+  // be escaped. This happens when the previous code unit was also escaped.
+  bool __escape = true;
   while (!__view.__at_end()) {
     auto __first                                  = __view.__position();
     typename __unicode::__consume_result __result = __view.__consume();
     if (__result.__status == __unicode::__consume_result::__ok) {
-      if (!__formatter::__is_escaped_sequence_written(__str, __result.__code_point, __mark))
+      __escape = __formatter::__is_escaped_sequence_written(__str, __result.__code_point, __escape, __mark);
+      if (!__escape)
         // 2.2.1.3 - Add the character
         ranges::copy(__first, __view.__position(), std::back_insert_iterator(__str));
     } else {
diff --git a/libcxx/include/__numeric/saturation_arithmetic.h b/libcxx/include/__numeric/saturation_arithmetic.h
index 41596a0c58e2..2390b42aaec3 100644
--- a/libcxx/include/__numeric/saturation_arithmetic.h
+++ b/libcxx/include/__numeric/saturation_arithmetic.h
@@ -25,10 +25,10 @@ _LIBCPP_PUSH_MACROS
 
 _LIBCPP_BEGIN_NAMESPACE_STD
 
-#if _LIBCPP_STD_VER >= 26
+#if _LIBCPP_STD_VER >= 20
 
 template <__libcpp_integer _Tp>
-_LIBCPP_HIDE_FROM_ABI constexpr _Tp add_sat(_Tp __x, _Tp __y) noexcept {
+_LIBCPP_HIDE_FROM_ABI constexpr _Tp __add_sat(_Tp __x, _Tp __y) noexcept {
   if (_Tp __sum; !__builtin_add_overflow(__x, __y, &__sum))
     return __sum;
   // Handle overflow
@@ -46,7 +46,7 @@ _LIBCPP_HIDE_FROM_ABI constexpr _Tp add_sat(_Tp __x, _Tp __y) noexcept {
 }
 
 template <__libcpp_integer _Tp>
-_LIBCPP_HIDE_FROM_ABI constexpr _Tp sub_sat(_Tp __x, _Tp __y) noexcept {
+_LIBCPP_HIDE_FROM_ABI constexpr _Tp __sub_sat(_Tp __x, _Tp __y) noexcept {
   if (_Tp __sub; !__builtin_sub_overflow(__x, __y, &__sub))
     return __sub;
   // Handle overflow
@@ -65,7 +65,7 @@ _LIBCPP_HIDE_FROM_ABI constexpr _Tp sub_sat(_Tp __x, _Tp __y) noexcept {
 }
 
 template <__libcpp_integer _Tp>
-_LIBCPP_HIDE_FROM_ABI constexpr _Tp mul_sat(_Tp __x, _Tp __y) noexcept {
+_LIBCPP_HIDE_FROM_ABI constexpr _Tp __mul_sat(_Tp __x, _Tp __y) noexcept {
   if (_Tp __mul; !__builtin_mul_overflow(__x, __y, &__mul))
     return __mul;
   // Handle overflow
@@ -81,7 +81,7 @@ _LIBCPP_HIDE_FROM_ABI constexpr _Tp mul_sat(_Tp __x, _Tp __y) noexcept {
 }
 
 template <__libcpp_integer _Tp>
-_LIBCPP_HIDE_FROM_ABI constexpr _Tp div_sat(_Tp __x, _Tp __y) noexcept {
+_LIBCPP_HIDE_FROM_ABI constexpr _Tp __div_sat(_Tp __x, _Tp __y) noexcept {
   _LIBCPP_ASSERT_UNCATEGORIZED(__y != 0, "Division by 0 is undefined");
   if constexpr (__libcpp_unsigned_integer<_Tp>) {
     return __x / __y;
@@ -94,7 +94,7 @@ _LIBCPP_HIDE_FROM_ABI constexpr _Tp div_sat(_Tp __x, _Tp __y) noexcept {
 }
 
 template <__libcpp_integer _Rp, __libcpp_integer _Tp>
-_LIBCPP_HIDE_FROM_ABI constexpr _Rp saturate_cast(_Tp __x) noexcept {
+_LIBCPP_HIDE_FROM_ABI constexpr _Rp __saturate_cast(_Tp __x) noexcept {
   // Saturation is impossible edge case when ((min _Rp) < (min _Tp) && (max _Rp) > (max _Tp)) and it is expected to be
   // optimized out by the compiler.
 
@@ -107,6 +107,35 @@ _LIBCPP_HIDE_FROM_ABI constexpr _Rp saturate_cast(_Tp __x) noexcept {
   return static_cast<_Rp>(__x);
 }
 
+#endif // _LIBCPP_STD_VER >= 20
+
+#if _LIBCPP_STD_VER >= 26
+
+template <__libcpp_integer _Tp>
+_LIBCPP_HIDE_FROM_ABI constexpr _Tp add_sat(_Tp __x, _Tp __y) noexcept {
+  return std::__add_sat(__x, __y);
+}
+
+template <__libcpp_integer _Tp>
+_LIBCPP_HIDE_FROM_ABI constexpr _Tp sub_sat(_Tp __x, _Tp __y) noexcept {
+  return std::__sub_sat(__x, __y);
+}
+
+template <__libcpp_integer _Tp>
+_LIBCPP_HIDE_FROM_ABI constexpr _Tp mul_sat(_Tp __x, _Tp __y) noexcept {
+  return std::__mul_sat(__x, __y);
+}
+
+template <__libcpp_integer _Tp>
+_LIBCPP_HIDE_FROM_ABI constexpr _Tp div_sat(_Tp __x, _Tp __y) noexcept {
+  return std::__div_sat(__x, __y);
+}
+
+template <__libcpp_integer _Rp, __libcpp_integer _Tp>
+_LIBCPP_HIDE_FROM_ABI constexpr _Rp saturate_cast(_Tp __x) noexcept {
+  return std::__saturate_cast<_Rp>(__x);
+}
+
 #endif // _LIBCPP_STD_VER >= 26
 
 _LIBCPP_END_NAMESPACE_STD
diff --git a/libcxx/include/__ranges/range_adaptor.h b/libcxx/include/__ranges/range_adaptor.h
index 726b7eda019e..2da246f24e1d 100644
--- a/libcxx/include/__ranges/range_adaptor.h
+++ b/libcxx/include/__ranges/range_adaptor.h
@@ -19,6 +19,7 @@
 #include <__functional/invoke.h>
 #include <__ranges/concepts.h>
 #include <__type_traits/decay.h>
+#include <__type_traits/is_class.h>
 #include <__type_traits/is_nothrow_constructible.h>
 #include <__type_traits/remove_cvref.h>
 #include <__utility/forward.h>
@@ -35,12 +36,15 @@ _LIBCPP_BEGIN_NAMESPACE_STD
 
 #if _LIBCPP_STD_VER >= 20
 
+namespace ranges {
+
 // CRTP base that one can derive from in order to be considered a range adaptor closure
 // by the library. When deriving from this class, a pipe operator will be provided to
 // make the following hold:
 // - `x | f` is equivalent to `f(x)`
 // - `f1 | f2` is an adaptor closure `g` such that `g(x)` is equivalent to `f2(f1(x))`
 template <class _Tp>
+  requires is_class_v<_Tp> && same_as<_Tp, remove_cv_t<_Tp>>
 struct __range_adaptor_closure;
 
 // Type that wraps an arbitrary function object and makes it into a range adaptor closure,
@@ -52,27 +56,42 @@ struct __range_adaptor_closure_t : _Fn, __range_adaptor_closure<__range_adaptor_
 _LIBCPP_CTAD_SUPPORTED_FOR_TYPE(__range_adaptor_closure_t);
 
 template <class _Tp>
-concept _RangeAdaptorClosure = derived_from<remove_cvref_t<_Tp>, __range_adaptor_closure<remove_cvref_t<_Tp>>>;
+_Tp __derived_from_range_adaptor_closure(__range_adaptor_closure<_Tp>*);
 
 template <class _Tp>
-struct __range_adaptor_closure {
-  template <ranges::viewable_range _View, _RangeAdaptorClosure _Closure>
-    requires same_as<_Tp, remove_cvref_t<_Closure>> && invocable<_Closure, _View>
-  [[nodiscard]] _LIBCPP_HIDE_FROM_ABI friend constexpr decltype(auto)
-  operator|(_View&& __view, _Closure&& __closure) noexcept(is_nothrow_invocable_v<_Closure, _View>) {
-    return std::invoke(std::forward<_Closure>(__closure), std::forward<_View>(__view));
-  }
-
-  template <_RangeAdaptorClosure _Closure, _RangeAdaptorClosure _OtherClosure>
-    requires same_as<_Tp, remove_cvref_t<_Closure>> && constructible_from<decay_t<_Closure>, _Closure> &&
-             constructible_from<decay_t<_OtherClosure>, _OtherClosure>
-  [[nodiscard]] _LIBCPP_HIDE_FROM_ABI friend constexpr auto operator|(_Closure&& __c1, _OtherClosure&& __c2) noexcept(
-      is_nothrow_constructible_v<decay_t<_Closure>, _Closure> &&
-      is_nothrow_constructible_v<decay_t<_OtherClosure>, _OtherClosure>) {
-    return __range_adaptor_closure_t(std::__compose(std::forward<_OtherClosure>(__c2), std::forward<_Closure>(__c1)));
-  }
+concept _RangeAdaptorClosure = !ranges::range<remove_cvref_t<_Tp>> && requires {
+  // Ensure that `remove_cvref_t<_Tp>` is derived from `__range_adaptor_closure<remove_cvref_t<_Tp>>` and isn't derived
+  // from `__range_adaptor_closure<U>` for any other type `U`.
+  { ranges::__derived_from_range_adaptor_closure((remove_cvref_t<_Tp>*)nullptr) } -> same_as<remove_cvref_t<_Tp>>;
 };
 
+template <ranges::range _Range, _RangeAdaptorClosure _Closure>
+  requires invocable<_Closure, _Range>
+[[nodiscard]] _LIBCPP_HIDE_FROM_ABI constexpr decltype(auto)
+operator|(_Range&& __range, _Closure&& __closure) noexcept(is_nothrow_invocable_v<_Closure, _Range>) {
+  return std::invoke(std::forward<_Closure>(__closure), std::forward<_Range>(__range));
+}
+
+template <_RangeAdaptorClosure _Closure, _RangeAdaptorClosure _OtherClosure>
+  requires constructible_from<decay_t<_Closure>, _Closure> && constructible_from<decay_t<_OtherClosure>, _OtherClosure>
+[[nodiscard]] _LIBCPP_HIDE_FROM_ABI constexpr auto operator|(_Closure&& __c1, _OtherClosure&& __c2) noexcept(
+    is_nothrow_constructible_v<decay_t<_Closure>, _Closure> &&
+    is_nothrow_constructible_v<decay_t<_OtherClosure>, _OtherClosure>) {
+  return __range_adaptor_closure_t(std::__compose(std::forward<_OtherClosure>(__c2), std::forward<_Closure>(__c1)));
+}
+
+template <class _Tp>
+  requires is_class_v<_Tp> && same_as<_Tp, remove_cv_t<_Tp>>
+struct __range_adaptor_closure {};
+
+#  if _LIBCPP_STD_VER >= 23
+template <class _Tp>
+  requires is_class_v<_Tp> && same_as<_Tp, remove_cv_t<_Tp>>
+class range_adaptor_closure : public __range_adaptor_closure<_Tp> {};
+#  endif // _LIBCPP_STD_VER >= 23
+
+} // namespace ranges
+
 #endif // _LIBCPP_STD_VER >= 20
 
 _LIBCPP_END_NAMESPACE_STD
diff --git a/libcxx/include/__ranges/to.h b/libcxx/include/__ranges/to.h
index 8a815bce5811..e0abe6290b8f 100644
--- a/libcxx/include/__ranges/to.h
+++ b/libcxx/include/__ranges/to.h
@@ -24,6 +24,7 @@
 #include <__ranges/concepts.h>
 #include <__ranges/from_range.h>
 #include <__ranges/range_adaptor.h>
+#include <__ranges/ref_view.h>
 #include <__ranges/size.h>
 #include <__ranges/transform_view.h>
 #include <__type_traits/add_pointer.h>
@@ -129,7 +130,7 @@ template <class _Container, input_range _Range, class... _Args>
     // Try the recursive case.
   } else if constexpr (input_range<range_reference_t<_Range>>) {
     return ranges::to<_Container>(
-        __range | views::transform([](auto&& __elem) {
+        ref_view(__range) | views::transform([](auto&& __elem) {
           return ranges::to<range_value_t<_Container>>(std::forward<decltype(__elem)>(__elem));
         }),
         std::forward<_Args>(__args)...);
diff --git a/libcxx/include/__string/char_traits.h b/libcxx/include/__string/char_traits.h
index 1fd22d518e1a..9d347b188ee1 100644
--- a/libcxx/include/__string/char_traits.h
+++ b/libcxx/include/__string/char_traits.h
@@ -344,7 +344,7 @@ struct _LIBCPP_TEMPLATE_VIS char_traits<char16_t> {
   _LIBCPP_HIDE_FROM_ABI static _LIBCPP_CONSTEXPR_SINCE_CXX17 const char_type*
   find(const char_type* __s, size_t __n, const char_type& __a) _NOEXCEPT {
     __identity __proj;
-    const char_type* __match = std::__find_impl(__s, __s + __n, __a, __proj);
+    const char_type* __match = std::__find(__s, __s + __n, __a, __proj);
     if (__match == __s + __n)
       return nullptr;
     return __match;
@@ -430,7 +430,7 @@ struct _LIBCPP_TEMPLATE_VIS char_traits<char32_t> {
   _LIBCPP_HIDE_FROM_ABI static _LIBCPP_CONSTEXPR_SINCE_CXX17 const char_type*
   find(const char_type* __s, size_t __n, const char_type& __a) _NOEXCEPT {
     __identity __proj;
-    const char_type* __match = std::__find_impl(__s, __s + __n, __a, __proj);
+    const char_type* __match = std::__find(__s, __s + __n, __a, __proj);
     if (__match == __s + __n)
       return nullptr;
     return __match;
diff --git a/libcxx/include/__string/constexpr_c_functions.h b/libcxx/include/__string/constexpr_c_functions.h
index 72c6ce69b60b..4da8542e3807 100644
--- a/libcxx/include/__string/constexpr_c_functions.h
+++ b/libcxx/include/__string/constexpr_c_functions.h
@@ -224,7 +224,7 @@ __constexpr_memmove(_Tp* __dest, _Up* __src, __element_count __n) {
         std::__assign_trivially_copyable(__dest[__i], __src[__i]);
     }
   } else if (__count > 0) {
-    ::__builtin_memmove(__dest, __src, (__count - 1) * sizeof(_Tp) + __libcpp_datasizeof<_Tp>::value);
+    ::__builtin_memmove(__dest, __src, (__count - 1) * sizeof(_Tp) + __datasizeof_v<_Tp>);
   }
   return __dest;
 }
diff --git a/libcxx/include/__type_traits/datasizeof.h b/libcxx/include/__type_traits/datasizeof.h
index 3a8b15160107..54fde242ebcd 100644
--- a/libcxx/include/__type_traits/datasizeof.h
+++ b/libcxx/include/__type_traits/datasizeof.h
@@ -26,39 +26,38 @@
 
 _LIBCPP_BEGIN_NAMESPACE_STD
 
-template <class _Tp>
-struct __libcpp_datasizeof {
 #if __has_extension(datasizeof)
-  static const size_t value = __datasizeof(_Tp);
+template <class _Tp>
+inline const size_t __datasizeof_v = __datasizeof(_Tp);
 #else
 // NOLINTNEXTLINE(readability-redundant-preprocessor) This is https://llvm.org/PR64825
 #  if __has_cpp_attribute(__no_unique_address__)
-  template <class = char>
-  struct _FirstPaddingByte {
-    [[__no_unique_address__]] _Tp __v_;
-    char __first_padding_byte_;
-  };
+template <class _Tp>
+struct _FirstPaddingByte {
+  [[__no_unique_address__]] _Tp __v_;
+  char __first_padding_byte_;
+};
 #  else
-  template <bool = __libcpp_is_final<_Tp>::value || !is_class<_Tp>::value>
-  struct _FirstPaddingByte : _Tp {
-    char __first_padding_byte_;
-  };
+template <class _Tp, bool = __libcpp_is_final<_Tp>::value || !is_class<_Tp>::value>
+struct _FirstPaddingByte : _Tp {
+  char __first_padding_byte_;
+};
 
-  template <>
-  struct _FirstPaddingByte<true> {
-    _Tp __v_;
-    char __first_padding_byte_;
-  };
+template <class _Tp>
+struct _FirstPaddingByte<_Tp, true> {
+  _Tp __v_;
+  char __first_padding_byte_;
+};
 #  endif // __has_cpp_attribute(__no_unique_address__)
 
-  // _FirstPaddingByte<> is sometimes non-standard layout. Using `offsetof` is UB in that case, but GCC and Clang allow
-  // the use as an extension.
-  _LIBCPP_DIAGNOSTIC_PUSH
-  _LIBCPP_CLANG_DIAGNOSTIC_IGNORED("-Winvalid-offsetof")
-  static const size_t value = offsetof(_FirstPaddingByte<>, __first_padding_byte_);
-  _LIBCPP_DIAGNOSTIC_POP
+// _FirstPaddingByte<> is sometimes non-standard layout. Using `offsetof` is UB in that case, but GCC and Clang allow
+// the use as an extension.
+_LIBCPP_DIAGNOSTIC_PUSH
+_LIBCPP_CLANG_DIAGNOSTIC_IGNORED("-Winvalid-offsetof")
+template <class _Tp>
+inline const size_t __datasizeof_v = offsetof(_FirstPaddingByte<_Tp>, __first_padding_byte_);
+_LIBCPP_DIAGNOSTIC_POP
 #endif   // __has_extension(datasizeof)
-};
 
 _LIBCPP_END_NAMESPACE_STD
 
diff --git a/libcxx/include/__utility/no_destroy.h b/libcxx/include/__utility/no_destroy.h
index f9c1eb7bed45..8edd194577d7 100644
--- a/libcxx/include/__utility/no_destroy.h
+++ b/libcxx/include/__utility/no_destroy.h
@@ -12,6 +12,7 @@
 #include <__config>
 #include <__type_traits/is_constant_evaluated.h>
 #include <__utility/forward.h>
+#include <new>
 
 #if !defined(_LIBCPP_HAS_NO_PRAGMA_SYSTEM_HEADER)
 #  pragma GCC system_header
@@ -29,33 +30,23 @@ struct __uninitialized_tag {};
 // initialization using __emplace.
 template <class _Tp>
 struct __no_destroy {
-  _LIBCPP_CONSTEXPR_SINCE_CXX14 _LIBCPP_HIDE_FROM_ABI explicit __no_destroy(__uninitialized_tag) : __dummy_() {
-    if (__libcpp_is_constant_evaluated()) {
-      __dummy_ = char();
-    }
-  }
-  _LIBCPP_HIDE_FROM_ABI ~__no_destroy() {
-    // nothing
-  }
+  _LIBCPP_HIDE_FROM_ABI _LIBCPP_CONSTEXPR explicit __no_destroy(__uninitialized_tag) : __obj_() {}
 
   template <class... _Args>
-  _LIBCPP_CONSTEXPR _LIBCPP_HIDE_FROM_ABI explicit __no_destroy(_Args&&... __args)
-      : __obj_(std::forward<_Args>(__args)...) {}
+  _LIBCPP_HIDE_FROM_ABI explicit __no_destroy(_Args&&... __args) {
+    ::new ((void*)__obj_) _Tp(std::forward<_Args>(__args)...);
+  }
 
   template <class... _Args>
-  _LIBCPP_CONSTEXPR_SINCE_CXX14 _LIBCPP_HIDE_FROM_ABI _Tp& __emplace(_Args&&... __args) {
-    new (&__obj_) _Tp(std::forward<_Args>(__args)...);
-    return __obj_;
+  _LIBCPP_HIDE_FROM_ABI _Tp& __emplace(_Args&&... __args) {
+    return *(::new ((void*)__obj_) _Tp(std::forward<_Args>(__args)...));
   }
 
-  _LIBCPP_CONSTEXPR_SINCE_CXX14 _LIBCPP_HIDE_FROM_ABI _Tp& __get() { return __obj_; }
-  _LIBCPP_CONSTEXPR_SINCE_CXX14 _LIBCPP_HIDE_FROM_ABI _Tp const& __get() const { return __obj_; }
+  _LIBCPP_HIDE_FROM_ABI _Tp& __get() { return *reinterpret_cast<_Tp*>(__obj_); }
+  _LIBCPP_HIDE_FROM_ABI _Tp const& __get() const { return *reinterpret_cast<const _Tp*>(__obj_); }
 
 private:
-  union {
-    _Tp __obj_;
-    char __dummy_; // so we can initialize a member even with __uninitialized_tag for constexpr-friendliness
-  };
+  _ALIGNAS_TYPE(_Tp) char __obj_[sizeof(_Tp)];
 };
 
 _LIBCPP_END_NAMESPACE_STD
diff --git a/libcxx/include/format b/libcxx/include/format
index f1e87de0f830..07c2ba083199 100644
--- a/libcxx/include/format
+++ b/libcxx/include/format
@@ -193,6 +193,8 @@ namespace std {
 #include <__format/concepts.h>
 #include <__format/container_adaptor.h>
 #include <__format/enable_insertable.h>
+#include <__format/escaped_output_table.h>
+#include <__format/extended_grapheme_cluster_table.h>
 #include <__format/format_arg.h>
 #include <__format/format_arg_store.h>
 #include <__format/format_args.h>
diff --git a/libcxx/include/ranges b/libcxx/include/ranges
index 167d2137eaf4..07a525ed8641 100644
--- a/libcxx/include/ranges
+++ b/libcxx/include/ranges
@@ -93,6 +93,11 @@ namespace std::ranges {
   template<class T>
   concept viewable_range = see below;
 
+  // [range.adaptor.object], range adaptor objects
+  template<class D>
+    requires is_class_v<D> && same_as<D, remove_cv_t<D>>
+  class range_adaptor_closure { };  // Since c++23
+
   // [view.interface], class template view_interface
   template<class D>
     requires is_class_v<D> && same_as<D, remove_cv_t<D>>
diff --git a/libcxx/include/streambuf b/libcxx/include/streambuf
index 7964758c908f..a5b4ab9520ae 100644
--- a/libcxx/include/streambuf
+++ b/libcxx/include/streambuf
@@ -107,10 +107,12 @@ protected:
 
 */
 
+#include <__assert>
 #include <__config>
 #include <__fwd/streambuf.h>
 #include <__locale>
 #include <__type_traits/is_same.h>
+#include <__utility/is_valid_range.h>
 #include <climits>
 #include <ios>
 #include <iosfwd>
@@ -234,6 +236,9 @@ protected:
   inline _LIBCPP_HIDE_FROM_ABI_AFTER_V1 void gbump(int __n) { __ninp_ += __n; }
 
   inline _LIBCPP_HIDE_FROM_ABI_AFTER_V1 void setg(char_type* __gbeg, char_type* __gnext, char_type* __gend) {
+    _LIBCPP_ASSERT_VALID_INPUT_RANGE(std::__is_valid_range(__gbeg, __gnext), "[gbeg, gnext) must be a valid range");
+    _LIBCPP_ASSERT_VALID_INPUT_RANGE(std::__is_valid_range(__gbeg, __gend), "[gbeg, gend) must be a valid range");
+    _LIBCPP_ASSERT_VALID_INPUT_RANGE(std::__is_valid_range(__gnext, __gend), "[gnext, gend) must be a valid range");
     __binp_ = __gbeg;
     __ninp_ = __gnext;
     __einp_ = __gend;
@@ -249,6 +254,7 @@ protected:
   _LIBCPP_HIDE_FROM_ABI void __pbump(streamsize __n) { __nout_ += __n; }
 
   inline _LIBCPP_HIDE_FROM_ABI_AFTER_V1 void setp(char_type* __pbeg, char_type* __pend) {
+    _LIBCPP_ASSERT_VALID_INPUT_RANGE(std::__is_valid_range(__pbeg, __pend), "[pbeg, pend) must be a valid range");
     __bout_ = __nout_ = __pbeg;
     __eout_           = __pend;
   }
diff --git a/libcxx/include/variant b/libcxx/include/variant
index 858a49b980bd..34150bd45284 100644
--- a/libcxx/include/variant
+++ b/libcxx/include/variant
@@ -657,6 +657,10 @@ private:
 
 } // namespace __visitation
 
+// Adding semi-colons in macro expansions helps clang-format to do a better job.
+// This macro is used to avoid compilation errors due to "stray" semi-colons.
+#  define _LIBCPP_EAT_SEMICOLON static_assert(true, "")
+
 template <size_t _Index, class _Tp>
 struct _LIBCPP_TEMPLATE_VIS __alt {
   using __value_type = _Tp;
@@ -691,11 +695,10 @@ union _LIBCPP_TEMPLATE_VIS __union<_DestructibleTrait, _Index> {};
       __union(const __union&) = default;                                                                               \
       __union(__union&&)      = default;                                                                               \
                                                                                                                        \
-      destructor                                                                                                       \
+      destructor;                                                                                                      \
                                                                                                                        \
-          __union&                                                                                                     \
-          operator=(const __union&) = default;                                                                         \
-      __union& operator=(__union&&) = default;                                                                         \
+      __union& operator=(const __union&) = default;                                                                    \
+      __union& operator=(__union&&)      = default;                                                                    \
                                                                                                                        \
     private:                                                                                                           \
       char __dummy;                                                                                                    \
@@ -705,9 +708,10 @@ union _LIBCPP_TEMPLATE_VIS __union<_DestructibleTrait, _Index> {};
       friend struct __access::__union;                                                                                 \
     }
 
-_LIBCPP_VARIANT_UNION(_Trait::_TriviallyAvailable, ~__union() = default;);
-_LIBCPP_VARIANT_UNION(_Trait::_Available, ~__union(){});
-_LIBCPP_VARIANT_UNION(_Trait::_Unavailable, ~__union() = delete;);
+_LIBCPP_VARIANT_UNION(_Trait::_TriviallyAvailable, ~__union() = default);
+_LIBCPP_VARIANT_UNION(
+    _Trait::_Available, _LIBCPP_HIDE_FROM_ABI ~__union() {} _LIBCPP_EAT_SEMICOLON);
+_LIBCPP_VARIANT_UNION(_Trait::_Unavailable, ~__union() = delete);
 
 #  undef _LIBCPP_VARIANT_UNION
 
@@ -761,23 +765,27 @@ class _LIBCPP_TEMPLATE_VIS __dtor;
       using __base_type::__base_type;                                                                                  \
       using __base_type::operator=;                                                                                    \
                                                                                                                        \
-      __dtor(const __dtor&)                       = default;                                                           \
-      __dtor(__dtor&&)                            = default;                                                           \
-      destructor __dtor& operator=(const __dtor&) = default;                                                           \
-      __dtor& operator=(__dtor&&)                 = default;                                                           \
+      __dtor(const __dtor&)            = default;                                                                      \
+      __dtor(__dtor&&)                 = default;                                                                      \
+      __dtor& operator=(const __dtor&) = default;                                                                      \
+      __dtor& operator=(__dtor&&)      = default;                                                                      \
+      destructor;                                                                                                      \
                                                                                                                        \
     protected:                                                                                                         \
-      inline _LIBCPP_HIDE_FROM_ABI destroy                                                                             \
+      inline _LIBCPP_HIDE_FROM_ABI destroy;                                                                            \
     }
 
 _LIBCPP_VARIANT_DESTRUCTOR(
-    _Trait::_TriviallyAvailable, ~__dtor() = default;
-    , void __destroy() noexcept { this->__index = __variant_npos<__index_t>; });
+    _Trait::_TriviallyAvailable,
+    ~__dtor() = default, //
+    _LIBCPP_HIDE_FROM_ABI void __destroy() noexcept {
+      this->__index = __variant_npos<__index_t>;
+    } _LIBCPP_EAT_SEMICOLON);
 
 _LIBCPP_VARIANT_DESTRUCTOR(
     _Trait::_Available,
-    ~__dtor() { __destroy(); },
-    void __destroy() noexcept {
+    _LIBCPP_HIDE_FROM_ABI ~__dtor() { __destroy(); } _LIBCPP_EAT_SEMICOLON,
+    _LIBCPP_HIDE_FROM_ABI void __destroy() noexcept {
       if (!this->valueless_by_exception()) {
         __visitation::__base::__visit_alt(
             [](auto& __alt) noexcept {
@@ -787,9 +795,9 @@ _LIBCPP_VARIANT_DESTRUCTOR(
             *this);
       }
       this->__index = __variant_npos<__index_t>;
-    });
+    } _LIBCPP_EAT_SEMICOLON);
 
-_LIBCPP_VARIANT_DESTRUCTOR(_Trait::_Unavailable, ~__dtor() = delete;, void __destroy() noexcept = delete;);
+_LIBCPP_VARIANT_DESTRUCTOR(_Trait::_Unavailable, ~__dtor() = delete, void __destroy() noexcept = delete);
 
 #  undef _LIBCPP_VARIANT_DESTRUCTOR
 
@@ -839,20 +847,24 @@ class _LIBCPP_TEMPLATE_VIS __move_constructor;
       using __base_type::operator=;                                                                                    \
                                                                                                                        \
       __move_constructor(const __move_constructor&)            = default;                                              \
-      move_constructor ~__move_constructor()                   = default;                                              \
+      ~__move_constructor()                                    = default;                                              \
       __move_constructor& operator=(const __move_constructor&) = default;                                              \
       __move_constructor& operator=(__move_constructor&&)      = default;                                              \
+      move_constructor;                                                                                                \
     }
 
 _LIBCPP_VARIANT_MOVE_CONSTRUCTOR(_Trait::_TriviallyAvailable,
-                                 __move_constructor(__move_constructor&& __that) = default;);
+                                 __move_constructor(__move_constructor&& __that) = default);
 
 _LIBCPP_VARIANT_MOVE_CONSTRUCTOR(
     _Trait::_Available,
-    __move_constructor(__move_constructor&& __that) noexcept(__all<is_nothrow_move_constructible_v<_Types>...>::value)
-    : __move_constructor(__valueless_t{}) { this->__generic_construct(*this, std::move(__that)); });
+    _LIBCPP_HIDE_FROM_ABI __move_constructor(__move_constructor&& __that) noexcept(
+        __all<is_nothrow_move_constructible_v<_Types>...>::value)
+    : __move_constructor(__valueless_t{}) {
+      this->__generic_construct(*this, std::move(__that));
+    } _LIBCPP_EAT_SEMICOLON);
 
-_LIBCPP_VARIANT_MOVE_CONSTRUCTOR(_Trait::_Unavailable, __move_constructor(__move_constructor&&) = delete;);
+_LIBCPP_VARIANT_MOVE_CONSTRUCTOR(_Trait::_Unavailable, __move_constructor(__move_constructor&&) = delete);
 
 #  undef _LIBCPP_VARIANT_MOVE_CONSTRUCTOR
 
@@ -869,20 +881,21 @@ class _LIBCPP_TEMPLATE_VIS __copy_constructor;
       using __base_type::__base_type;                                                                                  \
       using __base_type::operator=;                                                                                    \
                                                                                                                        \
-      copy_constructor __copy_constructor(__copy_constructor&&) = default;                                             \
-      ~__copy_constructor()                                     = default;                                             \
-      __copy_constructor& operator=(const __copy_constructor&)  = default;                                             \
-      __copy_constructor& operator=(__copy_constructor&&)       = default;                                             \
-    }
+      __copy_constructor(__copy_constructor&&)                 = default;                                              \
+      ~__copy_constructor()                                    = default;                                              \
+      __copy_constructor& operator=(const __copy_constructor&) = default;                                              \
+      __copy_constructor& operator=(__copy_constructor&&)      = default;                                              \
+      copy_constructor;                                                                                                \
+    } // namespace __variant_detail
 
 _LIBCPP_VARIANT_COPY_CONSTRUCTOR(_Trait::_TriviallyAvailable,
-                                 __copy_constructor(const __copy_constructor& __that) = default;);
+                                 __copy_constructor(const __copy_constructor& __that) = default);
 
 _LIBCPP_VARIANT_COPY_CONSTRUCTOR(
-    _Trait::_Available, __copy_constructor(const __copy_constructor& __that)
-    : __copy_constructor(__valueless_t{}) { this->__generic_construct(*this, __that); });
+    _Trait::_Available, _LIBCPP_HIDE_FROM_ABI __copy_constructor(const __copy_constructor& __that)
+    : __copy_constructor(__valueless_t{}) { this->__generic_construct(*this, __that); } _LIBCPP_EAT_SEMICOLON);
 
-_LIBCPP_VARIANT_COPY_CONSTRUCTOR(_Trait::_Unavailable, __copy_constructor(const __copy_constructor&) = delete;);
+_LIBCPP_VARIANT_COPY_CONSTRUCTOR(_Trait::_Unavailable, __copy_constructor(const __copy_constructor&) = delete);
 
 #  undef _LIBCPP_VARIANT_COPY_CONSTRUCTOR
 
@@ -955,22 +968,22 @@ class _LIBCPP_TEMPLATE_VIS __move_assignment;
       __move_assignment(__move_assignment&&)                 = default;                                                \
       ~__move_assignment()                                   = default;                                                \
       __move_assignment& operator=(const __move_assignment&) = default;                                                \
-      move_assignment                                                                                                  \
+      move_assignment;                                                                                                 \
     }
 
 _LIBCPP_VARIANT_MOVE_ASSIGNMENT(_Trait::_TriviallyAvailable,
-                                __move_assignment& operator=(__move_assignment&& __that) = default;);
+                                __move_assignment& operator=(__move_assignment&& __that) = default);
 
 _LIBCPP_VARIANT_MOVE_ASSIGNMENT(
     _Trait::_Available,
-    __move_assignment&
+    _LIBCPP_HIDE_FROM_ABI __move_assignment&
     operator=(__move_assignment&& __that) noexcept(
         __all<(is_nothrow_move_constructible_v<_Types> && is_nothrow_move_assignable_v<_Types>)...>::value) {
       this->__generic_assign(std::move(__that));
       return *this;
-    });
+    } _LIBCPP_EAT_SEMICOLON);
 
-_LIBCPP_VARIANT_MOVE_ASSIGNMENT(_Trait::_Unavailable, __move_assignment& operator=(__move_assignment&&) = delete;);
+_LIBCPP_VARIANT_MOVE_ASSIGNMENT(_Trait::_Unavailable, __move_assignment& operator=(__move_assignment&&) = delete);
 
 #  undef _LIBCPP_VARIANT_MOVE_ASSIGNMENT
 
@@ -987,22 +1000,23 @@ class _LIBCPP_TEMPLATE_VIS __copy_assignment;
       using __base_type::__base_type;                                                                                  \
       using __base_type::operator=;                                                                                    \
                                                                                                                        \
-      __copy_assignment(const __copy_assignment&)                       = default;                                     \
-      __copy_assignment(__copy_assignment&&)                            = default;                                     \
-      ~__copy_assignment()                                              = default;                                     \
-      copy_assignment __copy_assignment& operator=(__copy_assignment&&) = default;                                     \
+      __copy_assignment(const __copy_assignment&)       = default;                                                     \
+      __copy_assignment(__copy_assignment&&)            = default;                                                     \
+      ~__copy_assignment()                              = default;                                                     \
+      __copy_assignment& operator=(__copy_assignment&&) = default;                                                     \
+      copy_assignment;                                                                                                 \
     }
 
 _LIBCPP_VARIANT_COPY_ASSIGNMENT(_Trait::_TriviallyAvailable,
-                                __copy_assignment& operator=(const __copy_assignment& __that) = default;);
+                                __copy_assignment& operator=(const __copy_assignment& __that) = default);
 
 _LIBCPP_VARIANT_COPY_ASSIGNMENT(
-    _Trait::_Available, __copy_assignment& operator=(const __copy_assignment& __that) {
+    _Trait::_Available, _LIBCPP_HIDE_FROM_ABI __copy_assignment& operator=(const __copy_assignment& __that) {
       this->__generic_assign(__that);
       return *this;
-    });
+    } _LIBCPP_EAT_SEMICOLON);
 
-_LIBCPP_VARIANT_COPY_ASSIGNMENT(_Trait::_Unavailable, __copy_assignment& operator=(const __copy_assignment&) = delete;);
+_LIBCPP_VARIANT_COPY_ASSIGNMENT(_Trait::_Unavailable, __copy_assignment& operator=(const __copy_assignment&) = delete);
 
 #  undef _LIBCPP_VARIANT_COPY_ASSIGNMENT
 
diff --git a/libcxx/modules/std.compat/cstdlib.inc b/libcxx/modules/std.compat/cstdlib.inc
index a45a0a1caf8b..4783cbf51623 100644
--- a/libcxx/modules/std.compat/cstdlib.inc
+++ b/libcxx/modules/std.compat/cstdlib.inc
@@ -25,7 +25,7 @@ export {
   using ::system;
 
   // [c.malloc], C library memory allocation
-  using ::aligned_alloc;
+  using ::aligned_alloc _LIBCPP_USING_IF_EXISTS;
   using ::calloc;
   using ::free;
   using ::malloc;
diff --git a/libcxx/modules/std/ranges.inc b/libcxx/modules/std/ranges.inc
index 82c7d99f8979..80f31c79a1a4 100644
--- a/libcxx/modules/std/ranges.inc
+++ b/libcxx/modules/std/ranges.inc
@@ -138,8 +138,21 @@ export namespace std {
     }
 #endif // _LIBCPP_HAS_NO_LOCALIZATION
 
+#if _LIBCPP_STD_VER >= 23
     // [range.adaptor.object], range adaptor objects
-    // using std::ranges::range_adaptor_closure;
+    using std::ranges::range_adaptor_closure;
+    // Note: This declaration not in the synopsis or explicitly in the wording.
+    // However it is needed for the range adaptors.
+    // [range.adaptor.object]/3
+    //   The template parameter D for range_adaptor_closure may be an
+    //   incomplete type. If an expression of type cv D is used as an operand
+    //   to the | operator, D shall be complete and model
+    //   derived_from<range_adaptor_closure<D>>. The behavior of an expression
+    //   involving an object of type cv D as an operand to the | operator is
+    //   undefined if overload resolution selects a program-defined operator|
+    //   function.
+    using std::ranges::operator|;
+#endif
 
     // [range.all], all view
     namespace views {
diff --git a/libcxx/src/time_zone.cpp b/libcxx/src/time_zone.cpp
index aef6ac674a11..928f3d2855e4 100644
--- a/libcxx/src/time_zone.cpp
+++ b/libcxx/src/time_zone.cpp
@@ -567,11 +567,22 @@ __first_rule(seconds __stdoff, const vector<__tz::__rule>& __rules) {
         false};
   }
 
-  __named_rule_until __continuation_end{__continuation};
-  if (__time >= __continuation_end.__until() && !__continuation_end.__needs_adjustment())
-    // note std::unexpected<sys_seconds>(__end); is ambiguous with std::unexpected() in <exception>,
-    return __sys_info_result{std::unexpect, __continuation_end.__until()};
+  if (__rule->__save.__time != 0s) {
+    // another fix for America/Punta_Arenas when not at the start of the
+    // sys_info object.
+    seconds __save = __rule->__save.__time;
+    if (__continuation_begin >= __rule_begin - __save && __time < __next.first) {
+      return __sys_info{
+          sys_info{__continuation_begin,
+                   __next.first,
+                   __continuation.__stdoff + __save,
+                   chrono::duration_cast<minutes>(__save),
+                   chrono::__format(__continuation, __rule->__letters, __save)},
+          false};
+    }
+  }
 
+  __named_rule_until __continuation_end{__continuation};
   while (__next.second != __rules.end()) {
 #ifdef PRINT
     std::print(
diff --git a/libcxx/test/libcxx/transitive_includes/cxx20.csv b/libcxx/test/libcxx/transitive_includes/cxx20.csv
index 6b80790a9d19..7d31ba160ee1 100644
--- a/libcxx/test/libcxx/transitive_includes/cxx20.csv
+++ b/libcxx/test/libcxx/transitive_includes/cxx20.csv
@@ -129,6 +129,7 @@ chrono cwchar
 chrono forward_list
 chrono limits
 chrono locale
+chrono new
 chrono optional
 chrono ostream
 chrono ratio
diff --git a/libcxx/test/libcxx/type_traits/datasizeof.compile.pass.cpp b/libcxx/test/libcxx/type_traits/datasizeof.compile.pass.cpp
index 881b0bd85190..03dd0f6eac53 100644
--- a/libcxx/test/libcxx/type_traits/datasizeof.compile.pass.cpp
+++ b/libcxx/test/libcxx/type_traits/datasizeof.compile.pass.cpp
@@ -9,10 +9,10 @@
 #include <__type_traits/datasizeof.h>
 #include <cstdint>
 
-static_assert(std::__libcpp_datasizeof<std::int8_t>::value == 1, "");
-static_assert(std::__libcpp_datasizeof<std::int16_t>::value == 2, "");
-static_assert(std::__libcpp_datasizeof<std::int32_t>::value == 4, "");
-static_assert(std::__libcpp_datasizeof<std::int64_t>::value == 8, "");
+static_assert(std::__datasizeof_v<std::int8_t> == 1, "");
+static_assert(std::__datasizeof_v<std::int16_t> == 2, "");
+static_assert(std::__datasizeof_v<std::int32_t> == 4, "");
+static_assert(std::__datasizeof_v<std::int64_t> == 8, "");
 
 struct OneBytePadding {
   OneBytePadding() {}
@@ -22,9 +22,9 @@ struct OneBytePadding {
 };
 
 #if defined(_WIN32) && !defined(__MINGW32__)
-static_assert(std::__libcpp_datasizeof<OneBytePadding>::value == 4, "");
+static_assert(std::__datasizeof_v<OneBytePadding> == 4, "");
 #else
-static_assert(std::__libcpp_datasizeof<OneBytePadding>::value == 3, "");
+static_assert(std::__datasizeof_v<OneBytePadding> == 3, "");
 #endif
 
 struct InBetweenPadding {
@@ -35,4 +35,4 @@ struct InBetweenPadding {
   std::int16_t c;
 };
 
-static_assert(std::__libcpp_datasizeof<InBetweenPadding>::value == 8, "");
+static_assert(std::__datasizeof_v<InBetweenPadding> == 8, "");
diff --git a/libcxx/test/libcxx/utilities/expected/expected.expected/no_unique_address.compile.pass.cpp b/libcxx/test/libcxx/utilities/expected/expected.expected/no_unique_address.compile.pass.cpp
index cf1909b92873..580c0f4ae10c 100644
--- a/libcxx/test/libcxx/utilities/expected/expected.expected/no_unique_address.compile.pass.cpp
+++ b/libcxx/test/libcxx/utilities/expected/expected.expected/no_unique_address.compile.pass.cpp
@@ -47,28 +47,28 @@ static_assert(sizeof(std::expected<B, B>) == sizeof(B));
 
 // Check that `expected`'s datasize is large enough for the parameter type(s).
 static_assert(sizeof(std::expected<BoolWithPadding, Empty>) ==
-              std::__libcpp_datasizeof<std::expected<BoolWithPadding, Empty>>::value);
+              std::__datasizeof_v<std::expected<BoolWithPadding, Empty>>);
 static_assert(sizeof(std::expected<Empty, BoolWithPadding>) ==
-              std::__libcpp_datasizeof<std::expected<Empty, BoolWithPadding>>::value);
+              std::__datasizeof_v<std::expected<Empty, BoolWithPadding>>);
 
 // In this case, there should be tail padding in the `expected` because `A`
 // itself does _not_ have tail padding.
-static_assert(sizeof(std::expected<A, A>) > std::__libcpp_datasizeof<std::expected<A, A>>::value);
+static_assert(sizeof(std::expected<A, A>) > std::__datasizeof_v<std::expected<A, A>>);
 
 // Test with some real types.
 static_assert(sizeof(std::expected<std::optional<int>, int>) == 8);
-static_assert(std::__libcpp_datasizeof<std::expected<std::optional<int>, int>>::value == 8);
+static_assert(std::__datasizeof_v<std::expected<std::optional<int>, int>> == 8);
 
 static_assert(sizeof(std::expected<int, std::optional<int>>) == 8);
-static_assert(std::__libcpp_datasizeof<std::expected<int, std::optional<int>>>::value == 8);
+static_assert(std::__datasizeof_v<std::expected<int, std::optional<int>>> == 8);
 
 static_assert(sizeof(std::expected<int, int>) == 8);
-static_assert(std::__libcpp_datasizeof<std::expected<int, int>>::value == 5);
+static_assert(std::__datasizeof_v<std::expected<int, int>> == 5);
 
 // clang-format off
-static_assert(std::__libcpp_datasizeof<int>::value == 4);
-static_assert(std::__libcpp_datasizeof<std::expected<int, int>>::value == 5);
-static_assert(std::__libcpp_datasizeof<std::expected<std::expected<int, int>, int>>::value == 8);
-static_assert(std::__libcpp_datasizeof<std::expected<std::expected<std::expected<int, int>, int>, int>>::value == 9);
-static_assert(std::__libcpp_datasizeof<std::expected<std::expected<std::expected<std::expected<int, int>, int>, int>, int>>::value == 12);
+static_assert(std::__datasizeof_v<int> == 4);
+static_assert(std::__datasizeof_v<std::expected<int, int>> == 5);
+static_assert(std::__datasizeof_v<std::expected<std::expected<int, int>, int>> == 8);
+static_assert(std::__datasizeof_v<std::expected<std::expected<std::expected<int, int>, int>, int>> == 9);
+static_assert(std::__datasizeof_v<std::expected<std::expected<std::expected<std::expected<int, int>, int>, int>, int>> == 12);
 // clang-format on
diff --git a/libcxx/test/libcxx/utilities/expected/expected.void/no_unique_address.compile.pass.cpp b/libcxx/test/libcxx/utilities/expected/expected.void/no_unique_address.compile.pass.cpp
index fdee8b71e5d9..27da03c54ac4 100644
--- a/libcxx/test/libcxx/utilities/expected/expected.void/no_unique_address.compile.pass.cpp
+++ b/libcxx/test/libcxx/utilities/expected/expected.void/no_unique_address.compile.pass.cpp
@@ -45,23 +45,23 @@ static_assert(sizeof(std::expected<void, B>) == sizeof(B));
 
 // Check that `expected`'s datasize is large enough for the parameter type(s).
 static_assert(sizeof(std::expected<void, BoolWithPadding>) ==
-              std::__libcpp_datasizeof<std::expected<void, BoolWithPadding>>::value);
+              std::__datasizeof_v<std::expected<void, BoolWithPadding>>);
 
 // In this case, there should be tail padding in the `expected` because `A`
 // itself does _not_ have tail padding.
-static_assert(sizeof(std::expected<void, A>) > std::__libcpp_datasizeof<std::expected<void, A>>::value);
+static_assert(sizeof(std::expected<void, A>) > std::__datasizeof_v<std::expected<void, A>>);
 
 // Test with some real types.
 static_assert(sizeof(std::expected<void, std::optional<int>>) == 8);
-static_assert(std::__libcpp_datasizeof<std::expected<void, std::optional<int>>>::value == 8);
+static_assert(std::__datasizeof_v<std::expected<void, std::optional<int>>> == 8);
 
 static_assert(sizeof(std::expected<void, int>) == 8);
-static_assert(std::__libcpp_datasizeof<std::expected<void, int>>::value == 5);
+static_assert(std::__datasizeof_v<std::expected<void, int>> == 5);
 
 // clang-format off
-static_assert(std::__libcpp_datasizeof<int>::value == 4);
-static_assert(std::__libcpp_datasizeof<std::expected<void, int>>::value == 5);
-static_assert(std::__libcpp_datasizeof<std::expected<void, std::expected<void, int>>>::value == 8);
-static_assert(std::__libcpp_datasizeof<std::expected<void, std::expected<void, std::expected<void, int>>>>::value == 9);
-static_assert(std::__libcpp_datasizeof<std::expected<void, std::expected<void, std::expected<void, std::expected<void, int>>>>>::value == 12);
+static_assert(std::__datasizeof_v<int> == 4);
+static_assert(std::__datasizeof_v<std::expected<void, int>> == 5);
+static_assert(std::__datasizeof_v<std::expected<void, std::expected<void, int>>> == 8);
+static_assert(std::__datasizeof_v<std::expected<void, std::expected<void, std::expected<void, int>>>> == 9);
+static_assert(std::__datasizeof_v<std::expected<void, std::expected<void, std::expected<void, std::expected<void, int>>>>> == 12);
 // clang-format on
diff --git a/libcxx/test/libcxx/utilities/format/format.string/format.string.std/escaped_output.pass.cpp b/libcxx/test/libcxx/utilities/format/format.string/format.string.std/escaped_output.pass.cpp
new file mode 100644
index 000000000000..5b1191642c9a
--- /dev/null
+++ b/libcxx/test/libcxx/utilities/format/format.string/format.string.std/escaped_output.pass.cpp
@@ -0,0 +1,102 @@
+//===----------------------------------------------------------------------===//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+// UNSUPPORTED: c++03, c++11, c++14, c++17, c++20
+// UNSUPPORTED: GCC-ALWAYS_INLINE-FIXME
+
+// <format>
+
+// Tests the properties of the Unicode escaped output table.
+// The libc++ algorithm has size and speed optimizations based on the properties
+// of Unicode. This means updating the Unicode tables has a likilihood of
+// breaking test. This is an assert; it requires validating whether the
+// assumptions of the size and speed optimizations are still valid.
+
+#include <algorithm>
+#include <numeric>
+#include <format>
+#include <cassert>
+
+// Contains the entries for [format.string.escaped]/2.2.1.2.1
+//   CE is a Unicode encoding and C corresponds to a UCS scalar value whose
+//   Unicode property General_Category has a value in the groups Separator (Z)
+//   or Other (C), as described by table 12 of UAX #44
+//
+// Separator (Z) consists of General_Category
+// - Zs Space_Separator,
+// - Zl Line_Separator,
+// - Zp Paragraph_Separator.
+//
+// Other (C) consists of General_Category
+// - Cc Control,
+// - Cf Format,
+// - Cs Surrogate,
+// - Co Private_Use,
+// - Cn Unassigned.
+inline constexpr int Zs = 17;
+inline constexpr int Zl = 1;
+inline constexpr int Zp = 1;
+inline constexpr int Z  = Zs + Zl + Zp;
+
+inline constexpr int Cc = 65;
+inline constexpr int Cf = 170;
+inline constexpr int Cs = 2'048;
+inline constexpr int Co = 137'468;
+inline constexpr int Cn = 824'718;
+inline constexpr int C  = Cc + Cf + Cs + Co + Cn;
+
+// This is the final part of the Unicode properties table:
+//
+// 31350..323AF  ; Lo # [4192] CJK UNIFIED IDEOGRAPH-31350..CJK UNIFIED IDEOGRAPH-323AF
+// 323B0..E0000  ; Cn # [711761] <reserved-323B0>..<reserved-E0000>
+// E0001         ; Cf #       LANGUAGE TAG
+// E0002..E001F  ; Cn #  [30] <reserved-E0002>..<reserved-E001F>
+// E0020..E007F  ; Cf #  [96] TAG SPACE..CANCEL TAG
+// E0080..E00FF  ; Cn # [128] <reserved-E0080>..<reserved-E00FF>
+// E0100..E01EF  ; Mn # [240] VARIATION SELECTOR-17..VARIATION SELECTOR-256
+// E01F0..EFFFF  ; Cn # [65040] <reserved-E01F0>..<noncharacter-EFFFF>
+// F0000..FFFFD  ; Co # [65534] <private-use-F0000>..<private-use-FFFFD>
+// FFFFE..FFFFF  ; Cn #   [2] <noncharacter-FFFFE>..<noncharacter-FFFFF>
+// 100000..10FFFD; Co # [65534] <private-use-100000>..<private-use-10FFFD>
+// 10FFFE..10FFFF; Cn #   [2] <noncharacter-10FFFE>..<noncharacter-10FFFF>
+//
+// It can be observed all entries in the range 323B0..10FFFF are in the
+// categories Cf, Co, Cn, except a small range with the property Mn.
+// In order to reduce the size of the table only the entires in the range
+// [0000, 323B0) are stored in the table. The entries in the range
+// [323B0, 10FFFF] use a hand-crafted algorithm.
+//
+// This means a number of entries are omitted
+inline constexpr int excluded = ((0x10FFFF - 0x323B0) + 1) - 240;
+
+inline constexpr int entries = Z + C - excluded;
+
+static constexpr int count_entries() {
+  return std::transform_reduce(
+      std::begin(std::__escaped_output_table::__entries),
+      std::end(std::__escaped_output_table::__entries),
+      0,
+      std::plus{},
+      [](auto entry) { return 1 + static_cast<int>(entry & 0x3fffu); });
+}
+static_assert(count_entries() == entries);
+
+int main(int, char**) {
+  for (char32_t c = 0x31350; c <= 0x323AF; ++c) // 31350..323AF  ; Lo # [4192]
+    assert(std::__escaped_output_table::__needs_escape(c) == false);
+
+  for (char32_t c = 0x323B0; c <= 0xE00FF; ++c) // 323B0..E00FF ; C
+    assert(std::__escaped_output_table::__needs_escape(c) == true);
+
+  for (char32_t c = 0xE0100; c <= 0xE01EF; ++c) // E0100..E01EF  ; Mn # [240]
+    assert(std::__escaped_output_table::__needs_escape(c) == false);
+
+  for (char32_t c = 0xE01F0; c <= 0x10FFFF; ++c) // E01F0..10FFFF; C
+    assert(std::__escaped_output_table::__needs_escape(c) == true);
+
+  return 0;
+}
diff --git a/libcxx/test/libcxx/utilities/no_destroy.pass.cpp b/libcxx/test/libcxx/utilities/no_destroy.pass.cpp
new file mode 100644
index 000000000000..9a874a640753
--- /dev/null
+++ b/libcxx/test/libcxx/utilities/no_destroy.pass.cpp
@@ -0,0 +1,31 @@
+//===----------------------------------------------------------------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#include <__utility/no_destroy.h>
+#include <cassert>
+
+#include "test_macros.h"
+
+#if TEST_STD_VER > 17
+// Test constexpr-constructibility.
+constinit std::__no_destroy<int> nd_int_const(std::__uninitialized_tag{});
+#endif
+
+struct DestroyLast {
+  ~DestroyLast() { assert(*ptr == 5); }
+
+  int* ptr;
+} last;
+
+static std::__no_destroy<int> nd_int(5);
+
+int main(int, char**) {
+  last.ptr = &nd_int.__get();
+
+  return 0;
+}
diff --git a/libcxx/test/std/atomics/atomics.types.generic/atomics.types.float/fetch_add.pass.cpp b/libcxx/test/std/atomics/atomics.types.generic/atomics.types.float/fetch_add.pass.cpp
index 7350c1ddf0e9..4119c39772e5 100644
--- a/libcxx/test/std/atomics/atomics.types.generic/atomics.types.float/fetch_add.pass.cpp
+++ b/libcxx/test/std/atomics/atomics.types.generic/atomics.types.float/fetch_add.pass.cpp
@@ -6,7 +6,6 @@
 //
 //===----------------------------------------------------------------------===//
 // UNSUPPORTED: c++03, c++11, c++14, c++17
-// UNSUPPORTED: LIBCXX-AIX-FIXME
 // XFAIL: !has-64-bit-atomics
 
 // https://github.com/llvm/llvm-project/issues/72893
diff --git a/libcxx/test/std/atomics/atomics.types.generic/atomics.types.float/fetch_sub.pass.cpp b/libcxx/test/std/atomics/atomics.types.generic/atomics.types.float/fetch_sub.pass.cpp
index 84dcde5f2784..2460765a3c86 100644
--- a/libcxx/test/std/atomics/atomics.types.generic/atomics.types.float/fetch_sub.pass.cpp
+++ b/libcxx/test/std/atomics/atomics.types.generic/atomics.types.float/fetch_sub.pass.cpp
@@ -6,7 +6,6 @@
 //
 //===----------------------------------------------------------------------===//
 // UNSUPPORTED: c++03, c++11, c++14, c++17
-// UNSUPPORTED: LIBCXX-AIX-FIXME
 // XFAIL: !has-64-bit-atomics
 
 // https://github.com/llvm/llvm-project/issues/72893
diff --git a/libcxx/test/std/atomics/atomics.types.generic/atomics.types.float/operator.minus_equals.pass.cpp b/libcxx/test/std/atomics/atomics.types.generic/atomics.types.float/operator.minus_equals.pass.cpp
index 386a393e3550..4bd303022c0d 100644
--- a/libcxx/test/std/atomics/atomics.types.generic/atomics.types.float/operator.minus_equals.pass.cpp
+++ b/libcxx/test/std/atomics/atomics.types.generic/atomics.types.float/operator.minus_equals.pass.cpp
@@ -6,7 +6,6 @@
 //
 //===----------------------------------------------------------------------===//
 // UNSUPPORTED: c++03, c++11, c++14, c++17
-// UNSUPPORTED: LIBCXX-AIX-FIXME
 // XFAIL: !has-64-bit-atomics
 
 // floating-point-type operator-=(floating-point-type) volatile noexcept;
diff --git a/libcxx/test/std/atomics/atomics.types.generic/atomics.types.float/operator.plus_equals.pass.cpp b/libcxx/test/std/atomics/atomics.types.generic/atomics.types.float/operator.plus_equals.pass.cpp
index afd06d537c7a..69abb9ae63c3 100644
--- a/libcxx/test/std/atomics/atomics.types.generic/atomics.types.float/operator.plus_equals.pass.cpp
+++ b/libcxx/test/std/atomics/atomics.types.generic/atomics.types.float/operator.plus_equals.pass.cpp
@@ -6,7 +6,6 @@
 //
 //===----------------------------------------------------------------------===//
 // UNSUPPORTED: c++03, c++11, c++14, c++17
-// UNSUPPORTED: LIBCXX-AIX-FIXME
 // XFAIL: !has-64-bit-atomics
 
 // floating-point-type operator+=(floating-point-type) volatile noexcept;
diff --git a/libcxx/test/std/containers/sequences/array/size_and_alignment.compile.pass.cpp b/libcxx/test/std/containers/sequences/array/size_and_alignment.compile.pass.cpp
index 209e24964807..7ba56577d1bb 100644
--- a/libcxx/test/std/containers/sequences/array/size_and_alignment.compile.pass.cpp
+++ b/libcxx/test/std/containers/sequences/array/size_and_alignment.compile.pass.cpp
@@ -46,7 +46,7 @@ void test_type() {
     static_assert(!std::is_empty<Array>::value, "");
 
     // Make sure empty arrays don't have padding bytes
-    LIBCPP_STATIC_ASSERT(std::__libcpp_datasizeof<Array>::value == sizeof(Array), "");
+    LIBCPP_STATIC_ASSERT(std::__datasizeof_v<Array> == sizeof(Array), "");
   }
 
   {
diff --git a/libcxx/test/std/input.output/stream.buffers/streambuf/streambuf.cons/copy.pass.cpp b/libcxx/test/std/input.output/stream.buffers/streambuf/streambuf.cons/copy.pass.cpp
index 580675119507..b458f93601a1 100644
--- a/libcxx/test/std/input.output/stream.buffers/streambuf/streambuf.cons/copy.pass.cpp
+++ b/libcxx/test/std/input.output/stream.buffers/streambuf/streambuf.cons/copy.pass.cpp
@@ -57,18 +57,20 @@ int main(int, char**)
         test<char> t2 = t;
     }
     {
-        char g1, g2, g3, p1, p3;
+        char g[3];
+        char p[3];
         test<char> t;
-        t.setg(&g1, &g2, &g3);
-        t.setp(&p1, &p3);
+        t.setg(&g[0], &g[1], &g[2]);
+        t.setp(&p[0], &p[2]);
         test<char> t2 = t;
     }
 #ifndef TEST_HAS_NO_WIDE_CHARACTERS
     {
-        wchar_t g1, g2, g3, p1, p3;
+        wchar_t g[3];
+        wchar_t p[3];
         test<wchar_t> t;
-        t.setg(&g1, &g2, &g3);
-        t.setp(&p1, &p3);
+        t.setg(&g[0], &g[1], &g[2]);
+        t.setp(&p[0], &p[2]);
         test<wchar_t> t2 = t;
     }
     {
diff --git a/libcxx/test/std/input.output/stream.buffers/streambuf/streambuf.protected/streambuf.assign/assign.pass.cpp b/libcxx/test/std/input.output/stream.buffers/streambuf/streambuf.protected/streambuf.assign/assign.pass.cpp
index 8a976e77f0f1..45a8cdf3a23f 100644
--- a/libcxx/test/std/input.output/stream.buffers/streambuf/streambuf.protected/streambuf.assign/assign.pass.cpp
+++ b/libcxx/test/std/input.output/stream.buffers/streambuf/streambuf.protected/streambuf.assign/assign.pass.cpp
@@ -59,10 +59,11 @@ int main(int, char**)
         t2 = t;
     }
     {
-        char g1, g2, g3, p1, p3;
+        char g[3];
+        char p[3];
         test<char> t;
-        t.setg(&g1, &g2, &g3);
-        t.setp(&p1, &p3);
+        t.setg(&g[0], &g[1], &g[2]);
+        t.setp(&p[0], &p[2]);
         test<char> t2;
         t2 = t;
     }
@@ -73,10 +74,11 @@ int main(int, char**)
         t2 = t;
     }
     {
-        wchar_t g1, g2, g3, p1, p3;
+        wchar_t g[3];
+        wchar_t p[3];
         test<wchar_t> t;
-        t.setg(&g1, &g2, &g3);
-        t.setp(&p1, &p3);
+        t.setg(&g[0], &g[1], &g[2]);
+        t.setp(&p[0], &p[2]);
         test<wchar_t> t2;
         t2 = t;
     }
diff --git a/libcxx/test/std/input.output/stream.buffers/streambuf/streambuf.protected/streambuf.assign/swap.pass.cpp b/libcxx/test/std/input.output/stream.buffers/streambuf/streambuf.protected/streambuf.assign/swap.pass.cpp
index c575c2cb1271..b90c4c053c91 100644
--- a/libcxx/test/std/input.output/stream.buffers/streambuf/streambuf.protected/streambuf.assign/swap.pass.cpp
+++ b/libcxx/test/std/input.output/stream.buffers/streambuf/streambuf.protected/streambuf.assign/swap.pass.cpp
@@ -68,10 +68,11 @@ int main(int, char**)
         t2.swap(t);
     }
     {
-        char g1, g2, g3, p1, p3;
+        char g[3];
+        char p[3];
         test<char> t;
-        t.setg(&g1, &g2, &g3);
-        t.setp(&p1, &p3);
+        t.setg(&g[0], &g[1], &g[2]);
+        t.setp(&p[0], &p[2]);
         test<char> t2;
         t2.swap(t);
     }
@@ -82,10 +83,11 @@ int main(int, char**)
         t2.swap(t);
     }
     {
-        wchar_t g1, g2, g3, p1, p3;
+        wchar_t g[3];
+        wchar_t p[3];
         test<wchar_t> t;
-        t.setg(&g1, &g2, &g3);
-        t.setp(&p1, &p3);
+        t.setg(&g[0], &g[1], &g[2]);
+        t.setp(&p[0], &p[2]);
         test<wchar_t> t2;
         t2.swap(t);
     }
diff --git a/libcxx/test/std/input.output/stream.buffers/streambuf/streambuf.protected/streambuf.get.area/setg.assert.pass.cpp b/libcxx/test/std/input.output/stream.buffers/streambuf/streambuf.protected/streambuf.get.area/setg.assert.pass.cpp
new file mode 100644
index 000000000000..becf89b12fdd
--- /dev/null
+++ b/libcxx/test/std/input.output/stream.buffers/streambuf/streambuf.protected/streambuf.get.area/setg.assert.pass.cpp
@@ -0,0 +1,68 @@
+//===----------------------------------------------------------------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+// REQUIRES: has-unix-headers
+// UNSUPPORTED: libcpp-hardening-mode=none
+// XFAIL: libcpp-hardening-mode=debug && availability-verbose_abort-missing
+
+// <streambuf>
+
+// template <class charT, class traits = char_traits<charT> >
+// class basic_streambuf;
+
+// void setg(char_type* gbeg, char_type* gnext, char_type* gend);
+
+#include <algorithm>
+#include <iterator>
+#include <streambuf>
+#include <string>
+
+#include "check_assertion.h"
+#include "make_string.h"
+#include "test_macros.h"
+
+template <class CharT>
+struct streambuf : public std::basic_streambuf<CharT> {
+  typedef std::basic_streambuf<CharT> base;
+
+  streambuf() {}
+
+  void setg(CharT* gbeg, CharT* gnext, CharT* gend) { base::setg(gbeg, gnext, gend); }
+};
+
+template <class CharT>
+void test() {
+  std::basic_string<CharT> str = MAKE_STRING(CharT, "ABCDEF");
+  CharT arr[6];
+  std::copy(str.begin(), str.end(), arr);
+
+  {
+    streambuf<CharT> buff;
+    TEST_LIBCPP_ASSERT_FAILURE(
+        buff.setg(std::begin(arr) + 1, std::begin(arr), std::end(arr)), "[gbeg, gnext) must be a valid range");
+  }
+  {
+    streambuf<CharT> buff;
+    TEST_LIBCPP_ASSERT_FAILURE(
+        buff.setg(std::begin(arr) + 1, std::begin(arr) + 1, std::begin(arr)), "[gbeg, gend) must be a valid range");
+  }
+  {
+    streambuf<CharT> buff;
+    TEST_LIBCPP_ASSERT_FAILURE(
+        buff.setg(std::begin(arr), std::begin(arr) + 3, std::begin(arr) + 2), "[gnext, gend) must be a valid range");
+  }
+}
+
+int main(int, char**) {
+  test<char>();
+#ifndef TEST_HAS_NO_WIDE_CHARACTERS
+  test<wchar_t>();
+#endif
+
+  return 0;
+}
diff --git a/libcxx/test/std/input.output/stream.buffers/streambuf/streambuf.protected/streambuf.put.area/setp.assert.pass.cpp b/libcxx/test/std/input.output/stream.buffers/streambuf/streambuf.protected/streambuf.put.area/setp.assert.pass.cpp
new file mode 100644
index 000000000000..abd42272de50
--- /dev/null
+++ b/libcxx/test/std/input.output/stream.buffers/streambuf/streambuf.protected/streambuf.put.area/setp.assert.pass.cpp
@@ -0,0 +1,57 @@
+//===----------------------------------------------------------------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+// REQUIRES: has-unix-headers
+// UNSUPPORTED: libcpp-hardening-mode=none
+// XFAIL: libcpp-hardening-mode=debug && availability-verbose_abort-missing
+
+// <streambuf>
+
+// template <class charT, class traits = char_traits<charT> >
+// class basic_streambuf;
+
+// void setp(char_type* pbeg, char_type* pend);
+
+#include <algorithm>
+#include <iterator>
+#include <streambuf>
+#include <string>
+
+#include "check_assertion.h"
+#include "make_string.h"
+#include "test_macros.h"
+
+template <class CharT>
+struct streambuf : public std::basic_streambuf<CharT> {
+  typedef std::basic_streambuf<CharT> base;
+
+  streambuf() {}
+
+  void setp(CharT* pbeg, CharT* pend) { base::setp(pbeg, pend); }
+};
+
+template <class CharT>
+void test() {
+  std::basic_string<CharT> str = MAKE_STRING(CharT, "ABCDEF");
+  CharT arr[6];
+  std::copy(str.begin(), str.end(), arr);
+
+  {
+    streambuf<CharT> buff;
+    TEST_LIBCPP_ASSERT_FAILURE(buff.setp(std::begin(arr) + 3, std::begin(arr)), "[pbeg, pend) must be a valid range");
+  }
+}
+
+int main(int, char**) {
+  test<char>();
+#ifndef TEST_HAS_NO_WIDE_CHARACTERS
+  test<wchar_t>();
+#endif
+
+  return 0;
+}
diff --git a/libcxx/test/std/language.support/support.dynamic/hardware_inference_size.compile.pass.cpp b/libcxx/test/std/language.support/support.dynamic/hardware_inference_size.compile.pass.cpp
index ae277d53e46f..2656f0595bf5 100644
--- a/libcxx/test/std/language.support/support.dynamic/hardware_inference_size.compile.pass.cpp
+++ b/libcxx/test/std/language.support/support.dynamic/hardware_inference_size.compile.pass.cpp
@@ -7,7 +7,7 @@
 //===----------------------------------------------------------------------===//
 
 // UNSUPPORTED: c++03, c++11, c++14
-// XFAIL: (clang || apple-clang) && stdlib=libc++
+// UNSUPPORTED: (clang || apple-clang) && stdlib=libc++
 
 #include <new>
 
diff --git a/libcxx/test/std/ranges/range.adaptors/range.adaptor.object/range_adaptor_closure.pass.cpp b/libcxx/test/std/ranges/range.adaptors/range.adaptor.object/range_adaptor_closure.pass.cpp
new file mode 100644
index 000000000000..9d1eb1243458
--- /dev/null
+++ b/libcxx/test/std/ranges/range.adaptors/range.adaptor.object/range_adaptor_closure.pass.cpp
@@ -0,0 +1,142 @@
+//===----------------------------------------------------------------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+// UNSUPPORTED: c++03, c++11, c++14, c++17, c++20
+
+// std::ranges::range_adaptor_closure;
+
+#include <ranges>
+
+#include <algorithm>
+#include <vector>
+
+#include "test_range.h"
+
+template <class T>
+concept CanDeriveFromRangeAdaptorClosure = requires { typename std::ranges::range_adaptor_closure<T>; };
+static_assert(!CanDeriveFromRangeAdaptorClosure<int>);
+
+struct Foo {};
+static_assert(CanDeriveFromRangeAdaptorClosure<Foo>);
+static_assert(!CanDeriveFromRangeAdaptorClosure<Foo&>);
+static_assert(!CanDeriveFromRangeAdaptorClosure<const Foo>);
+static_assert(!CanDeriveFromRangeAdaptorClosure<volatile Foo>);
+static_assert(!CanDeriveFromRangeAdaptorClosure<const volatile Foo&&>);
+
+struct incomplete_t;
+static_assert(CanDeriveFromRangeAdaptorClosure<incomplete_t>);
+
+using range_t = std::vector<int>;
+
+template <class T>
+concept RangeAdaptorClosure =
+    CanBePiped<range_t, T&> && CanBePiped<range_t, const T&> && CanBePiped<range_t, T&&> &&
+    CanBePiped<range_t, const T&&>;
+
+struct callable : std::ranges::range_adaptor_closure<callable> {
+  static void operator()(const range_t&) {}
+};
+static_assert(RangeAdaptorClosure<callable>);
+
+// `not_callable_1` doesn't have an `operator()`
+struct not_callable_1 : std::ranges::range_adaptor_closure<not_callable_1> {};
+static_assert(!RangeAdaptorClosure<not_callable_1>);
+
+// `not_callable_2` doesn't have an `operator()` that accepts a `range` argument
+struct not_callable_2 : std::ranges::range_adaptor_closure<not_callable_2> {
+  static void operator()() {}
+};
+static_assert(!RangeAdaptorClosure<not_callable_2>);
+
+// `not_derived_from_1` doesn't derive from `std::ranges::range_adaptor_closure`
+struct not_derived_from_1 {
+  static void operator()(const range_t&) {}
+};
+static_assert(!RangeAdaptorClosure<not_derived_from_1>);
+
+// `not_derived_from_2` doesn't publicly derive from `std::ranges::range_adaptor_closure`
+struct not_derived_from_2 : private std::ranges::range_adaptor_closure<not_derived_from_2> {
+  static void operator()(const range_t&) {}
+};
+static_assert(!RangeAdaptorClosure<not_derived_from_2>);
+
+// `not_derived_from_3` doesn't derive from the correct specialization of `std::ranges::range_adaptor_closure`
+struct not_derived_from_3 : std::ranges::range_adaptor_closure<callable> {
+  static void operator()(const range_t&) {}
+};
+static_assert(!RangeAdaptorClosure<not_derived_from_3>);
+
+// `not_derived_from_4` doesn't derive from exactly one specialization of `std::ranges::range_adaptor_closure`
+struct not_derived_from_4
+    : std::ranges::range_adaptor_closure<not_derived_from_4>,
+      std::ranges::range_adaptor_closure<callable> {
+  static void operator()(const range_t&) {}
+};
+static_assert(!RangeAdaptorClosure<not_derived_from_4>);
+
+// `is_range` models `range`
+struct is_range : std::ranges::range_adaptor_closure<is_range> {
+  static void operator()(const range_t&) {}
+  int* begin() const { return nullptr; }
+  int* end() const { return nullptr; }
+};
+static_assert(std::ranges::range<is_range> && std::ranges::range<const is_range>);
+static_assert(!RangeAdaptorClosure<is_range>);
+
+// user-defined range adaptor closure object
+struct negate_fn : std::ranges::range_adaptor_closure<negate_fn> {
+  template <std::ranges::range Range>
+  static constexpr decltype(auto) operator()(Range&& range) {
+    return std::forward<Range>(range) | std::views::transform([](auto element) { return -element; });
+  }
+};
+static_assert(RangeAdaptorClosure<negate_fn>);
+constexpr auto negate = negate_fn{};
+
+// user-defined range adaptor closure object
+struct plus_1_fn : std::ranges::range_adaptor_closure<plus_1_fn> {
+  template <std::ranges::range Range>
+  static constexpr decltype(auto) operator()(Range&& range) {
+    return std::forward<Range>(range) | std::views::transform([](auto element) { return element + 1; });
+  }
+};
+static_assert(RangeAdaptorClosure<plus_1_fn>);
+constexpr auto plus_1 = plus_1_fn{};
+
+constexpr bool test() {
+  const std::vector<int> n{1, 2, 3, 4, 5};
+  const std::vector<int> n_negate{-1, -2, -3, -4, -5};
+
+  assert(std::ranges::equal(n | negate, n_negate));
+  assert(std::ranges::equal(negate(n), n_negate));
+
+  assert(std::ranges::equal(n | negate | negate, n));
+  assert(std::ranges::equal(n | (negate | negate), n));
+  assert(std::ranges::equal((n | negate) | negate, n));
+  assert(std::ranges::equal(negate(n) | negate, n));
+  assert(std::ranges::equal(negate(n | negate), n));
+  assert(std::ranges::equal((negate | negate)(n), n));
+  assert(std::ranges::equal(negate(negate(n)), n));
+
+  const std::vector<int> n_plus_1_negate{-2, -3, -4, -5, -6};
+  assert(std::ranges::equal(n | plus_1 | negate, n_plus_1_negate));
+  assert(std::ranges::equal(
+      n | plus_1 | std::views::transform([](auto element) { return element; }) | negate, n_plus_1_negate));
+
+  const std::vector<int> n_negate_plus_1{0, -1, -2, -3, -4};
+  assert(std::ranges::equal(n | negate | plus_1, n_negate_plus_1));
+  assert(std::ranges::equal(n | std::views::reverse | negate | plus_1 | std::views::reverse, n_negate_plus_1));
+  return true;
+}
+
+int main(int, char**) {
+  test();
+  static_assert(test());
+
+  return 0;
+}
diff --git a/libcxx/test/std/ranges/range.utility/range.utility.conv/to.pass.cpp b/libcxx/test/std/ranges/range.utility/range.utility.conv/to.pass.cpp
index 3df88d6a2dcc..7f816bb21a19 100644
--- a/libcxx/test/std/ranges/range.utility/range.utility.conv/to.pass.cpp
+++ b/libcxx/test/std/ranges/range.utility/range.utility.conv/to.pass.cpp
@@ -560,6 +560,11 @@ constexpr void test_recursive() {
   }
 
   assert((in | std::ranges::to<C4>()) == result);
+
+  // LWG3984: ranges::to's recursion branch may be ill-formed
+  auto in_owning_view = std::views::all(std::move(in));
+  static_assert(!std::ranges::viewable_range<decltype((in_owning_view))>);
+  assert(std::ranges::to<C4>(in_owning_view) == result);
 }
 
 constexpr bool test() {
diff --git a/libcxx/test/std/time/time.clock/time.clock.file/ostream.pass.cpp b/libcxx/test/std/time/time.clock/time.clock.file/ostream.pass.cpp
index 18a4506b9156..11eab1dddfe3 100644
--- a/libcxx/test/std/time/time.clock/time.clock.file/ostream.pass.cpp
+++ b/libcxx/test/std/time/time.clock/time.clock.file/ostream.pass.cpp
@@ -71,8 +71,45 @@ template <class CharT>
 static void test_c() {
   using namespace std::literals::chrono_literals;
 
+  assert(stream_c_locale<CharT>(std::chrono::file_time<std::chrono::nanoseconds>{-946'688'523'123'456'789ns}) ==
+         SV("1940-01-01 22:57:56.876543211"));
+
+  assert(stream_c_locale<CharT>(std::chrono::file_time<std::chrono::microseconds>{-946'688'523'123'456us}) ==
+         SV("1940-01-01 22:57:56.876544"));
+
+  assert(stream_c_locale<CharT>(std::chrono::file_time<std::chrono::milliseconds>{-946'688'523'123ms}) ==
+         SV("1940-01-01 22:57:56.877"));
+
+  assert(stream_c_locale<CharT>(std::chrono::file_time<std::chrono::nanoseconds>{-1000000000ns}) ==
+         SV("1969-12-31 23:59:59.000000000"));
+
+  assert(stream_c_locale<CharT>(std::chrono::file_time<std::chrono::microseconds>{-1000000us}) ==
+         SV("1969-12-31 23:59:59.000000"));
+
+  assert(stream_c_locale<CharT>(std::chrono::file_time<std::chrono::milliseconds>{-1000ms}) ==
+         SV("1969-12-31 23:59:59.000"));
+
+  assert(stream_c_locale<CharT>(std::chrono::file_time<std::chrono::nanoseconds>{-1ns}) ==
+         SV("1969-12-31 23:59:59.999999999"));
+
+  assert(stream_c_locale<CharT>(std::chrono::file_time<std::chrono::nanoseconds>{0ns}) ==
+         SV("1970-01-01 00:00:00.000000000"));
+
+  assert(stream_c_locale<CharT>(std::chrono::file_time<std::chrono::nanoseconds>{1ns}) ==
+         SV("1970-01-01 00:00:00.000000001"));
+
+  assert(stream_c_locale<CharT>(std::chrono::file_time<std::chrono::nanoseconds>{1000000000ns}) ==
+         SV("1970-01-01 00:00:01.000000000"));
+
+  assert(stream_c_locale<CharT>(std::chrono::file_time<std::chrono::microseconds>{1000000us}) ==
+         SV("1970-01-01 00:00:01.000000"));
+
+  assert(stream_c_locale<CharT>(std::chrono::file_time<std::chrono::milliseconds>{1000ms}) ==
+         SV("1970-01-01 00:00:01.000"));
+
   assert(stream_c_locale<CharT>(file_time<std::chrono::nanoseconds>{946'688'523'123'456'789ns}) ==
          SV("2000-01-01 01:02:03.123456789"));
+
   assert(stream_c_locale<CharT>(file_time<std::chrono::microseconds>{946'688'523'123'456us}) ==
          SV("2000-01-01 01:02:03.123456"));
 
@@ -107,6 +144,42 @@ template <class CharT>
 static void test_fr_FR() {
   using namespace std::literals::chrono_literals;
 
+  assert(stream_fr_FR_locale<CharT>(std::chrono::file_time<std::chrono::nanoseconds>{-946'688'523'123'456'789ns}) ==
+         SV("1940-01-01 22:57:56,876543211"));
+
+  assert(stream_fr_FR_locale<CharT>(std::chrono::file_time<std::chrono::microseconds>{-946'688'523'123'456us}) ==
+         SV("1940-01-01 22:57:56,876544"));
+
+  assert(stream_fr_FR_locale<CharT>(std::chrono::file_time<std::chrono::milliseconds>{-946'688'523'123ms}) ==
+         SV("1940-01-01 22:57:56,877"));
+
+  assert(stream_fr_FR_locale<CharT>(std::chrono::file_time<std::chrono::nanoseconds>{-1000000000ns}) ==
+         SV("1969-12-31 23:59:59,000000000"));
+
+  assert(stream_fr_FR_locale<CharT>(std::chrono::file_time<std::chrono::microseconds>{-1000000us}) ==
+         SV("1969-12-31 23:59:59,000000"));
+
+  assert(stream_fr_FR_locale<CharT>(std::chrono::file_time<std::chrono::milliseconds>{-1000ms}) ==
+         SV("1969-12-31 23:59:59,000"));
+
+  assert(stream_fr_FR_locale<CharT>(std::chrono::file_time<std::chrono::nanoseconds>{-1ns}) ==
+         SV("1969-12-31 23:59:59,999999999"));
+
+  assert(stream_fr_FR_locale<CharT>(std::chrono::file_time<std::chrono::nanoseconds>{0ns}) ==
+         SV("1970-01-01 00:00:00,000000000"));
+
+  assert(stream_fr_FR_locale<CharT>(std::chrono::file_time<std::chrono::nanoseconds>{1ns}) ==
+         SV("1970-01-01 00:00:00,000000001"));
+
+  assert(stream_fr_FR_locale<CharT>(std::chrono::file_time<std::chrono::nanoseconds>{1000000000ns}) ==
+         SV("1970-01-01 00:00:01,000000000"));
+
+  assert(stream_fr_FR_locale<CharT>(std::chrono::file_time<std::chrono::microseconds>{1000000us}) ==
+         SV("1970-01-01 00:00:01,000000"));
+
+  assert(stream_fr_FR_locale<CharT>(std::chrono::file_time<std::chrono::milliseconds>{1000ms}) ==
+         SV("1970-01-01 00:00:01,000"));
+
   assert(stream_fr_FR_locale<CharT>(file_time<std::chrono::nanoseconds>{946'688'523'123'456'789ns}) ==
          SV("2000-01-01 01:02:03,123456789"));
   assert(stream_fr_FR_locale<CharT>(file_time<std::chrono::microseconds>{946'688'523'123'456us}) ==
@@ -144,6 +217,42 @@ template <class CharT>
 static void test_ja_JP() {
   using namespace std::literals::chrono_literals;
 
+  assert(stream_ja_JP_locale<CharT>(std::chrono::file_time<std::chrono::nanoseconds>{-946'688'523'123'456'789ns}) ==
+         SV("1940-01-01 22:57:56.876543211"));
+
+  assert(stream_ja_JP_locale<CharT>(std::chrono::file_time<std::chrono::microseconds>{-946'688'523'123'456us}) ==
+         SV("1940-01-01 22:57:56.876544"));
+
+  assert(stream_ja_JP_locale<CharT>(std::chrono::file_time<std::chrono::milliseconds>{-946'688'523'123ms}) ==
+         SV("1940-01-01 22:57:56.877"));
+
+  assert(stream_ja_JP_locale<CharT>(std::chrono::file_time<std::chrono::nanoseconds>{-1000000000ns}) ==
+         SV("1969-12-31 23:59:59.000000000"));
+
+  assert(stream_ja_JP_locale<CharT>(std::chrono::file_time<std::chrono::microseconds>{-1000000us}) ==
+         SV("1969-12-31 23:59:59.000000"));
+
+  assert(stream_ja_JP_locale<CharT>(std::chrono::file_time<std::chrono::milliseconds>{-1000ms}) ==
+         SV("1969-12-31 23:59:59.000"));
+
+  assert(stream_ja_JP_locale<CharT>(std::chrono::file_time<std::chrono::nanoseconds>{-1ns}) ==
+         SV("1969-12-31 23:59:59.999999999"));
+
+  assert(stream_ja_JP_locale<CharT>(std::chrono::file_time<std::chrono::nanoseconds>{0ns}) ==
+         SV("1970-01-01 00:00:00.000000000"));
+
+  assert(stream_ja_JP_locale<CharT>(std::chrono::file_time<std::chrono::nanoseconds>{1ns}) ==
+         SV("1970-01-01 00:00:00.000000001"));
+
+  assert(stream_ja_JP_locale<CharT>(std::chrono::file_time<std::chrono::nanoseconds>{1000000000ns}) ==
+         SV("1970-01-01 00:00:01.000000000"));
+
+  assert(stream_ja_JP_locale<CharT>(std::chrono::file_time<std::chrono::microseconds>{1000000us}) ==
+         SV("1970-01-01 00:00:01.000000"));
+
+  assert(stream_ja_JP_locale<CharT>(std::chrono::file_time<std::chrono::milliseconds>{1000ms}) ==
+         SV("1970-01-01 00:00:01.000"));
+
   assert(stream_ja_JP_locale<CharT>(file_time<std::chrono::nanoseconds>{946'688'523'123'456'789ns}) ==
          SV("2000-01-01 01:02:03.123456789"));
   assert(stream_ja_JP_locale<CharT>(file_time<std::chrono::microseconds>{946'688'523'123'456us}) ==
diff --git a/libcxx/test/std/time/time.clock/time.clock.local/ostream.pass.cpp b/libcxx/test/std/time/time.clock/time.clock.local/ostream.pass.cpp
index 9fdef8d5adc7..6ec63a14fbbd 100644
--- a/libcxx/test/std/time/time.clock/time.clock.local/ostream.pass.cpp
+++ b/libcxx/test/std/time/time.clock/time.clock.local/ostream.pass.cpp
@@ -64,6 +64,24 @@ template <class CharT>
 static void test_c() {
   using namespace std::literals::chrono_literals;
 
+  assert(stream_c_locale<CharT>(std::chrono::local_time<std::chrono::nanoseconds>{-946'688'523'123'456'789ns}) ==
+         SV("1940-01-01 22:57:56.876543211"));
+
+  assert(stream_c_locale<CharT>(std::chrono::local_time<std::chrono::microseconds>{-946'688'523'123'456us}) ==
+         SV("1940-01-01 22:57:56.876544"));
+
+  assert(stream_c_locale<CharT>(std::chrono::local_time<std::chrono::milliseconds>{-946'688'523'123ms}) ==
+         SV("1940-01-01 22:57:56.877"));
+
+  assert(stream_c_locale<CharT>(std::chrono::local_time<std::chrono::nanoseconds>{-1ns}) ==
+         SV("1969-12-31 23:59:59.999999999"));
+
+  assert(stream_c_locale<CharT>(std::chrono::local_time<std::chrono::nanoseconds>{0ns}) ==
+         SV("1970-01-01 00:00:00.000000000"));
+
+  assert(stream_c_locale<CharT>(std::chrono::local_time<std::chrono::nanoseconds>{1ns}) ==
+         SV("1970-01-01 00:00:00.000000001"));
+
   assert(stream_c_locale<CharT>(std::chrono::local_time<std::chrono::nanoseconds>{946'688'523'123'456'789ns}) ==
          SV("2000-01-01 01:02:03.123456789"));
   assert(stream_c_locale<CharT>(std::chrono::local_time<std::chrono::microseconds>{946'688'523'123'456us}) ==
@@ -97,6 +115,24 @@ template <class CharT>
 static void test_fr_FR() {
   using namespace std::literals::chrono_literals;
 
+  assert(stream_fr_FR_locale<CharT>(std::chrono::local_time<std::chrono::nanoseconds>{-946'688'523'123'456'789ns}) ==
+         SV("1940-01-01 22:57:56,876543211"));
+
+  assert(stream_fr_FR_locale<CharT>(std::chrono::local_time<std::chrono::microseconds>{-946'688'523'123'456us}) ==
+         SV("1940-01-01 22:57:56,876544"));
+
+  assert(stream_fr_FR_locale<CharT>(std::chrono::local_time<std::chrono::milliseconds>{-946'688'523'123ms}) ==
+         SV("1940-01-01 22:57:56,877"));
+
+  assert(stream_fr_FR_locale<CharT>(std::chrono::local_time<std::chrono::nanoseconds>{-1ns}) ==
+         SV("1969-12-31 23:59:59,999999999"));
+
+  assert(stream_fr_FR_locale<CharT>(std::chrono::local_time<std::chrono::nanoseconds>{0ns}) ==
+         SV("1970-01-01 00:00:00,000000000"));
+
+  assert(stream_fr_FR_locale<CharT>(std::chrono::local_time<std::chrono::nanoseconds>{1ns}) ==
+         SV("1970-01-01 00:00:00,000000001"));
+
   assert(stream_fr_FR_locale<CharT>(std::chrono::local_time<std::chrono::nanoseconds>{946'688'523'123'456'789ns}) ==
          SV("2000-01-01 01:02:03,123456789"));
   assert(stream_fr_FR_locale<CharT>(std::chrono::local_time<std::chrono::microseconds>{946'688'523'123'456us}) ==
@@ -131,6 +167,24 @@ template <class CharT>
 static void test_ja_JP() {
   using namespace std::literals::chrono_literals;
 
+  assert(stream_ja_JP_locale<CharT>(std::chrono::local_time<std::chrono::nanoseconds>{-946'688'523'123'456'789ns}) ==
+         SV("1940-01-01 22:57:56.876543211"));
+
+  assert(stream_ja_JP_locale<CharT>(std::chrono::local_time<std::chrono::microseconds>{-946'688'523'123'456us}) ==
+         SV("1940-01-01 22:57:56.876544"));
+
+  assert(stream_ja_JP_locale<CharT>(std::chrono::local_time<std::chrono::milliseconds>{-946'688'523'123ms}) ==
+         SV("1940-01-01 22:57:56.877"));
+
+  assert(stream_ja_JP_locale<CharT>(std::chrono::local_time<std::chrono::nanoseconds>{-1ns}) ==
+         SV("1969-12-31 23:59:59.999999999"));
+
+  assert(stream_ja_JP_locale<CharT>(std::chrono::local_time<std::chrono::nanoseconds>{0ns}) ==
+         SV("1970-01-01 00:00:00.000000000"));
+
+  assert(stream_ja_JP_locale<CharT>(std::chrono::local_time<std::chrono::nanoseconds>{1ns}) ==
+         SV("1970-01-01 00:00:00.000000001"));
+
   assert(stream_ja_JP_locale<CharT>(std::chrono::local_time<std::chrono::nanoseconds>{946'688'523'123'456'789ns}) ==
          SV("2000-01-01 01:02:03.123456789"));
   assert(stream_ja_JP_locale<CharT>(std::chrono::local_time<std::chrono::microseconds>{946'688'523'123'456us}) ==
diff --git a/libcxx/test/std/time/time.clock/time.clock.system/sys_time.ostream.pass.cpp b/libcxx/test/std/time/time.clock/time.clock.system/sys_time.ostream.pass.cpp
index 78d8da57c150..e596ddefde51 100644
--- a/libcxx/test/std/time/time.clock/time.clock.system/sys_time.ostream.pass.cpp
+++ b/libcxx/test/std/time/time.clock/time.clock.system/sys_time.ostream.pass.cpp
@@ -64,6 +64,24 @@ template <class CharT>
 static void test_c() {
   using namespace std::literals::chrono_literals;
 
+  assert(stream_c_locale<CharT>(std::chrono::sys_time<std::chrono::nanoseconds>{-946'688'523'123'456'789ns}) ==
+         SV("1940-01-01 22:57:56.876543211"));
+
+  assert(stream_c_locale<CharT>(std::chrono::sys_time<std::chrono::microseconds>{-946'688'523'123'456us}) ==
+         SV("1940-01-01 22:57:56.876544"));
+
+  assert(stream_c_locale<CharT>(std::chrono::sys_time<std::chrono::milliseconds>{-946'688'523'123ms}) ==
+         SV("1940-01-01 22:57:56.877"));
+
+  assert(stream_c_locale<CharT>(std::chrono::sys_time<std::chrono::nanoseconds>{-1ns}) ==
+         SV("1969-12-31 23:59:59.999999999"));
+
+  assert(stream_c_locale<CharT>(std::chrono::sys_time<std::chrono::nanoseconds>{0ns}) ==
+         SV("1970-01-01 00:00:00.000000000"));
+
+  assert(stream_c_locale<CharT>(std::chrono::sys_time<std::chrono::nanoseconds>{1ns}) ==
+         SV("1970-01-01 00:00:00.000000001"));
+
   assert(stream_c_locale<CharT>(std::chrono::sys_time<std::chrono::nanoseconds>{946'688'523'123'456'789ns}) ==
          SV("2000-01-01 01:02:03.123456789"));
   assert(stream_c_locale<CharT>(std::chrono::sys_time<std::chrono::microseconds>{946'688'523'123'456us}) ==
@@ -92,6 +110,24 @@ template <class CharT>
 static void test_fr_FR() {
   using namespace std::literals::chrono_literals;
 
+  assert(stream_fr_FR_locale<CharT>(std::chrono::sys_time<std::chrono::nanoseconds>{-946'688'523'123'456'789ns}) ==
+         SV("1940-01-01 22:57:56,876543211"));
+
+  assert(stream_fr_FR_locale<CharT>(std::chrono::sys_time<std::chrono::microseconds>{-946'688'523'123'456us}) ==
+         SV("1940-01-01 22:57:56,876544"));
+
+  assert(stream_fr_FR_locale<CharT>(std::chrono::sys_time<std::chrono::milliseconds>{-946'688'523'123ms}) ==
+         SV("1940-01-01 22:57:56,877"));
+
+  assert(stream_fr_FR_locale<CharT>(std::chrono::sys_time<std::chrono::nanoseconds>{-1ns}) ==
+         SV("1969-12-31 23:59:59,999999999"));
+
+  assert(stream_fr_FR_locale<CharT>(std::chrono::sys_time<std::chrono::nanoseconds>{0ns}) ==
+         SV("1970-01-01 00:00:00,000000000"));
+
+  assert(stream_fr_FR_locale<CharT>(std::chrono::sys_time<std::chrono::nanoseconds>{1ns}) ==
+         SV("1970-01-01 00:00:00,000000001"));
+
   assert(stream_fr_FR_locale<CharT>(std::chrono::sys_time<std::chrono::nanoseconds>{946'688'523'123'456'789ns}) ==
          SV("2000-01-01 01:02:03,123456789"));
   assert(stream_fr_FR_locale<CharT>(std::chrono::sys_time<std::chrono::microseconds>{946'688'523'123'456us}) ==
@@ -120,6 +156,24 @@ template <class CharT>
 static void test_ja_JP() {
   using namespace std::literals::chrono_literals;
 
+  assert(stream_ja_JP_locale<CharT>(std::chrono::sys_time<std::chrono::nanoseconds>{-946'688'523'123'456'789ns}) ==
+         SV("1940-01-01 22:57:56.876543211"));
+
+  assert(stream_ja_JP_locale<CharT>(std::chrono::sys_time<std::chrono::microseconds>{-946'688'523'123'456us}) ==
+         SV("1940-01-01 22:57:56.876544"));
+
+  assert(stream_ja_JP_locale<CharT>(std::chrono::sys_time<std::chrono::milliseconds>{-946'688'523'123ms}) ==
+         SV("1940-01-01 22:57:56.877"));
+
+  assert(stream_ja_JP_locale<CharT>(std::chrono::sys_time<std::chrono::nanoseconds>{-1ns}) ==
+         SV("1969-12-31 23:59:59.999999999"));
+
+  assert(stream_ja_JP_locale<CharT>(std::chrono::sys_time<std::chrono::nanoseconds>{0ns}) ==
+         SV("1970-01-01 00:00:00.000000000"));
+
+  assert(stream_ja_JP_locale<CharT>(std::chrono::sys_time<std::chrono::nanoseconds>{1ns}) ==
+         SV("1970-01-01 00:00:00.000000001"));
+
   assert(stream_ja_JP_locale<CharT>(std::chrono::sys_time<std::chrono::nanoseconds>{946'688'523'123'456'789ns}) ==
          SV("2000-01-01 01:02:03.123456789"));
   assert(stream_ja_JP_locale<CharT>(std::chrono::sys_time<std::chrono::microseconds>{946'688'523'123'456us}) ==
diff --git a/libcxx/test/std/time/time.zone/time.zone.timezone/time.zone.members/get_info.sys_time.pass.cpp b/libcxx/test/std/time/time.zone/time.zone.timezone/time.zone.members/get_info.sys_time.pass.cpp
index 25d2ff11d093..1a1705d5ae59 100644
--- a/libcxx/test/std/time/time.zone/time.zone.timezone/time.zone.members/get_info.sys_time.pass.cpp
+++ b/libcxx/test/std/time/time.zone/time.zone.timezone/time.zone.members/get_info.sys_time.pass.cpp
@@ -1299,6 +1299,78 @@ static void test_america_indiana_knox() {
                tz->get_info(to_sys_seconds(2006y, std::chrono::October, 29d, 6h, 59min, 59s)));
 }
 
+static void test_america_punta_arenas() {
+  // Z America/Punta_Arenas -4:43:40 - LMT 1890
+  // ...
+  // -4 - -04 1919 Jul
+  // -4:42:45 - SMT 1927 S
+  // -5 x -05/-04 1932 S
+  // ...
+  //
+  // R x 1927 1931 - S 1 0 1 -
+  // R x 1928 1932 - Ap 1 0 0 -
+  // ...
+
+  using namespace std::literals::chrono_literals;
+  const std::chrono::time_zone* tz = std::chrono::locate_zone("America/Punta_Arenas");
+
+  assert_equal(
+      std::chrono::sys_info(
+          to_sys_seconds(1927y, std::chrono::September, 1d, 4h, 42min, 45s),
+          to_sys_seconds(1928y, std::chrono::April, 1d, 4h),
+          -4h,
+          60min,
+          "-04"),
+      tz->get_info(to_sys_seconds(1927y, std::chrono::September, 1d, 4h, 42min, 45s)));
+
+  assert_equal(
+      std::chrono::sys_info(
+          to_sys_seconds(1927y, std::chrono::September, 1d, 4h, 42min, 45s),
+          to_sys_seconds(1928y, std::chrono::April, 1d, 4h),
+          -4h,
+          60min,
+          "-04"),
+      tz->get_info(to_sys_seconds(1928y, std::chrono::April, 1d, 3h, 59min, 59s)));
+}
+
+static void test_europ_ljubljana() {
+  // Z Europe/Ljubljana 0:58:4 - LMT 1884
+  // 1 - CET 1941 Ap 18 23
+  // 1 c CE%sT 1945 May 8 2s
+  // 1 1 CEST 1945 S 16 2s
+  // 1 - CET 1982 N 27
+  // 1 E CE%sT
+  //
+  // ...
+  // R c 1943 o - O 4 2s 0 -
+  // R c 1944 1945 - Ap M>=1 2s 1 S
+  // R c 1944 o - O 2 2s 0 -
+  // R c 1945 o - S 16 2s 0 -
+  // R c 1977 1980 - Ap Su>=1 2s 1 S
+  // ...
+
+  using namespace std::literals::chrono_literals;
+  const std::chrono::time_zone* tz = std::chrono::locate_zone("Europe/Ljubljana");
+
+  assert_equal(
+      std::chrono::sys_info(
+          to_sys_seconds(1945y, std::chrono::April, 2d, 1h),
+          to_sys_seconds(1945y, std::chrono::September, 16d, 1h),
+          2h,
+          60min,
+          "CEST"),
+      tz->get_info(to_sys_seconds(1945y, std::chrono::April, 2d, 1h)));
+
+  assert_equal(
+      std::chrono::sys_info(
+          to_sys_seconds(1945y, std::chrono::April, 2d, 1h),
+          to_sys_seconds(1945y, std::chrono::September, 16d, 1h),
+          2h,
+          60min,
+          "CEST"),
+      tz->get_info(to_sys_seconds(1945y, std::chrono::September, 16d, 0h, 59min, 59s)));
+}
+
 int main(int, const char**) {
   // Basic tests
   test_gmt();
@@ -1333,5 +1405,9 @@ int main(int, const char**) {
   test_america_ciudad_juarez();
   test_america_indiana_knox();
 
+  // Reverse search bugs
+  test_america_punta_arenas();
+  test_europ_ljubljana();
+
   return 0;
 }
diff --git a/libcxx/test/std/utilities/format/format.functions/escaped_output.unicode.pass.cpp b/libcxx/test/std/utilities/format/format.functions/escaped_output.unicode.pass.cpp
index bf5c0a51f944..96c1e2664f7a 100644
--- a/libcxx/test/std/utilities/format/format.functions/escaped_output.unicode.pass.cpp
+++ b/libcxx/test/std/utilities/format/format.functions/escaped_output.unicode.pass.cpp
@@ -223,7 +223,7 @@ void test_char() {
     static_assert(sizeof(CharT) == 4, "add support for unexpected size");
     // Unicode fitting in a 32-bit wchar_t
 
-    constexpr wchar_t x  = 0x1ffff;
+    constexpr wchar_t x       = 0x1ffff;
     constexpr std::uint32_t y = 0x1ffff;
     static_assert(x == y);
 
@@ -290,7 +290,7 @@ void test_string() {
     test_format(SV("[\"\ud7ff\"]"), SV("[{:?}]"), "\xed\x9f\xbf");              // U+D7FF last valid
 #else
     /* U+D800..D+DFFFF surrogate range */
-    test_format(SV(R"(["\u{d7ff}"])"), SV("[{:?}]"), "\xed\x9f\xbf");           // U+D7FF last valid
+    test_format(SV(R"(["\u{d7ff}"])"), SV("[{:?}]"), "\xed\x9f\xbf"); // U+D7FF last valid
 #endif
     test_format(SV(R"(["\x{ed}\x{a0}\x{80}"])"), SV("[{:?}]"), "\xed\xa0\x80"); // U+D800
     test_format(SV(R"(["\x{ed}\x{af}\x{bf}"])"), SV("[{:?}]"), "\xed\xaf\xbf"); // U+DBFF
@@ -319,7 +319,8 @@ void test_string() {
     test_format(SV("[\"\u00c3(\"]"), SV("[{:?}]"), L"\xc3\x28");
   }
 
-  test_format(SV(R"(["🤷🏻\u{200d}♂\u{fe0f}"])"), SV("[{:?}]"), SV("🤷🏻‍♂️"));
+  // LWG-3965
+  test_format(SV(R"(["🤷🏻\u{200d}♂️"])"), SV("[{:?}]"), SV("🤷🏻‍♂️"));
 
   // *** Special cases ***
   test_format(SV(R"("\t\n\r\\'\" ")"), SV("{:?}"), SV("\t\n\r\\'\" "));
@@ -336,6 +337,11 @@ void test_string() {
   if constexpr (sizeof(CharT) == 1)
     test_format(SV(R"("\x{80}")"), SV("{:?}"), SV("\x80"));
 
+  // *** P2713R1 examples ***
+  test_format(SV(R"(["\u{301}"])"), SV("[{:?}]"), SV("\u0301"));
+  test_format(SV(R"(["\\\u{301}"])"), SV("[{:?}]"), SV("\\\u0301"));
+  test_format(SV(R"(["ẹ́"])"), SV("[{:?}]"), SV("e\u0301\u0323"));
+
 #ifndef TEST_HAS_NO_WIDE_CHARACTERS
   if constexpr (sizeof(CharT) > 1) {
     using V = std::basic_string_view<CharT>;
@@ -373,7 +379,7 @@ void test_string() {
     static_assert(sizeof(CharT) == 4, "add support for unexpected size");
     // Unicode fitting in a 32-bit wchar_t
 
-    constexpr wchar_t x  = 0x1ffff;
+    constexpr wchar_t x       = 0x1ffff;
     constexpr std::uint32_t y = 0x1ffff;
     static_assert(x == y);
 
@@ -406,20 +412,18 @@ void test_format_functions(TestFunction check) {
   check(SV(R"(*"hellö"**)"), SV("{:*^10?}"), SV("hellö"));
   check(SV(R"("hellö"***)"), SV("{:*<10?}"), SV("hellö"));
 
-  check(SV(R"("hello\u{308}")"), SV("{:*>10?}"), SV("hello\u0308"));
-  check(SV(R"(***"hello\u{308}")"), SV("{:*>17?}"), SV("hello\u0308"));
-  check(SV(R"(*"hello\u{308}"**)"), SV("{:*^17?}"), SV("hello\u0308"));
-  check(SV(R"("hello\u{308}"***)"), SV("{:*<17?}"), SV("hello\u0308"));
+  check(SV(R"(***"hellö")"), SV("{:*>10?}"), SV("hello\u0308"));
+  check(SV(R"(*"hellö"**)"), SV("{:*^10?}"), SV("hello\u0308"));
+  check(SV(R"("hellö"***)"), SV("{:*<10?}"), SV("hello\u0308"));
 
-  check(SV(R"("hello 🤷🏻\u{200d}♂\u{fe0f}")"), SV("{:*>10?}"), SV("hello 🤷🏻‍♂️"));
-  check(SV(R"(***"hello 🤷🏻\u{200d}♂\u{fe0f}")"), SV("{:*>30?}"), SV("hello 🤷🏻‍♂️"));
-  check(SV(R"(*"hello 🤷🏻\u{200d}♂\u{fe0f}"**)"), SV("{:*^30?}"), SV("hello 🤷🏻‍♂️"));
-  check(SV(R"("hello 🤷🏻\u{200d}♂\u{fe0f}"***)"), SV("{:*<30?}"), SV("hello 🤷🏻‍♂️"));
+  check(SV(R"(***"hello 🤷🏻\u{200d}♂️")"), SV("{:*>22?}"), SV("hello 🤷🏻‍♂️"));
+  check(SV(R"(*"hello 🤷🏻\u{200d}♂️"**)"), SV("{:*^22?}"), SV("hello 🤷🏻‍♂️"));
+  check(SV(R"("hello 🤷🏻\u{200d}♂️"***)"), SV("{:*<22?}"), SV("hello 🤷🏻‍♂️"));
 
   // *** width ***
   check(SV(R"("hellö"   )"), SV("{:10?}"), SV("hellö"));
-  check(SV(R"("hello\u{308}"   )"), SV("{:17?}"), SV("hello\u0308"));
-  check(SV(R"("hello 🤷🏻\u{200d}♂\u{fe0f}"   )"), SV("{:30?}"), SV("hello 🤷🏻‍♂️"));
+  check(SV(R"("hellö"   )"), SV("{:10?}"), SV("hello\u0308"));
+  check(SV(R"("hello 🤷🏻\u{200d}♂️"   )"), SV("{:22?}"), SV("hello 🤷🏻‍♂️"));
 
   // *** precision ***
   check(SV(R"("hell)"), SV("{:.5?}"), SV("hellö"));
@@ -431,9 +435,8 @@ void test_format_functions(TestFunction check) {
   check(SV(R"("hello 🤷🏻)"), SV("{:.9?}"), SV("hello 🤷🏻‍♂️"));
   check(SV(R"("hello 🤷🏻\)"), SV("{:.10?}"), SV("hello 🤷🏻‍♂️"));
   check(SV(R"("hello 🤷🏻\u{200d})"), SV("{:.17?}"), SV("hello 🤷🏻‍♂️"));
-  check(SV(R"("hello 🤷🏻\u{200d}♂)"), SV("{:.18?}"), SV("hello 🤷🏻‍♂️"));
-  check(SV(R"("hello 🤷🏻\u{200d}♂\)"), SV("{:.19?}"), SV("hello 🤷🏻‍♂️"));
-  check(SV(R"("hello 🤷🏻\u{200d}♂\u{fe0f}")"), SV("{:.28?}"), SV("hello 🤷🏻‍♂️"));
+  check(SV(R"("hello 🤷🏻\u{200d}♂️)"), SV("{:.18?}"), SV("hello 🤷🏻‍♂️"));
+  check(SV(R"("hello 🤷🏻\u{200d}♂️")"), SV("{:.19?}"), SV("hello 🤷🏻‍♂️"));
 
   // *** width & precision ***
   check(SV(R"("hell#########################)"), SV("{:#<30.5?}"), SV("hellö"));
@@ -445,9 +448,8 @@ void test_format_functions(TestFunction check) {
   check(SV(R"("hello 🤷🏻#####################)"), SV("{:#<30.9?}"), SV("hello 🤷🏻‍♂️"));
   check(SV(R"("hello 🤷🏻\####################)"), SV("{:#<30.10?}"), SV("hello 🤷🏻‍♂️"));
   check(SV(R"("hello 🤷🏻\u{200d}#############)"), SV("{:#<30.17?}"), SV("hello 🤷🏻‍♂️"));
-  check(SV(R"("hello 🤷🏻\u{200d}♂############)"), SV("{:#<30.18?}"), SV("hello 🤷🏻‍♂️"));
-  check(SV(R"("hello 🤷🏻\u{200d}♂\###########)"), SV("{:#<30.19?}"), SV("hello 🤷🏻‍♂️"));
-  check(SV(R"("hello 🤷🏻\u{200d}♂\u{fe0f}"###)"), SV("{:#<30.28?}"), SV("hello 🤷🏻‍♂️"));
+  check(SV(R"("hello 🤷🏻\u{200d}♂️############)"), SV("{:#<30.18?}"), SV("hello 🤷🏻‍♂️"));
+  check(SV(R"("hello 🤷🏻\u{200d}♂️"###########)"), SV("{:#<30.19?}"), SV("hello 🤷🏻‍♂️"));
 }
 
 template <class CharT>
diff --git a/libcxx/utils/generate_escaped_output_table.py b/libcxx/utils/generate_escaped_output_table.py
index c6bde8f2411c..523a0be3a451 100755
--- a/libcxx/utils/generate_escaped_output_table.py
+++ b/libcxx/utils/generate_escaped_output_table.py
@@ -39,12 +39,6 @@ LINE_REGEX = re.compile(
 )
 
 
-def filterCoreProperty(element: PropertyRange) -> Optional[PropertyRange]:
-    if element.prop == "Grapheme_Extend":
-        return element
-    return None
-
-
 # https://www.unicode.org/reports/tr44/#GC_Values_Table
 def filterGeneralProperty(element: PropertyRange) -> Optional[PropertyRange]:
     if element.prop in ["Zs", "Zl", "Zp", "Cc", "Cf", "Cs", "Co", "Cn"]:
@@ -94,10 +88,9 @@ DATA_ARRAY_TEMPLATE = """
 /// The entries of the characters to escape in format's debug string.
 ///
 /// Contains the entries for [format.string.escaped]/2.2.1.2.1
-///   CE is a Unicode encoding and C corresponds to either a UCS scalar value
-///   whose Unicode property General_Category has a value in the groups
-///   Separator (Z) or Other (C) or to a UCS scalar value which has the Unicode
-///   property Grapheme_Extend=Yes, as described by table 12 of UAX #44
+///   CE is a Unicode encoding and C corresponds to a UCS scalar value whose
+///   Unicode property General_Category has a value in the groups Separator (Z)
+///   or Other (C), as described by table 12 of UAX #44
 ///
 /// Separator (Z) consists of General_Category
 /// - Space_Separator,
@@ -112,7 +105,6 @@ DATA_ARRAY_TEMPLATE = """
 /// - Unassigned.
 ///
 /// The data is generated from
-/// - https://www.unicode.org/Public/UCD/latest/ucd/DerivedCoreProperties.txt
 /// - https://www.unicode.org/Public/UCD/latest/ucd/extracted/DerivedGeneralCategory.txt
 ///
 /// The table is similar to the table
@@ -121,34 +113,41 @@ DATA_ARRAY_TEMPLATE = """
 /// table lacks a property, thus having more bits available for the size.
 ///
 /// The data has 2 values:
-/// - bits [0, 10] The size of the range, allowing 2048 elements.
-/// - bits [11, 31] The lower bound code point of the range. The upper bound of
-///   the range is lower bound + size.
+/// - bits [0, 13] The size of the range, allowing 16384 elements.
+/// - bits [14, 31] The lower bound code point of the range. The upper bound of
+///   the range is lower bound + size. Note the code expects code units the fit
+///   into 18 bits, instead of the 21 bits needed for the full Unicode range.
 _LIBCPP_HIDE_FROM_ABI inline constexpr uint32_t __entries[{size}] = {{
 {entries}}};
 
+/// Returns whether the code unit needs to be escaped.
+///
 /// At the end of the valid Unicode code points space a lot of code points are
 /// either reserved or a noncharacter. Adding all these entries to the
-/// lookup table would add 446 entries to the table (in Unicode 14).
-/// Instead the only the start of the region is stored, every code point in
-/// this region needs to be escaped.
-inline constexpr uint32_t __unallocated_region_lower_bound = 0x{unallocated:08x};
+/// lookup table would greatly increase the size of the table. Instead these
+/// entries are manually processed. In this large area of reserved code points,
+/// there is a small area of extended graphemes that should not be escaped
+/// unconditionally. This is also manually coded. See the generation script for
+/// more details.
 
-/// Returns whether the code unit needs to be escaped.
 ///
 /// \pre The code point is a valid Unicode code point.
 [[nodiscard]] _LIBCPP_HIDE_FROM_ABI constexpr bool __needs_escape(const char32_t __code_point) noexcept {{
-  // Since __unallocated_region_lower_bound contains the unshifted range do the
-  // comparison without shifting.
-  if (__code_point >= __unallocated_region_lower_bound)
+
+  // The entries in the gap at the end.
+  if(__code_point >= 0x{gap_lower:08x} && __code_point <= 0x{gap_upper:08x})
+     return false;
+
+  // The entries at the end.
+  if (__code_point >= 0x{unallocated:08x})
     return true;
 
-  ptrdiff_t __i = std::ranges::upper_bound(__entries, (__code_point << 11) | 0x7ffu) - __entries;
+  ptrdiff_t __i = std::ranges::upper_bound(__entries, (__code_point << 14) | 0x3fffu) - __entries;
   if (__i == 0)
     return false;
 
   --__i;
-  uint32_t __upper_bound = (__entries[__i] >> 11) + (__entries[__i] & 0x7ffu);
+  uint32_t __upper_bound = (__entries[__i] >> 14) + (__entries[__i] & 0x3fffu);
   return __code_point <= __upper_bound;
 }}
 """
@@ -253,28 +252,43 @@ def property_ranges_to_table(ranges: list[PropertyRange]) -> list[Entry]:
 
         while True:
             e = Entry(range.lower, range.upper - range.lower)
-            if e.offset <= 2047:
+            if e.offset <= 16383:
                 result.append(e)
                 break
-            e.offset = 2047
+            e.offset = 16383
             result.append(e)
-            range.lower += 2048
+            range.lower += 16384
     return result
 
 
-cpp_entrytemplate = "    0x{:08x}"
+cpp_entrytemplate = "    0x{:08x} /* {:08x} - {:08x} [{:>5}] */"
 
 
-def generate_cpp_data(ranges: list[PropertyRange], unallocated: int) -> str:
+def generate_cpp_data(
+    ranges: list[PropertyRange], unallocated: int, gap_lower: int, gap_upper: int
+) -> str:
     result = StringIO()
     table = property_ranges_to_table(ranges)
+    # Validates all entries fit in 18 bits.
+    for x in table:
+        assert x.lower + x.offset < 0x3FFFF
     result.write(
         DATA_ARRAY_TEMPLATE.format(
             size=len(table),
             entries=",\n".join(
-                [cpp_entrytemplate.format(x.lower << 11 | x.offset) for x in table]
+                [
+                    cpp_entrytemplate.format(
+                        x.lower << 14 | x.offset,
+                        x.lower,
+                        x.lower + x.offset,
+                        x.offset + 1,
+                    )
+                    for x in table
+                ]
             ),
             unallocated=unallocated,
+            gap_lower=gap_lower,
+            gap_upper=gap_upper,
         )
     )
 
@@ -291,12 +305,6 @@ def generate_data_tables() -> str:
         / "unicode"
         / "DerivedGeneralCategory.txt"
     )
-    derived_core_catagory_path = (
-        Path(__file__).absolute().parent
-        / "data"
-        / "unicode"
-        / "DerivedCoreProperties.txt"
-    )
 
     properties = list()
     with derived_general_catagory_path.open(encoding="utf-8") as f:
@@ -308,30 +316,31 @@ def generate_data_tables() -> str:
                 )
             )
         )
-    with derived_core_catagory_path.open(encoding="utf-8") as f:
-        properties.extend(
-            list(
-                filter(
-                    filterCoreProperty,
-                    [x for line in f if (x := parsePropertyLine(line))],
-                )
-            )
-        )
 
     data = compactPropertyRanges(sorted(properties, key=lambda x: x.lower))
 
-    # The last entry is large. In Unicode 14 it contains the entries
-    # 3134B..0FFFF 912564 elements
-    # This are 446 entries of 1325 entries in the table.
-    # Based on the nature of these entries it is expected they remain for the
-    # forseeable future. Therefore we only store the lower bound of this section.
-    #
-    # When this region becomes substantially smaller we need to investigate
-    # this design.
-    assert data[-1].upper == 0x10FFFF
-    assert data[-1].upper - data[-1].lower > 900000
-
-    return "\n".join([generate_cpp_data(data[:-1], data[-1].lower)])
+    # The output table has two large entries at the end, with a small "gap"
+    #   E0100..E01EF  ; Grapheme_Extend # Mn [240] VARIATION SELECTOR-17..VARIATION SELECTOR-256
+    # Based on Unicode 15.1.0:
+    # - Encoding all these entries in the table requires 1173 entries.
+    # - Manually handling these last two blocks reduces the size to 729 entries.
+    # This not only reduces the binary size, but also improves the performance
+    # by having fewer elements to search.
+    # The exact entries may differ between Unicode versions. When these numbers
+    # change the test needs to be updated too.
+    #   libcxx/test/libcxx/utilities/format/format.string/format.string.std/escaped_output.pass.cpp
+    assert (data[-2].lower) == 0x323B0
+    assert (data[-2].upper) == 0xE00FF
+    assert (data[-1].lower) == 0xE01F0
+    assert (data[-1].upper) == 0x10FFFF
+
+    return "\n".join(
+        [
+            generate_cpp_data(
+                data[:-2], data[-2].lower, data[-2].upper + 1, data[-1].lower - 1
+            )
+        ]
+    )
 
 
 if __name__ == "__main__":
diff --git a/libcxx/utils/libcxx/header_information.py b/libcxx/utils/libcxx/header_information.py
index bccae353b0c6..e2165d6ab80b 100644
--- a/libcxx/utils/libcxx/header_information.py
+++ b/libcxx/utils/libcxx/header_information.py
@@ -161,6 +161,10 @@ def is_header(file):
     ]
 
 
+def is_public_header(header):
+    return "__" not in header and not header.startswith("ext/")
+
+
 def is_modulemap_header(header):
     """Returns whether a header should be listed in the modulemap"""
     # TODO: Should `__config_site` be in the modulemap?
@@ -192,17 +196,18 @@ test = pathlib.Path(os.path.join(libcxx_root, "test"))
 assert libcxx_root.exists()
 
 all_headers = sorted(
-    p.relative_to(include).as_posix() for p in include.rglob("[a-z]*") if is_header(p)
+    p.relative_to(include).as_posix() for p in include.rglob("[_a-z]*") if is_header(p)
 )
 toplevel_headers = sorted(
-    p.relative_to(include).as_posix() for p in include.glob("[a-z]*") if is_header(p)
+    p.relative_to(include).as_posix() for p in include.glob("[_a-z]*") if is_header(p)
 )
 experimental_headers = sorted(
     p.relative_to(include).as_posix()
     for p in include.glob("experimental/[a-z]*")
     if is_header(p)
 )
-public_headers = toplevel_headers + experimental_headers
+
+public_headers = [p for p in all_headers if is_public_header(p)]
 
 # The headers used in the std and std.compat modules.
 #
@@ -210,7 +215,7 @@ public_headers = toplevel_headers + experimental_headers
 module_headers = [
     header
     for header in toplevel_headers
-    if not header.endswith(".h")
+    if not header.endswith(".h") and is_public_header(header)
     # These headers have been removed in C++20 so are never part of a module.
     and not header in ["ccomplex", "ciso646", "cstdbool", "ctgmath"]
 ]
diff --git a/lld/COFF/MinGW.cpp b/lld/COFF/MinGW.cpp
index e46f5277a8c3..29c01da9e28f 100644
--- a/lld/COFF/MinGW.cpp
+++ b/lld/COFF/MinGW.cpp
@@ -50,7 +50,6 @@ AutoExporter::AutoExporter(
       "libclang_rt.profile-x86_64",
       "libc++",
       "libc++abi",
-      "libFortran_main",
       "libFortranRuntime",
       "libFortranDecimal",
       "libunwind",
diff --git a/lld/ELF/Arch/RISCV.cpp b/lld/ELF/Arch/RISCV.cpp
index 20de1b9b7bde..e4d63250135e 100644
--- a/lld/ELF/Arch/RISCV.cpp
+++ b/lld/ELF/Arch/RISCV.cpp
@@ -15,8 +15,8 @@
 #include "llvm/Support/LEB128.h"
 #include "llvm/Support/RISCVAttributeParser.h"
 #include "llvm/Support/RISCVAttributes.h"
-#include "llvm/Support/RISCVISAInfo.h"
 #include "llvm/Support/TimeProfiler.h"
+#include "llvm/TargetParser/RISCVISAInfo.h"
 
 using namespace llvm;
 using namespace llvm::object;
@@ -1057,7 +1057,7 @@ public:
 };
 } // namespace
 
-static void mergeArch(RISCVISAInfo::OrderedExtensionMap &mergedExts,
+static void mergeArch(RISCVISAUtils::OrderedExtensionMap &mergedExts,
                       unsigned &mergedXlen, const InputSectionBase *sec,
                       StringRef s) {
   auto maybeInfo = RISCVISAInfo::parseNormalizedArchString(s);
@@ -1086,7 +1086,7 @@ static void mergeArch(RISCVISAInfo::OrderedExtensionMap &mergedExts,
 
 static RISCVAttributesSection *
 mergeAttributesSection(const SmallVector<InputSectionBase *, 0> &sections) {
-  RISCVISAInfo::OrderedExtensionMap exts;
+  RISCVISAUtils::OrderedExtensionMap exts;
   const InputSectionBase *firstStackAlign = nullptr;
   unsigned firstStackAlignValue = 0, xlen = 0;
   bool hasArch = false;
diff --git a/lld/test/ELF/lto/i386-global-offset-table.ll b/lld/test/ELF/lto/i386-global-offset-table.ll
new file mode 100644
index 000000000000..3fa11f6e6d40
--- /dev/null
+++ b/lld/test/ELF/lto/i386-global-offset-table.ll
@@ -0,0 +1,29 @@
+; REQUIRES: x86
+;; LTO-generated relocatable files may reference _GLOBAL_OFFSET_TABLE_ while
+;; the IR does not mention _GLOBAL_OFFSET_TABLE_.
+;; Test that there is no spurious "undefined symbol" error.
+
+; RUN: rm -rf %t && mkdir %t && cd %t
+; RUN: llvm-as %s -o a.bc
+; RUN: ld.lld -pie a.bc -o a
+; RUN: llvm-nm a | FileCheck %s
+
+; CHECK: d _GLOBAL_OFFSET_TABLE_
+
+target datalayout = "e-m:e-p:32:32-p270:32:32-p271:32:32-p272:64:64-i128:128-f64:32:64-f80:32-n8:16:32-S128"
+target triple = "i386-pc-linux-gnu"
+
+@i = global i32 0
+
+define dso_local void @_start() {
+entry:
+  %0 = load i32, ptr @i
+  %inc = add nsw i32 %0, 1
+  store i32 %inc, ptr @i
+  ret void
+}
+
+!llvm.module.flags = !{!0, !1}
+
+!0 = !{i32 8, !"PIC Level", i32 2}
+!1 = !{i32 7, !"PIE Level", i32 2}
diff --git a/lld/test/ELF/lto/x86-64-global-offset-table.ll b/lld/test/ELF/lto/x86-64-global-offset-table.ll
new file mode 100644
index 000000000000..4b0f638b409e
--- /dev/null
+++ b/lld/test/ELF/lto/x86-64-global-offset-table.ll
@@ -0,0 +1,71 @@
+; REQUIRES: x86
+;; LTO-generated relocatable files may reference _GLOBAL_OFFSET_TABLE_ while
+;; the IR does not mention _GLOBAL_OFFSET_TABLE_.
+;; Test that there is no spurious "undefined symbol" error.
+
+; RUN: rm -rf %t && split-file %s %t && cd %t
+; RUN: opt -module-summary b.ll -o b.bc
+
+;; Test Thin LTO.
+; RUN: cat a.ll medium.ll | opt -module-summary - -o medium.bc
+; RUN: ld.lld -pie --no-relax medium.bc b.bc -o medium
+; RUN: llvm-objdump -dt medium | FileCheck %s
+
+;; Test regular LTO.
+; RUN: cat a.ll large.ll | llvm-as - -o large.bc
+; RUN: ld.lld -pie large.bc b.bc -o large
+; RUN: llvm-objdump -dt large | FileCheck %s
+
+;; Explicit reference of _GLOBAL_OFFSET_TABLE_ is fine.
+; RUN: cat a.ll medium.ll ref.ll | opt -module-summary - -o ref.bc
+; RUN: ld.lld -pie -u ref ref.bc b.bc -y _GLOBAL_OFFSET_TABLE_ -o ref 2>&1 | FileCheck %s --check-prefix=TRACE
+; RUN: llvm-objdump -dt ref | FileCheck %s
+
+; TRACE:      ref.bc: reference to _GLOBAL_OFFSET_TABLE_
+; TRACE-NEXT: ref.bc: reference to _GLOBAL_OFFSET_TABLE_
+; TRACE-NEXT: <internal>: definition of _GLOBAL_OFFSET_TABLE_
+; TRACE-NEXT: ref.lto.ref.o: reference to _GLOBAL_OFFSET_TABLE_
+
+;; The IR symbol table references _GLOBAL_OFFSET_TABLE_, which causes lld to define the symbol.
+; CHECK: .got.plt       0000000000000000 .hidden _GLOBAL_OFFSET_TABLE_
+; CHECK: movabsq
+
+;--- a.ll
+target datalayout = "e-m:e-p270:32:32-p271:32:32-p272:64:64-i64:64-i128:128-f80:128-n8:16:32:64-S128"
+target triple = "x86_64-unknown-linux-gnu"
+
+@i = external global i32
+
+define dso_local void @_start() {
+entry:
+  %0 = load i32, ptr @i
+  %inc = add nsw i32 %0, 1
+  store i32 %inc, ptr @i
+  ret void
+}
+
+!llvm.module.flags = !{!0, !1, !2, !3}
+
+!0 = !{i32 8, !"PIC Level", i32 2}
+!1 = !{i32 7, !"PIE Level", i32 2}
+!2 = !{i32 1, !"Large Data Threshold", i64 0}
+
+;--- medium.ll
+!3 = !{i32 1, !"Code Model", i32 3}
+
+;--- large.ll
+!3 = !{i32 1, !"Code Model", i32 4}
+
+;--- ref.ll
+@_GLOBAL_OFFSET_TABLE_ = external global [0 x i8]
+
+define dso_local ptr @ref() {
+entry:
+  ret ptr @_GLOBAL_OFFSET_TABLE_
+}
+
+;--- b.ll
+target datalayout = "e-m:e-p270:32:32-p271:32:32-p272:64:64-i64:64-i128:128-f80:128-n8:16:32:64-S128"
+target triple = "x86_64-unknown-linux-gnu"
+
+@i = global i32 0
diff --git a/lld/test/ELF/mips-eh_frame-pic.s b/lld/test/ELF/mips-eh_frame-pic.s
index a84c36b0e5ec..c04dbdf57b08 100644
--- a/lld/test/ELF/mips-eh_frame-pic.s
+++ b/lld/test/ELF/mips-eh_frame-pic.s
@@ -36,8 +36,8 @@
 # RELOCS:            .rel{{a?}}.eh_frame {
 # ABS32-RELOCS-NEXT:   0x1C R_MIPS_32 .text
 # ABS64-RELOCS-NEXT:   0x1C R_MIPS_64/R_MIPS_NONE/R_MIPS_NONE .text
-# PIC64-RELOCS-NEXT:   0x1C R_MIPS_PC32/R_MIPS_NONE/R_MIPS_NONE <null>
-# PIC32-RELOCS-NEXT:   0x1C R_MIPS_PC32 <null>
+# PIC64-RELOCS-NEXT:   0x1C R_MIPS_PC32/R_MIPS_NONE/R_MIPS_NONE .L0
+# PIC32-RELOCS-NEXT:   0x1C R_MIPS_PC32 .L0
 # RELOCS-NEXT:       }
 
 # ABS64-EH-FRAME: Augmentation data: 0C
diff --git a/lld/test/ELF/riscv-branch.s b/lld/test/ELF/riscv-branch.s
index dbf39dc0bb8f..1a2b446b5a43 100644
--- a/lld/test/ELF/riscv-branch.s
+++ b/lld/test/ELF/riscv-branch.s
@@ -7,19 +7,19 @@
 # RUN: ld.lld %t.rv64.o --defsym foo=_start+4 --defsym bar=_start -o %t.rv64
 # RUN: llvm-objdump -d %t.rv32 | FileCheck %s --check-prefix=CHECK-32
 # RUN: llvm-objdump -d %t.rv64 | FileCheck %s --check-prefix=CHECK-64
-# CHECK-32: 63 02 00 00     beqz    zero, 0x110b8
-# CHECK-32: e3 1e 00 fe     bnez    zero, 0x110b4
-# CHECK-64: 63 02 00 00     beqz    zero, 0x11124
-# CHECK-64: e3 1e 00 fe     bnez    zero, 0x11120
+# CHECK-32: 00000263     beqz    zero, 0x110b8
+# CHECK-32: fe001ee3     bnez    zero, 0x110b4
+# CHECK-64: 00000263     beqz    zero, 0x11124
+# CHECK-64: fe001ee3     bnez    zero, 0x11120
 #
 # RUN: ld.lld %t.rv32.o --defsym foo=_start+0xffe --defsym bar=_start+4-0x1000 -o %t.rv32.limits
 # RUN: ld.lld %t.rv64.o --defsym foo=_start+0xffe --defsym bar=_start+4-0x1000 -o %t.rv64.limits
 # RUN: llvm-objdump -d %t.rv32.limits | FileCheck --check-prefix=LIMITS-32 %s
 # RUN: llvm-objdump -d %t.rv64.limits | FileCheck --check-prefix=LIMITS-64 %s
-# LIMITS-32:      e3 0f 00 7e     beqz    zero, 0x120b2
-# LIMITS-32-NEXT: 63 10 00 80     bnez    zero, 0x100b8
-# LIMITS-64:      e3 0f 00 7e     beqz    zero, 0x1211e
-# LIMITS-64-NEXT: 63 10 00 80     bnez    zero, 0x10124
+# LIMITS-32:      7e000fe3     beqz    zero, 0x120b2
+# LIMITS-32-NEXT: 80001063     bnez    zero, 0x100b8
+# LIMITS-64:      7e000fe3     beqz    zero, 0x1211e
+# LIMITS-64-NEXT: 80001063     bnez    zero, 0x10124
 
 # RUN: not ld.lld %t.rv32.o --defsym foo=_start+0x1000 --defsym bar=_start+4-0x1002 -o /dev/null 2>&1 | FileCheck --check-prefix=ERROR-RANGE %s
 # RUN: not ld.lld %t.rv64.o --defsym foo=_start+0x1000 --defsym bar=_start+4-0x1002 -o /dev/null 2>&1 | FileCheck --check-prefix=ERROR-RANGE %s
diff --git a/lld/test/ELF/riscv-call.s b/lld/test/ELF/riscv-call.s
index 5fef156df0bb..0e81e9b4710e 100644
--- a/lld/test/ELF/riscv-call.s
+++ b/lld/test/ELF/riscv-call.s
@@ -7,19 +7,19 @@
 # RUN: ld.lld %t.rv64.o --defsym foo=_start+8 --defsym bar=_start -o %t.rv64
 # RUN: llvm-objdump -d %t.rv32 | FileCheck %s
 # RUN: llvm-objdump -d %t.rv64 | FileCheck %s
-# CHECK:      97 00 00 00     auipc   ra, 0x0
-# CHECK-NEXT: e7 80 80 00     jalr    0x8(ra)
-# CHECK:      97 00 00 00     auipc   ra, 0x0
-# CHECK-NEXT: e7 80 80 ff     jalr    -0x8(ra)
+# CHECK:      00000097     auipc   ra, 0x0
+# CHECK-NEXT: 008080e7     jalr    0x8(ra)
+# CHECK:      00000097     auipc   ra, 0x0
+# CHECK-NEXT: ff8080e7     jalr    -0x8(ra)
 
 # RUN: ld.lld %t.rv32.o --defsym foo=_start+0x7ffff7ff --defsym bar=_start+8-0x80000800 -o %t.rv32.limits
 # RUN: ld.lld %t.rv64.o --defsym foo=_start+0x7ffff7ff --defsym bar=_start+8-0x80000800 -o %t.rv64.limits
 # RUN: llvm-objdump -d %t.rv32.limits | FileCheck --check-prefix=LIMITS %s
 # RUN: llvm-objdump -d %t.rv64.limits | FileCheck --check-prefix=LIMITS %s
-# LIMITS:      97 f0 ff 7f     auipc   ra, 0x7ffff
-# LIMITS-NEXT: e7 80 f0 7f     jalr    0x7ff(ra)
-# LIMITS-NEXT: 97 00 00 80     auipc   ra, 0x80000
-# LIMITS-NEXT: e7 80 00 80     jalr    -0x800(ra)
+# LIMITS:      7ffff097     auipc   ra, 0x7ffff
+# LIMITS-NEXT: 7ff080e7     jalr    0x7ff(ra)
+# LIMITS-NEXT: 80000097     auipc   ra, 0x80000
+# LIMITS-NEXT: 800080e7     jalr    -0x800(ra)
 
 # RUN: ld.lld %t.rv32.o --defsym foo=_start+0x7ffff800 --defsym bar=_start+8-0x80000801 -o %t
 # RUN: not ld.lld %t.rv64.o --defsym foo=_start+0x7ffff800 --defsym bar=_start+8-0x80000801 -o /dev/null 2>&1 | \
diff --git a/lld/test/ELF/riscv-hi20-lo12.s b/lld/test/ELF/riscv-hi20-lo12.s
index 85861432db0b..b9786f563f28 100644
--- a/lld/test/ELF/riscv-hi20-lo12.s
+++ b/lld/test/ELF/riscv-hi20-lo12.s
@@ -7,23 +7,23 @@
 # RUN: ld.lld %t.rv64.o --defsym foo=0 --defsym bar=42 -o %t.rv64
 # RUN: llvm-objdump -d %t.rv32 | FileCheck %s
 # RUN: llvm-objdump -d %t.rv64 | FileCheck %s
-# CHECK:      37 05 00 00     lui     a0, 0x0
-# CHECK-NEXT: 13 05 05 00     mv      a0, a0
-# CHECK-NEXT: 23 20 a5 00     sw      a0, 0x0(a0)
-# CHECK-NEXT: b7 05 00 00     lui     a1, 0x0
-# CHECK-NEXT: 93 85 a5 02     addi    a1, a1, 0x2a
-# CHECK-NEXT: 23 a5 b5 02     sw      a1, 0x2a(a1)
+# CHECK:      00000537     lui     a0, 0x0
+# CHECK-NEXT: 00050513     mv      a0, a0
+# CHECK-NEXT: 00a52023     sw      a0, 0x0(a0)
+# CHECK-NEXT: 000005b7     lui     a1, 0x0
+# CHECK-NEXT: 02a58593     addi    a1, a1, 0x2a
+# CHECK-NEXT: 02b5a523     sw      a1, 0x2a(a1)
 
 # RUN: ld.lld %t.rv32.o --defsym foo=0x7ffff7ff --defsym bar=0x7ffff800 -o %t.rv32.limits
 # RUN: ld.lld %t.rv64.o --defsym foo=0x7ffff7ff --defsym bar=0xffffffff7ffff800 -o %t.rv64.limits
 # RUN: llvm-objdump -d %t.rv32.limits | FileCheck --check-prefix=LIMITS %s
 # RUN: llvm-objdump -d %t.rv64.limits | FileCheck --check-prefix=LIMITS %s
-# LIMITS:      37 f5 ff 7f     lui     a0, 0x7ffff
-# LIMITS-NEXT: 13 05 f5 7f     addi    a0, a0, 0x7ff
-# LIMITS-NEXT: a3 2f a5 7e     sw      a0, 0x7ff(a0)
-# LIMITS-NEXT: b7 05 00 80     lui     a1, 0x80000
-# LIMITS-NEXT: 93 85 05 80     addi    a1, a1, -0x800
-# LIMITS-NEXT: 23 a0 b5 80     sw      a1, -0x800(a1)
+# LIMITS:      7ffff537     lui     a0, 0x7ffff
+# LIMITS-NEXT: 7ff50513     addi    a0, a0, 0x7ff
+# LIMITS-NEXT: 7ea52fa3     sw      a0, 0x7ff(a0)
+# LIMITS-NEXT: 800005b7     lui     a1, 0x80000
+# LIMITS-NEXT: 80058593     addi    a1, a1, -0x800
+# LIMITS-NEXT: 80b5a023     sw      a1, -0x800(a1)
 
 # RUN: not ld.lld %t.rv64.o --defsym foo=0x7ffff800 --defsym bar=0xffffffff7ffff7ff -o /dev/null 2>&1 | FileCheck --check-prefix ERROR %s
 # ERROR: relocation R_RISCV_HI20 out of range: 524288 is not in [-524288, 524287]; references 'foo'
diff --git a/lld/test/ELF/riscv-jal.s b/lld/test/ELF/riscv-jal.s
index cd3b842aad60..2129e4454706 100644
--- a/lld/test/ELF/riscv-jal.s
+++ b/lld/test/ELF/riscv-jal.s
@@ -7,19 +7,19 @@
 # RUN: ld.lld %t.rv64.o --defsym foo=_start+4 --defsym bar=_start -o %t.rv64
 # RUN: llvm-objdump -d %t.rv32 | FileCheck %s --check-prefix=CHECK-32
 # RUN: llvm-objdump -d %t.rv64 | FileCheck %s --check-prefix=CHECK-64
-# CHECK-32: 6f 00 40 00    j   0x110b8
-# CHECK-32: ef f0 df ff    jal 0x110b4
-# CHECK-64: 6f 00 40 00    j   0x11124
-# CHECK-64: ef f0 df ff    jal 0x11120
+# CHECK-32: 0040006f    j   0x110b8
+# CHECK-32: ffdff0ef    jal 0x110b4
+# CHECK-64: 0040006f    j   0x11124
+# CHECK-64: ffdff0ef    jal 0x11120
 
 # RUN: ld.lld %t.rv32.o --defsym foo=_start+0xffffe --defsym bar=_start+4-0x100000 -o %t.rv32.limits
 # RUN: ld.lld %t.rv64.o --defsym foo=_start+0xffffe --defsym bar=_start+4-0x100000 -o %t.rv64.limits
 # RUN: llvm-objdump -d %t.rv32.limits | FileCheck --check-prefix=LIMITS-32 %s
 # RUN: llvm-objdump -d %t.rv64.limits | FileCheck --check-prefix=LIMITS-64 %s
-# LIMITS-32:      6f f0 ff 7f j   0x1110b2
-# LIMITS-32-NEXT: ef 00 00 80 jal 0xfff110b8
-# LIMITS-64:      6f f0 ff 7f j   0x11111e
-# LIMITS-64-NEXT: ef 00 00 80 jal 0xfffffffffff11124
+# LIMITS-32:      7ffff06f j   0x1110b2
+# LIMITS-32-NEXT: 800000ef jal 0xfff110b8
+# LIMITS-64:      7ffff06f j   0x11111e
+# LIMITS-64-NEXT: 800000ef jal 0xfffffffffff11124
 
 # RUN: not ld.lld %t.rv32.o --defsym foo=_start+0x100000 --defsym bar=_start+4-0x100002 -o /dev/null 2>&1 | FileCheck --check-prefix=ERROR-RANGE %s
 # RUN: not ld.lld %t.rv64.o --defsym foo=_start+0x100000 --defsym bar=_start+4-0x100002 -o /dev/null 2>&1 | FileCheck --check-prefix=ERROR-RANGE %s
diff --git a/lld/test/wasm/init-fini.ll b/lld/test/wasm/init-fini.ll
index 3d2e9a78043e..ef2f41f96e89 100644
--- a/lld/test/wasm/init-fini.ll
+++ b/lld/test/wasm/init-fini.ll
@@ -78,7 +78,7 @@ entry:
 ; CHECK-NEXT:         Body:            10041005100A100F1012100F10141004100C100F10161002100E0B
 ; CHECK:            - Index:           22
 ; CHECK-NEXT:         Locals:
-; CHECK-NEXT:         Body:            02404186808080004100418088808000108080808000450D0000000B0B
+; CHECK-NEXT:         Body:            02404186808080004100418088808000108080808000450D00000B0B
 ; CHECK-NEXT:   - Type:            CUSTOM
 ; CHECK-NEXT:     Name:            name
 ; CHECK-NEXT:     FunctionNames:
diff --git a/lldb/CMakeLists.txt b/lldb/CMakeLists.txt
index 7844d93d78d2..b0764f105327 100644
--- a/lldb/CMakeLists.txt
+++ b/lldb/CMakeLists.txt
@@ -75,6 +75,14 @@ if (LLDB_ENABLE_PYTHON)
       endif()
     endif()
   endforeach()
+  # Make sure lldb extension has "_d" suffix on Windows in Debug mode.
+  if(WIN32 AND CMAKE_BUILD_TYPE STREQUAL Debug)
+    string(SUBSTRING ${LLDB_PYTHON_EXT_SUFFIX} 0 2 FIRST_2_CHARS)
+    # Add "_d" manually if LLDB_PYTHON_EXT_SUFFIX lacks it due to release Python configuration.
+    if(NOT FIRST_2_CHARS STREQUAL "_d")
+      set(LLDB_PYTHON_EXT_SUFFIX "_d${LLDB_PYTHON_EXT_SUFFIX}")
+    endif()
+  endif()
 endif ()
 
 if (LLDB_ENABLE_LUA)
diff --git a/lldb/bindings/interface/SBValueDocstrings.i b/lldb/bindings/interface/SBValueDocstrings.i
index 6bab923e8b35..59fa807f5ec9 100644
--- a/lldb/bindings/interface/SBValueDocstrings.i
+++ b/lldb/bindings/interface/SBValueDocstrings.i
@@ -135,6 +135,26 @@ linked list."
 %feature("docstring", "Expands nested expressions like .a->b[0].c[1]->d."
 ) lldb::SBValue::GetValueForExpressionPath;
 
+%feature("docstring", "
+      Return the value as an address.  On failure, LLDB_INVALID_ADDRESS
+      will be returned.  On architectures like AArch64, where the
+      top (unaddressable) bits can be used for authentication,
+      memory tagging, or top byte ignore,  this method will return
+      the value with those top bits cleared.
+
+      GetValueAsUnsigned returns the actual value, with the
+      authentication/Top Byte Ignore/Memory Tagging Extension bits.
+
+      Calling this on a random value which is not a pointer is
+      incorrect.  Call GetType().IsPointerType() if in doubt.
+
+      An SB API program may want to show both the literal byte value
+      and the address it refers to in memory.  These two SBValue
+      methods allow SB API writers to behave appropriately for their
+      interface."
+) lldb::SBValue::GetValueAsAddress;
+
+
 %feature("doctstring", "
     Returns the number for children.
 
diff --git a/lldb/docs/conf.py b/lldb/docs/conf.py
index ec7f93710ab6..27a1cd7c3c31 100644
--- a/lldb/docs/conf.py
+++ b/lldb/docs/conf.py
@@ -13,6 +13,9 @@
 import sys, os, re, shutil
 from datetime import date
 
+# Add path for llvm_slug module.
+sys.path.insert(0, os.path.abspath(os.path.join("..", "..", "llvm", "docs")))
+
 building_man_page = tags.has("builder-man")
 
 # For the website we need to setup the path to the generated LLDB module that
@@ -42,6 +45,23 @@ automodapi_toctreedirnm = "python_api"
 # coming with Sphinx (named 'sphinx.ext.*') or your custom ones.
 extensions = ["sphinx.ext.todo", "sphinx.ext.mathjax", "sphinx.ext.intersphinx"]
 
+# When building man pages, we do not use the markdown pages,
+# So, we can continue without the myst_parser dependencies.
+# Doing so reduces dependencies of some packaged llvm distributions.
+try:
+    import myst_parser
+
+    extensions.append("myst_parser")
+except ImportError:
+    if not tags.has("builder-man"):
+        raise
+
+# Automatic anchors for markdown titles
+from llvm_slug import make_slug
+
+myst_heading_anchors = 6
+myst_heading_slug_func = make_slug
+
 autodoc_default_options = {"special-members": True}
 
 # Unless we only generate the basic manpage we need the plugin for generating
@@ -69,6 +89,7 @@ templates_path = ["_templates"]
 # The suffix of source filenames.
 source_suffix = {
     ".rst": "restructuredtext",
+    ".md": "markdown",
 }
 
 # The encoding of source files.
diff --git a/lldb/docs/index.rst b/lldb/docs/index.rst
index c378ab97d97b..7a27f6914fa8 100644
--- a/lldb/docs/index.rst
+++ b/lldb/docs/index.rst
@@ -157,6 +157,8 @@ interesting areas to contribute to lldb.
    resources/sbapi
    resources/dataformatters
    resources/extensions
+   resources/lldbgdbremote
+   resources/lldbplatformpackets
    resources/caveats
    resources/projects
    Public C++ API <https://lldb.llvm.org/cpp_reference/namespacelldb.html>
diff --git a/lldb/docs/lldb-gdb-remote.txt b/lldb/docs/lldb-gdb-remote.txt
deleted file mode 100644
index 6c29de61daba..000000000000
--- a/lldb/docs/lldb-gdb-remote.txt
+++ /dev/null
@@ -1,2286 +0,0 @@
-LLDB has added new GDB server packets to better support multi-threaded and
-remote debugging. Why? Normally you need to start the correct GDB and the
-correct GDB server when debugging. If you have mismatch, then things go wrong
-very quickly. LLDB makes extensive use of the GDB remote protocol and we
-wanted to make sure that the experience was a bit more dynamic where we can
-discover information about a remote target without having to know anything up
-front. We also ran into performance issues with the existing GDB remote
-protocol that can be overcome when using a reliable communications layer.
-Some packets improve performance, others allow for remote process launching
-(if you have an OS), and others allow us to dynamically figure out what
-registers a thread might have. Again with GDB, both sides pre-agree on how the
-registers will look (how many, their register number,name and offsets). We
-prefer to be able to dynamically determine what kind of architecture, OS and
-vendor we are debugging, as well as how things are laid out when it comes to
-the thread register contexts. Below are the details on the new packets we have
-added above and beyond the standard GDB remote protocol packets.
-
-//----------------------------------------------------------------------
-// "QStartNoAckMode"
-//
-// BRIEF
-//  Try to enable no ACK mode to skip sending ACKs and NACKs.
-//
-// PRIORITY TO IMPLEMENT
-//  High. Any GDB remote server that can implement this should if the
-//  connection is reliable. This improves packet throughput and increases
-//  the performance of the connection.
-//----------------------------------------------------------------------
-Having to send an ACK/NACK after every packet slows things down a bit, so we
-have a way to disable ACK packets to minimize the traffic for reliable
-communication interfaces (like sockets). Below GDB or LLDB will send this
-packet to try and disable ACKs. All lines that start with "send packet: " are
-from GDB/LLDB, and all lines that start with "read packet: " are from the GDB
-remote server:
-
-send packet: $QStartNoAckMode#b0
-read packet: +
-read packet: $OK#9a
-send packet: +
-
-//----------------------------------------------------------------------
-// "QSupported"
-//
-// BRIEF
-//  Query the GDB remote server for features it supports
-//
-// PRIORITY TO IMPLEMENT
-//  Optional.
-//----------------------------------------------------------------------
-
-QSupported is a standard GDB Remote Serial Protocol packet, but
-there are several additions to the response that lldb can parse.
-They are not all listed here.
-
-An example exchange:
-
-send packet: qSupported:xmlRegisters=i386,arm,mips,arc;multiprocess+;fork-events+;vfork-events+
-
-read packet: qXfer:features:read+;PacketSize=20000;qEcho+;native-signals+;SupportedCompressions=lzfse,zlib-deflate,lz4,lzma;SupportedWatchpointTypes=aarch64-mask,aarch64-bas;
-
-In the example above, three lldb extensions are shown:
-
-  PacketSize=20000
-    The base 16 maximum packet size that the stub can handle.
-  SupportedCompressions=<item,item,...>
-    A list of compression types that the stub can use to compress packets 
-    when the QEnableCompression packet is used to request one of them.
-  SupportedWatchpointTypes=<item,item,...>
-    A list of watchpoint types that this stub can manage.
-    Currently defined names are:
-        x86_64       64-bit x86-64 watchpoints
-                     (1, 2, 4, 8 byte watchpoints aligned to those amounts)
-        aarch64-bas  AArch64 Byte Address Select watchpoints
-                     (any number of contiguous bytes within a doubleword)
-        aarch64-mask AArch64 MASK watchpoints
-                     (any power-of-2 region of memory from 8 to 2GB, aligned)
-    If nothing is specified, lldb will default to sending power-of-2 
-    watchpoints, up to a pointer size, `sizeof(void*)`, a reasonable 
-    baseline assumption.
-
-//----------------------------------------------------------------------
-// "A" - launch args packet
-//
-// BRIEF
-//  Launch a program using the supplied arguments
-//
-// PRIORITY TO IMPLEMENT
-//  Low. Only needed if the remote target wants to launch a target after
-//  making a connection to a GDB server that isn't already connected to
-//  an inferior process.
-//----------------------------------------------------------------------
-
-We have added support for the "set program arguments" packet where we can
-start a connection to a remote server and then later supply the path to the
-executable and the arguments to use when executing:
-
-GDB remote docs for this:
-
-set program arguments(reserved) Aarglen,argnum,arg,...
-
-Where A is followed by the length in bytes of the hex encoded argument,
-followed by an argument integer, and followed by the ASCII characters
-converted into hex bytes foreach arg
-
-send packet: $A98,0,2f566f6c756d65732f776f726b2f67636c6179746f6e2f446f63756d656e74732f7372632f6174746163682f612e6f7574#00
-read packet: $OK#00
-
-The above packet helps when you have remote debugging abilities where you
-could launch a process on a remote host, this isn't needed for bare board
-debugging.
-
-//----------------------------------------------------------------------
-// "QEnvironment:NAME=VALUE"
-//
-// BRIEF
-//  Setup the environment up for a new child process that will soon be
-//  launched using the "A" packet.
-//
-// NB: key/value pairs are sent as-is so gdb-remote protocol meta characters
-//     (e.g. '#' or '$') are not acceptable.  If any non-printable or
-//     metacharacters are present in the strings, QEnvironmentHexEncoded
-//     should be used instead if it is available.  If you don't want to
-//     scan the environment strings before sending, prefer
-//     the QEnvironmentHexEncoded packet over QEnvironment, if it is
-//     available.
-//
-// PRIORITY TO IMPLEMENT
-//  Low. Only needed if the remote target wants to launch a target after
-//  making a connection to a GDB server that isn't already connected to
-//  an inferior process.
-//----------------------------------------------------------------------
-
-Both GDB and LLDB support passing down environment variables. Is it ok to
-respond with a "$#00" (unimplemented):
-
-send packet: $QEnvironment:ACK_COLOR_FILENAME=bold yellow#00
-read packet: $OK#00
-
-This packet can be sent one or more times _prior_ to sending a "A" packet.
-
-//----------------------------------------------------------------------
-// "QEnvironmentHexEncoded:HEX-ENCODING(NAME=VALUE)"
-//
-// BRIEF
-//  Setup the environment up for a new child process that will soon be
-//  launched using the "A" packet.
-//
-// The only difference between this packet and QEnvironment is that the
-// environment key-value pair is ascii hex encoded for transmission.
-// This allows values with gdb-remote metacharacters like '#' to be sent.
-//
-// PRIORITY TO IMPLEMENT
-//  Low. Only needed if the remote target wants to launch a target after
-//  making a connection to a GDB server that isn't already connected to
-//  an inferior process.
-//----------------------------------------------------------------------
-
-Both GDB and LLDB support passing down environment variables. Is it ok to
-respond with a "$#00" (unimplemented):
-
-send packet: $QEnvironment:41434b5f434f4c4f525f46494c454e414d453d626f6c642379656c6c6f77#00
-read packet: $OK#00
-
-This packet can be sent one or more times _prior_ to sending a "A" packet.
-
-//----------------------------------------------------------------------
-// "QEnableErrorStrings"
-//
-// BRIEF
-//  This packet enables reporting of Error strings in remote packet
-//  replies from the server to client. If the server supports this
-//  feature, it should send an OK response. The client can expect the
-//  following error replies if this feature is enabled in the server ->
-//
-//  EXX;AAAAAAAAA
-//
-//  where AAAAAAAAA will be a hex encoded ASCII string.
-//  XX is hex encoded byte number.
-//
-//  It must be noted that even if the client has enabled reporting
-//  strings in error replies, it must not expect error strings to all
-//  error replies.
-//
-// PRIORITY TO IMPLEMENT
-//  Low. Only needed if the remote target wants to provide strings that
-//  are human readable along with an error code.
-//----------------------------------------------------------------------
-
-send packet: $QEnableErrorStrings
-read packet: $OK#00
-
-//----------------------------------------------------------------------
-// "QSetSTDIN:<ascii-hex-path>"
-// "QSetSTDOUT:<ascii-hex-path>"
-// "QSetSTDERR:<ascii-hex-path>"
-//
-// BRIEF
-//  Setup where STDIN, STDOUT, and STDERR go prior to sending an "A"
-//  packet.
-//
-// PRIORITY TO IMPLEMENT
-//  Low. Only needed if the remote target wants to launch a target after
-//  making a connection to a GDB server that isn't already connected to
-//  an inferior process.
-//----------------------------------------------------------------------
-
-When launching a program through the GDB remote protocol with the "A" packet,
-you might also want to specify where stdin/out/err go:
-
-QSetSTDIN:<ascii-hex-path>
-QSetSTDOUT:<ascii-hex-path>
-QSetSTDERR:<ascii-hex-path>
-
-These packets must be sent  _prior_ to sending a "A" packet.
-
-//----------------------------------------------------------------------
-// "QSetWorkingDir:<ascii-hex-path>"
-//
-// BRIEF
-//  Set the working directory prior to sending an "A" packet.
-//
-// PRIORITY TO IMPLEMENT
-//  Low. Only needed if the remote target wants to launch a target after
-//  making a connection to a GDB server that isn't already connected to
-//  an inferior process.
-//----------------------------------------------------------------------
-
-Or specify the working directory:
-
-QSetWorkingDir:<ascii-hex-path>
-
-This packet must be sent  _prior_ to sending a "A" packet.
-
-//----------------------------------------------------------------------
-// "QSetDisableASLR:<bool>"
-//
-// BRIEF
-//  Enable or disable ASLR on the next "A" packet.
-//
-// PRIORITY TO IMPLEMENT
-//  Low. Only needed if the remote target wants to launch a target after
-//  making a connection to a GDB server that isn't already connected to
-//  an inferior process and if the target supports disabling ASLR
-//  (Address space layout randomization).
-//----------------------------------------------------------------------
-
-Or control if ASLR is enabled/disabled:
-
-send packet: QSetDisableASLR:1
-read packet: OK
-
-send packet: QSetDisableASLR:0
-read packet: OK
-
-This packet must be sent  _prior_ to sending a "A" packet.
-
-//----------------------------------------------------------------------
-// QListThreadsInStopReply
-//
-// BRIEF
-//  Enable the threads: and thread-pcs: data in the question-mark packet
-//  ("T packet") responses when the stub reports that a program has
-//  stopped executing.
-//
-// PRIORITY TO IMPLEMENT
-//  Performance.  This is a performance benefit to lldb if the thread id's
-//  and thread pc values are provided to lldb in the T stop packet -- if
-//  they are not provided to lldb, lldb will likely need to send one to
-//  two packets per thread to fetch the data at every private stop.
-//----------------------------------------------------------------------
-
-send packet: QListThreadsInStopReply
-read packet: OK
-
-//----------------------------------------------------------------------
-// jLLDBTraceSupported
-//
-// BRIEF
-//  Get the processor tracing type supported by the gdb-server for the current
-//  inferior. Responses might be different depending on the architecture and
-//  capabilities of the underlying OS.
-//
-//  OUTPUT SCHEMA
-//   {
-//     "name": <string>,
-//         Tracing technology name, e.g. intel-pt, arm-etm.
-//     "description": <string>,
-//         Description for this technology.
-//   }
-//
-//   If no tracing technology is supported for the inferior, or no process is
-//   running, then an error message is returned.
-//
-// NOTE
-//  This packet is used by Trace plug-ins (see lldb_private::Trace.h) to
-//  do live tracing. Specifically, the name of the plug-in should match the name
-//  of the tracing technology returned by this packet.
-//----------------------------------------------------------------------
-
-send packet: jLLDBTraceSupported
-read packet: {"name":<name>, "description":<description>}/E<error code>;AAAAAAAAA
-
-//----------------------------------------------------------------------
-// jLLDBTraceStart
-//
-// BRIEF
-//  Start tracing a process or its threads using a provided tracing technology.
-//  The input and output are specified as JSON objects. In case of success, an OK
-//  response is returned, or an error otherwise.
-//
-// PROCESS TRACING
-//  This traces existing and future threads of the current process. An error is
-//  returned if the process is already being traced.
-//
-// THREAD TRACING
-//  This traces specific threads.
-//
-// INPUT SCHEMA
-//  {
-//    "type": <string>,
-//        Tracing technology name, e.g. intel-pt, arm-etm.
-//
-//    /* thread tracing only */
-//    "tids"?: [<decimal integer>],
-//        Individual threads to trace.
-//
-//    ... other parameters specific to the provided tracing type
-//  }
-//
-// NOTES
-//  - If "tids" is not provided, then the operation is "process tracing",
-//    otherwise it's "thread tracing".
-//  - Each tracing technology can have different levels of support for "thread
-//    tracing" and "process tracing".
-//
-// INTEL-PT
-//  intel-pt supports both "thread tracing" and "process tracing".
-//
-//  "Process tracing" is implemented in two different ways. If the
-//  "perCpuTracing" option is false, then each thread is traced individually
-//  but managed by the same "process trace" instance. This means that the
-//  amount of trace buffers used is proportional to the number of running
-//  threads. This is the recommended option unless the number of threads is
-//  huge. If "perCpuTracing" is true, then each cpu core is traced invidually
-//  instead of each thread, which uses a fixed number of trace buffers, but
-//  might result in less data available for less frequent threads. See
-//  "perCpuTracing" below for more information.
-//
-//  Each actual intel pt trace buffer, either from "process tracing" or "thread
-//  tracing", is stored in an in-memory circular buffer, which keeps the most
-//  recent data.
-//
-//  Additional params in the input schema:
-//   {
-//     "iptTraceSize": <decimal integer>,
-//         Size in bytes used by each individual per-thread or per-cpu trace
-//         buffer. It must be a power of 2 greater than or equal to 4096 (2^12)
-//         bytes.
-//
-//     "enableTsc": <boolean>,
-//         Whether to enable TSC timestamps or not. This is supported on
-//         all devices that support intel-pt. A TSC timestamp is generated along
-//         with PSB (synchronization) packets, whose frequency can be configured
-//         with the "psbPeriod" parameter.
-//
-//     "psbPeriod"?: <Optional decimal integer>,
-//         This value defines the period in which PSB packets will be generated.
-//         A PSB packet is a synchronization packet that contains a TSC
-//         timestamp and the current absolute instruction pointer.
-//
-//         This parameter can only be used if
-//
-//             /sys/bus/event_source/devices/intel_pt/caps/psb_cyc
-//
-//         is 1. Otherwise, the PSB period will be defined by the processor.
-//
-//         If supported, valid values for this period can be found in
-/
-//             /sys/bus/event_source/devices/intel_pt/caps/psb_periods
-//
-//         which contains a hexadecimal number, whose bits represent valid
-//         values e.g. if bit 2 is set, then value 2 is valid.
-//
-//         The psb_period value is converted to the approximate number of
-//         raw trace bytes between PSB packets as:
-//
-//             2 ^ (value + 11)
-//
-//          e.g. value 3 means 16KiB between PSB packets. Defaults to
-//          0 if supported.
-//
-//     /* process tracing only */
-//     "perCpuTracing": <boolean>
-//         Instead of having an individual trace buffer per thread, this option
-//         triggers the collection on a per cpu core basis. This effectively
-//         traces the entire activity on all cores. At decoding time, in order
-//         to correctly associate a decoded instruction with a thread, the
-//         context switch trace of each core is needed, as well as a record per
-//         cpu indicating which thread was running on each core when tracing
-//         started. These secondary traces are correlated with the intel-pt
-//         trace by comparing TSC timestamps.
-//
-//         This option forces the capture of TSC timestamps (see "enableTsc").
-//
-//         Note: This option can't be used simulatenously with any other trace
-//         sessions because of its system-wide nature.
-//
-//     /* process tracing only */
-//     "processBufferSizeLimit": <decimal integer>,
-//         Maximum total buffer size per process in bytes.
-//         This limit applies to the sum of the sizes of all thread or cpu core
-//         buffers for the current process, excluding the ones started with
-//         "thread tracing".
-//
-//         If "perCpuTracing" is false, whenever a thread is attempted to be
-//         traced due to "process tracing" and the limit would be reached, the
-//         process is stopped with a "tracing" reason along with a meaningful
-//         description, so that the user can retrace the process if needed.
-//
-//         If "perCpuTracing" is true, then starting the system-wide trace
-//         session fails if all the individual per-cpu trace buffers require
-//         in total more memory that the limit impossed by this parameter.
-//   }
-//
-//  Notes:
-//   - Modifying the parameters of an existing trace is not supported. The user
-//     needs to stop the trace and start a new one.
-//   - If "process tracing" is attempted and there are individual threads
-//     already being traced with "thread tracing", these traces are left
-//     unaffected and the threads not traced twice.
-//   - If "thread tracing" is attempted on a thread already being traced with
-//     either "thread tracing" or "process tracing", it fails.
-//----------------------------------------------------------------------
-
-Process tracing:
-send packet: jLLDBTraceStart:{"type":<type>,...other params}]
-read packet: OK/E<error code>;AAAAAAAAA
-
-Thread tracing:
-send packet: jLLDBTraceStart:{"type":<type>,"tids":<tids>,...other params}]
-read packet: OK/E<error code>;AAAAAAAAA
-
-//----------------------------------------------------------------------
-// jLLDBTraceStop
-//
-// BRIEF
-//  Stop tracing a process or its threads using a provided tracing technology.
-//  The input and output are specified as JSON objects. In case of success, an OK
-//  response is returned, or an error otherwise.
-//
-// PROCESS TRACE STOPPING
-//  Stopping a process trace stops the active traces initiated with
-//  "thread tracing".
-//
-// THREAD TRACE STOPPING
-//  This is a best effort request, which tries to stop as many traces as
-//  possible.
-//
-// INPUT SCHEMA
-//  The schema for the input is
-//
-//  {
-//    "type": <string>
-//       Tracing technology name, e.g. intel-pt, arm-etm.
-//
-//    /* thread trace stopping only */
-//    "tids":  [<decimal integer>]
-//       Individual thread traces to stop.
-//  }
-//
-// NOTES
-//  - If "tids" is not provided, then the operation is "process trace stopping".
-//
-// INTEL PT
-//  Stopping a specific thread trace started with "process tracing" is allowed.
-//----------------------------------------------------------------------
-
-Process trace stopping:
-send packet: jLLDBTraceStop:{"type":<type>}]
-read packet: OK/E<error code>;AAAAAAAAA
-
-Thread trace stopping:
-send packet: jLLDBTraceStop:{"type":<type>,"tids":<tids>}]
-read packet: OK/E<error code>;AAAAAAAAA
-
-//----------------------------------------------------------------------
-// jLLDBTraceGetState
-//
-// BRIEF
-//  Get the current state of the process and its threads being traced by
-//  a given trace technology. The response is a JSON object with custom
-//  information depending on the trace technology. In case of errors, an
-//  error message is returned.
-//
-// INPUT SCHEMA
-//  {
-//     "type": <string>
-//        Tracing technology name, e.g. intel-pt, arm-etm.
-//  }
-//
-// OUTPUT SCHEMA
-//  {
-//    "tracedThreads": [{
-//      "tid": <decimal integer>,
-//      "binaryData": [
-//        {
-//          "kind": <string>,
-//              Identifier for some binary data related to this thread to
-//              fetch with the jLLDBTraceGetBinaryData packet.
-//          "size": <decimal integer>,
-//              Size in bytes of this thread data.
-//        },
-//      ]
-//    }],
-//    "processBinaryData": [
-//      {
-//        "kind": <string>,
-//            Identifier for some binary data related to this process to
-//            fetch with the jLLDBTraceGetBinaryData packet.
-//        "size": <decimal integer>,
-//            Size in bytes of this thread data.
-//      },
-//    ],
-//    "cpus"?: [
-//      "id": <decimal integer>,
-//          Identifier for this CPU logical core.
-//      "binaryData": [
-//        {
-//          "kind": <string>,
-//              Identifier for some binary data related to this thread to
-//              fetch with the jLLDBTraceGetBinaryData packet.
-//          "size": <decimal integer>,
-//              Size in bytes of this cpu core data.
-//        },
-//      ]
-//    ],
-//    "warnings"?: [<string>],
-//        Non-fatal messages useful for troubleshooting.
-//
-//    ... other attributes specific to the given tracing technology
-//  }
-//
-// NOTES
-//   - "traceThreads" includes all thread traced by both "process tracing" and
-//     "thread tracing".
-//
-// INTEL PT
-//
-//  If per-cpu process tracing is enabled, "tracedThreads" will contain all
-//  the threads of the process without any trace buffers. Besides that, the
-//  "cpus" field will also be returned with per cpu core trace buffers.
-//  A side effect of per-cpu tracing is that all the threads of unrelated
-//  processes will also be traced, thus polluting the tracing data.
-//
-//  Binary data kinds:
-//    - iptTrace: trace buffer for a thread or a cpu.
-//    - perfContextSwitchTrace: context switch trace for a cpu generated by
-//                              perf_event_open.
-//    - procfsCpuInfo: contents of the /proc/cpuinfo file.
-//
-//  Additional attributes:
-//    tscPerfZeroConversion:
-//
-//    This field allows converting Intel processor's TSC values to nanoseconds.
-//    It is available through the Linux perf_event API when cap_user_time and cap_user_time_zero
-//    are set.
-//    See the documentation of time_zero in
-//    https://man7.org/linux/man-pages/man2/perf_event_open.2.html for more information about
-//    the calculation and the meaning of the values in the schema below.
-///
-//    Schema for this field:
-//
-//    "tscPerfZeroConversion": {
-//      "timeMult": <decimal integer>,
-//      "timeShift": <decimal integer>,
-//      "timeZero": <decimal integer>,
-//    }
-//----------------------------------------------------------------------
-
-send packet: jLLDBTraceGetState:{"type":<type>}]
-read packet: {...object}/E<error code>;AAAAAAAAA
-
-//----------------------------------------------------------------------
-// jLLDBTraceGetBinaryData
-//
-// BRIEF
-//  Get binary data given a trace technology and a data identifier.
-//  The input is specified as a JSON object and the response has the same format
-//  as the "binary memory read" (aka "x") packet. In case of failures, an error
-//  message is returned.
-//
-// SCHEMA
-//  The schema for the input is
-//
-//  {
-//   "type": <string>,
-//       Tracing technology name, e.g. intel-pt, arm-etm.
-//   "kind": <string>,
-//       Identifier for the data.
-//   "cpuId": <Optional decimal>,
-//       Core id in decimal if the data belongs to a CPU core.
-//   "tid"?: <Optional decimal>,
-//       Tid in decimal if the data belongs to a thread.
-//  }
-//----------------------------------------------------------------------
-
-send packet: jLLDBTraceGetBinaryData:{"type":<type>,"kind":<query>,"tid":<tid>,"offset":<offset>,"size":<size>}]
-read packet: <binary data>/E<error code>;AAAAAAAAA
-
-//----------------------------------------------------------------------
-// "qRegisterInfo<hex-reg-id>"
-//
-// BRIEF
-//  Discover register information from the remote GDB server.
-//
-// PRIORITY TO IMPLEMENT
-//  High. Any target that can self describe its registers, should do so.
-//  This means if new registers are ever added to a remote target, they
-//  will get picked up automatically, and allows registers to change
-//  depending on the actual CPU type that is used.
-//
-//  NB: qRegisterInfo is deprecated in favor of the standard gdb remote
-//  serial protocol register description method, 
-//  "qXfer:features:read:target.xml".
-//  If qXfer:features:read:target.xml is supported, qRegisterInfo does
-//  not need to be implemented.  The target.xml format is used by most
-//  gdb RSP stubs whereas qRegisterInfo was an lldb-only design.
-//  qRegisterInfo requires one packet per register and can have undesirable
-//  performance costs at the start of a debug session, whereas target.xml
-//  may be able to describe all registers in a single packet.
-//----------------------------------------------------------------------
-
-With LLDB, for register information, remote GDB servers can add
-support for the "qRegisterInfoN" packet where "N" is a zero based
-base 16 register number that must start at zero and increase by one
-for each register that is supported.  The response is done in typical
-GDB remote fashion where a series of "KEY:VALUE;" pairs are returned.
-An example for the x86_64 registers is included below:
-
-send packet: $qRegisterInfo0#00
-read packet: $name:rax;bitsize:64;offset:0;encoding:uint;format:hex;set:General Purpose Registers;gcc:0;dwarf:0;#00
-send packet: $qRegisterInfo1#00
-read packet: $name:rbx;bitsize:64;offset:8;encoding:uint;format:hex;set:General Purpose Registers;gcc:3;dwarf:3;#00
-send packet: $qRegisterInfo2#00
-read packet: $name:rcx;bitsize:64;offset:16;encoding:uint;format:hex;set:General Purpose Registers;gcc:2;dwarf:2;#00
-send packet: $qRegisterInfo3#00
-read packet: $name:rdx;bitsize:64;offset:24;encoding:uint;format:hex;set:General Purpose Registers;gcc:1;dwarf:1;#00
-send packet: $qRegisterInfo4#00
-read packet: $name:rdi;bitsize:64;offset:32;encoding:uint;format:hex;set:General Purpose Registers;gcc:5;dwarf:5;#00
-send packet: $qRegisterInfo5#00
-read packet: $name:rsi;bitsize:64;offset:40;encoding:uint;format:hex;set:General Purpose Registers;gcc:4;dwarf:4;#00
-send packet: $qRegisterInfo6#00
-read packet: $name:rbp;alt-name:fp;bitsize:64;offset:48;encoding:uint;format:hex;set:General Purpose Registers;gcc:6;dwarf:6;generic:fp;#00
-send packet: $qRegisterInfo7#00
-read packet: $name:rsp;alt-name:sp;bitsize:64;offset:56;encoding:uint;format:hex;set:General Purpose Registers;gcc:7;dwarf:7;generic:sp;#00
-send packet: $qRegisterInfo8#00
-read packet: $name:r8;bitsize:64;offset:64;encoding:uint;format:hex;set:General Purpose Registers;gcc:8;dwarf:8;#00
-send packet: $qRegisterInfo9#00
-read packet: $name:r9;bitsize:64;offset:72;encoding:uint;format:hex;set:General Purpose Registers;gcc:9;dwarf:9;#00
-send packet: $qRegisterInfoa#00
-read packet: $name:r10;bitsize:64;offset:80;encoding:uint;format:hex;set:General Purpose Registers;gcc:10;dwarf:10;#00
-send packet: $qRegisterInfob#00
-read packet: $name:r11;bitsize:64;offset:88;encoding:uint;format:hex;set:General Purpose Registers;gcc:11;dwarf:11;#00
-send packet: $qRegisterInfoc#00
-read packet: $name:r12;bitsize:64;offset:96;encoding:uint;format:hex;set:General Purpose Registers;gcc:12;dwarf:12;#00
-send packet: $qRegisterInfod#00
-read packet: $name:r13;bitsize:64;offset:104;encoding:uint;format:hex;set:General Purpose Registers;gcc:13;dwarf:13;#00
-send packet: $qRegisterInfoe#00
-read packet: $name:r14;bitsize:64;offset:112;encoding:uint;format:hex;set:General Purpose Registers;gcc:14;dwarf:14;#00
-send packet: $qRegisterInfof#00
-read packet: $name:r15;bitsize:64;offset:120;encoding:uint;format:hex;set:General Purpose Registers;gcc:15;dwarf:15;#00
-send packet: $qRegisterInfo10#00
-read packet: $name:rip;alt-name:pc;bitsize:64;offset:128;encoding:uint;format:hex;set:General Purpose Registers;gcc:16;dwarf:16;generic:pc;#00
-send packet: $qRegisterInfo11#00
-read packet: $name:rflags;alt-name:flags;bitsize:64;offset:136;encoding:uint;format:hex;set:General Purpose Registers;#00
-send packet: $qRegisterInfo12#00
-read packet: $name:cs;bitsize:64;offset:144;encoding:uint;format:hex;set:General Purpose Registers;#00
-send packet: $qRegisterInfo13#00
-read packet: $name:fs;bitsize:64;offset:152;encoding:uint;format:hex;set:General Purpose Registers;#00
-send packet: $qRegisterInfo14#00
-read packet: $name:gs;bitsize:64;offset:160;encoding:uint;format:hex;set:General Purpose Registers;#00
-send packet: $qRegisterInfo15#00
-read packet: $name:fctrl;bitsize:16;offset:176;encoding:uint;format:hex;set:Floating Point Registers;#00
-send packet: $qRegisterInfo16#00
-read packet: $name:fstat;bitsize:16;offset:178;encoding:uint;format:hex;set:Floating Point Registers;#00
-send packet: $qRegisterInfo17#00
-read packet: $name:ftag;bitsize:8;offset:180;encoding:uint;format:hex;set:Floating Point Registers;#00
-send packet: $qRegisterInfo18#00
-read packet: $name:fop;bitsize:16;offset:182;encoding:uint;format:hex;set:Floating Point Registers;#00
-send packet: $qRegisterInfo19#00
-read packet: $name:fioff;bitsize:32;offset:184;encoding:uint;format:hex;set:Floating Point Registers;#00
-send packet: $qRegisterInfo1a#00
-read packet: $name:fiseg;bitsize:16;offset:188;encoding:uint;format:hex;set:Floating Point Registers;#00
-send packet: $qRegisterInfo1b#00
-read packet: $name:fooff;bitsize:32;offset:192;encoding:uint;format:hex;set:Floating Point Registers;#00
-send packet: $qRegisterInfo1c#00
-read packet: $name:foseg;bitsize:16;offset:196;encoding:uint;format:hex;set:Floating Point Registers;#00
-send packet: $qRegisterInfo1d#00
-read packet: $name:mxcsr;bitsize:32;offset:200;encoding:uint;format:hex;set:Floating Point Registers;#00
-send packet: $qRegisterInfo1e#00
-read packet: $name:mxcsrmask;bitsize:32;offset:204;encoding:uint;format:hex;set:Floating Point Registers;#00
-send packet: $qRegisterInfo1f#00
-read packet: $name:stmm0;bitsize:80;offset:208;encoding:vector;format:vector-uint8;set:Floating Point Registers;gcc:33;dwarf:33;#00
-send packet: $qRegisterInfo20#00
-read packet: $name:stmm1;bitsize:80;offset:224;encoding:vector;format:vector-uint8;set:Floating Point Registers;gcc:34;dwarf:34;#00
-send packet: $qRegisterInfo21#00
-read packet: $name:stmm2;bitsize:80;offset:240;encoding:vector;format:vector-uint8;set:Floating Point Registers;gcc:35;dwarf:35;#00
-send packet: $qRegisterInfo22#00
-read packet: $name:stmm3;bitsize:80;offset:256;encoding:vector;format:vector-uint8;set:Floating Point Registers;gcc:36;dwarf:36;#00
-send packet: $qRegisterInfo23#00
-read packet: $name:stmm4;bitsize:80;offset:272;encoding:vector;format:vector-uint8;set:Floating Point Registers;gcc:37;dwarf:37;#00
-send packet: $qRegisterInfo24#00
-read packet: $name:stmm5;bitsize:80;offset:288;encoding:vector;format:vector-uint8;set:Floating Point Registers;gcc:38;dwarf:38;#00
-send packet: $qRegisterInfo25#00
-read packet: $name:stmm6;bitsize:80;offset:304;encoding:vector;format:vector-uint8;set:Floating Point Registers;gcc:39;dwarf:39;#00
-send packet: $qRegisterInfo26#00
-read packet: $name:stmm7;bitsize:80;offset:320;encoding:vector;format:vector-uint8;set:Floating Point Registers;gcc:40;dwarf:40;#00
-send packet: $qRegisterInfo27#00
-read packet: $name:xmm0;bitsize:128;offset:336;encoding:vector;format:vector-uint8;set:Floating Point Registers;gcc:17;dwarf:17;#00
-send packet: $qRegisterInfo28#00
-read packet: $name:xmm1;bitsize:128;offset:352;encoding:vector;format:vector-uint8;set:Floating Point Registers;gcc:18;dwarf:18;#00
-send packet: $qRegisterInfo29#00
-read packet: $name:xmm2;bitsize:128;offset:368;encoding:vector;format:vector-uint8;set:Floating Point Registers;gcc:19;dwarf:19;#00
-send packet: $qRegisterInfo2a#00
-read packet: $name:xmm3;bitsize:128;offset:384;encoding:vector;format:vector-uint8;set:Floating Point Registers;gcc:20;dwarf:20;#00
-send packet: $qRegisterInfo2b#00
-read packet: $name:xmm4;bitsize:128;offset:400;encoding:vector;format:vector-uint8;set:Floating Point Registers;gcc:21;dwarf:21;#00
-send packet: $qRegisterInfo2c#00
-read packet: $name:xmm5;bitsize:128;offset:416;encoding:vector;format:vector-uint8;set:Floating Point Registers;gcc:22;dwarf:22;#00
-send packet: $qRegisterInfo2d#00
-read packet: $name:xmm6;bitsize:128;offset:432;encoding:vector;format:vector-uint8;set:Floating Point Registers;gcc:23;dwarf:23;#00
-send packet: $qRegisterInfo2e#00
-read packet: $name:xmm7;bitsize:128;offset:448;encoding:vector;format:vector-uint8;set:Floating Point Registers;gcc:24;dwarf:24;#00
-send packet: $qRegisterInfo2f#00
-read packet: $name:xmm8;bitsize:128;offset:464;encoding:vector;format:vector-uint8;set:Floating Point Registers;gcc:25;dwarf:25;#00
-send packet: $qRegisterInfo30#00
-read packet: $name:xmm9;bitsize:128;offset:480;encoding:vector;format:vector-uint8;set:Floating Point Registers;gcc:26;dwarf:26;#00
-send packet: $qRegisterInfo31#00
-read packet: $name:xmm10;bitsize:128;offset:496;encoding:vector;format:vector-uint8;set:Floating Point Registers;gcc:27;dwarf:27;#00
-send packet: $qRegisterInfo32#00
-read packet: $name:xmm11;bitsize:128;offset:512;encoding:vector;format:vector-uint8;set:Floating Point Registers;gcc:28;dwarf:28;#00
-send packet: $qRegisterInfo33#00
-read packet: $name:xmm12;bitsize:128;offset:528;encoding:vector;format:vector-uint8;set:Floating Point Registers;gcc:29;dwarf:29;#00
-send packet: $qRegisterInfo34#00
-read packet: $name:xmm13;bitsize:128;offset:544;encoding:vector;format:vector-uint8;set:Floating Point Registers;gcc:30;dwarf:30;#00
-send packet: $qRegisterInfo35#00
-read packet: $name:xmm14;bitsize:128;offset:560;encoding:vector;format:vector-uint8;set:Floating Point Registers;gcc:31;dwarf:31;#00
-send packet: $qRegisterInfo36#00
-read packet: $name:xmm15;bitsize:128;offset:576;encoding:vector;format:vector-uint8;set:Floating Point Registers;gcc:32;dwarf:32;#00
-send packet: $qRegisterInfo37#00
-read packet: $name:trapno;bitsize:32;offset:696;encoding:uint;format:hex;set:Exception State Registers;#00
-send packet: $qRegisterInfo38#00
-read packet: $name:err;bitsize:32;offset:700;encoding:uint;format:hex;set:Exception State Registers;#00
-send packet: $qRegisterInfo39#00
-read packet: $name:faultvaddr;bitsize:64;offset:704;encoding:uint;format:hex;set:Exception State Registers;#00
-send packet: $qRegisterInfo3a#00
-read packet: $E45#00
-
-As we see above we keep making subsequent calls to the remote server to
-discover all registers by increasing the number appended to qRegisterInfo and
-we get a response back that is a series of "key=value;" strings.
-
-The offset: fields should not leave a gap anywhere in the g/G packet -- the
-register values should be appended one after another.  For instance, if the
-register context for a thread looks like
-
-struct rctx {
-    uint32_t gpr1;  // offset 0
-    uint32_t gpr2;  // offset 4
-    uint32_t gpr3;  // offset 8
-    uint64_t fp1;   // offset 16
-};
-
-You may end up with a 4-byte gap between gpr3 and fp1 on architectures
-that align values like this.  The correct offset: value for fp1 is 12 -
-in the g/G packet fp1 will immediately follow gpr3, even though the
-in-memory thread structure has an empty 4 bytes for alignment between
-these two registers.
-
-The keys and values are detailed below:
-
-Key         Value
-==========  ================================================================
-name        The primary register name as a string ("rbp" for example)
-
-alt-name    An alternate name for a register as a string ("fp" for example for
-            the above "rbp")
-
-bitsize     Size in bits of a register (32, 64, etc).  Base 10.
-
-offset      The offset within the "g" and "G" packet of the register data for
-            this register.  This is the byte offset once the data has been
-            transformed into binary, not the character offset into the g/G
-            packet.  Base 10.
-
-encoding    The encoding type of the register which must be one of:
-
-                 uint (unsigned integer)
-                 sint (signed integer)
-                 ieee754 (IEEE 754 float)
-                 vector (vector register)
-
-format      The preferred format for display of this register. The value must
-            be one of:
-
-                binary
-                decimal
-                hex
-                float
-                vector-sint8
-                vector-uint8
-                vector-sint16
-                vector-uint16
-                vector-sint32
-                vector-uint32
-                vector-float32
-                vector-uint128
-
-set         The register set name as a string that this register belongs to.
-
-gcc         The GCC compiler registers number for this register (used for
-            EH frame and other compiler information that is encoded in the
-            executable files). The supplied number will be decoded like a
-            string passed to strtoul() with a base of zero, so the number
-            can be decimal, or hex if it is prefixed with "0x".
-
-            NOTE: If the compiler doesn't have a register number for this
-            register, this key/value pair should be omitted.
-
-dwarf       The DWARF register number for this register that is used for this
-            register in the debug information. The supplied number will be decoded
-            like a string passed to strtoul() with a base of zero, so the number
-            can be decimal, or hex if it is prefixed with "0x".
-
-            NOTE: If the compiler doesn't have a register number for this
-            register, this key/value pair should be omitted.
-
-generic     If the register is a generic register that most CPUs have, classify
-            it correctly so the debugger knows. Valid values are one of:
-             pc  (a program counter register. for example "name=eip;" (i386),
-                  "name=rip;" (x86_64), "name=r15;" (32 bit arm) would
-                  include a "generic=pc;" key value pair)
-             sp  (a stack pointer register. for example "name=esp;" (i386),
-                  "name=rsp;" (x86_64), "name=r13;" (32 bit arm) would
-                  include a "generic=sp;" key value pair)
-             fp  (a frame pointer register. for example "name=ebp;" (i386),
-                   "name=rbp;" (x86_64), "name=r7;" (32 bit arm with macosx
-                   ABI) would include a "generic=fp;" key value pair)
-             ra  (a return address register. for example "name=lr;" (32 bit ARM)
-                  would include a "generic=ra;" key value pair)
-             flags  (a CPU flags register. for example "name=eflags;" (i386),
-                  "name=rflags;" (x86_64), "name=cpsr;" (32 bit ARM)
-                  would include a "generic=flags;" key value pair)
-             arg1 - arg8 (specified for registers that contain function
-                      arguments when the argument fits into a register)
-
-container-regs
-            The value for this key is a comma separated list of raw hex (optional
-            leading "0x") register numbers.
-
-            This specifies that this register is contained in other concrete
-            register values. For example "eax" is in the lower 32 bits of the
-            "rax" register value for x86_64, so "eax" could specify that it is
-            contained in "rax" by specifying the register number for "rax" (whose
-            register number is 0x00)
-
-            "container-regs:00;"
-
-            If a register is comprised of one or more registers, like "d0" is ARM
-            which is a 64 bit register, it might be made up of "s0" and "s1". If
-            the register number for "s0" is 0x20, and the register number of "s1"
-            is "0x21", the "container-regs" key/value pair would be:
-
-            "container-regs:20,21;"
-
-            This is handy for defining what GDB used to call "pseudo" registers.
-            These registers are never requested by LLDB via the register read
-            or write packets, the container registers will be requested on behalf
-            of this register.
-
-invalidate-regs
-            The value for this key is a comma separated list of raw hex (optional
-            leading "0x") register numbers.
-
-            This specifies which register values should be invalidated when this
-            register is modified. For example if modifying "eax" would cause "rax",
-            "eax", "ax", "ah", and "al" to be modified where rax is 0x0, eax is 0x15,
-            ax is 0x25, ah is 0x35, and al is 0x39, the "invalidate-regs" key/value
-            pair would be:
-
-            "invalidate-regs:0,15,25,35,39;"
-
-            If there is a single register that gets invalidated, then omit the comma
-            and just list a single register:
-
-            "invalidate-regs:0;"
-
-            This is handy when modifying a specific register can cause other
-            register values to change. For example, when debugging an ARM target,
-            modifying the CPSR register can cause the r8 - r14 and cpsr value to
-            change depending on if the mode has changed.
-
-//----------------------------------------------------------------------
-// "qPlatform_shell"
-//
-// BRIEF
-//  Run a command in a shell on the connected remote machine.
-//
-// PRIORITY TO IMPLEMENT
-//  High. This command allows LLDB clients to run arbitrary shell
-//  commands on a remote host.
-//
-/----------------------------------------------------------------------
-
-The request consists of the command to be executed encoded in ASCII characters
-converted into hex bytes.
-
-The response to this packet consists of the letter F followed by the return code,
-followed by the signal number (or 0 if no signal was delivered), and escaped bytes
-of captured program output.
-
-Below is an example communication from a client sending an "ls -la" command:
-
-send packet: $qPlatform_shell:6c73202d6c61,00000002#ec
-read packet: $F,00000000,00000000,total 4736
-drwxrwxr-x 16 username groupname    4096 Aug 15 21:36 .
-drwxr-xr-x 17 username groupname    4096 Aug 10 16:39 ..
--rw-rw-r--  1 username groupname   73875 Aug 12 16:46 notes.txt
-drwxrwxr-x  5 username groupname    4096 Aug 15 21:36 source.cpp
--rw-r--r--  1 username groupname    2792 Aug 12 16:46 a.out
--rw-r--r--  1 username groupname    3190 Aug 12 16:46 Makefile
-
-//----------------------------------------------------------------------
-// "qPlatform_mkdir"
-//
-// BRIEF
-//  Creates a new directory on the connected remote machine.
-//
-// PRIORITY TO IMPLEMENT
-//  Low. This command allows LLDB clients to create new directories on
-//  a remote host.
-//
-/----------------------------------------------------------------------
-
-Request:
-    qPlatform_mkdir:<hex-file-mode>,<ascii-hex-path>
-
-Reply:
-    F<mkdir-return-code>
-        mkdir called successfully and returned with the given return code
-    Exx
-        An error occurred
-
-//----------------------------------------------------------------------
-// "qPlatform_chmod"
-//
-// BRIEF
-//  Change the permissions of a file on the connected remote machine.
-//
-// PRIORITY TO IMPLEMENT
-//  Low. This command allows LLDB clients to change the permissions of
-//  a file on the remote host.
-//
-/----------------------------------------------------------------------
-
-Request:
-    qPlatform_chmod:<hex-file-mode>,<ascii-hex-path>
-
-Reply:
-    F<chmod-return-code>
-        chmod called successfully and returned with the given return code
-    Exx
-        An error occurred
-
-//----------------------------------------------------------------------
-// "qHostInfo"
-//
-// BRIEF
-//  Get information about the host we are remotely connected to.
-//
-// PRIORITY TO IMPLEMENT
-//  High. This packet is usually very easy to implement and can help
-//  LLDB select the correct plug-ins for the job based on the target
-//  triple information that is supplied.
-//----------------------------------------------------------------------
-
-LLDB supports a host info call that gets all sorts of details of the system
-that is being debugged:
-
-send packet: $qHostInfo#00
-read packet: $cputype:16777223;cpusubtype:3;ostype:darwin;vendor:apple;endian:little;ptrsize:8;#00
-
-Key value pairs are one of:
-
-cputype: is a number that is the mach-o CPU type that is being debugged (base 10)
-cpusubtype: is a number that is the mach-o CPU subtype type that is being debugged (base 10)
-triple: a string for the target triple (x86_64-apple-macosx) that can be used to specify arch + vendor + os in one entry
-vendor: a string for the vendor (apple), not needed if "triple" is specified
-ostype: a string for the OS being debugged (macosx, linux, freebsd, ios, watchos), not needed if "triple" is specified
-endian: is one of "little", "big", or "pdp"
-ptrsize: an unsigned number that represents how big pointers are in bytes on the debug target
-hostname: the hostname of the host that is running the GDB server if available
-os_build: a string for the OS build for the remote host as a string value
-os_kernel: a string describing the kernel version
-os_version: a version string that represents the current OS version (10.8.2)
-watchpoint_exceptions_received: one of "before" or "after" to specify if a watchpoint is triggered before or after the pc when it stops
-default_packet_timeout: an unsigned number that specifies the default timeout in seconds
-distribution_id: optional. For linux, specifies distribution id (e.g. ubuntu, fedora, etc.)
-osmajor: optional, specifies the major version number of the OS (e.g. for macOS 10.12.2, it would be 10)
-osminor: optional, specifies the minor version number of the OS (e.g. for macOS 10.12.2, it would be 12)
-ospatch: optional, specifies the patch level number of the OS (e.g. for macOS 10.12.2, it would be 2)
-vm-page-size: optional, specifies the target system VM page size, base 10.
-           Needed for the "dirty-pages:" list in the qMemoryRegionInfo
-           packet, where a list of dirty pages is sent from the remote
-           stub.  This page size tells lldb how large each dirty page is.
-addressing_bits: optional, specifies how many bits in addresses are
-		 significant for addressing, base 10.  If bits 38..0
-		 in a 64-bit pointer are significant for addressing,
-		 then the value is 39.  This is needed on e.g. AArch64
-		 v8.3 ABIs that use pointer authentication, so lldb
-		 knows which bits to clear/set to get the actual
-		 addresses.
-low_mem_addressing_bits: optional, specifies how many bits in 
-     addresses in low memory are significant for addressing, base 10.  
-     AArch64 can have different page table setups for low and high
-     memory, and therefore a different number of bits used for addressing.
-high_mem_addressing_bits: optional, specifies how many bits in 
-     addresses in high memory are significant for addressing, base 10.  
-     AArch64 can have different page table setups for low and high
-     memory, and therefore a different number of bits used for addressing.
-
-//----------------------------------------------------------------------
-// "qGDBServerVersion"
-//
-// BRIEF
-//  Get version information about this implementation of the gdb-remote
-//  protocol.
-//
-// PRIORITY TO IMPLEMENT
-//  High. This packet is usually very easy to implement and can help
-//  LLDB to work around bugs in a server's implementation when they
-//  are found.
-//----------------------------------------------------------------------
-
-The goal of this packet is to provide enough information about an
-implementation of the gdb-remote-protocol server that lldb can
-work around implementation problems that are discovered after the
-version has been released/deployed.  The name and version number
-should be sufficiently unique that lldb can unambiguously identify
-the origin of the program (for instance, debugserver from lldb) and
-the version/submission number/patch level of the program - whatever
-is appropriate for your server implementation.
-
-The packet follows the key-value pair model, semicolon separated.
-
-send packet: $qGDBServerVersion#00
-read packet: $name:debugserver;version:310.2;#00
-
-Other clients may find other key-value pairs to be useful for identifying
-a gdb stub.  Patch level, release name, build number may all be keys that
-better describe your implementation's version.
-Suggested key names:
-
-  name   : the name of your remote server - "debugserver" is the lldb standard
-           implementation
-
-  version : identifies the version number of this server
-
-  patch_level : the patch level of this server
-
-  release_name : the name of this release, if your project uses names
-
-  build_number : if you use a build system with increasing build numbers,
-                 this may be the right key name for your server
-
-  major_version : major version number
-  minor_version : minor version number
-
-//----------------------------------------------------------------------
-// "qProcessInfo"
-//
-// BRIEF
-//  Get information about the process we are currently debugging.
-//
-// PRIORITY TO IMPLEMENT
-//  Medium.  On systems which can launch multiple different architecture processes,
-//  the qHostInfo may not disambiguate sufficiently to know what kind of
-//  process is being debugged.
-//  e.g. on a 64-bit x86 Mac system both 32-bit and 64-bit user processes are possible,
-//  and with Mach-O universal files, the executable file may contain both 32- and
-//  64-bit slices so it may be impossible to know until you're attached to a real
-//  process to know what you're working with.
-//
-//  All numeric fields return base 16 numbers without any "0x" prefix.
-//----------------------------------------------------------------------
-
-An i386 process:
-
-send packet: $qProcessInfo#00
-read packet: $pid:42a8;parent-pid:42bf;real-uid:ecf;real-gid:b;effective-uid:ecf;effective-gid:b;cputype:7;cpusubtype:3;ostype:macosx;vendor:apple;endian:little;ptrsize:4;#00
-
-An x86_64 process:
-
-send packet: $qProcessInfo#00
-read packet: $pid:d22c;parent-pid:d34d;real-uid:ecf;real-gid:b;effective-uid:ecf;effective-gid:b;cputype:1000007;cpusubtype:3;ostype:macosx;vendor:apple;endian:little;ptrsize:8;#00
-
-Key value pairs include:
-
-pid: the process id
-parent-pid: the process of the parent process (often debugserver will become the parent when attaching)
-real-uid: the real user id of the process
-real-gid: the real group id of the process
-effective-uid: the effective user id of the process
-effective-gid: the effective group id of the process
-cputype: the Mach-O CPU type of the process  (base 16)
-cpusubtype: the Mach-O CPU subtype of the process  (base 16)
-ostype: is a string the represents the OS being debugged (darwin, linux, freebsd)
-vendor: is a string that represents the vendor (apple)
-endian: is one of "little", "big", or "pdp"
-ptrsize: is a number that represents how big pointers are in bytes
-
-main-binary-uuid: is the UUID of a firmware type binary that the gdb stub knows about
-main-binary-address: is the load address of the firmware type binary
-main-binary-slide: is the slide of the firmware type binary, if address isn't known
-
-binary-addresses: A comma-separated list of binary load addresses base 16.  
-                  lldb will parse the binaries in memory to get UUIDs, then
-                  try to find the binaries & debug info by UUID.  Intended for
-                  use with a small number of firmware type binaries where the 
-                  search for binary/debug info may be expensive.
-
-//----------------------------------------------------------------------
-// "qShlibInfoAddr"
-//
-// BRIEF
-//  Get an address where the dynamic linker stores information about
-//  where shared libraries are loaded.
-//
-// PRIORITY TO IMPLEMENT
-//  High if you have a dynamic loader plug-in in LLDB for your target
-//  triple (see the "qHostInfo" packet) that can use this information.
-//  Many times address load randomization can make it hard to detect
-//  where the dynamic loader binary and data structures are located and
-//  some platforms know, or can find out where this information is.
-//
-//  Low if you have a debug target where all object and symbol files
-//  contain static load addresses.
-//----------------------------------------------------------------------
-
-LLDB and GDB both support the "qShlibInfoAddr" packet which is a hint to each
-debugger as to where to find the dynamic loader information. For darwin
-binaries that run in user land this is the address of the "all_image_infos"
-structure in the "/usr/lib/dyld" executable, or the result of a TASK_DYLD_INFO
-call. The result is returned as big endian hex bytes that are the address
-value:
-
-send packet: $qShlibInfoAddr#00
-read packet: $7fff5fc40040#00
-
-
-
-//----------------------------------------------------------------------
-// "qThreadStopInfo<tid>"
-//
-// BRIEF
-//  Get information about why a thread, whose ID is "<tid>", is stopped.
-//
-// PRIORITY TO IMPLEMENT
-//  High if you need to support multi-threaded or multi-core debugging.
-//  Many times one thread will hit a breakpoint and while the debugger
-//  is in the process of suspending the other threads, other threads
-//  will also hit a breakpoint. This packet allows LLDB to know why all
-//  threads (live system debug) / cores (JTAG) in your program have
-//  stopped and allows LLDB to display and control your program
-//  correctly.
-//----------------------------------------------------------------------
-
-LLDB tries to use the "qThreadStopInfo" packet which is formatted as
-"qThreadStopInfo%x" where %x is the hex thread ID. This requests information
-about why a thread is stopped. The response is the same as the stop reply
-packets and tells us what happened to the other threads. The standard GDB
-remote packets love to think that there is only _one_ reason that _one_ thread
-stops at a time. This allows us to see why all threads stopped and allows us
-to implement better multi-threaded debugging support.
-
-//----------------------------------------------------------------------
-// "QThreadSuffixSupported"
-//
-// BRIEF
-//  Try to enable thread suffix support for the 'g', 'G', 'p', and 'P'
-//  packets.
-//
-// PRIORITY TO IMPLEMENT
-//  High. Adding a thread suffix allows us to read and write registers
-//  more efficiently and stops us from having to select a thread with
-//  one packet and then read registers with a second packet. It also
-//  makes sure that no errors can occur where the debugger thinks it
-//  already has a thread selected (see the "Hg" packet from the standard
-//  GDB remote protocol documentation) yet the remote GDB server actually
-//  has another thread selected.
-//----------------------------------------------------------------------
-
-When reading thread registers, you currently need to set the current
-thread, then read the registers. This is kind of cumbersome, so we added the
-ability to query if the remote GDB server supports adding a "thread:<tid>;"
-suffix to all packets that request information for a thread. To test if the
-remote GDB server supports this feature:
-
-send packet: $QThreadSuffixSupported#00
-read packet: OK
-
-If "OK" is returned, then the 'g', 'G', 'p' and 'P' packets can accept a
-thread suffix. So to send a 'g' packet (read all register values):
-
-send packet: $g;thread:<tid>;#00
-read packet: ....
-
-send packet: $G;thread:<tid>;#00
-read packet: ....
-
-send packet: $p1a;thread:<tid>;#00
-read packet: ....
-
-send packet: $P1a=1234abcd;thread:<tid>;#00
-read packet: ....
-
-
-otherwise, without this you would need to always send two packets:
-
-send packet: $Hg<tid>#00
-read packet: ....
-send packet: $g#00
-read packet: ....
-
-We also added support for allocating and deallocating memory. We use this to
-allocate memory so we can run JITed code.
-
-//----------------------------------------------------------------------
-// "_M<size>,<permissions>"
-//
-// BRIEF
-//  Allocate memory on the remote target with the specified size and
-//  permissions.
-//
-// PRIORITY TO IMPLEMENT
-//  High if you want LLDB to be able to JIT code and run that code. JIT
-//  code also needs data which is also allocated and tracked.
-//
-//  Low if you don't support running JIT'ed code.
-//----------------------------------------------------------------------
-
-The allocate memory packet starts with "_M<size>,<permissions>". It returns a
-raw big endian address value, or "" for unimplemented, or "EXX" for an error
-code. The packet is formatted as:
-
-char packet[256];
-int packet_len;
-packet_len = ::snprintf (
-    packet,
-    sizeof(packet),
-    "_M%zx,%s%s%s",
-    (size_t)size,
-    permissions & lldb::ePermissionsReadable ? "r" : "",
-    permissions & lldb::ePermissionsWritable ? "w" : "",
-    permissions & lldb::ePermissionsExecutable ? "x" : "");
-
-You request a size and give the permissions. This packet does NOT need to be
-implemented if you don't want to support running JITed code. The return value
-is just the address of the newly allocated memory as raw big endian hex bytes.
-
-//----------------------------------------------------------------------
-// "_m<addr>"
-//
-// BRIEF
-//  Deallocate memory that was previously allocated using an allocate
-//  memory pack.
-//
-// PRIORITY TO IMPLEMENT
-//  High if you want LLDB to be able to JIT code and run that code. JIT
-//  code also needs data which is also allocated and tracked.
-//
-//  Low if you don't support running JIT'ed code.
-//----------------------------------------------------------------------
-
-The deallocate memory packet is "_m<addr>" where you pass in the address you
-got back from a previous call to the allocate memory packet. It returns "OK"
-if the memory was successfully deallocated, or "EXX" for an error, or "" if
-not supported.
-
-//----------------------------------------------------------------------
-// "qMemoryRegionInfo:<addr>"
-//
-// BRIEF
-//  Get information about the address range that contains "<addr>"
-//
-// PRIORITY TO IMPLEMENT
-//  Medium. This is nice to have, but it isn't necessary. It helps LLDB
-//  do stack unwinding when we branch into memory that isn't executable.
-//  If we can detect that the code we are stopped in isn't executable,
-//  then we can recover registers for stack frames above the current
-//  frame. Otherwise we must assume we are in some JIT'ed code (not JIT
-//  code that LLDB has made) and assume that no registers are available
-//  in higher stack frames.
-//----------------------------------------------------------------------
-
-We added a way to get information for a memory region. The packet is:
-
-    qMemoryRegionInfo:<addr>
-
-Where <addr> is a big endian hex address. The response is returned in a series
-of tuples like the data returned in a stop reply packet. The currently valid
-tuples to return are:
-
-    start:<start-addr>; // <start-addr> is a big endian hex address that is
-                        // the start address of the range that contains <addr>
-
-    size:<size>;    // <size> is a big endian hex byte size of the address
-                    // of the range that contains <addr>
-
-    permissions:<permissions>;  // <permissions> is a string that contains one
-                                // or more of the characters from "rwx"
-
-    name:<name>; // <name> is a hex encoded string that contains the name of
-                 // the memory region mapped at the given address. In case of
-                 // regions backed by a file it have to be the absolute path of
-                 // the file while for anonymous regions it have to be the name
-                 // associated to the region if that is available.
-
-    flags:<flags-string>; // where <flags-string> is a space separated string
-                          // of flag names. Currently the only supported flag
-                          // is "mt" for AArch64 memory tagging. lldb will
-                          // ignore any other flags in this field.
-
-    type:[<type>][,<type>]; // memory types that apply to this region, e.g.
-                 // "stack" for stack memory.
-
-    error:<ascii-byte-error-string>; // where <ascii-byte-error-string> is
-                                     // a hex encoded string value that
-                                     // contains an error string
-
-    dirty-pages:[<hexaddr>][,<hexaddr]; // A list of memory pages within this
-                 // region that are "dirty" -- they have been modified.
-                 // Page addresses are in base 16.  The size of a page can
-                 // be found from the qHostInfo's page-size key-value.
-                 //
-                 // If the stub supports identifying dirty pages within a
-                 // memory region, this key should always be present for all
-                 // qMemoryRegionInfo replies.  This key with no pages
-                 // listed ("dirty-pages:;") indicates no dirty pages in
-                 // this memory region.  The *absence* of this key means
-                 // that this stub cannot determine dirty pages.
-
-If the address requested is not in a mapped region (e.g. we've jumped through
-a NULL pointer and are at 0x0) currently lldb expects to get back the size
-of the unmapped region -- that is, the distance to the next valid region.
-For instance, with a macOS process which has nothing mapped in the first
-4GB of its address space, if we're asking about address 0x2,
-
-  qMemoryRegionInfo:2
-  start:2;size:fffffffe;
-
-The lack of 'permissions:' indicates that none of read/write/execute are valid
-for this region.
-
-//----------------------------------------------------------------------
-// "x" - Binary memory read
-//
-// Like the 'm' (read) and 'M' (write) packets, this is a partner to the
-// 'X' (write binary data) packet, 'x'.
-//
-// It is called like
-//
-// xADDRESS,LENGTH
-//
-// where both ADDRESS and LENGTH are big-endian base 16 values.
-//
-// To test if this packet is available, send a addr/len of 0:
-//
-// x0,0
-//
-// and you will get an "OK" response.
-//
-// The reply will be the data requested in 8-bit binary data format.
-// The standard quoting is applied to the payload -- characters
-//   }  #  $  *
-// will all be escaped with '}' (0x7d) character and then XOR'ed with 0x20.
-//
-// A typical use to read 512 bytes at 0x1000 would look like
-//
-// x0x1000,0x200
-//
-// The "0x" prefixes are optional - like most of the gdb-remote packets,
-// omitting them will work fine; these numbers are always base 16.
-//
-// The length of the payload is not provided.  A reliable, 8-bit clean,
-// transport layer is assumed.
-//----------------------------------------------------------------------
-
-//----------------------------------------------------------------------
-// Detach and stay stopped:
-//
-// We extended the "D" packet to specify that the monitor should keep the
-// target suspended on detach.  The normal behavior is to resume execution
-// on detach.  We will send:
-//
-//  qSupportsDetachAndStayStopped:
-//
-// to query whether the monitor supports the extended detach, and if it does,
-// when we want the monitor to detach but not resume the target, we will
-// send:
-//
-//   D1
-//
-// In any case, if we want the normal detach behavior we will just send:
-//
-//   D
-//----------------------------------------------------------------------
-
-//----------------------------------------------------------------------
-// QSaveRegisterState
-// QSaveRegisterState;thread:XXXX;
-//
-// BRIEF
-//  The QSaveRegisterState packet tells the remote debugserver to save
-//  all registers and return a non-zero unique integer ID that
-//  represents these save registers. If thread suffixes are enabled the
-//  second form of this packet is used, otherwise the first form is
-//  used. This packet is called prior to executing an expression, so
-//  the remote GDB server should do anything it needs to in order to
-//  ensure the registers that are saved are correct. On macOS this
-//  involves calling "thread_abort_safely(mach_port_t thread)" to
-//  ensure we get the correct registers for a thread in case it is
-//  currently having code run on its behalf in the kernel.
-//
-// RESPONSE
-//  unsigned - The save_id result is a non-zero unsigned integer value
-//             that can be passed back to the GDB server using a
-//             QRestoreRegisterState packet to restore the registers
-//             one time.
-//  "EXX" - or an error code in the form of EXX where XX is a
-//  hex error code.
-//
-// PRIORITY TO IMPLEMENT
-//  Low, this is mostly a convenience packet to avoid having to send all
-//  registers via a g packet. It should only be implemented if support
-//  for the QRestoreRegisterState is added.
-//----------------------------------------------------------------------
-
-//----------------------------------------------------------------------
-// QRestoreRegisterState:<save_id>
-// QRestoreRegisterState:<save_id>;thread:XXXX;
-//
-// BRIEF
-//  The QRestoreRegisterState packet tells the remote debugserver to
-//  restore all registers using the "save_id" which is an unsigned
-//  integer that was returned from a previous call to
-//  QSaveRegisterState. The restoration process can only be done once
-//  as the data backing the register state will be freed upon the
-//  completion of the QRestoreRegisterState command.
-//
-//  If thread suffixes are enabled the second form of this packet is
-//  used, otherwise the first form is used.
-//
-// RESPONSE
-//  "OK" - if all registers were successfully restored
-//  "EXX" - for any errors
-//
-// PRIORITY TO IMPLEMENT
-//  Low, this is mostly a convenience packet to avoid having to send all
-//  registers via a g packet. It should only be implemented if support
-//  for the QSaveRegisterState is added.
-//----------------------------------------------------------------------
-
-//----------------------------------------------------------------------
-// qFileLoadAddress:<file_path>
-//
-// BRIEF
-//  Get the load address of a memory mapped file.
-//  The load address is defined as the address of the first memory
-//  region what contains data mapped from the specified file.
-//
-// RESPONSE
-//  <unsigned-hex64> - Load address of the file in big endian encoding
-//  "E01" - the requested file isn't loaded
-//  "EXX" - for any other errors
-//
-// PRIORITY TO IMPLEMENT
-//  Low, required if dynamic linker don't fill in the load address of
-//  some object file in the rendezvous data structure.
-//----------------------------------------------------------------------
-
-//----------------------------------------------------------------------
-// qModuleInfo:<module_path>;<arch triple>
-//
-// BRIEF
-//  Get information for a module by given module path and architecture.
-//
-// RESPONSE
-//  "(uuid|md5):...;triple:...;file_offset:...;file_size...;"
-//  "EXX" - for any errors
-//
-// PRIORITY TO IMPLEMENT
-//  Optional, required if dynamic loader cannot fetch module's information like
-//  UUID directly from inferior's memory.
-//----------------------------------------------------------------------
-
-//----------------------------------------------------------------------
-// jModulesInfo:[{"file":"...",triple:"..."}, ...]
-//
-// BRIEF
-//  Get information for a list of modules by given module path and
-//  architecture.
-//
-// RESPONSE
-//  A JSON array of dictionaries containing the following keys: uuid,
-//  triple, file_path, file_offset, file_size. The meaning of the fields
-//  is the same as in the qModuleInfo packet. The server signals the
-//  failure to retrieve the module info for a file by ommiting the
-//  corresponding array entry from the response. The server may also
-//  include entries the client did not ask for, if it has reason to
-//  the modules will be interesting to the client.
-//
-// PRIORITY TO IMPLEMENT
-//  Optional. If not implemented, qModuleInfo packet will be used, which
-//  may be slower if the target contains a large number of modules and
-//  the communication link has a non-negligible latency.
-//----------------------------------------------------------------------
-
-//----------------------------------------------------------------------
-// Stop reply packet extensions
-//
-// BRIEF
-//  This section describes some of the additional information you can
-//  specify in stop reply packets that help LLDB to know more detailed
-//  information about your threads.
-//
-// DESCRIPTION
-//  Standard GDB remote stop reply packets are reply packets sent in
-//  response to a packet  that made the program run. They come in the
-//  following forms:
-//
-//  "SAA"
-//  "S" means signal and "AA" is a hex signal number that describes why
-//  the thread or stopped. It doesn't specify which thread, so the "T"
-//  packet is recommended to use instead of the "S" packet.
-//
-//  "TAAkey1:value1;key2:value2;..."
-//  "T" means a thread stopped due to a unix signal where "AA" is a hex
-//  signal number that describes why the program stopped. This is
-//  followed by a series of key/value pairs:
-//      - If key is a hex number, it is a register number and value is
-//        the hex value of the register in debuggee endian byte order.
-//      - If key == "thread", then the value is the big endian hex
-//        thread-id of the stopped thread.
-//      - If key == "core", then value is a hex number of the core on
-//        which the stop was detected.
-//      - If key == "watch" or key == "rwatch" or key == "awatch", then
-//        value is the data address in big endian hex
-//      - If key == "library", then value is ignore and "qXfer:libraries:read"
-//        packets should be used to detect any newly loaded shared libraries
-//
-//  "WAA"
-//  "W" means the process exited and "AA" is the exit status.
-//
-//  "XAA"
-//  "X" means the process exited and "AA" is signal that caused the program
-//  to exit.
-//
-//  "O<ascii-hex-string>"
-//  "O" means STDOUT has data that was written to its console and is
-//  being delivered to the debugger. This packet happens asynchronously
-//  and the debugger is expected to continue to wait for another stop reply
-//  packet.
-//
-// LLDB EXTENSIONS
-//
-//  We have extended the "T" packet to be able to also understand the
-//  following keys and values:
-//
-//  KEY           VALUE     DESCRIPTION
-//  ===========   ========  ================================================
-//  "metype"      unsigned  mach exception type (the value of the EXC_XXX enumerations)
-//                          as an unsigned integer. For targets with mach
-//                          kernels only.
-//
-//  "mecount"     unsigned  mach exception data count as an unsigned integer
-//                          For targets with mach kernels only.
-//
-//  "medata"      unsigned  There should be "mecount" of these and it is the data
-//                          that goes along with a mach exception (as an unsigned
-//                          integer). For targets with mach kernels only.
-//
-//  "name"        string    The name of the thread as a plain string. The string
-//                          must not contain an special packet characters or
-//                          contain a ':' or a ';'. Use "hexname" if the thread
-//                          name has special characters.
-//
-//  "hexname"     ascii-hex An ASCII hex string that contains the name of the thread
-//
-//  "qaddr"       hex       Big endian hex value that contains the libdispatch
-//                          queue address for the queue of the thread.
-//
-//  "reason"      enum      The enumeration must be one of:
-//                          "trace" the program stopped after a single instruction
-//                              was executed on a core. Usually done when single
-//                              stepping past a breakpoint
-//                          "breakpoint" a breakpoint set using a 'z' packet was hit.
-//                          "trap" stopped due to user interruption
-//                          "signal" stopped due to an actual unix signal, not
-//                              just the debugger using a unix signal to keep
-//                              the GDB remote client happy.
-//                          "watchpoint". Can be used with of the 
-//                              "watch"/"rwatch"/"awatch" key value pairs.
-//                              Or can be used *instead* of those keys, 
-//                              with the specially formatted "description" field.
-//                          "exception" an exception stop reason. Use with
-//                              the "description" key/value pair to describe the
-//                              exceptional event the user should see as the stop
-//                              reason.
-//  "description" ascii-hex An ASCII hex string that contains a more descriptive
-//           reason that the thread stopped. This is only needed
-//           if none of the key/value pairs are enough to
-//           describe why something stopped.
-//
-//           For "reason:watchpoint", "description" is an ascii-hex
-//           encoded string with between one and three base 10 numbers,
-//           space separated.  The three numbers are
-//             1. watchpoint address.  This address should always be within
-//                a memory region lldb has a watchpoint on.  
-//                On architectures where the actual reported hit address may
-//                be outside the watchpoint that was triggered, the remote
-//                stub should determine which watchpoint was triggered and
-//                report an address from within its range.
-//             2. watchpoint hardware register index number.
-//             3. actual watchpoint trap address, which may be outside
-//                the range of any watched region of memory. On MIPS, an addr
-//                outside a watched range means lldb should disable the wp, 
-//                step, re-enable the wp and continue silently.
-//
-//           On MIPS, the low 3 bits are masked so if a watchpoint is on 
-//           0x1004, a 2-byte write to 0x1000 will trigger the watchpoint 
-//           (a false positive hit), and lldb needs to disable the 
-//           watchpoint at 0x1004, inst-step, then re-enable the watchpoint
-//           and not make this a user visible event. The description here 
-//           would be "0x1004 0 0x1000". lldb needs a known watchpoint address
-//           in the first field, so it can disable it & step.
-//
-//           On AArch64 we have a related issue, where you watch 4 bytes at 
-//           0x1004, an instruction does an 8-byte write starting at 
-//           0x1000 (a true watchpoint hit) and the hardware may report the 
-//           trap address as 0x1000 - before the watched memory region - 
-//           with the write extending into the watched region.  This can 
-//           be reported as "0x1004 0 0x1000".  lldb will use 0x1004 to 
-//           identify which Watchpoint was triggered, and can report 0x1000 
-//           to the user.  The behavior of silently stepping over the 
-//           watchpoint, with an 3rd field addr outside the range, is 
-//           restricted to MIPS.
-//           There may be false-positive watchpoint hits on AArch64 as well,
-//           in the SVE Streaming Mode, but that is less common (see ESR 
-//           register flag "WPF", "Watchpoint might be False-Positive") and
-//           not currently handled by lldb.
-//           
-//  "threads"     comma-sep-base16  A list of thread ids for all threads (including
-//                                  the thread that we're reporting as stopped) that
-//                                  are live in the process right now.  lldb may
-//                                  request that this be included in the T packet via
-//                                  the QListThreadsInStopReply packet earlier in
-//                                  the debug session.
-//
-//                                  Example:
-//                                  threads:63387,633b2,63424,63462,63486;
-//
-//  "thread-pcs"  comma-sep-base16  A list of pc values for all threads that currently
-//                                  exist in the process, including the thread that
-//                                  this T packet is reporting as stopped.
-//                                  This key-value pair will only be emitted when the
-//                                  "threads" key is already included in the T packet.
-//                                  The pc values correspond to the threads reported
-//                                  in the "threads" list.  The number of pcs in the
-//                                  "thread-pcs" list will be the same as the number of
-//                                  threads in the "threads" list.
-//                                  lldb may request that this be included in the T
-//                                  packet via the QListThreadsInStopReply packet
-//                                  earlier in the debug session.
-//
-//                                  Example:
-//                                  thread-pcs:dec14,2cf872b0,2cf8681c,2d02d68c,2cf716a8;
-//
-//  "addressing_bits" unsigned optional  Specifies how many bits in addresses 
-//                                       are significant for addressing, base 
-//                                       10.  If bits 38..0 in a 64-bit 
-//                                       pointer are significant for 
-//                                       addressing, then the value is 39.  
-//                                       This is needed on e.g. AArch64
-//                                       v8.3 ABIs that use pointer 
-//                                       authentication in the high bits.
-//                                       This value is normally sent in the
-//                                       qHostInfo packet response, and if the
-//                                       value cannot change during the process
-//                                       lifetime, it does not need to be 
-//                                       duplicated here in the stop packet.
-//                                       For a firmware environment with early
-//                                       start code that may be changing the
-//                                       page table setup, a dynamically set
-//                                       value may be needed.
-//  "low_mem_addressing_bits" unsigned optional, specifies how many bits in 
-//                                       addresses in low memory are significant 
-//                                       for addressing, base 10.  AArch64 can 
-//                                       have different page table setups for low 
-//                                       and high memory, and therefore a different 
-//                                       number of bits used for addressing.
-//  "high_mem_addressing_bits" unsigned optional, specifies how many bits in 
-//                                       addresses in high memory are significant 
-//                                       for addressing, base 10.  AArch64 can have 
-//                                       different page table setups for low and 
-//                                       high memory, and therefore a different 
-//                                       number of bits used for addressing.
-//
-// BEST PRACTICES:
-//  Since register values can be supplied with this packet, it is often useful
-//  to return the PC, SP, FP, LR (if any), and FLAGS registers so that separate
-//  packets don't need to be sent to read each of these registers from each
-//  thread.
-//
-//  If a thread is stopped for no reason (like just because another thread
-//  stopped, or because when one core stops all cores should stop), use a
-//  "T" packet with "00" as the signal number and fill in as many key values
-//  and registers as possible.
-//
-//  LLDB likes to know why a thread stopped since many thread control
-//  operations like stepping over a source line, actually are implemented
-//  by running the process multiple times. If a breakpoint is hit while
-//  trying to step over a source line and LLDB finds out that a breakpoint
-//  is hit in the "reason", we will know to stop trying to do the step
-//  over because something happened that should stop us from trying to
-//  do the step. If we are at a breakpoint and we disable the breakpoint
-//  at the current PC and do an instruction single step, knowing that
-//  we stopped due to a "trace" helps us know that we can continue
-//  running versus stopping due to a "breakpoint" (if we have two
-//  breakpoint instruction on consecutive instructions). So the more info
-//  we can get about the reason a thread stops, the better job LLDB can
-//  do when controlling your process. A typical GDB server behavior is
-//  to send a SIGTRAP for breakpoints _and_ also when instruction single
-//  stepping, in this case the debugger doesn't really know why we
-//  stopped and it can make it hard for the debugger to control your
-//  program correctly. What if a real SIGTRAP was delivered to a thread
-//  while we were trying to single step? We wouldn't know the difference
-//  with a standard GDB remote server and we could do the wrong thing.
-//
-// PRIORITY TO IMPLEMENT
-//  High. Having the extra information in your stop reply packets makes
-//  your debug session more reliable and informative.
-//----------------------------------------------------------------------
-
-
-//----------------------------------------------------------------------
-// PLATFORM EXTENSION - for use as a GDB remote platform
-//----------------------------------------------------------------------
-// "qfProcessInfo"
-// "qsProcessInfo"
-//
-// BRIEF
-//  Get the first process info (qfProcessInfo) or subsequent process
-//  info (qsProcessInfo) for one or more processes on the remote
-//  platform. The first call gets the first match and subsequent calls
-//  to qsProcessInfo gets the subsequent matches. Return an error EXX,
-//  where XX are two hex digits, when no more matches are available.
-//
-// PRIORITY TO IMPLEMENT
-//  Required. The qfProcessInfo packet can be followed by a ':' and
-//  some key value pairs. The key value pairs in the command are:
-//
-//  KEY           VALUE     DESCRIPTION
-//  ===========   ========  ================================================
-//  "name"        ascii-hex An ASCII hex string that contains the name of
-//                          the process that will be matched.
-//  "name_match"  enum      One of: "equals", "starts_with", "ends_with",
-//                          "contains" or "regex"
-//  "pid"         integer   A string value containing the decimal process ID
-//  "parent_pid"  integer   A string value containing the decimal parent
-//                          process ID
-//  "uid"         integer   A string value containing the decimal user ID
-//  "gid"         integer   A string value containing the decimal group ID
-//  "euid"        integer   A string value containing the decimal effective user ID
-//  "egid"        integer   A string value containing the decimal effective group ID
-//  "all_users"   bool      A boolean value that specifies if processes should
-//                          be listed for all users, not just the user that the
-//                          platform is running as
-//  "triple"      string    An ASCII triple string ("x86_64",
-//                          "x86_64-apple-macosx", "armv7-apple-ios")
-//  "args"        string    A string value containing the process arguments
-//                          separated by the character '-', where each argument is
-//                          hex-encoded. It includes argv[0].
-//
-// The response consists of key/value pairs where the key is separated from the
-// values with colons and each pair is terminated with a semi colon. For a list
-// of the key/value pairs in the response see the "qProcessInfoPID" packet
-// documentation.
-//
-// Sample packet/response:
-// send packet: $qfProcessInfo#00
-// read packet: $pid:60001;ppid:59948;uid:7746;gid:11;euid:7746;egid:11;name:6c6c6462;triple:x86_64-apple-macosx;#00
-// send packet: $qsProcessInfo#00
-// read packet: $pid:59992;ppid:192;uid:7746;gid:11;euid:7746;egid:11;name:6d64776f726b6572;triple:x86_64-apple-macosx;#00
-// send packet: $qsProcessInfo#00
-// read packet: $E04#00
-//----------------------------------------------------------------------
-
-
-//----------------------------------------------------------------------
-// PLATFORM EXTENSION - for use as a GDB remote platform
-//----------------------------------------------------------------------
-// "qLaunchGDBServer"
-//
-// BRIEF
-//  Have the remote platform launch a GDB server.
-//
-// PRIORITY TO IMPLEMENT
-//  Required. The qLaunchGDBServer packet must be followed by a ':' and
-//  some key value pairs. The key value pairs in the command are:
-//
-//  KEY           VALUE     DESCRIPTION
-//  ===========   ========  ================================================
-//  "port"        integer   A string value containing the decimal port ID or
-//                          zero if the port should be bound and returned
-//
-//  "host"        integer   The host that connections should be limited to
-//                          when the GDB server is connected to.
-//
-// The response consists of key/value pairs where the key is separated from the
-// values with colons and each pair is terminated with a semi colon.
-//
-// Sample packet/response:
-// send packet: $qLaunchGDBServer:port:0;host:lldb.apple.com;#00
-// read packet: $pid:60025;port:50776;#00
-//
-// The "pid" key/value pair is only specified if the remote platform launched
-// a separate process for the GDB remote server and can be omitted if no
-// process was separately launched.
-//
-// The "port" key/value pair in the response lets clients know what port number
-// to attach to in case zero was specified as the "port" in the sent command.
-//----------------------------------------------------------------------
-
-
-//----------------------------------------------------------------------
-// PLATFORM EXTENSION - for use as a GDB remote platform
-//----------------------------------------------------------------------
-// "qProcessInfoPID:PID"
-//
-// BRIEF
-//  Have the remote platform get detailed information on a process by
-//  ID. PID is specified as a decimal integer.
-//
-// PRIORITY TO IMPLEMENT
-//  Optional.
-//
-// The response consists of key/value pairs where the key is separated from the
-// values with colons and each pair is terminated with a semi colon.
-//
-// The key value pairs in the response are:
-//
-//  KEY           VALUE     DESCRIPTION
-//  ===========   ========  ================================================
-//  "pid"         integer   Process ID as a decimal integer string
-//  "ppid"        integer   Parent process ID as a decimal integer string
-//  "uid"         integer   A string value containing the decimal user ID
-//  "gid"         integer   A string value containing the decimal group ID
-//  "euid"        integer   A string value containing the decimal effective user ID
-//  "egid"        integer   A string value containing the decimal effective group ID
-//  "name"        ascii-hex An ASCII hex string that contains the name of the process
-//  "triple"      string    A target triple ("x86_64-apple-macosx", "armv7-apple-ios")
-//
-// Sample packet/response:
-// send packet: $qProcessInfoPID:60050#00
-// read packet: $pid:60050;ppid:59948;uid:7746;gid:11;euid:7746;egid:11;name:6c6c6462;triple:x86_64-apple-macosx;#00
-//----------------------------------------------------------------------
-
-//----------------------------------------------------------------------
-// "vAttachName"
-//
-// BRIEF
-//  Same as vAttach, except instead of a "pid" you send a process name.
-//
-// PRIORITY TO IMPLEMENT
-//  Low. Only needed for "process attach -n".  If the packet isn't supported
-//  then "process attach -n" will fail gracefully.  So you need only to support
-//  it if attaching to a process by name makes sense for your environment.
-//----------------------------------------------------------------------
-
-//----------------------------------------------------------------------
-// "vAttachWait"
-//
-// BRIEF
-//  Same as vAttachName, except that the stub should wait for the next instance
-//  of a process by that name to be launched and attach to that.
-//
-// PRIORITY TO IMPLEMENT
-//  Low. Only needed to support "process attach -w -n" which will fail
-//  gracefully if the packet is not supported.
-//----------------------------------------------------------------------
-
-//----------------------------------------------------------------------
-// "qAttachOrWaitSupported"
-//
-// BRIEF
-//  This is a binary "is it supported" query.  Return OK if you support
-//  vAttachOrWait
-//
-// PRIORITY TO IMPLEMENT
-//  Low. This is required if you support vAttachOrWait, otherwise no support
-//  is needed since the standard "I don't recognize this packet" response
-//  will do the right thing.
-//----------------------------------------------------------------------
-
-//----------------------------------------------------------------------
-// "vAttachOrWait"
-//
-// BRIEF
-//  Same as vAttachWait, except that the stub will attach to a process
-//  by name if it exists, and if it does not, it will wait for a process
-//  of that name to appear and attach to it.
-//
-// PRIORITY TO IMPLEMENT
-//  Low. Only needed to implement "process attach -w -i false -n".  If
-//  you don't implement it but do implement -n AND lldb can somehow get
-//  a process list from your device, it will fall back on scanning the
-//  process list, and sending vAttach or vAttachWait depending on
-//  whether the requested process exists already.  This is racy,
-//  however, so if you want to support this behavior it is better to
-//  support this packet.
-//----------------------------------------------------------------------
-
-//----------------------------------------------------------------------
-// "jThreadExtendedInfo"
-//
-// BRIEF
-//  This packet, which takes its arguments as JSON and sends its reply as
-//  JSON, allows the gdb remote stub to provide additional information
-//  about a given thread.
-//
-// PRIORITY TO IMPLEMENT
-//  Low.  This packet is only needed if the gdb remote stub wants to
-//  provide interesting additional information about a thread for the
-//  user.
-//
-// This packet takes its arguments in JSON form ( http://www.json.org ).
-// At a minimum, a thread must be specified, for example:
-//
-//  jThreadExtendedInfo:{"thread":612910}
-//
-// Because this is a JSON string, the thread number is provided in base 10.
-// Additional key-value pairs may be provided by lldb to the gdb remote
-// stub.  For instance, on some versions of macOS, lldb can read offset
-// information out of the system libraries.  Using those offsets, debugserver
-// is able to find the Thread Specific Address (TSD) for a thread and include
-// that in the return information.  So lldb will send these additional fields
-// like so:
-//
-//   jThreadExtendedInfo:{"plo_pthread_tsd_base_address_offset":0,"plo_pthread_tsd_base_offset":224,"plo_pthread_tsd_entry_size":8,"thread":612910}
-//
-// There are no requirements for what is included in the response.  A simple
-// reply on a OS X Yosemite / iOS 8 may include the pthread_t value, the
-// Thread Specific Data (TSD) address, the dispatch_queue_t value if the thread
-// is associated with a GCD queue, and the requested Quality of Service (QoS)
-// information about that thread.  For instance, a reply may look like:
-//
-//  {"tsd_address":4371349728,"requested_qos":{"enum_value":33,"constant_name":"QOS_CLASS_USER_INTERACTIVE","printable_name":"User Interactive"},"pthread_t":4371349504,"dispatch_queue_t":140735087127872}
-//
-// tsd_address, pthread_t, and dispatch_queue_t are all simple key-value pairs.
-// The JSON standard requires that numbers be expressed in base 10 - so all of
-// these are.  requested_qos is a dictionary with three key-value pairs in it -
-// so the UI layer may choose the form most appropriate for displaying to the user.
-//
-// Sending JSON over gdb-remote protocol introduces some problems.  We may be
-// sending strings with arbitrary contents in them, including the '#', '$', and '*'
-// characters that have special meaning in gdb-remote protocol and cannot occur
-// in the middle of the string.  The standard solution for this would be to require
-// ascii-hex encoding of all strings, or ascii-hex encode the entire JSON payload.
-//
-// Instead, the binary escaping convention is used for JSON data.  This convention
-// (e.g. used for the X packet) says that if '#', '$', '*', or '}' are to occur in
-// the payload, the character '}' (0x7d) is emitted, then the metacharacter is emitted
-// xor'ed by 0x20.  The '}' character occurs in every JSON payload at least once, and
-// '}' ^ 0x20 happens to be ']' so the raw packet characters for a request will look
-// like
-//
-//  jThreadExtendedInfo:{"thread":612910}]
-//
-// on the wire.
-//----------------------------------------------------------------------
-
-//----------------------------------------------------------------------
-// "QEnableCompression"
-//
-// BRIEF
-//  This packet enables compression of the packets that the debug stub sends to lldb.
-//  If the debug stub can support compression, it indictes this in the reply of the
-//  "qSupported" packet.  e.g.
-//   LLDB SENDS:    qSupported:xmlRegisters=i386,arm,mips
-//   STUB REPLIES:  qXfer:features:read+;SupportedCompressions=lzfse,zlib-deflate,lz4,lzma;
-//
-//  If lldb knows how to use any of these compression algorithms, it can ask that this
-//  compression mode be enabled.  
-//
-//  QEnableCompression:type:zlib-deflate;
-//
-//  The debug stub should reply with an uncompressed "OK" packet to indicate that the
-//  request was accepted.  All further packets the stub sends will use this compression.
-//
-//  Packets are compressed as the last step before they are sent from the stub, and
-//  decompressed as the first step after they are received.  The packet format in compressed
-//  mode becomes one of two:
-//
-//   $N<uncompressed payload>#00
-//
-//   $C<size of uncompressed payload in base 10>:<compressed payload>#00
-//
-//  Where "#00" is the actual checksum value if noack mode is not enabled.  The checksum
-//  value is for the "N<uncompressed payload>" or
-//  "C<size of uncompressed payload in base 10>:<compressed payload>" bytes in the packet.
-//
-//  The size of the uncompressed payload in base 10 is provided because it will simplify
-//  decompression if the final buffer size needed is known ahead of time.
-//
-//  Compression on low-latency connections is unlikely to be an improvement.  Particularly
-//  when the debug stub and lldb are running on the same host.  It should only be used
-//  for slow connections, and likely only for larger packets.
-//
-//  Example compression algorithsm that may be used include
-//
-//    zlib-deflate
-//       The raw DEFLATE format as described in IETF RFC 1951.  With the ZLIB library, you
-//       can compress to this format with an initialization like
-//           deflateInit2 (&stream, 5, Z_DEFLATED, -15, 8, Z_DEFAULT_STRATEGY)
-//       and you can decompress with an initialization like
-//           inflateInit2 (&stream, -15)
-//
-//    lz4
-//       https://en.wikipedia.org/wiki/LZ4_(compression_algorithm)
-//       https://github.com/Cyan4973/lz4
-//       The libcompression APIs on darwin systems call this COMPRESSION_LZ4_RAW.
-//
-//    lzfse
-//       Compression algorithm added in macOS 10.11, with open source C reference
-//       implementation on github.
-//       https://en.wikipedia.org/wiki/LZFSE
-//       https://github.com/lzfse/lzfse
-//
-//    lzma
-//       libcompression implements "LZMA level 6", the default compression for the
-//       open source LZMA implementation.
-//----------------------------------------------------------------------
-
-//----------------------------------------------------------------------
-// "jGetLoadedDynamicLibrariesInfos"
-//
-// BRIEF
-//  This packet asks the remote debug stub to send the details about libraries
-//  being added/removed from the process as a performance optimization.
-//
-//  There are two ways this packet can be used.  Both return a dictionary of
-//  binary images formatted the same way.
-//
-//  One requests information on all shared libraries:
-//       jGetLoadedDynamicLibrariesInfos:{"fetch_all_solibs":true}
-//  with an optional `"report_load_commands":false` which can be added, asking
-//  that only the dyld SPI information (load addresses, filenames) be returned.
-//  The default behavior is that debugserver scans the mach-o header and load 
-//  commands of each binary, and returns it in the JSON reply.
-//
-//  And the second requests information about a list of shared libraries, given their load addresses:
-//       jGetLoadedDynamicLibrariesInfos:{"solib_addresses":[8382824135,3258302053,830202858503]}
-//
-//  The second call is both a performance optimization (instead of having lldb read the mach-o header/load commands
-//  out of memory with generic read packets) but also adds additional information in the form of the
-//  filename of the shared libraries (which is not available in the mach-o header/load commands.)
-//
-//  An example using the OS X 10.11 style call:
-//
-//  LLDB SENDS: jGetLoadedDynamicLibrariesInfos:{"image_count":1,"image_list_address":140734800075128}
-//  STUB REPLIES: ${"images":[{"load_address":4294967296,"mod_date":0,"pathname":"/tmp/a.out","uuid":"02CF262C-ED6F-3965-9E14-63538B465CFF","mach_header":{"magic":4277009103,"cputype":16777223,"cpusubtype":18446744071562067971,"filetype":2},"segments":{"name":"__PAGEZERO","vmaddr":0,"vmsize":4294967296,"fileoff":0,"filesize":0,"maxprot":0},{"name":"__TEXT","vmaddr":4294967296,"vmsize":4096,"fileoff":0,"filesize":4096,"maxprot":7},{"name":"__LINKEDIT","vmaddr":4294971392,"vmsize":4096,"fileoff":4096,"filesize":152,"maxprot":7}}]}#00
-//
-//  Or pretty-printed,
-//
-//  STUB REPLIES: ${"images":
-//                  [
-//                      {"load_address":4294967296,
-//                       "mod_date":0,
-//                       "pathname":"/tmp/a.out",
-//                       "uuid":"02CF262C-ED6F-3965-9E14-63538B465CFF",
-//                       "mach_header":
-//                          {"magic":4277009103,
-//                           "cputype":16777223,
-//                           "cpusubtype":18446744071562067971,
-//                           "filetype":2
-//                           },
-//                       "segments":
-//                        [
-//                          {"name":"__PAGEZERO",
-//                           "vmaddr":0,
-//                           "vmsize":4294967296,
-//                           "fileoff":0,
-//                           "filesize":0,
-//                           "maxprot":0
-//                          },
-//                          {"name":"__TEXT",
-//                           "vmaddr":4294967296,
-//                           "vmsize":4096,
-//                           "fileoff":0,
-//                           "filesize":4096,
-//                           "maxprot":7
-//                          },
-//                          {"name":"__LINKEDIT",
-//                           "vmaddr":4294971392,
-//                           "vmsize":4096,
-//                           "fileoff":4096,
-//                           "filesize":152,
-//                           "maxprot":7
-//                          }
-//                        ]
-//                      }
-//                  ]
-//              }
-//
-//
-// This is similar to the qXfer:libraries:read packet, and it could
-// be argued that it should be merged into that packet.  A separate
-// packet was created primarily because lldb needs to specify the
-// number of images to be read and the address from which the initial
-// information is read.  Also the XML DTD would need to be extended
-// quite a bit to provide all the information that the DynamicLoaderMacOSX
-// would need to work correctly on this platform.
-//
-// PRIORITY TO IMPLEMENT
-//  On OS X 10.11, iOS 9, tvOS 9, watchOS 2 and older: Low.  If this packet is absent,
-//  lldb will read the Mach-O headers/load commands out of memory.
-//  On macOS 10.12, iOS 10, tvOS 10, watchOS 3 and newer: High.  If this packet is absent,
-//  lldb will not know anything about shared libraries in the inferior, or where the main
-//  executable loaded.
-//----------------------------------------------------------------------
-
-//----------------------------------------------------------------------
-// "jThreadsInfo"
-//
-// BRIEF
-//  Ask for the server for thread stop information of all threads.
-//
-// PRIORITY TO IMPLEMENT
-//  Low. This is a performance optimization, which speeds up debugging by avoiding
-//  multiple round-trips for retrieving thread information. The information from this
-//  packet can be retrieved using a combination of qThreadStopInfo and m packets.
-//----------------------------------------------------------------------
-
-The data in this packet is very similar to the stop reply packets, but is packaged in
-JSON and uses JSON arrays where applicable. The JSON output looks like:
-    [
-      { "tid":1580681,
-        "metype":6,
-        "medata":[2,0],
-        "reason":"exception",
-        "qaddr":140735118423168,
-        "registers": {
-          "0":"8000000000000000",
-          "1":"0000000000000000",
-          "2":"20fabf5fff7f0000",
-          "3":"e8f8bf5fff7f0000",
-          "4":"0100000000000000",
-          "5":"d8f8bf5fff7f0000",
-          "6":"b0f8bf5fff7f0000",
-          "7":"20f4bf5fff7f0000",
-          "8":"8000000000000000",
-          "9":"61a8db78a61500db",
-          "10":"3200000000000000",
-          "11":"4602000000000000",
-          "12":"0000000000000000",
-          "13":"0000000000000000",
-          "14":"0000000000000000",
-          "15":"0000000000000000",
-          "16":"960b000001000000",
-          "17":"0202000000000000",
-          "18":"2b00000000000000",
-          "19":"0000000000000000",
-          "20":"0000000000000000"
-        },
-        "memory":[
-          {"address":140734799804592,"bytes":"c8f8bf5fff7f0000c9a59e8cff7f0000"},
-          {"address":140734799804616,"bytes":"00000000000000000100000000000000"}
-        ]
-      }
-    ]
-
-It contains an array of dictionaries with all of the key value pairs that are
-normally in the stop reply packet, including the expedited registers. The registers are
-passed as hex-encoded JSON string in debuggee-endian byte order. Note that the register
-numbers are decimal numbers, unlike the stop-reply packet, where they are written in
-hex. The packet also contains expedited memory in the "memory" key.  This allows the
-server to expedite memory that the client is likely to use (e.g., areas around the
-stack pointer, which are needed for computing backtraces) and it reduces the packet
-count.
-
-On macOS with debugserver, we expedite the frame pointer backchain for a thread
-(up to 256 entries) by reading 2 pointers worth of bytes at the frame pointer (for
-the previous FP and PC), and follow the backchain. Most backtraces on macOS and
-iOS now don't require us to read any memory!
-
-//----------------------------------------------------------------------
-// "jGetSharedCacheInfo"
-//
-// BRIEF
-//  This packet asks the remote debug stub to send the details about the inferior's
-//  shared cache. The shared cache is a collection of common libraries/frameworks that
-//  are mapped into every process at the same address on Darwin systems, and can be
-//  identified by a load address and UUID.
-//
-//
-//  LLDB SENDS: jGetSharedCacheInfo:{}
-//  STUB REPLIES: ${"shared_cache_base_address":140735683125248,"shared_cache_uuid":"DDB8D70C-C9A2-3561-B2C8-BE48A4F33F96","no_shared_cache":false,"shared_cache_private_cache":false]}#00
-//
-// PRIORITY TO IMPLEMENT
-//  Low.  When both lldb and the inferior process are running on the same computer, and lldb
-//  and the inferior process have the same shared cache, lldb may (as an optimization) read
-//  the shared cache out of its own memory instead of using gdb-remote read packets to read
-//  them from the inferior process.
-//----------------------------------------------------------------------
-
-//----------------------------------------------------------------------
-// "qQueryGDBServer"
-//
-// BRIEF
-//  Ask the platform for the list of gdbservers we have to connect
-//
-// PRIORITY TO IMPLEMENT
-//  Low. The packet is required to support connecting to gdbserver started
-//  by the platform instance automatically.
-//----------------------------------------------------------------------
-
-If the remote platform automatically started one or more gdbserver instance (without
-lldb asking it) then it have to return the list of port number or socket name for
-each of them what can be used by lldb to connect to those instances.
-
-The data in this packet is a JSON array of JSON objects with the following keys:
-"port":        <the port number to connect>        (optional)
-"socket_name": <the name of the socket to connect> (optional)
-
-Example packet:
-[
-    { "port": 1234 },
-    { "port": 5432 },
-    { "socket_name": "foo" }
-]
-
-//----------------------------------------------------------------------
-// "QSetDetachOnError"
-//
-// BRIEF
-//  Sets what the server should do when the communication channel with LLDB
-//  goes down. Either kill the inferior process (0) or remove breakpoints and
-//  detach (1).
-//
-// PRIORITY TO IMPLEMENT
-//  Low. Only required if the target wants to keep the inferior process alive
-//  when the communication channel goes down.
-//----------------------------------------------------------------------
-
-The data in this packet is a single a character, which should be '0' if the
-inferior process should be killed, or '1' if the server should remove all
-breakpoints and detach from the inferior.
-
-//----------------------------------------------------------------------
-// "jGetDyldProcessState"
-//
-// BRIEF
-//  This packet fetches the process launch state, as reported by libdyld on
-//  Darwin systems, most importantly to indicate when the system libraries 
-//  have initialized sufficiently to safely call utility functions.
-//
-//
-//  LLDB SENDS: jGetDyldProcessState
-//  STUB REPLIES: {"process_state_value":48,"process_state string":"dyld_process_state_libSystem_initialized"}
-//
-// PRIORITY TO IMPLEMENT
-//  Low. This packet is needed to prevent lldb's utility functions for
-//  scanning the Objective-C class list from running very early in 
-//  process startup.
-//----------------------------------------------------------------------
diff --git a/lldb/docs/lldb-platform-packets.txt b/lldb/docs/lldb-platform-packets.txt
deleted file mode 100644
index 4cf575e5ee8a..000000000000
--- a/lldb/docs/lldb-platform-packets.txt
+++ /dev/null
@@ -1,451 +0,0 @@
-Here is a brief overview of the packets that an lldb platform server
-needs to implement for the lldb testsuite to be run on a remote
-target device/system.
-
-These are almost all lldb extensions to the gdb-remote serial
-protocol.  Many of the vFile: packets are described to the "Host
-I/O Packets" detailed in the gdb-remote protocol documentation,
-although the lldb platform extensions include packets that are not
-defined there (vFile:size:, vFile:mode:, vFile:symlink, vFile:chmod:).
-Most importantly, the flags that lldb passes to vFile:open: are 
-incompatible with the flags that gdb specifies.
-
-
-//----------------------------------------------------------------------
-// QStartNoAckMode
-//
-// BRIEF
-//   A request to stop sending ACK packets for each properly formatted packet.
-//
-// EXAMPLE 
-//   A platform session will typically start like this:
-//
-//   receive: +$QStartNoAckMode#b0
-//   send:    +       <-- ACKing the properly formatted QStartNoAckMode packet
-//   send:    $OK#9a
-//   receive: +       <-- Our OK packet getting ACKed
-//
-//   ACK mode is now disabled.
-
-//----------------------------------------------------------------------
-// qHostInfo
-//
-// BRIEF
-//   Describe the hardware and OS of the target system
-//
-// EXAMPLE
-//
-//  receive: qHostInfo
-//  send:    cputype:16777228;cpusubtype:1;ostype:ios;watchpoint_exceptions_received:before;os_version:12.1;vendor:apple;default_packet_timeout:5;
-//
-//  All numbers are base 10, os_version is a string that will be parsed as major.minor.patch.
-
-//----------------------------------------------------------------------
-// qModuleInfo
-//
-// BRIEF
-//   Report information about a binary on the target system
-//
-// EXAMPLE
-//  receive: qModuleInfo:2f62696e2f6c73;
-//
-// FIXME finish this packet description, v. GDBRemoteCommunicationServerCommon::Handle_qModuleInfo
-
-
-//----------------------------------------------------------------------
-// qGetWorkingDir
-//
-// BRIEF
-//   Get the current working directory of the platform stub in
-//   ASCII hex encoding.
-//
-// EXAMPLE
-// 
-//  receive: qGetWorkingDir
-//  send:    2f4170706c65496e7465726e616c2f6c6c64622f73657474696e67732f342f5465737453657474696e67732e746573745f646973617373656d626c65725f73657474696e6773 
-
-
-
-//----------------------------------------------------------------------
-// QSetWorkingDir:
-//
-// BRIEF
-//   Set the current working directory of the platform stub in
-//   ASCII hex encoding.
-//
-// EXAMPLE
-// 
-//  receive: QSetWorkingDir:2f4170706c65496e7465726e616c2f6c6c64622f73657474696e67732f342f5465737453657474696e67732e746573745f646973617373656d626c65725f73657474696e6773 
-//  send:    OK
-
-//----------------------------------------------------------------------
-// qPlatform_mkdir:
-//
-// BRIEF
-//   Create a directory on the target system.
-//
-// EXAMPLE
-// 
-//  receive: qPlatform_mkdir:000001fd,2f746d702f6131
-//  send:    F0
-//
-//  request packet has the fields:
-//     1. mode bits in base 16
-//     2. file path in ascii-hex encoding
-//
-//  response is F followed by the return value of the mkdir() call,
-//  base 16 encoded.
-
-//----------------------------------------------------------------------
-// qPlatform_shell:
-//
-// BRIEF
-//   Run a shell command on the target system, return the output.
-//
-// EXAMPLE
-// 
-//  receive: qPlatform_shell:6c73202f746d702f,0000000a
-//  send:    F,0,0,<OUTPUT>
-//
-//  request packet has the fields:
-//     1. shell command ascii-hex encoded
-//     2. timeout 
-//     3. {optional} working directory ascii-hex encoded
-//
-//  Response is F followed by the return value of the command (base 16),
-//  followed by another number, followed by the output of the command
-/   in binary-escaped-data encoding.
-
-//----------------------------------------------------------------------
-// qLaunchGDBServer
-//
-// BRIEF
-//   Start a gdbserver process (gdbserver, debugserver, lldb-server)
-//   on the target system.
-//
-// EXAMPLE
-// 
-//  receive: qLaunchGDBServer;host:<HOSTNAME_LLDB_IS_ON>;
-//  send:    pid:1337;port:43001;
-//
-//  request packet hostname field is not ascii-hex encoded. Hostnames 
-//  don't have $ or # characters in them.
-//
-//  response to the packet is the pid of the newly launched gdbserver,
-//  and the port it is listening for a connection on.
-//
-//  When the testsuite is running, lldb may use the pid to kill off a 
-//  debugserver that doesn't seem to be responding, etc.
-
-//----------------------------------------------------------------------
-// qKillSpawnedProcess:
-//
-// BRIEF
-//   Kill a process running on the target system.
-//
-// EXAMPLE
-// 
-//  receive: qKillSpawnedProcess:1337
-//  send:    OK
-//
-//  The request packet has the process ID in base 10.
-
-//----------------------------------------------------------------------
-// qProcessInfoPID:
-//
-// BRIEF
-//   Gather information about a process running on the target
-//
-// EXAMPLE
-// 
-//  receive: qProcessInfoPID:71964
-//  send:    pid:71964;name:612e6f7574;
-//
-//  The request packet has the pid encoded in base 10.
-//
-//  The reply has semicolon-separated name:value fields, two are
-//  shown here.  pid is base 10 encoded.  name is ascii hex encoded.
-//  lldb-server can reply with many additional fields, but I think
-//  this is enough for the testsuite.
-
-//----------------------------------------------------------------------
-// qfProcessInfo:
-//
-// BRIEF
-//   Search the process table for processes matching criteria, 
-//   respond with them in multiple packets.
-//
-// EXAMPLE
-// 
-//  receive: qfProcessInfo:name_match:equals;name:6e6f70726f6365737365786973747377697468746869736e616d65;
-//  send:    pid:3500;name:612e6f7574;
-//
-//  The request packet has a criteria to search for, followed by
-//  a specific name.  
-//
-//  KEY           VALUE     DESCRIPTION
-//  ===========   ========  ================================================
-//  "name"        ascii-hex An ASCII hex string that contains the name of 
-//                          the process that will be matched.
-//  "name_match"  enum      One of: "equals", "starts_with", "ends_with", 
-//                          "contains" or "regex"
-//  "pid"         integer   A string value containing the decimal process ID
-//  "parent_pid"  integer   A string value containing the decimal parent 
-//                          process ID
-//  "uid"         integer   A string value containing the decimal user ID
-//  "gid"         integer   A string value containing the decimal group ID
-//  "euid"        integer   A string value containing the decimal effective user ID
-//  "egid"        integer   A string value containing the decimal effective group ID
-//  "all_users"   bool      A boolean value that specifies if processes should
-//                          be listed for all users, not just the user that the 
-//                          platform is running as
-//  "triple"      ascii-hex An ASCII hex target triple string ("x86_64", 
-//                          "x86_64-apple-macosx", "armv7-apple-ios")
-//
-//  If no criteria is given, qfProcessInfo will request a list of every process.
-//
-//  The lldb testsuite currently only uses name_match:equals and the
-//  no-criteria mode to list every process.
-//
-//  The response should include any information about the process that
-//  can be retrieved in semicolon-separated name:value fields.
-//  In this example, pid is base 10, name is ascii-hex encoded.
-//  The testsuite seems to only require these two.
-//
-//  This packet only responds with one process.  To get further matches to
-//  the search, qsProcessInfo should be sent.
-//
-//  If no process match is found, Exx should be returned.
-//
-//  Sample packet/response:
-//  send packet: $qfProcessInfo#00
-//  read packet: $pid:60001;ppid:59948;uid:7746;gid:11;euid:7746;egid:11;name:6c6c6462;triple:7838365f36342d6170706c652d6d61636f7378;#00
-//  send packet: $qsProcessInfo#00
-//  read packet: $pid:59992;ppid:192;uid:7746;gid:11;euid:7746;egid:11;name:6d64776f726b6572;triple:7838365f36342d6170706c652d6d61636f7378;#00
-//  send packet: $qsProcessInfo#00
-//  read packet: $E04#00
-
-//----------------------------------------------------------------------
-// qsProcessInfo
-//
-// BRIEF
-//   Return the next process info found by the most recent qfProcessInfo:
-//   packet.
-//
-// EXAMPLE
-// 
-//  Continues to return the results of the qfProcessInfo.  Once all matches
-//  have been sent, Exx is returned to indicate end of matches.
-
-//----------------------------------------------------------------------
-// qPathComplete
-//
-// BRIEF
-//   Get a list of matched disk files/directories by passing a boolean flag
-//   and a partial path.
-//
-// EXAMPLE
-//
-//   receive: qPathComplete:0,6d61696e
-//   send:    M6d61696e2e637070
-//   receive: qPathComplete:1,746573
-//   send:    M746573742f,74657374732f
-//
-//   If the first argument is zero, the result should contain all
-//   files (including directories) starting with the given path. If the
-//   argument is one, the result should contain only directories.
-//
-//   The result should be a comma-separated list of hex-encoded paths.
-//   Paths denoting a directory should end with a directory separator ('/' or '\').
-
-//----------------------------------------------------------------------
-// vFile:size:
-//
-// BRIEF
-//   Get the size of a file on the target system, filename in ASCII hex.
-//
-// EXAMPLE
-// 
-//  receive: vFile:size:2f746d702f61
-//  send:    Fc008
-//
-//  response is "F" followed by the file size in base 16.
-//  "F-1,errno" with the errno if an error occurs, base 16.
-
-
-//----------------------------------------------------------------------
-// vFile:mode:
-//
-// BRIEF
-//   Get the mode bits of a file on the target system, filename in ASCII hex.
-//
-// EXAMPLE
-// 
-//  receive: vFile:mode:2f746d702f61
-//  send:    F1ed
-//
-//  response is "F" followed by the mode bits in base 16, this 0x1ed would 
-//  correspond to 0755 in octal.  
-//  "F-1,errno" with the errno if an error occurs, base 16.
-
-//----------------------------------------------------------------------
-// vFile:unlink:
-//
-// BRIEF
-//   Remove a file on the target system.
-//
-// EXAMPLE
-// 
-//  receive: vFile:unlink:2f746d702f61
-//  send:    F0
-//
-//  Argument is a file path in ascii-hex encoding.
-//  Response is "F" plus the return value of unlink(), base 16 encoding.
-//  Return value may optionally be followed by a comma and the base16
-//  value of errno if unlink failed.
-
-//----------------------------------------------------------------------
-// vFile:symlink:
-//
-// BRIEF
-//   Create a symbolic link (symlink, soft-link) on the target system.
-//
-// EXAMPLE
-// 
-//  receive: vFile:symlink:<SRC-FILE>,<DST-NAME>
-//  send:    F0,0
-//
-//  Argument file paths are in ascii-hex encoding.
-//  Response is "F" plus the return value of symlink(), base 16 encoding, 
-//  optionally followed by the value of errno if it failed, also base 16.
-
-//----------------------------------------------------------------------
-// vFile:chmod:
-// qPlatform_chmod:
-//
-// BRIEF
-//   Change the permission mode bits on a file on the target
-//
-// EXAMPLE
-// 
-//  receive: vFile:chmod:180,2f746d702f61
-//  send:    F0
-//
-//  Arguments are the mode bits to set, base 16, and a file path in 
-//  ascii-hex encoding.
-//  Response is "F" plus the return value of chmod(), base 16 encoding.
-//
-//  I don't know why there are two packets for the same thing, v.
-//  vFile:chmod:.
-
-//----------------------------------------------------------------------
-// vFile:chmod:
-//
-// BRIEF
-//   Change the permission mode bits on a file on the target
-//
-// EXAMPLE
-// 
-//  receive: vFile:chmod:180,2f746d702f61
-//  send:    F0
-//
-//  Arguments are the mode bits to set, base 16, and a file path in 
-//  ascii-hex encoding.
-//  Response is "F" plus the return value of chmod(), base 10 encoding.
-
-
-//----------------------------------------------------------------------
-// vFile:open:
-//
-// BRIEF
-//   Open a file on the remote system and return the file descriptor of it.
-//
-// EXAMPLE
-// 
-//  receive: vFile:open:2f746d702f61,00000001,00000180
-//  send:    F8
-//
-//  request packet has the fields:
-//     1. ASCII hex encoded filename
-//     2. flags passed to the open call, base 16.
-//        Note that these are not the oflags that open(2) takes, but
-//        are the constant values in enum OpenOptions from lldb's File.h
-//     3. mode bits, base 16
-//  
-//  response is F followed by the opened file descriptor in base 16.
-//  "F-1,errno" with the errno if an error occurs, base 16.
-//
-//----------------------------------------------------------------------
-// vFile:close:
-//
-// BRIEF
-//   Close a previously opened file descriptor.
-//
-// EXAMPLE
-// 
-//  receive: vFile:close:7
-//  send:    F0
-//
-//  File descriptor is in base 16.
-//  "F-1,errno" with the errno if an error occurs,
-//  errno is base 16.
-
-
-//----------------------------------------------------------------------
-// vFile:pread:
-//
-// BRIEF
-//   Read data from an opened file descriptor.
-//
-// EXAMPLE
-// 
-//  receive: vFile:pread:7,1024,0
-//  send:    F4;a'b\00
-//
-//  request packet has the fields:
-//     1. file descriptor, base 16
-//     2. number of bytes to be read, base 16
-//     3. offset into file to start from, base 16
-//
-//  Response is F, followed by the number of bytes read (base 16), a
-//  semicolon, followed by the data in the binary-escaped-data encoding.
-
-
-//----------------------------------------------------------------------
-// vFile:pwrite:
-//
-// BRIEF
-//   Write data to a previously opened file descriptor.
-//
-// EXAMPLE
-// 
-//  receive: vFile:pwrite:8,0,\cf\fa\ed\fe\0c\00\00
-//  send:    F1024
-//
-//  request packet has the fields:
-//     1. file descriptor, base 16
-//     2. offset into file to start from, base 16
-//     3. binary-escaped-data to be written
-//
-//  Response is F, followed by the number of bytes written (base 16)
-
-
-
-
-
-Finally, the platform must be able to launch processes so that debugserver
-can attach to them.  To do this, the following packets should be handled:
-
-QSetDisableASLR
-QSetDetachOnError
-QSetSTDOUT
-QSetSTDERR
-QSetSTDIN
-QEnvironment
-QEnvironmentHexEncoded
-A
-qLaunchSuccess
-qProcessInfo
-
-Most of these are documented in the standard gdb-remote protocol
-and/or the lldb-gdb-remote.txt documentation.
diff --git a/lldb/docs/resources/lldbgdbremote.md b/lldb/docs/resources/lldbgdbremote.md
new file mode 100644
index 000000000000..a9fa2a432b70
--- /dev/null
+++ b/lldb/docs/resources/lldbgdbremote.md
@@ -0,0 +1,2473 @@
+# GDB Remote Protocol Extensions
+
+LLDB has added new GDB server packets to better support multi-threaded and
+remote debugging.
+
+Why? Normally you need to start the correct GDB and the
+correct GDB server when debugging. If you have mismatch, then things go wrong
+very quickly. LLDB makes extensive use of the GDB remote protocol and we
+wanted to make sure that the experience was a bit more dynamic where we can
+discover information about a remote target without having to know anything up
+front.
+
+We also ran into performance issues with the existing GDB remote
+protocol that can be overcome when using a reliable communications layer.
+
+Some packets improve performance, others allow for remote process launching
+(if you have an OS), and others allow us to dynamically figure out what
+registers a thread might have. Again with GDB, both sides pre-agree on how the
+registers will look (how many, their register number,name and offsets).
+
+We prefer to be able to dynamically determine what kind of architecture, OS and
+vendor we are debugging, as well as how things are laid out when it comes to
+the thread register contexts.
+
+Below are the details on the new packets we have added above and beyond the
+standard GDB remote protocol packets.
+
+## QStartNoAckMode
+
+Try to enable no ACK mode to skip sending ACKs and NACKs.
+
+Having to send an ACK/NACK after every packet slows things down a bit, so we
+have a way to disable ACK packets to minimize the traffic for reliable
+communication interfaces (like sockets). Below GDB or LLDB will send this
+packet to try and disable ACKs. All lines that start with "send packet: " are
+from GDB/LLDB, and all lines that start with "read packet: " are from the GDB
+remote server:
+```
+send packet: $QStartNoAckMode#b0
+read packet: +
+read packet: $OK#9a
+send packet: +
+```
+
+### Priority To Implement
+
+High. Any GDB remote server that can implement this should if the
+connection is reliable. This improves packet throughput and increases
+the performance of the connection.
+
+## QSupported
+
+Query the GDB remote server for features it supports
+
+QSupported is a standard GDB Remote Serial Protocol packet, but
+there are several additions to the response that lldb can parse.
+They are not all listed here.
+
+An example exchange:
+```
+send packet: qSupported:xmlRegisters=i386,arm,mips,arc;multiprocess+;fork-events+;vfork-events+
+
+read packet: qXfer:features:read+;PacketSize=20000;qEcho+;native-signals+;SupportedCompressions=lzfse,zlib-deflate,lz4,lzma;SupportedWatchpointTypes=aarch64-mask,aarch64-bas;
+```
+
+In the example above, three lldb extensions are shown:
+
+  * `PacketSize=20000`
+    * The base 16 maximum packet size that the stub can handle.
+  * `SupportedCompressions=<item,item,...>`
+    * A list of compression types that the stub can use to compress packets
+    when the QEnableCompression packet is used to request one of them.
+  * `SupportedWatchpointTypes=<item,item,...>`
+    * A list of watchpoint types that this stub can manage. Currently defined 
+      names are:
+        * `x86_64` - 64-bit x86-64 watchpoints (1, 2, 4, 8 byte watchpoints
+          aligned to those amounts)
+        * `aarch64-bas`  AArch64 Byte Address Select watchpoints
+                     (any number of contiguous bytes within a doubleword)
+        * `aarch64-mask` AArch64 MASK watchpoints
+                     (any power-of-2 region of memory from 8 to 2GB, aligned)
+
+      If nothing is specified, lldb will default to sending power-of-2
+      watchpoints, up to a pointer size, `sizeof(void*)`, a reasonable
+      baseline assumption.
+
+### Priority To Implement
+
+Optional.
+
+## "A" - launch args packet
+
+Launch a program using the supplied arguments
+
+We have added support for the "set program arguments" packet where we can
+start a connection to a remote server and then later supply the path to the
+executable and the arguments to use when executing:
+
+GDB remote docs for this:
+```
+set program arguments(reserved) Aarglen,argnum,arg,...
+```
+Where A is followed by the length in bytes of the hex encoded argument,
+followed by an argument integer, and followed by the ASCII characters
+converted into hex bytes for each arg:
+```
+send packet: $A98,0,2f566f6c756d65732f776f726b2f67636c6179746f6e2f446f63756d656e74732f7372632f6174746163682f612e6f7574#00
+read packet: $OK#00
+```
+The above packet helps when you have remote debugging abilities where you
+could launch a process on a remote host, this isn't needed for bare board
+debugging.
+
+### Priority To Implement
+
+Low. Only needed if the remote target wants to launch a target after
+making a connection to a GDB server that isn't already connected to
+an inferior process.
+
+## qLaunchSuccess
+
+Check whether launching a process with the `A` packet succeeded.
+
+Returns the status of the last attempt to launch a process.
+Either `OK` if no error ocurred, or `E` followed by a string
+describing the error.
+
+### Priority To Implement
+
+High, launching processes is a key part of LLDB's platform mode.
+
+## QEnvironment:NAME=VALUE
+
+Setup the environment up for a new child process that will soon be
+launched using the "A" packet.
+
+NB: key/value pairs are sent as-is so gdb-remote protocol meta characters
+(e.g. `#` or `$`) are not acceptable.  If any non-printable or
+metacharacters are present in the strings, `QEnvironmentHexEncoded`
+should be used instead if it is available.  If you don't want to
+scan the environment strings before sending, prefer
+the `QEnvironmentHexEncoded` packet over `QEnvironment`, if it is
+available.
+
+Both GDB and LLDB support passing down environment variables. Is it ok to
+respond with a `$#00` (unimplemented):
+```
+send packet: $QEnvironment:ACK_COLOR_FILENAME=bold yellow#00
+read packet: $OK#00
+```
+This packet can be sent one or more times _prior_ to sending a "A" packet.
+
+### Priority To Implement
+
+Low. Only needed if the remote target wants to launch a target after
+making a connection to a GDB server that isn't already connected to
+an inferior process.
+
+## QEnvironmentHexEncoded:HEX-ENCODING(NAME=VALUE)
+
+Setup the environment up for a new child process that will soon be
+launched using the "A" packet.
+
+The only difference between this packet and `QEnvironment` is that the
+environment key-value pair is ascii hex encoded for transmission.
+This allows values with gdb-remote metacharacters like `#` to be sent.
+
+Both GDB and LLDB support passing down environment variables. Is it ok to
+respond with a `$#00` (unimplemented):
+```
+send packet: $QEnvironment:41434b5f434f4c4f525f46494c454e414d453d626f6c642379656c6c6f77#00
+read packet: $OK#00
+```
+This packet can be sent one or more times _prior_ to sending a "A" packet.
+
+### Priority To Implement
+
+Low. Only needed if the remote target wants to launch a target after
+making a connection to a GDB server that isn't already connected to
+an inferior process.
+
+## QEnableErrorStrings
+
+This packet enables reporting of Error strings in remote packet
+replies from the server to client. If the server supports this
+feature, it should send an OK response. The client can expect the
+following error replies if this feature is enabled in the server:
+```
+EXX;AAAAAAAAA
+```
+where `AAAAAAAAA` will be a hex encoded ASCII string.
+`XX`` is hex encoded byte number.
+
+It must be noted that even if the client has enabled reporting
+strings in error replies, it must not expect error strings to all
+error replies.
+
+### Priority To Implement
+
+Low. Only needed if the remote target wants to provide strings that
+are human readable along with an error code.
+
+### Example
+
+```
+send packet: $QEnableErrorStrings
+read packet: $OK#00
+```
+
+## QSetSTDIN:\<ascii-hex-path\> / QSetSTDOUT:\<ascii-hex-path\> / QSetSTDERR:\<ascii-hex-path\>
+
+Setup where STDIN, STDOUT, and STDERR go prior to sending an "A"
+packet.
+
+When launching a program through the GDB remote protocol with the "A" packet,
+you might also want to specify where stdin/out/err go:
+```
+QSetSTDIN:<ascii-hex-path>
+QSetSTDOUT:<ascii-hex-path>
+QSetSTDERR:<ascii-hex-path>
+```
+These packets must be sent  _prior_ to sending a "A" packet.
+
+### Priority To Implement
+
+Low. Only needed if the remote target wants to launch a target after
+making a connection to a GDB server that isn't already connected to
+an inferior process.
+
+
+## QSetWorkingDir:\<ascii-hex-path\>
+
+Set the working directory prior to sending an "A" packet.
+
+Or specify the working directory:
+```
+QSetWorkingDir:<ascii-hex-path>
+```
+This packet must be sent  _prior_ to sending a "A" packet.
+
+### Priority To Implement
+
+Low. Only needed if the remote target wants to launch a target after
+making a connection to a GDB server that isn't already connected to
+an inferior process.
+
+## qGetWorkingDir
+
+Get the current working directory of the platform stub in
+ASCII hex encoding.
+
+### Example
+
+```
+receive: qGetWorkingDir
+send:    2f4170706c65496e7465726e616c2f6c6c64622f73657474696e67732f342f5465737453657474696e67732e746573745f646973617373656d626c65725f73657474696e6773
+```
+
+## QSetDisableASLR:\<bool\>
+
+Enable or disable ASLR on the next "A" packet.
+
+Or control if ASLR is enabled/disabled:
+```
+send packet: QSetDisableASLR:1
+read packet: OK
+
+send packet: QSetDisableASLR:0
+read packet: OK
+```
+This packet must be sent  _prior_ to sending a "A" packet.
+
+### Priority To Implement
+
+Low. Only needed if the remote target wants to launch a target after
+making a connection to a GDB server that isn't already connected to
+an inferior process and if the target supports disabling ASLR
+(Address space layout randomization).
+
+## QListThreadsInStopReply
+
+Enable the `threads:` and `thread-pcs:` data in the question-mark packet
+("T packet") responses when the stub reports that a program has
+stopped executing.
+
+### Priority To Implement
+
+Performance.  This is a performance benefit to lldb if the thread id's
+and thread pc values are provided to lldb in the T stop packet -- if
+they are not provided to lldb, lldb will likely need to send one to
+two packets per thread to fetch the data at every private stop.
+
+### Example
+
+```
+send packet: QListThreadsInStopReply
+read packet: OK
+```
+
+## jLLDBTraceSupported
+
+Get the processor tracing type supported by the gdb-server for the current
+inferior. Responses might be different depending on the architecture and
+capabilities of the underlying OS.
+
+### Output Schema
+
+```
+ {
+   "name": <string>,
+       Tracing technology name, e.g. intel-pt, arm-etm.
+   "description": <string>,
+       Description for this technology.
+ }
+```
+
+If no tracing technology is supported for the inferior, or no process is
+running, then an error message is returned.
+
+### Note
+
+This packet is used by Trace plug-ins (see `lldb_private::Trace.h`) to
+do live tracing. Specifically, the name of the plug-in should match the name
+of the tracing technology returned by this packet.
+
+### Example
+
+```
+send packet: jLLDBTraceSupported
+read packet: {"name":<name>, "description":<description>}/E<error code>;AAAAAAAAA
+```
+
+## jLLDBTraceStart
+
+Start tracing a process or its threads using a provided tracing technology.
+The input and output are specified as JSON objects. In case of success, an OK
+response is returned, or an error otherwise.
+
+### Process Tracing
+
+This traces existing and future threads of the current process. An error is
+returned if the process is already being traced.
+
+### Thread Tracing
+
+This traces specific threads.
+
+### Input Schema
+
+```
+{
+  "type": <string>,
+      Tracing technology name, e.g. intel-pt, arm-etm.
+
+  /* thread tracing only */
+  "tids"?: [<decimal integer>],
+      Individual threads to trace.
+
+  ... other parameters specific to the provided tracing type
+}
+```
+
+### Notes
+
+- If "tids" is not provided, then the operation is "process tracing",
+  otherwise it's "thread tracing".
+- Each tracing technology can have different levels of support for "thread
+  tracing" and "process tracing".
+
+### Intel-Pt
+
+intel-pt supports both "thread tracing" and "process tracing".
+
+"Process tracing" is implemented in two different ways. If the
+"perCpuTracing" option is false, then each thread is traced individually
+but managed by the same "process trace" instance. This means that the
+amount of trace buffers used is proportional to the number of running
+threads. This is the recommended option unless the number of threads is
+huge. If "perCpuTracing" is true, then each cpu core is traced invidually
+instead of each thread, which uses a fixed number of trace buffers, but
+might result in less data available for less frequent threads. See
+"perCpuTracing" below for more information.
+
+Each actual intel pt trace buffer, either from "process tracing" or "thread
+tracing", is stored in an in-memory circular buffer, which keeps the most
+recent data.
+
+Additional params in the input schema:
+```
+ {
+   "iptTraceSize": <decimal integer>,
+       Size in bytes used by each individual per-thread or per-cpu trace
+       buffer. It must be a power of 2 greater than or equal to 4096 (2^12)
+       bytes.
+
+   "enableTsc": <boolean>,
+       Whether to enable TSC timestamps or not. This is supported on
+       all devices that support intel-pt. A TSC timestamp is generated along
+       with PSB (synchronization) packets, whose frequency can be configured
+       with the "psbPeriod" parameter.
+
+   "psbPeriod"?: <Optional decimal integer>,
+       This value defines the period in which PSB packets will be generated.
+       A PSB packet is a synchronization packet that contains a TSC
+       timestamp and the current absolute instruction pointer.
+
+       This parameter can only be used if
+
+           /sys/bus/event_source/devices/intel_pt/caps/psb_cyc
+
+       is 1. Otherwise, the PSB period will be defined by the processor.
+
+       If supported, valid values for this period can be found in
+
+           /sys/bus/event_source/devices/intel_pt/caps/psb_periods
+
+       which contains a hexadecimal number, whose bits represent valid
+       values e.g. if bit 2 is set, then value 2 is valid.
+
+       The psb_period value is converted to the approximate number of
+       raw trace bytes between PSB packets as:
+
+           2 ^ (value + 11)
+
+        e.g. value 3 means 16KiB between PSB packets. Defaults to
+        0 if supported.
+
+   /* process tracing only */
+   "perCpuTracing": <boolean>
+       Instead of having an individual trace buffer per thread, this option
+       triggers the collection on a per cpu core basis. This effectively
+       traces the entire activity on all cores. At decoding time, in order
+       to correctly associate a decoded instruction with a thread, the
+       context switch trace of each core is needed, as well as a record per
+       cpu indicating which thread was running on each core when tracing
+       started. These secondary traces are correlated with the intel-pt
+       trace by comparing TSC timestamps.
+
+       This option forces the capture of TSC timestamps (see "enableTsc").
+
+       Note: This option can't be used simulatenously with any other trace
+       sessions because of its system-wide nature.
+
+   /* process tracing only */
+   "processBufferSizeLimit": <decimal integer>,
+       Maximum total buffer size per process in bytes.
+       This limit applies to the sum of the sizes of all thread or cpu core
+       buffers for the current process, excluding the ones started with
+       "thread tracing".
+
+       If "perCpuTracing" is false, whenever a thread is attempted to be
+       traced due to "process tracing" and the limit would be reached, the
+       process is stopped with a "tracing" reason along with a meaningful
+       description, so that the user can retrace the process if needed.
+
+       If "perCpuTracing" is true, then starting the system-wide trace
+       session fails if all the individual per-cpu trace buffers require
+       in total more memory that the limit impossed by this parameter.
+ }
+```
+
+Notes:
+ - Modifying the parameters of an existing trace is not supported. The user
+   needs to stop the trace and start a new one.
+ - If "process tracing" is attempted and there are individual threads
+   already being traced with "thread tracing", these traces are left
+   unaffected and the threads not traced twice.
+ - If "thread tracing" is attempted on a thread already being traced with
+   either "thread tracing" or "process tracing", it fails.
+
+### Examples
+
+Process tracing:
+```
+send packet: jLLDBTraceStart:{"type":<type>,...other params}]
+read packet: OK/E<error code>;AAAAAAAAA
+```
+
+Thread tracing:
+```
+send packet: jLLDBTraceStart:{"type":<type>,"tids":<tids>,...other params}]
+read packet: OK/E<error code>;AAAAAAAAA
+```
+
+## jLLDBTraceStop
+
+Stop tracing a process or its threads using a provided tracing technology.
+The input and output are specified as JSON objects. In case of success, an OK
+response is returned, or an error otherwise.
+
+### Process Trace Stopping
+
+Stopping a process trace stops the active traces initiated with
+"thread tracing".
+
+### Thread Trace Stopping
+
+This is a best effort request, which tries to stop as many traces as
+possible.
+
+### Input Schema
+
+The schema for the input is
+```
+{
+  "type": <string>
+     Tracing technology name, e.g. intel-pt, arm-etm.
+
+  /* thread trace stopping only */
+  "tids":  [<decimal integer>]
+     Individual thread traces to stop.
+}
+```
+
+### Notes
+
+- If "tids" is not provided, then the operation is "process trace stopping".
+
+### Intel Pt
+
+Stopping a specific thread trace started with "process tracing" is allowed.
+
+### Examples
+
+Process trace stopping:
+```
+send packet: jLLDBTraceStop:{"type":<type>}]
+read packet: OK/E<error code>;AAAAAAAAA
+```
+Thread trace stopping:
+```
+send packet: jLLDBTraceStop:{"type":<type>,"tids":<tids>}]
+read packet: OK/E<error code>;AAAAAAAAA
+```
+
+## jLLDBTraceGetState
+
+Get the current state of the process and its threads being traced by
+a given trace technology. The response is a JSON object with custom
+information depending on the trace technology. In case of errors, an
+error message is returned.
+
+### Input Schema
+
+```
+{
+   "type": <string>
+      Tracing technology name, e.g. intel-pt, arm-etm.
+}
+```
+
+### Output Schema
+
+```
+{
+  "tracedThreads": [{
+    "tid": <decimal integer>,
+    "binaryData": [
+      {
+        "kind": <string>,
+            Identifier for some binary data related to this thread to
+            fetch with the jLLDBTraceGetBinaryData packet.
+        "size": <decimal integer>,
+            Size in bytes of this thread data.
+      },
+    ]
+  }],
+  "processBinaryData": [
+    {
+      "kind": <string>,
+          Identifier for some binary data related to this process to
+          fetch with the jLLDBTraceGetBinaryData packet.
+      "size": <decimal integer>,
+          Size in bytes of this thread data.
+    },
+  ],
+  "cpus"?: [
+    "id": <decimal integer>,
+        Identifier for this CPU logical core.
+    "binaryData": [
+      {
+        "kind": <string>,
+            Identifier for some binary data related to this thread to
+            fetch with the jLLDBTraceGetBinaryData packet.
+        "size": <decimal integer>,
+            Size in bytes of this cpu core data.
+      },
+    ]
+  ],
+  "warnings"?: [<string>],
+      Non-fatal messages useful for troubleshooting.
+
+  ... other attributes specific to the given tracing technology
+}
+```
+
+### Notes
+
+ - "traceThreads" includes all thread traced by both "process tracing" and
+   "thread tracing".
+
+### Intel Pt
+
+If per-cpu process tracing is enabled, "tracedThreads" will contain all
+the threads of the process without any trace buffers. Besides that, the
+"cpus" field will also be returned with per cpu core trace buffers.
+A side effect of per-cpu tracing is that all the threads of unrelated
+processes will also be traced, thus polluting the tracing data.
+
+Binary data kinds:
+  - iptTrace: trace buffer for a thread or a cpu.
+  - perfContextSwitchTrace: context switch trace for a cpu generated by
+                            perf_event_open.
+  - procfsCpuInfo: contents of the /proc/cpuinfo file.
+
+Additional attributes:
+  * tscPerfZeroConversion
+    * This field allows converting Intel processor's TSC values to nanoseconds.
+      It is available through the Linux perf_event API when cap_user_time and cap_user_time_zero
+      are set.
+      See the documentation of time_zero in
+      https://man7.org/linux/man-pages/man2/perf_event_open.2.html for more information about
+      the calculation and the meaning of the values in the schema below.
+
+      Schema for this field:
+      ```
+      "tscPerfZeroConversion": {
+        "timeMult": <decimal integer>,
+        "timeShift": <decimal integer>,
+        "timeZero": <decimal integer>,
+      }
+      ```
+
+### Example
+
+```
+send packet: jLLDBTraceGetState:{"type":<type>}]
+read packet: {...object}/E<error code>;AAAAAAAAA
+```
+
+## jLLDBTraceGetBinaryData
+
+Get binary data given a trace technology and a data identifier.
+The input is specified as a JSON object and the response has the same format
+as the "binary memory read" (aka "x") packet. In case of failures, an error
+message is returned.
+
+### Schema
+
+The schema for the input is:
+```
+{
+ "type": <string>,
+     Tracing technology name, e.g. intel-pt, arm-etm.
+ "kind": <string>,
+     Identifier for the data.
+ "cpuId": <Optional decimal>,
+     Core id in decimal if the data belongs to a CPU core.
+ "tid"?: <Optional decimal>,
+     Tid in decimal if the data belongs to a thread.
+}
+```
+
+### Example
+
+```
+send packet: jLLDBTraceGetBinaryData:{"type":<type>,"kind":<query>,"tid":<tid>,"offset":<offset>,"size":<size>}]
+read packet: <binary data>/E<error code>;AAAAAAAAA
+```
+
+## qRegisterInfo\<hex-reg-id\>
+
+Discover register information from the remote GDB server.
+
+With LLDB, for register information, remote GDB servers can add
+support for the "qRegisterInfoN" packet where "N" is a zero based
+base 16 register number that must start at zero and increase by one
+for each register that is supported.  The response is done in typical
+GDB remote fashion where a series of "KEY:VALUE;" pairs are returned.
+An example for the x86_64 registers is included below:
+```
+send packet: $qRegisterInfo0#00
+read packet: $name:rax;bitsize:64;offset:0;encoding:uint;format:hex;set:General Purpose Registers;gcc:0;dwarf:0;#00
+send packet: $qRegisterInfo1#00
+read packet: $name:rbx;bitsize:64;offset:8;encoding:uint;format:hex;set:General Purpose Registers;gcc:3;dwarf:3;#00
+send packet: $qRegisterInfo2#00
+read packet: $name:rcx;bitsize:64;offset:16;encoding:uint;format:hex;set:General Purpose Registers;gcc:2;dwarf:2;#00
+send packet: $qRegisterInfo3#00
+read packet: $name:rdx;bitsize:64;offset:24;encoding:uint;format:hex;set:General Purpose Registers;gcc:1;dwarf:1;#00
+send packet: $qRegisterInfo4#00
+read packet: $name:rdi;bitsize:64;offset:32;encoding:uint;format:hex;set:General Purpose Registers;gcc:5;dwarf:5;#00
+send packet: $qRegisterInfo5#00
+read packet: $name:rsi;bitsize:64;offset:40;encoding:uint;format:hex;set:General Purpose Registers;gcc:4;dwarf:4;#00
+send packet: $qRegisterInfo6#00
+read packet: $name:rbp;alt-name:fp;bitsize:64;offset:48;encoding:uint;format:hex;set:General Purpose Registers;gcc:6;dwarf:6;generic:fp;#00
+send packet: $qRegisterInfo7#00
+read packet: $name:rsp;alt-name:sp;bitsize:64;offset:56;encoding:uint;format:hex;set:General Purpose Registers;gcc:7;dwarf:7;generic:sp;#00
+send packet: $qRegisterInfo8#00
+read packet: $name:r8;bitsize:64;offset:64;encoding:uint;format:hex;set:General Purpose Registers;gcc:8;dwarf:8;#00
+send packet: $qRegisterInfo9#00
+read packet: $name:r9;bitsize:64;offset:72;encoding:uint;format:hex;set:General Purpose Registers;gcc:9;dwarf:9;#00
+send packet: $qRegisterInfoa#00
+read packet: $name:r10;bitsize:64;offset:80;encoding:uint;format:hex;set:General Purpose Registers;gcc:10;dwarf:10;#00
+send packet: $qRegisterInfob#00
+read packet: $name:r11;bitsize:64;offset:88;encoding:uint;format:hex;set:General Purpose Registers;gcc:11;dwarf:11;#00
+send packet: $qRegisterInfoc#00
+read packet: $name:r12;bitsize:64;offset:96;encoding:uint;format:hex;set:General Purpose Registers;gcc:12;dwarf:12;#00
+send packet: $qRegisterInfod#00
+read packet: $name:r13;bitsize:64;offset:104;encoding:uint;format:hex;set:General Purpose Registers;gcc:13;dwarf:13;#00
+send packet: $qRegisterInfoe#00
+read packet: $name:r14;bitsize:64;offset:112;encoding:uint;format:hex;set:General Purpose Registers;gcc:14;dwarf:14;#00
+send packet: $qRegisterInfof#00
+read packet: $name:r15;bitsize:64;offset:120;encoding:uint;format:hex;set:General Purpose Registers;gcc:15;dwarf:15;#00
+send packet: $qRegisterInfo10#00
+read packet: $name:rip;alt-name:pc;bitsize:64;offset:128;encoding:uint;format:hex;set:General Purpose Registers;gcc:16;dwarf:16;generic:pc;#00
+send packet: $qRegisterInfo11#00
+read packet: $name:rflags;alt-name:flags;bitsize:64;offset:136;encoding:uint;format:hex;set:General Purpose Registers;#00
+send packet: $qRegisterInfo12#00
+read packet: $name:cs;bitsize:64;offset:144;encoding:uint;format:hex;set:General Purpose Registers;#00
+send packet: $qRegisterInfo13#00
+read packet: $name:fs;bitsize:64;offset:152;encoding:uint;format:hex;set:General Purpose Registers;#00
+send packet: $qRegisterInfo14#00
+read packet: $name:gs;bitsize:64;offset:160;encoding:uint;format:hex;set:General Purpose Registers;#00
+send packet: $qRegisterInfo15#00
+read packet: $name:fctrl;bitsize:16;offset:176;encoding:uint;format:hex;set:Floating Point Registers;#00
+send packet: $qRegisterInfo16#00
+read packet: $name:fstat;bitsize:16;offset:178;encoding:uint;format:hex;set:Floating Point Registers;#00
+send packet: $qRegisterInfo17#00
+read packet: $name:ftag;bitsize:8;offset:180;encoding:uint;format:hex;set:Floating Point Registers;#00
+send packet: $qRegisterInfo18#00
+read packet: $name:fop;bitsize:16;offset:182;encoding:uint;format:hex;set:Floating Point Registers;#00
+send packet: $qRegisterInfo19#00
+read packet: $name:fioff;bitsize:32;offset:184;encoding:uint;format:hex;set:Floating Point Registers;#00
+send packet: $qRegisterInfo1a#00
+read packet: $name:fiseg;bitsize:16;offset:188;encoding:uint;format:hex;set:Floating Point Registers;#00
+send packet: $qRegisterInfo1b#00
+read packet: $name:fooff;bitsize:32;offset:192;encoding:uint;format:hex;set:Floating Point Registers;#00
+send packet: $qRegisterInfo1c#00
+read packet: $name:foseg;bitsize:16;offset:196;encoding:uint;format:hex;set:Floating Point Registers;#00
+send packet: $qRegisterInfo1d#00
+read packet: $name:mxcsr;bitsize:32;offset:200;encoding:uint;format:hex;set:Floating Point Registers;#00
+send packet: $qRegisterInfo1e#00
+read packet: $name:mxcsrmask;bitsize:32;offset:204;encoding:uint;format:hex;set:Floating Point Registers;#00
+send packet: $qRegisterInfo1f#00
+read packet: $name:stmm0;bitsize:80;offset:208;encoding:vector;format:vector-uint8;set:Floating Point Registers;gcc:33;dwarf:33;#00
+send packet: $qRegisterInfo20#00
+read packet: $name:stmm1;bitsize:80;offset:224;encoding:vector;format:vector-uint8;set:Floating Point Registers;gcc:34;dwarf:34;#00
+send packet: $qRegisterInfo21#00
+read packet: $name:stmm2;bitsize:80;offset:240;encoding:vector;format:vector-uint8;set:Floating Point Registers;gcc:35;dwarf:35;#00
+send packet: $qRegisterInfo22#00
+read packet: $name:stmm3;bitsize:80;offset:256;encoding:vector;format:vector-uint8;set:Floating Point Registers;gcc:36;dwarf:36;#00
+send packet: $qRegisterInfo23#00
+read packet: $name:stmm4;bitsize:80;offset:272;encoding:vector;format:vector-uint8;set:Floating Point Registers;gcc:37;dwarf:37;#00
+send packet: $qRegisterInfo24#00
+read packet: $name:stmm5;bitsize:80;offset:288;encoding:vector;format:vector-uint8;set:Floating Point Registers;gcc:38;dwarf:38;#00
+send packet: $qRegisterInfo25#00
+read packet: $name:stmm6;bitsize:80;offset:304;encoding:vector;format:vector-uint8;set:Floating Point Registers;gcc:39;dwarf:39;#00
+send packet: $qRegisterInfo26#00
+read packet: $name:stmm7;bitsize:80;offset:320;encoding:vector;format:vector-uint8;set:Floating Point Registers;gcc:40;dwarf:40;#00
+send packet: $qRegisterInfo27#00
+read packet: $name:xmm0;bitsize:128;offset:336;encoding:vector;format:vector-uint8;set:Floating Point Registers;gcc:17;dwarf:17;#00
+send packet: $qRegisterInfo28#00
+read packet: $name:xmm1;bitsize:128;offset:352;encoding:vector;format:vector-uint8;set:Floating Point Registers;gcc:18;dwarf:18;#00
+send packet: $qRegisterInfo29#00
+read packet: $name:xmm2;bitsize:128;offset:368;encoding:vector;format:vector-uint8;set:Floating Point Registers;gcc:19;dwarf:19;#00
+send packet: $qRegisterInfo2a#00
+read packet: $name:xmm3;bitsize:128;offset:384;encoding:vector;format:vector-uint8;set:Floating Point Registers;gcc:20;dwarf:20;#00
+send packet: $qRegisterInfo2b#00
+read packet: $name:xmm4;bitsize:128;offset:400;encoding:vector;format:vector-uint8;set:Floating Point Registers;gcc:21;dwarf:21;#00
+send packet: $qRegisterInfo2c#00
+read packet: $name:xmm5;bitsize:128;offset:416;encoding:vector;format:vector-uint8;set:Floating Point Registers;gcc:22;dwarf:22;#00
+send packet: $qRegisterInfo2d#00
+read packet: $name:xmm6;bitsize:128;offset:432;encoding:vector;format:vector-uint8;set:Floating Point Registers;gcc:23;dwarf:23;#00
+send packet: $qRegisterInfo2e#00
+read packet: $name:xmm7;bitsize:128;offset:448;encoding:vector;format:vector-uint8;set:Floating Point Registers;gcc:24;dwarf:24;#00
+send packet: $qRegisterInfo2f#00
+read packet: $name:xmm8;bitsize:128;offset:464;encoding:vector;format:vector-uint8;set:Floating Point Registers;gcc:25;dwarf:25;#00
+send packet: $qRegisterInfo30#00
+read packet: $name:xmm9;bitsize:128;offset:480;encoding:vector;format:vector-uint8;set:Floating Point Registers;gcc:26;dwarf:26;#00
+send packet: $qRegisterInfo31#00
+read packet: $name:xmm10;bitsize:128;offset:496;encoding:vector;format:vector-uint8;set:Floating Point Registers;gcc:27;dwarf:27;#00
+send packet: $qRegisterInfo32#00
+read packet: $name:xmm11;bitsize:128;offset:512;encoding:vector;format:vector-uint8;set:Floating Point Registers;gcc:28;dwarf:28;#00
+send packet: $qRegisterInfo33#00
+read packet: $name:xmm12;bitsize:128;offset:528;encoding:vector;format:vector-uint8;set:Floating Point Registers;gcc:29;dwarf:29;#00
+send packet: $qRegisterInfo34#00
+read packet: $name:xmm13;bitsize:128;offset:544;encoding:vector;format:vector-uint8;set:Floating Point Registers;gcc:30;dwarf:30;#00
+send packet: $qRegisterInfo35#00
+read packet: $name:xmm14;bitsize:128;offset:560;encoding:vector;format:vector-uint8;set:Floating Point Registers;gcc:31;dwarf:31;#00
+send packet: $qRegisterInfo36#00
+read packet: $name:xmm15;bitsize:128;offset:576;encoding:vector;format:vector-uint8;set:Floating Point Registers;gcc:32;dwarf:32;#00
+send packet: $qRegisterInfo37#00
+read packet: $name:trapno;bitsize:32;offset:696;encoding:uint;format:hex;set:Exception State Registers;#00
+send packet: $qRegisterInfo38#00
+read packet: $name:err;bitsize:32;offset:700;encoding:uint;format:hex;set:Exception State Registers;#00
+send packet: $qRegisterInfo39#00
+read packet: $name:faultvaddr;bitsize:64;offset:704;encoding:uint;format:hex;set:Exception State Registers;#00
+send packet: $qRegisterInfo3a#00
+read packet: $E45#00
+```
+
+As we see above we keep making subsequent calls to the remote server to
+discover all registers by increasing the number appended to `qRegisterInfo` and
+we get a response back that is a series of `key=value;` strings.
+
+The `offset:` fields should not leave a gap anywhere in the g/G packet -- the
+register values should be appended one after another.  For instance, if the
+register context for a thread looks like:
+```
+struct rctx {
+    uint32_t gpr1;  // offset 0
+    uint32_t gpr2;  // offset 4
+    uint32_t gpr3;  // offset 8
+    uint64_t fp1;   // offset 16
+};
+```
+
+You may end up with a 4-byte gap between gpr3 and fp1 on architectures
+that align values like this.  The correct offset: value for fp1 is 12 -
+in the g/G packet fp1 will immediately follow gpr3, even though the
+in-memory thread structure has an empty 4 bytes for alignment between
+these two registers.
+
+The keys and values are detailed below:
+
+* `name` -
+  The primary register name as a string ("rbp" for example)
+* `alt-name` -
+  An alternate name for a register as a string ("fp" for example
+  for the above "rbp")
+* `bitsize` - Size in bits of a register (32, 64, etc).  Base 10.
+* `offset` -
+  The offset within the "g" and "G" packet of the register data for
+  this register.  This is the byte offset once the data has been
+  transformed into binary, not the character offset into the g/G
+  packet.  Base 10.
+* `encoding` -
+  The encoding type of the register which must be one of:
+  * `uint` (unsigned integer)
+  * `sint` (signed integer)
+  * `ieee754` (IEEE 754 float)
+  * `vector` (vector register)
+* format -
+  The preferred format for display of this register. The value must be one of:
+  * `binary`
+  * `decimal`
+  * `hex`
+  * `float`
+  * `vector-sint8`
+  * `vector-uint8`
+  * `vector-sint16`
+  * `vector-uint16`
+  * `vector-sint32`
+  * `vector-uint32`
+  * `vector-float32`
+  * `vector-uint128`
+* `set`-
+  The register set name as a string that this register belongs to.
+* `gcc` -
+  The GCC compiler registers number for this register (used for
+  EH frame and other compiler information that is encoded in the
+  executable files). The supplied number will be decoded like a
+  string passed to strtoul() with a base of zero, so the number
+  can be decimal, or hex if it is prefixed with "0x".
+
+  **Note:** If the compiler doesn't have a register number for this
+  register, this key/value pair should be omitted.
+* `dwarf` -
+  The DWARF register number for this register that is used for this
+  register in the debug information. The supplied number will be decoded
+  like a string passed to strtoul() with a base of zero, so the number
+  can be decimal, or hex if it is prefixed with "0x".
+
+  **Note:** If the compiler doesn't have a register number for this
+  register, this key/value pair should be omitted.
+* `generic` -
+  If the register is a generic register that most CPUs have, classify
+  it correctly so the debugger knows. Valid values are one of:
+  * `pc` (a program counter register. for example `name=eip;` (i386),
+    `name=rip;` (x86_64), `name=r15;` (32 bit arm) would
+    include a `generic=pc;` key value pair)
+  * `sp` (a stack pointer register. for example `name=esp;` (i386),
+    `name=rsp;` (x86_64), `name=r13;` (32 bit arm) would
+    include a `generic=sp;` key value pair)
+  * `fp` (a frame pointer register. for example `name=ebp;` (i386),
+    `name=rbp;` (x86_64), `name=r7;` (32 bit arm with macosx
+    ABI) would include a `generic=fp;` key value pair)
+  * `ra` (a return address register. for example `name=lr;` (32 bit ARM)
+    would include a `generic=ra;` key value pair)
+  * `flags` (a CPU flags register. for example `name=eflags;` (i386),
+    `name=rflags;` (x86_64), `name=cpsr;` (32 bit ARM)
+    would include a `generic=flags;` key value pair)
+  * `arg1` - `arg8` (specified for registers that contain function
+    arguments when the argument fits into a register)
+* `container-regs` -
+  The value for this key is a comma separated list of raw hex (optional
+  leading "0x") register numbers.
+
+  This specifies that this register is contained in other concrete
+  register values. For example "eax" is in the lower 32 bits of the
+  "rax" register value for x86_64, so "eax" could specify that it is
+  contained in "rax" by specifying the register number for "rax" (whose
+  register number is 0x00):
+  ```
+  container-regs:00;
+  ```
+  If a register is comprised of one or more registers, like "d0" is ARM
+  which is a 64 bit register, it might be made up of "s0" and "s1". If
+  the register number for "s0" is 0x20, and the register number of "s1"
+  is "0x21", the "container-regs" key/value pair would be:
+  ```
+  container-regs:20,21;
+  ```
+  This is handy for defining what GDB used to call "pseudo" registers.
+  These registers are never requested by LLDB via the register read
+  or write packets, the container registers will be requested on behalf
+  of this register.
+* `invalidate-regs` -
+  The value for this key is a comma separated list of raw hex (optional
+  leading "0x") register numbers.
+
+  This specifies which register values should be invalidated when this
+  register is modified. For example if modifying "eax" would cause "rax",
+  "eax", "ax", "ah", and "al" to be modified where rax is 0x0, eax is 0x15,
+  ax is 0x25, ah is 0x35, and al is 0x39, the "invalidate-regs" key/value
+  pair would be:
+  ```
+  invalidate-regs:0,15,25,35,39;
+  ```
+  If there is a single register that gets invalidated, then omit the comma
+  and just list a single register:
+  ```
+  invalidate-regs:0;
+  ```
+  This is handy when modifying a specific register can cause other
+  register values to change. For example, when debugging an ARM target,
+  modifying the CPSR register can cause the r8 - r14 and cpsr value to
+  change depending on if the mode has changed.
+
+### Priority To Implement
+
+High. Any target that can self describe its registers, should do so.
+This means if new registers are ever added to a remote target, they
+will get picked up automatically, and allows registers to change
+depending on the actual CPU type that is used.
+
+NB: `qRegisterInfo` is deprecated in favor of the standard gdb remote
+serial protocol register description method,
+`qXfer:features:read:target.xml`.
+If `qXfer:features:read:target.xml` is supported, `qRegisterInfo` does
+not need to be implemented.  The target.xml format is used by most
+gdb RSP stubs whereas `qRegisterInfo` was an lldb-only design.
+`qRegisterInfo` requires one packet per register and can have undesirable
+performance costs at the start of a debug session, whereas target.xml
+may be able to describe all registers in a single packet.
+
+## qPlatform_shell
+
+Run a command in a shell on the connected remote machine.
+
+The request consists of the command to be executed encoded in ASCII characters
+converted into hex bytes.
+
+The response to this packet consists of the letter F followed by the return code,
+followed by the signal number (or 0 if no signal was delivered), and escaped bytes
+of captured program output.
+
+Below is an example communication from a client sending an "ls -la" command:
+```
+send packet: $qPlatform_shell:6c73202d6c61,00000002#ec
+read packet: $F,00000000,00000000,total 4736
+drwxrwxr-x 16 username groupname    4096 Aug 15 21:36 .
+drwxr-xr-x 17 username groupname    4096 Aug 10 16:39 ..
+-rw-rw-r--  1 username groupname   73875 Aug 12 16:46 notes.txt
+drwxrwxr-x  5 username groupname    4096 Aug 15 21:36 source.cpp
+-rw-r--r--  1 username groupname    2792 Aug 12 16:46 a.out
+-rw-r--r--  1 username groupname    3190 Aug 12 16:46 Makefile
+```
+
+### Priority To Implement
+
+High. This command allows LLDB clients to run arbitrary shell
+commands on a remote host.
+
+## qPlatform_mkdir
+
+Creates a new directory on the connected remote machine.
+
+Request: `qPlatform_mkdir:<hex-file-mode>,<ascii-hex-path>`
+
+The request packet has the fields:
+   1. mode bits in base 16
+   2. file path in ascii-hex encoding
+
+Reply: 
+  * `F<mkdir-return-code>`
+    (mkdir called successfully and returned with the given return code)
+  * `Exx` (An error occurred)
+
+### Priority To Implement
+
+Low. This command allows LLDB clients to create new directories on
+a remote host.
+
+## vFile:chmod / qPlatform_chmod
+
+Change the permissions of a file on the connected remote machine.
+
+Request: `qPlatform_chmod:<hex-file-mode>,<ascii-hex-path>`
+
+Reply:
+* `F<chmod-return-code>`
+  (chmod called successfully and returned with the given return code)
+* `Exx` (An error occurred)
+
+### Priority To Implement
+
+Low.
+
+## qHostInfo
+
+Get information about the host we are remotely connected to.
+
+LLDB supports a host info call that gets all sorts of details of the system
+that is being debugged:
+```
+send packet: $qHostInfo#00
+read packet: $cputype:16777223;cpusubtype:3;ostype:darwin;vendor:apple;endian:little;ptrsize:8;#00
+```
+
+Key value pairs are one of:
+* `cputype`: is a number that is the mach-o CPU type that is being debugged (base 10)
+* `cpusubtype`: is a number that is the mach-o CPU subtype type that is being debugged (base 10)
+* `triple`: a string for the target triple (x86_64-apple-macosx) that can be used to specify arch + vendor + os in one entry
+* `vendor`: a string for the vendor (apple), not needed if "triple" is specified
+* `ostype`: a string for the OS being debugged (macosx, linux, freebsd, ios, watchos), not needed if "triple" is specified
+* `endian`: is one of "little", "big", or "pdp"
+* `ptrsize`: an unsigned number that represents how big pointers are in bytes on the debug target
+* `hostname`: the hostname of the host that is running the GDB server if available
+* `os_build`: a string for the OS build for the remote host as a string value
+* `os_kernel`: a string describing the kernel version
+* `os_version`: a version string that represents the current OS version (10.8.2)
+* `watchpoint_exceptions_received`: one of "before" or "after" to specify if a watchpoint is triggered before or after the pc when it stops
+* `default_packet_timeout`: an unsigned number that specifies the default timeout in seconds
+* `distribution_id`: optional. For linux, specifies distribution id (e.g. ubuntu, fedora, etc.)
+* `osmajor`: optional, specifies the major version number of the OS (e.g. for macOS 10.12.2, it would be 10)
+* `osminor`: optional, specifies the minor version number of the OS (e.g. for macOS 10.12.2, it would be 12)
+* `ospatch`: optional, specifies the patch level number of the OS (e.g. for macOS 10.12.2, it would be 2)
+* `vm-page-size`: optional, specifies the target system VM page size, base 10.
+  Needed for the "dirty-pages:" list in the qMemoryRegionInfo
+  packet, where a list of dirty pages is sent from the remote
+  stub.  This page size tells lldb how large each dirty page is.
+* `addressing_bits`: optional, specifies how many bits in addresses are
+	significant for addressing, base 10.  If bits 38..0
+	in a 64-bit pointer are significant for addressing,
+	then the value is 39.  This is needed on e.g. AArch64
+	v8.3 ABIs that use pointer authentication, so lldb
+	knows which bits to clear/set to get the actual
+	addresses.
+* `low_mem_addressing_bits`: optional, specifies how many bits in
+  addresses in low memory are significant for addressing, base 10.
+  AArch64 can have different page table setups for low and high
+  memory, and therefore a different number of bits used for addressing.
+* `high_mem_addressing_bits`: optional, specifies how many bits in
+  addresses in high memory are significant for addressing, base 10.
+  AArch64 can have different page table setups for low and high
+  memory, and therefore a different number of bits used for addressing.
+
+### Priority To Implement
+
+High. This packet is usually very easy to implement and can help
+LLDB select the correct plug-ins for the job based on the target
+triple information that is supplied.
+
+## qGDBServerVersion
+
+Get version information about this implementation of the gdb-remote
+protocol.
+
+The goal of this packet is to provide enough information about an
+implementation of the gdb-remote-protocol server that lldb can
+work around implementation problems that are discovered after the
+version has been released/deployed.  The name and version number
+should be sufficiently unique that lldb can unambiguously identify
+the origin of the program (for instance, debugserver from lldb) and
+the version/submission number/patch level of the program - whatever
+is appropriate for your server implementation.
+
+The packet follows the key-value pair model, semicolon separated.
+```
+send packet: $qGDBServerVersion#00
+read packet: $name:debugserver;version:310.2;#00
+```
+
+Other clients may find other key-value pairs to be useful for identifying
+a gdb stub.  Patch level, release name, build number may all be keys that
+better describe your implementation's version.
+
+Suggested key names:
+* `name`: the name of your remote server - "debugserver" is the lldb standard
+          implementation
+* `version`: identifies the version number of this server
+* `patch_level`: the patch level of this server
+* `release_name`: the name of this release, if your project uses names
+* `build_number`: if you use a build system with increasing build numbers,
+                  this may be the right key name for your server
+* `major_version`: major version number
+* `minor_version`: minor version number
+
+### Priority To Implement
+
+High. This packet is usually very easy to implement and can help
+LLDB to work around bugs in a server's implementation when they
+are found.
+
+## qProcessInfo
+
+Get information about the process we are currently debugging.
+
+### Priority To Implement
+
+Medium.  On systems which can launch multiple different architecture processes,
+the qHostInfo may not disambiguate sufficiently to know what kind of
+process is being debugged.
+
+For example on a 64-bit x86 Mac system both 32-bit and 64-bit user processes are possible,
+and with Mach-O universal files, the executable file may contain both 32- and
+64-bit slices so it may be impossible to know until you're attached to a real
+process to know what you're working with.
+
+All numeric fields return base 16 numbers without any "0x" prefix.
+
+An i386 process:
+```
+send packet: $qProcessInfo#00
+read packet: $pid:42a8;parent-pid:42bf;real-uid:ecf;real-gid:b;effective-uid:ecf;effective-gid:b;cputype:7;cpusubtype:3;ostype:macosx;vendor:apple;endian:little;ptrsize:4;#00
+```
+
+An x86_64 process:
+```
+send packet: $qProcessInfo#00
+read packet: $pid:d22c;parent-pid:d34d;real-uid:ecf;real-gid:b;effective-uid:ecf;effective-gid:b;cputype:1000007;cpusubtype:3;ostype:macosx;vendor:apple;endian:little;ptrsize:8;#00
+```
+
+Key value pairs include:
+* `pid`: the process id
+* `parent-pid`: the process of the parent process (often debugserver will become the parent when attaching)
+* `real-uid`: the real user id of the process
+* `real-gid`: the real group id of the process
+* `effective-uid`: the effective user id of the process
+* `effective-gid`: the effective group id of the process
+* `cputype`: the Mach-O CPU type of the process  (base 16)
+* `cpusubtype`: the Mach-O CPU subtype of the process  (base 16)
+* `ostype`: is a string the represents the OS being debugged (darwin, linux, freebsd)
+* `vendor`: is a string that represents the vendor (apple)
+* `endian`: is one of "little", "big", or "pdp"
+* `ptrsize`: is a number that represents how big pointers are in bytes
+* `main-binary-uuid`: is the UUID of a firmware type binary that the gdb stub knows about
+* `main-binary-address`: is the load address of the firmware type binary
+* `main-binary-slide`: is the slide of the firmware type binary, if address isn't known
+* `binary-addresses`: A comma-separated list of binary load addresses base 16.
+                      lldb will parse the binaries in memory to get UUIDs, then
+                      try to find the binaries & debug info by UUID.  Intended for
+                      use with a small number of firmware type binaries where the
+                      search for binary/debug info may be expensive.
+
+## qShlibInfoAddr
+
+Get an address where the dynamic linker stores information about
+where shared libraries are loaded.
+
+LLDB and GDB both support the `qShlibInfoAddr` packet which is a hint to each
+debugger as to where to find the dynamic loader information. For darwin
+binaries that run in user land this is the address of the `all_image_infos`
+structure in the `/usr/lib/dyld` executable, or the result of a `TASK_DYLD_INFO`
+call. The result is returned as big endian hex bytes that are the address
+value:
+```
+send packet: $qShlibInfoAddr#00
+read packet: $7fff5fc40040#00
+```
+
+### Priority To Implement
+
+High if you have a dynamic loader plug-in in LLDB for your target
+triple (see the "qHostInfo" packet) that can use this information.
+Many times address load randomization can make it hard to detect
+where the dynamic loader binary and data structures are located and
+some platforms know, or can find out where this information is.
+
+Low if you have a debug target where all object and symbol files
+contain static load addresses.
+
+## qThreadStopInfo\<tid\>
+
+Get information about why a thread, whose ID is `<tid>`, is stopped.
+
+LLDB tries to use the `qThreadStopInfo` packet which is formatted as
+`qThreadStopInfo%x` where `%x` is the hex thread ID. This requests information
+about why a thread is stopped. The response is the same as the stop reply
+packets and tells us what happened to the other threads. The standard GDB
+remote packets love to think that there is only _one_ reason that _one_ thread
+stops at a time. This allows us to see why all threads stopped and allows us
+to implement better multi-threaded debugging support.
+
+### Priority To Implement
+
+High if you need to support multi-threaded or multi-core debugging.
+Many times one thread will hit a breakpoint and while the debugger
+is in the process of suspending the other threads, other threads
+will also hit a breakpoint. This packet allows LLDB to know why all
+threads (live system debug) / cores (JTAG) in your program have
+stopped and allows LLDB to display and control your program
+correctly.
+
+## QThreadSuffixSupported
+
+Try to enable thread suffix support for the `g`, `G`, `p`, and `P` packets.
+
+When reading thread registers, you currently need to set the current
+thread, then read the registers. This is kind of cumbersome, so we added the
+ability to query if the remote GDB server supports adding a `thread:<tid>;`
+suffix to all packets that request information for a thread. To test if the
+remote GDB server supports this feature:
+```
+send packet: $QThreadSuffixSupported#00
+read packet: OK
+```
+
+If `OK` is returned, then the `g`, `G`, `p` and `P` packets can accept a
+thread suffix. So to send a `g` packet (read all register values):
+```
+send packet: $g;thread:<tid>;#00
+read packet: ....
+
+send packet: $G;thread:<tid>;#00
+read packet: ....
+
+send packet: $p1a;thread:<tid>;#00
+read packet: ....
+
+send packet: $P1a=1234abcd;thread:<tid>;#00
+read packet: ....
+```
+
+otherwise, without this you would need to always send two packets:
+```
+send packet: $Hg<tid>#00
+read packet: ....
+send packet: $g#00
+read packet: ....
+```
+
+We also added support for allocating and deallocating memory. We use this to
+allocate memory so we can run JITed code.
+
+### Priority To Implement
+
+High. Adding a thread suffix allows us to read and write registers
+more efficiently and stops us from having to select a thread with
+one packet and then read registers with a second packet. It also
+makes sure that no errors can occur where the debugger thinks it
+already has a thread selected (see the `Hg` packet from the standard
+GDB remote protocol documentation) yet the remote GDB server actually
+has another thread selected.
+
+## _M\<size\>,\<permissions\>
+
+Allocate memory on the remote target with the specified size and
+permissions.
+
+The allocate memory packet starts with `_M<size>,<permissions>`. It returns a
+raw big endian address value, or an empty response for unimplemented, or `EXX` for an error
+code. The packet is formatted as:
+```
+char packet[256];
+int packet_len;
+packet_len = ::snprintf (
+    packet,
+    sizeof(packet),
+    "_M%zx,%s%s%s",
+    (size_t)size,
+    permissions & lldb::ePermissionsReadable ? "r" : "",
+    permissions & lldb::ePermissionsWritable ? "w" : "",
+    permissions & lldb::ePermissionsExecutable ? "x" : "");
+```
+
+You request a size and give the permissions. This packet does NOT need to be
+implemented if you don't want to support running JITed code. The return value
+is just the address of the newly allocated memory as raw big endian hex bytes.
+
+### Priority To Implement
+
+High if you want LLDB to be able to JIT code and run that code. JIT
+code also needs data which is also allocated and tracked.
+
+Low if you don't support running JIT'ed code.
+
+## _m\<addr\>
+
+Deallocate memory that was previously allocated using an allocate
+memory pack.
+
+The deallocate memory packet is `_m<addr>` where you pass in the address you
+got back from a previous call to the allocate memory packet. It returns `OK`
+if the memory was successfully deallocated, or `EXX`" for an error, or an
+empty response if not supported.
+
+### Priority To Implement
+
+High if you want LLDB to be able to JIT code and run that code. JIT
+code also needs data which is also allocated and tracked.
+
+Low if you don't support running JIT'ed code.
+
+## qMemoryRegionInfo:\<addr\>
+
+Get information about the address range that contains `<addr>`.
+
+We added a way to get information for a memory region. The packet is:
+```
+qMemoryRegionInfo:<addr>
+```
+
+Where `<addr>` is a big endian hex address. The response is returned in a series
+of tuples like the data returned in a stop reply packet. The currently valid
+tuples to return are:
+* `start:<start-addr>;` - `<start-addr>` is a big endian hex address that is
+                          the start address of the range that contains `<addr>`
+* `size:<size>;` - `<size>` is a big endian hex byte size of the address
+                   of the range that contains `<addr>`
+* `permissions:<permissions>;` - `<permissions>` is a string that contains one
+                                 or more of the characters from `rwx`
+* `name:<name>;` - `<name>` is a hex encoded string that contains the name of
+                   the memory region mapped at the given address. In case of
+                   regions backed by a file it have to be the absolute path of
+                   the file while for anonymous regions it have to be the name
+                   associated to the region if that is available.
+* `flags:<flags-string>;` - where `<flags-string>` is a space separated string
+                            of flag names. Currently the only supported flag
+                            is `mt` for AArch64 memory tagging. lldb will
+                            ignore any other flags in this field.
+* `type:[<type>][,<type>];` - memory types that apply to this region, e.g.
+                              `stack` for stack memory.
+* `error:<ascii-byte-error-string>;` - where `<ascii-byte-error-string>` is
+                                       a hex encoded string value that
+                                       contains an error string
+* `dirty-pages:[<hexaddr>][,<hexaddr];` -
+  A list of memory pages within this
+  region that are "dirty" -- they have been modified.
+  Page addresses are in base 16. The size of a page can
+  be found from the `qHostInfo`'s `page-size` key-value.
+  
+  If the stub supports identifying dirty pages within a
+  memory region, this key should always be present for all
+  `qMemoryRegionInfo` replies.  This key with no pages
+  listed (`dirty-pages:;`) indicates no dirty pages in
+  this memory region.  The *absence* of this key means
+  that this stub cannot determine dirty pages.
+
+If the address requested is not in a mapped region (e.g. we've jumped through
+a NULL pointer and are at 0x0) currently lldb expects to get back the size
+of the unmapped region -- that is, the distance to the next valid region.
+For instance, with a macOS process which has nothing mapped in the first
+4GB of its address space, if we're asking about address 0x2:
+```
+  qMemoryRegionInfo:2
+  start:2;size:fffffffe;
+```
+
+The lack of `permissions:` indicates that none of read/write/execute are valid
+for this region.
+
+### Priority To Implement
+
+Medium. This is nice to have, but it isn't necessary. It helps LLDB
+do stack unwinding when we branch into memory that isn't executable.
+If we can detect that the code we are stopped in isn't executable,
+then we can recover registers for stack frames above the current
+frame. Otherwise we must assume we are in some JIT'ed code (not JIT
+code that LLDB has made) and assume that no registers are available
+in higher stack frames.
+
+## "x" - Binary memory read
+
+Like the `m` (read) and `M` (write) packets, this is a partner to the
+`X` (write binary data) packet, `x`.
+
+It is called like
+```
+xADDRESS,LENGTH
+```
+
+where both `ADDRESS` and `LENGTH` are big-endian base 16 values.
+
+To test if this packet is available, send a addr/len of 0:
+```
+x0,0
+```
+You will get an `OK` response if it is supported.
+
+The reply will be the data requested in 8-bit binary data format.
+The standard quoting is applied to the payload. Characters `}  #  $  *`
+will all be escaped with `}` (`0x7d`) character and then XOR'ed with `0x20`.
+
+A typical use to read 512 bytes at 0x1000 would look like:
+```
+x0x1000,0x200
+```
+The `0x` prefixes are optional - like most of the gdb-remote packets,
+omitting them will work fine; these numbers are always base 16.
+
+The length of the payload is not provided.  A reliable, 8-bit clean,
+transport layer is assumed.
+
+## Detach and stay stopped
+
+We extended the "D" packet to specify that the monitor should keep the
+target suspended on detach.  The normal behavior is to resume execution
+on detach.  We will send:
+```
+qSupportsDetachAndStayStopped:
+```
+
+to query whether the monitor supports the extended detach, and if it does,
+when we want the monitor to detach but not resume the target, we will
+send:
+```
+D1
+```
+In any case, if we want the normal detach behavior we will just send:
+```
+D
+```
+
+## QSaveRegisterState / QSaveRegisterState;thread:XXXX;
+
+The `QSaveRegisterState` packet tells the remote debugserver to save
+all registers and return a non-zero unique integer ID that
+represents these save registers. If thread suffixes are enabled the
+second form of this packet is used, otherwise the first form is
+used. This packet is called prior to executing an expression, so
+the remote GDB server should do anything it needs to in order to
+ensure the registers that are saved are correct. On macOS this
+involves calling `thread_abort_safely(mach_port_t thread)` to
+ensure we get the correct registers for a thread in case it is
+currently having code run on its behalf in the kernel.
+
+### Response
+
+* `<unsigned int>` - The save_id result is a non-zero unsigned integer value
+                 that can be passed back to the GDB server using a
+                 `QRestoreRegisterState` packet to restore the registers
+                 one time.
+* `EXX` - or an error code in the form of `EXX` where `XX` is a
+          hex error code.
+
+### Priority To Implement
+
+Low, this is mostly a convenience packet to avoid having to send all
+registers with a `g` packet. It should only be implemented if support
+for the `QRestoreRegisterState` is added.
+
+## QRestoreRegisterState:\<save_id\> / QRestoreRegisterState:\<save_id\>;thread:XXXX;
+
+The `QRestoreRegisterState` packet tells the remote debugserver to
+restore all registers using the `save_id` which is an unsigned
+integer that was returned from a previous call to
+`QSaveRegisterState`. The restoration process can only be done once
+as the data backing the register state will be freed upon the
+completion of the `QRestoreRegisterState` command.
+
+If thread suffixes are enabled the second form of this packet is
+used, otherwise the first form is used.
+
+### Response
+
+* `OK` - if all registers were successfully restored
+* `EXX` - for any errors
+
+### Priority To Implement
+
+Low, this is mostly a convenience packet to avoid having to send all
+registers with a `g` packet. It should only be implemented if support
+for the `QSaveRegisterState` is added.
+
+## qFileLoadAddress:\<file_path\>
+
+Get the load address of a memory mapped file.
+The load address is defined as the address of the first memory
+region what contains data mapped from the specified file.
+
+### Response
+
+* `<unsigned-hex64>` - Load address of the file in big endian encoding
+* `E01` - the requested file isn't loaded
+* `EXX` - for any other errors
+
+### Priority To Implement
+
+Low, required if dynamic linker don't fill in the load address of
+some object file in the rendezvous data structure.
+
+## qModuleInfo:\<module_path\>;\<arch triple\>
+
+Get information for a module by given module path and architecture.
+
+### Response
+
+* `(uuid|md5):...;triple:...;file_offset:...;file_size...;`
+* `EXX` - for any errors
+
+### Priority To Implement
+
+Optional, required if dynamic loader cannot fetch module's information like
+UUID directly from inferior's memory.
+
+## jModulesInfo:[{"file":"...",triple:"..."}, ...]
+
+Get information for a list of modules by given module path and
+architecture.
+
+### Response
+
+A JSON array of dictionaries containing the following keys:
+* `uuid`
+* `triple`
+* `file_path`
+* `file_offset`
+* `file_size`
+
+The meaning of the fields is the same as in the `qModuleInfo` packet. The server
+signals the failure to retrieve the module info for a file by ommiting the
+corresponding array entry from the response. The server may also
+include entries the client did not ask for, if it has reason to
+the modules will be interesting to the client.
+
+### Priority To Implement
+
+Optional. If not implemented, `qModuleInfo` packet will be used, which
+may be slower if the target contains a large number of modules and
+the communication link has a non-negligible latency.
+
+## Stop reply packet extensions
+
+This section describes some of the additional information you can
+specify in stop reply packets that help LLDB to know more detailed
+information about your threads.
+
+Standard GDB remote stop reply packets are reply packets sent in
+response to a packet  that made the program run. They come in the
+following forms:
+
+* `SAA` -
+  `S` means signal and `AA` is a hex signal number that describes why
+  the thread or stopped. It doesn't specify which thread, so the `T`
+  packet is recommended to use instead of the `S` packet.
+
+* `TAAkey1:value1;key2:value2;...` -
+  `T` means a thread stopped due to a unix signal where `AA` is a hex
+  signal number that describes why the program stopped. This is
+  followed by a series of key/value pairs:
+    * If key is a hex number, it is a register number and value is
+      the hex value of the register in debuggee endian byte order.
+    * If key == "thread", then the value is the big endian hex
+      thread-id of the stopped thread.
+    * If key == "core", then value is a hex number of the core on
+      which the stop was detected.
+    * If key == "watch" or key == "rwatch" or key == "awatch", then
+      value is the data address in big endian hex
+    * If key == "library", then value is ignore and "qXfer:libraries:read"
+      packets should be used to detect any newly loaded shared libraries
+
+* `WAA` - `W` means the process exited and `AA` is the exit status.
+
+* `XAA` - `X` means the process exited and `AA` is signal that caused the program
+  to exit.
+
+* `O<ascii-hex-string>` - `O` means `STDOUT` has data that was written to its
+  console and is being delivered to the debugger. This packet happens asynchronously
+  and the debugger is expected to continue to wait for another stop reply
+  packet.
+
+### Lldb Extensions
+
+We have extended the `T` packet to be able to also understand the
+following keys and values:
+
+* `metype` - `unsigned` -
+  mach exception type (the value of the `EXC_XXX` enumerations)
+  as an unsigned integer. For targets with mach
+  kernels only.
+* `mecount` - `unsigned` -
+  mach exception data count as an unsigned integer
+  For targets with mach kernels only.
+* `medata` - `unsigned` -
+  There should be `mecount` of these and it is the data
+  that goes along with a mach exception (as an unsigned
+  integer). For targets with mach kernels only.
+* `name` - `string` -
+  The name of the thread as a plain string. The string
+  must not contain an special packet characters or
+  contain a `:` or a `;`. Use `hexname` if the thread
+  name has special characters.
+* `hexname` - `ascii-hex` -  An ASCII hex string that contains the name of the thread
+* `qaddr` - `hex` -
+  Big endian hex value that contains the `libdispatch`
+  queue address for the queue of the thread.
+* `reason` - `enum` - The enumeration must be one of:
+  * `trace` -
+    the program stopped after a single instruction
+    was executed on a core. Usually done when single
+    stepping past a breakpoint
+  * `breakpoint` - a breakpoint set using a `z` packet was hit.
+  * `trap` - stopped due to user interruption
+  * `signal` -
+    stopped due to an actual unix signal, not
+    just the debugger using a unix signal to keep
+    the GDB remote client happy.
+  * `watchpoint` - Can be used with of the `watch`/`rwatch`/`awatch` key value
+    pairs. Or can be used *instead* of those keys, with the specially formatted
+    `description` field.
+  * `exception` - an exception stop reason. Use with
+    the `description` key/value pair to describe the
+    exceptional event the user should see as the stop
+    reason.
+  * `description` -
+    An ASCII hex string that contains a more descriptive
+    reason that the thread stopped. This is only needed
+    if none of the key/value pairs are enough to
+    describe why something stopped.
+
+    For `reason:watchpoint`, `description` is an ascii-hex
+    encoded string with between one and three base 10 numbers,
+    space separated.  The three numbers are:
+      1. Watchpoint address. This address should always be within
+         a memory region lldb has a watchpoint on.
+         On architectures where the actual reported hit address may
+         be outside the watchpoint that was triggered, the remote
+         stub should determine which watchpoint was triggered and
+         report an address from within its range.
+      2. Wwatchpoint hardware register index number.
+      3. Actual watchpoint trap address, which may be outside
+         the range of any watched region of memory. On MIPS, an addr
+         outside a watched range means lldb should disable the wp,
+         step, re-enable the wp and continue silently.
+
+    On MIPS, the low 3 bits are masked so if a watchpoint is on
+    0x1004, a 2-byte write to 0x1000 will trigger the watchpoint
+    (a false positive hit), and lldb needs to disable the
+    watchpoint at 0x1004, inst-step, then re-enable the watchpoint
+    and not make this a user visible event. The description here
+    would be "0x1004 0 0x1000". lldb needs a known watchpoint address
+    in the first field, so it can disable it and step.
+
+    On AArch64 we have a related issue, where you watch 4 bytes at
+    0x1004, an instruction does an 8-byte write starting at
+    0x1000 (a true watchpoint hit) and the hardware may report the
+    trap address as 0x1000 - before the watched memory region -
+    with the write extending into the watched region.  This can
+    be reported as "0x1004 0 0x1000".  lldb will use 0x1004 to
+    identify which Watchpoint was triggered, and can report 0x1000
+    to the user.  The behavior of silently stepping over the
+    watchpoint, with an 3rd field addr outside the range, is
+    restricted to MIPS.
+
+    There may be false-positive watchpoint hits on AArch64 as well,
+    in the SVE Streaming Mode, but that is less common (see ESR
+    register flag "WPF", "Watchpoint might be False-Positive") and
+    not currently handled by lldb.
+* `threads` - `comma-sep-base16` -
+  A list of thread ids for all threads (including
+  the thread that we're reporting as stopped) that
+  are live in the process right now.  lldb may
+  request that this be included in the T packet via
+  the QListThreadsInStopReply packet earlier in
+  the debug session.
+
+  Example:
+  ```
+  threads:63387,633b2,63424,63462,63486;
+  ```
+* `thread-pcs` - `comma-sep-base16` -
+  A list of pc values for all threads that currently
+  exist in the process, including the thread that
+  this `T` packet is reporting as stopped.
+  This key-value pair will only be emitted when the
+  `threads` key is already included in the `T` packet.
+  The pc values correspond to the threads reported
+  in the `threads` list.  The number of pcs in the
+  `thread-pcs` list will be the same as the number of
+  threads in the `threads` list.
+  lldb may request that this be included in the `T`
+  packet via the `QListThreadsInStopReply` packet
+  earlier in the debug session.
+
+  Example:
+  ```
+  thread-pcs:dec14,2cf872b0,2cf8681c,2d02d68c,2cf716a8;
+  ```
+* `addressing_bits` - `unsigned` (optional) -
+  Specifies how many bits in addresses are significant for addressing, base
+  10.  If bits 38..0 in a 64-bit pointer are significant for addressing, then the
+  value is 39. This is needed on e.g. AArch64 v8.3 ABIs that use pointer
+  authentication in the high bits. This value is normally sent in the `qHostInfo`
+  packet response, and if the value cannot change during the process lifetime,
+  it does not need to be duplicated here in the stop packet. For a firmware
+  environment with early start code that may be changing the page table setup,
+  a dynamically set value may be needed.
+* `low_mem_addressing_bits` - `unsigned` (optional) -
+  Specifies how many bits in addresses in low memory are significant for
+  addressing, base 10.  AArch64 can have different page table setups for low
+  and high memory, and therefore a different number of bits used for addressing.
+* `high_mem_addressing_bits` - `unsigned` (optional) -
+  Specifies how many bits in addresses in high memory are significant for
+  addressing, base 10.  AArch64 can have different page table setups for low and
+  high memory, and therefore a different number of bits used for addressing.
+
+### Best Practices
+
+Since register values can be supplied with this packet, it is often useful
+to return the PC, SP, FP, LR (if any), and FLAGS registers so that separate
+packets don't need to be sent to read each of these registers from each
+thread.
+
+If a thread is stopped for no reason (like just because another thread
+stopped, or because when one core stops all cores should stop), use a
+`T` packet with `00` as the signal number and fill in as many key values
+and registers as possible.
+
+LLDB likes to know why a thread stopped since many thread control
+operations like stepping over a source line, actually are implemented
+by running the process multiple times. If a breakpoint is hit while
+trying to step over a source line and LLDB finds out that a breakpoint
+is hit in the "reason", we will know to stop trying to do the step
+over because something happened that should stop us from trying to
+do the step. If we are at a breakpoint and we disable the breakpoint
+at the current PC and do an instruction single step, knowing that
+we stopped due to a "trace" helps us know that we can continue
+running versus stopping due to a "breakpoint" (if we have two
+breakpoint instruction on consecutive instructions). So the more info
+we can get about the reason a thread stops, the better job LLDB can
+do when controlling your process. A typical GDB server behavior is
+to send a SIGTRAP for breakpoints _and_ also when instruction single
+stepping, in this case the debugger doesn't really know why we
+stopped and it can make it hard for the debugger to control your
+program correctly. What if a real SIGTRAP was delivered to a thread
+while we were trying to single step? We wouldn't know the difference
+with a standard GDB remote server and we could do the wrong thing.
+
+### Priority To Implement
+
+High. Having the extra information in your stop reply packets makes
+your debug session more reliable and informative.
+
+## qfProcessInfo / qsProcessInfo (Platform Extension)
+
+Get the first process info (`qfProcessInfo`) or subsequent process
+info (`qsProcessInfo`) for one or more processes on the remote
+platform. The first call gets the first match and subsequent calls
+to `qsProcessInfo` gets the subsequent matches. Return an error `EXX`,
+where `XX` are two hex digits, when no more matches are available.
+
+
+ The `qfProcessInfo` packet can be followed by a `:` and
+some key value pairs. The key value pairs in the command are:
+* `name` - `ascii-hex` -
+  An ASCII hex string that contains the name of the process that will be matched.
+* `name_match` - `enum` -
+  One of:
+    * `equals`
+    * `starts_with`
+    * `ends_with`
+    * `contains`
+    * `regex`
+* `pid` - `integer`- A string value containing the decimal process ID
+* `parent_pid` - `integer` - A string value containing the decimal parent process ID
+* `uid` - `integer` - A string value containing the decimal user ID
+* `gid` - `integer` - A string value containing the decimal group ID
+* `euid` - `integer` - A string value containing the decimal effective user ID
+* `egid` - `integer` - A string value containing the decimal effective group ID
+* `all_users` - `bool` -
+  A boolean value that specifies if processes should
+  be listed for all users, not just the user that the
+  platform is running as
+* `triple` - `string` -
+  An ASCII triple string (`x86_64`, `x86_64-apple-macosx`, `armv7-apple-ios`)
+* `args` - `string` -
+  A string value containing the process arguments separated by the character `-`,
+  where each argument is hex-encoded. It includes `argv[0]`.
+
+The response consists of key/value pairs where the key is separated from the
+values with colons and each pair is terminated with a semi colon. For a list
+of the key/value pairs in the response see the `qProcessInfoPID` packet
+documentation.
+
+Sample packet/response:
+```
+send packet: $qfProcessInfo#00
+read packet: $pid:60001;ppid:59948;uid:7746;gid:11;euid:7746;egid:11;name:6c6c6462;triple:x86_64-apple-macosx;#00
+send packet: $qsProcessInfo#00
+read packet: $pid:59992;ppid:192;uid:7746;gid:11;euid:7746;egid:11;name:6d64776f726b6572;triple:x86_64-apple-macosx;#00
+send packet: $qsProcessInfo#00
+read packet: $E04#00
+```
+
+### Priority To Implement
+
+Required.
+
+## qPathComplete (Platform Extension)
+
+Get a list of matched disk files/directories by passing a boolean flag
+and a partial path.
+
+### Example
+
+```
+receive: qPathComplete:0,6d61696e
+send:    M6d61696e2e637070
+receive: qPathComplete:1,746573
+send:    M746573742f,74657374732f
+```
+
+If the first argument is zero, the result should contain all
+files (including directories) starting with the given path. If the
+argument is one, the result should contain only directories.
+
+The result should be a comma-separated list of hex-encoded paths.
+Paths denoting a directory should end with a directory separator (`/` or `\`.
+
+## qKillSpawnedProcess (Platform Extension)
+
+Kill a process running on the target system.
+
+### Example
+
+```
+receive: qKillSpawnedProcess:1337
+send:    OK
+```
+The request packet has the process ID in base 10.
+
+## qLaunchGDBServer (Platform Extension)
+
+Have the remote platform launch a GDB server.
+
+The `qLaunchGDBServer` packet must be followed by a `:` and
+some key value pairs. The key value pairs in the command are:
+* `port` - `integer` -
+  A string value containing the decimal port ID or zero if the port should be
+  bound and returned
+* `host` - `integer` -
+  The host that connections should be limited to when the GDB server is connected to.
+
+Sample packet/response:
+```
+send packet: $qLaunchGDBServer:port:0;host:lldb.apple.com;#00
+read packet: $pid:60025;port:50776;#00
+```
+
+The `pid` key/value pair is only specified if the remote platform launched
+a separate process for the GDB remote server and can be omitted if no
+process was separately launched.
+
+The `port` key/value pair in the response lets clients know what port number
+to attach to in case zero was specified as the "port" in the sent command.
+
+### Priority To Implement
+
+Required.
+
+## qProcessInfoPID:PID (Platform Extension)
+
+Have the remote platform get detailed information on a process by
+ID. PID is specified as a decimal integer.
+
+The response consists of key/value pairs where the key is separated from the
+values with colons and each pair is terminated with a semi colon.
+
+The key value pairs in the response are:
+* `pid` - `integer` - Process ID as a decimal integer string
+* `ppid` - `integer` - Parent process ID as a decimal integer string
+* `uid` - `integer` - A string value containing the decimal user ID
+* `gid` - `integer` - A string value containing the decimal group ID
+* `euid` - `integer` - A string value containing the decimal effective user ID
+* `egid` - `integer` - A string value containing the decimal effective group ID
+* `name` - `ascii-hex` - An ASCII hex string that contains the name of the process
+* `triple` - `string` - A target triple (`x86_64-apple-macosx`, `armv7-apple-ios`)
+
+Sample packet/response:
+```
+send packet: $qProcessInfoPID:60050#00
+read packet: $pid:60050;ppid:59948;uid:7746;gid:11;euid:7746;egid:11;name:6c6c6462;triple:x86_64-apple-macosx;#00
+```
+
+### Priority To Implement
+
+Optional.
+
+## vAttachName
+
+Same as `vAttach`, except instead of a `pid` you send a process name.
+
+### Priority To Implement
+
+Low. Only needed for `process attach -n`.  If the packet isn't supported
+then `process attach -n` will fail gracefully.  So you need only to support
+it if attaching to a process by name makes sense for your environment.
+
+## vAttachWait
+
+Same as `vAttachName`, except that the stub should wait for the next instance
+of a process by that name to be launched and attach to that.
+
+### Priority To Implement
+
+Low. Only needed to support `process attach -w -n` which will fail
+gracefully if the packet is not supported.
+
+## qAttachOrWaitSupported
+
+This is a binary "is it supported" query. Return OK if you support
+`vAttachOrWait`.
+
+### Priority To Implement
+
+Low. This is required if you support `vAttachOrWait`, otherwise no support
+is needed since the standard "I don't recognize this packet" response
+will do the right thing.
+
+## vAttachOrWait
+
+Same as `vAttachWait`, except that the stub will attach to a process
+by name if it exists, and if it does not, it will wait for a process
+of that name to appear and attach to it.
+
+### Priority To Implement
+
+Low. Only needed to implement `process attach -w -i false -n`.  If
+you don't implement it but do implement `-n` AND lldb can somehow get
+a process list from your device, it will fall back on scanning the
+process list, and sending `vAttach` or `vAttachWait` depending on
+whether the requested process exists already.  This is racy,
+however, so if you want to support this behavior it is better to
+support this packet.
+
+## jThreadExtendedInfo
+
+This packet, which takes its arguments as JSON and sends its reply as
+JSON, allows the gdb remote stub to provide additional information
+about a given thread.
+
+This packet takes its arguments in [JSON](http://www.json.org).
+At a minimum, a thread must be specified, for example:
+```
+jThreadExtendedInfo:{"thread":612910}
+```
+
+Because this is a JSON string, the thread number is provided in base 10.
+Additional key-value pairs may be provided by lldb to the gdb remote
+stub.  For instance, on some versions of macOS, lldb can read offset
+information out of the system libraries.  Using those offsets, debugserver
+is able to find the Thread Specific Address (TSD) for a thread and include
+that in the return information.  So lldb will send these additional fields
+like so:
+```
+jThreadExtendedInfo:{"plo_pthread_tsd_base_address_offset":0,"plo_pthread_tsd_base_offset":224,"plo_pthread_tsd_entry_size":8,"thread":612910}
+```
+
+There are no requirements for what is included in the response.  A simple
+reply on a OS X Yosemite / iOS 8 may include the pthread_t value, the
+Thread Specific Data (TSD) address, the dispatch_queue_t value if the thread
+is associated with a GCD queue, and the requested Quality of Service (QoS)
+information about that thread.  For instance, a reply may look like:
+```
+{"tsd_address":4371349728,"requested_qos":{"enum_value":33,"constant_name":"QOS_CLASS_USER_INTERACTIVE","printable_name":"User Interactive"},"pthread_t":4371349504,"dispatch_queue_t":140735087127872}
+```
+
+`tsd_address`, `pthread_t`, and `dispatch_queue_t` are all simple key-value pairs.
+The JSON standard requires that numbers be expressed in base 10 - so all of
+these are. `requested_qos` is a dictionary with three key-value pairs in it -
+so the UI layer may choose the form most appropriate for displaying to the user.
+
+Sending JSON over gdb-remote protocol introduces some problems.  We may be
+sending strings with arbitrary contents in them, including the `#`, `$`, and `*`
+characters that have special meaning in gdb-remote protocol and cannot occur
+in the middle of the string. The standard solution for this would be to require
+ascii-hex encoding of all strings, or ascii-hex encode the entire JSON payload.
+
+Instead, the binary escaping convention is used for JSON data.  This convention
+(e.g. used for the `X` packet) says that if `#`, `$`, `*`, or `}` are to occur in
+the payload, the character `}` (`0x7d`) is emitted, then the metacharacter is emitted
+xor'ed by `0x20`. The `}` character occurs in every JSON payload at least once, and
+`} ^ 0x20` happens to be `]` so the raw packet characters for a request will look
+like:
+```
+jThreadExtendedInfo:{"thread":612910}]
+```
+
+### Priority To Implement
+
+Low.  This packet is only needed if the gdb remote stub wants to
+provide interesting additional information about a thread for the
+user.
+
+## QEnableCompression
+
+This packet enables compression of the packets that the debug stub sends to lldb.
+If the debug stub can support compression, it indictes this in the reply of the
+"qSupported" packet. For example:
+```
+LLDB SENDS:    qSupported:xmlRegisters=i386,arm,mips
+STUB REPLIES:  qXfer:features:read+;SupportedCompressions=lzfse,zlib-deflate,lz4,lzma;
+```
+
+If lldb knows how to use any of these compression algorithms, it can ask that this
+compression mode be enabled.
+```
+QEnableCompression:type:zlib-deflate;
+```
+
+The debug stub should reply with an uncompressed `OK` packet to indicate that the
+request was accepted.  All further packets the stub sends will use this compression.
+
+Packets are compressed as the last step before they are sent from the stub, and
+decompressed as the first step after they are received.  The packet format in compressed
+mode becomes one of two:
+```
+$N<uncompressed payload>#00
+
+$C<size of uncompressed payload in base 10>:<compressed payload>#00
+```
+
+Where `#00` is the actual checksum value if noack mode is not enabled. The checksum
+value is for the `N<uncompressed payload>` or
+`C<size of uncompressed payload in base 10>:<compressed payload>` bytes in the packet.
+
+The size of the uncompressed payload in base 10 is provided because it will simplify
+decompression if the final buffer size needed is known ahead of time.
+
+Compression on low-latency connections is unlikely to be an improvement. Particularly
+when the debug stub and lldb are running on the same host. It should only be used
+for slow connections, and likely only for larger packets.
+
+Example compression algorithms that may be used include:
+* `zlib-deflate` -
+  The raw DEFLATE format as described in IETF RFC 1951.  With the ZLIB library, you
+  can compress to this format with an initialization like
+      deflateInit2 (&stream, 5, Z_DEFLATED, -15, 8, Z_DEFAULT_STRATEGY)
+  and you can decompress with an initialization like
+      inflateInit2 (&stream, -15).
+* `lz4` -
+  https://en.wikipedia.org/wiki/LZ4_(compression_algorithm)
+  https://github.com/Cyan4973/lz4
+  The libcompression APIs on darwin systems call this `COMPRESSION_LZ4_RAW`.
+* `lzfse` -
+  Compression algorithm added in macOS 10.11, with open source C reference
+  implementation on github.
+  https://en.wikipedia.org/wiki/LZFSE
+  https://github.com/lzfse/lzfse
+* `lzma` -
+  libcompression implements "LZMA level 6", the default compression for the
+  open source LZMA implementation.
+
+## jGetLoadedDynamicLibrariesInfos
+
+This packet asks the remote debug stub to send the details about libraries
+being added/removed from the process as a performance optimization.
+
+There are two ways this packet can be used.  Both return a dictionary of
+binary images formatted the same way.
+
+One requests information on all shared libraries:
+```
+jGetLoadedDynamicLibrariesInfos:{"fetch_all_solibs":true}
+```
+with an optional `"report_load_commands":false` which can be added, asking
+that only the dyld SPI information (load addresses, filenames) be returned.
+The default behavior is that debugserver scans the mach-o header and load
+commands of each binary, and returns it in the JSON reply.
+
+And the second requests information about a list of shared libraries, given their load addresses:
+```
+jGetLoadedDynamicLibrariesInfos:{"solib_addresses":[8382824135,3258302053,830202858503]}
+```
+
+The second call is both a performance optimization (instead of having lldb read the mach-o header/load commands
+out of memory with generic read packets) but also adds additional information in the form of the
+filename of the shared libraries (which is not available in the mach-o header/load commands.)
+
+An example using the OS X 10.11 style call:
+```
+LLDB SENDS: jGetLoadedDynamicLibrariesInfos:{"image_count":1,"image_list_address":140734800075128}
+STUB REPLIES: ${"images":[{"load_address":4294967296,"mod_date":0,"pathname":"/tmp/a.out","uuid":"02CF262C-ED6F-3965-9E14-63538B465CFF","mach_header":{"magic":4277009103,"cputype":16777223,"cpusubtype":18446744071562067971,"filetype":2},"segments":{"name":"__PAGEZERO","vmaddr":0,"vmsize":4294967296,"fileoff":0,"filesize":0,"maxprot":0},{"name":"__TEXT","vmaddr":4294967296,"vmsize":4096,"fileoff":0,"filesize":4096,"maxprot":7},{"name":"__LINKEDIT","vmaddr":4294971392,"vmsize":4096,"fileoff":4096,"filesize":152,"maxprot":7}}]}#00
+```
+
+Or pretty-printed:
+```
+STUB REPLIES: ${"images":
+                [
+                    {"load_address":4294967296,
+                     "mod_date":0,
+                     "pathname":"/tmp/a.out",
+                     "uuid":"02CF262C-ED6F-3965-9E14-63538B465CFF",
+                     "mach_header":
+                        {"magic":4277009103,
+                         "cputype":16777223,
+                         "cpusubtype":18446744071562067971,
+                         "filetype":2
+                         },
+                     "segments":
+                      [
+                        {"name":"__PAGEZERO",
+                         "vmaddr":0,
+                         "vmsize":4294967296,
+                         "fileoff":0,
+                         "filesize":0,
+                         "maxprot":0
+                        },
+                        {"name":"__TEXT",
+                         "vmaddr":4294967296,
+                         "vmsize":4096,
+                         "fileoff":0,
+                         "filesize":4096,
+                         "maxprot":7
+                        },
+                        {"name":"__LINKEDIT",
+                         "vmaddr":4294971392,
+                         "vmsize":4096,
+                         "fileoff":4096,
+                         "filesize":152,
+                         "maxprot":7
+                        }
+                      ]
+                    }
+                ]
+            }
+```
+
+This is similar to the `qXfer:libraries:read` packet, and it could
+be argued that it should be merged into that packet.  A separate
+packet was created primarily because lldb needs to specify the
+number of images to be read and the address from which the initial
+information is read.  Also the XML DTD would need to be extended
+quite a bit to provide all the information that the `DynamicLoaderMacOSX`
+would need to work correctly on this platform.
+
+### Priority To Implement
+
+On OS X 10.11, iOS 9, tvOS 9, watchOS 2 and older: Low.  If this packet is absent,
+lldb will read the Mach-O headers/load commands out of memory.
+On macOS 10.12, iOS 10, tvOS 10, watchOS 3 and newer: High.  If this packet is absent,
+lldb will not know anything about shared libraries in the inferior, or where the main
+executable loaded.
+
+## jThreadsInfo
+
+Ask for the server for thread stop information of all threads.
+
+The data in this packet is very similar to the stop reply packets, but is packaged in
+JSON and uses JSON arrays where applicable. The JSON output looks like:
+```
+    [
+      { "tid":1580681,
+        "metype":6,
+        "medata":[2,0],
+        "reason":"exception",
+        "qaddr":140735118423168,
+        "registers": {
+          "0":"8000000000000000",
+          "1":"0000000000000000",
+          "2":"20fabf5fff7f0000",
+          "3":"e8f8bf5fff7f0000",
+          "4":"0100000000000000",
+          "5":"d8f8bf5fff7f0000",
+          "6":"b0f8bf5fff7f0000",
+          "7":"20f4bf5fff7f0000",
+          "8":"8000000000000000",
+          "9":"61a8db78a61500db",
+          "10":"3200000000000000",
+          "11":"4602000000000000",
+          "12":"0000000000000000",
+          "13":"0000000000000000",
+          "14":"0000000000000000",
+          "15":"0000000000000000",
+          "16":"960b000001000000",
+          "17":"0202000000000000",
+          "18":"2b00000000000000",
+          "19":"0000000000000000",
+          "20":"0000000000000000"
+        },
+        "memory":[
+          {"address":140734799804592,"bytes":"c8f8bf5fff7f0000c9a59e8cff7f0000"},
+          {"address":140734799804616,"bytes":"00000000000000000100000000000000"}
+        ]
+      }
+    ]
+```
+
+It contains an array of dictionaries with all of the key value pairs that are
+normally in the stop reply packet, including the expedited registers. The registers are
+passed as hex-encoded JSON string in debuggee-endian byte order. Note that the register
+numbers are decimal numbers, unlike the stop-reply packet, where they are written in
+hex. The packet also contains expedited memory in the `memory` key.  This allows the
+server to expedite memory that the client is likely to use (e.g., areas around the
+stack pointer, which are needed for computing backtraces) and it reduces the packet
+count.
+
+On macOS with debugserver, we expedite the frame pointer backchain for a thread
+(up to 256 entries) by reading 2 pointers worth of bytes at the frame pointer (for
+the previous FP and PC), and follow the backchain. Most backtraces on macOS and
+iOS now don't require us to read any memory!
+
+### Priority To Implement
+
+Low. This is a performance optimization, which speeds up debugging by avoiding
+multiple round-trips for retrieving thread information. The information from this
+packet can be retrieved using a combination of `qThreadStopInfo` and `m` packets.
+
+## jGetSharedCacheInfo
+
+This packet asks the remote debug stub to send the details about the inferior's
+shared cache. The shared cache is a collection of common libraries/frameworks that
+are mapped into every process at the same address on Darwin systems, and can be
+identified by a load address and UUID.
+
+```
+LLDB SENDS: jGetSharedCacheInfo:{}
+STUB REPLIES: ${"shared_cache_base_address":140735683125248,"shared_cache_uuid":"DDB8D70C-C9A2-3561-B2C8-BE48A4F33F96","no_shared_cache":false,"shared_cache_private_cache":false]}#00
+```
+
+### Priority To Implement
+
+Low.  When both lldb and the inferior process are running on the same computer, and lldb
+and the inferior process have the same shared cache, lldb may (as an optimization) read
+the shared cache out of its own memory instead of using gdb-remote read packets to read
+them from the inferior process.
+
+## qQueryGDBServer
+
+Ask the platform for the list of gdbservers we have to connect
+
+If the remote platform automatically started one or more gdbserver instance (without
+lldb asking it) then it have to return the list of port number or socket name for
+each of them what can be used by lldb to connect to those instances.
+
+The data in this packet is a JSON array of JSON objects with the following keys:
+* `port`: `<the port number to connect>` (optional)
+* `socket_name`: `<the name of the socket to connect>` (optional)
+
+Example packet:
+```
+[
+    { "port": 1234 },
+    { "port": 5432 },
+    { "socket_name": "foo" }
+]
+```
+
+### Priority To Implement
+
+Low. The packet is required to support connecting to gdbserver started
+by the platform instance automatically.
+
+## QSetDetachOnError
+
+Sets what the server should do when the communication channel with LLDB
+goes down. Either kill the inferior process (`0`) or remove breakpoints and
+detach (`1`).
+
+The data in this packet is a single a character, which should be `0` if the
+inferior process should be killed, or `1` if the server should remove all
+breakpoints and detach from the inferior.
+
+### Priority To Implement
+
+Low. Only required if the target wants to keep the inferior process alive
+when the communication channel goes down.
+
+## jGetDyldProcessState
+
+This packet fetches the process launch state, as reported by libdyld on
+Darwin systems, most importantly to indicate when the system libraries
+have initialized sufficiently to safely call utility functions.
+
+```
+LLDB SENDS: jGetDyldProcessState
+STUB REPLIES: {"process_state_value":48,"process_state string":"dyld_process_state_libSystem_initialized"}
+```
+
+### Priority To Implement
+
+Low. This packet is needed to prevent lldb's utility functions for
+scanning the Objective-C class list from running very early in
+process startup.
+
+## vFile Packets
+
+Though some of these may match the ones described in GDB's protocol
+documentation, we include our own expectations here in case of
+mismatches or extensions.
+
+### vFile:size
+
+Get the size of a file on the target system, filename in ASCII hex.
+
+#### Example
+
+```
+receive: vFile:size:2f746d702f61
+send:    Fc008
+```
+
+response is `F` followed by the file size in base 16.
+`F-1,errno` with the errno if an error occurs, base 16.
+
+### vFile:mode
+
+Get the mode bits of a file on the target system, filename in ASCII hex.
+
+#### Example
+
+```
+receive: vFile:mode:2f746d702f61
+send:    F1ed
+```
+
+response is `F` followed by the mode bits in base 16, this `0x1ed` would
+correspond to `0755` in octal.
+`F-1,errno` with the errno if an error occurs, base 16.
+
+### vFile:unlink
+
+Remove a file on the target system.
+
+#### Example
+
+```
+receive: vFile:unlink:2f746d702f61
+send:    F0
+```
+
+Argument is a file path in ascii-hex encoding.
+Response is `F` plus the return value of `unlink()`, base 16 encoding.
+Return value may optionally be followed by a comma and the base16
+value of errno if unlink failed.
+
+### vFile:symlink
+
+Create a symbolic link (symlink, soft-link) on the target system.
+
+#### Example
+
+```
+receive: vFile:symlink:<SRC-FILE>,<DST-NAME>
+send:    F0,0
+```
+
+Argument file paths are in ascii-hex encoding.
+Response is `F` plus the return value of `symlink()`, base 16 encoding,
+optionally followed by the value of errno if it failed, also base 16.
+
+### vFile:open
+
+Open a file on the remote system and return the file descriptor of it.
+
+#### Example
+
+```
+receive: vFile:open:2f746d702f61,00000001,00000180
+send:    F8
+```
+
+request packet has the fields:
+   1. ASCII hex encoded filename
+   2. Flags passed to the open call, base 16.
+      Note that these are not the `oflags` that `open(2)` takes, but
+      are the constant values in `enum OpenOptions` from LLDB's
+      [`File.h`](https://github.com/llvm/llvm-project/blob/main/lldb/include/lldb/Host/File.h).
+   3. Mode bits, base 16
+
+response is `F` followed by the opened file descriptor in base 16.
+`F-1,errno` with the errno if an error occurs, base 16.
+
+### vFile:close
+
+Close a previously opened file descriptor.
+
+#### Example
+
+```
+receive: vFile:close:7
+send:    F0
+```
+
+File descriptor is in base 16. `F-1,errno` with the errno if an error occurs,
+errno is base 16.
+
+### vFile:pread
+
+Read data from an opened file descriptor.
+
+#### Example
+
+```
+receive: vFile:pread:7,1024,0
+send:    F4;a'b\00
+```
+
+Request packet has the fields:
+   1. File descriptor, base 16
+   2. Number of bytes to be read, base 16
+   3. Offset into file to start from, base 16
+
+Response is `F`, followed by the number of bytes read (base 16), a
+semicolon, followed by the data in the binary-escaped-data encoding.
+
+### vFile:pwrite
+
+Write data to a previously opened file descriptor.
+
+#### Example
+
+```
+receive: vFile:pwrite:8,0,\cf\fa\ed\fe\0c\00\00
+send:    F1024
+```
+
+Request packet has the fields:
+   1. File descriptor, base 16
+   2. Offset into file to start from, base 16
+   3. binary-escaped-data to be written
+
+Response is `F`, followed by the number of bytes written (base 16).
+
+### vFile:MD5
+
+Generate an MD5 hash of the file at the given path.
+
+#### Example
+
+```
+receive: vFile:MD5:2f746d702f61
+send (success): F,00000000000000001111111111111111
+send (failure): F,x
+```
+
+Request packet contains the ASCII hex encoded filename.
+
+If the hash succeeded, the response is `F,` followed by the low 64
+bits of the result, and finally the high 64 bits of the result. Both are in
+hex format without a prefix.
+
+The response is `F,`, followed by `x` if the file did not exist
+or failed to hash.
+
+### vFile:exists
+
+Check whether the file at the given path exists.
+
+#### Example
+
+```
+receive: vFile:exists:2f746d702f61
+send         (exists): F,1
+send (does not exist): F,0
+```
+
+Request packet contains the ASCII hex encoded filename.
+
+The response is a return code where 1 means the file exists
+and 0 means it does not.
\ No newline at end of file
diff --git a/lldb/docs/resources/lldbplatformpackets.md b/lldb/docs/resources/lldbplatformpackets.md
new file mode 100644
index 000000000000..f7262e193d94
--- /dev/null
+++ b/lldb/docs/resources/lldbplatformpackets.md
@@ -0,0 +1,49 @@
+# LLDB Platform Packets
+
+This is a list of the packets that an lldb platform server
+needs to implement for the lldb testsuite to be run on a remote
+target device/system.
+
+These are almost all lldb extensions to the gdb-remote serial
+protocol. Many of the `vFile:` packets are also described in the "Host
+I/O Packets" detailed in the gdb-remote protocol documentation,
+although the lldb platform extensions include packets that are not
+defined there (`vFile:size:`, `vFile:mode:`, `vFile:symlink`, `vFile:chmod:`).
+
+Most importantly, the flags that LLDB passes to `vFile:open:` are
+incompatible with the flags that GDB specifies.
+
+* [QStartNoAckMode](./lldbgdbremote.md#qstartnoackmode)
+* [qHostInfo](./lldbgdbremote.md#qhostinfo)
+* [qModuleInfo](./lldbgdbremote.md#qmoduleinfo-module-path-arch-triple)
+* [qGetWorkingDir](./lldbgdbremote.md#qgetworkingdir)
+* [QSetWorkingDir](./lldbgdbremote.md#qsetworkingdir-ascii-hex-path)
+* [qPlatform_mkdir](./lldbgdbremote.md#qplatform-mkdir)
+* [qPlatform_shell](./lldbgdbremote.md#qplatform-shell)
+* [qLaunchGDBServer](./lldbgdbremote.md#qlaunchgdbserver-platform-extension)
+* [qKillSpawnedProcess](./lldbgdbremote.md#qkillspawnedprocess-platform-extension)
+* [qProcessInfoPID](./lldbgdbremote.md#qprocessinfopid-pid-platform-extension)
+  * It is likely that you only need to support the `pid` and `name` fields.
+* [qProcessInfo](./lldbgdbremote.md#qprocessinfo)
+  * The lldb test suite currently only uses `name_match:equals` and the no-criteria mode to list every process.
+* [qPathComplete](./lldbgdbremote.md#qpathcomplete-platform-extension)
+* [vFile:chmod](./lldbgdbremote.md#vfile-chmod-qplatform-chmod)
+* [vFile:size](./lldbgdbremote.md#vfile-size)
+* [vFile:mode](./lldbgdbremote.md#vfile-mode)
+* [vFile:unlink](./lldbgdbremote.md#vfile-unlink)
+* [vFile:symlink](./lldbgdbremote.md#vfile-symlink)
+* [vFile:open](./lldbgdbremote.md#vfile-open)
+* [vFile:close](./lldbgdbremote.md#vfile-close)
+* [vFile:pread](./lldbgdbremote.md#vfile-pread)
+* [vFile:pwrite](./lldbgdbremote.md#vfile-pwrite)
+
+The remote platform must be able to launch processes so that debugserver
+can attach to them. This requires the following packets in addition to the
+previous list:
+* [QSetDisableASLR](./lldbgdbremote.md#qsetdisableaslr-bool)
+* [QSetDetatchOnError](./lldbgdbremote.md#qsetdetachonerror)
+* [QSetSTDIN / QSetSTDOUT / QSetSTDERR](./lldbgdbremote.md#qsetstdin-ascii-hex-path-qsetstdout-ascii-hex-path-qsetstderr-ascii-hex-path) (all 3)
+* [QEnvironment](./lldbgdbremote.md#qenvironment-name-value)
+* [QEnvironmentHexEncoded](./lldbgdbremote.md#qenvironmenthexencoded-hex-encoding-name-value)
+* [A](./lldbgdbremote.md#a-launch-args-packet)
+* [qLaunchSuccess](./lldbgdbremote.md#qlaunchsuccess)
diff --git a/lldb/docs/resources/test.rst b/lldb/docs/resources/test.rst
index 094fde8b1b5a..382e42bf22b1 100644
--- a/lldb/docs/resources/test.rst
+++ b/lldb/docs/resources/test.rst
@@ -618,9 +618,9 @@ On non-Windows platforms, you can use the ``-d`` option to ``dotest.py`` which
 will cause the script to print out the pid of the test and wait for a while
 until a debugger is attached. Then run ``lldb -p <pid>`` to attach.
 
-To instead debug a test's python source, edit the test and insert
-``import pdb; pdb.set_trace()`` at the point you want to start debugging. In
-addition to pdb's debugging facilities, lldb commands can be executed with the
+To instead debug a test's python source, edit the test and insert ``import pdb; pdb.set_trace()`` or ``breakpoint()`` (Python 3 only) at the point you want to start debugging. The ``breakpoint()`` command can be used for any LLDB Python script, not just for API tests.
+
+In addition to pdb's debugging facilities, lldb commands can be executed with the
 help of a pdb alias. For example ``lldb bt`` and ``lldb v some_var``. Add this
 line to your ``~/.pdbrc``:
 
diff --git a/lldb/include/lldb/API/SBDebugger.h b/lldb/include/lldb/API/SBDebugger.h
index cf5409a12a05..7333cd57ad31 100644
--- a/lldb/include/lldb/API/SBDebugger.h
+++ b/lldb/include/lldb/API/SBDebugger.h
@@ -42,6 +42,13 @@ public:
 
 class LLDB_API SBDebugger {
 public:
+  FLAGS_ANONYMOUS_ENUM(){
+      eBroadcastBitProgress = lldb::DebuggerBroadcastBit::eBroadcastBitProgress,
+      eBroadcastBitWarning = lldb::DebuggerBroadcastBit::eBroadcastBitWarning,
+      eBroadcastBitError = lldb::DebuggerBroadcastBit::eBroadcastBitError,
+      eBroadcastBitProgressCategory =
+          lldb::DebuggerBroadcastBit::eBroadcastBitProgressCategory,
+  };
   SBDebugger();
 
   SBDebugger(const lldb::SBDebugger &rhs);
diff --git a/lldb/include/lldb/API/SBLineEntry.h b/lldb/include/lldb/API/SBLineEntry.h
index 7c2431ba3c8a..d70c4fac6ec7 100644
--- a/lldb/include/lldb/API/SBLineEntry.h
+++ b/lldb/include/lldb/API/SBLineEntry.h
@@ -29,6 +29,9 @@ public:
 
   lldb::SBAddress GetEndAddress() const;
 
+  lldb::SBAddress
+  GetSameLineContiguousAddressRangeEnd(bool include_inlined_functions) const;
+
   explicit operator bool() const;
 
   bool IsValid() const;
diff --git a/lldb/include/lldb/API/SBProcess.h b/lldb/include/lldb/API/SBProcess.h
index 7da3335a7234..f1b5d1fb92ce 100644
--- a/lldb/include/lldb/API/SBProcess.h
+++ b/lldb/include/lldb/API/SBProcess.h
@@ -562,6 +562,8 @@ public:
 
   lldb::SBScriptObject GetScriptedImplementation();
 
+  void GetStatus(SBStream &status);
+
 protected:
   friend class SBAddress;
   friend class SBBreakpoint;
diff --git a/lldb/include/lldb/API/SBSymbolContextList.h b/lldb/include/lldb/API/SBSymbolContextList.h
index 4026afc21357..95100d219df2 100644
--- a/lldb/include/lldb/API/SBSymbolContextList.h
+++ b/lldb/include/lldb/API/SBSymbolContextList.h
@@ -44,6 +44,7 @@ public:
 protected:
   friend class SBModule;
   friend class SBTarget;
+  friend class SBCompileUnit;
 
   lldb_private::SymbolContextList *operator->() const;
 
diff --git a/lldb/include/lldb/API/SBTarget.h b/lldb/include/lldb/API/SBTarget.h
index 3644ac056da3..feeaa1cb7113 100644
--- a/lldb/include/lldb/API/SBTarget.h
+++ b/lldb/include/lldb/API/SBTarget.h
@@ -879,6 +879,10 @@ public:
                                            uint32_t count,
                                            const char *flavor_string);
 
+  lldb::SBInstructionList ReadInstructions(lldb::SBAddress start_addr,
+                                           lldb::SBAddress end_addr,
+                                           const char *flavor_string);
+
   lldb::SBInstructionList GetInstructions(lldb::SBAddress base_addr,
                                           const void *buf, size_t size);
 
@@ -954,6 +958,7 @@ protected:
   friend class SBSection;
   friend class SBSourceManager;
   friend class SBSymbol;
+  friend class SBTypeStaticField;
   friend class SBValue;
   friend class SBVariablesOptions;
 
diff --git a/lldb/include/lldb/API/SBType.h b/lldb/include/lldb/API/SBType.h
index 9980fe121830..5b9ff2170b2b 100644
--- a/lldb/include/lldb/API/SBType.h
+++ b/lldb/include/lldb/API/SBType.h
@@ -107,6 +107,35 @@ protected:
   lldb::TypeMemberFunctionImplSP m_opaque_sp;
 };
 
+class LLDB_API SBTypeStaticField {
+public:
+  SBTypeStaticField();
+
+  SBTypeStaticField(const lldb::SBTypeStaticField &rhs);
+  lldb::SBTypeStaticField &operator=(const lldb::SBTypeStaticField &rhs);
+
+  ~SBTypeStaticField();
+
+  explicit operator bool() const;
+
+  bool IsValid() const;
+
+  const char *GetName();
+
+  const char *GetMangledName();
+
+  lldb::SBType GetType();
+
+  lldb::SBValue GetConstantValue(lldb::SBTarget target);
+
+protected:
+  friend class SBType;
+
+  explicit SBTypeStaticField(lldb_private::CompilerDecl decl);
+
+  std::unique_ptr<lldb_private::CompilerDecl> m_opaque_up;
+};
+
 class SBType {
 public:
   SBType();
@@ -182,6 +211,8 @@ public:
 
   lldb::SBTypeMember GetVirtualBaseClassAtIndex(uint32_t idx);
 
+  lldb::SBTypeStaticField GetStaticFieldWithName(const char *name);
+
   lldb::SBTypeEnumMemberList GetEnumMembers();
 
   uint32_t GetNumberOfTemplateArguments();
@@ -242,6 +273,7 @@ protected:
   friend class SBTypeNameSpecifier;
   friend class SBTypeMember;
   friend class SBTypeMemberFunction;
+  friend class SBTypeStaticField;
   friend class SBTypeList;
   friend class SBValue;
   friend class SBWatchpoint;
diff --git a/lldb/include/lldb/API/SBValue.h b/lldb/include/lldb/API/SBValue.h
index bbcccaab51aa..8f4c4fd56dfb 100644
--- a/lldb/include/lldb/API/SBValue.h
+++ b/lldb/include/lldb/API/SBValue.h
@@ -68,6 +68,8 @@ public:
 
   uint64_t GetValueAsUnsigned(uint64_t fail_value = 0);
 
+  lldb::addr_t GetValueAsAddress();
+
   ValueType GetValueType();
 
   // If you call this on a newly created ValueObject, it will always return
@@ -426,6 +428,7 @@ protected:
   friend class SBModule;
   friend class SBTarget;
   friend class SBThread;
+  friend class SBTypeStaticField;
   friend class SBTypeSummary;
   friend class SBValueList;
 
diff --git a/lldb/include/lldb/Core/Debugger.h b/lldb/include/lldb/Core/Debugger.h
index 418c2403d020..49ff0737acef 100644
--- a/lldb/include/lldb/Core/Debugger.h
+++ b/lldb/include/lldb/Core/Debugger.h
@@ -89,7 +89,7 @@ public:
 
   using DebuggerList = std::vector<lldb::DebuggerSP>;
 
-  static ConstString GetStaticBroadcasterClass();
+  static llvm::StringRef GetStaticBroadcasterClass();
 
   /// Get the public broadcaster for this debugger.
   Broadcaster &GetBroadcaster() { return m_broadcaster; }
diff --git a/lldb/include/lldb/Core/ThreadedCommunication.h b/lldb/include/lldb/Core/ThreadedCommunication.h
index 7ebb77beb77f..24412b202793 100644
--- a/lldb/include/lldb/Core/ThreadedCommunication.h
+++ b/lldb/include/lldb/Core/ThreadedCommunication.h
@@ -216,9 +216,9 @@ public:
   ///
   void SynchronizeWithReadThread();
 
-  static ConstString &GetStaticBroadcasterClass();
+  static llvm::StringRef GetStaticBroadcasterClass();
 
-  ConstString &GetBroadcasterClass() const override {
+  llvm::StringRef GetBroadcasterClass() const override {
     return GetStaticBroadcasterClass();
   }
 
diff --git a/lldb/include/lldb/Interpreter/CommandInterpreter.h b/lldb/include/lldb/Interpreter/CommandInterpreter.h
index d190bcdcab44..70a55a77465b 100644
--- a/lldb/include/lldb/Interpreter/CommandInterpreter.h
+++ b/lldb/include/lldb/Interpreter/CommandInterpreter.h
@@ -255,9 +255,9 @@ public:
 
   // These two functions fill out the Broadcaster interface:
 
-  static ConstString &GetStaticBroadcasterClass();
+  static llvm::StringRef GetStaticBroadcasterClass();
 
-  ConstString &GetBroadcasterClass() const override {
+  llvm::StringRef GetBroadcasterClass() const override {
     return GetStaticBroadcasterClass();
   }
 
diff --git a/lldb/include/lldb/Symbol/CompilerDecl.h b/lldb/include/lldb/Symbol/CompilerDecl.h
index 825a4f15836f..5c99cae3781c 100644
--- a/lldb/include/lldb/Symbol/CompilerDecl.h
+++ b/lldb/include/lldb/Symbol/CompilerDecl.h
@@ -73,6 +73,9 @@ public:
 
   CompilerDeclContext GetDeclContext() const;
 
+  // If this decl has a type, return it.
+  CompilerType GetType() const;
+
   // If this decl represents a function, return the return type
   CompilerType GetFunctionReturnType() const;
 
@@ -91,6 +94,10 @@ public:
   /// the subsequent entry, so the topmost entry is the global namespace.
   std::vector<lldb_private::CompilerContext> GetCompilerContext() const;
 
+  // If decl represents a constant value, return it. Otherwise, return an
+  // invalid/empty Scalar.
+  Scalar GetConstantValue() const;
+
 private:
   TypeSystem *m_type_system = nullptr;
   void *m_opaque_decl = nullptr;
diff --git a/lldb/include/lldb/Symbol/CompilerType.h b/lldb/include/lldb/Symbol/CompilerType.h
index b71c531f2163..9e889a53086b 100644
--- a/lldb/include/lldb/Symbol/CompilerType.h
+++ b/lldb/include/lldb/Symbol/CompilerType.h
@@ -416,6 +416,8 @@ public:
   CompilerType GetVirtualBaseClassAtIndex(size_t idx,
                                           uint32_t *bit_offset_ptr) const;
 
+  CompilerDecl GetStaticFieldWithName(llvm::StringRef name) const;
+
   uint32_t GetIndexOfFieldWithName(const char *name,
                                    CompilerType *field_compiler_type = nullptr,
                                    uint64_t *bit_offset_ptr = nullptr,
diff --git a/lldb/include/lldb/Symbol/TypeSystem.h b/lldb/include/lldb/Symbol/TypeSystem.h
index 3a927d313b82..ad3b853227a9 100644
--- a/lldb/include/lldb/Symbol/TypeSystem.h
+++ b/lldb/include/lldb/Symbol/TypeSystem.h
@@ -27,6 +27,8 @@
 #include "lldb/Symbol/CompilerDecl.h"
 #include "lldb/Symbol/CompilerDeclContext.h"
 #include "lldb/Symbol/Type.h"
+#include "lldb/Utility/Scalar.h"
+#include "lldb/lldb-forward.h"
 #include "lldb/lldb-private.h"
 #include "lldb/lldb-types.h"
 
@@ -110,6 +112,8 @@ public:
   virtual std::vector<lldb_private::CompilerContext>
   DeclGetCompilerContext(void *opaque_decl);
 
+  virtual Scalar DeclGetConstantValue(void *opaque_decl) { return Scalar(); }
+
   virtual CompilerType GetTypeForDecl(void *opaque_decl) = 0;
 
   // CompilerDeclContext functions
@@ -339,6 +343,11 @@ public:
   GetVirtualBaseClassAtIndex(lldb::opaque_compiler_type_t type, size_t idx,
                              uint32_t *bit_offset_ptr) = 0;
 
+  virtual CompilerDecl GetStaticFieldWithName(lldb::opaque_compiler_type_t type,
+                                              llvm::StringRef name) {
+    return CompilerDecl();
+  }
+
   virtual CompilerType GetChildCompilerTypeAtIndex(
       lldb::opaque_compiler_type_t type, ExecutionContext *exe_ctx, size_t idx,
       bool transparent_pointers, bool omit_empty_base_classes,
diff --git a/lldb/include/lldb/Target/Process.h b/lldb/include/lldb/Target/Process.h
index 2f3a3c22422e..aac0cf51680a 100644
--- a/lldb/include/lldb/Target/Process.h
+++ b/lldb/include/lldb/Target/Process.h
@@ -381,7 +381,7 @@ public:
 
   // These two functions fill out the Broadcaster interface:
 
-  static ConstString &GetStaticBroadcasterClass();
+  static llvm::StringRef GetStaticBroadcasterClass();
 
   static constexpr llvm::StringRef AttachSynchronousHijackListenerName =
       "lldb.internal.Process.AttachSynchronous.hijack";
@@ -390,7 +390,7 @@ public:
   static constexpr llvm::StringRef ResumeSynchronousHijackListenerName =
       "lldb.internal.Process.ResumeSynchronous.hijack";
 
-  ConstString &GetBroadcasterClass() const override {
+  llvm::StringRef GetBroadcasterClass() const override {
     return GetStaticBroadcasterClass();
   }
 
diff --git a/lldb/include/lldb/Target/Target.h b/lldb/include/lldb/Target/Target.h
index 2c2e6b2831cc..cade60f3cc8c 100644
--- a/lldb/include/lldb/Target/Target.h
+++ b/lldb/include/lldb/Target/Target.h
@@ -499,9 +499,9 @@ public:
 
   // These two functions fill out the Broadcaster interface:
 
-  static ConstString &GetStaticBroadcasterClass();
+  static llvm::StringRef GetStaticBroadcasterClass();
 
-  ConstString &GetBroadcasterClass() const override {
+  llvm::StringRef GetBroadcasterClass() const override {
     return GetStaticBroadcasterClass();
   }
 
diff --git a/lldb/include/lldb/Target/TargetList.h b/lldb/include/lldb/Target/TargetList.h
index a0bc6f1f820b..a0cddc6b2966 100644
--- a/lldb/include/lldb/Target/TargetList.h
+++ b/lldb/include/lldb/Target/TargetList.h
@@ -37,9 +37,9 @@ public:
 
   // These two functions fill out the Broadcaster interface:
 
-  static ConstString &GetStaticBroadcasterClass();
+  static llvm::StringRef GetStaticBroadcasterClass();
 
-  ConstString &GetBroadcasterClass() const override {
+  llvm::StringRef GetBroadcasterClass() const override {
     return GetStaticBroadcasterClass();
   }
 
diff --git a/lldb/include/lldb/Target/Thread.h b/lldb/include/lldb/Target/Thread.h
index 1efef93b17de..c17bddf4d98b 100644
--- a/lldb/include/lldb/Target/Thread.h
+++ b/lldb/include/lldb/Target/Thread.h
@@ -74,9 +74,9 @@ public:
     eBroadcastBitThreadSelected = (1 << 4)
   };
 
-  static ConstString &GetStaticBroadcasterClass();
+  static llvm::StringRef GetStaticBroadcasterClass();
 
-  ConstString &GetBroadcasterClass() const override {
+  llvm::StringRef GetBroadcasterClass() const override {
     return GetStaticBroadcasterClass();
   }
 
diff --git a/lldb/include/lldb/Utility/Broadcaster.h b/lldb/include/lldb/Utility/Broadcaster.h
index f39e677fe9ee..58436ddb9f26 100644
--- a/lldb/include/lldb/Utility/Broadcaster.h
+++ b/lldb/include/lldb/Utility/Broadcaster.h
@@ -39,12 +39,12 @@ namespace lldb_private {
 /// Debugger maintains a list of BroadcastEventSpec's and when it is made
 class BroadcastEventSpec {
 public:
-  BroadcastEventSpec(const ConstString &broadcaster_class, uint32_t event_bits)
+  BroadcastEventSpec(llvm::StringRef broadcaster_class, uint32_t event_bits)
       : m_broadcaster_class(broadcaster_class), m_event_bits(event_bits) {}
 
   ~BroadcastEventSpec() = default;
 
-  ConstString GetBroadcasterClass() const { return m_broadcaster_class; }
+  const std::string &GetBroadcasterClass() const { return m_broadcaster_class; }
 
   uint32_t GetEventBits() const { return m_event_bits; }
 
@@ -67,7 +67,7 @@ public:
   bool operator<(const BroadcastEventSpec &rhs) const;
 
 private:
-  ConstString m_broadcaster_class;
+  std::string m_broadcaster_class;
   uint32_t m_event_bits;
 };
 
@@ -307,7 +307,7 @@ public:
   /// FIXME: Probably should make a ManagedBroadcaster subclass with all the
   /// bits needed to work with the BroadcasterManager, so that it is clearer
   /// how to add one.
-  virtual ConstString &GetBroadcasterClass() const;
+  virtual llvm::StringRef GetBroadcasterClass() const;
 
   lldb::BroadcasterManagerSP GetManager();
 
diff --git a/lldb/include/lldb/Utility/ProcessInfo.h b/lldb/include/lldb/Utility/ProcessInfo.h
index e9fe71e1b851..54ac000dc7fc 100644
--- a/lldb/include/lldb/Utility/ProcessInfo.h
+++ b/lldb/include/lldb/Utility/ProcessInfo.h
@@ -234,7 +234,7 @@ public:
 
   bool CumulativeSystemTimeIsValid() const {
     return m_cumulative_system_time.tv_sec > 0 ||
-           m_cumulative_system_time.tv_sec > 0;
+           m_cumulative_system_time.tv_usec > 0;
   }
 
   void Dump(Stream &s, UserIDResolver &resolver) const;
diff --git a/lldb/packages/Python/lldbsuite/test/tools/lldb-dap/dap_server.py b/lldb/packages/Python/lldbsuite/test/tools/lldb-dap/dap_server.py
index 27a76a652f40..5838281bcb1a 100644
--- a/lldb/packages/Python/lldbsuite/test/tools/lldb-dap/dap_server.py
+++ b/lldb/packages/Python/lldbsuite/test/tools/lldb-dap/dap_server.py
@@ -811,23 +811,34 @@ class DebugCommunication(object):
         command_dict = {"command": "next", "type": "request", "arguments": args_dict}
         return self.send_recv(command_dict)
 
-    def request_stepIn(self, threadId):
+    def request_stepIn(self, threadId, targetId):
         if self.exit_status is not None:
-            raise ValueError("request_continue called after process exited")
-        args_dict = {"threadId": threadId}
+            raise ValueError("request_stepIn called after process exited")
+        args_dict = {"threadId": threadId, "targetId": targetId}
         command_dict = {"command": "stepIn", "type": "request", "arguments": args_dict}
         return self.send_recv(command_dict)
 
+    def request_stepInTargets(self, frameId):
+        if self.exit_status is not None:
+            raise ValueError("request_stepInTargets called after process exited")
+        args_dict = {"frameId": frameId}
+        command_dict = {
+            "command": "stepInTargets",
+            "type": "request",
+            "arguments": args_dict,
+        }
+        return self.send_recv(command_dict)
+
     def request_stepOut(self, threadId):
         if self.exit_status is not None:
-            raise ValueError("request_continue called after process exited")
+            raise ValueError("request_stepOut called after process exited")
         args_dict = {"threadId": threadId}
         command_dict = {"command": "stepOut", "type": "request", "arguments": args_dict}
         return self.send_recv(command_dict)
 
     def request_pause(self, threadId=None):
         if self.exit_status is not None:
-            raise ValueError("request_continue called after process exited")
+            raise ValueError("request_pause called after process exited")
         if threadId is None:
             threadId = self.get_thread_id()
         args_dict = {"threadId": threadId}
diff --git a/lldb/packages/Python/lldbsuite/test/tools/lldb-dap/lldbdap_testcase.py b/lldb/packages/Python/lldbsuite/test/tools/lldb-dap/lldbdap_testcase.py
index 23f650d2d36f..d56ea5dca14b 100644
--- a/lldb/packages/Python/lldbsuite/test/tools/lldb-dap/lldbdap_testcase.py
+++ b/lldb/packages/Python/lldbsuite/test/tools/lldb-dap/lldbdap_testcase.py
@@ -218,8 +218,8 @@ class DAPTestCaseBase(TestBase):
         """Set a top level global variable only."""
         return self.dap_server.request_setVariable(2, name, str(value), id=id)
 
-    def stepIn(self, threadId=None, waitForStop=True):
-        self.dap_server.request_stepIn(threadId=threadId)
+    def stepIn(self, threadId=None, targetId=None, waitForStop=True):
+        self.dap_server.request_stepIn(threadId=threadId, targetId=targetId)
         if waitForStop:
             return self.dap_server.wait_for_stopped()
         return None
diff --git a/lldb/source/API/SBCommandInterpreter.cpp b/lldb/source/API/SBCommandInterpreter.cpp
index 7b87dc507e4b..83c0951c56db 100644
--- a/lldb/source/API/SBCommandInterpreter.cpp
+++ b/lldb/source/API/SBCommandInterpreter.cpp
@@ -512,7 +512,8 @@ SBBroadcaster SBCommandInterpreter::GetBroadcaster() {
 const char *SBCommandInterpreter::GetBroadcasterClass() {
   LLDB_INSTRUMENT();
 
-  return CommandInterpreter::GetStaticBroadcasterClass().AsCString();
+  return ConstString(CommandInterpreter::GetStaticBroadcasterClass())
+      .AsCString();
 }
 
 const char *SBCommandInterpreter::GetArgumentTypeAsCString(
diff --git a/lldb/source/API/SBCommunication.cpp b/lldb/source/API/SBCommunication.cpp
index f93898718be6..ee33e2abd854 100644
--- a/lldb/source/API/SBCommunication.cpp
+++ b/lldb/source/API/SBCommunication.cpp
@@ -170,5 +170,6 @@ SBBroadcaster SBCommunication::GetBroadcaster() {
 const char *SBCommunication::GetBroadcasterClass() {
   LLDB_INSTRUMENT();
 
-  return ThreadedCommunication::GetStaticBroadcasterClass().AsCString();
+  return ConstString(ThreadedCommunication::GetStaticBroadcasterClass())
+      .AsCString();
 }
diff --git a/lldb/source/API/SBDebugger.cpp b/lldb/source/API/SBDebugger.cpp
index fbcf30e67fc1..9c662dfbf441 100644
--- a/lldb/source/API/SBDebugger.cpp
+++ b/lldb/source/API/SBDebugger.cpp
@@ -112,7 +112,7 @@ SBDebugger &SBDebugger::operator=(const SBDebugger &rhs) {
 const char *SBDebugger::GetBroadcasterClass() {
   LLDB_INSTRUMENT();
 
-  return Debugger::GetStaticBroadcasterClass().AsCString();
+  return ConstString(Debugger::GetStaticBroadcasterClass()).AsCString();
 }
 
 const char *SBDebugger::GetProgressFromEvent(const lldb::SBEvent &event,
diff --git a/lldb/source/API/SBEvent.cpp b/lldb/source/API/SBEvent.cpp
index cc611449e250..aa9c0ff097d4 100644
--- a/lldb/source/API/SBEvent.cpp
+++ b/lldb/source/API/SBEvent.cpp
@@ -95,7 +95,8 @@ const char *SBEvent::GetBroadcasterClass() const {
 
   const Event *lldb_event = get();
   if (lldb_event)
-    return lldb_event->GetBroadcaster()->GetBroadcasterClass().AsCString();
+    return ConstString(lldb_event->GetBroadcaster()->GetBroadcasterClass())
+        .AsCString();
   else
     return "unknown class";
 }
diff --git a/lldb/source/API/SBLineEntry.cpp b/lldb/source/API/SBLineEntry.cpp
index 99a7b8fe644c..216ea6d18eab 100644
--- a/lldb/source/API/SBLineEntry.cpp
+++ b/lldb/source/API/SBLineEntry.cpp
@@ -67,6 +67,21 @@ SBAddress SBLineEntry::GetEndAddress() const {
   return sb_address;
 }
 
+SBAddress SBLineEntry::GetSameLineContiguousAddressRangeEnd(
+    bool include_inlined_functions) const {
+  LLDB_INSTRUMENT_VA(this);
+
+  SBAddress sb_address;
+  if (m_opaque_up) {
+    AddressRange line_range = m_opaque_up->GetSameLineContiguousAddressRange(
+        include_inlined_functions);
+
+    sb_address.SetAddress(line_range.GetBaseAddress());
+    sb_address.OffsetAddress(line_range.GetByteSize());
+  }
+  return sb_address;
+}
+
 bool SBLineEntry::IsValid() const {
   LLDB_INSTRUMENT_VA(this);
   return this->operator bool();
diff --git a/lldb/source/API/SBProcess.cpp b/lldb/source/API/SBProcess.cpp
index b80664882ebc..c37c111c5a58 100644
--- a/lldb/source/API/SBProcess.cpp
+++ b/lldb/source/API/SBProcess.cpp
@@ -77,7 +77,7 @@ SBProcess::~SBProcess() = default;
 const char *SBProcess::GetBroadcasterClassName() {
   LLDB_INSTRUMENT();
 
-  return Process::GetStaticBroadcasterClass().AsCString();
+  return ConstString(Process::GetStaticBroadcasterClass()).AsCString();
 }
 
 const char *SBProcess::GetPluginName() {
@@ -807,7 +807,7 @@ SBBroadcaster SBProcess::GetBroadcaster() const {
 const char *SBProcess::GetBroadcasterClass() {
   LLDB_INSTRUMENT();
 
-  return Process::GetStaticBroadcasterClass().AsCString();
+  return ConstString(Process::GetStaticBroadcasterClass()).AsCString();
 }
 
 size_t SBProcess::ReadMemory(addr_t addr, void *dst, size_t dst_len,
@@ -928,6 +928,14 @@ size_t SBProcess::WriteMemory(addr_t addr, const void *src, size_t src_len,
   return bytes_written;
 }
 
+void SBProcess::GetStatus(SBStream &status) {
+  LLDB_INSTRUMENT_VA(this, status);
+
+  ProcessSP process_sp(GetSP());
+  if (process_sp)
+    process_sp->GetStatus(status.ref());
+}
+
 bool SBProcess::GetDescription(SBStream &description) {
   LLDB_INSTRUMENT_VA(this, description);
 
diff --git a/lldb/source/API/SBTarget.cpp b/lldb/source/API/SBTarget.cpp
index cc9f1fdd76af..962ce9ba83cc 100644
--- a/lldb/source/API/SBTarget.cpp
+++ b/lldb/source/API/SBTarget.cpp
@@ -147,7 +147,7 @@ SBModule SBTarget::GetModuleAtIndexFromEvent(const uint32_t idx,
 const char *SBTarget::GetBroadcasterClassName() {
   LLDB_INSTRUMENT();
 
-  return Target::GetStaticBroadcasterClass().AsCString();
+  return ConstString(Target::GetStaticBroadcasterClass()).AsCString();
 }
 
 bool SBTarget::IsValid() const {
@@ -2011,6 +2011,30 @@ lldb::SBInstructionList SBTarget::ReadInstructions(lldb::SBAddress base_addr,
   return sb_instructions;
 }
 
+lldb::SBInstructionList SBTarget::ReadInstructions(lldb::SBAddress start_addr,
+                                                   lldb::SBAddress end_addr,
+                                                   const char *flavor_string) {
+  LLDB_INSTRUMENT_VA(this, start_addr, end_addr, flavor_string);
+
+  SBInstructionList sb_instructions;
+
+  TargetSP target_sp(GetSP());
+  if (target_sp) {
+    lldb::addr_t start_load_addr = start_addr.GetLoadAddress(*this);
+    lldb::addr_t end_load_addr = end_addr.GetLoadAddress(*this);
+    if (end_load_addr > start_load_addr) {
+      lldb::addr_t size = end_load_addr - start_load_addr;
+
+      AddressRange range(start_load_addr, size);
+      const bool force_live_memory = true;
+      sb_instructions.SetDisassembler(Disassembler::DisassembleRange(
+          target_sp->GetArchitecture(), nullptr, flavor_string, *target_sp,
+          range, force_live_memory));
+    }
+  }
+  return sb_instructions;
+}
+
 lldb::SBInstructionList SBTarget::GetInstructions(lldb::SBAddress base_addr,
                                                   const void *buf,
                                                   size_t size) {
diff --git a/lldb/source/API/SBThread.cpp b/lldb/source/API/SBThread.cpp
index eb9cf063802c..ac3e2cd25daa 100644
--- a/lldb/source/API/SBThread.cpp
+++ b/lldb/source/API/SBThread.cpp
@@ -53,7 +53,7 @@ using namespace lldb_private;
 const char *SBThread::GetBroadcasterClassName() {
   LLDB_INSTRUMENT();
 
-  return Thread::GetStaticBroadcasterClass().AsCString();
+  return ConstString(Thread::GetStaticBroadcasterClass()).AsCString();
 }
 
 // Constructors
diff --git a/lldb/source/API/SBType.cpp b/lldb/source/API/SBType.cpp
index ac0e56303fae..6cecb5c9ea81 100644
--- a/lldb/source/API/SBType.cpp
+++ b/lldb/source/API/SBType.cpp
@@ -7,16 +7,21 @@
 //===----------------------------------------------------------------------===//
 
 #include "lldb/API/SBType.h"
+#include "Utils.h"
 #include "lldb/API/SBDefines.h"
 #include "lldb/API/SBModule.h"
 #include "lldb/API/SBStream.h"
 #include "lldb/API/SBTypeEnumMember.h"
 #include "lldb/Core/Mangled.h"
+#include "lldb/Core/ValueObjectConstResult.h"
+#include "lldb/Symbol/CompilerDecl.h"
 #include "lldb/Symbol/CompilerType.h"
 #include "lldb/Symbol/Type.h"
 #include "lldb/Symbol/TypeSystem.h"
 #include "lldb/Utility/ConstString.h"
+#include "lldb/Utility/DataExtractor.h"
 #include "lldb/Utility/Instrumentation.h"
+#include "lldb/Utility/Scalar.h"
 #include "lldb/Utility/Stream.h"
 
 #include "llvm/ADT/APSInt.h"
@@ -325,6 +330,79 @@ lldb::SBTypeMemberFunction SBType::GetMemberFunctionAtIndex(uint32_t idx) {
   return sb_func_type;
 }
 
+SBTypeStaticField::SBTypeStaticField() { LLDB_INSTRUMENT_VA(this); }
+
+SBTypeStaticField::SBTypeStaticField(lldb_private::CompilerDecl decl)
+    : m_opaque_up(decl ? std::make_unique<CompilerDecl>(decl) : nullptr) {}
+
+SBTypeStaticField::SBTypeStaticField(const SBTypeStaticField &rhs) {
+  LLDB_INSTRUMENT_VA(this, rhs);
+
+  m_opaque_up = clone(rhs.m_opaque_up);
+}
+
+SBTypeStaticField &SBTypeStaticField::operator=(const SBTypeStaticField &rhs) {
+  LLDB_INSTRUMENT_VA(this, rhs);
+
+  m_opaque_up = clone(rhs.m_opaque_up);
+  return *this;
+}
+
+SBTypeStaticField::~SBTypeStaticField() { LLDB_INSTRUMENT_VA(this); }
+
+SBTypeStaticField::operator bool() const {
+  LLDB_INSTRUMENT_VA(this);
+
+  return IsValid();
+}
+
+bool SBTypeStaticField::IsValid() const {
+  LLDB_INSTRUMENT_VA(this);
+
+  return m_opaque_up != nullptr;
+}
+
+const char *SBTypeStaticField::GetName() {
+  LLDB_INSTRUMENT_VA(this);
+
+  if (!IsValid())
+    return "";
+  return m_opaque_up->GetName().GetCString();
+}
+
+const char *SBTypeStaticField::GetMangledName() {
+  LLDB_INSTRUMENT_VA(this);
+
+  if (!IsValid())
+    return "";
+  return m_opaque_up->GetMangledName().GetCString();
+}
+
+SBType SBTypeStaticField::GetType() {
+  LLDB_INSTRUMENT_VA(this);
+
+  if (!IsValid())
+    return SBType();
+  return SBType(m_opaque_up->GetType());
+}
+
+SBValue SBTypeStaticField::GetConstantValue(lldb::SBTarget target) {
+  LLDB_INSTRUMENT_VA(this, target);
+
+  if (!IsValid())
+    return SBValue();
+
+  Scalar value = m_opaque_up->GetConstantValue();
+  if (!value.IsValid())
+    return SBValue();
+  DataExtractor data;
+  value.GetData(data);
+  auto value_obj_sp = ValueObjectConstResult::Create(
+      target.GetSP().get(), m_opaque_up->GetType(), m_opaque_up->GetName(),
+      data);
+  return SBValue(std::move(value_obj_sp));
+}
+
 lldb::SBType SBType::GetUnqualifiedType() {
   LLDB_INSTRUMENT_VA(this);
 
@@ -438,6 +516,16 @@ SBTypeMember SBType::GetVirtualBaseClassAtIndex(uint32_t idx) {
   return sb_type_member;
 }
 
+SBTypeStaticField SBType::GetStaticFieldWithName(const char *name) {
+  LLDB_INSTRUMENT_VA(this, name);
+
+  if (!IsValid() || !name)
+    return SBTypeStaticField();
+
+  return SBTypeStaticField(m_opaque_sp->GetCompilerType(/*prefer_dynamic=*/true)
+                               .GetStaticFieldWithName(name));
+}
+
 SBTypeEnumMemberList SBType::GetEnumMembers() {
   LLDB_INSTRUMENT_VA(this);
 
diff --git a/lldb/source/API/SBValue.cpp b/lldb/source/API/SBValue.cpp
index 94a8f3ea319e..c53ec5a74648 100644
--- a/lldb/source/API/SBValue.cpp
+++ b/lldb/source/API/SBValue.cpp
@@ -909,6 +909,25 @@ uint64_t SBValue::GetValueAsUnsigned(uint64_t fail_value) {
   return fail_value;
 }
 
+lldb::addr_t SBValue::GetValueAsAddress() {
+  addr_t fail_value = LLDB_INVALID_ADDRESS;
+  ValueLocker locker;
+  lldb::ValueObjectSP value_sp(GetSP(locker));
+  if (value_sp) {
+    bool success = true;
+    uint64_t ret_val = fail_value;
+    ret_val = value_sp->GetValueAsUnsigned(fail_value, &success);
+    if (!success)
+      return fail_value;
+    ProcessSP process_sp = m_opaque_sp->GetProcessSP();
+    if (!process_sp)
+      return ret_val;
+    return process_sp->FixDataAddress(ret_val);
+  }
+
+  return fail_value;
+}
+
 bool SBValue::MightHaveChildren() {
   LLDB_INSTRUMENT_VA(this);
 
diff --git a/lldb/source/Core/Debugger.cpp b/lldb/source/Core/Debugger.cpp
index 19b3cf3bbf46..cac4642873b7 100644
--- a/lldb/source/Core/Debugger.cpp
+++ b/lldb/source/Core/Debugger.cpp
@@ -823,8 +823,8 @@ TargetSP Debugger::FindTargetWithProcess(Process *process) {
   return target_sp;
 }
 
-ConstString Debugger::GetStaticBroadcasterClass() {
-  static ConstString class_name("lldb.debugger");
+llvm::StringRef Debugger::GetStaticBroadcasterClass() {
+  static constexpr llvm::StringLiteral class_name("lldb.debugger");
   return class_name;
 }
 
@@ -846,7 +846,7 @@ Debugger::Debugger(lldb::LogOutputCallback log_callback, void *baton)
       m_loaded_plugins(), m_event_handler_thread(), m_io_handler_thread(),
       m_sync_broadcaster(nullptr, "lldb.debugger.sync"),
       m_broadcaster(m_broadcaster_manager_sp,
-                    GetStaticBroadcasterClass().AsCString()),
+                    GetStaticBroadcasterClass().str()),
       m_forward_listener_sp(), m_clear_once() {
   // Initialize the debugger properties as early as possible as other parts of
   // LLDB will start querying them during construction.
diff --git a/lldb/source/Core/ThreadedCommunication.cpp b/lldb/source/Core/ThreadedCommunication.cpp
index 7d8aae5d8ff6..2f3dada3ac93 100644
--- a/lldb/source/Core/ThreadedCommunication.cpp
+++ b/lldb/source/Core/ThreadedCommunication.cpp
@@ -32,8 +32,8 @@
 using namespace lldb;
 using namespace lldb_private;
 
-ConstString &ThreadedCommunication::GetStaticBroadcasterClass() {
-  static ConstString class_name("lldb.communication");
+llvm::StringRef ThreadedCommunication::GetStaticBroadcasterClass() {
+  static constexpr llvm::StringLiteral class_name("lldb.communication");
   return class_name;
 }
 
diff --git a/lldb/source/Interpreter/CommandInterpreter.cpp b/lldb/source/Interpreter/CommandInterpreter.cpp
index 8c3972a2ba4c..4c58ecc3c184 100644
--- a/lldb/source/Interpreter/CommandInterpreter.cpp
+++ b/lldb/source/Interpreter/CommandInterpreter.cpp
@@ -119,15 +119,15 @@ enum {
 #include "InterpreterPropertiesEnum.inc"
 };
 
-ConstString &CommandInterpreter::GetStaticBroadcasterClass() {
-  static ConstString class_name("lldb.commandInterpreter");
+llvm::StringRef CommandInterpreter::GetStaticBroadcasterClass() {
+  static constexpr llvm::StringLiteral class_name("lldb.commandInterpreter");
   return class_name;
 }
 
 CommandInterpreter::CommandInterpreter(Debugger &debugger,
                                        bool synchronous_execution)
     : Broadcaster(debugger.GetBroadcasterManager(),
-                  CommandInterpreter::GetStaticBroadcasterClass().AsCString()),
+                  CommandInterpreter::GetStaticBroadcasterClass().str()),
       Properties(
           OptionValuePropertiesSP(new OptionValueProperties("interpreter"))),
       IOHandlerDelegate(IOHandlerDelegate::Completion::LLDBCommand),
diff --git a/lldb/source/Plugins/ExpressionParser/Clang/ASTUtils.h b/lldb/source/Plugins/ExpressionParser/Clang/ASTUtils.h
index cefec15a7980..17f1506036c6 100644
--- a/lldb/source/Plugins/ExpressionParser/Clang/ASTUtils.h
+++ b/lldb/source/Plugins/ExpressionParser/Clang/ASTUtils.h
@@ -30,7 +30,7 @@ public:
 
   ~ExternalASTSourceWrapper() override;
 
-  clang::Decl *GetExternalDecl(uint32_t ID) override {
+  clang::Decl *GetExternalDecl(clang::GlobalDeclID ID) override {
     return m_Source->GetExternalDecl(ID);
   }
 
@@ -266,7 +266,7 @@ public:
   // ExternalASTSource.
   //===--------------------------------------------------------------------===//
 
-  clang::Decl *GetExternalDecl(uint32_t ID) override {
+  clang::Decl *GetExternalDecl(clang::GlobalDeclID ID) override {
     for (size_t i = 0; i < Sources.size(); ++i)
       if (clang::Decl *Result = Sources[i]->GetExternalDecl(ID))
         return Result;
diff --git a/lldb/source/Plugins/ExpressionParser/Clang/ClangASTSource.h b/lldb/source/Plugins/ExpressionParser/Clang/ClangASTSource.h
index f34e4661a81c..83c910477acc 100644
--- a/lldb/source/Plugins/ExpressionParser/Clang/ClangASTSource.h
+++ b/lldb/source/Plugins/ExpressionParser/Clang/ClangASTSource.h
@@ -49,7 +49,7 @@ public:
   ~ClangASTSource() override;
 
   /// Interface stubs.
-  clang::Decl *GetExternalDecl(uint32_t) override { return nullptr; }
+  clang::Decl *GetExternalDecl(clang::GlobalDeclID) override { return nullptr; }
   clang::Stmt *GetExternalDeclStmt(uint64_t) override { return nullptr; }
   clang::Selector GetExternalSelector(uint32_t) override {
     return clang::Selector();
diff --git a/lldb/source/Plugins/LanguageRuntime/ObjC/AppleObjCRuntime/AppleObjCRuntimeV2.cpp b/lldb/source/Plugins/LanguageRuntime/ObjC/AppleObjCRuntime/AppleObjCRuntimeV2.cpp
index d3fc487aed43..9409497f1c81 100644
--- a/lldb/source/Plugins/LanguageRuntime/ObjC/AppleObjCRuntime/AppleObjCRuntimeV2.cpp
+++ b/lldb/source/Plugins/LanguageRuntime/ObjC/AppleObjCRuntime/AppleObjCRuntimeV2.cpp
@@ -1869,15 +1869,15 @@ AppleObjCRuntimeV2::DynamicClassInfoExtractor::ComputeHelper(
       if (loader->IsFullyInitialized()) {
         switch (exe_ctx.GetTargetRef().GetDynamicClassInfoHelper()) {
         case eDynamicClassInfoHelperAuto:
-          LLVM_FALLTHROUGH;
+          [[fallthrough]];
         case eDynamicClassInfoHelperGetRealizedClassList:
           if (m_runtime.m_has_objc_getRealizedClassList_trylock)
             return DynamicClassInfoExtractor::objc_getRealizedClassList_trylock;
-          LLVM_FALLTHROUGH;
+          [[fallthrough]];
         case eDynamicClassInfoHelperCopyRealizedClassList:
           if (m_runtime.m_has_objc_copyRealizedClassList)
             return DynamicClassInfoExtractor::objc_copyRealizedClassList;
-          LLVM_FALLTHROUGH;
+          [[fallthrough]];
         case eDynamicClassInfoHelperRealizedClassesStruct:
           return DynamicClassInfoExtractor::gdb_objc_realized_classes;
         }
diff --git a/lldb/source/Plugins/Process/gdb-remote/GDBRemoteCommunicationServerLLGS.cpp b/lldb/source/Plugins/Process/gdb-remote/GDBRemoteCommunicationServerLLGS.cpp
index 3d37bb226a65..ae1a77e5be83 100644
--- a/lldb/source/Plugins/Process/gdb-remote/GDBRemoteCommunicationServerLLGS.cpp
+++ b/lldb/source/Plugins/Process/gdb-remote/GDBRemoteCommunicationServerLLGS.cpp
@@ -2087,7 +2087,7 @@ void GDBRemoteCommunicationServerLLGS::AddProcessThreads(
 GDBRemoteCommunication::PacketResult
 GDBRemoteCommunicationServerLLGS::Handle_qfThreadInfo(
     StringExtractorGDBRemote &packet) {
-  assert(m_debugged_processes.size() == 1 ||
+  assert(m_debugged_processes.size() <= 1 ||
          bool(m_extensions_supported &
               NativeProcessProtocol::Extension::multiprocess));
 
diff --git a/lldb/source/Plugins/ScriptInterpreter/Python/PythonDataObjects.cpp b/lldb/source/Plugins/ScriptInterpreter/Python/PythonDataObjects.cpp
index ea0a1cdff40f..7c7035e0c86c 100644
--- a/lldb/source/Plugins/ScriptInterpreter/Python/PythonDataObjects.cpp
+++ b/lldb/source/Plugins/ScriptInterpreter/Python/PythonDataObjects.cpp
@@ -61,7 +61,7 @@ Expected<std::string> python::As<std::string>(Expected<PythonObject> &&obj) {
   if (!obj)
     return obj.takeError();
   PyObject *str_obj = PyObject_Str(obj.get().get());
-  if (!obj)
+  if (!str_obj)
     return llvm::make_error<PythonException>();
   auto str = Take<PythonString>(str_obj);
   auto utf8 = str.AsUTF8();
diff --git a/lldb/source/Plugins/SymbolFile/CTF/SymbolFileCTF.cpp b/lldb/source/Plugins/SymbolFile/CTF/SymbolFileCTF.cpp
index 65f5b1a5f1b0..386ba44c5ea6 100644
--- a/lldb/source/Plugins/SymbolFile/CTF/SymbolFileCTF.cpp
+++ b/lldb/source/Plugins/SymbolFile/CTF/SymbolFileCTF.cpp
@@ -342,7 +342,7 @@ SymbolFileCTF::CreateInteger(const CTFInteger &ctf_integer) {
 
   CompilerType compiler_type = m_ast->GetBasicType(basic_type);
 
-  if (basic_type != eBasicTypeVoid) {
+  if (basic_type != eBasicTypeVoid && basic_type != eBasicTypeBool) {
     // Make sure the type we got is an integer type.
     bool compiler_type_is_signed = false;
     if (!compiler_type.IsIntegerType(compiler_type_is_signed))
@@ -802,7 +802,8 @@ size_t SymbolFileCTF::ParseFunctions(CompileUnit &cu) {
       }
 
       Type *arg_type = ResolveTypeUID(arg_uid);
-      arg_types.push_back(arg_type->GetFullCompilerType());
+      arg_types.push_back(arg_type ? arg_type->GetFullCompilerType()
+                                   : CompilerType());
     }
 
     if (symbol) {
@@ -813,8 +814,9 @@ size_t SymbolFileCTF::ParseFunctions(CompileUnit &cu) {
 
       // Create function type.
       CompilerType func_type = m_ast->CreateFunctionType(
-          ret_type->GetFullCompilerType(), arg_types.data(), arg_types.size(),
-          is_variadic, 0, clang::CallingConv::CC_C);
+          ret_type ? ret_type->GetFullCompilerType() : CompilerType(),
+          arg_types.data(), arg_types.size(), is_variadic, 0,
+          clang::CallingConv::CC_C);
       lldb::user_id_t function_type_uid = m_types.size() + 1;
       TypeSP type_sp =
           MakeType(function_type_uid, symbol->GetName(), 0, nullptr,
diff --git a/lldb/source/Plugins/SymbolFile/DWARF/DWARFASTParserClang.cpp b/lldb/source/Plugins/SymbolFile/DWARF/DWARFASTParserClang.cpp
index 41d81fbcf1b0..12dafd3f5d5d 100644
--- a/lldb/source/Plugins/SymbolFile/DWARF/DWARFASTParserClang.cpp
+++ b/lldb/source/Plugins/SymbolFile/DWARF/DWARFASTParserClang.cpp
@@ -495,6 +495,7 @@ TypeSP DWARFASTParserClang::ParseTypeFromDWARF(const SymbolContext &sc,
   case DW_TAG_const_type:
   case DW_TAG_restrict_type:
   case DW_TAG_volatile_type:
+  case DW_TAG_LLVM_ptrauth_type:
   case DW_TAG_atomic_type:
   case DW_TAG_unspecified_type: {
     type_sp = ParseTypeModifier(sc, die, attrs);
diff --git a/lldb/source/Plugins/SymbolFile/DWARF/DWARFCompileUnit.h b/lldb/source/Plugins/SymbolFile/DWARF/DWARFCompileUnit.h
index dd130977d4b1..b8344f548ac3 100644
--- a/lldb/source/Plugins/SymbolFile/DWARF/DWARFCompileUnit.h
+++ b/lldb/source/Plugins/SymbolFile/DWARF/DWARFCompileUnit.h
@@ -32,7 +32,7 @@ public:
 
 private:
   DWARFCompileUnit(SymbolFileDWARF &dwarf, lldb::user_id_t uid,
-                   const DWARFUnitHeader &header,
+                   const llvm::DWARFUnitHeader &header,
                    const llvm::DWARFAbbreviationDeclarationSet &abbrevs,
                    DIERef::Section section, bool is_dwo)
       : DWARFUnit(dwarf, uid, header, abbrevs, section, is_dwo) {}
diff --git a/lldb/source/Plugins/SymbolFile/DWARF/DWARFTypeUnit.h b/lldb/source/Plugins/SymbolFile/DWARF/DWARFTypeUnit.h
index 7b58c632c6c5..8c1f932d8c7f 100644
--- a/lldb/source/Plugins/SymbolFile/DWARF/DWARFTypeUnit.h
+++ b/lldb/source/Plugins/SymbolFile/DWARF/DWARFTypeUnit.h
@@ -24,15 +24,15 @@ public:
 
   void Dump(Stream *s) const override;
 
-  uint64_t GetTypeHash() { return m_header.GetTypeHash(); }
+  uint64_t GetTypeHash() { return m_header.getTypeHash(); }
 
-  dw_offset_t GetTypeOffset() { return GetOffset() + m_header.GetTypeOffset(); }
+  dw_offset_t GetTypeOffset() { return GetOffset() + m_header.getTypeOffset(); }
 
   static bool classof(const DWARFUnit *unit) { return unit->IsTypeUnit(); }
 
 private:
   DWARFTypeUnit(SymbolFileDWARF &dwarf, lldb::user_id_t uid,
-                const DWARFUnitHeader &header,
+                const llvm::DWARFUnitHeader &header,
                 const llvm::DWARFAbbreviationDeclarationSet &abbrevs,
                 DIERef::Section section, bool is_dwo)
       : DWARFUnit(dwarf, uid, header, abbrevs, section, is_dwo) {}
diff --git a/lldb/source/Plugins/SymbolFile/DWARF/DWARFUnit.cpp b/lldb/source/Plugins/SymbolFile/DWARF/DWARFUnit.cpp
index e28036d34b34..dabc595427df 100644
--- a/lldb/source/Plugins/SymbolFile/DWARF/DWARFUnit.cpp
+++ b/lldb/source/Plugins/SymbolFile/DWARF/DWARFUnit.cpp
@@ -33,12 +33,12 @@ using namespace lldb_private::plugin::dwarf;
 extern int g_verbose;
 
 DWARFUnit::DWARFUnit(SymbolFileDWARF &dwarf, lldb::user_id_t uid,
-                     const DWARFUnitHeader &header,
+                     const llvm::DWARFUnitHeader &header,
                      const llvm::DWARFAbbreviationDeclarationSet &abbrevs,
                      DIERef::Section section, bool is_dwo)
     : UserID(uid), m_dwarf(dwarf), m_header(header), m_abbrevs(&abbrevs),
       m_cancel_scopes(false), m_section(section), m_is_dwo(is_dwo),
-      m_has_parsed_non_skeleton_unit(false), m_dwo_id(header.GetDWOId()) {}
+      m_has_parsed_non_skeleton_unit(false), m_dwo_id(header.getDWOId()) {}
 
 DWARFUnit::~DWARFUnit() = default;
 
@@ -345,7 +345,7 @@ void DWARFUnit::ExtractDIEsRWLocked() {
 void DWARFUnit::SetDwoStrOffsetsBase() {
   lldb::offset_t baseOffset = 0;
 
-  if (const llvm::DWARFUnitIndex::Entry *entry = m_header.GetIndexEntry()) {
+  if (const llvm::DWARFUnitIndex::Entry *entry = m_header.getIndexEntry()) {
     if (const auto *contribution =
             entry->getContribution(llvm::DW_SECT_STR_OFFSETS))
       baseOffset = contribution->getOffset();
@@ -489,7 +489,7 @@ ParseListTableHeader(const llvm::DWARFDataExtractor &data, uint64_t offset,
 
 void DWARFUnit::SetLoclistsBase(dw_addr_t loclists_base) {
   uint64_t offset = 0;
-  if (const llvm::DWARFUnitIndex::Entry *entry = m_header.GetIndexEntry()) {
+  if (const llvm::DWARFUnitIndex::Entry *entry = m_header.getIndexEntry()) {
     const auto *contribution = entry->getContribution(llvm::DW_SECT_LOCLISTS);
     if (!contribution) {
       GetSymbolFileDWARF().GetObjectFile()->GetModule()->ReportError(
@@ -533,7 +533,7 @@ DWARFDataExtractor DWARFUnit::GetLocationData() const {
   DWARFContext &Ctx = GetSymbolFileDWARF().GetDWARFContext();
   const DWARFDataExtractor &data =
       GetVersion() >= 5 ? Ctx.getOrLoadLocListsData() : Ctx.getOrLoadLocData();
-  if (const llvm::DWARFUnitIndex::Entry *entry = m_header.GetIndexEntry()) {
+  if (const llvm::DWARFUnitIndex::Entry *entry = m_header.getIndexEntry()) {
     if (const auto *contribution = entry->getContribution(
             GetVersion() >= 5 ? llvm::DW_SECT_LOCLISTS : llvm::DW_SECT_EXT_LOC))
       return DWARFDataExtractor(data, contribution->getOffset(),
@@ -546,7 +546,7 @@ DWARFDataExtractor DWARFUnit::GetLocationData() const {
 DWARFDataExtractor DWARFUnit::GetRnglistData() const {
   DWARFContext &Ctx = GetSymbolFileDWARF().GetDWARFContext();
   const DWARFDataExtractor &data = Ctx.getOrLoadRngListsData();
-  if (const llvm::DWARFUnitIndex::Entry *entry = m_header.GetIndexEntry()) {
+  if (const llvm::DWARFUnitIndex::Entry *entry = m_header.getIndexEntry()) {
     if (const auto *contribution =
             entry->getContribution(llvm::DW_SECT_RNGLISTS))
       return DWARFDataExtractor(data, contribution->getOffset(),
@@ -924,84 +924,6 @@ const DWARFDebugAranges &DWARFUnit::GetFunctionAranges() {
   return *m_func_aranges_up;
 }
 
-llvm::Error DWARFUnitHeader::ApplyIndexEntry(
-    const llvm::DWARFUnitIndex::Entry *index_entry) {
-  // We should only be calling this function when the index entry is not set and
-  // we have a valid one to set it to.
-  assert(index_entry);
-  assert(!m_index_entry);
-
-  if (m_abbr_offset)
-    return llvm::createStringError(
-        llvm::inconvertibleErrorCode(),
-        "Package unit with a non-zero abbreviation offset");
-
-  auto *unit_contrib = index_entry->getContribution();
-  if (!unit_contrib || unit_contrib->getLength32() != m_length + 4)
-    return llvm::createStringError(llvm::inconvertibleErrorCode(),
-                                   "Inconsistent DWARF package unit index");
-
-  auto *abbr_entry = index_entry->getContribution(llvm::DW_SECT_ABBREV);
-  if (!abbr_entry)
-    return llvm::createStringError(
-        llvm::inconvertibleErrorCode(),
-        "DWARF package index missing abbreviation column");
-
-  m_abbr_offset = abbr_entry->getOffset();
-  m_index_entry = index_entry;
-  return llvm::Error::success();
-}
-
-llvm::Expected<DWARFUnitHeader>
-DWARFUnitHeader::extract(const DWARFDataExtractor &data,
-                         DIERef::Section section, DWARFContext &context,
-                         lldb::offset_t *offset_ptr) {
-  DWARFUnitHeader header;
-  header.m_offset = *offset_ptr;
-  header.m_length = data.GetDWARFInitialLength(offset_ptr);
-  header.m_version = data.GetU16(offset_ptr);
-  if (header.m_version == 5) {
-    header.m_unit_type = data.GetU8(offset_ptr);
-    header.m_addr_size = data.GetU8(offset_ptr);
-    header.m_abbr_offset = data.GetDWARFOffset(offset_ptr);
-    if (header.m_unit_type == llvm::dwarf::DW_UT_skeleton ||
-        header.m_unit_type == llvm::dwarf::DW_UT_split_compile)
-      header.m_dwo_id = data.GetU64(offset_ptr);
-  } else {
-    header.m_abbr_offset = data.GetDWARFOffset(offset_ptr);
-    header.m_addr_size = data.GetU8(offset_ptr);
-    header.m_unit_type =
-        section == DIERef::Section::DebugTypes ? DW_UT_type : DW_UT_compile;
-  }
-
-  if (header.IsTypeUnit()) {
-    header.m_type_hash = data.GetU64(offset_ptr);
-    header.m_type_offset = data.GetDWARFOffset(offset_ptr);
-  }
-
-  bool length_OK = data.ValidOffset(header.GetNextUnitOffset() - 1);
-  bool version_OK = SymbolFileDWARF::SupportedVersion(header.m_version);
-  bool addr_size_OK = (header.m_addr_size == 2) || (header.m_addr_size == 4) ||
-                      (header.m_addr_size == 8);
-  bool type_offset_OK =
-      !header.IsTypeUnit() || (header.m_type_offset <= header.GetLength());
-
-  if (!length_OK)
-    return llvm::make_error<llvm::object::GenericBinaryError>(
-        "Invalid unit length");
-  if (!version_OK)
-    return llvm::make_error<llvm::object::GenericBinaryError>(
-        "Unsupported unit version");
-  if (!addr_size_OK)
-    return llvm::make_error<llvm::object::GenericBinaryError>(
-        "Invalid unit address size");
-  if (!type_offset_OK)
-    return llvm::make_error<llvm::object::GenericBinaryError>(
-        "Type offset out of range");
-
-  return header;
-}
-
 llvm::Expected<DWARFUnitSP>
 DWARFUnit::extract(SymbolFileDWARF &dwarf, user_id_t uid,
                    const DWARFDataExtractor &debug_info,
@@ -1009,26 +931,35 @@ DWARFUnit::extract(SymbolFileDWARF &dwarf, user_id_t uid,
   assert(debug_info.ValidOffset(*offset_ptr));
 
   DWARFContext &context = dwarf.GetDWARFContext();
-  auto expected_header =
-      DWARFUnitHeader::extract(debug_info, section, context, offset_ptr);
-  if (!expected_header)
-    return expected_header.takeError();
+
+  // FIXME: Either properly map between DIERef::Section and
+  // llvm::DWARFSectionKind or switch to llvm's definition entirely.
+  llvm::DWARFSectionKind section_kind_llvm =
+      section == DIERef::Section::DebugInfo
+          ? llvm::DWARFSectionKind::DW_SECT_INFO
+          : llvm::DWARFSectionKind::DW_SECT_EXT_TYPES;
+
+  llvm::DWARFDataExtractor debug_info_llvm = debug_info.GetAsLLVMDWARF();
+  llvm::DWARFUnitHeader header;
+  if (llvm::Error extract_err = header.extract(
+          context.GetAsLLVM(), debug_info_llvm, offset_ptr, section_kind_llvm))
+    return std::move(extract_err);
 
   if (context.isDwo()) {
     const llvm::DWARFUnitIndex::Entry *entry = nullptr;
-    const llvm::DWARFUnitIndex &index = expected_header->IsTypeUnit()
+    const llvm::DWARFUnitIndex &index = header.isTypeUnit()
                                             ? context.GetAsLLVM().getTUIndex()
                                             : context.GetAsLLVM().getCUIndex();
     if (index) {
-      if (expected_header->IsTypeUnit())
-        entry = index.getFromHash(expected_header->GetTypeHash());
-      else if (auto dwo_id = expected_header->GetDWOId())
+      if (header.isTypeUnit())
+        entry = index.getFromHash(header.getTypeHash());
+      else if (auto dwo_id = header.getDWOId())
         entry = index.getFromHash(*dwo_id);
     }
     if (!entry)
-      entry = index.getFromOffset(expected_header->GetOffset());
+      entry = index.getFromOffset(header.getOffset());
     if (entry)
-      if (llvm::Error err = expected_header->ApplyIndexEntry(entry))
+      if (llvm::Error err = header.applyIndexEntry(entry))
         return std::move(err);
   }
 
@@ -1039,13 +970,13 @@ DWARFUnit::extract(SymbolFileDWARF &dwarf, user_id_t uid,
 
   bool abbr_offset_OK =
       dwarf.GetDWARFContext().getOrLoadAbbrevData().ValidOffset(
-          expected_header->GetAbbrOffset());
+          header.getAbbrOffset());
   if (!abbr_offset_OK)
     return llvm::make_error<llvm::object::GenericBinaryError>(
         "Abbreviation offset for unit is not valid");
 
   llvm::Expected<const llvm::DWARFAbbreviationDeclarationSet *> abbrevs_or_err =
-      abbr->getAbbreviationDeclarationSet(expected_header->GetAbbrOffset());
+      abbr->getAbbreviationDeclarationSet(header.getAbbrOffset());
   if (!abbrevs_or_err)
     return abbrevs_or_err.takeError();
 
@@ -1055,11 +986,11 @@ DWARFUnit::extract(SymbolFileDWARF &dwarf, user_id_t uid,
         "No abbrev exists at the specified offset.");
 
   bool is_dwo = dwarf.GetDWARFContext().isDwo();
-  if (expected_header->IsTypeUnit())
-    return DWARFUnitSP(new DWARFTypeUnit(dwarf, uid, *expected_header, *abbrevs,
-                                         section, is_dwo));
-  return DWARFUnitSP(new DWARFCompileUnit(dwarf, uid, *expected_header,
-                                          *abbrevs, section, is_dwo));
+  if (header.isTypeUnit())
+    return DWARFUnitSP(
+        new DWARFTypeUnit(dwarf, uid, header, *abbrevs, section, is_dwo));
+  return DWARFUnitSP(
+      new DWARFCompileUnit(dwarf, uid, header, *abbrevs, section, is_dwo));
 }
 
 const lldb_private::DWARFDataExtractor &DWARFUnit::GetData() const {
@@ -1069,7 +1000,7 @@ const lldb_private::DWARFDataExtractor &DWARFUnit::GetData() const {
 }
 
 uint32_t DWARFUnit::GetHeaderByteSize() const {
-  switch (m_header.GetUnitType()) {
+  switch (m_header.getUnitType()) {
   case llvm::dwarf::DW_UT_compile:
   case llvm::dwarf::DW_UT_partial:
     return GetVersion() < 5 ? 11 : 12;
@@ -1106,7 +1037,7 @@ DWARFUnit::FindRnglistFromOffset(dw_offset_t offset) {
   llvm::DWARFDataExtractor data = GetRnglistData().GetAsLLVMDWARF();
 
   // As DW_AT_rnglists_base may be missing we need to call setAddressSize.
-  data.setAddressSize(m_header.GetAddressByteSize());
+  data.setAddressSize(m_header.getAddressByteSize());
   auto range_list_or_error = GetRnglistTable()->findList(data, offset);
   if (!range_list_or_error)
     return range_list_or_error.takeError();
diff --git a/lldb/source/Plugins/SymbolFile/DWARF/DWARFUnit.h b/lldb/source/Plugins/SymbolFile/DWARF/DWARFUnit.h
index 28981b51bfcb..85c37971ced8 100644
--- a/lldb/source/Plugins/SymbolFile/DWARF/DWARFUnit.h
+++ b/lldb/source/Plugins/SymbolFile/DWARF/DWARFUnit.h
@@ -38,54 +38,6 @@ enum DWARFProducer {
   eProducerOther
 };
 
-/// Base class describing the header of any kind of "unit."  Some information
-/// is specific to certain unit types.  We separate this class out so we can
-/// parse the header before deciding what specific kind of unit to construct.
-class DWARFUnitHeader {
-  dw_offset_t m_offset = 0;
-  dw_offset_t m_length = 0;
-  uint16_t m_version = 0;
-  dw_offset_t m_abbr_offset = 0;
-
-  const llvm::DWARFUnitIndex::Entry *m_index_entry = nullptr;
-
-  uint8_t m_unit_type = 0;
-  uint8_t m_addr_size = 0;
-
-  uint64_t m_type_hash = 0;
-  uint32_t m_type_offset = 0;
-
-  std::optional<uint64_t> m_dwo_id;
-
-  DWARFUnitHeader() = default;
-
-public:
-  dw_offset_t GetOffset() const { return m_offset; }
-  uint16_t GetVersion() const { return m_version; }
-  uint16_t GetAddressByteSize() const { return m_addr_size; }
-  dw_offset_t GetLength() const { return m_length; }
-  dw_offset_t GetAbbrOffset() const { return m_abbr_offset; }
-  uint8_t GetUnitType() const { return m_unit_type; }
-  const llvm::DWARFUnitIndex::Entry *GetIndexEntry() const {
-    return m_index_entry;
-  }
-  uint64_t GetTypeHash() const { return m_type_hash; }
-  dw_offset_t GetTypeOffset() const { return m_type_offset; }
-  std::optional<uint64_t> GetDWOId() const { return m_dwo_id; }
-  bool IsTypeUnit() const {
-    return m_unit_type == llvm::dwarf::DW_UT_type ||
-           m_unit_type == llvm::dwarf::DW_UT_split_type;
-  }
-  dw_offset_t GetNextUnitOffset() const { return m_offset + m_length + 4; }
-
-  llvm::Error ApplyIndexEntry(const llvm::DWARFUnitIndex::Entry *index_entry);
-
-  static llvm::Expected<DWARFUnitHeader> extract(const DWARFDataExtractor &data,
-                                                 DIERef::Section section,
-                                                 DWARFContext &dwarf_context,
-                                                 lldb::offset_t *offset_ptr);
-};
-
 class DWARFUnit : public UserID {
   using die_iterator_range =
       llvm::iterator_range<DWARFDebugInfoEntry::collection::iterator>;
@@ -105,7 +57,7 @@ public:
   /// the DWO ID in the compile unit header and we sometimes only want to access
   /// this cheap value without causing the more expensive attribute fetches that
   /// GetDWOId() uses.
-  std::optional<uint64_t> GetHeaderDWOId() { return m_header.GetDWOId(); }
+  std::optional<uint64_t> GetHeaderDWOId() { return m_header.getDWOId(); }
   void ExtractUnitDIEIfNeeded();
   void ExtractUnitDIENoDwoIfNeeded();
   void ExtractDIEsIfNeeded();
@@ -143,7 +95,7 @@ public:
   uint32_t GetHeaderByteSize() const;
 
   // Offset of the initial length field.
-  dw_offset_t GetOffset() const { return m_header.GetOffset(); }
+  dw_offset_t GetOffset() const { return m_header.getOffset(); }
   /// Get the size in bytes of the length field in the header.
   ///
   /// In DWARF32 this is just 4 bytes
@@ -159,15 +111,15 @@ public:
   dw_offset_t GetFirstDIEOffset() const {
     return GetOffset() + GetHeaderByteSize();
   }
-  dw_offset_t GetNextUnitOffset() const { return m_header.GetNextUnitOffset(); }
+  dw_offset_t GetNextUnitOffset() const { return m_header.getNextUnitOffset(); }
   // Size of the CU data (without initial length and without header).
   size_t GetDebugInfoSize() const;
   // Size of the CU data incl. header but without initial length.
-  dw_offset_t GetLength() const { return m_header.GetLength(); }
-  uint16_t GetVersion() const { return m_header.GetVersion(); }
+  dw_offset_t GetLength() const { return m_header.getLength(); }
+  uint16_t GetVersion() const { return m_header.getVersion(); }
   const llvm::DWARFAbbreviationDeclarationSet *GetAbbreviations() const;
   dw_offset_t GetAbbrevOffset() const;
-  uint8_t GetAddressByteSize() const { return m_header.GetAddressByteSize(); }
+  uint8_t GetAddressByteSize() const { return m_header.getAddressByteSize(); }
   dw_addr_t GetAddrBase() const { return m_addr_base.value_or(0); }
   dw_addr_t GetBaseAddress() const { return m_base_addr; }
   dw_offset_t GetLineTableOffset();
@@ -250,8 +202,8 @@ public:
 
   DIERef::Section GetDebugSection() const { return m_section; }
 
-  uint8_t GetUnitType() const { return m_header.GetUnitType(); }
-  bool IsTypeUnit() const { return m_header.IsTypeUnit(); }
+  uint8_t GetUnitType() const { return m_header.getUnitType(); }
+  bool IsTypeUnit() const { return m_header.isTypeUnit(); }
   /// Note that this check only works for DWARF5+.
   bool IsSkeletonUnit() const {
     return GetUnitType() == llvm::dwarf::DW_UT_skeleton;
@@ -320,7 +272,7 @@ public:
 
 protected:
   DWARFUnit(SymbolFileDWARF &dwarf, lldb::user_id_t uid,
-            const DWARFUnitHeader &header,
+            const llvm::DWARFUnitHeader &header,
             const llvm::DWARFAbbreviationDeclarationSet &abbrevs,
             DIERef::Section section, bool is_dwo);
 
@@ -352,7 +304,7 @@ protected:
 
   SymbolFileDWARF &m_dwarf;
   std::shared_ptr<DWARFUnit> m_dwo;
-  DWARFUnitHeader m_header;
+  llvm::DWARFUnitHeader m_header;
   const llvm::DWARFAbbreviationDeclarationSet *m_abbrevs = nullptr;
   lldb_private::CompileUnit *m_lldb_cu = nullptr;
   // If this is a DWO file, we have a backlink to our skeleton compile unit.
diff --git a/lldb/source/Plugins/TypeSystem/Clang/TypeSystemClang.cpp b/lldb/source/Plugins/TypeSystem/Clang/TypeSystemClang.cpp
index 2621f682011b..8fc0f9103f55 100644
--- a/lldb/source/Plugins/TypeSystem/Clang/TypeSystemClang.cpp
+++ b/lldb/source/Plugins/TypeSystem/Clang/TypeSystemClang.cpp
@@ -9,6 +9,7 @@
 #include "TypeSystemClang.h"
 
 #include "clang/AST/DeclBase.h"
+#include "clang/AST/ExprCXX.h"
 #include "llvm/Support/Casting.h"
 #include "llvm/Support/FormatAdapters.h"
 #include "llvm/Support/FormatVariadic.h"
@@ -838,8 +839,11 @@ lldb::BasicType TypeSystemClang::GetBasicTypeEnumeration(llvm::StringRef name) {
       {"__int128_t", eBasicTypeInt128},
       {"__uint128_t", eBasicTypeUnsignedInt128},
 
-      // Miscellaneous
+      // "bool"
       {"bool", eBasicTypeBool},
+      {"_Bool", eBasicTypeBool},
+
+      // Miscellaneous
       {"float", eBasicTypeFloat},
       {"double", eBasicTypeDouble},
       {"long double", eBasicTypeLongDouble},
@@ -1145,6 +1149,8 @@ CompilerType TypeSystemClang::GetTypeForDecl(clang::NamedDecl *decl) {
     return GetTypeForDecl(interface_decl);
   if (clang::TagDecl *tag_decl = llvm::dyn_cast<clang::TagDecl>(decl))
     return GetTypeForDecl(tag_decl);
+  if (clang::ValueDecl *value_decl = llvm::dyn_cast<clang::ValueDecl>(decl))
+    return GetTypeForDecl(value_decl);
   return CompilerType();
 }
 
@@ -1156,6 +1162,10 @@ CompilerType TypeSystemClang::GetTypeForDecl(ObjCInterfaceDecl *decl) {
   return GetType(getASTContext().getObjCInterfaceType(decl));
 }
 
+CompilerType TypeSystemClang::GetTypeForDecl(clang::ValueDecl *value_decl) {
+  return GetType(value_decl->getType());
+}
+
 #pragma mark Structure, Unions, Classes
 
 void TypeSystemClang::SetOwningModule(clang::Decl *decl,
@@ -1227,7 +1237,7 @@ CompilerType TypeSystemClang::CreateRecordType(
   // complete definition just in case.
 
   bool has_name = !name.empty();
-  CXXRecordDecl *decl = CXXRecordDecl::CreateDeserialized(ast, 0);
+  CXXRecordDecl *decl = CXXRecordDecl::CreateDeserialized(ast, GlobalDeclID());
   decl->setTagKind(static_cast<TagDecl::TagKind>(kind));
   decl->setDeclContext(decl_ctx);
   if (has_name)
@@ -1399,7 +1409,7 @@ clang::FunctionTemplateDecl *TypeSystemClang::CreateFunctionTemplateDecl(
   TemplateParameterList *template_param_list = CreateTemplateParameterList(
       ast, template_param_infos, template_param_decls);
   FunctionTemplateDecl *func_tmpl_decl =
-      FunctionTemplateDecl::CreateDeserialized(ast, 0);
+      FunctionTemplateDecl::CreateDeserialized(ast, GlobalDeclID());
   func_tmpl_decl->setDeclContext(decl_ctx);
   func_tmpl_decl->setLocation(func_decl->getLocation());
   func_tmpl_decl->setDeclName(func_decl->getDeclName());
@@ -1561,7 +1571,8 @@ ClassTemplateDecl *TypeSystemClang::CreateClassTemplateDecl(
   TemplateParameterList *template_param_list = CreateTemplateParameterList(
       ast, template_param_infos, template_param_decls);
 
-  CXXRecordDecl *template_cxx_decl = CXXRecordDecl::CreateDeserialized(ast, 0);
+  CXXRecordDecl *template_cxx_decl =
+      CXXRecordDecl::CreateDeserialized(ast, GlobalDeclID());
   template_cxx_decl->setTagKind(static_cast<TagDecl::TagKind>(kind));
   // What decl context do we use here? TU? The actual decl context?
   template_cxx_decl->setDeclContext(decl_ctx);
@@ -1578,7 +1589,8 @@ ClassTemplateDecl *TypeSystemClang::CreateClassTemplateDecl(
   // template_cxx_decl->startDefinition();
   // template_cxx_decl->completeDefinition();
 
-  class_template_decl = ClassTemplateDecl::CreateDeserialized(ast, 0);
+  class_template_decl =
+      ClassTemplateDecl::CreateDeserialized(ast, GlobalDeclID());
   // What decl context do we use here? TU? The actual decl context?
   class_template_decl->setDeclContext(decl_ctx);
   class_template_decl->setDeclName(decl_name);
@@ -1639,7 +1651,7 @@ TypeSystemClang::CreateClassTemplateSpecializationDecl(
         ast, template_param_infos.GetParameterPackArgs());
   }
   ClassTemplateSpecializationDecl *class_template_specialization_decl =
-      ClassTemplateSpecializationDecl::CreateDeserialized(ast, 0);
+      ClassTemplateSpecializationDecl::CreateDeserialized(ast, GlobalDeclID());
   class_template_specialization_decl->setTagKind(
       static_cast<TagDecl::TagKind>(kind));
   class_template_specialization_decl->setDeclContext(decl_ctx);
@@ -1789,7 +1801,8 @@ CompilerType TypeSystemClang::CreateObjCClass(
   if (!decl_ctx)
     decl_ctx = ast.getTranslationUnitDecl();
 
-  ObjCInterfaceDecl *decl = ObjCInterfaceDecl::CreateDeserialized(ast, 0);
+  ObjCInterfaceDecl *decl =
+      ObjCInterfaceDecl::CreateDeserialized(ast, GlobalDeclID());
   decl->setDeclContext(decl_ctx);
   decl->setDeclName(&ast.Idents.get(name));
   /*isForwardDecl,*/
@@ -1897,7 +1910,7 @@ TypeSystemClang::CreateBlockDeclaration(clang::DeclContext *ctx,
                                         OptionalClangModuleID owning_module) {
   if (ctx) {
     clang::BlockDecl *decl =
-        clang::BlockDecl::CreateDeserialized(getASTContext(), 0);
+        clang::BlockDecl::CreateDeserialized(getASTContext(), GlobalDeclID());
     decl->setDeclContext(ctx);
     ctx->addDecl(decl);
     SetOwningModule(decl, owning_module);
@@ -1966,7 +1979,7 @@ clang::VarDecl *TypeSystemClang::CreateVariableDeclaration(
     const char *name, clang::QualType type) {
   if (decl_context) {
     clang::VarDecl *var_decl =
-        clang::VarDecl::CreateDeserialized(getASTContext(), 0);
+        clang::VarDecl::CreateDeserialized(getASTContext(), GlobalDeclID());
     var_decl->setDeclContext(decl_context);
     if (name && name[0])
       var_decl->setDeclName(&getASTContext().Idents.getOwn(name));
@@ -2126,7 +2139,7 @@ FunctionDecl *TypeSystemClang::CreateFunctionDeclaration(
 
   clang::DeclarationName declarationName =
       GetDeclarationName(name, function_clang_type);
-  func_decl = FunctionDecl::CreateDeserialized(ast, 0);
+  func_decl = FunctionDecl::CreateDeserialized(ast, GlobalDeclID());
   func_decl->setDeclContext(decl_ctx);
   func_decl->setDeclName(declarationName);
   func_decl->setType(ClangUtil::GetQualType(function_clang_type));
@@ -2187,7 +2200,7 @@ ParmVarDecl *TypeSystemClang::CreateParameterDeclaration(
     const char *name, const CompilerType &param_type, int storage,
     bool add_decl) {
   ASTContext &ast = getASTContext();
-  auto *decl = ParmVarDecl::CreateDeserialized(ast, 0);
+  auto *decl = ParmVarDecl::CreateDeserialized(ast, GlobalDeclID());
   decl->setDeclContext(decl_ctx);
   if (name && name[0])
     decl->setDeclName(&ast.Idents.get(name));
@@ -2292,7 +2305,7 @@ CompilerType TypeSystemClang::CreateEnumerationType(
 
   // TODO: ask about these...
   //    const bool IsFixed = false;
-  EnumDecl *enum_decl = EnumDecl::CreateDeserialized(ast, 0);
+  EnumDecl *enum_decl = EnumDecl::CreateDeserialized(ast, GlobalDeclID());
   enum_decl->setDeclContext(decl_ctx);
   if (!name.empty())
     enum_decl->setDeclName(&ast.Idents.get(name));
@@ -4531,7 +4544,7 @@ CompilerType TypeSystemClang::CreateTypedef(
       decl_ctx = getASTContext().getTranslationUnitDecl();
 
     clang::TypedefDecl *decl =
-        clang::TypedefDecl::CreateDeserialized(clang_ast, 0);
+        clang::TypedefDecl::CreateDeserialized(clang_ast, GlobalDeclID());
     decl->setDeclContext(decl_ctx);
     decl->setDeclName(&clang_ast.Idents.get(typedef_name));
     decl->setTypeSourceInfo(clang_ast.getTrivialTypeSourceInfo(qual_type));
@@ -4848,7 +4861,7 @@ lldb::Encoding TypeSystemClang::GetEncoding(lldb::opaque_compiler_type_t type,
     case clang::BuiltinType::Kind::OCLQueue:
     case clang::BuiltinType::Kind::OCLReserveID:
     case clang::BuiltinType::Kind::OCLSampler:
-    case clang::BuiltinType::Kind::OMPArraySection:
+    case clang::BuiltinType::Kind::ArraySection:
     case clang::BuiltinType::Kind::OMPArrayShaping:
     case clang::BuiltinType::Kind::OMPIterator:
     case clang::BuiltinType::Kind::Overload:
@@ -5900,6 +5913,36 @@ CompilerType TypeSystemClang::GetVirtualBaseClassAtIndex(
   return CompilerType();
 }
 
+CompilerDecl
+TypeSystemClang::GetStaticFieldWithName(lldb::opaque_compiler_type_t type,
+                                        llvm::StringRef name) {
+  clang::QualType qual_type = RemoveWrappingTypes(GetCanonicalQualType(type));
+  switch (qual_type->getTypeClass()) {
+  case clang::Type::Record: {
+    if (!GetCompleteType(type))
+      return CompilerDecl();
+
+    const clang::RecordType *record_type =
+        llvm::cast<clang::RecordType>(qual_type.getTypePtr());
+    const clang::RecordDecl *record_decl = record_type->getDecl();
+
+    clang::DeclarationName decl_name(&getASTContext().Idents.get(name));
+    for (NamedDecl *decl : record_decl->lookup(decl_name)) {
+      auto *var_decl = dyn_cast<clang::VarDecl>(decl);
+      if (!var_decl || var_decl->getStorageClass() != clang::SC_Static)
+        continue;
+
+      return CompilerDecl(this, var_decl);
+    }
+    break;
+  }
+
+  default:
+    break;
+  }
+  return CompilerDecl();
+}
+
 // If a pointer to a pointee type (the clang_type arg) says that it has no
 // children, then we either need to trust it, or override it and return a
 // different result. For example, an "int *" has one child that is an integer,
@@ -5970,7 +6013,7 @@ uint32_t TypeSystemClang::GetNumPointeeChildren(clang::QualType type) {
     case clang::BuiltinType::ARCUnbridgedCast:
     case clang::BuiltinType::PseudoObject:
     case clang::BuiltinType::BuiltinFn:
-    case clang::BuiltinType::OMPArraySection:
+    case clang::BuiltinType::ArraySection:
       return 1;
     default:
       return 0;
@@ -7288,7 +7331,7 @@ clang::FieldDecl *TypeSystemClang::AddFieldToRecordType(
 
   clang::RecordDecl *record_decl = ast->GetAsRecordDecl(type);
   if (record_decl) {
-    field = clang::FieldDecl::CreateDeserialized(clang_ast, 0);
+    field = clang::FieldDecl::CreateDeserialized(clang_ast, GlobalDeclID());
     field->setDeclContext(record_decl);
     field->setDeclName(ident);
     field->setType(ClangUtil::GetQualType(field_clang_type));
@@ -7335,7 +7378,8 @@ clang::FieldDecl *TypeSystemClang::AddFieldToRecordType(
 
       field_clang_type.GetCompleteType();
 
-      auto *ivar = clang::ObjCIvarDecl::CreateDeserialized(clang_ast, 0);
+      auto *ivar =
+          clang::ObjCIvarDecl::CreateDeserialized(clang_ast, GlobalDeclID());
       ivar->setDeclContext(class_interface_decl);
       ivar->setDeclName(ident);
       ivar->setType(ClangUtil::GetQualType(field_clang_type));
@@ -7501,7 +7545,8 @@ clang::VarDecl *TypeSystemClang::AddVariableToRecordType(
   if (!name.empty())
     ident = &ast->getASTContext().Idents.get(name);
 
-  var_decl = clang::VarDecl::CreateDeserialized(ast->getASTContext(), 0);
+  var_decl =
+      clang::VarDecl::CreateDeserialized(ast->getASTContext(), GlobalDeclID());
   var_decl->setDeclContext(record_decl);
   var_decl->setDeclName(ident);
   var_decl->setType(ClangUtil::GetQualType(var_type));
@@ -7602,8 +7647,8 @@ clang::CXXMethodDecl *TypeSystemClang::AddMethodToCXXRecordType(
                                     : clang::ExplicitSpecKind::ResolvedFalse);
 
   if (name.starts_with("~")) {
-    cxx_dtor_decl =
-        clang::CXXDestructorDecl::CreateDeserialized(getASTContext(), 0);
+    cxx_dtor_decl = clang::CXXDestructorDecl::CreateDeserialized(
+        getASTContext(), GlobalDeclID());
     cxx_dtor_decl->setDeclContext(cxx_record_decl);
     cxx_dtor_decl->setDeclName(
         getASTContext().DeclarationNames.getCXXDestructorName(
@@ -7615,7 +7660,7 @@ clang::CXXMethodDecl *TypeSystemClang::AddMethodToCXXRecordType(
     cxx_method_decl = cxx_dtor_decl;
   } else if (decl_name == cxx_record_decl->getDeclName()) {
     cxx_ctor_decl = clang::CXXConstructorDecl::CreateDeserialized(
-        getASTContext(), 0, 0);
+        getASTContext(), GlobalDeclID(), 0);
     cxx_ctor_decl->setDeclContext(cxx_record_decl);
     cxx_ctor_decl->setDeclName(
         getASTContext().DeclarationNames.getCXXConstructorName(
@@ -7641,8 +7686,8 @@ clang::CXXMethodDecl *TypeSystemClang::AddMethodToCXXRecordType(
         if (!TypeSystemClang::CheckOverloadedOperatorKindParameterCount(
                 is_method, op_kind, num_params))
           return nullptr;
-        cxx_method_decl =
-            clang::CXXMethodDecl::CreateDeserialized(getASTContext(), 0);
+        cxx_method_decl = clang::CXXMethodDecl::CreateDeserialized(
+            getASTContext(), GlobalDeclID());
         cxx_method_decl->setDeclContext(cxx_record_decl);
         cxx_method_decl->setDeclName(
             getASTContext().DeclarationNames.getCXXOperatorName(op_kind));
@@ -7653,7 +7698,8 @@ clang::CXXMethodDecl *TypeSystemClang::AddMethodToCXXRecordType(
       } else if (num_params == 0) {
         // Conversion operators don't take params...
         auto *cxx_conversion_decl =
-            clang::CXXConversionDecl::CreateDeserialized(getASTContext(), 0);
+            clang::CXXConversionDecl::CreateDeserialized(getASTContext(),
+                                                         GlobalDeclID());
         cxx_conversion_decl->setDeclContext(cxx_record_decl);
         cxx_conversion_decl->setDeclName(
             getASTContext().DeclarationNames.getCXXConversionFunctionName(
@@ -7668,8 +7714,8 @@ clang::CXXMethodDecl *TypeSystemClang::AddMethodToCXXRecordType(
     }
 
     if (cxx_method_decl == nullptr) {
-      cxx_method_decl =
-          clang::CXXMethodDecl::CreateDeserialized(getASTContext(), 0);
+      cxx_method_decl = clang::CXXMethodDecl::CreateDeserialized(
+          getASTContext(), GlobalDeclID());
       cxx_method_decl->setDeclContext(cxx_record_decl);
       cxx_method_decl->setDeclName(decl_name);
       cxx_method_decl->setType(method_qual_type);
@@ -7852,7 +7898,7 @@ bool TypeSystemClang::AddObjCClassProperty(
         ClangUtil::GetQualType(property_clang_type));
 
   clang::ObjCPropertyDecl *property_decl =
-      clang::ObjCPropertyDecl::CreateDeserialized(clang_ast, 0);
+      clang::ObjCPropertyDecl::CreateDeserialized(clang_ast, GlobalDeclID());
   property_decl->setDeclContext(class_interface_decl);
   property_decl->setDeclName(&clang_ast.Idents.get(property_name));
   property_decl->setType(ivar_decl
@@ -7941,7 +7987,8 @@ bool TypeSystemClang::AddObjCClassProperty(
         clang::ObjCImplementationControl::None;
     const bool HasRelatedResultType = false;
 
-    getter = clang::ObjCMethodDecl::CreateDeserialized(clang_ast, 0);
+    getter =
+        clang::ObjCMethodDecl::CreateDeserialized(clang_ast, GlobalDeclID());
     getter->setDeclName(getter_sel);
     getter->setReturnType(ClangUtil::GetQualType(property_clang_type_to_access));
     getter->setDeclContext(class_interface_decl);
@@ -7983,7 +8030,8 @@ bool TypeSystemClang::AddObjCClassProperty(
         clang::ObjCImplementationControl::None;
     const bool HasRelatedResultType = false;
 
-    setter = clang::ObjCMethodDecl::CreateDeserialized(clang_ast, 0);
+    setter =
+        clang::ObjCMethodDecl::CreateDeserialized(clang_ast, GlobalDeclID());
     setter->setDeclName(setter_sel);
     setter->setReturnType(result_type);
     setter->setDeclContext(class_interface_decl);
@@ -8111,7 +8159,8 @@ clang::ObjCMethodDecl *TypeSystemClang::AddMethodToObjCObjectType(
     return nullptr; // some debug information is corrupt.  We are not going to
                     // deal with it.
 
-  auto *objc_method_decl = clang::ObjCMethodDecl::CreateDeserialized(ast, 0);
+  auto *objc_method_decl =
+      clang::ObjCMethodDecl::CreateDeserialized(ast, GlobalDeclID());
   objc_method_decl->setDeclName(method_selector);
   objc_method_decl->setReturnType(method_function_prototype->getReturnType());
   objc_method_decl->setDeclContext(
@@ -8357,7 +8406,8 @@ clang::EnumConstantDecl *TypeSystemClang::AddEnumerationValueToEnumerationType(
     return nullptr;
 
   clang::EnumConstantDecl *enumerator_decl =
-      clang::EnumConstantDecl::CreateDeserialized(getASTContext(), 0);
+      clang::EnumConstantDecl::CreateDeserialized(getASTContext(),
+                                                  GlobalDeclID());
   enumerator_decl->setDeclContext(enutype->getDecl());
   if (name && name[0])
     enumerator_decl->setDeclName(&getASTContext().Idents.get(name));
@@ -9074,6 +9124,21 @@ CompilerType TypeSystemClang::DeclGetFunctionArgumentType(void *opaque_decl,
   return CompilerType();
 }
 
+Scalar TypeSystemClang::DeclGetConstantValue(void *opaque_decl) {
+  clang::Decl *decl = static_cast<clang::Decl *>(opaque_decl);
+  clang::VarDecl *var_decl = llvm::dyn_cast<clang::VarDecl>(decl);
+  if (!var_decl)
+    return Scalar();
+  clang::Expr *init_expr = var_decl->getInit();
+  if (!init_expr)
+    return Scalar();
+  std::optional<llvm::APSInt> value =
+      init_expr->getIntegerConstantExpr(getASTContext());
+  if (!value)
+    return Scalar();
+  return Scalar(*value);
+}
+
 // CompilerDeclContext functions
 
 std::vector<CompilerDecl> TypeSystemClang::DeclContextFindDeclByName(
diff --git a/lldb/source/Plugins/TypeSystem/Clang/TypeSystemClang.h b/lldb/source/Plugins/TypeSystem/Clang/TypeSystemClang.h
index 68b82e9688f1..62f14df7638d 100644
--- a/lldb/source/Plugins/TypeSystem/Clang/TypeSystemClang.h
+++ b/lldb/source/Plugins/TypeSystem/Clang/TypeSystemClang.h
@@ -22,6 +22,7 @@
 
 #include "clang/AST/ASTContext.h"
 #include "clang/AST/ASTFwd.h"
+#include "clang/AST/Decl.h"
 #include "clang/AST/TemplateBase.h"
 #include "clang/AST/Type.h"
 #include "clang/Basic/TargetInfo.h"
@@ -251,6 +252,8 @@ public:
 
   CompilerType GetTypeForDecl(clang::ObjCInterfaceDecl *objc_decl);
 
+  CompilerType GetTypeForDecl(clang::ValueDecl *value_decl);
+
   template <typename RecordDeclType>
   CompilerType
   GetTypeForIdentifier(llvm::StringRef type_name,
@@ -559,6 +562,8 @@ public:
   std::vector<lldb_private::CompilerContext>
   DeclGetCompilerContext(void *opaque_decl) override;
 
+  Scalar DeclGetConstantValue(void *opaque_decl) override;
+
   CompilerType GetTypeForDecl(void *opaque_decl) override;
 
   // CompilerDeclContext override functions
@@ -868,6 +873,9 @@ public:
                                           size_t idx,
                                           uint32_t *bit_offset_ptr) override;
 
+  CompilerDecl GetStaticFieldWithName(lldb::opaque_compiler_type_t type,
+                                      llvm::StringRef name) override;
+
   static uint32_t GetNumPointeeChildren(clang::QualType type);
 
   CompilerType GetChildCompilerTypeAtIndex(
diff --git a/lldb/source/Symbol/CompilerDecl.cpp b/lldb/source/Symbol/CompilerDecl.cpp
index 0eb630e5b9e1..5fa0a32f041a 100644
--- a/lldb/source/Symbol/CompilerDecl.cpp
+++ b/lldb/source/Symbol/CompilerDecl.cpp
@@ -9,6 +9,7 @@
 #include "lldb/Symbol/CompilerDecl.h"
 #include "lldb/Symbol/CompilerDeclContext.h"
 #include "lldb/Symbol/TypeSystem.h"
+#include "lldb/Utility/Scalar.h"
 
 using namespace lldb_private;
 
@@ -24,6 +25,10 @@ CompilerDeclContext CompilerDecl::GetDeclContext() const {
   return m_type_system->DeclGetDeclContext(m_opaque_decl);
 }
 
+CompilerType CompilerDecl::GetType() const {
+  return m_type_system->GetTypeForDecl(m_opaque_decl);
+}
+
 CompilerType CompilerDecl::GetFunctionReturnType() const {
   return m_type_system->DeclGetFunctionReturnType(m_opaque_decl);
 }
@@ -52,3 +57,7 @@ std::vector<lldb_private::CompilerContext>
 CompilerDecl::GetCompilerContext() const {
   return m_type_system->DeclGetCompilerContext(m_opaque_decl);
 }
+
+Scalar CompilerDecl::GetConstantValue() const {
+  return m_type_system->DeclGetConstantValue(m_opaque_decl);
+}
diff --git a/lldb/source/Symbol/CompilerType.cpp b/lldb/source/Symbol/CompilerType.cpp
index 96e74b890d2d..9523fb6ea77c 100644
--- a/lldb/source/Symbol/CompilerType.cpp
+++ b/lldb/source/Symbol/CompilerType.cpp
@@ -849,6 +849,12 @@ CompilerType::GetVirtualBaseClassAtIndex(size_t idx,
   return CompilerType();
 }
 
+CompilerDecl CompilerType::GetStaticFieldWithName(llvm::StringRef name) const {
+  if (IsValid())
+    return GetTypeSystem()->GetStaticFieldWithName(m_type, name);
+  return CompilerDecl();
+}
+
 uint32_t CompilerType::GetIndexOfFieldWithName(
     const char *name, CompilerType *field_compiler_type_ptr,
     uint64_t *bit_offset_ptr, uint32_t *bitfield_bit_size_ptr,
diff --git a/lldb/source/Target/Process.cpp b/lldb/source/Target/Process.cpp
index 606518ca5412..30c240b064b5 100644
--- a/lldb/source/Target/Process.cpp
+++ b/lldb/source/Target/Process.cpp
@@ -408,8 +408,8 @@ ProcessSP Process::FindPlugin(lldb::TargetSP target_sp,
   return process_sp;
 }
 
-ConstString &Process::GetStaticBroadcasterClass() {
-  static ConstString class_name("lldb.process");
+llvm::StringRef Process::GetStaticBroadcasterClass() {
+  static constexpr llvm::StringLiteral class_name("lldb.process");
   return class_name;
 }
 
@@ -423,7 +423,7 @@ Process::Process(lldb::TargetSP target_sp, ListenerSP listener_sp,
                  const UnixSignalsSP &unix_signals_sp)
     : ProcessProperties(this),
       Broadcaster((target_sp->GetDebugger().GetBroadcasterManager()),
-                  Process::GetStaticBroadcasterClass().AsCString()),
+                  Process::GetStaticBroadcasterClass().str()),
       m_target_wp(target_sp), m_public_state(eStateUnloaded),
       m_private_state(eStateUnloaded),
       m_private_state_broadcaster(nullptr,
diff --git a/lldb/source/Target/Target.cpp b/lldb/source/Target/Target.cpp
index 09b0ac42631d..aa4895bb5a6d 100644
--- a/lldb/source/Target/Target.cpp
+++ b/lldb/source/Target/Target.cpp
@@ -87,8 +87,8 @@ const Target::Arch &Target::Arch::operator=(const ArchSpec &spec) {
   return *this;
 }
 
-ConstString &Target::GetStaticBroadcasterClass() {
-  static ConstString class_name("lldb.target");
+llvm::StringRef Target::GetStaticBroadcasterClass() {
+  static constexpr llvm::StringLiteral class_name("lldb.target");
   return class_name;
 }
 
@@ -96,7 +96,7 @@ Target::Target(Debugger &debugger, const ArchSpec &target_arch,
                const lldb::PlatformSP &platform_sp, bool is_dummy_target)
     : TargetProperties(this),
       Broadcaster(debugger.GetBroadcasterManager(),
-                  Target::GetStaticBroadcasterClass().AsCString()),
+                  Target::GetStaticBroadcasterClass().str()),
       ExecutionContextScope(), m_debugger(debugger), m_platform_sp(platform_sp),
       m_mutex(), m_arch(target_arch), m_images(this), m_section_load_history(),
       m_breakpoint_list(false), m_internal_breakpoint_list(true),
diff --git a/lldb/source/Target/TargetList.cpp b/lldb/source/Target/TargetList.cpp
index b5d308739d0f..10467753666f 100644
--- a/lldb/source/Target/TargetList.cpp
+++ b/lldb/source/Target/TargetList.cpp
@@ -29,15 +29,15 @@
 using namespace lldb;
 using namespace lldb_private;
 
-ConstString &TargetList::GetStaticBroadcasterClass() {
-  static ConstString class_name("lldb.targetList");
+llvm::StringRef TargetList::GetStaticBroadcasterClass() {
+  static constexpr llvm::StringLiteral class_name("lldb.targetList");
   return class_name;
 }
 
 // TargetList constructor
 TargetList::TargetList(Debugger &debugger)
     : Broadcaster(debugger.GetBroadcasterManager(),
-                  TargetList::GetStaticBroadcasterClass().AsCString()),
+                  TargetList::GetStaticBroadcasterClass().str()),
       m_target_list(), m_target_list_mutex(), m_selected_target_idx(0) {
   CheckInWithManager();
 }
diff --git a/lldb/source/Target/Thread.cpp b/lldb/source/Target/Thread.cpp
index 412e44ede9c1..e75f5a356cec 100644
--- a/lldb/source/Target/Thread.cpp
+++ b/lldb/source/Target/Thread.cpp
@@ -205,15 +205,15 @@ Thread::ThreadEventData::GetStackFrameFromEvent(const Event *event_ptr) {
 
 // Thread class
 
-ConstString &Thread::GetStaticBroadcasterClass() {
-  static ConstString class_name("lldb.thread");
+llvm::StringRef Thread::GetStaticBroadcasterClass() {
+  static constexpr llvm::StringLiteral class_name("lldb.thread");
   return class_name;
 }
 
 Thread::Thread(Process &process, lldb::tid_t tid, bool use_invalid_index_id)
     : ThreadProperties(false), UserID(tid),
       Broadcaster(process.GetTarget().GetDebugger().GetBroadcasterManager(),
-                  Thread::GetStaticBroadcasterClass().AsCString()),
+                  Thread::GetStaticBroadcasterClass().str()),
       m_process_wp(process.shared_from_this()), m_stop_info_sp(),
       m_stop_info_stop_id(0), m_stop_info_override_stop_id(0),
       m_should_run_before_public_stop(false),
diff --git a/lldb/source/Utility/Broadcaster.cpp b/lldb/source/Utility/Broadcaster.cpp
index 12903edc36b1..bd65ffd86a1d 100644
--- a/lldb/source/Utility/Broadcaster.cpp
+++ b/lldb/source/Utility/Broadcaster.cpp
@@ -373,8 +373,8 @@ void Broadcaster::BroadcasterImpl::RestoreBroadcaster() {
     m_hijacking_masks.pop_back();
 }
 
-ConstString &Broadcaster::GetBroadcasterClass() const {
-  static ConstString class_name("lldb.anonymous");
+llvm::StringRef Broadcaster::GetBroadcasterClass() const {
+  static constexpr llvm::StringLiteral class_name("lldb.anonymous");
   return class_name;
 }
 
diff --git a/lldb/test/API/clear-sbvalue-nonaddressable-bits/Makefile b/lldb/test/API/clear-sbvalue-nonaddressable-bits/Makefile
new file mode 100644
index 000000000000..10495940055b
--- /dev/null
+++ b/lldb/test/API/clear-sbvalue-nonaddressable-bits/Makefile
@@ -0,0 +1,3 @@
+C_SOURCES := main.c
+
+include Makefile.rules
diff --git a/lldb/test/API/clear-sbvalue-nonaddressable-bits/TestClearSBValueNonAddressableBits.py b/lldb/test/API/clear-sbvalue-nonaddressable-bits/TestClearSBValueNonAddressableBits.py
new file mode 100644
index 000000000000..382b0e7a81d2
--- /dev/null
+++ b/lldb/test/API/clear-sbvalue-nonaddressable-bits/TestClearSBValueNonAddressableBits.py
@@ -0,0 +1,59 @@
+"""Test that SBValue clears non-addressable bits"""
+
+import lldb
+from lldbsuite.test.decorators import *
+from lldbsuite.test.lldbtest import *
+from lldbsuite.test import lldbutil
+
+
+class TestClearSBValueNonAddressableBits(TestBase):
+    NO_DEBUG_INFO_TESTCASE = True
+
+    # On AArch64 systems, the top bits that are not used for
+    # addressing may be used for TBI, MTE, and/or pointer
+    # authentication.
+    @skipIf(archs=no_match(["aarch64", "arm64", "arm64e"]))
+
+    # Only run this test on systems where TBI is known to be
+    # enabled, so the address mask will clear the TBI bits.
+    @skipUnlessPlatform(["linux"] + lldbplatformutil.getDarwinOSTriples())
+    def test(self):
+        self.source = "main.c"
+        self.build()
+        (target, process, thread, bkpt) = lldbutil.run_to_source_breakpoint(
+            self, "break here", lldb.SBFileSpec(self.source, False)
+        )
+
+        if self.TraceOn():
+            self.runCmd("frame variable")
+            self.runCmd("frame variable &count &global")
+
+        frame = thread.GetFrameAtIndex(0)
+
+        count_p = frame.FindVariable("count_p")
+        count_invalid_p = frame.FindVariable("count_invalid_p")
+        self.assertEqual(
+            count_p.GetValueAsUnsigned(), count_invalid_p.GetValueAsAddress()
+        )
+        self.assertNotEqual(
+            count_invalid_p.GetValueAsUnsigned(), count_invalid_p.GetValueAsAddress()
+        )
+        self.assertEqual(5, count_p.Dereference().GetValueAsUnsigned())
+        self.assertEqual(5, count_invalid_p.Dereference().GetValueAsUnsigned())
+
+        global_p = frame.FindVariable("global_p")
+        global_invalid_p = frame.FindVariable("global_invalid_p")
+        self.assertEqual(
+            global_p.GetValueAsUnsigned(), global_invalid_p.GetValueAsAddress()
+        )
+        self.assertNotEqual(
+            global_invalid_p.GetValueAsUnsigned(), global_invalid_p.GetValueAsAddress()
+        )
+        self.assertEqual(10, global_p.Dereference().GetValueAsUnsigned())
+        self.assertEqual(10, global_invalid_p.Dereference().GetValueAsUnsigned())
+
+        main_p = frame.FindVariable("main_p")
+        main_invalid_p = frame.FindVariable("main_invalid_p")
+        self.assertEqual(
+            main_p.GetValueAsUnsigned(), main_invalid_p.GetValueAsAddress()
+        )
diff --git a/lldb/test/API/clear-sbvalue-nonaddressable-bits/main.c b/lldb/test/API/clear-sbvalue-nonaddressable-bits/main.c
new file mode 100644
index 000000000000..1b0e42c50dd6
--- /dev/null
+++ b/lldb/test/API/clear-sbvalue-nonaddressable-bits/main.c
@@ -0,0 +1,27 @@
+#include <stdint.h>
+
+int global = 10;
+
+int main() {
+  int count = 5;
+  int *count_p = &count;
+
+  // Add some metadata in the top byte (this will crash unless the
+  // test is running with TBI enabled, but we won't dereference it)
+
+  intptr_t scratch = (intptr_t)count_p;
+  scratch |= (3ULL << 60);
+  int *count_invalid_p = (int *)scratch;
+
+  int (*main_p)() = main;
+  scratch = (intptr_t)main_p;
+  scratch |= (3ULL << 60);
+  int (*main_invalid_p)() = (int (*)())scratch;
+
+  int *global_p = &global;
+  scratch = (intptr_t)global_p;
+  scratch |= (3ULL << 60);
+  int *global_invalid_p = (int *)scratch;
+
+  return count; // break here
+}
diff --git a/lldb/test/API/functionalities/diagnostic_reporting/TestDiagnosticReporting.py b/lldb/test/API/functionalities/diagnostic_reporting/TestDiagnosticReporting.py
index 6353e3e8cbed..36a3be695628 100644
--- a/lldb/test/API/functionalities/diagnostic_reporting/TestDiagnosticReporting.py
+++ b/lldb/test/API/functionalities/diagnostic_reporting/TestDiagnosticReporting.py
@@ -15,7 +15,7 @@ class TestDiagnosticReporting(TestBase):
         self.broadcaster = self.dbg.GetBroadcaster()
         self.listener = lldbutil.start_listening_from(
             self.broadcaster,
-            lldb.eBroadcastBitWarning | lldb.eBroadcastBitError,
+            lldb.SBDebugger.eBroadcastBitWarning | lldb.SBDebugger.eBroadcastBitError,
         )
 
     def test_dwarf_symbol_loading_diagnostic_report(self):
diff --git a/lldb/test/API/functionalities/fork/concurrent_vfork/TestConcurrentVFork.py b/lldb/test/API/functionalities/fork/concurrent_vfork/TestConcurrentVFork.py
index 2dcbb728549f..dd9500c186b2 100644
--- a/lldb/test/API/functionalities/fork/concurrent_vfork/TestConcurrentVFork.py
+++ b/lldb/test/API/functionalities/fork/concurrent_vfork/TestConcurrentVFork.py
@@ -48,6 +48,8 @@ class TestConcurrentVFork(TestBase):
         self.expect("continue", patterns=[r"exited with status = 1[0-4]"])
 
     @skipUnlessPlatform(["linux"])
+    # https://github.com/llvm/llvm-project/issues/85084.
+    @skipIf(oslist=["linux"], archs=["aarch64", "arm"])
     def test_follow_parent_vfork_no_exec(self):
         """
         Make sure that debugging concurrent vfork() from multiple threads won't crash lldb during follow-parent.
@@ -56,6 +58,8 @@ class TestConcurrentVFork(TestBase):
         self.follow_parent_helper(use_fork=False, call_exec=False)
 
     @skipUnlessPlatform(["linux"])
+    # https://github.com/llvm/llvm-project/issues/85084.
+    @skipIf(oslist=["linux"], archs=["aarch64", "arm"])
     def test_follow_parent_fork_no_exec(self):
         """
         Make sure that debugging concurrent fork() from multiple threads won't crash lldb during follow-parent.
@@ -64,6 +68,8 @@ class TestConcurrentVFork(TestBase):
         self.follow_parent_helper(use_fork=True, call_exec=False)
 
     @skipUnlessPlatform(["linux"])
+    # https://github.com/llvm/llvm-project/issues/85084.
+    @skipIf(oslist=["linux"], archs=["aarch64", "arm"])
     def test_follow_parent_vfork_call_exec(self):
         """
         Make sure that debugging concurrent vfork() from multiple threads won't crash lldb during follow-parent.
@@ -72,6 +78,8 @@ class TestConcurrentVFork(TestBase):
         self.follow_parent_helper(use_fork=False, call_exec=True)
 
     @skipUnlessPlatform(["linux"])
+    # https://github.com/llvm/llvm-project/issues/85084.
+    @skipIf(oslist=["linux"], archs=["aarch64", "arm"])
     def test_follow_parent_fork_call_exec(self):
         """
         Make sure that debugging concurrent vfork() from multiple threads won't crash lldb during follow-parent.
@@ -80,6 +88,8 @@ class TestConcurrentVFork(TestBase):
         self.follow_parent_helper(use_fork=True, call_exec=True)
 
     @skipUnlessPlatform(["linux"])
+    # https://github.com/llvm/llvm-project/issues/85084.
+    @skipIf(oslist=["linux"], archs=["aarch64", "arm"])
     def test_follow_child_vfork_no_exec(self):
         """
         Make sure that debugging concurrent vfork() from multiple threads won't crash lldb during follow-child.
@@ -88,6 +98,8 @@ class TestConcurrentVFork(TestBase):
         self.follow_child_helper(use_fork=False, call_exec=False)
 
     @skipUnlessPlatform(["linux"])
+    # https://github.com/llvm/llvm-project/issues/85084.
+    @skipIf(oslist=["linux"], archs=["aarch64", "arm"])
     def test_follow_child_fork_no_exec(self):
         """
         Make sure that debugging concurrent fork() from multiple threads won't crash lldb during follow-child.
@@ -96,6 +108,8 @@ class TestConcurrentVFork(TestBase):
         self.follow_child_helper(use_fork=True, call_exec=False)
 
     @skipUnlessPlatform(["linux"])
+    # https://github.com/llvm/llvm-project/issues/85084.
+    @skipIf(oslist=["linux"], archs=["aarch64", "arm"])
     def test_follow_child_vfork_call_exec(self):
         """
         Make sure that debugging concurrent vfork() from multiple threads won't crash lldb during follow-child.
@@ -104,6 +118,8 @@ class TestConcurrentVFork(TestBase):
         self.follow_child_helper(use_fork=False, call_exec=True)
 
     @skipUnlessPlatform(["linux"])
+    # https://github.com/llvm/llvm-project/issues/85084.
+    @skipIf(oslist=["linux"], archs=["aarch64", "arm"])
     def test_follow_child_fork_call_exec(self):
         """
         Make sure that debugging concurrent fork() from multiple threads won't crash lldb during follow-child.
diff --git a/lldb/test/API/functionalities/progress_reporting/TestProgressReporting.py b/lldb/test/API/functionalities/progress_reporting/TestProgressReporting.py
index 98988d7624da..9af53845ca1b 100644
--- a/lldb/test/API/functionalities/progress_reporting/TestProgressReporting.py
+++ b/lldb/test/API/functionalities/progress_reporting/TestProgressReporting.py
@@ -13,7 +13,7 @@ class TestProgressReporting(TestBase):
         TestBase.setUp(self)
         self.broadcaster = self.dbg.GetBroadcaster()
         self.listener = lldbutil.start_listening_from(
-            self.broadcaster, lldb.eBroadcastBitProgress
+            self.broadcaster, lldb.SBDebugger.eBroadcastBitProgress
         )
 
     def test_dwarf_symbol_loading_progress_report(self):
diff --git a/lldb/test/API/functionalities/progress_reporting/clang_modules/TestClangModuleBuildProgress.py b/lldb/test/API/functionalities/progress_reporting/clang_modules/TestClangModuleBuildProgress.py
index 33c7c269c081..228f676aedf6 100644
--- a/lldb/test/API/functionalities/progress_reporting/clang_modules/TestClangModuleBuildProgress.py
+++ b/lldb/test/API/functionalities/progress_reporting/clang_modules/TestClangModuleBuildProgress.py
@@ -34,7 +34,7 @@ class TestCase(TestBase):
         # other unrelated progress events.
         broadcaster = self.dbg.GetBroadcaster()
         listener = lldbutil.start_listening_from(
-            broadcaster, lldb.eBroadcastBitProgress
+            broadcaster, lldb.SBDebugger.eBroadcastBitProgress
         )
 
         # Trigger module builds.
diff --git a/lldb/test/API/macosx/ctf/Makefile b/lldb/test/API/macosx/ctf/Makefile
index afe6ab1b5db0..0857e234837e 100644
--- a/lldb/test/API/macosx/ctf/Makefile
+++ b/lldb/test/API/macosx/ctf/Makefile
@@ -4,7 +4,7 @@ MAKE_DSYM := YES
 ifeq "$(COMPRESS_CTF)" "YES"
 	COMPRESS := -c
 else
-	 COMPRESS :=
+	COMPRESS :=
 endif
 
 all: a.out a.ctf
diff --git a/lldb/test/API/macosx/ctf/TestCTF.py b/lldb/test/API/macosx/ctf/TestCTF.py
index f5fd29f6ed96..fed3a8886dd3 100644
--- a/lldb/test/API/macosx/ctf/TestCTF.py
+++ b/lldb/test/API/macosx/ctf/TestCTF.py
@@ -53,6 +53,7 @@ class TestCTF(TestBase):
                 "[2] = 'b'",
                 "[3] = 'c'",
                 'u = (i = 1, s = "")',
+                "b = false",
                 "f = 0x0000000000000000",
             ],
         )
diff --git a/lldb/test/API/macosx/ctf/test.c b/lldb/test/API/macosx/ctf/test.c
index 358006646e76..a15f7a516133 100644
--- a/lldb/test/API/macosx/ctf/test.c
+++ b/lldb/test/API/macosx/ctf/test.c
@@ -1,3 +1,4 @@
+#include <stdbool.h>
 #include <stdio.h>
 
 struct ForwardDecl;
@@ -24,6 +25,7 @@ typedef struct MyNestedStruct {
   char a[4];
   MyEnumT e;
   MyUnionT u;
+  _Bool b;
 } MyNestedStructT;
 
 typedef struct MyStruct {
@@ -54,6 +56,7 @@ void populate(MyInt i) {
   foo.n.a[2] = 'c';
   foo.n.a[3] = 'd';
   foo.n.e = eOne;
+  foo.n.b = false;
   foo.f = NULL;
   forward = NULL;
   bar.b = i;
diff --git a/lldb/test/API/macosx/rosetta/TestRosetta.py b/lldb/test/API/macosx/rosetta/TestRosetta.py
index 669db95a1624..ce40de475ef1 100644
--- a/lldb/test/API/macosx/rosetta/TestRosetta.py
+++ b/lldb/test/API/macosx/rosetta/TestRosetta.py
@@ -49,7 +49,7 @@ class TestRosetta(TestBase):
         if rosetta_debugserver_installed():
             broadcaster = self.dbg.GetBroadcaster()
             listener = lldbutil.start_listening_from(
-                broadcaster, lldb.eBroadcastBitWarning
+                broadcaster, lldb.SBDebugger.eBroadcastBitWarning
             )
 
             target, process, thread, bkpt = lldbutil.run_to_source_breakpoint(
diff --git a/lldb/test/API/python_api/type/TestTypeList.py b/lldb/test/API/python_api/type/TestTypeList.py
index 09c1dee80ef6..81c44f7a39d6 100644
--- a/lldb/test/API/python_api/type/TestTypeList.py
+++ b/lldb/test/API/python_api/type/TestTypeList.py
@@ -18,7 +18,7 @@ class TypeAndTypeListTestCase(TestBase):
         self.source = "main.cpp"
         self.line = line_number(self.source, "// Break at this line")
 
-    def _find_nested_type_in_Task_pointer(self, pointer_type):
+    def _find_nested_type_in_Pointer_template_arg(self, pointer_type):
         self.assertTrue(pointer_type)
         self.DebugSBType(pointer_type)
         pointer_info_type = pointer_type.template_args[1]
@@ -33,6 +33,32 @@ class TypeAndTypeListTestCase(TestBase):
         self.assertTrue(pointer_masks2_type)
         self.DebugSBType(pointer_masks2_type)
 
+    def _find_static_field_in_Task_pointer(self, task_pointer):
+        self.assertTrue(task_pointer)
+        self.DebugSBType(task_pointer)
+
+        task_type = task_pointer.GetPointeeType()
+        self.assertTrue(task_type)
+        self.DebugSBType(task_type)
+
+        static_constexpr_field = task_type.GetStaticFieldWithName(
+            "static_constexpr_field"
+        )
+        self.assertTrue(static_constexpr_field)
+        self.assertEqual(static_constexpr_field.GetName(), "static_constexpr_field")
+        self.assertEqual(static_constexpr_field.GetType().GetName(), "const long")
+
+        value = static_constexpr_field.GetConstantValue(self.target())
+        self.DebugSBValue(value)
+        self.assertEqual(value.GetValueAsSigned(), 47)
+
+        static_mutable_field = task_type.GetStaticFieldWithName("static_mutable_field")
+        self.assertTrue(static_mutable_field)
+        self.assertEqual(static_mutable_field.GetName(), "static_mutable_field")
+        self.assertEqual(static_mutable_field.GetType().GetName(), "int")
+
+        self.assertFalse(static_mutable_field.GetConstantValue(self.target()))
+
     @skipIf(compiler="clang", compiler_version=["<", "17.0"])
     def test(self):
         """Exercise SBType and SBTypeList API."""
@@ -168,11 +194,20 @@ class TypeAndTypeListTestCase(TestBase):
 
         # Check that FindDirectNestedType works with types from module and
         # expression ASTs.
-        self._find_nested_type_in_Task_pointer(frame0.FindVariable("pointer").GetType())
-        self._find_nested_type_in_Task_pointer(
+        self._find_nested_type_in_Pointer_template_arg(
+            frame0.FindVariable("pointer").GetType()
+        )
+        self._find_nested_type_in_Pointer_template_arg(
             frame0.EvaluateExpression("pointer").GetType()
         )
 
+        self._find_static_field_in_Task_pointer(
+            frame0.FindVariable("task_head").GetType()
+        )
+        self._find_static_field_in_Task_pointer(
+            frame0.EvaluateExpression("task_head").GetType()
+        )
+
         # We'll now get the child member 'id' from 'task_head'.
         id = task_head.GetChildMemberWithName("id")
         self.DebugSBValue(id)
diff --git a/lldb/test/API/python_api/type/main.cpp b/lldb/test/API/python_api/type/main.cpp
index 391f58e3e587..c86644d91827 100644
--- a/lldb/test/API/python_api/type/main.cpp
+++ b/lldb/test/API/python_api/type/main.cpp
@@ -27,12 +27,15 @@ public:
     enum E : unsigned char {} e;
     union U {
     } u;
+    static constexpr long static_constexpr_field = 47;
+    static int static_mutable_field;
     Task(int i, Task *n):
         id(i),
         next(n),
         type(TASK_TYPE_1)
     {}
 };
+int Task::static_mutable_field = 42;
 
 template <unsigned Value> struct PointerInfo {
   enum Masks1 { pointer_mask };
diff --git a/lldb/test/API/tools/lldb-dap/console/TestDAP_console.py b/lldb/test/API/tools/lldb-dap/console/TestDAP_console.py
index ffa0dc943e06..8f456aaf890c 100644
--- a/lldb/test/API/tools/lldb-dap/console/TestDAP_console.py
+++ b/lldb/test/API/tools/lldb-dap/console/TestDAP_console.py
@@ -4,11 +4,23 @@ Test lldb-dap setBreakpoints request
 
 import dap_server
 import lldbdap_testcase
+import psutil
+from collections import deque
 from lldbsuite.test import lldbutil
 from lldbsuite.test.decorators import *
 from lldbsuite.test.lldbtest import *
 
 
+def get_subprocess(process_name):
+    queue = deque([psutil.Process(os.getpid())])
+    while queue:
+        process = queue.popleft()
+        if process.name() == process_name:
+            return process
+        queue.extend(process.children())
+
+    self.assertTrue(False, "No subprocess with name %s found" % process_name)
+
 class TestDAP_console(lldbdap_testcase.DAPTestCaseBase):
     def check_lldb_command(
         self, lldb_command, contains_string, assert_msg, command_escape_prefix="`"
@@ -104,3 +116,49 @@ class TestDAP_console(lldbdap_testcase.DAPTestCaseBase):
             "Help can be invoked",
             command_escape_prefix="",
         )
+
+    @skipIfWindows
+    @skipIfRemote
+    def test_exit_status_message_sigterm(self):
+        source = "main.cpp"
+        program = self.getBuildArtifact("a.out")
+        self.build_and_launch(program, commandEscapePrefix="")
+        breakpoint1_line = line_number(source, "// breakpoint 1")
+        breakpoint_ids = self.set_source_breakpoints(source, [breakpoint1_line])
+        self.continue_to_breakpoints(breakpoint_ids)
+
+        # Kill lldb-server process.
+        process_name = (
+            "debugserver" if platform.system() in ["Darwin"] else "lldb-server"
+        )
+        process = get_subprocess(process_name)
+        process.terminate()
+        process.wait()
+
+        # Get the console output
+        console_output = self.collect_console(1.0)
+
+        # Verify the exit status message is printed.
+        self.assertIn(
+            "exited with status = -1 (0xffffffff) debugserver died with signal SIGTERM",
+            console_output,
+            "Exit status does not contain message 'exited with status'",
+        )
+
+    @skipIfWindows
+    @skipIfRemote
+    def test_exit_status_message_ok(self):
+        source = "main.cpp"
+        program = self.getBuildArtifact("a.out")
+        self.build_and_launch(program, commandEscapePrefix="")
+        self.continue_to_exit()
+
+        # Get the console output
+        console_output = self.collect_console(1.0)
+
+        # Verify the exit status message is printed.
+        self.assertIn(
+            "exited with status = 0 (0x00000000)",
+            console_output,
+            "Exit status does not contain message 'exited with status'",
+        )
diff --git a/lldb/test/API/tools/lldb-dap/stepInTargets/Makefile b/lldb/test/API/tools/lldb-dap/stepInTargets/Makefile
new file mode 100644
index 000000000000..f772575cd561
--- /dev/null
+++ b/lldb/test/API/tools/lldb-dap/stepInTargets/Makefile
@@ -0,0 +1,6 @@
+	
+ENABLE_THREADS := YES
+
+CXX_SOURCES := main.cpp
+
+include Makefile.rules
diff --git a/lldb/test/API/tools/lldb-dap/stepInTargets/TestDAP_stepInTargets.py b/lldb/test/API/tools/lldb-dap/stepInTargets/TestDAP_stepInTargets.py
new file mode 100644
index 000000000000..6296f6554d07
--- /dev/null
+++ b/lldb/test/API/tools/lldb-dap/stepInTargets/TestDAP_stepInTargets.py
@@ -0,0 +1,68 @@
+"""
+Test lldb-dap stepInTargets request
+"""
+
+import dap_server
+from lldbsuite.test.decorators import *
+from lldbsuite.test.lldbtest import *
+import lldbdap_testcase
+from lldbsuite.test import lldbutil
+
+
+class TestDAP_stepInTargets(lldbdap_testcase.DAPTestCaseBase):
+    @skipIf(
+        archs=no_match(["x86_64"])
+    )  # InstructionControlFlowKind for ARM is not supported yet.
+    def test_basic(self):
+        """
+        Tests the basic stepping in targets with directly calls.
+        """
+        program = self.getBuildArtifact("a.out")
+        self.build_and_launch(program)
+        source = "main.cpp"
+
+        breakpoint_line = line_number(source, "// set breakpoint here")
+        lines = [breakpoint_line]
+        # Set breakpoint in the thread function so we can step the threads
+        breakpoint_ids = self.set_source_breakpoints(source, lines)
+        self.assertEqual(
+            len(breakpoint_ids), len(lines), "expect correct number of breakpoints"
+        )
+        self.continue_to_breakpoints(breakpoint_ids)
+
+        threads = self.dap_server.get_threads()
+        self.assertEqual(len(threads), 1, "expect one thread")
+        tid = threads[0]["id"]
+
+        leaf_frame = self.dap_server.get_stackFrame()
+        self.assertIsNotNone(leaf_frame, "expect a leaf frame")
+
+        # Request all step in targets list and verify the response.
+        step_in_targets_response = self.dap_server.request_stepInTargets(
+            leaf_frame["id"]
+        )
+        self.assertEqual(step_in_targets_response["success"], True, "expect success")
+        self.assertIn(
+            "body", step_in_targets_response, "expect body field in response body"
+        )
+        self.assertIn(
+            "targets",
+            step_in_targets_response["body"],
+            "expect targets field in response body",
+        )
+
+        step_in_targets = step_in_targets_response["body"]["targets"]
+        self.assertEqual(len(step_in_targets), 3, "expect 3 step in targets")
+
+        # Verify the target names are correct.
+        self.assertEqual(step_in_targets[0]["label"], "bar()", "expect bar()")
+        self.assertEqual(step_in_targets[1]["label"], "bar2()", "expect bar2()")
+        self.assertEqual(
+            step_in_targets[2]["label"], "foo(int, int)", "expect foo(int, int)"
+        )
+
+        # Choose to step into second target and verify that we are in bar2()
+        self.stepIn(threadId=tid, targetId=step_in_targets[1]["id"], waitForStop=True)
+        leaf_frame = self.dap_server.get_stackFrame()
+        self.assertIsNotNone(leaf_frame, "expect a leaf frame")
+        self.assertEqual(leaf_frame["name"], "bar2()")
diff --git a/lldb/test/API/tools/lldb-dap/stepInTargets/main.cpp b/lldb/test/API/tools/lldb-dap/stepInTargets/main.cpp
new file mode 100644
index 000000000000..d3c3dbcc139e
--- /dev/null
+++ b/lldb/test/API/tools/lldb-dap/stepInTargets/main.cpp
@@ -0,0 +1,11 @@
+
+int foo(int val, int extra) { return val + extra; }
+
+int bar() { return 22; }
+
+int bar2() { return 54; }
+
+int main(int argc, char const *argv[]) {
+  foo(bar(), bar2()); // set breakpoint here
+  return 0;
+}
diff --git a/lldb/test/Shell/SymbolFile/DWARF/x86/find-basic-function.cpp b/lldb/test/Shell/SymbolFile/DWARF/x86/find-basic-function.cpp
index 30143a41d5e7..b9a63525d071 100644
--- a/lldb/test/Shell/SymbolFile/DWARF/x86/find-basic-function.cpp
+++ b/lldb/test/Shell/SymbolFile/DWARF/x86/find-basic-function.cpp
@@ -59,6 +59,11 @@
 // RUN: lldb-test symbols --name=not_there --find=function %t | \
 // RUN:   FileCheck --check-prefix=EMPTY %s
 
+/// Test a per-module index built by lld.
+// RUN: ld.lld --debug-names %t.o -o %t
+// RUN: lldb-test symbols --name=foo --find=function --function-flags=base %t | \
+// RUN:   FileCheck --check-prefix=BASE %s
+
 // NAMES: Name: .debug_names
 
 // BASE: Found 4 functions:
diff --git a/lldb/test/Shell/SymbolFile/DWARF/x86/find-basic-variable.cpp b/lldb/test/Shell/SymbolFile/DWARF/x86/find-basic-variable.cpp
index 98b4097cdda0..e46fa14489d3 100644
--- a/lldb/test/Shell/SymbolFile/DWARF/x86/find-basic-variable.cpp
+++ b/lldb/test/Shell/SymbolFile/DWARF/x86/find-basic-variable.cpp
@@ -33,6 +33,11 @@
 // RUN: lldb-test symbols --name=not_there --find=variable %t | \
 // RUN:   FileCheck --check-prefix=EMPTY %s
 
+/// Test a per-module index built by lld.
+// RUN: ld.lld --debug-names %t.o -o %t
+// RUN: lldb-test symbols --name=foo --find=variable --context=context %t | \
+// RUN:   FileCheck --check-prefix=CONTEXT %s
+
 // NAMES: Name: .debug_names
 
 // EMPTY: Found 0 variables:
diff --git a/lldb/test/Shell/Unwind/eh-frame-dwarf-unwind-abort.test b/lldb/test/Shell/Unwind/eh-frame-dwarf-unwind-abort.test
index 477a656a711f..d5e66ca5e263 100644
--- a/lldb/test/Shell/Unwind/eh-frame-dwarf-unwind-abort.test
+++ b/lldb/test/Shell/Unwind/eh-frame-dwarf-unwind-abort.test
@@ -9,12 +9,12 @@ process launch
 # CHECK: stop reason = signal SIGTRAP
 
 thread backtrace
-# CHECK: frame #0: {{.*}}`asm_main + 23
+# CHECK: frame #0: {{.*}}`asm_main + 19
 # CHECK: frame #1: {{.*}}`main + {{.*}}
 
 target modules show-unwind -n asm_main
 # CHECK: eh_frame UnwindPlan:
 # CHECK: row[0]:    0: CFA=rsp +8 => rip=[CFA-8]
-# CHECK: row[1]:   14: CFA=rsp+16 => rbp=[CFA-16] rip=[CFA-8]
-# CHECK: row[2]:   17: CFA=rbp+16 => rbp=[CFA-16] rip=[CFA-8]
-# CHECK: row[3]:   22: CFA=rsp +8 => rip=[CFA-8]
+# CHECK: row[1]:   10: CFA=rsp+16 => rbp=[CFA-16] rip=[CFA-8]
+# CHECK: row[2]:   13: CFA=rbp+16 => rbp=[CFA-16] rip=[CFA-8]
+# CHECK: row[3]:   18: CFA=rsp +8 => rip=[CFA-8]
diff --git a/lldb/tools/lldb-dap/DAP.h b/lldb/tools/lldb-dap/DAP.h
index 8015dec9ba8f..5c70a056fea4 100644
--- a/lldb/tools/lldb-dap/DAP.h
+++ b/lldb/tools/lldb-dap/DAP.h
@@ -162,6 +162,8 @@ struct DAP {
   std::vector<std::string> exit_commands;
   std::vector<std::string> stop_commands;
   std::vector<std::string> terminate_commands;
+  // Map step in target id to list of function targets that user can choose.
+  llvm::DenseMap<lldb::addr_t, std::string> step_in_targets;
   // A copy of the last LaunchRequest or AttachRequest so we can reuse its
   // arguments if we get a RestartRequest.
   std::optional<llvm::json::Object> last_launch_or_attach_request;
diff --git a/lldb/tools/lldb-dap/lldb-dap.cpp b/lldb/tools/lldb-dap/lldb-dap.cpp
index 25c5ad56e3d6..8000d68dea7e 100644
--- a/lldb/tools/lldb-dap/lldb-dap.cpp
+++ b/lldb/tools/lldb-dap/lldb-dap.cpp
@@ -50,6 +50,7 @@
 #include <thread>
 #include <vector>
 
+#include "lldb/API/SBStream.h"
 #include "lldb/Host/Config.h"
 #include "llvm/ADT/ArrayRef.h"
 #include "llvm/ADT/DenseMap.h"
@@ -420,8 +421,8 @@ void SendStdOutStdErr(lldb::SBProcess &process) {
 
 void ProgressEventThreadFunction() {
   lldb::SBListener listener("lldb-dap.progress.listener");
-  g_dap.debugger.GetBroadcaster().AddListener(listener,
-                                              lldb::eBroadcastBitProgress);
+  g_dap.debugger.GetBroadcaster().AddListener(
+      listener, lldb::SBDebugger::eBroadcastBitProgress);
   g_dap.broadcaster.AddListener(listener, eBroadcastBitStopProgressThread);
   lldb::SBEvent event;
   bool done = false;
@@ -503,6 +504,10 @@ void EventThreadFunction() {
             SendContinuedEvent();
             break;
           case lldb::eStateExited:
+            lldb::SBStream stream;
+            process.GetStatus(stream);
+            g_dap.SendOutput(OutputType::Console, stream.GetData());
+
             // When restarting, we can get an "exited" event for the process we
             // just killed with the old PID, or even with no PID. In that case
             // we don't have to terminate the session.
@@ -1645,7 +1650,7 @@ void request_initialize(const llvm::json::Object &request) {
   // The debug adapter supports the gotoTargetsRequest.
   body.try_emplace("supportsGotoTargetsRequest", false);
   // The debug adapter supports the stepInTargetsRequest.
-  body.try_emplace("supportsStepInTargetsRequest", false);
+  body.try_emplace("supportsStepInTargetsRequest", true);
   // The debug adapter supports the completions request.
   body.try_emplace("supportsCompletionsRequest", true);
   // The debug adapter supports the disassembly request.
@@ -3180,14 +3185,155 @@ void request_stepIn(const llvm::json::Object &request) {
   llvm::json::Object response;
   FillResponse(request, response);
   auto arguments = request.getObject("arguments");
+
+  std::string step_in_target;
+  uint64_t target_id = GetUnsigned(arguments, "targetId", 0);
+  auto it = g_dap.step_in_targets.find(target_id);
+  if (it != g_dap.step_in_targets.end())
+    step_in_target = it->second;
+
+  const bool single_thread = GetBoolean(arguments, "singleThread", false);
+  lldb::RunMode run_mode =
+      single_thread ? lldb::eOnlyThisThread : lldb::eOnlyDuringStepping;
   lldb::SBThread thread = g_dap.GetLLDBThread(*arguments);
   if (thread.IsValid()) {
     // Remember the thread ID that caused the resume so we can set the
     // "threadCausedFocus" boolean value in the "stopped" events.
     g_dap.focus_tid = thread.GetThreadID();
-    thread.StepInto();
+    thread.StepInto(step_in_target.c_str(), run_mode);
+  } else {
+    response["success"] = llvm::json::Value(false);
+  }
+  g_dap.SendJSON(llvm::json::Value(std::move(response)));
+}
+
+// "StepInTargetsRequest": {
+//   "allOf": [ { "$ref": "#/definitions/Request" }, {
+//     "type": "object",
+//     "description": "This request retrieves the possible step-in targets for
+//     the specified stack frame.\nThese targets can be used in the `stepIn`
+//     request.\nClients should only call this request if the corresponding
+//     capability `supportsStepInTargetsRequest` is true.", "properties": {
+//       "command": {
+//         "type": "string",
+//         "enum": [ "stepInTargets" ]
+//       },
+//       "arguments": {
+//         "$ref": "#/definitions/StepInTargetsArguments"
+//       }
+//     },
+//     "required": [ "command", "arguments"  ]
+//   }]
+// },
+// "StepInTargetsArguments": {
+//   "type": "object",
+//   "description": "Arguments for `stepInTargets` request.",
+//   "properties": {
+//     "frameId": {
+//       "type": "integer",
+//       "description": "The stack frame for which to retrieve the possible
+//       step-in targets."
+//     }
+//   },
+//   "required": [ "frameId" ]
+// },
+// "StepInTargetsResponse": {
+//   "allOf": [ { "$ref": "#/definitions/Response" }, {
+//     "type": "object",
+//     "description": "Response to `stepInTargets` request.",
+//     "properties": {
+//       "body": {
+//         "type": "object",
+//         "properties": {
+//           "targets": {
+//             "type": "array",
+//             "items": {
+//               "$ref": "#/definitions/StepInTarget"
+//             },
+//             "description": "The possible step-in targets of the specified
+//             source location."
+//           }
+//         },
+//         "required": [ "targets" ]
+//       }
+//     },
+//     "required": [ "body" ]
+//   }]
+// }
+void request_stepInTargets(const llvm::json::Object &request) {
+  llvm::json::Object response;
+  FillResponse(request, response);
+  auto arguments = request.getObject("arguments");
+
+  g_dap.step_in_targets.clear();
+  lldb::SBFrame frame = g_dap.GetLLDBFrame(*arguments);
+  if (frame.IsValid()) {
+    lldb::SBAddress pc_addr = frame.GetPCAddress();
+    lldb::SBAddress line_end_addr =
+        pc_addr.GetLineEntry().GetSameLineContiguousAddressRangeEnd(true);
+    lldb::SBInstructionList insts = g_dap.target.ReadInstructions(
+        pc_addr, line_end_addr, /*flavor_string=*/nullptr);
+
+    if (!insts.IsValid()) {
+      response["success"] = false;
+      response["message"] = "Failed to get instructions for frame.";
+      g_dap.SendJSON(llvm::json::Value(std::move(response)));
+      return;
+    }
+
+    llvm::json::Array step_in_targets;
+    const auto num_insts = insts.GetSize();
+    for (size_t i = 0; i < num_insts; ++i) {
+      lldb::SBInstruction inst = insts.GetInstructionAtIndex(i);
+      if (!inst.IsValid())
+        break;
+
+      lldb::addr_t inst_addr = inst.GetAddress().GetLoadAddress(g_dap.target);
+
+      // Note: currently only x86/x64 supports flow kind.
+      lldb::InstructionControlFlowKind flow_kind =
+          inst.GetControlFlowKind(g_dap.target);
+      if (flow_kind == lldb::eInstructionControlFlowKindCall) {
+        // Use call site instruction address as id which is easy to debug.
+        llvm::json::Object step_in_target;
+        step_in_target["id"] = inst_addr;
+
+        llvm::StringRef call_operand_name = inst.GetOperands(g_dap.target);
+        lldb::addr_t call_target_addr;
+        if (call_operand_name.getAsInteger(0, call_target_addr))
+          continue;
+
+        lldb::SBAddress call_target_load_addr =
+            g_dap.target.ResolveLoadAddress(call_target_addr);
+        if (!call_target_load_addr.IsValid())
+          continue;
+
+        // The existing ThreadPlanStepInRange only accept step in target
+        // function with debug info.
+        lldb::SBSymbolContext sc = g_dap.target.ResolveSymbolContextForAddress(
+            call_target_load_addr, lldb::eSymbolContextFunction);
+
+        // The existing ThreadPlanStepInRange only accept step in target
+        // function with debug info.
+        std::string step_in_target_name;
+        if (sc.IsValid() && sc.GetFunction().IsValid())
+          step_in_target_name = sc.GetFunction().GetDisplayName();
+
+        // Skip call sites if we fail to resolve its symbol name.
+        if (step_in_target_name.empty())
+          continue;
+
+        g_dap.step_in_targets.try_emplace(inst_addr, step_in_target_name);
+        step_in_target.try_emplace("label", step_in_target_name);
+        step_in_targets.emplace_back(std::move(step_in_target));
+      }
+    }
+    llvm::json::Object body;
+    body.try_emplace("targets", std::move(step_in_targets));
+    response.try_emplace("body", std::move(body));
   } else {
     response["success"] = llvm::json::Value(false);
+    response["message"] = "Failed to get frame for input frameId.";
   }
   g_dap.SendJSON(llvm::json::Value(std::move(response)));
 }
@@ -3904,6 +4050,7 @@ void RegisterRequestCallbacks() {
   g_dap.RegisterRequestCallback("source", request_source);
   g_dap.RegisterRequestCallback("stackTrace", request_stackTrace);
   g_dap.RegisterRequestCallback("stepIn", request_stepIn);
+  g_dap.RegisterRequestCallback("stepInTargets", request_stepInTargets);
   g_dap.RegisterRequestCallback("stepOut", request_stepOut);
   g_dap.RegisterRequestCallback("threads", request_threads);
   g_dap.RegisterRequestCallback("variables", request_variables);
diff --git a/lldb/unittests/Host/HostTest.cpp b/lldb/unittests/Host/HostTest.cpp
index 5e01a6835c03..a1d8a3b7f485 100644
--- a/lldb/unittests/Host/HostTest.cpp
+++ b/lldb/unittests/Host/HostTest.cpp
@@ -7,6 +7,7 @@
 //===----------------------------------------------------------------------===//
 
 #include "lldb/Host/Host.h"
+#include "lldb/Utility/ProcessInfo.h"
 #include "gtest/gtest.h"
 
 using namespace lldb_private;
@@ -25,3 +26,23 @@ TEST(Host, GetEnvironment) {
   ASSERT_EQ("Host::GetEnvironment",
             Host::GetEnvironment().lookup("LLDB_TEST_ENVIRONMENT_VAR"));
 }
+
+TEST(Host, ProcessInstanceInfoCumulativeUserTimeIsValid) {
+  ProcessInstanceInfo info;
+  info.SetCumulativeUserTime(ProcessInstanceInfo::timespec{0, 0});
+  EXPECT_FALSE(info.CumulativeUserTimeIsValid());
+  info.SetCumulativeUserTime(ProcessInstanceInfo::timespec{0, 1});
+  EXPECT_TRUE(info.CumulativeUserTimeIsValid());
+  info.SetCumulativeUserTime(ProcessInstanceInfo::timespec{1, 0});
+  EXPECT_TRUE(info.CumulativeUserTimeIsValid());
+}
+
+TEST(Host, ProcessInstanceInfoCumulativeSystemTimeIsValid) {
+  ProcessInstanceInfo info;
+  info.SetCumulativeSystemTime(ProcessInstanceInfo::timespec{0, 0});
+  EXPECT_FALSE(info.CumulativeSystemTimeIsValid());
+  info.SetCumulativeSystemTime(ProcessInstanceInfo::timespec{0, 1});
+  EXPECT_TRUE(info.CumulativeSystemTimeIsValid());
+  info.SetCumulativeSystemTime(ProcessInstanceInfo::timespec{1, 0});
+  EXPECT_TRUE(info.CumulativeSystemTimeIsValid());
+}
\ No newline at end of file
diff --git a/lldb/unittests/Host/linux/HostTest.cpp b/lldb/unittests/Host/linux/HostTest.cpp
index 733909902474..8ecaf3ec0dec 100644
--- a/lldb/unittests/Host/linux/HostTest.cpp
+++ b/lldb/unittests/Host/linux/HostTest.cpp
@@ -69,17 +69,21 @@ TEST_F(HostTest, GetProcessInfo) {
   EXPECT_EQ(HostInfo::GetArchitecture(HostInfo::eArchKindDefault),
             Info.GetArchitecture());
   // Test timings
-  /*
-   * This is flaky in the buildbots on all archs
+  // In some sense this is a pretty trivial test. What it is trying to
+  // accomplish is just to validate that these values are never decreasing
+  // which would be unambiguously wrong. We can not reliably show them
+  // to be always increasing because the microsecond granularity means that,
+  // with hardware variations the number of loop iterations need to always
+  // be increasing for faster and faster machines.
   ASSERT_TRUE(Host::GetProcessInfo(getpid(), Info));
   ProcessInstanceInfo::timespec user_time = Info.GetUserTime();
   static volatile unsigned u = 0;
   for (unsigned i = 0; i < 10'000'000; i++) {
-    u = i;
+    u += i;
   }
+  ASSERT_TRUE(u > 0);
   ASSERT_TRUE(Host::GetProcessInfo(getpid(), Info));
   ProcessInstanceInfo::timespec next_user_time = Info.GetUserTime();
-  ASSERT_TRUE(user_time.tv_sec < next_user_time.tv_sec ||
-              user_time.tv_usec < next_user_time.tv_usec);
-  */
+  ASSERT_TRUE(user_time.tv_sec <= next_user_time.tv_sec ||
+              user_time.tv_usec <= next_user_time.tv_usec);
 }
diff --git a/llvm/cmake/modules/llvm-driver-template.cpp.in b/llvm/cmake/modules/llvm-driver-template.cpp.in
index 71aca6cd140c..1470ef1f0616 100644
--- a/llvm/cmake/modules/llvm-driver-template.cpp.in
+++ b/llvm/cmake/modules/llvm-driver-template.cpp.in
@@ -6,9 +6,9 @@
 //
 //===----------------------------------------------------------------------===//
 
-#include "llvm/Support/LLVMDriver.h"
 #include "llvm/ADT/ArrayRef.h"
 #include "llvm/Support/InitLLVM.h"
+#include "llvm/Support/LLVMDriver.h"
 
 int @TOOL_NAME@_main(int argc, char **, const llvm::ToolContext &);
 
diff --git a/llvm/docs/AMDGPUDwarfExtensionsForHeterogeneousDebugging.rst b/llvm/docs/AMDGPUDwarfExtensionsForHeterogeneousDebugging.rst
index 5dd8df231d30..7d4e06ebf049 100644
--- a/llvm/docs/AMDGPUDwarfExtensionsForHeterogeneousDebugging.rst
+++ b/llvm/docs/AMDGPUDwarfExtensionsForHeterogeneousDebugging.rst
@@ -2487,7 +2487,7 @@ type.
     .. note::
 
       Could also consider adding ``DW_OP_LLVM_aspace_breg0,
-      DW_OP_LLVM_aspace_breg1, ..., DW_OP_LLVM_aspace_bref31`` which would save
+      DW_OP_LLVM_aspace_breg1, ..., DW_OP_LLVM_aspace_breg31`` which would save
       encoding size.
 
 .. _amdgpu-dwarf-register-location-description-operations:
diff --git a/llvm/docs/CommandGuide/llvm-objcopy.rst b/llvm/docs/CommandGuide/llvm-objcopy.rst
index 57d6280d57c8..a62acfc8fdcd 100644
--- a/llvm/docs/CommandGuide/llvm-objcopy.rst
+++ b/llvm/docs/CommandGuide/llvm-objcopy.rst
@@ -366,12 +366,12 @@ them.
 
 .. option:: --keep-global-symbol <symbol>, -G
 
- Make all symbols local in the output, except for symbols with the name
+ Mark all symbols local in the output, except for symbols with the name
  ``<symbol>``. Can be specified multiple times to ignore multiple symbols.
 
 .. option:: --keep-global-symbols <filename>
 
- Make all symbols local in the output, except for symbols named in the file
+ Mark all symbols local in the output, except for symbols named in the file
  ``<filename>``. In the file, each line represents a single symbol, with leading
  and trailing whitespace ignored, as is anything following a '#'. Can be
  specified multiple times to read names from multiple files.
@@ -395,7 +395,7 @@ them.
 
 .. option:: --localize-hidden
 
- Make all symbols with hidden or internal visibility local in the output.
+ Mark all symbols with hidden or internal visibility local in the output.
 
 .. option:: --localize-symbol <symbol>, -L
 
diff --git a/llvm/docs/ConvergenceAndUniformity.rst b/llvm/docs/ConvergenceAndUniformity.rst
index 0e97595508f9..863cebd91a20 100644
--- a/llvm/docs/ConvergenceAndUniformity.rst
+++ b/llvm/docs/ConvergenceAndUniformity.rst
@@ -10,34 +10,61 @@ Convergence And Uniformity
 Introduction
 ============
 
-Some parallel environments execute threads in groups that allow
-communication within the group using special primitives called
-*convergent* operations. The outcome of a convergent operation is
-sensitive to the set of threads that executes it "together", i.e.,
-convergently.
-
-A value is said to be *uniform* across a set of threads if it is the
-same across those threads, and *divergent* otherwise. Correspondingly,
-a branch is said to be a uniform branch if its condition is uniform,
-and it is a divergent branch otherwise.
-
-Whether threads are *converged* or not depends on the paths they take
-through the control flow graph. Threads take different outgoing edges
-at a *divergent branch*. Divergent branches constrain
+In some environments, groups of threads execute the same program in parallel,
+where efficient communication within a group is established using special
+primitives called :ref:`convergent operations<convergent_operations>`. The
+outcome of a convergent operation is sensitive to the set of threads that
+participate in it.
+
+The intuitive picture of *convergence* is built around threads executing in
+"lock step" --- a set of threads is thought of as *converged* if they are all
+executing "the same sequence of instructions together". Such threads may
+*diverge* at a *divergent branch*, and they may later *reconverge* at some
+common program point.
+
+In this intuitive picture, when converged threads execute an instruction, the
+resulting value is said to be *uniform* if it is the same in those threads, and
+*divergent* otherwise. Correspondingly, a branch is said to be a uniform branch
+if its condition is uniform, and it is a divergent branch otherwise.
+
+But the assumption of lock-step execution is not necessary for describing
+communication at convergent operations. It also constrains the implementation
+(compiler as well as hardware) by overspecifying how threads execute in such a
+parallel environment. To eliminate this assumption:
+
+- We define convergence as a relation between the execution of each instruction
+  by different threads and not as a relation between the threads themselves.
+  This definition is reasonable for known targets and is compatible with the
+  semantics of :ref:`convergent operations<convergent_operations>` in LLVM IR.
+- We also define uniformity in terms of this convergence. The output of an
+  instruction can be examined for uniformity across multiple threads only if the
+  corresponding executions of that instruction are converged.
+
+This document decribes a static analysis for determining convergence at each
+instruction in a function. The analysis extends previous work on divergence
+analysis [DivergenceSPMD]_ to cover irreducible control-flow. The described
+analysis is used in LLVM to implement a UniformityAnalysis that determines the
+uniformity of value(s) computed at each instruction in an LLVM IR or MIR
+function.
+
+.. [DivergenceSPMD] Julian Rosemann, Simon Moll, and Sebastian
+   Hack. 2021. An Abstract Interpretation for SPMD Divergence on
+   Reducible Control Flow Graphs. Proc. ACM Program. Lang. 5, POPL,
+   Article 31 (January 2021), 35 pages.
+   https://doi.org/10.1145/3434312
+
+Motivation
+==========
+
+Divergent branches constrain
 program transforms such as changing the CFG or moving a convergent
 operation to a different point of the CFG. Performing these
 transformations across a divergent branch can change the sets of
 threads that execute convergent operations convergently. While these
-constraints are out of scope for this document, the described
-*uniformity analysis* allows these transformations to identify
+constraints are out of scope for this document,
+uniformity analysis allows these transformations to identify
 uniform branches where these constraints do not hold.
 
-Convergence and
-uniformity are inter-dependent: When threads diverge at a divergent
-branch, they may later *reconverge* at a common program point.
-Subsequent operations are performed convergently, but the inputs may
-be non-uniform, thus producing divergent outputs.
-
 Uniformity is also useful by itself on targets that execute threads in
 groups with shared execution resources (e.g. waves, warps, or
 subgroups):
@@ -50,18 +77,6 @@ subgroups):
   branches, since the whole group of threads follows either one side
   of the branch or the other.
 
-This document presents a definition of convergence that is reasonable
-for real targets and is compatible with the currently implicit
-semantics of convergent operations in LLVM IR. This is accompanied by
-a *uniformity analysis* that extends previous work on divergence analysis
-[DivergenceSPMD]_ to cover irreducible control-flow.
-
-.. [DivergenceSPMD] Julian Rosemann, Simon Moll, and Sebastian
-   Hack. 2021. An Abstract Interpretation for SPMD Divergence on
-   Reducible Control Flow Graphs. Proc. ACM Program. Lang. 5, POPL,
-   Article 31 (January 2021), 35 pages.
-   https://doi.org/10.1145/3434312
-
 Terminology
 ===========
 
@@ -133,12 +148,6 @@ meaning. Dynamic instances listed in the same column are converged.
 Convergence
 ===========
 
-*Converged-with* is a transitive symmetric relation over dynamic
-instances produced by *different threads* for the *same static
-instance*. Informally, two threads that produce converged dynamic
-instances are said to be *converged*, and they are said to execute
-that static instance *convergently*, at that point in the execution.
-
 *Convergence-before* is a strict partial order over dynamic instances
 that is defined as the transitive closure of:
 
@@ -171,11 +180,16 @@ to be converged (i.e., related to each other in the converged-with
 relation). The resulting convergence order includes the edges ``P ->
 Q2``, ``Q1 -> R``, ``P -> R``, ``P -> T``, etc.
 
-The fact that *convergence-before* is a strict partial order is a
-constraint on the *converged-with* relation. It is trivially satisfied
-if different dynamic instances are never converged. It is also
-trivially satisfied for all known implementations for which
-convergence plays some role.
+*Converged-with* is a transitive symmetric relation over dynamic instances
+produced by *different threads* for the *same static instance*.
+
+It is impractical to provide any one definition for the *converged-with*
+relation, since different environments may wish to relate dynamic instances in
+different ways. The fact that *convergence-before* is a strict partial order is
+a constraint on the *converged-with* relation. It is trivially satisfied if
+different dynamic instances are never converged. Below, we provide a relation
+called :ref:`maximal converged-with<convergence-maximal>`, which satisifies
+*convergence-before* and is suitable for known targets.
 
 .. _convergence-note-convergence:
 
@@ -217,14 +231,16 @@ iterations of parent cycles as well.
 
    Dynamic instances ``X1`` and ``X2`` produced by different threads
    for the same static instance ``X`` are converged in the maximal
-   converged-with relation if and only if for every cycle ``C`` with
-   header ``H`` that contains ``X``:
-
-   - every dynamic instance ``H1`` of ``H`` that precedes ``X1`` in
-     the respective thread is convergence-before ``X2``, and,
-   - every dynamic instance ``H2`` of ``H`` that precedes ``X2`` in
-     the respective thread is convergence-before ``X1``,
-   - without assuming that ``X1`` is converged with ``X2``.
+   converged-with relation if and only if:
+
+   - ``X`` is not contained in any cycle, or,
+   - For every cycle ``C`` with header ``H`` that contains ``X``:
+
+     - every dynamic instance ``H1`` of ``H`` that precedes ``X1`` in
+       the respective thread is convergence-before ``X2``, and,
+     - every dynamic instance ``H2`` of ``H`` that precedes ``X2`` in
+       the respective thread is convergence-before ``X1``,
+     - without assuming that ``X1`` is converged with ``X2``.
 
 .. note::
 
diff --git a/llvm/docs/ConvergentOperations.rst b/llvm/docs/ConvergentOperations.rst
index 332675f3edef..5081efffc89a 100644
--- a/llvm/docs/ConvergentOperations.rst
+++ b/llvm/docs/ConvergentOperations.rst
@@ -936,7 +936,8 @@ property <uniformity-analysis>` of static instances in the convergence region of
      1. Both threads executed converged dynamic instances of every token
         definition ``D`` such that ``X`` is in the convergence region of ``D``,
         and,
-     2. For every cycle ``C`` with header ``H`` that contains ``X``:
+     2. Either ``X`` is not contained in any cycle, or, for every cycle ``C``
+        with header ``H`` that contains ``X``:
 
         - every dynamic instance ``H1`` of ``H`` that precedes ``X1`` in the
           respective thread is convergence-before ``X2``, and,
diff --git a/llvm/docs/GettingInvolved.rst b/llvm/docs/GettingInvolved.rst
index a4247796cb65..93be3bd1d854 100644
--- a/llvm/docs/GettingInvolved.rst
+++ b/llvm/docs/GettingInvolved.rst
@@ -228,7 +228,8 @@ what to add to your calendar invite.
      - `Meeting details/agenda <https://docs.google.com/document/d/1Ry8O4-Tm5BFj9AMjr8qTQFU80z-ptiNQ62687NaIvLs/edit?usp=sharing>`__

    * - Floating Point Working Group

      - Every 3rd Wednesday of the month

-     - `gcal <https://calendar.google.com/calendar/u/0?cid=MDI1ODI1MDdiYWM3OWQxODY5MDA3MTI1NjZlYzNmYzY5YjMzYWMyNGQ3ZGUwYThjNzZjN2IxOTk3NmYxOTBjMEBncm91cC5jYWxlbmRhci5nb29nbGUuY29t>`__

+     - `ics <https://calendar.google.com/calendar/ical/02582507bac79d186900712566ec3fc69b33ac24d7de0a8c76c7b19976f190c0%40group.calendar.google.com/private-6e35506dbfe13812e92e9aa8cd5d761d/basic.ics>`__

+       `gcal <https://calendar.google.com/calendar/u/0?cid=MDI1ODI1MDdiYWM3OWQxODY5MDA3MTI1NjZlYzNmYzY5YjMzYWMyNGQ3ZGUwYThjNzZjN2IxOTk3NmYxOTBjMEBncm91cC5jYWxlbmRhci5nb29nbGUuY29t>`__

      - `Meeting details/agenda: <https://docs.google.com/document/d/1QcmUlWftPlBi-Wz6b6PipqJfvjpJ-OuRMRnN9Dm2t0c>`__

 

 .. _office-hours:

diff --git a/llvm/docs/GlobalISel/IRTranslator.rst b/llvm/docs/GlobalISel/IRTranslator.rst
index 6268ebb4f4a1..0cb596c516e2 100644
--- a/llvm/docs/GlobalISel/IRTranslator.rst
+++ b/llvm/docs/GlobalISel/IRTranslator.rst
@@ -6,7 +6,7 @@ IRTranslator
 .. contents::
   :local:
 
-This pass translates the input LLVM-IR ``Function`` to a GMIR
+This pass translates the input LLVM-IR ``Function`` to a :doc:`GMIR`
 ``MachineFunction``. This is typically a direct translation but does
 occasionally get a bit more involved. For example:
 
@@ -51,8 +51,37 @@ Translating Function Calls
 
 The ``IRTranslator`` also implements the ABI's calling convention by lowering
 calls, returns, and arguments to the appropriate physical register usage and
-instruction sequences. This is achieved using the ``CallLowering``
-implementation,
+instruction sequences. This is achieved using the ``CallLowering`` interface,
+which provides several hooks that targets should implement:
+``lowerFormalArguments``, ``lowerReturn``, ``lowerCall`` etc.
+
+In essence, all of these hooks need to find a way to move the argument/return
+values between the virtual registers used in the rest of the function and either
+physical registers or the stack, as dictated by the ABI. This may involve
+splitting large types into smaller ones, introducing sign/zero extensions etc.
+In order to share as much of this code as possible between the different
+backends, ``CallLowering`` makes available a few helpers and interfaces:
+
+* ``ArgInfo`` - used for formal arguments, but also return values, actual
+  arguments and call results; contains info such as the IR type, the virtual
+  registers etc; large values will likely have to be split into several
+  ``ArgInfo`` objects (``CallLowering::splitToValueTypes`` can help with that);
+
+* ``ValueAssigner`` - uses a ``CCAssignFn``, usually generated by TableGen (see
+  :ref:`backend-calling-convs`), to decide where to put each
+  ``ArgInfo`` (physical register or stack); backends can use the provided
+  ``IncomingValueAssigner`` (for formal arguments and call results) and
+  ``OutgoingValueAssigner`` (for actual arguments and function returns), but
+  it's also possible to subclass them;
+
+* ``ValueHandler`` - inserts the necessary instructions for putting each value
+  where it belongs; it has pure virtual methods for assigning values to
+  registers or to addresses, and a host of other helpers;
+
+* ``determineAndHandleAssignments`` (or for more fine grained control,
+  ``determineAssignments`` and ``handleAssignments``) - contains some boilerplate
+  for invoking a given ``ValueAssigner`` and ``ValueHandler`` on a series of
+  ``ArgInfo`` objects.
 
 .. _irtranslator-aggregates:
 
diff --git a/llvm/docs/GlobalISel/MIRPatterns.rst b/llvm/docs/GlobalISel/MIRPatterns.rst
index 728e32470144..d7dce1b978cd 100644
--- a/llvm/docs/GlobalISel/MIRPatterns.rst
+++ b/llvm/docs/GlobalISel/MIRPatterns.rst
@@ -514,3 +514,40 @@ of operands.
     (match (does_not_bind $tmp, $x)
            (G_MUL $dst, $x, $tmp)),
     (apply (COPY $dst, $x))>;
+
+
+
+
+Gallery
+=======
+
+We should use precise patterns that state our intentions. Please avoid
+using wip_match_opcode in patterns.
+
+.. code-block:: text
+  :caption: Example fold zext(trunc:nuw)
+
+  // Imprecise: matches any G_ZEXT
+  def zext : GICombineRule<
+    (defs root:$root),
+    (match (wip_match_opcode G_ZEXT):$root,
+    [{ return Helper.matchZextOfTrunc(*${root}, ${matchinfo}); }]),
+    (apply [{ Helper.applyBuildFn(*${root}, ${matchinfo}); }])>;
+
+
+  // Imprecise: matches G_ZEXT of G_TRUNC
+  def zext_of_trunc : GICombineRule<
+    (defs root:$root),
+    (match (G_TRUNC $src, $x),
+           (G_ZEXT $root, $src),
+    [{ return Helper.matchZextOfTrunc(${root}, ${matchinfo}); }]),
+    (apply [{ Helper.applyBuildFnMO(${root}, ${matchinfo}); }])>;
+
+
+  // Precise: matches G_ZEXT of G_TRUNC with nuw flag
+  def zext_of_trunc_nuw : GICombineRule<
+    (defs root:$root),
+    (match (G_TRUNC $src, $x, (MIFlags NoUWrap)),
+           (G_ZEXT $root, $src),
+    [{ return Helper.matchZextOfTrunc(${root}, ${matchinfo}); }]),
+    (apply [{ Helper.applyBuildFnMO(${root}, ${matchinfo}); }])>;
diff --git a/llvm/docs/LangRef.rst b/llvm/docs/LangRef.rst
index 9592929d79fe..37662f79145d 100644
--- a/llvm/docs/LangRef.rst
+++ b/llvm/docs/LangRef.rst
@@ -12517,7 +12517,7 @@ This instruction requires several arguments:
       ``llvm::GuaranteedTailCallOpt`` is ``true``, or the calling convention
       is ``tailcc``
    -  `Platform-specific constraints are
-      met. <CodeGenerator.html#tailcallopt>`_
+      met. <CodeGenerator.html#tail-call-optimization>`_
 
 #. The optional ``notail`` marker indicates that the optimizers should not add
    ``tail`` or ``musttail`` markers to the call. It is used to prevent tail
@@ -14139,6 +14139,41 @@ Semantics:
 """"""""""
 See description of '``llvm.instrprof.increment``' intrinsic.
 
+'``llvm.instrprof.callsite``' Intrinsic
+^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
+
+Syntax:
+"""""""
+
+::
+
+      declare void @llvm.instrprof.callsite(ptr <name>, i64 <hash>,
+                                            i32 <num-counters>,
+                                            i32 <index>, ptr <callsite>)
+
+Overview:
+"""""""""
+
+.. FIXME: detail when it's emitted once the support is added
+
+The '``llvm.instrprof.callsite``' intrinsic should be emitted before a callsite
+that's not to a "fake" callee (like another intrinsic or asm).
+
+Arguments:
+""""""""""
+The first 4 arguments are similar to ``llvm.instrprof.increment``. The indexing
+is specific to callsites, meaning callsites are indexed from 0, independent from
+the indexes used by the other intrinsics (such as 
+``llvm.instrprof.increment[.step]``).
+
+The last argument is the called value of the callsite this intrinsic precedes.
+
+Semantics:
+""""""""""
+.. FIXME: detail how when the lowering pass is added.
+
+This is lowered by contextual profiling.
+
 '``llvm.instrprof.timestamp``' Intrinsic
 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
 
@@ -18770,7 +18805,7 @@ runtime, then the result vector is a :ref:`poison value <poisonvalues>`. The
 ``idx`` parameter must be a vector index constant type (for most targets this
 will be an integer pointer type).
 
-'``llvm.experimental.vector.reverse``' Intrinsic
+'``llvm.vector.reverse``' Intrinsic
 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
 
 Syntax:
@@ -18779,25 +18814,26 @@ This is an overloaded intrinsic.
 
 ::
 
-      declare <2 x i8> @llvm.experimental.vector.reverse.v2i8(<2 x i8> %a)
-      declare <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> %a)
+      declare <2 x i8> @llvm.vector.reverse.v2i8(<2 x i8> %a)
+      declare <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> %a)
 
 Overview:
 """""""""
 
-The '``llvm.experimental.vector.reverse.*``' intrinsics reverse a vector.
+The '``llvm.vector.reverse.*``' intrinsics reverse a vector.
 The intrinsic takes a single vector and returns a vector of matching type but
 with the original lane order reversed. These intrinsics work for both fixed
-and scalable vectors. While this intrinsic is marked as experimental the
-recommended way to express reverse operations for fixed-width vectors is still
-to use a shufflevector, as that may allow for more optimization opportunities.
+and scalable vectors. While this intrinsic supports all vector types
+the recommended way to express this operation for fixed-width vectors is
+still to use a shufflevector, as that may allow for more optimization
+opportunities.
 
 Arguments:
 """"""""""
 
 The argument to this intrinsic must be a vector.
 
-'``llvm.experimental.vector.deinterleave2``' Intrinsic
+'``llvm.vector.deinterleave2``' Intrinsic
 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
 
 Syntax:
@@ -18806,13 +18842,13 @@ This is an overloaded intrinsic.
 
 ::
 
-      declare {<2 x double>, <2 x double>} @llvm.experimental.vector.deinterleave2.v4f64(<4 x double> %vec1)
-      declare {<vscale x 4 x i32>, <vscale x 4 x i32>}  @llvm.experimental.vector.deinterleave2.nxv8i32(<vscale x 8 x i32> %vec1)
+      declare {<2 x double>, <2 x double>} @llvm.vector.deinterleave2.v4f64(<4 x double> %vec1)
+      declare {<vscale x 4 x i32>, <vscale x 4 x i32>}  @llvm.vector.deinterleave2.nxv8i32(<vscale x 8 x i32> %vec1)
 
 Overview:
 """""""""
 
-The '``llvm.experimental.vector.deinterleave2``' intrinsic constructs two
+The '``llvm.vector.deinterleave2``' intrinsic constructs two
 vectors by deinterleaving the even and odd lanes of the input vector.
 
 This intrinsic works for both fixed and scalable vectors. While this intrinsic
@@ -18824,7 +18860,7 @@ For example:
 
 .. code-block:: text
 
-  {<2 x i64>, <2 x i64>} llvm.experimental.vector.deinterleave2.v4i64(<4 x i64> <i64 0, i64 1, i64 2, i64 3>); ==> {<2 x i64> <i64 0, i64 2>, <2 x i64> <i64 1, i64 3>}
+  {<2 x i64>, <2 x i64>} llvm.vector.deinterleave2.v4i64(<4 x i64> <i64 0, i64 1, i64 2, i64 3>); ==> {<2 x i64> <i64 0, i64 2>, <2 x i64> <i64 1, i64 3>}
 
 Arguments:
 """"""""""
@@ -18832,7 +18868,7 @@ Arguments:
 The argument is a vector whose type corresponds to the logical concatenation of
 the two result types.
 
-'``llvm.experimental.vector.interleave2``' Intrinsic
+'``llvm.vector.interleave2``' Intrinsic
 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
 
 Syntax:
@@ -18841,13 +18877,13 @@ This is an overloaded intrinsic.
 
 ::
 
-      declare <4 x double> @llvm.experimental.vector.interleave2.v4f64(<2 x double> %vec1, <2 x double> %vec2)
-      declare <vscale x 8 x i32> @llvm.experimental.vector.interleave2.nxv8i32(<vscale x 4 x i32> %vec1, <vscale x 4 x i32> %vec2)
+      declare <4 x double> @llvm.vector.interleave2.v4f64(<2 x double> %vec1, <2 x double> %vec2)
+      declare <vscale x 8 x i32> @llvm.vector.interleave2.nxv8i32(<vscale x 4 x i32> %vec1, <vscale x 4 x i32> %vec2)
 
 Overview:
 """""""""
 
-The '``llvm.experimental.vector.interleave2``' intrinsic constructs a vector
+The '``llvm.vector.interleave2``' intrinsic constructs a vector
 by interleaving two input vectors.
 
 This intrinsic works for both fixed and scalable vectors. While this intrinsic
@@ -18859,7 +18895,7 @@ For example:
 
 .. code-block:: text
 
-   <4 x i64> llvm.experimental.vector.interleave2.v4i64(<2 x i64> <i64 0, i64 2>, <2 x i64> <i64 1, i64 3>); ==> <4 x i64> <i64 0, i64 1, i64 2, i64 3>
+   <4 x i64> llvm.vector.interleave2.v4i64(<2 x i64> <i64 0, i64 2>, <2 x i64> <i64 1, i64 3>); ==> <4 x i64> <i64 0, i64 1, i64 2, i64 3>
 
 Arguments:
 """"""""""
@@ -18905,7 +18941,7 @@ The '``llvm.experimental.cttz.elts``' intrinsic counts the trailing (least
 significant) zero elements in a vector. If ``src == 0`` the result is the
 number of elements in the input vector.
 
-'``llvm.experimental.vector.splice``' Intrinsic
+'``llvm.vector.splice``' Intrinsic
 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
 
 Syntax:
@@ -18914,13 +18950,13 @@ This is an overloaded intrinsic.
 
 ::
 
-      declare <2 x double> @llvm.experimental.vector.splice.v2f64(<2 x double> %vec1, <2 x double> %vec2, i32 %imm)
-      declare <vscale x 4 x i32> @llvm.experimental.vector.splice.nxv4i32(<vscale x 4 x i32> %vec1, <vscale x 4 x i32> %vec2, i32 %imm)
+      declare <2 x double> @llvm.vector.splice.v2f64(<2 x double> %vec1, <2 x double> %vec2, i32 %imm)
+      declare <vscale x 4 x i32> @llvm.vector.splice.nxv4i32(<vscale x 4 x i32> %vec1, <vscale x 4 x i32> %vec2, i32 %imm)
 
 Overview:
 """""""""
 
-The '``llvm.experimental.vector.splice.*``' intrinsics construct a vector by
+The '``llvm.vector.splice.*``' intrinsics construct a vector by
 concatenating elements from the first input vector with elements of the second
 input vector, returning a vector of the same type as the input vectors. The
 signed immediate, modulo the number of elements in the vector, is the index
@@ -18931,7 +18967,7 @@ immediate, it extracts ``-imm`` trailing elements from the first vector, and
 the remaining elements from ``%vec2``.
 
 These intrinsics work for both fixed and scalable vectors. While this intrinsic
-is marked as experimental, the recommended way to express this operation for
+supports all vector types the recommended way to express this operation for
 fixed-width vectors is still to use a shufflevector, as that may allow for more
 optimization opportunities.
 
@@ -18939,8 +18975,8 @@ For example:
 
 .. code-block:: text
 
- llvm.experimental.vector.splice(<A,B,C,D>, <E,F,G,H>, 1);  ==> <B, C, D, E> index
- llvm.experimental.vector.splice(<A,B,C,D>, <E,F,G,H>, -3); ==> <B, C, D, E> trailing elements
+ llvm.vector.splice(<A,B,C,D>, <E,F,G,H>, 1);  ==> <B, C, D, E> index
+ llvm.vector.splice(<A,B,C,D>, <E,F,G,H>, -3); ==> <B, C, D, E> trailing elements
 
 
 Arguments:
@@ -22163,7 +22199,7 @@ Overview:
 """""""""
 
 The '``llvm.experimental.vp.splice.*``' intrinsic is the vector length
-predicated version of the '``llvm.experimental.vector.splice.*``' intrinsic.
+predicated version of the '``llvm.vector.splice.*``' intrinsic.
 
 Arguments:
 """"""""""
@@ -22222,7 +22258,7 @@ Overview:
 """""""""
 
 The '``llvm.experimental.vp.reverse.*``' intrinsic is the vector length
-predicated version of the '``llvm.experimental.vector.reverse.*``' intrinsic.
+predicated version of the '``llvm.vector.reverse.*``' intrinsic.
 
 Arguments:
 """"""""""
diff --git a/llvm/docs/MemoryModelRelaxationAnnotations.rst b/llvm/docs/MemoryModelRelaxationAnnotations.rst
new file mode 100644
index 000000000000..c79675d30f62
--- /dev/null
+++ b/llvm/docs/MemoryModelRelaxationAnnotations.rst
@@ -0,0 +1,481 @@
+===================================
+Memory Model Relaxation Annotations
+===================================
+
+.. contents::
+   :local:
+
+Introduction
+============
+
+Memory Model Relaxation Annotations (MMRAs) are target-defined properties
+on instructions that can be used to selectively relax constraints placed
+by the memory model. For example:
+
+* The use of ``VulkanMemoryModel`` in a SPIRV program allows certain
+  memory operations to be reordered across ``acquire`` or ``release``
+  operations.
+* OpenCL APIs expose primitives to only fence a specific set of address
+  spaces. Carrying that information to the backend can enable the
+  use of faster synchronization instructions, rather than fencing all
+  address spaces everytime.
+
+MMRAs offer an opt-in system for targets to relax the default LLVM
+memory model.
+As such, they are attached to an operation using LLVM metadata which
+can always be dropped without affecting correctness.
+
+Definitions
+===========
+
+memory operation
+    A load, a store, an atomic, or a function call that is marked as
+    accessing memory.
+
+synchronizing operation
+    An instruction that synchronizes memory with other threads (e.g.
+    an atomic or a fence).
+
+tag
+    Metadata attached to a memory or synchronizing operation
+    that represents some target-defined property regarding memory
+    synchronization.
+
+    An operation may have multiple tags that each represent a different
+    property.
+
+    A tag is composed of a pair of metadata string: a *prefix* and a *suffix*.
+
+    In LLVM IR, the pair is represented using a metadata tuple.
+    In other cases (comments, documentation, etc.), we may use the
+    ``prefix:suffix`` notation.
+    For example:
+
+    .. code-block::
+      :caption: Example: Tags in Metadata
+
+      !0 = !{!"scope", !"workgroup"}  # scope:workgroup
+      !1 = !{!"scope", !"device"}     # scope:device
+      !2 = !{!"scope", !"system"}     # scope:system
+
+    .. note::
+
+      The only semantics relevant to the optimizer is the
+      "compatibility" relation defined below. All other
+      semantics are target defined.
+
+    Tags can also be organised in lists to allow operations
+    to specify all of the tags they belong to. Such a list
+    is referred to as a "set of tags".
+
+    .. code-block::
+      :caption: Example: Set of Tags in Metadata
+
+      !0 = !{!"scope", !"workgroup"}
+      !1 = !{!"sync-as", !"private"}
+      !2 = !{!0, !2}
+
+    .. note::
+
+      If an operation does not have MMRA metadata, it's treated as if
+      it has an empty list (``!{}``) of tags.
+
+    Note that it is not an error if a tag is not recognized by the
+    instruction it is applied to, or by the current target.
+    Such tags are simply ignored.
+
+    Both synchronizing operations and memory operations can have
+    zero or more tags attached to them using the ``!mmra`` syntax.
+
+    For the sake of readability in examples below,
+    we use a (non-functional) short syntax to represent MMMRA metadata:
+
+    .. code-block::
+      :caption: Short Syntax Example
+
+      store %ptr1 # foo:bar
+      store %ptr1 !mmra !{!"foo", !"bar"}
+
+    These two notations can be used in this document and are strictly
+    equivalent. However, only the second version is functional.
+
+compatibility
+    Two sets of tags are said to be *compatible* iff, for every unique
+    tag prefix P present in at least one set:
+
+    - the other set contains no tag with prefix P, or
+    - at least one tag with prefix P is common to both sets.
+
+    The above definition implies that an empty set is always compatible
+    with any other set. This is an important property as it ensures that
+    if a transform drops the metadata on an operation, it can never affect
+    correctness. In other words, the memory model cannot be relaxed further
+    by deleting metadata from instructions.
+
+.. _HappensBefore:
+
+The *happens-before* Relation
+==============================
+
+Compatibility checks can be used to opt out of the *happens-before* relation
+established between two instructions.
+
+Ordering
+    When two instructions' metadata are not compatible, any program order
+    between them are not in *happens-before*.
+
+    For example, consider two tags ``foo:bar`` and
+    ``foo:baz`` exposed by a target:
+
+    .. code-block::
+
+       A: store %ptr1                 # foo:bar
+       B: store %ptr2                 # foo:baz
+       X: store atomic release %ptr3  # foo:bar
+
+    In the above figure, ``A`` is compatible with ``X``, and hence ``A``
+    happens-before ``X``. But ``B`` is not compatible with
+    ``X``, and hence it is not happens-before ``X``.
+
+Synchronization
+    If an synchronizing operation has one or more tags, then whether it
+    synchronizes-with and participates in the  ``seq_cst`` order with
+    other operations is target dependent.
+
+    Whether the following example synchronizes with another sequence depends
+    on the target-defined semantics of ``foo:bar`` and ``foo:bux``.
+
+    .. code-block::
+
+       fence release               # foo:bar
+       store atomic %ptr1          # foo:bux
+
+Examples
+--------
+
+Example 1:
+    .. code-block::
+
+      A: store ptr addrspace(1) %ptr2                  # sync-as:1 vulkan:nonprivate
+      B: store atomic release ptr addrspace(1) %ptr3   # sync-as:0 vulkan:nonprivate
+
+    A and B are not ordered relative to each other
+    (no *happens-before*) because their sets of tags are not compatible.
+
+    Note that the ``sync-as`` value does not have to match the ``addrspace`` value.
+    e.g. In Example 1, a store-release to a location in ``addrspace(1)`` wants to
+    only synchronize with operations happening in ``addrspace(0)``.
+
+Example 2:
+    .. code-block::
+
+      A: store ptr addrspace(1) %ptr2                 # sync-as:1 vulkan:nonprivate
+      B: store atomic release ptr addrspace(1) %ptr3  # sync-as:1 vulkan:nonprivate
+
+    The ordering of A and B is unaffected because their set of tags are
+    compatible.
+
+    Note that A and B may or may not be in *happens-before* due to other reasons.
+
+Example 3:
+    .. code-block::
+
+      A: store ptr addrspace(1) %ptr2                 # sync-as:1 vulkan:nonprivate
+      B: store atomic release ptr addrspace(1) %ptr3  # vulkan:nonprivate
+
+    The ordering of A and B is unaffected because their set of tags are
+    compatible.
+
+Example 4:
+    .. code-block::
+
+      A: store ptr addrspace(1) %ptr2                 # sync-as:1
+      B: store atomic release ptr addrspace(1) %ptr3  # sync-as:2
+
+    A and B do not have to be ordered relative to each other
+    (no *happens-before*) because their sets of tags are not compatible.
+
+Use-cases
+=========
+
+SPIRV ``NonPrivatePointer``
+---------------------------
+
+MMRAs can support the SPIRV capability
+``VulkanMemoryModel``, where synchronizing operations only affect
+memory operations that specify ``NonPrivatePointer`` semantics.
+
+The example below is generated from a SPIRV program using the
+following recipe:
+
+- Add ``vulkan:nonprivate`` to every synchronizing operation.
+- Add ``vulkan:nonprivate`` to every non-atomic memory operation
+  that is marked ``NonPrivatePointer``.
+- Add ``vulkan:private`` to tags of every non-atomic memory operation
+  that is not marked ``NonPrivatePointer``.
+
+.. code-block::
+
+   Thread T1:
+    A: store %ptr1                 # vulkan:nonprivate
+    B: store %ptr2                 # vulkan:private
+    X: store atomic release %ptr3  # vulkan:nonprivate
+
+   Thread T2:
+    Y: load atomic acquire %ptr3   # vulkan:nonprivate
+    C: load %ptr2                  # vulkan:private
+    D: load %ptr1                  # vulkan:nonprivate
+
+Compatibility ensures that operation ``A`` is ordered
+relative to ``X`` while operation ``D`` is ordered relative to ``Y``.
+If ``X`` synchronizes with ``Y``, then ``A`` happens-before ``D``.
+No such relation can be inferred about operations ``B`` and ``C``.
+
+.. note::
+   The `Vulkan Memory Model <https://registry.khronos.org/vulkan/specs/1.3-extensions/html/vkspec.html#memory-model-non-private>`_
+   considers all atomic operation non-private.
+
+   Whether ``vulkan:nonprivate`` would be specified on atomic operations is
+   an implementation detail, as an atomic operation is always ``nonprivate``.
+   The implementation may choose to be explicit and emit IR with
+   ``vulkan:nonprivate`` on every atomic operation, or it could choose to
+   only emit ``vulkan::private`` and assume ``vulkan:nonprivate``
+   by default.
+
+Operations marked with ``vulkan:private`` effectively opt out of the
+happens-before order in a SPIRV program since they are incompatible
+with every synchronizing operation. Note that SPIRV operations that
+are not marked ``NonPrivatePointer`` are not entirely private to the
+thread --- they are implicitly synchronized at the start or end of a
+thread by the Vulkan *system-synchronizes-with* relationship. This
+example assumes that the target-defined semantics of
+``vulkan:private`` correctly implements this property.
+
+This scheme is general enough to express the interoperability of SPIRV
+programs with other environments.
+
+.. code-block::
+
+   Thread T1:
+   A: store %ptr1                 # vulkan:nonprivate
+   X: store atomic release %ptr2  # vulkan:nonprivate
+
+   Thread T2:
+   Y: load atomic acquire %ptr2   # foo:bar
+   B: load %ptr1
+
+In the above example, thread ``T1`` originates from a SPIRV program
+while thread ``T2`` originates from a non-SPIRV program. Whether ``X``
+can synchronize with ``Y`` is target defined.  If ``X`` synchronizes
+with ``Y``, then ``A`` happens before ``B`` (because A/X and
+Y/B are compatible).
+
+Implementation Example
+~~~~~~~~~~~~~~~~~~~~~~
+
+Consider the implementation of SPIRV ``NonPrivatePointer`` on a target
+where all memory operations are cached, and the entire cache is
+flushed or invalidated at a ``release`` or ``acquire`` respectively. A
+possible scheme is that when translating a SPIRV program, memory
+operations marked ``NonPrivatePointer`` should not be cached, and the
+cache contents should not be touched during an ``acquire`` and
+``release`` operation.
+
+This could be implemented using the tags that share the ``vulkan:`` prefix,
+as follows:
+
+- For memory operations:
+
+  - Operations with ``vulkan:nonprivate`` should bypass the cache.
+  - Operations with ``vulkan:private`` should be cached.
+  - Operations that specify neither or both should conservatively
+    bypass the cache to ensure correctness.
+
+- For synchronizing operations:
+
+  - Operations with ``vulkan:nonprivate`` should not flush or
+    invalidate the cache.
+  - Operations with ``vulkan:private`` should flush or invalidate the cache.
+  - Operations that specify neither or both should conservatively
+    flush or invalidate the cache to ensure correctness.
+
+.. note::
+   In such an implementation, dropping the metadata on an operation, while
+   not affecting correctness, may have big performance implications.
+   e.g. an operation bypasses the cache when it shouldn't.
+
+Memory Types
+------------
+
+MMRAs may express the selective synchronization of
+different memory types.
+
+As an example, a target may expose an ``sync-as:<N>`` tag to
+pass information about which address spaces are synchronized by the
+execution of a synchronizing operation.
+
+.. note::
+  Address spaces are used here as a common example, but this concept
+  can apply for other "memory types". What "memory types" means here is
+  up to the target.
+
+.. code-block::
+
+   # let 1 = global address space
+   # let 3 = local address space
+
+   Thread T1:
+   A: store %ptr1                                  # sync-as:1
+   B: store %ptr2                                  # sync-as:3
+   X: store atomic release ptr addrspace(0) %ptr3  # sync-as:3
+
+   Thread T2:
+   Y: load atomic acquire ptr addrspace(0) %ptr3   # sync-as:3
+   C: load %ptr2                                   # sync-as:3
+   D: load %ptr1                                   # sync-as:1
+
+In the above figure, ``X`` and ``Y`` are atomic operations on a
+location in the ``global``  address space. If ``X`` synchronizes with
+``Y``, then ``B`` happens-before ``C`` in the ``local`` address
+space. But no such statement can be made about operations ``A`` and
+``D``, although they are peformed on a location in the ``global``
+address space.
+
+Implementation Example: Adding Address Space Information to Fences
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+Languages such as OpenCL C provide fence operations such as
+``atomic_work_item_fence`` that can take an explicit address
+space to fence.
+
+By default, LLVM has no means to carry that information in the IR, so
+the information is lost during lowering to LLVM IR. This means that
+targets such as AMDGPU have to conservatively emit instructions to
+fence all address spaces in all cases, which can have a noticeable
+performance impact in high-performance applications.
+
+MMRAs may be used to preserve that information at the IR level, all the
+way through code generation. For example, a fence that only affects the
+global address space ``addrspace(1)`` may be lowered as
+
+.. code-block::
+
+    fence release # sync-as:1
+
+and the target may use the presence of ``sync-as:1`` to infer that it
+must only emit instruction to fence the global address space.
+
+Note that as MMRAs are opt in, a fence that does not have MMRA metadata
+could still be lowered conservatively, so this optimization would only
+apply if the front-end emits the MMRA metadata on the fence instructions.
+
+Additional Topics
+=================
+
+.. note::
+
+  The following sections are informational.
+
+Performance Impact
+------------------
+
+MMRAs are a way to capture optimization opportunities in the program.
+But when an operation mentions no tags or conflicting tags,
+the target may need to produce conservative code to ensure correctness
+at the cost of performance. This can happen in the following situations:
+
+1. When a target first introduces MMRAs, the
+   frontend might not have been updated to emit them.
+2. An optimization may drop MMRA metadata.
+3. An optimization may add arbitrary tags to an operation.
+
+Note that targets can always choose to ignore (or even drop) MMRAs
+and revert to the default behavior/codegen heuristics without
+affecting correctness.
+
+Consequences of the Absence of *happens-before*
+-----------------------------------------------
+
+In the :ref:`happens-before<HappensBefore>` section, we defined how an
+*happens-before* relation between two instruction can be broken
+by leveraging compatibility between MMRAs. When the instructions
+are incompatible and there is no *happens-before* relation, we say
+that the instructions "do not have to be ordered relative to each
+other".
+
+"Ordering" in this context is a very broad term which covers both
+static and runtime aspects.
+
+When there is no ordering constraint, we *could* statically reorder
+the instructions in an optimizer transform if the reordering does
+not break other constraints as single location coherence.
+Static reordering is one consequence of breaking *happens-before*,
+but is not the most interesting one.
+
+Run-time consequences are more interesting. When there is an
+*happens-before* relation between instructions, the target has to emit
+synchronization code to ensure other threads will observe the effects of
+the instructions in the right order.
+
+For instance, the target may have to wait for previous loads & stores to
+finish before starting a fence-release, or there may be a need to flush a
+memory cache before executing the next instruction.
+In the absence of *happens-before*, there is no such requirement and
+no waiting or flushing is required. This may noticeably speed up
+execution in some cases.
+
+Combining Operations
+--------------------
+
+If a pass can combine multiple memory or synchronizing operations
+into one, it needs to be able to combine MMRAs. One possible way to
+achieve this is by doing a prefix-wise union of the tag sets.
+
+Let A and B be two tags set, and U be the prefix-wise union of A and B.
+For every unique tag prefix P present in A or B:
+
+* If either A or B has no tags with prefix P, no tags with prefix
+  P are added to U.
+* If both A and B have at least one tag with prefix P, all tags with prefix
+  P from both sets are added to U.
+
+Passes should avoid aggressively combining MMRAs, as this can result
+in significant losses of information. While this cannot affect
+correctness, it may affect performance.
+
+As a general rule of thumb, common passes such as SimplifyCFG that
+aggressively combine/reorder operations should only combine
+instructions that have identical sets of tags.
+Passes that combine less frequently, or that are well aware of the cost
+of combining the MMRAs can use the prefix-wise union described above.
+
+Examples:
+
+.. code-block::
+
+    A: store release %ptr1  # foo:x, foo:y, bar:x
+    B: store release %ptr2  # foo:x, bar:y
+
+    # Unique prefixes P = [foo, bar]
+    # "foo:x" is common to A and B so it's added to U.
+    # "bar:x" != "bar:y" so it's not added to U.
+    U: store release %ptr3  # foo:x
+
+.. code-block::
+
+    A: store release %ptr1  # foo:x, foo:y
+    B: store release %ptr2  # foo:x, bux:y
+
+    # Unique prefixes P = [foo, bux]
+    # "foo:x" is common to A and B so it's added to U.
+    # No tags have the prefix "bux" in A.
+    U: store release %ptr3  # foo:x
+
+.. code-block::
+
+    A: store release %ptr1
+    B: store release %ptr2  # foo:x, bar:y
+
+    # Unique prefixes P = [foo, bar]
+    # No tags with "foo" or "bar" in A, so no tags added.
+    U: store release %ptr3
diff --git a/llvm/docs/Reference.rst b/llvm/docs/Reference.rst
index 3a1d1665be43..1661c8c533db 100644
--- a/llvm/docs/Reference.rst
+++ b/llvm/docs/Reference.rst
@@ -39,6 +39,7 @@ LLVM and API reference documentation.
    PDB/index

    PointerAuth

    ScudoHardenedAllocator

+   MemoryModelRelaxationAnnotations

    MemTagSanitizer

    Security

    SecurityTransparencyReports

@@ -194,6 +195,9 @@ Additional Topics
 :doc:`ScudoHardenedAllocator`

   A library that implements a security-hardened `malloc()`.

 

+:doc:`MemoryModelRelaxationAnnotations`

+  Target-defined relaxation to LLVM's concurrency model.

+

 :doc:`MemTagSanitizer`

   Security hardening for production code aiming to mitigate memory

   related vulnerabilities. Based on the Armv8.5-A Memory Tagging Extension.

diff --git a/llvm/docs/ReleaseNotes.rst b/llvm/docs/ReleaseNotes.rst
index 580dc512d969..46d79d6c5822 100644
--- a/llvm/docs/ReleaseNotes.rst
+++ b/llvm/docs/ReleaseNotes.rst
@@ -50,6 +50,12 @@ Update on required toolchains to build LLVM
 Changes to the LLVM IR
 ----------------------
 
+* Added Memory Model Relaxation Annotations (MMRAs).
+* Renamed ``llvm.experimental.vector.reverse`` intrinsic to ``llvm.vector.reverse``.
+* Renamed ``llvm.experimental.vector.splice`` intrinsic to ``llvm.vector.splice``.
+* Renamed ``llvm.experimental.vector.interleave2`` intrinsic to ``llvm.vector.interleave2``.
+* Renamed ``llvm.experimental.vector.deinterleave2`` intrinsic to ``llvm.vector.deinterleave2``.
+
 Changes to LLVM infrastructure
 ------------------------------
 
@@ -67,7 +73,8 @@ Changes to Interprocedural Optimizations
 Changes to the AArch64 Backend
 ------------------------------
 
-* Added support for Cortex-A78AE, Cortex-A520AE and Cortex-A720AE CPUs.
+* Added support for Cortex-A78AE, Cortex-A520AE, Cortex-A720AE,
+  Neoverse-N3, Neoverse-V3 and Neoverse-V3AE CPUs.
 
 Changes to the AMDGPU Backend
 -----------------------------
@@ -110,6 +117,8 @@ Changes to the RISC-V Backend
 * The experimental Ssqosid extension is supported.
 * Zacas is no longer experimental.
 * Added the CSR names from the Resumable Non-Maskable Interrupts (Smrnmi) extension.
+* llvm-objdump now prints disassembled opcode bytes in groups of 2 or 4 bytes to
+  match GNU objdump. The bytes within the groups are in big endian order.
 
 Changes to the WebAssembly Backend
 ----------------------------------
@@ -133,7 +142,7 @@ Changes to the C API
   functions for accessing the values in a blockaddress constant.
 
 * Added ``LLVMConstStringInContext2`` function, which better matches the C++
-  API by using ``size_t`` for string length. Deprecated ``LLVMConstStringInContext``. 
+  API by using ``size_t`` for string length. Deprecated ``LLVMConstStringInContext``.
 
 * Added the following functions for accessing a function's prefix data:
 
diff --git a/llvm/docs/Security.rst b/llvm/docs/Security.rst
index 7f3493d4563b..9140923e5e8c 100644
--- a/llvm/docs/Security.rst
+++ b/llvm/docs/Security.rst
@@ -53,9 +53,9 @@ username for an individual isn't available, the brackets will be empty.
 * Peter Smith (ARM) [@smithp35]
 * Pietro Albini (Ferrous Systems; Rust) [@pietroalbini]
 * Serge Guelton (Mozilla) [@serge-sans-paille]
-* Sergey Maslov (Intel) [@smaslov-intel]
 * Shayne Hiet-Block (Microsoft) [@GreatKeeper]
 * Tim Penge (Sony) []
+* Will Huhn (Intel) [@wphuhn-intel]
 
 Criteria
 --------
diff --git a/llvm/docs/WritingAnLLVMBackend.rst b/llvm/docs/WritingAnLLVMBackend.rst
index 31ebc6204c98..f1f07e4681d5 100644
--- a/llvm/docs/WritingAnLLVMBackend.rst
+++ b/llvm/docs/WritingAnLLVMBackend.rst
@@ -1503,6 +1503,8 @@ non-v9 SPARC implementations.
   if (TM.getSubtarget<SparcSubtarget>().isV9())
     setOperationAction(ISD::CTPOP, MVT::i32, Legal);
 
+.. _backend-calling-convs:
+
 Calling Conventions
 -------------------
 
diff --git a/llvm/examples/OrcV2Examples/LLJITWithRemoteDebugging/RemoteJITUtils.cpp b/llvm/examples/OrcV2Examples/LLJITWithRemoteDebugging/RemoteJITUtils.cpp
index b11d875c6f2d..7c896ab6d88e 100644
--- a/llvm/examples/OrcV2Examples/LLJITWithRemoteDebugging/RemoteJITUtils.cpp
+++ b/llvm/examples/OrcV2Examples/LLJITWithRemoteDebugging/RemoteJITUtils.cpp
@@ -129,7 +129,7 @@ launchLocalExecutor(StringRef ExecutablePath) {
   close(FromExecutor[WriteEnd]);
 
   auto EPC = SimpleRemoteEPC::Create<FDSimpleRemoteEPCTransport>(
-      std::make_unique<DynamicThreadPoolTaskDispatcher>(),
+      std::make_unique<DynamicThreadPoolTaskDispatcher>(std::nullopt),
       SimpleRemoteEPC::Setup(),
       FromExecutor[ReadEnd], ToExecutor[WriteEnd]);
   if (!EPC)
@@ -201,7 +201,7 @@ connectTCPSocket(StringRef NetworkAddress) {
     return CreateErr(toString(SockFD.takeError()));
 
   return SimpleRemoteEPC::Create<FDSimpleRemoteEPCTransport>(
-      std::make_unique<DynamicThreadPoolTaskDispatcher>(),
+      std::make_unique<DynamicThreadPoolTaskDispatcher>(std::nullopt),
       SimpleRemoteEPC::Setup(), *SockFD);
 }
 
diff --git a/llvm/examples/SpeculativeJIT/SpeculativeJIT.cpp b/llvm/examples/SpeculativeJIT/SpeculativeJIT.cpp
index 0d97d379d227..1659e5c5c8b4 100644
--- a/llvm/examples/SpeculativeJIT/SpeculativeJIT.cpp
+++ b/llvm/examples/SpeculativeJIT/SpeculativeJIT.cpp
@@ -49,7 +49,9 @@ public:
     if (!DL)
       return DL.takeError();
 
-    auto EPC = SelfExecutorProcessControl::Create();
+    auto EPC = SelfExecutorProcessControl::Create(
+        nullptr,
+        std::make_unique<DynamicThreadPoolTaskDispatcher>(std::nullopt));
     if (!EPC)
       return EPC.takeError();
 
@@ -116,14 +118,6 @@ private:
                  std::move(ISMBuilder)) {
     MainJD.addGenerator(std::move(ProcessSymbolsGenerator));
     this->CODLayer.setImplMap(&Imps);
-    this->ES->setDispatchTask(
-        [this](std::unique_ptr<Task> T) {
-          CompileThreads.async(
-              [UnownedT = T.release()]() {
-                std::unique_ptr<Task> T(UnownedT);
-                T->run();
-              });
-        });
     ExitOnErr(S.addSpeculationRuntime(MainJD, Mangle));
     LocalCXXRuntimeOverrides CXXRuntimeoverrides;
     ExitOnErr(CXXRuntimeoverrides.enable(MainJD, Mangle));
diff --git a/llvm/include/llvm-c/DebugInfo.h b/llvm/include/llvm-c/DebugInfo.h
index dab1d697761b..2c3c75e246c0 100644
--- a/llvm/include/llvm-c/DebugInfo.h
+++ b/llvm/include/llvm-c/DebugInfo.h
@@ -125,7 +125,20 @@ typedef enum {
   LLVMDWARFSourceLanguageFortran18,
   LLVMDWARFSourceLanguageAda2005,
   LLVMDWARFSourceLanguageAda2012,
+  LLVMDWARFSourceLanguageHIP,
+  LLVMDWARFSourceLanguageAssembly,
+  LLVMDWARFSourceLanguageC_sharp,
   LLVMDWARFSourceLanguageMojo,
+  LLVMDWARFSourceLanguageGLSL,
+  LLVMDWARFSourceLanguageGLSL_ES,
+  LLVMDWARFSourceLanguageHLSL,
+  LLVMDWARFSourceLanguageOpenCL_CPP,
+  LLVMDWARFSourceLanguageCPP_for_OpenCL,
+  LLVMDWARFSourceLanguageSYCL,
+  LLVMDWARFSourceLanguageRuby,
+  LLVMDWARFSourceLanguageMove,
+  LLVMDWARFSourceLanguageHylo,
+
   // Vendor extensions:
   LLVMDWARFSourceLanguageMips_Assembler,
   LLVMDWARFSourceLanguageGOOGLE_RenderScript,
diff --git a/llvm/include/llvm/ADT/StringRef.h b/llvm/include/llvm/ADT/StringRef.h
index 1c6c96678b5d..04496c76e072 100644
--- a/llvm/include/llvm/ADT/StringRef.h
+++ b/llvm/include/llvm/ADT/StringRef.h
@@ -258,10 +258,8 @@ namespace llvm {
       return Length >= Prefix.Length &&
              compareMemory(Data, Prefix.Data, Prefix.Length) == 0;
     }
-    [[nodiscard]] LLVM_DEPRECATED(
-        "Use starts_with instead",
-        "starts_with") bool startswith(StringRef Prefix) const {
-      return starts_with(Prefix);
+    [[nodiscard]] bool starts_with(char Prefix) const {
+      return !empty() && front() == Prefix;
     }
 
     /// Check if this string starts with the given \p Prefix, ignoring case.
@@ -273,10 +271,8 @@ namespace llvm {
              compareMemory(end() - Suffix.Length, Suffix.Data, Suffix.Length) ==
                  0;
     }
-    [[nodiscard]] LLVM_DEPRECATED(
-        "Use ends_with instead",
-        "ends_with") bool endswith(StringRef Suffix) const {
-      return ends_with(Suffix);
+    [[nodiscard]] bool ends_with(char Suffix) const {
+      return !empty() && back() == Suffix;
     }
 
     /// Check if this string ends with the given \p Suffix, ignoring case.
diff --git a/llvm/include/llvm/Analysis/InstructionSimplify.h b/llvm/include/llvm/Analysis/InstructionSimplify.h
index 03d7ad12c12d..719cffc118f5 100644
--- a/llvm/include/llvm/Analysis/InstructionSimplify.h
+++ b/llvm/include/llvm/Analysis/InstructionSimplify.h
@@ -32,6 +32,7 @@
 #define LLVM_ANALYSIS_INSTRUCTIONSIMPLIFY_H
 
 #include "llvm/Analysis/SimplifyQuery.h"
+#include "llvm/IR/FPEnv.h"
 
 namespace llvm {
 
@@ -44,6 +45,7 @@ class DataLayout;
 class DominatorTree;
 class Function;
 class Instruction;
+class LoadInst;
 struct LoopStandardAnalysisResults;
 class Pass;
 template <class T, unsigned n> class SmallSetVector;
diff --git a/llvm/include/llvm/Analysis/Loads.h b/llvm/include/llvm/Analysis/Loads.h
index 0926093bba99..a8d954b9872d 100644
--- a/llvm/include/llvm/Analysis/Loads.h
+++ b/llvm/include/llvm/Analysis/Loads.h
@@ -173,14 +173,17 @@ Value *findAvailablePtrLoadStore(const MemoryLocation &Loc, Type *AccessTy,
                                  unsigned MaxInstsToScan, BatchAAResults *AA,
                                  bool *IsLoadCSE, unsigned *NumScanedInst);
 
-/// Returns true if a pointer value \p A can be replace with another pointer
-/// value \B if they are deemed equal through some means (e.g. information from
+/// Returns true if a pointer value \p From can be replaced with another pointer
+/// value \To if they are deemed equal through some means (e.g. information from
 /// conditions).
-/// NOTE: the current implementations is incomplete and unsound. It does not
-/// reject all invalid cases yet, but will be made stricter in the future. In
-/// particular this means returning true means unknown if replacement is safe.
-bool canReplacePointersIfEqual(Value *A, Value *B, const DataLayout &DL,
-                               Instruction *CtxI);
+/// NOTE: The current implementation allows replacement in Icmp and PtrToInt
+/// instructions, as well as when we are replacing with a null pointer.
+/// Additionally it also allows replacement of pointers when both pointers have
+/// the same underlying object.
+bool canReplacePointersIfEqual(const Value *From, const Value *To,
+                               const DataLayout &DL);
+bool canReplacePointersInUseIfEqual(const Use &U, const Value *To,
+                                    const DataLayout &DL);
 }
 
 #endif
diff --git a/llvm/include/llvm/Analysis/SimplifyQuery.h b/llvm/include/llvm/Analysis/SimplifyQuery.h
index a10c0dc49fa2..25b8f9b5eaf1 100644
--- a/llvm/include/llvm/Analysis/SimplifyQuery.h
+++ b/llvm/include/llvm/Analysis/SimplifyQuery.h
@@ -9,7 +9,7 @@
 #ifndef LLVM_ANALYSIS_SIMPLIFYQUERY_H
 #define LLVM_ANALYSIS_SIMPLIFYQUERY_H
 
-#include "llvm/IR/PatternMatch.h"
+#include "llvm/IR/Operator.h"
 
 namespace llvm {
 
@@ -106,13 +106,7 @@ struct SimplifyQuery {
 
   /// If CanUseUndef is true, returns whether \p V is undef.
   /// Otherwise always return false.
-  bool isUndefValue(Value *V) const {
-    if (!CanUseUndef)
-      return false;
-
-    using namespace PatternMatch;
-    return match(V, m_Undef());
-  }
+  bool isUndefValue(Value *V) const;
 
   SimplifyQuery getWithoutDomCondCache() const {
     SimplifyQuery Copy(*this);
diff --git a/llvm/include/llvm/Analysis/TargetTransformInfo.h b/llvm/include/llvm/Analysis/TargetTransformInfo.h
index 58c69ac93976..1c76821fe5e4 100644
--- a/llvm/include/llvm/Analysis/TargetTransformInfo.h
+++ b/llvm/include/llvm/Analysis/TargetTransformInfo.h
@@ -1267,7 +1267,7 @@ public:
       TTI::TargetCostKind CostKind = TTI::TCK_RecipThroughput,
       TTI::OperandValueInfo Opd1Info = {TTI::OK_AnyValue, TTI::OP_None},
       TTI::OperandValueInfo Opd2Info = {TTI::OK_AnyValue, TTI::OP_None},
-      ArrayRef<const Value *> Args = ArrayRef<const Value *>(),
+      ArrayRef<const Value *> Args = std::nullopt,
       const Instruction *CxtI = nullptr,
       const TargetLibraryInfo *TLibInfo = nullptr) const;
 
diff --git a/llvm/include/llvm/Analysis/ValueTracking.h b/llvm/include/llvm/Analysis/ValueTracking.h
index bab7c8868532..afd18e7e56ba 100644
--- a/llvm/include/llvm/Analysis/ValueTracking.h
+++ b/llvm/include/llvm/Analysis/ValueTracking.h
@@ -20,6 +20,7 @@
 #include "llvm/IR/Constants.h"
 #include "llvm/IR/DataLayout.h"
 #include "llvm/IR/FMF.h"
+#include "llvm/IR/Instructions.h"
 #include "llvm/IR/InstrTypes.h"
 #include "llvm/IR/Intrinsics.h"
 #include <cassert>
@@ -130,7 +131,8 @@ bool isKnownNonZero(const Value *V, const SimplifyQuery &Q, unsigned Depth = 0);
 /// Currently can recoginze Value pair:
 /// 1: <X, Y> if X = sub (0, Y) or Y = sub (0, X)
 /// 2: <X, Y> if X = sub (A, B) and Y = sub (B, A)
-bool isKnownNegation(const Value *X, const Value *Y, bool NeedNSW = false);
+bool isKnownNegation(const Value *X, const Value *Y, bool NeedNSW = false,
+                     bool AllowPoison = true);
 
 /// Returns true if the give value is known to be non-negative.
 bool isKnownNonNegative(const Value *V, const SimplifyQuery &SQ,
@@ -858,7 +860,8 @@ enum class OverflowResult {
 };
 
 OverflowResult computeOverflowForUnsignedMul(const Value *LHS, const Value *RHS,
-                                             const SimplifyQuery &SQ);
+                                             const SimplifyQuery &SQ,
+                                             bool IsNSW = false);
 OverflowResult computeOverflowForSignedMul(const Value *LHS, const Value *RHS,
                                            const SimplifyQuery &SQ);
 OverflowResult
diff --git a/llvm/include/llvm/Analysis/VectorUtils.h b/llvm/include/llvm/Analysis/VectorUtils.h
index c6eb66cc9660..424b73e375b5 100644
--- a/llvm/include/llvm/Analysis/VectorUtils.h
+++ b/llvm/include/llvm/Analysis/VectorUtils.h
@@ -301,7 +301,7 @@ MDNode *intersectAccessGroups(const Instruction *Inst1,
                               const Instruction *Inst2);
 
 /// Specifically, let Kinds = [MD_tbaa, MD_alias_scope, MD_noalias, MD_fpmath,
-/// MD_nontemporal, MD_access_group].
+/// MD_nontemporal, MD_access_group, MD_mmra].
 /// For K in Kinds, we get the MDNode for K from each of the
 /// elements of VL, compute their "intersection" (i.e., the most generic
 /// metadata value that covers all of the individual values), and set I's
diff --git a/llvm/include/llvm/BinaryFormat/Dwarf.def b/llvm/include/llvm/BinaryFormat/Dwarf.def
index 460a9264536b..adcf24eb83b0 100644
--- a/llvm/include/llvm/BinaryFormat/Dwarf.def
+++ b/llvm/include/llvm/BinaryFormat/Dwarf.def
@@ -11,19 +11,20 @@
 //===----------------------------------------------------------------------===//
 
 // TODO: Add other DW-based macros.
-#if !(                                                                         \
-    defined HANDLE_DW_TAG || defined HANDLE_DW_AT || defined HANDLE_DW_FORM || \
-    defined HANDLE_DW_OP || defined HANDLE_DW_OP_LLVM_USEROP ||                    \
-    defined HANDLE_DW_LANG || defined HANDLE_DW_ATE ||                         \
-    defined HANDLE_DW_VIRTUALITY || defined HANDLE_DW_DEFAULTED ||             \
-    defined HANDLE_DW_CC || defined HANDLE_DW_LNS || defined HANDLE_DW_LNE ||  \
-    defined HANDLE_DW_LNCT || defined HANDLE_DW_MACRO ||                       \
-    defined HANDLE_DW_MACRO_GNU || defined HANDLE_MACRO_FLAG ||                \
-    defined HANDLE_DW_RLE || defined HANDLE_DW_LLE ||                          \
-    (defined HANDLE_DW_CFA && defined HANDLE_DW_CFA_PRED) ||                   \
-    defined HANDLE_DW_APPLE_PROPERTY || defined HANDLE_DW_UT ||                \
-    defined HANDLE_DWARF_SECTION || defined HANDLE_DW_IDX ||                   \
-    defined HANDLE_DW_END || defined HANDLE_DW_SECT)
+#if !(defined HANDLE_DW_TAG || defined HANDLE_DW_AT ||                         \
+      defined HANDLE_DW_FORM || defined HANDLE_DW_OP ||                        \
+      defined HANDLE_DW_OP_LLVM_USEROP || defined HANDLE_DW_LANG ||            \
+      defined HANDLE_DW_LNAME || defined HANDLE_DW_ATE ||                      \
+      defined HANDLE_DW_VIRTUALITY || defined HANDLE_DW_DEFAULTED ||           \
+      defined HANDLE_DW_CC || defined HANDLE_DW_LNS ||                         \
+      defined HANDLE_DW_LNE || defined HANDLE_DW_LNCT ||                       \
+      defined HANDLE_DW_MACRO || defined HANDLE_DW_MACRO_GNU ||                \
+      defined HANDLE_MACRO_FLAG || defined HANDLE_DW_RLE ||                    \
+      defined HANDLE_DW_LLE ||                                                 \
+      (defined HANDLE_DW_CFA && defined HANDLE_DW_CFA_PRED) ||                 \
+      defined HANDLE_DW_APPLE_PROPERTY || defined HANDLE_DW_UT ||              \
+      defined HANDLE_DWARF_SECTION || defined HANDLE_DW_IDX ||                 \
+      defined HANDLE_DW_END || defined HANDLE_DW_SECT)
 #error "Missing macro definition of HANDLE_DW*"
 #endif
 
@@ -61,6 +62,10 @@
 #define HANDLE_DW_LANG(ID, NAME, LOWER_BOUND, VERSION, VENDOR)
 #endif
 
+#ifndef HANDLE_DW_LNAME
+#define HANDLE_DW_LNAME(ID, NAME, DESC, LOWER_BOUND)
+#endif
+
 #ifndef HANDLE_DW_ATE
 #define HANDLE_DW_ATE(ID, NAME, VERSION, VENDOR)
 #endif
@@ -950,12 +955,81 @@ HANDLE_DW_LANG(0x002c, C17, 0, 0, DWARF)
 HANDLE_DW_LANG(0x002d, Fortran18, 0, 0, DWARF)
 HANDLE_DW_LANG(0x002e, Ada2005, 0, 0, DWARF)
 HANDLE_DW_LANG(0x002f, Ada2012, 0, 0, DWARF)
+HANDLE_DW_LANG(0x0030, HIP, 0, 0, DWARF)
+HANDLE_DW_LANG(0x0031, Assembly, 0, 0, DWARF)
+HANDLE_DW_LANG(0x0032, C_sharp, 0, 0, DWARF)
 HANDLE_DW_LANG(0x0033, Mojo, 0, 0, DWARF)
+HANDLE_DW_LANG(0x0034, GLSL, 0, 0, DWARF)
+HANDLE_DW_LANG(0x0035, GLSL_ES, 0, 0, DWARF)
+HANDLE_DW_LANG(0x0036, HLSL, 0, 0, DWARF)
+HANDLE_DW_LANG(0x0037, OpenCL_CPP, 0, 0, DWARF)
+HANDLE_DW_LANG(0x0038, CPP_for_OpenCL, 0, 0, DWARF)
+HANDLE_DW_LANG(0x0039, SYCL, 0, 0, DWARF)
+HANDLE_DW_LANG(0x0040, Ruby, 0, 0, DWARF)
+HANDLE_DW_LANG(0x0041, Move, 0, 0, DWARF)
+HANDLE_DW_LANG(0x0042, Hylo, 0, 0, DWARF)
+
 // Vendor extensions:
 HANDLE_DW_LANG(0x8001, Mips_Assembler, std::nullopt, 0, MIPS)
 HANDLE_DW_LANG(0x8e57, GOOGLE_RenderScript, 0, 0, GOOGLE)
 HANDLE_DW_LANG(0xb000, BORLAND_Delphi, 0, 0, BORLAND)
 
+// Tentative DWARF 6 language codes. This list is subject to change.
+HANDLE_DW_LNAME(0x0001, Ada, "ISO Ada", 1) // YYYY
+HANDLE_DW_LNAME(0x0002, BLISS, "BLISS", 0)
+//     YYYYMM
+// K&R 000000
+// C89 198912
+// C99 199901
+// C11 201112
+// C17 201710
+// C23 202311
+HANDLE_DW_LNAME(0x0003, C, "C (K&R and ISO)", 0)
+//       YYYYMM
+// C++98 199711
+// C++03 200310
+// C++11 201103
+// C++14 201402
+// C++17 201703
+// C++20 202002
+HANDLE_DW_LNAME(0x0004, C_plus_plus, "ISO C++", 0)
+HANDLE_DW_LNAME(0x0005, Cobol, "ISO Cobol", 1) // YYYY
+HANDLE_DW_LNAME(0x0006, Crystal, "Crystal", 0)
+HANDLE_DW_LNAME(0x0007, D, "D", 0)
+HANDLE_DW_LNAME(0x0008, Dylan, "Dylan", 0)
+HANDLE_DW_LNAME(0x0009, Fortran, "ISO Fortran", 1) // YYYY
+HANDLE_DW_LNAME(0x000a, Go, "Go", 0)
+HANDLE_DW_LNAME(0x000b, Haskell, "Haskell", 0)
+HANDLE_DW_LNAME(0x000c, Java, "Java", 0)
+HANDLE_DW_LNAME(0x000d, Julia, "Julia", 1)
+HANDLE_DW_LNAME(0x000e, Kotlin, "Kotlin", 0)
+HANDLE_DW_LNAME(0x000f, Modula2, "Modula 2", 1)
+HANDLE_DW_LNAME(0x0010, Modula3, "Modula 3", 1)
+HANDLE_DW_LNAME(0x0011, ObjC, "Objective C", 0) // YYYYMM
+HANDLE_DW_LNAME(0x0012, ObjC_plus_plus, "Objective C++", 0) // YYYYMM
+HANDLE_DW_LNAME(0x0013, OCaml, "OCaml", 0)
+HANDLE_DW_LNAME(0x0014, OpenCL_C, "OpenCL C", 0)
+HANDLE_DW_LNAME(0x0015, Pascal, "ISO Pascal", 1) // YYYY
+HANDLE_DW_LNAME(0x0016, PLI, "ANSI PL/I", 1)
+HANDLE_DW_LNAME(0x0017, Python, "Python", 0)
+HANDLE_DW_LNAME(0x0018, RenderScript, "RenderScript Kernel Language", 0)
+HANDLE_DW_LNAME(0x0019, Rust, "Rust", 0)
+HANDLE_DW_LNAME(0x001a, Swift, "Swift", 0) // VVMM
+HANDLE_DW_LNAME(0x001b, UPC, "Unified Parallel C (UPC)", 0)
+HANDLE_DW_LNAME(0x001c, Zig, "Zig", 0)
+HANDLE_DW_LNAME(0x001d, Assembly, "Assembly", 0)
+// Conflict: HANDLE_DW_LNAME(0x001d, HIP, "HIP", 0)
+HANDLE_DW_LNAME(0x001e, C_sharp, "C#", 0)
+HANDLE_DW_LNAME(0x001f, Mojo, "Mojo", 0)
+HANDLE_DW_LNAME(0x0020, GLSL, "OpenGL Shading Language", 0) // VVMMPP
+HANDLE_DW_LNAME(0x0021, GLSL_ES, "OpenGL ES Shading Language", 0) // VVMMPP
+HANDLE_DW_LNAME(0x0022, HLSL, "High Level Shading Language", 0) // YYYY
+HANDLE_DW_LNAME(0x0023, OpenCL_CPP, "OpenCL C++", 0) // VVMM
+HANDLE_DW_LNAME(0x0024, CPP_for_OpenCL, "C++ for OpenCL", 0) // VVMM
+HANDLE_DW_LNAME(0x0025, SYCL, "SYCL", 0) // YYYYRR
+HANDLE_DW_LNAME(0x0026, Ruby, "Ruby", 0) // VVMMPP
+HANDLE_DW_LNAME(0x0027, Move, "Move", 0) // YYYYMM
+HANDLE_DW_LNAME(0x0028, Hylo, "Hylo", 0)
 
 // DWARF attribute type encodings.
 HANDLE_DW_ATE(0x01, address, 2, DWARF)
@@ -1267,6 +1341,7 @@ HANDLE_DW_SECT(8, RNGLISTS)
 #undef HANDLE_DW_OP
 #undef HANDLE_DW_OP_LLVM_USEROP
 #undef HANDLE_DW_LANG
+#undef HANDLE_DW_LNAME
 #undef HANDLE_DW_ATE
 #undef HANDLE_DW_VIRTUALITY
 #undef HANDLE_DW_DEFAULTED
diff --git a/llvm/include/llvm/BinaryFormat/Dwarf.h b/llvm/include/llvm/BinaryFormat/Dwarf.h
index 298700c8941e..74c4d6ff3a71 100644
--- a/llvm/include/llvm/BinaryFormat/Dwarf.h
+++ b/llvm/include/llvm/BinaryFormat/Dwarf.h
@@ -209,6 +209,284 @@ enum SourceLanguage {
   DW_LANG_hi_user = 0xffff
 };
 
+enum SourceLanguageName : uint16_t {
+#define HANDLE_DW_LNAME(ID, NAME, DESC, LOWER_BOUND) DW_LNAME_##NAME = ID,
+#include "llvm/BinaryFormat/Dwarf.def"
+};
+
+/// Convert a DWARF 6 pair of language name and version to a DWARF 5 DW_LANG.
+/// If the version number doesn't exactly match a known version it is
+/// rounded up to the next-highest known version number.
+inline std::optional<SourceLanguage> toDW_LANG(SourceLanguageName name,
+                                               uint32_t version) {
+  switch (name) {
+  case DW_LNAME_Ada: // YYYY
+    if (version <= 1983)
+      return DW_LANG_Ada83;
+    if (version <= 1995)
+      return DW_LANG_Ada95;
+    if (version <= 2005)
+      return DW_LANG_Ada2005;
+    if (version <= 2012)
+      return DW_LANG_Ada2012;
+    return {};
+  case DW_LNAME_BLISS:
+    return DW_LANG_BLISS;
+  case DW_LNAME_C: // YYYYMM, K&R 000000
+    if (version == 0)
+      return DW_LANG_C;
+    if (version <= 198912)
+      return DW_LANG_C89;
+    if (version <= 199901)
+      return DW_LANG_C99;
+    if (version <= 201112)
+      return DW_LANG_C11;
+    if (version <= 201710)
+      return DW_LANG_C17;
+    return {};
+  case DW_LNAME_C_plus_plus: // YYYYMM
+    if (version == 0)
+      return DW_LANG_C_plus_plus;
+    if (version <= 199711)
+      return DW_LANG_C_plus_plus;
+    if (version <= 200310)
+      return DW_LANG_C_plus_plus_03;
+    if (version <= 201103)
+      return DW_LANG_C_plus_plus_11;
+    if (version <= 201402)
+      return DW_LANG_C_plus_plus_14;
+    if (version <= 201703)
+      return DW_LANG_C_plus_plus_17;
+    if (version <= 202002)
+      return DW_LANG_C_plus_plus_20;
+    return {};
+  case DW_LNAME_Cobol: // YYYY
+    if (version <= 1974)
+      return DW_LANG_Cobol74;
+    if (version <= 1985)
+      return DW_LANG_Cobol85;
+    return {};
+  case DW_LNAME_Crystal:
+    return DW_LANG_Crystal;
+  case DW_LNAME_D:
+    return DW_LANG_D;
+  case DW_LNAME_Dylan:
+    return DW_LANG_Dylan;
+  case DW_LNAME_Fortran: // YYYY
+    if (version <= 1977)
+      return DW_LANG_Fortran77;
+    if (version <= 1990)
+      return DW_LANG_Fortran90;
+    if (version <= 1995)
+      return DW_LANG_Fortran95;
+    if (version <= 2003)
+      return DW_LANG_Fortran03;
+    if (version <= 2008)
+      return DW_LANG_Fortran08;
+    if (version <= 2018)
+      return DW_LANG_Fortran18;
+    return {};
+  case DW_LNAME_Go:
+    return DW_LANG_Go;
+  case DW_LNAME_Haskell:
+    return DW_LANG_Haskell;
+  // case DW_LNAME_HIP:
+  //   return DW_LANG_HIP;
+  case DW_LNAME_Java:
+    return DW_LANG_Java;
+  case DW_LNAME_Julia:
+    return DW_LANG_Julia;
+  case DW_LNAME_Kotlin:
+    return DW_LANG_Kotlin;
+  case DW_LNAME_Modula2:
+    return DW_LANG_Modula2;
+  case DW_LNAME_Modula3:
+    return DW_LANG_Modula3;
+  case DW_LNAME_ObjC:
+    return DW_LANG_ObjC;
+  case DW_LNAME_ObjC_plus_plus:
+    return DW_LANG_ObjC_plus_plus;
+  case DW_LNAME_OCaml:
+    return DW_LANG_OCaml;
+  case DW_LNAME_OpenCL_C:
+    return DW_LANG_OpenCL;
+  case DW_LNAME_Pascal:
+    return DW_LANG_Pascal83;
+  case DW_LNAME_PLI:
+    return DW_LANG_PLI;
+  case DW_LNAME_Python:
+    return DW_LANG_Python;
+  case DW_LNAME_RenderScript:
+    return DW_LANG_RenderScript;
+  case DW_LNAME_Rust:
+    return DW_LANG_Rust;
+  case DW_LNAME_Swift:
+    return DW_LANG_Swift;
+  case DW_LNAME_UPC:
+    return DW_LANG_UPC;
+  case DW_LNAME_Zig:
+    return DW_LANG_Zig;
+  case DW_LNAME_Assembly:
+    return DW_LANG_Assembly;
+  case DW_LNAME_C_sharp:
+    return DW_LANG_C_sharp;
+  case DW_LNAME_Mojo:
+    return DW_LANG_Mojo;
+  case DW_LNAME_GLSL:
+    return DW_LANG_GLSL;
+  case DW_LNAME_GLSL_ES:
+    return DW_LANG_GLSL_ES;
+  case DW_LNAME_HLSL:
+    return DW_LANG_HLSL;
+  case DW_LNAME_OpenCL_CPP:
+    return DW_LANG_OpenCL_CPP;
+  case DW_LNAME_CPP_for_OpenCL:
+    return {};
+  case DW_LNAME_SYCL:
+    return DW_LANG_SYCL;
+  case DW_LNAME_Ruby:
+    return DW_LANG_Ruby;
+  case DW_LNAME_Move:
+    return DW_LANG_Move;
+  case DW_LNAME_Hylo:
+    return DW_LANG_Hylo;
+  }
+  return {};
+}
+
+/// Convert a DWARF 5 DW_LANG to a DWARF 6 pair of language name and version.
+inline std::optional<std::pair<SourceLanguageName, uint32_t>>
+toDW_LNAME(SourceLanguage language) {
+  switch (language) {
+  case DW_LANG_Ada83:
+    return {{DW_LNAME_Ada, 1983}};
+  case DW_LANG_Ada95:
+    return {{DW_LNAME_Ada, 1995}};
+  case DW_LANG_Ada2005:
+    return {{DW_LNAME_Ada, 2005}};
+  case DW_LANG_Ada2012:
+    return {{DW_LNAME_Ada, 2012}};
+  case DW_LANG_BLISS:
+    return {{DW_LNAME_BLISS, 0}};
+  case DW_LANG_C:
+    return {{DW_LNAME_C, 0}};
+  case DW_LANG_C89:
+    return {{DW_LNAME_C, 198912}};
+  case DW_LANG_C99:
+    return {{DW_LNAME_C, 199901}};
+  case DW_LANG_C11:
+    return {{DW_LNAME_C, 201112}};
+  case DW_LANG_C17:
+    return {{DW_LNAME_C, 201712}};
+  case DW_LANG_C_plus_plus:
+    return {{DW_LNAME_C_plus_plus, 0}};
+  case DW_LANG_C_plus_plus_03:
+    return {{DW_LNAME_C_plus_plus, 200310}};
+  case DW_LANG_C_plus_plus_11:
+    return {{DW_LNAME_C_plus_plus, 201103}};
+  case DW_LANG_C_plus_plus_14:
+    return {{DW_LNAME_C_plus_plus, 201402}};
+  case DW_LANG_C_plus_plus_17:
+    return {{DW_LNAME_C_plus_plus, 201703}};
+  case DW_LANG_C_plus_plus_20:
+    return {{DW_LNAME_C_plus_plus, 202002}};
+  case DW_LANG_Cobol74:
+    return {{DW_LNAME_Cobol, 1974}};
+  case DW_LANG_Cobol85:
+    return {{DW_LNAME_Cobol, 1985}};
+  case DW_LANG_Crystal:
+    return {{DW_LNAME_Crystal, 0}};
+  case DW_LANG_D:
+    return {{DW_LNAME_D, 0}};
+  case DW_LANG_Dylan:
+    return {{DW_LNAME_Dylan, 0}};
+  case DW_LANG_Fortran77:
+    return {{DW_LNAME_Fortran, 1977}};
+  case DW_LANG_Fortran90:
+    return {{DW_LNAME_Fortran, 1990}};
+  case DW_LANG_Fortran95:
+    return {{DW_LNAME_Fortran, 1995}};
+  case DW_LANG_Fortran03:
+    return {{DW_LNAME_Fortran, 2003}};
+  case DW_LANG_Fortran08:
+    return {{DW_LNAME_Fortran, 2008}};
+  case DW_LANG_Fortran18:
+    return {{DW_LNAME_Fortran, 2018}};
+  case DW_LANG_Go:
+    return {{DW_LNAME_Go, 0}};
+  case DW_LANG_Haskell:
+    return {{DW_LNAME_Haskell, 0}};
+  case DW_LANG_HIP:
+    return {}; // return {{DW_LNAME_HIP, 0}};
+  case DW_LANG_Java:
+    return {{DW_LNAME_Java, 0}};
+  case DW_LANG_Julia:
+    return {{DW_LNAME_Julia, 0}};
+  case DW_LANG_Kotlin:
+    return {{DW_LNAME_Kotlin, 0}};
+  case DW_LANG_Modula2:
+    return {{DW_LNAME_Modula2, 0}};
+  case DW_LANG_Modula3:
+    return {{DW_LNAME_Modula3, 0}};
+  case DW_LANG_ObjC:
+    return {{DW_LNAME_ObjC, 0}};
+  case DW_LANG_ObjC_plus_plus:
+    return {{DW_LNAME_ObjC_plus_plus, 0}};
+  case DW_LANG_OCaml:
+    return {{DW_LNAME_OCaml, 0}};
+  case DW_LANG_OpenCL:
+    return {{DW_LNAME_OpenCL_C, 0}};
+  case DW_LANG_Pascal83:
+    return {{DW_LNAME_Pascal, 1983}};
+  case DW_LANG_PLI:
+    return {{DW_LNAME_PLI, 0}};
+  case DW_LANG_Python:
+    return {{DW_LNAME_Python, 0}};
+  case DW_LANG_RenderScript:
+  case DW_LANG_GOOGLE_RenderScript:
+    return {{DW_LNAME_RenderScript, 0}};
+  case DW_LANG_Rust:
+    return {{DW_LNAME_Rust, 0}};
+  case DW_LANG_Swift:
+    return {{DW_LNAME_Swift, 0}};
+  case DW_LANG_UPC:
+    return {{DW_LNAME_UPC, 0}};
+  case DW_LANG_Zig:
+    return {{DW_LNAME_Zig, 0}};
+  case DW_LANG_Assembly:
+  case DW_LANG_Mips_Assembler:
+    return {{DW_LNAME_Assembly, 0}};
+  case DW_LANG_C_sharp:
+    return {{DW_LNAME_C_sharp, 0}};
+  case DW_LANG_Mojo:
+    return {{DW_LNAME_Mojo, 0}};
+  case DW_LANG_GLSL:
+    return {{DW_LNAME_GLSL, 0}};
+  case DW_LANG_GLSL_ES:
+    return {{DW_LNAME_GLSL_ES, 0}};
+  case DW_LANG_HLSL:
+    return {{DW_LNAME_HLSL, 0}};
+  case DW_LANG_OpenCL_CPP:
+    return {{DW_LNAME_OpenCL_CPP, 0}};
+  case DW_LANG_SYCL:
+    return {{DW_LNAME_SYCL, 0}};
+  case DW_LANG_Ruby:
+    return {{DW_LNAME_Ruby, 0}};
+  case DW_LANG_Move:
+    return {{DW_LNAME_Move, 0}};
+  case DW_LANG_Hylo:
+    return {{DW_LNAME_Hylo, 0}};
+  case DW_LANG_BORLAND_Delphi:
+  case DW_LANG_CPP_for_OpenCL:
+  case DW_LANG_lo_user:
+  case DW_LANG_hi_user:
+    return {};
+  }
+  return {};
+}
+
+llvm::StringRef LanguageDescription(SourceLanguageName name);
+
 inline bool isCPlusPlus(SourceLanguage S) {
   bool result = false;
   // Deliberately enumerate all the language options so we get a warning when
@@ -268,7 +546,19 @@ inline bool isCPlusPlus(SourceLanguage S) {
   case DW_LANG_Fortran18:
   case DW_LANG_Ada2005:
   case DW_LANG_Ada2012:
+  case DW_LANG_HIP:
+  case DW_LANG_Assembly:
+  case DW_LANG_C_sharp:
   case DW_LANG_Mojo:
+  case DW_LANG_GLSL:
+  case DW_LANG_GLSL_ES:
+  case DW_LANG_HLSL:
+  case DW_LANG_OpenCL_CPP:
+  case DW_LANG_CPP_for_OpenCL:
+  case DW_LANG_SYCL:
+  case DW_LANG_Ruby:
+  case DW_LANG_Move:
+  case DW_LANG_Hylo:
     result = false;
     break;
   }
@@ -335,7 +625,19 @@ inline bool isFortran(SourceLanguage S) {
   case DW_LANG_C17:
   case DW_LANG_Ada2005:
   case DW_LANG_Ada2012:
+  case DW_LANG_HIP:
+  case DW_LANG_Assembly:
+  case DW_LANG_C_sharp:
   case DW_LANG_Mojo:
+  case DW_LANG_GLSL:
+  case DW_LANG_GLSL_ES:
+  case DW_LANG_HLSL:
+  case DW_LANG_OpenCL_CPP:
+  case DW_LANG_CPP_for_OpenCL:
+  case DW_LANG_SYCL:
+  case DW_LANG_Ruby:
+  case DW_LANG_Move:
+  case DW_LANG_Hylo:
     result = false;
     break;
   }
@@ -400,7 +702,19 @@ inline bool isC(SourceLanguage S) {
   case DW_LANG_Fortran18:
   case DW_LANG_Ada2005:
   case DW_LANG_Ada2012:
+  case DW_LANG_HIP:
+  case DW_LANG_Assembly:
+  case DW_LANG_C_sharp:
   case DW_LANG_Mojo:
+  case DW_LANG_GLSL:
+  case DW_LANG_GLSL_ES:
+  case DW_LANG_HLSL:
+  case DW_LANG_OpenCL_CPP:
+  case DW_LANG_CPP_for_OpenCL:
+  case DW_LANG_SYCL:
+  case DW_LANG_Ruby:
+  case DW_LANG_Move:
+  case DW_LANG_Hylo:
     return false;
   }
   llvm_unreachable("Unknown language kind.");
diff --git a/llvm/include/llvm/CodeGen/BasicTTIImpl.h b/llvm/include/llvm/CodeGen/BasicTTIImpl.h
index 06a19c75cf87..92b51438b4cb 100644
--- a/llvm/include/llvm/CodeGen/BasicTTIImpl.h
+++ b/llvm/include/llvm/CodeGen/BasicTTIImpl.h
@@ -892,7 +892,7 @@ public:
       unsigned Opcode, Type *Ty, TTI::TargetCostKind CostKind,
       TTI::OperandValueInfo Opd1Info = {TTI::OK_AnyValue, TTI::OP_None},
       TTI::OperandValueInfo Opd2Info = {TTI::OK_AnyValue, TTI::OP_None},
-      ArrayRef<const Value *> Args = ArrayRef<const Value *>(),
+      ArrayRef<const Value *> Args = std::nullopt,
       const Instruction *CxtI = nullptr) {
     // Check if any of the operands are vector operands.
     const TargetLoweringBase *TLI = getTLI();
@@ -1662,12 +1662,12 @@ public:
           TTI::SK_InsertSubvector, cast<VectorType>(Args[0]->getType()),
           std::nullopt, CostKind, Index, cast<VectorType>(Args[1]->getType()));
     }
-    case Intrinsic::experimental_vector_reverse: {
+    case Intrinsic::vector_reverse: {
       return thisT()->getShuffleCost(
           TTI::SK_Reverse, cast<VectorType>(Args[0]->getType()), std::nullopt,
           CostKind, 0, cast<VectorType>(RetTy));
     }
-    case Intrinsic::experimental_vector_splice: {
+    case Intrinsic::vector_splice: {
       unsigned Index = cast<ConstantInt>(Args[2])->getZExtValue();
       return thisT()->getShuffleCost(
           TTI::SK_Splice, cast<VectorType>(Args[0]->getType()), std::nullopt,
diff --git a/llvm/include/llvm/CodeGen/GlobalISel/CombinerHelper.h b/llvm/include/llvm/CodeGen/GlobalISel/CombinerHelper.h
index 4b8aec8e8a5d..76e8d1166ae0 100644
--- a/llvm/include/llvm/CodeGen/GlobalISel/CombinerHelper.h
+++ b/llvm/include/llvm/CodeGen/GlobalISel/CombinerHelper.h
@@ -853,6 +853,9 @@ public:
   bool matchExtractVectorElementWithDifferentIndices(const MachineOperand &MO,
                                                      BuildFnTy &MatchInfo);
 
+  /// Combine insert vector element OOB.
+  bool matchInsertVectorElementOOB(MachineInstr &MI, BuildFnTy &MatchInfo);
+
 private:
   /// Checks for legality of an indexed variant of \p LdSt.
   bool isIndexedLoadStoreLegal(GLoadStore &LdSt) const;
diff --git a/llvm/include/llvm/CodeGen/GlobalISel/GIMatchTableExecutor.h b/llvm/include/llvm/CodeGen/GlobalISel/GIMatchTableExecutor.h
index 29a46f04fd5d..8eddc6a6a531 100644
--- a/llvm/include/llvm/CodeGen/GlobalISel/GIMatchTableExecutor.h
+++ b/llvm/include/llvm/CodeGen/GlobalISel/GIMatchTableExecutor.h
@@ -217,6 +217,8 @@ enum {
   /// - OpIdx(ULEB128) - Operand index
   /// - Ty(1) - Expected type
   GIM_CheckType,
+  /// GIM_CheckType but InsnID is omitted and defaults to zero.
+  GIM_RootCheckType,
 
   /// Check the type of a pointer to any address space.
   /// - InsnID(ULEB128) - Instruction ID
@@ -229,6 +231,8 @@ enum {
   /// - OpIdx(ULEB128) - Operand index
   /// - RC(2) - Expected register bank (specified as a register class)
   GIM_CheckRegBankForClass,
+  /// GIM_CheckRegBankForClass but InsnID is omitted and defaults to zero.
+  GIM_RootCheckRegBankForClass,
 
   /// Check the operand matches a complex predicate
   /// - InsnID(ULEB128) - Instruction ID
@@ -278,9 +282,9 @@ enum {
   /// - OpIdx(ULEB128) - Operand index
   GIM_CheckIsImm,
 
-  /// Check if the specified operand is safe to fold into the current
-  /// instruction.
-  /// - InsnID(ULEB128) - Instruction ID
+  /// Checks if the matched instructions numbered [1, 1+N) can
+  /// be folded into the root (inst 0).
+  /// - Num(1)
   GIM_CheckIsSafeToFold,
 
   /// Check the specified operands are identical.
@@ -338,6 +342,8 @@ enum {
   /// - InsnID(ULEB128) - Instruction ID to define
   /// - Opcode(2) - The new opcode to use
   GIR_BuildMI,
+  /// GIR_BuildMI but InsnID is omitted and defaults to zero.
+  GIR_BuildRootMI,
 
   /// Builds a constant and stores its result in a TempReg.
   /// - TempRegID(ULEB128) - Temp Register to define.
@@ -349,6 +355,8 @@ enum {
   /// - OldInsnID(ULEB128) - Instruction ID to copy from
   /// - OpIdx(ULEB128) - The operand to copy
   GIR_Copy,
+  /// GIR_Copy but with both New/OldInsnIDs omitted and defaulting to zero.
+  GIR_RootToRootCopy,
 
   /// Copy an operand to the specified instruction or add a zero register if the
   /// operand is a zero immediate.
@@ -506,6 +514,9 @@ enum {
   /// description.
   /// - InsnID(ULEB128) - Instruction ID to modify
   GIR_ConstrainSelectedInstOperands,
+  /// GIR_ConstrainSelectedInstOperands but InsnID is omitted and defaults to
+  /// zero.
+  GIR_RootConstrainSelectedInstOperands,
 
   /// Merge all memory operands into instruction.
   /// - InsnID(ULEB128) - Instruction ID to modify
@@ -518,6 +529,9 @@ enum {
   /// - InsnID(ULEB128) - Instruction ID to erase
   GIR_EraseFromParent,
 
+  /// Combines both a GIR_EraseFromParent 0 + GIR_Done
+  GIR_EraseRootFromParent_Done,
+
   /// Create a new temporary register that's not constrained.
   /// - TempRegID(ULEB128) - The temporary register ID to initialize.
   /// - Ty(1) - Expected type
diff --git a/llvm/include/llvm/CodeGen/GlobalISel/GIMatchTableExecutorImpl.h b/llvm/include/llvm/CodeGen/GlobalISel/GIMatchTableExecutorImpl.h
index c73ac2c9f55b..dec2d97bb1fa 100644
--- a/llvm/include/llvm/CodeGen/GlobalISel/GIMatchTableExecutorImpl.h
+++ b/llvm/include/llvm/CodeGen/GlobalISel/GIMatchTableExecutorImpl.h
@@ -131,6 +131,16 @@ bool GIMatchTableExecutor::executeMatchTable(
     return V;
   };
 
+  const auto eraseImpl = [&](MachineInstr *MI) {
+    // If we're erasing the insertion point, ensure we don't leave a dangling
+    // pointer in the builder.
+    if (Builder.getInsertPt() == MI)
+      Builder.setInsertPt(*MI->getParent(), ++MI->getIterator());
+    if (Observer)
+      Observer->erasingInstr(*MI);
+    MI->eraseFromParent();
+  };
+
   while (true) {
     assert(CurrentIdx != ~0u && "Invalid MatchTable index");
     uint8_t MatcherOpcode = MatchTable[CurrentIdx++];
@@ -661,8 +671,9 @@ bool GIMatchTableExecutor::executeMatchTable(
 
       break;
     }
+    case GIM_RootCheckType:
     case GIM_CheckType: {
-      uint64_t InsnID = readULEB();
+      uint64_t InsnID = (MatcherOpcode == GIM_RootCheckType) ? 0 : readULEB();
       uint64_t OpIdx = readULEB();
       int TypeID = readS8();
       DEBUG_WITH_TYPE(TgtExecutor::getName(),
@@ -741,8 +752,11 @@ bool GIMatchTableExecutor::executeMatchTable(
       State.RecordedTypes[TypeIdx] = MRI.getType(Op.getReg());
       break;
     }
+
+    case GIM_RootCheckRegBankForClass:
     case GIM_CheckRegBankForClass: {
-      uint64_t InsnID = readULEB();
+      uint64_t InsnID =
+          (MatcherOpcode == GIM_RootCheckRegBankForClass) ? 0 : readULEB();
       uint64_t OpIdx = readULEB();
       uint16_t RCEnum = readU16();
       DEBUG_WITH_TYPE(TgtExecutor::getName(),
@@ -898,14 +912,16 @@ bool GIMatchTableExecutor::executeMatchTable(
       break;
     }
     case GIM_CheckIsSafeToFold: {
-      uint64_t InsnID = readULEB();
+      uint64_t NumInsn = MatchTable[CurrentIdx++];
       DEBUG_WITH_TYPE(TgtExecutor::getName(),
-                      dbgs() << CurrentIdx << ": GIM_CheckIsSafeToFold(MIs["
-                             << InsnID << "])\n");
-      assert(State.MIs[InsnID] != nullptr && "Used insn before defined");
-      if (!isObviouslySafeToFold(*State.MIs[InsnID], *State.MIs[0])) {
-        if (handleReject() == RejectAndGiveUp)
-          return false;
+                      dbgs() << CurrentIdx << ": GIM_CheckIsSafeToFold(N = "
+                             << NumInsn << ")\n");
+      MachineInstr &Root = *State.MIs[0];
+      for (unsigned K = 1, E = NumInsn + 1; K < E; ++K) {
+        if (!isObviouslySafeToFold(*State.MIs[K], Root)) {
+          if (handleReject() == RejectAndGiveUp)
+            return false;
+        }
       }
       break;
     }
@@ -1011,8 +1027,9 @@ bool GIMatchTableExecutor::executeMatchTable(
       break;
     }
 
+    case GIR_BuildRootMI:
     case GIR_BuildMI: {
-      uint64_t NewInsnID = readULEB();
+      uint64_t NewInsnID = (MatcherOpcode == GIR_BuildRootMI) ? 0 : readULEB();
       uint16_t Opcode = readU16();
       if (NewInsnID >= OutMIs.size())
         OutMIs.resize(NewInsnID + 1);
@@ -1034,9 +1051,12 @@ bool GIMatchTableExecutor::executeMatchTable(
       break;
     }
 
+    case GIR_RootToRootCopy:
     case GIR_Copy: {
-      uint64_t NewInsnID = readULEB();
-      uint64_t OldInsnID = readULEB();
+      uint64_t NewInsnID =
+          (MatcherOpcode == GIR_RootToRootCopy) ? 0 : readULEB();
+      uint64_t OldInsnID =
+          (MatcherOpcode == GIR_RootToRootCopy) ? 0 : readULEB();
       uint64_t OpIdx = readULEB();
       assert(OutMIs[NewInsnID] && "Attempted to add to undefined instruction");
       OutMIs[NewInsnID].add(State.MIs[OldInsnID]->getOperand(OpIdx));
@@ -1361,8 +1381,11 @@ bool GIMatchTableExecutor::executeMatchTable(
       break;
     }
 
+    case GIR_RootConstrainSelectedInstOperands:
     case GIR_ConstrainSelectedInstOperands: {
-      uint64_t InsnID = readULEB();
+      uint64_t InsnID = (MatcherOpcode == GIR_RootConstrainSelectedInstOperands)
+                            ? 0
+                            : readULEB();
       assert(OutMIs[InsnID] && "Attempted to add to undefined instruction");
       constrainSelectedInstRegOperands(*OutMIs[InsnID].getInstr(), TII, TRI,
                                        RBI);
@@ -1372,7 +1395,6 @@ bool GIMatchTableExecutor::executeMatchTable(
                              << InsnID << "])\n");
       break;
     }
-
     case GIR_MergeMemOperands: {
       uint64_t InsnID = readULEB();
       uint64_t NumInsn = MatchTable[CurrentIdx++];
@@ -1391,7 +1413,6 @@ bool GIMatchTableExecutor::executeMatchTable(
       DEBUG_WITH_TYPE(TgtExecutor::getName(), dbgs() << ")\n");
       break;
     }
-
     case GIR_EraseFromParent: {
       uint64_t InsnID = readULEB();
       MachineInstr *MI = State.MIs[InsnID];
@@ -1399,16 +1420,17 @@ bool GIMatchTableExecutor::executeMatchTable(
       DEBUG_WITH_TYPE(TgtExecutor::getName(),
                       dbgs() << CurrentIdx << ": GIR_EraseFromParent(MIs["
                              << InsnID << "])\n");
-      // If we're erasing the insertion point, ensure we don't leave a dangling
-      // pointer in the builder.
-      if (Builder.getInsertPt() == MI)
-        Builder.setInsertPt(*MI->getParent(), ++MI->getIterator());
-      if (Observer)
-        Observer->erasingInstr(*MI);
-      MI->eraseFromParent();
+      eraseImpl(MI);
       break;
     }
-
+    case GIR_EraseRootFromParent_Done: {
+      DEBUG_WITH_TYPE(TgtExecutor::getName(),
+                      dbgs()
+                          << CurrentIdx << ": GIR_EraseRootFromParent_Done\n");
+      eraseImpl(State.MIs[0]);
+      propagateFlags();
+      return true;
+    }
     case GIR_MakeTempReg: {
       uint64_t TempRegID = readULEB();
       int TypeID = readS8();
diff --git a/llvm/include/llvm/CodeGen/GlobalISel/IRTranslator.h b/llvm/include/llvm/CodeGen/GlobalISel/IRTranslator.h
index 5f28908e998a..deae2c55d26e 100644
--- a/llvm/include/llvm/CodeGen/GlobalISel/IRTranslator.h
+++ b/llvm/include/llvm/CodeGen/GlobalISel/IRTranslator.h
@@ -247,8 +247,8 @@ private:
   bool translateTrap(const CallInst &U, MachineIRBuilder &MIRBuilder,
                      unsigned Opcode);
 
-  // Translate @llvm.experimental.vector.interleave2 and
-  // @llvm.experimental.vector.deinterleave2 intrinsics for fixed-width vector
+  // Translate @llvm.vector.interleave2 and
+  // @llvm.vector.deinterleave2 intrinsics for fixed-width vector
   // types into vector shuffles.
   bool translateVectorInterleave2Intrinsic(const CallInst &CI,
                                            MachineIRBuilder &MIRBuilder);
diff --git a/llvm/include/llvm/CodeGen/GlobalISel/LegalizationArtifactCombiner.h b/llvm/include/llvm/CodeGen/GlobalISel/LegalizationArtifactCombiner.h
index ca62f38061b1..305bef7dd3ea 100644
--- a/llvm/include/llvm/CodeGen/GlobalISel/LegalizationArtifactCombiner.h
+++ b/llvm/include/llvm/CodeGen/GlobalISel/LegalizationArtifactCombiner.h
@@ -906,7 +906,8 @@ public:
                                         unsigned &DefOperandIdx) {
       if (Register Def = findValueFromDefImpl(Reg, 0, Size)) {
         if (auto *Unmerge = dyn_cast<GUnmerge>(MRI.getVRegDef(Def))) {
-          DefOperandIdx = Unmerge->findRegisterDefOperandIdx(Def);
+          DefOperandIdx =
+              Unmerge->findRegisterDefOperandIdx(Def, /*TRI=*/nullptr);
           return Unmerge;
         }
       }
diff --git a/llvm/include/llvm/CodeGen/GlobalISel/MachineIRBuilder.h b/llvm/include/llvm/CodeGen/GlobalISel/MachineIRBuilder.h
index be39eb7891f3..e15f7a7172e1 100644
--- a/llvm/include/llvm/CodeGen/GlobalISel/MachineIRBuilder.h
+++ b/llvm/include/llvm/CodeGen/GlobalISel/MachineIRBuilder.h
@@ -53,6 +53,8 @@ struct MachineIRBuilderState {
   DebugLoc DL;
   /// PC sections metadata to be set to any instruction we create.
   MDNode *PCSections = nullptr;
+  /// MMRA Metadata to be set on any instruction we create.
+  MDNode *MMRA = nullptr;
 
   /// \name Fields describing the insertion point.
   /// @{
@@ -354,6 +356,7 @@ public:
     setMBB(*MI.getParent());
     State.II = MI.getIterator();
     setPCSections(MI.getPCSections());
+    setMMRAMetadata(MI.getMMRAMetadata());
   }
   /// @}
 
@@ -387,6 +390,12 @@ public:
   /// Get the current instruction's PC sections metadata.
   MDNode *getPCSections() { return State.PCSections; }
 
+  /// Set the PC sections metadata to \p MD for all the next build instructions.
+  void setMMRAMetadata(MDNode *MMRA) { State.MMRA = MMRA; }
+
+  /// Get the current instruction's MMRA metadata.
+  MDNode *getMMRAMetadata() { return State.MMRA; }
+
   /// Build and insert <empty> = \p Opcode <empty>.
   /// The insertion point is the one set by the last call of either
   /// setBasicBlock or setMI.
diff --git a/llvm/include/llvm/CodeGen/GlobalISel/Utils.h b/llvm/include/llvm/CodeGen/GlobalISel/Utils.h
index c4174cee5e10..70421a518ab7 100644
--- a/llvm/include/llvm/CodeGen/GlobalISel/Utils.h
+++ b/llvm/include/llvm/CodeGen/GlobalISel/Utils.h
@@ -559,5 +559,31 @@ void salvageDebugInfo(const MachineRegisterInfo &MRI, MachineInstr &MI);
 /// having only floating-point operands.
 bool isPreISelGenericFloatingPointOpcode(unsigned Opc);
 
+/// Returns true if \p Reg can create undef or poison from non-undef &
+/// non-poison operands. \p ConsiderFlagsAndMetadata controls whether poison
+/// producing flags and metadata on the instruction are considered. This can be
+/// used to see if the instruction could still introduce undef or poison even
+/// without poison generating flags and metadata which might be on the
+/// instruction.
+bool canCreateUndefOrPoison(Register Reg, const MachineRegisterInfo &MRI,
+                            bool ConsiderFlagsAndMetadata = true);
+
+/// Returns true if \p Reg can create poison from non-poison operands.
+bool canCreatePoison(Register Reg, const MachineRegisterInfo &MRI,
+                     bool ConsiderFlagsAndMetadata = true);
+
+/// Returns true if \p Reg cannot be poison and undef.
+bool isGuaranteedNotToBeUndefOrPoison(Register Reg,
+                                      const MachineRegisterInfo &MRI,
+                                      unsigned Depth = 0);
+
+/// Returns true if \p Reg cannot be poison, but may be undef.
+bool isGuaranteedNotToBePoison(Register Reg, const MachineRegisterInfo &MRI,
+                               unsigned Depth = 0);
+
+/// Returns true if \p Reg cannot be undef, but may be poison.
+bool isGuaranteedNotToBeUndef(Register Reg, const MachineRegisterInfo &MRI,
+                              unsigned Depth = 0);
+
 } // End namespace llvm.
 #endif
diff --git a/llvm/include/llvm/CodeGen/ISDOpcodes.h b/llvm/include/llvm/CodeGen/ISDOpcodes.h
index 078a936b061a..6429947958ee 100644
--- a/llvm/include/llvm/CodeGen/ISDOpcodes.h
+++ b/llvm/include/llvm/CodeGen/ISDOpcodes.h
@@ -205,6 +205,7 @@ enum NodeType {
   /// CopyFromReg - This node indicates that the input value is a virtual or
   /// physical register that is defined outside of the scope of this
   /// SelectionDAG.  The register is available from the RegisterSDNode object.
+  /// Note that CopyFromReg is considered as also freezing the value.
   CopyFromReg,
 
   /// UNDEF - An undefined node.
diff --git a/llvm/include/llvm/CodeGen/MachineFrameInfo.h b/llvm/include/llvm/CodeGen/MachineFrameInfo.h
index 0fe73fec7ee6..a2c78e9e093d 100644
--- a/llvm/include/llvm/CodeGen/MachineFrameInfo.h
+++ b/llvm/include/llvm/CodeGen/MachineFrameInfo.h
@@ -701,6 +701,13 @@ public:
     return Objects[ObjectIdx+NumFixedObjects].isAliased;
   }
 
+  /// Set "maybe pointed to by an LLVM IR value" for an object.
+  void setIsAliasedObjectIndex(int ObjectIdx, bool IsAliased) {
+    assert(unsigned(ObjectIdx+NumFixedObjects) < Objects.size() &&
+           "Invalid Object Idx!");
+    Objects[ObjectIdx+NumFixedObjects].isAliased = IsAliased;
+  }
+
   /// Returns true if the specified index corresponds to an immutable object.
   bool isImmutableObjectIndex(int ObjectIdx) const {
     // Tail calling functions can clobber their function arguments.
diff --git a/llvm/include/llvm/CodeGen/MachineFunction.h b/llvm/include/llvm/CodeGen/MachineFunction.h
index 470997b31fe8..9f8e846cac45 100644
--- a/llvm/include/llvm/CodeGen/MachineFunction.h
+++ b/llvm/include/llvm/CodeGen/MachineFunction.h
@@ -1125,7 +1125,8 @@ public:
   MachineInstr::ExtraInfo *createMIExtraInfo(
       ArrayRef<MachineMemOperand *> MMOs, MCSymbol *PreInstrSymbol = nullptr,
       MCSymbol *PostInstrSymbol = nullptr, MDNode *HeapAllocMarker = nullptr,
-      MDNode *PCSections = nullptr, uint32_t CFIType = 0);
+      MDNode *PCSections = nullptr, uint32_t CFIType = 0,
+      MDNode *MMRAs = nullptr);
 
   /// Allocate a string and populate it with the given external symbol name.
   const char *createExternalSymbolName(StringRef Name);
diff --git a/llvm/include/llvm/CodeGen/MachineInstr.h b/llvm/include/llvm/CodeGen/MachineInstr.h
index 7249f812d2cc..2b0c5d166d88 100644
--- a/llvm/include/llvm/CodeGen/MachineInstr.h
+++ b/llvm/include/llvm/CodeGen/MachineInstr.h
@@ -160,37 +160,41 @@ private:
                              MCSymbol *PreInstrSymbol = nullptr,
                              MCSymbol *PostInstrSymbol = nullptr,
                              MDNode *HeapAllocMarker = nullptr,
-                             MDNode *PCSections = nullptr,
-                             uint32_t CFIType = 0) {
+                             MDNode *PCSections = nullptr, uint32_t CFIType = 0,
+                             MDNode *MMRAs = nullptr) {
       bool HasPreInstrSymbol = PreInstrSymbol != nullptr;
       bool HasPostInstrSymbol = PostInstrSymbol != nullptr;
       bool HasHeapAllocMarker = HeapAllocMarker != nullptr;
+      bool HasMMRAs = MMRAs != nullptr;
       bool HasCFIType = CFIType != 0;
       bool HasPCSections = PCSections != nullptr;
       auto *Result = new (Allocator.Allocate(
           totalSizeToAlloc<MachineMemOperand *, MCSymbol *, MDNode *, uint32_t>(
               MMOs.size(), HasPreInstrSymbol + HasPostInstrSymbol,
-              HasHeapAllocMarker + HasPCSections, HasCFIType),
+              HasHeapAllocMarker + HasPCSections + HasMMRAs, HasCFIType),
           alignof(ExtraInfo)))
           ExtraInfo(MMOs.size(), HasPreInstrSymbol, HasPostInstrSymbol,
-                    HasHeapAllocMarker, HasPCSections, HasCFIType);
+                    HasHeapAllocMarker, HasPCSections, HasCFIType, HasMMRAs);
 
       // Copy the actual data into the trailing objects.
       std::copy(MMOs.begin(), MMOs.end(),
                 Result->getTrailingObjects<MachineMemOperand *>());
 
+      unsigned MDNodeIdx = 0;
+
       if (HasPreInstrSymbol)
         Result->getTrailingObjects<MCSymbol *>()[0] = PreInstrSymbol;
       if (HasPostInstrSymbol)
         Result->getTrailingObjects<MCSymbol *>()[HasPreInstrSymbol] =
             PostInstrSymbol;
       if (HasHeapAllocMarker)
-        Result->getTrailingObjects<MDNode *>()[0] = HeapAllocMarker;
+        Result->getTrailingObjects<MDNode *>()[MDNodeIdx++] = HeapAllocMarker;
       if (HasPCSections)
-        Result->getTrailingObjects<MDNode *>()[HasHeapAllocMarker] =
-            PCSections;
+        Result->getTrailingObjects<MDNode *>()[MDNodeIdx++] = PCSections;
       if (HasCFIType)
         Result->getTrailingObjects<uint32_t>()[0] = CFIType;
+      if (HasMMRAs)
+        Result->getTrailingObjects<MDNode *>()[MDNodeIdx++] = MMRAs;
 
       return Result;
     }
@@ -223,6 +227,12 @@ private:
       return HasCFIType ? getTrailingObjects<uint32_t>()[0] : 0;
     }
 
+    MDNode *getMMRAMetadata() const {
+      return HasMMRAs ? getTrailingObjects<MDNode *>()[HasHeapAllocMarker +
+                                                       HasPCSections]
+                      : nullptr;
+    }
+
   private:
     friend TrailingObjects;
 
@@ -237,6 +247,7 @@ private:
     const bool HasHeapAllocMarker;
     const bool HasPCSections;
     const bool HasCFIType;
+    const bool HasMMRAs;
 
     // Implement the `TrailingObjects` internal API.
     size_t numTrailingObjects(OverloadToken<MachineMemOperand *>) const {
@@ -255,11 +266,12 @@ private:
     // Just a boring constructor to allow us to initialize the sizes. Always use
     // the `create` routine above.
     ExtraInfo(int NumMMOs, bool HasPreInstrSymbol, bool HasPostInstrSymbol,
-              bool HasHeapAllocMarker, bool HasPCSections, bool HasCFIType)
+              bool HasHeapAllocMarker, bool HasPCSections, bool HasCFIType,
+              bool HasMMRAs)
         : NumMMOs(NumMMOs), HasPreInstrSymbol(HasPreInstrSymbol),
           HasPostInstrSymbol(HasPostInstrSymbol),
           HasHeapAllocMarker(HasHeapAllocMarker), HasPCSections(HasPCSections),
-          HasCFIType(HasCFIType) {}
+          HasCFIType(HasCFIType), HasMMRAs(HasMMRAs) {}
   };
 
   /// Enumeration of the kinds of inline extra info available. It is important
@@ -838,6 +850,15 @@ public:
     return nullptr;
   }
 
+  /// Helper to extract mmra.op metadata.
+  MDNode *getMMRAMetadata() const {
+    if (!Info)
+      return nullptr;
+    if (ExtraInfo *EI = Info.get<EIIK_OutOfLine>())
+      return EI->getMMRAMetadata();
+    return nullptr;
+  }
+
   /// Helper to extract a CFI type hash if one has been added.
   uint32_t getCFIType() const {
     if (!Info)
@@ -1445,9 +1466,8 @@ public:
   /// is a read of a super-register.
   /// This does not count partial redefines of virtual registers as reads:
   ///   %reg1024:6 = OP.
-  bool readsRegister(Register Reg,
-                     const TargetRegisterInfo *TRI = nullptr) const {
-    return findRegisterUseOperandIdx(Reg, false, TRI) != -1;
+  bool readsRegister(Register Reg, const TargetRegisterInfo *TRI) const {
+    return findRegisterUseOperandIdx(Reg, TRI, false) != -1;
   }
 
   /// Return true if the MachineInstr reads the specified virtual register.
@@ -1466,34 +1486,30 @@ public:
   /// Return true if the MachineInstr kills the specified register.
   /// If TargetRegisterInfo is non-null, then it also checks if there is
   /// a kill of a super-register.
-  bool killsRegister(Register Reg,
-                     const TargetRegisterInfo *TRI = nullptr) const {
-    return findRegisterUseOperandIdx(Reg, true, TRI) != -1;
+  bool killsRegister(Register Reg, const TargetRegisterInfo *TRI) const {
+    return findRegisterUseOperandIdx(Reg, TRI, true) != -1;
   }
 
   /// Return true if the MachineInstr fully defines the specified register.
   /// If TargetRegisterInfo is non-null, then it also checks
   /// if there is a def of a super-register.
   /// NOTE: It's ignoring subreg indices on virtual registers.
-  bool definesRegister(Register Reg,
-                       const TargetRegisterInfo *TRI = nullptr) const {
-    return findRegisterDefOperandIdx(Reg, false, false, TRI) != -1;
+  bool definesRegister(Register Reg, const TargetRegisterInfo *TRI) const {
+    return findRegisterDefOperandIdx(Reg, TRI, false, false) != -1;
   }
 
   /// Return true if the MachineInstr modifies (fully define or partially
   /// define) the specified register.
   /// NOTE: It's ignoring subreg indices on virtual registers.
-  bool modifiesRegister(Register Reg,
-                        const TargetRegisterInfo *TRI = nullptr) const {
-    return findRegisterDefOperandIdx(Reg, false, true, TRI) != -1;
+  bool modifiesRegister(Register Reg, const TargetRegisterInfo *TRI) const {
+    return findRegisterDefOperandIdx(Reg, TRI, false, true) != -1;
   }
 
   /// Returns true if the register is dead in this machine instruction.
   /// If TargetRegisterInfo is non-null, then it also checks
   /// if there is a dead def of a super-register.
-  bool registerDefIsDead(Register Reg,
-                         const TargetRegisterInfo *TRI = nullptr) const {
-    return findRegisterDefOperandIdx(Reg, true, false, TRI) != -1;
+  bool registerDefIsDead(Register Reg, const TargetRegisterInfo *TRI) const {
+    return findRegisterDefOperandIdx(Reg, TRI, true, false) != -1;
   }
 
   /// Returns true if the MachineInstr has an implicit-use operand of exactly
@@ -1503,22 +1519,23 @@ public:
   /// Returns the operand index that is a use of the specific register or -1
   /// if it is not found. It further tightens the search criteria to a use
   /// that kills the register if isKill is true.
-  int findRegisterUseOperandIdx(Register Reg, bool isKill = false,
-                                const TargetRegisterInfo *TRI = nullptr) const;
+  int findRegisterUseOperandIdx(Register Reg, const TargetRegisterInfo *TRI,
+                                bool isKill = false) const;
 
   /// Wrapper for findRegisterUseOperandIdx, it returns
   /// a pointer to the MachineOperand rather than an index.
-  MachineOperand *findRegisterUseOperand(Register Reg, bool isKill = false,
-                                      const TargetRegisterInfo *TRI = nullptr) {
-    int Idx = findRegisterUseOperandIdx(Reg, isKill, TRI);
+  MachineOperand *findRegisterUseOperand(Register Reg,
+                                         const TargetRegisterInfo *TRI,
+                                         bool isKill = false) {
+    int Idx = findRegisterUseOperandIdx(Reg, TRI, isKill);
     return (Idx == -1) ? nullptr : &getOperand(Idx);
   }
 
-  const MachineOperand *findRegisterUseOperand(
-    Register Reg, bool isKill = false,
-    const TargetRegisterInfo *TRI = nullptr) const {
-    return const_cast<MachineInstr *>(this)->
-      findRegisterUseOperand(Reg, isKill, TRI);
+  const MachineOperand *findRegisterUseOperand(Register Reg,
+                                               const TargetRegisterInfo *TRI,
+                                               bool isKill = false) const {
+    return const_cast<MachineInstr *>(this)->findRegisterUseOperand(Reg, TRI,
+                                                                    isKill);
   }
 
   /// Returns the operand index that is a def of the specified register or
@@ -1527,26 +1544,26 @@ public:
   /// overlap the specified register. If TargetRegisterInfo is non-null,
   /// then it also checks if there is a def of a super-register.
   /// This may also return a register mask operand when Overlap is true.
-  int findRegisterDefOperandIdx(Register Reg,
-                                bool isDead = false, bool Overlap = false,
-                                const TargetRegisterInfo *TRI = nullptr) const;
+  int findRegisterDefOperandIdx(Register Reg, const TargetRegisterInfo *TRI,
+                                bool isDead = false,
+                                bool Overlap = false) const;
 
   /// Wrapper for findRegisterDefOperandIdx, it returns
   /// a pointer to the MachineOperand rather than an index.
-  MachineOperand *
-  findRegisterDefOperand(Register Reg, bool isDead = false,
-                         bool Overlap = false,
-                         const TargetRegisterInfo *TRI = nullptr) {
-    int Idx = findRegisterDefOperandIdx(Reg, isDead, Overlap, TRI);
+  MachineOperand *findRegisterDefOperand(Register Reg,
+                                         const TargetRegisterInfo *TRI,
+                                         bool isDead = false,
+                                         bool Overlap = false) {
+    int Idx = findRegisterDefOperandIdx(Reg, TRI, isDead, Overlap);
     return (Idx == -1) ? nullptr : &getOperand(Idx);
   }
 
-  const MachineOperand *
-  findRegisterDefOperand(Register Reg, bool isDead = false,
-                         bool Overlap = false,
-                         const TargetRegisterInfo *TRI = nullptr) const {
+  const MachineOperand *findRegisterDefOperand(Register Reg,
+                                               const TargetRegisterInfo *TRI,
+                                               bool isDead = false,
+                                               bool Overlap = false) const {
     return const_cast<MachineInstr *>(this)->findRegisterDefOperand(
-        Reg, isDead, Overlap, TRI);
+        Reg, TRI, isDead, Overlap);
   }
 
   /// Find the index of the first operand in the
@@ -1902,6 +1919,8 @@ public:
   // addresses into.
   void setPCSections(MachineFunction &MF, MDNode *MD);
 
+  void setMMRAMetadata(MachineFunction &MF, MDNode *MMRAs);
+
   /// Set the CFI type for the instruction.
   void setCFIType(MachineFunction &MF, uint32_t Type);
 
@@ -2014,7 +2033,7 @@ private:
   void setExtraInfo(MachineFunction &MF, ArrayRef<MachineMemOperand *> MMOs,
                     MCSymbol *PreInstrSymbol, MCSymbol *PostInstrSymbol,
                     MDNode *HeapAllocMarker, MDNode *PCSections,
-                    uint32_t CFIType);
+                    uint32_t CFIType, MDNode *MMRAs);
 };
 
 /// Special DenseMapInfo traits to compare MachineInstr* by *value* of the
diff --git a/llvm/include/llvm/CodeGen/MachineInstrBuilder.h b/llvm/include/llvm/CodeGen/MachineInstrBuilder.h
index 954d8e6770a2..a5b8d3af3cc9 100644
--- a/llvm/include/llvm/CodeGen/MachineInstrBuilder.h
+++ b/llvm/include/llvm/CodeGen/MachineInstrBuilder.h
@@ -322,6 +322,12 @@ public:
     return *this;
   }
 
+  const MachineInstrBuilder &setMMRAMetadata(MDNode *MMRA) const {
+    if (MMRA)
+      MI->setMMRAMetadata(*MF, MMRA);
+    return *this;
+  }
+
   /// Copy all the implicit operands from OtherMI onto this one.
   const MachineInstrBuilder &
   copyImplicitOps(const MachineInstr &OtherMI) const {
@@ -337,14 +343,15 @@ public:
 };
 
 /// Set of metadata that should be preserved when using BuildMI(). This provides
-/// a more convenient way of preserving DebugLoc and PCSections.
+/// a more convenient way of preserving DebugLoc, PCSections and MMRA.
 class MIMetadata {
 public:
   MIMetadata() = default;
-  MIMetadata(DebugLoc DL, MDNode *PCSections = nullptr)
-      : DL(std::move(DL)), PCSections(PCSections) {}
-  MIMetadata(const DILocation *DI, MDNode *PCSections = nullptr)
-      : DL(DI), PCSections(PCSections) {}
+  MIMetadata(DebugLoc DL, MDNode *PCSections = nullptr, MDNode *MMRA = nullptr)
+      : DL(std::move(DL)), PCSections(PCSections), MMRA(MMRA) {}
+  MIMetadata(const DILocation *DI, MDNode *PCSections = nullptr,
+             MDNode *MMRA = nullptr)
+      : DL(DI), PCSections(PCSections), MMRA(MMRA) {}
   explicit MIMetadata(const Instruction &From)
       : DL(From.getDebugLoc()),
         PCSections(From.getMetadata(LLVMContext::MD_pcsections)) {}
@@ -353,17 +360,20 @@ public:
 
   const DebugLoc &getDL() const { return DL; }
   MDNode *getPCSections() const { return PCSections; }
+  MDNode *getMMRAMetadata() const { return MMRA; }
 
 private:
   DebugLoc DL;
   MDNode *PCSections = nullptr;
+  MDNode *MMRA = nullptr;
 };
 
 /// Builder interface. Specify how to create the initial instruction itself.
 inline MachineInstrBuilder BuildMI(MachineFunction &MF, const MIMetadata &MIMD,
                                    const MCInstrDesc &MCID) {
   return MachineInstrBuilder(MF, MF.CreateMachineInstr(MCID, MIMD.getDL()))
-           .setPCSections(MIMD.getPCSections());
+      .setPCSections(MIMD.getPCSections())
+      .setMMRAMetadata(MIMD.getMMRAMetadata());
 }
 
 /// This version of the builder sets up the first operand as a
@@ -371,8 +381,9 @@ inline MachineInstrBuilder BuildMI(MachineFunction &MF, const MIMetadata &MIMD,
 inline MachineInstrBuilder BuildMI(MachineFunction &MF, const MIMetadata &MIMD,
                                    const MCInstrDesc &MCID, Register DestReg) {
   return MachineInstrBuilder(MF, MF.CreateMachineInstr(MCID, MIMD.getDL()))
-           .setPCSections(MIMD.getPCSections())
-           .addReg(DestReg, RegState::Define);
+      .setPCSections(MIMD.getPCSections())
+      .setMMRAMetadata(MIMD.getMMRAMetadata())
+      .addReg(DestReg, RegState::Define);
 }
 
 /// This version of the builder inserts the newly-built instruction before
@@ -386,8 +397,9 @@ inline MachineInstrBuilder BuildMI(MachineBasicBlock &BB,
   MachineInstr *MI = MF.CreateMachineInstr(MCID, MIMD.getDL());
   BB.insert(I, MI);
   return MachineInstrBuilder(MF, MI)
-           .setPCSections(MIMD.getPCSections())
-           .addReg(DestReg, RegState::Define);
+      .setPCSections(MIMD.getPCSections())
+      .setMMRAMetadata(MIMD.getMMRAMetadata())
+      .addReg(DestReg, RegState::Define);
 }
 
 /// This version of the builder inserts the newly-built instruction before
@@ -404,8 +416,9 @@ inline MachineInstrBuilder BuildMI(MachineBasicBlock &BB,
   MachineInstr *MI = MF.CreateMachineInstr(MCID, MIMD.getDL());
   BB.insert(I, MI);
   return MachineInstrBuilder(MF, MI)
-           .setPCSections(MIMD.getPCSections())
-           .addReg(DestReg, RegState::Define);
+      .setPCSections(MIMD.getPCSections())
+      .setMMRAMetadata(MIMD.getMMRAMetadata())
+      .addReg(DestReg, RegState::Define);
 }
 
 inline MachineInstrBuilder BuildMI(MachineBasicBlock &BB, MachineInstr &I,
@@ -435,7 +448,9 @@ inline MachineInstrBuilder BuildMI(MachineBasicBlock &BB,
   MachineFunction &MF = *BB.getParent();
   MachineInstr *MI = MF.CreateMachineInstr(MCID, MIMD.getDL());
   BB.insert(I, MI);
-  return MachineInstrBuilder(MF, MI).setPCSections(MIMD.getPCSections());
+  return MachineInstrBuilder(MF, MI)
+      .setPCSections(MIMD.getPCSections())
+      .setMMRAMetadata(MIMD.getMMRAMetadata());
 }
 
 inline MachineInstrBuilder BuildMI(MachineBasicBlock &BB,
@@ -445,7 +460,9 @@ inline MachineInstrBuilder BuildMI(MachineBasicBlock &BB,
   MachineFunction &MF = *BB.getParent();
   MachineInstr *MI = MF.CreateMachineInstr(MCID, MIMD.getDL());
   BB.insert(I, MI);
-  return MachineInstrBuilder(MF, MI).setPCSections(MIMD.getPCSections());
+  return MachineInstrBuilder(MF, MI)
+      .setPCSections(MIMD.getPCSections())
+      .setMMRAMetadata(MIMD.getMMRAMetadata());
 }
 
 inline MachineInstrBuilder BuildMI(MachineBasicBlock &BB, MachineInstr &I,
diff --git a/llvm/include/llvm/CodeGen/SelectionDAG.h b/llvm/include/llvm/CodeGen/SelectionDAG.h
index f347131be080..f353aef1f446 100644
--- a/llvm/include/llvm/CodeGen/SelectionDAG.h
+++ b/llvm/include/llvm/CodeGen/SelectionDAG.h
@@ -284,6 +284,7 @@ class SelectionDAG {
     CallSiteInfo CSInfo;
     MDNode *HeapAllocSite = nullptr;
     MDNode *PCSections = nullptr;
+    MDNode *MMRA = nullptr;
     bool NoMerge = false;
   };
   /// Out-of-line extra information for SDNodes.
@@ -2145,6 +2146,10 @@ public:
   const APInt *getValidShiftAmountConstant(SDValue V,
                                            const APInt &DemandedElts) const;
 
+  /// If a SHL/SRA/SRL node \p V has a constant or splat constant shift amount
+  /// that is less than the element bit-width of the shift node, return it.
+  const APInt *getValidShiftAmountConstant(SDValue V) const;
+
   /// If a SHL/SRA/SRL node \p V has constant shift amounts that are all less
   /// than the element bit-width of the shift node, return the minimum value.
   const APInt *
@@ -2152,11 +2157,21 @@ public:
                                      const APInt &DemandedElts) const;
 
   /// If a SHL/SRA/SRL node \p V has constant shift amounts that are all less
+  /// than the element bit-width of the shift node, return the minimum value.
+  const APInt *
+  getValidMinimumShiftAmountConstant(SDValue V) const;
+
+  /// If a SHL/SRA/SRL node \p V has constant shift amounts that are all less
   /// than the element bit-width of the shift node, return the maximum value.
   const APInt *
   getValidMaximumShiftAmountConstant(SDValue V,
                                      const APInt &DemandedElts) const;
 
+  /// If a SHL/SRA/SRL node \p V has constant shift amounts that are all less
+  /// than the element bit-width of the shift node, return the maximum value.
+  const APInt *
+  getValidMaximumShiftAmountConstant(SDValue V) const;
+
   /// Match a binop + shuffle pyramid that represents a horizontal reduction
   /// over the elements of a vector starting from the EXTRACT_VECTOR_ELT node /p
   /// Extract. The reduction must use one of the opcodes listed in /p
@@ -2279,11 +2294,21 @@ public:
   void addPCSections(const SDNode *Node, MDNode *MD) {
     SDEI[Node].PCSections = MD;
   }
+  /// Set MMRAMetadata to be associated with Node.
+  void addMMRAMetadata(const SDNode *Node, MDNode *MMRA) {
+    SDEI[Node].MMRA = MMRA;
+  }
   /// Return PCSections associated with Node, or nullptr if none exists.
   MDNode *getPCSections(const SDNode *Node) const {
     auto It = SDEI.find(Node);
     return It != SDEI.end() ? It->second.PCSections : nullptr;
   }
+  /// Return the MMRA MDNode associated with Node, or nullptr if none
+  /// exists.
+  MDNode *getMMRAMetadata(const SDNode *Node) const {
+    auto It = SDEI.find(Node);
+    return It != SDEI.end() ? It->second.MMRA : nullptr;
+  }
   /// Set NoMergeSiteInfo to be associated with Node if NoMerge is true.
   void addNoMergeSiteInfo(const SDNode *Node, bool NoMerge) {
     if (NoMerge)
diff --git a/llvm/include/llvm/CodeGen/SelectionDAGNodes.h b/llvm/include/llvm/CodeGen/SelectionDAGNodes.h
index 261f7e49e5c8..e7c710414545 100644
--- a/llvm/include/llvm/CodeGen/SelectionDAGNodes.h
+++ b/llvm/include/llvm/CodeGen/SelectionDAGNodes.h
@@ -999,6 +999,13 @@ public:
   /// If Flags is not in a defined state then this has no effect.
   void intersectFlagsWith(const SDNodeFlags Flags);
 
+  bool hasPoisonGeneratingFlags() const {
+    SDNodeFlags Flags = getFlags();
+    return Flags.hasNoUnsignedWrap() || Flags.hasNoSignedWrap() ||
+           Flags.hasExact() || Flags.hasDisjoint() || Flags.hasNonNeg() ||
+           Flags.hasNoNaNs() || Flags.hasNoInfs();
+  }
+
   void setCFIType(uint32_t Type) { CFIType = Type; }
   uint32_t getCFIType() const { return CFIType; }
 
@@ -1283,8 +1290,10 @@ private:
   unsigned DestAddrSpace;
 
 public:
-  AddrSpaceCastSDNode(unsigned Order, const DebugLoc &dl, EVT VT,
-                      unsigned SrcAS, unsigned DestAS);
+  AddrSpaceCastSDNode(unsigned Order, const DebugLoc &dl, SDVTList VTs,
+                      unsigned SrcAS, unsigned DestAS)
+      : SDNode(ISD::ADDRSPACECAST, Order, dl, VTs), SrcAddrSpace(SrcAS),
+        DestAddrSpace(DestAS) {}
 
   unsigned getSrcAddressSpace() const { return SrcAddrSpace; }
   unsigned getDestAddressSpace() const { return DestAddrSpace; }
@@ -1573,8 +1582,9 @@ class ShuffleVectorSDNode : public SDNode {
 protected:
   friend class SelectionDAG;
 
-  ShuffleVectorSDNode(EVT VT, unsigned Order, const DebugLoc &dl, const int *M)
-      : SDNode(ISD::VECTOR_SHUFFLE, Order, dl, getSDVTList(VT)), Mask(M) {}
+  ShuffleVectorSDNode(SDVTList VTs, unsigned Order, const DebugLoc &dl,
+                      const int *M)
+      : SDNode(ISD::VECTOR_SHUFFLE, Order, dl, VTs), Mask(M) {}
 
 public:
   ArrayRef<int> getMask() const {
@@ -1628,9 +1638,10 @@ class ConstantSDNode : public SDNode {
 
   const ConstantInt *Value;
 
-  ConstantSDNode(bool isTarget, bool isOpaque, const ConstantInt *val, EVT VT)
+  ConstantSDNode(bool isTarget, bool isOpaque, const ConstantInt *val,
+                 SDVTList VTs)
       : SDNode(isTarget ? ISD::TargetConstant : ISD::Constant, 0, DebugLoc(),
-               getSDVTList(VT)),
+               VTs),
         Value(val) {
     ConstantSDNodeBits.IsOpaque = isOpaque;
   }
@@ -1681,9 +1692,9 @@ class ConstantFPSDNode : public SDNode {
 
   const ConstantFP *Value;
 
-  ConstantFPSDNode(bool isTarget, const ConstantFP *val, EVT VT)
+  ConstantFPSDNode(bool isTarget, const ConstantFP *val, SDVTList VTs)
       : SDNode(isTarget ? ISD::TargetConstantFP : ISD::ConstantFP, 0,
-               DebugLoc(), getSDVTList(VT)),
+               DebugLoc(), VTs),
         Value(val) {}
 
 public:
@@ -1816,8 +1827,10 @@ class GlobalAddressSDNode : public SDNode {
   unsigned TargetFlags;
 
   GlobalAddressSDNode(unsigned Opc, unsigned Order, const DebugLoc &DL,
-                      const GlobalValue *GA, EVT VT, int64_t o,
-                      unsigned TF);
+                      const GlobalValue *GA, SDVTList VTs, int64_t o,
+                      unsigned TF)
+      : SDNode(Opc, Order, DL, VTs), TheGlobal(GA), Offset(o), TargetFlags(TF) {
+  }
 
 public:
   const GlobalValue *getGlobal() const { return TheGlobal; }
@@ -1839,10 +1852,10 @@ class FrameIndexSDNode : public SDNode {
 
   int FI;
 
-  FrameIndexSDNode(int fi, EVT VT, bool isTarg)
-    : SDNode(isTarg ? ISD::TargetFrameIndex : ISD::FrameIndex,
-      0, DebugLoc(), getSDVTList(VT)), FI(fi) {
-  }
+  FrameIndexSDNode(int fi, SDVTList VTs, bool isTarg)
+      : SDNode(isTarg ? ISD::TargetFrameIndex : ISD::FrameIndex, 0, DebugLoc(),
+               VTs),
+        FI(fi) {}
 
 public:
   int getIndex() const { return FI; }
@@ -1917,10 +1930,10 @@ class JumpTableSDNode : public SDNode {
   int JTI;
   unsigned TargetFlags;
 
-  JumpTableSDNode(int jti, EVT VT, bool isTarg, unsigned TF)
-    : SDNode(isTarg ? ISD::TargetJumpTable : ISD::JumpTable,
-      0, DebugLoc(), getSDVTList(VT)), JTI(jti), TargetFlags(TF) {
-  }
+  JumpTableSDNode(int jti, SDVTList VTs, bool isTarg, unsigned TF)
+      : SDNode(isTarg ? ISD::TargetJumpTable : ISD::JumpTable, 0, DebugLoc(),
+               VTs),
+        JTI(jti), TargetFlags(TF) {}
 
 public:
   int getIndex() const { return JTI; }
@@ -1943,19 +1956,19 @@ class ConstantPoolSDNode : public SDNode {
   Align Alignment; // Minimum alignment requirement of CP.
   unsigned TargetFlags;
 
-  ConstantPoolSDNode(bool isTarget, const Constant *c, EVT VT, int o,
+  ConstantPoolSDNode(bool isTarget, const Constant *c, SDVTList VTs, int o,
                      Align Alignment, unsigned TF)
       : SDNode(isTarget ? ISD::TargetConstantPool : ISD::ConstantPool, 0,
-               DebugLoc(), getSDVTList(VT)),
+               DebugLoc(), VTs),
         Offset(o), Alignment(Alignment), TargetFlags(TF) {
     assert(Offset >= 0 && "Offset is too large");
     Val.ConstVal = c;
   }
 
-  ConstantPoolSDNode(bool isTarget, MachineConstantPoolValue *v, EVT VT, int o,
-                     Align Alignment, unsigned TF)
+  ConstantPoolSDNode(bool isTarget, MachineConstantPoolValue *v, SDVTList VTs,
+                     int o, Align Alignment, unsigned TF)
       : SDNode(isTarget ? ISD::TargetConstantPool : ISD::ConstantPool, 0,
-               DebugLoc(), getSDVTList(VT)),
+               DebugLoc(), VTs),
         Offset(o), Alignment(Alignment), TargetFlags(TF) {
     assert(Offset >= 0 && "Offset is too large");
     Val.MachineCPVal = v;
@@ -2003,9 +2016,9 @@ class TargetIndexSDNode : public SDNode {
   int64_t Offset;
 
 public:
-  TargetIndexSDNode(int Idx, EVT VT, int64_t Ofs, unsigned TF)
-      : SDNode(ISD::TargetIndex, 0, DebugLoc(), getSDVTList(VT)),
-        TargetFlags(TF), Index(Idx), Offset(Ofs) {}
+  TargetIndexSDNode(int Idx, SDVTList VTs, int64_t Ofs, unsigned TF)
+      : SDNode(ISD::TargetIndex, 0, DebugLoc(), VTs), TargetFlags(TF),
+        Index(Idx), Offset(Ofs) {}
 
   unsigned getTargetFlags() const { return TargetFlags; }
   int getIndex() const { return Index; }
@@ -2215,8 +2228,8 @@ class RegisterSDNode : public SDNode {
 
   Register Reg;
 
-  RegisterSDNode(Register reg, EVT VT)
-    : SDNode(ISD::Register, 0, DebugLoc(), getSDVTList(VT)), Reg(reg) {}
+  RegisterSDNode(Register reg, SDVTList VTs)
+      : SDNode(ISD::Register, 0, DebugLoc(), VTs), Reg(reg) {}
 
 public:
   Register getReg() const { return Reg; }
@@ -2251,10 +2264,10 @@ class BlockAddressSDNode : public SDNode {
   int64_t Offset;
   unsigned TargetFlags;
 
-  BlockAddressSDNode(unsigned NodeTy, EVT VT, const BlockAddress *ba,
+  BlockAddressSDNode(unsigned NodeTy, SDVTList VTs, const BlockAddress *ba,
                      int64_t o, unsigned Flags)
-    : SDNode(NodeTy, 0, DebugLoc(), getSDVTList(VT)),
-             BA(ba), Offset(o), TargetFlags(Flags) {}
+      : SDNode(NodeTy, 0, DebugLoc(), VTs), BA(ba), Offset(o),
+        TargetFlags(Flags) {}
 
 public:
   const BlockAddress *getBlockAddress() const { return BA; }
@@ -2292,9 +2305,10 @@ class ExternalSymbolSDNode : public SDNode {
   const char *Symbol;
   unsigned TargetFlags;
 
-  ExternalSymbolSDNode(bool isTarget, const char *Sym, unsigned TF, EVT VT)
+  ExternalSymbolSDNode(bool isTarget, const char *Sym, unsigned TF,
+                       SDVTList VTs)
       : SDNode(isTarget ? ISD::TargetExternalSymbol : ISD::ExternalSymbol, 0,
-               DebugLoc(), getSDVTList(VT)),
+               DebugLoc(), VTs),
         Symbol(Sym), TargetFlags(TF) {}
 
 public:
@@ -2312,8 +2326,8 @@ class MCSymbolSDNode : public SDNode {
 
   MCSymbol *Symbol;
 
-  MCSymbolSDNode(MCSymbol *Symbol, EVT VT)
-      : SDNode(ISD::MCSymbol, 0, DebugLoc(), getSDVTList(VT)), Symbol(Symbol) {}
+  MCSymbolSDNode(MCSymbol *Symbol, SDVTList VTs)
+      : SDNode(ISD::MCSymbol, 0, DebugLoc(), VTs), Symbol(Symbol) {}
 
 public:
   MCSymbol *getMCSymbol() const { return Symbol; }
@@ -3026,8 +3040,8 @@ class AssertAlignSDNode : public SDNode {
   Align Alignment;
 
 public:
-  AssertAlignSDNode(unsigned Order, const DebugLoc &DL, EVT VT, Align A)
-      : SDNode(ISD::AssertAlign, Order, DL, getSDVTList(VT)), Alignment(A) {}
+  AssertAlignSDNode(unsigned Order, const DebugLoc &DL, SDVTList VTs, Align A)
+      : SDNode(ISD::AssertAlign, Order, DL, VTs), Alignment(A) {}
 
   Align getAlign() const { return Alignment; }
 
diff --git a/llvm/include/llvm/CodeGen/TargetInstrInfo.h b/llvm/include/llvm/CodeGen/TargetInstrInfo.h
index d4a83e3753d9..d5b1df2114e9 100644
--- a/llvm/include/llvm/CodeGen/TargetInstrInfo.h
+++ b/llvm/include/llvm/CodeGen/TargetInstrInfo.h
@@ -31,6 +31,7 @@
 #include "llvm/MC/MCInstrInfo.h"
 #include "llvm/Support/BranchProbability.h"
 #include "llvm/Support/ErrorHandling.h"
+#include <array>
 #include <cassert>
 #include <cstddef>
 #include <cstdint>
@@ -1271,11 +1272,20 @@ public:
     return true;
   }
 
+  /// The returned array encodes the operand index for each parameter because
+  /// the operands may be commuted; the operand indices for associative
+  /// operations might also be target-specific. Each element specifies the index
+  /// of {Prev, A, B, X, Y}.
+  virtual void
+  getReassociateOperandIndices(const MachineInstr &Root, unsigned Pattern,
+                               std::array<unsigned, 5> &OperandIndices) const;
+
   /// Attempt to reassociate \P Root and \P Prev according to \P Pattern to
   /// reduce critical path length.
   void reassociateOps(MachineInstr &Root, MachineInstr &Prev, unsigned Pattern,
                       SmallVectorImpl<MachineInstr *> &InsInstrs,
                       SmallVectorImpl<MachineInstr *> &DelInstrs,
+                      ArrayRef<unsigned> OperandIndices,
                       DenseMap<unsigned, unsigned> &InstrIdxForVirtReg) const;
 
   /// Reassociation of some instructions requires inverse operations (e.g.
diff --git a/llvm/include/llvm/CodeGen/TargetLowering.h b/llvm/include/llvm/CodeGen/TargetLowering.h
index 2dd978c7b584..661b2841c6ac 100644
--- a/llvm/include/llvm/CodeGen/TargetLowering.h
+++ b/llvm/include/llvm/CodeGen/TargetLowering.h
@@ -3146,7 +3146,7 @@ public:
 
   /// Lower a deinterleave intrinsic to a target specific load intrinsic.
   /// Return true on success. Currently only supports
-  /// llvm.experimental.vector.deinterleave2
+  /// llvm.vector.deinterleave2
   ///
   /// \p DI is the deinterleave intrinsic.
   /// \p LI is the accompanying load instruction
@@ -3157,7 +3157,7 @@ public:
 
   /// Lower an interleave intrinsic to a target specific store intrinsic.
   /// Return true on success. Currently only supports
-  /// llvm.experimental.vector.interleave2
+  /// llvm.vector.interleave2
   ///
   /// \p II is the interleave intrinsic.
   /// \p SI is the accompanying store instruction
@@ -5238,6 +5238,9 @@ public:
   /// Expand fminnum/fmaxnum into fminnum_ieee/fmaxnum_ieee with quieted inputs.
   SDValue expandFMINNUM_FMAXNUM(SDNode *N, SelectionDAG &DAG) const;
 
+  /// Expand fminimum/fmaximum into multiple comparison with selects.
+  SDValue expandFMINIMUM_FMAXIMUM(SDNode *N, SelectionDAG &DAG) const;
+
   /// Expand FP_TO_[US]INT_SAT into FP_TO_[US]INT and selects or min/max.
   /// \param N Node to expand
   /// \returns The expansion result
diff --git a/llvm/include/llvm/DebugInfo/DWARF/DWARFUnit.h b/llvm/include/llvm/DebugInfo/DWARF/DWARFUnit.h
index f20e71781f46..80c27aea8931 100644
--- a/llvm/include/llvm/DebugInfo/DWARF/DWARFUnit.h
+++ b/llvm/include/llvm/DebugInfo/DWARF/DWARFUnit.h
@@ -85,7 +85,7 @@ public:
                 uint64_t *offset_ptr, DWARFSectionKind SectionKind);
   // For units in DWARF Package File, remember the index entry and update
   // the abbreviation offset read by extract().
-  bool applyIndexEntry(const DWARFUnitIndex::Entry *Entry);
+  Error applyIndexEntry(const DWARFUnitIndex::Entry *Entry);
   uint64_t getOffset() const { return Offset; }
   const dwarf::FormParams &getFormParams() const { return FormParams; }
   uint16_t getVersion() const { return FormParams.Version; }
diff --git a/llvm/include/llvm/ExecutionEngine/Orc/Core.h b/llvm/include/llvm/ExecutionEngine/Orc/Core.h
index 7121b3fe7627..bac923aba02a 100644
--- a/llvm/include/llvm/ExecutionEngine/Orc/Core.h
+++ b/llvm/include/llvm/ExecutionEngine/Orc/Core.h
@@ -1443,9 +1443,6 @@ public:
   /// Send a result to the remote.
   using SendResultFunction = unique_function<void(shared::WrapperFunctionResult)>;
 
-  /// For dispatching ORC tasks (typically materialization tasks).
-  using DispatchTaskFunction = unique_function<void(std::unique_ptr<Task> T)>;
-
   /// An asynchronous wrapper-function callable from the executor via
   /// jit-dispatch.
   using JITDispatchHandlerFunction = unique_function<void(
@@ -1568,12 +1565,6 @@ public:
   /// Unhandled errors can be sent here to log them.
   void reportError(Error Err) { ReportError(std::move(Err)); }
 
-  /// Set the task dispatch function.
-  ExecutionSession &setDispatchTask(DispatchTaskFunction DispatchTask) {
-    this->DispatchTask = std::move(DispatchTask);
-    return *this;
-  }
-
   /// Search the given JITDylibs to find the flags associated with each of the
   /// given symbols.
   void lookupFlags(LookupKind K, JITDylibSearchOrder SearchOrder,
@@ -1648,7 +1639,7 @@ public:
   void dispatchTask(std::unique_ptr<Task> T) {
     assert(T && "T must be non-null");
     DEBUG_WITH_TYPE("orc", dumpDispatchInfo(*T));
-    DispatchTask(std::move(T));
+    EPC->getDispatcher().dispatch(std::move(T));
   }
 
   /// Run a wrapper function in the executor.
@@ -1762,8 +1753,6 @@ private:
     logAllUnhandledErrors(std::move(Err), errs(), "JIT session error: ");
   }
 
-  static void runOnCurrentThread(std::unique_ptr<Task> T) { T->run(); }
-
   void dispatchOutstandingMUs();
 
   static std::unique_ptr<MaterializationResponsibility>
@@ -1869,7 +1858,6 @@ private:
   std::unique_ptr<ExecutorProcessControl> EPC;
   std::unique_ptr<Platform> P;
   ErrorReporter ReportError = logErrorsToStdErr;
-  DispatchTaskFunction DispatchTask = runOnCurrentThread;
 
   std::vector<ResourceManager *> ResourceManagers;
 
diff --git a/llvm/include/llvm/ExecutionEngine/Orc/LLJIT.h b/llvm/include/llvm/ExecutionEngine/Orc/LLJIT.h
index 810a38f4a6ac..3a71ddc88ce9 100644
--- a/llvm/include/llvm/ExecutionEngine/Orc/LLJIT.h
+++ b/llvm/include/llvm/ExecutionEngine/Orc/LLJIT.h
@@ -254,7 +254,6 @@ protected:
 
   DataLayout DL;
   Triple TT;
-  std::unique_ptr<DefaultThreadPool> CompileThreads;
 
   std::unique_ptr<ObjectLayer> ObjLinkingLayer;
   std::unique_ptr<ObjectTransformLayer> ObjTransformLayer;
@@ -325,6 +324,7 @@ public:
   PlatformSetupFunction SetUpPlatform;
   NotifyCreatedFunction NotifyCreated;
   unsigned NumCompileThreads = 0;
+  std::optional<bool> SupportConcurrentCompilation;
 
   /// Called prior to JIT class construcion to fix up defaults.
   Error prepareForConstruction();
@@ -333,7 +333,7 @@ public:
 template <typename JITType, typename SetterImpl, typename State>
 class LLJITBuilderSetters {
 public:
-  /// Set a ExecutorProcessControl for this instance.
+  /// Set an ExecutorProcessControl for this instance.
   /// This should not be called if ExecutionSession has already been set.
   SetterImpl &
   setExecutorProcessControl(std::unique_ptr<ExecutorProcessControl> EPC) {
@@ -462,19 +462,26 @@ public:
   ///
   /// If this method is not called, behavior will be as if it were called with
   /// a zero argument.
+  ///
+  /// This setting should not be used if a custom ExecutionSession or
+  /// ExecutorProcessControl object is set: in those cases a custom
+  /// TaskDispatcher should be used instead.
   SetterImpl &setNumCompileThreads(unsigned NumCompileThreads) {
     impl().NumCompileThreads = NumCompileThreads;
     return impl();
   }
 
-  /// Set an ExecutorProcessControl object.
+  /// If set, this forces LLJIT concurrent compilation support to be either on
+  /// or off. This controls the selection of compile function (concurrent vs
+  /// single threaded) and whether or not sub-modules are cloned to new
+  /// contexts for lazy emission.
   ///
-  /// If the platform uses ObjectLinkingLayer by default and no
-  /// ObjectLinkingLayerCreator has been set then the ExecutorProcessControl
-  /// object will be used to supply the memory manager for the
-  /// ObjectLinkingLayer.
-  SetterImpl &setExecutorProcessControl(ExecutorProcessControl &EPC) {
-    impl().EPC = &EPC;
+  /// If not explicitly set then concurrency support will be turned on if
+  /// NumCompileThreads is set to a non-zero value, or if a custom
+  /// ExecutionSession or ExecutorProcessControl instance is provided.
+  SetterImpl &setSupportConcurrentCompilation(
+      std::optional<bool> SupportConcurrentCompilation) {
+    impl().SupportConcurrentCompilation = SupportConcurrentCompilation;
     return impl();
   }
 
diff --git a/llvm/include/llvm/ExecutionEngine/Orc/TaskDispatch.h b/llvm/include/llvm/ExecutionEngine/Orc/TaskDispatch.h
index 8c287f9fec0e..8c65677aae25 100644
--- a/llvm/include/llvm/ExecutionEngine/Orc/TaskDispatch.h
+++ b/llvm/include/llvm/ExecutionEngine/Orc/TaskDispatch.h
@@ -23,6 +23,7 @@
 
 #if LLVM_ENABLE_THREADS
 #include <condition_variable>
+#include <deque>
 #include <mutex>
 #include <thread>
 #endif
@@ -114,6 +115,9 @@ public:
 
 class DynamicThreadPoolTaskDispatcher : public TaskDispatcher {
 public:
+  DynamicThreadPoolTaskDispatcher(
+      std::optional<size_t> MaxMaterializationThreads)
+      : MaxMaterializationThreads(MaxMaterializationThreads) {}
   void dispatch(std::unique_ptr<Task> T) override;
   void shutdown() override;
 private:
@@ -121,6 +125,10 @@ private:
   bool Running = true;
   size_t Outstanding = 0;
   std::condition_variable OutstandingCV;
+
+  std::optional<size_t> MaxMaterializationThreads;
+  size_t NumMaterializationThreads = 0;
+  std::deque<std::unique_ptr<Task>> MaterializationTaskQueue;
 };
 
 #endif // LLVM_ENABLE_THREADS
diff --git a/llvm/include/llvm/Frontend/OpenMP/OMP.h b/llvm/include/llvm/Frontend/OpenMP/OMP.h
index 4ed47f15dfe5..6f7a39acac1d 100644
--- a/llvm/include/llvm/Frontend/OpenMP/OMP.h
+++ b/llvm/include/llvm/Frontend/OpenMP/OMP.h
@@ -16,10 +16,20 @@
 #include "llvm/Frontend/OpenMP/OMP.h.inc"
 
 #include "llvm/ADT/ArrayRef.h"
+#include "llvm/ADT/SmallVector.h"
 
 namespace llvm::omp {
 ArrayRef<Directive> getLeafConstructs(Directive D);
+ArrayRef<Directive> getLeafConstructsOrSelf(Directive D);
+
+ArrayRef<Directive>
+getLeafOrCompositeConstructs(Directive D, SmallVectorImpl<Directive> &Output);
+
 Directive getCompoundConstruct(ArrayRef<Directive> Parts);
+
+bool isLeafConstruct(Directive D);
+bool isCompositeConstruct(Directive D);
+bool isCombinedConstruct(Directive D);
 } // namespace llvm::omp
 
 #endif // LLVM_FRONTEND_OPENMP_OMP_H
diff --git a/llvm/include/llvm/Frontend/OpenMP/OMPKinds.def b/llvm/include/llvm/Frontend/OpenMP/OMPKinds.def
index d22d2a8e948b..fe09bb8177c2 100644
--- a/llvm/include/llvm/Frontend/OpenMP/OMPKinds.def
+++ b/llvm/include/llvm/Frontend/OpenMP/OMPKinds.def
@@ -436,7 +436,6 @@ __OMP_RTL(__tgt_target_kernel, false, Int32, IdentPtr, Int64, Int32, Int32,
           VoidPtr, KernelArgsPtr)
 __OMP_RTL(__tgt_target_kernel_nowait, false, Int32, IdentPtr, Int64, Int32,
           Int32, VoidPtr, KernelArgsPtr, Int32, VoidPtr, Int32, VoidPtr)
-__OMP_RTL(__tgt_register_requires, false, Void, Int64)
 __OMP_RTL(__tgt_target_data_begin_mapper, false, Void, IdentPtr, Int64, Int32, VoidPtrPtr,
           VoidPtrPtr, Int64Ptr, Int64Ptr, VoidPtrPtr, VoidPtrPtr)
 __OMP_RTL(__tgt_target_data_begin_nowait_mapper, false, Void, IdentPtr, Int64, Int32,
@@ -1025,8 +1024,6 @@ __OMP_RTL_ATTRS(__tgt_target_kernel_nowait, ForkAttrs, SExt,
                 ParamAttrs(AttributeSet(), AttributeSet(), SExt, SExt,
                            AttributeSet(), AttributeSet(), SExt, AttributeSet(),
                            SExt))
-__OMP_RTL_ATTRS(__tgt_register_requires, ForkAttrs, AttributeSet(),
-                ParamAttrs())
 __OMP_RTL_ATTRS(__tgt_target_data_begin_mapper, ForkAttrs, AttributeSet(),
                 ParamAttrs(AttributeSet(), AttributeSet(), SExt))
 __OMP_RTL_ATTRS(__tgt_target_data_begin_nowait_mapper, ForkAttrs, AttributeSet(),
diff --git a/llvm/include/llvm/IR/Attributes.h b/llvm/include/llvm/IR/Attributes.h
index 7dd8a329029a..5e3ba1f32e6a 100644
--- a/llvm/include/llvm/IR/Attributes.h
+++ b/llvm/include/llvm/IR/Attributes.h
@@ -747,6 +747,11 @@ public:
   addDereferenceableOrNullParamAttr(LLVMContext &C, unsigned ArgNo,
                                     uint64_t Bytes) const;
 
+  /// Add the range attribute to the attribute set at the return value index.
+  /// Returns a new list because attribute lists are immutable.
+  [[nodiscard]] AttributeList addRangeRetAttr(LLVMContext &C,
+                                              const ConstantRange &CR) const;
+
   /// Add the allocsize attribute to the attribute set at the given arg index.
   /// Returns a new list because attribute lists are immutable.
   [[nodiscard]] AttributeList
diff --git a/llvm/include/llvm/IR/EHPersonalities.h b/llvm/include/llvm/IR/EHPersonalities.h
index bd768440bfb9..c70f832de40b 100644
--- a/llvm/include/llvm/IR/EHPersonalities.h
+++ b/llvm/include/llvm/IR/EHPersonalities.h
@@ -32,7 +32,8 @@ enum class EHPersonality {
   CoreCLR,
   Rust,
   Wasm_CXX,
-  XL_CXX
+  XL_CXX,
+  ZOS_CXX,
 };
 
 /// See if the given exception handling personality function is one
diff --git a/llvm/include/llvm/IR/FixedMetadataKinds.def b/llvm/include/llvm/IR/FixedMetadataKinds.def
index b375d0f09120..5f4cc230a0f5 100644
--- a/llvm/include/llvm/IR/FixedMetadataKinds.def
+++ b/llvm/include/llvm/IR/FixedMetadataKinds.def
@@ -51,3 +51,4 @@ LLVM_FIXED_MD_KIND(MD_kcfi_type, "kcfi_type", 36)
 LLVM_FIXED_MD_KIND(MD_pcsections, "pcsections", 37)
 LLVM_FIXED_MD_KIND(MD_DIAssignID, "DIAssignID", 38)
 LLVM_FIXED_MD_KIND(MD_coro_outside_frame, "coro.outside.frame", 39)
+LLVM_FIXED_MD_KIND(MD_mmra, "mmra", 40)
diff --git a/llvm/include/llvm/IR/Function.h b/llvm/include/llvm/IR/Function.h
index 60f41b30e91c..cb514cde95b5 100644
--- a/llvm/include/llvm/IR/Function.h
+++ b/llvm/include/llvm/IR/Function.h
@@ -46,6 +46,7 @@ typedef unsigned ID;
 
 class AssemblyAnnotationWriter;
 class Constant;
+class ConstantRange;
 struct DenormalMode;
 class DISubprogram;
 enum LibFunc : unsigned;
@@ -462,6 +463,9 @@ public:
   /// attributes for the given arg.
   void addDereferenceableOrNullParamAttr(unsigned ArgNo, uint64_t Bytes);
 
+  /// adds the range attribute to the list of attributes for the return value.
+  void addRangeRetAttr(const ConstantRange &CR);
+
   MaybeAlign getParamAlign(unsigned ArgNo) const {
     return AttributeSets.getParamAlignment(ArgNo);
   }
diff --git a/llvm/include/llvm/IR/GenericConvergenceVerifierImpl.h b/llvm/include/llvm/IR/GenericConvergenceVerifierImpl.h
index a3ebde709ae6..7525c9eb758b 100644
--- a/llvm/include/llvm/IR/GenericConvergenceVerifierImpl.h
+++ b/llvm/include/llvm/IR/GenericConvergenceVerifierImpl.h
@@ -76,7 +76,7 @@ void GenericConvergenceVerifier<ContextT>::visit(const InstructionT &I) {
           "Entry intrinsic cannot be preceded by a convergent operation in the "
           "same basic block.",
           {Context.print(&I)});
-    LLVM_FALLTHROUGH;
+    [[fallthrough]];
   case CONV_ANCHOR:
     Check(!TokenDef,
           "Entry or anchor intrinsic cannot have a convergencectrl token "
diff --git a/llvm/include/llvm/IR/InstrTypes.h b/llvm/include/llvm/IR/InstrTypes.h
index 8e6bef69218c..b9af3a6ca42c 100644
--- a/llvm/include/llvm/IR/InstrTypes.h
+++ b/llvm/include/llvm/IR/InstrTypes.h
@@ -1941,6 +1941,11 @@ public:
     Attrs = Attrs.addDereferenceableRetAttr(getContext(), Bytes);
   }
 
+  /// adds the range attribute to the list of attributes.
+  void addRangeRetAttr(const ConstantRange &CR) {
+    Attrs = Attrs.addRangeRetAttr(getContext(), CR);
+  }
+
   /// Determine whether the return value has the given attribute.
   bool hasRetAttr(Attribute::AttrKind Kind) const {
     return hasRetAttrImpl(Kind);
diff --git a/llvm/include/llvm/IR/IntrinsicInst.h b/llvm/include/llvm/IR/IntrinsicInst.h
index 4f22720f1c55..2e99c9e2ee3e 100644
--- a/llvm/include/llvm/IR/IntrinsicInst.h
+++ b/llvm/include/llvm/IR/IntrinsicInst.h
@@ -311,7 +311,8 @@ public:
 
   Value *getVariableLocationOp(unsigned OpIdx) const;
 
-  void replaceVariableLocationOp(Value *OldValue, Value *NewValue);
+  void replaceVariableLocationOp(Value *OldValue, Value *NewValue,
+                                 bool AllowEmpty = false);
   void replaceVariableLocationOp(unsigned OpIdx, Value *NewValue);
   /// Adding a new location operand will always result in this intrinsic using
   /// an ArgList, and must always be accompanied by a new expression that uses
@@ -1435,6 +1436,7 @@ protected:
     case Intrinsic::instrprof_cover:
     case Intrinsic::instrprof_increment:
     case Intrinsic::instrprof_increment_step:
+    case Intrinsic::instrprof_callsite:
     case Intrinsic::instrprof_timestamp:
     case Intrinsic::instrprof_value_profile:
       return true;
@@ -1519,6 +1521,21 @@ public:
   }
 };
 
+/// This represents the llvm.instrprof.callsite intrinsic.
+/// It is structurally like the increment or step counters, hence the
+/// inheritance relationship, albeit somewhat tenuous (it's not 'counting' per
+/// se)
+class InstrProfCallsite : public InstrProfCntrInstBase {
+public:
+  static bool classof(const IntrinsicInst *I) {
+    return I->getIntrinsicID() == Intrinsic::instrprof_callsite;
+  }
+  static bool classof(const Value *V) {
+    return isa<IntrinsicInst>(V) && classof(cast<IntrinsicInst>(V));
+  }
+  Value *getCallee() const;
+};
+
 /// This represents the llvm.instrprof.timestamp intrinsic.
 class InstrProfTimestampInst : public InstrProfCntrInstBase {
 public:
diff --git a/llvm/include/llvm/IR/Intrinsics.td b/llvm/include/llvm/IR/Intrinsics.td
index 1d20f7e1b198..a2678d69ce40 100644
--- a/llvm/include/llvm/IR/Intrinsics.td
+++ b/llvm/include/llvm/IR/Intrinsics.td
@@ -914,6 +914,11 @@ def int_instrprof_increment_step : Intrinsic<[],
                                         [llvm_ptr_ty, llvm_i64_ty,
                                          llvm_i32_ty, llvm_i32_ty, llvm_i64_ty]>;
 
+// Callsite instrumentation for contextual profiling
+def int_instrprof_callsite : Intrinsic<[],
+                                        [llvm_ptr_ty, llvm_i64_ty,
+                                         llvm_i32_ty, llvm_i32_ty, llvm_ptr_ty]>;
+
 // A timestamp for instrumentation based profiling.
 def int_instrprof_timestamp : Intrinsic<[], [llvm_ptr_ty, llvm_i64_ty,
                                              llvm_i32_ty, llvm_i32_ty]>;
@@ -2572,15 +2577,15 @@ def int_preserve_static_offset : DefaultAttrsIntrinsic<[llvm_ptr_ty],
 
 //===------------ Intrinsics to perform common vector shuffles ------------===//
 
-def int_experimental_vector_reverse : DefaultAttrsIntrinsic<[llvm_anyvector_ty],
-                                                            [LLVMMatchType<0>],
-                                                            [IntrNoMem]>;
+def int_vector_reverse : DefaultAttrsIntrinsic<[llvm_anyvector_ty],
+                                               [LLVMMatchType<0>],
+                                               [IntrNoMem]>;
 
-def int_experimental_vector_splice : DefaultAttrsIntrinsic<[llvm_anyvector_ty],
-                                                           [LLVMMatchType<0>,
-                                                            LLVMMatchType<0>,
-                                                            llvm_i32_ty],
-                                                           [IntrNoMem, ImmArg<ArgIndex<2>>]>;
+def int_vector_splice : DefaultAttrsIntrinsic<[llvm_anyvector_ty],
+                                              [LLVMMatchType<0>,
+                                               LLVMMatchType<0>,
+                                               llvm_i32_ty],
+                                              [IntrNoMem, ImmArg<ArgIndex<2>>]>;
 
 //===---------- Intrinsics to query properties of scalable vectors --------===//
 def int_vscale : DefaultAttrsIntrinsic<[llvm_anyint_ty], [], [IntrNoMem]>;
@@ -2595,15 +2600,15 @@ def int_vector_extract : DefaultAttrsIntrinsic<[llvm_anyvector_ty],
                                                [IntrNoMem, IntrSpeculatable, ImmArg<ArgIndex<1>>]>;
 
 
-def int_experimental_vector_interleave2   : DefaultAttrsIntrinsic<[llvm_anyvector_ty],
-                                                                  [LLVMHalfElementsVectorType<0>,
-                                                                   LLVMHalfElementsVectorType<0>],
-                                                                  [IntrNoMem]>;
+def int_vector_interleave2   : DefaultAttrsIntrinsic<[llvm_anyvector_ty],
+                                                     [LLVMHalfElementsVectorType<0>,
+                                                      LLVMHalfElementsVectorType<0>],
+                                                     [IntrNoMem]>;
 
-def int_experimental_vector_deinterleave2 : DefaultAttrsIntrinsic<[LLVMHalfElementsVectorType<0>,
-                                                                   LLVMHalfElementsVectorType<0>],
-                                                                  [llvm_anyvector_ty],
-                                                                  [IntrNoMem]>;
+def int_vector_deinterleave2 : DefaultAttrsIntrinsic<[LLVMHalfElementsVectorType<0>,
+                                                      LLVMHalfElementsVectorType<0>],
+                                                     [llvm_anyvector_ty],
+                                                     [IntrNoMem]>;
 
 //===----------------- Pointer Authentication Intrinsics ------------------===//
 //
diff --git a/llvm/include/llvm/IR/IntrinsicsAArch64.td b/llvm/include/llvm/IR/IntrinsicsAArch64.td
index bcaa37de74b6..e31e00a9c76f 100644
--- a/llvm/include/llvm/IR/IntrinsicsAArch64.td
+++ b/llvm/include/llvm/IR/IntrinsicsAArch64.td
@@ -1762,6 +1762,7 @@ def int_aarch64_sve_uqsub_x   : AdvSIMD_2VectorArg_Intrinsic;
 def int_aarch64_sve_orqv      : AdvSIMD_SVE_V128_Reduce_Intrinsic;
 def int_aarch64_sve_eorqv     : AdvSIMD_SVE_V128_Reduce_Intrinsic;
 def int_aarch64_sve_andqv     : AdvSIMD_SVE_V128_Reduce_Intrinsic;
+def int_aarch64_sve_addqv     : AdvSIMD_SVE_V128_Reduce_Intrinsic;
 def int_aarch64_sve_smaxqv    : AdvSIMD_SVE_V128_Reduce_Intrinsic;
 def int_aarch64_sve_umaxqv    : AdvSIMD_SVE_V128_Reduce_Intrinsic;
 def int_aarch64_sve_sminqv    : AdvSIMD_SVE_V128_Reduce_Intrinsic;
@@ -2079,11 +2080,12 @@ def int_aarch64_sve_fmaxv   : AdvSIMD_SVE_Reduce_Intrinsic;
 def int_aarch64_sve_fmaxnmv : AdvSIMD_SVE_Reduce_Intrinsic;
 def int_aarch64_sve_fminv   : AdvSIMD_SVE_Reduce_Intrinsic;
 def int_aarch64_sve_fminnmv : AdvSIMD_SVE_Reduce_Intrinsic;
-def int_aarch64_sve_addqv : AdvSIMD_SVE_V128_Reduce_Intrinsic;
+
+def int_aarch64_sve_faddqv   : AdvSIMD_SVE_V128_Reduce_Intrinsic;
 def int_aarch64_sve_fmaxnmqv : AdvSIMD_SVE_V128_Reduce_Intrinsic;
 def int_aarch64_sve_fminnmqv : AdvSIMD_SVE_V128_Reduce_Intrinsic;
-def int_aarch64_sve_fmaxqv : AdvSIMD_SVE_V128_Reduce_Intrinsic;
-def int_aarch64_sve_fminqv : AdvSIMD_SVE_V128_Reduce_Intrinsic;
+def int_aarch64_sve_fmaxqv   : AdvSIMD_SVE_V128_Reduce_Intrinsic;
+def int_aarch64_sve_fminqv   : AdvSIMD_SVE_V128_Reduce_Intrinsic;
 
 //
 // Floating-point conversions
@@ -3646,4 +3648,4 @@ def int_aarch64_sve_pmov_to_pred_lane_zero : SVE2_1VectorArg_Pred_Intrinsic;
 
 def int_aarch64_sve_pmov_to_vector_lane_merging : SVE2_Pred_1VectorArgIndexed_Intrinsic;
    
-def int_aarch64_sve_pmov_to_vector_lane_zeroing : SVE2_Pred_1VectorArg_Intrinsic;
\ No newline at end of file
+def int_aarch64_sve_pmov_to_vector_lane_zeroing : SVE2_Pred_1VectorArg_Intrinsic;
diff --git a/llvm/include/llvm/IR/IntrinsicsNVVM.td b/llvm/include/llvm/IR/IntrinsicsNVVM.td
index 726cea004606..0a9139e0062b 100644
--- a/llvm/include/llvm/IR/IntrinsicsNVVM.td
+++ b/llvm/include/llvm/IR/IntrinsicsNVVM.td
@@ -1003,6 +1003,8 @@ let TargetPrefix = "nvvm" in {
 
   def int_nvvm_rsqrt_approx_ftz_f : ClangBuiltin<"__nvvm_rsqrt_approx_ftz_f">,
       DefaultAttrsIntrinsic<[llvm_float_ty], [llvm_float_ty], [IntrNoMem]>;
+  def int_nvvm_rsqrt_approx_ftz_d : ClangBuiltin<"__nvvm_rsqrt_approx_ftz_d">,
+      DefaultAttrsIntrinsic<[llvm_double_ty], [llvm_double_ty], [IntrNoMem]>;
   def int_nvvm_rsqrt_approx_f : ClangBuiltin<"__nvvm_rsqrt_approx_f">,
       DefaultAttrsIntrinsic<[llvm_float_ty], [llvm_float_ty], [IntrNoMem]>;
   def int_nvvm_rsqrt_approx_d : ClangBuiltin<"__nvvm_rsqrt_approx_d">,
diff --git a/llvm/include/llvm/IR/MemoryModelRelaxationAnnotations.h b/llvm/include/llvm/IR/MemoryModelRelaxationAnnotations.h
new file mode 100644
index 000000000000..a9ded6034d0b
--- /dev/null
+++ b/llvm/include/llvm/IR/MemoryModelRelaxationAnnotations.h
@@ -0,0 +1,132 @@
+//===- MemoryModelRelaxationAnnotations.h -----------------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+/// \file
+/// This file provides utility for Memory Model Relaxation Annotations (MMRAs).
+/// Those annotations are represented using Metadata. The MMRATagSet class
+/// offers a simple API to parse the metadata and perform common operations on
+/// it. The MMRAMetadata class is a simple tuple of MDNode that provides easy
+/// access to all MMRA annotations on an instruction.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_IR_MEMORYMODELRELAXATIONANNOTATIONS_H
+#define LLVM_IR_MEMORYMODELRELAXATIONANNOTATIONS_H
+
+#include "llvm/ADT/DenseSet.h"
+#include "llvm/ADT/StringRef.h"
+#include <tuple> // for std::pair
+
+namespace llvm {
+
+template <typename T> class ArrayRef;
+
+class MDNode;
+class MDTuple;
+class Metadata;
+class raw_ostream;
+class LLVMContext;
+class Instruction;
+
+/// Helper class to manipulate `!mmra` metadata nodes.
+///
+/// This can be visualized as a set of "tags", with each tag
+/// representing a particular property of an instruction, as
+/// explained in the MemoryModelRelaxationAnnotations docs.
+///
+/// This class (and the optimizer in general) does not reason
+/// about the exact nature of the tags and the properties they
+/// imply. It just sees the metadata as a collection of tags, which
+/// are a prefix/suffix pair of strings.
+class MMRAMetadata {
+public:
+  using TagT = std::pair<StringRef, StringRef>;
+  using SetT = DenseSet<TagT>;
+  using const_iterator = SetT::const_iterator;
+
+  /// \name Constructors
+  /// @{
+  MMRAMetadata() = default;
+  MMRAMetadata(const Instruction &I);
+  MMRAMetadata(MDNode *MD);
+  /// @}
+
+  /// \name Metadata Helpers & Builders
+  /// @{
+
+  /// Combines \p A and \p B according to MMRA semantics.
+  /// \returns !mmra metadata for the combined MMRAs.
+  static MDNode *combine(LLVMContext &Ctx, const MMRAMetadata &A,
+                         const MMRAMetadata &B);
+
+  /// Creates !mmra metadata for a single tag.
+  ///
+  /// !mmra metadata can either be a single tag, or a MDTuple containing
+  /// multiple tags.
+  static MDTuple *getTagMD(LLVMContext &Ctx, StringRef Prefix,
+                           StringRef Suffix);
+  static MDTuple *getTagMD(LLVMContext &Ctx, const TagT &T) {
+    return getTagMD(Ctx, T.first, T.second);
+  }
+
+  /// Creates !mmra metadata from \p Tags.
+  /// \returns nullptr or a MDTuple* from \p Tags.
+  static MDTuple *getMD(LLVMContext &Ctx, ArrayRef<TagT> Tags);
+
+  /// \returns true if \p MD is a well-formed MMRA tag.
+  static bool isTagMD(const Metadata *MD);
+
+  /// @}
+
+  /// \name Compatibility Helpers
+  /// @{
+
+  /// \returns whether the MMRAs on \p A and \p B are compatible.
+  static bool checkCompatibility(const Instruction &A, const Instruction &B) {
+    return MMRAMetadata(A).isCompatibleWith(B);
+  }
+
+  /// \returns whether this set of tags is compatible with \p Other.
+  bool isCompatibleWith(const MMRAMetadata &Other) const;
+
+  /// @}
+
+  /// \name Content Queries
+  /// @{
+
+  bool hasTag(StringRef Prefix, StringRef Suffix) const;
+  bool hasTagWithPrefix(StringRef Prefix) const;
+
+  const_iterator begin() const;
+  const_iterator end() const;
+  bool empty() const;
+  unsigned size() const;
+
+  /// @}
+
+  void print(raw_ostream &OS) const;
+  void dump() const;
+
+  operator bool() const { return !Tags.empty(); }
+  bool operator==(const MMRAMetadata &Other) const {
+    return Tags == Other.Tags;
+  }
+  bool operator!=(const MMRAMetadata &Other) const {
+    return Tags != Other.Tags;
+  }
+
+private:
+  SetT Tags;
+};
+
+/// \returns true if \p I can have !mmra metadata.
+bool canInstructionHaveMMRAs(const Instruction &I);
+
+} // namespace llvm
+
+#endif
diff --git a/llvm/include/llvm/IR/PatternMatch.h b/llvm/include/llvm/IR/PatternMatch.h
index 1fee1901fabb..739208e74dcb 100644
--- a/llvm/include/llvm/IR/PatternMatch.h
+++ b/llvm/include/llvm/IR/PatternMatch.h
@@ -350,8 +350,9 @@ template <int64_t Val> inline constantint_match<Val> m_ConstantInt() {
 
 /// This helper class is used to match constant scalars, vector splats,
 /// and fixed width vectors that satisfy a specified predicate.
-/// For fixed width vector constants, poison elements are ignored.
-template <typename Predicate, typename ConstantVal>
+/// For fixed width vector constants, poison elements are ignored if AllowPoison
+/// is true.
+template <typename Predicate, typename ConstantVal, bool AllowPoison>
 struct cstval_pred_ty : public Predicate {
   template <typename ITy> bool match(ITy *V) {
     if (const auto *CV = dyn_cast<ConstantVal>(V))
@@ -374,7 +375,7 @@ struct cstval_pred_ty : public Predicate {
           Constant *Elt = C->getAggregateElement(i);
           if (!Elt)
             return false;
-          if (isa<PoisonValue>(Elt))
+          if (AllowPoison && isa<PoisonValue>(Elt))
             continue;
           auto *CV = dyn_cast<ConstantVal>(Elt);
           if (!CV || !this->isValue(CV->getValue()))
@@ -389,12 +390,13 @@ struct cstval_pred_ty : public Predicate {
 };
 
 /// specialization of cstval_pred_ty for ConstantInt
-template <typename Predicate>
-using cst_pred_ty = cstval_pred_ty<Predicate, ConstantInt>;
+template <typename Predicate, bool AllowPoison = true>
+using cst_pred_ty = cstval_pred_ty<Predicate, ConstantInt, AllowPoison>;
 
 /// specialization of cstval_pred_ty for ConstantFP
 template <typename Predicate>
-using cstfp_pred_ty = cstval_pred_ty<Predicate, ConstantFP>;
+using cstfp_pred_ty = cstval_pred_ty<Predicate, ConstantFP,
+                                     /*AllowPoison=*/true>;
 
 /// This helper class is used to match scalar and vector constants that
 /// satisfy a specified predicate, and bind them to an APInt.
@@ -484,6 +486,10 @@ inline cst_pred_ty<is_all_ones> m_AllOnes() {
   return cst_pred_ty<is_all_ones>();
 }
 
+inline cst_pred_ty<is_all_ones, false> m_AllOnesForbidPoison() {
+  return cst_pred_ty<is_all_ones, false>();
+}
+
 struct is_maxsignedvalue {
   bool isValue(const APInt &C) { return C.isMaxSignedValue(); }
 };
@@ -2507,7 +2513,7 @@ inline typename m_Intrinsic_Ty<Opnd0, Opnd1>::Ty m_CopySign(const Opnd0 &Op0,
 
 template <typename Opnd0>
 inline typename m_Intrinsic_Ty<Opnd0>::Ty m_VecReverse(const Opnd0 &Op0) {
-  return m_Intrinsic<Intrinsic::experimental_vector_reverse>(Op0);
+  return m_Intrinsic<Intrinsic::vector_reverse>(Op0);
 }
 
 //===----------------------------------------------------------------------===//
@@ -2596,6 +2602,13 @@ m_Not(const ValTy &V) {
   return m_c_Xor(m_AllOnes(), V);
 }
 
+template <typename ValTy>
+inline BinaryOp_match<cst_pred_ty<is_all_ones, false>, ValTy, Instruction::Xor,
+                      true>
+m_NotForbidPoison(const ValTy &V) {
+  return m_c_Xor(m_AllOnesForbidPoison(), V);
+}
+
 /// Matches an SMin with LHS and RHS in either order.
 template <typename LHS, typename RHS>
 inline MaxMin_match<ICmpInst, LHS, RHS, smin_pred_ty, true>
diff --git a/llvm/include/llvm/MC/MCELFStreamer.h b/llvm/include/llvm/MC/MCELFStreamer.h
index 1309b17bff9c..1ff029d44d37 100644
--- a/llvm/include/llvm/MC/MCELFStreamer.h
+++ b/llvm/include/llvm/MC/MCELFStreamer.h
@@ -156,7 +156,7 @@ MCELFStreamer *createARMELFStreamer(MCContext &Context,
                                     std::unique_ptr<MCAsmBackend> TAB,
                                     std::unique_ptr<MCObjectWriter> OW,
                                     std::unique_ptr<MCCodeEmitter> Emitter,
-                                    bool RelaxAll, bool IsThumb, bool IsAndroid);
+                                    bool IsThumb, bool IsAndroid);
 
 } // end namespace llvm
 
diff --git a/llvm/include/llvm/MC/TargetRegistry.h b/llvm/include/llvm/MC/TargetRegistry.h
index 47051447404d..5038b87cd1dc 100644
--- a/llvm/include/llvm/MC/TargetRegistry.h
+++ b/llvm/include/llvm/MC/TargetRegistry.h
@@ -92,39 +92,33 @@ createAsmStreamer(MCContext &Ctx, std::unique_ptr<formatted_raw_ostream> OS,
 MCStreamer *createELFStreamer(MCContext &Ctx,
                               std::unique_ptr<MCAsmBackend> &&TAB,
                               std::unique_ptr<MCObjectWriter> &&OW,
-                              std::unique_ptr<MCCodeEmitter> &&CE,
-                              bool RelaxAll);
+                              std::unique_ptr<MCCodeEmitter> &&CE);
 MCStreamer *createGOFFStreamer(MCContext &Ctx,
                                std::unique_ptr<MCAsmBackend> &&TAB,
                                std::unique_ptr<MCObjectWriter> &&OW,
-                               std::unique_ptr<MCCodeEmitter> &&CE,
-                               bool RelaxAll);
+                               std::unique_ptr<MCCodeEmitter> &&CE);
 MCStreamer *createMachOStreamer(MCContext &Ctx,
                                 std::unique_ptr<MCAsmBackend> &&TAB,
                                 std::unique_ptr<MCObjectWriter> &&OW,
                                 std::unique_ptr<MCCodeEmitter> &&CE,
-                                bool RelaxAll, bool DWARFMustBeAtTheEnd,
+                                bool DWARFMustBeAtTheEnd,
                                 bool LabelSections = false);
 MCStreamer *createWasmStreamer(MCContext &Ctx,
                                std::unique_ptr<MCAsmBackend> &&TAB,
                                std::unique_ptr<MCObjectWriter> &&OW,
-                               std::unique_ptr<MCCodeEmitter> &&CE,
-                               bool RelaxAll);
+                               std::unique_ptr<MCCodeEmitter> &&CE);
 MCStreamer *createXCOFFStreamer(MCContext &Ctx,
                                 std::unique_ptr<MCAsmBackend> &&TAB,
                                 std::unique_ptr<MCObjectWriter> &&OW,
-                                std::unique_ptr<MCCodeEmitter> &&CE,
-                                bool RelaxAll);
+                                std::unique_ptr<MCCodeEmitter> &&CE);
 MCStreamer *createSPIRVStreamer(MCContext &Ctx,
                                 std::unique_ptr<MCAsmBackend> &&TAB,
                                 std::unique_ptr<MCObjectWriter> &&OW,
-                                std::unique_ptr<MCCodeEmitter> &&CE,
-                                bool RelaxAll);
+                                std::unique_ptr<MCCodeEmitter> &&CE);
 MCStreamer *createDXContainerStreamer(MCContext &Ctx,
                                       std::unique_ptr<MCAsmBackend> &&TAB,
                                       std::unique_ptr<MCObjectWriter> &&OW,
-                                      std::unique_ptr<MCCodeEmitter> &&CE,
-                                      bool RelaxAll);
+                                      std::unique_ptr<MCCodeEmitter> &&CE);
 
 MCRelocationInfo *createMCRelocationInfo(const Triple &TT, MCContext &Ctx);
 
@@ -199,42 +193,42 @@ public:
       MCStreamer *(*)(const Triple &T, MCContext &Ctx,
                       std::unique_ptr<MCAsmBackend> &&TAB,
                       std::unique_ptr<MCObjectWriter> &&OW,
-                      std::unique_ptr<MCCodeEmitter> &&Emitter, bool RelaxAll);
+                      std::unique_ptr<MCCodeEmitter> &&Emitter);
   using GOFFStreamerCtorTy =
       MCStreamer *(*)(MCContext &Ctx, std::unique_ptr<MCAsmBackend> &&TAB,
                       std::unique_ptr<MCObjectWriter> &&OW,
-                      std::unique_ptr<MCCodeEmitter> &&Emitter, bool RelaxAll);
+                      std::unique_ptr<MCCodeEmitter> &&Emitter);
   using MachOStreamerCtorTy =
       MCStreamer *(*)(MCContext &Ctx, std::unique_ptr<MCAsmBackend> &&TAB,
                       std::unique_ptr<MCObjectWriter> &&OW,
-                      std::unique_ptr<MCCodeEmitter> &&Emitter, bool RelaxAll,
+                      std::unique_ptr<MCCodeEmitter> &&Emitter,
                       bool DWARFMustBeAtTheEnd);
   using COFFStreamerCtorTy =
       MCStreamer *(*)(MCContext &Ctx, std::unique_ptr<MCAsmBackend> &&TAB,
                       std::unique_ptr<MCObjectWriter> &&OW,
-                      std::unique_ptr<MCCodeEmitter> &&Emitter, bool RelaxAll,
+                      std::unique_ptr<MCCodeEmitter> &&Emitter,
                       bool IncrementalLinkerCompatible);
   using WasmStreamerCtorTy =
       MCStreamer *(*)(const Triple &T, MCContext &Ctx,
                       std::unique_ptr<MCAsmBackend> &&TAB,
                       std::unique_ptr<MCObjectWriter> &&OW,
-                      std::unique_ptr<MCCodeEmitter> &&Emitter, bool RelaxAll);
+                      std::unique_ptr<MCCodeEmitter> &&Emitter);
   using XCOFFStreamerCtorTy =
       MCStreamer *(*)(const Triple &T, MCContext &Ctx,
                       std::unique_ptr<MCAsmBackend> &&TAB,
                       std::unique_ptr<MCObjectWriter> &&OW,
-                      std::unique_ptr<MCCodeEmitter> &&Emitter, bool RelaxAll);
+                      std::unique_ptr<MCCodeEmitter> &&Emitter);
   using SPIRVStreamerCtorTy =
       MCStreamer *(*)(const Triple &T, MCContext &Ctx,
                       std::unique_ptr<MCAsmBackend> &&TAB,
                       std::unique_ptr<MCObjectWriter> &&OW,
-                      std::unique_ptr<MCCodeEmitter> &&Emitter, bool RelaxAll);
-  
+                      std::unique_ptr<MCCodeEmitter> &&Emitter);
+
   using DXContainerStreamerCtorTy =
       MCStreamer *(*)(const Triple &T, MCContext &Ctx,
                       std::unique_ptr<MCAsmBackend> &&TAB,
                       std::unique_ptr<MCObjectWriter> &&OW,
-                      std::unique_ptr<MCCodeEmitter> &&Emitter, bool RelaxAll);
+                      std::unique_ptr<MCCodeEmitter> &&Emitter);
 
   using NullTargetStreamerCtorTy = MCTargetStreamer *(*)(MCStreamer &S);
   using AsmTargetStreamerCtorTy = MCTargetStreamer *(*)(
@@ -566,7 +560,7 @@ public:
                                      std::unique_ptr<MCAsmBackend> &&TAB,
                                      std::unique_ptr<MCObjectWriter> &&OW,
                                      std::unique_ptr<MCCodeEmitter> &&Emitter,
-                                     const MCSubtargetInfo &STI, bool RelaxAll,
+                                     const MCSubtargetInfo &STI, bool,
                                      bool IncrementalLinkerCompatible,
                                      bool DWARFMustBeAtTheEnd) const {
     MCStreamer *S = nullptr;
@@ -577,66 +571,63 @@ public:
       assert((T.isOSWindows() || T.isUEFI()) &&
              "only Windows and UEFI COFF are supported");
       S = COFFStreamerCtorFn(Ctx, std::move(TAB), std::move(OW),
-                             std::move(Emitter), RelaxAll,
-                             IncrementalLinkerCompatible);
+                             std::move(Emitter), IncrementalLinkerCompatible);
       break;
     case Triple::MachO:
       if (MachOStreamerCtorFn)
         S = MachOStreamerCtorFn(Ctx, std::move(TAB), std::move(OW),
-                                std::move(Emitter), RelaxAll,
-                                DWARFMustBeAtTheEnd);
+                                std::move(Emitter), DWARFMustBeAtTheEnd);
       else
         S = createMachOStreamer(Ctx, std::move(TAB), std::move(OW),
-                                std::move(Emitter), RelaxAll,
-                                DWARFMustBeAtTheEnd);
+                                std::move(Emitter), DWARFMustBeAtTheEnd);
       break;
     case Triple::ELF:
       if (ELFStreamerCtorFn)
         S = ELFStreamerCtorFn(T, Ctx, std::move(TAB), std::move(OW),
-                              std::move(Emitter), RelaxAll);
+                              std::move(Emitter));
       else
         S = createELFStreamer(Ctx, std::move(TAB), std::move(OW),
-                              std::move(Emitter), RelaxAll);
+                              std::move(Emitter));
       break;
     case Triple::Wasm:
       if (WasmStreamerCtorFn)
         S = WasmStreamerCtorFn(T, Ctx, std::move(TAB), std::move(OW),
-                               std::move(Emitter), RelaxAll);
+                               std::move(Emitter));
       else
         S = createWasmStreamer(Ctx, std::move(TAB), std::move(OW),
-                               std::move(Emitter), RelaxAll);
+                               std::move(Emitter));
       break;
     case Triple::GOFF:
       if (GOFFStreamerCtorFn)
         S = GOFFStreamerCtorFn(Ctx, std::move(TAB), std::move(OW),
-                               std::move(Emitter), RelaxAll);
+                               std::move(Emitter));
       else
         S = createGOFFStreamer(Ctx, std::move(TAB), std::move(OW),
-                               std::move(Emitter), RelaxAll);
+                               std::move(Emitter));
       break;
     case Triple::XCOFF:
       if (XCOFFStreamerCtorFn)
         S = XCOFFStreamerCtorFn(T, Ctx, std::move(TAB), std::move(OW),
-                                std::move(Emitter), RelaxAll);
+                                std::move(Emitter));
       else
         S = createXCOFFStreamer(Ctx, std::move(TAB), std::move(OW),
-                                std::move(Emitter), RelaxAll);
+                                std::move(Emitter));
       break;
     case Triple::SPIRV:
       if (SPIRVStreamerCtorFn)
         S = SPIRVStreamerCtorFn(T, Ctx, std::move(TAB), std::move(OW),
-                                std::move(Emitter), RelaxAll);
+                                std::move(Emitter));
       else
         S = createSPIRVStreamer(Ctx, std::move(TAB), std::move(OW),
-                                std::move(Emitter), RelaxAll);
+                                std::move(Emitter));
       break;
     case Triple::DXContainer:
       if (DXContainerStreamerCtorFn)
         S = DXContainerStreamerCtorFn(T, Ctx, std::move(TAB), std::move(OW),
-                              std::move(Emitter), RelaxAll);
+                                      std::move(Emitter));
       else
         S = createDXContainerStreamer(Ctx, std::move(TAB), std::move(OW),
-                                      std::move(Emitter), RelaxAll);
+                                      std::move(Emitter));
       break;
     }
     if (ObjectTargetStreamerCtorFn)
diff --git a/llvm/include/llvm/Object/ELFObjectFile.h b/llvm/include/llvm/Object/ELFObjectFile.h
index 1d457be93741..8cc09e7fd7d5 100644
--- a/llvm/include/llvm/Object/ELFObjectFile.h
+++ b/llvm/include/llvm/Object/ELFObjectFile.h
@@ -199,6 +199,14 @@ public:
   }
 };
 
+inline bool operator<(const ELFSymbolRef &A, const ELFSymbolRef &B) {
+  const DataRefImpl &DRIA = A.getRawDataRefImpl();
+  const DataRefImpl &DRIB = B.getRawDataRefImpl();
+  if (DRIA.d.a == DRIB.d.a)
+    return DRIA.d.b < DRIB.d.b;
+  return DRIA.d.a < DRIB.d.a;
+}
+
 class elf_symbol_iterator : public symbol_iterator {
 public:
   elf_symbol_iterator(const basic_symbol_iterator &B)
@@ -801,9 +809,8 @@ Expected<uint32_t> ELFObjectFile<ELFT>::getSymbolFlags(DataRefImpl Sym) const {
   } else if (EF.getHeader().e_machine == ELF::EM_RISCV) {
     if (Expected<StringRef> NameOrErr = getSymbolName(Sym)) {
       StringRef Name = *NameOrErr;
-      // Mark empty name symbols (used for label differences) and mapping
-      // symbols.
-      if (Name.empty() || Name.starts_with("$d") || Name.starts_with("$x"))
+      // Mark fake labels (used for label differences) and mapping symbols.
+      if (Name == ".L0 " || Name.starts_with("$d") || Name.starts_with("$x"))
         Result |= SymbolRef::SF_FormatSpecific;
     } else {
       // TODO: Actually report errors helpfully.
diff --git a/llvm/include/llvm/ProfileData/InstrProfWriter.h b/llvm/include/llvm/ProfileData/InstrProfWriter.h
index b0ae8f364fca..08db8fa6e7ef 100644
--- a/llvm/include/llvm/ProfileData/InstrProfWriter.h
+++ b/llvm/include/llvm/ProfileData/InstrProfWriter.h
@@ -85,11 +85,15 @@ private:
   // The MemProf version we should write.
   memprof::IndexedVersion MemProfVersionRequested;
 
+  // Whether to serialize the full schema.
+  bool MemProfFullSchema;
+
 public:
   InstrProfWriter(
       bool Sparse = false, uint64_t TemporalProfTraceReservoirSize = 0,
       uint64_t MaxTemporalProfTraceLength = 0, bool WritePrevVersion = false,
-      memprof::IndexedVersion MemProfVersionRequested = memprof::Version0);
+      memprof::IndexedVersion MemProfVersionRequested = memprof::Version0,
+      bool MemProfFullSchema = false);
   ~InstrProfWriter();
 
   StringMap<ProfilingData> &getProfileData() { return FunctionData; }
@@ -203,6 +207,7 @@ public:
   void setMemProfVersionRequested(memprof::IndexedVersion Version) {
     MemProfVersionRequested = Version;
   }
+  void setMemProfFullSchema(bool Full) { MemProfFullSchema = Full; }
   // Compute the overlap b/w this object and Other. Program level result is
   // stored in Overlap and function level result is stored in FuncLevelOverlap.
   void overlapRecord(NamedInstrProfRecord &&Other, OverlapStats &Overlap,
diff --git a/llvm/include/llvm/ProfileData/MemProf.h b/llvm/include/llvm/ProfileData/MemProf.h
index aa6cdf198485..4274f2a6849b 100644
--- a/llvm/include/llvm/ProfileData/MemProf.h
+++ b/llvm/include/llvm/ProfileData/MemProf.h
@@ -2,6 +2,7 @@
 #define LLVM_PROFILEDATA_MEMPROF_H_
 
 #include "llvm/ADT/MapVector.h"
+#include "llvm/ADT/STLForwardCompat.h"
 #include "llvm/ADT/STLFunctionalExtras.h"
 #include "llvm/ADT/SmallVector.h"
 #include "llvm/IR/GlobalValue.h"
@@ -10,6 +11,7 @@
 #include "llvm/Support/EndianStream.h"
 #include "llvm/Support/raw_ostream.h"
 
+#include <bitset>
 #include <cstdint>
 #include <optional>
 
@@ -44,12 +46,21 @@ enum class Meta : uint64_t {
 
 using MemProfSchema = llvm::SmallVector<Meta, static_cast<int>(Meta::Size)>;
 
+// Returns the full schema currently in use.
+MemProfSchema getFullSchema();
+
+// Returns the schema consisting of the fields used for hot cold memory hinting.
+MemProfSchema getHotColdSchema();
+
 // Holds the actual MemInfoBlock data with all fields. Contents may be read or
 // written partially by providing an appropriate schema to the serialize and
 // deserialize methods.
 struct PortableMemInfoBlock {
   PortableMemInfoBlock() = default;
-  explicit PortableMemInfoBlock(const MemInfoBlock &Block) {
+  explicit PortableMemInfoBlock(const MemInfoBlock &Block,
+                                const MemProfSchema &IncomingSchema) {
+    for (const Meta Id : IncomingSchema)
+      Schema.set(llvm::to_underlying(Id));
 #define MIBEntryDef(NameTag, Name, Type) Name = Block.Name;
 #include "llvm/ProfileData/MIBEntryDef.inc"
 #undef MIBEntryDef
@@ -61,10 +72,12 @@ struct PortableMemInfoBlock {
 
   // Read the contents of \p Ptr based on the \p Schema to populate the
   // MemInfoBlock member.
-  void deserialize(const MemProfSchema &Schema, const unsigned char *Ptr) {
+  void deserialize(const MemProfSchema &IncomingSchema,
+                   const unsigned char *Ptr) {
     using namespace support;
 
-    for (const Meta Id : Schema) {
+    Schema.reset();
+    for (const Meta Id : IncomingSchema) {
       switch (Id) {
 #define MIBEntryDef(NameTag, Name, Type)                                       \
   case Meta::Name: {                                                           \
@@ -76,6 +89,8 @@ struct PortableMemInfoBlock {
         llvm_unreachable("Unknown meta type id, is the profile collected from "
                          "a newer version of the runtime?");
       }
+
+      Schema.set(llvm::to_underlying(Id));
     }
   }
 
@@ -108,26 +123,29 @@ struct PortableMemInfoBlock {
 #undef MIBEntryDef
   }
 
+  // Return the schema, only for unit tests.
+  std::bitset<llvm::to_underlying(Meta::Size)> getSchema() const {
+    return Schema;
+  }
+
   // Define getters for each type which can be called by analyses.
 #define MIBEntryDef(NameTag, Name, Type)                                       \
-  Type get##Name() const { return Name; }
+  Type get##Name() const {                                                     \
+    assert(Schema[llvm::to_underlying(Meta::Name)]);                           \
+    return Name;                                                               \
+  }
 #include "llvm/ProfileData/MIBEntryDef.inc"
 #undef MIBEntryDef
 
   void clear() { *this = PortableMemInfoBlock(); }
 
-  // Returns the full schema currently in use.
-  static MemProfSchema getSchema() {
-    MemProfSchema List;
-#define MIBEntryDef(NameTag, Name, Type) List.push_back(Meta::Name);
-#include "llvm/ProfileData/MIBEntryDef.inc"
-#undef MIBEntryDef
-    return List;
-  }
-
   bool operator==(const PortableMemInfoBlock &Other) const {
+    if (Other.Schema != Schema)
+      return false;
+
 #define MIBEntryDef(NameTag, Name, Type)                                       \
-  if (Other.get##Name() != get##Name())                                        \
+  if (Schema[llvm::to_underlying(Meta::Name)] &&                               \
+      Other.get##Name() != get##Name())                                        \
     return false;
 #include "llvm/ProfileData/MIBEntryDef.inc"
 #undef MIBEntryDef
@@ -138,15 +156,29 @@ struct PortableMemInfoBlock {
     return !operator==(Other);
   }
 
-  static constexpr size_t serializedSize() {
+  static size_t serializedSize(const MemProfSchema &Schema) {
     size_t Result = 0;
-#define MIBEntryDef(NameTag, Name, Type) Result += sizeof(Type);
+
+    for (const Meta Id : Schema) {
+      switch (Id) {
+#define MIBEntryDef(NameTag, Name, Type)                                       \
+  case Meta::Name: {                                                           \
+    Result += sizeof(Type);                                                    \
+  } break;
 #include "llvm/ProfileData/MIBEntryDef.inc"
 #undef MIBEntryDef
+      default:
+        llvm_unreachable("Unknown meta type id, invalid input?");
+      }
+    }
+
     return Result;
   }
 
 private:
+  // The set of available fields, indexed by Meta::Name.
+  std::bitset<llvm::to_underlying(Meta::Size)> Schema;
+
 #define MIBEntryDef(NameTag, Name, Type) Type Name = Type();
 #include "llvm/ProfileData/MIBEntryDef.inc"
 #undef MIBEntryDef
@@ -288,11 +320,13 @@ struct IndexedAllocationInfo {
 
   IndexedAllocationInfo() = default;
   IndexedAllocationInfo(ArrayRef<FrameId> CS, CallStackId CSId,
-                        const MemInfoBlock &MB)
-      : CallStack(CS.begin(), CS.end()), CSId(CSId), Info(MB) {}
+                        const MemInfoBlock &MB,
+                        const MemProfSchema &Schema = getFullSchema())
+      : CallStack(CS.begin(), CS.end()), CSId(CSId), Info(MB, Schema) {}
 
   // Returns the size in bytes when this allocation info struct is serialized.
-  size_t serializedSize(IndexedVersion Version) const;
+  size_t serializedSize(const MemProfSchema &Schema,
+                        IndexedVersion Version) const;
 
   bool operator==(const IndexedAllocationInfo &Other) const {
     if (Other.Info != Info)
@@ -367,7 +401,8 @@ struct IndexedMemProfRecord {
     CallSites.append(Other.CallSites);
   }
 
-  size_t serializedSize(IndexedVersion Version) const;
+  size_t serializedSize(const MemProfSchema &Schema,
+                        IndexedVersion Version) const;
 
   bool operator==(const IndexedMemProfRecord &Other) const {
     if (Other.AllocSites != AllocSites)
@@ -535,7 +570,7 @@ public:
     endian::Writer LE(Out, llvm::endianness::little);
     offset_type N = sizeof(K);
     LE.write<offset_type>(N);
-    offset_type M = V.serializedSize(Version);
+    offset_type M = V.serializedSize(*Schema, Version);
     LE.write<offset_type>(M);
     return std::make_pair(N, M);
   }
@@ -727,6 +762,64 @@ public:
 // Compute a CallStackId for a given call stack.
 CallStackId hashCallStack(ArrayRef<FrameId> CS);
 
+namespace detail {
+// "Dereference" the iterator from DenseMap or OnDiskChainedHashTable.  We have
+// to do so in one of two different ways depending on the type of the hash
+// table.
+template <typename value_type, typename IterTy>
+value_type DerefIterator(IterTy Iter) {
+  using deref_type = llvm::remove_cvref_t<decltype(*Iter)>;
+  if constexpr (std::is_same_v<deref_type, value_type>)
+    return *Iter;
+  else
+    return Iter->second;
+}
+} // namespace detail
+
+// A function object that returns a frame for a given FrameId.
+template <typename MapTy> struct FrameIdConverter {
+  std::optional<FrameId> LastUnmappedId;
+  MapTy &Map;
+
+  FrameIdConverter() = delete;
+  FrameIdConverter(MapTy &Map) : Map(Map) {}
+
+  Frame operator()(FrameId Id) {
+    auto Iter = Map.find(Id);
+    if (Iter == Map.end()) {
+      LastUnmappedId = Id;
+      return Frame(0, 0, 0, false);
+    }
+    return detail::DerefIterator<Frame>(Iter);
+  }
+};
+
+// A function object that returns a call stack for a given CallStackId.
+template <typename MapTy> struct CallStackIdConverter {
+  std::optional<CallStackId> LastUnmappedId;
+  MapTy &Map;
+  std::function<Frame(FrameId)> FrameIdToFrame;
+
+  CallStackIdConverter() = delete;
+  CallStackIdConverter(MapTy &Map, std::function<Frame(FrameId)> FrameIdToFrame)
+      : Map(Map), FrameIdToFrame(FrameIdToFrame) {}
+
+  llvm::SmallVector<Frame> operator()(CallStackId CSId) {
+    llvm::SmallVector<Frame> Frames;
+    auto CSIter = Map.find(CSId);
+    if (CSIter == Map.end()) {
+      LastUnmappedId = CSId;
+    } else {
+      llvm::SmallVector<FrameId> CS =
+          detail::DerefIterator<llvm::SmallVector<FrameId>>(CSIter);
+      Frames.reserve(CS.size());
+      for (FrameId Id : CS)
+        Frames.push_back(FrameIdToFrame(Id));
+    }
+    return Frames;
+  }
+};
+
 // Verify that each CallStackId is computed with hashCallStack.  This function
 // is intended to help transition from CallStack to CSId in
 // IndexedAllocationInfo.
diff --git a/llvm/include/llvm/ProfileData/MemProfReader.h b/llvm/include/llvm/ProfileData/MemProfReader.h
index 444c58e8bdc8..b42e4f597774 100644
--- a/llvm/include/llvm/ProfileData/MemProfReader.h
+++ b/llvm/include/llvm/ProfileData/MemProfReader.h
@@ -76,20 +76,16 @@ public:
       Callback =
           std::bind(&MemProfReader::idToFrame, this, std::placeholders::_1);
 
-    auto CallStackCallback = [&](CallStackId CSId) {
-      llvm::SmallVector<Frame> CallStack;
-      auto Iter = CSIdToCallStack.find(CSId);
-      assert(Iter != CSIdToCallStack.end());
-      for (FrameId Id : Iter->second)
-        CallStack.push_back(Callback(Id));
-      return CallStack;
-    };
+    memprof::CallStackIdConverter<decltype(CSIdToCallStack)> CSIdConv(
+        CSIdToCallStack, Callback);
 
     const IndexedMemProfRecord &IndexedRecord = Iter->second;
     GuidRecord = {
         Iter->first,
-        IndexedRecord.toMemProfRecord(CallStackCallback),
+        IndexedRecord.toMemProfRecord(CSIdConv),
     };
+    if (CSIdConv.LastUnmappedId)
+      return make_error<InstrProfError>(instrprof_error::hash_mismatch);
     Iter++;
     return Error::success();
   }
diff --git a/llvm/include/llvm/Support/RISCVISAUtils.h b/llvm/include/llvm/Support/RISCVISAUtils.h
new file mode 100644
index 000000000000..77f8c3e45f1a
--- /dev/null
+++ b/llvm/include/llvm/Support/RISCVISAUtils.h
@@ -0,0 +1,49 @@
+//===-- RISCVISAUtils.h - RISC-V ISA Utilities ------------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// Utilities shared by TableGen and RISCVISAInfo.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_SUPPORT_RISCVISAUTILS_H
+#define LLVM_SUPPORT_RISCVISAUTILS_H
+
+#include "llvm/ADT/StringRef.h"
+#include <map>
+#include <string>
+
+namespace llvm {
+
+namespace RISCVISAUtils {
+constexpr StringLiteral AllStdExts = "mafdqlcbkjtpvnh";
+
+/// Represents the major and version number components of a RISC-V extension.
+struct ExtensionVersion {
+  unsigned Major;
+  unsigned Minor;
+};
+
+bool compareExtension(const std::string &LHS, const std::string &RHS);
+
+/// Helper class for OrderedExtensionMap.
+struct ExtensionComparator {
+  bool operator()(const std::string &LHS, const std::string &RHS) const {
+    return compareExtension(LHS, RHS);
+  }
+};
+
+/// OrderedExtensionMap is std::map, it's specialized to keep entries
+/// in canonical order of extension.
+typedef std::map<std::string, ExtensionVersion, ExtensionComparator>
+    OrderedExtensionMap;
+
+} // namespace RISCVISAUtils
+
+} // namespace llvm
+
+#endif
diff --git a/llvm/include/llvm/Support/YAMLTraits.h b/llvm/include/llvm/Support/YAMLTraits.h
index 3b1f4bad57fc..33aeb039320d 100644
--- a/llvm/include/llvm/Support/YAMLTraits.h
+++ b/llvm/include/llvm/Support/YAMLTraits.h
@@ -671,7 +671,11 @@ inline bool isBool(StringRef S) {
 // (except for TAB #x9, LF #xA, and CR #xD which are allowed), DEL #x7F, the C1
 // control block #x80-#x9F (except for NEL #x85 which is allowed), the surrogate
 // block #xD800-#xDFFF, #xFFFE, and #xFFFF.
-inline QuotingType needsQuotes(StringRef S) {
+//
+// Some strings are valid YAML values even unquoted, but without quotes are
+// interpreted as non-string type, for instance null, boolean or numeric values.
+// If ForcePreserveAsString is set, such strings are quoted.
+inline QuotingType needsQuotes(StringRef S, bool ForcePreserveAsString = true) {
   if (S.empty())
     return QuotingType::Single;
 
@@ -679,12 +683,14 @@ inline QuotingType needsQuotes(StringRef S) {
   if (isSpace(static_cast<unsigned char>(S.front())) ||
       isSpace(static_cast<unsigned char>(S.back())))
     MaxQuotingNeeded = QuotingType::Single;
-  if (isNull(S))
-    MaxQuotingNeeded = QuotingType::Single;
-  if (isBool(S))
-    MaxQuotingNeeded = QuotingType::Single;
-  if (isNumeric(S))
-    MaxQuotingNeeded = QuotingType::Single;
+  if (ForcePreserveAsString) {
+    if (isNull(S))
+      MaxQuotingNeeded = QuotingType::Single;
+    if (isBool(S))
+      MaxQuotingNeeded = QuotingType::Single;
+    if (isNumeric(S))
+      MaxQuotingNeeded = QuotingType::Single;
+  }
 
   // 7.3.3 Plain Style
   // Plain scalars must not begin with most indicators, as this would cause
@@ -1636,6 +1642,7 @@ public:
 
 private:
   void output(StringRef s);
+  void output(StringRef, QuotingType);
   void outputUpToEndOfLine(StringRef s);
   void newLineCheck(bool EmptySequence = false);
   void outputNewLine();
diff --git a/llvm/include/llvm/Target/GlobalISel/Combine.td b/llvm/include/llvm/Target/GlobalISel/Combine.td
index 31b903e63d99..dbbb3abaa830 100644
--- a/llvm/include/llvm/Target/GlobalISel/Combine.td
+++ b/llvm/include/llvm/Target/GlobalISel/Combine.td
@@ -1525,11 +1525,39 @@ def combine_shuffle_concat : GICombineRule<
         [{ return Helper.matchCombineShuffleConcat(*${root}, ${matchinfo}); }]),
   (apply [{ Helper.applyCombineShuffleConcat(*${root}, ${matchinfo}); }])>;
 
-// match_extract_of_element must be the first!
+def insert_vector_element_idx_undef : GICombineRule<
+   (defs root:$root, build_fn_matchinfo:$matchinfo),
+   (match (G_IMPLICIT_DEF $idx),
+          (G_INSERT_VECTOR_ELT $root, $src, $elt, $idx)),
+   (apply (G_IMPLICIT_DEF $root))>;
+
+def insert_vector_element_elt_undef : GICombineRule<
+   (defs root:$root, build_fn_matchinfo:$matchinfo),
+   (match (G_IMPLICIT_DEF $elt),
+          (G_INSERT_VECTOR_ELT $root, $src, $elt, $idx),
+          [{ return isGuaranteedNotToBePoison(${src}.getReg(), MRI); }]),
+   (apply (GIReplaceReg $root, $src))>;
+
+def insert_vector_element_extract_vector_element : GICombineRule<
+   (defs root:$root, build_fn_matchinfo:$matchinfo),
+   (match (G_EXTRACT_VECTOR_ELT $elt, $src, $idx),
+          (G_INSERT_VECTOR_ELT $root, $src, $elt, $idx)),
+   (apply (GIReplaceReg $root, $src))>;
+
+def insert_vector_elt_oob : GICombineRule<
+  (defs root:$root, build_fn_matchinfo:$matchinfo),
+  (match (wip_match_opcode G_INSERT_VECTOR_ELT):$root,
+         [{ return Helper.matchInsertVectorElementOOB(*${root}, ${matchinfo}); }]),
+  (apply [{ Helper.applyBuildFn(*${root}, ${matchinfo}); }])>;
+
+// match_extract_of_element and insert_vector_elt_oob must be the first!
 def vector_ops_combines: GICombineGroup<[
 match_extract_of_element_undef_vector,
 match_extract_of_element_undef_index,
+insert_vector_element_idx_undef,
+insert_vector_element_elt_undef,
 match_extract_of_element,
+insert_vector_elt_oob,
 extract_vector_element_not_const,
 extract_vector_element_different_indices,
 extract_vector_element_build_vector2,
@@ -1553,7 +1581,8 @@ extract_vector_element_build_vector_trunc5,
 extract_vector_element_build_vector_trunc6,
 extract_vector_element_build_vector_trunc7,
 extract_vector_element_build_vector_trunc8,
-extract_vector_element_freeze
+extract_vector_element_freeze,
+insert_vector_element_extract_vector_element
 ]>;
 
 // FIXME: These should use the custom predicate feature once it lands.
diff --git a/llvm/include/llvm/Target/TargetSelectionDAG.td b/llvm/include/llvm/Target/TargetSelectionDAG.td
index ea3520835fa0..1684b424e3b4 100644
--- a/llvm/include/llvm/Target/TargetSelectionDAG.td
+++ b/llvm/include/llvm/Target/TargetSelectionDAG.td
@@ -1979,6 +1979,7 @@ class Pattern<dag patternToMatch, list<dag> resultInstrs> {
   list<dag>       ResultInstrs    = resultInstrs;
   list<Predicate> Predicates      = [];  // See class Instruction in Target.td.
   int             AddedComplexity = 0;   // See class Instruction in Target.td.
+  bit           GISelShouldIgnore = 0;
 }
 
 // Pat - A simple (but common) form of a pattern, which produces a simple result
diff --git a/llvm/include/llvm/TargetParser/AArch64TargetParser.h b/llvm/include/llvm/TargetParser/AArch64TargetParser.h
index 805b963a7a13..0d1cfd152151 100644
--- a/llvm/include/llvm/TargetParser/AArch64TargetParser.h
+++ b/llvm/include/llvm/TargetParser/AArch64TargetParser.h
@@ -268,7 +268,7 @@ inline constexpr ExtensionInfo Extensions[] = {
     {"sha3", AArch64::AEK_SHA3, "+sha3", "-sha3", FEAT_SHA3, "+sha3,+sha2,+fp-armv8,+neon", 140},
     {"simd", AArch64::AEK_SIMD, "+neon", "-neon", FEAT_SIMD, "+fp-armv8,+neon", 100},
     {"sm4", AArch64::AEK_SM4, "+sm4", "-sm4", FEAT_SM4, "+sm4,+fp-armv8,+neon", 106},
-    {"sme-f16f16", AArch64::AEK_SMEF16F16, "+sme-f16f16", "-sme-f16f16", FEAT_INIT, "", 0},
+    {"sme-f16f16", AArch64::AEK_SMEF16F16, "+sme-f16f16", "-sme-f16f16", FEAT_INIT, "+sme2,+sme-f16f16", 0},
     {"sme-f64f64", AArch64::AEK_SMEF64F64, "+sme-f64f64", "-sme-f64f64", FEAT_SME_F64, "+sme,+sme-f64f64,+bf16", 560},
     {"sme-i16i64", AArch64::AEK_SMEI16I64, "+sme-i16i64", "-sme-i16i64", FEAT_SME_I64, "+sme,+sme-i16i64,+bf16", 570},
     {"sme", AArch64::AEK_SME, "+sme", "-sme", FEAT_SME, "+sme,+bf16", 430},
@@ -302,7 +302,7 @@ inline constexpr ExtensionInfo Extensions[] = {
     {"ssve-fp8dot4", AArch64::AEK_SSVE_FP8DOT4, "+ssve-fp8dot4", "-ssve-fp8dot4", FEAT_INIT, "+sme2", 0},
     {"lut", AArch64::AEK_LUT, "+lut", "-lut", FEAT_INIT, "", 0},
     {"sme-lutv2", AArch64::AEK_SME_LUTv2, "+sme-lutv2", "-sme-lutv2", FEAT_INIT, "", 0},
-    {"sme-f8f16", AArch64::AEK_SMEF8F16, "+sme-f8f16", "-sme-f8f16", FEAT_INIT, "+sme2,+fp8", 0},
+    {"sme-f8f16", AArch64::AEK_SMEF8F16, "+sme-f8f16", "-sme-f8f16", FEAT_INIT, "+fp8,+sme2", 0},
     {"sme-f8f32", AArch64::AEK_SMEF8F32, "+sme-f8f32", "-sme-f8f32", FEAT_INIT, "+sme2,+fp8", 0},
     {"sme-fa64",  AArch64::AEK_SMEFA64,  "+sme-fa64", "-sme-fa64",  FEAT_INIT, "", 0},
     {"cpa", AArch64::AEK_CPA, "+cpa", "-cpa", FEAT_INIT, "", 0},
@@ -677,6 +677,13 @@ inline constexpr CpuInfo CpuInfos[] = {
           AArch64::AEK_FP16FML, AArch64::AEK_I8MM, AArch64::AEK_MTE,
           AArch64::AEK_SB, AArch64::AEK_SSBS, AArch64::AEK_SVE,
           AArch64::AEK_SVE2, AArch64::AEK_SVE2BITPERM})},
+    {"neoverse-n3", ARMV9_2A,
+     AArch64::ExtensionBitset({AArch64::AEK_MTE, AArch64::AEK_SSBS,
+                               AArch64::AEK_SB, AArch64::AEK_PREDRES,
+                               AArch64::AEK_FP16FML, AArch64::AEK_PAUTH,
+                               AArch64::AEK_FLAGM, AArch64::AEK_PERFMON,
+                               AArch64::AEK_RAND, AArch64::AEK_SVE2BITPERM,
+                               AArch64::AEK_PROFILE, AArch64::AEK_PERFMON})},
     {"neoverse-512tvb", ARMV8_4A,
      AArch64::ExtensionBitset(
          {AArch64::AEK_AES, AArch64::AEK_SHA2, AArch64::AEK_SHA3,
@@ -697,6 +704,20 @@ inline constexpr CpuInfo CpuInfos[] = {
           AArch64::AEK_FP16, AArch64::AEK_BF16, AArch64::AEK_RAND,
           AArch64::AEK_DOTPROD, AArch64::AEK_PROFILE, AArch64::AEK_SVE2BITPERM,
           AArch64::AEK_FP16FML, AArch64::AEK_I8MM, AArch64::AEK_MTE})},
+    {"neoverse-v3", ARMV9_2A,
+     AArch64::ExtensionBitset(
+         {AArch64::AEK_PROFILE, AArch64::AEK_MTE, AArch64::AEK_SSBS,
+          AArch64::AEK_SB, AArch64::AEK_PREDRES, AArch64::AEK_LS64,
+          AArch64::AEK_BRBE, AArch64::AEK_PAUTH, AArch64::AEK_FLAGM,
+          AArch64::AEK_PERFMON, AArch64::AEK_RAND, AArch64::AEK_SVE2BITPERM,
+          AArch64::AEK_FP16FML})},
+    {"neoverse-v3ae", ARMV9_2A,
+     (AArch64::ExtensionBitset(
+         {AArch64::AEK_PROFILE, AArch64::AEK_MTE, AArch64::AEK_SSBS,
+          AArch64::AEK_SB, AArch64::AEK_PREDRES, AArch64::AEK_LS64,
+          AArch64::AEK_BRBE, AArch64::AEK_PAUTH, AArch64::AEK_FLAGM,
+          AArch64::AEK_PERFMON, AArch64::AEK_RAND, AArch64::AEK_SVE2BITPERM,
+          AArch64::AEK_FP16FML}))},
     {"cyclone", ARMV8A,
      AArch64::ExtensionBitset(
          {AArch64::AEK_AES, AArch64::AEK_SHA2, AArch64::AEK_NONE})},
diff --git a/llvm/include/llvm/TargetParser/CMakeLists.txt b/llvm/include/llvm/TargetParser/CMakeLists.txt
index 7f080e01548c..f89d4eb5ea16 100644
--- a/llvm/include/llvm/TargetParser/CMakeLists.txt
+++ b/llvm/include/llvm/TargetParser/CMakeLists.txt
@@ -1,3 +1,12 @@
+set(LLVM_TARGET_DEFINITIONS ${PROJECT_SOURCE_DIR}/lib/Target/ARM/ARM.td)
+tablegen(LLVM ARMTargetParserDef.inc -gen-arm-target-def -I ${PROJECT_SOURCE_DIR}/lib/Target/ARM/)
+add_public_tablegen_target(ARMTargetParserTableGen)
+
+set(LLVM_TARGET_DEFINITIONS ${PROJECT_SOURCE_DIR}/lib/Target/AArch64/AArch64.td)
+tablegen(LLVM AArch64TargetParserDef.inc -gen-arm-target-def -I ${PROJECT_SOURCE_DIR}/lib/Target/AArch64/)
+add_public_tablegen_target(AArch64TargetParserTableGen)
+
 set(LLVM_TARGET_DEFINITIONS ${PROJECT_SOURCE_DIR}/lib/Target/RISCV/RISCV.td)
 tablegen(LLVM RISCVTargetParserDef.inc -gen-riscv-target-def -I ${PROJECT_SOURCE_DIR}/lib/Target/RISCV/)
 add_public_tablegen_target(RISCVTargetParserTableGen)
+
diff --git a/llvm/include/llvm/Support/RISCVISAInfo.h b/llvm/include/llvm/TargetParser/RISCVISAInfo.h
index 46df93d75226..0d5637155daa 100644
--- a/llvm/include/llvm/Support/RISCVISAInfo.h
+++ b/llvm/include/llvm/TargetParser/RISCVISAInfo.h
@@ -12,6 +12,7 @@
 #include "llvm/ADT/StringMap.h"
 #include "llvm/ADT/StringRef.h"
 #include "llvm/Support/Error.h"
+#include "llvm/Support/RISCVISAUtils.h"
 
 #include <map>
 #include <string>
@@ -25,27 +26,7 @@ public:
   RISCVISAInfo(const RISCVISAInfo &) = delete;
   RISCVISAInfo &operator=(const RISCVISAInfo &) = delete;
 
-  /// Represents the major and version number components of a RISC-V extension.
-  struct ExtensionVersion {
-    unsigned Major;
-    unsigned Minor;
-  };
-
-  static bool compareExtension(const std::string &LHS, const std::string &RHS);
-
-  /// Helper class for OrderedExtensionMap.
-  struct ExtensionComparator {
-    bool operator()(const std::string &LHS, const std::string &RHS) const {
-      return compareExtension(LHS, RHS);
-    }
-  };
-
-  /// OrderedExtensionMap is std::map, it's specialized to keep entries
-  /// in canonical order of extension.
-  typedef std::map<std::string, ExtensionVersion, ExtensionComparator>
-      OrderedExtensionMap;
-
-  RISCVISAInfo(unsigned XLen, OrderedExtensionMap &Exts)
+  RISCVISAInfo(unsigned XLen, RISCVISAUtils::OrderedExtensionMap &Exts)
       : XLen(XLen), FLen(0), MinVLen(0), MaxELen(0), MaxELenFp(0), Exts(Exts) {}
 
   /// Parse RISC-V ISA info from arch string.
@@ -72,7 +53,9 @@ public:
   std::vector<std::string> toFeatures(bool AddAllExtensions = false,
                                       bool IgnoreUnknown = true) const;
 
-  const OrderedExtensionMap &getExtensions() const { return Exts; }
+  const RISCVISAUtils::OrderedExtensionMap &getExtensions() const {
+    return Exts;
+  }
 
   unsigned getXLen() const { return XLen; }
   unsigned getFLen() const { return FLen; }
@@ -103,9 +86,9 @@ private:
   unsigned MinVLen;
   unsigned MaxELen, MaxELenFp;
 
-  OrderedExtensionMap Exts;
+  RISCVISAUtils::OrderedExtensionMap Exts;
 
-  void addExtension(StringRef ExtName, ExtensionVersion Version);
+  void addExtension(StringRef ExtName, RISCVISAUtils::ExtensionVersion Version);
 
   Error checkDependency();
 
diff --git a/llvm/include/llvm/Transforms/InstCombine/InstCombiner.h b/llvm/include/llvm/Transforms/InstCombine/InstCombiner.h
index ea1f4fc3b85d..855d1aeddfae 100644
--- a/llvm/include/llvm/Transforms/InstCombine/InstCombiner.h
+++ b/llvm/include/llvm/Transforms/InstCombine/InstCombiner.h
@@ -461,9 +461,10 @@ public:
 
   OverflowResult computeOverflowForUnsignedMul(const Value *LHS,
                                                const Value *RHS,
-                                               const Instruction *CxtI) const {
-    return llvm::computeOverflowForUnsignedMul(LHS, RHS,
-                                               SQ.getWithInstruction(CxtI));
+                                               const Instruction *CxtI,
+                                               bool IsNSW = false) const {
+    return llvm::computeOverflowForUnsignedMul(
+        LHS, RHS, SQ.getWithInstruction(CxtI), IsNSW);
   }
 
   OverflowResult computeOverflowForSignedMul(const Value *LHS, const Value *RHS,
diff --git a/llvm/include/llvm/Transforms/Scalar/GVN.h b/llvm/include/llvm/Transforms/Scalar/GVN.h
index 4ba9b74ccb00..debe2ee79917 100644
--- a/llvm/include/llvm/Transforms/Scalar/GVN.h
+++ b/llvm/include/llvm/Transforms/Scalar/GVN.h
@@ -232,13 +232,67 @@ private:
 
   /// A mapping from value numbers to lists of Value*'s that
   /// have that value number.  Use findLeader to query it.
-  struct LeaderTableEntry {
-    Value *Val;
-    const BasicBlock *BB;
-    LeaderTableEntry *Next;
+  class LeaderMap {
+  public:
+    struct LeaderTableEntry {
+      Value *Val;
+      const BasicBlock *BB;
+    };
+
+  private:
+    struct LeaderListNode {
+      LeaderTableEntry Entry;
+      LeaderListNode *Next;
+    };
+    DenseMap<uint32_t, LeaderListNode> NumToLeaders;
+    BumpPtrAllocator TableAllocator;
+
+  public:
+    class leader_iterator {
+      const LeaderListNode *Current;
+
+    public:
+      using iterator_category = std::forward_iterator_tag;
+      using value_type = const LeaderTableEntry;
+      using difference_type = std::ptrdiff_t;
+      using pointer = value_type *;
+      using reference = value_type &;
+
+      leader_iterator(const LeaderListNode *C) : Current(C) {}
+      leader_iterator &operator++() {
+        assert(Current && "Dereferenced end of leader list!");
+        Current = Current->Next;
+        return *this;
+      }
+      bool operator==(const leader_iterator &Other) const {
+        return Current == Other.Current;
+      }
+      bool operator!=(const leader_iterator &Other) const {
+        return Current != Other.Current;
+      }
+      reference operator*() const { return Current->Entry; }
+    };
+
+    iterator_range<leader_iterator> getLeaders(uint32_t N) {
+      auto I = NumToLeaders.find(N);
+      if (I == NumToLeaders.end()) {
+        return iterator_range(leader_iterator(nullptr),
+                              leader_iterator(nullptr));
+      }
+
+      return iterator_range(leader_iterator(&I->second),
+                            leader_iterator(nullptr));
+    }
+
+    void insert(uint32_t N, Value *V, const BasicBlock *BB);
+    void erase(uint32_t N, Instruction *I, const BasicBlock *BB);
+    void verifyRemoved(const Value *Inst) const;
+    void clear() {
+      NumToLeaders.clear();
+      TableAllocator.Reset();
+    }
   };
-  DenseMap<uint32_t, LeaderTableEntry> LeaderTable;
-  BumpPtrAllocator TableAllocator;
+  LeaderMap LeaderTable;
 
   // Block-local map of equivalent values to their leader, does not
   // propagate to any successors. Entries added mid-block are applied
@@ -264,51 +318,6 @@ private:
                MemoryDependenceResults *RunMD, LoopInfo &LI,
                OptimizationRemarkEmitter *ORE, MemorySSA *MSSA = nullptr);
 
-  /// Push a new Value to the LeaderTable onto the list for its value number.
-  void addToLeaderTable(uint32_t N, Value *V, const BasicBlock *BB) {
-    LeaderTableEntry &Curr = LeaderTable[N];
-    if (!Curr.Val) {
-      Curr.Val = V;
-      Curr.BB = BB;
-      return;
-    }
-
-    LeaderTableEntry *Node = TableAllocator.Allocate<LeaderTableEntry>();
-    Node->Val = V;
-    Node->BB = BB;
-    Node->Next = Curr.Next;
-    Curr.Next = Node;
-  }
-
-  /// Scan the list of values corresponding to a given
-  /// value number, and remove the given instruction if encountered.
-  void removeFromLeaderTable(uint32_t N, Instruction *I, BasicBlock *BB) {
-    LeaderTableEntry *Prev = nullptr;
-    LeaderTableEntry *Curr = &LeaderTable[N];
-
-    while (Curr && (Curr->Val != I || Curr->BB != BB)) {
-      Prev = Curr;
-      Curr = Curr->Next;
-    }
-
-    if (!Curr)
-      return;
-
-    if (Prev) {
-      Prev->Next = Curr->Next;
-    } else {
-      if (!Curr->Next) {
-        Curr->Val = nullptr;
-        Curr->BB = nullptr;
-      } else {
-        LeaderTableEntry *Next = Curr->Next;
-        Curr->Val = Next->Val;
-        Curr->BB = Next->BB;
-        Curr->Next = Next->Next;
-      }
-    }
-  }
-
   // List of critical edges to be split between iterations.
   SmallVector<std::pair<Instruction *, unsigned>, 4> toSplit;
 
diff --git a/llvm/include/llvm/Transforms/Scalar/JumpThreading.h b/llvm/include/llvm/Transforms/Scalar/JumpThreading.h
index 3364d7eaee42..f7358ac9b1ee 100644
--- a/llvm/include/llvm/Transforms/Scalar/JumpThreading.h
+++ b/llvm/include/llvm/Transforms/Scalar/JumpThreading.h
@@ -22,6 +22,7 @@
 #include "llvm/Analysis/BranchProbabilityInfo.h"
 #include "llvm/Analysis/DomTreeUpdater.h"
 #include "llvm/IR/ValueHandle.h"
+#include "llvm/Transforms/Utils/ValueMapper.h"
 #include <optional>
 #include <utility>
 
@@ -114,11 +115,10 @@ public:
   bool processBlock(BasicBlock *BB);
   bool maybeMergeBasicBlockIntoOnlyPred(BasicBlock *BB);
   void updateSSA(BasicBlock *BB, BasicBlock *NewBB,
-                 DenseMap<Instruction *, Value *> &ValueMapping);
-  DenseMap<Instruction *, Value *> cloneInstructions(BasicBlock::iterator BI,
-                                                     BasicBlock::iterator BE,
-                                                     BasicBlock *NewBB,
-                                                     BasicBlock *PredBB);
+                 ValueToValueMapTy &ValueMapping);
+  void cloneInstructions(ValueToValueMapTy &ValueMapping,
+                         BasicBlock::iterator BI, BasicBlock::iterator BE,
+                         BasicBlock *NewBB, BasicBlock *PredBB);
   bool tryThreadEdge(BasicBlock *BB,
                      const SmallVectorImpl<BasicBlock *> &PredBBs,
                      BasicBlock *SuccBB);
diff --git a/llvm/include/llvm/Transforms/Utils/GlobalStatus.h b/llvm/include/llvm/Transforms/Utils/GlobalStatus.h
index 60c91fc30174..c001e587313c 100644
--- a/llvm/include/llvm/Transforms/Utils/GlobalStatus.h
+++ b/llvm/include/llvm/Transforms/Utils/GlobalStatus.h
@@ -24,9 +24,9 @@ class Value;
 ///
 bool isSafeToDestroyConstant(const Constant *C);
 
-/// As we analyze each global, keep track of some information about it.  If we
-/// find out that the address of the global is taken, none of this info will be
-/// accurate.
+/// As we analyze each global or thread-local variable, keep track of some
+/// information about it.  If we find out that the address of the global is
+/// taken, none of this info will be accurate.
 struct GlobalStatus {
   /// True if the global's address is used in a comparison.
   bool IsCompared = false;
diff --git a/llvm/include/llvm/Transforms/Utils/Local.h b/llvm/include/llvm/Transforms/Utils/Local.h
index 9ae026fa95d2..6937ec8dfd21 100644
--- a/llvm/include/llvm/Transforms/Utils/Local.h
+++ b/llvm/include/llvm/Transforms/Utils/Local.h
@@ -18,6 +18,7 @@
 #include "llvm/IR/Dominators.h"
 #include "llvm/Support/CommandLine.h"
 #include "llvm/Transforms/Utils/SimplifyCFGOptions.h"
+#include "llvm/Transforms/Utils/ValueMapper.h"
 #include <cstdint>
 
 namespace llvm {
@@ -439,6 +440,18 @@ unsigned replaceDominatedUsesWith(Value *From, Value *To, DominatorTree &DT,
 /// the end of the given BasicBlock. Returns the number of replacements made.
 unsigned replaceDominatedUsesWith(Value *From, Value *To, DominatorTree &DT,
                                   const BasicBlock *BB);
+/// Replace each use of 'From' with 'To' if that use is dominated by
+/// the given edge and the callback ShouldReplace returns true. Returns the
+/// number of replacements made.
+unsigned replaceDominatedUsesWithIf(
+    Value *From, Value *To, DominatorTree &DT, const BasicBlockEdge &Edge,
+    function_ref<bool(const Use &U, const Value *To)> ShouldReplace);
+/// Replace each use of 'From' with 'To' if that use is dominated by
+/// the end of the given BasicBlock and the callback ShouldReplace returns true.
+/// Returns the number of replacements made.
+unsigned replaceDominatedUsesWithIf(
+    Value *From, Value *To, DominatorTree &DT, const BasicBlock *BB,
+    function_ref<bool(const Use &U, const Value *To)> ShouldReplace);
 
 /// Return true if this call calls a gc leaf function.
 ///
@@ -478,6 +491,10 @@ void hoistAllInstructionsInto(BasicBlock *DomBlock, Instruction *InsertPt,
 DIExpression *getExpressionForConstant(DIBuilder &DIB, const Constant &C,
                                        Type &Ty);
 
+/// Remap the operands of the debug records attached to \p Inst, and the
+/// operands of \p Inst itself if it's a debug intrinsic.
+void remapDebugVariable(ValueToValueMapTy &Mapping, Instruction *Inst);
+
 //===----------------------------------------------------------------------===//
 //  Intrinsic pattern matching
 //
diff --git a/llvm/include/llvm/Transforms/Utils/MemoryTaggingSupport.h b/llvm/include/llvm/Transforms/Utils/MemoryTaggingSupport.h
index fb3ab33a0629..16589a605e60 100644
--- a/llvm/include/llvm/Transforms/Utils/MemoryTaggingSupport.h
+++ b/llvm/include/llvm/Transforms/Utils/MemoryTaggingSupport.h
@@ -86,6 +86,8 @@ Value *getFP(IRBuilder<> &IRB);
 Value *getPC(const Triple &TargetTriple, IRBuilder<> &IRB);
 Value *getAndroidSlotPtr(IRBuilder<> &IRB, int Slot);
 
+void annotateDebugRecords(AllocaInfo &Info, unsigned int Tag);
+
 } // namespace memtag
 } // namespace llvm
 
diff --git a/llvm/include/llvm/Transforms/Vectorize/SLPVectorizer.h b/llvm/include/llvm/Transforms/Vectorize/SLPVectorizer.h
index 326006fbb880..4f99d171469e 100644
--- a/llvm/include/llvm/Transforms/Vectorize/SLPVectorizer.h
+++ b/llvm/include/llvm/Transforms/Vectorize/SLPVectorizer.h
@@ -153,10 +153,15 @@ private:
   /// a vectorization chain.
   bool vectorizeChainsInBlock(BasicBlock *BB, slpvectorizer::BoUpSLP &R);
 
-  bool vectorizeStoreChain(ArrayRef<Value *> Chain, slpvectorizer::BoUpSLP &R,
-                           unsigned Idx, unsigned MinVF);
-
-  bool vectorizeStores(ArrayRef<StoreInst *> Stores, slpvectorizer::BoUpSLP &R);
+  std::optional<bool> vectorizeStoreChain(ArrayRef<Value *> Chain,
+                                          slpvectorizer::BoUpSLP &R,
+                                          unsigned Idx, unsigned MinVF,
+                                          unsigned &Size);
+
+  bool vectorizeStores(
+      ArrayRef<StoreInst *> Stores, slpvectorizer::BoUpSLP &R,
+      DenseSet<std::tuple<Value *, Value *, Value *, Value *, unsigned>>
+          &Visited);
 
   /// The store instructions in a basic block organized by base pointer.
   StoreListMap Stores;
diff --git a/llvm/lib/Analysis/BasicAliasAnalysis.cpp b/llvm/lib/Analysis/BasicAliasAnalysis.cpp
index 16ee2ca49d0e..3f456db1c51a 100644
--- a/llvm/lib/Analysis/BasicAliasAnalysis.cpp
+++ b/llvm/lib/Analysis/BasicAliasAnalysis.cpp
@@ -44,6 +44,7 @@
 #include "llvm/IR/IntrinsicInst.h"
 #include "llvm/IR/Intrinsics.h"
 #include "llvm/IR/Operator.h"
+#include "llvm/IR/PatternMatch.h"
 #include "llvm/IR/Type.h"
 #include "llvm/IR/User.h"
 #include "llvm/IR/Value.h"
diff --git a/llvm/lib/Analysis/InstructionSimplify.cpp b/llvm/lib/Analysis/InstructionSimplify.cpp
index 1bba75d237fb..4061dae83c10 100644
--- a/llvm/lib/Analysis/InstructionSimplify.cpp
+++ b/llvm/lib/Analysis/InstructionSimplify.cpp
@@ -6281,11 +6281,11 @@ static Value *simplifyUnaryIntrinsic(Function *F, Value *Op0,
                m_Intrinsic<Intrinsic::pow>(m_SpecificFP(10.0), m_Value(X)))))
       return X;
     break;
-  case Intrinsic::experimental_vector_reverse:
-    // experimental.vector.reverse(experimental.vector.reverse(x)) -> x
+  case Intrinsic::vector_reverse:
+    // vector.reverse(vector.reverse(x)) -> x
     if (match(Op0, m_VecReverse(m_Value(X))))
       return X;
-    // experimental.vector.reverse(splat(X)) -> splat(X)
+    // vector.reverse(splat(X)) -> splat(X)
     if (isSplatValue(Op0))
       return Op0;
     break;
@@ -7238,6 +7238,14 @@ const SimplifyQuery getBestSimplifyQuery(AnalysisManager<T, TArgs...> &AM,
 }
 template const SimplifyQuery getBestSimplifyQuery(AnalysisManager<Function> &,
                                                   Function &);
+
+bool SimplifyQuery::isUndefValue(Value *V) const {
+  if (!CanUseUndef)
+    return false;
+
+  return match(V, m_Undef());
+}
+
 } // namespace llvm
 
 void InstSimplifyFolder::anchor() {}
diff --git a/llvm/lib/Analysis/Loads.cpp b/llvm/lib/Analysis/Loads.cpp
index ac508e19c9e0..478302d687b5 100644
--- a/llvm/lib/Analysis/Loads.cpp
+++ b/llvm/lib/Analysis/Loads.cpp
@@ -710,22 +710,62 @@ Value *llvm::FindAvailableLoadedValue(LoadInst *Load, BatchAAResults &AA,
   return Available;
 }
 
-bool llvm::canReplacePointersIfEqual(Value *A, Value *B, const DataLayout &DL,
-                                     Instruction *CtxI) {
-  Type *Ty = A->getType();
-  assert(Ty == B->getType() && Ty->isPointerTy() &&
-         "values must have matching pointer types");
-
-  // NOTE: The checks in the function are incomplete and currently miss illegal
-  // cases! The current implementation is a starting point and the
-  // implementation should be made stricter over time.
-  if (auto *C = dyn_cast<Constant>(B)) {
-    // Do not allow replacing a pointer with a constant pointer, unless it is
-    // either null or at least one byte is dereferenceable.
-    APInt OneByte(DL.getPointerTypeSizeInBits(Ty), 1);
-    return C->isNullValue() ||
-           isDereferenceableAndAlignedPointer(B, Align(1), OneByte, DL, CtxI);
+// Returns true if a use is either in an ICmp/PtrToInt or a Phi/Select that only
+// feeds into them.
+static bool isPointerUseReplacable(const Use &U) {
+  unsigned Limit = 40;
+  SmallVector<const User *> Worklist({U.getUser()});
+  SmallPtrSet<const User *, 8> Visited;
+
+  while (!Worklist.empty() && --Limit) {
+    auto *User = Worklist.pop_back_val();
+    if (!Visited.insert(User).second)
+      continue;
+    if (isa<ICmpInst, PtrToIntInst>(User))
+      continue;
+    if (isa<PHINode, SelectInst>(User))
+      Worklist.append(User->user_begin(), User->user_end());
+    else
+      return false;
   }
 
-  return true;
+  return Limit != 0;
+}
+
+// Returns true if `To` is a null pointer, constant dereferenceable pointer or
+// both pointers have the same underlying objects.
+static bool isPointerAlwaysReplaceable(const Value *From, const Value *To,
+                                       const DataLayout &DL) {
+  // This is not strictly correct, but we do it for now to retain important
+  // optimizations.
+  if (isa<ConstantPointerNull>(To))
+    return true;
+  if (isa<Constant>(To) &&
+      isDereferenceablePointer(To, Type::getInt8Ty(To->getContext()), DL))
+    return true;
+  if (getUnderlyingObject(From) == getUnderlyingObject(To))
+    return true;
+  return false;
+}
+
+bool llvm::canReplacePointersInUseIfEqual(const Use &U, const Value *To,
+                                          const DataLayout &DL) {
+  assert(U->getType() == To->getType() && "values must have matching types");
+  // Not a pointer, just return true.
+  if (!To->getType()->isPointerTy())
+    return true;
+
+  if (isPointerAlwaysReplaceable(&*U, To, DL))
+    return true;
+  return isPointerUseReplacable(U);
+}
+
+bool llvm::canReplacePointersIfEqual(const Value *From, const Value *To,
+                                     const DataLayout &DL) {
+  assert(From->getType() == To->getType() && "values must have matching types");
+  // Not a pointer, just return true.
+  if (!From->getType()->isPointerTy())
+    return true;
+
+  return isPointerAlwaysReplaceable(From, To, DL);
 }
diff --git a/llvm/lib/Analysis/LoopAccessAnalysis.cpp b/llvm/lib/Analysis/LoopAccessAnalysis.cpp
index b1ba8e7c0f60..f65515ca3872 100644
--- a/llvm/lib/Analysis/LoopAccessAnalysis.cpp
+++ b/llvm/lib/Analysis/LoopAccessAnalysis.cpp
@@ -1920,20 +1920,21 @@ isLoopVariantIndirectAddress(ArrayRef<const Value *> UnderlyingObjects,
 namespace {
 struct DepDistanceStrideAndSizeInfo {
   const SCEV *Dist;
-  uint64_t Stride;
+  uint64_t StrideA;
+  uint64_t StrideB;
   uint64_t TypeByteSize;
   bool AIsWrite;
   bool BIsWrite;
 
-  DepDistanceStrideAndSizeInfo(const SCEV *Dist, uint64_t Stride,
-                               uint64_t TypeByteSize, bool AIsWrite,
-                               bool BIsWrite)
-      : Dist(Dist), Stride(Stride), TypeByteSize(TypeByteSize),
-        AIsWrite(AIsWrite), BIsWrite(BIsWrite) {}
+  DepDistanceStrideAndSizeInfo(const SCEV *Dist, uint64_t StrideA,
+                               uint64_t StrideB, uint64_t TypeByteSize,
+                               bool AIsWrite, bool BIsWrite)
+      : Dist(Dist), StrideA(StrideA), StrideB(StrideB),
+        TypeByteSize(TypeByteSize), AIsWrite(AIsWrite), BIsWrite(BIsWrite) {}
 };
 } // namespace
 
-// Get the dependence distance, stride, type size and whether it is a write for
+// Get the dependence distance, strides, type size and whether it is a write for
 // the dependence between A and B. Returns a DepType, if we can prove there's
 // no dependence or the analysis fails. Outlined to lambda to limit he scope
 // of various temporary variables, like A/BPtr, StrideA/BPtr and others.
@@ -1995,10 +1996,11 @@ getDependenceDistanceStrideAndSize(
                                    InnermostLoop))
     return MemoryDepChecker::Dependence::IndirectUnsafe;
 
-  // Need accesses with constant stride. We don't want to vectorize
-  // "A[B[i]] += ..." and similar code or pointer arithmetic that could wrap
-  // in the address space.
-  if (!StrideAPtr || !StrideBPtr || StrideAPtr != StrideBPtr) {
+  // Need accesses with constant strides and the same direction. We don't want
+  // to vectorize "A[B[i]] += ..." and similar code or pointer arithmetic that
+  // could wrap in the address space.
+  if (!StrideAPtr || !StrideBPtr || (StrideAPtr > 0 && StrideBPtr < 0) ||
+      (StrideAPtr < 0 && StrideBPtr > 0)) {
     LLVM_DEBUG(dbgs() << "Pointer access with non-constant stride\n");
     return MemoryDepChecker::Dependence::Unknown;
   }
@@ -2008,9 +2010,9 @@ getDependenceDistanceStrideAndSize(
       DL.getTypeStoreSizeInBits(ATy) == DL.getTypeStoreSizeInBits(BTy);
   if (!HasSameSize)
     TypeByteSize = 0;
-  uint64_t Stride = std::abs(StrideAPtr);
-  return DepDistanceStrideAndSizeInfo(Dist, Stride, TypeByteSize, AIsWrite,
-                                      BIsWrite);
+  return DepDistanceStrideAndSizeInfo(Dist, std::abs(StrideAPtr),
+                                      std::abs(StrideBPtr), TypeByteSize,
+                                      AIsWrite, BIsWrite);
 }
 
 MemoryDepChecker::Dependence::DepType MemoryDepChecker::isDependent(
@@ -2028,41 +2030,63 @@ MemoryDepChecker::Dependence::DepType MemoryDepChecker::isDependent(
   if (std::holds_alternative<Dependence::DepType>(Res))
     return std::get<Dependence::DepType>(Res);
 
-  const auto &[Dist, Stride, TypeByteSize, AIsWrite, BIsWrite] =
+  const auto &[Dist, StrideA, StrideB, TypeByteSize, AIsWrite, BIsWrite] =
       std::get<DepDistanceStrideAndSizeInfo>(Res);
   bool HasSameSize = TypeByteSize > 0;
 
+  std::optional<uint64_t> CommonStride =
+      StrideA == StrideB ? std::make_optional(StrideA) : std::nullopt;
+  if (isa<SCEVCouldNotCompute>(Dist)) {
+    // TODO: Relax requirement that there is a common stride to retry with
+    // non-constant distance dependencies.
+    FoundNonConstantDistanceDependence |= !!CommonStride;
+    LLVM_DEBUG(dbgs() << "LAA: Dependence because of uncomputable distance.\n");
+    return Dependence::Unknown;
+  }
+
   ScalarEvolution &SE = *PSE.getSE();
   auto &DL = InnermostLoop->getHeader()->getModule()->getDataLayout();
+
   // If the distance between the acecsses is larger than their absolute stride
   // multiplied by the backedge taken count, the accesses are independet, i.e.
   // they are far enough appart that accesses won't access the same location
   // across all loop ierations.
-  if (!isa<SCEVCouldNotCompute>(Dist) && HasSameSize &&
+  if (HasSameSize && CommonStride &&
       isSafeDependenceDistance(DL, SE, *(PSE.getBackedgeTakenCount()), *Dist,
-                               Stride, TypeByteSize))
+                               *CommonStride, TypeByteSize))
     return Dependence::NoDep;
 
   const SCEVConstant *C = dyn_cast<SCEVConstant>(Dist);
-  if (!C) {
-    LLVM_DEBUG(dbgs() << "LAA: Dependence because of non-constant distance\n");
-    FoundNonConstantDistanceDependence = true;
-    return Dependence::Unknown;
-  }
 
-  const APInt &Val = C->getAPInt();
-  int64_t Distance = Val.getSExtValue();
-
-  // If the distance between accesses and their strides are known constants,
-  // check whether the accesses interlace each other.
-  if (std::abs(Distance) > 0 && Stride > 1 && HasSameSize &&
-      areStridedAccessesIndependent(std::abs(Distance), Stride, TypeByteSize)) {
-    LLVM_DEBUG(dbgs() << "LAA: Strided accesses are independent\n");
-    return Dependence::NoDep;
+  // Attempt to prove strided accesses independent.
+  if (C) {
+    const APInt &Val = C->getAPInt();
+    int64_t Distance = Val.getSExtValue();
+
+    // If the distance between accesses and their strides are known constants,
+    // check whether the accesses interlace each other.
+    if (std::abs(Distance) > 0 && CommonStride && *CommonStride > 1 &&
+        HasSameSize &&
+        areStridedAccessesIndependent(std::abs(Distance), *CommonStride,
+                                      TypeByteSize)) {
+      LLVM_DEBUG(dbgs() << "LAA: Strided accesses are independent\n");
+      return Dependence::NoDep;
+    }
   }
 
   // Negative distances are not plausible dependencies.
-  if (Val.isNegative()) {
+  if (SE.isKnownNonPositive(Dist)) {
+    if (SE.isKnownNonNegative(Dist)) {
+      if (HasSameSize) {
+        // Write to the same location with the same size.
+        return Dependence::Forward;
+      } else {
+        LLVM_DEBUG(dbgs() << "LAA: possibly zero dependence difference but "
+                             "different type sizes\n");
+        return Dependence::Unknown;
+      }
+    }
+
     bool IsTrueDataDependence = (AIsWrite && !BIsWrite);
     // Check if the first access writes to a location that is read in a later
     // iteration, where the distance between them is not a multiple of a vector
@@ -2071,27 +2095,40 @@ MemoryDepChecker::Dependence::DepType MemoryDepChecker::isDependent(
     // NOTE: There is no need to update MaxSafeVectorWidthInBits after call to
     // couldPreventStoreLoadForward, even if it changed MinDepDistBytes, since a
     // forward dependency will allow vectorization using any width.
-    if (IsTrueDataDependence && EnableForwardingConflictDetection &&
-        (!HasSameSize || couldPreventStoreLoadForward(Val.abs().getZExtValue(),
-                                                      TypeByteSize))) {
-      LLVM_DEBUG(dbgs() << "LAA: Forward but may prevent st->ld forwarding\n");
-      return Dependence::ForwardButPreventsForwarding;
+
+    if (IsTrueDataDependence && EnableForwardingConflictDetection) {
+      if (!C) {
+        // TODO: FoundNonConstantDistanceDependence is used as a necessary
+        // condition to consider retrying with runtime checks. Historically, we
+        // did not set it when strides were different but there is no inherent
+        // reason to.
+        FoundNonConstantDistanceDependence |= CommonStride.has_value();
+        return Dependence::Unknown;
+      }
+      if (!HasSameSize ||
+          couldPreventStoreLoadForward(C->getAPInt().abs().getZExtValue(),
+                                       TypeByteSize)) {
+        LLVM_DEBUG(
+            dbgs() << "LAA: Forward but may prevent st->ld forwarding\n");
+        return Dependence::ForwardButPreventsForwarding;
+      }
     }
 
     LLVM_DEBUG(dbgs() << "LAA: Dependence is negative\n");
     return Dependence::Forward;
   }
 
-  // Write to the same location with the same size.
-  if (Val == 0) {
-    if (HasSameSize)
-      return Dependence::Forward;
-    LLVM_DEBUG(
-        dbgs() << "LAA: Zero dependence difference but different type sizes\n");
+  if (!C) {
+    // TODO: FoundNonConstantDistanceDependence is used as a necessary condition
+    // to consider retrying with runtime checks. Historically, we did not set it
+    // when strides were different but there is no inherent reason to.
+    FoundNonConstantDistanceDependence |= CommonStride.has_value();
+    LLVM_DEBUG(dbgs() << "LAA: Dependence because of non-constant distance\n");
     return Dependence::Unknown;
   }
 
-  assert(Val.isStrictlyPositive() && "Expect a positive value");
+  if (!SE.isKnownPositive(Dist))
+    return Dependence::Unknown;
 
   if (!HasSameSize) {
     LLVM_DEBUG(dbgs() << "LAA: ReadWrite-Write positive dependency with "
@@ -2099,6 +2136,14 @@ MemoryDepChecker::Dependence::DepType MemoryDepChecker::isDependent(
     return Dependence::Unknown;
   }
 
+  // The logic below currently only supports StrideA ==  StrideB, i.e. there's a
+  // common stride.
+  if (!CommonStride)
+    return Dependence::Unknown;
+
+  const APInt &Val = C->getAPInt();
+  int64_t Distance = Val.getSExtValue();
+
   // Bail out early if passed-in parameters make vectorization not feasible.
   unsigned ForcedFactor = (VectorizerParams::VectorizationFactor ?
                            VectorizerParams::VectorizationFactor : 1);
@@ -2134,7 +2179,7 @@ MemoryDepChecker::Dependence::DepType MemoryDepChecker::isDependent(
   // the minimum distance needed is 28, which is greater than distance. It is
   // not safe to do vectorization.
   uint64_t MinDistanceNeeded =
-      TypeByteSize * Stride * (MinNumIter - 1) + TypeByteSize;
+      TypeByteSize * (*CommonStride) * (MinNumIter - 1) + TypeByteSize;
   if (MinDistanceNeeded > static_cast<uint64_t>(Distance)) {
     LLVM_DEBUG(dbgs() << "LAA: Failure because of positive distance "
                       << Distance << '\n');
@@ -2183,7 +2228,7 @@ MemoryDepChecker::Dependence::DepType MemoryDepChecker::isDependent(
 
   // An update to MinDepDistBytes requires an update to MaxSafeVectorWidthInBits
   // since there is a backwards dependency.
-  uint64_t MaxVF = MinDepDistBytes / (TypeByteSize * Stride);
+  uint64_t MaxVF = MinDepDistBytes / (TypeByteSize * (*CommonStride));
   LLVM_DEBUG(dbgs() << "LAA: Positive distance " << Val.getSExtValue()
                     << " with max VF = " << MaxVF << '\n');
   uint64_t MaxVFInBits = MaxVF * TypeByteSize * 8;
diff --git a/llvm/lib/Analysis/ValueTracking.cpp b/llvm/lib/Analysis/ValueTracking.cpp
index 21e3f8a4cc52..1b461e7cfd01 100644
--- a/llvm/lib/Analysis/ValueTracking.cpp
+++ b/llvm/lib/Analysis/ValueTracking.cpp
@@ -2632,6 +2632,13 @@ static bool isKnownNonZeroFromOperator(const Operator *I,
             Q.DL.getTypeSizeInBits(I->getType()).getFixedValue())
       return isKnownNonZero(I->getOperand(0), Q, Depth);
     break;
+  case Instruction::Trunc:
+    // nuw/nsw trunc preserves zero/non-zero status of input.
+    if (auto *TI = dyn_cast<TruncInst>(I))
+      if (TI->hasNoSignedWrap() || TI->hasNoUnsignedWrap())
+        return isKnownNonZero(TI->getOperand(0), Q, Depth);
+    break;
+
   case Instruction::Sub:
     return isNonZeroSub(DemandedElts, Depth, Q, BitWidth, I->getOperand(0),
                         I->getOperand(1));
@@ -6679,9 +6686,15 @@ llvm::computeConstantRangeIncludingKnownBits(const WithCache<const Value *> &V,
 
 OverflowResult llvm::computeOverflowForUnsignedMul(const Value *LHS,
                                                    const Value *RHS,
-                                                   const SimplifyQuery &SQ) {
+                                                   const SimplifyQuery &SQ,
+                                                   bool IsNSW) {
   KnownBits LHSKnown = computeKnownBits(LHS, /*Depth=*/0, SQ);
   KnownBits RHSKnown = computeKnownBits(RHS, /*Depth=*/0, SQ);
+
+  // mul nsw of two non-negative numbers is also nuw.
+  if (IsNSW && LHSKnown.isNonNegative() && RHSKnown.isNonNegative())
+    return OverflowResult::NeverOverflows;
+
   ConstantRange LHSRange = ConstantRange::fromKnownBits(LHSKnown, false);
   ConstantRange RHSRange = ConstantRange::fromKnownBits(RHSKnown, false);
   return mapOverflowResult(LHSRange.unsignedMulMayOverflow(RHSRange));
@@ -8042,17 +8055,27 @@ static SelectPatternResult matchMinMax(CmpInst::Predicate Pred,
   return {SPF_UNKNOWN, SPNB_NA, false};
 }
 
-bool llvm::isKnownNegation(const Value *X, const Value *Y, bool NeedNSW) {
+bool llvm::isKnownNegation(const Value *X, const Value *Y, bool NeedNSW,
+                           bool AllowPoison) {
   assert(X && Y && "Invalid operand");
 
-  // X = sub (0, Y) || X = sub nsw (0, Y)
-  if ((!NeedNSW && match(X, m_Sub(m_ZeroInt(), m_Specific(Y)))) ||
-      (NeedNSW && match(X, m_NSWNeg(m_Specific(Y)))))
+  auto IsNegationOf = [&](const Value *X, const Value *Y) {
+    if (!match(X, m_Neg(m_Specific(Y))))
+      return false;
+
+    auto *BO = cast<BinaryOperator>(X);
+    if (NeedNSW && !BO->hasNoSignedWrap())
+      return false;
+
+    auto *Zero = cast<Constant>(BO->getOperand(0));
+    if (!AllowPoison && !Zero->isNullValue())
+      return false;
+
     return true;
+  };
 
-  // Y = sub (0, X) || Y = sub nsw (0, X)
-  if ((!NeedNSW && match(Y, m_Sub(m_ZeroInt(), m_Specific(X)))) ||
-      (NeedNSW && match(Y, m_NSWNeg(m_Specific(X)))))
+  // X = -Y or Y = -X
+  if (IsNegationOf(X, Y) || IsNegationOf(Y, X))
     return true;
 
   // X = sub (A, B), Y = sub (B, A) || X = sub nsw (A, B), Y = sub nsw (B, A)
diff --git a/llvm/lib/Analysis/VectorUtils.cpp b/llvm/lib/Analysis/VectorUtils.cpp
index bf7bc0ba84a0..917094267d05 100644
--- a/llvm/lib/Analysis/VectorUtils.cpp
+++ b/llvm/lib/Analysis/VectorUtils.cpp
@@ -23,6 +23,7 @@
 #include "llvm/IR/Constants.h"
 #include "llvm/IR/DerivedTypes.h"
 #include "llvm/IR/IRBuilder.h"
+#include "llvm/IR/MemoryModelRelaxationAnnotations.h"
 #include "llvm/IR/PatternMatch.h"
 #include "llvm/IR/Value.h"
 #include "llvm/Support/CommandLine.h"
@@ -793,13 +794,17 @@ Instruction *llvm::propagateMetadata(Instruction *Inst, ArrayRef<Value *> VL) {
   for (auto Kind : {LLVMContext::MD_tbaa, LLVMContext::MD_alias_scope,
                     LLVMContext::MD_noalias, LLVMContext::MD_fpmath,
                     LLVMContext::MD_nontemporal, LLVMContext::MD_invariant_load,
-                    LLVMContext::MD_access_group}) {
+                    LLVMContext::MD_access_group, LLVMContext::MD_mmra}) {
     MDNode *MD = I0->getMetadata(Kind);
-
     for (int J = 1, E = VL.size(); MD && J != E; ++J) {
       const Instruction *IJ = cast<Instruction>(VL[J]);
       MDNode *IMD = IJ->getMetadata(Kind);
+
       switch (Kind) {
+      case LLVMContext::MD_mmra: {
+        MD = MMRAMetadata::combine(Inst->getContext(), MD, IMD);
+        break;
+      }
       case LLVMContext::MD_tbaa:
         MD = MDNode::getMostGenericTBAA(MD, IMD);
         break;
diff --git a/llvm/lib/BinaryFormat/Dwarf.cpp b/llvm/lib/BinaryFormat/Dwarf.cpp
index e4e5b5dd8c0e..732426617268 100644
--- a/llvm/lib/BinaryFormat/Dwarf.cpp
+++ b/llvm/lib/BinaryFormat/Dwarf.cpp
@@ -411,6 +411,16 @@ llvm::dwarf::LanguageLowerBound(dwarf::SourceLanguage Lang) {
   }
 }
 
+StringRef llvm::dwarf::LanguageDescription(dwarf::SourceLanguageName lname) {
+  switch (lname) {
+#define HANDLE_DW_LNAME(ID, NAME, DESC, LOWER_BOUND)                           \
+  case DW_LNAME_##NAME:                                                        \
+    return DESC;
+#include "llvm/BinaryFormat/Dwarf.def"
+  }
+  return "Unknown";
+}
+
 StringRef llvm::dwarf::CaseString(unsigned Case) {
   switch (Case) {
   case DW_ID_case_sensitive:
diff --git a/llvm/lib/CodeGen/AggressiveAntiDepBreaker.cpp b/llvm/lib/CodeGen/AggressiveAntiDepBreaker.cpp
index ed6ce6bc73d3..e91750afd281 100644
--- a/llvm/lib/CodeGen/AggressiveAntiDepBreaker.cpp
+++ b/llvm/lib/CodeGen/AggressiveAntiDepBreaker.cpp
@@ -231,9 +231,9 @@ bool AggressiveAntiDepBreaker::IsImplicitDefUse(MachineInstr &MI,
 
   MachineOperand *Op = nullptr;
   if (MO.isDef())
-    Op = MI.findRegisterUseOperand(Reg, true);
+    Op = MI.findRegisterUseOperand(Reg, /*TRI=*/nullptr, true);
   else
-    Op = MI.findRegisterDefOperand(Reg);
+    Op = MI.findRegisterDefOperand(Reg, /*TRI=*/nullptr);
 
   return(Op && Op->isImplicit());
 }
@@ -679,7 +679,7 @@ bool AggressiveAntiDepBreaker::FindSuitableFreeRegisters(
       // defines 'NewReg' via an early-clobber operand.
       for (const auto &Q : make_range(RegRefs.equal_range(Reg))) {
         MachineInstr *UseMI = Q.second.Operand->getParent();
-        int Idx = UseMI->findRegisterDefOperandIdx(NewReg, false, true, TRI);
+        int Idx = UseMI->findRegisterDefOperandIdx(NewReg, TRI, false, true);
         if (Idx == -1)
           continue;
 
@@ -846,7 +846,8 @@ unsigned AggressiveAntiDepBreaker::BreakAntiDependencies(
           continue;
         } else {
           // No anti-dep breaking for implicit deps
-          MachineOperand *AntiDepOp = MI.findRegisterDefOperand(AntiDepReg);
+          MachineOperand *AntiDepOp =
+              MI.findRegisterDefOperand(AntiDepReg, /*TRI=*/nullptr);
           assert(AntiDepOp && "Can't find index for defined register operand");
           if (!AntiDepOp || AntiDepOp->isImplicit()) {
             LLVM_DEBUG(dbgs() << " (implicit)\n");
diff --git a/llvm/lib/CodeGen/AsmPrinter/DwarfCompileUnit.cpp b/llvm/lib/CodeGen/AsmPrinter/DwarfCompileUnit.cpp
index 14f2a363f9be..6022afbae574 100644
--- a/llvm/lib/CodeGen/AsmPrinter/DwarfCompileUnit.cpp
+++ b/llvm/lib/CodeGen/AsmPrinter/DwarfCompileUnit.cpp
@@ -32,6 +32,7 @@
 #include "llvm/MC/MCSymbol.h"
 #include "llvm/MC/MCSymbolWasm.h"
 #include "llvm/MC/MachineLocation.h"
+#include "llvm/Support/CommandLine.h"
 #include "llvm/Target/TargetLoweringObjectFile.h"
 #include "llvm/Target/TargetMachine.h"
 #include "llvm/Target/TargetOptions.h"
@@ -42,6 +43,20 @@
 
 using namespace llvm;
 
+/// Query value using AddLinkageNamesToDeclCallOriginsForTuning.
+cl::opt<cl::boolOrDefault> AddLinkageNamesToDeclCallOrigins(
+    "add-linkage-names-to-declaration-call-origins", cl::Hidden,
+    cl::desc("Add DW_AT_linkage_name to function declaration DIEs "
+             "referenced by DW_AT_call_origin attributes. Enabled by default "
+             "for -gsce debugger tuning."));
+
+static bool AddLinkageNamesToDeclCallOriginsForTuning(const DwarfDebug *DD) {
+  bool EnabledByDefault = DD->tuneForSCE();
+  if (EnabledByDefault)
+    return AddLinkageNamesToDeclCallOrigins != cl::boolOrDefault::BOU_FALSE;
+  return AddLinkageNamesToDeclCallOrigins == cl::boolOrDefault::BOU_TRUE;
+}
+
 static dwarf::Tag GetCompileUnitType(UnitKind Kind, DwarfDebug *DW) {
 
   //  According to DWARF Debugging Information Format Version 5,
@@ -1260,6 +1275,12 @@ DIE &DwarfCompileUnit::constructCallSiteEntryDIE(DIE &ScopeDIE,
   } else {
     DIE *CalleeDIE = getOrCreateSubprogramDIE(CalleeSP);
     assert(CalleeDIE && "Could not create DIE for call site entry origin");
+    if (AddLinkageNamesToDeclCallOriginsForTuning(DD) &&
+        !CalleeSP->isDefinition() &&
+        !CalleeDIE->findAttribute(dwarf::DW_AT_linkage_name)) {
+      addLinkageName(*CalleeDIE, CalleeSP->getLinkageName());
+    }
+
     addDIEEntry(CallSiteDIE, getDwarf5OrGNUAttr(dwarf::DW_AT_call_origin),
                 *CalleeDIE);
   }
diff --git a/llvm/lib/CodeGen/AtomicExpandPass.cpp b/llvm/lib/CodeGen/AtomicExpandPass.cpp
index 0aa89ea94335..f3b8097396e2 100644
--- a/llvm/lib/CodeGen/AtomicExpandPass.cpp
+++ b/llvm/lib/CodeGen/AtomicExpandPass.cpp
@@ -37,6 +37,7 @@
 #include "llvm/IR/InstIterator.h"
 #include "llvm/IR/Instruction.h"
 #include "llvm/IR/Instructions.h"
+#include "llvm/IR/MemoryModelRelaxationAnnotations.h"
 #include "llvm/IR/Module.h"
 #include "llvm/IR/Type.h"
 #include "llvm/IR/User.h"
@@ -133,15 +134,27 @@ public:
 };
 
 // IRBuilder to be used for replacement atomic instructions.
-struct ReplacementIRBuilder : IRBuilder<InstSimplifyFolder> {
+struct ReplacementIRBuilder
+    : IRBuilder<InstSimplifyFolder, IRBuilderCallbackInserter> {
+  MDNode *MMRAMD = nullptr;
+
   // Preserves the DebugLoc from I, and preserves still valid metadata.
   // Enable StrictFP builder mode when appropriate.
   explicit ReplacementIRBuilder(Instruction *I, const DataLayout &DL)
-      : IRBuilder(I->getContext(), DL) {
+      : IRBuilder(I->getContext(), DL,
+                  IRBuilderCallbackInserter(
+                      [this](Instruction *I) { addMMRAMD(I); })) {
     SetInsertPoint(I);
     this->CollectMetadataToCopy(I, {LLVMContext::MD_pcsections});
     if (BB->getParent()->getAttributes().hasFnAttr(Attribute::StrictFP))
       this->setIsFPConstrained(true);
+
+    MMRAMD = I->getMetadata(LLVMContext::MD_mmra);
+  }
+
+  void addMMRAMD(Instruction *I) {
+    if (canInstructionHaveMMRAs(*I))
+      I->setMetadata(LLVMContext::MD_mmra, MMRAMD);
   }
 };
 
@@ -421,8 +434,9 @@ AtomicExpandImpl::convertAtomicXchgToIntegerType(AtomicRMWInst *RMWI) {
                       ? Builder.CreatePtrToInt(Val, NewTy)
                       : Builder.CreateBitCast(Val, NewTy);
 
-  auto *NewRMWI = Builder.CreateAtomicRMW(
-      AtomicRMWInst::Xchg, Addr, NewVal, RMWI->getAlign(), RMWI->getOrdering());
+  auto *NewRMWI = Builder.CreateAtomicRMW(AtomicRMWInst::Xchg, Addr, NewVal,
+                                          RMWI->getAlign(), RMWI->getOrdering(),
+                                          RMWI->getSyncScopeID());
   NewRMWI->setVolatile(RMWI->isVolatile());
   LLVM_DEBUG(dbgs() << "Replaced " << *RMWI << " with " << *NewRMWI << "\n");
 
@@ -943,7 +957,7 @@ AtomicRMWInst *AtomicExpandImpl::widenPartwordAtomicRMW(AtomicRMWInst *AI) {
 
   if (Op == AtomicRMWInst::And)
     NewOperand =
-        Builder.CreateOr(PMV.Inv_Mask, ValOperand_Shifted, "AndOperand");
+        Builder.CreateOr(ValOperand_Shifted, PMV.Inv_Mask, "AndOperand");
   else
     NewOperand = ValOperand_Shifted;
 
diff --git a/llvm/lib/CodeGen/CalcSpillWeights.cpp b/llvm/lib/CodeGen/CalcSpillWeights.cpp
index f3cb7fa5af61..1d767a3484bc 100644
--- a/llvm/lib/CodeGen/CalcSpillWeights.cpp
+++ b/llvm/lib/CodeGen/CalcSpillWeights.cpp
@@ -251,7 +251,8 @@ float VirtRegAuxInfo::weightCalcHelper(LiveInterval &LI, SlotIndex *Start,
 
     // For terminators that produce values, ask the backend if the register is
     // not spillable.
-    if (TII.isUnspillableTerminator(MI) && MI->definesRegister(LI.reg())) {
+    if (TII.isUnspillableTerminator(MI) &&
+        MI->definesRegister(LI.reg(), /*TRI=*/nullptr)) {
       LI.markNotSpillable();
       return -1.0f;
     }
diff --git a/llvm/lib/CodeGen/CodeGenCommonISel.cpp b/llvm/lib/CodeGen/CodeGenCommonISel.cpp
index 577c5dbc8e2d..fe144d3c1820 100644
--- a/llvm/lib/CodeGen/CodeGenCommonISel.cpp
+++ b/llvm/lib/CodeGen/CodeGenCommonISel.cpp
@@ -260,7 +260,8 @@ void llvm::salvageDebugInfoForDbgValue(const MachineRegisterInfo &MRI,
       continue;
     }
 
-    int UseMOIdx = DbgMI->findRegisterUseOperandIdx(DefMO->getReg());
+    int UseMOIdx =
+        DbgMI->findRegisterUseOperandIdx(DefMO->getReg(), /*TRI=*/nullptr);
     assert(UseMOIdx != -1 && DbgMI->hasDebugOperandForReg(DefMO->getReg()) &&
            "Must use salvaged instruction as its location");
 
diff --git a/llvm/lib/CodeGen/CodeGenPrepare.cpp b/llvm/lib/CodeGen/CodeGenPrepare.cpp
index 22a766f8d625..339a1f1f2f00 100644
--- a/llvm/lib/CodeGen/CodeGenPrepare.cpp
+++ b/llvm/lib/CodeGen/CodeGenPrepare.cpp
@@ -1431,10 +1431,8 @@ static bool SinkCast(CastInst *CI) {
     if (!InsertedCast) {
       BasicBlock::iterator InsertPt = UserBB->getFirstInsertionPt();
       assert(InsertPt != UserBB->end());
-      InsertedCast = CastInst::Create(CI->getOpcode(), CI->getOperand(0),
-                                      CI->getType(), "");
+      InsertedCast = cast<CastInst>(CI->clone());
       InsertedCast->insertBefore(*UserBB, InsertPt);
-      InsertedCast->setDebugLoc(CI->getDebugLoc());
     }
 
     // Replace a use of the cast with a use of the new cast.
@@ -8272,6 +8270,7 @@ static bool optimizeBranch(BranchInst *Branch, const TargetLowering &TLI,
       IRBuilder<> Builder(Branch);
       if (UI->getParent() != Branch->getParent())
         UI->moveBefore(Branch);
+      UI->dropPoisonGeneratingFlags();
       Value *NewCmp = Builder.CreateCmp(ICmpInst::ICMP_EQ, UI,
                                         ConstantInt::get(UI->getType(), 0));
       LLVM_DEBUG(dbgs() << "Converting " << *Cmp << "\n");
@@ -8285,6 +8284,7 @@ static bool optimizeBranch(BranchInst *Branch, const TargetLowering &TLI,
       IRBuilder<> Builder(Branch);
       if (UI->getParent() != Branch->getParent())
         UI->moveBefore(Branch);
+      UI->dropPoisonGeneratingFlags();
       Value *NewCmp = Builder.CreateCmp(Cmp->getPredicate(), UI,
                                         ConstantInt::get(UI->getType(), 0));
       LLVM_DEBUG(dbgs() << "Converting " << *Cmp << "\n");
diff --git a/llvm/lib/CodeGen/ComplexDeinterleavingPass.cpp b/llvm/lib/CodeGen/ComplexDeinterleavingPass.cpp
index 031a271de5bd..8573b016d1e5 100644
--- a/llvm/lib/CodeGen/ComplexDeinterleavingPass.cpp
+++ b/llvm/lib/CodeGen/ComplexDeinterleavingPass.cpp
@@ -1639,8 +1639,7 @@ bool ComplexDeinterleavingGraph::checkNodes() {
 ComplexDeinterleavingGraph::NodePtr
 ComplexDeinterleavingGraph::identifyRoot(Instruction *RootI) {
   if (auto *Intrinsic = dyn_cast<IntrinsicInst>(RootI)) {
-    if (Intrinsic->getIntrinsicID() !=
-        Intrinsic::experimental_vector_interleave2)
+    if (Intrinsic->getIntrinsicID() != Intrinsic::vector_interleave2)
       return nullptr;
 
     auto *Real = dyn_cast<Instruction>(Intrinsic->getOperand(0));
@@ -1675,7 +1674,7 @@ ComplexDeinterleavingGraph::identifyDeinterleave(Instruction *Real,
   Value *FinalValue = nullptr;
   if (match(Real, m_ExtractValue<0>(m_Instruction(I))) &&
       match(Imag, m_ExtractValue<1>(m_Specific(I))) &&
-      match(I, m_Intrinsic<Intrinsic::experimental_vector_deinterleave2>(
+      match(I, m_Intrinsic<Intrinsic::vector_deinterleave2>(
                    m_Value(FinalValue)))) {
     NodePtr PlaceholderNode = prepareCompositeNode(
         llvm::ComplexDeinterleavingOperation::Deinterleave, Real, Imag);
@@ -1960,13 +1959,11 @@ Value *ComplexDeinterleavingGraph::replaceNode(IRBuilderBase &Builder,
       // Splats that are not constant are interleaved where they are located
       Instruction *InsertPoint = (I->comesBefore(R) ? R : I)->getNextNode();
       IRBuilder<> IRB(InsertPoint);
-      ReplacementNode =
-          IRB.CreateIntrinsic(Intrinsic::experimental_vector_interleave2, NewTy,
-                              {Node->Real, Node->Imag});
+      ReplacementNode = IRB.CreateIntrinsic(Intrinsic::vector_interleave2,
+                                            NewTy, {Node->Real, Node->Imag});
     } else {
-      ReplacementNode =
-          Builder.CreateIntrinsic(Intrinsic::experimental_vector_interleave2,
-                                  NewTy, {Node->Real, Node->Imag});
+      ReplacementNode = Builder.CreateIntrinsic(
+          Intrinsic::vector_interleave2, NewTy, {Node->Real, Node->Imag});
     }
     break;
   }
@@ -1991,9 +1988,8 @@ Value *ComplexDeinterleavingGraph::replaceNode(IRBuilderBase &Builder,
     auto *B = replaceNode(Builder, Node->Operands[1]);
     auto *NewMaskTy = VectorType::getDoubleElementsVectorType(
         cast<VectorType>(MaskReal->getType()));
-    auto *NewMask =
-        Builder.CreateIntrinsic(Intrinsic::experimental_vector_interleave2,
-                                NewMaskTy, {MaskReal, MaskImag});
+    auto *NewMask = Builder.CreateIntrinsic(Intrinsic::vector_interleave2,
+                                            NewMaskTy, {MaskReal, MaskImag});
     ReplacementNode = Builder.CreateSelect(NewMask, A, B);
     break;
   }
@@ -2021,8 +2017,8 @@ void ComplexDeinterleavingGraph::processReductionOperation(
   Value *InitImag = OldPHIImag->getIncomingValueForBlock(Incoming);
 
   IRBuilder<> Builder(Incoming->getTerminator());
-  auto *NewInit = Builder.CreateIntrinsic(
-      Intrinsic::experimental_vector_interleave2, NewVTy, {InitReal, InitImag});
+  auto *NewInit = Builder.CreateIntrinsic(Intrinsic::vector_interleave2, NewVTy,
+                                          {InitReal, InitImag});
 
   NewPHI->addIncoming(NewInit, Incoming);
   NewPHI->addIncoming(OperationReplacement, BackEdge);
@@ -2034,9 +2030,9 @@ void ComplexDeinterleavingGraph::processReductionOperation(
 
   Builder.SetInsertPoint(
       &*FinalReductionReal->getParent()->getFirstInsertionPt());
-  auto *Deinterleave = Builder.CreateIntrinsic(
-      Intrinsic::experimental_vector_deinterleave2,
-      OperationReplacement->getType(), OperationReplacement);
+  auto *Deinterleave = Builder.CreateIntrinsic(Intrinsic::vector_deinterleave2,
+                                               OperationReplacement->getType(),
+                                               OperationReplacement);
 
   auto *NewReal = Builder.CreateExtractValue(Deinterleave, (uint64_t)0);
   FinalReductionReal->replaceUsesOfWith(Real, NewReal);
diff --git a/llvm/lib/CodeGen/EarlyIfConversion.cpp b/llvm/lib/CodeGen/EarlyIfConversion.cpp
index 31e107ade1cc..2a7bee1618de 100644
--- a/llvm/lib/CodeGen/EarlyIfConversion.cpp
+++ b/llvm/lib/CodeGen/EarlyIfConversion.cpp
@@ -599,8 +599,8 @@ static bool hasSameValue(const MachineRegisterInfo &MRI,
     return false;
 
   // Further, check that the two defs come from corresponding operands.
-  int TIdx = TDef->findRegisterDefOperandIdx(TReg);
-  int FIdx = FDef->findRegisterDefOperandIdx(FReg);
+  int TIdx = TDef->findRegisterDefOperandIdx(TReg, /*TRI=*/nullptr);
+  int FIdx = FDef->findRegisterDefOperandIdx(FReg, /*TRI=*/nullptr);
   if (TIdx == -1 || FIdx == -1)
     return false;
 
diff --git a/llvm/lib/CodeGen/FixupStatepointCallerSaved.cpp b/llvm/lib/CodeGen/FixupStatepointCallerSaved.cpp
index 4d668c53f715..3bb9da5f1a37 100644
--- a/llvm/lib/CodeGen/FixupStatepointCallerSaved.cpp
+++ b/llvm/lib/CodeGen/FixupStatepointCallerSaved.cpp
@@ -112,7 +112,7 @@ static Register performCopyPropagation(Register Reg,
                                        bool &IsKill, const TargetInstrInfo &TII,
                                        const TargetRegisterInfo &TRI) {
   // First check if statepoint itself uses Reg in non-meta operands.
-  int Idx = RI->findRegisterUseOperandIdx(Reg, false, &TRI);
+  int Idx = RI->findRegisterUseOperandIdx(Reg, &TRI, false);
   if (Idx >= 0 && (unsigned)Idx < StatepointOpers(&*RI).getNumDeoptArgsIdx()) {
     IsKill = false;
     return Reg;
diff --git a/llvm/lib/CodeGen/GlobalISel/CombinerHelper.cpp b/llvm/lib/CodeGen/GlobalISel/CombinerHelper.cpp
index 5545ec3b3ed0..653e7689b577 100644
--- a/llvm/lib/CodeGen/GlobalISel/CombinerHelper.cpp
+++ b/llvm/lib/CodeGen/GlobalISel/CombinerHelper.cpp
@@ -2800,8 +2800,8 @@ bool CombinerHelper::matchEqualDefs(const MachineOperand &MOP1,
     // %5:_(s8), %6:_(s8), %7:_(s8), %8:_(s8) = G_UNMERGE_VALUES %4:_(<4 x s8>)
     // I1 and I2 are different instructions but produce same values,
     // %1 and %6 are same, %1 and %7 are not the same value.
-    return I1->findRegisterDefOperandIdx(InstAndDef1->Reg) ==
-           I2->findRegisterDefOperandIdx(InstAndDef2->Reg);
+    return I1->findRegisterDefOperandIdx(InstAndDef1->Reg, /*TRI=*/nullptr) ==
+           I2->findRegisterDefOperandIdx(InstAndDef2->Reg, /*TRI=*/nullptr);
   }
   return false;
 }
@@ -5069,6 +5069,9 @@ MachineInstr *CombinerHelper::buildUDivUsingMul(MachineInstr &MI) {
   const unsigned EltBits = ScalarTy.getScalarSizeInBits();
   LLT ShiftAmtTy = getTargetLowering().getPreferredShiftAmountTy(Ty);
   LLT ScalarShiftAmtTy = ShiftAmtTy.getScalarType();
+
+  unsigned KnownLeadingZeros =
+      KB ? KB->getKnownBits(LHS).countMinLeadingZeros() : 0;
   auto &MIB = Builder;
 
   bool UseNPQ = false;
@@ -5086,8 +5089,12 @@ MachineInstr *CombinerHelper::buildUDivUsingMul(MachineInstr &MI) {
     // at the end.
     // TODO: Use undef values for divisor of 1.
     if (!Divisor.isOne()) {
+
+      // UnsignedDivisionByConstantInfo doesn't work correctly if leading zeros
+      // in the dividend exceeds the leading zeros for the divisor.
       UnsignedDivisionByConstantInfo magics =
-          UnsignedDivisionByConstantInfo::get(Divisor);
+          UnsignedDivisionByConstantInfo::get(
+              Divisor, std::min(KnownLeadingZeros, Divisor.countl_zero()));
 
       Magic = std::move(magics.Magic);
 
diff --git a/llvm/lib/CodeGen/GlobalISel/CombinerHelperVectorOps.cpp b/llvm/lib/CodeGen/GlobalISel/CombinerHelperVectorOps.cpp
index 123bf21f657c..fb33801a3a33 100644
--- a/llvm/lib/CodeGen/GlobalISel/CombinerHelperVectorOps.cpp
+++ b/llvm/lib/CodeGen/GlobalISel/CombinerHelperVectorOps.cpp
@@ -77,7 +77,7 @@ bool CombinerHelper::matchExtractVectorElement(MachineInstr &MI,
 
   // Fold extractVectorElement(Vector, TOOLARGE) -> undef
   if (IndexC && VectorTy.isFixedVector() &&
-      IndexC->getZExtValue() >= VectorTy.getNumElements() &&
+      IndexC->uge(VectorTy.getNumElements()) &&
       isLegalOrBeforeLegalizer({TargetOpcode::G_IMPLICIT_DEF, {DstTy}})) {
     // For fixed-length vectors, it's invalid to extract out-of-range elements.
     MatchInfo = [=](MachineIRBuilder &B) { B.buildUndef(Dst); };
@@ -324,3 +324,26 @@ bool CombinerHelper::matchExtractVectorElementWithBuildVectorTrunc(
 
   return true;
 }
+
+bool CombinerHelper::matchInsertVectorElementOOB(MachineInstr &MI,
+                                                 BuildFnTy &MatchInfo) {
+  GInsertVectorElement *Insert = cast<GInsertVectorElement>(&MI);
+
+  Register Dst = Insert->getReg(0);
+  LLT DstTy = MRI.getType(Dst);
+  Register Index = Insert->getIndexReg();
+
+  if (!DstTy.isFixedVector())
+    return false;
+
+  std::optional<ValueAndVReg> MaybeIndex =
+      getIConstantVRegValWithLookThrough(Index, MRI);
+
+  if (MaybeIndex && MaybeIndex->Value.uge(DstTy.getNumElements()) &&
+      isLegalOrBeforeLegalizer({TargetOpcode::G_IMPLICIT_DEF, {DstTy}})) {
+    MatchInfo = [=](MachineIRBuilder &B) { B.buildUndef(Dst); };
+    return true;
+  }
+
+  return false;
+}
diff --git a/llvm/lib/CodeGen/GlobalISel/GISelKnownBits.cpp b/llvm/lib/CodeGen/GlobalISel/GISelKnownBits.cpp
index 51ab7b6262c6..529e50c8ebe0 100644
--- a/llvm/lib/CodeGen/GlobalISel/GISelKnownBits.cpp
+++ b/llvm/lib/CodeGen/GlobalISel/GISelKnownBits.cpp
@@ -694,6 +694,20 @@ unsigned GISelKnownBits::computeNumSignBits(Register R,
     const MachineMemOperand *MMO = *MI.memoperands_begin();
     return TyBits - MMO->getSizeInBits().getValue();
   }
+  case TargetOpcode::G_AND:
+  case TargetOpcode::G_OR:
+  case TargetOpcode::G_XOR: {
+    Register Src1 = MI.getOperand(1).getReg();
+    unsigned Src1NumSignBits =
+        computeNumSignBits(Src1, DemandedElts, Depth + 1);
+    if (Src1NumSignBits != 1) {
+      Register Src2 = MI.getOperand(2).getReg();
+      unsigned Src2NumSignBits =
+          computeNumSignBits(Src2, DemandedElts, Depth + 1);
+      FirstAnswer = std::min(Src1NumSignBits, Src2NumSignBits);
+    }
+    break;
+  }
   case TargetOpcode::G_TRUNC: {
     Register Src = MI.getOperand(1).getReg();
     LLT SrcTy = MRI.getType(Src);
diff --git a/llvm/lib/CodeGen/GlobalISel/IRTranslator.cpp b/llvm/lib/CodeGen/GlobalISel/IRTranslator.cpp
index 0b6aae375975..e26c6ca3d616 100644
--- a/llvm/lib/CodeGen/GlobalISel/IRTranslator.cpp
+++ b/llvm/lib/CodeGen/GlobalISel/IRTranslator.cpp
@@ -1804,7 +1804,7 @@ bool IRTranslator::translateTrap(const CallInst &CI,
 
 bool IRTranslator::translateVectorInterleave2Intrinsic(
     const CallInst &CI, MachineIRBuilder &MIRBuilder) {
-  assert(CI.getIntrinsicID() == Intrinsic::experimental_vector_interleave2 &&
+  assert(CI.getIntrinsicID() == Intrinsic::vector_interleave2 &&
          "This function can only be called on the interleave2 intrinsic!");
   // Canonicalize interleave2 to G_SHUFFLE_VECTOR (similar to SelectionDAG).
   Register Op0 = getOrCreateVReg(*CI.getOperand(0));
@@ -1820,7 +1820,7 @@ bool IRTranslator::translateVectorInterleave2Intrinsic(
 
 bool IRTranslator::translateVectorDeinterleave2Intrinsic(
     const CallInst &CI, MachineIRBuilder &MIRBuilder) {
-  assert(CI.getIntrinsicID() == Intrinsic::experimental_vector_deinterleave2 &&
+  assert(CI.getIntrinsicID() == Intrinsic::vector_deinterleave2 &&
          "This function can only be called on the deinterleave2 intrinsic!");
   // Canonicalize deinterleave2 to shuffles that extract sub-vectors (similar to
   // SelectionDAG).
@@ -2223,7 +2223,7 @@ bool IRTranslator::translateKnownIntrinsic(const CallInst &CI, Intrinsic::ID ID,
     // addresses. We can treat it like a normal dbg_value intrinsic here; to
     // benefit from the full analysis of stack/SSA locations, GlobalISel would
     // need to register for and use the AssignmentTrackingAnalysis pass.
-    LLVM_FALLTHROUGH;
+    [[fallthrough]];
   case Intrinsic::dbg_value: {
     // This form of DBG_VALUE is target-independent.
     const DbgValueInst &DI = cast<DbgValueInst>(CI);
@@ -2572,15 +2572,15 @@ bool IRTranslator::translateKnownIntrinsic(const CallInst &CI, Intrinsic::ID ID,
     return true;
   }
 
-  case Intrinsic::experimental_vector_interleave2:
-  case Intrinsic::experimental_vector_deinterleave2: {
+  case Intrinsic::vector_interleave2:
+  case Intrinsic::vector_deinterleave2: {
     // Both intrinsics have at least one operand.
     Value *Op0 = CI.getOperand(0);
     LLT ResTy = getLLTForType(*Op0->getType(), MIRBuilder.getDataLayout());
     if (!ResTy.isFixedVector())
       return false;
 
-    if (CI.getIntrinsicID() == Intrinsic::experimental_vector_interleave2)
+    if (CI.getIntrinsicID() == Intrinsic::vector_interleave2)
       return translateVectorInterleave2Intrinsic(CI, MIRBuilder);
 
     return translateVectorDeinterleave2Intrinsic(CI, MIRBuilder);
@@ -3446,6 +3446,7 @@ void IRTranslator::translateDbgInfo(const Instruction &Inst,
 bool IRTranslator::translate(const Instruction &Inst) {
   CurBuilder->setDebugLoc(Inst.getDebugLoc());
   CurBuilder->setPCSections(Inst.getMetadata(LLVMContext::MD_pcsections));
+  CurBuilder->setMMRAMetadata(Inst.getMetadata(LLVMContext::MD_mmra));
 
   if (TLI->fallBackToDAGISel(Inst))
     return false;
diff --git a/llvm/lib/CodeGen/GlobalISel/LegalizerHelper.cpp b/llvm/lib/CodeGen/GlobalISel/LegalizerHelper.cpp
index d55091e2e717..6a76ad7f5db7 100644
--- a/llvm/lib/CodeGen/GlobalISel/LegalizerHelper.cpp
+++ b/llvm/lib/CodeGen/GlobalISel/LegalizerHelper.cpp
@@ -4764,6 +4764,8 @@ LegalizerHelper::fewerElementsVector(MachineInstr &MI, unsigned TypeIdx,
     return fewerElementsVectorMultiEltType(GMI, NumElts, {2 /*pow*/});
   case G_BITCAST:
     return fewerElementsBitcast(MI, TypeIdx, NarrowTy);
+  case G_INTRINSIC_FPTRUNC_ROUND:
+    return fewerElementsVectorMultiEltType(GMI, NumElts, {2});
   default:
     return UnableToLegalize;
   }
diff --git a/llvm/lib/CodeGen/GlobalISel/MachineIRBuilder.cpp b/llvm/lib/CodeGen/GlobalISel/MachineIRBuilder.cpp
index 6b35caf83491..2e8407813ba6 100644
--- a/llvm/lib/CodeGen/GlobalISel/MachineIRBuilder.cpp
+++ b/llvm/lib/CodeGen/GlobalISel/MachineIRBuilder.cpp
@@ -28,6 +28,7 @@ void MachineIRBuilder::setMF(MachineFunction &MF) {
   State.TII = MF.getSubtarget().getInstrInfo();
   State.DL = DebugLoc();
   State.PCSections = nullptr;
+  State.MMRA = nullptr;
   State.II = MachineBasicBlock::iterator();
   State.Observer = nullptr;
 }
@@ -37,7 +38,8 @@ void MachineIRBuilder::setMF(MachineFunction &MF) {
 //------------------------------------------------------------------------------
 
 MachineInstrBuilder MachineIRBuilder::buildInstrNoInsert(unsigned Opcode) {
-  return BuildMI(getMF(), {getDL(), getPCSections()}, getTII().get(Opcode));
+  return BuildMI(getMF(), {getDL(), getPCSections(), getMMRAMetadata()},
+                 getTII().get(Opcode));
 }
 
 MachineInstrBuilder MachineIRBuilder::insertInstr(MachineInstrBuilder MIB) {
diff --git a/llvm/lib/CodeGen/GlobalISel/RegBankSelect.cpp b/llvm/lib/CodeGen/GlobalISel/RegBankSelect.cpp
index bb5363fb2527..383cb61aed41 100644
--- a/llvm/lib/CodeGen/GlobalISel/RegBankSelect.cpp
+++ b/llvm/lib/CodeGen/GlobalISel/RegBankSelect.cpp
@@ -420,7 +420,8 @@ void RegBankSelect::tryAvoidingSplit(
       // If the next terminator uses Reg, this means we have
       // to split right after MI and thus we need a way to ask
       // which outgoing edges are affected.
-      assert(!Next->readsRegister(Reg) && "Need to split between terminators");
+      assert(!Next->readsRegister(Reg, /*TRI=*/nullptr) &&
+             "Need to split between terminators");
     // We will split all the edges and repair there.
   } else {
     // This is a virtual register defined by a terminator.
diff --git a/llvm/lib/CodeGen/GlobalISel/Utils.cpp b/llvm/lib/CodeGen/GlobalISel/Utils.cpp
index ae43e9ccf611..4e3781cb4e9d 100644
--- a/llvm/lib/CodeGen/GlobalISel/Utils.cpp
+++ b/llvm/lib/CodeGen/GlobalISel/Utils.cpp
@@ -12,6 +12,7 @@
 #include "llvm/CodeGen/GlobalISel/Utils.h"
 #include "llvm/ADT/APFloat.h"
 #include "llvm/ADT/APInt.h"
+#include "llvm/Analysis/ValueTracking.h"
 #include "llvm/CodeGen/CodeGenCommonISel.h"
 #include "llvm/CodeGen/GlobalISel/GISelChangeObserver.h"
 #include "llvm/CodeGen/GlobalISel/GISelKnownBits.h"
@@ -28,6 +29,7 @@
 #include "llvm/CodeGen/StackProtector.h"
 #include "llvm/CodeGen/TargetInstrInfo.h"
 #include "llvm/CodeGen/TargetLowering.h"
+#include "llvm/CodeGen/TargetOpcodes.h"
 #include "llvm/CodeGen/TargetPassConfig.h"
 #include "llvm/CodeGen/TargetRegisterInfo.h"
 #include "llvm/IR/Constants.h"
@@ -1709,3 +1711,84 @@ bool llvm::isPreISelGenericFloatingPointOpcode(unsigned Opc) {
     return false;
   }
 }
+
+namespace {
+enum class UndefPoisonKind {
+  PoisonOnly = (1 << 0),
+  UndefOnly = (1 << 1),
+  UndefOrPoison = PoisonOnly | UndefOnly,
+};
+}
+
+[[maybe_unused]] static bool includesPoison(UndefPoisonKind Kind) {
+  return (unsigned(Kind) & unsigned(UndefPoisonKind::PoisonOnly)) != 0;
+}
+
+[[maybe_unused]] static bool includesUndef(UndefPoisonKind Kind) {
+  return (unsigned(Kind) & unsigned(UndefPoisonKind::UndefOnly)) != 0;
+}
+
+static bool canCreateUndefOrPoison(Register Reg, const MachineRegisterInfo &MRI,
+                                   bool ConsiderFlagsAndMetadata,
+                                   UndefPoisonKind Kind) {
+  MachineInstr *RegDef = MRI.getVRegDef(Reg);
+
+  switch (RegDef->getOpcode()) {
+  case TargetOpcode::G_FREEZE:
+    return false;
+  default:
+    return true;
+  }
+}
+
+static bool isGuaranteedNotToBeUndefOrPoison(Register Reg,
+                                             const MachineRegisterInfo &MRI,
+                                             unsigned Depth,
+                                             UndefPoisonKind Kind) {
+  if (Depth >= MaxAnalysisRecursionDepth)
+    return false;
+
+  MachineInstr *RegDef = MRI.getVRegDef(Reg);
+
+  switch (RegDef->getOpcode()) {
+  case TargetOpcode::G_FREEZE:
+    return true;
+  case TargetOpcode::G_IMPLICIT_DEF:
+    return !includesUndef(Kind);
+  default:
+    return false;
+  }
+}
+
+bool llvm::canCreateUndefOrPoison(Register Reg, const MachineRegisterInfo &MRI,
+                                  bool ConsiderFlagsAndMetadata) {
+  return ::canCreateUndefOrPoison(Reg, MRI, ConsiderFlagsAndMetadata,
+                                  UndefPoisonKind::UndefOrPoison);
+}
+
+bool canCreatePoison(Register Reg, const MachineRegisterInfo &MRI,
+                     bool ConsiderFlagsAndMetadata = true) {
+  return ::canCreateUndefOrPoison(Reg, MRI, ConsiderFlagsAndMetadata,
+                                  UndefPoisonKind::PoisonOnly);
+}
+
+bool llvm::isGuaranteedNotToBeUndefOrPoison(Register Reg,
+                                            const MachineRegisterInfo &MRI,
+                                            unsigned Depth) {
+  return ::isGuaranteedNotToBeUndefOrPoison(Reg, MRI, Depth,
+                                            UndefPoisonKind::UndefOrPoison);
+}
+
+bool llvm::isGuaranteedNotToBePoison(Register Reg,
+                                     const MachineRegisterInfo &MRI,
+                                     unsigned Depth) {
+  return ::isGuaranteedNotToBeUndefOrPoison(Reg, MRI, Depth,
+                                            UndefPoisonKind::PoisonOnly);
+}
+
+bool llvm::isGuaranteedNotToBeUndef(Register Reg,
+                                    const MachineRegisterInfo &MRI,
+                                    unsigned Depth) {
+  return ::isGuaranteedNotToBeUndefOrPoison(Reg, MRI, Depth,
+                                            UndefPoisonKind::UndefOnly);
+}
diff --git a/llvm/lib/CodeGen/InlineSpiller.cpp b/llvm/lib/CodeGen/InlineSpiller.cpp
index c46b1fe18ca7..69c671220db3 100644
--- a/llvm/lib/CodeGen/InlineSpiller.cpp
+++ b/llvm/lib/CodeGen/InlineSpiller.cpp
@@ -869,7 +869,7 @@ static void dumpMachineInstrRangeWithSlotIndex(MachineBasicBlock::iterator B,
     // destination that is marked as an early clobber, print the
     // early-clobber slot index.
     if (VReg) {
-      MachineOperand *MO = I->findRegisterDefOperand(VReg);
+      MachineOperand *MO = I->findRegisterDefOperand(VReg, /*TRI=*/nullptr);
       if (MO && MO->isEarlyClobber())
         Idx = Idx.getRegSlot(true);
     }
diff --git a/llvm/lib/CodeGen/InterleavedAccessPass.cpp b/llvm/lib/CodeGen/InterleavedAccessPass.cpp
index 8989eabbe6df..8c9065aec7fa 100644
--- a/llvm/lib/CodeGen/InterleavedAccessPass.cpp
+++ b/llvm/lib/CodeGen/InterleavedAccessPass.cpp
@@ -535,9 +535,9 @@ bool InterleavedAccessImpl::runOnFunction(Function &F) {
     if (auto *II = dyn_cast<IntrinsicInst>(&I)) {
       // At present, we only have intrinsics to represent (de)interleaving
       // with a factor of 2.
-      if (II->getIntrinsicID() == Intrinsic::experimental_vector_deinterleave2)
+      if (II->getIntrinsicID() == Intrinsic::vector_deinterleave2)
         Changed |= lowerDeinterleaveIntrinsic(II, DeadInsts);
-      if (II->getIntrinsicID() == Intrinsic::experimental_vector_interleave2)
+      if (II->getIntrinsicID() == Intrinsic::vector_interleave2)
         Changed |= lowerInterleaveIntrinsic(II, DeadInsts);
     }
   }
diff --git a/llvm/lib/CodeGen/LiveVariables.cpp b/llvm/lib/CodeGen/LiveVariables.cpp
index b85526cfb380..f44db575a925 100644
--- a/llvm/lib/CodeGen/LiveVariables.cpp
+++ b/llvm/lib/CodeGen/LiveVariables.cpp
@@ -258,7 +258,7 @@ void LiveVariables::HandlePhysRegUse(Register Reg, MachineInstr &MI) {
       }
     }
   } else if (LastDef && !PhysRegUse[Reg] &&
-             !LastDef->findRegisterDefOperand(Reg))
+             !LastDef->findRegisterDefOperand(Reg, /*TRI=*/nullptr))
     // Last def defines the super register, add an implicit def of reg.
     LastDef->addOperand(MachineOperand::CreateReg(Reg, true/*IsDef*/,
                                                   true/*IsImp*/));
@@ -361,7 +361,8 @@ bool LiveVariables::HandlePhysRegKill(Register Reg, MachineInstr *MI) {
         continue;
       bool NeedDef = true;
       if (PhysRegDef[Reg] == PhysRegDef[SubReg]) {
-        MachineOperand *MO = PhysRegDef[Reg]->findRegisterDefOperand(SubReg);
+        MachineOperand *MO =
+            PhysRegDef[Reg]->findRegisterDefOperand(SubReg, /*TRI=*/nullptr);
         if (MO) {
           NeedDef = false;
           assert(!MO->isDead());
@@ -388,7 +389,7 @@ bool LiveVariables::HandlePhysRegKill(Register Reg, MachineInstr *MI) {
                                                 true/*IsImp*/, true/*IsKill*/));
     else {
       MachineOperand *MO =
-        LastRefOrPartRef->findRegisterDefOperand(Reg, false, false, TRI);
+          LastRefOrPartRef->findRegisterDefOperand(Reg, TRI, false, false);
       bool NeedEC = MO->isEarlyClobber() && MO->getReg() != Reg;
       // If the last reference is the last def, then it's not used at all.
       // That is, unless we are currently processing the last reference itself.
@@ -396,7 +397,7 @@ bool LiveVariables::HandlePhysRegKill(Register Reg, MachineInstr *MI) {
       if (NeedEC) {
         // If we are adding a subreg def and the superreg def is marked early
         // clobber, add an early clobber marker to the subreg def.
-        MO = LastRefOrPartRef->findRegisterDefOperand(Reg);
+        MO = LastRefOrPartRef->findRegisterDefOperand(Reg, /*TRI=*/nullptr);
         if (MO)
           MO->setIsEarlyClobber();
       }
@@ -727,7 +728,7 @@ void LiveVariables::recomputeForSingleDefVirtReg(Register Reg) {
       if (MI.isPHI())
         break;
       if (MI.readsVirtualRegister(Reg)) {
-        assert(!MI.killsRegister(Reg));
+        assert(!MI.killsRegister(Reg, /*TRI=*/nullptr));
         MI.addRegisterKilled(Reg, nullptr);
         VI.Kills.push_back(&MI);
         break;
diff --git a/llvm/lib/CodeGen/MIRPrinter.cpp b/llvm/lib/CodeGen/MIRPrinter.cpp
index bf3aee67ec00..6751fcf97087 100644
--- a/llvm/lib/CodeGen/MIRPrinter.cpp
+++ b/llvm/lib/CodeGen/MIRPrinter.cpp
@@ -856,6 +856,13 @@ void MIPrinter::print(const MachineInstr &MI) {
     PCSections->printAsOperand(OS, MST);
     NeedComma = true;
   }
+  if (MDNode *MMRA = MI.getMMRAMetadata()) {
+    if (NeedComma)
+      OS << ',';
+    OS << " mmra ";
+    MMRA->printAsOperand(OS, MST);
+    NeedComma = true;
+  }
   if (uint32_t CFIType = MI.getCFIType()) {
     if (NeedComma)
       OS << ',';
diff --git a/llvm/lib/CodeGen/MachineCSE.cpp b/llvm/lib/CodeGen/MachineCSE.cpp
index 26a8d00e6626..42cdcaa5bbf4 100644
--- a/llvm/lib/CodeGen/MachineCSE.cpp
+++ b/llvm/lib/CodeGen/MachineCSE.cpp
@@ -709,7 +709,7 @@ bool MachineCSE::ProcessBlockCSE(MachineBasicBlock *MBB) {
         for (MachineBasicBlock::iterator II = CSMI, IE = &MI; II != IE; ++II)
           for (auto ImplicitDef : ImplicitDefs)
             if (MachineOperand *MO = II->findRegisterUseOperand(
-                    ImplicitDef, /*isKill=*/true, TRI))
+                    ImplicitDef, TRI, /*isKill=*/true))
               MO->setIsKill(false);
       } else {
         // If the instructions aren't in the same BB, bail out and clear the
diff --git a/llvm/lib/CodeGen/MachineCombiner.cpp b/llvm/lib/CodeGen/MachineCombiner.cpp
index ac58162bbfb4..c11263163a34 100644
--- a/llvm/lib/CodeGen/MachineCombiner.cpp
+++ b/llvm/lib/CodeGen/MachineCombiner.cpp
@@ -229,8 +229,10 @@ MachineCombiner::getDepth(SmallVectorImpl<MachineInstr *> &InsInstrs,
         assert(DefInstr &&
                "There must be a definition for a new virtual register");
         DepthOp = InstrDepth[II->second];
-        int DefIdx = DefInstr->findRegisterDefOperandIdx(MO.getReg());
-        int UseIdx = InstrPtr->findRegisterUseOperandIdx(MO.getReg());
+        int DefIdx =
+            DefInstr->findRegisterDefOperandIdx(MO.getReg(), /*TRI=*/nullptr);
+        int UseIdx =
+            InstrPtr->findRegisterUseOperandIdx(MO.getReg(), /*TRI=*/nullptr);
         LatencyOp = TSchedModel.computeOperandLatency(DefInstr, DefIdx,
                                                       InstrPtr, UseIdx);
       } else {
@@ -241,8 +243,12 @@ MachineCombiner::getDepth(SmallVectorImpl<MachineInstr *> &InsInstrs,
           DepthOp = BlockTrace.getInstrCycles(*DefInstr).Depth;
           if (!isTransientMI(DefInstr))
             LatencyOp = TSchedModel.computeOperandLatency(
-                DefInstr, DefInstr->findRegisterDefOperandIdx(MO.getReg()),
-                InstrPtr, InstrPtr->findRegisterUseOperandIdx(MO.getReg()));
+                DefInstr,
+                DefInstr->findRegisterDefOperandIdx(MO.getReg(),
+                                                    /*TRI=*/nullptr),
+                InstrPtr,
+                InstrPtr->findRegisterUseOperandIdx(MO.getReg(),
+                                                    /*TRI=*/nullptr));
         }
       }
       IDepth = std::max(IDepth, DepthOp + LatencyOp);
@@ -280,8 +286,10 @@ unsigned MachineCombiner::getLatency(MachineInstr *Root, MachineInstr *NewRoot,
     unsigned LatencyOp = 0;
     if (UseMO && BlockTrace.isDepInTrace(*Root, *UseMO)) {
       LatencyOp = TSchedModel.computeOperandLatency(
-          NewRoot, NewRoot->findRegisterDefOperandIdx(MO.getReg()), UseMO,
-          UseMO->findRegisterUseOperandIdx(MO.getReg()));
+          NewRoot,
+          NewRoot->findRegisterDefOperandIdx(MO.getReg(), /*TRI=*/nullptr),
+          UseMO,
+          UseMO->findRegisterUseOperandIdx(MO.getReg(), /*TRI=*/nullptr));
     } else {
       LatencyOp = TSchedModel.computeInstrLatency(NewRoot);
     }
diff --git a/llvm/lib/CodeGen/MachineCopyPropagation.cpp b/llvm/lib/CodeGen/MachineCopyPropagation.cpp
index 8dc6781fcb01..c82f00316147 100644
--- a/llvm/lib/CodeGen/MachineCopyPropagation.cpp
+++ b/llvm/lib/CodeGen/MachineCopyPropagation.cpp
@@ -737,7 +737,7 @@ void MachineCopyPropagation::forwardUses(MachineInstr &MI) {
     // cannot cope with that.
     if (isCopyInstr(MI, *TII, UseCopyInstr) &&
         MI.modifiesRegister(CopySrcReg, TRI) &&
-        !MI.definesRegister(CopySrcReg)) {
+        !MI.definesRegister(CopySrcReg, /*TRI=*/nullptr)) {
       LLVM_DEBUG(dbgs() << "MCP: Copy source overlap with dest in " << MI);
       continue;
     }
diff --git a/llvm/lib/CodeGen/MachineFunction.cpp b/llvm/lib/CodeGen/MachineFunction.cpp
index ad5321499266..8366ad285906 100644
--- a/llvm/lib/CodeGen/MachineFunction.cpp
+++ b/llvm/lib/CodeGen/MachineFunction.cpp
@@ -573,10 +573,10 @@ MachineFunction::getMachineMemOperand(const MachineMemOperand *MMO,
 MachineInstr::ExtraInfo *MachineFunction::createMIExtraInfo(
     ArrayRef<MachineMemOperand *> MMOs, MCSymbol *PreInstrSymbol,
     MCSymbol *PostInstrSymbol, MDNode *HeapAllocMarker, MDNode *PCSections,
-    uint32_t CFIType) {
+    uint32_t CFIType, MDNode *MMRAs) {
   return MachineInstr::ExtraInfo::create(Allocator, MMOs, PreInstrSymbol,
                                          PostInstrSymbol, HeapAllocMarker,
-                                         PCSections, CFIType);
+                                         PCSections, CFIType, MMRAs);
 }
 
 const char *MachineFunction::createExternalSymbolName(StringRef Name) {
diff --git a/llvm/lib/CodeGen/MachineInstr.cpp b/llvm/lib/CodeGen/MachineInstr.cpp
index 83604003a038..02479f31f0b6 100644
--- a/llvm/lib/CodeGen/MachineInstr.cpp
+++ b/llvm/lib/CodeGen/MachineInstr.cpp
@@ -318,14 +318,15 @@ void MachineInstr::setExtraInfo(MachineFunction &MF,
                                 MCSymbol *PreInstrSymbol,
                                 MCSymbol *PostInstrSymbol,
                                 MDNode *HeapAllocMarker, MDNode *PCSections,
-                                uint32_t CFIType) {
+                                uint32_t CFIType, MDNode *MMRAs) {
   bool HasPreInstrSymbol = PreInstrSymbol != nullptr;
   bool HasPostInstrSymbol = PostInstrSymbol != nullptr;
   bool HasHeapAllocMarker = HeapAllocMarker != nullptr;
   bool HasPCSections = PCSections != nullptr;
   bool HasCFIType = CFIType != 0;
+  bool HasMMRAs = MMRAs != nullptr;
   int NumPointers = MMOs.size() + HasPreInstrSymbol + HasPostInstrSymbol +
-                    HasHeapAllocMarker + HasPCSections + HasCFIType;
+                    HasHeapAllocMarker + HasPCSections + HasCFIType + HasMMRAs;
 
   // Drop all extra info if there is none.
   if (NumPointers <= 0) {
@@ -337,11 +338,11 @@ void MachineInstr::setExtraInfo(MachineFunction &MF,
   // out of line because PointerSumType cannot hold more than 4 tag types with
   // 32-bit pointers.
   // FIXME: Maybe we should make the symbols in the extra info mutable?
-  else if (NumPointers > 1 || HasHeapAllocMarker || HasPCSections ||
+  else if (NumPointers > 1 || HasMMRAs || HasHeapAllocMarker || HasPCSections ||
            HasCFIType) {
     Info.set<EIIK_OutOfLine>(
         MF.createMIExtraInfo(MMOs, PreInstrSymbol, PostInstrSymbol,
-                             HeapAllocMarker, PCSections, CFIType));
+                             HeapAllocMarker, PCSections, CFIType, MMRAs));
     return;
   }
 
@@ -359,7 +360,8 @@ void MachineInstr::dropMemRefs(MachineFunction &MF) {
     return;
 
   setExtraInfo(MF, {}, getPreInstrSymbol(), getPostInstrSymbol(),
-               getHeapAllocMarker(), getPCSections(), getCFIType());
+               getHeapAllocMarker(), getPCSections(), getCFIType(),
+               getMMRAMetadata());
 }
 
 void MachineInstr::setMemRefs(MachineFunction &MF,
@@ -370,7 +372,8 @@ void MachineInstr::setMemRefs(MachineFunction &MF,
   }
 
   setExtraInfo(MF, MMOs, getPreInstrSymbol(), getPostInstrSymbol(),
-               getHeapAllocMarker(), getPCSections(), getCFIType());
+               getHeapAllocMarker(), getPCSections(), getCFIType(),
+               getMMRAMetadata());
 }
 
 void MachineInstr::addMemOperand(MachineFunction &MF,
@@ -394,7 +397,8 @@ void MachineInstr::cloneMemRefs(MachineFunction &MF, const MachineInstr &MI) {
   if (getPreInstrSymbol() == MI.getPreInstrSymbol() &&
       getPostInstrSymbol() == MI.getPostInstrSymbol() &&
       getHeapAllocMarker() == MI.getHeapAllocMarker() &&
-      getPCSections() == MI.getPCSections()) {
+      getPCSections() == MI.getPCSections() && getMMRAMetadata() &&
+      MI.getMMRAMetadata()) {
     Info = MI.Info;
     return;
   }
@@ -479,7 +483,8 @@ void MachineInstr::setPreInstrSymbol(MachineFunction &MF, MCSymbol *Symbol) {
   }
 
   setExtraInfo(MF, memoperands(), Symbol, getPostInstrSymbol(),
-               getHeapAllocMarker(), getPCSections(), getCFIType());
+               getHeapAllocMarker(), getPCSections(), getCFIType(),
+               getMMRAMetadata());
 }
 
 void MachineInstr::setPostInstrSymbol(MachineFunction &MF, MCSymbol *Symbol) {
@@ -494,7 +499,8 @@ void MachineInstr::setPostInstrSymbol(MachineFunction &MF, MCSymbol *Symbol) {
   }
 
   setExtraInfo(MF, memoperands(), getPreInstrSymbol(), Symbol,
-               getHeapAllocMarker(), getPCSections(), getCFIType());
+               getHeapAllocMarker(), getPCSections(), getCFIType(),
+               getMMRAMetadata());
 }
 
 void MachineInstr::setHeapAllocMarker(MachineFunction &MF, MDNode *Marker) {
@@ -503,7 +509,7 @@ void MachineInstr::setHeapAllocMarker(MachineFunction &MF, MDNode *Marker) {
     return;
 
   setExtraInfo(MF, memoperands(), getPreInstrSymbol(), getPostInstrSymbol(),
-               Marker, getPCSections(), getCFIType());
+               Marker, getPCSections(), getCFIType(), getMMRAMetadata());
 }
 
 void MachineInstr::setPCSections(MachineFunction &MF, MDNode *PCSections) {
@@ -512,7 +518,8 @@ void MachineInstr::setPCSections(MachineFunction &MF, MDNode *PCSections) {
     return;
 
   setExtraInfo(MF, memoperands(), getPreInstrSymbol(), getPostInstrSymbol(),
-               getHeapAllocMarker(), PCSections, getCFIType());
+               getHeapAllocMarker(), PCSections, getCFIType(),
+               getMMRAMetadata());
 }
 
 void MachineInstr::setCFIType(MachineFunction &MF, uint32_t Type) {
@@ -521,7 +528,16 @@ void MachineInstr::setCFIType(MachineFunction &MF, uint32_t Type) {
     return;
 
   setExtraInfo(MF, memoperands(), getPreInstrSymbol(), getPostInstrSymbol(),
-               getHeapAllocMarker(), getPCSections(), Type);
+               getHeapAllocMarker(), getPCSections(), Type, getMMRAMetadata());
+}
+
+void MachineInstr::setMMRAMetadata(MachineFunction &MF, MDNode *MMRAs) {
+  // Do nothing if old and new symbols are the same.
+  if (MMRAs == getMMRAMetadata())
+    return;
+
+  setExtraInfo(MF, memoperands(), getPreInstrSymbol(), getPostInstrSymbol(),
+               getHeapAllocMarker(), getPCSections(), getCFIType(), MMRAs);
 }
 
 void MachineInstr::cloneInstrSymbols(MachineFunction &MF,
@@ -537,6 +553,7 @@ void MachineInstr::cloneInstrSymbols(MachineFunction &MF,
   setPostInstrSymbol(MF, MI.getPostInstrSymbol());
   setHeapAllocMarker(MF, MI.getHeapAllocMarker());
   setPCSections(MF, MI.getPCSections());
+  setMMRAMetadata(MF, MI.getMMRAMetadata());
 }
 
 uint32_t MachineInstr::mergeFlagsWith(const MachineInstr &Other) const {
@@ -1028,8 +1045,9 @@ bool MachineInstr::hasRegisterImplicitUseOperand(Register Reg) const {
 /// findRegisterUseOperandIdx() - Returns the MachineOperand that is a use of
 /// the specific register or -1 if it is not found. It further tightens
 /// the search criteria to a use that kills the register if isKill is true.
-int MachineInstr::findRegisterUseOperandIdx(
-    Register Reg, bool isKill, const TargetRegisterInfo *TRI) const {
+int MachineInstr::findRegisterUseOperandIdx(Register Reg,
+                                            const TargetRegisterInfo *TRI,
+                                            bool isKill) const {
   for (unsigned i = 0, e = getNumOperands(); i != e; ++i) {
     const MachineOperand &MO = getOperand(i);
     if (!MO.isReg() || !MO.isUse())
@@ -1076,9 +1094,9 @@ MachineInstr::readsWritesVirtualRegister(Register Reg,
 /// the specified register or -1 if it is not found. If isDead is true, defs
 /// that are not dead are skipped. If TargetRegisterInfo is non-null, then it
 /// also checks if there is a def of a super-register.
-int
-MachineInstr::findRegisterDefOperandIdx(Register Reg, bool isDead, bool Overlap,
-                                        const TargetRegisterInfo *TRI) const {
+int MachineInstr::findRegisterDefOperandIdx(Register Reg,
+                                            const TargetRegisterInfo *TRI,
+                                            bool isDead, bool Overlap) const {
   bool isPhys = Reg.isPhysical();
   for (unsigned i = 0, e = getNumOperands(); i != e; ++i) {
     const MachineOperand &MO = getOperand(i);
@@ -1881,6 +1899,14 @@ void MachineInstr::print(raw_ostream &OS, ModuleSlotTracker &MST,
     OS << " pcsections ";
     PCSections->printAsOperand(OS, MST);
   }
+  if (MDNode *MMRA = getMMRAMetadata()) {
+    if (!FirstOp) {
+      FirstOp = false;
+      OS << ',';
+    }
+    OS << " mmra ";
+    MMRA->printAsOperand(OS, MST);
+  }
   if (uint32_t CFIType = getCFIType()) {
     if (!FirstOp)
       OS << ',';
@@ -2111,7 +2137,7 @@ void MachineInstr::setRegisterDefReadUndef(Register Reg, bool IsUndef) {
 void MachineInstr::addRegisterDefined(Register Reg,
                                       const TargetRegisterInfo *RegInfo) {
   if (Reg.isPhysical()) {
-    MachineOperand *MO = findRegisterDefOperand(Reg, false, false, RegInfo);
+    MachineOperand *MO = findRegisterDefOperand(Reg, RegInfo, false, false);
     if (MO)
       return;
   } else {
diff --git a/llvm/lib/CodeGen/MachineLateInstrsCleanup.cpp b/llvm/lib/CodeGen/MachineLateInstrsCleanup.cpp
index aa1eb7c35425..1f596cd1bd2e 100644
--- a/llvm/lib/CodeGen/MachineLateInstrsCleanup.cpp
+++ b/llvm/lib/CodeGen/MachineLateInstrsCleanup.cpp
@@ -230,7 +230,7 @@ bool MachineLateInstrsCleanup::processBlock(MachineBasicBlock *MBB) {
       if (MI.modifiesRegister(Reg, TRI)) {
         MBBDefs.erase(Reg);
         MBBKills.erase(Reg);
-      } else if (MI.findRegisterUseOperandIdx(Reg, true /*isKill*/, TRI) != -1)
+      } else if (MI.findRegisterUseOperandIdx(Reg, TRI, true /*isKill*/) != -1)
         // Keep track of register kills.
         MBBKills[Reg] = &MI;
     }
diff --git a/llvm/lib/CodeGen/MachineSink.cpp b/llvm/lib/CodeGen/MachineSink.cpp
index c3a1d3759882..3d40130b92c4 100644
--- a/llvm/lib/CodeGen/MachineSink.cpp
+++ b/llvm/lib/CodeGen/MachineSink.cpp
@@ -309,7 +309,7 @@ static bool blockPrologueInterferes(const MachineBasicBlock *BB,
         if (PI->readsRegister(Reg, TRI))
           return true;
         // Check for interference with non-dead defs
-        auto *DefOp = PI->findRegisterDefOperand(Reg, false, true, TRI);
+        auto *DefOp = PI->findRegisterDefOperand(Reg, TRI, false, true);
         if (DefOp && !DefOp->isDead())
           return true;
       }
diff --git a/llvm/lib/CodeGen/ModuloSchedule.cpp b/llvm/lib/CodeGen/ModuloSchedule.cpp
index bdae94c4e6f8..b912112b1636 100644
--- a/llvm/lib/CodeGen/ModuloSchedule.cpp
+++ b/llvm/lib/CodeGen/ModuloSchedule.cpp
@@ -814,7 +814,7 @@ void ModuloScheduleExpander::splitLifetimes(MachineBasicBlock *KernelBB,
         unsigned SplitReg = 0;
         for (auto &BBJ : make_range(MachineBasicBlock::instr_iterator(MI),
                                     KernelBB->instr_end()))
-          if (BBJ.readsRegister(Def)) {
+          if (BBJ.readsRegister(Def, /*TRI=*/nullptr)) {
             // We split the lifetime when we find the first use.
             if (SplitReg == 0) {
               SplitReg = MRI.createVirtualRegister(MRI.getRegClass(Def));
@@ -829,7 +829,7 @@ void ModuloScheduleExpander::splitLifetimes(MachineBasicBlock *KernelBB,
         // Search through each of the epilog blocks for any uses to be renamed.
         for (auto &Epilog : EpilogBBs)
           for (auto &I : *Epilog)
-            if (I.readsRegister(Def))
+            if (I.readsRegister(Def, /*TRI=*/nullptr))
               I.substituteRegister(Def, SplitReg, 0, *TRI);
         break;
       }
@@ -1673,7 +1673,8 @@ void PeelingModuloScheduleExpander::moveStageBetweenBlocks(
     // we don't need the phi anymore.
     if (getStage(Def) == Stage) {
       Register PhiReg = MI.getOperand(0).getReg();
-      assert(Def->findRegisterDefOperandIdx(MI.getOperand(1).getReg()) != -1);
+      assert(Def->findRegisterDefOperandIdx(MI.getOperand(1).getReg(),
+                                            /*TRI=*/nullptr) != -1);
       MRI.replaceRegWith(MI.getOperand(0).getReg(), MI.getOperand(1).getReg());
       MI.getOperand(0).setReg(PhiReg);
       PhiToDelete.push_back(&MI);
@@ -1899,7 +1900,7 @@ Register
 PeelingModuloScheduleExpander::getEquivalentRegisterIn(Register Reg,
                                                        MachineBasicBlock *BB) {
   MachineInstr *MI = MRI.getUniqueVRegDef(Reg);
-  unsigned OpIdx = MI->findRegisterDefOperandIdx(Reg);
+  unsigned OpIdx = MI->findRegisterDefOperandIdx(Reg, /*TRI=*/nullptr);
   return BlockMIs[{BB, CanonicalMIs[MI]}]->getOperand(OpIdx).getReg();
 }
 
diff --git a/llvm/lib/CodeGen/PHIElimination.cpp b/llvm/lib/CodeGen/PHIElimination.cpp
index 18f8c001bd78..3254ec0b77fe 100644
--- a/llvm/lib/CodeGen/PHIElimination.cpp
+++ b/llvm/lib/CodeGen/PHIElimination.cpp
@@ -549,7 +549,7 @@ void PHIElimination::LowerPHINode(MachineBasicBlock &MBB,
       MachineBasicBlock::iterator KillInst = opBlock.end();
       for (MachineBasicBlock::iterator Term = InsertPos; Term != opBlock.end();
            ++Term) {
-        if (Term->readsRegister(SrcReg))
+        if (Term->readsRegister(SrcReg, /*TRI=*/nullptr))
           KillInst = Term;
       }
 
@@ -563,7 +563,7 @@ void PHIElimination::LowerPHINode(MachineBasicBlock &MBB,
             --KillInst;
             if (KillInst->isDebugInstr())
               continue;
-            if (KillInst->readsRegister(SrcReg))
+            if (KillInst->readsRegister(SrcReg, /*TRI=*/nullptr))
               break;
           }
         } else {
@@ -571,7 +571,8 @@ void PHIElimination::LowerPHINode(MachineBasicBlock &MBB,
           KillInst = NewSrcInstr;
         }
       }
-      assert(KillInst->readsRegister(SrcReg) && "Cannot find kill instruction");
+      assert(KillInst->readsRegister(SrcReg, /*TRI=*/nullptr) &&
+             "Cannot find kill instruction");
 
       // Finally, mark it killed.
       LV->addVirtualRegisterKilled(SrcReg, *KillInst);
@@ -607,7 +608,7 @@ void PHIElimination::LowerPHINode(MachineBasicBlock &MBB,
           MachineBasicBlock::iterator KillInst = opBlock.end();
           for (MachineBasicBlock::iterator Term = InsertPos;
                Term != opBlock.end(); ++Term) {
-            if (Term->readsRegister(SrcReg))
+            if (Term->readsRegister(SrcReg, /*TRI=*/nullptr))
               KillInst = Term;
           }
 
@@ -621,7 +622,7 @@ void PHIElimination::LowerPHINode(MachineBasicBlock &MBB,
                 --KillInst;
                 if (KillInst->isDebugInstr())
                   continue;
-                if (KillInst->readsRegister(SrcReg))
+                if (KillInst->readsRegister(SrcReg, /*TRI=*/nullptr))
                   break;
               }
             } else {
@@ -629,7 +630,7 @@ void PHIElimination::LowerPHINode(MachineBasicBlock &MBB,
               KillInst = std::prev(InsertPos);
             }
           }
-          assert(KillInst->readsRegister(SrcReg) &&
+          assert(KillInst->readsRegister(SrcReg, /*TRI=*/nullptr) &&
                  "Cannot find kill instruction");
 
           SlotIndex LastUseIndex = LIS->getInstructionIndex(*KillInst);
diff --git a/llvm/lib/CodeGen/PeepholeOptimizer.cpp b/llvm/lib/CodeGen/PeepholeOptimizer.cpp
index 1b1f22e827cb..477a86dbe3f8 100644
--- a/llvm/lib/CodeGen/PeepholeOptimizer.cpp
+++ b/llvm/lib/CodeGen/PeepholeOptimizer.cpp
@@ -1577,7 +1577,7 @@ bool PeepholeOptimizer::findTargetRecurrence(
     return false;
 
   MachineInstr &MI = *(MRI->use_instr_nodbg_begin(Reg));
-  unsigned Idx = MI.findRegisterUseOperandIdx(Reg);
+  unsigned Idx = MI.findRegisterUseOperandIdx(Reg, /*TRI=*/nullptr);
 
   // Only interested in recurrences whose instructions have only one def, which
   // is a virtual register.
diff --git a/llvm/lib/CodeGen/RegisterCoalescer.cpp b/llvm/lib/CodeGen/RegisterCoalescer.cpp
index 7e9c992031f8..3397bd0a6060 100644
--- a/llvm/lib/CodeGen/RegisterCoalescer.cpp
+++ b/llvm/lib/CodeGen/RegisterCoalescer.cpp
@@ -723,7 +723,8 @@ bool RegisterCoalescer::adjustCopiesBackFrom(const CoalescerPair &CP,
 
   // If the source instruction was killing the source register before the
   // merge, unset the isKill marker given the live range has been extended.
-  int UIdx = ValSEndInst->findRegisterUseOperandIdx(IntB.reg(), true);
+  int UIdx =
+      ValSEndInst->findRegisterUseOperandIdx(IntB.reg(), /*TRI=*/nullptr, true);
   if (UIdx != -1) {
     ValSEndInst->getOperand(UIdx).setIsKill(false);
   }
@@ -848,7 +849,7 @@ RegisterCoalescer::removeCopyByCommutingDef(const CoalescerPair &CP,
     return { false, false };
   // If DefMI is a two-address instruction then commuting it will change the
   // destination register.
-  int DefIdx = DefMI->findRegisterDefOperandIdx(IntA.reg());
+  int DefIdx = DefMI->findRegisterDefOperandIdx(IntA.reg(), /*TRI=*/nullptr);
   assert(DefIdx != -1);
   unsigned UseOpIdx;
   if (!DefMI->isRegTiedToUseOperand(DefIdx, &UseOpIdx))
diff --git a/llvm/lib/CodeGen/RenameIndependentSubregs.cpp b/llvm/lib/CodeGen/RenameIndependentSubregs.cpp
index bc3ef1c0329a..e888f290df51 100644
--- a/llvm/lib/CodeGen/RenameIndependentSubregs.cpp
+++ b/llvm/lib/CodeGen/RenameIndependentSubregs.cpp
@@ -334,10 +334,17 @@ void RenameIndependentSubregs::computeMainRangesFixFlags(
                                                DebugLoc(), MCDesc, Reg);
           SlotIndex DefIdx = LIS->InsertMachineInstrInMaps(*ImpDef);
           SlotIndex RegDefIdx = DefIdx.getRegSlot();
+          LaneBitmask Mask = MRI->getMaxLaneMaskForVReg(Reg);
           for (LiveInterval::SubRange &SR : LI.subranges()) {
+            Mask = Mask & ~SR.LaneMask;
             VNInfo *SRVNI = SR.getNextValue(RegDefIdx, Allocator);
             SR.addSegment(LiveRange::Segment(RegDefIdx, PredEnd, SRVNI));
           }
+
+          if (!Mask.none()) {
+            LiveInterval::SubRange *SR = LI.createSubRange(Allocator, Mask);
+            SR->createDeadDef(RegDefIdx, Allocator);
+          }
         }
       }
     }
diff --git a/llvm/lib/CodeGen/SelectionDAG/DAGCombiner.cpp b/llvm/lib/CodeGen/SelectionDAG/DAGCombiner.cpp
index e6e0a1fc7d82..4b81185c6e31 100644
--- a/llvm/lib/CodeGen/SelectionDAG/DAGCombiner.cpp
+++ b/llvm/lib/CodeGen/SelectionDAG/DAGCombiner.cpp
@@ -530,6 +530,7 @@ namespace {
     bool refineExtractVectorEltIntoMultipleNarrowExtractVectorElts(SDNode *N);
 
     SDValue visitSTORE(SDNode *N);
+    SDValue visitATOMIC_STORE(SDNode *N);
     SDValue visitLIFETIME_END(SDNode *N);
     SDValue visitINSERT_VECTOR_ELT(SDNode *N);
     SDValue visitEXTRACT_VECTOR_ELT(SDNode *N);
@@ -1909,6 +1910,7 @@ SDValue DAGCombiner::visit(SDNode *N) {
   case ISD::BR_CC:              return visitBR_CC(N);
   case ISD::LOAD:               return visitLOAD(N);
   case ISD::STORE:              return visitSTORE(N);
+  case ISD::ATOMIC_STORE:       return visitATOMIC_STORE(N);
   case ISD::INSERT_VECTOR_ELT:  return visitINSERT_VECTOR_ELT(N);
   case ISD::EXTRACT_VECTOR_ELT: return visitEXTRACT_VECTOR_ELT(N);
   case ISD::BUILD_VECTOR:       return visitBUILD_VECTOR(N);
@@ -7620,6 +7622,8 @@ SDValue DAGCombiner::visitORLike(SDValue N0, SDValue N1, const SDLoc &DL) {
 static SDValue visitORCommutative(SelectionDAG &DAG, SDValue N0, SDValue N1,
                                   SDNode *N) {
   EVT VT = N0.getValueType();
+  unsigned BW = VT.getScalarSizeInBits();
+  SDLoc DL(N);
 
   auto peekThroughResize = [](SDValue V) {
     if (V->getOpcode() == ISD::ZERO_EXTEND || V->getOpcode() == ISD::TRUNCATE)
@@ -7642,16 +7646,16 @@ static SDValue visitORCommutative(SelectionDAG &DAG, SDValue N0, SDValue N1,
     if (SDValue NotOperand = getBitwiseNotOperand(N01, N00,
                                                   /* AllowUndefs */ false)) {
       if (peekThroughResize(NotOperand) == N1Resized)
-        return DAG.getNode(ISD::OR, SDLoc(N), VT,
-                           DAG.getZExtOrTrunc(N00, SDLoc(N), VT), N1);
+        return DAG.getNode(ISD::OR, DL, VT, DAG.getZExtOrTrunc(N00, DL, VT),
+                           N1);
     }
 
     // fold (or (and (xor Y, -1), X), Y) -> (or X, Y)
     if (SDValue NotOperand = getBitwiseNotOperand(N00, N01,
                                                   /* AllowUndefs */ false)) {
       if (peekThroughResize(NotOperand) == N1Resized)
-        return DAG.getNode(ISD::OR, SDLoc(N), VT,
-                           DAG.getZExtOrTrunc(N01, SDLoc(N), VT), N1);
+        return DAG.getNode(ISD::OR, DL, VT, DAG.getZExtOrTrunc(N01, DL, VT),
+                           N1);
     }
   }
 
@@ -7659,13 +7663,13 @@ static SDValue visitORCommutative(SelectionDAG &DAG, SDValue N0, SDValue N1,
 
   // fold or (xor X, N1), N1 --> or X, N1
   if (sd_match(N0, m_Xor(m_Value(X), m_Specific(N1))))
-    return DAG.getNode(ISD::OR, SDLoc(N), VT, X, N1);
+    return DAG.getNode(ISD::OR, DL, VT, X, N1);
 
   // fold or (xor x, y), (x and/or y) --> or x, y
   if (sd_match(N0, m_Xor(m_Value(X), m_Value(Y))) &&
       (sd_match(N1, m_And(m_Specific(X), m_Specific(Y))) ||
        sd_match(N1, m_Or(m_Specific(X), m_Specific(Y)))))
-    return DAG.getNode(ISD::OR, SDLoc(N), VT, X, Y);
+    return DAG.getNode(ISD::OR, DL, VT, X, Y);
 
   if (SDValue R = foldLogicOfShifts(N, N0, N1, DAG))
     return R;
@@ -7688,6 +7692,26 @@ static SDValue visitORCommutative(SelectionDAG &DAG, SDValue N0, SDValue N1,
       peekThroughZext(N0.getOperand(2)) == peekThroughZext(N1.getOperand(1)))
     return N0;
 
+  // Attempt to match a legalized build_pair-esque pattern:
+  // or(shl(aext(Hi),BW/2),zext(Lo))
+  SDValue Lo, Hi;
+  if (sd_match(N0,
+               m_OneUse(m_Shl(m_AnyExt(m_Value(Hi)), m_SpecificInt(BW / 2)))) &&
+      sd_match(N1, m_ZExt(m_Value(Lo))) &&
+      Lo.getScalarValueSizeInBits() == (BW / 2) &&
+      Lo.getValueType() == Hi.getValueType()) {
+    // Fold build_pair(not(Lo),not(Hi)) -> not(build_pair(Lo,Hi)).
+    SDValue NotLo, NotHi;
+    if (sd_match(Lo, m_OneUse(m_Not(m_Value(NotLo)))) &&
+        sd_match(Hi, m_OneUse(m_Not(m_Value(NotHi))))) {
+      Lo = DAG.getNode(ISD::ZERO_EXTEND, DL, VT, NotLo);
+      Hi = DAG.getNode(ISD::ANY_EXTEND, DL, VT, NotHi);
+      Hi = DAG.getNode(ISD::SHL, DL, VT, Hi,
+                       DAG.getShiftAmountConstant(BW / 2, VT, DL));
+      return DAG.getNOT(DL, DAG.getNode(ISD::OR, DL, VT, Lo, Hi), VT);
+    }
+  }
+
   return SDValue();
 }
 
@@ -9503,8 +9527,15 @@ static SDValue combineShiftOfShiftedLogic(SDNode *Shift, SelectionDAG &DAG) {
     if (ShiftAmtVal->getBitWidth() != C1Val.getBitWidth())
       return false;
 
+    // The fold is not valid if the sum of the shift values doesn't fit in the
+    // given shift amount type.
+    bool Overflow = false;
+    APInt NewShiftAmt = C1Val.uadd_ov(*ShiftAmtVal, Overflow);
+    if (Overflow)
+      return false;
+
     // The fold is not valid if the sum of the shift values exceeds bitwidth.
-    if ((*ShiftAmtVal + C1Val).uge(V.getScalarValueSizeInBits()))
+    if (NewShiftAmt.uge(V.getScalarValueSizeInBits()))
       return false;
 
     return true;
@@ -15428,6 +15459,12 @@ SDValue DAGCombiner::visitFREEZE(SDNode *N) {
   if (DAG.isGuaranteedNotToBeUndefOrPoison(N0, /*PoisonOnly*/ false))
     return N0;
 
+  // We currently avoid folding freeze over SRA/SRL, due to the problems seen
+  // with (freeze (assert ext)) blocking simplifications of SRA/SRL. See for
+  // example https://reviews.llvm.org/D136529#4120959.
+  if (N0.getOpcode() == ISD::SRA || N0.getOpcode() == ISD::SRL)
+    return SDValue();
+
   // Fold freeze(op(x, ...)) -> op(freeze(x), ...).
   // Try to push freeze through instructions that propagate but don't produce
   // poison as far as possible. If an operand of freeze follows three
@@ -15444,6 +15481,26 @@ SDValue DAGCombiner::visitFREEZE(SDNode *N) {
                                           N0.getOpcode() == ISD::BUILD_PAIR ||
                                           N0.getOpcode() == ISD::CONCAT_VECTORS;
 
+  // Avoid turning a BUILD_VECTOR that can be recognized as "all zeros", "all
+  // ones" or "constant" into something that depends on FrozenUndef. We can
+  // instead pick undef values to keep those properties, while at the same time
+  // folding away the freeze.
+  // If we implement a more general solution for folding away freeze(undef) in
+  // the future, then this special handling can be removed.
+  if (N0.getOpcode() == ISD::BUILD_VECTOR) {
+    SDLoc DL(N0);
+    EVT VT = N0.getValueType();
+    if (llvm::ISD::isBuildVectorAllOnes(N0.getNode()))
+      return DAG.getAllOnesConstant(DL, VT);
+    if (llvm::ISD::isBuildVectorOfConstantSDNodes(N0.getNode())) {
+      SmallVector<SDValue, 8> NewVecC;
+      for (const SDValue &Op : N0->op_values())
+        NewVecC.push_back(
+            Op.isUndef() ? DAG.getConstant(0, DL, Op.getValueType()) : Op);
+      return DAG.getBuildVector(VT, DL, NewVecC);
+    }
+  }
+
   SmallSetVector<SDValue, 8> MaybePoisonOperands;
   for (SDValue Op : N0->ops()) {
     if (DAG.isGuaranteedNotToBeUndefOrPoison(Op, /*PoisonOnly*/ false,
@@ -21088,6 +21145,24 @@ SDValue DAGCombiner::replaceStoreOfInsertLoad(StoreSDNode *ST) {
                       ST->getMemOperand()->getFlags());
 }
 
+SDValue DAGCombiner::visitATOMIC_STORE(SDNode *N) {
+  AtomicSDNode *ST = cast<AtomicSDNode>(N);
+  SDValue Val = ST->getVal();
+  EVT VT = Val.getValueType();
+  EVT MemVT = ST->getMemoryVT();
+
+  if (MemVT.bitsLT(VT)) { // Is truncating store
+    APInt TruncDemandedBits = APInt::getLowBitsSet(VT.getScalarSizeInBits(),
+                                                   MemVT.getScalarSizeInBits());
+    // See if we can simplify the operation with SimplifyDemandedBits, which
+    // only works if the value has a single use.
+    if (SimplifyDemandedBits(Val, TruncDemandedBits))
+      return SDValue(N, 0);
+  }
+
+  return SDValue();
+}
+
 SDValue DAGCombiner::visitSTORE(SDNode *N) {
   StoreSDNode *ST  = cast<StoreSDNode>(N);
   SDValue Chain = ST->getChain();
diff --git a/llvm/lib/CodeGen/SelectionDAG/FastISel.cpp b/llvm/lib/CodeGen/SelectionDAG/FastISel.cpp
index aefedd060f89..ef9f78335519 100644
--- a/llvm/lib/CodeGen/SelectionDAG/FastISel.cpp
+++ b/llvm/lib/CodeGen/SelectionDAG/FastISel.cpp
@@ -1424,7 +1424,7 @@ bool FastISel::selectIntrinsicCall(const IntrinsicInst *II) {
     // happened (such as an optimised function being always-inlined into an
     // optnone function). We will not be using the extra information in the
     // dbg.assign in that case, just use its dbg.value fields.
-    LLVM_FALLTHROUGH;
+    [[fallthrough]];
   case Intrinsic::dbg_value: {
     // This form of DBG_VALUE is target-independent.
     const DbgValueInst *DI = cast<DbgValueInst>(II);
diff --git a/llvm/lib/CodeGen/SelectionDAG/InstrEmitter.cpp b/llvm/lib/CodeGen/SelectionDAG/InstrEmitter.cpp
index 54409cbf91f1..759368a67a16 100644
--- a/llvm/lib/CodeGen/SelectionDAG/InstrEmitter.cpp
+++ b/llvm/lib/CodeGen/SelectionDAG/InstrEmitter.cpp
@@ -1420,7 +1420,7 @@ EmitSpecialNode(SDNode *Node, bool IsClone, bool IsCloned,
     for (unsigned Reg : ECRegs) {
       if (MIB->readsRegister(Reg, TRI)) {
         MachineOperand *MO =
-            MIB->findRegisterDefOperand(Reg, false, false, TRI);
+            MIB->findRegisterDefOperand(Reg, TRI, false, false);
         assert(MO && "No def operand for clobbered register?");
         MO->setIsEarlyClobber(false);
       }
diff --git a/llvm/lib/CodeGen/SelectionDAG/LegalizeDAG.cpp b/llvm/lib/CodeGen/SelectionDAG/LegalizeDAG.cpp
index 24f69ea1b742..46e54b5366d6 100644
--- a/llvm/lib/CodeGen/SelectionDAG/LegalizeDAG.cpp
+++ b/llvm/lib/CodeGen/SelectionDAG/LegalizeDAG.cpp
@@ -3195,7 +3195,7 @@ bool SelectionDAGLegalize::ExpandNode(SDNode *Node) {
       break;
     }
 
-    LLVM_FALLTHROUGH;
+    [[fallthrough]];
   }
   case ISD::BITCAST:
     if ((Tmp1 = EmitStackConvert(Node->getOperand(0), Node->getValueType(0),
@@ -3556,6 +3556,12 @@ bool SelectionDAGLegalize::ExpandNode(SDNode *Node) {
       Results.push_back(Expanded);
     break;
   }
+  case ISD::FMINIMUM:
+  case ISD::FMAXIMUM: {
+    if (SDValue Expanded = TLI.expandFMINIMUM_FMAXIMUM(Node, DAG))
+      Results.push_back(Expanded);
+    break;
+  }
   case ISD::FSIN:
   case ISD::FCOS: {
     EVT VT = Node->getValueType(0);
@@ -5575,6 +5581,21 @@ void SelectionDAGLegalize::PromoteNode(SDNode *Node) {
     Results.push_back(NewAtomic.getValue(1));
     break;
   }
+  case ISD::ATOMIC_LOAD: {
+    AtomicSDNode *AM = cast<AtomicSDNode>(Node);
+    SDLoc SL(Node);
+    assert(NVT.getSizeInBits() == OVT.getSizeInBits() &&
+           "unexpected promotion type");
+    assert(AM->getMemoryVT().getSizeInBits() == NVT.getSizeInBits() &&
+           "unexpected atomic_load with illegal type");
+
+    SDValue NewAtomic =
+        DAG.getAtomic(ISD::ATOMIC_LOAD, SL, NVT, DAG.getVTList(NVT, MVT::Other),
+                      {AM->getChain(), AM->getBasePtr()}, AM->getMemOperand());
+    Results.push_back(DAG.getNode(ISD::BITCAST, SL, OVT, NewAtomic));
+    Results.push_back(NewAtomic.getValue(1));
+    break;
+  }
   case ISD::SPLAT_VECTOR: {
     SDValue Scalar = Node->getOperand(0);
     MVT ScalarType = Scalar.getSimpleValueType();
diff --git a/llvm/lib/CodeGen/SelectionDAG/LegalizeFloatTypes.cpp b/llvm/lib/CodeGen/SelectionDAG/LegalizeFloatTypes.cpp
index 7685bc73cf96..abe5be763825 100644
--- a/llvm/lib/CodeGen/SelectionDAG/LegalizeFloatTypes.cpp
+++ b/llvm/lib/CodeGen/SelectionDAG/LegalizeFloatTypes.cpp
@@ -2449,6 +2449,9 @@ void DAGTypeLegalizer::PromoteFloatResult(SDNode *N, unsigned ResNo) {
       R = PromoteFloatRes_STRICT_FP_ROUND(N);
       break;
     case ISD::LOAD:       R = PromoteFloatRes_LOAD(N); break;
+    case ISD::ATOMIC_LOAD:
+      R = PromoteFloatRes_ATOMIC_LOAD(N);
+      break;
     case ISD::SELECT:     R = PromoteFloatRes_SELECT(N); break;
     case ISD::SELECT_CC:  R = PromoteFloatRes_SELECT_CC(N); break;
 
@@ -2695,6 +2698,25 @@ SDValue DAGTypeLegalizer::PromoteFloatRes_LOAD(SDNode *N) {
   return DAG.getNode(GetPromotionOpcode(VT, NVT), SDLoc(N), NVT, newL);
 }
 
+SDValue DAGTypeLegalizer::PromoteFloatRes_ATOMIC_LOAD(SDNode *N) {
+  AtomicSDNode *AM = cast<AtomicSDNode>(N);
+  EVT VT = AM->getValueType(0);
+
+  // Load the value as an integer value with the same number of bits.
+  EVT IVT = EVT::getIntegerVT(*DAG.getContext(), VT.getSizeInBits());
+  SDValue newL = DAG.getAtomic(
+      ISD::ATOMIC_LOAD, SDLoc(N), IVT, DAG.getVTList(IVT, MVT::Other),
+      {AM->getChain(), AM->getBasePtr()}, AM->getMemOperand());
+
+  // Legalize the chain result by replacing uses of the old value chain with the
+  // new one
+  ReplaceValueWith(SDValue(N, 1), newL.getValue(1));
+
+  // Convert the integer value to the desired FP type
+  EVT NVT = TLI.getTypeToTransformTo(*DAG.getContext(), VT);
+  return DAG.getNode(GetPromotionOpcode(VT, IVT), SDLoc(N), NVT, newL);
+}
+
 // Construct a new SELECT node with the promoted true- and false- values.
 SDValue DAGTypeLegalizer::PromoteFloatRes_SELECT(SDNode *N) {
   SDValue TrueVal = GetPromotedFloat(N->getOperand(1));
@@ -2855,6 +2877,9 @@ void DAGTypeLegalizer::SoftPromoteHalfResult(SDNode *N, unsigned ResNo) {
   case ISD::FFREXP:      R = SoftPromoteHalfRes_FFREXP(N); break;
 
   case ISD::LOAD:        R = SoftPromoteHalfRes_LOAD(N); break;
+  case ISD::ATOMIC_LOAD:
+    R = SoftPromoteHalfRes_ATOMIC_LOAD(N);
+    break;
   case ISD::SELECT:      R = SoftPromoteHalfRes_SELECT(N); break;
   case ISD::SELECT_CC:   R = SoftPromoteHalfRes_SELECT_CC(N); break;
   case ISD::SINT_TO_FP:
@@ -3039,6 +3064,20 @@ SDValue DAGTypeLegalizer::SoftPromoteHalfRes_LOAD(SDNode *N) {
   return NewL;
 }
 
+SDValue DAGTypeLegalizer::SoftPromoteHalfRes_ATOMIC_LOAD(SDNode *N) {
+  AtomicSDNode *AM = cast<AtomicSDNode>(N);
+
+  // Load the value as an integer value with the same number of bits.
+  SDValue NewL = DAG.getAtomic(
+      ISD::ATOMIC_LOAD, SDLoc(N), MVT::i16, DAG.getVTList(MVT::i16, MVT::Other),
+      {AM->getChain(), AM->getBasePtr()}, AM->getMemOperand());
+
+  // Legalize the chain result by replacing uses of the old value chain with the
+  // new one
+  ReplaceValueWith(SDValue(N, 1), NewL.getValue(1));
+  return NewL;
+}
+
 SDValue DAGTypeLegalizer::SoftPromoteHalfRes_SELECT(SDNode *N) {
   SDValue Op1 = GetSoftPromotedHalf(N->getOperand(1));
   SDValue Op2 = GetSoftPromotedHalf(N->getOperand(2));
diff --git a/llvm/lib/CodeGen/SelectionDAG/LegalizeTypes.h b/llvm/lib/CodeGen/SelectionDAG/LegalizeTypes.h
index 0483f7c74f91..4a2c7b355eb5 100644
--- a/llvm/lib/CodeGen/SelectionDAG/LegalizeTypes.h
+++ b/llvm/lib/CodeGen/SelectionDAG/LegalizeTypes.h
@@ -691,6 +691,7 @@ private:
   SDValue PromoteFloatRes_FP_ROUND(SDNode *N);
   SDValue PromoteFloatRes_STRICT_FP_ROUND(SDNode *N);
   SDValue PromoteFloatRes_LOAD(SDNode *N);
+  SDValue PromoteFloatRes_ATOMIC_LOAD(SDNode *N);
   SDValue PromoteFloatRes_SELECT(SDNode *N);
   SDValue PromoteFloatRes_SELECT_CC(SDNode *N);
   SDValue PromoteFloatRes_UnaryOp(SDNode *N);
@@ -734,6 +735,7 @@ private:
   SDValue SoftPromoteHalfRes_FFREXP(SDNode *N);
   SDValue SoftPromoteHalfRes_FP_ROUND(SDNode *N);
   SDValue SoftPromoteHalfRes_LOAD(SDNode *N);
+  SDValue SoftPromoteHalfRes_ATOMIC_LOAD(SDNode *N);
   SDValue SoftPromoteHalfRes_SELECT(SDNode *N);
   SDValue SoftPromoteHalfRes_SELECT_CC(SDNode *N);
   SDValue SoftPromoteHalfRes_UnaryOp(SDNode *N);
@@ -984,7 +986,7 @@ private:
   SDValue WidenVecRes_FP_TO_XINT_SAT(SDNode *N);
   SDValue WidenVecRes_XRINT(SDNode *N);
   SDValue WidenVecRes_FCOPYSIGN(SDNode *N);
-  SDValue WidenVecRes_IS_FPCLASS(SDNode *N);
+  SDValue WidenVecRes_UnarySameEltsWithScalarArg(SDNode *N);
   SDValue WidenVecRes_ExpOp(SDNode *N);
   SDValue WidenVecRes_Unary(SDNode *N);
   SDValue WidenVecRes_InregOp(SDNode *N);
diff --git a/llvm/lib/CodeGen/SelectionDAG/LegalizeVectorOps.cpp b/llvm/lib/CodeGen/SelectionDAG/LegalizeVectorOps.cpp
index 7a9cfdf5c3fd..8f87ee8e0939 100644
--- a/llvm/lib/CodeGen/SelectionDAG/LegalizeVectorOps.cpp
+++ b/llvm/lib/CodeGen/SelectionDAG/LegalizeVectorOps.cpp
@@ -420,6 +420,7 @@ SDValue VectorLegalizer::LegalizeOp(SDValue Op) {
   case ISD::FFLOOR:
   case ISD::FP_ROUND:
   case ISD::FP_EXTEND:
+  case ISD::FPTRUNC_ROUND:
   case ISD::FMA:
   case ISD::SIGN_EXTEND_INREG:
   case ISD::ANY_EXTEND_VECTOR_INREG:
@@ -1048,6 +1049,13 @@ void VectorLegalizer::Expand(SDNode *Node, SmallVectorImpl<SDValue> &Results) {
       return;
     }
     break;
+  case ISD::FMINIMUM:
+  case ISD::FMAXIMUM:
+    if (SDValue Expanded = TLI.expandFMINIMUM_FMAXIMUM(Node, DAG)) {
+      Results.push_back(Expanded);
+      return;
+    }
+    break;
   case ISD::SMIN:
   case ISD::SMAX:
   case ISD::UMIN:
diff --git a/llvm/lib/CodeGen/SelectionDAG/LegalizeVectorTypes.cpp b/llvm/lib/CodeGen/SelectionDAG/LegalizeVectorTypes.cpp
index 8776d89f4c5b..985c9f16ab97 100644
--- a/llvm/lib/CodeGen/SelectionDAG/LegalizeVectorTypes.cpp
+++ b/llvm/lib/CodeGen/SelectionDAG/LegalizeVectorTypes.cpp
@@ -4242,7 +4242,8 @@ void DAGTypeLegalizer::WidenVectorResult(SDNode *N, unsigned ResNo) {
     break;
 
   case ISD::IS_FPCLASS:
-    Res = WidenVecRes_IS_FPCLASS(N);
+  case ISD::FPTRUNC_ROUND:
+    Res = WidenVecRes_UnarySameEltsWithScalarArg(N);
     break;
 
   case ISD::FLDEXP:
@@ -5004,7 +5005,10 @@ SDValue DAGTypeLegalizer::WidenVecRes_FCOPYSIGN(SDNode *N) {
   return DAG.UnrollVectorOp(N, WidenVT.getVectorNumElements());
 }
 
-SDValue DAGTypeLegalizer::WidenVecRes_IS_FPCLASS(SDNode *N) {
+/// Result and first source operand are different scalar types, but must have
+/// the same number of elements. There is an additional control argument which
+/// should be passed through unchanged.
+SDValue DAGTypeLegalizer::WidenVecRes_UnarySameEltsWithScalarArg(SDNode *N) {
   SDValue FpValue = N->getOperand(0);
   EVT WidenVT = TLI.getTypeToTransformTo(*DAG.getContext(), N->getValueType(0));
   if (getTypeAction(FpValue.getValueType()) != TargetLowering::TypeWidenVector)
diff --git a/llvm/lib/CodeGen/SelectionDAG/ScheduleDAGSDNodes.cpp b/llvm/lib/CodeGen/SelectionDAG/ScheduleDAGSDNodes.cpp
index 379456e3a9b6..f44added89a7 100644
--- a/llvm/lib/CodeGen/SelectionDAG/ScheduleDAGSDNodes.cpp
+++ b/llvm/lib/CodeGen/SelectionDAG/ScheduleDAGSDNodes.cpp
@@ -27,6 +27,7 @@
 #include "llvm/CodeGen/TargetRegisterInfo.h"
 #include "llvm/CodeGen/TargetSubtargetInfo.h"
 #include "llvm/Config/llvm-config.h"
+#include "llvm/IR/MemoryModelRelaxationAnnotations.h"
 #include "llvm/MC/MCInstrItineraries.h"
 #include "llvm/Support/CommandLine.h"
 #include "llvm/Support/Debug.h"
@@ -899,6 +900,14 @@ EmitSchedule(MachineBasicBlock::iterator &InsertPos) {
     if (MDNode *MD = DAG->getPCSections(Node))
       MI->setPCSections(MF, MD);
 
+    // Set MMRAs on _all_ added instructions.
+    if (MDNode *MMRA = DAG->getMMRAMetadata(Node)) {
+      for (MachineBasicBlock::iterator It = MI->getIterator(),
+                                       End = std::next(After);
+           It != End; ++It)
+        It->setMMRAMetadata(MF, MMRA);
+    }
+
     return MI;
   };
 
diff --git a/llvm/lib/CodeGen/SelectionDAG/SelectionDAG.cpp b/llvm/lib/CodeGen/SelectionDAG/SelectionDAG.cpp
index b63b8b893fdb..dfbfaa8c894f 100644
--- a/llvm/lib/CodeGen/SelectionDAG/SelectionDAG.cpp
+++ b/llvm/lib/CodeGen/SelectionDAG/SelectionDAG.cpp
@@ -1418,10 +1418,8 @@ void SelectionDAG::clear() {
   TargetExternalSymbols.clear();
   MCSymbols.clear();
   SDEI.clear();
-  std::fill(CondCodeNodes.begin(), CondCodeNodes.end(),
-            static_cast<CondCodeSDNode*>(nullptr));
-  std::fill(ValueTypeNodes.begin(), ValueTypeNodes.end(),
-            static_cast<SDNode*>(nullptr));
+  std::fill(CondCodeNodes.begin(), CondCodeNodes.end(), nullptr);
+  std::fill(ValueTypeNodes.begin(), ValueTypeNodes.end(), nullptr);
 
   EntryNode.UseList = nullptr;
   InsertNode(&EntryNode);
@@ -1703,8 +1701,9 @@ SDValue SelectionDAG::getConstant(const ConstantInt &Val, const SDLoc &DL,
   assert(Elt->getBitWidth() == EltVT.getSizeInBits() &&
          "APInt size does not match type size!");
   unsigned Opc = isT ? ISD::TargetConstant : ISD::Constant;
+  SDVTList VTs = getVTList(EltVT);
   FoldingSetNodeID ID;
-  AddNodeIDNode(ID, Opc, getVTList(EltVT), std::nullopt);
+  AddNodeIDNode(ID, Opc, VTs, std::nullopt);
   ID.AddPointer(Elt);
   ID.AddBoolean(isO);
   void *IP = nullptr;
@@ -1714,7 +1713,7 @@ SDValue SelectionDAG::getConstant(const ConstantInt &Val, const SDLoc &DL,
       return SDValue(N, 0);
 
   if (!N) {
-    N = newSDNode<ConstantSDNode>(isT, isO, Elt, EltVT);
+    N = newSDNode<ConstantSDNode>(isT, isO, Elt, VTs);
     CSEMap.InsertNode(N, IP);
     InsertNode(N);
     NewSDValueDbgMsg(SDValue(N, 0), "Creating constant: ", this);
@@ -1764,8 +1763,9 @@ SDValue SelectionDAG::getConstantFP(const ConstantFP &V, const SDLoc &DL,
   // value, so that we don't have problems with 0.0 comparing equal to -0.0, and
   // we don't have issues with SNANs.
   unsigned Opc = isTarget ? ISD::TargetConstantFP : ISD::ConstantFP;
+  SDVTList VTs = getVTList(EltVT);
   FoldingSetNodeID ID;
-  AddNodeIDNode(ID, Opc, getVTList(EltVT), std::nullopt);
+  AddNodeIDNode(ID, Opc, VTs, std::nullopt);
   ID.AddPointer(&V);
   void *IP = nullptr;
   SDNode *N = nullptr;
@@ -1774,7 +1774,7 @@ SDValue SelectionDAG::getConstantFP(const ConstantFP &V, const SDLoc &DL,
       return SDValue(N, 0);
 
   if (!N) {
-    N = newSDNode<ConstantFPSDNode>(isTarget, &V, EltVT);
+    N = newSDNode<ConstantFPSDNode>(isTarget, &V, VTs);
     CSEMap.InsertNode(N, IP);
     InsertNode(N);
   }
@@ -1821,8 +1821,9 @@ SDValue SelectionDAG::getGlobalAddress(const GlobalValue *GV, const SDLoc &DL,
   else
     Opc = isTargetGA ? ISD::TargetGlobalAddress : ISD::GlobalAddress;
 
+  SDVTList VTs = getVTList(VT);
   FoldingSetNodeID ID;
-  AddNodeIDNode(ID, Opc, getVTList(VT), std::nullopt);
+  AddNodeIDNode(ID, Opc, VTs, std::nullopt);
   ID.AddPointer(GV);
   ID.AddInteger(Offset);
   ID.AddInteger(TargetFlags);
@@ -1831,7 +1832,7 @@ SDValue SelectionDAG::getGlobalAddress(const GlobalValue *GV, const SDLoc &DL,
     return SDValue(E, 0);
 
   auto *N = newSDNode<GlobalAddressSDNode>(
-      Opc, DL.getIROrder(), DL.getDebugLoc(), GV, VT, Offset, TargetFlags);
+      Opc, DL.getIROrder(), DL.getDebugLoc(), GV, VTs, Offset, TargetFlags);
   CSEMap.InsertNode(N, IP);
     InsertNode(N);
   return SDValue(N, 0);
@@ -1839,14 +1840,15 @@ SDValue SelectionDAG::getGlobalAddress(const GlobalValue *GV, const SDLoc &DL,
 
 SDValue SelectionDAG::getFrameIndex(int FI, EVT VT, bool isTarget) {
   unsigned Opc = isTarget ? ISD::TargetFrameIndex : ISD::FrameIndex;
+  SDVTList VTs = getVTList(VT);
   FoldingSetNodeID ID;
-  AddNodeIDNode(ID, Opc, getVTList(VT), std::nullopt);
+  AddNodeIDNode(ID, Opc, VTs, std::nullopt);
   ID.AddInteger(FI);
   void *IP = nullptr;
   if (SDNode *E = FindNodeOrInsertPos(ID, IP))
     return SDValue(E, 0);
 
-  auto *N = newSDNode<FrameIndexSDNode>(FI, VT, isTarget);
+  auto *N = newSDNode<FrameIndexSDNode>(FI, VTs, isTarget);
   CSEMap.InsertNode(N, IP);
   InsertNode(N);
   return SDValue(N, 0);
@@ -1857,15 +1859,16 @@ SDValue SelectionDAG::getJumpTable(int JTI, EVT VT, bool isTarget,
   assert((TargetFlags == 0 || isTarget) &&
          "Cannot set target flags on target-independent jump tables");
   unsigned Opc = isTarget ? ISD::TargetJumpTable : ISD::JumpTable;
+  SDVTList VTs = getVTList(VT);
   FoldingSetNodeID ID;
-  AddNodeIDNode(ID, Opc, getVTList(VT), std::nullopt);
+  AddNodeIDNode(ID, Opc, VTs, std::nullopt);
   ID.AddInteger(JTI);
   ID.AddInteger(TargetFlags);
   void *IP = nullptr;
   if (SDNode *E = FindNodeOrInsertPos(ID, IP))
     return SDValue(E, 0);
 
-  auto *N = newSDNode<JumpTableSDNode>(JTI, VT, isTarget, TargetFlags);
+  auto *N = newSDNode<JumpTableSDNode>(JTI, VTs, isTarget, TargetFlags);
   CSEMap.InsertNode(N, IP);
   InsertNode(N);
   return SDValue(N, 0);
@@ -1888,8 +1891,9 @@ SDValue SelectionDAG::getConstantPool(const Constant *C, EVT VT,
                     ? getDataLayout().getABITypeAlign(C->getType())
                     : getDataLayout().getPrefTypeAlign(C->getType());
   unsigned Opc = isTarget ? ISD::TargetConstantPool : ISD::ConstantPool;
+  SDVTList VTs = getVTList(VT);
   FoldingSetNodeID ID;
-  AddNodeIDNode(ID, Opc, getVTList(VT), std::nullopt);
+  AddNodeIDNode(ID, Opc, VTs, std::nullopt);
   ID.AddInteger(Alignment->value());
   ID.AddInteger(Offset);
   ID.AddPointer(C);
@@ -1898,7 +1902,7 @@ SDValue SelectionDAG::getConstantPool(const Constant *C, EVT VT,
   if (SDNode *E = FindNodeOrInsertPos(ID, IP))
     return SDValue(E, 0);
 
-  auto *N = newSDNode<ConstantPoolSDNode>(isTarget, C, VT, Offset, *Alignment,
+  auto *N = newSDNode<ConstantPoolSDNode>(isTarget, C, VTs, Offset, *Alignment,
                                           TargetFlags);
   CSEMap.InsertNode(N, IP);
   InsertNode(N);
@@ -1915,8 +1919,9 @@ SDValue SelectionDAG::getConstantPool(MachineConstantPoolValue *C, EVT VT,
   if (!Alignment)
     Alignment = getDataLayout().getPrefTypeAlign(C->getType());
   unsigned Opc = isTarget ? ISD::TargetConstantPool : ISD::ConstantPool;
+  SDVTList VTs = getVTList(VT);
   FoldingSetNodeID ID;
-  AddNodeIDNode(ID, Opc, getVTList(VT), std::nullopt);
+  AddNodeIDNode(ID, Opc, VTs, std::nullopt);
   ID.AddInteger(Alignment->value());
   ID.AddInteger(Offset);
   C->addSelectionDAGCSEId(ID);
@@ -1925,7 +1930,7 @@ SDValue SelectionDAG::getConstantPool(MachineConstantPoolValue *C, EVT VT,
   if (SDNode *E = FindNodeOrInsertPos(ID, IP))
     return SDValue(E, 0);
 
-  auto *N = newSDNode<ConstantPoolSDNode>(isTarget, C, VT, Offset, *Alignment,
+  auto *N = newSDNode<ConstantPoolSDNode>(isTarget, C, VTs, Offset, *Alignment,
                                           TargetFlags);
   CSEMap.InsertNode(N, IP);
   InsertNode(N);
@@ -1963,7 +1968,7 @@ SDValue SelectionDAG::getValueType(EVT VT) {
 SDValue SelectionDAG::getExternalSymbol(const char *Sym, EVT VT) {
   SDNode *&N = ExternalSymbols[Sym];
   if (N) return SDValue(N, 0);
-  N = newSDNode<ExternalSymbolSDNode>(false, Sym, 0, VT);
+  N = newSDNode<ExternalSymbolSDNode>(false, Sym, 0, getVTList(VT));
   InsertNode(N);
   return SDValue(N, 0);
 }
@@ -1972,7 +1977,7 @@ SDValue SelectionDAG::getMCSymbol(MCSymbol *Sym, EVT VT) {
   SDNode *&N = MCSymbols[Sym];
   if (N)
     return SDValue(N, 0);
-  N = newSDNode<MCSymbolSDNode>(Sym, VT);
+  N = newSDNode<MCSymbolSDNode>(Sym, getVTList(VT));
   InsertNode(N);
   return SDValue(N, 0);
 }
@@ -1982,7 +1987,7 @@ SDValue SelectionDAG::getTargetExternalSymbol(const char *Sym, EVT VT,
   SDNode *&N =
       TargetExternalSymbols[std::pair<std::string, unsigned>(Sym, TargetFlags)];
   if (N) return SDValue(N, 0);
-  N = newSDNode<ExternalSymbolSDNode>(true, Sym, TargetFlags, VT);
+  N = newSDNode<ExternalSymbolSDNode>(true, Sym, TargetFlags, getVTList(VT));
   InsertNode(N);
   return SDValue(N, 0);
 }
@@ -2198,9 +2203,10 @@ SDValue SelectionDAG::getVectorShuffle(EVT VT, const SDLoc &dl, SDValue N1,
     }
   }
 
+  SDVTList VTs = getVTList(VT);
   FoldingSetNodeID ID;
   SDValue Ops[2] = { N1, N2 };
-  AddNodeIDNode(ID, ISD::VECTOR_SHUFFLE, getVTList(VT), Ops);
+  AddNodeIDNode(ID, ISD::VECTOR_SHUFFLE, VTs, Ops);
   for (int i = 0; i != NElts; ++i)
     ID.AddInteger(MaskVec[i]);
 
@@ -2214,7 +2220,7 @@ SDValue SelectionDAG::getVectorShuffle(EVT VT, const SDLoc &dl, SDValue N1,
   int *MaskAlloc = OperandAllocator.Allocate<int>(NElts);
   llvm::copy(MaskVec, MaskAlloc);
 
-  auto *N = newSDNode<ShuffleVectorSDNode>(VT, dl.getIROrder(),
+  auto *N = newSDNode<ShuffleVectorSDNode>(VTs, dl.getIROrder(),
                                            dl.getDebugLoc(), MaskAlloc);
   createOperands(N, Ops);
 
@@ -2236,14 +2242,15 @@ SDValue SelectionDAG::getCommutedVectorShuffle(const ShuffleVectorSDNode &SV) {
 }
 
 SDValue SelectionDAG::getRegister(unsigned RegNo, EVT VT) {
+  SDVTList VTs = getVTList(VT);
   FoldingSetNodeID ID;
-  AddNodeIDNode(ID, ISD::Register, getVTList(VT), std::nullopt);
+  AddNodeIDNode(ID, ISD::Register, VTs, std::nullopt);
   ID.AddInteger(RegNo);
   void *IP = nullptr;
   if (SDNode *E = FindNodeOrInsertPos(ID, IP))
     return SDValue(E, 0);
 
-  auto *N = newSDNode<RegisterSDNode>(RegNo, VT);
+  auto *N = newSDNode<RegisterSDNode>(RegNo, VTs);
   N->SDNodeBits.IsDivergent = TLI->isSDNodeSourceOfDivergence(N, FLI, UA);
   CSEMap.InsertNode(N, IP);
   InsertNode(N);
@@ -2292,9 +2299,10 @@ SDValue SelectionDAG::getBlockAddress(const BlockAddress *BA, EVT VT,
                                       int64_t Offset, bool isTarget,
                                       unsigned TargetFlags) {
   unsigned Opc = isTarget ? ISD::TargetBlockAddress : ISD::BlockAddress;
+  SDVTList VTs = getVTList(VT);
 
   FoldingSetNodeID ID;
-  AddNodeIDNode(ID, Opc, getVTList(VT), std::nullopt);
+  AddNodeIDNode(ID, Opc, VTs, std::nullopt);
   ID.AddPointer(BA);
   ID.AddInteger(Offset);
   ID.AddInteger(TargetFlags);
@@ -2302,7 +2310,7 @@ SDValue SelectionDAG::getBlockAddress(const BlockAddress *BA, EVT VT,
   if (SDNode *E = FindNodeOrInsertPos(ID, IP))
     return SDValue(E, 0);
 
-  auto *N = newSDNode<BlockAddressSDNode>(Opc, VT, BA, Offset, TargetFlags);
+  auto *N = newSDNode<BlockAddressSDNode>(Opc, VTs, BA, Offset, TargetFlags);
   CSEMap.InsertNode(N, IP);
   InsertNode(N);
   return SDValue(N, 0);
@@ -2347,9 +2355,10 @@ SDValue SelectionDAG::getBitcast(EVT VT, SDValue V) {
 
 SDValue SelectionDAG::getAddrSpaceCast(const SDLoc &dl, EVT VT, SDValue Ptr,
                                        unsigned SrcAS, unsigned DestAS) {
+  SDVTList VTs = getVTList(VT);
   SDValue Ops[] = {Ptr};
   FoldingSetNodeID ID;
-  AddNodeIDNode(ID, ISD::ADDRSPACECAST, getVTList(VT), Ops);
+  AddNodeIDNode(ID, ISD::ADDRSPACECAST, VTs, Ops);
   ID.AddInteger(SrcAS);
   ID.AddInteger(DestAS);
 
@@ -2358,7 +2367,7 @@ SDValue SelectionDAG::getAddrSpaceCast(const SDLoc &dl, EVT VT, SDValue Ptr,
     return SDValue(E, 0);
 
   auto *N = newSDNode<AddrSpaceCastSDNode>(dl.getIROrder(), dl.getDebugLoc(),
-                                           VT, SrcAS, DestAS);
+                                           VTs, SrcAS, DestAS);
   createOperands(N, Ops);
 
   CSEMap.InsertNode(N, IP);
@@ -2995,6 +3004,14 @@ SelectionDAG::getValidShiftAmountConstant(SDValue V,
   return nullptr;
 }
 
+const APInt *SelectionDAG::getValidShiftAmountConstant(SDValue V) const {
+  EVT VT = V.getValueType();
+  APInt DemandedElts = VT.isFixedLengthVector()
+                           ? APInt::getAllOnes(VT.getVectorNumElements())
+                           : APInt(1, 1);
+  return getValidShiftAmountConstant(V, DemandedElts);
+}
+
 const APInt *SelectionDAG::getValidMinimumShiftAmountConstant(
     SDValue V, const APInt &DemandedElts) const {
   assert((V.getOpcode() == ISD::SHL || V.getOpcode() == ISD::SRL ||
@@ -3024,6 +3041,14 @@ const APInt *SelectionDAG::getValidMinimumShiftAmountConstant(
   return MinShAmt;
 }
 
+const APInt *SelectionDAG::getValidMinimumShiftAmountConstant(SDValue V) const {
+  EVT VT = V.getValueType();
+  APInt DemandedElts = VT.isFixedLengthVector()
+                           ? APInt::getAllOnes(VT.getVectorNumElements())
+                           : APInt(1, 1);
+  return getValidMinimumShiftAmountConstant(V, DemandedElts);
+}
+
 const APInt *SelectionDAG::getValidMaximumShiftAmountConstant(
     SDValue V, const APInt &DemandedElts) const {
   assert((V.getOpcode() == ISD::SHL || V.getOpcode() == ISD::SRL ||
@@ -3053,6 +3078,14 @@ const APInt *SelectionDAG::getValidMaximumShiftAmountConstant(
   return MaxShAmt;
 }
 
+const APInt *SelectionDAG::getValidMaximumShiftAmountConstant(SDValue V) const {
+  EVT VT = V.getValueType();
+  APInt DemandedElts = VT.isFixedLengthVector()
+                           ? APInt::getAllOnes(VT.getVectorNumElements())
+                           : APInt(1, 1);
+  return getValidMaximumShiftAmountConstant(V, DemandedElts);
+}
+
 /// Determine which bits of Op are known to be either zero or one and return
 /// them in Known. For vectors, the known bits are those that are shared by
 /// every vector element.
@@ -5030,6 +5063,7 @@ bool SelectionDAG::isGuaranteedNotToBeUndefOrPoison(SDValue Op,
   case ISD::VALUETYPE:
   case ISD::FrameIndex:
   case ISD::TargetFrameIndex:
+  case ISD::CopyFromReg:
     return true;
 
   case ISD::UNDEF:
@@ -5095,11 +5129,24 @@ bool SelectionDAG::canCreateUndefOrPoison(SDValue Op, const APInt &DemandedElts,
   if (VT.isScalableVector())
     return true;
 
+  if (ConsiderFlags && Op->hasPoisonGeneratingFlags())
+    return true;
+
   unsigned Opcode = Op.getOpcode();
   switch (Opcode) {
   case ISD::FREEZE:
   case ISD::CONCAT_VECTORS:
   case ISD::INSERT_SUBVECTOR:
+  case ISD::SADDSAT:
+  case ISD::UADDSAT:
+  case ISD::SSUBSAT:
+  case ISD::USUBSAT:
+  case ISD::MULHU:
+  case ISD::MULHS:
+  case ISD::SMIN:
+  case ISD::SMAX:
+  case ISD::UMIN:
+  case ISD::UMAX:
   case ISD::AND:
   case ISD::XOR:
   case ISD::ROTL:
@@ -5120,6 +5167,7 @@ bool SelectionDAG::canCreateUndefOrPoison(SDValue Op, const APInt &DemandedElts,
   case ISD::BUILD_PAIR:
     return false;
 
+  case ISD::SELECT_CC:
   case ISD::SETCC: {
     // Integer setcc cannot create undef or poison.
     if (Op.getOperand(0).getValueType().isInteger())
@@ -5129,39 +5177,28 @@ bool SelectionDAG::canCreateUndefOrPoison(SDValue Op, const APInt &DemandedElts,
     // based on options and flags. The options and flags also cause special
     // nonan condition codes to be used. Those condition codes may be preserved
     // even if the nonan flag is dropped somewhere.
-    ISD::CondCode CCCode = cast<CondCodeSDNode>(Op.getOperand(2))->get();
+    unsigned CCOp = Opcode == ISD::SETCC ? 2 : 4;
+    ISD::CondCode CCCode = cast<CondCodeSDNode>(Op.getOperand(CCOp))->get();
     if (((unsigned)CCCode & 0x10U))
       return true;
 
     const TargetOptions &Options = getTarget().Options;
-    return Options.NoNaNsFPMath || Options.NoInfsFPMath ||
-           (ConsiderFlags &&
-            (Op->getFlags().hasNoNaNs() || Op->getFlags().hasNoInfs()));
+    return Options.NoNaNsFPMath || Options.NoInfsFPMath;
   }
 
-  // Matches hasPoisonGeneratingFlags().
+  case ISD::OR:
   case ISD::ZERO_EXTEND:
-    return ConsiderFlags && Op->getFlags().hasNonNeg();
-
   case ISD::ADD:
   case ISD::SUB:
   case ISD::MUL:
-    // Matches hasPoisonGeneratingFlags().
-    return ConsiderFlags && (Op->getFlags().hasNoSignedWrap() ||
-                             Op->getFlags().hasNoUnsignedWrap());
+    // No poison except from flags (which is handled above)
+    return false;
 
   case ISD::SHL:
+  case ISD::SRL:
+  case ISD::SRA:
     // If the max shift amount isn't in range, then the shift can create poison.
-    if (!getValidMaximumShiftAmountConstant(Op, DemandedElts))
-      return true;
-
-    // Matches hasPoisonGeneratingFlags().
-    return ConsiderFlags && (Op->getFlags().hasNoSignedWrap() ||
-                             Op->getFlags().hasNoUnsignedWrap());
-
-  // Matches hasPoisonGeneratingFlags().
-  case ISD::OR:
-    return ConsiderFlags && Op->getFlags().hasDisjoint();
+    return !getValidMaximumShiftAmountConstant(Op, DemandedElts);
 
   case ISD::SCALAR_TO_VECTOR:
     // Check if we demand any upper (undef) elements.
@@ -5691,14 +5728,14 @@ static SDValue foldCONCAT_VECTORS(const SDLoc &DL, EVT VT,
 
 /// Gets or creates the specified node.
 SDValue SelectionDAG::getNode(unsigned Opcode, const SDLoc &DL, EVT VT) {
+  SDVTList VTs = getVTList(VT);
   FoldingSetNodeID ID;
-  AddNodeIDNode(ID, Opcode, getVTList(VT), std::nullopt);
+  AddNodeIDNode(ID, Opcode, VTs, std::nullopt);
   void *IP = nullptr;
   if (SDNode *E = FindNodeOrInsertPos(ID, DL, IP))
     return SDValue(E, 0);
 
-  auto *N = newSDNode<SDNode>(Opcode, DL.getIROrder(), DL.getDebugLoc(),
-                              getVTList(VT));
+  auto *N = newSDNode<SDNode>(Opcode, DL.getIROrder(), DL.getDebugLoc(), VTs);
   CSEMap.InsertNode(N, IP);
 
   InsertNode(N);
@@ -6645,16 +6682,17 @@ SDValue SelectionDAG::getAssertAlign(const SDLoc &DL, SDValue Val, Align A) {
   if (A == Align(1))
     return Val;
 
+  SDVTList VTs = getVTList(Val.getValueType());
   FoldingSetNodeID ID;
-  AddNodeIDNode(ID, ISD::AssertAlign, getVTList(Val.getValueType()), {Val});
+  AddNodeIDNode(ID, ISD::AssertAlign, VTs, {Val});
   ID.AddInteger(A.value());
 
   void *IP = nullptr;
   if (SDNode *E = FindNodeOrInsertPos(ID, DL, IP))
     return SDValue(E, 0);
 
-  auto *N = newSDNode<AssertAlignSDNode>(DL.getIROrder(), DL.getDebugLoc(),
-                                         Val.getValueType(), A);
+  auto *N =
+      newSDNode<AssertAlignSDNode>(DL.getIROrder(), DL.getDebugLoc(), VTs, A);
   createOperands(N, {Val});
 
   CSEMap.InsertNode(N, IP);
@@ -11773,20 +11811,6 @@ HandleSDNode::~HandleSDNode() {
   DropOperands();
 }
 
-GlobalAddressSDNode::GlobalAddressSDNode(unsigned Opc, unsigned Order,
-                                         const DebugLoc &DL,
-                                         const GlobalValue *GA, EVT VT,
-                                         int64_t o, unsigned TF)
-    : SDNode(Opc, Order, DL, getSDVTList(VT)), Offset(o), TargetFlags(TF) {
-  TheGlobal = GA;
-}
-
-AddrSpaceCastSDNode::AddrSpaceCastSDNode(unsigned Order, const DebugLoc &dl,
-                                         EVT VT, unsigned SrcAS,
-                                         unsigned DestAS)
-    : SDNode(ISD::ADDRSPACECAST, Order, dl, getSDVTList(VT)),
-      SrcAddrSpace(SrcAS), DestAddrSpace(DestAS) {}
-
 MemSDNode::MemSDNode(unsigned Opc, unsigned Order, const DebugLoc &dl,
                      SDVTList VTs, EVT memvt, MachineMemOperand *mmo)
     : SDNode(Opc, Order, dl, VTs), MemoryVT(memvt), MMO(mmo) {
@@ -12946,7 +12970,7 @@ void SelectionDAG::copyExtraInfo(SDNode *From, SDNode *To) {
   // Use of operator[] on the DenseMap may cause an insertion, which invalidates
   // the iterator, hence the need to make a copy to prevent a use-after-free.
   NodeExtraInfo NEI = I->second;
-  if (LLVM_LIKELY(!NEI.PCSections)) {
+  if (LLVM_LIKELY(!NEI.PCSections) && LLVM_LIKELY(!NEI.MMRA)) {
     // No deep copy required for the types of extra info set.
     //
     // FIXME: Investigate if other types of extra info also need deep copy. This
diff --git a/llvm/lib/CodeGen/SelectionDAG/SelectionDAGBuilder.cpp b/llvm/lib/CodeGen/SelectionDAG/SelectionDAGBuilder.cpp
index 3f69f7ad5447..5caf868c83a2 100644
--- a/llvm/lib/CodeGen/SelectionDAG/SelectionDAGBuilder.cpp
+++ b/llvm/lib/CodeGen/SelectionDAG/SelectionDAGBuilder.cpp
@@ -80,6 +80,7 @@
 #include "llvm/IR/IntrinsicsAMDGPU.h"
 #include "llvm/IR/IntrinsicsWebAssembly.h"
 #include "llvm/IR/LLVMContext.h"
+#include "llvm/IR/MemoryModelRelaxationAnnotations.h"
 #include "llvm/IR/Metadata.h"
 #include "llvm/IR/Module.h"
 #include "llvm/IR/Operator.h"
@@ -1326,7 +1327,8 @@ void SelectionDAGBuilder::visit(const Instruction &I) {
   bool NodeInserted = false;
   std::unique_ptr<SelectionDAG::DAGNodeInsertedListener> InsertedListener;
   MDNode *PCSectionsMD = I.getMetadata(LLVMContext::MD_pcsections);
-  if (PCSectionsMD) {
+  MDNode *MMRA = I.getMetadata(LLVMContext::MD_mmra);
+  if (PCSectionsMD || MMRA) {
     InsertedListener = std::make_unique<SelectionDAG::DAGNodeInsertedListener>(
         DAG, [&](SDNode *) { NodeInserted = true; });
   }
@@ -1338,14 +1340,17 @@ void SelectionDAGBuilder::visit(const Instruction &I) {
     CopyToExportRegsIfNeeded(&I);
 
   // Handle metadata.
-  if (PCSectionsMD) {
+  if (PCSectionsMD || MMRA) {
     auto It = NodeMap.find(&I);
     if (It != NodeMap.end()) {
-      DAG.addPCSections(It->second.getNode(), PCSectionsMD);
+      if (PCSectionsMD)
+        DAG.addPCSections(It->second.getNode(), PCSectionsMD);
+      if (MMRA)
+        DAG.addMMRAMetadata(It->second.getNode(), MMRA);
     } else if (NodeInserted) {
       // This should not happen; if it does, don't let it go unnoticed so we can
       // fix it. Relevant visit*() function is probably missing a setValue().
-      errs() << "warning: loosing !pcsections metadata ["
+      errs() << "warning: loosing !pcsections and/or !mmra metadata ["
              << I.getModule()->getName() << "]\n";
       LLVM_DEBUG(I.dump());
       assert(false);
@@ -5294,9 +5299,9 @@ void SelectionDAGBuilder::visitTargetIntrinsic(const CallInst &I,
       Result =
           DAG.getAssertAlign(getCurSDLoc(), Result, Alignment.valueOrOne());
     }
-
-    setValue(&I, Result);
   }
+
+  setValue(&I, Result);
 }
 
 /// GetSignificand - Get the significand and build it into a floating-point
@@ -7925,19 +7930,19 @@ void SelectionDAGBuilder::visitIntrinsicCall(const CallInst &I,
              DAG.getNode(ISD::EXTRACT_SUBVECTOR, sdl, ResultVT, Vec, Index));
     return;
   }
-  case Intrinsic::experimental_vector_reverse:
+  case Intrinsic::vector_reverse:
     visitVectorReverse(I);
     return;
-  case Intrinsic::experimental_vector_splice:
+  case Intrinsic::vector_splice:
     visitVectorSplice(I);
     return;
   case Intrinsic::callbr_landingpad:
     visitCallBrLandingPad(I);
     return;
-  case Intrinsic::experimental_vector_interleave2:
+  case Intrinsic::vector_interleave2:
     visitVectorInterleave(I);
     return;
-  case Intrinsic::experimental_vector_deinterleave2:
+  case Intrinsic::vector_deinterleave2:
     visitVectorDeinterleave(I);
     return;
   case Intrinsic::experimental_convergence_anchor:
@@ -11128,7 +11133,7 @@ static void tryToElideArgumentCopy(
   }
 
   // Perform the elision. Delete the old stack object and replace its only use
-  // in the variable info map. Mark the stack object as mutable.
+  // in the variable info map. Mark the stack object as mutable and aliased.
   LLVM_DEBUG({
     dbgs() << "Eliding argument copy from " << Arg << " to " << *AI << '\n'
            << "  Replacing frame index " << OldIndex << " with " << FixedIndex
@@ -11136,6 +11141,7 @@ static void tryToElideArgumentCopy(
   });
   MFI.RemoveStackObject(OldIndex);
   MFI.setIsImmutableObjectIndex(FixedIndex, false);
+  MFI.setIsAliasedObjectIndex(FixedIndex, true);
   AllocaIndex = FixedIndex;
   ArgCopyElisionFrameIndexMap.insert({OldIndex, FixedIndex});
   for (SDValue ArgVal : ArgVals)
diff --git a/llvm/lib/CodeGen/SelectionDAG/SelectionDAGDumper.cpp b/llvm/lib/CodeGen/SelectionDAG/SelectionDAGDumper.cpp
index 6691aa41face..4ad4a938ca97 100644
--- a/llvm/lib/CodeGen/SelectionDAG/SelectionDAGDumper.cpp
+++ b/llvm/lib/CodeGen/SelectionDAG/SelectionDAGDumper.cpp
@@ -299,6 +299,7 @@ std::string SDNode::getOperationName(const SelectionDAG *G) const {
   case ISD::SETCCCARRY:                 return "setcccarry";
   case ISD::STRICT_FSETCC:              return "strict_fsetcc";
   case ISD::STRICT_FSETCCS:             return "strict_fsetccs";
+  case ISD::FPTRUNC_ROUND:              return "fptrunc_round";
   case ISD::SELECT:                     return "select";
   case ISD::VSELECT:                    return "vselect";
   case ISD::SELECT_CC:                  return "select_cc";
@@ -905,6 +906,13 @@ void SDNode::print_details(raw_ostream &OS, const SelectionDAG *G) const {
       MD->printAsOperand(OS, G->getMachineFunction().getFunction().getParent());
       OS << ']';
     }
+
+    if (MDNode *MMRA = G ? G->getMMRAMetadata(this) : nullptr) {
+      OS << " [mmra ";
+      MMRA->printAsOperand(OS,
+                           G->getMachineFunction().getFunction().getParent());
+      OS << ']';
+    }
   }
 }
 
diff --git a/llvm/lib/CodeGen/SelectionDAG/SelectionDAGISel.cpp b/llvm/lib/CodeGen/SelectionDAG/SelectionDAGISel.cpp
index d629c36bc792..b5694c955b8c 100644
--- a/llvm/lib/CodeGen/SelectionDAG/SelectionDAGISel.cpp
+++ b/llvm/lib/CodeGen/SelectionDAG/SelectionDAGISel.cpp
@@ -1059,6 +1059,8 @@ public:
     SDNode *CurNode = &*ISelPosition;
     if (MDNode *MD = DAG.getPCSections(CurNode))
       DAG.addPCSections(N, MD);
+    if (MDNode *MMRA = DAG.getMMRAMetadata(CurNode))
+      DAG.addMMRAMetadata(N, MMRA);
   }
 };
 
diff --git a/llvm/lib/CodeGen/SelectionDAG/TargetLowering.cpp b/llvm/lib/CodeGen/SelectionDAG/TargetLowering.cpp
index c938b3996be3..cdc1227fd572 100644
--- a/llvm/lib/CodeGen/SelectionDAG/TargetLowering.cpp
+++ b/llvm/lib/CodeGen/SelectionDAG/TargetLowering.cpp
@@ -8381,6 +8381,64 @@ SDValue TargetLowering::expandFMINNUM_FMAXNUM(SDNode *Node,
   return SDValue();
 }
 
+SDValue TargetLowering::expandFMINIMUM_FMAXIMUM(SDNode *N,
+                                                SelectionDAG &DAG) const {
+  SDLoc DL(N);
+  SDValue LHS = N->getOperand(0);
+  SDValue RHS = N->getOperand(1);
+  unsigned Opc = N->getOpcode();
+  EVT VT = N->getValueType(0);
+  EVT CCVT = getSetCCResultType(DAG.getDataLayout(), *DAG.getContext(), VT);
+  bool IsMax = Opc == ISD::FMAXIMUM;
+
+  if (VT.isVector() &&
+      isOperationLegalOrCustomOrPromote(Opc, VT.getScalarType()))
+    return SDValue();
+
+  // First, implement comparison not propagating NaN. If no native fmin or fmax
+  // available, use plain select with setcc instead.
+  SDValue MinMax;
+  unsigned CompOpcIeee = IsMax ? ISD::FMAXNUM_IEEE : ISD::FMINNUM_IEEE;
+  unsigned CompOpc = IsMax ? ISD::FMAXNUM : ISD::FMINNUM;
+  if (isOperationLegalOrCustom(CompOpcIeee, VT)) {
+    MinMax = DAG.getNode(CompOpcIeee, DL, VT, LHS, RHS);
+  } else if (isOperationLegalOrCustom(CompOpc, VT)) {
+    MinMax = DAG.getNode(CompOpc, DL, VT, LHS, RHS);
+  } else {
+    // NaN (if exists) will be propagated later, so orderness doesn't matter.
+    SDValue Compare =
+        DAG.getSetCC(DL, CCVT, LHS, RHS, IsMax ? ISD::SETGT : ISD::SETLT);
+    MinMax = DAG.getSelect(DL, VT, Compare, LHS, RHS);
+  }
+
+  // Propagate any NaN of both operands
+  if (!N->getFlags().hasNoNaNs() &&
+      (!DAG.isKnownNeverNaN(RHS) || !DAG.isKnownNeverNaN(LHS))) {
+    ConstantFP *FPNaN = ConstantFP::get(
+        *DAG.getContext(), APFloat::getNaN(DAG.EVTToAPFloatSemantics(VT)));
+    MinMax = DAG.getSelect(DL, VT, DAG.getSetCC(DL, CCVT, LHS, RHS, ISD::SETUO),
+                           DAG.getConstantFP(*FPNaN, DL, VT), MinMax);
+  }
+
+  // fminimum/fmaximum requires -0.0 less than +0.0
+  if (!N->getFlags().hasNoSignedZeros() && !DAG.isKnownNeverZeroFloat(RHS) &&
+      !DAG.isKnownNeverZeroFloat(LHS)) {
+    SDValue IsZero = DAG.getSetCC(DL, CCVT, MinMax,
+                                  DAG.getConstantFP(0.0, DL, VT), ISD::SETEQ);
+    SDValue TestZero =
+        DAG.getTargetConstant(IsMax ? fcPosZero : fcNegZero, DL, MVT::i32);
+    SDValue LCmp = DAG.getSelect(
+        DL, VT, DAG.getNode(ISD::IS_FPCLASS, DL, CCVT, LHS, TestZero), LHS,
+        MinMax);
+    SDValue RCmp = DAG.getSelect(
+        DL, VT, DAG.getNode(ISD::IS_FPCLASS, DL, CCVT, RHS, TestZero), RHS,
+        LCmp);
+    MinMax = DAG.getSelect(DL, VT, IsZero, RCmp, MinMax);
+  }
+
+  return MinMax;
+}
+
 /// Returns a true value if if this FPClassTest can be performed with an ordered
 /// fcmp to 0, and a false value if it's an unordered fcmp to 0. Returns
 /// std::nullopt if it cannot be performed as a compare with 0.
diff --git a/llvm/lib/CodeGen/StackSlotColoring.cpp b/llvm/lib/CodeGen/StackSlotColoring.cpp
index 6d3fc740b292..9fdc8a338b52 100644
--- a/llvm/lib/CodeGen/StackSlotColoring.cpp
+++ b/llvm/lib/CodeGen/StackSlotColoring.cpp
@@ -486,7 +486,8 @@ bool StackSlotColoring::RemoveDeadStores(MachineBasicBlock* MBB) {
     ++NumDead;
     changed = true;
 
-    if (NextMI->findRegisterUseOperandIdx(LoadReg, true, nullptr) != -1) {
+    if (NextMI->findRegisterUseOperandIdx(LoadReg, /*TRI=*/nullptr, true) !=
+        -1) {
       ++NumDead;
       toErase.push_back(&*ProbableLoadMI);
     }
diff --git a/llvm/lib/CodeGen/TargetInstrInfo.cpp b/llvm/lib/CodeGen/TargetInstrInfo.cpp
index 14b2e4268eb0..e01e7b388891 100644
--- a/llvm/lib/CodeGen/TargetInstrInfo.cpp
+++ b/llvm/lib/CodeGen/TargetInstrInfo.cpp
@@ -1055,12 +1055,34 @@ static std::pair<bool, bool> mustSwapOperands(unsigned Pattern) {
   }
 }
 
+void TargetInstrInfo::getReassociateOperandIndices(
+    const MachineInstr &Root, unsigned Pattern,
+    std::array<unsigned, 5> &OperandIndices) const {
+  switch (Pattern) {
+  case MachineCombinerPattern::REASSOC_AX_BY:
+    OperandIndices = {1, 1, 1, 2, 2};
+    break;
+  case MachineCombinerPattern::REASSOC_AX_YB:
+    OperandIndices = {2, 1, 2, 2, 1};
+    break;
+  case MachineCombinerPattern::REASSOC_XA_BY:
+    OperandIndices = {1, 2, 1, 1, 2};
+    break;
+  case MachineCombinerPattern::REASSOC_XA_YB:
+    OperandIndices = {2, 2, 2, 1, 1};
+    break;
+  default:
+    llvm_unreachable("unexpected MachineCombinerPattern");
+  }
+}
+
 /// Attempt the reassociation transformation to reduce critical path length.
 /// See the above comments before getMachineCombinerPatterns().
 void TargetInstrInfo::reassociateOps(
     MachineInstr &Root, MachineInstr &Prev, unsigned Pattern,
     SmallVectorImpl<MachineInstr *> &InsInstrs,
     SmallVectorImpl<MachineInstr *> &DelInstrs,
+    ArrayRef<unsigned> OperandIndices,
     DenseMap<unsigned, unsigned> &InstrIdxForVirtReg) const {
   MachineFunction *MF = Root.getMF();
   MachineRegisterInfo &MRI = MF->getRegInfo();
@@ -1068,29 +1090,10 @@ void TargetInstrInfo::reassociateOps(
   const TargetRegisterInfo *TRI = MF->getSubtarget().getRegisterInfo();
   const TargetRegisterClass *RC = Root.getRegClassConstraint(0, TII, TRI);
 
-  // This array encodes the operand index for each parameter because the
-  // operands may be commuted. Each row corresponds to a pattern value,
-  // and each column specifies the index of A, B, X, Y.
-  unsigned OpIdx[4][4] = {
-    { 1, 1, 2, 2 },
-    { 1, 2, 2, 1 },
-    { 2, 1, 1, 2 },
-    { 2, 2, 1, 1 }
-  };
-
-  int Row;
-  switch (Pattern) {
-  case MachineCombinerPattern::REASSOC_AX_BY: Row = 0; break;
-  case MachineCombinerPattern::REASSOC_AX_YB: Row = 1; break;
-  case MachineCombinerPattern::REASSOC_XA_BY: Row = 2; break;
-  case MachineCombinerPattern::REASSOC_XA_YB: Row = 3; break;
-  default: llvm_unreachable("unexpected MachineCombinerPattern");
-  }
-
-  MachineOperand &OpA = Prev.getOperand(OpIdx[Row][0]);
-  MachineOperand &OpB = Root.getOperand(OpIdx[Row][1]);
-  MachineOperand &OpX = Prev.getOperand(OpIdx[Row][2]);
-  MachineOperand &OpY = Root.getOperand(OpIdx[Row][3]);
+  MachineOperand &OpA = Prev.getOperand(OperandIndices[1]);
+  MachineOperand &OpB = Root.getOperand(OperandIndices[2]);
+  MachineOperand &OpX = Prev.getOperand(OperandIndices[3]);
+  MachineOperand &OpY = Root.getOperand(OperandIndices[4]);
   MachineOperand &OpC = Root.getOperand(0);
 
   Register RegA = OpA.getReg();
@@ -1129,11 +1132,62 @@ void TargetInstrInfo::reassociateOps(
     std::swap(KillX, KillY);
   }
 
+  unsigned PrevFirstOpIdx, PrevSecondOpIdx;
+  unsigned RootFirstOpIdx, RootSecondOpIdx;
+  switch (Pattern) {
+  case MachineCombinerPattern::REASSOC_AX_BY:
+    PrevFirstOpIdx = OperandIndices[1];
+    PrevSecondOpIdx = OperandIndices[3];
+    RootFirstOpIdx = OperandIndices[2];
+    RootSecondOpIdx = OperandIndices[4];
+    break;
+  case MachineCombinerPattern::REASSOC_AX_YB:
+    PrevFirstOpIdx = OperandIndices[1];
+    PrevSecondOpIdx = OperandIndices[3];
+    RootFirstOpIdx = OperandIndices[4];
+    RootSecondOpIdx = OperandIndices[2];
+    break;
+  case MachineCombinerPattern::REASSOC_XA_BY:
+    PrevFirstOpIdx = OperandIndices[3];
+    PrevSecondOpIdx = OperandIndices[1];
+    RootFirstOpIdx = OperandIndices[2];
+    RootSecondOpIdx = OperandIndices[4];
+    break;
+  case MachineCombinerPattern::REASSOC_XA_YB:
+    PrevFirstOpIdx = OperandIndices[3];
+    PrevSecondOpIdx = OperandIndices[1];
+    RootFirstOpIdx = OperandIndices[4];
+    RootSecondOpIdx = OperandIndices[2];
+    break;
+  default:
+    llvm_unreachable("unexpected MachineCombinerPattern");
+  }
+
+  // Basically BuildMI but doesn't add implicit operands by default.
+  auto buildMINoImplicit = [](MachineFunction &MF, const MIMetadata &MIMD,
+                              const MCInstrDesc &MCID, Register DestReg) {
+    return MachineInstrBuilder(
+               MF, MF.CreateMachineInstr(MCID, MIMD.getDL(), /*NoImpl=*/true))
+        .setPCSections(MIMD.getPCSections())
+        .addReg(DestReg, RegState::Define);
+  };
+
   // Create new instructions for insertion.
   MachineInstrBuilder MIB1 =
-      BuildMI(*MF, MIMetadata(Prev), TII->get(NewPrevOpc), NewVR)
-          .addReg(RegX, getKillRegState(KillX))
-          .addReg(RegY, getKillRegState(KillY));
+      buildMINoImplicit(*MF, MIMetadata(Prev), TII->get(NewPrevOpc), NewVR);
+  for (const auto &MO : Prev.explicit_operands()) {
+    unsigned Idx = MO.getOperandNo();
+    // Skip the result operand we'd already added.
+    if (Idx == 0)
+      continue;
+    if (Idx == PrevFirstOpIdx)
+      MIB1.addReg(RegX, getKillRegState(KillX));
+    else if (Idx == PrevSecondOpIdx)
+      MIB1.addReg(RegY, getKillRegState(KillY));
+    else
+      MIB1.add(MO);
+  }
+  MIB1.copyImplicitOps(Prev);
 
   if (SwapRootOperands) {
     std::swap(RegA, NewVR);
@@ -1141,9 +1195,20 @@ void TargetInstrInfo::reassociateOps(
   }
 
   MachineInstrBuilder MIB2 =
-      BuildMI(*MF, MIMetadata(Root), TII->get(NewRootOpc), RegC)
-          .addReg(RegA, getKillRegState(KillA))
-          .addReg(NewVR, getKillRegState(KillNewVR));
+      buildMINoImplicit(*MF, MIMetadata(Root), TII->get(NewRootOpc), RegC);
+  for (const auto &MO : Root.explicit_operands()) {
+    unsigned Idx = MO.getOperandNo();
+    // Skip the result operand.
+    if (Idx == 0)
+      continue;
+    if (Idx == RootFirstOpIdx)
+      MIB2 = MIB2.addReg(RegA, getKillRegState(KillA));
+    else if (Idx == RootSecondOpIdx)
+      MIB2 = MIB2.addReg(NewVR, getKillRegState(KillNewVR));
+    else
+      MIB2 = MIB2.add(MO);
+  }
+  MIB2.copyImplicitOps(Root);
 
   // Propagate FP flags from the original instructions.
   // But clear poison-generating flags because those may not be valid now.
@@ -1187,25 +1252,17 @@ void TargetInstrInfo::genAlternativeCodeSequence(
   MachineRegisterInfo &MRI = Root.getMF()->getRegInfo();
 
   // Select the previous instruction in the sequence based on the input pattern.
-  MachineInstr *Prev = nullptr;
-  switch (Pattern) {
-  case MachineCombinerPattern::REASSOC_AX_BY:
-  case MachineCombinerPattern::REASSOC_XA_BY:
-    Prev = MRI.getUniqueVRegDef(Root.getOperand(1).getReg());
-    break;
-  case MachineCombinerPattern::REASSOC_AX_YB:
-  case MachineCombinerPattern::REASSOC_XA_YB:
-    Prev = MRI.getUniqueVRegDef(Root.getOperand(2).getReg());
-    break;
-  default:
-    llvm_unreachable("Unknown pattern for machine combiner");
-  }
+  std::array<unsigned, 5> OperandIndices;
+  getReassociateOperandIndices(Root, Pattern, OperandIndices);
+  MachineInstr *Prev =
+      MRI.getUniqueVRegDef(Root.getOperand(OperandIndices[0]).getReg());
 
   // Don't reassociate if Prev and Root are in different blocks.
   if (Prev->getParent() != Root.getParent())
     return;
 
-  reassociateOps(Root, *Prev, Pattern, InsInstrs, DelInstrs, InstIdxForVirtReg);
+  reassociateOps(Root, *Prev, Pattern, InsInstrs, DelInstrs, OperandIndices,
+                 InstIdxForVirtReg);
 }
 
 MachineTraceStrategy TargetInstrInfo::getMachineCombinerTraceStrategy() const {
diff --git a/llvm/lib/CodeGen/TwoAddressInstructionPass.cpp b/llvm/lib/CodeGen/TwoAddressInstructionPass.cpp
index ebacbc420f85..b9b2841e7c9e 100644
--- a/llvm/lib/CodeGen/TwoAddressInstructionPass.cpp
+++ b/llvm/lib/CodeGen/TwoAddressInstructionPass.cpp
@@ -339,7 +339,7 @@ bool TwoAddressInstructionPass::isPlainlyKilled(const MachineInstr *MI,
     });
   }
 
-  return MI->killsRegister(Reg);
+  return MI->killsRegister(Reg, /*TRI=*/nullptr);
 }
 
 /// Test if the register used by the given operand is killed by the operand's
@@ -1355,8 +1355,10 @@ tryInstructionTransform(MachineBasicBlock::iterator &mi,
                           << "2addr:    NEW INST: " << *NewMIs[1]);
 
         // Transform the instruction, now that it no longer has a load.
-        unsigned NewDstIdx = NewMIs[1]->findRegisterDefOperandIdx(regA);
-        unsigned NewSrcIdx = NewMIs[1]->findRegisterUseOperandIdx(regB);
+        unsigned NewDstIdx =
+            NewMIs[1]->findRegisterDefOperandIdx(regA, /*TRI=*/nullptr);
+        unsigned NewSrcIdx =
+            NewMIs[1]->findRegisterUseOperandIdx(regB, /*TRI=*/nullptr);
         MachineBasicBlock::iterator NewMI = NewMIs[1];
         bool TransformResult =
           tryInstructionTransform(NewMI, mi, NewSrcIdx, NewDstIdx, Dist, true);
@@ -1371,19 +1373,22 @@ tryInstructionTransform(MachineBasicBlock::iterator &mi,
               if (MO.isReg() && MO.getReg().isVirtual()) {
                 if (MO.isUse()) {
                   if (MO.isKill()) {
-                    if (NewMIs[0]->killsRegister(MO.getReg()))
+                    if (NewMIs[0]->killsRegister(MO.getReg(), /*TRI=*/nullptr))
                       LV->replaceKillInstruction(MO.getReg(), MI, *NewMIs[0]);
                     else {
-                      assert(NewMIs[1]->killsRegister(MO.getReg()) &&
+                      assert(NewMIs[1]->killsRegister(MO.getReg(),
+                                                      /*TRI=*/nullptr) &&
                              "Kill missing after load unfold!");
                       LV->replaceKillInstruction(MO.getReg(), MI, *NewMIs[1]);
                     }
                   }
                 } else if (LV->removeVirtualRegisterDead(MO.getReg(), MI)) {
-                  if (NewMIs[1]->registerDefIsDead(MO.getReg()))
+                  if (NewMIs[1]->registerDefIsDead(MO.getReg(),
+                                                   /*TRI=*/nullptr))
                     LV->addVirtualRegisterDead(MO.getReg(), *NewMIs[1]);
                   else {
-                    assert(NewMIs[0]->registerDefIsDead(MO.getReg()) &&
+                    assert(NewMIs[0]->registerDefIsDead(MO.getReg(),
+                                                        /*TRI=*/nullptr) &&
                            "Dead flag missing after load unfold!");
                     LV->addVirtualRegisterDead(MO.getReg(), *NewMIs[0]);
                   }
diff --git a/llvm/lib/DebugInfo/DWARF/DWARFUnit.cpp b/llvm/lib/DebugInfo/DWARF/DWARFUnit.cpp
index 9f455fa7e96a..bdd04b00f557 100644
--- a/llvm/lib/DebugInfo/DWARF/DWARFUnit.cpp
+++ b/llvm/lib/DebugInfo/DWARF/DWARFUnit.cpp
@@ -98,8 +98,12 @@ void DWARFUnitVector::addUnitsImpl(
         if (!IndexEntry)
           IndexEntry = Index.getFromOffset(Header.getOffset());
       }
-      if (IndexEntry && !Header.applyIndexEntry(IndexEntry))
-        return nullptr;
+      if (IndexEntry) {
+        if (Error ApplicationErr = Header.applyIndexEntry(IndexEntry)) {
+          Context.getWarningHandler()(std::move(ApplicationErr));
+          return nullptr;
+        }
+      }
       std::unique_ptr<DWARFUnit> U;
       if (Header.isTypeUnit())
         U = std::make_unique<DWARFTypeUnit>(Context, InfoSection, Header, DA,
@@ -334,21 +338,40 @@ Error DWARFUnitHeader::extract(DWARFContext &Context,
   return Error::success();
 }
 
-bool DWARFUnitHeader::applyIndexEntry(const DWARFUnitIndex::Entry *Entry) {
+Error DWARFUnitHeader::applyIndexEntry(const DWARFUnitIndex::Entry *Entry) {
   assert(Entry);
   assert(!IndexEntry);
   IndexEntry = Entry;
   if (AbbrOffset)
-    return false;
+    return createStringError(errc::invalid_argument,
+                             "DWARF package unit at offset 0x%8.8" PRIx64
+                             " has a non-zero abbreviation offset",
+                             Offset);
+
   auto *UnitContrib = IndexEntry->getContribution();
-  if (!UnitContrib ||
-      UnitContrib->getLength() != (getLength() + getUnitLengthFieldByteSize()))
-    return false;
+  if (!UnitContrib)
+    return createStringError(errc::invalid_argument,
+                             "DWARF package unit at offset 0x%8.8" PRIx64
+                             " has no contribution index",
+                             Offset);
+
+  uint64_t IndexLength = getLength() + getUnitLengthFieldByteSize();
+  if (UnitContrib->getLength() != IndexLength)
+    return createStringError(errc::invalid_argument,
+                             "DWARF package unit at offset 0x%8.8" PRIx64
+                             " has an inconsistent index (expected: %" PRIu64
+                             ", actual: %" PRIu64 ")",
+                             Offset, UnitContrib->getLength(), IndexLength);
+
   auto *AbbrEntry = IndexEntry->getContribution(DW_SECT_ABBREV);
   if (!AbbrEntry)
-    return false;
+    return createStringError(errc::invalid_argument,
+                             "DWARF package unit at offset 0x%8.8" PRIx64
+                             " missing abbreviation column",
+                             Offset);
+
   AbbrOffset = AbbrEntry->getOffset();
-  return true;
+  return Error::success();
 }
 
 Error DWARFUnit::extractRangeList(uint64_t RangeListOffset,
diff --git a/llvm/lib/ExecutionEngine/Orc/ExecutorProcessControl.cpp b/llvm/lib/ExecutionEngine/Orc/ExecutorProcessControl.cpp
index efafca949e61..0df7c4f25eb8 100644
--- a/llvm/lib/ExecutionEngine/Orc/ExecutorProcessControl.cpp
+++ b/llvm/lib/ExecutionEngine/Orc/ExecutorProcessControl.cpp
@@ -61,13 +61,8 @@ SelfExecutorProcessControl::Create(
   if (!SSP)
     SSP = std::make_shared<SymbolStringPool>();
 
-  if (!D) {
-#if LLVM_ENABLE_THREADS
-    D = std::make_unique<DynamicThreadPoolTaskDispatcher>();
-#else
+  if (!D)
     D = std::make_unique<InPlaceTaskDispatcher>();
-#endif
-  }
 
   auto PageSize = sys::Process::getPageSize();
   if (!PageSize)
diff --git a/llvm/lib/ExecutionEngine/Orc/LLJIT.cpp b/llvm/lib/ExecutionEngine/Orc/LLJIT.cpp
index 79adda5b7bc0..568b2ececaa0 100644
--- a/llvm/lib/ExecutionEngine/Orc/LLJIT.cpp
+++ b/llvm/lib/ExecutionEngine/Orc/LLJIT.cpp
@@ -667,6 +667,40 @@ Error LLJITBuilderState::prepareForConstruction() {
       return JTMBOrErr.takeError();
   }
 
+  if ((ES || EPC) && NumCompileThreads)
+    return make_error<StringError>(
+        "NumCompileThreads cannot be used with a custom ExecutionSession or "
+        "ExecutorProcessControl",
+        inconvertibleErrorCode());
+
+#if !LLVM_ENABLE_THREADS
+  if (NumCompileThreads)
+    return make_error<StringError>(
+        "LLJIT num-compile-threads is " + Twine(NumCompileThreads) +
+            " but LLVM was compiled with LLVM_ENABLE_THREADS=Off",
+        inconvertibleErrorCode());
+#endif // !LLVM_ENABLE_THREADS
+
+  // Only used in debug builds.
+  [[maybe_unused]] bool ConcurrentCompilationSettingDefaulted =
+      !SupportConcurrentCompilation;
+
+  if (!SupportConcurrentCompilation) {
+#if LLVM_ENABLE_THREADS
+    SupportConcurrentCompilation = NumCompileThreads || ES || EPC;
+#else
+    SupportConcurrentCompilation = false;
+#endif // LLVM_ENABLE_THREADS
+  } else {
+#if !LLVM_ENABLE_THREADS
+    if (*SupportConcurrentCompilation)
+      return make_error<StringError>(
+          "LLJIT concurrent compilation support requested, but LLVM was built "
+          "with LLVM_ENABLE_THREADS=Off",
+          inconvertibleErrorCode());
+#endif // !LLVM_ENABLE_THREADS
+  }
+
   LLVM_DEBUG({
     dbgs() << "  JITTargetMachineBuilder is "
            << JITTargetMachineBuilderPrinter(*JTMB, "  ")
@@ -684,11 +718,13 @@ Error LLJITBuilderState::prepareForConstruction() {
            << (CreateCompileFunction ? "Yes" : "No") << "\n"
            << "  Custom platform-setup function: "
            << (SetUpPlatform ? "Yes" : "No") << "\n"
-           << "  Number of compile threads: " << NumCompileThreads;
-    if (!NumCompileThreads)
-      dbgs() << " (code will be compiled on the execution thread)\n";
+           << "  Support concurrent compilation: "
+           << (*SupportConcurrentCompilation ? "Yes" : "No");
+    if (ConcurrentCompilationSettingDefaulted)
+      dbgs() << " (defaulted based on ES / EPC / NumCompileThreads)\n";
     else
       dbgs() << "\n";
+    dbgs() << "  Number of compile threads: " << NumCompileThreads << "\n";
   });
 
   // Create DL if not specified.
@@ -705,7 +741,19 @@ Error LLJITBuilderState::prepareForConstruction() {
       dbgs() << "ExecutorProcessControl not specified, "
                 "Creating SelfExecutorProcessControl instance\n";
     });
-    if (auto EPCOrErr = SelfExecutorProcessControl::Create())
+
+    std::unique_ptr<TaskDispatcher> D = nullptr;
+#if LLVM_ENABLE_THREADS
+    if (*SupportConcurrentCompilation) {
+      std::optional<size_t> NumThreads = std ::nullopt;
+      if (NumCompileThreads)
+        NumThreads = NumCompileThreads;
+      D = std::make_unique<DynamicThreadPoolTaskDispatcher>(NumThreads);
+    } else
+      D = std::make_unique<InPlaceTaskDispatcher>();
+#endif // LLVM_ENABLE_THREADS
+    if (auto EPCOrErr =
+            SelfExecutorProcessControl::Create(nullptr, std::move(D), nullptr))
       EPC = std::move(*EPCOrErr);
     else
       return EPCOrErr.takeError();
@@ -790,8 +838,6 @@ Error LLJITBuilderState::prepareForConstruction() {
 }
 
 LLJIT::~LLJIT() {
-  if (CompileThreads)
-    CompileThreads->wait();
   if (auto Err = ES->endSession())
     ES->reportError(std::move(Err));
 }
@@ -916,9 +962,8 @@ LLJIT::createCompileFunction(LLJITBuilderState &S,
   if (S.CreateCompileFunction)
     return S.CreateCompileFunction(std::move(JTMB));
 
-  // Otherwise default to creating a SimpleCompiler, or ConcurrentIRCompiler,
-  // depending on the number of threads requested.
-  if (S.NumCompileThreads > 0)
+  // If using a custom EPC then use a ConcurrentIRCompiler by default.
+  if (*S.SupportConcurrentCompilation)
     return std::make_unique<ConcurrentIRCompiler>(std::move(JTMB));
 
   auto TM = JTMB.createTargetMachine();
@@ -970,21 +1015,8 @@ LLJIT::LLJIT(LLJITBuilderState &S, Error &Err)
         std::make_unique<IRTransformLayer>(*ES, *TransformLayer);
   }
 
-  if (S.NumCompileThreads > 0) {
+  if (*S.SupportConcurrentCompilation)
     InitHelperTransformLayer->setCloneToNewContextOnEmit(true);
-    CompileThreads = std::make_unique<DefaultThreadPool>(
-        hardware_concurrency(S.NumCompileThreads));
-    ES->setDispatchTask([this](std::unique_ptr<Task> T) {
-      // FIXME: We should be able to use move-capture here, but ThreadPool's
-      // AsyncTaskTys are std::functions rather than unique_functions
-      // (because MSVC's std::packaged_tasks don't support move-only types).
-      // Fix this when all the above gets sorted out.
-      CompileThreads->async([UnownedT = T.release()]() mutable {
-        std::unique_ptr<Task> T(UnownedT);
-        T->run();
-      });
-    });
-  }
 
   if (S.SetupProcessSymbolsJITDylib) {
     if (auto ProcSymsJD = S.SetupProcessSymbolsJITDylib(*this)) {
@@ -1240,7 +1272,7 @@ LLLazyJIT::LLLazyJIT(LLLazyJITBuilderState &S, Error &Err) : LLJIT(S, Err) {
   CODLayer = std::make_unique<CompileOnDemandLayer>(
       *ES, *InitHelperTransformLayer, *LCTMgr, std::move(ISMBuilder));
 
-  if (S.NumCompileThreads > 0)
+  if (*S.SupportConcurrentCompilation)
     CODLayer->setCloneToNewContextOnEmit(true);
 }
 
diff --git a/llvm/lib/ExecutionEngine/Orc/MemoryMapper.cpp b/llvm/lib/ExecutionEngine/Orc/MemoryMapper.cpp
index 2c87b344083e..bba3329e8cc2 100644
--- a/llvm/lib/ExecutionEngine/Orc/MemoryMapper.cpp
+++ b/llvm/lib/ExecutionEngine/Orc/MemoryMapper.cpp
@@ -16,6 +16,10 @@
 #if defined(LLVM_ON_UNIX) && !defined(__ANDROID__)
 #include <fcntl.h>
 #include <sys/mman.h>
+#if defined(__MVS__)
+#include "llvm/Support/BLAKE3.h"
+#include <sys/shm.h>
+#endif
 #include <unistd.h>
 #elif defined(_WIN32)
 #include <windows.h>
@@ -239,6 +243,24 @@ void SharedMemoryMapper::reserve(size_t NumBytes,
 
 #if defined(LLVM_ON_UNIX)
 
+#if defined(__MVS__)
+        ArrayRef<uint8_t> Data(
+            reinterpret_cast<const uint8_t *>(SharedMemoryName.c_str()),
+            SharedMemoryName.size());
+        auto HashedName = BLAKE3::hash<sizeof(key_t)>(Data);
+        key_t Key = *reinterpret_cast<key_t *>(HashedName.data());
+        int SharedMemoryId =
+            shmget(Key, NumBytes, IPC_CREAT | __IPC_SHAREAS | 0700);
+        if (SharedMemoryId < 0) {
+          return OnReserved(errorCodeToError(
+              std::error_code(errno, std::generic_category())));
+        }
+        LocalAddr = shmat(SharedMemoryId, nullptr, 0);
+        if (LocalAddr == reinterpret_cast<void *>(-1)) {
+          return OnReserved(errorCodeToError(
+              std::error_code(errno, std::generic_category())));
+        }
+#else
         int SharedMemoryFile = shm_open(SharedMemoryName.c_str(), O_RDWR, 0700);
         if (SharedMemoryFile < 0) {
           return OnReserved(errorCodeToError(errnoAsErrorCode()));
@@ -254,6 +276,7 @@ void SharedMemoryMapper::reserve(size_t NumBytes,
         }
 
         close(SharedMemoryFile);
+#endif
 
 #elif defined(_WIN32)
 
@@ -373,8 +396,13 @@ void SharedMemoryMapper::release(ArrayRef<ExecutorAddr> Bases,
 
 #if defined(LLVM_ON_UNIX)
 
+#if defined(__MVS__)
+      if (shmdt(Reservations[Base].LocalAddr) < 0)
+        Err = joinErrors(std::move(Err), errorCodeToError(errnoAsErrorCode()));
+#else
       if (munmap(Reservations[Base].LocalAddr, Reservations[Base].Size) != 0)
         Err = joinErrors(std::move(Err), errorCodeToError(errnoAsErrorCode()));
+#endif
 
 #elif defined(_WIN32)
 
@@ -415,7 +443,11 @@ SharedMemoryMapper::~SharedMemoryMapper() {
 
 #if defined(LLVM_ON_UNIX) && !defined(__ANDROID__)
 
+#if defined(__MVS__)
+    shmdt(R.second.LocalAddr);
+#else
     munmap(R.second.LocalAddr, R.second.Size);
+#endif
 
 #elif defined(_WIN32)
 
diff --git a/llvm/lib/ExecutionEngine/Orc/TargetProcess/ExecutorSharedMemoryMapperService.cpp b/llvm/lib/ExecutionEngine/Orc/TargetProcess/ExecutorSharedMemoryMapperService.cpp
index 6614beec760f..f5118c0f2bfa 100644
--- a/llvm/lib/ExecutionEngine/Orc/TargetProcess/ExecutorSharedMemoryMapperService.cpp
+++ b/llvm/lib/ExecutionEngine/Orc/TargetProcess/ExecutorSharedMemoryMapperService.cpp
@@ -18,6 +18,10 @@
 #include <errno.h>
 #include <fcntl.h>
 #include <sys/mman.h>
+#if defined(__MVS__)
+#include "llvm/Support/BLAKE3.h"
+#include <sys/shm.h>
+#endif
 #include <unistd.h>
 #endif
 
@@ -59,6 +63,21 @@ ExecutorSharedMemoryMapperService::reserve(uint64_t Size) {
     SharedMemoryName = SharedMemoryNameStream.str();
   }
 
+#if defined(__MVS__)
+  ArrayRef<uint8_t> Data(
+      reinterpret_cast<const uint8_t *>(SharedMemoryName.c_str()),
+      SharedMemoryName.size());
+  auto HashedName = BLAKE3::hash<sizeof(key_t)>(Data);
+  key_t Key = *reinterpret_cast<key_t *>(HashedName.data());
+  int SharedMemoryId =
+      shmget(Key, Size, IPC_CREAT | IPC_EXCL | __IPC_SHAREAS | 0700);
+  if (SharedMemoryId < 0)
+    return errorCodeToError(errnoAsErrorCode());
+
+  void *Addr = shmat(SharedMemoryId, nullptr, 0);
+  if (Addr == reinterpret_cast<void *>(-1))
+    return errorCodeToError(errnoAsErrorCode());
+#else
   int SharedMemoryFile =
       shm_open(SharedMemoryName.c_str(), O_RDWR | O_CREAT | O_EXCL, 0700);
   if (SharedMemoryFile < 0)
@@ -73,6 +92,7 @@ ExecutorSharedMemoryMapperService::reserve(uint64_t Size) {
     return errorCodeToError(errnoAsErrorCode());
 
   close(SharedMemoryFile);
+#endif
 
 #elif defined(_WIN32)
 
@@ -131,6 +151,9 @@ Expected<ExecutorAddr> ExecutorSharedMemoryMapperService::initialize(
 
 #if defined(LLVM_ON_UNIX)
 
+#if defined(__MVS__)
+      // TODO Is it possible to change the protection level?
+#else
     int NativeProt = 0;
     if ((Segment.RAG.Prot & MemProt::Read) == MemProt::Read)
       NativeProt |= PROT_READ;
@@ -141,6 +164,7 @@ Expected<ExecutorAddr> ExecutorSharedMemoryMapperService::initialize(
 
     if (mprotect(Segment.Addr.toPtr<void *>(), Segment.Size, NativeProt))
       return errorCodeToError(errnoAsErrorCode());
+#endif
 
 #elif defined(_WIN32)
 
@@ -239,8 +263,15 @@ Error ExecutorSharedMemoryMapperService::release(
 
 #if defined(LLVM_ON_UNIX)
 
+#if defined(__MVS__)
+    (void)Size;
+
+    if (shmdt(Base.toPtr<void *>()) < 0)
+      Err = joinErrors(std::move(Err), errorCodeToError(errnoAsErrorCode()));
+#else
     if (munmap(Base.toPtr<void *>(), Size) != 0)
       Err = joinErrors(std::move(Err), errorCodeToError(errnoAsErrorCode()));
+#endif
 
 #elif defined(_WIN32)
     (void)Size;
diff --git a/llvm/lib/ExecutionEngine/Orc/TaskDispatch.cpp b/llvm/lib/ExecutionEngine/Orc/TaskDispatch.cpp
index 11a99986f2ee..4ac2a4209185 100644
--- a/llvm/lib/ExecutionEngine/Orc/TaskDispatch.cpp
+++ b/llvm/lib/ExecutionEngine/Orc/TaskDispatch.cpp
@@ -7,6 +7,7 @@
 //===----------------------------------------------------------------------===//
 
 #include "llvm/ExecutionEngine/Orc/TaskDispatch.h"
+#include "llvm/ExecutionEngine/Orc/Core.h"
 
 namespace llvm {
 namespace orc {
@@ -24,16 +25,52 @@ void InPlaceTaskDispatcher::shutdown() {}
 
 #if LLVM_ENABLE_THREADS
 void DynamicThreadPoolTaskDispatcher::dispatch(std::unique_ptr<Task> T) {
+  bool IsMaterializationTask = isa<MaterializationTask>(*T);
+
   {
     std::lock_guard<std::mutex> Lock(DispatchMutex);
+
+    if (IsMaterializationTask) {
+
+      // If this is a materialization task and there are too many running
+      // already then queue this one up and return early.
+      if (MaxMaterializationThreads &&
+          NumMaterializationThreads == *MaxMaterializationThreads) {
+        MaterializationTaskQueue.push_back(std::move(T));
+        return;
+      }
+
+      // Otherwise record that we have a materialization task running.
+      ++NumMaterializationThreads;
+    }
+
     ++Outstanding;
   }
 
-  std::thread([this, T = std::move(T)]() mutable {
-    T->run();
-    std::lock_guard<std::mutex> Lock(DispatchMutex);
-    --Outstanding;
-    OutstandingCV.notify_all();
+  std::thread([this, T = std::move(T), IsMaterializationTask]() mutable {
+    while (true) {
+
+      // Run the task.
+      T->run();
+
+      std::lock_guard<std::mutex> Lock(DispatchMutex);
+      if (!MaterializationTaskQueue.empty()) {
+        // If there are any materialization tasks running then steal that work.
+        T = std::move(MaterializationTaskQueue.front());
+        MaterializationTaskQueue.pop_front();
+        if (!IsMaterializationTask) {
+          ++NumMaterializationThreads;
+          IsMaterializationTask = true;
+        }
+      } else {
+        // Otherwise decrement work counters.
+        if (IsMaterializationTask)
+          --NumMaterializationThreads;
+        --Outstanding;
+        OutstandingCV.notify_all();
+        return;
+      }
+    }
   }).detach();
 }
 
diff --git a/llvm/lib/Frontend/OpenMP/OMP.cpp b/llvm/lib/Frontend/OpenMP/OMP.cpp
index e097510592c9..c1556ff3c74d 100644
--- a/llvm/lib/Frontend/OpenMP/OMP.cpp
+++ b/llvm/lib/Frontend/OpenMP/OMP.cpp
@@ -25,6 +25,54 @@ using namespace llvm::omp;
 #define GEN_DIRECTIVES_IMPL
 #include "llvm/Frontend/OpenMP/OMP.inc"
 
+static iterator_range<ArrayRef<Directive>::iterator>
+getFirstCompositeRange(iterator_range<ArrayRef<Directive>::iterator> Leafs) {
+  // OpenMP Spec 5.2: [17.3, 8-9]
+  // If directive-name-A and directive-name-B both correspond to loop-
+  // associated constructs then directive-name is a composite construct
+  // otherwise directive-name is a combined construct.
+  //
+  // In the list of leaf constructs, find the first loop-associated construct,
+  // this is the beginning of the returned range. Then, starting from the
+  // immediately following leaf construct, find the first sequence of adjacent
+  // loop-associated constructs. The last of those is the last one of the
+  // range, that is, the end of the range is one past that element.
+  // If such a sequence of adjacent loop-associated directives does not exist,
+  // return an empty range.
+  //
+  // The end of the returned range (including empty range) is intended to be
+  // a point from which the search for the next range could resume.
+  //
+  // Consequently, this function can't return a range with a single leaf
+  // construct in it.
+
+  auto firstLoopAssociated =
+      [](iterator_range<ArrayRef<Directive>::iterator> List) {
+        for (auto It = List.begin(), End = List.end(); It != End; ++It) {
+          if (getDirectiveAssociation(*It) == Association::Loop)
+            return It;
+        }
+        return List.end();
+      };
+
+  auto Empty = llvm::make_range(Leafs.end(), Leafs.end());
+
+  auto Begin = firstLoopAssociated(Leafs);
+  if (Begin == Leafs.end())
+    return Empty;
+
+  auto End =
+      firstLoopAssociated(llvm::make_range(std::next(Begin), Leafs.end()));
+  if (End == Leafs.end())
+    return Empty;
+
+  for (; End != Leafs.end(); ++End) {
+    if (getDirectiveAssociation(*End) != Association::Loop)
+      break;
+  }
+  return llvm::make_range(Begin, End);
+}
+
 namespace llvm::omp {
 ArrayRef<Directive> getLeafConstructs(Directive D) {
   auto Idx = static_cast<std::size_t>(D);
@@ -34,6 +82,44 @@ ArrayRef<Directive> getLeafConstructs(Directive D) {
   return ArrayRef(&Row[2], static_cast<int>(Row[1]));
 }
 
+ArrayRef<Directive> getLeafConstructsOrSelf(Directive D) {
+  if (auto Leafs = getLeafConstructs(D); !Leafs.empty())
+    return Leafs;
+  auto Idx = static_cast<size_t>(D);
+  assert(Idx < Directive_enumSize && "Invalid directive");
+  const auto *Row = LeafConstructTable[LeafConstructTableOrdering[Idx]];
+  // The first entry in the row is the directive itself.
+  return ArrayRef(&Row[0], &Row[0] + 1);
+}
+
+ArrayRef<Directive>
+getLeafOrCompositeConstructs(Directive D, SmallVectorImpl<Directive> &Output) {
+  using ArrayTy = ArrayRef<Directive>;
+  using IteratorTy = ArrayTy::iterator;
+  ArrayRef<Directive> Leafs = getLeafConstructsOrSelf(D);
+
+  IteratorTy Iter = Leafs.begin();
+  do {
+    auto Range = getFirstCompositeRange(llvm::make_range(Iter, Leafs.end()));
+    // All directives before the range are leaf constructs.
+    for (; Iter != Range.begin(); ++Iter)
+      Output.push_back(*Iter);
+    if (!Range.empty()) {
+      Directive Comp =
+          getCompoundConstruct(ArrayTy(Range.begin(), Range.end()));
+      assert(Comp != OMPD_unknown);
+      Output.push_back(Comp);
+      Iter = Range.end();
+      // As of now, a composite construct must contain all constituent leaf
+      // constructs from some point until the end of all constituent leaf
+      // constructs.
+      assert(Iter == Leafs.end() && "Malformed directive");
+    }
+  } while (Iter != Leafs.end());
+
+  return Output;
+}
+
 Directive getCompoundConstruct(ArrayRef<Directive> Parts) {
   if (Parts.empty())
     return OMPD_unknown;
@@ -84,4 +170,20 @@ Directive getCompoundConstruct(ArrayRef<Directive> Parts) {
     return Found;
   return OMPD_unknown;
 }
+
+bool isLeafConstruct(Directive D) { return getLeafConstructs(D).empty(); }
+
+bool isCompositeConstruct(Directive D) {
+  ArrayRef<Directive> Leafs = getLeafConstructsOrSelf(D);
+  if (Leafs.size() <= 1)
+    return false;
+  auto Range = getFirstCompositeRange(Leafs);
+  return Range.begin() == Leafs.begin() && Range.end() == Leafs.end();
+}
+
+bool isCombinedConstruct(Directive D) {
+  // OpenMP Spec 5.2: [17.3, 9-10]
+  // Otherwise directive-name is a combined construct.
+  return !getLeafConstructs(D).empty() && !isCompositeConstruct(D);
+}
 } // namespace llvm::omp
diff --git a/llvm/lib/IR/Attributes.cpp b/llvm/lib/IR/Attributes.cpp
index b2d9992cdc02..9c48a481de1f 100644
--- a/llvm/lib/IR/Attributes.cpp
+++ b/llvm/lib/IR/Attributes.cpp
@@ -1530,6 +1530,13 @@ AttributeList::addDereferenceableOrNullParamAttr(LLVMContext &C, unsigned Index,
   return addParamAttributes(C, Index, B);
 }
 
+AttributeList AttributeList::addRangeRetAttr(LLVMContext &C,
+                                             const ConstantRange &CR) const {
+  AttrBuilder B(C);
+  B.addRangeAttr(CR);
+  return addRetAttributes(C, B);
+}
+
 AttributeList AttributeList::addAllocSizeParamAttr(
     LLVMContext &C, unsigned Index, unsigned ElemSizeArg,
     const std::optional<unsigned> &NumElemsArg) {
diff --git a/llvm/lib/IR/AutoUpgrade.cpp b/llvm/lib/IR/AutoUpgrade.cpp
index 634b2dd5119e..5b02b0e94dda 100644
--- a/llvm/lib/IR/AutoUpgrade.cpp
+++ b/llvm/lib/IR/AutoUpgrade.cpp
@@ -846,6 +846,18 @@ static bool upgradeArmOrAarch64IntrinsicFunction(bool IsArm, Function *F,
         return false; // No other 'aarch64.sve.bf*'.
       }
 
+      if (Name.consume_front("addqv")) {
+        // 'aarch64.sve.addqv'.
+        if (!F->getReturnType()->isFPOrFPVectorTy())
+          return false;
+
+        auto Args = F->getFunctionType()->params();
+        Type *Tys[] = {F->getReturnType(), Args[1]};
+        NewFn = Intrinsic::getDeclaration(F->getParent(),
+                                          Intrinsic::aarch64_sve_faddqv, Tys);
+        return true;
+      }
+
       if (Name.consume_front("ld")) {
         // 'aarch64.sve.ld*'.
         static const Regex LdRegex("^[234](.nxv[a-z0-9]+|$)");
@@ -1080,17 +1092,24 @@ static bool upgradeIntrinsicFunction1(Function *F, Function *&NewFn,
     break;
   case 'e':
     if (Name.consume_front("experimental.vector.")) {
-      Intrinsic::ID ID = StringSwitch<Intrinsic::ID>(Name)
-                             .StartsWith("extract.", Intrinsic::vector_extract)
-                             .StartsWith("insert.", Intrinsic::vector_insert)
-                             .Default(Intrinsic::not_intrinsic);
+      Intrinsic::ID ID =
+          StringSwitch<Intrinsic::ID>(Name)
+              .StartsWith("extract.", Intrinsic::vector_extract)
+              .StartsWith("insert.", Intrinsic::vector_insert)
+              .StartsWith("splice.", Intrinsic::vector_splice)
+              .StartsWith("reverse.", Intrinsic::vector_reverse)
+              .StartsWith("interleave2.", Intrinsic::vector_interleave2)
+              .StartsWith("deinterleave2.", Intrinsic::vector_deinterleave2)
+              .Default(Intrinsic::not_intrinsic);
       if (ID != Intrinsic::not_intrinsic) {
         const auto *FT = F->getFunctionType();
         SmallVector<Type *, 2> Tys;
-        if (ID == Intrinsic::vector_extract)
+        if (ID == Intrinsic::vector_extract ||
+            ID == Intrinsic::vector_interleave2)
           // Extracting overloads the return type.
           Tys.push_back(FT->getReturnType());
-        Tys.push_back(FT->getParamType(0));
+        if (ID != Intrinsic::vector_interleave2)
+          Tys.push_back(FT->getParamType(0));
         if (ID == Intrinsic::vector_insert)
           // Inserting overloads the inserted type.
           Tys.push_back(FT->getParamType(1));
diff --git a/llvm/lib/IR/CMakeLists.txt b/llvm/lib/IR/CMakeLists.txt
index f1668ee3be63..b5fb7409d8e8 100644
--- a/llvm/lib/IR/CMakeLists.txt
+++ b/llvm/lib/IR/CMakeLists.txt
@@ -41,6 +41,7 @@ add_llvm_component_library(LLVMCore
   LLVMRemarkStreamer.cpp
   LegacyPassManager.cpp
   MDBuilder.cpp
+  MemoryModelRelaxationAnnotations.cpp
   Mangler.cpp
   Metadata.cpp
   Module.cpp
diff --git a/llvm/lib/IR/EHPersonalities.cpp b/llvm/lib/IR/EHPersonalities.cpp
index fb5e2d5c517e..7c32601b8a83 100644
--- a/llvm/lib/IR/EHPersonalities.cpp
+++ b/llvm/lib/IR/EHPersonalities.cpp
@@ -42,6 +42,7 @@ EHPersonality llvm::classifyEHPersonality(const Value *Pers) {
       .Case("rust_eh_personality", EHPersonality::Rust)
       .Case("__gxx_wasm_personality_v0", EHPersonality::Wasm_CXX)
       .Case("__xlcxx_personality_v1", EHPersonality::XL_CXX)
+      .Case("__zos_cxx_personality_v2", EHPersonality::ZOS_CXX)
       .Default(EHPersonality::Unknown);
 }
 
@@ -73,6 +74,8 @@ StringRef llvm::getEHPersonalityName(EHPersonality Pers) {
     return "__gxx_wasm_personality_v0";
   case EHPersonality::XL_CXX:
     return "__xlcxx_personality_v1";
+  case EHPersonality::ZOS_CXX:
+    return "__zos_cxx_personality_v2";
   case EHPersonality::Unknown:
     llvm_unreachable("Unknown EHPersonality!");
   }
diff --git a/llvm/lib/IR/Function.cpp b/llvm/lib/IR/Function.cpp
index e66fe73425e8..545940dd86f9 100644
--- a/llvm/lib/IR/Function.cpp
+++ b/llvm/lib/IR/Function.cpp
@@ -735,6 +735,10 @@ void Function::addDereferenceableOrNullParamAttr(unsigned ArgNo,
                                                                   ArgNo, Bytes);
 }
 
+void Function::addRangeRetAttr(const ConstantRange &CR) {
+  AttributeSets = AttributeSets.addRangeRetAttr(getContext(), CR);
+}
+
 DenormalMode Function::getDenormalMode(const fltSemantics &FPType) const {
   if (&FPType == &APFloat::IEEEsingle()) {
     DenormalMode Mode = getDenormalModeF32Raw();
diff --git a/llvm/lib/IR/IRBuilder.cpp b/llvm/lib/IR/IRBuilder.cpp
index d6746d1d4382..9ec5a7deeec6 100644
--- a/llvm/lib/IR/IRBuilder.cpp
+++ b/llvm/lib/IR/IRBuilder.cpp
@@ -1171,8 +1171,7 @@ Value *IRBuilderBase::CreateVectorReverse(Value *V, const Twine &Name) {
   auto *Ty = cast<VectorType>(V->getType());
   if (isa<ScalableVectorType>(Ty)) {
     Module *M = BB->getParent()->getParent();
-    Function *F = Intrinsic::getDeclaration(
-        M, Intrinsic::experimental_vector_reverse, Ty);
+    Function *F = Intrinsic::getDeclaration(M, Intrinsic::vector_reverse, Ty);
     return Insert(CallInst::Create(F, V), Name);
   }
   // Keep the original behaviour for fixed vector
@@ -1191,8 +1190,7 @@ Value *IRBuilderBase::CreateVectorSplice(Value *V1, Value *V2, int64_t Imm,
 
   if (auto *VTy = dyn_cast<ScalableVectorType>(V1->getType())) {
     Module *M = BB->getParent()->getParent();
-    Function *F = Intrinsic::getDeclaration(
-        M, Intrinsic::experimental_vector_splice, VTy);
+    Function *F = Intrinsic::getDeclaration(M, Intrinsic::vector_splice, VTy);
 
     Value *Ops[] = {V1, V2, getInt32(Imm)};
     return Insert(CallInst::Create(F, Ops), Name);
diff --git a/llvm/lib/IR/Instruction.cpp b/llvm/lib/IR/Instruction.cpp
index eb1c5f445eb8..678edc58ad84 100644
--- a/llvm/lib/IR/Instruction.cpp
+++ b/llvm/lib/IR/Instruction.cpp
@@ -19,6 +19,7 @@
 #include "llvm/IR/Instructions.h"
 #include "llvm/IR/IntrinsicInst.h"
 #include "llvm/IR/Intrinsics.h"
+#include "llvm/IR/MemoryModelRelaxationAnnotations.h"
 #include "llvm/IR/Operator.h"
 #include "llvm/IR/ProfDataUtils.h"
 #include "llvm/IR/Type.h"
diff --git a/llvm/lib/IR/Instructions.cpp b/llvm/lib/IR/Instructions.cpp
index d2babc748731..7ad1ad4cddb7 100644
--- a/llvm/lib/IR/Instructions.cpp
+++ b/llvm/lib/IR/Instructions.cpp
@@ -2889,7 +2889,7 @@ bool ShuffleVectorInst::isOneUseSingleSourceMask(int VF) const {
 bool ShuffleVectorInst::isInterleave(unsigned Factor) {
   FixedVectorType *OpTy = dyn_cast<FixedVectorType>(getOperand(0)->getType());
   // shuffle_vector can only interleave fixed length vectors - for scalable
-  // vectors, see the @llvm.experimental.vector.interleave2 intrinsic
+  // vectors, see the @llvm.vector.interleave2 intrinsic
   if (!OpTy)
     return false;
   unsigned OpNumElts = OpTy->getNumElements();
diff --git a/llvm/lib/IR/IntrinsicInst.cpp b/llvm/lib/IR/IntrinsicInst.cpp
index 89403e1d7fcb..6b6420ae41c9 100644
--- a/llvm/lib/IR/IntrinsicInst.cpp
+++ b/llvm/lib/IR/IntrinsicInst.cpp
@@ -119,7 +119,8 @@ static ValueAsMetadata *getAsMetadata(Value *V) {
 }
 
 void DbgVariableIntrinsic::replaceVariableLocationOp(Value *OldValue,
-                                                     Value *NewValue) {
+                                                     Value *NewValue,
+                                                     bool AllowEmpty) {
   // If OldValue is used as the address part of a dbg.assign intrinsic replace
   // it with NewValue and return true.
   auto ReplaceDbgAssignAddress = [this, OldValue, NewValue]() -> bool {
@@ -136,6 +137,8 @@ void DbgVariableIntrinsic::replaceVariableLocationOp(Value *OldValue,
   auto Locations = location_ops();
   auto OldIt = find(Locations, OldValue);
   if (OldIt == Locations.end()) {
+    if (AllowEmpty || DbgAssignAddrReplaced)
+      return;
     assert(DbgAssignAddrReplaced &&
            "OldValue must be dbg.assign addr if unused in DIArgList");
     return;
@@ -291,6 +294,12 @@ Value *InstrProfIncrementInst::getStep() const {
   return ConstantInt::get(Type::getInt64Ty(Context), 1);
 }
 
+Value *InstrProfCallsite::getCallee() const {
+  if (isa<InstrProfCallsite>(this))
+    return getArgOperand(4);
+  return nullptr;
+}
+
 std::optional<RoundingMode> ConstrainedFPIntrinsic::getRoundingMode() const {
   unsigned NumOperands = arg_size();
   Metadata *MD = nullptr;
diff --git a/llvm/lib/IR/MemoryModelRelaxationAnnotations.cpp b/llvm/lib/IR/MemoryModelRelaxationAnnotations.cpp
new file mode 100644
index 000000000000..19f438d890ae
--- /dev/null
+++ b/llvm/lib/IR/MemoryModelRelaxationAnnotations.cpp
@@ -0,0 +1,170 @@
+//===- MemoryModelRelaxationAnnotations.cpp ---------------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#include "llvm/IR/MemoryModelRelaxationAnnotations.h"
+#include "llvm/ADT/StringSet.h"
+#include "llvm/IR/Instructions.h"
+#include "llvm/IR/Metadata.h"
+#include "llvm/Support/Debug.h"
+#include "llvm/Support/raw_ostream.h"
+
+using namespace llvm;
+
+//===- MMRAMetadata -------------------------------------------------------===//
+
+MMRAMetadata::MMRAMetadata(const Instruction &I)
+    : MMRAMetadata(I.getMetadata(LLVMContext::MD_mmra)) {}
+
+MMRAMetadata::MMRAMetadata(MDNode *MD) {
+  if (!MD)
+    return;
+
+  // TODO: Split this into a "tryParse" function that can return an err.
+  // CTor can use the tryParse & just fatal on err.
+
+  MDTuple *Tuple = dyn_cast<MDTuple>(MD);
+  assert(Tuple && "Invalid MMRA structure");
+
+  const auto HandleTagMD = [this](MDNode *TagMD) {
+    Tags.insert({cast<MDString>(TagMD->getOperand(0))->getString(),
+                 cast<MDString>(TagMD->getOperand(1))->getString()});
+  };
+
+  if (isTagMD(Tuple)) {
+    HandleTagMD(Tuple);
+    return;
+  }
+
+  for (const MDOperand &Op : Tuple->operands()) {
+    MDNode *MDOp = cast<MDNode>(Op.get());
+    assert(isTagMD(MDOp));
+    HandleTagMD(MDOp);
+  }
+}
+
+bool MMRAMetadata::isTagMD(const Metadata *MD) {
+  if (auto *Tuple = dyn_cast<MDTuple>(MD)) {
+    return Tuple->getNumOperands() == 2 &&
+           isa<MDString>(Tuple->getOperand(0)) &&
+           isa<MDString>(Tuple->getOperand(1));
+  }
+  return false;
+}
+
+MDTuple *MMRAMetadata::getTagMD(LLVMContext &Ctx, StringRef Prefix,
+                                StringRef Suffix) {
+  return MDTuple::get(Ctx,
+                      {MDString::get(Ctx, Prefix), MDString::get(Ctx, Suffix)});
+}
+
+MDTuple *MMRAMetadata::getMD(LLVMContext &Ctx,
+                             ArrayRef<MMRAMetadata::TagT> Tags) {
+  if (Tags.empty())
+    return nullptr;
+
+  if (Tags.size() == 1)
+    return getTagMD(Ctx, Tags.front());
+
+  SmallVector<Metadata *> MMRAs;
+  for (const auto &Tag : Tags)
+    MMRAs.push_back(getTagMD(Ctx, Tag));
+  return MDTuple::get(Ctx, MMRAs);
+}
+
+MDNode *MMRAMetadata::combine(LLVMContext &Ctx, const MMRAMetadata &A,
+                              const MMRAMetadata &B) {
+  // Let A and B be two tags set, and U be the prefix-wise union of A and B.
+  // For every unique tag prefix P present in A or B:
+  // * If either A or B has no tags with prefix P, no tags with prefix
+  //   P are added to U.
+  // * If both A and B have at least one tag with prefix P, all tags with prefix
+  //   P from both sets are added to U.
+
+  SmallVector<Metadata *> Result;
+
+  for (const auto &[P, S] : A) {
+    if (B.hasTagWithPrefix(P))
+      Result.push_back(getTagMD(Ctx, P, S));
+  }
+  for (const auto &[P, S] : B) {
+    if (A.hasTagWithPrefix(P))
+      Result.push_back(getTagMD(Ctx, P, S));
+  }
+
+  return MDTuple::get(Ctx, Result);
+}
+
+bool MMRAMetadata::hasTag(StringRef Prefix, StringRef Suffix) const {
+  return Tags.count({Prefix, Suffix});
+}
+
+bool MMRAMetadata::isCompatibleWith(const MMRAMetadata &Other) const {
+  // Two sets of tags are compatible iff, for every unique tag prefix P
+  // present in at least one set:
+  //   - the other set contains no tag with prefix P, or
+  //   - at least one tag with prefix P is common to both sets.
+
+  StringMap<bool> PrefixStatuses;
+  for (const auto &[P, S] : Tags)
+    PrefixStatuses[P] |= (Other.hasTag(P, S) || !Other.hasTagWithPrefix(P));
+  for (const auto &[P, S] : Other)
+    PrefixStatuses[P] |= (hasTag(P, S) || !hasTagWithPrefix(P));
+
+  for (auto &[Prefix, Status] : PrefixStatuses) {
+    if (!Status)
+      return false;
+  }
+
+  return true;
+}
+
+bool MMRAMetadata::hasTagWithPrefix(StringRef Prefix) const {
+  for (const auto &[P, S] : Tags)
+    if (P == Prefix)
+      return true;
+  return false;
+}
+
+MMRAMetadata::const_iterator MMRAMetadata::begin() const {
+  return Tags.begin();
+}
+
+MMRAMetadata::const_iterator MMRAMetadata::end() const { return Tags.end(); }
+
+bool MMRAMetadata::empty() const { return Tags.empty(); }
+
+unsigned MMRAMetadata::size() const { return Tags.size(); }
+
+void MMRAMetadata::print(raw_ostream &OS) const {
+  bool IsFirst = true;
+  // TODO: use map_iter + join
+  for (const auto &[P, S] : Tags) {
+    if (IsFirst)
+      IsFirst = false;
+    else
+      OS << ", ";
+    OS << P << ":" << S;
+  }
+}
+
+LLVM_DUMP_METHOD
+void MMRAMetadata::dump() const { print(dbgs()); }
+
+//===- Helpers ------------------------------------------------------------===//
+
+static bool isReadWriteMemCall(const Instruction &I) {
+  if (const auto *C = dyn_cast<CallBase>(&I))
+    return C->mayReadOrWriteMemory() ||
+           !C->getMemoryEffects().doesNotAccessMemory();
+  return false;
+}
+
+bool llvm::canInstructionHaveMMRAs(const Instruction &I) {
+  return isa<LoadInst>(I) || isa<StoreInst>(I) || isa<AtomicCmpXchgInst>(I) ||
+         isa<AtomicRMWInst>(I) || isa<FenceInst>(I) || isReadWriteMemCall(I);
+}
diff --git a/llvm/lib/IR/Verifier.cpp b/llvm/lib/IR/Verifier.cpp
index edad3be9a3e0..430e2ce89f6a 100644
--- a/llvm/lib/IR/Verifier.cpp
+++ b/llvm/lib/IR/Verifier.cpp
@@ -99,6 +99,7 @@
 #include "llvm/IR/IntrinsicsNVPTX.h"
 #include "llvm/IR/IntrinsicsWebAssembly.h"
 #include "llvm/IR/LLVMContext.h"
+#include "llvm/IR/MemoryModelRelaxationAnnotations.h"
 #include "llvm/IR/Metadata.h"
 #include "llvm/IR/Module.h"
 #include "llvm/IR/ModuleSlotTracker.h"
@@ -116,6 +117,7 @@
 #include "llvm/Support/CommandLine.h"
 #include "llvm/Support/ErrorHandling.h"
 #include "llvm/Support/MathExtras.h"
+#include "llvm/Support/ModRef.h"
 #include "llvm/Support/raw_ostream.h"
 #include <algorithm>
 #include <cassert>
@@ -529,6 +531,7 @@ private:
   void visitMemProfMetadata(Instruction &I, MDNode *MD);
   void visitCallsiteMetadata(Instruction &I, MDNode *MD);
   void visitDIAssignIDMetadata(Instruction &I, MDNode *MD);
+  void visitMMRAMetadata(Instruction &I, MDNode *MD);
   void visitAnnotationMetadata(MDNode *Annotation);
   void visitAliasScopeMetadata(const MDNode *MD);
   void visitAliasScopeListMetadata(const MDNode *MD);
@@ -4844,6 +4847,24 @@ void Verifier::visitDIAssignIDMetadata(Instruction &I, MDNode *MD) {
   }
 }
 
+void Verifier::visitMMRAMetadata(Instruction &I, MDNode *MD) {
+  Check(canInstructionHaveMMRAs(I),
+        "!mmra metadata attached to unexpected instruction kind", I, MD);
+
+  // MMRA Metadata should either be a tag, e.g. !{!"foo", !"bar"}, or a
+  // list of tags such as !2 in the following example:
+  //    !0 = !{!"a", !"b"}
+  //    !1 = !{!"c", !"d"}
+  //    !2 = !{!0, !1}
+  if (MMRAMetadata::isTagMD(MD))
+    return;
+
+  Check(isa<MDTuple>(MD), "!mmra expected to be a metadata tuple", I, MD);
+  for (const MDOperand &MDOp : MD->operands())
+    Check(MMRAMetadata::isTagMD(MDOp.get()),
+          "!mmra metadata tuple operand is not an MMRA tag", I, MDOp.get());
+}
+
 void Verifier::visitCallStackMetadata(MDNode *MD) {
   // Call stack metadata should consist of a list of at least 1 constant int
   // (representing a hash of the location).
@@ -5161,6 +5182,9 @@ void Verifier::visitInstruction(Instruction &I) {
   if (MDNode *MD = I.getMetadata(LLVMContext::MD_DIAssignID))
     visitDIAssignIDMetadata(I, MD);
 
+  if (MDNode *MMRA = I.getMetadata(LLVMContext::MD_mmra))
+    visitMMRAMetadata(I, MMRA);
+
   if (MDNode *Annotation = I.getMetadata(LLVMContext::MD_annotation))
     visitAnnotationMetadata(Annotation);
 
@@ -5995,7 +6019,7 @@ void Verifier::visitIntrinsicCall(Intrinsic::ID ID, CallBase &Call) {
 
     break;
   }
-  case Intrinsic::experimental_vector_splice: {
+  case Intrinsic::vector_splice: {
     VectorType *VecTy = cast<VectorType>(Call.getType());
     int64_t Idx = cast<ConstantInt>(Call.getArgOperand(2))->getSExtValue();
     int64_t KnownMinNumElements = VecTy->getElementCount().getKnownMinValue();
@@ -6202,7 +6226,6 @@ void Verifier::visitIntrinsicCall(Intrinsic::ID ID, CallBase &Call) {
     break;
   }
   case Intrinsic::experimental_convergence_entry:
-    LLVM_FALLTHROUGH;
   case Intrinsic::experimental_convergence_anchor:
     break;
   case Intrinsic::experimental_convergence_loop:
diff --git a/llvm/lib/LTO/LTO.cpp b/llvm/lib/LTO/LTO.cpp
index 53060df7f503..21cad1de0ced 100644
--- a/llvm/lib/LTO/LTO.cpp
+++ b/llvm/lib/LTO/LTO.cpp
@@ -114,12 +114,12 @@ void llvm::computeLTOCacheKey(
   auto AddUnsigned = [&](unsigned I) {
     uint8_t Data[4];
     support::endian::write32le(Data, I);
-    Hasher.update(ArrayRef<uint8_t>{Data, 4});
+    Hasher.update(Data);
   };
   auto AddUint64 = [&](uint64_t I) {
     uint8_t Data[8];
     support::endian::write64le(Data, I);
-    Hasher.update(ArrayRef<uint8_t>{Data, 8});
+    Hasher.update(Data);
   };
   AddString(Conf.CPU);
   // FIXME: Hash more of Options. For now all clients initialize Options from
diff --git a/llvm/lib/MC/ELFObjectWriter.cpp b/llvm/lib/MC/ELFObjectWriter.cpp
index 005521bad6e0..b8ef2654ed6e 100644
--- a/llvm/lib/MC/ELFObjectWriter.cpp
+++ b/llvm/lib/MC/ELFObjectWriter.cpp
@@ -725,7 +725,13 @@ void ELFWriter::computeSymbolTable(
         HasLargeSectionIndex = true;
     }
 
+    // Temporary symbols generated for certain assembler features (.eh_frame,
+    // .debug_line) of an empty name may be referenced by relocations due to
+    // linker relaxation. Rename them to ".L0 " to match the gas fake label name
+    // and allow ld/objcopy --discard-locals to discard such symbols.
     StringRef Name = Symbol.getName();
+    if (Name.empty())
+      Name = ".L0 ";
 
     // Sections have their own string table
     if (Symbol.getType() != ELF::STT_SECTION) {
diff --git a/llvm/lib/MC/MCDXContainerStreamer.cpp b/llvm/lib/MC/MCDXContainerStreamer.cpp
index 3cb452f3dfa5..a596c9a16d49 100644
--- a/llvm/lib/MC/MCDXContainerStreamer.cpp
+++ b/llvm/lib/MC/MCDXContainerStreamer.cpp
@@ -21,11 +21,8 @@ void MCDXContainerStreamer::emitInstToData(const MCInst &,
 
 MCStreamer *llvm::createDXContainerStreamer(
     MCContext &Context, std::unique_ptr<MCAsmBackend> &&MAB,
-    std::unique_ptr<MCObjectWriter> &&OW, std::unique_ptr<MCCodeEmitter> &&CE,
-    bool RelaxAll) {
+    std::unique_ptr<MCObjectWriter> &&OW, std::unique_ptr<MCCodeEmitter> &&CE) {
   auto *S = new MCDXContainerStreamer(Context, std::move(MAB), std::move(OW),
                                       std::move(CE));
-  if (RelaxAll)
-    S->getAssembler().setRelaxAll(true);
   return S;
 }
diff --git a/llvm/lib/MC/MCELFStreamer.cpp b/llvm/lib/MC/MCELFStreamer.cpp
index e541090769e9..23e926c3a9d1 100644
--- a/llvm/lib/MC/MCELFStreamer.cpp
+++ b/llvm/lib/MC/MCELFStreamer.cpp
@@ -892,11 +892,8 @@ void MCELFStreamer::createAttributesSection(
 MCStreamer *llvm::createELFStreamer(MCContext &Context,
                                     std::unique_ptr<MCAsmBackend> &&MAB,
                                     std::unique_ptr<MCObjectWriter> &&OW,
-                                    std::unique_ptr<MCCodeEmitter> &&CE,
-                                    bool RelaxAll) {
+                                    std::unique_ptr<MCCodeEmitter> &&CE) {
   MCELFStreamer *S =
       new MCELFStreamer(Context, std::move(MAB), std::move(OW), std::move(CE));
-  if (RelaxAll)
-    S->getAssembler().setRelaxAll(true);
   return S;
 }
diff --git a/llvm/lib/MC/MCGOFFStreamer.cpp b/llvm/lib/MC/MCGOFFStreamer.cpp
index 58d13c9f3788..2b6d5c8e75a7 100644
--- a/llvm/lib/MC/MCGOFFStreamer.cpp
+++ b/llvm/lib/MC/MCGOFFStreamer.cpp
@@ -24,11 +24,8 @@ MCGOFFStreamer::~MCGOFFStreamer() {}
 MCStreamer *llvm::createGOFFStreamer(MCContext &Context,
                                      std::unique_ptr<MCAsmBackend> &&MAB,
                                      std::unique_ptr<MCObjectWriter> &&OW,
-                                     std::unique_ptr<MCCodeEmitter> &&CE,
-                                     bool RelaxAll) {
+                                     std::unique_ptr<MCCodeEmitter> &&CE) {
   MCGOFFStreamer *S =
       new MCGOFFStreamer(Context, std::move(MAB), std::move(OW), std::move(CE));
-  if (RelaxAll)
-    S->getAssembler().setRelaxAll(true);
   return S;
 }
diff --git a/llvm/lib/MC/MCMachOStreamer.cpp b/llvm/lib/MC/MCMachOStreamer.cpp
index d7d343f15eaa..10f9988b9d16 100644
--- a/llvm/lib/MC/MCMachOStreamer.cpp
+++ b/llvm/lib/MC/MCMachOStreamer.cpp
@@ -564,7 +564,7 @@ MCStreamer *llvm::createMachOStreamer(MCContext &Context,
                                       std::unique_ptr<MCAsmBackend> &&MAB,
                                       std::unique_ptr<MCObjectWriter> &&OW,
                                       std::unique_ptr<MCCodeEmitter> &&CE,
-                                      bool RelaxAll, bool DWARFMustBeAtTheEnd,
+                                      bool DWARFMustBeAtTheEnd,
                                       bool LabelSections) {
   MCMachOStreamer *S =
       new MCMachOStreamer(Context, std::move(MAB), std::move(OW), std::move(CE),
@@ -574,8 +574,6 @@ MCStreamer *llvm::createMachOStreamer(MCContext &Context,
       Target, Context.getObjectFileInfo()->getSDKVersion(),
       Context.getObjectFileInfo()->getDarwinTargetVariantTriple(),
       Context.getObjectFileInfo()->getDarwinTargetVariantSDKVersion());
-  if (RelaxAll)
-    S->getAssembler().setRelaxAll(true);
   return S;
 }
 
diff --git a/llvm/lib/MC/MCObjectStreamer.cpp b/llvm/lib/MC/MCObjectStreamer.cpp
index 490e0a4dd404..d2da5d0d3f90 100644
--- a/llvm/lib/MC/MCObjectStreamer.cpp
+++ b/llvm/lib/MC/MCObjectStreamer.cpp
@@ -34,6 +34,8 @@ MCObjectStreamer::MCObjectStreamer(MCContext &Context,
       EmitEHFrame(true), EmitDebugFrame(false) {
   if (Assembler->getBackendPtr())
     setAllowAutoPadding(Assembler->getBackend().allowAutoPadding());
+  if (Context.getTargetOptions() && Context.getTargetOptions()->MCRelaxAll)
+    Assembler->setRelaxAll(true);
 }
 
 MCObjectStreamer::~MCObjectStreamer() = default;
diff --git a/llvm/lib/MC/MCSPIRVStreamer.cpp b/llvm/lib/MC/MCSPIRVStreamer.cpp
index 0bb73c7ff7ee..3b75a2e17a4a 100644
--- a/llvm/lib/MC/MCSPIRVStreamer.cpp
+++ b/llvm/lib/MC/MCSPIRVStreamer.cpp
@@ -34,11 +34,8 @@ void MCSPIRVStreamer::emitInstToData(const MCInst &Inst,
 MCStreamer *llvm::createSPIRVStreamer(MCContext &Context,
                                       std::unique_ptr<MCAsmBackend> &&MAB,
                                       std::unique_ptr<MCObjectWriter> &&OW,
-                                      std::unique_ptr<MCCodeEmitter> &&CE,
-                                      bool RelaxAll) {
+                                      std::unique_ptr<MCCodeEmitter> &&CE) {
   MCSPIRVStreamer *S = new MCSPIRVStreamer(Context, std::move(MAB),
                                            std::move(OW), std::move(CE));
-  if (RelaxAll)
-    S->getAssembler().setRelaxAll(true);
   return S;
 }
diff --git a/llvm/lib/MC/MCWasmStreamer.cpp b/llvm/lib/MC/MCWasmStreamer.cpp
index fbab72fb5f3d..c553ede77555 100644
--- a/llvm/lib/MC/MCWasmStreamer.cpp
+++ b/llvm/lib/MC/MCWasmStreamer.cpp
@@ -275,11 +275,8 @@ void MCWasmStreamer::emitTBSSSymbol(MCSection *Section, MCSymbol *Symbol,
 MCStreamer *llvm::createWasmStreamer(MCContext &Context,
                                      std::unique_ptr<MCAsmBackend> &&MAB,
                                      std::unique_ptr<MCObjectWriter> &&OW,
-                                     std::unique_ptr<MCCodeEmitter> &&CE,
-                                     bool RelaxAll) {
+                                     std::unique_ptr<MCCodeEmitter> &&CE) {
   MCWasmStreamer *S =
       new MCWasmStreamer(Context, std::move(MAB), std::move(OW), std::move(CE));
-  if (RelaxAll)
-    S->getAssembler().setRelaxAll(true);
   return S;
 }
diff --git a/llvm/lib/MC/MCXCOFFStreamer.cpp b/llvm/lib/MC/MCXCOFFStreamer.cpp
index 458b4be61983..175d7d6b6c31 100644
--- a/llvm/lib/MC/MCXCOFFStreamer.cpp
+++ b/llvm/lib/MC/MCXCOFFStreamer.cpp
@@ -162,12 +162,9 @@ void MCXCOFFStreamer::emitInstToData(const MCInst &Inst,
 MCStreamer *llvm::createXCOFFStreamer(MCContext &Context,
                                       std::unique_ptr<MCAsmBackend> &&MAB,
                                       std::unique_ptr<MCObjectWriter> &&OW,
-                                      std::unique_ptr<MCCodeEmitter> &&CE,
-                                      bool RelaxAll) {
+                                      std::unique_ptr<MCCodeEmitter> &&CE) {
   MCXCOFFStreamer *S = new MCXCOFFStreamer(Context, std::move(MAB),
                                            std::move(OW), std::move(CE));
-  if (RelaxAll)
-    S->getAssembler().setRelaxAll(true);
   return S;
 }
 
diff --git a/llvm/lib/MCA/InstrBuilder.cpp b/llvm/lib/MCA/InstrBuilder.cpp
index 1a82e45763a2..2e3ebe3d9073 100644
--- a/llvm/lib/MCA/InstrBuilder.cpp
+++ b/llvm/lib/MCA/InstrBuilder.cpp
@@ -542,8 +542,7 @@ InstrBuilder::createInstrDescImpl(const MCInst &MCI,
   const MCSchedClassDesc &SCDesc = *SM.getSchedClassDesc(SchedClassID);
   if (SCDesc.NumMicroOps == MCSchedClassDesc::InvalidNumMicroOps) {
     return make_error<InstructionError<MCInst>>(
-        "found an unsupported instruction in the input assembly sequence.",
-        MCI);
+        "found an unsupported instruction in the input assembly sequence", MCI);
   }
 
   LLVM_DEBUG(dbgs() << "\n\t\tOpcode Name= " << MCII.getName(Opcode) << '\n');
diff --git a/llvm/lib/Object/ELFObjectFile.cpp b/llvm/lib/Object/ELFObjectFile.cpp
index efec612957de..24d7a7a280fd 100644
--- a/llvm/lib/Object/ELFObjectFile.cpp
+++ b/llvm/lib/Object/ELFObjectFile.cpp
@@ -24,7 +24,7 @@
 #include "llvm/Support/MathExtras.h"
 #include "llvm/Support/RISCVAttributeParser.h"
 #include "llvm/Support/RISCVAttributes.h"
-#include "llvm/Support/RISCVISAInfo.h"
+#include "llvm/TargetParser/RISCVISAInfo.h"
 #include "llvm/TargetParser/SubtargetFeature.h"
 #include "llvm/TargetParser/Triple.h"
 #include <algorithm>
diff --git a/llvm/lib/Object/ModuleSymbolTable.cpp b/llvm/lib/Object/ModuleSymbolTable.cpp
index 07f76688fa43..d8f520ad02c2 100644
--- a/llvm/lib/Object/ModuleSymbolTable.cpp
+++ b/llvm/lib/Object/ModuleSymbolTable.cpp
@@ -175,6 +175,20 @@ void ModuleSymbolTable::CollectAsmSymbols(
       AsmSymbol(Key, BasicSymbolRef::Flags(Res));
     }
   });
+
+  // In ELF, object code generated for x86-32 and some code models of x86-64 may
+  // reference the special symbol _GLOBAL_OFFSET_TABLE_ that is not used in the
+  // IR. Record it like inline asm symbols.
+  Triple TT(M.getTargetTriple());
+  if (!TT.isOSBinFormatELF() || !TT.isX86())
+    return;
+  auto CM = M.getCodeModel();
+  if (TT.getArch() == Triple::x86 || CM == CodeModel::Medium ||
+      CM == CodeModel::Large) {
+    AsmSymbol("_GLOBAL_OFFSET_TABLE_",
+              BasicSymbolRef::Flags(BasicSymbolRef::SF_Undefined |
+                                    BasicSymbolRef::SF_Global));
+  }
 }
 
 void ModuleSymbolTable::CollectAsmSymvers(
diff --git a/llvm/lib/ProfileData/InstrProfReader.cpp b/llvm/lib/ProfileData/InstrProfReader.cpp
index cefb6af12d00..ba21e01abfba 100644
--- a/llvm/lib/ProfileData/InstrProfReader.cpp
+++ b/llvm/lib/ProfileData/InstrProfReader.cpp
@@ -1520,54 +1520,38 @@ IndexedMemProfReader::getMemProfRecord(const uint64_t FuncNameHash) const {
 
   // Setup a callback to convert from frame ids to frame using the on-disk
   // FrameData hash table.
-  std::optional<memprof::FrameId> LastUnmappedFrameId;
-  auto IdToFrameCallback = [&](const memprof::FrameId Id) {
-    auto FrIter = MemProfFrameTable->find(Id);
-    if (FrIter == MemProfFrameTable->end()) {
-      LastUnmappedFrameId = Id;
-      return memprof::Frame(0, 0, 0, false);
-    }
-    return *FrIter;
-  };
-
-  // Setup a callback to convert call stack ids to call stacks using the on-disk
-  // hash table.
-  std::optional<memprof::CallStackId> LastUnmappedCSId;
-  auto CSIdToCallStackCallback = [&](memprof::CallStackId CSId) {
-    llvm::SmallVector<memprof::Frame> Frames;
-    auto CSIter = MemProfCallStackTable->find(CSId);
-    if (CSIter == MemProfCallStackTable->end()) {
-      LastUnmappedCSId = CSId;
-    } else {
-      const llvm::SmallVector<memprof::FrameId> &CS = *CSIter;
-      Frames.reserve(CS.size());
-      for (memprof::FrameId Id : CS)
-        Frames.push_back(IdToFrameCallback(Id));
-    }
-    return Frames;
-  };
+  memprof::FrameIdConverter<MemProfFrameHashTable> FrameIdConv(
+      *MemProfFrameTable.get());
 
   const memprof::IndexedMemProfRecord IndexedRecord = *Iter;
   memprof::MemProfRecord Record;
-  if (MemProfCallStackTable)
-    Record = IndexedRecord.toMemProfRecord(CSIdToCallStackCallback);
-  else
-    Record = memprof::MemProfRecord(IndexedRecord, IdToFrameCallback);
+  if (MemProfCallStackTable) {
+    // Setup a callback to convert call stack ids to call stacks using the
+    // on-disk hash table.
+    memprof::CallStackIdConverter<MemProfCallStackHashTable> CSIdConv(
+        *MemProfCallStackTable.get(), FrameIdConv);
 
-  // Check that all frame ids were successfully converted to frames.
-  if (LastUnmappedFrameId) {
-    return make_error<InstrProfError>(instrprof_error::hash_mismatch,
-                                      "memprof frame not found for frame id " +
-                                          Twine(*LastUnmappedFrameId));
+    Record = IndexedRecord.toMemProfRecord(CSIdConv);
+
+    // Check that all call stack ids were successfully converted to call stacks.
+    if (CSIdConv.LastUnmappedId) {
+      return make_error<InstrProfError>(
+          instrprof_error::hash_mismatch,
+          "memprof call stack not found for call stack id " +
+              Twine(*CSIdConv.LastUnmappedId));
+    }
+  } else {
+    Record = memprof::MemProfRecord(IndexedRecord, FrameIdConv);
   }
 
-  // Check that all call stack ids were successfully converted to call stacks.
-  if (LastUnmappedCSId) {
+  // Check that all frame ids were successfully converted to frames.
+  if (FrameIdConv.LastUnmappedId) {
     return make_error<InstrProfError>(
         instrprof_error::hash_mismatch,
-        "memprof call stack not found for call stack id " +
-            Twine(*LastUnmappedCSId));
+        "memprof frame not found for frame id " +
+            Twine(*FrameIdConv.LastUnmappedId));
   }
+
   return Record;
 }
 
diff --git a/llvm/lib/ProfileData/InstrProfWriter.cpp b/llvm/lib/ProfileData/InstrProfWriter.cpp
index 4a6fc9d64b69..b61c59aacc0f 100644
--- a/llvm/lib/ProfileData/InstrProfWriter.cpp
+++ b/llvm/lib/ProfileData/InstrProfWriter.cpp
@@ -184,12 +184,13 @@ public:
 InstrProfWriter::InstrProfWriter(
     bool Sparse, uint64_t TemporalProfTraceReservoirSize,
     uint64_t MaxTemporalProfTraceLength, bool WritePrevVersion,
-    memprof::IndexedVersion MemProfVersionRequested)
+    memprof::IndexedVersion MemProfVersionRequested, bool MemProfFullSchema)
     : Sparse(Sparse), MaxTemporalProfTraceLength(MaxTemporalProfTraceLength),
       TemporalProfTraceReservoirSize(TemporalProfTraceReservoirSize),
       InfoObj(new InstrProfRecordWriterTrait()),
       WritePrevVersion(WritePrevVersion),
-      MemProfVersionRequested(MemProfVersionRequested) {}
+      MemProfVersionRequested(MemProfVersionRequested),
+      MemProfFullSchema(MemProfFullSchema) {}
 
 InstrProfWriter::~InstrProfWriter() { delete InfoObj; }
 
@@ -507,7 +508,7 @@ static Error writeMemProfV0(
   OS.write(0ULL); // Reserve space for the memprof frame payload offset.
   OS.write(0ULL); // Reserve space for the memprof frame table offset.
 
-  auto Schema = memprof::PortableMemInfoBlock::getSchema();
+  auto Schema = memprof::getFullSchema();
   writeMemProfSchema(OS, Schema);
 
   uint64_t RecordTableOffset =
@@ -533,7 +534,7 @@ static Error writeMemProfV1(
   OS.write(0ULL); // Reserve space for the memprof frame payload offset.
   OS.write(0ULL); // Reserve space for the memprof frame table offset.
 
-  auto Schema = memprof::PortableMemInfoBlock::getSchema();
+  auto Schema = memprof::getFullSchema();
   writeMemProfSchema(OS, Schema);
 
   uint64_t RecordTableOffset =
@@ -554,7 +555,8 @@ static Error writeMemProfV2(
         &MemProfRecordData,
     llvm::MapVector<memprof::FrameId, memprof::Frame> &MemProfFrameData,
     llvm::MapVector<memprof::CallStackId, llvm::SmallVector<memprof::FrameId>>
-        &MemProfCallStackData) {
+        &MemProfCallStackData,
+    bool MemProfFullSchema) {
   OS.write(memprof::Version2);
   uint64_t HeaderUpdatePos = OS.tell();
   OS.write(0ULL); // Reserve space for the memprof record table offset.
@@ -563,7 +565,9 @@ static Error writeMemProfV2(
   OS.write(0ULL); // Reserve space for the memprof call stack payload offset.
   OS.write(0ULL); // Reserve space for the memprof call stack table offset.
 
-  auto Schema = memprof::PortableMemInfoBlock::getSchema();
+  auto Schema = memprof::getHotColdSchema();
+  if (MemProfFullSchema)
+    Schema = memprof::getFullSchema();
   writeMemProfSchema(OS, Schema);
 
   uint64_t RecordTableOffset =
@@ -605,7 +609,7 @@ static Error writeMemProf(
     llvm::MapVector<memprof::FrameId, memprof::Frame> &MemProfFrameData,
     llvm::MapVector<memprof::CallStackId, llvm::SmallVector<memprof::FrameId>>
         &MemProfCallStackData,
-    memprof::IndexedVersion MemProfVersionRequested) {
+    memprof::IndexedVersion MemProfVersionRequested, bool MemProfFullSchema) {
 
   switch (MemProfVersionRequested) {
   case memprof::Version0:
@@ -614,7 +618,7 @@ static Error writeMemProf(
     return writeMemProfV1(OS, MemProfRecordData, MemProfFrameData);
   case memprof::Version2:
     return writeMemProfV2(OS, MemProfRecordData, MemProfFrameData,
-                          MemProfCallStackData);
+                          MemProfCallStackData, MemProfFullSchema);
   }
 
   return make_error<InstrProfError>(
@@ -653,8 +657,8 @@ Error InstrProfWriter::writeImpl(ProfOStream &OS) {
                        : IndexedInstrProf::ProfVersion::CurrentVersion;
   // The WritePrevVersion handling will either need to be removed or updated
   // if the version is advanced beyond 12.
-  assert(IndexedInstrProf::ProfVersion::CurrentVersion ==
-         IndexedInstrProf::ProfVersion::Version12);
+  static_assert(IndexedInstrProf::ProfVersion::CurrentVersion ==
+                IndexedInstrProf::ProfVersion::Version12);
   if (static_cast<bool>(ProfileKind & InstrProfKind::IRInstrumentation))
     Header.Version |= VARIANT_MASK_IR_PROF;
   if (static_cast<bool>(ProfileKind & InstrProfKind::ContextSensitive))
@@ -733,7 +737,8 @@ Error InstrProfWriter::writeImpl(ProfOStream &OS) {
   if (static_cast<bool>(ProfileKind & InstrProfKind::MemProf)) {
     MemProfSectionStart = OS.tell();
     if (auto E = writeMemProf(OS, MemProfRecordData, MemProfFrameData,
-                              MemProfCallStackData, MemProfVersionRequested))
+                              MemProfCallStackData, MemProfVersionRequested,
+                              MemProfFullSchema))
       return E;
   }
 
diff --git a/llvm/lib/ProfileData/MemProf.cpp b/llvm/lib/ProfileData/MemProf.cpp
index 8e0402dd16e6..4667778ca11d 100644
--- a/llvm/lib/ProfileData/MemProf.cpp
+++ b/llvm/lib/ProfileData/MemProf.cpp
@@ -10,42 +10,59 @@
 
 namespace llvm {
 namespace memprof {
-static size_t serializedSizeV0(const IndexedAllocationInfo &IAI) {
+MemProfSchema getFullSchema() {
+  MemProfSchema List;
+#define MIBEntryDef(NameTag, Name, Type) List.push_back(Meta::Name);
+#include "llvm/ProfileData/MIBEntryDef.inc"
+#undef MIBEntryDef
+  return List;
+}
+
+MemProfSchema getHotColdSchema() {
+  return {Meta::AllocCount, Meta::TotalSize, Meta::TotalLifetime,
+          Meta::TotalLifetimeAccessDensity};
+}
+
+static size_t serializedSizeV0(const IndexedAllocationInfo &IAI,
+                               const MemProfSchema &Schema) {
   size_t Size = 0;
   // The number of frames to serialize.
   Size += sizeof(uint64_t);
   // The callstack frame ids.
   Size += sizeof(FrameId) * IAI.CallStack.size();
   // The size of the payload.
-  Size += PortableMemInfoBlock::serializedSize();
+  Size += PortableMemInfoBlock::serializedSize(Schema);
   return Size;
 }
 
-static size_t serializedSizeV2(const IndexedAllocationInfo &IAI) {
+static size_t serializedSizeV2(const IndexedAllocationInfo &IAI,
+                               const MemProfSchema &Schema) {
   size_t Size = 0;
   // The CallStackId
   Size += sizeof(CallStackId);
   // The size of the payload.
-  Size += PortableMemInfoBlock::serializedSize();
+  Size += PortableMemInfoBlock::serializedSize(Schema);
   return Size;
 }
 
-size_t IndexedAllocationInfo::serializedSize(IndexedVersion Version) const {
+size_t IndexedAllocationInfo::serializedSize(const MemProfSchema &Schema,
+                                             IndexedVersion Version) const {
   switch (Version) {
   case Version0:
   case Version1:
-    return serializedSizeV0(*this);
+    return serializedSizeV0(*this, Schema);
   case Version2:
-    return serializedSizeV2(*this);
+    return serializedSizeV2(*this, Schema);
   }
   llvm_unreachable("unsupported MemProf version");
 }
 
-static size_t serializedSizeV0(const IndexedMemProfRecord &Record) {
+static size_t serializedSizeV0(const IndexedMemProfRecord &Record,
+                               const MemProfSchema &Schema) {
   // The number of alloc sites to serialize.
   size_t Result = sizeof(uint64_t);
   for (const IndexedAllocationInfo &N : Record.AllocSites)
-    Result += N.serializedSize(Version0);
+    Result += N.serializedSize(Schema, Version0);
 
   // The number of callsites we have information for.
   Result += sizeof(uint64_t);
@@ -57,11 +74,12 @@ static size_t serializedSizeV0(const IndexedMemProfRecord &Record) {
   return Result;
 }
 
-static size_t serializedSizeV2(const IndexedMemProfRecord &Record) {
+static size_t serializedSizeV2(const IndexedMemProfRecord &Record,
+                               const MemProfSchema &Schema) {
   // The number of alloc sites to serialize.
   size_t Result = sizeof(uint64_t);
   for (const IndexedAllocationInfo &N : Record.AllocSites)
-    Result += N.serializedSize(Version2);
+    Result += N.serializedSize(Schema, Version2);
 
   // The number of callsites we have information for.
   Result += sizeof(uint64_t);
@@ -70,13 +88,14 @@ static size_t serializedSizeV2(const IndexedMemProfRecord &Record) {
   return Result;
 }
 
-size_t IndexedMemProfRecord::serializedSize(IndexedVersion Version) const {
+size_t IndexedMemProfRecord::serializedSize(const MemProfSchema &Schema,
+                                            IndexedVersion Version) const {
   switch (Version) {
   case Version0:
   case Version1:
-    return serializedSizeV0(*this);
+    return serializedSizeV0(*this, Schema);
   case Version2:
-    return serializedSizeV2(*this);
+    return serializedSizeV2(*this, Schema);
   }
   llvm_unreachable("unsupported MemProf version");
 }
@@ -156,7 +175,7 @@ static IndexedMemProfRecord deserializeV0(const MemProfSchema &Schema,
     }
     Node.CSId = hashCallStack(Node.CallStack);
     Node.Info.deserialize(Schema, Ptr);
-    Ptr += PortableMemInfoBlock::serializedSize();
+    Ptr += PortableMemInfoBlock::serializedSize(Schema);
     Record.AllocSites.push_back(Node);
   }
 
@@ -193,7 +212,7 @@ static IndexedMemProfRecord deserializeV2(const MemProfSchema &Schema,
     IndexedAllocationInfo Node;
     Node.CSId = endian::readNext<CallStackId, llvm::endianness::little>(Ptr);
     Node.Info.deserialize(Schema, Ptr);
-    Ptr += PortableMemInfoBlock::serializedSize();
+    Ptr += PortableMemInfoBlock::serializedSize(Schema);
     Record.AllocSites.push_back(Node);
   }
 
diff --git a/llvm/lib/Support/CMakeLists.txt b/llvm/lib/Support/CMakeLists.txt
index e18beddf7bc5..03e888958a07 100644
--- a/llvm/lib/Support/CMakeLists.txt
+++ b/llvm/lib/Support/CMakeLists.txt
@@ -219,7 +219,7 @@ add_llvm_component_library(LLVMSupport
   Regex.cpp
   RISCVAttributes.cpp
   RISCVAttributeParser.cpp
-  RISCVISAInfo.cpp
+  RISCVISAUtils.cpp
   ScaledNumber.cpp
   ScopedPrinter.cpp
   SHA1.cpp
diff --git a/llvm/lib/Support/RISCVISAUtils.cpp b/llvm/lib/Support/RISCVISAUtils.cpp
new file mode 100644
index 000000000000..ca7518f71907
--- /dev/null
+++ b/llvm/lib/Support/RISCVISAUtils.cpp
@@ -0,0 +1,88 @@
+//===-- RISCVISAUtils.cpp - RISC-V ISA Utilities --------------------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// Utilities shared by TableGen and RISCVISAInfo.
+//
+//===----------------------------------------------------------------------===//
+
+#include "llvm/Support/RISCVISAUtils.h"
+#include <cassert>
+
+using namespace llvm;
+
+// We rank extensions in the following order:
+// -Single letter extensions in canonical order.
+// -Unknown single letter extensions in alphabetical order.
+// -Multi-letter extensions starting with 'z' sorted by canonical order of
+//  the second letter then sorted alphabetically.
+// -Multi-letter extensions starting with 's' in alphabetical order.
+// -(TODO) Multi-letter extensions starting with 'zxm' in alphabetical order.
+// -X extensions in alphabetical order.
+// These flags are used to indicate the category. The first 6 bits store the
+// single letter extension rank for single letter and multi-letter extensions
+// starting with 'z'.
+enum RankFlags {
+  RF_Z_EXTENSION = 1 << 6,
+  RF_S_EXTENSION = 1 << 7,
+  RF_X_EXTENSION = 1 << 8,
+};
+
+// Get the rank for single-letter extension, lower value meaning higher
+// priority.
+static unsigned singleLetterExtensionRank(char Ext) {
+  assert(Ext >= 'a' && Ext <= 'z');
+  switch (Ext) {
+  case 'i':
+    return 0;
+  case 'e':
+    return 1;
+  }
+
+  size_t Pos = RISCVISAUtils::AllStdExts.find(Ext);
+  if (Pos != StringRef::npos)
+    return Pos + 2; // Skip 'e' and 'i' from above.
+
+  // If we got an unknown extension letter, then give it an alphabetical
+  // order, but after all known standard extensions.
+  return 2 + RISCVISAUtils::AllStdExts.size() + (Ext - 'a');
+}
+
+// Get the rank for multi-letter extension, lower value meaning higher
+// priority/order in canonical order.
+static unsigned getExtensionRank(const std::string &ExtName) {
+  assert(ExtName.size() >= 1);
+  switch (ExtName[0]) {
+  case 's':
+    return RF_S_EXTENSION;
+  case 'z':
+    assert(ExtName.size() >= 2);
+    // `z` extension must be sorted by canonical order of second letter.
+    // e.g. zmx has higher rank than zax.
+    return RF_Z_EXTENSION | singleLetterExtensionRank(ExtName[1]);
+  case 'x':
+    return RF_X_EXTENSION;
+  default:
+    assert(ExtName.size() == 1);
+    return singleLetterExtensionRank(ExtName[0]);
+  }
+}
+
+// Compare function for extension.
+// Only compare the extension name, ignore version comparison.
+bool llvm::RISCVISAUtils::compareExtension(const std::string &LHS,
+                                           const std::string &RHS) {
+  unsigned LHSRank = getExtensionRank(LHS);
+  unsigned RHSRank = getExtensionRank(RHS);
+
+  // If the ranks differ, pick the lower rank.
+  if (LHSRank != RHSRank)
+    return LHSRank < RHSRank;
+
+  // If the rank is same, it must be sorted by lexicographic order.
+  return LHS < RHS;
+}
diff --git a/llvm/lib/Support/SuffixTree.cpp b/llvm/lib/Support/SuffixTree.cpp
index eaa653078e09..c00c7989d1a6 100644
--- a/llvm/lib/Support/SuffixTree.cpp
+++ b/llvm/lib/Support/SuffixTree.cpp
@@ -242,8 +242,8 @@ void SuffixTree::RepeatedSubstringIterator::advance() {
     unsigned Length = Curr->getConcatLen();
 
     // Iterate over each child, saving internal nodes for visiting, and
-    // leaf nodes in LeafChildren. Internal nodes represent individual
-    // strings, which may repeat.
+    // leaf nodes' SuffixIdx in RepeatedSubstringStarts. Internal nodes
+    // represent individual strings, which may repeat.
     for (auto &ChildPair : Curr->Children) {
       // Save all of this node's children for processing.
       if (auto *InternalChild =
diff --git a/llvm/lib/Support/YAMLTraits.cpp b/llvm/lib/Support/YAMLTraits.cpp
index 4aaf59be2ce5..7bb60894b335 100644
--- a/llvm/lib/Support/YAMLTraits.cpp
+++ b/llvm/lib/Support/YAMLTraits.cpp
@@ -718,40 +718,8 @@ void Output::scalarString(StringRef &S, QuotingType MustQuote) {
     outputUpToEndOfLine("''");
     return;
   }
-  if (MustQuote == QuotingType::None) {
-    // Only quote if we must.
-    outputUpToEndOfLine(S);
-    return;
-  }
-
-  const char *const Quote = MustQuote == QuotingType::Single ? "'" : "\"";
-  output(Quote); // Starting quote.
-
-  // When using double-quoted strings (and only in that case), non-printable characters may be
-  // present, and will be escaped using a variety of unicode-scalar and special short-form
-  // escapes. This is handled in yaml::escape.
-  if (MustQuote == QuotingType::Double) {
-    output(yaml::escape(S, /* EscapePrintable= */ false));
-    outputUpToEndOfLine(Quote);
-    return;
-  }
-
-  unsigned i = 0;
-  unsigned j = 0;
-  unsigned End = S.size();
-  const char *Base = S.data();
-
-  // When using single-quoted strings, any single quote ' must be doubled to be escaped.
-  while (j < End) {
-    if (S[j] == '\'') {                    // Escape quotes.
-      output(StringRef(&Base[i], j - i));  // "flush".
-      output(StringLiteral("''"));         // Print it as ''
-      i = j + 1;
-    }
-    ++j;
-  }
-  output(StringRef(&Base[i], j - i));
-  outputUpToEndOfLine(Quote); // Ending quote.
+  output(S, MustQuote);
+  outputUpToEndOfLine("");
 }
 
 void Output::blockScalarString(StringRef &S) {
@@ -801,6 +769,46 @@ void Output::output(StringRef s) {
   Out << s;
 }
 
+void Output::output(StringRef S, QuotingType MustQuote) {
+  if (MustQuote == QuotingType::None) {
+    // Only quote if we must.
+    output(S);
+    return;
+  }
+
+  StringLiteral Quote = MustQuote == QuotingType::Single ? StringLiteral("'")
+                                                         : StringLiteral("\"");
+  output(Quote); // Starting quote.
+
+  // When using double-quoted strings (and only in that case), non-printable
+  // characters may be present, and will be escaped using a variety of
+  // unicode-scalar and special short-form escapes. This is handled in
+  // yaml::escape.
+  if (MustQuote == QuotingType::Double) {
+    output(yaml::escape(S, /* EscapePrintable= */ false));
+    output(Quote);
+    return;
+  }
+
+  unsigned i = 0;
+  unsigned j = 0;
+  unsigned End = S.size();
+  const char *Base = S.data();
+
+  // When using single-quoted strings, any single quote ' must be doubled to be
+  // escaped.
+  while (j < End) {
+    if (S[j] == '\'') {                   // Escape quotes.
+      output(StringRef(&Base[i], j - i)); // "flush".
+      output(StringLiteral("''"));        // Print it as ''
+      i = j + 1;
+    }
+    ++j;
+  }
+  output(StringRef(&Base[i], j - i));
+  output(Quote); // Ending quote.
+}
+
 void Output::outputUpToEndOfLine(StringRef s) {
   output(s);
   if (StateStack.empty() || (!inFlowSeqAnyElement(StateStack.back()) &&
@@ -853,7 +861,7 @@ void Output::newLineCheck(bool EmptySequence) {
 }
 
 void Output::paddedKey(StringRef key) {
-  output(key);
+  output(key, needsQuotes(key, false));
   output(":");
   const char *spaces = "                ";
   if (key.size() < strlen(spaces))
@@ -872,7 +880,7 @@ void Output::flowKey(StringRef Key) {
     Column = ColumnAtMapFlowStart;
     output("  ");
   }
-  output(Key);
+  output(Key, needsQuotes(Key, false));
   output(": ");
 }
 
diff --git a/llvm/lib/Target/AArch64/AArch64.td b/llvm/lib/Target/AArch64/AArch64.td
index 741c97a3dc00..4b2ce0d73949 100644
--- a/llvm/lib/Target/AArch64/AArch64.td
+++ b/llvm/lib/Target/AArch64/AArch64.td
@@ -16,727 +16,9 @@
 include "llvm/Target/Target.td"
 
 //===----------------------------------------------------------------------===//
-// AArch64 Subtarget features.
-//
-
-// Each SubtargetFeature which corresponds to an Arm Architecture feature should
-// be annotated with the respective FEAT_ feature name from the Architecture
-// Reference Manual. If a SubtargetFeature enables instructions from multiple
-// Arm Architecture Features, it should list all the relevant features. Not all
-// FEAT_ features have a corresponding SubtargetFeature.
-
-def FeatureFPARMv8 : SubtargetFeature<"fp-armv8", "HasFPARMv8", "true",
-                                       "Enable ARMv8 FP (FEAT_FP)">;
-
-def FeatureNEON : SubtargetFeature<"neon", "HasNEON", "true",
-  "Enable Advanced SIMD instructions (FEAT_AdvSIMD)", [FeatureFPARMv8]>;
-
-def FeatureSM4 : SubtargetFeature<
-    "sm4", "HasSM4", "true",
-    "Enable SM3 and SM4 support (FEAT_SM4, FEAT_SM3)", [FeatureNEON]>;
-
-def FeatureSHA2 : SubtargetFeature<
-    "sha2", "HasSHA2", "true",
-    "Enable SHA1 and SHA256 support (FEAT_SHA1, FEAT_SHA256)", [FeatureNEON]>;
-
-def FeatureSHA3 : SubtargetFeature<
-    "sha3", "HasSHA3", "true",
-    "Enable SHA512 and SHA3 support (FEAT_SHA3, FEAT_SHA512)", [FeatureNEON, FeatureSHA2]>;
-
-def FeatureAES : SubtargetFeature<
-    "aes", "HasAES", "true",
-    "Enable AES support (FEAT_AES, FEAT_PMULL)", [FeatureNEON]>;
-
-// Crypto has been split up and any combination is now valid (see the
-// crypto definitions above). Also, crypto is now context sensitive:
-// it has a different meaning for e.g. Armv8.4 than it has for Armv8.2.
-// Therefore, we rely on Clang, the user interfacing tool, to pass on the
-// appropriate crypto options. But here in the backend, crypto has very little
-// meaning anymore. We kept the Crypto definition here for backward
-// compatibility, and now imply features SHA2 and AES, which was the
-// "traditional" meaning of Crypto.
-def FeatureCrypto : SubtargetFeature<"crypto", "HasCrypto", "true",
-  "Enable cryptographic instructions", [FeatureNEON, FeatureSHA2, FeatureAES]>;
-
-def FeatureCRC : SubtargetFeature<"crc", "HasCRC", "true",
-  "Enable ARMv8 CRC-32 checksum instructions (FEAT_CRC32)">;
-
-def FeatureRAS : SubtargetFeature<"ras", "HasRAS", "true",
-  "Enable ARMv8 Reliability, Availability and Serviceability Extensions (FEAT_RAS, FEAT_RASv1p1)">;
-
-def FeatureRASv2 : SubtargetFeature<"rasv2", "HasRASv2", "true",
-  "Enable ARMv8.9-A Reliability, Availability and Serviceability Extensions (FEAT_RASv2)",
-  [FeatureRAS]>;
-
-def FeatureLSE : SubtargetFeature<"lse", "HasLSE", "true",
-  "Enable ARMv8.1 Large System Extension (LSE) atomic instructions (FEAT_LSE)">;
-
-def FeatureLSE2 : SubtargetFeature<"lse2", "HasLSE2", "true",
-  "Enable ARMv8.4 Large System Extension 2 (LSE2) atomicity rules (FEAT_LSE2)">;
-
-def FeatureOutlineAtomics : SubtargetFeature<"outline-atomics", "OutlineAtomics", "true",
-  "Enable out of line atomics to support LSE instructions">;
-
-def FeatureFMV : SubtargetFeature<"fmv", "HasFMV", "true",
-  "Enable Function Multi Versioning support.">;
-
-def FeatureRDM : SubtargetFeature<"rdm", "HasRDM", "true",
-  "Enable ARMv8.1 Rounding Double Multiply Add/Subtract instructions (FEAT_RDM)",
-  [FeatureNEON]>;
-
-def FeaturePAN : SubtargetFeature<
-    "pan", "HasPAN", "true",
-    "Enables ARM v8.1 Privileged Access-Never extension (FEAT_PAN)">;
-
-def FeatureLOR : SubtargetFeature<
-    "lor", "HasLOR", "true",
-    "Enables ARM v8.1 Limited Ordering Regions extension (FEAT_LOR)">;
-
-def FeatureCONTEXTIDREL2 : SubtargetFeature<"CONTEXTIDREL2", "HasCONTEXTIDREL2",
-    "true", "Enable RW operand CONTEXTIDR_EL2" >;
-
-def FeatureVH : SubtargetFeature<"vh", "HasVH", "true",
-    "Enables ARM v8.1 Virtual Host extension (FEAT_VHE)", [FeatureCONTEXTIDREL2] >;
-
-// This SubtargetFeature is special. It controls only whether codegen will turn
-// `llvm.readcyclecounter()` into an access to a PMUv3 System Register. The
-// `FEAT_PMUv3*` system registers are always available for assembly/disassembly.
-def FeaturePerfMon : SubtargetFeature<"perfmon", "HasPerfMon", "true",
-  "Enable Code Generation for ARMv8 PMUv3 Performance Monitors extension (FEAT_PMUv3)">;
-
-def FeatureFullFP16 : SubtargetFeature<"fullfp16", "HasFullFP16", "true",
-  "Full FP16 (FEAT_FP16)", [FeatureFPARMv8]>;
-
-def FeatureFP16FML : SubtargetFeature<"fp16fml", "HasFP16FML", "true",
-  "Enable FP16 FML instructions (FEAT_FHM)", [FeatureFullFP16]>;
-
-def FeatureSPE : SubtargetFeature<"spe", "HasSPE", "true",
-  "Enable Statistical Profiling extension (FEAT_SPE)">;
-
-def FeaturePAN_RWV : SubtargetFeature<
-    "pan-rwv", "HasPAN_RWV", "true",
-    "Enable v8.2 PAN s1e1R and s1e1W Variants (FEAT_PAN2)",
-    [FeaturePAN]>;
-
-// UAO PState
-def FeaturePsUAO : SubtargetFeature< "uaops", "HasPsUAO", "true",
-    "Enable v8.2 UAO PState (FEAT_UAO)">;
-
-def FeatureCCPP : SubtargetFeature<"ccpp", "HasCCPP",
-    "true", "Enable v8.2 data Cache Clean to Point of Persistence (FEAT_DPB)" >;
-
-def FeatureSVE : SubtargetFeature<"sve", "HasSVE", "true",
-  "Enable Scalable Vector Extension (SVE) instructions (FEAT_SVE)", [FeatureFullFP16]>;
-
-def FeatureFPMR : SubtargetFeature<"fpmr", "HasFPMR", "true",
-  "Enable FPMR Register (FEAT_FPMR)">;
-
-def FeatureFP8 : SubtargetFeature<"fp8", "HasFP8", "true",
-  "Enable FP8 instructions (FEAT_FP8)">;
-
-// This flag is currently still labeled as Experimental, but when fully
-// implemented this should tell the compiler to use the zeroing pseudos to
-// benefit from the reverse instructions (e.g. SUB vs SUBR) if the inactive
-// lanes are known to be zero. The pseudos will then be expanded using the
-// MOVPRFX instruction to zero the inactive lanes. This feature should only be
-// enabled if MOVPRFX instructions are known to merge with the destructive
-// operations they prefix.
-//
-// This feature could similarly be extended to support cheap merging of _any_
-// value into the inactive lanes using the MOVPRFX instruction that uses
-// merging-predication.
-def FeatureExperimentalZeroingPseudos
-    : SubtargetFeature<"use-experimental-zeroing-pseudos",
-                       "UseExperimentalZeroingPseudos", "true",
-                       "Hint to the compiler that the MOVPRFX instruction is "
-                       "merged with destructive operations",
-                       []>;
-
-def FeatureUseScalarIncVL : SubtargetFeature<"use-scalar-inc-vl",
-  "UseScalarIncVL", "true", "Prefer inc/dec over add+cnt">;
-
-def FeatureBF16 : SubtargetFeature<"bf16", "HasBF16",
-    "true", "Enable BFloat16 Extension (FEAT_BF16)" >;
-
-def FeatureNoSVEFPLD1R : SubtargetFeature<"no-sve-fp-ld1r",
-  "NoSVEFPLD1R", "true", "Avoid using LD1RX instructions for FP">;
-
-def FeatureSVE2 : SubtargetFeature<"sve2", "HasSVE2", "true",
-  "Enable Scalable Vector Extension 2 (SVE2) instructions (FEAT_SVE2)",
-  [FeatureSVE, FeatureUseScalarIncVL]>;
-
-def FeatureSVE2AES : SubtargetFeature<"sve2-aes", "HasSVE2AES", "true",
-  "Enable AES SVE2 instructions (FEAT_SVE_AES, FEAT_SVE_PMULL128)",
-  [FeatureSVE2, FeatureAES]>;
-
-def FeatureSVE2SM4 : SubtargetFeature<"sve2-sm4", "HasSVE2SM4", "true",
-  "Enable SM4 SVE2 instructions (FEAT_SVE_SM4)", [FeatureSVE2, FeatureSM4]>;
-
-def FeatureSVE2SHA3 : SubtargetFeature<"sve2-sha3", "HasSVE2SHA3", "true",
-  "Enable SHA3 SVE2 instructions (FEAT_SVE_SHA3)", [FeatureSVE2, FeatureSHA3]>;
-
-def FeatureSVE2BitPerm : SubtargetFeature<"sve2-bitperm", "HasSVE2BitPerm", "true",
-  "Enable bit permutation SVE2 instructions (FEAT_SVE_BitPerm)", [FeatureSVE2]>;
-
-def FeatureSVE2p1: SubtargetFeature<"sve2p1", "HasSVE2p1", "true",
-  "Enable Scalable Vector Extension 2.1 instructions", [FeatureSVE2]>;
-
-def FeatureB16B16 : SubtargetFeature<"b16b16", "HasB16B16", "true",
-  "Enable SVE2.1 or SME2.1 non-widening BFloat16 to BFloat16 instructions (FEAT_B16B16)", [FeatureBF16]>;
-
-def FeatureZCRegMove : SubtargetFeature<"zcm", "HasZeroCycleRegMove", "true",
-                                        "Has zero-cycle register moves">;
-
-def FeatureZCZeroingGP : SubtargetFeature<"zcz-gp", "HasZeroCycleZeroingGP", "true",
-                                        "Has zero-cycle zeroing instructions for generic registers">;
-
-// It is generally beneficial to rewrite "fmov s0, wzr" to "movi d0, #0".
-// as movi is more efficient across all cores. Newer cores can eliminate
-// fmovs early and there is no difference with movi, but this not true for
-// all implementations.
-def FeatureNoZCZeroingFP : SubtargetFeature<"no-zcz-fp", "HasZeroCycleZeroingFP", "false",
-                                        "Has no zero-cycle zeroing instructions for FP registers">;
-
-def FeatureZCZeroing : SubtargetFeature<"zcz", "HasZeroCycleZeroing", "true",
-                                        "Has zero-cycle zeroing instructions",
-                                        [FeatureZCZeroingGP]>;
-
-/// ... but the floating-point version doesn't quite work in rare cases on older
-/// CPUs.
-def FeatureZCZeroingFPWorkaround : SubtargetFeature<"zcz-fp-workaround",
-    "HasZeroCycleZeroingFPWorkaround", "true",
-    "The zero-cycle floating-point zeroing instruction has a bug">;
-
-def FeatureStrictAlign : SubtargetFeature<"strict-align",
-                                          "RequiresStrictAlign", "true",
-                                          "Disallow all unaligned memory "
-                                          "access">;
-
-foreach i = {1-7,9-15,18,20-28} in
-    def FeatureReserveX#i : SubtargetFeature<"reserve-x"#i, "ReserveXRegister["#i#"]", "true",
-                                             "Reserve X"#i#", making it unavailable "
-                                             "as a GPR">;
-
-foreach i = {8-15,18} in
-    def FeatureCallSavedX#i : SubtargetFeature<"call-saved-x"#i,
-         "CustomCallSavedXRegs["#i#"]", "true", "Make X"#i#" callee saved.">;
-
-def FeatureBalanceFPOps : SubtargetFeature<"balance-fp-ops", "BalanceFPOps",
-    "true",
-    "balance mix of odd and even D-registers for fp multiply(-accumulate) ops">;
-
-def FeaturePredictableSelectIsExpensive : SubtargetFeature<
-    "predictable-select-expensive", "PredictableSelectIsExpensive", "true",
-    "Prefer likely predicted branches over selects">;
-
-def FeatureEnableSelectOptimize : SubtargetFeature<
-    "enable-select-opt", "EnableSelectOptimize", "true",
-    "Enable the select optimize pass for select loop heuristics">;
-
-def FeatureExynosCheapAsMoveHandling : SubtargetFeature<"exynos-cheap-as-move",
-    "HasExynosCheapAsMoveHandling", "true",
-    "Use Exynos specific handling of cheap instructions">;
-
-def FeaturePostRAScheduler : SubtargetFeature<"use-postra-scheduler",
-    "UsePostRAScheduler", "true", "Schedule again after register allocation">;
-
-def FeatureSlowMisaligned128Store : SubtargetFeature<"slow-misaligned-128store",
-    "IsMisaligned128StoreSlow", "true", "Misaligned 128 bit stores are slow">;
-
-def FeatureSlowPaired128 : SubtargetFeature<"slow-paired-128",
-    "IsPaired128Slow", "true", "Paired 128 bit loads and stores are slow">;
-
-def FeatureAscendStoreAddress : SubtargetFeature<"ascend-store-address",
-    "IsStoreAddressAscend", "true",
-    "Schedule vector stores by ascending address">;
-
-def FeatureSlowSTRQro : SubtargetFeature<"slow-strqro-store", "IsSTRQroSlow",
-    "true", "STR of Q register with register offset is slow">;
-
-def FeatureAlternateSExtLoadCVTF32Pattern : SubtargetFeature<
-    "alternate-sextload-cvt-f32-pattern", "UseAlternateSExtLoadCVTF32Pattern",
-    "true", "Use alternative pattern for sextload convert to f32">;
-
-def FeatureArithmeticBccFusion : SubtargetFeature<
-    "arith-bcc-fusion", "HasArithmeticBccFusion", "true",
-    "CPU fuses arithmetic+bcc operations">;
-
-def FeatureArithmeticCbzFusion : SubtargetFeature<
-    "arith-cbz-fusion", "HasArithmeticCbzFusion", "true",
-    "CPU fuses arithmetic + cbz/cbnz operations">;
-
-def FeatureCmpBccFusion : SubtargetFeature<
-    "cmp-bcc-fusion", "HasCmpBccFusion", "true",
-    "CPU fuses cmp+bcc operations">;
-
-def FeatureFuseAddress : SubtargetFeature<
-    "fuse-address", "HasFuseAddress", "true",
-    "CPU fuses address generation and memory operations">;
-
-def FeatureFuseAES : SubtargetFeature<
-    "fuse-aes", "HasFuseAES", "true",
-    "CPU fuses AES crypto operations">;
-
-def FeatureFuseArithmeticLogic : SubtargetFeature<
-    "fuse-arith-logic", "HasFuseArithmeticLogic", "true",
-    "CPU fuses arithmetic and logic operations">;
-
-def FeatureFuseCCSelect : SubtargetFeature<
-    "fuse-csel", "HasFuseCCSelect", "true",
-    "CPU fuses conditional select operations">;
-
-def FeatureFuseCryptoEOR : SubtargetFeature<
-    "fuse-crypto-eor", "HasFuseCryptoEOR", "true",
-    "CPU fuses AES/PMULL and EOR operations">;
-
-def FeatureFuseAdrpAdd : SubtargetFeature<
-    "fuse-adrp-add", "HasFuseAdrpAdd", "true",
-    "CPU fuses adrp+add operations">;
-
-def FeatureFuseLiterals : SubtargetFeature<
-    "fuse-literals", "HasFuseLiterals", "true",
-    "CPU fuses literal generation operations">;
-
-def FeatureFuseAddSub2RegAndConstOne : SubtargetFeature<
-   "fuse-addsub-2reg-const1", "HasFuseAddSub2RegAndConstOne", "true",
-   "CPU fuses (a + b + 1) and (a - b - 1)">;
-
-def FeatureDisableLatencySchedHeuristic : SubtargetFeature<
-    "disable-latency-sched-heuristic", "DisableLatencySchedHeuristic", "true",
-    "Disable latency scheduling heuristic">;
-
-def FeatureStorePairSuppress : SubtargetFeature<
-    "store-pair-suppress", "EnableStorePairSuppress", "true",
-    "Enable Store Pair Suppression heuristics">;
-
-def FeatureForce32BitJumpTables
-   : SubtargetFeature<"force-32bit-jump-tables", "Force32BitJumpTables", "true",
-                      "Force jump table entries to be 32-bits wide except at MinSize">;
-
-def FeatureRCPC : SubtargetFeature<"rcpc", "HasRCPC", "true",
-                                   "Enable support for RCPC extension (FEAT_LRCPC)">;
-
-def FeatureUseRSqrt : SubtargetFeature<
-    "use-reciprocal-square-root", "UseRSqrt", "true",
-    "Use the reciprocal square root approximation">;
-
-def FeatureDotProd : SubtargetFeature<
-    "dotprod", "HasDotProd", "true",
-    "Enable dot product support (FEAT_DotProd)", [FeatureNEON]>;
-
-def FeaturePAuth : SubtargetFeature<
-    "pauth", "HasPAuth", "true",
-    "Enable v8.3-A Pointer Authentication extension (FEAT_PAuth)">;
-
-def FeatureJS : SubtargetFeature<
-    "jsconv", "HasJS", "true",
-    "Enable v8.3-A JavaScript FP conversion instructions (FEAT_JSCVT)",
-    [FeatureFPARMv8]>;
-
-def FeatureCCIDX : SubtargetFeature<
-    "ccidx", "HasCCIDX", "true",
-    "Enable v8.3-A Extend of the CCSIDR number of sets (FEAT_CCIDX)">;
-
-def FeatureComplxNum : SubtargetFeature<
-    "complxnum", "HasComplxNum", "true",
-    "Enable v8.3-A Floating-point complex number support (FEAT_FCMA)",
-    [FeatureNEON]>;
-
-def FeatureNV : SubtargetFeature<
-    "nv", "HasNV", "true",
-    "Enable v8.4-A Nested Virtualization Enchancement (FEAT_NV, FEAT_NV2)">;
-
-def FeatureMPAM : SubtargetFeature<
-    "mpam", "HasMPAM", "true",
-    "Enable v8.4-A Memory system Partitioning and Monitoring extension (FEAT_MPAM)">;
-
-def FeatureDIT : SubtargetFeature<
-    "dit", "HasDIT", "true",
-    "Enable v8.4-A Data Independent Timing instructions (FEAT_DIT)">;
-
-def FeatureTRACEV8_4 : SubtargetFeature<
-    "tracev8.4", "HasTRACEV8_4", "true",
-    "Enable v8.4-A Trace extension (FEAT_TRF)">;
-
-def FeatureAM : SubtargetFeature<
-    "am", "HasAM", "true",
-    "Enable v8.4-A Activity Monitors extension (FEAT_AMUv1)">;
-
-def FeatureAMVS : SubtargetFeature<
-    "amvs", "HasAMVS", "true",
-    "Enable v8.6-A Activity Monitors Virtualization support (FEAT_AMUv1p1)",
-    [FeatureAM]>;
-
-def FeatureSEL2 : SubtargetFeature<
-    "sel2", "HasSEL2", "true",
-    "Enable v8.4-A Secure Exception Level 2 extension (FEAT_SEL2)">;
-
-def FeatureTLB_RMI : SubtargetFeature<
-    "tlb-rmi", "HasTLB_RMI", "true",
-    "Enable v8.4-A TLB Range and Maintenance Instructions (FEAT_TLBIOS, FEAT_TLBIRANGE)">;
-
-def FeatureFlagM : SubtargetFeature<
-    "flagm", "HasFlagM", "true",
-    "Enable v8.4-A Flag Manipulation Instructions (FEAT_FlagM)">;
-
-// 8.4 RCPC enchancements: LDAPR & STLR instructions with Immediate Offset
-def FeatureRCPC_IMMO : SubtargetFeature<"rcpc-immo", "HasRCPC_IMMO", "true",
-    "Enable v8.4-A RCPC instructions with Immediate Offsets (FEAT_LRCPC2)",
-    [FeatureRCPC]>;
-
-def FeatureNoNegativeImmediates : SubtargetFeature<"no-neg-immediates",
-                                        "NegativeImmediates", "false",
-                                        "Convert immediates and instructions "
-                                        "to their negated or complemented "
-                                        "equivalent when the immediate does "
-                                        "not fit in the encoding.">;
-
-// Address operands with shift amount 2 or 3 are fast on all Arm chips except
-// some old Apple cores (A7-A10?) which handle all shifts slowly. Cortex-A57
-// and derived designs through Cortex-X1 take an extra micro-op for shifts
-// of 1 or 4. Other Arm chips handle all shifted operands at the same speed
-// as unshifted operands.
-//
-// We don't try to model the behavior of the old Apple cores because new code
-// targeting A7 is very unlikely to actually run on an A7. The Cortex cores
-// are modeled by FeatureAddrLSLSlow14.
-def FeatureAddrLSLSlow14 : SubtargetFeature<
-    "addr-lsl-slow-14", "HasAddrLSLSlow14", "true",
-    "Address operands with shift amount of 1 or 4 are slow">;
-
-def FeatureALULSLFast : SubtargetFeature<
-    "alu-lsl-fast", "HasALULSLFast", "true",
-    "Add/Sub operations with lsl shift <= 4 are cheap">;
-
-def FeatureAggressiveFMA :
-  SubtargetFeature<"aggressive-fma",
-                   "HasAggressiveFMA",
-                   "true",
-                   "Enable Aggressive FMA for floating-point.">;
-
-def FeatureAltFPCmp : SubtargetFeature<"altnzcv", "HasAlternativeNZCV", "true",
-  "Enable alternative NZCV format for floating point comparisons (FEAT_FlagM2)">;
-
-def FeatureFRInt3264 : SubtargetFeature<"fptoint", "HasFRInt3264", "true",
-  "Enable FRInt[32|64][Z|X] instructions that round a floating-point number to "
-  "an integer (in FP format) forcing it to fit into a 32- or 64-bit int (FEAT_FRINTTS)" >;
-
-def FeatureSpecRestrict : SubtargetFeature<"specrestrict", "HasSpecRestrict",
-  "true", "Enable architectural speculation restriction (FEAT_CSV2_2)">;
-
-def FeatureSB : SubtargetFeature<"sb", "HasSB",
-  "true", "Enable v8.5 Speculation Barrier (FEAT_SB)" >;
-
-def FeatureSSBS : SubtargetFeature<"ssbs", "HasSSBS",
-  "true", "Enable Speculative Store Bypass Safe bit (FEAT_SSBS, FEAT_SSBS2)" >;
-
-def FeaturePredRes : SubtargetFeature<"predres", "HasPredRes", "true",
-  "Enable v8.5a execution and data prediction invalidation instructions (FEAT_SPECRES)" >;
-
-def FeatureCacheDeepPersist : SubtargetFeature<"ccdp", "HasCCDP",
-    "true", "Enable v8.5 Cache Clean to Point of Deep Persistence (FEAT_DPB2)" >;
-
-def FeatureBranchTargetId : SubtargetFeature<"bti", "HasBTI",
-    "true", "Enable Branch Target Identification (FEAT_BTI)" >;
-
-def FeatureRandGen : SubtargetFeature<"rand", "HasRandGen",
-    "true", "Enable Random Number generation instructions (FEAT_RNG)" >;
-
-def FeatureMTE : SubtargetFeature<"mte", "HasMTE",
-    "true", "Enable Memory Tagging Extension (FEAT_MTE, FEAT_MTE2)" >;
-
-def FeatureTRBE : SubtargetFeature<"trbe", "HasTRBE",
-    "true", "Enable Trace Buffer Extension (FEAT_TRBE)">;
-
-def FeatureETE : SubtargetFeature<"ete", "HasETE",
-    "true", "Enable Embedded Trace Extension (FEAT_ETE)",
-    [FeatureTRBE]>;
-
-def FeatureTME : SubtargetFeature<"tme", "HasTME",
-    "true", "Enable Transactional Memory Extension (FEAT_TME)" >;
-
-def FeatureTaggedGlobals : SubtargetFeature<"tagged-globals",
-    "AllowTaggedGlobals",
-    "true", "Use an instruction sequence for taking the address of a global "
-    "that allows a memory tag in the upper address bits">;
-
-def FeatureMatMulInt8 : SubtargetFeature<"i8mm", "HasMatMulInt8",
-    "true", "Enable Matrix Multiply Int8 Extension (FEAT_I8MM)">;
-
-def FeatureMatMulFP32 : SubtargetFeature<"f32mm", "HasMatMulFP32",
-    "true", "Enable Matrix Multiply FP32 Extension (FEAT_F32MM)", [FeatureSVE]>;
-
-def FeatureMatMulFP64 : SubtargetFeature<"f64mm", "HasMatMulFP64",
-    "true", "Enable Matrix Multiply FP64 Extension (FEAT_F64MM)", [FeatureSVE]>;
-
-def FeatureXS : SubtargetFeature<"xs", "HasXS",
-    "true", "Enable Armv8.7-A limited-TLB-maintenance instruction (FEAT_XS)">;
-
-def FeatureWFxT : SubtargetFeature<"wfxt", "HasWFxT",
-    "true", "Enable Armv8.7-A WFET and WFIT instruction (FEAT_WFxT)">;
-
-def FeatureHCX : SubtargetFeature<
-    "hcx", "HasHCX", "true", "Enable Armv8.7-A HCRX_EL2 system register (FEAT_HCX)">;
-
-def FeatureLS64 : SubtargetFeature<"ls64", "HasLS64",
-    "true", "Enable Armv8.7-A LD64B/ST64B Accelerator Extension (FEAT_LS64, FEAT_LS64_V, FEAT_LS64_ACCDATA)">;
-
-def FeatureHBC : SubtargetFeature<"hbc", "HasHBC",
-    "true", "Enable Armv8.8-A Hinted Conditional Branches Extension (FEAT_HBC)">;
-
-def FeatureMOPS : SubtargetFeature<"mops", "HasMOPS",
-    "true", "Enable Armv8.8-A memcpy and memset acceleration instructions (FEAT_MOPS)">;
-
-def FeatureNMI : SubtargetFeature<"nmi", "HasNMI",
-    "true", "Enable Armv8.8-A Non-maskable Interrupts (FEAT_NMI, FEAT_GICv3_NMI)">;
-
-def FeatureBRBE : SubtargetFeature<"brbe", "HasBRBE",
-    "true", "Enable Branch Record Buffer Extension (FEAT_BRBE)">;
-
-def FeatureSPE_EEF : SubtargetFeature<"spe-eef", "HasSPE_EEF",
-    "true", "Enable extra register in the Statistical Profiling Extension (FEAT_SPEv1p2)">;
-
-def FeatureFineGrainedTraps : SubtargetFeature<"fgt", "HasFineGrainedTraps",
-    "true", "Enable fine grained virtualization traps extension (FEAT_FGT)">;
-
-def FeatureEnhancedCounterVirtualization :
-      SubtargetFeature<"ecv", "HasEnhancedCounterVirtualization",
-      "true", "Enable enhanced counter virtualization extension (FEAT_ECV)">;
-
-def FeatureRME : SubtargetFeature<"rme", "HasRME",
-    "true", "Enable Realm Management Extension (FEAT_RME)">;
-
-def FeatureSME : SubtargetFeature<"sme", "HasSME", "true",
-  "Enable Scalable Matrix Extension (SME) (FEAT_SME)", [FeatureBF16, FeatureUseScalarIncVL]>;
-
-def FeatureSMEF64F64 : SubtargetFeature<"sme-f64f64", "HasSMEF64F64", "true",
-  "Enable Scalable Matrix Extension (SME) F64F64 instructions (FEAT_SME_F64F64)", [FeatureSME]>;
-
-def FeatureSMEI16I64 : SubtargetFeature<"sme-i16i64", "HasSMEI16I64", "true",
-  "Enable Scalable Matrix Extension (SME) I16I64 instructions (FEAT_SME_I16I64)", [FeatureSME]>;
-
-def FeatureSMEF16F16 : SubtargetFeature<"sme-f16f16", "HasSMEF16F16", "true",
-  "Enable SME2.1 non-widening Float16 instructions (FEAT_SME_F16F16)", []>;
-
-def FeatureSMEFA64 : SubtargetFeature<"sme-fa64", "HasSMEFA64", "true",
-  "Enable the full A64 instruction set in streaming SVE mode (FEAT_SME_FA64)", [FeatureSME, FeatureSVE2]>;
-
-def FeatureSME2 : SubtargetFeature<"sme2", "HasSME2", "true",
-  "Enable Scalable Matrix Extension 2 (SME2) instructions", [FeatureSME]>;
-
-def FeatureSME2p1 : SubtargetFeature<"sme2p1", "HasSME2p1", "true",
-  "Enable Scalable Matrix Extension 2.1 (FEAT_SME2p1) instructions", [FeatureSME2]>;
-
-def FeatureFAMINMAX: SubtargetFeature<"faminmax", "HasFAMINMAX", "true",
-   "Enable FAMIN and FAMAX instructions (FEAT_FAMINMAX)">;
-
-def FeatureFP8FMA : SubtargetFeature<"fp8fma", "HasFP8FMA", "true",
-  "Enable fp8 multiply-add instructions (FEAT_FP8FMA)">;
-
-def FeatureSSVE_FP8FMA : SubtargetFeature<"ssve-fp8fma", "HasSSVE_FP8FMA", "true",
-  "Enable SVE2 fp8 multiply-add instructions (FEAT_SSVE_FP8FMA)", [FeatureSME2]>;
-
-def FeatureFP8DOT2: SubtargetFeature<"fp8dot2", "HasFP8DOT2", "true",
-   "Enable fp8 2-way dot instructions (FEAT_FP8DOT2)">;
-
-def FeatureSSVE_FP8DOT2 : SubtargetFeature<"ssve-fp8dot2", "HasSSVE_FP8DOT2", "true",
-  "Enable SVE2 fp8 2-way dot product instructions (FEAT_SSVE_FP8DOT2)", [FeatureSME2]>;
-
-def FeatureFP8DOT4: SubtargetFeature<"fp8dot4", "HasFP8DOT4", "true",
-   "Enable fp8 4-way dot instructions (FEAT_FP8DOT4)">;
-
-def FeatureSSVE_FP8DOT4 : SubtargetFeature<"ssve-fp8dot4", "HasSSVE_FP8DOT4", "true",
-  "Enable SVE2 fp8 4-way dot product instructions (FEAT_SSVE_FP8DOT4)", [FeatureSME2]>;
-def FeatureLUT: SubtargetFeature<"lut", "HasLUT", "true",
-   "Enable Lookup Table instructions (FEAT_LUT)">;
-
-def FeatureSME_LUTv2 : SubtargetFeature<"sme-lutv2", "HasSME_LUTv2", "true",
-  "Enable Scalable Matrix Extension (SME) LUTv2 instructions (FEAT_SME_LUTv2)">;
-
-def FeatureSMEF8F16 : SubtargetFeature<"sme-f8f16", "HasSMEF8F16", "true",
-  "Enable Scalable Matrix Extension (SME) F8F16 instructions(FEAT_SME_F8F16)", [FeatureSME2, FeatureFP8]>;
-
-def FeatureSMEF8F32 : SubtargetFeature<"sme-f8f32", "HasSMEF8F32", "true",
-  "Enable Scalable Matrix Extension (SME) F8F32 instructions (FEAT_SME_F8F32)", [FeatureSME2, FeatureFP8]>;
-
-def FeatureAppleA7SysReg  : SubtargetFeature<"apple-a7-sysreg", "HasAppleA7SysReg", "true",
-  "Apple A7 (the CPU formerly known as Cyclone)">;
-
-def FeatureEL2VMSA : SubtargetFeature<"el2vmsa", "HasEL2VMSA", "true",
-  "Enable Exception Level 2 Virtual Memory System Architecture">;
-
-def FeatureEL3 : SubtargetFeature<"el3", "HasEL3", "true",
-  "Enable Exception Level 3">;
-
-def FeatureCSSC : SubtargetFeature<"cssc", "HasCSSC", "true",
-  "Enable Common Short Sequence Compression (CSSC) instructions (FEAT_CSSC)">;
-
-def FeatureFixCortexA53_835769 : SubtargetFeature<"fix-cortex-a53-835769",
-  "FixCortexA53_835769", "true", "Mitigate Cortex-A53 Erratum 835769">;
-
-def FeatureNoBTIAtReturnTwice : SubtargetFeature<"no-bti-at-return-twice",
-                                                 "NoBTIAtReturnTwice", "true",
-                                                 "Don't place a BTI instruction "
-                                                 "after a return-twice">;
-
-def FeatureCHK : SubtargetFeature<"chk", "HasCHK",
-    "true", "Enable Armv8.0-A Check Feature Status Extension (FEAT_CHK)">;
-
-def FeatureGCS : SubtargetFeature<"gcs", "HasGCS",
-    "true", "Enable Armv9.4-A Guarded Call Stack Extension", [FeatureCHK]>;
-
-def FeatureCLRBHB : SubtargetFeature<"clrbhb", "HasCLRBHB",
-    "true", "Enable Clear BHB instruction (FEAT_CLRBHB)">;
-
-def FeaturePRFM_SLC : SubtargetFeature<"prfm-slc-target", "HasPRFM_SLC",
-    "true", "Enable SLC target for PRFM instruction">;
-
-def FeatureSPECRES2 : SubtargetFeature<"specres2", "HasSPECRES2",
-    "true", "Enable Speculation Restriction Instruction (FEAT_SPECRES2)",
-    [FeaturePredRes]>;
-
-def FeatureMEC : SubtargetFeature<"mec", "HasMEC",
-    "true", "Enable Memory Encryption Contexts Extension", [FeatureRME]>;
-
-def FeatureITE : SubtargetFeature<"ite", "HasITE",
-    "true", "Enable Armv9.4-A Instrumentation Extension FEAT_ITE", [FeatureETE,
-    FeatureTRBE]>;
-
-def FeatureRCPC3 : SubtargetFeature<"rcpc3", "HasRCPC3",
-    "true", "Enable Armv8.9-A RCPC instructions for A64 and Advanced SIMD and floating-point instruction set (FEAT_LRCPC3)",
-    [FeatureRCPC_IMMO]>;
-
-def FeatureTHE : SubtargetFeature<"the", "HasTHE",
-    "true", "Enable Armv8.9-A Translation Hardening Extension (FEAT_THE)">;
-
-def FeatureLSE128 : SubtargetFeature<"lse128", "HasLSE128",
-    "true", "Enable Armv9.4-A 128-bit Atomic Instructions (FEAT_LSE128)",
-    [FeatureLSE]>;
-
-// FEAT_D128, FEAT_LVA3, FEAT_SYSREG128, and FEAT_SYSINSTR128 are mutually implicit.
-// Therefore group them all under a single feature flag, d128:
-def FeatureD128 : SubtargetFeature<"d128", "HasD128",
-    "true", "Enable Armv9.4-A 128-bit Page Table Descriptors, System Registers "
-    "and Instructions (FEAT_D128, FEAT_LVA3, FEAT_SYSREG128, FEAT_SYSINSTR128)",
-    [FeatureLSE128]>;
-
-def FeatureDisableLdp : SubtargetFeature<"disable-ldp", "HasDisableLdp",
-    "true", "Do not emit ldp">;
-
-def FeatureDisableStp : SubtargetFeature<"disable-stp", "HasDisableStp",
-    "true", "Do not emit stp">;
-
-def FeatureLdpAlignedOnly : SubtargetFeature<"ldp-aligned-only", "HasLdpAlignedOnly",
-    "true", "In order to emit ldp, first check if the load will be aligned to 2 * element_size">;
-
-def FeatureStpAlignedOnly : SubtargetFeature<"stp-aligned-only", "HasStpAlignedOnly",
-    "true", "In order to emit stp, first check if the store will be aligned to 2 * element_size">;
-
-// AArch64 2023 Architecture Extensions (v9.5-A)
-
-def FeatureCPA : SubtargetFeature<"cpa", "HasCPA", "true",
-    "Enable Armv9.5-A Checked Pointer Arithmetic (FEAT_CPA)">;
-
-def FeaturePAuthLR : SubtargetFeature<"pauth-lr", "HasPAuthLR",
-    "true", "Enable Armv9.5-A PAC enhancements (FEAT_PAuth_LR)">;
-
-def FeatureTLBIW : SubtargetFeature<"tlbiw", "HasTLBIW", "true",
-  "Enable ARMv9.5-A TLBI VMALL for Dirty State (FEAT_TLBIW)">;
-
+// Subtarget features.
 //===----------------------------------------------------------------------===//
-// Architectures.
-//
-def HasV8_0aOps : SubtargetFeature<"v8a", "HasV8_0aOps", "true",
-  "Support ARM v8.0a instructions", [FeatureEL2VMSA, FeatureEL3]>;
-
-def HasV8_1aOps : SubtargetFeature<"v8.1a", "HasV8_1aOps", "true",
-  "Support ARM v8.1a instructions", [HasV8_0aOps, FeatureCRC, FeatureLSE,
-  FeatureRDM, FeaturePAN, FeatureLOR, FeatureVH]>;
-
-def HasV8_2aOps : SubtargetFeature<"v8.2a", "HasV8_2aOps", "true",
-  "Support ARM v8.2a instructions", [HasV8_1aOps, FeaturePsUAO,
-  FeaturePAN_RWV, FeatureRAS, FeatureCCPP]>;
-
-def HasV8_3aOps : SubtargetFeature<"v8.3a", "HasV8_3aOps", "true",
-  "Support ARM v8.3a instructions", [HasV8_2aOps, FeatureRCPC, FeaturePAuth,
-  FeatureJS, FeatureCCIDX, FeatureComplxNum]>;
-
-def HasV8_4aOps : SubtargetFeature<"v8.4a", "HasV8_4aOps", "true",
-  "Support ARM v8.4a instructions", [HasV8_3aOps, FeatureDotProd,
-  FeatureNV, FeatureMPAM, FeatureDIT,
-  FeatureTRACEV8_4, FeatureAM, FeatureSEL2, FeatureTLB_RMI,
-  FeatureFlagM, FeatureRCPC_IMMO, FeatureLSE2]>;
-
-def HasV8_5aOps : SubtargetFeature<
-  "v8.5a", "HasV8_5aOps", "true", "Support ARM v8.5a instructions",
-  [HasV8_4aOps, FeatureAltFPCmp, FeatureFRInt3264, FeatureSpecRestrict,
-   FeatureSSBS, FeatureSB, FeaturePredRes, FeatureCacheDeepPersist,
-   FeatureBranchTargetId]>;
-
-def HasV8_6aOps : SubtargetFeature<
-  "v8.6a", "HasV8_6aOps", "true", "Support ARM v8.6a instructions",
-  [HasV8_5aOps, FeatureAMVS, FeatureBF16, FeatureFineGrainedTraps,
-   FeatureEnhancedCounterVirtualization, FeatureMatMulInt8]>;
-
-def HasV8_7aOps : SubtargetFeature<
-  "v8.7a", "HasV8_7aOps", "true", "Support ARM v8.7a instructions",
-  [HasV8_6aOps, FeatureXS, FeatureWFxT, FeatureHCX]>;
-
-def HasV8_8aOps : SubtargetFeature<
-  "v8.8a", "HasV8_8aOps", "true", "Support ARM v8.8a instructions",
-  [HasV8_7aOps, FeatureHBC, FeatureMOPS, FeatureNMI]>;
-
-def HasV8_9aOps : SubtargetFeature<
-  "v8.9a", "HasV8_9aOps", "true", "Support ARM v8.9a instructions",
-  [HasV8_8aOps, FeatureCLRBHB, FeaturePRFM_SLC, FeatureSPECRES2,
-   FeatureCSSC, FeatureRASv2, FeatureCHK]>;
-
-def HasV9_0aOps : SubtargetFeature<
-  "v9a", "HasV9_0aOps", "true", "Support ARM v9a instructions",
-  [HasV8_5aOps, FeatureMEC, FeatureSVE2]>;
-
-def HasV9_1aOps : SubtargetFeature<
-  "v9.1a", "HasV9_1aOps", "true", "Support ARM v9.1a instructions",
-  [HasV8_6aOps, HasV9_0aOps]>;
-
-def HasV9_2aOps : SubtargetFeature<
-  "v9.2a", "HasV9_2aOps", "true", "Support ARM v9.2a instructions",
-  [HasV8_7aOps, HasV9_1aOps]>;
-
-def HasV9_3aOps : SubtargetFeature<
-  "v9.3a", "HasV9_3aOps", "true", "Support ARM v9.3a instructions",
-  [HasV8_8aOps, HasV9_2aOps]>;
-
-def HasV9_4aOps : SubtargetFeature<
-  "v9.4a", "HasV9_4aOps", "true", "Support ARM v9.4a instructions",
-  [HasV8_9aOps, HasV9_3aOps]>;
-
-def HasV9_5aOps : SubtargetFeature<
-  "v9.5a", "HasV9_5aOps", "true", "Support ARM v9.5a instructions",
-  [HasV9_4aOps, FeatureCPA]>;
-
-def HasV8_0rOps : SubtargetFeature<
-  "v8r", "HasV8_0rOps", "true", "Support ARM v8r instructions",
-  [//v8.1
-  FeatureCRC, FeaturePAN, FeatureLSE, FeatureCONTEXTIDREL2,
-  //v8.2
-  FeatureRAS, FeaturePsUAO, FeatureCCPP, FeaturePAN_RWV,
-  //v8.3
-  FeatureCCIDX, FeaturePAuth, FeatureRCPC,
-  //v8.4
-  FeatureTRACEV8_4, FeatureTLB_RMI, FeatureFlagM, FeatureDIT, FeatureSEL2,
-  FeatureRCPC_IMMO,
-  // Not mandatory in v8.0-R, but included here on the grounds that it
-  // only enables names of system registers
-  FeatureSpecRestrict
-  ]>;
-
-// Only intended to be used by disassemblers.
-def FeatureAll
-    : SubtargetFeature<"all", "IsAll", "true", "Enable all instructions", []>;
-
-class AssemblerPredicateWithAll<dag cond, string name="">
-    : AssemblerPredicate<(any_of FeatureAll, cond), name>;
+include "AArch64Features.td"
 
 //===----------------------------------------------------------------------===//
 // Register File Description
@@ -766,30 +48,6 @@ def AArch64InstrInfo : InstrInfo;
 include "AArch64SystemOperands.td"
 
 //===----------------------------------------------------------------------===//
-// Access to privileged registers
-//===----------------------------------------------------------------------===//
-
-foreach i = 1-3 in
-def FeatureUseEL#i#ForTP : SubtargetFeature<"tpidr-el"#i, "UseEL"#i#"ForTP",
-  "true", "Permit use of TPIDR_EL"#i#" for the TLS base">;
-def FeatureUseROEL0ForTP : SubtargetFeature<"tpidrro-el0", "UseROEL0ForTP",
-  "true", "Permit use of TPIDRRO_EL0 for the TLS base">;
-
-//===----------------------------------------------------------------------===//
-// Control codegen mitigation against Straight Line Speculation vulnerability.
-//===----------------------------------------------------------------------===//
-
-def FeatureHardenSlsRetBr : SubtargetFeature<"harden-sls-retbr",
-  "HardenSlsRetBr", "true",
-  "Harden against straight line speculation across RET and BR instructions">;
-def FeatureHardenSlsBlr : SubtargetFeature<"harden-sls-blr",
-  "HardenSlsBlr", "true",
-  "Harden against straight line speculation across BLR instructions">;
-def FeatureHardenSlsNoComdat : SubtargetFeature<"harden-sls-nocomdat",
-  "HardenSlsNoComdat", "true",
-  "Generate thunk code for SLS mitigation in the normal text section">;
-
-//===----------------------------------------------------------------------===//
 // AArch64 Processors supported.
 //
 
@@ -818,7 +76,7 @@ def SME2p1Unsupported : AArch64Unsupported;
 
 def SME2Unsupported : AArch64Unsupported {
   let F = !listconcat([HasSME2, HasSVE2orSME2, HasSVE2p1_or_HasSME2, HasSSVE_FP8FMA,
-                      HasSMEF8F16, HasSMEF8F32],
+                      HasSMEF8F16, HasSMEF8F32, HasSMEF16F16orSMEF8F16],
                       SME2p1Unsupported.F);
 }
 
@@ -852,920 +110,7 @@ include "AArch64SchedNeoverseN2.td"
 include "AArch64SchedNeoverseV1.td"
 include "AArch64SchedNeoverseV2.td"
 
-def TuneA35     : SubtargetFeature<"a35", "ARMProcFamily", "CortexA35",
-                                "Cortex-A35 ARM processors">;
-
-def TuneA53     : SubtargetFeature<"a53", "ARMProcFamily", "CortexA53",
-                                   "Cortex-A53 ARM processors", [
-                                   FeatureFuseAES,
-                                   FeatureFuseAdrpAdd,
-                                   FeatureBalanceFPOps,
-                                   FeaturePostRAScheduler]>;
-
-def TuneA55     : SubtargetFeature<"a55", "ARMProcFamily", "CortexA55",
-                                   "Cortex-A55 ARM processors", [
-                                   FeatureFuseAES,
-                                   FeatureFuseAdrpAdd,
-                                   FeaturePostRAScheduler,
-                                   FeatureFuseAddress]>;
-
-def TuneA510    : SubtargetFeature<"a510", "ARMProcFamily", "CortexA510",
-                                   "Cortex-A510 ARM processors", [
-                                   FeatureFuseAES,
-                                   FeatureFuseAdrpAdd,
-                                   FeaturePostRAScheduler
-                                   ]>;
-
-def TuneA520    : SubtargetFeature<"a520", "ARMProcFamily", "CortexA520",
-                                   "Cortex-A520 ARM processors", [
-                                   FeatureFuseAES,
-                                   FeatureFuseAdrpAdd,
-                                   FeaturePostRAScheduler]>;
-
-def TuneA520AE  : SubtargetFeature<"a520ae", "ARMProcFamily", "CortexA520",
-                                   "Cortex-A520AE ARM processors", [
-                                   FeatureFuseAES,
-                                   FeatureFuseAdrpAdd,
-                                   FeaturePostRAScheduler]>;
-
-def TuneA57     : SubtargetFeature<"a57", "ARMProcFamily", "CortexA57",
-                                   "Cortex-A57 ARM processors", [
-                                   FeatureFuseAES,
-                                   FeatureBalanceFPOps,
-                                   FeatureFuseAdrpAdd,
-                                   FeatureFuseLiterals,
-                                   FeatureAddrLSLSlow14,
-                                   FeaturePostRAScheduler,
-                                   FeatureEnableSelectOptimize,
-                                   FeaturePredictableSelectIsExpensive]>;
-
-def TuneA65     : SubtargetFeature<"a65", "ARMProcFamily", "CortexA65",
-                                   "Cortex-A65 ARM processors", [
-                                   FeatureFuseAES,
-                                   FeatureFuseAddress,
-                                   FeatureFuseAdrpAdd,
-                                   FeatureFuseLiterals,
-                                   FeatureEnableSelectOptimize,
-                                   FeaturePredictableSelectIsExpensive]>;
-
-def TuneA72     : SubtargetFeature<"a72", "ARMProcFamily", "CortexA72",
-                                   "Cortex-A72 ARM processors", [
-                                   FeatureFuseAES,
-                                   FeatureFuseAdrpAdd,
-                                   FeatureFuseLiterals,
-                                   FeatureAddrLSLSlow14,
-                                   FeatureEnableSelectOptimize,
-                                   FeaturePredictableSelectIsExpensive]>;
-
-def TuneA73     : SubtargetFeature<"a73", "ARMProcFamily", "CortexA73",
-                                   "Cortex-A73 ARM processors", [
-                                   FeatureFuseAES,
-                                   FeatureFuseAdrpAdd,
-                                   FeatureAddrLSLSlow14,
-                                   FeatureEnableSelectOptimize,
-                                   FeaturePredictableSelectIsExpensive]>;
-
-def TuneA75     : SubtargetFeature<"a75", "ARMProcFamily", "CortexA75",
-                                   "Cortex-A75 ARM processors", [
-                                   FeatureFuseAES,
-                                   FeatureFuseAdrpAdd,
-                                   FeatureAddrLSLSlow14,
-                                   FeatureEnableSelectOptimize,
-                                   FeaturePredictableSelectIsExpensive]>;
-
-def TuneA76     : SubtargetFeature<"a76", "ARMProcFamily", "CortexA76",
-                                   "Cortex-A76 ARM processors", [
-                                   FeatureFuseAES,
-                                   FeatureFuseAdrpAdd,
-                                   FeatureAddrLSLSlow14,
-                                   FeatureALULSLFast,
-                                   FeatureEnableSelectOptimize,
-                                   FeaturePredictableSelectIsExpensive]>;
-
-def TuneA77     : SubtargetFeature<"a77", "ARMProcFamily", "CortexA77",
-                                   "Cortex-A77 ARM processors", [
-                                   FeatureCmpBccFusion,
-                                   FeatureFuseAES,
-                                   FeatureFuseAdrpAdd,
-                                   FeatureAddrLSLSlow14,
-                                   FeatureALULSLFast,
-                                   FeatureEnableSelectOptimize,
-                                   FeaturePredictableSelectIsExpensive]>;
-
-def TuneA78 : SubtargetFeature<"a78", "ARMProcFamily", "CortexA78",
-                               "Cortex-A78 ARM processors", [
-                               FeatureCmpBccFusion,
-                               FeatureFuseAES,
-                               FeatureFuseAdrpAdd,
-                               FeatureAddrLSLSlow14,
-                               FeatureALULSLFast,
-                               FeaturePostRAScheduler,
-                               FeatureEnableSelectOptimize,
-                               FeaturePredictableSelectIsExpensive]>;
-
-def TuneA78AE : SubtargetFeature<"a78ae", "ARMProcFamily",
-                                 "CortexA78AE",
-                                 "Cortex-A78AE ARM processors", [
-                                 FeatureCmpBccFusion,
-                                 FeatureFuseAES,
-                                 FeatureFuseAdrpAdd,
-                                 FeatureAddrLSLSlow14,
-                                 FeatureALULSLFast,
-                                 FeaturePostRAScheduler,
-                                 FeatureEnableSelectOptimize,
-                                 FeaturePredictableSelectIsExpensive]>;
-
-def TuneA78C : SubtargetFeature<"a78c", "ARMProcFamily",
-                                "CortexA78C",
-                                "Cortex-A78C ARM processors", [
-                                FeatureCmpBccFusion,
-                                FeatureFuseAES,
-                                FeatureFuseAdrpAdd,
-                                FeatureAddrLSLSlow14,
-                                FeatureALULSLFast,
-                                FeaturePostRAScheduler,
-                                FeatureEnableSelectOptimize,
-                                FeaturePredictableSelectIsExpensive]>;
-
-def TuneA710    : SubtargetFeature<"a710", "ARMProcFamily", "CortexA710",
-                                   "Cortex-A710 ARM processors", [
-                                   FeatureCmpBccFusion,
-                                   FeatureFuseAES,
-                                   FeatureFuseAdrpAdd,
-                                   FeatureALULSLFast,
-                                   FeaturePostRAScheduler,
-                                   FeatureEnableSelectOptimize,
-                                   FeaturePredictableSelectIsExpensive]>;
-
-def TuneA715 : SubtargetFeature<"a715", "ARMProcFamily", "CortexA715",
-                                 "Cortex-A715 ARM processors", [
-                                 FeatureFuseAES,
-                                 FeaturePostRAScheduler,
-                                 FeatureCmpBccFusion,
-                                 FeatureALULSLFast,
-                                 FeatureFuseAdrpAdd,
-                                 FeatureEnableSelectOptimize,
-                                 FeaturePredictableSelectIsExpensive]>;
-
-def TuneA720 : SubtargetFeature<"a720", "ARMProcFamily", "CortexA720",
-                                "Cortex-A720 ARM processors", [
-                                 FeatureFuseAES,
-                                 FeaturePostRAScheduler,
-                                 FeatureCmpBccFusion,
-                                 FeatureALULSLFast,
-                                 FeatureFuseAdrpAdd,
-                                 FeatureEnableSelectOptimize,
-                                 FeaturePredictableSelectIsExpensive]>;
-
-def TuneA720AE : SubtargetFeature<"a720ae", "ARMProcFamily", "CortexA720",
-                                "Cortex-A720AE ARM processors", [
-                                 FeatureFuseAES,
-                                 FeaturePostRAScheduler,
-                                 FeatureCmpBccFusion,
-                                 FeatureALULSLFast,
-                                 FeatureFuseAdrpAdd,
-                                 FeatureEnableSelectOptimize,
-                                 FeaturePredictableSelectIsExpensive]>;
-
-def TuneR82 : SubtargetFeature<"cortex-r82", "ARMProcFamily",
-                               "CortexR82",
-                               "Cortex-R82 ARM processors", [
-                               FeaturePostRAScheduler]>;
-
-def TuneX1 : SubtargetFeature<"cortex-x1", "ARMProcFamily", "CortexX1",
-                                  "Cortex-X1 ARM processors", [
-                                  FeatureCmpBccFusion,
-                                  FeatureFuseAES,
-                                  FeatureFuseAdrpAdd,
-                                  FeatureAddrLSLSlow14,
-                                  FeatureALULSLFast,
-                                  FeaturePostRAScheduler,
-                                  FeatureEnableSelectOptimize,
-                                  FeaturePredictableSelectIsExpensive]>;
-
-def TuneX2 : SubtargetFeature<"cortex-x2", "ARMProcFamily", "CortexX2",
-                                  "Cortex-X2 ARM processors", [
-                                  FeatureCmpBccFusion,
-                                  FeatureFuseAES,
-                                  FeatureFuseAdrpAdd,
-                                  FeatureALULSLFast,
-                                  FeaturePostRAScheduler,
-                                  FeatureEnableSelectOptimize,
-                                  FeaturePredictableSelectIsExpensive]>;
-
-def TuneX3 : SubtargetFeature<"cortex-x3", "ARMProcFamily", "CortexX3",
-                              "Cortex-X3 ARM processors", [
-                               FeatureALULSLFast,
-                               FeatureFuseAdrpAdd,
-                               FeatureFuseAES,
-                               FeaturePostRAScheduler,
-                               FeatureEnableSelectOptimize,
-                               FeaturePredictableSelectIsExpensive]>;
-
-def TuneX4 : SubtargetFeature<"cortex-x4", "ARMProcFamily", "CortexX4",
-                              "Cortex-X4 ARM processors", [
-                               FeatureALULSLFast,
-                               FeatureFuseAdrpAdd,
-                               FeatureFuseAES,
-                               FeaturePostRAScheduler,
-                               FeatureEnableSelectOptimize,
-                               FeaturePredictableSelectIsExpensive]>;
-
-def TuneA64FX : SubtargetFeature<"a64fx", "ARMProcFamily", "A64FX",
-                                 "Fujitsu A64FX processors", [
-                                 FeaturePostRAScheduler,
-                                 FeatureAggressiveFMA,
-                                 FeatureArithmeticBccFusion,
-                                 FeatureStorePairSuppress,
-                                 FeaturePredictableSelectIsExpensive]>;
-
-def TuneCarmel : SubtargetFeature<"carmel", "ARMProcFamily", "Carmel",
-                                  "Nvidia Carmel processors">;
-
-// Note that cyclone does not fuse AES instructions, but newer apple chips do
-// perform the fusion and cyclone is used by default when targetting apple OSes.
-def TuneAppleA7  : SubtargetFeature<"apple-a7", "ARMProcFamily", "AppleA7",
-                                    "Apple A7 (the CPU formerly known as Cyclone)", [
-                                    FeatureAlternateSExtLoadCVTF32Pattern,
-                                    FeatureArithmeticBccFusion,
-                                    FeatureArithmeticCbzFusion,
-                                    FeatureDisableLatencySchedHeuristic,
-                                    FeatureFuseAES, FeatureFuseCryptoEOR,
-                                    FeatureStorePairSuppress,
-                                    FeatureZCRegMove,
-                                    FeatureZCZeroing,
-                                    FeatureZCZeroingFPWorkaround]>;
-
-def TuneAppleA10 : SubtargetFeature<"apple-a10", "ARMProcFamily", "AppleA10",
-                                    "Apple A10", [
-                                    FeatureAlternateSExtLoadCVTF32Pattern,
-                                    FeatureArithmeticBccFusion,
-                                    FeatureArithmeticCbzFusion,
-                                    FeatureDisableLatencySchedHeuristic,
-                                    FeatureFuseAES,
-                                    FeatureFuseCryptoEOR,
-                                    FeatureStorePairSuppress,
-                                    FeatureZCRegMove,
-                                    FeatureZCZeroing]>;
-
-def TuneAppleA11 : SubtargetFeature<"apple-a11", "ARMProcFamily", "AppleA11",
-                                    "Apple A11", [
-                                    FeatureAlternateSExtLoadCVTF32Pattern,
-                                    FeatureArithmeticBccFusion,
-                                    FeatureArithmeticCbzFusion,
-                                    FeatureDisableLatencySchedHeuristic,
-                                    FeatureFuseAES,
-                                    FeatureFuseCryptoEOR,
-                                    FeatureStorePairSuppress,
-                                    FeatureZCRegMove,
-                                    FeatureZCZeroing]>;
-
-def TuneAppleA12 : SubtargetFeature<"apple-a12", "ARMProcFamily", "AppleA12",
-                                    "Apple A12", [
-                                    FeatureAlternateSExtLoadCVTF32Pattern,
-                                    FeatureArithmeticBccFusion,
-                                    FeatureArithmeticCbzFusion,
-                                    FeatureDisableLatencySchedHeuristic,
-                                    FeatureFuseAES,
-                                    FeatureFuseCryptoEOR,
-                                    FeatureStorePairSuppress,
-                                    FeatureZCRegMove,
-                                    FeatureZCZeroing]>;
-
-def TuneAppleA13 : SubtargetFeature<"apple-a13", "ARMProcFamily", "AppleA13",
-                                    "Apple A13", [
-                                    FeatureAlternateSExtLoadCVTF32Pattern,
-                                    FeatureArithmeticBccFusion,
-                                    FeatureArithmeticCbzFusion,
-                                    FeatureDisableLatencySchedHeuristic,
-                                    FeatureFuseAES,
-                                    FeatureFuseCryptoEOR,
-                                    FeatureStorePairSuppress,
-                                    FeatureZCRegMove,
-                                    FeatureZCZeroing]>;
-
-def TuneAppleA14 : SubtargetFeature<"apple-a14", "ARMProcFamily", "AppleA14",
-                                    "Apple A14", [
-                                    FeatureAggressiveFMA,
-                                    FeatureAlternateSExtLoadCVTF32Pattern,
-                                    FeatureArithmeticBccFusion,
-                                    FeatureArithmeticCbzFusion,
-                                    FeatureDisableLatencySchedHeuristic,
-                                    FeatureFuseAddress,
-                                    FeatureFuseAES,
-                                    FeatureFuseArithmeticLogic,
-                                    FeatureFuseCCSelect,
-                                    FeatureFuseCryptoEOR,
-                                    FeatureFuseLiterals,
-                                    FeatureStorePairSuppress,
-                                    FeatureZCRegMove,
-                                    FeatureZCZeroing]>;
-
-def TuneAppleA15 : SubtargetFeature<"apple-a15", "ARMProcFamily", "AppleA15",
-                                    "Apple A15", [
-                                    FeatureAlternateSExtLoadCVTF32Pattern,
-                                    FeatureArithmeticBccFusion,
-                                    FeatureArithmeticCbzFusion,
-                                    FeatureDisableLatencySchedHeuristic,
-                                    FeatureFuseAddress,
-                                    FeatureFuseAES,
-                                    FeatureFuseArithmeticLogic,
-                                    FeatureFuseCCSelect,
-                                    FeatureFuseCryptoEOR,
-                                    FeatureFuseLiterals,
-                                    FeatureStorePairSuppress,
-                                    FeatureZCRegMove,
-                                    FeatureZCZeroing]>;
-
-def TuneAppleA16 : SubtargetFeature<"apple-a16", "ARMProcFamily", "AppleA16",
-                                    "Apple A16", [
-                                    FeatureAlternateSExtLoadCVTF32Pattern,
-                                    FeatureArithmeticBccFusion,
-                                    FeatureArithmeticCbzFusion,
-                                    FeatureDisableLatencySchedHeuristic,
-                                    FeatureFuseAddress,
-                                    FeatureFuseAdrpAdd,
-                                    FeatureFuseAES,
-                                    FeatureFuseArithmeticLogic,
-                                    FeatureFuseCCSelect,
-                                    FeatureFuseCryptoEOR,
-                                    FeatureFuseLiterals,
-                                    FeatureStorePairSuppress,
-                                    FeatureZCRegMove,
-                                    FeatureZCZeroing]>;
-
-def TuneAppleA17 : SubtargetFeature<"apple-a17", "ARMProcFamily", "AppleA17",
-                                    "Apple A17", [
-                                    FeatureAlternateSExtLoadCVTF32Pattern,
-                                    FeatureArithmeticBccFusion,
-                                    FeatureArithmeticCbzFusion,
-                                    FeatureDisableLatencySchedHeuristic,
-                                    FeatureFuseAddress,
-                                    FeatureFuseAdrpAdd,
-                                    FeatureFuseAES,
-                                    FeatureFuseArithmeticLogic,
-                                    FeatureFuseCCSelect,
-                                    FeatureFuseCryptoEOR,
-                                    FeatureFuseLiterals,
-                                    FeatureStorePairSuppress,
-                                    FeatureZCRegMove,
-                                    FeatureZCZeroing]>;
-
-def TuneExynosM3 : SubtargetFeature<"exynosm3", "ARMProcFamily", "ExynosM3",
-                                    "Samsung Exynos-M3 processors",
-                                    [FeatureExynosCheapAsMoveHandling,
-                                     FeatureForce32BitJumpTables,
-                                     FeatureFuseAddress,
-                                     FeatureFuseAES,
-                                     FeatureFuseCCSelect,
-                                     FeatureFuseAdrpAdd,
-                                     FeatureFuseLiterals,
-                                     FeatureStorePairSuppress,
-                                     FeatureALULSLFast,
-                                     FeaturePostRAScheduler,
-                                     FeaturePredictableSelectIsExpensive]>;
-
-// Re-uses some scheduling and tunings from the ExynosM3 proc family.
-def TuneExynosM4 : SubtargetFeature<"exynosm4", "ARMProcFamily", "ExynosM3",
-                                    "Samsung Exynos-M4 processors",
-                                    [FeatureArithmeticBccFusion,
-                                     FeatureArithmeticCbzFusion,
-                                     FeatureExynosCheapAsMoveHandling,
-                                     FeatureForce32BitJumpTables,
-                                     FeatureFuseAddress,
-                                     FeatureFuseAES,
-                                     FeatureFuseArithmeticLogic,
-                                     FeatureFuseCCSelect,
-                                     FeatureFuseAdrpAdd,
-                                     FeatureFuseLiterals,
-                                     FeatureStorePairSuppress,
-                                     FeatureALULSLFast,
-                                     FeaturePostRAScheduler,
-                                     FeatureZCZeroing]>;
-
-def TuneKryo    : SubtargetFeature<"kryo", "ARMProcFamily", "Kryo",
-                                   "Qualcomm Kryo processors", [
-                                   FeaturePostRAScheduler,
-                                   FeaturePredictableSelectIsExpensive,
-                                   FeatureZCZeroing,
-                                   FeatureALULSLFast,
-                                   FeatureStorePairSuppress]>;
-
-def TuneFalkor  : SubtargetFeature<"falkor", "ARMProcFamily", "Falkor",
-                                   "Qualcomm Falkor processors", [
-                                   FeaturePostRAScheduler,
-                                   FeaturePredictableSelectIsExpensive,
-                                   FeatureZCZeroing,
-                                   FeatureStorePairSuppress,
-                                   FeatureALULSLFast,
-                                   FeatureSlowSTRQro]>;
-
-def TuneNeoverseE1 : SubtargetFeature<"neoversee1", "ARMProcFamily", "NeoverseE1",
-                                      "Neoverse E1 ARM processors", [
-                                      FeatureFuseAES,
-                                      FeatureFuseAdrpAdd,
-                                      FeaturePostRAScheduler]>;
-
-def TuneNeoverseN1 : SubtargetFeature<"neoversen1", "ARMProcFamily", "NeoverseN1",
-                                      "Neoverse N1 ARM processors", [
-                                      FeatureFuseAES,
-                                      FeatureFuseAdrpAdd,
-                                      FeatureAddrLSLSlow14,
-                                      FeatureALULSLFast,
-                                      FeaturePostRAScheduler,
-                                      FeatureEnableSelectOptimize,
-                                      FeaturePredictableSelectIsExpensive]>;
-
-def TuneNeoverseN2 : SubtargetFeature<"neoversen2", "ARMProcFamily", "NeoverseN2",
-                                      "Neoverse N2 ARM processors", [
-                                      FeatureFuseAES,
-                                      FeatureFuseAdrpAdd,
-                                      FeatureALULSLFast,
-                                      FeaturePostRAScheduler,
-                                      FeatureEnableSelectOptimize,
-                                      FeaturePredictableSelectIsExpensive]>;
-
-def TuneNeoverse512TVB : SubtargetFeature<"neoverse512tvb", "ARMProcFamily", "Neoverse512TVB",
-                                      "Neoverse 512-TVB ARM processors", [
-                                      FeatureFuseAES,
-                                      FeatureFuseAdrpAdd,
-                                      FeatureALULSLFast,
-                                      FeaturePostRAScheduler,
-                                      FeatureEnableSelectOptimize,
-                                      FeaturePredictableSelectIsExpensive]>;
-
-def TuneNeoverseV1 : SubtargetFeature<"neoversev1", "ARMProcFamily", "NeoverseV1",
-                                      "Neoverse V1 ARM processors", [
-                                      FeatureFuseAES,
-                                      FeatureFuseAdrpAdd,
-                                      FeatureAddrLSLSlow14,
-                                      FeatureALULSLFast,
-                                      FeaturePostRAScheduler,
-                                      FeatureEnableSelectOptimize,
-                                      FeaturePredictableSelectIsExpensive,
-                                      FeatureNoSVEFPLD1R]>;
-
-def TuneNeoverseV2 : SubtargetFeature<"neoversev2", "ARMProcFamily", "NeoverseV2",
-                                      "Neoverse V2 ARM processors", [
-                                      FeatureFuseAES,
-                                      FeatureFuseAdrpAdd,
-                                      FeatureALULSLFast,
-                                      FeaturePostRAScheduler,
-                                      FeatureEnableSelectOptimize,
-                                      FeaturePredictableSelectIsExpensive]>;
-
-def TuneSaphira  : SubtargetFeature<"saphira", "ARMProcFamily", "Saphira",
-                                   "Qualcomm Saphira processors", [
-                                   FeaturePostRAScheduler,
-                                   FeaturePredictableSelectIsExpensive,
-                                   FeatureZCZeroing,
-                                   FeatureStorePairSuppress,
-                                   FeatureALULSLFast]>;
-
-def TuneThunderX2T99  : SubtargetFeature<"thunderx2t99", "ARMProcFamily", "ThunderX2T99",
-                                         "Cavium ThunderX2 processors", [
-                                          FeatureAggressiveFMA,
-                                          FeatureArithmeticBccFusion,
-                                          FeaturePostRAScheduler,
-                                          FeatureStorePairSuppress,
-                                          FeaturePredictableSelectIsExpensive]>;
-
-def TuneThunderX3T110  : SubtargetFeature<"thunderx3t110", "ARMProcFamily",
-                                          "ThunderX3T110",
-                                          "Marvell ThunderX3 processors", [
-                                           FeatureAggressiveFMA,
-                                           FeatureArithmeticBccFusion,
-                                           FeaturePostRAScheduler,
-                                           FeaturePredictableSelectIsExpensive,
-                                           FeatureBalanceFPOps,
-                                           FeatureStorePairSuppress,
-                                           FeatureStrictAlign]>;
-
-def TuneThunderX : SubtargetFeature<"thunderx", "ARMProcFamily", "ThunderX",
-                                    "Cavium ThunderX processors", [
-                                    FeaturePostRAScheduler,
-                                    FeatureStorePairSuppress,
-                                    FeaturePredictableSelectIsExpensive]>;
-
-def TuneThunderXT88 : SubtargetFeature<"thunderxt88", "ARMProcFamily",
-                                       "ThunderXT88",
-                                       "Cavium ThunderX processors", [
-                                       FeaturePostRAScheduler,
-                                       FeatureStorePairSuppress,
-                                       FeaturePredictableSelectIsExpensive]>;
-
-def TuneThunderXT81 : SubtargetFeature<"thunderxt81", "ARMProcFamily",
-                                       "ThunderXT81",
-                                       "Cavium ThunderX processors", [
-                                       FeaturePostRAScheduler,
-                                       FeatureStorePairSuppress,
-                                       FeaturePredictableSelectIsExpensive]>;
-
-def TuneThunderXT83 : SubtargetFeature<"thunderxt83", "ARMProcFamily",
-                                       "ThunderXT83",
-                                       "Cavium ThunderX processors", [
-                                       FeaturePostRAScheduler,
-                                       FeatureStorePairSuppress,
-                                       FeaturePredictableSelectIsExpensive]>;
-
-def TuneTSV110 : SubtargetFeature<"tsv110", "ARMProcFamily", "TSV110",
-                                  "HiSilicon TS-V110 processors", [
-                                  FeatureFuseAES,
-                                  FeatureStorePairSuppress,
-                                  FeaturePostRAScheduler]>;
-
-def TuneAmpere1 : SubtargetFeature<"ampere1", "ARMProcFamily", "Ampere1",
-                                   "Ampere Computing Ampere-1 processors", [
-                                   FeaturePostRAScheduler,
-                                   FeatureFuseAES,
-                                   FeatureFuseAdrpAdd,
-                                   FeatureALULSLFast,
-                                   FeatureAggressiveFMA,
-                                   FeatureArithmeticBccFusion,
-                                   FeatureCmpBccFusion,
-                                   FeatureFuseAddress,
-                                   FeatureFuseLiterals,
-                                   FeatureStorePairSuppress,
-                                   FeatureLdpAlignedOnly,
-                                   FeatureStpAlignedOnly]>;
-
-def TuneAmpere1A : SubtargetFeature<"ampere1a", "ARMProcFamily", "Ampere1A",
-                                    "Ampere Computing Ampere-1A processors", [
-                                    FeaturePostRAScheduler,
-                                    FeatureFuseAES,
-                                    FeatureFuseAdrpAdd,
-                                    FeatureALULSLFast,
-                                    FeatureAggressiveFMA,
-                                    FeatureArithmeticBccFusion,
-                                    FeatureCmpBccFusion,
-                                    FeatureFuseAddress,
-                                    FeatureFuseLiterals,
-                                    FeatureFuseAddSub2RegAndConstOne,
-                                    FeatureStorePairSuppress,
-                                    FeatureLdpAlignedOnly,
-                                    FeatureStpAlignedOnly]>;
-
-def TuneAmpere1B : SubtargetFeature<"ampere1b", "ARMProcFamily", "Ampere1B",
-                                    "Ampere Computing Ampere-1B processors", [
-                                    FeaturePostRAScheduler,
-                                    FeatureFuseAES,
-                                    FeatureFuseAdrpAdd,
-                                    FeatureALULSLFast,
-                                    FeatureAggressiveFMA,
-                                    FeatureArithmeticBccFusion,
-                                    FeatureCmpBccFusion,
-                                    FeatureFuseAddress,
-                                    FeatureFuseLiterals,
-                                    FeatureStorePairSuppress,
-                                    FeatureEnableSelectOptimize,
-                                    FeaturePredictableSelectIsExpensive,
-                                    FeatureLdpAlignedOnly,
-                                    FeatureStpAlignedOnly]>;
-
-def ProcessorFeatures {
-  list<SubtargetFeature> A53  = [HasV8_0aOps, FeatureCRC, FeatureCrypto,
-                                 FeatureFPARMv8, FeatureNEON, FeaturePerfMon];
-  list<SubtargetFeature> A55  = [HasV8_2aOps, FeatureCrypto, FeatureFPARMv8,
-                                 FeatureNEON, FeatureFullFP16, FeatureDotProd,
-                                 FeatureRCPC, FeaturePerfMon];
-  list<SubtargetFeature> A510 = [HasV9_0aOps, FeatureNEON, FeaturePerfMon,
-                                 FeatureMatMulInt8, FeatureBF16, FeatureAM,
-                                 FeatureMTE, FeatureETE, FeatureSVE2BitPerm,
-                                 FeatureFP16FML];
-  list<SubtargetFeature> A520 = [HasV9_2aOps, FeaturePerfMon, FeatureAM,
-                                 FeatureMTE, FeatureETE, FeatureSVE2BitPerm,
-                                 FeatureFP16FML];
-  list<SubtargetFeature> A520AE = [HasV9_2aOps, FeaturePerfMon, FeatureAM,
-                                 FeatureMTE, FeatureETE, FeatureSVE2BitPerm,
-                                 FeatureFP16FML];
-  list<SubtargetFeature> A65  = [HasV8_2aOps, FeatureCrypto, FeatureFPARMv8,
-                                 FeatureNEON, FeatureFullFP16, FeatureDotProd,
-                                 FeatureRCPC, FeatureSSBS, FeatureRAS,
-                                 FeaturePerfMon];
-  list<SubtargetFeature> A76  = [HasV8_2aOps, FeatureCrypto, FeatureFPARMv8,
-                                 FeatureNEON, FeatureFullFP16, FeatureDotProd,
-                                 FeatureRCPC, FeatureSSBS, FeaturePerfMon];
-  list<SubtargetFeature> A77  = [HasV8_2aOps, FeatureCrypto, FeatureFPARMv8,
-                                 FeatureNEON, FeatureFullFP16, FeatureDotProd,
-                                 FeatureRCPC, FeaturePerfMon, FeatureSSBS];
-  list<SubtargetFeature> A78  = [HasV8_2aOps, FeatureCrypto, FeatureFPARMv8,
-                                 FeatureNEON, FeatureFullFP16, FeatureDotProd,
-                                 FeatureRCPC, FeaturePerfMon, FeatureSPE,
-                                 FeatureSSBS];
-  list<SubtargetFeature> A78AE = [HasV8_2aOps, FeatureCrypto, FeatureFPARMv8,
-                                  FeatureNEON, FeatureFullFP16, FeatureDotProd,
-                                  FeatureRCPC, FeaturePerfMon, FeatureSPE,
-                                  FeatureSSBS];
-  list<SubtargetFeature> A78C = [HasV8_2aOps, FeatureCrypto, FeatureFPARMv8,
-                                 FeatureNEON, FeatureFullFP16, FeatureDotProd,
-                                 FeatureFlagM, FeaturePAuth,
-                                 FeaturePerfMon, FeatureRCPC, FeatureSPE,
-                                 FeatureSSBS];
-  list<SubtargetFeature> A710 = [HasV9_0aOps, FeatureNEON, FeaturePerfMon,
-                                 FeatureETE, FeatureMTE, FeatureFP16FML,
-                                 FeatureSVE2BitPerm, FeatureBF16, FeatureMatMulInt8];
-  list<SubtargetFeature> A715 = [HasV9_0aOps, FeatureNEON, FeatureMTE,
-                                 FeatureFP16FML, FeatureSVE, FeatureTRBE,
-                                 FeatureSVE2BitPerm, FeatureBF16, FeatureETE,
-                                 FeaturePerfMon, FeatureMatMulInt8, FeatureSPE];
-  list<SubtargetFeature> A720 = [HasV9_2aOps, FeatureMTE, FeatureFP16FML,
-                                 FeatureTRBE, FeatureSVE2BitPerm, FeatureETE,
-                                 FeaturePerfMon, FeatureSPE, FeatureSPE_EEF];
-  list<SubtargetFeature> A720AE = [HasV9_2aOps, FeatureMTE, FeatureFP16FML,
-                                 FeatureTRBE, FeatureSVE2BitPerm, FeatureETE,
-                                 FeaturePerfMon, FeatureSPE, FeatureSPE_EEF];
-  list<SubtargetFeature> R82  = [HasV8_0rOps, FeaturePerfMon, FeatureFullFP16,
-                                 FeatureFP16FML, FeatureSSBS, FeaturePredRes,
-                                 FeatureSB, FeatureRDM, FeatureDotProd,
-                                 FeatureComplxNum, FeatureJS];
-  list<SubtargetFeature> X1   = [HasV8_2aOps, FeatureCrypto, FeatureFPARMv8,
-                                 FeatureNEON, FeatureRCPC, FeaturePerfMon,
-                                 FeatureSPE, FeatureFullFP16, FeatureDotProd,
-                                 FeatureSSBS];
-  list<SubtargetFeature> X1C  = [HasV8_2aOps, FeatureCrypto, FeatureFPARMv8,
-                                 FeatureNEON, FeatureRCPC_IMMO, FeaturePerfMon,
-                                 FeatureSPE, FeatureFullFP16, FeatureDotProd,
-                                 FeaturePAuth, FeatureSSBS, FeatureFlagM,
-                                 FeatureLSE2];
-  list<SubtargetFeature> X2   = [HasV9_0aOps, FeatureNEON, FeaturePerfMon,
-                                 FeatureMatMulInt8, FeatureBF16, FeatureAM,
-                                 FeatureMTE, FeatureETE, FeatureSVE2BitPerm,
-                                 FeatureFP16FML];
-  list<SubtargetFeature> X3 =   [HasV9_0aOps, FeatureSVE, FeatureNEON,
-                                 FeaturePerfMon, FeatureETE, FeatureTRBE,
-                                 FeatureSPE, FeatureBF16, FeatureMatMulInt8,
-                                 FeatureMTE, FeatureSVE2BitPerm, FeatureFullFP16,
-                                 FeatureFP16FML];
-  list<SubtargetFeature> X4 =   [HasV9_2aOps,
-                                 FeaturePerfMon, FeatureETE, FeatureTRBE,
-                                 FeatureSPE, FeatureMTE, FeatureSVE2BitPerm,
-                                 FeatureFP16FML, FeatureSPE_EEF];
-  list<SubtargetFeature> A64FX    = [HasV8_2aOps, FeatureFPARMv8, FeatureNEON,
-                                     FeatureSHA2, FeaturePerfMon, FeatureFullFP16,
-                                     FeatureSVE, FeatureComplxNum];
-  list<SubtargetFeature> Carmel   = [HasV8_2aOps, FeatureNEON, FeatureCrypto,
-                                     FeatureFullFP16];
-  list<SubtargetFeature> AppleA7  = [HasV8_0aOps, FeatureCrypto, FeatureFPARMv8,
-                                     FeatureNEON,FeaturePerfMon, FeatureAppleA7SysReg];
-  list<SubtargetFeature> AppleA10 = [HasV8_0aOps, FeatureCrypto, FeatureFPARMv8,
-                                     FeatureNEON, FeaturePerfMon, FeatureCRC,
-                                     FeatureRDM, FeaturePAN, FeatureLOR, FeatureVH];
-  list<SubtargetFeature> AppleA11 = [HasV8_2aOps, FeatureCrypto, FeatureFPARMv8,
-                                     FeatureNEON, FeaturePerfMon, FeatureFullFP16];
-  list<SubtargetFeature> AppleA12 = [HasV8_3aOps, FeatureCrypto, FeatureFPARMv8,
-                                     FeatureNEON, FeaturePerfMon, FeatureFullFP16];
-  list<SubtargetFeature> AppleA13 = [HasV8_4aOps, FeatureCrypto, FeatureFPARMv8,
-                                     FeatureNEON, FeaturePerfMon, FeatureFullFP16,
-                                     FeatureFP16FML, FeatureSHA3];
-  list<SubtargetFeature> AppleA14 = [HasV8_4aOps, FeatureCrypto, FeatureFPARMv8,
-                                     FeatureNEON, FeaturePerfMon, FeatureFRInt3264,
-                                     FeatureSpecRestrict, FeatureSSBS, FeatureSB,
-                                     FeaturePredRes, FeatureCacheDeepPersist,
-                                     FeatureFullFP16, FeatureFP16FML, FeatureSHA3,
-                                     FeatureAltFPCmp];
-  list<SubtargetFeature> AppleA15 = [HasV8_6aOps, FeatureCrypto, FeatureFPARMv8,
-                                     FeatureNEON, FeaturePerfMon, FeatureSHA3,
-                                     FeatureFullFP16, FeatureFP16FML];
-  list<SubtargetFeature> AppleA16 = [HasV8_6aOps, FeatureCrypto, FeatureFPARMv8,
-                                     FeatureNEON, FeaturePerfMon, FeatureSHA3,
-                                     FeatureFullFP16, FeatureFP16FML,
-                                     FeatureHCX];
-  list<SubtargetFeature> AppleA17 = [HasV8_6aOps, FeatureCrypto, FeatureFPARMv8,
-                                     FeatureNEON, FeaturePerfMon, FeatureSHA3,
-                                     FeatureFullFP16, FeatureFP16FML,
-                                     FeatureHCX];
-  list<SubtargetFeature> ExynosM3 = [HasV8_0aOps, FeatureCRC, FeatureCrypto,
-                                     FeaturePerfMon];
-  list<SubtargetFeature> ExynosM4 = [HasV8_2aOps, FeatureCrypto, FeatureDotProd,
-                                     FeatureFullFP16, FeaturePerfMon];
-  list<SubtargetFeature> Falkor   = [HasV8_0aOps, FeatureCRC, FeatureCrypto,
-                                     FeatureFPARMv8, FeatureNEON, FeaturePerfMon,
-                                     FeatureRDM];
-  list<SubtargetFeature> NeoverseE1 = [HasV8_2aOps, FeatureCrypto, FeatureDotProd,
-                                       FeatureFPARMv8, FeatureFullFP16, FeatureNEON,
-                                       FeatureRCPC, FeatureSSBS, FeaturePerfMon];
-  list<SubtargetFeature> NeoverseN1 = [HasV8_2aOps, FeatureCrypto, FeatureDotProd,
-                                       FeatureFPARMv8, FeatureFullFP16, FeatureNEON,
-                                       FeatureRCPC, FeatureSPE, FeatureSSBS,
-                                       FeaturePerfMon];
-  list<SubtargetFeature> NeoverseN2 = [HasV9_0aOps, FeatureBF16, FeatureETE, FeatureFP16FML,
-                                       FeatureMatMulInt8, FeatureMTE, FeatureSVE2,
-                                       FeatureSVE2BitPerm, FeatureTRBE,
-                                       FeaturePerfMon];
-  list<SubtargetFeature> Neoverse512TVB = [HasV8_4aOps, FeatureBF16, FeatureCacheDeepPersist,
-                                           FeatureCrypto, FeatureFPARMv8, FeatureFP16FML,
-                                           FeatureFullFP16, FeatureMatMulInt8, FeatureNEON,
-                                           FeaturePerfMon, FeatureRandGen, FeatureSPE,
-                                           FeatureSSBS, FeatureSVE];
-  list<SubtargetFeature> NeoverseV1 = [HasV8_4aOps, FeatureBF16, FeatureCacheDeepPersist,
-                                       FeatureCrypto, FeatureFPARMv8, FeatureFP16FML,
-                                       FeatureFullFP16, FeatureMatMulInt8, FeatureNEON,
-                                       FeaturePerfMon, FeatureRandGen, FeatureSPE,
-                                       FeatureSSBS, FeatureSVE];
-  list<SubtargetFeature> NeoverseV2 = [HasV9_0aOps, FeatureBF16, FeatureSPE,
-                                       FeaturePerfMon, FeatureETE, FeatureMatMulInt8,
-                                       FeatureNEON, FeatureSVE2BitPerm, FeatureFP16FML,
-                                       FeatureMTE, FeatureRandGen];
-  list<SubtargetFeature> Saphira    = [HasV8_4aOps, FeatureCrypto, FeatureFPARMv8,
-                                       FeatureNEON, FeatureSPE, FeaturePerfMon];
-  list<SubtargetFeature> ThunderX   = [HasV8_0aOps, FeatureCRC, FeatureCrypto,
-                                       FeatureFPARMv8, FeaturePerfMon, FeatureNEON];
-  list<SubtargetFeature> ThunderX2T99  = [HasV8_1aOps, FeatureCRC, FeatureCrypto,
-                                          FeatureFPARMv8, FeatureNEON, FeatureLSE];
-  list<SubtargetFeature> ThunderX3T110 = [HasV8_3aOps, FeatureCRC, FeatureCrypto,
-                                          FeatureFPARMv8, FeatureNEON, FeatureLSE,
-                                          FeaturePAuth, FeaturePerfMon];
-  list<SubtargetFeature> TSV110 = [HasV8_2aOps, FeatureCrypto, FeatureFPARMv8,
-                                   FeatureNEON, FeaturePerfMon, FeatureSPE,
-                                   FeatureFullFP16, FeatureFP16FML, FeatureDotProd,
-                                   FeatureJS, FeatureComplxNum];
-  list<SubtargetFeature> Ampere1 = [HasV8_6aOps, FeatureNEON, FeaturePerfMon,
-                                    FeatureSSBS, FeatureRandGen, FeatureSB,
-                                    FeatureSHA2, FeatureSHA3, FeatureAES];
-  list<SubtargetFeature> Ampere1A = [HasV8_6aOps, FeatureNEON, FeaturePerfMon,
-                                     FeatureMTE, FeatureSSBS, FeatureRandGen,
-                                     FeatureSB, FeatureSM4, FeatureSHA2,
-                                     FeatureSHA3, FeatureAES];
-  list<SubtargetFeature> Ampere1B = [HasV8_7aOps, FeatureNEON, FeaturePerfMon,
-                                     FeatureMTE, FeatureSSBS, FeatureRandGen,
-                                     FeatureSB, FeatureSM4, FeatureSHA2,
-                                     FeatureSHA3, FeatureAES, FeatureCSSC,
-                                     FeatureWFxT, FeatureFullFP16];
-
-  // ETE and TRBE are future architecture extensions. We temporarily enable them
-  // by default for users targeting generic AArch64. The extensions do not
-  // affect code generated by the compiler and can be used only by explicitly
-  // mentioning the new system register names in assembly.
-  list<SubtargetFeature> Generic = [FeatureFPARMv8, FeatureNEON, FeatureETE];
-}
-
-// FeatureFuseAdrpAdd is enabled under Generic to allow linker merging
-// optimizations.
-def : ProcessorModel<"generic", CortexA510Model, ProcessorFeatures.Generic,
-                     [FeatureFuseAES, FeatureFuseAdrpAdd, FeaturePostRAScheduler,
-                      FeatureEnableSelectOptimize]>;
-def : ProcessorModel<"cortex-a35", CortexA53Model, ProcessorFeatures.A53,
-                     [TuneA35]>;
-def : ProcessorModel<"cortex-a34", CortexA53Model, ProcessorFeatures.A53,
-                     [TuneA35]>;
-def : ProcessorModel<"cortex-a53", CortexA53Model, ProcessorFeatures.A53,
-                     [TuneA53]>;
-def : ProcessorModel<"cortex-a55", CortexA55Model, ProcessorFeatures.A55,
-                     [TuneA55]>;
-def : ProcessorModel<"cortex-a510", CortexA510Model, ProcessorFeatures.A510,
-                     [TuneA510]>;
-def : ProcessorModel<"cortex-a520", CortexA510Model, ProcessorFeatures.A520,
-                     [TuneA520]>;
-def : ProcessorModel<"cortex-a520ae", CortexA510Model, ProcessorFeatures.A520AE,
-                     [TuneA520AE]>;
-def : ProcessorModel<"cortex-a57", CortexA57Model, ProcessorFeatures.A53,
-                     [TuneA57]>;
-def : ProcessorModel<"cortex-a65", CortexA53Model, ProcessorFeatures.A65,
-                     [TuneA65]>;
-def : ProcessorModel<"cortex-a65ae", CortexA53Model, ProcessorFeatures.A65,
-                     [TuneA65]>;
-def : ProcessorModel<"cortex-a72", CortexA57Model, ProcessorFeatures.A53,
-                     [TuneA72]>;
-def : ProcessorModel<"cortex-a73", CortexA57Model, ProcessorFeatures.A53,
-                     [TuneA73]>;
-def : ProcessorModel<"cortex-a75", CortexA57Model, ProcessorFeatures.A55,
-                     [TuneA75]>;
-def : ProcessorModel<"cortex-a76", CortexA57Model, ProcessorFeatures.A76,
-                     [TuneA76]>;
-def : ProcessorModel<"cortex-a76ae", CortexA57Model, ProcessorFeatures.A76,
-                     [TuneA76]>;
-def : ProcessorModel<"cortex-a77", CortexA57Model, ProcessorFeatures.A77,
-                     [TuneA77]>;
-def : ProcessorModel<"cortex-a78", CortexA57Model, ProcessorFeatures.A78,
-                     [TuneA78]>;
-def : ProcessorModel<"cortex-a78ae", CortexA57Model, ProcessorFeatures.A78AE,
-                     [TuneA78AE]>;
-def : ProcessorModel<"cortex-a78c", CortexA57Model, ProcessorFeatures.A78C,
-                     [TuneA78C]>;
-def : ProcessorModel<"cortex-a710", NeoverseN2Model, ProcessorFeatures.A710,
-                     [TuneA710]>;
-def : ProcessorModel<"cortex-a715", NeoverseN2Model, ProcessorFeatures.A715,
-                     [TuneA715]>;
-def : ProcessorModel<"cortex-a720", NeoverseN2Model, ProcessorFeatures.A720,
-                     [TuneA720]>;
-def : ProcessorModel<"cortex-a720ae", NeoverseN2Model, ProcessorFeatures.A720AE,
-                     [TuneA720AE]>;
-def : ProcessorModel<"cortex-r82", CortexA55Model, ProcessorFeatures.R82,
-                     [TuneR82]>;
-def : ProcessorModel<"cortex-x1", CortexA57Model, ProcessorFeatures.X1,
-                     [TuneX1]>;
-def : ProcessorModel<"cortex-x1c", CortexA57Model, ProcessorFeatures.X1C,
-                     [TuneX1]>;
-def : ProcessorModel<"cortex-x2", NeoverseN2Model, ProcessorFeatures.X2,
-                     [TuneX2]>;
-def : ProcessorModel<"cortex-x3", NeoverseN2Model, ProcessorFeatures.X3,
-                     [TuneX3]>;
-def : ProcessorModel<"cortex-x4", NeoverseN2Model, ProcessorFeatures.X4,
-                     [TuneX4]>;
-def : ProcessorModel<"neoverse-e1", CortexA53Model,
-                     ProcessorFeatures.NeoverseE1, [TuneNeoverseE1]>;
-def : ProcessorModel<"neoverse-n1", NeoverseN1Model,
-                     ProcessorFeatures.NeoverseN1, [TuneNeoverseN1]>;
-def : ProcessorModel<"neoverse-n2", NeoverseN2Model,
-                     ProcessorFeatures.NeoverseN2, [TuneNeoverseN2]>;
-def : ProcessorModel<"neoverse-512tvb", NeoverseV1Model,
-                     ProcessorFeatures.Neoverse512TVB, [TuneNeoverse512TVB]>;
-def : ProcessorModel<"neoverse-v1", NeoverseV1Model,
-                     ProcessorFeatures.NeoverseV1, [TuneNeoverseV1]>;
-def : ProcessorModel<"neoverse-v2", NeoverseV2Model,
-                     ProcessorFeatures.NeoverseV2, [TuneNeoverseV2]>;
-def : ProcessorModel<"exynos-m3", ExynosM3Model, ProcessorFeatures.ExynosM3,
-                     [TuneExynosM3]>;
-def : ProcessorModel<"exynos-m4", ExynosM4Model, ProcessorFeatures.ExynosM4,
-                     [TuneExynosM4]>;
-def : ProcessorModel<"exynos-m5", ExynosM5Model, ProcessorFeatures.ExynosM4,
-                     [TuneExynosM4]>;
-def : ProcessorModel<"falkor", FalkorModel, ProcessorFeatures.Falkor,
-                     [TuneFalkor]>;
-def : ProcessorModel<"saphira", FalkorModel, ProcessorFeatures.Saphira,
-                     [TuneSaphira]>;
-def : ProcessorModel<"kryo", KryoModel, ProcessorFeatures.A53, [TuneKryo]>;
-
-// Cavium ThunderX/ThunderX T8X  Processors
-def : ProcessorModel<"thunderx", ThunderXT8XModel,  ProcessorFeatures.ThunderX,
-                     [TuneThunderX]>;
-def : ProcessorModel<"thunderxt88", ThunderXT8XModel,
-                     ProcessorFeatures.ThunderX, [TuneThunderXT88]>;
-def : ProcessorModel<"thunderxt81", ThunderXT8XModel,
-                     ProcessorFeatures.ThunderX, [TuneThunderXT81]>;
-def : ProcessorModel<"thunderxt83", ThunderXT8XModel,
-                     ProcessorFeatures.ThunderX, [TuneThunderXT83]>;
-// Cavium ThunderX2T9X  Processors. Formerly Broadcom Vulcan.
-def : ProcessorModel<"thunderx2t99", ThunderX2T99Model,
-                     ProcessorFeatures.ThunderX2T99, [TuneThunderX2T99]>;
-// Marvell ThunderX3T110 Processors.
-def : ProcessorModel<"thunderx3t110", ThunderX3T110Model,
-                     ProcessorFeatures.ThunderX3T110, [TuneThunderX3T110]>;
-def : ProcessorModel<"tsv110", TSV110Model, ProcessorFeatures.TSV110,
-                     [TuneTSV110]>;
-
-// Support cyclone as an alias for apple-a7 so we can still LTO old bitcode.
-def : ProcessorModel<"cyclone", CycloneModel, ProcessorFeatures.AppleA7,
-                     [TuneAppleA7]>;
-
-// iPhone and iPad CPUs
-def : ProcessorModel<"apple-a7", CycloneModel, ProcessorFeatures.AppleA7,
-                     [TuneAppleA7]>;
-def : ProcessorModel<"apple-a8", CycloneModel, ProcessorFeatures.AppleA7,
-                     [TuneAppleA7]>;
-def : ProcessorModel<"apple-a9", CycloneModel, ProcessorFeatures.AppleA7,
-                     [TuneAppleA7]>;
-def : ProcessorModel<"apple-a10", CycloneModel, ProcessorFeatures.AppleA10,
-                     [TuneAppleA10]>;
-def : ProcessorModel<"apple-a11", CycloneModel, ProcessorFeatures.AppleA11,
-                     [TuneAppleA11]>;
-def : ProcessorModel<"apple-a12", CycloneModel, ProcessorFeatures.AppleA12,
-                     [TuneAppleA12]>;
-def : ProcessorModel<"apple-a13", CycloneModel, ProcessorFeatures.AppleA13,
-                     [TuneAppleA13]>;
-def : ProcessorModel<"apple-a14", CycloneModel, ProcessorFeatures.AppleA14,
-                     [TuneAppleA14]>;
-def : ProcessorModel<"apple-a15", CycloneModel, ProcessorFeatures.AppleA15,
-                     [TuneAppleA15]>;
-def : ProcessorModel<"apple-a16", CycloneModel, ProcessorFeatures.AppleA16,
-                     [TuneAppleA16]>;
-def : ProcessorModel<"apple-a17", CycloneModel, ProcessorFeatures.AppleA17,
-                     [TuneAppleA17]>;
-// Mac CPUs
-def : ProcessorModel<"apple-m1", CycloneModel, ProcessorFeatures.AppleA14,
-                     [TuneAppleA14]>;
-def : ProcessorModel<"apple-m2", CycloneModel, ProcessorFeatures.AppleA15,
-                     [TuneAppleA15]>;
-def : ProcessorModel<"apple-m3", CycloneModel, ProcessorFeatures.AppleA16,
-                     [TuneAppleA16]>;
-
-// watch CPUs.
-def : ProcessorModel<"apple-s4", CycloneModel, ProcessorFeatures.AppleA12,
-                     [TuneAppleA12]>;
-def : ProcessorModel<"apple-s5", CycloneModel, ProcessorFeatures.AppleA12,
-                     [TuneAppleA12]>;
-
-// Alias for the latest Apple processor model supported by LLVM.
-def : ProcessorModel<"apple-latest", CycloneModel, ProcessorFeatures.AppleA16,
-                     [TuneAppleA16]>;
-
-// Fujitsu A64FX
-def : ProcessorModel<"a64fx", A64FXModel, ProcessorFeatures.A64FX,
-                     [TuneA64FX]>;
-
-// Nvidia Carmel
-def : ProcessorModel<"carmel", NoSchedModel, ProcessorFeatures.Carmel,
-                     [TuneCarmel]>;
-
-// Ampere Computing
-def : ProcessorModel<"ampere1", Ampere1Model, ProcessorFeatures.Ampere1,
-                     [TuneAmpere1]>;
-
-def : ProcessorModel<"ampere1a", Ampere1Model, ProcessorFeatures.Ampere1A,
-                     [TuneAmpere1A]>;
-
-def : ProcessorModel<"ampere1b", Ampere1BModel, ProcessorFeatures.Ampere1B,
-                     [TuneAmpere1B]>;
+include "AArch64Processors.td"
 
 //===----------------------------------------------------------------------===//
 // Assembly parser
diff --git a/llvm/lib/Target/AArch64/AArch64Arm64ECCallLowering.cpp b/llvm/lib/Target/AArch64/AArch64Arm64ECCallLowering.cpp
index 3bf6283b79e9..dddc181b0314 100644
--- a/llvm/lib/Target/AArch64/AArch64Arm64ECCallLowering.cpp
+++ b/llvm/lib/Target/AArch64/AArch64Arm64ECCallLowering.cpp
@@ -178,13 +178,14 @@ void AArch64Arm64ECCallLowering::getThunkArgTypes(
   }
 
   for (unsigned E = FT->getNumParams(); I != E; ++I) {
-    Align ParamAlign = AttrList.getParamAlignment(I).valueOrOne();
 #if 0
     // FIXME: Need more information about argument size; see
     // https://reviews.llvm.org/D132926
     uint64_t ArgSizeBytes = AttrList.getParamArm64ECArgSizeBytes(I);
+    Align ParamAlign = AttrList.getParamAlignment(I).valueOrOne();
 #else
     uint64_t ArgSizeBytes = 0;
+    Align ParamAlign = Align();
 #endif
     Type *Arm64Ty, *X64Ty;
     canonicalizeThunkType(FT->getParamType(I), ParamAlign,
@@ -294,7 +295,7 @@ void AArch64Arm64ECCallLowering::canonicalizeThunkType(
     uint64_t TotalSizeBytes = ElementCnt * ElementSizePerBytes;
     if (ElementTy->isFloatTy() || ElementTy->isDoubleTy()) {
       Out << (ElementTy->isFloatTy() ? "F" : "D") << TotalSizeBytes;
-      if (Alignment.value() >= 8 && !T->isPointerTy())
+      if (Alignment.value() >= 16 && !Ret)
         Out << "a" << Alignment.value();
       Arm64Ty = T;
       if (TotalSizeBytes <= 8) {
@@ -325,7 +326,7 @@ void AArch64Arm64ECCallLowering::canonicalizeThunkType(
   Out << "m";
   if (TypeSize != 4)
     Out << TypeSize;
-  if (Alignment.value() >= 8 && !T->isPointerTy())
+  if (Alignment.value() >= 16 && !Ret)
     Out << "a" << Alignment.value();
   // FIXME: Try to canonicalize Arm64Ty more thoroughly?
   Arm64Ty = T;
diff --git a/llvm/lib/Target/AArch64/AArch64ConditionOptimizer.cpp b/llvm/lib/Target/AArch64/AArch64ConditionOptimizer.cpp
index 1c20e24e41d7..2a4a3c0df08f 100644
--- a/llvm/lib/Target/AArch64/AArch64ConditionOptimizer.cpp
+++ b/llvm/lib/Target/AArch64/AArch64ConditionOptimizer.cpp
@@ -163,7 +163,7 @@ MachineInstr *AArch64ConditionOptimizer::findSuitableCompare(
     MachineInstr &I = *It;
     assert(!I.isTerminator() && "Spurious terminator");
     // Check if there is any use of NZCV between CMP and Bcc.
-    if (I.readsRegister(AArch64::NZCV))
+    if (I.readsRegister(AArch64::NZCV, /*TRI=*/nullptr))
       return nullptr;
     switch (I.getOpcode()) {
     // cmp is an alias for subs with a dead destination register.
diff --git a/llvm/lib/Target/AArch64/AArch64ConditionalCompares.cpp b/llvm/lib/Target/AArch64/AArch64ConditionalCompares.cpp
index 17e0e3072db6..8c16a88a13a4 100644
--- a/llvm/lib/Target/AArch64/AArch64ConditionalCompares.cpp
+++ b/llvm/lib/Target/AArch64/AArch64ConditionalCompares.cpp
@@ -299,7 +299,7 @@ MachineInstr *SSACCmpConv::findConvertibleCompare(MachineBasicBlock *MBB) {
   if (I == MBB->end())
     return nullptr;
   // The terminator must be controlled by the flags.
-  if (!I->readsRegister(AArch64::NZCV)) {
+  if (!I->readsRegister(AArch64::NZCV, /*TRI=*/nullptr)) {
     switch (I->getOpcode()) {
     case AArch64::CBZW:
     case AArch64::CBZX:
diff --git a/llvm/lib/Target/AArch64/AArch64DeadRegisterDefinitionsPass.cpp b/llvm/lib/Target/AArch64/AArch64DeadRegisterDefinitionsPass.cpp
index 3e04cbae8acf..2bc14f9821e6 100644
--- a/llvm/lib/Target/AArch64/AArch64DeadRegisterDefinitionsPass.cpp
+++ b/llvm/lib/Target/AArch64/AArch64DeadRegisterDefinitionsPass.cpp
@@ -124,7 +124,8 @@ void AArch64DeadRegisterDefinitions::processMachineBasicBlock(
       LLVM_DEBUG(dbgs() << "    Ignoring, operand is frame index\n");
       continue;
     }
-    if (MI.definesRegister(AArch64::XZR) || MI.definesRegister(AArch64::WZR)) {
+    if (MI.definesRegister(AArch64::XZR, /*TRI=*/nullptr) ||
+        MI.definesRegister(AArch64::WZR, /*TRI=*/nullptr)) {
       // It is not allowed to write to the same register (not even the zero
       // register) twice in a single instruction.
       LLVM_DEBUG(
diff --git a/llvm/lib/Target/AArch64/AArch64Features.td b/llvm/lib/Target/AArch64/AArch64Features.td
new file mode 100644
index 000000000000..3a3751a85afd
--- /dev/null
+++ b/llvm/lib/Target/AArch64/AArch64Features.td
@@ -0,0 +1,752 @@
+//=- AArch64Features.td - Describe AArch64 SubtargetFeatures -*- tablegen -*-=//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+//
+//===----------------------------------------------------------------------===//
+
+// Each SubtargetFeature which corresponds to an Arm Architecture feature should
+// be annotated with the respective FEAT_ feature name from the Architecture
+// Reference Manual. If a SubtargetFeature enables instructions from multiple
+// Arm Architecture Features, it should list all the relevant features. Not all
+// FEAT_ features have a corresponding SubtargetFeature.
+
+def FeatureFPARMv8 : SubtargetFeature<"fp-armv8", "HasFPARMv8", "true",
+                                       "Enable ARMv8 FP (FEAT_FP)">;
+
+def FeatureNEON : SubtargetFeature<"neon", "HasNEON", "true",
+  "Enable Advanced SIMD instructions (FEAT_AdvSIMD)", [FeatureFPARMv8]>;
+
+def FeatureSM4 : SubtargetFeature<
+    "sm4", "HasSM4", "true",
+    "Enable SM3 and SM4 support (FEAT_SM4, FEAT_SM3)", [FeatureNEON]>;
+
+def FeatureSHA2 : SubtargetFeature<
+    "sha2", "HasSHA2", "true",
+    "Enable SHA1 and SHA256 support (FEAT_SHA1, FEAT_SHA256)", [FeatureNEON]>;
+
+def FeatureSHA3 : SubtargetFeature<
+    "sha3", "HasSHA3", "true",
+    "Enable SHA512 and SHA3 support (FEAT_SHA3, FEAT_SHA512)", [FeatureNEON, FeatureSHA2]>;
+
+def FeatureAES : SubtargetFeature<
+    "aes", "HasAES", "true",
+    "Enable AES support (FEAT_AES, FEAT_PMULL)", [FeatureNEON]>;
+
+// Crypto has been split up and any combination is now valid (see the
+// crypto definitions above). Also, crypto is now context sensitive:
+// it has a different meaning for e.g. Armv8.4 than it has for Armv8.2.
+// Therefore, we rely on Clang, the user interfacing tool, to pass on the
+// appropriate crypto options. But here in the backend, crypto has very little
+// meaning anymore. We kept the Crypto definition here for backward
+// compatibility, and now imply features SHA2 and AES, which was the
+// "traditional" meaning of Crypto.
+def FeatureCrypto : SubtargetFeature<"crypto", "HasCrypto", "true",
+  "Enable cryptographic instructions", [FeatureNEON, FeatureSHA2, FeatureAES]>;
+
+def FeatureCRC : SubtargetFeature<"crc", "HasCRC", "true",
+  "Enable ARMv8 CRC-32 checksum instructions (FEAT_CRC32)">;
+
+def FeatureRAS : SubtargetFeature<"ras", "HasRAS", "true",
+  "Enable ARMv8 Reliability, Availability and Serviceability Extensions (FEAT_RAS, FEAT_RASv1p1)">;
+
+def FeatureRASv2 : SubtargetFeature<"rasv2", "HasRASv2", "true",
+  "Enable ARMv8.9-A Reliability, Availability and Serviceability Extensions (FEAT_RASv2)",
+  [FeatureRAS]>;
+
+def FeatureLSE : SubtargetFeature<"lse", "HasLSE", "true",
+  "Enable ARMv8.1 Large System Extension (LSE) atomic instructions (FEAT_LSE)">;
+
+def FeatureLSE2 : SubtargetFeature<"lse2", "HasLSE2", "true",
+  "Enable ARMv8.4 Large System Extension 2 (LSE2) atomicity rules (FEAT_LSE2)">;
+
+def FeatureOutlineAtomics : SubtargetFeature<"outline-atomics", "OutlineAtomics", "true",
+  "Enable out of line atomics to support LSE instructions">;
+
+def FeatureFMV : SubtargetFeature<"fmv", "HasFMV", "true",
+  "Enable Function Multi Versioning support.">;
+
+def FeatureRDM : SubtargetFeature<"rdm", "HasRDM", "true",
+  "Enable ARMv8.1 Rounding Double Multiply Add/Subtract instructions (FEAT_RDM)",
+  [FeatureNEON]>;
+
+def FeaturePAN : SubtargetFeature<
+    "pan", "HasPAN", "true",
+    "Enables ARM v8.1 Privileged Access-Never extension (FEAT_PAN)">;
+
+def FeatureLOR : SubtargetFeature<
+    "lor", "HasLOR", "true",
+    "Enables ARM v8.1 Limited Ordering Regions extension (FEAT_LOR)">;
+
+def FeatureCONTEXTIDREL2 : SubtargetFeature<"CONTEXTIDREL2", "HasCONTEXTIDREL2",
+    "true", "Enable RW operand CONTEXTIDR_EL2" >;
+
+def FeatureVH : SubtargetFeature<"vh", "HasVH", "true",
+    "Enables ARM v8.1 Virtual Host extension (FEAT_VHE)", [FeatureCONTEXTIDREL2] >;
+
+// This SubtargetFeature is special. It controls only whether codegen will turn
+// `llvm.readcyclecounter()` into an access to a PMUv3 System Register. The
+// `FEAT_PMUv3*` system registers are always available for assembly/disassembly.
+def FeaturePerfMon : SubtargetFeature<"perfmon", "HasPerfMon", "true",
+  "Enable Code Generation for ARMv8 PMUv3 Performance Monitors extension (FEAT_PMUv3)">;
+
+def FeatureFullFP16 : SubtargetFeature<"fullfp16", "HasFullFP16", "true",
+  "Full FP16 (FEAT_FP16)", [FeatureFPARMv8]>;
+
+def FeatureFP16FML : SubtargetFeature<"fp16fml", "HasFP16FML", "true",
+  "Enable FP16 FML instructions (FEAT_FHM)", [FeatureFullFP16]>;
+
+def FeatureSPE : SubtargetFeature<"spe", "HasSPE", "true",
+  "Enable Statistical Profiling extension (FEAT_SPE)">;
+
+def FeaturePAN_RWV : SubtargetFeature<
+    "pan-rwv", "HasPAN_RWV", "true",
+    "Enable v8.2 PAN s1e1R and s1e1W Variants (FEAT_PAN2)",
+    [FeaturePAN]>;
+
+// UAO PState
+def FeaturePsUAO : SubtargetFeature< "uaops", "HasPsUAO", "true",
+    "Enable v8.2 UAO PState (FEAT_UAO)">;
+
+def FeatureCCPP : SubtargetFeature<"ccpp", "HasCCPP",
+    "true", "Enable v8.2 data Cache Clean to Point of Persistence (FEAT_DPB)" >;
+
+def FeatureSVE : SubtargetFeature<"sve", "HasSVE", "true",
+  "Enable Scalable Vector Extension (SVE) instructions (FEAT_SVE)", [FeatureFullFP16]>;
+
+def FeatureFPMR : SubtargetFeature<"fpmr", "HasFPMR", "true",
+  "Enable FPMR Register (FEAT_FPMR)">;
+
+def FeatureFP8 : SubtargetFeature<"fp8", "HasFP8", "true",
+  "Enable FP8 instructions (FEAT_FP8)">;
+
+// This flag is currently still labeled as Experimental, but when fully
+// implemented this should tell the compiler to use the zeroing pseudos to
+// benefit from the reverse instructions (e.g. SUB vs SUBR) if the inactive
+// lanes are known to be zero. The pseudos will then be expanded using the
+// MOVPRFX instruction to zero the inactive lanes. This feature should only be
+// enabled if MOVPRFX instructions are known to merge with the destructive
+// operations they prefix.
+//
+// This feature could similarly be extended to support cheap merging of _any_
+// value into the inactive lanes using the MOVPRFX instruction that uses
+// merging-predication.
+def FeatureExperimentalZeroingPseudos
+    : SubtargetFeature<"use-experimental-zeroing-pseudos",
+                       "UseExperimentalZeroingPseudos", "true",
+                       "Hint to the compiler that the MOVPRFX instruction is "
+                       "merged with destructive operations",
+                       []>;
+
+def FeatureUseScalarIncVL : SubtargetFeature<"use-scalar-inc-vl",
+  "UseScalarIncVL", "true", "Prefer inc/dec over add+cnt">;
+
+def FeatureBF16 : SubtargetFeature<"bf16", "HasBF16",
+    "true", "Enable BFloat16 Extension (FEAT_BF16)" >;
+
+def FeatureNoSVEFPLD1R : SubtargetFeature<"no-sve-fp-ld1r",
+  "NoSVEFPLD1R", "true", "Avoid using LD1RX instructions for FP">;
+
+def FeatureSVE2 : SubtargetFeature<"sve2", "HasSVE2", "true",
+  "Enable Scalable Vector Extension 2 (SVE2) instructions (FEAT_SVE2)",
+  [FeatureSVE, FeatureUseScalarIncVL]>;
+
+def FeatureSVE2AES : SubtargetFeature<"sve2-aes", "HasSVE2AES", "true",
+  "Enable AES SVE2 instructions (FEAT_SVE_AES, FEAT_SVE_PMULL128)",
+  [FeatureSVE2, FeatureAES]>;
+
+def FeatureSVE2SM4 : SubtargetFeature<"sve2-sm4", "HasSVE2SM4", "true",
+  "Enable SM4 SVE2 instructions (FEAT_SVE_SM4)", [FeatureSVE2, FeatureSM4]>;
+
+def FeatureSVE2SHA3 : SubtargetFeature<"sve2-sha3", "HasSVE2SHA3", "true",
+  "Enable SHA3 SVE2 instructions (FEAT_SVE_SHA3)", [FeatureSVE2, FeatureSHA3]>;
+
+def FeatureSVE2BitPerm : SubtargetFeature<"sve2-bitperm", "HasSVE2BitPerm", "true",
+  "Enable bit permutation SVE2 instructions (FEAT_SVE_BitPerm)", [FeatureSVE2]>;
+
+def FeatureSVE2p1: SubtargetFeature<"sve2p1", "HasSVE2p1", "true",
+  "Enable Scalable Vector Extension 2.1 instructions", [FeatureSVE2]>;
+
+def FeatureB16B16 : SubtargetFeature<"b16b16", "HasB16B16", "true",
+  "Enable SVE2.1 or SME2.1 non-widening BFloat16 to BFloat16 instructions (FEAT_B16B16)", [FeatureBF16]>;
+
+def FeatureZCRegMove : SubtargetFeature<"zcm", "HasZeroCycleRegMove", "true",
+                                        "Has zero-cycle register moves">;
+
+def FeatureZCZeroingGP : SubtargetFeature<"zcz-gp", "HasZeroCycleZeroingGP", "true",
+                                        "Has zero-cycle zeroing instructions for generic registers">;
+
+// It is generally beneficial to rewrite "fmov s0, wzr" to "movi d0, #0".
+// as movi is more efficient across all cores. Newer cores can eliminate
+// fmovs early and there is no difference with movi, but this not true for
+// all implementations.
+def FeatureNoZCZeroingFP : SubtargetFeature<"no-zcz-fp", "HasZeroCycleZeroingFP", "false",
+                                        "Has no zero-cycle zeroing instructions for FP registers">;
+
+def FeatureZCZeroing : SubtargetFeature<"zcz", "HasZeroCycleZeroing", "true",
+                                        "Has zero-cycle zeroing instructions",
+                                        [FeatureZCZeroingGP]>;
+
+/// ... but the floating-point version doesn't quite work in rare cases on older
+/// CPUs.
+def FeatureZCZeroingFPWorkaround : SubtargetFeature<"zcz-fp-workaround",
+    "HasZeroCycleZeroingFPWorkaround", "true",
+    "The zero-cycle floating-point zeroing instruction has a bug">;
+
+def FeatureStrictAlign : SubtargetFeature<"strict-align",
+                                          "RequiresStrictAlign", "true",
+                                          "Disallow all unaligned memory "
+                                          "access">;
+
+foreach i = {1-7,9-15,18,20-28} in
+    def FeatureReserveX#i : SubtargetFeature<"reserve-x"#i, "ReserveXRegister["#i#"]", "true",
+                                             "Reserve X"#i#", making it unavailable "
+                                             "as a GPR">;
+
+foreach i = {8-15,18} in
+    def FeatureCallSavedX#i : SubtargetFeature<"call-saved-x"#i,
+         "CustomCallSavedXRegs["#i#"]", "true", "Make X"#i#" callee saved.">;
+
+def FeatureBalanceFPOps : SubtargetFeature<"balance-fp-ops", "BalanceFPOps",
+    "true",
+    "balance mix of odd and even D-registers for fp multiply(-accumulate) ops">;
+
+def FeaturePredictableSelectIsExpensive : SubtargetFeature<
+    "predictable-select-expensive", "PredictableSelectIsExpensive", "true",
+    "Prefer likely predicted branches over selects">;
+
+def FeatureEnableSelectOptimize : SubtargetFeature<
+    "enable-select-opt", "EnableSelectOptimize", "true",
+    "Enable the select optimize pass for select loop heuristics">;
+
+def FeatureExynosCheapAsMoveHandling : SubtargetFeature<"exynos-cheap-as-move",
+    "HasExynosCheapAsMoveHandling", "true",
+    "Use Exynos specific handling of cheap instructions">;
+
+def FeaturePostRAScheduler : SubtargetFeature<"use-postra-scheduler",
+    "UsePostRAScheduler", "true", "Schedule again after register allocation">;
+
+def FeatureSlowMisaligned128Store : SubtargetFeature<"slow-misaligned-128store",
+    "IsMisaligned128StoreSlow", "true", "Misaligned 128 bit stores are slow">;
+
+def FeatureSlowPaired128 : SubtargetFeature<"slow-paired-128",
+    "IsPaired128Slow", "true", "Paired 128 bit loads and stores are slow">;
+
+def FeatureAscendStoreAddress : SubtargetFeature<"ascend-store-address",
+    "IsStoreAddressAscend", "true",
+    "Schedule vector stores by ascending address">;
+
+def FeatureSlowSTRQro : SubtargetFeature<"slow-strqro-store", "IsSTRQroSlow",
+    "true", "STR of Q register with register offset is slow">;
+
+def FeatureAlternateSExtLoadCVTF32Pattern : SubtargetFeature<
+    "alternate-sextload-cvt-f32-pattern", "UseAlternateSExtLoadCVTF32Pattern",
+    "true", "Use alternative pattern for sextload convert to f32">;
+
+def FeatureArithmeticBccFusion : SubtargetFeature<
+    "arith-bcc-fusion", "HasArithmeticBccFusion", "true",
+    "CPU fuses arithmetic+bcc operations">;
+
+def FeatureArithmeticCbzFusion : SubtargetFeature<
+    "arith-cbz-fusion", "HasArithmeticCbzFusion", "true",
+    "CPU fuses arithmetic + cbz/cbnz operations">;
+
+def FeatureCmpBccFusion : SubtargetFeature<
+    "cmp-bcc-fusion", "HasCmpBccFusion", "true",
+    "CPU fuses cmp+bcc operations">;
+
+def FeatureFuseAddress : SubtargetFeature<
+    "fuse-address", "HasFuseAddress", "true",
+    "CPU fuses address generation and memory operations">;
+
+def FeatureFuseAES : SubtargetFeature<
+    "fuse-aes", "HasFuseAES", "true",
+    "CPU fuses AES crypto operations">;
+
+def FeatureFuseArithmeticLogic : SubtargetFeature<
+    "fuse-arith-logic", "HasFuseArithmeticLogic", "true",
+    "CPU fuses arithmetic and logic operations">;
+
+def FeatureFuseCCSelect : SubtargetFeature<
+    "fuse-csel", "HasFuseCCSelect", "true",
+    "CPU fuses conditional select operations">;
+
+def FeatureFuseCryptoEOR : SubtargetFeature<
+    "fuse-crypto-eor", "HasFuseCryptoEOR", "true",
+    "CPU fuses AES/PMULL and EOR operations">;
+
+def FeatureFuseAdrpAdd : SubtargetFeature<
+    "fuse-adrp-add", "HasFuseAdrpAdd", "true",
+    "CPU fuses adrp+add operations">;
+
+def FeatureFuseLiterals : SubtargetFeature<
+    "fuse-literals", "HasFuseLiterals", "true",
+    "CPU fuses literal generation operations">;
+
+def FeatureFuseAddSub2RegAndConstOne : SubtargetFeature<
+   "fuse-addsub-2reg-const1", "HasFuseAddSub2RegAndConstOne", "true",
+   "CPU fuses (a + b + 1) and (a - b - 1)">;
+
+def FeatureDisableLatencySchedHeuristic : SubtargetFeature<
+    "disable-latency-sched-heuristic", "DisableLatencySchedHeuristic", "true",
+    "Disable latency scheduling heuristic">;
+
+def FeatureStorePairSuppress : SubtargetFeature<
+    "store-pair-suppress", "EnableStorePairSuppress", "true",
+    "Enable Store Pair Suppression heuristics">;
+
+def FeatureForce32BitJumpTables
+   : SubtargetFeature<"force-32bit-jump-tables", "Force32BitJumpTables", "true",
+                      "Force jump table entries to be 32-bits wide except at MinSize">;
+
+def FeatureRCPC : SubtargetFeature<"rcpc", "HasRCPC", "true",
+                                   "Enable support for RCPC extension (FEAT_LRCPC)">;
+
+def FeatureUseRSqrt : SubtargetFeature<
+    "use-reciprocal-square-root", "UseRSqrt", "true",
+    "Use the reciprocal square root approximation">;
+
+def FeatureDotProd : SubtargetFeature<
+    "dotprod", "HasDotProd", "true",
+    "Enable dot product support (FEAT_DotProd)", [FeatureNEON]>;
+
+def FeaturePAuth : SubtargetFeature<
+    "pauth", "HasPAuth", "true",
+    "Enable v8.3-A Pointer Authentication extension (FEAT_PAuth)">;
+
+def FeatureJS : SubtargetFeature<
+    "jsconv", "HasJS", "true",
+    "Enable v8.3-A JavaScript FP conversion instructions (FEAT_JSCVT)",
+    [FeatureFPARMv8]>;
+
+def FeatureCCIDX : SubtargetFeature<
+    "ccidx", "HasCCIDX", "true",
+    "Enable v8.3-A Extend of the CCSIDR number of sets (FEAT_CCIDX)">;
+
+def FeatureComplxNum : SubtargetFeature<
+    "complxnum", "HasComplxNum", "true",
+    "Enable v8.3-A Floating-point complex number support (FEAT_FCMA)",
+    [FeatureNEON]>;
+
+def FeatureNV : SubtargetFeature<
+    "nv", "HasNV", "true",
+    "Enable v8.4-A Nested Virtualization Enchancement (FEAT_NV, FEAT_NV2)">;
+
+def FeatureMPAM : SubtargetFeature<
+    "mpam", "HasMPAM", "true",
+    "Enable v8.4-A Memory system Partitioning and Monitoring extension (FEAT_MPAM)">;
+
+def FeatureDIT : SubtargetFeature<
+    "dit", "HasDIT", "true",
+    "Enable v8.4-A Data Independent Timing instructions (FEAT_DIT)">;
+
+def FeatureTRACEV8_4 : SubtargetFeature<
+    "tracev8.4", "HasTRACEV8_4", "true",
+    "Enable v8.4-A Trace extension (FEAT_TRF)">;
+
+def FeatureAM : SubtargetFeature<
+    "am", "HasAM", "true",
+    "Enable v8.4-A Activity Monitors extension (FEAT_AMUv1)">;
+
+def FeatureAMVS : SubtargetFeature<
+    "amvs", "HasAMVS", "true",
+    "Enable v8.6-A Activity Monitors Virtualization support (FEAT_AMUv1p1)",
+    [FeatureAM]>;
+
+def FeatureSEL2 : SubtargetFeature<
+    "sel2", "HasSEL2", "true",
+    "Enable v8.4-A Secure Exception Level 2 extension (FEAT_SEL2)">;
+
+def FeatureTLB_RMI : SubtargetFeature<
+    "tlb-rmi", "HasTLB_RMI", "true",
+    "Enable v8.4-A TLB Range and Maintenance Instructions (FEAT_TLBIOS, FEAT_TLBIRANGE)">;
+
+def FeatureFlagM : SubtargetFeature<
+    "flagm", "HasFlagM", "true",
+    "Enable v8.4-A Flag Manipulation Instructions (FEAT_FlagM)">;
+
+// 8.4 RCPC enchancements: LDAPR & STLR instructions with Immediate Offset
+def FeatureRCPC_IMMO : SubtargetFeature<"rcpc-immo", "HasRCPC_IMMO", "true",
+    "Enable v8.4-A RCPC instructions with Immediate Offsets (FEAT_LRCPC2)",
+    [FeatureRCPC]>;
+
+def FeatureNoNegativeImmediates : SubtargetFeature<"no-neg-immediates",
+                                        "NegativeImmediates", "false",
+                                        "Convert immediates and instructions "
+                                        "to their negated or complemented "
+                                        "equivalent when the immediate does "
+                                        "not fit in the encoding.">;
+
+// Address operands with shift amount 2 or 3 are fast on all Arm chips except
+// some old Apple cores (A7-A10?) which handle all shifts slowly. Cortex-A57
+// and derived designs through Cortex-X1 take an extra micro-op for shifts
+// of 1 or 4. Other Arm chips handle all shifted operands at the same speed
+// as unshifted operands.
+//
+// We don't try to model the behavior of the old Apple cores because new code
+// targeting A7 is very unlikely to actually run on an A7. The Cortex cores
+// are modeled by FeatureAddrLSLSlow14.
+def FeatureAddrLSLSlow14 : SubtargetFeature<
+    "addr-lsl-slow-14", "HasAddrLSLSlow14", "true",
+    "Address operands with shift amount of 1 or 4 are slow">;
+
+def FeatureALULSLFast : SubtargetFeature<
+    "alu-lsl-fast", "HasALULSLFast", "true",
+    "Add/Sub operations with lsl shift <= 4 are cheap">;
+
+def FeatureAggressiveFMA :
+  SubtargetFeature<"aggressive-fma",
+                   "HasAggressiveFMA",
+                   "true",
+                   "Enable Aggressive FMA for floating-point.">;
+
+def FeatureAltFPCmp : SubtargetFeature<"altnzcv", "HasAlternativeNZCV", "true",
+  "Enable alternative NZCV format for floating point comparisons (FEAT_FlagM2)">;
+
+def FeatureFRInt3264 : SubtargetFeature<"fptoint", "HasFRInt3264", "true",
+  "Enable FRInt[32|64][Z|X] instructions that round a floating-point number to "
+  "an integer (in FP format) forcing it to fit into a 32- or 64-bit int (FEAT_FRINTTS)" >;
+
+def FeatureSpecRestrict : SubtargetFeature<"specrestrict", "HasSpecRestrict",
+  "true", "Enable architectural speculation restriction (FEAT_CSV2_2)">;
+
+def FeatureSB : SubtargetFeature<"sb", "HasSB",
+  "true", "Enable v8.5 Speculation Barrier (FEAT_SB)" >;
+
+def FeatureSSBS : SubtargetFeature<"ssbs", "HasSSBS",
+  "true", "Enable Speculative Store Bypass Safe bit (FEAT_SSBS, FEAT_SSBS2)" >;
+
+def FeaturePredRes : SubtargetFeature<"predres", "HasPredRes", "true",
+  "Enable v8.5a execution and data prediction invalidation instructions (FEAT_SPECRES)" >;
+
+def FeatureCacheDeepPersist : SubtargetFeature<"ccdp", "HasCCDP",
+    "true", "Enable v8.5 Cache Clean to Point of Deep Persistence (FEAT_DPB2)" >;
+
+def FeatureBranchTargetId : SubtargetFeature<"bti", "HasBTI",
+    "true", "Enable Branch Target Identification (FEAT_BTI)" >;
+
+def FeatureRandGen : SubtargetFeature<"rand", "HasRandGen",
+    "true", "Enable Random Number generation instructions (FEAT_RNG)" >;
+
+def FeatureMTE : SubtargetFeature<"mte", "HasMTE",
+    "true", "Enable Memory Tagging Extension (FEAT_MTE, FEAT_MTE2)" >;
+
+def FeatureTRBE : SubtargetFeature<"trbe", "HasTRBE",
+    "true", "Enable Trace Buffer Extension (FEAT_TRBE)">;
+
+def FeatureETE : SubtargetFeature<"ete", "HasETE",
+    "true", "Enable Embedded Trace Extension (FEAT_ETE)",
+    [FeatureTRBE]>;
+
+def FeatureTME : SubtargetFeature<"tme", "HasTME",
+    "true", "Enable Transactional Memory Extension (FEAT_TME)" >;
+
+def FeatureTaggedGlobals : SubtargetFeature<"tagged-globals",
+    "AllowTaggedGlobals",
+    "true", "Use an instruction sequence for taking the address of a global "
+    "that allows a memory tag in the upper address bits">;
+
+def FeatureMatMulInt8 : SubtargetFeature<"i8mm", "HasMatMulInt8",
+    "true", "Enable Matrix Multiply Int8 Extension (FEAT_I8MM)">;
+
+def FeatureMatMulFP32 : SubtargetFeature<"f32mm", "HasMatMulFP32",
+    "true", "Enable Matrix Multiply FP32 Extension (FEAT_F32MM)", [FeatureSVE]>;
+
+def FeatureMatMulFP64 : SubtargetFeature<"f64mm", "HasMatMulFP64",
+    "true", "Enable Matrix Multiply FP64 Extension (FEAT_F64MM)", [FeatureSVE]>;
+
+def FeatureXS : SubtargetFeature<"xs", "HasXS",
+    "true", "Enable Armv8.7-A limited-TLB-maintenance instruction (FEAT_XS)">;
+
+def FeatureWFxT : SubtargetFeature<"wfxt", "HasWFxT",
+    "true", "Enable Armv8.7-A WFET and WFIT instruction (FEAT_WFxT)">;
+
+def FeatureHCX : SubtargetFeature<
+    "hcx", "HasHCX", "true", "Enable Armv8.7-A HCRX_EL2 system register (FEAT_HCX)">;
+
+def FeatureLS64 : SubtargetFeature<"ls64", "HasLS64",
+    "true", "Enable Armv8.7-A LD64B/ST64B Accelerator Extension (FEAT_LS64, FEAT_LS64_V, FEAT_LS64_ACCDATA)">;
+
+def FeatureHBC : SubtargetFeature<"hbc", "HasHBC",
+    "true", "Enable Armv8.8-A Hinted Conditional Branches Extension (FEAT_HBC)">;
+
+def FeatureMOPS : SubtargetFeature<"mops", "HasMOPS",
+    "true", "Enable Armv8.8-A memcpy and memset acceleration instructions (FEAT_MOPS)">;
+
+def FeatureNMI : SubtargetFeature<"nmi", "HasNMI",
+    "true", "Enable Armv8.8-A Non-maskable Interrupts (FEAT_NMI, FEAT_GICv3_NMI)">;
+
+def FeatureBRBE : SubtargetFeature<"brbe", "HasBRBE",
+    "true", "Enable Branch Record Buffer Extension (FEAT_BRBE)">;
+
+def FeatureSPE_EEF : SubtargetFeature<"spe-eef", "HasSPE_EEF",
+    "true", "Enable extra register in the Statistical Profiling Extension (FEAT_SPEv1p2)">;
+
+def FeatureFineGrainedTraps : SubtargetFeature<"fgt", "HasFineGrainedTraps",
+    "true", "Enable fine grained virtualization traps extension (FEAT_FGT)">;
+
+def FeatureEnhancedCounterVirtualization :
+      SubtargetFeature<"ecv", "HasEnhancedCounterVirtualization",
+      "true", "Enable enhanced counter virtualization extension (FEAT_ECV)">;
+
+def FeatureRME : SubtargetFeature<"rme", "HasRME",
+    "true", "Enable Realm Management Extension (FEAT_RME)">;
+
+def FeatureSME : SubtargetFeature<"sme", "HasSME", "true",
+  "Enable Scalable Matrix Extension (SME) (FEAT_SME)", [FeatureBF16, FeatureUseScalarIncVL]>;
+
+def FeatureSMEF64F64 : SubtargetFeature<"sme-f64f64", "HasSMEF64F64", "true",
+  "Enable Scalable Matrix Extension (SME) F64F64 instructions (FEAT_SME_F64F64)", [FeatureSME]>;
+
+def FeatureSMEI16I64 : SubtargetFeature<"sme-i16i64", "HasSMEI16I64", "true",
+  "Enable Scalable Matrix Extension (SME) I16I64 instructions (FEAT_SME_I16I64)", [FeatureSME]>;
+
+def FeatureSMEFA64 : SubtargetFeature<"sme-fa64", "HasSMEFA64", "true",
+  "Enable the full A64 instruction set in streaming SVE mode (FEAT_SME_FA64)", [FeatureSME, FeatureSVE2]>;
+
+def FeatureSME2 : SubtargetFeature<"sme2", "HasSME2", "true",
+  "Enable Scalable Matrix Extension 2 (SME2) instructions", [FeatureSME]>;
+
+def FeatureSMEF16F16 : SubtargetFeature<"sme-f16f16", "HasSMEF16F16", "true",
+  "Enable SME non-widening Float16 instructions (FEAT_SME_F16F16)", [FeatureSME2]>;
+
+def FeatureSME2p1 : SubtargetFeature<"sme2p1", "HasSME2p1", "true",
+  "Enable Scalable Matrix Extension 2.1 (FEAT_SME2p1) instructions", [FeatureSME2]>;
+
+def FeatureFAMINMAX: SubtargetFeature<"faminmax", "HasFAMINMAX", "true",
+   "Enable FAMIN and FAMAX instructions (FEAT_FAMINMAX)">;
+
+def FeatureFP8FMA : SubtargetFeature<"fp8fma", "HasFP8FMA", "true",
+  "Enable fp8 multiply-add instructions (FEAT_FP8FMA)">;
+
+def FeatureSSVE_FP8FMA : SubtargetFeature<"ssve-fp8fma", "HasSSVE_FP8FMA", "true",
+  "Enable SVE2 fp8 multiply-add instructions (FEAT_SSVE_FP8FMA)", [FeatureSME2]>;
+
+def FeatureFP8DOT2: SubtargetFeature<"fp8dot2", "HasFP8DOT2", "true",
+   "Enable fp8 2-way dot instructions (FEAT_FP8DOT2)">;
+
+def FeatureSSVE_FP8DOT2 : SubtargetFeature<"ssve-fp8dot2", "HasSSVE_FP8DOT2", "true",
+  "Enable SVE2 fp8 2-way dot product instructions (FEAT_SSVE_FP8DOT2)", [FeatureSME2]>;
+
+def FeatureFP8DOT4: SubtargetFeature<"fp8dot4", "HasFP8DOT4", "true",
+   "Enable fp8 4-way dot instructions (FEAT_FP8DOT4)">;
+
+def FeatureSSVE_FP8DOT4 : SubtargetFeature<"ssve-fp8dot4", "HasSSVE_FP8DOT4", "true",
+  "Enable SVE2 fp8 4-way dot product instructions (FEAT_SSVE_FP8DOT4)", [FeatureSME2]>;
+def FeatureLUT: SubtargetFeature<"lut", "HasLUT", "true",
+   "Enable Lookup Table instructions (FEAT_LUT)">;
+
+def FeatureSME_LUTv2 : SubtargetFeature<"sme-lutv2", "HasSME_LUTv2", "true",
+  "Enable Scalable Matrix Extension (SME) LUTv2 instructions (FEAT_SME_LUTv2)">;
+
+def FeatureSMEF8F16 : SubtargetFeature<"sme-f8f16", "HasSMEF8F16", "true",
+  "Enable Scalable Matrix Extension (SME) F8F16 instructions(FEAT_SME_F8F16)", [FeatureSME2, FeatureFP8]>;
+
+def FeatureSMEF8F32 : SubtargetFeature<"sme-f8f32", "HasSMEF8F32", "true",
+  "Enable Scalable Matrix Extension (SME) F8F32 instructions (FEAT_SME_F8F32)", [FeatureSME2, FeatureFP8]>;
+
+def FeatureAppleA7SysReg  : SubtargetFeature<"apple-a7-sysreg", "HasAppleA7SysReg", "true",
+  "Apple A7 (the CPU formerly known as Cyclone)">;
+
+def FeatureEL2VMSA : SubtargetFeature<"el2vmsa", "HasEL2VMSA", "true",
+  "Enable Exception Level 2 Virtual Memory System Architecture">;
+
+def FeatureEL3 : SubtargetFeature<"el3", "HasEL3", "true",
+  "Enable Exception Level 3">;
+
+def FeatureCSSC : SubtargetFeature<"cssc", "HasCSSC", "true",
+  "Enable Common Short Sequence Compression (CSSC) instructions (FEAT_CSSC)">;
+
+def FeatureFixCortexA53_835769 : SubtargetFeature<"fix-cortex-a53-835769",
+  "FixCortexA53_835769", "true", "Mitigate Cortex-A53 Erratum 835769">;
+
+def FeatureNoBTIAtReturnTwice : SubtargetFeature<"no-bti-at-return-twice",
+                                                 "NoBTIAtReturnTwice", "true",
+                                                 "Don't place a BTI instruction "
+                                                 "after a return-twice">;
+
+def FeatureCHK : SubtargetFeature<"chk", "HasCHK",
+    "true", "Enable Armv8.0-A Check Feature Status Extension (FEAT_CHK)">;
+
+def FeatureGCS : SubtargetFeature<"gcs", "HasGCS",
+    "true", "Enable Armv9.4-A Guarded Call Stack Extension", [FeatureCHK]>;
+
+def FeatureCLRBHB : SubtargetFeature<"clrbhb", "HasCLRBHB",
+    "true", "Enable Clear BHB instruction (FEAT_CLRBHB)">;
+
+def FeaturePRFM_SLC : SubtargetFeature<"prfm-slc-target", "HasPRFM_SLC",
+    "true", "Enable SLC target for PRFM instruction">;
+
+def FeatureSPECRES2 : SubtargetFeature<"specres2", "HasSPECRES2",
+    "true", "Enable Speculation Restriction Instruction (FEAT_SPECRES2)",
+    [FeaturePredRes]>;
+
+def FeatureMEC : SubtargetFeature<"mec", "HasMEC",
+    "true", "Enable Memory Encryption Contexts Extension", [FeatureRME]>;
+
+def FeatureITE : SubtargetFeature<"ite", "HasITE",
+    "true", "Enable Armv9.4-A Instrumentation Extension FEAT_ITE", [FeatureETE,
+    FeatureTRBE]>;
+
+def FeatureRCPC3 : SubtargetFeature<"rcpc3", "HasRCPC3",
+    "true", "Enable Armv8.9-A RCPC instructions for A64 and Advanced SIMD and floating-point instruction set (FEAT_LRCPC3)",
+    [FeatureRCPC_IMMO]>;
+
+def FeatureTHE : SubtargetFeature<"the", "HasTHE",
+    "true", "Enable Armv8.9-A Translation Hardening Extension (FEAT_THE)">;
+
+def FeatureLSE128 : SubtargetFeature<"lse128", "HasLSE128",
+    "true", "Enable Armv9.4-A 128-bit Atomic Instructions (FEAT_LSE128)",
+    [FeatureLSE]>;
+
+// FEAT_D128, FEAT_LVA3, FEAT_SYSREG128, and FEAT_SYSINSTR128 are mutually implicit.
+// Therefore group them all under a single feature flag, d128:
+def FeatureD128 : SubtargetFeature<"d128", "HasD128",
+    "true", "Enable Armv9.4-A 128-bit Page Table Descriptors, System Registers "
+    "and Instructions (FEAT_D128, FEAT_LVA3, FEAT_SYSREG128, FEAT_SYSINSTR128)",
+    [FeatureLSE128]>;
+
+def FeatureDisableLdp : SubtargetFeature<"disable-ldp", "HasDisableLdp",
+    "true", "Do not emit ldp">;
+
+def FeatureDisableStp : SubtargetFeature<"disable-stp", "HasDisableStp",
+    "true", "Do not emit stp">;
+
+def FeatureLdpAlignedOnly : SubtargetFeature<"ldp-aligned-only", "HasLdpAlignedOnly",
+    "true", "In order to emit ldp, first check if the load will be aligned to 2 * element_size">;
+
+def FeatureStpAlignedOnly : SubtargetFeature<"stp-aligned-only", "HasStpAlignedOnly",
+    "true", "In order to emit stp, first check if the store will be aligned to 2 * element_size">;
+
+// AArch64 2023 Architecture Extensions (v9.5-A)
+
+def FeatureCPA : SubtargetFeature<"cpa", "HasCPA", "true",
+    "Enable Armv9.5-A Checked Pointer Arithmetic (FEAT_CPA)">;
+
+def FeaturePAuthLR : SubtargetFeature<"pauth-lr", "HasPAuthLR",
+    "true", "Enable Armv9.5-A PAC enhancements (FEAT_PAuth_LR)">;
+
+def FeatureTLBIW : SubtargetFeature<"tlbiw", "HasTLBIW", "true",
+  "Enable ARMv9.5-A TLBI VMALL for Dirty State (FEAT_TLBIW)">;
+
+
+//===----------------------------------------------------------------------===//
+// Architectures.
+//
+def HasV8_0aOps : SubtargetFeature<"v8a", "HasV8_0aOps", "true",
+  "Support ARM v8.0a instructions", [FeatureEL2VMSA, FeatureEL3]>;
+
+def HasV8_1aOps : SubtargetFeature<"v8.1a", "HasV8_1aOps", "true",
+  "Support ARM v8.1a instructions", [HasV8_0aOps, FeatureCRC, FeatureLSE,
+  FeatureRDM, FeaturePAN, FeatureLOR, FeatureVH]>;
+
+def HasV8_2aOps : SubtargetFeature<"v8.2a", "HasV8_2aOps", "true",
+  "Support ARM v8.2a instructions", [HasV8_1aOps, FeaturePsUAO,
+  FeaturePAN_RWV, FeatureRAS, FeatureCCPP]>;
+
+def HasV8_3aOps : SubtargetFeature<"v8.3a", "HasV8_3aOps", "true",
+  "Support ARM v8.3a instructions", [HasV8_2aOps, FeatureRCPC, FeaturePAuth,
+  FeatureJS, FeatureCCIDX, FeatureComplxNum]>;
+
+def HasV8_4aOps : SubtargetFeature<"v8.4a", "HasV8_4aOps", "true",
+  "Support ARM v8.4a instructions", [HasV8_3aOps, FeatureDotProd,
+  FeatureNV, FeatureMPAM, FeatureDIT,
+  FeatureTRACEV8_4, FeatureAM, FeatureSEL2, FeatureTLB_RMI,
+  FeatureFlagM, FeatureRCPC_IMMO, FeatureLSE2]>;
+
+def HasV8_5aOps : SubtargetFeature<
+  "v8.5a", "HasV8_5aOps", "true", "Support ARM v8.5a instructions",
+  [HasV8_4aOps, FeatureAltFPCmp, FeatureFRInt3264, FeatureSpecRestrict,
+   FeatureSSBS, FeatureSB, FeaturePredRes, FeatureCacheDeepPersist,
+   FeatureBranchTargetId]>;
+
+def HasV8_6aOps : SubtargetFeature<
+  "v8.6a", "HasV8_6aOps", "true", "Support ARM v8.6a instructions",
+  [HasV8_5aOps, FeatureAMVS, FeatureBF16, FeatureFineGrainedTraps,
+   FeatureEnhancedCounterVirtualization, FeatureMatMulInt8]>;
+
+def HasV8_7aOps : SubtargetFeature<
+  "v8.7a", "HasV8_7aOps", "true", "Support ARM v8.7a instructions",
+  [HasV8_6aOps, FeatureXS, FeatureWFxT, FeatureHCX]>;
+
+def HasV8_8aOps : SubtargetFeature<
+  "v8.8a", "HasV8_8aOps", "true", "Support ARM v8.8a instructions",
+  [HasV8_7aOps, FeatureHBC, FeatureMOPS, FeatureNMI]>;
+
+def HasV8_9aOps : SubtargetFeature<
+  "v8.9a", "HasV8_9aOps", "true", "Support ARM v8.9a instructions",
+  [HasV8_8aOps, FeatureCLRBHB, FeaturePRFM_SLC, FeatureSPECRES2,
+   FeatureCSSC, FeatureRASv2, FeatureCHK]>;
+
+def HasV9_0aOps : SubtargetFeature<
+  "v9a", "HasV9_0aOps", "true", "Support ARM v9a instructions",
+  [HasV8_5aOps, FeatureMEC, FeatureSVE2]>;
+
+def HasV9_1aOps : SubtargetFeature<
+  "v9.1a", "HasV9_1aOps", "true", "Support ARM v9.1a instructions",
+  [HasV8_6aOps, HasV9_0aOps]>;
+
+def HasV9_2aOps : SubtargetFeature<
+  "v9.2a", "HasV9_2aOps", "true", "Support ARM v9.2a instructions",
+  [HasV8_7aOps, HasV9_1aOps]>;
+
+def HasV9_3aOps : SubtargetFeature<
+  "v9.3a", "HasV9_3aOps", "true", "Support ARM v9.3a instructions",
+  [HasV8_8aOps, HasV9_2aOps]>;
+
+def HasV9_4aOps : SubtargetFeature<
+  "v9.4a", "HasV9_4aOps", "true", "Support ARM v9.4a instructions",
+  [HasV8_9aOps, HasV9_3aOps]>;
+
+def HasV9_5aOps : SubtargetFeature<
+  "v9.5a", "HasV9_5aOps", "true", "Support ARM v9.5a instructions",
+  [HasV9_4aOps, FeatureCPA]>;
+
+def HasV8_0rOps : SubtargetFeature<
+  "v8r", "HasV8_0rOps", "true", "Support ARM v8r instructions",
+  [//v8.1
+  FeatureCRC, FeaturePAN, FeatureLSE, FeatureCONTEXTIDREL2,
+  //v8.2
+  FeatureRAS, FeaturePsUAO, FeatureCCPP, FeaturePAN_RWV,
+  //v8.3
+  FeatureCCIDX, FeaturePAuth, FeatureRCPC,
+  //v8.4
+  FeatureTRACEV8_4, FeatureTLB_RMI, FeatureFlagM, FeatureDIT, FeatureSEL2,
+  FeatureRCPC_IMMO,
+  // Not mandatory in v8.0-R, but included here on the grounds that it
+  // only enables names of system registers
+  FeatureSpecRestrict
+  ]>;
+
+//===----------------------------------------------------------------------===//
+// Access to privileged registers
+//===----------------------------------------------------------------------===//
+
+foreach i = 1-3 in
+def FeatureUseEL#i#ForTP : SubtargetFeature<"tpidr-el"#i, "UseEL"#i#"ForTP",
+  "true", "Permit use of TPIDR_EL"#i#" for the TLS base">;
+def FeatureUseROEL0ForTP : SubtargetFeature<"tpidrro-el0", "UseROEL0ForTP",
+  "true", "Permit use of TPIDRRO_EL0 for the TLS base">;
+
+//===----------------------------------------------------------------------===//
+// Control codegen mitigation against Straight Line Speculation vulnerability.
+//===----------------------------------------------------------------------===//
+
+def FeatureHardenSlsRetBr : SubtargetFeature<"harden-sls-retbr",
+  "HardenSlsRetBr", "true",
+  "Harden against straight line speculation across RET and BR instructions">;
+def FeatureHardenSlsBlr : SubtargetFeature<"harden-sls-blr",
+  "HardenSlsBlr", "true",
+  "Harden against straight line speculation across BLR instructions">;
+def FeatureHardenSlsNoComdat : SubtargetFeature<"harden-sls-nocomdat",
+  "HardenSlsNoComdat", "true",
+  "Generate thunk code for SLS mitigation in the normal text section">;
+
+
+// Only intended to be used by disassemblers.
+def FeatureAll
+    : SubtargetFeature<"all", "IsAll", "true", "Enable all instructions", []>;
diff --git a/llvm/lib/Target/AArch64/AArch64FrameLowering.cpp b/llvm/lib/Target/AArch64/AArch64FrameLowering.cpp
index 419c141121c3..c86c98eed24f 100644
--- a/llvm/lib/Target/AArch64/AArch64FrameLowering.cpp
+++ b/llvm/lib/Target/AArch64/AArch64FrameLowering.cpp
@@ -1296,7 +1296,7 @@ static MachineBasicBlock::iterator InsertSEH(MachineBasicBlock::iterator MBBI,
   }
   case AArch64::LDPQpost:
     Imm = -Imm;
-    LLVM_FALLTHROUGH;
+    [[fallthrough]];
   case AArch64::STPQpre: {
     unsigned Reg0 = RegInfo->getSEHRegNum(MBBI->getOperand(1).getReg());
     unsigned Reg1 = RegInfo->getSEHRegNum(MBBI->getOperand(2).getReg());
diff --git a/llvm/lib/Target/AArch64/AArch64ISelLowering.cpp b/llvm/lib/Target/AArch64/AArch64ISelLowering.cpp
index 3d1453e3beb9..cb7930f0cdee 100644
--- a/llvm/lib/Target/AArch64/AArch64ISelLowering.cpp
+++ b/llvm/lib/Target/AArch64/AArch64ISelLowering.cpp
@@ -842,6 +842,9 @@ AArch64TargetLowering::AArch64TargetLowering(const TargetMachine &TM,
 
   setOperationAction(ISD::GET_ROUNDING, MVT::i32, Custom);
   setOperationAction(ISD::SET_ROUNDING, MVT::Other, Custom);
+  setOperationAction(ISD::GET_FPMODE, MVT::i32, Custom);
+  setOperationAction(ISD::SET_FPMODE, MVT::i32, Custom);
+  setOperationAction(ISD::RESET_FPMODE, MVT::Other, Custom);
 
   setOperationAction(ISD::ATOMIC_CMP_SWAP, MVT::i128, Custom);
   if (!Subtarget->hasLSE() && !Subtarget->outlineAtomics()) {
@@ -1603,39 +1606,19 @@ AArch64TargetLowering::AArch64TargetLowering(const TargetMachine &TM,
         setOperationAction(ISD::VECREDUCE_SEQ_FADD, VT, Custom);
     }
 
-    if (!Subtarget->isNeonAvailable()) {
-      setTruncStoreAction(MVT::v2f32, MVT::v2bf16, Custom);
-      setTruncStoreAction(MVT::v4f32, MVT::v4bf16, Custom);
-      setTruncStoreAction(MVT::v8f32, MVT::v8bf16, Custom);
-      setTruncStoreAction(MVT::v2f64, MVT::v2bf16, Custom);
-      setTruncStoreAction(MVT::v4f64, MVT::v4bf16, Custom);
-      setTruncStoreAction(MVT::v2f32, MVT::v2f16, Custom);
-      setTruncStoreAction(MVT::v4f32, MVT::v4f16, Custom);
-      setTruncStoreAction(MVT::v8f32, MVT::v8f16, Custom);
-      setTruncStoreAction(MVT::v1f64, MVT::v1f16, Custom);
-      setTruncStoreAction(MVT::v2f64, MVT::v2f16, Custom);
-      setTruncStoreAction(MVT::v4f64, MVT::v4f16, Custom);
-      setTruncStoreAction(MVT::v1f64, MVT::v1f32, Custom);
-      setTruncStoreAction(MVT::v2f64, MVT::v2f32, Custom);
-      setTruncStoreAction(MVT::v4f64, MVT::v4f32, Custom);
-      for (MVT VT : {MVT::v8i8, MVT::v16i8, MVT::v4i16, MVT::v8i16, MVT::v2i32,
-                     MVT::v4i32, MVT::v1i64, MVT::v2i64})
-        addTypeForFixedLengthSVE(VT, /*StreamingSVE=*/ true);
-
-      for (MVT VT :
-           {MVT::v4f16, MVT::v8f16, MVT::v2f32, MVT::v4f32, MVT::v2f64})
-        addTypeForFixedLengthSVE(VT, /*StreamingSVE=*/ true);
-    }
-
     // NOTE: Currently this has to happen after computeRegisterProperties rather
     // than the preferred option of combining it with the addRegisterClass call.
     if (Subtarget->useSVEForFixedLengthVectors()) {
-      for (MVT VT : MVT::integer_fixedlen_vector_valuetypes())
-        if (useSVEForFixedLengthVectorVT(VT))
-          addTypeForFixedLengthSVE(VT, /*StreamingSVE=*/ false);
-      for (MVT VT : MVT::fp_fixedlen_vector_valuetypes())
-        if (useSVEForFixedLengthVectorVT(VT))
-          addTypeForFixedLengthSVE(VT, /*StreamingSVE=*/ false);
+      for (MVT VT : MVT::integer_fixedlen_vector_valuetypes()) {
+        if (useSVEForFixedLengthVectorVT(
+                VT, /*OverrideNEON=*/!Subtarget->isNeonAvailable()))
+          addTypeForFixedLengthSVE(VT);
+      }
+      for (MVT VT : MVT::fp_fixedlen_vector_valuetypes()) {
+        if (useSVEForFixedLengthVectorVT(
+                VT, /*OverrideNEON=*/!Subtarget->isNeonAvailable()))
+          addTypeForFixedLengthSVE(VT);
+      }
 
       // 64bit results can mean a bigger than NEON input.
       for (auto VT : {MVT::v8i8, MVT::v4i16})
@@ -1869,8 +1852,7 @@ bool AArch64TargetLowering::shouldExpandCttzElements(EVT VT) const {
   return !Subtarget->hasSVEorSME() || VT != MVT::nxv16i1;
 }
 
-void AArch64TargetLowering::addTypeForFixedLengthSVE(MVT VT,
-                                                     bool StreamingSVE) {
+void AArch64TargetLowering::addTypeForFixedLengthSVE(MVT VT) {
   assert(VT.isFixedLengthVector() && "Expected fixed length vector type!");
 
   // By default everything must be expanded.
@@ -1889,13 +1871,17 @@ void AArch64TargetLowering::addTypeForFixedLengthSVE(MVT VT,
     setCondCodeAction(ISD::SETONE, VT, Expand);
   }
 
+  TargetLoweringBase::LegalizeAction Default =
+      VT == MVT::v1f64 ? Expand : Custom;
+
   // Mark integer truncating stores/extending loads as having custom lowering
   if (VT.isInteger()) {
     MVT InnerVT = VT.changeVectorElementType(MVT::i8);
     while (InnerVT != VT) {
-      setTruncStoreAction(VT, InnerVT, Custom);
-      setLoadExtAction(ISD::ZEXTLOAD, VT, InnerVT, Custom);
-      setLoadExtAction(ISD::SEXTLOAD, VT, InnerVT, Custom);
+      setTruncStoreAction(VT, InnerVT, Default);
+      setLoadExtAction(ISD::ZEXTLOAD, VT, InnerVT, Default);
+      setLoadExtAction(ISD::SEXTLOAD, VT, InnerVT, Default);
+      setLoadExtAction(ISD::EXTLOAD, VT, InnerVT, Default);
       InnerVT = InnerVT.changeVectorElementType(
           MVT::getIntegerVT(2 * InnerVT.getScalarSizeInBits()));
     }
@@ -1907,101 +1893,103 @@ void AArch64TargetLowering::addTypeForFixedLengthSVE(MVT VT,
     MVT InnerVT = VT.changeVectorElementType(MVT::f16);
     while (InnerVT != VT) {
       setTruncStoreAction(VT, InnerVT, Custom);
-      setLoadExtAction(ISD::EXTLOAD, VT, InnerVT, Custom);
+      setLoadExtAction(ISD::EXTLOAD, VT, InnerVT, Default);
       InnerVT = InnerVT.changeVectorElementType(
           MVT::getFloatingPointVT(2 * InnerVT.getScalarSizeInBits()));
     }
   }
 
+  bool PreferNEON = VT.is64BitVector() || VT.is128BitVector();
+  bool PreferSVE = !PreferNEON && Subtarget->isSVEAvailable();
+
   // Lower fixed length vector operations to scalable equivalents.
-  setOperationAction(ISD::ABS, VT, Custom);
-  setOperationAction(ISD::ADD, VT, Custom);
-  setOperationAction(ISD::AND, VT, Custom);
-  setOperationAction(ISD::ANY_EXTEND, VT, Custom);
-  setOperationAction(ISD::BITCAST, VT, StreamingSVE ? Legal : Custom);
-  setOperationAction(ISD::BITREVERSE, VT, Custom);
-  setOperationAction(ISD::BSWAP, VT, Custom);
-  setOperationAction(ISD::BUILD_VECTOR, VT, Custom);
-  setOperationAction(ISD::CONCAT_VECTORS, VT, Custom);
-  setOperationAction(ISD::CTLZ, VT, Custom);
-  setOperationAction(ISD::CTPOP, VT, Custom);
-  setOperationAction(ISD::CTTZ, VT, Custom);
-  setOperationAction(ISD::EXTRACT_SUBVECTOR, VT, Custom);
-  setOperationAction(ISD::EXTRACT_VECTOR_ELT, VT, Custom);
-  setOperationAction(ISD::FABS, VT, Custom);
-  setOperationAction(ISD::FADD, VT, Custom);
-  setOperationAction(ISD::FCEIL, VT, Custom);
-  setOperationAction(ISD::FCOPYSIGN, VT, Custom);
-  setOperationAction(ISD::FDIV, VT, Custom);
-  setOperationAction(ISD::FFLOOR, VT, Custom);
-  setOperationAction(ISD::FMA, VT, Custom);
-  setOperationAction(ISD::FMAXIMUM, VT, Custom);
-  setOperationAction(ISD::FMAXNUM, VT, Custom);
-  setOperationAction(ISD::FMINIMUM, VT, Custom);
-  setOperationAction(ISD::FMINNUM, VT, Custom);
-  setOperationAction(ISD::FMUL, VT, Custom);
-  setOperationAction(ISD::FNEARBYINT, VT, Custom);
-  setOperationAction(ISD::FNEG, VT, Custom);
-  setOperationAction(ISD::FP_EXTEND, VT, Custom);
-  setOperationAction(ISD::FP_ROUND, VT, Custom);
-  setOperationAction(ISD::FP_TO_SINT, VT, Custom);
-  setOperationAction(ISD::FP_TO_UINT, VT, Custom);
-  setOperationAction(ISD::FRINT, VT, Custom);
-  setOperationAction(ISD::FROUND, VT, Custom);
-  setOperationAction(ISD::FROUNDEVEN, VT, Custom);
-  setOperationAction(ISD::FSQRT, VT, Custom);
-  setOperationAction(ISD::FSUB, VT, Custom);
-  setOperationAction(ISD::FTRUNC, VT, Custom);
-  setOperationAction(ISD::INSERT_VECTOR_ELT, VT, Custom);
-  setOperationAction(ISD::LOAD, VT, StreamingSVE ? Legal : Custom);
-  setOperationAction(ISD::MGATHER, VT, StreamingSVE ? Expand : Custom);
-  setOperationAction(ISD::MLOAD, VT, Custom);
-  setOperationAction(ISD::MSCATTER, VT, StreamingSVE ? Expand : Custom);
-  setOperationAction(ISD::MSTORE, VT, Custom);
-  setOperationAction(ISD::MUL, VT, Custom);
-  setOperationAction(ISD::MULHS, VT, Custom);
-  setOperationAction(ISD::MULHU, VT, Custom);
-  setOperationAction(ISD::OR, VT, Custom);
-  setOperationAction(ISD::SCALAR_TO_VECTOR, VT, StreamingSVE ? Legal : Expand);
-  setOperationAction(ISD::SDIV, VT, Custom);
-  setOperationAction(ISD::SELECT, VT, Custom);
-  setOperationAction(ISD::SETCC, VT, Custom);
-  setOperationAction(ISD::SHL, VT, Custom);
-  setOperationAction(ISD::SIGN_EXTEND, VT, Custom);
-  setOperationAction(ISD::SIGN_EXTEND_INREG, VT, Custom);
-  setOperationAction(ISD::SINT_TO_FP, VT, Custom);
-  setOperationAction(ISD::SMAX, VT, Custom);
-  setOperationAction(ISD::SMIN, VT, Custom);
-  setOperationAction(ISD::SPLAT_VECTOR, VT, Custom);
-  setOperationAction(ISD::SRA, VT, Custom);
-  setOperationAction(ISD::SRL, VT, Custom);
-  setOperationAction(ISD::STORE, VT, StreamingSVE ? Legal : Custom);
-  setOperationAction(ISD::SUB, VT, Custom);
-  setOperationAction(ISD::TRUNCATE, VT, Custom);
-  setOperationAction(ISD::UDIV, VT, Custom);
-  setOperationAction(ISD::UINT_TO_FP, VT, Custom);
-  setOperationAction(ISD::UMAX, VT, Custom);
-  setOperationAction(ISD::UMIN, VT, Custom);
-  setOperationAction(ISD::VECREDUCE_ADD, VT, Custom);
-  setOperationAction(ISD::VECREDUCE_AND, VT, Custom);
-  setOperationAction(ISD::VECREDUCE_FADD, VT, Custom);
-  setOperationAction(ISD::VECREDUCE_FMAX, VT, Custom);
-  setOperationAction(ISD::VECREDUCE_FMIN, VT, Custom);
-  setOperationAction(ISD::VECREDUCE_FMAXIMUM, VT, Custom);
-  setOperationAction(ISD::VECREDUCE_FMINIMUM, VT, Custom);
-  setOperationAction(ISD::VECREDUCE_OR, VT, Custom);
-  setOperationAction(ISD::VECREDUCE_SEQ_FADD, VT,
-                     StreamingSVE ? Expand : Custom);
-  setOperationAction(ISD::VECREDUCE_SMAX, VT, Custom);
-  setOperationAction(ISD::VECREDUCE_SMIN, VT, Custom);
-  setOperationAction(ISD::VECREDUCE_UMAX, VT, Custom);
-  setOperationAction(ISD::VECREDUCE_UMIN, VT, Custom);
-  setOperationAction(ISD::VECREDUCE_XOR, VT, Custom);
-  setOperationAction(ISD::VECTOR_SHUFFLE, VT, Custom);
-  setOperationAction(ISD::VECTOR_SPLICE, VT, Custom);
-  setOperationAction(ISD::VSELECT, VT, Custom);
-  setOperationAction(ISD::XOR, VT, Custom);
-  setOperationAction(ISD::ZERO_EXTEND, VT, Custom);
+  setOperationAction(ISD::ABS, VT, Default);
+  setOperationAction(ISD::ADD, VT, Default);
+  setOperationAction(ISD::AND, VT, Default);
+  setOperationAction(ISD::ANY_EXTEND, VT, Default);
+  setOperationAction(ISD::BITCAST, VT, PreferNEON ? Legal : Default);
+  setOperationAction(ISD::BITREVERSE, VT, Default);
+  setOperationAction(ISD::BSWAP, VT, Default);
+  setOperationAction(ISD::BUILD_VECTOR, VT, Default);
+  setOperationAction(ISD::CONCAT_VECTORS, VT, Default);
+  setOperationAction(ISD::CTLZ, VT, Default);
+  setOperationAction(ISD::CTPOP, VT, Default);
+  setOperationAction(ISD::CTTZ, VT, Default);
+  setOperationAction(ISD::EXTRACT_SUBVECTOR, VT, Default);
+  setOperationAction(ISD::EXTRACT_VECTOR_ELT, VT, Default);
+  setOperationAction(ISD::FABS, VT, Default);
+  setOperationAction(ISD::FADD, VT, Default);
+  setOperationAction(ISD::FCEIL, VT, Default);
+  setOperationAction(ISD::FCOPYSIGN, VT, Default);
+  setOperationAction(ISD::FDIV, VT, Default);
+  setOperationAction(ISD::FFLOOR, VT, Default);
+  setOperationAction(ISD::FMA, VT, Default);
+  setOperationAction(ISD::FMAXIMUM, VT, Default);
+  setOperationAction(ISD::FMAXNUM, VT, Default);
+  setOperationAction(ISD::FMINIMUM, VT, Default);
+  setOperationAction(ISD::FMINNUM, VT, Default);
+  setOperationAction(ISD::FMUL, VT, Default);
+  setOperationAction(ISD::FNEARBYINT, VT, Default);
+  setOperationAction(ISD::FNEG, VT, Default);
+  setOperationAction(ISD::FP_EXTEND, VT, Default);
+  setOperationAction(ISD::FP_ROUND, VT, Default);
+  setOperationAction(ISD::FP_TO_SINT, VT, Default);
+  setOperationAction(ISD::FP_TO_UINT, VT, Default);
+  setOperationAction(ISD::FRINT, VT, Default);
+  setOperationAction(ISD::FROUND, VT, Default);
+  setOperationAction(ISD::FROUNDEVEN, VT, Default);
+  setOperationAction(ISD::FSQRT, VT, Default);
+  setOperationAction(ISD::FSUB, VT, Default);
+  setOperationAction(ISD::FTRUNC, VT, Default);
+  setOperationAction(ISD::INSERT_VECTOR_ELT, VT, Default);
+  setOperationAction(ISD::LOAD, VT, PreferNEON ? Legal : Default);
+  setOperationAction(ISD::MGATHER, VT, PreferSVE ? Default : Expand);
+  setOperationAction(ISD::MLOAD, VT, Default);
+  setOperationAction(ISD::MSCATTER, VT, PreferSVE ? Default : Expand);
+  setOperationAction(ISD::MSTORE, VT, Default);
+  setOperationAction(ISD::MUL, VT, Default);
+  setOperationAction(ISD::MULHS, VT, Default);
+  setOperationAction(ISD::MULHU, VT, Default);
+  setOperationAction(ISD::OR, VT, Default);
+  setOperationAction(ISD::SCALAR_TO_VECTOR, VT, PreferNEON ? Legal : Expand);
+  setOperationAction(ISD::SDIV, VT, Default);
+  setOperationAction(ISD::SELECT, VT, Default);
+  setOperationAction(ISD::SETCC, VT, Default);
+  setOperationAction(ISD::SHL, VT, Default);
+  setOperationAction(ISD::SIGN_EXTEND, VT, Default);
+  setOperationAction(ISD::SIGN_EXTEND_INREG, VT, Default);
+  setOperationAction(ISD::SINT_TO_FP, VT, Default);
+  setOperationAction(ISD::SMAX, VT, Default);
+  setOperationAction(ISD::SMIN, VT, Default);
+  setOperationAction(ISD::SPLAT_VECTOR, VT, Default);
+  setOperationAction(ISD::SRA, VT, Default);
+  setOperationAction(ISD::SRL, VT, Default);
+  setOperationAction(ISD::STORE, VT, PreferNEON ? Legal : Default);
+  setOperationAction(ISD::SUB, VT, Default);
+  setOperationAction(ISD::TRUNCATE, VT, Default);
+  setOperationAction(ISD::UDIV, VT, Default);
+  setOperationAction(ISD::UINT_TO_FP, VT, Default);
+  setOperationAction(ISD::UMAX, VT, Default);
+  setOperationAction(ISD::UMIN, VT, Default);
+  setOperationAction(ISD::VECREDUCE_ADD, VT, Default);
+  setOperationAction(ISD::VECREDUCE_AND, VT, Default);
+  setOperationAction(ISD::VECREDUCE_FADD, VT, Default);
+  setOperationAction(ISD::VECREDUCE_FMAX, VT, Default);
+  setOperationAction(ISD::VECREDUCE_FMIN, VT, Default);
+  setOperationAction(ISD::VECREDUCE_FMAXIMUM, VT, Default);
+  setOperationAction(ISD::VECREDUCE_FMINIMUM, VT, Default);
+  setOperationAction(ISD::VECREDUCE_OR, VT, Default);
+  setOperationAction(ISD::VECREDUCE_SEQ_FADD, VT, PreferSVE ? Default : Expand);
+  setOperationAction(ISD::VECREDUCE_SMAX, VT, Default);
+  setOperationAction(ISD::VECREDUCE_SMIN, VT, Default);
+  setOperationAction(ISD::VECREDUCE_UMAX, VT, Default);
+  setOperationAction(ISD::VECREDUCE_UMIN, VT, Default);
+  setOperationAction(ISD::VECREDUCE_XOR, VT, Default);
+  setOperationAction(ISD::VECTOR_SHUFFLE, VT, Default);
+  setOperationAction(ISD::VECTOR_SPLICE, VT, Default);
+  setOperationAction(ISD::VSELECT, VT, Default);
+  setOperationAction(ISD::XOR, VT, Default);
+  setOperationAction(ISD::ZERO_EXTEND, VT, Default);
 }
 
 void AArch64TargetLowering::addDRTypeForNEON(MVT VT) {
@@ -4885,6 +4873,65 @@ SDValue AArch64TargetLowering::LowerSET_ROUNDING(SDValue Op,
   return DAG.getNode(ISD::INTRINSIC_VOID, DL, MVT::Other, Ops2);
 }
 
+SDValue AArch64TargetLowering::LowerGET_FPMODE(SDValue Op,
+                                               SelectionDAG &DAG) const {
+  SDLoc DL(Op);
+  SDValue Chain = Op->getOperand(0);
+
+  // Get current value of FPCR.
+  SDValue Ops[] = {
+      Chain, DAG.getTargetConstant(Intrinsic::aarch64_get_fpcr, DL, MVT::i64)};
+  SDValue FPCR =
+      DAG.getNode(ISD::INTRINSIC_W_CHAIN, DL, {MVT::i64, MVT::Other}, Ops);
+  Chain = FPCR.getValue(1);
+  FPCR = FPCR.getValue(0);
+
+  // Truncate FPCR to 32 bits.
+  SDValue Result = DAG.getNode(ISD::TRUNCATE, DL, MVT::i32, FPCR);
+
+  return DAG.getMergeValues({Result, Chain}, DL);
+}
+
+SDValue AArch64TargetLowering::LowerSET_FPMODE(SDValue Op,
+                                               SelectionDAG &DAG) const {
+  SDLoc DL(Op);
+  SDValue Chain = Op->getOperand(0);
+  SDValue Mode = Op->getOperand(1);
+
+  // Extend the specified value to 64 bits.
+  SDValue FPCR = DAG.getZExtOrTrunc(Mode, DL, MVT::i64);
+
+  // Set new value of FPCR.
+  SDValue Ops2[] = {
+      Chain, DAG.getConstant(Intrinsic::aarch64_set_fpcr, DL, MVT::i64), FPCR};
+  return DAG.getNode(ISD::INTRINSIC_VOID, DL, MVT::Other, Ops2);
+}
+
+SDValue AArch64TargetLowering::LowerRESET_FPMODE(SDValue Op,
+                                                 SelectionDAG &DAG) const {
+  SDLoc DL(Op);
+  SDValue Chain = Op->getOperand(0);
+
+  // Get current value of FPCR.
+  SDValue Ops[] = {
+      Chain, DAG.getTargetConstant(Intrinsic::aarch64_get_fpcr, DL, MVT::i64)};
+  SDValue FPCR =
+      DAG.getNode(ISD::INTRINSIC_W_CHAIN, DL, {MVT::i64, MVT::Other}, Ops);
+  Chain = FPCR.getValue(1);
+  FPCR = FPCR.getValue(0);
+
+  // Clear bits that are not reserved.
+  SDValue FPSCRMasked = DAG.getNode(
+      ISD::AND, DL, MVT::i64, FPCR,
+      DAG.getConstant(AArch64::ReservedFPControlBits, DL, MVT::i64));
+
+  // Set new value of FPCR.
+  SDValue Ops2[] = {Chain,
+                    DAG.getConstant(Intrinsic::aarch64_set_fpcr, DL, MVT::i64),
+                    FPSCRMasked};
+  return DAG.getNode(ISD::INTRINSIC_VOID, DL, MVT::Other, Ops2);
+}
+
 static unsigned selectUmullSmull(SDValue &N0, SDValue &N1, SelectionDAG &DAG,
                                  SDLoc DL, bool &IsMLA) {
   bool IsN0SExt = isSignExtended(N0, DAG);
@@ -6499,6 +6546,12 @@ SDValue AArch64TargetLowering::LowerOperation(SDValue Op,
     return LowerGET_ROUNDING(Op, DAG);
   case ISD::SET_ROUNDING:
     return LowerSET_ROUNDING(Op, DAG);
+  case ISD::GET_FPMODE:
+    return LowerGET_FPMODE(Op, DAG);
+  case ISD::SET_FPMODE:
+    return LowerSET_FPMODE(Op, DAG);
+  case ISD::RESET_FPMODE:
+    return LowerRESET_FPMODE(Op, DAG);
   case ISD::MUL:
     return LowerMUL(Op, DAG);
   case ISD::MULHS:
@@ -13896,16 +13949,14 @@ SDValue AArch64TargetLowering::LowerINSERT_SUBVECTOR(SDValue Op,
                        DAG.getVectorIdxConstant(0, DL));
       Hi = DAG.getNode(ISD::EXTRACT_SUBVECTOR, DL, HalfVT, Vec0,
                        DAG.getVectorIdxConstant(NumElts / 2, DL));
-      if (Idx < (NumElts / 2)) {
-        SDValue NewLo = DAG.getNode(ISD::INSERT_SUBVECTOR, DL, HalfVT, Lo, Vec1,
-                                    DAG.getVectorIdxConstant(Idx, DL));
-        return DAG.getNode(AArch64ISD::UZP1, DL, VT, NewLo, Hi);
-      } else {
-        SDValue NewHi =
-            DAG.getNode(ISD::INSERT_SUBVECTOR, DL, HalfVT, Hi, Vec1,
-                        DAG.getVectorIdxConstant(Idx - (NumElts / 2), DL));
-        return DAG.getNode(AArch64ISD::UZP1, DL, VT, Lo, NewHi);
-      }
+      if (Idx < (NumElts / 2))
+        Lo = DAG.getNode(ISD::INSERT_SUBVECTOR, DL, HalfVT, Lo, Vec1,
+                         DAG.getVectorIdxConstant(Idx, DL));
+      else
+        Hi = DAG.getNode(ISD::INSERT_SUBVECTOR, DL, HalfVT, Hi, Vec1,
+                         DAG.getVectorIdxConstant(Idx - (NumElts / 2), DL));
+
+      return DAG.getNode(ISD::CONCAT_VECTORS, DL, VT, Lo, Hi);
     }
 
     // Ensure the subvector is half the size of the main vector.
@@ -16347,7 +16398,7 @@ bool AArch64TargetLowering::lowerInterleavedStore(StoreInst *SI,
 bool AArch64TargetLowering::lowerDeinterleaveIntrinsicToLoad(
     IntrinsicInst *DI, LoadInst *LI) const {
   // Only deinterleave2 supported at present.
-  if (DI->getIntrinsicID() != Intrinsic::experimental_vector_deinterleave2)
+  if (DI->getIntrinsicID() != Intrinsic::vector_deinterleave2)
     return false;
 
   // Only a factor of 2 supported at present.
@@ -16422,7 +16473,7 @@ bool AArch64TargetLowering::lowerDeinterleaveIntrinsicToLoad(
 bool AArch64TargetLowering::lowerInterleaveIntrinsicToStore(
     IntrinsicInst *II, StoreInst *SI) const {
   // Only interleave2 supported at present.
-  if (II->getIntrinsicID() != Intrinsic::experimental_vector_interleave2)
+  if (II->getIntrinsicID() != Intrinsic::vector_interleave2)
     return false;
 
   // Only a factor of 2 supported at present.
@@ -17602,12 +17653,32 @@ static SDValue performMulCombine(SDNode *N, SelectionDAG &DAG,
     return false;
   };
 
+  // Can the const C be decomposed into (2^M + 1) * 2^N + 1), eg:
+  // C = 11 is equal to (1+4)*2+1, we don't decompose it into (1+2)*4-1 as
+  // the (2^N - 1) can't be execused via a single instruction.
+  auto isPowPlusPlusOneConst = [](APInt C, APInt &M, APInt &N) {
+    APInt CVMinus1 = C - 1;
+    if (CVMinus1.isNegative())
+      return false;
+    unsigned TrailingZeroes = CVMinus1.countr_zero();
+    APInt SCVMinus1 = CVMinus1.ashr(TrailingZeroes) - 1;
+    if (SCVMinus1.isPowerOf2()) {
+      unsigned BitWidth = SCVMinus1.getBitWidth();
+      M = APInt(BitWidth, SCVMinus1.logBase2());
+      N = APInt(BitWidth, TrailingZeroes);
+      return true;
+    }
+    return false;
+  };
+
   if (ConstValue.isNonNegative()) {
     // (mul x, (2^N + 1) * 2^M) => (shl (add (shl x, N), x), M)
     // (mul x, 2^N - 1) => (sub (shl x, N), x)
     // (mul x, (2^(N-M) - 1) * 2^M) => (sub (shl x, N), (shl x, M))
     // (mul x, (2^M + 1) * (2^N + 1))
     //     => MV = (add (shl x, M), x); (add (shl MV, N), MV)
+    // (mul x, (2^M + 1) * 2^N + 1))
+    //     =>  MV = add (shl x, M), x); add (shl MV, N), x)
     APInt SCVMinus1 = ShiftedConstValue - 1;
     APInt SCVPlus1 = ShiftedConstValue + 1;
     APInt CVPlus1 = ConstValue + 1;
@@ -17621,18 +17692,29 @@ static SDValue performMulCombine(SDNode *N, SelectionDAG &DAG,
     } else if (SCVPlus1.isPowerOf2()) {
       ShiftAmt = SCVPlus1.logBase2() + TrailingZeroes;
       return Sub(Shl(N0, ShiftAmt), Shl(N0, TrailingZeroes));
-    } else if (Subtarget->hasALULSLFast() &&
-               isPowPlusPlusConst(ConstValue, CVM, CVN)) {
+    }
+    if (Subtarget->hasALULSLFast() &&
+        isPowPlusPlusConst(ConstValue, CVM, CVN)) {
       APInt CVMMinus1 = CVM - 1;
       APInt CVNMinus1 = CVN - 1;
       unsigned ShiftM1 = CVMMinus1.logBase2();
       unsigned ShiftN1 = CVNMinus1.logBase2();
-      // LSLFast implicate that Shifts <= 3 places are fast
-      if (ShiftM1 <= 3 && ShiftN1 <= 3) {
+      // ALULSLFast implicate that Shifts <= 4 places are fast
+      if (ShiftM1 <= 4 && ShiftN1 <= 4) {
         SDValue MVal = Add(Shl(N0, ShiftM1), N0);
         return Add(Shl(MVal, ShiftN1), MVal);
       }
     }
+    if (Subtarget->hasALULSLFast() &&
+        isPowPlusPlusOneConst(ConstValue, CVM, CVN)) {
+      unsigned ShiftM = CVM.getZExtValue();
+      unsigned ShiftN = CVN.getZExtValue();
+      // ALULSLFast implicate that Shifts <= 4 places are fast
+      if (ShiftM <= 4 && ShiftN <= 4) {
+        SDValue MVal = Add(Shl(N0, CVM.getZExtValue()), N0);
+        return Add(Shl(MVal, CVN.getZExtValue()), N0);
+      }
+    }
   } else {
     // (mul x, -(2^N - 1)) => (sub x, (shl x, N))
     // (mul x, -(2^N + 1)) => - (add (shl x, N), x)
@@ -18649,14 +18731,12 @@ static SDValue performConcatVectorsCombine(SDNode *N,
   if (DCI.isBeforeLegalizeOps())
     return SDValue();
 
-  // Optimise concat_vectors of two [us]avgceils or [us]avgfloors with a 128-bit
-  // destination size, combine into an avg of two contacts of the source
-  // vectors. eg: concat(uhadd(a,b), uhadd(c, d)) -> uhadd(concat(a, c),
-  // concat(b, d))
+  // Optimise concat_vectors of two identical binops with a 128-bit destination
+  // size, combine into an binop of two contacts of the source vectors. eg:
+  // concat(uhadd(a,b), uhadd(c, d)) -> uhadd(concat(a, c), concat(b, d))
   if (N->getNumOperands() == 2 && N0Opc == N1Opc && VT.is128BitVector() &&
-      (N0Opc == ISD::AVGCEILU || N0Opc == ISD::AVGCEILS ||
-       N0Opc == ISD::AVGFLOORU || N0Opc == ISD::AVGFLOORS) &&
-      N0->hasOneUse() && N1->hasOneUse()) {
+      DAG.getTargetLoweringInfo().isBinOp(N0Opc) && N0->hasOneUse() &&
+      N1->hasOneUse()) {
     SDValue N00 = N0->getOperand(0);
     SDValue N01 = N0->getOperand(1);
     SDValue N10 = N1->getOperand(0);
diff --git a/llvm/lib/Target/AArch64/AArch64ISelLowering.h b/llvm/lib/Target/AArch64/AArch64ISelLowering.h
index db6e8a00d2fb..fbdc4de5617f 100644
--- a/llvm/lib/Target/AArch64/AArch64ISelLowering.h
+++ b/llvm/lib/Target/AArch64/AArch64ISelLowering.h
@@ -523,6 +523,9 @@ enum Rounding {
 // Bit position of rounding mode bits in FPCR.
 const unsigned RoundingBitsPos = 22;
 
+// Reserved bits should be preserved when modifying FPCR.
+const uint64_t ReservedFPControlBits = 0xfffffffff80040f8;
+
 // Registers used to pass function arguments.
 ArrayRef<MCPhysReg> getGPRArgRegs();
 ArrayRef<MCPhysReg> getFPRArgRegs();
@@ -1013,7 +1016,7 @@ private:
   bool isExtFreeImpl(const Instruction *Ext) const override;
 
   void addTypeForNEON(MVT VT);
-  void addTypeForFixedLengthSVE(MVT VT, bool StreamingSVE);
+  void addTypeForFixedLengthSVE(MVT VT);
   void addDRTypeForNEON(MVT VT);
   void addQRTypeForNEON(MVT VT);
 
@@ -1128,6 +1131,9 @@ private:
   SDValue LowerRETURNADDR(SDValue Op, SelectionDAG &DAG) const;
   SDValue LowerGET_ROUNDING(SDValue Op, SelectionDAG &DAG) const;
   SDValue LowerSET_ROUNDING(SDValue Op, SelectionDAG &DAG) const;
+  SDValue LowerGET_FPMODE(SDValue Op, SelectionDAG &DAG) const;
+  SDValue LowerSET_FPMODE(SDValue Op, SelectionDAG &DAG) const;
+  SDValue LowerRESET_FPMODE(SDValue Op, SelectionDAG &DAG) const;
   SDValue LowerINSERT_VECTOR_ELT(SDValue Op, SelectionDAG &DAG) const;
   SDValue LowerEXTRACT_VECTOR_ELT(SDValue Op, SelectionDAG &DAG) const;
   SDValue LowerBUILD_VECTOR(SDValue Op, SelectionDAG &DAG) const;
diff --git a/llvm/lib/Target/AArch64/AArch64InstrInfo.cpp b/llvm/lib/Target/AArch64/AArch64InstrInfo.cpp
index 9518d573bccd..55fecc4b4845 100644
--- a/llvm/lib/Target/AArch64/AArch64InstrInfo.cpp
+++ b/llvm/lib/Target/AArch64/AArch64InstrInfo.cpp
@@ -643,7 +643,8 @@ static unsigned canFoldIntoCSel(const MachineRegisterInfo &MRI, unsigned VReg,
   case AArch64::ADDSXri:
   case AArch64::ADDSWri:
     // if NZCV is used, do not fold.
-    if (DefMI->findRegisterDefOperandIdx(AArch64::NZCV, true) == -1)
+    if (DefMI->findRegisterDefOperandIdx(AArch64::NZCV, /*TRI=*/nullptr,
+                                         true) == -1)
       return 0;
     // fall-through to ADDXri and ADDWri.
     [[fallthrough]];
@@ -671,7 +672,8 @@ static unsigned canFoldIntoCSel(const MachineRegisterInfo &MRI, unsigned VReg,
   case AArch64::SUBSXrr:
   case AArch64::SUBSWrr:
     // if NZCV is used, do not fold.
-    if (DefMI->findRegisterDefOperandIdx(AArch64::NZCV, true) == -1)
+    if (DefMI->findRegisterDefOperandIdx(AArch64::NZCV, /*TRI=*/nullptr,
+                                         true) == -1)
       return 0;
     // fall-through to SUBXrr and SUBWrr.
     [[fallthrough]];
@@ -1275,7 +1277,8 @@ static unsigned convertToNonFlagSettingOpc(const MachineInstr &MI) {
   // Don't convert all compare instructions, because for some the zero register
   // encoding becomes the sp register.
   bool MIDefinesZeroReg = false;
-  if (MI.definesRegister(AArch64::WZR) || MI.definesRegister(AArch64::XZR))
+  if (MI.definesRegister(AArch64::WZR, /*TRI=*/nullptr) ||
+      MI.definesRegister(AArch64::XZR, /*TRI=*/nullptr))
     MIDefinesZeroReg = true;
 
   switch (MI.getOpcode()) {
@@ -1519,10 +1522,11 @@ bool AArch64InstrInfo::optimizeCompareInstr(
   assert(MRI);
 
   // Replace SUBSWrr with SUBWrr if NZCV is not used.
-  int DeadNZCVIdx = CmpInstr.findRegisterDefOperandIdx(AArch64::NZCV, true);
+  int DeadNZCVIdx =
+      CmpInstr.findRegisterDefOperandIdx(AArch64::NZCV, /*TRI=*/nullptr, true);
   if (DeadNZCVIdx != -1) {
-    if (CmpInstr.definesRegister(AArch64::WZR) ||
-        CmpInstr.definesRegister(AArch64::XZR)) {
+    if (CmpInstr.definesRegister(AArch64::WZR, /*TRI=*/nullptr) ||
+        CmpInstr.definesRegister(AArch64::XZR, /*TRI=*/nullptr)) {
       CmpInstr.eraseFromParent();
       return true;
     }
@@ -1623,7 +1627,7 @@ findCondCodeUseOperandIdxForBranchOrSelect(const MachineInstr &Instr) {
     return -1;
 
   case AArch64::Bcc: {
-    int Idx = Instr.findRegisterUseOperandIdx(AArch64::NZCV);
+    int Idx = Instr.findRegisterUseOperandIdx(AArch64::NZCV, /*TRI=*/nullptr);
     assert(Idx >= 2);
     return Idx - 2;
   }
@@ -1638,7 +1642,7 @@ findCondCodeUseOperandIdxForBranchOrSelect(const MachineInstr &Instr) {
   case AArch64::CSNEGXr:
   case AArch64::FCSELSrrr:
   case AArch64::FCSELDrrr: {
-    int Idx = Instr.findRegisterUseOperandIdx(AArch64::NZCV);
+    int Idx = Instr.findRegisterUseOperandIdx(AArch64::NZCV, /*TRI=*/nullptr);
     assert(Idx >= 1);
     return Idx - 1;
   }
@@ -1846,7 +1850,7 @@ static bool canCmpInstrBeRemoved(MachineInstr &MI, MachineInstr &CmpInstr,
     return false;
 
   // NZCV needs to be defined
-  if (MI.findRegisterDefOperandIdx(AArch64::NZCV, true) != -1)
+  if (MI.findRegisterDefOperandIdx(AArch64::NZCV, /*TRI=*/nullptr, true) != -1)
     return false;
 
   // CmpInstr is 'ADDS %vreg, 0' or 'SUBS %vreg, 0' or 'SUBS %vreg, 1'
@@ -5913,7 +5917,7 @@ static bool canCombine(MachineBasicBlock &MBB, MachineOperand &MO,
   }
 
   if (isCombineInstrSettingFlag(CombineOpc) &&
-      MI->findRegisterDefOperandIdx(AArch64::NZCV, true) == -1)
+      MI->findRegisterDefOperandIdx(AArch64::NZCV, /*TRI=*/nullptr, true) == -1)
     return false;
 
   return true;
@@ -6051,7 +6055,8 @@ static bool getMaddPatterns(MachineInstr &Root,
   if (!isCombineInstrCandidate(Opc))
     return false;
   if (isCombineInstrSettingFlag(Opc)) {
-    int Cmp_NZCV = Root.findRegisterDefOperandIdx(AArch64::NZCV, true);
+    int Cmp_NZCV =
+        Root.findRegisterDefOperandIdx(AArch64::NZCV, /*TRI=*/nullptr, true);
     // When NZCV is live bail out.
     if (Cmp_NZCV == -1)
       return false;
@@ -6546,7 +6551,8 @@ static bool getMiscPatterns(MachineInstr &Root,
   }
 
   if (isCombineInstrSettingFlag(Opc) &&
-      Root.findRegisterDefOperandIdx(AArch64::NZCV, true) == -1)
+      Root.findRegisterDefOperandIdx(AArch64::NZCV, /*TRI=*/nullptr, true) ==
+          -1)
     return false;
 
   if (canCombine(MBB, Root.getOperand(2), AArch64::ADDWrr) ||
@@ -6918,19 +6924,26 @@ genSubAdd2SubSub(MachineFunction &MF, MachineRegisterInfo &MRI,
     assert((Opcode == AArch64::SUBWrr || Opcode == AArch64::SUBXrr) &&
            "Unexpected instruction opcode.");
 
+  uint32_t Flags = Root.mergeFlagsWith(*AddMI);
+  Flags &= ~MachineInstr::NoSWrap;
+  Flags &= ~MachineInstr::NoUWrap;
+
   MachineInstrBuilder MIB1 =
       BuildMI(MF, MIMetadata(Root), TII->get(Opcode), NewVR)
           .addReg(RegA, getKillRegState(RegAIsKill))
-          .addReg(RegB, getKillRegState(RegBIsKill));
+          .addReg(RegB, getKillRegState(RegBIsKill))
+          .setMIFlags(Flags);
   MachineInstrBuilder MIB2 =
       BuildMI(MF, MIMetadata(Root), TII->get(Opcode), ResultReg)
           .addReg(NewVR, getKillRegState(true))
-          .addReg(RegC, getKillRegState(RegCIsKill));
+          .addReg(RegC, getKillRegState(RegCIsKill))
+          .setMIFlags(Flags);
 
   InstrIdxForVirtReg.insert(std::make_pair(NewVR, 0));
   InsInstrs.push_back(MIB1);
   InsInstrs.push_back(MIB2);
   DelInstrs.push_back(AddMI);
+  DelInstrs.push_back(&Root);
 }
 
 /// When getMachineCombinerPatterns() finds potential patterns,
@@ -6960,13 +6973,13 @@ void AArch64InstrInfo::genAlternativeCodeSequence(
     // ==> (A - B) - C
     genSubAdd2SubSub(MF, MRI, TII, Root, InsInstrs, DelInstrs, 1,
                      InstrIdxForVirtReg);
-    break;
+    return;
   case AArch64MachineCombinerPattern::SUBADD_OP2:
     // A - (B + C)
     // ==> (A - C) - B
     genSubAdd2SubSub(MF, MRI, TII, Root, InsInstrs, DelInstrs, 2,
                      InstrIdxForVirtReg);
-    break;
+    return;
   case AArch64MachineCombinerPattern::MULADDW_OP1:
   case AArch64MachineCombinerPattern::MULADDX_OP1:
     // MUL I=A,B,0
@@ -8031,7 +8044,8 @@ bool AArch64InstrInfo::optimizeCondBranch(MachineInstr &MI) const {
           DefMI->getOperand(2).getReg() == AArch64::XZR))
       return false;
 
-    if (DefMI->findRegisterDefOperandIdx(AArch64::NZCV, true) != -1)
+    if (DefMI->findRegisterDefOperandIdx(AArch64::NZCV, /*TRI=*/nullptr,
+                                         true) != -1)
       return false;
 
     AArch64CC::CondCode CC = (AArch64CC::CondCode)DefMI->getOperand(3).getImm();
@@ -9238,7 +9252,8 @@ AArch64InstrInfo::isCopyInstrImpl(const MachineInstr &MI) const {
        MI.getOperand(0).getSubReg() == 0) &&
       (!MI.getOperand(0).getReg().isPhysical() ||
        MI.findRegisterDefOperandIdx(MI.getOperand(0).getReg() - AArch64::W0 +
-                                    AArch64::X0) == -1))
+                                        AArch64::X0,
+                                    /*TRI=*/nullptr) == -1))
     return DestSourcePair{MI.getOperand(0), MI.getOperand(2)};
 
   if (MI.getOpcode() == AArch64::ORRXrs &&
diff --git a/llvm/lib/Target/AArch64/AArch64InstrInfo.h b/llvm/lib/Target/AArch64/AArch64InstrInfo.h
index 9a2914891675..f434799c3982 100644
--- a/llvm/lib/Target/AArch64/AArch64InstrInfo.h
+++ b/llvm/lib/Target/AArch64/AArch64InstrInfo.h
@@ -725,6 +725,7 @@ static inline unsigned getAUTOpcodeForKey(AArch64PACKey::ID K, bool Zero) {
   case DA: return Zero ? AArch64::AUTDZA : AArch64::AUTDA;
   case DB: return Zero ? AArch64::AUTDZB : AArch64::AUTDB;
   }
+  llvm_unreachable("Unhandled AArch64PACKey::ID enum");
 }
 
 /// Return PAC opcode to be used for a ptrauth sign using the given key, or its
@@ -737,6 +738,7 @@ static inline unsigned getPACOpcodeForKey(AArch64PACKey::ID K, bool Zero) {
   case DA: return Zero ? AArch64::PACDZA : AArch64::PACDA;
   case DB: return Zero ? AArch64::PACDZB : AArch64::PACDB;
   }
+  llvm_unreachable("Unhandled AArch64PACKey::ID enum");
 }
 
 // struct TSFlags {
diff --git a/llvm/lib/Target/AArch64/AArch64InstrInfo.td b/llvm/lib/Target/AArch64/AArch64InstrInfo.td
index ebddbefeb94f..2159116d1ab7 100644
--- a/llvm/lib/Target/AArch64/AArch64InstrInfo.td
+++ b/llvm/lib/Target/AArch64/AArch64InstrInfo.td
@@ -10,9 +10,14 @@
 //
 //===----------------------------------------------------------------------===//
 
+
 //===----------------------------------------------------------------------===//
 // ARM Instruction Predicate Definitions.
 //
+
+class AssemblerPredicateWithAll<dag cond, string name="">
+    : AssemblerPredicate<(any_of FeatureAll, cond), name>;
+
 def HasV8_0a         : Predicate<"Subtarget->hasV8_0aOps()">,
                                  AssemblerPredicate<(all_of HasV8_0aOps), "armv8.0a">;
 def HasV8_1a         : Predicate<"Subtarget->hasV8_1aOps()">,
@@ -225,6 +230,12 @@ def HasSVE2p1_or_HasSME2
 def HasSVE2p1_or_HasSME2p1
     : Predicate<"Subtarget->hasSVE2p1() || Subtarget->hasSME2p1()">,
                  AssemblerPredicateWithAll<(any_of FeatureSME2p1, FeatureSVE2p1), "sme2p1 or sve2p1">;
+
+def HasSMEF16F16orSMEF8F16
+    : Predicate<"Subtarget->hasSMEF16F16() || Subtarget->hasSMEF8F16()">,
+                AssemblerPredicateWithAll<(any_of FeatureSMEF16F16, FeatureSMEF8F16),
+                "sme-f16f16 or sme-f8f16">;
+
 // A subset of NEON instructions are legal in Streaming SVE execution mode,
 // they should be enabled if either has been specified.
 def HasNEONorSME
@@ -310,6 +321,8 @@ def UseNegativeImmediates
 
 def UseScalarIncVL : Predicate<"Subtarget->useScalarIncVL()">;
 
+def NoUseScalarIncVL : Predicate<"!Subtarget->useScalarIncVL()">;
+
 def UseSVEFPLD1R : Predicate<"!Subtarget->noSVEFPLD1R()">;
 
 def IsNeonAvailable : Predicate<"Subtarget->isNeonAvailable()">;
@@ -733,6 +746,8 @@ def AArch64vsli : SDNode<"AArch64ISD::VSLI", SDT_AArch64vshiftinsert>;
 def AArch64vsri : SDNode<"AArch64ISD::VSRI", SDT_AArch64vshiftinsert>;
 
 def AArch64bsp: SDNode<"AArch64ISD::BSP", SDT_AArch64trivec>;
+def AArch64nbsl: PatFrag<(ops node:$Op1, node:$Op2, node:$Op3),
+                         (vnot (AArch64bsp node:$Op1, node:$Op2, node:$Op3))>;
 
 def AArch64cmeq: SDNode<"AArch64ISD::CMEQ", SDT_AArch64binvec>;
 def AArch64cmge: SDNode<"AArch64ISD::CMGE", SDT_AArch64binvec>;
diff --git a/llvm/lib/Target/AArch64/AArch64MacroFusion.cpp b/llvm/lib/Target/AArch64/AArch64MacroFusion.cpp
index 05d60872bf51..ff7a0d1faedf 100644
--- a/llvm/lib/Target/AArch64/AArch64MacroFusion.cpp
+++ b/llvm/lib/Target/AArch64/AArch64MacroFusion.cpp
@@ -245,7 +245,7 @@ static bool isCCSelectPair(const MachineInstr *FirstMI,
     if (FirstMI == nullptr)
       return true;
 
-    if (FirstMI->definesRegister(AArch64::WZR))
+    if (FirstMI->definesRegister(AArch64::WZR, /*TRI=*/nullptr))
       switch (FirstMI->getOpcode()) {
       case AArch64::SUBSWrs:
         return !AArch64InstrInfo::hasShiftedReg(*FirstMI);
@@ -263,7 +263,7 @@ static bool isCCSelectPair(const MachineInstr *FirstMI,
     if (FirstMI == nullptr)
       return true;
 
-    if (FirstMI->definesRegister(AArch64::XZR))
+    if (FirstMI->definesRegister(AArch64::XZR, /*TRI=*/nullptr))
       switch (FirstMI->getOpcode()) {
       case AArch64::SUBSXrs:
         return !AArch64InstrInfo::hasShiftedReg(*FirstMI);
diff --git a/llvm/lib/Target/AArch64/AArch64PerfectShuffle.h b/llvm/lib/Target/AArch64/AArch64PerfectShuffle.h
index 7abaead694d1..a143243a8d3b 100644
--- a/llvm/lib/Target/AArch64/AArch64PerfectShuffle.h
+++ b/llvm/lib/Target/AArch64/AArch64PerfectShuffle.h
@@ -6620,11 +6620,28 @@ static unsigned getPerfectShuffleCost(llvm::ArrayRef<int> M) {
   return (PFEntry >> 30) + 1;
 }
 
-inline bool isZIPMask(ArrayRef<int> M, EVT VT, unsigned &WhichResult) {
+/// Return true for zip1 or zip2 masks of the form:
+///  <0,  8, 1,  9, 2, 10, 3, 11> or
+///  <4, 12, 5, 13, 6, 14, 7, 15>
+inline bool isZIPMask(ArrayRef<int> M, EVT VT, unsigned &WhichResultOut) {
   unsigned NumElts = VT.getVectorNumElements();
   if (NumElts % 2 != 0)
     return false;
-  WhichResult = (M[0] == 0 ? 0 : 1);
+  // Check the first non-undef element for which half to use.
+  unsigned WhichResult = 2;
+  for (unsigned i = 0; i != NumElts / 2; i++) {
+    if (M[i * 2] >= 0) {
+      WhichResult = ((unsigned)M[i * 2] == i ? 0 : 1);
+      break;
+    } else if (M[i * 2 + 1] >= 0) {
+      WhichResult = ((unsigned)M[i * 2 + 1] == NumElts + i ? 0 : 1);
+      break;
+    }
+  }
+  if (WhichResult == 2)
+    return false;
+
+  // Check all elements match.
   unsigned Idx = WhichResult * NumElts / 2;
   for (unsigned i = 0; i != NumElts; i += 2) {
     if ((M[i] >= 0 && (unsigned)M[i] != Idx) ||
@@ -6632,20 +6649,34 @@ inline bool isZIPMask(ArrayRef<int> M, EVT VT, unsigned &WhichResult) {
       return false;
     Idx += 1;
   }
-
+  WhichResultOut = WhichResult;
   return true;
 }
 
-inline bool isUZPMask(ArrayRef<int> M, EVT VT, unsigned &WhichResult) {
+/// Return true for uzp1 or uzp2 masks of the form:
+///  <0, 2, 4, 6, 8, 10, 12, 14> or
+///  <1, 3, 5, 7, 9, 11, 13, 15>
+inline bool isUZPMask(ArrayRef<int> M, EVT VT, unsigned &WhichResultOut) {
   unsigned NumElts = VT.getVectorNumElements();
-  WhichResult = (M[0] == 0 ? 0 : 1);
+  // Check the first non-undef element for which half to use.
+  unsigned WhichResult = 2;
+  for (unsigned i = 0; i != NumElts; i++) {
+    if (M[i] >= 0) {
+      WhichResult = ((unsigned)M[i] == i * 2 ? 0 : 1);
+      break;
+    }
+  }
+  if (WhichResult == 2)
+    return false;
+
+  // Check all elements match.
   for (unsigned i = 0; i != NumElts; ++i) {
     if (M[i] < 0)
       continue; // ignore UNDEF indices
     if ((unsigned)M[i] != 2 * i + WhichResult)
       return false;
   }
-
+  WhichResultOut = WhichResult;
   return true;
 }
 
diff --git a/llvm/lib/Target/AArch64/AArch64Processors.td b/llvm/lib/Target/AArch64/AArch64Processors.td
new file mode 100644
index 000000000000..8772e51bf0ab
--- /dev/null
+++ b/llvm/lib/Target/AArch64/AArch64Processors.td
@@ -0,0 +1,976 @@
+//=- AArch64Processors.td - Describe AArch64 Processors ------*- tablegen -*-=//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+//
+//===----------------------------------------------------------------------===//
+
+//===----------------------------------------------------------------------===//
+// AArch64 Processor subtarget features.
+//===----------------------------------------------------------------------===//
+
+
+def TuneA35     : SubtargetFeature<"a35", "ARMProcFamily", "CortexA35",
+                                "Cortex-A35 ARM processors">;
+
+def TuneA53     : SubtargetFeature<"a53", "ARMProcFamily", "CortexA53",
+                                   "Cortex-A53 ARM processors", [
+                                   FeatureFuseAES,
+                                   FeatureFuseAdrpAdd,
+                                   FeatureBalanceFPOps,
+                                   FeaturePostRAScheduler]>;
+
+def TuneA55     : SubtargetFeature<"a55", "ARMProcFamily", "CortexA55",
+                                   "Cortex-A55 ARM processors", [
+                                   FeatureFuseAES,
+                                   FeatureFuseAdrpAdd,
+                                   FeaturePostRAScheduler,
+                                   FeatureFuseAddress]>;
+
+def TuneA510    : SubtargetFeature<"a510", "ARMProcFamily", "CortexA510",
+                                   "Cortex-A510 ARM processors", [
+                                   FeatureFuseAES,
+                                   FeatureFuseAdrpAdd,
+                                   FeaturePostRAScheduler
+                                   ]>;
+
+def TuneA520    : SubtargetFeature<"a520", "ARMProcFamily", "CortexA520",
+                                   "Cortex-A520 ARM processors", [
+                                   FeatureFuseAES,
+                                   FeatureFuseAdrpAdd,
+                                   FeaturePostRAScheduler]>;
+
+def TuneA520AE  : SubtargetFeature<"a520ae", "ARMProcFamily", "CortexA520",
+                                   "Cortex-A520AE ARM processors", [
+                                   FeatureFuseAES,
+                                   FeatureFuseAdrpAdd,
+                                   FeaturePostRAScheduler]>;
+
+def TuneA57     : SubtargetFeature<"a57", "ARMProcFamily", "CortexA57",
+                                   "Cortex-A57 ARM processors", [
+                                   FeatureFuseAES,
+                                   FeatureBalanceFPOps,
+                                   FeatureFuseAdrpAdd,
+                                   FeatureFuseLiterals,
+                                   FeatureAddrLSLSlow14,
+                                   FeaturePostRAScheduler,
+                                   FeatureEnableSelectOptimize,
+                                   FeaturePredictableSelectIsExpensive]>;
+
+def TuneA65     : SubtargetFeature<"a65", "ARMProcFamily", "CortexA65",
+                                   "Cortex-A65 ARM processors", [
+                                   FeatureFuseAES,
+                                   FeatureFuseAddress,
+                                   FeatureFuseAdrpAdd,
+                                   FeatureFuseLiterals,
+                                   FeatureEnableSelectOptimize,
+                                   FeaturePredictableSelectIsExpensive]>;
+
+def TuneA72     : SubtargetFeature<"a72", "ARMProcFamily", "CortexA72",
+                                   "Cortex-A72 ARM processors", [
+                                   FeatureFuseAES,
+                                   FeatureFuseAdrpAdd,
+                                   FeatureFuseLiterals,
+                                   FeatureAddrLSLSlow14,
+                                   FeatureEnableSelectOptimize,
+                                   FeaturePredictableSelectIsExpensive]>;
+
+def TuneA73     : SubtargetFeature<"a73", "ARMProcFamily", "CortexA73",
+                                   "Cortex-A73 ARM processors", [
+                                   FeatureFuseAES,
+                                   FeatureFuseAdrpAdd,
+                                   FeatureAddrLSLSlow14,
+                                   FeatureEnableSelectOptimize,
+                                   FeaturePredictableSelectIsExpensive]>;
+
+def TuneA75     : SubtargetFeature<"a75", "ARMProcFamily", "CortexA75",
+                                   "Cortex-A75 ARM processors", [
+                                   FeatureFuseAES,
+                                   FeatureFuseAdrpAdd,
+                                   FeatureAddrLSLSlow14,
+                                   FeatureEnableSelectOptimize,
+                                   FeaturePredictableSelectIsExpensive]>;
+
+def TuneA76     : SubtargetFeature<"a76", "ARMProcFamily", "CortexA76",
+                                   "Cortex-A76 ARM processors", [
+                                   FeatureFuseAES,
+                                   FeatureFuseAdrpAdd,
+                                   FeatureAddrLSLSlow14,
+                                   FeatureALULSLFast,
+                                   FeatureEnableSelectOptimize,
+                                   FeaturePredictableSelectIsExpensive]>;
+
+def TuneA77     : SubtargetFeature<"a77", "ARMProcFamily", "CortexA77",
+                                   "Cortex-A77 ARM processors", [
+                                   FeatureCmpBccFusion,
+                                   FeatureFuseAES,
+                                   FeatureFuseAdrpAdd,
+                                   FeatureAddrLSLSlow14,
+                                   FeatureALULSLFast,
+                                   FeatureEnableSelectOptimize,
+                                   FeaturePredictableSelectIsExpensive]>;
+
+def TuneA78 : SubtargetFeature<"a78", "ARMProcFamily", "CortexA78",
+                               "Cortex-A78 ARM processors", [
+                               FeatureCmpBccFusion,
+                               FeatureFuseAES,
+                               FeatureFuseAdrpAdd,
+                               FeatureAddrLSLSlow14,
+                               FeatureALULSLFast,
+                               FeaturePostRAScheduler,
+                               FeatureEnableSelectOptimize,
+                               FeaturePredictableSelectIsExpensive]>;
+
+def TuneA78AE : SubtargetFeature<"a78ae", "ARMProcFamily",
+                                 "CortexA78AE",
+                                 "Cortex-A78AE ARM processors", [
+                                 FeatureCmpBccFusion,
+                                 FeatureFuseAES,
+                                 FeatureFuseAdrpAdd,
+                                 FeatureAddrLSLSlow14,
+                                 FeatureALULSLFast,
+                                 FeaturePostRAScheduler,
+                                 FeatureEnableSelectOptimize,
+                                 FeaturePredictableSelectIsExpensive]>;
+
+def TuneA78C : SubtargetFeature<"a78c", "ARMProcFamily",
+                                "CortexA78C",
+                                "Cortex-A78C ARM processors", [
+                                FeatureCmpBccFusion,
+                                FeatureFuseAES,
+                                FeatureFuseAdrpAdd,
+                                FeatureAddrLSLSlow14,
+                                FeatureALULSLFast,
+                                FeaturePostRAScheduler,
+                                FeatureEnableSelectOptimize,
+                                FeaturePredictableSelectIsExpensive]>;
+
+def TuneA710    : SubtargetFeature<"a710", "ARMProcFamily", "CortexA710",
+                                   "Cortex-A710 ARM processors", [
+                                   FeatureCmpBccFusion,
+                                   FeatureFuseAES,
+                                   FeatureFuseAdrpAdd,
+                                   FeatureALULSLFast,
+                                   FeaturePostRAScheduler,
+                                   FeatureEnableSelectOptimize,
+                                   FeaturePredictableSelectIsExpensive]>;
+
+def TuneA715 : SubtargetFeature<"a715", "ARMProcFamily", "CortexA715",
+                                 "Cortex-A715 ARM processors", [
+                                 FeatureFuseAES,
+                                 FeaturePostRAScheduler,
+                                 FeatureCmpBccFusion,
+                                 FeatureALULSLFast,
+                                 FeatureFuseAdrpAdd,
+                                 FeatureEnableSelectOptimize,
+                                 FeaturePredictableSelectIsExpensive]>;
+
+def TuneA720 : SubtargetFeature<"a720", "ARMProcFamily", "CortexA720",
+                                "Cortex-A720 ARM processors", [
+                                 FeatureFuseAES,
+                                 FeaturePostRAScheduler,
+                                 FeatureCmpBccFusion,
+                                 FeatureALULSLFast,
+                                 FeatureFuseAdrpAdd,
+                                 FeatureEnableSelectOptimize,
+                                 FeaturePredictableSelectIsExpensive]>;
+
+def TuneA720AE : SubtargetFeature<"a720ae", "ARMProcFamily", "CortexA720",
+                                "Cortex-A720AE ARM processors", [
+                                 FeatureFuseAES,
+                                 FeaturePostRAScheduler,
+                                 FeatureCmpBccFusion,
+                                 FeatureALULSLFast,
+                                 FeatureFuseAdrpAdd,
+                                 FeatureEnableSelectOptimize,
+                                 FeaturePredictableSelectIsExpensive]>;
+
+def TuneR82 : SubtargetFeature<"cortex-r82", "ARMProcFamily",
+                               "CortexR82",
+                               "Cortex-R82 ARM processors", [
+                               FeaturePostRAScheduler]>;
+
+def TuneX1 : SubtargetFeature<"cortex-x1", "ARMProcFamily", "CortexX1",
+                                  "Cortex-X1 ARM processors", [
+                                  FeatureCmpBccFusion,
+                                  FeatureFuseAES,
+                                  FeatureFuseAdrpAdd,
+                                  FeatureAddrLSLSlow14,
+                                  FeatureALULSLFast,
+                                  FeaturePostRAScheduler,
+                                  FeatureEnableSelectOptimize,
+                                  FeaturePredictableSelectIsExpensive]>;
+
+def TuneX2 : SubtargetFeature<"cortex-x2", "ARMProcFamily", "CortexX2",
+                                  "Cortex-X2 ARM processors", [
+                                  FeatureCmpBccFusion,
+                                  FeatureFuseAES,
+                                  FeatureFuseAdrpAdd,
+                                  FeatureALULSLFast,
+                                  FeaturePostRAScheduler,
+                                  FeatureEnableSelectOptimize,
+                                  FeaturePredictableSelectIsExpensive]>;
+
+def TuneX3 : SubtargetFeature<"cortex-x3", "ARMProcFamily", "CortexX3",
+                              "Cortex-X3 ARM processors", [
+                               FeatureALULSLFast,
+                               FeatureFuseAdrpAdd,
+                               FeatureFuseAES,
+                               FeaturePostRAScheduler,
+                               FeatureEnableSelectOptimize,
+                               FeaturePredictableSelectIsExpensive]>;
+
+def TuneX4 : SubtargetFeature<"cortex-x4", "ARMProcFamily", "CortexX4",
+                              "Cortex-X4 ARM processors", [
+                               FeatureALULSLFast,
+                               FeatureFuseAdrpAdd,
+                               FeatureFuseAES,
+                               FeaturePostRAScheduler,
+                               FeatureEnableSelectOptimize,
+                               FeaturePredictableSelectIsExpensive]>;
+
+def TuneA64FX : SubtargetFeature<"a64fx", "ARMProcFamily", "A64FX",
+                                 "Fujitsu A64FX processors", [
+                                 FeaturePostRAScheduler,
+                                 FeatureAggressiveFMA,
+                                 FeatureArithmeticBccFusion,
+                                 FeatureStorePairSuppress,
+                                 FeaturePredictableSelectIsExpensive]>;
+
+def TuneCarmel : SubtargetFeature<"carmel", "ARMProcFamily", "Carmel",
+                                  "Nvidia Carmel processors">;
+
+// Note that cyclone does not fuse AES instructions, but newer apple chips do
+// perform the fusion and cyclone is used by default when targetting apple OSes.
+def TuneAppleA7  : SubtargetFeature<"apple-a7", "ARMProcFamily", "AppleA7",
+                                    "Apple A7 (the CPU formerly known as Cyclone)", [
+                                    FeatureAlternateSExtLoadCVTF32Pattern,
+                                    FeatureArithmeticBccFusion,
+                                    FeatureArithmeticCbzFusion,
+                                    FeatureDisableLatencySchedHeuristic,
+                                    FeatureFuseAES, FeatureFuseCryptoEOR,
+                                    FeatureStorePairSuppress,
+                                    FeatureZCRegMove,
+                                    FeatureZCZeroing,
+                                    FeatureZCZeroingFPWorkaround]>;
+
+def TuneAppleA10 : SubtargetFeature<"apple-a10", "ARMProcFamily", "AppleA10",
+                                    "Apple A10", [
+                                    FeatureAlternateSExtLoadCVTF32Pattern,
+                                    FeatureArithmeticBccFusion,
+                                    FeatureArithmeticCbzFusion,
+                                    FeatureDisableLatencySchedHeuristic,
+                                    FeatureFuseAES,
+                                    FeatureFuseCryptoEOR,
+                                    FeatureStorePairSuppress,
+                                    FeatureZCRegMove,
+                                    FeatureZCZeroing]>;
+
+def TuneAppleA11 : SubtargetFeature<"apple-a11", "ARMProcFamily", "AppleA11",
+                                    "Apple A11", [
+                                    FeatureAlternateSExtLoadCVTF32Pattern,
+                                    FeatureArithmeticBccFusion,
+                                    FeatureArithmeticCbzFusion,
+                                    FeatureDisableLatencySchedHeuristic,
+                                    FeatureFuseAES,
+                                    FeatureFuseCryptoEOR,
+                                    FeatureStorePairSuppress,
+                                    FeatureZCRegMove,
+                                    FeatureZCZeroing]>;
+
+def TuneAppleA12 : SubtargetFeature<"apple-a12", "ARMProcFamily", "AppleA12",
+                                    "Apple A12", [
+                                    FeatureAlternateSExtLoadCVTF32Pattern,
+                                    FeatureArithmeticBccFusion,
+                                    FeatureArithmeticCbzFusion,
+                                    FeatureDisableLatencySchedHeuristic,
+                                    FeatureFuseAES,
+                                    FeatureFuseCryptoEOR,
+                                    FeatureStorePairSuppress,
+                                    FeatureZCRegMove,
+                                    FeatureZCZeroing]>;
+
+def TuneAppleA13 : SubtargetFeature<"apple-a13", "ARMProcFamily", "AppleA13",
+                                    "Apple A13", [
+                                    FeatureAlternateSExtLoadCVTF32Pattern,
+                                    FeatureArithmeticBccFusion,
+                                    FeatureArithmeticCbzFusion,
+                                    FeatureDisableLatencySchedHeuristic,
+                                    FeatureFuseAES,
+                                    FeatureFuseCryptoEOR,
+                                    FeatureStorePairSuppress,
+                                    FeatureZCRegMove,
+                                    FeatureZCZeroing]>;
+
+def TuneAppleA14 : SubtargetFeature<"apple-a14", "ARMProcFamily", "AppleA14",
+                                    "Apple A14", [
+                                    FeatureAggressiveFMA,
+                                    FeatureAlternateSExtLoadCVTF32Pattern,
+                                    FeatureArithmeticBccFusion,
+                                    FeatureArithmeticCbzFusion,
+                                    FeatureDisableLatencySchedHeuristic,
+                                    FeatureFuseAddress,
+                                    FeatureFuseAES,
+                                    FeatureFuseArithmeticLogic,
+                                    FeatureFuseCCSelect,
+                                    FeatureFuseCryptoEOR,
+                                    FeatureFuseLiterals,
+                                    FeatureStorePairSuppress,
+                                    FeatureZCRegMove,
+                                    FeatureZCZeroing]>;
+
+def TuneAppleA15 : SubtargetFeature<"apple-a15", "ARMProcFamily", "AppleA15",
+                                    "Apple A15", [
+                                    FeatureAlternateSExtLoadCVTF32Pattern,
+                                    FeatureArithmeticBccFusion,
+                                    FeatureArithmeticCbzFusion,
+                                    FeatureDisableLatencySchedHeuristic,
+                                    FeatureFuseAddress,
+                                    FeatureFuseAES,
+                                    FeatureFuseArithmeticLogic,
+                                    FeatureFuseCCSelect,
+                                    FeatureFuseCryptoEOR,
+                                    FeatureFuseLiterals,
+                                    FeatureStorePairSuppress,
+                                    FeatureZCRegMove,
+                                    FeatureZCZeroing]>;
+
+def TuneAppleA16 : SubtargetFeature<"apple-a16", "ARMProcFamily", "AppleA16",
+                                    "Apple A16", [
+                                    FeatureAlternateSExtLoadCVTF32Pattern,
+                                    FeatureArithmeticBccFusion,
+                                    FeatureArithmeticCbzFusion,
+                                    FeatureDisableLatencySchedHeuristic,
+                                    FeatureFuseAddress,
+                                    FeatureFuseAdrpAdd,
+                                    FeatureFuseAES,
+                                    FeatureFuseArithmeticLogic,
+                                    FeatureFuseCCSelect,
+                                    FeatureFuseCryptoEOR,
+                                    FeatureFuseLiterals,
+                                    FeatureStorePairSuppress,
+                                    FeatureZCRegMove,
+                                    FeatureZCZeroing]>;
+
+def TuneAppleA17 : SubtargetFeature<"apple-a17", "ARMProcFamily", "AppleA17",
+                                    "Apple A17", [
+                                    FeatureAlternateSExtLoadCVTF32Pattern,
+                                    FeatureArithmeticBccFusion,
+                                    FeatureArithmeticCbzFusion,
+                                    FeatureDisableLatencySchedHeuristic,
+                                    FeatureFuseAddress,
+                                    FeatureFuseAdrpAdd,
+                                    FeatureFuseAES,
+                                    FeatureFuseArithmeticLogic,
+                                    FeatureFuseCCSelect,
+                                    FeatureFuseCryptoEOR,
+                                    FeatureFuseLiterals,
+                                    FeatureStorePairSuppress,
+                                    FeatureZCRegMove,
+                                    FeatureZCZeroing]>;
+
+def TuneExynosM3 : SubtargetFeature<"exynosm3", "ARMProcFamily", "ExynosM3",
+                                    "Samsung Exynos-M3 processors",
+                                    [FeatureExynosCheapAsMoveHandling,
+                                     FeatureForce32BitJumpTables,
+                                     FeatureFuseAddress,
+                                     FeatureFuseAES,
+                                     FeatureFuseCCSelect,
+                                     FeatureFuseAdrpAdd,
+                                     FeatureFuseLiterals,
+                                     FeatureStorePairSuppress,
+                                     FeatureALULSLFast,
+                                     FeaturePostRAScheduler,
+                                     FeaturePredictableSelectIsExpensive]>;
+
+// Re-uses some scheduling and tunings from the ExynosM3 proc family.
+def TuneExynosM4 : SubtargetFeature<"exynosm4", "ARMProcFamily", "ExynosM3",
+                                    "Samsung Exynos-M4 processors",
+                                    [FeatureArithmeticBccFusion,
+                                     FeatureArithmeticCbzFusion,
+                                     FeatureExynosCheapAsMoveHandling,
+                                     FeatureForce32BitJumpTables,
+                                     FeatureFuseAddress,
+                                     FeatureFuseAES,
+                                     FeatureFuseArithmeticLogic,
+                                     FeatureFuseCCSelect,
+                                     FeatureFuseAdrpAdd,
+                                     FeatureFuseLiterals,
+                                     FeatureStorePairSuppress,
+                                     FeatureALULSLFast,
+                                     FeaturePostRAScheduler,
+                                     FeatureZCZeroing]>;
+
+def TuneKryo    : SubtargetFeature<"kryo", "ARMProcFamily", "Kryo",
+                                   "Qualcomm Kryo processors", [
+                                   FeaturePostRAScheduler,
+                                   FeaturePredictableSelectIsExpensive,
+                                   FeatureZCZeroing,
+                                   FeatureALULSLFast,
+                                   FeatureStorePairSuppress]>;
+
+def TuneFalkor  : SubtargetFeature<"falkor", "ARMProcFamily", "Falkor",
+                                   "Qualcomm Falkor processors", [
+                                   FeaturePostRAScheduler,
+                                   FeaturePredictableSelectIsExpensive,
+                                   FeatureZCZeroing,
+                                   FeatureStorePairSuppress,
+                                   FeatureALULSLFast,
+                                   FeatureSlowSTRQro]>;
+
+def TuneNeoverseE1 : SubtargetFeature<"neoversee1", "ARMProcFamily", "NeoverseE1",
+                                      "Neoverse E1 ARM processors", [
+                                      FeatureFuseAES,
+                                      FeatureFuseAdrpAdd,
+                                      FeaturePostRAScheduler]>;
+
+def TuneNeoverseN1 : SubtargetFeature<"neoversen1", "ARMProcFamily", "NeoverseN1",
+                                      "Neoverse N1 ARM processors", [
+                                      FeatureFuseAES,
+                                      FeatureFuseAdrpAdd,
+                                      FeatureAddrLSLSlow14,
+                                      FeatureALULSLFast,
+                                      FeaturePostRAScheduler,
+                                      FeatureEnableSelectOptimize,
+                                      FeaturePredictableSelectIsExpensive]>;
+
+def TuneNeoverseN2 : SubtargetFeature<"neoversen2", "ARMProcFamily", "NeoverseN2",
+                                      "Neoverse N2 ARM processors", [
+                                      FeatureFuseAES,
+                                      FeatureFuseAdrpAdd,
+                                      FeatureALULSLFast,
+                                      FeaturePostRAScheduler,
+                                      FeatureEnableSelectOptimize,
+                                      FeaturePredictableSelectIsExpensive]>;
+
+def TuneNeoverseN3 : SubtargetFeature<"neoversen3", "ARMProcFamily", "NeoverseN3",
+                                      "Neoverse N3 ARM processors", [
+                                      FeatureFuseAES,
+                                      FeaturePostRAScheduler,
+                                      FeatureALULSLFast,
+                                      FeatureFuseAdrpAdd,
+                                      FeatureEnableSelectOptimize,
+                                      FeaturePredictableSelectIsExpensive]>;
+
+def TuneNeoverse512TVB : SubtargetFeature<"neoverse512tvb", "ARMProcFamily", "Neoverse512TVB",
+                                      "Neoverse 512-TVB ARM processors", [
+                                      FeatureFuseAES,
+                                      FeatureFuseAdrpAdd,
+                                      FeatureALULSLFast,
+                                      FeaturePostRAScheduler,
+                                      FeatureEnableSelectOptimize,
+                                      FeaturePredictableSelectIsExpensive]>;
+
+def TuneNeoverseV1 : SubtargetFeature<"neoversev1", "ARMProcFamily", "NeoverseV1",
+                                      "Neoverse V1 ARM processors", [
+                                      FeatureFuseAES,
+                                      FeatureFuseAdrpAdd,
+                                      FeatureAddrLSLSlow14,
+                                      FeatureALULSLFast,
+                                      FeaturePostRAScheduler,
+                                      FeatureEnableSelectOptimize,
+                                      FeaturePredictableSelectIsExpensive,
+                                      FeatureNoSVEFPLD1R]>;
+
+def TuneNeoverseV2 : SubtargetFeature<"neoversev2", "ARMProcFamily", "NeoverseV2",
+                                      "Neoverse V2 ARM processors", [
+                                      FeatureFuseAES,
+                                      FeatureFuseAdrpAdd,
+                                      FeatureALULSLFast,
+                                      FeaturePostRAScheduler,
+                                      FeatureEnableSelectOptimize,
+                                      FeaturePredictableSelectIsExpensive]>;
+
+def TuneNeoverseV3 : SubtargetFeature<"neoversev3", "ARMProcFamily", "NeoverseV3",
+                                      "Neoverse V3 ARM processors", [
+                                      FeatureFuseAES,
+                                      FeatureALULSLFast,
+                                      FeatureFuseAdrpAdd,
+                                      FeaturePostRAScheduler,
+                                      FeatureEnableSelectOptimize,
+                                      FeaturePredictableSelectIsExpensive]>;
+
+def TuneNeoverseV3AE : SubtargetFeature<"neoversev3AE", "ARMProcFamily", "NeoverseV3",
+                                      "Neoverse V3AE ARM processors", [
+                                      FeatureFuseAES,
+                                      FeatureALULSLFast,
+                                      FeatureFuseAdrpAdd,
+                                      FeaturePostRAScheduler,
+                                      FeatureEnableSelectOptimize,
+                                      FeaturePredictableSelectIsExpensive]>;
+
+def TuneSaphira  : SubtargetFeature<"saphira", "ARMProcFamily", "Saphira",
+                                   "Qualcomm Saphira processors", [
+                                   FeaturePostRAScheduler,
+                                   FeaturePredictableSelectIsExpensive,
+                                   FeatureZCZeroing,
+                                   FeatureStorePairSuppress,
+                                   FeatureALULSLFast]>;
+
+def TuneThunderX2T99  : SubtargetFeature<"thunderx2t99", "ARMProcFamily", "ThunderX2T99",
+                                         "Cavium ThunderX2 processors", [
+                                          FeatureAggressiveFMA,
+                                          FeatureArithmeticBccFusion,
+                                          FeaturePostRAScheduler,
+                                          FeatureStorePairSuppress,
+                                          FeaturePredictableSelectIsExpensive]>;
+
+def TuneThunderX3T110  : SubtargetFeature<"thunderx3t110", "ARMProcFamily",
+                                          "ThunderX3T110",
+                                          "Marvell ThunderX3 processors", [
+                                           FeatureAggressiveFMA,
+                                           FeatureArithmeticBccFusion,
+                                           FeaturePostRAScheduler,
+                                           FeaturePredictableSelectIsExpensive,
+                                           FeatureBalanceFPOps,
+                                           FeatureStorePairSuppress,
+                                           FeatureStrictAlign]>;
+
+def TuneThunderX : SubtargetFeature<"thunderx", "ARMProcFamily", "ThunderX",
+                                    "Cavium ThunderX processors", [
+                                    FeaturePostRAScheduler,
+                                    FeatureStorePairSuppress,
+                                    FeaturePredictableSelectIsExpensive]>;
+
+def TuneThunderXT88 : SubtargetFeature<"thunderxt88", "ARMProcFamily",
+                                       "ThunderXT88",
+                                       "Cavium ThunderX processors", [
+                                       FeaturePostRAScheduler,
+                                       FeatureStorePairSuppress,
+                                       FeaturePredictableSelectIsExpensive]>;
+
+def TuneThunderXT81 : SubtargetFeature<"thunderxt81", "ARMProcFamily",
+                                       "ThunderXT81",
+                                       "Cavium ThunderX processors", [
+                                       FeaturePostRAScheduler,
+                                       FeatureStorePairSuppress,
+                                       FeaturePredictableSelectIsExpensive]>;
+
+def TuneThunderXT83 : SubtargetFeature<"thunderxt83", "ARMProcFamily",
+                                       "ThunderXT83",
+                                       "Cavium ThunderX processors", [
+                                       FeaturePostRAScheduler,
+                                       FeatureStorePairSuppress,
+                                       FeaturePredictableSelectIsExpensive]>;
+
+def TuneTSV110 : SubtargetFeature<"tsv110", "ARMProcFamily", "TSV110",
+                                  "HiSilicon TS-V110 processors", [
+                                  FeatureFuseAES,
+                                  FeatureStorePairSuppress,
+                                  FeaturePostRAScheduler]>;
+
+def TuneAmpere1 : SubtargetFeature<"ampere1", "ARMProcFamily", "Ampere1",
+                                   "Ampere Computing Ampere-1 processors", [
+                                   FeaturePostRAScheduler,
+                                   FeatureFuseAES,
+                                   FeatureFuseAdrpAdd,
+                                   FeatureALULSLFast,
+                                   FeatureAggressiveFMA,
+                                   FeatureArithmeticBccFusion,
+                                   FeatureCmpBccFusion,
+                                   FeatureFuseAddress,
+                                   FeatureFuseLiterals,
+                                   FeatureStorePairSuppress,
+                                   FeatureLdpAlignedOnly,
+                                   FeatureStpAlignedOnly]>;
+
+def TuneAmpere1A : SubtargetFeature<"ampere1a", "ARMProcFamily", "Ampere1A",
+                                    "Ampere Computing Ampere-1A processors", [
+                                    FeaturePostRAScheduler,
+                                    FeatureFuseAES,
+                                    FeatureFuseAdrpAdd,
+                                    FeatureALULSLFast,
+                                    FeatureAggressiveFMA,
+                                    FeatureArithmeticBccFusion,
+                                    FeatureCmpBccFusion,
+                                    FeatureFuseAddress,
+                                    FeatureFuseLiterals,
+                                    FeatureFuseAddSub2RegAndConstOne,
+                                    FeatureStorePairSuppress,
+                                    FeatureLdpAlignedOnly,
+                                    FeatureStpAlignedOnly]>;
+
+def TuneAmpere1B : SubtargetFeature<"ampere1b", "ARMProcFamily", "Ampere1B",
+                                    "Ampere Computing Ampere-1B processors", [
+                                    FeaturePostRAScheduler,
+                                    FeatureFuseAES,
+                                    FeatureFuseAdrpAdd,
+                                    FeatureALULSLFast,
+                                    FeatureAggressiveFMA,
+                                    FeatureArithmeticBccFusion,
+                                    FeatureCmpBccFusion,
+                                    FeatureFuseAddress,
+                                    FeatureFuseLiterals,
+                                    FeatureStorePairSuppress,
+                                    FeatureEnableSelectOptimize,
+                                    FeaturePredictableSelectIsExpensive,
+                                    FeatureLdpAlignedOnly,
+                                    FeatureStpAlignedOnly]>;
+
+
+def ProcessorFeatures {
+  list<SubtargetFeature> A53  = [HasV8_0aOps, FeatureCRC, FeatureCrypto,
+                                 FeatureFPARMv8, FeatureNEON, FeaturePerfMon];
+  list<SubtargetFeature> A55  = [HasV8_2aOps, FeatureCrypto, FeatureFPARMv8,
+                                 FeatureNEON, FeatureFullFP16, FeatureDotProd,
+                                 FeatureRCPC, FeaturePerfMon];
+  list<SubtargetFeature> A510 = [HasV9_0aOps, FeatureNEON, FeaturePerfMon,
+                                 FeatureMatMulInt8, FeatureBF16, FeatureAM,
+                                 FeatureMTE, FeatureETE, FeatureSVE2BitPerm,
+                                 FeatureFP16FML];
+  list<SubtargetFeature> A520 = [HasV9_2aOps, FeaturePerfMon, FeatureAM,
+                                 FeatureMTE, FeatureETE, FeatureSVE2BitPerm,
+                                 FeatureFP16FML];
+  list<SubtargetFeature> A520AE = [HasV9_2aOps, FeaturePerfMon, FeatureAM,
+                                 FeatureMTE, FeatureETE, FeatureSVE2BitPerm,
+                                 FeatureFP16FML];
+  list<SubtargetFeature> A65  = [HasV8_2aOps, FeatureCrypto, FeatureFPARMv8,
+                                 FeatureNEON, FeatureFullFP16, FeatureDotProd,
+                                 FeatureRCPC, FeatureSSBS, FeatureRAS,
+                                 FeaturePerfMon];
+  list<SubtargetFeature> A76  = [HasV8_2aOps, FeatureCrypto, FeatureFPARMv8,
+                                 FeatureNEON, FeatureFullFP16, FeatureDotProd,
+                                 FeatureRCPC, FeatureSSBS, FeaturePerfMon];
+  list<SubtargetFeature> A77  = [HasV8_2aOps, FeatureCrypto, FeatureFPARMv8,
+                                 FeatureNEON, FeatureFullFP16, FeatureDotProd,
+                                 FeatureRCPC, FeaturePerfMon, FeatureSSBS];
+  list<SubtargetFeature> A78  = [HasV8_2aOps, FeatureCrypto, FeatureFPARMv8,
+                                 FeatureNEON, FeatureFullFP16, FeatureDotProd,
+                                 FeatureRCPC, FeaturePerfMon, FeatureSPE,
+                                 FeatureSSBS];
+  list<SubtargetFeature> A78AE = [HasV8_2aOps, FeatureCrypto, FeatureFPARMv8,
+                                  FeatureNEON, FeatureFullFP16, FeatureDotProd,
+                                  FeatureRCPC, FeaturePerfMon, FeatureSPE,
+                                  FeatureSSBS];
+  list<SubtargetFeature> A78C = [HasV8_2aOps, FeatureCrypto, FeatureFPARMv8,
+                                 FeatureNEON, FeatureFullFP16, FeatureDotProd,
+                                 FeatureFlagM, FeaturePAuth,
+                                 FeaturePerfMon, FeatureRCPC, FeatureSPE,
+                                 FeatureSSBS];
+  list<SubtargetFeature> A710 = [HasV9_0aOps, FeatureNEON, FeaturePerfMon,
+                                 FeatureETE, FeatureMTE, FeatureFP16FML,
+                                 FeatureSVE2BitPerm, FeatureBF16, FeatureMatMulInt8];
+  list<SubtargetFeature> A715 = [HasV9_0aOps, FeatureNEON, FeatureMTE,
+                                 FeatureFP16FML, FeatureSVE, FeatureTRBE,
+                                 FeatureSVE2BitPerm, FeatureBF16, FeatureETE,
+                                 FeaturePerfMon, FeatureMatMulInt8, FeatureSPE];
+  list<SubtargetFeature> A720 = [HasV9_2aOps, FeatureMTE, FeatureFP16FML,
+                                 FeatureTRBE, FeatureSVE2BitPerm, FeatureETE,
+                                 FeaturePerfMon, FeatureSPE, FeatureSPE_EEF];
+  list<SubtargetFeature> A720AE = [HasV9_2aOps, FeatureMTE, FeatureFP16FML,
+                                 FeatureTRBE, FeatureSVE2BitPerm, FeatureETE,
+                                 FeaturePerfMon, FeatureSPE, FeatureSPE_EEF];
+  list<SubtargetFeature> R82  = [HasV8_0rOps, FeaturePerfMon, FeatureFullFP16,
+                                 FeatureFP16FML, FeatureSSBS, FeaturePredRes,
+                                 FeatureSB, FeatureRDM, FeatureDotProd,
+                                 FeatureComplxNum, FeatureJS];
+  list<SubtargetFeature> X1   = [HasV8_2aOps, FeatureCrypto, FeatureFPARMv8,
+                                 FeatureNEON, FeatureRCPC, FeaturePerfMon,
+                                 FeatureSPE, FeatureFullFP16, FeatureDotProd,
+                                 FeatureSSBS];
+  list<SubtargetFeature> X1C  = [HasV8_2aOps, FeatureCrypto, FeatureFPARMv8,
+                                 FeatureNEON, FeatureRCPC_IMMO, FeaturePerfMon,
+                                 FeatureSPE, FeatureFullFP16, FeatureDotProd,
+                                 FeaturePAuth, FeatureSSBS, FeatureFlagM,
+                                 FeatureLSE2];
+  list<SubtargetFeature> X2   = [HasV9_0aOps, FeatureNEON, FeaturePerfMon,
+                                 FeatureMatMulInt8, FeatureBF16, FeatureAM,
+                                 FeatureMTE, FeatureETE, FeatureSVE2BitPerm,
+                                 FeatureFP16FML];
+  list<SubtargetFeature> X3 =   [HasV9_0aOps, FeatureSVE, FeatureNEON,
+                                 FeaturePerfMon, FeatureETE, FeatureTRBE,
+                                 FeatureSPE, FeatureBF16, FeatureMatMulInt8,
+                                 FeatureMTE, FeatureSVE2BitPerm, FeatureFullFP16,
+                                 FeatureFP16FML];
+  list<SubtargetFeature> X4 =   [HasV9_2aOps,
+                                 FeaturePerfMon, FeatureETE, FeatureTRBE,
+                                 FeatureSPE, FeatureMTE, FeatureSVE2BitPerm,
+                                 FeatureFP16FML, FeatureSPE_EEF];
+  list<SubtargetFeature> A64FX    = [HasV8_2aOps, FeatureFPARMv8, FeatureNEON,
+                                     FeatureSHA2, FeaturePerfMon, FeatureFullFP16,
+                                     FeatureSVE, FeatureComplxNum];
+  list<SubtargetFeature> Carmel   = [HasV8_2aOps, FeatureNEON, FeatureCrypto,
+                                     FeatureFullFP16];
+  list<SubtargetFeature> AppleA7  = [HasV8_0aOps, FeatureCrypto, FeatureFPARMv8,
+                                     FeatureNEON,FeaturePerfMon, FeatureAppleA7SysReg];
+  list<SubtargetFeature> AppleA10 = [HasV8_0aOps, FeatureCrypto, FeatureFPARMv8,
+                                     FeatureNEON, FeaturePerfMon, FeatureCRC,
+                                     FeatureRDM, FeaturePAN, FeatureLOR, FeatureVH];
+  list<SubtargetFeature> AppleA11 = [HasV8_2aOps, FeatureCrypto, FeatureFPARMv8,
+                                     FeatureNEON, FeaturePerfMon, FeatureFullFP16];
+  list<SubtargetFeature> AppleA12 = [HasV8_3aOps, FeatureCrypto, FeatureFPARMv8,
+                                     FeatureNEON, FeaturePerfMon, FeatureFullFP16];
+  list<SubtargetFeature> AppleA13 = [HasV8_4aOps, FeatureCrypto, FeatureFPARMv8,
+                                     FeatureNEON, FeaturePerfMon, FeatureFullFP16,
+                                     FeatureFP16FML, FeatureSHA3];
+  list<SubtargetFeature> AppleA14 = [HasV8_4aOps, FeatureCrypto, FeatureFPARMv8,
+                                     FeatureNEON, FeaturePerfMon, FeatureFRInt3264,
+                                     FeatureSpecRestrict, FeatureSSBS, FeatureSB,
+                                     FeaturePredRes, FeatureCacheDeepPersist,
+                                     FeatureFullFP16, FeatureFP16FML, FeatureSHA3,
+                                     FeatureAltFPCmp];
+  list<SubtargetFeature> AppleA15 = [HasV8_6aOps, FeatureCrypto, FeatureFPARMv8,
+                                     FeatureNEON, FeaturePerfMon, FeatureSHA3,
+                                     FeatureFullFP16, FeatureFP16FML];
+  list<SubtargetFeature> AppleA16 = [HasV8_6aOps, FeatureCrypto, FeatureFPARMv8,
+                                     FeatureNEON, FeaturePerfMon, FeatureSHA3,
+                                     FeatureFullFP16, FeatureFP16FML,
+                                     FeatureHCX];
+  list<SubtargetFeature> AppleA17 = [HasV8_6aOps, FeatureCrypto, FeatureFPARMv8,
+                                     FeatureNEON, FeaturePerfMon, FeatureSHA3,
+                                     FeatureFullFP16, FeatureFP16FML,
+                                     FeatureHCX];
+  list<SubtargetFeature> ExynosM3 = [HasV8_0aOps, FeatureCRC, FeatureCrypto,
+                                     FeaturePerfMon];
+  list<SubtargetFeature> ExynosM4 = [HasV8_2aOps, FeatureCrypto, FeatureDotProd,
+                                     FeatureFullFP16, FeaturePerfMon];
+  list<SubtargetFeature> Falkor   = [HasV8_0aOps, FeatureCRC, FeatureCrypto,
+                                     FeatureFPARMv8, FeatureNEON, FeaturePerfMon,
+                                     FeatureRDM];
+  list<SubtargetFeature> NeoverseE1 = [HasV8_2aOps, FeatureCrypto, FeatureDotProd,
+                                       FeatureFPARMv8, FeatureFullFP16, FeatureNEON,
+                                       FeatureRCPC, FeatureSSBS, FeaturePerfMon];
+  list<SubtargetFeature> NeoverseN1 = [HasV8_2aOps, FeatureCrypto, FeatureDotProd,
+                                       FeatureFPARMv8, FeatureFullFP16, FeatureNEON,
+                                       FeatureRCPC, FeatureSPE, FeatureSSBS,
+                                       FeaturePerfMon];
+  list<SubtargetFeature> NeoverseN2 = [HasV9_0aOps, FeatureBF16, FeatureETE, FeatureFP16FML,
+                                       FeatureMatMulInt8, FeatureMTE, FeatureSVE2,
+                                       FeatureSVE2BitPerm, FeatureTRBE,
+                                       FeaturePerfMon];
+  list<SubtargetFeature> NeoverseN3 = [HasV9_2aOps, FeatureETE, FeatureFP16FML,
+                                      FeatureFullFP16, FeatureMTE, FeaturePerfMon,
+                                      FeatureRandGen, FeatureSPE, FeatureSPE_EEF,
+                                      FeatureSVE2BitPerm];
+  list<SubtargetFeature> Neoverse512TVB = [HasV8_4aOps, FeatureBF16, FeatureCacheDeepPersist,
+                                           FeatureCrypto, FeatureFPARMv8, FeatureFP16FML,
+                                           FeatureFullFP16, FeatureMatMulInt8, FeatureNEON,
+                                           FeaturePerfMon, FeatureRandGen, FeatureSPE,
+                                           FeatureSSBS, FeatureSVE];
+  list<SubtargetFeature> NeoverseV1 = [HasV8_4aOps, FeatureBF16, FeatureCacheDeepPersist,
+                                       FeatureCrypto, FeatureFPARMv8, FeatureFP16FML,
+                                       FeatureFullFP16, FeatureMatMulInt8, FeatureNEON,
+                                       FeaturePerfMon, FeatureRandGen, FeatureSPE,
+                                       FeatureSSBS, FeatureSVE];
+  list<SubtargetFeature> NeoverseV2 = [HasV9_0aOps, FeatureBF16, FeatureSPE,
+                                       FeaturePerfMon, FeatureETE, FeatureMatMulInt8,
+                                       FeatureNEON, FeatureSVE2BitPerm, FeatureFP16FML,
+                                       FeatureMTE, FeatureRandGen];
+  list<SubtargetFeature> NeoverseV3 = [HasV9_2aOps, FeatureETE, FeatureFP16FML,
+                                      FeatureFullFP16, FeatureLS64, FeatureMTE,
+                                      FeaturePerfMon, FeatureRandGen, FeatureSPE,
+                                      FeatureSPE_EEF, FeatureSVE2BitPerm, FeatureBRBE];
+  list<SubtargetFeature> NeoverseV3AE = [HasV9_2aOps, FeatureETE, FeatureFP16FML,
+                                      FeatureFullFP16, FeatureLS64, FeatureMTE,
+                                      FeaturePerfMon, FeatureRandGen, FeatureSPE,
+                                      FeatureSPE_EEF, FeatureSVE2BitPerm, FeatureBRBE];
+  list<SubtargetFeature> Saphira    = [HasV8_4aOps, FeatureCrypto, FeatureFPARMv8,
+                                       FeatureNEON, FeatureSPE, FeaturePerfMon];
+  list<SubtargetFeature> ThunderX   = [HasV8_0aOps, FeatureCRC, FeatureCrypto,
+                                       FeatureFPARMv8, FeaturePerfMon, FeatureNEON];
+  list<SubtargetFeature> ThunderX2T99  = [HasV8_1aOps, FeatureCRC, FeatureCrypto,
+                                          FeatureFPARMv8, FeatureNEON, FeatureLSE];
+  list<SubtargetFeature> ThunderX3T110 = [HasV8_3aOps, FeatureCRC, FeatureCrypto,
+                                          FeatureFPARMv8, FeatureNEON, FeatureLSE,
+                                          FeaturePAuth, FeaturePerfMon];
+  list<SubtargetFeature> TSV110 = [HasV8_2aOps, FeatureCrypto, FeatureFPARMv8,
+                                   FeatureNEON, FeaturePerfMon, FeatureSPE,
+                                   FeatureFullFP16, FeatureFP16FML, FeatureDotProd,
+                                   FeatureJS, FeatureComplxNum];
+  list<SubtargetFeature> Ampere1 = [HasV8_6aOps, FeatureNEON, FeaturePerfMon,
+                                    FeatureSSBS, FeatureRandGen, FeatureSB,
+                                    FeatureSHA2, FeatureSHA3, FeatureAES];
+  list<SubtargetFeature> Ampere1A = [HasV8_6aOps, FeatureNEON, FeaturePerfMon,
+                                     FeatureMTE, FeatureSSBS, FeatureRandGen,
+                                     FeatureSB, FeatureSM4, FeatureSHA2,
+                                     FeatureSHA3, FeatureAES];
+  list<SubtargetFeature> Ampere1B = [HasV8_7aOps, FeatureNEON, FeaturePerfMon,
+                                     FeatureMTE, FeatureSSBS, FeatureRandGen,
+                                     FeatureSB, FeatureSM4, FeatureSHA2,
+                                     FeatureSHA3, FeatureAES, FeatureCSSC,
+                                     FeatureWFxT, FeatureFullFP16];
+
+  // ETE and TRBE are future architecture extensions. We temporarily enable them
+  // by default for users targeting generic AArch64. The extensions do not
+  // affect code generated by the compiler and can be used only by explicitly
+  // mentioning the new system register names in assembly.
+  list<SubtargetFeature> Generic = [FeatureFPARMv8, FeatureNEON, FeatureETE];
+}
+
+// FeatureFuseAdrpAdd is enabled under Generic to allow linker merging
+// optimizations.
+def : ProcessorModel<"generic", CortexA510Model, ProcessorFeatures.Generic,
+                     [FeatureFuseAES, FeatureFuseAdrpAdd, FeaturePostRAScheduler,
+                      FeatureEnableSelectOptimize]>;
+def : ProcessorModel<"cortex-a35", CortexA53Model, ProcessorFeatures.A53,
+                     [TuneA35]>;
+def : ProcessorModel<"cortex-a34", CortexA53Model, ProcessorFeatures.A53,
+                     [TuneA35]>;
+def : ProcessorModel<"cortex-a53", CortexA53Model, ProcessorFeatures.A53,
+                     [TuneA53]>;
+def : ProcessorModel<"cortex-a55", CortexA55Model, ProcessorFeatures.A55,
+                     [TuneA55]>;
+def : ProcessorModel<"cortex-a510", CortexA510Model, ProcessorFeatures.A510,
+                     [TuneA510]>;
+def : ProcessorModel<"cortex-a520", CortexA510Model, ProcessorFeatures.A520,
+                     [TuneA520]>;
+def : ProcessorModel<"cortex-a520ae", CortexA510Model, ProcessorFeatures.A520AE,
+                     [TuneA520AE]>;
+def : ProcessorModel<"cortex-a57", CortexA57Model, ProcessorFeatures.A53,
+                     [TuneA57]>;
+def : ProcessorModel<"cortex-a65", CortexA53Model, ProcessorFeatures.A65,
+                     [TuneA65]>;
+def : ProcessorModel<"cortex-a65ae", CortexA53Model, ProcessorFeatures.A65,
+                     [TuneA65]>;
+def : ProcessorModel<"cortex-a72", CortexA57Model, ProcessorFeatures.A53,
+                     [TuneA72]>;
+def : ProcessorModel<"cortex-a73", CortexA57Model, ProcessorFeatures.A53,
+                     [TuneA73]>;
+def : ProcessorModel<"cortex-a75", CortexA57Model, ProcessorFeatures.A55,
+                     [TuneA75]>;
+def : ProcessorModel<"cortex-a76", CortexA57Model, ProcessorFeatures.A76,
+                     [TuneA76]>;
+def : ProcessorModel<"cortex-a76ae", CortexA57Model, ProcessorFeatures.A76,
+                     [TuneA76]>;
+def : ProcessorModel<"cortex-a77", CortexA57Model, ProcessorFeatures.A77,
+                     [TuneA77]>;
+def : ProcessorModel<"cortex-a78", CortexA57Model, ProcessorFeatures.A78,
+                     [TuneA78]>;
+def : ProcessorModel<"cortex-a78ae", CortexA57Model, ProcessorFeatures.A78AE,
+                     [TuneA78AE]>;
+def : ProcessorModel<"cortex-a78c", CortexA57Model, ProcessorFeatures.A78C,
+                     [TuneA78C]>;
+def : ProcessorModel<"cortex-a710", NeoverseN2Model, ProcessorFeatures.A710,
+                     [TuneA710]>;
+def : ProcessorModel<"cortex-a715", NeoverseN2Model, ProcessorFeatures.A715,
+                     [TuneA715]>;
+def : ProcessorModel<"cortex-a720", NeoverseN2Model, ProcessorFeatures.A720,
+                     [TuneA720]>;
+def : ProcessorModel<"cortex-a720ae", NeoverseN2Model, ProcessorFeatures.A720AE,
+                     [TuneA720AE]>;
+def : ProcessorModel<"cortex-r82", CortexA55Model, ProcessorFeatures.R82,
+                     [TuneR82]>;
+def : ProcessorModel<"cortex-x1", CortexA57Model, ProcessorFeatures.X1,
+                     [TuneX1]>;
+def : ProcessorModel<"cortex-x1c", CortexA57Model, ProcessorFeatures.X1C,
+                     [TuneX1]>;
+def : ProcessorModel<"cortex-x2", NeoverseN2Model, ProcessorFeatures.X2,
+                     [TuneX2]>;
+def : ProcessorModel<"cortex-x3", NeoverseN2Model, ProcessorFeatures.X3,
+                     [TuneX3]>;
+def : ProcessorModel<"cortex-x4", NeoverseN2Model, ProcessorFeatures.X4,
+                     [TuneX4]>;
+def : ProcessorModel<"neoverse-e1", CortexA53Model,
+                     ProcessorFeatures.NeoverseE1, [TuneNeoverseE1]>;
+def : ProcessorModel<"neoverse-n1", NeoverseN1Model,
+                     ProcessorFeatures.NeoverseN1, [TuneNeoverseN1]>;
+def : ProcessorModel<"neoverse-n2", NeoverseN2Model,
+                     ProcessorFeatures.NeoverseN2, [TuneNeoverseN2]>;
+def : ProcessorModel<"neoverse-n3", NeoverseN2Model,
+                     ProcessorFeatures.NeoverseN3, [TuneNeoverseN3]>;
+def : ProcessorModel<"neoverse-512tvb", NeoverseV1Model,
+                     ProcessorFeatures.Neoverse512TVB, [TuneNeoverse512TVB]>;
+def : ProcessorModel<"neoverse-v1", NeoverseV1Model,
+                     ProcessorFeatures.NeoverseV1, [TuneNeoverseV1]>;
+def : ProcessorModel<"neoverse-v2", NeoverseV2Model,
+                     ProcessorFeatures.NeoverseV2, [TuneNeoverseV2]>;
+def : ProcessorModel<"neoverse-v3", NeoverseV2Model,
+                     ProcessorFeatures.NeoverseV3, [TuneNeoverseV3]>;
+def : ProcessorModel<"neoverse-v3ae", NeoverseV2Model,
+                     ProcessorFeatures.NeoverseV3AE, [TuneNeoverseV3AE]>;
+def : ProcessorModel<"exynos-m3", ExynosM3Model, ProcessorFeatures.ExynosM3,
+                     [TuneExynosM3]>;
+def : ProcessorModel<"exynos-m4", ExynosM4Model, ProcessorFeatures.ExynosM4,
+                     [TuneExynosM4]>;
+def : ProcessorModel<"exynos-m5", ExynosM5Model, ProcessorFeatures.ExynosM4,
+                     [TuneExynosM4]>;
+def : ProcessorModel<"falkor", FalkorModel, ProcessorFeatures.Falkor,
+                     [TuneFalkor]>;
+def : ProcessorModel<"saphira", FalkorModel, ProcessorFeatures.Saphira,
+                     [TuneSaphira]>;
+def : ProcessorModel<"kryo", KryoModel, ProcessorFeatures.A53, [TuneKryo]>;
+
+// Cavium ThunderX/ThunderX T8X  Processors
+def : ProcessorModel<"thunderx", ThunderXT8XModel,  ProcessorFeatures.ThunderX,
+                     [TuneThunderX]>;
+def : ProcessorModel<"thunderxt88", ThunderXT8XModel,
+                     ProcessorFeatures.ThunderX, [TuneThunderXT88]>;
+def : ProcessorModel<"thunderxt81", ThunderXT8XModel,
+                     ProcessorFeatures.ThunderX, [TuneThunderXT81]>;
+def : ProcessorModel<"thunderxt83", ThunderXT8XModel,
+                     ProcessorFeatures.ThunderX, [TuneThunderXT83]>;
+// Cavium ThunderX2T9X  Processors. Formerly Broadcom Vulcan.
+def : ProcessorModel<"thunderx2t99", ThunderX2T99Model,
+                     ProcessorFeatures.ThunderX2T99, [TuneThunderX2T99]>;
+// Marvell ThunderX3T110 Processors.
+def : ProcessorModel<"thunderx3t110", ThunderX3T110Model,
+                     ProcessorFeatures.ThunderX3T110, [TuneThunderX3T110]>;
+def : ProcessorModel<"tsv110", TSV110Model, ProcessorFeatures.TSV110,
+                     [TuneTSV110]>;
+
+// Support cyclone as an alias for apple-a7 so we can still LTO old bitcode.
+def : ProcessorModel<"cyclone", CycloneModel, ProcessorFeatures.AppleA7,
+                     [TuneAppleA7]>;
+
+// iPhone and iPad CPUs
+def : ProcessorModel<"apple-a7", CycloneModel, ProcessorFeatures.AppleA7,
+                     [TuneAppleA7]>;
+def : ProcessorModel<"apple-a8", CycloneModel, ProcessorFeatures.AppleA7,
+                     [TuneAppleA7]>;
+def : ProcessorModel<"apple-a9", CycloneModel, ProcessorFeatures.AppleA7,
+                     [TuneAppleA7]>;
+def : ProcessorModel<"apple-a10", CycloneModel, ProcessorFeatures.AppleA10,
+                     [TuneAppleA10]>;
+def : ProcessorModel<"apple-a11", CycloneModel, ProcessorFeatures.AppleA11,
+                     [TuneAppleA11]>;
+def : ProcessorModel<"apple-a12", CycloneModel, ProcessorFeatures.AppleA12,
+                     [TuneAppleA12]>;
+def : ProcessorModel<"apple-a13", CycloneModel, ProcessorFeatures.AppleA13,
+                     [TuneAppleA13]>;
+def : ProcessorModel<"apple-a14", CycloneModel, ProcessorFeatures.AppleA14,
+                     [TuneAppleA14]>;
+def : ProcessorModel<"apple-a15", CycloneModel, ProcessorFeatures.AppleA15,
+                     [TuneAppleA15]>;
+def : ProcessorModel<"apple-a16", CycloneModel, ProcessorFeatures.AppleA16,
+                     [TuneAppleA16]>;
+def : ProcessorModel<"apple-a17", CycloneModel, ProcessorFeatures.AppleA17,
+                     [TuneAppleA17]>;
+// Mac CPUs
+def : ProcessorModel<"apple-m1", CycloneModel, ProcessorFeatures.AppleA14,
+                     [TuneAppleA14]>;
+def : ProcessorModel<"apple-m2", CycloneModel, ProcessorFeatures.AppleA15,
+                     [TuneAppleA15]>;
+def : ProcessorModel<"apple-m3", CycloneModel, ProcessorFeatures.AppleA16,
+                     [TuneAppleA16]>;
+
+// watch CPUs.
+def : ProcessorModel<"apple-s4", CycloneModel, ProcessorFeatures.AppleA12,
+                     [TuneAppleA12]>;
+def : ProcessorModel<"apple-s5", CycloneModel, ProcessorFeatures.AppleA12,
+                     [TuneAppleA12]>;
+
+// Alias for the latest Apple processor model supported by LLVM.
+def : ProcessorModel<"apple-latest", CycloneModel, ProcessorFeatures.AppleA16,
+                     [TuneAppleA16]>;
+
+// Fujitsu A64FX
+def : ProcessorModel<"a64fx", A64FXModel, ProcessorFeatures.A64FX,
+                     [TuneA64FX]>;
+
+// Nvidia Carmel
+def : ProcessorModel<"carmel", NoSchedModel, ProcessorFeatures.Carmel,
+                     [TuneCarmel]>;
+
+// Ampere Computing
+def : ProcessorModel<"ampere1", Ampere1Model, ProcessorFeatures.Ampere1,
+                     [TuneAmpere1]>;
+
+def : ProcessorModel<"ampere1a", Ampere1Model, ProcessorFeatures.Ampere1A,
+                     [TuneAmpere1A]>;
+
+def : ProcessorModel<"ampere1b", Ampere1BModel, ProcessorFeatures.Ampere1B,
+                     [TuneAmpere1B]>;
diff --git a/llvm/lib/Target/AArch64/AArch64RedundantCopyElimination.cpp b/llvm/lib/Target/AArch64/AArch64RedundantCopyElimination.cpp
index 1494312886a4..69fc13883f6b 100644
--- a/llvm/lib/Target/AArch64/AArch64RedundantCopyElimination.cpp
+++ b/llvm/lib/Target/AArch64/AArch64RedundantCopyElimination.cpp
@@ -265,7 +265,7 @@ bool AArch64RedundantCopyElimination::knownRegValInBlock(
     }
 
     // Bail if we see an instruction that defines NZCV that we don't handle.
-    if (PredI.definesRegister(AArch64::NZCV))
+    if (PredI.definesRegister(AArch64::NZCV, /*TRI=*/nullptr))
       return false;
 
     // Track clobbered and used registers.
diff --git a/llvm/lib/Target/AArch64/AArch64SMEInstrInfo.td b/llvm/lib/Target/AArch64/AArch64SMEInstrInfo.td
index 2db0fa253434..574178c8d524 100644
--- a/llvm/lib/Target/AArch64/AArch64SMEInstrInfo.td
+++ b/llvm/lib/Target/AArch64/AArch64SMEInstrInfo.td
@@ -792,12 +792,14 @@ defm LUTI4_S_2ZTZI : sme2p1_luti4_vector_vg2_index<"luti4">;
 defm LUTI4_S_4ZTZI : sme2p1_luti4_vector_vg4_index<"luti4">;
 }
 
-let Predicates = [HasSME2p1, HasSMEF16F16] in {
+let Predicates = [HasSMEF16F16orSMEF8F16] in {
 defm FADD_VG2_M2Z_H : sme2_multivec_accum_add_sub_vg2<"fadd", 0b0100, MatrixOp16, ZZ_h_mul_r, nxv8f16, null_frag>;
 defm FADD_VG4_M4Z_H : sme2_multivec_accum_add_sub_vg4<"fadd", 0b0100, MatrixOp16, ZZZZ_h_mul_r, nxv8f16, null_frag>;
 defm FSUB_VG2_M2Z_H : sme2_multivec_accum_add_sub_vg2<"fsub", 0b0101, MatrixOp16, ZZ_h_mul_r, nxv8f16, null_frag>;
 defm FSUB_VG4_M4Z_H : sme2_multivec_accum_add_sub_vg4<"fsub", 0b0101, MatrixOp16, ZZZZ_h_mul_r, nxv8f16, null_frag>;
+}
 
+let Predicates = [HasSMEF16F16] in {
 defm FMLA_VG2_M2ZZI_H : sme2p1_multi_vec_array_vg2_index_16b<"fmla", 0b00, 0b100, ZZ_h_mul_r, ZPR4b16>;
 defm FMLA_VG4_M4ZZI_H : sme2p1_multi_vec_array_vg4_index_16b<"fmla", 0b000, ZZZZ_h_mul_r, ZPR4b16>;
 defm FMLA_VG2_M2ZZ_H :  sme2_dot_mla_add_sub_array_vg24_single<"fmla", 0b0011100, MatrixOp16, ZZ_h, ZPR4b16>;
diff --git a/llvm/lib/Target/AArch64/AArch64SVEInstrInfo.td b/llvm/lib/Target/AArch64/AArch64SVEInstrInfo.td
index 6972acd985cb..b90ac0ff1fe0 100644
--- a/llvm/lib/Target/AArch64/AArch64SVEInstrInfo.td
+++ b/llvm/lib/Target/AArch64/AArch64SVEInstrInfo.td
@@ -2037,15 +2037,15 @@ let Predicates = [HasSVEorSME] in {
   defm FCMEQ_PPzZ0 : sve_fp_2op_p_pd<0b100, "fcmeq", SETOEQ, SETEQ, SETOEQ, SETEQ>;
   defm FCMNE_PPzZ0 : sve_fp_2op_p_pd<0b110, "fcmne", SETUNE, SETNE, SETUNE, SETNE>;
 
-  defm WHILELT_PWW : sve_int_while4_rr<0b010, "whilelt", int_aarch64_sve_whilelt>;
-  defm WHILELE_PWW : sve_int_while4_rr<0b011, "whilele", int_aarch64_sve_whilele>;
-  defm WHILELO_PWW : sve_int_while4_rr<0b110, "whilelo", int_aarch64_sve_whilelo>;
-  defm WHILELS_PWW : sve_int_while4_rr<0b111, "whilels", int_aarch64_sve_whilels>;
+  defm WHILELT_PWW : sve_int_while4_rr<0b010, "whilelt", int_aarch64_sve_whilelt, int_aarch64_sve_whilegt>;
+  defm WHILELE_PWW : sve_int_while4_rr<0b011, "whilele", int_aarch64_sve_whilele, null_frag>;
+  defm WHILELO_PWW : sve_int_while4_rr<0b110, "whilelo", int_aarch64_sve_whilelo, int_aarch64_sve_whilehi>;
+  defm WHILELS_PWW : sve_int_while4_rr<0b111, "whilels", int_aarch64_sve_whilels, null_frag>;
 
-  defm WHILELT_PXX : sve_int_while8_rr<0b010, "whilelt", int_aarch64_sve_whilelt>;
-  defm WHILELE_PXX : sve_int_while8_rr<0b011, "whilele", int_aarch64_sve_whilele>;
-  defm WHILELO_PXX : sve_int_while8_rr<0b110, "whilelo", int_aarch64_sve_whilelo>;
-  defm WHILELS_PXX : sve_int_while8_rr<0b111, "whilels", int_aarch64_sve_whilels>;
+  defm WHILELT_PXX : sve_int_while8_rr<0b010, "whilelt", int_aarch64_sve_whilelt, int_aarch64_sve_whilegt>;
+  defm WHILELE_PXX : sve_int_while8_rr<0b011, "whilele", int_aarch64_sve_whilele, null_frag>;
+  defm WHILELO_PXX : sve_int_while8_rr<0b110, "whilelo", int_aarch64_sve_whilelo, int_aarch64_sve_whilehi>;
+  defm WHILELS_PXX : sve_int_while8_rr<0b111, "whilels", int_aarch64_sve_whilels, null_frag>;
 
   def CTERMEQ_WW : sve_int_cterm<0b0, 0b0, "ctermeq", GPR32>;
   def CTERMNE_WW : sve_int_cterm<0b0, 0b1, "ctermne", GPR32>;
@@ -2517,6 +2517,23 @@ let Predicates = [HasSVEorSME] in {
     def : Pat<(vscale (sve_cntd_imm_neg i32:$imm)), (SUBXrs XZR, (CNTD_XPiI 31, $imm), 0)>;
   }
 
+  // Add NoUseScalarIncVL to avoid affecting for patterns with UseScalarIncVL
+  let Predicates = [HasSVEorSME, NoUseScalarIncVL] in {
+    def : Pat<(add GPR64:$op, (vscale (sve_cnth_imm_neg i32:$imm))),
+              (SUBXrs GPR64:$op, (CNTH_XPiI 31, $imm), 0)>;
+    def : Pat<(add GPR64:$op, (vscale (sve_cntw_imm_neg i32:$imm))),
+              (SUBXrs GPR64:$op, (CNTW_XPiI 31, $imm), 0)>;
+    def : Pat<(add GPR64:$op, (vscale (sve_cntd_imm_neg i32:$imm))),
+              (SUBXrs GPR64:$op, (CNTD_XPiI 31, $imm), 0)>;
+
+    def : Pat<(add GPR32:$op, (i32 (trunc (vscale (sve_cnth_imm_neg i32:$imm))))),
+              (SUBSWrr GPR32:$op, (EXTRACT_SUBREG (CNTH_XPiI 31, $imm), sub_32))>;
+    def : Pat<(add GPR32:$op, (i32 (trunc (vscale (sve_cntw_imm_neg i32:$imm))))),
+              (SUBSWrr GPR32:$op, (EXTRACT_SUBREG (CNTW_XPiI 31, $imm), sub_32))>;
+    def : Pat<(add GPR32:$op, (i32 (trunc (vscale (sve_cntd_imm_neg i32:$imm))))),
+              (SUBSWrr GPR32:$op, (EXTRACT_SUBREG (CNTD_XPiI 31, $imm), sub_32))>;
+  }
+
   let AddedComplexity = 5 in {
     def : Pat<(nxv8i16 (add ZPR:$op, (nxv8i16 (splat_vector (i32 (trunc (vscale (sve_cnth_imm i32:$imm)))))))),
               (INCH_ZPiI ZPR:$op, 31, $imm)>;
@@ -3743,7 +3760,7 @@ let Predicates = [HasSVE2orSME] in {
   defm BSL_ZZZZ   : sve2_int_bitwise_ternary_op<0b001, "bsl",   int_aarch64_sve_bsl, AArch64bsp>;
   defm BSL1N_ZZZZ : sve2_int_bitwise_ternary_op<0b011, "bsl1n", int_aarch64_sve_bsl1n>;
   defm BSL2N_ZZZZ : sve2_int_bitwise_ternary_op<0b101, "bsl2n", int_aarch64_sve_bsl2n>;
-  defm NBSL_ZZZZ  : sve2_int_bitwise_ternary_op<0b111, "nbsl",  int_aarch64_sve_nbsl>;
+  defm NBSL_ZZZZ  : sve2_int_bitwise_ternary_op<0b111, "nbsl",  int_aarch64_sve_nbsl, AArch64nbsl>;
 
   // SVE2 bitwise xor and rotate right by immediate
   defm XAR_ZZZI : sve2_int_rotate_right_imm<"xar", int_aarch64_sve_xar>;
@@ -3792,15 +3809,15 @@ let Predicates = [HasSVE2orSME] in {
   defm TBX_ZZZ  : sve2_int_perm_tbx<"tbx", 0b01, int_aarch64_sve_tbx>;
 
   // SVE2 integer compare scalar count and limit
-  defm WHILEGE_PWW : sve_int_while4_rr<0b000, "whilege", int_aarch64_sve_whilege>;
-  defm WHILEGT_PWW : sve_int_while4_rr<0b001, "whilegt", int_aarch64_sve_whilegt>;
-  defm WHILEHS_PWW : sve_int_while4_rr<0b100, "whilehs", int_aarch64_sve_whilehs>;
-  defm WHILEHI_PWW : sve_int_while4_rr<0b101, "whilehi", int_aarch64_sve_whilehi>;
+  defm WHILEGE_PWW : sve_int_while4_rr<0b000, "whilege", int_aarch64_sve_whilege, null_frag>;
+  defm WHILEGT_PWW : sve_int_while4_rr<0b001, "whilegt", int_aarch64_sve_whilegt, int_aarch64_sve_whilelt>;
+  defm WHILEHS_PWW : sve_int_while4_rr<0b100, "whilehs", int_aarch64_sve_whilehs, null_frag>;
+  defm WHILEHI_PWW : sve_int_while4_rr<0b101, "whilehi", int_aarch64_sve_whilehi, int_aarch64_sve_whilelo>;
 
-  defm WHILEGE_PXX : sve_int_while8_rr<0b000, "whilege", int_aarch64_sve_whilege>;
-  defm WHILEGT_PXX : sve_int_while8_rr<0b001, "whilegt", int_aarch64_sve_whilegt>;
-  defm WHILEHS_PXX : sve_int_while8_rr<0b100, "whilehs", int_aarch64_sve_whilehs>;
-  defm WHILEHI_PXX : sve_int_while8_rr<0b101, "whilehi", int_aarch64_sve_whilehi>;
+  defm WHILEGE_PXX : sve_int_while8_rr<0b000, "whilege", int_aarch64_sve_whilege, null_frag>;
+  defm WHILEGT_PXX : sve_int_while8_rr<0b001, "whilegt", int_aarch64_sve_whilegt, int_aarch64_sve_whilelt>;
+  defm WHILEHS_PXX : sve_int_while8_rr<0b100, "whilehs", int_aarch64_sve_whilehs, null_frag>;
+  defm WHILEHI_PXX : sve_int_while8_rr<0b101, "whilehi", int_aarch64_sve_whilehi, int_aarch64_sve_whilelo>;
 
   // SVE2 pointer conflict compare
   defm WHILEWR_PXX : sve2_int_while_rr<0b0, "whilewr", "int_aarch64_sve_whilewr">;
@@ -4102,7 +4119,7 @@ defm BFCLAMP_ZZZ : sve2p1_bfclamp<"bfclamp", AArch64fclamp>;
 // SME2.1 or SVE2.1 instructions
 //===----------------------------------------------------------------------===//
 let Predicates = [HasSVE2p1_or_HasSME2p1] in {
-defm FADDQV   : sve2p1_fp_reduction_q<0b000, "faddqv", int_aarch64_sve_addqv>;
+defm FADDQV   : sve2p1_fp_reduction_q<0b000, "faddqv", int_aarch64_sve_faddqv>;
 defm FMAXNMQV : sve2p1_fp_reduction_q<0b100, "fmaxnmqv", int_aarch64_sve_fmaxnmqv>;
 defm FMINNMQV : sve2p1_fp_reduction_q<0b101, "fminnmqv", int_aarch64_sve_fminnmqv>;
 defm FMAXQV   : sve2p1_fp_reduction_q<0b110, "fmaxqv", int_aarch64_sve_fmaxqv>;
diff --git a/llvm/lib/Target/AArch64/AArch64Subtarget.cpp b/llvm/lib/Target/AArch64/AArch64Subtarget.cpp
index 9eb1ecbb0315..7d34dd1c7768 100644
--- a/llvm/lib/Target/AArch64/AArch64Subtarget.cpp
+++ b/llvm/lib/Target/AArch64/AArch64Subtarget.cpp
@@ -144,7 +144,6 @@ void AArch64Subtarget::initializeProperties(bool HasMinSize) {
   case CortexA78C:
   case CortexR82:
   case CortexX1:
-  case CortexX1C:
     PrefFunctionAlignment = Align(16);
     PrefLoopAlignment = Align(32);
     MaxBytesForLoopAlignment = 16;
@@ -235,7 +234,9 @@ void AArch64Subtarget::initializeProperties(bool HasMinSize) {
     MaxBytesForLoopAlignment = 16;
     break;
   case NeoverseN2:
+  case NeoverseN3:
   case NeoverseV2:
+  case NeoverseV3:
     PrefFunctionAlignment = Align(16);
     PrefLoopAlignment = Align(32);
     MaxBytesForLoopAlignment = 16;
diff --git a/llvm/lib/Target/AArch64/AArch64Subtarget.h b/llvm/lib/Target/AArch64/AArch64Subtarget.h
index c5ebd4c6cc61..3f3eefc4f680 100644
--- a/llvm/lib/Target/AArch64/AArch64Subtarget.h
+++ b/llvm/lib/Target/AArch64/AArch64Subtarget.h
@@ -39,61 +39,9 @@ class AArch64Subtarget final : public AArch64GenSubtargetInfo {
 public:
   enum ARMProcFamilyEnum : uint8_t {
     Others,
-    A64FX,
-    Ampere1,
-    Ampere1A,
-    Ampere1B,
-    AppleA7,
-    AppleA10,
-    AppleA11,
-    AppleA12,
-    AppleA13,
-    AppleA14,
-    AppleA15,
-    AppleA16,
-    AppleA17,
-    Carmel,
-    CortexA35,
-    CortexA53,
-    CortexA55,
-    CortexA510,
-    CortexA520,
-    CortexA57,
-    CortexA65,
-    CortexA72,
-    CortexA73,
-    CortexA75,
-    CortexA76,
-    CortexA77,
-    CortexA78,
-    CortexA78AE,
-    CortexA78C,
-    CortexA710,
-    CortexA715,
-    CortexA720,
-    CortexR82,
-    CortexX1,
-    CortexX1C,
-    CortexX2,
-    CortexX3,
-    CortexX4,
-    ExynosM3,
-    Falkor,
-    Kryo,
-    NeoverseE1,
-    NeoverseN1,
-    NeoverseN2,
-    Neoverse512TVB,
-    NeoverseV1,
-    NeoverseV2,
-    Saphira,
-    ThunderX2T99,
-    ThunderX,
-    ThunderXT81,
-    ThunderXT83,
-    ThunderXT88,
-    ThunderX3T110,
-    TSV110
+#define ARM_PROCESSOR_FAMILY(ENUM) ENUM,
+#include "llvm/TargetParser/AArch64TargetParserDef.inc"
+#undef ARM_PROCESSOR_FAMILY
   };
 
 protected:
diff --git a/llvm/lib/Target/AArch64/AArch64TargetTransformInfo.cpp b/llvm/lib/Target/AArch64/AArch64TargetTransformInfo.cpp
index 700242b88346..af3a94a0faec 100644
--- a/llvm/lib/Target/AArch64/AArch64TargetTransformInfo.cpp
+++ b/llvm/lib/Target/AArch64/AArch64TargetTransformInfo.cpp
@@ -790,6 +790,27 @@ AArch64TTIImpl::getIntrinsicInstrCost(const IntrinsicCostAttributes &ICA,
       break;
     return TyL.first + ExtraCost;
   }
+  case Intrinsic::get_active_lane_mask: {
+    auto *RetTy = dyn_cast<FixedVectorType>(ICA.getReturnType());
+    if (RetTy) {
+      EVT RetVT = getTLI()->getValueType(DL, RetTy);
+      EVT OpVT = getTLI()->getValueType(DL, ICA.getArgTypes()[0]);
+      if (!getTLI()->shouldExpandGetActiveLaneMask(RetVT, OpVT) &&
+          !getTLI()->isTypeLegal(RetVT)) {
+        // We don't have enough context at this point to determine if the mask
+        // is going to be kept live after the block, which will force the vXi1
+        // type to be expanded to legal vectors of integers, e.g. v4i1->v4i32.
+        // For now, we just assume the vectorizer created this intrinsic and
+        // the result will be the input for a PHI. In this case the cost will
+        // be extremely high for fixed-width vectors.
+        // NOTE: getScalarizationOverhead returns a cost that's far too
+        // pessimistic for the actual generated codegen. In reality there are
+        // two instructions generated per lane.
+        return RetTy->getNumElements() * 2;
+      }
+    }
+    break;
+  }
   default:
     break;
   }
diff --git a/llvm/lib/Target/AArch64/AArch64TargetTransformInfo.h b/llvm/lib/Target/AArch64/AArch64TargetTransformInfo.h
index dba384481f6a..678c132e6a80 100644
--- a/llvm/lib/Target/AArch64/AArch64TargetTransformInfo.h
+++ b/llvm/lib/Target/AArch64/AArch64TargetTransformInfo.h
@@ -203,7 +203,7 @@ public:
       unsigned Opcode, Type *Ty, TTI::TargetCostKind CostKind,
       TTI::OperandValueInfo Op1Info = {TTI::OK_AnyValue, TTI::OP_None},
       TTI::OperandValueInfo Op2Info = {TTI::OK_AnyValue, TTI::OP_None},
-      ArrayRef<const Value *> Args = ArrayRef<const Value *>(),
+      ArrayRef<const Value *> Args = std::nullopt,
       const Instruction *CxtI = nullptr);
 
   InstructionCost getAddressComputationCost(Type *Ty, ScalarEvolution *SE,
diff --git a/llvm/lib/Target/AArch64/GISel/AArch64LegalizerInfo.cpp b/llvm/lib/Target/AArch64/GISel/AArch64LegalizerInfo.cpp
index 85dd0f2eb192..243891249668 100644
--- a/llvm/lib/Target/AArch64/GISel/AArch64LegalizerInfo.cpp
+++ b/llvm/lib/Target/AArch64/GISel/AArch64LegalizerInfo.cpp
@@ -93,18 +93,12 @@ AArch64LegalizerInfo::AArch64LegalizerInfo(const AArch64Subtarget &ST)
       .legalFor(PackedVectorAllTypeList)
       .widenScalarToNextPow2(0)
       .clampScalar(0, s8, s64)
-      .fewerElementsIf(
-          [=](const LegalityQuery &Query) {
-            return Query.Types[0].isVector() &&
-                   (Query.Types[0].getElementType() != s64 ||
-                    Query.Types[0].getNumElements() != 2);
-          },
-          [=](const LegalityQuery &Query) {
-            LLT EltTy = Query.Types[0].getElementType();
-            if (EltTy == s64)
-              return std::make_pair(0, LLT::fixed_vector(2, 64));
-            return std::make_pair(0, EltTy);
-          });
+      .moreElementsToNextPow2(0)
+      .widenVectorEltsToVectorMinSize(0, 64)
+      .clampNumElements(0, v8s8, v16s8)
+      .clampNumElements(0, v4s16, v8s16)
+      .clampNumElements(0, v2s32, v4s32)
+      .clampNumElements(0, v2s64, v2s64);
 
   getActionDefinitionsBuilder(G_PHI)
       .legalFor({p0, s16, s32, s64})
@@ -746,6 +740,9 @@ AArch64LegalizerInfo::AArch64LegalizerInfo(const AArch64Subtarget &ST)
       .legalForCartesianProduct({s32, v2s16, v4s8})
       .legalForCartesianProduct({s64, v8s8, v4s16, v2s32})
       .legalForCartesianProduct({s128, v16s8, v8s16, v4s32, v2s64, v2p0})
+      .lowerIf([=](const LegalityQuery &Query) {
+        return Query.Types[0].isVector() != Query.Types[1].isVector();
+      })
       .moreElementsToNextPow2(0)
       .clampNumElements(0, v8s8, v16s8)
       .clampNumElements(0, v4s16, v8s16)
@@ -1573,7 +1570,7 @@ bool AArch64LegalizerInfo::legalizeIntrinsic(LegalizerHelper &Helper,
     MI.eraseFromParent();
     return true;
   }
-  case Intrinsic::experimental_vector_reverse:
+  case Intrinsic::vector_reverse:
     // TODO: Add support for vector_reverse
     return false;
   }
diff --git a/llvm/lib/Target/AArch64/GISel/AArch64PostSelectOptimize.cpp b/llvm/lib/Target/AArch64/GISel/AArch64PostSelectOptimize.cpp
index 94584e20f5ab..11866f2dd186 100644
--- a/llvm/lib/Target/AArch64/GISel/AArch64PostSelectOptimize.cpp
+++ b/llvm/lib/Target/AArch64/GISel/AArch64PostSelectOptimize.cpp
@@ -199,10 +199,11 @@ bool AArch64PostSelectOptimize::optimizeNZCVDefs(MachineBasicBlock &MBB) {
 
   for (auto &II : instructionsWithoutDebug(MBB.rbegin(), MBB.rend())) {
     bool NZCVDead = LRU.available(AArch64::NZCV);
-    if (NZCVDead && II.definesRegister(AArch64::NZCV)) {
+    if (NZCVDead && II.definesRegister(AArch64::NZCV, /*TRI=*/nullptr)) {
       // The instruction defines NZCV, but NZCV is dead.
       unsigned NewOpc = getNonFlagSettingVariant(II.getOpcode());
-      int DeadNZCVIdx = II.findRegisterDefOperandIdx(AArch64::NZCV);
+      int DeadNZCVIdx =
+          II.findRegisterDefOperandIdx(AArch64::NZCV, /*TRI=*/nullptr);
       if (DeadNZCVIdx != -1) {
         if (NewOpc) {
           // If there is an equivalent non-flag-setting op, we convert.
diff --git a/llvm/lib/Target/AArch64/MCTargetDesc/AArch64ELFObjectWriter.cpp b/llvm/lib/Target/AArch64/MCTargetDesc/AArch64ELFObjectWriter.cpp
index 6e074b6a63c4..b4c5cde5fd88 100644
--- a/llvm/lib/Target/AArch64/MCTargetDesc/AArch64ELFObjectWriter.cpp
+++ b/llvm/lib/Target/AArch64/MCTargetDesc/AArch64ELFObjectWriter.cpp
@@ -146,8 +146,8 @@ unsigned AArch64ELFObjectWriter::getRelocType(MCContext &Ctx,
                         "ILP32 8 byte PC relative data "
                         "relocation not supported (LP64 eqv: PREL64)");
         return ELF::R_AARCH64_NONE;
-      } else
-        return ELF::R_AARCH64_PREL64;
+      }
+      return ELF::R_AARCH64_PREL64;
     case AArch64::fixup_aarch64_pcrel_adr_imm21:
       if (SymLoc != AArch64MCExpr::VK_ABS)
         Ctx.reportError(Fixup.getLoc(),
@@ -162,9 +162,8 @@ unsigned AArch64ELFObjectWriter::getRelocType(MCContext &Ctx,
                           "invalid fixup for 32-bit pcrel ADRP instruction "
                           "VK_ABS VK_NC");
           return ELF::R_AARCH64_NONE;
-        } else {
-          return ELF::R_AARCH64_ADR_PREL_PG_HI21_NC;
         }
+        return ELF::R_AARCH64_ADR_PREL_PG_HI21_NC;
       }
       if (SymLoc == AArch64MCExpr::VK_GOT && !IsNC)
         return R_CLS(ADR_GOT_PAGE);
@@ -211,18 +210,18 @@ unsigned AArch64ELFObjectWriter::getRelocType(MCContext &Ctx,
               Target.getAccessVariant() == MCSymbolRefExpr::VK_GOTPCREL)
                  ? ELF::R_AARCH64_GOTPCREL32
                  : R_CLS(ABS32);
-    case FK_Data_8:
+    case FK_Data_8: {
+      bool IsAuth = (RefKind == AArch64MCExpr::VK_AUTH ||
+                     RefKind == AArch64MCExpr::VK_AUTHADDR);
       if (IsILP32) {
         Ctx.reportError(Fixup.getLoc(),
-                        "ILP32 8 byte absolute data "
-                        "relocation not supported (LP64 eqv: ABS64)");
+                        Twine("ILP32 8 byte absolute data "
+                              "relocation not supported (LP64 eqv: ") +
+                            (IsAuth ? "AUTH_ABS64" : "ABS64") + Twine(')'));
         return ELF::R_AARCH64_NONE;
-      } else {
-        if (RefKind == AArch64MCExpr::VK_AUTH ||
-            RefKind == AArch64MCExpr::VK_AUTHADDR)
-          return ELF::R_AARCH64_AUTH_ABS64;
-        return ELF::R_AARCH64_ABS64;
       }
+      return (IsAuth ? ELF::R_AARCH64_AUTH_ABS64 : ELF::R_AARCH64_ABS64);
+    }
     case AArch64::fixup_aarch64_add_imm12:
       if (RefKind == AArch64MCExpr::VK_DTPREL_HI12)
         return R_CLS(TLSLD_ADD_DTPREL_HI12);
@@ -286,14 +285,12 @@ unsigned AArch64ELFObjectWriter::getRelocType(MCContext &Ctx,
       if (SymLoc == AArch64MCExpr::VK_TPREL && IsNC)
         return R_CLS(TLSLE_LDST32_TPREL_LO12_NC);
       if (SymLoc == AArch64MCExpr::VK_GOT && IsNC) {
-        if (IsILP32) {
+        if (IsILP32)
           return ELF::R_AARCH64_P32_LD32_GOT_LO12_NC;
-        } else {
-          Ctx.reportError(Fixup.getLoc(),
-                          "LP64 4 byte unchecked GOT load/store relocation "
-                          "not supported (ILP32 eqv: LD32_GOT_LO12_NC");
-          return ELF::R_AARCH64_NONE;
-        }
+        Ctx.reportError(Fixup.getLoc(),
+                        "LP64 4 byte unchecked GOT load/store relocation "
+                        "not supported (ILP32 eqv: LD32_GOT_LO12_NC");
+        return ELF::R_AARCH64_NONE;
       }
       if (SymLoc == AArch64MCExpr::VK_GOT && !IsNC) {
         if (IsILP32) {
@@ -309,25 +306,20 @@ unsigned AArch64ELFObjectWriter::getRelocType(MCContext &Ctx,
         return ELF::R_AARCH64_NONE;
       }
       if (SymLoc == AArch64MCExpr::VK_GOTTPREL && IsNC) {
-        if (IsILP32) {
+        if (IsILP32)
           return ELF::R_AARCH64_P32_TLSIE_LD32_GOTTPREL_LO12_NC;
-        } else {
-          Ctx.reportError(Fixup.getLoc(),
-                          "LP64 32-bit load/store "
-                          "relocation not supported (ILP32 eqv: "
-                          "TLSIE_LD32_GOTTPREL_LO12_NC)");
-          return ELF::R_AARCH64_NONE;
-        }
+        Ctx.reportError(Fixup.getLoc(), "LP64 32-bit load/store "
+                                        "relocation not supported (ILP32 eqv: "
+                                        "TLSIE_LD32_GOTTPREL_LO12_NC)");
+        return ELF::R_AARCH64_NONE;
       }
       if (SymLoc == AArch64MCExpr::VK_TLSDESC && !IsNC) {
-        if (IsILP32) {
+        if (IsILP32)
           return ELF::R_AARCH64_P32_TLSDESC_LD32_LO12;
-        } else {
-          Ctx.reportError(Fixup.getLoc(),
-                          "LP64 4 byte TLSDESC load/store relocation "
-                          "not supported (ILP32 eqv: TLSDESC_LD64_LO12)");
-          return ELF::R_AARCH64_NONE;
-        }
+        Ctx.reportError(Fixup.getLoc(),
+                        "LP64 4 byte TLSDESC load/store relocation "
+                        "not supported (ILP32 eqv: TLSDESC_LD64_LO12)");
+        return ELF::R_AARCH64_NONE;
       }
 
       Ctx.reportError(Fixup.getLoc(),
@@ -344,12 +336,11 @@ unsigned AArch64ELFObjectWriter::getRelocType(MCContext &Ctx,
           if (AddressLoc == AArch64MCExpr::VK_LO15)
             return ELF::R_AARCH64_LD64_GOTPAGE_LO15;
           return ELF::R_AARCH64_LD64_GOT_LO12_NC;
-        } else {
-          Ctx.reportError(Fixup.getLoc(), "ILP32 64-bit load/store "
-                                          "relocation not supported (LP64 eqv: "
-                                          "LD64_GOT_LO12_NC)");
-          return ELF::R_AARCH64_NONE;
         }
+        Ctx.reportError(Fixup.getLoc(), "ILP32 64-bit load/store "
+                                        "relocation not supported (LP64 eqv: "
+                                        "LD64_GOT_LO12_NC)");
+        return ELF::R_AARCH64_NONE;
       }
       if (SymLoc == AArch64MCExpr::VK_DTPREL && !IsNC)
         return R_CLS(TLSLD_LDST64_DTPREL_LO12);
@@ -360,24 +351,20 @@ unsigned AArch64ELFObjectWriter::getRelocType(MCContext &Ctx,
       if (SymLoc == AArch64MCExpr::VK_TPREL && IsNC)
         return R_CLS(TLSLE_LDST64_TPREL_LO12_NC);
       if (SymLoc == AArch64MCExpr::VK_GOTTPREL && IsNC) {
-        if (!IsILP32) {
+        if (!IsILP32)
           return ELF::R_AARCH64_TLSIE_LD64_GOTTPREL_LO12_NC;
-        } else {
-          Ctx.reportError(Fixup.getLoc(), "ILP32 64-bit load/store "
-                                          "relocation not supported (LP64 eqv: "
-                                          "TLSIE_LD64_GOTTPREL_LO12_NC)");
-          return ELF::R_AARCH64_NONE;
-        }
+        Ctx.reportError(Fixup.getLoc(), "ILP32 64-bit load/store "
+                                        "relocation not supported (LP64 eqv: "
+                                        "TLSIE_LD64_GOTTPREL_LO12_NC)");
+        return ELF::R_AARCH64_NONE;
       }
       if (SymLoc == AArch64MCExpr::VK_TLSDESC) {
-        if (!IsILP32) {
+        if (!IsILP32)
           return ELF::R_AARCH64_TLSDESC_LD64_LO12;
-        } else {
-          Ctx.reportError(Fixup.getLoc(), "ILP32 64-bit load/store "
-                                          "relocation not supported (LP64 eqv: "
-                                          "TLSDESC_LD64_LO12)");
-          return ELF::R_AARCH64_NONE;
-        }
+        Ctx.reportError(Fixup.getLoc(), "ILP32 64-bit load/store "
+                                        "relocation not supported (LP64 eqv: "
+                                        "TLSDESC_LD64_LO12)");
+        return ELF::R_AARCH64_NONE;
       }
       Ctx.reportError(Fixup.getLoc(),
                       "invalid fixup for 64-bit load/store instruction");
diff --git a/llvm/lib/Target/AArch64/MCTargetDesc/AArch64ELFStreamer.cpp b/llvm/lib/Target/AArch64/MCTargetDesc/AArch64ELFStreamer.cpp
index ad21f2673a64..f5bea3336cbf 100644
--- a/llvm/lib/Target/AArch64/MCTargetDesc/AArch64ELFStreamer.cpp
+++ b/llvm/lib/Target/AArch64/MCTargetDesc/AArch64ELFStreamer.cpp
@@ -306,13 +306,12 @@ llvm::createAArch64AsmTargetStreamer(MCStreamer &S, formatted_raw_ostream &OS,
   return new AArch64TargetAsmStreamer(S, OS);
 }
 
-MCELFStreamer *llvm::createAArch64ELFStreamer(
-    MCContext &Context, std::unique_ptr<MCAsmBackend> TAB,
-    std::unique_ptr<MCObjectWriter> OW, std::unique_ptr<MCCodeEmitter> Emitter,
-    bool RelaxAll) {
+MCELFStreamer *
+llvm::createAArch64ELFStreamer(MCContext &Context,
+                               std::unique_ptr<MCAsmBackend> TAB,
+                               std::unique_ptr<MCObjectWriter> OW,
+                               std::unique_ptr<MCCodeEmitter> Emitter) {
   AArch64ELFStreamer *S = new AArch64ELFStreamer(
       Context, std::move(TAB), std::move(OW), std::move(Emitter));
-  if (RelaxAll)
-    S->getAssembler().setRelaxAll(true);
   return S;
 }
diff --git a/llvm/lib/Target/AArch64/MCTargetDesc/AArch64ELFStreamer.h b/llvm/lib/Target/AArch64/MCTargetDesc/AArch64ELFStreamer.h
index 25c609ee1496..e6df79ba19d4 100644
--- a/llvm/lib/Target/AArch64/MCTargetDesc/AArch64ELFStreamer.h
+++ b/llvm/lib/Target/AArch64/MCTargetDesc/AArch64ELFStreamer.h
@@ -20,8 +20,7 @@ namespace llvm {
 MCELFStreamer *createAArch64ELFStreamer(MCContext &Context,
                                         std::unique_ptr<MCAsmBackend> TAB,
                                         std::unique_ptr<MCObjectWriter> OW,
-                                        std::unique_ptr<MCCodeEmitter> Emitter,
-                                        bool RelaxAll);
+                                        std::unique_ptr<MCCodeEmitter> Emitter);
 }
 
 #endif
diff --git a/llvm/lib/Target/AArch64/MCTargetDesc/AArch64MCTargetDesc.cpp b/llvm/lib/Target/AArch64/MCTargetDesc/AArch64MCTargetDesc.cpp
index 043f0a03b797..0dd4a78f962d 100644
--- a/llvm/lib/Target/AArch64/MCTargetDesc/AArch64MCTargetDesc.cpp
+++ b/llvm/lib/Target/AArch64/MCTargetDesc/AArch64MCTargetDesc.cpp
@@ -378,30 +378,28 @@ static MCInstPrinter *createAArch64MCInstPrinter(const Triple &T,
 static MCStreamer *createELFStreamer(const Triple &T, MCContext &Ctx,
                                      std::unique_ptr<MCAsmBackend> &&TAB,
                                      std::unique_ptr<MCObjectWriter> &&OW,
-                                     std::unique_ptr<MCCodeEmitter> &&Emitter,
-                                     bool RelaxAll) {
+                                     std::unique_ptr<MCCodeEmitter> &&Emitter) {
   return createAArch64ELFStreamer(Ctx, std::move(TAB), std::move(OW),
-                                  std::move(Emitter), RelaxAll);
+                                  std::move(Emitter));
 }
 
 static MCStreamer *createMachOStreamer(MCContext &Ctx,
                                        std::unique_ptr<MCAsmBackend> &&TAB,
                                        std::unique_ptr<MCObjectWriter> &&OW,
                                        std::unique_ptr<MCCodeEmitter> &&Emitter,
-                                       bool RelaxAll,
                                        bool DWARFMustBeAtTheEnd) {
   return createMachOStreamer(Ctx, std::move(TAB), std::move(OW),
-                             std::move(Emitter), RelaxAll, DWARFMustBeAtTheEnd,
+                             std::move(Emitter), DWARFMustBeAtTheEnd,
                              /*LabelSections*/ true);
 }
 
 static MCStreamer *
 createWinCOFFStreamer(MCContext &Ctx, std::unique_ptr<MCAsmBackend> &&TAB,
                       std::unique_ptr<MCObjectWriter> &&OW,
-                      std::unique_ptr<MCCodeEmitter> &&Emitter, bool RelaxAll,
+                      std::unique_ptr<MCCodeEmitter> &&Emitter,
                       bool IncrementalLinkerCompatible) {
   return createAArch64WinCOFFStreamer(Ctx, std::move(TAB), std::move(OW),
-                                      std::move(Emitter), RelaxAll,
+                                      std::move(Emitter),
                                       IncrementalLinkerCompatible);
 }
 
diff --git a/llvm/lib/Target/AArch64/MCTargetDesc/AArch64WinCOFFStreamer.cpp b/llvm/lib/Target/AArch64/MCTargetDesc/AArch64WinCOFFStreamer.cpp
index 438ac6cc4788..c25cc2e99adc 100644
--- a/llvm/lib/Target/AArch64/MCTargetDesc/AArch64WinCOFFStreamer.cpp
+++ b/llvm/lib/Target/AArch64/MCTargetDesc/AArch64WinCOFFStreamer.cpp
@@ -294,7 +294,7 @@ void AArch64TargetWinCOFFStreamer::emitARM64WinCFISaveAnyRegQPX(unsigned Reg,
 MCWinCOFFStreamer *llvm::createAArch64WinCOFFStreamer(
     MCContext &Context, std::unique_ptr<MCAsmBackend> MAB,
     std::unique_ptr<MCObjectWriter> OW, std::unique_ptr<MCCodeEmitter> Emitter,
-    bool RelaxAll, bool IncrementalLinkerCompatible) {
+    bool IncrementalLinkerCompatible) {
   auto *S = new AArch64WinCOFFStreamer(Context, std::move(MAB),
                                        std::move(Emitter), std::move(OW));
   S->getAssembler().setIncrementalLinkerCompatible(IncrementalLinkerCompatible);
diff --git a/llvm/lib/Target/AArch64/MCTargetDesc/AArch64WinCOFFStreamer.h b/llvm/lib/Target/AArch64/MCTargetDesc/AArch64WinCOFFStreamer.h
index 8c0656652eed..a13b1a451be5 100644
--- a/llvm/lib/Target/AArch64/MCTargetDesc/AArch64WinCOFFStreamer.h
+++ b/llvm/lib/Target/AArch64/MCTargetDesc/AArch64WinCOFFStreamer.h
@@ -21,7 +21,7 @@ namespace llvm {
 MCWinCOFFStreamer *createAArch64WinCOFFStreamer(
     MCContext &Context, std::unique_ptr<MCAsmBackend> TAB,
     std::unique_ptr<MCObjectWriter> OW, std::unique_ptr<MCCodeEmitter> Emitter,
-    bool RelaxAll, bool IncrementalLinkerCompatible);
+    bool IncrementalLinkerCompatible);
 } // end llvm namespace
 
 #endif
diff --git a/llvm/lib/Target/AArch64/SVEInstrFormats.td b/llvm/lib/Target/AArch64/SVEInstrFormats.td
index 3317cf84aa6f..485b38a29a0f 100644
--- a/llvm/lib/Target/AArch64/SVEInstrFormats.td
+++ b/llvm/lib/Target/AArch64/SVEInstrFormats.td
@@ -5387,7 +5387,8 @@ class sve_int_while_rr<bits<2> sz8_64, bits<4> opc, string asm,
   let isWhile = 1;
 }
 
-multiclass sve_int_while4_rr<bits<3> opc, string asm, SDPatternOperator op> {
+multiclass sve_int_while4_rr<bits<3> opc, string asm, SDPatternOperator op,
+                             SDPatternOperator rev_op> {
   def _B : sve_int_while_rr<0b00, { 0, opc }, asm, GPR32, PPR8>;
   def _H : sve_int_while_rr<0b01, { 0, opc }, asm, GPR32, PPR16>;
   def _S : sve_int_while_rr<0b10, { 0, opc }, asm, GPR32, PPR32>;
@@ -5397,9 +5398,19 @@ multiclass sve_int_while4_rr<bits<3> opc, string asm, SDPatternOperator op> {
   def : SVE_2_Op_Pat<nxv8i1,  op, i32, i32, !cast<Instruction>(NAME # _H)>;
   def : SVE_2_Op_Pat<nxv4i1,  op, i32, i32, !cast<Instruction>(NAME # _S)>;
   def : SVE_2_Op_Pat<nxv2i1,  op, i32, i32, !cast<Instruction>(NAME # _D)>;
+
+  def : Pat<(nxv16i1 (vector_reverse (rev_op i32:$op2, i32:$op1))),
+            (!cast<Instruction>(NAME # "_B") $op1, $op2)>;
+  def : Pat<(nxv8i1 (vector_reverse (rev_op i32:$op2, i32:$op1))),
+            (!cast<Instruction>(NAME # "_H") $op1, $op2)>;
+  def : Pat<(nxv4i1 (vector_reverse (rev_op i32:$op2, i32:$op1))),
+            (!cast<Instruction>(NAME # "_S") $op1, $op2)>;
+  def : Pat<(nxv2i1 (vector_reverse (rev_op i32:$op2, i32:$op1))),
+            (!cast<Instruction>(NAME # "_D") $op1, $op2)>;
 }
 
-multiclass sve_int_while8_rr<bits<3> opc, string asm, SDPatternOperator op> {
+multiclass sve_int_while8_rr<bits<3> opc, string asm, SDPatternOperator op,
+                             SDPatternOperator rev_op> {
   def _B : sve_int_while_rr<0b00, { 1, opc }, asm, GPR64, PPR8>;
   def _H : sve_int_while_rr<0b01, { 1, opc }, asm, GPR64, PPR16>;
   def _S : sve_int_while_rr<0b10, { 1, opc }, asm, GPR64, PPR32>;
@@ -5409,6 +5420,15 @@ multiclass sve_int_while8_rr<bits<3> opc, string asm, SDPatternOperator op> {
   def : SVE_2_Op_Pat<nxv8i1,  op, i64, i64, !cast<Instruction>(NAME # _H)>;
   def : SVE_2_Op_Pat<nxv4i1,  op, i64, i64, !cast<Instruction>(NAME # _S)>;
   def : SVE_2_Op_Pat<nxv2i1,  op, i64, i64, !cast<Instruction>(NAME # _D)>;
+
+  def : Pat<(nxv16i1 (vector_reverse (rev_op i64:$op2, i64:$op1))),
+            (!cast<Instruction>(NAME # "_B") $op1, $op2)>;
+  def : Pat<(nxv8i1 (vector_reverse (rev_op i64:$op2, i64:$op1))),
+            (!cast<Instruction>(NAME # "_H") $op1, $op2)>;
+  def : Pat<(nxv4i1 (vector_reverse (rev_op i64:$op2, i64:$op1))),
+            (!cast<Instruction>(NAME # "_S") $op1, $op2)>;
+  def : Pat<(nxv2i1 (vector_reverse (rev_op i64:$op2, i64:$op1))),
+            (!cast<Instruction>(NAME # "_D") $op1, $op2)>;
 }
 
 class sve2_int_while_rr<bits<2> sz8_64, bits<1> rw, string asm,
diff --git a/llvm/lib/Target/AMDGPU/AMDGPU.td b/llvm/lib/Target/AMDGPU/AMDGPU.td
index 5c2c6d4b13c6..8abe9920c02c 100644
--- a/llvm/lib/Target/AMDGPU/AMDGPU.td
+++ b/llvm/lib/Target/AMDGPU/AMDGPU.td
@@ -307,6 +307,12 @@ def FeatureMSAALoadDstSelBug : SubtargetFeature<"msaa-load-dst-sel-bug",
   "MSAA loads not honoring dst_sel bug"
 >;
 
+def FeaturePrivEnabledTrap2NopBug : SubtargetFeature<"priv-enabled-trap2-nop-bug",
+  "HasPrivEnabledTrap2NopBug",
+  "true",
+  "Hardware that runs with PRIV=1 interpreting 's_trap 2' as a nop bug"
+>;
+
 class SubtargetFeatureLDSBankCount <int Value> : SubtargetFeature <
   "ldsbankcount"#Value,
   "LDSBankCount",
@@ -1487,13 +1493,16 @@ def FeatureISAVersion11_Generic: FeatureSet<
     [FeatureMSAALoadDstSelBug,
      FeatureVALUTransUseHazard,
      FeatureUserSGPRInit16Bug,
+     FeatureMADIntraFwdBug,
+     FeaturePrivEnabledTrap2NopBug,
      FeatureRequiresCOV6])>;
 
 def FeatureISAVersion11_0_Common : FeatureSet<
   !listconcat(FeatureISAVersion11_Common.Features,
     [FeatureMSAALoadDstSelBug,
      FeatureVALUTransUseHazard,
-     FeatureMADIntraFwdBug])>;
+     FeatureMADIntraFwdBug,
+     FeaturePrivEnabledTrap2NopBug])>;
 
 def FeatureISAVersion11_0_0 : FeatureSet<
   !listconcat(FeatureISAVersion11_0_Common.Features,
diff --git a/llvm/lib/Target/AMDGPU/AMDGPUISelLowering.cpp b/llvm/lib/Target/AMDGPU/AMDGPUISelLowering.cpp
index db69d50799e7..7f4a2437f62e 100644
--- a/llvm/lib/Target/AMDGPU/AMDGPUISelLowering.cpp
+++ b/llvm/lib/Target/AMDGPU/AMDGPUISelLowering.cpp
@@ -148,6 +148,19 @@ AMDGPUTargetLowering::AMDGPUTargetLowering(const TargetMachine &TM,
   setOperationAction(ISD::LOAD, MVT::i128, Promote);
   AddPromotedToType(ISD::LOAD, MVT::i128, MVT::v4i32);
 
+  // TODO: Would be better to consume as directly legal
+  setOperationAction(ISD::ATOMIC_LOAD, MVT::f32, Promote);
+  AddPromotedToType(ISD::ATOMIC_LOAD, MVT::f32, MVT::i32);
+
+  setOperationAction(ISD::ATOMIC_LOAD, MVT::f64, Promote);
+  AddPromotedToType(ISD::ATOMIC_LOAD, MVT::f64, MVT::i64);
+
+  setOperationAction(ISD::ATOMIC_LOAD, MVT::f16, Promote);
+  AddPromotedToType(ISD::ATOMIC_LOAD, MVT::f16, MVT::i16);
+
+  setOperationAction(ISD::ATOMIC_LOAD, MVT::bf16, Promote);
+  AddPromotedToType(ISD::ATOMIC_LOAD, MVT::bf16, MVT::i16);
+
   // There are no 64-bit extloads. These should be done as a 32-bit extload and
   // an extension to 64-bit.
   for (MVT VT : MVT::integer_valuetypes())
@@ -300,11 +313,16 @@ AMDGPUTargetLowering::AMDGPUTargetLowering(const TargetMachine &TM,
 
   setTruncStoreAction(MVT::f32, MVT::bf16, Expand);
   setTruncStoreAction(MVT::f32, MVT::f16, Expand);
+  setTruncStoreAction(MVT::v2f32, MVT::v2bf16, Expand);
   setTruncStoreAction(MVT::v2f32, MVT::v2f16, Expand);
   setTruncStoreAction(MVT::v3f32, MVT::v3f16, Expand);
+  setTruncStoreAction(MVT::v4f32, MVT::v4bf16, Expand);
   setTruncStoreAction(MVT::v4f32, MVT::v4f16, Expand);
+  setTruncStoreAction(MVT::v8f32, MVT::v8bf16, Expand);
   setTruncStoreAction(MVT::v8f32, MVT::v8f16, Expand);
+  setTruncStoreAction(MVT::v16f32, MVT::v16bf16, Expand);
   setTruncStoreAction(MVT::v16f32, MVT::v16f16, Expand);
+  setTruncStoreAction(MVT::v32f32, MVT::v32bf16, Expand);
   setTruncStoreAction(MVT::v32f32, MVT::v32f16, Expand);
 
   setTruncStoreAction(MVT::f64, MVT::bf16, Expand);
@@ -5377,6 +5395,7 @@ const char* AMDGPUTargetLowering::getTargetNodeName(unsigned Opcode) const {
   NODE_NAME_CASE(RETURN_TO_EPILOG)
   NODE_NAME_CASE(ENDPGM)
   NODE_NAME_CASE(ENDPGM_TRAP)
+  NODE_NAME_CASE(SIMULATED_TRAP)
   NODE_NAME_CASE(DWORDADDR)
   NODE_NAME_CASE(FRACT)
   NODE_NAME_CASE(SETCC)
diff --git a/llvm/lib/Target/AMDGPU/AMDGPUISelLowering.h b/llvm/lib/Target/AMDGPU/AMDGPUISelLowering.h
index f10a357125e5..269c414521db 100644
--- a/llvm/lib/Target/AMDGPU/AMDGPUISelLowering.h
+++ b/llvm/lib/Target/AMDGPU/AMDGPUISelLowering.h
@@ -230,6 +230,12 @@ public:
   bool isCheapToSpeculateCtlz(Type *Ty) const override;
 
   bool isSDNodeAlwaysUniform(const SDNode *N) const override;
+
+  // FIXME: This hook should not exist
+  AtomicExpansionKind shouldCastAtomicLoadInIR(LoadInst *LI) const override {
+    return AtomicExpansionKind::None;
+  }
+
   static CCAssignFn *CCAssignFnForCall(CallingConv::ID CC, bool IsVarArg);
   static CCAssignFn *CCAssignFnForReturn(CallingConv::ID CC, bool IsVarArg);
 
@@ -407,6 +413,9 @@ enum NodeType : unsigned {
   // s_endpgm, but we may want to insert it in the middle of the block.
   ENDPGM_TRAP,
 
+  // "s_trap 2" equivalent on hardware that does not support it.
+  SIMULATED_TRAP,
+
   // Return to a shader part's epilog code.
   RETURN_TO_EPILOG,
 
diff --git a/llvm/lib/Target/AMDGPU/AMDGPUInstrInfo.td b/llvm/lib/Target/AMDGPU/AMDGPUInstrInfo.td
index 82f58ea38fd0..702f6e67c552 100644
--- a/llvm/lib/Target/AMDGPU/AMDGPUInstrInfo.td
+++ b/llvm/lib/Target/AMDGPU/AMDGPUInstrInfo.td
@@ -377,6 +377,8 @@ def AMDGPUendpgm : SDNode<"AMDGPUISD::ENDPGM", SDTNone,
     [SDNPHasChain, SDNPOptInGlue]>;
 def AMDGPUendpgm_trap : SDNode<"AMDGPUISD::ENDPGM_TRAP", SDTNone,
     [SDNPHasChain]>;
+def AMDGPUsimulated_trap : SDNode<"AMDGPUISD::SIMULATED_TRAP", SDTNone,
+    [SDNPHasChain]>;
 
 def AMDGPUreturn_to_epilog : SDNode<"AMDGPUISD::RETURN_TO_EPILOG", SDTNone,
     [SDNPHasChain, SDNPOptInGlue, SDNPVariadic]>;
diff --git a/llvm/lib/Target/AMDGPU/AMDGPULegalizerInfo.cpp b/llvm/lib/Target/AMDGPU/AMDGPULegalizerInfo.cpp
index 780dfaae11ef..6cd93abff1a4 100644
--- a/llvm/lib/Target/AMDGPU/AMDGPULegalizerInfo.cpp
+++ b/llvm/lib/Target/AMDGPU/AMDGPULegalizerInfo.cpp
@@ -4248,7 +4248,8 @@ bool AMDGPULegalizerInfo::loadInputValue(
       AMDGPU::isEntryFunctionCC(CC) && !MFI->hasWorkGroupIDZ() ? ~0u : 0xFFFFu);
   const ArgDescriptor WorkGroupIDZ =
       ArgDescriptor::createRegister(AMDGPU::TTMP7, 0xFFFF0000u);
-  if (ST.hasArchitectedSGPRs() && AMDGPU::isCompute(CC)) {
+  if (ST.hasArchitectedSGPRs() &&
+      (AMDGPU::isCompute(CC) || CC == CallingConv::AMDGPU_Gfx)) {
     switch (ArgType) {
     case AMDGPUFunctionArgInfo::WORKGROUP_ID_X:
       Arg = &WorkGroupIDX;
@@ -6724,8 +6725,18 @@ bool AMDGPULegalizerInfo::legalizeTrapHsaQueuePtr(
   return true;
 }
 
-bool AMDGPULegalizerInfo::legalizeTrapHsa(
-    MachineInstr &MI, MachineRegisterInfo &MRI, MachineIRBuilder &B) const {
+bool AMDGPULegalizerInfo::legalizeTrapHsa(MachineInstr &MI,
+                                          MachineRegisterInfo &MRI,
+                                          MachineIRBuilder &B) const {
+  // We need to simulate the 's_trap 2' instruction on targets that run in
+  // PRIV=1 (where it is treated as a nop).
+  if (ST.hasPrivEnabledTrap2NopBug()) {
+    ST.getInstrInfo()->insertSimulatedTrap(MRI, B.getMBB(), MI,
+                                           MI.getDebugLoc());
+    MI.eraseFromParent();
+    return true;
+  }
+
   B.buildInstr(AMDGPU::S_TRAP)
       .addImm(static_cast<unsigned>(GCNSubtarget::TrapID::LLVMAMDHSATrap));
   MI.eraseFromParent();
diff --git a/llvm/lib/Target/AMDGPU/AMDGPURegisterBankInfo.cpp b/llvm/lib/Target/AMDGPU/AMDGPURegisterBankInfo.cpp
index aa4ec785bf02..56345d14a331 100644
--- a/llvm/lib/Target/AMDGPU/AMDGPURegisterBankInfo.cpp
+++ b/llvm/lib/Target/AMDGPU/AMDGPURegisterBankInfo.cpp
@@ -2261,7 +2261,7 @@ void AMDGPURegisterBankInfo::applyMappingImpl(
   case AMDGPU::G_FCMP:
     if (!Subtarget.hasSALUFloatInsts())
       break;
-    LLVM_FALLTHROUGH;
+    [[fallthrough]];
   case AMDGPU::G_ICMP:
   case AMDGPU::G_UADDO:
   case AMDGPU::G_USUBO:
diff --git a/llvm/lib/Target/AMDGPU/AMDGPUTargetTransformInfo.h b/llvm/lib/Target/AMDGPU/AMDGPUTargetTransformInfo.h
index 0dab3a982779..b423df17302c 100644
--- a/llvm/lib/Target/AMDGPU/AMDGPUTargetTransformInfo.h
+++ b/llvm/lib/Target/AMDGPU/AMDGPUTargetTransformInfo.h
@@ -155,7 +155,7 @@ public:
       unsigned Opcode, Type *Ty, TTI::TargetCostKind CostKind,
       TTI::OperandValueInfo Op1Info = {TTI::OK_AnyValue, TTI::OP_None},
       TTI::OperandValueInfo Op2Info = {TTI::OK_AnyValue, TTI::OP_None},
-      ArrayRef<const Value *> Args = ArrayRef<const Value *>(),
+      ArrayRef<const Value *> Args = std::nullopt,
       const Instruction *CxtI = nullptr);
 
   InstructionCost getCFInstrCost(unsigned Opcode, TTI::TargetCostKind CostKind,
diff --git a/llvm/lib/Target/AMDGPU/AsmParser/AMDGPUAsmParser.cpp b/llvm/lib/Target/AMDGPU/AsmParser/AMDGPUAsmParser.cpp
index 386672352114..510f5bbf2555 100644
--- a/llvm/lib/Target/AMDGPU/AsmParser/AMDGPUAsmParser.cpp
+++ b/llvm/lib/Target/AMDGPU/AsmParser/AMDGPUAsmParser.cpp
@@ -172,6 +172,7 @@ public:
     ImmTyWaitEXP,
     ImmTyWaitVAVDst,
     ImmTyWaitVMVSrc,
+    ImmTyByteSel,
   };
 
   // Immediate operand kind.
@@ -384,8 +385,8 @@ public:
   bool isIdxen() const { return isImmTy(ImmTyIdxen); }
   bool isAddr64() const { return isImmTy(ImmTyAddr64); }
   bool isOffset() const { return isImmTy(ImmTyOffset); }
-  bool isOffset0() const { return isImmTy(ImmTyOffset0) && isUInt<8>(getImm()); }
-  bool isOffset1() const { return isImmTy(ImmTyOffset1) && isUInt<8>(getImm()); }
+  bool isOffset0() const { return isImmTy(ImmTyOffset0); }
+  bool isOffset1() const { return isImmTy(ImmTyOffset1); }
   bool isSMEMOffsetMod() const { return isImmTy(ImmTySMEMOffsetMod); }
   bool isFlatOffset() const { return isImmTy(ImmTyOffset) || isImmTy(ImmTyInstOffset); }
   bool isGDS() const { return isImmTy(ImmTyGDS); }
@@ -410,6 +411,7 @@ public:
   bool isOpSelHi() const { return isImmTy(ImmTyOpSelHi); }
   bool isNegLo() const { return isImmTy(ImmTyNegLo); }
   bool isNegHi() const { return isImmTy(ImmTyNegHi); }
+  bool isByteSel() const { return isImmTy(ImmTyByteSel); }
 
   bool isRegOrImm() const {
     return isReg() || isImm();
@@ -1139,6 +1141,7 @@ public:
     case ImmTyWaitEXP: OS << "WaitEXP"; break;
     case ImmTyWaitVAVDst: OS << "WaitVAVDst"; break;
     case ImmTyWaitVMVSrc: OS << "WaitVMVSrc"; break;
+    case ImmTyByteSel: OS << "ByteSel" ; break;
     }
     // clang-format on
   }
@@ -8644,6 +8647,13 @@ void AMDGPUAsmParser::cvtVOP3(MCInst &Inst, const OperandVector &Operands,
     }
   }
 
+  if (AMDGPU::hasNamedOperand(Opc, AMDGPU::OpName::byte_sel)) {
+    assert(AMDGPU::hasNamedOperand(Opc, AMDGPU::OpName::vdst_in));
+    Inst.addOperand(Inst.getOperand(0));
+    addOptionalImmOperand(Inst, Operands, OptionalIdx,
+                          AMDGPUOperand::ImmTyByteSel);
+  }
+
   if (AMDGPU::hasNamedOperand(Opc, AMDGPU::OpName::clamp))
     addOptionalImmOperand(Inst, Operands, OptionalIdx,
                           AMDGPUOperand::ImmTyClampSI);
@@ -8680,8 +8690,8 @@ void AMDGPUAsmParser::cvtVOP3P(MCInst &Inst, const OperandVector &Operands,
 
   if (Opc == AMDGPU::V_CVT_SR_BF8_F32_vi ||
       Opc == AMDGPU::V_CVT_SR_FP8_F32_vi ||
-      Opc == AMDGPU::V_CVT_SR_BF8_F32_e64_gfx12 ||
-      Opc == AMDGPU::V_CVT_SR_FP8_F32_e64_gfx12) {
+      Opc == AMDGPU::V_CVT_SR_BF8_F32_gfx12_e64_gfx12 ||
+      Opc == AMDGPU::V_CVT_SR_FP8_F32_gfx12_e64_gfx12) {
     Inst.addOperand(MCOperand::createImm(0)); // Placeholder for src2_mods
     Inst.addOperand(Inst.getOperand(0));
   }
@@ -8692,7 +8702,11 @@ void AMDGPUAsmParser::cvtVOP3P(MCInst &Inst, const OperandVector &Operands,
       !(Opc == AMDGPU::V_CVT_PK_BF8_F32_e64_dpp_gfx12 ||
         Opc == AMDGPU::V_CVT_PK_FP8_F32_e64_dpp_gfx12 ||
         Opc == AMDGPU::V_CVT_PK_BF8_F32_e64_dpp8_gfx12 ||
-        Opc == AMDGPU::V_CVT_PK_FP8_F32_e64_dpp8_gfx12)) {
+        Opc == AMDGPU::V_CVT_PK_FP8_F32_e64_dpp8_gfx12 ||
+        Opc == AMDGPU::V_CVT_SR_FP8_F32_gfx12_e64_dpp_gfx12 ||
+        Opc == AMDGPU::V_CVT_SR_FP8_F32_gfx12_e64_dpp8_gfx12 ||
+        Opc == AMDGPU::V_CVT_SR_BF8_F32_gfx12_e64_dpp_gfx12 ||
+        Opc == AMDGPU::V_CVT_SR_BF8_F32_gfx12_e64_dpp8_gfx12)) {
     assert(!IsPacked);
     Inst.addOperand(Inst.getOperand(0));
   }
@@ -8923,11 +8937,11 @@ bool AMDGPUOperand::isBLGP() const {
 }
 
 bool AMDGPUOperand::isCBSZ() const {
-  return isImm() && getImmTy() == ImmTyCBSZ && isUInt<3>(getImm());
+  return isImm() && getImmTy() == ImmTyCBSZ;
 }
 
 bool AMDGPUOperand::isABID() const {
-  return isImm() && getImmTy() == ImmTyABID && isUInt<4>(getImm());
+  return isImm() && getImmTy() == ImmTyABID;
 }
 
 bool AMDGPUOperand::isS16Imm() const {
@@ -9207,10 +9221,11 @@ void AMDGPUAsmParser::cvtVOP3DPP(MCInst &Inst, const OperandVector &Operands,
       Inst.addOperand(Inst.getOperand(0));
     }
 
-    bool IsVOP3CvtSrDpp = Opc == AMDGPU::V_CVT_SR_BF8_F32_e64_dpp8_gfx12 ||
-                          Opc == AMDGPU::V_CVT_SR_FP8_F32_e64_dpp8_gfx12 ||
-                          Opc == AMDGPU::V_CVT_SR_BF8_F32_e64_dpp_gfx12 ||
-                          Opc == AMDGPU::V_CVT_SR_FP8_F32_e64_dpp_gfx12;
+    bool IsVOP3CvtSrDpp =
+        Opc == AMDGPU::V_CVT_SR_BF8_F32_gfx12_e64_dpp8_gfx12 ||
+        Opc == AMDGPU::V_CVT_SR_FP8_F32_gfx12_e64_dpp8_gfx12 ||
+        Opc == AMDGPU::V_CVT_SR_BF8_F32_gfx12_e64_dpp_gfx12 ||
+        Opc == AMDGPU::V_CVT_SR_FP8_F32_gfx12_e64_dpp_gfx12;
     if (IsVOP3CvtSrDpp) {
       if (Src2ModIdx == static_cast<int>(Inst.getNumOperands())) {
         Inst.addOperand(MCOperand::createImm(0));
@@ -9243,6 +9258,11 @@ void AMDGPUAsmParser::cvtVOP3DPP(MCInst &Inst, const OperandVector &Operands,
       llvm_unreachable("unhandled operand type");
     }
   }
+
+  if (AMDGPU::hasNamedOperand(Opc, AMDGPU::OpName::byte_sel))
+    addOptionalImmOperand(Inst, Operands, OptionalIdx,
+                          AMDGPUOperand::ImmTyByteSel);
+
   if (AMDGPU::hasNamedOperand(Opc, AMDGPU::OpName::clamp))
     addOptionalImmOperand(Inst, Operands, OptionalIdx, AMDGPUOperand::ImmTyClampSI);
 
@@ -9648,25 +9668,17 @@ bool AMDGPUOperand::isEndpgm() const { return isImmTy(ImmTyEndpgm); }
 // LDSDIR
 //===----------------------------------------------------------------------===//
 
-bool AMDGPUOperand::isWaitVDST() const {
-  return isImmTy(ImmTyWaitVDST) && isUInt<4>(getImm());
-}
+bool AMDGPUOperand::isWaitVDST() const { return isImmTy(ImmTyWaitVDST); }
 
-bool AMDGPUOperand::isWaitVAVDst() const {
-  return isImmTy(ImmTyWaitVAVDst) && isUInt<4>(getImm());
-}
+bool AMDGPUOperand::isWaitVAVDst() const { return isImmTy(ImmTyWaitVAVDst); }
 
-bool AMDGPUOperand::isWaitVMVSrc() const {
-  return isImmTy(ImmTyWaitVMVSrc) && isUInt<1>(getImm());
-}
+bool AMDGPUOperand::isWaitVMVSrc() const { return isImmTy(ImmTyWaitVMVSrc); }
 
 //===----------------------------------------------------------------------===//
 // VINTERP
 //===----------------------------------------------------------------------===//
 
-bool AMDGPUOperand::isWaitEXP() const {
-  return isImmTy(ImmTyWaitEXP) && isUInt<3>(getImm());
-}
+bool AMDGPUOperand::isWaitEXP() const { return isImmTy(ImmTyWaitEXP); }
 
 //===----------------------------------------------------------------------===//
 // Split Barrier
diff --git a/llvm/lib/Target/AMDGPU/Disassembler/AMDGPUDisassembler.cpp b/llvm/lib/Target/AMDGPU/Disassembler/AMDGPUDisassembler.cpp
index dc1bf92771b4..8fd36b84a00c 100644
--- a/llvm/lib/Target/AMDGPU/Disassembler/AMDGPUDisassembler.cpp
+++ b/llvm/lib/Target/AMDGPU/Disassembler/AMDGPUDisassembler.cpp
@@ -869,10 +869,6 @@ void AMDGPUDisassembler::convertDPP8Inst(MCInst &MI) const {
   if (VDstInIdx != -1)
     insertNamedMCOperand(MI, MI.getOperand(0), AMDGPU::OpName::vdst_in);
 
-  if (MI.getOpcode() == AMDGPU::V_CVT_SR_BF8_F32_e64_dpp8_gfx12 ||
-      MI.getOpcode() == AMDGPU::V_CVT_SR_FP8_F32_e64_dpp8_gfx12)
-    insertNamedMCOperand(MI, MI.getOperand(0), AMDGPU::OpName::src2);
-
   unsigned DescNumOps = MCII->get(Opc).getNumOperands();
   if (MI.getNumOperands() < DescNumOps &&
       AMDGPU::hasNamedOperand(Opc, AMDGPU::OpName::op_sel)) {
@@ -902,10 +898,6 @@ void AMDGPUDisassembler::convertVOP3DPPInst(MCInst &MI) const {
   if (VDstInIdx != -1)
     insertNamedMCOperand(MI, MI.getOperand(0), AMDGPU::OpName::vdst_in);
 
-  if (MI.getOpcode() == AMDGPU::V_CVT_SR_BF8_F32_e64_dpp_gfx12 ||
-      MI.getOpcode() == AMDGPU::V_CVT_SR_FP8_F32_e64_dpp_gfx12)
-    insertNamedMCOperand(MI, MI.getOperand(0), AMDGPU::OpName::src2);
-
   unsigned Opc = MI.getOpcode();
   unsigned DescNumOps = MCII->get(Opc).getNumOperands();
   if (MI.getNumOperands() < DescNumOps &&
diff --git a/llvm/lib/Target/AMDGPU/GCNDPPCombine.cpp b/llvm/lib/Target/AMDGPU/GCNDPPCombine.cpp
index 5090b0a07da4..91733c2933b4 100644
--- a/llvm/lib/Target/AMDGPU/GCNDPPCombine.cpp
+++ b/llvm/lib/Target/AMDGPU/GCNDPPCombine.cpp
@@ -409,6 +409,11 @@ MachineInstr *GCNDPPCombine::createDPPInst(MachineInstr &OrigMI,
       if (NegHiOpr && AMDGPU::hasNamedOperand(DPPOp, AMDGPU::OpName::neg_hi)) {
         DPPInst.addImm(NegHiOpr->getImm());
       }
+      auto *ByteSelOpr = TII->getNamedOperand(OrigMI, AMDGPU::OpName::byte_sel);
+      if (ByteSelOpr &&
+          AMDGPU::hasNamedOperand(DPPOp, AMDGPU::OpName::byte_sel)) {
+        DPPInst.addImm(ByteSelOpr->getImm());
+      }
     }
     DPPInst.add(*TII->getNamedOperand(MovMI, AMDGPU::OpName::dpp_ctrl));
     DPPInst.add(*TII->getNamedOperand(MovMI, AMDGPU::OpName::row_mask));
diff --git a/llvm/lib/Target/AMDGPU/GCNHazardRecognizer.cpp b/llvm/lib/Target/AMDGPU/GCNHazardRecognizer.cpp
index 75766b11ca82..f0c111eaf060 100644
--- a/llvm/lib/Target/AMDGPU/GCNHazardRecognizer.cpp
+++ b/llvm/lib/Target/AMDGPU/GCNHazardRecognizer.cpp
@@ -241,7 +241,7 @@ GCNHazardRecognizer::getHazardType(SUnit *SU, int Stalls) {
        (ST.hasReadM0SendMsgHazard() && isSendMsgTraceDataOrGDS(TII, *MI)) ||
        (ST.hasReadM0LdsDmaHazard() && isLdsDma(*MI)) ||
        (ST.hasReadM0LdsDirectHazard() &&
-        MI->readsRegister(AMDGPU::LDS_DIRECT))) &&
+        MI->readsRegister(AMDGPU::LDS_DIRECT, /*TRI=*/nullptr))) &&
       checkReadM0Hazards(MI) > 0)
     return HazardType;
 
@@ -381,7 +381,8 @@ unsigned GCNHazardRecognizer::PreEmitNoopsCommon(MachineInstr *MI) {
         MI->getOpcode() == AMDGPU::DS_READ_ADDTID_B32)) ||
       (ST.hasReadM0SendMsgHazard() && isSendMsgTraceDataOrGDS(TII, *MI)) ||
       (ST.hasReadM0LdsDmaHazard() && isLdsDma(*MI)) ||
-      (ST.hasReadM0LdsDirectHazard() && MI->readsRegister(AMDGPU::LDS_DIRECT)))
+      (ST.hasReadM0LdsDirectHazard() &&
+       MI->readsRegister(AMDGPU::LDS_DIRECT, /*TRI=*/nullptr)))
     return std::max(WaitStates, checkReadM0Hazards(MI));
 
   if (SIInstrInfo::isMAI(*MI))
@@ -1161,7 +1162,7 @@ bool GCNHazardRecognizer::fixVMEMtoScalarWriteHazards(MachineInstr *MI) {
 
     for (const MachineOperand &Def : MI->defs()) {
       const MachineOperand *Op =
-          I.findRegisterUseOperand(Def.getReg(), false, TRI);
+          I.findRegisterUseOperand(Def.getReg(), TRI, false);
       if (!Op)
         continue;
       return true;
diff --git a/llvm/lib/Target/AMDGPU/GCNSubtarget.h b/llvm/lib/Target/AMDGPU/GCNSubtarget.h
index 2ca5ae306b11..be337e0b2192 100644
--- a/llvm/lib/Target/AMDGPU/GCNSubtarget.h
+++ b/llvm/lib/Target/AMDGPU/GCNSubtarget.h
@@ -224,6 +224,7 @@ protected:
   bool HasImageStoreD16Bug = false;
   bool HasImageGather4D16Bug = false;
   bool HasMSAALoadDstSelBug = false;
+  bool HasPrivEnabledTrap2NopBug = false;
   bool Has1_5xVGPRs = false;
   bool HasMADIntraFwdBug = false;
   bool HasVOPDInsts = false;
@@ -1032,6 +1033,8 @@ public:
 
   bool hasMSAALoadDstSelBug() const { return HasMSAALoadDstSelBug; }
 
+  bool hasPrivEnabledTrap2NopBug() const { return HasPrivEnabledTrap2NopBug; }
+
   bool hasNSAEncoding() const { return HasNSAEncoding; }
 
   bool hasNonNSAEncoding() const { return getGeneration() < GFX12; }
diff --git a/llvm/lib/Target/AMDGPU/MCTargetDesc/AMDGPUELFStreamer.cpp b/llvm/lib/Target/AMDGPU/MCTargetDesc/AMDGPUELFStreamer.cpp
index 1ce7012040da..4e9a33227a5d 100644
--- a/llvm/lib/Target/AMDGPU/MCTargetDesc/AMDGPUELFStreamer.cpp
+++ b/llvm/lib/Target/AMDGPU/MCTargetDesc/AMDGPUELFStreamer.cpp
@@ -28,10 +28,11 @@ public:
 
 }
 
-MCELFStreamer *llvm::createAMDGPUELFStreamer(
-    const Triple &T, MCContext &Context, std::unique_ptr<MCAsmBackend> MAB,
-    std::unique_ptr<MCObjectWriter> OW, std::unique_ptr<MCCodeEmitter> Emitter,
-    bool RelaxAll) {
+MCELFStreamer *
+llvm::createAMDGPUELFStreamer(const Triple &T, MCContext &Context,
+                              std::unique_ptr<MCAsmBackend> MAB,
+                              std::unique_ptr<MCObjectWriter> OW,
+                              std::unique_ptr<MCCodeEmitter> Emitter) {
   return new AMDGPUELFStreamer(T, Context, std::move(MAB), std::move(OW),
                                std::move(Emitter));
 }
diff --git a/llvm/lib/Target/AMDGPU/MCTargetDesc/AMDGPUELFStreamer.h b/llvm/lib/Target/AMDGPU/MCTargetDesc/AMDGPUELFStreamer.h
index e09e2dca1b47..f9ece5f22b0f 100644
--- a/llvm/lib/Target/AMDGPU/MCTargetDesc/AMDGPUELFStreamer.h
+++ b/llvm/lib/Target/AMDGPU/MCTargetDesc/AMDGPUELFStreamer.h
@@ -26,8 +26,7 @@ class Triple;
 MCELFStreamer *createAMDGPUELFStreamer(const Triple &T, MCContext &Context,
                                        std::unique_ptr<MCAsmBackend> MAB,
                                        std::unique_ptr<MCObjectWriter> OW,
-                                       std::unique_ptr<MCCodeEmitter> Emitter,
-                                       bool RelaxAll);
+                                       std::unique_ptr<MCCodeEmitter> Emitter);
 } // namespace llvm.
 
 #endif
diff --git a/llvm/lib/Target/AMDGPU/MCTargetDesc/AMDGPUInstPrinter.cpp b/llvm/lib/Target/AMDGPU/MCTargetDesc/AMDGPUInstPrinter.cpp
index b6a95906bc45..883b6c4407fe 100644
--- a/llvm/lib/Target/AMDGPU/MCTargetDesc/AMDGPUInstPrinter.cpp
+++ b/llvm/lib/Target/AMDGPU/MCTargetDesc/AMDGPUInstPrinter.cpp
@@ -1806,4 +1806,14 @@ void AMDGPUInstPrinter::printEndpgm(const MCInst *MI, unsigned OpNo,
   O << ' ' << formatDec(Imm);
 }
 
+void AMDGPUInstPrinter::printByteSel(const MCInst *MI, unsigned OpNo,
+                                     const MCSubtargetInfo &STI,
+                                     raw_ostream &O) {
+  uint8_t Imm = MI->getOperand(OpNo).getImm();
+  if (!Imm)
+    return;
+
+  O << " byte_sel:" << formatDec(Imm);
+}
+
 #include "AMDGPUGenAsmWriter.inc"
diff --git a/llvm/lib/Target/AMDGPU/MCTargetDesc/AMDGPUInstPrinter.h b/llvm/lib/Target/AMDGPU/MCTargetDesc/AMDGPUInstPrinter.h
index c801eaf1111e..d6d7fd34b68c 100644
--- a/llvm/lib/Target/AMDGPU/MCTargetDesc/AMDGPUInstPrinter.h
+++ b/llvm/lib/Target/AMDGPU/MCTargetDesc/AMDGPUInstPrinter.h
@@ -186,6 +186,8 @@ private:
                     const MCSubtargetInfo &STI, raw_ostream &O);
   void printExpTgt(const MCInst *MI, unsigned OpNo,
                    const MCSubtargetInfo &STI, raw_ostream &O);
+  void printByteSel(const MCInst *MI, unsigned OpNo, const MCSubtargetInfo &STI,
+                    raw_ostream &O);
 
 public:
   static void printIfSet(const MCInst *MI, unsigned OpNo, raw_ostream &O,
diff --git a/llvm/lib/Target/AMDGPU/MCTargetDesc/AMDGPUMCTargetDesc.cpp b/llvm/lib/Target/AMDGPU/MCTargetDesc/AMDGPUMCTargetDesc.cpp
index 4700a984770b..30dd384051b9 100644
--- a/llvm/lib/Target/AMDGPU/MCTargetDesc/AMDGPUMCTargetDesc.cpp
+++ b/llvm/lib/Target/AMDGPU/MCTargetDesc/AMDGPUMCTargetDesc.cpp
@@ -112,10 +112,9 @@ static MCTargetStreamer *createAMDGPUNullTargetStreamer(MCStreamer &S) {
 static MCStreamer *createMCStreamer(const Triple &T, MCContext &Context,
                                     std::unique_ptr<MCAsmBackend> &&MAB,
                                     std::unique_ptr<MCObjectWriter> &&OW,
-                                    std::unique_ptr<MCCodeEmitter> &&Emitter,
-                                    bool RelaxAll) {
+                                    std::unique_ptr<MCCodeEmitter> &&Emitter) {
   return createAMDGPUELFStreamer(T, Context, std::move(MAB), std::move(OW),
-                                 std::move(Emitter), RelaxAll);
+                                 std::move(Emitter));
 }
 
 namespace {
diff --git a/llvm/lib/Target/AMDGPU/R600InstrInfo.cpp b/llvm/lib/Target/AMDGPU/R600InstrInfo.cpp
index 7f874b245b8f..98e735935789 100644
--- a/llvm/lib/Target/AMDGPU/R600InstrInfo.cpp
+++ b/llvm/lib/Target/AMDGPU/R600InstrInfo.cpp
@@ -207,11 +207,11 @@ bool R600InstrInfo::mustBeLastInClause(unsigned Opcode) const {
 }
 
 bool R600InstrInfo::usesAddressRegister(MachineInstr &MI) const {
-  return MI.findRegisterUseOperandIdx(R600::AR_X, false, &RI) != -1;
+  return MI.findRegisterUseOperandIdx(R600::AR_X, &RI, false) != -1;
 }
 
 bool R600InstrInfo::definesAddressRegister(MachineInstr &MI) const {
-  return MI.findRegisterDefOperandIdx(R600::AR_X, false, false, &RI) != -1;
+  return MI.findRegisterDefOperandIdx(R600::AR_X, &RI, false, false) != -1;
 }
 
 bool R600InstrInfo::readsLDSSrcReg(const MachineInstr &MI) const {
diff --git a/llvm/lib/Target/AMDGPU/SIFixSGPRCopies.cpp b/llvm/lib/Target/AMDGPU/SIFixSGPRCopies.cpp
index 8b21c22b4497..a00ca625fc73 100644
--- a/llvm/lib/Target/AMDGPU/SIFixSGPRCopies.cpp
+++ b/llvm/lib/Target/AMDGPU/SIFixSGPRCopies.cpp
@@ -947,8 +947,9 @@ void SIFixSGPRCopies::analyzeVGPRToSGPRCopy(MachineInstr* MI) {
         (Inst->isCopy() && Inst->getOperand(0).getReg() == AMDGPU::SCC)) {
       auto I = Inst->getIterator();
       auto E = Inst->getParent()->end();
-      while (++I != E && !I->findRegisterDefOperand(AMDGPU::SCC)) {
-        if (I->readsRegister(AMDGPU::SCC))
+      while (++I != E &&
+             !I->findRegisterDefOperand(AMDGPU::SCC, /*TRI=*/nullptr)) {
+        if (I->readsRegister(AMDGPU::SCC, /*TRI=*/nullptr))
           Users.push_back(&*I);
       }
     } else if (Inst->getNumExplicitDefs() != 0) {
diff --git a/llvm/lib/Target/AMDGPU/SIISelLowering.cpp b/llvm/lib/Target/AMDGPU/SIISelLowering.cpp
index 17b6e0cb9c3b..cb4efdc7cf65 100644
--- a/llvm/lib/Target/AMDGPU/SIISelLowering.cpp
+++ b/llvm/lib/Target/AMDGPU/SIISelLowering.cpp
@@ -2124,7 +2124,8 @@ SDValue SITargetLowering::getPreloadedValue(SelectionDAG &DAG,
       AMDGPU::isEntryFunctionCC(CC) && !MFI.hasWorkGroupIDZ() ? ~0u : 0xFFFFu);
   const ArgDescriptor WorkGroupIDZ =
       ArgDescriptor::createRegister(AMDGPU::TTMP7, 0xFFFF0000u);
-  if (Subtarget->hasArchitectedSGPRs() && AMDGPU::isCompute(CC)) {
+  if (Subtarget->hasArchitectedSGPRs() &&
+      (AMDGPU::isCompute(CC) || CC == CallingConv::AMDGPU_Gfx)) {
     switch (PVID) {
     case AMDGPUFunctionArgInfo::WORKGROUP_ID_X:
       Reg = &WorkGroupIDX;
@@ -2798,7 +2799,9 @@ SDValue SITargetLowering::LowerFormalArguments(
     (void)UserSGPRInfo;
     if (!Subtarget->enableFlatScratch())
       assert(!UserSGPRInfo.hasFlatScratchInit());
-    if (CallConv != CallingConv::AMDGPU_CS || !Subtarget->hasArchitectedSGPRs())
+    if ((CallConv != CallingConv::AMDGPU_CS &&
+         CallConv != CallingConv::AMDGPU_Gfx) ||
+        !Subtarget->hasArchitectedSGPRs())
       assert(!Info->hasWorkGroupIDX() && !Info->hasWorkGroupIDY() &&
              !Info->hasWorkGroupIDZ());
   }
@@ -5402,6 +5405,14 @@ MachineBasicBlock *SITargetLowering::EmitInstrWithCustomInserter(
     MI.eraseFromParent();
     return SplitBB;
   }
+  case AMDGPU::SIMULATED_TRAP: {
+    assert(Subtarget->hasPrivEnabledTrap2NopBug());
+    MachineRegisterInfo &MRI = BB->getParent()->getRegInfo();
+    MachineBasicBlock *SplitBB =
+        TII->insertSimulatedTrap(MRI, *BB, MI, MI.getDebugLoc());
+    MI.eraseFromParent();
+    return SplitBB;
+  }
   default:
     if (TII->isImage(MI) || TII->isMUBUF(MI)) {
       if (!MI.mayStore())
@@ -6620,6 +6631,11 @@ SDValue SITargetLowering::lowerTrapHsa(
   SDLoc SL(Op);
   SDValue Chain = Op.getOperand(0);
 
+  // We need to simulate the 's_trap 2' instruction on targets that run in
+  // PRIV=1 (where it is treated as a nop).
+  if (Subtarget->hasPrivEnabledTrap2NopBug())
+    return DAG.getNode(AMDGPUISD::SIMULATED_TRAP, SL, MVT::Other, Chain);
+
   uint64_t TrapID = static_cast<uint64_t>(GCNSubtarget::TrapID::LLVMAMDHSATrap);
   SDValue Ops[] = {
     Chain,
diff --git a/llvm/lib/Target/AMDGPU/SIInsertWaitcnts.cpp b/llvm/lib/Target/AMDGPU/SIInsertWaitcnts.cpp
index 36de5b89af02..91a1c40dd824 100644
--- a/llvm/lib/Target/AMDGPU/SIInsertWaitcnts.cpp
+++ b/llvm/lib/Target/AMDGPU/SIInsertWaitcnts.cpp
@@ -2252,12 +2252,12 @@ bool SIInsertWaitcnts::insertWaitcntInBlock(MachineFunction &MF,
 
     // Don't examine operands unless we need to track vccz correctness.
     if (ST->hasReadVCCZBug() || !ST->partialVCCWritesUpdateVCCZ()) {
-      if (Inst.definesRegister(AMDGPU::VCC_LO) ||
-          Inst.definesRegister(AMDGPU::VCC_HI)) {
+      if (Inst.definesRegister(AMDGPU::VCC_LO, /*TRI=*/nullptr) ||
+          Inst.definesRegister(AMDGPU::VCC_HI, /*TRI=*/nullptr)) {
         // Up to gfx9, writes to vcc_lo and vcc_hi don't update vccz.
         if (!ST->partialVCCWritesUpdateVCCZ())
           VCCZCorrect = false;
-      } else if (Inst.definesRegister(AMDGPU::VCC)) {
+      } else if (Inst.definesRegister(AMDGPU::VCC, /*TRI=*/nullptr)) {
         // There is a hardware bug on CI/SI where SMRD instruction may corrupt
         // vccz bit, so when we detect that an instruction may read from a
         // corrupt vccz bit, we need to:
diff --git a/llvm/lib/Target/AMDGPU/SIInstrInfo.cpp b/llvm/lib/Target/AMDGPU/SIInstrInfo.cpp
index f4b21b7dfac3..76b90042d65f 100644
--- a/llvm/lib/Target/AMDGPU/SIInstrInfo.cpp
+++ b/llvm/lib/Target/AMDGPU/SIInstrInfo.cpp
@@ -164,7 +164,7 @@ static bool resultDependsOnExec(const MachineInstr &MI) {
         break;
       case AMDGPU::S_AND_B32:
       case AMDGPU::S_AND_B64:
-        if (!Use.readsRegister(AMDGPU::EXEC))
+        if (!Use.readsRegister(AMDGPU::EXEC, /*TRI=*/nullptr))
           return true;
         break;
       default:
@@ -461,8 +461,10 @@ bool SIInstrInfo::getMemOperandsWithOffsetWidth(
     return true;
   }
 
-  if (isMIMG(LdSt)) {
-    int SRsrcIdx = AMDGPU::getNamedOperandIdx(Opc, AMDGPU::OpName::srsrc);
+  if (isImage(LdSt)) {
+    auto RsrcOpName =
+        isMIMG(LdSt) ? AMDGPU::OpName::srsrc : AMDGPU::OpName::rsrc;
+    int SRsrcIdx = AMDGPU::getNamedOperandIdx(Opc, RsrcOpName);
     BaseOps.push_back(&LdSt.getOperand(SRsrcIdx));
     int VAddr0Idx = AMDGPU::getNamedOperandIdx(Opc, AMDGPU::OpName::vaddr0);
     if (VAddr0Idx >= 0) {
@@ -2026,6 +2028,57 @@ void SIInstrInfo::insertReturn(MachineBasicBlock &MBB) const {
   }
 }
 
+MachineBasicBlock *SIInstrInfo::insertSimulatedTrap(MachineRegisterInfo &MRI,
+                                                    MachineBasicBlock &MBB,
+                                                    MachineInstr &MI,
+                                                    const DebugLoc &DL) const {
+  MachineFunction *MF = MBB.getParent();
+  MachineBasicBlock *SplitBB = MBB.splitAt(MI, /*UpdateLiveIns=*/false);
+  MachineBasicBlock *HaltLoop = MF->CreateMachineBasicBlock();
+  MF->push_back(HaltLoop);
+
+  constexpr unsigned DoorbellIDMask = 0x3ff;
+  constexpr unsigned ECQueueWaveAbort = 0x400;
+
+  // Start with a `s_trap 2`, if we're in PRIV=1 and we need the workaround this
+  // will be a nop.
+  BuildMI(MBB, MI, DL, get(AMDGPU::S_TRAP))
+      .addImm(static_cast<unsigned>(GCNSubtarget::TrapID::LLVMAMDHSATrap));
+  Register DoorbellReg = MRI.createVirtualRegister(&AMDGPU::SReg_32RegClass);
+  BuildMI(MBB, MI, DL, get(AMDGPU::S_SENDMSG_RTN_B32), DoorbellReg)
+      .addImm(AMDGPU::SendMsg::ID_RTN_GET_DOORBELL);
+  BuildMI(MBB, MI, DL, get(AMDGPU::S_MOV_B32), AMDGPU::TTMP2)
+      .addUse(AMDGPU::M0);
+  Register DoorbellRegMasked =
+      MRI.createVirtualRegister(&AMDGPU::SReg_32RegClass);
+  BuildMI(MBB, MI, DL, get(AMDGPU::S_AND_B32), DoorbellRegMasked)
+      .addUse(DoorbellReg)
+      .addImm(DoorbellIDMask);
+  Register SetWaveAbortBit =
+      MRI.createVirtualRegister(&AMDGPU::SReg_32RegClass);
+  BuildMI(MBB, MI, DL, get(AMDGPU::S_OR_B32), SetWaveAbortBit)
+      .addUse(DoorbellRegMasked)
+      .addImm(ECQueueWaveAbort);
+  BuildMI(MBB, MI, DL, get(AMDGPU::S_MOV_B32), AMDGPU::M0)
+      .addUse(SetWaveAbortBit);
+  BuildMI(MBB, MI, DL, get(AMDGPU::S_SENDMSG))
+      .addImm(AMDGPU::SendMsg::ID_INTERRUPT);
+  BuildMI(MBB, MI, DL, get(AMDGPU::S_MOV_B32), AMDGPU::M0)
+      .addUse(AMDGPU::TTMP2);
+  BuildMI(MBB, MI, DL, get(AMDGPU::S_BRANCH)).addMBB(HaltLoop);
+
+  BuildMI(*HaltLoop, HaltLoop->end(), DL, get(AMDGPU::S_SETHALT)).addImm(5);
+  BuildMI(*HaltLoop, HaltLoop->end(), DL, get(AMDGPU::S_BRANCH))
+      .addMBB(HaltLoop);
+
+  if (SplitBB != &MBB)
+    MBB.removeSuccessor(SplitBB);
+  MBB.addSuccessor(HaltLoop);
+  HaltLoop->addSuccessor(HaltLoop);
+
+  return SplitBB;
+}
+
 unsigned SIInstrInfo::getNumWaitStates(const MachineInstr &MI) {
   switch (MI.getOpcode()) {
   default:
@@ -6689,7 +6742,7 @@ SIInstrInfo::legalizeOperands(MachineInstr &MI,
       // Also include following copies of the return value
       ++End;
       while (End != MBB.end() && End->isCopy() && End->getOperand(1).isReg() &&
-             MI.definesRegister(End->getOperand(1).getReg()))
+             MI.definesRegister(End->getOperand(1).getReg(), /*TRI=*/nullptr))
         ++End;
       CreatedBB =
           loadMBUFScalarOperandsFromVGPR(*this, MI, {Dest}, MDT, Start, End);
@@ -7257,7 +7310,7 @@ void SIInstrInfo::moveToVALUImpl(SIInstrWorklist &Worklist,
           .add(Inst.getOperand(1));
     }
     legalizeOperands(*NewInstr, MDT);
-    int SCCIdx = Inst.findRegisterDefOperandIdx(AMDGPU::SCC);
+    int SCCIdx = Inst.findRegisterDefOperandIdx(AMDGPU::SCC, /*TRI=*/nullptr);
     MachineOperand SCCOp = Inst.getOperand(SCCIdx);
     addSCCDefUsersToVALUWorklist(SCCOp, Inst, Worklist, CondReg);
     Inst.eraseFromParent();
@@ -7523,7 +7576,7 @@ void SIInstrInfo::lowerSelect(SIInstrWorklist &Worklist, MachineInstr &Inst,
     for (MachineInstr &CandI :
          make_range(std::next(MachineBasicBlock::reverse_iterator(Inst)),
                     Inst.getParent()->rend())) {
-      if (CandI.findRegisterDefOperandIdx(AMDGPU::SCC, false, false, &RI) !=
+      if (CandI.findRegisterDefOperandIdx(AMDGPU::SCC, &RI, false, false) !=
           -1) {
         if (CandI.isCopy() && CandI.getOperand(0).getReg() == AMDGPU::SCC) {
           BuildMI(MBB, MII, DL, get(AMDGPU::COPY), NewCondReg)
@@ -8338,7 +8391,7 @@ void SIInstrInfo::addSCCDefUsersToVALUWorklist(MachineOperand &Op,
        make_range(std::next(MachineBasicBlock::iterator(SCCDefInst)),
                   SCCDefInst.getParent()->end())) {
     // Check if SCC is used first.
-    int SCCIdx = MI.findRegisterUseOperandIdx(AMDGPU::SCC, false, &RI);
+    int SCCIdx = MI.findRegisterUseOperandIdx(AMDGPU::SCC, &RI, false);
     if (SCCIdx != -1) {
       if (MI.isCopy()) {
         MachineRegisterInfo &MRI = MI.getParent()->getParent()->getRegInfo();
@@ -8355,7 +8408,7 @@ void SIInstrInfo::addSCCDefUsersToVALUWorklist(MachineOperand &Op,
       }
     }
     // Exit if we find another SCC def.
-    if (MI.findRegisterDefOperandIdx(AMDGPU::SCC, false, false, &RI) != -1)
+    if (MI.findRegisterDefOperandIdx(AMDGPU::SCC, &RI, false, false) != -1)
       break;
   }
   for (auto &Copy : CopyToDelete)
@@ -9408,7 +9461,7 @@ MachineInstr *SIInstrInfo::createPHIDestinationCopy(
   auto Cur = MBB.begin();
   if (Cur != MBB.end())
     do {
-      if (!Cur->isPHI() && Cur->readsRegister(Dst))
+      if (!Cur->isPHI() && Cur->readsRegister(Dst, /*TRI=*/nullptr))
         return BuildMI(MBB, Cur, DL, get(TargetOpcode::COPY), Dst).addReg(Src);
       ++Cur;
     } while (Cur != MBB.end() && Cur != LastPHIIt);
@@ -9424,7 +9477,7 @@ MachineInstr *SIInstrInfo::createPHISourceCopy(
       (InsPt->getOpcode() == AMDGPU::SI_IF ||
        InsPt->getOpcode() == AMDGPU::SI_ELSE ||
        InsPt->getOpcode() == AMDGPU::SI_IF_BREAK) &&
-      InsPt->definesRegister(Src)) {
+      InsPt->definesRegister(Src, /*TRI=*/nullptr)) {
     InsPt++;
     return BuildMI(MBB, InsPt, DL,
                    get(ST.isWave32() ? AMDGPU::S_MOV_B32_term
@@ -9796,7 +9849,8 @@ bool SIInstrInfo::optimizeCompareInstr(MachineInstr &CmpInstr, Register SrcReg,
         return false;
     }
 
-    MachineOperand *SccDef = Def->findRegisterDefOperand(AMDGPU::SCC);
+    MachineOperand *SccDef =
+        Def->findRegisterDefOperand(AMDGPU::SCC, /*TRI=*/nullptr);
     SccDef->setIsDead(false);
     CmpInstr.eraseFromParent();
 
diff --git a/llvm/lib/Target/AMDGPU/SIInstrInfo.h b/llvm/lib/Target/AMDGPU/SIInstrInfo.h
index 4c5978cdc666..b314b9b2fb51 100644
--- a/llvm/lib/Target/AMDGPU/SIInstrInfo.h
+++ b/llvm/lib/Target/AMDGPU/SIInstrInfo.h
@@ -1194,6 +1194,15 @@ public:
                    unsigned Quantity) const override;
 
   void insertReturn(MachineBasicBlock &MBB) const;
+
+  /// Build instructions that simulate the behavior of a `s_trap 2` instructions
+  /// for hardware (namely, gfx11) that runs in PRIV=1 mode. There, s_trap is
+  /// interpreted as a nop.
+  MachineBasicBlock *insertSimulatedTrap(MachineRegisterInfo &MRI,
+                                         MachineBasicBlock &MBB,
+                                         MachineInstr &MI,
+                                         const DebugLoc &DL) const;
+
   /// Return the number of wait states that result from executing this
   /// instruction.
   static unsigned getNumWaitStates(const MachineInstr &MI);
diff --git a/llvm/lib/Target/AMDGPU/SIInstrInfo.td b/llvm/lib/Target/AMDGPU/SIInstrInfo.td
index f1afbcc060b2..7189e6e40506 100644
--- a/llvm/lib/Target/AMDGPU/SIInstrInfo.td
+++ b/llvm/lib/Target/AMDGPU/SIInstrInfo.td
@@ -1000,8 +1000,10 @@ def SDWAVopcDst : BoolRC {
 }
 
 class NamedIntOperand<ValueType Type, string Prefix, bit Optional = 1,
-                      string name = NAME, string ConvertMethod = "nullptr">
+                      string name = NAME>
     : CustomOperand<Type, Optional, name> {
+  string Validator = "[](int64_t V) { return true; }";
+  string ConvertMethod = "[](int64_t &V) { return "#Validator#"(V); }";
   let ParserMethod =
     "[this](OperandVector &Operands) -> ParseStatus { "#
     "return parseIntWithPrefix(\""#Prefix#"\", Operands, "#
@@ -1045,8 +1047,10 @@ class ArrayOperand0<string Id, string Name = NAME>
 let ImmTy = "ImmTyOffset" in
 def flat_offset : CustomOperand<i32, 1, "FlatOffset">;
 def Offset : NamedIntOperand<i32, "offset">;
+let Validator = "isUInt<8>" in {
 def Offset0 : NamedIntOperand<i8, "offset0">;
 def Offset1 : NamedIntOperand<i8, "offset1">;
+}
 
 def gds : NamedBitOperand<"gds", "GDS">;
 
@@ -1103,25 +1107,41 @@ let DefaultValue = "0xf" in {
 def DppRowMask : NamedIntOperand<i32, "row_mask">;
 def DppBankMask : NamedIntOperand<i32, "bank_mask">;
 }
-def DppBoundCtrl : NamedIntOperand<i1, "bound_ctrl", 1, "DppBoundCtrl",
-    "[this] (int64_t &BC) -> bool { return convertDppBoundCtrl(BC); }">;
+def DppBoundCtrl : NamedIntOperand<i1, "bound_ctrl"> {
+  let ConvertMethod = "[this] (int64_t &BC) -> bool { return convertDppBoundCtrl(BC); }";
+}
 
 let DecoderMethod = "decodeDpp8FI" in
 def Dpp8FI : NamedIntOperand<i32, "fi", 1, "DppFI">;
 def Dpp16FI : NamedIntOperand<i32, "fi", 1, "DppFI">;
 
 def blgp : CustomOperand<i32, 1, "BLGP">;
-def CBSZ : NamedIntOperand<i32, "cbsz">;
-def ABID : NamedIntOperand<i32, "abid">;
-
+def CBSZ : NamedIntOperand<i32, "cbsz"> {
+  let Validator = "isUInt<3>";
+}
+def ABID : NamedIntOperand<i32, "abid"> {
+  let Validator = "isUInt<4>";
+}
 def hwreg : CustomOperand<i32, 0, "Hwreg">;
 
 def exp_tgt : CustomOperand<i32, 0, "ExpTgt">;
 
-def WaitVDST : NamedIntOperand<i8, "wait_vdst">;
-def WaitEXP : NamedIntOperand<i8, "wait_exp">;
-def WaitVAVDst : NamedIntOperand<i8, "wait_va_vdst">;
-def WaitVMVSrc : NamedIntOperand<i8, "wait_vm_vsrc">;
+def WaitVDST : NamedIntOperand<i8, "wait_vdst"> {
+  let Validator = "isUInt<4>";
+}
+def WaitEXP : NamedIntOperand<i8, "wait_exp"> {
+  let Validator = "isUInt<3>";
+}
+def WaitVAVDst : NamedIntOperand<i8, "wait_va_vdst"> {
+  let Validator = "isUInt<4>";
+}
+def WaitVMVSrc : NamedIntOperand<i8, "wait_vm_vsrc"> {
+  let Validator = "isUInt<1>";
+}
+
+def ByteSel : NamedIntOperand<i8, "byte_sel"> {
+  let Validator = "isUInt<2>";
+}
 
 class KImmFPOperand<ValueType vt> : ImmOperand<vt> {
   let OperandNamespace = "AMDGPU";
@@ -1700,9 +1720,9 @@ class getIns64 <RegisterOperand Src0RC, RegisterOperand Src1RC,
           (ins Src0Mod:$src0_modifiers, Src0RC:$src0,
                Src1Mod:$src1_modifiers, Src1RC:$src1,
                clampmod0:$clamp, omod0:$omod),
-           (ins Src0Mod:$src0_modifiers, Src0RC:$src0,
-               Src1Mod:$src1_modifiers, Src1RC:$src1,
-               clampmod0:$clamp))
+          !con((ins Src0Mod:$src0_modifiers, Src0RC:$src0,
+                    Src1Mod:$src1_modifiers, Src1RC:$src1),
+                !if(HasClamp, (ins clampmod0:$clamp), (ins))))
       /* else */,
         // VOP2 without modifiers
         !if (HasClamp,
@@ -2036,7 +2056,8 @@ class getAsmDPP8 <bit HasDst, int NumSrcArgs, bit HasModifiers, ValueType DstVT
 class getAsmVOP3Base <int NumSrcArgs, bit HasDst, bit HasClamp,
                        bit HasOpSel, bit HasOMod, bit IsVOP3P,
                        bit HasModifiers, bit Src0HasMods,
-                       bit Src1HasMods, bit Src2HasMods, ValueType DstVT = i32> {
+                       bit Src1HasMods, bit Src2HasMods, ValueType DstVT = i32,
+                       bit HasByteSel = 0> {
   string dst = !if(HasDst,
                    !if(!eq(DstVT.Size, 1),
                        "$sdst",
@@ -2058,6 +2079,7 @@ class getAsmVOP3Base <int NumSrcArgs, bit HasDst, bit HasClamp,
   string src1 = !if(Src1HasMods, src1mods, src1nomods);
   string src2 = !if(Src2HasMods, src2mods, src2nomods);
   string opsel = !if(HasOpSel, "$op_sel", "");
+  string bytesel = !if(HasByteSel, "$byte_sel", "");
   string 3PMods = !if(IsVOP3P,
                       !if(HasOpSel, "$op_sel_hi", "")
                         #!if(HasModifiers, "$neg_lo$neg_hi", ""),
@@ -2065,7 +2087,7 @@ class getAsmVOP3Base <int NumSrcArgs, bit HasDst, bit HasClamp,
   string clamp = !if(HasClamp, "$clamp", "");
   string omod = !if(HasOMod, "$omod", "");
 
-  string ret = dst#!if(!gt(NumSrcArgs,0),", "#src0#src1#src2#opsel#3PMods#clamp#omod, "");
+  string ret = dst#!if(!gt(NumSrcArgs,0),", "#src0#src1#src2#opsel#bytesel#3PMods#clamp#omod, "");
 
 }
 
@@ -2282,6 +2304,7 @@ class VOPProfile <list<ValueType> _ArgVT, bit _EnableClamp = 0> {
   field bit IsSWMMAC = 0;
 
   field bit IsFP8 = 0;
+  field bit IsFP8DstByteSel = 0;
 
   field bit HasDst = !ne(DstVT.Value, untyped.Value);
   field bit HasDst32 = HasDst;
@@ -2401,7 +2424,7 @@ class VOPProfile <list<ValueType> _ArgVT, bit _EnableClamp = 0> {
   field string AsmDPP8 = getAsmDPP8<HasDst, NumSrcArgs, 0 /*HasModifiers*/, DstVT>.ret;
   field string AsmVOP3Base = getAsmVOP3Base<NumSrcArgs, HasDst, HasClamp,
    HasOpSel, HasOMod, IsVOP3P, HasModifiers, HasModifiers, HasModifiers,
-   HasModifiers, DstVT>.ret;
+   HasModifiers, DstVT, IsFP8DstByteSel>.ret;
   field string Asm64 = AsmVOP3Base;
   field string AsmVOP3P = getAsmVOP3P<NumSrcArgs, HasModifiers, HasClamp, HasOpSel>.ret;
   field string AsmVOP3OpSel = getAsmVOP3OpSel<NumSrcArgs,
diff --git a/llvm/lib/Target/AMDGPU/SIInstructions.td b/llvm/lib/Target/AMDGPU/SIInstructions.td
index d6d49889656b..cca8d96f29c0 100644
--- a/llvm/lib/Target/AMDGPU/SIInstructions.td
+++ b/llvm/lib/Target/AMDGPU/SIInstructions.td
@@ -106,6 +106,12 @@ def ENDPGM_TRAP : SPseudoInstSI<
   let usesCustomInserter = 1;
 }
 
+def SIMULATED_TRAP : SPseudoInstSI<(outs), (ins), [(AMDGPUsimulated_trap)],
+                                   "SIMULATED_TRAP"> {
+  let hasSideEffects = 1;
+  let usesCustomInserter = 1;
+}
+
 def ATOMIC_FENCE : SPseudoInstSI<
   (outs), (ins i32imm:$ordering, i32imm:$scope),
   [(atomic_fence (i32 timm:$ordering), (i32 timm:$scope))],
diff --git a/llvm/lib/Target/AMDGPU/SIMachineFunctionInfo.cpp b/llvm/lib/Target/AMDGPU/SIMachineFunctionInfo.cpp
index 12433dc83c48..072c5aedc220 100644
--- a/llvm/lib/Target/AMDGPU/SIMachineFunctionInfo.cpp
+++ b/llvm/lib/Target/AMDGPU/SIMachineFunctionInfo.cpp
@@ -110,7 +110,8 @@ SIMachineFunctionInfo::SIMachineFunctionInfo(const Function &F,
   }
 
   if (!AMDGPU::isGraphics(CC) ||
-      (CC == CallingConv::AMDGPU_CS && ST.hasArchitectedSGPRs())) {
+      ((CC == CallingConv::AMDGPU_CS || CC == CallingConv::AMDGPU_Gfx) &&
+       ST.hasArchitectedSGPRs())) {
     if (IsKernel || !F.hasFnAttribute("amdgpu-no-workgroup-id-x"))
       WorkGroupIDX = true;
 
diff --git a/llvm/lib/Target/AMDGPU/SIOptimizeExecMaskingPreRA.cpp b/llvm/lib/Target/AMDGPU/SIOptimizeExecMaskingPreRA.cpp
index d2a5eb89da12..c91d241f81ab 100644
--- a/llvm/lib/Target/AMDGPU/SIOptimizeExecMaskingPreRA.cpp
+++ b/llvm/lib/Target/AMDGPU/SIOptimizeExecMaskingPreRA.cpp
@@ -456,7 +456,8 @@ bool SIOptimizeExecMaskingPreRA::runOnMachineFunction(MachineFunction &MF) {
       Register SavedExec = I->getOperand(0).getReg();
       if (SavedExec.isVirtual() && MRI->hasOneNonDBGUse(SavedExec)) {
         MachineInstr *SingleExecUser = &*MRI->use_instr_nodbg_begin(SavedExec);
-        int Idx = SingleExecUser->findRegisterUseOperandIdx(SavedExec);
+        int Idx = SingleExecUser->findRegisterUseOperandIdx(SavedExec,
+                                                            /*TRI=*/nullptr);
         assert(Idx != -1);
         if (SingleExecUser->getParent() == I->getParent() &&
             !SingleExecUser->getOperand(Idx).isImplicit() &&
diff --git a/llvm/lib/Target/AMDGPU/SIPreEmitPeephole.cpp b/llvm/lib/Target/AMDGPU/SIPreEmitPeephole.cpp
index 82da53d0c0eb..875bccb208c8 100644
--- a/llvm/lib/Target/AMDGPU/SIPreEmitPeephole.cpp
+++ b/llvm/lib/Target/AMDGPU/SIPreEmitPeephole.cpp
@@ -171,7 +171,7 @@ bool SIPreEmitPeephole::optimizeVccBranch(MachineInstr &MI) const {
   if (A->getOpcode() == AndN2)
     MaskValue = ~MaskValue;
 
-  if (!ReadsCond && A->registerDefIsDead(AMDGPU::SCC)) {
+  if (!ReadsCond && A->registerDefIsDead(AMDGPU::SCC, /*TRI=*/nullptr)) {
     if (!MI.killsRegister(CondReg, TRI)) {
       // Replace AND with MOV
       if (MaskValue == 0) {
@@ -235,7 +235,7 @@ bool SIPreEmitPeephole::optimizeVccBranch(MachineInstr &MI) const {
         TII->get(IsVCCZ ? AMDGPU::S_CBRANCH_EXECZ : AMDGPU::S_CBRANCH_EXECNZ));
   }
 
-  MI.removeOperand(MI.findRegisterUseOperandIdx(CondReg, false /*Kill*/, TRI));
+  MI.removeOperand(MI.findRegisterUseOperandIdx(CondReg, TRI, false /*Kill*/));
   MI.addImplicitDefUseOperands(*MBB.getParent());
 
   return true;
diff --git a/llvm/lib/Target/AMDGPU/SIRegisterInfo.cpp b/llvm/lib/Target/AMDGPU/SIRegisterInfo.cpp
index acb54fd10b90..ddb5f7193568 100644
--- a/llvm/lib/Target/AMDGPU/SIRegisterInfo.cpp
+++ b/llvm/lib/Target/AMDGPU/SIRegisterInfo.cpp
@@ -2366,8 +2366,8 @@ bool SIRegisterInfo::eliminateFrameIndex(MachineBasicBlock::iterator MI,
           return false;
         }
 
-        bool NeedSaveSCC =
-            RS->isRegUsed(AMDGPU::SCC) && !MI->definesRegister(AMDGPU::SCC);
+        bool NeedSaveSCC = RS->isRegUsed(AMDGPU::SCC) &&
+                           !MI->definesRegister(AMDGPU::SCC, /*TRI=*/nullptr);
 
         Register TmpSReg =
             UseSGPR ? TmpReg
@@ -2409,7 +2409,8 @@ bool SIRegisterInfo::eliminateFrameIndex(MachineBasicBlock::iterator MI,
 
         if (TmpSReg == FrameReg) {
           // Undo frame register modification.
-          if (NeedSaveSCC && !MI->registerDefIsDead(AMDGPU::SCC)) {
+          if (NeedSaveSCC &&
+              !MI->registerDefIsDead(AMDGPU::SCC, /*TRI=*/nullptr)) {
             MachineBasicBlock::iterator I =
                 BuildMI(*MBB, std::next(MI), DL, TII->get(AMDGPU::S_ADDC_U32),
                         TmpSReg)
@@ -2439,8 +2440,8 @@ bool SIRegisterInfo::eliminateFrameIndex(MachineBasicBlock::iterator MI,
         // Convert to a swizzled stack address by scaling by the wave size.
         // In an entry function/kernel the offset is already swizzled.
         bool IsSALU = isSGPRClass(TII->getOpRegClass(*MI, FIOperandNum));
-        bool LiveSCC =
-            RS->isRegUsed(AMDGPU::SCC) && !MI->definesRegister(AMDGPU::SCC);
+        bool LiveSCC = RS->isRegUsed(AMDGPU::SCC) &&
+                       !MI->definesRegister(AMDGPU::SCC, /*TRI=*/nullptr);
         const TargetRegisterClass *RC = IsSALU && !LiveSCC
                                             ? &AMDGPU::SReg_32RegClass
                                             : &AMDGPU::VGPR_32RegClass;
diff --git a/llvm/lib/Target/AMDGPU/SIShrinkInstructions.cpp b/llvm/lib/Target/AMDGPU/SIShrinkInstructions.cpp
index 3c6f6ddfd89d..647fae904d39 100644
--- a/llvm/lib/Target/AMDGPU/SIShrinkInstructions.cpp
+++ b/llvm/lib/Target/AMDGPU/SIShrinkInstructions.cpp
@@ -1014,7 +1014,7 @@ bool SIShrinkInstructions::runOnMachineFunction(MachineFunction &MF) {
 
       // Copy deadness from the old explicit vcc def to the new implicit def.
       if (SDst && SDst->isDead())
-        Inst32->findRegisterDefOperand(VCCReg)->setIsDead();
+        Inst32->findRegisterDefOperand(VCCReg, /*TRI=*/nullptr)->setIsDead();
 
       MI.eraseFromParent();
       foldImmediates(*Inst32);
diff --git a/llvm/lib/Target/AMDGPU/SIWholeQuadMode.cpp b/llvm/lib/Target/AMDGPU/SIWholeQuadMode.cpp
index 5e6c34992930..ea8109bbee9a 100644
--- a/llvm/lib/Target/AMDGPU/SIWholeQuadMode.cpp
+++ b/llvm/lib/Target/AMDGPU/SIWholeQuadMode.cpp
@@ -1525,10 +1525,10 @@ void SIWholeQuadMode::lowerCopyInstrs() {
         MI->getOperand(0).setIsEarlyClobber(false);
         LIS->createAndComputeVirtRegInterval(Reg);
       }
-      int Index = MI->findRegisterUseOperandIdx(AMDGPU::EXEC);
+      int Index = MI->findRegisterUseOperandIdx(AMDGPU::EXEC, /*TRI=*/nullptr);
       while (Index >= 0) {
         MI->removeOperand(Index);
-        Index = MI->findRegisterUseOperandIdx(AMDGPU::EXEC);
+        Index = MI->findRegisterUseOperandIdx(AMDGPU::EXEC, /*TRI=*/nullptr);
       }
       MI->setDesc(TII->get(AMDGPU::COPY));
       LLVM_DEBUG(dbgs() << "  -> " << *MI);
diff --git a/llvm/lib/Target/AMDGPU/SOPInstructions.td b/llvm/lib/Target/AMDGPU/SOPInstructions.td
index 0b7d45ee8c02..93b7e86b5f29 100644
--- a/llvm/lib/Target/AMDGPU/SOPInstructions.td
+++ b/llvm/lib/Target/AMDGPU/SOPInstructions.td
@@ -1803,7 +1803,7 @@ def : GCNPat<
 let SubtargetPredicate = isNotGFX12Plus in
   def : GCNPat <(int_amdgcn_s_wait_event_export_ready), (S_WAIT_EVENT (i16 0))>;
 let SubtargetPredicate = isGFX12Plus in
-  def : GCNPat <(int_amdgcn_s_wait_event_export_ready), (S_WAIT_EVENT (i16 1))>;
+  def : GCNPat <(int_amdgcn_s_wait_event_export_ready), (S_WAIT_EVENT (i16 2))>;
 
 // The first 10 bits of the mode register are the core FP mode on all
 // subtargets.
diff --git a/llvm/lib/Target/AMDGPU/VOP3Instructions.td b/llvm/lib/Target/AMDGPU/VOP3Instructions.td
index 647595d9ccab..616bc7684753 100644
--- a/llvm/lib/Target/AMDGPU/VOP3Instructions.td
+++ b/llvm/lib/Target/AMDGPU/VOP3Instructions.td
@@ -580,6 +580,22 @@ def VOP3_CVT_SR_F8_F32_Profile : VOP3_Profile<VOPProfile<[i32, f32, i32, f32]>,
                                           HasSrc2FloatMods>.ret>.ret);
 }
 
+class VOP3_CVT_SR_F8_ByteSel_Profile<ValueType SrcVT> :
+  VOP3_Profile<VOPProfile<[i32, SrcVT, i32, untyped]>> {
+  let IsFP8DstByteSel = 1;
+  let HasClamp = 0;
+  defvar bytesel = (ins VGPR_32:$vdst_in, ByteSel:$byte_sel);
+  let Ins64 = !con(getIns64<Src0RC64, Src1RC64, Src2RC64, NumSrcArgs,
+                            HasClamp, HasModifiers, HasSrc2Mods,
+                            HasOMod, Src0Mod, Src1Mod, Src2Mod>.ret,
+                   bytesel);
+  let InsVOP3Base = !con(
+    getInsVOP3Base<Src0VOP3DPP, Src1VOP3DPP,
+                   Src2VOP3DPP, NumSrcArgs, HasClamp, HasModifiers, HasSrc2Mods, HasOMod,
+                   Src0ModVOP3DPP, Src1ModVOP3DPP, Src2ModVOP3DPP, HasOpSel>.ret,
+    bytesel);
+}
+
 def IsPow2Plus1: PatLeaf<(i32 imm), [{
   uint32_t V = N->getZExtValue();
   return isPowerOf2_32(V - 1);
@@ -645,12 +661,17 @@ let OtherPredicates = [HasFP8ConversionInsts], mayRaiseFPException = 0,
   let Constraints = "$vdst = $vdst_in", DisableEncoding = "$vdst_in" in {
     defm V_CVT_PK_FP8_F32 : VOP3Inst<"v_cvt_pk_fp8_f32", VOP3_CVT_PK_F8_F32_Profile>;
     defm V_CVT_PK_BF8_F32 : VOP3Inst<"v_cvt_pk_bf8_f32", VOP3_CVT_PK_F8_F32_Profile>;
+
+    let SubtargetPredicate = isGFX12Plus in {
+      defm V_CVT_SR_FP8_F32_gfx12 : VOP3Inst<"v_cvt_sr_fp8_f32_gfx12", VOP3_CVT_SR_F8_ByteSel_Profile<f32>>;
+      defm V_CVT_SR_BF8_F32_gfx12 : VOP3Inst<"v_cvt_sr_bf8_f32_gfx12", VOP3_CVT_SR_F8_ByteSel_Profile<f32>>;
+    }
   }
 
   // These instructions have non-standard use of op_sel. In particular they are
   // using op_sel bits 2 and 3 while only having two sources. Therefore dummy
   // src2 is used to hold the op_sel value.
-  let Constraints = "$vdst = $src2", DisableEncoding = "$src2" in {
+  let Constraints = "$vdst = $src2", DisableEncoding = "$src2", SubtargetPredicate = isGFX940Plus in {
     defm V_CVT_SR_FP8_F32 : VOP3Inst<"v_cvt_sr_fp8_f32", VOP3_CVT_SR_F8_F32_Profile>;
     defm V_CVT_SR_BF8_F32 : VOP3Inst<"v_cvt_sr_bf8_f32", VOP3_CVT_SR_F8_F32_Profile>;
   }
@@ -667,15 +688,28 @@ class Cvt_SR_F8_F32_Pat<SDPatternOperator node, bits<2> index, VOP3_Pseudo inst>
           !if(index{0}, SRCMODS.OP_SEL_0, 0), $old, 0)
 >;
 
+class Cvt_SR_F8_ByteSel_Pat<SDPatternOperator node, VOP3_Pseudo inst, ValueType SrcVT> : GCNPat<
+    (i32 (node (VOP3Mods SrcVT:$src0, i32:$src0_modifiers), (VOP3Mods i32:$src1, i32:$src1_modifiers),
+          i32:$old, timm:$byte_sel)),
+    (inst $src0_modifiers, $src0, $src1_modifiers, $src1, $old, (as_i32timm $byte_sel))
+>;
+
 let OtherPredicates = [HasFP8ConversionInsts] in {
 foreach Index = [0, -1] in {
   def : Cvt_PK_F8_F32_Pat<int_amdgcn_cvt_pk_fp8_f32, Index, V_CVT_PK_FP8_F32_e64>;
   def : Cvt_PK_F8_F32_Pat<int_amdgcn_cvt_pk_bf8_f32, Index, V_CVT_PK_BF8_F32_e64>;
 }
 
-foreach Index = [0, 1, 2, 3] in {
-  def : Cvt_SR_F8_F32_Pat<int_amdgcn_cvt_sr_fp8_f32, Index, V_CVT_SR_FP8_F32_e64>;
-  def : Cvt_SR_F8_F32_Pat<int_amdgcn_cvt_sr_bf8_f32, Index, V_CVT_SR_BF8_F32_e64>;
+let SubtargetPredicate = isGFX940Plus in {
+  foreach Index = [0, 1, 2, 3] in {
+    def : Cvt_SR_F8_F32_Pat<int_amdgcn_cvt_sr_fp8_f32, Index, V_CVT_SR_FP8_F32_e64>;
+    def : Cvt_SR_F8_F32_Pat<int_amdgcn_cvt_sr_bf8_f32, Index, V_CVT_SR_BF8_F32_e64>;
+  }
+}
+
+let SubtargetPredicate = isGFX12Plus in {
+  def : Cvt_SR_F8_ByteSel_Pat<int_amdgcn_cvt_sr_fp8_f32, V_CVT_SR_FP8_F32_gfx12_e64, f32>;
+  def : Cvt_SR_F8_ByteSel_Pat<int_amdgcn_cvt_sr_bf8_f32, V_CVT_SR_BF8_F32_gfx12_e64, f32>;
 }
 }
 
@@ -1040,8 +1074,8 @@ defm V_PERMLANEX16_VAR_B32 : VOP3Only_Real_Base_gfx12<0x310>;
 
 defm V_CVT_PK_FP8_F32  : VOP3Only_Realtriple_gfx12<0x369>;
 defm V_CVT_PK_BF8_F32  : VOP3Only_Realtriple_gfx12<0x36a>;
-defm V_CVT_SR_FP8_F32  : VOP3Only_Realtriple_gfx12<0x36b>;
-defm V_CVT_SR_BF8_F32  : VOP3Only_Realtriple_gfx12<0x36c>;
+defm V_CVT_SR_FP8_F32_gfx12 : VOP3_Realtriple_with_name_gfx12<0x36b, "V_CVT_SR_FP8_F32_gfx12", "v_cvt_sr_fp8_f32" >;
+defm V_CVT_SR_BF8_F32_gfx12 : VOP3_Realtriple_with_name_gfx12<0x36c, "V_CVT_SR_BF8_F32_gfx12", "v_cvt_sr_bf8_f32">;
 
 //===----------------------------------------------------------------------===//
 // GFX11, GFX12
diff --git a/llvm/lib/Target/AMDGPU/VOPInstructions.td b/llvm/lib/Target/AMDGPU/VOPInstructions.td
index da16178cb58b..7cdb5cbfe297 100644
--- a/llvm/lib/Target/AMDGPU/VOPInstructions.td
+++ b/llvm/lib/Target/AMDGPU/VOPInstructions.td
@@ -311,6 +311,14 @@ class VOP3FP8OpSel_gfx11_gfx12<bits<10> op, VOPProfile p> : VOP3e_gfx10<op, p> {
   let Inst{12} = !if(p.HasSrc0, src0_modifiers{3}, 0);
 }
 
+ class VOP3FP8OpSel_dst_bytesel_gfx11_gfx12<bits<10> op, VOPProfile p> : VOP3e_gfx10<op, p> {
+   bits<2> byte_sel;
+
+   let Inst{11} = 0; // op_sel0
+   let Inst{12} = 0; // op_sel1
+   let Inst{14-13} = byte_sel;  // op_sel2/3
+ }
+
 class VOP3DotOpSel_gfx11_gfx12<bits<10> op, VOPProfile p> : VOP3OpSel_gfx11_gfx12<op, p>{
   let Inst{11} = ?;
   let Inst{12} = ?;
@@ -741,6 +749,7 @@ class VOP3_DPPe_Common_Base<bits<10> op, VOPProfile P> : Enc96 {
   bits<3> src2_modifiers;
   bits<1> clamp;
   bits<2> omod;
+  bits<2> byte_sel;
 
   let Inst{8}     = !if(P.HasSrc0Mods, src0_modifiers{1}, 0);
   let Inst{9}     = !if(P.HasSrc1Mods, src1_modifiers{1}, 0);
@@ -748,8 +757,8 @@ class VOP3_DPPe_Common_Base<bits<10> op, VOPProfile P> : Enc96 {
   // OPSEL must be set such that the low result only uses low inputs, and the high result only uses high inputs.
   let Inst{11} = !if(P.HasOpSel,!if(P.HasSrc0Mods, src0_modifiers{2}, 0),?);
   let Inst{12} = !if(P.HasOpSel,!if(P.HasSrc1Mods, src1_modifiers{2}, !if((P.IsFP8), src0_modifiers{3}, 0)), ?);
-  let Inst{13} = !if(P.HasOpSel,!if(P.HasSrc2Mods, src2_modifiers{2}, 0),?);
-  let Inst{14} = !if(P.HasOpSel,!if(P.HasSrc0Mods, src0_modifiers{3}, 0),?);
+  let Inst{13} = !if(P.HasOpSel,!if(P.HasSrc2Mods, src2_modifiers{2}, 0),!if(P.IsFP8DstByteSel, byte_sel{0}, ?));
+  let Inst{14} = !if(P.HasOpSel,!if(P.HasSrc0Mods, src0_modifiers{3}, 0),!if(P.IsFP8DstByteSel, byte_sel{1}, ?));
   let Inst{15}    = !if(P.HasClamp, clamp, 0);
   let Inst{25-16} = op;
   let Inst{31-26} = 0x35;
@@ -1388,7 +1397,11 @@ multiclass VOP3_Real_Base<GFXGen Gen, bits<10> op, string opName = NAME,
                           bit isSingle = 0> {
   defvar ps = !cast<VOP_Pseudo>(opName#"_e64");
   let IsSingle = !or(isSingle, ps.Pfl.IsSingle) in {
-    if ps.Pfl.HasOpSel then {
+    if ps.Pfl.IsFP8DstByteSel then {
+      def _e64#Gen.Suffix :
+        VOP3_Real_Gen<ps, Gen>,
+        VOP3FP8OpSel_dst_bytesel_gfx11_gfx12<op, ps.Pfl>;
+    } if ps.Pfl.HasOpSel then {
       def _e64#Gen.Suffix :
         VOP3_Real_Gen<ps, Gen>,
         VOP3OpSel_gfx11_gfx12<op, ps.Pfl>;
@@ -1419,6 +1432,10 @@ multiclass VOP3_Real_with_name<GFXGen Gen, bits<10> op, string opName,
       def _e64#Gen.Suffix :
         VOP3_Real_Gen<ps, Gen>,
         VOP3FP8OpSel_gfx11_gfx12<op, ps.Pfl>;
+    } else if ps.Pfl.IsFP8DstByteSel then {
+      def _e64#Gen.Suffix :
+        VOP3_Real_Gen<ps, Gen>,
+        VOP3FP8OpSel_dst_bytesel_gfx11_gfx12<op, ps.Pfl>;
     } else if ps.Pfl.HasOpSel then {
       def _e64#Gen.Suffix :
         VOP3_Real_Gen<ps, Gen>,
diff --git a/llvm/lib/Target/ARM/A15SDOptimizer.cpp b/llvm/lib/Target/ARM/A15SDOptimizer.cpp
index 3543cefeb399..be87707a297d 100644
--- a/llvm/lib/Target/ARM/A15SDOptimizer.cpp
+++ b/llvm/lib/Target/ARM/A15SDOptimizer.cpp
@@ -156,7 +156,7 @@ unsigned A15SDOptimizer::getPrefSPRLane(unsigned SReg) {
 
   MachineInstr *MI = MRI->getVRegDef(SReg);
   if (!MI) return ARM::ssub_0;
-  MachineOperand *MO = MI->findRegisterDefOperand(SReg);
+  MachineOperand *MO = MI->findRegisterDefOperand(SReg, /*TRI=*/nullptr);
   if (!MO) return ARM::ssub_0;
   assert(MO->isReg() && "Non-register operand found!");
 
@@ -192,7 +192,7 @@ void A15SDOptimizer::eraseInstrWithNoUses(MachineInstr *MI) {
       Register Reg = MO.getReg();
       if (!Reg.isVirtual())
         continue;
-      MachineOperand *Op = MI->findRegisterDefOperand(Reg);
+      MachineOperand *Op = MI->findRegisterDefOperand(Reg, /*TRI=*/nullptr);
 
       if (!Op)
         continue;
diff --git a/llvm/lib/Target/ARM/ARM.td b/llvm/lib/Target/ARM/ARM.td
index 66596dbda83c..570aae9b3c7a 100644
--- a/llvm/lib/Target/ARM/ARM.td
+++ b/llvm/lib/Target/ARM/ARM.td
@@ -15,1203 +15,8 @@
 
 include "llvm/Target/Target.td"
 
-//===----------------------------------------------------------------------===//
-// ARM Subtarget state.
-//
-
-// True if compiling for Thumb, false for ARM.
-def ModeThumb             : SubtargetFeature<"thumb-mode", "IsThumb",
-                                             "true", "Thumb mode">;
-
-// True if we're using software floating point features.
-def ModeSoftFloat         : SubtargetFeature<"soft-float","UseSoftFloat",
-                                             "true", "Use software floating "
-                                             "point features.">;
-
-
-//===----------------------------------------------------------------------===//
-// ARM Subtarget features.
-//
-
-// Floating Point, HW Division and Neon Support
-
-// FP loads/stores/moves, shared between VFP and MVE (even in the integer-only
-// version).
-def FeatureFPRegs         : SubtargetFeature<"fpregs", "HasFPRegs", "true",
-                                             "Enable FP registers">;
-
-// 16-bit FP loads/stores/moves, shared between VFP (with the v8.2A FP16
-// extension) and MVE (even in the integer-only version).
-def FeatureFPRegs16       : SubtargetFeature<"fpregs16", "HasFPRegs16", "true",
-                                             "Enable 16-bit FP registers",
-                                             [FeatureFPRegs]>;
-
-def FeatureFPRegs64       : SubtargetFeature<"fpregs64", "HasFPRegs64", "true",
-                                             "Enable 64-bit FP registers",
-                                             [FeatureFPRegs]>;
-
-// True if the floating point unit supports double precision.
-def FeatureFP64           : SubtargetFeature<"fp64", "HasFP64", "true",
-                                             "Floating point unit supports "
-                                             "double precision",
-                                             [FeatureFPRegs64]>;
-
-// True if subtarget has the full 32 double precision FP registers for VFPv3.
-def FeatureD32            : SubtargetFeature<"d32", "HasD32", "true",
-                                             "Extend FP to 32 double registers">;
-
-/// Versions of the VFP flags restricted to single precision, or to
-/// 16 d-registers, or both.
-multiclass VFPver<string name, string query, string description,
-                  list<SubtargetFeature> prev,
-                  list<SubtargetFeature> otherimplies,
-                  list<SubtargetFeature> vfp2prev = []> {
-  def _D16_SP: SubtargetFeature<
-    name#"d16sp", query#"D16SP", "true",
-    description#" with only 16 d-registers and no double precision",
-    !foreach(v, prev, !cast<SubtargetFeature>(v # "_D16_SP")) #
-      !foreach(v, vfp2prev, !cast<SubtargetFeature>(v # "_SP")) #
-      otherimplies>;
-  def _SP: SubtargetFeature<
-    name#"sp", query#"SP", "true",
-    description#" with no double precision",
-    !foreach(v, prev, !cast<SubtargetFeature>(v # "_SP")) #
-      otherimplies # [FeatureD32, !cast<SubtargetFeature>(NAME # "_D16_SP")]>;
-  def _D16: SubtargetFeature<
-    name#"d16", query#"D16", "true",
-    description#" with only 16 d-registers",
-    !foreach(v, prev, !cast<SubtargetFeature>(v # "_D16")) #
-      vfp2prev #
-      otherimplies # [FeatureFP64, !cast<SubtargetFeature>(NAME # "_D16_SP")]>;
-  def "": SubtargetFeature<
-    name, query, "true", description,
-    prev # otherimplies # [
-        !cast<SubtargetFeature>(NAME # "_D16"),
-        !cast<SubtargetFeature>(NAME # "_SP")]>;
-}
-
-def FeatureVFP2_SP        : SubtargetFeature<"vfp2sp", "HasVFPv2SP", "true",
-                                             "Enable VFP2 instructions with "
-                                             "no double precision",
-                                             [FeatureFPRegs]>;
-
-def FeatureVFP2           : SubtargetFeature<"vfp2", "HasVFPv2", "true",
-                                             "Enable VFP2 instructions",
-                                             [FeatureFP64, FeatureVFP2_SP]>;
-
-defm FeatureVFP3: VFPver<"vfp3", "HasVFPv3", "Enable VFP3 instructions",
-                         [], [], [FeatureVFP2]>;
-
-def FeatureNEON           : SubtargetFeature<"neon", "HasNEON", "true",
-                                             "Enable NEON instructions",
-                                             [FeatureVFP3]>;
-
-// True if subtarget supports half-precision FP conversions.
-def FeatureFP16           : SubtargetFeature<"fp16", "HasFP16", "true",
-                                             "Enable half-precision "
-                                             "floating point">;
-
-defm FeatureVFP4: VFPver<"vfp4", "HasVFPv4", "Enable VFP4 instructions",
-                         [FeatureVFP3], [FeatureFP16]>;
-
-defm FeatureFPARMv8: VFPver<"fp-armv8", "HasFPARMv8", "Enable ARMv8 FP",
-                         [FeatureVFP4], []>;
-
-// True if subtarget supports half-precision FP operations.
-def FeatureFullFP16       : SubtargetFeature<"fullfp16", "HasFullFP16", "true",
-                                             "Enable full half-precision "
-                                             "floating point",
-                                             [FeatureFPARMv8_D16_SP, FeatureFPRegs16]>;
-
-// True if subtarget supports half-precision FP fml operations.
-def FeatureFP16FML        : SubtargetFeature<"fp16fml", "HasFP16FML", "true",
-                                             "Enable full half-precision "
-                                             "floating point fml instructions",
-                                             [FeatureFullFP16]>;
-
-// True if subtarget supports [su]div in Thumb mode.
-def FeatureHWDivThumb     : SubtargetFeature<"hwdiv",
-                                             "HasDivideInThumbMode", "true",
-                                             "Enable divide instructions in Thumb">;
-
-// True if subtarget supports [su]div in ARM mode.
-def FeatureHWDivARM       : SubtargetFeature<"hwdiv-arm",
-                                             "HasDivideInARMMode", "true",
-                                             "Enable divide instructions in ARM mode">;
-
-// Atomic Support
-
-// True if the subtarget supports DMB / DSB data barrier instructions.
-def FeatureDB             : SubtargetFeature<"db", "HasDataBarrier", "true",
-                                             "Has data barrier (dmb/dsb) instructions">;
-
-// True if the subtarget supports CLREX instructions.
-def FeatureV7Clrex        : SubtargetFeature<"v7clrex", "HasV7Clrex", "true",
-                                             "Has v7 clrex instruction">;
-
-// True if the subtarget supports DFB data barrier instruction.
-def FeatureDFB  : SubtargetFeature<"dfb", "HasFullDataBarrier", "true",
-                                   "Has full data barrier (dfb) instruction">;
-
-// True if the subtarget supports v8 atomics (LDA/LDAEX etc) instructions.
-def FeatureAcquireRelease : SubtargetFeature<"acquire-release",
-                                             "HasAcquireRelease", "true",
-                                             "Has v8 acquire/release (lda/ldaex "
-                                             " etc) instructions">;
-
-
-// True if floating point compare + branch is slow.
-def FeatureSlowFPBrcc     : SubtargetFeature<"slow-fp-brcc", "IsFPBrccSlow", "true",
-                                             "FP compare + branch is slow">;
-
-// True if the processor supports the Performance Monitor Extensions. These
-// include a generic cycle-counter as well as more fine-grained (often
-// implementation-specific) events.
-def FeaturePerfMon        : SubtargetFeature<"perfmon", "HasPerfMon", "true",
-                                             "Enable support for Performance "
-                                             "Monitor extensions">;
-
-
-// TrustZone Security Extensions
-
-// True if processor supports TrustZone security extensions.
-def FeatureTrustZone      : SubtargetFeature<"trustzone", "HasTrustZone", "true",
-                                             "Enable support for TrustZone "
-                                             "security extensions">;
-
-// True if processor supports ARMv8-M Security Extensions.
-def Feature8MSecExt       : SubtargetFeature<"8msecext", "Has8MSecExt", "true",
-                                             "Enable support for ARMv8-M "
-                                             "Security Extensions">;
-
-// True if processor supports SHA1 and SHA256.
-def FeatureSHA2           : SubtargetFeature<"sha2", "HasSHA2", "true",
-                                             "Enable SHA1 and SHA256 support", [FeatureNEON]>;
-
-def FeatureAES            : SubtargetFeature<"aes", "HasAES", "true",
-                                             "Enable AES support", [FeatureNEON]>;
-
-// True if processor supports Cryptography extensions.
-def FeatureCrypto         : SubtargetFeature<"crypto", "HasCrypto", "true",
-                                             "Enable support for "
-                                             "Cryptography extensions",
-                                             [FeatureNEON, FeatureSHA2, FeatureAES]>;
-
-// True if processor supports CRC instructions.
-def FeatureCRC            : SubtargetFeature<"crc", "HasCRC", "true",
-                                             "Enable support for CRC instructions">;
-
-// True if the ARMv8.2A dot product instructions are supported.
-def FeatureDotProd        : SubtargetFeature<"dotprod", "HasDotProd", "true",
-                                             "Enable support for dot product instructions",
-                                             [FeatureNEON]>;
-
-// True if the processor supports RAS extensions.
-// Not to be confused with FeatureHasRetAddrStack (return address stack).
-def FeatureRAS            : SubtargetFeature<"ras", "HasRAS", "true",
-                                             "Enable Reliability, Availability "
-                                             "and Serviceability extensions">;
-
-// Fast computation of non-negative address offsets.
-// True if processor does positive address offset computation faster.
-def FeatureFPAO           : SubtargetFeature<"fpao", "HasFPAO", "true",
-                                             "Enable fast computation of "
-                                             "positive address offsets">;
-
-// Fast execution of AES crypto operations.
-// True if processor executes back to back AES instruction pairs faster.
-def FeatureFuseAES        : SubtargetFeature<"fuse-aes", "HasFuseAES", "true",
-                                             "CPU fuses AES crypto operations">;
-
-// Fast execution of bottom and top halves of literal generation.
-// True if processor executes back to back bottom and top halves of literal generation faster.
-def FeatureFuseLiterals   : SubtargetFeature<"fuse-literals", "HasFuseLiterals", "true",
-                                             "CPU fuses literal generation operations">;
-
-// Choice of hardware register to use as the thread pointer, if any.
-def FeatureReadTpTPIDRURW :  SubtargetFeature<"read-tp-tpidrurw", "IsReadTPTPIDRURW", "true",
-                                      "Reading thread pointer from TPIDRURW register">;
-def FeatureReadTpTPIDRURO :  SubtargetFeature<"read-tp-tpidruro", "IsReadTPTPIDRURO", "true",
-                                      "Reading thread pointer from TPIDRURO register">;
-def FeatureReadTpTPIDRPRW :  SubtargetFeature<"read-tp-tpidrprw", "IsReadTPTPIDRPRW", "true",
-                                      "Reading thread pointer from TPIDRPRW register">;
-
-// Cyclone can zero VFP registers in 0 cycles.
-// True if the instructions "vmov.i32 d0, #0" and "vmov.i32 q0, #0" are
-// particularly effective at zeroing a VFP register.
-def FeatureZCZeroing      : SubtargetFeature<"zcz", "HasZeroCycleZeroing", "true",
-                                             "Has zero-cycle zeroing instructions">;
-
-// Whether it is profitable to unpredicate certain instructions during if-conversion.
-// True if if conversion may decide to leave some instructions unpredicated.
-def FeatureProfUnpredicate : SubtargetFeature<"prof-unpr",
-                                              "IsProfitableToUnpredicate", "true",
-                                              "Is profitable to unpredicate">;
-
-// Some targets (e.g. Swift) have microcoded VGETLNi32.
-// True if VMOV will be favored over VGETLNi32.
-def FeatureSlowVGETLNi32  : SubtargetFeature<"slow-vgetlni32",
-                                             "HasSlowVGETLNi32", "true",
-                                             "Has slow VGETLNi32 - prefer VMOV">;
-
-// Some targets (e.g. Swift) have microcoded VDUP32.
-// True if VMOV will be favored over VDUP.
-def FeatureSlowVDUP32     : SubtargetFeature<"slow-vdup32", "HasSlowVDUP32",
-                                             "true",
-                                             "Has slow VDUP32 - prefer VMOV">;
-
-// Some targets (e.g. Cortex-A9) prefer VMOVSR to VMOVDRR even when using NEON
-// for scalar FP, as this allows more effective execution domain optimization.
-// True if VMOVSR will be favored over VMOVDRR.
-def FeaturePreferVMOVSR   : SubtargetFeature<"prefer-vmovsr", "PreferVMOVSR",
-                                             "true", "Prefer VMOVSR">;
-
-// Swift has ISHST barriers compatible with Atomic Release semantics but weaker
-// than ISH.
-// True if ISHST barriers will be used for Release semantics.
-def FeaturePrefISHSTBarrier : SubtargetFeature<"prefer-ishst", "PreferISHSTBarriers",
-                                               "true", "Prefer ISHST barriers">;
-
-// Some targets (e.g. Cortex-A9) have muxed AGU and NEON/FPU.
-// True if the AGU and NEON/FPU units are multiplexed.
-def FeatureMuxedUnits     : SubtargetFeature<"muxed-units", "HasMuxedUnits",
-                                             "true",
-                                             "Has muxed AGU and NEON/FPU">;
-
-// Whether VLDM/VSTM starting with odd register number need more microops
-// than single VLDRS.
-// True if a VLDM/VSTM starting with an odd register number is considered to
-// take more microops than single VLDRS/VSTRS.
-def FeatureSlowOddRegister : SubtargetFeature<"slow-odd-reg", "HasSlowOddRegister",
-                                              "true", "VLDM/VSTM starting "
-                                              "with an odd register is slow">;
-
-// Some targets have a renaming dependency when loading into D subregisters.
-// True if loading into a D subregister will be penalized.
-def FeatureSlowLoadDSubreg : SubtargetFeature<"slow-load-D-subreg",
-                                              "HasSlowLoadDSubregister", "true",
-                                              "Loading into D subregs is slow">;
-
-// True if use a wider stride when allocating VFP registers.
-def FeatureUseWideStrideVFP : SubtargetFeature<"wide-stride-vfp",
-                                               "UseWideStrideVFP", "true",
-                                               "Use a wide stride when allocating VFP registers">;
-
-// Some targets (e.g. Cortex-A15) never want VMOVS to be widened to VMOVD.
-// True if VMOVS will never be widened to VMOVD.
-def FeatureDontWidenVMOVS : SubtargetFeature<"dont-widen-vmovs",
-                                             "DontWidenVMOVS", "true",
-                                             "Don't widen VMOVS to VMOVD">;
-
-// Some targets (e.g. Cortex-A15) prefer to avoid mixing operations on different
-// VFP register widths.
-// True if splat a register between VFP and NEON instructions.
-def FeatureSplatVFPToNeon : SubtargetFeature<"splat-vfp-neon",
-                                             "UseSplatVFPToNeon", "true",
-                                             "Splat register from VFP to NEON",
-                                             [FeatureDontWidenVMOVS]>;
-
-// Whether or not it is profitable to expand VFP/NEON MLA/MLS instructions.
-// True if run the MLx expansion pass.
-def FeatureExpandMLx      : SubtargetFeature<"expand-fp-mlx",
-                                             "ExpandMLx", "true",
-                                             "Expand VFP/NEON MLA/MLS instructions">;
-
-// Some targets have special RAW hazards for VFP/NEON VMLA/VMLS.
-// True if VFP/NEON VMLA/VMLS have special RAW hazards.
-def FeatureHasVMLxHazards : SubtargetFeature<"vmlx-hazards", "HasVMLxHazards",
-                                             "true", "Has VMLx hazards">;
-
-// Some targets (e.g. Cortex-A9) want to convert VMOVRS, VMOVSR and VMOVS from
-// VFP to NEON, as an execution domain optimization.
-// True if VMOVRS, VMOVSR and VMOVS will be converted from VFP to NEON.
-def FeatureNEONForFPMovs  : SubtargetFeature<"neon-fpmovs",
-                                             "UseNEONForFPMovs", "true",
-                                             "Convert VMOVSR, VMOVRS, "
-                                             "VMOVS to NEON">;
-
-// Some processors benefit from using NEON instructions for scalar
-// single-precision FP operations. This affects instruction selection and should
-// only be enabled if the handling of denormals is not important.
-// Use the method useNEONForSinglePrecisionFP() to determine if NEON should actually be used.
-def FeatureNEONForFP      : SubtargetFeature<"neonfp",
-                                             "HasNEONForFP",
-                                             "true",
-                                             "Use NEON for single precision FP">;
-
-// On some processors, VLDn instructions that access unaligned data take one
-// extra cycle. Take that into account when computing operand latencies.
-// True if VLDn instructions take an extra cycle for unaligned accesses.
-def FeatureCheckVLDnAlign : SubtargetFeature<"vldn-align", "CheckVLDnAccessAlignment",
-                                             "true",
-                                             "Check for VLDn unaligned access">;
-
-// Some processors have a nonpipelined VFP coprocessor.
-// True if VFP instructions are not pipelined.
-def FeatureNonpipelinedVFP : SubtargetFeature<"nonpipelined-vfp",
-                                              "NonpipelinedVFP", "true",
-                                              "VFP instructions are not pipelined">;
-
-// Some processors have FP multiply-accumulate instructions that don't
-// play nicely with other VFP / NEON instructions, and it's generally better
-// to just not use them.
-// If the VFP2 / NEON instructions are available, indicates
-// whether the FP VML[AS] instructions are slow (if so, don't use them).
-def FeatureHasSlowFPVMLx  : SubtargetFeature<"slowfpvmlx", "SlowFPVMLx", "true",
-                                             "Disable VFP / NEON MAC instructions">;
-
-// VFPv4 added VFMA instructions that can similarly be fast or slow.
-// If the VFP4 / NEON instructions are available, indicates
-// whether the FP VFM[AS] instructions are slow (if so, don't use them).
-def FeatureHasSlowFPVFMx  : SubtargetFeature<"slowfpvfmx", "SlowFPVFMx", "true",
-                                             "Disable VFP / NEON FMA instructions">;
-
-// Cortex-A8 / A9 Advanced SIMD has multiplier accumulator forwarding.
-/// True if NEON has special multiplier accumulator
-/// forwarding to allow mul + mla being issued back to back.
-def FeatureVMLxForwarding : SubtargetFeature<"vmlx-forwarding",
-                                             "HasVMLxForwarding", "true",
-                                             "Has multiplier accumulator forwarding">;
-
-// Disable 32-bit to 16-bit narrowing for experimentation.
-// True if codegen would prefer 32-bit Thumb instructions over 16-bit ones.
-def FeaturePref32BitThumb : SubtargetFeature<"32bit", "Prefers32BitThumb", "true",
-                                             "Prefer 32-bit Thumb instrs">;
-
-def FeaturePrefLoopAlign32 : SubtargetFeature<"loop-align", "PrefLoopLogAlignment","2",
-                                              "Prefer 32-bit alignment for loops">;
-
-def FeatureMVEVectorCostFactor1 : SubtargetFeature<"mve1beat", "MVEVectorCostFactor", "4",
-                        "Model MVE instructions as a 1 beat per tick architecture">;
-
-def FeatureMVEVectorCostFactor2 : SubtargetFeature<"mve2beat", "MVEVectorCostFactor", "2",
-                        "Model MVE instructions as a 2 beats per tick architecture">;
-
-def FeatureMVEVectorCostFactor4 : SubtargetFeature<"mve4beat", "MVEVectorCostFactor", "1",
-                        "Model MVE instructions as a 4 beats per tick architecture">;
-
-/// Some instructions update CPSR partially, which can add false dependency for
-/// out-of-order implementation, e.g. Cortex-A9, unless each individual bit is
-/// mapped to a separate physical register. Avoid partial CPSR update for these
-/// processors.
-/// True if codegen would avoid using instructions
-/// that partially update CPSR and add false dependency on the previous
-/// CPSR setting instruction.
-def FeatureAvoidPartialCPSR : SubtargetFeature<"avoid-partial-cpsr",
-                                               "AvoidCPSRPartialUpdate", "true",
-                                 "Avoid CPSR partial update for OOO execution">;
-
-/// Disable +1 predication cost for instructions updating CPSR.
-/// Enabled for Cortex-A57.
-/// True if disable +1 predication cost for instructions updating CPSR. Enabled for Cortex-A57.
-def FeatureCheapPredicableCPSR : SubtargetFeature<"cheap-predicable-cpsr",
-                                                  "CheapPredicableCPSRDef",
-                                                  "true",
-                  "Disable +1 predication cost for instructions updating CPSR">;
-
-// True if codegen should avoid using flag setting movs with shifter operand (i.e. asr, lsl, lsr).
-def FeatureAvoidMOVsShOp  : SubtargetFeature<"avoid-movs-shop",
-                                             "AvoidMOVsShifterOperand", "true",
-                                             "Avoid movs instructions with "
-                                             "shifter operand">;
-
-// Some processors perform return stack prediction. CodeGen should avoid issue
-// "normal" call instructions to callees which do not return.
-def FeatureHasRetAddrStack : SubtargetFeature<"ret-addr-stack",
-                                              "HasRetAddrStack", "true",
-                                              "Has return address stack">;
-
-// Some processors have no branch predictor, which changes the expected cost of
-// taking a branch which affects the choice of whether to use predicated
-// instructions.
-// True if the subtarget has a branch predictor. Having
-// a branch predictor or not changes the expected cost of taking a branch
-// which affects the choice of whether to use predicated instructions.
-def FeatureHasNoBranchPredictor : SubtargetFeature<"no-branch-predictor",
-                                                   "HasBranchPredictor", "false",
-                                                   "Has no branch predictor">;
-
-/// DSP extension.
-/// True if the subtarget supports the DSP (saturating arith and such) instructions.
-def FeatureDSP            : SubtargetFeature<"dsp", "HasDSP", "true",
-                                             "Supports DSP instructions in "
-                                             "ARM and/or Thumb2">;
-
-// True if the subtarget supports Multiprocessing extension (ARMv7 only).
-def FeatureMP             : SubtargetFeature<"mp", "HasMPExtension", "true",
-                                        "Supports Multiprocessing extension">;
-
-// Virtualization extension - requires HW divide (ARMv7-AR ARMARM - 4.4.8).
-def FeatureVirtualization : SubtargetFeature<"virtualization",
-                                             "HasVirtualization", "true",
-                                             "Supports Virtualization extension",
-                                             [FeatureHWDivThumb, FeatureHWDivARM]>;
-
-// Special TRAP encoding for NaCl, which looks like a TRAP in Thumb too.
-// See ARMInstrInfo.td for details.
-// True if NaCl TRAP instruction is generated instead of the regular TRAP.
-def FeatureNaClTrap       : SubtargetFeature<"nacl-trap", "UseNaClTrap", "true",
-                                             "NaCl trap">;
-
-// True if the subtarget disallows unaligned memory
-// accesses for some types.  For details, see
-// ARMTargetLowering::allowsMisalignedMemoryAccesses().
-def FeatureStrictAlign    : SubtargetFeature<"strict-align",
-                                             "StrictAlign", "true",
-                                             "Disallow all unaligned memory "
-                                             "access">;
-
-// Generate calls via indirect call instructions.
-def FeatureLongCalls      : SubtargetFeature<"long-calls", "GenLongCalls", "true",
-                                             "Generate calls via indirect call "
-                                             "instructions">;
-
-// Generate code that does not contain data access to code sections.
-def FeatureExecuteOnly    : SubtargetFeature<"execute-only",
-                                             "GenExecuteOnly", "true",
-                                             "Enable the generation of "
-                                             "execute only code.">;
-
-// True if R9 is not available as a general purpose register.
-def FeatureReserveR9      : SubtargetFeature<"reserve-r9", "ReserveR9", "true",
-                                             "Reserve R9, making it unavailable"
-                                             " as GPR">;
-
-// True if MOVT / MOVW pairs are not used for materialization of
-// 32-bit imms (including global addresses).
-def FeatureNoMovt         : SubtargetFeature<"no-movt", "NoMovt", "true",
-                                             "Don't use movt/movw pairs for "
-                                             "32-bit imms">;
-
-/// Implicitly convert an instruction to a different one if its immediates
-/// cannot be encoded. For example, ADD r0, r1, #FFFFFFFF -> SUB r0, r1, #1.
-def FeatureNoNegativeImmediates
-                          : SubtargetFeature<"no-neg-immediates",
-                                             "NegativeImmediates", "false",
-                                             "Convert immediates and instructions "
-                                             "to their negated or complemented "
-                                             "equivalent when the immediate does "
-                                             "not fit in the encoding.">;
-
-// Use the MachineScheduler for instruction scheduling for the subtarget.
-def FeatureUseMISched: SubtargetFeature<"use-misched", "UseMISched", "true",
-                                        "Use the MachineScheduler">;
-
-// Use the MachinePipeliner for instruction scheduling for the subtarget.
-def FeatureUseMIPipeliner: SubtargetFeature<"use-mipipeliner", "UseMIPipeliner", "true",
-                                            "Use the MachinePipeliner">;
-
-// False if scheduling should happen again after register allocation.
-def FeatureNoPostRASched : SubtargetFeature<"disable-postra-scheduler",
-    "DisablePostRAScheduler", "true",
-    "Don't schedule again after register allocation">;
-
-// Armv8.5-A extensions
-
-// Has speculation barrier.
-def FeatureSB       : SubtargetFeature<"sb", "HasSB", "true",
-  "Enable v8.5a Speculation Barrier" >;
-
-// Armv8.6-A extensions
-
-// True if subtarget supports BFloat16 floating point operations.
-def FeatureBF16     : SubtargetFeature<"bf16", "HasBF16", "true",
-  "Enable support for BFloat16 instructions",  [FeatureNEON]>;
-
-// True if subtarget supports 8-bit integer matrix multiply.
-def FeatureMatMulInt8 : SubtargetFeature<"i8mm", "HasMatMulInt8",
-    "true", "Enable Matrix Multiply Int8 Extension", [FeatureNEON]>;
-
-// Armv8.1-M extensions
-
-// True if the processor supports the Low Overhead Branch extension.
-def FeatureLOB            : SubtargetFeature<"lob", "HasLOB", "true",
-                                             "Enable Low Overhead Branch "
-                                             "extensions">;
-
-// Mitigate against the cve-2021-35465 security vulnurability.
-def FeatureFixCMSE_CVE_2021_35465 : SubtargetFeature<"fix-cmse-cve-2021-35465",
-                                        "FixCMSE_CVE_2021_35465", "true",
-                                        "Mitigate against the cve-2021-35465 "
-                                        "security vulnurability">;
-
-def FeaturePACBTI         : SubtargetFeature<"pacbti", "HasPACBTI", "true",
-                                             "Enable Pointer Authentication and Branch "
-                                             "Target Identification">;
-
-/// Don't place a BTI instruction after return-twice constructs (setjmp).
-def FeatureNoBTIAtReturnTwice : SubtargetFeature<"no-bti-at-return-twice",
-                                                 "NoBTIAtReturnTwice", "true",
-                                                 "Don't place a BTI instruction "
-                                                 "after a return-twice">;
-
-// Armv8.9-A/Armv9.4-A 2022 Architecture Extensions
-def FeatureCLRBHB : SubtargetFeature<"clrbhb", "HasCLRBHB", "true",
-                                     "Enable Clear BHB instruction">;
-
-
-def FeatureFixCortexA57AES1742098 : SubtargetFeature<"fix-cortex-a57-aes-1742098",
-  "FixCortexA57AES1742098", "true",
-  "Work around Cortex-A57 Erratum 1742098 / Cortex-A72 Erratum 1655431 (AES)">;
-
-def FeatureAAPCSFrameChain : SubtargetFeature<"aapcs-frame-chain",
-                                              "CreateAAPCSFrameChain", "true",
-                                              "Create an AAPCS compliant frame chain">;
-
-def FeatureAAPCSFrameChainLeaf : SubtargetFeature<"aapcs-frame-chain-leaf",
-                                                  "CreateAAPCSFrameChainLeaf", "true",
-                                                  "Create an AAPCS compliant frame chain "
-                                                  "for leaf functions",
-                                                  [FeatureAAPCSFrameChain]>;
-
-// Assume that lock-free 32-bit atomics are available, even if the target
-// and operating system combination would not usually provide them. The user
-// is responsible for providing any necessary __sync implementations. Code
-// built with this feature is not ABI-compatible with code built without this
-// feature, if atomic variables are exposed across the ABI boundary.
-def FeatureAtomics32 : SubtargetFeature<
-    "atomics-32", "HasForced32BitAtomics", "true",
-    "Assume that lock-free 32-bit atomics are available">;
-
-//===----------------------------------------------------------------------===//
-// ARM architecture class
-//
-
-// A-series ISA
-def FeatureAClass : SubtargetFeature<"aclass", "ARMProcClass", "AClass",
-                                     "Is application profile ('A' series)">;
-
-// R-series ISA
-def FeatureRClass : SubtargetFeature<"rclass", "ARMProcClass", "RClass",
-                                     "Is realtime profile ('R' series)">;
-
-// M-series ISA
-def FeatureMClass : SubtargetFeature<"mclass", "ARMProcClass", "MClass",
-                                     "Is microcontroller profile ('M' series)">;
-
-// True if Thumb2 instructions are supported.
-def FeatureThumb2 : SubtargetFeature<"thumb2", "HasThumb2", "true",
-                                     "Enable Thumb2 instructions">;
-
-// True if subtarget does not support ARM mode execution.
-def FeatureNoARM  : SubtargetFeature<"noarm", "NoARM", "true",
-                                     "Does not support ARM mode execution">;
-
-//===----------------------------------------------------------------------===//
-// ARM ISAa.
-//
-// Specify whether target support specific ARM ISA variants.
-
-def HasV4TOps   : SubtargetFeature<"v4t", "HasV4TOps", "true",
-                                   "Support ARM v4T instructions">;
-
-def HasV5TOps   : SubtargetFeature<"v5t", "HasV5TOps", "true",
-                                   "Support ARM v5T instructions",
-                                   [HasV4TOps]>;
-
-def HasV5TEOps  : SubtargetFeature<"v5te", "HasV5TEOps", "true",
-                                   "Support ARM v5TE, v5TEj, and "
-                                   "v5TExp instructions",
-                                   [HasV5TOps]>;
-
-def HasV6Ops    : SubtargetFeature<"v6", "HasV6Ops", "true",
-                                   "Support ARM v6 instructions",
-                                   [HasV5TEOps]>;
-
-def HasV6MOps   : SubtargetFeature<"v6m", "HasV6MOps", "true",
-                                   "Support ARM v6M instructions",
-                                   [HasV6Ops]>;
-
-def HasV8MBaselineOps : SubtargetFeature<"v8m", "HasV8MBaselineOps", "true",
-                                         "Support ARM v8M Baseline instructions",
-                                         [HasV6MOps]>;
-
-def HasV6KOps   : SubtargetFeature<"v6k", "HasV6KOps", "true",
-                                   "Support ARM v6k instructions",
-                                   [HasV6Ops]>;
-
-def HasV6T2Ops  : SubtargetFeature<"v6t2", "HasV6T2Ops", "true",
-                                   "Support ARM v6t2 instructions",
-                                   [HasV8MBaselineOps, HasV6KOps, FeatureThumb2]>;
-
-def HasV7Ops    : SubtargetFeature<"v7", "HasV7Ops", "true",
-                                   "Support ARM v7 instructions",
-                                   [HasV6T2Ops, FeatureV7Clrex]>;
-
-def HasV8MMainlineOps :
-                  SubtargetFeature<"v8m.main", "HasV8MMainlineOps", "true",
-                                   "Support ARM v8M Mainline instructions",
-                                   [HasV7Ops]>;
-
-def HasV8Ops    : SubtargetFeature<"v8", "HasV8Ops", "true",
-                                   "Support ARM v8 instructions",
-                                   [HasV7Ops, FeaturePerfMon, FeatureAcquireRelease]>;
-
-def HasV8_1aOps : SubtargetFeature<"v8.1a", "HasV8_1aOps", "true",
-                                   "Support ARM v8.1a instructions",
-                                   [HasV8Ops]>;
-
-def HasV8_2aOps : SubtargetFeature<"v8.2a", "HasV8_2aOps", "true",
-                                   "Support ARM v8.2a instructions",
-                                   [HasV8_1aOps]>;
-
-def HasV8_3aOps   : SubtargetFeature<"v8.3a", "HasV8_3aOps", "true",
-                                   "Support ARM v8.3a instructions",
-                                   [HasV8_2aOps]>;
-
-def HasV8_4aOps   : SubtargetFeature<"v8.4a", "HasV8_4aOps", "true",
-                                   "Support ARM v8.4a instructions",
-                                   [HasV8_3aOps, FeatureDotProd]>;
-
-def HasV8_5aOps   : SubtargetFeature<"v8.5a", "HasV8_5aOps", "true",
-                                   "Support ARM v8.5a instructions",
-                                   [HasV8_4aOps, FeatureSB]>;
-
-def HasV8_6aOps   : SubtargetFeature<"v8.6a", "HasV8_6aOps", "true",
-                                   "Support ARM v8.6a instructions",
-                                   [HasV8_5aOps, FeatureBF16,
-                                    FeatureMatMulInt8]>;
-
-def HasV8_7aOps   : SubtargetFeature<"v8.7a", "HasV8_7aOps", "true",
-                                   "Support ARM v8.7a instructions",
-                                   [HasV8_6aOps]>;
-
-def HasV8_8aOps   : SubtargetFeature<"v8.8a", "HasV8_8aOps", "true",
-                                   "Support ARM v8.8a instructions",
-                                   [HasV8_7aOps]>;
-
-def HasV8_9aOps   : SubtargetFeature<"v8.9a", "HasV8_9aOps", "true",
-                                   "Support ARM v8.9a instructions",
-                                   [HasV8_8aOps, FeatureCLRBHB]>;
-
-def HasV9_0aOps   : SubtargetFeature<"v9a", "HasV9_0aOps", "true",
-                                   "Support ARM v9a instructions",
-                                   [HasV8_5aOps]>;
-
-def HasV9_1aOps   : SubtargetFeature<"v9.1a", "HasV9_1aOps", "true",
-                                   "Support ARM v9.1a instructions",
-                                   [HasV8_6aOps, HasV9_0aOps]>;
-
-def HasV9_2aOps   : SubtargetFeature<"v9.2a", "HasV9_2aOps", "true",
-                                   "Support ARM v9.2a instructions",
-                                   [HasV8_7aOps, HasV9_1aOps]>;
-
-def HasV9_3aOps   : SubtargetFeature<"v9.3a", "HasV9_3aOps", "true",
-                                   "Support ARM v9.3a instructions",
-                                   [HasV8_8aOps, HasV9_2aOps]>;
-
-def HasV9_4aOps   : SubtargetFeature<"v9.4a", "HasV9_4aOps", "true",
-                                   "Support ARM v9.4a instructions",
-                                   [HasV8_9aOps, HasV9_3aOps]>;
-
-// Armv9.5-A is a v9-only architecture. From v9.5-A onwards there's no mapping
-// to an equivalent v8.x version.
-def HasV9_5aOps   : SubtargetFeature<"v9.5a", "HasV9_5aOps", "true",
-                                   "Support ARM v9.5a instructions",
-                                   [HasV9_4aOps]>;
-
-def HasV8_1MMainlineOps : SubtargetFeature<
-               "v8.1m.main", "HasV8_1MMainlineOps", "true",
-               "Support ARM v8-1M Mainline instructions",
-               [HasV8MMainlineOps]>;
-def HasMVEIntegerOps : SubtargetFeature<
-               "mve", "HasMVEIntegerOps", "true",
-               "Support M-Class Vector Extension with integer ops",
-               [HasV8_1MMainlineOps, FeatureDSP, FeatureFPRegs16, FeatureFPRegs64]>;
-def HasMVEFloatOps : SubtargetFeature<
-               "mve.fp", "HasMVEFloatOps", "true",
-               "Support M-Class Vector Extension with integer and floating ops",
-               [HasMVEIntegerOps, FeatureFPARMv8_D16_SP, FeatureFullFP16]>;
-
-def HasCDEOps : SubtargetFeature<"cde", "HasCDEOps", "true",
-                                 "Support CDE instructions",
-                                 [HasV8MMainlineOps]>;
-
-foreach i = {0-7} in
-    def FeatureCoprocCDE#i : SubtargetFeature<"cdecp"#i,
-                                              "CoprocCDE["#i#"]", "true",
-                                              "Coprocessor "#i#" ISA is CDEv1",
-                                              [HasCDEOps]>;
-
-//===----------------------------------------------------------------------===//
-// Control codegen mitigation against Straight Line Speculation vulnerability.
-//===----------------------------------------------------------------------===//
-
-/// Harden against Straight Line Speculation for Returns and Indirect Branches.
-def FeatureHardenSlsRetBr : SubtargetFeature<"harden-sls-retbr",
-  "HardenSlsRetBr", "true",
-  "Harden against straight line speculation across RETurn and BranchRegister "
-  "instructions">;
-/// Harden against Straight Line Speculation for indirect calls.
-def FeatureHardenSlsBlr : SubtargetFeature<"harden-sls-blr",
-  "HardenSlsBlr", "true",
-  "Harden against straight line speculation across indirect calls">;
-/// Generate thunk code for SLS mitigation in the normal text section.
-def FeatureHardenSlsNoComdat : SubtargetFeature<"harden-sls-nocomdat",
-  "HardenSlsNoComdat", "true",
-  "Generate thunk code for SLS mitigation in the normal text section">;
-
-//===----------------------------------------------------------------------===//
-// Endianness of instruction encodings in memory.
-//
-// In the current Arm architecture, this is usually little-endian regardless of
-// data endianness. But before Armv7 it was typical for instruction endianness
-// to match data endianness, so that a big-endian system was consistently big-
-// endian. And Armv7-R can be configured to use big-endian instructions.
-//
-// Additionally, even when targeting Armv7-A, big-endian instructions can be
-// found in relocatable object files, because the Arm ABI specifies that the
-// linker byte-reverses them depending on the target architecture.
-//
-// So we have a feature here to indicate that instructions are stored big-
-// endian, which you can set when instantiating an MCDisassembler.
-def ModeBigEndianInstructions : SubtargetFeature<"big-endian-instructions",
-    "BigEndianInstructions", "true",
-     "Expect instructions to be stored big-endian.">;
-
-//===----------------------------------------------------------------------===//
-// ARM Processor subtarget features.
-//
-
-def ProcA5      : SubtargetFeature<"a5", "ARMProcFamily", "CortexA5",
-                                   "Cortex-A5 ARM processors", []>;
-def ProcA7      : SubtargetFeature<"a7", "ARMProcFamily", "CortexA7",
-                                   "Cortex-A7 ARM processors", []>;
-def ProcA8      : SubtargetFeature<"a8", "ARMProcFamily", "CortexA8",
-                                   "Cortex-A8 ARM processors", []>;
-def ProcA9      : SubtargetFeature<"a9", "ARMProcFamily", "CortexA9",
-                                   "Cortex-A9 ARM processors", []>;
-def ProcA12     : SubtargetFeature<"a12", "ARMProcFamily", "CortexA12",
-                                   "Cortex-A12 ARM processors", []>;
-def ProcA15     : SubtargetFeature<"a15", "ARMProcFamily", "CortexA15",
-                                   "Cortex-A15 ARM processors", []>;
-def ProcA17     : SubtargetFeature<"a17", "ARMProcFamily", "CortexA17",
-                                   "Cortex-A17 ARM processors", []>;
-def ProcA32     : SubtargetFeature<"a32", "ARMProcFamily", "CortexA32",
-                                   "Cortex-A32 ARM processors", []>;
-def ProcA35     : SubtargetFeature<"a35", "ARMProcFamily", "CortexA35",
-                                   "Cortex-A35 ARM processors", []>;
-def ProcA53     : SubtargetFeature<"a53", "ARMProcFamily", "CortexA53",
-                                   "Cortex-A53 ARM processors", []>;
-def ProcA55     : SubtargetFeature<"a55", "ARMProcFamily", "CortexA55",
-                                   "Cortex-A55 ARM processors", []>;
-def ProcA57     : SubtargetFeature<"a57", "ARMProcFamily", "CortexA57",
-                                   "Cortex-A57 ARM processors", []>;
-def ProcA72     : SubtargetFeature<"a72", "ARMProcFamily", "CortexA72",
-                                   "Cortex-A72 ARM processors", []>;
-def ProcA73     : SubtargetFeature<"a73", "ARMProcFamily", "CortexA73",
-                                   "Cortex-A73 ARM processors", []>;
-def ProcA75     : SubtargetFeature<"a75", "ARMProcFamily", "CortexA75",
-                                   "Cortex-A75 ARM processors", []>;
-def ProcA76     : SubtargetFeature<"a76", "ARMProcFamily", "CortexA76",
-                                   "Cortex-A76 ARM processors", []>;
-def ProcA77     : SubtargetFeature<"a77", "ARMProcFamily", "CortexA77",
-                                   "Cortex-A77 ARM processors", []>;
-def ProcA78     : SubtargetFeature<"cortex-a78", "ARMProcFamily", "CortexA78",
-                                   "Cortex-A78 ARM processors", []>;
-def ProcA78AE   : SubtargetFeature<"cortex-a78ae", "ARMProcFamily", "CortexA78AE",
-                                   "Cortex-A78AE ARM processors", []>;
-def ProcA78C    : SubtargetFeature<"a78c", "ARMProcFamily", "CortexA78C",
-                                   "Cortex-A78C ARM processors", []>;
-def ProcA710    : SubtargetFeature<"cortex-a710", "ARMProcFamily",
-                                   "CortexA710", "Cortex-A710 ARM processors", []>;
-def ProcX1      : SubtargetFeature<"cortex-x1", "ARMProcFamily", "CortexX1",
-                                   "Cortex-X1 ARM processors", []>;
-def ProcX1C     : SubtargetFeature<"cortex-x1c", "ARMProcFamily", "CortexX1C",
-                                   "Cortex-X1C ARM processors", []>;
-
-def ProcV1      : SubtargetFeature<"neoverse-v1", "ARMProcFamily",
-                                   "NeoverseV1", "Neoverse-V1 ARM processors", []>;
-
-def ProcKrait   : SubtargetFeature<"krait", "ARMProcFamily", "Krait",
-                                   "Qualcomm Krait processors", []>;
-def ProcKryo    : SubtargetFeature<"kryo", "ARMProcFamily", "Kryo",
-                                   "Qualcomm Kryo processors", []>;
-def ProcSwift   : SubtargetFeature<"swift", "ARMProcFamily", "Swift",
-                                   "Swift ARM processors", []>;
-
-def ProcExynos  : SubtargetFeature<"exynos", "ARMProcFamily", "Exynos",
-                                   "Samsung Exynos processors",
-                                   [FeatureZCZeroing,
-                                    FeatureUseWideStrideVFP,
-                                    FeatureSplatVFPToNeon,
-                                    FeatureSlowVGETLNi32,
-                                    FeatureSlowVDUP32,
-                                    FeatureSlowFPBrcc,
-                                    FeatureProfUnpredicate,
-                                    FeatureHWDivThumb,
-                                    FeatureHWDivARM,
-                                    FeatureHasSlowFPVMLx,
-                                    FeatureHasSlowFPVFMx,
-                                    FeatureHasRetAddrStack,
-                                    FeatureFuseLiterals,
-                                    FeatureFuseAES,
-                                    FeatureExpandMLx,
-                                    FeatureCrypto,
-                                    FeatureCRC]>;
-
-def ProcR4      : SubtargetFeature<"r4", "ARMProcFamily", "CortexR4",
-                                   "Cortex-R4 ARM processors", []>;
-def ProcR5      : SubtargetFeature<"r5", "ARMProcFamily", "CortexR5",
-                                   "Cortex-R5 ARM processors", []>;
-def ProcR7      : SubtargetFeature<"r7", "ARMProcFamily", "CortexR7",
-                                   "Cortex-R7 ARM processors", []>;
-def ProcR52     : SubtargetFeature<"r52", "ARMProcFamily", "CortexR52",
-                                   "Cortex-R52 ARM processors", []>;
-
-def ProcM3      : SubtargetFeature<"m3", "ARMProcFamily", "CortexM3",
-                                   "Cortex-M3 ARM processors", []>;
-def ProcM7      : SubtargetFeature<"m7", "ARMProcFamily", "CortexM7",
-                                   "Cortex-M7 ARM processors", []>;
-
-//===----------------------------------------------------------------------===//
-// ARM Helper classes.
-//
-
-class Architecture<string fname, string aname, list<SubtargetFeature> features>
-  : SubtargetFeature<fname, "ARMArch", aname,
-                     !strconcat(aname, " architecture"), features>;
-
-class ProcNoItin<string Name, list<SubtargetFeature> Features>
-  : Processor<Name, NoItineraries, Features>;
-
-
-//===----------------------------------------------------------------------===//
-// ARM architectures
-//
-
-def ARMv4     : Architecture<"armv4",     "ARMv4",    []>;
-
-def ARMv4t    : Architecture<"armv4t",    "ARMv4t",   [HasV4TOps]>;
-
-def ARMv5t    : Architecture<"armv5t",    "ARMv5t",   [HasV5TOps]>;
-
-def ARMv5te   : Architecture<"armv5te",   "ARMv5te",  [HasV5TEOps]>;
-
-def ARMv5tej  : Architecture<"armv5tej",  "ARMv5tej", [HasV5TEOps]>;
-
-def ARMv6     : Architecture<"armv6",     "ARMv6",    [HasV6Ops,
-                                                       FeatureDSP]>;
-
-def ARMv6t2   : Architecture<"armv6t2",   "ARMv6t2",  [HasV6T2Ops,
-                                                       FeatureDSP]>;
-
-def ARMv6k    : Architecture<"armv6k",    "ARMv6k",   [HasV6KOps]>;
-
-def ARMv6kz   : Architecture<"armv6kz",   "ARMv6kz",  [HasV6KOps,
-                                                       FeatureTrustZone]>;
-
-def ARMv6m    : Architecture<"armv6-m",   "ARMv6m",   [HasV6MOps,
-                                                       FeatureNoARM,
-                                                       ModeThumb,
-                                                       FeatureDB,
-                                                       FeatureMClass,
-                                                       FeatureStrictAlign]>;
-
-def ARMv6sm   : Architecture<"armv6s-m",  "ARMv6sm",  [HasV6MOps,
-                                                       FeatureNoARM,
-                                                       ModeThumb,
-                                                       FeatureDB,
-                                                       FeatureMClass,
-                                                       FeatureStrictAlign]>;
-
-def ARMv7a    : Architecture<"armv7-a",   "ARMv7a",   [HasV7Ops,
-                                                       FeatureNEON,
-                                                       FeatureDB,
-                                                       FeatureDSP,
-                                                       FeatureAClass,
-                                                       FeaturePerfMon]>;
-
-def ARMv7ve   : Architecture<"armv7ve",   "ARMv7ve",  [HasV7Ops,
-                                                       FeatureNEON,
-                                                       FeatureDB,
-                                                       FeatureDSP,
-                                                       FeatureTrustZone,
-                                                       FeatureMP,
-                                                       FeatureVirtualization,
-                                                       FeatureAClass,
-                                                       FeaturePerfMon]>;
-
-def ARMv7r    : Architecture<"armv7-r",   "ARMv7r",   [HasV7Ops,
-                                                       FeatureDB,
-                                                       FeatureDSP,
-                                                       FeatureHWDivThumb,
-                                                       FeatureRClass,
-                                                       FeaturePerfMon]>;
-
-def ARMv7m    : Architecture<"armv7-m",   "ARMv7m",   [HasV7Ops,
-                                                       FeatureThumb2,
-                                                       FeatureNoARM,
-                                                       ModeThumb,
-                                                       FeatureDB,
-                                                       FeatureHWDivThumb,
-                                                       FeatureMClass]>;
-
-def ARMv7em   : Architecture<"armv7e-m",  "ARMv7em",  [HasV7Ops,
-                                                       FeatureThumb2,
-                                                       FeatureNoARM,
-                                                       ModeThumb,
-                                                       FeatureDB,
-                                                       FeatureHWDivThumb,
-                                                       FeatureMClass,
-                                                       FeatureDSP]>;
-
-def ARMv8a    : Architecture<"armv8-a",   "ARMv8a",   [HasV8Ops,
-                                                       FeatureAClass,
-                                                       FeatureDB,
-                                                       FeatureFPARMv8,
-                                                       FeatureNEON,
-                                                       FeatureDSP,
-                                                       FeatureTrustZone,
-                                                       FeatureMP,
-                                                       FeatureVirtualization,
-                                                       FeatureCrypto,
-                                                       FeatureCRC]>;
-
-def ARMv81a   : Architecture<"armv8.1-a", "ARMv81a",  [HasV8_1aOps,
-                                                       FeatureAClass,
-                                                       FeatureDB,
-                                                       FeatureFPARMv8,
-                                                       FeatureNEON,
-                                                       FeatureDSP,
-                                                       FeatureTrustZone,
-                                                       FeatureMP,
-                                                       FeatureVirtualization,
-                                                       FeatureCrypto,
-                                                       FeatureCRC]>;
-
-def ARMv82a   : Architecture<"armv8.2-a", "ARMv82a",  [HasV8_2aOps,
-                                                       FeatureAClass,
-                                                       FeatureDB,
-                                                       FeatureFPARMv8,
-                                                       FeatureNEON,
-                                                       FeatureDSP,
-                                                       FeatureTrustZone,
-                                                       FeatureMP,
-                                                       FeatureVirtualization,
-                                                       FeatureCrypto,
-                                                       FeatureCRC,
-                                                       FeatureRAS]>;
-
-def ARMv83a   : Architecture<"armv8.3-a", "ARMv83a",  [HasV8_3aOps,
-                                                       FeatureAClass,
-                                                       FeatureDB,
-                                                       FeatureFPARMv8,
-                                                       FeatureNEON,
-                                                       FeatureDSP,
-                                                       FeatureTrustZone,
-                                                       FeatureMP,
-                                                       FeatureVirtualization,
-                                                       FeatureCrypto,
-                                                       FeatureCRC,
-                                                       FeatureRAS]>;
-
-def ARMv84a   : Architecture<"armv8.4-a", "ARMv84a",  [HasV8_4aOps,
-                                                       FeatureAClass,
-                                                       FeatureDB,
-                                                       FeatureFPARMv8,
-                                                       FeatureNEON,
-                                                       FeatureDSP,
-                                                       FeatureTrustZone,
-                                                       FeatureMP,
-                                                       FeatureVirtualization,
-                                                       FeatureCrypto,
-                                                       FeatureCRC,
-                                                       FeatureRAS,
-                                                       FeatureDotProd]>;
-
-def ARMv85a   : Architecture<"armv8.5-a", "ARMv85a",  [HasV8_5aOps,
-                                                       FeatureAClass,
-                                                       FeatureDB,
-                                                       FeatureFPARMv8,
-                                                       FeatureNEON,
-                                                       FeatureDSP,
-                                                       FeatureTrustZone,
-                                                       FeatureMP,
-                                                       FeatureVirtualization,
-                                                       FeatureCrypto,
-                                                       FeatureCRC,
-                                                       FeatureRAS,
-                                                       FeatureDotProd]>;
-def ARMv86a   : Architecture<"armv8.6-a", "ARMv86a",  [HasV8_6aOps,
-                                                       FeatureAClass,
-                                                       FeatureDB,
-                                                       FeatureFPARMv8,
-                                                       FeatureNEON,
-                                                       FeatureDSP,
-                                                       FeatureTrustZone,
-                                                       FeatureMP,
-                                                       FeatureVirtualization,
-                                                       FeatureCrypto,
-                                                       FeatureCRC,
-                                                       FeatureRAS,
-                                                       FeatureDotProd]>;
-def ARMv87a   : Architecture<"armv8.7-a", "ARMv87a",  [HasV8_7aOps,
-                                                       FeatureAClass,
-                                                       FeatureDB,
-                                                       FeatureFPARMv8,
-                                                       FeatureNEON,
-                                                       FeatureDSP,
-                                                       FeatureTrustZone,
-                                                       FeatureMP,
-                                                       FeatureVirtualization,
-                                                       FeatureCrypto,
-                                                       FeatureCRC,
-                                                       FeatureRAS,
-                                                       FeatureDotProd]>;
-def ARMv88a   : Architecture<"armv8.8-a", "ARMv88a",  [HasV8_8aOps,
-                                                       FeatureAClass,
-                                                       FeatureDB,
-                                                       FeatureFPARMv8,
-                                                       FeatureNEON,
-                                                       FeatureDSP,
-                                                       FeatureTrustZone,
-                                                       FeatureMP,
-                                                       FeatureVirtualization,
-                                                       FeatureCrypto,
-                                                       FeatureCRC,
-                                                       FeatureRAS,
-                                                       FeatureDotProd]>;
-def ARMv89a   : Architecture<"armv8.9-a", "ARMv89a",  [HasV8_9aOps,
-                                                       FeatureAClass,
-                                                       FeatureDB,
-                                                       FeatureFPARMv8,
-                                                       FeatureNEON,
-                                                       FeatureDSP,
-                                                       FeatureTrustZone,
-                                                       FeatureMP,
-                                                       FeatureVirtualization,
-                                                       FeatureCrypto,
-                                                       FeatureCRC,
-                                                       FeatureRAS,
-                                                       FeatureDotProd]>;
-
-def ARMv9a   : Architecture<"armv9-a", "ARMv9a",       [HasV9_0aOps,
-                                                       FeatureAClass,
-                                                       FeatureDB,
-                                                       FeatureFPARMv8,
-                                                       FeatureNEON,
-                                                       FeatureDSP,
-                                                       FeatureTrustZone,
-                                                       FeatureMP,
-                                                       FeatureVirtualization,
-                                                       FeatureCRC,
-                                                       FeatureRAS,
-                                                       FeatureDotProd]>;
-def ARMv91a   : Architecture<"armv9.1-a", "ARMv91a",   [HasV9_1aOps,
-                                                       FeatureAClass,
-                                                       FeatureDB,
-                                                       FeatureFPARMv8,
-                                                       FeatureNEON,
-                                                       FeatureDSP,
-                                                       FeatureTrustZone,
-                                                       FeatureMP,
-                                                       FeatureVirtualization,
-                                                       FeatureCRC,
-                                                       FeatureRAS,
-                                                       FeatureDotProd]>;
-def ARMv92a   : Architecture<"armv9.2-a", "ARMv92a",  [HasV9_2aOps,
-                                                       FeatureAClass,
-                                                       FeatureDB,
-                                                       FeatureFPARMv8,
-                                                       FeatureNEON,
-                                                       FeatureDSP,
-                                                       FeatureTrustZone,
-                                                       FeatureMP,
-                                                       FeatureVirtualization,
-                                                       FeatureCRC,
-                                                       FeatureRAS,
-                                                       FeatureDotProd]>;
-def ARMv93a   : Architecture<"armv9.3-a", "ARMv93a",  [HasV9_3aOps,
-                                                       FeatureAClass,
-                                                       FeatureDB,
-                                                       FeatureFPARMv8,
-                                                       FeatureNEON,
-                                                       FeatureDSP,
-                                                       FeatureTrustZone,
-                                                       FeatureMP,
-                                                       FeatureVirtualization,
-                                                       FeatureCrypto,
-                                                       FeatureCRC,
-                                                       FeatureRAS,
-                                                       FeatureDotProd]>;
-def ARMv94a   : Architecture<"armv9.4-a", "ARMv94a",  [HasV9_4aOps,
-                                                       FeatureAClass,
-                                                       FeatureDB,
-                                                       FeatureFPARMv8,
-                                                       FeatureNEON,
-                                                       FeatureDSP,
-                                                       FeatureTrustZone,
-                                                       FeatureMP,
-                                                       FeatureVirtualization,
-                                                       FeatureCRC,
-                                                       FeatureRAS,
-                                                       FeatureDotProd]>;
-def ARMv95a   : Architecture<"armv9.5-a", "ARMv95a",  [HasV9_5aOps,
-                                                       FeatureAClass,
-                                                       FeatureDB,
-                                                       FeatureFPARMv8,
-                                                       FeatureNEON,
-                                                       FeatureDSP,
-                                                       FeatureTrustZone,
-                                                       FeatureMP,
-                                                       FeatureVirtualization,
-                                                       FeatureCRC,
-                                                       FeatureRAS,
-                                                       FeatureDotProd]>;
-
-def ARMv8r    : Architecture<"armv8-r",   "ARMv8r",   [HasV8Ops,
-                                                       FeatureRClass,
-                                                       FeatureDB,
-                                                       FeatureDFB,
-                                                       FeatureDSP,
-                                                       FeatureCRC,
-                                                       FeatureMP,
-                                                       FeatureVirtualization,
-                                                       FeatureFPARMv8,
-                                                       FeatureNEON]>;
-
-def ARMv8mBaseline : Architecture<"armv8-m.base", "ARMv8mBaseline",
-                                                      [HasV8MBaselineOps,
-                                                       FeatureNoARM,
-                                                       ModeThumb,
-                                                       FeatureDB,
-                                                       FeatureHWDivThumb,
-                                                       FeatureV7Clrex,
-                                                       Feature8MSecExt,
-                                                       FeatureAcquireRelease,
-                                                       FeatureMClass,
-                                                       FeatureStrictAlign]>;
-
-def ARMv8mMainline : Architecture<"armv8-m.main", "ARMv8mMainline",
-                                                      [HasV8MMainlineOps,
-                                                       FeatureNoARM,
-                                                       ModeThumb,
-                                                       FeatureDB,
-                                                       FeatureHWDivThumb,
-                                                       Feature8MSecExt,
-                                                       FeatureAcquireRelease,
-                                                       FeatureMClass]>;
-
-def ARMv81mMainline : Architecture<"armv8.1-m.main", "ARMv81mMainline",
-                                                      [HasV8_1MMainlineOps,
-                                                       FeatureNoARM,
-                                                       ModeThumb,
-                                                       FeatureDB,
-                                                       FeatureHWDivThumb,
-                                                       Feature8MSecExt,
-                                                       FeatureAcquireRelease,
-                                                       FeatureMClass,
-                                                       FeatureRAS,
-                                                       FeatureLOB]>;
-
-// Aliases
-def IWMMXT   : Architecture<"iwmmxt",      "ARMv5te",  [ARMv5te]>;
-def IWMMXT2  : Architecture<"iwmmxt2",     "ARMv5te",  [ARMv5te]>;
-def XScale   : Architecture<"xscale",      "ARMv5te",  [ARMv5te]>;
-def ARMv6j   : Architecture<"armv6j",      "ARMv7a",   [ARMv6]>;
-def ARMv7k   : Architecture<"armv7k",      "ARMv7a",   [ARMv7a]>;
-def ARMv7s   : Architecture<"armv7s",      "ARMv7a",   [ARMv7a]>;
+include "ARMFeatures.td"
+include "ARMArchitectures.td"
 
 //===----------------------------------------------------------------------===//
 // Register File Description
@@ -1249,485 +54,7 @@ include "ARMScheduleM55.td"
 include "ARMScheduleM7.td"
 include "ARMScheduleM85.td"
 
-//===----------------------------------------------------------------------===//
-// ARM processors
-//
-// Dummy CPU, used to target architectures
-def : ProcessorModel<"generic",     CortexA8Model,      []>;
-
-// FIXME: Several processors below are not using their own scheduler
-// model, but one of similar/previous processor. These should be fixed.
-
-def : ProcNoItin<"arm8",                                [ARMv4]>;
-def : ProcNoItin<"arm810",                              [ARMv4]>;
-def : ProcNoItin<"strongarm",                           [ARMv4]>;
-def : ProcNoItin<"strongarm110",                        [ARMv4]>;
-def : ProcNoItin<"strongarm1100",                       [ARMv4]>;
-def : ProcNoItin<"strongarm1110",                       [ARMv4]>;
-
-def : ProcNoItin<"arm7tdmi",                            [ARMv4t]>;
-def : ProcNoItin<"arm7tdmi-s",                          [ARMv4t]>;
-def : ProcNoItin<"arm710t",                             [ARMv4t]>;
-def : ProcNoItin<"arm720t",                             [ARMv4t]>;
-def : ProcNoItin<"arm9",                                [ARMv4t]>;
-def : ProcNoItin<"arm9tdmi",                            [ARMv4t]>;
-def : ProcNoItin<"arm920",                              [ARMv4t]>;
-def : ProcNoItin<"arm920t",                             [ARMv4t]>;
-def : ProcNoItin<"arm922t",                             [ARMv4t]>;
-def : ProcNoItin<"arm940t",                             [ARMv4t]>;
-def : ProcNoItin<"ep9312",                              [ARMv4t]>;
-
-def : ProcNoItin<"arm10tdmi",                           [ARMv5t]>;
-def : ProcNoItin<"arm1020t",                            [ARMv5t]>;
-
-def : ProcNoItin<"arm9e",                               [ARMv5te]>;
-def : ProcNoItin<"arm926ej-s",                          [ARMv5te]>;
-def : ProcNoItin<"arm946e-s",                           [ARMv5te]>;
-def : ProcNoItin<"arm966e-s",                           [ARMv5te]>;
-def : ProcNoItin<"arm968e-s",                           [ARMv5te]>;
-def : ProcNoItin<"arm10e",                              [ARMv5te]>;
-def : ProcNoItin<"arm1020e",                            [ARMv5te]>;
-def : ProcNoItin<"arm1022e",                            [ARMv5te]>;
-def : ProcNoItin<"xscale",                              [ARMv5te]>;
-def : ProcNoItin<"iwmmxt",                              [ARMv5te]>;
-
-def : Processor<"arm1136j-s",       ARMV6Itineraries,   [ARMv6]>;
-def : Processor<"arm1136jf-s",      ARMV6Itineraries,   [ARMv6,
-                                                         FeatureVFP2,
-                                                         FeatureHasSlowFPVMLx]>;
-
-def : Processor<"cortex-m0",        ARMV6Itineraries,   [ARMv6m,
-                                                         FeatureHasNoBranchPredictor]>;
-def : Processor<"cortex-m0plus",    ARMV6Itineraries,   [ARMv6m,
-                                                         FeatureHasNoBranchPredictor]>;
-def : Processor<"cortex-m1",        ARMV6Itineraries,   [ARMv6m,
-                                                         FeatureHasNoBranchPredictor]>;
-def : Processor<"sc000",            ARMV6Itineraries,   [ARMv6m,
-                                                         FeatureHasNoBranchPredictor]>;
-
-def : Processor<"arm1176jz-s",      ARMV6Itineraries,   [ARMv6kz]>;
-def : Processor<"arm1176jzf-s",     ARMV6Itineraries,   [ARMv6kz,
-                                                         FeatureVFP2,
-                                                         FeatureHasSlowFPVMLx]>;
-
-def : Processor<"mpcorenovfp",      ARMV6Itineraries,   [ARMv6k]>;
-def : Processor<"mpcore",           ARMV6Itineraries,   [ARMv6k,
-                                                         FeatureVFP2,
-                                                         FeatureHasSlowFPVMLx]>;
-
-def : Processor<"arm1156t2-s",      ARMV6Itineraries,   [ARMv6t2]>;
-def : Processor<"arm1156t2f-s",     ARMV6Itineraries,   [ARMv6t2,
-                                                         FeatureVFP2,
-                                                         FeatureHasSlowFPVMLx]>;
-
-def : ProcessorModel<"cortex-a5",   CortexA8Model,      [ARMv7a, ProcA5,
-                                                         FeatureHasRetAddrStack,
-                                                         FeatureTrustZone,
-                                                         FeatureSlowFPBrcc,
-                                                         FeatureHasSlowFPVMLx,
-                                                         FeatureHasSlowFPVFMx,
-                                                         FeatureVMLxForwarding,
-                                                         FeatureMP,
-                                                         FeatureVFP4]>;
-
-def : ProcessorModel<"cortex-a7",   CortexA8Model,      [ARMv7a, ProcA7,
-                                                         FeatureHasRetAddrStack,
-                                                         FeatureTrustZone,
-                                                         FeatureSlowFPBrcc,
-                                                         FeatureHasVMLxHazards,
-                                                         FeatureHasSlowFPVMLx,
-                                                         FeatureHasSlowFPVFMx,
-                                                         FeatureVMLxForwarding,
-                                                         FeatureMP,
-                                                         FeatureVFP4,
-                                                         FeatureVirtualization]>;
-
-def : ProcessorModel<"cortex-a8",   CortexA8Model,      [ARMv7a, ProcA8,
-                                                         FeatureHasRetAddrStack,
-                                                         FeatureNonpipelinedVFP,
-                                                         FeatureTrustZone,
-                                                         FeatureSlowFPBrcc,
-                                                         FeatureHasVMLxHazards,
-                                                         FeatureHasSlowFPVMLx,
-                                                         FeatureHasSlowFPVFMx,
-                                                         FeatureVMLxForwarding]>;
-
-def : ProcessorModel<"cortex-a9",   CortexA9Model,      [ARMv7a, ProcA9,
-                                                         FeatureHasRetAddrStack,
-                                                         FeatureTrustZone,
-                                                         FeatureHasVMLxHazards,
-                                                         FeatureVMLxForwarding,
-                                                         FeatureFP16,
-                                                         FeatureAvoidPartialCPSR,
-                                                         FeatureExpandMLx,
-                                                         FeaturePreferVMOVSR,
-                                                         FeatureMuxedUnits,
-                                                         FeatureNEONForFPMovs,
-                                                         FeatureCheckVLDnAlign,
-                                                         FeatureMP]>;
-
-def : ProcessorModel<"cortex-a12",  CortexA9Model,      [ARMv7a, ProcA12,
-                                                         FeatureHasRetAddrStack,
-                                                         FeatureTrustZone,
-                                                         FeatureVMLxForwarding,
-                                                         FeatureVFP4,
-                                                         FeatureAvoidPartialCPSR,
-                                                         FeatureVirtualization,
-                                                         FeatureMP]>;
-
-def : ProcessorModel<"cortex-a15",  CortexA9Model,      [ARMv7a, ProcA15,
-                                                         FeatureDontWidenVMOVS,
-                                                         FeatureSplatVFPToNeon,
-                                                         FeatureHasRetAddrStack,
-                                                         FeatureMuxedUnits,
-                                                         FeatureTrustZone,
-                                                         FeatureVFP4,
-                                                         FeatureMP,
-                                                         FeatureCheckVLDnAlign,
-                                                         FeatureAvoidPartialCPSR,
-                                                         FeatureVirtualization]>;
-
-def : ProcessorModel<"cortex-a17",  CortexA9Model,      [ARMv7a, ProcA17,
-                                                         FeatureHasRetAddrStack,
-                                                         FeatureTrustZone,
-                                                         FeatureMP,
-                                                         FeatureVMLxForwarding,
-                                                         FeatureVFP4,
-                                                         FeatureAvoidPartialCPSR,
-                                                         FeatureVirtualization]>;
-
-// FIXME: krait has currently the same features as A9 plus VFP4 and  HWDiv
-def : ProcessorModel<"krait",       CortexA9Model,      [ARMv7a, ProcKrait,
-                                                         FeatureHasRetAddrStack,
-                                                         FeatureMuxedUnits,
-                                                         FeatureCheckVLDnAlign,
-                                                         FeatureVMLxForwarding,
-                                                         FeatureFP16,
-                                                         FeatureAvoidPartialCPSR,
-                                                         FeatureVFP4,
-                                                         FeatureHWDivThumb,
-                                                         FeatureHWDivARM]>;
-
-def : ProcessorModel<"swift",       SwiftModel,         [ARMv7a, ProcSwift,
-                                                         FeatureHasRetAddrStack,
-                                                         FeatureNEONForFP,
-                                                         FeatureVFP4,
-                                                         FeatureUseWideStrideVFP,
-                                                         FeatureMP,
-                                                         FeatureHWDivThumb,
-                                                         FeatureHWDivARM,
-                                                         FeatureAvoidPartialCPSR,
-                                                         FeatureAvoidMOVsShOp,
-                                                         FeatureHasSlowFPVMLx,
-                                                         FeatureHasSlowFPVFMx,
-                                                         FeatureHasVMLxHazards,
-                                                         FeatureProfUnpredicate,
-                                                         FeaturePrefISHSTBarrier,
-                                                         FeatureSlowOddRegister,
-                                                         FeatureSlowLoadDSubreg,
-                                                         FeatureSlowVGETLNi32,
-                                                         FeatureSlowVDUP32,
-                                                         FeatureUseMISched,
-                                                         FeatureNoPostRASched]>;
-
-def : ProcessorModel<"cortex-r4",   CortexA8Model,      [ARMv7r, ProcR4,
-                                                         FeatureHasRetAddrStack,
-                                                         FeatureAvoidPartialCPSR]>;
-
-def : ProcessorModel<"cortex-r4f",  CortexA8Model,      [ARMv7r, ProcR4,
-                                                         FeatureHasRetAddrStack,
-                                                         FeatureSlowFPBrcc,
-                                                         FeatureHasSlowFPVMLx,
-                                                         FeatureHasSlowFPVFMx,
-                                                         FeatureVFP3_D16,
-                                                         FeatureAvoidPartialCPSR]>;
-
-def : ProcessorModel<"cortex-r5",   CortexA8Model,      [ARMv7r, ProcR5,
-                                                         FeatureHasRetAddrStack,
-                                                         FeatureVFP3_D16,
-                                                         FeatureSlowFPBrcc,
-                                                         FeatureHWDivARM,
-                                                         FeatureHasSlowFPVMLx,
-                                                         FeatureHasSlowFPVFMx,
-                                                         FeatureAvoidPartialCPSR]>;
-
-def : ProcessorModel<"cortex-r7",   CortexA8Model,      [ARMv7r, ProcR7,
-                                                         FeatureHasRetAddrStack,
-                                                         FeatureVFP3_D16,
-                                                         FeatureFP16,
-                                                         FeatureMP,
-                                                         FeatureSlowFPBrcc,
-                                                         FeatureHWDivARM,
-                                                         FeatureHasSlowFPVMLx,
-                                                         FeatureHasSlowFPVFMx,
-                                                         FeatureAvoidPartialCPSR]>;
-
-def : ProcessorModel<"cortex-r8",   CortexA8Model,      [ARMv7r,
-                                                         FeatureHasRetAddrStack,
-                                                         FeatureVFP3_D16,
-                                                         FeatureFP16,
-                                                         FeatureMP,
-                                                         FeatureSlowFPBrcc,
-                                                         FeatureHWDivARM,
-                                                         FeatureHasSlowFPVMLx,
-                                                         FeatureHasSlowFPVFMx,
-                                                         FeatureAvoidPartialCPSR]>;
-
-def : ProcessorModel<"cortex-m3",   CortexM4Model,      [ARMv7m,
-                                                         ProcM3,
-                                                         FeaturePrefLoopAlign32,
-                                                         FeatureUseMISched,
-                                                         FeatureHasNoBranchPredictor]>;
-
-def : ProcessorModel<"sc300",       CortexM4Model,      [ARMv7m,
-                                                         ProcM3,
-                                                         FeatureUseMISched,
-                                                         FeatureHasNoBranchPredictor]>;
-
-def : ProcessorModel<"cortex-m4", CortexM4Model,        [ARMv7em,
-                                                         FeatureVFP4_D16_SP,
-                                                         FeaturePrefLoopAlign32,
-                                                         FeatureHasSlowFPVMLx,
-                                                         FeatureHasSlowFPVFMx,
-                                                         FeatureUseMISched,
-                                                         FeatureHasNoBranchPredictor]>;
-
-def : ProcessorModel<"cortex-m7", CortexM7Model,        [ARMv7em,
-                                                         ProcM7,
-                                                         FeatureFPARMv8_D16,
-                                                         FeatureUseMIPipeliner,
-                                                         FeatureUseMISched]>;
-
-def : ProcNoItin<"cortex-m23",                          [ARMv8mBaseline,
-                                                         FeatureNoMovt,
-                                                         FeatureHasNoBranchPredictor]>;
-
-def : ProcessorModel<"cortex-m33", CortexM4Model,       [ARMv8mMainline,
-                                                         FeatureDSP,
-                                                         FeatureFPARMv8_D16_SP,
-                                                         FeaturePrefLoopAlign32,
-                                                         FeatureHasSlowFPVMLx,
-                                                         FeatureHasSlowFPVFMx,
-                                                         FeatureUseMISched,
-                                                         FeatureHasNoBranchPredictor,
-                                                         FeatureFixCMSE_CVE_2021_35465]>;
-
-def : ProcessorModel<"cortex-m35p", CortexM4Model,      [ARMv8mMainline,
-                                                         FeatureDSP,
-                                                         FeatureFPARMv8_D16_SP,
-                                                         FeaturePrefLoopAlign32,
-                                                         FeatureHasSlowFPVMLx,
-                                                         FeatureHasSlowFPVFMx,
-                                                         FeatureUseMISched,
-                                                         FeatureHasNoBranchPredictor,
-                                                         FeatureFixCMSE_CVE_2021_35465]>;
-
-def : ProcessorModel<"cortex-m55", CortexM55Model,      [ARMv81mMainline,
-                                                         FeatureDSP,
-                                                         FeatureFPARMv8_D16,
-                                                         FeatureUseMISched,
-                                                         FeatureHasNoBranchPredictor,
-                                                         FeaturePrefLoopAlign32,
-                                                         FeatureHasSlowFPVMLx,
-                                                         HasMVEFloatOps,
-                                                         FeatureFixCMSE_CVE_2021_35465]>;
-
-def : ProcessorModel<"cortex-m85", CortexM85Model,      [ARMv81mMainline,
-                                                         FeatureDSP,
-                                                         FeatureFPARMv8_D16,
-                                                         FeaturePACBTI,
-                                                         FeatureUseMISched,
-                                                         HasMVEFloatOps]>;
-
-def : ProcessorModel<"cortex-m52", CortexM55Model,      [ARMv81mMainline,
-                                                         FeatureDSP,
-                                                         FeatureFPARMv8_D16,
-                                                         FeatureHasNoBranchPredictor,
-                                                         FeaturePACBTI,
-                                                         FeatureUseMISched,
-                                                         FeaturePrefLoopAlign32,
-                                                         FeatureHasSlowFPVMLx,
-                                                         FeatureMVEVectorCostFactor1,
-                                                         HasMVEFloatOps]>;
-
-def : ProcNoItin<"cortex-a32",                           [ARMv8a,
-                                                         FeatureHWDivThumb,
-                                                         FeatureHWDivARM,
-                                                         FeatureCrypto,
-                                                         FeatureCRC]>;
-
-def : ProcNoItin<"cortex-a35",                          [ARMv8a, ProcA35,
-                                                         FeatureHWDivThumb,
-                                                         FeatureHWDivARM,
-                                                         FeatureCrypto,
-                                                         FeatureCRC]>;
-
-def : ProcNoItin<"cortex-a53",                          [ARMv8a, ProcA53,
-                                                         FeatureHWDivThumb,
-                                                         FeatureHWDivARM,
-                                                         FeatureCrypto,
-                                                         FeatureCRC,
-                                                         FeatureFPAO]>;
-
-def : ProcNoItin<"cortex-a55",                          [ARMv82a, ProcA55,
-                                                         FeatureHWDivThumb,
-                                                         FeatureHWDivARM,
-                                                         FeatureDotProd]>;
-
-def : ProcessorModel<"cortex-a57",  CortexA57Model,     [ARMv8a, ProcA57,
-                                                         FeatureHWDivThumb,
-                                                         FeatureHWDivARM,
-                                                         FeatureCrypto,
-                                                         FeatureCRC,
-                                                         FeatureFPAO,
-                                                         FeatureAvoidPartialCPSR,
-                                                         FeatureCheapPredicableCPSR,
-                                                         FeatureFixCortexA57AES1742098]>;
-
-def : ProcessorModel<"cortex-a72",  CortexA57Model,     [ARMv8a, ProcA72,
-                                                         FeatureHWDivThumb,
-                                                         FeatureHWDivARM,
-                                                         FeatureCrypto,
-                                                         FeatureCRC,
-                                                         FeatureFixCortexA57AES1742098]>;
-
-def : ProcNoItin<"cortex-a73",                          [ARMv8a, ProcA73,
-                                                         FeatureHWDivThumb,
-                                                         FeatureHWDivARM,
-                                                         FeatureCrypto,
-                                                         FeatureCRC]>;
-
-def : ProcNoItin<"cortex-a75",                          [ARMv82a, ProcA75,
-                                                         FeatureHWDivThumb,
-                                                         FeatureHWDivARM,
-                                                         FeatureDotProd]>;
-
-def : ProcNoItin<"cortex-a76",                          [ARMv82a, ProcA76,
-                                                         FeatureHWDivThumb,
-                                                         FeatureHWDivARM,
-                                                         FeatureCrypto,
-                                                         FeatureCRC,
-                                                         FeatureFullFP16,
-                                                         FeatureDotProd]>;
-
-def : ProcNoItin<"cortex-a76ae",                        [ARMv82a, ProcA76,
-                                                         FeatureHWDivThumb,
-                                                         FeatureHWDivARM,
-                                                         FeatureCrypto,
-                                                         FeatureCRC,
-                                                         FeatureFullFP16,
-                                                         FeatureDotProd]>;
-
-def : ProcNoItin<"cortex-a77",                          [ARMv82a, ProcA77,
-                                                         FeatureHWDivThumb,
-                                                         FeatureHWDivARM,
-                                                         FeatureCrypto,
-                                                         FeatureCRC,
-                                                         FeatureFullFP16,
-                                                         FeatureDotProd]>;
-
-def : ProcNoItin<"cortex-a78",                          [ARMv82a, ProcA78,
-                                                         FeatureHWDivThumb,
-                                                         FeatureHWDivARM,
-                                                         FeatureCrypto,
-                                                         FeatureCRC,
-                                                         FeatureFullFP16,
-                                                         FeatureDotProd]>;
-
-def : ProcNoItin<"cortex-a78ae",                        [ARMv82a, ProcA78AE,
-                                                         FeatureHWDivThumb,
-                                                         FeatureHWDivARM,
-                                                         FeatureCrypto,
-                                                         FeatureCRC,
-                                                         FeatureFullFP16,
-                                                         FeatureDotProd]>;
-
-def : ProcNoItin<"cortex-a78c",                         [ARMv82a, ProcA78C,
-                                                         FeatureHWDivThumb,
-                                                         FeatureHWDivARM,
-                                                         FeatureCrypto,
-                                                         FeatureCRC,
-                                                         FeatureDotProd,
-                                                         FeatureFullFP16]>;
-
-def : ProcNoItin<"cortex-a710",                         [ARMv9a, ProcA710,
-                                                         FeatureHWDivThumb,
-                                                         FeatureHWDivARM,
-                                                         FeatureFP16FML,
-                                                         FeatureBF16,
-                                                         FeatureMatMulInt8,
-                                                         FeatureSB]>;
-
-def : ProcNoItin<"cortex-x1",                           [ARMv82a, ProcX1,
-                                                         FeatureHWDivThumb,
-                                                         FeatureHWDivARM,
-                                                         FeatureCrypto,
-                                                         FeatureCRC,
-                                                         FeatureFullFP16,
-                                                         FeatureDotProd]>;
-
-def : ProcNoItin<"cortex-x1c",                          [ARMv82a, ProcX1C,
-                                                         FeatureHWDivThumb,
-                                                         FeatureHWDivARM,
-                                                         FeatureCrypto,
-                                                         FeatureCRC,
-                                                         FeatureFullFP16,
-                                                         FeatureDotProd]>;
-
-def : ProcNoItin<"neoverse-v1",                         [ARMv84a,
-                                                         FeatureHWDivThumb,
-                                                         FeatureHWDivARM,
-                                                         FeatureCrypto,
-                                                         FeatureCRC,
-                                                         FeatureFullFP16,
-                                                         FeatureBF16,
-                                                         FeatureMatMulInt8]>;
-
-def : ProcNoItin<"neoverse-n1",                         [ARMv82a,
-                                                         FeatureHWDivThumb,
-                                                         FeatureHWDivARM,
-                                                         FeatureCrypto,
-                                                         FeatureCRC,
-                                                         FeatureDotProd]>;
-
-def : ProcNoItin<"neoverse-n2",                         [ARMv9a,
-                                                         FeatureBF16,
-                                                         FeatureFP16FML,
-                                                         FeatureMatMulInt8]>;
-
-def : ProcessorModel<"cyclone",     SwiftModel,         [ARMv8a, ProcSwift,
-                                                         FeatureHasRetAddrStack,
-                                                         FeatureNEONForFP,
-                                                         FeatureVFP4,
-                                                         FeatureMP,
-                                                         FeatureHWDivThumb,
-                                                         FeatureHWDivARM,
-                                                         FeatureAvoidPartialCPSR,
-                                                         FeatureAvoidMOVsShOp,
-                                                         FeatureHasSlowFPVMLx,
-                                                         FeatureHasSlowFPVFMx,
-                                                         FeatureCrypto,
-                                                         FeatureUseMISched,
-                                                         FeatureZCZeroing,
-                                                         FeatureNoPostRASched]>;
-
-def : ProcNoItin<"exynos-m3",                           [ARMv8a, ProcExynos]>;
-def : ProcNoItin<"exynos-m4",                           [ARMv82a, ProcExynos,
-                                                         FeatureFullFP16,
-                                                         FeatureDotProd]>;
-def : ProcNoItin<"exynos-m5",                           [ARMv82a, ProcExynos,
-                                                         FeatureFullFP16,
-                                                         FeatureDotProd]>;
-
-def : ProcNoItin<"kryo",                                [ARMv8a, ProcKryo,
-                                                         FeatureHWDivThumb,
-                                                         FeatureHWDivARM,
-                                                         FeatureCrypto,
-                                                         FeatureCRC]>;
-
-def : ProcessorModel<"cortex-r52", CortexR52Model,      [ARMv8r, ProcR52,
-                                                         FeatureUseMISched,
-                                                         FeatureFPAO]>;
+include "ARMProcessors.td"
 
 //===----------------------------------------------------------------------===//
 // Declare the target which we are implementing
diff --git a/llvm/lib/Target/ARM/ARMArchitectures.td b/llvm/lib/Target/ARM/ARMArchitectures.td
new file mode 100644
index 000000000000..daf54f457b3b
--- /dev/null
+++ b/llvm/lib/Target/ARM/ARMArchitectures.td
@@ -0,0 +1,340 @@
+class Architecture<string fname, string aname, list<SubtargetFeature> features>
+  : SubtargetFeature<fname, "ARMArch", aname,
+                     !strconcat(aname, " architecture"), features>;
+
+//===----------------------------------------------------------------------===//
+// ARM architectures
+//
+
+def ARMv4     : Architecture<"armv4",     "ARMv4",    []>;
+
+def ARMv4t    : Architecture<"armv4t",    "ARMv4t",   [HasV4TOps]>;
+
+def ARMv5t    : Architecture<"armv5t",    "ARMv5t",   [HasV5TOps]>;
+
+def ARMv5te   : Architecture<"armv5te",   "ARMv5te",  [HasV5TEOps]>;
+
+def ARMv5tej  : Architecture<"armv5tej",  "ARMv5tej", [HasV5TEOps]>;
+
+def ARMv6     : Architecture<"armv6",     "ARMv6",    [HasV6Ops,
+                                                       FeatureDSP]>;
+
+def ARMv6t2   : Architecture<"armv6t2",   "ARMv6t2",  [HasV6T2Ops,
+                                                       FeatureDSP]>;
+
+def ARMv6k    : Architecture<"armv6k",    "ARMv6k",   [HasV6KOps]>;
+
+def ARMv6kz   : Architecture<"armv6kz",   "ARMv6kz",  [HasV6KOps,
+                                                       FeatureTrustZone]>;
+
+def ARMv6m    : Architecture<"armv6-m",   "ARMv6m",   [HasV6MOps,
+                                                       FeatureNoARM,
+                                                       ModeThumb,
+                                                       FeatureDB,
+                                                       FeatureMClass,
+                                                       FeatureStrictAlign]>;
+
+def ARMv6sm   : Architecture<"armv6s-m",  "ARMv6sm",  [HasV6MOps,
+                                                       FeatureNoARM,
+                                                       ModeThumb,
+                                                       FeatureDB,
+                                                       FeatureMClass,
+                                                       FeatureStrictAlign]>;
+
+def ARMv7a    : Architecture<"armv7-a",   "ARMv7a",   [HasV7Ops,
+                                                       FeatureNEON,
+                                                       FeatureDB,
+                                                       FeatureDSP,
+                                                       FeatureAClass,
+                                                       FeaturePerfMon]>;
+
+def ARMv7ve   : Architecture<"armv7ve",   "ARMv7ve",  [HasV7Ops,
+                                                       FeatureNEON,
+                                                       FeatureDB,
+                                                       FeatureDSP,
+                                                       FeatureTrustZone,
+                                                       FeatureMP,
+                                                       FeatureVirtualization,
+                                                       FeatureAClass,
+                                                       FeaturePerfMon]>;
+
+def ARMv7r    : Architecture<"armv7-r",   "ARMv7r",   [HasV7Ops,
+                                                       FeatureDB,
+                                                       FeatureDSP,
+                                                       FeatureHWDivThumb,
+                                                       FeatureRClass,
+                                                       FeaturePerfMon]>;
+
+def ARMv7m    : Architecture<"armv7-m",   "ARMv7m",   [HasV7Ops,
+                                                       FeatureThumb2,
+                                                       FeatureNoARM,
+                                                       ModeThumb,
+                                                       FeatureDB,
+                                                       FeatureHWDivThumb,
+                                                       FeatureMClass]>;
+
+def ARMv7em   : Architecture<"armv7e-m",  "ARMv7em",  [HasV7Ops,
+                                                       FeatureThumb2,
+                                                       FeatureNoARM,
+                                                       ModeThumb,
+                                                       FeatureDB,
+                                                       FeatureHWDivThumb,
+                                                       FeatureMClass,
+                                                       FeatureDSP]>;
+
+def ARMv8a    : Architecture<"armv8-a",   "ARMv8a",   [HasV8Ops,
+                                                       FeatureAClass,
+                                                       FeatureDB,
+                                                       FeatureFPARMv8,
+                                                       FeatureNEON,
+                                                       FeatureDSP,
+                                                       FeatureTrustZone,
+                                                       FeatureMP,
+                                                       FeatureVirtualization,
+                                                       FeatureCrypto,
+                                                       FeatureCRC]>;
+
+def ARMv81a   : Architecture<"armv8.1-a", "ARMv81a",  [HasV8_1aOps,
+                                                       FeatureAClass,
+                                                       FeatureDB,
+                                                       FeatureFPARMv8,
+                                                       FeatureNEON,
+                                                       FeatureDSP,
+                                                       FeatureTrustZone,
+                                                       FeatureMP,
+                                                       FeatureVirtualization,
+                                                       FeatureCrypto,
+                                                       FeatureCRC]>;
+
+def ARMv82a   : Architecture<"armv8.2-a", "ARMv82a",  [HasV8_2aOps,
+                                                       FeatureAClass,
+                                                       FeatureDB,
+                                                       FeatureFPARMv8,
+                                                       FeatureNEON,
+                                                       FeatureDSP,
+                                                       FeatureTrustZone,
+                                                       FeatureMP,
+                                                       FeatureVirtualization,
+                                                       FeatureCrypto,
+                                                       FeatureCRC,
+                                                       FeatureRAS]>;
+
+def ARMv83a   : Architecture<"armv8.3-a", "ARMv83a",  [HasV8_3aOps,
+                                                       FeatureAClass,
+                                                       FeatureDB,
+                                                       FeatureFPARMv8,
+                                                       FeatureNEON,
+                                                       FeatureDSP,
+                                                       FeatureTrustZone,
+                                                       FeatureMP,
+                                                       FeatureVirtualization,
+                                                       FeatureCrypto,
+                                                       FeatureCRC,
+                                                       FeatureRAS]>;
+
+def ARMv84a   : Architecture<"armv8.4-a", "ARMv84a",  [HasV8_4aOps,
+                                                       FeatureAClass,
+                                                       FeatureDB,
+                                                       FeatureFPARMv8,
+                                                       FeatureNEON,
+                                                       FeatureDSP,
+                                                       FeatureTrustZone,
+                                                       FeatureMP,
+                                                       FeatureVirtualization,
+                                                       FeatureCrypto,
+                                                       FeatureCRC,
+                                                       FeatureRAS,
+                                                       FeatureDotProd]>;
+
+def ARMv85a   : Architecture<"armv8.5-a", "ARMv85a",  [HasV8_5aOps,
+                                                       FeatureAClass,
+                                                       FeatureDB,
+                                                       FeatureFPARMv8,
+                                                       FeatureNEON,
+                                                       FeatureDSP,
+                                                       FeatureTrustZone,
+                                                       FeatureMP,
+                                                       FeatureVirtualization,
+                                                       FeatureCrypto,
+                                                       FeatureCRC,
+                                                       FeatureRAS,
+                                                       FeatureDotProd]>;
+def ARMv86a   : Architecture<"armv8.6-a", "ARMv86a",  [HasV8_6aOps,
+                                                       FeatureAClass,
+                                                       FeatureDB,
+                                                       FeatureFPARMv8,
+                                                       FeatureNEON,
+                                                       FeatureDSP,
+                                                       FeatureTrustZone,
+                                                       FeatureMP,
+                                                       FeatureVirtualization,
+                                                       FeatureCrypto,
+                                                       FeatureCRC,
+                                                       FeatureRAS,
+                                                       FeatureDotProd]>;
+def ARMv87a   : Architecture<"armv8.7-a", "ARMv87a",  [HasV8_7aOps,
+                                                       FeatureAClass,
+                                                       FeatureDB,
+                                                       FeatureFPARMv8,
+                                                       FeatureNEON,
+                                                       FeatureDSP,
+                                                       FeatureTrustZone,
+                                                       FeatureMP,
+                                                       FeatureVirtualization,
+                                                       FeatureCrypto,
+                                                       FeatureCRC,
+                                                       FeatureRAS,
+                                                       FeatureDotProd]>;
+def ARMv88a   : Architecture<"armv8.8-a", "ARMv88a",  [HasV8_8aOps,
+                                                       FeatureAClass,
+                                                       FeatureDB,
+                                                       FeatureFPARMv8,
+                                                       FeatureNEON,
+                                                       FeatureDSP,
+                                                       FeatureTrustZone,
+                                                       FeatureMP,
+                                                       FeatureVirtualization,
+                                                       FeatureCrypto,
+                                                       FeatureCRC,
+                                                       FeatureRAS,
+                                                       FeatureDotProd]>;
+def ARMv89a   : Architecture<"armv8.9-a", "ARMv89a",  [HasV8_9aOps,
+                                                       FeatureAClass,
+                                                       FeatureDB,
+                                                       FeatureFPARMv8,
+                                                       FeatureNEON,
+                                                       FeatureDSP,
+                                                       FeatureTrustZone,
+                                                       FeatureMP,
+                                                       FeatureVirtualization,
+                                                       FeatureCrypto,
+                                                       FeatureCRC,
+                                                       FeatureRAS,
+                                                       FeatureDotProd]>;
+
+def ARMv9a   : Architecture<"armv9-a", "ARMv9a",       [HasV9_0aOps,
+                                                       FeatureAClass,
+                                                       FeatureDB,
+                                                       FeatureFPARMv8,
+                                                       FeatureNEON,
+                                                       FeatureDSP,
+                                                       FeatureTrustZone,
+                                                       FeatureMP,
+                                                       FeatureVirtualization,
+                                                       FeatureCRC,
+                                                       FeatureRAS,
+                                                       FeatureDotProd]>;
+def ARMv91a   : Architecture<"armv9.1-a", "ARMv91a",   [HasV9_1aOps,
+                                                       FeatureAClass,
+                                                       FeatureDB,
+                                                       FeatureFPARMv8,
+                                                       FeatureNEON,
+                                                       FeatureDSP,
+                                                       FeatureTrustZone,
+                                                       FeatureMP,
+                                                       FeatureVirtualization,
+                                                       FeatureCRC,
+                                                       FeatureRAS,
+                                                       FeatureDotProd]>;
+def ARMv92a   : Architecture<"armv9.2-a", "ARMv92a",  [HasV9_2aOps,
+                                                       FeatureAClass,
+                                                       FeatureDB,
+                                                       FeatureFPARMv8,
+                                                       FeatureNEON,
+                                                       FeatureDSP,
+                                                       FeatureTrustZone,
+                                                       FeatureMP,
+                                                       FeatureVirtualization,
+                                                       FeatureCRC,
+                                                       FeatureRAS,
+                                                       FeatureDotProd]>;
+def ARMv93a   : Architecture<"armv9.3-a", "ARMv93a",  [HasV9_3aOps,
+                                                       FeatureAClass,
+                                                       FeatureDB,
+                                                       FeatureFPARMv8,
+                                                       FeatureNEON,
+                                                       FeatureDSP,
+                                                       FeatureTrustZone,
+                                                       FeatureMP,
+                                                       FeatureVirtualization,
+                                                       FeatureCrypto,
+                                                       FeatureCRC,
+                                                       FeatureRAS,
+                                                       FeatureDotProd]>;
+def ARMv94a   : Architecture<"armv9.4-a", "ARMv94a",  [HasV9_4aOps,
+                                                       FeatureAClass,
+                                                       FeatureDB,
+                                                       FeatureFPARMv8,
+                                                       FeatureNEON,
+                                                       FeatureDSP,
+                                                       FeatureTrustZone,
+                                                       FeatureMP,
+                                                       FeatureVirtualization,
+                                                       FeatureCRC,
+                                                       FeatureRAS,
+                                                       FeatureDotProd]>;
+def ARMv95a   : Architecture<"armv9.5-a", "ARMv95a",  [HasV9_5aOps,
+                                                       FeatureAClass,
+                                                       FeatureDB,
+                                                       FeatureFPARMv8,
+                                                       FeatureNEON,
+                                                       FeatureDSP,
+                                                       FeatureTrustZone,
+                                                       FeatureMP,
+                                                       FeatureVirtualization,
+                                                       FeatureCRC,
+                                                       FeatureRAS,
+                                                       FeatureDotProd]>;
+
+def ARMv8r    : Architecture<"armv8-r",   "ARMv8r",   [HasV8Ops,
+                                                       FeatureRClass,
+                                                       FeatureDB,
+                                                       FeatureDFB,
+                                                       FeatureDSP,
+                                                       FeatureCRC,
+                                                       FeatureMP,
+                                                       FeatureVirtualization,
+                                                       FeatureFPARMv8,
+                                                       FeatureNEON]>;
+
+def ARMv8mBaseline : Architecture<"armv8-m.base", "ARMv8mBaseline",
+                                                      [HasV8MBaselineOps,
+                                                       FeatureNoARM,
+                                                       ModeThumb,
+                                                       FeatureDB,
+                                                       FeatureHWDivThumb,
+                                                       FeatureV7Clrex,
+                                                       Feature8MSecExt,
+                                                       FeatureAcquireRelease,
+                                                       FeatureMClass,
+                                                       FeatureStrictAlign]>;
+
+def ARMv8mMainline : Architecture<"armv8-m.main", "ARMv8mMainline",
+                                                      [HasV8MMainlineOps,
+                                                       FeatureNoARM,
+                                                       ModeThumb,
+                                                       FeatureDB,
+                                                       FeatureHWDivThumb,
+                                                       Feature8MSecExt,
+                                                       FeatureAcquireRelease,
+                                                       FeatureMClass]>;
+
+def ARMv81mMainline : Architecture<"armv8.1-m.main", "ARMv81mMainline",
+                                                      [HasV8_1MMainlineOps,
+                                                       FeatureNoARM,
+                                                       ModeThumb,
+                                                       FeatureDB,
+                                                       FeatureHWDivThumb,
+                                                       Feature8MSecExt,
+                                                       FeatureAcquireRelease,
+                                                       FeatureMClass,
+                                                       FeatureRAS,
+                                                       FeatureLOB]>;
+
+// Aliases
+def IWMMXT   : Architecture<"iwmmxt",      "ARMv5te",  [ARMv5te]>;
+def IWMMXT2  : Architecture<"iwmmxt2",     "ARMv5te",  [ARMv5te]>;
+def XScale   : Architecture<"xscale",      "ARMv5te",  [ARMv5te]>;
+def ARMv6j   : Architecture<"armv6j",      "ARMv7a",   [ARMv6]>;
+def ARMv7k   : Architecture<"armv7k",      "ARMv7a",   [ARMv7a]>;
+def ARMv7s   : Architecture<"armv7s",      "ARMv7a",   [ARMv7a]>;
diff --git a/llvm/lib/Target/ARM/ARMBaseInstrInfo.cpp b/llvm/lib/Target/ARM/ARMBaseInstrInfo.cpp
index 5d0468948dfb..8f873bee484a 100644
--- a/llvm/lib/Target/ARM/ARMBaseInstrInfo.cpp
+++ b/llvm/lib/Target/ARM/ARMBaseInstrInfo.cpp
@@ -326,7 +326,7 @@ ARMBaseInstrInfo::convertToThreeAddress(MachineInstr &MI, LiveVariables *LV,
           for (unsigned j = 0; j < 2; ++j) {
             // Look at the two new MI's in reverse order.
             MachineInstr *NewMI = NewMIs[j];
-            if (!NewMI->readsRegister(Reg))
+            if (!NewMI->readsRegister(Reg, /*TRI=*/nullptr))
               continue;
             LV->addVirtualRegisterKilled(Reg, *NewMI);
             if (VI.removeKill(MI))
@@ -1732,7 +1732,7 @@ bool ARMBaseInstrInfo::expandPostRAPseudo(MachineInstr &MI) const {
 
   // Get rid of the old implicit-def of DstRegD.  Leave it if it defines a Q-reg
   // or some other super-register.
-  int ImpDefIdx = MI.findRegisterDefOperandIdx(DstRegD);
+  int ImpDefIdx = MI.findRegisterDefOperandIdx(DstRegD, /*TRI=*/nullptr);
   if (ImpDefIdx != -1)
     MI.removeOperand(ImpDefIdx);
 
@@ -2085,7 +2085,7 @@ bool ARMBaseInstrInfo::isSchedulingBoundary(const MachineInstr &MI,
   // Calls don't actually change the stack pointer, even if they have imp-defs.
   // No ARM calling conventions change the stack pointer. (X86 calling
   // conventions sometimes do).
-  if (!MI.isCall() && MI.definesRegister(ARM::SP))
+  if (!MI.isCall() && MI.definesRegister(ARM::SP, /*TRI=*/nullptr))
     return true;
 
   return false;
@@ -4137,7 +4137,7 @@ static const MachineInstr *getBundledDefMI(const TargetRegisterInfo *TRI,
 
   int Idx = -1;
   while (II->isInsideBundle()) {
-    Idx = II->findRegisterDefOperandIdx(Reg, false, true, TRI);
+    Idx = II->findRegisterDefOperandIdx(Reg, TRI, false, true);
     if (Idx != -1)
       break;
     --II;
@@ -4161,7 +4161,7 @@ static const MachineInstr *getBundledUseMI(const TargetRegisterInfo *TRI,
   // FIXME: This doesn't properly handle multiple uses.
   int Idx = -1;
   while (II != E && II->isInsideBundle()) {
-    Idx = II->findRegisterUseOperandIdx(Reg, false, TRI);
+    Idx = II->findRegisterUseOperandIdx(Reg, TRI, false);
     if (Idx != -1)
       break;
     if (II->getOpcode() != ARM::t2IT)
@@ -5361,7 +5361,7 @@ unsigned ARMBaseInstrInfo::getPartialRegUpdateClearance(
   case ARM::VMOVv2i32:
   case ARM::VMOVv2f32:
   case ARM::VMOVv1i64:
-    UseOp = MI.findRegisterUseOperandIdx(Reg, false, TRI);
+    UseOp = MI.findRegisterUseOperandIdx(Reg, TRI, false);
     break;
 
     // Explicitly reads the dependency.
@@ -6092,7 +6092,7 @@ ARMBaseInstrInfo::getOutliningCandidateInfo(
 bool ARMBaseInstrInfo::checkAndUpdateStackOffset(MachineInstr *MI,
                                                  int64_t Fixup,
                                                  bool Updt) const {
-  int SPIdx = MI->findRegisterUseOperandIdx(ARM::SP);
+  int SPIdx = MI->findRegisterUseOperandIdx(ARM::SP, /*TRI=*/nullptr);
   unsigned AddrMode = (MI->getDesc().TSFlags & ARMII::AddrModeMask);
   if (SPIdx < 0)
     // No SP operand
diff --git a/llvm/lib/Target/ARM/ARMConstantIslandPass.cpp b/llvm/lib/Target/ARM/ARMConstantIslandPass.cpp
index 7a3ba5870bc6..9579053943f9 100644
--- a/llvm/lib/Target/ARM/ARMConstantIslandPass.cpp
+++ b/llvm/lib/Target/ARM/ARMConstantIslandPass.cpp
@@ -1937,7 +1937,7 @@ bool ARMConstantIslands::optimizeThumb2Branches() {
 
     // If the conditional branch doesn't kill CPSR, then CPSR can be liveout
     // so this transformation is not safe.
-    if (!Br.MI->killsRegister(ARM::CPSR))
+    if (!Br.MI->killsRegister(ARM::CPSR, /*TRI=*/nullptr))
       return false;
 
     Register PredReg;
diff --git a/llvm/lib/Target/ARM/ARMFeatures.td b/llvm/lib/Target/ARM/ARMFeatures.td
new file mode 100644
index 000000000000..111c87838291
--- /dev/null
+++ b/llvm/lib/Target/ARM/ARMFeatures.td
@@ -0,0 +1,753 @@
+//===----------------------------------------------------------------------===//
+// ARM Subtarget state.
+//
+
+// True if compiling for Thumb, false for ARM.
+def ModeThumb             : SubtargetFeature<"thumb-mode", "IsThumb",
+                                             "true", "Thumb mode">;
+
+// True if we're using software floating point features.
+def ModeSoftFloat         : SubtargetFeature<"soft-float","UseSoftFloat",
+                                             "true", "Use software floating "
+                                             "point features.">;
+
+//===----------------------------------------------------------------------===//
+// ARM Subtarget features.
+//
+
+// Floating Point, HW Division and Neon Support
+
+// FP loads/stores/moves, shared between VFP and MVE (even in the integer-only
+// version).
+def FeatureFPRegs         : SubtargetFeature<"fpregs", "HasFPRegs", "true",
+                                             "Enable FP registers">;
+
+// 16-bit FP loads/stores/moves, shared between VFP (with the v8.2A FP16
+// extension) and MVE (even in the integer-only version).
+def FeatureFPRegs16       : SubtargetFeature<"fpregs16", "HasFPRegs16", "true",
+                                             "Enable 16-bit FP registers",
+                                             [FeatureFPRegs]>;
+
+def FeatureFPRegs64       : SubtargetFeature<"fpregs64", "HasFPRegs64", "true",
+                                             "Enable 64-bit FP registers",
+                                             [FeatureFPRegs]>;
+
+// True if the floating point unit supports double precision.
+def FeatureFP64           : SubtargetFeature<"fp64", "HasFP64", "true",
+                                             "Floating point unit supports "
+                                             "double precision",
+                                             [FeatureFPRegs64]>;
+
+// True if subtarget has the full 32 double precision FP registers for VFPv3.
+def FeatureD32            : SubtargetFeature<"d32", "HasD32", "true",
+                                             "Extend FP to 32 double registers">;
+
+/// Versions of the VFP flags restricted to single precision, or to
+/// 16 d-registers, or both.
+multiclass VFPver<string name, string query, string description,
+                  list<SubtargetFeature> prev,
+                  list<SubtargetFeature> otherimplies,
+                  list<SubtargetFeature> vfp2prev = []> {
+  def _D16_SP: SubtargetFeature<
+    name#"d16sp", query#"D16SP", "true",
+    description#" with only 16 d-registers and no double precision",
+    !foreach(v, prev, !cast<SubtargetFeature>(v # "_D16_SP")) #
+      !foreach(v, vfp2prev, !cast<SubtargetFeature>(v # "_SP")) #
+      otherimplies>;
+  def _SP: SubtargetFeature<
+    name#"sp", query#"SP", "true",
+    description#" with no double precision",
+    !foreach(v, prev, !cast<SubtargetFeature>(v # "_SP")) #
+      otherimplies # [FeatureD32, !cast<SubtargetFeature>(NAME # "_D16_SP")]>;
+  def _D16: SubtargetFeature<
+    name#"d16", query#"D16", "true",
+    description#" with only 16 d-registers",
+    !foreach(v, prev, !cast<SubtargetFeature>(v # "_D16")) #
+      vfp2prev #
+      otherimplies # [FeatureFP64, !cast<SubtargetFeature>(NAME # "_D16_SP")]>;
+  def "": SubtargetFeature<
+    name, query, "true", description,
+    prev # otherimplies # [
+        !cast<SubtargetFeature>(NAME # "_D16"),
+        !cast<SubtargetFeature>(NAME # "_SP")]>;
+}
+
+def FeatureVFP2_SP        : SubtargetFeature<"vfp2sp", "HasVFPv2SP", "true",
+                                             "Enable VFP2 instructions with "
+                                             "no double precision",
+                                             [FeatureFPRegs]>;
+
+def FeatureVFP2           : SubtargetFeature<"vfp2", "HasVFPv2", "true",
+                                             "Enable VFP2 instructions",
+                                             [FeatureFP64, FeatureVFP2_SP]>;
+
+defm FeatureVFP3: VFPver<"vfp3", "HasVFPv3", "Enable VFP3 instructions",
+                         [], [], [FeatureVFP2]>;
+
+def FeatureNEON           : SubtargetFeature<"neon", "HasNEON", "true",
+                                             "Enable NEON instructions",
+                                             [FeatureVFP3]>;
+
+// True if subtarget supports half-precision FP conversions.
+def FeatureFP16           : SubtargetFeature<"fp16", "HasFP16", "true",
+                                             "Enable half-precision "
+                                             "floating point">;
+
+defm FeatureVFP4: VFPver<"vfp4", "HasVFPv4", "Enable VFP4 instructions",
+                         [FeatureVFP3], [FeatureFP16]>;
+
+defm FeatureFPARMv8: VFPver<"fp-armv8", "HasFPARMv8", "Enable ARMv8 FP",
+                         [FeatureVFP4], []>;
+
+// True if subtarget supports half-precision FP operations.
+def FeatureFullFP16       : SubtargetFeature<"fullfp16", "HasFullFP16", "true",
+                                             "Enable full half-precision "
+                                             "floating point",
+                                             [FeatureFPARMv8_D16_SP, FeatureFPRegs16]>;
+
+// True if subtarget supports half-precision FP fml operations.
+def FeatureFP16FML        : SubtargetFeature<"fp16fml", "HasFP16FML", "true",
+                                             "Enable full half-precision "
+                                             "floating point fml instructions",
+                                             [FeatureFullFP16]>;
+
+// True if subtarget supports [su]div in Thumb mode.
+def FeatureHWDivThumb     : SubtargetFeature<"hwdiv",
+                                             "HasDivideInThumbMode", "true",
+                                             "Enable divide instructions in Thumb">;
+
+// True if subtarget supports [su]div in ARM mode.
+def FeatureHWDivARM       : SubtargetFeature<"hwdiv-arm",
+                                             "HasDivideInARMMode", "true",
+                                             "Enable divide instructions in ARM mode">;
+
+// Atomic Support
+
+// True if the subtarget supports DMB / DSB data barrier instructions.
+def FeatureDB             : SubtargetFeature<"db", "HasDataBarrier", "true",
+                                             "Has data barrier (dmb/dsb) instructions">;
+
+// True if the subtarget supports CLREX instructions.
+def FeatureV7Clrex        : SubtargetFeature<"v7clrex", "HasV7Clrex", "true",
+                                             "Has v7 clrex instruction">;
+
+// True if the subtarget supports DFB data barrier instruction.
+def FeatureDFB  : SubtargetFeature<"dfb", "HasFullDataBarrier", "true",
+                                   "Has full data barrier (dfb) instruction">;
+
+// True if the subtarget supports v8 atomics (LDA/LDAEX etc) instructions.
+def FeatureAcquireRelease : SubtargetFeature<"acquire-release",
+                                             "HasAcquireRelease", "true",
+                                             "Has v8 acquire/release (lda/ldaex "
+                                             " etc) instructions">;
+
+
+// True if floating point compare + branch is slow.
+def FeatureSlowFPBrcc     : SubtargetFeature<"slow-fp-brcc", "IsFPBrccSlow", "true",
+                                             "FP compare + branch is slow">;
+
+// True if the processor supports the Performance Monitor Extensions. These
+// include a generic cycle-counter as well as more fine-grained (often
+// implementation-specific) events.
+def FeaturePerfMon        : SubtargetFeature<"perfmon", "HasPerfMon", "true",
+                                             "Enable support for Performance "
+                                             "Monitor extensions">;
+
+
+// TrustZone Security Extensions
+
+// True if processor supports TrustZone security extensions.
+def FeatureTrustZone      : SubtargetFeature<"trustzone", "HasTrustZone", "true",
+                                             "Enable support for TrustZone "
+                                             "security extensions">;
+
+// True if processor supports ARMv8-M Security Extensions.
+def Feature8MSecExt       : SubtargetFeature<"8msecext", "Has8MSecExt", "true",
+                                             "Enable support for ARMv8-M "
+                                             "Security Extensions">;
+
+// True if processor supports SHA1 and SHA256.
+def FeatureSHA2           : SubtargetFeature<"sha2", "HasSHA2", "true",
+                                             "Enable SHA1 and SHA256 support", [FeatureNEON]>;
+
+def FeatureAES            : SubtargetFeature<"aes", "HasAES", "true",
+                                             "Enable AES support", [FeatureNEON]>;
+
+// True if processor supports Cryptography extensions.
+def FeatureCrypto         : SubtargetFeature<"crypto", "HasCrypto", "true",
+                                             "Enable support for "
+                                             "Cryptography extensions",
+                                             [FeatureNEON, FeatureSHA2, FeatureAES]>;
+
+// True if processor supports CRC instructions.
+def FeatureCRC            : SubtargetFeature<"crc", "HasCRC", "true",
+                                             "Enable support for CRC instructions">;
+
+// True if the ARMv8.2A dot product instructions are supported.
+def FeatureDotProd        : SubtargetFeature<"dotprod", "HasDotProd", "true",
+                                             "Enable support for dot product instructions",
+                                             [FeatureNEON]>;
+
+// True if the processor supports RAS extensions.
+// Not to be confused with FeatureHasRetAddrStack (return address stack).
+def FeatureRAS            : SubtargetFeature<"ras", "HasRAS", "true",
+                                             "Enable Reliability, Availability "
+                                             "and Serviceability extensions">;
+
+// Fast computation of non-negative address offsets.
+// True if processor does positive address offset computation faster.
+def FeatureFPAO           : SubtargetFeature<"fpao", "HasFPAO", "true",
+                                             "Enable fast computation of "
+                                             "positive address offsets">;
+
+// Fast execution of AES crypto operations.
+// True if processor executes back to back AES instruction pairs faster.
+def FeatureFuseAES        : SubtargetFeature<"fuse-aes", "HasFuseAES", "true",
+                                             "CPU fuses AES crypto operations">;
+
+// Fast execution of bottom and top halves of literal generation.
+// True if processor executes back to back bottom and top halves of literal generation faster.
+def FeatureFuseLiterals   : SubtargetFeature<"fuse-literals", "HasFuseLiterals", "true",
+                                             "CPU fuses literal generation operations">;
+
+// Choice of hardware register to use as the thread pointer, if any.
+def FeatureReadTpTPIDRURW :  SubtargetFeature<"read-tp-tpidrurw", "IsReadTPTPIDRURW", "true",
+                                      "Reading thread pointer from TPIDRURW register">;
+def FeatureReadTpTPIDRURO :  SubtargetFeature<"read-tp-tpidruro", "IsReadTPTPIDRURO", "true",
+                                      "Reading thread pointer from TPIDRURO register">;
+def FeatureReadTpTPIDRPRW :  SubtargetFeature<"read-tp-tpidrprw", "IsReadTPTPIDRPRW", "true",
+                                      "Reading thread pointer from TPIDRPRW register">;
+
+// Cyclone can zero VFP registers in 0 cycles.
+// True if the instructions "vmov.i32 d0, #0" and "vmov.i32 q0, #0" are
+// particularly effective at zeroing a VFP register.
+def FeatureZCZeroing      : SubtargetFeature<"zcz", "HasZeroCycleZeroing", "true",
+                                             "Has zero-cycle zeroing instructions">;
+
+// Whether it is profitable to unpredicate certain instructions during if-conversion.
+// True if if conversion may decide to leave some instructions unpredicated.
+def FeatureProfUnpredicate : SubtargetFeature<"prof-unpr",
+                                              "IsProfitableToUnpredicate", "true",
+                                              "Is profitable to unpredicate">;
+
+// Some targets (e.g. Swift) have microcoded VGETLNi32.
+// True if VMOV will be favored over VGETLNi32.
+def FeatureSlowVGETLNi32  : SubtargetFeature<"slow-vgetlni32",
+                                             "HasSlowVGETLNi32", "true",
+                                             "Has slow VGETLNi32 - prefer VMOV">;
+
+// Some targets (e.g. Swift) have microcoded VDUP32.
+// True if VMOV will be favored over VDUP.
+def FeatureSlowVDUP32     : SubtargetFeature<"slow-vdup32", "HasSlowVDUP32",
+                                             "true",
+                                             "Has slow VDUP32 - prefer VMOV">;
+
+// Some targets (e.g. Cortex-A9) prefer VMOVSR to VMOVDRR even when using NEON
+// for scalar FP, as this allows more effective execution domain optimization.
+// True if VMOVSR will be favored over VMOVDRR.
+def FeaturePreferVMOVSR   : SubtargetFeature<"prefer-vmovsr", "PreferVMOVSR",
+                                             "true", "Prefer VMOVSR">;
+
+// Swift has ISHST barriers compatible with Atomic Release semantics but weaker
+// than ISH.
+// True if ISHST barriers will be used for Release semantics.
+def FeaturePrefISHSTBarrier : SubtargetFeature<"prefer-ishst", "PreferISHSTBarriers",
+                                               "true", "Prefer ISHST barriers">;
+
+// Some targets (e.g. Cortex-A9) have muxed AGU and NEON/FPU.
+// True if the AGU and NEON/FPU units are multiplexed.
+def FeatureMuxedUnits     : SubtargetFeature<"muxed-units", "HasMuxedUnits",
+                                             "true",
+                                             "Has muxed AGU and NEON/FPU">;
+
+// Whether VLDM/VSTM starting with odd register number need more microops
+// than single VLDRS.
+// True if a VLDM/VSTM starting with an odd register number is considered to
+// take more microops than single VLDRS/VSTRS.
+def FeatureSlowOddRegister : SubtargetFeature<"slow-odd-reg", "HasSlowOddRegister",
+                                              "true", "VLDM/VSTM starting "
+                                              "with an odd register is slow">;
+
+// Some targets have a renaming dependency when loading into D subregisters.
+// True if loading into a D subregister will be penalized.
+def FeatureSlowLoadDSubreg : SubtargetFeature<"slow-load-D-subreg",
+                                              "HasSlowLoadDSubregister", "true",
+                                              "Loading into D subregs is slow">;
+
+// True if use a wider stride when allocating VFP registers.
+def FeatureUseWideStrideVFP : SubtargetFeature<"wide-stride-vfp",
+                                               "UseWideStrideVFP", "true",
+                                               "Use a wide stride when allocating VFP registers">;
+
+// Some targets (e.g. Cortex-A15) never want VMOVS to be widened to VMOVD.
+// True if VMOVS will never be widened to VMOVD.
+def FeatureDontWidenVMOVS : SubtargetFeature<"dont-widen-vmovs",
+                                             "DontWidenVMOVS", "true",
+                                             "Don't widen VMOVS to VMOVD">;
+
+// Some targets (e.g. Cortex-A15) prefer to avoid mixing operations on different
+// VFP register widths.
+// True if splat a register between VFP and NEON instructions.
+def FeatureSplatVFPToNeon : SubtargetFeature<"splat-vfp-neon",
+                                             "UseSplatVFPToNeon", "true",
+                                             "Splat register from VFP to NEON",
+                                             [FeatureDontWidenVMOVS]>;
+
+// Whether or not it is profitable to expand VFP/NEON MLA/MLS instructions.
+// True if run the MLx expansion pass.
+def FeatureExpandMLx      : SubtargetFeature<"expand-fp-mlx",
+                                             "ExpandMLx", "true",
+                                             "Expand VFP/NEON MLA/MLS instructions">;
+
+// Some targets have special RAW hazards for VFP/NEON VMLA/VMLS.
+// True if VFP/NEON VMLA/VMLS have special RAW hazards.
+def FeatureHasVMLxHazards : SubtargetFeature<"vmlx-hazards", "HasVMLxHazards",
+                                             "true", "Has VMLx hazards">;
+
+// Some targets (e.g. Cortex-A9) want to convert VMOVRS, VMOVSR and VMOVS from
+// VFP to NEON, as an execution domain optimization.
+// True if VMOVRS, VMOVSR and VMOVS will be converted from VFP to NEON.
+def FeatureNEONForFPMovs  : SubtargetFeature<"neon-fpmovs",
+                                             "UseNEONForFPMovs", "true",
+                                             "Convert VMOVSR, VMOVRS, "
+                                             "VMOVS to NEON">;
+
+// Some processors benefit from using NEON instructions for scalar
+// single-precision FP operations. This affects instruction selection and should
+// only be enabled if the handling of denormals is not important.
+// Use the method useNEONForSinglePrecisionFP() to determine if NEON should actually be used.
+def FeatureNEONForFP      : SubtargetFeature<"neonfp",
+                                             "HasNEONForFP",
+                                             "true",
+                                             "Use NEON for single precision FP">;
+
+// On some processors, VLDn instructions that access unaligned data take one
+// extra cycle. Take that into account when computing operand latencies.
+// True if VLDn instructions take an extra cycle for unaligned accesses.
+def FeatureCheckVLDnAlign : SubtargetFeature<"vldn-align", "CheckVLDnAccessAlignment",
+                                             "true",
+                                             "Check for VLDn unaligned access">;
+
+// Some processors have a nonpipelined VFP coprocessor.
+// True if VFP instructions are not pipelined.
+def FeatureNonpipelinedVFP : SubtargetFeature<"nonpipelined-vfp",
+                                              "NonpipelinedVFP", "true",
+                                              "VFP instructions are not pipelined">;
+
+// Some processors have FP multiply-accumulate instructions that don't
+// play nicely with other VFP / NEON instructions, and it's generally better
+// to just not use them.
+// If the VFP2 / NEON instructions are available, indicates
+// whether the FP VML[AS] instructions are slow (if so, don't use them).
+def FeatureHasSlowFPVMLx  : SubtargetFeature<"slowfpvmlx", "SlowFPVMLx", "true",
+                                             "Disable VFP / NEON MAC instructions">;
+
+// VFPv4 added VFMA instructions that can similarly be fast or slow.
+// If the VFP4 / NEON instructions are available, indicates
+// whether the FP VFM[AS] instructions are slow (if so, don't use them).
+def FeatureHasSlowFPVFMx  : SubtargetFeature<"slowfpvfmx", "SlowFPVFMx", "true",
+                                             "Disable VFP / NEON FMA instructions">;
+
+// Cortex-A8 / A9 Advanced SIMD has multiplier accumulator forwarding.
+/// True if NEON has special multiplier accumulator
+/// forwarding to allow mul + mla being issued back to back.
+def FeatureVMLxForwarding : SubtargetFeature<"vmlx-forwarding",
+                                             "HasVMLxForwarding", "true",
+                                             "Has multiplier accumulator forwarding">;
+
+// Disable 32-bit to 16-bit narrowing for experimentation.
+// True if codegen would prefer 32-bit Thumb instructions over 16-bit ones.
+def FeaturePref32BitThumb : SubtargetFeature<"32bit", "Prefers32BitThumb", "true",
+                                             "Prefer 32-bit Thumb instrs">;
+
+def FeaturePrefLoopAlign32 : SubtargetFeature<"loop-align", "PrefLoopLogAlignment","2",
+                                              "Prefer 32-bit alignment for loops">;
+
+def FeatureMVEVectorCostFactor1 : SubtargetFeature<"mve1beat", "MVEVectorCostFactor", "4",
+                        "Model MVE instructions as a 1 beat per tick architecture">;
+
+def FeatureMVEVectorCostFactor2 : SubtargetFeature<"mve2beat", "MVEVectorCostFactor", "2",
+                        "Model MVE instructions as a 2 beats per tick architecture">;
+
+def FeatureMVEVectorCostFactor4 : SubtargetFeature<"mve4beat", "MVEVectorCostFactor", "1",
+                        "Model MVE instructions as a 4 beats per tick architecture">;
+
+/// Some instructions update CPSR partially, which can add false dependency for
+/// out-of-order implementation, e.g. Cortex-A9, unless each individual bit is
+/// mapped to a separate physical register. Avoid partial CPSR update for these
+/// processors.
+/// True if codegen would avoid using instructions
+/// that partially update CPSR and add false dependency on the previous
+/// CPSR setting instruction.
+def FeatureAvoidPartialCPSR : SubtargetFeature<"avoid-partial-cpsr",
+                                               "AvoidCPSRPartialUpdate", "true",
+                                 "Avoid CPSR partial update for OOO execution">;
+
+/// Disable +1 predication cost for instructions updating CPSR.
+/// Enabled for Cortex-A57.
+/// True if disable +1 predication cost for instructions updating CPSR. Enabled for Cortex-A57.
+def FeatureCheapPredicableCPSR : SubtargetFeature<"cheap-predicable-cpsr",
+                                                  "CheapPredicableCPSRDef",
+                                                  "true",
+                  "Disable +1 predication cost for instructions updating CPSR">;
+
+// True if codegen should avoid using flag setting movs with shifter operand (i.e. asr, lsl, lsr).
+def FeatureAvoidMOVsShOp  : SubtargetFeature<"avoid-movs-shop",
+                                             "AvoidMOVsShifterOperand", "true",
+                                             "Avoid movs instructions with "
+                                             "shifter operand">;
+
+// Some processors perform return stack prediction. CodeGen should avoid issue
+// "normal" call instructions to callees which do not return.
+def FeatureHasRetAddrStack : SubtargetFeature<"ret-addr-stack",
+                                              "HasRetAddrStack", "true",
+                                              "Has return address stack">;
+
+// Some processors have no branch predictor, which changes the expected cost of
+// taking a branch which affects the choice of whether to use predicated
+// instructions.
+// True if the subtarget has a branch predictor. Having
+// a branch predictor or not changes the expected cost of taking a branch
+// which affects the choice of whether to use predicated instructions.
+def FeatureHasNoBranchPredictor : SubtargetFeature<"no-branch-predictor",
+                                                   "HasBranchPredictor", "false",
+                                                   "Has no branch predictor">;
+
+/// DSP extension.
+/// True if the subtarget supports the DSP (saturating arith and such) instructions.
+def FeatureDSP            : SubtargetFeature<"dsp", "HasDSP", "true",
+                                             "Supports DSP instructions in "
+                                             "ARM and/or Thumb2">;
+
+// True if the subtarget supports Multiprocessing extension (ARMv7 only).
+def FeatureMP             : SubtargetFeature<"mp", "HasMPExtension", "true",
+                                        "Supports Multiprocessing extension">;
+
+// Virtualization extension - requires HW divide (ARMv7-AR ARMARM - 4.4.8).
+def FeatureVirtualization : SubtargetFeature<"virtualization",
+                                             "HasVirtualization", "true",
+                                             "Supports Virtualization extension",
+                                             [FeatureHWDivThumb, FeatureHWDivARM]>;
+
+// Special TRAP encoding for NaCl, which looks like a TRAP in Thumb too.
+// See ARMInstrInfo.td for details.
+// True if NaCl TRAP instruction is generated instead of the regular TRAP.
+def FeatureNaClTrap       : SubtargetFeature<"nacl-trap", "UseNaClTrap", "true",
+                                             "NaCl trap">;
+
+// True if the subtarget disallows unaligned memory
+// accesses for some types.  For details, see
+// ARMTargetLowering::allowsMisalignedMemoryAccesses().
+def FeatureStrictAlign    : SubtargetFeature<"strict-align",
+                                             "StrictAlign", "true",
+                                             "Disallow all unaligned memory "
+                                             "access">;
+
+// Generate calls via indirect call instructions.
+def FeatureLongCalls      : SubtargetFeature<"long-calls", "GenLongCalls", "true",
+                                             "Generate calls via indirect call "
+                                             "instructions">;
+
+// Generate code that does not contain data access to code sections.
+def FeatureExecuteOnly    : SubtargetFeature<"execute-only",
+                                             "GenExecuteOnly", "true",
+                                             "Enable the generation of "
+                                             "execute only code.">;
+
+// True if R9 is not available as a general purpose register.
+def FeatureReserveR9      : SubtargetFeature<"reserve-r9", "ReserveR9", "true",
+                                             "Reserve R9, making it unavailable"
+                                             " as GPR">;
+
+// True if MOVT / MOVW pairs are not used for materialization of
+// 32-bit imms (including global addresses).
+def FeatureNoMovt         : SubtargetFeature<"no-movt", "NoMovt", "true",
+                                             "Don't use movt/movw pairs for "
+                                             "32-bit imms">;
+
+/// Implicitly convert an instruction to a different one if its immediates
+/// cannot be encoded. For example, ADD r0, r1, #FFFFFFFF -> SUB r0, r1, #1.
+def FeatureNoNegativeImmediates
+                          : SubtargetFeature<"no-neg-immediates",
+                                             "NegativeImmediates", "false",
+                                             "Convert immediates and instructions "
+                                             "to their negated or complemented "
+                                             "equivalent when the immediate does "
+                                             "not fit in the encoding.">;
+
+// Use the MachineScheduler for instruction scheduling for the subtarget.
+def FeatureUseMISched: SubtargetFeature<"use-misched", "UseMISched", "true",
+                                        "Use the MachineScheduler">;
+
+// Use the MachinePipeliner for instruction scheduling for the subtarget.
+def FeatureUseMIPipeliner: SubtargetFeature<"use-mipipeliner", "UseMIPipeliner", "true",
+                                            "Use the MachinePipeliner">;
+
+// False if scheduling should happen again after register allocation.
+def FeatureNoPostRASched : SubtargetFeature<"disable-postra-scheduler",
+    "DisablePostRAScheduler", "true",
+    "Don't schedule again after register allocation">;
+
+// Armv8.5-A extensions
+
+// Has speculation barrier.
+def FeatureSB       : SubtargetFeature<"sb", "HasSB", "true",
+  "Enable v8.5a Speculation Barrier" >;
+
+// Armv8.6-A extensions
+
+// True if subtarget supports BFloat16 floating point operations.
+def FeatureBF16     : SubtargetFeature<"bf16", "HasBF16", "true",
+  "Enable support for BFloat16 instructions",  [FeatureNEON]>;
+
+// True if subtarget supports 8-bit integer matrix multiply.
+def FeatureMatMulInt8 : SubtargetFeature<"i8mm", "HasMatMulInt8",
+    "true", "Enable Matrix Multiply Int8 Extension", [FeatureNEON]>;
+
+// Armv8.1-M extensions
+
+// True if the processor supports the Low Overhead Branch extension.
+def FeatureLOB            : SubtargetFeature<"lob", "HasLOB", "true",
+                                             "Enable Low Overhead Branch "
+                                             "extensions">;
+
+// Mitigate against the cve-2021-35465 security vulnurability.
+def FeatureFixCMSE_CVE_2021_35465 : SubtargetFeature<"fix-cmse-cve-2021-35465",
+                                        "FixCMSE_CVE_2021_35465", "true",
+                                        "Mitigate against the cve-2021-35465 "
+                                        "security vulnurability">;
+
+def FeaturePACBTI         : SubtargetFeature<"pacbti", "HasPACBTI", "true",
+                                             "Enable Pointer Authentication and Branch "
+                                             "Target Identification">;
+
+/// Don't place a BTI instruction after return-twice constructs (setjmp).
+def FeatureNoBTIAtReturnTwice : SubtargetFeature<"no-bti-at-return-twice",
+                                                 "NoBTIAtReturnTwice", "true",
+                                                 "Don't place a BTI instruction "
+                                                 "after a return-twice">;
+
+// Armv8.9-A/Armv9.4-A 2022 Architecture Extensions
+def FeatureCLRBHB : SubtargetFeature<"clrbhb", "HasCLRBHB", "true",
+                                     "Enable Clear BHB instruction">;
+
+
+def FeatureFixCortexA57AES1742098 : SubtargetFeature<"fix-cortex-a57-aes-1742098",
+  "FixCortexA57AES1742098", "true",
+  "Work around Cortex-A57 Erratum 1742098 / Cortex-A72 Erratum 1655431 (AES)">;
+
+def FeatureAAPCSFrameChain : SubtargetFeature<"aapcs-frame-chain",
+                                              "CreateAAPCSFrameChain", "true",
+                                              "Create an AAPCS compliant frame chain">;
+
+def FeatureAAPCSFrameChainLeaf : SubtargetFeature<"aapcs-frame-chain-leaf",
+                                                  "CreateAAPCSFrameChainLeaf", "true",
+                                                  "Create an AAPCS compliant frame chain "
+                                                  "for leaf functions",
+                                                  [FeatureAAPCSFrameChain]>;
+
+// Assume that lock-free 32-bit atomics are available, even if the target
+// and operating system combination would not usually provide them. The user
+// is responsible for providing any necessary __sync implementations. Code
+// built with this feature is not ABI-compatible with code built without this
+// feature, if atomic variables are exposed across the ABI boundary.
+def FeatureAtomics32 : SubtargetFeature<
+    "atomics-32", "HasForced32BitAtomics", "true",
+    "Assume that lock-free 32-bit atomics are available">;
+
+//===----------------------------------------------------------------------===//
+// ARM architecture class
+//
+
+// A-series ISA
+def FeatureAClass : SubtargetFeature<"aclass", "ARMProcClass", "AClass",
+                                     "Is application profile ('A' series)">;
+
+// R-series ISA
+def FeatureRClass : SubtargetFeature<"rclass", "ARMProcClass", "RClass",
+                                     "Is realtime profile ('R' series)">;
+
+// M-series ISA
+def FeatureMClass : SubtargetFeature<"mclass", "ARMProcClass", "MClass",
+                                     "Is microcontroller profile ('M' series)">;
+
+// True if Thumb2 instructions are supported.
+def FeatureThumb2 : SubtargetFeature<"thumb2", "HasThumb2", "true",
+                                     "Enable Thumb2 instructions">;
+
+// True if subtarget does not support ARM mode execution.
+def FeatureNoARM  : SubtargetFeature<"noarm", "NoARM", "true",
+                                     "Does not support ARM mode execution">;
+
+//===----------------------------------------------------------------------===//
+// ARM ISAa.
+//
+// Specify whether target support specific ARM ISA variants.
+
+def HasV4TOps   : SubtargetFeature<"v4t", "HasV4TOps", "true",
+                                   "Support ARM v4T instructions">;
+
+def HasV5TOps   : SubtargetFeature<"v5t", "HasV5TOps", "true",
+                                   "Support ARM v5T instructions",
+                                   [HasV4TOps]>;
+
+def HasV5TEOps  : SubtargetFeature<"v5te", "HasV5TEOps", "true",
+                                   "Support ARM v5TE, v5TEj, and "
+                                   "v5TExp instructions",
+                                   [HasV5TOps]>;
+
+def HasV6Ops    : SubtargetFeature<"v6", "HasV6Ops", "true",
+                                   "Support ARM v6 instructions",
+                                   [HasV5TEOps]>;
+
+def HasV6MOps   : SubtargetFeature<"v6m", "HasV6MOps", "true",
+                                   "Support ARM v6M instructions",
+                                   [HasV6Ops]>;
+
+def HasV8MBaselineOps : SubtargetFeature<"v8m", "HasV8MBaselineOps", "true",
+                                         "Support ARM v8M Baseline instructions",
+                                         [HasV6MOps]>;
+
+def HasV6KOps   : SubtargetFeature<"v6k", "HasV6KOps", "true",
+                                   "Support ARM v6k instructions",
+                                   [HasV6Ops]>;
+
+def HasV6T2Ops  : SubtargetFeature<"v6t2", "HasV6T2Ops", "true",
+                                   "Support ARM v6t2 instructions",
+                                   [HasV8MBaselineOps, HasV6KOps, FeatureThumb2]>;
+
+def HasV7Ops    : SubtargetFeature<"v7", "HasV7Ops", "true",
+                                   "Support ARM v7 instructions",
+                                   [HasV6T2Ops, FeatureV7Clrex]>;
+
+def HasV8MMainlineOps :
+                  SubtargetFeature<"v8m.main", "HasV8MMainlineOps", "true",
+                                   "Support ARM v8M Mainline instructions",
+                                   [HasV7Ops]>;
+
+def HasV8Ops    : SubtargetFeature<"v8", "HasV8Ops", "true",
+                                   "Support ARM v8 instructions",
+                                   [HasV7Ops, FeaturePerfMon, FeatureAcquireRelease]>;
+
+def HasV8_1aOps : SubtargetFeature<"v8.1a", "HasV8_1aOps", "true",
+                                   "Support ARM v8.1a instructions",
+                                   [HasV8Ops]>;
+
+def HasV8_2aOps : SubtargetFeature<"v8.2a", "HasV8_2aOps", "true",
+                                   "Support ARM v8.2a instructions",
+                                   [HasV8_1aOps]>;
+
+def HasV8_3aOps   : SubtargetFeature<"v8.3a", "HasV8_3aOps", "true",
+                                   "Support ARM v8.3a instructions",
+                                   [HasV8_2aOps]>;
+
+def HasV8_4aOps   : SubtargetFeature<"v8.4a", "HasV8_4aOps", "true",
+                                   "Support ARM v8.4a instructions",
+                                   [HasV8_3aOps, FeatureDotProd]>;
+
+def HasV8_5aOps   : SubtargetFeature<"v8.5a", "HasV8_5aOps", "true",
+                                   "Support ARM v8.5a instructions",
+                                   [HasV8_4aOps, FeatureSB]>;
+
+def HasV8_6aOps   : SubtargetFeature<"v8.6a", "HasV8_6aOps", "true",
+                                   "Support ARM v8.6a instructions",
+                                   [HasV8_5aOps, FeatureBF16,
+                                    FeatureMatMulInt8]>;
+
+def HasV8_7aOps   : SubtargetFeature<"v8.7a", "HasV8_7aOps", "true",
+                                   "Support ARM v8.7a instructions",
+                                   [HasV8_6aOps]>;
+
+def HasV8_8aOps   : SubtargetFeature<"v8.8a", "HasV8_8aOps", "true",
+                                   "Support ARM v8.8a instructions",
+                                   [HasV8_7aOps]>;
+
+def HasV8_9aOps   : SubtargetFeature<"v8.9a", "HasV8_9aOps", "true",
+                                   "Support ARM v8.9a instructions",
+                                   [HasV8_8aOps, FeatureCLRBHB]>;
+
+def HasV9_0aOps   : SubtargetFeature<"v9a", "HasV9_0aOps", "true",
+                                   "Support ARM v9a instructions",
+                                   [HasV8_5aOps]>;
+
+def HasV9_1aOps   : SubtargetFeature<"v9.1a", "HasV9_1aOps", "true",
+                                   "Support ARM v9.1a instructions",
+                                   [HasV8_6aOps, HasV9_0aOps]>;
+
+def HasV9_2aOps   : SubtargetFeature<"v9.2a", "HasV9_2aOps", "true",
+                                   "Support ARM v9.2a instructions",
+                                   [HasV8_7aOps, HasV9_1aOps]>;
+
+def HasV9_3aOps   : SubtargetFeature<"v9.3a", "HasV9_3aOps", "true",
+                                   "Support ARM v9.3a instructions",
+                                   [HasV8_8aOps, HasV9_2aOps]>;
+
+def HasV9_4aOps   : SubtargetFeature<"v9.4a", "HasV9_4aOps", "true",
+                                   "Support ARM v9.4a instructions",
+                                   [HasV8_9aOps, HasV9_3aOps]>;
+
+// Armv9.5-A is a v9-only architecture. From v9.5-A onwards there's no mapping
+// to an equivalent v8.x version.
+def HasV9_5aOps   : SubtargetFeature<"v9.5a", "HasV9_5aOps", "true",
+                                   "Support ARM v9.5a instructions",
+                                   [HasV9_4aOps]>;
+
+def HasV8_1MMainlineOps : SubtargetFeature<
+               "v8.1m.main", "HasV8_1MMainlineOps", "true",
+               "Support ARM v8-1M Mainline instructions",
+               [HasV8MMainlineOps]>;
+def HasMVEIntegerOps : SubtargetFeature<
+               "mve", "HasMVEIntegerOps", "true",
+               "Support M-Class Vector Extension with integer ops",
+               [HasV8_1MMainlineOps, FeatureDSP, FeatureFPRegs16, FeatureFPRegs64]>;
+def HasMVEFloatOps : SubtargetFeature<
+               "mve.fp", "HasMVEFloatOps", "true",
+               "Support M-Class Vector Extension with integer and floating ops",
+               [HasMVEIntegerOps, FeatureFPARMv8_D16_SP, FeatureFullFP16]>;
+
+def HasCDEOps : SubtargetFeature<"cde", "HasCDEOps", "true",
+                                 "Support CDE instructions",
+                                 [HasV8MMainlineOps]>;
+
+foreach i = {0-7} in
+    def FeatureCoprocCDE#i : SubtargetFeature<"cdecp"#i,
+                                              "CoprocCDE["#i#"]", "true",
+                                              "Coprocessor "#i#" ISA is CDEv1",
+                                              [HasCDEOps]>;
+
+//===----------------------------------------------------------------------===//
+// Control codegen mitigation against Straight Line Speculation vulnerability.
+//===----------------------------------------------------------------------===//
+
+/// Harden against Straight Line Speculation for Returns and Indirect Branches.
+def FeatureHardenSlsRetBr : SubtargetFeature<"harden-sls-retbr",
+  "HardenSlsRetBr", "true",
+  "Harden against straight line speculation across RETurn and BranchRegister "
+  "instructions">;
+/// Harden against Straight Line Speculation for indirect calls.
+def FeatureHardenSlsBlr : SubtargetFeature<"harden-sls-blr",
+  "HardenSlsBlr", "true",
+  "Harden against straight line speculation across indirect calls">;
+/// Generate thunk code for SLS mitigation in the normal text section.
+def FeatureHardenSlsNoComdat : SubtargetFeature<"harden-sls-nocomdat",
+  "HardenSlsNoComdat", "true",
+  "Generate thunk code for SLS mitigation in the normal text section">;
+
+//===----------------------------------------------------------------------===//
+// Endianness of instruction encodings in memory.
+//
+// In the current Arm architecture, this is usually little-endian regardless of
+// data endianness. But before Armv7 it was typical for instruction endianness
+// to match data endianness, so that a big-endian system was consistently big-
+// endian. And Armv7-R can be configured to use big-endian instructions.
+//
+// Additionally, even when targeting Armv7-A, big-endian instructions can be
+// found in relocatable object files, because the Arm ABI specifies that the
+// linker byte-reverses them depending on the target architecture.
+//
+// So we have a feature here to indicate that instructions are stored big-
+// endian, which you can set when instantiating an MCDisassembler.
+def ModeBigEndianInstructions : SubtargetFeature<"big-endian-instructions",
+    "BigEndianInstructions", "true",
+     "Expect instructions to be stored big-endian.">;
+
diff --git a/llvm/lib/Target/ARM/ARMFrameLowering.cpp b/llvm/lib/Target/ARM/ARMFrameLowering.cpp
index 9b54dd4e4e61..a332f743f495 100644
--- a/llvm/lib/Target/ARM/ARMFrameLowering.cpp
+++ b/llvm/lib/Target/ARM/ARMFrameLowering.cpp
@@ -1873,7 +1873,7 @@ skipAlignedDPRCS2Spills(MachineBasicBlock::iterator MI,
   case 1:
   case 2:
   case 4:
-    assert(MI->killsRegister(ARM::R4) && "Missed kill flag");
+    assert(MI->killsRegister(ARM::R4, /*TRI=*/nullptr) && "Missed kill flag");
     ++MI;
   }
   return MI;
diff --git a/llvm/lib/Target/ARM/ARMISelLowering.cpp b/llvm/lib/Target/ARM/ARMISelLowering.cpp
index 3907131be6d1..f67a68acbf23 100644
--- a/llvm/lib/Target/ARM/ARMISelLowering.cpp
+++ b/llvm/lib/Target/ARM/ARMISelLowering.cpp
@@ -1555,15 +1555,11 @@ ARMTargetLowering::ARMTargetLowering(const TargetMachine &TM,
 
   if (Subtarget->hasNEON()) {
     // vmin and vmax aren't available in a scalar form, so we can use
-    // a NEON instruction with an undef lane instead.  This has a performance
-    // penalty on some cores, so we don't do this unless we have been
-    // asked to by the core tuning model.
-    if (Subtarget->useNEONForSinglePrecisionFP()) {
-      setOperationAction(ISD::FMINIMUM, MVT::f32, Legal);
-      setOperationAction(ISD::FMAXIMUM, MVT::f32, Legal);
-      setOperationAction(ISD::FMINIMUM, MVT::f16, Legal);
-      setOperationAction(ISD::FMAXIMUM, MVT::f16, Legal);
-    }
+    // a NEON instruction with an undef lane instead.
+    setOperationAction(ISD::FMINIMUM, MVT::f32, Legal);
+    setOperationAction(ISD::FMAXIMUM, MVT::f32, Legal);
+    setOperationAction(ISD::FMINIMUM, MVT::f16, Legal);
+    setOperationAction(ISD::FMAXIMUM, MVT::f16, Legal);
     setOperationAction(ISD::FMINIMUM, MVT::v2f32, Legal);
     setOperationAction(ISD::FMAXIMUM, MVT::v2f32, Legal);
     setOperationAction(ISD::FMINIMUM, MVT::v4f32, Legal);
@@ -11796,9 +11792,9 @@ static bool checkAndUpdateCPSRKill(MachineBasicBlock::iterator SelectItr,
   MachineBasicBlock::iterator miI(std::next(SelectItr));
   for (MachineBasicBlock::iterator miE = BB->end(); miI != miE; ++miI) {
     const MachineInstr& mi = *miI;
-    if (mi.readsRegister(ARM::CPSR))
+    if (mi.readsRegister(ARM::CPSR, /*TRI=*/nullptr))
       return false;
-    if (mi.definesRegister(ARM::CPSR))
+    if (mi.definesRegister(ARM::CPSR, /*TRI=*/nullptr))
       break; // Should have kill-flag - update below.
   }
 
@@ -12157,7 +12153,7 @@ ARMTargetLowering::EmitInstrWithCustomInserter(MachineInstr &MI,
 
     // Check whether CPSR is live past the tMOVCCr_pseudo.
     const TargetRegisterInfo *TRI = Subtarget->getRegisterInfo();
-    if (!MI.killsRegister(ARM::CPSR) &&
+    if (!MI.killsRegister(ARM::CPSR, /*TRI=*/nullptr) &&
         !checkAndUpdateCPSRKill(MI, thisMBB, TRI)) {
       copy0MBB->addLiveIn(ARM::CPSR);
       sinkMBB->addLiveIn(ARM::CPSR);
diff --git a/llvm/lib/Target/ARM/ARMLoadStoreOptimizer.cpp b/llvm/lib/Target/ARM/ARMLoadStoreOptimizer.cpp
index 469340784284..4a5b672f862b 100644
--- a/llvm/lib/Target/ARM/ARMLoadStoreOptimizer.cpp
+++ b/llvm/lib/Target/ARM/ARMLoadStoreOptimizer.cpp
@@ -495,7 +495,7 @@ void ARMLoadStoreOpt::UpdateBaseRegUses(MachineBasicBlock &MBB,
     bool InsertSub = false;
     unsigned Opc = MBBI->getOpcode();
 
-    if (MBBI->readsRegister(Base)) {
+    if (MBBI->readsRegister(Base, /*TRI=*/nullptr)) {
       int Offset;
       bool IsLoad =
         Opc == ARM::tLDRi || Opc == ARM::tLDRHi || Opc == ARM::tLDRBi;
@@ -560,7 +560,8 @@ void ARMLoadStoreOpt::UpdateBaseRegUses(MachineBasicBlock &MBB,
       return;
     }
 
-    if (MBBI->killsRegister(Base) || MBBI->definesRegister(Base))
+    if (MBBI->killsRegister(Base, /*TRI=*/nullptr) ||
+        MBBI->definesRegister(Base, /*TRI=*/nullptr))
       // Register got killed. Stop updating.
       return;
   }
@@ -888,7 +889,7 @@ MachineInstr *ARMLoadStoreOpt::MergeOpsUpdate(const MergeCandidate &Cand) {
         if (is_contained(ImpDefs, DefReg))
           continue;
         // We can ignore cases where the super-reg is read and written.
-        if (MI->readsRegister(DefReg))
+        if (MI->readsRegister(DefReg, /*TRI=*/nullptr))
           continue;
         ImpDefs.push_back(DefReg);
       }
@@ -903,7 +904,7 @@ MachineInstr *ARMLoadStoreOpt::MergeOpsUpdate(const MergeCandidate &Cand) {
   MachineBasicBlock &MBB = *LatestMI->getParent();
   unsigned Offset = getMemoryOpOffset(*First);
   Register Base = getLoadStoreBaseOp(*First).getReg();
-  bool BaseKill = LatestMI->killsRegister(Base);
+  bool BaseKill = LatestMI->killsRegister(Base, /*TRI=*/nullptr);
   Register PredReg;
   ARMCC::CondCodes Pred = getInstrPredicate(*First, PredReg);
   DebugLoc DL = First->getDebugLoc();
@@ -2076,7 +2077,8 @@ bool ARMLoadStoreOpt::CombineMovBx(MachineBasicBlock &MBB) {
 
   MachineBasicBlock::iterator Prev = MBBI;
   --Prev;
-  if (Prev->getOpcode() != ARM::tMOVr || !Prev->definesRegister(ARM::LR))
+  if (Prev->getOpcode() != ARM::tMOVr ||
+      !Prev->definesRegister(ARM::LR, /*TRI=*/nullptr))
     return false;
 
   for (auto Use : Prev->uses())
@@ -3176,7 +3178,7 @@ bool ARMPreAllocLoadStoreOpt::DistributeIncrements(Register Base) {
     if (PrePostInc || BaseAccess->getParent() != Increment->getParent())
       return false;
     Register PredReg;
-    if (Increment->definesRegister(ARM::CPSR) ||
+    if (Increment->definesRegister(ARM::CPSR, /*TRI=*/nullptr) ||
         getInstrPredicate(*Increment, PredReg) != ARMCC::AL)
       return false;
 
diff --git a/llvm/lib/Target/ARM/ARMLowOverheadLoops.cpp b/llvm/lib/Target/ARM/ARMLowOverheadLoops.cpp
index ea5dd5427ce7..91a36898aecb 100644
--- a/llvm/lib/Target/ARM/ARMLowOverheadLoops.cpp
+++ b/llvm/lib/Target/ARM/ARMLowOverheadLoops.cpp
@@ -91,11 +91,11 @@ static bool isVectorPredicated(MachineInstr *MI) {
 }
 
 static bool isVectorPredicate(MachineInstr *MI) {
-  return MI->findRegisterDefOperandIdx(ARM::VPR) != -1;
+  return MI->findRegisterDefOperandIdx(ARM::VPR, /*TRI=*/nullptr) != -1;
 }
 
 static bool hasVPRUse(MachineInstr &MI) {
-  return MI.findRegisterUseOperandIdx(ARM::VPR) != -1;
+  return MI.findRegisterUseOperandIdx(ARM::VPR, /*TRI=*/nullptr) != -1;
 }
 
 static bool isDomainMVE(MachineInstr *MI) {
@@ -564,7 +564,8 @@ static bool TryRemove(MachineInstr *MI, ReachingDefAnalysis &RDA,
     SmallPtrSet<MachineInstr *, 2> ModifiedITs;
     SmallPtrSet<MachineInstr *, 2> RemoveITs;
     for (auto *Dead : Killed) {
-      if (MachineOperand *MO = Dead->findRegisterUseOperand(ARM::ITSTATE)) {
+      if (MachineOperand *MO =
+              Dead->findRegisterUseOperand(ARM::ITSTATE, /*TRI=*/nullptr)) {
         MachineInstr *IT = RDA.getMIOperand(Dead, *MO);
         RemoveITs.insert(IT);
         auto &CurrentBlock = ITBlocks[IT];
diff --git a/llvm/lib/Target/ARM/ARMProcessors.td b/llvm/lib/Target/ARM/ARMProcessors.td
new file mode 100644
index 000000000000..2c5594976400
--- /dev/null
+++ b/llvm/lib/Target/ARM/ARMProcessors.td
@@ -0,0 +1,577 @@
+class ProcNoItin<string Name, list<SubtargetFeature> Features>
+  : Processor<Name, NoItineraries, Features>;
+
+//===----------------------------------------------------------------------===//
+// ARM Processor subtarget features.
+//
+
+def ProcA5      : SubtargetFeature<"a5", "ARMProcFamily", "CortexA5",
+                                   "Cortex-A5 ARM processors", []>;
+def ProcA7      : SubtargetFeature<"a7", "ARMProcFamily", "CortexA7",
+                                   "Cortex-A7 ARM processors", []>;
+def ProcA8      : SubtargetFeature<"a8", "ARMProcFamily", "CortexA8",
+                                   "Cortex-A8 ARM processors", []>;
+def ProcA9      : SubtargetFeature<"a9", "ARMProcFamily", "CortexA9",
+                                   "Cortex-A9 ARM processors", []>;
+def ProcA12     : SubtargetFeature<"a12", "ARMProcFamily", "CortexA12",
+                                   "Cortex-A12 ARM processors", []>;
+def ProcA15     : SubtargetFeature<"a15", "ARMProcFamily", "CortexA15",
+                                   "Cortex-A15 ARM processors", []>;
+def ProcA17     : SubtargetFeature<"a17", "ARMProcFamily", "CortexA17",
+                                   "Cortex-A17 ARM processors", []>;
+def ProcA32     : SubtargetFeature<"a32", "ARMProcFamily", "CortexA32",
+                                   "Cortex-A32 ARM processors", []>;
+def ProcA35     : SubtargetFeature<"a35", "ARMProcFamily", "CortexA35",
+                                   "Cortex-A35 ARM processors", []>;
+def ProcA53     : SubtargetFeature<"a53", "ARMProcFamily", "CortexA53",
+                                   "Cortex-A53 ARM processors", []>;
+def ProcA55     : SubtargetFeature<"a55", "ARMProcFamily", "CortexA55",
+                                   "Cortex-A55 ARM processors", []>;
+def ProcA57     : SubtargetFeature<"a57", "ARMProcFamily", "CortexA57",
+                                   "Cortex-A57 ARM processors", []>;
+def ProcA72     : SubtargetFeature<"a72", "ARMProcFamily", "CortexA72",
+                                   "Cortex-A72 ARM processors", []>;
+def ProcA73     : SubtargetFeature<"a73", "ARMProcFamily", "CortexA73",
+                                   "Cortex-A73 ARM processors", []>;
+def ProcA75     : SubtargetFeature<"a75", "ARMProcFamily", "CortexA75",
+                                   "Cortex-A75 ARM processors", []>;
+def ProcA76     : SubtargetFeature<"a76", "ARMProcFamily", "CortexA76",
+                                   "Cortex-A76 ARM processors", []>;
+def ProcA77     : SubtargetFeature<"a77", "ARMProcFamily", "CortexA77",
+                                   "Cortex-A77 ARM processors", []>;
+def ProcA78     : SubtargetFeature<"cortex-a78", "ARMProcFamily", "CortexA78",
+                                   "Cortex-A78 ARM processors", []>;
+def ProcA78AE   : SubtargetFeature<"cortex-a78ae", "ARMProcFamily", "CortexA78AE",
+                                   "Cortex-A78AE ARM processors", []>;
+def ProcA78C    : SubtargetFeature<"a78c", "ARMProcFamily", "CortexA78C",
+                                   "Cortex-A78C ARM processors", []>;
+def ProcA710    : SubtargetFeature<"cortex-a710", "ARMProcFamily",
+                                   "CortexA710", "Cortex-A710 ARM processors", []>;
+def ProcX1      : SubtargetFeature<"cortex-x1", "ARMProcFamily", "CortexX1",
+                                   "Cortex-X1 ARM processors", []>;
+def ProcX1C     : SubtargetFeature<"cortex-x1c", "ARMProcFamily", "CortexX1C",
+                                   "Cortex-X1C ARM processors", []>;
+
+def ProcV1      : SubtargetFeature<"neoverse-v1", "ARMProcFamily",
+                                   "NeoverseV1", "Neoverse-V1 ARM processors", []>;
+
+def ProcKrait   : SubtargetFeature<"krait", "ARMProcFamily", "Krait",
+                                   "Qualcomm Krait processors", []>;
+def ProcKryo    : SubtargetFeature<"kryo", "ARMProcFamily", "Kryo",
+                                   "Qualcomm Kryo processors", []>;
+def ProcSwift   : SubtargetFeature<"swift", "ARMProcFamily", "Swift",
+                                   "Swift ARM processors", []>;
+
+def ProcExynos  : SubtargetFeature<"exynos", "ARMProcFamily", "Exynos",
+                                   "Samsung Exynos processors",
+                                   [FeatureZCZeroing,
+                                    FeatureUseWideStrideVFP,
+                                    FeatureSplatVFPToNeon,
+                                    FeatureSlowVGETLNi32,
+                                    FeatureSlowVDUP32,
+                                    FeatureSlowFPBrcc,
+                                    FeatureProfUnpredicate,
+                                    FeatureHWDivThumb,
+                                    FeatureHWDivARM,
+                                    FeatureHasSlowFPVMLx,
+                                    FeatureHasSlowFPVFMx,
+                                    FeatureHasRetAddrStack,
+                                    FeatureFuseLiterals,
+                                    FeatureFuseAES,
+                                    FeatureExpandMLx,
+                                    FeatureCrypto,
+                                    FeatureCRC]>;
+
+def ProcR4      : SubtargetFeature<"r4", "ARMProcFamily", "CortexR4",
+                                   "Cortex-R4 ARM processors", []>;
+def ProcR5      : SubtargetFeature<"r5", "ARMProcFamily", "CortexR5",
+                                   "Cortex-R5 ARM processors", []>;
+def ProcR7      : SubtargetFeature<"r7", "ARMProcFamily", "CortexR7",
+                                   "Cortex-R7 ARM processors", []>;
+def ProcR52     : SubtargetFeature<"r52", "ARMProcFamily", "CortexR52",
+                                   "Cortex-R52 ARM processors", []>;
+
+def ProcM3      : SubtargetFeature<"m3", "ARMProcFamily", "CortexM3",
+                                   "Cortex-M3 ARM processors", []>;
+def ProcM7      : SubtargetFeature<"m7", "ARMProcFamily", "CortexM7",
+                                   "Cortex-M7 ARM processors", []>;
+
+//===----------------------------------------------------------------------===//
+// ARM processors
+//
+// Dummy CPU, used to target architectures
+def : ProcessorModel<"generic",     CortexA8Model,      []>;
+
+// FIXME: Several processors below are not using their own scheduler
+// model, but one of similar/previous processor. These should be fixed.
+
+def : ProcNoItin<"arm8",                                [ARMv4]>;
+def : ProcNoItin<"arm810",                              [ARMv4]>;
+def : ProcNoItin<"strongarm",                           [ARMv4]>;
+def : ProcNoItin<"strongarm110",                        [ARMv4]>;
+def : ProcNoItin<"strongarm1100",                       [ARMv4]>;
+def : ProcNoItin<"strongarm1110",                       [ARMv4]>;
+
+def : ProcNoItin<"arm7tdmi",                            [ARMv4t]>;
+def : ProcNoItin<"arm7tdmi-s",                          [ARMv4t]>;
+def : ProcNoItin<"arm710t",                             [ARMv4t]>;
+def : ProcNoItin<"arm720t",                             [ARMv4t]>;
+def : ProcNoItin<"arm9",                                [ARMv4t]>;
+def : ProcNoItin<"arm9tdmi",                            [ARMv4t]>;
+def : ProcNoItin<"arm920",                              [ARMv4t]>;
+def : ProcNoItin<"arm920t",                             [ARMv4t]>;
+def : ProcNoItin<"arm922t",                             [ARMv4t]>;
+def : ProcNoItin<"arm940t",                             [ARMv4t]>;
+def : ProcNoItin<"ep9312",                              [ARMv4t]>;
+
+def : ProcNoItin<"arm10tdmi",                           [ARMv5t]>;
+def : ProcNoItin<"arm1020t",                            [ARMv5t]>;
+
+def : ProcNoItin<"arm9e",                               [ARMv5te]>;
+def : ProcNoItin<"arm926ej-s",                          [ARMv5te]>;
+def : ProcNoItin<"arm946e-s",                           [ARMv5te]>;
+def : ProcNoItin<"arm966e-s",                           [ARMv5te]>;
+def : ProcNoItin<"arm968e-s",                           [ARMv5te]>;
+def : ProcNoItin<"arm10e",                              [ARMv5te]>;
+def : ProcNoItin<"arm1020e",                            [ARMv5te]>;
+def : ProcNoItin<"arm1022e",                            [ARMv5te]>;
+def : ProcNoItin<"xscale",                              [ARMv5te]>;
+def : ProcNoItin<"iwmmxt",                              [ARMv5te]>;
+
+def : Processor<"arm1136j-s",       ARMV6Itineraries,   [ARMv6]>;
+def : Processor<"arm1136jf-s",      ARMV6Itineraries,   [ARMv6,
+                                                         FeatureVFP2,
+                                                         FeatureHasSlowFPVMLx]>;
+
+def : Processor<"cortex-m0",        ARMV6Itineraries,   [ARMv6m,
+                                                         FeatureHasNoBranchPredictor]>;
+def : Processor<"cortex-m0plus",    ARMV6Itineraries,   [ARMv6m,
+                                                         FeatureHasNoBranchPredictor]>;
+def : Processor<"cortex-m1",        ARMV6Itineraries,   [ARMv6m,
+                                                         FeatureHasNoBranchPredictor]>;
+def : Processor<"sc000",            ARMV6Itineraries,   [ARMv6m,
+                                                         FeatureHasNoBranchPredictor]>;
+
+def : Processor<"arm1176jz-s",      ARMV6Itineraries,   [ARMv6kz]>;
+def : Processor<"arm1176jzf-s",     ARMV6Itineraries,   [ARMv6kz,
+                                                         FeatureVFP2,
+                                                         FeatureHasSlowFPVMLx]>;
+
+def : Processor<"mpcorenovfp",      ARMV6Itineraries,   [ARMv6k]>;
+def : Processor<"mpcore",           ARMV6Itineraries,   [ARMv6k,
+                                                         FeatureVFP2,
+                                                         FeatureHasSlowFPVMLx]>;
+
+def : Processor<"arm1156t2-s",      ARMV6Itineraries,   [ARMv6t2]>;
+def : Processor<"arm1156t2f-s",     ARMV6Itineraries,   [ARMv6t2,
+                                                         FeatureVFP2,
+                                                         FeatureHasSlowFPVMLx]>;
+
+def : ProcessorModel<"cortex-a5",   CortexA8Model,      [ARMv7a, ProcA5,
+                                                         FeatureHasRetAddrStack,
+                                                         FeatureTrustZone,
+                                                         FeatureSlowFPBrcc,
+                                                         FeatureHasSlowFPVMLx,
+                                                         FeatureHasSlowFPVFMx,
+                                                         FeatureVMLxForwarding,
+                                                         FeatureMP,
+                                                         FeatureVFP4]>;
+
+def : ProcessorModel<"cortex-a7",   CortexA8Model,      [ARMv7a, ProcA7,
+                                                         FeatureHasRetAddrStack,
+                                                         FeatureTrustZone,
+                                                         FeatureSlowFPBrcc,
+                                                         FeatureHasVMLxHazards,
+                                                         FeatureHasSlowFPVMLx,
+                                                         FeatureHasSlowFPVFMx,
+                                                         FeatureVMLxForwarding,
+                                                         FeatureMP,
+                                                         FeatureVFP4,
+                                                         FeatureVirtualization]>;
+
+def : ProcessorModel<"cortex-a8",   CortexA8Model,      [ARMv7a, ProcA8,
+                                                         FeatureHasRetAddrStack,
+                                                         FeatureNonpipelinedVFP,
+                                                         FeatureTrustZone,
+                                                         FeatureSlowFPBrcc,
+                                                         FeatureHasVMLxHazards,
+                                                         FeatureHasSlowFPVMLx,
+                                                         FeatureHasSlowFPVFMx,
+                                                         FeatureVMLxForwarding]>;
+
+def : ProcessorModel<"cortex-a9",   CortexA9Model,      [ARMv7a, ProcA9,
+                                                         FeatureHasRetAddrStack,
+                                                         FeatureTrustZone,
+                                                         FeatureHasVMLxHazards,
+                                                         FeatureVMLxForwarding,
+                                                         FeatureFP16,
+                                                         FeatureAvoidPartialCPSR,
+                                                         FeatureExpandMLx,
+                                                         FeaturePreferVMOVSR,
+                                                         FeatureMuxedUnits,
+                                                         FeatureNEONForFPMovs,
+                                                         FeatureCheckVLDnAlign,
+                                                         FeatureMP]>;
+
+def : ProcessorModel<"cortex-a12",  CortexA9Model,      [ARMv7a, ProcA12,
+                                                         FeatureHasRetAddrStack,
+                                                         FeatureTrustZone,
+                                                         FeatureVMLxForwarding,
+                                                         FeatureVFP4,
+                                                         FeatureAvoidPartialCPSR,
+                                                         FeatureVirtualization,
+                                                         FeatureMP]>;
+
+def : ProcessorModel<"cortex-a15",  CortexA9Model,      [ARMv7a, ProcA15,
+                                                         FeatureDontWidenVMOVS,
+                                                         FeatureSplatVFPToNeon,
+                                                         FeatureHasRetAddrStack,
+                                                         FeatureMuxedUnits,
+                                                         FeatureTrustZone,
+                                                         FeatureVFP4,
+                                                         FeatureMP,
+                                                         FeatureCheckVLDnAlign,
+                                                         FeatureAvoidPartialCPSR,
+                                                         FeatureVirtualization]>;
+
+def : ProcessorModel<"cortex-a17",  CortexA9Model,      [ARMv7a, ProcA17,
+                                                         FeatureHasRetAddrStack,
+                                                         FeatureTrustZone,
+                                                         FeatureMP,
+                                                         FeatureVMLxForwarding,
+                                                         FeatureVFP4,
+                                                         FeatureAvoidPartialCPSR,
+                                                         FeatureVirtualization]>;
+
+// FIXME: krait has currently the same features as A9 plus VFP4 and  HWDiv
+def : ProcessorModel<"krait",       CortexA9Model,      [ARMv7a, ProcKrait,
+                                                         FeatureHasRetAddrStack,
+                                                         FeatureMuxedUnits,
+                                                         FeatureCheckVLDnAlign,
+                                                         FeatureVMLxForwarding,
+                                                         FeatureFP16,
+                                                         FeatureAvoidPartialCPSR,
+                                                         FeatureVFP4,
+                                                         FeatureHWDivThumb,
+                                                         FeatureHWDivARM]>;
+
+def : ProcessorModel<"swift",       SwiftModel,         [ARMv7a, ProcSwift,
+                                                         FeatureHasRetAddrStack,
+                                                         FeatureNEONForFP,
+                                                         FeatureVFP4,
+                                                         FeatureUseWideStrideVFP,
+                                                         FeatureMP,
+                                                         FeatureHWDivThumb,
+                                                         FeatureHWDivARM,
+                                                         FeatureAvoidPartialCPSR,
+                                                         FeatureAvoidMOVsShOp,
+                                                         FeatureHasSlowFPVMLx,
+                                                         FeatureHasSlowFPVFMx,
+                                                         FeatureHasVMLxHazards,
+                                                         FeatureProfUnpredicate,
+                                                         FeaturePrefISHSTBarrier,
+                                                         FeatureSlowOddRegister,
+                                                         FeatureSlowLoadDSubreg,
+                                                         FeatureSlowVGETLNi32,
+                                                         FeatureSlowVDUP32,
+                                                         FeatureUseMISched,
+                                                         FeatureNoPostRASched]>;
+
+def : ProcessorModel<"cortex-r4",   CortexA8Model,      [ARMv7r, ProcR4,
+                                                         FeatureHasRetAddrStack,
+                                                         FeatureAvoidPartialCPSR]>;
+
+def : ProcessorModel<"cortex-r4f",  CortexA8Model,      [ARMv7r, ProcR4,
+                                                         FeatureHasRetAddrStack,
+                                                         FeatureSlowFPBrcc,
+                                                         FeatureHasSlowFPVMLx,
+                                                         FeatureHasSlowFPVFMx,
+                                                         FeatureVFP3_D16,
+                                                         FeatureAvoidPartialCPSR]>;
+
+def : ProcessorModel<"cortex-r5",   CortexA8Model,      [ARMv7r, ProcR5,
+                                                         FeatureHasRetAddrStack,
+                                                         FeatureVFP3_D16,
+                                                         FeatureSlowFPBrcc,
+                                                         FeatureHWDivARM,
+                                                         FeatureHasSlowFPVMLx,
+                                                         FeatureHasSlowFPVFMx,
+                                                         FeatureAvoidPartialCPSR]>;
+
+def : ProcessorModel<"cortex-r7",   CortexA8Model,      [ARMv7r, ProcR7,
+                                                         FeatureHasRetAddrStack,
+                                                         FeatureVFP3_D16,
+                                                         FeatureFP16,
+                                                         FeatureMP,
+                                                         FeatureSlowFPBrcc,
+                                                         FeatureHWDivARM,
+                                                         FeatureHasSlowFPVMLx,
+                                                         FeatureHasSlowFPVFMx,
+                                                         FeatureAvoidPartialCPSR]>;
+
+def : ProcessorModel<"cortex-r8",   CortexA8Model,      [ARMv7r,
+                                                         FeatureHasRetAddrStack,
+                                                         FeatureVFP3_D16,
+                                                         FeatureFP16,
+                                                         FeatureMP,
+                                                         FeatureSlowFPBrcc,
+                                                         FeatureHWDivARM,
+                                                         FeatureHasSlowFPVMLx,
+                                                         FeatureHasSlowFPVFMx,
+                                                         FeatureAvoidPartialCPSR]>;
+
+def : ProcessorModel<"cortex-m3",   CortexM4Model,      [ARMv7m,
+                                                         ProcM3,
+                                                         FeaturePrefLoopAlign32,
+                                                         FeatureUseMISched,
+                                                         FeatureHasNoBranchPredictor]>;
+
+def : ProcessorModel<"sc300",       CortexM4Model,      [ARMv7m,
+                                                         ProcM3,
+                                                         FeatureUseMISched,
+                                                         FeatureHasNoBranchPredictor]>;
+
+def : ProcessorModel<"cortex-m4", CortexM4Model,        [ARMv7em,
+                                                         FeatureVFP4_D16_SP,
+                                                         FeaturePrefLoopAlign32,
+                                                         FeatureHasSlowFPVMLx,
+                                                         FeatureHasSlowFPVFMx,
+                                                         FeatureUseMISched,
+                                                         FeatureHasNoBranchPredictor]>;
+
+def : ProcessorModel<"cortex-m7", CortexM7Model,        [ARMv7em,
+                                                         ProcM7,
+                                                         FeatureFPARMv8_D16,
+                                                         FeatureUseMIPipeliner,
+                                                         FeatureUseMISched]>;
+
+def : ProcNoItin<"cortex-m23",                          [ARMv8mBaseline,
+                                                         FeatureNoMovt,
+                                                         FeatureHasNoBranchPredictor]>;
+
+def : ProcessorModel<"cortex-m33", CortexM4Model,       [ARMv8mMainline,
+                                                         FeatureDSP,
+                                                         FeatureFPARMv8_D16_SP,
+                                                         FeaturePrefLoopAlign32,
+                                                         FeatureHasSlowFPVMLx,
+                                                         FeatureHasSlowFPVFMx,
+                                                         FeatureUseMISched,
+                                                         FeatureHasNoBranchPredictor,
+                                                         FeatureFixCMSE_CVE_2021_35465]>;
+
+def : ProcessorModel<"cortex-m35p", CortexM4Model,      [ARMv8mMainline,
+                                                         FeatureDSP,
+                                                         FeatureFPARMv8_D16_SP,
+                                                         FeaturePrefLoopAlign32,
+                                                         FeatureHasSlowFPVMLx,
+                                                         FeatureHasSlowFPVFMx,
+                                                         FeatureUseMISched,
+                                                         FeatureHasNoBranchPredictor,
+                                                         FeatureFixCMSE_CVE_2021_35465]>;
+
+def : ProcessorModel<"cortex-m55", CortexM55Model,      [ARMv81mMainline,
+                                                         FeatureDSP,
+                                                         FeatureFPARMv8_D16,
+                                                         FeatureUseMISched,
+                                                         FeatureHasNoBranchPredictor,
+                                                         FeaturePrefLoopAlign32,
+                                                         FeatureHasSlowFPVMLx,
+                                                         HasMVEFloatOps,
+                                                         FeatureFixCMSE_CVE_2021_35465]>;
+
+def : ProcessorModel<"cortex-m85", CortexM85Model,      [ARMv81mMainline,
+                                                         FeatureDSP,
+                                                         FeatureFPARMv8_D16,
+                                                         FeaturePACBTI,
+                                                         FeatureUseMISched,
+                                                         HasMVEFloatOps]>;
+
+def : ProcessorModel<"cortex-m52", CortexM55Model,      [ARMv81mMainline,
+                                                         FeatureDSP,
+                                                         FeatureFPARMv8_D16,
+                                                         FeatureHasNoBranchPredictor,
+                                                         FeaturePACBTI,
+                                                         FeatureUseMISched,
+                                                         FeaturePrefLoopAlign32,
+                                                         FeatureHasSlowFPVMLx,
+                                                         FeatureMVEVectorCostFactor1,
+                                                         HasMVEFloatOps]>;
+
+def : ProcNoItin<"cortex-a32",                           [ARMv8a,
+                                                         FeatureHWDivThumb,
+                                                         FeatureHWDivARM,
+                                                         FeatureCrypto,
+                                                         FeatureCRC]>;
+
+def : ProcNoItin<"cortex-a35",                          [ARMv8a, ProcA35,
+                                                         FeatureHWDivThumb,
+                                                         FeatureHWDivARM,
+                                                         FeatureCrypto,
+                                                         FeatureCRC]>;
+
+def : ProcNoItin<"cortex-a53",                          [ARMv8a, ProcA53,
+                                                         FeatureHWDivThumb,
+                                                         FeatureHWDivARM,
+                                                         FeatureCrypto,
+                                                         FeatureCRC,
+                                                         FeatureFPAO]>;
+
+def : ProcNoItin<"cortex-a55",                          [ARMv82a, ProcA55,
+                                                         FeatureHWDivThumb,
+                                                         FeatureHWDivARM,
+                                                         FeatureDotProd]>;
+
+def : ProcessorModel<"cortex-a57",  CortexA57Model,     [ARMv8a, ProcA57,
+                                                         FeatureHWDivThumb,
+                                                         FeatureHWDivARM,
+                                                         FeatureCrypto,
+                                                         FeatureCRC,
+                                                         FeatureFPAO,
+                                                         FeatureAvoidPartialCPSR,
+                                                         FeatureCheapPredicableCPSR,
+                                                         FeatureFixCortexA57AES1742098]>;
+
+def : ProcessorModel<"cortex-a72",  CortexA57Model,     [ARMv8a, ProcA72,
+                                                         FeatureHWDivThumb,
+                                                         FeatureHWDivARM,
+                                                         FeatureCrypto,
+                                                         FeatureCRC,
+                                                         FeatureFixCortexA57AES1742098]>;
+
+def : ProcNoItin<"cortex-a73",                          [ARMv8a, ProcA73,
+                                                         FeatureHWDivThumb,
+                                                         FeatureHWDivARM,
+                                                         FeatureCrypto,
+                                                         FeatureCRC]>;
+
+def : ProcNoItin<"cortex-a75",                          [ARMv82a, ProcA75,
+                                                         FeatureHWDivThumb,
+                                                         FeatureHWDivARM,
+                                                         FeatureDotProd]>;
+
+def : ProcNoItin<"cortex-a76",                          [ARMv82a, ProcA76,
+                                                         FeatureHWDivThumb,
+                                                         FeatureHWDivARM,
+                                                         FeatureCrypto,
+                                                         FeatureCRC,
+                                                         FeatureFullFP16,
+                                                         FeatureDotProd]>;
+
+def : ProcNoItin<"cortex-a76ae",                        [ARMv82a, ProcA76,
+                                                         FeatureHWDivThumb,
+                                                         FeatureHWDivARM,
+                                                         FeatureCrypto,
+                                                         FeatureCRC,
+                                                         FeatureFullFP16,
+                                                         FeatureDotProd]>;
+
+def : ProcNoItin<"cortex-a77",                          [ARMv82a, ProcA77,
+                                                         FeatureHWDivThumb,
+                                                         FeatureHWDivARM,
+                                                         FeatureCrypto,
+                                                         FeatureCRC,
+                                                         FeatureFullFP16,
+                                                         FeatureDotProd]>;
+
+def : ProcNoItin<"cortex-a78",                          [ARMv82a, ProcA78,
+                                                         FeatureHWDivThumb,
+                                                         FeatureHWDivARM,
+                                                         FeatureCrypto,
+                                                         FeatureCRC,
+                                                         FeatureFullFP16,
+                                                         FeatureDotProd]>;
+
+def : ProcNoItin<"cortex-a78ae",                        [ARMv82a, ProcA78AE,
+                                                         FeatureHWDivThumb,
+                                                         FeatureHWDivARM,
+                                                         FeatureCrypto,
+                                                         FeatureCRC,
+                                                         FeatureFullFP16,
+                                                         FeatureDotProd]>;
+
+def : ProcNoItin<"cortex-a78c",                         [ARMv82a, ProcA78C,
+                                                         FeatureHWDivThumb,
+                                                         FeatureHWDivARM,
+                                                         FeatureCrypto,
+                                                         FeatureCRC,
+                                                         FeatureDotProd,
+                                                         FeatureFullFP16]>;
+
+def : ProcNoItin<"cortex-a710",                         [ARMv9a, ProcA710,
+                                                         FeatureHWDivThumb,
+                                                         FeatureHWDivARM,
+                                                         FeatureFP16FML,
+                                                         FeatureBF16,
+                                                         FeatureMatMulInt8,
+                                                         FeatureSB]>;
+
+def : ProcNoItin<"cortex-x1",                           [ARMv82a, ProcX1,
+                                                         FeatureHWDivThumb,
+                                                         FeatureHWDivARM,
+                                                         FeatureCrypto,
+                                                         FeatureCRC,
+                                                         FeatureFullFP16,
+                                                         FeatureDotProd]>;
+
+def : ProcNoItin<"cortex-x1c",                          [ARMv82a, ProcX1C,
+                                                         FeatureHWDivThumb,
+                                                         FeatureHWDivARM,
+                                                         FeatureCrypto,
+                                                         FeatureCRC,
+                                                         FeatureFullFP16,
+                                                         FeatureDotProd]>;
+
+def : ProcNoItin<"neoverse-v1",                         [ARMv84a,
+                                                         FeatureHWDivThumb,
+                                                         FeatureHWDivARM,
+                                                         FeatureCrypto,
+                                                         FeatureCRC,
+                                                         FeatureFullFP16,
+                                                         FeatureBF16,
+                                                         FeatureMatMulInt8]>;
+
+def : ProcNoItin<"neoverse-n1",                         [ARMv82a,
+                                                         FeatureHWDivThumb,
+                                                         FeatureHWDivARM,
+                                                         FeatureCrypto,
+                                                         FeatureCRC,
+                                                         FeatureDotProd]>;
+
+def : ProcNoItin<"neoverse-n2",                         [ARMv9a,
+                                                         FeatureBF16,
+                                                         FeatureFP16FML,
+                                                         FeatureMatMulInt8]>;
+
+def : ProcessorModel<"cyclone",     SwiftModel,         [ARMv8a, ProcSwift,
+                                                         FeatureHasRetAddrStack,
+                                                         FeatureNEONForFP,
+                                                         FeatureVFP4,
+                                                         FeatureMP,
+                                                         FeatureHWDivThumb,
+                                                         FeatureHWDivARM,
+                                                         FeatureAvoidPartialCPSR,
+                                                         FeatureAvoidMOVsShOp,
+                                                         FeatureHasSlowFPVMLx,
+                                                         FeatureHasSlowFPVFMx,
+                                                         FeatureCrypto,
+                                                         FeatureUseMISched,
+                                                         FeatureZCZeroing,
+                                                         FeatureNoPostRASched]>;
+
+def : ProcNoItin<"exynos-m3",                           [ARMv8a, ProcExynos]>;
+def : ProcNoItin<"exynos-m4",                           [ARMv82a, ProcExynos,
+                                                         FeatureFullFP16,
+                                                         FeatureDotProd]>;
+def : ProcNoItin<"exynos-m5",                           [ARMv82a, ProcExynos,
+                                                         FeatureFullFP16,
+                                                         FeatureDotProd]>;
+
+def : ProcNoItin<"kryo",                                [ARMv8a, ProcKryo,
+                                                         FeatureHWDivThumb,
+                                                         FeatureHWDivARM,
+                                                         FeatureCrypto,
+                                                         FeatureCRC]>;
+
+def : ProcessorModel<"cortex-r52", CortexR52Model,      [ARMv8r, ProcR52,
+                                                         FeatureUseMISched,
+                                                         FeatureFPAO]>;
diff --git a/llvm/lib/Target/ARM/ARMSubtarget.cpp b/llvm/lib/Target/ARM/ARMSubtarget.cpp
index 04ba20a17187..5e13d8fabe04 100644
--- a/llvm/lib/Target/ARM/ARMSubtarget.cpp
+++ b/llvm/lib/Target/ARM/ARMSubtarget.cpp
@@ -293,13 +293,11 @@ void ARMSubtarget::initSubtargetFeatures(StringRef CPU, StringRef FS) {
   case CortexA78C:
   case CortexA710:
   case CortexR4:
-  case CortexR4F:
   case CortexR5:
   case CortexR7:
   case CortexM3:
   case CortexM7:
   case CortexR52:
-  case CortexM52:
   case CortexX1:
   case CortexX1C:
     break;
@@ -314,8 +312,6 @@ void ARMSubtarget::initSubtargetFeatures(StringRef CPU, StringRef FS) {
   case Krait:
     PreISelOperandLatencyAdjustment = 1;
     break;
-  case NeoverseN1:
-  case NeoverseN2:
   case NeoverseV1:
     break;
   case Swift:
diff --git a/llvm/lib/Target/ARM/ARMSubtarget.h b/llvm/lib/Target/ARM/ARMSubtarget.h
index 497ae160fde2..00239ff94b7b 100644
--- a/llvm/lib/Target/ARM/ARMSubtarget.h
+++ b/llvm/lib/Target/ARM/ARMSubtarget.h
@@ -49,45 +49,9 @@ class ARMSubtarget : public ARMGenSubtargetInfo {
 protected:
   enum ARMProcFamilyEnum {
     Others,
-
-    CortexA12,
-    CortexA15,
-    CortexA17,
-    CortexA32,
-    CortexA35,
-    CortexA5,
-    CortexA53,
-    CortexA55,
-    CortexA57,
-    CortexA7,
-    CortexA72,
-    CortexA73,
-    CortexA75,
-    CortexA76,
-    CortexA77,
-    CortexA78,
-    CortexA78AE,
-    CortexA78C,
-    CortexA710,
-    CortexA8,
-    CortexA9,
-    CortexM3,
-    CortexM7,
-    CortexM52,
-    CortexR4,
-    CortexR4F,
-    CortexR5,
-    CortexR52,
-    CortexR7,
-    CortexX1,
-    CortexX1C,
-    Exynos,
-    Krait,
-    Kryo,
-    NeoverseN1,
-    NeoverseN2,
-    NeoverseV1,
-    Swift
+#define ARM_PROCESSOR_FAMILY(ENUM) ENUM,
+#include "llvm/TargetParser/ARMTargetParserDef.inc"
+#undef ARM_PROCESSOR_FAMILY
   };
   enum ARMProcClassEnum {
     None,
@@ -97,43 +61,9 @@ protected:
     RClass
   };
   enum ARMArchEnum {
-    ARMv4,
-    ARMv4t,
-    ARMv5,
-    ARMv5t,
-    ARMv5te,
-    ARMv5tej,
-    ARMv6,
-    ARMv6k,
-    ARMv6kz,
-    ARMv6m,
-    ARMv6sm,
-    ARMv6t2,
-    ARMv7a,
-    ARMv7em,
-    ARMv7m,
-    ARMv7r,
-    ARMv7ve,
-    ARMv81a,
-    ARMv82a,
-    ARMv83a,
-    ARMv84a,
-    ARMv85a,
-    ARMv86a,
-    ARMv87a,
-    ARMv88a,
-    ARMv89a,
-    ARMv8a,
-    ARMv8mBaseline,
-    ARMv8mMainline,
-    ARMv8r,
-    ARMv81mMainline,
-    ARMv9a,
-    ARMv91a,
-    ARMv92a,
-    ARMv93a,
-    ARMv94a,
-    ARMv95a,
+#define ARM_ARCHITECTURE(ENUM) ENUM,
+#include "llvm/TargetParser/ARMTargetParserDef.inc"
+#undef ARM_ARCHITECTURE
   };
 
 public:
diff --git a/llvm/lib/Target/ARM/ARMTargetTransformInfo.h b/llvm/lib/Target/ARM/ARMTargetTransformInfo.h
index 04b32194f806..58eab45b9641 100644
--- a/llvm/lib/Target/ARM/ARMTargetTransformInfo.h
+++ b/llvm/lib/Target/ARM/ARMTargetTransformInfo.h
@@ -256,7 +256,7 @@ public:
       unsigned Opcode, Type *Ty, TTI::TargetCostKind CostKind,
       TTI::OperandValueInfo Op1Info = {TTI::OK_AnyValue, TTI::OP_None},
       TTI::OperandValueInfo Op2Info = {TTI::OK_AnyValue, TTI::OP_None},
-      ArrayRef<const Value *> Args = ArrayRef<const Value *>(),
+      ArrayRef<const Value *> Args = std::nullopt,
       const Instruction *CxtI = nullptr);
 
   InstructionCost
diff --git a/llvm/lib/Target/ARM/AsmParser/ARMAsmParser.cpp b/llvm/lib/Target/ARM/AsmParser/ARMAsmParser.cpp
index 028db9d17e30..e54314cc7d00 100644
--- a/llvm/lib/Target/ARM/AsmParser/ARMAsmParser.cpp
+++ b/llvm/lib/Target/ARM/AsmParser/ARMAsmParser.cpp
@@ -11104,7 +11104,7 @@ ARMAsmParser::checkEarlyTargetMatchPredicate(MCInst &Inst,
       return Match_MnemonicFail;
     }
   }
-    LLVM_FALLTHROUGH;
+    [[fallthrough]];
   default:
     return Match_Success;
   }
diff --git a/llvm/lib/Target/ARM/MCTargetDesc/ARMELFStreamer.cpp b/llvm/lib/Target/ARM/MCTargetDesc/ARMELFStreamer.cpp
index 1d80af590d16..afd7dccbeca9 100644
--- a/llvm/lib/Target/ARM/MCTargetDesc/ARMELFStreamer.cpp
+++ b/llvm/lib/Target/ARM/MCTargetDesc/ARMELFStreamer.cpp
@@ -1487,8 +1487,7 @@ MCELFStreamer *createARMELFStreamer(MCContext &Context,
                                     std::unique_ptr<MCAsmBackend> TAB,
                                     std::unique_ptr<MCObjectWriter> OW,
                                     std::unique_ptr<MCCodeEmitter> Emitter,
-                                    bool RelaxAll, bool IsThumb,
-                                    bool IsAndroid) {
+                                    bool IsThumb, bool IsAndroid) {
   ARMELFStreamer *S =
       new ARMELFStreamer(Context, std::move(TAB), std::move(OW),
                          std::move(Emitter), IsThumb, IsAndroid);
@@ -1497,8 +1496,6 @@ MCELFStreamer *createARMELFStreamer(MCContext &Context,
   // the status quo for ARM and setting EF_ARM_EABI_VER5 as the default.
   S->getAssembler().setELFHeaderEFlags(ELF::EF_ARM_EABI_VER5);
 
-  if (RelaxAll)
-    S->getAssembler().setRelaxAll(true);
   return S;
 }
 
diff --git a/llvm/lib/Target/ARM/MCTargetDesc/ARMMCTargetDesc.cpp b/llvm/lib/Target/ARM/MCTargetDesc/ARMMCTargetDesc.cpp
index 8d9959a9457d..20603b6cf1b0 100644
--- a/llvm/lib/Target/ARM/MCTargetDesc/ARMMCTargetDesc.cpp
+++ b/llvm/lib/Target/ARM/MCTargetDesc/ARMMCTargetDesc.cpp
@@ -359,10 +359,9 @@ static MCAsmInfo *createARMMCAsmInfo(const MCRegisterInfo &MRI,
 static MCStreamer *createELFStreamer(const Triple &T, MCContext &Ctx,
                                      std::unique_ptr<MCAsmBackend> &&MAB,
                                      std::unique_ptr<MCObjectWriter> &&OW,
-                                     std::unique_ptr<MCCodeEmitter> &&Emitter,
-                                     bool RelaxAll) {
+                                     std::unique_ptr<MCCodeEmitter> &&Emitter) {
   return createARMELFStreamer(
-      Ctx, std::move(MAB), std::move(OW), std::move(Emitter), false,
+      Ctx, std::move(MAB), std::move(OW), std::move(Emitter),
       (T.getArch() == Triple::thumb || T.getArch() == Triple::thumbeb),
       T.isAndroid());
 }
@@ -370,10 +369,10 @@ static MCStreamer *createELFStreamer(const Triple &T, MCContext &Ctx,
 static MCStreamer *
 createARMMachOStreamer(MCContext &Ctx, std::unique_ptr<MCAsmBackend> &&MAB,
                        std::unique_ptr<MCObjectWriter> &&OW,
-                       std::unique_ptr<MCCodeEmitter> &&Emitter, bool RelaxAll,
+                       std::unique_ptr<MCCodeEmitter> &&Emitter,
                        bool DWARFMustBeAtTheEnd) {
   return createMachOStreamer(Ctx, std::move(MAB), std::move(OW),
-                             std::move(Emitter), false, DWARFMustBeAtTheEnd);
+                             std::move(Emitter), DWARFMustBeAtTheEnd);
 }
 
 static MCInstPrinter *createARMMCInstPrinter(const Triple &T,
diff --git a/llvm/lib/Target/ARM/MCTargetDesc/ARMMCTargetDesc.h b/llvm/lib/Target/ARM/MCTargetDesc/ARMMCTargetDesc.h
index 3066d9ba6783..a673d590419e 100644
--- a/llvm/lib/Target/ARM/MCTargetDesc/ARMMCTargetDesc.h
+++ b/llvm/lib/Target/ARM/MCTargetDesc/ARMMCTargetDesc.h
@@ -94,7 +94,6 @@ MCStreamer *createARMWinCOFFStreamer(MCContext &Context,
                                      std::unique_ptr<MCAsmBackend> &&MAB,
                                      std::unique_ptr<MCObjectWriter> &&OW,
                                      std::unique_ptr<MCCodeEmitter> &&Emitter,
-                                     bool RelaxAll,
                                      bool IncrementalLinkerCompatible);
 
 /// Construct an ELF Mach-O object writer.
diff --git a/llvm/lib/Target/ARM/MCTargetDesc/ARMWinCOFFStreamer.cpp b/llvm/lib/Target/ARM/MCTargetDesc/ARMWinCOFFStreamer.cpp
index cdd7f6fb715a..0fcf6eb1a5ab 100644
--- a/llvm/lib/Target/ARM/MCTargetDesc/ARMWinCOFFStreamer.cpp
+++ b/llvm/lib/Target/ARM/MCTargetDesc/ARMWinCOFFStreamer.cpp
@@ -66,11 +66,12 @@ void ARMWinCOFFStreamer::finishImpl() {
 }
 }
 
-MCStreamer *llvm::createARMWinCOFFStreamer(
-    MCContext &Context, std::unique_ptr<MCAsmBackend> &&MAB,
-    std::unique_ptr<MCObjectWriter> &&OW,
-    std::unique_ptr<MCCodeEmitter> &&Emitter, bool RelaxAll,
-    bool IncrementalLinkerCompatible) {
+MCStreamer *
+llvm::createARMWinCOFFStreamer(MCContext &Context,
+                               std::unique_ptr<MCAsmBackend> &&MAB,
+                               std::unique_ptr<MCObjectWriter> &&OW,
+                               std::unique_ptr<MCCodeEmitter> &&Emitter,
+                               bool IncrementalLinkerCompatible) {
   auto *S = new ARMWinCOFFStreamer(Context, std::move(MAB), std::move(Emitter),
                                    std::move(OW));
   S->getAssembler().setIncrementalLinkerCompatible(IncrementalLinkerCompatible);
diff --git a/llvm/lib/Target/ARM/MVETPAndVPTOptimisationsPass.cpp b/llvm/lib/Target/ARM/MVETPAndVPTOptimisationsPass.cpp
index e8d2cba7ee55..c9bbc41ac13b 100644
--- a/llvm/lib/Target/ARM/MVETPAndVPTOptimisationsPass.cpp
+++ b/llvm/lib/Target/ARM/MVETPAndVPTOptimisationsPass.cpp
@@ -667,17 +667,18 @@ static bool MoveVPNOTBeforeFirstUser(MachineBasicBlock &MBB,
   MachineOperand *VPNOTOperandKiller = nullptr;
   for (; Iter != MBB.end(); ++Iter) {
     if (MachineOperand *MO =
-            Iter->findRegisterUseOperand(VPNOTOperand, /*isKill*/ true)) {
+            Iter->findRegisterUseOperand(VPNOTOperand, /*TRI=*/nullptr,
+                                         /*isKill*/ true)) {
       // If we find the operand that kills the VPNOTOperand's result, save it.
       VPNOTOperandKiller = MO;
     }
 
-    if (Iter->findRegisterUseOperandIdx(Reg) != -1) {
+    if (Iter->findRegisterUseOperandIdx(Reg, /*TRI=*/nullptr) != -1) {
       MustMove = true;
       continue;
     }
 
-    if (Iter->findRegisterUseOperandIdx(VPNOTResult) == -1)
+    if (Iter->findRegisterUseOperandIdx(VPNOTResult, /*TRI=*/nullptr) == -1)
       continue;
 
     HasUser = true;
@@ -731,7 +732,7 @@ bool MVETPAndVPTOptimisations::ReduceOldVCCRValueUses(MachineBasicBlock &MBB) {
       // If we already have a VCCRValue, and this is a VPNOT on VCCRValue, we've
       // found what we were looking for.
       if (VCCRValue && Iter->getOpcode() == ARM::MVE_VPNOT &&
-          Iter->findRegisterUseOperandIdx(VCCRValue) != -1) {
+          Iter->findRegisterUseOperandIdx(VCCRValue, /*TRI=*/nullptr) != -1) {
         // Move the VPNOT closer to its first user if needed, and ignore if it
         // has no users.
         if (!MoveVPNOTBeforeFirstUser(MBB, Iter, VCCRValue))
@@ -763,7 +764,8 @@ bool MVETPAndVPTOptimisations::ReduceOldVCCRValueUses(MachineBasicBlock &MBB) {
     for (; Iter != End; ++Iter) {
       bool IsInteresting = false;
 
-      if (MachineOperand *MO = Iter->findRegisterUseOperand(VCCRValue)) {
+      if (MachineOperand *MO =
+              Iter->findRegisterUseOperand(VCCRValue, /*TRI=*/nullptr)) {
         IsInteresting = true;
 
         // - If the instruction is a VPNOT, it can be removed, and we can just
@@ -794,8 +796,8 @@ bool MVETPAndVPTOptimisations::ReduceOldVCCRValueUses(MachineBasicBlock &MBB) {
       } else {
         // If the instr uses OppositeVCCRValue, make it use LastVPNOTResult
         // instead as they contain the same value.
-        if (MachineOperand *MO =
-                Iter->findRegisterUseOperand(OppositeVCCRValue)) {
+        if (MachineOperand *MO = Iter->findRegisterUseOperand(
+                OppositeVCCRValue, /*TRI=*/nullptr)) {
           IsInteresting = true;
 
           // This is pointless if LastVPNOTResult == OppositeVCCRValue.
@@ -855,8 +857,9 @@ bool MVETPAndVPTOptimisations::ReplaceVCMPsByVPNOTs(MachineBasicBlock &MBB) {
 
   for (MachineInstr &Instr : MBB.instrs()) {
     if (PrevVCMP) {
-      if (MachineOperand *MO = Instr.findRegisterUseOperand(
-              PrevVCMP->getOperand(0).getReg(), /*isKill*/ true)) {
+      if (MachineOperand *MO =
+              Instr.findRegisterUseOperand(PrevVCMP->getOperand(0).getReg(),
+                                           /*TRI=*/nullptr, /*isKill*/ true)) {
         // If we come accross the instr that kills PrevVCMP's result, record it
         // so we can remove the kill flag later if we need to.
         PrevVCMPResultKiller = MO;
diff --git a/llvm/lib/Target/ARM/MVEVPTBlockPass.cpp b/llvm/lib/Target/ARM/MVEVPTBlockPass.cpp
index d2b0bcf1250f..197eca421e21 100644
--- a/llvm/lib/Target/ARM/MVEVPTBlockPass.cpp
+++ b/llvm/lib/Target/ARM/MVEVPTBlockPass.cpp
@@ -131,7 +131,8 @@ static bool StepOverPredicatedInstrs(MachineBasicBlock::instr_iterator &Iter,
 static bool IsVPRDefinedOrKilledByBlock(MachineBasicBlock::iterator Iter,
                                         MachineBasicBlock::iterator End) {
   for (; Iter != End; ++Iter)
-    if (Iter->definesRegister(ARM::VPR) || Iter->killsRegister(ARM::VPR))
+    if (Iter->definesRegister(ARM::VPR, /*TRI=*/nullptr) ||
+        Iter->killsRegister(ARM::VPR, /*TRI=*/nullptr))
       return true;
   return false;
 }
diff --git a/llvm/lib/Target/ARM/Thumb2ITBlockPass.cpp b/llvm/lib/Target/ARM/Thumb2ITBlockPass.cpp
index 2945b5eaae3e..147bf751945d 100644
--- a/llvm/lib/Target/ARM/Thumb2ITBlockPass.cpp
+++ b/llvm/lib/Target/ARM/Thumb2ITBlockPass.cpp
@@ -269,7 +269,8 @@ bool Thumb2ITBlock::InsertITInstructions(MachineBasicBlock &MBB) {
     MIB.addImm(Mask);
 
     // Last instruction in IT block kills ITSTATE.
-    LastITMI->findRegisterUseOperand(ARM::ITSTATE)->setIsKill();
+    LastITMI->findRegisterUseOperand(ARM::ITSTATE, /*TRI=*/nullptr)
+        ->setIsKill();
 
     // Finalize the bundle.
     finalizeBundle(MBB, InsertPos.getInstrIterator(),
diff --git a/llvm/lib/Target/ARM/Thumb2InstrInfo.cpp b/llvm/lib/Target/ARM/Thumb2InstrInfo.cpp
index fc2834cb0b45..8ef5c3d9d6ba 100644
--- a/llvm/lib/Target/ARM/Thumb2InstrInfo.cpp
+++ b/llvm/lib/Target/ARM/Thumb2InstrInfo.cpp
@@ -571,7 +571,7 @@ bool llvm::rewriteT2FrameIndex(MachineInstr &MI, unsigned FrameRegIdx,
 
     Register PredReg;
     if (Offset == 0 && getInstrPredicate(MI, PredReg) == ARMCC::AL &&
-        !MI.definesRegister(ARM::CPSR)) {
+        !MI.definesRegister(ARM::CPSR, /*TRI=*/nullptr)) {
       // Turn it into a move.
       MI.setDesc(TII.get(ARM::tMOVr));
       MI.getOperand(FrameRegIdx).ChangeToRegister(FrameReg, false);
diff --git a/llvm/lib/Target/ARM/Thumb2SizeReduction.cpp b/llvm/lib/Target/ARM/Thumb2SizeReduction.cpp
index 286010e2ba23..f572af986007 100644
--- a/llvm/lib/Target/ARM/Thumb2SizeReduction.cpp
+++ b/llvm/lib/Target/ARM/Thumb2SizeReduction.cpp
@@ -1097,12 +1097,13 @@ bool Thumb2SizeReduce::ReduceMBB(MachineBasicBlock &MBB,
       // marker is only on the BUNDLE instruction. Process the BUNDLE
       // instruction as we finish with the bundled instruction to work around
       // the inconsistency.
-      if (BundleMI->killsRegister(ARM::CPSR))
+      if (BundleMI->killsRegister(ARM::CPSR, /*TRI=*/nullptr))
         LiveCPSR = false;
-      MachineOperand *MO = BundleMI->findRegisterDefOperand(ARM::CPSR);
+      MachineOperand *MO =
+          BundleMI->findRegisterDefOperand(ARM::CPSR, /*TRI=*/nullptr);
       if (MO && !MO->isDead())
         LiveCPSR = true;
-      MO = BundleMI->findRegisterUseOperand(ARM::CPSR);
+      MO = BundleMI->findRegisterUseOperand(ARM::CPSR, /*TRI=*/nullptr);
       if (MO && !MO->isKill())
         LiveCPSR = true;
     }
diff --git a/llvm/lib/Target/AVR/MCTargetDesc/AVRMCTargetDesc.cpp b/llvm/lib/Target/AVR/MCTargetDesc/AVRMCTargetDesc.cpp
index ba370261e284..119baff83dae 100644
--- a/llvm/lib/Target/AVR/MCTargetDesc/AVRMCTargetDesc.cpp
+++ b/llvm/lib/Target/AVR/MCTargetDesc/AVRMCTargetDesc.cpp
@@ -72,10 +72,9 @@ static MCInstPrinter *createAVRMCInstPrinter(const Triple &T,
 static MCStreamer *createMCStreamer(const Triple &T, MCContext &Context,
                                     std::unique_ptr<MCAsmBackend> &&MAB,
                                     std::unique_ptr<MCObjectWriter> &&OW,
-                                    std::unique_ptr<MCCodeEmitter> &&Emitter,
-                                    bool RelaxAll) {
+                                    std::unique_ptr<MCCodeEmitter> &&Emitter) {
   return createELFStreamer(Context, std::move(MAB), std::move(OW),
-                           std::move(Emitter), RelaxAll);
+                           std::move(Emitter));
 }
 
 static MCTargetStreamer *
diff --git a/llvm/lib/Target/BPF/BPFTargetTransformInfo.h b/llvm/lib/Target/BPF/BPFTargetTransformInfo.h
index 5aa9ec283406..0ecff32a7525 100644
--- a/llvm/lib/Target/BPF/BPFTargetTransformInfo.h
+++ b/llvm/lib/Target/BPF/BPFTargetTransformInfo.h
@@ -59,14 +59,14 @@ public:
       unsigned Opcode, Type *Ty, TTI::TargetCostKind CostKind,
       TTI::OperandValueInfo Op1Info = {TTI::OK_AnyValue, TTI::OP_None},
       TTI::OperandValueInfo Op2Info = {TTI::OK_AnyValue, TTI::OP_None},
-    ArrayRef<const Value *> Args = ArrayRef<const Value *>(),
-    const Instruction *CxtI = nullptr) {
-      int ISD = TLI->InstructionOpcodeToISD(Opcode);
-      if (ISD == ISD::ADD && CostKind == TTI::TCK_RecipThroughput)
-        return SCEVCheapExpansionBudget.getValue() + 1;
-
-      return BaseT::getArithmeticInstrCost(Opcode, Ty, CostKind, Op1Info,
-                                           Op2Info);
+      ArrayRef<const Value *> Args = std::nullopt,
+      const Instruction *CxtI = nullptr) {
+    int ISD = TLI->InstructionOpcodeToISD(Opcode);
+    if (ISD == ISD::ADD && CostKind == TTI::TCK_RecipThroughput)
+      return SCEVCheapExpansionBudget.getValue() + 1;
+
+    return BaseT::getArithmeticInstrCost(Opcode, Ty, CostKind, Op1Info,
+                                         Op2Info);
   }
 
   TTI::MemCmpExpansionOptions enableMemCmpExpansion(bool OptSize,
diff --git a/llvm/lib/Target/BPF/BTFDebug.cpp b/llvm/lib/Target/BPF/BTFDebug.cpp
index ebd8447eba85..8c9f5c4dc554 100644
--- a/llvm/lib/Target/BPF/BTFDebug.cpp
+++ b/llvm/lib/Target/BPF/BTFDebug.cpp
@@ -973,8 +973,7 @@ void BTFDebug::visitMapDefType(const DIType *Ty, uint32_t &TypeId) {
 }
 
 /// Read file contents from the actual file or from the source
-std::string BTFDebug::populateFileContent(const DISubprogram *SP) {
-  auto File = SP->getFile();
+std::string BTFDebug::populateFileContent(const DIFile *File) {
   std::string FileName;
 
   if (!File->getFilename().starts_with("/") && File->getDirectory().size())
@@ -1005,9 +1004,9 @@ std::string BTFDebug::populateFileContent(const DISubprogram *SP) {
   return FileName;
 }
 
-void BTFDebug::constructLineInfo(const DISubprogram *SP, MCSymbol *Label,
+void BTFDebug::constructLineInfo(MCSymbol *Label, const DIFile *File,
                                  uint32_t Line, uint32_t Column) {
-  std::string FileName = populateFileContent(SP);
+  std::string FileName = populateFileContent(File);
   BTFLineInfo LineInfo;
 
   LineInfo.Label = Label;
@@ -1366,10 +1365,10 @@ void BTFDebug::beginInstruction(const MachineInstr *MI) {
   if (!CurMI) // no debug info
     return;
 
-  // Skip this instruction if no DebugLoc or the DebugLoc
-  // is the same as the previous instruction.
+  // Skip this instruction if no DebugLoc, the DebugLoc
+  // is the same as the previous instruction or Line is 0.
   const DebugLoc &DL = MI->getDebugLoc();
-  if (!DL || PrevInstLoc == DL) {
+  if (!DL || PrevInstLoc == DL || DL.getLine() == 0) {
     // This instruction will be skipped, no LineInfo has
     // been generated, construct one based on function signature.
     if (LineInfoGenerated == false) {
@@ -1377,7 +1376,7 @@ void BTFDebug::beginInstruction(const MachineInstr *MI) {
       if (!S)
         return;
       MCSymbol *FuncLabel = Asm->getFunctionBegin();
-      constructLineInfo(S, FuncLabel, S->getLine(), 0);
+      constructLineInfo(FuncLabel, S->getFile(), S->getLine(), 0);
       LineInfoGenerated = true;
     }
 
@@ -1389,8 +1388,7 @@ void BTFDebug::beginInstruction(const MachineInstr *MI) {
   OS.emitLabel(LineSym);
 
   // Construct the lineinfo.
-  auto SP = DL->getScope()->getSubprogram();
-  constructLineInfo(SP, LineSym, DL.getLine(), DL.getCol());
+  constructLineInfo(LineSym, DL->getFile(), DL.getLine(), DL.getCol());
 
   LineInfoGenerated = true;
   PrevInstLoc = DL;
diff --git a/llvm/lib/Target/BPF/BTFDebug.h b/llvm/lib/Target/BPF/BTFDebug.h
index 7536006ed21c..11a0c59ba6c9 100644
--- a/llvm/lib/Target/BPF/BTFDebug.h
+++ b/llvm/lib/Target/BPF/BTFDebug.h
@@ -343,10 +343,10 @@ class BTFDebug : public DebugHandlerBase {
 
   /// Get the file content for the subprogram. Certain lines of the file
   /// later may be put into string table and referenced by line info.
-  std::string populateFileContent(const DISubprogram *SP);
+  std::string populateFileContent(const DIFile *File);
 
   /// Construct a line info.
-  void constructLineInfo(const DISubprogram *SP, MCSymbol *Label, uint32_t Line,
+  void constructLineInfo(MCSymbol *Label, const DIFile *File, uint32_t Line,
                          uint32_t Column);
 
   /// Generate types and variables for globals.
diff --git a/llvm/lib/Target/BPF/MCTargetDesc/BPFMCTargetDesc.cpp b/llvm/lib/Target/BPF/MCTargetDesc/BPFMCTargetDesc.cpp
index 44932383fb43..caf84701b999 100644
--- a/llvm/lib/Target/BPF/MCTargetDesc/BPFMCTargetDesc.cpp
+++ b/llvm/lib/Target/BPF/MCTargetDesc/BPFMCTargetDesc.cpp
@@ -50,13 +50,13 @@ static MCSubtargetInfo *createBPFMCSubtargetInfo(const Triple &TT,
   return createBPFMCSubtargetInfoImpl(TT, CPU, /*TuneCPU*/ CPU, FS);
 }
 
-static MCStreamer *createBPFMCStreamer(const Triple &T, MCContext &Ctx,
-                                       std::unique_ptr<MCAsmBackend> &&MAB,
-                                       std::unique_ptr<MCObjectWriter> &&OW,
-                                       std::unique_ptr<MCCodeEmitter> &&Emitter,
-                                       bool RelaxAll) {
-  return createELFStreamer(Ctx, std::move(MAB), std::move(OW), std::move(Emitter),
-                           RelaxAll);
+static MCStreamer *
+createBPFMCStreamer(const Triple &T, MCContext &Ctx,
+                    std::unique_ptr<MCAsmBackend> &&MAB,
+                    std::unique_ptr<MCObjectWriter> &&OW,
+                    std::unique_ptr<MCCodeEmitter> &&Emitter) {
+  return createELFStreamer(Ctx, std::move(MAB), std::move(OW),
+                           std::move(Emitter));
 }
 
 static MCInstPrinter *createBPFMCInstPrinter(const Triple &T,
diff --git a/llvm/lib/Target/CSKY/MCTargetDesc/CSKYMCTargetDesc.cpp b/llvm/lib/Target/CSKY/MCTargetDesc/CSKYMCTargetDesc.cpp
index 64f01cd1c9fa..c3403ade389c 100644
--- a/llvm/lib/Target/CSKY/MCTargetDesc/CSKYMCTargetDesc.cpp
+++ b/llvm/lib/Target/CSKY/MCTargetDesc/CSKYMCTargetDesc.cpp
@@ -88,13 +88,10 @@ createCSKYObjectTargetStreamer(MCStreamer &S, const MCSubtargetInfo &STI) {
 static MCStreamer *createELFStreamer(const Triple &T, MCContext &Ctx,
                                      std::unique_ptr<MCAsmBackend> &&MAB,
                                      std::unique_ptr<MCObjectWriter> &&OW,
-                                     std::unique_ptr<MCCodeEmitter> &&Emitter,
-                                     bool RelaxAll) {
+                                     std::unique_ptr<MCCodeEmitter> &&Emitter) {
   CSKYELFStreamer *S = new CSKYELFStreamer(Ctx, std::move(MAB), std::move(OW),
                                            std::move(Emitter));
 
-  if (RelaxAll)
-    S->getAssembler().setRelaxAll(true);
   return S;
 }
 
diff --git a/llvm/lib/Target/DirectX/DXILIntrinsicExpansion.cpp b/llvm/lib/Target/DirectX/DXILIntrinsicExpansion.cpp
index 4d99bc006900..4b162a35365c 100644
--- a/llvm/lib/Target/DirectX/DXILIntrinsicExpansion.cpp
+++ b/llvm/lib/Target/DirectX/DXILIntrinsicExpansion.cpp
@@ -77,8 +77,8 @@ static bool expandIntegerDot(CallInst *Orig, Intrinsic::ID DotIntrinsic) {
                                    : Intrinsic::dx_umad;
   Value *A = Orig->getOperand(0);
   Value *B = Orig->getOperand(1);
-  Type *ATy = A->getType();
-  Type *BTy = B->getType();
+  [[maybe_unused]] Type *ATy = A->getType();
+  [[maybe_unused]] Type *BTy = B->getType();
   assert(ATy->isVectorTy() && BTy->isVectorTy());
 
   IRBuilder<> Builder(Orig->getParent());
diff --git a/llvm/lib/Target/Hexagon/CMakeLists.txt b/llvm/lib/Target/Hexagon/CMakeLists.txt
index cdc062eee72b..9e4ca08aea40 100644
--- a/llvm/lib/Target/Hexagon/CMakeLists.txt
+++ b/llvm/lib/Target/Hexagon/CMakeLists.txt
@@ -26,6 +26,7 @@ add_llvm_target(HexagonCodeGen
   HexagonCommonGEP.cpp
   HexagonConstExtenders.cpp
   HexagonConstPropagation.cpp
+  HexagonCopyHoisting.cpp
   HexagonCopyToCombine.cpp
   HexagonEarlyIfConv.cpp
   HexagonExpandCondsets.cpp
diff --git a/llvm/lib/Target/Hexagon/HexagonCopyHoisting.cpp b/llvm/lib/Target/Hexagon/HexagonCopyHoisting.cpp
new file mode 100644
index 000000000000..97917270601b
--- /dev/null
+++ b/llvm/lib/Target/Hexagon/HexagonCopyHoisting.cpp
@@ -0,0 +1,272 @@
+//===--------- HexagonCopyHoisting.cpp - Hexagon Copy Hoisting  ----------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+// The purpose of this pass is to move the copy instructions that are
+// present in all the successor of a basic block (BB) to the end of BB.
+//===----------------------------------------------------------------------===//
+
+#include "HexagonTargetMachine.h"
+#include "llvm/ADT/DenseMap.h"
+#include "llvm/ADT/PostOrderIterator.h"
+#include "llvm/ADT/StringRef.h"
+#include "llvm/ADT/Twine.h"
+#include "llvm/CodeGen/LiveInterval.h"
+#include "llvm/CodeGen/LiveIntervals.h"
+#include "llvm/CodeGen/MachineDominators.h"
+#include "llvm/CodeGen/MachineRegisterInfo.h"
+#include "llvm/Support/CommandLine.h"
+#include "llvm/Support/Debug.h"
+
+#define DEBUG_TYPE "CopyHoist"
+
+using namespace llvm;
+
+static cl::opt<std::string> CPHoistFn("cphoistfn", cl::Hidden, cl::desc(""),
+                                      cl::init(""));
+
+namespace llvm {
+void initializeHexagonCopyHoistingPass(PassRegistry &Registry);
+FunctionPass *createHexagonCopyHoisting();
+} // namespace llvm
+
+namespace {
+
+class HexagonCopyHoisting : public MachineFunctionPass {
+
+public:
+  static char ID;
+  HexagonCopyHoisting() : MachineFunctionPass(ID), MFN(nullptr), MRI(nullptr) {
+    initializeHexagonCopyHoistingPass(*PassRegistry::getPassRegistry());
+  }
+
+  StringRef getPassName() const override { return "Hexagon Copy Hoisting"; }
+
+  void getAnalysisUsage(AnalysisUsage &AU) const override {
+    AU.addRequired<SlotIndexes>();
+    AU.addRequired<LiveIntervals>();
+    AU.addPreserved<SlotIndexes>();
+    AU.addPreserved<LiveIntervals>();
+    AU.addRequired<MachineDominatorTree>();
+    AU.addPreserved<MachineDominatorTree>();
+    MachineFunctionPass::getAnalysisUsage(AU);
+  }
+
+  bool runOnMachineFunction(MachineFunction &Fn) override;
+  void collectCopyInst();
+  void addMItoCopyList(MachineInstr *MI);
+  bool analyzeCopy(MachineBasicBlock *BB);
+  bool isSafetoMove(MachineInstr *CandMI);
+  void moveCopyInstr(MachineBasicBlock *DestBB,
+                     std::pair<Register, Register> Key, MachineInstr *MI);
+
+  MachineFunction *MFN;
+  MachineRegisterInfo *MRI;
+  std::vector<DenseMap<std::pair<Register, Register>, MachineInstr *>>
+      CopyMIList;
+};
+
+} // namespace
+
+char HexagonCopyHoisting::ID = 0;
+
+namespace llvm {
+char &HexagonCopyHoistingID = HexagonCopyHoisting::ID;
+} // namespace llvm
+
+bool HexagonCopyHoisting::runOnMachineFunction(MachineFunction &Fn) {
+
+  if ((CPHoistFn != "") && (CPHoistFn != Fn.getFunction().getName()))
+    return false;
+
+  MFN = &Fn;
+  MRI = &Fn.getRegInfo();
+
+  LLVM_DEBUG(dbgs() << "\nCopy Hoisting:" << "\'" << Fn.getName() << "\'\n");
+
+  CopyMIList.clear();
+  CopyMIList.resize(Fn.getNumBlockIDs());
+
+  // Traverse through all basic blocks and collect copy instructions.
+  collectCopyInst();
+
+  // Traverse through the basic blocks again and move the COPY instructions
+  // that are present in all the successors of BB to BB.
+  bool Changed = false;
+  for (MachineBasicBlock *BB : post_order(&Fn)) {
+    if (!BB->empty()) {
+      if (BB->pred_size() != 1)
+        continue;
+      auto &BBCopyInst = CopyMIList[BB->getNumber()];
+      if (BBCopyInst.size() > 0)
+        Changed |= analyzeCopy(*BB->pred_begin());
+    }
+  }
+  // Re-compute liveness
+  if (Changed) {
+    LiveIntervals &LIS = getAnalysis<LiveIntervals>();
+    SlotIndexes *SI = LIS.getSlotIndexes();
+    SI->releaseMemory();
+    SI->runOnMachineFunction(Fn);
+    LIS.releaseMemory();
+    LIS.runOnMachineFunction(Fn);
+  }
+  return Changed;
+}
+
+//===----------------------------------------------------------------------===//
+// Save all COPY instructions for each basic block in CopyMIList vector.
+//===----------------------------------------------------------------------===//
+void HexagonCopyHoisting::collectCopyInst() {
+  for (MachineBasicBlock &BB : *MFN) {
+#ifndef NDEBUG
+    auto &BBCopyInst = CopyMIList[BB.getNumber()];
+    LLVM_DEBUG(dbgs() << "Visiting BB#" << BB.getNumber() << ":\n");
+#endif
+
+    for (MachineInstr &MI : BB) {
+      if (MI.getOpcode() == TargetOpcode::COPY)
+        addMItoCopyList(&MI);
+    }
+    LLVM_DEBUG(dbgs() << "\tNumber of copies: " << BBCopyInst.size() << "\n");
+  }
+}
+
+void HexagonCopyHoisting::addMItoCopyList(MachineInstr *MI) {
+  unsigned BBNum = MI->getParent()->getNumber();
+  auto &BBCopyInst = CopyMIList[BBNum];
+  Register DstReg = MI->getOperand(0).getReg();
+  Register SrcReg = MI->getOperand(1).getReg();
+
+  if (!Register::isVirtualRegister(DstReg) ||
+      !Register::isVirtualRegister(SrcReg) ||
+      MRI->getRegClass(DstReg) != &Hexagon::IntRegsRegClass ||
+      MRI->getRegClass(SrcReg) != &Hexagon::IntRegsRegClass)
+    return;
+
+  BBCopyInst.insert(std::pair(std::pair(SrcReg, DstReg), MI));
+#ifndef NDEBUG
+  LLVM_DEBUG(dbgs() << "\tAdding Copy Instr to the list: " << MI << "\n");
+  for (auto II : BBCopyInst) {
+    MachineInstr *TempMI = II.getSecond();
+    LLVM_DEBUG(dbgs() << "\tIn the list: " << TempMI << "\n");
+  }
+#endif
+}
+
+//===----------------------------------------------------------------------===//
+// Look at the COPY instructions of all the successors of BB. If the same
+// instruction is present in every successor and can be safely moved,
+// pull it into BB.
+//===----------------------------------------------------------------------===//
+bool HexagonCopyHoisting::analyzeCopy(MachineBasicBlock *BB) {
+
+  bool Changed = false;
+  if (BB->succ_size() < 2)
+    return false;
+
+  for (MachineBasicBlock *SB : BB->successors()) {
+    if (SB->pred_size() != 1 || SB->isEHPad() || SB->hasAddressTaken())
+      return false;
+  }
+
+  MachineBasicBlock *SBB1 = *BB->succ_begin();
+  auto &BBCopyInst1 = CopyMIList[SBB1->getNumber()];
+
+  for (auto II : BBCopyInst1) {
+    std::pair<Register, Register> Key = II.getFirst();
+    MachineInstr *MI = II.getSecond();
+    bool IsSafetoMove = true;
+    for (MachineBasicBlock *SuccBB : BB->successors()) {
+      auto &SuccBBCopyInst = CopyMIList[SuccBB->getNumber()];
+      if (!SuccBBCopyInst.count(Key)) {
+        // Same copy not present in this successor
+        IsSafetoMove = false;
+        break;
+      }
+      // If present, make sure that it's safe to pull this copy instruction
+      // into the predecessor.
+      MachineInstr *SuccMI = SuccBBCopyInst[Key];
+      if (!isSafetoMove(SuccMI)) {
+        IsSafetoMove = false;
+        break;
+      }
+    }
+    // If we have come this far, this copy instruction can be safely
+    // moved to the predecessor basic block.
+    if (IsSafetoMove) {
+      LLVM_DEBUG(dbgs() << "\t\t Moving instr to BB#" << BB->getNumber() << ": "
+                        << MI << "\n");
+      moveCopyInstr(BB, Key, MI);
+      // Add my into BB copyMI list.
+      Changed = true;
+    }
+  }
+
+#ifndef NDEBUG
+  auto &BBCopyInst = CopyMIList[BB->getNumber()];
+  for (auto II : BBCopyInst) {
+    MachineInstr *TempMI = II.getSecond();
+    LLVM_DEBUG(dbgs() << "\tIn the list: " << TempMI << "\n");
+  }
+#endif
+  return Changed;
+}
+
+bool HexagonCopyHoisting::isSafetoMove(MachineInstr *CandMI) {
+  // Make sure that it's safe to move this 'copy' instruction to the predecessor
+  // basic block.
+  assert(CandMI->getOperand(0).isReg() && CandMI->getOperand(1).isReg());
+  Register DefR = CandMI->getOperand(0).getReg();
+  Register UseR = CandMI->getOperand(1).getReg();
+
+  MachineBasicBlock *BB = CandMI->getParent();
+  // There should not be a def/use of DefR between the start of BB and CandMI.
+  MachineBasicBlock::iterator MII, MIE;
+  for (MII = BB->begin(), MIE = CandMI; MII != MIE; ++MII) {
+    MachineInstr *OtherMI = &*MII;
+    for (const MachineOperand &Mo : OtherMI->operands())
+      if (Mo.isReg() && Mo.getReg() == DefR)
+        return false;
+  }
+  // There should not be a def of UseR between the start of BB and CandMI.
+  for (MII = BB->begin(), MIE = CandMI; MII != MIE; ++MII) {
+    MachineInstr *OtherMI = &*MII;
+    for (const MachineOperand &Mo : OtherMI->operands())
+      if (Mo.isReg() && Mo.isDef() && Mo.getReg() == UseR)
+        return false;
+  }
+  return true;
+}
+
+void HexagonCopyHoisting::moveCopyInstr(MachineBasicBlock *DestBB,
+                                        std::pair<Register, Register> Key,
+                                        MachineInstr *MI) {
+  MachineBasicBlock::iterator FirstTI = DestBB->getFirstTerminator();
+  assert(FirstTI != DestBB->end());
+
+  DestBB->splice(FirstTI, MI->getParent(), MI);
+
+  addMItoCopyList(MI);
+  for (auto I = ++(DestBB->succ_begin()), E = DestBB->succ_end(); I != E; ++I) {
+    MachineBasicBlock *SuccBB = *I;
+    auto &BBCopyInst = CopyMIList[SuccBB->getNumber()];
+    MachineInstr *SuccMI = BBCopyInst[Key];
+    SuccMI->eraseFromParent();
+    BBCopyInst.erase(Key);
+  }
+}
+
+//===----------------------------------------------------------------------===//
+//                         Public Constructor Functions
+//===----------------------------------------------------------------------===//
+
+INITIALIZE_PASS(HexagonCopyHoisting, "hexagon-move-phicopy",
+                "Hexagon move phi copy", false, false)
+
+FunctionPass *llvm::createHexagonCopyHoisting() {
+  return new HexagonCopyHoisting();
+}
diff --git a/llvm/lib/Target/Hexagon/HexagonCopyToCombine.cpp b/llvm/lib/Target/Hexagon/HexagonCopyToCombine.cpp
index 310993662b67..99ac0c346d0c 100644
--- a/llvm/lib/Target/Hexagon/HexagonCopyToCombine.cpp
+++ b/llvm/lib/Target/Hexagon/HexagonCopyToCombine.cpp
@@ -283,7 +283,7 @@ bool HexagonCopyToCombine::isSafeToMoveTogether(MachineInstr &I1,
     // uses I2's use reg we need to modify that (first) instruction to now kill
     // this reg.
     unsigned KilledOperand = 0;
-    if (I2.killsRegister(I2UseReg))
+    if (I2.killsRegister(I2UseReg, /*TRI=*/nullptr))
       KilledOperand = I2UseReg;
     MachineInstr *KillingInstr = nullptr;
 
@@ -360,11 +360,12 @@ bool HexagonCopyToCombine::isSafeToMoveTogether(MachineInstr &I1,
 
       if (isUnsafeToMoveAcross(MI, I1UseReg, I1DestReg, TRI) ||
           // Check for an aliased register kill. Bail out if we see one.
-          (!MI.killsRegister(I1UseReg) && MI.killsRegister(I1UseReg, TRI)))
+          (!MI.killsRegister(I1UseReg, /*TRI=*/nullptr) &&
+           MI.killsRegister(I1UseReg, TRI)))
         return false;
 
       // Check for an exact kill (registers match).
-      if (I1UseReg && MI.killsRegister(I1UseReg)) {
+      if (I1UseReg && MI.killsRegister(I1UseReg, /*TRI=*/nullptr)) {
         assert(!KillingInstr && "Should only see one killing instruction");
         KilledOperand = I1UseReg;
         KillingInstr = &MI;
diff --git a/llvm/lib/Target/Hexagon/HexagonExpandCondsets.cpp b/llvm/lib/Target/Hexagon/HexagonExpandCondsets.cpp
index e1005296d637..204f3b6b20c7 100644
--- a/llvm/lib/Target/Hexagon/HexagonExpandCondsets.cpp
+++ b/llvm/lib/Target/Hexagon/HexagonExpandCondsets.cpp
@@ -779,7 +779,8 @@ MachineInstr *HexagonExpandCondsets::getReachingDefForPred(RegisterRef RD,
     // Check if this instruction can be ignored, i.e. if it is predicated
     // on the complementary condition.
     if (PredValid && HII->isPredicated(*MI)) {
-      if (MI->readsRegister(PredR) && (Cond != HII->isPredicatedTrue(*MI)))
+      if (MI->readsRegister(PredR, /*TRI=*/nullptr) &&
+          (Cond != HII->isPredicatedTrue(*MI)))
         continue;
     }
 
@@ -937,7 +938,8 @@ void HexagonExpandCondsets::renameInRange(RegisterRef RO, RegisterRef RN,
     // on the opposite condition.
     if (!HII->isPredicated(MI))
       continue;
-    if (!MI.readsRegister(PredR) || (Cond != HII->isPredicatedTrue(MI)))
+    if (!MI.readsRegister(PredR, /*TRI=*/nullptr) ||
+        (Cond != HII->isPredicatedTrue(MI)))
       continue;
 
     for (auto &Op : MI.operands()) {
@@ -1007,7 +1009,8 @@ bool HexagonExpandCondsets::predicate(MachineInstr &TfrI, bool Cond,
     // By default assume that the instruction executes on the same condition
     // as TfrI (Exec_Then), and also on the opposite one (Exec_Else).
     unsigned Exec = Exec_Then | Exec_Else;
-    if (PredValid && HII->isPredicated(MI) && MI.readsRegister(PredR))
+    if (PredValid && HII->isPredicated(MI) &&
+        MI.readsRegister(PredR, /*TRI=*/nullptr))
       Exec = (Cond == HII->isPredicatedTrue(MI)) ? Exec_Then : Exec_Else;
 
     for (auto &Op : MI.operands()) {
diff --git a/llvm/lib/Target/Hexagon/HexagonInstrInfo.cpp b/llvm/lib/Target/Hexagon/HexagonInstrInfo.cpp
index b9bf26ba7cca..e04f9c6faacd 100644
--- a/llvm/lib/Target/Hexagon/HexagonInstrInfo.cpp
+++ b/llvm/lib/Target/Hexagon/HexagonInstrInfo.cpp
@@ -3517,7 +3517,7 @@ unsigned HexagonInstrInfo::getCompoundOpcode(const MachineInstr &GA,
       (GB.getOpcode() != Hexagon::J2_jumptnew))
     return -1u;
   Register DestReg = GA.getOperand(0).getReg();
-  if (!GB.readsRegister(DestReg))
+  if (!GB.readsRegister(DestReg, /*TRI=*/nullptr))
     return -1u;
   if (DestReg != Hexagon::P0 && DestReg != Hexagon::P1)
     return -1u;
@@ -4334,7 +4334,7 @@ std::optional<unsigned> HexagonInstrInfo::getOperandLatency(
   if (DefMO.isReg() && DefMO.getReg().isPhysical()) {
     if (DefMO.isImplicit()) {
       for (MCPhysReg SR : HRI.superregs(DefMO.getReg())) {
-        int Idx = DefMI.findRegisterDefOperandIdx(SR, false, false, &HRI);
+        int Idx = DefMI.findRegisterDefOperandIdx(SR, &HRI, false, false);
         if (Idx != -1) {
           DefIdx = Idx;
           break;
@@ -4345,7 +4345,7 @@ std::optional<unsigned> HexagonInstrInfo::getOperandLatency(
     const MachineOperand &UseMO = UseMI.getOperand(UseIdx);
     if (UseMO.isImplicit()) {
       for (MCPhysReg SR : HRI.superregs(UseMO.getReg())) {
-        int Idx = UseMI.findRegisterUseOperandIdx(SR, false, &HRI);
+        int Idx = UseMI.findRegisterUseOperandIdx(SR, &HRI, false);
         if (Idx != -1) {
           UseIdx = Idx;
           break;
diff --git a/llvm/lib/Target/Hexagon/HexagonTargetMachine.cpp b/llvm/lib/Target/Hexagon/HexagonTargetMachine.cpp
index e64d7e52a9aa..3a792ecfd03d 100644
--- a/llvm/lib/Target/Hexagon/HexagonTargetMachine.cpp
+++ b/llvm/lib/Target/Hexagon/HexagonTargetMachine.cpp
@@ -43,8 +43,9 @@ cl::opt<unsigned> RDFFuncBlockLimit(
     "rdf-bb-limit", cl::Hidden, cl::init(1000),
     cl::desc("Basic block limit for a function for RDF optimizations"));
 
-static cl::opt<bool> DisableHardwareLoops("disable-hexagon-hwloops",
-  cl::Hidden, cl::desc("Disable Hardware Loops for Hexagon target"));
+static cl::opt<bool>
+    DisableHardwareLoops("disable-hexagon-hwloops", cl::Hidden,
+                         cl::desc("Disable Hardware Loops for Hexagon target"));
 
 static cl::opt<bool>
     DisableAModeOpt("disable-hexagon-amodeopt", cl::Hidden,
@@ -58,8 +59,9 @@ static cl::opt<bool>
     DisableHCP("disable-hcp", cl::Hidden,
                cl::desc("Disable Hexagon constant propagation"));
 
-static cl::opt<bool> DisableStoreWidening("disable-store-widen",
-  cl::Hidden, cl::init(false), cl::desc("Disable store widening"));
+static cl::opt<bool> DisableStoreWidening("disable-store-widen", cl::Hidden,
+                                          cl::init(false),
+                                          cl::desc("Disable store widening"));
 
 static cl::opt<bool> EnableExpandCondsets("hexagon-expand-condsets",
                                           cl::init(true), cl::Hidden,
@@ -72,42 +74,53 @@ static cl::opt<bool> EnableTfrCleanup("hexagon-tfr-cleanup", cl::init(true),
 static cl::opt<bool> EnableEarlyIf("hexagon-eif", cl::init(true), cl::Hidden,
                                    cl::desc("Enable early if-conversion"));
 
-static cl::opt<bool> EnableGenInsert("hexagon-insert", cl::init(true),
-  cl::Hidden, cl::desc("Generate \"insert\" instructions"));
+static cl::opt<bool> EnableCopyHoist("hexagon-copy-hoist", cl::init(true),
+                                     cl::Hidden, cl::ZeroOrMore,
+                                     cl::desc("Enable Hexagon copy hoisting"));
+
+static cl::opt<bool>
+    EnableGenInsert("hexagon-insert", cl::init(true), cl::Hidden,
+                    cl::desc("Generate \"insert\" instructions"));
 
 static cl::opt<bool>
     EnableCommGEP("hexagon-commgep", cl::init(true), cl::Hidden,
                   cl::desc("Enable commoning of GEP instructions"));
 
-static cl::opt<bool> EnableGenExtract("hexagon-extract", cl::init(true),
-  cl::Hidden, cl::desc("Generate \"extract\" instructions"));
+static cl::opt<bool>
+    EnableGenExtract("hexagon-extract", cl::init(true), cl::Hidden,
+                     cl::desc("Generate \"extract\" instructions"));
 
-static cl::opt<bool> EnableGenMux("hexagon-mux", cl::init(true), cl::Hidden,
-  cl::desc("Enable converting conditional transfers into MUX instructions"));
+static cl::opt<bool> EnableGenMux(
+    "hexagon-mux", cl::init(true), cl::Hidden,
+    cl::desc("Enable converting conditional transfers into MUX instructions"));
 
-static cl::opt<bool> EnableGenPred("hexagon-gen-pred", cl::init(true),
-  cl::Hidden, cl::desc("Enable conversion of arithmetic operations to "
-  "predicate instructions"));
+static cl::opt<bool>
+    EnableGenPred("hexagon-gen-pred", cl::init(true), cl::Hidden,
+                  cl::desc("Enable conversion of arithmetic operations to "
+                           "predicate instructions"));
 
 static cl::opt<bool>
     EnableLoopPrefetch("hexagon-loop-prefetch", cl::Hidden,
                        cl::desc("Enable loop data prefetch on Hexagon"));
 
-static cl::opt<bool> DisableHSDR("disable-hsdr", cl::init(false), cl::Hidden,
-  cl::desc("Disable splitting double registers"));
+static cl::opt<bool>
+    DisableHSDR("disable-hsdr", cl::init(false), cl::Hidden,
+                cl::desc("Disable splitting double registers"));
 
 static cl::opt<bool>
     EnableGenMemAbs("hexagon-mem-abs", cl::init(true), cl::Hidden,
                     cl::desc("Generate absolute set instructions"));
 
 static cl::opt<bool> EnableBitSimplify("hexagon-bit", cl::init(true),
-  cl::Hidden, cl::desc("Bit simplification"));
+                                       cl::Hidden,
+                                       cl::desc("Bit simplification"));
 
 static cl::opt<bool> EnableLoopResched("hexagon-loop-resched", cl::init(true),
-  cl::Hidden, cl::desc("Loop rescheduling"));
+                                       cl::Hidden,
+                                       cl::desc("Loop rescheduling"));
 
-static cl::opt<bool> HexagonNoOpt("hexagon-noopt", cl::init(false),
-  cl::Hidden, cl::desc("Disable backend optimizations"));
+static cl::opt<bool> HexagonNoOpt("hexagon-noopt", cl::init(false), cl::Hidden,
+                                  cl::desc("Disable backend optimizations"));
 
 static cl::opt<bool>
     EnableVectorPrint("enable-hexagon-vector-print", cl::Hidden,
@@ -148,69 +161,72 @@ static ScheduleDAGInstrs *createVLIWMachineSched(MachineSchedContext *C) {
 }
 
 static MachineSchedRegistry
-SchedCustomRegistry("hexagon", "Run Hexagon's custom scheduler",
-                    createVLIWMachineSched);
+    SchedCustomRegistry("hexagon", "Run Hexagon's custom scheduler",
+                        createVLIWMachineSched);
 
 namespace llvm {
-  extern char &HexagonExpandCondsetsID;
-  extern char &HexagonTfrCleanupID;
-  void initializeHexagonBitSimplifyPass(PassRegistry&);
-  void initializeHexagonConstExtendersPass(PassRegistry&);
-  void initializeHexagonConstPropagationPass(PassRegistry&);
-  void initializeHexagonCopyToCombinePass(PassRegistry&);
-  void initializeHexagonEarlyIfConversionPass(PassRegistry&);
-  void initializeHexagonExpandCondsetsPass(PassRegistry&);
-  void initializeHexagonGenMemAbsolutePass(PassRegistry &);
-  void initializeHexagonGenMuxPass(PassRegistry&);
-  void initializeHexagonHardwareLoopsPass(PassRegistry&);
-  void initializeHexagonLoopIdiomRecognizeLegacyPassPass(PassRegistry &);
-  void initializeHexagonLoopAlignPass(PassRegistry &);
-  void initializeHexagonNewValueJumpPass(PassRegistry&);
-  void initializeHexagonOptAddrModePass(PassRegistry&);
-  void initializeHexagonPacketizerPass(PassRegistry&);
-  void initializeHexagonRDFOptPass(PassRegistry&);
-  void initializeHexagonSplitDoubleRegsPass(PassRegistry&);
-  void initializeHexagonTfrCleanupPass(PassRegistry &);
-  void initializeHexagonVExtractPass(PassRegistry &);
-  void initializeHexagonVectorCombineLegacyPass(PassRegistry&);
-  void initializeHexagonVectorLoopCarriedReuseLegacyPassPass(PassRegistry &);
-  Pass *createHexagonLoopIdiomPass();
-  Pass *createHexagonVectorLoopCarriedReuseLegacyPass();
-
-  FunctionPass *createHexagonBitSimplify();
-  FunctionPass *createHexagonBranchRelaxation();
-  FunctionPass *createHexagonCallFrameInformation();
-  FunctionPass *createHexagonCFGOptimizer();
-  FunctionPass *createHexagonCommonGEP();
-  FunctionPass *createHexagonConstExtenders();
-  FunctionPass *createHexagonConstPropagationPass();
-  FunctionPass *createHexagonCopyToCombine();
-  FunctionPass *createHexagonEarlyIfConversion();
-  FunctionPass *createHexagonFixupHwLoops();
-  FunctionPass *createHexagonGenExtract();
-  FunctionPass *createHexagonGenInsert();
-  FunctionPass *createHexagonGenMemAbsolute();
-  FunctionPass *createHexagonGenMux();
-  FunctionPass *createHexagonGenPredicate();
-  FunctionPass *createHexagonHardwareLoops();
-  FunctionPass *createHexagonISelDag(HexagonTargetMachine &TM,
-                                     CodeGenOptLevel OptLevel);
-  FunctionPass *createHexagonLoopAlign();
-  FunctionPass *createHexagonLoopRescheduling();
-  FunctionPass *createHexagonNewValueJump();
-  FunctionPass *createHexagonOptAddrMode();
-  FunctionPass *createHexagonOptimizeSZextends();
-  FunctionPass *createHexagonPacketizer(bool Minimal);
-  FunctionPass *createHexagonPeephole();
-  FunctionPass *createHexagonRDFOpt();
-  FunctionPass *createHexagonSplitConst32AndConst64();
-  FunctionPass *createHexagonSplitDoubleRegs();
-  FunctionPass *createHexagonStoreWidening();
-  FunctionPass *createHexagonTfrCleanup();
-  FunctionPass *createHexagonVectorCombineLegacyPass();
-  FunctionPass *createHexagonVectorPrint();
-  FunctionPass *createHexagonVExtract();
-} // end namespace llvm;
+extern char &HexagonCopyHoistingID;
+extern char &HexagonExpandCondsetsID;
+extern char &HexagonTfrCleanupID;
+void initializeHexagonBitSimplifyPass(PassRegistry &);
+void initializeHexagonCopyHoistingPass(PassRegistry &);
+void initializeHexagonConstExtendersPass(PassRegistry &);
+void initializeHexagonConstPropagationPass(PassRegistry &);
+void initializeHexagonCopyToCombinePass(PassRegistry &);
+void initializeHexagonEarlyIfConversionPass(PassRegistry &);
+void initializeHexagonExpandCondsetsPass(PassRegistry &);
+void initializeHexagonGenMemAbsolutePass(PassRegistry &);
+void initializeHexagonGenMuxPass(PassRegistry &);
+void initializeHexagonHardwareLoopsPass(PassRegistry &);
+void initializeHexagonLoopIdiomRecognizeLegacyPassPass(PassRegistry &);
+void initializeHexagonLoopAlignPass(PassRegistry &);
+void initializeHexagonNewValueJumpPass(PassRegistry &);
+void initializeHexagonOptAddrModePass(PassRegistry &);
+void initializeHexagonPacketizerPass(PassRegistry &);
+void initializeHexagonRDFOptPass(PassRegistry &);
+void initializeHexagonSplitDoubleRegsPass(PassRegistry &);
+void initializeHexagonTfrCleanupPass(PassRegistry &);
+void initializeHexagonVExtractPass(PassRegistry &);
+void initializeHexagonVectorCombineLegacyPass(PassRegistry &);
+void initializeHexagonVectorLoopCarriedReuseLegacyPassPass(PassRegistry &);
+Pass *createHexagonLoopIdiomPass();
+Pass *createHexagonVectorLoopCarriedReuseLegacyPass();
+
+FunctionPass *createHexagonBitSimplify();
+FunctionPass *createHexagonBranchRelaxation();
+FunctionPass *createHexagonCallFrameInformation();
+FunctionPass *createHexagonCFGOptimizer();
+FunctionPass *createHexagonCommonGEP();
+FunctionPass *createHexagonConstExtenders();
+FunctionPass *createHexagonConstPropagationPass();
+FunctionPass *createHexagonCopyHoisting();
+FunctionPass *createHexagonCopyToCombine();
+FunctionPass *createHexagonEarlyIfConversion();
+FunctionPass *createHexagonFixupHwLoops();
+FunctionPass *createHexagonGenExtract();
+FunctionPass *createHexagonGenInsert();
+FunctionPass *createHexagonGenMemAbsolute();
+FunctionPass *createHexagonGenMux();
+FunctionPass *createHexagonGenPredicate();
+FunctionPass *createHexagonHardwareLoops();
+FunctionPass *createHexagonISelDag(HexagonTargetMachine &TM,
+                                   CodeGenOptLevel OptLevel);
+FunctionPass *createHexagonLoopAlign();
+FunctionPass *createHexagonLoopRescheduling();
+FunctionPass *createHexagonNewValueJump();
+FunctionPass *createHexagonOptAddrMode();
+FunctionPass *createHexagonOptimizeSZextends();
+FunctionPass *createHexagonPacketizer(bool Minimal);
+FunctionPass *createHexagonPeephole();
+FunctionPass *createHexagonRDFOpt();
+FunctionPass *createHexagonSplitConst32AndConst64();
+FunctionPass *createHexagonSplitDoubleRegs();
+FunctionPass *createHexagonStoreWidening();
+FunctionPass *createHexagonTfrCleanup();
+FunctionPass *createHexagonVectorCombineLegacyPass();
+FunctionPass *createHexagonVectorPrint();
+FunctionPass *createHexagonVExtract();
+} // namespace llvm
 
 static Reloc::Model getEffectiveRelocModel(std::optional<Reloc::Model> RM) {
   return RM.value_or(Reloc::Static);
@@ -260,6 +276,7 @@ HexagonTargetMachine::HexagonTargetMachine(const Target &T, const Triple &TT,
           (HexagonNoOpt ? CodeGenOptLevel::None : OL)),
       TLOF(std::make_unique<HexagonTargetObjectFile>()),
       Subtarget(Triple(TT), CPU, FS, *this) {
+  initializeHexagonCopyHoistingPass(*PassRegistry::getPassRegistry());
   initializeHexagonExpandCondsetsPass(*PassRegistry::getPassRegistry());
   initializeHexagonLoopAlignPass(*PassRegistry::getPassRegistry());
   initializeHexagonTfrCleanupPass(*PassRegistry::getPassRegistry());
@@ -269,10 +286,8 @@ HexagonTargetMachine::HexagonTargetMachine(const Target &T, const Triple &TT,
 const HexagonSubtarget *
 HexagonTargetMachine::getSubtargetImpl(const Function &F) const {
   AttributeList FnAttrs = F.getAttributes();
-  Attribute CPUAttr =
-      FnAttrs.getFnAttr("target-cpu");
-  Attribute FSAttr =
-      FnAttrs.getFnAttr("target-features");
+  Attribute CPUAttr = FnAttrs.getFnAttr("target-cpu");
+  Attribute FSAttr = FnAttrs.getFnAttr("target-features");
 
   std::string CPU =
       CPUAttr.isValid() ? CPUAttr.getValueAsString().str() : TargetCPU;
@@ -331,7 +346,7 @@ namespace {
 class HexagonPassConfig : public TargetPassConfig {
 public:
   HexagonPassConfig(HexagonTargetMachine &TM, PassManagerBase &PM)
-    : TargetPassConfig(TM, PM) {}
+      : TargetPassConfig(TM, PM) {}
 
   HexagonTargetMachine &getHexagonTargetMachine() const {
     return getTM<HexagonTargetMachine>();
@@ -433,6 +448,8 @@ void HexagonPassConfig::addPreRegAlloc() {
       addPass(createHexagonConstExtenders());
     if (EnableExpandCondsets)
       insertPass(&RegisterCoalescerID, &HexagonExpandCondsetsID);
+    if (EnableCopyHoist)
+      insertPass(&RegisterCoalescerID, &HexagonCopyHoistingID);
     if (EnableTfrCleanup)
       insertPass(&VirtRegRewriterID, &HexagonTfrCleanupID);
     if (!DisableStoreWidening)
diff --git a/llvm/lib/Target/Hexagon/HexagonTargetTransformInfo.h b/llvm/lib/Target/Hexagon/HexagonTargetTransformInfo.h
index 9689f2f5bb86..90c402876a57 100644
--- a/llvm/lib/Target/Hexagon/HexagonTargetTransformInfo.h
+++ b/llvm/lib/Target/Hexagon/HexagonTargetTransformInfo.h
@@ -141,7 +141,7 @@ public:
       unsigned Opcode, Type *Ty, TTI::TargetCostKind CostKind,
       TTI::OperandValueInfo Op1Info = {TTI::OK_AnyValue, TTI::OP_None},
       TTI::OperandValueInfo Op2Info = {TTI::OK_AnyValue, TTI::OP_None},
-      ArrayRef<const Value *> Args = ArrayRef<const Value *>(),
+      ArrayRef<const Value *> Args = std::nullopt,
       const Instruction *CxtI = nullptr);
   InstructionCost getCastInstrCost(unsigned Opcode, Type *Dst, Type *Src,
                                    TTI::CastContextHint CCH,
diff --git a/llvm/lib/Target/Hexagon/MCTargetDesc/HexagonMCTargetDesc.cpp b/llvm/lib/Target/Hexagon/MCTargetDesc/HexagonMCTargetDesc.cpp
index dc8328a6705d..0a948402fb89 100644
--- a/llvm/lib/Target/Hexagon/MCTargetDesc/HexagonMCTargetDesc.cpp
+++ b/llvm/lib/Target/Hexagon/MCTargetDesc/HexagonMCTargetDesc.cpp
@@ -385,8 +385,7 @@ createMCAsmTargetStreamer(MCStreamer &S, formatted_raw_ostream &OS,
 static MCStreamer *createMCStreamer(Triple const &T, MCContext &Context,
                                     std::unique_ptr<MCAsmBackend> &&MAB,
                                     std::unique_ptr<MCObjectWriter> &&OW,
-                                    std::unique_ptr<MCCodeEmitter> &&Emitter,
-                                    bool RelaxAll) {
+                                    std::unique_ptr<MCCodeEmitter> &&Emitter) {
   return createHexagonELFStreamer(T, Context, std::move(MAB), std::move(OW),
                                   std::move(Emitter));
 }
diff --git a/llvm/lib/Target/Lanai/LanaiTargetTransformInfo.h b/llvm/lib/Target/Lanai/LanaiTargetTransformInfo.h
index bc612963b0ad..b064f4d5b1ed 100644
--- a/llvm/lib/Target/Lanai/LanaiTargetTransformInfo.h
+++ b/llvm/lib/Target/Lanai/LanaiTargetTransformInfo.h
@@ -94,7 +94,7 @@ public:
       unsigned Opcode, Type *Ty, TTI::TargetCostKind CostKind,
       TTI::OperandValueInfo Op1Info = {TTI::OK_AnyValue, TTI::OP_None},
       TTI::OperandValueInfo Op2Info = {TTI::OK_AnyValue, TTI::OP_None},
-      ArrayRef<const Value *> Args = ArrayRef<const Value *>(),
+      ArrayRef<const Value *> Args = std::nullopt,
       const Instruction *CxtI = nullptr) {
     int ISD = TLI->InstructionOpcodeToISD(Opcode);
 
diff --git a/llvm/lib/Target/Lanai/MCTargetDesc/LanaiMCTargetDesc.cpp b/llvm/lib/Target/Lanai/MCTargetDesc/LanaiMCTargetDesc.cpp
index 8f83c883e822..4a381c033b38 100644
--- a/llvm/lib/Target/Lanai/MCTargetDesc/LanaiMCTargetDesc.cpp
+++ b/llvm/lib/Target/Lanai/MCTargetDesc/LanaiMCTargetDesc.cpp
@@ -63,13 +63,12 @@ createLanaiMCSubtargetInfo(const Triple &TT, StringRef CPU, StringRef FS) {
 static MCStreamer *createMCStreamer(const Triple &T, MCContext &Context,
                                     std::unique_ptr<MCAsmBackend> &&MAB,
                                     std::unique_ptr<MCObjectWriter> &&OW,
-                                    std::unique_ptr<MCCodeEmitter> &&Emitter,
-                                    bool RelaxAll) {
+                                    std::unique_ptr<MCCodeEmitter> &&Emitter) {
   if (!T.isOSBinFormatELF())
     llvm_unreachable("OS not supported");
 
   return createELFStreamer(Context, std::move(MAB), std::move(OW),
-                           std::move(Emitter), RelaxAll);
+                           std::move(Emitter));
 }
 
 static MCInstPrinter *createLanaiMCInstPrinter(const Triple & /*T*/,
diff --git a/llvm/lib/Target/LoongArch/LoongArchSubtarget.h b/llvm/lib/Target/LoongArch/LoongArchSubtarget.h
index cecb4a50aa76..a6e40840517f 100644
--- a/llvm/lib/Target/LoongArch/LoongArchSubtarget.h
+++ b/llvm/lib/Target/LoongArch/LoongArchSubtarget.h
@@ -31,21 +31,11 @@ class StringRef;
 
 class LoongArchSubtarget : public LoongArchGenSubtargetInfo {
   virtual void anchor();
-  bool HasLA32 = false;
-  bool HasLA64 = false;
-  bool HasBasicF = false;
-  bool HasBasicD = false;
-  bool HasExtLSX = false;
-  bool HasExtLASX = false;
-  bool HasExtLVZ = false;
-  bool HasExtLBT = false;
-  bool HasLaGlobalWithPcrel = false;
-  bool HasLaGlobalWithAbs = false;
-  bool HasLaLocalWithAbs = false;
-  bool HasUAL = false;
-  bool HasLinkerRelax = false;
-  bool HasExpAutoVec = false;
-  bool HasFrecipe = false;
+
+#define GET_SUBTARGETINFO_MACRO(ATTRIBUTE, DEFAULT, GETTER) \
+  bool ATTRIBUTE = DEFAULT;
+#include "LoongArchGenSubtargetInfo.inc"
+
   unsigned GRLen = 32;
   MVT GRLenVT = MVT::i32;
   LoongArchABI::ABI TargetABI = LoongArchABI::ABI_Unknown;
@@ -92,20 +82,12 @@ public:
   const SelectionDAGTargetInfo *getSelectionDAGInfo() const override {
     return &TSInfo;
   }
+
+#define GET_SUBTARGETINFO_MACRO(ATTRIBUTE, DEFAULT, GETTER) \
+  bool GETTER() const { return ATTRIBUTE; }
+#include "LoongArchGenSubtargetInfo.inc"
+
   bool is64Bit() const { return HasLA64; }
-  bool hasBasicF() const { return HasBasicF; }
-  bool hasBasicD() const { return HasBasicD; }
-  bool hasExtLSX() const { return HasExtLSX; }
-  bool hasExtLASX() const { return HasExtLASX; }
-  bool hasExtLVZ() const { return HasExtLVZ; }
-  bool hasExtLBT() const { return HasExtLBT; }
-  bool hasLaGlobalWithPcrel() const { return HasLaGlobalWithPcrel; }
-  bool hasLaGlobalWithAbs() const { return HasLaGlobalWithAbs; }
-  bool hasLaLocalWithAbs() const { return HasLaLocalWithAbs; }
-  bool hasUAL() const { return HasUAL; }
-  bool hasLinkerRelax() const { return HasLinkerRelax; }
-  bool hasExpAutoVec() const { return HasExpAutoVec; }
-  bool hasFrecipe() const { return HasFrecipe; }
   MVT getGRLenVT() const { return GRLenVT; }
   unsigned getGRLen() const { return GRLen; }
   LoongArchABI::ABI getTargetABI() const { return TargetABI; }
diff --git a/llvm/lib/Target/LoongArch/MCTargetDesc/LoongArchELFStreamer.cpp b/llvm/lib/Target/LoongArch/MCTargetDesc/LoongArchELFStreamer.cpp
index a6e15e09463d..9e56333e5fd9 100644
--- a/llvm/lib/Target/LoongArch/MCTargetDesc/LoongArchELFStreamer.cpp
+++ b/llvm/lib/Target/LoongArch/MCTargetDesc/LoongArchELFStreamer.cpp
@@ -82,11 +82,9 @@ namespace llvm {
 MCELFStreamer *createLoongArchELFStreamer(MCContext &C,
                                           std::unique_ptr<MCAsmBackend> MAB,
                                           std::unique_ptr<MCObjectWriter> MOW,
-                                          std::unique_ptr<MCCodeEmitter> MCE,
-                                          bool RelaxAll) {
+                                          std::unique_ptr<MCCodeEmitter> MCE) {
   LoongArchELFStreamer *S = new LoongArchELFStreamer(
       C, std::move(MAB), std::move(MOW), std::move(MCE));
-  S->getAssembler().setRelaxAll(RelaxAll);
   return S;
 }
 } // end namespace llvm
diff --git a/llvm/lib/Target/LoongArch/MCTargetDesc/LoongArchELFStreamer.h b/llvm/lib/Target/LoongArch/MCTargetDesc/LoongArchELFStreamer.h
index 220b54092c72..e220729d8923 100644
--- a/llvm/lib/Target/LoongArch/MCTargetDesc/LoongArchELFStreamer.h
+++ b/llvm/lib/Target/LoongArch/MCTargetDesc/LoongArchELFStreamer.h
@@ -25,7 +25,6 @@ public:
 MCELFStreamer *createLoongArchELFStreamer(MCContext &C,
                                           std::unique_ptr<MCAsmBackend> MAB,
                                           std::unique_ptr<MCObjectWriter> MOW,
-                                          std::unique_ptr<MCCodeEmitter> MCE,
-                                          bool RelaxAll);
+                                          std::unique_ptr<MCCodeEmitter> MCE);
 } // end namespace llvm
 #endif
diff --git a/llvm/lib/Target/LoongArch/MCTargetDesc/LoongArchMCTargetDesc.cpp b/llvm/lib/Target/LoongArch/MCTargetDesc/LoongArchMCTargetDesc.cpp
index a4e6a09863e6..e40981f5b5cd 100644
--- a/llvm/lib/Target/LoongArch/MCTargetDesc/LoongArchMCTargetDesc.cpp
+++ b/llvm/lib/Target/LoongArch/MCTargetDesc/LoongArchMCTargetDesc.cpp
@@ -193,10 +193,9 @@ namespace {
 MCStreamer *createLoongArchELFStreamer(const Triple &T, MCContext &Context,
                                        std::unique_ptr<MCAsmBackend> &&MAB,
                                        std::unique_ptr<MCObjectWriter> &&MOW,
-                                       std::unique_ptr<MCCodeEmitter> &&MCE,
-                                       bool RelaxAll) {
+                                       std::unique_ptr<MCCodeEmitter> &&MCE) {
   return createLoongArchELFStreamer(Context, std::move(MAB), std::move(MOW),
-                                    std::move(MCE), RelaxAll);
+                                    std::move(MCE));
 }
 } // end namespace
 
diff --git a/llvm/lib/Target/M68k/M68kExpandPseudo.cpp b/llvm/lib/Target/M68k/M68kExpandPseudo.cpp
index 7fcc65beaa65..c7fdd7d7c350 100644
--- a/llvm/lib/Target/M68k/M68kExpandPseudo.cpp
+++ b/llvm/lib/Target/M68k/M68kExpandPseudo.cpp
@@ -80,6 +80,13 @@ bool M68kExpandPseudo::ExpandMI(MachineBasicBlock &MBB,
   default:
     return false;
 
+  case M68k::MOVI8di:
+    return TII->ExpandMOVI(MIB, MVT::i8);
+  case M68k::MOVI16ri:
+    return TII->ExpandMOVI(MIB, MVT::i16);
+  case M68k::MOVI32ri:
+    return TII->ExpandMOVI(MIB, MVT::i32);
+
   case M68k::MOVXd16d8:
     return TII->ExpandMOVX_RR(MIB, MVT::i16, MVT::i8);
   case M68k::MOVXd32d8:
diff --git a/llvm/lib/Target/M68k/M68kISelLowering.cpp b/llvm/lib/Target/M68k/M68kISelLowering.cpp
index 786aa7bcb64e..62e4b36b5c9a 100644
--- a/llvm/lib/Target/M68k/M68kISelLowering.cpp
+++ b/llvm/lib/Target/M68k/M68kISelLowering.cpp
@@ -3075,9 +3075,9 @@ static bool checkAndUpdateCCRKill(MachineBasicBlock::iterator SelectItr,
   MachineBasicBlock::iterator miI(std::next(SelectItr));
   for (MachineBasicBlock::iterator miE = BB->end(); miI != miE; ++miI) {
     const MachineInstr &mi = *miI;
-    if (mi.readsRegister(M68k::CCR))
+    if (mi.readsRegister(M68k::CCR, /*TRI=*/nullptr))
       return false;
-    if (mi.definesRegister(M68k::CCR))
+    if (mi.definesRegister(M68k::CCR, /*TRI=*/nullptr))
       break; // Should have kill-flag - update below.
   }
 
@@ -3208,7 +3208,7 @@ M68kTargetLowering::EmitLoweredSelect(MachineInstr &MI,
   const TargetRegisterInfo *TRI = Subtarget.getRegisterInfo();
 
   MachineInstr *LastCCRSUser = CascadedCMOV ? CascadedCMOV : LastCMOV;
-  if (!LastCCRSUser->killsRegister(M68k::CCR) &&
+  if (!LastCCRSUser->killsRegister(M68k::CCR, /*TRI=*/nullptr) &&
       !checkAndUpdateCCRKill(LastCCRSUser, MBB, TRI)) {
     Copy0MBB->addLiveIn(M68k::CCR);
     SinkMBB->addLiveIn(M68k::CCR);
diff --git a/llvm/lib/Target/M68k/M68kInstrData.td b/llvm/lib/Target/M68k/M68kInstrData.td
index fa7e7aa0ed46..dc777a933e27 100644
--- a/llvm/lib/Target/M68k/M68kInstrData.td
+++ b/llvm/lib/Target/M68k/M68kInstrData.td
@@ -19,7 +19,7 @@
 ///
 ///  Pseudo:
 ///
-///     MOVSX [x]     MOVZX [x]     MOVX   [x]
+///     MOVI  [x]     MOVSX [x]     MOVZX [x]     MOVX   [x]
 ///
 ///  Map:
 ///
@@ -165,11 +165,12 @@ foreach AM = MxMoveSupportedAMs in {
 } // foreach AM
 
 // R <- I
+// No pattern, as all immediate -> register moves are matched to the MOVI pseudo
 class MxMove_RI<MxType TYPE, string DST_REG, MxMoveEncoding ENC,
                 MxImmOpBundle SRC = !cast<MxImmOpBundle>("MxOp"#TYPE.Size#"AddrMode_i"),
                 MxOpBundle DST = !cast<MxOpBundle>("MxOp"#TYPE.Size#"AddrMode_"#DST_REG)>
     : MxMove<TYPE.Prefix, (outs DST.Op:$dst), (ins SRC.Op:$src),
-              [(set TYPE.VT:$dst, SRC.ImmPat:$src)], ENC>;
+              [(null_frag)], ENC>;
 
 foreach REG = ["r", "a", "d"] in {
   foreach TYPE = !if(!eq(REG, "d"), [MxType8, MxType16, MxType32], [MxType16, MxType32]) in
@@ -243,6 +244,24 @@ def : Pat<(store   MxType32.BPat :$src, MxType32.JPat :$dst),
           (MOV32ji MxType32.JOp  :$dst, MxType32.IOp  :$src)>;
 
 //===----------------------------------------------------------------------===//
+// MOVEQ
+//===----------------------------------------------------------------------===//
+
+/// ------------+---------+---+-----------------------
+///  F  E  D  C | B  A  9 | 8 | 7  6  5  4  3  2  1  0
+/// ------------+---------+---+-----------------------
+///  0  1  1  1 |   REG   | 0 |          DATA
+/// ------------+---------+---+-----------------------
+
+// No pattern, as all immediate -> register moves are matched to the MOVI pseudo
+let Defs = [CCR] in
+def MOVQ : MxInst<(outs MxDRD32:$dst), (ins Mxi8imm:$imm),
+                  "moveq\t$imm, $dst",
+                  [(null_frag)]> {
+  let Inst = (descend 0b0111, (operand "$dst", 3), 0b0, (operand "$imm", 8));
+}
+
+//===----------------------------------------------------------------------===//
 // MOVEM
 //
 // The mask is already pre-processed by the save/restore spill hook
@@ -496,7 +515,23 @@ class MxPseudoMove_RR<MxType DST, MxType SRC, list<dag> PAT = []>
 
 class MxPseudoMove_RM<MxType DST, MxOperand SRCOpd, list<dag> PAT = []>
     : MxPseudo<(outs DST.ROp:$dst), (ins SRCOpd:$src), PAT>;
-}
+
+
+// These Pseudos handle loading immediates to registers.
+// They are expanded post-RA into either move or moveq instructions,
+// depending on size, destination register class, and immediate value.
+// This is done with pseudoinstructions in order to not constrain RA to
+// data registers if moveq matches.
+class MxPseudoMove_DI<MxType TYPE>
+    : MxPseudo<(outs TYPE.ROp:$dst), (ins TYPE.IOp:$src),
+               [(set TYPE.ROp:$dst, imm:$src)]>;
+
+// i8 imm -> reg can always be converted to moveq,
+// but we still emit a pseudo for consistency.
+def MOVI8di  : MxPseudoMove_DI<MxType8d>;
+def MOVI16ri : MxPseudoMove_DI<MxType16r>;
+def MOVI32ri : MxPseudoMove_DI<MxType32r>;
+} // let Defs = [CCR]
 
 /// This group of Pseudos is analogues to the real x86 extending moves, but
 /// since M68k does not have those we need to emulate. These instructions
diff --git a/llvm/lib/Target/M68k/M68kInstrInfo.cpp b/llvm/lib/Target/M68k/M68kInstrInfo.cpp
index d56fef9e9029..338db45782c9 100644
--- a/llvm/lib/Target/M68k/M68kInstrInfo.cpp
+++ b/llvm/lib/Target/M68k/M68kInstrInfo.cpp
@@ -346,6 +346,40 @@ void M68kInstrInfo::AddZExt(MachineBasicBlock &MBB,
   BuildMI(MBB, I, DL, get(And), Reg).addReg(Reg).addImm(Mask);
 }
 
+// Convert MOVI to MOVQ if the target is a data register and the immediate
+// fits in a sign-extended i8, otherwise emit a plain MOV.
+bool M68kInstrInfo::ExpandMOVI(MachineInstrBuilder &MIB, MVT MVTSize) const {
+  Register Reg = MIB->getOperand(0).getReg();
+  int64_t Imm = MIB->getOperand(1).getImm();
+  bool IsAddressReg = false;
+
+  const auto *DR32 = RI.getRegClass(M68k::DR32RegClassID);
+  const auto *AR32 = RI.getRegClass(M68k::AR32RegClassID);
+  const auto *AR16 = RI.getRegClass(M68k::AR16RegClassID);
+
+  if (AR16->contains(Reg) || AR32->contains(Reg))
+    IsAddressReg = true;
+
+  LLVM_DEBUG(dbgs() << "Expand " << *MIB.getInstr() << " to ");
+
+  if (MVTSize == MVT::i8 || (!IsAddressReg && Imm >= -128 && Imm <= 127)) {
+    LLVM_DEBUG(dbgs() << "MOVEQ\n");
+
+    // We need to assign to the full register to make IV happy
+    Register SReg =
+        MVTSize == MVT::i32 ? Reg : Register(RI.getMatchingMegaReg(Reg, DR32));
+    assert(SReg && "No viable MEGA register available");
+
+    MIB->setDesc(get(M68k::MOVQ));
+    MIB->getOperand(0).setReg(SReg);
+  } else {
+    LLVM_DEBUG(dbgs() << "MOVE\n");
+    MIB->setDesc(get(MVTSize == MVT::i16 ? M68k::MOV16ri : M68k::MOV32ri));
+  }
+
+  return true;
+}
+
 bool M68kInstrInfo::ExpandMOVX_RR(MachineInstrBuilder &MIB, MVT MVTDst,
                                   MVT MVTSrc) const {
   unsigned Move = MVTDst == MVT::i16 ? M68k::MOV16rr : M68k::MOV32rr;
diff --git a/llvm/lib/Target/M68k/M68kInstrInfo.h b/llvm/lib/Target/M68k/M68kInstrInfo.h
index 577967f2fdfc..d1e1e1cd9998 100644
--- a/llvm/lib/Target/M68k/M68kInstrInfo.h
+++ b/llvm/lib/Target/M68k/M68kInstrInfo.h
@@ -302,6 +302,9 @@ public:
   void AddZExt(MachineBasicBlock &MBB, MachineBasicBlock::iterator I,
                DebugLoc DL, unsigned Reg, MVT From, MVT To) const;
 
+  /// Move immediate to register
+  bool ExpandMOVI(MachineInstrBuilder &MIB, MVT MVTSize) const;
+
   /// Move across register classes without extension
   bool ExpandMOVX_RR(MachineInstrBuilder &MIB, MVT MVTDst, MVT MVTSrc) const;
 
diff --git a/llvm/lib/Target/Mips/MCTargetDesc/MipsELFStreamer.cpp b/llvm/lib/Target/Mips/MCTargetDesc/MipsELFStreamer.cpp
index 9843b6144343..e907e8d8a700 100644
--- a/llvm/lib/Target/Mips/MCTargetDesc/MipsELFStreamer.cpp
+++ b/llvm/lib/Target/Mips/MCTargetDesc/MipsELFStreamer.cpp
@@ -112,10 +112,11 @@ void MipsELFStreamer::EmitMipsOptionRecords() {
     I->EmitMipsOptionRecord();
 }
 
-MCELFStreamer *llvm::createMipsELFStreamer(
-    MCContext &Context, std::unique_ptr<MCAsmBackend> MAB,
-    std::unique_ptr<MCObjectWriter> OW, std::unique_ptr<MCCodeEmitter> Emitter,
-    bool RelaxAll) {
+MCELFStreamer *
+llvm::createMipsELFStreamer(MCContext &Context,
+                            std::unique_ptr<MCAsmBackend> MAB,
+                            std::unique_ptr<MCObjectWriter> OW,
+                            std::unique_ptr<MCCodeEmitter> Emitter) {
   return new MipsELFStreamer(Context, std::move(MAB), std::move(OW),
                              std::move(Emitter));
 }
diff --git a/llvm/lib/Target/Mips/MCTargetDesc/MipsELFStreamer.h b/llvm/lib/Target/Mips/MCTargetDesc/MipsELFStreamer.h
index ac70e40d4dfe..051806d2cfe8 100644
--- a/llvm/lib/Target/Mips/MCTargetDesc/MipsELFStreamer.h
+++ b/llvm/lib/Target/Mips/MCTargetDesc/MipsELFStreamer.h
@@ -75,8 +75,7 @@ public:
 MCELFStreamer *createMipsELFStreamer(MCContext &Context,
                                      std::unique_ptr<MCAsmBackend> MAB,
                                      std::unique_ptr<MCObjectWriter> OW,
-                                     std::unique_ptr<MCCodeEmitter> Emitter,
-                                     bool RelaxAll);
+                                     std::unique_ptr<MCCodeEmitter> Emitter);
 } // end namespace llvm
 
 #endif // LLVM_LIB_TARGET_MIPS_MCTARGETDESC_MIPSELFSTREAMER_H
diff --git a/llvm/lib/Target/Mips/MCTargetDesc/MipsMCNaCl.h b/llvm/lib/Target/Mips/MCTargetDesc/MipsMCNaCl.h
index a84ca8ccfb2d..2722e34b3f62 100644
--- a/llvm/lib/Target/Mips/MCTargetDesc/MipsMCNaCl.h
+++ b/llvm/lib/Target/Mips/MCTargetDesc/MipsMCNaCl.h
@@ -22,11 +22,10 @@ bool isBasePlusOffsetMemoryAccess(unsigned Opcode, unsigned *AddrIdx,
 bool baseRegNeedsLoadStoreMask(unsigned Reg);
 
 // This function creates an MCELFStreamer for Mips NaCl.
-MCELFStreamer *createMipsNaClELFStreamer(MCContext &Context,
-                                         std::unique_ptr<MCAsmBackend> TAB,
-                                         std::unique_ptr<MCObjectWriter> OW,
-                                         std::unique_ptr<MCCodeEmitter> Emitter,
-                                         bool RelaxAll);
+MCELFStreamer *
+createMipsNaClELFStreamer(MCContext &Context, std::unique_ptr<MCAsmBackend> TAB,
+                          std::unique_ptr<MCObjectWriter> OW,
+                          std::unique_ptr<MCCodeEmitter> Emitter);
 }
 
 #endif
diff --git a/llvm/lib/Target/Mips/MCTargetDesc/MipsMCTargetDesc.cpp b/llvm/lib/Target/Mips/MCTargetDesc/MipsMCTargetDesc.cpp
index d38b89f9a1f2..499cbd873e29 100644
--- a/llvm/lib/Target/Mips/MCTargetDesc/MipsMCTargetDesc.cpp
+++ b/llvm/lib/Target/Mips/MCTargetDesc/MipsMCTargetDesc.cpp
@@ -104,15 +104,14 @@ static MCInstPrinter *createMipsMCInstPrinter(const Triple &T,
 static MCStreamer *createMCStreamer(const Triple &T, MCContext &Context,
                                     std::unique_ptr<MCAsmBackend> &&MAB,
                                     std::unique_ptr<MCObjectWriter> &&OW,
-                                    std::unique_ptr<MCCodeEmitter> &&Emitter,
-                                    bool RelaxAll) {
+                                    std::unique_ptr<MCCodeEmitter> &&Emitter) {
   MCStreamer *S;
   if (!T.isOSNaCl())
     S = createMipsELFStreamer(Context, std::move(MAB), std::move(OW),
-                              std::move(Emitter), RelaxAll);
+                              std::move(Emitter));
   else
     S = createMipsNaClELFStreamer(Context, std::move(MAB), std::move(OW),
-                                  std::move(Emitter), RelaxAll);
+                                  std::move(Emitter));
   return S;
 }
 
diff --git a/llvm/lib/Target/Mips/MCTargetDesc/MipsNaClELFStreamer.cpp b/llvm/lib/Target/Mips/MCTargetDesc/MipsNaClELFStreamer.cpp
index 4ba0ae91e2f6..86194a9ebb61 100644
--- a/llvm/lib/Target/Mips/MCTargetDesc/MipsNaClELFStreamer.cpp
+++ b/llvm/lib/Target/Mips/MCTargetDesc/MipsNaClELFStreamer.cpp
@@ -259,15 +259,12 @@ bool baseRegNeedsLoadStoreMask(unsigned Reg) {
   return Reg != Mips::SP && Reg != Mips::T8;
 }
 
-MCELFStreamer *createMipsNaClELFStreamer(MCContext &Context,
-                                         std::unique_ptr<MCAsmBackend> TAB,
-                                         std::unique_ptr<MCObjectWriter> OW,
-                                         std::unique_ptr<MCCodeEmitter> Emitter,
-                                         bool RelaxAll) {
+MCELFStreamer *
+createMipsNaClELFStreamer(MCContext &Context, std::unique_ptr<MCAsmBackend> TAB,
+                          std::unique_ptr<MCObjectWriter> OW,
+                          std::unique_ptr<MCCodeEmitter> Emitter) {
   MipsNaClELFStreamer *S = new MipsNaClELFStreamer(
       Context, std::move(TAB), std::move(OW), std::move(Emitter));
-  if (RelaxAll)
-    S->getAssembler().setRelaxAll(true);
 
   // Set bundle-alignment as required by the NaCl ABI for the target.
   S->emitBundleAlignMode(MIPS_NACL_BUNDLE_ALIGN);
diff --git a/llvm/lib/Target/Mips/Mips32r6InstrInfo.td b/llvm/lib/Target/Mips/Mips32r6InstrInfo.td
index bef7607118ce..f609305bfee4 100644
--- a/llvm/lib/Target/Mips/Mips32r6InstrInfo.td
+++ b/llvm/lib/Target/Mips/Mips32r6InstrInfo.td
@@ -1117,6 +1117,22 @@ def : MipsPat<(select i32:$cond, immz, i32:$f),
               ISA_MIPS32R6;
 }
 
+// llvm.fmin/fmax operations.
+let AdditionalPredicates = [NotInMicroMips] in {
+  def : MipsPat<(fmaxnum f32:$lhs, f32:$rhs),
+                (MAX_S   f32:$lhs, f32:$rhs)>,
+                ISA_MIPS32R6;
+  def : MipsPat<(fmaxnum f64:$lhs, f64:$rhs),
+                (MAX_D   f64:$lhs, f64:$rhs)>,
+                ISA_MIPS32R6;
+  def : MipsPat<(fminnum f32:$lhs, f32:$rhs),
+                (MIN_S   f32:$lhs, f32:$rhs)>,
+                ISA_MIPS32R6;
+  def : MipsPat<(fminnum f64:$lhs, f64:$rhs),
+                (MIN_D   f64:$lhs, f64:$rhs)>,
+                ISA_MIPS32R6;
+}
+
 // Pseudo instructions
 let isCall = 1, isTerminator = 1, isReturn = 1, isBarrier = 1, hasDelaySlot = 1,
     hasExtraSrcRegAllocReq = 1, isCTI = 1, Defs = [AT], hasPostISelHook = 1 in {
diff --git a/llvm/lib/Target/Mips/MipsDelaySlotFiller.cpp b/llvm/lib/Target/Mips/MipsDelaySlotFiller.cpp
index cb98c04ff4e5..b525606b1f8f 100644
--- a/llvm/lib/Target/Mips/MipsDelaySlotFiller.cpp
+++ b/llvm/lib/Target/Mips/MipsDelaySlotFiller.cpp
@@ -365,7 +365,8 @@ void RegDefsUses::setCallerSaved(const MachineInstr &MI) {
   // Add RA/RA_64 to Defs to prevent users of RA/RA_64 from going into
   // the delay slot. The reason is that RA/RA_64 must not be changed
   // in the delay slot so that the callee can return to the caller.
-  if (MI.definesRegister(Mips::RA) || MI.definesRegister(Mips::RA_64)) {
+  if (MI.definesRegister(Mips::RA, /*TRI=*/nullptr) ||
+      MI.definesRegister(Mips::RA_64, /*TRI=*/nullptr)) {
     Defs.set(Mips::RA);
     Defs.set(Mips::RA_64);
   }
diff --git a/llvm/lib/Target/Mips/MipsExpandPseudo.cpp b/llvm/lib/Target/Mips/MipsExpandPseudo.cpp
index d33852a04baf..199474fbd82d 100644
--- a/llvm/lib/Target/Mips/MipsExpandPseudo.cpp
+++ b/llvm/lib/Target/Mips/MipsExpandPseudo.cpp
@@ -479,13 +479,13 @@ bool MipsExpandPseudo::expandAtomicBinOpSubword(
     BuildMI(loopMBB, DL, TII->get(Mips::SRAV), StoreVal)
         .addReg(OldVal)
         .addReg(ShiftAmnt);
-    if (STI->hasMips32r2() && !IsUnsigned) {
-      BuildMI(loopMBB, DL, TII->get(SEOp), StoreVal).addReg(StoreVal);
-    } else if (STI->hasMips32r2() && IsUnsigned) {
+    if (IsUnsigned) {
       const unsigned OpMask = SEOp == Mips::SEH ? 0xffff : 0xff;
       BuildMI(loopMBB, DL, TII->get(Mips::ANDi), StoreVal)
           .addReg(StoreVal)
           .addImm(OpMask);
+    } else if (STI->hasMips32r2()) {
+      BuildMI(loopMBB, DL, TII->get(SEOp), StoreVal).addReg(StoreVal);
     } else {
       const unsigned ShiftImm = SEOp == Mips::SEH ? 16 : 24;
       const unsigned SROp = IsUnsigned ? Mips::SRL : Mips::SRA;
diff --git a/llvm/lib/Target/Mips/MipsISelLowering.cpp b/llvm/lib/Target/Mips/MipsISelLowering.cpp
index 7bc66b2d9f4b..8f7c47370ee5 100644
--- a/llvm/lib/Target/Mips/MipsISelLowering.cpp
+++ b/llvm/lib/Target/Mips/MipsISelLowering.cpp
@@ -358,6 +358,15 @@ MipsTargetLowering::MipsTargetLowering(const MipsTargetMachine &TM,
   setOperationAction(ISD::FCOPYSIGN,          MVT::f64,   Custom);
   setOperationAction(ISD::FP_TO_SINT,         MVT::i32,   Custom);
 
+  // Lower fmin and fmax operations for MIPS R6.
+  // Instructions are defined but never used.
+  if (Subtarget.hasMips32r6() || Subtarget.hasMips64r6()) {
+    setOperationAction(ISD::FMINNUM, MVT::f32, Legal);
+    setOperationAction(ISD::FMINNUM, MVT::f64, Legal);
+    setOperationAction(ISD::FMAXNUM, MVT::f32, Legal);
+    setOperationAction(ISD::FMAXNUM, MVT::f64, Legal);
+  }
+
   if (Subtarget.isGP64bit()) {
     setOperationAction(ISD::GlobalAddress,      MVT::i64,   Custom);
     setOperationAction(ISD::BlockAddress,       MVT::i64,   Custom);
diff --git a/llvm/lib/Target/Mips/MipsInstrInfo.cpp b/llvm/lib/Target/Mips/MipsInstrInfo.cpp
index 392cc15d7943..f4fba5e53132 100644
--- a/llvm/lib/Target/Mips/MipsInstrInfo.cpp
+++ b/llvm/lib/Target/Mips/MipsInstrInfo.cpp
@@ -619,7 +619,7 @@ bool MipsInstrInfo::SafeInLoadDelaySlot(const MachineInstr &MIInSlot,
     return false;
 
   return !llvm::any_of(LoadMI.defs(), [&](const MachineOperand &Op) {
-    return Op.isReg() && MIInSlot.readsRegister(Op.getReg());
+    return Op.isReg() && MIInSlot.readsRegister(Op.getReg(), /*TRI=*/nullptr);
   });
 }
 
@@ -699,7 +699,7 @@ MipsInstrInfo::genInstrWithNewOpc(unsigned NewOpc,
   bool BranchWithZeroOperand = false;
   if (I->isBranch() && !I->isPseudo()) {
     auto TRI = I->getParent()->getParent()->getSubtarget().getRegisterInfo();
-    ZeroOperandPosition = I->findRegisterUseOperandIdx(Mips::ZERO, false, TRI);
+    ZeroOperandPosition = I->findRegisterUseOperandIdx(Mips::ZERO, TRI, false);
     BranchWithZeroOperand = ZeroOperandPosition != -1;
   }
 
diff --git a/llvm/lib/Target/NVPTX/NVPTXISelDAGToDAG.cpp b/llvm/lib/Target/NVPTX/NVPTXISelDAGToDAG.cpp
index a362709c98ef..595395bb1b4b 100644
--- a/llvm/lib/Target/NVPTX/NVPTXISelDAGToDAG.cpp
+++ b/llvm/lib/Target/NVPTX/NVPTXISelDAGToDAG.cpp
@@ -30,6 +30,10 @@ using namespace llvm;
 #define DEBUG_TYPE "nvptx-isel"
 #define PASS_NAME "NVPTX DAG->DAG Pattern Instruction Selection"
 
+static cl::opt<bool>
+    EnableRsqrtOpt("nvptx-rsqrt-approx-opt", cl::init(true), cl::Hidden,
+                   cl::desc("Enable reciprocal sqrt optimization"));
+
 /// createNVPTXISelDag - This pass converts a legalized DAG into a
 /// NVPTX-specific DAG, ready for instruction scheduling.
 FunctionPass *llvm::createNVPTXISelDag(NVPTXTargetMachine &TM,
@@ -74,6 +78,8 @@ bool NVPTXDAGToDAGISel::allowUnsafeFPMath() const {
   return TL->allowUnsafeFPMath(*MF);
 }
 
+bool NVPTXDAGToDAGISel::doRsqrtOpt() const { return EnableRsqrtOpt; }
+
 /// Select - Select instructions not customized! Used for
 /// expanded, promoted and normal instructions.
 void NVPTXDAGToDAGISel::Select(SDNode *N) {
diff --git a/llvm/lib/Target/NVPTX/NVPTXISelDAGToDAG.h b/llvm/lib/Target/NVPTX/NVPTXISelDAGToDAG.h
index 10822f87cef3..7a7774744bc7 100644
--- a/llvm/lib/Target/NVPTX/NVPTXISelDAGToDAG.h
+++ b/llvm/lib/Target/NVPTX/NVPTXISelDAGToDAG.h
@@ -36,6 +36,7 @@ class LLVM_LIBRARY_VISIBILITY NVPTXDAGToDAGISel : public SelectionDAGISel {
   bool useF32FTZ() const;
   bool allowFMA() const;
   bool allowUnsafeFPMath() const;
+  bool doRsqrtOpt() const;
 
 public:
   static char ID;
diff --git a/llvm/lib/Target/NVPTX/NVPTXInstrInfo.td b/llvm/lib/Target/NVPTX/NVPTXInstrInfo.td
index 931292c7fd60..897ee89323f0 100644
--- a/llvm/lib/Target/NVPTX/NVPTXInstrInfo.td
+++ b/llvm/lib/Target/NVPTX/NVPTXInstrInfo.td
@@ -142,6 +142,7 @@ def hasLDU : Predicate<"Subtarget->hasLDU()">;
 
 def doF32FTZ : Predicate<"useF32FTZ()">;
 def doNoF32FTZ : Predicate<"!useF32FTZ()">;
+def doRsqrtOpt : Predicate<"doRsqrtOpt()">;
 
 def doMulWide      : Predicate<"doMulWide">;
 
diff --git a/llvm/lib/Target/NVPTX/NVPTXIntrinsics.td b/llvm/lib/Target/NVPTX/NVPTXIntrinsics.td
index ec9170b4e41e..5f6e28283c5d 100644
--- a/llvm/lib/Target/NVPTX/NVPTXIntrinsics.td
+++ b/llvm/lib/Target/NVPTX/NVPTXIntrinsics.td
@@ -1171,11 +1171,36 @@ def : Pat<(int_nvvm_sqrt_f Float32Regs:$a),
 def INT_NVVM_RSQRT_APPROX_FTZ_F
   : F_MATH_1<"rsqrt.approx.ftz.f32 \t$dst, $src0;", Float32Regs, Float32Regs,
     int_nvvm_rsqrt_approx_ftz_f>;
+def INT_NVVM_RSQRT_APPROX_FTZ_D
+  : F_MATH_1<"rsqrt.approx.ftz.f64 \t$dst, $src0;", Float64Regs, Float64Regs,
+    int_nvvm_rsqrt_approx_ftz_d>;
+
 def INT_NVVM_RSQRT_APPROX_F : F_MATH_1<"rsqrt.approx.f32 \t$dst, $src0;",
   Float32Regs, Float32Regs, int_nvvm_rsqrt_approx_f>;
 def INT_NVVM_RSQRT_APPROX_D : F_MATH_1<"rsqrt.approx.f64 \t$dst, $src0;",
   Float64Regs, Float64Regs, int_nvvm_rsqrt_approx_d>;
 
+// 1.0f / sqrt_approx -> rsqrt_approx
+def: Pat<(fdiv FloatConst1, (int_nvvm_sqrt_approx_f Float32Regs:$a)),
+         (INT_NVVM_RSQRT_APPROX_F Float32Regs:$a)>,
+         Requires<[doRsqrtOpt]>;
+def: Pat<(fdiv FloatConst1, (int_nvvm_sqrt_approx_ftz_f Float32Regs:$a)),
+         (INT_NVVM_RSQRT_APPROX_FTZ_F Float32Regs:$a)>,
+         Requires<[doRsqrtOpt]>;
+// same for int_nvvm_sqrt_f when non-precision sqrt is requested
+def: Pat<(fdiv FloatConst1, (int_nvvm_sqrt_f Float32Regs:$a)),
+         (INT_NVVM_RSQRT_APPROX_F Float32Regs:$a)>,
+         Requires<[doRsqrtOpt, do_SQRTF32_APPROX, doNoF32FTZ]>;
+def: Pat<(fdiv FloatConst1, (int_nvvm_sqrt_f Float32Regs:$a)),
+         (INT_NVVM_RSQRT_APPROX_FTZ_F Float32Regs:$a)>,
+         Requires<[doRsqrtOpt, do_SQRTF32_APPROX, doF32FTZ]>;
+
+def: Pat<(fdiv FloatConst1, (fsqrt Float32Regs:$a)),
+         (INT_NVVM_RSQRT_APPROX_F Float32Regs:$a)>,
+         Requires<[doRsqrtOpt, do_SQRTF32_APPROX, doNoF32FTZ]>;
+def: Pat<(fdiv FloatConst1, (fsqrt Float32Regs:$a)),
+         (INT_NVVM_RSQRT_APPROX_FTZ_F Float32Regs:$a)>,
+         Requires<[doRsqrtOpt, do_SQRTF32_APPROX, doF32FTZ]>;
 //
 // Add
 //
diff --git a/llvm/lib/Target/NVPTX/NVPTXTargetTransformInfo.h b/llvm/lib/Target/NVPTX/NVPTXTargetTransformInfo.h
index 3ce2675560c4..5a4fbab97f95 100644
--- a/llvm/lib/Target/NVPTX/NVPTXTargetTransformInfo.h
+++ b/llvm/lib/Target/NVPTX/NVPTXTargetTransformInfo.h
@@ -98,7 +98,7 @@ public:
       unsigned Opcode, Type *Ty, TTI::TargetCostKind CostKind,
       TTI::OperandValueInfo Op1Info = {TTI::OK_AnyValue, TTI::OP_None},
       TTI::OperandValueInfo Op2Info = {TTI::OK_AnyValue, TTI::OP_None},
-      ArrayRef<const Value *> Args = ArrayRef<const Value *>(),
+      ArrayRef<const Value *> Args = std::nullopt,
       const Instruction *CxtI = nullptr);
 
   void getUnrollingPreferences(Loop *L, ScalarEvolution &SE,
diff --git a/llvm/lib/Target/PowerPC/MCTargetDesc/PPCMCTargetDesc.cpp b/llvm/lib/Target/PowerPC/MCTargetDesc/PPCMCTargetDesc.cpp
index b849b7be7b7b..241078b03873 100644
--- a/llvm/lib/Target/PowerPC/MCTargetDesc/PPCMCTargetDesc.cpp
+++ b/llvm/lib/Target/PowerPC/MCTargetDesc/PPCMCTargetDesc.cpp
@@ -203,15 +203,16 @@ static MCStreamer *
 createPPCELFStreamer(const Triple &T, MCContext &Context,
                      std::unique_ptr<MCAsmBackend> &&MAB,
                      std::unique_ptr<MCObjectWriter> &&OW,
-                     std::unique_ptr<MCCodeEmitter> &&Emitter, bool RelaxAll) {
+                     std::unique_ptr<MCCodeEmitter> &&Emitter) {
   return createPPCELFStreamer(Context, std::move(MAB), std::move(OW),
                               std::move(Emitter));
 }
 
-static MCStreamer *createPPCXCOFFStreamer(
-    const Triple &T, MCContext &Context, std::unique_ptr<MCAsmBackend> &&MAB,
-    std::unique_ptr<MCObjectWriter> &&OW,
-    std::unique_ptr<MCCodeEmitter> &&Emitter, bool RelaxAll) {
+static MCStreamer *
+createPPCXCOFFStreamer(const Triple &T, MCContext &Context,
+                       std::unique_ptr<MCAsmBackend> &&MAB,
+                       std::unique_ptr<MCObjectWriter> &&OW,
+                       std::unique_ptr<MCCodeEmitter> &&Emitter) {
   return createPPCXCOFFStreamer(Context, std::move(MAB), std::move(OW),
                                 std::move(Emitter));
 }
diff --git a/llvm/lib/Target/PowerPC/PPCAsmPrinter.cpp b/llvm/lib/Target/PowerPC/PPCAsmPrinter.cpp
index 6e1002c45d81..51b79dc2b04b 100644
--- a/llvm/lib/Target/PowerPC/PPCAsmPrinter.cpp
+++ b/llvm/lib/Target/PowerPC/PPCAsmPrinter.cpp
@@ -27,7 +27,7 @@
 #include "PPCTargetStreamer.h"
 #include "TargetInfo/PowerPCTargetInfo.h"
 #include "llvm/ADT/MapVector.h"
-#include "llvm/ADT/SmallPtrSet.h"
+#include "llvm/ADT/SetVector.h"
 #include "llvm/ADT/Statistic.h"
 #include "llvm/ADT/StringExtras.h"
 #include "llvm/ADT/StringRef.h"
@@ -242,7 +242,7 @@ class PPCAIXAsmPrinter : public PPCAsmPrinter {
 private:
   /// Symbols lowered from ExternalSymbolSDNodes, we will need to emit extern
   /// linkage for them in AIX.
-  SmallPtrSet<MCSymbol *, 8> ExtSymSDNodeSymbols;
+  SmallSetVector<MCSymbol *, 8> ExtSymSDNodeSymbols;
 
   /// A format indicator and unique trailing identifier to form part of the
   /// sinit/sterm function names.
diff --git a/llvm/lib/Target/PowerPC/PPCCTRLoops.cpp b/llvm/lib/Target/PowerPC/PPCCTRLoops.cpp
index 5299c0c924b3..491779124e8a 100644
--- a/llvm/lib/Target/PowerPC/PPCCTRLoops.cpp
+++ b/llvm/lib/Target/PowerPC/PPCCTRLoops.cpp
@@ -123,10 +123,12 @@ bool PPCCTRLoops::isCTRClobber(MachineInstr *MI, bool CheckReads) const {
     // CTR defination inside the callee of a call instruction will not impact
     // the defination of MTCTRloop, so we can use definesRegister() for the
     // check, no need to check the regmask.
-    return MI->definesRegister(PPC::CTR) || MI->definesRegister(PPC::CTR8);
+    return MI->definesRegister(PPC::CTR, /*TRI=*/nullptr) ||
+           MI->definesRegister(PPC::CTR8, /*TRI=*/nullptr);
   }
 
-  if (MI->modifiesRegister(PPC::CTR) || MI->modifiesRegister(PPC::CTR8))
+  if (MI->modifiesRegister(PPC::CTR, /*TRI=*/nullptr) ||
+      MI->modifiesRegister(PPC::CTR8, /*TRI=*/nullptr))
     return true;
 
   if (MI->getDesc().isCall())
@@ -134,7 +136,8 @@ bool PPCCTRLoops::isCTRClobber(MachineInstr *MI, bool CheckReads) const {
 
   // We define the CTR in the loop preheader, so if there is any CTR reader in
   // the loop, we also can not use CTR loop form.
-  if (MI->readsRegister(PPC::CTR) || MI->readsRegister(PPC::CTR8))
+  if (MI->readsRegister(PPC::CTR, /*TRI=*/nullptr) ||
+      MI->readsRegister(PPC::CTR8, /*TRI=*/nullptr))
     return true;
 
   return false;
diff --git a/llvm/lib/Target/PowerPC/PPCInstr64Bit.td b/llvm/lib/Target/PowerPC/PPCInstr64Bit.td
index a9359794a641..9af8ada78376 100644
--- a/llvm/lib/Target/PowerPC/PPCInstr64Bit.td
+++ b/llvm/lib/Target/PowerPC/PPCInstr64Bit.td
@@ -76,7 +76,7 @@ let Interpretation64Bit = 1, isCodeGenOnly = 1 in {
 let isTerminator = 1, isBarrier = 1, PPC970_Unit = 7, hasSideEffects = 0 in {
   let isReturn = 1, isPredicable = 1, Uses = [LR8, RM] in
     def BLR8 : XLForm_2_ext<19, 16, 20, 0, 0, (outs), (ins), "blr", IIC_BrB,
-                            [(retglue)]>, Requires<[In64BitMode]>;
+                            [(PPCretglue)]>, Requires<[In64BitMode]>;
   let isBranch = 1, isIndirectBranch = 1, Uses = [CTR8] in {
     let isPredicable = 1 in
       def BCTR8 : XLForm_2_ext<19, 528, 20, 0, 0, (outs), (ins), "bctr", IIC_BrB,
diff --git a/llvm/lib/Target/PowerPC/PPCInstrInfo.cpp b/llvm/lib/Target/PowerPC/PPCInstrInfo.cpp
index b32f178ca38e..9e56de732c58 100644
--- a/llvm/lib/Target/PowerPC/PPCInstrInfo.cpp
+++ b/llvm/lib/Target/PowerPC/PPCInstrInfo.cpp
@@ -2125,7 +2125,8 @@ bool PPCInstrInfo::foldImmediate(MachineInstr &UseMI, MachineInstr &DefMI,
 
 static bool MBBDefinesCTR(MachineBasicBlock &MBB) {
   for (MachineInstr &MI : MBB)
-    if (MI.definesRegister(PPC::CTR) || MI.definesRegister(PPC::CTR8))
+    if (MI.definesRegister(PPC::CTR, /*TRI=*/nullptr) ||
+        MI.definesRegister(PPC::CTR8, /*TRI=*/nullptr))
       return true;
   return false;
 }
@@ -2731,19 +2732,19 @@ bool PPCInstrInfo::optimizeCompareInstr(MachineInstr &CmpInstr, Register SrcReg,
     MI->setDesc(NewDesc);
 
     for (MCPhysReg ImpDef : NewDesc.implicit_defs()) {
-      if (!MI->definesRegister(ImpDef)) {
+      if (!MI->definesRegister(ImpDef, /*TRI=*/nullptr)) {
         MI->addOperand(*MI->getParent()->getParent(),
                        MachineOperand::CreateReg(ImpDef, true, true));
       }
     }
     for (MCPhysReg ImpUse : NewDesc.implicit_uses()) {
-      if (!MI->readsRegister(ImpUse)) {
+      if (!MI->readsRegister(ImpUse, /*TRI=*/nullptr)) {
         MI->addOperand(*MI->getParent()->getParent(),
                        MachineOperand::CreateReg(ImpUse, false, true));
       }
     }
   }
-  assert(MI->definesRegister(PPC::CR0) &&
+  assert(MI->definesRegister(PPC::CR0, /*TRI=*/nullptr) &&
          "Record-form instruction does not define cr0?");
 
   // Modify the condition code of operands in OperandsToUpdate.
@@ -2793,7 +2794,7 @@ bool PPCInstrInfo::optimizeCmpPostRA(MachineInstr &CmpMI) const {
 
   bool SrcRegHasOtherUse = false;
   MachineInstr *SrcMI = getDefMIPostRA(SrcReg, CmpMI, SrcRegHasOtherUse);
-  if (!SrcMI || !SrcMI->definesRegister(SrcReg))
+  if (!SrcMI || !SrcMI->definesRegister(SrcReg, /*TRI=*/nullptr))
     return false;
 
   MachineOperand RegMO = CmpMI.getOperand(0);
@@ -2806,7 +2807,7 @@ bool PPCInstrInfo::optimizeCmpPostRA(MachineInstr &CmpMI) const {
   bool IsCRRegKilled = false;
   if (!isRegElgibleForForwarding(RegMO, *SrcMI, CmpMI, false, IsCRRegKilled,
                                  SeenUseOfCRReg) ||
-      SrcMI->definesRegister(CRReg) || SeenUseOfCRReg)
+      SrcMI->definesRegister(CRReg, /*TRI=*/nullptr) || SeenUseOfCRReg)
     return false;
 
   int SrcMIOpc = SrcMI->getOpcode();
@@ -2823,7 +2824,7 @@ bool PPCInstrInfo::optimizeCmpPostRA(MachineInstr &CmpMI) const {
       .addReg(CRReg, RegState::ImplicitDefine);
   SrcMI->clearRegisterDeads(CRReg);
 
-  assert(SrcMI->definesRegister(PPC::CR0) &&
+  assert(SrcMI->definesRegister(PPC::CR0, /*TRI=*/nullptr) &&
          "Record-form instruction does not define cr0?");
 
   LLVM_DEBUG(dbgs() << "with: ");
@@ -3293,7 +3294,7 @@ void PPCInstrInfo::replaceInstrOperandWithImm(MachineInstr &MI,
   // result its number of explicit operands may be changed, thus the begin of
   // implicit operand is changed.
   const TargetRegisterInfo *TRI = &getRegisterInfo();
-  int UseOpIdx = MI.findRegisterUseOperandIdx(InUseReg, false, TRI);
+  int UseOpIdx = MI.findRegisterUseOperandIdx(InUseReg, TRI, false);
   if (UseOpIdx >= 0) {
     MachineOperand &MO = MI.getOperand(UseOpIdx);
     if (MO.isImplicit())
diff --git a/llvm/lib/Target/PowerPC/PPCInstrInfo.td b/llvm/lib/Target/PowerPC/PPCInstrInfo.td
index 261b9a3d1dff..7929a781dbda 100644
--- a/llvm/lib/Target/PowerPC/PPCInstrInfo.td
+++ b/llvm/lib/Target/PowerPC/PPCInstrInfo.td
@@ -336,7 +336,7 @@ def PPCbctrl_load_toc_rm : SDNode<"PPCISD::BCTRL_LOAD_TOC_RM",
                                   [SDNPHasChain, SDNPOptInGlue, SDNPOutGlue,
                                    SDNPVariadic]>;
 
-def retglue       : SDNode<"PPCISD::RET_GLUE", SDTNone,
+def PPCretglue   : SDNode<"PPCISD::RET_GLUE", SDTNone,
                            [SDNPHasChain, SDNPOptInGlue, SDNPVariadic]>;
 
 def PPCtc_return : SDNode<"PPCISD::TC_RETURN", SDT_PPCTC_ret,
@@ -1299,7 +1299,7 @@ def RESTORE_CRBIT : PPCEmitTimePseudo<(outs crbitrc:$cond), (ins memri:$F),
 let isTerminator = 1, isBarrier = 1, PPC970_Unit = 7, hasSideEffects = 0 in {
   let isPredicable = 1, isReturn = 1, Uses = [LR, RM] in
     def BLR : XLForm_2_ext<19, 16, 20, 0, 0, (outs), (ins), "blr", IIC_BrB,
-                           [(retglue)]>, Requires<[In32BitMode]>;
+                           [(PPCretglue)]>, Requires<[In32BitMode]>;
   let isBranch = 1, isIndirectBranch = 1, Uses = [CTR] in {
     let isPredicable = 1 in
       def BCTR : XLForm_2_ext<19, 528, 20, 0, 0, (outs), (ins), "bctr", IIC_BrB,
diff --git a/llvm/lib/Target/PowerPC/PPCMIPeephole.cpp b/llvm/lib/Target/PowerPC/PPCMIPeephole.cpp
index 494e4b52a5b5..c6db8a7bbeb8 100644
--- a/llvm/lib/Target/PowerPC/PPCMIPeephole.cpp
+++ b/llvm/lib/Target/PowerPC/PPCMIPeephole.cpp
@@ -45,6 +45,7 @@
 #include "llvm/CodeGen/MachineRegisterInfo.h"
 #include "llvm/InitializePasses.h"
 #include "llvm/Support/Debug.h"
+#include "llvm/Support/DebugCounter.h"
 
 using namespace llvm;
 
@@ -95,6 +96,13 @@ static cl::opt<bool>
                            cl::desc("enable optimization of conditional traps"),
                            cl::init(false), cl::Hidden);
 
+DEBUG_COUNTER(
+    PeepholeXToICounter, "ppc-xtoi-peephole",
+    "Controls whether PPC reg+reg to reg+imm peephole is performed on a MI");
+
+DEBUG_COUNTER(PeepholePerOpCounter, "ppc-per-op-peephole",
+              "Controls whether PPC per opcode peephole is performed on a MI");
+
 namespace {
 
 struct PPCMIPeephole : public MachineFunctionPass {
@@ -469,6 +477,9 @@ bool PPCMIPeephole::simplifyCode() {
           if (MI.isDebugInstr())
             continue;
 
+          if (!DebugCounter::shouldExecute(PeepholeXToICounter))
+            continue;
+
           SmallSet<Register, 4> RRToRIRegsToUpdate;
           if (!TII->convertToImmediateForm(MI, RRToRIRegsToUpdate))
             continue;
@@ -538,6 +549,9 @@ bool PPCMIPeephole::simplifyCode() {
       if (MI.isDebugInstr())
         continue;
 
+      if (!DebugCounter::shouldExecute(PeepholePerOpCounter))
+        continue;
+
       // Per-opcode peepholes.
       switch (MI.getOpcode()) {
 
diff --git a/llvm/lib/Target/PowerPC/PPCPreEmitPeephole.cpp b/llvm/lib/Target/PowerPC/PPCPreEmitPeephole.cpp
index 6e3bf26a598a..a182be3ea712 100644
--- a/llvm/lib/Target/PowerPC/PPCPreEmitPeephole.cpp
+++ b/llvm/lib/Target/PowerPC/PPCPreEmitPeephole.cpp
@@ -158,7 +158,7 @@ static bool hasPCRelativeForm(MachineInstr &Use) {
              ++AfterBBI) {
           // Track the operand that kill Reg. We would unset the kill flag of
           // the operand if there is a following redundant load immediate.
-          int KillIdx = AfterBBI->findRegisterUseOperandIdx(Reg, true, TRI);
+          int KillIdx = AfterBBI->findRegisterUseOperandIdx(Reg, TRI, true);
 
           // We can't just clear implicit kills, so if we encounter one, stop
           // looking further.
@@ -204,7 +204,7 @@ static bool hasPCRelativeForm(MachineInstr &Use) {
               DeadOrKillToUnset->setIsKill(false);
           }
           DeadOrKillToUnset =
-              AfterBBI->findRegisterDefOperand(Reg, true, true, TRI);
+              AfterBBI->findRegisterDefOperand(Reg, TRI, true, true);
           if (DeadOrKillToUnset)
             LLVM_DEBUG(dbgs()
                        << " Dead flag of " << *DeadOrKillToUnset << " from "
diff --git a/llvm/lib/Target/PowerPC/PPCRegisterInfo.cpp b/llvm/lib/Target/PowerPC/PPCRegisterInfo.cpp
index 0f450a4bf969..7e4cd6c72aa8 100644
--- a/llvm/lib/Target/PowerPC/PPCRegisterInfo.cpp
+++ b/llvm/lib/Target/PowerPC/PPCRegisterInfo.cpp
@@ -1013,8 +1013,8 @@ void PPCRegisterInfo::lowerCRRestore(MachineBasicBlock::iterator II,
 
   Register Reg = MF.getRegInfo().createVirtualRegister(LP64 ? G8RC : GPRC);
   Register DestReg = MI.getOperand(0).getReg();
-  assert(MI.definesRegister(DestReg) &&
-    "RESTORE_CR does not define its destination");
+  assert(MI.definesRegister(DestReg, /*TRI=*/nullptr) &&
+         "RESTORE_CR does not define its destination");
 
   addFrameReference(BuildMI(MBB, II, dl, TII.get(LP64 ? PPC::LWZ8 : PPC::LWZ),
                               Reg), FrameIndex);
@@ -1175,8 +1175,8 @@ void PPCRegisterInfo::lowerCRBitRestore(MachineBasicBlock::iterator II,
 
   Register Reg = MF.getRegInfo().createVirtualRegister(LP64 ? G8RC : GPRC);
   Register DestReg = MI.getOperand(0).getReg();
-  assert(MI.definesRegister(DestReg) &&
-    "RESTORE_CRBIT does not define its destination");
+  assert(MI.definesRegister(DestReg, /*TRI=*/nullptr) &&
+         "RESTORE_CRBIT does not define its destination");
 
   addFrameReference(BuildMI(MBB, II, dl, TII.get(LP64 ? PPC::LWZ8 : PPC::LWZ),
                               Reg), FrameIndex);
@@ -1363,7 +1363,7 @@ void PPCRegisterInfo::lowerACCRestore(MachineBasicBlock::iterator II,
   DebugLoc DL = MI.getDebugLoc();
 
   Register DestReg = MI.getOperand(0).getReg();
-  assert(MI.definesRegister(DestReg) &&
+  assert(MI.definesRegister(DestReg, /*TRI=*/nullptr) &&
          "RESTORE_ACC does not define its destination");
 
   bool IsPrimed = PPC::ACCRCRegClass.contains(DestReg);
@@ -1491,7 +1491,7 @@ void PPCRegisterInfo::lowerQuadwordRestore(MachineBasicBlock::iterator II,
   DebugLoc DL = MI.getDebugLoc();
 
   Register DestReg = MI.getOperand(0).getReg();
-  assert(MI.definesRegister(DestReg) &&
+  assert(MI.definesRegister(DestReg, /*TRI=*/nullptr) &&
          "RESTORE_QUADWORD does not define its destination");
 
   Register Reg = PPC::X0 + (DestReg - PPC::G8p0) * 2;
diff --git a/llvm/lib/Target/PowerPC/PPCTargetTransformInfo.h b/llvm/lib/Target/PowerPC/PPCTargetTransformInfo.h
index 36006dd7df73..061f5da5daf5 100644
--- a/llvm/lib/Target/PowerPC/PPCTargetTransformInfo.h
+++ b/llvm/lib/Target/PowerPC/PPCTargetTransformInfo.h
@@ -106,7 +106,7 @@ public:
       unsigned Opcode, Type *Ty, TTI::TargetCostKind CostKind,
       TTI::OperandValueInfo Op1Info = {TTI::OK_AnyValue, TTI::OP_None},
       TTI::OperandValueInfo Op2Info = {TTI::OK_AnyValue, TTI::OP_None},
-      ArrayRef<const Value *> Args = ArrayRef<const Value *>(),
+      ArrayRef<const Value *> Args = std::nullopt,
       const Instruction *CxtI = nullptr);
   InstructionCost getShuffleCost(TTI::ShuffleKind Kind, Type *Tp,
                                  ArrayRef<int> Mask,
diff --git a/llvm/lib/Target/RISCV/AsmParser/RISCVAsmParser.cpp b/llvm/lib/Target/RISCV/AsmParser/RISCVAsmParser.cpp
index d926ccdb59e1..8ac79ddce595 100644
--- a/llvm/lib/Target/RISCV/AsmParser/RISCVAsmParser.cpp
+++ b/llvm/lib/Target/RISCV/AsmParser/RISCVAsmParser.cpp
@@ -38,7 +38,7 @@
 #include "llvm/Support/CommandLine.h"
 #include "llvm/Support/MathExtras.h"
 #include "llvm/Support/RISCVAttributes.h"
-#include "llvm/Support/RISCVISAInfo.h"
+#include "llvm/TargetParser/RISCVISAInfo.h"
 
 #include <limits>
 
@@ -117,7 +117,7 @@ class RISCVAsmParser : public MCTargetAsmParser {
 
   ParseStatus parseDirective(AsmToken DirectiveID) override;
 
-  bool parseVTypeToken(StringRef Identifier, VTypeState &State, unsigned &Sew,
+  bool parseVTypeToken(const AsmToken &Tok, VTypeState &State, unsigned &Sew,
                        unsigned &Lmul, bool &Fractional, bool &TailAgnostic,
                        bool &MaskAgnostic);
   bool generateVTypeError(SMLoc ErrorLoc);
@@ -2125,10 +2125,15 @@ ParseStatus RISCVAsmParser::parseJALOffset(OperandVector &Operands) {
   return parseImmediate(Operands);
 }
 
-bool RISCVAsmParser::parseVTypeToken(StringRef Identifier, VTypeState &State,
+bool RISCVAsmParser::parseVTypeToken(const AsmToken &Tok, VTypeState &State,
                                      unsigned &Sew, unsigned &Lmul,
                                      bool &Fractional, bool &TailAgnostic,
                                      bool &MaskAgnostic) {
+  if (Tok.isNot(AsmToken::Identifier))
+    return true;
+
+  StringRef Identifier = Tok.getIdentifier();
+
   switch (State) {
   case VTypeState_SEW:
     if (!Identifier.consume_front("e"))
@@ -2187,24 +2192,14 @@ ParseStatus RISCVAsmParser::parseVTypeI(OperandVector &Operands) {
 
   VTypeState State = VTypeState_SEW;
 
-  if (getLexer().isNot(AsmToken::Identifier))
-    return ParseStatus::NoMatch;
-
-  StringRef Identifier = getTok().getIdentifier();
-
-  if (parseVTypeToken(Identifier, State, Sew, Lmul, Fractional, TailAgnostic,
+  if (parseVTypeToken(getTok(), State, Sew, Lmul, Fractional, TailAgnostic,
                       MaskAgnostic))
     return ParseStatus::NoMatch;
 
   getLexer().Lex();
 
   while (parseOptionalToken(AsmToken::Comma)) {
-    if (getLexer().isNot(AsmToken::Identifier))
-      break;
-
-    Identifier = getTok().getIdentifier();
-
-    if (parseVTypeToken(Identifier, State, Sew, Lmul, Fractional, TailAgnostic,
+    if (parseVTypeToken(getTok(), State, Sew, Lmul, Fractional, TailAgnostic,
                         MaskAgnostic))
       break;
 
diff --git a/llvm/lib/Target/RISCV/Disassembler/RISCVDisassembler.cpp b/llvm/lib/Target/RISCV/Disassembler/RISCVDisassembler.cpp
index 998b9181efe6..b9e8e1f33d3a 100644
--- a/llvm/lib/Target/RISCV/Disassembler/RISCVDisassembler.cpp
+++ b/llvm/lib/Target/RISCV/Disassembler/RISCVDisassembler.cpp
@@ -44,6 +44,13 @@ public:
 
 private:
   void addSPOperands(MCInst &MI) const;
+
+  DecodeStatus getInstruction32(MCInst &Instr, uint64_t &Size,
+                                ArrayRef<uint8_t> Bytes, uint64_t Address,
+                                raw_ostream &CStream) const;
+  DecodeStatus getInstruction16(MCInst &Instr, uint64_t &Size,
+                                ArrayRef<uint8_t> Bytes, uint64_t Address,
+                                raw_ostream &CStream) const;
 };
 } // end anonymous namespace
 
@@ -182,7 +189,7 @@ static DecodeStatus DecodeGPRPairRegisterClass(MCInst &Inst, uint32_t RegNo,
   return MCDisassembler::Success;
 }
 
-static DecodeStatus DecodeSR07RegisterClass(MCInst &Inst, uint64_t RegNo,
+static DecodeStatus DecodeSR07RegisterClass(MCInst &Inst, uint32_t RegNo,
                                             uint64_t Address,
                                             const void *Decoder) {
   if (RegNo >= 8)
@@ -255,12 +262,12 @@ static DecodeStatus DecodeVRM8RegisterClass(MCInst &Inst, uint32_t RegNo,
   return MCDisassembler::Success;
 }
 
-static DecodeStatus decodeVMaskReg(MCInst &Inst, uint64_t RegNo,
+static DecodeStatus decodeVMaskReg(MCInst &Inst, uint32_t RegNo,
                                    uint64_t Address,
                                    const MCDisassembler *Decoder) {
-  if (RegNo > 2) {
+  if (RegNo >= 2)
     return MCDisassembler::Fail;
-  }
+
   MCRegister Reg = (RegNo == 0) ? RISCV::V0 : RISCV::NoRegister;
 
   Inst.addOperand(MCOperand::createReg(Reg));
@@ -361,13 +368,13 @@ static DecodeStatus decodeXTHeadMemPair(MCInst &Inst, uint32_t Insn,
                                         uint64_t Address,
                                         const MCDisassembler *Decoder);
 
-static DecodeStatus decodeZcmpRlist(MCInst &Inst, unsigned Imm,
+static DecodeStatus decodeZcmpRlist(MCInst &Inst, uint32_t Imm,
                                     uint64_t Address, const void *Decoder);
 
 static DecodeStatus decodeRegReg(MCInst &Inst, uint32_t Insn, uint64_t Address,
                                  const MCDisassembler *Decoder);
 
-static DecodeStatus decodeZcmpSpimm(MCInst &Inst, unsigned Imm,
+static DecodeStatus decodeZcmpSpimm(MCInst &Inst, uint32_t Imm,
                                     uint64_t Address, const void *Decoder);
 
 static DecodeStatus decodeCSSPushPopchk(MCInst &Inst, uint32_t Insn,
@@ -470,7 +477,7 @@ static DecodeStatus decodeXTHeadMemPair(MCInst &Inst, uint32_t Insn,
   return MCDisassembler::Success;
 }
 
-static DecodeStatus decodeZcmpRlist(MCInst &Inst, unsigned Imm,
+static DecodeStatus decodeZcmpRlist(MCInst &Inst, uint32_t Imm,
                                     uint64_t Address, const void *Decoder) {
   if (Imm <= 3)
     return MCDisassembler::Fail;
@@ -487,7 +494,7 @@ static DecodeStatus decodeRegReg(MCInst &Inst, uint32_t Insn, uint64_t Address,
   return MCDisassembler::Success;
 }
 
-static DecodeStatus decodeZcmpSpimm(MCInst &Inst, unsigned Imm,
+static DecodeStatus decodeZcmpSpimm(MCInst &Inst, uint32_t Imm,
                                     uint64_t Address, const void *Decoder) {
   Inst.addOperand(MCOperand::createImm(Imm));
   return MCDisassembler::Success;
@@ -502,21 +509,13 @@ void RISCVDisassembler::addSPOperands(MCInst &MI) const {
       MI.insert(MI.begin() + i, MCOperand::createReg(RISCV::X2));
 }
 
-DecodeStatus RISCVDisassembler::getInstruction(MCInst &MI, uint64_t &Size,
-                                               ArrayRef<uint8_t> Bytes,
-                                               uint64_t Address,
-                                               raw_ostream &CS) const {
-  // TODO: This will need modification when supporting instruction set
-  // extensions with instructions > 32-bits (up to 176 bits wide).
-  uint32_t Insn;
-  DecodeStatus Result;
-
 #define TRY_TO_DECODE_WITH_ADDITIONAL_OPERATION(FEATURE_CHECKS, DECODER_TABLE, \
                                                 DESC, ADDITIONAL_OPERATION)    \
   do {                                                                         \
     if (FEATURE_CHECKS) {                                                      \
       LLVM_DEBUG(dbgs() << "Trying " DESC ":\n");                              \
-      Result = decodeInstruction(DECODER_TABLE, MI, Insn, Address, this, STI); \
+      DecodeStatus Result =                                                    \
+          decodeInstruction(DECODER_TABLE, MI, Insn, Address, this, STI);      \
       if (Result != MCDisassembler::Fail) {                                    \
         ADDITIONAL_OPERATION;                                                  \
         return Result;                                                         \
@@ -532,104 +531,111 @@ DecodeStatus RISCVDisassembler::getInstruction(MCInst &MI, uint64_t &Size,
 #define TRY_TO_DECODE_FEATURE(FEATURE, DECODER_TABLE, DESC)                    \
   TRY_TO_DECODE(STI.hasFeature(FEATURE), DECODER_TABLE, DESC)
 
-  // It's a 32 bit instruction if bit 0 and 1 are 1.
-  if ((Bytes[0] & 0x3) == 0x3) {
-    if (Bytes.size() < 4) {
-      Size = 0;
-      return MCDisassembler::Fail;
-    }
-    Size = 4;
-
-    Insn = support::endian::read32le(Bytes.data());
-
-    TRY_TO_DECODE(STI.hasFeature(RISCV::FeatureStdExtZdinx) &&
-                      !STI.hasFeature(RISCV::Feature64Bit),
-                  DecoderTableRV32Zdinx32,
-                  "RV32Zdinx table (Double in Integer and rv32)");
-    TRY_TO_DECODE(STI.hasFeature(RISCV::FeatureStdExtZacas) &&
-                      !STI.hasFeature(RISCV::Feature64Bit),
-                  DecoderTableRV32Zacas32,
-                  "RV32Zacas table (Compare-And-Swap and rv32)");
-    TRY_TO_DECODE_FEATURE(RISCV::FeatureStdExtZfinx, DecoderTableRVZfinx32,
-                          "RVZfinx table (Float in Integer)");
-    TRY_TO_DECODE_FEATURE(RISCV::FeatureVendorXVentanaCondOps,
-                          DecoderTableXVentana32, "Ventana custom opcode table");
-    TRY_TO_DECODE_FEATURE(RISCV::FeatureVendorXTHeadBa, DecoderTableXTHeadBa32,
-                          "XTHeadBa custom opcode table");
-    TRY_TO_DECODE_FEATURE(RISCV::FeatureVendorXTHeadBb, DecoderTableXTHeadBb32,
-                          "XTHeadBb custom opcode table");
-    TRY_TO_DECODE_FEATURE(RISCV::FeatureVendorXTHeadBs, DecoderTableXTHeadBs32,
-                          "XTHeadBs custom opcode table");
-    TRY_TO_DECODE_FEATURE(RISCV::FeatureVendorXTHeadCondMov,
-                          DecoderTableXTHeadCondMov32,
-                          "XTHeadCondMov custom opcode table");
-    TRY_TO_DECODE_FEATURE(RISCV::FeatureVendorXTHeadCmo, DecoderTableXTHeadCmo32,
-                          "XTHeadCmo custom opcode table");
-    TRY_TO_DECODE_FEATURE(RISCV::FeatureVendorXTHeadFMemIdx,
-                          DecoderTableXTHeadFMemIdx32,
-                          "XTHeadFMemIdx custom opcode table");
-    TRY_TO_DECODE_FEATURE(RISCV::FeatureVendorXTHeadMac, DecoderTableXTHeadMac32,
-                          "XTHeadMac custom opcode table");
-    TRY_TO_DECODE_FEATURE(RISCV::FeatureVendorXTHeadMemIdx,
-                          DecoderTableXTHeadMemIdx32,
-                          "XTHeadMemIdx custom opcode table");
-    TRY_TO_DECODE_FEATURE(RISCV::FeatureVendorXTHeadMemPair,
-                          DecoderTableXTHeadMemPair32,
-                          "XTHeadMemPair custom opcode table");
-    TRY_TO_DECODE_FEATURE(RISCV::FeatureVendorXTHeadSync,
-                          DecoderTableXTHeadSync32,
-                          "XTHeadSync custom opcode table");
-    TRY_TO_DECODE_FEATURE(RISCV::FeatureVendorXTHeadVdot, DecoderTableXTHeadVdot32,
-                          "XTHeadVdot custom opcode table");
-    TRY_TO_DECODE_FEATURE(RISCV::FeatureVendorXSfvcp, DecoderTableXSfvcp32,
-                          "SiFive VCIX custom opcode table");
-    TRY_TO_DECODE_FEATURE(
-        RISCV::FeatureVendorXSfvqmaccdod, DecoderTableXSfvqmaccdod32,
-        "SiFive Matrix Multiplication (2x8 and 8x2) Instruction opcode table");
-    TRY_TO_DECODE_FEATURE(
-        RISCV::FeatureVendorXSfvqmaccqoq, DecoderTableXSfvqmaccqoq32,
-        "SiFive Matrix Multiplication (4x8 and 8x4) Instruction opcode table");
-    TRY_TO_DECODE_FEATURE(
-        RISCV::FeatureVendorXSfvfwmaccqqq, DecoderTableXSfvfwmaccqqq32,
-        "SiFive Matrix Multiplication Instruction opcode table");
-    TRY_TO_DECODE_FEATURE(
-        RISCV::FeatureVendorXSfvfnrclipxfqf, DecoderTableXSfvfnrclipxfqf32,
-        "SiFive FP32-to-int8 Ranged Clip Instructions opcode table");
-    TRY_TO_DECODE_FEATURE(RISCV::FeatureVendorXSiFivecdiscarddlone,
-                          DecoderTableXSiFivecdiscarddlone32,
-                          "SiFive sf.cdiscard.d.l1 custom opcode table");
-    TRY_TO_DECODE_FEATURE(RISCV::FeatureVendorXSiFivecflushdlone,
-                          DecoderTableXSiFivecflushdlone32,
-                          "SiFive sf.cflush.d.l1 custom opcode table");
-    TRY_TO_DECODE_FEATURE(RISCV::FeatureVendorXSfcease, DecoderTableXSfcease32,
-                          "SiFive sf.cease custom opcode table");
-    TRY_TO_DECODE_FEATURE(RISCV::FeatureVendorXCVbitmanip,
-                          DecoderTableXCVbitmanip32,
-                          "CORE-V Bit Manipulation custom opcode table");
-    TRY_TO_DECODE_FEATURE(RISCV::FeatureVendorXCVelw, DecoderTableXCVelw32,
-                          "CORE-V Event load custom opcode table");
-    TRY_TO_DECODE_FEATURE(RISCV::FeatureVendorXCVmac, DecoderTableXCVmac32,
-                          "CORE-V MAC custom opcode table");
-    TRY_TO_DECODE_FEATURE(RISCV::FeatureVendorXCVmem, DecoderTableXCVmem32,
-                          "CORE-V MEM custom opcode table");
-    TRY_TO_DECODE_FEATURE(RISCV::FeatureVendorXCValu, DecoderTableXCValu32,
-                          "CORE-V ALU custom opcode table");
-    TRY_TO_DECODE_FEATURE(RISCV::FeatureVendorXCVsimd, DecoderTableXCVsimd32,
-                          "CORE-V SIMD extensions custom opcode table");
-    TRY_TO_DECODE_FEATURE(RISCV::FeatureVendorXCVbi, DecoderTableXCVbi32,
-                          "CORE-V Immediate Branching custom opcode table");
-    TRY_TO_DECODE(true, DecoderTable32, "RISCV32 table");
-
+DecodeStatus RISCVDisassembler::getInstruction32(MCInst &MI, uint64_t &Size,
+                                                 ArrayRef<uint8_t> Bytes,
+                                                 uint64_t Address,
+                                                 raw_ostream &CS) const {
+  if (Bytes.size() < 4) {
+    Size = 0;
     return MCDisassembler::Fail;
   }
+  Size = 4;
+
+  uint32_t Insn = support::endian::read32le(Bytes.data());
+
+  TRY_TO_DECODE(STI.hasFeature(RISCV::FeatureStdExtZdinx) &&
+                    !STI.hasFeature(RISCV::Feature64Bit),
+                DecoderTableRV32Zdinx32,
+                "RV32Zdinx table (Double in Integer and rv32)");
+  TRY_TO_DECODE(STI.hasFeature(RISCV::FeatureStdExtZacas) &&
+                    !STI.hasFeature(RISCV::Feature64Bit),
+                DecoderTableRV32Zacas32,
+                "RV32Zacas table (Compare-And-Swap and rv32)");
+  TRY_TO_DECODE_FEATURE(RISCV::FeatureStdExtZfinx, DecoderTableRVZfinx32,
+                        "RVZfinx table (Float in Integer)");
+  TRY_TO_DECODE_FEATURE(RISCV::FeatureVendorXVentanaCondOps,
+                        DecoderTableXVentana32, "Ventana custom opcode table");
+  TRY_TO_DECODE_FEATURE(RISCV::FeatureVendorXTHeadBa, DecoderTableXTHeadBa32,
+                        "XTHeadBa custom opcode table");
+  TRY_TO_DECODE_FEATURE(RISCV::FeatureVendorXTHeadBb, DecoderTableXTHeadBb32,
+                        "XTHeadBb custom opcode table");
+  TRY_TO_DECODE_FEATURE(RISCV::FeatureVendorXTHeadBs, DecoderTableXTHeadBs32,
+                        "XTHeadBs custom opcode table");
+  TRY_TO_DECODE_FEATURE(RISCV::FeatureVendorXTHeadCondMov,
+                        DecoderTableXTHeadCondMov32,
+                        "XTHeadCondMov custom opcode table");
+  TRY_TO_DECODE_FEATURE(RISCV::FeatureVendorXTHeadCmo, DecoderTableXTHeadCmo32,
+                        "XTHeadCmo custom opcode table");
+  TRY_TO_DECODE_FEATURE(RISCV::FeatureVendorXTHeadFMemIdx,
+                        DecoderTableXTHeadFMemIdx32,
+                        "XTHeadFMemIdx custom opcode table");
+  TRY_TO_DECODE_FEATURE(RISCV::FeatureVendorXTHeadMac, DecoderTableXTHeadMac32,
+                        "XTHeadMac custom opcode table");
+  TRY_TO_DECODE_FEATURE(RISCV::FeatureVendorXTHeadMemIdx,
+                        DecoderTableXTHeadMemIdx32,
+                        "XTHeadMemIdx custom opcode table");
+  TRY_TO_DECODE_FEATURE(RISCV::FeatureVendorXTHeadMemPair,
+                        DecoderTableXTHeadMemPair32,
+                        "XTHeadMemPair custom opcode table");
+  TRY_TO_DECODE_FEATURE(RISCV::FeatureVendorXTHeadSync,
+                        DecoderTableXTHeadSync32,
+                        "XTHeadSync custom opcode table");
+  TRY_TO_DECODE_FEATURE(RISCV::FeatureVendorXTHeadVdot,
+                        DecoderTableXTHeadVdot32,
+                        "XTHeadVdot custom opcode table");
+  TRY_TO_DECODE_FEATURE(RISCV::FeatureVendorXSfvcp, DecoderTableXSfvcp32,
+                        "SiFive VCIX custom opcode table");
+  TRY_TO_DECODE_FEATURE(
+      RISCV::FeatureVendorXSfvqmaccdod, DecoderTableXSfvqmaccdod32,
+      "SiFive Matrix Multiplication (2x8 and 8x2) Instruction opcode table");
+  TRY_TO_DECODE_FEATURE(
+      RISCV::FeatureVendorXSfvqmaccqoq, DecoderTableXSfvqmaccqoq32,
+      "SiFive Matrix Multiplication (4x8 and 8x4) Instruction opcode table");
+  TRY_TO_DECODE_FEATURE(
+      RISCV::FeatureVendorXSfvfwmaccqqq, DecoderTableXSfvfwmaccqqq32,
+      "SiFive Matrix Multiplication Instruction opcode table");
+  TRY_TO_DECODE_FEATURE(
+      RISCV::FeatureVendorXSfvfnrclipxfqf, DecoderTableXSfvfnrclipxfqf32,
+      "SiFive FP32-to-int8 Ranged Clip Instructions opcode table");
+  TRY_TO_DECODE_FEATURE(RISCV::FeatureVendorXSiFivecdiscarddlone,
+                        DecoderTableXSiFivecdiscarddlone32,
+                        "SiFive sf.cdiscard.d.l1 custom opcode table");
+  TRY_TO_DECODE_FEATURE(RISCV::FeatureVendorXSiFivecflushdlone,
+                        DecoderTableXSiFivecflushdlone32,
+                        "SiFive sf.cflush.d.l1 custom opcode table");
+  TRY_TO_DECODE_FEATURE(RISCV::FeatureVendorXSfcease, DecoderTableXSfcease32,
+                        "SiFive sf.cease custom opcode table");
+  TRY_TO_DECODE_FEATURE(RISCV::FeatureVendorXCVbitmanip,
+                        DecoderTableXCVbitmanip32,
+                        "CORE-V Bit Manipulation custom opcode table");
+  TRY_TO_DECODE_FEATURE(RISCV::FeatureVendorXCVelw, DecoderTableXCVelw32,
+                        "CORE-V Event load custom opcode table");
+  TRY_TO_DECODE_FEATURE(RISCV::FeatureVendorXCVmac, DecoderTableXCVmac32,
+                        "CORE-V MAC custom opcode table");
+  TRY_TO_DECODE_FEATURE(RISCV::FeatureVendorXCVmem, DecoderTableXCVmem32,
+                        "CORE-V MEM custom opcode table");
+  TRY_TO_DECODE_FEATURE(RISCV::FeatureVendorXCValu, DecoderTableXCValu32,
+                        "CORE-V ALU custom opcode table");
+  TRY_TO_DECODE_FEATURE(RISCV::FeatureVendorXCVsimd, DecoderTableXCVsimd32,
+                        "CORE-V SIMD extensions custom opcode table");
+  TRY_TO_DECODE_FEATURE(RISCV::FeatureVendorXCVbi, DecoderTableXCVbi32,
+                        "CORE-V Immediate Branching custom opcode table");
+  TRY_TO_DECODE(true, DecoderTable32, "RISCV32 table");
 
+  return MCDisassembler::Fail;
+}
+
+DecodeStatus RISCVDisassembler::getInstruction16(MCInst &MI, uint64_t &Size,
+                                                 ArrayRef<uint8_t> Bytes,
+                                                 uint64_t Address,
+                                                 raw_ostream &CS) const {
   if (Bytes.size() < 2) {
     Size = 0;
     return MCDisassembler::Fail;
   }
   Size = 2;
 
-  Insn = support::endian::read16le(Bytes.data());
+  uint32_t Insn = support::endian::read16le(Bytes.data());
   TRY_TO_DECODE_AND_ADD_SP(!STI.hasFeature(RISCV::Feature64Bit),
                            DecoderTableRISCV32Only_16,
                            "RISCV32Only_16 table (16-bit Instruction)");
@@ -645,3 +651,49 @@ DecodeStatus RISCVDisassembler::getInstruction(MCInst &MI, uint64_t &Size,
 
   return MCDisassembler::Fail;
 }
+
+DecodeStatus RISCVDisassembler::getInstruction(MCInst &MI, uint64_t &Size,
+                                               ArrayRef<uint8_t> Bytes,
+                                               uint64_t Address,
+                                               raw_ostream &CS) const {
+  // It's a 16 bit instruction if bit 0 and 1 are not 0b11.
+  if ((Bytes[0] & 0b11) != 0b11)
+    return getInstruction16(MI, Size, Bytes, Address, CS);
+
+  // It's a 32 bit instruction if bit 1:0 are 0b11(checked above) and bits 4:2
+  // are not 0b111.
+  if ((Bytes[0] & 0b1'1100) != 0b1'1100)
+    return getInstruction32(MI, Size, Bytes, Address, CS);
+
+  // 48-bit instructions are encoded as 0bxx011111.
+  if ((Bytes[0] & 0b11'1111) == 0b01'1111) {
+    Size = Bytes.size() >= 6 ? 6 : 0;
+    return MCDisassembler::Fail;
+  }
+
+  // 64-bit instructions are encoded as 0x0111111.
+  if ((Bytes[0] & 0b111'1111) == 0b011'1111) {
+    Size = Bytes.size() >= 8 ? 8 : 0;
+    return MCDisassembler::Fail;
+  }
+
+  // Remaining cases need to check a second byte.
+  if (Bytes.size() < 2) {
+    Size = 0;
+    return MCDisassembler::Fail;
+  }
+
+  // 80-bit through 176-bit instructions are encoded as 0bxnnnxxxx_x1111111.
+  // Where the number of bits is (80 + (nnn * 16)) for nnn != 0b111.
+  unsigned nnn = (Bytes[1] >> 4) & 0b111;
+  if (nnn != 0b111) {
+    Size = 10 + (nnn * 2);
+    if (Bytes.size() < Size)
+      Size = 0;
+    return MCDisassembler::Fail;
+  }
+
+  // Remaining encodings are reserved for > 176-bit instructions.
+  Size = 0;
+  return MCDisassembler::Fail;
+}
diff --git a/llvm/lib/Target/RISCV/MCA/RISCVCustomBehaviour.cpp b/llvm/lib/Target/RISCV/MCA/RISCVCustomBehaviour.cpp
index 8d97c5ffd20a..fb0dc482e608 100644
--- a/llvm/lib/Target/RISCV/MCA/RISCVCustomBehaviour.cpp
+++ b/llvm/lib/Target/RISCV/MCA/RISCVCustomBehaviour.cpp
@@ -20,26 +20,6 @@
 
 #define DEBUG_TYPE "llvm-mca-riscv-custombehaviour"
 
-// This brings in a table with primary key of
-// base instruction opcode and lmul and maps
-// to the opcode of the pseudo instruction.
-namespace RISCVVInversePseudosTable {
-using namespace llvm;
-using namespace llvm::RISCV;
-
-struct PseudoInfo {
-  uint16_t Pseudo;
-  uint16_t BaseInstr;
-  uint8_t VLMul;
-  uint8_t SEW;
-};
-
-#define GET_RISCVVInversePseudosTable_IMPL
-#define GET_RISCVVInversePseudosTable_DECL
-#include "RISCVGenSearchableTables.inc"
-
-} // end namespace RISCVVInversePseudosTable
-
 namespace llvm {
 namespace mca {
 
diff --git a/llvm/lib/Target/RISCV/MCTargetDesc/RISCVBaseInfo.cpp b/llvm/lib/Target/RISCV/MCTargetDesc/RISCVBaseInfo.cpp
index 67c9060b5157..4c59474df883 100644
--- a/llvm/lib/Target/RISCV/MCTargetDesc/RISCVBaseInfo.cpp
+++ b/llvm/lib/Target/RISCV/MCTargetDesc/RISCVBaseInfo.cpp
@@ -16,7 +16,6 @@
 #include "llvm/MC/MCInst.h"
 #include "llvm/MC/MCRegisterInfo.h"
 #include "llvm/MC/MCSubtargetInfo.h"
-#include "llvm/Support/RISCVISAInfo.h"
 #include "llvm/Support/raw_ostream.h"
 #include "llvm/TargetParser/TargetParser.h"
 #include "llvm/TargetParser/Triple.h"
diff --git a/llvm/lib/Target/RISCV/MCTargetDesc/RISCVBaseInfo.h b/llvm/lib/Target/RISCV/MCTargetDesc/RISCVBaseInfo.h
index aa641bc866aa..08f056f78979 100644
--- a/llvm/lib/Target/RISCV/MCTargetDesc/RISCVBaseInfo.h
+++ b/llvm/lib/Target/RISCV/MCTargetDesc/RISCVBaseInfo.h
@@ -19,7 +19,7 @@
 #include "llvm/ADT/StringRef.h"
 #include "llvm/ADT/StringSwitch.h"
 #include "llvm/MC/MCInstrDesc.h"
-#include "llvm/Support/RISCVISAInfo.h"
+#include "llvm/TargetParser/RISCVISAInfo.h"
 #include "llvm/TargetParser/RISCVTargetParser.h"
 #include "llvm/TargetParser/SubtargetFeature.h"
 
diff --git a/llvm/lib/Target/RISCV/MCTargetDesc/RISCVELFStreamer.cpp b/llvm/lib/Target/RISCV/MCTargetDesc/RISCVELFStreamer.cpp
index cdf7c048a4bf..ae7ce476fff2 100644
--- a/llvm/lib/Target/RISCV/MCTargetDesc/RISCVELFStreamer.cpp
+++ b/llvm/lib/Target/RISCV/MCTargetDesc/RISCVELFStreamer.cpp
@@ -197,11 +197,9 @@ namespace llvm {
 MCELFStreamer *createRISCVELFStreamer(MCContext &C,
                                       std::unique_ptr<MCAsmBackend> MAB,
                                       std::unique_ptr<MCObjectWriter> MOW,
-                                      std::unique_ptr<MCCodeEmitter> MCE,
-                                      bool RelaxAll) {
+                                      std::unique_ptr<MCCodeEmitter> MCE) {
   RISCVELFStreamer *S =
       new RISCVELFStreamer(C, std::move(MAB), std::move(MOW), std::move(MCE));
-  S->getAssembler().setRelaxAll(RelaxAll);
   return S;
 }
 } // namespace llvm
diff --git a/llvm/lib/Target/RISCV/MCTargetDesc/RISCVELFStreamer.h b/llvm/lib/Target/RISCV/MCTargetDesc/RISCVELFStreamer.h
index e8f29cd8449b..212d731889f1 100644
--- a/llvm/lib/Target/RISCV/MCTargetDesc/RISCVELFStreamer.h
+++ b/llvm/lib/Target/RISCV/MCTargetDesc/RISCVELFStreamer.h
@@ -75,7 +75,6 @@ public:
 MCELFStreamer *createRISCVELFStreamer(MCContext &C,
                                       std::unique_ptr<MCAsmBackend> MAB,
                                       std::unique_ptr<MCObjectWriter> MOW,
-                                      std::unique_ptr<MCCodeEmitter> MCE,
-                                      bool RelaxAll);
+                                      std::unique_ptr<MCCodeEmitter> MCE);
 }
 #endif
diff --git a/llvm/lib/Target/RISCV/MCTargetDesc/RISCVMCTargetDesc.cpp b/llvm/lib/Target/RISCV/MCTargetDesc/RISCVMCTargetDesc.cpp
index 79e56a7a6d03..691a5892ae82 100644
--- a/llvm/lib/Target/RISCV/MCTargetDesc/RISCVMCTargetDesc.cpp
+++ b/llvm/lib/Target/RISCV/MCTargetDesc/RISCVMCTargetDesc.cpp
@@ -43,6 +43,15 @@
 #define GET_SUBTARGETINFO_MC_DESC
 #include "RISCVGenSubtargetInfo.inc"
 
+namespace llvm::RISCVVInversePseudosTable {
+
+using namespace RISCV;
+
+#define GET_RISCVVInversePseudosTable_IMPL
+#include "RISCVGenSearchableTables.inc"
+
+} // namespace llvm::RISCVVInversePseudosTable
+
 using namespace llvm;
 
 static MCInstrInfo *createRISCVMCInstrInfo() {
@@ -323,10 +332,9 @@ namespace {
 MCStreamer *createRISCVELFStreamer(const Triple &T, MCContext &Context,
                                    std::unique_ptr<MCAsmBackend> &&MAB,
                                    std::unique_ptr<MCObjectWriter> &&MOW,
-                                   std::unique_ptr<MCCodeEmitter> &&MCE,
-                                   bool RelaxAll) {
+                                   std::unique_ptr<MCCodeEmitter> &&MCE) {
   return createRISCVELFStreamer(Context, std::move(MAB), std::move(MOW),
-                                std::move(MCE), RelaxAll);
+                                std::move(MCE));
 }
 } // end anonymous namespace
 
diff --git a/llvm/lib/Target/RISCV/MCTargetDesc/RISCVMCTargetDesc.h b/llvm/lib/Target/RISCV/MCTargetDesc/RISCVMCTargetDesc.h
index 3cfddb530cdf..d4aa0fe99078 100644
--- a/llvm/lib/Target/RISCV/MCTargetDesc/RISCVMCTargetDesc.h
+++ b/llvm/lib/Target/RISCV/MCTargetDesc/RISCVMCTargetDesc.h
@@ -37,7 +37,21 @@ MCAsmBackend *createRISCVAsmBackend(const Target &T, const MCSubtargetInfo &STI,
 
 std::unique_ptr<MCObjectTargetWriter> createRISCVELFObjectWriter(uint8_t OSABI,
                                                                  bool Is64Bit);
-}
+
+namespace RISCVVInversePseudosTable {
+
+struct PseudoInfo {
+  uint16_t Pseudo;
+  uint16_t BaseInstr;
+  uint8_t VLMul;
+  uint8_t SEW;
+};
+
+#define GET_RISCVVInversePseudosTable_DECL
+#include "RISCVGenSearchableTables.inc"
+
+} // namespace RISCVVInversePseudosTable
+} // namespace llvm
 
 // Defines symbolic names for RISC-V registers.
 #define GET_REGINFO_ENUM
diff --git a/llvm/lib/Target/RISCV/MCTargetDesc/RISCVTargetStreamer.cpp b/llvm/lib/Target/RISCV/MCTargetDesc/RISCVTargetStreamer.cpp
index 4a4b1e13c2b9..0f92e9ed6a64 100644
--- a/llvm/lib/Target/RISCV/MCTargetDesc/RISCVTargetStreamer.cpp
+++ b/llvm/lib/Target/RISCV/MCTargetDesc/RISCVTargetStreamer.cpp
@@ -16,7 +16,7 @@
 #include "llvm/MC/MCSymbol.h"
 #include "llvm/Support/FormattedStream.h"
 #include "llvm/Support/RISCVAttributes.h"
-#include "llvm/Support/RISCVISAInfo.h"
+#include "llvm/TargetParser/RISCVISAInfo.h"
 
 using namespace llvm;
 
diff --git a/llvm/lib/Target/RISCV/RISCV.h b/llvm/lib/Target/RISCV/RISCV.h
index 7af543f018cc..d405395dcf9e 100644
--- a/llvm/lib/Target/RISCV/RISCV.h
+++ b/llvm/lib/Target/RISCV/RISCV.h
@@ -61,6 +61,9 @@ void initializeRISCVExpandAtomicPseudoPass(PassRegistry &);
 FunctionPass *createRISCVInsertVSETVLIPass();
 void initializeRISCVInsertVSETVLIPass(PassRegistry &);
 
+FunctionPass *createRISCVCoalesceVSETVLIPass();
+void initializeRISCVCoalesceVSETVLIPass(PassRegistry &);
+
 FunctionPass *createRISCVPostRAExpandPseudoPass();
 void initializeRISCVPostRAExpandPseudoPass(PassRegistry &);
 FunctionPass *createRISCVInsertReadWriteCSRPass();
diff --git a/llvm/lib/Target/RISCV/RISCV.td b/llvm/lib/Target/RISCV/RISCV.td
index 9fb84efd5b6f..09f496574d64 100644
--- a/llvm/lib/Target/RISCV/RISCV.td
+++ b/llvm/lib/Target/RISCV/RISCV.td
@@ -15,6 +15,12 @@ include "llvm/Target/Target.td"
 include "RISCVFeatures.td"
 
 //===----------------------------------------------------------------------===//
+// RISC-V profiles supported.
+//===----------------------------------------------------------------------===//
+
+include "RISCVProfiles.td"
+
+//===----------------------------------------------------------------------===//
 // Named operands for CSR instructions.
 //===----------------------------------------------------------------------===//
 
diff --git a/llvm/lib/Target/RISCV/RISCVAsmPrinter.cpp b/llvm/lib/Target/RISCV/RISCVAsmPrinter.cpp
index 779f179dff61..6eceaddc747d 100644
--- a/llvm/lib/Target/RISCV/RISCVAsmPrinter.cpp
+++ b/llvm/lib/Target/RISCV/RISCVAsmPrinter.cpp
@@ -37,8 +37,8 @@
 #include "llvm/MC/MCStreamer.h"
 #include "llvm/MC/MCSymbol.h"
 #include "llvm/MC/TargetRegistry.h"
-#include "llvm/Support/RISCVISAInfo.h"
 #include "llvm/Support/raw_ostream.h"
+#include "llvm/TargetParser/RISCVISAInfo.h"
 #include "llvm/Transforms/Instrumentation/HWAddressSanitizer.h"
 
 using namespace llvm;
diff --git a/llvm/lib/Target/RISCV/RISCVFeatures.td b/llvm/lib/Target/RISCV/RISCVFeatures.td
index b064191b838e..c3dc4ea53697 100644
--- a/llvm/lib/Target/RISCV/RISCVFeatures.td
+++ b/llvm/lib/Target/RISCV/RISCVFeatures.td
@@ -391,7 +391,7 @@ def HasStdExtZcb : Predicate<"Subtarget->hasStdExtZcb()">,
 def FeatureStdExtZcd
     : RISCVExtension<"zcd", 1, 0,
                      "'Zcd' (Compressed Double-Precision Floating-Point Instructions)",
-                     [FeatureStdExtZca]>;
+                     [FeatureStdExtD, FeatureStdExtZca]>;
 
 def HasStdExtCOrZcd
     : Predicate<"Subtarget->hasStdExtCOrZcd()">,
@@ -402,7 +402,7 @@ def HasStdExtCOrZcd
 def FeatureStdExtZcf
     : RISCVExtension<"zcf", 1, 0,
                      "'Zcf' (Compressed Single-Precision Floating-Point Instructions)",
-                     [FeatureStdExtZca]>;
+                     [FeatureStdExtF, FeatureStdExtZca]>;
 
 def FeatureStdExtZcmp
     : RISCVExtension<"zcmp", 1, 0,
@@ -423,8 +423,7 @@ def HasStdExtZcmt : Predicate<"Subtarget->hasStdExtZcmt()">,
 def FeatureStdExtZce
     : RISCVExtension<"zce", 1, 0,
                      "'Zce' (Compressed extensions for microcontrollers)",
-                     [FeatureStdExtZca, FeatureStdExtZcb, FeatureStdExtZcmp,
-                      FeatureStdExtZcmt]>;
+                     [FeatureStdExtZcb, FeatureStdExtZcmp, FeatureStdExtZcmt]>;
 
 def HasStdExtCOrZcfOrZce
     : Predicate<"Subtarget->hasStdExtC() || Subtarget->hasStdExtZcf() "
@@ -1016,8 +1015,7 @@ def HasVendorXTHeadCmo : Predicate<"Subtarget->hasVendorXTHeadCmo()">,
 
 def FeatureVendorXTHeadFMemIdx
     : RISCVExtension<"xtheadfmemidx", 1, 0,
-                     "'xtheadfmemidx' (T-Head FP Indexed Memory Operations)",
-                     [FeatureStdExtF]>;
+                     "'xtheadfmemidx' (T-Head FP Indexed Memory Operations)">;
 def HasVendorXTHeadFMemIdx : Predicate<"Subtarget->hasVendorXTHeadFMemIdx()">,
                              AssemblerPredicate<(all_of FeatureVendorXTHeadFMemIdx),
                                  "'xtheadfmemidx' (T-Head FP Indexed Memory Operations)">;
@@ -1089,7 +1087,7 @@ def HasVendorXSfvqmaccqoq
 def FeatureVendorXSfvfwmaccqqq
     : RISCVExtension<"xsfvfwmaccqqq", 1, 0,
                      "'XSfvfwmaccqqq' (SiFive Matrix Multiply Accumulate Instruction and 4-by-4))",
-                     [FeatureStdExtZve32f, FeatureStdExtZvfbfmin]>;
+                     [FeatureStdExtZvfbfmin]>;
 def HasVendorXSfvfwmaccqqq
     : Predicate<"Subtarget->hasVendorXSfvfwmaccqqq()">,
       AssemblerPredicate<(all_of FeatureVendorXSfvfwmaccqqq),
diff --git a/llvm/lib/Target/RISCV/RISCVISelDAGToDAG.cpp b/llvm/lib/Target/RISCV/RISCVISelDAGToDAG.cpp
index f99dc0b85763..b0568297a470 100644
--- a/llvm/lib/Target/RISCV/RISCVISelDAGToDAG.cpp
+++ b/llvm/lib/Target/RISCV/RISCVISelDAGToDAG.cpp
@@ -43,7 +43,6 @@ namespace llvm::RISCV {
 #define GET_RISCVVSETable_IMPL
 #define GET_RISCVVLXTable_IMPL
 #define GET_RISCVVSXTable_IMPL
-#define GET_RISCVMaskedPseudosTable_IMPL
 #include "RISCVGenSearchableTables.inc"
 } // namespace llvm::RISCV
 
diff --git a/llvm/lib/Target/RISCV/RISCVISelDAGToDAG.h b/llvm/lib/Target/RISCV/RISCVISelDAGToDAG.h
index 92f818b0dc48..7d4aec2dfdc9 100644
--- a/llvm/lib/Target/RISCV/RISCVISelDAGToDAG.h
+++ b/llvm/lib/Target/RISCV/RISCVISelDAGToDAG.h
@@ -261,13 +261,6 @@ struct VLX_VSXPseudo {
   uint16_t Pseudo;
 };
 
-struct RISCVMaskedPseudoInfo {
-  uint16_t MaskedPseudo;
-  uint16_t UnmaskedPseudo;
-  uint8_t MaskOpIdx;
-  uint8_t MaskAffectsResult : 1;
-};
-
 #define GET_RISCVVSSEGTable_DECL
 #define GET_RISCVVLSEGTable_DECL
 #define GET_RISCVVLXSEGTable_DECL
@@ -276,8 +269,6 @@ struct RISCVMaskedPseudoInfo {
 #define GET_RISCVVSETable_DECL
 #define GET_RISCVVLXTable_DECL
 #define GET_RISCVVSXTable_DECL
-#define GET_RISCVMaskedPseudosTable_DECL
-#include "RISCVGenSearchableTables.inc"
 } // namespace RISCV
 
 } // namespace llvm
diff --git a/llvm/lib/Target/RISCV/RISCVISelLowering.cpp b/llvm/lib/Target/RISCV/RISCVISelLowering.cpp
index 41483c49ae03..662ff86df600 100644
--- a/llvm/lib/Target/RISCV/RISCVISelLowering.cpp
+++ b/llvm/lib/Target/RISCV/RISCVISelLowering.cpp
@@ -524,8 +524,8 @@ RISCVTargetLowering::RISCVTargetLowering(const TargetMachine &TM,
     if (Subtarget.is64Bit())
       setOperationAction(ISD::FPOWI, MVT::i32, Custom);
 
-    if (!Subtarget.hasStdExtZfa())
-      setOperationAction({ISD::FMAXIMUM, ISD::FMINIMUM}, MVT::f16, Custom);
+    setOperationAction({ISD::FMAXIMUM, ISD::FMINIMUM}, MVT::f16,
+                       Subtarget.hasStdExtZfa() ? Legal : Custom);
   }
 
   if (Subtarget.hasStdExtFOrZfinx()) {
@@ -548,10 +548,12 @@ RISCVTargetLowering::RISCVTargetLowering(const TargetMachine &TM,
     setOperationAction(ISD::FP_TO_FP16, MVT::f32, Custom);
     setOperationAction(ISD::FP16_TO_FP, MVT::f32, Custom);
 
-    if (Subtarget.hasStdExtZfa())
+    if (Subtarget.hasStdExtZfa()) {
       setOperationAction(ISD::FNEARBYINT, MVT::f32, Legal);
-    else
+      setOperationAction({ISD::FMAXIMUM, ISD::FMINIMUM}, MVT::f32, Legal);
+    } else {
       setOperationAction({ISD::FMAXIMUM, ISD::FMINIMUM}, MVT::f32, Custom);
+    }
   }
 
   if (Subtarget.hasStdExtFOrZfinx() && Subtarget.is64Bit())
@@ -566,6 +568,7 @@ RISCVTargetLowering::RISCVTargetLowering(const TargetMachine &TM,
     if (Subtarget.hasStdExtZfa()) {
       setOperationAction(FPRndMode, MVT::f64, Legal);
       setOperationAction(ISD::FNEARBYINT, MVT::f64, Legal);
+      setOperationAction({ISD::FMAXIMUM, ISD::FMINIMUM}, MVT::f64, Legal);
     } else {
       if (Subtarget.is64Bit())
         setOperationAction(FPRndMode, MVT::f64, Custom);
@@ -13416,12 +13419,33 @@ static SDValue expandMul(SDNode *N, SelectionDAG &DAG,
     return SDValue();
   uint64_t MulAmt = CNode->getZExtValue();
 
-  // 3/5/9 * 2^N -> shXadd (sll X, C), (sll X, C)
-  // Matched in tablegen, avoid perturbing patterns.
-  for (uint64_t Divisor : {3, 5, 9})
-    if (MulAmt % Divisor == 0 && isPowerOf2_64(MulAmt / Divisor))
+  // WARNING: The code below is knowingly incorrect with regards to undef semantics.
+  // We're adding additional uses of X here, and in principle, we should be freezing
+  // X before doing so.  However, adding freeze here causes real regressions, and no
+  // other target properly freezes X in these cases either.
+  SDValue X = N->getOperand(0);
+
+  for (uint64_t Divisor : {3, 5, 9}) {
+    if (MulAmt % Divisor != 0)
+      continue;
+    uint64_t MulAmt2 = MulAmt / Divisor;
+    // 3/5/9 * 2^N -> shXadd (sll X, C), (sll X, C)
+    // Matched in tablegen, avoid perturbing patterns.
+    if (isPowerOf2_64(MulAmt2))
       return SDValue();
 
+    // 3/5/9 * 3/5/9 -> shXadd (shYadd X, X), (shYadd X, X)
+    if (MulAmt2 == 3 || MulAmt2 == 5 || MulAmt2 == 9) {
+      SDLoc DL(N);
+      SDValue Mul359 =
+          DAG.getNode(RISCVISD::SHL_ADD, DL, VT, X,
+                      DAG.getConstant(Log2_64(Divisor - 1), DL, VT), X);
+      return DAG.getNode(RISCVISD::SHL_ADD, DL, VT, Mul359,
+                         DAG.getConstant(Log2_64(MulAmt2 - 1), DL, VT),
+                         Mul359);
+    }
+  }
+
   // If this is a power 2 + 2/4/8, we can use a shift followed by a single
   // shXadd. First check if this a sum of two power of 2s because that's
   // easy. Then count how many zeros are up to the first bit.
@@ -13430,33 +13454,31 @@ static SDValue expandMul(SDNode *N, SelectionDAG &DAG,
     if (ScaleShift >= 1 && ScaleShift < 4) {
       unsigned ShiftAmt = Log2_64((MulAmt & (MulAmt - 1)));
       SDLoc DL(N);
-      SDValue X = DAG.getFreeze(N->getOperand(0));
       SDValue Shift1 =
           DAG.getNode(ISD::SHL, DL, VT, X, DAG.getConstant(ShiftAmt, DL, VT));
-      SDValue Shift2 =
-          DAG.getNode(ISD::SHL, DL, VT, X, DAG.getConstant(ScaleShift, DL, VT));
-      return DAG.getNode(ISD::ADD, DL, VT, Shift1, Shift2);
+      return DAG.getNode(RISCVISD::SHL_ADD, DL, VT, X,
+                         DAG.getConstant(ScaleShift, DL, VT), Shift1);
     }
   }
 
   // 2^(1,2,3) * 3,5,9 + 1 -> (shXadd (shYadd x, x), x)
-  // Matched in tablegen, avoid perturbing patterns.
-  switch (MulAmt) {
-  case 11:
-  case 13:
-  case 19:
-  case 21:
-  case 25:
-  case 27:
-  case 29:
-  case 37:
-  case 41:
-  case 45:
-  case 73:
-  case 91:
-    return SDValue();
-  default:
-    break;
+  // This is the two instruction form, there are also three instruction
+  // variants we could implement.  e.g.
+  //   (2^(1,2,3) * 3,5,9 + 1) << C2
+  //   2^(C1>3) * 3,5,9 +/- 1
+  for (uint64_t Divisor : {3, 5, 9}) {
+    uint64_t C = MulAmt - 1;
+    if (C <= Divisor)
+      continue;
+    unsigned TZ = llvm::countr_zero(C);
+    if ((C >> TZ) == Divisor && (TZ == 1 || TZ == 2 || TZ == 3)) {
+      SDLoc DL(N);
+      SDValue Mul359 =
+          DAG.getNode(RISCVISD::SHL_ADD, DL, VT, X,
+                      DAG.getConstant(Log2_64(Divisor - 1), DL, VT), X);
+      return DAG.getNode(RISCVISD::SHL_ADD, DL, VT, Mul359,
+                         DAG.getConstant(TZ, DL, VT), X);
+    }
   }
 
   // 2^n + 2/4/8 + 1 -> (add (shl X, C1), (shXadd X, X))
@@ -13465,13 +13487,25 @@ static SDValue expandMul(SDNode *N, SelectionDAG &DAG,
     if (ScaleShift >= 1 && ScaleShift < 4) {
       unsigned ShiftAmt = Log2_64(((MulAmt - 1) & (MulAmt - 2)));
       SDLoc DL(N);
-      SDValue X = DAG.getFreeze(N->getOperand(0));
       SDValue Shift1 =
           DAG.getNode(ISD::SHL, DL, VT, X, DAG.getConstant(ShiftAmt, DL, VT));
-      SDValue Shift2 =
-          DAG.getNode(ISD::SHL, DL, VT, X, DAG.getConstant(ScaleShift, DL, VT));
       return DAG.getNode(ISD::ADD, DL, VT, Shift1,
-                         DAG.getNode(ISD::ADD, DL, VT, Shift2, X));
+                         DAG.getNode(RISCVISD::SHL_ADD, DL, VT, X,
+                                     DAG.getConstant(ScaleShift, DL, VT), X));
+    }
+  }
+
+  // 2^N - 3/5/9 --> (sub (shl X, C1), (shXadd X, x))
+  for (uint64_t Offset : {3, 5, 9}) {
+    if (isPowerOf2_64(MulAmt + Offset)) {
+      SDLoc DL(N);
+      SDValue Shift1 =
+          DAG.getNode(ISD::SHL, DL, VT, X,
+                      DAG.getConstant(Log2_64(MulAmt + Offset), DL, VT));
+      SDValue Mul359 = DAG.getNode(RISCVISD::SHL_ADD, DL, VT, X,
+                                   DAG.getConstant(Log2_64(Offset - 1), DL, VT),
+                                   X);
+      return DAG.getNode(ISD::SUB, DL, VT, Shift1, Mul359);
     }
   }
 
@@ -16133,23 +16167,31 @@ SDValue RISCVTargetLowering::PerformDAGCombine(SDNode *N,
   case ISD::SELECT:
     return performSELECTCombine(N, DAG, Subtarget);
   case RISCVISD::CZERO_EQZ:
-  case RISCVISD::CZERO_NEZ:
+  case RISCVISD::CZERO_NEZ: {
+    SDValue LHS = N->getOperand(0);
+    SDValue RHS = N->getOperand(1);
     // czero_eq X, (xor Y, 1) -> czero_ne X, Y if Y is 0 or 1.
     // czero_ne X, (xor Y, 1) -> czero_eq X, Y if Y is 0 or 1.
-    if (N->getOperand(1).getOpcode() == ISD::XOR &&
-        isOneConstant(N->getOperand(1).getOperand(1))) {
-      SDValue Cond = N->getOperand(1).getOperand(0);
+    if (RHS.getOpcode() == ISD::XOR && isOneConstant(RHS.getOperand(1))) {
+      SDValue Cond = RHS.getOperand(0);
       APInt Mask = APInt::getBitsSetFrom(Cond.getValueSizeInBits(), 1);
       if (DAG.MaskedValueIsZero(Cond, Mask)) {
         unsigned NewOpc = N->getOpcode() == RISCVISD::CZERO_EQZ
                               ? RISCVISD::CZERO_NEZ
                               : RISCVISD::CZERO_EQZ;
-        return DAG.getNode(NewOpc, SDLoc(N), N->getValueType(0),
-                           N->getOperand(0), Cond);
+        return DAG.getNode(NewOpc, SDLoc(N), N->getValueType(0), LHS, Cond);
       }
     }
+    // czero_eqz x, (setcc x, 0, ne) -> x
+    // czero_nez x, (setcc x, 0, eq) -> x
+    if (RHS.getOpcode() == ISD::SETCC && isNullConstant(RHS.getOperand(1)) &&
+        cast<CondCodeSDNode>(RHS.getOperand(2))->get() ==
+            (N->getOpcode() == RISCVISD::CZERO_EQZ ? ISD::CondCode::SETNE
+                                                   : ISD::CondCode::SETEQ) &&
+        LHS == RHS.getOperand(0))
+      return LHS;
     return SDValue();
-
+  }
   case RISCVISD::SELECT_CC: {
     // Transform
     SDValue LHS = N->getOperand(0);
@@ -16817,7 +16859,7 @@ SDValue RISCVTargetLowering::PerformDAGCombine(SDNode *N,
           StrideC && StrideC->getZExtValue() == ElementSize)
         return DAG.getMaskedStore(Store->getChain(), DL, Value, Base,
                                   DAG.getUNDEF(XLenVT), Mask,
-                                  Store->getMemoryVT(), Store->getMemOperand(),
+                                  Value.getValueType(), Store->getMemOperand(),
                                   ISD::UNINDEXED, false);
       return SDValue();
     }
@@ -17758,6 +17800,18 @@ static MachineBasicBlock *emitSelectPseudo(MachineInstr &MI,
   return TailMBB;
 }
 
+// Helper to find Masked Pseudo instruction from MC instruction, LMUL and SEW.
+static const RISCV::RISCVMaskedPseudoInfo *
+lookupMaskedIntrinsic(uint16_t MCOpcode, RISCVII::VLMUL LMul, unsigned SEW) {
+  const RISCVVInversePseudosTable::PseudoInfo *Inverse =
+      RISCVVInversePseudosTable::getBaseInfo(MCOpcode, LMul, SEW);
+  assert(Inverse && "Unexpected LMUL and SEW pair for instruction");
+  const RISCV::RISCVMaskedPseudoInfo *Masked =
+      RISCV::lookupMaskedIntrinsicByUnmasked(Inverse->Pseudo);
+  assert(Masked && "Could not find masked instruction for LMUL and SEW pair");
+  return Masked;
+}
+
 static MachineBasicBlock *emitVFROUND_NOEXCEPT_MASK(MachineInstr &MI,
                                                     MachineBasicBlock *BB,
                                                     unsigned CVTXOpc) {
@@ -17795,80 +17849,9 @@ static MachineBasicBlock *emitVFROUND_NOEXCEPT_MASK(MachineInstr &MI,
   unsigned Log2SEW = MI.getOperand(RISCVII::getSEWOpNum(MI.getDesc())).getImm();
   // There is no E8 variant for VFCVT_F_X.
   assert(Log2SEW >= 4);
-  // Since MI (VFROUND) isn't SEW specific, we cannot use a macro to make
-  // handling of different (LMUL, SEW) pairs easier because we need to pull the
-  // SEW immediate from MI, and that information is not avaliable during macro
-  // expansion.
-  unsigned CVTFOpc;
-  if (Log2SEW == 4) {
-    switch (LMul) {
-    case RISCVII::LMUL_1:
-      CVTFOpc = RISCV::PseudoVFCVT_F_X_V_M1_E16_MASK;
-      break;
-    case RISCVII::LMUL_2:
-      CVTFOpc = RISCV::PseudoVFCVT_F_X_V_M2_E16_MASK;
-      break;
-    case RISCVII::LMUL_4:
-      CVTFOpc = RISCV::PseudoVFCVT_F_X_V_M4_E16_MASK;
-      break;
-    case RISCVII::LMUL_8:
-      CVTFOpc = RISCV::PseudoVFCVT_F_X_V_M8_E16_MASK;
-      break;
-    case RISCVII::LMUL_F2:
-      CVTFOpc = RISCV::PseudoVFCVT_F_X_V_MF2_E16_MASK;
-      break;
-    case RISCVII::LMUL_F4:
-      CVTFOpc = RISCV::PseudoVFCVT_F_X_V_MF4_E16_MASK;
-      break;
-    case RISCVII::LMUL_F8:
-    case RISCVII::LMUL_RESERVED:
-      llvm_unreachable("Unexpected LMUL and SEW combination value for MI.");
-    }
-  } else if (Log2SEW == 5) {
-    switch (LMul) {
-    case RISCVII::LMUL_1:
-      CVTFOpc = RISCV::PseudoVFCVT_F_X_V_M1_E32_MASK;
-      break;
-    case RISCVII::LMUL_2:
-      CVTFOpc = RISCV::PseudoVFCVT_F_X_V_M2_E32_MASK;
-      break;
-    case RISCVII::LMUL_4:
-      CVTFOpc = RISCV::PseudoVFCVT_F_X_V_M4_E32_MASK;
-      break;
-    case RISCVII::LMUL_8:
-      CVTFOpc = RISCV::PseudoVFCVT_F_X_V_M8_E32_MASK;
-      break;
-    case RISCVII::LMUL_F2:
-      CVTFOpc = RISCV::PseudoVFCVT_F_X_V_MF2_E32_MASK;
-      break;
-    case RISCVII::LMUL_F4:
-    case RISCVII::LMUL_F8:
-    case RISCVII::LMUL_RESERVED:
-      llvm_unreachable("Unexpected LMUL and SEW combination value for MI.");
-    }
-  } else if (Log2SEW == 6) {
-    switch (LMul) {
-    case RISCVII::LMUL_1:
-      CVTFOpc = RISCV::PseudoVFCVT_F_X_V_M1_E64_MASK;
-      break;
-    case RISCVII::LMUL_2:
-      CVTFOpc = RISCV::PseudoVFCVT_F_X_V_M2_E64_MASK;
-      break;
-    case RISCVII::LMUL_4:
-      CVTFOpc = RISCV::PseudoVFCVT_F_X_V_M4_E64_MASK;
-      break;
-    case RISCVII::LMUL_8:
-      CVTFOpc = RISCV::PseudoVFCVT_F_X_V_M8_E64_MASK;
-      break;
-    case RISCVII::LMUL_F2:
-    case RISCVII::LMUL_F4:
-    case RISCVII::LMUL_F8:
-    case RISCVII::LMUL_RESERVED:
-      llvm_unreachable("Unexpected LMUL and SEW combination value for MI.");
-    }
-  } else {
-    llvm_unreachable("Unexpected LMUL and SEW combination value for MI.");
-  }
+  unsigned CVTFOpc =
+      lookupMaskedIntrinsic(RISCV::VFCVT_F_X_V, LMul, 1 << Log2SEW)
+          ->MaskedPseudo;
 
   BuildMI(*BB, MI, DL, TII.get(CVTFOpc))
       .add(MI.getOperand(0))
@@ -18132,7 +18115,7 @@ void RISCVTargetLowering::AdjustInstrPostInstrSelection(MachineInstr &MI,
   if (MI.getOperand(Idx).getImm() != RISCVFPRndMode::DYN)
     return;
   // If the instruction already reads FRM, don't add another read.
-  if (MI.readsRegister(RISCV::FRM))
+  if (MI.readsRegister(RISCV::FRM, /*TRI=*/nullptr))
     return;
   MI.addOperand(
       MachineOperand::CreateReg(RISCV::FRM, /*isDef*/ false, /*isImp*/ true));
@@ -19669,6 +19652,7 @@ const char *RISCVTargetLowering::getTargetNodeName(unsigned Opcode) const {
   NODE_NAME_CASE(LLA)
   NODE_NAME_CASE(ADD_TPREL)
   NODE_NAME_CASE(MULHSU)
+  NODE_NAME_CASE(SHL_ADD)
   NODE_NAME_CASE(SLLW)
   NODE_NAME_CASE(SRAW)
   NODE_NAME_CASE(SRLW)
@@ -21031,7 +21015,7 @@ bool RISCVTargetLowering::lowerDeinterleaveIntrinsicToLoad(IntrinsicInst *DI,
   IRBuilder<> Builder(LI);
 
   // Only deinterleave2 supported at present.
-  if (DI->getIntrinsicID() != Intrinsic::experimental_vector_deinterleave2)
+  if (DI->getIntrinsicID() != Intrinsic::vector_deinterleave2)
     return false;
 
   unsigned Factor = 2;
@@ -21081,7 +21065,7 @@ bool RISCVTargetLowering::lowerInterleaveIntrinsicToStore(IntrinsicInst *II,
   IRBuilder<> Builder(SI);
 
   // Only interleave2 supported at present.
-  if (II->getIntrinsicID() != Intrinsic::experimental_vector_interleave2)
+  if (II->getIntrinsicID() != Intrinsic::vector_interleave2)
     return false;
 
   unsigned Factor = 2;
diff --git a/llvm/lib/Target/RISCV/RISCVISelLowering.h b/llvm/lib/Target/RISCV/RISCVISelLowering.h
index b10da3d40bef..ed14fd453943 100644
--- a/llvm/lib/Target/RISCV/RISCVISelLowering.h
+++ b/llvm/lib/Target/RISCV/RISCVISelLowering.h
@@ -59,6 +59,12 @@ enum NodeType : unsigned {
 
   // Multiply high for signedxunsigned.
   MULHSU,
+
+  // Represents (ADD (SHL a, b), c) with the arguments appearing in the order
+  // a, b, c.  'b' must be a constant.  Maps to sh1add/sh2add/sh3add with zba
+  // or addsl with XTheadBa.
+  SHL_ADD,
+
   // RV64I shifts, directly matching the semantics of the named RISC-V
   // instructions.
   SLLW,
diff --git a/llvm/lib/Target/RISCV/RISCVInsertReadWriteCSR.cpp b/llvm/lib/Target/RISCV/RISCVInsertReadWriteCSR.cpp
index aac0ecc1cfc9..7b9e9fb988bc 100644
--- a/llvm/lib/Target/RISCV/RISCVInsertReadWriteCSR.cpp
+++ b/llvm/lib/Target/RISCV/RISCVInsertReadWriteCSR.cpp
@@ -82,7 +82,8 @@ bool RISCVInsertReadWriteCSR::emitWriteRoundingModeOpt(MachineBasicBlock &MBB) {
       continue;
     }
 
-    if (MI.isCall() || MI.isInlineAsm() || MI.readsRegister(RISCV::FRM)) {
+    if (MI.isCall() || MI.isInlineAsm() ||
+        MI.readsRegister(RISCV::FRM, /*TRI=*/nullptr)) {
       // Restore FRM before unknown operations.
       if (SavedFRM.isValid())
         BuildMI(MBB, MI, MI.getDebugLoc(), TII->get(RISCV::WriteFRM))
@@ -92,7 +93,7 @@ bool RISCVInsertReadWriteCSR::emitWriteRoundingModeOpt(MachineBasicBlock &MBB) {
       continue;
     }
 
-    assert(!MI.modifiesRegister(RISCV::FRM) &&
+    assert(!MI.modifiesRegister(RISCV::FRM, /*TRI=*/nullptr) &&
            "Expected that MI could not modify FRM.");
 
     int FRMIdx = RISCVII::getFRMOpNum(MI.getDesc());
diff --git a/llvm/lib/Target/RISCV/RISCVInsertVSETVLI.cpp b/llvm/lib/Target/RISCV/RISCVInsertVSETVLI.cpp
index 331253e39c0a..b27e1dd258eb 100644
--- a/llvm/lib/Target/RISCV/RISCVInsertVSETVLI.cpp
+++ b/llvm/lib/Target/RISCV/RISCVInsertVSETVLI.cpp
@@ -27,16 +27,19 @@
 #include "RISCV.h"
 #include "RISCVSubtarget.h"
 #include "llvm/ADT/Statistic.h"
+#include "llvm/CodeGen/LiveDebugVariables.h"
 #include "llvm/CodeGen/LiveIntervals.h"
+#include "llvm/CodeGen/LiveStacks.h"
 #include "llvm/CodeGen/MachineFunctionPass.h"
 #include <queue>
 using namespace llvm;
 
 #define DEBUG_TYPE "riscv-insert-vsetvli"
 #define RISCV_INSERT_VSETVLI_NAME "RISC-V Insert VSETVLI pass"
+#define RISCV_COALESCE_VSETVLI_NAME "RISC-V Coalesce VSETVLI pass"
 
 STATISTIC(NumInsertedVSETVL, "Number of VSETVL inst inserted");
-STATISTIC(NumRemovedVSETVL, "Number of VSETVL inst removed");
+STATISTIC(NumCoalescedVSETVL, "Number of VSETVL inst coalesced");
 
 static cl::opt<bool> DisableInsertVSETVLPHIOpt(
     "riscv-disable-insert-vsetvl-phi-opt", cl::init(false), cl::Hidden,
@@ -153,7 +156,7 @@ static std::optional<unsigned> getEEWForLoadStore(const MachineInstr &MI) {
   }
 }
 
-static bool isNonZeroLoadImmediate(MachineInstr &MI) {
+static bool isNonZeroLoadImmediate(const MachineInstr &MI) {
   return MI.getOpcode() == RISCV::ADDI &&
     MI.getOperand(1).isReg() && MI.getOperand(2).isImm() &&
     MI.getOperand(1).getReg() == RISCV::X0 &&
@@ -190,6 +193,11 @@ static bool hasUndefinedMergeOp(const MachineInstr &MI,
   if (UseMO.getReg() == RISCV::NoRegister)
     return true;
 
+  if (UseMO.isUndef())
+    return true;
+  if (UseMO.getReg().isPhysical())
+    return false;
+
   if (MachineInstr *UseMI = MRI.getVRegDef(UseMO.getReg())) {
     if (UseMI->isImplicitDef())
       return true;
@@ -254,6 +262,17 @@ struct DemandedFields {
     VLZeroness = true;
   }
 
+  // Make this the result of demanding both the fields in this and B.
+  void doUnion(const DemandedFields &B) {
+    VLAny |= B.VLAny;
+    VLZeroness |= B.VLZeroness;
+    SEW = std::max(SEW, B.SEW);
+    LMUL |= B.LMUL;
+    SEWLMULRatio |= B.SEWLMULRatio;
+    TailPolicy |= B.TailPolicy;
+    MaskPolicy |= B.MaskPolicy;
+  }
+
 #if !defined(NDEBUG) || defined(LLVM_ENABLE_DUMP)
   /// Support for debugging, callable in GDB: V->dump()
   LLVM_DUMP_METHOD void dump() const {
@@ -356,9 +375,11 @@ DemandedFields getDemanded(const MachineInstr &MI,
   // Most instructions don't use any of these subfeilds.
   DemandedFields Res;
   // Start conservative if registers are used
-  if (MI.isCall() || MI.isInlineAsm() || MI.readsRegister(RISCV::VL))
+  if (MI.isCall() || MI.isInlineAsm() ||
+      MI.readsRegister(RISCV::VL, /*TRI=*/nullptr))
     Res.demandVL();
-  if (MI.isCall() || MI.isInlineAsm() || MI.readsRegister(RISCV::VTYPE))
+  if (MI.isCall() || MI.isInlineAsm() ||
+      MI.readsRegister(RISCV::VTYPE, /*TRI=*/nullptr))
     Res.demandVTYPE();
   // Start conservative on the unlowered form too
   uint64_t TSFlags = MI.getDesc().TSFlags;
@@ -433,8 +454,12 @@ DemandedFields getDemanded(const MachineInstr &MI,
 /// Defines the abstract state with which the forward dataflow models the
 /// values of the VL and VTYPE registers after insertion.
 class VSETVLIInfo {
+  struct AVLDef {
+    const MachineInstr *DefMI;
+    Register DefReg;
+  };
   union {
-    Register AVLReg;
+    AVLDef AVLRegDef;
     unsigned AVLImm;
   };
 
@@ -442,6 +467,8 @@ class VSETVLIInfo {
     Uninitialized,
     AVLIsReg,
     AVLIsImm,
+    AVLIsVLMAX,
+    AVLIsIgnored,
     Unknown,
   } State = Uninitialized;
 
@@ -467,9 +494,10 @@ public:
   void setUnknown() { State = Unknown; }
   bool isUnknown() const { return State == Unknown; }
 
-  void setAVLReg(Register Reg) {
-    assert(Reg.isVirtual() || Reg == RISCV::X0 || Reg == RISCV::NoRegister);
-    AVLReg = Reg;
+  void setAVLRegDef(const MachineInstr *DefMI, Register AVLReg) {
+    assert(DefMI && AVLReg.isVirtual());
+    AVLRegDef.DefMI = DefMI;
+    AVLRegDef.DefReg = AVLReg;
     State = AVLIsReg;
   }
 
@@ -478,23 +506,37 @@ public:
     State = AVLIsImm;
   }
 
+  void setAVLVLMAX() { State = AVLIsVLMAX; }
+
+  void setAVLIgnored() { State = AVLIsIgnored; }
+
   bool hasAVLImm() const { return State == AVLIsImm; }
   bool hasAVLReg() const { return State == AVLIsReg; }
+  bool hasAVLVLMAX() const { return State == AVLIsVLMAX; }
+  bool hasAVLIgnored() const { return State == AVLIsIgnored; }
   Register getAVLReg() const {
-    assert(hasAVLReg());
-    return AVLReg;
+    assert(hasAVLReg() && AVLRegDef.DefReg.isVirtual());
+    return AVLRegDef.DefReg;
   }
   unsigned getAVLImm() const {
     assert(hasAVLImm());
     return AVLImm;
   }
+  const MachineInstr &getAVLDefMI() const {
+    assert(hasAVLReg() && AVLRegDef.DefMI);
+    return *AVLRegDef.DefMI;
+  }
 
   void setAVL(VSETVLIInfo Info) {
     assert(Info.isValid());
     if (Info.isUnknown())
       setUnknown();
     else if (Info.hasAVLReg())
-      setAVLReg(Info.getAVLReg());
+      setAVLRegDef(&Info.getAVLDefMI(), Info.getAVLReg());
+    else if (Info.hasAVLVLMAX())
+      setAVLVLMAX();
+    else if (Info.hasAVLIgnored())
+      setAVLIgnored();
     else {
       assert(Info.hasAVLImm());
       setAVLImm(Info.getAVLImm());
@@ -506,34 +548,38 @@ public:
   bool getTailAgnostic() const { return TailAgnostic; }
   bool getMaskAgnostic() const { return MaskAgnostic; }
 
-  bool hasNonZeroAVL(const MachineRegisterInfo &MRI) const {
+  bool hasNonZeroAVL() const {
     if (hasAVLImm())
       return getAVLImm() > 0;
-    if (hasAVLReg()) {
-      if (getAVLReg() == RISCV::X0)
-        return true;
-      if (MachineInstr *MI = MRI.getVRegDef(getAVLReg());
-          MI && isNonZeroLoadImmediate(*MI))
-        return true;
+    if (hasAVLReg())
+      return isNonZeroLoadImmediate(getAVLDefMI());
+    if (hasAVLVLMAX())
+      return true;
+    if (hasAVLIgnored())
       return false;
-    }
     return false;
   }
 
-  bool hasEquallyZeroAVL(const VSETVLIInfo &Other,
-                         const MachineRegisterInfo &MRI) const {
+  bool hasEquallyZeroAVL(const VSETVLIInfo &Other) const {
     if (hasSameAVL(Other))
       return true;
-    return (hasNonZeroAVL(MRI) && Other.hasNonZeroAVL(MRI));
+    return (hasNonZeroAVL() && Other.hasNonZeroAVL());
   }
 
   bool hasSameAVL(const VSETVLIInfo &Other) const {
     if (hasAVLReg() && Other.hasAVLReg())
-      return getAVLReg() == Other.getAVLReg();
+      return AVLRegDef.DefMI == Other.AVLRegDef.DefMI &&
+             AVLRegDef.DefReg == Other.AVLRegDef.DefReg;
 
     if (hasAVLImm() && Other.hasAVLImm())
       return getAVLImm() == Other.getAVLImm();
 
+    if (hasAVLVLMAX())
+      return Other.hasAVLVLMAX() && hasSameVLMAX(Other);
+
+    if (hasAVLIgnored())
+      return Other.hasAVLIgnored();
+
     return false;
   }
 
@@ -619,7 +665,7 @@ public:
     if (Used.VLAny && !(hasSameAVL(Require) && hasSameVLMAX(Require)))
       return false;
 
-    if (Used.VLZeroness && !hasEquallyZeroAVL(Require, MRI))
+    if (Used.VLZeroness && !hasEquallyZeroAVL(Require))
       return false;
 
     return hasCompatibleVTYPE(Used, Require);
@@ -704,9 +750,13 @@ public:
     if (isUnknown())
       OS << "unknown";
     if (hasAVLReg())
-      OS << "AVLReg=" << (unsigned)AVLReg;
+      OS << "AVLReg=" << (unsigned)getAVLReg();
     if (hasAVLImm())
       OS << "AVLImm=" << (unsigned)AVLImm;
+    if (hasAVLVLMAX())
+      OS << "AVLVLMAX";
+    if (hasAVLIgnored())
+      OS << "AVLIgnored";
     OS << ", "
        << "VLMul=" << (unsigned)VLMul << ", "
        << "SEW=" << (unsigned)SEW << ", "
@@ -778,11 +828,40 @@ private:
                              VSETVLIInfo &Info) const;
   void computeIncomingVLVTYPE(const MachineBasicBlock &MBB);
   void emitVSETVLIs(MachineBasicBlock &MBB);
-  void doLocalPostpass(MachineBasicBlock &MBB);
   void doPRE(MachineBasicBlock &MBB);
   void insertReadVL(MachineBasicBlock &MBB);
 };
 
+class RISCVCoalesceVSETVLI : public MachineFunctionPass {
+public:
+  static char ID;
+  const RISCVSubtarget *ST;
+  const TargetInstrInfo *TII;
+  MachineRegisterInfo *MRI;
+  LiveIntervals *LIS;
+
+  RISCVCoalesceVSETVLI() : MachineFunctionPass(ID) {}
+  bool runOnMachineFunction(MachineFunction &MF) override;
+
+  void getAnalysisUsage(AnalysisUsage &AU) const override {
+    AU.setPreservesCFG();
+
+    AU.addRequired<LiveIntervals>();
+    AU.addPreserved<LiveIntervals>();
+    AU.addRequired<SlotIndexes>();
+    AU.addPreserved<SlotIndexes>();
+    AU.addPreserved<LiveDebugVariables>();
+    AU.addPreserved<LiveStacks>();
+
+    MachineFunctionPass::getAnalysisUsage(AU);
+  }
+
+  StringRef getPassName() const override { return RISCV_COALESCE_VSETVLI_NAME; }
+
+private:
+  bool coalesceVSETVLIs(MachineBasicBlock &MBB);
+};
+
 } // end anonymous namespace
 
 char RISCVInsertVSETVLI::ID = 0;
@@ -790,9 +869,15 @@ char RISCVInsertVSETVLI::ID = 0;
 INITIALIZE_PASS(RISCVInsertVSETVLI, DEBUG_TYPE, RISCV_INSERT_VSETVLI_NAME,
                 false, false)
 
+char RISCVCoalesceVSETVLI::ID = 0;
+
+INITIALIZE_PASS(RISCVCoalesceVSETVLI, "riscv-coalesce-vsetvli",
+                RISCV_COALESCE_VSETVLI_NAME, false, false)
+
 // Return a VSETVLIInfo representing the changes made by this VSETVLI or
 // VSETIVLI instruction.
-static VSETVLIInfo getInfoForVSETVLI(const MachineInstr &MI) {
+static VSETVLIInfo getInfoForVSETVLI(const MachineInstr &MI,
+                                     const MachineRegisterInfo &MRI) {
   VSETVLIInfo NewInfo;
   if (MI.getOpcode() == RISCV::PseudoVSETIVLI) {
     NewInfo.setAVLImm(MI.getOperand(1).getImm());
@@ -802,7 +887,10 @@ static VSETVLIInfo getInfoForVSETVLI(const MachineInstr &MI) {
     Register AVLReg = MI.getOperand(1).getReg();
     assert((AVLReg != RISCV::X0 || MI.getOperand(0).getReg() != RISCV::X0) &&
            "Can't handle X0, X0 vsetvli yet");
-    NewInfo.setAVLReg(AVLReg);
+    if (AVLReg == RISCV::X0)
+      NewInfo.setAVLVLMAX();
+    else
+      NewInfo.setAVLRegDef(MRI.getVRegDef(AVLReg), AVLReg);
   }
   NewInfo.setVTYPE(MI.getOperand(2).getImm());
 
@@ -869,16 +957,19 @@ static VSETVLIInfo computeInfoForInstr(const MachineInstr &MI, uint64_t TSFlags,
         if (ST.getRealMinVLen() == ST.getRealMaxVLen() && VLMAX <= 31)
           InstrInfo.setAVLImm(VLMAX);
         else
-          InstrInfo.setAVLReg(RISCV::X0);
+          InstrInfo.setAVLVLMAX();
       }
       else
         InstrInfo.setAVLImm(Imm);
     } else {
-      InstrInfo.setAVLReg(VLOp.getReg());
+      InstrInfo.setAVLRegDef(MRI->getVRegDef(VLOp.getReg()), VLOp.getReg());
     }
   } else {
     assert(isScalarExtractInstr(MI));
-    InstrInfo.setAVLReg(RISCV::NoRegister);
+    // TODO: If we are more clever about x0,x0 insertion then we should be able
+    // to deduce that the VL is ignored based off of DemandedFields, and remove
+    // the AVLIsIgnored state. Then we can just use an arbitrary immediate AVL.
+    InstrInfo.setAVLIgnored();
   }
 #ifndef NDEBUG
   if (std::optional<unsigned> EEW = getEEWForLoadStore(MI)) {
@@ -891,14 +982,13 @@ static VSETVLIInfo computeInfoForInstr(const MachineInstr &MI, uint64_t TSFlags,
   // AVL operand with the AVL of the defining vsetvli.  We avoid general
   // register AVLs to avoid extending live ranges without being sure we can
   // kill the original source reg entirely.
-  if (InstrInfo.hasAVLReg() && InstrInfo.getAVLReg().isVirtual()) {
-    MachineInstr *DefMI = MRI->getVRegDef(InstrInfo.getAVLReg());
-    if (DefMI && isVectorConfigInstr(*DefMI)) {
-      VSETVLIInfo DefInstrInfo = getInfoForVSETVLI(*DefMI);
+  if (InstrInfo.hasAVLReg()) {
+    const MachineInstr &DefMI = InstrInfo.getAVLDefMI();
+    if (isVectorConfigInstr(DefMI)) {
+      VSETVLIInfo DefInstrInfo = getInfoForVSETVLI(DefMI, *MRI);
       if (DefInstrInfo.hasSameVLMAX(InstrInfo) &&
-          (DefInstrInfo.hasAVLImm() || DefInstrInfo.getAVLReg() == RISCV::X0)) {
+          (DefInstrInfo.hasAVLImm() || DefInstrInfo.hasAVLVLMAX()))
         InstrInfo.setAVL(DefInstrInfo);
-      }
     }
   }
 
@@ -932,19 +1022,17 @@ void RISCVInsertVSETVLI::insertVSETVLI(MachineBasicBlock &MBB,
     // If our AVL is a virtual register, it might be defined by a VSET(I)VLI. If
     // it has the same VLMAX we want and the last VL/VTYPE we observed is the
     // same, we can use the X0, X0 form.
-    if (Info.hasSameVLMAX(PrevInfo) && Info.hasAVLReg() &&
-        Info.getAVLReg().isVirtual()) {
-      if (MachineInstr *DefMI = MRI->getVRegDef(Info.getAVLReg())) {
-        if (isVectorConfigInstr(*DefMI)) {
-          VSETVLIInfo DefInfo = getInfoForVSETVLI(*DefMI);
-          if (DefInfo.hasSameAVL(PrevInfo) && DefInfo.hasSameVLMAX(PrevInfo)) {
-            BuildMI(MBB, InsertPt, DL, TII->get(RISCV::PseudoVSETVLIX0))
-                .addReg(RISCV::X0, RegState::Define | RegState::Dead)
-                .addReg(RISCV::X0, RegState::Kill)
-                .addImm(Info.encodeVTYPE())
-                .addReg(RISCV::VL, RegState::Implicit);
-            return;
-          }
+    if (Info.hasSameVLMAX(PrevInfo) && Info.hasAVLReg()) {
+      const MachineInstr &DefMI = Info.getAVLDefMI();
+      if (isVectorConfigInstr(DefMI)) {
+        VSETVLIInfo DefInfo = getInfoForVSETVLI(DefMI, *MRI);
+        if (DefInfo.hasSameAVL(PrevInfo) && DefInfo.hasSameVLMAX(PrevInfo)) {
+          BuildMI(MBB, InsertPt, DL, TII->get(RISCV::PseudoVSETVLIX0))
+              .addReg(RISCV::X0, RegState::Define | RegState::Dead)
+              .addReg(RISCV::X0, RegState::Kill)
+              .addImm(Info.encodeVTYPE())
+              .addReg(RISCV::VL, RegState::Implicit);
+          return;
         }
       }
     }
@@ -958,8 +1046,7 @@ void RISCVInsertVSETVLI::insertVSETVLI(MachineBasicBlock &MBB,
     return;
   }
 
-  Register AVLReg = Info.getAVLReg();
-  if (AVLReg == RISCV::NoRegister) {
+  if (Info.hasAVLIgnored()) {
     // We can only use x0, x0 if there's no chance of the vtype change causing
     // the previous vl to become invalid.
     if (PrevInfo.isValid() && !PrevInfo.isUnknown() &&
@@ -979,20 +1066,19 @@ void RISCVInsertVSETVLI::insertVSETVLI(MachineBasicBlock &MBB,
     return;
   }
 
-  if (AVLReg.isVirtual())
-    MRI->constrainRegClass(AVLReg, &RISCV::GPRNoX0RegClass);
-
-  // Use X0 as the DestReg unless AVLReg is X0. We also need to change the
-  // opcode if the AVLReg is X0 as they have different register classes for
-  // the AVL operand.
-  Register DestReg = RISCV::X0;
-  unsigned Opcode = RISCV::PseudoVSETVLI;
-  if (AVLReg == RISCV::X0) {
-    DestReg = MRI->createVirtualRegister(&RISCV::GPRRegClass);
-    Opcode = RISCV::PseudoVSETVLIX0;
+  if (Info.hasAVLVLMAX()) {
+    Register DestReg = MRI->createVirtualRegister(&RISCV::GPRRegClass);
+    BuildMI(MBB, InsertPt, DL, TII->get(RISCV::PseudoVSETVLIX0))
+        .addReg(DestReg, RegState::Define | RegState::Dead)
+        .addReg(RISCV::X0, RegState::Kill)
+        .addImm(Info.encodeVTYPE());
+    return;
   }
-  BuildMI(MBB, InsertPt, DL, TII->get(Opcode))
-      .addReg(DestReg, RegState::Define | RegState::Dead)
+
+  Register AVLReg = Info.getAVLReg();
+  MRI->constrainRegClass(AVLReg, &RISCV::GPRNoX0RegClass);
+  BuildMI(MBB, InsertPt, DL, TII->get(RISCV::PseudoVSETVLI))
+      .addReg(RISCV::X0, RegState::Define | RegState::Dead)
       .addReg(AVLReg)
       .addImm(Info.encodeVTYPE());
 }
@@ -1054,14 +1140,12 @@ bool RISCVInsertVSETVLI::needVSETVLI(const MachineInstr &MI,
   // it might be defined by a VSET(I)VLI. If it has the same VLMAX we need
   // and the last VL/VTYPE we observed is the same, we don't need a
   // VSETVLI here.
-  if (Require.hasAVLReg() && Require.getAVLReg().isVirtual() &&
-      CurInfo.hasCompatibleVTYPE(Used, Require)) {
-    if (MachineInstr *DefMI = MRI->getVRegDef(Require.getAVLReg())) {
-      if (isVectorConfigInstr(*DefMI)) {
-        VSETVLIInfo DefInfo = getInfoForVSETVLI(*DefMI);
-        if (DefInfo.hasSameAVL(CurInfo) && DefInfo.hasSameVLMAX(CurInfo))
-          return false;
-      }
+  if (Require.hasAVLReg() && CurInfo.hasCompatibleVTYPE(Used, Require)) {
+    const MachineInstr &DefMI = Require.getAVLDefMI();
+    if (isVectorConfigInstr(DefMI)) {
+      VSETVLIInfo DefInfo = getInfoForVSETVLI(DefMI, *MRI);
+      if (DefInfo.hasSameAVL(CurInfo) && DefInfo.hasSameVLMAX(CurInfo))
+        return false;
     }
   }
 
@@ -1114,7 +1198,7 @@ void RISCVInsertVSETVLI::transferBefore(VSETVLIInfo &Info,
   // variant, so we avoid the transform to prevent extending live range of an
   // avl register operand.
   // TODO: We can probably relax this for immediates.
-  bool EquallyZero = IncomingInfo.hasEquallyZeroAVL(PrevInfo, *MRI) &&
+  bool EquallyZero = IncomingInfo.hasEquallyZeroAVL(PrevInfo) &&
                      IncomingInfo.hasSameVLMAX(PrevInfo);
   if (Demanded.VLAny || (Demanded.VLZeroness && !EquallyZero))
     Info.setAVL(IncomingInfo);
@@ -1145,20 +1229,22 @@ void RISCVInsertVSETVLI::transferBefore(VSETVLIInfo &Info,
 void RISCVInsertVSETVLI::transferAfter(VSETVLIInfo &Info,
                                        const MachineInstr &MI) const {
   if (isVectorConfigInstr(MI)) {
-    Info = getInfoForVSETVLI(MI);
+    Info = getInfoForVSETVLI(MI, *MRI);
     return;
   }
 
   if (RISCV::isFaultFirstLoad(MI)) {
     // Update AVL to vl-output of the fault first load.
-    Info.setAVLReg(MI.getOperand(1).getReg());
+    Info.setAVLRegDef(MRI->getVRegDef(MI.getOperand(1).getReg()),
+                      MI.getOperand(1).getReg());
     return;
   }
 
   // If this is something that updates VL/VTYPE that we don't know about, set
   // the state to unknown.
-  if (MI.isCall() || MI.isInlineAsm() || MI.modifiesRegister(RISCV::VL) ||
-      MI.modifiesRegister(RISCV::VTYPE))
+  if (MI.isCall() || MI.isInlineAsm() ||
+      MI.modifiesRegister(RISCV::VL, /*TRI=*/nullptr) ||
+      MI.modifiesRegister(RISCV::VTYPE, /*TRI=*/nullptr))
     Info = VSETVLIInfo::getUnknown();
 }
 
@@ -1244,24 +1330,16 @@ bool RISCVInsertVSETVLI::needVSETVLIPHI(const VSETVLIInfo &Require,
   if (!Require.hasAVLReg())
     return true;
 
-  Register AVLReg = Require.getAVLReg();
-  if (!AVLReg.isVirtual())
-    return true;
-
   // We need the AVL to be produce by a PHI node in this basic block.
-  MachineInstr *PHI = MRI->getVRegDef(AVLReg);
-  if (!PHI || PHI->getOpcode() != RISCV::PHI || PHI->getParent() != &MBB)
+  const MachineInstr *PHI = &Require.getAVLDefMI();
+  if (PHI->getOpcode() != RISCV::PHI || PHI->getParent() != &MBB)
     return true;
 
   for (unsigned PHIOp = 1, NumOps = PHI->getNumOperands(); PHIOp != NumOps;
        PHIOp += 2) {
     Register InReg = PHI->getOperand(PHIOp).getReg();
     MachineBasicBlock *PBB = PHI->getOperand(PHIOp + 1).getMBB();
-    const BlockData &PBBInfo = BlockInfo[PBB->getNumber()];
-    // If the exit from the predecessor has the VTYPE we are looking for
-    // we might be able to avoid a VSETVLI.
-    if (PBBInfo.Exit.isUnknown() || !PBBInfo.Exit.hasSameVTYPE(Require))
-      return true;
+    const VSETVLIInfo &PBBExit = BlockInfo[PBB->getNumber()].Exit;
 
     // We need the PHI input to the be the output of a VSET(I)VLI.
     MachineInstr *DefMI = MRI->getVRegDef(InReg);
@@ -1270,9 +1348,14 @@ bool RISCVInsertVSETVLI::needVSETVLIPHI(const VSETVLIInfo &Require,
 
     // We found a VSET(I)VLI make sure it matches the output of the
     // predecessor block.
-    VSETVLIInfo DefInfo = getInfoForVSETVLI(*DefMI);
-    if (!DefInfo.hasSameAVL(PBBInfo.Exit) ||
-        !DefInfo.hasSameVTYPE(PBBInfo.Exit))
+    VSETVLIInfo DefInfo = getInfoForVSETVLI(*DefMI, *MRI);
+    if (DefInfo != PBBExit)
+      return true;
+
+    // Require has the same VL as PBBExit, so if the exit from the
+    // predecessor has the VTYPE we are looking for we might be able
+    // to avoid a VSETVLI.
+    if (PBBExit.isUnknown() || !PBBExit.hasSameVTYPE(Require))
       return true;
   }
 
@@ -1341,8 +1424,9 @@ void RISCVInsertVSETVLI::emitVSETVLIs(MachineBasicBlock &MBB) {
                                               /*isImp*/ true));
     }
 
-    if (MI.isCall() || MI.isInlineAsm() || MI.modifiesRegister(RISCV::VL) ||
-        MI.modifiesRegister(RISCV::VTYPE))
+    if (MI.isCall() || MI.isInlineAsm() ||
+        MI.modifiesRegister(RISCV::VL, /*TRI=*/nullptr) ||
+        MI.modifiesRegister(RISCV::VTYPE, /*TRI=*/nullptr))
       PrefixTransparent = false;
 
     transferAfter(CurInfo, MI);
@@ -1417,10 +1501,8 @@ void RISCVInsertVSETVLI::doPRE(MachineBasicBlock &MBB) {
   // If the AVL value is a register (other than our VLMAX sentinel),
   // we need to prove the value is available at the point we're going
   // to insert the vsetvli at.
-  if (AvailableInfo.hasAVLReg() && RISCV::X0 != AvailableInfo.getAVLReg()) {
-    MachineInstr *AVLDefMI = MRI->getVRegDef(AvailableInfo.getAVLReg());
-    if (!AVLDefMI)
-      return;
+  if (AvailableInfo.hasAVLReg()) {
+    const MachineInstr *AVLDefMI = &AvailableInfo.getAVLDefMI();
     // This is an inline dominance check which covers the case of
     // UnavailablePred being the preheader of a loop.
     if (AVLDefMI->getParent() != UnavailablePred)
@@ -1430,6 +1512,11 @@ void RISCVInsertVSETVLI::doPRE(MachineBasicBlock &MBB) {
         return;
   }
 
+  // If the AVL isn't used in its predecessors then bail, since we have no AVL
+  // to insert a vsetvli with.
+  if (AvailableInfo.hasAVLIgnored())
+    return;
+
   // Model the effect of changing the input state of the block MBB to
   // AvailableInfo.  We're looking for two issues here; one legality,
   // one profitability.
@@ -1478,16 +1565,6 @@ void RISCVInsertVSETVLI::doPRE(MachineBasicBlock &MBB) {
                 AvailableInfo, OldExit);
 }
 
-static void doUnion(DemandedFields &A, DemandedFields B) {
-  A.VLAny |= B.VLAny;
-  A.VLZeroness |= B.VLZeroness;
-  A.SEW = std::max(A.SEW, B.SEW);
-  A.LMUL |= B.LMUL;
-  A.SEWLMULRatio |= B.SEWLMULRatio;
-  A.TailPolicy |= B.TailPolicy;
-  A.MaskPolicy |= B.MaskPolicy;
-}
-
 // Return true if we can mutate PrevMI to match MI without changing any the
 // fields which would be observed.
 static bool canMutatePriorConfig(const MachineInstr &PrevMI,
@@ -1504,19 +1581,19 @@ static bool canMutatePriorConfig(const MachineInstr &PrevMI,
     if (Used.VLZeroness) {
       if (isVLPreservingConfig(PrevMI))
         return false;
-      if (!getInfoForVSETVLI(PrevMI).hasEquallyZeroAVL(getInfoForVSETVLI(MI),
-                                                       MRI))
+      if (!getInfoForVSETVLI(PrevMI, MRI)
+               .hasEquallyZeroAVL(getInfoForVSETVLI(MI, MRI)))
         return false;
     }
 
     auto &AVL = MI.getOperand(1);
     auto &PrevAVL = PrevMI.getOperand(1);
-    assert(MRI.isSSA());
 
     // If the AVL is a register, we need to make sure MI's AVL dominates PrevMI.
     // For now just check that PrevMI uses the same virtual register.
     if (AVL.isReg() && AVL.getReg() != RISCV::X0 &&
-        (!PrevAVL.isReg() || PrevAVL.getReg() != AVL.getReg()))
+        (!MRI.hasOneDef(AVL.getReg()) || !PrevAVL.isReg() ||
+         PrevAVL.getReg() != AVL.getReg()))
       return false;
   }
 
@@ -1526,7 +1603,7 @@ static bool canMutatePriorConfig(const MachineInstr &PrevMI,
   return areCompatibleVTYPEs(PriorVType, VType, Used);
 }
 
-void RISCVInsertVSETVLI::doLocalPostpass(MachineBasicBlock &MBB) {
+bool RISCVCoalesceVSETVLI::coalesceVSETVLIs(MachineBasicBlock &MBB) {
   MachineInstr *NextMI = nullptr;
   // We can have arbitrary code in successors, so VL and VTYPE
   // must be considered demanded.
@@ -1537,9 +1614,10 @@ void RISCVInsertVSETVLI::doLocalPostpass(MachineBasicBlock &MBB) {
   for (MachineInstr &MI : make_range(MBB.rbegin(), MBB.rend())) {
 
     if (!isVectorConfigInstr(MI)) {
-      doUnion(Used, getDemanded(MI, MRI, ST));
-      if (MI.isCall() || MI.isInlineAsm() || MI.modifiesRegister(RISCV::VL) ||
-          MI.modifiesRegister(RISCV::VTYPE))
+      Used.doUnion(getDemanded(MI, MRI, ST));
+      if (MI.isCall() || MI.isInlineAsm() ||
+          MI.modifiesRegister(RISCV::VL, /*TRI=*/nullptr) ||
+          MI.modifiesRegister(RISCV::VTYPE, /*TRI=*/nullptr))
         NextMI = nullptr;
       continue;
     }
@@ -1558,8 +1636,28 @@ void RISCVInsertVSETVLI::doLocalPostpass(MachineBasicBlock &MBB) {
 
       if (canMutatePriorConfig(MI, *NextMI, Used, *MRI)) {
         if (!isVLPreservingConfig(*NextMI)) {
-          MI.getOperand(0).setReg(NextMI->getOperand(0).getReg());
+          Register DefReg = NextMI->getOperand(0).getReg();
+
+          MI.getOperand(0).setReg(DefReg);
           MI.getOperand(0).setIsDead(false);
+
+          // The def of DefReg moved to MI, so extend the LiveInterval up to
+          // it.
+          if (DefReg.isVirtual()) {
+            LiveInterval &DefLI = LIS->getInterval(DefReg);
+            SlotIndex MISlot = LIS->getInstructionIndex(MI).getRegSlot();
+            VNInfo *DefVNI = DefLI.getVNInfoAt(DefLI.beginIndex());
+            LiveInterval::Segment S(MISlot, DefLI.beginIndex(), DefVNI);
+            DefLI.addSegment(S);
+            DefVNI->def = MISlot;
+            // Mark DefLI as spillable if it was previously unspillable
+            DefLI.setWeight(0);
+
+            // DefReg may have had no uses, in which case we need to shrink
+            // the LiveInterval up to MI.
+            LIS->shrinkToUses(&DefLI);
+          }
+
           Register OldVLReg;
           if (MI.getOperand(1).isReg())
             OldVLReg = MI.getOperand(1).getReg();
@@ -1567,11 +1665,21 @@ void RISCVInsertVSETVLI::doLocalPostpass(MachineBasicBlock &MBB) {
             MI.getOperand(1).ChangeToImmediate(NextMI->getOperand(1).getImm());
           else
             MI.getOperand(1).ChangeToRegister(NextMI->getOperand(1).getReg(), false);
-          if (OldVLReg) {
+
+          // Clear NextMI's AVL early so we're not counting it as a use.
+          if (NextMI->getOperand(1).isReg())
+            NextMI->getOperand(1).setReg(RISCV::NoRegister);
+
+          if (OldVLReg && OldVLReg.isVirtual()) {
+            // NextMI no longer uses OldVLReg so shrink its LiveInterval.
+            LIS->shrinkToUses(&LIS->getInterval(OldVLReg));
+
             MachineInstr *VLOpDef = MRI->getUniqueVRegDef(OldVLReg);
             if (VLOpDef && TII->isAddImmediate(*VLOpDef, OldVLReg) &&
-                MRI->use_nodbg_empty(OldVLReg))
+                MRI->use_nodbg_empty(OldVLReg)) {
               VLOpDef->eraseFromParent();
+              LIS->removeInterval(OldVLReg);
+            }
           }
           MI.setDesc(NextMI->getDesc());
         }
@@ -1584,9 +1692,13 @@ void RISCVInsertVSETVLI::doLocalPostpass(MachineBasicBlock &MBB) {
     Used = getDemanded(MI, MRI, ST);
   }
 
-  NumRemovedVSETVL += ToDelete.size();
-  for (auto *MI : ToDelete)
+  NumCoalescedVSETVL += ToDelete.size();
+  for (auto *MI : ToDelete) {
+    LIS->RemoveMachineInstrFromMaps(*MI);
     MI->eraseFromParent();
+  }
+
+  return !ToDelete.empty();
 }
 
 void RISCVInsertVSETVLI::insertReadVL(MachineBasicBlock &MBB) {
@@ -1661,15 +1773,6 @@ bool RISCVInsertVSETVLI::runOnMachineFunction(MachineFunction &MF) {
   for (MachineBasicBlock &MBB : MF)
     emitVSETVLIs(MBB);
 
-  // Now that all vsetvlis are explicit, go through and do block local
-  // DSE and peephole based demanded fields based transforms.  Note that
-  // this *must* be done outside the main dataflow so long as we allow
-  // any cross block analysis within the dataflow.  We can't have both
-  // demanded fields based mutation and non-local analysis in the
-  // dataflow at the same time without introducing inconsistencies.
-  for (MachineBasicBlock &MBB : MF)
-    doLocalPostpass(MBB);
-
   // Insert PseudoReadVL after VLEFF/VLSEGFF and replace it with the vl output
   // of VLEFF/VLSEGFF.
   for (MachineBasicBlock &MBB : MF)
@@ -1683,3 +1786,29 @@ bool RISCVInsertVSETVLI::runOnMachineFunction(MachineFunction &MF) {
 FunctionPass *llvm::createRISCVInsertVSETVLIPass() {
   return new RISCVInsertVSETVLI();
 }
+
+// Now that all vsetvlis are explicit, go through and do block local
+// DSE and peephole based demanded fields based transforms.  Note that
+// this *must* be done outside the main dataflow so long as we allow
+// any cross block analysis within the dataflow.  We can't have both
+// demanded fields based mutation and non-local analysis in the
+// dataflow at the same time without introducing inconsistencies.
+bool RISCVCoalesceVSETVLI::runOnMachineFunction(MachineFunction &MF) {
+  // Skip if the vector extension is not enabled.
+  ST = &MF.getSubtarget<RISCVSubtarget>();
+  if (!ST->hasVInstructions())
+    return false;
+  TII = ST->getInstrInfo();
+  MRI = &MF.getRegInfo();
+  LIS = &getAnalysis<LiveIntervals>();
+
+  bool Changed = false;
+  for (MachineBasicBlock &MBB : MF)
+    Changed |= coalesceVSETVLIs(MBB);
+
+  return Changed;
+}
+
+FunctionPass *llvm::createRISCVCoalesceVSETVLIPass() {
+  return new RISCVCoalesceVSETVLI();
+}
diff --git a/llvm/lib/Target/RISCV/RISCVInsertWriteVXRM.cpp b/llvm/lib/Target/RISCV/RISCVInsertWriteVXRM.cpp
index e487cc8b2e20..f72ba2d5c667 100644
--- a/llvm/lib/Target/RISCV/RISCVInsertWriteVXRM.cpp
+++ b/llvm/lib/Target/RISCV/RISCVInsertWriteVXRM.cpp
@@ -225,7 +225,8 @@ bool RISCVInsertWriteVXRM::computeVXRMChanges(const MachineBasicBlock &MBB) {
       continue;
     }
 
-    if (MI.isCall() || MI.isInlineAsm() || MI.modifiesRegister(RISCV::VXRM)) {
+    if (MI.isCall() || MI.isInlineAsm() ||
+        MI.modifiesRegister(RISCV::VXRM, /*TRI=*/nullptr)) {
       if (!BBInfo.VXRMUse.isValid())
         BBInfo.VXRMUse.setUnknown();
 
@@ -386,7 +387,8 @@ void RISCVInsertWriteVXRM::emitWriteVXRM(MachineBasicBlock &MBB) {
       continue;
     }
 
-    if (MI.isCall() || MI.isInlineAsm() || MI.modifiesRegister(RISCV::VXRM))
+    if (MI.isCall() || MI.isInlineAsm() ||
+        MI.modifiesRegister(RISCV::VXRM, /*TRI=*/nullptr))
       Info.setUnknown();
   }
 
diff --git a/llvm/lib/Target/RISCV/RISCVInstrInfo.cpp b/llvm/lib/Target/RISCV/RISCVInstrInfo.cpp
index 8331fc0b8c30..8cb9a40a98bc 100644
--- a/llvm/lib/Target/RISCV/RISCVInstrInfo.cpp
+++ b/llvm/lib/Target/RISCV/RISCVInstrInfo.cpp
@@ -66,6 +66,13 @@ using namespace RISCV;
 
 } // namespace llvm::RISCVVPseudosTable
 
+namespace llvm::RISCV {
+
+#define GET_RISCVMaskedPseudosTable_IMPL
+#include "RISCVGenSearchableTables.inc"
+
+} // end namespace llvm::RISCV
+
 RISCVInstrInfo::RISCVInstrInfo(RISCVSubtarget &STI)
     : RISCVGenInstrInfo(RISCV::ADJCALLSTACKDOWN, RISCV::ADJCALLSTACKUP),
       STI(STI) {}
@@ -239,7 +246,7 @@ static bool isConvertibleToVMV_V_V(const RISCVSubtarget &STI,
     } else if (MBBI->getNumDefs()) {
       // Check all the instructions which will change VL.
       // For example, vleff has implicit def VL.
-      if (MBBI->modifiesRegister(RISCV::VL))
+      if (MBBI->modifiesRegister(RISCV::VL, /*TRI=*/nullptr))
         return false;
 
       // Only converting whole register copies to vmv.v.v when the defining
@@ -1582,10 +1589,10 @@ void RISCVInstrInfo::finalizeInsInstrs(
   MachineFunction &MF = *Root.getMF();
 
   for (auto *NewMI : InsInstrs) {
-    assert(static_cast<unsigned>(RISCV::getNamedOperandIdx(
-               NewMI->getOpcode(), RISCV::OpName::frm)) ==
-               NewMI->getNumOperands() &&
-           "Instruction has unexpected number of operands");
+    // We'd already added the FRM operand.
+    if (static_cast<unsigned>(RISCV::getNamedOperandIdx(
+            NewMI->getOpcode(), RISCV::OpName::frm)) != NewMI->getNumOperands())
+      continue;
     MachineInstrBuilder MIB(MF, NewMI);
     MIB.add(FRM);
     if (FRM.getImm() == RISCVFPRndMode::DYN)
@@ -1626,8 +1633,230 @@ static bool isFMUL(unsigned Opc) {
   }
 }
 
+bool RISCVInstrInfo::isVectorAssociativeAndCommutative(const MachineInstr &Inst,
+                                                       bool Invert) const {
+#define OPCODE_LMUL_CASE(OPC)                                                  \
+  case RISCV::OPC##_M1:                                                        \
+  case RISCV::OPC##_M2:                                                        \
+  case RISCV::OPC##_M4:                                                        \
+  case RISCV::OPC##_M8:                                                        \
+  case RISCV::OPC##_MF2:                                                       \
+  case RISCV::OPC##_MF4:                                                       \
+  case RISCV::OPC##_MF8
+
+#define OPCODE_LMUL_MASK_CASE(OPC)                                             \
+  case RISCV::OPC##_M1_MASK:                                                   \
+  case RISCV::OPC##_M2_MASK:                                                   \
+  case RISCV::OPC##_M4_MASK:                                                   \
+  case RISCV::OPC##_M8_MASK:                                                   \
+  case RISCV::OPC##_MF2_MASK:                                                  \
+  case RISCV::OPC##_MF4_MASK:                                                  \
+  case RISCV::OPC##_MF8_MASK
+
+  unsigned Opcode = Inst.getOpcode();
+  if (Invert) {
+    if (auto InvOpcode = getInverseOpcode(Opcode))
+      Opcode = *InvOpcode;
+    else
+      return false;
+  }
+
+  // clang-format off
+  switch (Opcode) {
+  default:
+    return false;
+  OPCODE_LMUL_CASE(PseudoVADD_VV):
+  OPCODE_LMUL_MASK_CASE(PseudoVADD_VV):
+  OPCODE_LMUL_CASE(PseudoVMUL_VV):
+  OPCODE_LMUL_MASK_CASE(PseudoVMUL_VV):
+    return true;
+  }
+  // clang-format on
+
+#undef OPCODE_LMUL_MASK_CASE
+#undef OPCODE_LMUL_CASE
+}
+
+bool RISCVInstrInfo::areRVVInstsReassociable(const MachineInstr &Root,
+                                             const MachineInstr &Prev) const {
+  if (!areOpcodesEqualOrInverse(Root.getOpcode(), Prev.getOpcode()))
+    return false;
+
+  assert(Root.getMF() == Prev.getMF());
+  const MachineRegisterInfo *MRI = &Root.getMF()->getRegInfo();
+  const TargetRegisterInfo *TRI = MRI->getTargetRegisterInfo();
+
+  // Make sure vtype operands are also the same.
+  const MCInstrDesc &Desc = get(Root.getOpcode());
+  const uint64_t TSFlags = Desc.TSFlags;
+
+  auto checkImmOperand = [&](unsigned OpIdx) {
+    return Root.getOperand(OpIdx).getImm() == Prev.getOperand(OpIdx).getImm();
+  };
+
+  auto checkRegOperand = [&](unsigned OpIdx) {
+    return Root.getOperand(OpIdx).getReg() == Prev.getOperand(OpIdx).getReg();
+  };
+
+  // PassThru
+  // TODO: Potentially we can loosen the condition to consider Root to be
+  // associable with Prev if Root has NoReg as passthru. In which case we
+  // also need to loosen the condition on vector policies between these.
+  if (!checkRegOperand(1))
+    return false;
+
+  // SEW
+  if (RISCVII::hasSEWOp(TSFlags) &&
+      !checkImmOperand(RISCVII::getSEWOpNum(Desc)))
+    return false;
+
+  // Mask
+  if (RISCVII::usesMaskPolicy(TSFlags)) {
+    const MachineBasicBlock *MBB = Root.getParent();
+    const MachineBasicBlock::const_reverse_iterator It1(&Root);
+    const MachineBasicBlock::const_reverse_iterator It2(&Prev);
+    Register MI1VReg;
+
+    bool SeenMI2 = false;
+    for (auto End = MBB->rend(), It = It1; It != End; ++It) {
+      if (It == It2) {
+        SeenMI2 = true;
+        if (!MI1VReg.isValid())
+          // There is no V0 def between Root and Prev; they're sharing the
+          // same V0.
+          break;
+      }
+
+      if (It->modifiesRegister(RISCV::V0, TRI)) {
+        Register SrcReg = It->getOperand(1).getReg();
+        // If it's not VReg it'll be more difficult to track its defs, so
+        // bailing out here just to be safe.
+        if (!SrcReg.isVirtual())
+          return false;
+
+        if (!MI1VReg.isValid()) {
+          // This is the V0 def for Root.
+          MI1VReg = SrcReg;
+          continue;
+        }
+
+        // Some random mask updates.
+        if (!SeenMI2)
+          continue;
+
+        // This is the V0 def for Prev; check if it's the same as that of
+        // Root.
+        if (MI1VReg != SrcReg)
+          return false;
+        else
+          break;
+      }
+    }
+
+    // If we haven't encountered Prev, it's likely that this function was
+    // called in a wrong way (e.g. Root is before Prev).
+    assert(SeenMI2 && "Prev is expected to appear before Root");
+  }
+
+  // Tail / Mask policies
+  if (RISCVII::hasVecPolicyOp(TSFlags) &&
+      !checkImmOperand(RISCVII::getVecPolicyOpNum(Desc)))
+    return false;
+
+  // VL
+  if (RISCVII::hasVLOp(TSFlags)) {
+    unsigned OpIdx = RISCVII::getVLOpNum(Desc);
+    const MachineOperand &Op1 = Root.getOperand(OpIdx);
+    const MachineOperand &Op2 = Prev.getOperand(OpIdx);
+    if (Op1.getType() != Op2.getType())
+      return false;
+    switch (Op1.getType()) {
+    case MachineOperand::MO_Register:
+      if (Op1.getReg() != Op2.getReg())
+        return false;
+      break;
+    case MachineOperand::MO_Immediate:
+      if (Op1.getImm() != Op2.getImm())
+        return false;
+      break;
+    default:
+      llvm_unreachable("Unrecognized VL operand type");
+    }
+  }
+
+  // Rounding modes
+  if (RISCVII::hasRoundModeOp(TSFlags) &&
+      !checkImmOperand(RISCVII::getVLOpNum(Desc) - 1))
+    return false;
+
+  return true;
+}
+
+// Most of our RVV pseudos have passthru operand, so the real operands
+// start from index = 2.
+bool RISCVInstrInfo::hasReassociableVectorSibling(const MachineInstr &Inst,
+                                                  bool &Commuted) const {
+  const MachineBasicBlock *MBB = Inst.getParent();
+  const MachineRegisterInfo &MRI = MBB->getParent()->getRegInfo();
+  assert(RISCVII::isFirstDefTiedToFirstUse(get(Inst.getOpcode())) &&
+         "Expect the present of passthrough operand.");
+  MachineInstr *MI1 = MRI.getUniqueVRegDef(Inst.getOperand(2).getReg());
+  MachineInstr *MI2 = MRI.getUniqueVRegDef(Inst.getOperand(3).getReg());
+
+  // If only one operand has the same or inverse opcode and it's the second
+  // source operand, the operands must be commuted.
+  Commuted = !areRVVInstsReassociable(Inst, *MI1) &&
+             areRVVInstsReassociable(Inst, *MI2);
+  if (Commuted)
+    std::swap(MI1, MI2);
+
+  return areRVVInstsReassociable(Inst, *MI1) &&
+         (isVectorAssociativeAndCommutative(*MI1) ||
+          isVectorAssociativeAndCommutative(*MI1, /* Invert */ true)) &&
+         hasReassociableOperands(*MI1, MBB) &&
+         MRI.hasOneNonDBGUse(MI1->getOperand(0).getReg());
+}
+
+bool RISCVInstrInfo::hasReassociableOperands(
+    const MachineInstr &Inst, const MachineBasicBlock *MBB) const {
+  if (!isVectorAssociativeAndCommutative(Inst) &&
+      !isVectorAssociativeAndCommutative(Inst, /*Invert=*/true))
+    return TargetInstrInfo::hasReassociableOperands(Inst, MBB);
+
+  const MachineOperand &Op1 = Inst.getOperand(2);
+  const MachineOperand &Op2 = Inst.getOperand(3);
+  const MachineRegisterInfo &MRI = MBB->getParent()->getRegInfo();
+
+  // We need virtual register definitions for the operands that we will
+  // reassociate.
+  MachineInstr *MI1 = nullptr;
+  MachineInstr *MI2 = nullptr;
+  if (Op1.isReg() && Op1.getReg().isVirtual())
+    MI1 = MRI.getUniqueVRegDef(Op1.getReg());
+  if (Op2.isReg() && Op2.getReg().isVirtual())
+    MI2 = MRI.getUniqueVRegDef(Op2.getReg());
+
+  // And at least one operand must be defined in MBB.
+  return MI1 && MI2 && (MI1->getParent() == MBB || MI2->getParent() == MBB);
+}
+
+void RISCVInstrInfo::getReassociateOperandIndices(
+    const MachineInstr &Root, unsigned Pattern,
+    std::array<unsigned, 5> &OperandIndices) const {
+  TargetInstrInfo::getReassociateOperandIndices(Root, Pattern, OperandIndices);
+  if (RISCV::getRVVMCOpcode(Root.getOpcode())) {
+    // Skip the passthrough operand, so increment all indices by one.
+    for (unsigned I = 0; I < 5; ++I)
+      ++OperandIndices[I];
+  }
+}
+
 bool RISCVInstrInfo::hasReassociableSibling(const MachineInstr &Inst,
                                             bool &Commuted) const {
+  if (isVectorAssociativeAndCommutative(Inst) ||
+      isVectorAssociativeAndCommutative(Inst, /*Invert=*/true))
+    return hasReassociableVectorSibling(Inst, Commuted);
+
   if (!TargetInstrInfo::hasReassociableSibling(Inst, Commuted))
     return false;
 
@@ -1647,6 +1876,9 @@ bool RISCVInstrInfo::hasReassociableSibling(const MachineInstr &Inst,
 
 bool RISCVInstrInfo::isAssociativeAndCommutative(const MachineInstr &Inst,
                                                  bool Invert) const {
+  if (isVectorAssociativeAndCommutative(Inst, Invert))
+    return true;
+
   unsigned Opc = Inst.getOpcode();
   if (Invert) {
     auto InverseOpcode = getInverseOpcode(Opc);
@@ -1699,6 +1931,38 @@ bool RISCVInstrInfo::isAssociativeAndCommutative(const MachineInstr &Inst,
 
 std::optional<unsigned>
 RISCVInstrInfo::getInverseOpcode(unsigned Opcode) const {
+#define RVV_OPC_LMUL_CASE(OPC, INV)                                            \
+  case RISCV::OPC##_M1:                                                        \
+    return RISCV::INV##_M1;                                                    \
+  case RISCV::OPC##_M2:                                                        \
+    return RISCV::INV##_M2;                                                    \
+  case RISCV::OPC##_M4:                                                        \
+    return RISCV::INV##_M4;                                                    \
+  case RISCV::OPC##_M8:                                                        \
+    return RISCV::INV##_M8;                                                    \
+  case RISCV::OPC##_MF2:                                                       \
+    return RISCV::INV##_MF2;                                                   \
+  case RISCV::OPC##_MF4:                                                       \
+    return RISCV::INV##_MF4;                                                   \
+  case RISCV::OPC##_MF8:                                                       \
+    return RISCV::INV##_MF8
+
+#define RVV_OPC_LMUL_MASK_CASE(OPC, INV)                                       \
+  case RISCV::OPC##_M1_MASK:                                                   \
+    return RISCV::INV##_M1_MASK;                                               \
+  case RISCV::OPC##_M2_MASK:                                                   \
+    return RISCV::INV##_M2_MASK;                                               \
+  case RISCV::OPC##_M4_MASK:                                                   \
+    return RISCV::INV##_M4_MASK;                                               \
+  case RISCV::OPC##_M8_MASK:                                                   \
+    return RISCV::INV##_M8_MASK;                                               \
+  case RISCV::OPC##_MF2_MASK:                                                  \
+    return RISCV::INV##_MF2_MASK;                                              \
+  case RISCV::OPC##_MF4_MASK:                                                  \
+    return RISCV::INV##_MF4_MASK;                                              \
+  case RISCV::OPC##_MF8_MASK:                                                  \
+    return RISCV::INV##_MF8_MASK
+
   switch (Opcode) {
   default:
     return std::nullopt;
@@ -1722,7 +1986,16 @@ RISCVInstrInfo::getInverseOpcode(unsigned Opcode) const {
     return RISCV::SUBW;
   case RISCV::SUBW:
     return RISCV::ADDW;
+    // clang-format off
+  RVV_OPC_LMUL_CASE(PseudoVADD_VV, PseudoVSUB_VV);
+  RVV_OPC_LMUL_MASK_CASE(PseudoVADD_VV, PseudoVSUB_VV);
+  RVV_OPC_LMUL_CASE(PseudoVSUB_VV, PseudoVADD_VV);
+  RVV_OPC_LMUL_MASK_CASE(PseudoVSUB_VV, PseudoVADD_VV);
+    // clang-format on
   }
+
+#undef RVV_OPC_LMUL_MASK_CASE
+#undef RVV_OPC_LMUL_CASE
 }
 
 static bool canCombineFPFusedMultiply(const MachineInstr &Root,
@@ -2859,6 +3132,11 @@ bool RISCVInstrInfo::findCommutedOpIndices(const MachineInstr &MI,
   case CASE_RVV_OPCODE_WIDEN(VWMACC_VV):
   case CASE_RVV_OPCODE_WIDEN(VWMACCU_VV):
   case CASE_RVV_OPCODE_UNMASK(VADC_VVM):
+  case CASE_RVV_OPCODE(VSADD_VV):
+  case CASE_RVV_OPCODE(VSADDU_VV):
+  case CASE_RVV_OPCODE(VAADD_VV):
+  case CASE_RVV_OPCODE(VAADDU_VV):
+  case CASE_RVV_OPCODE(VSMUL_VV):
     // Operands 2 and 3 are commutable.
     return fixCommutedOpIndices(SrcOpIdx1, SrcOpIdx2, 2, 3);
   case CASE_VFMA_SPLATS(FMADD):
@@ -3145,6 +3423,16 @@ MachineInstr *RISCVInstrInfo::commuteInstructionImpl(MachineInstr &MI,
   return TargetInstrInfo::commuteInstructionImpl(MI, NewMI, OpIdx1, OpIdx2);
 }
 
+#undef CASE_RVV_OPCODE_UNMASK_LMUL
+#undef CASE_RVV_OPCODE_MASK_LMUL
+#undef CASE_RVV_OPCODE_LMUL
+#undef CASE_RVV_OPCODE_UNMASK_WIDEN
+#undef CASE_RVV_OPCODE_UNMASK
+#undef CASE_RVV_OPCODE_MASK_WIDEN
+#undef CASE_RVV_OPCODE_MASK
+#undef CASE_RVV_OPCODE_WIDEN
+#undef CASE_RVV_OPCODE
+
 #undef CASE_VMA_OPCODE_COMMON
 #undef CASE_VMA_OPCODE_LMULS_M1
 #undef CASE_VMA_OPCODE_LMULS_MF2
@@ -3546,8 +3834,8 @@ RISCV::isRVVSpillForZvlsseg(unsigned Opcode) {
 }
 
 bool RISCV::isFaultFirstLoad(const MachineInstr &MI) {
-  return MI.getNumExplicitDefs() == 2 && MI.modifiesRegister(RISCV::VL) &&
-         !MI.isInlineAsm();
+  return MI.getNumExplicitDefs() == 2 &&
+         MI.modifiesRegister(RISCV::VL, /*TRI=*/nullptr) && !MI.isInlineAsm();
 }
 
 bool RISCV::hasEqualFRM(const MachineInstr &MI1, const MachineInstr &MI2) {
diff --git a/llvm/lib/Target/RISCV/RISCVInstrInfo.h b/llvm/lib/Target/RISCV/RISCVInstrInfo.h
index 70fe7da85be0..170f813eb10d 100644
--- a/llvm/lib/Target/RISCV/RISCVInstrInfo.h
+++ b/llvm/lib/Target/RISCV/RISCVInstrInfo.h
@@ -266,6 +266,9 @@ public:
       SmallVectorImpl<MachineInstr *> &DelInstrs,
       DenseMap<unsigned, unsigned> &InstrIdxForVirtReg) const override;
 
+  bool hasReassociableOperands(const MachineInstr &Inst,
+                               const MachineBasicBlock *MBB) const override;
+
   bool hasReassociableSibling(const MachineInstr &Inst,
                               bool &Commuted) const override;
 
@@ -274,6 +277,10 @@ public:
 
   std::optional<unsigned> getInverseOpcode(unsigned Opcode) const override;
 
+  void getReassociateOperandIndices(
+      const MachineInstr &Root, unsigned Pattern,
+      std::array<unsigned, 5> &OperandIndices) const override;
+
   ArrayRef<std::pair<MachineMemOperand::Flags, const char *>>
   getSerializableMachineMemOperandTargetFlags() const override;
 
@@ -297,6 +304,13 @@ protected:
 
 private:
   unsigned getInstBundleLength(const MachineInstr &MI) const;
+
+  bool isVectorAssociativeAndCommutative(const MachineInstr &MI,
+                                         bool Invert = false) const;
+  bool areRVVInstsReassociable(const MachineInstr &MI1,
+                               const MachineInstr &MI2) const;
+  bool hasReassociableVectorSibling(const MachineInstr &Inst,
+                                    bool &Commuted) const;
 };
 
 namespace RISCV {
@@ -359,5 +373,17 @@ struct PseudoInfo {
 
 } // end namespace RISCVVPseudosTable
 
+namespace RISCV {
+
+struct RISCVMaskedPseudoInfo {
+  uint16_t MaskedPseudo;
+  uint16_t UnmaskedPseudo;
+  uint8_t MaskOpIdx;
+  uint8_t MaskAffectsResult : 1;
+};
+#define GET_RISCVMaskedPseudosTable_DECL
+#include "RISCVGenSearchableTables.inc"
+} // end namespace RISCV
+
 } // end namespace llvm
 #endif
diff --git a/llvm/lib/Target/RISCV/RISCVInstrInfo.td b/llvm/lib/Target/RISCV/RISCVInstrInfo.td
index f9dadc6c0d48..da4020758eb6 100644
--- a/llvm/lib/Target/RISCV/RISCVInstrInfo.td
+++ b/llvm/lib/Target/RISCV/RISCVInstrInfo.td
@@ -1813,6 +1813,8 @@ def : Pat<(binop_allwusers<srl> (sext_inreg GPR:$rs1, i32), uimm5:$shamt),
 
 // Use binop_allwusers to recover immediates that may have been broken by
 // SimplifyDemandedBits.
+def : Pat<(binop_allwusers<and> GPR:$rs1, 0xffffffff),
+          (COPY GPR:$rs1)>;
 def : Pat<(binop_allwusers<and> GPR:$rs1, u32simm12:$imm),
           (ANDI GPR:$rs1, u32simm12:$imm)>;
 
diff --git a/llvm/lib/Target/RISCV/RISCVInstrInfoC.td b/llvm/lib/Target/RISCV/RISCVInstrInfoC.td
index 18d38348f721..f4e50d7aa45c 100644
--- a/llvm/lib/Target/RISCV/RISCVInstrInfoC.td
+++ b/llvm/lib/Target/RISCV/RISCVInstrInfoC.td
@@ -841,7 +841,7 @@ def : InstAlias<".insn_cj $opcode, $funct3, $imm11",
 //===----------------------------------------------------------------------===//
 
 // Patterns are defined in the same order the compressed instructions appear
-// on page 82 of the ISA manual.
+// under the "RVC Instruction Set Listings" section of the ISA manual.
 
 // Quadrant 0
 let Predicates = [HasStdExtCOrZca] in {
diff --git a/llvm/lib/Target/RISCV/RISCVInstrInfoVPseudos.td b/llvm/lib/Target/RISCV/RISCVInstrInfoVPseudos.td
index 435cd7f84c61..fc60a9cc7cd3 100644
--- a/llvm/lib/Target/RISCV/RISCVInstrInfoVPseudos.td
+++ b/llvm/lib/Target/RISCV/RISCVInstrInfoVPseudos.td
@@ -2146,8 +2146,9 @@ multiclass VPseudoBinaryRoundingMode<VReg RetClass,
                                      string Constraint = "",
                                      int sew = 0,
                                      int UsesVXRM = 1,
-                                     int TargetConstraintType = 1> {
-  let VLMul = MInfo.value, SEW=sew in {
+                                     int TargetConstraintType = 1,
+                                     bit Commutable = 0> {
+  let VLMul = MInfo.value, SEW=sew, isCommutable = Commutable in {
     defvar suffix = !if(sew, "_" # MInfo.MX # "_E" # sew, "_" # MInfo.MX);
     def suffix : VPseudoBinaryNoMaskRoundingMode<RetClass, Op1Class, Op2Class,
                                                  Constraint, UsesVXRM,
@@ -2232,8 +2233,9 @@ multiclass VPseudoBinaryV_VV<LMULInfo m, string Constraint = "", int sew = 0, bi
   defm _VV : VPseudoBinary<m.vrclass, m.vrclass, m.vrclass, m, Constraint, sew, Commutable=Commutable>;
 }
 
-multiclass VPseudoBinaryV_VV_RM<LMULInfo m, string Constraint = ""> {
-  defm _VV : VPseudoBinaryRoundingMode<m.vrclass, m.vrclass, m.vrclass, m, Constraint>;
+multiclass VPseudoBinaryV_VV_RM<LMULInfo m, string Constraint = "", bit Commutable = 0> {
+  defm _VV : VPseudoBinaryRoundingMode<m.vrclass, m.vrclass, m.vrclass, m, Constraint,
+                                       Commutable=Commutable>;
 }
 
 // Similar to VPseudoBinaryV_VV, but uses MxListF.
@@ -2715,10 +2717,11 @@ multiclass VPseudoVGTR_VV_VX_VI<Operand ImmType = simm5, string Constraint = "">
   }
 }
 
-multiclass VPseudoVSALU_VV_VX_VI<Operand ImmType = simm5, string Constraint = ""> {
+multiclass VPseudoVSALU_VV_VX_VI<Operand ImmType = simm5, string Constraint = "",
+                                 bit Commutable = 0> {
   foreach m = MxList in {
     defvar mx = m.MX;
-    defm "" : VPseudoBinaryV_VV<m, Constraint>,
+    defm "" : VPseudoBinaryV_VV<m, Constraint, Commutable=Commutable>,
               SchedBinary<"WriteVSALUV", "ReadVSALUV", "ReadVSALUX", mx,
                           forceMergeOpRead=true>;
     defm "" : VPseudoBinaryV_VX<m, Constraint>,
@@ -2788,7 +2791,7 @@ multiclass VPseudoVSALU_VV_VX {
 multiclass VPseudoVSMUL_VV_VX_RM {
   foreach m = MxList in {
     defvar mx = m.MX;
-    defm "" : VPseudoBinaryV_VV_RM<m>,
+    defm "" : VPseudoBinaryV_VV_RM<m, Commutable=1>,
               SchedBinary<"WriteVSMulV", "ReadVSMulV", "ReadVSMulV", mx,
                           forceMergeOpRead=true>;
     defm "" : VPseudoBinaryV_VX_RM<m>,
@@ -2797,10 +2800,10 @@ multiclass VPseudoVSMUL_VV_VX_RM {
   }
 }
 
-multiclass VPseudoVAALU_VV_VX_RM {
+multiclass VPseudoVAALU_VV_VX_RM<bit Commutable = 0> {
   foreach m = MxList in {
     defvar mx = m.MX;
-    defm "" : VPseudoBinaryV_VV_RM<m>,
+    defm "" : VPseudoBinaryV_VV_RM<m, Commutable=Commutable>,
               SchedBinary<"WriteVAALUV", "ReadVAALUV", "ReadVAALUV", mx,
                           forceMergeOpRead=true>;
     defm "" : VPseudoBinaryV_VX_RM<m>,
@@ -3593,7 +3596,7 @@ multiclass VPseudoConversion<VReg RetClass,
                              int sew = 0,
                              int TargetConstraintType = 1> {
   defvar suffix = !if(sew, "_" # MInfo.MX # "_E" # sew, "_" # MInfo.MX);
-  let VLMul = MInfo.value in {
+  let VLMul = MInfo.value, SEW=sew in {
     def suffix : VPseudoUnaryNoMask<RetClass, Op1Class, Constraint, TargetConstraintType>;
     def suffix # "_MASK" : VPseudoUnaryMask<RetClass, Op1Class,
                                             Constraint, TargetConstraintType>,
@@ -3607,7 +3610,7 @@ multiclass VPseudoConversionRoundingMode<VReg RetClass,
                              string Constraint = "",
                              int sew = 0,
                              int TargetConstraintType = 1> {
-  let VLMul = MInfo.value in {
+  let VLMul = MInfo.value, SEW=sew in {
     defvar suffix = !if(sew, "_" # MInfo.MX # "_E" # sew, "_" # MInfo.MX);
     def suffix : VPseudoUnaryNoMaskRoundingMode<RetClass, Op1Class, Constraint, TargetConstraintType>;
     def suffix # "_MASK" : VPseudoUnaryMaskRoundingMode<RetClass, Op1Class,
@@ -3624,7 +3627,7 @@ multiclass VPseudoConversionRM<VReg RetClass,
                                string Constraint = "",
                                int sew = 0,
                                int TargetConstraintType = 1> {
-  let VLMul = MInfo.value in {
+  let VLMul = MInfo.value, SEW=sew in {
     defvar suffix = !if(sew, "_" # MInfo.MX # "_E" # sew, "_" # MInfo.MX);
     def suffix : VPseudoUnaryNoMask_FRM<RetClass, Op1Class,
                                         Constraint, TargetConstraintType>;
@@ -6448,8 +6451,8 @@ defm PseudoVMV_V : VPseudoUnaryVMV_V_X_I;
 // 12.1. Vector Single-Width Saturating Add and Subtract
 //===----------------------------------------------------------------------===//
 let Defs = [VXSAT], hasSideEffects = 1 in {
-  defm PseudoVSADDU : VPseudoVSALU_VV_VX_VI;
-  defm PseudoVSADD  : VPseudoVSALU_VV_VX_VI;
+  defm PseudoVSADDU : VPseudoVSALU_VV_VX_VI<Commutable=1>;
+  defm PseudoVSADD  : VPseudoVSALU_VV_VX_VI<Commutable=1>;
   defm PseudoVSSUBU : VPseudoVSALU_VV_VX;
   defm PseudoVSSUB  : VPseudoVSALU_VV_VX;
 }
@@ -6457,8 +6460,8 @@ let Defs = [VXSAT], hasSideEffects = 1 in {
 //===----------------------------------------------------------------------===//
 // 12.2. Vector Single-Width Averaging Add and Subtract
 //===----------------------------------------------------------------------===//
-defm PseudoVAADDU : VPseudoVAALU_VV_VX_RM;
-defm PseudoVAADD  : VPseudoVAALU_VV_VX_RM;
+defm PseudoVAADDU : VPseudoVAALU_VV_VX_RM<Commutable=1>;
+defm PseudoVAADD  : VPseudoVAALU_VV_VX_RM<Commutable=1>;
 defm PseudoVASUBU : VPseudoVAALU_VV_VX_RM;
 defm PseudoVASUB  : VPseudoVAALU_VV_VX_RM;
 
diff --git a/llvm/lib/Target/RISCV/RISCVInstrInfoXTHead.td b/llvm/lib/Target/RISCV/RISCVInstrInfoXTHead.td
index 79ced3864363..b398c5e7fec2 100644
--- a/llvm/lib/Target/RISCV/RISCVInstrInfoXTHead.td
+++ b/llvm/lib/Target/RISCV/RISCVInstrInfoXTHead.td
@@ -538,6 +538,8 @@ multiclass VPatTernaryVMAQA_VV_VX<string intrinsic, string instruction,
 let Predicates = [HasVendorXTHeadBa] in {
 def : Pat<(add (XLenVT GPR:$rs1), (shl GPR:$rs2, uimm2:$uimm2)),
           (TH_ADDSL GPR:$rs1, GPR:$rs2, uimm2:$uimm2)>;
+def : Pat<(XLenVT (riscv_shl_add GPR:$rs1, uimm2:$uimm2, GPR:$rs2)),
+          (TH_ADDSL GPR:$rs2, GPR:$rs1, uimm2:$uimm2)>;
 
 // Reuse complex patterns from StdExtZba
 def : Pat<(add_non_imm12 sh1add_op:$rs1, (XLenVT GPR:$rs2)),
@@ -581,30 +583,6 @@ def : Pat<(mul (XLenVT GPR:$r), C9LeftShift:$i),
           (SLLI (XLenVT (TH_ADDSL GPR:$r, GPR:$r, 3)),
                 (TrailingZeros C9LeftShift:$i))>;
 
-def : Pat<(mul_const_oneuse GPR:$r, (XLenVT 11)),
-          (TH_ADDSL GPR:$r, (XLenVT (TH_ADDSL GPR:$r, GPR:$r, 2)), 1)>;
-def : Pat<(mul_const_oneuse GPR:$r, (XLenVT 19)),
-          (TH_ADDSL GPR:$r, (XLenVT (TH_ADDSL GPR:$r, GPR:$r, 3)), 1)>;
-def : Pat<(mul_const_oneuse GPR:$r, (XLenVT 13)),
-          (TH_ADDSL GPR:$r, (XLenVT (TH_ADDSL GPR:$r, GPR:$r, 1)), 2)>;
-def : Pat<(mul_const_oneuse GPR:$r, (XLenVT 21)),
-          (TH_ADDSL GPR:$r, (XLenVT (TH_ADDSL GPR:$r, GPR:$r, 2)), 2)>;
-def : Pat<(mul_const_oneuse GPR:$r, (XLenVT 37)),
-          (TH_ADDSL GPR:$r, (XLenVT (TH_ADDSL GPR:$r, GPR:$r, 3)), 2)>;
-def : Pat<(mul_const_oneuse GPR:$r, (XLenVT 25)),
-          (TH_ADDSL (XLenVT (TH_ADDSL GPR:$r, GPR:$r, 2)),
-                    (XLenVT (TH_ADDSL GPR:$r, GPR:$r, 2)), 2)>;
-def : Pat<(mul_const_oneuse GPR:$r, (XLenVT 41)),
-          (TH_ADDSL GPR:$r, (XLenVT (TH_ADDSL GPR:$r, GPR:$r, 2)), 3)>;
-def : Pat<(mul_const_oneuse GPR:$r, (XLenVT 73)),
-          (TH_ADDSL GPR:$r, (XLenVT (TH_ADDSL GPR:$r, GPR:$r, 3)), 3)>;
-def : Pat<(mul_const_oneuse GPR:$r, (XLenVT 27)),
-          (TH_ADDSL (XLenVT (TH_ADDSL GPR:$r, GPR:$r, 3)), (XLenVT (TH_ADDSL GPR:$r, GPR:$r, 3)), 1)>;
-def : Pat<(mul_const_oneuse GPR:$r, (XLenVT 45)),
-          (TH_ADDSL (XLenVT (TH_ADDSL GPR:$r, GPR:$r, 3)), (XLenVT (TH_ADDSL GPR:$r, GPR:$r, 3)), 2)>;
-def : Pat<(mul_const_oneuse GPR:$r, (XLenVT 81)),
-          (TH_ADDSL (XLenVT (TH_ADDSL GPR:$r, GPR:$r, 3)), (XLenVT (TH_ADDSL GPR:$r, GPR:$r, 3)), 3)>;
-
 def : Pat<(mul_const_oneuse GPR:$r, (XLenVT 200)),
           (SLLI (XLenVT (TH_ADDSL (XLenVT (TH_ADDSL GPR:$r, GPR:$r, 2)),
                                   (XLenVT (TH_ADDSL GPR:$r, GPR:$r, 2)), 2)), 3)>;
diff --git a/llvm/lib/Target/RISCV/RISCVInstrInfoZb.td b/llvm/lib/Target/RISCV/RISCVInstrInfoZb.td
index 434b071e628a..ffe2b7e27120 100644
--- a/llvm/lib/Target/RISCV/RISCVInstrInfoZb.td
+++ b/llvm/lib/Target/RISCV/RISCVInstrInfoZb.td
@@ -26,18 +26,24 @@
 // Operand and SDNode transformation definitions.
 //===----------------------------------------------------------------------===//
 
-def riscv_clzw   : SDNode<"RISCVISD::CLZW",   SDT_RISCVIntUnaryOpW>;
-def riscv_ctzw   : SDNode<"RISCVISD::CTZW",   SDT_RISCVIntUnaryOpW>;
-def riscv_rolw   : SDNode<"RISCVISD::ROLW",   SDT_RISCVIntBinOpW>;
-def riscv_rorw   : SDNode<"RISCVISD::RORW",   SDT_RISCVIntBinOpW>;
-def riscv_brev8  : SDNode<"RISCVISD::BREV8",  SDTIntUnaryOp>;
-def riscv_orc_b  : SDNode<"RISCVISD::ORC_B",  SDTIntUnaryOp>;
-def riscv_zip    : SDNode<"RISCVISD::ZIP",    SDTIntUnaryOp>;
-def riscv_unzip  : SDNode<"RISCVISD::UNZIP",  SDTIntUnaryOp>;
-def riscv_absw   : SDNode<"RISCVISD::ABSW",   SDTIntUnaryOp>;
-def riscv_clmul  : SDNode<"RISCVISD::CLMUL",  SDTIntBinOp>;
-def riscv_clmulh : SDNode<"RISCVISD::CLMULH", SDTIntBinOp>;
-def riscv_clmulr : SDNode<"RISCVISD::CLMULR", SDTIntBinOp>;
+def SDTIntShiftAddOp : SDTypeProfile<1, 3, [   // shl_add
+  SDTCisSameAs<0, 1>, SDTCisSameAs<0, 3>, SDTCisInt<0>, SDTCisInt<2>,
+  SDTCisInt<3>
+]>;
+
+def riscv_shl_add : SDNode<"RISCVISD::SHL_ADD", SDTIntShiftAddOp>;
+def riscv_clzw    : SDNode<"RISCVISD::CLZW",    SDT_RISCVIntUnaryOpW>;
+def riscv_ctzw    : SDNode<"RISCVISD::CTZW",    SDT_RISCVIntUnaryOpW>;
+def riscv_rolw    : SDNode<"RISCVISD::ROLW",    SDT_RISCVIntBinOpW>;
+def riscv_rorw    : SDNode<"RISCVISD::RORW",    SDT_RISCVIntBinOpW>;
+def riscv_brev8   : SDNode<"RISCVISD::BREV8",   SDTIntUnaryOp>;
+def riscv_orc_b   : SDNode<"RISCVISD::ORC_B",   SDTIntUnaryOp>;
+def riscv_zip     : SDNode<"RISCVISD::ZIP",     SDTIntUnaryOp>;
+def riscv_unzip   : SDNode<"RISCVISD::UNZIP",   SDTIntUnaryOp>;
+def riscv_absw    : SDNode<"RISCVISD::ABSW",    SDTIntUnaryOp>;
+def riscv_clmul   : SDNode<"RISCVISD::CLMUL",   SDTIntBinOp>;
+def riscv_clmulh  : SDNode<"RISCVISD::CLMULH",  SDTIntBinOp>;
+def riscv_clmulr  : SDNode<"RISCVISD::CLMULR",  SDTIntBinOp>;
 
 def UImmLog2XLenHalfAsmOperand : AsmOperandClass {
   let Name = "UImmLog2XLenHalf";
@@ -678,6 +684,8 @@ foreach i = {1,2,3} in {
   defvar shxadd = !cast<Instruction>("SH"#i#"ADD");
   def : Pat<(XLenVT (add_like_non_imm12 (shl GPR:$rs1, (XLenVT i)), GPR:$rs2)),
             (shxadd GPR:$rs1, GPR:$rs2)>;
+  def : Pat<(XLenVT (riscv_shl_add GPR:$rs1, (XLenVT i), GPR:$rs2)),
+            (shxadd GPR:$rs1, GPR:$rs2)>;
 
   defvar pat = !cast<ComplexPattern>("sh"#i#"add_op");
   // More complex cases use a ComplexPattern.
@@ -721,31 +729,6 @@ def : Pat<(mul (XLenVT GPR:$r), C9LeftShift:$i),
           (SLLI (XLenVT (SH3ADD GPR:$r, GPR:$r)),
                 (TrailingZeros C9LeftShift:$i))>;
 
-def : Pat<(mul_const_oneuse GPR:$r, (XLenVT 11)),
-          (SH1ADD (XLenVT (SH2ADD GPR:$r, GPR:$r)), GPR:$r)>;
-def : Pat<(mul_const_oneuse GPR:$r, (XLenVT 19)),
-          (SH1ADD (XLenVT (SH3ADD GPR:$r, GPR:$r)), GPR:$r)>;
-def : Pat<(mul_const_oneuse GPR:$r, (XLenVT 13)),
-          (SH2ADD (XLenVT (SH1ADD GPR:$r, GPR:$r)), GPR:$r)>;
-def : Pat<(mul_const_oneuse GPR:$r, (XLenVT 21)),
-          (SH2ADD (XLenVT (SH2ADD GPR:$r, GPR:$r)), GPR:$r)>;
-def : Pat<(mul_const_oneuse GPR:$r, (XLenVT 37)),
-          (SH2ADD (XLenVT (SH3ADD GPR:$r, GPR:$r)), GPR:$r)>;
-def : Pat<(mul_const_oneuse GPR:$r, (XLenVT 25)),
-          (SH3ADD (XLenVT (SH1ADD GPR:$r, GPR:$r)), GPR:$r)>;
-def : Pat<(mul_const_oneuse GPR:$r, (XLenVT 41)),
-          (SH3ADD (XLenVT (SH2ADD GPR:$r, GPR:$r)), GPR:$r)>;
-def : Pat<(mul_const_oneuse GPR:$r, (XLenVT 73)),
-          (SH3ADD (XLenVT (SH3ADD GPR:$r, GPR:$r)), GPR:$r)>;
-def : Pat<(mul_const_oneuse GPR:$r, (XLenVT 27)),
-          (SH1ADD (XLenVT (SH3ADD GPR:$r, GPR:$r)),
-                  (XLenVT (SH3ADD GPR:$r, GPR:$r)))>;
-def : Pat<(mul_const_oneuse GPR:$r, (XLenVT 45)),
-          (SH2ADD (XLenVT (SH3ADD GPR:$r, GPR:$r)),
-                  (XLenVT (SH3ADD GPR:$r, GPR:$r)))>;
-def : Pat<(mul_const_oneuse GPR:$r, (XLenVT 81)),
-          (SH3ADD (XLenVT (SH3ADD GPR:$r, GPR:$r)),
-                  (XLenVT (SH3ADD GPR:$r, GPR:$r)))>;
 } // Predicates = [HasStdExtZba]
 
 let Predicates = [HasStdExtZba, IsRV64] in {
@@ -764,6 +747,8 @@ foreach i = {1,2,3} in {
   defvar shxadd_uw = !cast<Instruction>("SH"#i#"ADD_UW");
   def : Pat<(i64 (add_like_non_imm12 (shl (and GPR:$rs1, 0xFFFFFFFF), (i64 i)), (XLenVT GPR:$rs2))),
             (shxadd_uw GPR:$rs1, GPR:$rs2)>;
+  def : Pat<(i64 (riscv_shl_add (and GPR:$rs1, 0xFFFFFFFF), (i64 i), GPR:$rs2)),
+            (shxadd_uw GPR:$rs1, GPR:$rs2)>;
 }
 
 def : Pat<(i64 (add_like_non_imm12 (and (shl GPR:$rs1, (i64 1)), 0x1FFFFFFFF), (XLenVT GPR:$rs2))),
@@ -881,6 +866,8 @@ foreach i = {1,2,3} in {
   defvar shxadd = !cast<Instruction>("SH"#i#"ADD");
   def : Pat<(i32 (add_like_non_imm12 (shl GPR:$rs1, (i64 i)), GPR:$rs2)),
             (shxadd GPR:$rs1, GPR:$rs2)>;
+  def : Pat<(i32 (riscv_shl_add GPR:$rs1, (i32 i), GPR:$rs2)),
+            (shxadd GPR:$rs1, GPR:$rs2)>;
 }
 }
 
diff --git a/llvm/lib/Target/RISCV/RISCVInstrInfoZvk.td b/llvm/lib/Target/RISCV/RISCVInstrInfoZvk.td
index aac7dc444a2d..aaf9c019aedf 100644
--- a/llvm/lib/Target/RISCV/RISCVInstrInfoZvk.td
+++ b/llvm/lib/Target/RISCV/RISCVInstrInfoZvk.td
@@ -1,4 +1,4 @@
-//===-- RISCVInstrInfoZvk.td - RISC-V 'Zvk' instructions -------*- tablegen -*-===//
+//===-- RISCVInstrInfoZvk.td - RISC-V 'Zvk' instructions ---*- tablegen -*-===//
 //
 // Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
 // See https://llvm.org/LICENSE.txt for license information.
diff --git a/llvm/lib/Target/RISCV/RISCVOptWInstrs.cpp b/llvm/lib/Target/RISCV/RISCVOptWInstrs.cpp
index ead91c5656be..788d8f9cfc85 100644
--- a/llvm/lib/Target/RISCV/RISCVOptWInstrs.cpp
+++ b/llvm/lib/Target/RISCV/RISCVOptWInstrs.cpp
@@ -420,7 +420,7 @@ static bool isSignExtendedW(Register SrcReg, const RISCVSubtarget &ST,
     if (!MI)
       continue;
 
-    int OpNo = MI->findRegisterDefOperandIdx(Reg);
+    int OpNo = MI->findRegisterDefOperandIdx(Reg, /*TRI=*/nullptr);
     assert(OpNo != -1 && "Couldn't find register");
 
     // If this is a sign extending operation we don't need to look any further.
diff --git a/llvm/lib/Target/RISCV/RISCVProfiles.td b/llvm/lib/Target/RISCV/RISCVProfiles.td
new file mode 100644
index 000000000000..5c13710faf65
--- /dev/null
+++ b/llvm/lib/Target/RISCV/RISCVProfiles.td
@@ -0,0 +1,204 @@
+//===------ RISCVProfiles.td - RISC-V Profiles -------------*- tablegen -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+class RISCVProfile<string name, list<SubtargetFeature> features>
+    : SubtargetFeature<name, "Is" # NAME, "true",
+                       "RISC-V " # name # " profile", features>;
+
+defvar RVI20U32Features = [Feature32Bit, FeatureStdExtI];
+defvar RVI20U64Features = [Feature64Bit, FeatureStdExtI];
+
+defvar RVA20U64Features = [Feature64Bit,
+                           FeatureStdExtI,
+                           FeatureStdExtM,
+                           FeatureStdExtA,
+                           FeatureStdExtF,
+                           FeatureStdExtD,
+                           FeatureStdExtC,
+                           FeatureStdExtZicntr,
+                           FeatureStdExtZiccif,
+                           FeatureStdExtZiccrse,
+                           FeatureStdExtZiccamoa,
+                           FeatureStdExtZa128rs,
+                           FeatureStdExtZicclsm];
+
+defvar RVA20S64Features = !listconcat(RVA20U64Features,
+                                      [FeatureStdExtZifencei,
+                                       FeatureStdExtSvbare,
+                                       FeatureStdExtSvade,
+                                       FeatureStdExtSsccptr,
+                                       FeatureStdExtSstvecd,
+                                       FeatureStdExtSstvala]);
+
+defvar RVA22U64Features = [Feature64Bit,
+                           FeatureStdExtI,
+                           FeatureStdExtM,
+                           FeatureStdExtA,
+                           FeatureStdExtF,
+                           FeatureStdExtD,
+                           FeatureStdExtC,
+                           FeatureStdExtZicntr,
+                           FeatureStdExtZiccif,
+                           FeatureStdExtZiccrse,
+                           FeatureStdExtZiccamoa,
+                           FeatureStdExtZicclsm,
+                           FeatureStdExtZa64rs,
+                           FeatureStdExtZihpm,
+                           FeatureStdExtZihintpause,
+                           FeatureStdExtZba,
+                           FeatureStdExtZbb,
+                           FeatureStdExtZbs,
+                           FeatureStdExtZic64b,
+                           FeatureStdExtZicbom,
+                           FeatureStdExtZicbop,
+                           FeatureStdExtZicboz,
+                           FeatureStdExtZfhmin,
+                           FeatureStdExtZkt];
+
+defvar RVA22S64Features = !listconcat(RVA22U64Features,
+                                      [FeatureStdExtZifencei,
+                                       FeatureStdExtSvbare,
+                                       FeatureStdExtSvade,
+                                       FeatureStdExtSsccptr,
+                                       FeatureStdExtSstvecd,
+                                       FeatureStdExtSstvala,
+                                       FeatureStdExtSscounterenw,
+                                       FeatureStdExtSvpbmt,
+                                       FeatureStdExtSvinval]);
+
+defvar RVA23U64Features = [Feature64Bit,
+                           FeatureStdExtI,
+                           FeatureStdExtM,
+                           FeatureStdExtA,
+                           FeatureStdExtF,
+                           FeatureStdExtD,
+                           FeatureStdExtC,
+                           FeatureStdExtZicntr,
+                           FeatureStdExtZihpm,
+                           FeatureStdExtZiccif,
+                           FeatureStdExtZiccrse,
+                           FeatureStdExtZiccamoa,
+                           FeatureStdExtZicclsm,
+                           FeatureStdExtZa64rs,
+                           FeatureStdExtZihintpause,
+                           FeatureStdExtZba,
+                           FeatureStdExtZbb,
+                           FeatureStdExtZbs,
+                           FeatureStdExtZic64b,
+                           FeatureStdExtZicbom,
+                           FeatureStdExtZicbop,
+                           FeatureStdExtZicboz,
+                           FeatureStdExtZfhmin,
+                           FeatureStdExtZkt,
+                           FeatureStdExtV,
+                           FeatureStdExtZvfhmin,
+                           FeatureStdExtZvbb,
+                           FeatureStdExtZvkt,
+                           FeatureStdExtZihintntl,
+                           FeatureStdExtZicond,
+                           FeatureStdExtZimop,
+                           FeatureStdExtZcmop,
+                           FeatureStdExtZcb,
+                           FeatureStdExtZfa,
+                           FeatureStdExtZawrs];
+
+defvar RVA23S64Features = !listconcat(RVA23U64Features,
+                                      [FeatureStdExtZifencei,
+                                       FeatureStdExtSvbare,
+                                       FeatureStdExtSvade,
+                                       FeatureStdExtSsccptr,
+                                       FeatureStdExtSstvecd,
+                                       FeatureStdExtSstvala,
+                                       FeatureStdExtSscounterenw,
+                                       FeatureStdExtSvpbmt,
+                                       FeatureStdExtSvinval,
+                                       FeatureStdExtSvnapot,
+                                       FeatureStdExtSstc,
+                                       FeatureStdExtSscofpmf,
+                                       FeatureStdExtSsnpm,
+                                       FeatureStdExtSsu64xl,
+                                       FeatureStdExtH,
+                                       FeatureStdExtSsstateen,
+                                       FeatureStdExtShcounterenw,
+                                       FeatureStdExtShvstvala,
+                                       FeatureStdExtShtvala,
+                                       FeatureStdExtShvstvecd,
+                                       FeatureStdExtShvsatpa,
+                                       FeatureStdExtShgatpa]);
+
+defvar RVB23U64Features = [Feature64Bit,
+                           FeatureStdExtI,
+                           FeatureStdExtM,
+                           FeatureStdExtA,
+                           FeatureStdExtF,
+                           FeatureStdExtD,
+                           FeatureStdExtC,
+                           FeatureStdExtZicntr,
+                           FeatureStdExtZihpm,
+                           FeatureStdExtZiccif,
+                           FeatureStdExtZiccrse,
+                           FeatureStdExtZiccamoa,
+                           FeatureStdExtZicclsm,
+                           FeatureStdExtZa64rs,
+                           FeatureStdExtZihintpause,
+                           FeatureStdExtZba,
+                           FeatureStdExtZbb,
+                           FeatureStdExtZbs,
+                           FeatureStdExtZic64b,
+                           FeatureStdExtZicbom,
+                           FeatureStdExtZicbop,
+                           FeatureStdExtZicboz,
+                           FeatureStdExtZkt,
+                           FeatureStdExtZihintntl,
+                           FeatureStdExtZicond,
+                           FeatureStdExtZimop,
+                           FeatureStdExtZcmop,
+                           FeatureStdExtZcb,
+                           FeatureStdExtZfa,
+                           FeatureStdExtZawrs];
+
+defvar RVB23S64Features = !listconcat(RVB23U64Features,
+                                      [FeatureStdExtZifencei,
+                                       FeatureStdExtSvnapot,
+                                       FeatureStdExtSvbare,
+                                       FeatureStdExtSvade,
+                                       FeatureStdExtSsccptr,
+                                       FeatureStdExtSstvecd,
+                                       FeatureStdExtSstvala,
+                                       FeatureStdExtSscounterenw,
+                                       FeatureStdExtSvpbmt,
+                                       FeatureStdExtSvinval,
+                                       FeatureStdExtSstc,
+                                       FeatureStdExtSscofpmf,
+                                       FeatureStdExtSsu64xl]);
+
+defvar RVM23U32Features = [Feature32Bit,
+                           FeatureStdExtI,
+                           FeatureStdExtM,
+                           FeatureStdExtZba,
+                           FeatureStdExtZbb,
+                           FeatureStdExtZbs,
+                           FeatureStdExtZicond,
+                           FeatureStdExtZihintpause,
+                           FeatureStdExtZihintntl,
+                           FeatureStdExtZce,
+                           FeatureStdExtZicbop,
+                           FeatureStdExtZimop,
+                           FeatureStdExtZcmop];
+
+def RVI20U32 : RISCVProfile<"rvi20u32", RVI20U32Features>;
+def RVI20U64 : RISCVProfile<"rvi20u64", RVI20U64Features>;
+def RVA20U64 : RISCVProfile<"rva20u64", RVA20U64Features>;
+def RVA20S64 : RISCVProfile<"rva20s64", RVA20S64Features>;
+def RVA22U64 : RISCVProfile<"rva22u64", RVA22U64Features>;
+def RVA22S64 : RISCVProfile<"rva22s64", RVA22S64Features>;
+def RVA23U64 : RISCVProfile<"rva23u64", RVA23U64Features>;
+def RVA23S64 : RISCVProfile<"rva23s64", RVA23S64Features>;
+def RVB23U64 : RISCVProfile<"rvb23u64", RVB23U64Features>;
+def RVB23S64 : RISCVProfile<"rvb23s64", RVB23S64Features>;
+def RVM23U32 : RISCVProfile<"rvm23u32", RVM23U32Features>;
diff --git a/llvm/lib/Target/RISCV/RISCVTargetMachine.cpp b/llvm/lib/Target/RISCV/RISCVTargetMachine.cpp
index ae1a6f179a49..0876f46728a1 100644
--- a/llvm/lib/Target/RISCV/RISCVTargetMachine.cpp
+++ b/llvm/lib/Target/RISCV/RISCVTargetMachine.cpp
@@ -91,11 +91,6 @@ static cl::opt<bool>
                            cl::desc("Enable the loop data prefetch pass"),
                            cl::init(true));
 
-static cl::opt<bool>
-    EnableSplitRegAlloc("riscv-split-regalloc", cl::Hidden,
-                        cl::desc("Enable Split RegisterAlloc for RVV"),
-                        cl::init(true));
-
 static cl::opt<bool> EnableMISchedLoadClustering(
     "riscv-misched-load-clustering", cl::Hidden,
     cl::desc("Enable load clustering in the machine scheduler"),
@@ -121,6 +116,7 @@ extern "C" LLVM_EXTERNAL_VISIBILITY void LLVMInitializeRISCVTarget() {
   initializeRISCVExpandPseudoPass(*PR);
   initializeRISCVFoldMasksPass(*PR);
   initializeRISCVInsertVSETVLIPass(*PR);
+  initializeRISCVCoalesceVSETVLIPass(*PR);
   initializeRISCVInsertReadWriteCSRPass(*PR);
   initializeRISCVInsertWriteVXRMPass(*PR);
   initializeRISCVDAGToDAGISelPass(*PR);
@@ -392,16 +388,15 @@ FunctionPass *RISCVPassConfig::createRVVRegAllocPass(bool Optimized) {
 }
 
 bool RISCVPassConfig::addRegAssignAndRewriteFast() {
-  if (EnableSplitRegAlloc)
-    addPass(createRVVRegAllocPass(false));
+  addPass(createRVVRegAllocPass(false));
+  addPass(createRISCVCoalesceVSETVLIPass());
   return TargetPassConfig::addRegAssignAndRewriteFast();
 }
 
 bool RISCVPassConfig::addRegAssignAndRewriteOptimized() {
-  if (EnableSplitRegAlloc) {
-    addPass(createRVVRegAllocPass(true));
-    addPass(createVirtRegRewriter(false));
-  }
+  addPass(createRVVRegAllocPass(true));
+  addPass(createVirtRegRewriter(false));
+  addPass(createRISCVCoalesceVSETVLIPass());
   return TargetPassConfig::addRegAssignAndRewriteOptimized();
 }
 
diff --git a/llvm/lib/Target/RISCV/RISCVTargetTransformInfo.h b/llvm/lib/Target/RISCV/RISCVTargetTransformInfo.h
index 2f9281ab8924..a4d139087509 100644
--- a/llvm/lib/Target/RISCV/RISCVTargetTransformInfo.h
+++ b/llvm/lib/Target/RISCV/RISCVTargetTransformInfo.h
@@ -210,7 +210,7 @@ public:
       unsigned Opcode, Type *Ty, TTI::TargetCostKind CostKind,
       TTI::OperandValueInfo Op1Info = {TTI::OK_AnyValue, TTI::OP_None},
       TTI::OperandValueInfo Op2Info = {TTI::OK_AnyValue, TTI::OP_None},
-      ArrayRef<const Value *> Args = ArrayRef<const Value *>(),
+      ArrayRef<const Value *> Args = std::nullopt,
       const Instruction *CxtI = nullptr);
 
   bool isElementTypeLegalForScalableVector(Type *Ty) const {
diff --git a/llvm/lib/Target/SPIRV/MCTargetDesc/SPIRVMCTargetDesc.cpp b/llvm/lib/Target/SPIRV/MCTargetDesc/SPIRVMCTargetDesc.cpp
index 78dfbf4ec932..74ebaa9d0c00 100644
--- a/llvm/lib/Target/SPIRV/MCTargetDesc/SPIRVMCTargetDesc.cpp
+++ b/llvm/lib/Target/SPIRV/MCTargetDesc/SPIRVMCTargetDesc.cpp
@@ -53,9 +53,9 @@ static MCStreamer *
 createSPIRVMCStreamer(const Triple &T, MCContext &Ctx,
                       std::unique_ptr<MCAsmBackend> &&MAB,
                       std::unique_ptr<MCObjectWriter> &&OW,
-                      std::unique_ptr<MCCodeEmitter> &&Emitter, bool RelaxAll) {
+                      std::unique_ptr<MCCodeEmitter> &&Emitter) {
   return createSPIRVStreamer(Ctx, std::move(MAB), std::move(OW),
-                             std::move(Emitter), RelaxAll);
+                             std::move(Emitter));
 }
 
 static MCTargetStreamer *createTargetAsmStreamer(MCStreamer &S,
diff --git a/llvm/lib/Target/SPIRV/SPIRVBuiltins.cpp b/llvm/lib/Target/SPIRV/SPIRVBuiltins.cpp
index 4b07d7e61fa1..7439d0fefa98 100644
--- a/llvm/lib/Target/SPIRV/SPIRVBuiltins.cpp
+++ b/llvm/lib/Target/SPIRV/SPIRVBuiltins.cpp
@@ -189,6 +189,10 @@ lookupBuiltin(StringRef DemangledCall,
   std::string BuiltinName =
       DemangledCall.substr(0, DemangledCall.find('(')).str();
 
+  // Account for possible "__spirv_ocl_" prefix in SPIR-V friendly LLVM IR
+  if (BuiltinName.rfind("__spirv_ocl_", 0) == 0)
+    BuiltinName = BuiltinName.substr(12);
+
   // Check if the extracted name contains type information between angle
   // brackets. If so, the builtin is an instantiated template - needs to have
   // the information after angle brackets and return type removed.
@@ -2008,6 +2012,13 @@ static bool generateAsyncCopy(const SPIRV::IncomingCall *Call,
   const SPIRV::DemangledBuiltin *Builtin = Call->Builtin;
   unsigned Opcode =
       SPIRV::lookupNativeBuiltin(Builtin->Name, Builtin->Set)->Opcode;
+
+  bool IsSet = Opcode == SPIRV::OpGroupAsyncCopy;
+  Register TypeReg = GR->getSPIRVTypeID(Call->ReturnType);
+  if (Call->isSpirvOp())
+    return buildOpFromWrapper(MIRBuilder, Opcode, Call,
+                              IsSet ? TypeReg : Register(0));
+
   auto Scope = buildConstantIntReg(SPIRV::Scope::Workgroup, MIRBuilder, GR);
 
   switch (Opcode) {
@@ -2306,7 +2317,7 @@ Type *parseBuiltinCallArgumentBaseType(const StringRef DemangledCall,
     // parseBuiltinCallArgumentBaseType(...) as this function only retrieves the
     // base types.
     if (TypeStr.ends_with("*"))
-      TypeStr = TypeStr.slice(0, TypeStr.find_first_of(" "));
+      TypeStr = TypeStr.slice(0, TypeStr.find_first_of(" *"));
 
     return parseBuiltinTypeNameToTargetExtType("opencl." + TypeStr.str() + "_t",
                                                Ctx);
diff --git a/llvm/lib/Target/SPIRV/SPIRVBuiltins.td b/llvm/lib/Target/SPIRV/SPIRVBuiltins.td
index 660000fb548d..564028547821 100644
--- a/llvm/lib/Target/SPIRV/SPIRVBuiltins.td
+++ b/llvm/lib/Target/SPIRV/SPIRVBuiltins.td
@@ -585,9 +585,9 @@ defm : DemangledNativeBuiltin<"__spirv_SpecConstantComposite", OpenCL_std, SpecC
 
 // Async Copy and Prefetch builtin records:
 defm : DemangledNativeBuiltin<"async_work_group_copy", OpenCL_std, AsyncCopy, 4, 4, OpGroupAsyncCopy>;
-defm : DemangledNativeBuiltin<"__spirv_GroupAsyncCopy", OpenCL_std, AsyncCopy, 4, 4, OpGroupAsyncCopy>;
+defm : DemangledNativeBuiltin<"__spirv_GroupAsyncCopy", OpenCL_std, AsyncCopy, 6, 6, OpGroupAsyncCopy>;
 defm : DemangledNativeBuiltin<"wait_group_events", OpenCL_std, AsyncCopy, 2, 2, OpGroupWaitEvents>;
-defm : DemangledNativeBuiltin<"__spirv_GroupWaitEvents", OpenCL_std, AsyncCopy, 2, 2, OpGroupWaitEvents>;
+defm : DemangledNativeBuiltin<"__spirv_GroupWaitEvents", OpenCL_std, AsyncCopy, 3, 3, OpGroupWaitEvents>;
 
 // Load and store builtin records:
 defm : DemangledNativeBuiltin<"__spirv_Load", OpenCL_std, LoadStore, 1, 3, OpLoad>;
diff --git a/llvm/lib/Target/SPIRV/SPIRVEmitIntrinsics.cpp b/llvm/lib/Target/SPIRV/SPIRVEmitIntrinsics.cpp
index 472bc8638c9a..0d539b1ed9a8 100644
--- a/llvm/lib/Target/SPIRV/SPIRVEmitIntrinsics.cpp
+++ b/llvm/lib/Target/SPIRV/SPIRVEmitIntrinsics.cpp
@@ -98,6 +98,8 @@ class SPIRVEmitIntrinsics
     return B.CreateIntrinsic(IntrID, {Types}, Args);
   }
 
+  void buildAssignPtr(IRBuilder<> &B, Type *ElemTy, Value *Arg);
+
   void replaceMemInstrUses(Instruction *Old, Instruction *New, IRBuilder<> &B);
   void processInstrAfterVisit(Instruction *I, IRBuilder<> &B);
   void insertAssignPtrTypeIntrs(Instruction *I, IRBuilder<> &B);
@@ -111,6 +113,7 @@ class SPIRVEmitIntrinsics
   void insertPtrCastOrAssignTypeInstr(Instruction *I, IRBuilder<> &B);
   void processGlobalValue(GlobalVariable &GV, IRBuilder<> &B);
   void processParamTypes(Function *F, IRBuilder<> &B);
+  void processParamTypesByFunHeader(Function *F, IRBuilder<> &B);
   Type *deduceFunParamElementType(Function *F, unsigned OpIdx);
   Type *deduceFunParamElementType(Function *F, unsigned OpIdx,
                                   std::unordered_set<Function *> &FVisited);
@@ -194,6 +197,17 @@ static inline void reportFatalOnTokenType(const Instruction *I) {
                        false);
 }
 
+void SPIRVEmitIntrinsics::buildAssignPtr(IRBuilder<> &B, Type *ElemTy,
+                                         Value *Arg) {
+  CallInst *AssignPtrTyCI =
+      buildIntrWithMD(Intrinsic::spv_assign_ptr_type, {Arg->getType()},
+                      Constant::getNullValue(ElemTy), Arg,
+                      {B.getInt32(getPointerAddressSpace(Arg->getType()))}, B);
+  GR->addDeducedElementType(AssignPtrTyCI, ElemTy);
+  GR->addDeducedElementType(Arg, ElemTy);
+  AssignPtrTypeInstr[Arg] = AssignPtrTyCI;
+}
+
 // Set element pointer type to the given value of ValueTy and tries to
 // specify this type further (recursively) by Operand value, if needed.
 Type *SPIRVEmitIntrinsics::deduceElementTypeByValueDeep(
@@ -232,6 +246,19 @@ Type *SPIRVEmitIntrinsics::deduceElementTypeByUsersDeep(
   return nullptr;
 }
 
+// Implements what we know in advance about intrinsics and builtin calls
+// TODO: consider feasibility of this particular case to be generalized by
+// encoding knowledge about intrinsics and builtin calls by corresponding
+// specification rules
+static Type *getPointeeTypeByCallInst(StringRef DemangledName,
+                                      Function *CalledF, unsigned OpIdx) {
+  if ((DemangledName.starts_with("__spirv_ocl_printf(") ||
+       DemangledName.starts_with("printf(")) &&
+      OpIdx == 0)
+    return IntegerType::getInt8Ty(CalledF->getContext());
+  return nullptr;
+}
+
 // Deduce and return a successfully deduced Type of the Instruction,
 // or nullptr otherwise.
 Type *SPIRVEmitIntrinsics::deduceElementTypeHelper(Value *I) {
@@ -795,6 +822,8 @@ void SPIRVEmitIntrinsics::insertPtrCastOrAssignTypeInstr(Instruction *I,
     return;
 
   // collect information about formal parameter types
+  std::string DemangledName =
+      getOclOrSpirvBuiltinDemangledName(CI->getCalledFunction()->getName());
   Function *CalledF = CI->getCalledFunction();
   SmallVector<Type *, 4> CalledArgTys;
   bool HaveTypes = false;
@@ -811,10 +840,15 @@ void SPIRVEmitIntrinsics::insertPtrCastOrAssignTypeInstr(Instruction *I,
       if (!ElemTy && hasPointeeTypeAttr(CalledArg))
         ElemTy = getPointeeTypeByAttr(CalledArg);
       if (!ElemTy) {
-        for (User *U : CalledArg->users()) {
-          if (Instruction *Inst = dyn_cast<Instruction>(U)) {
-            if ((ElemTy = deduceElementTypeHelper(Inst)) != nullptr)
-              break;
+        ElemTy = getPointeeTypeByCallInst(DemangledName, CalledF, OpIdx);
+        if (ElemTy) {
+          GR->addDeducedElementType(CalledArg, ElemTy);
+        } else {
+          for (User *U : CalledArg->users()) {
+            if (Instruction *Inst = dyn_cast<Instruction>(U)) {
+              if ((ElemTy = deduceElementTypeHelper(Inst)) != nullptr)
+                break;
+            }
           }
         }
       }
@@ -823,8 +857,6 @@ void SPIRVEmitIntrinsics::insertPtrCastOrAssignTypeInstr(Instruction *I,
     }
   }
 
-  std::string DemangledName =
-      getOclOrSpirvBuiltinDemangledName(CI->getCalledFunction()->getName());
   if (DemangledName.empty() && !HaveTypes)
     return;
 
@@ -835,8 +867,14 @@ void SPIRVEmitIntrinsics::insertPtrCastOrAssignTypeInstr(Instruction *I,
       continue;
 
     // Constants (nulls/undefs) are handled in insertAssignPtrTypeIntrs()
-    if (!isa<Instruction>(ArgOperand) && !isa<Argument>(ArgOperand))
-      continue;
+    if (!isa<Instruction>(ArgOperand) && !isa<Argument>(ArgOperand)) {
+      // However, we may have assumptions about the formal argument's type and
+      // may have a need to insert a ptr cast for the actual parameter of this
+      // call.
+      Argument *CalledArg = CalledF->getArg(OpIdx);
+      if (!GR->findDeducedElementType(CalledArg))
+        continue;
+    }
 
     Type *ExpectedType =
         OpIdx < CalledArgTys.size() ? CalledArgTys[OpIdx] : nullptr;
@@ -1102,9 +1140,13 @@ void SPIRVEmitIntrinsics::processInstrAfterVisit(Instruction *I,
                  (II->paramHasAttr(OpNo, Attribute::ImmArg))))
         continue;
       B.SetInsertPoint(I);
-      auto *NewOp =
-          buildIntrWithMD(Intrinsic::spv_track_constant,
-                          {Op->getType(), Op->getType()}, Op, Op, {}, B);
+      Value *OpTyVal = Op;
+      if (Op->getType()->isTargetExtTy())
+        OpTyVal = Constant::getNullValue(
+            IntegerType::get(I->getContext(), GR->getPointerSize()));
+      auto *NewOp = buildIntrWithMD(Intrinsic::spv_track_constant,
+                                    {Op->getType(), OpTyVal->getType()}, Op,
+                                    OpTyVal, {}, B);
       I->setOperand(OpNo, NewOp);
     }
   }
@@ -1179,28 +1221,29 @@ Type *SPIRVEmitIntrinsics::deduceFunParamElementType(
   return nullptr;
 }
 
-void SPIRVEmitIntrinsics::processParamTypes(Function *F, IRBuilder<> &B) {
+void SPIRVEmitIntrinsics::processParamTypesByFunHeader(Function *F,
+                                                       IRBuilder<> &B) {
   B.SetInsertPointPastAllocas(F);
   for (unsigned OpIdx = 0; OpIdx < F->arg_size(); ++OpIdx) {
     Argument *Arg = F->getArg(OpIdx);
     if (!isUntypedPointerTy(Arg->getType()))
       continue;
+    Type *ElemTy = GR->findDeducedElementType(Arg);
+    if (!ElemTy && hasPointeeTypeAttr(Arg) &&
+        (ElemTy = getPointeeTypeByAttr(Arg)) != nullptr)
+      buildAssignPtr(B, ElemTy, Arg);
+  }
+}
 
+void SPIRVEmitIntrinsics::processParamTypes(Function *F, IRBuilder<> &B) {
+  B.SetInsertPointPastAllocas(F);
+  for (unsigned OpIdx = 0; OpIdx < F->arg_size(); ++OpIdx) {
+    Argument *Arg = F->getArg(OpIdx);
+    if (!isUntypedPointerTy(Arg->getType()))
+      continue;
     Type *ElemTy = GR->findDeducedElementType(Arg);
-    if (!ElemTy) {
-      if (hasPointeeTypeAttr(Arg) &&
-          (ElemTy = getPointeeTypeByAttr(Arg)) != nullptr) {
-        GR->addDeducedElementType(Arg, ElemTy);
-      } else if ((ElemTy = deduceFunParamElementType(F, OpIdx)) != nullptr) {
-        CallInst *AssignPtrTyCI = buildIntrWithMD(
-            Intrinsic::spv_assign_ptr_type, {Arg->getType()},
-            Constant::getNullValue(ElemTy), Arg,
-            {B.getInt32(getPointerAddressSpace(Arg->getType()))}, B);
-        GR->addDeducedElementType(AssignPtrTyCI, ElemTy);
-        GR->addDeducedElementType(Arg, ElemTy);
-        AssignPtrTypeInstr[Arg] = AssignPtrTyCI;
-      }
-    }
+    if (!ElemTy && (ElemTy = deduceFunParamElementType(F, OpIdx)) != nullptr)
+      buildAssignPtr(B, ElemTy, Arg);
   }
 }
 
@@ -1217,6 +1260,8 @@ bool SPIRVEmitIntrinsics::runOnFunction(Function &Func) {
   AggrConstTypes.clear();
   AggrStores.clear();
 
+  processParamTypesByFunHeader(F, B);
+
   // StoreInst's operand type can be changed during the next transformations,
   // so we need to store it in the set. Also store already transformed types.
   for (auto &I : instructions(Func)) {
diff --git a/llvm/lib/Target/SPIRV/SPIRVISelLowering.cpp b/llvm/lib/Target/SPIRV/SPIRVISelLowering.cpp
index b8296c3f6eea..96b4a570a26b 100644
--- a/llvm/lib/Target/SPIRV/SPIRVISelLowering.cpp
+++ b/llvm/lib/Target/SPIRV/SPIRVISelLowering.cpp
@@ -314,6 +314,16 @@ void SPIRVTargetLowering::finalizeLowering(MachineFunction &MF) const {
                                       SPIRV::OpTypeBool))
           MI.setDesc(STI.getInstrInfo()->get(SPIRV::OpLogicalNotEqual));
         break;
+      case SPIRV::OpConstantI: {
+        SPIRVType *Type = GR.getSPIRVTypeForVReg(MI.getOperand(1).getReg());
+        if (Type->getOpcode() != SPIRV::OpTypeInt && MI.getOperand(2).isImm() &&
+            MI.getOperand(2).getImm() == 0) {
+          // Validate the null constant of a target extension type
+          MI.setDesc(STI.getInstrInfo()->get(SPIRV::OpConstantNull));
+          for (unsigned i = MI.getNumOperands() - 1; i > 1; --i)
+            MI.removeOperand(i);
+        }
+      } break;
       }
     }
   }
diff --git a/llvm/lib/Target/SPIRV/SPIRVInstructionSelector.cpp b/llvm/lib/Target/SPIRV/SPIRVInstructionSelector.cpp
index 21a69fc3ad9b..9994a966c82c 100644
--- a/llvm/lib/Target/SPIRV/SPIRVInstructionSelector.cpp
+++ b/llvm/lib/Target/SPIRV/SPIRVInstructionSelector.cpp
@@ -646,6 +646,37 @@ bool SPIRVInstructionSelector::selectUnOp(Register ResVReg,
                                           const SPIRVType *ResType,
                                           MachineInstr &I,
                                           unsigned Opcode) const {
+  if (STI.isOpenCLEnv() && I.getOperand(1).isReg()) {
+    Register SrcReg = I.getOperand(1).getReg();
+    bool IsGV = false;
+    for (MachineRegisterInfo::def_instr_iterator DefIt =
+             MRI->def_instr_begin(SrcReg);
+         DefIt != MRI->def_instr_end(); DefIt = std::next(DefIt)) {
+      if ((*DefIt).getOpcode() == TargetOpcode::G_GLOBAL_VALUE) {
+        IsGV = true;
+        break;
+      }
+    }
+    if (IsGV) {
+      uint32_t SpecOpcode = 0;
+      switch (Opcode) {
+      case SPIRV::OpConvertPtrToU:
+        SpecOpcode = static_cast<uint32_t>(SPIRV::Opcode::ConvertPtrToU);
+        break;
+      case SPIRV::OpConvertUToPtr:
+        SpecOpcode = static_cast<uint32_t>(SPIRV::Opcode::ConvertUToPtr);
+        break;
+      }
+      if (SpecOpcode)
+        return BuildMI(*I.getParent(), I, I.getDebugLoc(),
+                       TII.get(SPIRV::OpSpecConstantOp))
+            .addDef(ResVReg)
+            .addUse(GR.getSPIRVTypeID(ResType))
+            .addImm(SpecOpcode)
+            .addUse(SrcReg)
+            .constrainAllUses(TII, TRI, RBI);
+    }
+  }
   return selectUnOpWithSrc(ResVReg, ResType, I, I.getOperand(1).getReg(),
                            Opcode);
 }
@@ -1587,8 +1618,18 @@ bool SPIRVInstructionSelector::selectIToF(Register ResVReg,
 bool SPIRVInstructionSelector::selectExt(Register ResVReg,
                                          const SPIRVType *ResType,
                                          MachineInstr &I, bool IsSigned) const {
-  if (GR.isScalarOrVectorOfType(I.getOperand(1).getReg(), SPIRV::OpTypeBool))
+  Register SrcReg = I.getOperand(1).getReg();
+  if (GR.isScalarOrVectorOfType(SrcReg, SPIRV::OpTypeBool))
     return selectSelect(ResVReg, ResType, I, IsSigned);
+
+  SPIRVType *SrcType = GR.getSPIRVTypeForVReg(SrcReg);
+  if (SrcType == ResType)
+    return BuildMI(*I.getParent(), I, I.getDebugLoc(),
+                   TII.get(TargetOpcode::COPY))
+        .addDef(ResVReg)
+        .addUse(SrcReg)
+        .constrainAllUses(TII, TRI, RBI);
+
   unsigned Opcode = IsSigned ? SPIRV::OpSConvert : SPIRV::OpUConvert;
   return selectUnOp(ResVReg, ResType, I, Opcode);
 }
@@ -1622,11 +1663,16 @@ bool SPIRVInstructionSelector::selectIntToBool(Register IntReg,
 bool SPIRVInstructionSelector::selectTrunc(Register ResVReg,
                                            const SPIRVType *ResType,
                                            MachineInstr &I) const {
-  if (GR.isScalarOrVectorOfType(ResVReg, SPIRV::OpTypeBool)) {
-    Register IntReg = I.getOperand(1).getReg();
-    const SPIRVType *ArgType = GR.getSPIRVTypeForVReg(IntReg);
+  Register IntReg = I.getOperand(1).getReg();
+  const SPIRVType *ArgType = GR.getSPIRVTypeForVReg(IntReg);
+  if (GR.isScalarOrVectorOfType(ResVReg, SPIRV::OpTypeBool))
     return selectIntToBool(IntReg, ResVReg, I, ArgType, ResType);
-  }
+  if (ArgType == ResType)
+    return BuildMI(*I.getParent(), I, I.getDebugLoc(),
+                   TII.get(TargetOpcode::COPY))
+        .addDef(ResVReg)
+        .addUse(IntReg)
+        .constrainAllUses(TII, TRI, RBI);
   bool IsSigned = GR.isScalarOrVectorSigned(ResType);
   unsigned Opcode = IsSigned ? SPIRV::OpSConvert : SPIRV::OpUConvert;
   return selectUnOp(ResVReg, ResType, I, Opcode);
diff --git a/llvm/lib/Target/SPIRV/SPIRVPreLegalizer.cpp b/llvm/lib/Target/SPIRV/SPIRVPreLegalizer.cpp
index d16f6d5bf67e..84508fb5fe09 100644
--- a/llvm/lib/Target/SPIRV/SPIRVPreLegalizer.cpp
+++ b/llvm/lib/Target/SPIRV/SPIRVPreLegalizer.cpp
@@ -38,7 +38,9 @@ public:
 };
 } // namespace
 
-static void addConstantsToTrack(MachineFunction &MF, SPIRVGlobalRegistry *GR) {
+static void
+addConstantsToTrack(MachineFunction &MF, SPIRVGlobalRegistry *GR,
+                    DenseMap<MachineInstr *, Type *> &TargetExtConstTypes) {
   MachineRegisterInfo &MRI = MF.getRegInfo();
   DenseMap<MachineInstr *, Register> RegsAlreadyAddedToDT;
   SmallVector<MachineInstr *, 10> ToErase, ToEraseComposites;
@@ -47,6 +49,7 @@ static void addConstantsToTrack(MachineFunction &MF, SPIRVGlobalRegistry *GR) {
       if (!isSpvIntrinsic(MI, Intrinsic::spv_track_constant))
         continue;
       ToErase.push_back(&MI);
+      Register SrcReg = MI.getOperand(2).getReg();
       auto *Const =
           cast<Constant>(cast<ConstantAsMetadata>(
                              MI.getOperand(3).getMetadata()->getOperand(0))
@@ -54,14 +57,14 @@ static void addConstantsToTrack(MachineFunction &MF, SPIRVGlobalRegistry *GR) {
       if (auto *GV = dyn_cast<GlobalValue>(Const)) {
         Register Reg = GR->find(GV, &MF);
         if (!Reg.isValid())
-          GR->add(GV, &MF, MI.getOperand(2).getReg());
+          GR->add(GV, &MF, SrcReg);
         else
           RegsAlreadyAddedToDT[&MI] = Reg;
       } else {
         Register Reg = GR->find(Const, &MF);
         if (!Reg.isValid()) {
           if (auto *ConstVec = dyn_cast<ConstantDataVector>(Const)) {
-            auto *BuildVec = MRI.getVRegDef(MI.getOperand(2).getReg());
+            auto *BuildVec = MRI.getVRegDef(SrcReg);
             assert(BuildVec &&
                    BuildVec->getOpcode() == TargetOpcode::G_BUILD_VECTOR);
             for (unsigned i = 0; i < ConstVec->getNumElements(); ++i) {
@@ -75,7 +78,13 @@ static void addConstantsToTrack(MachineFunction &MF, SPIRVGlobalRegistry *GR) {
                 BuildVec->getOperand(1 + i).setReg(ElemReg);
             }
           }
-          GR->add(Const, &MF, MI.getOperand(2).getReg());
+          GR->add(Const, &MF, SrcReg);
+          if (Const->getType()->isTargetExtTy()) {
+            // remember association so that we can restore it when assign types
+            MachineInstr *SrcMI = MRI.getVRegDef(SrcReg);
+            if (SrcMI && SrcMI->getOpcode() == TargetOpcode::G_CONSTANT)
+              TargetExtConstTypes[SrcMI] = Const->getType();
+          }
         } else {
           RegsAlreadyAddedToDT[&MI] = Reg;
           // This MI is unused and will be removed. If the MI uses
@@ -224,6 +233,10 @@ static SPIRVType *propagateSPIRVType(MachineInstr *MI, SPIRVGlobalRegistry *GR,
         }
         break;
       }
+      case TargetOpcode::G_PTRTOINT:
+        SpirvTy = GR->getOrCreateSPIRVIntegerType(
+            MRI.getType(Reg).getScalarSizeInBits(), MIB);
+        break;
       case TargetOpcode::G_TRUNC:
       case TargetOpcode::G_ADDRSPACE_CAST:
       case TargetOpcode::G_PTR_ADD:
@@ -252,6 +265,7 @@ createNewIdReg(SPIRVType *SpvType, Register SrcReg, MachineRegisterInfo &MRI,
   if (!SpvType)
     SpvType = GR.getSPIRVTypeForVReg(SrcReg);
   assert(SpvType && "VReg is expected to have SPIRV type");
+  LLT SrcLLT = MRI.getType(SrcReg);
   LLT NewT = LLT::scalar(32);
   bool IsFloat = SpvType->getOpcode() == SPIRV::OpTypeFloat;
   bool IsVectorFloat =
@@ -261,10 +275,10 @@ createNewIdReg(SPIRVType *SpvType, Register SrcReg, MachineRegisterInfo &MRI,
   IsFloat |= IsVectorFloat;
   auto GetIdOp = IsFloat ? SPIRV::GET_fID : SPIRV::GET_ID;
   auto DstClass = IsFloat ? &SPIRV::fIDRegClass : &SPIRV::IDRegClass;
-  if (MRI.getType(SrcReg).isPointer()) {
+  if (SrcLLT.isPointer()) {
     unsigned PtrSz = GR.getPointerSize();
     NewT = LLT::pointer(0, PtrSz);
-    bool IsVec = MRI.getType(SrcReg).isVector();
+    bool IsVec = SrcLLT.isVector();
     if (IsVec)
       NewT = LLT::fixed_vector(2, NewT);
     if (PtrSz == 64) {
@@ -284,7 +298,7 @@ createNewIdReg(SPIRVType *SpvType, Register SrcReg, MachineRegisterInfo &MRI,
         DstClass = &SPIRV::pID32RegClass;
       }
     }
-  } else if (MRI.getType(SrcReg).isVector()) {
+  } else if (SrcLLT.isVector()) {
     NewT = LLT::fixed_vector(2, NewT);
     if (IsFloat) {
       GetIdOp = SPIRV::GET_vfID;
@@ -359,8 +373,10 @@ void processInstr(MachineInstr &MI, MachineIRBuilder &MIB,
 }
 } // namespace llvm
 
-static void generateAssignInstrs(MachineFunction &MF, SPIRVGlobalRegistry *GR,
-                                 MachineIRBuilder MIB) {
+static void
+generateAssignInstrs(MachineFunction &MF, SPIRVGlobalRegistry *GR,
+                     MachineIRBuilder MIB,
+                     DenseMap<MachineInstr *, Type *> &TargetExtConstTypes) {
   // Get access to information about available extensions
   const SPIRVSubtarget *ST =
       static_cast<const SPIRVSubtarget *>(&MIB.getMF().getSubtarget());
@@ -417,11 +433,14 @@ static void generateAssignInstrs(MachineFunction &MF, SPIRVGlobalRegistry *GR,
             continue;
         }
         Type *Ty = nullptr;
-        if (MI.getOpcode() == TargetOpcode::G_CONSTANT)
-          Ty = MI.getOperand(1).getCImm()->getType();
-        else if (MI.getOpcode() == TargetOpcode::G_FCONSTANT)
+        if (MI.getOpcode() == TargetOpcode::G_CONSTANT) {
+          auto TargetExtIt = TargetExtConstTypes.find(&MI);
+          Ty = TargetExtIt == TargetExtConstTypes.end()
+                   ? MI.getOperand(1).getCImm()->getType()
+                   : TargetExtIt->second;
+        } else if (MI.getOpcode() == TargetOpcode::G_FCONSTANT) {
           Ty = MI.getOperand(1).getFPImm()->getType();
-        else {
+        } else {
           assert(MI.getOpcode() == TargetOpcode::G_BUILD_VECTOR);
           Type *ElemTy = nullptr;
           MachineInstr *ElemMI = MRI.getVRegDef(MI.getOperand(1).getReg());
@@ -440,6 +459,7 @@ static void generateAssignInstrs(MachineFunction &MF, SPIRVGlobalRegistry *GR,
         insertAssignInstr(Reg, Ty, nullptr, GR, MIB, MRI);
       } else if (MI.getOpcode() == TargetOpcode::G_TRUNC ||
                  MI.getOpcode() == TargetOpcode::G_ZEXT ||
+                 MI.getOpcode() == TargetOpcode::G_PTRTOINT ||
                  MI.getOpcode() == TargetOpcode::G_GLOBAL_VALUE ||
                  MI.getOpcode() == TargetOpcode::COPY ||
                  MI.getOpcode() == TargetOpcode::G_ADDRSPACE_CAST) {
@@ -483,7 +503,8 @@ static void processInstrsWithTypeFolding(MachineFunction &MF,
         continue;
       Register DstReg = MI.getOperand(0).getReg();
       bool IsDstPtr = MRI.getType(DstReg).isPointer();
-      if (IsDstPtr || MRI.getType(DstReg).isVector())
+      bool isDstVec = MRI.getType(DstReg).isVector();
+      if (IsDstPtr || isDstVec)
         MRI.setRegClass(DstReg, &SPIRV::IDRegClass);
       // Don't need to reset type of register holding constant and used in
       // G_ADDRSPACE_CAST, since it breaks legalizer.
@@ -609,10 +630,12 @@ bool SPIRVPreLegalizer::runOnMachineFunction(MachineFunction &MF) {
   SPIRVGlobalRegistry *GR = ST.getSPIRVGlobalRegistry();
   GR->setCurrentFunc(MF);
   MachineIRBuilder MIB(MF);
-  addConstantsToTrack(MF, GR);
+  // a registry of target extension constants
+  DenseMap<MachineInstr *, Type *> TargetExtConstTypes;
+  addConstantsToTrack(MF, GR, TargetExtConstTypes);
   foldConstantsIntoIntrinsics(MF);
   insertBitcasts(MF, GR, MIB);
-  generateAssignInstrs(MF, GR, MIB);
+  generateAssignInstrs(MF, GR, MIB, TargetExtConstTypes);
   processSwitches(MF, GR, MIB);
   processInstrsWithTypeFolding(MF, GR, MIB);
   removeImplicitFallthroughs(MF, MIB);
diff --git a/llvm/lib/Target/SPIRV/SPIRVSymbolicOperands.td b/llvm/lib/Target/SPIRV/SPIRVSymbolicOperands.td
index ff102e318469..31e19ad8630c 100644
--- a/llvm/lib/Target/SPIRV/SPIRVSymbolicOperands.td
+++ b/llvm/lib/Target/SPIRV/SPIRVSymbolicOperands.td
@@ -1612,3 +1612,5 @@ multiclass OpcodeOperand<bits<32> value> {
 defm InBoundsPtrAccessChain : OpcodeOperand<70>;
 defm PtrCastToGeneric : OpcodeOperand<121>;
 defm Bitcast : OpcodeOperand<124>;
+defm ConvertPtrToU : OpcodeOperand<117>;
+defm ConvertUToPtr : OpcodeOperand<120>;
diff --git a/llvm/lib/Target/SystemZ/SystemZElimCompare.cpp b/llvm/lib/Target/SystemZ/SystemZElimCompare.cpp
index e58f50e471fc..99067e3ef187 100644
--- a/llvm/lib/Target/SystemZ/SystemZElimCompare.cpp
+++ b/llvm/lib/Target/SystemZ/SystemZElimCompare.cpp
@@ -633,7 +633,7 @@ bool SystemZElimCompare::fuseCompareOperations(
     RegMask = MBBI->getOperand(3).getRegMask();
 
   // Clear out all current operands.
-  int CCUse = MBBI->findRegisterUseOperandIdx(SystemZ::CC, false, TRI);
+  int CCUse = MBBI->findRegisterUseOperandIdx(SystemZ::CC, TRI, false);
   assert(CCUse >= 0 && "BRC/BCR must use CC");
   Branch->removeOperand(CCUse);
   // Remove regmask (sibcall).
@@ -707,11 +707,11 @@ bool SystemZElimCompare::processBlock(MachineBasicBlock &MBB) {
       continue;
     }
 
-    if (MI.definesRegister(SystemZ::CC)) {
+    if (MI.definesRegister(SystemZ::CC, /*TRI=*/nullptr)) {
       CCUsers.clear();
       CompleteCCUsers = true;
     }
-    if (MI.readsRegister(SystemZ::CC) && CompleteCCUsers)
+    if (MI.readsRegister(SystemZ::CC, /*TRI=*/nullptr) && CompleteCCUsers)
       CCUsers.push_back(&MI);
   }
   return Changed;
diff --git a/llvm/lib/Target/SystemZ/SystemZISelLowering.cpp b/llvm/lib/Target/SystemZ/SystemZISelLowering.cpp
index 48956b571dc3..2da4431cf077 100644
--- a/llvm/lib/Target/SystemZ/SystemZISelLowering.cpp
+++ b/llvm/lib/Target/SystemZ/SystemZISelLowering.cpp
@@ -8092,9 +8092,9 @@ static bool checkCCKill(MachineInstr &MI, MachineBasicBlock *MBB) {
   MachineBasicBlock::iterator miI(std::next(MachineBasicBlock::iterator(MI)));
   for (MachineBasicBlock::iterator miE = MBB->end(); miI != miE; ++miI) {
     const MachineInstr& mi = *miI;
-    if (mi.readsRegister(SystemZ::CC))
+    if (mi.readsRegister(SystemZ::CC, /*TRI=*/nullptr))
       return false;
-    if (mi.definesRegister(SystemZ::CC))
+    if (mi.definesRegister(SystemZ::CC, /*TRI=*/nullptr))
       break; // Should have kill-flag - update below.
   }
 
@@ -8233,7 +8233,8 @@ SystemZTargetLowering::emitSelect(MachineInstr &MI,
       }
       break;
     }
-    if (NextMI.definesRegister(SystemZ::CC) || NextMI.usesCustomInsertionHook())
+    if (NextMI.definesRegister(SystemZ::CC, /*TRI=*/nullptr) ||
+        NextMI.usesCustomInsertionHook())
       break;
     bool User = false;
     for (auto *SelMI : Selects)
@@ -8251,8 +8252,8 @@ SystemZTargetLowering::emitSelect(MachineInstr &MI,
   }
 
   MachineInstr *LastMI = Selects.back();
-  bool CCKilled =
-      (LastMI->killsRegister(SystemZ::CC) || checkCCKill(*LastMI, MBB));
+  bool CCKilled = (LastMI->killsRegister(SystemZ::CC, /*TRI=*/nullptr) ||
+                   checkCCKill(*LastMI, MBB));
   MachineBasicBlock *StartMBB = MBB;
   MachineBasicBlock *JoinMBB  = SystemZ::splitBlockAfter(LastMI, MBB);
   MachineBasicBlock *FalseMBB = SystemZ::emitBlockAfter(StartMBB);
@@ -8352,7 +8353,8 @@ MachineBasicBlock *SystemZTargetLowering::emitCondStore(MachineInstr &MI,
 
   // Unless CC was killed in the CondStore instruction, mark it as
   // live-in to both FalseMBB and JoinMBB.
-  if (!MI.killsRegister(SystemZ::CC) && !checkCCKill(MI, JoinMBB)) {
+  if (!MI.killsRegister(SystemZ::CC, /*TRI=*/nullptr) &&
+      !checkCCKill(MI, JoinMBB)) {
     FalseMBB->addLiveIn(SystemZ::CC);
     JoinMBB->addLiveIn(SystemZ::CC);
   }
@@ -8755,7 +8757,7 @@ SystemZTargetLowering::emitAtomicCmpSwapW(MachineInstr &MI,
   // If the CC def wasn't dead in the ATOMIC_CMP_SWAPW, mark CC as live-in
   // to the block after the loop.  At this point, CC may have been defined
   // either by the CR in LoopMBB or by the CS in SetMBB.
-  if (!MI.registerDefIsDead(SystemZ::CC))
+  if (!MI.registerDefIsDead(SystemZ::CC, /*TRI=*/nullptr))
     DoneMBB->addLiveIn(SystemZ::CC);
 
   MI.eraseFromParent();
@@ -9629,7 +9631,7 @@ SDValue SystemZTargetLowering::lowerVECREDUCE_ADD(SDValue Op,
   case 8:
   case 16:
     Op = DAG.getNode(SystemZISD::VSUM, DL, MVT::v4i32, Op, Zero);
-    LLVM_FALLTHROUGH;
+    [[fallthrough]];
   case 32:
   case 64:
     Op = DAG.getNode(SystemZISD::VSUM, DL, MVT::i128, Op,
diff --git a/llvm/lib/Target/SystemZ/SystemZInstrInfo.cpp b/llvm/lib/Target/SystemZ/SystemZInstrInfo.cpp
index 950548abcfa9..6b75c30943b4 100644
--- a/llvm/lib/Target/SystemZ/SystemZInstrInfo.cpp
+++ b/llvm/lib/Target/SystemZ/SystemZInstrInfo.cpp
@@ -938,8 +938,9 @@ static LogicOp interpretAndImmediate(unsigned Opcode) {
 }
 
 static void transferDeadCC(MachineInstr *OldMI, MachineInstr *NewMI) {
-  if (OldMI->registerDefIsDead(SystemZ::CC)) {
-    MachineOperand *CCDef = NewMI->findRegisterDefOperand(SystemZ::CC);
+  if (OldMI->registerDefIsDead(SystemZ::CC, /*TRI=*/nullptr)) {
+    MachineOperand *CCDef =
+        NewMI->findRegisterDefOperand(SystemZ::CC, /*TRI=*/nullptr);
     if (CCDef != nullptr)
       CCDef->setIsDead(true);
   }
@@ -1034,7 +1035,8 @@ MachineInstr *SystemZInstrInfo::foldMemoryOperandImpl(
         .addFrameIndex(FrameIndex)
         .addImm(0)
         .addImm(MI.getOperand(2).getImm());
-      BuiltMI->findRegisterDefOperand(SystemZ::CC)->setIsDead(true);
+      BuiltMI->findRegisterDefOperand(SystemZ::CC, /*TRI=*/nullptr)
+          ->setIsDead(true);
       CCLiveRange->createDeadDef(MISlot, LIS->getVNInfoAllocator());
       return BuiltMI;
     }
@@ -1195,7 +1197,7 @@ MachineInstr *SystemZInstrInfo::foldMemoryOperandImpl(
   unsigned NumOps = MI.getNumExplicitOperands();
   int MemOpcode = SystemZ::getMemOpcode(Opcode);
   if (MemOpcode == -1 ||
-      (CCLiveAtMI && !MI.definesRegister(SystemZ::CC) &&
+      (CCLiveAtMI && !MI.definesRegister(SystemZ::CC, /*TRI=*/nullptr) &&
        get(MemOpcode).hasImplicitDefOfPhysReg(SystemZ::CC)))
     return nullptr;
 
@@ -1303,9 +1305,9 @@ MachineInstr *SystemZInstrInfo::foldMemoryOperandImpl(
       MIB.addImm(CCValid);
       MIB.addImm(NeedsCommute ? CCMask ^ CCValid : CCMask);
     }
-    if (MIB->definesRegister(SystemZ::CC) &&
-        (!MI.definesRegister(SystemZ::CC) ||
-         MI.registerDefIsDead(SystemZ::CC))) {
+    if (MIB->definesRegister(SystemZ::CC, /*TRI=*/nullptr) &&
+        (!MI.definesRegister(SystemZ::CC, /*TRI=*/nullptr) ||
+         MI.registerDefIsDead(SystemZ::CC, /*TRI=*/nullptr))) {
       MIB->addRegisterDead(SystemZ::CC, TRI);
       if (CCLiveRange)
         CCLiveRange->createDeadDef(MISlot, LIS->getVNInfoAllocator());
@@ -1861,14 +1863,14 @@ prepareCompareSwapOperands(MachineBasicBlock::iterator const MBBI) const {
   bool CCLive = true;
   SmallVector<MachineInstr *, 4> CCUsers;
   for (MachineInstr &MI : llvm::make_range(std::next(MBBI), MBB->end())) {
-    if (MI.readsRegister(SystemZ::CC)) {
+    if (MI.readsRegister(SystemZ::CC, /*TRI=*/nullptr)) {
       unsigned Flags = MI.getDesc().TSFlags;
       if ((Flags & SystemZII::CCMaskFirst) || (Flags & SystemZII::CCMaskLast))
         CCUsers.push_back(&MI);
       else
         return false;
     }
-    if (MI.definesRegister(SystemZ::CC)) {
+    if (MI.definesRegister(SystemZ::CC, /*TRI=*/nullptr)) {
       CCLive = false;
       break;
     }
diff --git a/llvm/lib/Target/SystemZ/SystemZTargetTransformInfo.h b/llvm/lib/Target/SystemZ/SystemZTargetTransformInfo.h
index 696d887c1d5d..3cf4a69ac281 100644
--- a/llvm/lib/Target/SystemZ/SystemZTargetTransformInfo.h
+++ b/llvm/lib/Target/SystemZ/SystemZTargetTransformInfo.h
@@ -89,7 +89,7 @@ public:
       unsigned Opcode, Type *Ty, TTI::TargetCostKind CostKind,
       TTI::OperandValueInfo Op1Info = {TTI::OK_AnyValue, TTI::OP_None},
       TTI::OperandValueInfo Op2Info = {TTI::OK_AnyValue, TTI::OP_None},
-      ArrayRef<const Value *> Args = ArrayRef<const Value *>(),
+      ArrayRef<const Value *> Args = std::nullopt,
       const Instruction *CxtI = nullptr);
   InstructionCost getShuffleCost(TTI::ShuffleKind Kind, VectorType *Tp,
                                  ArrayRef<int> Mask,
diff --git a/llvm/lib/Target/WebAssembly/CMakeLists.txt b/llvm/lib/Target/WebAssembly/CMakeLists.txt
index f430be2653b4..1e83cbeac50d 100644
--- a/llvm/lib/Target/WebAssembly/CMakeLists.txt
+++ b/llvm/lib/Target/WebAssembly/CMakeLists.txt
@@ -19,6 +19,7 @@ add_llvm_target(WebAssemblyCodeGen
   WebAssemblyArgumentMove.cpp
   WebAssemblyAsmPrinter.cpp
   WebAssemblyCFGStackify.cpp
+  WebAssemblyCleanCodeAfterTrap.cpp
   WebAssemblyCFGSort.cpp
   WebAssemblyDebugFixup.cpp
   WebAssemblyDebugValueManager.cpp
diff --git a/llvm/lib/Target/WebAssembly/MCTargetDesc/WebAssemblyMCTypeUtilities.cpp b/llvm/lib/Target/WebAssembly/MCTargetDesc/WebAssemblyMCTypeUtilities.cpp
index b7b5b2a97c59..8ea02bd2ad1f 100644
--- a/llvm/lib/Target/WebAssembly/MCTargetDesc/WebAssemblyMCTypeUtilities.cpp
+++ b/llvm/lib/Target/WebAssembly/MCTargetDesc/WebAssemblyMCTypeUtilities.cpp
@@ -18,24 +18,16 @@
 using namespace llvm;
 
 std::optional<wasm::ValType> WebAssembly::parseType(StringRef Type) {
-  // FIXME: can't use StringSwitch because wasm::ValType doesn't have a
-  // "invalid" value.
-  if (Type == "i32")
-    return wasm::ValType::I32;
-  if (Type == "i64")
-    return wasm::ValType::I64;
-  if (Type == "f32")
-    return wasm::ValType::F32;
-  if (Type == "f64")
-    return wasm::ValType::F64;
-  if (Type == "v128" || Type == "i8x16" || Type == "i16x8" || Type == "i32x4" ||
-      Type == "i64x2" || Type == "f32x4" || Type == "f64x2")
-    return wasm::ValType::V128;
-  if (Type == "funcref")
-    return wasm::ValType::FUNCREF;
-  if (Type == "externref")
-    return wasm::ValType::EXTERNREF;
-  return std::nullopt;
+  return llvm::StringSwitch<std::optional<wasm::ValType>>{Type}
+      .Case("i32", wasm::ValType::I32)
+      .Case("i64", wasm::ValType::I64)
+      .Case("f32", wasm::ValType::F32)
+      .Case("f64", wasm::ValType::F64)
+      .Cases("v128", "i8x16", "i16x8", "i32x4", "i64x2", "f32x4", "f64x2",
+             wasm::ValType::V128)
+      .Case("funcref", wasm::ValType::FUNCREF)
+      .Case("externref", wasm::ValType::EXTERNREF)
+      .Default(std::nullopt);
 }
 
 WebAssembly::BlockType WebAssembly::parseBlockType(StringRef Type) {
diff --git a/llvm/lib/Target/WebAssembly/WebAssembly.h b/llvm/lib/Target/WebAssembly/WebAssembly.h
index 1c40addb6d6f..7fc8546248f1 100644
--- a/llvm/lib/Target/WebAssembly/WebAssembly.h
+++ b/llvm/lib/Target/WebAssembly/WebAssembly.h
@@ -37,6 +37,7 @@ FunctionPass *createWebAssemblyISelDag(WebAssemblyTargetMachine &TM,
                                        CodeGenOptLevel OptLevel);
 FunctionPass *createWebAssemblyArgumentMove();
 FunctionPass *createWebAssemblySetP2AlignOperands();
+FunctionPass *createWebAssemblyCleanCodeAfterTrap();
 
 // Late passes.
 FunctionPass *createWebAssemblyReplacePhysRegs();
@@ -63,6 +64,7 @@ void initializeOptimizeReturnedPass(PassRegistry &);
 void initializeWebAssemblyRefTypeMem2LocalPass(PassRegistry &);
 void initializeWebAssemblyAddMissingPrototypesPass(PassRegistry &);
 void initializeWebAssemblyArgumentMovePass(PassRegistry &);
+void initializeWebAssemblyCleanCodeAfterTrapPass(PassRegistry &);
 void initializeWebAssemblyCFGSortPass(PassRegistry &);
 void initializeWebAssemblyCFGStackifyPass(PassRegistry &);
 void initializeWebAssemblyDAGToDAGISelPass(PassRegistry &);
diff --git a/llvm/lib/Target/WebAssembly/WebAssembly.td b/llvm/lib/Target/WebAssembly/WebAssembly.td
index d538197450b6..f00974531209 100644
--- a/llvm/lib/Target/WebAssembly/WebAssembly.td
+++ b/llvm/lib/Target/WebAssembly/WebAssembly.td
@@ -28,6 +28,9 @@ def FeatureSIMD128 : SubtargetFeature<"simd128", "SIMDLevel", "SIMD128",
 def FeatureRelaxedSIMD : SubtargetFeature<"relaxed-simd", "SIMDLevel", "RelaxedSIMD",
                                       "Enable relaxed-simd instructions">;
 
+def FeatureHalfPrecision : SubtargetFeature<"half-precision", "HasHalfPrecision", "true",
+                                            "Enable half precision instructions">;
+
 def FeatureAtomics : SubtargetFeature<"atomics", "HasAtomics", "true",
                                       "Enable Atomics">;
 
diff --git a/llvm/lib/Target/WebAssembly/WebAssemblyCleanCodeAfterTrap.cpp b/llvm/lib/Target/WebAssembly/WebAssemblyCleanCodeAfterTrap.cpp
new file mode 100644
index 000000000000..e5cba3c48547
--- /dev/null
+++ b/llvm/lib/Target/WebAssembly/WebAssemblyCleanCodeAfterTrap.cpp
@@ -0,0 +1,80 @@
+//===-- WebAssemblyCleanCodeAfterTrap.cpp - Clean Code After Trap ---------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+///
+/// \file
+/// This file remove instruction after trap.
+/// ``llvm.trap`` will be convert as ``unreachable`` which is terminator.
+/// Instruction after terminator will cause validation failed.
+///
+//===----------------------------------------------------------------------===//
+
+#include "WebAssembly.h"
+#include "WebAssemblyUtilities.h"
+#include "llvm/ADT/SmallVector.h"
+#include "llvm/CodeGen/MachineBlockFrequencyInfo.h"
+#include "llvm/CodeGen/Passes.h"
+#include "llvm/MC/MCInstrDesc.h"
+#include "llvm/Support/Debug.h"
+#include "llvm/Support/raw_ostream.h"
+using namespace llvm;
+
+#define DEBUG_TYPE "wasm-clean-code-after-trap"
+
+namespace {
+class WebAssemblyCleanCodeAfterTrap final : public MachineFunctionPass {
+public:
+  static char ID; // Pass identification, replacement for typeid
+  WebAssemblyCleanCodeAfterTrap() : MachineFunctionPass(ID) {}
+
+  StringRef getPassName() const override {
+    return "WebAssembly Clean Code After Trap";
+  }
+
+  bool runOnMachineFunction(MachineFunction &MF) override;
+};
+} // end anonymous namespace
+
+char WebAssemblyCleanCodeAfterTrap::ID = 0;
+INITIALIZE_PASS(WebAssemblyCleanCodeAfterTrap, DEBUG_TYPE,
+                "WebAssembly Clean Code After Trap", false, false)
+
+FunctionPass *llvm::createWebAssemblyCleanCodeAfterTrap() {
+  return new WebAssemblyCleanCodeAfterTrap();
+}
+
+bool WebAssemblyCleanCodeAfterTrap::runOnMachineFunction(MachineFunction &MF) {
+  LLVM_DEBUG({
+    dbgs() << "********** CleanCodeAfterTrap **********\n"
+           << "********** Function: " << MF.getName() << '\n';
+  });
+
+  bool Changed = false;
+
+  for (MachineBasicBlock &BB : MF) {
+    bool HasTerminator = false;
+    llvm::SmallVector<MachineInstr *> RemoveMI{};
+    for (MachineInstr &MI : BB) {
+      if (HasTerminator)
+        RemoveMI.push_back(&MI);
+      if (MI.hasProperty(MCID::Trap) && MI.isTerminator())
+        HasTerminator = true;
+    }
+    if (!RemoveMI.empty()) {
+      Changed = true;
+      LLVM_DEBUG({
+        for (MachineInstr *MI : RemoveMI) {
+          llvm::dbgs() << "* remove ";
+          MI->print(llvm::dbgs());
+        }
+      });
+      for (MachineInstr *MI : RemoveMI)
+        MI->eraseFromParent();
+    }
+  }
+  return Changed;
+}
diff --git a/llvm/lib/Target/WebAssembly/WebAssemblyDebugValueManager.cpp b/llvm/lib/Target/WebAssembly/WebAssemblyDebugValueManager.cpp
index a2a054127d5f..da3717499689 100644
--- a/llvm/lib/Target/WebAssembly/WebAssemblyDebugValueManager.cpp
+++ b/llvm/lib/Target/WebAssembly/WebAssemblyDebugValueManager.cpp
@@ -37,7 +37,7 @@ WebAssemblyDebugValueManager::WebAssemblyDebugValueManager(MachineInstr *Def)
                                    ME = Def->getParent()->end();
        MI != ME; ++MI) {
     // If another definition appears, stop
-    if (MI->definesRegister(CurrentReg))
+    if (MI->definesRegister(CurrentReg, /*TRI=*/nullptr))
       break;
     if (MI->isDebugValue() && MI->hasDebugOperandForReg(CurrentReg))
       DbgValues.push_back(&*MI);
diff --git a/llvm/lib/Target/WebAssembly/WebAssemblyISelLowering.cpp b/llvm/lib/Target/WebAssembly/WebAssemblyISelLowering.cpp
index 905ff3b90184..64bcadf3f567 100644
--- a/llvm/lib/Target/WebAssembly/WebAssemblyISelLowering.cpp
+++ b/llvm/lib/Target/WebAssembly/WebAssemblyISelLowering.cpp
@@ -1288,7 +1288,7 @@ bool WebAssemblyTargetLowering::CanLowerReturn(
     const SmallVectorImpl<ISD::OutputArg> &Outs,
     LLVMContext & /*Context*/) const {
   // WebAssembly can only handle returning tuples with multivalue enabled
-  return Subtarget->hasMultivalue() || Outs.size() <= 1;
+  return WebAssembly::canLowerReturn(Outs.size(), Subtarget);
 }
 
 SDValue WebAssemblyTargetLowering::LowerReturn(
@@ -1296,7 +1296,7 @@ SDValue WebAssemblyTargetLowering::LowerReturn(
     const SmallVectorImpl<ISD::OutputArg> &Outs,
     const SmallVectorImpl<SDValue> &OutVals, const SDLoc &DL,
     SelectionDAG &DAG) const {
-  assert((Subtarget->hasMultivalue() || Outs.size() <= 1) &&
+  assert(WebAssembly::canLowerReturn(Outs.size(), Subtarget) &&
          "MVP WebAssembly can only return up to one value");
   if (!callingConvSupported(CallConv))
     fail(DL, DAG, "WebAssembly doesn't support non-C calling conventions");
diff --git a/llvm/lib/Target/WebAssembly/WebAssemblyInstrInfo.td b/llvm/lib/Target/WebAssembly/WebAssemblyInstrInfo.td
index 59ea9247bd86..7b57f8ce90e0 100644
--- a/llvm/lib/Target/WebAssembly/WebAssemblyInstrInfo.td
+++ b/llvm/lib/Target/WebAssembly/WebAssemblyInstrInfo.td
@@ -30,6 +30,10 @@ def HasRelaxedSIMD :
     Predicate<"Subtarget->hasRelaxedSIMD()">,
     AssemblerPredicate<(all_of FeatureRelaxedSIMD), "relaxed-simd">;
 
+def HasHalfPrecision :
+    Predicate<"Subtarget->hasHalfPrecision()">,
+    AssemblerPredicate<(all_of FeatureHalfPrecision), "half-precision">;
+
 def HasAtomics :
     Predicate<"Subtarget->hasAtomics()">,
     AssemblerPredicate<(all_of FeatureAtomics), "atomics">;
diff --git a/llvm/lib/Target/WebAssembly/WebAssemblyMachineFunctionInfo.cpp b/llvm/lib/Target/WebAssembly/WebAssemblyMachineFunctionInfo.cpp
index 6f4e7d876c69..7505c2995cf7 100644
--- a/llvm/lib/Target/WebAssembly/WebAssemblyMachineFunctionInfo.cpp
+++ b/llvm/lib/Target/WebAssembly/WebAssemblyMachineFunctionInfo.cpp
@@ -17,6 +17,7 @@
 #include "Utils/WebAssemblyTypeUtilities.h"
 #include "WebAssemblyISelLowering.h"
 #include "WebAssemblySubtarget.h"
+#include "WebAssemblyUtilities.h"
 #include "llvm/CodeGen/Analysis.h"
 #include "llvm/CodeGen/WasmEHFuncInfo.h"
 #include "llvm/Target/TargetMachine.h"
@@ -70,8 +71,9 @@ void llvm::computeSignatureVTs(const FunctionType *Ty,
   computeLegalValueVTs(ContextFunc, TM, Ty->getReturnType(), Results);
 
   MVT PtrVT = MVT::getIntegerVT(TM.createDataLayout().getPointerSizeInBits());
-  if (Results.size() > 1 &&
-      !TM.getSubtarget<WebAssemblySubtarget>(ContextFunc).hasMultivalue()) {
+  if (!WebAssembly::canLowerReturn(
+          Results.size(),
+          &TM.getSubtarget<WebAssemblySubtarget>(ContextFunc))) {
     // WebAssembly can't lower returns of multiple values without demoting to
     // sret unless multivalue is enabled (see
     // WebAssemblyTargetLowering::CanLowerReturn). So replace multiple return
diff --git a/llvm/lib/Target/WebAssembly/WebAssemblyRefTypeMem2Local.cpp b/llvm/lib/Target/WebAssembly/WebAssemblyRefTypeMem2Local.cpp
index d3c60ee289df..04b4c7d78aab 100644
--- a/llvm/lib/Target/WebAssembly/WebAssemblyRefTypeMem2Local.cpp
+++ b/llvm/lib/Target/WebAssembly/WebAssemblyRefTypeMem2Local.cpp
@@ -86,6 +86,9 @@ bool WebAssemblyRefTypeMem2Local::runOnFunction(Function &F) {
                        "********** Function: "
                     << F.getName() << '\n');
 
-  visit(F);
+  if (F.getFnAttribute("target-features")
+          .getValueAsString()
+          .contains("+reference-types"))
+    visit(F);
   return Changed;
 }
diff --git a/llvm/lib/Target/WebAssembly/WebAssemblyRegStackify.cpp b/llvm/lib/Target/WebAssembly/WebAssemblyRegStackify.cpp
index 3046f9476f91..ef174e1716ef 100644
--- a/llvm/lib/Target/WebAssembly/WebAssemblyRegStackify.cpp
+++ b/llvm/lib/Target/WebAssembly/WebAssemblyRegStackify.cpp
@@ -80,13 +80,13 @@ FunctionPass *llvm::createWebAssemblyRegStackify() {
 // the expression stack.
 static void imposeStackOrdering(MachineInstr *MI) {
   // Write the opaque VALUE_STACK register.
-  if (!MI->definesRegister(WebAssembly::VALUE_STACK))
+  if (!MI->definesRegister(WebAssembly::VALUE_STACK, /*TRI=*/nullptr))
     MI->addOperand(MachineOperand::CreateReg(WebAssembly::VALUE_STACK,
                                              /*isDef=*/true,
                                              /*isImp=*/true));
 
   // Also read the opaque VALUE_STACK register.
-  if (!MI->readsRegister(WebAssembly::VALUE_STACK))
+  if (!MI->readsRegister(WebAssembly::VALUE_STACK, /*TRI=*/nullptr))
     MI->addOperand(MachineOperand::CreateReg(WebAssembly::VALUE_STACK,
                                              /*isDef=*/false,
                                              /*isImp=*/true));
@@ -371,8 +371,8 @@ static bool isSafeToMove(const MachineOperand *Def, const MachineOperand *Use,
     Register Reg = MO.getReg();
 
     // If the register is dead here and at Insert, ignore it.
-    if (MO.isDead() && Insert->definesRegister(Reg) &&
-        !Insert->readsRegister(Reg))
+    if (MO.isDead() && Insert->definesRegister(Reg, /*TRI=*/nullptr) &&
+        !Insert->readsRegister(Reg, /*TRI=*/nullptr))
       continue;
 
     if (Reg.isPhysical()) {
@@ -864,7 +864,8 @@ bool WebAssemblyRegStackify::runOnMachineFunction(MachineFunction &MF) {
         if (WebAssembly::isArgument(DefI->getOpcode()))
           continue;
 
-        MachineOperand *Def = DefI->findRegisterDefOperand(Reg);
+        MachineOperand *Def =
+            DefI->findRegisterDefOperand(Reg, /*TRI=*/nullptr);
         assert(Def != nullptr);
 
         // Decide which strategy to take. Prefer to move a single-use value
diff --git a/llvm/lib/Target/WebAssembly/WebAssemblyRuntimeLibcallSignatures.cpp b/llvm/lib/Target/WebAssembly/WebAssemblyRuntimeLibcallSignatures.cpp
index 1896ac631d96..d9936557776b 100644
--- a/llvm/lib/Target/WebAssembly/WebAssemblyRuntimeLibcallSignatures.cpp
+++ b/llvm/lib/Target/WebAssembly/WebAssemblyRuntimeLibcallSignatures.cpp
@@ -20,6 +20,7 @@
 
 #include "WebAssemblyRuntimeLibcallSignatures.h"
 #include "WebAssemblySubtarget.h"
+#include "WebAssemblyUtilities.h"
 #include "llvm/CodeGen/RuntimeLibcalls.h"
 
 using namespace llvm;
@@ -694,7 +695,7 @@ void WebAssembly::getLibcallSignature(const WebAssemblySubtarget &Subtarget,
     Params.push_back(PtrTy);
     break;
   case i64_i64_func_f32:
-    if (Subtarget.hasMultivalue()) {
+    if (WebAssembly::canLowerMultivalueReturn(&Subtarget)) {
       Rets.push_back(wasm::ValType::I64);
       Rets.push_back(wasm::ValType::I64);
     } else {
@@ -703,7 +704,7 @@ void WebAssembly::getLibcallSignature(const WebAssemblySubtarget &Subtarget,
     Params.push_back(wasm::ValType::F32);
     break;
   case i64_i64_func_f64:
-    if (Subtarget.hasMultivalue()) {
+    if (WebAssembly::canLowerMultivalueReturn(&Subtarget)) {
       Rets.push_back(wasm::ValType::I64);
       Rets.push_back(wasm::ValType::I64);
     } else {
@@ -712,7 +713,7 @@ void WebAssembly::getLibcallSignature(const WebAssemblySubtarget &Subtarget,
     Params.push_back(wasm::ValType::F64);
     break;
   case i16_i16_func_i16_i16:
-    if (Subtarget.hasMultivalue()) {
+    if (WebAssembly::canLowerMultivalueReturn(&Subtarget)) {
       Rets.push_back(wasm::ValType::I32);
       Rets.push_back(wasm::ValType::I32);
     } else {
@@ -722,7 +723,7 @@ void WebAssembly::getLibcallSignature(const WebAssemblySubtarget &Subtarget,
     Params.push_back(wasm::ValType::I32);
     break;
   case i32_i32_func_i32_i32:
-    if (Subtarget.hasMultivalue()) {
+    if (WebAssembly::canLowerMultivalueReturn(&Subtarget)) {
       Rets.push_back(wasm::ValType::I32);
       Rets.push_back(wasm::ValType::I32);
     } else {
@@ -732,7 +733,7 @@ void WebAssembly::getLibcallSignature(const WebAssemblySubtarget &Subtarget,
     Params.push_back(wasm::ValType::I32);
     break;
   case i64_i64_func_i64_i64:
-    if (Subtarget.hasMultivalue()) {
+    if (WebAssembly::canLowerMultivalueReturn(&Subtarget)) {
       Rets.push_back(wasm::ValType::I64);
       Rets.push_back(wasm::ValType::I64);
     } else {
@@ -742,7 +743,7 @@ void WebAssembly::getLibcallSignature(const WebAssemblySubtarget &Subtarget,
     Params.push_back(wasm::ValType::I64);
     break;
   case i64_i64_func_i64_i64_i64_i64:
-    if (Subtarget.hasMultivalue()) {
+    if (WebAssembly::canLowerMultivalueReturn(&Subtarget)) {
       Rets.push_back(wasm::ValType::I64);
       Rets.push_back(wasm::ValType::I64);
     } else {
@@ -754,7 +755,7 @@ void WebAssembly::getLibcallSignature(const WebAssemblySubtarget &Subtarget,
     Params.push_back(wasm::ValType::I64);
     break;
   case i64_i64_func_i64_i64_i64_i64_iPTR:
-    if (Subtarget.hasMultivalue()) {
+    if (WebAssembly::canLowerMultivalueReturn(&Subtarget)) {
       Rets.push_back(wasm::ValType::I64);
       Rets.push_back(wasm::ValType::I64);
     } else {
@@ -767,7 +768,7 @@ void WebAssembly::getLibcallSignature(const WebAssemblySubtarget &Subtarget,
     Params.push_back(PtrTy);
     break;
   case i64_i64_i64_i64_func_i64_i64_i64_i64:
-    if (Subtarget.hasMultivalue()) {
+    if (WebAssembly::canLowerMultivalueReturn(&Subtarget)) {
       Rets.push_back(wasm::ValType::I64);
       Rets.push_back(wasm::ValType::I64);
       Rets.push_back(wasm::ValType::I64);
@@ -781,7 +782,7 @@ void WebAssembly::getLibcallSignature(const WebAssemblySubtarget &Subtarget,
     Params.push_back(wasm::ValType::I64);
     break;
   case i64_i64_func_i64_i64_i32:
-    if (Subtarget.hasMultivalue()) {
+    if (WebAssembly::canLowerMultivalueReturn(&Subtarget)) {
       Rets.push_back(wasm::ValType::I64);
       Rets.push_back(wasm::ValType::I64);
     } else {
@@ -851,7 +852,7 @@ void WebAssembly::getLibcallSignature(const WebAssemblySubtarget &Subtarget,
     Params.push_back(wasm::ValType::I64);
     break;
   case i64_i64_func_i64_i64_i64_i64_i64_i64:
-    if (Subtarget.hasMultivalue()) {
+    if (WebAssembly::canLowerMultivalueReturn(&Subtarget)) {
       Rets.push_back(wasm::ValType::I64);
       Rets.push_back(wasm::ValType::I64);
     } else {
@@ -865,7 +866,7 @@ void WebAssembly::getLibcallSignature(const WebAssemblySubtarget &Subtarget,
     Params.push_back(wasm::ValType::I64);
     break;
   case i64_i64_func_i32:
-    if (Subtarget.hasMultivalue()) {
+    if (WebAssembly::canLowerMultivalueReturn(&Subtarget)) {
       Rets.push_back(wasm::ValType::I64);
       Rets.push_back(wasm::ValType::I64);
     } else {
@@ -874,7 +875,7 @@ void WebAssembly::getLibcallSignature(const WebAssemblySubtarget &Subtarget,
     Params.push_back(wasm::ValType::I32);
     break;
   case i64_i64_func_i64:
-    if (Subtarget.hasMultivalue()) {
+    if (WebAssembly::canLowerMultivalueReturn(&Subtarget)) {
       Rets.push_back(wasm::ValType::I64);
       Rets.push_back(wasm::ValType::I64);
     } else {
diff --git a/llvm/lib/Target/WebAssembly/WebAssemblySubtarget.h b/llvm/lib/Target/WebAssembly/WebAssemblySubtarget.h
index 85d02b087c78..8b0d9fa12de8 100644
--- a/llvm/lib/Target/WebAssembly/WebAssemblySubtarget.h
+++ b/llvm/lib/Target/WebAssembly/WebAssemblySubtarget.h
@@ -50,6 +50,7 @@ class WebAssemblySubtarget final : public WebAssemblyGenSubtargetInfo {
   bool HasReferenceTypes = false;
   bool HasExtendedConst = false;
   bool HasMultiMemory = false;
+  bool HasHalfPrecision = false;
 
   /// What processor and OS we're targeting.
   Triple TargetTriple;
@@ -93,6 +94,7 @@ public:
   bool hasAddr64() const { return TargetTriple.isArch64Bit(); }
   bool hasSIMD128() const { return SIMDLevel >= SIMD128; }
   bool hasRelaxedSIMD() const { return SIMDLevel >= RelaxedSIMD; }
+  bool hasHalfPrecision() const { return HasHalfPrecision; }
   bool hasAtomics() const { return HasAtomics; }
   bool hasNontrappingFPToInt() const { return HasNontrappingFPToInt; }
   bool hasSignExt() const { return HasSignExt; }
diff --git a/llvm/lib/Target/WebAssembly/WebAssemblyTargetMachine.cpp b/llvm/lib/Target/WebAssembly/WebAssemblyTargetMachine.cpp
index 769ee765e190..de342e896573 100644
--- a/llvm/lib/Target/WebAssembly/WebAssemblyTargetMachine.cpp
+++ b/llvm/lib/Target/WebAssembly/WebAssemblyTargetMachine.cpp
@@ -128,7 +128,8 @@ WebAssemblyTargetMachine::WebAssemblyTargetMachine(
                                        "n32:64-S128-ni:1:10:20"),
           TT, CPU, FS, Options, getEffectiveRelocModel(RM, TT),
           getEffectiveCodeModel(CM, CodeModel::Large), OL),
-      TLOF(new WebAssemblyTargetObjectFile()) {
+      TLOF(new WebAssemblyTargetObjectFile()),
+      UsesMultivalueABI(Options.MCOptions.getABIName() == "experimental-mv") {
   // WebAssembly type-checks instructions, but a noreturn function with a return
   // type that doesn't match the context will cause a check failure. So we lower
   // LLVM 'unreachable' to ISD::TRAP and then lower that to WebAssembly's
@@ -483,16 +484,9 @@ void WebAssemblyPassConfig::addIRPasses() {
 }
 
 void WebAssemblyPassConfig::addISelPrepare() {
-  WebAssemblyTargetMachine *WasmTM =
-      static_cast<WebAssemblyTargetMachine *>(TM);
-  const WebAssemblySubtarget *Subtarget =
-      WasmTM->getSubtargetImpl(std::string(WasmTM->getTargetCPU()),
-                               std::string(WasmTM->getTargetFeatureString()));
-  if (Subtarget->hasReferenceTypes()) {
-    // We need to move reference type allocas to WASM_ADDRESS_SPACE_VAR so that
-    // loads and stores are promoted to local.gets/local.sets.
-    addPass(createWebAssemblyRefTypeMem2Local());
-  }
+  // We need to move reference type allocas to WASM_ADDRESS_SPACE_VAR so that
+  // loads and stores are promoted to local.gets/local.sets.
+  addPass(createWebAssemblyRefTypeMem2Local());
   // Lower atomics and TLS if necessary
   addPass(new CoalesceFeaturesAndStripAtomics(&getWebAssemblyTargetMachine()));
 
@@ -518,6 +512,10 @@ bool WebAssemblyPassConfig::addInstSelector() {
   // Eliminate range checks and add default targets to br_table instructions.
   addPass(createWebAssemblyFixBrTableDefaults());
 
+  // unreachable is terminator, non-terminator instruction after it is not
+  // allowed.
+  addPass(createWebAssemblyCleanCodeAfterTrap());
+
   return false;
 }
 
diff --git a/llvm/lib/Target/WebAssembly/WebAssemblyTargetMachine.h b/llvm/lib/Target/WebAssembly/WebAssemblyTargetMachine.h
index 2e8cd43840e3..1ff2e175978c 100644
--- a/llvm/lib/Target/WebAssembly/WebAssemblyTargetMachine.h
+++ b/llvm/lib/Target/WebAssembly/WebAssemblyTargetMachine.h
@@ -24,6 +24,7 @@ namespace llvm {
 class WebAssemblyTargetMachine final : public LLVMTargetMachine {
   std::unique_ptr<TargetLoweringObjectFile> TLOF;
   mutable StringMap<std::unique_ptr<WebAssemblySubtarget>> SubtargetMap;
+  bool UsesMultivalueABI = false;
 
 public:
   WebAssemblyTargetMachine(const Target &T, const Triple &TT, StringRef CPU,
@@ -62,6 +63,8 @@ public:
                                 PerFunctionMIParsingState &PFS,
                                 SMDiagnostic &Error,
                                 SMRange &SourceRange) const override;
+
+  bool usesMultivalueABI() const { return UsesMultivalueABI; }
 };
 
 } // end namespace llvm
diff --git a/llvm/lib/Target/WebAssembly/WebAssemblyTargetTransformInfo.h b/llvm/lib/Target/WebAssembly/WebAssemblyTargetTransformInfo.h
index a803fe5c1bbe..801f905d377e 100644
--- a/llvm/lib/Target/WebAssembly/WebAssemblyTargetTransformInfo.h
+++ b/llvm/lib/Target/WebAssembly/WebAssemblyTargetTransformInfo.h
@@ -63,7 +63,7 @@ public:
       unsigned Opcode, Type *Ty, TTI::TargetCostKind CostKind,
       TTI::OperandValueInfo Op1Info = {TTI::OK_AnyValue, TTI::OP_None},
       TTI::OperandValueInfo Op2Info = {TTI::OK_AnyValue, TTI::OP_None},
-      ArrayRef<const Value *> Args = ArrayRef<const Value *>(),
+      ArrayRef<const Value *> Args = std::nullopt,
       const Instruction *CxtI = nullptr);
   using BaseT::getVectorInstrCost;
   InstructionCost getVectorInstrCost(unsigned Opcode, Type *Val,
diff --git a/llvm/lib/Target/WebAssembly/WebAssemblyUtilities.cpp b/llvm/lib/Target/WebAssembly/WebAssemblyUtilities.cpp
index ac7cf5b37fca..60e872549f87 100644
--- a/llvm/lib/Target/WebAssembly/WebAssemblyUtilities.cpp
+++ b/llvm/lib/Target/WebAssembly/WebAssemblyUtilities.cpp
@@ -13,7 +13,7 @@
 
 #include "WebAssemblyUtilities.h"
 #include "WebAssemblyMachineFunctionInfo.h"
-#include "WebAssemblySubtarget.h"
+#include "WebAssemblyTargetMachine.h"
 #include "llvm/CodeGen/MachineInstr.h"
 #include "llvm/CodeGen/MachineLoopInfo.h"
 #include "llvm/IR/Function.h"
@@ -179,3 +179,15 @@ unsigned WebAssembly::getCopyOpcodeForRegClass(const TargetRegisterClass *RC) {
     llvm_unreachable("Unexpected register class");
   }
 }
+
+bool WebAssembly::canLowerMultivalueReturn(
+    const WebAssemblySubtarget *Subtarget) {
+  const auto &TM = static_cast<const WebAssemblyTargetMachine &>(
+      Subtarget->getTargetLowering()->getTargetMachine());
+  return Subtarget->hasMultivalue() && TM.usesMultivalueABI();
+}
+
+bool WebAssembly::canLowerReturn(size_t ResultSize,
+                                 const WebAssemblySubtarget *Subtarget) {
+  return ResultSize <= 1 || canLowerMultivalueReturn(Subtarget);
+}
diff --git a/llvm/lib/Target/WebAssembly/WebAssemblyUtilities.h b/llvm/lib/Target/WebAssembly/WebAssemblyUtilities.h
index 7f28fb1858a6..046b1b5db2a7 100644
--- a/llvm/lib/Target/WebAssembly/WebAssemblyUtilities.h
+++ b/llvm/lib/Target/WebAssembly/WebAssemblyUtilities.h
@@ -63,6 +63,16 @@ MachineInstr *findCatch(MachineBasicBlock *EHPad);
 /// Returns the appropriate copy opcode for the given register class.
 unsigned getCopyOpcodeForRegClass(const TargetRegisterClass *RC);
 
+/// Returns true if multivalue returns of a function can be lowered directly,
+/// i.e., not indirectly via a pointer parameter that points to the value in
+/// memory.
+bool canLowerMultivalueReturn(const WebAssemblySubtarget *Subtarget);
+
+/// Returns true if the function's return value(s) can be lowered directly,
+/// i.e., not indirectly via a pointer parameter that points to the value in
+/// memory.
+bool canLowerReturn(size_t ResultSize, const WebAssemblySubtarget *Subtarget);
+
 } // end namespace WebAssembly
 
 } // end namespace llvm
diff --git a/llvm/lib/Target/X86/MCTargetDesc/X86BaseInfo.h b/llvm/lib/Target/X86/MCTargetDesc/X86BaseInfo.h
index c0a75e215a40..8e4015783641 100644
--- a/llvm/lib/Target/X86/MCTargetDesc/X86BaseInfo.h
+++ b/llvm/lib/Target/X86/MCTargetDesc/X86BaseInfo.h
@@ -1168,33 +1168,34 @@ inline int getMemoryOperandNo(uint64_t TSFlags) {
 
 /// \returns true if the register is a XMM.
 inline bool isXMMReg(unsigned RegNo) {
-  assert(X86::XMM15 - X86::XMM0 == 15 &&
-         "XMM0-15 registers are not continuous");
-  assert(X86::XMM31 - X86::XMM16 == 15 &&
-         "XMM16-31 registers are not continuous");
+  static_assert(X86::XMM15 - X86::XMM0 == 15,
+                "XMM0-15 registers are not continuous");
+  static_assert(X86::XMM31 - X86::XMM16 == 15,
+                "XMM16-31 registers are not continuous");
   return (RegNo >= X86::XMM0 && RegNo <= X86::XMM15) ||
          (RegNo >= X86::XMM16 && RegNo <= X86::XMM31);
 }
 
 /// \returns true if the register is a YMM.
 inline bool isYMMReg(unsigned RegNo) {
-  assert(X86::YMM15 - X86::YMM0 == 15 &&
-         "YMM0-15 registers are not continuous");
-  assert(X86::YMM31 - X86::YMM16 == 15 &&
-         "YMM16-31 registers are not continuous");
+  static_assert(X86::YMM15 - X86::YMM0 == 15,
+                "YMM0-15 registers are not continuous");
+  static_assert(X86::YMM31 - X86::YMM16 == 15,
+                "YMM16-31 registers are not continuous");
   return (RegNo >= X86::YMM0 && RegNo <= X86::YMM15) ||
          (RegNo >= X86::YMM16 && RegNo <= X86::YMM31);
 }
 
 /// \returns true if the register is a ZMM.
 inline bool isZMMReg(unsigned RegNo) {
-  assert(X86::ZMM31 - X86::ZMM0 == 31 && "ZMM registers are not continuous");
+  static_assert(X86::ZMM31 - X86::ZMM0 == 31,
+                "ZMM registers are not continuous");
   return RegNo >= X86::ZMM0 && RegNo <= X86::ZMM31;
 }
 
 /// \returns true if \p RegNo is an apx extended register.
 inline bool isApxExtendedReg(unsigned RegNo) {
-  assert(X86::R31WH - X86::R16 == 95 && "EGPRs are not continuous");
+  static_assert(X86::R31WH - X86::R16 == 95, "EGPRs are not continuous");
   return RegNo >= X86::R16 && RegNo <= X86::R31WH;
 }
 
diff --git a/llvm/lib/Target/X86/MCTargetDesc/X86MCTargetDesc.h b/llvm/lib/Target/X86/MCTargetDesc/X86MCTargetDesc.h
index 437a7bd6ff6c..18ecca34943f 100644
--- a/llvm/lib/Target/X86/MCTargetDesc/X86MCTargetDesc.h
+++ b/llvm/lib/Target/X86/MCTargetDesc/X86MCTargetDesc.h
@@ -121,7 +121,6 @@ MCStreamer *createX86WinCOFFStreamer(MCContext &C,
                                      std::unique_ptr<MCAsmBackend> &&AB,
                                      std::unique_ptr<MCObjectWriter> &&OW,
                                      std::unique_ptr<MCCodeEmitter> &&CE,
-                                     bool RelaxAll,
                                      bool IncrementalLinkerCompatible);
 
 /// Construct an X86 Mach-O object writer.
diff --git a/llvm/lib/Target/X86/MCTargetDesc/X86WinCOFFStreamer.cpp b/llvm/lib/Target/X86/MCTargetDesc/X86WinCOFFStreamer.cpp
index 36945d1f6746..dac8bc1fb1be 100644
--- a/llvm/lib/Target/X86/MCTargetDesc/X86WinCOFFStreamer.cpp
+++ b/llvm/lib/Target/X86/MCTargetDesc/X86WinCOFFStreamer.cpp
@@ -70,11 +70,9 @@ MCStreamer *llvm::createX86WinCOFFStreamer(MCContext &C,
                                            std::unique_ptr<MCAsmBackend> &&AB,
                                            std::unique_ptr<MCObjectWriter> &&OW,
                                            std::unique_ptr<MCCodeEmitter> &&CE,
-                                           bool RelaxAll,
                                            bool IncrementalLinkerCompatible) {
   X86WinCOFFStreamer *S =
       new X86WinCOFFStreamer(C, std::move(AB), std::move(CE), std::move(OW));
-  S->getAssembler().setRelaxAll(RelaxAll);
   S->getAssembler().setIncrementalLinkerCompatible(IncrementalLinkerCompatible);
   return S;
 }
diff --git a/llvm/lib/Target/X86/X86CmovConversion.cpp b/llvm/lib/Target/X86/X86CmovConversion.cpp
index 8dc3b91f08e2..297acf07115a 100644
--- a/llvm/lib/Target/X86/X86CmovConversion.cpp
+++ b/llvm/lib/Target/X86/X86CmovConversion.cpp
@@ -355,7 +355,7 @@ bool X86CmovConverterPass::collectCmovCandidates(
       FoundNonCMOVInst = true;
       // Check if this instruction define EFLAGS, to determine end of processed
       // range, as there would be no more instructions using current EFLAGS def.
-      if (I.definesRegister(X86::EFLAGS)) {
+      if (I.definesRegister(X86::EFLAGS, /*TRI=*/nullptr)) {
         // Check if current processed CMOV-group should not be skipped and add
         // it as a CMOV-group-candidate.
         if (!SkipGroup)
@@ -582,7 +582,7 @@ bool X86CmovConverterPass::checkForProfitableCmovCandidates(
 }
 
 static bool checkEFLAGSLive(MachineInstr *MI) {
-  if (MI->killsRegister(X86::EFLAGS))
+  if (MI->killsRegister(X86::EFLAGS, /*TRI=*/nullptr))
     return false;
 
   // The EFLAGS operand of MI might be missing a kill marker.
@@ -592,9 +592,9 @@ static bool checkEFLAGSLive(MachineInstr *MI) {
 
   // Scan forward through BB for a use/def of EFLAGS.
   for (auto I = std::next(ItrMI), E = BB->end(); I != E; ++I) {
-    if (I->readsRegister(X86::EFLAGS))
+    if (I->readsRegister(X86::EFLAGS, /*TRI=*/nullptr))
       return true;
-    if (I->definesRegister(X86::EFLAGS))
+    if (I->definesRegister(X86::EFLAGS, /*TRI=*/nullptr))
       return false;
   }
 
diff --git a/llvm/lib/Target/X86/X86FixupSetCC.cpp b/llvm/lib/Target/X86/X86FixupSetCC.cpp
index 269f8ce6bd7a..5c7105988070 100644
--- a/llvm/lib/Target/X86/X86FixupSetCC.cpp
+++ b/llvm/lib/Target/X86/X86FixupSetCC.cpp
@@ -69,7 +69,7 @@ bool X86FixupSetCCPass::runOnMachineFunction(MachineFunction &MF) {
     MachineInstr *FlagsDefMI = nullptr;
     for (auto &MI : MBB) {
       // Remember the most recent preceding eflags defining instruction.
-      if (MI.definesRegister(X86::EFLAGS))
+      if (MI.definesRegister(X86::EFLAGS, /*TRI=*/nullptr))
         FlagsDefMI = &MI;
 
       // Find a setcc that is used by a zext.
@@ -94,7 +94,7 @@ bool X86FixupSetCCPass::runOnMachineFunction(MachineFunction &MF) {
       // it, itself, by definition, clobbers eflags. But it may happen that
       // FlagsDefMI also *uses* eflags, in which case the transformation is
       // invalid.
-      if (FlagsDefMI->readsRegister(X86::EFLAGS))
+      if (FlagsDefMI->readsRegister(X86::EFLAGS, /*TRI=*/nullptr))
         continue;
 
       // On 32-bit, we need to be careful to force an ABCD register.
diff --git a/llvm/lib/Target/X86/X86FlagsCopyLowering.cpp b/llvm/lib/Target/X86/X86FlagsCopyLowering.cpp
index d96613d7bb7e..78355d355083 100644
--- a/llvm/lib/Target/X86/X86FlagsCopyLowering.cpp
+++ b/llvm/lib/Target/X86/X86FlagsCopyLowering.cpp
@@ -442,7 +442,8 @@ bool X86FlagsCopyLoweringPass::runOnMachineFunction(MachineFunction &MF) {
           llvm::reverse(llvm::make_range(Begin, End)), [&](MachineInstr &MI) {
             // Flag any instruction (other than the copy we are
             // currently rewriting) that defs EFLAGS.
-            return &MI != CopyI && MI.findRegisterDefOperand(X86::EFLAGS);
+            return &MI != CopyI &&
+                   MI.findRegisterDefOperand(X86::EFLAGS, /*TRI=*/nullptr);
           });
     };
     auto HasEFLAGSClobberPath = [&](MachineBasicBlock *BeginMBB,
@@ -500,7 +501,7 @@ bool X86FlagsCopyLoweringPass::runOnMachineFunction(MachineFunction &MF) {
       auto DefIt = llvm::find_if(
           llvm::reverse(llvm::make_range(TestMBB->instr_begin(), TestPos)),
           [&](MachineInstr &MI) {
-            return MI.findRegisterDefOperand(X86::EFLAGS);
+            return MI.findRegisterDefOperand(X86::EFLAGS, /*TRI=*/nullptr);
           });
       if (DefIt.base() != TestMBB->instr_begin()) {
         dbgs() << "  Using EFLAGS defined by: ";
@@ -562,9 +563,10 @@ bool X86FlagsCopyLoweringPass::runOnMachineFunction(MachineFunction &MF) {
           break;
         }
 
-        MachineOperand *FlagUse = MI.findRegisterUseOperand(X86::EFLAGS);
+        MachineOperand *FlagUse =
+            MI.findRegisterUseOperand(X86::EFLAGS, /*TRI=*/nullptr);
         if (!FlagUse) {
-          if (MI.findRegisterDefOperand(X86::EFLAGS)) {
+          if (MI.findRegisterDefOperand(X86::EFLAGS, /*TRI=*/nullptr)) {
             // If EFLAGS are defined, it's as-if they were killed. We can stop
             // scanning here.
             //
@@ -615,7 +617,7 @@ bool X86FlagsCopyLoweringPass::runOnMachineFunction(MachineFunction &MF) {
           rewriteCopy(MI, *FlagUse, CopyDefI);
         } else {
           // We assume all other instructions that use flags also def them.
-          assert(MI.findRegisterDefOperand(X86::EFLAGS) &&
+          assert(MI.findRegisterDefOperand(X86::EFLAGS, /*TRI=*/nullptr) &&
                  "Expected a def of EFLAGS for this instruction!");
 
           // NB!!! Several arithmetic instructions only *partially* update
@@ -734,7 +736,7 @@ CondRegArray X86FlagsCopyLoweringPass::collectCondsInRegs(
 
     // Stop scanning when we see the first definition of the EFLAGS as prior to
     // this we would potentially capture the wrong flag state.
-    if (MI.findRegisterDefOperand(X86::EFLAGS))
+    if (MI.findRegisterDefOperand(X86::EFLAGS, /*TRI=*/nullptr))
       break;
   }
   return CondRegs;
@@ -914,7 +916,7 @@ void X86FlagsCopyLoweringPass::rewriteCondJmp(
   // Rewrite the jump to use the !ZF flag from the test, and kill its use of
   // flags afterward.
   JmpI.getOperand(1).setImm(Inverted ? X86::COND_E : X86::COND_NE);
-  JmpI.findRegisterUseOperand(X86::EFLAGS)->setIsKill(true);
+  JmpI.findRegisterUseOperand(X86::EFLAGS, /*TRI=*/nullptr)->setIsKill(true);
   LLVM_DEBUG(dbgs() << "    fixed jCC: "; JmpI.dump());
 }
 
diff --git a/llvm/lib/Target/X86/X86FloatingPoint.cpp b/llvm/lib/Target/X86/X86FloatingPoint.cpp
index 260879ffaa4f..02c3ca9839fc 100644
--- a/llvm/lib/Target/X86/X86FloatingPoint.cpp
+++ b/llvm/lib/Target/X86/X86FloatingPoint.cpp
@@ -829,7 +829,8 @@ static const TableEntry PopTable[] = {
 };
 
 static bool doesInstructionSetFPSW(MachineInstr &MI) {
-  if (const MachineOperand *MO = MI.findRegisterDefOperand(X86::FPSW))
+  if (const MachineOperand *MO =
+          MI.findRegisterDefOperand(X86::FPSW, /*TRI=*/nullptr))
     if (!MO->isDead())
       return true;
   return false;
@@ -872,7 +873,7 @@ void FPS::popStackAfter(MachineBasicBlock::iterator &I) {
     if (doesInstructionSetFPSW(MI)) {
       MachineBasicBlock &MBB = *MI.getParent();
       MachineBasicBlock::iterator Next = getNextFPInstruction(I);
-      if (Next != MBB.end() && Next->readsRegister(X86::FPSW))
+      if (Next != MBB.end() && Next->readsRegister(X86::FPSW, /*TRI=*/nullptr))
         I = Next;
     }
     I = BuildMI(*MBB, ++I, dl, TII->get(X86::ST_FPrr)).addReg(X86::ST0);
@@ -1082,9 +1083,10 @@ void FPS::handleReturn(MachineBasicBlock::iterator &I) {
     // FP Register uses must be kills unless there are two uses of the same
     // register, in which case only one will be a kill.
     assert(Op.isUse() &&
-           (Op.isKill() ||                    // Marked kill.
-            getFPReg(Op) == FirstFPRegOp ||   // Second instance.
-            MI.killsRegister(Op.getReg())) && // Later use is marked kill.
+           (Op.isKill() ||                  // Marked kill.
+            getFPReg(Op) == FirstFPRegOp || // Second instance.
+            MI.killsRegister(Op.getReg(),
+                             /*TRI=*/nullptr)) && // Later use is marked kill.
            "Ret only defs operands, and values aren't live beyond it");
 
     if (FirstFPRegOp == ~0U)
@@ -1181,7 +1183,7 @@ void FPS::handleOneArgFP(MachineBasicBlock::iterator &I) {
 
   // Is this the last use of the source register?
   unsigned Reg = getFPReg(MI.getOperand(NumOps - 1));
-  bool KillsSrc = MI.killsRegister(X86::FP0 + Reg);
+  bool KillsSrc = MI.killsRegister(X86::FP0 + Reg, /*TRI=*/nullptr);
 
   // FISTP64m is strange because there isn't a non-popping versions.
   // If we have one _and_ we don't want to pop the operand, duplicate the value
@@ -1244,7 +1246,7 @@ void FPS::handleOneArgFPRW(MachineBasicBlock::iterator &I) {
 
   // Is this the last use of the source register?
   unsigned Reg = getFPReg(MI.getOperand(1));
-  bool KillsSrc = MI.killsRegister(X86::FP0 + Reg);
+  bool KillsSrc = MI.killsRegister(X86::FP0 + Reg, /*TRI=*/nullptr);
 
   if (KillsSrc) {
     // If this is the last use of the source register, just make sure it's on
@@ -1355,8 +1357,8 @@ void FPS::handleTwoArgFP(MachineBasicBlock::iterator &I) {
   unsigned Dest = getFPReg(MI.getOperand(0));
   unsigned Op0 = getFPReg(MI.getOperand(NumOperands - 2));
   unsigned Op1 = getFPReg(MI.getOperand(NumOperands - 1));
-  bool KillsOp0 = MI.killsRegister(X86::FP0 + Op0);
-  bool KillsOp1 = MI.killsRegister(X86::FP0 + Op1);
+  bool KillsOp0 = MI.killsRegister(X86::FP0 + Op0, /*TRI=*/nullptr);
+  bool KillsOp1 = MI.killsRegister(X86::FP0 + Op1, /*TRI=*/nullptr);
   const DebugLoc &dl = MI.getDebugLoc();
 
   unsigned TOS = getStackEntry(0);
@@ -1453,8 +1455,8 @@ void FPS::handleCompareFP(MachineBasicBlock::iterator &I) {
   assert(NumOperands == 2 && "Illegal FUCOM* instruction!");
   unsigned Op0 = getFPReg(MI.getOperand(NumOperands - 2));
   unsigned Op1 = getFPReg(MI.getOperand(NumOperands - 1));
-  bool KillsOp0 = MI.killsRegister(X86::FP0 + Op0);
-  bool KillsOp1 = MI.killsRegister(X86::FP0 + Op1);
+  bool KillsOp0 = MI.killsRegister(X86::FP0 + Op0, /*TRI=*/nullptr);
+  bool KillsOp1 = MI.killsRegister(X86::FP0 + Op1, /*TRI=*/nullptr);
 
   // Make sure the first operand is on the top of stack, the other one can be
   // anywhere.
@@ -1480,7 +1482,7 @@ void FPS::handleCondMovFP(MachineBasicBlock::iterator &I) {
 
   unsigned Op0 = getFPReg(MI.getOperand(0));
   unsigned Op1 = getFPReg(MI.getOperand(2));
-  bool KillsOp1 = MI.killsRegister(X86::FP0 + Op1);
+  bool KillsOp1 = MI.killsRegister(X86::FP0 + Op1, /*TRI=*/nullptr);
 
   // The first operand *must* be on the top of the stack.
   moveToTop(Op0, I);
@@ -1524,7 +1526,7 @@ void FPS::handleSpecialFP(MachineBasicBlock::iterator &Inst) {
     // We handle three kinds of copies: FP <- FP, FP <- ST, and ST <- FP.
     const MachineOperand &MO1 = MI.getOperand(1);
     const MachineOperand &MO0 = MI.getOperand(0);
-    bool KillsSrc = MI.killsRegister(MO1.getReg());
+    bool KillsSrc = MI.killsRegister(MO1.getReg(), /*TRI=*/nullptr);
 
     // FP <- FP copy.
     unsigned DstFP = getFPReg(MO0);
diff --git a/llvm/lib/Target/X86/X86ISelLowering.cpp b/llvm/lib/Target/X86/X86ISelLowering.cpp
index dd40d079c7e2..a811ce43422e 100644
--- a/llvm/lib/Target/X86/X86ISelLowering.cpp
+++ b/llvm/lib/Target/X86/X86ISelLowering.cpp
@@ -4441,10 +4441,8 @@ static SDValue concatSubVectors(SDValue V1, SDValue V2, SelectionDAG &DAG,
 static SDValue getOnesVector(EVT VT, SelectionDAG &DAG, const SDLoc &dl) {
   assert((VT.is128BitVector() || VT.is256BitVector() || VT.is512BitVector()) &&
          "Expected a 128/256/512-bit vector type");
-
-  APInt Ones = APInt::getAllOnes(32);
   unsigned NumElts = VT.getSizeInBits() / 32;
-  SDValue Vec = DAG.getConstant(Ones, dl, MVT::getVectorVT(MVT::i32, NumElts));
+  SDValue Vec = DAG.getAllOnesConstant(dl, MVT::getVectorVT(MVT::i32, NumElts));
   return DAG.getBitcast(VT, Vec);
 }
 
@@ -5179,9 +5177,13 @@ static void getHorizDemandedElts(EVT VT, const APInt &DemandedElts,
 /// IsUnary for shuffles which use a single input multiple times, and in those
 /// cases it will adjust the mask to only have indices within that single input.
 /// It is an error to call this with non-empty Mask/Ops vectors.
-static bool getTargetShuffleMask(SDNode *N, MVT VT, bool AllowSentinelZero,
+static bool getTargetShuffleMask(SDValue N, bool AllowSentinelZero,
                                  SmallVectorImpl<SDValue> &Ops,
                                  SmallVectorImpl<int> &Mask, bool &IsUnary) {
+  if (!isTargetShuffle(N.getOpcode()))
+    return false;
+
+  MVT VT = N.getSimpleValueType();
   unsigned NumElems = VT.getVectorNumElements();
   unsigned MaskEltSize = VT.getScalarSizeInBits();
   SmallVector<uint64_t, 32> RawMask;
@@ -5193,129 +5195,129 @@ static bool getTargetShuffleMask(SDNode *N, MVT VT, bool AllowSentinelZero,
 
   IsUnary = false;
   bool IsFakeUnary = false;
-  switch (N->getOpcode()) {
+  switch (N.getOpcode()) {
   case X86ISD::BLENDI:
-    assert(N->getOperand(0).getValueType() == VT && "Unexpected value type");
-    assert(N->getOperand(1).getValueType() == VT && "Unexpected value type");
-    ImmN = N->getConstantOperandVal(N->getNumOperands() - 1);
+    assert(N.getOperand(0).getValueType() == VT && "Unexpected value type");
+    assert(N.getOperand(1).getValueType() == VT && "Unexpected value type");
+    ImmN = N.getConstantOperandVal(N.getNumOperands() - 1);
     DecodeBLENDMask(NumElems, ImmN, Mask);
-    IsUnary = IsFakeUnary = N->getOperand(0) == N->getOperand(1);
+    IsUnary = IsFakeUnary = N.getOperand(0) == N.getOperand(1);
     break;
   case X86ISD::SHUFP:
-    assert(N->getOperand(0).getValueType() == VT && "Unexpected value type");
-    assert(N->getOperand(1).getValueType() == VT && "Unexpected value type");
-    ImmN = N->getConstantOperandVal(N->getNumOperands() - 1);
+    assert(N.getOperand(0).getValueType() == VT && "Unexpected value type");
+    assert(N.getOperand(1).getValueType() == VT && "Unexpected value type");
+    ImmN = N.getConstantOperandVal(N.getNumOperands() - 1);
     DecodeSHUFPMask(NumElems, MaskEltSize, ImmN, Mask);
-    IsUnary = IsFakeUnary = N->getOperand(0) == N->getOperand(1);
+    IsUnary = IsFakeUnary = N.getOperand(0) == N.getOperand(1);
     break;
   case X86ISD::INSERTPS:
-    assert(N->getOperand(0).getValueType() == VT && "Unexpected value type");
-    assert(N->getOperand(1).getValueType() == VT && "Unexpected value type");
-    ImmN = N->getConstantOperandVal(N->getNumOperands() - 1);
+    assert(N.getOperand(0).getValueType() == VT && "Unexpected value type");
+    assert(N.getOperand(1).getValueType() == VT && "Unexpected value type");
+    ImmN = N.getConstantOperandVal(N.getNumOperands() - 1);
     DecodeINSERTPSMask(ImmN, Mask);
-    IsUnary = IsFakeUnary = N->getOperand(0) == N->getOperand(1);
+    IsUnary = IsFakeUnary = N.getOperand(0) == N.getOperand(1);
     break;
   case X86ISD::EXTRQI:
-    assert(N->getOperand(0).getValueType() == VT && "Unexpected value type");
-    if (isa<ConstantSDNode>(N->getOperand(1)) &&
-        isa<ConstantSDNode>(N->getOperand(2))) {
-      int BitLen = N->getConstantOperandVal(1);
-      int BitIdx = N->getConstantOperandVal(2);
+    assert(N.getOperand(0).getValueType() == VT && "Unexpected value type");
+    if (isa<ConstantSDNode>(N.getOperand(1)) &&
+        isa<ConstantSDNode>(N.getOperand(2))) {
+      int BitLen = N.getConstantOperandVal(1);
+      int BitIdx = N.getConstantOperandVal(2);
       DecodeEXTRQIMask(NumElems, MaskEltSize, BitLen, BitIdx, Mask);
       IsUnary = true;
     }
     break;
   case X86ISD::INSERTQI:
-    assert(N->getOperand(0).getValueType() == VT && "Unexpected value type");
-    assert(N->getOperand(1).getValueType() == VT && "Unexpected value type");
-    if (isa<ConstantSDNode>(N->getOperand(2)) &&
-        isa<ConstantSDNode>(N->getOperand(3))) {
-      int BitLen = N->getConstantOperandVal(2);
-      int BitIdx = N->getConstantOperandVal(3);
+    assert(N.getOperand(0).getValueType() == VT && "Unexpected value type");
+    assert(N.getOperand(1).getValueType() == VT && "Unexpected value type");
+    if (isa<ConstantSDNode>(N.getOperand(2)) &&
+        isa<ConstantSDNode>(N.getOperand(3))) {
+      int BitLen = N.getConstantOperandVal(2);
+      int BitIdx = N.getConstantOperandVal(3);
       DecodeINSERTQIMask(NumElems, MaskEltSize, BitLen, BitIdx, Mask);
-      IsUnary = IsFakeUnary = N->getOperand(0) == N->getOperand(1);
+      IsUnary = IsFakeUnary = N.getOperand(0) == N.getOperand(1);
     }
     break;
   case X86ISD::UNPCKH:
-    assert(N->getOperand(0).getValueType() == VT && "Unexpected value type");
-    assert(N->getOperand(1).getValueType() == VT && "Unexpected value type");
+    assert(N.getOperand(0).getValueType() == VT && "Unexpected value type");
+    assert(N.getOperand(1).getValueType() == VT && "Unexpected value type");
     DecodeUNPCKHMask(NumElems, MaskEltSize, Mask);
-    IsUnary = IsFakeUnary = N->getOperand(0) == N->getOperand(1);
+    IsUnary = IsFakeUnary = N.getOperand(0) == N.getOperand(1);
     break;
   case X86ISD::UNPCKL:
-    assert(N->getOperand(0).getValueType() == VT && "Unexpected value type");
-    assert(N->getOperand(1).getValueType() == VT && "Unexpected value type");
+    assert(N.getOperand(0).getValueType() == VT && "Unexpected value type");
+    assert(N.getOperand(1).getValueType() == VT && "Unexpected value type");
     DecodeUNPCKLMask(NumElems, MaskEltSize, Mask);
-    IsUnary = IsFakeUnary = N->getOperand(0) == N->getOperand(1);
+    IsUnary = IsFakeUnary = N.getOperand(0) == N.getOperand(1);
     break;
   case X86ISD::MOVHLPS:
-    assert(N->getOperand(0).getValueType() == VT && "Unexpected value type");
-    assert(N->getOperand(1).getValueType() == VT && "Unexpected value type");
+    assert(N.getOperand(0).getValueType() == VT && "Unexpected value type");
+    assert(N.getOperand(1).getValueType() == VT && "Unexpected value type");
     DecodeMOVHLPSMask(NumElems, Mask);
-    IsUnary = IsFakeUnary = N->getOperand(0) == N->getOperand(1);
+    IsUnary = IsFakeUnary = N.getOperand(0) == N.getOperand(1);
     break;
   case X86ISD::MOVLHPS:
-    assert(N->getOperand(0).getValueType() == VT && "Unexpected value type");
-    assert(N->getOperand(1).getValueType() == VT && "Unexpected value type");
+    assert(N.getOperand(0).getValueType() == VT && "Unexpected value type");
+    assert(N.getOperand(1).getValueType() == VT && "Unexpected value type");
     DecodeMOVLHPSMask(NumElems, Mask);
-    IsUnary = IsFakeUnary = N->getOperand(0) == N->getOperand(1);
+    IsUnary = IsFakeUnary = N.getOperand(0) == N.getOperand(1);
     break;
   case X86ISD::VALIGN:
     assert((VT.getScalarType() == MVT::i32 || VT.getScalarType() == MVT::i64) &&
            "Only 32-bit and 64-bit elements are supported!");
-    assert(N->getOperand(0).getValueType() == VT && "Unexpected value type");
-    assert(N->getOperand(1).getValueType() == VT && "Unexpected value type");
-    ImmN = N->getConstantOperandVal(N->getNumOperands() - 1);
+    assert(N.getOperand(0).getValueType() == VT && "Unexpected value type");
+    assert(N.getOperand(1).getValueType() == VT && "Unexpected value type");
+    ImmN = N.getConstantOperandVal(N.getNumOperands() - 1);
     DecodeVALIGNMask(NumElems, ImmN, Mask);
-    IsUnary = IsFakeUnary = N->getOperand(0) == N->getOperand(1);
-    Ops.push_back(N->getOperand(1));
-    Ops.push_back(N->getOperand(0));
+    IsUnary = IsFakeUnary = N.getOperand(0) == N.getOperand(1);
+    Ops.push_back(N.getOperand(1));
+    Ops.push_back(N.getOperand(0));
     break;
   case X86ISD::PALIGNR:
     assert(VT.getScalarType() == MVT::i8 && "Byte vector expected");
-    assert(N->getOperand(0).getValueType() == VT && "Unexpected value type");
-    assert(N->getOperand(1).getValueType() == VT && "Unexpected value type");
-    ImmN = N->getConstantOperandVal(N->getNumOperands() - 1);
+    assert(N.getOperand(0).getValueType() == VT && "Unexpected value type");
+    assert(N.getOperand(1).getValueType() == VT && "Unexpected value type");
+    ImmN = N.getConstantOperandVal(N.getNumOperands() - 1);
     DecodePALIGNRMask(NumElems, ImmN, Mask);
-    IsUnary = IsFakeUnary = N->getOperand(0) == N->getOperand(1);
-    Ops.push_back(N->getOperand(1));
-    Ops.push_back(N->getOperand(0));
+    IsUnary = IsFakeUnary = N.getOperand(0) == N.getOperand(1);
+    Ops.push_back(N.getOperand(1));
+    Ops.push_back(N.getOperand(0));
     break;
   case X86ISD::VSHLDQ:
     assert(VT.getScalarType() == MVT::i8 && "Byte vector expected");
-    assert(N->getOperand(0).getValueType() == VT && "Unexpected value type");
-    ImmN = N->getConstantOperandVal(N->getNumOperands() - 1);
+    assert(N.getOperand(0).getValueType() == VT && "Unexpected value type");
+    ImmN = N.getConstantOperandVal(N.getNumOperands() - 1);
     DecodePSLLDQMask(NumElems, ImmN, Mask);
     IsUnary = true;
     break;
   case X86ISD::VSRLDQ:
     assert(VT.getScalarType() == MVT::i8 && "Byte vector expected");
-    assert(N->getOperand(0).getValueType() == VT && "Unexpected value type");
-    ImmN = N->getConstantOperandVal(N->getNumOperands() - 1);
+    assert(N.getOperand(0).getValueType() == VT && "Unexpected value type");
+    ImmN = N.getConstantOperandVal(N.getNumOperands() - 1);
     DecodePSRLDQMask(NumElems, ImmN, Mask);
     IsUnary = true;
     break;
   case X86ISD::PSHUFD:
   case X86ISD::VPERMILPI:
-    assert(N->getOperand(0).getValueType() == VT && "Unexpected value type");
-    ImmN = N->getConstantOperandVal(N->getNumOperands() - 1);
+    assert(N.getOperand(0).getValueType() == VT && "Unexpected value type");
+    ImmN = N.getConstantOperandVal(N.getNumOperands() - 1);
     DecodePSHUFMask(NumElems, MaskEltSize, ImmN, Mask);
     IsUnary = true;
     break;
   case X86ISD::PSHUFHW:
-    assert(N->getOperand(0).getValueType() == VT && "Unexpected value type");
-    ImmN = N->getConstantOperandVal(N->getNumOperands() - 1);
+    assert(N.getOperand(0).getValueType() == VT && "Unexpected value type");
+    ImmN = N.getConstantOperandVal(N.getNumOperands() - 1);
     DecodePSHUFHWMask(NumElems, ImmN, Mask);
     IsUnary = true;
     break;
   case X86ISD::PSHUFLW:
-    assert(N->getOperand(0).getValueType() == VT && "Unexpected value type");
-    ImmN = N->getConstantOperandVal(N->getNumOperands() - 1);
+    assert(N.getOperand(0).getValueType() == VT && "Unexpected value type");
+    ImmN = N.getConstantOperandVal(N.getNumOperands() - 1);
     DecodePSHUFLWMask(NumElems, ImmN, Mask);
     IsUnary = true;
     break;
   case X86ISD::VZEXT_MOVL:
-    assert(N->getOperand(0).getValueType() == VT && "Unexpected value type");
+    assert(N.getOperand(0).getValueType() == VT && "Unexpected value type");
     DecodeZeroMoveLowMask(NumElems, Mask);
     IsUnary = true;
     break;
@@ -5323,16 +5325,16 @@ static bool getTargetShuffleMask(SDNode *N, MVT VT, bool AllowSentinelZero,
     // We only decode broadcasts of same-sized vectors, peeking through to
     // extracted subvectors is likely to cause hasOneUse issues with
     // SimplifyDemandedBits etc.
-    if (N->getOperand(0).getValueType() == VT) {
+    if (N.getOperand(0).getValueType() == VT) {
       DecodeVectorBroadcast(NumElems, Mask);
       IsUnary = true;
       break;
     }
     return false;
   case X86ISD::VPERMILPV: {
-    assert(N->getOperand(0).getValueType() == VT && "Unexpected value type");
+    assert(N.getOperand(0).getValueType() == VT && "Unexpected value type");
     IsUnary = true;
-    SDValue MaskNode = N->getOperand(1);
+    SDValue MaskNode = N.getOperand(1);
     if (getTargetShuffleMaskIndices(MaskNode, MaskEltSize, RawMask,
                                     RawUndefs)) {
       DecodeVPERMILPMask(NumElems, MaskEltSize, RawMask, RawUndefs, Mask);
@@ -5342,10 +5344,10 @@ static bool getTargetShuffleMask(SDNode *N, MVT VT, bool AllowSentinelZero,
   }
   case X86ISD::PSHUFB: {
     assert(VT.getScalarType() == MVT::i8 && "Byte vector expected");
-    assert(N->getOperand(0).getValueType() == VT && "Unexpected value type");
-    assert(N->getOperand(1).getValueType() == VT && "Unexpected value type");
+    assert(N.getOperand(0).getValueType() == VT && "Unexpected value type");
+    assert(N.getOperand(1).getValueType() == VT && "Unexpected value type");
     IsUnary = true;
-    SDValue MaskNode = N->getOperand(1);
+    SDValue MaskNode = N.getOperand(1);
     if (getTargetShuffleMaskIndices(MaskNode, 8, RawMask, RawUndefs)) {
       DecodePSHUFBMask(RawMask, RawUndefs, Mask);
       break;
@@ -5353,53 +5355,53 @@ static bool getTargetShuffleMask(SDNode *N, MVT VT, bool AllowSentinelZero,
     return false;
   }
   case X86ISD::VPERMI:
-    assert(N->getOperand(0).getValueType() == VT && "Unexpected value type");
-    ImmN = N->getConstantOperandVal(N->getNumOperands() - 1);
+    assert(N.getOperand(0).getValueType() == VT && "Unexpected value type");
+    ImmN = N.getConstantOperandVal(N.getNumOperands() - 1);
     DecodeVPERMMask(NumElems, ImmN, Mask);
     IsUnary = true;
     break;
   case X86ISD::MOVSS:
   case X86ISD::MOVSD:
   case X86ISD::MOVSH:
-    assert(N->getOperand(0).getValueType() == VT && "Unexpected value type");
-    assert(N->getOperand(1).getValueType() == VT && "Unexpected value type");
+    assert(N.getOperand(0).getValueType() == VT && "Unexpected value type");
+    assert(N.getOperand(1).getValueType() == VT && "Unexpected value type");
     DecodeScalarMoveMask(NumElems, /* IsLoad */ false, Mask);
     break;
   case X86ISD::VPERM2X128:
-    assert(N->getOperand(0).getValueType() == VT && "Unexpected value type");
-    assert(N->getOperand(1).getValueType() == VT && "Unexpected value type");
-    ImmN = N->getConstantOperandVal(N->getNumOperands() - 1);
+    assert(N.getOperand(0).getValueType() == VT && "Unexpected value type");
+    assert(N.getOperand(1).getValueType() == VT && "Unexpected value type");
+    ImmN = N.getConstantOperandVal(N.getNumOperands() - 1);
     DecodeVPERM2X128Mask(NumElems, ImmN, Mask);
-    IsUnary = IsFakeUnary = N->getOperand(0) == N->getOperand(1);
+    IsUnary = IsFakeUnary = N.getOperand(0) == N.getOperand(1);
     break;
   case X86ISD::SHUF128:
-    assert(N->getOperand(0).getValueType() == VT && "Unexpected value type");
-    assert(N->getOperand(1).getValueType() == VT && "Unexpected value type");
-    ImmN = N->getConstantOperandVal(N->getNumOperands() - 1);
+    assert(N.getOperand(0).getValueType() == VT && "Unexpected value type");
+    assert(N.getOperand(1).getValueType() == VT && "Unexpected value type");
+    ImmN = N.getConstantOperandVal(N.getNumOperands() - 1);
     decodeVSHUF64x2FamilyMask(NumElems, MaskEltSize, ImmN, Mask);
-    IsUnary = IsFakeUnary = N->getOperand(0) == N->getOperand(1);
+    IsUnary = IsFakeUnary = N.getOperand(0) == N.getOperand(1);
     break;
   case X86ISD::MOVSLDUP:
-    assert(N->getOperand(0).getValueType() == VT && "Unexpected value type");
+    assert(N.getOperand(0).getValueType() == VT && "Unexpected value type");
     DecodeMOVSLDUPMask(NumElems, Mask);
     IsUnary = true;
     break;
   case X86ISD::MOVSHDUP:
-    assert(N->getOperand(0).getValueType() == VT && "Unexpected value type");
+    assert(N.getOperand(0).getValueType() == VT && "Unexpected value type");
     DecodeMOVSHDUPMask(NumElems, Mask);
     IsUnary = true;
     break;
   case X86ISD::MOVDDUP:
-    assert(N->getOperand(0).getValueType() == VT && "Unexpected value type");
+    assert(N.getOperand(0).getValueType() == VT && "Unexpected value type");
     DecodeMOVDDUPMask(NumElems, Mask);
     IsUnary = true;
     break;
   case X86ISD::VPERMIL2: {
-    assert(N->getOperand(0).getValueType() == VT && "Unexpected value type");
-    assert(N->getOperand(1).getValueType() == VT && "Unexpected value type");
-    IsUnary = IsFakeUnary = N->getOperand(0) == N->getOperand(1);
-    SDValue MaskNode = N->getOperand(2);
-    SDValue CtrlNode = N->getOperand(3);
+    assert(N.getOperand(0).getValueType() == VT && "Unexpected value type");
+    assert(N.getOperand(1).getValueType() == VT && "Unexpected value type");
+    IsUnary = IsFakeUnary = N.getOperand(0) == N.getOperand(1);
+    SDValue MaskNode = N.getOperand(2);
+    SDValue CtrlNode = N.getOperand(3);
     if (ConstantSDNode *CtrlOp = dyn_cast<ConstantSDNode>(CtrlNode)) {
       unsigned CtrlImm = CtrlOp->getZExtValue();
       if (getTargetShuffleMaskIndices(MaskNode, MaskEltSize, RawMask,
@@ -5412,10 +5414,10 @@ static bool getTargetShuffleMask(SDNode *N, MVT VT, bool AllowSentinelZero,
     return false;
   }
   case X86ISD::VPPERM: {
-    assert(N->getOperand(0).getValueType() == VT && "Unexpected value type");
-    assert(N->getOperand(1).getValueType() == VT && "Unexpected value type");
-    IsUnary = IsFakeUnary = N->getOperand(0) == N->getOperand(1);
-    SDValue MaskNode = N->getOperand(2);
+    assert(N.getOperand(0).getValueType() == VT && "Unexpected value type");
+    assert(N.getOperand(1).getValueType() == VT && "Unexpected value type");
+    IsUnary = IsFakeUnary = N.getOperand(0) == N.getOperand(1);
+    SDValue MaskNode = N.getOperand(2);
     if (getTargetShuffleMaskIndices(MaskNode, 8, RawMask, RawUndefs)) {
       DecodeVPPERMMask(RawMask, RawUndefs, Mask);
       break;
@@ -5423,11 +5425,11 @@ static bool getTargetShuffleMask(SDNode *N, MVT VT, bool AllowSentinelZero,
     return false;
   }
   case X86ISD::VPERMV: {
-    assert(N->getOperand(1).getValueType() == VT && "Unexpected value type");
+    assert(N.getOperand(1).getValueType() == VT && "Unexpected value type");
     IsUnary = true;
     // Unlike most shuffle nodes, VPERMV's mask operand is operand 0.
-    Ops.push_back(N->getOperand(1));
-    SDValue MaskNode = N->getOperand(0);
+    Ops.push_back(N.getOperand(1));
+    SDValue MaskNode = N.getOperand(0);
     if (getTargetShuffleMaskIndices(MaskNode, MaskEltSize, RawMask,
                                     RawUndefs)) {
       DecodeVPERMVMask(RawMask, RawUndefs, Mask);
@@ -5436,13 +5438,13 @@ static bool getTargetShuffleMask(SDNode *N, MVT VT, bool AllowSentinelZero,
     return false;
   }
   case X86ISD::VPERMV3: {
-    assert(N->getOperand(0).getValueType() == VT && "Unexpected value type");
-    assert(N->getOperand(2).getValueType() == VT && "Unexpected value type");
-    IsUnary = IsFakeUnary = N->getOperand(0) == N->getOperand(2);
+    assert(N.getOperand(0).getValueType() == VT && "Unexpected value type");
+    assert(N.getOperand(2).getValueType() == VT && "Unexpected value type");
+    IsUnary = IsFakeUnary = N.getOperand(0) == N.getOperand(2);
     // Unlike most shuffle nodes, VPERMV3's mask operand is the middle one.
-    Ops.push_back(N->getOperand(0));
-    Ops.push_back(N->getOperand(2));
-    SDValue MaskNode = N->getOperand(1);
+    Ops.push_back(N.getOperand(0));
+    Ops.push_back(N.getOperand(2));
+    SDValue MaskNode = N.getOperand(1);
     if (getTargetShuffleMaskIndices(MaskNode, MaskEltSize, RawMask,
                                     RawUndefs)) {
       DecodeVPERMV3Mask(RawMask, RawUndefs, Mask);
@@ -5450,7 +5452,8 @@ static bool getTargetShuffleMask(SDNode *N, MVT VT, bool AllowSentinelZero,
     }
     return false;
   }
-  default: llvm_unreachable("unknown target shuffle node");
+  default:
+    llvm_unreachable("unknown target shuffle node");
   }
 
   // Empty mask indicates the decode failed.
@@ -5472,20 +5475,20 @@ static bool getTargetShuffleMask(SDNode *N, MVT VT, bool AllowSentinelZero,
   // If we didn't already add operands in the opcode-specific code, default to
   // adding 1 or 2 operands starting at 0.
   if (Ops.empty()) {
-    Ops.push_back(N->getOperand(0));
+    Ops.push_back(N.getOperand(0));
     if (!IsUnary || IsFakeUnary)
-      Ops.push_back(N->getOperand(1));
+      Ops.push_back(N.getOperand(1));
   }
 
   return true;
 }
 
 // Wrapper for getTargetShuffleMask with InUnary;
-static bool getTargetShuffleMask(SDNode *N, MVT VT, bool AllowSentinelZero,
+static bool getTargetShuffleMask(SDValue N, bool AllowSentinelZero,
                                  SmallVectorImpl<SDValue> &Ops,
                                  SmallVectorImpl<int> &Mask) {
   bool IsUnary;
-  return getTargetShuffleMask(N, VT, AllowSentinelZero, Ops, Mask, IsUnary);
+  return getTargetShuffleMask(N, AllowSentinelZero, Ops, Mask, IsUnary);
 }
 
 /// Compute whether each element of a shuffle is zeroable.
@@ -5587,7 +5590,7 @@ static bool getTargetShuffleAndZeroables(SDValue N, SmallVectorImpl<int> &Mask,
     return false;
 
   MVT VT = N.getSimpleValueType();
-  if (!getTargetShuffleMask(N.getNode(), VT, true, Ops, Mask, IsUnary))
+  if (!getTargetShuffleMask(N, true, Ops, Mask, IsUnary))
     return false;
 
   int Size = Mask.size();
@@ -6372,8 +6375,7 @@ static SDValue getShuffleScalarElt(SDValue Op, unsigned Index,
     int NumElems = (int)ShufVT.getVectorNumElements();
     SmallVector<int, 16> ShuffleMask;
     SmallVector<SDValue, 16> ShuffleOps;
-    if (!getTargetShuffleMask(Op.getNode(), ShufVT, true, ShuffleOps,
-                              ShuffleMask))
+    if (!getTargetShuffleMask(Op, true, ShuffleOps, ShuffleMask))
       return SDValue();
 
     int Elt = ShuffleMask[Index];
@@ -20390,14 +20392,16 @@ static SDValue matchTruncateWithPACK(unsigned &PackOpcode, EVT DstVT,
   EVT SrcVT = In.getValueType();
   EVT DstSVT = DstVT.getVectorElementType();
   EVT SrcSVT = SrcVT.getVectorElementType();
+  unsigned NumDstEltBits = DstSVT.getSizeInBits();
+  unsigned NumSrcEltBits = SrcSVT.getSizeInBits();
 
   // Check we have a truncation suited for PACKSS/PACKUS.
   if (!((SrcSVT == MVT::i16 || SrcSVT == MVT::i32 || SrcSVT == MVT::i64) &&
         (DstSVT == MVT::i8 || DstSVT == MVT::i16 || DstSVT == MVT::i32)))
     return SDValue();
 
-  assert(SrcSVT.getSizeInBits() > DstSVT.getSizeInBits() && "Bad truncation");
-  unsigned NumStages = Log2_32(SrcSVT.getSizeInBits() / DstSVT.getSizeInBits());
+  assert(NumSrcEltBits > NumDstEltBits && "Bad truncation");
+  unsigned NumStages = Log2_32(NumSrcEltBits / NumDstEltBits);
 
   // Truncation from 128-bit to vXi32 can be better handled with PSHUFD.
   // Truncation to sub-64-bit vXi16 can be better handled with PSHUFD/PSHUFLW.
@@ -20418,8 +20422,7 @@ static SDValue matchTruncateWithPACK(unsigned &PackOpcode, EVT DstVT,
   if (Subtarget.hasAVX512() && NumStages > 1)
     return SDValue();
 
-  unsigned NumSrcEltBits = SrcVT.getScalarSizeInBits();
-  unsigned NumPackedSignBits = std::min<unsigned>(DstSVT.getSizeInBits(), 16);
+  unsigned NumPackedSignBits = std::min<unsigned>(NumDstEltBits, 16);
   unsigned NumPackedZeroBits = Subtarget.hasSSE41() ? NumPackedSignBits : 8;
 
   // Truncate with PACKUS if we are truncating a vector with leading zero
@@ -20441,7 +20444,7 @@ static SDValue matchTruncateWithPACK(unsigned &PackOpcode, EVT DstVT,
   // a sign splat (or AVX512 VPSRAQ support). ComputeNumSignBits struggles to
   // see through BITCASTs later on and combines/simplifications can't then use
   // it.
-  if (DstSVT == MVT::i32 && NumSignBits != SrcSVT.getSizeInBits() &&
+  if (DstSVT == MVT::i32 && NumSignBits != NumSrcEltBits &&
       !Subtarget.hasAVX512())
     return SDValue();
 
@@ -20455,8 +20458,7 @@ static SDValue matchTruncateWithPACK(unsigned &PackOpcode, EVT DstVT,
   // the truncation then we can use PACKSS by converting the srl to a sra.
   // SimplifyDemandedBits often relaxes sra to srl so we need to reverse it.
   if (In.getOpcode() == ISD::SRL && In->hasOneUse())
-    if (const APInt *ShAmt = DAG.getValidShiftAmountConstant(
-            In, APInt::getAllOnes(SrcVT.getVectorNumElements()))) {
+    if (const APInt *ShAmt = DAG.getValidShiftAmountConstant(In)) {
       if (*ShAmt == MinSignBits) {
         PackOpcode = X86ISD::PACKSS;
         return DAG.getNode(ISD::SRA, DL, SrcVT, In->ops());
@@ -24137,8 +24139,7 @@ SDValue X86TargetLowering::LowerSELECT(SDValue Op, SelectionDAG &DAG) const {
               DAG.getConstant(1, DL, VT));
         else
           Neg = CmpOp0;
-        SDValue Mask = DAG.getNode(ISD::SUB, DL, VT, DAG.getConstant(0, DL, VT),
-                                   Neg); // -(and (x, 0x1))
+        SDValue Mask = DAG.getNegative(Neg, DL, VT); // -(and (x, 0x1))
         SDValue And = DAG.getNode(ISD::AND, DL, VT, Mask, Src1); // Mask & z
         return DAG.getNode(Op2.getOpcode(), DL, VT, And, Src2);  // And Op y
       }
@@ -27893,7 +27894,7 @@ static SDValue LowerVectorCTLZInRegLUT(SDValue Op, const SDLoc &DL,
   SDValue InRegLUT = DAG.getBuildVector(CurrVT, DL, LUTVec);
 
   // Begin by bitcasting the input to byte vector, then split those bytes
-  // into lo/hi nibbles and use the PSHUFB LUT to perform CLTZ on each of them.
+  // into lo/hi nibbles and use the PSHUFB LUT to perform CTLZ on each of them.
   // If the hi input nibble is zero then we add both results together, otherwise
   // we just take the hi result (by masking the lo result to zero before the
   // add).
@@ -28147,9 +28148,8 @@ static SDValue LowerABS(SDValue Op, const X86Subtarget &Subtarget,
   // ABS(vXi64 X) --> VPBLENDVPD(X, 0-X, X).
   if ((VT == MVT::v2i64 || VT == MVT::v4i64) && Subtarget.hasSSE41()) {
     SDValue Src = Op.getOperand(0);
-    SDValue Sub =
-        DAG.getNode(ISD::SUB, DL, VT, DAG.getConstant(0, DL, VT), Src);
-    return DAG.getNode(X86ISD::BLENDV, DL, VT, Src, Sub, Src);
+    SDValue Neg = DAG.getNegative(Src, DL, VT);
+    return DAG.getNode(X86ISD::BLENDV, DL, VT, Src, Neg, Src);
   }
 
   if (VT.is256BitVector() && !Subtarget.hasInt256()) {
@@ -29370,10 +29370,8 @@ static SDValue LowerShift(SDValue Op, const X86Subtarget &Subtarget,
   // +ve/-ve Amt = shift left/right.
   if (Subtarget.hasXOP() && (VT == MVT::v2i64 || VT == MVT::v4i32 ||
                              VT == MVT::v8i16 || VT == MVT::v16i8)) {
-    if (Opc == ISD::SRL || Opc == ISD::SRA) {
-      SDValue Zero = DAG.getConstant(0, dl, VT);
-      Amt = DAG.getNode(ISD::SUB, dl, VT, Zero, Amt);
-    }
+    if (Opc == ISD::SRL || Opc == ISD::SRA)
+      Amt = DAG.getNegative(Amt, dl, VT);
     if (Opc == ISD::SHL || Opc == ISD::SRL)
       return DAG.getNode(X86ISD::VPSHL, dl, VT, R, Amt);
     if (Opc == ISD::SRA)
@@ -34224,10 +34222,10 @@ static bool isEFLAGSLiveAfter(MachineBasicBlock::iterator Itr,
                               MachineBasicBlock *BB) {
   // Scan forward through BB for a use/def of EFLAGS.
   for (const MachineInstr &mi : llvm::make_range(std::next(Itr), BB->end())) {
-    if (mi.readsRegister(X86::EFLAGS))
+    if (mi.readsRegister(X86::EFLAGS, /*TRI=*/nullptr))
       return true;
     // If we found a def, we can stop searching.
-    if (mi.definesRegister(X86::EFLAGS))
+    if (mi.definesRegister(X86::EFLAGS, /*TRI=*/nullptr))
       return false;
   }
 
@@ -34813,7 +34811,7 @@ X86TargetLowering::EmitLoweredCascadedSelect(MachineInstr &FirstCMOV,
   // If the EFLAGS register isn't dead in the terminator, then claim that it's
   // live into the sink and copy blocks.
   const TargetRegisterInfo *TRI = Subtarget.getRegisterInfo();
-  if (!SecondCascadedCMOV.killsRegister(X86::EFLAGS) &&
+  if (!SecondCascadedCMOV.killsRegister(X86::EFLAGS, /*TRI=*/nullptr) &&
       !checkAndUpdateEFLAGSKill(SecondCascadedCMOV, ThisMBB, TRI)) {
     SecondInsertedMBB->addLiveIn(X86::EFLAGS);
     SinkMBB->addLiveIn(X86::EFLAGS);
@@ -34969,7 +34967,7 @@ X86TargetLowering::EmitLoweredSelect(MachineInstr &MI,
   // If the EFLAGS register isn't dead in the terminator, then claim that it's
   // live into the sink and copy blocks.
   const TargetRegisterInfo *TRI = Subtarget.getRegisterInfo();
-  if (!LastCMOV->killsRegister(X86::EFLAGS) &&
+  if (!LastCMOV->killsRegister(X86::EFLAGS, /*TRI=*/nullptr) &&
       !checkAndUpdateEFLAGSKill(LastCMOV, ThisMBB, TRI)) {
     FalseMBB->addLiveIn(X86::EFLAGS);
     SinkMBB->addLiveIn(X86::EFLAGS);
@@ -36548,10 +36546,11 @@ X86TargetLowering::EmitInstrWithCustomInserter(MachineInstr &MI,
     // four operand definitions that are E[ABCD] registers. We skip them and
     // then insert the LEA.
     MachineBasicBlock::reverse_iterator RMBBI(MI.getReverseIterator());
-    while (RMBBI != BB->rend() && (RMBBI->definesRegister(X86::EAX) ||
-                                   RMBBI->definesRegister(X86::EBX) ||
-                                   RMBBI->definesRegister(X86::ECX) ||
-                                   RMBBI->definesRegister(X86::EDX))) {
+    while (RMBBI != BB->rend() &&
+           (RMBBI->definesRegister(X86::EAX, /*TRI=*/nullptr) ||
+            RMBBI->definesRegister(X86::EBX, /*TRI=*/nullptr) ||
+            RMBBI->definesRegister(X86::ECX, /*TRI=*/nullptr) ||
+            RMBBI->definesRegister(X86::EDX, /*TRI=*/nullptr))) {
       ++RMBBI;
     }
     MachineBasicBlock::iterator MBBI(RMBBI);
@@ -37206,7 +37205,7 @@ void X86TargetLowering::computeKnownBitsForTargetNode(const SDValue Op,
   if (isTargetShuffle(Opc)) {
     SmallVector<int, 64> Mask;
     SmallVector<SDValue, 2> Ops;
-    if (getTargetShuffleMask(Op.getNode(), VT.getSimpleVT(), true, Ops, Mask)) {
+    if (getTargetShuffleMask(Op, true, Ops, Mask)) {
       unsigned NumOps = Ops.size();
       unsigned NumElts = VT.getVectorNumElements();
       if (Mask.size() == NumElts) {
@@ -37375,7 +37374,7 @@ unsigned X86TargetLowering::ComputeNumSignBitsForTargetNode(
   if (isTargetShuffle(Opcode)) {
     SmallVector<int, 64> Mask;
     SmallVector<SDValue, 2> Ops;
-    if (getTargetShuffleMask(Op.getNode(), VT.getSimpleVT(), true, Ops, Mask)) {
+    if (getTargetShuffleMask(Op, true, Ops, Mask)) {
       unsigned NumOps = Ops.size();
       unsigned NumElts = VT.getVectorNumElements();
       if (Mask.size() == NumElts) {
@@ -39777,8 +39776,7 @@ static SmallVector<int, 4> getPSHUFShuffleMask(SDValue N) {
   MVT VT = N.getSimpleValueType();
   SmallVector<int, 4> Mask;
   SmallVector<SDValue, 2> Ops;
-  bool HaveMask =
-      getTargetShuffleMask(N.getNode(), VT, false, Ops, Mask);
+  bool HaveMask = getTargetShuffleMask(N, false, Ops, Mask);
   (void)HaveMask;
   assert(HaveMask);
 
@@ -40047,8 +40045,7 @@ static SDValue canonicalizeShuffleWithOp(SDValue N, SelectionDAG &DAG,
     // Don't merge PSHUFB if it contains zero'd elements.
     SmallVector<int> Mask;
     SmallVector<SDValue> Ops;
-    if (!getTargetShuffleMask(N.getNode(), ShuffleVT.getSimpleVT(), false, Ops,
-                              Mask))
+    if (!getTargetShuffleMask(N, false, Ops, Mask))
       break;
     [[fallthrough]];
   }
@@ -40598,8 +40595,7 @@ static SDValue combineTargetShuffle(SDValue N, const SDLoc &DL,
       bool Updated = false;
       SmallVector<int> Mask;
       SmallVector<SDValue> Ops;
-      if (getTargetShuffleMask(N.getNode(), VT, false, Ops, Mask) &&
-          Ops.size() == 2) {
+      if (getTargetShuffleMask(N, false, Ops, Mask) && Ops.size() == 2) {
         for (int i = 0; i != 2; ++i) {
           SmallVector<SDValue> SubOps;
           SmallVector<int> SubMask, SubScaledMask;
@@ -42745,8 +42741,7 @@ bool X86TargetLowering::isGuaranteedNotToBeUndefOrPoisonForTargetNode(
   case X86ISD::VPERMILPI: {
     SmallVector<int, 8> Mask;
     SmallVector<SDValue, 2> Ops;
-    if (getTargetShuffleMask(Op.getNode(), Op.getSimpleValueType(), true, Ops,
-                             Mask)) {
+    if (getTargetShuffleMask(Op, true, Ops, Mask)) {
       SmallVector<APInt, 2> DemandedSrcElts(Ops.size(),
                                             APInt::getZero(NumElts));
       for (auto M : enumerate(Mask)) {
@@ -45272,7 +45267,7 @@ static SDValue commuteSelect(SDNode *N, SelectionDAG &DAG,
       ISD::getSetCCInverse(cast<CondCodeSDNode>(Cond.getOperand(2))->get(),
                            Cond.getOperand(0).getValueType());
   Cond = DAG.getSetCC(SDLoc(Cond), Cond.getValueType(), Cond.getOperand(0),
-		                        Cond.getOperand(1), NewCC);
+                      Cond.getOperand(1), NewCC);
   return DAG.getSelect(DL, LHS.getValueType(), Cond, RHS, LHS);
 }
 
@@ -45332,8 +45327,8 @@ static SDValue combineSelect(SDNode *N, SelectionDAG &DAG,
     SmallVector<SDValue, 1> LHSOps, RHSOps;
     SmallVector<int, 64> LHSMask, RHSMask, CondMask;
     if (createShuffleMaskFromVSELECT(CondMask, Cond) &&
-        getTargetShuffleMask(LHS.getNode(), SimpleVT, true, LHSOps, LHSMask) &&
-        getTargetShuffleMask(RHS.getNode(), SimpleVT, true, RHSOps, RHSMask)) {
+        getTargetShuffleMask(LHS, true, LHSOps, LHSMask) &&
+        getTargetShuffleMask(RHS, true, RHSOps, RHSMask)) {
       int NumElts = VT.getVectorNumElements();
       for (int i = 0; i != NumElts; ++i) {
         // getConstVector sets negative shuffle mask values as undef, so ensure
@@ -55436,8 +55431,7 @@ static SDValue combineConcatVectorOps(const SDLoc &DL, MVT VT,
         for (unsigned i = 0; i != NumOps; ++i) {
           SmallVector<int, 64> SubMask;
           SmallVector<SDValue, 2> SubOps;
-          if (!getTargetShuffleMask(Ops[i].getNode(), OpVT, false, SubOps,
-                                    SubMask))
+          if (!getTargetShuffleMask(Ops[i], false, SubOps, SubMask))
             break;
           for (int M : SubMask) {
             if (0 <= M)
@@ -55463,8 +55457,7 @@ static SDValue combineConcatVectorOps(const SDLoc &DL, MVT VT,
         for (unsigned i = 0; i != NumOps; ++i) {
           SmallVector<int, 64> SubMask;
           SmallVector<SDValue, 2> SubOps;
-          if (!getTargetShuffleMask(Ops[i].getNode(), OpVT, false, SubOps,
-                                    SubMask))
+          if (!getTargetShuffleMask(Ops[i], false, SubOps, SubMask))
             break;
           for (int M : SubMask) {
             if (0 <= M) {
diff --git a/llvm/lib/Target/X86/X86InstrInfo.cpp b/llvm/lib/Target/X86/X86InstrInfo.cpp
index 510b08f9901a..3d80c43b571f 100644
--- a/llvm/lib/Target/X86/X86InstrInfo.cpp
+++ b/llvm/lib/Target/X86/X86InstrInfo.cpp
@@ -1168,7 +1168,7 @@ bool X86InstrInfo::classifyLEAReg(MachineInstr &MI, const MachineOperand &Src,
     RC = Opc != X86::LEA32r ? &X86::GR64_NOSPRegClass : &X86::GR32_NOSPRegClass;
   }
   Register SrcReg = Src.getReg();
-  isKill = MI.killsRegister(SrcReg);
+  isKill = MI.killsRegister(SrcReg, /*TRI=*/nullptr);
 
   // For both LEA64 and LEA32 the register already has essentially the right
   // type (32-bit or 64-bit) we may just need to forbid SP.
@@ -3727,7 +3727,7 @@ bool X86InstrInfo::analyzeBranchImpl(
 
     // In practice we should never have an undef eflags operand, if we do
     // abort here as we are not prepared to preserve the flag.
-    if (I->findRegisterUseOperand(X86::EFLAGS)->isUndef())
+    if (I->findRegisterUseOperand(X86::EFLAGS, /*TRI=*/nullptr)->isUndef())
       return true;
 
     // Working from the bottom, handle the first conditional branch.
@@ -5472,7 +5472,8 @@ bool X86InstrInfo::optimizeCompareInstr(MachineInstr &CmpInstr, Register SrcReg,
   }
 
   // Make sure Sub instruction defines EFLAGS and mark the def live.
-  MachineOperand *FlagDef = Sub->findRegisterDefOperand(X86::EFLAGS);
+  MachineOperand *FlagDef =
+      Sub->findRegisterDefOperand(X86::EFLAGS, /*TRI=*/nullptr);
   assert(FlagDef && "Unable to locate a def EFLAGS operand");
   FlagDef->setIsDead(false);
 
@@ -5629,7 +5630,7 @@ bool X86InstrInfo::foldImmediateImpl(MachineInstr &UseMI, MachineInstr *DefMI,
       return false;
   }
 
-  if (UseMI.findRegisterUseOperand(Reg)->getSubReg())
+  if (UseMI.findRegisterUseOperand(Reg, /*TRI=*/nullptr)->getSubReg())
     return false;
   // Immediate has larger code size than register. So avoid folding the
   // immediate if it has more than 1 use and we are optimizing for size.
@@ -5676,7 +5677,8 @@ bool X86InstrInfo::foldImmediateImpl(MachineInstr &UseMI, MachineInstr *DefMI,
         if (!MakeChange)
           return true;
         UseMI.setDesc(get(X86::MOV32r0));
-        UseMI.removeOperand(UseMI.findRegisterUseOperandIdx(Reg));
+        UseMI.removeOperand(
+            UseMI.findRegisterUseOperandIdx(Reg, /*TRI=*/nullptr));
         UseMI.addOperand(MachineOperand::CreateReg(X86::EFLAGS, /*isDef=*/true,
                                                    /*isImp=*/true,
                                                    /*isKill=*/false,
@@ -5698,18 +5700,18 @@ bool X86InstrInfo::foldImmediateImpl(MachineInstr &UseMI, MachineInstr *DefMI,
        NewOpc == X86::SBB64ri32 || NewOpc == X86::SBB32ri ||
        NewOpc == X86::SUB64ri32_ND || NewOpc == X86::SUB32ri_ND ||
        NewOpc == X86::SBB64ri32_ND || NewOpc == X86::SBB32ri_ND) &&
-      UseMI.findRegisterUseOperandIdx(Reg) != 2)
+      UseMI.findRegisterUseOperandIdx(Reg, /*TRI=*/nullptr) != 2)
     return false;
   // For CMP instructions the immediate can only be at index 1.
   if (((NewOpc == X86::CMP64ri32 || NewOpc == X86::CMP32ri) ||
        (NewOpc == X86::CCMP64ri32 || NewOpc == X86::CCMP32ri)) &&
-      UseMI.findRegisterUseOperandIdx(Reg) != 1)
+      UseMI.findRegisterUseOperandIdx(Reg, /*TRI=*/nullptr) != 1)
     return false;
 
   using namespace X86;
   if (isSHL(Opc) || isSHR(Opc) || isSAR(Opc) || isROL(Opc) || isROR(Opc) ||
       isRCL(Opc) || isRCR(Opc)) {
-    unsigned RegIdx = UseMI.findRegisterUseOperandIdx(Reg);
+    unsigned RegIdx = UseMI.findRegisterUseOperandIdx(Reg, /*TRI=*/nullptr);
     if (RegIdx < 2)
       return false;
     if (!isInt<8>(ImmVal))
@@ -5733,13 +5735,15 @@ bool X86InstrInfo::foldImmediateImpl(MachineInstr &UseMI, MachineInstr *DefMI,
   if (!Modified) {
     // Modify the instruction.
     if (ImmVal == 0 && canConvert2Copy(NewOpc) &&
-        UseMI.registerDefIsDead(X86::EFLAGS)) {
+        UseMI.registerDefIsDead(X86::EFLAGS, /*TRI=*/nullptr)) {
       //          %100 = add %101, 0
       //    ==>
       //          %100 = COPY %101
       UseMI.setDesc(get(TargetOpcode::COPY));
-      UseMI.removeOperand(UseMI.findRegisterUseOperandIdx(Reg));
-      UseMI.removeOperand(UseMI.findRegisterDefOperandIdx(X86::EFLAGS));
+      UseMI.removeOperand(
+          UseMI.findRegisterUseOperandIdx(Reg, /*TRI=*/nullptr));
+      UseMI.removeOperand(
+          UseMI.findRegisterDefOperandIdx(X86::EFLAGS, /*TRI=*/nullptr));
       UseMI.untieRegOperand(0);
       UseMI.clearFlag(MachineInstr::MIFlag::NoSWrap);
       UseMI.clearFlag(MachineInstr::MIFlag::NoUWrap);
@@ -9538,7 +9542,8 @@ bool X86InstrInfo::hasReassociableOperands(const MachineInstr &Inst,
   // not change anything because rearranging the operands could affect other
   // instructions that depend on the exact status flags (zero, sign, etc.)
   // that are set by using these particular operands with this operation.
-  const MachineOperand *FlagDef = Inst.findRegisterDefOperand(X86::EFLAGS);
+  const MachineOperand *FlagDef =
+      Inst.findRegisterDefOperand(X86::EFLAGS, /*TRI=*/nullptr);
   assert((Inst.getNumDefs() == 1 || FlagDef) && "Implicit def isn't flags?");
   if (FlagDef && !FlagDef->isDead())
     return false;
@@ -10060,8 +10065,10 @@ void X86InstrInfo::setSpecialOperandAttr(MachineInstr &OldMI1,
                                          MachineInstr &NewMI1,
                                          MachineInstr &NewMI2) const {
   // Integer instructions may define an implicit EFLAGS dest register operand.
-  MachineOperand *OldFlagDef1 = OldMI1.findRegisterDefOperand(X86::EFLAGS);
-  MachineOperand *OldFlagDef2 = OldMI2.findRegisterDefOperand(X86::EFLAGS);
+  MachineOperand *OldFlagDef1 =
+      OldMI1.findRegisterDefOperand(X86::EFLAGS, /*TRI=*/nullptr);
+  MachineOperand *OldFlagDef2 =
+      OldMI2.findRegisterDefOperand(X86::EFLAGS, /*TRI=*/nullptr);
 
   assert(!OldFlagDef1 == !OldFlagDef2 &&
          "Unexpected instruction type for reassociation");
@@ -10072,8 +10079,10 @@ void X86InstrInfo::setSpecialOperandAttr(MachineInstr &OldMI1,
   assert(OldFlagDef1->isDead() && OldFlagDef2->isDead() &&
          "Must have dead EFLAGS operand in reassociable instruction");
 
-  MachineOperand *NewFlagDef1 = NewMI1.findRegisterDefOperand(X86::EFLAGS);
-  MachineOperand *NewFlagDef2 = NewMI2.findRegisterDefOperand(X86::EFLAGS);
+  MachineOperand *NewFlagDef1 =
+      NewMI1.findRegisterDefOperand(X86::EFLAGS, /*TRI=*/nullptr);
+  MachineOperand *NewFlagDef2 =
+      NewMI2.findRegisterDefOperand(X86::EFLAGS, /*TRI=*/nullptr);
 
   assert(NewFlagDef1 && NewFlagDef2 &&
          "Unexpected operand in reassociable instruction");
diff --git a/llvm/lib/Target/X86/X86LowerTileCopy.cpp b/llvm/lib/Target/X86/X86LowerTileCopy.cpp
index e7afc49240e5..fd05e16ac1ce 100644
--- a/llvm/lib/Target/X86/X86LowerTileCopy.cpp
+++ b/llvm/lib/Target/X86/X86LowerTileCopy.cpp
@@ -20,6 +20,7 @@
 #include "X86InstrBuilder.h"
 #include "X86InstrInfo.h"
 #include "X86Subtarget.h"
+#include "llvm/CodeGen/LiveRegUnits.h"
 #include "llvm/CodeGen/MachineBasicBlock.h"
 #include "llvm/CodeGen/MachineFrameInfo.h"
 #include "llvm/CodeGen/MachineFunction.h"
@@ -72,10 +73,28 @@ FunctionPass *llvm::createX86LowerTileCopyPass() {
 bool X86LowerTileCopy::runOnMachineFunction(MachineFunction &MF) {
   const X86Subtarget &ST = MF.getSubtarget<X86Subtarget>();
   const X86InstrInfo *TII = ST.getInstrInfo();
+  const TargetRegisterInfo *TRI = ST.getRegisterInfo();
+  BitVector GR64Regs =
+      TRI->getAllocatableSet(MF, TRI->getRegClass(X86::GR64RegClassID));
+  BitVector TILERegs =
+      TRI->getAllocatableSet(MF, TRI->getRegClass(X86::TILERegClassID));
   bool Changed = false;
 
   for (MachineBasicBlock &MBB : MF) {
-    for (MachineInstr &MI : llvm::make_early_inc_range(MBB)) {
+    // There won't be a tile copy if no tile register live in.
+    bool HasTileCopy = false;
+    for (const auto &LI : MBB.liveins()) {
+      if (TILERegs.test(LI.PhysReg)) {
+        HasTileCopy = true;
+        break;
+      }
+    }
+    if (!HasTileCopy)
+      continue;
+    LiveRegUnits UsedRegs(*TRI);
+    UsedRegs.addLiveOuts(MBB);
+    for (MachineInstr &MI : llvm::make_early_inc_range(reverse(MBB))) {
+      UsedRegs.stepBackward(MI);
       if (!MI.isCopy())
         continue;
       MachineOperand &DstMO = MI.getOperand(0);
@@ -85,27 +104,41 @@ bool X86LowerTileCopy::runOnMachineFunction(MachineFunction &MF) {
       if (!X86::TILERegClass.contains(DstReg, SrcReg))
         continue;
 
-      const TargetRegisterInfo *TRI = ST.getRegisterInfo();
       // Allocate stack slot for tile register
       unsigned Size = TRI->getSpillSize(X86::TILERegClass);
       Align Alignment = TRI->getSpillAlign(X86::TILERegClass);
       int TileSS = MF.getFrameInfo().CreateSpillStackObject(Size, Alignment);
-      // Allocate stack slot for stride register
-      Size = TRI->getSpillSize(X86::GR64RegClass);
-      Alignment = TRI->getSpillAlign(X86::GR64RegClass);
-      int StrideSS = MF.getFrameInfo().CreateSpillStackObject(Size, Alignment);
 
-      // TODO: Pick a killed regiter to avoid save/reload. There is problem
-      // to get live interval in this stage.
-      Register GR64Cand = X86::RAX;
+      int StrideSS = 0;
+
+      // Pick a killed register to avoid a save/reload.
+      Register GR64Cand = X86::NoRegister;
+      for (auto RegT : GR64Regs.set_bits()) {
+        if (UsedRegs.available(RegT)) {
+          GR64Cand = RegT;
+          break;
+        }
+      }
 
       const DebugLoc &DL = MI.getDebugLoc();
-      // mov %rax (%sp)
-      BuildMI(MBB, MI, DL, TII->get(X86::IMPLICIT_DEF), GR64Cand);
-      addFrameReference(BuildMI(MBB, MI, DL, TII->get(X86::MOV64mr)), StrideSS)
-          .addReg(GR64Cand);
-      // mov 64 %rax
-      BuildMI(MBB, MI, DL, TII->get(X86::MOV64ri), GR64Cand).addImm(64);
+      if (GR64Cand) {
+        // mov 64 %reg
+        BuildMI(MBB, MI, DL, TII->get(X86::MOV64ri), GR64Cand).addImm(64);
+      } else {
+        // No available register? Save RAX and reload it after use.
+
+        // Allocate stack slot for stride register
+        Size = TRI->getSpillSize(X86::GR64RegClass);
+        Alignment = TRI->getSpillAlign(X86::GR64RegClass);
+        StrideSS = MF.getFrameInfo().CreateSpillStackObject(Size, Alignment);
+
+        // mov %reg (%sp)
+        addFrameReference(BuildMI(MBB, MI, DL, TII->get(X86::MOV64mr)),
+                          StrideSS)
+            .addReg(X86::RAX);
+        // mov 64 %reg
+        BuildMI(MBB, MI, DL, TII->get(X86::MOV64ri), X86::RAX).addImm(64);
+      }
       // tilestored %tmm, (%sp, %idx)
 #define GET_EGPR_IF_ENABLED(OPC) (ST.hasEGPR() ? OPC##_EVEX : OPC)
       unsigned Opc = GET_EGPR_IF_ENABLED(X86::TILESTORED);
@@ -120,10 +153,12 @@ bool X86LowerTileCopy::runOnMachineFunction(MachineFunction &MF) {
 #undef GET_EGPR_IF_ENABLED
       NewMI = addFrameReference(BuildMI(MBB, MI, DL, TII->get(Opc), DstReg),
                                 TileSS);
-      // restore %rax
-      // mov (%sp) %rax
-      addFrameReference(BuildMI(MBB, MI, DL, TII->get(X86::MOV64rm), GR64Cand),
-                        StrideSS);
+      if (!GR64Cand) {
+        // restore %rax
+        // mov (%sp) %rax
+        addFrameReference(
+            BuildMI(MBB, MI, DL, TII->get(X86::MOV64rm), GR64Cand), StrideSS);
+      }
       MI.eraseFromParent();
       Changed = true;
     }
diff --git a/llvm/lib/Target/X86/X86MCInstLower.cpp b/llvm/lib/Target/X86/X86MCInstLower.cpp
index e6510be6b9af..1d699b42dc67 100644
--- a/llvm/lib/Target/X86/X86MCInstLower.cpp
+++ b/llvm/lib/Target/X86/X86MCInstLower.cpp
@@ -506,7 +506,8 @@ void X86MCInstLower::Lower(const MachineInstr *MI, MCInst &OutMI) const {
     // recognize as TZCNT, which has better performance than BSF.
     // BSF and TZCNT have different interpretations on ZF bit. So make sure
     // it won't be used later.
-    const MachineOperand *FlagDef = MI->findRegisterDefOperand(X86::EFLAGS);
+    const MachineOperand *FlagDef =
+        MI->findRegisterDefOperand(X86::EFLAGS, /*TRI=*/nullptr);
     if (!MF.getFunction().hasOptSize() && FlagDef && FlagDef->isDead())
       OutMI.setFlags(X86::IP_HAS_REPEAT);
     break;
diff --git a/llvm/lib/Target/X86/X86PfmCounters.td b/llvm/lib/Target/X86/X86PfmCounters.td
index d87a559aa353..2b1dac411c99 100644
--- a/llvm/lib/Target/X86/X86PfmCounters.td
+++ b/llvm/lib/Target/X86/X86PfmCounters.td
@@ -204,6 +204,22 @@ def : PfmCountersBinding<"icelake-server", IceLakePfmCounters>;
 def : PfmCountersBinding<"rocketlake", IceLakePfmCounters>;
 def : PfmCountersBinding<"tigerlake", IceLakePfmCounters>;
 
+def AlderLakePfmCounters : ProcPfmCounters {
+  let CycleCounter = UnhaltedCoreCyclesPfmCounter;
+  let UopsCounter = UopsIssuedPfmCounter;
+  let IssueCounters = [
+    PfmIssueCounter<"ADLPPort00", "uops_dispatched_port:port_0">,
+    PfmIssueCounter<"ADLPPort01", "uops_dispatched_port:port_1">,
+    PfmIssueCounter<"ADLPPort02_03_10", "uops_dispatched_port:port_2_3_10">,
+    PfmIssueCounter<"ADLPPort04_09", "uops_dispatched_port:port_4_9">,
+    PfmIssueCounter<"ADLPPort05_11", "uops_dispatched_port:port_5_11">,
+    PfmIssueCounter<"ADLPPort06", "uops_dispatched_port:port_6">,
+    PfmIssueCounter<"ADLPPort07_08", "uops_dispatched_port:port_7_8">
+  ];
+  let ValidationCounters = DefaultIntelPfmValidationCounters;
+}
+def : PfmCountersBinding<"alderlake", AlderLakePfmCounters>;
+
 // AMD X86 Counters.
 defvar DefaultAMDPfmValidationCounters = [
   PfmValidationCounter<InstructionRetired, "RETIRED_INSTRUCTIONS">,
diff --git a/llvm/lib/Target/X86/X86RegisterInfo.cpp b/llvm/lib/Target/X86/X86RegisterInfo.cpp
index be0cf1596d0d..555ede9e9540 100644
--- a/llvm/lib/Target/X86/X86RegisterInfo.cpp
+++ b/llvm/lib/Target/X86/X86RegisterInfo.cpp
@@ -649,10 +649,11 @@ unsigned X86RegisterInfo::getNumSupportedRegs(const MachineFunction &MF) const {
   // APX registers (R16-R31)
   //
   // and try to return the minimum number of registers supported by the target.
-  assert((X86::R15WH + 1 == X86 ::YMM0) && (X86::YMM15 + 1 == X86::K0) &&
-         (X86::K6_K7 + 1 == X86::TMMCFG) && (X86::TMM7 + 1 == X86::R16) &&
-         (X86::R31WH + 1 == X86::NUM_TARGET_REGS) &&
-         "Register number may be incorrect");
+  static_assert((X86::R15WH + 1 == X86::YMM0) && (X86::YMM15 + 1 == X86::K0) &&
+                    (X86::K6_K7 + 1 == X86::TMMCFG) &&
+                    (X86::TMM7 + 1 == X86::R16) &&
+                    (X86::R31WH + 1 == X86::NUM_TARGET_REGS),
+                "Register number may be incorrect");
 
   const X86Subtarget &ST = MF.getSubtarget<X86Subtarget>();
   if (ST.hasEGPR())
diff --git a/llvm/lib/Target/X86/X86SchedAlderlakeP.td b/llvm/lib/Target/X86/X86SchedAlderlakeP.td
index 6f9d2cf7ffdf..7756cd57cf7f 100644
--- a/llvm/lib/Target/X86/X86SchedAlderlakeP.td
+++ b/llvm/lib/Target/X86/X86SchedAlderlakeP.td
@@ -60,6 +60,8 @@ def ADLPPort01_05_10       : ProcResGroup<[ADLPPort01, ADLPPort05, ADLPPort10]>;
 def ADLPPort02_03          : ProcResGroup<[ADLPPort02, ADLPPort03]>;
 def ADLPPort02_03_07       : ProcResGroup<[ADLPPort02, ADLPPort03, ADLPPort07]>;
 def ADLPPort02_03_11       : ProcResGroup<[ADLPPort02, ADLPPort03, ADLPPort11]>;
+def ADLPPort02_03_10       : ProcResGroup<[ADLPPort02, ADLPPort03, ADLPPort10]>;
+def ADLPPort05_11          : ProcResGroup<[ADLPPort05, ADLPPort11]>;
 def ADLPPort07_08          : ProcResGroup<[ADLPPort07, ADLPPort08]>;
 
 // EU has 112 reservation stations.
@@ -79,6 +81,10 @@ def ADLPPort02_03_07_08_11 : ProcResGroup<[ADLPPort02, ADLPPort03, ADLPPort07,
   let BufferSize = 72;
 }
 
+def ADLPPortAny : ProcResGroup<[ADLPPort00, ADLPPort01, ADLPPort02, ADLPPort03,
+                                ADLPPort04, ADLPPort05, ADLPPort06, ADLPPort07,
+                                ADLPPort08, ADLPPort09, ADLPPort10, ADLPPort11]>;
+
 // Integer loads are 5 cycles, so ReadAfterLd registers needn't be available
 // until 5 cycles after the memory operand.
 def : ReadAdvance<ReadAfterLd, 5>;
diff --git a/llvm/lib/Target/X86/X86SpeculativeLoadHardening.cpp b/llvm/lib/Target/X86/X86SpeculativeLoadHardening.cpp
index 69a54e7667b5..489c8f492524 100644
--- a/llvm/lib/Target/X86/X86SpeculativeLoadHardening.cpp
+++ b/llvm/lib/Target/X86/X86SpeculativeLoadHardening.cpp
@@ -483,7 +483,7 @@ bool X86SpeculativeLoadHardeningPass::runOnMachineFunction(
                          PredStateSubReg);
     ++NumInstsInserted;
     MachineOperand *ZeroEFLAGSDefOp =
-        ZeroI->findRegisterDefOperand(X86::EFLAGS);
+        ZeroI->findRegisterDefOperand(X86::EFLAGS, /*TRI=*/nullptr);
     assert(ZeroEFLAGSDefOp && ZeroEFLAGSDefOp->isImplicit() &&
            "Must have an implicit def of EFLAGS!");
     ZeroEFLAGSDefOp->setIsDead(true);
@@ -762,7 +762,8 @@ X86SpeculativeLoadHardeningPass::tracePredStateThroughCFG(
             // If this is the last cmov and the EFLAGS weren't originally
             // live-in, mark them as killed.
             if (!LiveEFLAGS && Cond == Conds.back())
-              CMovI->findRegisterUseOperand(X86::EFLAGS)->setIsKill(true);
+              CMovI->findRegisterUseOperand(X86::EFLAGS, /*TRI=*/nullptr)
+                  ->setIsKill(true);
 
             ++NumInstsInserted;
             LLVM_DEBUG(dbgs() << "  Inserting cmov: "; CMovI->dump();
@@ -1185,7 +1186,8 @@ X86SpeculativeLoadHardeningPass::tracePredStateThroughIndirectBranches(
             .addReg(PS->InitialReg)
             .addReg(PS->PoisonReg)
             .addImm(X86::COND_NE);
-    CMovI->findRegisterUseOperand(X86::EFLAGS)->setIsKill(true);
+    CMovI->findRegisterUseOperand(X86::EFLAGS, /*TRI=*/nullptr)
+        ->setIsKill(true);
     ++NumInstsInserted;
     LLVM_DEBUG(dbgs() << "  Inserting cmov: "; CMovI->dump(); dbgs() << "\n");
     CMovs.push_back(&*CMovI);
@@ -1202,7 +1204,8 @@ X86SpeculativeLoadHardeningPass::tracePredStateThroughIndirectBranches(
 // Returns true if the MI has EFLAGS as a register def operand and it's live,
 // otherwise it returns false
 static bool isEFLAGSDefLive(const MachineInstr &MI) {
-  if (const MachineOperand *DefOp = MI.findRegisterDefOperand(X86::EFLAGS)) {
+  if (const MachineOperand *DefOp =
+          MI.findRegisterDefOperand(X86::EFLAGS, /*TRI=*/nullptr)) {
     return !DefOp->isDead();
   }
   return false;
@@ -1213,7 +1216,8 @@ static bool isEFLAGSLive(MachineBasicBlock &MBB, MachineBasicBlock::iterator I,
   // Check if EFLAGS are alive by seeing if there is a def of them or they
   // live-in, and then seeing if that def is in turn used.
   for (MachineInstr &MI : llvm::reverse(llvm::make_range(MBB.begin(), I))) {
-    if (MachineOperand *DefOp = MI.findRegisterDefOperand(X86::EFLAGS)) {
+    if (MachineOperand *DefOp =
+            MI.findRegisterDefOperand(X86::EFLAGS, /*TRI=*/nullptr)) {
       // If the def is dead, then EFLAGS is not live.
       if (DefOp->isDead())
         return false;
@@ -2182,7 +2186,7 @@ void X86SpeculativeLoadHardeningPass::tracePredStateThroughCall(
                    .addReg(NewStateReg, RegState::Kill)
                    .addReg(PS->PoisonReg)
                    .addImm(X86::COND_NE);
-  CMovI->findRegisterUseOperand(X86::EFLAGS)->setIsKill(true);
+  CMovI->findRegisterUseOperand(X86::EFLAGS, /*TRI=*/nullptr)->setIsKill(true);
   ++NumInstsInserted;
   LLVM_DEBUG(dbgs() << "  Inserting cmov: "; CMovI->dump(); dbgs() << "\n");
 
diff --git a/llvm/lib/Target/X86/X86Subtarget.h b/llvm/lib/Target/X86/X86Subtarget.h
index a458b5f9ec8f..4d55a084b730 100644
--- a/llvm/lib/Target/X86/X86Subtarget.h
+++ b/llvm/lib/Target/X86/X86Subtarget.h
@@ -244,7 +244,8 @@ public:
   // TODO: Currently we're always allowing widening on CPUs without VLX,
   // because for many cases we don't have a better option.
   bool canExtendTo512DQ() const {
-    return hasAVX512() && (!hasVLX() || getPreferVectorWidth() >= 512);
+    return hasAVX512() && hasEVEX512() &&
+           (!hasVLX() || getPreferVectorWidth() >= 512);
   }
   bool canExtendTo512BW() const  {
     return hasBWI() && canExtendTo512DQ();
diff --git a/llvm/lib/Target/X86/X86TargetTransformInfo.cpp b/llvm/lib/Target/X86/X86TargetTransformInfo.cpp
index 515847617e61..aac355713f90 100644
--- a/llvm/lib/Target/X86/X86TargetTransformInfo.cpp
+++ b/llvm/lib/Target/X86/X86TargetTransformInfo.cpp
@@ -1490,12 +1490,20 @@ InstructionCost X86TTIImpl::getShuffleCost(
   if (Kind == TTI::SK_Transpose)
     Kind = TTI::SK_PermuteTwoSrc;
 
-  // For Broadcasts we are splatting the first element from the first input
-  // register, so only need to reference that input and all the output
-  // registers are the same.
-  if (Kind == TTI::SK_Broadcast)
+  if (Kind == TTI::SK_Broadcast) {
+    // For Broadcasts we are splatting the first element from the first input
+    // register, so only need to reference that input and all the output
+    // registers are the same.
     LT.first = 1;
 
+    // If we're broadcasting a load then AVX/AVX2 can do this for free.
+    using namespace PatternMatch;
+    if (!Args.empty() && match(Args[0], m_OneUse(m_Load(m_Value()))) &&
+        (ST->hasAVX2() ||
+         (ST->hasAVX() && LT.second.getScalarSizeInBits() >= 32)))
+      return TTI::TCC_Free;
+  }
+
   // Treat <X x bfloat> shuffles as <X x half>.
   if (LT.second.isVector() && LT.second.getScalarType() == MVT::bf16)
     LT.second = LT.second.changeVectorElementType(MVT::f16);
diff --git a/llvm/lib/Target/X86/X86TargetTransformInfo.h b/llvm/lib/Target/X86/X86TargetTransformInfo.h
index 8ef9b4f86ffd..b50193074573 100644
--- a/llvm/lib/Target/X86/X86TargetTransformInfo.h
+++ b/llvm/lib/Target/X86/X86TargetTransformInfo.h
@@ -139,7 +139,7 @@ public:
       unsigned Opcode, Type *Ty, TTI::TargetCostKind CostKind,
       TTI::OperandValueInfo Op1Info = {TTI::OK_AnyValue, TTI::OP_None},
       TTI::OperandValueInfo Op2Info = {TTI::OK_AnyValue, TTI::OP_None},
-      ArrayRef<const Value *> Args = ArrayRef<const Value *>(),
+      ArrayRef<const Value *> Args = std::nullopt,
       const Instruction *CxtI = nullptr);
   InstructionCost getAltInstrCost(VectorType *VecTy, unsigned Opcode0,
                                   unsigned Opcode1,
diff --git a/llvm/lib/Target/Xtensa/AsmParser/XtensaAsmParser.cpp b/llvm/lib/Target/Xtensa/AsmParser/XtensaAsmParser.cpp
index 1fa00af2111e..eaf046630299 100644
--- a/llvm/lib/Target/Xtensa/AsmParser/XtensaAsmParser.cpp
+++ b/llvm/lib/Target/Xtensa/AsmParser/XtensaAsmParser.cpp
@@ -8,7 +8,9 @@
 //
 //===----------------------------------------------------------------------===//
 
+#include "MCTargetDesc/XtensaMCExpr.h"
 #include "MCTargetDesc/XtensaMCTargetDesc.h"
+#include "MCTargetDesc/XtensaTargetStreamer.h"
 #include "TargetInfo/XtensaTargetInfo.h"
 #include "llvm/ADT/STLExtras.h"
 #include "llvm/ADT/StringSwitch.h"
@@ -22,6 +24,7 @@
 #include "llvm/MC/MCRegisterInfo.h"
 #include "llvm/MC/MCStreamer.h"
 #include "llvm/MC/MCSubtargetInfo.h"
+#include "llvm/MC/MCSymbol.h"
 #include "llvm/MC/TargetRegistry.h"
 #include "llvm/Support/Casting.h"
 
@@ -35,6 +38,12 @@ class XtensaAsmParser : public MCTargetAsmParser {
 
   SMLoc getLoc() const { return getParser().getTok().getLoc(); }
 
+  XtensaTargetStreamer &getTargetStreamer() {
+    MCTargetStreamer &TS = *getParser().getStreamer().getTargetStreamer();
+    return static_cast<XtensaTargetStreamer &>(TS);
+  }
+
+  ParseStatus parseDirective(AsmToken DirectiveID) override;
   bool parseRegister(MCRegister &Reg, SMLoc &StartLoc, SMLoc &EndLoc) override;
   bool ParseInstruction(ParseInstructionInfo &Info, StringRef Name,
                         SMLoc NameLoc, OperandVector &Operands) override;
@@ -45,6 +54,9 @@ class XtensaAsmParser : public MCTargetAsmParser {
   unsigned validateTargetOperandClass(MCParsedAsmOperand &Op,
                                       unsigned Kind) override;
 
+  bool processInstruction(MCInst &Inst, SMLoc IDLoc, MCStreamer &Out,
+                          const MCSubtargetInfo *STI);
+
 // Auto-generated instruction matching functions
 #define GET_ASSEMBLER_HEADER
 #include "XtensaGenAsmMatcher.inc"
@@ -62,6 +74,7 @@ class XtensaAsmParser : public MCTargetAsmParser {
     return ParseStatus::NoMatch;
   }
   ParseStatus parsePCRelTarget(OperandVector &Operands);
+  bool parseLiteralDirective(SMLoc L);
 
 public:
   enum XtensaMatchResultTy {
@@ -148,7 +161,8 @@ public:
 
   bool isImm12() const { return isImm(-2048, 2047); }
 
-  bool isImm12m() const { return isImm(-2048, 2047); }
+  // Convert MOVI to literal load, when immediate is not in range (-2048, 2047)
+  bool isImm12m() const { return Kind == Immediate; }
 
   bool isOffset4m32() const {
     return isImm(0, 60) &&
@@ -348,6 +362,69 @@ static SMLoc RefineErrorLoc(const SMLoc Loc, const OperandVector &Operands,
   return Loc;
 }
 
+bool XtensaAsmParser::processInstruction(MCInst &Inst, SMLoc IDLoc,
+                                         MCStreamer &Out,
+                                         const MCSubtargetInfo *STI) {
+  Inst.setLoc(IDLoc);
+  const unsigned Opcode = Inst.getOpcode();
+  switch (Opcode) {
+  case Xtensa::L32R: {
+    const MCSymbolRefExpr *OpExpr =
+        static_cast<const MCSymbolRefExpr *>(Inst.getOperand(1).getExpr());
+    XtensaMCExpr::VariantKind Kind = XtensaMCExpr::VK_Xtensa_None;
+    const MCExpr *NewOpExpr = XtensaMCExpr::create(OpExpr, Kind, getContext());
+    Inst.getOperand(1).setExpr(NewOpExpr);
+    break;
+  }
+  case Xtensa::MOVI: {
+    XtensaTargetStreamer &TS = this->getTargetStreamer();
+
+    // Expand MOVI operand
+    if (!Inst.getOperand(1).isExpr()) {
+      uint64_t ImmOp64 = Inst.getOperand(1).getImm();
+      int32_t Imm = ImmOp64;
+      if (!isInt<12>(Imm)) {
+        XtensaTargetStreamer &TS = this->getTargetStreamer();
+        MCInst TmpInst;
+        TmpInst.setLoc(IDLoc);
+        TmpInst.setOpcode(Xtensa::L32R);
+        const MCExpr *Value = MCConstantExpr::create(ImmOp64, getContext());
+        MCSymbol *Sym = getContext().createTempSymbol();
+        const MCExpr *Expr = MCSymbolRefExpr::create(
+            Sym, MCSymbolRefExpr::VK_None, getContext());
+        const MCExpr *OpExpr = XtensaMCExpr::create(
+            Expr, XtensaMCExpr::VK_Xtensa_None, getContext());
+        TmpInst.addOperand(Inst.getOperand(0));
+        MCOperand Op1 = MCOperand::createExpr(OpExpr);
+        TmpInst.addOperand(Op1);
+        TS.emitLiteral(Sym, Value, true, IDLoc);
+        Inst = TmpInst;
+      }
+    } else {
+      MCInst TmpInst;
+      TmpInst.setLoc(IDLoc);
+      TmpInst.setOpcode(Xtensa::L32R);
+      const MCExpr *Value = Inst.getOperand(1).getExpr();
+      MCSymbol *Sym = getContext().createTempSymbol();
+      const MCExpr *Expr =
+          MCSymbolRefExpr::create(Sym, MCSymbolRefExpr::VK_None, getContext());
+      const MCExpr *OpExpr = XtensaMCExpr::create(
+          Expr, XtensaMCExpr::VK_Xtensa_None, getContext());
+      TmpInst.addOperand(Inst.getOperand(0));
+      MCOperand Op1 = MCOperand::createExpr(OpExpr);
+      TmpInst.addOperand(Op1);
+      Inst = TmpInst;
+      TS.emitLiteral(Sym, Value, true, IDLoc);
+    }
+    break;
+  }
+  default:
+    break;
+  }
+
+  return true;
+}
+
 bool XtensaAsmParser::MatchAndEmitInstruction(SMLoc IDLoc, unsigned &Opcode,
                                               OperandVector &Operands,
                                               MCStreamer &Out,
@@ -361,6 +438,7 @@ bool XtensaAsmParser::MatchAndEmitInstruction(SMLoc IDLoc, unsigned &Opcode,
   default:
     break;
   case Match_Success:
+    processInstruction(Inst, IDLoc, Out, STI);
     Inst.setLoc(IDLoc);
     Out.emitInstruction(Inst, getSTI());
     return false;
@@ -686,6 +764,57 @@ bool XtensaAsmParser::ParseInstruction(ParseInstructionInfo &Info,
   return false;
 }
 
+bool XtensaAsmParser::parseLiteralDirective(SMLoc L) {
+  MCAsmParser &Parser = getParser();
+  const MCExpr *Value;
+  SMLoc LiteralLoc = getLexer().getLoc();
+  XtensaTargetStreamer &TS = this->getTargetStreamer();
+
+  if (Parser.parseExpression(Value))
+    return true;
+
+  const MCSymbolRefExpr *SE = dyn_cast<MCSymbolRefExpr>(Value);
+
+  if (!SE)
+    return Error(LiteralLoc, "literal label must be a symbol");
+
+  if (Parser.parseComma())
+    return true;
+
+  SMLoc OpcodeLoc = getLexer().getLoc();
+  if (parseOptionalToken(AsmToken::EndOfStatement))
+    return Error(OpcodeLoc, "expected value");
+
+  if (Parser.parseExpression(Value))
+    return true;
+
+  if (parseEOL())
+    return true;
+
+  MCSymbol *Sym = getContext().getOrCreateSymbol(SE->getSymbol().getName());
+
+  TS.emitLiteral(Sym, Value, true, LiteralLoc);
+
+  return false;
+}
+
+ParseStatus XtensaAsmParser::parseDirective(AsmToken DirectiveID) {
+  StringRef IDVal = DirectiveID.getString();
+  SMLoc Loc = getLexer().getLoc();
+
+  if (IDVal == ".literal_position") {
+    XtensaTargetStreamer &TS = this->getTargetStreamer();
+    TS.emitLiteralPosition();
+    return parseEOL();
+  }
+
+  if (IDVal == ".literal") {
+    return parseLiteralDirective(Loc);
+  }
+
+  return ParseStatus::NoMatch;
+}
+
 // Force static initialization.
 extern "C" LLVM_EXTERNAL_VISIBILITY void LLVMInitializeXtensaAsmParser() {
   RegisterMCAsmParser<XtensaAsmParser> X(getTheXtensaTarget());
diff --git a/llvm/lib/Target/Xtensa/CMakeLists.txt b/llvm/lib/Target/Xtensa/CMakeLists.txt
index 2064511e75b8..726efadc87c0 100644
--- a/llvm/lib/Target/Xtensa/CMakeLists.txt
+++ b/llvm/lib/Target/Xtensa/CMakeLists.txt
@@ -4,6 +4,7 @@ set(LLVM_TARGET_DEFINITIONS Xtensa.td)
 
 tablegen(LLVM XtensaGenAsmMatcher.inc -gen-asm-matcher)
 tablegen(LLVM XtensaGenAsmWriter.inc -gen-asm-writer)
+tablegen(LLVM XtensaGenCallingConv.inc -gen-callingconv)
 tablegen(LLVM XtensaGenDAGISel.inc -gen-dag-isel)
 tablegen(LLVM XtensaGenDisassemblerTables.inc -gen-disassembler)
 tablegen(LLVM XtensaGenInstrInfo.inc -gen-instr-info)
@@ -15,6 +16,7 @@ add_public_tablegen_target(XtensaCommonTableGen)
 
 add_llvm_target(XtensaCodeGen
   XtensaAsmPrinter.cpp
+  XtensaConstantPoolValue.cpp
   XtensaFrameLowering.cpp
   XtensaInstrInfo.cpp
   XtensaISelDAGToDAG.cpp
@@ -22,6 +24,7 @@ add_llvm_target(XtensaCodeGen
   XtensaRegisterInfo.cpp
   XtensaSubtarget.cpp
   XtensaTargetMachine.cpp
+  XtensaUtils.cpp
 
   LINK_COMPONENTS
   AsmPrinter
diff --git a/llvm/lib/Target/Xtensa/MCTargetDesc/CMakeLists.txt b/llvm/lib/Target/Xtensa/MCTargetDesc/CMakeLists.txt
index 6841b44f9d56..dc12863394c7 100644
--- a/llvm/lib/Target/Xtensa/MCTargetDesc/CMakeLists.txt
+++ b/llvm/lib/Target/Xtensa/MCTargetDesc/CMakeLists.txt
@@ -6,6 +6,7 @@ add_llvm_component_library(LLVMXtensaDesc
   XtensaMCCodeEmitter.cpp
   XtensaMCExpr.cpp
   XtensaMCTargetDesc.cpp
+  XtensaTargetStreamer.cpp
 
   LINK_COMPONENTS
   MC
diff --git a/llvm/lib/Target/Xtensa/MCTargetDesc/XtensaMCTargetDesc.cpp b/llvm/lib/Target/Xtensa/MCTargetDesc/XtensaMCTargetDesc.cpp
index 48674d15bdfb..87ef66ba742b 100644
--- a/llvm/lib/Target/Xtensa/MCTargetDesc/XtensaMCTargetDesc.cpp
+++ b/llvm/lib/Target/Xtensa/MCTargetDesc/XtensaMCTargetDesc.cpp
@@ -8,9 +8,10 @@
 //
 //===----------------------------------------------------------------------===//
 #include "XtensaMCTargetDesc.h"
+#include "TargetInfo/XtensaTargetInfo.h"
 #include "XtensaInstPrinter.h"
 #include "XtensaMCAsmInfo.h"
-#include "TargetInfo/XtensaTargetInfo.h"
+#include "XtensaTargetStreamer.h"
 #include "llvm/ADT/STLExtras.h"
 #include "llvm/MC/MCAsmInfo.h"
 #include "llvm/MC/MCInstrInfo.h"
@@ -63,16 +64,29 @@ createXtensaMCSubtargetInfo(const Triple &TT, StringRef CPU, StringRef FS) {
   return createXtensaMCSubtargetInfoImpl(TT, CPU, CPU, FS);
 }
 
+static MCTargetStreamer *
+createXtensaAsmTargetStreamer(MCStreamer &S, formatted_raw_ostream &OS,
+                              MCInstPrinter *InstPrint, bool isVerboseAsm) {
+  return new XtensaTargetAsmStreamer(S, OS);
+}
+
+static MCTargetStreamer *
+createXtensaObjectTargetStreamer(MCStreamer &S, const MCSubtargetInfo &STI) {
+  return new XtensaTargetELFStreamer(S);
+}
+
 extern "C" LLVM_EXTERNAL_VISIBILITY void LLVMInitializeXtensaTargetMC() {
   // Register the MCAsmInfo.
-  TargetRegistry::RegisterMCAsmInfo(getTheXtensaTarget(), createXtensaMCAsmInfo);
+  TargetRegistry::RegisterMCAsmInfo(getTheXtensaTarget(),
+                                    createXtensaMCAsmInfo);
 
   // Register the MCCodeEmitter.
   TargetRegistry::RegisterMCCodeEmitter(getTheXtensaTarget(),
                                         createXtensaMCCodeEmitter);
 
   // Register the MCInstrInfo.
-  TargetRegistry::RegisterMCInstrInfo(getTheXtensaTarget(), createXtensaMCInstrInfo);
+  TargetRegistry::RegisterMCInstrInfo(getTheXtensaTarget(),
+                                      createXtensaMCInstrInfo);
 
   // Register the MCInstPrinter.
   TargetRegistry::RegisterMCInstPrinter(getTheXtensaTarget(),
@@ -89,4 +103,12 @@ extern "C" LLVM_EXTERNAL_VISIBILITY void LLVMInitializeXtensaTargetMC() {
   // Register the MCAsmBackend.
   TargetRegistry::RegisterMCAsmBackend(getTheXtensaTarget(),
                                        createXtensaMCAsmBackend);
+
+  // Register the asm target streamer.
+  TargetRegistry::RegisterAsmTargetStreamer(getTheXtensaTarget(),
+                                            createXtensaAsmTargetStreamer);
+
+  // Register the ELF target streamer.
+  TargetRegistry::RegisterObjectTargetStreamer(
+      getTheXtensaTarget(), createXtensaObjectTargetStreamer);
 }
diff --git a/llvm/lib/Target/Xtensa/MCTargetDesc/XtensaTargetStreamer.cpp b/llvm/lib/Target/Xtensa/MCTargetDesc/XtensaTargetStreamer.cpp
new file mode 100644
index 000000000000..0ea70cff4d40
--- /dev/null
+++ b/llvm/lib/Target/Xtensa/MCTargetDesc/XtensaTargetStreamer.cpp
@@ -0,0 +1,119 @@
+//===-- XtensaTargetStreamer.cpp - Xtensa Target Streamer Methods ---------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file provides Xtensa specific target streamer methods.
+//
+//===----------------------------------------------------------------------===//
+
+#include "XtensaTargetStreamer.h"
+#include "XtensaInstPrinter.h"
+#include "llvm/BinaryFormat/ELF.h"
+#include "llvm/MC/MCAssembler.h"
+#include "llvm/MC/MCContext.h"
+#include "llvm/MC/MCObjectFileInfo.h"
+#include "llvm/MC/MCSectionELF.h"
+#include "llvm/Support/Casting.h"
+#include "llvm/Support/FormattedStream.h"
+
+using namespace llvm;
+
+static std::string getLiteralSectionName(StringRef CSectionName) {
+  std::size_t Pos = CSectionName.find(".text");
+  std::string SectionName;
+  if (Pos != std::string::npos) {
+    SectionName = CSectionName.substr(0, Pos);
+
+    if (Pos > 0)
+      SectionName += ".text";
+
+    CSectionName = CSectionName.drop_front(Pos);
+    CSectionName.consume_front(".text");
+
+    SectionName += ".literal";
+    SectionName += CSectionName;
+  } else {
+    SectionName = CSectionName;
+    SectionName += ".literal";
+  }
+  return SectionName;
+}
+
+XtensaTargetStreamer::XtensaTargetStreamer(MCStreamer &S)
+    : MCTargetStreamer(S) {}
+
+XtensaTargetAsmStreamer::XtensaTargetAsmStreamer(MCStreamer &S,
+                                                 formatted_raw_ostream &OS)
+    : XtensaTargetStreamer(S), OS(OS) {}
+
+void XtensaTargetAsmStreamer::emitLiteral(MCSymbol *LblSym, const MCExpr *Value,
+                                          bool SwitchLiteralSection, SMLoc L) {
+  SmallString<60> Str;
+  raw_svector_ostream LiteralStr(Str);
+
+  LiteralStr << "\t.literal " << LblSym->getName() << ", ";
+
+  if (auto CE = dyn_cast<MCConstantExpr>(Value)) {
+    LiteralStr << CE->getValue() << "\n";
+  } else if (auto SRE = dyn_cast<MCSymbolRefExpr>(Value)) {
+    const MCSymbol &Sym = SRE->getSymbol();
+    LiteralStr << Sym.getName() << "\n";
+  } else {
+    llvm_unreachable("unexpected constant pool entry type");
+  }
+
+  OS << LiteralStr.str();
+}
+
+void XtensaTargetAsmStreamer::emitLiteralPosition() {
+  OS << "\t.literal_position\n";
+}
+
+void XtensaTargetAsmStreamer::startLiteralSection(MCSection *BaseSection) {
+  emitLiteralPosition();
+}
+
+XtensaTargetELFStreamer::XtensaTargetELFStreamer(MCStreamer &S)
+    : XtensaTargetStreamer(S) {}
+
+void XtensaTargetELFStreamer::emitLiteral(MCSymbol *LblSym, const MCExpr *Value,
+                                          bool SwitchLiteralSection, SMLoc L) {
+  MCStreamer &OutStreamer = getStreamer();
+  if (SwitchLiteralSection) {
+    MCContext &Context = OutStreamer.getContext();
+    auto *CS = static_cast<MCSectionELF *>(OutStreamer.getCurrentSectionOnly());
+    std::string SectionName = getLiteralSectionName(CS->getName());
+
+    MCSection *ConstSection = Context.getELFSection(
+        SectionName, ELF::SHT_PROGBITS, ELF::SHF_EXECINSTR | ELF::SHF_ALLOC);
+
+    OutStreamer.pushSection();
+    OutStreamer.switchSection(ConstSection);
+  }
+
+  OutStreamer.emitLabel(LblSym, L);
+  OutStreamer.emitValue(Value, 4, L);
+
+  if (SwitchLiteralSection) {
+    OutStreamer.popSection();
+  }
+}
+
+void XtensaTargetELFStreamer::startLiteralSection(MCSection *BaseSection) {
+  MCContext &Context = getStreamer().getContext();
+
+  std::string SectionName = getLiteralSectionName(BaseSection->getName());
+
+  MCSection *ConstSection = Context.getELFSection(
+      SectionName, ELF::SHT_PROGBITS, ELF::SHF_EXECINSTR | ELF::SHF_ALLOC);
+
+  ConstSection->setAlignment(Align(4));
+}
+
+MCELFStreamer &XtensaTargetELFStreamer::getStreamer() {
+  return static_cast<MCELFStreamer &>(Streamer);
+}
diff --git a/llvm/lib/Target/Xtensa/MCTargetDesc/XtensaTargetStreamer.h b/llvm/lib/Target/Xtensa/MCTargetDesc/XtensaTargetStreamer.h
new file mode 100644
index 000000000000..817940e880b3
--- /dev/null
+++ b/llvm/lib/Target/Xtensa/MCTargetDesc/XtensaTargetStreamer.h
@@ -0,0 +1,58 @@
+//===-- XtensaTargetStreamer.h - Xtensa Target Streamer --------*- C++ -*--===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_LIB_TARGET_XTENSA_XTENSATARGETSTREAMER_H
+#define LLVM_LIB_TARGET_XTENSA_XTENSATARGETSTREAMER_H
+
+#include "llvm/MC/MCELFStreamer.h"
+#include "llvm/MC/MCStreamer.h"
+#include "llvm/Support/SMLoc.h"
+
+namespace llvm {
+class formatted_raw_ostream;
+
+class XtensaTargetStreamer : public MCTargetStreamer {
+public:
+  XtensaTargetStreamer(MCStreamer &S);
+
+  // Emit literal label and literal Value to the literal section. If literal
+  // section is not switched yet (SwitchLiteralSection is true) then switch to
+  // literal section.
+  virtual void emitLiteral(MCSymbol *LblSym, const MCExpr *Value,
+                           bool SwitchLiteralSection, SMLoc L = SMLoc()) = 0;
+
+  virtual void emitLiteralPosition() = 0;
+
+  // Switch to the literal section. The BaseSection name is used to construct
+  // literal section name.
+  virtual void startLiteralSection(MCSection *BaseSection) = 0;
+};
+
+class XtensaTargetAsmStreamer : public XtensaTargetStreamer {
+  formatted_raw_ostream &OS;
+
+public:
+  XtensaTargetAsmStreamer(MCStreamer &S, formatted_raw_ostream &OS);
+  void emitLiteral(MCSymbol *LblSym, const MCExpr *Value,
+                   bool SwitchLiteralSection, SMLoc L) override;
+  void emitLiteralPosition() override;
+  void startLiteralSection(MCSection *Section) override;
+};
+
+class XtensaTargetELFStreamer : public XtensaTargetStreamer {
+public:
+  XtensaTargetELFStreamer(MCStreamer &S);
+  MCELFStreamer &getStreamer();
+  void emitLiteral(MCSymbol *LblSym, const MCExpr *Value,
+                   bool SwitchLiteralSection, SMLoc L) override;
+  void emitLiteralPosition() override {}
+  void startLiteralSection(MCSection *Section) override;
+};
+} // end namespace llvm
+
+#endif // LLVM_LIB_TARGET_XTENSA_XTENSATARGETSTREAMER_H
diff --git a/llvm/lib/Target/Xtensa/Xtensa.td b/llvm/lib/Target/Xtensa/Xtensa.td
index b953540be94d..460a15e808b3 100644
--- a/llvm/lib/Target/Xtensa/Xtensa.td
+++ b/llvm/lib/Target/Xtensa/Xtensa.td
@@ -36,6 +36,12 @@ def : Proc<"generic", []>;
 include "XtensaRegisterInfo.td"
 
 //===----------------------------------------------------------------------===//
+// Calling Convention Description
+//===----------------------------------------------------------------------===//
+
+include "XtensaCallingConv.td"
+
+//===----------------------------------------------------------------------===//
 // Instruction Descriptions
 //===----------------------------------------------------------------------===//
 
diff --git a/llvm/lib/Target/Xtensa/XtensaAsmPrinter.cpp b/llvm/lib/Target/Xtensa/XtensaAsmPrinter.cpp
index 87dbf2eb5166..e222919b28dc 100644
--- a/llvm/lib/Target/Xtensa/XtensaAsmPrinter.cpp
+++ b/llvm/lib/Target/Xtensa/XtensaAsmPrinter.cpp
@@ -12,8 +12,13 @@
 //===----------------------------------------------------------------------===//
 
 #include "XtensaAsmPrinter.h"
+#include "MCTargetDesc/XtensaMCExpr.h"
+#include "MCTargetDesc/XtensaTargetStreamer.h"
 #include "TargetInfo/XtensaTargetInfo.h"
+#include "XtensaConstantPoolValue.h"
+#include "llvm/ADT/StringExtras.h"
 #include "llvm/BinaryFormat/ELF.h"
+#include "llvm/CodeGen/MachineConstantPool.h"
 #include "llvm/CodeGen/MachineModuleInfoImpls.h"
 #include "llvm/CodeGen/TargetLoweringObjectFileImpl.h"
 #include "llvm/MC/MCExpr.h"
@@ -25,12 +30,152 @@
 
 using namespace llvm;
 
+static MCSymbolRefExpr::VariantKind
+getModifierVariantKind(XtensaCP::XtensaCPModifier Modifier) {
+  switch (Modifier) {
+  case XtensaCP::no_modifier:
+    return MCSymbolRefExpr::VK_None;
+  case XtensaCP::TPOFF:
+    return MCSymbolRefExpr::VK_TPOFF;
+  }
+  report_fatal_error("Invalid XtensaCPModifier!");
+}
+
 void XtensaAsmPrinter::emitInstruction(const MachineInstr *MI) {
   MCInst LoweredMI;
   lowerToMCInst(MI, LoweredMI);
   EmitToStreamer(*OutStreamer, LoweredMI);
 }
 
+void XtensaAsmPrinter::emitMachineConstantPoolValue(
+    MachineConstantPoolValue *MCPV) {
+  XtensaConstantPoolValue *ACPV = static_cast<XtensaConstantPoolValue *>(MCPV);
+  MCSymbol *MCSym;
+
+  assert(ACPV->isExtSymbol() && "unrecognized constant pool value");
+
+  XtensaConstantPoolSymbol *XtensaSym = cast<XtensaConstantPoolSymbol>(ACPV);
+  const char *Sym = XtensaSym->getSymbol();
+  std::string SymName(Sym);
+
+  if (XtensaSym->isPrivateLinkage())
+    SymName = ".L" + SymName;
+
+  MCSym = GetExternalSymbolSymbol(StringRef(SymName));
+  MCSymbol *LblSym = GetCPISymbol(ACPV->getLabelId());
+  auto *TS =
+      static_cast<XtensaTargetStreamer *>(OutStreamer->getTargetStreamer());
+  MCSymbolRefExpr::VariantKind VK = getModifierVariantKind(ACPV->getModifier());
+
+  if (ACPV->getModifier() != XtensaCP::no_modifier) {
+    std::string SymName(MCSym->getName());
+    StringRef Modifier = ACPV->getModifierText();
+    SymName += Modifier;
+    MCSym = GetExternalSymbolSymbol(StringRef(SymName));
+  }
+
+  const MCExpr *Expr = MCSymbolRefExpr::create(MCSym, VK, OutContext);
+  TS->emitLiteral(LblSym, Expr, false);
+}
+
+void XtensaAsmPrinter::emitMachineConstantPoolEntry(
+    const MachineConstantPoolEntry &CPE, int i) {
+  if (CPE.isMachineConstantPoolEntry()) {
+    XtensaConstantPoolValue *ACPV =
+        static_cast<XtensaConstantPoolValue *>(CPE.Val.MachineCPVal);
+    ACPV->setLabelId(i);
+    emitMachineConstantPoolValue(CPE.Val.MachineCPVal);
+  } else {
+    MCSymbol *LblSym = GetCPISymbol(i);
+    auto *TS =
+        static_cast<XtensaTargetStreamer *>(OutStreamer->getTargetStreamer());
+    const Constant *C = CPE.Val.ConstVal;
+    const MCExpr *Value = nullptr;
+
+    Type *Ty = C->getType();
+    if (const auto *CFP = dyn_cast<ConstantFP>(C)) {
+      Value = MCConstantExpr::create(
+          CFP->getValueAPF().bitcastToAPInt().getSExtValue(), OutContext);
+    } else if (const auto *CI = dyn_cast<ConstantInt>(C)) {
+      Value = MCConstantExpr::create(CI->getValue().getSExtValue(), OutContext);
+    } else if (isa<PointerType>(Ty)) {
+      Value = lowerConstant(C);
+    } else {
+      llvm_unreachable("unexpected constant pool entry type");
+    }
+
+    TS->emitLiteral(LblSym, Value, false);
+  }
+}
+
+// EmitConstantPool - Print to the current output stream assembly
+// representations of the constants in the constant pool MCP. This is
+// used to print out constants which have been "spilled to memory" by
+// the code generator.
+void XtensaAsmPrinter::emitConstantPool() {
+  const Function &F = MF->getFunction();
+  const MachineConstantPool *MCP = MF->getConstantPool();
+  const std::vector<MachineConstantPoolEntry> &CP = MCP->getConstants();
+  if (CP.empty())
+    return;
+
+  OutStreamer->pushSection();
+
+  auto *TS =
+      static_cast<XtensaTargetStreamer *>(OutStreamer->getTargetStreamer());
+  MCSection *CS = getObjFileLowering().SectionForGlobal(&F, TM);
+  TS->startLiteralSection(CS);
+
+  int CPIdx = 0;
+  for (const MachineConstantPoolEntry &CPE : CP) {
+    emitMachineConstantPoolEntry(CPE, CPIdx++);
+  }
+
+  OutStreamer->popSection();
+}
+
+MCSymbol *
+XtensaAsmPrinter::GetConstantPoolIndexSymbol(const MachineOperand &MO) const {
+  // Create a symbol for the name.
+  return GetCPISymbol(MO.getIndex());
+}
+
+MCOperand
+XtensaAsmPrinter::LowerSymbolOperand(const MachineOperand &MO,
+                                     MachineOperand::MachineOperandType MOTy,
+                                     unsigned Offset) const {
+  const MCSymbol *Symbol;
+  XtensaMCExpr::VariantKind Kind = XtensaMCExpr::VK_Xtensa_None;
+
+  switch (MOTy) {
+  case MachineOperand::MO_GlobalAddress:
+    Symbol = getSymbol(MO.getGlobal());
+    Offset += MO.getOffset();
+    break;
+  case MachineOperand::MO_ConstantPoolIndex:
+    Symbol = GetConstantPoolIndexSymbol(MO);
+    Offset += MO.getOffset();
+    break;
+  default:
+    report_fatal_error("<unknown operand type>");
+  }
+
+  const MCExpr *ME =
+      MCSymbolRefExpr::create(Symbol, MCSymbolRefExpr::VK_None, OutContext);
+  ME = XtensaMCExpr::create(ME, Kind, OutContext);
+
+  if (Offset) {
+    // Assume offset is never negative.
+    assert(Offset > 0);
+
+    const MCConstantExpr *OffsetExpr =
+        MCConstantExpr::create(Offset, OutContext);
+    ME = MCBinaryExpr::createAdd(ME, OffsetExpr, OutContext);
+  }
+
+  return MCOperand::createExpr(ME);
+}
+
 MCOperand XtensaAsmPrinter::lowerOperand(const MachineOperand &MO,
                                          unsigned Offset) const {
   MachineOperand::MachineOperandType MOTy = MO.getType();
@@ -45,6 +190,9 @@ MCOperand XtensaAsmPrinter::lowerOperand(const MachineOperand &MO,
     return MCOperand::createImm(MO.getImm() + Offset);
   case MachineOperand::MO_RegisterMask:
     break;
+  case MachineOperand::MO_GlobalAddress:
+  case MachineOperand::MO_ConstantPoolIndex:
+    return LowerSymbolOperand(MO, MOTy, Offset);
   default:
     report_fatal_error("unknown operand type");
   }
diff --git a/llvm/lib/Target/Xtensa/XtensaAsmPrinter.h b/llvm/lib/Target/Xtensa/XtensaAsmPrinter.h
index dec2a1ee4954..f3fec19724aa 100644
--- a/llvm/lib/Target/Xtensa/XtensaAsmPrinter.h
+++ b/llvm/lib/Target/Xtensa/XtensaAsmPrinter.h
@@ -15,6 +15,7 @@
 
 #include "XtensaTargetMachine.h"
 #include "llvm/CodeGen/AsmPrinter.h"
+#include "llvm/CodeGen/MachineConstantPool.h"
 #include "llvm/Support/Compiler.h"
 
 namespace llvm {
@@ -35,6 +36,18 @@ public:
   StringRef getPassName() const override { return "Xtensa Assembly Printer"; }
   void emitInstruction(const MachineInstr *MI) override;
 
+  void emitConstantPool() override;
+
+  void emitMachineConstantPoolEntry(const MachineConstantPoolEntry &CPE, int i);
+
+  void emitMachineConstantPoolValue(MachineConstantPoolValue *MCPV) override;
+
+  MCSymbol *GetConstantPoolIndexSymbol(const MachineOperand &MO) const;
+
+  MCOperand LowerSymbolOperand(const MachineOperand &MO,
+                               MachineOperand::MachineOperandType MOTy,
+                               unsigned Offset) const;
+
   // Lower MachineInstr MI to MCInst OutMI.
   void lowerToMCInst(const MachineInstr *MI, MCInst &OutMI) const;
 
diff --git a/llvm/lib/Target/Xtensa/XtensaCallingConv.td b/llvm/lib/Target/Xtensa/XtensaCallingConv.td
new file mode 100644
index 000000000000..a348b4c890b2
--- /dev/null
+++ b/llvm/lib/Target/Xtensa/XtensaCallingConv.td
@@ -0,0 +1,24 @@
+//===- XtensaCallingConv.td - Xtensa Calling Conventions -*- tablegen ---*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+// This describes the calling conventions for the Xtensa ABI.
+//===----------------------------------------------------------------------===//
+
+//===----------------------------------------------------------------------===//
+// Xtensa return value calling convention
+//===----------------------------------------------------------------------===//
+def RetCC_Xtensa : CallingConv<[
+  // First two return values go in a2, a3, a4, a5
+  CCIfType<[i32], CCAssignToReg<[A2, A3, A4, A5]>>,
+  CCIfType<[i64], CCAssignToRegWithShadow<[A2, A4], [A3, A5]>>
+]>;
+
+//===----------------------------------------------------------------------===//
+// Callee-saved register lists.
+//===----------------------------------------------------------------------===//
+
+def CSR_Xtensa : CalleeSavedRegs<(add A0, A12, A13, A14, A15)>;
diff --git a/llvm/lib/Target/Xtensa/XtensaConstantPoolValue.cpp b/llvm/lib/Target/Xtensa/XtensaConstantPoolValue.cpp
new file mode 100644
index 000000000000..4e53aa5736c7
--- /dev/null
+++ b/llvm/lib/Target/Xtensa/XtensaConstantPoolValue.cpp
@@ -0,0 +1,207 @@
+//===- XtensaConstantPoolValue.cpp - Xtensa constantpool value ------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file implements the Xtensa specific constantpool value class.
+//
+//===----------------------------------------------------------------------===//
+
+#include "XtensaConstantPoolValue.h"
+#include "llvm/ADT/FoldingSet.h"
+#include "llvm/CodeGen/MachineBasicBlock.h"
+#include "llvm/IR/Constant.h"
+#include "llvm/IR/Constants.h"
+#include "llvm/IR/GlobalValue.h"
+#include "llvm/IR/Type.h"
+#include "llvm/Support/raw_ostream.h"
+#include <cstdlib>
+using namespace llvm;
+
+XtensaConstantPoolValue::XtensaConstantPoolValue(
+    Type *Ty, unsigned ID, XtensaCP::XtensaCPKind Kind,
+    XtensaCP::XtensaCPModifier modifier)
+    : MachineConstantPoolValue(Ty), LabelId(ID), Kind(Kind),
+      Modifier(modifier) {}
+
+XtensaConstantPoolValue::XtensaConstantPoolValue(
+    LLVMContext &C, unsigned ID, XtensaCP::XtensaCPKind Kind,
+    XtensaCP::XtensaCPModifier Modifier)
+    : MachineConstantPoolValue((Type *)Type::getInt32Ty(C)), LabelId(ID),
+      Kind(Kind), Modifier(Modifier) {}
+
+XtensaConstantPoolValue::~XtensaConstantPoolValue() {}
+
+StringRef XtensaConstantPoolValue::getModifierText() const {
+  switch (Modifier) {
+  case XtensaCP::no_modifier:
+    return "";
+  case XtensaCP::TPOFF:
+    return "@TPOFF";
+  }
+  report_fatal_error("Unknown modifier!");
+}
+
+int XtensaConstantPoolValue::getExistingMachineCPValue(MachineConstantPool *CP,
+                                                       Align Alignment) {
+  report_fatal_error("Shouldn't be calling this directly!");
+}
+
+void XtensaConstantPoolValue::addSelectionDAGCSEId(FoldingSetNodeID &ID) {
+  ID.AddInteger(LabelId);
+}
+
+bool XtensaConstantPoolValue::hasSameValue(XtensaConstantPoolValue *ACPV) {
+  if (ACPV->Kind == Kind) {
+    if (ACPV->LabelId == LabelId)
+      return true;
+  }
+  return false;
+}
+
+#if !defined(NDEBUG) || defined(LLVM_ENABLE_DUMP)
+void XtensaConstantPoolValue::dump() const { errs() << "  " << *this; }
+#endif
+
+void XtensaConstantPoolValue::print(raw_ostream &O) const {}
+
+//===----------------------------------------------------------------------===//
+// XtensaConstantPoolConstant
+//===----------------------------------------------------------------------===//
+
+XtensaConstantPoolConstant::XtensaConstantPoolConstant(
+    const Constant *C, unsigned ID, XtensaCP::XtensaCPKind Kind)
+    : XtensaConstantPoolValue(C->getType(), ID, Kind), CVal(C) {}
+
+XtensaConstantPoolConstant *
+XtensaConstantPoolConstant::Create(const Constant *C, unsigned ID,
+                                   XtensaCP::XtensaCPKind Kind) {
+  return new XtensaConstantPoolConstant(C, ID, Kind);
+}
+
+const BlockAddress *XtensaConstantPoolConstant::getBlockAddress() const {
+  return dyn_cast_or_null<BlockAddress>(CVal);
+}
+
+int XtensaConstantPoolConstant::getExistingMachineCPValue(
+    MachineConstantPool *CP, Align Alignment) {
+  return getExistingMachineCPValueImpl<XtensaConstantPoolConstant>(CP,
+                                                                   Alignment);
+}
+
+bool XtensaConstantPoolConstant::hasSameValue(XtensaConstantPoolValue *ACPV) {
+  const XtensaConstantPoolConstant *ACPC =
+      dyn_cast<XtensaConstantPoolConstant>(ACPV);
+  return ACPC && ACPC->CVal == CVal &&
+         XtensaConstantPoolValue::hasSameValue(ACPV);
+}
+
+void XtensaConstantPoolConstant::addSelectionDAGCSEId(FoldingSetNodeID &ID) {
+  ID.AddPointer(CVal);
+  XtensaConstantPoolValue::addSelectionDAGCSEId(ID);
+}
+
+void XtensaConstantPoolConstant::print(raw_ostream &O) const {
+  O << CVal->getName();
+  XtensaConstantPoolValue::print(O);
+}
+
+XtensaConstantPoolSymbol::XtensaConstantPoolSymbol(
+    LLVMContext &C, const char *Str, unsigned ID, bool PrivLinkage,
+    XtensaCP::XtensaCPModifier Modifier)
+    : XtensaConstantPoolValue(C, ID, XtensaCP::CPExtSymbol, Modifier), S(Str),
+      PrivateLinkage(PrivLinkage) {}
+
+XtensaConstantPoolSymbol *
+XtensaConstantPoolSymbol::Create(LLVMContext &C, const char *Str, unsigned ID,
+                                 bool PrivLinkage,
+                                 XtensaCP::XtensaCPModifier Modifier)
+
+{
+  return new XtensaConstantPoolSymbol(C, Str, ID, PrivLinkage, Modifier);
+}
+
+int XtensaConstantPoolSymbol::getExistingMachineCPValue(MachineConstantPool *CP,
+                                                        Align Alignment) {
+  return getExistingMachineCPValueImpl<XtensaConstantPoolSymbol>(CP, Alignment);
+}
+
+bool XtensaConstantPoolSymbol::hasSameValue(XtensaConstantPoolValue *ACPV) {
+  const XtensaConstantPoolSymbol *ACPS =
+      dyn_cast<XtensaConstantPoolSymbol>(ACPV);
+  return ACPS && ACPS->S == S && XtensaConstantPoolValue::hasSameValue(ACPV);
+}
+
+void XtensaConstantPoolSymbol::addSelectionDAGCSEId(FoldingSetNodeID &ID) {
+  ID.AddString(S);
+  XtensaConstantPoolValue::addSelectionDAGCSEId(ID);
+}
+
+void XtensaConstantPoolSymbol::print(raw_ostream &O) const {
+  O << S;
+  XtensaConstantPoolValue::print(O);
+}
+
+XtensaConstantPoolMBB::XtensaConstantPoolMBB(LLVMContext &C,
+                                             const MachineBasicBlock *M,
+                                             unsigned Id)
+    : XtensaConstantPoolValue(C, 0, XtensaCP::CPMachineBasicBlock), MBB(M) {}
+
+XtensaConstantPoolMBB *XtensaConstantPoolMBB::Create(LLVMContext &C,
+                                                     const MachineBasicBlock *M,
+                                                     unsigned Idx) {
+  return new XtensaConstantPoolMBB(C, M, Idx);
+}
+
+int XtensaConstantPoolMBB::getExistingMachineCPValue(MachineConstantPool *CP,
+                                                     Align Alignment) {
+  return getExistingMachineCPValueImpl<XtensaConstantPoolMBB>(CP, Alignment);
+}
+
+bool XtensaConstantPoolMBB::hasSameValue(XtensaConstantPoolValue *ACPV) {
+  const XtensaConstantPoolMBB *ACPMBB = dyn_cast<XtensaConstantPoolMBB>(ACPV);
+  return ACPMBB && ACPMBB->MBB == MBB &&
+         XtensaConstantPoolValue::hasSameValue(ACPV);
+}
+
+void XtensaConstantPoolMBB::addSelectionDAGCSEId(FoldingSetNodeID &ID) {
+  ID.AddPointer(MBB);
+  XtensaConstantPoolValue::addSelectionDAGCSEId(ID);
+}
+
+void XtensaConstantPoolMBB::print(raw_ostream &O) const {
+  O << "BB#" << MBB->getNumber();
+  XtensaConstantPoolValue::print(O);
+}
+
+XtensaConstantPoolJumpTable::XtensaConstantPoolJumpTable(LLVMContext &C,
+                                                         unsigned Index)
+    : XtensaConstantPoolValue(C, 0, XtensaCP::CPJumpTable), Idx(Index) {}
+
+XtensaConstantPoolJumpTable *XtensaConstantPoolJumpTable::Create(LLVMContext &C,
+                                                                 unsigned Idx) {
+  return new XtensaConstantPoolJumpTable(C, Idx);
+}
+
+int XtensaConstantPoolJumpTable::getExistingMachineCPValue(
+    MachineConstantPool *CP, Align Alignment) {
+  return getExistingMachineCPValueImpl<XtensaConstantPoolJumpTable>(CP,
+                                                                    Alignment);
+}
+
+bool XtensaConstantPoolJumpTable::hasSameValue(XtensaConstantPoolValue *ACPV) {
+  const XtensaConstantPoolJumpTable *ACPJT =
+      dyn_cast<XtensaConstantPoolJumpTable>(ACPV);
+  return ACPJT && ACPJT->Idx == Idx &&
+         XtensaConstantPoolValue::hasSameValue(ACPV);
+}
+
+void XtensaConstantPoolJumpTable::addSelectionDAGCSEId(FoldingSetNodeID &ID) {}
+
+void XtensaConstantPoolJumpTable::print(raw_ostream &O) const {
+  O << "JT" << Idx;
+  XtensaConstantPoolValue::print(O);
+}
diff --git a/llvm/lib/Target/Xtensa/XtensaConstantPoolValue.h b/llvm/lib/Target/Xtensa/XtensaConstantPoolValue.h
new file mode 100644
index 000000000000..5580de484474
--- /dev/null
+++ b/llvm/lib/Target/Xtensa/XtensaConstantPoolValue.h
@@ -0,0 +1,263 @@
+//===- XtensaConstantPoolValue.h - Xtensa constantpool value ----*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file implements the Xtensa specific constantpool value class.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_LIB_TARGET_XTENSA_XTENSACONSTANTPOOLVALUE_H
+#define LLVM_LIB_TARGET_XTENSA_XTENSACONSTANTPOOLVALUE_H
+
+#include "llvm/CodeGen/MachineConstantPool.h"
+#include "llvm/Support/Casting.h"
+#include "llvm/Support/ErrorHandling.h"
+#include <cstddef>
+#include <string>
+#include <vector>
+
+namespace llvm {
+
+class BlockAddress;
+class Constant;
+class GlobalValue;
+class LLVMContext;
+class MachineBasicBlock;
+
+namespace XtensaCP {
+enum XtensaCPKind {
+  CPExtSymbol,
+  CPBlockAddress,
+  CPMachineBasicBlock,
+  CPJumpTable
+};
+
+enum XtensaCPModifier {
+  no_modifier, // None
+  TPOFF        // Thread Pointer Offset
+};
+} // namespace XtensaCP
+
+/// XtensaConstantPoolValue - Xtensa specific constantpool value. This is used
+/// to represent PC-relative displacement between the address of the load
+/// instruction and the constant being loaded.
+class XtensaConstantPoolValue : public MachineConstantPoolValue {
+  unsigned LabelId;                    // Label id of the load.
+  XtensaCP::XtensaCPKind Kind;         // Kind of constant.
+  XtensaCP::XtensaCPModifier Modifier; // Symbol name modifier
+                                       //(for example Global Variable name)
+
+protected:
+  XtensaConstantPoolValue(
+      Type *Ty, unsigned ID, XtensaCP::XtensaCPKind Kind,
+      XtensaCP::XtensaCPModifier Modifier = XtensaCP::no_modifier);
+
+  XtensaConstantPoolValue(
+      LLVMContext &C, unsigned id, XtensaCP::XtensaCPKind Kind,
+      XtensaCP::XtensaCPModifier Modifier = XtensaCP::no_modifier);
+
+  template <typename Derived>
+  int getExistingMachineCPValueImpl(MachineConstantPool *CP, Align Alignment) {
+    const std::vector<MachineConstantPoolEntry> &Constants = CP->getConstants();
+    for (unsigned i = 0, e = Constants.size(); i != e; ++i) {
+      if (Constants[i].isMachineConstantPoolEntry() &&
+          (Constants[i].getAlign() >= Alignment)) {
+        auto *CPV = static_cast<XtensaConstantPoolValue *>(
+            Constants[i].Val.MachineCPVal);
+        if (Derived *APC = dyn_cast<Derived>(CPV))
+          if (cast<Derived>(this)->equals(APC))
+            return i;
+      }
+    }
+
+    return -1;
+  }
+
+public:
+  ~XtensaConstantPoolValue() override;
+
+  XtensaCP::XtensaCPModifier getModifier() const { return Modifier; }
+  bool hasModifier() const { return Modifier != XtensaCP::no_modifier; }
+  StringRef getModifierText() const;
+
+  unsigned getLabelId() const { return LabelId; }
+  void setLabelId(unsigned ID) { LabelId = ID; }
+
+  bool isExtSymbol() const { return Kind == XtensaCP::CPExtSymbol; }
+  bool isBlockAddress() const { return Kind == XtensaCP::CPBlockAddress; }
+  bool isMachineBasicBlock() const {
+    return Kind == XtensaCP::CPMachineBasicBlock;
+  }
+  bool isJumpTable() const { return Kind == XtensaCP::CPJumpTable; }
+
+  int getExistingMachineCPValue(MachineConstantPool *CP,
+                                Align Alignment) override;
+
+  void addSelectionDAGCSEId(FoldingSetNodeID &ID) override;
+
+  /// hasSameValue - Return true if this Xtensa constpool value can share the
+  /// same constantpool entry as another Xtensa constpool value.
+  virtual bool hasSameValue(XtensaConstantPoolValue *ACPV);
+
+  bool equals(const XtensaConstantPoolValue *A) const {
+    return this->LabelId == A->LabelId && this->Modifier == A->Modifier;
+  }
+
+  void print(raw_ostream &O) const override;
+
+#if !defined(NDEBUG) || defined(LLVM_ENABLE_DUMP)
+  void dump() const;
+#endif
+};
+
+inline raw_ostream &operator<<(raw_ostream &O,
+                               const XtensaConstantPoolValue &V) {
+  V.print(O);
+  return O;
+}
+
+/// XtensaConstantPoolConstant - Xtensa-specific constant pool values for
+/// Constants (for example BlockAddresses).
+class XtensaConstantPoolConstant : public XtensaConstantPoolValue {
+  const Constant *CVal; // Constant being loaded.
+
+  XtensaConstantPoolConstant(const Constant *C, unsigned ID,
+                             XtensaCP::XtensaCPKind Kind);
+
+public:
+  static XtensaConstantPoolConstant *Create(const Constant *C, unsigned ID,
+                                            XtensaCP::XtensaCPKind Kind);
+
+  const BlockAddress *getBlockAddress() const;
+
+  int getExistingMachineCPValue(MachineConstantPool *CP,
+                                Align Alignment) override;
+
+  /// hasSameValue - Return true if this Xtensa constpool value can share the
+  /// same constantpool entry as another Xtensa constpool value.
+  bool hasSameValue(XtensaConstantPoolValue *ACPV) override;
+
+  void addSelectionDAGCSEId(FoldingSetNodeID &ID) override;
+
+  void print(raw_ostream &O) const override;
+  static bool classof(const XtensaConstantPoolValue *APV) {
+    return APV->isBlockAddress();
+  }
+
+  bool equals(const XtensaConstantPoolConstant *A) const {
+    return CVal == A->CVal && XtensaConstantPoolValue::equals(A);
+  }
+};
+
+/// XtensaConstantPoolSymbol - Xtensa-specific constantpool values for external
+/// symbols.
+class XtensaConstantPoolSymbol : public XtensaConstantPoolValue {
+  const std::string S; // ExtSymbol being loaded.
+  bool PrivateLinkage;
+
+  XtensaConstantPoolSymbol(
+      LLVMContext &C, const char *S, unsigned Id, bool PrivLinkage,
+      XtensaCP::XtensaCPModifier Modifier = XtensaCP::no_modifier);
+
+public:
+  static XtensaConstantPoolSymbol *
+  Create(LLVMContext &C, const char *S, unsigned ID, bool PrivLinkage,
+         XtensaCP::XtensaCPModifier Modifier = XtensaCP::no_modifier);
+
+  const char *getSymbol() const { return S.c_str(); }
+
+  int getExistingMachineCPValue(MachineConstantPool *CP,
+                                Align Alignment) override;
+
+  void addSelectionDAGCSEId(FoldingSetNodeID &ID) override;
+
+  /// hasSameValue - Return true if this Xtensa constpool value can share the
+  /// same constantpool entry as another Xtensa constpool value.
+  bool hasSameValue(XtensaConstantPoolValue *ACPV) override;
+
+  bool isPrivateLinkage() { return PrivateLinkage; }
+
+  void print(raw_ostream &O) const override;
+
+  static bool classof(const XtensaConstantPoolValue *ACPV) {
+    return ACPV->isExtSymbol();
+  }
+
+  bool equals(const XtensaConstantPoolSymbol *A) const {
+    return S == A->S && XtensaConstantPoolValue::equals(A);
+  }
+};
+
+/// XtensaConstantPoolMBB - Xtensa-specific constantpool value of a machine
+/// basic block.
+class XtensaConstantPoolMBB : public XtensaConstantPoolValue {
+  const MachineBasicBlock *MBB; // Machine basic block.
+
+  XtensaConstantPoolMBB(LLVMContext &C, const MachineBasicBlock *M,
+                        unsigned ID);
+
+public:
+  static XtensaConstantPoolMBB *Create(LLVMContext &C,
+                                       const MachineBasicBlock *M, unsigned ID);
+
+  const MachineBasicBlock *getMBB() const { return MBB; }
+
+  int getExistingMachineCPValue(MachineConstantPool *CP,
+                                Align Alignment) override;
+
+  void addSelectionDAGCSEId(FoldingSetNodeID &ID) override;
+
+  /// hasSameValue - Return true if this Xtensa constpool value can share the
+  /// same constantpool entry as another Xtensa constpool value.
+  bool hasSameValue(XtensaConstantPoolValue *ACPV) override;
+
+  void print(raw_ostream &O) const override;
+
+  static bool classof(const XtensaConstantPoolValue *ACPV) {
+    return ACPV->isMachineBasicBlock();
+  }
+
+  bool equals(const XtensaConstantPoolMBB *A) const {
+    return MBB == A->MBB && XtensaConstantPoolValue::equals(A);
+  }
+};
+
+/// XtensaConstantPoolJumpTable - Xtensa-specific constantpool values for Jump
+/// Table symbols.
+class XtensaConstantPoolJumpTable : public XtensaConstantPoolValue {
+  unsigned Idx; // Jump Table Index.
+
+  XtensaConstantPoolJumpTable(LLVMContext &C, unsigned Idx);
+
+public:
+  static XtensaConstantPoolJumpTable *Create(LLVMContext &C, unsigned Idx);
+
+  unsigned getIndex() const { return Idx; }
+
+  int getExistingMachineCPValue(MachineConstantPool *CP,
+                                Align Alignment) override;
+
+  void addSelectionDAGCSEId(FoldingSetNodeID &ID) override;
+
+  /// hasSameValue - Return true if this Xtensa constpool value can share the
+  /// same constantpool entry as another Xtensa constpool value.
+  bool hasSameValue(XtensaConstantPoolValue *ACPV) override;
+
+  void print(raw_ostream &O) const override;
+
+  static bool classof(const XtensaConstantPoolValue *ACPV) {
+    return ACPV->isJumpTable();
+  }
+
+  bool equals(const XtensaConstantPoolJumpTable *A) const {
+    return Idx == A->Idx && XtensaConstantPoolValue::equals(A);
+  }
+};
+
+} // namespace llvm
+
+#endif /* LLVM_LIB_TARGET_XTENSA_XTENSACONSTANTPOOLVALUE_H */
diff --git a/llvm/lib/Target/Xtensa/XtensaFrameLowering.cpp b/llvm/lib/Target/Xtensa/XtensaFrameLowering.cpp
index 2092a2d947f8..ab37c09bf8bf 100644
--- a/llvm/lib/Target/Xtensa/XtensaFrameLowering.cpp
+++ b/llvm/lib/Target/Xtensa/XtensaFrameLowering.cpp
@@ -37,3 +37,23 @@ void XtensaFrameLowering::emitPrologue(MachineFunction &MF,
 
 void XtensaFrameLowering::emitEpilogue(MachineFunction &MF,
                                        MachineBasicBlock &MBB) const {}
+
+// Eliminate ADJCALLSTACKDOWN, ADJCALLSTACKUP pseudo instructions
+MachineBasicBlock::iterator XtensaFrameLowering::eliminateCallFramePseudoInstr(
+    MachineFunction &MF, MachineBasicBlock &MBB,
+    MachineBasicBlock::iterator I) const {
+  const XtensaInstrInfo &TII =
+      *static_cast<const XtensaInstrInfo *>(MF.getSubtarget().getInstrInfo());
+
+  if (!hasReservedCallFrame(MF)) {
+    int64_t Amount = I->getOperand(0).getImm();
+
+    if (I->getOpcode() == Xtensa::ADJCALLSTACKDOWN)
+      Amount = -Amount;
+
+    unsigned SP = Xtensa::SP;
+    TII.adjustStackPtr(SP, Amount, MBB, I);
+  }
+
+  return MBB.erase(I);
+}
diff --git a/llvm/lib/Target/Xtensa/XtensaFrameLowering.h b/llvm/lib/Target/Xtensa/XtensaFrameLowering.h
index 19e52310a99d..2da88ab14073 100644
--- a/llvm/lib/Target/Xtensa/XtensaFrameLowering.h
+++ b/llvm/lib/Target/Xtensa/XtensaFrameLowering.h
@@ -25,6 +25,10 @@ public:
   /// the function.
   void emitPrologue(MachineFunction &, MachineBasicBlock &) const override;
   void emitEpilogue(MachineFunction &MF, MachineBasicBlock &MBB) const override;
+
+  MachineBasicBlock::iterator
+  eliminateCallFramePseudoInstr(MachineFunction &MF, MachineBasicBlock &MBB,
+                                MachineBasicBlock::iterator I) const override;
 };
 
 } // namespace llvm
diff --git a/llvm/lib/Target/Xtensa/XtensaISelDAGToDAG.cpp b/llvm/lib/Target/Xtensa/XtensaISelDAGToDAG.cpp
index 300737275452..5ebedefafc16 100644
--- a/llvm/lib/Target/Xtensa/XtensaISelDAGToDAG.cpp
+++ b/llvm/lib/Target/Xtensa/XtensaISelDAGToDAG.cpp
@@ -12,9 +12,11 @@
 
 #include "Xtensa.h"
 #include "XtensaTargetMachine.h"
+#include "XtensaUtils.h"
 #include "llvm/CodeGen/MachineFunction.h"
 #include "llvm/CodeGen/MachineRegisterInfo.h"
 #include "llvm/CodeGen/SelectionDAGISel.h"
+#include "llvm/IR/DiagnosticInfo.h"
 #include "llvm/Support/Debug.h"
 #include "llvm/Support/raw_ostream.h"
 
@@ -37,9 +39,57 @@ public:
 
   void Select(SDNode *Node) override;
 
+  // For load/store instructions generate (base+offset) pair from
+  // memory address. The offset must be a multiple of scale argument.
   bool selectMemRegAddr(SDValue Addr, SDValue &Base, SDValue &Offset,
                         int Scale) {
-    report_fatal_error("MemReg address is not implemented yet");
+    EVT ValTy = Addr.getValueType();
+
+    // if Address is FI, get the TargetFrameIndex.
+    if (FrameIndexSDNode *FIN = dyn_cast<FrameIndexSDNode>(Addr)) {
+      Base = CurDAG->getTargetFrameIndex(FIN->getIndex(), ValTy);
+      Offset = CurDAG->getTargetConstant(0, SDLoc(Addr), ValTy);
+
+      return true;
+    }
+
+    if (TM.isPositionIndependent()) {
+      DiagnosticInfoUnsupported Diag(CurDAG->getMachineFunction().getFunction(),
+                                     "PIC relocations are not supported ",
+                                     Addr.getDebugLoc());
+      CurDAG->getContext()->diagnose(Diag);
+    }
+
+    if ((Addr.getOpcode() == ISD::TargetExternalSymbol ||
+         Addr.getOpcode() == ISD::TargetGlobalAddress))
+      return false;
+
+    // Addresses of the form FI+const
+    bool Valid = false;
+    if (CurDAG->isBaseWithConstantOffset(Addr)) {
+      ConstantSDNode *CN = dyn_cast<ConstantSDNode>(Addr.getOperand(1));
+      int64_t OffsetVal = CN->getSExtValue();
+
+      Valid = isValidAddrOffset(Scale, OffsetVal);
+
+      if (Valid) {
+        // If the first operand is a FI, get the TargetFI Node
+        if (FrameIndexSDNode *FIN =
+                dyn_cast<FrameIndexSDNode>(Addr.getOperand(0)))
+          Base = CurDAG->getTargetFrameIndex(FIN->getIndex(), ValTy);
+        else
+          Base = Addr.getOperand(0);
+
+        Offset =
+            CurDAG->getTargetConstant(CN->getZExtValue(), SDLoc(Addr), ValTy);
+        return true;
+      }
+    }
+
+    // Last case
+    Base = Addr;
+    Offset = CurDAG->getTargetConstant(0, SDLoc(Addr), Addr.getValueType());
+    return true;
   }
 
   bool selectMemRegAddrISH1(SDValue Addr, SDValue &Base, SDValue &Offset) {
diff --git a/llvm/lib/Target/Xtensa/XtensaISelLowering.cpp b/llvm/lib/Target/Xtensa/XtensaISelLowering.cpp
index 276fab838d17..64b996b6a42e 100644
--- a/llvm/lib/Target/Xtensa/XtensaISelLowering.cpp
+++ b/llvm/lib/Target/Xtensa/XtensaISelLowering.cpp
@@ -12,6 +12,7 @@
 //===----------------------------------------------------------------------===//
 
 #include "XtensaISelLowering.h"
+#include "XtensaConstantPoolValue.h"
 #include "XtensaSubtarget.h"
 #include "XtensaTargetMachine.h"
 #include "llvm/CodeGen/CallingConvLower.h"
@@ -22,15 +23,26 @@
 #include "llvm/CodeGen/TargetLoweringObjectFileImpl.h"
 #include "llvm/Support/Debug.h"
 #include "llvm/Support/ErrorHandling.h"
+#include "llvm/Support/MathExtras.h"
 #include "llvm/Support/raw_ostream.h"
+#include <deque>
 
 using namespace llvm;
 
 #define DEBUG_TYPE "xtensa-lower"
 
+// Return true if we must use long (in fact, indirect) function call.
+// It's simplified version, production implimentation must
+// resolve a functions in ROM (usually glibc functions)
+static bool isLongCall(const char *str) {
+  // Currently always use long calls
+  return true;
+}
+
 XtensaTargetLowering::XtensaTargetLowering(const TargetMachine &TM,
                                            const XtensaSubtarget &STI)
     : TargetLowering(TM), Subtarget(STI) {
+  MVT PtrVT = MVT::i32;
   // Set up the register classes.
   addRegisterClass(MVT::i32, &Xtensa::ARRegClass);
 
@@ -41,18 +53,507 @@ XtensaTargetLowering::XtensaTargetLowering(const TargetMachine &TM,
 
   setMinFunctionAlignment(Align(4));
 
+  setOperationAction(ISD::Constant, MVT::i32, Custom);
+  setOperationAction(ISD::Constant, MVT::i64, Expand);
+
+  setBooleanContents(ZeroOrOneBooleanContent);
+
+  setOperationAction(ISD::SIGN_EXTEND_INREG, MVT::i1, Expand);
+  setOperationAction(ISD::SIGN_EXTEND_INREG, MVT::i8, Expand);
+  setOperationAction(ISD::SIGN_EXTEND_INREG, MVT::i16, Expand);
+
+  setOperationAction(ISD::BITCAST, MVT::i32, Expand);
+  setOperationAction(ISD::BITCAST, MVT::f32, Expand);
+  setOperationAction(ISD::UINT_TO_FP, MVT::i32, Expand);
+  setOperationAction(ISD::SINT_TO_FP, MVT::i32, Expand);
+  setOperationAction(ISD::FP_TO_UINT, MVT::i32, Expand);
+  setOperationAction(ISD::FP_TO_SINT, MVT::i32, Expand);
+
+  // No sign extend instructions for i1
+  for (MVT VT : MVT::integer_valuetypes()) {
+    setLoadExtAction(ISD::SEXTLOAD, VT, MVT::i1, Promote);
+    setLoadExtAction(ISD::ZEXTLOAD, VT, MVT::i1, Promote);
+    setLoadExtAction(ISD::EXTLOAD, VT, MVT::i1, Promote);
+  }
+
+  setOperationAction(ISD::ConstantPool, PtrVT, Custom);
+
   // Compute derived properties from the register classes
   computeRegisterProperties(STI.getRegisterInfo());
 }
 
+//===----------------------------------------------------------------------===//
+// Calling conventions
+//===----------------------------------------------------------------------===//
+
+#include "XtensaGenCallingConv.inc"
+
+static bool CC_Xtensa_Custom(unsigned ValNo, MVT ValVT, MVT LocVT,
+                             CCValAssign::LocInfo LocInfo,
+                             ISD::ArgFlagsTy ArgFlags, CCState &State) {
+  static const MCPhysReg IntRegs[] = {Xtensa::A2, Xtensa::A3, Xtensa::A4,
+                                      Xtensa::A5, Xtensa::A6, Xtensa::A7};
+
+  if (ArgFlags.isByVal()) {
+    Align ByValAlign = ArgFlags.getNonZeroByValAlign();
+    unsigned ByValSize = ArgFlags.getByValSize();
+    if (ByValSize < 4) {
+      ByValSize = 4;
+    }
+    if (ByValAlign < Align(4)) {
+      ByValAlign = Align(4);
+    }
+    unsigned Offset = State.AllocateStack(ByValSize, ByValAlign);
+    State.addLoc(CCValAssign::getMem(ValNo, ValVT, Offset, LocVT, LocInfo));
+    // Mark all unused registers as allocated to avoid misuse
+    // of such registers.
+    while (State.AllocateReg(IntRegs))
+      ;
+    return false;
+  }
+
+  // Promote i8 and i16
+  if (LocVT == MVT::i8 || LocVT == MVT::i16) {
+    LocVT = MVT::i32;
+    if (ArgFlags.isSExt())
+      LocInfo = CCValAssign::SExt;
+    else if (ArgFlags.isZExt())
+      LocInfo = CCValAssign::ZExt;
+    else
+      LocInfo = CCValAssign::AExt;
+  }
+
+  unsigned Register;
+
+  Align OrigAlign = ArgFlags.getNonZeroOrigAlign();
+  bool needs64BitAlign = (ValVT == MVT::i32 && OrigAlign == Align(8));
+  bool needs128BitAlign = (ValVT == MVT::i32 && OrigAlign == Align(16));
+
+  if (ValVT == MVT::i32) {
+    Register = State.AllocateReg(IntRegs);
+    // If this is the first part of an i64 arg,
+    // the allocated register must be either A2, A4 or A6.
+    if (needs64BitAlign && (Register == Xtensa::A3 || Register == Xtensa::A5 ||
+                            Register == Xtensa::A7))
+      Register = State.AllocateReg(IntRegs);
+    // arguments with 16byte alignment must be passed in the first register or
+    // passed via stack
+    if (needs128BitAlign && (Register != Xtensa::A2))
+      while ((Register = State.AllocateReg(IntRegs)))
+        ;
+    LocVT = MVT::i32;
+  } else if (ValVT == MVT::f64) {
+    // Allocate int register and shadow next int register.
+    Register = State.AllocateReg(IntRegs);
+    if (Register == Xtensa::A3 || Register == Xtensa::A5 ||
+        Register == Xtensa::A7)
+      Register = State.AllocateReg(IntRegs);
+    State.AllocateReg(IntRegs);
+    LocVT = MVT::i32;
+  } else {
+    report_fatal_error("Cannot handle this ValVT.");
+  }
+
+  if (!Register) {
+    unsigned Offset = State.AllocateStack(ValVT.getStoreSize(), OrigAlign);
+    State.addLoc(CCValAssign::getMem(ValNo, ValVT, Offset, LocVT, LocInfo));
+  } else {
+    State.addLoc(CCValAssign::getReg(ValNo, ValVT, Register, LocVT, LocInfo));
+  }
+
+  return false;
+}
+
+CCAssignFn *XtensaTargetLowering::CCAssignFnForCall(CallingConv::ID CC,
+                                                    bool IsVarArg) const {
+  return CC_Xtensa_Custom;
+}
+
+SDValue XtensaTargetLowering::LowerFormalArguments(
+    SDValue Chain, CallingConv::ID CallConv, bool IsVarArg,
+    const SmallVectorImpl<ISD::InputArg> &Ins, const SDLoc &DL,
+    SelectionDAG &DAG, SmallVectorImpl<SDValue> &InVals) const {
+  MachineFunction &MF = DAG.getMachineFunction();
+  MachineFrameInfo &MFI = MF.getFrameInfo();
+
+  // Used with vargs to acumulate store chains.
+  std::vector<SDValue> OutChains;
+
+  if (IsVarArg)
+    report_fatal_error("Var arg not supported by FormalArguments Lowering");
+
+  // Assign locations to all of the incoming arguments.
+  SmallVector<CCValAssign, 16> ArgLocs;
+  CCState CCInfo(CallConv, IsVarArg, DAG.getMachineFunction(), ArgLocs,
+                 *DAG.getContext());
+
+  CCInfo.AnalyzeFormalArguments(Ins, CCAssignFnForCall(CallConv, IsVarArg));
+
+  for (unsigned i = 0, e = ArgLocs.size(); i != e; ++i) {
+    CCValAssign &VA = ArgLocs[i];
+    // Arguments stored on registers
+    if (VA.isRegLoc()) {
+      EVT RegVT = VA.getLocVT();
+      const TargetRegisterClass *RC;
+
+      if (RegVT == MVT::i32)
+        RC = &Xtensa::ARRegClass;
+      else
+        report_fatal_error("RegVT not supported by FormalArguments Lowering");
+
+      // Transform the arguments stored on
+      // physical registers into virtual ones
+      unsigned Register = MF.addLiveIn(VA.getLocReg(), RC);
+      SDValue ArgValue = DAG.getCopyFromReg(Chain, DL, Register, RegVT);
+
+      // If this is an 8 or 16-bit value, it has been passed promoted
+      // to 32 bits.  Insert an assert[sz]ext to capture this, then
+      // truncate to the right size.
+      if (VA.getLocInfo() != CCValAssign::Full) {
+        unsigned Opcode = 0;
+        if (VA.getLocInfo() == CCValAssign::SExt)
+          Opcode = ISD::AssertSext;
+        else if (VA.getLocInfo() == CCValAssign::ZExt)
+          Opcode = ISD::AssertZext;
+        if (Opcode)
+          ArgValue = DAG.getNode(Opcode, DL, RegVT, ArgValue,
+                                 DAG.getValueType(VA.getValVT()));
+        ArgValue = DAG.getNode((VA.getValVT() == MVT::f32) ? ISD::BITCAST
+                                                           : ISD::TRUNCATE,
+                               DL, VA.getValVT(), ArgValue);
+      }
+
+      InVals.push_back(ArgValue);
+
+    } else {
+      assert(VA.isMemLoc());
+
+      EVT ValVT = VA.getValVT();
+
+      // The stack pointer offset is relative to the caller stack frame.
+      int FI = MFI.CreateFixedObject(ValVT.getStoreSize(), VA.getLocMemOffset(),
+                                     true);
+
+      if (Ins[VA.getValNo()].Flags.isByVal()) {
+        // Assume that in this case load operation is created
+        SDValue FIN = DAG.getFrameIndex(FI, MVT::i32);
+        InVals.push_back(FIN);
+      } else {
+        // Create load nodes to retrieve arguments from the stack
+        SDValue FIN =
+            DAG.getFrameIndex(FI, getFrameIndexTy(DAG.getDataLayout()));
+        InVals.push_back(DAG.getLoad(
+            ValVT, DL, Chain, FIN,
+            MachinePointerInfo::getFixedStack(DAG.getMachineFunction(), FI)));
+      }
+    }
+  }
+
+  // All stores are grouped in one node to allow the matching between
+  // the size of Ins and InVals. This only happens when on varg functions
+  if (!OutChains.empty()) {
+    OutChains.push_back(Chain);
+    Chain = DAG.getNode(ISD::TokenFactor, DL, MVT::Other, OutChains);
+  }
+
+  return Chain;
+}
+
+SDValue
+XtensaTargetLowering::LowerCall(CallLoweringInfo &CLI,
+                                SmallVectorImpl<SDValue> &InVals) const {
+  SelectionDAG &DAG = CLI.DAG;
+  SDLoc &DL = CLI.DL;
+  SmallVector<ISD::OutputArg, 32> &Outs = CLI.Outs;
+  SmallVector<SDValue, 32> &OutVals = CLI.OutVals;
+  SmallVector<ISD::InputArg, 32> &Ins = CLI.Ins;
+  SDValue Chain = CLI.Chain;
+  SDValue Callee = CLI.Callee;
+  bool &IsTailCall = CLI.IsTailCall;
+  CallingConv::ID CallConv = CLI.CallConv;
+  bool IsVarArg = CLI.IsVarArg;
+
+  MachineFunction &MF = DAG.getMachineFunction();
+  EVT PtrVT = getPointerTy(DAG.getDataLayout());
+  const TargetFrameLowering *TFL = Subtarget.getFrameLowering();
+
+  // TODO: Support tail call optimization.
+  IsTailCall = false;
+
+  // Analyze the operands of the call, assigning locations to each operand.
+  SmallVector<CCValAssign, 16> ArgLocs;
+  CCState CCInfo(CallConv, IsVarArg, MF, ArgLocs, *DAG.getContext());
+
+  CCAssignFn *CC = CCAssignFnForCall(CallConv, IsVarArg);
+
+  CCInfo.AnalyzeCallOperands(Outs, CC);
+
+  // Get a count of how many bytes are to be pushed on the stack.
+  unsigned NumBytes = CCInfo.getStackSize();
+
+  Align StackAlignment = TFL->getStackAlign();
+  unsigned NextStackOffset = alignTo(NumBytes, StackAlignment);
+
+  Chain = DAG.getCALLSEQ_START(Chain, NextStackOffset, 0, DL);
+
+  // Copy argument values to their designated locations.
+  std::deque<std::pair<unsigned, SDValue>> RegsToPass;
+  SmallVector<SDValue, 8> MemOpChains;
+  SDValue StackPtr;
+  for (unsigned I = 0, E = ArgLocs.size(); I != E; ++I) {
+    CCValAssign &VA = ArgLocs[I];
+    SDValue ArgValue = OutVals[I];
+    ISD::ArgFlagsTy Flags = Outs[I].Flags;
+
+    if (VA.isRegLoc())
+      // Queue up the argument copies and emit them at the end.
+      RegsToPass.push_back(std::make_pair(VA.getLocReg(), ArgValue));
+    else if (Flags.isByVal()) {
+      assert(VA.isMemLoc());
+      assert(Flags.getByValSize() &&
+             "ByVal args of size 0 should have been ignored by front-end.");
+      assert(!IsTailCall &&
+             "Do not tail-call optimize if there is a byval argument.");
+
+      if (!StackPtr.getNode())
+        StackPtr = DAG.getCopyFromReg(Chain, DL, Xtensa::SP, PtrVT);
+      unsigned Offset = VA.getLocMemOffset();
+      SDValue Address = DAG.getNode(ISD::ADD, DL, PtrVT, StackPtr,
+                                    DAG.getIntPtrConstant(Offset, DL));
+      SDValue SizeNode = DAG.getConstant(Flags.getByValSize(), DL, MVT::i32);
+      SDValue Memcpy = DAG.getMemcpy(
+          Chain, DL, Address, ArgValue, SizeNode, Flags.getNonZeroByValAlign(),
+          /*isVolatile=*/false, /*AlwaysInline=*/false,
+          /*isTailCall=*/false, MachinePointerInfo(), MachinePointerInfo());
+      MemOpChains.push_back(Memcpy);
+    } else {
+      assert(VA.isMemLoc() && "Argument not register or memory");
+
+      // Work out the address of the stack slot.  Unpromoted ints and
+      // floats are passed as right-justified 8-byte values.
+      if (!StackPtr.getNode())
+        StackPtr = DAG.getCopyFromReg(Chain, DL, Xtensa::SP, PtrVT);
+      unsigned Offset = VA.getLocMemOffset();
+      SDValue Address = DAG.getNode(ISD::ADD, DL, PtrVT, StackPtr,
+                                    DAG.getIntPtrConstant(Offset, DL));
+
+      // Emit the store.
+      MemOpChains.push_back(
+          DAG.getStore(Chain, DL, ArgValue, Address, MachinePointerInfo()));
+    }
+  }
+
+  // Join the stores, which are independent of one another.
+  if (!MemOpChains.empty())
+    Chain = DAG.getNode(ISD::TokenFactor, DL, MVT::Other, MemOpChains);
+
+  // Build a sequence of copy-to-reg nodes, chained and glued together.
+  SDValue Glue;
+  for (unsigned I = 0, E = RegsToPass.size(); I != E; ++I) {
+    unsigned Reg = RegsToPass[I].first;
+    Chain = DAG.getCopyToReg(Chain, DL, Reg, RegsToPass[I].second, Glue);
+    Glue = Chain.getValue(1);
+  }
+  std::string name;
+  unsigned char TF = 0;
+
+  // Accept direct calls by converting symbolic call addresses to the
+  // associated Target* opcodes.
+  if (ExternalSymbolSDNode *E = dyn_cast<ExternalSymbolSDNode>(Callee)) {
+    name = E->getSymbol();
+    TF = E->getTargetFlags();
+    if (isPositionIndependent()) {
+      report_fatal_error("PIC relocations is not supported");
+    } else
+      Callee = DAG.getTargetExternalSymbol(E->getSymbol(), PtrVT, TF);
+  } else if (GlobalAddressSDNode *G = dyn_cast<GlobalAddressSDNode>(Callee)) {
+    const GlobalValue *GV = G->getGlobal();
+    name = GV->getName().str();
+  }
+
+  if ((!name.empty()) && isLongCall(name.c_str())) {
+    // Create a constant pool entry for the callee address
+    XtensaCP::XtensaCPModifier Modifier = XtensaCP::no_modifier;
+
+    XtensaConstantPoolValue *CPV = XtensaConstantPoolSymbol::Create(
+        *DAG.getContext(), name.c_str(), 0 /* XtensaCLabelIndex */, false,
+        Modifier);
+
+    // Get the address of the callee into a register
+    SDValue CPAddr = DAG.getTargetConstantPool(CPV, PtrVT, Align(4), 0, TF);
+    SDValue CPWrap = getAddrPCRel(CPAddr, DAG);
+    Callee = CPWrap;
+  }
+
+  // The first call operand is the chain and the second is the target address.
+  SmallVector<SDValue, 8> Ops;
+  Ops.push_back(Chain);
+  Ops.push_back(Callee);
+
+  // Add a register mask operand representing the call-preserved registers.
+  const TargetRegisterInfo *TRI = Subtarget.getRegisterInfo();
+  const uint32_t *Mask = TRI->getCallPreservedMask(MF, CallConv);
+  assert(Mask && "Missing call preserved mask for calling convention");
+  Ops.push_back(DAG.getRegisterMask(Mask));
+
+  // Add argument registers to the end of the list so that they are
+  // known live into the call.
+  for (unsigned I = 0, E = RegsToPass.size(); I != E; ++I) {
+    unsigned Reg = RegsToPass[I].first;
+    Ops.push_back(DAG.getRegister(Reg, RegsToPass[I].second.getValueType()));
+  }
+
+  // Glue the call to the argument copies, if any.
+  if (Glue.getNode())
+    Ops.push_back(Glue);
+
+  SDVTList NodeTys = DAG.getVTList(MVT::Other, MVT::Glue);
+  Chain = DAG.getNode(XtensaISD::CALL, DL, NodeTys, Ops);
+  Glue = Chain.getValue(1);
+
+  // Mark the end of the call, which is glued to the call itself.
+  Chain = DAG.getCALLSEQ_END(Chain, DAG.getConstant(NumBytes, DL, PtrVT, true),
+                             DAG.getConstant(0, DL, PtrVT, true), Glue, DL);
+  Glue = Chain.getValue(1);
+
+  // Assign locations to each value returned by this call.
+  SmallVector<CCValAssign, 16> RetLocs;
+  CCState RetCCInfo(CallConv, IsVarArg, MF, RetLocs, *DAG.getContext());
+  RetCCInfo.AnalyzeCallResult(Ins, RetCC_Xtensa);
+
+  // Copy all of the result registers out of their specified physreg.
+  for (unsigned I = 0, E = RetLocs.size(); I != E; ++I) {
+    CCValAssign &VA = RetLocs[I];
+
+    // Copy the value out, gluing the copy to the end of the call sequence.
+    unsigned Reg = VA.getLocReg();
+    SDValue RetValue = DAG.getCopyFromReg(Chain, DL, Reg, VA.getLocVT(), Glue);
+    Chain = RetValue.getValue(1);
+    Glue = RetValue.getValue(2);
+
+    InVals.push_back(RetValue);
+  }
+  return Chain;
+}
+
+bool XtensaTargetLowering::CanLowerReturn(
+    CallingConv::ID CallConv, MachineFunction &MF, bool IsVarArg,
+    const SmallVectorImpl<ISD::OutputArg> &Outs, LLVMContext &Context) const {
+  SmallVector<CCValAssign, 16> RVLocs;
+  CCState CCInfo(CallConv, IsVarArg, MF, RVLocs, Context);
+  return CCInfo.CheckReturn(Outs, RetCC_Xtensa);
+}
+
+SDValue
+XtensaTargetLowering::LowerReturn(SDValue Chain, CallingConv::ID CallConv,
+                                  bool IsVarArg,
+                                  const SmallVectorImpl<ISD::OutputArg> &Outs,
+                                  const SmallVectorImpl<SDValue> &OutVals,
+                                  const SDLoc &DL, SelectionDAG &DAG) const {
+  if (IsVarArg)
+    report_fatal_error("VarArg not supported");
+
+  MachineFunction &MF = DAG.getMachineFunction();
+
+  // Assign locations to each returned value.
+  SmallVector<CCValAssign, 16> RetLocs;
+  CCState RetCCInfo(CallConv, IsVarArg, MF, RetLocs, *DAG.getContext());
+  RetCCInfo.AnalyzeReturn(Outs, RetCC_Xtensa);
+
+  SDValue Glue;
+  // Quick exit for void returns
+  if (RetLocs.empty())
+    return DAG.getNode(XtensaISD::RET, DL, MVT::Other, Chain);
+
+  // Copy the result values into the output registers.
+  SmallVector<SDValue, 4> RetOps;
+  RetOps.push_back(Chain);
+  for (unsigned I = 0, E = RetLocs.size(); I != E; ++I) {
+    CCValAssign &VA = RetLocs[I];
+    SDValue RetValue = OutVals[I];
+
+    // Make the return register live on exit.
+    assert(VA.isRegLoc() && "Can only return in registers!");
+
+    // Chain and glue the copies together.
+    unsigned Register = VA.getLocReg();
+    Chain = DAG.getCopyToReg(Chain, DL, Register, RetValue, Glue);
+    Glue = Chain.getValue(1);
+    RetOps.push_back(DAG.getRegister(Register, VA.getLocVT()));
+  }
+
+  // Update chain and glue.
+  RetOps[0] = Chain;
+  if (Glue.getNode())
+    RetOps.push_back(Glue);
+
+  return DAG.getNode(XtensaISD::RET, DL, MVT::Other, RetOps);
+}
+
+SDValue XtensaTargetLowering::LowerImmediate(SDValue Op,
+                                             SelectionDAG &DAG) const {
+  const ConstantSDNode *CN = cast<ConstantSDNode>(Op);
+  SDLoc DL(CN);
+  APInt APVal = CN->getAPIntValue();
+  int64_t Value = APVal.getSExtValue();
+  if (Op.getValueType() == MVT::i32) {
+    // Check if use node maybe lowered to the MOVI instruction
+    if (Value > -2048 && Value <= 2047)
+      return Op;
+    // Check if use node maybe lowered to the ADDMI instruction
+    SDNode &OpNode = *Op.getNode();
+    if ((OpNode.hasOneUse() && OpNode.use_begin()->getOpcode() == ISD::ADD) &&
+        isShiftedInt<16, 8>(Value))
+      return Op;
+    Type *Ty = Type::getInt32Ty(*DAG.getContext());
+    Constant *CV = ConstantInt::get(Ty, Value);
+    SDValue CP = DAG.getConstantPool(CV, MVT::i32);
+    return CP;
+  }
+  return Op;
+}
+
+SDValue XtensaTargetLowering::getAddrPCRel(SDValue Op,
+                                           SelectionDAG &DAG) const {
+  SDLoc DL(Op);
+  EVT Ty = Op.getValueType();
+  return DAG.getNode(XtensaISD::PCREL_WRAPPER, DL, Ty, Op);
+}
+
+SDValue XtensaTargetLowering::LowerConstantPool(ConstantPoolSDNode *CP,
+                                                SelectionDAG &DAG) const {
+  EVT PtrVT = getPointerTy(DAG.getDataLayout());
+  SDValue Result;
+  if (!CP->isMachineConstantPoolEntry()) {
+    Result = DAG.getTargetConstantPool(CP->getConstVal(), PtrVT, CP->getAlign(),
+                                       CP->getOffset());
+  } else {
+    report_fatal_error("This constantpool type is not supported yet");
+  }
+
+  return getAddrPCRel(Result, DAG);
+}
+
 SDValue XtensaTargetLowering::LowerOperation(SDValue Op,
                                              SelectionDAG &DAG) const {
   switch (Op.getOpcode()) {
+  case ISD::Constant:
+    return LowerImmediate(Op, DAG);
+  case ISD::ConstantPool:
+    return LowerConstantPool(cast<ConstantPoolSDNode>(Op), DAG);
   default:
     report_fatal_error("Unexpected node to lower");
   }
 }
 
 const char *XtensaTargetLowering::getTargetNodeName(unsigned Opcode) const {
+  switch (Opcode) {
+  case XtensaISD::CALL:
+    return "XtensaISD::CALL";
+  case XtensaISD::PCREL_WRAPPER:
+    return "XtensaISD::PCREL_WRAPPER";
+  case XtensaISD::RET:
+    return "XtensaISD::RET";
+  }
   return nullptr;
 }
diff --git a/llvm/lib/Target/Xtensa/XtensaISelLowering.h b/llvm/lib/Target/Xtensa/XtensaISelLowering.h
index 8b03712efc9b..077559e2d612 100644
--- a/llvm/lib/Target/Xtensa/XtensaISelLowering.h
+++ b/llvm/lib/Target/Xtensa/XtensaISelLowering.h
@@ -19,6 +19,23 @@
 #include "llvm/CodeGen/TargetLowering.h"
 
 namespace llvm {
+
+namespace XtensaISD {
+enum {
+  FIRST_NUMBER = ISD::BUILTIN_OP_END,
+
+  // Calls a function.  Operand 0 is the chain operand and operand 1
+  // is the target address.  The arguments start at operand 2.
+  // There is an optional glue operand at the end.
+  CALL,
+
+  // Wraps a TargetGlobalAddress that should be loaded using PC-relative
+  // accesses.  Operand 0 is the address.
+  PCREL_WRAPPER,
+  RET
+};
+}
+
 class XtensaSubtarget;
 
 class XtensaTargetLowering : public TargetLowering {
@@ -30,10 +47,37 @@ public:
 
   SDValue LowerOperation(SDValue Op, SelectionDAG &DAG) const override;
 
+  SDValue LowerFormalArguments(SDValue Chain, CallingConv::ID CallConv,
+                               bool isVarArg,
+                               const SmallVectorImpl<ISD::InputArg> &Ins,
+                               const SDLoc &DL, SelectionDAG &DAG,
+                               SmallVectorImpl<SDValue> &InVals) const override;
+
+  SDValue LowerCall(CallLoweringInfo &CLI,
+                    SmallVectorImpl<SDValue> &InVals) const override;
+
+  bool CanLowerReturn(CallingConv::ID CallConv, MachineFunction &MF,
+                      bool isVarArg,
+                      const SmallVectorImpl<ISD::OutputArg> &Outs,
+                      LLVMContext &Context) const override;
+
+  SDValue LowerReturn(SDValue Chain, CallingConv::ID CallConv, bool IsVarArg,
+                      const SmallVectorImpl<ISD::OutputArg> &Outs,
+                      const SmallVectorImpl<SDValue> &OutVals, const SDLoc &DL,
+                      SelectionDAG &DAG) const override;
+
   const XtensaSubtarget &getSubtarget() const { return Subtarget; }
 
 private:
   const XtensaSubtarget &Subtarget;
+
+  SDValue LowerImmediate(SDValue Op, SelectionDAG &DAG) const;
+
+  SDValue LowerConstantPool(ConstantPoolSDNode *CP, SelectionDAG &DAG) const;
+
+  SDValue getAddrPCRel(SDValue Op, SelectionDAG &DAG) const;
+
+  CCAssignFn *CCAssignFnForCall(CallingConv::ID CC, bool IsVarArg) const;
 };
 
 } // end namespace llvm
diff --git a/llvm/lib/Target/Xtensa/XtensaInstrInfo.cpp b/llvm/lib/Target/Xtensa/XtensaInstrInfo.cpp
index 41b794d64fdb..26d8727ce1d3 100644
--- a/llvm/lib/Target/Xtensa/XtensaInstrInfo.cpp
+++ b/llvm/lib/Target/Xtensa/XtensaInstrInfo.cpp
@@ -15,6 +15,7 @@
 #include "XtensaInstrInfo.h"
 #include "XtensaTargetMachine.h"
 #include "llvm/CodeGen/MachineConstantPool.h"
+#include "llvm/CodeGen/MachineFrameInfo.h"
 #include "llvm/CodeGen/MachineInstrBuilder.h"
 #include "llvm/CodeGen/MachineRegisterInfo.h"
 
@@ -23,5 +24,139 @@
 
 using namespace llvm;
 
+static const MachineInstrBuilder &
+addFrameReference(const MachineInstrBuilder &MIB, int FI) {
+  MachineInstr *MI = MIB;
+  MachineFunction &MF = *MI->getParent()->getParent();
+  MachineFrameInfo &MFFrame = MF.getFrameInfo();
+  const MCInstrDesc &MCID = MI->getDesc();
+  MachineMemOperand::Flags Flags = MachineMemOperand::MONone;
+  if (MCID.mayLoad())
+    Flags |= MachineMemOperand::MOLoad;
+  if (MCID.mayStore())
+    Flags |= MachineMemOperand::MOStore;
+  int64_t Offset = 0;
+  Align Alignment = MFFrame.getObjectAlign(FI);
+
+  MachineMemOperand *MMO =
+      MF.getMachineMemOperand(MachinePointerInfo::getFixedStack(MF, FI, Offset),
+                              Flags, MFFrame.getObjectSize(FI), Alignment);
+  return MIB.addFrameIndex(FI).addImm(Offset).addMemOperand(MMO);
+}
+
 XtensaInstrInfo::XtensaInstrInfo(const XtensaSubtarget &STI)
-    : XtensaGenInstrInfo(), RI(STI), STI(STI) {}
+    : XtensaGenInstrInfo(Xtensa::ADJCALLSTACKDOWN, Xtensa::ADJCALLSTACKUP),
+      RI(STI), STI(STI) {}
+
+/// Adjust SP by Amount bytes.
+void XtensaInstrInfo::adjustStackPtr(unsigned SP, int64_t Amount,
+                                     MachineBasicBlock &MBB,
+                                     MachineBasicBlock::iterator I) const {
+  DebugLoc DL = I != MBB.end() ? I->getDebugLoc() : DebugLoc();
+
+  if (Amount == 0)
+    return;
+
+  MachineRegisterInfo &RegInfo = MBB.getParent()->getRegInfo();
+  const TargetRegisterClass *RC = &Xtensa::ARRegClass;
+
+  // create virtual reg to store immediate
+  unsigned Reg = RegInfo.createVirtualRegister(RC);
+
+  if (isInt<8>(Amount)) { // addi sp, sp, amount
+    BuildMI(MBB, I, DL, get(Xtensa::ADDI), Reg).addReg(SP).addImm(Amount);
+  } else { // Expand immediate that doesn't fit in 8-bit.
+    unsigned Reg1;
+    loadImmediate(MBB, I, &Reg1, Amount);
+    BuildMI(MBB, I, DL, get(Xtensa::ADD), Reg)
+        .addReg(SP)
+        .addReg(Reg1, RegState::Kill);
+  }
+
+  BuildMI(MBB, I, DL, get(Xtensa::OR), SP)
+      .addReg(Reg, RegState::Kill)
+      .addReg(Reg, RegState::Kill);
+}
+
+void XtensaInstrInfo::copyPhysReg(MachineBasicBlock &MBB,
+                                  MachineBasicBlock::iterator MBBI,
+                                  const DebugLoc &DL, MCRegister DestReg,
+                                  MCRegister SrcReg, bool KillSrc) const {
+  // The MOV instruction is not present in core ISA,
+  // so use OR instruction.
+  if (Xtensa::ARRegClass.contains(DestReg, SrcReg))
+    BuildMI(MBB, MBBI, DL, get(Xtensa::OR), DestReg)
+        .addReg(SrcReg, getKillRegState(KillSrc))
+        .addReg(SrcReg, getKillRegState(KillSrc));
+  else
+    report_fatal_error("Impossible reg-to-reg copy");
+}
+
+void XtensaInstrInfo::storeRegToStackSlot(
+    MachineBasicBlock &MBB, MachineBasicBlock::iterator MBBI, Register SrcReg,
+    bool isKill, int FrameIdx, const TargetRegisterClass *RC,
+    const TargetRegisterInfo *TRI, Register VReg) const {
+  DebugLoc DL = MBBI != MBB.end() ? MBBI->getDebugLoc() : DebugLoc();
+  unsigned LoadOpcode, StoreOpcode;
+  getLoadStoreOpcodes(RC, LoadOpcode, StoreOpcode, FrameIdx);
+  MachineInstrBuilder MIB = BuildMI(MBB, MBBI, DL, get(StoreOpcode))
+                                .addReg(SrcReg, getKillRegState(isKill));
+  addFrameReference(MIB, FrameIdx);
+}
+
+void XtensaInstrInfo::loadRegFromStackSlot(MachineBasicBlock &MBB,
+                                           MachineBasicBlock::iterator MBBI,
+                                           Register DestReg, int FrameIdx,
+                                           const TargetRegisterClass *RC,
+                                           const TargetRegisterInfo *TRI,
+                                           Register VReg) const {
+  DebugLoc DL = MBBI != MBB.end() ? MBBI->getDebugLoc() : DebugLoc();
+  unsigned LoadOpcode, StoreOpcode;
+  getLoadStoreOpcodes(RC, LoadOpcode, StoreOpcode, FrameIdx);
+  addFrameReference(BuildMI(MBB, MBBI, DL, get(LoadOpcode), DestReg), FrameIdx);
+}
+
+void XtensaInstrInfo::getLoadStoreOpcodes(const TargetRegisterClass *RC,
+                                          unsigned &LoadOpcode,
+                                          unsigned &StoreOpcode,
+                                          int64_t offset) const {
+  assert((RC == &Xtensa::ARRegClass) &&
+         "Unsupported regclass to load or store");
+
+  LoadOpcode = Xtensa::L32I;
+  StoreOpcode = Xtensa::S32I;
+}
+
+void XtensaInstrInfo::loadImmediate(MachineBasicBlock &MBB,
+                                    MachineBasicBlock::iterator MBBI,
+                                    unsigned *Reg, int64_t Value) const {
+  DebugLoc DL = MBBI != MBB.end() ? MBBI->getDebugLoc() : DebugLoc();
+  MachineRegisterInfo &RegInfo = MBB.getParent()->getRegInfo();
+  const TargetRegisterClass *RC = &Xtensa::ARRegClass;
+
+  // create virtual reg to store immediate
+  *Reg = RegInfo.createVirtualRegister(RC);
+  if (Value >= -2048 && Value <= 2047) {
+    BuildMI(MBB, MBBI, DL, get(Xtensa::MOVI), *Reg).addImm(Value);
+  } else if (Value >= -32768 && Value <= 32767) {
+    int Low = Value & 0xFF;
+    int High = Value & ~0xFF;
+
+    BuildMI(MBB, MBBI, DL, get(Xtensa::MOVI), *Reg).addImm(Low);
+    BuildMI(MBB, MBBI, DL, get(Xtensa::ADDMI), *Reg).addReg(*Reg).addImm(High);
+  } else if (Value >= -4294967296LL && Value <= 4294967295LL) {
+    // 32 bit arbirary constant
+    MachineConstantPool *MCP = MBB.getParent()->getConstantPool();
+    uint64_t UVal = ((uint64_t)Value) & 0xFFFFFFFFLL;
+    const Constant *CVal = ConstantInt::get(
+        Type::getInt32Ty(MBB.getParent()->getFunction().getContext()), UVal,
+        false);
+    unsigned Idx = MCP->getConstantPoolIndex(CVal, Align(2U));
+    //	MCSymbol MSym
+    BuildMI(MBB, MBBI, DL, get(Xtensa::L32R), *Reg).addConstantPoolIndex(Idx);
+  } else {
+    // use L32R to let assembler load immediate best
+    // TODO replace to L32R
+    report_fatal_error("Unsupported load immediate value");
+  }
+}
diff --git a/llvm/lib/Target/Xtensa/XtensaInstrInfo.h b/llvm/lib/Target/Xtensa/XtensaInstrInfo.h
index 8c73c9bd7940..1acd314e2720 100644
--- a/llvm/lib/Target/Xtensa/XtensaInstrInfo.h
+++ b/llvm/lib/Target/Xtensa/XtensaInstrInfo.h
@@ -35,9 +35,38 @@ class XtensaInstrInfo : public XtensaGenInstrInfo {
 public:
   XtensaInstrInfo(const XtensaSubtarget &STI);
 
+  void adjustStackPtr(unsigned SP, int64_t Amount, MachineBasicBlock &MBB,
+                      MachineBasicBlock::iterator I) const;
+
   // Return the XtensaRegisterInfo, which this class owns.
   const XtensaRegisterInfo &getRegisterInfo() const { return RI; }
 
+  void copyPhysReg(MachineBasicBlock &MBB, MachineBasicBlock::iterator MBBI,
+                   const DebugLoc &DL, MCRegister DestReg, MCRegister SrcReg,
+                   bool KillSrc) const override;
+
+  void storeRegToStackSlot(MachineBasicBlock &MBB,
+                           MachineBasicBlock::iterator MBBI, Register SrcReg,
+                           bool isKill, int FrameIndex,
+                           const TargetRegisterClass *RC,
+                           const TargetRegisterInfo *TRI,
+                           Register VReg) const override;
+
+  void loadRegFromStackSlot(MachineBasicBlock &MBB,
+                            MachineBasicBlock::iterator MBBI, Register DestReg,
+                            int FrameIdx, const TargetRegisterClass *RC,
+                            const TargetRegisterInfo *TRI,
+                            Register VReg) const override;
+
+  // Get the load and store opcodes for a given register class and offset.
+  void getLoadStoreOpcodes(const TargetRegisterClass *RC, unsigned &LoadOpcode,
+                           unsigned &StoreOpcode, int64_t offset) const;
+
+  // Emit code before MBBI in MI to move immediate value Value into
+  // physical register Reg.
+  void loadImmediate(MachineBasicBlock &MBB, MachineBasicBlock::iterator MBBI,
+                     unsigned *Reg, int64_t Value) const;
+
   const XtensaSubtarget &getSubtarget() const { return STI; }
 };
 } // end namespace llvm
diff --git a/llvm/lib/Target/Xtensa/XtensaInstrInfo.td b/llvm/lib/Target/Xtensa/XtensaInstrInfo.td
index 268a9943d8c1..6e9e75257ccf 100644
--- a/llvm/lib/Target/Xtensa/XtensaInstrInfo.td
+++ b/llvm/lib/Target/Xtensa/XtensaInstrInfo.td
@@ -14,6 +14,7 @@
 
 include "XtensaInstrFormats.td"
 include "XtensaOperands.td"
+include "XtensaOperators.td"
 
 //===----------------------------------------------------------------------===//
 // Arithmetic & Logical instructions
@@ -238,6 +239,34 @@ def L32R : RI16_Inst<0x01, (outs AR:$t), (ins L32Rtarget:$label),
   let imm16 = label;
 }
 
+// pcrel addr loading using L32R
+def : Pat<(Xtensa_pcrel_wrapper tconstpool : $in), (L32R tconstpool : $in)>;
+
+// FrameIndexes are legalized when they are operands from load/store
+// instructions. The same not happens for stack address copies, so an
+// add op with mem ComplexPattern is used and the stack address copy
+// can be matched.
+// Setting of attribute mayLoad is trick to process instruction operands
+// in function XtensaRegisterInfo::eliminateFI
+
+let isCodeGenOnly = 1, mayLoad = 1 in {
+
+  def LEA_ADD : RRI8_Inst<0x02, (outs AR:$t), (ins mem32:$addr),
+       "addi\t$t, $addr",
+       [(set AR:$t, addr_ish4:$addr)]> {
+    bits<12> addr;
+
+    let r = 0x0C;
+    let imm8{7-0} = addr{11-4};
+    let s{3-0} = addr{3-0};
+  }
+}
+
+//extending loads
+def : Pat<(i32 (extloadi1  addr_ish1:$addr)), (L8UI addr_ish1:$addr)>;
+def : Pat<(i32 (extloadi8  addr_ish1:$addr)), (L8UI addr_ish1:$addr)>;
+def : Pat<(i32 (extloadi16 addr_ish2:$addr)), (L16UI addr_ish2:$addr)>;
+
 //===----------------------------------------------------------------------===//
 // Conditional branch instructions
 //===----------------------------------------------------------------------===//
@@ -426,7 +455,7 @@ let isReturn = 1, isTerminator = 1,
     isBarrier = 1, Uses = [A0] in {
 
   def RET : CALLX_Inst<0x00, 0x00, 0x00, (outs), (ins),
-                      "ret", []> {
+                      "ret", [(Xtensa_ret)]> {
     let m = 0x2;
     let n = 0x0;
     let s = 0;
@@ -434,6 +463,14 @@ let isReturn = 1, isTerminator = 1,
   }
 }
 
+// Call patterns
+def : Pat<(Xtensa_call (i32 tglobaladdr:$dst)),
+          (CALL0 tglobaladdr:$dst)>;
+def : Pat<(Xtensa_call (i32 texternalsym:$dst)),
+          (CALL0 texternalsym:$dst)>;
+def : Pat<(Xtensa_call AR:$dst),
+          (CALLX0 AR:$dst)>;
+
 //===----------------------------------------------------------------------===//
 // Mem barrier instructions
 //===----------------------------------------------------------------------===//
@@ -506,3 +543,19 @@ def XSR : RSR_Inst<0x00, 0x01, 0x06, (outs AR:$ard, SR:$srd), (ins AR:$t, SR:$sr
                   "xsr\t$t, $sr", []> {
   let Constraints = "$ard = $t, $srd = $sr";
 }
+
+//===----------------------------------------------------------------------===//
+// Stack allocation
+//===----------------------------------------------------------------------===//
+
+// ADJCALLSTACKDOWN/UP implicitly use/def SP because they may be expanded into
+// a stack adjustment and the codegen must know that they may modify the stack
+// pointer before prolog-epilog rewriting occurs.
+let Defs = [SP], Uses = [SP] in {
+  def ADJCALLSTACKDOWN : Pseudo<(outs), (ins i32imm:$amt1, i32imm:$amt2),
+                               "#ADJCALLSTACKDOWN",
+                               [(Xtensa_callseq_start timm:$amt1, timm:$amt2)]>;
+  def ADJCALLSTACKUP   : Pseudo<(outs), (ins i32imm:$amt1, i32imm:$amt2),
+                               "#ADJCALLSTACKUP",
+                               [(Xtensa_callseq_end timm:$amt1, timm:$amt2)]>;
+}
diff --git a/llvm/lib/Target/Xtensa/XtensaOperands.td b/llvm/lib/Target/Xtensa/XtensaOperands.td
index 7a1a2e86e8c2..f41081f9bf2f 100644
--- a/llvm/lib/Target/Xtensa/XtensaOperands.td
+++ b/llvm/lib/Target/Xtensa/XtensaOperands.td
@@ -195,7 +195,7 @@ def jumptarget : Operand<OtherVT> {
   let ParserMatchClass = XtensaPCRelTargetAsmOperand;
 }
 
-def L32Rtarget : Operand<OtherVT> {
+def L32Rtarget : Operand<i32> {
   let PrintMethod = "printL32RTarget";
   let EncoderMethod = "getL32RTargetEncoding";
   let DecoderMethod = "decodeL32ROperand";
diff --git a/llvm/lib/Target/Xtensa/XtensaOperators.td b/llvm/lib/Target/Xtensa/XtensaOperators.td
new file mode 100644
index 000000000000..cd4d831c85b5
--- /dev/null
+++ b/llvm/lib/Target/Xtensa/XtensaOperators.td
@@ -0,0 +1,36 @@
+//===- XtensaOperators.td - Xtensa-specific operators ---------*- tblgen-*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+//===----------------------------------------------------------------------===//
+// Type profiles
+//===----------------------------------------------------------------------===//
+def SDT_XtensaCallSeqStart        : SDCallSeqStart<[SDTCisVT<0, i32>, SDTCisVT<1, i32>]>;
+def SDT_XtensaCallSeqEnd          : SDCallSeqEnd<[SDTCisVT<0, i32>, SDTCisVT<1, i32>]>;
+def SDT_XtensaCall                : SDTypeProfile<0, -1, [SDTCisPtrTy<0>]>;
+
+def SDT_XtensaWrapPtr             : SDTypeProfile<1, 1,
+                                                 [SDTCisSameAs<0, 1>,
+                                                  SDTCisPtrTy<0>]>;
+
+//===----------------------------------------------------------------------===//
+// Node definitions
+//===----------------------------------------------------------------------===//
+def Xtensa_call: SDNode<"XtensaISD::CALL", SDT_XtensaCall,
+                       [SDNPHasChain, SDNPOutGlue, SDNPOptInGlue, SDNPVariadic]>;
+
+def Xtensa_ret: SDNode<"XtensaISD::RET", SDTNone,
+                       [SDNPHasChain, SDNPOptInGlue, SDNPVariadic]>;
+
+def Xtensa_pcrel_wrapper: SDNode<"XtensaISD::PCREL_WRAPPER", SDT_XtensaWrapPtr, []>;
+
+def Xtensa_callseq_start: SDNode<"ISD::CALLSEQ_START", SDT_XtensaCallSeqStart,
+                                [SDNPHasChain, SDNPSideEffect, SDNPOutGlue]>;
+
+def Xtensa_callseq_end  : SDNode<"ISD::CALLSEQ_END",   SDT_XtensaCallSeqEnd,
+                                [SDNPHasChain, SDNPSideEffect, SDNPOptInGlue,
+                                 SDNPOutGlue]>;
diff --git a/llvm/lib/Target/Xtensa/XtensaRegisterInfo.cpp b/llvm/lib/Target/Xtensa/XtensaRegisterInfo.cpp
index f749cc51f96a..bced2d4ad009 100644
--- a/llvm/lib/Target/Xtensa/XtensaRegisterInfo.cpp
+++ b/llvm/lib/Target/Xtensa/XtensaRegisterInfo.cpp
@@ -13,6 +13,9 @@
 #include "XtensaRegisterInfo.h"
 #include "XtensaInstrInfo.h"
 #include "XtensaSubtarget.h"
+#include "XtensaUtils.h"
+#include "llvm/CodeGen/MachineFrameInfo.h"
+#include "llvm/CodeGen/MachineFunction.h"
 #include "llvm/CodeGen/MachineInstrBuilder.h"
 #include "llvm/CodeGen/MachineRegisterInfo.h"
 #include "llvm/Support/Debug.h"
@@ -31,15 +34,13 @@ XtensaRegisterInfo::XtensaRegisterInfo(const XtensaSubtarget &STI)
 
 const uint16_t *
 XtensaRegisterInfo::getCalleeSavedRegs(const MachineFunction *MF) const {
-  // Calling convention is not implemented yet
-  return nullptr;
+  return CSR_Xtensa_SaveList;
 }
 
 const uint32_t *
 XtensaRegisterInfo::getCallPreservedMask(const MachineFunction &MF,
                                          CallingConv::ID) const {
-  // Calling convention is not implemented yet
-  return nullptr;
+  return CSR_Xtensa_RegMask;
 }
 
 BitVector XtensaRegisterInfo::getReservedRegs(const MachineFunction &MF) const {
@@ -60,7 +61,70 @@ BitVector XtensaRegisterInfo::getReservedRegs(const MachineFunction &MF) const {
 bool XtensaRegisterInfo::eliminateFrameIndex(MachineBasicBlock::iterator II,
                                              int SPAdj, unsigned FIOperandNum,
                                              RegScavenger *RS) const {
-  report_fatal_error("Eliminate frame index not supported yet");
+  MachineInstr &MI = *II;
+  MachineFunction &MF = *MI.getParent()->getParent();
+  int FrameIndex = MI.getOperand(FIOperandNum).getIndex();
+  uint64_t StackSize = MF.getFrameInfo().getStackSize();
+  int64_t SPOffset = MF.getFrameInfo().getObjectOffset(FrameIndex);
+  MachineFrameInfo &MFI = MF.getFrameInfo();
+  const std::vector<CalleeSavedInfo> &CSI = MFI.getCalleeSavedInfo();
+  int MinCSFI = 0;
+  int MaxCSFI = -1;
+
+  if (CSI.size()) {
+    MinCSFI = CSI[0].getFrameIdx();
+    MaxCSFI = CSI[CSI.size() - 1].getFrameIdx();
+  }
+  // The following stack frame objects are always referenced relative to $sp:
+  //  1. Outgoing arguments.
+  //  2. Pointer to dynamically allocated stack space.
+  //  3. Locations for callee-saved registers.
+  //  4. Locations for eh data registers.
+  // Everything else is referenced relative to whatever register
+  // getFrameRegister() returns.
+  unsigned FrameReg;
+  if ((FrameIndex >= MinCSFI && FrameIndex <= MaxCSFI))
+    FrameReg = Xtensa::SP;
+  else
+    FrameReg = getFrameRegister(MF);
+
+  // Calculate final offset.
+  // - There is no need to change the offset if the frame object is one of the
+  //   following: an outgoing argument, pointer to a dynamically allocated
+  //   stack space or a $gp restore location,
+  // - If the frame object is any of the following, its offset must be adjusted
+  //   by adding the size of the stack:
+  //   incoming argument, callee-saved register location or local variable.
+  bool IsKill = false;
+  int64_t Offset =
+      SPOffset + (int64_t)StackSize + MI.getOperand(FIOperandNum + 1).getImm();
+
+  bool Valid = isValidAddrOffset(MI, Offset);
+
+  // If MI is not a debug value, make sure Offset fits in the 16-bit immediate
+  // field.
+  if (!MI.isDebugValue() && !Valid) {
+    MachineBasicBlock &MBB = *MI.getParent();
+    DebugLoc DL = II->getDebugLoc();
+    unsigned ADD = Xtensa::ADD;
+    unsigned Reg;
+    const XtensaInstrInfo &TII = *static_cast<const XtensaInstrInfo *>(
+        MBB.getParent()->getSubtarget().getInstrInfo());
+
+    TII.loadImmediate(MBB, II, &Reg, Offset);
+    BuildMI(MBB, II, DL, TII.get(ADD), Reg)
+        .addReg(FrameReg)
+        .addReg(Reg, RegState::Kill);
+
+    FrameReg = Reg;
+    Offset = 0;
+    IsKill = true;
+  }
+
+  MI.getOperand(FIOperandNum).ChangeToRegister(FrameReg, false, false, IsKill);
+  MI.getOperand(FIOperandNum + 1).ChangeToImmediate(Offset);
+
+  return false;
 }
 
 Register XtensaRegisterInfo::getFrameRegister(const MachineFunction &MF) const {
diff --git a/llvm/lib/Target/Xtensa/XtensaUtils.cpp b/llvm/lib/Target/Xtensa/XtensaUtils.cpp
new file mode 100644
index 000000000000..98e424f6ea44
--- /dev/null
+++ b/llvm/lib/Target/Xtensa/XtensaUtils.cpp
@@ -0,0 +1,59 @@
+//===--- XtensaUtils.cpp ---- Xtensa Utility Functions ----------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file contains miscellaneous utility functions.
+//
+//===----------------------------------------------------------------------===//
+
+#include "XtensaUtils.h"
+
+namespace llvm {
+
+bool isValidAddrOffset(int Scale, int64_t OffsetVal) {
+  bool Valid = false;
+
+  switch (Scale) {
+  case 1:
+    Valid = (OffsetVal >= 0 && OffsetVal <= 255);
+    break;
+  case 2:
+    Valid = (OffsetVal >= 0 && OffsetVal <= 510) && ((OffsetVal & 0x1) == 0);
+    break;
+  case 4:
+    Valid = (OffsetVal >= 0 && OffsetVal <= 1020) && ((OffsetVal & 0x3) == 0);
+    break;
+  default:
+    break;
+  }
+  return Valid;
+}
+
+bool isValidAddrOffset(MachineInstr &MI, int64_t Offset) {
+  int Scale = 0;
+
+  switch (MI.getOpcode()) {
+  case Xtensa::L8UI:
+  case Xtensa::S8I:
+    Scale = 1;
+    break;
+  case Xtensa::L16SI:
+  case Xtensa::L16UI:
+  case Xtensa::S16I:
+    Scale = 2;
+    break;
+  case Xtensa::LEA_ADD:
+    return (Offset >= -128 && Offset <= 127);
+  default:
+    // assume that MI is 32-bit load/store operation
+    Scale = 4;
+    break;
+  }
+  return isValidAddrOffset(Scale, Offset);
+}
+
+} // namespace llvm
diff --git a/llvm/lib/Target/Xtensa/XtensaUtils.h b/llvm/lib/Target/Xtensa/XtensaUtils.h
new file mode 100644
index 000000000000..2b0ac37a6971
--- /dev/null
+++ b/llvm/lib/Target/Xtensa/XtensaUtils.h
@@ -0,0 +1,27 @@
+//===--- XtensaUtils.h ---- Xtensa Utility Functions ------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file contains miscellaneous utility functions.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_LIB_TARGET_XTENSA_XTENSAUTILS_H
+#define LLVM_LIB_TARGET_XTENSA_XTENSAUTILS_H
+
+#include "XtensaInstrInfo.h"
+#include "llvm/CodeGen/MachineInstr.h"
+
+namespace llvm {
+// Check address offset for load/store instructions.
+// The offset should be multiple of scale.
+bool isValidAddrOffset(int Scale, int64_t OffsetVal);
+
+// Check address offset for load/store instructions.
+bool isValidAddrOffset(MachineInstr &MI, int64_t Offset);
+} // namespace llvm
+#endif // LLVM_LIB_TARGET_XTENSA_XTENSAUTILS_H
diff --git a/llvm/lib/TargetParser/CMakeLists.txt b/llvm/lib/TargetParser/CMakeLists.txt
index da1e352b0373..4b5d582d57a4 100644
--- a/llvm/lib/TargetParser/CMakeLists.txt
+++ b/llvm/lib/TargetParser/CMakeLists.txt
@@ -20,6 +20,7 @@ add_llvm_component_library(LLVMTargetParser
   CSKYTargetParser.cpp
   Host.cpp
   LoongArchTargetParser.cpp
+  RISCVISAInfo.cpp
   RISCVTargetParser.cpp
   SubtargetFeature.cpp
   TargetParser.cpp
@@ -37,5 +38,7 @@ add_llvm_component_library(LLVMTargetParser
   Support
 
   DEPENDS
+  ARMTargetParserTableGen
+  AArch64TargetParserTableGen
   RISCVTargetParserTableGen
   )
diff --git a/llvm/lib/TargetParser/Host.cpp b/llvm/lib/TargetParser/Host.cpp
index d2c9bae97364..0a93b06f40c2 100644
--- a/llvm/lib/TargetParser/Host.cpp
+++ b/llvm/lib/TargetParser/Host.cpp
@@ -245,8 +245,11 @@ StringRef sys::detail::getHostCPUNameForARM(StringRef ProcCpuinfoContent) {
         .Case("0xd4a", "neoverse-e1")
         .Case("0xd0c", "neoverse-n1")
         .Case("0xd49", "neoverse-n2")
+        .Case("0xd8e", "neoverse-n3")
         .Case("0xd40", "neoverse-v1")
         .Case("0xd4f", "neoverse-v2")
+        .Case("0xd84", "neoverse-v3")
+        .Case("0xd83", "neoverse-v3ae")
         .Default("generic");
   }
 
diff --git a/llvm/lib/Support/RISCVISAInfo.cpp b/llvm/lib/TargetParser/RISCVISAInfo.cpp
index fa967403ea44..20182fb06037 100644
--- a/llvm/lib/Support/RISCVISAInfo.cpp
+++ b/llvm/lib/TargetParser/RISCVISAInfo.cpp
@@ -1,4 +1,4 @@
-//===-- RISCVISAInfo.cpp - RISC-V Arch String Parser ------------*- C++ -*-===//
+//===-- RISCVISAInfo.cpp - RISC-V Arch String Parser ----------------------===//
 //
 // Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
 // See https://llvm.org/LICENSE.txt for license information.
@@ -6,7 +6,7 @@
 //
 //===----------------------------------------------------------------------===//
 
-#include "llvm/Support/RISCVISAInfo.h"
+#include "llvm/TargetParser/RISCVISAInfo.h"
 #include "llvm/ADT/MapVector.h"
 #include "llvm/ADT/STLExtras.h"
 #include "llvm/ADT/SetVector.h"
@@ -29,7 +29,7 @@ namespace {
 struct RISCVSupportedExtension {
   const char *Name;
   /// Supported version.
-  RISCVISAInfo::ExtensionVersion Version;
+  RISCVISAUtils::ExtensionVersion Version;
 
   bool operator<(const RISCVSupportedExtension &RHS) const {
     return StringRef(Name) < StringRef(RHS.Name);
@@ -43,246 +43,15 @@ struct RISCVProfile {
 
 } // end anonymous namespace
 
-static constexpr StringLiteral AllStdExts = "mafdqlcbkjtpvnh";
-
 static const char *RISCVGImplications[] = {
   "i", "m", "a", "f", "d", "zicsr", "zifencei"
 };
 
-// NOTE: This table should be sorted alphabetically by extension name.
-static const RISCVSupportedExtension SupportedExtensions[] = {
-    {"a", {2, 1}},
-    {"c", {2, 0}},
-    {"d", {2, 2}},
-    {"e", {2, 0}},
-    {"f", {2, 2}},
-    {"h", {1, 0}},
-    {"i", {2, 1}},
-    {"m", {2, 0}},
-
-    {"shcounterenw", {1, 0}},
-    {"shgatpa", {1, 0}},
-    {"shtvala", {1, 0}},
-    {"shvsatpa", {1, 0}},
-    {"shvstvala", {1, 0}},
-    {"shvstvecd", {1, 0}},
-    {"smaia", {1, 0}},
-    {"smepmp", {1, 0}},
-    {"ssaia", {1, 0}},
-    {"ssccptr", {1, 0}},
-    {"sscofpmf", {1, 0}},
-    {"sscounterenw", {1, 0}},
-    {"ssstateen", {1, 0}},
-    {"ssstrict", {1, 0}},
-    {"sstc", {1, 0}},
-    {"sstvala", {1, 0}},
-    {"sstvecd", {1, 0}},
-    {"ssu64xl", {1, 0}},
-    {"svade", {1, 0}},
-    {"svadu", {1, 0}},
-    {"svbare", {1, 0}},
-    {"svinval", {1, 0}},
-    {"svnapot", {1, 0}},
-    {"svpbmt", {1, 0}},
-
-    {"v", {1, 0}},
-
-    // vendor-defined ('X') extensions
-    {"xcvalu", {1, 0}},
-    {"xcvbi", {1, 0}},
-    {"xcvbitmanip", {1, 0}},
-    {"xcvelw", {1, 0}},
-    {"xcvmac", {1, 0}},
-    {"xcvmem", {1, 0}},
-    {"xcvsimd", {1, 0}},
-    {"xsfcease", {1, 0}},
-    {"xsfvcp", {1, 0}},
-    {"xsfvfnrclipxfqf", {1, 0}},
-    {"xsfvfwmaccqqq", {1, 0}},
-    {"xsfvqmaccdod", {1, 0}},
-    {"xsfvqmaccqoq", {1, 0}},
-    {"xsifivecdiscarddlone", {1, 0}},
-    {"xsifivecflushdlone", {1, 0}},
-    {"xtheadba", {1, 0}},
-    {"xtheadbb", {1, 0}},
-    {"xtheadbs", {1, 0}},
-    {"xtheadcmo", {1, 0}},
-    {"xtheadcondmov", {1, 0}},
-    {"xtheadfmemidx", {1, 0}},
-    {"xtheadmac", {1, 0}},
-    {"xtheadmemidx", {1, 0}},
-    {"xtheadmempair", {1, 0}},
-    {"xtheadsync", {1, 0}},
-    {"xtheadvdot", {1, 0}},
-    {"xventanacondops", {1, 0}},
-
-    {"za128rs", {1, 0}},
-    {"za64rs", {1, 0}},
-    {"zacas", {1, 0}},
-    {"zama16b", {1, 0}},
-    {"zawrs", {1, 0}},
-
-    {"zba", {1, 0}},
-    {"zbb", {1, 0}},
-    {"zbc", {1, 0}},
-    {"zbkb", {1, 0}},
-    {"zbkc", {1, 0}},
-    {"zbkx", {1, 0}},
-    {"zbs", {1, 0}},
-
-    {"zca", {1, 0}},
-    {"zcb", {1, 0}},
-    {"zcd", {1, 0}},
-    {"zce", {1, 0}},
-    {"zcf", {1, 0}},
-    {"zcmop", {1, 0}},
-    {"zcmp", {1, 0}},
-    {"zcmt", {1, 0}},
-
-    {"zdinx", {1, 0}},
-
-    {"zfa", {1, 0}},
-    {"zfh", {1, 0}},
-    {"zfhmin", {1, 0}},
-    {"zfinx", {1, 0}},
-
-    {"zhinx", {1, 0}},
-    {"zhinxmin", {1, 0}},
-
-    {"zic64b", {1, 0}},
-    {"zicbom", {1, 0}},
-    {"zicbop", {1, 0}},
-    {"zicboz", {1, 0}},
-    {"ziccamoa", {1, 0}},
-    {"ziccif", {1, 0}},
-    {"zicclsm", {1, 0}},
-    {"ziccrse", {1, 0}},
-    {"zicntr", {2, 0}},
-    {"zicond", {1, 0}},
-    {"zicsr", {2, 0}},
-    {"zifencei", {2, 0}},
-    {"zihintntl", {1, 0}},
-    {"zihintpause", {2, 0}},
-    {"zihpm", {2, 0}},
-    {"zimop", {1, 0}},
-
-    {"zk", {1, 0}},
-    {"zkn", {1, 0}},
-    {"zknd", {1, 0}},
-    {"zkne", {1, 0}},
-    {"zknh", {1, 0}},
-    {"zkr", {1, 0}},
-    {"zks", {1, 0}},
-    {"zksed", {1, 0}},
-    {"zksh", {1, 0}},
-    {"zkt", {1, 0}},
-
-    {"zmmul", {1, 0}},
-
-    {"zvbb", {1, 0}},
-    {"zvbc", {1, 0}},
-
-    {"zve32f", {1, 0}},
-    {"zve32x", {1, 0}},
-    {"zve64d", {1, 0}},
-    {"zve64f", {1, 0}},
-    {"zve64x", {1, 0}},
-
-    {"zvfh", {1, 0}},
-    {"zvfhmin", {1, 0}},
-
-    // vector crypto
-    {"zvkb", {1, 0}},
-    {"zvkg", {1, 0}},
-    {"zvkn", {1, 0}},
-    {"zvknc", {1, 0}},
-    {"zvkned", {1, 0}},
-    {"zvkng", {1, 0}},
-    {"zvknha", {1, 0}},
-    {"zvknhb", {1, 0}},
-    {"zvks", {1, 0}},
-    {"zvksc", {1, 0}},
-    {"zvksed", {1, 0}},
-    {"zvksg", {1, 0}},
-    {"zvksh", {1, 0}},
-    {"zvkt", {1, 0}},
-
-    {"zvl1024b", {1, 0}},
-    {"zvl128b", {1, 0}},
-    {"zvl16384b", {1, 0}},
-    {"zvl2048b", {1, 0}},
-    {"zvl256b", {1, 0}},
-    {"zvl32768b", {1, 0}},
-    {"zvl32b", {1, 0}},
-    {"zvl4096b", {1, 0}},
-    {"zvl512b", {1, 0}},
-    {"zvl64b", {1, 0}},
-    {"zvl65536b", {1, 0}},
-    {"zvl8192b", {1, 0}},
-};
-
-// NOTE: This table should be sorted alphabetically by extension name.
-// clang-format off
-static const RISCVSupportedExtension SupportedExperimentalExtensions[] = {
-    {"smmpm", {0, 8}},
-    {"smnpm", {0, 8}},
-    {"ssnpm", {0, 8}},
-    {"sspm", {0, 8}},
-    {"ssqosid", {1, 0}},
-    {"supm", {0, 8}},
-
-    {"zaamo", {0, 2}},
-    {"zabha", {1, 0}},
-    {"zalasr", {0, 1}},
-    {"zalrsc", {0, 2}},
-
-    {"zfbfmin", {1, 0}},
-
-    {"zicfilp", {0, 4}},
-    {"zicfiss", {0, 4}},
+#define GET_SUPPORTED_EXTENSIONS
+#include "llvm/TargetParser/RISCVTargetParserDef.inc"
 
-    {"ztso", {0, 1}},
-
-    {"zvfbfmin", {1, 0}},
-    {"zvfbfwma", {1, 0}},
-};
-// clang-format on
-
-static constexpr RISCVProfile SupportedProfiles[] = {
-    {"rvi20u32", "rv32i"},
-    {"rvi20u64", "rv64i"},
-    {"rva20u64", "rv64imafdc_ziccamoa_ziccif_zicclsm_ziccrse_zicntr_za128rs"},
-    {"rva20s64", "rv64imafdc_ziccamoa_ziccif_zicclsm_ziccrse_zicntr_zifencei_"
-                 "za128rs_ssccptr_sstvala_sstvecd_svade_svbare"},
-    {"rva22u64",
-     "rv64imafdc_zic64b_zicbom_zicbop_zicboz_ziccamoa_ziccif_zicclsm_ziccrse_"
-     "zicntr_zihintpause_zihpm_za64rs_zfhmin_zba_zbb_zbs_zkt"},
-    {"rva22s64",
-     "rv64imafdc_zic64b_zicbom_zicbop_zicboz_ziccamoa_ziccif_zicclsm_ziccrse_"
-     "zicntr_zifencei_zihintpause_zihpm_za64rs_zfhmin_zba_zbb_zbs_zkt_ssccptr_"
-     "sscounterenw_sstvala_sstvecd_svade_svbare_svinval_svpbmt"},
-    {"rva23u64",
-     "rv64imafdcv_zic64b_zicbom_zicbop_zicboz_ziccamoa_ziccif_zicclsm_ziccrse_"
-     "zicntr_zicond_zihintntl_zihintpause_zihpm_zimop_za64rs_zawrs_zfa_zfhmin_"
-     "zcb_zcmop_zba_zbb_zbs_zkt_zvbb_zvfhmin_zvkt"},
-    {"rva23s64",
-     "rv64imafdcvh_zic64b_zicbom_zicbop_zicboz_ziccamoa_ziccif_zicclsm_ziccrse_"
-     "zicntr_zicond_zifencei_zihintntl_zihintpause_zihpm_zimop_za64rs_zawrs_"
-     "zfa_zfhmin_zcb_zcmop_zba_zbb_zbs_zkt_zvbb_zvfhmin_zvkt_shcounterenw_"
-     "shgatpa_shtvala_shvsatpa_shvstvala_shvstvecd_ssccptr_sscofpmf_"
-     "sscounterenw_ssnpm0p8_ssstateen_sstc_sstvala_sstvecd_ssu64xl_svade_"
-     "svbare_svinval_svnapot_svpbmt"},
-    {"rvb23u64", "rv64imafdc_zic64b_zicbom_zicbop_zicboz_ziccamoa_ziccif_"
-                 "zicclsm_ziccrse_zicntr_zicond_zihintntl_zihintpause_zihpm_"
-                 "zimop_za64rs_zawrs_zfa_zcb_zcmop_zba_zbb_zbs_zkt"},
-    {"rvb23s64",
-     "rv64imafdc_zic64b_zicbom_zicbop_zicboz_ziccamoa_ziccif_zicclsm_ziccrse_"
-     "zicntr_zicond_zifencei_zihintntl_zihintpause_zihpm_zimop_za64rs_zawrs_"
-     "zfa_zcb_zcmop_zba_zbb_zbs_zkt_ssccptr_sscofpmf_sscounterenw_sstc_sstvala_"
-     "sstvecd_ssu64xl_svade_svbare_svinval_svnapot_svpbmt"},
-    {"rvm23u32", "rv32im_zicbop_zicond_zicsr_zihintntl_zihintpause_zimop_zca_"
-                 "zcb_zce_zcmop_zcmp_zcmt_zba_zbb_zbs"},
-};
+#define GET_SUPPORTED_PROFILES
+#include "llvm/TargetParser/RISCVTargetParserDef.inc"
 
 static void verifyTables() {
 #ifndef NDEBUG
@@ -310,7 +79,7 @@ void llvm::riscvExtensionsHelp(StringMap<StringRef> DescMap) {
   outs() << "All available -march extensions for RISC-V\n\n";
   PrintExtension("Name", "Version", (DescMap.empty() ? "" : "Description"));
 
-  RISCVISAInfo::OrderedExtensionMap ExtMap;
+  RISCVISAUtils::OrderedExtensionMap ExtMap;
   for (const auto &E : SupportedExtensions)
     ExtMap[E.Name] = {E.Version.Major, E.Version.Minor};
   for (const auto &E : ExtMap) {
@@ -370,7 +139,7 @@ struct LessExtName {
 };
 } // namespace
 
-static std::optional<RISCVISAInfo::ExtensionVersion>
+static std::optional<RISCVISAUtils::ExtensionVersion>
 findDefaultVersion(StringRef ExtName) {
   // Find default version of an extension.
   // TODO: We might set default version based on profile or ISA spec.
@@ -387,7 +156,7 @@ findDefaultVersion(StringRef ExtName) {
 }
 
 void RISCVISAInfo::addExtension(StringRef ExtName,
-                                RISCVISAInfo::ExtensionVersion Version) {
+                                RISCVISAUtils::ExtensionVersion Version) {
   Exts[ExtName.str()] = Version;
 }
 
@@ -411,7 +180,7 @@ static StringRef getExtensionType(StringRef Ext) {
   return StringRef();
 }
 
-static std::optional<RISCVISAInfo::ExtensionVersion>
+static std::optional<RISCVISAUtils::ExtensionVersion>
 isExperimentalExtension(StringRef Ext) {
   auto I =
       llvm::lower_bound(SupportedExperimentalExtensions, Ext, LessExtName());
@@ -468,78 +237,6 @@ bool RISCVISAInfo::hasExtension(StringRef Ext) const {
   return Exts.count(Ext.str()) != 0;
 }
 
-// We rank extensions in the following order:
-// -Single letter extensions in canonical order.
-// -Unknown single letter extensions in alphabetical order.
-// -Multi-letter extensions starting with 'z' sorted by canonical order of
-//  the second letter then sorted alphabetically.
-// -Multi-letter extensions starting with 's' in alphabetical order.
-// -(TODO) Multi-letter extensions starting with 'zxm' in alphabetical order.
-// -X extensions in alphabetical order.
-// These flags are used to indicate the category. The first 6 bits store the
-// single letter extension rank for single letter and multi-letter extensions
-// starting with 'z'.
-enum RankFlags {
-  RF_Z_EXTENSION = 1 << 6,
-  RF_S_EXTENSION = 1 << 7,
-  RF_X_EXTENSION = 1 << 8,
-};
-
-// Get the rank for single-letter extension, lower value meaning higher
-// priority.
-static unsigned singleLetterExtensionRank(char Ext) {
-  assert(Ext >= 'a' && Ext <= 'z');
-  switch (Ext) {
-  case 'i':
-    return 0;
-  case 'e':
-    return 1;
-  }
-
-  size_t Pos = AllStdExts.find(Ext);
-  if (Pos != StringRef::npos)
-    return Pos + 2; // Skip 'e' and 'i' from above.
-
-  // If we got an unknown extension letter, then give it an alphabetical
-  // order, but after all known standard extensions.
-  return 2 + AllStdExts.size() + (Ext - 'a');
-}
-
-// Get the rank for multi-letter extension, lower value meaning higher
-// priority/order in canonical order.
-static unsigned getExtensionRank(const std::string &ExtName) {
-  assert(ExtName.size() >= 1);
-  switch (ExtName[0]) {
-  case 's':
-    return RF_S_EXTENSION;
-  case 'z':
-    assert(ExtName.size() >= 2);
-    // `z` extension must be sorted by canonical order of second letter.
-    // e.g. zmx has higher rank than zax.
-    return RF_Z_EXTENSION | singleLetterExtensionRank(ExtName[1]);
-  case 'x':
-    return RF_X_EXTENSION;
-  default:
-    assert(ExtName.size() == 1);
-    return singleLetterExtensionRank(ExtName[0]);
-  }
-}
-
-// Compare function for extension.
-// Only compare the extension name, ignore version comparison.
-bool RISCVISAInfo::compareExtension(const std::string &LHS,
-                                    const std::string &RHS) {
-  unsigned LHSRank = getExtensionRank(LHS);
-  unsigned RHSRank = getExtensionRank(RHS);
-
-  // If the ranks differ, pick the lower rank.
-  if (LHSRank != RHSRank)
-    return LHSRank < RHSRank;
-
-  // If the rank is same, it must be sorted by lexicographic order.
-  return LHS < RHS;
-}
-
 std::vector<std::string> RISCVISAInfo::toFeatures(bool AddAllExtensions,
                                                   bool IgnoreUnknown) const {
   std::vector<std::string> Features;
@@ -808,7 +505,7 @@ static Error splitExtsByUnderscore(StringRef Exts,
 
 static Error processMultiLetterExtension(
     StringRef RawExt,
-    MapVector<std::string, RISCVISAInfo::ExtensionVersion,
+    MapVector<std::string, RISCVISAUtils::ExtensionVersion,
               std::map<std::string, unsigned>> &SeenExtMap,
     bool IgnoreUnknown, bool EnableExperimentalExtension,
     bool ExperimentalExtensionVersionCheck) {
@@ -854,7 +551,7 @@ static Error processMultiLetterExtension(
 
 static Error processSingleLetterExtension(
     StringRef &RawExt,
-    MapVector<std::string, RISCVISAInfo::ExtensionVersion,
+    MapVector<std::string, RISCVISAUtils::ExtensionVersion,
               std::map<std::string, unsigned>> &SeenExtMap,
     bool IgnoreUnknown, bool EnableExperimentalExtension,
     bool ExperimentalExtensionVersionCheck) {
@@ -930,7 +627,7 @@ RISCVISAInfo::parseArchString(StringRef Arch, bool EnableExperimentalExtension,
 
   unsigned XLen = HasRV64 ? 64 : 32;
   std::unique_ptr<RISCVISAInfo> ISAInfo(new RISCVISAInfo(XLen));
-  MapVector<std::string, RISCVISAInfo::ExtensionVersion,
+  MapVector<std::string, RISCVISAUtils::ExtensionVersion,
             std::map<std::string, unsigned>>
       SeenExtMap;
 
@@ -942,7 +639,8 @@ RISCVISAInfo::parseArchString(StringRef Arch, bool EnableExperimentalExtension,
   switch (Baseline) {
   default:
     return createStringError(errc::invalid_argument,
-                             "first letter should be 'e', 'i' or 'g'");
+                             "first letter after \'" + Arch.slice(0, 4) +
+                                 "\' should be 'e', 'i' or 'g'");
   case 'e':
   case 'i':
     break;
@@ -1007,7 +705,7 @@ RISCVISAInfo::parseArchString(StringRef Arch, bool EnableExperimentalExtension,
   for (auto &Ext : SplitExts) {
     StringRef CurrExt = Ext;
     while (!CurrExt.empty()) {
-      if (AllStdExts.contains(CurrExt.front())) {
+      if (RISCVISAUtils::AllStdExts.contains(CurrExt.front())) {
         if (auto E = processSingleLetterExtension(
                 CurrExt, SeenExtMap, IgnoreUnknown, EnableExperimentalExtension,
                 ExperimentalExtensionVersionCheck))
@@ -1041,7 +739,7 @@ RISCVISAInfo::parseArchString(StringRef Arch, bool EnableExperimentalExtension,
   // Check all Extensions are supported.
   for (auto &SeenExtAndVers : SeenExtMap) {
     const std::string &ExtName = SeenExtAndVers.first;
-    RISCVISAInfo::ExtensionVersion ExtVers = SeenExtAndVers.second;
+    RISCVISAUtils::ExtensionVersion ExtVers = SeenExtAndVers.second;
 
     if (!RISCVISAInfo::isSupportedExtension(ExtName))
       return getStringErrorForInvalidExt(ExtName);
@@ -1102,145 +800,38 @@ Error RISCVISAInfo::checkDependency() {
     return createStringError(errc::invalid_argument,
                              "'zcf' is only supported for 'rv32'");
 
+  if (Exts.count("zacas") && !(Exts.count("a") || Exts.count("zamo")))
+    return createStringError(
+        errc::invalid_argument,
+        "'zacas' requires 'a' or 'zaamo' extension to also be specified");
+
+  if (Exts.count("zabha") && !(Exts.count("a") || Exts.count("zamo")))
+    return createStringError(
+        errc::invalid_argument,
+        "'zabha' requires 'a' or 'zaamo' extension to also be specified");
+
   return Error::success();
 }
 
-static const char *ImpliedExtsD[] = {"f"};
-static const char *ImpliedExtsF[] = {"zicsr"};
-static const char *ImpliedExtsV[] = {"zvl128b", "zve64d"};
-static const char *ImpliedExtsXTHeadVdot[] = {"v"};
-static const char *ImpliedExtsXSfvcp[] = {"zve32x"};
-static const char *ImpliedExtsXSfvfnrclipxfqf[] = {"zve32f"};
-static const char *ImpliedExtsXSfvfwmaccqqq[] = {"zvfbfmin"};
-static const char *ImpliedExtsXSfvqmaccdod[] = {"zve32x"};
-static const char *ImpliedExtsXSfvqmaccqoq[] = {"zve32x"};
-static const char *ImpliedExtsZabha[] = {"a"};
-static const char *ImpliedExtsZacas[] = {"a"};
-static const char *ImpliedExtsZcb[] = {"zca"};
-static const char *ImpliedExtsZcd[] = {"d", "zca"};
-static const char *ImpliedExtsZce[] = {"zcb", "zcmp", "zcmt"};
-static const char *ImpliedExtsZcf[] = {"f", "zca"};
-static const char *ImpliedExtsZcmop[] = {"zca"};
-static const char *ImpliedExtsZcmp[] = {"zca"};
-static const char *ImpliedExtsZcmt[] = {"zca", "zicsr"};
-static const char *ImpliedExtsZdinx[] = {"zfinx"};
-static const char *ImpliedExtsZfa[] = {"f"};
-static const char *ImpliedExtsZfbfmin[] = {"f"};
-static const char *ImpliedExtsZfh[] = {"zfhmin"};
-static const char *ImpliedExtsZfhmin[] = {"f"};
-static const char *ImpliedExtsZfinx[] = {"zicsr"};
-static const char *ImpliedExtsZhinx[] = {"zhinxmin"};
-static const char *ImpliedExtsZhinxmin[] = {"zfinx"};
-static const char *ImpliedExtsZicntr[] = {"zicsr"};
-static const char *ImpliedExtsZicfiss[] = {"zicsr", "zimop"};
-static const char *ImpliedExtsZihpm[] = {"zicsr"};
-static const char *ImpliedExtsZk[] = {"zkn", "zkt", "zkr"};
-static const char *ImpliedExtsZkn[] = {"zbkb", "zbkc", "zbkx",
-                                       "zkne", "zknd", "zknh"};
-static const char *ImpliedExtsZks[] = {"zbkb", "zbkc", "zbkx", "zksed", "zksh"};
-static const char *ImpliedExtsZvbb[] = {"zvkb"};
-static const char *ImpliedExtsZve32f[] = {"zve32x", "f"};
-static const char *ImpliedExtsZve32x[] = {"zvl32b", "zicsr"};
-static const char *ImpliedExtsZve64d[] = {"zve64f", "d"};
-static const char *ImpliedExtsZve64f[] = {"zve64x", "zve32f"};
-static const char *ImpliedExtsZve64x[] = {"zve32x", "zvl64b"};
-static const char *ImpliedExtsZvfbfmin[] = {"zve32f"};
-static const char *ImpliedExtsZvfbfwma[] = {"zvfbfmin", "zfbfmin"};
-static const char *ImpliedExtsZvfh[] = {"zvfhmin", "zfhmin"};
-static const char *ImpliedExtsZvfhmin[] = {"zve32f"};
-static const char *ImpliedExtsZvkn[] = {"zvkb", "zvkned", "zvknhb", "zvkt"};
-static const char *ImpliedExtsZvknc[] = {"zvbc", "zvkn"};
-static const char *ImpliedExtsZvkng[] = {"zvkg", "zvkn"};
-static const char *ImpliedExtsZvknhb[] = {"zve64x"};
-static const char *ImpliedExtsZvks[] = {"zvkb", "zvksed", "zvksh", "zvkt"};
-static const char *ImpliedExtsZvksc[] = {"zvbc", "zvks"};
-static const char *ImpliedExtsZvksg[] = {"zvkg", "zvks"};
-static const char *ImpliedExtsZvl1024b[] = {"zvl512b"};
-static const char *ImpliedExtsZvl128b[] = {"zvl64b"};
-static const char *ImpliedExtsZvl16384b[] = {"zvl8192b"};
-static const char *ImpliedExtsZvl2048b[] = {"zvl1024b"};
-static const char *ImpliedExtsZvl256b[] = {"zvl128b"};
-static const char *ImpliedExtsZvl32768b[] = {"zvl16384b"};
-static const char *ImpliedExtsZvl4096b[] = {"zvl2048b"};
-static const char *ImpliedExtsZvl512b[] = {"zvl256b"};
-static const char *ImpliedExtsZvl64b[] = {"zvl32b"};
-static const char *ImpliedExtsZvl65536b[] = {"zvl32768b"};
-static const char *ImpliedExtsZvl8192b[] = {"zvl4096b"};
-
 struct ImpliedExtsEntry {
   StringLiteral Name;
-  ArrayRef<const char *> Exts;
+  const char *ImpliedExt;
 
   bool operator<(const ImpliedExtsEntry &Other) const {
     return Name < Other.Name;
   }
-
-  bool operator<(StringRef Other) const { return Name < Other; }
 };
 
-// Note: The table needs to be sorted by name.
-static constexpr ImpliedExtsEntry ImpliedExts[] = {
-    {{"d"}, {ImpliedExtsD}},
-    {{"f"}, {ImpliedExtsF}},
-    {{"v"}, {ImpliedExtsV}},
-    {{"xsfvcp"}, {ImpliedExtsXSfvcp}},
-    {{"xsfvfnrclipxfqf"}, {ImpliedExtsXSfvfnrclipxfqf}},
-    {{"xsfvfwmaccqqq"}, {ImpliedExtsXSfvfwmaccqqq}},
-    {{"xsfvqmaccdod"}, {ImpliedExtsXSfvqmaccdod}},
-    {{"xsfvqmaccqoq"}, {ImpliedExtsXSfvqmaccqoq}},
-    {{"xtheadvdot"}, {ImpliedExtsXTHeadVdot}},
-    {{"zabha"}, {ImpliedExtsZabha}},
-    {{"zacas"}, {ImpliedExtsZacas}},
-    {{"zcb"}, {ImpliedExtsZcb}},
-    {{"zcd"}, {ImpliedExtsZcd}},
-    {{"zce"}, {ImpliedExtsZce}},
-    {{"zcf"}, {ImpliedExtsZcf}},
-    {{"zcmop"}, {ImpliedExtsZcmop}},
-    {{"zcmp"}, {ImpliedExtsZcmp}},
-    {{"zcmt"}, {ImpliedExtsZcmt}},
-    {{"zdinx"}, {ImpliedExtsZdinx}},
-    {{"zfa"}, {ImpliedExtsZfa}},
-    {{"zfbfmin"}, {ImpliedExtsZfbfmin}},
-    {{"zfh"}, {ImpliedExtsZfh}},
-    {{"zfhmin"}, {ImpliedExtsZfhmin}},
-    {{"zfinx"}, {ImpliedExtsZfinx}},
-    {{"zhinx"}, {ImpliedExtsZhinx}},
-    {{"zhinxmin"}, {ImpliedExtsZhinxmin}},
-    {{"zicfiss"}, {ImpliedExtsZicfiss}},
-    {{"zicntr"}, {ImpliedExtsZicntr}},
-    {{"zihpm"}, {ImpliedExtsZihpm}},
-    {{"zk"}, {ImpliedExtsZk}},
-    {{"zkn"}, {ImpliedExtsZkn}},
-    {{"zks"}, {ImpliedExtsZks}},
-    {{"zvbb"}, {ImpliedExtsZvbb}},
-    {{"zve32f"}, {ImpliedExtsZve32f}},
-    {{"zve32x"}, {ImpliedExtsZve32x}},
-    {{"zve64d"}, {ImpliedExtsZve64d}},
-    {{"zve64f"}, {ImpliedExtsZve64f}},
-    {{"zve64x"}, {ImpliedExtsZve64x}},
-    {{"zvfbfmin"}, {ImpliedExtsZvfbfmin}},
-    {{"zvfbfwma"}, {ImpliedExtsZvfbfwma}},
-    {{"zvfh"}, {ImpliedExtsZvfh}},
-    {{"zvfhmin"}, {ImpliedExtsZvfhmin}},
-    {{"zvkn"}, {ImpliedExtsZvkn}},
-    {{"zvknc"}, {ImpliedExtsZvknc}},
-    {{"zvkng"}, {ImpliedExtsZvkng}},
-    {{"zvknhb"}, {ImpliedExtsZvknhb}},
-    {{"zvks"}, {ImpliedExtsZvks}},
-    {{"zvksc"}, {ImpliedExtsZvksc}},
-    {{"zvksg"}, {ImpliedExtsZvksg}},
-    {{"zvl1024b"}, {ImpliedExtsZvl1024b}},
-    {{"zvl128b"}, {ImpliedExtsZvl128b}},
-    {{"zvl16384b"}, {ImpliedExtsZvl16384b}},
-    {{"zvl2048b"}, {ImpliedExtsZvl2048b}},
-    {{"zvl256b"}, {ImpliedExtsZvl256b}},
-    {{"zvl32768b"}, {ImpliedExtsZvl32768b}},
-    {{"zvl4096b"}, {ImpliedExtsZvl4096b}},
-    {{"zvl512b"}, {ImpliedExtsZvl512b}},
-    {{"zvl64b"}, {ImpliedExtsZvl64b}},
-    {{"zvl65536b"}, {ImpliedExtsZvl65536b}},
-    {{"zvl8192b"}, {ImpliedExtsZvl8192b}},
-};
+static bool operator<(const ImpliedExtsEntry &LHS, StringRef RHS) {
+  return LHS.Name < RHS;
+}
+
+static bool operator<(StringRef LHS, const ImpliedExtsEntry &RHS) {
+  return LHS < RHS.Name;
+}
+
+#define GET_IMPLIED_EXTENSIONS
+#include "llvm/TargetParser/RISCVTargetParserDef.inc"
 
 void RISCVISAInfo::updateImplication() {
   bool HasE = Exts.count("e") != 0;
@@ -1263,18 +854,19 @@ void RISCVISAInfo::updateImplication() {
 
   while (!WorkList.empty()) {
     StringRef ExtName = WorkList.pop_back_val();
-    auto I = llvm::lower_bound(ImpliedExts, ExtName);
-    if (I != std::end(ImpliedExts) && I->Name == ExtName) {
-      for (const char *ImpliedExt : I->Exts) {
-        if (WorkList.count(ImpliedExt))
-          continue;
-        if (Exts.count(ImpliedExt))
-          continue;
-        auto Version = findDefaultVersion(ImpliedExt);
-        addExtension(ImpliedExt, Version.value());
-        WorkList.insert(ImpliedExt);
-      }
-    }
+    auto Range = std::equal_range(std::begin(ImpliedExts),
+                                  std::end(ImpliedExts), ExtName);
+    std::for_each(Range.first, Range.second,
+                  [&](const ImpliedExtsEntry &Implied) {
+                    const char *ImpliedExt = Implied.ImpliedExt;
+                    if (WorkList.count(ImpliedExt))
+                      return;
+                    if (Exts.count(ImpliedExt))
+                      return;
+                    auto Version = findDefaultVersion(ImpliedExt);
+                    addExtension(ImpliedExt, Version.value());
+                    WorkList.insert(ImpliedExt);
+                  });
   }
 
   // Add Zcf if Zce and F are enabled on RV32.
@@ -1285,42 +877,34 @@ void RISCVISAInfo::updateImplication() {
   }
 }
 
-struct CombinedExtsEntry {
-  StringLiteral CombineExt;
-  ArrayRef<const char *> RequiredExts;
-};
-
-static constexpr CombinedExtsEntry CombineIntoExts[] = {
-    {{"zk"}, {ImpliedExtsZk}},
-    {{"zkn"}, {ImpliedExtsZkn}},
-    {{"zks"}, {ImpliedExtsZks}},
-    {{"zvkn"}, {ImpliedExtsZvkn}},
-    {{"zvknc"}, {ImpliedExtsZvknc}},
-    {{"zvkng"}, {ImpliedExtsZvkng}},
-    {{"zvks"}, {ImpliedExtsZvks}},
-    {{"zvksc"}, {ImpliedExtsZvksc}},
-    {{"zvksg"}, {ImpliedExtsZvksg}},
+static constexpr StringLiteral CombineIntoExts[] = {
+    {"zk"},    {"zkn"},  {"zks"},   {"zvkn"},  {"zvknc"},
+    {"zvkng"}, {"zvks"}, {"zvksc"}, {"zvksg"},
 };
 
 void RISCVISAInfo::updateCombination() {
-  bool IsNewCombine = false;
+  bool MadeChange = false;
   do {
-    IsNewCombine = false;
-    for (CombinedExtsEntry CombineIntoExt : CombineIntoExts) {
-      auto CombineExt = CombineIntoExt.CombineExt;
-      auto RequiredExts = CombineIntoExt.RequiredExts;
+    MadeChange = false;
+    for (StringRef CombineExt : CombineIntoExts) {
       if (hasExtension(CombineExt))
         continue;
-      bool IsAllRequiredFeatureExist = true;
-      for (const char *Ext : RequiredExts)
-        IsAllRequiredFeatureExist &= hasExtension(Ext);
-      if (IsAllRequiredFeatureExist) {
+
+      // Look up the extension in the ImpliesExt table to find everything it
+      // depends on.
+      auto Range = std::equal_range(std::begin(ImpliedExts),
+                                    std::end(ImpliedExts), CombineExt);
+      bool HasAllRequiredFeatures = std::all_of(
+          Range.first, Range.second, [&](const ImpliedExtsEntry &Implied) {
+            return hasExtension(Implied.ImpliedExt);
+          });
+      if (HasAllRequiredFeatures) {
         auto Version = findDefaultVersion(CombineExt);
         addExtension(CombineExt, Version.value());
-        IsNewCombine = true;
+        MadeChange = true;
       }
     }
-  } while (IsNewCombine);
+  } while (MadeChange);
 }
 
 void RISCVISAInfo::updateFLen() {
diff --git a/llvm/lib/TargetParser/RISCVTargetParser.cpp b/llvm/lib/TargetParser/RISCVTargetParser.cpp
index 0d95e3a9b819..9003f9beffa7 100644
--- a/llvm/lib/TargetParser/RISCVTargetParser.cpp
+++ b/llvm/lib/TargetParser/RISCVTargetParser.cpp
@@ -14,7 +14,7 @@
 #include "llvm/TargetParser/RISCVTargetParser.h"
 #include "llvm/ADT/SmallVector.h"
 #include "llvm/ADT/StringSwitch.h"
-#include "llvm/Support/RISCVISAInfo.h"
+#include "llvm/TargetParser/RISCVISAInfo.h"
 #include "llvm/TargetParser/Triple.h"
 
 namespace llvm {
diff --git a/llvm/lib/Transforms/Coroutines/CoroInstr.h b/llvm/lib/Transforms/Coroutines/CoroInstr.h
index 79e745bb162c..a31703fe0130 100644
--- a/llvm/lib/Transforms/Coroutines/CoroInstr.h
+++ b/llvm/lib/Transforms/Coroutines/CoroInstr.h
@@ -78,10 +78,10 @@ public:
   }
 };
 
-/// This represents the llvm.coro.await.suspend instruction.
+/// This represents the llvm.coro.await.suspend.{void,bool,handle} instructions.
 // FIXME: add callback metadata
 // FIXME: make a proper IntrinisicInst. Currently this is not possible,
-// because llvm.coro.await.suspend can be invoked.
+// because llvm.coro.await.suspend.* can be invoked.
 class LLVM_LIBRARY_VISIBILITY CoroAwaitSuspendInst : public CallBase {
   enum { AwaiterArg, FrameArg, WrapperArg };
 
diff --git a/llvm/lib/Transforms/IPO/FunctionAttrs.cpp b/llvm/lib/Transforms/IPO/FunctionAttrs.cpp
index 7ebf265e17ba..8e11cbf1cee4 100644
--- a/llvm/lib/Transforms/IPO/FunctionAttrs.cpp
+++ b/llvm/lib/Transforms/IPO/FunctionAttrs.cpp
@@ -1287,7 +1287,8 @@ static void addNoUndefAttrs(const SCCNodeSet &SCCNodes,
   // values.
   for (Function *F : SCCNodes) {
     // Already noundef.
-    if (F->getAttributes().hasRetAttr(Attribute::NoUndef))
+    AttributeList Attrs = F->getAttributes();
+    if (Attrs.hasRetAttr(Attribute::NoUndef))
       continue;
 
     // We can infer and propagate function attributes only when we know that the
@@ -1305,10 +1306,30 @@ static void addNoUndefAttrs(const SCCNodeSet &SCCNodes,
     if (F->getReturnType()->isVoidTy())
       continue;
 
-    if (all_of(*F, [](BasicBlock &BB) {
+    const DataLayout &DL = F->getParent()->getDataLayout();
+    if (all_of(*F, [&](BasicBlock &BB) {
           if (auto *Ret = dyn_cast<ReturnInst>(BB.getTerminator())) {
             // TODO: perform context-sensitive analysis?
-            return isGuaranteedNotToBeUndefOrPoison(Ret->getReturnValue());
+            Value *RetVal = Ret->getReturnValue();
+            if (!isGuaranteedNotToBeUndefOrPoison(RetVal))
+              return false;
+
+            // We know the original return value is not poison now, but it
+            // could still be converted to poison by another return attribute.
+            // Try to explicitly re-prove the relevant attributes.
+            if (Attrs.hasRetAttr(Attribute::NonNull) &&
+                !isKnownNonZero(RetVal, DL))
+              return false;
+
+            if (MaybeAlign Align = Attrs.getRetAlignment())
+              if (RetVal->getPointerAlignment(DL) < *Align)
+                return false;
+
+            Attribute Attr = Attrs.getRetAttr(Attribute::Range);
+            if (Attr.isValid() &&
+                !Attr.getRange().contains(
+                    computeConstantRange(RetVal, /*ForSigned=*/false)))
+              return false;
           }
           return true;
         })) {
diff --git a/llvm/lib/Transforms/IPO/GlobalOpt.cpp b/llvm/lib/Transforms/IPO/GlobalOpt.cpp
index da714c9a7570..fbb83e787f63 100644
--- a/llvm/lib/Transforms/IPO/GlobalOpt.cpp
+++ b/llvm/lib/Transforms/IPO/GlobalOpt.cpp
@@ -306,6 +306,10 @@ static bool CleanupConstantGlobalUsers(GlobalVariable *GV,
       APInt Offset(DL.getIndexTypeSizeInBits(PtrOp->getType()), 0);
       PtrOp = PtrOp->stripAndAccumulateConstantOffsets(
           DL, Offset, /* AllowNonInbounds */ true);
+      if (IntrinsicInst *II = dyn_cast<IntrinsicInst>(PtrOp)) {
+        if (II->getIntrinsicID() == Intrinsic::threadlocal_address)
+          PtrOp = II->getArgOperand(0);
+      }
       if (PtrOp == GV) {
         if (auto *Value = ConstantFoldLoadFromConst(Init, Ty, Offset, DL)) {
           LI->replaceAllUsesWith(Value);
@@ -318,6 +322,9 @@ static bool CleanupConstantGlobalUsers(GlobalVariable *GV,
     } else if (MemIntrinsic *MI = dyn_cast<MemIntrinsic>(U)) { // memset/cpy/mv
       if (getUnderlyingObject(MI->getRawDest()) == GV)
         EraseFromParent(MI);
+    } else if (IntrinsicInst *II = dyn_cast<IntrinsicInst>(U)) {
+      if (II->getIntrinsicID() == Intrinsic::threadlocal_address)
+        append_range(WorkList, II->users());
     }
   }
 
diff --git a/llvm/lib/Transforms/IPO/SCCP.cpp b/llvm/lib/Transforms/IPO/SCCP.cpp
index f8920541e6fd..e591a8e73b1c 100644
--- a/llvm/lib/Transforms/IPO/SCCP.cpp
+++ b/llvm/lib/Transforms/IPO/SCCP.cpp
@@ -281,32 +281,21 @@ static bool runIPSCCP(
     Function *F = I.first;
     const ValueLatticeElement &ReturnValue = I.second;
 
-    // If there is a known constant range for the return value, add !range
-    // metadata to the function's call sites.
+    // If there is a known constant range for the return value, add range
+    // attribute to the return value.
     if (ReturnValue.isConstantRange() &&
         !ReturnValue.getConstantRange().isSingleElement()) {
       // Do not add range metadata if the return value may include undef.
       if (ReturnValue.isConstantRangeIncludingUndef())
         continue;
 
+      // Do not touch existing attribute for now.
+      // TODO: We should be able to take the intersection of the existing
+      // attribute and the inferred range.
+      if (F->hasRetAttribute(Attribute::Range))
+        continue;
       auto &CR = ReturnValue.getConstantRange();
-      for (User *User : F->users()) {
-        auto *CB = dyn_cast<CallBase>(User);
-        if (!CB || CB->getCalledFunction() != F)
-          continue;
-
-        // Do not touch existing metadata for now.
-        // TODO: We should be able to take the intersection of the existing
-        // metadata and the inferred range.
-        if (CB->getMetadata(LLVMContext::MD_range))
-          continue;
-
-        LLVMContext &Context = CB->getParent()->getContext();
-        Metadata *RangeMD[] = {
-            ConstantAsMetadata::get(ConstantInt::get(Context, CR.getLower())),
-            ConstantAsMetadata::get(ConstantInt::get(Context, CR.getUpper()))};
-        CB->setMetadata(LLVMContext::MD_range, MDNode::get(Context, RangeMD));
-      }
+      F->addRangeRetAttr(CR);
       continue;
     }
     if (F->getReturnType()->isVoidTy())
diff --git a/llvm/lib/Transforms/IPO/SampleProfile.cpp b/llvm/lib/Transforms/IPO/SampleProfile.cpp
index 0b3a6931e779..6cbd138842c8 100644
--- a/llvm/lib/Transforms/IPO/SampleProfile.cpp
+++ b/llvm/lib/Transforms/IPO/SampleProfile.cpp
@@ -252,20 +252,21 @@ static cl::opt<unsigned> PrecentMismatchForStalenessError(
 
 static cl::opt<bool> CallsitePrioritizedInline(
     "sample-profile-prioritized-inline", cl::Hidden,
-
     cl::desc("Use call site prioritized inlining for sample profile loader."
              "Currently only CSSPGO is supported."));
 
 static cl::opt<bool> UsePreInlinerDecision(
     "sample-profile-use-preinliner", cl::Hidden,
-
     cl::desc("Use the preinliner decisions stored in profile context."));
 
 static cl::opt<bool> AllowRecursiveInline(
     "sample-profile-recursive-inline", cl::Hidden,
-
     cl::desc("Allow sample loader inliner to inline recursive calls."));
 
+static cl::opt<bool> RemoveProbeAfterProfileAnnotation(
+    "sample-profile-remove-probe", cl::Hidden, cl::init(false),
+    cl::desc("Remove pseudo-probe after sample profile annotation."));
+
 static cl::opt<std::string> ProfileInlineReplayFile(
     "sample-profile-inline-replay", cl::init(""), cl::value_desc("filename"),
     cl::desc(
@@ -518,6 +519,7 @@ protected:
   void generateMDProfMetadata(Function &F);
   bool rejectHighStalenessProfile(Module &M, ProfileSummaryInfo *PSI,
                                   const SampleProfileMap &Profiles);
+  void removePseudoProbeInsts(Module &M);
 
   /// Map from function name to Function *. Used to find the function from
   /// the function name. If the function name contains suffix, additional
@@ -2127,6 +2129,20 @@ bool SampleProfileLoader::rejectHighStalenessProfile(
   return false;
 }
 
+void SampleProfileLoader::removePseudoProbeInsts(Module &M) {
+  for (auto &F : M) {
+    std::vector<Instruction *> InstsToDel;
+    for (auto &BB : F) {
+      for (auto &I : BB) {
+        if (isa<PseudoProbeInst>(&I))
+          InstsToDel.push_back(&I);
+      }
+    }
+    for (auto *I : InstsToDel)
+      I->eraseFromParent();
+  }
+}
+
 bool SampleProfileLoader::runOnModule(Module &M, ModuleAnalysisManager *AM,
                                       ProfileSummaryInfo *_PSI,
                                       LazyCallGraph &CG) {
@@ -2196,6 +2212,9 @@ bool SampleProfileLoader::runOnModule(Module &M, ModuleAnalysisManager *AM,
          notInlinedCallInfo)
       updateProfileCallee(pair.first, pair.second.entryCount);
 
+  if (RemoveProbeAfterProfileAnnotation && FunctionSamples::ProfileIsProbeBased)
+    removePseudoProbeInsts(M);
+
   return retval;
 }
 
diff --git a/llvm/lib/Transforms/InstCombine/InstCombineAddSub.cpp b/llvm/lib/Transforms/InstCombine/InstCombineAddSub.cpp
index fc284bc61cce..51ac77348ed9 100644
--- a/llvm/lib/Transforms/InstCombine/InstCombineAddSub.cpp
+++ b/llvm/lib/Transforms/InstCombine/InstCombineAddSub.cpp
@@ -1134,6 +1134,8 @@ static bool MulWillOverflow(APInt &C0, APInt &C1, bool IsSigned) {
 
 // Simplifies X % C0 + (( X / C0 ) % C1) * C0 to X % (C0 * C1), where (C0 * C1)
 // does not overflow.
+// Simplifies (X / C0) * C1 + (X % C0) * C2 to
+// (X / C0) * (C1 - C2 * C0) + X * C2
 Value *InstCombinerImpl::SimplifyAddWithRemainder(BinaryOperator &I) {
   Value *LHS = I.getOperand(0), *RHS = I.getOperand(1);
   Value *X, *MulOpV;
@@ -1161,6 +1163,33 @@ Value *InstCombinerImpl::SimplifyAddWithRemainder(BinaryOperator &I) {
     }
   }
 
+  // Match I = (X / C0) * C1 + (X % C0) * C2
+  Value *Div, *Rem;
+  APInt C1, C2;
+  if (!LHS->hasOneUse() || !MatchMul(LHS, Div, C1))
+    Div = LHS, C1 = APInt(I.getType()->getScalarSizeInBits(), 1);
+  if (!RHS->hasOneUse() || !MatchMul(RHS, Rem, C2))
+    Rem = RHS, C2 = APInt(I.getType()->getScalarSizeInBits(), 1);
+  if (match(Div, m_IRem(m_Value(), m_Value()))) {
+    std::swap(Div, Rem);
+    std::swap(C1, C2);
+  }
+  Value *DivOpV;
+  APInt DivOpC;
+  if (MatchRem(Rem, X, C0, IsSigned) &&
+      MatchDiv(Div, DivOpV, DivOpC, IsSigned) && X == DivOpV && C0 == DivOpC) {
+    APInt NewC = C1 - C2 * C0;
+    if (!NewC.isZero() && !Rem->hasOneUse())
+      return nullptr;
+    if (!isGuaranteedNotToBeUndef(X, &AC, &I, &DT))
+      return nullptr;
+    Value *MulXC2 = Builder.CreateMul(X, ConstantInt::get(X->getType(), C2));
+    if (NewC.isZero())
+      return MulXC2;
+    return Builder.CreateAdd(
+        Builder.CreateMul(Div, ConstantInt::get(X->getType(), NewC)), MulXC2);
+  }
+
   return nullptr;
 }
 
@@ -1972,43 +2001,30 @@ Value *InstCombinerImpl::OptimizePointerDifference(Value *LHS, Value *RHS,
   if (!GEP1)
     return nullptr;
 
-  if (GEP2) {
-    // (gep X, ...) - (gep X, ...)
-    //
-    // Avoid duplicating the arithmetic if there are more than one non-constant
-    // indices between the two GEPs and either GEP has a non-constant index and
-    // multiple users. If zero non-constant index, the result is a constant and
-    // there is no duplication. If one non-constant index, the result is an add
-    // or sub with a constant, which is no larger than the original code, and
-    // there's no duplicated arithmetic, even if either GEP has multiple
-    // users. If more than one non-constant indices combined, as long as the GEP
-    // with at least one non-constant index doesn't have multiple users, there
-    // is no duplication.
-    unsigned NumNonConstantIndices1 = GEP1->countNonConstantIndices();
-    unsigned NumNonConstantIndices2 = GEP2->countNonConstantIndices();
-    if (NumNonConstantIndices1 + NumNonConstantIndices2 > 1 &&
-        ((NumNonConstantIndices1 > 0 && !GEP1->hasOneUse()) ||
-         (NumNonConstantIndices2 > 0 && !GEP2->hasOneUse()))) {
-      return nullptr;
-    }
-  }
+  // To avoid duplicating the offset arithmetic, rewrite the GEP to use the
+  // computed offset. This may erase the original GEP, so be sure to cache the
+  // inbounds flag before emitting the offset.
+  // TODO: We should probably do this even if there is only one GEP.
+  bool RewriteGEPs = GEP2 != nullptr;
 
   // Emit the offset of the GEP and an intptr_t.
-  Value *Result = EmitGEPOffset(GEP1);
+  bool GEP1IsInBounds = GEP1->isInBounds();
+  Value *Result = EmitGEPOffset(GEP1, RewriteGEPs);
 
   // If this is a single inbounds GEP and the original sub was nuw,
   // then the final multiplication is also nuw.
   if (auto *I = dyn_cast<Instruction>(Result))
-    if (IsNUW && !GEP2 && !Swapped && GEP1->isInBounds() &&
+    if (IsNUW && !GEP2 && !Swapped && GEP1IsInBounds &&
         I->getOpcode() == Instruction::Mul)
       I->setHasNoUnsignedWrap();
 
   // If we have a 2nd GEP of the same base pointer, subtract the offsets.
   // If both GEPs are inbounds, then the subtract does not have signed overflow.
   if (GEP2) {
-    Value *Offset = EmitGEPOffset(GEP2);
+    bool GEP2IsInBounds = GEP2->isInBounds();
+    Value *Offset = EmitGEPOffset(GEP2, RewriteGEPs);
     Result = Builder.CreateSub(Result, Offset, "gepdiff", /* NUW */ false,
-                               GEP1->isInBounds() && GEP2->isInBounds());
+                               GEP1IsInBounds && GEP2IsInBounds);
   }
 
   // If we have p - gep(p, ...)  then we have to negate the result.
@@ -2752,6 +2768,16 @@ Instruction *InstCombinerImpl::visitFNeg(UnaryOperator &I) {
       propagateSelectFMF(NewSel, P == X);
       return NewSel;
     }
+
+    // -(Cond ? X : C) --> Cond ? -X : -C
+    // -(Cond ? C : Y) --> Cond ? -C : -Y
+    if (match(X, m_ImmConstant()) || match(Y, m_ImmConstant())) {
+      Value *NegX = Builder.CreateFNegFMF(X, &I, X->getName() + ".neg");
+      Value *NegY = Builder.CreateFNegFMF(Y, &I, Y->getName() + ".neg");
+      SelectInst *NewSel = SelectInst::Create(Cond, NegX, NegY);
+      propagateSelectFMF(NewSel, /*CommonOperand=*/true);
+      return NewSel;
+    }
   }
 
   // fneg (copysign x, y) -> copysign x, (fneg y)
diff --git a/llvm/lib/Transforms/InstCombine/InstCombineAndOrXor.cpp b/llvm/lib/Transforms/InstCombine/InstCombineAndOrXor.cpp
index e1923a344179..ed9a89b14efc 100644
--- a/llvm/lib/Transforms/InstCombine/InstCombineAndOrXor.cpp
+++ b/llvm/lib/Transforms/InstCombine/InstCombineAndOrXor.cpp
@@ -3141,20 +3141,20 @@ Value *InstCombinerImpl::getSelectCondition(Value *A, Value *B,
   return nullptr;
 }
 
-/// We have an expression of the form (A & C) | (B & D). Try to simplify this
-/// to "A' ? C : D", where A' is a boolean or vector of booleans.
+/// We have an expression of the form (A & B) | (C & D). Try to simplify this
+/// to "A' ? B : D", where A' is a boolean or vector of booleans.
 /// When InvertFalseVal is set to true, we try to match the pattern
-/// where we have peeked through a 'not' op and A and B are the same:
-/// (A & C) | ~(A | D) --> (A & C) | (~A & ~D) --> A' ? C : ~D
-Value *InstCombinerImpl::matchSelectFromAndOr(Value *A, Value *C, Value *B,
+/// where we have peeked through a 'not' op and A and C are the same:
+/// (A & B) | ~(A | D) --> (A & B) | (~A & ~D) --> A' ? B : ~D
+Value *InstCombinerImpl::matchSelectFromAndOr(Value *A, Value *B, Value *C,
                                               Value *D, bool InvertFalseVal) {
   // The potential condition of the select may be bitcasted. In that case, look
   // through its bitcast and the corresponding bitcast of the 'not' condition.
   Type *OrigType = A->getType();
   A = peekThroughBitcast(A, true);
-  B = peekThroughBitcast(B, true);
-  if (Value *Cond = getSelectCondition(A, B, InvertFalseVal)) {
-    // ((bc Cond) & C) | ((bc ~Cond) & D) --> bc (select Cond, (bc C), (bc D))
+  C = peekThroughBitcast(C, true);
+  if (Value *Cond = getSelectCondition(A, C, InvertFalseVal)) {
+    // ((bc Cond) & B) | ((bc ~Cond) & D) --> bc (select Cond, (bc B), (bc D))
     // If this is a vector, we may need to cast to match the condition's length.
     // The bitcasts will either all exist or all not exist. The builder will
     // not create unnecessary casts if the types already match.
@@ -3168,11 +3168,11 @@ Value *InstCombinerImpl::matchSelectFromAndOr(Value *A, Value *C, Value *B,
       Type *EltTy = Builder.getIntNTy(SelEltSize / Elts);
       SelTy = VectorType::get(EltTy, VecTy->getElementCount());
     }
-    Value *BitcastC = Builder.CreateBitCast(C, SelTy);
+    Value *BitcastB = Builder.CreateBitCast(B, SelTy);
     if (InvertFalseVal)
       D = Builder.CreateNot(D);
     Value *BitcastD = Builder.CreateBitCast(D, SelTy);
-    Value *Select = Builder.CreateSelect(Cond, BitcastC, BitcastD);
+    Value *Select = Builder.CreateSelect(Cond, BitcastB, BitcastD);
     return Builder.CreateBitCast(Select, OrigType);
   }
 
@@ -3958,6 +3958,10 @@ Instruction *InstCombinerImpl::visitOr(BinaryOperator &I) {
                                       /*SimplifyOnly*/ false, *this))
     return BinaryOperator::CreateOr(Op0, V);
 
+  if (cast<PossiblyDisjointInst>(I).isDisjoint())
+    if (Value *V = SimplifyAddWithRemainder(I))
+      return replaceInstUsesWith(I, V);
+
   return nullptr;
 }
 
diff --git a/llvm/lib/Transforms/InstCombine/InstCombineCalls.cpp b/llvm/lib/Transforms/InstCombine/InstCombineCalls.cpp
index a37a4cde96f8..1913ef92c16c 100644
--- a/llvm/lib/Transforms/InstCombine/InstCombineCalls.cpp
+++ b/llvm/lib/Transforms/InstCombine/InstCombineCalls.cpp
@@ -606,14 +606,13 @@ static Instruction *foldCttzCtlz(IntrinsicInst &II, InstCombinerImpl &IC) {
       return IC.replaceOperand(II, 1, IC.Builder.getTrue());
   }
 
-  // Add range metadata since known bits can't completely reflect what we know.
-  auto *IT = cast<IntegerType>(Op0->getType()->getScalarType());
-  if (IT && IT->getBitWidth() != 1 && !II.getMetadata(LLVMContext::MD_range)) {
-    Metadata *LowAndHigh[] = {
-        ConstantAsMetadata::get(ConstantInt::get(IT, DefiniteZeros)),
-        ConstantAsMetadata::get(ConstantInt::get(IT, PossibleZeros + 1))};
-    II.setMetadata(LLVMContext::MD_range,
-                   MDNode::get(II.getContext(), LowAndHigh));
+  // Add range attribute since known bits can't completely reflect what we know.
+  unsigned BitWidth = Op0->getType()->getScalarSizeInBits();
+  if (BitWidth != 1 && !II.hasRetAttr(Attribute::Range) &&
+      !II.getMetadata(LLVMContext::MD_range)) {
+    ConstantRange Range(APInt(BitWidth, DefiniteZeros),
+                        APInt(BitWidth, PossibleZeros + 1));
+    II.addRangeRetAttr(Range);
     return &II;
   }
 
@@ -685,16 +684,12 @@ static Instruction *foldCtpop(IntrinsicInst &II, InstCombinerImpl &IC) {
                                                   Constant::getNullValue(Ty)),
                             Ty);
 
-  // Add range metadata since known bits can't completely reflect what we know.
-  auto *IT = cast<IntegerType>(Ty->getScalarType());
-  unsigned MinCount = Known.countMinPopulation();
-  unsigned MaxCount = Known.countMaxPopulation();
-  if (IT->getBitWidth() != 1 && !II.getMetadata(LLVMContext::MD_range)) {
-    Metadata *LowAndHigh[] = {
-        ConstantAsMetadata::get(ConstantInt::get(IT, MinCount)),
-        ConstantAsMetadata::get(ConstantInt::get(IT, MaxCount + 1))};
-    II.setMetadata(LLVMContext::MD_range,
-                   MDNode::get(II.getContext(), LowAndHigh));
+  // Add range attribute since known bits can't completely reflect what we know.
+  if (BitWidth != 1 && !II.hasRetAttr(Attribute::Range) &&
+      !II.getMetadata(LLVMContext::MD_range)) {
+    ConstantRange Range(APInt(BitWidth, Known.countMinPopulation()),
+                        APInt(BitWidth, Known.countMaxPopulation() + 1));
+    II.addRangeRetAttr(Range);
     return &II;
   }
 
@@ -3173,7 +3168,7 @@ Instruction *InstCombinerImpl::visitCallInst(CallInst &CI) {
     }
     break;
   }
-  case Intrinsic::experimental_vector_reverse: {
+  case Intrinsic::vector_reverse: {
     Value *BO0, *BO1, *X, *Y;
     Value *Vec = II->getArgOperand(0);
     if (match(Vec, m_OneUse(m_BinOp(m_Value(BO0), m_Value(BO1))))) {
diff --git a/llvm/lib/Transforms/InstCombine/InstCombineCasts.cpp b/llvm/lib/Transforms/InstCombine/InstCombineCasts.cpp
index 6ce39be59bda..567b27b46304 100644
--- a/llvm/lib/Transforms/InstCombine/InstCombineCasts.cpp
+++ b/llvm/lib/Transforms/InstCombine/InstCombineCasts.cpp
@@ -2050,9 +2050,9 @@ Instruction *InstCombinerImpl::visitPtrToInt(PtrToIntInst &CI) {
     // the GEP otherwise.
     if (GEP->hasOneUse() &&
         isa<ConstantPointerNull>(GEP->getPointerOperand())) {
-      return replaceInstUsesWith(CI,
-                                 Builder.CreateIntCast(EmitGEPOffset(GEP), Ty,
-                                                       /*isSigned=*/false));
+      return replaceInstUsesWith(
+          CI, Builder.CreateIntCast(EmitGEPOffset(cast<GEPOperator>(GEP)), Ty,
+                                    /*isSigned=*/false));
     }
   }
 
diff --git a/llvm/lib/Transforms/InstCombine/InstCombineCompares.cpp b/llvm/lib/Transforms/InstCombine/InstCombineCompares.cpp
index f0278f207549..f66883de8dd5 100644
--- a/llvm/lib/Transforms/InstCombine/InstCombineCompares.cpp
+++ b/llvm/lib/Transforms/InstCombine/InstCombineCompares.cpp
@@ -816,29 +816,9 @@ Instruction *InstCombinerImpl::foldGEPICmp(GEPOperator *GEPLHS, Value *RHS,
     }
 
     if (GEPsInBounds || CmpInst::isEquality(Cond)) {
-      auto EmitGEPOffsetAndRewrite = [&](GEPOperator *GEP) {
-        IRBuilderBase::InsertPointGuard Guard(Builder);
-        auto *Inst = dyn_cast<Instruction>(GEP);
-        if (Inst)
-          Builder.SetInsertPoint(Inst);
-
-        Value *Offset = EmitGEPOffset(GEP);
-        // If a non-trivial GEP has other uses, rewrite it to avoid duplicating
-        // the offset arithmetic.
-        if (Inst && !GEP->hasOneUse() && !GEP->hasAllConstantIndices() &&
-            !GEP->getSourceElementType()->isIntegerTy(8)) {
-          replaceInstUsesWith(*Inst,
-                              Builder.CreateGEP(Builder.getInt8Ty(),
-                                                GEP->getPointerOperand(),
-                                                Offset, "", GEPsInBounds));
-          eraseInstFromFunction(*Inst);
-        }
-        return Offset;
-      };
-
       // ((gep Ptr, OFFSET1) cmp (gep Ptr, OFFSET2)  --->  (OFFSET1 cmp OFFSET2)
-      Value *L = EmitGEPOffsetAndRewrite(GEPLHS);
-      Value *R = EmitGEPOffsetAndRewrite(GEPRHS);
+      Value *L = EmitGEPOffset(GEPLHS, /*RewriteGEP=*/true);
+      Value *R = EmitGEPOffset(GEPRHS, /*RewriteGEP=*/true);
       return new ICmpInst(ICmpInst::getSignedPredicate(Cond), L, R);
     }
   }
@@ -6909,8 +6889,8 @@ static Instruction *foldVectorCmp(CmpInst &Cmp,
     if (auto *I = dyn_cast<Instruction>(V))
       I->copyIRFlags(&Cmp);
     Module *M = Cmp.getModule();
-    Function *F = Intrinsic::getDeclaration(
-        M, Intrinsic::experimental_vector_reverse, V->getType());
+    Function *F =
+        Intrinsic::getDeclaration(M, Intrinsic::vector_reverse, V->getType());
     return CallInst::Create(F, V);
   };
 
@@ -8052,6 +8032,8 @@ Instruction *InstCombinerImpl::visitFCmpInst(FCmpInst &I) {
                  m_Select(m_Value(), m_Value(X), m_FNeg(m_Deferred(X)))) ||
            match(LHSI, m_Select(m_Value(), m_FNeg(m_Value(X)), m_Deferred(X)))))
         return replaceOperand(I, 0, X);
+      if (Instruction *NV = FoldOpIntoSelect(I, cast<SelectInst>(LHSI)))
+        return NV;
       break;
     case Instruction::PHI:
       if (Instruction *NV = foldOpIntoPhi(I, cast<PHINode>(LHSI)))
diff --git a/llvm/lib/Transforms/InstCombine/InstCombineInternal.h b/llvm/lib/Transforms/InstCombine/InstCombineInternal.h
index 4479afbd09af..db7838bbe3c2 100644
--- a/llvm/lib/Transforms/InstCombine/InstCombineInternal.h
+++ b/llvm/lib/Transforms/InstCombine/InstCombineInternal.h
@@ -354,8 +354,9 @@ private:
   }
 
   bool willNotOverflowUnsignedMul(const Value *LHS, const Value *RHS,
-                                  const Instruction &CxtI) const {
-    return computeOverflowForUnsignedMul(LHS, RHS, &CxtI) ==
+                                  const Instruction &CxtI,
+                                  bool IsNSW = false) const {
+    return computeOverflowForUnsignedMul(LHS, RHS, &CxtI, IsNSW) ==
            OverflowResult::NeverOverflows;
   }
 
@@ -376,7 +377,7 @@ private:
     }
   }
 
-  Value *EmitGEPOffset(User *GEP);
+  Value *EmitGEPOffset(GEPOperator *GEP, bool RewriteGEP = false);
   Instruction *scalarizePHI(ExtractElementInst &EI, PHINode *PN);
   Instruction *foldBitcastExtElt(ExtractElementInst &ExtElt);
   Instruction *foldCastedBitwiseLogic(BinaryOperator &I);
diff --git a/llvm/lib/Transforms/InstCombine/InstCombineMulDivRem.cpp b/llvm/lib/Transforms/InstCombine/InstCombineMulDivRem.cpp
index 4ed4c36e21e0..ca1b1921404d 100644
--- a/llvm/lib/Transforms/InstCombine/InstCombineMulDivRem.cpp
+++ b/llvm/lib/Transforms/InstCombine/InstCombineMulDivRem.cpp
@@ -530,7 +530,7 @@ Instruction *InstCombinerImpl::visitMul(BinaryOperator &I) {
     I.setHasNoSignedWrap(true);
   }
 
-  if (!HasNUW && willNotOverflowUnsignedMul(Op0, Op1, I)) {
+  if (!HasNUW && willNotOverflowUnsignedMul(Op0, Op1, I, I.hasNoSignedWrap())) {
     Changed = true;
     I.setHasNoUnsignedWrap(true);
   }
diff --git a/llvm/lib/Transforms/InstCombine/InstCombineNegator.cpp b/llvm/lib/Transforms/InstCombine/InstCombineNegator.cpp
index d697f361dec0..ed2a98ba4ae4 100644
--- a/llvm/lib/Transforms/InstCombine/InstCombineNegator.cpp
+++ b/llvm/lib/Transforms/InstCombine/InstCombineNegator.cpp
@@ -320,7 +320,8 @@ std::array<Value *, 2> Negator::getSortedOperandsOfBinOp(Instruction *I) {
     return NegatedPHI;
   }
   case Instruction::Select: {
-    if (isKnownNegation(I->getOperand(1), I->getOperand(2))) {
+    if (isKnownNegation(I->getOperand(1), I->getOperand(2), /*NeedNSW=*/false,
+                        /*AllowPoison=*/false)) {
       // Of one hand of select is known to be negation of another hand,
       // just swap the hands around.
       auto *NewSelect = cast<SelectInst>(I->clone());
diff --git a/llvm/lib/Transforms/InstCombine/InstCombineSelect.cpp b/llvm/lib/Transforms/InstCombine/InstCombineSelect.cpp
index 73600206a55c..8818369e7945 100644
--- a/llvm/lib/Transforms/InstCombine/InstCombineSelect.cpp
+++ b/llvm/lib/Transforms/InstCombine/InstCombineSelect.cpp
@@ -1722,11 +1722,11 @@ static Instruction *foldSelectICmpEq(SelectInst &SI, ICmpInst *ICI,
       return match(CmpRHS, m_Zero()) && match(FalseVal, matchInner);
 
     if (NotMask == NotInner) {
-      return match(FalseVal,
-                   m_c_BinOp(OuterOpc, m_Not(matchInner), m_Specific(CmpRHS)));
+      return match(FalseVal, m_c_BinOp(OuterOpc, m_NotForbidPoison(matchInner),
+                                       m_Specific(CmpRHS)));
     } else if (NotMask == NotRHS) {
-      return match(FalseVal,
-                   m_c_BinOp(OuterOpc, matchInner, m_Not(m_Specific(CmpRHS))));
+      return match(FalseVal, m_c_BinOp(OuterOpc, matchInner,
+                                       m_NotForbidPoison(m_Specific(CmpRHS))));
     } else {
       return match(FalseVal,
                    m_c_BinOp(OuterOpc, matchInner, m_Specific(CmpRHS)));
@@ -2537,8 +2537,8 @@ Instruction *InstCombinerImpl::foldVectorSelect(SelectInst &Sel) {
       if (auto *I = dyn_cast<Instruction>(V))
         I->copyIRFlags(&Sel);
       Module *M = Sel.getModule();
-      Function *F = Intrinsic::getDeclaration(
-          M, Intrinsic::experimental_vector_reverse, V->getType());
+      Function *F =
+          Intrinsic::getDeclaration(M, Intrinsic::vector_reverse, V->getType());
       return CallInst::Create(F, V);
     };
 
diff --git a/llvm/lib/Transforms/InstCombine/InstCombineShifts.cpp b/llvm/lib/Transforms/InstCombine/InstCombineShifts.cpp
index 244f03a1bc2b..1cb21a1d81af 100644
--- a/llvm/lib/Transforms/InstCombine/InstCombineShifts.cpp
+++ b/llvm/lib/Transforms/InstCombine/InstCombineShifts.cpp
@@ -1120,14 +1120,6 @@ Instruction *InstCombinerImpl::visitShl(BinaryOperator &I) {
       return BinaryOperator::CreateAnd(Trunc, ConstantInt::get(Ty, Mask));
     }
 
-    if (match(Op0, m_Shl(m_Value(X), m_APInt(C1))) && C1->ult(BitWidth)) {
-      unsigned AmtSum = ShAmtC + C1->getZExtValue();
-      // Oversized shifts are simplified to zero in InstSimplify.
-      if (AmtSum < BitWidth)
-        // (X << C1) << C2 --> X << (C1 + C2)
-        return BinaryOperator::CreateShl(X, ConstantInt::get(Ty, AmtSum));
-    }
-
     // If we have an opposite shift by the same amount, we may be able to
     // reorder binops and shifts to eliminate math/logic.
     auto isSuitableBinOpcode = [](Instruction::BinaryOps BinOpcode) {
@@ -1394,14 +1386,6 @@ Instruction *InstCombinerImpl::visitLShr(BinaryOperator &I) {
       }
     }
 
-    // (X >>u C1) >>u C --> X >>u (C1 + C)
-    if (match(Op0, m_LShr(m_Value(X), m_APInt(C1)))) {
-      // Oversized shifts are simplified to zero in InstSimplify.
-      unsigned AmtSum = ShAmtC + C1->getZExtValue();
-      if (AmtSum < BitWidth)
-        return BinaryOperator::CreateLShr(X, ConstantInt::get(Ty, AmtSum));
-    }
-
     Instruction *TruncSrc;
     if (match(Op0, m_OneUse(m_Trunc(m_Instruction(TruncSrc)))) &&
         match(TruncSrc, m_LShr(m_Value(X), m_APInt(C1)))) {
diff --git a/llvm/lib/Transforms/InstCombine/InstructionCombining.cpp b/llvm/lib/Transforms/InstCombine/InstructionCombining.cpp
index 3ac5c2559ddf..7356941be645 100644
--- a/llvm/lib/Transforms/InstCombine/InstructionCombining.cpp
+++ b/llvm/lib/Transforms/InstCombine/InstructionCombining.cpp
@@ -190,8 +190,26 @@ bool InstCombiner::isValidAddrSpaceCast(unsigned FromAS, unsigned ToAS) const {
   return TTI.isValidAddrSpaceCast(FromAS, ToAS);
 }
 
-Value *InstCombinerImpl::EmitGEPOffset(User *GEP) {
-  return llvm::emitGEPOffset(&Builder, DL, GEP);
+Value *InstCombinerImpl::EmitGEPOffset(GEPOperator *GEP, bool RewriteGEP) {
+  if (!RewriteGEP)
+    return llvm::emitGEPOffset(&Builder, DL, GEP);
+
+  IRBuilderBase::InsertPointGuard Guard(Builder);
+  auto *Inst = dyn_cast<Instruction>(GEP);
+  if (Inst)
+    Builder.SetInsertPoint(Inst);
+
+  Value *Offset = EmitGEPOffset(GEP);
+  // If a non-trivial GEP has other uses, rewrite it to avoid duplicating
+  // the offset arithmetic.
+  if (Inst && !GEP->hasOneUse() && !GEP->hasAllConstantIndices() &&
+      !GEP->getSourceElementType()->isIntegerTy(8)) {
+    replaceInstUsesWith(
+        *Inst, Builder.CreateGEP(Builder.getInt8Ty(), GEP->getPointerOperand(),
+                                 Offset, "", GEP->isInBounds()));
+    eraseInstFromFunction(*Inst);
+  }
+  return Offset;
 }
 
 /// Legal integers and common types are considered desirable. This is used to
@@ -2025,8 +2043,8 @@ Instruction *InstCombinerImpl::foldVectorBinop(BinaryOperator &Inst) {
     if (auto *BO = dyn_cast<BinaryOperator>(V))
       BO->copyIRFlags(&Inst);
     Module *M = Inst.getModule();
-    Function *F = Intrinsic::getDeclaration(
-        M, Intrinsic::experimental_vector_reverse, V->getType());
+    Function *F =
+        Intrinsic::getDeclaration(M, Intrinsic::vector_reverse, V->getType());
     return CallInst::Create(F, V);
   };
 
@@ -2321,6 +2339,43 @@ static Instruction *foldSelectGEP(GetElementPtrInst &GEP,
   return SelectInst::Create(Cond, NewTrueC, NewFalseC, "", nullptr, Sel);
 }
 
+// Canonicalization:
+// gep T, (gep i8, base, C1), (Index + C2) into
+// gep T, (gep i8, base, C1 + C2 * sizeof(T)), Index
+static Instruction *canonicalizeGEPOfConstGEPI8(GetElementPtrInst &GEP,
+                                                GEPOperator *Src,
+                                                InstCombinerImpl &IC) {
+  if (GEP.getNumIndices() != 1)
+    return nullptr;
+  auto &DL = IC.getDataLayout();
+  Value *Base;
+  const APInt *C1;
+  if (!match(Src, m_PtrAdd(m_Value(Base), m_APInt(C1))))
+    return nullptr;
+  Value *VarIndex;
+  const APInt *C2;
+  Type *PtrTy = Src->getType()->getScalarType();
+  unsigned IndexSizeInBits = DL.getIndexTypeSizeInBits(PtrTy);
+  if (!match(GEP.getOperand(1), m_AddLike(m_Value(VarIndex), m_APInt(C2))))
+    return nullptr;
+  if (C1->getBitWidth() != IndexSizeInBits ||
+      C2->getBitWidth() != IndexSizeInBits)
+    return nullptr;
+  Type *BaseType = GEP.getSourceElementType();
+  if (isa<ScalableVectorType>(BaseType))
+    return nullptr;
+  APInt TypeSize(IndexSizeInBits, DL.getTypeAllocSize(BaseType));
+  APInt NewOffset = TypeSize * *C2 + *C1;
+  if (NewOffset.isZero() ||
+      (Src->hasOneUse() && GEP.getOperand(1)->hasOneUse())) {
+    Value *GEPConst =
+        IC.Builder.CreatePtrAdd(Base, IC.Builder.getInt(NewOffset));
+    return GetElementPtrInst::Create(BaseType, GEPConst, VarIndex);
+  }
+
+  return nullptr;
+}
+
 Instruction *InstCombinerImpl::visitGEPOfGEP(GetElementPtrInst &GEP,
                                              GEPOperator *Src) {
   // Combine Indices - If the source pointer to this getelementptr instruction
@@ -2329,6 +2384,9 @@ Instruction *InstCombinerImpl::visitGEPOfGEP(GetElementPtrInst &GEP,
   if (!shouldMergeGEPs(*cast<GEPOperator>(&GEP), *Src))
     return nullptr;
 
+  if (auto *I = canonicalizeGEPOfConstGEPI8(GEP, Src, *this))
+    return I;
+
   // For constant GEPs, use a more general offset-based folding approach.
   Type *PtrTy = Src->getType()->getScalarType();
   if (GEP.hasAllConstantIndices() &&
@@ -2890,6 +2948,14 @@ Instruction *InstCombinerImpl::visitGetElementPtrInst(GetElementPtrInst &GEP) {
     return nullptr;
 
   if (GEP.getNumIndices() == 1) {
+    // We can only preserve inbounds if the original gep is inbounds, the add
+    // is nsw, and the add operands are non-negative.
+    auto CanPreserveInBounds = [&](bool AddIsNSW, Value *Idx1, Value *Idx2) {
+      SimplifyQuery Q = SQ.getWithInstruction(&GEP);
+      return GEP.isInBounds() && AddIsNSW && isKnownNonNegative(Idx1, Q) &&
+             isKnownNonNegative(Idx2, Q);
+    };
+
     // Try to replace ADD + GEP with GEP + GEP.
     Value *Idx1, *Idx2;
     if (match(GEP.getOperand(1),
@@ -2899,10 +2965,15 @@ Instruction *InstCombinerImpl::visitGetElementPtrInst(GetElementPtrInst &GEP) {
       // as:
       //   %newptr = getelementptr i32, ptr %ptr, i64 %idx1
       //   %newgep = getelementptr i32, ptr %newptr, i64 %idx2
-      auto *NewPtr = Builder.CreateGEP(GEP.getResultElementType(),
-                                       GEP.getPointerOperand(), Idx1);
-      return GetElementPtrInst::Create(GEP.getResultElementType(), NewPtr,
-                                       Idx2);
+      bool IsInBounds = CanPreserveInBounds(
+          cast<OverflowingBinaryOperator>(GEP.getOperand(1))->hasNoSignedWrap(),
+          Idx1, Idx2);
+      auto *NewPtr =
+          Builder.CreateGEP(GEP.getResultElementType(), GEP.getPointerOperand(),
+                            Idx1, "", IsInBounds);
+      return replaceInstUsesWith(
+          GEP, Builder.CreateGEP(GEP.getResultElementType(), NewPtr, Idx2, "",
+                                 IsInBounds));
     }
     ConstantInt *C;
     if (match(GEP.getOperand(1), m_OneUse(m_SExtLike(m_OneUse(m_NSWAdd(
@@ -2913,12 +2984,17 @@ Instruction *InstCombinerImpl::visitGetElementPtrInst(GetElementPtrInst &GEP) {
       // as:
       // %newptr = getelementptr i32, ptr %ptr, i32 %idx1
       // %newgep = getelementptr i32, ptr %newptr, i32 idx2
+      bool IsInBounds = CanPreserveInBounds(
+          /*IsNSW=*/true, Idx1, C);
       auto *NewPtr = Builder.CreateGEP(
           GEP.getResultElementType(), GEP.getPointerOperand(),
-          Builder.CreateSExt(Idx1, GEP.getOperand(1)->getType()));
-      return GetElementPtrInst::Create(
-          GEP.getResultElementType(), NewPtr,
-          Builder.CreateSExt(C, GEP.getOperand(1)->getType()));
+          Builder.CreateSExt(Idx1, GEP.getOperand(1)->getType()), "",
+          IsInBounds);
+      return replaceInstUsesWith(
+          GEP,
+          Builder.CreateGEP(GEP.getResultElementType(), NewPtr,
+                            Builder.CreateSExt(C, GEP.getOperand(1)->getType()),
+                            "", IsInBounds));
     }
   }
 
@@ -3972,6 +4048,7 @@ static bool isCatchAll(EHPersonality Personality, Constant *TypeInfo) {
   case EHPersonality::CoreCLR:
   case EHPersonality::Wasm_CXX:
   case EHPersonality::XL_CXX:
+  case EHPersonality::ZOS_CXX:
     return TypeInfo->isNullValue();
   }
   llvm_unreachable("invalid enum");
diff --git a/llvm/lib/Transforms/Instrumentation/HWAddressSanitizer.cpp b/llvm/lib/Transforms/Instrumentation/HWAddressSanitizer.cpp
index a35f24447cc3..fa661b17c13a 100644
--- a/llvm/lib/Transforms/Instrumentation/HWAddressSanitizer.cpp
+++ b/llvm/lib/Transforms/Instrumentation/HWAddressSanitizer.cpp
@@ -930,11 +930,33 @@ void HWAddressSanitizer::instrumentMemAccessOutline(Value *Ptr, bool IsWrite,
 
   IRBuilder<> IRB(InsertBefore);
   Module *M = IRB.GetInsertBlock()->getParent()->getParent();
-  IRB.CreateCall(Intrinsic::getDeclaration(
-                     M, UseShortGranules
-                            ? Intrinsic::hwasan_check_memaccess_shortgranules
-                            : Intrinsic::hwasan_check_memaccess),
-                 {ShadowBase, Ptr, ConstantInt::get(Int32Ty, AccessInfo)});
+  bool useFixedShadowIntrinsic = false;
+  // The memaccess fixed shadow intrinsic is only supported on AArch64,
+  // which allows a 16-bit immediate to be left-shifted by 32.
+  // Since kShadowBaseAlignment == 32, and Linux by default will not
+  // mmap above 48-bits, practically any valid shadow offset is
+  // representable.
+  // In particular, an offset of 4TB (1024 << 32) is representable, and
+  // ought to be good enough for anybody.
+  if (TargetTriple.isAArch64() && Mapping.Offset != kDynamicShadowSentinel) {
+    uint16_t offset_shifted = Mapping.Offset >> 32;
+    useFixedShadowIntrinsic = (uint64_t)offset_shifted << 32 == Mapping.Offset;
+  }
+
+  if (useFixedShadowIntrinsic)
+    IRB.CreateCall(
+        Intrinsic::getDeclaration(
+            M, UseShortGranules
+                   ? Intrinsic::hwasan_check_memaccess_shortgranules_fixedshadow
+                   : Intrinsic::hwasan_check_memaccess_fixedshadow),
+        {Ptr, ConstantInt::get(Int32Ty, AccessInfo),
+         ConstantInt::get(Int64Ty, Mapping.Offset)});
+  else
+    IRB.CreateCall(Intrinsic::getDeclaration(
+                       M, UseShortGranules
+                              ? Intrinsic::hwasan_check_memaccess_shortgranules
+                              : Intrinsic::hwasan_check_memaccess),
+                   {ShadowBase, Ptr, ConstantInt::get(Int32Ty, AccessInfo)});
 }
 
 void HWAddressSanitizer::instrumentMemAccessInline(Value *Ptr, bool IsWrite,
@@ -1363,14 +1385,6 @@ bool HWAddressSanitizer::instrumentLandingPads(
   return true;
 }
 
-static DbgAssignIntrinsic *DynCastToDbgAssign(DbgVariableIntrinsic *DVI) {
-  return dyn_cast<DbgAssignIntrinsic>(DVI);
-}
-
-static DbgVariableRecord *DynCastToDbgAssign(DbgVariableRecord *DVR) {
-  return DVR->isDbgAssign() ? DVR : nullptr;
-}
-
 bool HWAddressSanitizer::instrumentStack(memtag::StackInfo &SInfo,
                                          Value *StackTag, Value *UARTag,
                                          const DominatorTree &DT,
@@ -1426,28 +1440,7 @@ bool HWAddressSanitizer::instrumentStack(memtag::StackInfo &SInfo,
              !memtag::isLifetimeIntrinsic(User);
     });
 
-    // Helper utility for adding DW_OP_LLVM_tag_offset to debug-info records,
-    // abstracted over whether they're intrinsic-stored or DbgVariableRecord
-    // stored.
-    auto AnnotateDbgRecord = [&](auto *DPtr) {
-      // Prepend "tag_offset, N" to the dwarf expression.
-      // Tag offset logically applies to the alloca pointer, and it makes sense
-      // to put it at the beginning of the expression.
-      SmallVector<uint64_t, 8> NewOps = {dwarf::DW_OP_LLVM_tag_offset,
-                                         retagMask(N)};
-      for (size_t LocNo = 0; LocNo < DPtr->getNumVariableLocationOps(); ++LocNo)
-        if (DPtr->getVariableLocationOp(LocNo) == AI)
-          DPtr->setExpression(DIExpression::appendOpsToArg(
-              DPtr->getExpression(), NewOps, LocNo));
-      if (auto *DAI = DynCastToDbgAssign(DPtr)) {
-        if (DAI->getAddress() == AI)
-          DAI->setAddressExpression(DIExpression::prependOpcodes(
-              DAI->getAddressExpression(), NewOps));
-      }
-    };
-
-    llvm::for_each(Info.DbgVariableIntrinsics, AnnotateDbgRecord);
-    llvm::for_each(Info.DbgVariableRecords, AnnotateDbgRecord);
+    memtag::annotateDebugRecords(Info, retagMask(N));
 
     auto TagEnd = [&](Instruction *Node) {
       IRB.SetInsertPoint(Node);
diff --git a/llvm/lib/Transforms/Instrumentation/MemorySanitizer.cpp b/llvm/lib/Transforms/Instrumentation/MemorySanitizer.cpp
index b1fd64fa3b29..cc2295c44023 100644
--- a/llvm/lib/Transforms/Instrumentation/MemorySanitizer.cpp
+++ b/llvm/lib/Transforms/Instrumentation/MemorySanitizer.cpp
@@ -152,6 +152,7 @@
 #include "llvm/ADT/DenseMap.h"
 #include "llvm/ADT/DepthFirstIterator.h"
 #include "llvm/ADT/SetVector.h"
+#include "llvm/ADT/SmallPtrSet.h"
 #include "llvm/ADT/SmallVector.h"
 #include "llvm/ADT/StringExtras.h"
 #include "llvm/ADT/StringRef.h"
@@ -212,6 +213,9 @@ using namespace llvm;
 DEBUG_COUNTER(DebugInsertCheck, "msan-insert-check",
               "Controls which checks to insert");
 
+DEBUG_COUNTER(DebugInstrumentInstruction, "msan-instrument-instruction",
+              "Controls which instruction to instrument");
+
 static const unsigned kOriginSize = 4;
 static const Align kMinOriginAlignment = Align(4);
 static const Align kShadowTLSAlignment = Align(8);
@@ -1336,7 +1340,7 @@ struct MemorySanitizerVisitor : public InstVisitor<MemorySanitizerVisitor> {
     }
   }
 
-  // Returns true if Debug Location curresponds to multiple warnings.
+  // Returns true if Debug Location corresponds to multiple warnings.
   bool shouldDisambiguateWarningLocation(const DebugLoc &DebugLoc) {
     if (MS.TrackOrigins < 2)
       return false;
@@ -1461,19 +1465,21 @@ struct MemorySanitizerVisitor : public InstVisitor<MemorySanitizerVisitor> {
   }
 
   void materializeChecks() {
-    llvm::stable_sort(InstrumentationList,
-                      [](const ShadowOriginAndInsertPoint &L,
-                         const ShadowOriginAndInsertPoint &R) {
-                        return L.OrigIns < R.OrigIns;
-                      });
+#ifndef NDEBUG
+    // For assert below.
+    SmallPtrSet<Instruction *, 16> Done;
+#endif
 
     for (auto I = InstrumentationList.begin();
          I != InstrumentationList.end();) {
-      auto J =
-          std::find_if(I + 1, InstrumentationList.end(),
-                       [L = I->OrigIns](const ShadowOriginAndInsertPoint &R) {
-                         return L != R.OrigIns;
-                       });
+      auto OrigIns = I->OrigIns;
+      // Checks are grouped by the original instruction. We call all
+      // `insertShadowCheck` for an instruction at once.
+      assert(Done.insert(OrigIns).second);
+      auto J = std::find_if(I + 1, InstrumentationList.end(),
+                            [OrigIns](const ShadowOriginAndInsertPoint &R) {
+                              return OrigIns != R.OrigIns;
+                            });
       // Process all checks of instruction at once.
       materializeInstructionChecks(ArrayRef<ShadowOriginAndInsertPoint>(I, J));
       I = J;
@@ -1949,8 +1955,15 @@ struct MemorySanitizerVisitor : public InstVisitor<MemorySanitizerVisitor> {
       unsigned ArgOffset = 0;
       const DataLayout &DL = F->getParent()->getDataLayout();
       for (auto &FArg : F->args()) {
-        if (!FArg.getType()->isSized()) {
-          LLVM_DEBUG(dbgs() << "Arg is not sized\n");
+        if (!FArg.getType()->isSized() || FArg.getType()->isScalableTy()) {
+          LLVM_DEBUG(dbgs() << (FArg.getType()->isScalableTy()
+                                    ? "vscale not fully supported\n"
+                                    : "Arg is not sized\n"));
+          if (A == &FArg) {
+            ShadowPtr = getCleanShadow(V);
+            setOrigin(A, getCleanOrigin());
+            break;
+          }
           continue;
         }
 
@@ -2176,6 +2189,13 @@ struct MemorySanitizerVisitor : public InstVisitor<MemorySanitizerVisitor> {
     // Don't want to visit if we're in the prologue
     if (isInPrologue(I))
       return;
+    if (!DebugCounter::shouldExecute(DebugInstrumentInstruction)) {
+      LLVM_DEBUG(dbgs() << "Skipping instruction: " << I << "\n");
+      // We still need to set the shadow and origin to clean values.
+      setShadow(&I, getCleanShadow(&I));
+      setOrigin(&I, getCleanOrigin());
+      return;
+    }
     InstVisitor<MemorySanitizerVisitor>::visit(I);
   }
 
@@ -2493,6 +2513,8 @@ struct MemorySanitizerVisitor : public InstVisitor<MemorySanitizerVisitor> {
   Value *CreateShadowCast(IRBuilder<> &IRB, Value *V, Type *dstTy,
                           bool Signed = false) {
     Type *srcTy = V->getType();
+    if (srcTy == dstTy)
+      return V;
     size_t srcSizeInBits = VectorOrPrimitiveTypeSizeInBits(srcTy);
     size_t dstSizeInBits = VectorOrPrimitiveTypeSizeInBits(dstTy);
     if (srcSizeInBits > 1 && dstSizeInBits == 1)
@@ -4183,6 +4205,14 @@ struct MemorySanitizerVisitor : public InstVisitor<MemorySanitizerVisitor> {
         LLVM_DEBUG(dbgs() << "Arg " << i << " is not sized: " << CB << "\n");
         continue;
       }
+
+      if (A->getType()->isScalableTy()) {
+        LLVM_DEBUG(dbgs() << "Arg  " << i << " is vscale: " << CB << "\n");
+        // Handle as noundef, but don't reserve tls slots.
+        insertShadowCheck(A, &CB);
+        continue;
+      }
+
       unsigned Size = 0;
       const DataLayout &DL = F.getParent()->getDataLayout();
 
diff --git a/llvm/lib/Transforms/Instrumentation/PGOInstrumentation.cpp b/llvm/lib/Transforms/Instrumentation/PGOInstrumentation.cpp
index 98c6f8cbf5af..a7b7556685e4 100644
--- a/llvm/lib/Transforms/Instrumentation/PGOInstrumentation.cpp
+++ b/llvm/lib/Transforms/Instrumentation/PGOInstrumentation.cpp
@@ -1367,6 +1367,7 @@ void PGOUseFunc::populateCoverage(IndexedInstrProfReader *PGOReader) {
     handleInstrProfError(std::move(Err), MismatchedFuncSum);
     return;
   }
+  IsCS ? NumOfCSPGOFunc++ : NumOfPGOFunc++;
 
   std::vector<uint64_t> &CountsFromProfile = Result.get().Counts;
   DenseMap<const BasicBlock *, bool> Coverage;
diff --git a/llvm/lib/Transforms/Scalar/DFAJumpThreading.cpp b/llvm/lib/Transforms/Scalar/DFAJumpThreading.cpp
index 1caed93b1b66..ba2546b8db0e 100644
--- a/llvm/lib/Transforms/Scalar/DFAJumpThreading.cpp
+++ b/llvm/lib/Transforms/Scalar/DFAJumpThreading.cpp
@@ -131,7 +131,7 @@ public:
   explicit operator bool() const { return SI && SIUse; }
 };
 
-void unfold(DomTreeUpdater *DTU, SelectInstToUnfold SIToUnfold,
+void unfold(DomTreeUpdater *DTU, LoopInfo *LI, SelectInstToUnfold SIToUnfold,
             std::vector<SelectInstToUnfold> *NewSIsToUnfold,
             std::vector<BasicBlock *> *NewBBs);
 
@@ -142,6 +142,7 @@ public:
       : AC(AC), DT(DT), LI(LI), TTI(TTI), ORE(ORE) {}
 
   bool run(Function &F);
+  bool LoopInfoBroken;
 
 private:
   void
@@ -157,7 +158,7 @@ private:
 
       std::vector<SelectInstToUnfold> NewSIsToUnfold;
       std::vector<BasicBlock *> NewBBs;
-      unfold(&DTU, SIToUnfold, &NewSIsToUnfold, &NewBBs);
+      unfold(&DTU, LI, SIToUnfold, &NewSIsToUnfold, &NewBBs);
 
       // Put newly discovered select instructions into the work list.
       for (const SelectInstToUnfold &NewSIToUnfold : NewSIsToUnfold)
@@ -201,7 +202,7 @@ void createBasicBlockAndSinkSelectInst(
 /// created basic blocks into \p NewBBs.
 ///
 /// TODO: merge it with CodeGenPrepare::optimizeSelectInst() if possible.
-void unfold(DomTreeUpdater *DTU, SelectInstToUnfold SIToUnfold,
+void unfold(DomTreeUpdater *DTU, LoopInfo *LI, SelectInstToUnfold SIToUnfold,
             std::vector<SelectInstToUnfold> *NewSIsToUnfold,
             std::vector<BasicBlock *> *NewBBs) {
   SelectInst *SI = SIToUnfold.getInst();
@@ -307,6 +308,12 @@ void unfold(DomTreeUpdater *DTU, SelectInstToUnfold SIToUnfold,
   DTU->applyUpdates({{DominatorTree::Insert, StartBlock, TT},
                      {DominatorTree::Insert, StartBlock, FT}});
 
+  // Preserve loop info
+  if (Loop *L = LI->getLoopFor(SI->getParent())) {
+    for (BasicBlock *NewBB : *NewBBs)
+      L->addBasicBlockToLoop(NewBB, *LI);
+  }
+
   // The select is now dead.
   assert(SI->use_empty() && "Select must be dead now");
   SI->eraseFromParent();
@@ -522,9 +529,10 @@ private:
 };
 
 struct AllSwitchPaths {
-  AllSwitchPaths(const MainSwitch *MSwitch, OptimizationRemarkEmitter *ORE)
-      : Switch(MSwitch->getInstr()), SwitchBlock(Switch->getParent()),
-        ORE(ORE) {}
+  AllSwitchPaths(const MainSwitch *MSwitch, OptimizationRemarkEmitter *ORE,
+                 LoopInfo *LI)
+      : Switch(MSwitch->getInstr()), SwitchBlock(Switch->getParent()), ORE(ORE),
+        LI(LI) {}
 
   std::vector<ThreadingPath> &getThreadingPaths() { return TPaths; }
   unsigned getNumThreadingPaths() { return TPaths.size(); }
@@ -545,7 +553,7 @@ struct AllSwitchPaths {
       return;
     }
 
-    for (PathType Path : LoopPaths) {
+    for (const PathType &Path : LoopPaths) {
       ThreadingPath TPath;
 
       const BasicBlock *PrevBB = Path.back();
@@ -596,6 +604,12 @@ private:
 
     Visited.insert(BB);
 
+    // Stop if we have reached the BB out of loop, since its successors have no
+    // impact on the DFA.
+    // TODO: Do we need to stop exploring if BB is the outer loop of the switch?
+    if (!LI->getLoopFor(BB))
+      return Res;
+
     // Some blocks have multiple edges to the same successor, and this set
     // is used to prevent a duplicate path from being generated
     SmallSet<BasicBlock *, 4> Successors;
@@ -737,6 +751,7 @@ private:
   BasicBlock *SwitchBlock;
   OptimizationRemarkEmitter *ORE;
   std::vector<ThreadingPath> TPaths;
+  LoopInfo *LI;
 };
 
 struct TransformDFA {
@@ -1283,6 +1298,7 @@ bool DFAJumpThreading::run(Function &F) {
 
   SmallVector<AllSwitchPaths, 2> ThreadableLoops;
   bool MadeChanges = false;
+  LoopInfoBroken = false;
 
   for (BasicBlock &BB : F) {
     auto *SI = dyn_cast<SwitchInst>(BB.getTerminator());
@@ -1304,7 +1320,7 @@ bool DFAJumpThreading::run(Function &F) {
     if (!Switch.getSelectInsts().empty())
       MadeChanges = true;
 
-    AllSwitchPaths SwitchPaths(&Switch, ORE);
+    AllSwitchPaths SwitchPaths(&Switch, ORE, LI);
     SwitchPaths.run();
 
     if (SwitchPaths.getNumThreadingPaths() > 0) {
@@ -1315,10 +1331,15 @@ bool DFAJumpThreading::run(Function &F) {
       // strict requirement but it can cause buggy behavior if there is an
       // overlap of blocks in different opportunities. There is a lot of room to
       // experiment with catching more opportunities here.
+      // NOTE: To release this contraint, we must handle LoopInfo invalidation
       break;
     }
   }
 
+#ifdef NDEBUG
+  LI->verify(*DT);
+#endif
+
   SmallPtrSet<const Value *, 32> EphValues;
   if (ThreadableLoops.size() > 0)
     CodeMetrics::collectEphemeralValues(&F, AC, EphValues);
@@ -1327,6 +1348,7 @@ bool DFAJumpThreading::run(Function &F) {
     TransformDFA Transform(&SwitchPaths, DT, AC, TTI, ORE, EphValues);
     Transform.run();
     MadeChanges = true;
+    LoopInfoBroken = true;
   }
 
 #ifdef EXPENSIVE_CHECKS
@@ -1347,11 +1369,13 @@ PreservedAnalyses DFAJumpThreadingPass::run(Function &F,
   LoopInfo &LI = AM.getResult<LoopAnalysis>(F);
   TargetTransformInfo &TTI = AM.getResult<TargetIRAnalysis>(F);
   OptimizationRemarkEmitter ORE(&F);
-
-  if (!DFAJumpThreading(&AC, &DT, &LI, &TTI, &ORE).run(F))
+  DFAJumpThreading ThreadImpl(&AC, &DT, &LI, &TTI, &ORE);
+  if (!ThreadImpl.run(F))
     return PreservedAnalyses::all();
 
   PreservedAnalyses PA;
   PA.preserve<DominatorTreeAnalysis>();
+  if (!ThreadImpl.LoopInfoBroken)
+    PA.preserve<LoopAnalysis>();
   return PA;
 }
diff --git a/llvm/lib/Transforms/Scalar/GVN.cpp b/llvm/lib/Transforms/Scalar/GVN.cpp
index 86d5c9909f3d..b5be8ac24941 100644
--- a/llvm/lib/Transforms/Scalar/GVN.cpp
+++ b/llvm/lib/Transforms/Scalar/GVN.cpp
@@ -33,6 +33,7 @@
 #include "llvm/Analysis/GlobalsModRef.h"
 #include "llvm/Analysis/InstructionPrecedenceTracking.h"
 #include "llvm/Analysis/InstructionSimplify.h"
+#include "llvm/Analysis/Loads.h"
 #include "llvm/Analysis/LoopInfo.h"
 #include "llvm/Analysis/MemoryBuiltins.h"
 #include "llvm/Analysis/MemoryDependenceAnalysis.h"
@@ -725,6 +726,69 @@ void GVNPass::ValueTable::verifyRemoved(const Value *V) const {
 }
 
 //===----------------------------------------------------------------------===//
+//                     LeaderMap External Functions
+//===----------------------------------------------------------------------===//
+
+/// Push a new Value to the LeaderTable onto the list for its value number.
+void GVNPass::LeaderMap::insert(uint32_t N, Value *V, const BasicBlock *BB) {
+  LeaderListNode &Curr = NumToLeaders[N];
+  if (!Curr.Entry.Val) {
+    Curr.Entry.Val = V;
+    Curr.Entry.BB = BB;
+    return;
+  }
+
+  LeaderListNode *Node = TableAllocator.Allocate<LeaderListNode>();
+  Node->Entry.Val = V;
+  Node->Entry.BB = BB;
+  Node->Next = Curr.Next;
+  Curr.Next = Node;
+}
+
+/// Scan the list of values corresponding to a given
+/// value number, and remove the given instruction if encountered.
+void GVNPass::LeaderMap::erase(uint32_t N, Instruction *I,
+                               const BasicBlock *BB) {
+  LeaderListNode *Prev = nullptr;
+  LeaderListNode *Curr = &NumToLeaders[N];
+
+  while (Curr && (Curr->Entry.Val != I || Curr->Entry.BB != BB)) {
+    Prev = Curr;
+    Curr = Curr->Next;
+  }
+
+  if (!Curr)
+    return;
+
+  if (Prev) {
+    Prev->Next = Curr->Next;
+  } else {
+    if (!Curr->Next) {
+      Curr->Entry.Val = nullptr;
+      Curr->Entry.BB = nullptr;
+    } else {
+      LeaderListNode *Next = Curr->Next;
+      Curr->Entry.Val = Next->Entry.Val;
+      Curr->Entry.BB = Next->Entry.BB;
+      Curr->Next = Next->Next;
+    }
+  }
+}
+
+void GVNPass::LeaderMap::verifyRemoved(const Value *V) const {
+  // Walk through the value number scope to make sure the instruction isn't
+  // ferreted away in it.
+  for (const auto &I : NumToLeaders) {
+    (void)I;
+    assert(I.second.Entry.Val != V && "Inst still in value numbering scope!");
+    assert(
+        std::none_of(leader_iterator(&I.second), leader_iterator(nullptr),
+                     [=](const LeaderTableEntry &E) { return E.Val == V; }) &&
+        "Inst still in value numbering scope!");
+  }
+}
+
+//===----------------------------------------------------------------------===//
 //                                GVN Pass
 //===----------------------------------------------------------------------===//
 
@@ -1466,7 +1530,7 @@ void GVNPass::eliminatePartiallyRedundantLoad(
         OldLoad->replaceAllUsesWith(NewLoad);
         replaceValuesPerBlockEntry(ValuesPerBlock, OldLoad, NewLoad);
         if (uint32_t ValNo = VN.lookup(OldLoad, false))
-          removeFromLeaderTable(ValNo, OldLoad, OldLoad->getParent());
+          LeaderTable.erase(ValNo, OldLoad, OldLoad->getParent());
         VN.erase(OldLoad);
         removeInstruction(OldLoad);
       }
@@ -2203,10 +2267,9 @@ GVNPass::ValueTable::assignExpNewValueNum(Expression &Exp) {
 /// defined in \p BB.
 bool GVNPass::ValueTable::areAllValsInBB(uint32_t Num, const BasicBlock *BB,
                                          GVNPass &Gvn) {
-  LeaderTableEntry *Vals = &Gvn.LeaderTable[Num];
-  while (Vals && Vals->BB == BB)
-    Vals = Vals->Next;
-  return !Vals;
+  return all_of(
+      Gvn.LeaderTable.getLeaders(Num),
+      [=](const LeaderMap::LeaderTableEntry &L) { return L.BB == BB; });
 }
 
 /// Wrap phiTranslateImpl to provide caching functionality.
@@ -2228,12 +2291,11 @@ bool GVNPass::ValueTable::areCallValsEqual(uint32_t Num, uint32_t NewNum,
                                            const BasicBlock *PhiBlock,
                                            GVNPass &Gvn) {
   CallInst *Call = nullptr;
-  LeaderTableEntry *Vals = &Gvn.LeaderTable[Num];
-  while (Vals) {
-    Call = dyn_cast<CallInst>(Vals->Val);
+  auto Leaders = Gvn.LeaderTable.getLeaders(Num);
+  for (const auto &Entry : Leaders) {
+    Call = dyn_cast<CallInst>(Entry.Val);
     if (Call && Call->getParent() == PhiBlock)
       break;
-    Vals = Vals->Next;
   }
 
   if (AA->doesNotAccessMemory(Call))
@@ -2326,23 +2388,17 @@ void GVNPass::ValueTable::eraseTranslateCacheEntry(
 // question.  This is fast because dominator tree queries consist of only
 // a few comparisons of DFS numbers.
 Value *GVNPass::findLeader(const BasicBlock *BB, uint32_t num) {
-  LeaderTableEntry Vals = LeaderTable[num];
-  if (!Vals.Val) return nullptr;
+  auto Leaders = LeaderTable.getLeaders(num);
+  if (Leaders.empty())
+    return nullptr;
 
   Value *Val = nullptr;
-  if (DT->dominates(Vals.BB, BB)) {
-    Val = Vals.Val;
-    if (isa<Constant>(Val)) return Val;
-  }
-
-  LeaderTableEntry* Next = Vals.Next;
-  while (Next) {
-    if (DT->dominates(Next->BB, BB)) {
-      if (isa<Constant>(Next->Val)) return Next->Val;
-      if (!Val) Val = Next->Val;
+  for (const auto &Entry : Leaders) {
+    if (DT->dominates(Entry.BB, BB)) {
+      Val = Entry.Val;
+      if (isa<Constant>(Val))
+        return Val;
     }
-
-    Next = Next->Next;
   }
 
   return Val;
@@ -2419,6 +2475,10 @@ bool GVNPass::propagateEquality(Value *LHS, Value *RHS,
     if (isa<Constant>(LHS) || (isa<Argument>(LHS) && !isa<Constant>(RHS)))
       std::swap(LHS, RHS);
     assert((isa<Argument>(LHS) || isa<Instruction>(LHS)) && "Unexpected value!");
+    const DataLayout &DL =
+        isa<Argument>(LHS)
+            ? cast<Argument>(LHS)->getParent()->getParent()->getDataLayout()
+            : cast<Instruction>(LHS)->getModule()->getDataLayout();
 
     // If there is no obvious reason to prefer the left-hand side over the
     // right-hand side, ensure the longest lived term is on the right-hand side,
@@ -2445,23 +2505,32 @@ bool GVNPass::propagateEquality(Value *LHS, Value *RHS,
     // using the leader table is about compiling faster, not optimizing better).
     // The leader table only tracks basic blocks, not edges. Only add to if we
     // have the simple case where the edge dominates the end.
-    if (RootDominatesEnd && !isa<Instruction>(RHS))
-      addToLeaderTable(LVN, RHS, Root.getEnd());
+    if (RootDominatesEnd && !isa<Instruction>(RHS) &&
+        canReplacePointersIfEqual(LHS, RHS, DL))
+      LeaderTable.insert(LVN, RHS, Root.getEnd());
 
     // Replace all occurrences of 'LHS' with 'RHS' everywhere in the scope.  As
     // LHS always has at least one use that is not dominated by Root, this will
     // never do anything if LHS has only one use.
     if (!LHS->hasOneUse()) {
+      // Create a callback that captures the DL.
+      auto canReplacePointersCallBack = [&DL](const Use &U, const Value *To) {
+        return canReplacePointersInUseIfEqual(U, To, DL);
+      };
       unsigned NumReplacements =
           DominatesByEdge
-              ? replaceDominatedUsesWith(LHS, RHS, *DT, Root)
-              : replaceDominatedUsesWith(LHS, RHS, *DT, Root.getStart());
-
-      Changed |= NumReplacements > 0;
-      NumGVNEqProp += NumReplacements;
-      // Cached information for anything that uses LHS will be invalid.
-      if (MD)
-        MD->invalidateCachedPointerInfo(LHS);
+              ? replaceDominatedUsesWithIf(LHS, RHS, *DT, Root,
+                                           canReplacePointersCallBack)
+              : replaceDominatedUsesWithIf(LHS, RHS, *DT, Root.getStart(),
+                                           canReplacePointersCallBack);
+
+      if (NumReplacements > 0) {
+        Changed = true;
+        NumGVNEqProp += NumReplacements;
+        // Cached information for anything that uses LHS will be invalid.
+        if (MD)
+          MD->invalidateCachedPointerInfo(LHS);
+      }
     }
 
     // Now try to deduce additional equalities from this one. For example, if
@@ -2532,7 +2601,7 @@ bool GVNPass::propagateEquality(Value *LHS, Value *RHS,
       // The leader table only tracks basic blocks, not edges. Only add to if we
       // have the simple case where the edge dominates the end.
       if (RootDominatesEnd)
-        addToLeaderTable(Num, NotVal, Root.getEnd());
+        LeaderTable.insert(Num, NotVal, Root.getEnd());
 
       continue;
     }
@@ -2582,7 +2651,7 @@ bool GVNPass::processInstruction(Instruction *I) {
       return true;
 
     unsigned Num = VN.lookupOrAdd(Load);
-    addToLeaderTable(Num, Load, Load->getParent());
+    LeaderTable.insert(Num, Load, Load->getParent());
     return false;
   }
 
@@ -2650,7 +2719,7 @@ bool GVNPass::processInstruction(Instruction *I) {
   // Allocations are always uniquely numbered, so we can save time and memory
   // by fast failing them.
   if (isa<AllocaInst>(I) || I->isTerminator() || isa<PHINode>(I)) {
-    addToLeaderTable(Num, I, I->getParent());
+    LeaderTable.insert(Num, I, I->getParent());
     return false;
   }
 
@@ -2658,7 +2727,7 @@ bool GVNPass::processInstruction(Instruction *I) {
   // need to do a lookup to see if the number already exists
   // somewhere in the domtree: it can't!
   if (Num >= NextNum) {
-    addToLeaderTable(Num, I, I->getParent());
+    LeaderTable.insert(Num, I, I->getParent());
     return false;
   }
 
@@ -2667,7 +2736,7 @@ bool GVNPass::processInstruction(Instruction *I) {
   Value *Repl = findLeader(I->getParent(), Num);
   if (!Repl) {
     // Failure, just remember this instance for future use.
-    addToLeaderTable(Num, I, I->getParent());
+    LeaderTable.insert(Num, I, I->getParent());
     return false;
   }
 
@@ -2862,7 +2931,7 @@ bool GVNPass::performScalarPREInsertion(Instruction *Instr, BasicBlock *Pred,
   VN.add(Instr, Num);
 
   // Update the availability map to include the new instruction.
-  addToLeaderTable(Num, Instr, Pred);
+  LeaderTable.insert(Num, Instr, Pred);
   return true;
 }
 
@@ -3013,13 +3082,13 @@ bool GVNPass::performScalarPRE(Instruction *CurInst) {
   // After creating a new PHI for ValNo, the phi translate result for ValNo will
   // be changed, so erase the related stale entries in phi translate cache.
   VN.eraseTranslateCacheEntry(ValNo, *CurrentBlock);
-  addToLeaderTable(ValNo, Phi, CurrentBlock);
+  LeaderTable.insert(ValNo, Phi, CurrentBlock);
   Phi->setDebugLoc(CurInst->getDebugLoc());
   CurInst->replaceAllUsesWith(Phi);
   if (MD && Phi->getType()->isPtrOrPtrVectorTy())
     MD->invalidateCachedPointerInfo(Phi);
   VN.erase(CurInst);
-  removeFromLeaderTable(ValNo, CurInst, CurrentBlock);
+  LeaderTable.erase(ValNo, CurInst, CurrentBlock);
 
   LLVM_DEBUG(dbgs() << "GVN PRE removed: " << *CurInst << '\n');
   removeInstruction(CurInst);
@@ -3113,7 +3182,6 @@ void GVNPass::cleanupGlobalSets() {
   VN.clear();
   LeaderTable.clear();
   BlockRPONumber.clear();
-  TableAllocator.Reset();
   ICF->clear();
   InvalidBlockRPONumbers = true;
 }
@@ -3133,18 +3201,7 @@ void GVNPass::removeInstruction(Instruction *I) {
 /// internal data structures.
 void GVNPass::verifyRemoved(const Instruction *Inst) const {
   VN.verifyRemoved(Inst);
-
-  // Walk through the value number scope to make sure the instruction isn't
-  // ferreted away in it.
-  for (const auto &I : LeaderTable) {
-    const LeaderTableEntry *Node = &I.second;
-    assert(Node->Val != Inst && "Inst still in value numbering scope!");
-
-    while (Node->Next) {
-      Node = Node->Next;
-      assert(Node->Val != Inst && "Inst still in value numbering scope!");
-    }
-  }
+  LeaderTable.verifyRemoved(Inst);
 }
 
 /// BB is declared dead, which implied other blocks become dead as well. This
@@ -3271,7 +3328,7 @@ void GVNPass::assignValNumForDeadCode() {
   for (BasicBlock *BB : DeadBlocks) {
     for (Instruction &Inst : *BB) {
       unsigned ValNum = VN.lookupOrAdd(&Inst);
-      addToLeaderTable(ValNum, &Inst, BB);
+      LeaderTable.insert(ValNum, &Inst, BB);
     }
   }
 }
diff --git a/llvm/lib/Transforms/Scalar/InductiveRangeCheckElimination.cpp b/llvm/lib/Transforms/Scalar/InductiveRangeCheckElimination.cpp
index 9df28747570c..104e8ceb7967 100644
--- a/llvm/lib/Transforms/Scalar/InductiveRangeCheckElimination.cpp
+++ b/llvm/lib/Transforms/Scalar/InductiveRangeCheckElimination.cpp
@@ -279,6 +279,9 @@ bool InductiveRangeCheck::parseRangeCheckICmp(Loop *L, ICmpInst *ICI,
   Value *LHS = ICI->getOperand(0);
   Value *RHS = ICI->getOperand(1);
 
+  if (!LHS->getType()->isIntegerTy())
+    return false;
+
   // Canonicalize to the `Index Pred Invariant` comparison
   if (IsLoopInvariant(LHS)) {
     std::swap(LHS, RHS);
diff --git a/llvm/lib/Transforms/Scalar/JumpThreading.cpp b/llvm/lib/Transforms/Scalar/JumpThreading.cpp
index ffcb511e6a83..08d82fa66da3 100644
--- a/llvm/lib/Transforms/Scalar/JumpThreading.cpp
+++ b/llvm/lib/Transforms/Scalar/JumpThreading.cpp
@@ -1876,7 +1876,7 @@ bool JumpThreadingPass::processBranchOnXOR(BinaryOperator *BO) {
 static void addPHINodeEntriesForMappedBlock(BasicBlock *PHIBB,
                                             BasicBlock *OldPred,
                                             BasicBlock *NewPred,
-                                     DenseMap<Instruction*, Value*> &ValueMap) {
+                                            ValueToValueMapTy &ValueMap) {
   for (PHINode &PN : PHIBB->phis()) {
     // Ok, we have a PHI node.  Figure out what the incoming value was for the
     // DestBlock.
@@ -1884,7 +1884,7 @@ static void addPHINodeEntriesForMappedBlock(BasicBlock *PHIBB,
 
     // Remap the value if necessary.
     if (Instruction *Inst = dyn_cast<Instruction>(IV)) {
-      DenseMap<Instruction*, Value*>::iterator I = ValueMap.find(Inst);
+      ValueToValueMapTy::iterator I = ValueMap.find(Inst);
       if (I != ValueMap.end())
         IV = I->second;
     }
@@ -1945,9 +1945,8 @@ bool JumpThreadingPass::maybeMergeBasicBlockIntoOnlyPred(BasicBlock *BB) {
 
 /// Update the SSA form.  NewBB contains instructions that are copied from BB.
 /// ValueMapping maps old values in BB to new ones in NewBB.
-void JumpThreadingPass::updateSSA(
-    BasicBlock *BB, BasicBlock *NewBB,
-    DenseMap<Instruction *, Value *> &ValueMapping) {
+void JumpThreadingPass::updateSSA(BasicBlock *BB, BasicBlock *NewBB,
+                                  ValueToValueMapTy &ValueMapping) {
   // If there were values defined in BB that are used outside the block, then we
   // now have to update all uses of the value to use either the original value,
   // the cloned value, or some PHI derived value.  This can require arbitrary
@@ -2008,14 +2007,15 @@ void JumpThreadingPass::updateSSA(
 /// Clone instructions in range [BI, BE) to NewBB.  For PHI nodes, we only clone
 /// arguments that come from PredBB.  Return the map from the variables in the
 /// source basic block to the variables in the newly created basic block.
-DenseMap<Instruction *, Value *>
-JumpThreadingPass::cloneInstructions(BasicBlock::iterator BI,
-                                     BasicBlock::iterator BE, BasicBlock *NewBB,
-                                     BasicBlock *PredBB) {
+
+void JumpThreadingPass::cloneInstructions(ValueToValueMapTy &ValueMapping,
+                                          BasicBlock::iterator BI,
+                                          BasicBlock::iterator BE,
+                                          BasicBlock *NewBB,
+                                          BasicBlock *PredBB) {
   // We are going to have to map operands from the source basic block to the new
   // copy of the block 'NewBB'.  If there are PHI nodes in the source basic
   // block, evaluate them to account for entry from PredBB.
-  DenseMap<Instruction *, Value *> ValueMapping;
 
   // Retargets llvm.dbg.value to any renamed variables.
   auto RetargetDbgValueIfPossible = [&](Instruction *NewInst) -> bool {
@@ -2103,7 +2103,7 @@ JumpThreadingPass::cloneInstructions(BasicBlock::iterator BI,
     // Remap operands to patch up intra-block references.
     for (unsigned i = 0, e = New->getNumOperands(); i != e; ++i)
       if (Instruction *Inst = dyn_cast<Instruction>(New->getOperand(i))) {
-        DenseMap<Instruction *, Value *>::iterator I = ValueMapping.find(Inst);
+        ValueToValueMapTy::iterator I = ValueMapping.find(Inst);
         if (I != ValueMapping.end())
           New->setOperand(i, I->second);
       }
@@ -2120,7 +2120,7 @@ JumpThreadingPass::cloneInstructions(BasicBlock::iterator BI,
       RetargetDbgVariableRecordIfPossible(&DVR);
   }
 
-  return ValueMapping;
+  return;
 }
 
 /// Attempt to thread through two successive basic blocks.
@@ -2295,8 +2295,9 @@ void JumpThreadingPass::threadThroughTwoBasicBlocks(BasicBlock *PredPredBB,
   // We are going to have to map operands from the original BB block to the new
   // copy of the block 'NewBB'.  If there are PHI nodes in PredBB, evaluate them
   // to account for entry from PredPredBB.
-  DenseMap<Instruction *, Value *> ValueMapping =
-      cloneInstructions(PredBB->begin(), PredBB->end(), NewBB, PredPredBB);
+  ValueToValueMapTy ValueMapping;
+  cloneInstructions(ValueMapping, PredBB->begin(), PredBB->end(), NewBB,
+                    PredPredBB);
 
   // Copy the edge probabilities from PredBB to NewBB.
   if (BPI)
@@ -2419,8 +2420,9 @@ void JumpThreadingPass::threadEdge(BasicBlock *BB,
   }
 
   // Copy all the instructions from BB to NewBB except the terminator.
-  DenseMap<Instruction *, Value *> ValueMapping =
-      cloneInstructions(BB->begin(), std::prev(BB->end()), NewBB, PredBB);
+  ValueToValueMapTy ValueMapping;
+  cloneInstructions(ValueMapping, BB->begin(), std::prev(BB->end()), NewBB,
+                    PredBB);
 
   // We didn't copy the terminator from BB over to NewBB, because there is now
   // an unconditional jump to SuccBB.  Insert the unconditional jump.
@@ -2675,7 +2677,7 @@ bool JumpThreadingPass::duplicateCondBranchOnPHIIntoPred(
 
   // We are going to have to map operands from the original BB block into the
   // PredBB block.  Evaluate PHI nodes in BB.
-  DenseMap<Instruction*, Value*> ValueMapping;
+  ValueToValueMapTy ValueMapping;
 
   BasicBlock::iterator BI = BB->begin();
   for (; PHINode *PN = dyn_cast<PHINode>(BI); ++BI)
@@ -2689,11 +2691,14 @@ bool JumpThreadingPass::duplicateCondBranchOnPHIIntoPred(
     // Remap operands to patch up intra-block references.
     for (unsigned i = 0, e = New->getNumOperands(); i != e; ++i)
       if (Instruction *Inst = dyn_cast<Instruction>(New->getOperand(i))) {
-        DenseMap<Instruction*, Value*>::iterator I = ValueMapping.find(Inst);
+        ValueToValueMapTy::iterator I = ValueMapping.find(Inst);
         if (I != ValueMapping.end())
           New->setOperand(i, I->second);
       }
 
+    // Remap debug variable operands.
+    remapDebugVariable(ValueMapping, New);
+
     // If this instruction can be simplified after the operands are updated,
     // just use the simplified value instead.  This frequently happens due to
     // phi translation.
diff --git a/llvm/lib/Transforms/Scalar/LoopLoadElimination.cpp b/llvm/lib/Transforms/Scalar/LoopLoadElimination.cpp
index edddfb1b9240..059900f357e6 100644
--- a/llvm/lib/Transforms/Scalar/LoopLoadElimination.cpp
+++ b/llvm/lib/Transforms/Scalar/LoopLoadElimination.cpp
@@ -126,8 +126,10 @@ struct StoreToLoadForwardingCandidate {
 
     // We don't need to check non-wrapping here because forward/backward
     // dependence wouldn't be valid if these weren't monotonic accesses.
-    auto *Dist = cast<SCEVConstant>(
+    auto *Dist = dyn_cast<SCEVConstant>(
         PSE.getSE()->getMinusSCEV(StorePtrSCEV, LoadPtrSCEV));
+    if (!Dist)
+      return false;
     const APInt &Val = Dist->getAPInt();
     return Val == TypeByteSize * StrideLoad;
   }
diff --git a/llvm/lib/Transforms/Scalar/MemCpyOptimizer.cpp b/llvm/lib/Transforms/Scalar/MemCpyOptimizer.cpp
index 1036b8ae963a..7ef5dceffec0 100644
--- a/llvm/lib/Transforms/Scalar/MemCpyOptimizer.cpp
+++ b/llvm/lib/Transforms/Scalar/MemCpyOptimizer.cpp
@@ -99,7 +99,7 @@ struct MemsetRange {
   MaybeAlign Alignment;
 
   /// TheStores - The actual stores that make up this range.
-  SmallVector<Instruction*, 16> TheStores;
+  SmallVector<Instruction *, 16> TheStores;
 
   bool isProfitableToUseMemset(const DataLayout &DL) const;
 };
@@ -108,10 +108,12 @@ struct MemsetRange {
 
 bool MemsetRange::isProfitableToUseMemset(const DataLayout &DL) const {
   // If we found more than 4 stores to merge or 16 bytes, use memset.
-  if (TheStores.size() >= 4 || End-Start >= 16) return true;
+  if (TheStores.size() >= 4 || End - Start >= 16)
+    return true;
 
   // If there is nothing to merge, don't do anything.
-  if (TheStores.size() < 2) return false;
+  if (TheStores.size() < 2)
+    return false;
 
   // If any of the stores are a memset, then it is always good to extend the
   // memset.
@@ -121,7 +123,8 @@ bool MemsetRange::isProfitableToUseMemset(const DataLayout &DL) const {
 
   // Assume that the code generator is capable of merging pairs of stores
   // together if it wants to.
-  if (TheStores.size() == 2) return false;
+  if (TheStores.size() == 2)
+    return false;
 
   // If we have fewer than 8 stores, it can still be worthwhile to do this.
   // For example, merging 4 i8 stores into an i32 store is useful almost always.
@@ -133,7 +136,7 @@ bool MemsetRange::isProfitableToUseMemset(const DataLayout &DL) const {
   // the maximum GPR width is the same size as the largest legal integer
   // size. If so, check to see whether we will end up actually reducing the
   // number of stores used.
-  unsigned Bytes = unsigned(End-Start);
+  unsigned Bytes = unsigned(End - Start);
   unsigned MaxIntSize = DL.getLargestLegalIntTypeSizeInBits() / 8;
   if (MaxIntSize == 0)
     MaxIntSize = 1;
@@ -145,7 +148,7 @@ bool MemsetRange::isProfitableToUseMemset(const DataLayout &DL) const {
   // If we will reduce the # stores (according to this heuristic), do the
   // transformation.  This encourages merging 4 x i8 -> i32 and 2 x i16 -> i32
   // etc.
-  return TheStores.size() > NumPointerStores+NumByteStores;
+  return TheStores.size() > NumPointerStores + NumByteStores;
 }
 
 namespace {
@@ -197,7 +200,7 @@ public:
 /// existing ranges as appropriate.
 void MemsetRanges::addRange(int64_t Start, int64_t Size, Value *Ptr,
                             MaybeAlign Alignment, Instruction *Inst) {
-  int64_t End = Start+Size;
+  int64_t End = Start + Size;
 
   range_iterator I = partition_point(
       Ranges, [=](const MemsetRange &O) { return O.End < Start; });
@@ -207,10 +210,10 @@ void MemsetRanges::addRange(int64_t Start, int64_t Size, Value *Ptr,
   // to insert a new range.  Handle this now.
   if (I == Ranges.end() || End < I->Start) {
     MemsetRange &R = *Ranges.insert(I, MemsetRange());
-    R.Start        = Start;
-    R.End          = End;
-    R.StartPtr     = Ptr;
-    R.Alignment    = Alignment;
+    R.Start = Start;
+    R.End = End;
+    R.StartPtr = Ptr;
+    R.Alignment = Alignment;
     R.TheStores.push_back(Inst);
     return;
   }
@@ -397,7 +400,8 @@ Instruction *MemCpyOptPass::tryMergingIntoMemset(Instruction *StartInst,
 
     if (auto *NextStore = dyn_cast<StoreInst>(BI)) {
       // If this is a store, see if we can merge it in.
-      if (!NextStore->isSimple()) break;
+      if (!NextStore->isSimple())
+        break;
 
       Value *StoredVal = NextStore->getValueOperand();
 
@@ -460,7 +464,8 @@ Instruction *MemCpyOptPass::tryMergingIntoMemset(Instruction *StartInst,
   // emit memset's for anything big enough to be worthwhile.
   Instruction *AMemSet = nullptr;
   for (const MemsetRange &Range : Ranges) {
-    if (Range.TheStores.size() == 1) continue;
+    if (Range.TheStores.size() == 1)
+      continue;
 
     // If it is profitable to lower this range to memset, do so now.
     if (!Range.isProfitableToUseMemset(DL))
@@ -481,12 +486,10 @@ Instruction *MemCpyOptPass::tryMergingIntoMemset(Instruction *StartInst,
     if (!Range.TheStores.empty())
       AMemSet->setDebugLoc(Range.TheStores[0]->getDebugLoc());
 
-    auto *NewDef =
-        cast<MemoryDef>(MemInsertPoint->getMemoryInst() == &*BI
-                            ? MSSAU->createMemoryAccessBefore(
-                                  AMemSet, nullptr, MemInsertPoint)
-                            : MSSAU->createMemoryAccessAfter(
-                                  AMemSet, nullptr, MemInsertPoint));
+    auto *NewDef = cast<MemoryDef>(
+        MemInsertPoint->getMemoryInst() == &*BI
+            ? MSSAU->createMemoryAccessBefore(AMemSet, nullptr, MemInsertPoint)
+            : MSSAU->createMemoryAccessAfter(AMemSet, nullptr, MemInsertPoint));
     MSSAU->insertDef(NewDef, /*RenameUses=*/true);
     MemInsertPoint = NewDef;
 
@@ -512,12 +515,13 @@ bool MemCpyOptPass::moveUp(StoreInst *SI, Instruction *P, const LoadInst *LI) {
 
   // Keep track of the arguments of all instruction we plan to lift
   // so we can make sure to lift them as well if appropriate.
-  DenseSet<Instruction*> Args;
+  DenseSet<Instruction *> Args;
   auto AddArg = [&](Value *Arg) {
     auto *I = dyn_cast<Instruction>(Arg);
     if (I && I->getParent() == SI->getParent()) {
       // Cannot hoist user of P above P
-      if (I == P) return false;
+      if (I == P)
+        return false;
       Args.insert(I);
     }
     return true;
@@ -630,8 +634,7 @@ bool MemCpyOptPass::moveUp(StoreInst *SI, Instruction *P, const LoadInst *LI) {
 bool MemCpyOptPass::processStoreOfLoad(StoreInst *SI, LoadInst *LI,
                                        const DataLayout &DL,
                                        BasicBlock::iterator &BBI) {
-  if (!LI->isSimple() || !LI->hasOneUse() ||
-      LI->getParent() != SI->getParent())
+  if (!LI->isSimple() || !LI->hasOneUse() || LI->getParent() != SI->getParent())
     return false;
 
   auto *T = LI->getType();
@@ -678,21 +681,20 @@ bool MemCpyOptPass::processStoreOfLoad(StoreInst *SI, LoadInst *LI,
         UseMemMove = true;
 
       IRBuilder<> Builder(P);
-      Value *Size = Builder.CreateTypeSize(Builder.getInt64Ty(),
-                                           DL.getTypeStoreSize(T));
+      Value *Size =
+          Builder.CreateTypeSize(Builder.getInt64Ty(), DL.getTypeStoreSize(T));
       Instruction *M;
       if (UseMemMove)
-        M = Builder.CreateMemMove(
-            SI->getPointerOperand(), SI->getAlign(),
-            LI->getPointerOperand(), LI->getAlign(), Size);
+        M = Builder.CreateMemMove(SI->getPointerOperand(), SI->getAlign(),
+                                  LI->getPointerOperand(), LI->getAlign(),
+                                  Size);
       else
-        M = Builder.CreateMemCpy(
-            SI->getPointerOperand(), SI->getAlign(),
-            LI->getPointerOperand(), LI->getAlign(), Size);
+        M = Builder.CreateMemCpy(SI->getPointerOperand(), SI->getAlign(),
+                                 LI->getPointerOperand(), LI->getAlign(), Size);
       M->copyMetadata(*SI, LLVMContext::MD_DIAssignID);
 
-      LLVM_DEBUG(dbgs() << "Promoting " << *LI << " to " << *SI << " => "
-                        << *M << "\n");
+      LLVM_DEBUG(dbgs() << "Promoting " << *LI << " to " << *SI << " => " << *M
+                        << "\n");
 
       auto *LastDef =
           cast<MemoryDef>(MSSAU->getMemorySSA()->getMemoryAccess(SI));
@@ -755,7 +757,8 @@ bool MemCpyOptPass::processStoreOfLoad(StoreInst *SI, LoadInst *LI,
 }
 
 bool MemCpyOptPass::processStore(StoreInst *SI, BasicBlock::iterator &BBI) {
-  if (!SI->isSimple()) return false;
+  if (!SI->isSimple())
+    return false;
 
   // Avoid merging nontemporal stores since the resulting
   // memcpy/memset would not be able to preserve the nontemporal hint.
@@ -794,8 +797,8 @@ bool MemCpyOptPass::processStore(StoreInst *SI, BasicBlock::iterator &BBI) {
   // 0xA0A0A0A0 and 0.0.
   auto *V = SI->getOperand(0);
   if (Value *ByteVal = isBytewiseValue(V, DL)) {
-    if (Instruction *I = tryMergingIntoMemset(SI, SI->getPointerOperand(),
-                                              ByteVal)) {
+    if (Instruction *I =
+            tryMergingIntoMemset(SI, SI->getPointerOperand(), ByteVal)) {
       BBI = I->getIterator(); // Don't invalidate iterator.
       return true;
     }
@@ -816,8 +819,7 @@ bool MemCpyOptPass::processStore(StoreInst *SI, BasicBlock::iterator &BBI) {
       // The newly inserted memset is immediately overwritten by the original
       // store, so we do not need to rename uses.
       auto *StoreDef = cast<MemoryDef>(MSSA->getMemoryAccess(SI));
-      auto *NewAccess = MSSAU->createMemoryAccessBefore(
-          M, nullptr, StoreDef);
+      auto *NewAccess = MSSAU->createMemoryAccessBefore(M, nullptr, StoreDef);
       MSSAU->insertDef(cast<MemoryDef>(NewAccess), /*RenameUses=*/false);
 
       eraseInstruction(SI);
@@ -836,8 +838,8 @@ bool MemCpyOptPass::processMemSet(MemSetInst *MSI, BasicBlock::iterator &BBI) {
   // See if there is another memset or store neighboring this memset which
   // allows us to widen out the memset to do a single larger store.
   if (isa<ConstantInt>(MSI->getLength()) && !MSI->isVolatile())
-    if (Instruction *I = tryMergingIntoMemset(MSI, MSI->getDest(),
-                                              MSI->getValue())) {
+    if (Instruction *I =
+            tryMergingIntoMemset(MSI, MSI->getDest(), MSI->getValue())) {
       BBI = I->getIterator(); // Don't invalidate iterator.
       return true;
     }
@@ -850,7 +852,8 @@ bool MemCpyOptPass::processMemSet(MemSetInst *MSI, BasicBlock::iterator &BBI) {
 bool MemCpyOptPass::performCallSlotOptzn(Instruction *cpyLoad,
                                          Instruction *cpyStore, Value *cpyDest,
                                          Value *cpySrc, TypeSize cpySize,
-                                         Align cpyDestAlign, BatchAAResults &BAA,
+                                         Align cpyDestAlign,
+                                         BatchAAResults &BAA,
                                          std::function<CallInst *()> GetC) {
   // The general transformation to keep in mind is
   //
@@ -898,15 +901,15 @@ bool MemCpyOptPass::performCallSlotOptzn(Instruction *cpyLoad,
     if (F->isIntrinsic() && F->getIntrinsicID() == Intrinsic::lifetime_start)
       return false;
 
-
   if (C->getParent() != cpyStore->getParent()) {
     LLVM_DEBUG(dbgs() << "Call Slot: block local restriction\n");
     return false;
   }
 
-  MemoryLocation DestLoc = isa<StoreInst>(cpyStore) ?
-    MemoryLocation::get(cpyStore) :
-    MemoryLocation::getForDest(cast<MemCpyInst>(cpyStore));
+  MemoryLocation DestLoc =
+      isa<StoreInst>(cpyStore)
+          ? MemoryLocation::get(cpyStore)
+          : MemoryLocation::getForDest(cast<MemCpyInst>(cpyStore));
 
   // Check that nothing touches the dest of the copy between
   // the call and the store/memcpy.
@@ -1175,7 +1178,8 @@ bool MemCpyOptPass::processMemCpyMemCpyDependence(MemCpyInst *M,
 
   // If all checks passed, then we can transform M.
   LLVM_DEBUG(dbgs() << "MemCpyOptPass: Forwarding memcpy->memcpy src:\n"
-                    << *MDep << '\n' << *M << '\n');
+                    << *MDep << '\n'
+                    << *M << '\n');
 
   // TODO: Is this worth it if we're creating a less aligned memcpy? For
   // example we could be moving from movaps -> movq on x86.
@@ -1307,8 +1311,8 @@ bool MemCpyOptPass::processMemSetMemCpyDependence(MemCpyInst *MemCpy,
   // memcpy's defining access is the memset about to be removed.
   auto *LastDef =
       cast<MemoryDef>(MSSAU->getMemorySSA()->getMemoryAccess(MemCpy));
-  auto *NewAccess = MSSAU->createMemoryAccessBefore(
-      NewMemSet, nullptr, LastDef);
+  auto *NewAccess =
+      MSSAU->createMemoryAccessBefore(NewMemSet, nullptr, LastDef);
   MSSAU->insertDef(cast<MemoryDef>(NewAccess), /*RenameUses=*/true);
 
   eraseInstruction(MemSet);
@@ -1384,7 +1388,7 @@ bool MemCpyOptPass::performMemCpyToMemSetOptzn(MemCpyInst *MemCpy,
       return false;
 
     // A known memcpy size is also required.
-    auto  *CCopySize = dyn_cast<ConstantInt>(CopySize);
+    auto *CCopySize = dyn_cast<ConstantInt>(CopySize);
     if (!CCopySize)
       return false;
     if (CCopySize->getZExtValue() > CMemSetSize->getZExtValue()) {
@@ -1655,7 +1659,8 @@ static bool isZeroSize(Value *Size) {
 /// altogether.
 bool MemCpyOptPass::processMemCpy(MemCpyInst *M, BasicBlock::iterator &BBI) {
   // We can only optimize non-volatile memcpy's.
-  if (M->isVolatile()) return false;
+  if (M->isVolatile())
+    return false;
 
   // If the source and destination of the memcpy are the same, then zap it.
   if (M->getSource() == M->getDest()) {
@@ -1796,11 +1801,10 @@ bool MemCpyOptPass::processMemMove(MemMoveInst *M) {
                     << "\n");
 
   // If not, then we know we can transform this.
-  Type *ArgTys[3] = { M->getRawDest()->getType(),
-                      M->getRawSource()->getType(),
-                      M->getLength()->getType() };
-  M->setCalledFunction(Intrinsic::getDeclaration(M->getModule(),
-                                                 Intrinsic::memcpy, ArgTys));
+  Type *ArgTys[3] = {M->getRawDest()->getType(), M->getRawSource()->getType(),
+                     M->getLength()->getType()};
+  M->setCalledFunction(
+      Intrinsic::getDeclaration(M->getModule(), Intrinsic::memcpy, ArgTys));
 
   // For MemorySSA nothing really changes (except that memcpy may imply stricter
   // aliasing guarantees).
@@ -1843,7 +1847,8 @@ bool MemCpyOptPass::processByValArgument(CallBase &CB, unsigned ArgNo) {
   // Get the alignment of the byval.  If the call doesn't specify the alignment,
   // then it is some target specific value that we can't know.
   MaybeAlign ByValAlign = CB.getParamAlign(ArgNo);
-  if (!ByValAlign) return false;
+  if (!ByValAlign)
+    return false;
 
   // If it is greater than the memcpy, then we check to see if we can force the
   // source of the memcpy to the alignment we need.  If we fail, we bail out.
@@ -1987,7 +1992,7 @@ bool MemCpyOptPass::iterateOnFunction(Function &F) {
       continue;
 
     for (BasicBlock::iterator BI = BB.begin(), BE = BB.end(); BI != BE;) {
-        // Avoid invalidating the iterator.
+      // Avoid invalidating the iterator.
       Instruction *I = &*BI++;
 
       bool RepeatInstruction = false;
diff --git a/llvm/lib/Transforms/Utils/BasicBlockUtils.cpp b/llvm/lib/Transforms/Utils/BasicBlockUtils.cpp
index 5396038d8b92..51fc28ef90ef 100644
--- a/llvm/lib/Transforms/Utils/BasicBlockUtils.cpp
+++ b/llvm/lib/Transforms/Utils/BasicBlockUtils.cpp
@@ -333,6 +333,10 @@ bool llvm::MergeBlockIntoPredecessor(BasicBlock *BB, DomTreeUpdater *DTU,
   // Finally, erase the old block and update dominator info.
   DeleteDeadBlock(BB, DTU);
 
+  // Remove redundant "llvm.dbg" instrunctions after blocks have been merged.
+  if (PredBB->getParent()->getSubprogram())
+    RemoveRedundantDbgInstrs(PredBB);
+
   return true;
 }
 
@@ -1401,13 +1405,13 @@ SplitBlockPredecessorsImpl(BasicBlock *BB, ArrayRef<BasicBlock *> Preds,
   if (OldLatch) {
     BasicBlock *NewLatch = L->getLoopLatch();
     if (NewLatch != OldLatch) {
-      MDNode *MD = OldLatch->getTerminator()->getMetadata("llvm.loop");
-      NewLatch->getTerminator()->setMetadata("llvm.loop", MD);
+      MDNode *MD = OldLatch->getTerminator()->getMetadata(LLVMContext::MD_loop);
+      NewLatch->getTerminator()->setMetadata(LLVMContext::MD_loop, MD);
       // It's still possible that OldLatch is the latch of another inner loop,
       // in which case we do not remove the metadata.
       Loop *IL = LI->getLoopFor(OldLatch);
       if (IL && IL->getLoopLatch() != OldLatch)
-        OldLatch->getTerminator()->setMetadata("llvm.loop", nullptr);
+        OldLatch->getTerminator()->setMetadata(LLVMContext::MD_loop, nullptr);
     }
   }
 
diff --git a/llvm/lib/Transforms/Utils/CloneFunction.cpp b/llvm/lib/Transforms/Utils/CloneFunction.cpp
index 3eac726994ae..303a09805a9d 100644
--- a/llvm/lib/Transforms/Utils/CloneFunction.cpp
+++ b/llvm/lib/Transforms/Utils/CloneFunction.cpp
@@ -1131,6 +1131,9 @@ BasicBlock *llvm::DuplicateInstructionsInSplitBetween(
         if (I != ValueMapping.end())
           New->setOperand(i, I->second);
       }
+
+    // Remap debug variable operands.
+    remapDebugVariable(ValueMapping, New);
   }
 
   return NewBB;
diff --git a/llvm/lib/Transforms/Utils/EntryExitInstrumenter.cpp b/llvm/lib/Transforms/Utils/EntryExitInstrumenter.cpp
index f4207474e9a6..59a7dd1a00ed 100644
--- a/llvm/lib/Transforms/Utils/EntryExitInstrumenter.cpp
+++ b/llvm/lib/Transforms/Utils/EntryExitInstrumenter.cpp
@@ -137,7 +137,8 @@ static bool runOnFunction(Function &F, bool PostInlining) {
 
 PreservedAnalyses
 llvm::EntryExitInstrumenterPass::run(Function &F, FunctionAnalysisManager &AM) {
-  runOnFunction(F, PostInlining);
+  if (!runOnFunction(F, PostInlining))
+    return PreservedAnalyses::all();
   PreservedAnalyses PA;
   PA.preserveSet<CFGAnalyses>();
   return PA;
diff --git a/llvm/lib/Transforms/Utils/GlobalStatus.cpp b/llvm/lib/Transforms/Utils/GlobalStatus.cpp
index c5aded3c45f4..b177e048faae 100644
--- a/llvm/lib/Transforms/Utils/GlobalStatus.cpp
+++ b/llvm/lib/Transforms/Utils/GlobalStatus.cpp
@@ -172,9 +172,14 @@ static bool analyzeGlobalAux(const Value *V, GlobalStatus &GS,
           return true;
         GS.StoredType = GlobalStatus::Stored;
       } else if (const auto *CB = dyn_cast<CallBase>(I)) {
-        if (!CB->isCallee(&U))
-          return true;
-        GS.IsLoaded = true;
+        if (CB->getIntrinsicID() == Intrinsic::threadlocal_address) {
+          if (analyzeGlobalAux(I, GS, VisitedUsers))
+            return true;
+        } else {
+          if (!CB->isCallee(&U))
+            return true;
+          GS.IsLoaded = true;
+        }
       } else {
         return true; // Any other non-load instruction might take address!
       }
diff --git a/llvm/lib/Transforms/Utils/Local.cpp b/llvm/lib/Transforms/Utils/Local.cpp
index a42ef0c4e6ae..f3cd3104c312 100644
--- a/llvm/lib/Transforms/Utils/Local.cpp
+++ b/llvm/lib/Transforms/Utils/Local.cpp
@@ -59,6 +59,7 @@
 #include "llvm/IR/IntrinsicsWebAssembly.h"
 #include "llvm/IR/LLVMContext.h"
 #include "llvm/IR/MDBuilder.h"
+#include "llvm/IR/MemoryModelRelaxationAnnotations.h"
 #include "llvm/IR/Metadata.h"
 #include "llvm/IR/Module.h"
 #include "llvm/IR/PatternMatch.h"
@@ -3284,6 +3285,9 @@ void llvm::combineMetadata(Instruction *K, const Instruction *J,
       case LLVMContext::MD_invariant_group:
         // Preserve !invariant.group in K.
         break;
+      case LLVMContext::MD_mmra:
+        // Combine MMRAs
+        break;
       case LLVMContext::MD_align:
         if (DoesKMove || !K->hasMetadata(LLVMContext::MD_noundef))
           K->setMetadata(
@@ -3322,6 +3326,16 @@ void llvm::combineMetadata(Instruction *K, const Instruction *J,
   if (auto *JMD = J->getMetadata(LLVMContext::MD_invariant_group))
     if (isa<LoadInst>(K) || isa<StoreInst>(K))
       K->setMetadata(LLVMContext::MD_invariant_group, JMD);
+
+  // Merge MMRAs.
+  // This is handled separately because we also want to handle cases where K
+  // doesn't have tags but J does.
+  auto JMMRA = J->getMetadata(LLVMContext::MD_mmra);
+  auto KMMRA = K->getMetadata(LLVMContext::MD_mmra);
+  if (JMMRA || KMMRA) {
+    K->setMetadata(LLVMContext::MD_mmra,
+                   MMRAMetadata::combine(K->getContext(), JMMRA, KMMRA));
+  }
 }
 
 void llvm::combineMetadataForCSE(Instruction *K, const Instruction *J,
@@ -3341,7 +3355,8 @@ void llvm::combineMetadataForCSE(Instruction *K, const Instruction *J,
                          LLVMContext::MD_preserve_access_index,
                          LLVMContext::MD_prof,
                          LLVMContext::MD_nontemporal,
-                         LLVMContext::MD_noundef};
+                         LLVMContext::MD_noundef,
+                         LLVMContext::MD_mmra};
   combineMetadata(K, J, KnownIDs, KDominatesJ);
 }
 
@@ -3429,15 +3444,15 @@ void llvm::patchReplacementInstruction(Instruction *I, Value *Repl) {
   combineMetadataForCSE(ReplInst, I, false);
 }
 
-template <typename RootType, typename DominatesFn>
+template <typename RootType, typename ShouldReplaceFn>
 static unsigned replaceDominatedUsesWith(Value *From, Value *To,
                                          const RootType &Root,
-                                         const DominatesFn &Dominates) {
+                                         const ShouldReplaceFn &ShouldReplace) {
   assert(From->getType() == To->getType());
 
   unsigned Count = 0;
   for (Use &U : llvm::make_early_inc_range(From->uses())) {
-    if (!Dominates(Root, U))
+    if (!ShouldReplace(Root, U))
       continue;
     LLVM_DEBUG(dbgs() << "Replace dominated use of '";
                From->printAsOperand(dbgs());
@@ -3481,6 +3496,26 @@ unsigned llvm::replaceDominatedUsesWith(Value *From, Value *To,
   return ::replaceDominatedUsesWith(From, To, BB, Dominates);
 }
 
+unsigned llvm::replaceDominatedUsesWithIf(
+    Value *From, Value *To, DominatorTree &DT, const BasicBlockEdge &Root,
+    function_ref<bool(const Use &U, const Value *To)> ShouldReplace) {
+  auto DominatesAndShouldReplace =
+      [&DT, &ShouldReplace, To](const BasicBlockEdge &Root, const Use &U) {
+        return DT.dominates(Root, U) && ShouldReplace(U, To);
+      };
+  return ::replaceDominatedUsesWith(From, To, Root, DominatesAndShouldReplace);
+}
+
+unsigned llvm::replaceDominatedUsesWithIf(
+    Value *From, Value *To, DominatorTree &DT, const BasicBlock *BB,
+    function_ref<bool(const Use &U, const Value *To)> ShouldReplace) {
+  auto DominatesAndShouldReplace = [&DT, &ShouldReplace,
+                                    To](const BasicBlock *BB, const Use &U) {
+    return DT.dominates(BB, U) && ShouldReplace(U, To);
+  };
+  return ::replaceDominatedUsesWith(From, To, BB, DominatesAndShouldReplace);
+}
+
 bool llvm::callsGCLeafFunction(const CallBase *Call,
                                const TargetLibraryInfo &TLI) {
   // Check if the function is specifically marked as a gc leaf function.
@@ -3650,6 +3685,30 @@ DIExpression *llvm::getExpressionForConstant(DIBuilder &DIB, const Constant &C,
   return nullptr;
 }
 
+void llvm::remapDebugVariable(ValueToValueMapTy &Mapping, Instruction *Inst) {
+  auto RemapDebugOperands = [&Mapping](auto *DV, auto Set) {
+    for (auto *Op : Set) {
+      auto I = Mapping.find(Op);
+      if (I != Mapping.end())
+        DV->replaceVariableLocationOp(Op, I->second, /*AllowEmpty=*/true);
+    }
+  };
+  auto RemapAssignAddress = [&Mapping](auto *DA) {
+    auto I = Mapping.find(DA->getAddress());
+    if (I != Mapping.end())
+      DA->setAddress(I->second);
+  };
+  if (auto DVI = dyn_cast<DbgVariableIntrinsic>(Inst))
+    RemapDebugOperands(DVI, DVI->location_ops());
+  if (auto DAI = dyn_cast<DbgAssignIntrinsic>(Inst))
+    RemapAssignAddress(DAI);
+  for (DbgVariableRecord &DVR : filterDbgVars(Inst->getDbgRecordRange())) {
+    RemapDebugOperands(&DVR, DVR.location_ops());
+    if (DVR.isDbgAssign())
+      RemapAssignAddress(&DVR);
+  }
+}
+
 namespace {
 
 /// A potential constituent of a bitreverse or bswap expression. See
diff --git a/llvm/lib/Transforms/Utils/LoopUtils.cpp b/llvm/lib/Transforms/Utils/LoopUtils.cpp
index 73c5d6367822..e3e09d11ba8c 100644
--- a/llvm/lib/Transforms/Utils/LoopUtils.cpp
+++ b/llvm/lib/Transforms/Utils/LoopUtils.cpp
@@ -1930,10 +1930,12 @@ llvm::hasPartialIVCondition(const Loop &L, unsigned MSSAThreshold,
   if (!TI || !TI->isConditional())
     return {};
 
-  auto *CondI = dyn_cast<CmpInst>(TI->getCondition());
+  auto *CondI = dyn_cast<Instruction>(TI->getCondition());
   // The case with the condition outside the loop should already be handled
   // earlier.
-  if (!CondI || !L.contains(CondI))
+  // Allow CmpInst and TruncInsts as they may be users of load instructions
+  // and have potential for partial unswitching
+  if (!CondI || !isa<CmpInst, TruncInst>(CondI) || !L.contains(CondI))
     return {};
 
   SmallVector<Instruction *> InstToDuplicate;
diff --git a/llvm/lib/Transforms/Utils/MemoryTaggingSupport.cpp b/llvm/lib/Transforms/Utils/MemoryTaggingSupport.cpp
index 7b1eb70168d8..0464ba5e1811 100644
--- a/llvm/lib/Transforms/Utils/MemoryTaggingSupport.cpp
+++ b/llvm/lib/Transforms/Utils/MemoryTaggingSupport.cpp
@@ -17,6 +17,7 @@
 #include "llvm/Analysis/PostDominators.h"
 #include "llvm/Analysis/StackSafetyAnalysis.h"
 #include "llvm/Analysis/ValueTracking.h"
+#include "llvm/BinaryFormat/Dwarf.h"
 #include "llvm/IR/BasicBlock.h"
 #include "llvm/IR/IRBuilder.h"
 #include "llvm/IR/IntrinsicInst.h"
@@ -283,5 +284,37 @@ Value *getAndroidSlotPtr(IRBuilder<> &IRB, int Slot) {
                                 IRB.CreateCall(ThreadPointerFunc), 8 * Slot);
 }
 
+static DbgAssignIntrinsic *DynCastToDbgAssign(DbgVariableIntrinsic *DVI) {
+  return dyn_cast<DbgAssignIntrinsic>(DVI);
+}
+
+static DbgVariableRecord *DynCastToDbgAssign(DbgVariableRecord *DVR) {
+  return DVR->isDbgAssign() ? DVR : nullptr;
+}
+
+void annotateDebugRecords(AllocaInfo &Info, unsigned int Tag) {
+  // Helper utility for adding DW_OP_LLVM_tag_offset to debug-info records,
+  // abstracted over whether they're intrinsic-stored or DbgVariableRecord
+  // stored.
+  auto AnnotateDbgRecord = [&](auto *DPtr) {
+    // Prepend "tag_offset, N" to the dwarf expression.
+    // Tag offset logically applies to the alloca pointer, and it makes sense
+    // to put it at the beginning of the expression.
+    SmallVector<uint64_t, 8> NewOps = {dwarf::DW_OP_LLVM_tag_offset, Tag};
+    for (size_t LocNo = 0; LocNo < DPtr->getNumVariableLocationOps(); ++LocNo)
+      if (DPtr->getVariableLocationOp(LocNo) == Info.AI)
+        DPtr->setExpression(
+            DIExpression::appendOpsToArg(DPtr->getExpression(), NewOps, LocNo));
+    if (auto *DAI = DynCastToDbgAssign(DPtr)) {
+      if (DAI->getAddress() == Info.AI)
+        DAI->setAddressExpression(
+            DIExpression::prependOpcodes(DAI->getAddressExpression(), NewOps));
+    }
+  };
+
+  llvm::for_each(Info.DbgVariableIntrinsics, AnnotateDbgRecord);
+  llvm::for_each(Info.DbgVariableRecords, AnnotateDbgRecord);
+}
+
 } // namespace memtag
 } // namespace llvm
diff --git a/llvm/lib/Transforms/Utils/SimplifyCFG.cpp b/llvm/lib/Transforms/Utils/SimplifyCFG.cpp
index da3eb229c041..4db72461c95e 100644
--- a/llvm/lib/Transforms/Utils/SimplifyCFG.cpp
+++ b/llvm/lib/Transforms/Utils/SimplifyCFG.cpp
@@ -51,6 +51,7 @@
 #include "llvm/IR/IntrinsicInst.h"
 #include "llvm/IR/LLVMContext.h"
 #include "llvm/IR/MDBuilder.h"
+#include "llvm/IR/MemoryModelRelaxationAnnotations.h"
 #include "llvm/IR/Metadata.h"
 #include "llvm/IR/Module.h"
 #include "llvm/IR/NoFolder.h"
@@ -1677,7 +1678,8 @@ bool SimplifyCFGOpt::hoistCommonCodeFromSuccessors(BasicBlock *BB,
     for (auto &SuccIter : OtherSuccIterRange) {
       Instruction *I2 = &*SuccIter;
       HasTerminator |= I2->isTerminator();
-      if (AllInstsAreIdentical && !I1->isIdenticalToWhenDefined(I2))
+      if (AllInstsAreIdentical && (!I1->isIdenticalToWhenDefined(I2) ||
+                                   MMRAMetadata(*I1) != MMRAMetadata(*I2)))
         AllInstsAreIdentical = false;
     }
 
@@ -1964,6 +1966,7 @@ static bool canSinkInstructions(
   }
 
   const Instruction *I0 = Insts.front();
+  const auto I0MMRA = MMRAMetadata(*I0);
   for (auto *I : Insts) {
     if (!I->isSameOperationAs(I0))
       return false;
@@ -1975,6 +1978,11 @@ static bool canSinkInstructions(
       return false;
     if (isa<LoadInst>(I) && I->getOperand(0)->isSwiftError())
       return false;
+
+    // Treat MMRAs conservatively. This pass can be quite aggressive and
+    // could drop a lot of MMRAs otherwise.
+    if (MMRAMetadata(*I) != I0MMRA)
+      return false;
   }
 
   // All instructions in Insts are known to be the same opcode. If they have a
@@ -2888,7 +2896,8 @@ static Value *isSafeToSpeculateStore(Instruction *I, BasicBlock *BrBB,
       // simple, to avoid introducing a spurious non-atomic write after an
       // atomic write.
       if (SI->getPointerOperand() == StorePtr &&
-          SI->getValueOperand()->getType() == StoreTy && SI->isSimple())
+          SI->getValueOperand()->getType() == StoreTy && SI->isSimple() &&
+          SI->getAlign() >= StoreToHoist->getAlign())
         // Found the previous store, return its value operand.
         return SI->getValueOperand();
       return nullptr; // Unknown store.
@@ -2896,7 +2905,7 @@ static Value *isSafeToSpeculateStore(Instruction *I, BasicBlock *BrBB,
 
     if (auto *LI = dyn_cast<LoadInst>(&CurI)) {
       if (LI->getPointerOperand() == StorePtr && LI->getType() == StoreTy &&
-          LI->isSimple()) {
+          LI->isSimple() && LI->getAlign() >= StoreToHoist->getAlign()) {
         // Local objects (created by an `alloca` instruction) are always
         // writable, so once we are past a read from a location it is valid to
         // also write to that same location.
@@ -7515,6 +7524,13 @@ static bool passingValueIsAlwaysUndefined(Value *V, Instruction *I, bool PtrValu
                                       SI->getPointerAddressSpace())) &&
                SI->getPointerOperand() == I;
 
+    // llvm.assume(false/undef) always triggers immediate UB.
+    if (auto *Assume = dyn_cast<AssumeInst>(Use)) {
+      // Ignore assume operand bundles.
+      if (I == Assume->getArgOperand(0))
+        return true;
+    }
+
     if (auto *CB = dyn_cast<CallBase>(Use)) {
       if (C->isNullValue() && NullPointerIsDefined(CB->getFunction()))
         return false;
diff --git a/llvm/lib/Transforms/Vectorize/LoopVectorize.cpp b/llvm/lib/Transforms/Vectorize/LoopVectorize.cpp
index 33c4decd58a6..f6dcdbee590e 100644
--- a/llvm/lib/Transforms/Vectorize/LoopVectorize.cpp
+++ b/llvm/lib/Transforms/Vectorize/LoopVectorize.cpp
@@ -460,9 +460,9 @@ static Value *interleaveVectors(IRBuilderBase &Builder, ArrayRef<Value *> Vals,
   // must use intrinsics to interleave.
   if (VecTy->isScalableTy()) {
     VectorType *WideVecTy = VectorType::getDoubleElementsVectorType(VecTy);
-    return Builder.CreateIntrinsic(
-        WideVecTy, Intrinsic::experimental_vector_interleave2, Vals,
-        /*FMFSource=*/nullptr, Name);
+    return Builder.CreateIntrinsic(WideVecTy, Intrinsic::vector_interleave2,
+                                   Vals,
+                                   /*FMFSource=*/nullptr, Name);
   }
 
   // Fixed length. Start by concatenating all vectors into a wide vector.
@@ -2517,9 +2517,8 @@ void InnerLoopVectorizer::vectorizeInterleaveGroup(
       SmallVector<Value *, 2> Ops = {BlockInMaskPart, BlockInMaskPart};
       auto *MaskTy =
           VectorType::get(Builder.getInt1Ty(), VF.getKnownMinValue() * 2, true);
-      return Builder.CreateIntrinsic(
-          MaskTy, Intrinsic::experimental_vector_interleave2, Ops,
-          /*FMFSource=*/nullptr, "interleaved.mask");
+      return Builder.CreateIntrinsic(MaskTy, Intrinsic::vector_interleave2, Ops,
+                                     /*FMFSource=*/nullptr, "interleaved.mask");
     }
 
     if (!BlockInMask)
@@ -2571,7 +2570,7 @@ void InnerLoopVectorizer::vectorizeInterleaveGroup(
         // Scalable vectors cannot use arbitrary shufflevectors (only splats),
         // so must use intrinsics to deinterleave.
         Value *DI = Builder.CreateIntrinsic(
-            Intrinsic::experimental_vector_deinterleave2, VecTy, NewLoads[Part],
+            Intrinsic::vector_deinterleave2, VecTy, NewLoads[Part],
             /*FMFSource=*/nullptr, "strided.vec");
         unsigned J = 0;
         for (unsigned I = 0; I < InterleaveFactor; ++I) {
@@ -4167,7 +4166,6 @@ void LoopVectorizationCostModel::collectLoopUniforms(ElementCount VF) {
 
   // Worklist containing uniform instructions demanding lane 0.
   SetVector<Instruction *> Worklist;
-  BasicBlock *Latch = TheLoop->getLoopLatch();
 
   // Add uniform instructions demanding lane 0 to the worklist. Instructions
   // that are scalar with predication must not be considered uniform after
@@ -4189,12 +4187,16 @@ void LoopVectorizationCostModel::collectLoopUniforms(ElementCount VF) {
     Worklist.insert(I);
   };
 
-  // Start with the conditional branch. If the branch condition is an
-  // instruction contained in the loop that is only used by the branch, it is
-  // uniform.
-  auto *Cmp = dyn_cast<Instruction>(Latch->getTerminator()->getOperand(0));
-  if (Cmp && TheLoop->contains(Cmp) && Cmp->hasOneUse())
-    addToWorklistIfAllowed(Cmp);
+  // Start with the conditional branches exiting the loop. If the branch
+  // condition is an instruction contained in the loop that is only used by the
+  // branch, it is uniform.
+  SmallVector<BasicBlock *> Exiting;
+  TheLoop->getExitingBlocks(Exiting);
+  for (BasicBlock *E : Exiting) {
+    auto *Cmp = dyn_cast<Instruction>(E->getTerminator()->getOperand(0));
+    if (Cmp && TheLoop->contains(Cmp) && Cmp->hasOneUse())
+      addToWorklistIfAllowed(Cmp);
+  }
 
   auto PrevVF = VF.divideCoefficientBy(2);
   // Return true if all lanes perform the same memory operation, and we can
@@ -4335,6 +4337,7 @@ void LoopVectorizationCostModel::collectLoopUniforms(ElementCount VF) {
   // nodes separately. An induction variable will remain uniform if all users
   // of the induction variable and induction variable update remain uniform.
   // The code below handles both pointer and non-pointer induction variables.
+  BasicBlock *Latch = TheLoop->getLoopLatch();
   for (const auto &Induction : Legal->getInductionVars()) {
     auto *Ind = Induction.first;
     auto *IndUpdate = cast<Instruction>(Ind->getIncomingValueForBlock(Latch));
@@ -8817,12 +8820,24 @@ LoopVectorizationPlanner::tryToBuildVPlanWithVPRecipes(VFRange &Range) {
     // Only handle constant strides for now.
     if (!ScevStride)
       continue;
-    Constant *CI = ConstantInt::get(Stride->getType(), ScevStride->getAPInt());
 
-    auto *ConstVPV = Plan->getOrAddLiveIn(CI);
-    // The versioned value may not be used in the loop directly, so just add a
-    // new live-in in those cases.
-    Plan->getOrAddLiveIn(StrideV)->replaceAllUsesWith(ConstVPV);
+    auto *CI = Plan->getOrAddLiveIn(
+        ConstantInt::get(Stride->getType(), ScevStride->getAPInt()));
+    if (VPValue *StrideVPV = Plan->getLiveIn(StrideV))
+      StrideVPV->replaceAllUsesWith(CI);
+
+    // The versioned value may not be used in the loop directly but through a
+    // sext/zext. Add new live-ins in those cases.
+    for (Value *U : StrideV->users()) {
+      if (!isa<SExtInst, ZExtInst>(U))
+        continue;
+      VPValue *StrideVPV = Plan->getLiveIn(U);
+      if (!StrideVPV)
+        continue;
+      VPValue *CI = Plan->getOrAddLiveIn(ConstantInt::get(
+          U->getType(), ScevStride->getAPInt().getSExtValue()));
+      StrideVPV->replaceAllUsesWith(CI);
+    }
   }
 
   VPlanTransforms::dropPoisonGeneratingRecipes(*Plan, [this](BasicBlock *BB) {
diff --git a/llvm/lib/Transforms/Vectorize/SLPVectorizer.cpp b/llvm/lib/Transforms/Vectorize/SLPVectorizer.cpp
index 6ac380a6ab6c..e3a1b0d39a4d 100644
--- a/llvm/lib/Transforms/Vectorize/SLPVectorizer.cpp
+++ b/llvm/lib/Transforms/Vectorize/SLPVectorizer.cpp
@@ -1135,6 +1135,7 @@ public:
     ScalarToTreeEntry.clear();
     MultiNodeScalars.clear();
     MustGather.clear();
+    NonScheduledFirst.clear();
     EntryToLastInstruction.clear();
     ExternalUses.clear();
     ExternalUsesAsGEPs.clear();
@@ -1252,7 +1253,7 @@ public:
   /// effectively impossible for the backend to undo.
   /// TODO: If load combining is allowed in the IR optimizer, this analysis
   ///       may not be necessary.
-  bool isLoadCombineCandidate() const;
+  bool isLoadCombineCandidate(ArrayRef<Value *> Stores) const;
 
   /// Checks if the given array of loads can be represented as a vectorized,
   /// scatter or just simple gather.
@@ -2356,6 +2357,14 @@ public:
   bool isAnyGathered(const SmallDenseSet<Value *> &Vals) const {
     return any_of(MustGather, [&](Value *V) { return Vals.contains(V); });
   }
+  /// Checks if the given value is gathered in one of the nodes.
+  bool isGathered(const Value *V) const {
+    return MustGather.contains(V);
+  }
+  /// Checks if the specified value was not schedule.
+  bool isNotScheduled(const Value *V) const {
+    return NonScheduledFirst.contains(V);
+  }
 
   /// Check if the value is vectorized in the tree.
   bool isVectorized(Value *V) const { return getTreeEntry(V); }
@@ -2478,12 +2487,12 @@ private:
   /// which exploits values reused across lanes, and arranges the inserts
   /// for ease of later optimization.
   template <typename BVTy, typename ResTy, typename... Args>
-  ResTy processBuildVector(const TreeEntry *E, Args &...Params);
+  ResTy processBuildVector(const TreeEntry *E, Type *ScalarTy, Args &...Params);
 
   /// Create a new vector from a list of scalar values.  Produces a sequence
   /// which exploits values reused across lanes, and arranges the inserts
   /// for ease of later optimization.
-  Value *createBuildVector(const TreeEntry *E);
+  Value *createBuildVector(const TreeEntry *E, Type *ScalarTy);
 
   /// Returns the instruction in the bundle, which can be used as a base point
   /// for scheduling. Usually it is the last instruction in the bundle, except
@@ -2547,7 +2556,8 @@ private:
   /// this subtree gets vectorized, we may need to extract the values from the
   /// roots. This method calculates the cost of extracting the values.
   /// \param ForPoisonSrc true if initial vector is poison, false otherwise.
-  InstructionCost getGatherCost(ArrayRef<Value *> VL, bool ForPoisonSrc) const;
+  InstructionCost getGatherCost(ArrayRef<Value *> VL, bool ForPoisonSrc,
+                                Type *ScalarTy) const;
 
   /// Set the Builder insert point to one after the last instruction in
   /// the bundle
@@ -2555,7 +2565,7 @@ private:
 
   /// \returns a vector from a collection of scalars in \p VL. if \p Root is not
   /// specified, the starting vector value is poison.
-  Value *gather(ArrayRef<Value *> VL, Value *Root);
+  Value *gather(ArrayRef<Value *> VL, Value *Root, Type *ScalarTy);
 
   /// \returns whether the VectorizableTree is fully vectorizable and will
   /// be beneficial even the tree height is tiny.
@@ -3071,6 +3081,9 @@ private:
   /// A list of scalars that we found that we need to keep as scalars.
   ValueSet MustGather;
 
+  /// A set of first non-schedulable values.
+  ValueSet NonScheduledFirst;
+
   /// A map between the vectorized entries and the last instructions in the
   /// bundles. The bundles are built in use order, not in the def order of the
   /// instructions. So, we cannot rely directly on the last instruction in the
@@ -6646,6 +6659,7 @@ void BoUpSLP::buildTree_rec(ArrayRef<Value *> VL, unsigned Depth,
            "tryScheduleBundle should cancelScheduling on failure");
     newTreeEntry(VL, std::nullopt /*not vectorized*/, S, UserTreeIdx,
                  ReuseShuffleIndicies);
+    NonScheduledFirst.insert(VL.front());
     return;
   }
   LLVM_DEBUG(dbgs() << "SLP: We are able to schedule this bundle.\n");
@@ -7863,6 +7877,7 @@ class BoUpSLP::ShuffleCostEstimator : public BaseShuffleAnalysis {
   bool IsFinalized = false;
   SmallVector<int> CommonMask;
   SmallVector<PointerUnion<Value *, const TreeEntry *>, 2> InVectors;
+  Type *ScalarTy = nullptr;
   const TargetTransformInfo &TTI;
   InstructionCost Cost = 0;
   SmallDenseSet<Value *> VectorizedVals;
@@ -7892,13 +7907,13 @@ class BoUpSLP::ShuffleCostEstimator : public BaseShuffleAnalysis {
   InstructionCost getBuildVectorCost(ArrayRef<Value *> VL, Value *Root) {
     if ((!Root && allConstant(VL)) || all_of(VL, IsaPred<UndefValue>))
       return TTI::TCC_Free;
-    auto *VecTy = FixedVectorType::get(VL.front()->getType(), VL.size());
+    auto *VecTy = FixedVectorType::get(ScalarTy, VL.size());
     InstructionCost GatherCost = 0;
     SmallVector<Value *> Gathers(VL.begin(), VL.end());
     // Improve gather cost for gather of loads, if we can group some of the
     // loads into vector loads.
     InstructionsState S = getSameOpcode(VL, *R.TLI);
-    const unsigned Sz = R.DL->getTypeSizeInBits(VL.front()->getType());
+    const unsigned Sz = R.DL->getTypeSizeInBits(ScalarTy);
     unsigned MinVF = R.getMinVF(2 * Sz);
     if (VL.size() > 2 &&
         ((S.getOpcode() == Instruction::Load && !S.isAltShuffle()) ||
@@ -7912,7 +7927,7 @@ class BoUpSLP::ShuffleCostEstimator : public BaseShuffleAnalysis {
                  }))) &&
         !all_of(Gathers, [&](Value *V) { return R.getTreeEntry(V); }) &&
         !isSplat(Gathers)) {
-      InstructionCost BaseCost = R.getGatherCost(Gathers, !Root);
+      InstructionCost BaseCost = R.getGatherCost(Gathers, !Root, ScalarTy);
       SetVector<Value *> VectorizedLoads;
       SmallVector<std::pair<unsigned, LoadsState>> VectorizedStarts;
       SmallVector<unsigned> ScatterVectorized;
@@ -8040,7 +8055,8 @@ class BoUpSLP::ShuffleCostEstimator : public BaseShuffleAnalysis {
                                              VecTy, Mask, CostKind);
             }
           } else {
-            GatherCost += R.getGatherCost(PointerOps, /*ForPoisonSrc=*/true);
+            GatherCost += R.getGatherCost(PointerOps, /*ForPoisonSrc=*/true,
+                                          PointerOps.front()->getType());
           }
         }
         if (NeedInsertSubvectorAnalysis) {
@@ -8074,18 +8090,19 @@ class BoUpSLP::ShuffleCostEstimator : public BaseShuffleAnalysis {
       transform(VL, ShuffleMask.begin(), [](Value *V) {
         return isa<PoisonValue>(V) ? PoisonMaskElem : 0;
       });
-      InstructionCost InsertCost = TTI.getVectorInstrCost(
-          Instruction::InsertElement, VecTy, CostKind, 0,
-          PoisonValue::get(VecTy), *It);
-      return InsertCost +
-             TTI.getShuffleCost(TargetTransformInfo::SK_Broadcast, VecTy,
-                                ShuffleMask, CostKind, /*Index=*/0,
-                                /*SubTp=*/nullptr, /*Args=*/*It);
+      InstructionCost InsertCost =
+          TTI.getVectorInstrCost(Instruction::InsertElement, VecTy, CostKind, 0,
+                                 PoisonValue::get(VecTy), *It);
+      return InsertCost + TTI.getShuffleCost(TargetTransformInfo::SK_Broadcast,
+                                             VecTy, ShuffleMask, CostKind,
+                                             /*Index=*/0, /*SubTp=*/nullptr,
+                                             /*Args=*/*It);
     }
     return GatherCost +
            (all_of(Gathers, IsaPred<UndefValue>)
                 ? TTI::TCC_Free
-                : R.getGatherCost(Gathers, !Root && VL.equals(Gathers)));
+                : R.getGatherCost(Gathers, !Root && VL.equals(Gathers),
+                                  ScalarTy));
   };
 
   /// Compute the cost of creating a vector containing the extracted values from
@@ -8105,8 +8122,8 @@ class BoUpSLP::ShuffleCostEstimator : public BaseShuffleAnalysis {
             return Sz;
           return std::max(Sz, VecTy->getNumElements());
         });
-    unsigned NumSrcRegs = TTI.getNumberOfParts(
-        FixedVectorType::get(VL.front()->getType(), NumElts));
+    unsigned NumSrcRegs =
+        TTI.getNumberOfParts(FixedVectorType::get(ScalarTy, NumElts));
     if (NumSrcRegs == 0)
       NumSrcRegs = 1;
     // FIXME: this must be moved to TTI for better estimation.
@@ -8152,17 +8169,16 @@ class BoUpSLP::ShuffleCostEstimator : public BaseShuffleAnalysis {
       std::optional<TTI::ShuffleKind> RegShuffleKind =
           CheckPerRegistersShuffle(SubMask);
       if (!RegShuffleKind) {
-        Cost += ::getShuffleCost(
-            TTI, *ShuffleKinds[Part],
-            FixedVectorType::get(VL.front()->getType(), NumElts), MaskSlice);
+        Cost += ::getShuffleCost(TTI, *ShuffleKinds[Part],
+                                 FixedVectorType::get(ScalarTy, NumElts),
+                                 MaskSlice);
         continue;
       }
       if (*RegShuffleKind != TTI::SK_PermuteSingleSrc ||
           !ShuffleVectorInst::isIdentityMask(SubMask, EltsPerVector)) {
-        Cost += ::getShuffleCost(
-            TTI, *RegShuffleKind,
-            FixedVectorType::get(VL.front()->getType(), EltsPerVector),
-            SubMask);
+        Cost += ::getShuffleCost(TTI, *RegShuffleKind,
+                                 FixedVectorType::get(ScalarTy, EltsPerVector),
+                                 SubMask);
       }
     }
     return Cost;
@@ -8279,6 +8295,48 @@ class BoUpSLP::ShuffleCostEstimator : public BaseShuffleAnalysis {
     SmallVector<int> CommonMask(Mask.begin(), Mask.end());
     Value *V1 = P1.dyn_cast<Value *>(), *V2 = P2.dyn_cast<Value *>();
     unsigned CommonVF = Mask.size();
+    InstructionCost ExtraCost = 0;
+    auto GetNodeMinBWAffectedCost = [&](const TreeEntry &E,
+                                        unsigned VF) -> InstructionCost {
+      if (E.State == TreeEntry::NeedToGather && allConstant(E.Scalars))
+        return TTI::TCC_Free;
+      Type *EScalarTy = E.Scalars.front()->getType();
+      bool IsSigned = true;
+      if (auto It = R.MinBWs.find(&E); It != R.MinBWs.end()) {
+        EScalarTy = IntegerType::get(EScalarTy->getContext(), It->second.first);
+        IsSigned = It->second.second;
+      }
+      if (EScalarTy != ScalarTy) {
+        unsigned CastOpcode = Instruction::Trunc;
+        unsigned DstSz = R.DL->getTypeSizeInBits(ScalarTy);
+        unsigned SrcSz = R.DL->getTypeSizeInBits(EScalarTy);
+        if (DstSz > SrcSz)
+          CastOpcode = IsSigned ? Instruction::SExt : Instruction::ZExt;
+        return TTI.getCastInstrCost(CastOpcode,
+                                    FixedVectorType::get(ScalarTy, VF),
+                                    FixedVectorType::get(EScalarTy, VF),
+                                    TTI::CastContextHint::None, CostKind);
+      }
+      return TTI::TCC_Free;
+    };
+    auto GetValueMinBWAffectedCost = [&](const Value *V) -> InstructionCost {
+      if (isa<Constant>(V))
+        return TTI::TCC_Free;
+      auto *VecTy = cast<VectorType>(V->getType());
+      Type *EScalarTy = VecTy->getElementType();
+      if (EScalarTy != ScalarTy) {
+        bool IsSigned = !isKnownNonNegative(V, SimplifyQuery(*R.DL));
+        unsigned CastOpcode = Instruction::Trunc;
+        unsigned DstSz = R.DL->getTypeSizeInBits(ScalarTy);
+        unsigned SrcSz = R.DL->getTypeSizeInBits(EScalarTy);
+        if (DstSz > SrcSz)
+          CastOpcode = IsSigned ? Instruction::SExt : Instruction::ZExt;
+        return TTI.getCastInstrCost(
+            CastOpcode, VectorType::get(ScalarTy, VecTy->getElementCount()),
+            VecTy, TTI::CastContextHint::None, CostKind);
+      }
+      return TTI::TCC_Free;
+    };
     if (!V1 && !V2 && !P2.isNull()) {
       // Shuffle 2 entry nodes.
       const TreeEntry *E = P1.get<const TreeEntry *>();
@@ -8305,11 +8363,14 @@ class BoUpSLP::ShuffleCostEstimator : public BaseShuffleAnalysis {
           }
         }
         CommonVF = E->Scalars.size();
+        ExtraCost += GetNodeMinBWAffectedCost(*E, CommonVF) +
+                     GetNodeMinBWAffectedCost(*E2, CommonVF);
+      } else {
+        ExtraCost += GetNodeMinBWAffectedCost(*E, E->getVectorFactor()) +
+                     GetNodeMinBWAffectedCost(*E2, E2->getVectorFactor());
       }
-      V1 = Constant::getNullValue(
-          FixedVectorType::get(E->Scalars.front()->getType(), CommonVF));
-      V2 = getAllOnesValue(
-          *R.DL, FixedVectorType::get(E->Scalars.front()->getType(), CommonVF));
+      V1 = Constant::getNullValue(FixedVectorType::get(ScalarTy, CommonVF));
+      V2 = getAllOnesValue(*R.DL, FixedVectorType::get(ScalarTy, CommonVF));
     } else if (!V1 && P2.isNull()) {
       // Shuffle single entry node.
       const TreeEntry *E = P1.get<const TreeEntry *>();
@@ -8328,8 +8389,8 @@ class BoUpSLP::ShuffleCostEstimator : public BaseShuffleAnalysis {
         }
         CommonVF = E->Scalars.size();
       }
-      V1 = Constant::getNullValue(
-          FixedVectorType::get(E->Scalars.front()->getType(), CommonVF));
+      ExtraCost += GetNodeMinBWAffectedCost(*E, CommonVF);
+      V1 = Constant::getNullValue(FixedVectorType::get(ScalarTy, CommonVF));
       // Not identity/broadcast? Try to see if the original vector is better.
       if (!E->ReorderIndices.empty() && CommonVF == E->ReorderIndices.size() &&
           CommonVF == CommonMask.size() &&
@@ -8346,6 +8407,7 @@ class BoUpSLP::ShuffleCostEstimator : public BaseShuffleAnalysis {
       }
     } else if (V1 && P2.isNull()) {
       // Shuffle single vector.
+      ExtraCost += GetValueMinBWAffectedCost(V1);
       CommonVF = cast<FixedVectorType>(V1->getType())->getNumElements();
       assert(
           all_of(Mask,
@@ -8372,11 +8434,11 @@ class BoUpSLP::ShuffleCostEstimator : public BaseShuffleAnalysis {
         }
         CommonVF = VF;
       }
-      V1 = Constant::getNullValue(
-          FixedVectorType::get(E2->Scalars.front()->getType(), CommonVF));
-      V2 = getAllOnesValue(
-          *R.DL,
-          FixedVectorType::get(E2->Scalars.front()->getType(), CommonVF));
+      ExtraCost += GetValueMinBWAffectedCost(V1);
+      V1 = Constant::getNullValue(FixedVectorType::get(ScalarTy, CommonVF));
+      ExtraCost += GetNodeMinBWAffectedCost(
+          *E2, std::min(CommonVF, E2->getVectorFactor()));
+      V2 = getAllOnesValue(*R.DL, FixedVectorType::get(ScalarTy, CommonVF));
     } else if (!V1 && V2) {
       // Shuffle vector and tree node.
       unsigned VF = cast<FixedVectorType>(V2->getType())->getNumElements();
@@ -8400,11 +8462,11 @@ class BoUpSLP::ShuffleCostEstimator : public BaseShuffleAnalysis {
         }
         CommonVF = VF;
       }
-      V1 = Constant::getNullValue(
-          FixedVectorType::get(E1->Scalars.front()->getType(), CommonVF));
-      V2 = getAllOnesValue(
-          *R.DL,
-          FixedVectorType::get(E1->Scalars.front()->getType(), CommonVF));
+      ExtraCost += GetNodeMinBWAffectedCost(
+          *E1, std::min(CommonVF, E1->getVectorFactor()));
+      V1 = Constant::getNullValue(FixedVectorType::get(ScalarTy, CommonVF));
+      ExtraCost += GetValueMinBWAffectedCost(V2);
+      V2 = getAllOnesValue(*R.DL, FixedVectorType::get(ScalarTy, CommonVF));
     } else {
       assert(V1 && V2 && "Expected both vectors.");
       unsigned VF = cast<FixedVectorType>(V1->getType())->getNumElements();
@@ -8415,30 +8477,33 @@ class BoUpSLP::ShuffleCostEstimator : public BaseShuffleAnalysis {
                       return Idx < 2 * static_cast<int>(CommonVF);
                     }) &&
              "All elements in mask must be less than 2 * CommonVF.");
+      ExtraCost +=
+          GetValueMinBWAffectedCost(V1) + GetValueMinBWAffectedCost(V2);
       if (V1->getType() != V2->getType()) {
-        V1 = Constant::getNullValue(FixedVectorType::get(
-            cast<FixedVectorType>(V1->getType())->getElementType(), CommonVF));
-        V2 = getAllOnesValue(
-            *R.DL, FixedVectorType::get(
-                       cast<FixedVectorType>(V1->getType())->getElementType(),
-                       CommonVF));
+        V1 = Constant::getNullValue(FixedVectorType::get(ScalarTy, CommonVF));
+        V2 = getAllOnesValue(*R.DL, FixedVectorType::get(ScalarTy, CommonVF));
+      } else {
+        if (cast<VectorType>(V1->getType())->getElementType() != ScalarTy)
+          V1 = Constant::getNullValue(FixedVectorType::get(ScalarTy, CommonVF));
+        if (cast<VectorType>(V2->getType())->getElementType() != ScalarTy)
+          V2 = getAllOnesValue(*R.DL, FixedVectorType::get(ScalarTy, CommonVF));
       }
     }
-    InVectors.front() = Constant::getNullValue(FixedVectorType::get(
-        cast<FixedVectorType>(V1->getType())->getElementType(),
-        CommonMask.size()));
+    InVectors.front() = Constant::getNullValue(
+        FixedVectorType::get(ScalarTy, CommonMask.size()));
     if (InVectors.size() == 2)
       InVectors.pop_back();
-    return BaseShuffleAnalysis::createShuffle<InstructionCost>(
-        V1, V2, CommonMask, Builder);
+    return ExtraCost + BaseShuffleAnalysis::createShuffle<InstructionCost>(
+                           V1, V2, CommonMask, Builder);
   }
 
 public:
-  ShuffleCostEstimator(TargetTransformInfo &TTI,
+  ShuffleCostEstimator(Type *ScalarTy, TargetTransformInfo &TTI,
                        ArrayRef<Value *> VectorizedVals, BoUpSLP &R,
                        SmallPtrSetImpl<Value *> &CheckedExtracts)
-      : TTI(TTI), VectorizedVals(VectorizedVals.begin(), VectorizedVals.end()),
-        R(R), CheckedExtracts(CheckedExtracts) {}
+      : ScalarTy(ScalarTy), TTI(TTI),
+        VectorizedVals(VectorizedVals.begin(), VectorizedVals.end()), R(R),
+        CheckedExtracts(CheckedExtracts) {}
   Value *adjustExtracts(const TreeEntry *E, MutableArrayRef<int> Mask,
                         ArrayRef<std::optional<TTI::ShuffleKind>> ShuffleKinds,
                         unsigned NumParts, bool &UseVecBaseAsInput) {
@@ -8486,6 +8551,12 @@ public:
         const TreeEntry *VE = R.getTreeEntry(V);
         if (!CheckedExtracts.insert(V).second ||
             !R.areAllUsersVectorized(cast<Instruction>(V), &VectorizedVals) ||
+            any_of(EE->users(),
+                   [&](User *U) {
+                     return isa<GetElementPtrInst>(U) &&
+                            !R.areAllUsersVectorized(cast<Instruction>(U),
+                                                     &VectorizedVals);
+                   }) ||
             (VE && VE != E))
           continue;
         std::optional<unsigned> EEIdx = getExtractIndex(EE);
@@ -8528,7 +8599,7 @@ public:
     if (NumParts != 1 && UniqueBases.size() != 1) {
       UseVecBaseAsInput = true;
       VecBase = Constant::getNullValue(
-          FixedVectorType::get(VL.front()->getType(), CommonMask.size()));
+          FixedVectorType::get(ScalarTy, CommonMask.size()));
     }
     return VecBase;
   }
@@ -8556,8 +8627,7 @@ public:
       return;
     }
     assert(!CommonMask.empty() && "Expected non-empty common mask.");
-    auto *MaskVecTy =
-        FixedVectorType::get(E1.Scalars.front()->getType(), Mask.size());
+    auto *MaskVecTy = FixedVectorType::get(ScalarTy, Mask.size());
     unsigned NumParts = TTI.getNumberOfParts(MaskVecTy);
     if (NumParts == 0 || NumParts >= Mask.size())
       NumParts = 1;
@@ -8574,8 +8644,7 @@ public:
       return;
     }
     assert(!CommonMask.empty() && "Expected non-empty common mask.");
-    auto *MaskVecTy =
-        FixedVectorType::get(E1.Scalars.front()->getType(), Mask.size());
+    auto *MaskVecTy = FixedVectorType::get(ScalarTy, Mask.size());
     unsigned NumParts = TTI.getNumberOfParts(MaskVecTy);
     if (NumParts == 0 || NumParts >= Mask.size())
       NumParts = 1;
@@ -8675,7 +8744,7 @@ public:
     return ConstantVector::getSplat(
         ElementCount::getFixed(
             cast<FixedVectorType>(Root->getType())->getNumElements()),
-        getAllOnesValue(*R.DL, VL.front()->getType()));
+        getAllOnesValue(*R.DL, ScalarTy));
   }
   InstructionCost createFreeze(InstructionCost Cost) { return Cost; }
   /// Finalize emission of the shuffles.
@@ -8821,7 +8890,7 @@ BoUpSLP::getEntryCost(const TreeEntry *E, ArrayRef<Value *> VectorizedVals,
     if (isa<InsertElementInst>(VL[0]))
       return InstructionCost::getInvalid();
     return processBuildVector<ShuffleCostEstimator, InstructionCost>(
-        E, *TTI, VectorizedVals, *this, CheckedExtracts);
+        E, ScalarTy, *TTI, VectorizedVals, *this, CheckedExtracts);
   }
   InstructionCost CommonCost = 0;
   SmallVector<int> Mask;
@@ -9587,11 +9656,11 @@ bool BoUpSLP::isLoadCombineReductionCandidate(RecurKind RdxKind) const {
                                     /* MatchOr */ false);
 }
 
-bool BoUpSLP::isLoadCombineCandidate() const {
+bool BoUpSLP::isLoadCombineCandidate(ArrayRef<Value *> Stores) const {
   // Peek through a final sequence of stores and check if all operations are
   // likely to be load-combined.
-  unsigned NumElts = VectorizableTree[0]->Scalars.size();
-  for (Value *Scalar : VectorizableTree[0]->Scalars) {
+  unsigned NumElts = Stores.size();
+  for (Value *Scalar : Stores) {
     Value *X;
     if (!match(Scalar, m_Store(m_Value(X), m_Value())) ||
         !isLoadCombineCandidateImpl(X, NumElts, TTI, /* MatchOr */ true))
@@ -9640,6 +9709,7 @@ bool BoUpSLP::isTreeTinyAndNotFullyVectorizable(bool ForReduction) const {
   bool IsAllowedSingleBVNode =
       VectorizableTree.size() > 1 ||
       (VectorizableTree.size() == 1 && VectorizableTree.front()->getOpcode() &&
+       !VectorizableTree.front()->isAltShuffle() &&
        VectorizableTree.front()->getOpcode() != Instruction::PHI &&
        VectorizableTree.front()->getOpcode() != Instruction::GetElementPtr &&
        allSameBlock(VectorizableTree.front()->Scalars));
@@ -10860,12 +10930,8 @@ BoUpSLP::isGatherShuffledEntry(
   return Res;
 }
 
-InstructionCost BoUpSLP::getGatherCost(ArrayRef<Value *> VL,
-                                       bool ForPoisonSrc) const {
-  // Find the type of the operands in VL.
-  Type *ScalarTy = VL[0]->getType();
-  if (StoreInst *SI = dyn_cast<StoreInst>(VL[0]))
-    ScalarTy = SI->getValueOperand()->getType();
+InstructionCost BoUpSLP::getGatherCost(ArrayRef<Value *> VL, bool ForPoisonSrc,
+                                       Type *ScalarTy) const {
   auto *VecTy = FixedVectorType::get(ScalarTy, VL.size());
   bool DuplicateNonConst = false;
   // Find the cost of inserting/extracting values from the vector.
@@ -10876,6 +10942,11 @@ InstructionCost BoUpSLP::getGatherCost(ArrayRef<Value *> VL,
   constexpr TTI::TargetCostKind CostKind = TTI::TCK_RecipThroughput;
   InstructionCost Cost;
   auto EstimateInsertCost = [&](unsigned I, Value *V) {
+    if (V->getType() != ScalarTy) {
+      Cost += TTI->getCastInstrCost(Instruction::Trunc, ScalarTy, V->getType(),
+                                    TTI::CastContextHint::None, CostKind);
+      V = nullptr;
+    }
     if (!ForPoisonSrc)
       Cost +=
           TTI->getVectorInstrCost(Instruction::InsertElement, VecTy, CostKind,
@@ -11032,7 +11103,10 @@ Instruction &BoUpSLP::getLastInstructionInBundle(const TreeEntry *E) {
                         isUsedOutsideBlock(V);
                }) ||
         (E->State == TreeEntry::NeedToGather && E->Idx == 0 &&
-         all_of(E->Scalars, IsaPred<ExtractElementInst, UndefValue>)))
+         all_of(E->Scalars, [](Value *V) {
+           return isa<ExtractElementInst, UndefValue>(V) ||
+                  areAllOperandsNonInsts(V);
+         })))
       Res.second = FindLastInst();
     else
       Res.second = FindFirstInst();
@@ -11100,7 +11174,7 @@ void BoUpSLP::setInsertPointAfterBundle(const TreeEntry *E) {
   Builder.SetCurrentDebugLocation(Front->getDebugLoc());
 }
 
-Value *BoUpSLP::gather(ArrayRef<Value *> VL, Value *Root) {
+Value *BoUpSLP::gather(ArrayRef<Value *> VL, Value *Root, Type *ScalarTy) {
   // List of instructions/lanes from current block and/or the blocks which are
   // part of the current loop. These instructions will be inserted at the end to
   // make it possible to optimize loops and hoist invariant instructions out of
@@ -11126,14 +11200,11 @@ Value *BoUpSLP::gather(ArrayRef<Value *> VL, Value *Root) {
   auto &&CreateInsertElement = [this](Value *Vec, Value *V, unsigned Pos,
                                       Type *Ty) {
     Value *Scalar = V;
-    if (cast<VectorType>(Vec->getType())->getElementType() != Ty) {
-      assert(V->getType()->isIntegerTy() && Ty->isIntegerTy() &&
+    if (Scalar->getType() != Ty) {
+      assert(Scalar->getType()->isIntegerTy() && Ty->isIntegerTy() &&
              "Expected integer types only.");
-      Vec = Builder.CreateIntCast(
-          Vec,
-          VectorType::get(Ty,
-                          cast<VectorType>(Vec->getType())->getElementCount()),
-          !isKnownNonNegative(Vec, SimplifyQuery(*DL)));
+      Scalar = Builder.CreateIntCast(
+          Scalar, Ty, !isKnownNonNegative(Scalar, SimplifyQuery(*DL)));
     }
 
     Vec = Builder.CreateInsertElement(Vec, Scalar, Builder.getInt32(Pos));
@@ -11161,10 +11232,7 @@ Value *BoUpSLP::gather(ArrayRef<Value *> VL, Value *Root) {
     }
     return Vec;
   };
-  Value *Val0 =
-      isa<StoreInst>(VL[0]) ? cast<StoreInst>(VL[0])->getValueOperand() : VL[0];
-  Type *ScalarTy = Val0->getType();
-  FixedVectorType *VecTy = FixedVectorType::get(ScalarTy, VL.size());
+  auto *VecTy = FixedVectorType::get(ScalarTy, VL.size());
   Value *Vec = Root ? Root : PoisonValue::get(VecTy);
   SmallVector<int> NonConsts;
   // Insert constant values at first.
@@ -11243,6 +11311,7 @@ class BoUpSLP::ShuffleInstructionBuilder final : public BaseShuffleAnalysis {
   /// resulting shuffle and the second operand sets to be the newly added
   /// operand. The \p CommonMask is transformed in the proper way after that.
   SmallVector<Value *, 2> InVectors;
+  Type *ScalarTy = nullptr;
   IRBuilderBase &Builder;
   BoUpSLP &R;
 
@@ -11353,9 +11422,20 @@ class BoUpSLP::ShuffleInstructionBuilder final : public BaseShuffleAnalysis {
         CommonMask[Idx] = Idx;
   }
 
+  /// Cast value \p V to the vector type with the same number of elements, but
+  /// the base type \p ScalarTy.
+  Value *castToScalarTyElem(Value *V) {
+    auto *VecTy = cast<VectorType>(V->getType());
+    if (VecTy->getElementType() == ScalarTy)
+      return V;
+    return Builder.CreateIntCast(
+        V, VectorType::get(ScalarTy, VecTy->getElementCount()),
+        !isKnownNonNegative(V, SimplifyQuery(*R.DL)));
+  }
+
 public:
-  ShuffleInstructionBuilder(IRBuilderBase &Builder, BoUpSLP &R)
-      : Builder(Builder), R(R) {}
+  ShuffleInstructionBuilder(Type *ScalarTy, IRBuilderBase &Builder, BoUpSLP &R)
+      : ScalarTy(ScalarTy), Builder(Builder), R(R) {}
 
   /// Adjusts extractelements after reusing them.
   Value *adjustExtracts(const TreeEntry *E, MutableArrayRef<int> Mask,
@@ -11380,6 +11460,8 @@ public:
           any_of(EI->users(), [&](User *U) {
             const TreeEntry *UTE = R.getTreeEntry(U);
             return !UTE || R.MultiNodeScalars.contains(U) ||
+                   (isa<GetElementPtrInst>(U) &&
+                    !R.areAllUsersVectorized(cast<Instruction>(U))) ||
                    count_if(R.VectorizableTree,
                             [&](const std::unique_ptr<TreeEntry> &TE) {
                               return any_of(TE->UserTreeIndices,
@@ -11392,8 +11474,10 @@ public:
         continue;
       R.eraseInstruction(EI);
     }
-    if (NumParts == 1 || UniqueBases.size() == 1)
+    if (NumParts == 1 || UniqueBases.size() == 1) {
+      VecBase = castToScalarTyElem(VecBase);
       return VecBase;
+    }
     UseVecBaseAsInput = true;
     auto TransformToIdentity = [](MutableArrayRef<int> Mask) {
       for (auto [I, Idx] : enumerate(Mask))
@@ -11430,6 +11514,7 @@ public:
                "Expected vectors of the same size.");
         PrevSize = Size;
 #endif // NDEBUG
+        VecOp = castToScalarTyElem(VecOp);
         Bases[SubMask[I] < Size ? 0 : 1] = VecOp;
       }
       if (!Bases.front())
@@ -11485,10 +11570,10 @@ public:
       return std::nullopt;
     // Postpone gather emission, will be emitted after the end of the
     // process to keep correct order.
-    auto *VecTy = FixedVectorType::get(E->Scalars.front()->getType(),
-                                       E->getVectorFactor());
+    auto *ResVecTy = FixedVectorType::get(ScalarTy, E->getVectorFactor());
     return Builder.CreateAlignedLoad(
-        VecTy, PoisonValue::get(PointerType::getUnqual(VecTy->getContext())),
+        ResVecTy,
+        PoisonValue::get(PointerType::getUnqual(ScalarTy->getContext())),
         MaybeAlign());
   }
   /// Adds 2 input vectors (in form of tree entries) and the mask for their
@@ -11504,6 +11589,8 @@ public:
   /// Adds 2 input vectors and the mask for their shuffling.
   void add(Value *V1, Value *V2, ArrayRef<int> Mask) {
     assert(V1 && V2 && !Mask.empty() && "Expected non-empty input vectors.");
+    V1 = castToScalarTyElem(V1);
+    V2 = castToScalarTyElem(V2);
     if (InVectors.empty()) {
       InVectors.push_back(V1);
       InVectors.push_back(V2);
@@ -11531,6 +11618,7 @@ public:
   }
   /// Adds another one input vector and the mask for the shuffling.
   void add(Value *V1, ArrayRef<int> Mask, bool = false) {
+    V1 = castToScalarTyElem(V1);
     if (InVectors.empty()) {
       if (!isa<FixedVectorType>(V1->getType())) {
         V1 = createShuffle(V1, nullptr, CommonMask);
@@ -11594,7 +11682,7 @@ public:
   }
   Value *gather(ArrayRef<Value *> VL, unsigned MaskVF = 0,
                 Value *Root = nullptr) {
-    return R.gather(VL, Root);
+    return R.gather(VL, Root, ScalarTy);
   }
   Value *createFreeze(Value *V) { return Builder.CreateFreeze(V); }
   /// Finalize emission of the shuffles.
@@ -11694,7 +11782,8 @@ Value *BoUpSLP::vectorizeOperand(TreeEntry *E, unsigned NodeIdx,
     }
     if (IsSameVE) {
       auto FinalShuffle = [&](Value *V, ArrayRef<int> Mask) {
-        ShuffleInstructionBuilder ShuffleBuilder(Builder, *this);
+        ShuffleInstructionBuilder ShuffleBuilder(
+            cast<VectorType>(V->getType())->getElementType(), Builder, *this);
         ShuffleBuilder.add(V, Mask);
         return ShuffleBuilder.finalize(std::nullopt);
       };
@@ -11769,7 +11858,8 @@ Value *BoUpSLP::vectorizeOperand(TreeEntry *E, unsigned NodeIdx,
 }
 
 template <typename BVTy, typename ResTy, typename... Args>
-ResTy BoUpSLP::processBuildVector(const TreeEntry *E, Args &...Params) {
+ResTy BoUpSLP::processBuildVector(const TreeEntry *E, Type *ScalarTy,
+                                  Args &...Params) {
   assert(E->State == TreeEntry::NeedToGather && "Expected gather node.");
   unsigned VF = E->getVectorFactor();
 
@@ -11817,7 +11907,7 @@ ResTy BoUpSLP::processBuildVector(const TreeEntry *E, Args &...Params) {
     }
     return true;
   };
-  BVTy ShuffleBuilder(Params...);
+  BVTy ShuffleBuilder(ScalarTy, Params...);
   ResTy Res = ResTy();
   SmallVector<int> Mask;
   SmallVector<int> ExtractMask(GatheredScalars.size(), PoisonMaskElem);
@@ -11826,7 +11916,7 @@ ResTy BoUpSLP::processBuildVector(const TreeEntry *E, Args &...Params) {
   bool UseVecBaseAsInput = false;
   SmallVector<std::optional<TargetTransformInfo::ShuffleKind>> GatherShuffles;
   SmallVector<SmallVector<const TreeEntry *>> Entries;
-  Type *ScalarTy = GatheredScalars.front()->getType();
+  Type *OrigScalarTy = GatheredScalars.front()->getType();
   auto *VecTy = FixedVectorType::get(ScalarTy, GatheredScalars.size());
   unsigned NumParts = TTI->getNumberOfParts(VecTy);
   if (NumParts == 0 || NumParts >= GatheredScalars.size())
@@ -11861,7 +11951,7 @@ ResTy BoUpSLP::processBuildVector(const TreeEntry *E, Args &...Params) {
               GatheredScalars.size() != VF) {
             Resized = true;
             GatheredScalars.append(VF - GatheredScalars.size(),
-                                   PoisonValue::get(ScalarTy));
+                                   PoisonValue::get(OrigScalarTy));
           }
       }
     }
@@ -11921,12 +12011,12 @@ ResTy BoUpSLP::processBuildVector(const TreeEntry *E, Args &...Params) {
               });
             }))
           GatheredScalars.append(VF - GatheredScalars.size(),
-                                 PoisonValue::get(ScalarTy));
+                                 PoisonValue::get(OrigScalarTy));
       }
       // Remove shuffled elements from list of gathers.
       for (int I = 0, Sz = Mask.size(); I < Sz; ++I) {
         if (Mask[I] != PoisonMaskElem)
-          GatheredScalars[I] = PoisonValue::get(ScalarTy);
+          GatheredScalars[I] = PoisonValue::get(OrigScalarTy);
       }
     }
   }
@@ -11937,7 +12027,7 @@ ResTy BoUpSLP::processBuildVector(const TreeEntry *E, Args &...Params) {
     // such sequences.
     bool IsSplat = IsRootPoison && isSplat(Scalars) &&
                    (Scalars.size() > 2 || Scalars.front() == Scalars.back());
-    Scalars.append(VF - Scalars.size(), PoisonValue::get(ScalarTy));
+    Scalars.append(VF - Scalars.size(), PoisonValue::get(OrigScalarTy));
     SmallVector<int> UndefPos;
     DenseMap<Value *, unsigned> UniquePositions;
     // Gather unique non-const values and all constant values.
@@ -11959,7 +12049,7 @@ ResTy BoUpSLP::processBuildVector(const TreeEntry *E, Args &...Params) {
       ++NumNonConsts;
       SinglePos = I;
       Value *OrigV = V;
-      Scalars[I] = PoisonValue::get(ScalarTy);
+      Scalars[I] = PoisonValue::get(OrigScalarTy);
       if (IsSplat) {
         Scalars.front() = OrigV;
         ReuseMask[I] = 0;
@@ -11975,7 +12065,7 @@ ResTy BoUpSLP::processBuildVector(const TreeEntry *E, Args &...Params) {
         ReuseMask.assign(VF, PoisonMaskElem);
         std::swap(Scalars.front(), Scalars[SinglePos]);
         if (!UndefPos.empty() && UndefPos.front() == 0)
-          Scalars.front() = UndefValue::get(ScalarTy);
+          Scalars.front() = UndefValue::get(OrigScalarTy);
       }
       ReuseMask[SinglePos] = SinglePos;
     } else if (!UndefPos.empty() && IsSplat) {
@@ -12005,7 +12095,7 @@ ResTy BoUpSLP::processBuildVector(const TreeEntry *E, Args &...Params) {
           // Replace the undef by the poison, in the mask it is replaced by
           // non-poisoned scalar already.
           if (I != Pos)
-            Scalars[I] = PoisonValue::get(ScalarTy);
+            Scalars[I] = PoisonValue::get(OrigScalarTy);
         }
       } else {
         // Replace undefs by the poisons, emit broadcast and then emit
@@ -12013,7 +12103,7 @@ ResTy BoUpSLP::processBuildVector(const TreeEntry *E, Args &...Params) {
         for (int I : UndefPos) {
           ReuseMask[I] = PoisonMaskElem;
           if (isa<UndefValue>(Scalars[I]))
-            Scalars[I] = PoisonValue::get(ScalarTy);
+            Scalars[I] = PoisonValue::get(OrigScalarTy);
         }
         NeedFreeze = true;
       }
@@ -12068,9 +12158,8 @@ ResTy BoUpSLP::processBuildVector(const TreeEntry *E, Args &...Params) {
         IsNonPoisoned &= isGuaranteedNotToBePoison(Vec1);
       } else {
         IsUsedInExpr = false;
-        ShuffleBuilder.add(PoisonValue::get(FixedVectorType::get(
-                               ScalarTy, GatheredScalars.size())),
-                           ExtractMask, /*ForExtracts=*/true);
+        ShuffleBuilder.add(PoisonValue::get(VecTy), ExtractMask,
+                           /*ForExtracts=*/true);
       }
     }
     if (!GatherShuffles.empty()) {
@@ -12151,9 +12240,9 @@ ResTy BoUpSLP::processBuildVector(const TreeEntry *E, Args &...Params) {
     // contains only constant to build final vector and then shuffle.
     for (int I = 0, Sz = GatheredScalars.size(); I < Sz; ++I) {
       if (EnoughConstsForShuffle && isa<Constant>(GatheredScalars[I]))
-        NonConstants[I] = PoisonValue::get(ScalarTy);
+        NonConstants[I] = PoisonValue::get(OrigScalarTy);
       else
-        GatheredScalars[I] = PoisonValue::get(ScalarTy);
+        GatheredScalars[I] = PoisonValue::get(OrigScalarTy);
     }
     // Generate constants for final shuffle and build a mask for them.
     if (!all_of(GatheredScalars, IsaPred<PoisonValue>)) {
@@ -12199,9 +12288,9 @@ ResTy BoUpSLP::processBuildVector(const TreeEntry *E, Args &...Params) {
   return Res;
 }
 
-Value *BoUpSLP::createBuildVector(const TreeEntry *E) {
-  return processBuildVector<ShuffleInstructionBuilder, Value *>(E, Builder,
-                                                                *this);
+Value *BoUpSLP::createBuildVector(const TreeEntry *E, Type *ScalarTy) {
+  return processBuildVector<ShuffleInstructionBuilder, Value *>(E, ScalarTy,
+                                                                Builder, *this);
 }
 
 Value *BoUpSLP::vectorizeTree(TreeEntry *E, bool PostponedPHIs) {
@@ -12214,18 +12303,28 @@ Value *BoUpSLP::vectorizeTree(TreeEntry *E, bool PostponedPHIs) {
     return E->VectorizedValue;
   }
 
+  Value *V = E->Scalars.front();
+  Type *ScalarTy = V->getType();
+  if (auto *Store = dyn_cast<StoreInst>(V))
+    ScalarTy = Store->getValueOperand()->getType();
+  else if (auto *IE = dyn_cast<InsertElementInst>(V))
+    ScalarTy = IE->getOperand(1)->getType();
+  auto It = MinBWs.find(E);
+  if (It != MinBWs.end())
+    ScalarTy = IntegerType::get(F->getContext(), It->second.first);
+  auto *VecTy = FixedVectorType::get(ScalarTy, E->Scalars.size());
   if (E->State == TreeEntry::NeedToGather) {
     // Set insert point for non-reduction initial nodes.
     if (E->getMainOp() && E->Idx == 0 && !UserIgnoreList)
       setInsertPointAfterBundle(E);
-    Value *Vec = createBuildVector(E);
+    Value *Vec = createBuildVector(E, ScalarTy);
     E->VectorizedValue = Vec;
     return Vec;
   }
 
   bool IsReverseOrder = isReverseOrder(E->ReorderIndices);
   auto FinalShuffle = [&](Value *V, const TreeEntry *E, VectorType *VecTy) {
-    ShuffleInstructionBuilder ShuffleBuilder(Builder, *this);
+    ShuffleInstructionBuilder ShuffleBuilder(ScalarTy, Builder, *this);
     if (E->getOpcode() == Instruction::Store) {
       ArrayRef<int> Mask =
           ArrayRef(reinterpret_cast<const int *>(E->ReorderIndices.begin()),
@@ -12246,14 +12345,6 @@ Value *BoUpSLP::vectorizeTree(TreeEntry *E, bool PostponedPHIs) {
   unsigned ShuffleOrOp =
       E->isAltShuffle() ? (unsigned)Instruction::ShuffleVector : E->getOpcode();
   Instruction *VL0 = E->getMainOp();
-  Type *ScalarTy = VL0->getType();
-  if (auto *Store = dyn_cast<StoreInst>(VL0))
-    ScalarTy = Store->getValueOperand()->getType();
-  else if (auto *IE = dyn_cast<InsertElementInst>(VL0))
-    ScalarTy = IE->getOperand(1)->getType();
-  auto It = MinBWs.find(E);
-  if (It != MinBWs.end())
-    ScalarTy = IntegerType::get(F->getContext(), It->second.first);
   auto GetOperandSignedness = [&](unsigned Idx) {
     const TreeEntry *OpE = getOperandEntry(E, Idx);
     bool IsSigned = false;
@@ -12266,7 +12357,6 @@ Value *BoUpSLP::vectorizeTree(TreeEntry *E, bool PostponedPHIs) {
       });
     return IsSigned;
   };
-  auto *VecTy = FixedVectorType::get(ScalarTy, E->Scalars.size());
   switch (ShuffleOrOp) {
     case Instruction::PHI: {
       assert((E->ReorderIndices.empty() || !E->ReuseShuffleIndices.empty() ||
@@ -13141,7 +13231,8 @@ Value *BoUpSLP::vectorizeTree(
     auto *TE = const_cast<TreeEntry *>(E);
     if (auto *VecTE = getTreeEntry(TE->Scalars.front()))
       if (VecTE->isSame(TE->UserTreeIndices.front().UserTE->getOperand(
-              TE->UserTreeIndices.front().EdgeIdx)))
+              TE->UserTreeIndices.front().EdgeIdx)) &&
+          VecTE->isSame(TE->Scalars))
         // Found gather node which is absolutely the same as one of the
         // vectorized nodes. It may happen after reordering.
         continue;
@@ -13312,8 +13403,11 @@ Value *BoUpSLP::vectorizeTree(
             // Leave the GEPs as is, they are free in most cases and better to
             // keep them as GEPs.
             auto *CloneGEP = GEP->clone();
-            CloneGEP->insertBefore(*Builder.GetInsertBlock(),
-                                   Builder.GetInsertPoint());
+            if (isa<Instruction>(Vec))
+              CloneGEP->insertBefore(*Builder.GetInsertBlock(),
+                                     Builder.GetInsertPoint());
+            else
+              CloneGEP->insertBefore(GEP);
             if (GEP->hasName())
               CloneGEP->takeName(GEP);
             Ex = CloneGEP;
@@ -13516,7 +13610,8 @@ Value *BoUpSLP::vectorizeTree(
       else
         CombinedMask2[I] = Mask[I] - VF;
     }
-    ShuffleInstructionBuilder ShuffleBuilder(Builder, *this);
+    ShuffleInstructionBuilder ShuffleBuilder(
+        cast<VectorType>(V1->getType())->getElementType(), Builder, *this);
     ShuffleBuilder.add(V1, CombinedMask1);
     if (V2)
       ShuffleBuilder.add(V2, CombinedMask2);
@@ -14551,13 +14646,27 @@ bool BoUpSLP::collectValuesToDemote(
       return false;
     bool Res = all_of(
         E.Scalars, std::bind(IsPotentiallyTruncated, _1, std::ref(BitWidth)));
-    // Gather demoted constant operands.
-    if (Res && E.State == TreeEntry::NeedToGather &&
-        all_of(E.Scalars, IsaPred<Constant>))
-      ToDemote.push_back(E.Idx);
+    // Demote gathers.
+    if (Res && E.State == TreeEntry::NeedToGather) {
+      // Check possible extractelement instructions bases and final vector
+      // length.
+      SmallPtrSet<Value *, 4> UniqueBases;
+      for (Value *V : E.Scalars) {
+        auto *EE = dyn_cast<ExtractElementInst>(V);
+        if (!EE)
+          continue;
+        UniqueBases.insert(EE->getVectorOperand());
+      }
+      const unsigned VF = E.Scalars.size();
+      Type *OrigScalarTy = E.Scalars.front()->getType();
+      if (UniqueBases.size() <= 2 ||
+          TTI->getNumberOfParts(FixedVectorType::get(OrigScalarTy, VF)) ==
+              TTI->getNumberOfParts(FixedVectorType::get(
+                  IntegerType::get(OrigScalarTy->getContext(), BitWidth), VF)))
+        ToDemote.push_back(E.Idx);
+    }
     return Res;
   };
-  // TODO: improve handling of gathered values and others.
   if (E.State == TreeEntry::NeedToGather || !Visited.insert(&E).second ||
       any_of(E.Scalars, [&](Value *V) {
         return all_of(V->users(), [&](User *U) {
@@ -15068,11 +15177,16 @@ void BoUpSLP::computeMinimumValueSizes() {
       IsSignedCmp =
           NodeIdx < VectorizableTree.size() &&
           any_of(VectorizableTree[NodeIdx]->UserTreeIndices,
-                 [](const EdgeInfo &EI) {
+                 [&](const EdgeInfo &EI) {
                    return EI.UserTE->getOpcode() == Instruction::ICmp &&
-                          any_of(EI.UserTE->Scalars, [](Value *V) {
+                          any_of(EI.UserTE->Scalars, [&](Value *V) {
                             auto *IC = dyn_cast<ICmpInst>(V);
-                            return IC && IC->isSigned();
+                            return IC &&
+                                   (IC->isSigned() ||
+                                    !isKnownNonNegative(IC->getOperand(0),
+                                                        SimplifyQuery(*DL)) ||
+                                    !isKnownNonNegative(IC->getOperand(1),
+                                                        SimplifyQuery(*DL)));
                           });
                  });
     }
@@ -15201,8 +15315,11 @@ bool SLPVectorizerPass::runImpl(Function &F, ScalarEvolution *SE_,
   return Changed;
 }
 
-bool SLPVectorizerPass::vectorizeStoreChain(ArrayRef<Value *> Chain, BoUpSLP &R,
-                                            unsigned Idx, unsigned MinVF) {
+std::optional<bool>
+SLPVectorizerPass::vectorizeStoreChain(ArrayRef<Value *> Chain, BoUpSLP &R,
+                                       unsigned Idx, unsigned MinVF,
+                                       unsigned &Size) {
+  Size = 0;
   LLVM_DEBUG(dbgs() << "SLP: Analyzing a store chain of length " << Chain.size()
                     << "\n");
   const unsigned Sz = R.getVectorElementSize(Chain[0]);
@@ -15219,11 +15336,42 @@ bool SLPVectorizerPass::vectorizeStoreChain(ArrayRef<Value *> Chain, BoUpSLP &R,
   LLVM_DEBUG(dbgs() << "SLP: Analyzing " << VF << " stores at offset " << Idx
                     << "\n");
 
+  SetVector<Value *> ValOps;
+  for (Value *V : Chain)
+    ValOps.insert(cast<StoreInst>(V)->getValueOperand());
+  // Operands are not same/alt opcodes or non-power-of-2 uniques - exit.
+  InstructionsState S = getSameOpcode(ValOps.getArrayRef(), *TLI);
+  if (all_of(ValOps, IsaPred<Instruction>) && ValOps.size() > 1) {
+    DenseSet<Value *> Stores(Chain.begin(), Chain.end());
+    bool IsPowerOf2 =
+        isPowerOf2_32(ValOps.size()) ||
+        (VectorizeNonPowerOf2 && isPowerOf2_32(ValOps.size() + 1));
+    if ((!IsPowerOf2 && S.getOpcode() && S.getOpcode() != Instruction::Load &&
+         (!S.MainOp->isSafeToRemove() ||
+          any_of(ValOps.getArrayRef(),
+                 [&](Value *V) {
+                   return !isa<ExtractElementInst>(V) &&
+                          (V->getNumUses() > Chain.size() ||
+                           any_of(V->users(), [&](User *U) {
+                             return !Stores.contains(U);
+                           }));
+                 }))) ||
+        (ValOps.size() > Chain.size() / 2 && !S.getOpcode())) {
+      Size = (!IsPowerOf2 && S.getOpcode()) ? 1 : 2;
+      return false;
+    }
+  }
+  if (R.isLoadCombineCandidate(Chain))
+    return true;
   R.buildTree(Chain);
-  if (R.isTreeTinyAndNotFullyVectorizable())
-    return false;
-  if (R.isLoadCombineCandidate())
+  // Check if tree tiny and store itself or its value is not vectorized.
+  if (R.isTreeTinyAndNotFullyVectorizable()) {
+    if (R.isGathered(Chain.front()) ||
+        R.isNotScheduled(cast<StoreInst>(Chain.front())->getValueOperand()))
+      return std::nullopt;
+    Size = R.getTreeSize();
     return false;
+  }
   R.reorderTopToBottom();
   R.reorderBottomToTop();
   R.buildExternalUses();
@@ -15231,6 +15379,9 @@ bool SLPVectorizerPass::vectorizeStoreChain(ArrayRef<Value *> Chain, BoUpSLP &R,
   R.computeMinimumValueSizes();
   R.transformNodes();
 
+  Size = R.getTreeSize();
+  if (S.getOpcode() == Instruction::Load)
+    Size = 2; // cut off masked gather small trees
   InstructionCost Cost = R.getTreeCost();
 
   LLVM_DEBUG(dbgs() << "SLP: Found cost = " << Cost << " for VF=" << VF << "\n");
@@ -15252,17 +15403,45 @@ bool SLPVectorizerPass::vectorizeStoreChain(ArrayRef<Value *> Chain, BoUpSLP &R,
   return false;
 }
 
-bool SLPVectorizerPass::vectorizeStores(ArrayRef<StoreInst *> Stores,
-                                        BoUpSLP &R) {
+/// Checks if the quadratic mean deviation is less than 90% of the mean size.
+static bool checkTreeSizes(ArrayRef<std::pair<unsigned, unsigned>> Sizes,
+                           bool First) {
+  unsigned Num = 0;
+  uint64_t Sum = std::accumulate(
+      Sizes.begin(), Sizes.end(), static_cast<uint64_t>(0),
+      [&](uint64_t V, const std::pair<unsigned, unsigned> &Val) {
+        unsigned Size = First ? Val.first : Val.second;
+        if (Size == 1)
+          return V;
+        ++Num;
+        return V + Size;
+      });
+  if (Num == 0)
+    return true;
+  uint64_t Mean = Sum / Num;
+  if (Mean == 0)
+    return true;
+  uint64_t Dev = std::accumulate(
+                     Sizes.begin(), Sizes.end(), static_cast<uint64_t>(0),
+                     [&](uint64_t V, const std::pair<unsigned, unsigned> &Val) {
+                       unsigned P = First ? Val.first : Val.second;
+                       if (P == 1)
+                         return V;
+                       return V + (P - Mean) * (P - Mean);
+                     }) /
+                 Num;
+  return Dev * 81 / (Mean * Mean) == 0;
+}
+
+bool SLPVectorizerPass::vectorizeStores(
+    ArrayRef<StoreInst *> Stores, BoUpSLP &R,
+    DenseSet<std::tuple<Value *, Value *, Value *, Value *, unsigned>>
+        &Visited) {
   // We may run into multiple chains that merge into a single chain. We mark the
   // stores that we vectorized so that we don't visit the same store twice.
   BoUpSLP::ValueSet VectorizedStores;
   bool Changed = false;
 
-  // Stores the pair of stores (first_store, last_store) in a range, that were
-  // already tried to be vectorized. Allows to skip the store ranges that were
-  // already tried to be vectorized but the attempts were unsuccessful.
-  DenseSet<std::pair<Value *, Value *>> TriedSequences;
   struct StoreDistCompare {
     bool operator()(const std::pair<unsigned, int> &Op1,
                     const std::pair<unsigned, int> &Op2) const {
@@ -15290,7 +15469,14 @@ bool SLPVectorizerPass::vectorizeStores(ArrayRef<StoreInst *> Stores,
         PrevDist = DataVar.second;
       });
 
-      if (Operands.size() <= 1)
+      if (Operands.size() <= 1 ||
+          !Visited
+               .insert({Operands.front(),
+                        cast<StoreInst>(Operands.front())->getValueOperand(),
+                        Operands.back(),
+                        cast<StoreInst>(Operands.back())->getValueOperand(),
+                        Operands.size()})
+               .second)
         continue;
 
       unsigned MaxVecRegSize = R.getMaxVecRegSize();
@@ -15299,13 +15485,19 @@ bool SLPVectorizerPass::vectorizeStores(ArrayRef<StoreInst *> Stores,
 
       unsigned MaxVF =
           std::min(R.getMaximumVF(EltSize, Instruction::Store), MaxElts);
+      unsigned MaxRegVF = MaxVF;
       auto *Store = cast<StoreInst>(Operands[0]);
       Type *StoreTy = Store->getValueOperand()->getType();
       Type *ValueTy = StoreTy;
       if (auto *Trunc = dyn_cast<TruncInst>(Store->getValueOperand()))
         ValueTy = Trunc->getSrcTy();
-      unsigned MinVF = PowerOf2Ceil(TTI->getStoreMinimumVF(
-          R.getMinVF(DL->getTypeStoreSizeInBits(StoreTy)), StoreTy, ValueTy));
+      if (ValueTy == StoreTy &&
+          R.getVectorElementSize(Store->getValueOperand()) <= EltSize)
+        MaxVF = std::min<unsigned>(MaxVF, bit_floor(Operands.size()));
+      unsigned MinVF = std::max<unsigned>(
+          2, PowerOf2Ceil(TTI->getStoreMinimumVF(
+                 R.getMinVF(DL->getTypeStoreSizeInBits(StoreTy)), StoreTy,
+                 ValueTy)));
 
       if (MaxVF < MinVF) {
         LLVM_DEBUG(dbgs() << "SLP: Vectorization infeasible as MaxVF (" << MaxVF
@@ -15320,7 +15512,7 @@ bool SLPVectorizerPass::vectorizeStores(ArrayRef<StoreInst *> Stores,
         // consider cases where VF + 1 is a power-of-2, i.e. almost all vector
         // lanes are used.
         unsigned CandVF = Operands.size();
-        if (isPowerOf2_32(CandVF + 1) && CandVF <= MaxVF)
+        if (isPowerOf2_32(CandVF + 1) && CandVF <= MaxRegVF)
           NonPowerOf2VF = CandVF;
       }
 
@@ -15331,40 +15523,184 @@ bool SLPVectorizerPass::vectorizeStores(ArrayRef<StoreInst *> Stores,
         VF = Size > MaxVF ? NonPowerOf2VF : Size;
         Size *= 2;
       });
-      unsigned StartIdx = 0;
-      for (unsigned Size : CandidateVFs) {
-        for (unsigned Cnt = StartIdx, E = Operands.size(); Cnt + Size <= E;) {
-          ArrayRef<Value *> Slice = ArrayRef(Operands).slice(Cnt, Size);
-          assert(
-              all_of(
-                  Slice,
-                  [&](Value *V) {
-                    return cast<StoreInst>(V)->getValueOperand()->getType() ==
-                           cast<StoreInst>(Slice.front())
-                               ->getValueOperand()
-                               ->getType();
-                  }) &&
-              "Expected all operands of same type.");
-          if (!VectorizedStores.count(Slice.front()) &&
-              !VectorizedStores.count(Slice.back()) &&
-              TriedSequences.insert(std::make_pair(Slice.front(), Slice.back()))
-                  .second &&
-              vectorizeStoreChain(Slice, R, Cnt, MinVF)) {
-            // Mark the vectorized stores so that we don't vectorize them again.
-            VectorizedStores.insert(Slice.begin(), Slice.end());
-            Changed = true;
-            // If we vectorized initial block, no need to try to vectorize it
-            // again.
-            if (Cnt == StartIdx)
-              StartIdx += Size;
-            Cnt += Size;
-            continue;
+      unsigned End = Operands.size();
+      unsigned Repeat = 0;
+      constexpr unsigned MaxAttempts = 4;
+      OwningArrayRef<std::pair<unsigned, unsigned>> RangeSizes(Operands.size());
+      for_each(RangeSizes, [](std::pair<unsigned, unsigned> &P) {
+        P.first = P.second = 1;
+      });
+      DenseMap<Value *, std::pair<unsigned, unsigned>> NonSchedulable;
+      auto IsNotVectorized = [](bool First,
+                                const std::pair<unsigned, unsigned> &P) {
+        return First ? P.first > 0 : P.second > 0;
+      };
+      auto IsVectorized = [](bool First,
+                             const std::pair<unsigned, unsigned> &P) {
+        return First ? P.first == 0 : P.second == 0;
+      };
+      auto VFIsProfitable = [](bool First, unsigned Size,
+                               const std::pair<unsigned, unsigned> &P) {
+        return First ? Size >= P.first : Size >= P.second;
+      };
+      auto FirstSizeSame = [](unsigned Size,
+                              const std::pair<unsigned, unsigned> &P) {
+        return Size == P.first;
+      };
+      while (true) {
+        ++Repeat;
+        bool RepeatChanged = false;
+        bool AnyProfitableGraph;
+        for (unsigned Size : CandidateVFs) {
+          AnyProfitableGraph = false;
+          unsigned StartIdx = std::distance(
+              RangeSizes.begin(),
+              find_if(RangeSizes, std::bind(IsNotVectorized, Size >= MaxRegVF,
+                                            std::placeholders::_1)));
+          while (StartIdx < End) {
+            unsigned EndIdx =
+                std::distance(RangeSizes.begin(),
+                              find_if(RangeSizes.drop_front(StartIdx),
+                                      std::bind(IsVectorized, Size >= MaxRegVF,
+                                                std::placeholders::_1)));
+            unsigned Sz = EndIdx >= End ? End : EndIdx;
+            for (unsigned Cnt = StartIdx; Cnt + Size <= Sz;) {
+              if (!checkTreeSizes(RangeSizes.slice(Cnt, Size),
+                                  Size >= MaxRegVF)) {
+                ++Cnt;
+                continue;
+              }
+              ArrayRef<Value *> Slice = ArrayRef(Operands).slice(Cnt, Size);
+              assert(all_of(Slice,
+                            [&](Value *V) {
+                              return cast<StoreInst>(V)
+                                         ->getValueOperand()
+                                         ->getType() ==
+                                     cast<StoreInst>(Slice.front())
+                                         ->getValueOperand()
+                                         ->getType();
+                            }) &&
+                     "Expected all operands of same type.");
+              if (!NonSchedulable.empty()) {
+                auto [NonSchedSizeMax, NonSchedSizeMin] =
+                    NonSchedulable.lookup(Slice.front());
+                if (NonSchedSizeMax > 0 && NonSchedSizeMin <= Size) {
+                  Cnt += NonSchedSizeMax;
+                  continue;
+                }
+              }
+              unsigned TreeSize;
+              std::optional<bool> Res =
+                  vectorizeStoreChain(Slice, R, Cnt, MinVF, TreeSize);
+              if (!Res) {
+                NonSchedulable
+                    .try_emplace(Slice.front(), std::make_pair(Size, Size))
+                    .first->getSecond()
+                    .second = Size;
+              } else if (*Res) {
+                // Mark the vectorized stores so that we don't vectorize them
+                // again.
+                VectorizedStores.insert(Slice.begin(), Slice.end());
+                // Mark the vectorized stores so that we don't vectorize them
+                // again.
+                AnyProfitableGraph = RepeatChanged = Changed = true;
+                // If we vectorized initial block, no need to try to vectorize
+                // it again.
+                for_each(RangeSizes.slice(Cnt, Size),
+                         [](std::pair<unsigned, unsigned> &P) {
+                           P.first = P.second = 0;
+                         });
+                if (Cnt < StartIdx + MinVF) {
+                  for_each(RangeSizes.slice(StartIdx, Cnt - StartIdx),
+                           [](std::pair<unsigned, unsigned> &P) {
+                             P.first = P.second = 0;
+                           });
+                  StartIdx = Cnt + Size;
+                }
+                if (Cnt > Sz - Size - MinVF) {
+                  for_each(RangeSizes.slice(Cnt + Size, Sz - (Cnt + Size)),
+                           [](std::pair<unsigned, unsigned> &P) {
+                             P.first = P.second = 0;
+                           });
+                  if (Sz == End)
+                    End = Cnt;
+                  Sz = Cnt;
+                }
+                Cnt += Size;
+                continue;
+              }
+              if (Size > 2 && Res &&
+                  !all_of(RangeSizes.slice(Cnt, Size),
+                          std::bind(VFIsProfitable, Size >= MaxRegVF, TreeSize,
+                                    std::placeholders::_1))) {
+                Cnt += Size;
+                continue;
+              }
+              // Check for the very big VFs that we're not rebuilding same
+              // trees, just with larger number of elements.
+              if (Size > MaxRegVF && TreeSize > 1 &&
+                  all_of(RangeSizes.slice(Cnt, Size),
+                         std::bind(FirstSizeSame, TreeSize,
+                                   std::placeholders::_1))) {
+                Cnt += Size;
+                while (Cnt != Sz && RangeSizes[Cnt].first == TreeSize)
+                  ++Cnt;
+                continue;
+              }
+              if (TreeSize > 1)
+                for_each(RangeSizes.slice(Cnt, Size),
+                         [&](std::pair<unsigned, unsigned> &P) {
+                           if (Size >= MaxRegVF)
+                             P.second = std::max(P.second, TreeSize);
+                           else
+                             P.first = std::max(P.first, TreeSize);
+                         });
+              ++Cnt;
+              AnyProfitableGraph = true;
+            }
+            if (StartIdx >= End)
+              break;
+            if (Sz - StartIdx < Size && Sz - StartIdx >= MinVF)
+              AnyProfitableGraph = true;
+            StartIdx = std::distance(
+                RangeSizes.begin(),
+                find_if(RangeSizes.drop_front(Sz),
+                        std::bind(IsNotVectorized, Size >= MaxRegVF,
+                                  std::placeholders::_1)));
           }
-          ++Cnt;
+          if (!AnyProfitableGraph && Size >= MaxRegVF)
+            break;
         }
-        // Check if the whole array was vectorized already - exit.
-        if (StartIdx >= Operands.size())
+        // All values vectorized - exit.
+        if (all_of(RangeSizes, [](const std::pair<unsigned, unsigned> &P) {
+              return P.first == 0 && P.second == 0;
+            }))
+          break;
+        // Check if tried all attempts or no need for the last attempts at all.
+        if (Repeat >= MaxAttempts ||
+            (Repeat > 1 && (RepeatChanged || !AnyProfitableGraph)))
           break;
+        constexpr unsigned StoresLimit = 64;
+        const unsigned MaxTotalNum = bit_floor(std::min<unsigned>(
+            Operands.size(),
+            static_cast<unsigned>(
+                End -
+                std::distance(
+                    RangeSizes.begin(),
+                    find_if(RangeSizes, std::bind(IsNotVectorized, true,
+                                                  std::placeholders::_1))) +
+                1)));
+        unsigned VF = PowerOf2Ceil(CandidateVFs.front()) * 2;
+        if (VF > MaxTotalNum || VF >= StoresLimit)
+          break;
+        for_each(RangeSizes, [&](std::pair<unsigned, unsigned> &P) {
+          if (P.first != 0)
+            P.first = std::max(P.second, P.first);
+        });
+        // Last attempt to vectorize max number of elements, if all previous
+        // attempts were unsuccessful because of the cost issues.
+        CandidateVFs.clear();
+        CandidateVFs.push_back(VF);
       }
     }
   };
@@ -18182,6 +18518,7 @@ bool SLPVectorizerPass::vectorizeStoreChains(BoUpSLP &R) {
   };
 
   // Attempt to sort and vectorize each of the store-groups.
+  DenseSet<std::tuple<Value *, Value *, Value *, Value *, unsigned>> Attempted;
   for (auto &Pair : Stores) {
     if (Pair.second.size() < 2)
       continue;
@@ -18199,8 +18536,8 @@ bool SLPVectorizerPass::vectorizeStoreChains(BoUpSLP &R) {
                                             Pair.second.rend());
     Changed |= tryToVectorizeSequence<StoreInst>(
         ReversedStores, StoreSorter, AreCompatibleStores,
-        [this, &R](ArrayRef<StoreInst *> Candidates, bool) {
-          return vectorizeStores(Candidates, R);
+        [&](ArrayRef<StoreInst *> Candidates, bool) {
+          return vectorizeStores(Candidates, R, Attempted);
         },
         /*MaxVFOnly=*/false, R);
   }
diff --git a/llvm/lib/Transforms/Vectorize/VPlan.h b/llvm/lib/Transforms/Vectorize/VPlan.h
index c74329a0bcc4..71387bf5b7e9 100644
--- a/llvm/lib/Transforms/Vectorize/VPlan.h
+++ b/llvm/lib/Transforms/Vectorize/VPlan.h
@@ -3215,6 +3215,9 @@ public:
     return Value2VPValue[V];
   }
 
+  /// Return the live-in VPValue for \p V, if there is one or nullptr otherwise.
+  VPValue *getLiveIn(Value *V) const { return Value2VPValue.lookup(V); }
+
 #if !defined(NDEBUG) || defined(LLVM_ENABLE_DUMP)
   /// Print the live-ins of this VPlan to \p O.
   void printLiveIns(raw_ostream &O) const;
@@ -3278,8 +3281,8 @@ public:
 private:
   /// Add to the given dominator tree the header block and every new basic block
   /// that was created between it and the latch block, inclusive.
-  static void updateDominatorTree(DominatorTree *DT, BasicBlock *LoopLatchBB,
-                                  BasicBlock *LoopPreHeaderBB,
+  static void updateDominatorTree(DominatorTree *DT, BasicBlock *LoopHeaderBB,
+                                  BasicBlock *LoopLatchBB,
                                   BasicBlock *LoopExitBB);
 };
 
diff --git a/llvm/lib/Transforms/Vectorize/VPlanAnalysis.cpp b/llvm/lib/Transforms/Vectorize/VPlanAnalysis.cpp
index ad4ea648cd61..5f93339083f0 100644
--- a/llvm/lib/Transforms/Vectorize/VPlanAnalysis.cpp
+++ b/llvm/lib/Transforms/Vectorize/VPlanAnalysis.cpp
@@ -35,6 +35,7 @@ Type *VPTypeAnalysis::inferScalarTypeForRecipe(const VPInstruction *R) {
     CachedTypes[OtherV] = ResTy;
     return ResTy;
   }
+  case Instruction::Or:
   case Instruction::ICmp:
   case VPInstruction::FirstOrderRecurrenceSplice: {
     Type *ResTy = inferScalarType(R->getOperand(0));
@@ -44,6 +45,12 @@ Type *VPTypeAnalysis::inferScalarTypeForRecipe(const VPInstruction *R) {
     CachedTypes[OtherV] = ResTy;
     return ResTy;
   }
+  case VPInstruction::Not: {
+    Type *ResTy = inferScalarType(R->getOperand(0));
+    assert(IntegerType::get(Ctx, 1) == ResTy &&
+           "unexpected scalar type inferred for operand");
+    return ResTy;
+  }
   case VPInstruction::PtrAdd:
     // Return the type based on the pointer argument (i.e. first operand).
     return inferScalarType(R->getOperand(0));
diff --git a/llvm/lib/Transforms/Vectorize/VPlanPatternMatch.h b/llvm/lib/Transforms/Vectorize/VPlanPatternMatch.h
index ae7717eb7cc9..50b08bbb7ebf 100644
--- a/llvm/lib/Transforms/Vectorize/VPlanPatternMatch.h
+++ b/llvm/lib/Transforms/Vectorize/VPlanPatternMatch.h
@@ -51,8 +51,9 @@ template <typename Class> struct bind_ty {
 };
 
 /// Match a specified integer value or vector of all elements of that
-/// value.
-struct specific_intval {
+/// value. \p BitWidth optionally specifies the bitwidth the matched constant
+/// must have. If it is 0, the matched constant can have any bitwidth.
+template <unsigned BitWidth = 0> struct specific_intval {
   APInt Val;
 
   specific_intval(APInt V) : Val(std::move(V)) {}
@@ -66,15 +67,21 @@ struct specific_intval {
       if (const auto *C = dyn_cast<Constant>(V))
         CI = dyn_cast_or_null<ConstantInt>(
             C->getSplatValue(/*AllowPoison=*/false));
+    if (!CI)
+      return false;
 
-    return CI && APInt::isSameValue(CI->getValue(), Val);
+    assert((BitWidth == 0 || CI->getBitWidth() == BitWidth) &&
+           "Trying the match constant with unexpected bitwidth.");
+    return APInt::isSameValue(CI->getValue(), Val);
   }
 };
 
-inline specific_intval m_SpecificInt(uint64_t V) {
-  return specific_intval(APInt(64, V));
+inline specific_intval<0> m_SpecificInt(uint64_t V) {
+  return specific_intval<0>(APInt(64, V));
 }
 
+inline specific_intval<1> m_False() { return specific_intval<1>(APInt(64, 0)); }
+
 /// Matching combinators
 template <typename LTy, typename RTy> struct match_combine_or {
   LTy L;
diff --git a/llvm/lib/Transforms/Vectorize/VPlanTransforms.cpp b/llvm/lib/Transforms/Vectorize/VPlanTransforms.cpp
index 007dc3f89b3f..a7337f7aa94d 100644
--- a/llvm/lib/Transforms/Vectorize/VPlanTransforms.cpp
+++ b/llvm/lib/Transforms/Vectorize/VPlanTransforms.cpp
@@ -506,13 +506,12 @@ static void removeDeadRecipes(VPlan &Plan) {
   }
 }
 
-static VPValue *createScalarIVSteps(VPlan &Plan,
-                                    InductionDescriptor::InductionKind Kind,
-                                    Instruction::BinaryOps InductionOpcode,
-                                    FPMathOperator *FPBinOp,
-                                    ScalarEvolution &SE, Instruction *TruncI,
-                                    VPValue *StartV, VPValue *Step,
-                                    VPBasicBlock::iterator IP) {
+static VPScalarIVStepsRecipe *
+createScalarIVSteps(VPlan &Plan, InductionDescriptor::InductionKind Kind,
+                    Instruction::BinaryOps InductionOpcode,
+                    FPMathOperator *FPBinOp, ScalarEvolution &SE,
+                    Instruction *TruncI, VPValue *StartV, VPValue *Step,
+                    VPBasicBlock::iterator IP) {
   VPBasicBlock *HeaderVPBB = Plan.getVectorLoopRegion()->getEntryBasicBlock();
   VPCanonicalIVPHIRecipe *CanonicalIV = Plan.getCanonicalIV();
   VPSingleDefRecipe *BaseIV = CanonicalIV;
@@ -579,16 +578,13 @@ static void legalizeAndOptimizeInductions(VPlan &Plan, ScalarEvolution &SE) {
       VPValue *StartV =
           Plan.getOrAddLiveIn(ConstantInt::get(ID.getStep()->getType(), 0));
       VPValue *StepV = PtrIV->getOperand(1);
-      VPRecipeBase *Steps =
-          createScalarIVSteps(Plan, InductionDescriptor::IK_IntInduction,
-                              Instruction::Add, nullptr, SE, nullptr, StartV,
-                              StepV, InsertPt)
-              ->getDefiningRecipe();
+      VPScalarIVStepsRecipe *Steps = createScalarIVSteps(
+          Plan, InductionDescriptor::IK_IntInduction, Instruction::Add, nullptr,
+          SE, nullptr, StartV, StepV, InsertPt);
 
-      auto *Recipe =
-          new VPInstruction(VPInstruction::PtrAdd,
-                            {PtrIV->getStartValue(), Steps->getVPSingleValue()},
-                            PtrIV->getDebugLoc(), "next.gep");
+      auto *Recipe = new VPInstruction(VPInstruction::PtrAdd,
+                                       {PtrIV->getStartValue(), Steps},
+                                       PtrIV->getDebugLoc(), "next.gep");
 
       Recipe->insertAfter(Steps);
       PtrIV->replaceAllUsesWith(Recipe);
@@ -606,7 +602,7 @@ static void legalizeAndOptimizeInductions(VPlan &Plan, ScalarEvolution &SE) {
       continue;
 
     const InductionDescriptor &ID = WideIV->getInductionDescriptor();
-    VPValue *Steps = createScalarIVSteps(
+    VPScalarIVStepsRecipe *Steps = createScalarIVSteps(
         Plan, ID.getKind(), ID.getInductionOpcode(),
         dyn_cast_or_null<FPMathOperator>(ID.getInductionBinOp()), SE,
         WideIV->getTruncInst(), WideIV->getStartValue(), WideIV->getStepValue(),
@@ -884,18 +880,19 @@ void VPlanTransforms::clearReductionWrapFlags(VPlan &Plan) {
 
 /// Try to simplify recipe \p R.
 static void simplifyRecipe(VPRecipeBase &R, VPTypeAnalysis &TypeInfo) {
+  using namespace llvm::VPlanPatternMatch;
   // Try to remove redundant blend recipes.
   if (auto *Blend = dyn_cast<VPBlendRecipe>(&R)) {
     VPValue *Inc0 = Blend->getIncomingValue(0);
     for (unsigned I = 1; I != Blend->getNumIncomingValues(); ++I)
-      if (Inc0 != Blend->getIncomingValue(I))
+      if (Inc0 != Blend->getIncomingValue(I) &&
+          !match(Blend->getMask(I), m_False()))
         return;
     Blend->replaceAllUsesWith(Inc0);
     Blend->eraseFromParent();
     return;
   }
 
-  using namespace llvm::VPlanPatternMatch;
   VPValue *A;
   if (match(&R, m_Trunc(m_ZExtOrSExt(m_VPValue(A))))) {
     VPValue *Trunc = R.getVPSingleValue();
@@ -1031,7 +1028,9 @@ void VPlanTransforms::truncateToMinimalBitwidths(
       if (auto *VPW = dyn_cast<VPRecipeWithIRFlags>(&R))
         VPW->dropPoisonGeneratingFlags();
 
-      if (OldResSizeInBits != NewResSizeInBits) {
+      using namespace llvm::VPlanPatternMatch;
+      if (OldResSizeInBits != NewResSizeInBits &&
+          !match(&R, m_Binary<Instruction::ICmp>(m_VPValue(), m_VPValue()))) {
         // Extend result to original width.
         auto *Ext =
             new VPWidenCastRecipe(Instruction::ZExt, ResultVPV, OldResTy);
@@ -1040,8 +1039,9 @@ void VPlanTransforms::truncateToMinimalBitwidths(
         Ext->setOperand(0, ResultVPV);
         assert(OldResSizeInBits > NewResSizeInBits && "Nothing to shrink?");
       } else
-        assert(cast<VPWidenRecipe>(&R)->getOpcode() == Instruction::ICmp &&
-               "Only ICmps should not need extending the result.");
+        assert(
+            match(&R, m_Binary<Instruction::ICmp>(m_VPValue(), m_VPValue())) &&
+            "Only ICmps should not need extending the result.");
 
       assert(!isa<VPWidenStoreRecipe>(&R) && "stores cannot be narrowed");
       if (isa<VPWidenLoadRecipe>(&R))
diff --git a/llvm/lib/Transforms/Vectorize/VectorCombine.cpp b/llvm/lib/Transforms/Vectorize/VectorCombine.cpp
index df761f9b711a..da3c780550a0 100644
--- a/llvm/lib/Transforms/Vectorize/VectorCombine.cpp
+++ b/llvm/lib/Transforms/Vectorize/VectorCombine.cpp
@@ -1395,60 +1395,91 @@ bool VectorCombine::scalarizeLoadExtract(Instruction &I) {
   return true;
 }
 
-/// Try to convert "shuffle (binop), (binop)" with a shared binop operand into
-/// "binop (shuffle), (shuffle)".
+/// Try to convert "shuffle (binop), (binop)" into "binop (shuffle), (shuffle)".
 bool VectorCombine::foldShuffleOfBinops(Instruction &I) {
-  auto *VecTy = cast<FixedVectorType>(I.getType());
   BinaryOperator *B0, *B1;
-  ArrayRef<int> Mask;
+  ArrayRef<int> OldMask;
   if (!match(&I, m_Shuffle(m_OneUse(m_BinOp(B0)), m_OneUse(m_BinOp(B1)),
-                           m_Mask(Mask))) ||
-      B0->getOpcode() != B1->getOpcode() || B0->getType() != VecTy)
+                           m_Mask(OldMask))))
     return false;
 
   // Don't introduce poison into div/rem.
-  if (any_of(Mask, [](int M) { return M == PoisonMaskElem; }) &&
+  if (any_of(OldMask, [](int M) { return M == PoisonMaskElem; }) &&
       B0->isIntDivRem())
     return false;
 
-  // Try to replace a binop with a shuffle if the shuffle is not costly.
-  // The new shuffle will choose from a single, common operand, so it may be
-  // cheaper than the existing two-operand shuffle.
-  SmallVector<int> UnaryMask = createUnaryMask(Mask, Mask.size());
+  // TODO: Add support for addlike etc.
   Instruction::BinaryOps Opcode = B0->getOpcode();
-  InstructionCost BinopCost = TTI.getArithmeticInstrCost(Opcode, VecTy);
-  InstructionCost ShufCost = TTI.getShuffleCost(
-      TargetTransformInfo::SK_PermuteSingleSrc, VecTy, UnaryMask);
-  if (ShufCost > BinopCost)
+  if (Opcode != B1->getOpcode())
+    return false;
+
+  auto *ShuffleDstTy = dyn_cast<FixedVectorType>(I.getType());
+  auto *BinOpTy = dyn_cast<FixedVectorType>(B0->getType());
+  if (!ShuffleDstTy || !BinOpTy)
     return false;
 
+  unsigned NumSrcElts = BinOpTy->getNumElements();
+
   // If we have something like "add X, Y" and "add Z, X", swap ops to match.
   Value *X = B0->getOperand(0), *Y = B0->getOperand(1);
   Value *Z = B1->getOperand(0), *W = B1->getOperand(1);
-  if (BinaryOperator::isCommutative(Opcode) && X != Z && Y != W)
+  if (BinaryOperator::isCommutative(Opcode) && X != Z && Y != W &&
+      (X == W || Y == Z))
     std::swap(X, Y);
 
-  Value *Shuf0, *Shuf1;
+  auto ConvertToUnary = [NumSrcElts](int &M) {
+    if (M >= (int)NumSrcElts)
+      M -= NumSrcElts;
+  };
+
+  SmallVector<int> NewMask0(OldMask.begin(), OldMask.end());
+  TargetTransformInfo::ShuffleKind SK0 = TargetTransformInfo::SK_PermuteTwoSrc;
   if (X == Z) {
-    // shuf (bo X, Y), (bo X, W) --> bo (shuf X), (shuf Y, W)
-    Shuf0 = Builder.CreateShuffleVector(X, UnaryMask);
-    Shuf1 = Builder.CreateShuffleVector(Y, W, Mask);
-  } else if (Y == W) {
-    // shuf (bo X, Y), (bo Z, Y) --> bo (shuf X, Z), (shuf Y)
-    Shuf0 = Builder.CreateShuffleVector(X, Z, Mask);
-    Shuf1 = Builder.CreateShuffleVector(Y, UnaryMask);
-  } else {
-    return false;
+    llvm::for_each(NewMask0, ConvertToUnary);
+    SK0 = TargetTransformInfo::SK_PermuteSingleSrc;
+    Z = PoisonValue::get(BinOpTy);
+  }
+
+  SmallVector<int> NewMask1(OldMask.begin(), OldMask.end());
+  TargetTransformInfo::ShuffleKind SK1 = TargetTransformInfo::SK_PermuteTwoSrc;
+  if (Y == W) {
+    llvm::for_each(NewMask1, ConvertToUnary);
+    SK1 = TargetTransformInfo::SK_PermuteSingleSrc;
+    W = PoisonValue::get(BinOpTy);
   }
 
+  // Try to replace a binop with a shuffle if the shuffle is not costly.
+  TTI::TargetCostKind CostKind = TTI::TCK_RecipThroughput;
+
+  InstructionCost OldCost =
+      TTI.getArithmeticInstrCost(B0->getOpcode(), BinOpTy, CostKind) +
+      TTI.getArithmeticInstrCost(B1->getOpcode(), BinOpTy, CostKind) +
+      TTI.getShuffleCost(TargetTransformInfo::SK_PermuteTwoSrc, BinOpTy,
+                         OldMask, CostKind, 0, nullptr, {B0, B1}, &I);
+
+  InstructionCost NewCost =
+      TTI.getShuffleCost(SK0, BinOpTy, NewMask0, CostKind, 0, nullptr, {X, Z}) +
+      TTI.getShuffleCost(SK1, BinOpTy, NewMask1, CostKind, 0, nullptr, {Y, W}) +
+      TTI.getArithmeticInstrCost(Opcode, ShuffleDstTy, CostKind);
+
+  LLVM_DEBUG(dbgs() << "Found a shuffle feeding two binops: " << I
+                    << "\n  OldCost: " << OldCost << " vs NewCost: " << NewCost
+                    << "\n");
+  if (NewCost >= OldCost)
+    return false;
+
+  Value *Shuf0 = Builder.CreateShuffleVector(X, Z, NewMask0);
+  Value *Shuf1 = Builder.CreateShuffleVector(Y, W, NewMask1);
   Value *NewBO = Builder.CreateBinOp(Opcode, Shuf0, Shuf1);
+
   // Intersect flags from the old binops.
   if (auto *NewInst = dyn_cast<Instruction>(NewBO)) {
     NewInst->copyIRFlags(B0);
     NewInst->andIRFlags(B1);
   }
 
-  // TODO: Add Shuf0/Shuf1 to WorkList?
+  Worklist.pushValue(Shuf0);
+  Worklist.pushValue(Shuf1);
   replaceValue(I, *NewBO);
   return true;
 }
@@ -1566,6 +1597,8 @@ bool VectorCombine::foldShuffleOfShuffles(Instruction &I) {
                            m_Mask(OuterMask))))
     return false;
 
+  auto *ShufI0 = dyn_cast<Instruction>(I.getOperand(0));
+  auto *ShufI1 = dyn_cast<Instruction>(I.getOperand(1));
   auto *ShuffleDstTy = dyn_cast<FixedVectorType>(I.getType());
   auto *ShuffleSrcTy = dyn_cast<FixedVectorType>(V0->getType());
   auto *ShuffleImmTy = dyn_cast<FixedVectorType>(I.getOperand(0)->getType());
@@ -1607,14 +1640,15 @@ bool VectorCombine::foldShuffleOfShuffles(Instruction &I) {
 
   InstructionCost OldCost =
       TTI.getShuffleCost(TargetTransformInfo::SK_PermuteSingleSrc, ShuffleSrcTy,
-                         InnerMask0, CostKind) +
+                         InnerMask0, CostKind, 0, nullptr, {V0, U0}, ShufI0) +
       TTI.getShuffleCost(TargetTransformInfo::SK_PermuteSingleSrc, ShuffleSrcTy,
-                         InnerMask1, CostKind) +
+                         InnerMask1, CostKind, 0, nullptr, {V1, U1}, ShufI1) +
       TTI.getShuffleCost(TargetTransformInfo::SK_PermuteTwoSrc, ShuffleImmTy,
-                         OuterMask, CostKind, 0, nullptr, std::nullopt, &I);
+                         OuterMask, CostKind, 0, nullptr, {ShufI0, ShufI1}, &I);
 
-  InstructionCost NewCost = TTI.getShuffleCost(
-      TargetTransformInfo::SK_PermuteTwoSrc, ShuffleSrcTy, NewMask, CostKind);
+  InstructionCost NewCost =
+      TTI.getShuffleCost(TargetTransformInfo::SK_PermuteTwoSrc, ShuffleSrcTy,
+                         NewMask, CostKind, 0, nullptr, {V0, V1});
 
   LLVM_DEBUG(dbgs() << "Found a shuffle feeding two shuffles: " << I
                     << "\n  OldCost: " << OldCost << " vs NewCost: " << NewCost
diff --git a/llvm/test/Analysis/CostModel/AArch64/getIntrinsicInstrCost-vector-reverse.ll b/llvm/test/Analysis/CostModel/AArch64/getIntrinsicInstrCost-vector-reverse.ll
index 836a028ad6aa..f491b086107a 100644
--- a/llvm/test/Analysis/CostModel/AArch64/getIntrinsicInstrCost-vector-reverse.ll
+++ b/llvm/test/Analysis/CostModel/AArch64/getIntrinsicInstrCost-vector-reverse.ll
@@ -7,58 +7,58 @@ target datalayout = "e-m:e-i8:8:32-i16:16:32-i64:64-i128:128-n32:64-S128"
 
 define void @vector_reverse() #0{
 ; CHECK-LABEL: 'vector_reverse'
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %1 = call <16 x i8> @llvm.experimental.vector.reverse.v16i8(<16 x i8> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %2 = call <32 x i8> @llvm.experimental.vector.reverse.v32i8(<32 x i8> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %3 = call <8 x i16> @llvm.experimental.vector.reverse.v8i16(<8 x i16> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %4 = call <16 x i16> @llvm.experimental.vector.reverse.v16i16(<16 x i16> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %5 = call <4 x i32> @llvm.experimental.vector.reverse.v4i32(<4 x i32> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %6 = call <8 x i32> @llvm.experimental.vector.reverse.v8i32(<8 x i32> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %7 = call <2 x i64> @llvm.experimental.vector.reverse.v2i64(<2 x i64> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %8 = call <4 x i64> @llvm.experimental.vector.reverse.v4i64(<4 x i64> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %9 = call <8 x half> @llvm.experimental.vector.reverse.v8f16(<8 x half> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %10 = call <16 x half> @llvm.experimental.vector.reverse.v16f16(<16 x half> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %11 = call <4 x float> @llvm.experimental.vector.reverse.v4f32(<4 x float> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %12 = call <8 x float> @llvm.experimental.vector.reverse.v8f32(<8 x float> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %13 = call <2 x double> @llvm.experimental.vector.reverse.v2f64(<2 x double> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %14 = call <4 x double> @llvm.experimental.vector.reverse.v4f64(<4 x double> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 28 for instruction: %15 = call <8 x bfloat> @llvm.experimental.vector.reverse.v8bf16(<8 x bfloat> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 56 for instruction: %16 = call <16 x bfloat> @llvm.experimental.vector.reverse.v16bf16(<16 x bfloat> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %1 = call <16 x i8> @llvm.vector.reverse.v16i8(<16 x i8> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %2 = call <32 x i8> @llvm.vector.reverse.v32i8(<32 x i8> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %3 = call <8 x i16> @llvm.vector.reverse.v8i16(<8 x i16> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %4 = call <16 x i16> @llvm.vector.reverse.v16i16(<16 x i16> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %5 = call <4 x i32> @llvm.vector.reverse.v4i32(<4 x i32> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %6 = call <8 x i32> @llvm.vector.reverse.v8i32(<8 x i32> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %7 = call <2 x i64> @llvm.vector.reverse.v2i64(<2 x i64> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %8 = call <4 x i64> @llvm.vector.reverse.v4i64(<4 x i64> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %9 = call <8 x half> @llvm.vector.reverse.v8f16(<8 x half> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %10 = call <16 x half> @llvm.vector.reverse.v16f16(<16 x half> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %11 = call <4 x float> @llvm.vector.reverse.v4f32(<4 x float> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %12 = call <8 x float> @llvm.vector.reverse.v8f32(<8 x float> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %13 = call <2 x double> @llvm.vector.reverse.v2f64(<2 x double> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %14 = call <4 x double> @llvm.vector.reverse.v4f64(<4 x double> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 28 for instruction: %15 = call <8 x bfloat> @llvm.vector.reverse.v8bf16(<8 x bfloat> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 56 for instruction: %16 = call <16 x bfloat> @llvm.vector.reverse.v16bf16(<16 x bfloat> undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
 ;
 
-  call <16 x i8> @llvm.experimental.vector.reverse.v16i8(<16 x i8> undef)
-  call <32 x i8> @llvm.experimental.vector.reverse.v32i8(<32 x i8> undef)
-  call <8 x i16> @llvm.experimental.vector.reverse.v8i16(<8 x i16> undef)
-  call <16 x i16> @llvm.experimental.vector.reverse.v16i16(<16 x i16> undef)
-  call <4 x i32> @llvm.experimental.vector.reverse.v4i32(<4 x i32> undef)
-  call <8 x i32> @llvm.experimental.vector.reverse.v8i32(<8 x i32> undef)
-  call <2 x i64> @llvm.experimental.vector.reverse.v2i64(<2 x i64> undef)
-  call <4 x i64> @llvm.experimental.vector.reverse.v4i64(<4 x i64> undef)
-  call <8 x half> @llvm.experimental.vector.reverse.v8f16(<8 x half> undef)
-  call <16 x half> @llvm.experimental.vector.reverse.v16f16(<16 x half> undef)
-  call <4 x float> @llvm.experimental.vector.reverse.v4f32(<4 x float> undef)
-  call <8 x float> @llvm.experimental.vector.reverse.v8f32(<8 x float> undef)
-  call <2 x double> @llvm.experimental.vector.reverse.v2f64(<2 x double> undef)
-  call <4 x double> @llvm.experimental.vector.reverse.v4f64(<4 x double> undef)
-  call <8 x bfloat> @llvm.experimental.vector.reverse.v8bf16(<8 x bfloat> undef)
-  call <16 x bfloat> @llvm.experimental.vector.reverse.v16bf16(<16 x bfloat> undef)
+  call <16 x i8> @llvm.vector.reverse.v16i8(<16 x i8> undef)
+  call <32 x i8> @llvm.vector.reverse.v32i8(<32 x i8> undef)
+  call <8 x i16> @llvm.vector.reverse.v8i16(<8 x i16> undef)
+  call <16 x i16> @llvm.vector.reverse.v16i16(<16 x i16> undef)
+  call <4 x i32> @llvm.vector.reverse.v4i32(<4 x i32> undef)
+  call <8 x i32> @llvm.vector.reverse.v8i32(<8 x i32> undef)
+  call <2 x i64> @llvm.vector.reverse.v2i64(<2 x i64> undef)
+  call <4 x i64> @llvm.vector.reverse.v4i64(<4 x i64> undef)
+  call <8 x half> @llvm.vector.reverse.v8f16(<8 x half> undef)
+  call <16 x half> @llvm.vector.reverse.v16f16(<16 x half> undef)
+  call <4 x float> @llvm.vector.reverse.v4f32(<4 x float> undef)
+  call <8 x float> @llvm.vector.reverse.v8f32(<8 x float> undef)
+  call <2 x double> @llvm.vector.reverse.v2f64(<2 x double> undef)
+  call <4 x double> @llvm.vector.reverse.v4f64(<4 x double> undef)
+  call <8 x bfloat> @llvm.vector.reverse.v8bf16(<8 x bfloat> undef)
+  call <16 x bfloat> @llvm.vector.reverse.v16bf16(<16 x bfloat> undef)
   ret void
 }
 
 attributes #0 = { "target-features"="+sve,+bf16" }
-declare <16 x i8> @llvm.experimental.vector.reverse.v16i8(<16 x i8>)
-declare <32 x i8> @llvm.experimental.vector.reverse.v32i8(<32 x i8>)
-declare <8 x i16> @llvm.experimental.vector.reverse.v8i16(<8 x i16>)
-declare <16 x i16> @llvm.experimental.vector.reverse.v16i16(<16 x i16>)
-declare <4 x i32> @llvm.experimental.vector.reverse.v4i32(<4 x i32>)
-declare <8 x i32> @llvm.experimental.vector.reverse.v8i32(<8 x i32>)
-declare <2 x i64> @llvm.experimental.vector.reverse.v2i64(<2 x i64>)
-declare <4 x i64> @llvm.experimental.vector.reverse.v4i64(<4 x i64>)
-declare <8 x half> @llvm.experimental.vector.reverse.v8f16(<8 x half>)
-declare <16 x half> @llvm.experimental.vector.reverse.v16f16(<16 x half>)
-declare <4 x float> @llvm.experimental.vector.reverse.v4f32(<4 x float>)
-declare <8 x float> @llvm.experimental.vector.reverse.v8f32(<8 x float>)
-declare <2 x double> @llvm.experimental.vector.reverse.v2f64(<2 x double>)
-declare <4 x double> @llvm.experimental.vector.reverse.v4f64(<4 x double>)
-declare <8 x bfloat> @llvm.experimental.vector.reverse.v8bf16(<8 x bfloat>)
-declare <16 x bfloat> @llvm.experimental.vector.reverse.v16bf16(<16 x bfloat>)
+declare <16 x i8> @llvm.vector.reverse.v16i8(<16 x i8>)
+declare <32 x i8> @llvm.vector.reverse.v32i8(<32 x i8>)
+declare <8 x i16> @llvm.vector.reverse.v8i16(<8 x i16>)
+declare <16 x i16> @llvm.vector.reverse.v16i16(<16 x i16>)
+declare <4 x i32> @llvm.vector.reverse.v4i32(<4 x i32>)
+declare <8 x i32> @llvm.vector.reverse.v8i32(<8 x i32>)
+declare <2 x i64> @llvm.vector.reverse.v2i64(<2 x i64>)
+declare <4 x i64> @llvm.vector.reverse.v4i64(<4 x i64>)
+declare <8 x half> @llvm.vector.reverse.v8f16(<8 x half>)
+declare <16 x half> @llvm.vector.reverse.v16f16(<16 x half>)
+declare <4 x float> @llvm.vector.reverse.v4f32(<4 x float>)
+declare <8 x float> @llvm.vector.reverse.v8f32(<8 x float>)
+declare <2 x double> @llvm.vector.reverse.v2f64(<2 x double>)
+declare <4 x double> @llvm.vector.reverse.v4f64(<4 x double>)
+declare <8 x bfloat> @llvm.vector.reverse.v8bf16(<8 x bfloat>)
+declare <16 x bfloat> @llvm.vector.reverse.v16bf16(<16 x bfloat>)
diff --git a/llvm/test/Analysis/CostModel/AArch64/splice.ll b/llvm/test/Analysis/CostModel/AArch64/splice.ll
index f5afdff41b1d..1d76a4838cee 100644
--- a/llvm/test/Analysis/CostModel/AArch64/splice.ll
+++ b/llvm/test/Analysis/CostModel/AArch64/splice.ll
@@ -5,96 +5,96 @@ target datalayout = "e-m:e-i8:8:32-i16:16:32-i64:64-i128:128-n32:64-S128"
 
 define void @vector_splice() #0 {
 ; CHECK-LABEL: 'vector_splice'
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %splice.v16i8 = call <16 x i8> @llvm.experimental.vector.splice.v16i8(<16 x i8> zeroinitializer, <16 x i8> zeroinitializer, i32 1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.v32i8 = call <32 x i8> @llvm.experimental.vector.splice.v32i8(<32 x i8> zeroinitializer, <32 x i8> zeroinitializer, i32 1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %splice.v2i16 = call <2 x i16> @llvm.experimental.vector.splice.v2i16(<2 x i16> zeroinitializer, <2 x i16> zeroinitializer, i32 1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %splice.v4i16 = call <4 x i16> @llvm.experimental.vector.splice.v4i16(<4 x i16> zeroinitializer, <4 x i16> zeroinitializer, i32 1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %splice.v8i16 = call <8 x i16> @llvm.experimental.vector.splice.v8i16(<8 x i16> zeroinitializer, <8 x i16> zeroinitializer, i32 1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.v16i16 = call <16 x i16> @llvm.experimental.vector.splice.v16i16(<16 x i16> zeroinitializer, <16 x i16> zeroinitializer, i32 1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %splice.v4i32 = call <4 x i32> @llvm.experimental.vector.splice.v4i32(<4 x i32> zeroinitializer, <4 x i32> zeroinitializer, i32 1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.v8i32 = call <8 x i32> @llvm.experimental.vector.splice.v8i32(<8 x i32> zeroinitializer, <8 x i32> zeroinitializer, i32 1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %splice.v2i64 = call <2 x i64> @llvm.experimental.vector.splice.v2i64(<2 x i64> zeroinitializer, <2 x i64> zeroinitializer, i32 1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.v4i64 = call <4 x i64> @llvm.experimental.vector.splice.v4i64(<4 x i64> zeroinitializer, <4 x i64> zeroinitializer, i32 1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %splice.v2f16 = call <2 x half> @llvm.experimental.vector.splice.v2f16(<2 x half> zeroinitializer, <2 x half> zeroinitializer, i32 1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %splice.v4f16 = call <4 x half> @llvm.experimental.vector.splice.v4f16(<4 x half> zeroinitializer, <4 x half> zeroinitializer, i32 1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %splice.v8f16 = call <8 x half> @llvm.experimental.vector.splice.v8f16(<8 x half> zeroinitializer, <8 x half> zeroinitializer, i32 1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.v16f16 = call <16 x half> @llvm.experimental.vector.splice.v16f16(<16 x half> zeroinitializer, <16 x half> zeroinitializer, i32 1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %splice.v2f32 = call <2 x float> @llvm.experimental.vector.splice.v2f32(<2 x float> zeroinitializer, <2 x float> zeroinitializer, i32 1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %splice.v4f32 = call <4 x float> @llvm.experimental.vector.splice.v4f32(<4 x float> zeroinitializer, <4 x float> zeroinitializer, i32 1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.v8f32 = call <8 x float> @llvm.experimental.vector.splice.v8f32(<8 x float> zeroinitializer, <8 x float> zeroinitializer, i32 1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %splice.v2f64 = call <2 x double> @llvm.experimental.vector.splice.v2f64(<2 x double> zeroinitializer, <2 x double> zeroinitializer, i32 1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.v4f64 = call <4 x double> @llvm.experimental.vector.splice.v4f64(<4 x double> zeroinitializer, <4 x double> zeroinitializer, i32 1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %splice.v2bf16 = call <2 x bfloat> @llvm.experimental.vector.splice.v2bf16(<2 x bfloat> zeroinitializer, <2 x bfloat> zeroinitializer, i32 1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %splice.v4bf16 = call <4 x bfloat> @llvm.experimental.vector.splice.v4bf16(<4 x bfloat> zeroinitializer, <4 x bfloat> zeroinitializer, i32 1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %splice.v8bf16 = call <8 x bfloat> @llvm.experimental.vector.splice.v8bf16(<8 x bfloat> zeroinitializer, <8 x bfloat> zeroinitializer, i32 1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.v16bf16 = call <16 x bfloat> @llvm.experimental.vector.splice.v16bf16(<16 x bfloat> zeroinitializer, <16 x bfloat> zeroinitializer, i32 1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %splice.v16i1 = call <16 x i1> @llvm.experimental.vector.splice.v16i1(<16 x i1> zeroinitializer, <16 x i1> zeroinitializer, i32 1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %splice.v8i1 = call <8 x i1> @llvm.experimental.vector.splice.v8i1(<8 x i1> zeroinitializer, <8 x i1> zeroinitializer, i32 1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %splice.v4i1 = call <4 x i1> @llvm.experimental.vector.splice.v4i1(<4 x i1> zeroinitializer, <4 x i1> zeroinitializer, i32 1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %splice.v2i1 = call <2 x i1> @llvm.experimental.vector.splice.v2i1(<2 x i1> zeroinitializer, <2 x i1> zeroinitializer, i32 1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %splice.v2i128 = call <2 x i128> @llvm.experimental.vector.splice.v2i128(<2 x i128> zeroinitializer, <2 x i128> zeroinitializer, i32 1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %splice.v16i8 = call <16 x i8> @llvm.vector.splice.v16i8(<16 x i8> zeroinitializer, <16 x i8> zeroinitializer, i32 1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.v32i8 = call <32 x i8> @llvm.vector.splice.v32i8(<32 x i8> zeroinitializer, <32 x i8> zeroinitializer, i32 1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %splice.v2i16 = call <2 x i16> @llvm.vector.splice.v2i16(<2 x i16> zeroinitializer, <2 x i16> zeroinitializer, i32 1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %splice.v4i16 = call <4 x i16> @llvm.vector.splice.v4i16(<4 x i16> zeroinitializer, <4 x i16> zeroinitializer, i32 1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %splice.v8i16 = call <8 x i16> @llvm.vector.splice.v8i16(<8 x i16> zeroinitializer, <8 x i16> zeroinitializer, i32 1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.v16i16 = call <16 x i16> @llvm.vector.splice.v16i16(<16 x i16> zeroinitializer, <16 x i16> zeroinitializer, i32 1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %splice.v4i32 = call <4 x i32> @llvm.vector.splice.v4i32(<4 x i32> zeroinitializer, <4 x i32> zeroinitializer, i32 1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.v8i32 = call <8 x i32> @llvm.vector.splice.v8i32(<8 x i32> zeroinitializer, <8 x i32> zeroinitializer, i32 1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %splice.v2i64 = call <2 x i64> @llvm.vector.splice.v2i64(<2 x i64> zeroinitializer, <2 x i64> zeroinitializer, i32 1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.v4i64 = call <4 x i64> @llvm.vector.splice.v4i64(<4 x i64> zeroinitializer, <4 x i64> zeroinitializer, i32 1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %splice.v2f16 = call <2 x half> @llvm.vector.splice.v2f16(<2 x half> zeroinitializer, <2 x half> zeroinitializer, i32 1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %splice.v4f16 = call <4 x half> @llvm.vector.splice.v4f16(<4 x half> zeroinitializer, <4 x half> zeroinitializer, i32 1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %splice.v8f16 = call <8 x half> @llvm.vector.splice.v8f16(<8 x half> zeroinitializer, <8 x half> zeroinitializer, i32 1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.v16f16 = call <16 x half> @llvm.vector.splice.v16f16(<16 x half> zeroinitializer, <16 x half> zeroinitializer, i32 1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %splice.v2f32 = call <2 x float> @llvm.vector.splice.v2f32(<2 x float> zeroinitializer, <2 x float> zeroinitializer, i32 1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %splice.v4f32 = call <4 x float> @llvm.vector.splice.v4f32(<4 x float> zeroinitializer, <4 x float> zeroinitializer, i32 1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.v8f32 = call <8 x float> @llvm.vector.splice.v8f32(<8 x float> zeroinitializer, <8 x float> zeroinitializer, i32 1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %splice.v2f64 = call <2 x double> @llvm.vector.splice.v2f64(<2 x double> zeroinitializer, <2 x double> zeroinitializer, i32 1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.v4f64 = call <4 x double> @llvm.vector.splice.v4f64(<4 x double> zeroinitializer, <4 x double> zeroinitializer, i32 1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %splice.v2bf16 = call <2 x bfloat> @llvm.vector.splice.v2bf16(<2 x bfloat> zeroinitializer, <2 x bfloat> zeroinitializer, i32 1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %splice.v4bf16 = call <4 x bfloat> @llvm.vector.splice.v4bf16(<4 x bfloat> zeroinitializer, <4 x bfloat> zeroinitializer, i32 1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %splice.v8bf16 = call <8 x bfloat> @llvm.vector.splice.v8bf16(<8 x bfloat> zeroinitializer, <8 x bfloat> zeroinitializer, i32 1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.v16bf16 = call <16 x bfloat> @llvm.vector.splice.v16bf16(<16 x bfloat> zeroinitializer, <16 x bfloat> zeroinitializer, i32 1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %splice.v16i1 = call <16 x i1> @llvm.vector.splice.v16i1(<16 x i1> zeroinitializer, <16 x i1> zeroinitializer, i32 1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %splice.v8i1 = call <8 x i1> @llvm.vector.splice.v8i1(<8 x i1> zeroinitializer, <8 x i1> zeroinitializer, i32 1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %splice.v4i1 = call <4 x i1> @llvm.vector.splice.v4i1(<4 x i1> zeroinitializer, <4 x i1> zeroinitializer, i32 1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %splice.v2i1 = call <2 x i1> @llvm.vector.splice.v2i1(<2 x i1> zeroinitializer, <2 x i1> zeroinitializer, i32 1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %splice.v2i128 = call <2 x i128> @llvm.vector.splice.v2i128(<2 x i128> zeroinitializer, <2 x i128> zeroinitializer, i32 1)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
 ;
-  %splice.v16i8 = call <16 x i8> @llvm.experimental.vector.splice.v16i8(<16 x i8> zeroinitializer, <16 x i8> zeroinitializer, i32 1)
-  %splice.v32i8 = call <32 x i8> @llvm.experimental.vector.splice.v32i8(<32 x i8> zeroinitializer, <32 x i8> zeroinitializer, i32 1)
-  %splice.v2i16 = call <2 x i16> @llvm.experimental.vector.splice.v2i16(<2 x i16> zeroinitializer, <2 x i16> zeroinitializer, i32 1)
-  %splice.v4i16 = call <4 x i16> @llvm.experimental.vector.splice.v4i16(<4 x i16> zeroinitializer, <4 x i16> zeroinitializer, i32 1)
-  %splice.v8i16 = call <8 x i16> @llvm.experimental.vector.splice.v8i16(<8 x i16> zeroinitializer, <8 x i16> zeroinitializer, i32 1)
-  %splice.v16i16 = call <16 x i16> @llvm.experimental.vector.splice.v16i16(<16 x i16> zeroinitializer, <16 x i16> zeroinitializer, i32 1)
-  %splice.v4i32 = call <4 x i32> @llvm.experimental.vector.splice.v4i32(<4 x i32> zeroinitializer, <4 x i32> zeroinitializer, i32 1)
-  %splice.v8i32 = call <8 x i32> @llvm.experimental.vector.splice.v8i32(<8 x i32> zeroinitializer, <8 x i32> zeroinitializer, i32 1)
-  %splice.v2i64 = call <2 x i64> @llvm.experimental.vector.splice.v2i64(<2 x i64> zeroinitializer, <2 x i64> zeroinitializer, i32 1)
-  %splice.v4i64 = call <4 x i64> @llvm.experimental.vector.splice.v4i64(<4 x i64> zeroinitializer, <4 x i64> zeroinitializer, i32 1)
-  %splice.v2f16 = call <2 x half> @llvm.experimental.vector.splice.v2f16(<2 x half> zeroinitializer, <2 x half> zeroinitializer, i32 1)
-  %splice.v4f16 = call <4 x half> @llvm.experimental.vector.splice.v4f16(<4 x half> zeroinitializer, <4 x half> zeroinitializer, i32 1)
-  %splice.v8f16 = call <8 x half> @llvm.experimental.vector.splice.v8f16(<8 x half> zeroinitializer, <8 x half> zeroinitializer, i32 1)
-  %splice.v16f16 = call <16 x half> @llvm.experimental.vector.splice.v16f16(<16 x half> zeroinitializer, <16 x half> zeroinitializer, i32 1)
-  %splice.v2f32 = call <2 x float> @llvm.experimental.vector.splice.v2f32(<2 x float> zeroinitializer, <2 x float> zeroinitializer, i32 1)
-  %splice.v4f32 = call <4 x float> @llvm.experimental.vector.splice.v4f32(<4 x float> zeroinitializer, <4 x float> zeroinitializer, i32 1)
-  %splice.v8f32 = call <8 x float> @llvm.experimental.vector.splice.v8f32(<8 x float> zeroinitializer, <8 x float> zeroinitializer, i32 1)
-  %splice.v2f64 = call <2 x double> @llvm.experimental.vector.splice.v2f64(<2 x double> zeroinitializer, <2 x double> zeroinitializer, i32 1)
-  %splice.v4f64 = call <4 x double> @llvm.experimental.vector.splice.v4f64(<4 x double> zeroinitializer, <4 x double> zeroinitializer, i32 1)
-  %splice.v2bf16 = call <2 x bfloat> @llvm.experimental.vector.splice.v2bf16(<2 x bfloat> zeroinitializer, <2 x bfloat> zeroinitializer, i32 1)
-  %splice.v4bf16 = call <4 x bfloat> @llvm.experimental.vector.splice.v4bf16(<4 x bfloat> zeroinitializer, <4 x bfloat> zeroinitializer, i32 1)
-  %splice.v8bf16 = call <8 x bfloat> @llvm.experimental.vector.splice.v8bf16(<8 x bfloat> zeroinitializer, <8 x bfloat> zeroinitializer, i32 1)
-  %splice.v16bf16 = call <16 x bfloat> @llvm.experimental.vector.splice.v16bf16(<16 x bfloat> zeroinitializer, <16 x bfloat> zeroinitializer, i32 1)
-  %splice.v16i1 = call <16 x i1> @llvm.experimental.vector.splice.v16i1(<16 x i1> zeroinitializer, <16 x i1> zeroinitializer, i32 1)
-  %splice.v8i1 =  call <8 x i1> @llvm.experimental.vector.splice.v8i1(<8 x i1> zeroinitializer, <8 x i1> zeroinitializer, i32 1)
-  %splice.v4i1 = call <4 x i1> @llvm.experimental.vector.splice.v4i1(<4 x i1> zeroinitializer, <4 x i1> zeroinitializer, i32 1)
-  %splice.v2i1 = call <2 x i1> @llvm.experimental.vector.splice.v2i1(<2 x i1> zeroinitializer, <2 x i1> zeroinitializer, i32 1)
-  %splice.v2i128 = call <2 x i128> @llvm.experimental.vector.splice.v2i128(<2 x i128> zeroinitializer, <2 x i128> zeroinitializer, i32 1)
+  %splice.v16i8 = call <16 x i8> @llvm.vector.splice.v16i8(<16 x i8> zeroinitializer, <16 x i8> zeroinitializer, i32 1)
+  %splice.v32i8 = call <32 x i8> @llvm.vector.splice.v32i8(<32 x i8> zeroinitializer, <32 x i8> zeroinitializer, i32 1)
+  %splice.v2i16 = call <2 x i16> @llvm.vector.splice.v2i16(<2 x i16> zeroinitializer, <2 x i16> zeroinitializer, i32 1)
+  %splice.v4i16 = call <4 x i16> @llvm.vector.splice.v4i16(<4 x i16> zeroinitializer, <4 x i16> zeroinitializer, i32 1)
+  %splice.v8i16 = call <8 x i16> @llvm.vector.splice.v8i16(<8 x i16> zeroinitializer, <8 x i16> zeroinitializer, i32 1)
+  %splice.v16i16 = call <16 x i16> @llvm.vector.splice.v16i16(<16 x i16> zeroinitializer, <16 x i16> zeroinitializer, i32 1)
+  %splice.v4i32 = call <4 x i32> @llvm.vector.splice.v4i32(<4 x i32> zeroinitializer, <4 x i32> zeroinitializer, i32 1)
+  %splice.v8i32 = call <8 x i32> @llvm.vector.splice.v8i32(<8 x i32> zeroinitializer, <8 x i32> zeroinitializer, i32 1)
+  %splice.v2i64 = call <2 x i64> @llvm.vector.splice.v2i64(<2 x i64> zeroinitializer, <2 x i64> zeroinitializer, i32 1)
+  %splice.v4i64 = call <4 x i64> @llvm.vector.splice.v4i64(<4 x i64> zeroinitializer, <4 x i64> zeroinitializer, i32 1)
+  %splice.v2f16 = call <2 x half> @llvm.vector.splice.v2f16(<2 x half> zeroinitializer, <2 x half> zeroinitializer, i32 1)
+  %splice.v4f16 = call <4 x half> @llvm.vector.splice.v4f16(<4 x half> zeroinitializer, <4 x half> zeroinitializer, i32 1)
+  %splice.v8f16 = call <8 x half> @llvm.vector.splice.v8f16(<8 x half> zeroinitializer, <8 x half> zeroinitializer, i32 1)
+  %splice.v16f16 = call <16 x half> @llvm.vector.splice.v16f16(<16 x half> zeroinitializer, <16 x half> zeroinitializer, i32 1)
+  %splice.v2f32 = call <2 x float> @llvm.vector.splice.v2f32(<2 x float> zeroinitializer, <2 x float> zeroinitializer, i32 1)
+  %splice.v4f32 = call <4 x float> @llvm.vector.splice.v4f32(<4 x float> zeroinitializer, <4 x float> zeroinitializer, i32 1)
+  %splice.v8f32 = call <8 x float> @llvm.vector.splice.v8f32(<8 x float> zeroinitializer, <8 x float> zeroinitializer, i32 1)
+  %splice.v2f64 = call <2 x double> @llvm.vector.splice.v2f64(<2 x double> zeroinitializer, <2 x double> zeroinitializer, i32 1)
+  %splice.v4f64 = call <4 x double> @llvm.vector.splice.v4f64(<4 x double> zeroinitializer, <4 x double> zeroinitializer, i32 1)
+  %splice.v2bf16 = call <2 x bfloat> @llvm.vector.splice.v2bf16(<2 x bfloat> zeroinitializer, <2 x bfloat> zeroinitializer, i32 1)
+  %splice.v4bf16 = call <4 x bfloat> @llvm.vector.splice.v4bf16(<4 x bfloat> zeroinitializer, <4 x bfloat> zeroinitializer, i32 1)
+  %splice.v8bf16 = call <8 x bfloat> @llvm.vector.splice.v8bf16(<8 x bfloat> zeroinitializer, <8 x bfloat> zeroinitializer, i32 1)
+  %splice.v16bf16 = call <16 x bfloat> @llvm.vector.splice.v16bf16(<16 x bfloat> zeroinitializer, <16 x bfloat> zeroinitializer, i32 1)
+  %splice.v16i1 = call <16 x i1> @llvm.vector.splice.v16i1(<16 x i1> zeroinitializer, <16 x i1> zeroinitializer, i32 1)
+  %splice.v8i1 =  call <8 x i1> @llvm.vector.splice.v8i1(<8 x i1> zeroinitializer, <8 x i1> zeroinitializer, i32 1)
+  %splice.v4i1 = call <4 x i1> @llvm.vector.splice.v4i1(<4 x i1> zeroinitializer, <4 x i1> zeroinitializer, i32 1)
+  %splice.v2i1 = call <2 x i1> @llvm.vector.splice.v2i1(<2 x i1> zeroinitializer, <2 x i1> zeroinitializer, i32 1)
+  %splice.v2i128 = call <2 x i128> @llvm.vector.splice.v2i128(<2 x i128> zeroinitializer, <2 x i128> zeroinitializer, i32 1)
   ret void
 }
 
-declare <2 x i1> @llvm.experimental.vector.splice.v2i1(<2 x i1>, <2 x i1>, i32)
-declare <4 x i1> @llvm.experimental.vector.splice.v4i1(<4 x i1>, <4 x i1>, i32)
-declare <8 x i1> @llvm.experimental.vector.splice.v8i1(<8 x i1>, <8 x i1>, i32)
-declare <16 x i1> @llvm.experimental.vector.splice.v16i1(<16 x i1>, <16 x i1>, i32)
-declare <2 x i8> @llvm.experimental.vector.splice.v2i8(<2 x i8>, <2 x i8>, i32)
-declare <16 x i8> @llvm.experimental.vector.splice.v16i8(<16 x i8>, <16 x i8>, i32)
-declare <32 x i8> @llvm.experimental.vector.splice.v32i8(<32 x i8>, <32 x i8>, i32)
-declare <2 x i16> @llvm.experimental.vector.splice.v2i16(<2 x i16>, <2 x i16>, i32)
-declare <4 x i16> @llvm.experimental.vector.splice.v4i16(<4 x i16>, <4 x i16>, i32)
-declare <8 x i16> @llvm.experimental.vector.splice.v8i16(<8 x i16>, <8 x i16>, i32)
-declare <16 x i16> @llvm.experimental.vector.splice.v16i16(<16 x i16>, <16 x i16>, i32)
-declare <4 x i32> @llvm.experimental.vector.splice.v4i32(<4 x i32>, <4 x i32>, i32)
-declare <8 x i32> @llvm.experimental.vector.splice.v8i32(<8 x i32>, <8 x i32>, i32)
-declare <2 x i64> @llvm.experimental.vector.splice.v2i64(<2 x i64>, <2 x i64>, i32)
-declare <4 x i64> @llvm.experimental.vector.splice.v4i64(<4 x i64>, <4 x i64>, i32)
-declare <2 x half> @llvm.experimental.vector.splice.v2f16(<2 x half>, <2 x half>, i32)
-declare <4 x half> @llvm.experimental.vector.splice.v4f16(<4 x half>, <4 x half>, i32)
-declare <8 x half> @llvm.experimental.vector.splice.v8f16(<8 x half>, <8 x half>, i32)
-declare <16 x half> @llvm.experimental.vector.splice.v16f16(<16 x half>, <16 x half>, i32)
-declare <2 x bfloat> @llvm.experimental.vector.splice.v2bf16(<2 x bfloat>, <2 x bfloat>, i32)
-declare <4 x bfloat> @llvm.experimental.vector.splice.v4bf16(<4 x bfloat>, <4 x bfloat>, i32)
-declare <8 x bfloat> @llvm.experimental.vector.splice.v8bf16(<8 x bfloat>, <8 x bfloat>, i32)
-declare <16 x bfloat> @llvm.experimental.vector.splice.v16bf16(<16 x bfloat>, <16 x bfloat>, i32)
-declare <2 x float> @llvm.experimental.vector.splice.v2f32(<2 x float>, <2 x float>, i32)
-declare <4 x float> @llvm.experimental.vector.splice.v4f32(<4 x float>, <4 x float>, i32)
-declare <8 x float> @llvm.experimental.vector.splice.v8f32(<8 x float>, <8 x float>, i32)
-declare <16 x float> @llvm.experimental.vector.splice.v16f32(<16 x float>, <16 x float>, i32)
-declare <2 x double> @llvm.experimental.vector.splice.v2f64(<2 x double>, <2 x double>, i32)
-declare <4 x double> @llvm.experimental.vector.splice.v4f64(<4 x double>, <4 x double>, i32)
-declare <2 x i128> @llvm.experimental.vector.splice.v2i128(<2 x i128>, <2 x i128>, i32)
+declare <2 x i1> @llvm.vector.splice.v2i1(<2 x i1>, <2 x i1>, i32)
+declare <4 x i1> @llvm.vector.splice.v4i1(<4 x i1>, <4 x i1>, i32)
+declare <8 x i1> @llvm.vector.splice.v8i1(<8 x i1>, <8 x i1>, i32)
+declare <16 x i1> @llvm.vector.splice.v16i1(<16 x i1>, <16 x i1>, i32)
+declare <2 x i8> @llvm.vector.splice.v2i8(<2 x i8>, <2 x i8>, i32)
+declare <16 x i8> @llvm.vector.splice.v16i8(<16 x i8>, <16 x i8>, i32)
+declare <32 x i8> @llvm.vector.splice.v32i8(<32 x i8>, <32 x i8>, i32)
+declare <2 x i16> @llvm.vector.splice.v2i16(<2 x i16>, <2 x i16>, i32)
+declare <4 x i16> @llvm.vector.splice.v4i16(<4 x i16>, <4 x i16>, i32)
+declare <8 x i16> @llvm.vector.splice.v8i16(<8 x i16>, <8 x i16>, i32)
+declare <16 x i16> @llvm.vector.splice.v16i16(<16 x i16>, <16 x i16>, i32)
+declare <4 x i32> @llvm.vector.splice.v4i32(<4 x i32>, <4 x i32>, i32)
+declare <8 x i32> @llvm.vector.splice.v8i32(<8 x i32>, <8 x i32>, i32)
+declare <2 x i64> @llvm.vector.splice.v2i64(<2 x i64>, <2 x i64>, i32)
+declare <4 x i64> @llvm.vector.splice.v4i64(<4 x i64>, <4 x i64>, i32)
+declare <2 x half> @llvm.vector.splice.v2f16(<2 x half>, <2 x half>, i32)
+declare <4 x half> @llvm.vector.splice.v4f16(<4 x half>, <4 x half>, i32)
+declare <8 x half> @llvm.vector.splice.v8f16(<8 x half>, <8 x half>, i32)
+declare <16 x half> @llvm.vector.splice.v16f16(<16 x half>, <16 x half>, i32)
+declare <2 x bfloat> @llvm.vector.splice.v2bf16(<2 x bfloat>, <2 x bfloat>, i32)
+declare <4 x bfloat> @llvm.vector.splice.v4bf16(<4 x bfloat>, <4 x bfloat>, i32)
+declare <8 x bfloat> @llvm.vector.splice.v8bf16(<8 x bfloat>, <8 x bfloat>, i32)
+declare <16 x bfloat> @llvm.vector.splice.v16bf16(<16 x bfloat>, <16 x bfloat>, i32)
+declare <2 x float> @llvm.vector.splice.v2f32(<2 x float>, <2 x float>, i32)
+declare <4 x float> @llvm.vector.splice.v4f32(<4 x float>, <4 x float>, i32)
+declare <8 x float> @llvm.vector.splice.v8f32(<8 x float>, <8 x float>, i32)
+declare <16 x float> @llvm.vector.splice.v16f32(<16 x float>, <16 x float>, i32)
+declare <2 x double> @llvm.vector.splice.v2f64(<2 x double>, <2 x double>, i32)
+declare <4 x double> @llvm.vector.splice.v4f64(<4 x double>, <4 x double>, i32)
+declare <2 x i128> @llvm.vector.splice.v2i128(<2 x i128>, <2 x i128>, i32)
 
 attributes #0 = { "target-features"="+bf16" }
diff --git a/llvm/test/Analysis/CostModel/AArch64/sve-intrinsics.ll b/llvm/test/Analysis/CostModel/AArch64/sve-intrinsics.ll
index 8d5535e2a82f..15c278b060c9 100644
--- a/llvm/test/Analysis/CostModel/AArch64/sve-intrinsics.ll
+++ b/llvm/test/Analysis/CostModel/AArch64/sve-intrinsics.ll
@@ -270,122 +270,122 @@ declare <vscale x 4 x i32> @llvm.cttz.nxv4i32(<vscale x 4 x i32>, i1)
 
 define void @vector_reverse() #0 {
 ; CHECK-LABEL: 'vector_reverse'
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %reverse_nxv16i8 = call <vscale x 16 x i8> @llvm.experimental.vector.reverse.nxv16i8(<vscale x 16 x i8> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %reverse_nxv32i8 = call <vscale x 32 x i8> @llvm.experimental.vector.reverse.nxv32i8(<vscale x 32 x i8> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %reverse_nxv2i16 = call <vscale x 2 x i16> @llvm.experimental.vector.reverse.nxv2i16(<vscale x 2 x i16> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %reverse_nxv4i16 = call <vscale x 4 x i16> @llvm.experimental.vector.reverse.nxv4i16(<vscale x 4 x i16> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %reverse_nxv8i16 = call <vscale x 8 x i16> @llvm.experimental.vector.reverse.nxv8i16(<vscale x 8 x i16> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %reverse_nxv16i16 = call <vscale x 16 x i16> @llvm.experimental.vector.reverse.nxv16i16(<vscale x 16 x i16> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %reverse_nxv4i32 = call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %reverse_nxv8i32 = call <vscale x 8 x i32> @llvm.experimental.vector.reverse.nxv8i32(<vscale x 8 x i32> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %reverse_nxv2i64 = call <vscale x 2 x i64> @llvm.experimental.vector.reverse.nxv2i64(<vscale x 2 x i64> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %reverse_nxv4i64 = call <vscale x 4 x i64> @llvm.experimental.vector.reverse.nxv4i64(<vscale x 4 x i64> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %reverse_nxv2f16 = call <vscale x 2 x half> @llvm.experimental.vector.reverse.nxv2f16(<vscale x 2 x half> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %reverse_nxv4f16 = call <vscale x 4 x half> @llvm.experimental.vector.reverse.nxv4f16(<vscale x 4 x half> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %reverse_nxv8f16 = call <vscale x 8 x half> @llvm.experimental.vector.reverse.nxv8f16(<vscale x 8 x half> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %reverse_nxv16f16 = call <vscale x 16 x half> @llvm.experimental.vector.reverse.nxv16f16(<vscale x 16 x half> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %reverse_nxv2f32 = call <vscale x 2 x float> @llvm.experimental.vector.reverse.nxv2f32(<vscale x 2 x float> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %reverse_nxv4f32 = call <vscale x 4 x float> @llvm.experimental.vector.reverse.nxv4f32(<vscale x 4 x float> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %reverse_nxv8f32 = call <vscale x 8 x float> @llvm.experimental.vector.reverse.nxv8f32(<vscale x 8 x float> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %reverse_nxv2f64 = call <vscale x 2 x double> @llvm.experimental.vector.reverse.nxv2f64(<vscale x 2 x double> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %reverse_nxv4f64 = call <vscale x 4 x double> @llvm.experimental.vector.reverse.nxv4f64(<vscale x 4 x double> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %reverse_nxv2bf16 = call <vscale x 2 x bfloat> @llvm.experimental.vector.reverse.nxv2bf16(<vscale x 2 x bfloat> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %reverse_nxv4bf16 = call <vscale x 4 x bfloat> @llvm.experimental.vector.reverse.nxv4bf16(<vscale x 4 x bfloat> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %reverse_nxv8bf16 = call <vscale x 8 x bfloat> @llvm.experimental.vector.reverse.nxv8bf16(<vscale x 8 x bfloat> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %reverse_nxv16bf16 = call <vscale x 16 x bfloat> @llvm.experimental.vector.reverse.nxv16bf16(<vscale x 16 x bfloat> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %reverse_nxv16i1 = call <vscale x 16 x i1> @llvm.experimental.vector.reverse.nxv16i1(<vscale x 16 x i1> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %reverse_nxv8i1 = call <vscale x 8 x i1> @llvm.experimental.vector.reverse.nxv8i1(<vscale x 8 x i1> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %reverse_nxv4i1 = call <vscale x 4 x i1> @llvm.experimental.vector.reverse.nxv4i1(<vscale x 4 x i1> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %reverse_nxv2i1 = call <vscale x 2 x i1> @llvm.experimental.vector.reverse.nxv2i1(<vscale x 2 x i1> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %reverse_nxv16i8 = call <vscale x 16 x i8> @llvm.vector.reverse.nxv16i8(<vscale x 16 x i8> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %reverse_nxv32i8 = call <vscale x 32 x i8> @llvm.vector.reverse.nxv32i8(<vscale x 32 x i8> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %reverse_nxv2i16 = call <vscale x 2 x i16> @llvm.vector.reverse.nxv2i16(<vscale x 2 x i16> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %reverse_nxv4i16 = call <vscale x 4 x i16> @llvm.vector.reverse.nxv4i16(<vscale x 4 x i16> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %reverse_nxv8i16 = call <vscale x 8 x i16> @llvm.vector.reverse.nxv8i16(<vscale x 8 x i16> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %reverse_nxv16i16 = call <vscale x 16 x i16> @llvm.vector.reverse.nxv16i16(<vscale x 16 x i16> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %reverse_nxv4i32 = call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %reverse_nxv8i32 = call <vscale x 8 x i32> @llvm.vector.reverse.nxv8i32(<vscale x 8 x i32> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %reverse_nxv2i64 = call <vscale x 2 x i64> @llvm.vector.reverse.nxv2i64(<vscale x 2 x i64> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %reverse_nxv4i64 = call <vscale x 4 x i64> @llvm.vector.reverse.nxv4i64(<vscale x 4 x i64> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %reverse_nxv2f16 = call <vscale x 2 x half> @llvm.vector.reverse.nxv2f16(<vscale x 2 x half> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %reverse_nxv4f16 = call <vscale x 4 x half> @llvm.vector.reverse.nxv4f16(<vscale x 4 x half> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %reverse_nxv8f16 = call <vscale x 8 x half> @llvm.vector.reverse.nxv8f16(<vscale x 8 x half> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %reverse_nxv16f16 = call <vscale x 16 x half> @llvm.vector.reverse.nxv16f16(<vscale x 16 x half> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %reverse_nxv2f32 = call <vscale x 2 x float> @llvm.vector.reverse.nxv2f32(<vscale x 2 x float> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %reverse_nxv4f32 = call <vscale x 4 x float> @llvm.vector.reverse.nxv4f32(<vscale x 4 x float> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %reverse_nxv8f32 = call <vscale x 8 x float> @llvm.vector.reverse.nxv8f32(<vscale x 8 x float> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %reverse_nxv2f64 = call <vscale x 2 x double> @llvm.vector.reverse.nxv2f64(<vscale x 2 x double> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %reverse_nxv4f64 = call <vscale x 4 x double> @llvm.vector.reverse.nxv4f64(<vscale x 4 x double> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %reverse_nxv2bf16 = call <vscale x 2 x bfloat> @llvm.vector.reverse.nxv2bf16(<vscale x 2 x bfloat> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %reverse_nxv4bf16 = call <vscale x 4 x bfloat> @llvm.vector.reverse.nxv4bf16(<vscale x 4 x bfloat> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %reverse_nxv8bf16 = call <vscale x 8 x bfloat> @llvm.vector.reverse.nxv8bf16(<vscale x 8 x bfloat> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %reverse_nxv16bf16 = call <vscale x 16 x bfloat> @llvm.vector.reverse.nxv16bf16(<vscale x 16 x bfloat> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %reverse_nxv16i1 = call <vscale x 16 x i1> @llvm.vector.reverse.nxv16i1(<vscale x 16 x i1> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %reverse_nxv8i1 = call <vscale x 8 x i1> @llvm.vector.reverse.nxv8i1(<vscale x 8 x i1> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %reverse_nxv4i1 = call <vscale x 4 x i1> @llvm.vector.reverse.nxv4i1(<vscale x 4 x i1> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %reverse_nxv2i1 = call <vscale x 2 x i1> @llvm.vector.reverse.nxv2i1(<vscale x 2 x i1> undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
 ;
 ; TYPE_BASED_ONLY-LABEL: 'vector_reverse'
-; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %reverse_nxv16i8 = call <vscale x 16 x i8> @llvm.experimental.vector.reverse.nxv16i8(<vscale x 16 x i8> undef)
-; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %reverse_nxv32i8 = call <vscale x 32 x i8> @llvm.experimental.vector.reverse.nxv32i8(<vscale x 32 x i8> undef)
-; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %reverse_nxv2i16 = call <vscale x 2 x i16> @llvm.experimental.vector.reverse.nxv2i16(<vscale x 2 x i16> undef)
-; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %reverse_nxv4i16 = call <vscale x 4 x i16> @llvm.experimental.vector.reverse.nxv4i16(<vscale x 4 x i16> undef)
-; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %reverse_nxv8i16 = call <vscale x 8 x i16> @llvm.experimental.vector.reverse.nxv8i16(<vscale x 8 x i16> undef)
-; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %reverse_nxv16i16 = call <vscale x 16 x i16> @llvm.experimental.vector.reverse.nxv16i16(<vscale x 16 x i16> undef)
-; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %reverse_nxv4i32 = call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> undef)
-; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %reverse_nxv8i32 = call <vscale x 8 x i32> @llvm.experimental.vector.reverse.nxv8i32(<vscale x 8 x i32> undef)
-; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %reverse_nxv2i64 = call <vscale x 2 x i64> @llvm.experimental.vector.reverse.nxv2i64(<vscale x 2 x i64> undef)
-; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %reverse_nxv4i64 = call <vscale x 4 x i64> @llvm.experimental.vector.reverse.nxv4i64(<vscale x 4 x i64> undef)
-; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %reverse_nxv2f16 = call <vscale x 2 x half> @llvm.experimental.vector.reverse.nxv2f16(<vscale x 2 x half> undef)
-; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %reverse_nxv4f16 = call <vscale x 4 x half> @llvm.experimental.vector.reverse.nxv4f16(<vscale x 4 x half> undef)
-; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %reverse_nxv8f16 = call <vscale x 8 x half> @llvm.experimental.vector.reverse.nxv8f16(<vscale x 8 x half> undef)
-; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %reverse_nxv16f16 = call <vscale x 16 x half> @llvm.experimental.vector.reverse.nxv16f16(<vscale x 16 x half> undef)
-; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %reverse_nxv2f32 = call <vscale x 2 x float> @llvm.experimental.vector.reverse.nxv2f32(<vscale x 2 x float> undef)
-; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %reverse_nxv4f32 = call <vscale x 4 x float> @llvm.experimental.vector.reverse.nxv4f32(<vscale x 4 x float> undef)
-; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %reverse_nxv8f32 = call <vscale x 8 x float> @llvm.experimental.vector.reverse.nxv8f32(<vscale x 8 x float> undef)
-; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %reverse_nxv2f64 = call <vscale x 2 x double> @llvm.experimental.vector.reverse.nxv2f64(<vscale x 2 x double> undef)
-; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %reverse_nxv4f64 = call <vscale x 4 x double> @llvm.experimental.vector.reverse.nxv4f64(<vscale x 4 x double> undef)
-; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %reverse_nxv2bf16 = call <vscale x 2 x bfloat> @llvm.experimental.vector.reverse.nxv2bf16(<vscale x 2 x bfloat> undef)
-; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %reverse_nxv4bf16 = call <vscale x 4 x bfloat> @llvm.experimental.vector.reverse.nxv4bf16(<vscale x 4 x bfloat> undef)
-; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %reverse_nxv8bf16 = call <vscale x 8 x bfloat> @llvm.experimental.vector.reverse.nxv8bf16(<vscale x 8 x bfloat> undef)
-; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %reverse_nxv16bf16 = call <vscale x 16 x bfloat> @llvm.experimental.vector.reverse.nxv16bf16(<vscale x 16 x bfloat> undef)
-; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %reverse_nxv16i1 = call <vscale x 16 x i1> @llvm.experimental.vector.reverse.nxv16i1(<vscale x 16 x i1> undef)
-; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %reverse_nxv8i1 = call <vscale x 8 x i1> @llvm.experimental.vector.reverse.nxv8i1(<vscale x 8 x i1> undef)
-; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %reverse_nxv4i1 = call <vscale x 4 x i1> @llvm.experimental.vector.reverse.nxv4i1(<vscale x 4 x i1> undef)
-; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %reverse_nxv2i1 = call <vscale x 2 x i1> @llvm.experimental.vector.reverse.nxv2i1(<vscale x 2 x i1> undef)
+; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %reverse_nxv16i8 = call <vscale x 16 x i8> @llvm.vector.reverse.nxv16i8(<vscale x 16 x i8> undef)
+; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %reverse_nxv32i8 = call <vscale x 32 x i8> @llvm.vector.reverse.nxv32i8(<vscale x 32 x i8> undef)
+; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %reverse_nxv2i16 = call <vscale x 2 x i16> @llvm.vector.reverse.nxv2i16(<vscale x 2 x i16> undef)
+; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %reverse_nxv4i16 = call <vscale x 4 x i16> @llvm.vector.reverse.nxv4i16(<vscale x 4 x i16> undef)
+; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %reverse_nxv8i16 = call <vscale x 8 x i16> @llvm.vector.reverse.nxv8i16(<vscale x 8 x i16> undef)
+; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %reverse_nxv16i16 = call <vscale x 16 x i16> @llvm.vector.reverse.nxv16i16(<vscale x 16 x i16> undef)
+; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %reverse_nxv4i32 = call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> undef)
+; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %reverse_nxv8i32 = call <vscale x 8 x i32> @llvm.vector.reverse.nxv8i32(<vscale x 8 x i32> undef)
+; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %reverse_nxv2i64 = call <vscale x 2 x i64> @llvm.vector.reverse.nxv2i64(<vscale x 2 x i64> undef)
+; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %reverse_nxv4i64 = call <vscale x 4 x i64> @llvm.vector.reverse.nxv4i64(<vscale x 4 x i64> undef)
+; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %reverse_nxv2f16 = call <vscale x 2 x half> @llvm.vector.reverse.nxv2f16(<vscale x 2 x half> undef)
+; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %reverse_nxv4f16 = call <vscale x 4 x half> @llvm.vector.reverse.nxv4f16(<vscale x 4 x half> undef)
+; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %reverse_nxv8f16 = call <vscale x 8 x half> @llvm.vector.reverse.nxv8f16(<vscale x 8 x half> undef)
+; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %reverse_nxv16f16 = call <vscale x 16 x half> @llvm.vector.reverse.nxv16f16(<vscale x 16 x half> undef)
+; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %reverse_nxv2f32 = call <vscale x 2 x float> @llvm.vector.reverse.nxv2f32(<vscale x 2 x float> undef)
+; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %reverse_nxv4f32 = call <vscale x 4 x float> @llvm.vector.reverse.nxv4f32(<vscale x 4 x float> undef)
+; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %reverse_nxv8f32 = call <vscale x 8 x float> @llvm.vector.reverse.nxv8f32(<vscale x 8 x float> undef)
+; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %reverse_nxv2f64 = call <vscale x 2 x double> @llvm.vector.reverse.nxv2f64(<vscale x 2 x double> undef)
+; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %reverse_nxv4f64 = call <vscale x 4 x double> @llvm.vector.reverse.nxv4f64(<vscale x 4 x double> undef)
+; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %reverse_nxv2bf16 = call <vscale x 2 x bfloat> @llvm.vector.reverse.nxv2bf16(<vscale x 2 x bfloat> undef)
+; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %reverse_nxv4bf16 = call <vscale x 4 x bfloat> @llvm.vector.reverse.nxv4bf16(<vscale x 4 x bfloat> undef)
+; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %reverse_nxv8bf16 = call <vscale x 8 x bfloat> @llvm.vector.reverse.nxv8bf16(<vscale x 8 x bfloat> undef)
+; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %reverse_nxv16bf16 = call <vscale x 16 x bfloat> @llvm.vector.reverse.nxv16bf16(<vscale x 16 x bfloat> undef)
+; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %reverse_nxv16i1 = call <vscale x 16 x i1> @llvm.vector.reverse.nxv16i1(<vscale x 16 x i1> undef)
+; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %reverse_nxv8i1 = call <vscale x 8 x i1> @llvm.vector.reverse.nxv8i1(<vscale x 8 x i1> undef)
+; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %reverse_nxv4i1 = call <vscale x 4 x i1> @llvm.vector.reverse.nxv4i1(<vscale x 4 x i1> undef)
+; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %reverse_nxv2i1 = call <vscale x 2 x i1> @llvm.vector.reverse.nxv2i1(<vscale x 2 x i1> undef)
 ; TYPE_BASED_ONLY-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
 ;
 
-  %reverse_nxv16i8 = call <vscale x 16 x i8> @llvm.experimental.vector.reverse.nxv16i8(<vscale x 16 x i8> undef)
-  %reverse_nxv32i8 = call <vscale x 32 x i8> @llvm.experimental.vector.reverse.nxv32i8(<vscale x 32 x i8> undef)
-  %reverse_nxv2i16 = call <vscale x 2 x i16> @llvm.experimental.vector.reverse.nxv2i16(<vscale x 2 x i16> undef)
-  %reverse_nxv4i16 = call <vscale x 4 x i16> @llvm.experimental.vector.reverse.nxv4i16(<vscale x 4 x i16> undef)
-  %reverse_nxv8i16 = call <vscale x 8 x i16> @llvm.experimental.vector.reverse.nxv8i16(<vscale x 8 x i16> undef)
-  %reverse_nxv16i16 = call <vscale x 16 x i16> @llvm.experimental.vector.reverse.nxv16i16(<vscale x 16 x i16> undef)
-  %reverse_nxv4i32 = call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> undef)
-  %reverse_nxv8i32 = call <vscale x 8 x i32> @llvm.experimental.vector.reverse.nxv8i32(<vscale x 8 x i32> undef)
-  %reverse_nxv2i64 = call <vscale x 2 x i64> @llvm.experimental.vector.reverse.nxv2i64(<vscale x 2 x i64> undef)
-  %reverse_nxv4i64 = call <vscale x 4 x i64> @llvm.experimental.vector.reverse.nxv4i64(<vscale x 4 x i64> undef)
-  %reverse_nxv2f16 = call <vscale x 2 x half> @llvm.experimental.vector.reverse.nxv2f16(<vscale x 2 x half> undef)
-  %reverse_nxv4f16 = call <vscale x 4 x half> @llvm.experimental.vector.reverse.nxv4f16(<vscale x 4 x half> undef)
-  %reverse_nxv8f16 = call <vscale x 8 x half> @llvm.experimental.vector.reverse.nxv8f16(<vscale x 8 x half> undef)
-  %reverse_nxv16f16 = call <vscale x 16 x half> @llvm.experimental.vector.reverse.nxv16f16(<vscale x 16 x half> undef)
-  %reverse_nxv2f32 = call <vscale x 2 x float> @llvm.experimental.vector.reverse.nxv2f32(<vscale x 2 x float> undef)
-  %reverse_nxv4f32 = call <vscale x 4 x float> @llvm.experimental.vector.reverse.nxv4f32(<vscale x 4 x float> undef)
-  %reverse_nxv8f32 = call <vscale x 8 x float> @llvm.experimental.vector.reverse.nxv8f32(<vscale x 8 x float> undef)
-  %reverse_nxv2f64 = call <vscale x 2 x double> @llvm.experimental.vector.reverse.nxv2f64(<vscale x 2 x double> undef)
-  %reverse_nxv4f64 = call <vscale x 4 x double> @llvm.experimental.vector.reverse.nxv4f64(<vscale x 4 x double> undef)
-  %reverse_nxv2bf16 = call <vscale x 2 x bfloat> @llvm.experimental.vector.reverse.nxv2bf16(<vscale x 2 x bfloat> undef)
-  %reverse_nxv4bf16 = call <vscale x 4 x bfloat> @llvm.experimental.vector.reverse.nxv4bf16(<vscale x 4 x bfloat> undef)
-  %reverse_nxv8bf16 = call <vscale x 8 x bfloat> @llvm.experimental.vector.reverse.nxv8bf16(<vscale x 8 x bfloat> undef)
-  %reverse_nxv16bf16 = call <vscale x 16 x bfloat> @llvm.experimental.vector.reverse.nxv16bf16(<vscale x 16 x bfloat> undef)
-  %reverse_nxv16i1 = call <vscale x 16 x i1> @llvm.experimental.vector.reverse.nxv16i1(<vscale x 16 x i1> undef)
-  %reverse_nxv8i1 =  call <vscale x 8 x i1> @llvm.experimental.vector.reverse.nxv8i1(<vscale x 8 x i1> undef)
-  %reverse_nxv4i1 = call <vscale x 4 x i1> @llvm.experimental.vector.reverse.nxv4i1(<vscale x 4 x i1> undef)
-  %reverse_nxv2i1 = call <vscale x 2 x i1> @llvm.experimental.vector.reverse.nxv2i1(<vscale x 2 x i1> undef)
+  %reverse_nxv16i8 = call <vscale x 16 x i8> @llvm.vector.reverse.nxv16i8(<vscale x 16 x i8> undef)
+  %reverse_nxv32i8 = call <vscale x 32 x i8> @llvm.vector.reverse.nxv32i8(<vscale x 32 x i8> undef)
+  %reverse_nxv2i16 = call <vscale x 2 x i16> @llvm.vector.reverse.nxv2i16(<vscale x 2 x i16> undef)
+  %reverse_nxv4i16 = call <vscale x 4 x i16> @llvm.vector.reverse.nxv4i16(<vscale x 4 x i16> undef)
+  %reverse_nxv8i16 = call <vscale x 8 x i16> @llvm.vector.reverse.nxv8i16(<vscale x 8 x i16> undef)
+  %reverse_nxv16i16 = call <vscale x 16 x i16> @llvm.vector.reverse.nxv16i16(<vscale x 16 x i16> undef)
+  %reverse_nxv4i32 = call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> undef)
+  %reverse_nxv8i32 = call <vscale x 8 x i32> @llvm.vector.reverse.nxv8i32(<vscale x 8 x i32> undef)
+  %reverse_nxv2i64 = call <vscale x 2 x i64> @llvm.vector.reverse.nxv2i64(<vscale x 2 x i64> undef)
+  %reverse_nxv4i64 = call <vscale x 4 x i64> @llvm.vector.reverse.nxv4i64(<vscale x 4 x i64> undef)
+  %reverse_nxv2f16 = call <vscale x 2 x half> @llvm.vector.reverse.nxv2f16(<vscale x 2 x half> undef)
+  %reverse_nxv4f16 = call <vscale x 4 x half> @llvm.vector.reverse.nxv4f16(<vscale x 4 x half> undef)
+  %reverse_nxv8f16 = call <vscale x 8 x half> @llvm.vector.reverse.nxv8f16(<vscale x 8 x half> undef)
+  %reverse_nxv16f16 = call <vscale x 16 x half> @llvm.vector.reverse.nxv16f16(<vscale x 16 x half> undef)
+  %reverse_nxv2f32 = call <vscale x 2 x float> @llvm.vector.reverse.nxv2f32(<vscale x 2 x float> undef)
+  %reverse_nxv4f32 = call <vscale x 4 x float> @llvm.vector.reverse.nxv4f32(<vscale x 4 x float> undef)
+  %reverse_nxv8f32 = call <vscale x 8 x float> @llvm.vector.reverse.nxv8f32(<vscale x 8 x float> undef)
+  %reverse_nxv2f64 = call <vscale x 2 x double> @llvm.vector.reverse.nxv2f64(<vscale x 2 x double> undef)
+  %reverse_nxv4f64 = call <vscale x 4 x double> @llvm.vector.reverse.nxv4f64(<vscale x 4 x double> undef)
+  %reverse_nxv2bf16 = call <vscale x 2 x bfloat> @llvm.vector.reverse.nxv2bf16(<vscale x 2 x bfloat> undef)
+  %reverse_nxv4bf16 = call <vscale x 4 x bfloat> @llvm.vector.reverse.nxv4bf16(<vscale x 4 x bfloat> undef)
+  %reverse_nxv8bf16 = call <vscale x 8 x bfloat> @llvm.vector.reverse.nxv8bf16(<vscale x 8 x bfloat> undef)
+  %reverse_nxv16bf16 = call <vscale x 16 x bfloat> @llvm.vector.reverse.nxv16bf16(<vscale x 16 x bfloat> undef)
+  %reverse_nxv16i1 = call <vscale x 16 x i1> @llvm.vector.reverse.nxv16i1(<vscale x 16 x i1> undef)
+  %reverse_nxv8i1 =  call <vscale x 8 x i1> @llvm.vector.reverse.nxv8i1(<vscale x 8 x i1> undef)
+  %reverse_nxv4i1 = call <vscale x 4 x i1> @llvm.vector.reverse.nxv4i1(<vscale x 4 x i1> undef)
+  %reverse_nxv2i1 = call <vscale x 2 x i1> @llvm.vector.reverse.nxv2i1(<vscale x 2 x i1> undef)
   ret void
 }
-declare <vscale x 16 x i8> @llvm.experimental.vector.reverse.nxv16i8(<vscale x 16 x i8>)
-declare <vscale x 32 x i8> @llvm.experimental.vector.reverse.nxv32i8(<vscale x 32 x i8>)
-declare <vscale x 2 x i16> @llvm.experimental.vector.reverse.nxv2i16(<vscale x 2 x i16>)
-declare <vscale x 4 x i16> @llvm.experimental.vector.reverse.nxv4i16(<vscale x 4 x i16>)
-declare <vscale x 8 x i16> @llvm.experimental.vector.reverse.nxv8i16(<vscale x 8 x i16>)
-declare <vscale x 16 x i16> @llvm.experimental.vector.reverse.nxv16i16(<vscale x 16 x i16>)
-declare <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32>)
-declare <vscale x 8 x i32> @llvm.experimental.vector.reverse.nxv8i32(<vscale x 8 x i32>)
-declare <vscale x 2 x i64> @llvm.experimental.vector.reverse.nxv2i64(<vscale x 2 x i64>)
-declare <vscale x 4 x i64> @llvm.experimental.vector.reverse.nxv4i64(<vscale x 4 x i64>)
-declare <vscale x 2 x half> @llvm.experimental.vector.reverse.nxv2f16(<vscale x 2 x half>)
-declare <vscale x 4 x half> @llvm.experimental.vector.reverse.nxv4f16(<vscale x 4 x half>)
-declare <vscale x 8 x half> @llvm.experimental.vector.reverse.nxv8f16(<vscale x 8 x half>)
-declare <vscale x 16 x half> @llvm.experimental.vector.reverse.nxv16f16(<vscale x 16 x half>)
-declare <vscale x 2 x float> @llvm.experimental.vector.reverse.nxv2f32(<vscale x 2 x float>)
-declare <vscale x 4 x float> @llvm.experimental.vector.reverse.nxv4f32(<vscale x 4 x float>)
-declare <vscale x 8 x float> @llvm.experimental.vector.reverse.nxv8f32(<vscale x 8 x float>)
-declare <vscale x 2 x double> @llvm.experimental.vector.reverse.nxv2f64(<vscale x 2 x double>)
-declare <vscale x 4 x double> @llvm.experimental.vector.reverse.nxv4f64(<vscale x 4 x double>)
-declare <vscale x 2 x bfloat> @llvm.experimental.vector.reverse.nxv2bf16(<vscale x 2 x bfloat>)
-declare <vscale x 4 x bfloat> @llvm.experimental.vector.reverse.nxv4bf16(<vscale x 4 x bfloat>)
-declare <vscale x 8 x bfloat> @llvm.experimental.vector.reverse.nxv8bf16(<vscale x 8 x bfloat>)
-declare <vscale x 16 x bfloat> @llvm.experimental.vector.reverse.nxv16bf16(<vscale x 16 x bfloat>)
-declare <vscale x 16 x i1> @llvm.experimental.vector.reverse.nxv16i1(<vscale x 16 x i1>)
-declare <vscale x 8 x i1> @llvm.experimental.vector.reverse.nxv8i1(<vscale x 8 x i1>)
-declare <vscale x 4 x i1> @llvm.experimental.vector.reverse.nxv4i1(<vscale x 4 x i1>)
-declare <vscale x 2 x i1> @llvm.experimental.vector.reverse.nxv2i1(<vscale x 2 x i1>)
+declare <vscale x 16 x i8> @llvm.vector.reverse.nxv16i8(<vscale x 16 x i8>)
+declare <vscale x 32 x i8> @llvm.vector.reverse.nxv32i8(<vscale x 32 x i8>)
+declare <vscale x 2 x i16> @llvm.vector.reverse.nxv2i16(<vscale x 2 x i16>)
+declare <vscale x 4 x i16> @llvm.vector.reverse.nxv4i16(<vscale x 4 x i16>)
+declare <vscale x 8 x i16> @llvm.vector.reverse.nxv8i16(<vscale x 8 x i16>)
+declare <vscale x 16 x i16> @llvm.vector.reverse.nxv16i16(<vscale x 16 x i16>)
+declare <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32>)
+declare <vscale x 8 x i32> @llvm.vector.reverse.nxv8i32(<vscale x 8 x i32>)
+declare <vscale x 2 x i64> @llvm.vector.reverse.nxv2i64(<vscale x 2 x i64>)
+declare <vscale x 4 x i64> @llvm.vector.reverse.nxv4i64(<vscale x 4 x i64>)
+declare <vscale x 2 x half> @llvm.vector.reverse.nxv2f16(<vscale x 2 x half>)
+declare <vscale x 4 x half> @llvm.vector.reverse.nxv4f16(<vscale x 4 x half>)
+declare <vscale x 8 x half> @llvm.vector.reverse.nxv8f16(<vscale x 8 x half>)
+declare <vscale x 16 x half> @llvm.vector.reverse.nxv16f16(<vscale x 16 x half>)
+declare <vscale x 2 x float> @llvm.vector.reverse.nxv2f32(<vscale x 2 x float>)
+declare <vscale x 4 x float> @llvm.vector.reverse.nxv4f32(<vscale x 4 x float>)
+declare <vscale x 8 x float> @llvm.vector.reverse.nxv8f32(<vscale x 8 x float>)
+declare <vscale x 2 x double> @llvm.vector.reverse.nxv2f64(<vscale x 2 x double>)
+declare <vscale x 4 x double> @llvm.vector.reverse.nxv4f64(<vscale x 4 x double>)
+declare <vscale x 2 x bfloat> @llvm.vector.reverse.nxv2bf16(<vscale x 2 x bfloat>)
+declare <vscale x 4 x bfloat> @llvm.vector.reverse.nxv4bf16(<vscale x 4 x bfloat>)
+declare <vscale x 8 x bfloat> @llvm.vector.reverse.nxv8bf16(<vscale x 8 x bfloat>)
+declare <vscale x 16 x bfloat> @llvm.vector.reverse.nxv16bf16(<vscale x 16 x bfloat>)
+declare <vscale x 16 x i1> @llvm.vector.reverse.nxv16i1(<vscale x 16 x i1>)
+declare <vscale x 8 x i1> @llvm.vector.reverse.nxv8i1(<vscale x 8 x i1>)
+declare <vscale x 4 x i1> @llvm.vector.reverse.nxv4i1(<vscale x 4 x i1>)
+declare <vscale x 2 x i1> @llvm.vector.reverse.nxv2i1(<vscale x 2 x i1>)
 
 define void @unsupported_fp_ops(<vscale x 4 x float> %vec, i32 %extraarg) {
 ; CHECK-LABEL: 'unsupported_fp_ops'
@@ -450,236 +450,236 @@ declare <vscale x 4 x float> @llvm.log10.nxv4f32(<vscale x 4 x float>)
 
 define void @vector_splice() #0 {
 ; CHECK-LABEL: 'vector_splice'
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %splice_nxv16i8 = call <vscale x 16 x i8> @llvm.experimental.vector.splice.nxv16i8(<vscale x 16 x i8> zeroinitializer, <vscale x 16 x i8> zeroinitializer, i32 1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice_nxv32i8 = call <vscale x 32 x i8> @llvm.experimental.vector.splice.nxv32i8(<vscale x 32 x i8> zeroinitializer, <vscale x 32 x i8> zeroinitializer, i32 1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %splice_nxv2i16 = call <vscale x 2 x i16> @llvm.experimental.vector.splice.nxv2i16(<vscale x 2 x i16> zeroinitializer, <vscale x 2 x i16> zeroinitializer, i32 1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %splice_nxv4i16 = call <vscale x 4 x i16> @llvm.experimental.vector.splice.nxv4i16(<vscale x 4 x i16> zeroinitializer, <vscale x 4 x i16> zeroinitializer, i32 1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %splice_nxv8i16 = call <vscale x 8 x i16> @llvm.experimental.vector.splice.nxv8i16(<vscale x 8 x i16> zeroinitializer, <vscale x 8 x i16> zeroinitializer, i32 1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice_nxv16i16 = call <vscale x 16 x i16> @llvm.experimental.vector.splice.nxv16i16(<vscale x 16 x i16> zeroinitializer, <vscale x 16 x i16> zeroinitializer, i32 1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %splice_nxv4i32 = call <vscale x 4 x i32> @llvm.experimental.vector.splice.nxv4i32(<vscale x 4 x i32> zeroinitializer, <vscale x 4 x i32> zeroinitializer, i32 1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice_nxv8i32 = call <vscale x 8 x i32> @llvm.experimental.vector.splice.nxv8i32(<vscale x 8 x i32> zeroinitializer, <vscale x 8 x i32> zeroinitializer, i32 1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %splice_nxv2i64 = call <vscale x 2 x i64> @llvm.experimental.vector.splice.nxv2i64(<vscale x 2 x i64> zeroinitializer, <vscale x 2 x i64> zeroinitializer, i32 1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice_nxv4i64 = call <vscale x 4 x i64> @llvm.experimental.vector.splice.nxv4i64(<vscale x 4 x i64> zeroinitializer, <vscale x 4 x i64> zeroinitializer, i32 1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %splice_nxv2f16 = call <vscale x 2 x half> @llvm.experimental.vector.splice.nxv2f16(<vscale x 2 x half> zeroinitializer, <vscale x 2 x half> zeroinitializer, i32 1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %splice_nxv4f16 = call <vscale x 4 x half> @llvm.experimental.vector.splice.nxv4f16(<vscale x 4 x half> zeroinitializer, <vscale x 4 x half> zeroinitializer, i32 1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %splice_nxv8f16 = call <vscale x 8 x half> @llvm.experimental.vector.splice.nxv8f16(<vscale x 8 x half> zeroinitializer, <vscale x 8 x half> zeroinitializer, i32 1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice_nxv16f16 = call <vscale x 16 x half> @llvm.experimental.vector.splice.nxv16f16(<vscale x 16 x half> zeroinitializer, <vscale x 16 x half> zeroinitializer, i32 1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %splice_nxv2f32 = call <vscale x 2 x float> @llvm.experimental.vector.splice.nxv2f32(<vscale x 2 x float> zeroinitializer, <vscale x 2 x float> zeroinitializer, i32 1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %splice_nxv4f32 = call <vscale x 4 x float> @llvm.experimental.vector.splice.nxv4f32(<vscale x 4 x float> zeroinitializer, <vscale x 4 x float> zeroinitializer, i32 1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice_nxv8f32 = call <vscale x 8 x float> @llvm.experimental.vector.splice.nxv8f32(<vscale x 8 x float> zeroinitializer, <vscale x 8 x float> zeroinitializer, i32 1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %splice_nxv2f64 = call <vscale x 2 x double> @llvm.experimental.vector.splice.nxv2f64(<vscale x 2 x double> zeroinitializer, <vscale x 2 x double> zeroinitializer, i32 1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice_nxv4f64 = call <vscale x 4 x double> @llvm.experimental.vector.splice.nxv4f64(<vscale x 4 x double> zeroinitializer, <vscale x 4 x double> zeroinitializer, i32 1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %splice_nxv2bf16 = call <vscale x 2 x bfloat> @llvm.experimental.vector.splice.nxv2bf16(<vscale x 2 x bfloat> zeroinitializer, <vscale x 2 x bfloat> zeroinitializer, i32 1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %splice_nxv4bf16 = call <vscale x 4 x bfloat> @llvm.experimental.vector.splice.nxv4bf16(<vscale x 4 x bfloat> zeroinitializer, <vscale x 4 x bfloat> zeroinitializer, i32 1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %splice_nxv8bf16 = call <vscale x 8 x bfloat> @llvm.experimental.vector.splice.nxv8bf16(<vscale x 8 x bfloat> zeroinitializer, <vscale x 8 x bfloat> zeroinitializer, i32 1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice_nxv16bf16 = call <vscale x 16 x bfloat> @llvm.experimental.vector.splice.nxv16bf16(<vscale x 16 x bfloat> zeroinitializer, <vscale x 16 x bfloat> zeroinitializer, i32 1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %splice_nxv16i1 = call <vscale x 16 x i1> @llvm.experimental.vector.splice.nxv16i1(<vscale x 16 x i1> zeroinitializer, <vscale x 16 x i1> zeroinitializer, i32 1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %splice_nxv8i1 = call <vscale x 8 x i1> @llvm.experimental.vector.splice.nxv8i1(<vscale x 8 x i1> zeroinitializer, <vscale x 8 x i1> zeroinitializer, i32 1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %splice_nxv4i1 = call <vscale x 4 x i1> @llvm.experimental.vector.splice.nxv4i1(<vscale x 4 x i1> zeroinitializer, <vscale x 4 x i1> zeroinitializer, i32 1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %splice_nxv2i1 = call <vscale x 2 x i1> @llvm.experimental.vector.splice.nxv2i1(<vscale x 2 x i1> zeroinitializer, <vscale x 2 x i1> zeroinitializer, i32 1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %splice_nxv16i8_neg = call <vscale x 16 x i8> @llvm.experimental.vector.splice.nxv16i8(<vscale x 16 x i8> zeroinitializer, <vscale x 16 x i8> zeroinitializer, i32 -1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %splice_nxv32i8_neg = call <vscale x 32 x i8> @llvm.experimental.vector.splice.nxv32i8(<vscale x 32 x i8> zeroinitializer, <vscale x 32 x i8> zeroinitializer, i32 -1)
-; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv1i16_neg = call <vscale x 1 x i16> @llvm.experimental.vector.splice.nxv1i16(<vscale x 1 x i16> zeroinitializer, <vscale x 1 x i16> zeroinitializer, i32 -1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %splice_nxv2i16_neg = call <vscale x 2 x i16> @llvm.experimental.vector.splice.nxv2i16(<vscale x 2 x i16> zeroinitializer, <vscale x 2 x i16> zeroinitializer, i32 -1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %splice_nxv4i16_neg = call <vscale x 4 x i16> @llvm.experimental.vector.splice.nxv4i16(<vscale x 4 x i16> zeroinitializer, <vscale x 4 x i16> zeroinitializer, i32 -1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %splice_nxv8i16_neg = call <vscale x 8 x i16> @llvm.experimental.vector.splice.nxv8i16(<vscale x 8 x i16> zeroinitializer, <vscale x 8 x i16> zeroinitializer, i32 -1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %splice_nxv16i16_neg = call <vscale x 16 x i16> @llvm.experimental.vector.splice.nxv16i16(<vscale x 16 x i16> zeroinitializer, <vscale x 16 x i16> zeroinitializer, i32 -1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %splice_nxv4i32_neg = call <vscale x 4 x i32> @llvm.experimental.vector.splice.nxv4i32(<vscale x 4 x i32> zeroinitializer, <vscale x 4 x i32> zeroinitializer, i32 -1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %splice_nxv8i32_neg = call <vscale x 8 x i32> @llvm.experimental.vector.splice.nxv8i32(<vscale x 8 x i32> zeroinitializer, <vscale x 8 x i32> zeroinitializer, i32 -1)
-; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv1i64_neg = call <vscale x 1 x i64> @llvm.experimental.vector.splice.nxv1i64(<vscale x 1 x i64> zeroinitializer, <vscale x 1 x i64> zeroinitializer, i32 -1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %splice_nxv2i64_neg = call <vscale x 2 x i64> @llvm.experimental.vector.splice.nxv2i64(<vscale x 2 x i64> zeroinitializer, <vscale x 2 x i64> zeroinitializer, i32 -1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %splice_nxv4i64_neg = call <vscale x 4 x i64> @llvm.experimental.vector.splice.nxv4i64(<vscale x 4 x i64> zeroinitializer, <vscale x 4 x i64> zeroinitializer, i32 -1)
-; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv1f16_neg = call <vscale x 1 x half> @llvm.experimental.vector.splice.nxv1f16(<vscale x 1 x half> zeroinitializer, <vscale x 1 x half> zeroinitializer, i32 -1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %splice_nxv2f16_neg = call <vscale x 2 x half> @llvm.experimental.vector.splice.nxv2f16(<vscale x 2 x half> zeroinitializer, <vscale x 2 x half> zeroinitializer, i32 -1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %splice_nxv4f16_neg = call <vscale x 4 x half> @llvm.experimental.vector.splice.nxv4f16(<vscale x 4 x half> zeroinitializer, <vscale x 4 x half> zeroinitializer, i32 -1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %splice_nxv8f16_neg = call <vscale x 8 x half> @llvm.experimental.vector.splice.nxv8f16(<vscale x 8 x half> zeroinitializer, <vscale x 8 x half> zeroinitializer, i32 -1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %splice_nxv16f16_neg = call <vscale x 16 x half> @llvm.experimental.vector.splice.nxv16f16(<vscale x 16 x half> zeroinitializer, <vscale x 16 x half> zeroinitializer, i32 -1)
-; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv1f32_neg = call <vscale x 1 x float> @llvm.experimental.vector.splice.nxv1f32(<vscale x 1 x float> zeroinitializer, <vscale x 1 x float> zeroinitializer, i32 -1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %splice_nxv2f32_neg = call <vscale x 2 x float> @llvm.experimental.vector.splice.nxv2f32(<vscale x 2 x float> zeroinitializer, <vscale x 2 x float> zeroinitializer, i32 -1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %splice_nxv4f32_neg = call <vscale x 4 x float> @llvm.experimental.vector.splice.nxv4f32(<vscale x 4 x float> zeroinitializer, <vscale x 4 x float> zeroinitializer, i32 -1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %splice_nxv8f32_neg = call <vscale x 8 x float> @llvm.experimental.vector.splice.nxv8f32(<vscale x 8 x float> zeroinitializer, <vscale x 8 x float> zeroinitializer, i32 -1)
-; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv1f64_neg = call <vscale x 1 x double> @llvm.experimental.vector.splice.nxv1f64(<vscale x 1 x double> zeroinitializer, <vscale x 1 x double> zeroinitializer, i32 -1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %splice_nxv2f64_neg = call <vscale x 2 x double> @llvm.experimental.vector.splice.nxv2f64(<vscale x 2 x double> zeroinitializer, <vscale x 2 x double> zeroinitializer, i32 -1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %splice_nxv4f64_neg = call <vscale x 4 x double> @llvm.experimental.vector.splice.nxv4f64(<vscale x 4 x double> zeroinitializer, <vscale x 4 x double> zeroinitializer, i32 -1)
-; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv1bf16_neg = call <vscale x 1 x bfloat> @llvm.experimental.vector.splice.nxv1bf16(<vscale x 1 x bfloat> zeroinitializer, <vscale x 1 x bfloat> zeroinitializer, i32 -1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %splice_nxv2bf16_neg = call <vscale x 2 x bfloat> @llvm.experimental.vector.splice.nxv2bf16(<vscale x 2 x bfloat> zeroinitializer, <vscale x 2 x bfloat> zeroinitializer, i32 -1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %splice_nxv4bf16_neg = call <vscale x 4 x bfloat> @llvm.experimental.vector.splice.nxv4bf16(<vscale x 4 x bfloat> zeroinitializer, <vscale x 4 x bfloat> zeroinitializer, i32 -1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %splice_nxv8bf16_neg = call <vscale x 8 x bfloat> @llvm.experimental.vector.splice.nxv8bf16(<vscale x 8 x bfloat> zeroinitializer, <vscale x 8 x bfloat> zeroinitializer, i32 -1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %splice_nxv16bf16_neg = call <vscale x 16 x bfloat> @llvm.experimental.vector.splice.nxv16bf16(<vscale x 16 x bfloat> zeroinitializer, <vscale x 16 x bfloat> zeroinitializer, i32 -1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %splice_nxv16i1_neg = call <vscale x 16 x i1> @llvm.experimental.vector.splice.nxv16i1(<vscale x 16 x i1> zeroinitializer, <vscale x 16 x i1> zeroinitializer, i32 -1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %splice_nxv8i1_neg = call <vscale x 8 x i1> @llvm.experimental.vector.splice.nxv8i1(<vscale x 8 x i1> zeroinitializer, <vscale x 8 x i1> zeroinitializer, i32 -1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %splice_nxv4i1_neg = call <vscale x 4 x i1> @llvm.experimental.vector.splice.nxv4i1(<vscale x 4 x i1> zeroinitializer, <vscale x 4 x i1> zeroinitializer, i32 -1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %splice_nxv2i1_neg = call <vscale x 2 x i1> @llvm.experimental.vector.splice.nxv2i1(<vscale x 2 x i1> zeroinitializer, <vscale x 2 x i1> zeroinitializer, i32 -1)
-; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv1i1_neg = call <vscale x 1 x i1> @llvm.experimental.vector.splice.nxv1i1(<vscale x 1 x i1> zeroinitializer, <vscale x 1 x i1> zeroinitializer, i32 -1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %splice_nxv16i8 = call <vscale x 16 x i8> @llvm.vector.splice.nxv16i8(<vscale x 16 x i8> zeroinitializer, <vscale x 16 x i8> zeroinitializer, i32 1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice_nxv32i8 = call <vscale x 32 x i8> @llvm.vector.splice.nxv32i8(<vscale x 32 x i8> zeroinitializer, <vscale x 32 x i8> zeroinitializer, i32 1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %splice_nxv2i16 = call <vscale x 2 x i16> @llvm.vector.splice.nxv2i16(<vscale x 2 x i16> zeroinitializer, <vscale x 2 x i16> zeroinitializer, i32 1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %splice_nxv4i16 = call <vscale x 4 x i16> @llvm.vector.splice.nxv4i16(<vscale x 4 x i16> zeroinitializer, <vscale x 4 x i16> zeroinitializer, i32 1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %splice_nxv8i16 = call <vscale x 8 x i16> @llvm.vector.splice.nxv8i16(<vscale x 8 x i16> zeroinitializer, <vscale x 8 x i16> zeroinitializer, i32 1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice_nxv16i16 = call <vscale x 16 x i16> @llvm.vector.splice.nxv16i16(<vscale x 16 x i16> zeroinitializer, <vscale x 16 x i16> zeroinitializer, i32 1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %splice_nxv4i32 = call <vscale x 4 x i32> @llvm.vector.splice.nxv4i32(<vscale x 4 x i32> zeroinitializer, <vscale x 4 x i32> zeroinitializer, i32 1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice_nxv8i32 = call <vscale x 8 x i32> @llvm.vector.splice.nxv8i32(<vscale x 8 x i32> zeroinitializer, <vscale x 8 x i32> zeroinitializer, i32 1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %splice_nxv2i64 = call <vscale x 2 x i64> @llvm.vector.splice.nxv2i64(<vscale x 2 x i64> zeroinitializer, <vscale x 2 x i64> zeroinitializer, i32 1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice_nxv4i64 = call <vscale x 4 x i64> @llvm.vector.splice.nxv4i64(<vscale x 4 x i64> zeroinitializer, <vscale x 4 x i64> zeroinitializer, i32 1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %splice_nxv2f16 = call <vscale x 2 x half> @llvm.vector.splice.nxv2f16(<vscale x 2 x half> zeroinitializer, <vscale x 2 x half> zeroinitializer, i32 1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %splice_nxv4f16 = call <vscale x 4 x half> @llvm.vector.splice.nxv4f16(<vscale x 4 x half> zeroinitializer, <vscale x 4 x half> zeroinitializer, i32 1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %splice_nxv8f16 = call <vscale x 8 x half> @llvm.vector.splice.nxv8f16(<vscale x 8 x half> zeroinitializer, <vscale x 8 x half> zeroinitializer, i32 1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice_nxv16f16 = call <vscale x 16 x half> @llvm.vector.splice.nxv16f16(<vscale x 16 x half> zeroinitializer, <vscale x 16 x half> zeroinitializer, i32 1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %splice_nxv2f32 = call <vscale x 2 x float> @llvm.vector.splice.nxv2f32(<vscale x 2 x float> zeroinitializer, <vscale x 2 x float> zeroinitializer, i32 1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %splice_nxv4f32 = call <vscale x 4 x float> @llvm.vector.splice.nxv4f32(<vscale x 4 x float> zeroinitializer, <vscale x 4 x float> zeroinitializer, i32 1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice_nxv8f32 = call <vscale x 8 x float> @llvm.vector.splice.nxv8f32(<vscale x 8 x float> zeroinitializer, <vscale x 8 x float> zeroinitializer, i32 1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %splice_nxv2f64 = call <vscale x 2 x double> @llvm.vector.splice.nxv2f64(<vscale x 2 x double> zeroinitializer, <vscale x 2 x double> zeroinitializer, i32 1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice_nxv4f64 = call <vscale x 4 x double> @llvm.vector.splice.nxv4f64(<vscale x 4 x double> zeroinitializer, <vscale x 4 x double> zeroinitializer, i32 1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %splice_nxv2bf16 = call <vscale x 2 x bfloat> @llvm.vector.splice.nxv2bf16(<vscale x 2 x bfloat> zeroinitializer, <vscale x 2 x bfloat> zeroinitializer, i32 1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %splice_nxv4bf16 = call <vscale x 4 x bfloat> @llvm.vector.splice.nxv4bf16(<vscale x 4 x bfloat> zeroinitializer, <vscale x 4 x bfloat> zeroinitializer, i32 1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %splice_nxv8bf16 = call <vscale x 8 x bfloat> @llvm.vector.splice.nxv8bf16(<vscale x 8 x bfloat> zeroinitializer, <vscale x 8 x bfloat> zeroinitializer, i32 1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice_nxv16bf16 = call <vscale x 16 x bfloat> @llvm.vector.splice.nxv16bf16(<vscale x 16 x bfloat> zeroinitializer, <vscale x 16 x bfloat> zeroinitializer, i32 1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %splice_nxv16i1 = call <vscale x 16 x i1> @llvm.vector.splice.nxv16i1(<vscale x 16 x i1> zeroinitializer, <vscale x 16 x i1> zeroinitializer, i32 1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %splice_nxv8i1 = call <vscale x 8 x i1> @llvm.vector.splice.nxv8i1(<vscale x 8 x i1> zeroinitializer, <vscale x 8 x i1> zeroinitializer, i32 1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %splice_nxv4i1 = call <vscale x 4 x i1> @llvm.vector.splice.nxv4i1(<vscale x 4 x i1> zeroinitializer, <vscale x 4 x i1> zeroinitializer, i32 1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %splice_nxv2i1 = call <vscale x 2 x i1> @llvm.vector.splice.nxv2i1(<vscale x 2 x i1> zeroinitializer, <vscale x 2 x i1> zeroinitializer, i32 1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %splice_nxv16i8_neg = call <vscale x 16 x i8> @llvm.vector.splice.nxv16i8(<vscale x 16 x i8> zeroinitializer, <vscale x 16 x i8> zeroinitializer, i32 -1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %splice_nxv32i8_neg = call <vscale x 32 x i8> @llvm.vector.splice.nxv32i8(<vscale x 32 x i8> zeroinitializer, <vscale x 32 x i8> zeroinitializer, i32 -1)
+; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv1i16_neg = call <vscale x 1 x i16> @llvm.vector.splice.nxv1i16(<vscale x 1 x i16> zeroinitializer, <vscale x 1 x i16> zeroinitializer, i32 -1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %splice_nxv2i16_neg = call <vscale x 2 x i16> @llvm.vector.splice.nxv2i16(<vscale x 2 x i16> zeroinitializer, <vscale x 2 x i16> zeroinitializer, i32 -1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %splice_nxv4i16_neg = call <vscale x 4 x i16> @llvm.vector.splice.nxv4i16(<vscale x 4 x i16> zeroinitializer, <vscale x 4 x i16> zeroinitializer, i32 -1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %splice_nxv8i16_neg = call <vscale x 8 x i16> @llvm.vector.splice.nxv8i16(<vscale x 8 x i16> zeroinitializer, <vscale x 8 x i16> zeroinitializer, i32 -1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %splice_nxv16i16_neg = call <vscale x 16 x i16> @llvm.vector.splice.nxv16i16(<vscale x 16 x i16> zeroinitializer, <vscale x 16 x i16> zeroinitializer, i32 -1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %splice_nxv4i32_neg = call <vscale x 4 x i32> @llvm.vector.splice.nxv4i32(<vscale x 4 x i32> zeroinitializer, <vscale x 4 x i32> zeroinitializer, i32 -1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %splice_nxv8i32_neg = call <vscale x 8 x i32> @llvm.vector.splice.nxv8i32(<vscale x 8 x i32> zeroinitializer, <vscale x 8 x i32> zeroinitializer, i32 -1)
+; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv1i64_neg = call <vscale x 1 x i64> @llvm.vector.splice.nxv1i64(<vscale x 1 x i64> zeroinitializer, <vscale x 1 x i64> zeroinitializer, i32 -1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %splice_nxv2i64_neg = call <vscale x 2 x i64> @llvm.vector.splice.nxv2i64(<vscale x 2 x i64> zeroinitializer, <vscale x 2 x i64> zeroinitializer, i32 -1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %splice_nxv4i64_neg = call <vscale x 4 x i64> @llvm.vector.splice.nxv4i64(<vscale x 4 x i64> zeroinitializer, <vscale x 4 x i64> zeroinitializer, i32 -1)
+; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv1f16_neg = call <vscale x 1 x half> @llvm.vector.splice.nxv1f16(<vscale x 1 x half> zeroinitializer, <vscale x 1 x half> zeroinitializer, i32 -1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %splice_nxv2f16_neg = call <vscale x 2 x half> @llvm.vector.splice.nxv2f16(<vscale x 2 x half> zeroinitializer, <vscale x 2 x half> zeroinitializer, i32 -1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %splice_nxv4f16_neg = call <vscale x 4 x half> @llvm.vector.splice.nxv4f16(<vscale x 4 x half> zeroinitializer, <vscale x 4 x half> zeroinitializer, i32 -1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %splice_nxv8f16_neg = call <vscale x 8 x half> @llvm.vector.splice.nxv8f16(<vscale x 8 x half> zeroinitializer, <vscale x 8 x half> zeroinitializer, i32 -1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %splice_nxv16f16_neg = call <vscale x 16 x half> @llvm.vector.splice.nxv16f16(<vscale x 16 x half> zeroinitializer, <vscale x 16 x half> zeroinitializer, i32 -1)
+; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv1f32_neg = call <vscale x 1 x float> @llvm.vector.splice.nxv1f32(<vscale x 1 x float> zeroinitializer, <vscale x 1 x float> zeroinitializer, i32 -1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %splice_nxv2f32_neg = call <vscale x 2 x float> @llvm.vector.splice.nxv2f32(<vscale x 2 x float> zeroinitializer, <vscale x 2 x float> zeroinitializer, i32 -1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %splice_nxv4f32_neg = call <vscale x 4 x float> @llvm.vector.splice.nxv4f32(<vscale x 4 x float> zeroinitializer, <vscale x 4 x float> zeroinitializer, i32 -1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %splice_nxv8f32_neg = call <vscale x 8 x float> @llvm.vector.splice.nxv8f32(<vscale x 8 x float> zeroinitializer, <vscale x 8 x float> zeroinitializer, i32 -1)
+; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv1f64_neg = call <vscale x 1 x double> @llvm.vector.splice.nxv1f64(<vscale x 1 x double> zeroinitializer, <vscale x 1 x double> zeroinitializer, i32 -1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %splice_nxv2f64_neg = call <vscale x 2 x double> @llvm.vector.splice.nxv2f64(<vscale x 2 x double> zeroinitializer, <vscale x 2 x double> zeroinitializer, i32 -1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %splice_nxv4f64_neg = call <vscale x 4 x double> @llvm.vector.splice.nxv4f64(<vscale x 4 x double> zeroinitializer, <vscale x 4 x double> zeroinitializer, i32 -1)
+; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv1bf16_neg = call <vscale x 1 x bfloat> @llvm.vector.splice.nxv1bf16(<vscale x 1 x bfloat> zeroinitializer, <vscale x 1 x bfloat> zeroinitializer, i32 -1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %splice_nxv2bf16_neg = call <vscale x 2 x bfloat> @llvm.vector.splice.nxv2bf16(<vscale x 2 x bfloat> zeroinitializer, <vscale x 2 x bfloat> zeroinitializer, i32 -1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %splice_nxv4bf16_neg = call <vscale x 4 x bfloat> @llvm.vector.splice.nxv4bf16(<vscale x 4 x bfloat> zeroinitializer, <vscale x 4 x bfloat> zeroinitializer, i32 -1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %splice_nxv8bf16_neg = call <vscale x 8 x bfloat> @llvm.vector.splice.nxv8bf16(<vscale x 8 x bfloat> zeroinitializer, <vscale x 8 x bfloat> zeroinitializer, i32 -1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %splice_nxv16bf16_neg = call <vscale x 16 x bfloat> @llvm.vector.splice.nxv16bf16(<vscale x 16 x bfloat> zeroinitializer, <vscale x 16 x bfloat> zeroinitializer, i32 -1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %splice_nxv16i1_neg = call <vscale x 16 x i1> @llvm.vector.splice.nxv16i1(<vscale x 16 x i1> zeroinitializer, <vscale x 16 x i1> zeroinitializer, i32 -1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %splice_nxv8i1_neg = call <vscale x 8 x i1> @llvm.vector.splice.nxv8i1(<vscale x 8 x i1> zeroinitializer, <vscale x 8 x i1> zeroinitializer, i32 -1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %splice_nxv4i1_neg = call <vscale x 4 x i1> @llvm.vector.splice.nxv4i1(<vscale x 4 x i1> zeroinitializer, <vscale x 4 x i1> zeroinitializer, i32 -1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 5 for instruction: %splice_nxv2i1_neg = call <vscale x 2 x i1> @llvm.vector.splice.nxv2i1(<vscale x 2 x i1> zeroinitializer, <vscale x 2 x i1> zeroinitializer, i32 -1)
+; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv1i1_neg = call <vscale x 1 x i1> @llvm.vector.splice.nxv1i1(<vscale x 1 x i1> zeroinitializer, <vscale x 1 x i1> zeroinitializer, i32 -1)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
 ;
 ; TYPE_BASED_ONLY-LABEL: 'vector_splice'
-; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv16i8 = call <vscale x 16 x i8> @llvm.experimental.vector.splice.nxv16i8(<vscale x 16 x i8> zeroinitializer, <vscale x 16 x i8> zeroinitializer, i32 1)
-; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv32i8 = call <vscale x 32 x i8> @llvm.experimental.vector.splice.nxv32i8(<vscale x 32 x i8> zeroinitializer, <vscale x 32 x i8> zeroinitializer, i32 1)
-; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv2i16 = call <vscale x 2 x i16> @llvm.experimental.vector.splice.nxv2i16(<vscale x 2 x i16> zeroinitializer, <vscale x 2 x i16> zeroinitializer, i32 1)
-; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv4i16 = call <vscale x 4 x i16> @llvm.experimental.vector.splice.nxv4i16(<vscale x 4 x i16> zeroinitializer, <vscale x 4 x i16> zeroinitializer, i32 1)
-; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv8i16 = call <vscale x 8 x i16> @llvm.experimental.vector.splice.nxv8i16(<vscale x 8 x i16> zeroinitializer, <vscale x 8 x i16> zeroinitializer, i32 1)
-; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv16i16 = call <vscale x 16 x i16> @llvm.experimental.vector.splice.nxv16i16(<vscale x 16 x i16> zeroinitializer, <vscale x 16 x i16> zeroinitializer, i32 1)
-; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv4i32 = call <vscale x 4 x i32> @llvm.experimental.vector.splice.nxv4i32(<vscale x 4 x i32> zeroinitializer, <vscale x 4 x i32> zeroinitializer, i32 1)
-; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv8i32 = call <vscale x 8 x i32> @llvm.experimental.vector.splice.nxv8i32(<vscale x 8 x i32> zeroinitializer, <vscale x 8 x i32> zeroinitializer, i32 1)
-; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv2i64 = call <vscale x 2 x i64> @llvm.experimental.vector.splice.nxv2i64(<vscale x 2 x i64> zeroinitializer, <vscale x 2 x i64> zeroinitializer, i32 1)
-; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv4i64 = call <vscale x 4 x i64> @llvm.experimental.vector.splice.nxv4i64(<vscale x 4 x i64> zeroinitializer, <vscale x 4 x i64> zeroinitializer, i32 1)
-; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv2f16 = call <vscale x 2 x half> @llvm.experimental.vector.splice.nxv2f16(<vscale x 2 x half> zeroinitializer, <vscale x 2 x half> zeroinitializer, i32 1)
-; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv4f16 = call <vscale x 4 x half> @llvm.experimental.vector.splice.nxv4f16(<vscale x 4 x half> zeroinitializer, <vscale x 4 x half> zeroinitializer, i32 1)
-; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv8f16 = call <vscale x 8 x half> @llvm.experimental.vector.splice.nxv8f16(<vscale x 8 x half> zeroinitializer, <vscale x 8 x half> zeroinitializer, i32 1)
-; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv16f16 = call <vscale x 16 x half> @llvm.experimental.vector.splice.nxv16f16(<vscale x 16 x half> zeroinitializer, <vscale x 16 x half> zeroinitializer, i32 1)
-; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv2f32 = call <vscale x 2 x float> @llvm.experimental.vector.splice.nxv2f32(<vscale x 2 x float> zeroinitializer, <vscale x 2 x float> zeroinitializer, i32 1)
-; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv4f32 = call <vscale x 4 x float> @llvm.experimental.vector.splice.nxv4f32(<vscale x 4 x float> zeroinitializer, <vscale x 4 x float> zeroinitializer, i32 1)
-; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv8f32 = call <vscale x 8 x float> @llvm.experimental.vector.splice.nxv8f32(<vscale x 8 x float> zeroinitializer, <vscale x 8 x float> zeroinitializer, i32 1)
-; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv2f64 = call <vscale x 2 x double> @llvm.experimental.vector.splice.nxv2f64(<vscale x 2 x double> zeroinitializer, <vscale x 2 x double> zeroinitializer, i32 1)
-; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv4f64 = call <vscale x 4 x double> @llvm.experimental.vector.splice.nxv4f64(<vscale x 4 x double> zeroinitializer, <vscale x 4 x double> zeroinitializer, i32 1)
-; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv2bf16 = call <vscale x 2 x bfloat> @llvm.experimental.vector.splice.nxv2bf16(<vscale x 2 x bfloat> zeroinitializer, <vscale x 2 x bfloat> zeroinitializer, i32 1)
-; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv4bf16 = call <vscale x 4 x bfloat> @llvm.experimental.vector.splice.nxv4bf16(<vscale x 4 x bfloat> zeroinitializer, <vscale x 4 x bfloat> zeroinitializer, i32 1)
-; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv8bf16 = call <vscale x 8 x bfloat> @llvm.experimental.vector.splice.nxv8bf16(<vscale x 8 x bfloat> zeroinitializer, <vscale x 8 x bfloat> zeroinitializer, i32 1)
-; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv16bf16 = call <vscale x 16 x bfloat> @llvm.experimental.vector.splice.nxv16bf16(<vscale x 16 x bfloat> zeroinitializer, <vscale x 16 x bfloat> zeroinitializer, i32 1)
-; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv16i1 = call <vscale x 16 x i1> @llvm.experimental.vector.splice.nxv16i1(<vscale x 16 x i1> zeroinitializer, <vscale x 16 x i1> zeroinitializer, i32 1)
-; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv8i1 = call <vscale x 8 x i1> @llvm.experimental.vector.splice.nxv8i1(<vscale x 8 x i1> zeroinitializer, <vscale x 8 x i1> zeroinitializer, i32 1)
-; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv4i1 = call <vscale x 4 x i1> @llvm.experimental.vector.splice.nxv4i1(<vscale x 4 x i1> zeroinitializer, <vscale x 4 x i1> zeroinitializer, i32 1)
-; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv2i1 = call <vscale x 2 x i1> @llvm.experimental.vector.splice.nxv2i1(<vscale x 2 x i1> zeroinitializer, <vscale x 2 x i1> zeroinitializer, i32 1)
-; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv16i8_neg = call <vscale x 16 x i8> @llvm.experimental.vector.splice.nxv16i8(<vscale x 16 x i8> zeroinitializer, <vscale x 16 x i8> zeroinitializer, i32 -1)
-; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv32i8_neg = call <vscale x 32 x i8> @llvm.experimental.vector.splice.nxv32i8(<vscale x 32 x i8> zeroinitializer, <vscale x 32 x i8> zeroinitializer, i32 -1)
-; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv1i16_neg = call <vscale x 1 x i16> @llvm.experimental.vector.splice.nxv1i16(<vscale x 1 x i16> zeroinitializer, <vscale x 1 x i16> zeroinitializer, i32 -1)
-; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv2i16_neg = call <vscale x 2 x i16> @llvm.experimental.vector.splice.nxv2i16(<vscale x 2 x i16> zeroinitializer, <vscale x 2 x i16> zeroinitializer, i32 -1)
-; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv4i16_neg = call <vscale x 4 x i16> @llvm.experimental.vector.splice.nxv4i16(<vscale x 4 x i16> zeroinitializer, <vscale x 4 x i16> zeroinitializer, i32 -1)
-; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv8i16_neg = call <vscale x 8 x i16> @llvm.experimental.vector.splice.nxv8i16(<vscale x 8 x i16> zeroinitializer, <vscale x 8 x i16> zeroinitializer, i32 -1)
-; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv16i16_neg = call <vscale x 16 x i16> @llvm.experimental.vector.splice.nxv16i16(<vscale x 16 x i16> zeroinitializer, <vscale x 16 x i16> zeroinitializer, i32 -1)
-; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv4i32_neg = call <vscale x 4 x i32> @llvm.experimental.vector.splice.nxv4i32(<vscale x 4 x i32> zeroinitializer, <vscale x 4 x i32> zeroinitializer, i32 -1)
-; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv8i32_neg = call <vscale x 8 x i32> @llvm.experimental.vector.splice.nxv8i32(<vscale x 8 x i32> zeroinitializer, <vscale x 8 x i32> zeroinitializer, i32 -1)
-; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv1i64_neg = call <vscale x 1 x i64> @llvm.experimental.vector.splice.nxv1i64(<vscale x 1 x i64> zeroinitializer, <vscale x 1 x i64> zeroinitializer, i32 -1)
-; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv2i64_neg = call <vscale x 2 x i64> @llvm.experimental.vector.splice.nxv2i64(<vscale x 2 x i64> zeroinitializer, <vscale x 2 x i64> zeroinitializer, i32 -1)
-; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv4i64_neg = call <vscale x 4 x i64> @llvm.experimental.vector.splice.nxv4i64(<vscale x 4 x i64> zeroinitializer, <vscale x 4 x i64> zeroinitializer, i32 -1)
-; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv1f16_neg = call <vscale x 1 x half> @llvm.experimental.vector.splice.nxv1f16(<vscale x 1 x half> zeroinitializer, <vscale x 1 x half> zeroinitializer, i32 -1)
-; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv2f16_neg = call <vscale x 2 x half> @llvm.experimental.vector.splice.nxv2f16(<vscale x 2 x half> zeroinitializer, <vscale x 2 x half> zeroinitializer, i32 -1)
-; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv4f16_neg = call <vscale x 4 x half> @llvm.experimental.vector.splice.nxv4f16(<vscale x 4 x half> zeroinitializer, <vscale x 4 x half> zeroinitializer, i32 -1)
-; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv8f16_neg = call <vscale x 8 x half> @llvm.experimental.vector.splice.nxv8f16(<vscale x 8 x half> zeroinitializer, <vscale x 8 x half> zeroinitializer, i32 -1)
-; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv16f16_neg = call <vscale x 16 x half> @llvm.experimental.vector.splice.nxv16f16(<vscale x 16 x half> zeroinitializer, <vscale x 16 x half> zeroinitializer, i32 -1)
-; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv1f32_neg = call <vscale x 1 x float> @llvm.experimental.vector.splice.nxv1f32(<vscale x 1 x float> zeroinitializer, <vscale x 1 x float> zeroinitializer, i32 -1)
-; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv2f32_neg = call <vscale x 2 x float> @llvm.experimental.vector.splice.nxv2f32(<vscale x 2 x float> zeroinitializer, <vscale x 2 x float> zeroinitializer, i32 -1)
-; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv4f32_neg = call <vscale x 4 x float> @llvm.experimental.vector.splice.nxv4f32(<vscale x 4 x float> zeroinitializer, <vscale x 4 x float> zeroinitializer, i32 -1)
-; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv8f32_neg = call <vscale x 8 x float> @llvm.experimental.vector.splice.nxv8f32(<vscale x 8 x float> zeroinitializer, <vscale x 8 x float> zeroinitializer, i32 -1)
-; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv1f64_neg = call <vscale x 1 x double> @llvm.experimental.vector.splice.nxv1f64(<vscale x 1 x double> zeroinitializer, <vscale x 1 x double> zeroinitializer, i32 -1)
-; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv2f64_neg = call <vscale x 2 x double> @llvm.experimental.vector.splice.nxv2f64(<vscale x 2 x double> zeroinitializer, <vscale x 2 x double> zeroinitializer, i32 -1)
-; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv4f64_neg = call <vscale x 4 x double> @llvm.experimental.vector.splice.nxv4f64(<vscale x 4 x double> zeroinitializer, <vscale x 4 x double> zeroinitializer, i32 -1)
-; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv1bf16_neg = call <vscale x 1 x bfloat> @llvm.experimental.vector.splice.nxv1bf16(<vscale x 1 x bfloat> zeroinitializer, <vscale x 1 x bfloat> zeroinitializer, i32 -1)
-; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv2bf16_neg = call <vscale x 2 x bfloat> @llvm.experimental.vector.splice.nxv2bf16(<vscale x 2 x bfloat> zeroinitializer, <vscale x 2 x bfloat> zeroinitializer, i32 -1)
-; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv4bf16_neg = call <vscale x 4 x bfloat> @llvm.experimental.vector.splice.nxv4bf16(<vscale x 4 x bfloat> zeroinitializer, <vscale x 4 x bfloat> zeroinitializer, i32 -1)
-; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv8bf16_neg = call <vscale x 8 x bfloat> @llvm.experimental.vector.splice.nxv8bf16(<vscale x 8 x bfloat> zeroinitializer, <vscale x 8 x bfloat> zeroinitializer, i32 -1)
-; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv16bf16_neg = call <vscale x 16 x bfloat> @llvm.experimental.vector.splice.nxv16bf16(<vscale x 16 x bfloat> zeroinitializer, <vscale x 16 x bfloat> zeroinitializer, i32 -1)
-; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv16i1_neg = call <vscale x 16 x i1> @llvm.experimental.vector.splice.nxv16i1(<vscale x 16 x i1> zeroinitializer, <vscale x 16 x i1> zeroinitializer, i32 -1)
-; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv8i1_neg = call <vscale x 8 x i1> @llvm.experimental.vector.splice.nxv8i1(<vscale x 8 x i1> zeroinitializer, <vscale x 8 x i1> zeroinitializer, i32 -1)
-; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv4i1_neg = call <vscale x 4 x i1> @llvm.experimental.vector.splice.nxv4i1(<vscale x 4 x i1> zeroinitializer, <vscale x 4 x i1> zeroinitializer, i32 -1)
-; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv2i1_neg = call <vscale x 2 x i1> @llvm.experimental.vector.splice.nxv2i1(<vscale x 2 x i1> zeroinitializer, <vscale x 2 x i1> zeroinitializer, i32 -1)
-; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv1i1_neg = call <vscale x 1 x i1> @llvm.experimental.vector.splice.nxv1i1(<vscale x 1 x i1> zeroinitializer, <vscale x 1 x i1> zeroinitializer, i32 -1)
+; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv16i8 = call <vscale x 16 x i8> @llvm.vector.splice.nxv16i8(<vscale x 16 x i8> zeroinitializer, <vscale x 16 x i8> zeroinitializer, i32 1)
+; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv32i8 = call <vscale x 32 x i8> @llvm.vector.splice.nxv32i8(<vscale x 32 x i8> zeroinitializer, <vscale x 32 x i8> zeroinitializer, i32 1)
+; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv2i16 = call <vscale x 2 x i16> @llvm.vector.splice.nxv2i16(<vscale x 2 x i16> zeroinitializer, <vscale x 2 x i16> zeroinitializer, i32 1)
+; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv4i16 = call <vscale x 4 x i16> @llvm.vector.splice.nxv4i16(<vscale x 4 x i16> zeroinitializer, <vscale x 4 x i16> zeroinitializer, i32 1)
+; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv8i16 = call <vscale x 8 x i16> @llvm.vector.splice.nxv8i16(<vscale x 8 x i16> zeroinitializer, <vscale x 8 x i16> zeroinitializer, i32 1)
+; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv16i16 = call <vscale x 16 x i16> @llvm.vector.splice.nxv16i16(<vscale x 16 x i16> zeroinitializer, <vscale x 16 x i16> zeroinitializer, i32 1)
+; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv4i32 = call <vscale x 4 x i32> @llvm.vector.splice.nxv4i32(<vscale x 4 x i32> zeroinitializer, <vscale x 4 x i32> zeroinitializer, i32 1)
+; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv8i32 = call <vscale x 8 x i32> @llvm.vector.splice.nxv8i32(<vscale x 8 x i32> zeroinitializer, <vscale x 8 x i32> zeroinitializer, i32 1)
+; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv2i64 = call <vscale x 2 x i64> @llvm.vector.splice.nxv2i64(<vscale x 2 x i64> zeroinitializer, <vscale x 2 x i64> zeroinitializer, i32 1)
+; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv4i64 = call <vscale x 4 x i64> @llvm.vector.splice.nxv4i64(<vscale x 4 x i64> zeroinitializer, <vscale x 4 x i64> zeroinitializer, i32 1)
+; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv2f16 = call <vscale x 2 x half> @llvm.vector.splice.nxv2f16(<vscale x 2 x half> zeroinitializer, <vscale x 2 x half> zeroinitializer, i32 1)
+; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv4f16 = call <vscale x 4 x half> @llvm.vector.splice.nxv4f16(<vscale x 4 x half> zeroinitializer, <vscale x 4 x half> zeroinitializer, i32 1)
+; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv8f16 = call <vscale x 8 x half> @llvm.vector.splice.nxv8f16(<vscale x 8 x half> zeroinitializer, <vscale x 8 x half> zeroinitializer, i32 1)
+; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv16f16 = call <vscale x 16 x half> @llvm.vector.splice.nxv16f16(<vscale x 16 x half> zeroinitializer, <vscale x 16 x half> zeroinitializer, i32 1)
+; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv2f32 = call <vscale x 2 x float> @llvm.vector.splice.nxv2f32(<vscale x 2 x float> zeroinitializer, <vscale x 2 x float> zeroinitializer, i32 1)
+; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv4f32 = call <vscale x 4 x float> @llvm.vector.splice.nxv4f32(<vscale x 4 x float> zeroinitializer, <vscale x 4 x float> zeroinitializer, i32 1)
+; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv8f32 = call <vscale x 8 x float> @llvm.vector.splice.nxv8f32(<vscale x 8 x float> zeroinitializer, <vscale x 8 x float> zeroinitializer, i32 1)
+; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv2f64 = call <vscale x 2 x double> @llvm.vector.splice.nxv2f64(<vscale x 2 x double> zeroinitializer, <vscale x 2 x double> zeroinitializer, i32 1)
+; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv4f64 = call <vscale x 4 x double> @llvm.vector.splice.nxv4f64(<vscale x 4 x double> zeroinitializer, <vscale x 4 x double> zeroinitializer, i32 1)
+; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv2bf16 = call <vscale x 2 x bfloat> @llvm.vector.splice.nxv2bf16(<vscale x 2 x bfloat> zeroinitializer, <vscale x 2 x bfloat> zeroinitializer, i32 1)
+; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv4bf16 = call <vscale x 4 x bfloat> @llvm.vector.splice.nxv4bf16(<vscale x 4 x bfloat> zeroinitializer, <vscale x 4 x bfloat> zeroinitializer, i32 1)
+; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv8bf16 = call <vscale x 8 x bfloat> @llvm.vector.splice.nxv8bf16(<vscale x 8 x bfloat> zeroinitializer, <vscale x 8 x bfloat> zeroinitializer, i32 1)
+; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv16bf16 = call <vscale x 16 x bfloat> @llvm.vector.splice.nxv16bf16(<vscale x 16 x bfloat> zeroinitializer, <vscale x 16 x bfloat> zeroinitializer, i32 1)
+; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv16i1 = call <vscale x 16 x i1> @llvm.vector.splice.nxv16i1(<vscale x 16 x i1> zeroinitializer, <vscale x 16 x i1> zeroinitializer, i32 1)
+; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv8i1 = call <vscale x 8 x i1> @llvm.vector.splice.nxv8i1(<vscale x 8 x i1> zeroinitializer, <vscale x 8 x i1> zeroinitializer, i32 1)
+; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv4i1 = call <vscale x 4 x i1> @llvm.vector.splice.nxv4i1(<vscale x 4 x i1> zeroinitializer, <vscale x 4 x i1> zeroinitializer, i32 1)
+; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv2i1 = call <vscale x 2 x i1> @llvm.vector.splice.nxv2i1(<vscale x 2 x i1> zeroinitializer, <vscale x 2 x i1> zeroinitializer, i32 1)
+; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv16i8_neg = call <vscale x 16 x i8> @llvm.vector.splice.nxv16i8(<vscale x 16 x i8> zeroinitializer, <vscale x 16 x i8> zeroinitializer, i32 -1)
+; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv32i8_neg = call <vscale x 32 x i8> @llvm.vector.splice.nxv32i8(<vscale x 32 x i8> zeroinitializer, <vscale x 32 x i8> zeroinitializer, i32 -1)
+; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv1i16_neg = call <vscale x 1 x i16> @llvm.vector.splice.nxv1i16(<vscale x 1 x i16> zeroinitializer, <vscale x 1 x i16> zeroinitializer, i32 -1)
+; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv2i16_neg = call <vscale x 2 x i16> @llvm.vector.splice.nxv2i16(<vscale x 2 x i16> zeroinitializer, <vscale x 2 x i16> zeroinitializer, i32 -1)
+; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv4i16_neg = call <vscale x 4 x i16> @llvm.vector.splice.nxv4i16(<vscale x 4 x i16> zeroinitializer, <vscale x 4 x i16> zeroinitializer, i32 -1)
+; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv8i16_neg = call <vscale x 8 x i16> @llvm.vector.splice.nxv8i16(<vscale x 8 x i16> zeroinitializer, <vscale x 8 x i16> zeroinitializer, i32 -1)
+; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv16i16_neg = call <vscale x 16 x i16> @llvm.vector.splice.nxv16i16(<vscale x 16 x i16> zeroinitializer, <vscale x 16 x i16> zeroinitializer, i32 -1)
+; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv4i32_neg = call <vscale x 4 x i32> @llvm.vector.splice.nxv4i32(<vscale x 4 x i32> zeroinitializer, <vscale x 4 x i32> zeroinitializer, i32 -1)
+; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv8i32_neg = call <vscale x 8 x i32> @llvm.vector.splice.nxv8i32(<vscale x 8 x i32> zeroinitializer, <vscale x 8 x i32> zeroinitializer, i32 -1)
+; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv1i64_neg = call <vscale x 1 x i64> @llvm.vector.splice.nxv1i64(<vscale x 1 x i64> zeroinitializer, <vscale x 1 x i64> zeroinitializer, i32 -1)
+; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv2i64_neg = call <vscale x 2 x i64> @llvm.vector.splice.nxv2i64(<vscale x 2 x i64> zeroinitializer, <vscale x 2 x i64> zeroinitializer, i32 -1)
+; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv4i64_neg = call <vscale x 4 x i64> @llvm.vector.splice.nxv4i64(<vscale x 4 x i64> zeroinitializer, <vscale x 4 x i64> zeroinitializer, i32 -1)
+; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv1f16_neg = call <vscale x 1 x half> @llvm.vector.splice.nxv1f16(<vscale x 1 x half> zeroinitializer, <vscale x 1 x half> zeroinitializer, i32 -1)
+; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv2f16_neg = call <vscale x 2 x half> @llvm.vector.splice.nxv2f16(<vscale x 2 x half> zeroinitializer, <vscale x 2 x half> zeroinitializer, i32 -1)
+; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv4f16_neg = call <vscale x 4 x half> @llvm.vector.splice.nxv4f16(<vscale x 4 x half> zeroinitializer, <vscale x 4 x half> zeroinitializer, i32 -1)
+; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv8f16_neg = call <vscale x 8 x half> @llvm.vector.splice.nxv8f16(<vscale x 8 x half> zeroinitializer, <vscale x 8 x half> zeroinitializer, i32 -1)
+; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv16f16_neg = call <vscale x 16 x half> @llvm.vector.splice.nxv16f16(<vscale x 16 x half> zeroinitializer, <vscale x 16 x half> zeroinitializer, i32 -1)
+; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv1f32_neg = call <vscale x 1 x float> @llvm.vector.splice.nxv1f32(<vscale x 1 x float> zeroinitializer, <vscale x 1 x float> zeroinitializer, i32 -1)
+; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv2f32_neg = call <vscale x 2 x float> @llvm.vector.splice.nxv2f32(<vscale x 2 x float> zeroinitializer, <vscale x 2 x float> zeroinitializer, i32 -1)
+; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv4f32_neg = call <vscale x 4 x float> @llvm.vector.splice.nxv4f32(<vscale x 4 x float> zeroinitializer, <vscale x 4 x float> zeroinitializer, i32 -1)
+; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv8f32_neg = call <vscale x 8 x float> @llvm.vector.splice.nxv8f32(<vscale x 8 x float> zeroinitializer, <vscale x 8 x float> zeroinitializer, i32 -1)
+; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv1f64_neg = call <vscale x 1 x double> @llvm.vector.splice.nxv1f64(<vscale x 1 x double> zeroinitializer, <vscale x 1 x double> zeroinitializer, i32 -1)
+; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv2f64_neg = call <vscale x 2 x double> @llvm.vector.splice.nxv2f64(<vscale x 2 x double> zeroinitializer, <vscale x 2 x double> zeroinitializer, i32 -1)
+; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv4f64_neg = call <vscale x 4 x double> @llvm.vector.splice.nxv4f64(<vscale x 4 x double> zeroinitializer, <vscale x 4 x double> zeroinitializer, i32 -1)
+; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv1bf16_neg = call <vscale x 1 x bfloat> @llvm.vector.splice.nxv1bf16(<vscale x 1 x bfloat> zeroinitializer, <vscale x 1 x bfloat> zeroinitializer, i32 -1)
+; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv2bf16_neg = call <vscale x 2 x bfloat> @llvm.vector.splice.nxv2bf16(<vscale x 2 x bfloat> zeroinitializer, <vscale x 2 x bfloat> zeroinitializer, i32 -1)
+; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv4bf16_neg = call <vscale x 4 x bfloat> @llvm.vector.splice.nxv4bf16(<vscale x 4 x bfloat> zeroinitializer, <vscale x 4 x bfloat> zeroinitializer, i32 -1)
+; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv8bf16_neg = call <vscale x 8 x bfloat> @llvm.vector.splice.nxv8bf16(<vscale x 8 x bfloat> zeroinitializer, <vscale x 8 x bfloat> zeroinitializer, i32 -1)
+; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv16bf16_neg = call <vscale x 16 x bfloat> @llvm.vector.splice.nxv16bf16(<vscale x 16 x bfloat> zeroinitializer, <vscale x 16 x bfloat> zeroinitializer, i32 -1)
+; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv16i1_neg = call <vscale x 16 x i1> @llvm.vector.splice.nxv16i1(<vscale x 16 x i1> zeroinitializer, <vscale x 16 x i1> zeroinitializer, i32 -1)
+; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv8i1_neg = call <vscale x 8 x i1> @llvm.vector.splice.nxv8i1(<vscale x 8 x i1> zeroinitializer, <vscale x 8 x i1> zeroinitializer, i32 -1)
+; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv4i1_neg = call <vscale x 4 x i1> @llvm.vector.splice.nxv4i1(<vscale x 4 x i1> zeroinitializer, <vscale x 4 x i1> zeroinitializer, i32 -1)
+; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv2i1_neg = call <vscale x 2 x i1> @llvm.vector.splice.nxv2i1(<vscale x 2 x i1> zeroinitializer, <vscale x 2 x i1> zeroinitializer, i32 -1)
+; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %splice_nxv1i1_neg = call <vscale x 1 x i1> @llvm.vector.splice.nxv1i1(<vscale x 1 x i1> zeroinitializer, <vscale x 1 x i1> zeroinitializer, i32 -1)
 ; TYPE_BASED_ONLY-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
 ;
 
-  %splice_nxv16i8 = call <vscale x 16 x i8> @llvm.experimental.vector.splice.nxv16i8(<vscale x 16 x i8> zeroinitializer, <vscale x 16 x i8> zeroinitializer, i32 1)
-  %splice_nxv32i8 = call <vscale x 32 x i8> @llvm.experimental.vector.splice.nxv32i8(<vscale x 32 x i8> zeroinitializer, <vscale x 32 x i8> zeroinitializer, i32 1)
-  %splice_nxv2i16 = call <vscale x 2 x i16> @llvm.experimental.vector.splice.nxv2i16(<vscale x 2 x i16> zeroinitializer, <vscale x 2 x i16> zeroinitializer, i32 1)
-  %splice_nxv4i16 = call <vscale x 4 x i16> @llvm.experimental.vector.splice.nxv4i16(<vscale x 4 x i16> zeroinitializer, <vscale x 4 x i16> zeroinitializer, i32 1)
-  %splice_nxv8i16 = call <vscale x 8 x i16> @llvm.experimental.vector.splice.nxv8i16(<vscale x 8 x i16> zeroinitializer, <vscale x 8 x i16> zeroinitializer, i32 1)
-  %splice_nxv16i16 = call <vscale x 16 x i16> @llvm.experimental.vector.splice.nxv16i16(<vscale x 16 x i16> zeroinitializer, <vscale x 16 x i16> zeroinitializer, i32 1)
-  %splice_nxv4i32 = call <vscale x 4 x i32> @llvm.experimental.vector.splice.nxv4i32(<vscale x 4 x i32> zeroinitializer, <vscale x 4 x i32> zeroinitializer, i32 1)
-  %splice_nxv8i32 = call <vscale x 8 x i32> @llvm.experimental.vector.splice.nxv8i32(<vscale x 8 x i32> zeroinitializer, <vscale x 8 x i32> zeroinitializer, i32 1)
-  %splice_nxv2i64 = call <vscale x 2 x i64> @llvm.experimental.vector.splice.nxv2i64(<vscale x 2 x i64> zeroinitializer, <vscale x 2 x i64> zeroinitializer, i32 1)
-  %splice_nxv4i64 = call <vscale x 4 x i64> @llvm.experimental.vector.splice.nxv4i64(<vscale x 4 x i64> zeroinitializer, <vscale x 4 x i64> zeroinitializer, i32 1)
-  %splice_nxv2f16 = call <vscale x 2 x half> @llvm.experimental.vector.splice.nxv2f16(<vscale x 2 x half> zeroinitializer, <vscale x 2 x half> zeroinitializer, i32 1)
-  %splice_nxv4f16 = call <vscale x 4 x half> @llvm.experimental.vector.splice.nxv4f16(<vscale x 4 x half> zeroinitializer, <vscale x 4 x half> zeroinitializer, i32 1)
-  %splice_nxv8f16 = call <vscale x 8 x half> @llvm.experimental.vector.splice.nxv8f16(<vscale x 8 x half> zeroinitializer, <vscale x 8 x half> zeroinitializer, i32 1)
-  %splice_nxv16f16 = call <vscale x 16 x half> @llvm.experimental.vector.splice.nxv16f16(<vscale x 16 x half> zeroinitializer, <vscale x 16 x half> zeroinitializer, i32 1)
-  %splice_nxv2f32 = call <vscale x 2 x float> @llvm.experimental.vector.splice.nxv2f32(<vscale x 2 x float> zeroinitializer, <vscale x 2 x float> zeroinitializer, i32 1)
-  %splice_nxv4f32 = call <vscale x 4 x float> @llvm.experimental.vector.splice.nxv4f32(<vscale x 4 x float> zeroinitializer, <vscale x 4 x float> zeroinitializer, i32 1)
-  %splice_nxv8f32 = call <vscale x 8 x float> @llvm.experimental.vector.splice.nxv8f32(<vscale x 8 x float> zeroinitializer, <vscale x 8 x float> zeroinitializer, i32 1)
-  %splice_nxv2f64 = call <vscale x 2 x double> @llvm.experimental.vector.splice.nxv2f64(<vscale x 2 x double> zeroinitializer, <vscale x 2 x double> zeroinitializer, i32 1)
-  %splice_nxv4f64 = call <vscale x 4 x double> @llvm.experimental.vector.splice.nxv4f64(<vscale x 4 x double> zeroinitializer, <vscale x 4 x double> zeroinitializer, i32 1)
-  %splice_nxv2bf16 = call <vscale x 2 x bfloat> @llvm.experimental.vector.splice.nxv2bf16(<vscale x 2 x bfloat> zeroinitializer, <vscale x 2 x bfloat> zeroinitializer, i32 1)
-  %splice_nxv4bf16 = call <vscale x 4 x bfloat> @llvm.experimental.vector.splice.nxv4bf16(<vscale x 4 x bfloat> zeroinitializer, <vscale x 4 x bfloat> zeroinitializer, i32 1)
-  %splice_nxv8bf16 = call <vscale x 8 x bfloat> @llvm.experimental.vector.splice.nxv8bf16(<vscale x 8 x bfloat> zeroinitializer, <vscale x 8 x bfloat> zeroinitializer, i32 1)
-  %splice_nxv16bf16 = call <vscale x 16 x bfloat> @llvm.experimental.vector.splice.nxv16bf16(<vscale x 16 x bfloat> zeroinitializer, <vscale x 16 x bfloat> zeroinitializer, i32 1)
-  %splice_nxv16i1 = call <vscale x 16 x i1> @llvm.experimental.vector.splice.nxv16i1(<vscale x 16 x i1> zeroinitializer, <vscale x 16 x i1> zeroinitializer, i32 1)
-  %splice_nxv8i1 =  call <vscale x 8 x i1> @llvm.experimental.vector.splice.nxv8i1(<vscale x 8 x i1> zeroinitializer, <vscale x 8 x i1> zeroinitializer, i32 1)
-  %splice_nxv4i1 = call <vscale x 4 x i1> @llvm.experimental.vector.splice.nxv4i1(<vscale x 4 x i1> zeroinitializer, <vscale x 4 x i1> zeroinitializer, i32 1)
-  %splice_nxv2i1 = call <vscale x 2 x i1> @llvm.experimental.vector.splice.nxv2i1(<vscale x 2 x i1> zeroinitializer, <vscale x 2 x i1> zeroinitializer, i32 1)
+  %splice_nxv16i8 = call <vscale x 16 x i8> @llvm.vector.splice.nxv16i8(<vscale x 16 x i8> zeroinitializer, <vscale x 16 x i8> zeroinitializer, i32 1)
+  %splice_nxv32i8 = call <vscale x 32 x i8> @llvm.vector.splice.nxv32i8(<vscale x 32 x i8> zeroinitializer, <vscale x 32 x i8> zeroinitializer, i32 1)
+  %splice_nxv2i16 = call <vscale x 2 x i16> @llvm.vector.splice.nxv2i16(<vscale x 2 x i16> zeroinitializer, <vscale x 2 x i16> zeroinitializer, i32 1)
+  %splice_nxv4i16 = call <vscale x 4 x i16> @llvm.vector.splice.nxv4i16(<vscale x 4 x i16> zeroinitializer, <vscale x 4 x i16> zeroinitializer, i32 1)
+  %splice_nxv8i16 = call <vscale x 8 x i16> @llvm.vector.splice.nxv8i16(<vscale x 8 x i16> zeroinitializer, <vscale x 8 x i16> zeroinitializer, i32 1)
+  %splice_nxv16i16 = call <vscale x 16 x i16> @llvm.vector.splice.nxv16i16(<vscale x 16 x i16> zeroinitializer, <vscale x 16 x i16> zeroinitializer, i32 1)
+  %splice_nxv4i32 = call <vscale x 4 x i32> @llvm.vector.splice.nxv4i32(<vscale x 4 x i32> zeroinitializer, <vscale x 4 x i32> zeroinitializer, i32 1)
+  %splice_nxv8i32 = call <vscale x 8 x i32> @llvm.vector.splice.nxv8i32(<vscale x 8 x i32> zeroinitializer, <vscale x 8 x i32> zeroinitializer, i32 1)
+  %splice_nxv2i64 = call <vscale x 2 x i64> @llvm.vector.splice.nxv2i64(<vscale x 2 x i64> zeroinitializer, <vscale x 2 x i64> zeroinitializer, i32 1)
+  %splice_nxv4i64 = call <vscale x 4 x i64> @llvm.vector.splice.nxv4i64(<vscale x 4 x i64> zeroinitializer, <vscale x 4 x i64> zeroinitializer, i32 1)
+  %splice_nxv2f16 = call <vscale x 2 x half> @llvm.vector.splice.nxv2f16(<vscale x 2 x half> zeroinitializer, <vscale x 2 x half> zeroinitializer, i32 1)
+  %splice_nxv4f16 = call <vscale x 4 x half> @llvm.vector.splice.nxv4f16(<vscale x 4 x half> zeroinitializer, <vscale x 4 x half> zeroinitializer, i32 1)
+  %splice_nxv8f16 = call <vscale x 8 x half> @llvm.vector.splice.nxv8f16(<vscale x 8 x half> zeroinitializer, <vscale x 8 x half> zeroinitializer, i32 1)
+  %splice_nxv16f16 = call <vscale x 16 x half> @llvm.vector.splice.nxv16f16(<vscale x 16 x half> zeroinitializer, <vscale x 16 x half> zeroinitializer, i32 1)
+  %splice_nxv2f32 = call <vscale x 2 x float> @llvm.vector.splice.nxv2f32(<vscale x 2 x float> zeroinitializer, <vscale x 2 x float> zeroinitializer, i32 1)
+  %splice_nxv4f32 = call <vscale x 4 x float> @llvm.vector.splice.nxv4f32(<vscale x 4 x float> zeroinitializer, <vscale x 4 x float> zeroinitializer, i32 1)
+  %splice_nxv8f32 = call <vscale x 8 x float> @llvm.vector.splice.nxv8f32(<vscale x 8 x float> zeroinitializer, <vscale x 8 x float> zeroinitializer, i32 1)
+  %splice_nxv2f64 = call <vscale x 2 x double> @llvm.vector.splice.nxv2f64(<vscale x 2 x double> zeroinitializer, <vscale x 2 x double> zeroinitializer, i32 1)
+  %splice_nxv4f64 = call <vscale x 4 x double> @llvm.vector.splice.nxv4f64(<vscale x 4 x double> zeroinitializer, <vscale x 4 x double> zeroinitializer, i32 1)
+  %splice_nxv2bf16 = call <vscale x 2 x bfloat> @llvm.vector.splice.nxv2bf16(<vscale x 2 x bfloat> zeroinitializer, <vscale x 2 x bfloat> zeroinitializer, i32 1)
+  %splice_nxv4bf16 = call <vscale x 4 x bfloat> @llvm.vector.splice.nxv4bf16(<vscale x 4 x bfloat> zeroinitializer, <vscale x 4 x bfloat> zeroinitializer, i32 1)
+  %splice_nxv8bf16 = call <vscale x 8 x bfloat> @llvm.vector.splice.nxv8bf16(<vscale x 8 x bfloat> zeroinitializer, <vscale x 8 x bfloat> zeroinitializer, i32 1)
+  %splice_nxv16bf16 = call <vscale x 16 x bfloat> @llvm.vector.splice.nxv16bf16(<vscale x 16 x bfloat> zeroinitializer, <vscale x 16 x bfloat> zeroinitializer, i32 1)
+  %splice_nxv16i1 = call <vscale x 16 x i1> @llvm.vector.splice.nxv16i1(<vscale x 16 x i1> zeroinitializer, <vscale x 16 x i1> zeroinitializer, i32 1)
+  %splice_nxv8i1 =  call <vscale x 8 x i1> @llvm.vector.splice.nxv8i1(<vscale x 8 x i1> zeroinitializer, <vscale x 8 x i1> zeroinitializer, i32 1)
+  %splice_nxv4i1 = call <vscale x 4 x i1> @llvm.vector.splice.nxv4i1(<vscale x 4 x i1> zeroinitializer, <vscale x 4 x i1> zeroinitializer, i32 1)
+  %splice_nxv2i1 = call <vscale x 2 x i1> @llvm.vector.splice.nxv2i1(<vscale x 2 x i1> zeroinitializer, <vscale x 2 x i1> zeroinitializer, i32 1)
 ;; negative Index
-  %splice_nxv16i8_neg = call <vscale x 16 x i8> @llvm.experimental.vector.splice.nxv16i8(<vscale x 16 x i8> zeroinitializer, <vscale x 16 x i8> zeroinitializer, i32 -1)
-  %splice_nxv32i8_neg = call <vscale x 32 x i8> @llvm.experimental.vector.splice.nxv32i8(<vscale x 32 x i8> zeroinitializer, <vscale x 32 x i8> zeroinitializer, i32 -1)
-  %splice_nxv1i16_neg = call <vscale x 1 x i16> @llvm.experimental.vector.splice.nxv1i16(<vscale x 1 x i16> zeroinitializer, <vscale x 1 x i16> zeroinitializer, i32 -1)
-  %splice_nxv2i16_neg = call <vscale x 2 x i16> @llvm.experimental.vector.splice.nxv2i16(<vscale x 2 x i16> zeroinitializer, <vscale x 2 x i16> zeroinitializer, i32 -1)
-  %splice_nxv4i16_neg = call <vscale x 4 x i16> @llvm.experimental.vector.splice.nxv4i16(<vscale x 4 x i16> zeroinitializer, <vscale x 4 x i16> zeroinitializer, i32 -1)
-  %splice_nxv8i16_neg = call <vscale x 8 x i16> @llvm.experimental.vector.splice.nxv8i16(<vscale x 8 x i16> zeroinitializer, <vscale x 8 x i16> zeroinitializer, i32 -1)
-  %splice_nxv16i16_neg = call <vscale x 16 x i16> @llvm.experimental.vector.splice.nxv16i16(<vscale x 16 x i16> zeroinitializer, <vscale x 16 x i16> zeroinitializer, i32 -1)
-  %splice_nxv4i32_neg = call <vscale x 4 x i32> @llvm.experimental.vector.splice.nxv4i32(<vscale x 4 x i32> zeroinitializer, <vscale x 4 x i32> zeroinitializer, i32 -1)
-  %splice_nxv8i32_neg = call <vscale x 8 x i32> @llvm.experimental.vector.splice.nxv8i32(<vscale x 8 x i32> zeroinitializer, <vscale x 8 x i32> zeroinitializer, i32 -1)
-  %splice_nxv1i64_neg= call <vscale x 1 x i64> @llvm.experimental.vector.splice.nxv1i64(<vscale x 1 x i64> zeroinitializer, <vscale x 1 x i64> zeroinitializer, i32 -1)
-  %splice_nxv2i64_neg= call <vscale x 2 x i64> @llvm.experimental.vector.splice.nxv2i64(<vscale x 2 x i64> zeroinitializer, <vscale x 2 x i64> zeroinitializer, i32 -1)
-  %splice_nxv4i64_neg = call <vscale x 4 x i64> @llvm.experimental.vector.splice.nxv4i64(<vscale x 4 x i64> zeroinitializer, <vscale x 4 x i64> zeroinitializer, i32 -1)
-  %splice_nxv1f16_neg = call <vscale x 1 x half> @llvm.experimental.vector.splice.nxv1f16(<vscale x 1 x half> zeroinitializer, <vscale x 1 x half> zeroinitializer, i32 -1)
-  %splice_nxv2f16_neg = call <vscale x 2 x half> @llvm.experimental.vector.splice.nxv2f16(<vscale x 2 x half> zeroinitializer, <vscale x 2 x half> zeroinitializer, i32 -1)
-  %splice_nxv4f16_neg = call <vscale x 4 x half> @llvm.experimental.vector.splice.nxv4f16(<vscale x 4 x half> zeroinitializer, <vscale x 4 x half> zeroinitializer, i32 -1)
-  %splice_nxv8f16_neg = call <vscale x 8 x half> @llvm.experimental.vector.splice.nxv8f16(<vscale x 8 x half> zeroinitializer, <vscale x 8 x half> zeroinitializer, i32 -1)
-  %splice_nxv16f16_neg = call <vscale x 16 x half> @llvm.experimental.vector.splice.nxv16f16(<vscale x 16 x half> zeroinitializer, <vscale x 16 x half> zeroinitializer, i32 -1)
-  %splice_nxv1f32_neg = call <vscale x 1 x float> @llvm.experimental.vector.splice.nxv1f32(<vscale x 1 x float> zeroinitializer, <vscale x 1 x float> zeroinitializer, i32 -1)
-  %splice_nxv2f32_neg = call <vscale x 2 x float> @llvm.experimental.vector.splice.nxv2f32(<vscale x 2 x float> zeroinitializer, <vscale x 2 x float> zeroinitializer, i32 -1)
-  %splice_nxv4f32_neg = call <vscale x 4 x float> @llvm.experimental.vector.splice.nxv4f32(<vscale x 4 x float> zeroinitializer, <vscale x 4 x float> zeroinitializer, i32 -1)
-  %splice_nxv8f32_neg = call <vscale x 8 x float> @llvm.experimental.vector.splice.nxv8f32(<vscale x 8 x float> zeroinitializer, <vscale x 8 x float> zeroinitializer, i32 -1)
-  %splice_nxv1f64_neg = call <vscale x 1 x double> @llvm.experimental.vector.splice.nxv1f64(<vscale x 1 x double> zeroinitializer, <vscale x 1 x double> zeroinitializer, i32 -1)
-  %splice_nxv2f64_neg = call <vscale x 2 x double> @llvm.experimental.vector.splice.nxv2f64(<vscale x 2 x double> zeroinitializer, <vscale x 2 x double> zeroinitializer, i32 -1)
-  %splice_nxv4f64_neg = call <vscale x 4 x double> @llvm.experimental.vector.splice.nxv4f64(<vscale x 4 x double> zeroinitializer, <vscale x 4 x double> zeroinitializer, i32 -1)
-  %splice_nxv1bf16_neg = call <vscale x 1 x bfloat> @llvm.experimental.vector.splice.nxv1bf16(<vscale x 1 x bfloat> zeroinitializer, <vscale x 1 x bfloat> zeroinitializer, i32 -1)
-  %splice_nxv2bf16_neg = call <vscale x 2 x bfloat> @llvm.experimental.vector.splice.nxv2bf16(<vscale x 2 x bfloat> zeroinitializer, <vscale x 2 x bfloat> zeroinitializer, i32 -1)
-  %splice_nxv4bf16_neg = call <vscale x 4 x bfloat> @llvm.experimental.vector.splice.nxv4bf16(<vscale x 4 x bfloat> zeroinitializer, <vscale x 4 x bfloat> zeroinitializer, i32 -1)
-  %splice_nxv8bf16_neg = call <vscale x 8 x bfloat> @llvm.experimental.vector.splice.nxv8bf16(<vscale x 8 x bfloat> zeroinitializer, <vscale x 8 x bfloat> zeroinitializer, i32 -1)
-  %splice_nxv16bf16_neg = call <vscale x 16 x bfloat> @llvm.experimental.vector.splice.nxv16bf16(<vscale x 16 x bfloat> zeroinitializer, <vscale x 16 x bfloat> zeroinitializer, i32 -1)
-  %splice_nxv16i1_neg = call <vscale x 16 x i1> @llvm.experimental.vector.splice.nxv16i1(<vscale x 16 x i1> zeroinitializer, <vscale x 16 x i1> zeroinitializer, i32 -1)
-  %splice_nxv8i1_neg =  call <vscale x 8 x i1> @llvm.experimental.vector.splice.nxv8i1(<vscale x 8 x i1> zeroinitializer, <vscale x 8 x i1> zeroinitializer, i32 -1)
-  %splice_nxv4i1_neg = call <vscale x 4 x i1> @llvm.experimental.vector.splice.nxv4i1(<vscale x 4 x i1> zeroinitializer, <vscale x 4 x i1> zeroinitializer, i32 -1)
-  %splice_nxv2i1_neg = call <vscale x 2 x i1> @llvm.experimental.vector.splice.nxv2i1(<vscale x 2 x i1> zeroinitializer, <vscale x 2 x i1> zeroinitializer, i32 -1)
-  %splice_nxv1i1_neg = call <vscale x 1 x i1> @llvm.experimental.vector.splice.nxv1i1(<vscale x 1 x i1> zeroinitializer, <vscale x 1 x i1> zeroinitializer, i32 -1)
+  %splice_nxv16i8_neg = call <vscale x 16 x i8> @llvm.vector.splice.nxv16i8(<vscale x 16 x i8> zeroinitializer, <vscale x 16 x i8> zeroinitializer, i32 -1)
+  %splice_nxv32i8_neg = call <vscale x 32 x i8> @llvm.vector.splice.nxv32i8(<vscale x 32 x i8> zeroinitializer, <vscale x 32 x i8> zeroinitializer, i32 -1)
+  %splice_nxv1i16_neg = call <vscale x 1 x i16> @llvm.vector.splice.nxv1i16(<vscale x 1 x i16> zeroinitializer, <vscale x 1 x i16> zeroinitializer, i32 -1)
+  %splice_nxv2i16_neg = call <vscale x 2 x i16> @llvm.vector.splice.nxv2i16(<vscale x 2 x i16> zeroinitializer, <vscale x 2 x i16> zeroinitializer, i32 -1)
+  %splice_nxv4i16_neg = call <vscale x 4 x i16> @llvm.vector.splice.nxv4i16(<vscale x 4 x i16> zeroinitializer, <vscale x 4 x i16> zeroinitializer, i32 -1)
+  %splice_nxv8i16_neg = call <vscale x 8 x i16> @llvm.vector.splice.nxv8i16(<vscale x 8 x i16> zeroinitializer, <vscale x 8 x i16> zeroinitializer, i32 -1)
+  %splice_nxv16i16_neg = call <vscale x 16 x i16> @llvm.vector.splice.nxv16i16(<vscale x 16 x i16> zeroinitializer, <vscale x 16 x i16> zeroinitializer, i32 -1)
+  %splice_nxv4i32_neg = call <vscale x 4 x i32> @llvm.vector.splice.nxv4i32(<vscale x 4 x i32> zeroinitializer, <vscale x 4 x i32> zeroinitializer, i32 -1)
+  %splice_nxv8i32_neg = call <vscale x 8 x i32> @llvm.vector.splice.nxv8i32(<vscale x 8 x i32> zeroinitializer, <vscale x 8 x i32> zeroinitializer, i32 -1)
+  %splice_nxv1i64_neg= call <vscale x 1 x i64> @llvm.vector.splice.nxv1i64(<vscale x 1 x i64> zeroinitializer, <vscale x 1 x i64> zeroinitializer, i32 -1)
+  %splice_nxv2i64_neg= call <vscale x 2 x i64> @llvm.vector.splice.nxv2i64(<vscale x 2 x i64> zeroinitializer, <vscale x 2 x i64> zeroinitializer, i32 -1)
+  %splice_nxv4i64_neg = call <vscale x 4 x i64> @llvm.vector.splice.nxv4i64(<vscale x 4 x i64> zeroinitializer, <vscale x 4 x i64> zeroinitializer, i32 -1)
+  %splice_nxv1f16_neg = call <vscale x 1 x half> @llvm.vector.splice.nxv1f16(<vscale x 1 x half> zeroinitializer, <vscale x 1 x half> zeroinitializer, i32 -1)
+  %splice_nxv2f16_neg = call <vscale x 2 x half> @llvm.vector.splice.nxv2f16(<vscale x 2 x half> zeroinitializer, <vscale x 2 x half> zeroinitializer, i32 -1)
+  %splice_nxv4f16_neg = call <vscale x 4 x half> @llvm.vector.splice.nxv4f16(<vscale x 4 x half> zeroinitializer, <vscale x 4 x half> zeroinitializer, i32 -1)
+  %splice_nxv8f16_neg = call <vscale x 8 x half> @llvm.vector.splice.nxv8f16(<vscale x 8 x half> zeroinitializer, <vscale x 8 x half> zeroinitializer, i32 -1)
+  %splice_nxv16f16_neg = call <vscale x 16 x half> @llvm.vector.splice.nxv16f16(<vscale x 16 x half> zeroinitializer, <vscale x 16 x half> zeroinitializer, i32 -1)
+  %splice_nxv1f32_neg = call <vscale x 1 x float> @llvm.vector.splice.nxv1f32(<vscale x 1 x float> zeroinitializer, <vscale x 1 x float> zeroinitializer, i32 -1)
+  %splice_nxv2f32_neg = call <vscale x 2 x float> @llvm.vector.splice.nxv2f32(<vscale x 2 x float> zeroinitializer, <vscale x 2 x float> zeroinitializer, i32 -1)
+  %splice_nxv4f32_neg = call <vscale x 4 x float> @llvm.vector.splice.nxv4f32(<vscale x 4 x float> zeroinitializer, <vscale x 4 x float> zeroinitializer, i32 -1)
+  %splice_nxv8f32_neg = call <vscale x 8 x float> @llvm.vector.splice.nxv8f32(<vscale x 8 x float> zeroinitializer, <vscale x 8 x float> zeroinitializer, i32 -1)
+  %splice_nxv1f64_neg = call <vscale x 1 x double> @llvm.vector.splice.nxv1f64(<vscale x 1 x double> zeroinitializer, <vscale x 1 x double> zeroinitializer, i32 -1)
+  %splice_nxv2f64_neg = call <vscale x 2 x double> @llvm.vector.splice.nxv2f64(<vscale x 2 x double> zeroinitializer, <vscale x 2 x double> zeroinitializer, i32 -1)
+  %splice_nxv4f64_neg = call <vscale x 4 x double> @llvm.vector.splice.nxv4f64(<vscale x 4 x double> zeroinitializer, <vscale x 4 x double> zeroinitializer, i32 -1)
+  %splice_nxv1bf16_neg = call <vscale x 1 x bfloat> @llvm.vector.splice.nxv1bf16(<vscale x 1 x bfloat> zeroinitializer, <vscale x 1 x bfloat> zeroinitializer, i32 -1)
+  %splice_nxv2bf16_neg = call <vscale x 2 x bfloat> @llvm.vector.splice.nxv2bf16(<vscale x 2 x bfloat> zeroinitializer, <vscale x 2 x bfloat> zeroinitializer, i32 -1)
+  %splice_nxv4bf16_neg = call <vscale x 4 x bfloat> @llvm.vector.splice.nxv4bf16(<vscale x 4 x bfloat> zeroinitializer, <vscale x 4 x bfloat> zeroinitializer, i32 -1)
+  %splice_nxv8bf16_neg = call <vscale x 8 x bfloat> @llvm.vector.splice.nxv8bf16(<vscale x 8 x bfloat> zeroinitializer, <vscale x 8 x bfloat> zeroinitializer, i32 -1)
+  %splice_nxv16bf16_neg = call <vscale x 16 x bfloat> @llvm.vector.splice.nxv16bf16(<vscale x 16 x bfloat> zeroinitializer, <vscale x 16 x bfloat> zeroinitializer, i32 -1)
+  %splice_nxv16i1_neg = call <vscale x 16 x i1> @llvm.vector.splice.nxv16i1(<vscale x 16 x i1> zeroinitializer, <vscale x 16 x i1> zeroinitializer, i32 -1)
+  %splice_nxv8i1_neg =  call <vscale x 8 x i1> @llvm.vector.splice.nxv8i1(<vscale x 8 x i1> zeroinitializer, <vscale x 8 x i1> zeroinitializer, i32 -1)
+  %splice_nxv4i1_neg = call <vscale x 4 x i1> @llvm.vector.splice.nxv4i1(<vscale x 4 x i1> zeroinitializer, <vscale x 4 x i1> zeroinitializer, i32 -1)
+  %splice_nxv2i1_neg = call <vscale x 2 x i1> @llvm.vector.splice.nxv2i1(<vscale x 2 x i1> zeroinitializer, <vscale x 2 x i1> zeroinitializer, i32 -1)
+  %splice_nxv1i1_neg = call <vscale x 1 x i1> @llvm.vector.splice.nxv1i1(<vscale x 1 x i1> zeroinitializer, <vscale x 1 x i1> zeroinitializer, i32 -1)
   ret void
 }
 
-declare <vscale x 1 x i1> @llvm.experimental.vector.splice.nxv1i1(<vscale x 1 x i1>, <vscale x 1 x i1>, i32)
-declare <vscale x 2 x i1> @llvm.experimental.vector.splice.nxv2i1(<vscale x 2 x i1>, <vscale x 2 x i1>, i32)
-declare <vscale x 4 x i1> @llvm.experimental.vector.splice.nxv4i1(<vscale x 4 x i1>, <vscale x 4 x i1>, i32)
-declare <vscale x 8 x i1> @llvm.experimental.vector.splice.nxv8i1(<vscale x 8 x i1>, <vscale x 8 x i1>, i32)
-declare <vscale x 16 x i1> @llvm.experimental.vector.splice.nxv16i1(<vscale x 16 x i1>, <vscale x 16 x i1>, i32)
-declare <vscale x 1 x i8> @llvm.experimental.vector.splice.nxv1i8(<vscale x 1 x i8>, <vscale x 1 x i8>, i32)
-declare <vscale x 2 x i8> @llvm.experimental.vector.splice.nxv2i8(<vscale x 2 x i8>, <vscale x 2 x i8>, i32)
-declare <vscale x 16 x i8> @llvm.experimental.vector.splice.nxv16i8(<vscale x 16 x i8>, <vscale x 16 x i8>, i32)
-declare <vscale x 32 x i8> @llvm.experimental.vector.splice.nxv32i8(<vscale x 32 x i8>, <vscale x 32 x i8>, i32)
-declare <vscale x 1 x i16> @llvm.experimental.vector.splice.nxv1i16(<vscale x 1 x i16>, <vscale x 1 x i16>, i32)
-declare <vscale x 2 x i16> @llvm.experimental.vector.splice.nxv2i16(<vscale x 2 x i16>, <vscale x 2 x i16>, i32)
-declare <vscale x 4 x i16> @llvm.experimental.vector.splice.nxv4i16(<vscale x 4 x i16>, <vscale x 4 x i16>, i32)
-declare <vscale x 8 x i16> @llvm.experimental.vector.splice.nxv8i16(<vscale x 8 x i16>, <vscale x 8 x i16>, i32)
-declare <vscale x 16 x i16> @llvm.experimental.vector.splice.nxv16i16(<vscale x 16 x i16>, <vscale x 16 x i16>, i32)
-declare <vscale x 4 x i32> @llvm.experimental.vector.splice.nxv4i32(<vscale x 4 x i32>, <vscale x 4 x i32>, i32)
-declare <vscale x 8 x i32> @llvm.experimental.vector.splice.nxv8i32(<vscale x 8 x i32>, <vscale x 8 x i32>, i32)
-declare <vscale x 1 x i64> @llvm.experimental.vector.splice.nxv1i64(<vscale x 1 x i64>, <vscale x 1 x i64>, i32)
-declare <vscale x 2 x i64> @llvm.experimental.vector.splice.nxv2i64(<vscale x 2 x i64>, <vscale x 2 x i64>, i32)
-declare <vscale x 4 x i64> @llvm.experimental.vector.splice.nxv4i64(<vscale x 4 x i64>, <vscale x 4 x i64>, i32)
-declare <vscale x 1 x half> @llvm.experimental.vector.splice.nxv1f16(<vscale x 1 x half>, <vscale x 1 x half>, i32)
-declare <vscale x 2 x half> @llvm.experimental.vector.splice.nxv2f16(<vscale x 2 x half>, <vscale x 2 x half>, i32)
-declare <vscale x 4 x half> @llvm.experimental.vector.splice.nxv4f16(<vscale x 4 x half>, <vscale x 4 x half>, i32)
-declare <vscale x 8 x half> @llvm.experimental.vector.splice.nxv8f16(<vscale x 8 x half>, <vscale x 8 x half>, i32)
-declare <vscale x 16 x half> @llvm.experimental.vector.splice.nxv16f16(<vscale x 16 x half>, <vscale x 16 x half>, i32)
-declare <vscale x 1 x bfloat> @llvm.experimental.vector.splice.nxv1bf16(<vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i32)
-declare <vscale x 2 x bfloat> @llvm.experimental.vector.splice.nxv2bf16(<vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i32)
-declare <vscale x 4 x bfloat> @llvm.experimental.vector.splice.nxv4bf16(<vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i32)
-declare <vscale x 8 x bfloat> @llvm.experimental.vector.splice.nxv8bf16(<vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i32)
-declare <vscale x 16 x bfloat> @llvm.experimental.vector.splice.nxv16bf16(<vscale x 16 x bfloat>, <vscale x 16 x bfloat>, i32)
-declare <vscale x 1 x float> @llvm.experimental.vector.splice.nxv1f32(<vscale x 1 x float>, <vscale x 1 x float>, i32)
-declare <vscale x 2 x float> @llvm.experimental.vector.splice.nxv2f32(<vscale x 2 x float>, <vscale x 2 x float>, i32)
-declare <vscale x 4 x float> @llvm.experimental.vector.splice.nxv4f32(<vscale x 4 x float>, <vscale x 4 x float>, i32)
-declare <vscale x 8 x float> @llvm.experimental.vector.splice.nxv8f32(<vscale x 8 x float>, <vscale x 8 x float>, i32)
-declare <vscale x 16 x float> @llvm.experimental.vector.splice.nxv16f32(<vscale x 16 x float>, <vscale x 16 x float>, i32)
-declare <vscale x 1 x double> @llvm.experimental.vector.splice.nxv1f64(<vscale x 1 x double>, <vscale x 1 x double>, i32)
-declare <vscale x 2 x double> @llvm.experimental.vector.splice.nxv2f64(<vscale x 2 x double>, <vscale x 2 x double>, i32)
-declare <vscale x 4 x double> @llvm.experimental.vector.splice.nxv4f64(<vscale x 4 x double>, <vscale x 4 x double>, i32)
+declare <vscale x 1 x i1> @llvm.vector.splice.nxv1i1(<vscale x 1 x i1>, <vscale x 1 x i1>, i32)
+declare <vscale x 2 x i1> @llvm.vector.splice.nxv2i1(<vscale x 2 x i1>, <vscale x 2 x i1>, i32)
+declare <vscale x 4 x i1> @llvm.vector.splice.nxv4i1(<vscale x 4 x i1>, <vscale x 4 x i1>, i32)
+declare <vscale x 8 x i1> @llvm.vector.splice.nxv8i1(<vscale x 8 x i1>, <vscale x 8 x i1>, i32)
+declare <vscale x 16 x i1> @llvm.vector.splice.nxv16i1(<vscale x 16 x i1>, <vscale x 16 x i1>, i32)
+declare <vscale x 1 x i8> @llvm.vector.splice.nxv1i8(<vscale x 1 x i8>, <vscale x 1 x i8>, i32)
+declare <vscale x 2 x i8> @llvm.vector.splice.nxv2i8(<vscale x 2 x i8>, <vscale x 2 x i8>, i32)
+declare <vscale x 16 x i8> @llvm.vector.splice.nxv16i8(<vscale x 16 x i8>, <vscale x 16 x i8>, i32)
+declare <vscale x 32 x i8> @llvm.vector.splice.nxv32i8(<vscale x 32 x i8>, <vscale x 32 x i8>, i32)
+declare <vscale x 1 x i16> @llvm.vector.splice.nxv1i16(<vscale x 1 x i16>, <vscale x 1 x i16>, i32)
+declare <vscale x 2 x i16> @llvm.vector.splice.nxv2i16(<vscale x 2 x i16>, <vscale x 2 x i16>, i32)
+declare <vscale x 4 x i16> @llvm.vector.splice.nxv4i16(<vscale x 4 x i16>, <vscale x 4 x i16>, i32)
+declare <vscale x 8 x i16> @llvm.vector.splice.nxv8i16(<vscale x 8 x i16>, <vscale x 8 x i16>, i32)
+declare <vscale x 16 x i16> @llvm.vector.splice.nxv16i16(<vscale x 16 x i16>, <vscale x 16 x i16>, i32)
+declare <vscale x 4 x i32> @llvm.vector.splice.nxv4i32(<vscale x 4 x i32>, <vscale x 4 x i32>, i32)
+declare <vscale x 8 x i32> @llvm.vector.splice.nxv8i32(<vscale x 8 x i32>, <vscale x 8 x i32>, i32)
+declare <vscale x 1 x i64> @llvm.vector.splice.nxv1i64(<vscale x 1 x i64>, <vscale x 1 x i64>, i32)
+declare <vscale x 2 x i64> @llvm.vector.splice.nxv2i64(<vscale x 2 x i64>, <vscale x 2 x i64>, i32)
+declare <vscale x 4 x i64> @llvm.vector.splice.nxv4i64(<vscale x 4 x i64>, <vscale x 4 x i64>, i32)
+declare <vscale x 1 x half> @llvm.vector.splice.nxv1f16(<vscale x 1 x half>, <vscale x 1 x half>, i32)
+declare <vscale x 2 x half> @llvm.vector.splice.nxv2f16(<vscale x 2 x half>, <vscale x 2 x half>, i32)
+declare <vscale x 4 x half> @llvm.vector.splice.nxv4f16(<vscale x 4 x half>, <vscale x 4 x half>, i32)
+declare <vscale x 8 x half> @llvm.vector.splice.nxv8f16(<vscale x 8 x half>, <vscale x 8 x half>, i32)
+declare <vscale x 16 x half> @llvm.vector.splice.nxv16f16(<vscale x 16 x half>, <vscale x 16 x half>, i32)
+declare <vscale x 1 x bfloat> @llvm.vector.splice.nxv1bf16(<vscale x 1 x bfloat>, <vscale x 1 x bfloat>, i32)
+declare <vscale x 2 x bfloat> @llvm.vector.splice.nxv2bf16(<vscale x 2 x bfloat>, <vscale x 2 x bfloat>, i32)
+declare <vscale x 4 x bfloat> @llvm.vector.splice.nxv4bf16(<vscale x 4 x bfloat>, <vscale x 4 x bfloat>, i32)
+declare <vscale x 8 x bfloat> @llvm.vector.splice.nxv8bf16(<vscale x 8 x bfloat>, <vscale x 8 x bfloat>, i32)
+declare <vscale x 16 x bfloat> @llvm.vector.splice.nxv16bf16(<vscale x 16 x bfloat>, <vscale x 16 x bfloat>, i32)
+declare <vscale x 1 x float> @llvm.vector.splice.nxv1f32(<vscale x 1 x float>, <vscale x 1 x float>, i32)
+declare <vscale x 2 x float> @llvm.vector.splice.nxv2f32(<vscale x 2 x float>, <vscale x 2 x float>, i32)
+declare <vscale x 4 x float> @llvm.vector.splice.nxv4f32(<vscale x 4 x float>, <vscale x 4 x float>, i32)
+declare <vscale x 8 x float> @llvm.vector.splice.nxv8f32(<vscale x 8 x float>, <vscale x 8 x float>, i32)
+declare <vscale x 16 x float> @llvm.vector.splice.nxv16f32(<vscale x 16 x float>, <vscale x 16 x float>, i32)
+declare <vscale x 1 x double> @llvm.vector.splice.nxv1f64(<vscale x 1 x double>, <vscale x 1 x double>, i32)
+declare <vscale x 2 x double> @llvm.vector.splice.nxv2f64(<vscale x 2 x double>, <vscale x 2 x double>, i32)
+declare <vscale x 4 x double> @llvm.vector.splice.nxv4f64(<vscale x 4 x double>, <vscale x 4 x double>, i32)
 
 define void @get_lane_mask() #0 {
 ; CHECK-LABEL: 'get_lane_mask'
@@ -693,14 +693,14 @@ define void @get_lane_mask() #0 {
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %mask_nxv2i1_i32 = call <vscale x 2 x i1> @llvm.get.active.lane.mask.nxv2i1.i32(i32 undef, i32 undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %mask_nxv32i1_i64 = call <vscale x 32 x i1> @llvm.get.active.lane.mask.nxv32i1.i64(i64 undef, i64 undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %mask_nxv16i1_i16 = call <vscale x 16 x i1> @llvm.get.active.lane.mask.nxv16i1.i16(i16 undef, i16 undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %mask_v16i1_i64 = call <16 x i1> @llvm.get.active.lane.mask.v16i1.i64(i64 undef, i64 undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %mask_v8i1_i64 = call <8 x i1> @llvm.get.active.lane.mask.v8i1.i64(i64 undef, i64 undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %mask_v4i1_i64 = call <4 x i1> @llvm.get.active.lane.mask.v4i1.i64(i64 undef, i64 undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %mask_v2i1_i64 = call <2 x i1> @llvm.get.active.lane.mask.v2i1.i64(i64 undef, i64 undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %mask_v16i1_i32 = call <16 x i1> @llvm.get.active.lane.mask.v16i1.i32(i32 undef, i32 undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %mask_v8i1_i32 = call <8 x i1> @llvm.get.active.lane.mask.v8i1.i32(i32 undef, i32 undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %mask_v4i1_i32 = call <4 x i1> @llvm.get.active.lane.mask.v4i1.i32(i32 undef, i32 undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %mask_v2i1_i32 = call <2 x i1> @llvm.get.active.lane.mask.v2i1.i32(i32 undef, i32 undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %mask_v16i1_i64 = call <16 x i1> @llvm.get.active.lane.mask.v16i1.i64(i64 undef, i64 undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %mask_v8i1_i64 = call <8 x i1> @llvm.get.active.lane.mask.v8i1.i64(i64 undef, i64 undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %mask_v4i1_i64 = call <4 x i1> @llvm.get.active.lane.mask.v4i1.i64(i64 undef, i64 undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %mask_v2i1_i64 = call <2 x i1> @llvm.get.active.lane.mask.v2i1.i64(i64 undef, i64 undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %mask_v16i1_i32 = call <16 x i1> @llvm.get.active.lane.mask.v16i1.i32(i32 undef, i32 undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %mask_v8i1_i32 = call <8 x i1> @llvm.get.active.lane.mask.v8i1.i32(i32 undef, i32 undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %mask_v4i1_i32 = call <4 x i1> @llvm.get.active.lane.mask.v4i1.i32(i32 undef, i32 undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %mask_v2i1_i32 = call <2 x i1> @llvm.get.active.lane.mask.v2i1.i32(i32 undef, i32 undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 144 for instruction: %mask_v32i1_i64 = call <32 x i1> @llvm.get.active.lane.mask.v32i1.i64(i64 undef, i64 undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 22 for instruction: %mask_v16i1_i16 = call <16 x i1> @llvm.get.active.lane.mask.v16i1.i16(i16 undef, i16 undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
@@ -716,14 +716,14 @@ define void @get_lane_mask() #0 {
 ; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %mask_nxv2i1_i32 = call <vscale x 2 x i1> @llvm.get.active.lane.mask.nxv2i1.i32(i32 undef, i32 undef)
 ; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %mask_nxv32i1_i64 = call <vscale x 32 x i1> @llvm.get.active.lane.mask.nxv32i1.i64(i64 undef, i64 undef)
 ; TYPE_BASED_ONLY-NEXT:  Cost Model: Invalid cost for instruction: %mask_nxv16i1_i16 = call <vscale x 16 x i1> @llvm.get.active.lane.mask.nxv16i1.i16(i16 undef, i16 undef)
-; TYPE_BASED_ONLY-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %mask_v16i1_i64 = call <16 x i1> @llvm.get.active.lane.mask.v16i1.i64(i64 undef, i64 undef)
-; TYPE_BASED_ONLY-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %mask_v8i1_i64 = call <8 x i1> @llvm.get.active.lane.mask.v8i1.i64(i64 undef, i64 undef)
-; TYPE_BASED_ONLY-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %mask_v4i1_i64 = call <4 x i1> @llvm.get.active.lane.mask.v4i1.i64(i64 undef, i64 undef)
-; TYPE_BASED_ONLY-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %mask_v2i1_i64 = call <2 x i1> @llvm.get.active.lane.mask.v2i1.i64(i64 undef, i64 undef)
-; TYPE_BASED_ONLY-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %mask_v16i1_i32 = call <16 x i1> @llvm.get.active.lane.mask.v16i1.i32(i32 undef, i32 undef)
-; TYPE_BASED_ONLY-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %mask_v8i1_i32 = call <8 x i1> @llvm.get.active.lane.mask.v8i1.i32(i32 undef, i32 undef)
-; TYPE_BASED_ONLY-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %mask_v4i1_i32 = call <4 x i1> @llvm.get.active.lane.mask.v4i1.i32(i32 undef, i32 undef)
-; TYPE_BASED_ONLY-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %mask_v2i1_i32 = call <2 x i1> @llvm.get.active.lane.mask.v2i1.i32(i32 undef, i32 undef)
+; TYPE_BASED_ONLY-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %mask_v16i1_i64 = call <16 x i1> @llvm.get.active.lane.mask.v16i1.i64(i64 undef, i64 undef)
+; TYPE_BASED_ONLY-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %mask_v8i1_i64 = call <8 x i1> @llvm.get.active.lane.mask.v8i1.i64(i64 undef, i64 undef)
+; TYPE_BASED_ONLY-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %mask_v4i1_i64 = call <4 x i1> @llvm.get.active.lane.mask.v4i1.i64(i64 undef, i64 undef)
+; TYPE_BASED_ONLY-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %mask_v2i1_i64 = call <2 x i1> @llvm.get.active.lane.mask.v2i1.i64(i64 undef, i64 undef)
+; TYPE_BASED_ONLY-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %mask_v16i1_i32 = call <16 x i1> @llvm.get.active.lane.mask.v16i1.i32(i32 undef, i32 undef)
+; TYPE_BASED_ONLY-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %mask_v8i1_i32 = call <8 x i1> @llvm.get.active.lane.mask.v8i1.i32(i32 undef, i32 undef)
+; TYPE_BASED_ONLY-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %mask_v4i1_i32 = call <4 x i1> @llvm.get.active.lane.mask.v4i1.i32(i32 undef, i32 undef)
+; TYPE_BASED_ONLY-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %mask_v2i1_i32 = call <2 x i1> @llvm.get.active.lane.mask.v2i1.i32(i32 undef, i32 undef)
 ; TYPE_BASED_ONLY-NEXT:  Cost Model: Found an estimated cost of 96 for instruction: %mask_v32i1_i64 = call <32 x i1> @llvm.get.active.lane.mask.v32i1.i64(i64 undef, i64 undef)
 ; TYPE_BASED_ONLY-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %mask_v16i1_i16 = call <16 x i1> @llvm.get.active.lane.mask.v16i1.i16(i16 undef, i16 undef)
 ; TYPE_BASED_ONLY-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
diff --git a/llvm/test/Analysis/CostModel/RISCV/arith-fp.ll b/llvm/test/Analysis/CostModel/RISCV/arith-fp.ll
index 306277e46fa5..1dde88f366a3 100644
--- a/llvm/test/Analysis/CostModel/RISCV/arith-fp.ll
+++ b/llvm/test/Analysis/CostModel/RISCV/arith-fp.ll
@@ -1,5 +1,5 @@
 ; NOTE: Assertions have been autogenerated by utils/update_analyze_test_checks.py
-; RUN: opt -passes="print<cost-model>" 2>&1 -disable-output -mtriple=riscv64 -mattr=+v,+f,+d,+zfh,+zvfh -riscv-v-vector-bits-min=128 -riscv-v-fixed-length-vector-lmul-max=1 < %s | FileCheck %s
+; RUN: opt -passes="print<cost-model>" 2>&1 -disable-output -mtriple=riscv64 -mattr=+v,+f,+d,+zfh,+zvfh -riscv-v-vector-bits-min=128 < %s | FileCheck %s
 ; Check that we don't crash querying costs when vectors are not enabled.
 ; RUN: opt -passes="print<cost-model>" 2>&1 -disable-output -mtriple=riscv64
 
@@ -252,8 +252,8 @@ define i32 @fdiv() {
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V2F16 = fdiv <2 x half> undef, undef
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V4F16 = fdiv <4 x half> undef, undef
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V8F16 = fdiv <8 x half> undef, undef
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16F16 = fdiv <16 x half> undef, undef
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V32F16 = fdiv <32 x half> undef, undef
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V16F16 = fdiv <16 x half> undef, undef
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V32F16 = fdiv <32 x half> undef, undef
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %NXV1F16 = fdiv <vscale x 1 x half> undef, undef
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %NXV2F16 = fdiv <vscale x 2 x half> undef, undef
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %NXV4F16 = fdiv <vscale x 4 x half> undef, undef
@@ -263,8 +263,8 @@ define i32 @fdiv() {
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V1F32 = fdiv <1 x float> undef, undef
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V2F32 = fdiv <2 x float> undef, undef
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V4F32 = fdiv <4 x float> undef, undef
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V8F32 = fdiv <8 x float> undef, undef
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V16F32 = fdiv <16 x float> undef, undef
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V8F32 = fdiv <8 x float> undef, undef
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V16F32 = fdiv <16 x float> undef, undef
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %NXV1F32 = fdiv <vscale x 1 x float> undef, undef
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %NXV2F32 = fdiv <vscale x 2 x float> undef, undef
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %NXV4F32 = fdiv <vscale x 4 x float> undef, undef
@@ -272,8 +272,8 @@ define i32 @fdiv() {
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %NXV16F32 = fdiv <vscale x 16 x float> undef, undef
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V1F64 = fdiv <1 x double> undef, undef
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V2F64 = fdiv <2 x double> undef, undef
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V4F64 = fdiv <4 x double> undef, undef
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %V8F64 = fdiv <8 x double> undef, undef
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V4F64 = fdiv <4 x double> undef, undef
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V8F64 = fdiv <8 x double> undef, undef
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %NXV1F64 = fdiv <vscale x 1 x double> undef, undef
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %NXV2F64 = fdiv <vscale x 2 x double> undef, undef
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %NXV4F64 = fdiv <vscale x 4 x double> undef, undef
@@ -332,8 +332,8 @@ define i32 @frem() {
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V2F16 = frem <2 x half> undef, undef
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %V4F16 = frem <4 x half> undef, undef
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 31 for instruction: %V8F16 = frem <8 x half> undef, undef
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %V16F16 = frem <16 x half> undef, undef
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 96 for instruction: %V32F16 = frem <32 x half> undef, undef
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 63 for instruction: %V16F16 = frem <16 x half> undef, undef
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 127 for instruction: %V32F16 = frem <32 x half> undef, undef
 ; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %NXV1F16 = frem <vscale x 1 x half> undef, undef
 ; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %NXV2F16 = frem <vscale x 2 x half> undef, undef
 ; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %NXV4F16 = frem <vscale x 4 x half> undef, undef
@@ -343,8 +343,8 @@ define i32 @frem() {
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V1F32 = frem <1 x float> undef, undef
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V2F32 = frem <2 x float> undef, undef
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %V4F32 = frem <4 x float> undef, undef
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V8F32 = frem <8 x float> undef, undef
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 48 for instruction: %V16F32 = frem <16 x float> undef, undef
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 31 for instruction: %V8F32 = frem <8 x float> undef, undef
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 63 for instruction: %V16F32 = frem <16 x float> undef, undef
 ; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %NXV1F32 = frem <vscale x 1 x float> undef, undef
 ; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %NXV2F32 = frem <vscale x 2 x float> undef, undef
 ; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %NXV4F32 = frem <vscale x 4 x float> undef, undef
@@ -352,8 +352,8 @@ define i32 @frem() {
 ; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %NXV16F32 = frem <vscale x 16 x float> undef, undef
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %V1F64 = frem <1 x double> undef, undef
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %V2F64 = frem <2 x double> undef, undef
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %V4F64 = frem <4 x double> undef, undef
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %V8F64 = frem <8 x double> undef, undef
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 15 for instruction: %V4F64 = frem <4 x double> undef, undef
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 31 for instruction: %V8F64 = frem <8 x double> undef, undef
 ; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %NXV1F64 = frem <vscale x 1 x double> undef, undef
 ; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %NXV2F64 = frem <vscale x 2 x double> undef, undef
 ; CHECK-NEXT:  Cost Model: Invalid cost for instruction: %NXV4F64 = frem <vscale x 4 x double> undef, undef
@@ -492,8 +492,8 @@ define i32 @fcopysign() {
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2F16 = call <2 x half> @llvm.copysign.v2f16(<2 x half> undef, <2 x half> undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4F16 = call <4 x half> @llvm.copysign.v4f16(<4 x half> undef, <4 x half> undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8F16 = call <8 x half> @llvm.copysign.v8f16(<8 x half> undef, <8 x half> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V16F16 = call <16 x half> @llvm.copysign.v16f16(<16 x half> undef, <16 x half> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V32F16 = call <32 x half> @llvm.copysign.v32f16(<32 x half> undef, <32 x half> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16F16 = call <16 x half> @llvm.copysign.v16f16(<16 x half> undef, <16 x half> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32F16 = call <32 x half> @llvm.copysign.v32f16(<32 x half> undef, <32 x half> undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %NXV1F16 = call <vscale x 1 x half> @llvm.copysign.nxv1f16(<vscale x 1 x half> undef, <vscale x 1 x half> undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %NXV2F16 = call <vscale x 2 x half> @llvm.copysign.nxv2f16(<vscale x 2 x half> undef, <vscale x 2 x half> undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %NXV4F16 = call <vscale x 4 x half> @llvm.copysign.nxv4f16(<vscale x 4 x half> undef, <vscale x 4 x half> undef)
@@ -503,8 +503,8 @@ define i32 @fcopysign() {
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V1F32 = call <1 x float> @llvm.copysign.v1f32(<1 x float> undef, <1 x float> undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2F32 = call <2 x float> @llvm.copysign.v2f32(<2 x float> undef, <2 x float> undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4F32 = call <4 x float> @llvm.copysign.v4f32(<4 x float> undef, <4 x float> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V8F32 = call <8 x float> @llvm.copysign.v8f32(<8 x float> undef, <8 x float> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16F32 = call <16 x float> @llvm.copysign.v16f32(<16 x float> undef, <16 x float> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8F32 = call <8 x float> @llvm.copysign.v8f32(<8 x float> undef, <8 x float> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16F32 = call <16 x float> @llvm.copysign.v16f32(<16 x float> undef, <16 x float> undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %NXV1F32 = call <vscale x 1 x float> @llvm.copysign.nxv1f32(<vscale x 1 x float> undef, <vscale x 1 x float> undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %NXV2F32 = call <vscale x 2 x float> @llvm.copysign.nxv2f32(<vscale x 2 x float> undef, <vscale x 2 x float> undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %NXV4F32 = call <vscale x 4 x float> @llvm.copysign.nxv4f32(<vscale x 4 x float> undef, <vscale x 4 x float> undef)
@@ -512,8 +512,8 @@ define i32 @fcopysign() {
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %NXV16F32 = call <vscale x 16 x float> @llvm.copysign.nxv16f32(<vscale x 16 x float> undef, <vscale x 16 x float> undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V1F64 = call <1 x double> @llvm.copysign.v1f64(<1 x double> undef, <1 x double> undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2F64 = call <2 x double> @llvm.copysign.v2f64(<2 x double> undef, <2 x double> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V4F64 = call <4 x double> @llvm.copysign.v4f64(<4 x double> undef, <4 x double> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V8F64 = call <8 x double> @llvm.copysign.v8f64(<8 x double> undef, <8 x double> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4F64 = call <4 x double> @llvm.copysign.v4f64(<4 x double> undef, <4 x double> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8F64 = call <8 x double> @llvm.copysign.v8f64(<8 x double> undef, <8 x double> undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %NXV1F64 = call <vscale x 1 x double> @llvm.copysign.nxv1f64(<vscale x 1 x double> undef, <vscale x 1 x double> undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %NXV2F64 = call <vscale x 2 x double> @llvm.copysign.nxv2f64(<vscale x 2 x double> undef, <vscale x 2 x double> undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %NXV4F64 = call <vscale x 4 x double> @llvm.copysign.nxv4f64(<vscale x 4 x double> undef, <vscale x 4 x double> undef)
@@ -572,8 +572,8 @@ define i32 @fma() {
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2F16 = call <2 x half> @llvm.fma.v2f16(<2 x half> undef, <2 x half> undef, <2 x half> undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4F16 = call <4 x half> @llvm.fma.v4f16(<4 x half> undef, <4 x half> undef, <4 x half> undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8F16 = call <8 x half> @llvm.fma.v8f16(<8 x half> undef, <8 x half> undef, <8 x half> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V16F16 = call <16 x half> @llvm.fma.v16f16(<16 x half> undef, <16 x half> undef, <16 x half> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V32F16 = call <32 x half> @llvm.fma.v32f16(<32 x half> undef, <32 x half> undef, <32 x half> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16F16 = call <16 x half> @llvm.fma.v16f16(<16 x half> undef, <16 x half> undef, <16 x half> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32F16 = call <32 x half> @llvm.fma.v32f16(<32 x half> undef, <32 x half> undef, <32 x half> undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %NXV1F16 = call <vscale x 1 x half> @llvm.fma.nxv1f16(<vscale x 1 x half> undef, <vscale x 1 x half> undef, <vscale x 1 x half> undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %NXV2F16 = call <vscale x 2 x half> @llvm.fma.nxv2f16(<vscale x 2 x half> undef, <vscale x 2 x half> undef, <vscale x 2 x half> undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %NXV4F16 = call <vscale x 4 x half> @llvm.fma.nxv4f16(<vscale x 4 x half> undef, <vscale x 4 x half> undef, <vscale x 4 x half> undef)
@@ -583,8 +583,8 @@ define i32 @fma() {
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V1F32 = call <1 x float> @llvm.fma.v1f32(<1 x float> undef, <1 x float> undef, <1 x float> undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2F32 = call <2 x float> @llvm.fma.v2f32(<2 x float> undef, <2 x float> undef, <2 x float> undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4F32 = call <4 x float> @llvm.fma.v4f32(<4 x float> undef, <4 x float> undef, <4 x float> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V8F32 = call <8 x float> @llvm.fma.v8f32(<8 x float> undef, <8 x float> undef, <8 x float> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16F32 = call <16 x float> @llvm.fma.v16f32(<16 x float> undef, <16 x float> undef, <16 x float> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8F32 = call <8 x float> @llvm.fma.v8f32(<8 x float> undef, <8 x float> undef, <8 x float> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16F32 = call <16 x float> @llvm.fma.v16f32(<16 x float> undef, <16 x float> undef, <16 x float> undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %NXV1F32 = call <vscale x 1 x float> @llvm.fma.nxv1f32(<vscale x 1 x float> undef, <vscale x 1 x float> undef, <vscale x 1 x float> undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %NXV2F32 = call <vscale x 2 x float> @llvm.fma.nxv2f32(<vscale x 2 x float> undef, <vscale x 2 x float> undef, <vscale x 2 x float> undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %NXV4F32 = call <vscale x 4 x float> @llvm.fma.nxv4f32(<vscale x 4 x float> undef, <vscale x 4 x float> undef, <vscale x 4 x float> undef)
@@ -592,8 +592,8 @@ define i32 @fma() {
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %NXV16F32 = call <vscale x 16 x float> @llvm.fma.nxv16f32(<vscale x 16 x float> undef, <vscale x 16 x float> undef, <vscale x 16 x float> undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V1F64 = call <1 x double> @llvm.fma.v1f64(<1 x double> undef, <1 x double> undef, <1 x double> undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2F64 = call <2 x double> @llvm.fma.v2f64(<2 x double> undef, <2 x double> undef, <2 x double> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V4F64 = call <4 x double> @llvm.fma.v4f64(<4 x double> undef, <4 x double> undef, <4 x double> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V8F64 = call <8 x double> @llvm.fma.v8f64(<8 x double> undef, <8 x double> undef, <8 x double> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4F64 = call <4 x double> @llvm.fma.v4f64(<4 x double> undef, <4 x double> undef, <4 x double> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8F64 = call <8 x double> @llvm.fma.v8f64(<8 x double> undef, <8 x double> undef, <8 x double> undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %NXV1F64 = call <vscale x 1 x double> @llvm.fma.nxv1f64(<vscale x 1 x double> undef, <vscale x 1 x double> undef, <vscale x 1 x double> undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %NXV2F64 = call <vscale x 2 x double> @llvm.fma.nxv2f64(<vscale x 2 x double> undef, <vscale x 2 x double> undef, <vscale x 2 x double> undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %NXV4F64 = call <vscale x 4 x double> @llvm.fma.nxv4f64(<vscale x 4 x double> undef, <vscale x 4 x double> undef, <vscale x 4 x double> undef)
@@ -651,15 +651,15 @@ define void @fmuladd() {
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %4 = call <2 x half> @llvm.fmuladd.v2f16(<2 x half> undef, <2 x half> undef, <2 x half> undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %5 = call <4 x half> @llvm.fmuladd.v4f16(<4 x half> undef, <4 x half> undef, <4 x half> undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %6 = call <8 x half> @llvm.fmuladd.v8f16(<8 x half> undef, <8 x half> undef, <8 x half> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %7 = call <16 x half> @llvm.fmuladd.v16f16(<16 x half> undef, <16 x half> undef, <16 x half> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %7 = call <16 x half> @llvm.fmuladd.v16f16(<16 x half> undef, <16 x half> undef, <16 x half> undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %8 = call <2 x float> @llvm.fmuladd.v2f32(<2 x float> undef, <2 x float> undef, <2 x float> undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %9 = call <4 x float> @llvm.fmuladd.v4f32(<4 x float> undef, <4 x float> undef, <4 x float> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %10 = call <8 x float> @llvm.fmuladd.v8f32(<8 x float> undef, <8 x float> undef, <8 x float> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %11 = call <16 x float> @llvm.fmuladd.v16f32(<16 x float> undef, <16 x float> undef, <16 x float> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %10 = call <8 x float> @llvm.fmuladd.v8f32(<8 x float> undef, <8 x float> undef, <8 x float> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %11 = call <16 x float> @llvm.fmuladd.v16f32(<16 x float> undef, <16 x float> undef, <16 x float> undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %12 = call <2 x double> @llvm.fmuladd.v2f64(<2 x double> undef, <2 x double> undef, <2 x double> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %13 = call <4 x double> @llvm.fmuladd.v4f64(<4 x double> undef, <4 x double> undef, <4 x double> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %14 = call <8 x double> @llvm.fmuladd.v8f64(<8 x double> undef, <8 x double> undef, <8 x double> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %15 = call <16 x double> @llvm.fmuladd.v16f64(<16 x double> undef, <16 x double> undef, <16 x double> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %13 = call <4 x double> @llvm.fmuladd.v4f64(<4 x double> undef, <4 x double> undef, <4 x double> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %14 = call <8 x double> @llvm.fmuladd.v8f64(<8 x double> undef, <8 x double> undef, <8 x double> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %15 = call <16 x double> @llvm.fmuladd.v16f64(<16 x double> undef, <16 x double> undef, <16 x double> undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %16 = call <vscale x 1 x half> @llvm.fmuladd.nxv1f16(<vscale x 1 x half> undef, <vscale x 1 x half> undef, <vscale x 1 x half> undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %17 = call <vscale x 2 x half> @llvm.fmuladd.nxv2f16(<vscale x 2 x half> undef, <vscale x 2 x half> undef, <vscale x 2 x half> undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %18 = call <vscale x 4 x half> @llvm.fmuladd.nxv4f16(<vscale x 4 x half> undef, <vscale x 4 x half> undef, <vscale x 4 x half> undef)
diff --git a/llvm/test/Analysis/CostModel/RISCV/arith-int.ll b/llvm/test/Analysis/CostModel/RISCV/arith-int.ll
index 00f2cd7b63a4..b4afbb513166 100644
--- a/llvm/test/Analysis/CostModel/RISCV/arith-int.ll
+++ b/llvm/test/Analysis/CostModel/RISCV/arith-int.ll
@@ -1,6 +1,6 @@
 ; NOTE: Assertions have been autogenerated by utils/update_analyze_test_checks.py
-; RUN: opt -passes="print<cost-model>" 2>&1 -disable-output -mtriple=riscv64 -mattr=+v,+f,+d,+zfh,+zvfh -riscv-v-fixed-length-vector-lmul-max=1 < %s | FileCheck %s
-; RUN: opt -passes="print<cost-model>" 2>&1 -disable-output -mtriple=riscv64 -mcpu=sifive-x280 -riscv-v-fixed-length-vector-lmul-max=1 < %s | FileCheck %s --check-prefix=SIFIVE-X280
+; RUN: opt -passes="print<cost-model>" 2>&1 -disable-output -mtriple=riscv64 -mattr=+v,+f,+d,+zfh,+zvfh < %s | FileCheck %s
+; RUN: opt -passes="print<cost-model>" 2>&1 -disable-output -mtriple=riscv64 -mcpu=sifive-x280 < %s | FileCheck %s --check-prefix=SIFIVE-X280
 ; Check that we don't crash querying costs when vectors are not enabled.
 ; RUN: opt -passes="print<cost-model>" 2>&1 -disable-output -mtriple=riscv64
 
@@ -709,8 +709,8 @@ define i32 @udiv() {
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2I16 = udiv <2 x i16> undef, undef
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4I16 = udiv <4 x i16> undef, undef
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8I16 = udiv <8 x i16> undef, undef
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V16I16 = udiv <16 x i16> undef, undef
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V32I16 = udiv <32 x i16> undef, undef
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16I16 = udiv <16 x i16> undef, undef
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32I16 = udiv <32 x i16> undef, undef
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %NXV1I16 = udiv <vscale x 1 x i16> undef, undef
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %NXV2I16 = udiv <vscale x 2 x i16> undef, undef
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %NXV4I16 = udiv <vscale x 4 x i16> undef, undef
@@ -720,8 +720,8 @@ define i32 @udiv() {
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V1I32 = udiv <1 x i32> undef, undef
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2I32 = udiv <2 x i32> undef, undef
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4I32 = udiv <4 x i32> undef, undef
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V8I32 = udiv <8 x i32> undef, undef
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16I32 = udiv <16 x i32> undef, undef
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8I32 = udiv <8 x i32> undef, undef
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16I32 = udiv <16 x i32> undef, undef
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %NXV1I32 = udiv <vscale x 1 x i32> undef, undef
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %NXV2I32 = udiv <vscale x 2 x i32> undef, undef
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %NXV4I32 = udiv <vscale x 4 x i32> undef, undef
@@ -729,8 +729,8 @@ define i32 @udiv() {
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %NXV16I32 = udiv <vscale x 16 x i32> undef, undef
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V1I64 = udiv <1 x i64> undef, undef
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2I64 = udiv <2 x i64> undef, undef
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V4I64 = udiv <4 x i64> undef, undef
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V8I64 = udiv <8 x i64> undef, undef
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4I64 = udiv <4 x i64> undef, undef
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8I64 = udiv <8 x i64> undef, undef
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %NXV1I64 = udiv <vscale x 1 x i64> undef, undef
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %NXV2I64 = udiv <vscale x 2 x i64> undef, undef
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %NXV4I64 = udiv <vscale x 4 x i64> undef, undef
@@ -825,8 +825,8 @@ define i32 @urem() {
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2I16 = urem <2 x i16> undef, undef
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4I16 = urem <4 x i16> undef, undef
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8I16 = urem <8 x i16> undef, undef
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V16I16 = urem <16 x i16> undef, undef
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V32I16 = urem <32 x i16> undef, undef
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16I16 = urem <16 x i16> undef, undef
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32I16 = urem <32 x i16> undef, undef
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %NXV1I16 = urem <vscale x 1 x i16> undef, undef
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %NXV2I16 = urem <vscale x 2 x i16> undef, undef
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %NXV4I16 = urem <vscale x 4 x i16> undef, undef
@@ -836,8 +836,8 @@ define i32 @urem() {
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V1I32 = urem <1 x i32> undef, undef
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2I32 = urem <2 x i32> undef, undef
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4I32 = urem <4 x i32> undef, undef
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V8I32 = urem <8 x i32> undef, undef
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16I32 = urem <16 x i32> undef, undef
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8I32 = urem <8 x i32> undef, undef
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16I32 = urem <16 x i32> undef, undef
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %NXV1I32 = urem <vscale x 1 x i32> undef, undef
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %NXV2I32 = urem <vscale x 2 x i32> undef, undef
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %NXV4I32 = urem <vscale x 4 x i32> undef, undef
@@ -845,8 +845,8 @@ define i32 @urem() {
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %NXV16I32 = urem <vscale x 16 x i32> undef, undef
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V1I64 = urem <1 x i64> undef, undef
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2I64 = urem <2 x i64> undef, undef
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V4I64 = urem <4 x i64> undef, undef
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V8I64 = urem <8 x i64> undef, undef
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4I64 = urem <4 x i64> undef, undef
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8I64 = urem <8 x i64> undef, undef
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %NXV1I64 = urem <vscale x 1 x i64> undef, undef
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %NXV2I64 = urem <vscale x 2 x i64> undef, undef
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %NXV4I64 = urem <vscale x 4 x i64> undef, undef
@@ -941,8 +941,8 @@ define i32 @sdiv() {
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2I16 = sdiv <2 x i16> undef, undef
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4I16 = sdiv <4 x i16> undef, undef
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8I16 = sdiv <8 x i16> undef, undef
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V16I16 = sdiv <16 x i16> undef, undef
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V32I16 = sdiv <32 x i16> undef, undef
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16I16 = sdiv <16 x i16> undef, undef
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32I16 = sdiv <32 x i16> undef, undef
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %NXV1I16 = sdiv <vscale x 1 x i16> undef, undef
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %NXV2I16 = sdiv <vscale x 2 x i16> undef, undef
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %NXV4I16 = sdiv <vscale x 4 x i16> undef, undef
@@ -952,8 +952,8 @@ define i32 @sdiv() {
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V1I32 = sdiv <1 x i32> undef, undef
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2I32 = sdiv <2 x i32> undef, undef
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4I32 = sdiv <4 x i32> undef, undef
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V8I32 = sdiv <8 x i32> undef, undef
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16I32 = sdiv <16 x i32> undef, undef
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8I32 = sdiv <8 x i32> undef, undef
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16I32 = sdiv <16 x i32> undef, undef
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %NXV1I32 = sdiv <vscale x 1 x i32> undef, undef
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %NXV2I32 = sdiv <vscale x 2 x i32> undef, undef
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %NXV4I32 = sdiv <vscale x 4 x i32> undef, undef
@@ -961,8 +961,8 @@ define i32 @sdiv() {
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %NXV16I32 = sdiv <vscale x 16 x i32> undef, undef
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V1I64 = sdiv <1 x i64> undef, undef
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2I64 = sdiv <2 x i64> undef, undef
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V4I64 = sdiv <4 x i64> undef, undef
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V8I64 = sdiv <8 x i64> undef, undef
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4I64 = sdiv <4 x i64> undef, undef
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8I64 = sdiv <8 x i64> undef, undef
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %NXV1I64 = sdiv <vscale x 1 x i64> undef, undef
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %NXV2I64 = sdiv <vscale x 2 x i64> undef, undef
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %NXV4I64 = sdiv <vscale x 4 x i64> undef, undef
@@ -1057,8 +1057,8 @@ define i32 @srem() {
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2I16 = srem <2 x i16> undef, undef
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4I16 = srem <4 x i16> undef, undef
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8I16 = srem <8 x i16> undef, undef
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V16I16 = srem <16 x i16> undef, undef
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V32I16 = srem <32 x i16> undef, undef
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16I16 = srem <16 x i16> undef, undef
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V32I16 = srem <32 x i16> undef, undef
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %NXV1I16 = srem <vscale x 1 x i16> undef, undef
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %NXV2I16 = srem <vscale x 2 x i16> undef, undef
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %NXV4I16 = srem <vscale x 4 x i16> undef, undef
@@ -1068,8 +1068,8 @@ define i32 @srem() {
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V1I32 = srem <1 x i32> undef, undef
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2I32 = srem <2 x i32> undef, undef
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4I32 = srem <4 x i32> undef, undef
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V8I32 = srem <8 x i32> undef, undef
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V16I32 = srem <16 x i32> undef, undef
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8I32 = srem <8 x i32> undef, undef
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V16I32 = srem <16 x i32> undef, undef
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %NXV1I32 = srem <vscale x 1 x i32> undef, undef
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %NXV2I32 = srem <vscale x 2 x i32> undef, undef
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %NXV4I32 = srem <vscale x 4 x i32> undef, undef
@@ -1077,8 +1077,8 @@ define i32 @srem() {
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %NXV16I32 = srem <vscale x 16 x i32> undef, undef
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V1I64 = srem <1 x i64> undef, undef
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V2I64 = srem <2 x i64> undef, undef
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %V4I64 = srem <4 x i64> undef, undef
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %V8I64 = srem <8 x i64> undef, undef
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V4I64 = srem <4 x i64> undef, undef
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %V8I64 = srem <8 x i64> undef, undef
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %NXV1I64 = srem <vscale x 1 x i64> undef, undef
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %NXV2I64 = srem <vscale x 2 x i64> undef, undef
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %NXV4I64 = srem <vscale x 4 x i64> undef, undef
diff --git a/llvm/test/Analysis/CostModel/RISCV/rvv-shuffle.ll b/llvm/test/Analysis/CostModel/RISCV/rvv-shuffle.ll
index 7cc7cff0e6e8..e068ab638d3a 100644
--- a/llvm/test/Analysis/CostModel/RISCV/rvv-shuffle.ll
+++ b/llvm/test/Analysis/CostModel/RISCV/rvv-shuffle.ll
@@ -78,148 +78,148 @@ declare <vscale x 16 x i32> @llvm.vector.insert.nxv16i32.nxv4i32(<vscale x 16 x
 
 define void @vector_reverse() {
 ; CHECK-LABEL: 'vector_reverse'
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 11 for instruction: %reverse_nxv16i8 = call <vscale x 16 x i8> @llvm.experimental.vector.reverse.nxv16i8(<vscale x 16 x i8> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 27 for instruction: %reverse_nxv32i8 = call <vscale x 32 x i8> @llvm.experimental.vector.reverse.nxv32i8(<vscale x 32 x i8> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %reverse_nxv2i16 = call <vscale x 2 x i16> @llvm.experimental.vector.reverse.nxv2i16(<vscale x 2 x i16> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %reverse_nxv4i16 = call <vscale x 4 x i16> @llvm.experimental.vector.reverse.nxv4i16(<vscale x 4 x i16> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 11 for instruction: %reverse_nxv8i16 = call <vscale x 8 x i16> @llvm.experimental.vector.reverse.nxv8i16(<vscale x 8 x i16> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 27 for instruction: %reverse_nxv16i16 = call <vscale x 16 x i16> @llvm.experimental.vector.reverse.nxv16i16(<vscale x 16 x i16> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 11 for instruction: %reverse_nxv4i32 = call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 27 for instruction: %reverse_nxv8i32 = call <vscale x 8 x i32> @llvm.experimental.vector.reverse.nxv8i32(<vscale x 8 x i32> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 11 for instruction: %reverse_nxv2i64 = call <vscale x 2 x i64> @llvm.experimental.vector.reverse.nxv2i64(<vscale x 2 x i64> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 27 for instruction: %reverse_nxv4i64 = call <vscale x 4 x i64> @llvm.experimental.vector.reverse.nxv4i64(<vscale x 4 x i64> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 83 for instruction: %reverse_nxv8i64 = call <vscale x 8 x i64> @llvm.experimental.vector.reverse.nxv8i64(<vscale x 8 x i64> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 166 for instruction: %reverse_nxv16i64 = call <vscale x 16 x i64> @llvm.experimental.vector.reverse.nxv16i64(<vscale x 16 x i64> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 332 for instruction: %reverse_nxv32i64 = call <vscale x 32 x i64> @llvm.experimental.vector.reverse.nxv32i64(<vscale x 32 x i64> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 14 for instruction: %reverse_nxv16i1 = call <vscale x 16 x i1> @llvm.experimental.vector.reverse.nxv16i1(<vscale x 16 x i1> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %reverse_nxv8i1 = call <vscale x 8 x i1> @llvm.experimental.vector.reverse.nxv8i1(<vscale x 8 x i1> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %reverse_nxv4i1 = call <vscale x 4 x i1> @llvm.experimental.vector.reverse.nxv4i1(<vscale x 4 x i1> undef)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %reverse_nxv2i1 = call <vscale x 2 x i1> @llvm.experimental.vector.reverse.nxv2i1(<vscale x 2 x i1> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 11 for instruction: %reverse_nxv16i8 = call <vscale x 16 x i8> @llvm.vector.reverse.nxv16i8(<vscale x 16 x i8> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 27 for instruction: %reverse_nxv32i8 = call <vscale x 32 x i8> @llvm.vector.reverse.nxv32i8(<vscale x 32 x i8> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %reverse_nxv2i16 = call <vscale x 2 x i16> @llvm.vector.reverse.nxv2i16(<vscale x 2 x i16> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %reverse_nxv4i16 = call <vscale x 4 x i16> @llvm.vector.reverse.nxv4i16(<vscale x 4 x i16> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 11 for instruction: %reverse_nxv8i16 = call <vscale x 8 x i16> @llvm.vector.reverse.nxv8i16(<vscale x 8 x i16> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 27 for instruction: %reverse_nxv16i16 = call <vscale x 16 x i16> @llvm.vector.reverse.nxv16i16(<vscale x 16 x i16> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 11 for instruction: %reverse_nxv4i32 = call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 27 for instruction: %reverse_nxv8i32 = call <vscale x 8 x i32> @llvm.vector.reverse.nxv8i32(<vscale x 8 x i32> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 11 for instruction: %reverse_nxv2i64 = call <vscale x 2 x i64> @llvm.vector.reverse.nxv2i64(<vscale x 2 x i64> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 27 for instruction: %reverse_nxv4i64 = call <vscale x 4 x i64> @llvm.vector.reverse.nxv4i64(<vscale x 4 x i64> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 83 for instruction: %reverse_nxv8i64 = call <vscale x 8 x i64> @llvm.vector.reverse.nxv8i64(<vscale x 8 x i64> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 166 for instruction: %reverse_nxv16i64 = call <vscale x 16 x i64> @llvm.vector.reverse.nxv16i64(<vscale x 16 x i64> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 332 for instruction: %reverse_nxv32i64 = call <vscale x 32 x i64> @llvm.vector.reverse.nxv32i64(<vscale x 32 x i64> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 14 for instruction: %reverse_nxv16i1 = call <vscale x 16 x i1> @llvm.vector.reverse.nxv16i1(<vscale x 16 x i1> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %reverse_nxv8i1 = call <vscale x 8 x i1> @llvm.vector.reverse.nxv8i1(<vscale x 8 x i1> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %reverse_nxv4i1 = call <vscale x 4 x i1> @llvm.vector.reverse.nxv4i1(<vscale x 4 x i1> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %reverse_nxv2i1 = call <vscale x 2 x i1> @llvm.vector.reverse.nxv2i1(<vscale x 2 x i1> undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
 ;
 ; SIZE-LABEL: 'vector_reverse'
-; SIZE-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %reverse_nxv16i8 = call <vscale x 16 x i8> @llvm.experimental.vector.reverse.nxv16i8(<vscale x 16 x i8> undef)
-; SIZE-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %reverse_nxv32i8 = call <vscale x 32 x i8> @llvm.experimental.vector.reverse.nxv32i8(<vscale x 32 x i8> undef)
-; SIZE-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %reverse_nxv2i16 = call <vscale x 2 x i16> @llvm.experimental.vector.reverse.nxv2i16(<vscale x 2 x i16> undef)
-; SIZE-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %reverse_nxv4i16 = call <vscale x 4 x i16> @llvm.experimental.vector.reverse.nxv4i16(<vscale x 4 x i16> undef)
-; SIZE-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %reverse_nxv8i16 = call <vscale x 8 x i16> @llvm.experimental.vector.reverse.nxv8i16(<vscale x 8 x i16> undef)
-; SIZE-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %reverse_nxv16i16 = call <vscale x 16 x i16> @llvm.experimental.vector.reverse.nxv16i16(<vscale x 16 x i16> undef)
-; SIZE-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %reverse_nxv4i32 = call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> undef)
-; SIZE-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %reverse_nxv8i32 = call <vscale x 8 x i32> @llvm.experimental.vector.reverse.nxv8i32(<vscale x 8 x i32> undef)
-; SIZE-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %reverse_nxv2i64 = call <vscale x 2 x i64> @llvm.experimental.vector.reverse.nxv2i64(<vscale x 2 x i64> undef)
-; SIZE-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %reverse_nxv4i64 = call <vscale x 4 x i64> @llvm.experimental.vector.reverse.nxv4i64(<vscale x 4 x i64> undef)
-; SIZE-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %reverse_nxv8i64 = call <vscale x 8 x i64> @llvm.experimental.vector.reverse.nxv8i64(<vscale x 8 x i64> undef)
-; SIZE-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %reverse_nxv16i64 = call <vscale x 16 x i64> @llvm.experimental.vector.reverse.nxv16i64(<vscale x 16 x i64> undef)
-; SIZE-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %reverse_nxv32i64 = call <vscale x 32 x i64> @llvm.experimental.vector.reverse.nxv32i64(<vscale x 32 x i64> undef)
-; SIZE-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %reverse_nxv16i1 = call <vscale x 16 x i1> @llvm.experimental.vector.reverse.nxv16i1(<vscale x 16 x i1> undef)
-; SIZE-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %reverse_nxv8i1 = call <vscale x 8 x i1> @llvm.experimental.vector.reverse.nxv8i1(<vscale x 8 x i1> undef)
-; SIZE-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %reverse_nxv4i1 = call <vscale x 4 x i1> @llvm.experimental.vector.reverse.nxv4i1(<vscale x 4 x i1> undef)
-; SIZE-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %reverse_nxv2i1 = call <vscale x 2 x i1> @llvm.experimental.vector.reverse.nxv2i1(<vscale x 2 x i1> undef)
+; SIZE-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %reverse_nxv16i8 = call <vscale x 16 x i8> @llvm.vector.reverse.nxv16i8(<vscale x 16 x i8> undef)
+; SIZE-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %reverse_nxv32i8 = call <vscale x 32 x i8> @llvm.vector.reverse.nxv32i8(<vscale x 32 x i8> undef)
+; SIZE-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %reverse_nxv2i16 = call <vscale x 2 x i16> @llvm.vector.reverse.nxv2i16(<vscale x 2 x i16> undef)
+; SIZE-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %reverse_nxv4i16 = call <vscale x 4 x i16> @llvm.vector.reverse.nxv4i16(<vscale x 4 x i16> undef)
+; SIZE-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %reverse_nxv8i16 = call <vscale x 8 x i16> @llvm.vector.reverse.nxv8i16(<vscale x 8 x i16> undef)
+; SIZE-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %reverse_nxv16i16 = call <vscale x 16 x i16> @llvm.vector.reverse.nxv16i16(<vscale x 16 x i16> undef)
+; SIZE-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %reverse_nxv4i32 = call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> undef)
+; SIZE-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %reverse_nxv8i32 = call <vscale x 8 x i32> @llvm.vector.reverse.nxv8i32(<vscale x 8 x i32> undef)
+; SIZE-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %reverse_nxv2i64 = call <vscale x 2 x i64> @llvm.vector.reverse.nxv2i64(<vscale x 2 x i64> undef)
+; SIZE-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %reverse_nxv4i64 = call <vscale x 4 x i64> @llvm.vector.reverse.nxv4i64(<vscale x 4 x i64> undef)
+; SIZE-NEXT:  Cost Model: Found an estimated cost of 6 for instruction: %reverse_nxv8i64 = call <vscale x 8 x i64> @llvm.vector.reverse.nxv8i64(<vscale x 8 x i64> undef)
+; SIZE-NEXT:  Cost Model: Found an estimated cost of 12 for instruction: %reverse_nxv16i64 = call <vscale x 16 x i64> @llvm.vector.reverse.nxv16i64(<vscale x 16 x i64> undef)
+; SIZE-NEXT:  Cost Model: Found an estimated cost of 24 for instruction: %reverse_nxv32i64 = call <vscale x 32 x i64> @llvm.vector.reverse.nxv32i64(<vscale x 32 x i64> undef)
+; SIZE-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %reverse_nxv16i1 = call <vscale x 16 x i1> @llvm.vector.reverse.nxv16i1(<vscale x 16 x i1> undef)
+; SIZE-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %reverse_nxv8i1 = call <vscale x 8 x i1> @llvm.vector.reverse.nxv8i1(<vscale x 8 x i1> undef)
+; SIZE-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %reverse_nxv4i1 = call <vscale x 4 x i1> @llvm.vector.reverse.nxv4i1(<vscale x 4 x i1> undef)
+; SIZE-NEXT:  Cost Model: Found an estimated cost of 9 for instruction: %reverse_nxv2i1 = call <vscale x 2 x i1> @llvm.vector.reverse.nxv2i1(<vscale x 2 x i1> undef)
 ; SIZE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: ret void
 ;
-  %reverse_nxv16i8 = call <vscale x 16 x i8> @llvm.experimental.vector.reverse.nxv16i8(<vscale x 16 x i8> undef)
-  %reverse_nxv32i8 = call <vscale x 32 x i8> @llvm.experimental.vector.reverse.nxv32i8(<vscale x 32 x i8> undef)
-  %reverse_nxv2i16 = call <vscale x 2 x i16> @llvm.experimental.vector.reverse.nxv2i16(<vscale x 2 x i16> undef)
-  %reverse_nxv4i16 = call <vscale x 4 x i16> @llvm.experimental.vector.reverse.nxv4i16(<vscale x 4 x i16> undef)
-  %reverse_nxv8i16 = call <vscale x 8 x i16> @llvm.experimental.vector.reverse.nxv8i16(<vscale x 8 x i16> undef)
-  %reverse_nxv16i16 = call <vscale x 16 x i16> @llvm.experimental.vector.reverse.nxv16i16(<vscale x 16 x i16> undef)
-  %reverse_nxv4i32 = call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> undef)
-  %reverse_nxv8i32 = call <vscale x 8 x i32> @llvm.experimental.vector.reverse.nxv8i32(<vscale x 8 x i32> undef)
-  %reverse_nxv2i64 = call <vscale x 2 x i64> @llvm.experimental.vector.reverse.nxv2i64(<vscale x 2 x i64> undef)
-  %reverse_nxv4i64 = call <vscale x 4 x i64> @llvm.experimental.vector.reverse.nxv4i64(<vscale x 4 x i64> undef)
-  %reverse_nxv8i64 = call <vscale x 8 x i64> @llvm.experimental.vector.reverse.nxv8i64(<vscale x 8 x i64> undef)
-  %reverse_nxv16i64 = call <vscale x 16 x i64> @llvm.experimental.vector.reverse.nxv16i64(<vscale x 16 x i64> undef)
-  %reverse_nxv32i64 = call <vscale x 32 x i64> @llvm.experimental.vector.reverse.nxv32i64(<vscale x 32 x i64> undef)
-  %reverse_nxv16i1 = call <vscale x 16 x i1> @llvm.experimental.vector.reverse.nxv16i1(<vscale x 16 x i1> undef)
-  %reverse_nxv8i1 =  call <vscale x 8 x i1> @llvm.experimental.vector.reverse.nxv8i1(<vscale x 8 x i1> undef)
-  %reverse_nxv4i1 = call <vscale x 4 x i1> @llvm.experimental.vector.reverse.nxv4i1(<vscale x 4 x i1> undef)
-  %reverse_nxv2i1 = call <vscale x 2 x i1> @llvm.experimental.vector.reverse.nxv2i1(<vscale x 2 x i1> undef)
+  %reverse_nxv16i8 = call <vscale x 16 x i8> @llvm.vector.reverse.nxv16i8(<vscale x 16 x i8> undef)
+  %reverse_nxv32i8 = call <vscale x 32 x i8> @llvm.vector.reverse.nxv32i8(<vscale x 32 x i8> undef)
+  %reverse_nxv2i16 = call <vscale x 2 x i16> @llvm.vector.reverse.nxv2i16(<vscale x 2 x i16> undef)
+  %reverse_nxv4i16 = call <vscale x 4 x i16> @llvm.vector.reverse.nxv4i16(<vscale x 4 x i16> undef)
+  %reverse_nxv8i16 = call <vscale x 8 x i16> @llvm.vector.reverse.nxv8i16(<vscale x 8 x i16> undef)
+  %reverse_nxv16i16 = call <vscale x 16 x i16> @llvm.vector.reverse.nxv16i16(<vscale x 16 x i16> undef)
+  %reverse_nxv4i32 = call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> undef)
+  %reverse_nxv8i32 = call <vscale x 8 x i32> @llvm.vector.reverse.nxv8i32(<vscale x 8 x i32> undef)
+  %reverse_nxv2i64 = call <vscale x 2 x i64> @llvm.vector.reverse.nxv2i64(<vscale x 2 x i64> undef)
+  %reverse_nxv4i64 = call <vscale x 4 x i64> @llvm.vector.reverse.nxv4i64(<vscale x 4 x i64> undef)
+  %reverse_nxv8i64 = call <vscale x 8 x i64> @llvm.vector.reverse.nxv8i64(<vscale x 8 x i64> undef)
+  %reverse_nxv16i64 = call <vscale x 16 x i64> @llvm.vector.reverse.nxv16i64(<vscale x 16 x i64> undef)
+  %reverse_nxv32i64 = call <vscale x 32 x i64> @llvm.vector.reverse.nxv32i64(<vscale x 32 x i64> undef)
+  %reverse_nxv16i1 = call <vscale x 16 x i1> @llvm.vector.reverse.nxv16i1(<vscale x 16 x i1> undef)
+  %reverse_nxv8i1 =  call <vscale x 8 x i1> @llvm.vector.reverse.nxv8i1(<vscale x 8 x i1> undef)
+  %reverse_nxv4i1 = call <vscale x 4 x i1> @llvm.vector.reverse.nxv4i1(<vscale x 4 x i1> undef)
+  %reverse_nxv2i1 = call <vscale x 2 x i1> @llvm.vector.reverse.nxv2i1(<vscale x 2 x i1> undef)
   ret void
 }
 
-declare <vscale x 16 x i8> @llvm.experimental.vector.reverse.nxv16i8(<vscale x 16 x i8>)
-declare <vscale x 32 x i8> @llvm.experimental.vector.reverse.nxv32i8(<vscale x 32 x i8>)
-declare <vscale x 2 x i16> @llvm.experimental.vector.reverse.nxv2i16(<vscale x 2 x i16>)
-declare <vscale x 4 x i16> @llvm.experimental.vector.reverse.nxv4i16(<vscale x 4 x i16>)
-declare <vscale x 8 x i16> @llvm.experimental.vector.reverse.nxv8i16(<vscale x 8 x i16>)
-declare <vscale x 16 x i16> @llvm.experimental.vector.reverse.nxv16i16(<vscale x 16 x i16>)
-declare <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32>)
-declare <vscale x 8 x i32> @llvm.experimental.vector.reverse.nxv8i32(<vscale x 8 x i32>)
-declare <vscale x 2 x i64> @llvm.experimental.vector.reverse.nxv2i64(<vscale x 2 x i64>)
-declare <vscale x 4 x i64> @llvm.experimental.vector.reverse.nxv4i64(<vscale x 4 x i64>)
-declare <vscale x 8 x i64> @llvm.experimental.vector.reverse.nxv8i64(<vscale x 8 x i64>)
-declare <vscale x 16 x i64> @llvm.experimental.vector.reverse.nxv16i64(<vscale x 16 x i64>)
-declare <vscale x 32 x i64> @llvm.experimental.vector.reverse.nxv32i64(<vscale x 32 x i64>)
-declare <vscale x 16 x i1> @llvm.experimental.vector.reverse.nxv16i1(<vscale x 16 x i1>)
-declare <vscale x 8 x i1> @llvm.experimental.vector.reverse.nxv8i1(<vscale x 8 x i1>)
-declare <vscale x 4 x i1> @llvm.experimental.vector.reverse.nxv4i1(<vscale x 4 x i1>)
-declare <vscale x 2 x i1> @llvm.experimental.vector.reverse.nxv2i1(<vscale x 2 x i1>)
+declare <vscale x 16 x i8> @llvm.vector.reverse.nxv16i8(<vscale x 16 x i8>)
+declare <vscale x 32 x i8> @llvm.vector.reverse.nxv32i8(<vscale x 32 x i8>)
+declare <vscale x 2 x i16> @llvm.vector.reverse.nxv2i16(<vscale x 2 x i16>)
+declare <vscale x 4 x i16> @llvm.vector.reverse.nxv4i16(<vscale x 4 x i16>)
+declare <vscale x 8 x i16> @llvm.vector.reverse.nxv8i16(<vscale x 8 x i16>)
+declare <vscale x 16 x i16> @llvm.vector.reverse.nxv16i16(<vscale x 16 x i16>)
+declare <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32>)
+declare <vscale x 8 x i32> @llvm.vector.reverse.nxv8i32(<vscale x 8 x i32>)
+declare <vscale x 2 x i64> @llvm.vector.reverse.nxv2i64(<vscale x 2 x i64>)
+declare <vscale x 4 x i64> @llvm.vector.reverse.nxv4i64(<vscale x 4 x i64>)
+declare <vscale x 8 x i64> @llvm.vector.reverse.nxv8i64(<vscale x 8 x i64>)
+declare <vscale x 16 x i64> @llvm.vector.reverse.nxv16i64(<vscale x 16 x i64>)
+declare <vscale x 32 x i64> @llvm.vector.reverse.nxv32i64(<vscale x 32 x i64>)
+declare <vscale x 16 x i1> @llvm.vector.reverse.nxv16i1(<vscale x 16 x i1>)
+declare <vscale x 8 x i1> @llvm.vector.reverse.nxv8i1(<vscale x 8 x i1>)
+declare <vscale x 4 x i1> @llvm.vector.reverse.nxv4i1(<vscale x 4 x i1>)
+declare <vscale x 2 x i1> @llvm.vector.reverse.nxv2i1(<vscale x 2 x i1>)
 
 
 define void @vector_splice() {
 ; CHECK-LABEL: 'vector_splice'
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %splice_nxv16i8 = call <vscale x 16 x i8> @llvm.experimental.vector.splice.nxv16i8(<vscale x 16 x i8> zeroinitializer, <vscale x 16 x i8> zeroinitializer, i32 1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %splice_nxv32i8 = call <vscale x 32 x i8> @llvm.experimental.vector.splice.nxv32i8(<vscale x 32 x i8> zeroinitializer, <vscale x 32 x i8> zeroinitializer, i32 1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice_nxv2i16 = call <vscale x 2 x i16> @llvm.experimental.vector.splice.nxv2i16(<vscale x 2 x i16> zeroinitializer, <vscale x 2 x i16> zeroinitializer, i32 1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice_nxv4i16 = call <vscale x 4 x i16> @llvm.experimental.vector.splice.nxv4i16(<vscale x 4 x i16> zeroinitializer, <vscale x 4 x i16> zeroinitializer, i32 1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %splice_nxv8i16 = call <vscale x 8 x i16> @llvm.experimental.vector.splice.nxv8i16(<vscale x 8 x i16> zeroinitializer, <vscale x 8 x i16> zeroinitializer, i32 1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %splice_nxv16i16 = call <vscale x 16 x i16> @llvm.experimental.vector.splice.nxv16i16(<vscale x 16 x i16> zeroinitializer, <vscale x 16 x i16> zeroinitializer, i32 1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %splice_nxv4i32 = call <vscale x 4 x i32> @llvm.experimental.vector.splice.nxv4i32(<vscale x 4 x i32> zeroinitializer, <vscale x 4 x i32> zeroinitializer, i32 1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %splice_nxv8i32 = call <vscale x 8 x i32> @llvm.experimental.vector.splice.nxv8i32(<vscale x 8 x i32> zeroinitializer, <vscale x 8 x i32> zeroinitializer, i32 1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %splice_nxv2i64 = call <vscale x 2 x i64> @llvm.experimental.vector.splice.nxv2i64(<vscale x 2 x i64> zeroinitializer, <vscale x 2 x i64> zeroinitializer, i32 1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %splice_nxv4i64 = call <vscale x 4 x i64> @llvm.experimental.vector.splice.nxv4i64(<vscale x 4 x i64> zeroinitializer, <vscale x 4 x i64> zeroinitializer, i32 1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %splice_nxv16i1 = call <vscale x 16 x i1> @llvm.experimental.vector.splice.nxv16i1(<vscale x 16 x i1> zeroinitializer, <vscale x 16 x i1> zeroinitializer, i32 1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice_nxv8i1 = call <vscale x 8 x i1> @llvm.experimental.vector.splice.nxv8i1(<vscale x 8 x i1> zeroinitializer, <vscale x 8 x i1> zeroinitializer, i32 1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice_nxv4i1 = call <vscale x 4 x i1> @llvm.experimental.vector.splice.nxv4i1(<vscale x 4 x i1> zeroinitializer, <vscale x 4 x i1> zeroinitializer, i32 1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice_nxv2i1 = call <vscale x 2 x i1> @llvm.experimental.vector.splice.nxv2i1(<vscale x 2 x i1> zeroinitializer, <vscale x 2 x i1> zeroinitializer, i32 1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %splice_nxv16i8 = call <vscale x 16 x i8> @llvm.vector.splice.nxv16i8(<vscale x 16 x i8> zeroinitializer, <vscale x 16 x i8> zeroinitializer, i32 1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %splice_nxv32i8 = call <vscale x 32 x i8> @llvm.vector.splice.nxv32i8(<vscale x 32 x i8> zeroinitializer, <vscale x 32 x i8> zeroinitializer, i32 1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice_nxv2i16 = call <vscale x 2 x i16> @llvm.vector.splice.nxv2i16(<vscale x 2 x i16> zeroinitializer, <vscale x 2 x i16> zeroinitializer, i32 1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice_nxv4i16 = call <vscale x 4 x i16> @llvm.vector.splice.nxv4i16(<vscale x 4 x i16> zeroinitializer, <vscale x 4 x i16> zeroinitializer, i32 1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %splice_nxv8i16 = call <vscale x 8 x i16> @llvm.vector.splice.nxv8i16(<vscale x 8 x i16> zeroinitializer, <vscale x 8 x i16> zeroinitializer, i32 1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %splice_nxv16i16 = call <vscale x 16 x i16> @llvm.vector.splice.nxv16i16(<vscale x 16 x i16> zeroinitializer, <vscale x 16 x i16> zeroinitializer, i32 1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %splice_nxv4i32 = call <vscale x 4 x i32> @llvm.vector.splice.nxv4i32(<vscale x 4 x i32> zeroinitializer, <vscale x 4 x i32> zeroinitializer, i32 1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %splice_nxv8i32 = call <vscale x 8 x i32> @llvm.vector.splice.nxv8i32(<vscale x 8 x i32> zeroinitializer, <vscale x 8 x i32> zeroinitializer, i32 1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %splice_nxv2i64 = call <vscale x 2 x i64> @llvm.vector.splice.nxv2i64(<vscale x 2 x i64> zeroinitializer, <vscale x 2 x i64> zeroinitializer, i32 1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %splice_nxv4i64 = call <vscale x 4 x i64> @llvm.vector.splice.nxv4i64(<vscale x 4 x i64> zeroinitializer, <vscale x 4 x i64> zeroinitializer, i32 1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %splice_nxv16i1 = call <vscale x 16 x i1> @llvm.vector.splice.nxv16i1(<vscale x 16 x i1> zeroinitializer, <vscale x 16 x i1> zeroinitializer, i32 1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice_nxv8i1 = call <vscale x 8 x i1> @llvm.vector.splice.nxv8i1(<vscale x 8 x i1> zeroinitializer, <vscale x 8 x i1> zeroinitializer, i32 1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice_nxv4i1 = call <vscale x 4 x i1> @llvm.vector.splice.nxv4i1(<vscale x 4 x i1> zeroinitializer, <vscale x 4 x i1> zeroinitializer, i32 1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice_nxv2i1 = call <vscale x 2 x i1> @llvm.vector.splice.nxv2i1(<vscale x 2 x i1> zeroinitializer, <vscale x 2 x i1> zeroinitializer, i32 1)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
 ;
 ; SIZE-LABEL: 'vector_splice'
-; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice_nxv16i8 = call <vscale x 16 x i8> @llvm.experimental.vector.splice.nxv16i8(<vscale x 16 x i8> zeroinitializer, <vscale x 16 x i8> zeroinitializer, i32 1)
-; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice_nxv32i8 = call <vscale x 32 x i8> @llvm.experimental.vector.splice.nxv32i8(<vscale x 32 x i8> zeroinitializer, <vscale x 32 x i8> zeroinitializer, i32 1)
-; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice_nxv2i16 = call <vscale x 2 x i16> @llvm.experimental.vector.splice.nxv2i16(<vscale x 2 x i16> zeroinitializer, <vscale x 2 x i16> zeroinitializer, i32 1)
-; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice_nxv4i16 = call <vscale x 4 x i16> @llvm.experimental.vector.splice.nxv4i16(<vscale x 4 x i16> zeroinitializer, <vscale x 4 x i16> zeroinitializer, i32 1)
-; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice_nxv8i16 = call <vscale x 8 x i16> @llvm.experimental.vector.splice.nxv8i16(<vscale x 8 x i16> zeroinitializer, <vscale x 8 x i16> zeroinitializer, i32 1)
-; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice_nxv16i16 = call <vscale x 16 x i16> @llvm.experimental.vector.splice.nxv16i16(<vscale x 16 x i16> zeroinitializer, <vscale x 16 x i16> zeroinitializer, i32 1)
-; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice_nxv4i32 = call <vscale x 4 x i32> @llvm.experimental.vector.splice.nxv4i32(<vscale x 4 x i32> zeroinitializer, <vscale x 4 x i32> zeroinitializer, i32 1)
-; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice_nxv8i32 = call <vscale x 8 x i32> @llvm.experimental.vector.splice.nxv8i32(<vscale x 8 x i32> zeroinitializer, <vscale x 8 x i32> zeroinitializer, i32 1)
-; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice_nxv2i64 = call <vscale x 2 x i64> @llvm.experimental.vector.splice.nxv2i64(<vscale x 2 x i64> zeroinitializer, <vscale x 2 x i64> zeroinitializer, i32 1)
-; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice_nxv4i64 = call <vscale x 4 x i64> @llvm.experimental.vector.splice.nxv4i64(<vscale x 4 x i64> zeroinitializer, <vscale x 4 x i64> zeroinitializer, i32 1)
-; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice_nxv16i1 = call <vscale x 16 x i1> @llvm.experimental.vector.splice.nxv16i1(<vscale x 16 x i1> zeroinitializer, <vscale x 16 x i1> zeroinitializer, i32 1)
-; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice_nxv8i1 = call <vscale x 8 x i1> @llvm.experimental.vector.splice.nxv8i1(<vscale x 8 x i1> zeroinitializer, <vscale x 8 x i1> zeroinitializer, i32 1)
-; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice_nxv4i1 = call <vscale x 4 x i1> @llvm.experimental.vector.splice.nxv4i1(<vscale x 4 x i1> zeroinitializer, <vscale x 4 x i1> zeroinitializer, i32 1)
-; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice_nxv2i1 = call <vscale x 2 x i1> @llvm.experimental.vector.splice.nxv2i1(<vscale x 2 x i1> zeroinitializer, <vscale x 2 x i1> zeroinitializer, i32 1)
+; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice_nxv16i8 = call <vscale x 16 x i8> @llvm.vector.splice.nxv16i8(<vscale x 16 x i8> zeroinitializer, <vscale x 16 x i8> zeroinitializer, i32 1)
+; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice_nxv32i8 = call <vscale x 32 x i8> @llvm.vector.splice.nxv32i8(<vscale x 32 x i8> zeroinitializer, <vscale x 32 x i8> zeroinitializer, i32 1)
+; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice_nxv2i16 = call <vscale x 2 x i16> @llvm.vector.splice.nxv2i16(<vscale x 2 x i16> zeroinitializer, <vscale x 2 x i16> zeroinitializer, i32 1)
+; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice_nxv4i16 = call <vscale x 4 x i16> @llvm.vector.splice.nxv4i16(<vscale x 4 x i16> zeroinitializer, <vscale x 4 x i16> zeroinitializer, i32 1)
+; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice_nxv8i16 = call <vscale x 8 x i16> @llvm.vector.splice.nxv8i16(<vscale x 8 x i16> zeroinitializer, <vscale x 8 x i16> zeroinitializer, i32 1)
+; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice_nxv16i16 = call <vscale x 16 x i16> @llvm.vector.splice.nxv16i16(<vscale x 16 x i16> zeroinitializer, <vscale x 16 x i16> zeroinitializer, i32 1)
+; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice_nxv4i32 = call <vscale x 4 x i32> @llvm.vector.splice.nxv4i32(<vscale x 4 x i32> zeroinitializer, <vscale x 4 x i32> zeroinitializer, i32 1)
+; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice_nxv8i32 = call <vscale x 8 x i32> @llvm.vector.splice.nxv8i32(<vscale x 8 x i32> zeroinitializer, <vscale x 8 x i32> zeroinitializer, i32 1)
+; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice_nxv2i64 = call <vscale x 2 x i64> @llvm.vector.splice.nxv2i64(<vscale x 2 x i64> zeroinitializer, <vscale x 2 x i64> zeroinitializer, i32 1)
+; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice_nxv4i64 = call <vscale x 4 x i64> @llvm.vector.splice.nxv4i64(<vscale x 4 x i64> zeroinitializer, <vscale x 4 x i64> zeroinitializer, i32 1)
+; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice_nxv16i1 = call <vscale x 16 x i1> @llvm.vector.splice.nxv16i1(<vscale x 16 x i1> zeroinitializer, <vscale x 16 x i1> zeroinitializer, i32 1)
+; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice_nxv8i1 = call <vscale x 8 x i1> @llvm.vector.splice.nxv8i1(<vscale x 8 x i1> zeroinitializer, <vscale x 8 x i1> zeroinitializer, i32 1)
+; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice_nxv4i1 = call <vscale x 4 x i1> @llvm.vector.splice.nxv4i1(<vscale x 4 x i1> zeroinitializer, <vscale x 4 x i1> zeroinitializer, i32 1)
+; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice_nxv2i1 = call <vscale x 2 x i1> @llvm.vector.splice.nxv2i1(<vscale x 2 x i1> zeroinitializer, <vscale x 2 x i1> zeroinitializer, i32 1)
 ; SIZE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: ret void
 ;
-  %splice_nxv16i8 = call <vscale x 16 x i8> @llvm.experimental.vector.splice.nxv16i8(<vscale x 16 x i8> zeroinitializer, <vscale x 16 x i8> zeroinitializer, i32 1)
-  %splice_nxv32i8 = call <vscale x 32 x i8> @llvm.experimental.vector.splice.nxv32i8(<vscale x 32 x i8> zeroinitializer, <vscale x 32 x i8> zeroinitializer, i32 1)
-  %splice_nxv2i16 = call <vscale x 2 x i16> @llvm.experimental.vector.splice.nxv2i16(<vscale x 2 x i16> zeroinitializer, <vscale x 2 x i16> zeroinitializer, i32 1)
-  %splice_nxv4i16 = call <vscale x 4 x i16> @llvm.experimental.vector.splice.nxv4i16(<vscale x 4 x i16> zeroinitializer, <vscale x 4 x i16> zeroinitializer, i32 1)
-  %splice_nxv8i16 = call <vscale x 8 x i16> @llvm.experimental.vector.splice.nxv8i16(<vscale x 8 x i16> zeroinitializer, <vscale x 8 x i16> zeroinitializer, i32 1)
-  %splice_nxv16i16 = call <vscale x 16 x i16> @llvm.experimental.vector.splice.nxv16i16(<vscale x 16 x i16> zeroinitializer, <vscale x 16 x i16> zeroinitializer, i32 1)
-  %splice_nxv4i32 = call <vscale x 4 x i32> @llvm.experimental.vector.splice.nxv4i32(<vscale x 4 x i32> zeroinitializer, <vscale x 4 x i32> zeroinitializer, i32 1)
-  %splice_nxv8i32 = call <vscale x 8 x i32> @llvm.experimental.vector.splice.nxv8i32(<vscale x 8 x i32> zeroinitializer, <vscale x 8 x i32> zeroinitializer, i32 1)
-  %splice_nxv2i64 = call <vscale x 2 x i64> @llvm.experimental.vector.splice.nxv2i64(<vscale x 2 x i64> zeroinitializer, <vscale x 2 x i64> zeroinitializer, i32 1)
-  %splice_nxv4i64 = call <vscale x 4 x i64> @llvm.experimental.vector.splice.nxv4i64(<vscale x 4 x i64> zeroinitializer, <vscale x 4 x i64> zeroinitializer, i32 1)
-  %splice_nxv16i1 = call <vscale x 16 x i1> @llvm.experimental.vector.splice.nxv16i1(<vscale x 16 x i1> zeroinitializer, <vscale x 16 x i1> zeroinitializer, i32 1)
-  %splice_nxv8i1 =  call <vscale x 8 x i1> @llvm.experimental.vector.splice.nxv8i1(<vscale x 8 x i1> zeroinitializer, <vscale x 8 x i1> zeroinitializer, i32 1)
-  %splice_nxv4i1 = call <vscale x 4 x i1> @llvm.experimental.vector.splice.nxv4i1(<vscale x 4 x i1> zeroinitializer, <vscale x 4 x i1> zeroinitializer, i32 1)
-  %splice_nxv2i1 = call <vscale x 2 x i1> @llvm.experimental.vector.splice.nxv2i1(<vscale x 2 x i1> zeroinitializer, <vscale x 2 x i1> zeroinitializer, i32 1)
+  %splice_nxv16i8 = call <vscale x 16 x i8> @llvm.vector.splice.nxv16i8(<vscale x 16 x i8> zeroinitializer, <vscale x 16 x i8> zeroinitializer, i32 1)
+  %splice_nxv32i8 = call <vscale x 32 x i8> @llvm.vector.splice.nxv32i8(<vscale x 32 x i8> zeroinitializer, <vscale x 32 x i8> zeroinitializer, i32 1)
+  %splice_nxv2i16 = call <vscale x 2 x i16> @llvm.vector.splice.nxv2i16(<vscale x 2 x i16> zeroinitializer, <vscale x 2 x i16> zeroinitializer, i32 1)
+  %splice_nxv4i16 = call <vscale x 4 x i16> @llvm.vector.splice.nxv4i16(<vscale x 4 x i16> zeroinitializer, <vscale x 4 x i16> zeroinitializer, i32 1)
+  %splice_nxv8i16 = call <vscale x 8 x i16> @llvm.vector.splice.nxv8i16(<vscale x 8 x i16> zeroinitializer, <vscale x 8 x i16> zeroinitializer, i32 1)
+  %splice_nxv16i16 = call <vscale x 16 x i16> @llvm.vector.splice.nxv16i16(<vscale x 16 x i16> zeroinitializer, <vscale x 16 x i16> zeroinitializer, i32 1)
+  %splice_nxv4i32 = call <vscale x 4 x i32> @llvm.vector.splice.nxv4i32(<vscale x 4 x i32> zeroinitializer, <vscale x 4 x i32> zeroinitializer, i32 1)
+  %splice_nxv8i32 = call <vscale x 8 x i32> @llvm.vector.splice.nxv8i32(<vscale x 8 x i32> zeroinitializer, <vscale x 8 x i32> zeroinitializer, i32 1)
+  %splice_nxv2i64 = call <vscale x 2 x i64> @llvm.vector.splice.nxv2i64(<vscale x 2 x i64> zeroinitializer, <vscale x 2 x i64> zeroinitializer, i32 1)
+  %splice_nxv4i64 = call <vscale x 4 x i64> @llvm.vector.splice.nxv4i64(<vscale x 4 x i64> zeroinitializer, <vscale x 4 x i64> zeroinitializer, i32 1)
+  %splice_nxv16i1 = call <vscale x 16 x i1> @llvm.vector.splice.nxv16i1(<vscale x 16 x i1> zeroinitializer, <vscale x 16 x i1> zeroinitializer, i32 1)
+  %splice_nxv8i1 =  call <vscale x 8 x i1> @llvm.vector.splice.nxv8i1(<vscale x 8 x i1> zeroinitializer, <vscale x 8 x i1> zeroinitializer, i32 1)
+  %splice_nxv4i1 = call <vscale x 4 x i1> @llvm.vector.splice.nxv4i1(<vscale x 4 x i1> zeroinitializer, <vscale x 4 x i1> zeroinitializer, i32 1)
+  %splice_nxv2i1 = call <vscale x 2 x i1> @llvm.vector.splice.nxv2i1(<vscale x 2 x i1> zeroinitializer, <vscale x 2 x i1> zeroinitializer, i32 1)
   ret void
 }
 
-declare <vscale x 2 x i1> @llvm.experimental.vector.splice.nxv2i1(<vscale x 2 x i1>, <vscale x 2 x i1>, i32)
-declare <vscale x 4 x i1> @llvm.experimental.vector.splice.nxv4i1(<vscale x 4 x i1>, <vscale x 4 x i1>, i32)
-declare <vscale x 8 x i1> @llvm.experimental.vector.splice.nxv8i1(<vscale x 8 x i1>, <vscale x 8 x i1>, i32)
-declare <vscale x 16 x i1> @llvm.experimental.vector.splice.nxv16i1(<vscale x 16 x i1>, <vscale x 16 x i1>, i32)
-declare <vscale x 2 x i8> @llvm.experimental.vector.splice.nxv2i8(<vscale x 2 x i8>, <vscale x 2 x i8>, i32)
-declare <vscale x 16 x i8> @llvm.experimental.vector.splice.nxv16i8(<vscale x 16 x i8>, <vscale x 16 x i8>, i32)
-declare <vscale x 32 x i8> @llvm.experimental.vector.splice.nxv32i8(<vscale x 32 x i8>, <vscale x 32 x i8>, i32)
-declare <vscale x 2 x i16> @llvm.experimental.vector.splice.nxv2i16(<vscale x 2 x i16>, <vscale x 2 x i16>, i32)
-declare <vscale x 4 x i16> @llvm.experimental.vector.splice.nxv4i16(<vscale x 4 x i16>, <vscale x 4 x i16>, i32)
-declare <vscale x 8 x i16> @llvm.experimental.vector.splice.nxv8i16(<vscale x 8 x i16>, <vscale x 8 x i16>, i32)
-declare <vscale x 16 x i16> @llvm.experimental.vector.splice.nxv16i16(<vscale x 16 x i16>, <vscale x 16 x i16>, i32)
-declare <vscale x 4 x i32> @llvm.experimental.vector.splice.nxv4i32(<vscale x 4 x i32>, <vscale x 4 x i32>, i32)
-declare <vscale x 8 x i32> @llvm.experimental.vector.splice.nxv8i32(<vscale x 8 x i32>, <vscale x 8 x i32>, i32)
-declare <vscale x 2 x i64> @llvm.experimental.vector.splice.nxv2i64(<vscale x 2 x i64>, <vscale x 2 x i64>, i32)
-declare <vscale x 4 x i64> @llvm.experimental.vector.splice.nxv4i64(<vscale x 4 x i64>, <vscale x 4 x i64>, i32)
+declare <vscale x 2 x i1> @llvm.vector.splice.nxv2i1(<vscale x 2 x i1>, <vscale x 2 x i1>, i32)
+declare <vscale x 4 x i1> @llvm.vector.splice.nxv4i1(<vscale x 4 x i1>, <vscale x 4 x i1>, i32)
+declare <vscale x 8 x i1> @llvm.vector.splice.nxv8i1(<vscale x 8 x i1>, <vscale x 8 x i1>, i32)
+declare <vscale x 16 x i1> @llvm.vector.splice.nxv16i1(<vscale x 16 x i1>, <vscale x 16 x i1>, i32)
+declare <vscale x 2 x i8> @llvm.vector.splice.nxv2i8(<vscale x 2 x i8>, <vscale x 2 x i8>, i32)
+declare <vscale x 16 x i8> @llvm.vector.splice.nxv16i8(<vscale x 16 x i8>, <vscale x 16 x i8>, i32)
+declare <vscale x 32 x i8> @llvm.vector.splice.nxv32i8(<vscale x 32 x i8>, <vscale x 32 x i8>, i32)
+declare <vscale x 2 x i16> @llvm.vector.splice.nxv2i16(<vscale x 2 x i16>, <vscale x 2 x i16>, i32)
+declare <vscale x 4 x i16> @llvm.vector.splice.nxv4i16(<vscale x 4 x i16>, <vscale x 4 x i16>, i32)
+declare <vscale x 8 x i16> @llvm.vector.splice.nxv8i16(<vscale x 8 x i16>, <vscale x 8 x i16>, i32)
+declare <vscale x 16 x i16> @llvm.vector.splice.nxv16i16(<vscale x 16 x i16>, <vscale x 16 x i16>, i32)
+declare <vscale x 4 x i32> @llvm.vector.splice.nxv4i32(<vscale x 4 x i32>, <vscale x 4 x i32>, i32)
+declare <vscale x 8 x i32> @llvm.vector.splice.nxv8i32(<vscale x 8 x i32>, <vscale x 8 x i32>, i32)
+declare <vscale x 2 x i64> @llvm.vector.splice.nxv2i64(<vscale x 2 x i64>, <vscale x 2 x i64>, i32)
+declare <vscale x 4 x i64> @llvm.vector.splice.nxv4i64(<vscale x 4 x i64>, <vscale x 4 x i64>, i32)
diff --git a/llvm/test/Analysis/CostModel/RISCV/splice.ll b/llvm/test/Analysis/CostModel/RISCV/splice.ll
index c70c879dba5a..9acccef9c4f6 100644
--- a/llvm/test/Analysis/CostModel/RISCV/splice.ll
+++ b/llvm/test/Analysis/CostModel/RISCV/splice.ll
@@ -4,220 +4,220 @@
 
 define void @vector_splice() {
 ; CHECK-LABEL: 'vector_splice'
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv1i8 = call <vscale x 1 x i8> @llvm.experimental.vector.splice.nxv1i8(<vscale x 1 x i8> zeroinitializer, <vscale x 1 x i8> zeroinitializer, i32 -1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv2i8 = call <vscale x 2 x i8> @llvm.experimental.vector.splice.nxv2i8(<vscale x 2 x i8> zeroinitializer, <vscale x 2 x i8> zeroinitializer, i32 -1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv4i8 = call <vscale x 4 x i8> @llvm.experimental.vector.splice.nxv4i8(<vscale x 4 x i8> zeroinitializer, <vscale x 4 x i8> zeroinitializer, i32 -1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv8i8 = call <vscale x 8 x i8> @llvm.experimental.vector.splice.nxv8i8(<vscale x 8 x i8> zeroinitializer, <vscale x 8 x i8> zeroinitializer, i32 -1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %splice.nxv16i8 = call <vscale x 16 x i8> @llvm.experimental.vector.splice.nxv16i8(<vscale x 16 x i8> zeroinitializer, <vscale x 16 x i8> zeroinitializer, i32 -1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %splice.nxv32i8 = call <vscale x 32 x i8> @llvm.experimental.vector.splice.nxv32i8(<vscale x 32 x i8> zeroinitializer, <vscale x 32 x i8> zeroinitializer, i32 -1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %splice.nxv64i8 = call <vscale x 64 x i8> @llvm.experimental.vector.splice.nxv64i8(<vscale x 64 x i8> zeroinitializer, <vscale x 64 x i8> zeroinitializer, i32 -1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv1i16 = call <vscale x 1 x i16> @llvm.experimental.vector.splice.nxv1i16(<vscale x 1 x i16> zeroinitializer, <vscale x 1 x i16> zeroinitializer, i32 -1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv2i16 = call <vscale x 2 x i16> @llvm.experimental.vector.splice.nxv2i16(<vscale x 2 x i16> zeroinitializer, <vscale x 2 x i16> zeroinitializer, i32 -1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv4i16 = call <vscale x 4 x i16> @llvm.experimental.vector.splice.nxv4i16(<vscale x 4 x i16> zeroinitializer, <vscale x 4 x i16> zeroinitializer, i32 -1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %splice.nxv8i16 = call <vscale x 8 x i16> @llvm.experimental.vector.splice.nxv8i16(<vscale x 8 x i16> zeroinitializer, <vscale x 8 x i16> zeroinitializer, i32 -1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %splice.nxv16i16 = call <vscale x 16 x i16> @llvm.experimental.vector.splice.nxv16i16(<vscale x 16 x i16> zeroinitializer, <vscale x 16 x i16> zeroinitializer, i32 -1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %splice.nxv32i16 = call <vscale x 32 x i16> @llvm.experimental.vector.splice.nxv32i16(<vscale x 32 x i16> zeroinitializer, <vscale x 32 x i16> zeroinitializer, i32 -1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %splice.nxv64i16 = call <vscale x 64 x i16> @llvm.experimental.vector.splice.nxv64i16(<vscale x 64 x i16> zeroinitializer, <vscale x 64 x i16> zeroinitializer, i32 -1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv1i32 = call <vscale x 1 x i32> @llvm.experimental.vector.splice.nxv1i32(<vscale x 1 x i32> zeroinitializer, <vscale x 1 x i32> zeroinitializer, i32 -1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv2i32 = call <vscale x 2 x i32> @llvm.experimental.vector.splice.nxv2i32(<vscale x 2 x i32> zeroinitializer, <vscale x 2 x i32> zeroinitializer, i32 -1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %splice.nxv4i32 = call <vscale x 4 x i32> @llvm.experimental.vector.splice.nxv4i32(<vscale x 4 x i32> zeroinitializer, <vscale x 4 x i32> zeroinitializer, i32 -1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %splice.nxv8i32 = call <vscale x 8 x i32> @llvm.experimental.vector.splice.nxv8i32(<vscale x 8 x i32> zeroinitializer, <vscale x 8 x i32> zeroinitializer, i32 -1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %splice.nxv16i32 = call <vscale x 16 x i32> @llvm.experimental.vector.splice.nxv16i32(<vscale x 16 x i32> zeroinitializer, <vscale x 16 x i32> zeroinitializer, i32 -1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %splice.nxv32i32 = call <vscale x 32 x i32> @llvm.experimental.vector.splice.nxv32i32(<vscale x 32 x i32> zeroinitializer, <vscale x 32 x i32> zeroinitializer, i32 -1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %splice.nxv64i32 = call <vscale x 64 x i32> @llvm.experimental.vector.splice.nxv64i32(<vscale x 64 x i32> zeroinitializer, <vscale x 64 x i32> zeroinitializer, i32 -1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv1i64 = call <vscale x 1 x i64> @llvm.experimental.vector.splice.nxv1i64(<vscale x 1 x i64> zeroinitializer, <vscale x 1 x i64> zeroinitializer, i32 -1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %splice.nxv2i64 = call <vscale x 2 x i64> @llvm.experimental.vector.splice.nxv2i64(<vscale x 2 x i64> zeroinitializer, <vscale x 2 x i64> zeroinitializer, i32 -1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %splice.nxv4i64 = call <vscale x 4 x i64> @llvm.experimental.vector.splice.nxv4i64(<vscale x 4 x i64> zeroinitializer, <vscale x 4 x i64> zeroinitializer, i32 -1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %splice.nxv8i64 = call <vscale x 8 x i64> @llvm.experimental.vector.splice.nxv8i64(<vscale x 8 x i64> zeroinitializer, <vscale x 8 x i64> zeroinitializer, i32 -1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %splice.nxv16i64 = call <vscale x 16 x i64> @llvm.experimental.vector.splice.nxv16i64(<vscale x 16 x i64> zeroinitializer, <vscale x 16 x i64> zeroinitializer, i32 -1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %splice.nxv32i64 = call <vscale x 32 x i64> @llvm.experimental.vector.splice.nxv32i64(<vscale x 32 x i64> zeroinitializer, <vscale x 32 x i64> zeroinitializer, i32 -1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 128 for instruction: %splice.nxv64i64 = call <vscale x 64 x i64> @llvm.experimental.vector.splice.nxv64i64(<vscale x 64 x i64> zeroinitializer, <vscale x 64 x i64> zeroinitializer, i32 -1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv1f16 = call <vscale x 1 x half> @llvm.experimental.vector.splice.nxv1f16(<vscale x 1 x half> zeroinitializer, <vscale x 1 x half> zeroinitializer, i32 -1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv2f16 = call <vscale x 2 x half> @llvm.experimental.vector.splice.nxv2f16(<vscale x 2 x half> zeroinitializer, <vscale x 2 x half> zeroinitializer, i32 -1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv4f16 = call <vscale x 4 x half> @llvm.experimental.vector.splice.nxv4f16(<vscale x 4 x half> zeroinitializer, <vscale x 4 x half> zeroinitializer, i32 -1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %splice.nxv8f16 = call <vscale x 8 x half> @llvm.experimental.vector.splice.nxv8f16(<vscale x 8 x half> zeroinitializer, <vscale x 8 x half> zeroinitializer, i32 -1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %splice.nxv16f16 = call <vscale x 16 x half> @llvm.experimental.vector.splice.nxv16f16(<vscale x 16 x half> zeroinitializer, <vscale x 16 x half> zeroinitializer, i32 -1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %splice.nxv32f16 = call <vscale x 32 x half> @llvm.experimental.vector.splice.nxv32f16(<vscale x 32 x half> zeroinitializer, <vscale x 32 x half> zeroinitializer, i32 -1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %splice.nxv64f16 = call <vscale x 64 x half> @llvm.experimental.vector.splice.nxv64f16(<vscale x 64 x half> zeroinitializer, <vscale x 64 x half> zeroinitializer, i32 -1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv1f32 = call <vscale x 1 x float> @llvm.experimental.vector.splice.nxv1f32(<vscale x 1 x float> zeroinitializer, <vscale x 1 x float> zeroinitializer, i32 -1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv2f32 = call <vscale x 2 x float> @llvm.experimental.vector.splice.nxv2f32(<vscale x 2 x float> zeroinitializer, <vscale x 2 x float> zeroinitializer, i32 -1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %splice.nxv4f32 = call <vscale x 4 x float> @llvm.experimental.vector.splice.nxv4f32(<vscale x 4 x float> zeroinitializer, <vscale x 4 x float> zeroinitializer, i32 -1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %splice.nxv8f32 = call <vscale x 8 x float> @llvm.experimental.vector.splice.nxv8f32(<vscale x 8 x float> zeroinitializer, <vscale x 8 x float> zeroinitializer, i32 -1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %splice.nxv16f32 = call <vscale x 16 x float> @llvm.experimental.vector.splice.nxv16f32(<vscale x 16 x float> zeroinitializer, <vscale x 16 x float> zeroinitializer, i32 -1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %splice.nxv32f32 = call <vscale x 32 x float> @llvm.experimental.vector.splice.nxv32f32(<vscale x 32 x float> zeroinitializer, <vscale x 32 x float> zeroinitializer, i32 -1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %splice.nxv64f32 = call <vscale x 64 x float> @llvm.experimental.vector.splice.nxv64f32(<vscale x 64 x float> zeroinitializer, <vscale x 64 x float> zeroinitializer, i32 -1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv1f64 = call <vscale x 1 x double> @llvm.experimental.vector.splice.nxv1f64(<vscale x 1 x double> zeroinitializer, <vscale x 1 x double> zeroinitializer, i32 -1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %splice.nxv2f64 = call <vscale x 2 x double> @llvm.experimental.vector.splice.nxv2f64(<vscale x 2 x double> zeroinitializer, <vscale x 2 x double> zeroinitializer, i32 -1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %splice.nxv4f64 = call <vscale x 4 x double> @llvm.experimental.vector.splice.nxv4f64(<vscale x 4 x double> zeroinitializer, <vscale x 4 x double> zeroinitializer, i32 -1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %splice.nxv8f64 = call <vscale x 8 x double> @llvm.experimental.vector.splice.nxv8f64(<vscale x 8 x double> zeroinitializer, <vscale x 8 x double> zeroinitializer, i32 -1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %splice.nxv16f64 = call <vscale x 16 x double> @llvm.experimental.vector.splice.nxv16f64(<vscale x 16 x double> zeroinitializer, <vscale x 16 x double> zeroinitializer, i32 -1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %splice.nxv32f64 = call <vscale x 32 x double> @llvm.experimental.vector.splice.nxv32f64(<vscale x 32 x double> zeroinitializer, <vscale x 32 x double> zeroinitializer, i32 -1)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 128 for instruction: %splice.nxv64f64 = call <vscale x 64 x double> @llvm.experimental.vector.splice.nxv64f64(<vscale x 64 x double> zeroinitializer, <vscale x 64 x double> zeroinitializer, i32 -1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv1i8 = call <vscale x 1 x i8> @llvm.vector.splice.nxv1i8(<vscale x 1 x i8> zeroinitializer, <vscale x 1 x i8> zeroinitializer, i32 -1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv2i8 = call <vscale x 2 x i8> @llvm.vector.splice.nxv2i8(<vscale x 2 x i8> zeroinitializer, <vscale x 2 x i8> zeroinitializer, i32 -1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv4i8 = call <vscale x 4 x i8> @llvm.vector.splice.nxv4i8(<vscale x 4 x i8> zeroinitializer, <vscale x 4 x i8> zeroinitializer, i32 -1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv8i8 = call <vscale x 8 x i8> @llvm.vector.splice.nxv8i8(<vscale x 8 x i8> zeroinitializer, <vscale x 8 x i8> zeroinitializer, i32 -1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %splice.nxv16i8 = call <vscale x 16 x i8> @llvm.vector.splice.nxv16i8(<vscale x 16 x i8> zeroinitializer, <vscale x 16 x i8> zeroinitializer, i32 -1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %splice.nxv32i8 = call <vscale x 32 x i8> @llvm.vector.splice.nxv32i8(<vscale x 32 x i8> zeroinitializer, <vscale x 32 x i8> zeroinitializer, i32 -1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %splice.nxv64i8 = call <vscale x 64 x i8> @llvm.vector.splice.nxv64i8(<vscale x 64 x i8> zeroinitializer, <vscale x 64 x i8> zeroinitializer, i32 -1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv1i16 = call <vscale x 1 x i16> @llvm.vector.splice.nxv1i16(<vscale x 1 x i16> zeroinitializer, <vscale x 1 x i16> zeroinitializer, i32 -1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv2i16 = call <vscale x 2 x i16> @llvm.vector.splice.nxv2i16(<vscale x 2 x i16> zeroinitializer, <vscale x 2 x i16> zeroinitializer, i32 -1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv4i16 = call <vscale x 4 x i16> @llvm.vector.splice.nxv4i16(<vscale x 4 x i16> zeroinitializer, <vscale x 4 x i16> zeroinitializer, i32 -1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %splice.nxv8i16 = call <vscale x 8 x i16> @llvm.vector.splice.nxv8i16(<vscale x 8 x i16> zeroinitializer, <vscale x 8 x i16> zeroinitializer, i32 -1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %splice.nxv16i16 = call <vscale x 16 x i16> @llvm.vector.splice.nxv16i16(<vscale x 16 x i16> zeroinitializer, <vscale x 16 x i16> zeroinitializer, i32 -1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %splice.nxv32i16 = call <vscale x 32 x i16> @llvm.vector.splice.nxv32i16(<vscale x 32 x i16> zeroinitializer, <vscale x 32 x i16> zeroinitializer, i32 -1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %splice.nxv64i16 = call <vscale x 64 x i16> @llvm.vector.splice.nxv64i16(<vscale x 64 x i16> zeroinitializer, <vscale x 64 x i16> zeroinitializer, i32 -1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv1i32 = call <vscale x 1 x i32> @llvm.vector.splice.nxv1i32(<vscale x 1 x i32> zeroinitializer, <vscale x 1 x i32> zeroinitializer, i32 -1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv2i32 = call <vscale x 2 x i32> @llvm.vector.splice.nxv2i32(<vscale x 2 x i32> zeroinitializer, <vscale x 2 x i32> zeroinitializer, i32 -1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %splice.nxv4i32 = call <vscale x 4 x i32> @llvm.vector.splice.nxv4i32(<vscale x 4 x i32> zeroinitializer, <vscale x 4 x i32> zeroinitializer, i32 -1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %splice.nxv8i32 = call <vscale x 8 x i32> @llvm.vector.splice.nxv8i32(<vscale x 8 x i32> zeroinitializer, <vscale x 8 x i32> zeroinitializer, i32 -1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %splice.nxv16i32 = call <vscale x 16 x i32> @llvm.vector.splice.nxv16i32(<vscale x 16 x i32> zeroinitializer, <vscale x 16 x i32> zeroinitializer, i32 -1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %splice.nxv32i32 = call <vscale x 32 x i32> @llvm.vector.splice.nxv32i32(<vscale x 32 x i32> zeroinitializer, <vscale x 32 x i32> zeroinitializer, i32 -1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %splice.nxv64i32 = call <vscale x 64 x i32> @llvm.vector.splice.nxv64i32(<vscale x 64 x i32> zeroinitializer, <vscale x 64 x i32> zeroinitializer, i32 -1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv1i64 = call <vscale x 1 x i64> @llvm.vector.splice.nxv1i64(<vscale x 1 x i64> zeroinitializer, <vscale x 1 x i64> zeroinitializer, i32 -1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %splice.nxv2i64 = call <vscale x 2 x i64> @llvm.vector.splice.nxv2i64(<vscale x 2 x i64> zeroinitializer, <vscale x 2 x i64> zeroinitializer, i32 -1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %splice.nxv4i64 = call <vscale x 4 x i64> @llvm.vector.splice.nxv4i64(<vscale x 4 x i64> zeroinitializer, <vscale x 4 x i64> zeroinitializer, i32 -1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %splice.nxv8i64 = call <vscale x 8 x i64> @llvm.vector.splice.nxv8i64(<vscale x 8 x i64> zeroinitializer, <vscale x 8 x i64> zeroinitializer, i32 -1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %splice.nxv16i64 = call <vscale x 16 x i64> @llvm.vector.splice.nxv16i64(<vscale x 16 x i64> zeroinitializer, <vscale x 16 x i64> zeroinitializer, i32 -1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %splice.nxv32i64 = call <vscale x 32 x i64> @llvm.vector.splice.nxv32i64(<vscale x 32 x i64> zeroinitializer, <vscale x 32 x i64> zeroinitializer, i32 -1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 128 for instruction: %splice.nxv64i64 = call <vscale x 64 x i64> @llvm.vector.splice.nxv64i64(<vscale x 64 x i64> zeroinitializer, <vscale x 64 x i64> zeroinitializer, i32 -1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv1f16 = call <vscale x 1 x half> @llvm.vector.splice.nxv1f16(<vscale x 1 x half> zeroinitializer, <vscale x 1 x half> zeroinitializer, i32 -1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv2f16 = call <vscale x 2 x half> @llvm.vector.splice.nxv2f16(<vscale x 2 x half> zeroinitializer, <vscale x 2 x half> zeroinitializer, i32 -1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv4f16 = call <vscale x 4 x half> @llvm.vector.splice.nxv4f16(<vscale x 4 x half> zeroinitializer, <vscale x 4 x half> zeroinitializer, i32 -1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %splice.nxv8f16 = call <vscale x 8 x half> @llvm.vector.splice.nxv8f16(<vscale x 8 x half> zeroinitializer, <vscale x 8 x half> zeroinitializer, i32 -1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %splice.nxv16f16 = call <vscale x 16 x half> @llvm.vector.splice.nxv16f16(<vscale x 16 x half> zeroinitializer, <vscale x 16 x half> zeroinitializer, i32 -1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %splice.nxv32f16 = call <vscale x 32 x half> @llvm.vector.splice.nxv32f16(<vscale x 32 x half> zeroinitializer, <vscale x 32 x half> zeroinitializer, i32 -1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %splice.nxv64f16 = call <vscale x 64 x half> @llvm.vector.splice.nxv64f16(<vscale x 64 x half> zeroinitializer, <vscale x 64 x half> zeroinitializer, i32 -1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv1f32 = call <vscale x 1 x float> @llvm.vector.splice.nxv1f32(<vscale x 1 x float> zeroinitializer, <vscale x 1 x float> zeroinitializer, i32 -1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv2f32 = call <vscale x 2 x float> @llvm.vector.splice.nxv2f32(<vscale x 2 x float> zeroinitializer, <vscale x 2 x float> zeroinitializer, i32 -1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %splice.nxv4f32 = call <vscale x 4 x float> @llvm.vector.splice.nxv4f32(<vscale x 4 x float> zeroinitializer, <vscale x 4 x float> zeroinitializer, i32 -1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %splice.nxv8f32 = call <vscale x 8 x float> @llvm.vector.splice.nxv8f32(<vscale x 8 x float> zeroinitializer, <vscale x 8 x float> zeroinitializer, i32 -1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %splice.nxv16f32 = call <vscale x 16 x float> @llvm.vector.splice.nxv16f32(<vscale x 16 x float> zeroinitializer, <vscale x 16 x float> zeroinitializer, i32 -1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %splice.nxv32f32 = call <vscale x 32 x float> @llvm.vector.splice.nxv32f32(<vscale x 32 x float> zeroinitializer, <vscale x 32 x float> zeroinitializer, i32 -1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %splice.nxv64f32 = call <vscale x 64 x float> @llvm.vector.splice.nxv64f32(<vscale x 64 x float> zeroinitializer, <vscale x 64 x float> zeroinitializer, i32 -1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv1f64 = call <vscale x 1 x double> @llvm.vector.splice.nxv1f64(<vscale x 1 x double> zeroinitializer, <vscale x 1 x double> zeroinitializer, i32 -1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %splice.nxv2f64 = call <vscale x 2 x double> @llvm.vector.splice.nxv2f64(<vscale x 2 x double> zeroinitializer, <vscale x 2 x double> zeroinitializer, i32 -1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %splice.nxv4f64 = call <vscale x 4 x double> @llvm.vector.splice.nxv4f64(<vscale x 4 x double> zeroinitializer, <vscale x 4 x double> zeroinitializer, i32 -1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %splice.nxv8f64 = call <vscale x 8 x double> @llvm.vector.splice.nxv8f64(<vscale x 8 x double> zeroinitializer, <vscale x 8 x double> zeroinitializer, i32 -1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 32 for instruction: %splice.nxv16f64 = call <vscale x 16 x double> @llvm.vector.splice.nxv16f64(<vscale x 16 x double> zeroinitializer, <vscale x 16 x double> zeroinitializer, i32 -1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 64 for instruction: %splice.nxv32f64 = call <vscale x 32 x double> @llvm.vector.splice.nxv32f64(<vscale x 32 x double> zeroinitializer, <vscale x 32 x double> zeroinitializer, i32 -1)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 128 for instruction: %splice.nxv64f64 = call <vscale x 64 x double> @llvm.vector.splice.nxv64f64(<vscale x 64 x double> zeroinitializer, <vscale x 64 x double> zeroinitializer, i32 -1)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
 ;
 ; SIZE-LABEL: 'vector_splice'
-; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv1i8 = call <vscale x 1 x i8> @llvm.experimental.vector.splice.nxv1i8(<vscale x 1 x i8> zeroinitializer, <vscale x 1 x i8> zeroinitializer, i32 -1)
-; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv2i8 = call <vscale x 2 x i8> @llvm.experimental.vector.splice.nxv2i8(<vscale x 2 x i8> zeroinitializer, <vscale x 2 x i8> zeroinitializer, i32 -1)
-; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv4i8 = call <vscale x 4 x i8> @llvm.experimental.vector.splice.nxv4i8(<vscale x 4 x i8> zeroinitializer, <vscale x 4 x i8> zeroinitializer, i32 -1)
-; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv8i8 = call <vscale x 8 x i8> @llvm.experimental.vector.splice.nxv8i8(<vscale x 8 x i8> zeroinitializer, <vscale x 8 x i8> zeroinitializer, i32 -1)
-; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv16i8 = call <vscale x 16 x i8> @llvm.experimental.vector.splice.nxv16i8(<vscale x 16 x i8> zeroinitializer, <vscale x 16 x i8> zeroinitializer, i32 -1)
-; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv32i8 = call <vscale x 32 x i8> @llvm.experimental.vector.splice.nxv32i8(<vscale x 32 x i8> zeroinitializer, <vscale x 32 x i8> zeroinitializer, i32 -1)
-; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv64i8 = call <vscale x 64 x i8> @llvm.experimental.vector.splice.nxv64i8(<vscale x 64 x i8> zeroinitializer, <vscale x 64 x i8> zeroinitializer, i32 -1)
-; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv1i16 = call <vscale x 1 x i16> @llvm.experimental.vector.splice.nxv1i16(<vscale x 1 x i16> zeroinitializer, <vscale x 1 x i16> zeroinitializer, i32 -1)
-; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv2i16 = call <vscale x 2 x i16> @llvm.experimental.vector.splice.nxv2i16(<vscale x 2 x i16> zeroinitializer, <vscale x 2 x i16> zeroinitializer, i32 -1)
-; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv4i16 = call <vscale x 4 x i16> @llvm.experimental.vector.splice.nxv4i16(<vscale x 4 x i16> zeroinitializer, <vscale x 4 x i16> zeroinitializer, i32 -1)
-; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv8i16 = call <vscale x 8 x i16> @llvm.experimental.vector.splice.nxv8i16(<vscale x 8 x i16> zeroinitializer, <vscale x 8 x i16> zeroinitializer, i32 -1)
-; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv16i16 = call <vscale x 16 x i16> @llvm.experimental.vector.splice.nxv16i16(<vscale x 16 x i16> zeroinitializer, <vscale x 16 x i16> zeroinitializer, i32 -1)
-; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv32i16 = call <vscale x 32 x i16> @llvm.experimental.vector.splice.nxv32i16(<vscale x 32 x i16> zeroinitializer, <vscale x 32 x i16> zeroinitializer, i32 -1)
-; SIZE-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %splice.nxv64i16 = call <vscale x 64 x i16> @llvm.experimental.vector.splice.nxv64i16(<vscale x 64 x i16> zeroinitializer, <vscale x 64 x i16> zeroinitializer, i32 -1)
-; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv1i32 = call <vscale x 1 x i32> @llvm.experimental.vector.splice.nxv1i32(<vscale x 1 x i32> zeroinitializer, <vscale x 1 x i32> zeroinitializer, i32 -1)
-; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv2i32 = call <vscale x 2 x i32> @llvm.experimental.vector.splice.nxv2i32(<vscale x 2 x i32> zeroinitializer, <vscale x 2 x i32> zeroinitializer, i32 -1)
-; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv4i32 = call <vscale x 4 x i32> @llvm.experimental.vector.splice.nxv4i32(<vscale x 4 x i32> zeroinitializer, <vscale x 4 x i32> zeroinitializer, i32 -1)
-; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv8i32 = call <vscale x 8 x i32> @llvm.experimental.vector.splice.nxv8i32(<vscale x 8 x i32> zeroinitializer, <vscale x 8 x i32> zeroinitializer, i32 -1)
-; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv16i32 = call <vscale x 16 x i32> @llvm.experimental.vector.splice.nxv16i32(<vscale x 16 x i32> zeroinitializer, <vscale x 16 x i32> zeroinitializer, i32 -1)
-; SIZE-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %splice.nxv32i32 = call <vscale x 32 x i32> @llvm.experimental.vector.splice.nxv32i32(<vscale x 32 x i32> zeroinitializer, <vscale x 32 x i32> zeroinitializer, i32 -1)
-; SIZE-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %splice.nxv64i32 = call <vscale x 64 x i32> @llvm.experimental.vector.splice.nxv64i32(<vscale x 64 x i32> zeroinitializer, <vscale x 64 x i32> zeroinitializer, i32 -1)
-; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv1i64 = call <vscale x 1 x i64> @llvm.experimental.vector.splice.nxv1i64(<vscale x 1 x i64> zeroinitializer, <vscale x 1 x i64> zeroinitializer, i32 -1)
-; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv2i64 = call <vscale x 2 x i64> @llvm.experimental.vector.splice.nxv2i64(<vscale x 2 x i64> zeroinitializer, <vscale x 2 x i64> zeroinitializer, i32 -1)
-; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv4i64 = call <vscale x 4 x i64> @llvm.experimental.vector.splice.nxv4i64(<vscale x 4 x i64> zeroinitializer, <vscale x 4 x i64> zeroinitializer, i32 -1)
-; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv8i64 = call <vscale x 8 x i64> @llvm.experimental.vector.splice.nxv8i64(<vscale x 8 x i64> zeroinitializer, <vscale x 8 x i64> zeroinitializer, i32 -1)
-; SIZE-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %splice.nxv16i64 = call <vscale x 16 x i64> @llvm.experimental.vector.splice.nxv16i64(<vscale x 16 x i64> zeroinitializer, <vscale x 16 x i64> zeroinitializer, i32 -1)
-; SIZE-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %splice.nxv32i64 = call <vscale x 32 x i64> @llvm.experimental.vector.splice.nxv32i64(<vscale x 32 x i64> zeroinitializer, <vscale x 32 x i64> zeroinitializer, i32 -1)
-; SIZE-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %splice.nxv64i64 = call <vscale x 64 x i64> @llvm.experimental.vector.splice.nxv64i64(<vscale x 64 x i64> zeroinitializer, <vscale x 64 x i64> zeroinitializer, i32 -1)
-; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv1f16 = call <vscale x 1 x half> @llvm.experimental.vector.splice.nxv1f16(<vscale x 1 x half> zeroinitializer, <vscale x 1 x half> zeroinitializer, i32 -1)
-; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv2f16 = call <vscale x 2 x half> @llvm.experimental.vector.splice.nxv2f16(<vscale x 2 x half> zeroinitializer, <vscale x 2 x half> zeroinitializer, i32 -1)
-; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv4f16 = call <vscale x 4 x half> @llvm.experimental.vector.splice.nxv4f16(<vscale x 4 x half> zeroinitializer, <vscale x 4 x half> zeroinitializer, i32 -1)
-; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv8f16 = call <vscale x 8 x half> @llvm.experimental.vector.splice.nxv8f16(<vscale x 8 x half> zeroinitializer, <vscale x 8 x half> zeroinitializer, i32 -1)
-; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv16f16 = call <vscale x 16 x half> @llvm.experimental.vector.splice.nxv16f16(<vscale x 16 x half> zeroinitializer, <vscale x 16 x half> zeroinitializer, i32 -1)
-; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv32f16 = call <vscale x 32 x half> @llvm.experimental.vector.splice.nxv32f16(<vscale x 32 x half> zeroinitializer, <vscale x 32 x half> zeroinitializer, i32 -1)
-; SIZE-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %splice.nxv64f16 = call <vscale x 64 x half> @llvm.experimental.vector.splice.nxv64f16(<vscale x 64 x half> zeroinitializer, <vscale x 64 x half> zeroinitializer, i32 -1)
-; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv1f32 = call <vscale x 1 x float> @llvm.experimental.vector.splice.nxv1f32(<vscale x 1 x float> zeroinitializer, <vscale x 1 x float> zeroinitializer, i32 -1)
-; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv2f32 = call <vscale x 2 x float> @llvm.experimental.vector.splice.nxv2f32(<vscale x 2 x float> zeroinitializer, <vscale x 2 x float> zeroinitializer, i32 -1)
-; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv4f32 = call <vscale x 4 x float> @llvm.experimental.vector.splice.nxv4f32(<vscale x 4 x float> zeroinitializer, <vscale x 4 x float> zeroinitializer, i32 -1)
-; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv8f32 = call <vscale x 8 x float> @llvm.experimental.vector.splice.nxv8f32(<vscale x 8 x float> zeroinitializer, <vscale x 8 x float> zeroinitializer, i32 -1)
-; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv16f32 = call <vscale x 16 x float> @llvm.experimental.vector.splice.nxv16f32(<vscale x 16 x float> zeroinitializer, <vscale x 16 x float> zeroinitializer, i32 -1)
-; SIZE-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %splice.nxv32f32 = call <vscale x 32 x float> @llvm.experimental.vector.splice.nxv32f32(<vscale x 32 x float> zeroinitializer, <vscale x 32 x float> zeroinitializer, i32 -1)
-; SIZE-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %splice.nxv64f32 = call <vscale x 64 x float> @llvm.experimental.vector.splice.nxv64f32(<vscale x 64 x float> zeroinitializer, <vscale x 64 x float> zeroinitializer, i32 -1)
-; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv1f64 = call <vscale x 1 x double> @llvm.experimental.vector.splice.nxv1f64(<vscale x 1 x double> zeroinitializer, <vscale x 1 x double> zeroinitializer, i32 -1)
-; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv2f64 = call <vscale x 2 x double> @llvm.experimental.vector.splice.nxv2f64(<vscale x 2 x double> zeroinitializer, <vscale x 2 x double> zeroinitializer, i32 -1)
-; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv4f64 = call <vscale x 4 x double> @llvm.experimental.vector.splice.nxv4f64(<vscale x 4 x double> zeroinitializer, <vscale x 4 x double> zeroinitializer, i32 -1)
-; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv8f64 = call <vscale x 8 x double> @llvm.experimental.vector.splice.nxv8f64(<vscale x 8 x double> zeroinitializer, <vscale x 8 x double> zeroinitializer, i32 -1)
-; SIZE-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %splice.nxv16f64 = call <vscale x 16 x double> @llvm.experimental.vector.splice.nxv16f64(<vscale x 16 x double> zeroinitializer, <vscale x 16 x double> zeroinitializer, i32 -1)
-; SIZE-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %splice.nxv32f64 = call <vscale x 32 x double> @llvm.experimental.vector.splice.nxv32f64(<vscale x 32 x double> zeroinitializer, <vscale x 32 x double> zeroinitializer, i32 -1)
-; SIZE-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %splice.nxv64f64 = call <vscale x 64 x double> @llvm.experimental.vector.splice.nxv64f64(<vscale x 64 x double> zeroinitializer, <vscale x 64 x double> zeroinitializer, i32 -1)
+; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv1i8 = call <vscale x 1 x i8> @llvm.vector.splice.nxv1i8(<vscale x 1 x i8> zeroinitializer, <vscale x 1 x i8> zeroinitializer, i32 -1)
+; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv2i8 = call <vscale x 2 x i8> @llvm.vector.splice.nxv2i8(<vscale x 2 x i8> zeroinitializer, <vscale x 2 x i8> zeroinitializer, i32 -1)
+; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv4i8 = call <vscale x 4 x i8> @llvm.vector.splice.nxv4i8(<vscale x 4 x i8> zeroinitializer, <vscale x 4 x i8> zeroinitializer, i32 -1)
+; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv8i8 = call <vscale x 8 x i8> @llvm.vector.splice.nxv8i8(<vscale x 8 x i8> zeroinitializer, <vscale x 8 x i8> zeroinitializer, i32 -1)
+; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv16i8 = call <vscale x 16 x i8> @llvm.vector.splice.nxv16i8(<vscale x 16 x i8> zeroinitializer, <vscale x 16 x i8> zeroinitializer, i32 -1)
+; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv32i8 = call <vscale x 32 x i8> @llvm.vector.splice.nxv32i8(<vscale x 32 x i8> zeroinitializer, <vscale x 32 x i8> zeroinitializer, i32 -1)
+; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv64i8 = call <vscale x 64 x i8> @llvm.vector.splice.nxv64i8(<vscale x 64 x i8> zeroinitializer, <vscale x 64 x i8> zeroinitializer, i32 -1)
+; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv1i16 = call <vscale x 1 x i16> @llvm.vector.splice.nxv1i16(<vscale x 1 x i16> zeroinitializer, <vscale x 1 x i16> zeroinitializer, i32 -1)
+; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv2i16 = call <vscale x 2 x i16> @llvm.vector.splice.nxv2i16(<vscale x 2 x i16> zeroinitializer, <vscale x 2 x i16> zeroinitializer, i32 -1)
+; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv4i16 = call <vscale x 4 x i16> @llvm.vector.splice.nxv4i16(<vscale x 4 x i16> zeroinitializer, <vscale x 4 x i16> zeroinitializer, i32 -1)
+; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv8i16 = call <vscale x 8 x i16> @llvm.vector.splice.nxv8i16(<vscale x 8 x i16> zeroinitializer, <vscale x 8 x i16> zeroinitializer, i32 -1)
+; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv16i16 = call <vscale x 16 x i16> @llvm.vector.splice.nxv16i16(<vscale x 16 x i16> zeroinitializer, <vscale x 16 x i16> zeroinitializer, i32 -1)
+; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv32i16 = call <vscale x 32 x i16> @llvm.vector.splice.nxv32i16(<vscale x 32 x i16> zeroinitializer, <vscale x 32 x i16> zeroinitializer, i32 -1)
+; SIZE-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %splice.nxv64i16 = call <vscale x 64 x i16> @llvm.vector.splice.nxv64i16(<vscale x 64 x i16> zeroinitializer, <vscale x 64 x i16> zeroinitializer, i32 -1)
+; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv1i32 = call <vscale x 1 x i32> @llvm.vector.splice.nxv1i32(<vscale x 1 x i32> zeroinitializer, <vscale x 1 x i32> zeroinitializer, i32 -1)
+; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv2i32 = call <vscale x 2 x i32> @llvm.vector.splice.nxv2i32(<vscale x 2 x i32> zeroinitializer, <vscale x 2 x i32> zeroinitializer, i32 -1)
+; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv4i32 = call <vscale x 4 x i32> @llvm.vector.splice.nxv4i32(<vscale x 4 x i32> zeroinitializer, <vscale x 4 x i32> zeroinitializer, i32 -1)
+; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv8i32 = call <vscale x 8 x i32> @llvm.vector.splice.nxv8i32(<vscale x 8 x i32> zeroinitializer, <vscale x 8 x i32> zeroinitializer, i32 -1)
+; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv16i32 = call <vscale x 16 x i32> @llvm.vector.splice.nxv16i32(<vscale x 16 x i32> zeroinitializer, <vscale x 16 x i32> zeroinitializer, i32 -1)
+; SIZE-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %splice.nxv32i32 = call <vscale x 32 x i32> @llvm.vector.splice.nxv32i32(<vscale x 32 x i32> zeroinitializer, <vscale x 32 x i32> zeroinitializer, i32 -1)
+; SIZE-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %splice.nxv64i32 = call <vscale x 64 x i32> @llvm.vector.splice.nxv64i32(<vscale x 64 x i32> zeroinitializer, <vscale x 64 x i32> zeroinitializer, i32 -1)
+; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv1i64 = call <vscale x 1 x i64> @llvm.vector.splice.nxv1i64(<vscale x 1 x i64> zeroinitializer, <vscale x 1 x i64> zeroinitializer, i32 -1)
+; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv2i64 = call <vscale x 2 x i64> @llvm.vector.splice.nxv2i64(<vscale x 2 x i64> zeroinitializer, <vscale x 2 x i64> zeroinitializer, i32 -1)
+; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv4i64 = call <vscale x 4 x i64> @llvm.vector.splice.nxv4i64(<vscale x 4 x i64> zeroinitializer, <vscale x 4 x i64> zeroinitializer, i32 -1)
+; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv8i64 = call <vscale x 8 x i64> @llvm.vector.splice.nxv8i64(<vscale x 8 x i64> zeroinitializer, <vscale x 8 x i64> zeroinitializer, i32 -1)
+; SIZE-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %splice.nxv16i64 = call <vscale x 16 x i64> @llvm.vector.splice.nxv16i64(<vscale x 16 x i64> zeroinitializer, <vscale x 16 x i64> zeroinitializer, i32 -1)
+; SIZE-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %splice.nxv32i64 = call <vscale x 32 x i64> @llvm.vector.splice.nxv32i64(<vscale x 32 x i64> zeroinitializer, <vscale x 32 x i64> zeroinitializer, i32 -1)
+; SIZE-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %splice.nxv64i64 = call <vscale x 64 x i64> @llvm.vector.splice.nxv64i64(<vscale x 64 x i64> zeroinitializer, <vscale x 64 x i64> zeroinitializer, i32 -1)
+; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv1f16 = call <vscale x 1 x half> @llvm.vector.splice.nxv1f16(<vscale x 1 x half> zeroinitializer, <vscale x 1 x half> zeroinitializer, i32 -1)
+; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv2f16 = call <vscale x 2 x half> @llvm.vector.splice.nxv2f16(<vscale x 2 x half> zeroinitializer, <vscale x 2 x half> zeroinitializer, i32 -1)
+; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv4f16 = call <vscale x 4 x half> @llvm.vector.splice.nxv4f16(<vscale x 4 x half> zeroinitializer, <vscale x 4 x half> zeroinitializer, i32 -1)
+; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv8f16 = call <vscale x 8 x half> @llvm.vector.splice.nxv8f16(<vscale x 8 x half> zeroinitializer, <vscale x 8 x half> zeroinitializer, i32 -1)
+; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv16f16 = call <vscale x 16 x half> @llvm.vector.splice.nxv16f16(<vscale x 16 x half> zeroinitializer, <vscale x 16 x half> zeroinitializer, i32 -1)
+; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv32f16 = call <vscale x 32 x half> @llvm.vector.splice.nxv32f16(<vscale x 32 x half> zeroinitializer, <vscale x 32 x half> zeroinitializer, i32 -1)
+; SIZE-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %splice.nxv64f16 = call <vscale x 64 x half> @llvm.vector.splice.nxv64f16(<vscale x 64 x half> zeroinitializer, <vscale x 64 x half> zeroinitializer, i32 -1)
+; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv1f32 = call <vscale x 1 x float> @llvm.vector.splice.nxv1f32(<vscale x 1 x float> zeroinitializer, <vscale x 1 x float> zeroinitializer, i32 -1)
+; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv2f32 = call <vscale x 2 x float> @llvm.vector.splice.nxv2f32(<vscale x 2 x float> zeroinitializer, <vscale x 2 x float> zeroinitializer, i32 -1)
+; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv4f32 = call <vscale x 4 x float> @llvm.vector.splice.nxv4f32(<vscale x 4 x float> zeroinitializer, <vscale x 4 x float> zeroinitializer, i32 -1)
+; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv8f32 = call <vscale x 8 x float> @llvm.vector.splice.nxv8f32(<vscale x 8 x float> zeroinitializer, <vscale x 8 x float> zeroinitializer, i32 -1)
+; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv16f32 = call <vscale x 16 x float> @llvm.vector.splice.nxv16f32(<vscale x 16 x float> zeroinitializer, <vscale x 16 x float> zeroinitializer, i32 -1)
+; SIZE-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %splice.nxv32f32 = call <vscale x 32 x float> @llvm.vector.splice.nxv32f32(<vscale x 32 x float> zeroinitializer, <vscale x 32 x float> zeroinitializer, i32 -1)
+; SIZE-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %splice.nxv64f32 = call <vscale x 64 x float> @llvm.vector.splice.nxv64f32(<vscale x 64 x float> zeroinitializer, <vscale x 64 x float> zeroinitializer, i32 -1)
+; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv1f64 = call <vscale x 1 x double> @llvm.vector.splice.nxv1f64(<vscale x 1 x double> zeroinitializer, <vscale x 1 x double> zeroinitializer, i32 -1)
+; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv2f64 = call <vscale x 2 x double> @llvm.vector.splice.nxv2f64(<vscale x 2 x double> zeroinitializer, <vscale x 2 x double> zeroinitializer, i32 -1)
+; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv4f64 = call <vscale x 4 x double> @llvm.vector.splice.nxv4f64(<vscale x 4 x double> zeroinitializer, <vscale x 4 x double> zeroinitializer, i32 -1)
+; SIZE-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice.nxv8f64 = call <vscale x 8 x double> @llvm.vector.splice.nxv8f64(<vscale x 8 x double> zeroinitializer, <vscale x 8 x double> zeroinitializer, i32 -1)
+; SIZE-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %splice.nxv16f64 = call <vscale x 16 x double> @llvm.vector.splice.nxv16f64(<vscale x 16 x double> zeroinitializer, <vscale x 16 x double> zeroinitializer, i32 -1)
+; SIZE-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %splice.nxv32f64 = call <vscale x 32 x double> @llvm.vector.splice.nxv32f64(<vscale x 32 x double> zeroinitializer, <vscale x 32 x double> zeroinitializer, i32 -1)
+; SIZE-NEXT:  Cost Model: Found an estimated cost of 16 for instruction: %splice.nxv64f64 = call <vscale x 64 x double> @llvm.vector.splice.nxv64f64(<vscale x 64 x double> zeroinitializer, <vscale x 64 x double> zeroinitializer, i32 -1)
 ; SIZE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: ret void
 ;
-  %splice.nxv1i8 = call <vscale x 1 x i8> @llvm.experimental.vector.splice.nxv1i8(<vscale x 1 x i8> zeroinitializer, <vscale x 1 x i8> zeroinitializer, i32 -1)
-  %splice.nxv2i8 = call <vscale x 2 x i8> @llvm.experimental.vector.splice.nxv2i8(<vscale x 2 x i8> zeroinitializer, <vscale x 2 x i8> zeroinitializer, i32 -1)
-  %splice.nxv4i8 = call <vscale x 4 x i8> @llvm.experimental.vector.splice.nxv4i8(<vscale x 4 x i8> zeroinitializer, <vscale x 4 x i8> zeroinitializer, i32 -1)
-  %splice.nxv8i8 = call <vscale x 8 x i8> @llvm.experimental.vector.splice.nxv8i8(<vscale x 8 x i8> zeroinitializer, <vscale x 8 x i8> zeroinitializer, i32 -1)
-  %splice.nxv16i8 = call <vscale x 16 x i8> @llvm.experimental.vector.splice.nxv16i8(<vscale x 16 x i8> zeroinitializer, <vscale x 16 x i8> zeroinitializer, i32 -1)
-  %splice.nxv32i8 = call <vscale x 32 x i8> @llvm.experimental.vector.splice.nxv32i8(<vscale x 32 x i8> zeroinitializer, <vscale x 32 x i8> zeroinitializer, i32 -1)
-  %splice.nxv64i8 = call <vscale x 64 x i8> @llvm.experimental.vector.splice.nxv64i8(<vscale x 64 x i8> zeroinitializer, <vscale x 64 x i8> zeroinitializer, i32 -1)
+  %splice.nxv1i8 = call <vscale x 1 x i8> @llvm.vector.splice.nxv1i8(<vscale x 1 x i8> zeroinitializer, <vscale x 1 x i8> zeroinitializer, i32 -1)
+  %splice.nxv2i8 = call <vscale x 2 x i8> @llvm.vector.splice.nxv2i8(<vscale x 2 x i8> zeroinitializer, <vscale x 2 x i8> zeroinitializer, i32 -1)
+  %splice.nxv4i8 = call <vscale x 4 x i8> @llvm.vector.splice.nxv4i8(<vscale x 4 x i8> zeroinitializer, <vscale x 4 x i8> zeroinitializer, i32 -1)
+  %splice.nxv8i8 = call <vscale x 8 x i8> @llvm.vector.splice.nxv8i8(<vscale x 8 x i8> zeroinitializer, <vscale x 8 x i8> zeroinitializer, i32 -1)
+  %splice.nxv16i8 = call <vscale x 16 x i8> @llvm.vector.splice.nxv16i8(<vscale x 16 x i8> zeroinitializer, <vscale x 16 x i8> zeroinitializer, i32 -1)
+  %splice.nxv32i8 = call <vscale x 32 x i8> @llvm.vector.splice.nxv32i8(<vscale x 32 x i8> zeroinitializer, <vscale x 32 x i8> zeroinitializer, i32 -1)
+  %splice.nxv64i8 = call <vscale x 64 x i8> @llvm.vector.splice.nxv64i8(<vscale x 64 x i8> zeroinitializer, <vscale x 64 x i8> zeroinitializer, i32 -1)
 
-  %splice.nxv1i16 = call <vscale x 1 x i16> @llvm.experimental.vector.splice.nxv1i16(<vscale x 1 x i16> zeroinitializer, <vscale x 1 x i16> zeroinitializer, i32 -1)
-  %splice.nxv2i16 = call <vscale x 2 x i16> @llvm.experimental.vector.splice.nxv2i16(<vscale x 2 x i16> zeroinitializer, <vscale x 2 x i16> zeroinitializer, i32 -1)
-  %splice.nxv4i16 = call <vscale x 4 x i16> @llvm.experimental.vector.splice.nxv4i16(<vscale x 4 x i16> zeroinitializer, <vscale x 4 x i16> zeroinitializer, i32 -1)
-  %splice.nxv8i16 = call <vscale x 8 x i16> @llvm.experimental.vector.splice.nxv8i16(<vscale x 8 x i16> zeroinitializer, <vscale x 8 x i16> zeroinitializer, i32 -1)
-  %splice.nxv16i16 = call <vscale x 16 x i16> @llvm.experimental.vector.splice.nxv16i16(<vscale x 16 x i16> zeroinitializer, <vscale x 16 x i16> zeroinitializer, i32 -1)
-  %splice.nxv32i16 = call <vscale x 32 x i16> @llvm.experimental.vector.splice.nxv32i16(<vscale x 32 x i16> zeroinitializer, <vscale x 32 x i16> zeroinitializer, i32 -1)
-  %splice.nxv64i16 = call <vscale x 64 x i16> @llvm.experimental.vector.splice.nxv64i16(<vscale x 64 x i16> zeroinitializer, <vscale x 64 x i16> zeroinitializer, i32 -1)
+  %splice.nxv1i16 = call <vscale x 1 x i16> @llvm.vector.splice.nxv1i16(<vscale x 1 x i16> zeroinitializer, <vscale x 1 x i16> zeroinitializer, i32 -1)
+  %splice.nxv2i16 = call <vscale x 2 x i16> @llvm.vector.splice.nxv2i16(<vscale x 2 x i16> zeroinitializer, <vscale x 2 x i16> zeroinitializer, i32 -1)
+  %splice.nxv4i16 = call <vscale x 4 x i16> @llvm.vector.splice.nxv4i16(<vscale x 4 x i16> zeroinitializer, <vscale x 4 x i16> zeroinitializer, i32 -1)
+  %splice.nxv8i16 = call <vscale x 8 x i16> @llvm.vector.splice.nxv8i16(<vscale x 8 x i16> zeroinitializer, <vscale x 8 x i16> zeroinitializer, i32 -1)
+  %splice.nxv16i16 = call <vscale x 16 x i16> @llvm.vector.splice.nxv16i16(<vscale x 16 x i16> zeroinitializer, <vscale x 16 x i16> zeroinitializer, i32 -1)
+  %splice.nxv32i16 = call <vscale x 32 x i16> @llvm.vector.splice.nxv32i16(<vscale x 32 x i16> zeroinitializer, <vscale x 32 x i16> zeroinitializer, i32 -1)
+  %splice.nxv64i16 = call <vscale x 64 x i16> @llvm.vector.splice.nxv64i16(<vscale x 64 x i16> zeroinitializer, <vscale x 64 x i16> zeroinitializer, i32 -1)
 
-  %splice.nxv1i32 = call <vscale x 1 x i32> @llvm.experimental.vector.splice.nxv1i32(<vscale x 1 x i32> zeroinitializer, <vscale x 1 x i32> zeroinitializer, i32 -1)
-  %splice.nxv2i32 = call <vscale x 2 x i32> @llvm.experimental.vector.splice.nxv2i32(<vscale x 2 x i32> zeroinitializer, <vscale x 2 x i32> zeroinitializer, i32 -1)
-  %splice.nxv4i32 = call <vscale x 4 x i32> @llvm.experimental.vector.splice.nxv4i32(<vscale x 4 x i32> zeroinitializer, <vscale x 4 x i32> zeroinitializer, i32 -1)
-  %splice.nxv8i32 = call <vscale x 8 x i32> @llvm.experimental.vector.splice.nxv8i32(<vscale x 8 x i32> zeroinitializer, <vscale x 8 x i32> zeroinitializer, i32 -1)
-  %splice.nxv16i32 = call <vscale x 16 x i32> @llvm.experimental.vector.splice.nxv16i32(<vscale x 16 x i32> zeroinitializer, <vscale x 16 x i32> zeroinitializer, i32 -1)
-  %splice.nxv32i32 = call <vscale x 32 x i32> @llvm.experimental.vector.splice.nxv32i32(<vscale x 32 x i32> zeroinitializer, <vscale x 32 x i32> zeroinitializer, i32 -1)
-  %splice.nxv64i32 = call <vscale x 64 x i32> @llvm.experimental.vector.splice.nxv64i32(<vscale x 64 x i32> zeroinitializer, <vscale x 64 x i32> zeroinitializer, i32 -1)
+  %splice.nxv1i32 = call <vscale x 1 x i32> @llvm.vector.splice.nxv1i32(<vscale x 1 x i32> zeroinitializer, <vscale x 1 x i32> zeroinitializer, i32 -1)
+  %splice.nxv2i32 = call <vscale x 2 x i32> @llvm.vector.splice.nxv2i32(<vscale x 2 x i32> zeroinitializer, <vscale x 2 x i32> zeroinitializer, i32 -1)
+  %splice.nxv4i32 = call <vscale x 4 x i32> @llvm.vector.splice.nxv4i32(<vscale x 4 x i32> zeroinitializer, <vscale x 4 x i32> zeroinitializer, i32 -1)
+  %splice.nxv8i32 = call <vscale x 8 x i32> @llvm.vector.splice.nxv8i32(<vscale x 8 x i32> zeroinitializer, <vscale x 8 x i32> zeroinitializer, i32 -1)
+  %splice.nxv16i32 = call <vscale x 16 x i32> @llvm.vector.splice.nxv16i32(<vscale x 16 x i32> zeroinitializer, <vscale x 16 x i32> zeroinitializer, i32 -1)
+  %splice.nxv32i32 = call <vscale x 32 x i32> @llvm.vector.splice.nxv32i32(<vscale x 32 x i32> zeroinitializer, <vscale x 32 x i32> zeroinitializer, i32 -1)
+  %splice.nxv64i32 = call <vscale x 64 x i32> @llvm.vector.splice.nxv64i32(<vscale x 64 x i32> zeroinitializer, <vscale x 64 x i32> zeroinitializer, i32 -1)
 
-  %splice.nxv1i64 = call <vscale x 1 x i64> @llvm.experimental.vector.splice.nxv1i64(<vscale x 1 x i64> zeroinitializer, <vscale x 1 x i64> zeroinitializer, i32 -1)
-  %splice.nxv2i64 = call <vscale x 2 x i64> @llvm.experimental.vector.splice.nxv2i64(<vscale x 2 x i64> zeroinitializer, <vscale x 2 x i64> zeroinitializer, i32 -1)
-  %splice.nxv4i64 = call <vscale x 4 x i64> @llvm.experimental.vector.splice.nxv4i64(<vscale x 4 x i64> zeroinitializer, <vscale x 4 x i64> zeroinitializer, i32 -1)
-  %splice.nxv8i64 = call <vscale x 8 x i64> @llvm.experimental.vector.splice.nxv8i64(<vscale x 8 x i64> zeroinitializer, <vscale x 8 x i64> zeroinitializer, i32 -1)
-  %splice.nxv16i64 = call <vscale x 16 x i64> @llvm.experimental.vector.splice.nxv16i64(<vscale x 16 x i64> zeroinitializer, <vscale x 16 x i64> zeroinitializer, i32 -1)
-  %splice.nxv32i64 = call <vscale x 32 x i64> @llvm.experimental.vector.splice.nxv32i64(<vscale x 32 x i64> zeroinitializer, <vscale x 32 x i64> zeroinitializer, i32 -1)
-  %splice.nxv64i64 = call <vscale x 64 x i64> @llvm.experimental.vector.splice.nxv64i64(<vscale x 64 x i64> zeroinitializer, <vscale x 64 x i64> zeroinitializer, i32 -1)
+  %splice.nxv1i64 = call <vscale x 1 x i64> @llvm.vector.splice.nxv1i64(<vscale x 1 x i64> zeroinitializer, <vscale x 1 x i64> zeroinitializer, i32 -1)
+  %splice.nxv2i64 = call <vscale x 2 x i64> @llvm.vector.splice.nxv2i64(<vscale x 2 x i64> zeroinitializer, <vscale x 2 x i64> zeroinitializer, i32 -1)
+  %splice.nxv4i64 = call <vscale x 4 x i64> @llvm.vector.splice.nxv4i64(<vscale x 4 x i64> zeroinitializer, <vscale x 4 x i64> zeroinitializer, i32 -1)
+  %splice.nxv8i64 = call <vscale x 8 x i64> @llvm.vector.splice.nxv8i64(<vscale x 8 x i64> zeroinitializer, <vscale x 8 x i64> zeroinitializer, i32 -1)
+  %splice.nxv16i64 = call <vscale x 16 x i64> @llvm.vector.splice.nxv16i64(<vscale x 16 x i64> zeroinitializer, <vscale x 16 x i64> zeroinitializer, i32 -1)
+  %splice.nxv32i64 = call <vscale x 32 x i64> @llvm.vector.splice.nxv32i64(<vscale x 32 x i64> zeroinitializer, <vscale x 32 x i64> zeroinitializer, i32 -1)
+  %splice.nxv64i64 = call <vscale x 64 x i64> @llvm.vector.splice.nxv64i64(<vscale x 64 x i64> zeroinitializer, <vscale x 64 x i64> zeroinitializer, i32 -1)
 
-  %splice.nxv1f16 = call <vscale x 1 x half> @llvm.experimental.vector.splice.nxv1f16(<vscale x 1 x half> zeroinitializer, <vscale x 1 x half> zeroinitializer, i32 -1)
-  %splice.nxv2f16 = call <vscale x 2 x half> @llvm.experimental.vector.splice.nxv2f16(<vscale x 2 x half> zeroinitializer, <vscale x 2 x half> zeroinitializer, i32 -1)
-  %splice.nxv4f16 = call <vscale x 4 x half> @llvm.experimental.vector.splice.nxv4f16(<vscale x 4 x half> zeroinitializer, <vscale x 4 x half> zeroinitializer, i32 -1)
-  %splice.nxv8f16 = call <vscale x 8 x half> @llvm.experimental.vector.splice.nxv8f16(<vscale x 8 x half> zeroinitializer, <vscale x 8 x half> zeroinitializer, i32 -1)
-  %splice.nxv16f16 = call <vscale x 16 x half> @llvm.experimental.vector.splice.nxv16f16(<vscale x 16 x half> zeroinitializer, <vscale x 16 x half> zeroinitializer, i32 -1)
-  %splice.nxv32f16 = call <vscale x 32 x half> @llvm.experimental.vector.splice.nxv32f16(<vscale x 32 x half> zeroinitializer, <vscale x 32 x half> zeroinitializer, i32 -1)
-  %splice.nxv64f16 = call <vscale x 64 x half> @llvm.experimental.vector.splice.nxv64f16(<vscale x 64 x half> zeroinitializer, <vscale x 64 x half> zeroinitializer, i32 -1)
+  %splice.nxv1f16 = call <vscale x 1 x half> @llvm.vector.splice.nxv1f16(<vscale x 1 x half> zeroinitializer, <vscale x 1 x half> zeroinitializer, i32 -1)
+  %splice.nxv2f16 = call <vscale x 2 x half> @llvm.vector.splice.nxv2f16(<vscale x 2 x half> zeroinitializer, <vscale x 2 x half> zeroinitializer, i32 -1)
+  %splice.nxv4f16 = call <vscale x 4 x half> @llvm.vector.splice.nxv4f16(<vscale x 4 x half> zeroinitializer, <vscale x 4 x half> zeroinitializer, i32 -1)
+  %splice.nxv8f16 = call <vscale x 8 x half> @llvm.vector.splice.nxv8f16(<vscale x 8 x half> zeroinitializer, <vscale x 8 x half> zeroinitializer, i32 -1)
+  %splice.nxv16f16 = call <vscale x 16 x half> @llvm.vector.splice.nxv16f16(<vscale x 16 x half> zeroinitializer, <vscale x 16 x half> zeroinitializer, i32 -1)
+  %splice.nxv32f16 = call <vscale x 32 x half> @llvm.vector.splice.nxv32f16(<vscale x 32 x half> zeroinitializer, <vscale x 32 x half> zeroinitializer, i32 -1)
+  %splice.nxv64f16 = call <vscale x 64 x half> @llvm.vector.splice.nxv64f16(<vscale x 64 x half> zeroinitializer, <vscale x 64 x half> zeroinitializer, i32 -1)
 
-  %splice.nxv1f32 = call <vscale x 1 x float> @llvm.experimental.vector.splice.nxv1f32(<vscale x 1 x float> zeroinitializer, <vscale x 1 x float> zeroinitializer, i32 -1)
-  %splice.nxv2f32 = call <vscale x 2 x float> @llvm.experimental.vector.splice.nxv2f32(<vscale x 2 x float> zeroinitializer, <vscale x 2 x float> zeroinitializer, i32 -1)
-  %splice.nxv4f32 = call <vscale x 4 x float> @llvm.experimental.vector.splice.nxv4f32(<vscale x 4 x float> zeroinitializer, <vscale x 4 x float> zeroinitializer, i32 -1)
-  %splice.nxv8f32 = call <vscale x 8 x float> @llvm.experimental.vector.splice.nxv8f32(<vscale x 8 x float> zeroinitializer, <vscale x 8 x float> zeroinitializer, i32 -1)
-  %splice.nxv16f32 = call <vscale x 16 x float> @llvm.experimental.vector.splice.nxv16f32(<vscale x 16 x float> zeroinitializer, <vscale x 16 x float> zeroinitializer, i32 -1)
-  %splice.nxv32f32 = call <vscale x 32 x float> @llvm.experimental.vector.splice.nxv32f32(<vscale x 32 x float> zeroinitializer, <vscale x 32 x float> zeroinitializer, i32 -1)
-  %splice.nxv64f32 = call <vscale x 64 x float> @llvm.experimental.vector.splice.nxv64f32(<vscale x 64 x float> zeroinitializer, <vscale x 64 x float> zeroinitializer, i32 -1)
+  %splice.nxv1f32 = call <vscale x 1 x float> @llvm.vector.splice.nxv1f32(<vscale x 1 x float> zeroinitializer, <vscale x 1 x float> zeroinitializer, i32 -1)
+  %splice.nxv2f32 = call <vscale x 2 x float> @llvm.vector.splice.nxv2f32(<vscale x 2 x float> zeroinitializer, <vscale x 2 x float> zeroinitializer, i32 -1)
+  %splice.nxv4f32 = call <vscale x 4 x float> @llvm.vector.splice.nxv4f32(<vscale x 4 x float> zeroinitializer, <vscale x 4 x float> zeroinitializer, i32 -1)
+  %splice.nxv8f32 = call <vscale x 8 x float> @llvm.vector.splice.nxv8f32(<vscale x 8 x float> zeroinitializer, <vscale x 8 x float> zeroinitializer, i32 -1)
+  %splice.nxv16f32 = call <vscale x 16 x float> @llvm.vector.splice.nxv16f32(<vscale x 16 x float> zeroinitializer, <vscale x 16 x float> zeroinitializer, i32 -1)
+  %splice.nxv32f32 = call <vscale x 32 x float> @llvm.vector.splice.nxv32f32(<vscale x 32 x float> zeroinitializer, <vscale x 32 x float> zeroinitializer, i32 -1)
+  %splice.nxv64f32 = call <vscale x 64 x float> @llvm.vector.splice.nxv64f32(<vscale x 64 x float> zeroinitializer, <vscale x 64 x float> zeroinitializer, i32 -1)
 
-  %splice.nxv1f64 = call <vscale x 1 x double> @llvm.experimental.vector.splice.nxv1f64(<vscale x 1 x double> zeroinitializer, <vscale x 1 x double> zeroinitializer, i32 -1)
-  %splice.nxv2f64 = call <vscale x 2 x double> @llvm.experimental.vector.splice.nxv2f64(<vscale x 2 x double> zeroinitializer, <vscale x 2 x double> zeroinitializer, i32 -1)
-  %splice.nxv4f64 = call <vscale x 4 x double> @llvm.experimental.vector.splice.nxv4f64(<vscale x 4 x double> zeroinitializer, <vscale x 4 x double> zeroinitializer, i32 -1)
-  %splice.nxv8f64 = call <vscale x 8 x double> @llvm.experimental.vector.splice.nxv8f64(<vscale x 8 x double> zeroinitializer, <vscale x 8 x double> zeroinitializer, i32 -1)
-  %splice.nxv16f64 = call <vscale x 16 x double> @llvm.experimental.vector.splice.nxv16f64(<vscale x 16 x double> zeroinitializer, <vscale x 16 x double> zeroinitializer, i32 -1)
-  %splice.nxv32f64 = call <vscale x 32 x double> @llvm.experimental.vector.splice.nxv32f64(<vscale x 32 x double> zeroinitializer, <vscale x 32 x double> zeroinitializer, i32 -1)
-  %splice.nxv64f64 = call <vscale x 64 x double> @llvm.experimental.vector.splice.nxv64f64(<vscale x 64 x double> zeroinitializer, <vscale x 64 x double> zeroinitializer, i32 -1)
+  %splice.nxv1f64 = call <vscale x 1 x double> @llvm.vector.splice.nxv1f64(<vscale x 1 x double> zeroinitializer, <vscale x 1 x double> zeroinitializer, i32 -1)
+  %splice.nxv2f64 = call <vscale x 2 x double> @llvm.vector.splice.nxv2f64(<vscale x 2 x double> zeroinitializer, <vscale x 2 x double> zeroinitializer, i32 -1)
+  %splice.nxv4f64 = call <vscale x 4 x double> @llvm.vector.splice.nxv4f64(<vscale x 4 x double> zeroinitializer, <vscale x 4 x double> zeroinitializer, i32 -1)
+  %splice.nxv8f64 = call <vscale x 8 x double> @llvm.vector.splice.nxv8f64(<vscale x 8 x double> zeroinitializer, <vscale x 8 x double> zeroinitializer, i32 -1)
+  %splice.nxv16f64 = call <vscale x 16 x double> @llvm.vector.splice.nxv16f64(<vscale x 16 x double> zeroinitializer, <vscale x 16 x double> zeroinitializer, i32 -1)
+  %splice.nxv32f64 = call <vscale x 32 x double> @llvm.vector.splice.nxv32f64(<vscale x 32 x double> zeroinitializer, <vscale x 32 x double> zeroinitializer, i32 -1)
+  %splice.nxv64f64 = call <vscale x 64 x double> @llvm.vector.splice.nxv64f64(<vscale x 64 x double> zeroinitializer, <vscale x 64 x double> zeroinitializer, i32 -1)
 
   ret void
 }
 
-declare <vscale x 1 x i8> @llvm.experimental.vector.splice.nxv1i8(<vscale x 1 x i8>, <vscale x 1 x i8>, i32)
-declare <vscale x 2 x i8> @llvm.experimental.vector.splice.nxv2i8(<vscale x 2 x i8>, <vscale x 2 x i8>, i32)
-declare <vscale x 4 x i8> @llvm.experimental.vector.splice.nxv4i8(<vscale x 4 x i8>, <vscale x 4 x i8>, i32)
-declare <vscale x 8 x i8> @llvm.experimental.vector.splice.nxv8i8(<vscale x 8 x i8>, <vscale x 8 x i8>, i32)
-declare <vscale x 16 x i8> @llvm.experimental.vector.splice.nxv16i8(<vscale x 16 x i8>, <vscale x 16 x i8>, i32)
-declare <vscale x 32 x i8> @llvm.experimental.vector.splice.nxv32i8(<vscale x 32 x i8>, <vscale x 32 x i8>, i32)
-declare <vscale x 64 x i8> @llvm.experimental.vector.splice.nxv64i8(<vscale x 64 x i8>, <vscale x 64 x i8>, i32)
+declare <vscale x 1 x i8> @llvm.vector.splice.nxv1i8(<vscale x 1 x i8>, <vscale x 1 x i8>, i32)
+declare <vscale x 2 x i8> @llvm.vector.splice.nxv2i8(<vscale x 2 x i8>, <vscale x 2 x i8>, i32)
+declare <vscale x 4 x i8> @llvm.vector.splice.nxv4i8(<vscale x 4 x i8>, <vscale x 4 x i8>, i32)
+declare <vscale x 8 x i8> @llvm.vector.splice.nxv8i8(<vscale x 8 x i8>, <vscale x 8 x i8>, i32)
+declare <vscale x 16 x i8> @llvm.vector.splice.nxv16i8(<vscale x 16 x i8>, <vscale x 16 x i8>, i32)
+declare <vscale x 32 x i8> @llvm.vector.splice.nxv32i8(<vscale x 32 x i8>, <vscale x 32 x i8>, i32)
+declare <vscale x 64 x i8> @llvm.vector.splice.nxv64i8(<vscale x 64 x i8>, <vscale x 64 x i8>, i32)
 
-declare <vscale x 1 x i16> @llvm.experimental.vector.splice.nxv1i16(<vscale x 1 x i16>, <vscale x 1 x i16>, i32)
-declare <vscale x 2 x i16> @llvm.experimental.vector.splice.nxv2i16(<vscale x 2 x i16>, <vscale x 2 x i16>, i32)
-declare <vscale x 4 x i16> @llvm.experimental.vector.splice.nxv4i16(<vscale x 4 x i16>, <vscale x 4 x i16>, i32)
-declare <vscale x 8 x i16> @llvm.experimental.vector.splice.nxv8i16(<vscale x 8 x i16>, <vscale x 8 x i16>, i32)
-declare <vscale x 16 x i16> @llvm.experimental.vector.splice.nxv16i16(<vscale x 16 x i16>, <vscale x 16 x i16>, i32)
-declare <vscale x 32 x i16> @llvm.experimental.vector.splice.nxv32i16(<vscale x 32 x i16>, <vscale x 32 x i16>, i32)
-declare <vscale x 64 x i16> @llvm.experimental.vector.splice.nxv64i16(<vscale x 64 x i16>, <vscale x 64 x i16>, i32)
+declare <vscale x 1 x i16> @llvm.vector.splice.nxv1i16(<vscale x 1 x i16>, <vscale x 1 x i16>, i32)
+declare <vscale x 2 x i16> @llvm.vector.splice.nxv2i16(<vscale x 2 x i16>, <vscale x 2 x i16>, i32)
+declare <vscale x 4 x i16> @llvm.vector.splice.nxv4i16(<vscale x 4 x i16>, <vscale x 4 x i16>, i32)
+declare <vscale x 8 x i16> @llvm.vector.splice.nxv8i16(<vscale x 8 x i16>, <vscale x 8 x i16>, i32)
+declare <vscale x 16 x i16> @llvm.vector.splice.nxv16i16(<vscale x 16 x i16>, <vscale x 16 x i16>, i32)
+declare <vscale x 32 x i16> @llvm.vector.splice.nxv32i16(<vscale x 32 x i16>, <vscale x 32 x i16>, i32)
+declare <vscale x 64 x i16> @llvm.vector.splice.nxv64i16(<vscale x 64 x i16>, <vscale x 64 x i16>, i32)
 
-declare <vscale x 1 x i32> @llvm.experimental.vector.splice.nxv1i32(<vscale x 1 x i32>, <vscale x 1 x i32>, i32)
-declare <vscale x 2 x i32> @llvm.experimental.vector.splice.nxv2i32(<vscale x 2 x i32>, <vscale x 2 x i32>, i32)
-declare <vscale x 4 x i32> @llvm.experimental.vector.splice.nxv4i32(<vscale x 4 x i32>, <vscale x 4 x i32>, i32)
-declare <vscale x 8 x i32> @llvm.experimental.vector.splice.nxv8i32(<vscale x 8 x i32>, <vscale x 8 x i32>, i32)
-declare <vscale x 16 x i32> @llvm.experimental.vector.splice.nxv16i32(<vscale x 16 x i32>, <vscale x 16 x i32>, i32)
-declare <vscale x 32 x i32> @llvm.experimental.vector.splice.nxv32i32(<vscale x 32 x i32>, <vscale x 32 x i32>, i32)
-declare <vscale x 64 x i32> @llvm.experimental.vector.splice.nxv64i32(<vscale x 64 x i32>, <vscale x 64 x i32>, i32)
+declare <vscale x 1 x i32> @llvm.vector.splice.nxv1i32(<vscale x 1 x i32>, <vscale x 1 x i32>, i32)
+declare <vscale x 2 x i32> @llvm.vector.splice.nxv2i32(<vscale x 2 x i32>, <vscale x 2 x i32>, i32)
+declare <vscale x 4 x i32> @llvm.vector.splice.nxv4i32(<vscale x 4 x i32>, <vscale x 4 x i32>, i32)
+declare <vscale x 8 x i32> @llvm.vector.splice.nxv8i32(<vscale x 8 x i32>, <vscale x 8 x i32>, i32)
+declare <vscale x 16 x i32> @llvm.vector.splice.nxv16i32(<vscale x 16 x i32>, <vscale x 16 x i32>, i32)
+declare <vscale x 32 x i32> @llvm.vector.splice.nxv32i32(<vscale x 32 x i32>, <vscale x 32 x i32>, i32)
+declare <vscale x 64 x i32> @llvm.vector.splice.nxv64i32(<vscale x 64 x i32>, <vscale x 64 x i32>, i32)
 
-declare <vscale x 1 x i64> @llvm.experimental.vector.splice.nxv1i64(<vscale x 1 x i64>, <vscale x 1 x i64>, i32)
-declare <vscale x 2 x i64> @llvm.experimental.vector.splice.nxv2i64(<vscale x 2 x i64>, <vscale x 2 x i64>, i32)
-declare <vscale x 4 x i64> @llvm.experimental.vector.splice.nxv4i64(<vscale x 4 x i64>, <vscale x 4 x i64>, i32)
-declare <vscale x 8 x i64> @llvm.experimental.vector.splice.nxv8i64(<vscale x 8 x i64>, <vscale x 8 x i64>, i32)
-declare <vscale x 16 x i64> @llvm.experimental.vector.splice.nxv16i64(<vscale x 16 x i64>, <vscale x 16 x i64>, i32)
-declare <vscale x 32 x i64> @llvm.experimental.vector.splice.nxv32i64(<vscale x 32 x i64>, <vscale x 32 x i64>, i32)
-declare <vscale x 64 x i64> @llvm.experimental.vector.splice.nxv64i64(<vscale x 64 x i64>, <vscale x 64 x i64>, i32)
+declare <vscale x 1 x i64> @llvm.vector.splice.nxv1i64(<vscale x 1 x i64>, <vscale x 1 x i64>, i32)
+declare <vscale x 2 x i64> @llvm.vector.splice.nxv2i64(<vscale x 2 x i64>, <vscale x 2 x i64>, i32)
+declare <vscale x 4 x i64> @llvm.vector.splice.nxv4i64(<vscale x 4 x i64>, <vscale x 4 x i64>, i32)
+declare <vscale x 8 x i64> @llvm.vector.splice.nxv8i64(<vscale x 8 x i64>, <vscale x 8 x i64>, i32)
+declare <vscale x 16 x i64> @llvm.vector.splice.nxv16i64(<vscale x 16 x i64>, <vscale x 16 x i64>, i32)
+declare <vscale x 32 x i64> @llvm.vector.splice.nxv32i64(<vscale x 32 x i64>, <vscale x 32 x i64>, i32)
+declare <vscale x 64 x i64> @llvm.vector.splice.nxv64i64(<vscale x 64 x i64>, <vscale x 64 x i64>, i32)
 
-declare <vscale x 1 x half> @llvm.experimental.vector.splice.nxv1f16(<vscale x 1 x half>, <vscale x 1 x half>, i32)
-declare <vscale x 2 x half> @llvm.experimental.vector.splice.nxv2f16(<vscale x 2 x half>, <vscale x 2 x half>, i32)
-declare <vscale x 4 x half> @llvm.experimental.vector.splice.nxv4f16(<vscale x 4 x half>, <vscale x 4 x half>, i32)
-declare <vscale x 8 x half> @llvm.experimental.vector.splice.nxv8f16(<vscale x 8 x half>, <vscale x 8 x half>, i32)
-declare <vscale x 16 x half> @llvm.experimental.vector.splice.nxv16f16(<vscale x 16 x half>, <vscale x 16 x half>, i32)
-declare <vscale x 32 x half> @llvm.experimental.vector.splice.nxv32f16(<vscale x 32 x half>, <vscale x 32 x half>, i32)
-declare <vscale x 64 x half> @llvm.experimental.vector.splice.nxv64f16(<vscale x 64 x half>, <vscale x 64 x half>, i32)
+declare <vscale x 1 x half> @llvm.vector.splice.nxv1f16(<vscale x 1 x half>, <vscale x 1 x half>, i32)
+declare <vscale x 2 x half> @llvm.vector.splice.nxv2f16(<vscale x 2 x half>, <vscale x 2 x half>, i32)
+declare <vscale x 4 x half> @llvm.vector.splice.nxv4f16(<vscale x 4 x half>, <vscale x 4 x half>, i32)
+declare <vscale x 8 x half> @llvm.vector.splice.nxv8f16(<vscale x 8 x half>, <vscale x 8 x half>, i32)
+declare <vscale x 16 x half> @llvm.vector.splice.nxv16f16(<vscale x 16 x half>, <vscale x 16 x half>, i32)
+declare <vscale x 32 x half> @llvm.vector.splice.nxv32f16(<vscale x 32 x half>, <vscale x 32 x half>, i32)
+declare <vscale x 64 x half> @llvm.vector.splice.nxv64f16(<vscale x 64 x half>, <vscale x 64 x half>, i32)
 
-declare <vscale x 1 x float> @llvm.experimental.vector.splice.nxv1f32(<vscale x 1 x float>, <vscale x 1 x float>, i32)
-declare <vscale x 2 x float> @llvm.experimental.vector.splice.nxv2f32(<vscale x 2 x float>, <vscale x 2 x float>, i32)
-declare <vscale x 4 x float> @llvm.experimental.vector.splice.nxv4f32(<vscale x 4 x float>, <vscale x 4 x float>, i32)
-declare <vscale x 8 x float> @llvm.experimental.vector.splice.nxv8f32(<vscale x 8 x float>, <vscale x 8 x float>, i32)
-declare <vscale x 16 x float> @llvm.experimental.vector.splice.nxv16f32(<vscale x 16 x float>, <vscale x 16 x float>, i32)
-declare <vscale x 32 x float> @llvm.experimental.vector.splice.nxv32f32(<vscale x 32 x float>, <vscale x 32 x float>, i32)
-declare <vscale x 64 x float> @llvm.experimental.vector.splice.nxv64f32(<vscale x 64 x float>, <vscale x 64 x float>, i32)
+declare <vscale x 1 x float> @llvm.vector.splice.nxv1f32(<vscale x 1 x float>, <vscale x 1 x float>, i32)
+declare <vscale x 2 x float> @llvm.vector.splice.nxv2f32(<vscale x 2 x float>, <vscale x 2 x float>, i32)
+declare <vscale x 4 x float> @llvm.vector.splice.nxv4f32(<vscale x 4 x float>, <vscale x 4 x float>, i32)
+declare <vscale x 8 x float> @llvm.vector.splice.nxv8f32(<vscale x 8 x float>, <vscale x 8 x float>, i32)
+declare <vscale x 16 x float> @llvm.vector.splice.nxv16f32(<vscale x 16 x float>, <vscale x 16 x float>, i32)
+declare <vscale x 32 x float> @llvm.vector.splice.nxv32f32(<vscale x 32 x float>, <vscale x 32 x float>, i32)
+declare <vscale x 64 x float> @llvm.vector.splice.nxv64f32(<vscale x 64 x float>, <vscale x 64 x float>, i32)
 
-declare <vscale x 1 x double> @llvm.experimental.vector.splice.nxv1f64(<vscale x 1 x double>, <vscale x 1 x double>, i32)
-declare <vscale x 2 x double> @llvm.experimental.vector.splice.nxv2f64(<vscale x 2 x double>, <vscale x 2 x double>, i32)
-declare <vscale x 4 x double> @llvm.experimental.vector.splice.nxv4f64(<vscale x 4 x double>, <vscale x 4 x double>, i32)
-declare <vscale x 8 x double> @llvm.experimental.vector.splice.nxv8f64(<vscale x 8 x double>, <vscale x 8 x double>, i32)
-declare <vscale x 16 x double> @llvm.experimental.vector.splice.nxv16f64(<vscale x 16 x double>, <vscale x 16 x double>, i32)
-declare <vscale x 32 x double> @llvm.experimental.vector.splice.nxv32f64(<vscale x 32 x double>, <vscale x 32 x double>, i32)
-declare <vscale x 64 x double> @llvm.experimental.vector.splice.nxv64f64(<vscale x 64 x double>, <vscale x 64 x double>, i32)
+declare <vscale x 1 x double> @llvm.vector.splice.nxv1f64(<vscale x 1 x double>, <vscale x 1 x double>, i32)
+declare <vscale x 2 x double> @llvm.vector.splice.nxv2f64(<vscale x 2 x double>, <vscale x 2 x double>, i32)
+declare <vscale x 4 x double> @llvm.vector.splice.nxv4f64(<vscale x 4 x double>, <vscale x 4 x double>, i32)
+declare <vscale x 8 x double> @llvm.vector.splice.nxv8f64(<vscale x 8 x double>, <vscale x 8 x double>, i32)
+declare <vscale x 16 x double> @llvm.vector.splice.nxv16f64(<vscale x 16 x double>, <vscale x 16 x double>, i32)
+declare <vscale x 32 x double> @llvm.vector.splice.nxv32f64(<vscale x 32 x double>, <vscale x 32 x double>, i32)
+declare <vscale x 64 x double> @llvm.vector.splice.nxv64f64(<vscale x 64 x double>, <vscale x 64 x double>, i32)
diff --git a/llvm/test/Analysis/CostModel/X86/load-broadcast.ll b/llvm/test/Analysis/CostModel/X86/load-broadcast.ll
new file mode 100644
index 000000000000..4cd8e5b636e9
--- /dev/null
+++ b/llvm/test/Analysis/CostModel/X86/load-broadcast.ll
@@ -0,0 +1,556 @@
+; NOTE: Assertions have been autogenerated by utils/update_analyze_test_checks.py UTC_ARGS: --version 4
+; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -passes="print<cost-model>" 2>&1 -disable-output -mattr=+sse2 | FileCheck %s -check-prefixes=SSE,SSE2
+; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -passes="print<cost-model>" 2>&1 -disable-output -mattr=+ssse3 | FileCheck %s -check-prefixes=SSE,SSSE3
+; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -passes="print<cost-model>" 2>&1 -disable-output -mattr=+sse4.2 | FileCheck %s -check-prefixes=SSE,SSE42
+; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -passes="print<cost-model>" 2>&1 -disable-output -mattr=+avx | FileCheck %s -check-prefixes=AVX,AVX1
+; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -passes="print<cost-model>" 2>&1 -disable-output -mattr=+avx2 | FileCheck %s -check-prefixes=AVX,AVX2
+; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -passes="print<cost-model>" 2>&1 -disable-output -mattr=+avx512f | FileCheck %s --check-prefixes=AVX,AVX512
+; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -passes="print<cost-model>" 2>&1 -disable-output -mattr=+avx512f,+avx512bw | FileCheck %s --check-prefixes=AVX,AVX512
+; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -passes="print<cost-model>" 2>&1 -disable-output -mattr=+avx512f,+avx512bw,+avx512vbmi | FileCheck %s --check-prefixes=AVX,AVX512
+;
+; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -passes="print<cost-model>" 2>&1 -disable-output -mcpu=slm | FileCheck %s --check-prefixes=SSE,SSE42
+; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -passes="print<cost-model>" 2>&1 -disable-output -mcpu=goldmont | FileCheck %s --check-prefixes=SSE,SSE42
+; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -passes="print<cost-model>" 2>&1 -disable-output -mcpu=btver2 | FileCheck %s --check-prefixes=AVX,AVX1
+
+;
+; vXf64
+;
+
+define <2 x double> @broadcast_load_v2f64_v2f64(ptr %src) {
+; SSE2-LABEL: 'broadcast_load_v2f64_v2f64'
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %load = load <2 x double>, ptr %src, align 16
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %bcst = shufflevector <2 x double> %load, <2 x double> poison, <2 x i32> zeroinitializer
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <2 x double> %bcst
+;
+; SSSE3-LABEL: 'broadcast_load_v2f64_v2f64'
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %load = load <2 x double>, ptr %src, align 16
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %bcst = shufflevector <2 x double> %load, <2 x double> poison, <2 x i32> zeroinitializer
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <2 x double> %bcst
+;
+; SSE42-LABEL: 'broadcast_load_v2f64_v2f64'
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %load = load <2 x double>, ptr %src, align 16
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %bcst = shufflevector <2 x double> %load, <2 x double> poison, <2 x i32> zeroinitializer
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <2 x double> %bcst
+;
+; AVX-LABEL: 'broadcast_load_v2f64_v2f64'
+; AVX-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %load = load <2 x double>, ptr %src, align 16
+; AVX-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %bcst = shufflevector <2 x double> %load, <2 x double> poison, <2 x i32> zeroinitializer
+; AVX-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <2 x double> %bcst
+;
+  %load = load <2 x double>, ptr %src
+  %bcst = shufflevector <2 x double> %load, <2 x double> poison, <2 x i32> zeroinitializer
+  ret <2 x double> %bcst
+}
+
+define <4 x double> @broadcast_load_v2f64_v4f64(ptr %src) {
+; SSE2-LABEL: 'broadcast_load_v2f64_v4f64'
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %load = load <2 x double>, ptr %src, align 16
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %bcst = shufflevector <2 x double> %load, <2 x double> poison, <4 x i32> zeroinitializer
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <4 x double> %bcst
+;
+; SSSE3-LABEL: 'broadcast_load_v2f64_v4f64'
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %load = load <2 x double>, ptr %src, align 16
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %bcst = shufflevector <2 x double> %load, <2 x double> poison, <4 x i32> zeroinitializer
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <4 x double> %bcst
+;
+; SSE42-LABEL: 'broadcast_load_v2f64_v4f64'
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %load = load <2 x double>, ptr %src, align 16
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %bcst = shufflevector <2 x double> %load, <2 x double> poison, <4 x i32> zeroinitializer
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <4 x double> %bcst
+;
+; AVX-LABEL: 'broadcast_load_v2f64_v4f64'
+; AVX-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %load = load <2 x double>, ptr %src, align 16
+; AVX-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %bcst = shufflevector <2 x double> %load, <2 x double> poison, <4 x i32> zeroinitializer
+; AVX-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <4 x double> %bcst
+;
+  %load = load <2 x double>, ptr %src
+  %bcst = shufflevector <2 x double> %load, <2 x double> poison, <4 x i32> zeroinitializer
+  ret <4 x double> %bcst
+}
+
+define <8 x double> @broadcast_load_v2f64_v8f64(ptr %src) {
+; SSE2-LABEL: 'broadcast_load_v2f64_v8f64'
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %load = load <2 x double>, ptr %src, align 16
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %bcst = shufflevector <2 x double> %load, <2 x double> poison, <8 x i32> zeroinitializer
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <8 x double> %bcst
+;
+; SSSE3-LABEL: 'broadcast_load_v2f64_v8f64'
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %load = load <2 x double>, ptr %src, align 16
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %bcst = shufflevector <2 x double> %load, <2 x double> poison, <8 x i32> zeroinitializer
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <8 x double> %bcst
+;
+; SSE42-LABEL: 'broadcast_load_v2f64_v8f64'
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %load = load <2 x double>, ptr %src, align 16
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %bcst = shufflevector <2 x double> %load, <2 x double> poison, <8 x i32> zeroinitializer
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <8 x double> %bcst
+;
+; AVX-LABEL: 'broadcast_load_v2f64_v8f64'
+; AVX-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %load = load <2 x double>, ptr %src, align 16
+; AVX-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %bcst = shufflevector <2 x double> %load, <2 x double> poison, <8 x i32> zeroinitializer
+; AVX-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <8 x double> %bcst
+;
+  %load = load <2 x double>, ptr %src
+  %bcst = shufflevector <2 x double> %load, <2 x double> poison, <8 x i32> zeroinitializer
+  ret <8 x double> %bcst
+}
+
+define <2 x double> @broadcast_load_f64_v2f64(ptr %src) {
+; SSE-LABEL: 'broadcast_load_f64_v2f64'
+; SSE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %load = load double, ptr %src, align 8
+; SSE-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %insert = insertelement <2 x double> poison, double %load, i32 0
+; SSE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %bcst = shufflevector <2 x double> %insert, <2 x double> poison, <2 x i32> zeroinitializer
+; SSE-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <2 x double> %bcst
+;
+; AVX-LABEL: 'broadcast_load_f64_v2f64'
+; AVX-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %load = load double, ptr %src, align 8
+; AVX-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %insert = insertelement <2 x double> poison, double %load, i32 0
+; AVX-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %bcst = shufflevector <2 x double> %insert, <2 x double> poison, <2 x i32> zeroinitializer
+; AVX-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <2 x double> %bcst
+;
+  %load = load double, ptr %src
+  %insert = insertelement <2 x double> poison, double %load, i32 0
+  %bcst = shufflevector <2 x double> %insert, <2 x double> poison, <2 x i32> zeroinitializer
+  ret <2 x double> %bcst
+}
+
+define <4 x double> @broadcast_load_f64_v4f64(ptr %src) {
+; SSE-LABEL: 'broadcast_load_f64_v4f64'
+; SSE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %load = load double, ptr %src, align 8
+; SSE-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %insert = insertelement <4 x double> poison, double %load, i32 0
+; SSE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %bcst = shufflevector <4 x double> %insert, <4 x double> poison, <4 x i32> zeroinitializer
+; SSE-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <4 x double> %bcst
+;
+; AVX1-LABEL: 'broadcast_load_f64_v4f64'
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %load = load double, ptr %src, align 8
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %insert = insertelement <4 x double> poison, double %load, i32 0
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %bcst = shufflevector <4 x double> %insert, <4 x double> poison, <4 x i32> zeroinitializer
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <4 x double> %bcst
+;
+; AVX2-LABEL: 'broadcast_load_f64_v4f64'
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %load = load double, ptr %src, align 8
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %insert = insertelement <4 x double> poison, double %load, i32 0
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %bcst = shufflevector <4 x double> %insert, <4 x double> poison, <4 x i32> zeroinitializer
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <4 x double> %bcst
+;
+; AVX512-LABEL: 'broadcast_load_f64_v4f64'
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %load = load double, ptr %src, align 8
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %insert = insertelement <4 x double> poison, double %load, i32 0
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %bcst = shufflevector <4 x double> %insert, <4 x double> poison, <4 x i32> zeroinitializer
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <4 x double> %bcst
+;
+  %load = load double, ptr %src
+  %insert = insertelement <4 x double> poison, double %load, i32 0
+  %bcst = shufflevector <4 x double> %insert, <4 x double> poison, <4 x i32> zeroinitializer
+  ret <4 x double> %bcst
+}
+
+define <8 x double> @broadcast_load_f64_v8f64(ptr %src) {
+; SSE-LABEL: 'broadcast_load_f64_v8f64'
+; SSE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %load = load double, ptr %src, align 8
+; SSE-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %insert = insertelement <8 x double> poison, double %load, i32 0
+; SSE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %bcst = shufflevector <8 x double> %insert, <8 x double> poison, <8 x i32> zeroinitializer
+; SSE-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <8 x double> %bcst
+;
+; AVX1-LABEL: 'broadcast_load_f64_v8f64'
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %load = load double, ptr %src, align 8
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %insert = insertelement <8 x double> poison, double %load, i32 0
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %bcst = shufflevector <8 x double> %insert, <8 x double> poison, <8 x i32> zeroinitializer
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <8 x double> %bcst
+;
+; AVX2-LABEL: 'broadcast_load_f64_v8f64'
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %load = load double, ptr %src, align 8
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %insert = insertelement <8 x double> poison, double %load, i32 0
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %bcst = shufflevector <8 x double> %insert, <8 x double> poison, <8 x i32> zeroinitializer
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <8 x double> %bcst
+;
+; AVX512-LABEL: 'broadcast_load_f64_v8f64'
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %load = load double, ptr %src, align 8
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %insert = insertelement <8 x double> poison, double %load, i32 0
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %bcst = shufflevector <8 x double> %insert, <8 x double> poison, <8 x i32> zeroinitializer
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <8 x double> %bcst
+;
+  %load = load double, ptr %src
+  %insert = insertelement <8 x double> poison, double %load, i32 0
+  %bcst = shufflevector <8 x double> %insert, <8 x double> poison, <8 x i32> zeroinitializer
+  ret <8 x double> %bcst
+}
+
+;
+; vXf32
+;
+
+define <4 x float> @broadcast_load_v4f32_v4f32(ptr %src) {
+; SSE-LABEL: 'broadcast_load_v4f32_v4f32'
+; SSE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %load = load <4 x float>, ptr %src, align 16
+; SSE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %bcst = shufflevector <4 x float> %load, <4 x float> poison, <4 x i32> zeroinitializer
+; SSE-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <4 x float> %bcst
+;
+; AVX-LABEL: 'broadcast_load_v4f32_v4f32'
+; AVX-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %load = load <4 x float>, ptr %src, align 16
+; AVX-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %bcst = shufflevector <4 x float> %load, <4 x float> poison, <4 x i32> zeroinitializer
+; AVX-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <4 x float> %bcst
+;
+  %load = load <4 x float>, ptr %src
+  %bcst = shufflevector <4 x float> %load, <4 x float> poison, <4 x i32> zeroinitializer
+  ret <4 x float> %bcst
+}
+
+define <8 x float> @broadcast_load_v4f32_v8f32(ptr %src) {
+; SSE-LABEL: 'broadcast_load_v4f32_v8f32'
+; SSE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %load = load <4 x float>, ptr %src, align 16
+; SSE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %bcst = shufflevector <4 x float> %load, <4 x float> poison, <8 x i32> zeroinitializer
+; SSE-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <8 x float> %bcst
+;
+; AVX-LABEL: 'broadcast_load_v4f32_v8f32'
+; AVX-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %load = load <4 x float>, ptr %src, align 16
+; AVX-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %bcst = shufflevector <4 x float> %load, <4 x float> poison, <8 x i32> zeroinitializer
+; AVX-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <8 x float> %bcst
+;
+  %load = load <4 x float>, ptr %src
+  %bcst = shufflevector <4 x float> %load, <4 x float> poison, <8 x i32> zeroinitializer
+  ret <8 x float> %bcst
+}
+
+define <16 x float> @broadcast_load_v4f32_v16f32(ptr %src) {
+; SSE-LABEL: 'broadcast_load_v4f32_v16f32'
+; SSE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %load = load <4 x float>, ptr %src, align 16
+; SSE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %bcst = shufflevector <4 x float> %load, <4 x float> poison, <16 x i32> zeroinitializer
+; SSE-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <16 x float> %bcst
+;
+; AVX-LABEL: 'broadcast_load_v4f32_v16f32'
+; AVX-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %load = load <4 x float>, ptr %src, align 16
+; AVX-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %bcst = shufflevector <4 x float> %load, <4 x float> poison, <16 x i32> zeroinitializer
+; AVX-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <16 x float> %bcst
+;
+  %load = load <4 x float>, ptr %src
+  %bcst = shufflevector <4 x float> %load, <4 x float> poison, <16 x i32> zeroinitializer
+  ret <16 x float> %bcst
+}
+
+define <2 x i64> @broadcast_load_v2i64_v2i64(ptr %src) {
+; SSE-LABEL: 'broadcast_load_v2i64_v2i64'
+; SSE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %load = load <2 x i64>, ptr %src, align 16
+; SSE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %bcst = shufflevector <2 x i64> %load, <2 x i64> poison, <2 x i32> zeroinitializer
+; SSE-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <2 x i64> %bcst
+;
+; AVX-LABEL: 'broadcast_load_v2i64_v2i64'
+; AVX-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %load = load <2 x i64>, ptr %src, align 16
+; AVX-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %bcst = shufflevector <2 x i64> %load, <2 x i64> poison, <2 x i32> zeroinitializer
+; AVX-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <2 x i64> %bcst
+;
+  %load = load <2 x i64>, ptr %src
+  %bcst = shufflevector <2 x i64> %load, <2 x i64> poison, <2 x i32> zeroinitializer
+  ret <2 x i64> %bcst
+}
+
+;
+; vXi64
+;
+
+define <4 x i64> @broadcast_load_v2i64_v4i64(ptr %src) {
+; SSE-LABEL: 'broadcast_load_v2i64_v4i64'
+; SSE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %load = load <2 x i64>, ptr %src, align 16
+; SSE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %bcst = shufflevector <2 x i64> %load, <2 x i64> poison, <4 x i32> zeroinitializer
+; SSE-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i64> %bcst
+;
+; AVX-LABEL: 'broadcast_load_v2i64_v4i64'
+; AVX-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %load = load <2 x i64>, ptr %src, align 16
+; AVX-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %bcst = shufflevector <2 x i64> %load, <2 x i64> poison, <4 x i32> zeroinitializer
+; AVX-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i64> %bcst
+;
+  %load = load <2 x i64>, ptr %src
+  %bcst = shufflevector <2 x i64> %load, <2 x i64> poison, <4 x i32> zeroinitializer
+  ret <4 x i64> %bcst
+}
+
+define <8 x i64> @broadcast_load_v2i64_v8i64(ptr %src) {
+; SSE-LABEL: 'broadcast_load_v2i64_v8i64'
+; SSE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %load = load <2 x i64>, ptr %src, align 16
+; SSE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %bcst = shufflevector <2 x i64> %load, <2 x i64> poison, <8 x i32> zeroinitializer
+; SSE-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i64> %bcst
+;
+; AVX-LABEL: 'broadcast_load_v2i64_v8i64'
+; AVX-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %load = load <2 x i64>, ptr %src, align 16
+; AVX-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %bcst = shufflevector <2 x i64> %load, <2 x i64> poison, <8 x i32> zeroinitializer
+; AVX-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i64> %bcst
+;
+  %load = load <2 x i64>, ptr %src
+  %bcst = shufflevector <2 x i64> %load, <2 x i64> poison, <8 x i32> zeroinitializer
+  ret <8 x i64> %bcst
+}
+
+;
+; vXi32
+;
+
+define <4 x i32> @broadcast_load_v4i32_v4i32(ptr %src) {
+; SSE-LABEL: 'broadcast_load_v4i32_v4i32'
+; SSE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %load = load <4 x i32>, ptr %src, align 16
+; SSE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %bcst = shufflevector <4 x i32> %load, <4 x i32> poison, <4 x i32> zeroinitializer
+; SSE-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i32> %bcst
+;
+; AVX-LABEL: 'broadcast_load_v4i32_v4i32'
+; AVX-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %load = load <4 x i32>, ptr %src, align 16
+; AVX-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %bcst = shufflevector <4 x i32> %load, <4 x i32> poison, <4 x i32> zeroinitializer
+; AVX-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <4 x i32> %bcst
+;
+  %load = load <4 x i32>, ptr %src
+  %bcst = shufflevector <4 x i32> %load, <4 x i32> poison, <4 x i32> zeroinitializer
+  ret <4 x i32> %bcst
+}
+
+define <8 x i32> @broadcast_load_v4i32_v8i32(ptr %src) {
+; SSE-LABEL: 'broadcast_load_v4i32_v8i32'
+; SSE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %load = load <4 x i32>, ptr %src, align 16
+; SSE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %bcst = shufflevector <4 x i32> %load, <4 x i32> poison, <8 x i32> zeroinitializer
+; SSE-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i32> %bcst
+;
+; AVX-LABEL: 'broadcast_load_v4i32_v8i32'
+; AVX-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %load = load <4 x i32>, ptr %src, align 16
+; AVX-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %bcst = shufflevector <4 x i32> %load, <4 x i32> poison, <8 x i32> zeroinitializer
+; AVX-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i32> %bcst
+;
+  %load = load <4 x i32>, ptr %src
+  %bcst = shufflevector <4 x i32> %load, <4 x i32> poison, <8 x i32> zeroinitializer
+  ret <8 x i32> %bcst
+}
+
+define <16 x i32> @broadcast_load_v4i32_v16i32(ptr %src) {
+; SSE-LABEL: 'broadcast_load_v4i32_v16i32'
+; SSE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %load = load <4 x i32>, ptr %src, align 16
+; SSE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %bcst = shufflevector <4 x i32> %load, <4 x i32> poison, <16 x i32> zeroinitializer
+; SSE-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i32> %bcst
+;
+; AVX-LABEL: 'broadcast_load_v4i32_v16i32'
+; AVX-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %load = load <4 x i32>, ptr %src, align 16
+; AVX-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %bcst = shufflevector <4 x i32> %load, <4 x i32> poison, <16 x i32> zeroinitializer
+; AVX-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i32> %bcst
+;
+  %load = load <4 x i32>, ptr %src
+  %bcst = shufflevector <4 x i32> %load, <4 x i32> poison, <16 x i32> zeroinitializer
+  ret <16 x i32> %bcst
+}
+
+;
+; vXi16
+;
+
+define <8 x i16> @broadcast_load_v8i16_v8i16(ptr %src) {
+; SSE2-LABEL: 'broadcast_load_v8i16_v8i16'
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %load = load <8 x i16>, ptr %src, align 16
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %bcst = shufflevector <8 x i16> %load, <8 x i16> poison, <8 x i32> zeroinitializer
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i16> %bcst
+;
+; SSSE3-LABEL: 'broadcast_load_v8i16_v8i16'
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %load = load <8 x i16>, ptr %src, align 16
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %bcst = shufflevector <8 x i16> %load, <8 x i16> poison, <8 x i32> zeroinitializer
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i16> %bcst
+;
+; SSE42-LABEL: 'broadcast_load_v8i16_v8i16'
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %load = load <8 x i16>, ptr %src, align 16
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %bcst = shufflevector <8 x i16> %load, <8 x i16> poison, <8 x i32> zeroinitializer
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i16> %bcst
+;
+; AVX1-LABEL: 'broadcast_load_v8i16_v8i16'
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %load = load <8 x i16>, ptr %src, align 16
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %bcst = shufflevector <8 x i16> %load, <8 x i16> poison, <8 x i32> zeroinitializer
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i16> %bcst
+;
+; AVX2-LABEL: 'broadcast_load_v8i16_v8i16'
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %load = load <8 x i16>, ptr %src, align 16
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %bcst = shufflevector <8 x i16> %load, <8 x i16> poison, <8 x i32> zeroinitializer
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i16> %bcst
+;
+; AVX512-LABEL: 'broadcast_load_v8i16_v8i16'
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %load = load <8 x i16>, ptr %src, align 16
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %bcst = shufflevector <8 x i16> %load, <8 x i16> poison, <8 x i32> zeroinitializer
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <8 x i16> %bcst
+;
+  %load = load <8 x i16>, ptr %src
+  %bcst = shufflevector <8 x i16> %load, <8 x i16> poison, <8 x i32> zeroinitializer
+  ret <8 x i16> %bcst
+}
+
+define <16 x i16> @broadcast_load_v8i16_v16i16(ptr %src) {
+; SSE2-LABEL: 'broadcast_load_v8i16_v16i16'
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %load = load <8 x i16>, ptr %src, align 16
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %bcst = shufflevector <8 x i16> %load, <8 x i16> poison, <16 x i32> zeroinitializer
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i16> %bcst
+;
+; SSSE3-LABEL: 'broadcast_load_v8i16_v16i16'
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %load = load <8 x i16>, ptr %src, align 16
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %bcst = shufflevector <8 x i16> %load, <8 x i16> poison, <16 x i32> zeroinitializer
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i16> %bcst
+;
+; SSE42-LABEL: 'broadcast_load_v8i16_v16i16'
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %load = load <8 x i16>, ptr %src, align 16
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %bcst = shufflevector <8 x i16> %load, <8 x i16> poison, <16 x i32> zeroinitializer
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i16> %bcst
+;
+; AVX1-LABEL: 'broadcast_load_v8i16_v16i16'
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %load = load <8 x i16>, ptr %src, align 16
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %bcst = shufflevector <8 x i16> %load, <8 x i16> poison, <16 x i32> zeroinitializer
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i16> %bcst
+;
+; AVX2-LABEL: 'broadcast_load_v8i16_v16i16'
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %load = load <8 x i16>, ptr %src, align 16
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %bcst = shufflevector <8 x i16> %load, <8 x i16> poison, <16 x i32> zeroinitializer
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i16> %bcst
+;
+; AVX512-LABEL: 'broadcast_load_v8i16_v16i16'
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %load = load <8 x i16>, ptr %src, align 16
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %bcst = shufflevector <8 x i16> %load, <8 x i16> poison, <16 x i32> zeroinitializer
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i16> %bcst
+;
+  %load = load <8 x i16>, ptr %src
+  %bcst = shufflevector <8 x i16> %load, <8 x i16> poison, <16 x i32> zeroinitializer
+  ret <16 x i16> %bcst
+}
+
+define <32 x i16> @broadcast_load_v8i16_v32i16(ptr %src) {
+; SSE2-LABEL: 'broadcast_load_v8i16_v32i16'
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %load = load <8 x i16>, ptr %src, align 16
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %bcst = shufflevector <8 x i16> %load, <8 x i16> poison, <32 x i32> zeroinitializer
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i16> %bcst
+;
+; SSSE3-LABEL: 'broadcast_load_v8i16_v32i16'
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %load = load <8 x i16>, ptr %src, align 16
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %bcst = shufflevector <8 x i16> %load, <8 x i16> poison, <32 x i32> zeroinitializer
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i16> %bcst
+;
+; SSE42-LABEL: 'broadcast_load_v8i16_v32i16'
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %load = load <8 x i16>, ptr %src, align 16
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %bcst = shufflevector <8 x i16> %load, <8 x i16> poison, <32 x i32> zeroinitializer
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i16> %bcst
+;
+; AVX1-LABEL: 'broadcast_load_v8i16_v32i16'
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %load = load <8 x i16>, ptr %src, align 16
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %bcst = shufflevector <8 x i16> %load, <8 x i16> poison, <32 x i32> zeroinitializer
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i16> %bcst
+;
+; AVX2-LABEL: 'broadcast_load_v8i16_v32i16'
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %load = load <8 x i16>, ptr %src, align 16
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %bcst = shufflevector <8 x i16> %load, <8 x i16> poison, <32 x i32> zeroinitializer
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i16> %bcst
+;
+; AVX512-LABEL: 'broadcast_load_v8i16_v32i16'
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %load = load <8 x i16>, ptr %src, align 16
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %bcst = shufflevector <8 x i16> %load, <8 x i16> poison, <32 x i32> zeroinitializer
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i16> %bcst
+;
+  %load = load <8 x i16>, ptr %src
+  %bcst = shufflevector <8 x i16> %load, <8 x i16> poison, <32 x i32> zeroinitializer
+  ret <32 x i16> %bcst
+}
+
+;
+; vXi8
+;
+
+define <16 x i8> @broadcast_load_v16i8_v16i8(ptr %src) {
+; SSE2-LABEL: 'broadcast_load_v16i8_v16i8'
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %load = load <16 x i8>, ptr %src, align 16
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %bcst = shufflevector <16 x i8> %load, <16 x i8> poison, <16 x i32> zeroinitializer
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i8> %bcst
+;
+; SSSE3-LABEL: 'broadcast_load_v16i8_v16i8'
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %load = load <16 x i8>, ptr %src, align 16
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %bcst = shufflevector <16 x i8> %load, <16 x i8> poison, <16 x i32> zeroinitializer
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i8> %bcst
+;
+; SSE42-LABEL: 'broadcast_load_v16i8_v16i8'
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %load = load <16 x i8>, ptr %src, align 16
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %bcst = shufflevector <16 x i8> %load, <16 x i8> poison, <16 x i32> zeroinitializer
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i8> %bcst
+;
+; AVX1-LABEL: 'broadcast_load_v16i8_v16i8'
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %load = load <16 x i8>, ptr %src, align 16
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %bcst = shufflevector <16 x i8> %load, <16 x i8> poison, <16 x i32> zeroinitializer
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i8> %bcst
+;
+; AVX2-LABEL: 'broadcast_load_v16i8_v16i8'
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %load = load <16 x i8>, ptr %src, align 16
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %bcst = shufflevector <16 x i8> %load, <16 x i8> poison, <16 x i32> zeroinitializer
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i8> %bcst
+;
+; AVX512-LABEL: 'broadcast_load_v16i8_v16i8'
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %load = load <16 x i8>, ptr %src, align 16
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %bcst = shufflevector <16 x i8> %load, <16 x i8> poison, <16 x i32> zeroinitializer
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <16 x i8> %bcst
+;
+  %load = load <16 x i8>, ptr %src
+  %bcst = shufflevector <16 x i8> %load, <16 x i8> poison, <16 x i32> zeroinitializer
+  ret <16 x i8> %bcst
+}
+
+define <32 x i8> @broadcast_load_v16i8_v32i8(ptr %src) {
+; SSE2-LABEL: 'broadcast_load_v16i8_v32i8'
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %load = load <16 x i8>, ptr %src, align 16
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %bcst = shufflevector <16 x i8> %load, <16 x i8> poison, <32 x i32> zeroinitializer
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i8> %bcst
+;
+; SSSE3-LABEL: 'broadcast_load_v16i8_v32i8'
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %load = load <16 x i8>, ptr %src, align 16
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %bcst = shufflevector <16 x i8> %load, <16 x i8> poison, <32 x i32> zeroinitializer
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i8> %bcst
+;
+; SSE42-LABEL: 'broadcast_load_v16i8_v32i8'
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %load = load <16 x i8>, ptr %src, align 16
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %bcst = shufflevector <16 x i8> %load, <16 x i8> poison, <32 x i32> zeroinitializer
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i8> %bcst
+;
+; AVX1-LABEL: 'broadcast_load_v16i8_v32i8'
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %load = load <16 x i8>, ptr %src, align 16
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %bcst = shufflevector <16 x i8> %load, <16 x i8> poison, <32 x i32> zeroinitializer
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i8> %bcst
+;
+; AVX2-LABEL: 'broadcast_load_v16i8_v32i8'
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %load = load <16 x i8>, ptr %src, align 16
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %bcst = shufflevector <16 x i8> %load, <16 x i8> poison, <32 x i32> zeroinitializer
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i8> %bcst
+;
+; AVX512-LABEL: 'broadcast_load_v16i8_v32i8'
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %load = load <16 x i8>, ptr %src, align 16
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %bcst = shufflevector <16 x i8> %load, <16 x i8> poison, <32 x i32> zeroinitializer
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <32 x i8> %bcst
+;
+  %load = load <16 x i8>, ptr %src
+  %bcst = shufflevector <16 x i8> %load, <16 x i8> poison, <32 x i32> zeroinitializer
+  ret <32 x i8> %bcst
+}
+
+define <64 x i8> @broadcast_load_v16i8_v64i8(ptr %src) {
+; SSE2-LABEL: 'broadcast_load_v16i8_v64i8'
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %load = load <16 x i8>, ptr %src, align 16
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %bcst = shufflevector <16 x i8> %load, <16 x i8> poison, <64 x i32> zeroinitializer
+; SSE2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <64 x i8> %bcst
+;
+; SSSE3-LABEL: 'broadcast_load_v16i8_v64i8'
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %load = load <16 x i8>, ptr %src, align 16
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %bcst = shufflevector <16 x i8> %load, <16 x i8> poison, <64 x i32> zeroinitializer
+; SSSE3-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <64 x i8> %bcst
+;
+; SSE42-LABEL: 'broadcast_load_v16i8_v64i8'
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %load = load <16 x i8>, ptr %src, align 16
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %bcst = shufflevector <16 x i8> %load, <16 x i8> poison, <64 x i32> zeroinitializer
+; SSE42-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <64 x i8> %bcst
+;
+; AVX1-LABEL: 'broadcast_load_v16i8_v64i8'
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %load = load <16 x i8>, ptr %src, align 16
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %bcst = shufflevector <16 x i8> %load, <16 x i8> poison, <64 x i32> zeroinitializer
+; AVX1-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <64 x i8> %bcst
+;
+; AVX2-LABEL: 'broadcast_load_v16i8_v64i8'
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %load = load <16 x i8>, ptr %src, align 16
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %bcst = shufflevector <16 x i8> %load, <16 x i8> poison, <64 x i32> zeroinitializer
+; AVX2-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <64 x i8> %bcst
+;
+; AVX512-LABEL: 'broadcast_load_v16i8_v64i8'
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %load = load <16 x i8>, ptr %src, align 16
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %bcst = shufflevector <16 x i8> %load, <16 x i8> poison, <64 x i32> zeroinitializer
+; AVX512-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret <64 x i8> %bcst
+;
+  %load = load <16 x i8>, ptr %src
+  %bcst = shufflevector <16 x i8> %load, <16 x i8> poison, <64 x i32> zeroinitializer
+  ret <64 x i8> %bcst
+}
diff --git a/llvm/test/Analysis/LoopAccessAnalysis/different-strides-safe-dep-due-to-backedge-taken-count.ll b/llvm/test/Analysis/LoopAccessAnalysis/different-strides-safe-dep-due-to-backedge-taken-count.ll
new file mode 100644
index 000000000000..5312c36e436a
--- /dev/null
+++ b/llvm/test/Analysis/LoopAccessAnalysis/different-strides-safe-dep-due-to-backedge-taken-count.ll
@@ -0,0 +1,154 @@
+; NOTE: Assertions have been autogenerated by utils/update_analyze_test_checks.py UTC_ARGS: --version 4
+; RUN: opt -passes='print<access-info>' -disable-output %s 2>&1 | FileCheck %s
+
+target datalayout = "e-m:o-i64:64-i128:128-n32:64-S128"
+
+define void @forward_dep_known_safe_due_to_backedge_taken_count(ptr %A) {
+; CHECK-LABEL: 'forward_dep_known_safe_due_to_backedge_taken_count'
+; CHECK-NEXT:    loop:
+; CHECK-NEXT:      Memory dependences are safe
+; CHECK-NEXT:      Dependences:
+; CHECK-NEXT:        Forward:
+; CHECK-NEXT:            %l = load i32, ptr %gep.mul.2, align 4 ->
+; CHECK-NEXT:            store i32 %add, ptr %gep, align 4
+; CHECK-EMPTY:
+; CHECK-NEXT:      Run-time memory checks:
+; CHECK-NEXT:      Grouped accesses:
+; CHECK-EMPTY:
+; CHECK-NEXT:      Non vectorizable stores to invariant address were not found in loop.
+; CHECK-NEXT:      SCEV assumptions:
+; CHECK-EMPTY:
+; CHECK-NEXT:      Expressions re-written:
+;
+entry:
+  %A.511= getelementptr inbounds i32, ptr %A, i64 511
+  br label %loop
+
+loop:
+  %iv = phi i64 [ 0, %entry ], [ %iv.next, %loop ]
+  %iv.mul.2 = shl nuw nsw i64 %iv, 1
+  %gep.mul.2 = getelementptr inbounds i32, ptr %A.511, i64 %iv.mul.2
+  %l = load i32, ptr %gep.mul.2, align 4
+  %add = add nsw i32 %l, 5
+  %gep = getelementptr inbounds i32, ptr %A, i64 %iv
+  store i32 %add, ptr %gep, align 4
+  %iv.next = add nuw nsw i64 %iv, 1
+  %exitcond.not = icmp eq i64 %iv.next, 256
+  br i1 %exitcond.not, label %exit, label %loop
+
+exit:
+  ret void
+}
+
+define void @forward_dep_not_known_safe_due_to_backedge_taken_count(ptr %A) {
+; CHECK-LABEL: 'forward_dep_not_known_safe_due_to_backedge_taken_count'
+; CHECK-NEXT:    loop:
+; CHECK-NEXT:      Memory dependences are safe
+; CHECK-NEXT:      Dependences:
+; CHECK-NEXT:        Forward:
+; CHECK-NEXT:            %l = load i32, ptr %gep.mul.2, align 4 ->
+; CHECK-NEXT:            store i32 %add, ptr %gep, align 4
+; CHECK-EMPTY:
+; CHECK-NEXT:      Run-time memory checks:
+; CHECK-NEXT:      Grouped accesses:
+; CHECK-EMPTY:
+; CHECK-NEXT:      Non vectorizable stores to invariant address were not found in loop.
+; CHECK-NEXT:      SCEV assumptions:
+; CHECK-EMPTY:
+; CHECK-NEXT:      Expressions re-written:
+;
+entry:
+  %A.510 = getelementptr inbounds i32, ptr %A, i64 510
+  br label %loop
+
+loop:
+  %iv = phi i64 [ 0, %entry ], [ %iv.next, %loop ]
+  %iv.mul.2 = shl nuw nsw i64 %iv, 1
+  %gep.mul.2 = getelementptr inbounds i32, ptr %A.510, i64 %iv.mul.2
+  %l = load i32, ptr %gep.mul.2, align 4
+  %add = add nsw i32 %l, 5
+  %gep = getelementptr inbounds i32, ptr %A, i64 %iv
+  store i32 %add, ptr %gep, align 4
+  %iv.next = add nuw nsw i64 %iv, 1
+  %exitcond.not = icmp eq i64 %iv.next, 256
+  br i1 %exitcond.not, label %exit, label %loop
+
+exit:
+  ret void
+}
+
+define void @unknown_dep_known_safe_due_to_backedge_taken_count(ptr %A) {
+; CHECK-LABEL: 'unknown_dep_known_safe_due_to_backedge_taken_count'
+; CHECK-NEXT:    loop:
+; CHECK-NEXT:      Report: unsafe dependent memory operations in loop. Use #pragma clang loop distribute(enable) to allow loop distribution to attempt to isolate the offending operations into a separate loop
+; CHECK-NEXT:  Unknown data dependence.
+; CHECK-NEXT:      Dependences:
+; CHECK-NEXT:        Unknown:
+; CHECK-NEXT:            %l = load i32, ptr %gep, align 4 ->
+; CHECK-NEXT:            store i32 %add, ptr %gep.mul.2, align 4
+; CHECK-EMPTY:
+; CHECK-NEXT:      Run-time memory checks:
+; CHECK-NEXT:      Grouped accesses:
+; CHECK-EMPTY:
+; CHECK-NEXT:      Non vectorizable stores to invariant address were not found in loop.
+; CHECK-NEXT:      SCEV assumptions:
+; CHECK-EMPTY:
+; CHECK-NEXT:      Expressions re-written:
+;
+entry:
+  %A.511 = getelementptr inbounds i32, ptr %A, i64 511
+  br label %loop
+
+loop:
+  %iv = phi i64 [ 0, %entry ], [ %iv.next, %loop ]
+  %iv.mul.2 = shl nuw nsw i64 %iv, 1
+  %gep = getelementptr inbounds i32, ptr %A, i64 %iv
+  %l = load i32, ptr %gep, align 4
+  %add = add nsw i32 %l, 5
+  %gep.mul.2 = getelementptr inbounds i32, ptr %A.511, i64 %iv.mul.2
+  store i32 %add, ptr %gep.mul.2, align 4
+  %iv.next = add nuw nsw i64 %iv, 1
+  %exitcond.not = icmp eq i64 %iv.next, 256
+  br i1 %exitcond.not, label %exit, label %loop
+
+exit:
+  ret void
+}
+
+define void @unknown_dep_not_known_safe_due_to_backedge_taken_count(ptr %A) {
+; CHECK-LABEL: 'unknown_dep_not_known_safe_due_to_backedge_taken_count'
+; CHECK-NEXT:    loop:
+; CHECK-NEXT:      Report: unsafe dependent memory operations in loop. Use #pragma clang loop distribute(enable) to allow loop distribution to attempt to isolate the offending operations into a separate loop
+; CHECK-NEXT:  Unknown data dependence.
+; CHECK-NEXT:      Dependences:
+; CHECK-NEXT:        Unknown:
+; CHECK-NEXT:            %l = load i32, ptr %gep, align 4 ->
+; CHECK-NEXT:            store i32 %add, ptr %gep.mul.2, align 4
+; CHECK-EMPTY:
+; CHECK-NEXT:      Run-time memory checks:
+; CHECK-NEXT:      Grouped accesses:
+; CHECK-EMPTY:
+; CHECK-NEXT:      Non vectorizable stores to invariant address were not found in loop.
+; CHECK-NEXT:      SCEV assumptions:
+; CHECK-EMPTY:
+; CHECK-NEXT:      Expressions re-written:
+;
+entry:
+  %A.510 = getelementptr inbounds i32, ptr %A, i64 510
+  br label %loop
+
+loop:
+  %iv = phi i64 [ 0, %entry ], [ %iv.next, %loop ]
+  %iv.mul.2 = shl nuw nsw i64 %iv, 1
+  %gep = getelementptr inbounds i32, ptr %A, i64 %iv
+  %l = load i32, ptr %gep, align 4
+  %add = add nsw i32 %l, 5
+  %gep.mul.2 = getelementptr inbounds i32, ptr %A.510, i64 %iv.mul.2
+  store i32 %add, ptr %gep.mul.2, align 4
+  %iv.next = add nuw nsw i64 %iv, 1
+  %exitcond.not = icmp eq i64 %iv.next, 256
+  br i1 %exitcond.not, label %exit, label %loop
+
+exit:
+  ret void
+}
diff --git a/llvm/test/Analysis/LoopAccessAnalysis/non-constant-strides-forward.ll b/llvm/test/Analysis/LoopAccessAnalysis/non-constant-strides-forward.ll
index 51755314896b..5f4c732dc19d 100644
--- a/llvm/test/Analysis/LoopAccessAnalysis/non-constant-strides-forward.ll
+++ b/llvm/test/Analysis/LoopAccessAnalysis/non-constant-strides-forward.ll
@@ -8,10 +8,9 @@ declare void @llvm.assume(i1)
 define void @different_non_constant_strides_known_forward(ptr %A) {
 ; CHECK-LABEL: 'different_non_constant_strides_known_forward'
 ; CHECK-NEXT:    loop:
-; CHECK-NEXT:      Report: unsafe dependent memory operations in loop. Use #pragma clang loop distribute(enable) to allow loop distribution to attempt to isolate the offending operations into a separate loop
-; CHECK-NEXT:  Unknown data dependence.
+; CHECK-NEXT:      Memory dependences are safe
 ; CHECK-NEXT:      Dependences:
-; CHECK-NEXT:        Unknown:
+; CHECK-NEXT:        Forward:
 ; CHECK-NEXT:            %l = load i32, ptr %gep.mul.2, align 4 ->
 ; CHECK-NEXT:            store i32 %add, ptr %gep, align 4
 ; CHECK-EMPTY:
@@ -45,10 +44,9 @@ exit:
 define void @different_non_constant_strides_known_forward_min_distance_3(ptr %A) {
 ; CHECK-LABEL: 'different_non_constant_strides_known_forward_min_distance_3'
 ; CHECK-NEXT:    loop:
-; CHECK-NEXT:      Report: unsafe dependent memory operations in loop. Use #pragma clang loop distribute(enable) to allow loop distribution to attempt to isolate the offending operations into a separate loop
-; CHECK-NEXT:  Unknown data dependence.
+; CHECK-NEXT:      Memory dependences are safe
 ; CHECK-NEXT:      Dependences:
-; CHECK-NEXT:        Unknown:
+; CHECK-NEXT:        Forward:
 ; CHECK-NEXT:            %l = load i32, ptr %gep.mul.2, align 4 ->
 ; CHECK-NEXT:            store i32 %add, ptr %gep, align 4
 ; CHECK-EMPTY:
diff --git a/llvm/test/Analysis/ValueTracking/known-non-zero.ll b/llvm/test/Analysis/ValueTracking/known-non-zero.ll
index 0159050d925c..c00e47fba8c7 100644
--- a/llvm/test/Analysis/ValueTracking/known-non-zero.ll
+++ b/llvm/test/Analysis/ValueTracking/known-non-zero.ll
@@ -1202,7 +1202,6 @@ define <2 x i1> @cmp_excludes_zero_with_nonsplat_vec_wpoison(<2 x i8> %a, <2 x i
   ret <2 x i1> %r
 }
 
-
 define <2 x i1> @cmp_excludes_zero_with_nonsplat_vec_fail(<2 x i8> %a, <2 x i8> %b) {
 ; CHECK-LABEL: @cmp_excludes_zero_with_nonsplat_vec_fail(
 ; CHECK-NEXT:    [[C:%.*]] = icmp sge <2 x i8> [[A:%.*]], <i8 0, i8 4>
@@ -1314,8 +1313,8 @@ define i1 @range_attr(i8 range(i8 1, 0) %x, i8 %y) {
 
 define i1 @neg_range_attr(i8 range(i8 -1, 1) %x, i8 %y) {
 ; CHECK-LABEL: @neg_range_attr(
-; CHECK-NEXT:    [[I:%.*]] = or i8 [[Y:%.*]], [[X:%.*]]
-; CHECK-NEXT:    [[CMP:%.*]] = icmp eq i8 [[I]], 0
+; CHECK-NEXT:    [[OR:%.*]] = or i8 [[Y:%.*]], [[X:%.*]]
+; CHECK-NEXT:    [[CMP:%.*]] = icmp eq i8 [[OR]], 0
 ; CHECK-NEXT:    ret i1 [[CMP]]
 ;
   %or = or i8 %y, %x
@@ -1328,7 +1327,7 @@ declare range(i8 -1, 1) i8 @returns_contain_zero_range_helper()
 
 define i1 @range_return(i8 %y) {
 ; CHECK-LABEL: @range_return(
-; CHECK-NEXT:    [[I:%.*]] = call i8 @returns_non_zero_range_helper()
+; CHECK-NEXT:    [[X:%.*]] = call i8 @returns_non_zero_range_helper()
 ; CHECK-NEXT:    ret i1 false
 ;
   %x = call i8 @returns_non_zero_range_helper()
@@ -1339,8 +1338,8 @@ define i1 @range_return(i8 %y) {
 
 define i1 @neg_range_return(i8 %y) {
 ; CHECK-LABEL: @neg_range_return(
-; CHECK-NEXT:    [[I:%.*]] = call i8 @returns_contain_zero_range_helper()
-; CHECK-NEXT:    [[OR:%.*]] = or i8 [[Y:%.*]], [[I]]
+; CHECK-NEXT:    [[X:%.*]] = call i8 @returns_contain_zero_range_helper()
+; CHECK-NEXT:    [[OR:%.*]] = or i8 [[Y:%.*]], [[X]]
 ; CHECK-NEXT:    [[CMP:%.*]] = icmp eq i8 [[OR]], 0
 ; CHECK-NEXT:    ret i1 [[CMP]]
 ;
@@ -1354,7 +1353,7 @@ declare i8 @returns_i8_helper()
 
 define i1 @range_call(i8 %y) {
 ; CHECK-LABEL: @range_call(
-; CHECK-NEXT:    [[I:%.*]] = call range(i8 1, 0) i8 @returns_i8_helper()
+; CHECK-NEXT:    [[X:%.*]] = call range(i8 1, 0) i8 @returns_i8_helper()
 ; CHECK-NEXT:    ret i1 false
 ;
   %x = call range(i8 1, 0) i8 @returns_i8_helper()
@@ -1365,8 +1364,8 @@ define i1 @range_call(i8 %y) {
 
 define i1 @neg_range_call(i8 %y) {
 ; CHECK-LABEL: @neg_range_call(
-; CHECK-NEXT:    [[I:%.*]] = call range(i8 -1, 1) i8 @returns_i8_helper()
-; CHECK-NEXT:    [[OR:%.*]] = or i8 [[Y:%.*]], [[I]]
+; CHECK-NEXT:    [[X:%.*]] = call range(i8 -1, 1) i8 @returns_i8_helper()
+; CHECK-NEXT:    [[OR:%.*]] = or i8 [[Y:%.*]], [[X]]
 ; CHECK-NEXT:    [[CMP:%.*]] = icmp eq i8 [[OR]], 0
 ; CHECK-NEXT:    ret i1 [[CMP]]
 ;
@@ -1401,7 +1400,7 @@ declare range(i8 -1, 1) <2 x i8> @returns_contain_zero_range_helper_vec()
 
 define <2 x i1> @range_return_vec(<2 x i8> %y) {
 ; CHECK-LABEL: @range_return_vec(
-; CHECK-NEXT:    [[I:%.*]] = call <2 x i8> @returns_non_zero_range_helper_vec()
+; CHECK-NEXT:    [[X:%.*]] = call <2 x i8> @returns_non_zero_range_helper_vec()
 ; CHECK-NEXT:    ret <2 x i1> <i1 true, i1 true>
 ;
   %x = call <2 x i8> @returns_non_zero_range_helper_vec()
@@ -1412,8 +1411,8 @@ define <2 x i1> @range_return_vec(<2 x i8> %y) {
 
 define <2 x i1> @neg_range_return_vec(<2 x i8> %y) {
 ; CHECK-LABEL: @neg_range_return_vec(
-; CHECK-NEXT:    [[I:%.*]] = call <2 x i8> @returns_contain_zero_range_helper_vec()
-; CHECK-NEXT:    [[OR:%.*]] = or <2 x i8> [[Y:%.*]], [[I]]
+; CHECK-NEXT:    [[X:%.*]] = call <2 x i8> @returns_contain_zero_range_helper_vec()
+; CHECK-NEXT:    [[OR:%.*]] = or <2 x i8> [[Y:%.*]], [[X]]
 ; CHECK-NEXT:    [[CMP:%.*]] = icmp ne <2 x i8> [[OR]], zeroinitializer
 ; CHECK-NEXT:    ret <2 x i1> [[CMP]]
 ;
@@ -1427,7 +1426,7 @@ declare <2 x i8> @returns_i8_helper_vec()
 
 define <2 x i1> @range_call_vec(<2 x i8> %y) {
 ; CHECK-LABEL: @range_call_vec(
-; CHECK-NEXT:    [[I:%.*]] = call range(i8 1, 0) <2 x i8> @returns_i8_helper_vec()
+; CHECK-NEXT:    [[X:%.*]] = call range(i8 1, 0) <2 x i8> @returns_i8_helper_vec()
 ; CHECK-NEXT:    ret <2 x i1> <i1 true, i1 true>
 ;
   %x = call range(i8 1, 0) <2 x i8> @returns_i8_helper_vec()
@@ -1438,8 +1437,8 @@ define <2 x i1> @range_call_vec(<2 x i8> %y) {
 
 define <2 x i1> @neg_range_call_vec(<2 x i8> %y) {
 ; CHECK-LABEL: @neg_range_call_vec(
-; CHECK-NEXT:    [[I:%.*]] = call range(i8 -1, 1) <2 x i8> @returns_i8_helper_vec()
-; CHECK-NEXT:    [[OR:%.*]] = or <2 x i8> [[Y:%.*]], [[I]]
+; CHECK-NEXT:    [[X:%.*]] = call range(i8 -1, 1) <2 x i8> @returns_i8_helper_vec()
+; CHECK-NEXT:    [[OR:%.*]] = or <2 x i8> [[Y:%.*]], [[X]]
 ; CHECK-NEXT:    [[CMP:%.*]] = icmp ne <2 x i8> [[OR]], zeroinitializer
 ; CHECK-NEXT:    ret <2 x i1> [[CMP]]
 ;
@@ -1449,5 +1448,53 @@ define <2 x i1> @neg_range_call_vec(<2 x i8> %y) {
   ret <2 x i1> %cmp
 }
 
+define i1 @trunc_nsw_non_zero(i8 %x) {
+; CHECK-LABEL: @trunc_nsw_non_zero(
+; CHECK-NEXT:    [[X_NE_Z:%.*]] = icmp ne i8 [[X:%.*]], 0
+; CHECK-NEXT:    call void @llvm.assume(i1 [[X_NE_Z]])
+; CHECK-NEXT:    ret i1 true
+;
+  %x_ne_z = icmp ne i8 %x, 0
+  call void @llvm.assume(i1 %x_ne_z)
+  %v = trunc nsw i8 %x to i4
+  %r = icmp ne i4 %v, 0
+  ret i1 %r
+}
+
+define i1 @trunc_nuw_non_zero(i8 %xx) {
+; CHECK-LABEL: @trunc_nuw_non_zero(
+; CHECK-NEXT:    ret i1 false
+;
+  %x = add nuw i8 %xx, 1
+  %v = trunc nuw i8 %x to i4
+  %r = icmp eq i4 %v, 0
+  ret i1 %r
+}
+
+define i1 @trunc_non_zero_fail(i8 %x) {
+; CHECK-LABEL: @trunc_non_zero_fail(
+; CHECK-NEXT:    [[X_NE_Z:%.*]] = icmp ne i8 [[X:%.*]], 0
+; CHECK-NEXT:    call void @llvm.assume(i1 [[X_NE_Z]])
+; CHECK-NEXT:    [[R:%.*]] = trunc i8 [[X]] to i1
+; CHECK-NEXT:    ret i1 [[R]]
+;
+  %x_ne_z = icmp ne i8 %x, 0
+  call void @llvm.assume(i1 %x_ne_z)
+  %r = trunc i8 %x to i1
+  ret i1 %r
+}
+
+define i1 @trunc_nsw_nuw_non_zero_fail(i8 %xx) {
+; CHECK-LABEL: @trunc_nsw_nuw_non_zero_fail(
+; CHECK-NEXT:    [[X:%.*]] = add nsw i8 [[XX:%.*]], 1
+; CHECK-NEXT:    [[V:%.*]] = trunc nuw nsw i8 [[X]] to i4
+; CHECK-NEXT:    [[R:%.*]] = icmp eq i4 [[V]], 0
+; CHECK-NEXT:    ret i1 [[R]]
+;
+  %x = add nsw i8 %xx, 1
+  %v = trunc nsw nuw i8 %x to i4
+  %r = icmp eq i4 %v, 0
+  ret i1 %r
+}
 
 declare i32 @llvm.experimental.get.vector.length.i32(i32, i32, i1)
diff --git a/llvm/test/Bitcode/upgrade-vector-interleave2-deinterleave2-intrinsics.ll b/llvm/test/Bitcode/upgrade-vector-interleave2-deinterleave2-intrinsics.ll
new file mode 100644
index 000000000000..f06395945297
--- /dev/null
+++ b/llvm/test/Bitcode/upgrade-vector-interleave2-deinterleave2-intrinsics.ll
@@ -0,0 +1,46 @@
+; RUN: opt -S < %s | FileCheck %s
+; RUN: llvm-as %s -o - | llvm-dis | FileCheck %s
+
+define <8 x i32> @interleave_fixed(<4 x i32> %a, <4 x i32> %b) {
+; CHECK-LABEL: @interleave_fixed
+; CHECK: %res = call <8 x i32> @llvm.vector.interleave2.v8i32(<4 x i32> %a, <4 x i32> %b)
+
+  %res = call <8 x i32> @llvm.experimental.vector.interleave2.v8i32(<4 x i32> %a, <4 x i32> %b)
+  ret <8 x i32> %res
+}
+
+define { <4 x i32>, <4 x i32> } @deinterleave_fixed(<8 x i32> %a) {
+; CHECK-LABEL: @deinterleave_fixed
+; CHECK: %res = call { <4 x i32>, <4 x i32> } @llvm.vector.deinterleave2.v8i32(<8 x i32> %a)
+
+  %res = call { <4 x i32>, <4 x i32> } @llvm.experimental.vector.deinterleave2.v8i32(<8 x i32> %a)
+  ret { <4 x i32>, <4 x i32> } %res
+}
+
+define <vscale x 8 x i32> @interleave_scalable(<vscale x 4 x i32> %a, <vscale x 4 x i32> %b) {
+; CHECK-LABEL: @interleave_scalable
+; CHECK: %res = call <vscale x 8 x i32> @llvm.vector.interleave2.nxv8i32(<vscale x 4 x i32> %a, <vscale x 4 x i32> %b)
+
+  %res = call <vscale x 8 x i32> @llvm.experimental.vector.interleave2.nxv8i32(<vscale x 4 x i32> %a, <vscale x 4 x i32> %b)
+  ret <vscale x 8 x i32> %res
+}
+
+define { <vscale x 4 x i32>, <vscale x 4 x i32> } @deinterleave_scalable(<vscale x 8 x i32> %a) {
+; CHECK-LABEL: @deinterleave_scalable
+; CHECK: %res = call { <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.vector.deinterleave2.nxv8i32(<vscale x 8 x i32> %a)
+
+  %res = call { <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.experimental.vector.deinterleave2.nxv8i32(<vscale x 8 x i32> %a)
+  ret { <vscale x 4 x i32>, <vscale x 4 x i32> } %res
+}
+
+declare <8 x i32> @llvm.experimental.vector.interleave2.v8i32(<4 x i32>, <4 x i32>)
+; CHECK: <8 x i32> @llvm.vector.interleave2.v8i32(<4 x i32>, <4 x i32>)
+
+declare  { <4 x i32>, <4 x i32> } @llvm.experimental.vector.deinterleave2.v8i32(<8 x i32>)
+; CHECK: declare  { <4 x i32>, <4 x i32> } @llvm.vector.deinterleave2.v8i32(<8 x i32>)
+
+declare <vscale x 8 x i32> @llvm.experimental.vector.interleave2.nxv8i32(<vscale x 4 x i32>, <vscale x 4 x i32>)
+; CHECK: <vscale x 8 x i32> @llvm.vector.interleave2.nxv8i32(<vscale x 4 x i32>, <vscale x 4 x i32>)
+
+declare  { <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.experimental.vector.deinterleave2.nxv8i32(<vscale x 8 x i32>)
+; CHECK: declare  { <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.vector.deinterleave2.nxv8i32(<vscale x 8 x i32>)
diff --git a/llvm/test/Bitcode/upgrade-vector-reverse-intrinsic.ll b/llvm/test/Bitcode/upgrade-vector-reverse-intrinsic.ll
new file mode 100644
index 000000000000..6b853eaf4175
--- /dev/null
+++ b/llvm/test/Bitcode/upgrade-vector-reverse-intrinsic.ll
@@ -0,0 +1,24 @@
+; RUN: opt -S < %s | FileCheck %s
+; RUN: llvm-as %s -o - | llvm-dis | FileCheck %s
+
+define <16 x i8> @reverse_fixed(<16 x i8> %a) {
+; CHECK-LABEL: @reverse_fixed
+; CHECK: %res = call <16 x i8> @llvm.vector.reverse.v16i8(<16 x i8> %a)
+
+  %res = call <16 x i8> @llvm.experimental.vector.reverse.v16i8(<16 x i8> %a)
+  ret <16 x i8> %res
+}
+
+define <vscale x 16 x i8> @reverse_scalable(<vscale x 16 x i8> %a) {
+; CHECK-LABEL: @reverse_scalable
+; CHECK: %res = call <vscale x 16 x i8> @llvm.vector.reverse.nxv16i8(<vscale x 16 x i8> %a)
+
+  %res = call <vscale x 16 x i8> @llvm.experimental.vector.reverse.nxv16i8(<vscale x 16 x i8> %a)
+  ret <vscale x 16 x i8> %res
+}
+
+declare <16 x i8> @llvm.experimental.vector.reverse.v16i8(<16 x i8>)
+; CHECK: declare <16 x i8> @llvm.vector.reverse.v16i8(<16 x i8>)
+
+declare <vscale x 16 x i8> @llvm.experimental.vector.reverse.nxv16i8(<vscale x 16 x i8>)
+; CHECK: declare <vscale x 16 x i8> @llvm.vector.reverse.nxv16i8(<vscale x 16 x i8>)
diff --git a/llvm/test/Bitcode/upgrade-vector-splice-intrinsic.ll b/llvm/test/Bitcode/upgrade-vector-splice-intrinsic.ll
new file mode 100644
index 000000000000..1b55da21ecd2
--- /dev/null
+++ b/llvm/test/Bitcode/upgrade-vector-splice-intrinsic.ll
@@ -0,0 +1,24 @@
+; RUN: opt -S < %s | FileCheck %s
+; RUN: llvm-as %s -o - | llvm-dis | FileCheck %s
+
+define <8 x half> @splice_fixed(<8 x half> %a, <8 x half> %b) {
+; CHECK-LABEL: @splice_fixed
+; CHECK: %res = call <8 x half> @llvm.vector.splice.v8f16(<8 x half> %a, <8 x half> %b, i32 2)
+
+  %res = call <8 x half> @llvm.experimental.vector.splice.v8f16(<8 x half> %a, <8 x half> %b, i32 2)
+  ret <8 x half> %res
+}
+
+define <vscale x 8 x half> @splice_scalable(<vscale x 8 x half> %a, <vscale x 8 x half> %b) {
+; CHECK-LABEL: @splice_scalable
+; CHECK: %res = call <vscale x 8 x half> @llvm.vector.splice.nxv8f16(<vscale x 8 x half> %a, <vscale x 8 x half> %b, i32 2)
+
+  %res = call <vscale x 8 x half> @llvm.experimental.vector.splice.nxv8f16(<vscale x 8 x half> %a, <vscale x 8 x half> %b, i32 2)
+  ret <vscale x 8 x half> %res
+}
+
+declare <8 x half> @llvm.experimental.vector.splice.v8f16(<8 x half>, <8 x half>, i32 immarg)
+; CHECK: declare <8 x half> @llvm.vector.splice.v8f16(<8 x half>, <8 x half>, i32 immarg)
+
+declare <vscale x 8 x half> @llvm.experimental.vector.splice.nxv8f16(<vscale x 8 x half>, <vscale x 8 x half>, i32 immarg)
+; CHECK: declare <vscale x 8 x half> @llvm.vector.splice.nxv8f16(<vscale x 8 x half>, <vscale x 8 x half>, i32 immarg)
diff --git a/llvm/test/CodeGen/AArch64/GlobalISel/combine-insert-vec-elt.mir b/llvm/test/CodeGen/AArch64/GlobalISel/combine-insert-vec-elt.mir
index 06fb2ce161c2..0c67a867580c 100644
--- a/llvm/test/CodeGen/AArch64/GlobalISel/combine-insert-vec-elt.mir
+++ b/llvm/test/CodeGen/AArch64/GlobalISel/combine-insert-vec-elt.mir
@@ -201,3 +201,113 @@ body:             |
     RET_ReallyLR
 
 ...
+---
+name:            test_idx_undef
+body:             |
+  bb.1:
+    liveins: $x0
+    ; CHECK-LABEL: name: test_idx_undef
+    ; CHECK: liveins: $x0
+    ; CHECK-NEXT: {{  $}}
+    ; CHECK-NEXT: RET_ReallyLR
+    %3:_(s8) = G_CONSTANT i8 127
+    %2:_(<32 x s8>) = G_BUILD_VECTOR %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8)
+    %4:_(s8) = G_CONSTANT i8 -128
+    %5:_(s64) = G_IMPLICIT_DEF
+    %0:_(p0) = COPY $x0
+    %1:_(<32 x s8>) = G_INSERT_VECTOR_ELT %2, %4(s8), %5(s64)
+    G_STORE %1(<32 x s8>), %0(p0) :: (store (<32 x s8>))
+    RET_ReallyLR
+
+...
+---
+name:            test_elt_undef
+body:             |
+  bb.1:
+    liveins: $x0
+    ; CHECK-LABEL: name: test_elt_undef
+    ; CHECK: liveins: $x0
+    ; CHECK-NEXT: {{  $}}
+    ; CHECK-NEXT: [[C:%[0-9]+]]:_(s8) = G_CONSTANT i8 127
+    ; CHECK-NEXT: [[DEF:%[0-9]+]]:_(s8) = G_IMPLICIT_DEF
+    ; CHECK-NEXT: [[COPY:%[0-9]+]]:_(p0) = COPY $x0
+    ; CHECK-NEXT: [[BUILD_VECTOR:%[0-9]+]]:_(<32 x s8>) = G_BUILD_VECTOR [[C]](s8), [[C]](s8), [[C]](s8), [[DEF]](s8), [[C]](s8), [[C]](s8), [[C]](s8), [[C]](s8), [[C]](s8), [[C]](s8), [[C]](s8), [[C]](s8), [[C]](s8), [[C]](s8), [[C]](s8), [[C]](s8), [[C]](s8), [[C]](s8), [[C]](s8), [[C]](s8), [[C]](s8), [[C]](s8), [[C]](s8), [[C]](s8), [[C]](s8), [[C]](s8), [[C]](s8), [[C]](s8), [[C]](s8), [[C]](s8), [[C]](s8), [[C]](s8)
+    ; CHECK-NEXT: G_STORE [[BUILD_VECTOR]](<32 x s8>), [[COPY]](p0) :: (store (<32 x s8>))
+    ; CHECK-NEXT: RET_ReallyLR
+    %3:_(s8) = G_CONSTANT i8 127
+    %2:_(<32 x s8>) = G_BUILD_VECTOR %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8)
+    %4:_(s8) = G_IMPLICIT_DEF
+    %5:_(s64) = G_CONSTANT i64 3
+    %0:_(p0) = COPY $x0
+    %1:_(<32 x s8>) = G_INSERT_VECTOR_ELT %2, %4(s8), %5(s64)
+    G_STORE %1(<32 x s8>), %0(p0) :: (store (<32 x s8>))
+    RET_ReallyLR
+
+...
+---
+name:            test_elt_undef_with_freeze
+body:             |
+  bb.1:
+    liveins: $x0
+    ; CHECK-LABEL: name: test_elt_undef_with_freeze
+    ; CHECK: liveins: $x0
+    ; CHECK-NEXT: {{  $}}
+    ; CHECK-NEXT: [[C:%[0-9]+]]:_(s8) = G_CONSTANT i8 127
+    ; CHECK-NEXT: [[BUILD_VECTOR:%[0-9]+]]:_(<32 x s8>) = G_BUILD_VECTOR [[C]](s8), [[C]](s8), [[C]](s8), [[C]](s8), [[C]](s8), [[C]](s8), [[C]](s8), [[C]](s8), [[C]](s8), [[C]](s8), [[C]](s8), [[C]](s8), [[C]](s8), [[C]](s8), [[C]](s8), [[C]](s8), [[C]](s8), [[C]](s8), [[C]](s8), [[C]](s8), [[C]](s8), [[C]](s8), [[C]](s8), [[C]](s8), [[C]](s8), [[C]](s8), [[C]](s8), [[C]](s8), [[C]](s8), [[C]](s8), [[C]](s8), [[C]](s8)
+    ; CHECK-NEXT: [[COPY:%[0-9]+]]:_(p0) = COPY $x0
+    ; CHECK-NEXT: [[FREEZE:%[0-9]+]]:_(<32 x s8>) = G_FREEZE [[BUILD_VECTOR]]
+    ; CHECK-NEXT: G_STORE [[FREEZE]](<32 x s8>), [[COPY]](p0) :: (store (<32 x s8>))
+    ; CHECK-NEXT: RET_ReallyLR
+    %3:_(s8) = G_CONSTANT i8 127
+    %2:_(<32 x s8>) = G_BUILD_VECTOR %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8)
+    %4:_(s8) = G_IMPLICIT_DEF
+    %5:_(s64) = G_CONSTANT i64 3
+    %0:_(p0) = COPY $x0
+    %9:_(<32 x s8>) = G_FREEZE %2
+    %1:_(<32 x s8>) = G_INSERT_VECTOR_ELT %9, %4(s8), %5(s64)
+    G_STORE %1(<32 x s8>), %0(p0) :: (store (<32 x s8>))
+    RET_ReallyLR
+
+...
+---
+name:            test_insert_extract
+body:             |
+  bb.1:
+    liveins: $x0
+    ; CHECK-LABEL: name: test_insert_extract
+    ; CHECK: liveins: $x0
+    ; CHECK-NEXT: {{  $}}
+    ; CHECK-NEXT: [[C:%[0-9]+]]:_(s8) = G_CONSTANT i8 127
+    ; CHECK-NEXT: [[COPY:%[0-9]+]]:_(p0) = COPY $x0
+    ; CHECK-NEXT: [[BUILD_VECTOR:%[0-9]+]]:_(<32 x s8>) = G_BUILD_VECTOR [[C]](s8), [[C]](s8), [[C]](s8), [[C]](s8), [[C]](s8), [[C]](s8), [[C]](s8), [[C]](s8), [[C]](s8), [[C]](s8), [[C]](s8), [[C]](s8), [[C]](s8), [[C]](s8), [[C]](s8), [[C]](s8), [[C]](s8), [[C]](s8), [[C]](s8), [[C]](s8), [[C]](s8), [[C]](s8), [[C]](s8), [[C]](s8), [[C]](s8), [[C]](s8), [[C]](s8), [[C]](s8), [[C]](s8), [[C]](s8), [[C]](s8), [[C]](s8)
+    ; CHECK-NEXT: G_STORE [[BUILD_VECTOR]](<32 x s8>), [[COPY]](p0) :: (store (<32 x s8>))
+    ; CHECK-NEXT: RET_ReallyLR
+    %3:_(s8) = G_CONSTANT i8 127
+    %2:_(<32 x s8>) = G_BUILD_VECTOR %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8)
+    %5:_(s64) = G_CONSTANT i64 3
+    %4:_(s8) = G_EXTRACT_VECTOR_ELT %2, %5
+    %0:_(p0) = COPY $x0
+    %1:_(<32 x s8>) = G_INSERT_VECTOR_ELT %2, %4(s8), %5(s64)
+    G_STORE %1(<32 x s8>), %0(p0) :: (store (<32 x s8>))
+    RET_ReallyLR
+
+...
+---
+name:            test_idx_oob
+body:             |
+  bb.1:
+    liveins: $x0
+    ; CHECK-LABEL: name: test_idx_oob
+    ; CHECK: liveins: $x0
+    ; CHECK-NEXT: {{  $}}
+    ; CHECK-NEXT: RET_ReallyLR
+    %3:_(s8) = G_CONSTANT i8 127
+    %2:_(<32 x s8>) = G_BUILD_VECTOR %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8), %3(s8)
+    %4:_(s8) = G_CONSTANT i8 -128
+    %5:_(s64) = G_CONSTANT i64 1024
+    %0:_(p0) = COPY $x0
+    %1:_(<32 x s8>) = G_INSERT_VECTOR_ELT %2, %4(s8), %5(s64)
+    G_STORE %1(<32 x s8>), %0(p0) :: (store (<32 x s8>))
+    RET_ReallyLR
+
+...
diff --git a/llvm/test/CodeGen/AArch64/GlobalISel/combine-udiv.ll b/llvm/test/CodeGen/AArch64/GlobalISel/combine-udiv.ll
index 9a525151ca32..c97a00ccdd45 100644
--- a/llvm/test/CodeGen/AArch64/GlobalISel/combine-udiv.ll
+++ b/llvm/test/CodeGen/AArch64/GlobalISel/combine-udiv.ll
@@ -243,3 +243,29 @@ define <8 x i16> @pr38477(<8 x i16> %a0) {
   %1 = udiv <8 x i16> %a0, <i16 1, i16 119, i16 73, i16 -111, i16 -3, i16 118, i16 32, i16 31>
   ret <8 x i16> %1
 }
+
+define i32 @udiv_div_by_180(i32 %x)
+; SDAG-LABEL: udiv_div_by_180:
+; SDAG:       // %bb.0:
+; SDAG-NEXT:    mov w8, #5826 // =0x16c2
+; SDAG-NEXT:    and w9, w0, #0xff
+; SDAG-NEXT:    movk w8, #364, lsl #16
+; SDAG-NEXT:    umull x8, w9, w8
+; SDAG-NEXT:    lsr x0, x8, #32
+; SDAG-NEXT:    // kill: def $w0 killed $w0 killed $x0
+; SDAG-NEXT:    ret
+;
+; GISEL-LABEL: udiv_div_by_180:
+; GISEL:       // %bb.0:
+; GISEL-NEXT:    uxtb w8, w0
+; GISEL-NEXT:    mov w9, #5826 // =0x16c2
+; GISEL-NEXT:    movk w9, #364, lsl #16
+; GISEL-NEXT:    umull x8, w8, w9
+; GISEL-NEXT:    lsr x0, x8, #32
+; GISEL-NEXT:    // kill: def $w0 killed $w0 killed $x0
+; GISEL-NEXT:    ret
+{
+  %truncate = and i32 %x, 255
+  %udiv = udiv i32 %truncate, 180
+  ret i32 %udiv
+}
diff --git a/llvm/test/CodeGen/AArch64/GlobalISel/irtranslator-vector-deinterleave2.ll b/llvm/test/CodeGen/AArch64/GlobalISel/irtranslator-vector-deinterleave2.ll
index 10882a06af1b..0b7fae47a65a 100644
--- a/llvm/test/CodeGen/AArch64/GlobalISel/irtranslator-vector-deinterleave2.ll
+++ b/llvm/test/CodeGen/AArch64/GlobalISel/irtranslator-vector-deinterleave2.ll
@@ -11,7 +11,7 @@ define void @vector_deinterleave2_v4i32(<4 x i32> %a) {
   ; CHECK-NEXT:   [[SHUF:%[0-9]+]]:_(<2 x s32>) = G_SHUFFLE_VECTOR [[COPY]](<4 x s32>), [[DEF]], shufflemask(0, 2)
   ; CHECK-NEXT:   [[SHUF1:%[0-9]+]]:_(<2 x s32>) = G_SHUFFLE_VECTOR [[COPY]](<4 x s32>), [[DEF]], shufflemask(1, 3)
   ; CHECK-NEXT:   RET_ReallyLR
-    %res = call {<2 x i32>, <2 x i32>} @llvm.experimental.vector.deinterleave2.v4i32(<4 x i32> %a)
+    %res = call {<2 x i32>, <2 x i32>} @llvm.vector.deinterleave2.v4i32(<4 x i32> %a)
     ret void
 }
 
@@ -29,6 +29,6 @@ define void @vector_deinterleave2_v8f32(<8 x float> %a) {
   ; CHECK-NEXT:   [[SHUF:%[0-9]+]]:_(<4 x s32>) = G_SHUFFLE_VECTOR [[CONCAT_VECTORS]](<8 x s32>), [[DEF]], shufflemask(0, 2, 4, 6)
   ; CHECK-NEXT:   [[SHUF1:%[0-9]+]]:_(<4 x s32>) = G_SHUFFLE_VECTOR [[CONCAT_VECTORS]](<8 x s32>), [[DEF]], shufflemask(1, 3, 5, 7)
   ; CHECK-NEXT:   RET_ReallyLR
-    %res = call {<4 x float>, <4 x float>} @llvm.experimental.vector.deinterleave2.v8f32(<8 x float> %a)
+    %res = call {<4 x float>, <4 x float>} @llvm.vector.deinterleave2.v8f32(<8 x float> %a)
     ret void
 }
diff --git a/llvm/test/CodeGen/AArch64/GlobalISel/irtranslator-vector-interleave2.ll b/llvm/test/CodeGen/AArch64/GlobalISel/irtranslator-vector-interleave2.ll
index f51e47a428d1..0d8ac82c1051 100644
--- a/llvm/test/CodeGen/AArch64/GlobalISel/irtranslator-vector-interleave2.ll
+++ b/llvm/test/CodeGen/AArch64/GlobalISel/irtranslator-vector-interleave2.ll
@@ -10,7 +10,7 @@ define void @vector_interleave2_v4i32(<2 x i32> %a, <2 x i32> %b) {
   ; CHECK-NEXT:   [[COPY1:%[0-9]+]]:_(<2 x s32>) = COPY $d1
   ; CHECK-NEXT:   [[SHUF:%[0-9]+]]:_(<4 x s32>) = G_SHUFFLE_VECTOR [[COPY]](<2 x s32>), [[COPY1]], shufflemask(0, 2, 1, 3)
   ; CHECK-NEXT:   RET_ReallyLR
-    %res = call <4 x i32> @llvm.experimental.vector.interleave2.v4i32(<2 x i32> %a, <2 x i32> %b)
+    %res = call <4 x i32> @llvm.vector.interleave2.v4i32(<2 x i32> %a, <2 x i32> %b)
     ret void
 }
 
@@ -25,6 +25,6 @@ define void @vector_interleave2_v8f32(<4 x float> %a, <4 x float> %b) {
   ; CHECK-NEXT:   [[BITCAST1:%[0-9]+]]:_(<4 x s32>) = G_BITCAST [[COPY1]](<2 x s64>)
   ; CHECK-NEXT:   [[SHUF:%[0-9]+]]:_(<8 x s32>) = G_SHUFFLE_VECTOR [[BITCAST]](<4 x s32>), [[BITCAST1]], shufflemask(0, 4, 1, 5, 2, 6, 3, 7)
   ; CHECK-NEXT:   RET_ReallyLR
-    %res = call <8 x float> @llvm.experimental.vector.interleave2.v8f32(<4 x float> %a, <4 x float> %b)
+    %res = call <8 x float> @llvm.vector.interleave2.v8f32(<4 x float> %a, <4 x float> %b)
     ret void
 }
diff --git a/llvm/test/CodeGen/AArch64/GlobalISel/legalize-bitcast.mir b/llvm/test/CodeGen/AArch64/GlobalISel/legalize-bitcast.mir
new file mode 100644
index 000000000000..e3a633c9e035
--- /dev/null
+++ b/llvm/test/CodeGen/AArch64/GlobalISel/legalize-bitcast.mir
@@ -0,0 +1,50 @@
+# NOTE: Assertions have been autogenerated by utils/update_mir_test_checks.py UTC_ARGS: --version 4
+# RUN: llc -mtriple=aarch64 -run-pass=legalizer -global-isel-abort=1  %s -o - | FileCheck %s
+---
+name:            scalar_to_oversize_vector
+tracksRegLiveness: true
+body:             |
+  ; CHECK-LABEL: name: scalar_to_oversize_vector
+  ; CHECK: bb.0:
+  ; CHECK-NEXT:   successors: %bb.1(0x80000000)
+  ; CHECK-NEXT: {{  $}}
+  ; CHECK-NEXT:   G_BR %bb.1
+  ; CHECK-NEXT: {{  $}}
+  ; CHECK-NEXT: bb.1:
+  ; CHECK-NEXT:   successors: %bb.1(0x80000000)
+  ; CHECK-NEXT: {{  $}}
+  ; CHECK-NEXT:   [[C:%[0-9]+]]:_(s64) = G_CONSTANT i64 0
+  ; CHECK-NEXT:   [[UV:%[0-9]+]]:_(s32), [[UV1:%[0-9]+]]:_(s32) = G_UNMERGE_VALUES [[C]](s64)
+  ; CHECK-NEXT:   [[UV2:%[0-9]+]]:_(s32), [[UV3:%[0-9]+]]:_(s32) = G_UNMERGE_VALUES [[C]](s64)
+  ; CHECK-NEXT:   [[UV4:%[0-9]+]]:_(s32), [[UV5:%[0-9]+]]:_(s32) = G_UNMERGE_VALUES [[C]](s64)
+  ; CHECK-NEXT:   [[UV6:%[0-9]+]]:_(s32), [[UV7:%[0-9]+]]:_(s32) = G_UNMERGE_VALUES [[C]](s64)
+  ; CHECK-NEXT:   [[UV8:%[0-9]+]]:_(s32), [[UV9:%[0-9]+]]:_(s32) = G_UNMERGE_VALUES [[C]](s64)
+  ; CHECK-NEXT:   [[UV10:%[0-9]+]]:_(s32), [[UV11:%[0-9]+]]:_(s32) = G_UNMERGE_VALUES [[C]](s64)
+  ; CHECK-NEXT:   [[UV12:%[0-9]+]]:_(s32), [[UV13:%[0-9]+]]:_(s32) = G_UNMERGE_VALUES [[C]](s64)
+  ; CHECK-NEXT:   [[UV14:%[0-9]+]]:_(s32), [[UV15:%[0-9]+]]:_(s32) = G_UNMERGE_VALUES [[C]](s64)
+  ; CHECK-NEXT:   [[BUILD_VECTOR:%[0-9]+]]:_(<4 x s32>) = G_BUILD_VECTOR [[UV]](s32), [[UV1]](s32), [[UV2]](s32), [[UV3]](s32)
+  ; CHECK-NEXT:   [[BUILD_VECTOR1:%[0-9]+]]:_(<4 x s32>) = G_BUILD_VECTOR [[UV4]](s32), [[UV5]](s32), [[UV6]](s32), [[UV7]](s32)
+  ; CHECK-NEXT:   [[BUILD_VECTOR2:%[0-9]+]]:_(<4 x s32>) = G_BUILD_VECTOR [[UV8]](s32), [[UV9]](s32), [[UV10]](s32), [[UV11]](s32)
+  ; CHECK-NEXT:   [[BUILD_VECTOR3:%[0-9]+]]:_(<4 x s32>) = G_BUILD_VECTOR [[UV12]](s32), [[UV13]](s32), [[UV14]](s32), [[UV15]](s32)
+  ; CHECK-NEXT:   [[C1:%[0-9]+]]:_(p0) = G_CONSTANT i64 0
+  ; CHECK-NEXT:   G_STORE [[BUILD_VECTOR]](<4 x s32>), [[C1]](p0) :: (store (<4 x s32>), align 64)
+  ; CHECK-NEXT:   [[C2:%[0-9]+]]:_(p0) = G_CONSTANT i64 16
+  ; CHECK-NEXT:   G_STORE [[BUILD_VECTOR1]](<4 x s32>), [[C2]](p0) :: (store (<4 x s32>) into unknown-address + 16)
+  ; CHECK-NEXT:   [[C3:%[0-9]+]]:_(p0) = G_CONSTANT i64 32
+  ; CHECK-NEXT:   G_STORE [[BUILD_VECTOR2]](<4 x s32>), [[C3]](p0) :: (store (<4 x s32>) into unknown-address + 32, align 32)
+  ; CHECK-NEXT:   [[C4:%[0-9]+]]:_(p0) = G_CONSTANT i64 48
+  ; CHECK-NEXT:   G_STORE [[BUILD_VECTOR3]](<4 x s32>), [[C4]](p0) :: (store (<4 x s32>) into unknown-address + 48)
+  ; CHECK-NEXT:   G_BR %bb.1
+  bb.1:
+    %0:_(s512) = G_CONSTANT i512 0
+    %2:_(p0) = G_CONSTANT i64 0
+    G_BR %bb.2
+
+  bb.2:
+    %4:_(s512) = G_CONSTANT i512 0
+    %1:_(<16 x s32>) = G_BITCAST %4(s512)
+    %3:_(p0) = G_CONSTANT i64 0
+    G_STORE %1(<16 x s32>), %3(p0) :: (store (<16 x s32>))
+    G_BR %bb.2
+
+...
diff --git a/llvm/test/CodeGen/AArch64/GlobalISel/legalize-extract-vector-elt.mir b/llvm/test/CodeGen/AArch64/GlobalISel/legalize-extract-vector-elt.mir
index c03f51a89dfb..323a3993473f 100644
--- a/llvm/test/CodeGen/AArch64/GlobalISel/legalize-extract-vector-elt.mir
+++ b/llvm/test/CodeGen/AArch64/GlobalISel/legalize-extract-vector-elt.mir
@@ -368,10 +368,10 @@ body: |
   bb.0:
     liveins: $x0
     ; CHECK-LABEL: name: test_eve_v4p0
-    ; CHECK: [[DEF:%[0-9]+]]:_(p0) = G_IMPLICIT_DEF
-    ; CHECK-NEXT: [[BUILD_VECTOR:%[0-9]+]]:_(<2 x p0>) = G_BUILD_VECTOR [[DEF]](p0), [[DEF]](p0)
+    ; CHECK: %vec:_(<4 x p0>) = G_IMPLICIT_DEF
     ; CHECK-NEXT: %idx:_(s64) = G_CONSTANT i64 1
-    ; CHECK-NEXT: %eve:_(p0) = G_EXTRACT_VECTOR_ELT [[BUILD_VECTOR]](<2 x p0>), %idx(s64)
+    ; CHECK-NEXT: [[UV:%[0-9]+]]:_(<2 x p0>), [[UV1:%[0-9]+]]:_(<2 x p0>) = G_UNMERGE_VALUES %vec(<4 x p0>)
+    ; CHECK-NEXT: %eve:_(p0) = G_EXTRACT_VECTOR_ELT [[UV]](<2 x p0>), %idx(s64)
     ; CHECK-NEXT: $x0 = COPY %eve(p0)
     ; CHECK-NEXT: RET_ReallyLR
     %vec:_(<4 x p0>) = G_IMPLICIT_DEF
@@ -452,16 +452,15 @@ body: |
   bb.0:
     liveins: $x0
     ; CHECK-LABEL: name: test_eve_v4p0_unknown_idx
-    ; CHECK: [[DEF:%[0-9]+]]:_(p0) = G_IMPLICIT_DEF
-    ; CHECK-NEXT: [[BUILD_VECTOR:%[0-9]+]]:_(<2 x p0>) = G_BUILD_VECTOR [[DEF]](p0), [[DEF]](p0)
-    ; CHECK-NEXT: [[BUILD_VECTOR1:%[0-9]+]]:_(<2 x p0>) = G_BUILD_VECTOR [[DEF]](p0), [[DEF]](p0)
+    ; CHECK: %vec:_(<4 x p0>) = G_IMPLICIT_DEF
     ; CHECK-NEXT: %idx:_(s64) = COPY $x0
     ; CHECK-NEXT: [[FRAME_INDEX:%[0-9]+]]:_(p0) = G_FRAME_INDEX %stack.0
-    ; CHECK-NEXT: [[BITCAST:%[0-9]+]]:_(<2 x s64>) = G_BITCAST [[BUILD_VECTOR]](<2 x p0>)
+    ; CHECK-NEXT: [[UV:%[0-9]+]]:_(<2 x p0>), [[UV1:%[0-9]+]]:_(<2 x p0>) = G_UNMERGE_VALUES %vec(<4 x p0>)
+    ; CHECK-NEXT: [[BITCAST:%[0-9]+]]:_(<2 x s64>) = G_BITCAST [[UV]](<2 x p0>)
     ; CHECK-NEXT: G_STORE [[BITCAST]](<2 x s64>), [[FRAME_INDEX]](p0) :: (store (<2 x s64>) into %stack.0, align 32)
     ; CHECK-NEXT: [[C:%[0-9]+]]:_(s64) = G_CONSTANT i64 16
     ; CHECK-NEXT: [[PTR_ADD:%[0-9]+]]:_(p0) = G_PTR_ADD [[FRAME_INDEX]], [[C]](s64)
-    ; CHECK-NEXT: [[BITCAST1:%[0-9]+]]:_(<2 x s64>) = G_BITCAST [[BUILD_VECTOR1]](<2 x p0>)
+    ; CHECK-NEXT: [[BITCAST1:%[0-9]+]]:_(<2 x s64>) = G_BITCAST [[UV1]](<2 x p0>)
     ; CHECK-NEXT: G_STORE [[BITCAST1]](<2 x s64>), [[PTR_ADD]](p0) :: (store (<2 x s64>) into %stack.0 + 16, basealign 32)
     ; CHECK-NEXT: [[C1:%[0-9]+]]:_(s64) = G_CONSTANT i64 3
     ; CHECK-NEXT: [[AND:%[0-9]+]]:_(s64) = G_AND %idx, [[C1]]
diff --git a/llvm/test/CodeGen/AArch64/GlobalISel/legalize-fp-arith.mir b/llvm/test/CodeGen/AArch64/GlobalISel/legalize-fp-arith.mir
index c687039da0b7..be397ede0b89 100644
--- a/llvm/test/CodeGen/AArch64/GlobalISel/legalize-fp-arith.mir
+++ b/llvm/test/CodeGen/AArch64/GlobalISel/legalize-fp-arith.mir
@@ -96,10 +96,9 @@ name:            test_fmul_v8s32
 body:             |
   bb.0.entry:
     ; CHECK-LABEL: name: test_fmul_v8s32
-    ; CHECK: [[DEF:%[0-9]+]]:_(s32) = G_IMPLICIT_DEF
-    ; CHECK: [[BUILD_VECTOR:%[0-9]+]]:_(<4 x s32>) = G_BUILD_VECTOR [[DEF]](s32), [[DEF]](s32), [[DEF]](s32), [[DEF]](s32)
-    ; CHECK: [[FMUL:%[0-9]+]]:_(<4 x s32>) = G_FMUL [[BUILD_VECTOR]], [[BUILD_VECTOR]]
-    ; CHECK: [[FMUL1:%[0-9]+]]:_(<4 x s32>) = G_FMUL [[BUILD_VECTOR]], [[BUILD_VECTOR]]
+    ; CHECK: [[DEF:%[0-9]+]]:_(<4 x s32>) = G_IMPLICIT_DEF
+    ; CHECK: [[FMUL:%[0-9]+]]:_(<4 x s32>) = G_FMUL [[DEF]], [[DEF]]
+    ; CHECK: [[FMUL1:%[0-9]+]]:_(<4 x s32>) = G_FMUL [[DEF]], [[DEF]]
     ; CHECK: $q0 = COPY [[FMUL]](<4 x s32>)
     ; CHECK: $q1 = COPY [[FMUL1]](<4 x s32>)
     %0:_(<8 x s32>) = G_IMPLICIT_DEF
diff --git a/llvm/test/CodeGen/AArch64/GlobalISel/legalize-freeze.mir b/llvm/test/CodeGen/AArch64/GlobalISel/legalize-freeze.mir
index be674d79b54f..3e768c4d7a26 100644
--- a/llvm/test/CodeGen/AArch64/GlobalISel/legalize-freeze.mir
+++ b/llvm/test/CodeGen/AArch64/GlobalISel/legalize-freeze.mir
@@ -130,3 +130,95 @@ body: |
     %freeze:_(s2) = G_FREEZE %x
     %ext:_(s64) = G_ZEXT %freeze
     $x0 = COPY %ext(s64)
+...
+---
+name:            test_freeze_v4s1
+body: |
+  bb.0.entry:
+    liveins: $q0
+    ; CHECK-LABEL: name: test_freeze_v4s1
+    ; CHECK: liveins: $q0
+    ; CHECK-NEXT: {{  $}}
+    ; CHECK-NEXT: [[DEF:%[0-9]+]]:_(<4 x s16>) = G_IMPLICIT_DEF
+    ; CHECK-NEXT: [[FREEZE:%[0-9]+]]:_(<4 x s16>) = G_FREEZE [[DEF]]
+    ; CHECK-NEXT: [[ANYEXT:%[0-9]+]]:_(<4 x s32>) = G_ANYEXT [[FREEZE]](<4 x s16>)
+    ; CHECK-NEXT: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 1
+    ; CHECK-NEXT: [[BUILD_VECTOR:%[0-9]+]]:_(<4 x s32>) = G_BUILD_VECTOR [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32)
+    ; CHECK-NEXT: %ext:_(<4 x s32>) = G_AND [[ANYEXT]], [[BUILD_VECTOR]]
+    ; CHECK-NEXT: $q0 = COPY %ext(<4 x s32>)
+    %x:_(<4 x s1>) = G_IMPLICIT_DEF
+    %freeze:_(<4 x s1>) = G_FREEZE %x
+    %ext:_(<4 x s32>) = G_ZEXT %freeze
+    $q0 = COPY %ext(<4 x s32>)
+...
+---
+name:            test_freeze_v3s8
+body: |
+  bb.0.entry:
+    liveins: $q0
+    ; CHECK-LABEL: name: test_freeze_v3s8
+    ; CHECK: liveins: $q0
+    ; CHECK-NEXT: {{  $}}
+    ; CHECK-NEXT: [[DEF:%[0-9]+]]:_(<4 x s16>) = G_IMPLICIT_DEF
+    ; CHECK-NEXT: [[UV:%[0-9]+]]:_(s16), [[UV1:%[0-9]+]]:_(s16), [[UV2:%[0-9]+]]:_(s16), [[UV3:%[0-9]+]]:_(s16) = G_UNMERGE_VALUES [[DEF]](<4 x s16>)
+    ; CHECK-NEXT: [[TRUNC:%[0-9]+]]:_(s8) = G_TRUNC [[UV]](s16)
+    ; CHECK-NEXT: [[TRUNC1:%[0-9]+]]:_(s8) = G_TRUNC [[UV1]](s16)
+    ; CHECK-NEXT: [[TRUNC2:%[0-9]+]]:_(s8) = G_TRUNC [[UV2]](s16)
+    ; CHECK-NEXT: [[DEF1:%[0-9]+]]:_(s8) = G_IMPLICIT_DEF
+    ; CHECK-NEXT: [[BUILD_VECTOR:%[0-9]+]]:_(<8 x s8>) = G_BUILD_VECTOR [[TRUNC]](s8), [[TRUNC1]](s8), [[TRUNC2]](s8), [[DEF1]](s8), [[DEF1]](s8), [[DEF1]](s8), [[DEF1]](s8), [[DEF1]](s8)
+    ; CHECK-NEXT: [[ANYEXT:%[0-9]+]]:_(<8 x s16>) = G_ANYEXT [[BUILD_VECTOR]](<8 x s8>)
+    ; CHECK-NEXT: [[UV4:%[0-9]+]]:_(<4 x s16>), [[UV5:%[0-9]+]]:_(<4 x s16>) = G_UNMERGE_VALUES [[ANYEXT]](<8 x s16>)
+    ; CHECK-NEXT: [[FREEZE:%[0-9]+]]:_(<4 x s16>) = G_FREEZE [[UV4]]
+    ; CHECK-NEXT: [[UV6:%[0-9]+]]:_(s16), [[UV7:%[0-9]+]]:_(s16), [[UV8:%[0-9]+]]:_(s16), [[UV9:%[0-9]+]]:_(s16) = G_UNMERGE_VALUES [[FREEZE]](<4 x s16>)
+    ; CHECK-NEXT: %undef:_(s32) = G_IMPLICIT_DEF
+    ; CHECK-NEXT: [[ANYEXT1:%[0-9]+]]:_(s32) = G_ANYEXT [[UV6]](s16)
+    ; CHECK-NEXT: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 255
+    ; CHECK-NEXT: %ext0:_(s32) = G_AND [[ANYEXT1]], [[C]]
+    ; CHECK-NEXT: [[ANYEXT2:%[0-9]+]]:_(s32) = G_ANYEXT [[UV7]](s16)
+    ; CHECK-NEXT: %ext1:_(s32) = G_AND [[ANYEXT2]], [[C]]
+    ; CHECK-NEXT: [[ANYEXT3:%[0-9]+]]:_(s32) = G_ANYEXT [[UV8]](s16)
+    ; CHECK-NEXT: %ext2:_(s32) = G_AND [[ANYEXT3]], [[C]]
+    ; CHECK-NEXT: %res:_(<4 x s32>) = G_BUILD_VECTOR %ext0(s32), %ext1(s32), %ext2(s32), %undef(s32)
+    ; CHECK-NEXT: $q0 = COPY %res(<4 x s32>)
+    %x:_(<3 x s8>) = G_IMPLICIT_DEF
+    %freeze:_(<3 x s8>) = G_FREEZE %x
+    %ext:_(<3 x s32>) = G_ZEXT %freeze
+    %undef:_(s32) = G_IMPLICIT_DEF
+    %ext0:_(s32), %ext1:_(s32), %ext2:_(s32) = G_UNMERGE_VALUES %ext
+    %res:_(<4 x s32>) = G_BUILD_VECTOR %ext0, %ext1, %ext2, %undef
+    $q0 = COPY %res(<4 x s32>)
+...
+---
+name:            test_freeze_v4s1_select
+body: |
+  bb.0.entry:
+    liveins: $q0, $q1
+    ; CHECK-LABEL: name: test_freeze_v4s1_select
+    ; CHECK: liveins: $q0, $q1
+    ; CHECK-NEXT: {{  $}}
+    ; CHECK-NEXT: [[COPY:%[0-9]+]]:_(<4 x s32>) = COPY $q0
+    ; CHECK-NEXT: [[COPY1:%[0-9]+]]:_(<4 x s32>) = COPY $q1
+    ; CHECK-NEXT: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 0
+    ; CHECK-NEXT: [[BUILD_VECTOR:%[0-9]+]]:_(<4 x s32>) = G_BUILD_VECTOR [[C]](s32), [[C]](s32), [[C]](s32), [[C]](s32)
+    ; CHECK-NEXT: [[FCMP:%[0-9]+]]:_(<4 x s32>) = nnan ninf nsz arcp contract afn reassoc G_FCMP floatpred(olt), [[COPY]](<4 x s32>), [[BUILD_VECTOR]]
+    ; CHECK-NEXT: [[FCMP1:%[0-9]+]]:_(<4 x s32>) = nnan ninf nsz arcp contract afn reassoc G_FCMP floatpred(ogt), [[COPY1]](<4 x s32>), [[COPY]]
+    ; CHECK-NEXT: [[TRUNC:%[0-9]+]]:_(<4 x s16>) = G_TRUNC [[FCMP1]](<4 x s32>)
+    ; CHECK-NEXT: [[FREEZE:%[0-9]+]]:_(<4 x s16>) = G_FREEZE [[TRUNC]]
+    ; CHECK-NEXT: [[TRUNC1:%[0-9]+]]:_(<4 x s16>) = G_TRUNC [[FCMP]](<4 x s32>)
+    ; CHECK-NEXT: [[AND:%[0-9]+]]:_(<4 x s16>) = G_AND [[TRUNC1]], [[FREEZE]]
+    ; CHECK-NEXT: [[ANYEXT:%[0-9]+]]:_(<4 x s32>) = G_ANYEXT [[AND]](<4 x s16>)
+    ; CHECK-NEXT: [[C1:%[0-9]+]]:_(s32) = G_CONSTANT i32 1
+    ; CHECK-NEXT: [[BUILD_VECTOR1:%[0-9]+]]:_(<4 x s32>) = G_BUILD_VECTOR [[C1]](s32), [[C1]](s32), [[C1]](s32), [[C1]](s32)
+    ; CHECK-NEXT: [[AND1:%[0-9]+]]:_(<4 x s32>) = G_AND [[ANYEXT]], [[BUILD_VECTOR1]]
+    ; CHECK-NEXT: $q0 = COPY [[AND1]](<4 x s32>)
+    %1:_(<4 x s32>) = COPY $q0
+    %2:_(<4 x s32>) = COPY $q1
+    %3:_(s32) = G_CONSTANT i32 0
+    %4:_(<4 x s32>) = G_BUILD_VECTOR %3, %3, %3, %3
+    %5:_(s1) = G_CONSTANT i1 false
+    %6:_(<4 x s1>) = nnan ninf nsz arcp contract afn reassoc G_FCMP floatpred(olt), %1:_(<4 x s32>), %4:_
+    %7:_(<4 x s1>) = nnan ninf nsz arcp contract afn reassoc G_FCMP floatpred(ogt), %2:_(<4 x s32>), %1:_
+    %8:_(<4 x s1>) = G_FREEZE %7
+    %9:_(<4 x s1>) = G_AND %6, %8
+    %10:_(<4 x s32>) = G_ZEXT %9
+    $q0 = COPY %10
diff --git a/llvm/test/CodeGen/AArch64/GlobalISel/legalize-insert-vector-elt.mir b/llvm/test/CodeGen/AArch64/GlobalISel/legalize-insert-vector-elt.mir
index a74bf9a5438b..9a8697c1d9b8 100644
--- a/llvm/test/CodeGen/AArch64/GlobalISel/legalize-insert-vector-elt.mir
+++ b/llvm/test/CodeGen/AArch64/GlobalISel/legalize-insert-vector-elt.mir
@@ -248,19 +248,24 @@ body:             |
   ; CHECK-NEXT:   [[TRUNC3:%[0-9]+]]:_(s8) = G_TRUNC [[UV2]](s16)
   ; CHECK-NEXT:   [[TRUNC4:%[0-9]+]]:_(s8) = G_TRUNC [[UV3]](s16)
   ; CHECK-NEXT:   [[TRUNC5:%[0-9]+]]:_(s8) = G_TRUNC [[UV4]](s16)
-  ; CHECK-NEXT:   [[BUILD_VECTOR1:%[0-9]+]]:_(<16 x s8>) = G_BUILD_VECTOR [[TRUNC3]](s8), [[TRUNC4]](s8), [[TRUNC5]](s8), [[DEF1]](s8), [[DEF1]](s8), [[DEF1]](s8), [[DEF1]](s8), [[DEF1]](s8), [[DEF1]](s8), [[DEF1]](s8), [[DEF1]](s8), [[DEF1]](s8), [[DEF1]](s8), [[DEF1]](s8), [[DEF1]](s8), [[DEF1]](s8)
-  ; CHECK-NEXT:   [[BUILD_VECTOR2:%[0-9]+]]:_(<16 x s8>) = G_BUILD_VECTOR [[C]](s8), [[DEF]](s8), [[DEF]](s8), [[DEF1]](s8), [[DEF1]](s8), [[DEF1]](s8), [[DEF1]](s8), [[DEF1]](s8), [[DEF1]](s8), [[DEF1]](s8), [[DEF1]](s8), [[DEF1]](s8), [[DEF1]](s8), [[DEF1]](s8), [[DEF1]](s8), [[DEF1]](s8)
+  ; CHECK-NEXT:   [[DEF2:%[0-9]+]]:_(<4 x s16>) = G_IMPLICIT_DEF
+  ; CHECK-NEXT:   [[UV6:%[0-9]+]]:_(s16), [[UV7:%[0-9]+]]:_(s16), [[UV8:%[0-9]+]]:_(s16), [[UV9:%[0-9]+]]:_(s16) = G_UNMERGE_VALUES [[DEF2]](<4 x s16>)
+  ; CHECK-NEXT:   [[TRUNC6:%[0-9]+]]:_(s8) = G_TRUNC [[UV6]](s16)
+  ; CHECK-NEXT:   [[TRUNC7:%[0-9]+]]:_(s8) = G_TRUNC [[UV7]](s16)
+  ; CHECK-NEXT:   [[TRUNC8:%[0-9]+]]:_(s8) = G_TRUNC [[UV8]](s16)
+  ; CHECK-NEXT:   [[BUILD_VECTOR1:%[0-9]+]]:_(<16 x s8>) = G_BUILD_VECTOR [[TRUNC3]](s8), [[TRUNC4]](s8), [[TRUNC5]](s8), [[TRUNC6]](s8), [[TRUNC7]](s8), [[TRUNC8]](s8), [[TRUNC6]](s8), [[TRUNC7]](s8), [[TRUNC8]](s8), [[TRUNC6]](s8), [[TRUNC7]](s8), [[TRUNC8]](s8), [[DEF1]](s8), [[DEF1]](s8), [[DEF1]](s8), [[DEF1]](s8)
+  ; CHECK-NEXT:   [[BUILD_VECTOR2:%[0-9]+]]:_(<16 x s8>) = G_BUILD_VECTOR [[C]](s8), [[DEF]](s8), [[DEF]](s8), [[TRUNC6]](s8), [[TRUNC7]](s8), [[TRUNC8]](s8), [[TRUNC6]](s8), [[TRUNC7]](s8), [[TRUNC8]](s8), [[TRUNC6]](s8), [[TRUNC7]](s8), [[TRUNC8]](s8), [[DEF1]](s8), [[DEF1]](s8), [[DEF1]](s8), [[DEF1]](s8)
   ; CHECK-NEXT:   [[SHUF:%[0-9]+]]:_(<16 x s8>) = G_SHUFFLE_VECTOR [[BUILD_VECTOR1]](<16 x s8>), [[BUILD_VECTOR2]], shufflemask(0, 16, 16, 16, 1, 16, 16, 16, 2, 16, 16, 16, undef, undef, undef, undef)
   ; CHECK-NEXT:   [[BITCAST:%[0-9]+]]:_(<4 x s32>) = G_BITCAST [[SHUF]](<16 x s8>)
   ; CHECK-NEXT:   [[UITOFP:%[0-9]+]]:_(<4 x s32>) = G_UITOFP [[BITCAST]](<4 x s32>)
-  ; CHECK-NEXT:   [[UV6:%[0-9]+]]:_(s32), [[UV7:%[0-9]+]]:_(s32), [[UV8:%[0-9]+]]:_(s32), [[UV9:%[0-9]+]]:_(s32) = G_UNMERGE_VALUES [[UITOFP]](<4 x s32>)
-  ; CHECK-NEXT:   G_STORE [[UV6]](s32), [[COPY]](p0) :: (store (s32), align 16)
+  ; CHECK-NEXT:   [[UV10:%[0-9]+]]:_(s32), [[UV11:%[0-9]+]]:_(s32), [[UV12:%[0-9]+]]:_(s32), [[UV13:%[0-9]+]]:_(s32) = G_UNMERGE_VALUES [[UITOFP]](<4 x s32>)
+  ; CHECK-NEXT:   G_STORE [[UV10]](s32), [[COPY]](p0) :: (store (s32), align 16)
   ; CHECK-NEXT:   [[C3:%[0-9]+]]:_(s64) = G_CONSTANT i64 4
   ; CHECK-NEXT:   [[PTR_ADD:%[0-9]+]]:_(p0) = G_PTR_ADD [[COPY]], [[C3]](s64)
-  ; CHECK-NEXT:   G_STORE [[UV7]](s32), [[PTR_ADD]](p0) :: (store (s32) into unknown-address + 4)
+  ; CHECK-NEXT:   G_STORE [[UV11]](s32), [[PTR_ADD]](p0) :: (store (s32) into unknown-address + 4)
   ; CHECK-NEXT:   [[C4:%[0-9]+]]:_(s64) = G_CONSTANT i64 8
   ; CHECK-NEXT:   [[PTR_ADD1:%[0-9]+]]:_(p0) = G_PTR_ADD [[COPY]], [[C4]](s64)
-  ; CHECK-NEXT:   G_STORE [[UV8]](s32), [[PTR_ADD1]](p0) :: (store (s32) into unknown-address + 8, align 8)
+  ; CHECK-NEXT:   G_STORE [[UV12]](s32), [[PTR_ADD1]](p0) :: (store (s32) into unknown-address + 8, align 8)
   ; CHECK-NEXT:   G_BR %bb.1
   bb.1:
     liveins: $w1, $w2, $w3, $x0
diff --git a/llvm/test/CodeGen/AArch64/GlobalISel/legalize-itofp.mir b/llvm/test/CodeGen/AArch64/GlobalISel/legalize-itofp.mir
index fc417b2eca61..69fe50e03df0 100644
--- a/llvm/test/CodeGen/AArch64/GlobalISel/legalize-itofp.mir
+++ b/llvm/test/CodeGen/AArch64/GlobalISel/legalize-itofp.mir
@@ -274,12 +274,10 @@ body: |
     ; CHECK-LABEL: name: test_uitofp_v2s64_v2i1
     ; CHECK: liveins: $q0
     ; CHECK-NEXT: {{  $}}
-    ; CHECK-NEXT: [[DEF:%[0-9]+]]:_(s64) = G_IMPLICIT_DEF
-    ; CHECK-NEXT: [[COPY:%[0-9]+]]:_(s64) = COPY [[DEF]](s64)
-    ; CHECK-NEXT: [[BUILD_VECTOR:%[0-9]+]]:_(<2 x s64>) = G_BUILD_VECTOR [[DEF]](s64), [[COPY]](s64)
+    ; CHECK-NEXT: [[DEF:%[0-9]+]]:_(<2 x s64>) = G_IMPLICIT_DEF
     ; CHECK-NEXT: [[C:%[0-9]+]]:_(s64) = G_CONSTANT i64 1
-    ; CHECK-NEXT: [[BUILD_VECTOR1:%[0-9]+]]:_(<2 x s64>) = G_BUILD_VECTOR [[C]](s64), [[C]](s64)
-    ; CHECK-NEXT: [[AND:%[0-9]+]]:_(<2 x s64>) = G_AND [[BUILD_VECTOR]], [[BUILD_VECTOR1]]
+    ; CHECK-NEXT: [[BUILD_VECTOR:%[0-9]+]]:_(<2 x s64>) = G_BUILD_VECTOR [[C]](s64), [[C]](s64)
+    ; CHECK-NEXT: [[AND:%[0-9]+]]:_(<2 x s64>) = G_AND [[DEF]], [[BUILD_VECTOR]]
     ; CHECK-NEXT: [[UITOFP:%[0-9]+]]:_(<2 x s64>) = G_UITOFP [[AND]](<2 x s64>)
     ; CHECK-NEXT: $q0 = COPY [[UITOFP]](<2 x s64>)
     %0:_(<2 x s1>) = G_IMPLICIT_DEF
@@ -295,10 +293,8 @@ body: |
     ; CHECK-LABEL: name: test_sitofp_v2s64_v2i1
     ; CHECK: liveins: $q0
     ; CHECK-NEXT: {{  $}}
-    ; CHECK-NEXT: [[DEF:%[0-9]+]]:_(s64) = G_IMPLICIT_DEF
-    ; CHECK-NEXT: [[COPY:%[0-9]+]]:_(s64) = COPY [[DEF]](s64)
-    ; CHECK-NEXT: [[BUILD_VECTOR:%[0-9]+]]:_(<2 x s64>) = G_BUILD_VECTOR [[DEF]](s64), [[COPY]](s64)
-    ; CHECK-NEXT: [[SEXT_INREG:%[0-9]+]]:_(<2 x s64>) = G_SEXT_INREG [[BUILD_VECTOR]], 1
+    ; CHECK-NEXT: [[DEF:%[0-9]+]]:_(<2 x s64>) = G_IMPLICIT_DEF
+    ; CHECK-NEXT: [[SEXT_INREG:%[0-9]+]]:_(<2 x s64>) = G_SEXT_INREG [[DEF]], 1
     ; CHECK-NEXT: [[SITOFP:%[0-9]+]]:_(<2 x s64>) = G_SITOFP [[SEXT_INREG]](<2 x s64>)
     ; CHECK-NEXT: $q0 = COPY [[SITOFP]](<2 x s64>)
     %0:_(<2 x s1>) = G_IMPLICIT_DEF
diff --git a/llvm/test/CodeGen/AArch64/GlobalISel/legalize-load-store.mir b/llvm/test/CodeGen/AArch64/GlobalISel/legalize-load-store.mir
index b8328eda9a66..94bdcf74a8aa 100644
--- a/llvm/test/CodeGen/AArch64/GlobalISel/legalize-load-store.mir
+++ b/llvm/test/CodeGen/AArch64/GlobalISel/legalize-load-store.mir
@@ -328,14 +328,12 @@ body:             |
     ; CHECK-LABEL: name: store_32xs8
     ; CHECK: liveins: $x0
     ; CHECK-NEXT: {{  $}}
-    ; CHECK-NEXT: [[DEF:%[0-9]+]]:_(s8) = G_IMPLICIT_DEF
-    ; CHECK-NEXT: [[BUILD_VECTOR:%[0-9]+]]:_(<16 x s8>) = G_BUILD_VECTOR [[DEF]](s8), [[DEF]](s8), [[DEF]](s8), [[DEF]](s8), [[DEF]](s8), [[DEF]](s8), [[DEF]](s8), [[DEF]](s8), [[DEF]](s8), [[DEF]](s8), [[DEF]](s8), [[DEF]](s8), [[DEF]](s8), [[DEF]](s8), [[DEF]](s8), [[DEF]](s8)
-    ; CHECK-NEXT: [[BUILD_VECTOR1:%[0-9]+]]:_(<16 x s8>) = G_BUILD_VECTOR [[DEF]](s8), [[DEF]](s8), [[DEF]](s8), [[DEF]](s8), [[DEF]](s8), [[DEF]](s8), [[DEF]](s8), [[DEF]](s8), [[DEF]](s8), [[DEF]](s8), [[DEF]](s8), [[DEF]](s8), [[DEF]](s8), [[DEF]](s8), [[DEF]](s8), [[DEF]](s8)
+    ; CHECK-NEXT: [[DEF:%[0-9]+]]:_(<16 x s8>) = G_IMPLICIT_DEF
     ; CHECK-NEXT: %ptr:_(p0) = COPY $x0
-    ; CHECK-NEXT: G_STORE [[BUILD_VECTOR]](<16 x s8>), %ptr(p0) :: (store (<16 x s8>), align 32)
+    ; CHECK-NEXT: G_STORE [[DEF]](<16 x s8>), %ptr(p0) :: (store (<16 x s8>), align 32)
     ; CHECK-NEXT: [[C:%[0-9]+]]:_(s64) = G_CONSTANT i64 16
     ; CHECK-NEXT: [[PTR_ADD:%[0-9]+]]:_(p0) = G_PTR_ADD %ptr, [[C]](s64)
-    ; CHECK-NEXT: G_STORE [[BUILD_VECTOR1]](<16 x s8>), [[PTR_ADD]](p0) :: (store (<16 x s8>) into unknown-address + 16)
+    ; CHECK-NEXT: G_STORE [[DEF]](<16 x s8>), [[PTR_ADD]](p0) :: (store (<16 x s8>) into unknown-address + 16)
     ; CHECK-NEXT: RET_ReallyLR
     %val:_(<32 x s8>) = G_IMPLICIT_DEF
     %ptr:_(p0) = COPY $x0
@@ -353,14 +351,12 @@ body:             |
     ; CHECK-LABEL: name: store_16xs16
     ; CHECK: liveins: $x0
     ; CHECK-NEXT: {{  $}}
-    ; CHECK-NEXT: [[DEF:%[0-9]+]]:_(s16) = G_IMPLICIT_DEF
-    ; CHECK-NEXT: [[BUILD_VECTOR:%[0-9]+]]:_(<8 x s16>) = G_BUILD_VECTOR [[DEF]](s16), [[DEF]](s16), [[DEF]](s16), [[DEF]](s16), [[DEF]](s16), [[DEF]](s16), [[DEF]](s16), [[DEF]](s16)
-    ; CHECK-NEXT: [[BUILD_VECTOR1:%[0-9]+]]:_(<8 x s16>) = G_BUILD_VECTOR [[DEF]](s16), [[DEF]](s16), [[DEF]](s16), [[DEF]](s16), [[DEF]](s16), [[DEF]](s16), [[DEF]](s16), [[DEF]](s16)
+    ; CHECK-NEXT: [[DEF:%[0-9]+]]:_(<8 x s16>) = G_IMPLICIT_DEF
     ; CHECK-NEXT: %ptr:_(p0) = COPY $x0
-    ; CHECK-NEXT: G_STORE [[BUILD_VECTOR]](<8 x s16>), %ptr(p0) :: (store (<8 x s16>), align 32)
+    ; CHECK-NEXT: G_STORE [[DEF]](<8 x s16>), %ptr(p0) :: (store (<8 x s16>), align 32)
     ; CHECK-NEXT: [[C:%[0-9]+]]:_(s64) = G_CONSTANT i64 16
     ; CHECK-NEXT: [[PTR_ADD:%[0-9]+]]:_(p0) = G_PTR_ADD %ptr, [[C]](s64)
-    ; CHECK-NEXT: G_STORE [[BUILD_VECTOR1]](<8 x s16>), [[PTR_ADD]](p0) :: (store (<8 x s16>) into unknown-address + 16)
+    ; CHECK-NEXT: G_STORE [[DEF]](<8 x s16>), [[PTR_ADD]](p0) :: (store (<8 x s16>) into unknown-address + 16)
     ; CHECK-NEXT: RET_ReallyLR
     %val:_(<16 x s16>) = G_IMPLICIT_DEF
     %ptr:_(p0) = COPY $x0
@@ -378,14 +374,12 @@ body:             |
     ; CHECK-LABEL: name: store_8xs32
     ; CHECK: liveins: $x0
     ; CHECK-NEXT: {{  $}}
-    ; CHECK-NEXT: [[DEF:%[0-9]+]]:_(s32) = G_IMPLICIT_DEF
-    ; CHECK-NEXT: [[BUILD_VECTOR:%[0-9]+]]:_(<4 x s32>) = G_BUILD_VECTOR [[DEF]](s32), [[DEF]](s32), [[DEF]](s32), [[DEF]](s32)
-    ; CHECK-NEXT: [[BUILD_VECTOR1:%[0-9]+]]:_(<4 x s32>) = G_BUILD_VECTOR [[DEF]](s32), [[DEF]](s32), [[DEF]](s32), [[DEF]](s32)
+    ; CHECK-NEXT: [[DEF:%[0-9]+]]:_(<4 x s32>) = G_IMPLICIT_DEF
     ; CHECK-NEXT: %ptr:_(p0) = COPY $x0
-    ; CHECK-NEXT: G_STORE [[BUILD_VECTOR]](<4 x s32>), %ptr(p0) :: (store (<4 x s32>), align 32)
+    ; CHECK-NEXT: G_STORE [[DEF]](<4 x s32>), %ptr(p0) :: (store (<4 x s32>), align 32)
     ; CHECK-NEXT: [[C:%[0-9]+]]:_(s64) = G_CONSTANT i64 16
     ; CHECK-NEXT: [[PTR_ADD:%[0-9]+]]:_(p0) = G_PTR_ADD %ptr, [[C]](s64)
-    ; CHECK-NEXT: G_STORE [[BUILD_VECTOR1]](<4 x s32>), [[PTR_ADD]](p0) :: (store (<4 x s32>) into unknown-address + 16)
+    ; CHECK-NEXT: G_STORE [[DEF]](<4 x s32>), [[PTR_ADD]](p0) :: (store (<4 x s32>) into unknown-address + 16)
     ; CHECK-NEXT: RET_ReallyLR
     %val:_(<8 x s32>) = G_IMPLICIT_DEF
     %ptr:_(p0) = COPY $x0
diff --git a/llvm/test/CodeGen/AArch64/GlobalISel/legalize-min-max.mir b/llvm/test/CodeGen/AArch64/GlobalISel/legalize-min-max.mir
index cab8a91e447e..35c9538627b3 100644
--- a/llvm/test/CodeGen/AArch64/GlobalISel/legalize-min-max.mir
+++ b/llvm/test/CodeGen/AArch64/GlobalISel/legalize-min-max.mir
@@ -55,10 +55,9 @@ body: |
     ; CHECK-LABEL: name: v32s8_smin
     ; CHECK: liveins: $x0, $q0, $q1
     ; CHECK-NEXT: {{  $}}
-    ; CHECK-NEXT: [[DEF:%[0-9]+]]:_(s8) = G_IMPLICIT_DEF
-    ; CHECK-NEXT: [[BUILD_VECTOR:%[0-9]+]]:_(<16 x s8>) = G_BUILD_VECTOR [[DEF]](s8), [[DEF]](s8), [[DEF]](s8), [[DEF]](s8), [[DEF]](s8), [[DEF]](s8), [[DEF]](s8), [[DEF]](s8), [[DEF]](s8), [[DEF]](s8), [[DEF]](s8), [[DEF]](s8), [[DEF]](s8), [[DEF]](s8), [[DEF]](s8), [[DEF]](s8)
-    ; CHECK-NEXT: [[SMIN:%[0-9]+]]:_(<16 x s8>) = G_SMIN [[BUILD_VECTOR]], [[BUILD_VECTOR]]
-    ; CHECK-NEXT: [[SMIN1:%[0-9]+]]:_(<16 x s8>) = G_SMIN [[BUILD_VECTOR]], [[BUILD_VECTOR]]
+    ; CHECK-NEXT: [[DEF:%[0-9]+]]:_(<16 x s8>) = G_IMPLICIT_DEF
+    ; CHECK-NEXT: [[SMIN:%[0-9]+]]:_(<16 x s8>) = G_SMIN [[DEF]], [[DEF]]
+    ; CHECK-NEXT: [[SMIN1:%[0-9]+]]:_(<16 x s8>) = G_SMIN [[DEF]], [[DEF]]
     ; CHECK-NEXT: [[COPY:%[0-9]+]]:_(p0) = COPY $x0
     ; CHECK-NEXT: G_STORE [[SMIN]](<16 x s8>), [[COPY]](p0) :: (store (<16 x s8>), align 32)
     ; CHECK-NEXT: [[C:%[0-9]+]]:_(s64) = G_CONSTANT i64 16
@@ -125,10 +124,9 @@ body: |
     ; CHECK-LABEL: name: v16s16_smin
     ; CHECK: liveins: $x0, $q0, $q1
     ; CHECK-NEXT: {{  $}}
-    ; CHECK-NEXT: [[DEF:%[0-9]+]]:_(s16) = G_IMPLICIT_DEF
-    ; CHECK-NEXT: [[BUILD_VECTOR:%[0-9]+]]:_(<8 x s16>) = G_BUILD_VECTOR [[DEF]](s16), [[DEF]](s16), [[DEF]](s16), [[DEF]](s16), [[DEF]](s16), [[DEF]](s16), [[DEF]](s16), [[DEF]](s16)
-    ; CHECK-NEXT: [[SMIN:%[0-9]+]]:_(<8 x s16>) = G_SMIN [[BUILD_VECTOR]], [[BUILD_VECTOR]]
-    ; CHECK-NEXT: [[SMIN1:%[0-9]+]]:_(<8 x s16>) = G_SMIN [[BUILD_VECTOR]], [[BUILD_VECTOR]]
+    ; CHECK-NEXT: [[DEF:%[0-9]+]]:_(<8 x s16>) = G_IMPLICIT_DEF
+    ; CHECK-NEXT: [[SMIN:%[0-9]+]]:_(<8 x s16>) = G_SMIN [[DEF]], [[DEF]]
+    ; CHECK-NEXT: [[SMIN1:%[0-9]+]]:_(<8 x s16>) = G_SMIN [[DEF]], [[DEF]]
     ; CHECK-NEXT: [[COPY:%[0-9]+]]:_(p0) = COPY $x0
     ; CHECK-NEXT: G_STORE [[SMIN]](<8 x s16>), [[COPY]](p0) :: (store (<8 x s16>), align 32)
     ; CHECK-NEXT: [[C:%[0-9]+]]:_(s64) = G_CONSTANT i64 16
@@ -195,10 +193,9 @@ body: |
     ; CHECK-LABEL: name: v8s32_smin
     ; CHECK: liveins: $x0, $q0, $q1
     ; CHECK-NEXT: {{  $}}
-    ; CHECK-NEXT: [[DEF:%[0-9]+]]:_(s32) = G_IMPLICIT_DEF
-    ; CHECK-NEXT: [[BUILD_VECTOR:%[0-9]+]]:_(<4 x s32>) = G_BUILD_VECTOR [[DEF]](s32), [[DEF]](s32), [[DEF]](s32), [[DEF]](s32)
-    ; CHECK-NEXT: [[SMIN:%[0-9]+]]:_(<4 x s32>) = G_SMIN [[BUILD_VECTOR]], [[BUILD_VECTOR]]
-    ; CHECK-NEXT: [[SMIN1:%[0-9]+]]:_(<4 x s32>) = G_SMIN [[BUILD_VECTOR]], [[BUILD_VECTOR]]
+    ; CHECK-NEXT: [[DEF:%[0-9]+]]:_(<4 x s32>) = G_IMPLICIT_DEF
+    ; CHECK-NEXT: [[SMIN:%[0-9]+]]:_(<4 x s32>) = G_SMIN [[DEF]], [[DEF]]
+    ; CHECK-NEXT: [[SMIN1:%[0-9]+]]:_(<4 x s32>) = G_SMIN [[DEF]], [[DEF]]
     ; CHECK-NEXT: [[COPY:%[0-9]+]]:_(p0) = COPY $x0
     ; CHECK-NEXT: G_STORE [[SMIN]](<4 x s32>), [[COPY]](p0) :: (store (<4 x s32>), align 32)
     ; CHECK-NEXT: [[C:%[0-9]+]]:_(s64) = G_CONSTANT i64 16
@@ -331,10 +328,9 @@ body: |
     ; CHECK-LABEL: name: v32s8_umin
     ; CHECK: liveins: $x0, $q0, $q1
     ; CHECK-NEXT: {{  $}}
-    ; CHECK-NEXT: [[DEF:%[0-9]+]]:_(s8) = G_IMPLICIT_DEF
-    ; CHECK-NEXT: [[BUILD_VECTOR:%[0-9]+]]:_(<16 x s8>) = G_BUILD_VECTOR [[DEF]](s8), [[DEF]](s8), [[DEF]](s8), [[DEF]](s8), [[DEF]](s8), [[DEF]](s8), [[DEF]](s8), [[DEF]](s8), [[DEF]](s8), [[DEF]](s8), [[DEF]](s8), [[DEF]](s8), [[DEF]](s8), [[DEF]](s8), [[DEF]](s8), [[DEF]](s8)
-    ; CHECK-NEXT: [[UMIN:%[0-9]+]]:_(<16 x s8>) = G_UMIN [[BUILD_VECTOR]], [[BUILD_VECTOR]]
-    ; CHECK-NEXT: [[UMIN1:%[0-9]+]]:_(<16 x s8>) = G_UMIN [[BUILD_VECTOR]], [[BUILD_VECTOR]]
+    ; CHECK-NEXT: [[DEF:%[0-9]+]]:_(<16 x s8>) = G_IMPLICIT_DEF
+    ; CHECK-NEXT: [[UMIN:%[0-9]+]]:_(<16 x s8>) = G_UMIN [[DEF]], [[DEF]]
+    ; CHECK-NEXT: [[UMIN1:%[0-9]+]]:_(<16 x s8>) = G_UMIN [[DEF]], [[DEF]]
     ; CHECK-NEXT: [[COPY:%[0-9]+]]:_(p0) = COPY $x0
     ; CHECK-NEXT: G_STORE [[UMIN]](<16 x s8>), [[COPY]](p0) :: (store (<16 x s8>), align 32)
     ; CHECK-NEXT: [[C:%[0-9]+]]:_(s64) = G_CONSTANT i64 16
@@ -401,10 +397,9 @@ body: |
     ; CHECK-LABEL: name: v16s16_umin
     ; CHECK: liveins: $x0, $q0, $q1
     ; CHECK-NEXT: {{  $}}
-    ; CHECK-NEXT: [[DEF:%[0-9]+]]:_(s16) = G_IMPLICIT_DEF
-    ; CHECK-NEXT: [[BUILD_VECTOR:%[0-9]+]]:_(<8 x s16>) = G_BUILD_VECTOR [[DEF]](s16), [[DEF]](s16), [[DEF]](s16), [[DEF]](s16), [[DEF]](s16), [[DEF]](s16), [[DEF]](s16), [[DEF]](s16)
-    ; CHECK-NEXT: [[UMIN:%[0-9]+]]:_(<8 x s16>) = G_UMIN [[BUILD_VECTOR]], [[BUILD_VECTOR]]
-    ; CHECK-NEXT: [[UMIN1:%[0-9]+]]:_(<8 x s16>) = G_UMIN [[BUILD_VECTOR]], [[BUILD_VECTOR]]
+    ; CHECK-NEXT: [[DEF:%[0-9]+]]:_(<8 x s16>) = G_IMPLICIT_DEF
+    ; CHECK-NEXT: [[UMIN:%[0-9]+]]:_(<8 x s16>) = G_UMIN [[DEF]], [[DEF]]
+    ; CHECK-NEXT: [[UMIN1:%[0-9]+]]:_(<8 x s16>) = G_UMIN [[DEF]], [[DEF]]
     ; CHECK-NEXT: [[COPY:%[0-9]+]]:_(p0) = COPY $x0
     ; CHECK-NEXT: G_STORE [[UMIN]](<8 x s16>), [[COPY]](p0) :: (store (<8 x s16>), align 32)
     ; CHECK-NEXT: [[C:%[0-9]+]]:_(s64) = G_CONSTANT i64 16
@@ -471,10 +466,9 @@ body: |
     ; CHECK-LABEL: name: v8s32_umin
     ; CHECK: liveins: $x0, $q0, $q1
     ; CHECK-NEXT: {{  $}}
-    ; CHECK-NEXT: [[DEF:%[0-9]+]]:_(s32) = G_IMPLICIT_DEF
-    ; CHECK-NEXT: [[BUILD_VECTOR:%[0-9]+]]:_(<4 x s32>) = G_BUILD_VECTOR [[DEF]](s32), [[DEF]](s32), [[DEF]](s32), [[DEF]](s32)
-    ; CHECK-NEXT: [[UMIN:%[0-9]+]]:_(<4 x s32>) = G_UMIN [[BUILD_VECTOR]], [[BUILD_VECTOR]]
-    ; CHECK-NEXT: [[UMIN1:%[0-9]+]]:_(<4 x s32>) = G_UMIN [[BUILD_VECTOR]], [[BUILD_VECTOR]]
+    ; CHECK-NEXT: [[DEF:%[0-9]+]]:_(<4 x s32>) = G_IMPLICIT_DEF
+    ; CHECK-NEXT: [[UMIN:%[0-9]+]]:_(<4 x s32>) = G_UMIN [[DEF]], [[DEF]]
+    ; CHECK-NEXT: [[UMIN1:%[0-9]+]]:_(<4 x s32>) = G_UMIN [[DEF]], [[DEF]]
     ; CHECK-NEXT: [[COPY:%[0-9]+]]:_(p0) = COPY $x0
     ; CHECK-NEXT: G_STORE [[UMIN]](<4 x s32>), [[COPY]](p0) :: (store (<4 x s32>), align 32)
     ; CHECK-NEXT: [[C:%[0-9]+]]:_(s64) = G_CONSTANT i64 16
@@ -629,10 +623,9 @@ body: |
     ; CHECK-LABEL: name: v32s8_smax
     ; CHECK: liveins: $x0, $q0, $q1
     ; CHECK-NEXT: {{  $}}
-    ; CHECK-NEXT: [[DEF:%[0-9]+]]:_(s8) = G_IMPLICIT_DEF
-    ; CHECK-NEXT: [[BUILD_VECTOR:%[0-9]+]]:_(<16 x s8>) = G_BUILD_VECTOR [[DEF]](s8), [[DEF]](s8), [[DEF]](s8), [[DEF]](s8), [[DEF]](s8), [[DEF]](s8), [[DEF]](s8), [[DEF]](s8), [[DEF]](s8), [[DEF]](s8), [[DEF]](s8), [[DEF]](s8), [[DEF]](s8), [[DEF]](s8), [[DEF]](s8), [[DEF]](s8)
-    ; CHECK-NEXT: [[SMAX:%[0-9]+]]:_(<16 x s8>) = G_SMAX [[BUILD_VECTOR]], [[BUILD_VECTOR]]
-    ; CHECK-NEXT: [[SMAX1:%[0-9]+]]:_(<16 x s8>) = G_SMAX [[BUILD_VECTOR]], [[BUILD_VECTOR]]
+    ; CHECK-NEXT: [[DEF:%[0-9]+]]:_(<16 x s8>) = G_IMPLICIT_DEF
+    ; CHECK-NEXT: [[SMAX:%[0-9]+]]:_(<16 x s8>) = G_SMAX [[DEF]], [[DEF]]
+    ; CHECK-NEXT: [[SMAX1:%[0-9]+]]:_(<16 x s8>) = G_SMAX [[DEF]], [[DEF]]
     ; CHECK-NEXT: [[COPY:%[0-9]+]]:_(p0) = COPY $x0
     ; CHECK-NEXT: G_STORE [[SMAX]](<16 x s8>), [[COPY]](p0) :: (store (<16 x s8>), align 32)
     ; CHECK-NEXT: [[C:%[0-9]+]]:_(s64) = G_CONSTANT i64 16
@@ -677,10 +670,9 @@ body: |
     ; CHECK-LABEL: name: v16s16_smax
     ; CHECK: liveins: $x0, $q0, $q1
     ; CHECK-NEXT: {{  $}}
-    ; CHECK-NEXT: [[DEF:%[0-9]+]]:_(s16) = G_IMPLICIT_DEF
-    ; CHECK-NEXT: [[BUILD_VECTOR:%[0-9]+]]:_(<8 x s16>) = G_BUILD_VECTOR [[DEF]](s16), [[DEF]](s16), [[DEF]](s16), [[DEF]](s16), [[DEF]](s16), [[DEF]](s16), [[DEF]](s16), [[DEF]](s16)
-    ; CHECK-NEXT: [[SMAX:%[0-9]+]]:_(<8 x s16>) = G_SMAX [[BUILD_VECTOR]], [[BUILD_VECTOR]]
-    ; CHECK-NEXT: [[SMAX1:%[0-9]+]]:_(<8 x s16>) = G_SMAX [[BUILD_VECTOR]], [[BUILD_VECTOR]]
+    ; CHECK-NEXT: [[DEF:%[0-9]+]]:_(<8 x s16>) = G_IMPLICIT_DEF
+    ; CHECK-NEXT: [[SMAX:%[0-9]+]]:_(<8 x s16>) = G_SMAX [[DEF]], [[DEF]]
+    ; CHECK-NEXT: [[SMAX1:%[0-9]+]]:_(<8 x s16>) = G_SMAX [[DEF]], [[DEF]]
     ; CHECK-NEXT: [[COPY:%[0-9]+]]:_(p0) = COPY $x0
     ; CHECK-NEXT: G_STORE [[SMAX]](<8 x s16>), [[COPY]](p0) :: (store (<8 x s16>), align 32)
     ; CHECK-NEXT: [[C:%[0-9]+]]:_(s64) = G_CONSTANT i64 16
@@ -747,10 +739,9 @@ body: |
     ; CHECK-LABEL: name: v8s32_smax
     ; CHECK: liveins: $x0, $q0, $q1
     ; CHECK-NEXT: {{  $}}
-    ; CHECK-NEXT: [[DEF:%[0-9]+]]:_(s32) = G_IMPLICIT_DEF
-    ; CHECK-NEXT: [[BUILD_VECTOR:%[0-9]+]]:_(<4 x s32>) = G_BUILD_VECTOR [[DEF]](s32), [[DEF]](s32), [[DEF]](s32), [[DEF]](s32)
-    ; CHECK-NEXT: [[SMAX:%[0-9]+]]:_(<4 x s32>) = G_SMAX [[BUILD_VECTOR]], [[BUILD_VECTOR]]
-    ; CHECK-NEXT: [[SMAX1:%[0-9]+]]:_(<4 x s32>) = G_SMAX [[BUILD_VECTOR]], [[BUILD_VECTOR]]
+    ; CHECK-NEXT: [[DEF:%[0-9]+]]:_(<4 x s32>) = G_IMPLICIT_DEF
+    ; CHECK-NEXT: [[SMAX:%[0-9]+]]:_(<4 x s32>) = G_SMAX [[DEF]], [[DEF]]
+    ; CHECK-NEXT: [[SMAX1:%[0-9]+]]:_(<4 x s32>) = G_SMAX [[DEF]], [[DEF]]
     ; CHECK-NEXT: [[COPY:%[0-9]+]]:_(p0) = COPY $x0
     ; CHECK-NEXT: G_STORE [[SMAX]](<4 x s32>), [[COPY]](p0) :: (store (<4 x s32>), align 32)
     ; CHECK-NEXT: [[C:%[0-9]+]]:_(s64) = G_CONSTANT i64 16
@@ -883,10 +874,9 @@ body: |
     ; CHECK-LABEL: name: v32s8_umax
     ; CHECK: liveins: $x0, $q0, $q1
     ; CHECK-NEXT: {{  $}}
-    ; CHECK-NEXT: [[DEF:%[0-9]+]]:_(s8) = G_IMPLICIT_DEF
-    ; CHECK-NEXT: [[BUILD_VECTOR:%[0-9]+]]:_(<16 x s8>) = G_BUILD_VECTOR [[DEF]](s8), [[DEF]](s8), [[DEF]](s8), [[DEF]](s8), [[DEF]](s8), [[DEF]](s8), [[DEF]](s8), [[DEF]](s8), [[DEF]](s8), [[DEF]](s8), [[DEF]](s8), [[DEF]](s8), [[DEF]](s8), [[DEF]](s8), [[DEF]](s8), [[DEF]](s8)
-    ; CHECK-NEXT: [[UMAX:%[0-9]+]]:_(<16 x s8>) = G_UMAX [[BUILD_VECTOR]], [[BUILD_VECTOR]]
-    ; CHECK-NEXT: [[UMAX1:%[0-9]+]]:_(<16 x s8>) = G_UMAX [[BUILD_VECTOR]], [[BUILD_VECTOR]]
+    ; CHECK-NEXT: [[DEF:%[0-9]+]]:_(<16 x s8>) = G_IMPLICIT_DEF
+    ; CHECK-NEXT: [[UMAX:%[0-9]+]]:_(<16 x s8>) = G_UMAX [[DEF]], [[DEF]]
+    ; CHECK-NEXT: [[UMAX1:%[0-9]+]]:_(<16 x s8>) = G_UMAX [[DEF]], [[DEF]]
     ; CHECK-NEXT: [[COPY:%[0-9]+]]:_(p0) = COPY $x0
     ; CHECK-NEXT: G_STORE [[UMAX]](<16 x s8>), [[COPY]](p0) :: (store (<16 x s8>), align 32)
     ; CHECK-NEXT: [[C:%[0-9]+]]:_(s64) = G_CONSTANT i64 16
@@ -953,10 +943,9 @@ body: |
     ; CHECK-LABEL: name: v16s16_umax
     ; CHECK: liveins: $x0, $q0, $q1
     ; CHECK-NEXT: {{  $}}
-    ; CHECK-NEXT: [[DEF:%[0-9]+]]:_(s16) = G_IMPLICIT_DEF
-    ; CHECK-NEXT: [[BUILD_VECTOR:%[0-9]+]]:_(<8 x s16>) = G_BUILD_VECTOR [[DEF]](s16), [[DEF]](s16), [[DEF]](s16), [[DEF]](s16), [[DEF]](s16), [[DEF]](s16), [[DEF]](s16), [[DEF]](s16)
-    ; CHECK-NEXT: [[UMAX:%[0-9]+]]:_(<8 x s16>) = G_UMAX [[BUILD_VECTOR]], [[BUILD_VECTOR]]
-    ; CHECK-NEXT: [[UMAX1:%[0-9]+]]:_(<8 x s16>) = G_UMAX [[BUILD_VECTOR]], [[BUILD_VECTOR]]
+    ; CHECK-NEXT: [[DEF:%[0-9]+]]:_(<8 x s16>) = G_IMPLICIT_DEF
+    ; CHECK-NEXT: [[UMAX:%[0-9]+]]:_(<8 x s16>) = G_UMAX [[DEF]], [[DEF]]
+    ; CHECK-NEXT: [[UMAX1:%[0-9]+]]:_(<8 x s16>) = G_UMAX [[DEF]], [[DEF]]
     ; CHECK-NEXT: [[COPY:%[0-9]+]]:_(p0) = COPY $x0
     ; CHECK-NEXT: G_STORE [[UMAX]](<8 x s16>), [[COPY]](p0) :: (store (<8 x s16>), align 32)
     ; CHECK-NEXT: [[C:%[0-9]+]]:_(s64) = G_CONSTANT i64 16
@@ -1023,10 +1012,9 @@ body: |
     ; CHECK-LABEL: name: v8s32_umax
     ; CHECK: liveins: $x0, $q0, $q1
     ; CHECK-NEXT: {{  $}}
-    ; CHECK-NEXT: [[DEF:%[0-9]+]]:_(s32) = G_IMPLICIT_DEF
-    ; CHECK-NEXT: [[BUILD_VECTOR:%[0-9]+]]:_(<4 x s32>) = G_BUILD_VECTOR [[DEF]](s32), [[DEF]](s32), [[DEF]](s32), [[DEF]](s32)
-    ; CHECK-NEXT: [[UMAX:%[0-9]+]]:_(<4 x s32>) = G_UMAX [[BUILD_VECTOR]], [[BUILD_VECTOR]]
-    ; CHECK-NEXT: [[UMAX1:%[0-9]+]]:_(<4 x s32>) = G_UMAX [[BUILD_VECTOR]], [[BUILD_VECTOR]]
+    ; CHECK-NEXT: [[DEF:%[0-9]+]]:_(<4 x s32>) = G_IMPLICIT_DEF
+    ; CHECK-NEXT: [[UMAX:%[0-9]+]]:_(<4 x s32>) = G_UMAX [[DEF]], [[DEF]]
+    ; CHECK-NEXT: [[UMAX1:%[0-9]+]]:_(<4 x s32>) = G_UMAX [[DEF]], [[DEF]]
     ; CHECK-NEXT: [[COPY:%[0-9]+]]:_(p0) = COPY $x0
     ; CHECK-NEXT: G_STORE [[UMAX]](<4 x s32>), [[COPY]](p0) :: (store (<4 x s32>), align 32)
     ; CHECK-NEXT: [[C:%[0-9]+]]:_(s64) = G_CONSTANT i64 16
diff --git a/llvm/test/CodeGen/AArch64/GlobalISel/legalize-phi.mir b/llvm/test/CodeGen/AArch64/GlobalISel/legalize-phi.mir
index 8bd62c592254..7dbe3fefb6bd 100644
--- a/llvm/test/CodeGen/AArch64/GlobalISel/legalize-phi.mir
+++ b/llvm/test/CodeGen/AArch64/GlobalISel/legalize-phi.mir
@@ -766,22 +766,20 @@ body:             |
   ; CHECK-NEXT:   successors: %bb.1(0x50000000), %bb.2(0x30000000)
   ; CHECK-NEXT:   liveins: $x0, $x1
   ; CHECK-NEXT: {{  $}}
-  ; CHECK-NEXT:   [[DEF:%[0-9]+]]:_(s32) = G_IMPLICIT_DEF
-  ; CHECK-NEXT:   [[BUILD_VECTOR:%[0-9]+]]:_(<4 x s32>) = G_BUILD_VECTOR [[DEF]](s32), [[DEF]](s32), [[DEF]](s32), [[DEF]](s32)
-  ; CHECK-NEXT:   [[COPY:%[0-9]+]]:_(s32) = COPY [[DEF]](s32)
+  ; CHECK-NEXT:   [[DEF:%[0-9]+]]:_(<4 x s32>) = G_IMPLICIT_DEF
+  ; CHECK-NEXT:   [[DEF1:%[0-9]+]]:_(s32) = G_IMPLICIT_DEF
   ; CHECK-NEXT:   [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 1
-  ; CHECK-NEXT:   [[AND:%[0-9]+]]:_(s32) = G_AND [[COPY]], [[C]]
+  ; CHECK-NEXT:   [[AND:%[0-9]+]]:_(s32) = G_AND [[DEF1]], [[C]]
   ; CHECK-NEXT:   G_BRCOND [[AND]](s32), %bb.2
   ; CHECK-NEXT:   G_BR %bb.1
   ; CHECK-NEXT: {{  $}}
   ; CHECK-NEXT: bb.1:
   ; CHECK-NEXT:   successors: %bb.2(0x80000000)
   ; CHECK-NEXT: {{  $}}
-  ; CHECK-NEXT:   [[DEF1:%[0-9]+]]:_(s32) = G_IMPLICIT_DEF
-  ; CHECK-NEXT:   [[BUILD_VECTOR1:%[0-9]+]]:_(<4 x s32>) = G_BUILD_VECTOR [[DEF1]](s32), [[DEF1]](s32), [[DEF1]](s32), [[DEF1]](s32)
+  ; CHECK-NEXT:   [[DEF2:%[0-9]+]]:_(<4 x s32>) = G_IMPLICIT_DEF
   ; CHECK-NEXT: {{  $}}
   ; CHECK-NEXT: bb.2:
-  ; CHECK-NEXT:   [[PHI:%[0-9]+]]:_(<4 x s32>) = G_PHI [[BUILD_VECTOR1]](<4 x s32>), %bb.1, [[BUILD_VECTOR]](<4 x s32>), %bb.0
+  ; CHECK-NEXT:   [[PHI:%[0-9]+]]:_(<4 x s32>) = G_PHI [[DEF2]](<4 x s32>), %bb.1, [[DEF]](<4 x s32>), %bb.0
   ; CHECK-NEXT:   %one:_(s64) = G_CONSTANT i64 1
   ; CHECK-NEXT:   %extract:_(s32) = G_EXTRACT_VECTOR_ELT [[PHI]](<4 x s32>), %one(s64)
   ; CHECK-NEXT:   $w0 = COPY %extract(s32)
@@ -812,8 +810,7 @@ body:             |
   ; CHECK-NEXT:   successors: %bb.1(0x50000000), %bb.2(0x30000000)
   ; CHECK-NEXT:   liveins: $x0, $x1
   ; CHECK-NEXT: {{  $}}
-  ; CHECK-NEXT:   [[DEF:%[0-9]+]]:_(s16) = G_IMPLICIT_DEF
-  ; CHECK-NEXT:   [[BUILD_VECTOR:%[0-9]+]]:_(<8 x s16>) = G_BUILD_VECTOR [[DEF]](s16), [[DEF]](s16), [[DEF]](s16), [[DEF]](s16), [[DEF]](s16), [[DEF]](s16), [[DEF]](s16), [[DEF]](s16)
+  ; CHECK-NEXT:   [[DEF:%[0-9]+]]:_(<8 x s16>) = G_IMPLICIT_DEF
   ; CHECK-NEXT:   [[DEF1:%[0-9]+]]:_(s32) = G_IMPLICIT_DEF
   ; CHECK-NEXT:   [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 1
   ; CHECK-NEXT:   [[AND:%[0-9]+]]:_(s32) = G_AND [[DEF1]], [[C]]
@@ -823,11 +820,10 @@ body:             |
   ; CHECK-NEXT: bb.1:
   ; CHECK-NEXT:   successors: %bb.2(0x80000000)
   ; CHECK-NEXT: {{  $}}
-  ; CHECK-NEXT:   [[DEF2:%[0-9]+]]:_(s16) = G_IMPLICIT_DEF
-  ; CHECK-NEXT:   [[BUILD_VECTOR1:%[0-9]+]]:_(<8 x s16>) = G_BUILD_VECTOR [[DEF2]](s16), [[DEF2]](s16), [[DEF2]](s16), [[DEF2]](s16), [[DEF2]](s16), [[DEF2]](s16), [[DEF2]](s16), [[DEF2]](s16)
+  ; CHECK-NEXT:   [[DEF2:%[0-9]+]]:_(<8 x s16>) = G_IMPLICIT_DEF
   ; CHECK-NEXT: {{  $}}
   ; CHECK-NEXT: bb.2:
-  ; CHECK-NEXT:   [[PHI:%[0-9]+]]:_(<8 x s16>) = G_PHI [[BUILD_VECTOR1]](<8 x s16>), %bb.1, [[BUILD_VECTOR]](<8 x s16>), %bb.0
+  ; CHECK-NEXT:   [[PHI:%[0-9]+]]:_(<8 x s16>) = G_PHI [[DEF2]](<8 x s16>), %bb.1, [[DEF]](<8 x s16>), %bb.0
   ; CHECK-NEXT:   %one:_(s64) = G_CONSTANT i64 1
   ; CHECK-NEXT:   %extract:_(s16) = G_EXTRACT_VECTOR_ELT [[PHI]](<8 x s16>), %one(s64)
   ; CHECK-NEXT:   $h0 = COPY %extract(s16)
@@ -858,8 +854,7 @@ body:             |
   ; CHECK-NEXT:   successors: %bb.1(0x50000000), %bb.2(0x30000000)
   ; CHECK-NEXT:   liveins: $x0, $x1
   ; CHECK-NEXT: {{  $}}
-  ; CHECK-NEXT:   [[DEF:%[0-9]+]]:_(s8) = G_IMPLICIT_DEF
-  ; CHECK-NEXT:   [[BUILD_VECTOR:%[0-9]+]]:_(<16 x s8>) = G_BUILD_VECTOR [[DEF]](s8), [[DEF]](s8), [[DEF]](s8), [[DEF]](s8), [[DEF]](s8), [[DEF]](s8), [[DEF]](s8), [[DEF]](s8), [[DEF]](s8), [[DEF]](s8), [[DEF]](s8), [[DEF]](s8), [[DEF]](s8), [[DEF]](s8), [[DEF]](s8), [[DEF]](s8)
+  ; CHECK-NEXT:   [[DEF:%[0-9]+]]:_(<16 x s8>) = G_IMPLICIT_DEF
   ; CHECK-NEXT:   [[DEF1:%[0-9]+]]:_(s32) = G_IMPLICIT_DEF
   ; CHECK-NEXT:   [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 1
   ; CHECK-NEXT:   [[AND:%[0-9]+]]:_(s32) = G_AND [[DEF1]], [[C]]
@@ -869,11 +864,10 @@ body:             |
   ; CHECK-NEXT: bb.1:
   ; CHECK-NEXT:   successors: %bb.2(0x80000000)
   ; CHECK-NEXT: {{  $}}
-  ; CHECK-NEXT:   [[DEF2:%[0-9]+]]:_(s8) = G_IMPLICIT_DEF
-  ; CHECK-NEXT:   [[BUILD_VECTOR1:%[0-9]+]]:_(<16 x s8>) = G_BUILD_VECTOR [[DEF2]](s8), [[DEF2]](s8), [[DEF2]](s8), [[DEF2]](s8), [[DEF2]](s8), [[DEF2]](s8), [[DEF2]](s8), [[DEF2]](s8), [[DEF2]](s8), [[DEF2]](s8), [[DEF2]](s8), [[DEF2]](s8), [[DEF2]](s8), [[DEF2]](s8), [[DEF2]](s8), [[DEF2]](s8)
+  ; CHECK-NEXT:   [[DEF2:%[0-9]+]]:_(<16 x s8>) = G_IMPLICIT_DEF
   ; CHECK-NEXT: {{  $}}
   ; CHECK-NEXT: bb.2:
-  ; CHECK-NEXT:   [[PHI:%[0-9]+]]:_(<16 x s8>) = G_PHI [[BUILD_VECTOR1]](<16 x s8>), %bb.1, [[BUILD_VECTOR]](<16 x s8>), %bb.0
+  ; CHECK-NEXT:   [[PHI:%[0-9]+]]:_(<16 x s8>) = G_PHI [[DEF2]](<16 x s8>), %bb.1, [[DEF]](<16 x s8>), %bb.0
   ; CHECK-NEXT:   %one:_(s64) = G_CONSTANT i64 1
   ; CHECK-NEXT:   %extract:_(s8) = G_EXTRACT_VECTOR_ELT [[PHI]](<16 x s8>), %one(s64)
   ; CHECK-NEXT:   $b0 = COPY %extract(s8)
diff --git a/llvm/test/CodeGen/AArch64/GlobalISel/legalize-select.mir b/llvm/test/CodeGen/AArch64/GlobalISel/legalize-select.mir
index e49a94c12ed4..8b5cac1ec873 100644
--- a/llvm/test/CodeGen/AArch64/GlobalISel/legalize-select.mir
+++ b/llvm/test/CodeGen/AArch64/GlobalISel/legalize-select.mir
@@ -292,45 +292,47 @@ body:             |
     ; CHECK-NEXT: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 4100
     ; CHECK-NEXT: [[ICMP2:%[0-9]+]]:_(s32) = G_ICMP intpred(eq), %w0(s32), [[C]]
     ; CHECK-NEXT: [[SEXT_INREG:%[0-9]+]]:_(s32) = G_SEXT_INREG [[ICMP2]], 1
-    ; CHECK-NEXT: [[DEF:%[0-9]+]]:_(s8) = G_IMPLICIT_DEF
+    ; CHECK-NEXT: [[DEF:%[0-9]+]]:_(<4 x s16>) = G_IMPLICIT_DEF
     ; CHECK-NEXT: [[C1:%[0-9]+]]:_(s64) = G_CONSTANT i64 0
-    ; CHECK-NEXT: [[DEF1:%[0-9]+]]:_(s8) = G_IMPLICIT_DEF
-    ; CHECK-NEXT: [[BUILD_VECTOR:%[0-9]+]]:_(<8 x s8>) = G_BUILD_VECTOR [[DEF]](s8), [[DEF]](s8), [[DEF]](s8), [[DEF]](s8), [[DEF1]](s8), [[DEF1]](s8), [[DEF1]](s8), [[DEF1]](s8)
-    ; CHECK-NEXT: [[ANYEXT:%[0-9]+]]:_(<8 x s16>) = G_ANYEXT [[BUILD_VECTOR]](<8 x s8>)
-    ; CHECK-NEXT: [[UV:%[0-9]+]]:_(<4 x s16>), [[UV1:%[0-9]+]]:_(<4 x s16>) = G_UNMERGE_VALUES [[ANYEXT]](<8 x s16>)
     ; CHECK-NEXT: [[TRUNC:%[0-9]+]]:_(s16) = G_TRUNC [[SEXT_INREG]](s32)
-    ; CHECK-NEXT: [[IVEC:%[0-9]+]]:_(<4 x s16>) = G_INSERT_VECTOR_ELT [[UV]], [[TRUNC]](s16), [[C1]](s64)
-    ; CHECK-NEXT: [[UV2:%[0-9]+]]:_(s16), [[UV3:%[0-9]+]]:_(s16), [[UV4:%[0-9]+]]:_(s16), [[UV5:%[0-9]+]]:_(s16) = G_UNMERGE_VALUES [[IVEC]](<4 x s16>)
-    ; CHECK-NEXT: [[TRUNC1:%[0-9]+]]:_(s8) = G_TRUNC [[UV2]](s16)
-    ; CHECK-NEXT: [[TRUNC2:%[0-9]+]]:_(s8) = G_TRUNC [[UV3]](s16)
-    ; CHECK-NEXT: [[TRUNC3:%[0-9]+]]:_(s8) = G_TRUNC [[UV4]](s16)
-    ; CHECK-NEXT: [[TRUNC4:%[0-9]+]]:_(s8) = G_TRUNC [[UV5]](s16)
-    ; CHECK-NEXT: [[BUILD_VECTOR1:%[0-9]+]]:_(<8 x s8>) = G_BUILD_VECTOR [[TRUNC1]](s8), [[TRUNC2]](s8), [[TRUNC3]](s8), [[TRUNC4]](s8), [[DEF1]](s8), [[DEF1]](s8), [[DEF1]](s8), [[DEF1]](s8)
-    ; CHECK-NEXT: [[BUILD_VECTOR2:%[0-9]+]]:_(<8 x s8>) = G_BUILD_VECTOR [[DEF]](s8), [[DEF]](s8), [[DEF]](s8), [[DEF]](s8), [[DEF1]](s8), [[DEF1]](s8), [[DEF1]](s8), [[DEF1]](s8)
-    ; CHECK-NEXT: [[SHUF:%[0-9]+]]:_(<8 x s8>) = G_SHUFFLE_VECTOR [[BUILD_VECTOR1]](<8 x s8>), [[BUILD_VECTOR2]], shufflemask(0, 0, 0, 0, undef, undef, undef, undef)
-    ; CHECK-NEXT: [[UV6:%[0-9]+]]:_(<4 x s8>), [[UV7:%[0-9]+]]:_(<4 x s8>) = G_UNMERGE_VALUES [[SHUF]](<8 x s8>)
+    ; CHECK-NEXT: [[IVEC:%[0-9]+]]:_(<4 x s16>) = G_INSERT_VECTOR_ELT [[DEF]], [[TRUNC]](s16), [[C1]](s64)
+    ; CHECK-NEXT: [[UV:%[0-9]+]]:_(s16), [[UV1:%[0-9]+]]:_(s16), [[UV2:%[0-9]+]]:_(s16), [[UV3:%[0-9]+]]:_(s16) = G_UNMERGE_VALUES [[IVEC]](<4 x s16>)
+    ; CHECK-NEXT: [[TRUNC1:%[0-9]+]]:_(s8) = G_TRUNC [[UV]](s16)
+    ; CHECK-NEXT: [[TRUNC2:%[0-9]+]]:_(s8) = G_TRUNC [[UV1]](s16)
+    ; CHECK-NEXT: [[TRUNC3:%[0-9]+]]:_(s8) = G_TRUNC [[UV2]](s16)
+    ; CHECK-NEXT: [[TRUNC4:%[0-9]+]]:_(s8) = G_TRUNC [[UV3]](s16)
+    ; CHECK-NEXT: [[DEF1:%[0-9]+]]:_(s8) = G_IMPLICIT_DEF
+    ; CHECK-NEXT: [[BUILD_VECTOR:%[0-9]+]]:_(<8 x s8>) = G_BUILD_VECTOR [[TRUNC1]](s8), [[TRUNC2]](s8), [[TRUNC3]](s8), [[TRUNC4]](s8), [[DEF1]](s8), [[DEF1]](s8), [[DEF1]](s8), [[DEF1]](s8)
+    ; CHECK-NEXT: [[UV4:%[0-9]+]]:_(s16), [[UV5:%[0-9]+]]:_(s16), [[UV6:%[0-9]+]]:_(s16), [[UV7:%[0-9]+]]:_(s16) = G_UNMERGE_VALUES [[DEF]](<4 x s16>)
+    ; CHECK-NEXT: [[TRUNC5:%[0-9]+]]:_(s8) = G_TRUNC [[UV4]](s16)
+    ; CHECK-NEXT: [[TRUNC6:%[0-9]+]]:_(s8) = G_TRUNC [[UV5]](s16)
+    ; CHECK-NEXT: [[TRUNC7:%[0-9]+]]:_(s8) = G_TRUNC [[UV6]](s16)
+    ; CHECK-NEXT: [[TRUNC8:%[0-9]+]]:_(s8) = G_TRUNC [[UV7]](s16)
+    ; CHECK-NEXT: [[BUILD_VECTOR1:%[0-9]+]]:_(<8 x s8>) = G_BUILD_VECTOR [[TRUNC5]](s8), [[TRUNC6]](s8), [[TRUNC7]](s8), [[TRUNC8]](s8), [[DEF1]](s8), [[DEF1]](s8), [[DEF1]](s8), [[DEF1]](s8)
+    ; CHECK-NEXT: [[SHUF:%[0-9]+]]:_(<8 x s8>) = G_SHUFFLE_VECTOR [[BUILD_VECTOR]](<8 x s8>), [[BUILD_VECTOR1]], shufflemask(0, 0, 0, 0, undef, undef, undef, undef)
+    ; CHECK-NEXT: [[UV8:%[0-9]+]]:_(<4 x s8>), [[UV9:%[0-9]+]]:_(<4 x s8>) = G_UNMERGE_VALUES [[SHUF]](<8 x s8>)
     ; CHECK-NEXT: [[C2:%[0-9]+]]:_(s8) = G_CONSTANT i8 1
-    ; CHECK-NEXT: [[UV8:%[0-9]+]]:_(s8), [[UV9:%[0-9]+]]:_(s8), [[UV10:%[0-9]+]]:_(s8), [[UV11:%[0-9]+]]:_(s8) = G_UNMERGE_VALUES [[UV6]](<4 x s8>)
-    ; CHECK-NEXT: [[BUILD_VECTOR3:%[0-9]+]]:_(<8 x s8>) = G_BUILD_VECTOR [[UV8]](s8), [[UV9]](s8), [[UV10]](s8), [[UV11]](s8), [[DEF1]](s8), [[DEF1]](s8), [[DEF1]](s8), [[DEF1]](s8)
+    ; CHECK-NEXT: [[UV10:%[0-9]+]]:_(s8), [[UV11:%[0-9]+]]:_(s8), [[UV12:%[0-9]+]]:_(s8), [[UV13:%[0-9]+]]:_(s8) = G_UNMERGE_VALUES [[UV8]](<4 x s8>)
+    ; CHECK-NEXT: [[BUILD_VECTOR2:%[0-9]+]]:_(<8 x s8>) = G_BUILD_VECTOR [[UV10]](s8), [[UV11]](s8), [[UV12]](s8), [[UV13]](s8), [[DEF1]](s8), [[DEF1]](s8), [[DEF1]](s8), [[DEF1]](s8)
+    ; CHECK-NEXT: [[ANYEXT:%[0-9]+]]:_(<8 x s16>) = G_ANYEXT [[BUILD_VECTOR2]](<8 x s8>)
+    ; CHECK-NEXT: [[UV14:%[0-9]+]]:_(<4 x s16>), [[UV15:%[0-9]+]]:_(<4 x s16>) = G_UNMERGE_VALUES [[ANYEXT]](<8 x s16>)
+    ; CHECK-NEXT: [[BUILD_VECTOR3:%[0-9]+]]:_(<8 x s8>) = G_BUILD_VECTOR [[C2]](s8), [[C2]](s8), [[C2]](s8), [[C2]](s8), [[DEF1]](s8), [[DEF1]](s8), [[DEF1]](s8), [[DEF1]](s8)
     ; CHECK-NEXT: [[ANYEXT1:%[0-9]+]]:_(<8 x s16>) = G_ANYEXT [[BUILD_VECTOR3]](<8 x s8>)
-    ; CHECK-NEXT: [[UV12:%[0-9]+]]:_(<4 x s16>), [[UV13:%[0-9]+]]:_(<4 x s16>) = G_UNMERGE_VALUES [[ANYEXT1]](<8 x s16>)
-    ; CHECK-NEXT: [[BUILD_VECTOR4:%[0-9]+]]:_(<8 x s8>) = G_BUILD_VECTOR [[C2]](s8), [[C2]](s8), [[C2]](s8), [[C2]](s8), [[DEF1]](s8), [[DEF1]](s8), [[DEF1]](s8), [[DEF1]](s8)
+    ; CHECK-NEXT: [[UV16:%[0-9]+]]:_(<4 x s16>), [[UV17:%[0-9]+]]:_(<4 x s16>) = G_UNMERGE_VALUES [[ANYEXT1]](<8 x s16>)
+    ; CHECK-NEXT: [[XOR:%[0-9]+]]:_(<4 x s16>) = G_XOR [[UV14]], [[UV16]]
+    ; CHECK-NEXT: [[TRUNC9:%[0-9]+]]:_(<4 x s16>) = G_TRUNC [[ICMP]](<4 x s32>)
+    ; CHECK-NEXT: [[UV18:%[0-9]+]]:_(s8), [[UV19:%[0-9]+]]:_(s8), [[UV20:%[0-9]+]]:_(s8), [[UV21:%[0-9]+]]:_(s8) = G_UNMERGE_VALUES [[UV8]](<4 x s8>)
+    ; CHECK-NEXT: [[BUILD_VECTOR4:%[0-9]+]]:_(<8 x s8>) = G_BUILD_VECTOR [[UV18]](s8), [[UV19]](s8), [[UV20]](s8), [[UV21]](s8), [[DEF1]](s8), [[DEF1]](s8), [[DEF1]](s8), [[DEF1]](s8)
     ; CHECK-NEXT: [[ANYEXT2:%[0-9]+]]:_(<8 x s16>) = G_ANYEXT [[BUILD_VECTOR4]](<8 x s8>)
-    ; CHECK-NEXT: [[UV14:%[0-9]+]]:_(<4 x s16>), [[UV15:%[0-9]+]]:_(<4 x s16>) = G_UNMERGE_VALUES [[ANYEXT2]](<8 x s16>)
-    ; CHECK-NEXT: [[XOR:%[0-9]+]]:_(<4 x s16>) = G_XOR [[UV12]], [[UV14]]
-    ; CHECK-NEXT: [[TRUNC5:%[0-9]+]]:_(<4 x s16>) = G_TRUNC [[ICMP]](<4 x s32>)
-    ; CHECK-NEXT: [[UV16:%[0-9]+]]:_(s8), [[UV17:%[0-9]+]]:_(s8), [[UV18:%[0-9]+]]:_(s8), [[UV19:%[0-9]+]]:_(s8) = G_UNMERGE_VALUES [[UV6]](<4 x s8>)
-    ; CHECK-NEXT: [[BUILD_VECTOR5:%[0-9]+]]:_(<8 x s8>) = G_BUILD_VECTOR [[UV16]](s8), [[UV17]](s8), [[UV18]](s8), [[UV19]](s8), [[DEF1]](s8), [[DEF1]](s8), [[DEF1]](s8), [[DEF1]](s8)
-    ; CHECK-NEXT: [[ANYEXT3:%[0-9]+]]:_(<8 x s16>) = G_ANYEXT [[BUILD_VECTOR5]](<8 x s8>)
-    ; CHECK-NEXT: [[UV20:%[0-9]+]]:_(<4 x s16>), [[UV21:%[0-9]+]]:_(<4 x s16>) = G_UNMERGE_VALUES [[ANYEXT3]](<8 x s16>)
-    ; CHECK-NEXT: [[AND:%[0-9]+]]:_(<4 x s16>) = G_AND [[TRUNC5]], [[UV20]]
-    ; CHECK-NEXT: [[TRUNC6:%[0-9]+]]:_(<4 x s16>) = G_TRUNC [[ICMP1]](<4 x s32>)
-    ; CHECK-NEXT: [[AND1:%[0-9]+]]:_(<4 x s16>) = G_AND [[TRUNC6]], [[XOR]]
+    ; CHECK-NEXT: [[UV22:%[0-9]+]]:_(<4 x s16>), [[UV23:%[0-9]+]]:_(<4 x s16>) = G_UNMERGE_VALUES [[ANYEXT2]](<8 x s16>)
+    ; CHECK-NEXT: [[AND:%[0-9]+]]:_(<4 x s16>) = G_AND [[TRUNC9]], [[UV22]]
+    ; CHECK-NEXT: [[TRUNC10:%[0-9]+]]:_(<4 x s16>) = G_TRUNC [[ICMP1]](<4 x s32>)
+    ; CHECK-NEXT: [[AND1:%[0-9]+]]:_(<4 x s16>) = G_AND [[TRUNC10]], [[XOR]]
     ; CHECK-NEXT: [[OR:%[0-9]+]]:_(<4 x s16>) = G_OR [[AND]], [[AND1]]
-    ; CHECK-NEXT: [[ANYEXT4:%[0-9]+]]:_(<4 x s32>) = G_ANYEXT [[OR]](<4 x s16>)
+    ; CHECK-NEXT: [[ANYEXT3:%[0-9]+]]:_(<4 x s32>) = G_ANYEXT [[OR]](<4 x s16>)
     ; CHECK-NEXT: [[C3:%[0-9]+]]:_(s32) = G_CONSTANT i32 1
-    ; CHECK-NEXT: [[BUILD_VECTOR6:%[0-9]+]]:_(<4 x s32>) = G_BUILD_VECTOR [[C3]](s32), [[C3]](s32), [[C3]](s32), [[C3]](s32)
-    ; CHECK-NEXT: %zext_select:_(<4 x s32>) = G_AND [[ANYEXT4]], [[BUILD_VECTOR6]]
+    ; CHECK-NEXT: [[BUILD_VECTOR5:%[0-9]+]]:_(<4 x s32>) = G_BUILD_VECTOR [[C3]](s32), [[C3]](s32), [[C3]](s32), [[C3]](s32)
+    ; CHECK-NEXT: %zext_select:_(<4 x s32>) = G_AND [[ANYEXT3]], [[BUILD_VECTOR5]]
     ; CHECK-NEXT: $q0 = COPY %zext_select(<4 x s32>)
     ; CHECK-NEXT: RET_ReallyLR implicit $q0
     %w0:_(s32) = COPY $w0
diff --git a/llvm/test/CodeGen/AArch64/GlobalISel/legalize-shift.mir b/llvm/test/CodeGen/AArch64/GlobalISel/legalize-shift.mir
index 9a9b35cf293d..f1d345ff18f7 100644
--- a/llvm/test/CodeGen/AArch64/GlobalISel/legalize-shift.mir
+++ b/llvm/test/CodeGen/AArch64/GlobalISel/legalize-shift.mir
@@ -411,13 +411,9 @@ name: test_ashr_v8s32
 body:             |
   bb.0:
     ; CHECK-LABEL: name: test_ashr_v8s32
-    ; CHECK: [[DEF:%[0-9]+]]:_(s32) = G_IMPLICIT_DEF
-    ; CHECK-NEXT: [[BUILD_VECTOR:%[0-9]+]]:_(<4 x s32>) = G_BUILD_VECTOR [[DEF]](s32), [[DEF]](s32), [[DEF]](s32), [[DEF]](s32)
-    ; CHECK-NEXT: [[BUILD_VECTOR1:%[0-9]+]]:_(<4 x s32>) = G_BUILD_VECTOR [[DEF]](s32), [[DEF]](s32), [[DEF]](s32), [[DEF]](s32)
-    ; CHECK-NEXT: [[BUILD_VECTOR2:%[0-9]+]]:_(<4 x s32>) = G_BUILD_VECTOR [[DEF]](s32), [[DEF]](s32), [[DEF]](s32), [[DEF]](s32)
-    ; CHECK-NEXT: [[BUILD_VECTOR3:%[0-9]+]]:_(<4 x s32>) = G_BUILD_VECTOR [[DEF]](s32), [[DEF]](s32), [[DEF]](s32), [[DEF]](s32)
-    ; CHECK-NEXT: [[ASHR:%[0-9]+]]:_(<4 x s32>) = G_ASHR [[BUILD_VECTOR]], [[BUILD_VECTOR2]](<4 x s32>)
-    ; CHECK-NEXT: [[ASHR1:%[0-9]+]]:_(<4 x s32>) = G_ASHR [[BUILD_VECTOR1]], [[BUILD_VECTOR3]](<4 x s32>)
+    ; CHECK: [[DEF:%[0-9]+]]:_(<4 x s32>) = G_IMPLICIT_DEF
+    ; CHECK-NEXT: [[ASHR:%[0-9]+]]:_(<4 x s32>) = G_ASHR [[DEF]], [[DEF]](<4 x s32>)
+    ; CHECK-NEXT: [[ASHR1:%[0-9]+]]:_(<4 x s32>) = G_ASHR [[DEF]], [[DEF]](<4 x s32>)
     ; CHECK-NEXT: $q0 = COPY [[ASHR]](<4 x s32>)
     ; CHECK-NEXT: $q1 = COPY [[ASHR1]](<4 x s32>)
     %0:_(<8 x s32>) = G_IMPLICIT_DEF
diff --git a/llvm/test/CodeGen/AArch64/GlobalISel/legalize-shuffle-vector.mir b/llvm/test/CodeGen/AArch64/GlobalISel/legalize-shuffle-vector.mir
index 2cfee7bcc462..c92718f9e9b3 100644
--- a/llvm/test/CodeGen/AArch64/GlobalISel/legalize-shuffle-vector.mir
+++ b/llvm/test/CodeGen/AArch64/GlobalISel/legalize-shuffle-vector.mir
@@ -312,9 +312,10 @@ body:             |
     ; CHECK-NEXT: [[COPY:%[0-9]+]]:_(s32) = COPY $w0
     ; CHECK-NEXT: [[COPY1:%[0-9]+]]:_(s32) = COPY $w1
     ; CHECK-NEXT: [[COPY2:%[0-9]+]]:_(s32) = COPY $w2
-    ; CHECK-NEXT: [[DEF:%[0-9]+]]:_(s32) = G_IMPLICIT_DEF
-    ; CHECK-NEXT: [[BUILD_VECTOR:%[0-9]+]]:_(<4 x s32>) = G_BUILD_VECTOR [[COPY]](s32), [[COPY1]](s32), [[COPY2]](s32), [[DEF]](s32)
-    ; CHECK-NEXT: [[BUILD_VECTOR1:%[0-9]+]]:_(<4 x s32>) = G_BUILD_VECTOR [[COPY]](s32), [[COPY1]](s32), [[COPY2]](s32), [[DEF]](s32)
+    ; CHECK-NEXT: [[DEF:%[0-9]+]]:_(<4 x s32>) = G_IMPLICIT_DEF
+    ; CHECK-NEXT: [[UV:%[0-9]+]]:_(s32), [[UV1:%[0-9]+]]:_(s32), [[UV2:%[0-9]+]]:_(s32), [[UV3:%[0-9]+]]:_(s32) = G_UNMERGE_VALUES [[DEF]](<4 x s32>)
+    ; CHECK-NEXT: [[BUILD_VECTOR:%[0-9]+]]:_(<4 x s32>) = G_BUILD_VECTOR [[COPY]](s32), [[COPY1]](s32), [[COPY2]](s32), [[UV]](s32)
+    ; CHECK-NEXT: [[BUILD_VECTOR1:%[0-9]+]]:_(<4 x s32>) = G_BUILD_VECTOR [[COPY]](s32), [[COPY1]](s32), [[COPY2]](s32), [[UV]](s32)
     ; CHECK-NEXT: [[SHUF:%[0-9]+]]:_(<4 x s32>) = G_SHUFFLE_VECTOR [[BUILD_VECTOR]](<4 x s32>), [[BUILD_VECTOR1]], shufflemask(0, 1, 5, 6)
     ; CHECK-NEXT: [[C:%[0-9]+]]:_(s64) = G_CONSTANT i64 0
     ; CHECK-NEXT: [[EVEC:%[0-9]+]]:_(s32) = G_EXTRACT_VECTOR_ELT [[SHUF]](<4 x s32>), [[C]](s64)
diff --git a/llvm/test/CodeGen/AArch64/GlobalISel/legalizer-info-validation.mir b/llvm/test/CodeGen/AArch64/GlobalISel/legalizer-info-validation.mir
index 098726b0a980..20133158e4fa 100644
--- a/llvm/test/CodeGen/AArch64/GlobalISel/legalizer-info-validation.mir
+++ b/llvm/test/CodeGen/AArch64/GlobalISel/legalizer-info-validation.mir
@@ -71,8 +71,8 @@
 # DEBUG-NEXT: .. the first uncovered imm index: 0, OK
 #
 # DEBUG-NEXT: G_IMPLICIT_DEF (opcode {{[0-9]+}}): 1 type index, 0 imm indices
-# DEBUG-NEXT: .. type index coverage check SKIPPED: user-defined predicate detected
-# DEBUG-NEXT: .. imm index coverage check SKIPPED: user-defined predicate detected
+# DEBUG-NEXT: .. the first uncovered type index: {{[0-9]+}}, OK
+# DEBUG-NEXT: .. the first uncovered imm index: {{[0-9]+}}, OK
 #
 # DEBUG-NEXT: G_PHI (opcode {{[0-9]+}}): 1 type index, 0 imm indices
 # DEBUG-NEXT: .. the first uncovered type index: 1, OK
@@ -132,13 +132,13 @@
 #
 # DEBUG-NEXT: G_FREEZE (opcode {{[0-9]+}}): 1 type index, 0 imm indices
 # DEBUG-NEXT: .. opcode {{[0-9]+}} is aliased to {{[0-9]+}}
-# DEBUG-NEXT: .. type index coverage check SKIPPED: user-defined predicate detected
-# DEBUG-NEXT: .. imm index coverage check SKIPPED: user-defined predicate detected
+# DEBUG-NEXT: .. the first uncovered type index: {{[0-9]+}}, OK
+# DEBUG-NEXT: .. the first uncovered imm index: {{[0-9]+}}, OK
 
 # DEBUG-NEXT: G_CONSTANT_FOLD_BARRIER (opcode {{[0-9]+}}): 1 type index, 0 imm indices
 # DEBUG-NEXT: .. opcode {{[0-9]+}} is aliased to {{[0-9]+}}
-# DEBUG-NEXT: .. type index coverage check SKIPPED: user-defined predicate detected
-# DEBUG-NEXT: .. imm index coverage check SKIPPED: user-defined predicate detected
+# DEBUG-NEXT: .. the first uncovered type index: {{[0-9]+}}, OK
+# DEBUG-NEXT: .. the first uncovered imm index: {{[0-9]+}}, OK
 
 #
 # DEBUG-NEXT: G_INTRINSIC_FPTRUNC_ROUND (opcode {{[0-9]+}}): 2 type indices, 0 imm indices
diff --git a/llvm/test/CodeGen/AArch64/arm64-uzp.ll b/llvm/test/CodeGen/AArch64/arm64-uzp.ll
index 6e01ebc95a1c..49a51d96fbc8 100644
--- a/llvm/test/CodeGen/AArch64/arm64-uzp.ll
+++ b/llvm/test/CodeGen/AArch64/arm64-uzp.ll
@@ -110,13 +110,9 @@ define <8 x i16> @vuzpQi16_undef1(<8 x i16> %A, <8 x i16> %B) nounwind {
 define <8 x i16> @vuzpQi16_undef0(<8 x i16> %A, <8 x i16> %B) nounwind {
 ; CHECK-LABEL: vuzpQi16_undef0:
 ; CHECK:       // %bb.0:
-; CHECK-NEXT:    adrp x8, .LCPI8_0
-; CHECK-NEXT:    // kill: def $q1 killed $q1 killed $q0_q1 def $q0_q1
-; CHECK-NEXT:    ldr q2, [x8, :lo12:.LCPI8_0]
-; CHECK-NEXT:    // kill: def $q0 killed $q0 killed $q0_q1 def $q0_q1
-; CHECK-NEXT:    uzp2.8h v3, v0, v1
-; CHECK-NEXT:    tbl.16b v0, { v0, v1 }, v2
-; CHECK-NEXT:    add.8h v0, v0, v3
+; CHECK-NEXT:    uzp1.8h v2, v0, v1
+; CHECK-NEXT:    uzp2.8h v0, v0, v1
+; CHECK-NEXT:    add.8h v0, v2, v0
 ; CHECK-NEXT:    ret
   %tmp3 = shufflevector <8 x i16> %A, <8 x i16> %B, <8 x i32> <i32 undef, i32 2, i32 4, i32 6, i32 8, i32 10, i32 12, i32 14>
   %tmp4 = shufflevector <8 x i16> %A, <8 x i16> %B, <8 x i32> <i32 undef, i32 3, i32 5, i32 7, i32 9, i32 11, i32 13, i32 15>
@@ -127,13 +123,9 @@ define <8 x i16> @vuzpQi16_undef0(<8 x i16> %A, <8 x i16> %B) nounwind {
 define <8 x i16> @vuzpQi16_undef01(<8 x i16> %A, <8 x i16> %B) nounwind {
 ; CHECK-LABEL: vuzpQi16_undef01:
 ; CHECK:       // %bb.0:
-; CHECK-NEXT:    adrp x8, .LCPI9_0
-; CHECK-NEXT:    // kill: def $q1 killed $q1 killed $q0_q1 def $q0_q1
-; CHECK-NEXT:    ldr q2, [x8, :lo12:.LCPI9_0]
-; CHECK-NEXT:    // kill: def $q0 killed $q0 killed $q0_q1 def $q0_q1
-; CHECK-NEXT:    uzp2.8h v3, v0, v1
-; CHECK-NEXT:    tbl.16b v0, { v0, v1 }, v2
-; CHECK-NEXT:    add.8h v0, v0, v3
+; CHECK-NEXT:    uzp1.8h v2, v0, v1
+; CHECK-NEXT:    uzp2.8h v0, v0, v1
+; CHECK-NEXT:    add.8h v0, v2, v0
 ; CHECK-NEXT:    ret
   %tmp3 = shufflevector <8 x i16> %A, <8 x i16> %B, <8 x i32> <i32 undef, i32 undef, i32 4, i32 6, i32 8, i32 10, i32 12, i32 14>
   %tmp4 = shufflevector <8 x i16> %A, <8 x i16> %B, <8 x i32> <i32 undef, i32 undef, i32 5, i32 7, i32 9, i32 11, i32 13, i32 15>
@@ -144,13 +136,9 @@ define <8 x i16> @vuzpQi16_undef01(<8 x i16> %A, <8 x i16> %B) nounwind {
 define <8 x i16> @vuzpQi16_undef012(<8 x i16> %A, <8 x i16> %B) nounwind {
 ; CHECK-LABEL: vuzpQi16_undef012:
 ; CHECK:       // %bb.0:
-; CHECK-NEXT:    adrp x8, .LCPI10_0
-; CHECK-NEXT:    // kill: def $q1 killed $q1 killed $q0_q1 def $q0_q1
-; CHECK-NEXT:    ldr q2, [x8, :lo12:.LCPI10_0]
-; CHECK-NEXT:    // kill: def $q0 killed $q0 killed $q0_q1 def $q0_q1
-; CHECK-NEXT:    uzp2.8h v3, v0, v1
-; CHECK-NEXT:    tbl.16b v0, { v0, v1 }, v2
-; CHECK-NEXT:    add.8h v0, v0, v3
+; CHECK-NEXT:    uzp1.8h v2, v0, v1
+; CHECK-NEXT:    uzp2.8h v0, v0, v1
+; CHECK-NEXT:    add.8h v0, v2, v0
 ; CHECK-NEXT:    ret
   %tmp3 = shufflevector <8 x i16> %A, <8 x i16> %B, <8 x i32> <i32 undef, i32 undef, i32 undef, i32 6, i32 8, i32 10, i32 12, i32 14>
   %tmp4 = shufflevector <8 x i16> %A, <8 x i16> %B, <8 x i32> <i32 undef, i32 undef, i32 undef, i32 7, i32 9, i32 11, i32 13, i32 15>
diff --git a/llvm/test/CodeGen/AArch64/arm64-zip.ll b/llvm/test/CodeGen/AArch64/arm64-zip.ll
index 349751dda461..4c771cbd2966 100644
--- a/llvm/test/CodeGen/AArch64/arm64-zip.ll
+++ b/llvm/test/CodeGen/AArch64/arm64-zip.ll
@@ -142,11 +142,7 @@ define <16 x i8> @vzipQi8_undef(ptr %A, ptr %B) nounwind {
 define <8 x i16> @vzip1_undef_01(<8 x i16> %A, <8 x i16> %B) nounwind {
 ; CHECK-LABEL: vzip1_undef_01:
 ; CHECK:       // %bb.0:
-; CHECK-NEXT:    adrp x8, .LCPI8_0
-; CHECK-NEXT:    // kill: def $q1 killed $q1 killed $q0_q1 def $q0_q1
-; CHECK-NEXT:    ldr q2, [x8, :lo12:.LCPI8_0]
-; CHECK-NEXT:    // kill: def $q0 killed $q0 killed $q0_q1 def $q0_q1
-; CHECK-NEXT:    tbl.16b v0, { v0, v1 }, v2
+; CHECK-NEXT:    zip1.8h v0, v0, v1
 ; CHECK-NEXT:    ret
   %s = shufflevector <8 x i16> %A, <8 x i16> %B, <8 x i32> <i32 undef, i32 undef, i32 1, i32 9, i32 2, i32 10, i32 3, i32 11>
   ret <8 x i16> %s
@@ -155,11 +151,7 @@ define <8 x i16> @vzip1_undef_01(<8 x i16> %A, <8 x i16> %B) nounwind {
 define <8 x i16> @vzip1_undef_0(<8 x i16> %A, <8 x i16> %B) nounwind {
 ; CHECK-LABEL: vzip1_undef_0:
 ; CHECK:       // %bb.0:
-; CHECK-NEXT:    adrp x8, .LCPI9_0
-; CHECK-NEXT:    // kill: def $q1 killed $q1 killed $q0_q1 def $q0_q1
-; CHECK-NEXT:    ldr q2, [x8, :lo12:.LCPI9_0]
-; CHECK-NEXT:    // kill: def $q0 killed $q0 killed $q0_q1 def $q0_q1
-; CHECK-NEXT:    tbl.16b v0, { v0, v1 }, v2
+; CHECK-NEXT:    zip1.8h v0, v0, v1
 ; CHECK-NEXT:    ret
   %s = shufflevector <8 x i16> %A, <8 x i16> %B, <8 x i32> <i32 undef, i32 8, i32 1, i32 9, i32 2, i32 10, i32 3, i32 11>
   ret <8 x i16> %s
@@ -177,11 +169,7 @@ define <8 x i16> @vzip1_undef_1(<8 x i16> %A, <8 x i16> %B) nounwind {
 define <8 x i16> @vzip1_undef_012(<8 x i16> %A, <8 x i16> %B) nounwind {
 ; CHECK-LABEL: vzip1_undef_012:
 ; CHECK:       // %bb.0:
-; CHECK-NEXT:    adrp x8, .LCPI11_0
-; CHECK-NEXT:    // kill: def $q1 killed $q1 killed $q0_q1 def $q0_q1
-; CHECK-NEXT:    ldr q2, [x8, :lo12:.LCPI11_0]
-; CHECK-NEXT:    // kill: def $q0 killed $q0 killed $q0_q1 def $q0_q1
-; CHECK-NEXT:    tbl.16b v0, { v0, v1 }, v2
+; CHECK-NEXT:    zip1.8h v0, v0, v1
 ; CHECK-NEXT:    ret
   %s = shufflevector <8 x i16> %A, <8 x i16> %B, <8 x i32> <i32 undef, i32 undef, i32 undef, i32 9, i32 2, i32 10, i32 3, i32 11>
   ret <8 x i16> %s
diff --git a/llvm/test/CodeGen/AArch64/arm64ec-entry-thunks.ll b/llvm/test/CodeGen/AArch64/arm64ec-entry-thunks.ll
index bb9ba05f7a27..c00c9bfe127e 100644
--- a/llvm/test/CodeGen/AArch64/arm64ec-entry-thunks.ll
+++ b/llvm/test/CodeGen/AArch64/arm64ec-entry-thunks.ll
@@ -223,8 +223,8 @@ define i8 @matches_has_sret() nounwind {
 
 %TSRet = type { i64, i64 }
 define void @has_aligned_sret(ptr align 32 sret(%TSRet)) nounwind {
-; CHECK-LABEL:    .def    $ientry_thunk$cdecl$m16a32$v;
-; CHECK:          .section        .wowthk$aa,"xr",discard,$ientry_thunk$cdecl$m16a32$v
+; CHECK-LABEL:    .def    $ientry_thunk$cdecl$m16$v;
+; CHECK:          .section        .wowthk$aa,"xr",discard,$ientry_thunk$cdecl$m16$v
 ; CHECK:          // %bb.0:
 ; CHECK-NEXT:     stp     q6, q7, [sp, #-176]!            // 32-byte Folded Spill
 ; CHECK-NEXT:     .seh_save_any_reg_px    q6, 176
@@ -457,7 +457,7 @@ define %T2 @simple_struct(%T1 %0, %T2 %1, %T3, %T4) nounwind {
 ; CHECK-NEXT:     .symidx $ientry_thunk$cdecl$i8$v
 ; CHECK-NEXT:     .word   1
 ; CHECK-NEXT:     .symidx "#has_aligned_sret"
-; CHECK-NEXT:     .symidx $ientry_thunk$cdecl$m16a32$v
+; CHECK-NEXT:     .symidx $ientry_thunk$cdecl$m16$v
 ; CHECK-NEXT:     .word   1
 ; CHECK-NEXT:     .symidx "#small_array"
 ; CHECK-NEXT:     .symidx $ientry_thunk$cdecl$m2$m2F8
diff --git a/llvm/test/CodeGen/AArch64/arm64ec-exit-thunks.ll b/llvm/test/CodeGen/AArch64/arm64ec-exit-thunks.ll
index 3b911e78aff2..7a40fcd85ac5 100644
--- a/llvm/test/CodeGen/AArch64/arm64ec-exit-thunks.ll
+++ b/llvm/test/CodeGen/AArch64/arm64ec-exit-thunks.ll
@@ -236,8 +236,8 @@ declare void @has_sret(ptr sret([100 x i8])) nounwind;
 
 %TSRet = type { i64, i64 }
 declare void @has_aligned_sret(ptr align 32 sret(%TSRet)) nounwind;
-; CHECK-LABEL:    .def    $iexit_thunk$cdecl$m16a32$v;
-; CHECK:          .section        .wowthk$aa,"xr",discard,$iexit_thunk$cdecl$m16a32$v
+; CHECK-LABEL:    .def    $iexit_thunk$cdecl$m16$v;
+; CHECK:          .section        .wowthk$aa,"xr",discard,$iexit_thunk$cdecl$m16$v
 ; CHECK:          // %bb.0:
 ; CHECK-NEXT:     sub     sp, sp, #48
 ; CHECK-NEXT:     .seh_stackalloc 48
@@ -271,8 +271,8 @@ declare void @has_aligned_sret(ptr align 32 sret(%TSRet)) nounwind;
 ; CHECK:          adrp    x11, has_aligned_sret
 ; CHECK:          add     x11, x11, :lo12:has_aligned_sret
 ; CHECK:          ldr     x9, [x9, :lo12:__os_arm64x_check_icall]
-; CHECK:          adrp    x10, ($iexit_thunk$cdecl$m16a32$v)
-; CHECK:          add     x10, x10, :lo12:($iexit_thunk$cdecl$m16a32$v)
+; CHECK:          adrp    x10, ($iexit_thunk$cdecl$m16$v)
+; CHECK:          add     x10, x10, :lo12:($iexit_thunk$cdecl$m16$v)
 ; CHECK:          blr     x9
 ; CHECK:          .seh_startepilogue
 ; CHECK:          ldr     x30, [sp], #16                  // 8-byte Folded Reload
@@ -492,7 +492,7 @@ declare %T2 @simple_struct(%T1, %T2, %T3, %T4) nounwind;
 ; CHECK-NEXT:     .symidx has_sret
 ; CHECK-NEXT:     .word   0
 ; CHECK-NEXT:     .symidx has_aligned_sret
-; CHECK-NEXT:     .symidx $iexit_thunk$cdecl$m16a32$v
+; CHECK-NEXT:     .symidx $iexit_thunk$cdecl$m16$v
 ; CHECK-NEXT:     .word   4
 ; CHECK-NEXT:     .symidx "#has_aligned_sret$exit_thunk"
 ; CHECK-NEXT:     .symidx has_aligned_sret
diff --git a/llvm/test/CodeGen/AArch64/combine-mul.ll b/llvm/test/CodeGen/AArch64/combine-mul.ll
index a2b042530809..c49e5ae6620a 100644
--- a/llvm/test/CodeGen/AArch64/combine-mul.ll
+++ b/llvm/test/CodeGen/AArch64/combine-mul.ll
@@ -44,8 +44,7 @@ define <4 x i1> @PR48683_vec_undef(<4 x i32> %x) {
 define i64 @combine_mul_self_demandedbits(i64 %x) {
 ; CHECK-LABEL: combine_mul_self_demandedbits:
 ; CHECK:       // %bb.0:
-; CHECK-NEXT:    mul x8, x0, x0
-; CHECK-NEXT:    and x0, x8, #0xfffffffffffffffd
+; CHECK-NEXT:    mul x0, x0, x0
 ; CHECK-NEXT:    ret
   %1 = mul i64 %x, %x
   %2 = and i64 %1, -3
@@ -77,7 +76,7 @@ define i8 @one_demanded_bit(i8 %x) {
 define <2 x i64> @one_demanded_bit_splat(<2 x i64> %x) {
 ; CHECK-LABEL: one_demanded_bit_splat:
 ; CHECK:       // %bb.0:
-; CHECK-NEXT:    mov w8, #32
+; CHECK-NEXT:    mov w8, #32 // =0x20
 ; CHECK-NEXT:    shl v0.2d, v0.2d, #5
 ; CHECK-NEXT:    dup v1.2d, x8
 ; CHECK-NEXT:    and v0.16b, v0.16b, v1.16b
@@ -131,7 +130,7 @@ define i32 @squared_demanded_2_low_bits(i32 %x) {
 define <2 x i64> @squared_demanded_2_low_bits_splat(<2 x i64> %x) {
 ; CHECK-LABEL: squared_demanded_2_low_bits_splat:
 ; CHECK:       // %bb.0:
-; CHECK-NEXT:    mov x8, #-2
+; CHECK-NEXT:    mov x8, #-2 // =0xfffffffffffffffe
 ; CHECK-NEXT:    dup v1.2d, x8
 ; CHECK-NEXT:    orr v0.16b, v0.16b, v1.16b
 ; CHECK-NEXT:    ret
diff --git a/llvm/test/CodeGen/AArch64/complex-deinterleaving-add-mull-scalable-contract.ll b/llvm/test/CodeGen/AArch64/complex-deinterleaving-add-mull-scalable-contract.ll
index 86b1d5d195ff..0485d530fd06 100644
--- a/llvm/test/CodeGen/AArch64/complex-deinterleaving-add-mull-scalable-contract.ll
+++ b/llvm/test/CodeGen/AArch64/complex-deinterleaving-add-mull-scalable-contract.ll
@@ -25,10 +25,10 @@ define <vscale x 4 x double> @mull_add(<vscale x 4 x double> %a, <vscale x 4 x d
 ; CHECK-NEXT:    zip2 z1.d, z2.d, z1.d
 ; CHECK-NEXT:    ret
 entry:
-  %strided.vec = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.experimental.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %a)
+  %strided.vec = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %a)
   %0 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %strided.vec, 0
   %1 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %strided.vec, 1
-  %strided.vec29 = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.experimental.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %b)
+  %strided.vec29 = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %b)
   %2 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %strided.vec29, 0
   %3 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %strided.vec29, 1
   %4 = fmul contract <vscale x 2 x double> %0, %3
@@ -37,12 +37,12 @@ entry:
   %7 = fmul contract <vscale x 2 x double> %0, %2
   %8 = fmul contract <vscale x 2 x double> %1, %3
   %9 = fsub contract <vscale x 2 x double> %7, %8
-  %strided.vec31 = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.experimental.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %c)
+  %strided.vec31 = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %c)
   %10 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %strided.vec31, 0
   %11 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %strided.vec31, 1
   %12 = fadd contract <vscale x 2 x double> %10, %9
   %13 = fadd contract <vscale x 2 x double> %6, %11
-  %interleaved.vec = tail call <vscale x 4 x double> @llvm.experimental.vector.interleave2.nxv4f64(<vscale x 2 x double> %12, <vscale x 2 x double> %13)
+  %interleaved.vec = tail call <vscale x 4 x double> @llvm.vector.interleave2.nxv4f64(<vscale x 2 x double> %12, <vscale x 2 x double> %13)
   ret <vscale x 4 x double> %interleaved.vec
 }
 
@@ -67,10 +67,10 @@ define <vscale x 4 x double> @mul_add_mull(<vscale x 4 x double> %a, <vscale x 4
 ; CHECK-NEXT:    fadd z0.d, z25.d, z27.d
 ; CHECK-NEXT:    ret
 entry:
-  %strided.vec = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.experimental.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %a)
+  %strided.vec = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %a)
   %0 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %strided.vec, 0
   %1 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %strided.vec, 1
-  %strided.vec52 = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.experimental.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %b)
+  %strided.vec52 = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %b)
   %2 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %strided.vec52, 0
   %3 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %strided.vec52, 1
   %4 = fmul contract <vscale x 2 x double> %0, %3
@@ -79,10 +79,10 @@ entry:
   %7 = fmul contract <vscale x 2 x double> %0, %2
   %8 = fmul contract <vscale x 2 x double> %1, %3
   %9 = fsub contract <vscale x 2 x double> %7, %8
-  %strided.vec54 = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.experimental.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %c)
+  %strided.vec54 = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %c)
   %10 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %strided.vec54, 0
   %11 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %strided.vec54, 1
-  %strided.vec56 = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.experimental.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %d)
+  %strided.vec56 = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %d)
   %12 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %strided.vec56, 0
   %13 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %strided.vec56, 1
   %14 = fmul contract <vscale x 2 x double> %10, %13
@@ -93,7 +93,7 @@ entry:
   %19 = fsub contract <vscale x 2 x double> %17, %18
   %20 = fadd contract <vscale x 2 x double> %9, %19
   %21 = fadd contract <vscale x 2 x double> %6, %16
-  %interleaved.vec = tail call <vscale x 4 x double> @llvm.experimental.vector.interleave2.nxv4f64(<vscale x 2 x double> %20, <vscale x 2 x double> %21)
+  %interleaved.vec = tail call <vscale x 4 x double> @llvm.vector.interleave2.nxv4f64(<vscale x 2 x double> %20, <vscale x 2 x double> %21)
   ret <vscale x 4 x double> %interleaved.vec
 }
 
@@ -118,10 +118,10 @@ define <vscale x 4 x double> @mul_sub_mull(<vscale x 4 x double> %a, <vscale x 4
 ; CHECK-NEXT:    fsub z0.d, z25.d, z27.d
 ; CHECK-NEXT:    ret
 entry:
-  %strided.vec = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.experimental.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %a)
+  %strided.vec = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %a)
   %0 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %strided.vec, 0
   %1 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %strided.vec, 1
-  %strided.vec52 = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.experimental.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %b)
+  %strided.vec52 = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %b)
   %2 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %strided.vec52, 0
   %3 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %strided.vec52, 1
   %4 = fmul contract <vscale x 2 x double> %0, %3
@@ -130,10 +130,10 @@ entry:
   %7 = fmul contract <vscale x 2 x double> %0, %2
   %8 = fmul contract <vscale x 2 x double> %1, %3
   %9 = fsub contract <vscale x 2 x double> %7, %8
-  %strided.vec54 = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.experimental.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %c)
+  %strided.vec54 = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %c)
   %10 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %strided.vec54, 0
   %11 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %strided.vec54, 1
-  %strided.vec56 = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.experimental.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %d)
+  %strided.vec56 = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %d)
   %12 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %strided.vec56, 0
   %13 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %strided.vec56, 1
   %14 = fmul contract <vscale x 2 x double> %10, %13
@@ -144,7 +144,7 @@ entry:
   %19 = fsub contract <vscale x 2 x double> %17, %18
   %20 = fsub contract <vscale x 2 x double> %9, %19
   %21 = fsub contract <vscale x 2 x double> %6, %16
-  %interleaved.vec = tail call <vscale x 4 x double> @llvm.experimental.vector.interleave2.nxv4f64(<vscale x 2 x double> %20, <vscale x 2 x double> %21)
+  %interleaved.vec = tail call <vscale x 4 x double> @llvm.vector.interleave2.nxv4f64(<vscale x 2 x double> %20, <vscale x 2 x double> %21)
   ret <vscale x 4 x double> %interleaved.vec
 }
 
@@ -169,10 +169,10 @@ define <vscale x 4 x double> @mul_conj_mull(<vscale x 4 x double> %a, <vscale x
 ; CHECK-NEXT:    fadd z0.d, z25.d, z27.d
 ; CHECK-NEXT:    ret
 entry:
-  %strided.vec = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.experimental.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %a)
+  %strided.vec = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %a)
   %0 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %strided.vec, 0
   %1 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %strided.vec, 1
-  %strided.vec60 = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.experimental.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %b)
+  %strided.vec60 = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %b)
   %2 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %strided.vec60, 0
   %3 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %strided.vec60, 1
   %4 = fmul contract <vscale x 2 x double> %0, %3
@@ -181,10 +181,10 @@ entry:
   %7 = fmul contract <vscale x 2 x double> %0, %2
   %8 = fmul contract <vscale x 2 x double> %1, %3
   %9 = fsub contract <vscale x 2 x double> %7, %8
-  %strided.vec62 = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.experimental.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %c)
+  %strided.vec62 = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %c)
   %10 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %strided.vec62, 0
   %11 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %strided.vec62, 1
-  %strided.vec64 = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.experimental.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %d)
+  %strided.vec64 = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %d)
   %12 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %strided.vec64, 0
   %13 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %strided.vec64, 1
   %14 = fmul contract <vscale x 2 x double> %10, %13
@@ -195,7 +195,7 @@ entry:
   %19 = fadd contract <vscale x 2 x double> %17, %18
   %20 = fadd contract <vscale x 2 x double> %9, %19
   %21 = fadd contract <vscale x 2 x double> %6, %16
-  %interleaved.vec = tail call <vscale x 4 x double> @llvm.experimental.vector.interleave2.nxv4f64(<vscale x 2 x double> %20, <vscale x 2 x double> %21)
+  %interleaved.vec = tail call <vscale x 4 x double> @llvm.vector.interleave2.nxv4f64(<vscale x 2 x double> %20, <vscale x 2 x double> %21)
   ret <vscale x 4 x double> %interleaved.vec
 }
 
@@ -238,10 +238,10 @@ define <vscale x 4 x double> @mul_add_rot_mull(<vscale x 4 x double> %a, <vscale
 ; CHECK-NEXT:    zip2 z1.d, z2.d, z1.d
 ; CHECK-NEXT:    ret
 entry:
-  %strided.vec = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.experimental.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %a)
+  %strided.vec = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %a)
   %0 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %strided.vec, 0
   %1 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %strided.vec, 1
-  %strided.vec78 = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.experimental.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %b)
+  %strided.vec78 = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %b)
   %2 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %strided.vec78, 0
   %3 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %strided.vec78, 1
   %4 = fmul contract <vscale x 2 x double> %0, %3
@@ -250,14 +250,14 @@ entry:
   %7 = fmul contract <vscale x 2 x double> %0, %2
   %8 = fmul contract <vscale x 2 x double> %1, %3
   %9 = fsub contract <vscale x 2 x double> %7, %8
-  %strided.vec80 = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.experimental.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %c)
+  %strided.vec80 = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %c)
   %10 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %strided.vec80, 0
   %11 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %strided.vec80, 1
   %12 = tail call contract <vscale x 2 x double> @llvm.copysign.nxv2f64(<vscale x 2 x double> zeroinitializer, <vscale x 2 x double> %11)
   %13 = fadd contract <vscale x 2 x double> %10, %12
   %14 = tail call contract <vscale x 2 x double> @llvm.copysign.nxv2f64(<vscale x 2 x double> zeroinitializer, <vscale x 2 x double> %10)
   %15 = fsub contract <vscale x 2 x double> %14, %11
-  %strided.vec82 = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.experimental.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %d)
+  %strided.vec82 = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %d)
   %16 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %strided.vec82, 0
   %17 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %strided.vec82, 1
   %18 = fmul contract <vscale x 2 x double> %15, %17
@@ -268,10 +268,10 @@ entry:
   %23 = fsub contract <vscale x 2 x double> %21, %22
   %24 = fadd contract <vscale x 2 x double> %9, %23
   %25 = fadd contract <vscale x 2 x double> %6, %20
-  %interleaved.vec = tail call <vscale x 4 x double> @llvm.experimental.vector.interleave2.nxv4f64(<vscale x 2 x double> %24, <vscale x 2 x double> %25)
+  %interleaved.vec = tail call <vscale x 4 x double> @llvm.vector.interleave2.nxv4f64(<vscale x 2 x double> %24, <vscale x 2 x double> %25)
   ret <vscale x 4 x double> %interleaved.vec
 }
 
-declare { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.experimental.vector.deinterleave2.nxv4f64(<vscale x 4 x double>)
-declare <vscale x 4 x double> @llvm.experimental.vector.interleave2.nxv4f64(<vscale x 2 x double>, <vscale x 2 x double>)
+declare { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.vector.deinterleave2.nxv4f64(<vscale x 4 x double>)
+declare <vscale x 4 x double> @llvm.vector.interleave2.nxv4f64(<vscale x 2 x double>, <vscale x 2 x double>)
 declare <vscale x 2 x double> @llvm.copysign.nxv2f64(<vscale x 2 x double>, <vscale x 2 x double>)
diff --git a/llvm/test/CodeGen/AArch64/complex-deinterleaving-add-mull-scalable-fast.ll b/llvm/test/CodeGen/AArch64/complex-deinterleaving-add-mull-scalable-fast.ll
index edf580e334e8..c643ae9265c0 100644
--- a/llvm/test/CodeGen/AArch64/complex-deinterleaving-add-mull-scalable-fast.ll
+++ b/llvm/test/CodeGen/AArch64/complex-deinterleaving-add-mull-scalable-fast.ll
@@ -16,24 +16,24 @@ define <vscale x 4 x double> @mull_add(<vscale x 4 x double> %a, <vscale x 4 x d
 ; CHECK-NEXT:    mov z1.d, z5.d
 ; CHECK-NEXT:    ret
 entry:
-  %strided.vec = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.experimental.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %a)
+  %strided.vec = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %a)
   %0 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %strided.vec, 0
   %1 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %strided.vec, 1
-  %strided.vec29 = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.experimental.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %b)
+  %strided.vec29 = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %b)
   %2 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %strided.vec29, 0
   %3 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %strided.vec29, 1
   %4 = fmul fast <vscale x 2 x double> %3, %0
   %5 = fmul fast <vscale x 2 x double> %2, %1
   %6 = fadd fast <vscale x 2 x double> %4, %5
   %7 = fmul fast <vscale x 2 x double> %2, %0
-  %strided.vec31 = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.experimental.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %c)
+  %strided.vec31 = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %c)
   %8 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %strided.vec31, 0
   %9 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %strided.vec31, 1
   %10 = fadd fast <vscale x 2 x double> %8, %7
   %11 = fmul fast <vscale x 2 x double> %3, %1
   %12 = fsub fast <vscale x 2 x double> %10, %11
   %13 = fadd fast <vscale x 2 x double> %6, %9
-  %interleaved.vec = tail call <vscale x 4 x double> @llvm.experimental.vector.interleave2.nxv4f64(<vscale x 2 x double> %12, <vscale x 2 x double> %13)
+  %interleaved.vec = tail call <vscale x 4 x double> @llvm.vector.interleave2.nxv4f64(<vscale x 2 x double> %12, <vscale x 2 x double> %13)
   ret <vscale x 4 x double> %interleaved.vec
 }
 
@@ -56,20 +56,20 @@ define <vscale x 4 x double> @mul_add_mull(<vscale x 4 x double> %a, <vscale x 4
 ; CHECK-NEXT:    mov z0.d, z25.d
 ; CHECK-NEXT:    ret
 entry:
-  %strided.vec = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.experimental.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %a)
+  %strided.vec = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %a)
   %0 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %strided.vec, 0
   %1 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %strided.vec, 1
-  %strided.vec52 = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.experimental.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %b)
+  %strided.vec52 = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %b)
   %2 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %strided.vec52, 0
   %3 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %strided.vec52, 1
   %4 = fmul fast <vscale x 2 x double> %3, %0
   %5 = fmul fast <vscale x 2 x double> %2, %1
   %6 = fmul fast <vscale x 2 x double> %2, %0
   %7 = fmul fast <vscale x 2 x double> %3, %1
-  %strided.vec54 = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.experimental.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %c)
+  %strided.vec54 = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %c)
   %8 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %strided.vec54, 0
   %9 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %strided.vec54, 1
-  %strided.vec56 = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.experimental.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %d)
+  %strided.vec56 = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %d)
   %10 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %strided.vec56, 0
   %11 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %strided.vec56, 1
   %12 = fmul fast <vscale x 2 x double> %11, %8
@@ -82,7 +82,7 @@ entry:
   %19 = fadd fast <vscale x 2 x double> %4, %5
   %20 = fadd fast <vscale x 2 x double> %19, %13
   %21 = fadd fast <vscale x 2 x double> %20, %12
-  %interleaved.vec = tail call <vscale x 4 x double> @llvm.experimental.vector.interleave2.nxv4f64(<vscale x 2 x double> %18, <vscale x 2 x double> %21)
+  %interleaved.vec = tail call <vscale x 4 x double> @llvm.vector.interleave2.nxv4f64(<vscale x 2 x double> %18, <vscale x 2 x double> %21)
   ret <vscale x 4 x double> %interleaved.vec
 }
 
@@ -105,20 +105,20 @@ define <vscale x 4 x double> @mul_sub_mull(<vscale x 4 x double> %a, <vscale x 4
 ; CHECK-NEXT:    mov z0.d, z25.d
 ; CHECK-NEXT:    ret
 entry:
-  %strided.vec = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.experimental.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %a)
+  %strided.vec = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %a)
   %0 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %strided.vec, 0
   %1 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %strided.vec, 1
-  %strided.vec54 = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.experimental.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %b)
+  %strided.vec54 = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %b)
   %2 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %strided.vec54, 0
   %3 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %strided.vec54, 1
   %4 = fmul fast <vscale x 2 x double> %3, %0
   %5 = fmul fast <vscale x 2 x double> %2, %1
   %6 = fmul fast <vscale x 2 x double> %2, %0
   %7 = fmul fast <vscale x 2 x double> %3, %1
-  %strided.vec56 = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.experimental.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %c)
+  %strided.vec56 = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %c)
   %8 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %strided.vec56, 0
   %9 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %strided.vec56, 1
-  %strided.vec58 = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.experimental.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %d)
+  %strided.vec58 = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %d)
   %10 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %strided.vec58, 0
   %11 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %strided.vec58, 1
   %12 = fmul fast <vscale x 2 x double> %11, %9
@@ -131,7 +131,7 @@ entry:
   %19 = fadd fast <vscale x 2 x double> %18, %17
   %20 = fadd fast <vscale x 2 x double> %4, %5
   %21 = fsub fast <vscale x 2 x double> %20, %19
-  %interleaved.vec = tail call <vscale x 4 x double> @llvm.experimental.vector.interleave2.nxv4f64(<vscale x 2 x double> %16, <vscale x 2 x double> %21)
+  %interleaved.vec = tail call <vscale x 4 x double> @llvm.vector.interleave2.nxv4f64(<vscale x 2 x double> %16, <vscale x 2 x double> %21)
   ret <vscale x 4 x double> %interleaved.vec
 }
 
@@ -154,19 +154,19 @@ define <vscale x 4 x double> @mul_conj_mull(<vscale x 4 x double> %a, <vscale x
 ; CHECK-NEXT:    mov z0.d, z25.d
 ; CHECK-NEXT:    ret
 entry:
-  %strided.vec = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.experimental.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %a)
+  %strided.vec = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %a)
   %0 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %strided.vec, 0
   %1 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %strided.vec, 1
-  %strided.vec60 = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.experimental.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %b)
+  %strided.vec60 = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %b)
   %2 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %strided.vec60, 0
   %3 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %strided.vec60, 1
   %4 = fmul fast <vscale x 2 x double> %3, %0
   %5 = fmul fast <vscale x 2 x double> %2, %1
   %6 = fmul fast <vscale x 2 x double> %2, %0
-  %strided.vec62 = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.experimental.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %c)
+  %strided.vec62 = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %c)
   %7 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %strided.vec62, 0
   %8 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %strided.vec62, 1
-  %strided.vec64 = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.experimental.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %d)
+  %strided.vec64 = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %d)
   %9 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %strided.vec64, 0
   %10 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %strided.vec64, 1
   %11 = fmul fast <vscale x 2 x double> %10, %7
@@ -180,7 +180,7 @@ entry:
   %19 = fmul fast <vscale x 2 x double> %9, %8
   %20 = fsub fast <vscale x 2 x double> %18, %19
   %21 = fadd fast <vscale x 2 x double> %20, %11
-  %interleaved.vec = tail call <vscale x 4 x double> @llvm.experimental.vector.interleave2.nxv4f64(<vscale x 2 x double> %17, <vscale x 2 x double> %21)
+  %interleaved.vec = tail call <vscale x 4 x double> @llvm.vector.interleave2.nxv4f64(<vscale x 2 x double> %17, <vscale x 2 x double> %21)
   ret <vscale x 4 x double> %interleaved.vec
 }
 
@@ -211,20 +211,20 @@ define <vscale x 4 x double> @mul_add_rot_mull(<vscale x 4 x double> %a, <vscale
 ; CHECK-NEXT:    zip2 z1.d, z2.d, z1.d
 ; CHECK-NEXT:    ret
 entry:
-  %strided.vec = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.experimental.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %a)
+  %strided.vec = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %a)
   %0 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %strided.vec, 0
   %1 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %strided.vec, 1
-  %strided.vec80 = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.experimental.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %b)
+  %strided.vec80 = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %b)
   %2 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %strided.vec80, 0
   %3 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %strided.vec80, 1
   %4 = fmul fast <vscale x 2 x double> %3, %0
   %5 = fmul fast <vscale x 2 x double> %2, %1
   %6 = fmul fast <vscale x 2 x double> %2, %0
   %7 = fmul fast <vscale x 2 x double> %3, %1
-  %strided.vec82 = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.experimental.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %c)
+  %strided.vec82 = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %c)
   %8 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %strided.vec82, 0
   %9 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %strided.vec82, 1
-  %strided.vec84 = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.experimental.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %d)
+  %strided.vec84 = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %d)
   %10 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %strided.vec84, 0
   %11 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %strided.vec84, 1
   %12 = fmul fast <vscale x 2 x double> %10, %8
@@ -237,9 +237,9 @@ entry:
   %19 = fadd fast <vscale x 2 x double> %18, %12
   %20 = fmul fast <vscale x 2 x double> %11, %9
   %21 = fsub fast <vscale x 2 x double> %19, %20
-  %interleaved.vec = tail call <vscale x 4 x double> @llvm.experimental.vector.interleave2.nxv4f64(<vscale x 2 x double> %17, <vscale x 2 x double> %21)
+  %interleaved.vec = tail call <vscale x 4 x double> @llvm.vector.interleave2.nxv4f64(<vscale x 2 x double> %17, <vscale x 2 x double> %21)
   ret <vscale x 4 x double> %interleaved.vec
 }
 
-declare { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.experimental.vector.deinterleave2.nxv4f64(<vscale x 4 x double>)
-declare <vscale x 4 x double> @llvm.experimental.vector.interleave2.nxv4f64(<vscale x 2 x double>, <vscale x 2 x double>)
+declare { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.vector.deinterleave2.nxv4f64(<vscale x 4 x double>)
+declare <vscale x 4 x double> @llvm.vector.interleave2.nxv4f64(<vscale x 2 x double>, <vscale x 2 x double>)
diff --git a/llvm/test/CodeGen/AArch64/complex-deinterleaving-f16-add-scalable.ll b/llvm/test/CodeGen/AArch64/complex-deinterleaving-f16-add-scalable.ll
index 48b5756b01fb..dae8d9f89e99 100644
--- a/llvm/test/CodeGen/AArch64/complex-deinterleaving-f16-add-scalable.ll
+++ b/llvm/test/CodeGen/AArch64/complex-deinterleaving-f16-add-scalable.ll
@@ -24,15 +24,15 @@ define <vscale x 4 x half> @complex_add_v4f16(<vscale x 4 x half> %a, <vscale x
 ; CHECK-NEXT:    uzp1 z0.s, z0.s, z2.s
 ; CHECK-NEXT:    ret
 entry:
-  %a.deinterleaved = tail call { <vscale x 2 x half>, <vscale x 2 x half> } @llvm.experimental.vector.deinterleave2.nxv4f16(<vscale x 4 x half> %a)
+  %a.deinterleaved = tail call { <vscale x 2 x half>, <vscale x 2 x half> } @llvm.vector.deinterleave2.nxv4f16(<vscale x 4 x half> %a)
   %a.real = extractvalue { <vscale x 2 x half>, <vscale x 2 x half> } %a.deinterleaved, 0
   %a.imag = extractvalue { <vscale x 2 x half>, <vscale x 2 x half> } %a.deinterleaved, 1
-  %b.deinterleaved = tail call { <vscale x 2 x half>, <vscale x 2 x half> } @llvm.experimental.vector.deinterleave2.nxv4f16(<vscale x 4 x half> %b)
+  %b.deinterleaved = tail call { <vscale x 2 x half>, <vscale x 2 x half> } @llvm.vector.deinterleave2.nxv4f16(<vscale x 4 x half> %b)
   %b.real = extractvalue { <vscale x 2 x half>, <vscale x 2 x half> } %b.deinterleaved, 0
   %b.imag = extractvalue { <vscale x 2 x half>, <vscale x 2 x half> } %b.deinterleaved, 1
   %0 = fsub fast <vscale x 2 x half> %b.real, %a.imag
   %1 = fadd fast <vscale x 2 x half> %b.imag, %a.real
-  %interleaved.vec = tail call <vscale x 4 x half> @llvm.experimental.vector.interleave2.nxv4f16(<vscale x 2 x half> %0, <vscale x 2 x half> %1)
+  %interleaved.vec = tail call <vscale x 4 x half> @llvm.vector.interleave2.nxv4f16(<vscale x 2 x half> %0, <vscale x 2 x half> %1)
   ret <vscale x 4 x half> %interleaved.vec
 }
 
@@ -45,15 +45,15 @@ define <vscale x 8 x half> @complex_add_v8f16(<vscale x 8 x half> %a, <vscale x
 ; CHECK-NEXT:    mov z0.d, z1.d
 ; CHECK-NEXT:    ret
 entry:
-  %a.deinterleaved = tail call { <vscale x 4 x half>, <vscale x 4 x half> } @llvm.experimental.vector.deinterleave2.nxv8f16(<vscale x 8 x half> %a)
+  %a.deinterleaved = tail call { <vscale x 4 x half>, <vscale x 4 x half> } @llvm.vector.deinterleave2.nxv8f16(<vscale x 8 x half> %a)
   %a.real = extractvalue { <vscale x 4 x half>, <vscale x 4 x half> } %a.deinterleaved, 0
   %a.imag = extractvalue { <vscale x 4 x half>, <vscale x 4 x half> } %a.deinterleaved, 1
-  %b.deinterleaved = tail call { <vscale x 4 x half>, <vscale x 4 x half> } @llvm.experimental.vector.deinterleave2.nxv8f16(<vscale x 8 x half> %b)
+  %b.deinterleaved = tail call { <vscale x 4 x half>, <vscale x 4 x half> } @llvm.vector.deinterleave2.nxv8f16(<vscale x 8 x half> %b)
   %b.real = extractvalue { <vscale x 4 x half>, <vscale x 4 x half> } %b.deinterleaved, 0
   %b.imag = extractvalue { <vscale x 4 x half>, <vscale x 4 x half> } %b.deinterleaved, 1
   %0 = fsub fast <vscale x 4 x half> %b.real, %a.imag
   %1 = fadd fast <vscale x 4 x half> %b.imag, %a.real
-  %interleaved.vec = tail call <vscale x 8 x half> @llvm.experimental.vector.interleave2.nxv8f16(<vscale x 4 x half> %0, <vscale x 4 x half> %1)
+  %interleaved.vec = tail call <vscale x 8 x half> @llvm.vector.interleave2.nxv8f16(<vscale x 4 x half> %0, <vscale x 4 x half> %1)
   ret <vscale x 8 x half> %interleaved.vec
 }
 
@@ -68,15 +68,15 @@ define <vscale x 16 x half> @complex_add_v16f16(<vscale x 16 x half> %a, <vscale
 ; CHECK-NEXT:    mov z1.d, z3.d
 ; CHECK-NEXT:    ret
 entry:
-  %a.deinterleaved = tail call { <vscale x 8 x half>, <vscale x 8 x half> } @llvm.experimental.vector.deinterleave2.nxv16f16(<vscale x 16 x half> %a)
+  %a.deinterleaved = tail call { <vscale x 8 x half>, <vscale x 8 x half> } @llvm.vector.deinterleave2.nxv16f16(<vscale x 16 x half> %a)
   %a.real = extractvalue { <vscale x 8 x half>, <vscale x 8 x half> } %a.deinterleaved, 0
   %a.imag = extractvalue { <vscale x 8 x half>, <vscale x 8 x half> } %a.deinterleaved, 1
-  %b.deinterleaved = tail call { <vscale x 8 x half>, <vscale x 8 x half> } @llvm.experimental.vector.deinterleave2.nxv16f16(<vscale x 16 x half> %b)
+  %b.deinterleaved = tail call { <vscale x 8 x half>, <vscale x 8 x half> } @llvm.vector.deinterleave2.nxv16f16(<vscale x 16 x half> %b)
   %b.real = extractvalue { <vscale x 8 x half>, <vscale x 8 x half> } %b.deinterleaved, 0
   %b.imag = extractvalue { <vscale x 8 x half>, <vscale x 8 x half> } %b.deinterleaved, 1
   %0 = fsub fast <vscale x 8 x half> %b.real, %a.imag
   %1 = fadd fast <vscale x 8 x half> %b.imag, %a.real
-  %interleaved.vec = tail call <vscale x 16 x half> @llvm.experimental.vector.interleave2.nxv16f16(<vscale x 8 x half> %0, <vscale x 8 x half> %1)
+  %interleaved.vec = tail call <vscale x 16 x half> @llvm.vector.interleave2.nxv16f16(<vscale x 8 x half> %0, <vscale x 8 x half> %1)
   ret <vscale x 16 x half> %interleaved.vec
 }
 
@@ -95,26 +95,26 @@ define <vscale x 32 x half> @complex_add_v32f16(<vscale x 32 x half> %a, <vscale
 ; CHECK-NEXT:    mov z3.d, z7.d
 ; CHECK-NEXT:    ret
 entry:
-  %a.deinterleaved = tail call { <vscale x 16 x half>, <vscale x 16 x half> } @llvm.experimental.vector.deinterleave2.nxv32f16(<vscale x 32 x half> %a)
+  %a.deinterleaved = tail call { <vscale x 16 x half>, <vscale x 16 x half> } @llvm.vector.deinterleave2.nxv32f16(<vscale x 32 x half> %a)
   %a.real = extractvalue { <vscale x 16 x half>, <vscale x 16 x half> } %a.deinterleaved, 0
   %a.imag = extractvalue { <vscale x 16 x half>, <vscale x 16 x half> } %a.deinterleaved, 1
-  %b.deinterleaved = tail call { <vscale x 16 x half>, <vscale x 16 x half> } @llvm.experimental.vector.deinterleave2.nxv32f16(<vscale x 32 x half> %b)
+  %b.deinterleaved = tail call { <vscale x 16 x half>, <vscale x 16 x half> } @llvm.vector.deinterleave2.nxv32f16(<vscale x 32 x half> %b)
   %b.real = extractvalue { <vscale x 16 x half>, <vscale x 16 x half> } %b.deinterleaved, 0
   %b.imag = extractvalue { <vscale x 16 x half>, <vscale x 16 x half> } %b.deinterleaved, 1
   %0 = fsub fast <vscale x 16 x half> %b.real, %a.imag
   %1 = fadd fast <vscale x 16 x half> %b.imag, %a.real
-  %interleaved.vec = tail call <vscale x 32 x half> @llvm.experimental.vector.interleave2.nxv32f16(<vscale x 16 x half> %0, <vscale x 16 x half> %1)
+  %interleaved.vec = tail call <vscale x 32 x half> @llvm.vector.interleave2.nxv32f16(<vscale x 16 x half> %0, <vscale x 16 x half> %1)
   ret <vscale x 32 x half> %interleaved.vec
 }
 
-declare { <vscale x 2 x half>, <vscale x 2 x half> } @llvm.experimental.vector.deinterleave2.nxv4f16(<vscale x 4 x half>)
-declare <vscale x 4 x half> @llvm.experimental.vector.interleave2.nxv4f16(<vscale x 2 x half>, <vscale x 2 x half>)
+declare { <vscale x 2 x half>, <vscale x 2 x half> } @llvm.vector.deinterleave2.nxv4f16(<vscale x 4 x half>)
+declare <vscale x 4 x half> @llvm.vector.interleave2.nxv4f16(<vscale x 2 x half>, <vscale x 2 x half>)
 
-declare { <vscale x 4 x half>, <vscale x 4 x half> } @llvm.experimental.vector.deinterleave2.nxv8f16(<vscale x 8 x half>)
-declare <vscale x 8 x half> @llvm.experimental.vector.interleave2.nxv8f16(<vscale x 4 x half>, <vscale x 4 x half>)
+declare { <vscale x 4 x half>, <vscale x 4 x half> } @llvm.vector.deinterleave2.nxv8f16(<vscale x 8 x half>)
+declare <vscale x 8 x half> @llvm.vector.interleave2.nxv8f16(<vscale x 4 x half>, <vscale x 4 x half>)
 
-declare { <vscale x 8 x half>, <vscale x 8 x half> } @llvm.experimental.vector.deinterleave2.nxv16f16(<vscale x 16 x half>)
-declare <vscale x 16 x half> @llvm.experimental.vector.interleave2.nxv16f16(<vscale x 8 x half>, <vscale x 8 x half>)
+declare { <vscale x 8 x half>, <vscale x 8 x half> } @llvm.vector.deinterleave2.nxv16f16(<vscale x 16 x half>)
+declare <vscale x 16 x half> @llvm.vector.interleave2.nxv16f16(<vscale x 8 x half>, <vscale x 8 x half>)
 
-declare { <vscale x 16 x half>, <vscale x 16 x half> } @llvm.experimental.vector.deinterleave2.nxv32f16(<vscale x 32 x half>)
-declare <vscale x 32 x half> @llvm.experimental.vector.interleave2.nxv32f16(<vscale x 16 x half>, <vscale x 16 x half>)
+declare { <vscale x 16 x half>, <vscale x 16 x half> } @llvm.vector.deinterleave2.nxv32f16(<vscale x 32 x half>)
+declare <vscale x 32 x half> @llvm.vector.interleave2.nxv32f16(<vscale x 16 x half>, <vscale x 16 x half>)
diff --git a/llvm/test/CodeGen/AArch64/complex-deinterleaving-f16-add.ll b/llvm/test/CodeGen/AArch64/complex-deinterleaving-f16-add.ll
index 7cdb10e7159f..a5c64c0982d0 100644
--- a/llvm/test/CodeGen/AArch64/complex-deinterleaving-f16-add.ll
+++ b/llvm/test/CodeGen/AArch64/complex-deinterleaving-f16-add.ll
@@ -127,15 +127,15 @@ define <4 x half> @complex_add_v4f16_with_intrinsic(<4 x half> %a, <4 x half> %b
 ; CHECK-NEXT:    fcadd v0.4h, v1.4h, v0.4h, #90
 ; CHECK-NEXT:    ret
 entry:
-  %a.deinterleaved = tail call { <2 x half>, <2 x half> } @llvm.experimental.vector.deinterleave2.v4f16(<4 x half> %a)
+  %a.deinterleaved = tail call { <2 x half>, <2 x half> } @llvm.vector.deinterleave2.v4f16(<4 x half> %a)
   %a.real = extractvalue { <2 x half>, <2 x half> } %a.deinterleaved, 0
   %a.imag = extractvalue { <2 x half>, <2 x half> } %a.deinterleaved, 1
-  %b.deinterleaved = tail call { <2 x half>, <2 x half> } @llvm.experimental.vector.deinterleave2.v4f16(<4 x half> %b)
+  %b.deinterleaved = tail call { <2 x half>, <2 x half> } @llvm.vector.deinterleave2.v4f16(<4 x half> %b)
   %b.real = extractvalue { <2 x half>, <2 x half> } %b.deinterleaved, 0
   %b.imag = extractvalue { <2 x half>, <2 x half> } %b.deinterleaved, 1
   %0 = fsub fast <2 x half> %b.real, %a.imag
   %1 = fadd fast <2 x half> %b.imag, %a.real
-  %interleaved.vec = tail call <4 x half> @llvm.experimental.vector.interleave2.v4f16(<2 x half> %0, <2 x half> %1)
+  %interleaved.vec = tail call <4 x half> @llvm.vector.interleave2.v4f16(<2 x half> %0, <2 x half> %1)
   ret <4 x half> %interleaved.vec
 }
 
@@ -146,15 +146,15 @@ define <8 x half> @complex_add_v8f16_with_intrinsic(<8 x half> %a, <8 x half> %b
 ; CHECK-NEXT:    fcadd v0.8h, v1.8h, v0.8h, #90
 ; CHECK-NEXT:    ret
 entry:
-  %a.deinterleaved = tail call { <4 x half>, <4 x half> } @llvm.experimental.vector.deinterleave2.v8f16(<8 x half> %a)
+  %a.deinterleaved = tail call { <4 x half>, <4 x half> } @llvm.vector.deinterleave2.v8f16(<8 x half> %a)
   %a.real = extractvalue { <4 x half>, <4 x half> } %a.deinterleaved, 0
   %a.imag = extractvalue { <4 x half>, <4 x half> } %a.deinterleaved, 1
-  %b.deinterleaved = tail call { <4 x half>, <4 x half> } @llvm.experimental.vector.deinterleave2.v8f16(<8 x half> %b)
+  %b.deinterleaved = tail call { <4 x half>, <4 x half> } @llvm.vector.deinterleave2.v8f16(<8 x half> %b)
   %b.real = extractvalue { <4 x half>, <4 x half> } %b.deinterleaved, 0
   %b.imag = extractvalue { <4 x half>, <4 x half> } %b.deinterleaved, 1
   %0 = fsub fast <4 x half> %b.real, %a.imag
   %1 = fadd fast <4 x half> %b.imag, %a.real
-  %interleaved.vec = tail call <8 x half> @llvm.experimental.vector.interleave2.v8f16(<4 x half> %0, <4 x half> %1)
+  %interleaved.vec = tail call <8 x half> @llvm.vector.interleave2.v8f16(<4 x half> %0, <4 x half> %1)
   ret <8 x half> %interleaved.vec
 }
 
@@ -166,15 +166,15 @@ define <16 x half> @complex_add_v16f16_with_intrinsic(<16 x half> %a, <16 x half
 ; CHECK-NEXT:    fcadd v0.8h, v2.8h, v0.8h, #90
 ; CHECK-NEXT:    ret
 entry:
-  %a.deinterleaved = tail call { <8 x half>, <8 x half> } @llvm.experimental.vector.deinterleave2.v16f16(<16 x half> %a)
+  %a.deinterleaved = tail call { <8 x half>, <8 x half> } @llvm.vector.deinterleave2.v16f16(<16 x half> %a)
   %a.real = extractvalue { <8 x half>, <8 x half> } %a.deinterleaved, 0
   %a.imag = extractvalue { <8 x half>, <8 x half> } %a.deinterleaved, 1
-  %b.deinterleaved = tail call { <8 x half>, <8 x half> } @llvm.experimental.vector.deinterleave2.v16f16(<16 x half> %b)
+  %b.deinterleaved = tail call { <8 x half>, <8 x half> } @llvm.vector.deinterleave2.v16f16(<16 x half> %b)
   %b.real = extractvalue { <8 x half>, <8 x half> } %b.deinterleaved, 0
   %b.imag = extractvalue { <8 x half>, <8 x half> } %b.deinterleaved, 1
   %0 = fsub fast <8 x half> %b.real, %a.imag
   %1 = fadd fast <8 x half> %b.imag, %a.real
-  %interleaved.vec = tail call <16 x half> @llvm.experimental.vector.interleave2.v16f16(<8 x half> %0, <8 x half> %1)
+  %interleaved.vec = tail call <16 x half> @llvm.vector.interleave2.v16f16(<8 x half> %0, <8 x half> %1)
   ret <16 x half> %interleaved.vec
 }
 
@@ -216,11 +216,11 @@ entry:
 }
 
 
-declare { <2 x half>, <2 x half> } @llvm.experimental.vector.deinterleave2.v4f16(<4 x half>)
-declare <4 x half> @llvm.experimental.vector.interleave2.v4f16(<2 x half>, <2 x half>)
+declare { <2 x half>, <2 x half> } @llvm.vector.deinterleave2.v4f16(<4 x half>)
+declare <4 x half> @llvm.vector.interleave2.v4f16(<2 x half>, <2 x half>)
 
-declare { <4 x half>, <4 x half> } @llvm.experimental.vector.deinterleave2.v8f16(<8 x half>)
-declare <8 x half> @llvm.experimental.vector.interleave2.v8f16(<4 x half>, <4 x half>)
+declare { <4 x half>, <4 x half> } @llvm.vector.deinterleave2.v8f16(<8 x half>)
+declare <8 x half> @llvm.vector.interleave2.v8f16(<4 x half>, <4 x half>)
 
-declare { <8 x half>, <8 x half> } @llvm.experimental.vector.deinterleave2.v16f16(<16 x half>)
-declare <16 x half> @llvm.experimental.vector.interleave2.v16f16(<8 x half>, <8 x half>)
+declare { <8 x half>, <8 x half> } @llvm.vector.deinterleave2.v16f16(<16 x half>)
+declare <16 x half> @llvm.vector.interleave2.v16f16(<8 x half>, <8 x half>)
diff --git a/llvm/test/CodeGen/AArch64/complex-deinterleaving-f16-mul-scalable.ll b/llvm/test/CodeGen/AArch64/complex-deinterleaving-f16-mul-scalable.ll
index cb285c05b2e8..c09ec616b015 100644
--- a/llvm/test/CodeGen/AArch64/complex-deinterleaving-f16-mul-scalable.ll
+++ b/llvm/test/CodeGen/AArch64/complex-deinterleaving-f16-mul-scalable.ll
@@ -27,10 +27,10 @@ define <vscale x 4 x half> @complex_mul_v4f16(<vscale x 4 x half> %a, <vscale x
 ; CHECK-NEXT:    uzp1 z0.s, z0.s, z1.s
 ; CHECK-NEXT:    ret
 entry:
-  %a.deinterleaved = tail call { <vscale x 2 x half>, <vscale x 2 x half> } @llvm.experimental.vector.deinterleave2.nxv4f16(<vscale x 4 x half> %a)
+  %a.deinterleaved = tail call { <vscale x 2 x half>, <vscale x 2 x half> } @llvm.vector.deinterleave2.nxv4f16(<vscale x 4 x half> %a)
   %a.real = extractvalue { <vscale x 2 x half>, <vscale x 2 x half> } %a.deinterleaved, 0
   %a.imag = extractvalue { <vscale x 2 x half>, <vscale x 2 x half> } %a.deinterleaved, 1
-  %b.deinterleaved = tail call { <vscale x 2 x half>, <vscale x 2 x half> } @llvm.experimental.vector.deinterleave2.nxv4f16(<vscale x 4 x half> %b)
+  %b.deinterleaved = tail call { <vscale x 2 x half>, <vscale x 2 x half> } @llvm.vector.deinterleave2.nxv4f16(<vscale x 4 x half> %b)
   %b.real = extractvalue { <vscale x 2 x half>, <vscale x 2 x half> } %b.deinterleaved, 0
   %b.imag = extractvalue { <vscale x 2 x half>, <vscale x 2 x half> } %b.deinterleaved, 1
   %0 = fmul fast <vscale x 2 x half> %b.imag, %a.real
@@ -39,7 +39,7 @@ entry:
   %3 = fmul fast <vscale x 2 x half> %b.real, %a.real
   %4 = fmul fast <vscale x 2 x half> %a.imag, %b.imag
   %5 = fsub fast <vscale x 2 x half> %3, %4
-  %interleaved.vec = tail call <vscale x 4 x half> @llvm.experimental.vector.interleave2.nxv4f16(<vscale x 2 x half> %5, <vscale x 2 x half> %2)
+  %interleaved.vec = tail call <vscale x 4 x half> @llvm.vector.interleave2.nxv4f16(<vscale x 2 x half> %5, <vscale x 2 x half> %2)
   ret <vscale x 4 x half> %interleaved.vec
 }
 
@@ -54,10 +54,10 @@ define <vscale x 8 x half> @complex_mul_v8f16(<vscale x 8 x half> %a, <vscale x
 ; CHECK-NEXT:    mov z0.d, z2.d
 ; CHECK-NEXT:    ret
 entry:
-  %a.deinterleaved = tail call { <vscale x 4 x half>, <vscale x 4 x half> } @llvm.experimental.vector.deinterleave2.nxv8f16(<vscale x 8 x half> %a)
+  %a.deinterleaved = tail call { <vscale x 4 x half>, <vscale x 4 x half> } @llvm.vector.deinterleave2.nxv8f16(<vscale x 8 x half> %a)
   %a.real = extractvalue { <vscale x 4 x half>, <vscale x 4 x half> } %a.deinterleaved, 0
   %a.imag = extractvalue { <vscale x 4 x half>, <vscale x 4 x half> } %a.deinterleaved, 1
-  %b.deinterleaved = tail call { <vscale x 4 x half>, <vscale x 4 x half> } @llvm.experimental.vector.deinterleave2.nxv8f16(<vscale x 8 x half> %b)
+  %b.deinterleaved = tail call { <vscale x 4 x half>, <vscale x 4 x half> } @llvm.vector.deinterleave2.nxv8f16(<vscale x 8 x half> %b)
   %b.real = extractvalue { <vscale x 4 x half>, <vscale x 4 x half> } %b.deinterleaved, 0
   %b.imag = extractvalue { <vscale x 4 x half>, <vscale x 4 x half> } %b.deinterleaved, 1
   %0 = fmul fast <vscale x 4 x half> %b.imag, %a.real
@@ -66,7 +66,7 @@ entry:
   %3 = fmul fast <vscale x 4 x half> %b.real, %a.real
   %4 = fmul fast <vscale x 4 x half> %a.imag, %b.imag
   %5 = fsub fast <vscale x 4 x half> %3, %4
-  %interleaved.vec = tail call <vscale x 8 x half> @llvm.experimental.vector.interleave2.nxv8f16(<vscale x 4 x half> %5, <vscale x 4 x half> %2)
+  %interleaved.vec = tail call <vscale x 8 x half> @llvm.vector.interleave2.nxv8f16(<vscale x 4 x half> %5, <vscale x 4 x half> %2)
   ret <vscale x 8 x half> %interleaved.vec
 }
 ; Expected to transform
@@ -84,10 +84,10 @@ define <vscale x 16 x half> @complex_mul_v16f16(<vscale x 16 x half> %a, <vscale
 ; CHECK-NEXT:    mov z0.d, z5.d
 ; CHECK-NEXT:    ret
 entry:
-  %a.deinterleaved = tail call { <vscale x 8 x half>, <vscale x 8 x half> } @llvm.experimental.vector.deinterleave2.nxv16f16(<vscale x 16 x half> %a)
+  %a.deinterleaved = tail call { <vscale x 8 x half>, <vscale x 8 x half> } @llvm.vector.deinterleave2.nxv16f16(<vscale x 16 x half> %a)
   %a.real = extractvalue { <vscale x 8 x half>, <vscale x 8 x half> } %a.deinterleaved, 0
   %a.imag = extractvalue { <vscale x 8 x half>, <vscale x 8 x half> } %a.deinterleaved, 1
-  %b.deinterleaved = tail call { <vscale x 8 x half>, <vscale x 8 x half> } @llvm.experimental.vector.deinterleave2.nxv16f16(<vscale x 16 x half> %b)
+  %b.deinterleaved = tail call { <vscale x 8 x half>, <vscale x 8 x half> } @llvm.vector.deinterleave2.nxv16f16(<vscale x 16 x half> %b)
   %b.real = extractvalue { <vscale x 8 x half>, <vscale x 8 x half> } %b.deinterleaved, 0
   %b.imag = extractvalue { <vscale x 8 x half>, <vscale x 8 x half> } %b.deinterleaved, 1
   %0 = fmul fast <vscale x 8 x half> %b.imag, %a.real
@@ -96,7 +96,7 @@ entry:
   %3 = fmul fast <vscale x 8 x half> %b.real, %a.real
   %4 = fmul fast <vscale x 8 x half> %a.imag, %b.imag
   %5 = fsub fast <vscale x 8 x half> %3, %4
-  %interleaved.vec = tail call <vscale x 16 x half> @llvm.experimental.vector.interleave2.nxv16f16(<vscale x 8 x half> %5, <vscale x 8 x half> %2)
+  %interleaved.vec = tail call <vscale x 16 x half> @llvm.vector.interleave2.nxv16f16(<vscale x 8 x half> %5, <vscale x 8 x half> %2)
   ret <vscale x 16 x half> %interleaved.vec
 }
 
@@ -123,10 +123,10 @@ define <vscale x 32 x half> @complex_mul_v32f16(<vscale x 32 x half> %a, <vscale
 ; CHECK-NEXT:    mov z2.d, z27.d
 ; CHECK-NEXT:    ret
 entry:
-  %a.deinterleaved = tail call { <vscale x 16 x half>, <vscale x 16 x half> } @llvm.experimental.vector.deinterleave2.nxv32f16(<vscale x 32 x half> %a)
+  %a.deinterleaved = tail call { <vscale x 16 x half>, <vscale x 16 x half> } @llvm.vector.deinterleave2.nxv32f16(<vscale x 32 x half> %a)
   %a.real = extractvalue { <vscale x 16 x half>, <vscale x 16 x half> } %a.deinterleaved, 0
   %a.imag = extractvalue { <vscale x 16 x half>, <vscale x 16 x half> } %a.deinterleaved, 1
-  %b.deinterleaved = tail call { <vscale x 16 x half>, <vscale x 16 x half> } @llvm.experimental.vector.deinterleave2.nxv32f16(<vscale x 32 x half> %b)
+  %b.deinterleaved = tail call { <vscale x 16 x half>, <vscale x 16 x half> } @llvm.vector.deinterleave2.nxv32f16(<vscale x 32 x half> %b)
   %b.real = extractvalue { <vscale x 16 x half>, <vscale x 16 x half> } %b.deinterleaved, 0
   %b.imag = extractvalue { <vscale x 16 x half>, <vscale x 16 x half> } %b.deinterleaved, 1
   %0 = fmul fast <vscale x 16 x half> %b.imag, %a.real
@@ -135,20 +135,20 @@ entry:
   %3 = fmul fast <vscale x 16 x half> %b.real, %a.real
   %4 = fmul fast <vscale x 16 x half> %a.imag, %b.imag
   %5 = fsub fast <vscale x 16 x half> %3, %4
-  %interleaved.vec = tail call <vscale x 32 x half> @llvm.experimental.vector.interleave2.nxv32f16(<vscale x 16 x half> %5, <vscale x 16 x half> %2)
+  %interleaved.vec = tail call <vscale x 32 x half> @llvm.vector.interleave2.nxv32f16(<vscale x 16 x half> %5, <vscale x 16 x half> %2)
   ret <vscale x 32 x half> %interleaved.vec
 }
 
-declare { <vscale x 2 x half>, <vscale x 2 x half> } @llvm.experimental.vector.deinterleave2.nxv4f16(<vscale x 4 x half>)
-declare <vscale x 4 x half> @llvm.experimental.vector.interleave2.nxv4f16(<vscale x 2 x half>, <vscale x 2 x half>)
+declare { <vscale x 2 x half>, <vscale x 2 x half> } @llvm.vector.deinterleave2.nxv4f16(<vscale x 4 x half>)
+declare <vscale x 4 x half> @llvm.vector.interleave2.nxv4f16(<vscale x 2 x half>, <vscale x 2 x half>)
 
-declare { <vscale x 4 x half>, <vscale x 4 x half> } @llvm.experimental.vector.deinterleave2.nxv8f16(<vscale x 8 x half>)
-declare <vscale x 8 x half> @llvm.experimental.vector.interleave2.nxv8f16(<vscale x 4 x half>, <vscale x 4 x half>)
+declare { <vscale x 4 x half>, <vscale x 4 x half> } @llvm.vector.deinterleave2.nxv8f16(<vscale x 8 x half>)
+declare <vscale x 8 x half> @llvm.vector.interleave2.nxv8f16(<vscale x 4 x half>, <vscale x 4 x half>)
 
-declare { <vscale x 8 x half>, <vscale x 8 x half> } @llvm.experimental.vector.deinterleave2.nxv16f16(<vscale x 16 x half>)
-declare <vscale x 16 x half> @llvm.experimental.vector.interleave2.nxv16f16(<vscale x 8 x half>, <vscale x 8 x half>)
+declare { <vscale x 8 x half>, <vscale x 8 x half> } @llvm.vector.deinterleave2.nxv16f16(<vscale x 16 x half>)
+declare <vscale x 16 x half> @llvm.vector.interleave2.nxv16f16(<vscale x 8 x half>, <vscale x 8 x half>)
 
-declare { <vscale x 16 x half>, <vscale x 16 x half> } @llvm.experimental.vector.deinterleave2.nxv32f16(<vscale x 32 x half>)
-declare <vscale x 32 x half> @llvm.experimental.vector.interleave2.nxv32f16(<vscale x 16 x half>, <vscale x 16 x half>)
+declare { <vscale x 16 x half>, <vscale x 16 x half> } @llvm.vector.deinterleave2.nxv32f16(<vscale x 32 x half>)
+declare <vscale x 32 x half> @llvm.vector.interleave2.nxv32f16(<vscale x 16 x half>, <vscale x 16 x half>)
 
 
diff --git a/llvm/test/CodeGen/AArch64/complex-deinterleaving-f32-add-scalable.ll b/llvm/test/CodeGen/AArch64/complex-deinterleaving-f32-add-scalable.ll
index ab764a58a770..47ad9ea2451a 100644
--- a/llvm/test/CodeGen/AArch64/complex-deinterleaving-f32-add-scalable.ll
+++ b/llvm/test/CodeGen/AArch64/complex-deinterleaving-f32-add-scalable.ll
@@ -12,15 +12,15 @@ define <vscale x 4 x float> @complex_add_v4f32(<vscale x 4 x float> %a, <vscale
 ; CHECK-NEXT:    mov z0.d, z1.d
 ; CHECK-NEXT:    ret
 entry:
-  %a.deinterleaved = tail call { <vscale x 2 x float>, <vscale x 2 x float> } @llvm.experimental.vector.deinterleave2.nxv4f32(<vscale x 4 x float> %a)
+  %a.deinterleaved = tail call { <vscale x 2 x float>, <vscale x 2 x float> } @llvm.vector.deinterleave2.nxv4f32(<vscale x 4 x float> %a)
   %a.real = extractvalue { <vscale x 2 x float>, <vscale x 2 x float> } %a.deinterleaved, 0
   %a.imag = extractvalue { <vscale x 2 x float>, <vscale x 2 x float> } %a.deinterleaved, 1
-  %b.deinterleaved = tail call { <vscale x 2 x float>, <vscale x 2 x float> } @llvm.experimental.vector.deinterleave2.nxv4f32(<vscale x 4 x float> %b)
+  %b.deinterleaved = tail call { <vscale x 2 x float>, <vscale x 2 x float> } @llvm.vector.deinterleave2.nxv4f32(<vscale x 4 x float> %b)
   %b.real = extractvalue { <vscale x 2 x float>, <vscale x 2 x float> } %b.deinterleaved, 0
   %b.imag = extractvalue { <vscale x 2 x float>, <vscale x 2 x float> } %b.deinterleaved, 1
   %0 = fsub fast <vscale x 2 x float> %b.real, %a.imag
   %1 = fadd fast <vscale x 2 x float> %b.imag, %a.real
-  %interleaved.vec = tail call <vscale x 4 x float> @llvm.experimental.vector.interleave2.nxv4f32(<vscale x 2 x float> %0, <vscale x 2 x float> %1)
+  %interleaved.vec = tail call <vscale x 4 x float> @llvm.vector.interleave2.nxv4f32(<vscale x 2 x float> %0, <vscale x 2 x float> %1)
   ret <vscale x 4 x float> %interleaved.vec
 }
 
@@ -35,15 +35,15 @@ define <vscale x 8 x float> @complex_add_v8f32(<vscale x 8 x float> %a, <vscale
 ; CHECK-NEXT:    mov z1.d, z3.d
 ; CHECK-NEXT:    ret
 entry:
-  %a.deinterleaved = tail call { <vscale x 4 x float>, <vscale x 4 x float> } @llvm.experimental.vector.deinterleave2.nxv8f32(<vscale x 8 x float> %a)
+  %a.deinterleaved = tail call { <vscale x 4 x float>, <vscale x 4 x float> } @llvm.vector.deinterleave2.nxv8f32(<vscale x 8 x float> %a)
   %a.real = extractvalue { <vscale x 4 x float>, <vscale x 4 x float> } %a.deinterleaved, 0
   %a.imag = extractvalue { <vscale x 4 x float>, <vscale x 4 x float> } %a.deinterleaved, 1
-  %b.deinterleaved = tail call { <vscale x 4 x float>, <vscale x 4 x float> } @llvm.experimental.vector.deinterleave2.nxv8f32(<vscale x 8 x float> %b)
+  %b.deinterleaved = tail call { <vscale x 4 x float>, <vscale x 4 x float> } @llvm.vector.deinterleave2.nxv8f32(<vscale x 8 x float> %b)
   %b.real = extractvalue { <vscale x 4 x float>, <vscale x 4 x float> } %b.deinterleaved, 0
   %b.imag = extractvalue { <vscale x 4 x float>, <vscale x 4 x float> } %b.deinterleaved, 1
   %0 = fsub fast <vscale x 4 x float> %b.real, %a.imag
   %1 = fadd fast <vscale x 4 x float> %b.imag, %a.real
-  %interleaved.vec = tail call <vscale x 8 x float> @llvm.experimental.vector.interleave2.nxv8f32(<vscale x 4 x float> %0, <vscale x 4 x float> %1)
+  %interleaved.vec = tail call <vscale x 8 x float> @llvm.vector.interleave2.nxv8f32(<vscale x 4 x float> %0, <vscale x 4 x float> %1)
   ret <vscale x 8 x float> %interleaved.vec
 }
 ; Expected to transform
@@ -61,23 +61,23 @@ define <vscale x 16 x float> @complex_add_v16f32(<vscale x 16 x float> %a, <vsca
 ; CHECK-NEXT:    mov z3.d, z7.d
 ; CHECK-NEXT:    ret
 entry:
-  %a.deinterleaved = tail call { <vscale x 8 x float>, <vscale x 8 x float> } @llvm.experimental.vector.deinterleave2.nxv16f32(<vscale x 16 x float> %a)
+  %a.deinterleaved = tail call { <vscale x 8 x float>, <vscale x 8 x float> } @llvm.vector.deinterleave2.nxv16f32(<vscale x 16 x float> %a)
   %a.real = extractvalue { <vscale x 8 x float>, <vscale x 8 x float> } %a.deinterleaved, 0
   %a.imag = extractvalue { <vscale x 8 x float>, <vscale x 8 x float> } %a.deinterleaved, 1
-  %b.deinterleaved = tail call { <vscale x 8 x float>, <vscale x 8 x float> } @llvm.experimental.vector.deinterleave2.nxv16f32(<vscale x 16 x float> %b)
+  %b.deinterleaved = tail call { <vscale x 8 x float>, <vscale x 8 x float> } @llvm.vector.deinterleave2.nxv16f32(<vscale x 16 x float> %b)
   %b.real = extractvalue { <vscale x 8 x float>, <vscale x 8 x float> } %b.deinterleaved, 0
   %b.imag = extractvalue { <vscale x 8 x float>, <vscale x 8 x float> } %b.deinterleaved, 1
   %0 = fsub fast <vscale x 8 x float> %b.real, %a.imag
   %1 = fadd fast <vscale x 8 x float> %b.imag, %a.real
-  %interleaved.vec = tail call <vscale x 16 x float> @llvm.experimental.vector.interleave2.nxv16f32(<vscale x 8 x float> %0, <vscale x 8 x float> %1)
+  %interleaved.vec = tail call <vscale x 16 x float> @llvm.vector.interleave2.nxv16f32(<vscale x 8 x float> %0, <vscale x 8 x float> %1)
   ret <vscale x 16 x float> %interleaved.vec
 }
 
-declare { <vscale x 2 x float>, <vscale x 2 x float> } @llvm.experimental.vector.deinterleave2.nxv4f32(<vscale x 4 x float>)
-declare <vscale x 4 x float> @llvm.experimental.vector.interleave2.nxv4f32(<vscale x 2 x float>, <vscale x 2 x float>)
+declare { <vscale x 2 x float>, <vscale x 2 x float> } @llvm.vector.deinterleave2.nxv4f32(<vscale x 4 x float>)
+declare <vscale x 4 x float> @llvm.vector.interleave2.nxv4f32(<vscale x 2 x float>, <vscale x 2 x float>)
 
-declare { <vscale x 4 x float>, <vscale x 4 x float> } @llvm.experimental.vector.deinterleave2.nxv8f32(<vscale x 8 x float>)
-declare <vscale x 8 x float> @llvm.experimental.vector.interleave2.nxv8f32(<vscale x 4 x float>, <vscale x 4 x float>)
+declare { <vscale x 4 x float>, <vscale x 4 x float> } @llvm.vector.deinterleave2.nxv8f32(<vscale x 8 x float>)
+declare <vscale x 8 x float> @llvm.vector.interleave2.nxv8f32(<vscale x 4 x float>, <vscale x 4 x float>)
 
-declare { <vscale x 8 x float>, <vscale x 8 x float> } @llvm.experimental.vector.deinterleave2.nxv16f32(<vscale x 16 x float>)
-declare <vscale x 16 x float> @llvm.experimental.vector.interleave2.nxv16f32(<vscale x 8 x float>, <vscale x 8 x float>)
+declare { <vscale x 8 x float>, <vscale x 8 x float> } @llvm.vector.deinterleave2.nxv16f32(<vscale x 16 x float>)
+declare <vscale x 16 x float> @llvm.vector.interleave2.nxv16f32(<vscale x 8 x float>, <vscale x 8 x float>)
diff --git a/llvm/test/CodeGen/AArch64/complex-deinterleaving-f32-mul-scalable.ll b/llvm/test/CodeGen/AArch64/complex-deinterleaving-f32-mul-scalable.ll
index 1e2afb78de1b..bcd46aa182b5 100644
--- a/llvm/test/CodeGen/AArch64/complex-deinterleaving-f32-mul-scalable.ll
+++ b/llvm/test/CodeGen/AArch64/complex-deinterleaving-f32-mul-scalable.ll
@@ -14,10 +14,10 @@ define <vscale x 4 x float> @complex_mul_v4f32(<vscale x 4 x float> %a, <vscale
 ; CHECK-NEXT:    mov z0.d, z2.d
 ; CHECK-NEXT:    ret
 entry:
-  %a.deinterleaved = tail call { <vscale x 2 x float>, <vscale x 2 x float> } @llvm.experimental.vector.deinterleave2.nxv4f32(<vscale x 4 x float> %a)
+  %a.deinterleaved = tail call { <vscale x 2 x float>, <vscale x 2 x float> } @llvm.vector.deinterleave2.nxv4f32(<vscale x 4 x float> %a)
   %a.real = extractvalue { <vscale x 2 x float>, <vscale x 2 x float> } %a.deinterleaved, 0
   %a.imag = extractvalue { <vscale x 2 x float>, <vscale x 2 x float> } %a.deinterleaved, 1
-  %b.deinterleaved = tail call { <vscale x 2 x float>, <vscale x 2 x float> } @llvm.experimental.vector.deinterleave2.nxv4f32(<vscale x 4 x float> %b)
+  %b.deinterleaved = tail call { <vscale x 2 x float>, <vscale x 2 x float> } @llvm.vector.deinterleave2.nxv4f32(<vscale x 4 x float> %b)
   %b.real = extractvalue { <vscale x 2 x float>, <vscale x 2 x float> } %b.deinterleaved, 0
   %b.imag = extractvalue { <vscale x 2 x float>, <vscale x 2 x float> } %b.deinterleaved, 1
   %0 = fmul fast <vscale x 2 x float> %b.imag, %a.real
@@ -26,7 +26,7 @@ entry:
   %3 = fmul fast <vscale x 2 x float> %b.real, %a.real
   %4 = fmul fast <vscale x 2 x float> %a.imag, %b.imag
   %5 = fsub fast <vscale x 2 x float> %3, %4
-  %interleaved.vec = tail call <vscale x 4 x float> @llvm.experimental.vector.interleave2.nxv4f32(<vscale x 2 x float> %5, <vscale x 2 x float> %2)
+  %interleaved.vec = tail call <vscale x 4 x float> @llvm.vector.interleave2.nxv4f32(<vscale x 2 x float> %5, <vscale x 2 x float> %2)
   ret <vscale x 4 x float> %interleaved.vec
 }
 
@@ -45,10 +45,10 @@ define <vscale x 8 x float> @complex_mul_v8f32(<vscale x 8 x float> %a, <vscale
 ; CHECK-NEXT:    mov z0.d, z5.d
 ; CHECK-NEXT:    ret
 entry:
-  %a.deinterleaved = tail call { <vscale x 4 x float>, <vscale x 4 x float> } @llvm.experimental.vector.deinterleave2.nxv8f32(<vscale x 8 x float> %a)
+  %a.deinterleaved = tail call { <vscale x 4 x float>, <vscale x 4 x float> } @llvm.vector.deinterleave2.nxv8f32(<vscale x 8 x float> %a)
   %a.real = extractvalue { <vscale x 4 x float>, <vscale x 4 x float> } %a.deinterleaved, 0
   %a.imag = extractvalue { <vscale x 4 x float>, <vscale x 4 x float> } %a.deinterleaved, 1
-  %b.deinterleaved = tail call { <vscale x 4 x float>, <vscale x 4 x float> } @llvm.experimental.vector.deinterleave2.nxv8f32(<vscale x 8 x float> %b)
+  %b.deinterleaved = tail call { <vscale x 4 x float>, <vscale x 4 x float> } @llvm.vector.deinterleave2.nxv8f32(<vscale x 8 x float> %b)
   %b.real = extractvalue { <vscale x 4 x float>, <vscale x 4 x float> } %b.deinterleaved, 0
   %b.imag = extractvalue { <vscale x 4 x float>, <vscale x 4 x float> } %b.deinterleaved, 1
   %0 = fmul fast <vscale x 4 x float> %b.imag, %a.real
@@ -57,7 +57,7 @@ entry:
   %3 = fmul fast <vscale x 4 x float> %b.real, %a.real
   %4 = fmul fast <vscale x 4 x float> %a.imag, %b.imag
   %5 = fsub fast <vscale x 4 x float> %3, %4
-  %interleaved.vec = tail call <vscale x 8 x float> @llvm.experimental.vector.interleave2.nxv8f32(<vscale x 4 x float> %5, <vscale x 4 x float> %2)
+  %interleaved.vec = tail call <vscale x 8 x float> @llvm.vector.interleave2.nxv8f32(<vscale x 4 x float> %5, <vscale x 4 x float> %2)
   ret <vscale x 8 x float> %interleaved.vec
 }
 
@@ -84,10 +84,10 @@ define <vscale x 16 x float> @complex_mul_v16f32(<vscale x 16 x float> %a, <vsca
 ; CHECK-NEXT:    mov z2.d, z27.d
 ; CHECK-NEXT:    ret
 entry:
-  %a.deinterleaved = tail call { <vscale x 8 x float>, <vscale x 8 x float> } @llvm.experimental.vector.deinterleave2.nxv16f32(<vscale x 16 x float> %a)
+  %a.deinterleaved = tail call { <vscale x 8 x float>, <vscale x 8 x float> } @llvm.vector.deinterleave2.nxv16f32(<vscale x 16 x float> %a)
   %a.real = extractvalue { <vscale x 8 x float>, <vscale x 8 x float> } %a.deinterleaved, 0
   %a.imag = extractvalue { <vscale x 8 x float>, <vscale x 8 x float> } %a.deinterleaved, 1
-  %b.deinterleaved = tail call { <vscale x 8 x float>, <vscale x 8 x float> } @llvm.experimental.vector.deinterleave2.nxv16f32(<vscale x 16 x float> %b)
+  %b.deinterleaved = tail call { <vscale x 8 x float>, <vscale x 8 x float> } @llvm.vector.deinterleave2.nxv16f32(<vscale x 16 x float> %b)
   %b.real = extractvalue { <vscale x 8 x float>, <vscale x 8 x float> } %b.deinterleaved, 0
   %b.imag = extractvalue { <vscale x 8 x float>, <vscale x 8 x float> } %b.deinterleaved, 1
   %0 = fmul fast <vscale x 8 x float> %b.imag, %a.real
@@ -96,16 +96,16 @@ entry:
   %3 = fmul fast <vscale x 8 x float> %b.real, %a.real
   %4 = fmul fast <vscale x 8 x float> %a.imag, %b.imag
   %5 = fsub fast <vscale x 8 x float> %3, %4
-  %interleaved.vec = tail call <vscale x 16 x float> @llvm.experimental.vector.interleave2.nxv16f32(<vscale x 8 x float> %5, <vscale x 8 x float> %2)
+  %interleaved.vec = tail call <vscale x 16 x float> @llvm.vector.interleave2.nxv16f32(<vscale x 8 x float> %5, <vscale x 8 x float> %2)
   ret <vscale x 16 x float> %interleaved.vec
 }
 
-declare { <vscale x 2 x float>, <vscale x 2 x float> } @llvm.experimental.vector.deinterleave2.nxv4f32(<vscale x 4 x float>)
-declare <vscale x 4 x float> @llvm.experimental.vector.interleave2.nxv4f32(<vscale x 2 x float>, <vscale x 2 x float>)
+declare { <vscale x 2 x float>, <vscale x 2 x float> } @llvm.vector.deinterleave2.nxv4f32(<vscale x 4 x float>)
+declare <vscale x 4 x float> @llvm.vector.interleave2.nxv4f32(<vscale x 2 x float>, <vscale x 2 x float>)
 
-declare { <vscale x 4 x float>, <vscale x 4 x float> } @llvm.experimental.vector.deinterleave2.nxv8f32(<vscale x 8 x float>)
-declare <vscale x 8 x float> @llvm.experimental.vector.interleave2.nxv8f32(<vscale x 4 x float>, <vscale x 4 x float>)
+declare { <vscale x 4 x float>, <vscale x 4 x float> } @llvm.vector.deinterleave2.nxv8f32(<vscale x 8 x float>)
+declare <vscale x 8 x float> @llvm.vector.interleave2.nxv8f32(<vscale x 4 x float>, <vscale x 4 x float>)
 
-declare { <vscale x 8 x float>, <vscale x 8 x float> } @llvm.experimental.vector.deinterleave2.nxv16f32(<vscale x 16 x float>)
-declare <vscale x 16 x float> @llvm.experimental.vector.interleave2.nxv16f32(<vscale x 8 x float>, <vscale x 8 x float>)
+declare { <vscale x 8 x float>, <vscale x 8 x float> } @llvm.vector.deinterleave2.nxv16f32(<vscale x 16 x float>)
+declare <vscale x 16 x float> @llvm.vector.interleave2.nxv16f32(<vscale x 8 x float>, <vscale x 8 x float>)
 
diff --git a/llvm/test/CodeGen/AArch64/complex-deinterleaving-f64-add-scalable.ll b/llvm/test/CodeGen/AArch64/complex-deinterleaving-f64-add-scalable.ll
index 46a15f489d2b..c992d63ca283 100644
--- a/llvm/test/CodeGen/AArch64/complex-deinterleaving-f64-add-scalable.ll
+++ b/llvm/test/CodeGen/AArch64/complex-deinterleaving-f64-add-scalable.ll
@@ -12,15 +12,15 @@ define <vscale x 2 x double> @complex_add_v2f64(<vscale x 2 x double> %a, <vscal
 ; CHECK-NEXT:    mov z0.d, z1.d
 ; CHECK-NEXT:    ret
 entry:
-  %a.deinterleaved = tail call { <vscale x 1 x double>, <vscale x 1 x double> } @llvm.experimental.vector.deinterleave2.nxv2f64(<vscale x 2 x double> %a)
+  %a.deinterleaved = tail call { <vscale x 1 x double>, <vscale x 1 x double> } @llvm.vector.deinterleave2.nxv2f64(<vscale x 2 x double> %a)
   %a.real = extractvalue { <vscale x 1 x double>, <vscale x 1 x double> } %a.deinterleaved, 0
   %a.imag = extractvalue { <vscale x 1 x double>, <vscale x 1 x double> } %a.deinterleaved, 1
-  %b.deinterleaved = tail call { <vscale x 1 x double>, <vscale x 1 x double> } @llvm.experimental.vector.deinterleave2.nxv2f64(<vscale x 2 x double> %b)
+  %b.deinterleaved = tail call { <vscale x 1 x double>, <vscale x 1 x double> } @llvm.vector.deinterleave2.nxv2f64(<vscale x 2 x double> %b)
   %b.real = extractvalue { <vscale x 1 x double>, <vscale x 1 x double> } %b.deinterleaved, 0
   %b.imag = extractvalue { <vscale x 1 x double>, <vscale x 1 x double> } %b.deinterleaved, 1
   %0 = fsub fast <vscale x 1 x double> %b.real, %a.imag
   %1 = fadd fast <vscale x 1 x double> %b.imag, %a.real
-  %interleaved.vec = tail call <vscale x 2 x double> @llvm.experimental.vector.interleave2.nxv2f64(<vscale x 1 x double> %0, <vscale x 1 x double> %1)
+  %interleaved.vec = tail call <vscale x 2 x double> @llvm.vector.interleave2.nxv2f64(<vscale x 1 x double> %0, <vscale x 1 x double> %1)
   ret <vscale x 2 x double> %interleaved.vec
 }
 
@@ -35,15 +35,15 @@ define <vscale x 4 x double> @complex_add_v4f64(<vscale x 4 x double> %a, <vscal
 ; CHECK-NEXT:    mov z1.d, z3.d
 ; CHECK-NEXT:    ret
 entry:
-  %a.deinterleaved = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.experimental.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %a)
+  %a.deinterleaved = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %a)
   %a.real = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %a.deinterleaved, 0
   %a.imag = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %a.deinterleaved, 1
-  %b.deinterleaved = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.experimental.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %b)
+  %b.deinterleaved = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %b)
   %b.real = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %b.deinterleaved, 0
   %b.imag = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %b.deinterleaved, 1
   %0 = fsub fast <vscale x 2 x double> %b.real, %a.imag
   %1 = fadd fast <vscale x 2 x double> %b.imag, %a.real
-  %interleaved.vec = tail call <vscale x 4 x double> @llvm.experimental.vector.interleave2.nxv4f64(<vscale x 2 x double> %0, <vscale x 2 x double> %1)
+  %interleaved.vec = tail call <vscale x 4 x double> @llvm.vector.interleave2.nxv4f64(<vscale x 2 x double> %0, <vscale x 2 x double> %1)
   ret <vscale x 4 x double> %interleaved.vec
 }
 
@@ -62,23 +62,23 @@ define <vscale x 8 x double> @complex_add_v8f64(<vscale x 8 x double> %a, <vscal
 ; CHECK-NEXT:    mov z3.d, z7.d
 ; CHECK-NEXT:    ret
 entry:
-  %a.deinterleaved = tail call { <vscale x 4 x double>, <vscale x 4 x double> } @llvm.experimental.vector.deinterleave2.nxv8f64(<vscale x 8 x double> %a)
+  %a.deinterleaved = tail call { <vscale x 4 x double>, <vscale x 4 x double> } @llvm.vector.deinterleave2.nxv8f64(<vscale x 8 x double> %a)
   %a.real = extractvalue { <vscale x 4 x double>, <vscale x 4 x double> } %a.deinterleaved, 0
   %a.imag = extractvalue { <vscale x 4 x double>, <vscale x 4 x double> } %a.deinterleaved, 1
-  %b.deinterleaved = tail call { <vscale x 4 x double>, <vscale x 4 x double> } @llvm.experimental.vector.deinterleave2.nxv8f64(<vscale x 8 x double> %b)
+  %b.deinterleaved = tail call { <vscale x 4 x double>, <vscale x 4 x double> } @llvm.vector.deinterleave2.nxv8f64(<vscale x 8 x double> %b)
   %b.real = extractvalue { <vscale x 4 x double>, <vscale x 4 x double> } %b.deinterleaved, 0
   %b.imag = extractvalue { <vscale x 4 x double>, <vscale x 4 x double> } %b.deinterleaved, 1
   %0 = fsub fast <vscale x 4 x double> %b.real, %a.imag
   %1 = fadd fast <vscale x 4 x double> %b.imag, %a.real
-  %interleaved.vec = tail call <vscale x 8 x double> @llvm.experimental.vector.interleave2.nxv8f64(<vscale x 4 x double> %0, <vscale x 4 x double> %1)
+  %interleaved.vec = tail call <vscale x 8 x double> @llvm.vector.interleave2.nxv8f64(<vscale x 4 x double> %0, <vscale x 4 x double> %1)
   ret <vscale x 8 x double> %interleaved.vec
 }
 
-declare { <vscale x 1 x double>, <vscale x 1 x double> } @llvm.experimental.vector.deinterleave2.nxv2f64(<vscale x 2 x double>)
-declare <vscale x 2 x double> @llvm.experimental.vector.interleave2.nxv2f64(<vscale x 1 x double>, <vscale x 1 x double>)
+declare { <vscale x 1 x double>, <vscale x 1 x double> } @llvm.vector.deinterleave2.nxv2f64(<vscale x 2 x double>)
+declare <vscale x 2 x double> @llvm.vector.interleave2.nxv2f64(<vscale x 1 x double>, <vscale x 1 x double>)
 
-declare { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.experimental.vector.deinterleave2.nxv4f64(<vscale x 4 x double>)
-declare <vscale x 4 x double> @llvm.experimental.vector.interleave2.nxv4f64(<vscale x 2 x double>, <vscale x 2 x double>)
+declare { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.vector.deinterleave2.nxv4f64(<vscale x 4 x double>)
+declare <vscale x 4 x double> @llvm.vector.interleave2.nxv4f64(<vscale x 2 x double>, <vscale x 2 x double>)
 
-declare { <vscale x 4 x double>, <vscale x 4 x double> } @llvm.experimental.vector.deinterleave2.nxv8f64(<vscale x 8 x double>)
-declare <vscale x 8 x double> @llvm.experimental.vector.interleave2.nxv8f64(<vscale x 4 x double>, <vscale x 4 x double>)
+declare { <vscale x 4 x double>, <vscale x 4 x double> } @llvm.vector.deinterleave2.nxv8f64(<vscale x 8 x double>)
+declare <vscale x 8 x double> @llvm.vector.interleave2.nxv8f64(<vscale x 4 x double>, <vscale x 4 x double>)
diff --git a/llvm/test/CodeGen/AArch64/complex-deinterleaving-f64-mul-scalable.ll b/llvm/test/CodeGen/AArch64/complex-deinterleaving-f64-mul-scalable.ll
index 17a239a09a03..db28fa3997cb 100644
--- a/llvm/test/CodeGen/AArch64/complex-deinterleaving-f64-mul-scalable.ll
+++ b/llvm/test/CodeGen/AArch64/complex-deinterleaving-f64-mul-scalable.ll
@@ -14,10 +14,10 @@ define <vscale x 2 x double> @complex_mul_v2f64(<vscale x 2 x double> %a, <vscal
 ; CHECK-NEXT:    mov z0.d, z2.d
 ; CHECK-NEXT:    ret
 entry:
-  %a.deinterleaved = tail call { <vscale x 1 x double>, <vscale x 1 x double> } @llvm.experimental.vector.deinterleave2.nxv2f64(<vscale x 2 x double> %a)
+  %a.deinterleaved = tail call { <vscale x 1 x double>, <vscale x 1 x double> } @llvm.vector.deinterleave2.nxv2f64(<vscale x 2 x double> %a)
   %a.real = extractvalue { <vscale x 1 x double>, <vscale x 1 x double> } %a.deinterleaved, 0
   %a.imag = extractvalue { <vscale x 1 x double>, <vscale x 1 x double> } %a.deinterleaved, 1
-  %b.deinterleaved = tail call { <vscale x 1 x double>, <vscale x 1 x double> } @llvm.experimental.vector.deinterleave2.nxv2f64(<vscale x 2 x double> %b)
+  %b.deinterleaved = tail call { <vscale x 1 x double>, <vscale x 1 x double> } @llvm.vector.deinterleave2.nxv2f64(<vscale x 2 x double> %b)
   %b.real = extractvalue { <vscale x 1 x double>, <vscale x 1 x double> } %b.deinterleaved, 0
   %b.imag = extractvalue { <vscale x 1 x double>, <vscale x 1 x double> } %b.deinterleaved, 1
   %0 = fmul fast <vscale x 1 x double> %b.imag, %a.real
@@ -26,7 +26,7 @@ entry:
   %3 = fmul fast <vscale x 1 x double> %b.real, %a.real
   %4 = fmul fast <vscale x 1 x double> %a.imag, %b.imag
   %5 = fsub fast <vscale x 1 x double> %3, %4
-  %interleaved.vec = tail call <vscale x 2 x double> @llvm.experimental.vector.interleave2.nxv2f64(<vscale x 1 x double> %5, <vscale x 1 x double> %2)
+  %interleaved.vec = tail call <vscale x 2 x double> @llvm.vector.interleave2.nxv2f64(<vscale x 1 x double> %5, <vscale x 1 x double> %2)
   ret <vscale x 2 x double> %interleaved.vec
 }
 
@@ -45,10 +45,10 @@ define <vscale x 4 x double> @complex_mul_v4f64(<vscale x 4 x double> %a, <vscal
 ; CHECK-NEXT:    mov z0.d, z5.d
 ; CHECK-NEXT:    ret
 entry:
-  %a.deinterleaved = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.experimental.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %a)
+  %a.deinterleaved = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %a)
   %a.real = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %a.deinterleaved, 0
   %a.imag = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %a.deinterleaved, 1
-  %b.deinterleaved = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.experimental.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %b)
+  %b.deinterleaved = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %b)
   %b.real = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %b.deinterleaved, 0
   %b.imag = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %b.deinterleaved, 1
   %0 = fmul fast <vscale x 2 x double> %b.imag, %a.real
@@ -57,7 +57,7 @@ entry:
   %3 = fmul fast <vscale x 2 x double> %b.real, %a.real
   %4 = fmul fast <vscale x 2 x double> %a.imag, %b.imag
   %5 = fsub fast <vscale x 2 x double> %3, %4
-  %interleaved.vec = tail call <vscale x 4 x double> @llvm.experimental.vector.interleave2.nxv4f64(<vscale x 2 x double> %5, <vscale x 2 x double> %2)
+  %interleaved.vec = tail call <vscale x 4 x double> @llvm.vector.interleave2.nxv4f64(<vscale x 2 x double> %5, <vscale x 2 x double> %2)
   ret <vscale x 4 x double> %interleaved.vec
 }
 
@@ -84,10 +84,10 @@ define <vscale x 8 x double> @complex_mul_v8f64(<vscale x 8 x double> %a, <vscal
 ; CHECK-NEXT:    mov z2.d, z27.d
 ; CHECK-NEXT:    ret
 entry:
-  %a.deinterleaved = tail call { <vscale x 4 x double>, <vscale x 4 x double> } @llvm.experimental.vector.deinterleave2.nxv8f64(<vscale x 8 x double> %a)
+  %a.deinterleaved = tail call { <vscale x 4 x double>, <vscale x 4 x double> } @llvm.vector.deinterleave2.nxv8f64(<vscale x 8 x double> %a)
   %a.real = extractvalue { <vscale x 4 x double>, <vscale x 4 x double> } %a.deinterleaved, 0
   %a.imag = extractvalue { <vscale x 4 x double>, <vscale x 4 x double> } %a.deinterleaved, 1
-  %b.deinterleaved = tail call { <vscale x 4 x double>, <vscale x 4 x double> } @llvm.experimental.vector.deinterleave2.nxv8f64(<vscale x 8 x double> %b)
+  %b.deinterleaved = tail call { <vscale x 4 x double>, <vscale x 4 x double> } @llvm.vector.deinterleave2.nxv8f64(<vscale x 8 x double> %b)
   %b.real = extractvalue { <vscale x 4 x double>, <vscale x 4 x double> } %b.deinterleaved, 0
   %b.imag = extractvalue { <vscale x 4 x double>, <vscale x 4 x double> } %b.deinterleaved, 1
   %0 = fmul fast <vscale x 4 x double> %b.imag, %a.real
@@ -96,15 +96,15 @@ entry:
   %3 = fmul fast <vscale x 4 x double> %b.real, %a.real
   %4 = fmul fast <vscale x 4 x double> %a.imag, %b.imag
   %5 = fsub fast <vscale x 4 x double> %3, %4
-  %interleaved.vec = tail call <vscale x 8 x double> @llvm.experimental.vector.interleave2.nxv8f64(<vscale x 4 x double> %5, <vscale x 4 x double> %2)
+  %interleaved.vec = tail call <vscale x 8 x double> @llvm.vector.interleave2.nxv8f64(<vscale x 4 x double> %5, <vscale x 4 x double> %2)
   ret <vscale x 8 x double> %interleaved.vec
 }
 
-declare { <vscale x 1 x double>, <vscale x 1 x double> } @llvm.experimental.vector.deinterleave2.nxv2f64(<vscale x 2 x double>)
-declare <vscale x 2 x double> @llvm.experimental.vector.interleave2.nxv2f64(<vscale x 1 x double>, <vscale x 1 x double>)
+declare { <vscale x 1 x double>, <vscale x 1 x double> } @llvm.vector.deinterleave2.nxv2f64(<vscale x 2 x double>)
+declare <vscale x 2 x double> @llvm.vector.interleave2.nxv2f64(<vscale x 1 x double>, <vscale x 1 x double>)
 
-declare { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.experimental.vector.deinterleave2.nxv4f64(<vscale x 4 x double>)
-declare <vscale x 4 x double> @llvm.experimental.vector.interleave2.nxv4f64(<vscale x 2 x double>, <vscale x 2 x double>)
+declare { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.vector.deinterleave2.nxv4f64(<vscale x 4 x double>)
+declare <vscale x 4 x double> @llvm.vector.interleave2.nxv4f64(<vscale x 2 x double>, <vscale x 2 x double>)
 
-declare { <vscale x 4 x double>, <vscale x 4 x double> } @llvm.experimental.vector.deinterleave2.nxv8f64(<vscale x 8 x double>)
-declare <vscale x 8 x double> @llvm.experimental.vector.interleave2.nxv8f64(<vscale x 4 x double>, <vscale x 4 x double>)
+declare { <vscale x 4 x double>, <vscale x 4 x double> } @llvm.vector.deinterleave2.nxv8f64(<vscale x 8 x double>)
+declare <vscale x 8 x double> @llvm.vector.interleave2.nxv8f64(<vscale x 4 x double>, <vscale x 4 x double>)
diff --git a/llvm/test/CodeGen/AArch64/complex-deinterleaving-i16-add-scalable.ll b/llvm/test/CodeGen/AArch64/complex-deinterleaving-i16-add-scalable.ll
index 001046f8f397..f0569674c651 100644
--- a/llvm/test/CodeGen/AArch64/complex-deinterleaving-i16-add-scalable.ll
+++ b/llvm/test/CodeGen/AArch64/complex-deinterleaving-i16-add-scalable.ll
@@ -22,15 +22,15 @@ define <vscale x 4 x i16> @complex_add_v4i16(<vscale x 4 x i16> %a, <vscale x 4
 ; CHECK-NEXT:    uzp1 z0.s, z0.s, z2.s
 ; CHECK-NEXT:    ret
 entry:
-  %a.deinterleaved = tail call { <vscale x 2 x i16>, <vscale x 2 x i16> } @llvm.experimental.vector.deinterleave2.nxv4i16(<vscale x 4 x i16> %a)
+  %a.deinterleaved = tail call { <vscale x 2 x i16>, <vscale x 2 x i16> } @llvm.vector.deinterleave2.nxv4i16(<vscale x 4 x i16> %a)
   %a.real = extractvalue { <vscale x 2 x i16>, <vscale x 2 x i16> } %a.deinterleaved, 0
   %a.imag = extractvalue { <vscale x 2 x i16>, <vscale x 2 x i16> } %a.deinterleaved, 1
-  %b.deinterleaved = tail call { <vscale x 2 x i16>, <vscale x 2 x i16> } @llvm.experimental.vector.deinterleave2.nxv4i16(<vscale x 4 x i16> %b)
+  %b.deinterleaved = tail call { <vscale x 2 x i16>, <vscale x 2 x i16> } @llvm.vector.deinterleave2.nxv4i16(<vscale x 4 x i16> %b)
   %b.real = extractvalue { <vscale x 2 x i16>, <vscale x 2 x i16> } %b.deinterleaved, 0
   %b.imag = extractvalue { <vscale x 2 x i16>, <vscale x 2 x i16> } %b.deinterleaved, 1
   %0 = sub <vscale x 2 x i16> %b.real, %a.imag
   %1 = add <vscale x 2 x i16> %b.imag, %a.real
-  %interleaved.vec = tail call <vscale x 4 x i16> @llvm.experimental.vector.interleave2.nxv4i16(<vscale x 2 x i16> %0, <vscale x 2 x i16> %1)
+  %interleaved.vec = tail call <vscale x 4 x i16> @llvm.vector.interleave2.nxv4i16(<vscale x 2 x i16> %0, <vscale x 2 x i16> %1)
   ret <vscale x 4 x i16> %interleaved.vec
 }
 
@@ -42,15 +42,15 @@ define <vscale x 8 x i16> @complex_add_v8i16(<vscale x 8 x i16> %a, <vscale x 8
 ; CHECK-NEXT:    mov z0.d, z1.d
 ; CHECK-NEXT:    ret
 entry:
-  %a.deinterleaved = tail call { <vscale x 4 x i16>, <vscale x 4 x i16> } @llvm.experimental.vector.deinterleave2.nxv8i16(<vscale x 8 x i16> %a)
+  %a.deinterleaved = tail call { <vscale x 4 x i16>, <vscale x 4 x i16> } @llvm.vector.deinterleave2.nxv8i16(<vscale x 8 x i16> %a)
   %a.real = extractvalue { <vscale x 4 x i16>, <vscale x 4 x i16> } %a.deinterleaved, 0
   %a.imag = extractvalue { <vscale x 4 x i16>, <vscale x 4 x i16> } %a.deinterleaved, 1
-  %b.deinterleaved = tail call { <vscale x 4 x i16>, <vscale x 4 x i16> } @llvm.experimental.vector.deinterleave2.nxv8i16(<vscale x 8 x i16> %b)
+  %b.deinterleaved = tail call { <vscale x 4 x i16>, <vscale x 4 x i16> } @llvm.vector.deinterleave2.nxv8i16(<vscale x 8 x i16> %b)
   %b.real = extractvalue { <vscale x 4 x i16>, <vscale x 4 x i16> } %b.deinterleaved, 0
   %b.imag = extractvalue { <vscale x 4 x i16>, <vscale x 4 x i16> } %b.deinterleaved, 1
   %0 = sub <vscale x 4 x i16> %b.real, %a.imag
   %1 = add <vscale x 4 x i16> %b.imag, %a.real
-  %interleaved.vec = tail call <vscale x 8 x i16> @llvm.experimental.vector.interleave2.nxv8i16(<vscale x 4 x i16> %0, <vscale x 4 x i16> %1)
+  %interleaved.vec = tail call <vscale x 8 x i16> @llvm.vector.interleave2.nxv8i16(<vscale x 4 x i16> %0, <vscale x 4 x i16> %1)
   ret <vscale x 8 x i16> %interleaved.vec
 }
 
@@ -64,15 +64,15 @@ define <vscale x 16 x i16> @complex_add_v16i16(<vscale x 16 x i16> %a, <vscale x
 ; CHECK-NEXT:    mov z1.d, z3.d
 ; CHECK-NEXT:    ret
 entry:
-  %a.deinterleaved = tail call { <vscale x 8 x i16>, <vscale x 8 x i16> } @llvm.experimental.vector.deinterleave2.nxv16i16(<vscale x 16 x i16> %a)
+  %a.deinterleaved = tail call { <vscale x 8 x i16>, <vscale x 8 x i16> } @llvm.vector.deinterleave2.nxv16i16(<vscale x 16 x i16> %a)
   %a.real = extractvalue { <vscale x 8 x i16>, <vscale x 8 x i16> } %a.deinterleaved, 0
   %a.imag = extractvalue { <vscale x 8 x i16>, <vscale x 8 x i16> } %a.deinterleaved, 1
-  %b.deinterleaved = tail call { <vscale x 8 x i16>, <vscale x 8 x i16> } @llvm.experimental.vector.deinterleave2.nxv16i16(<vscale x 16 x i16> %b)
+  %b.deinterleaved = tail call { <vscale x 8 x i16>, <vscale x 8 x i16> } @llvm.vector.deinterleave2.nxv16i16(<vscale x 16 x i16> %b)
   %b.real = extractvalue { <vscale x 8 x i16>, <vscale x 8 x i16> } %b.deinterleaved, 0
   %b.imag = extractvalue { <vscale x 8 x i16>, <vscale x 8 x i16> } %b.deinterleaved, 1
   %0 = sub <vscale x 8 x i16> %b.real, %a.imag
   %1 = add <vscale x 8 x i16> %b.imag, %a.real
-  %interleaved.vec = tail call <vscale x 16 x i16> @llvm.experimental.vector.interleave2.nxv16i16(<vscale x 8 x i16> %0, <vscale x 8 x i16> %1)
+  %interleaved.vec = tail call <vscale x 16 x i16> @llvm.vector.interleave2.nxv16i16(<vscale x 8 x i16> %0, <vscale x 8 x i16> %1)
   ret <vscale x 16 x i16> %interleaved.vec
 }
 
@@ -90,26 +90,26 @@ define <vscale x 32 x i16> @complex_add_v32i16(<vscale x 32 x i16> %a, <vscale x
 ; CHECK-NEXT:    mov z3.d, z7.d
 ; CHECK-NEXT:    ret
 entry:
-  %a.deinterleaved = tail call { <vscale x 16 x i16>, <vscale x 16 x i16> } @llvm.experimental.vector.deinterleave2.nxv32i16(<vscale x 32 x i16> %a)
+  %a.deinterleaved = tail call { <vscale x 16 x i16>, <vscale x 16 x i16> } @llvm.vector.deinterleave2.nxv32i16(<vscale x 32 x i16> %a)
   %a.real = extractvalue { <vscale x 16 x i16>, <vscale x 16 x i16> } %a.deinterleaved, 0
   %a.imag = extractvalue { <vscale x 16 x i16>, <vscale x 16 x i16> } %a.deinterleaved, 1
-  %b.deinterleaved = tail call { <vscale x 16 x i16>, <vscale x 16 x i16> } @llvm.experimental.vector.deinterleave2.nxv32i16(<vscale x 32 x i16> %b)
+  %b.deinterleaved = tail call { <vscale x 16 x i16>, <vscale x 16 x i16> } @llvm.vector.deinterleave2.nxv32i16(<vscale x 32 x i16> %b)
   %b.real = extractvalue { <vscale x 16 x i16>, <vscale x 16 x i16> } %b.deinterleaved, 0
   %b.imag = extractvalue { <vscale x 16 x i16>, <vscale x 16 x i16> } %b.deinterleaved, 1
   %0 = sub <vscale x 16 x i16> %b.real, %a.imag
   %1 = add <vscale x 16 x i16> %b.imag, %a.real
-  %interleaved.vec = tail call <vscale x 32 x i16> @llvm.experimental.vector.interleave2.nxv32i16(<vscale x 16 x i16> %0, <vscale x 16 x i16> %1)
+  %interleaved.vec = tail call <vscale x 32 x i16> @llvm.vector.interleave2.nxv32i16(<vscale x 16 x i16> %0, <vscale x 16 x i16> %1)
   ret <vscale x 32 x i16> %interleaved.vec
 }
 
-declare { <vscale x 2 x i16>, <vscale x 2 x i16> } @llvm.experimental.vector.deinterleave2.nxv4i16(<vscale x 4 x i16>)
-declare <vscale x 4 x i16> @llvm.experimental.vector.interleave2.nxv4i16(<vscale x 2 x i16>, <vscale x 2 x i16>)
+declare { <vscale x 2 x i16>, <vscale x 2 x i16> } @llvm.vector.deinterleave2.nxv4i16(<vscale x 4 x i16>)
+declare <vscale x 4 x i16> @llvm.vector.interleave2.nxv4i16(<vscale x 2 x i16>, <vscale x 2 x i16>)
 
-declare { <vscale x 4 x i16>, <vscale x 4 x i16> } @llvm.experimental.vector.deinterleave2.nxv8i16(<vscale x 8 x i16>)
-declare <vscale x 8 x i16> @llvm.experimental.vector.interleave2.nxv8i16(<vscale x 4 x i16>, <vscale x 4 x i16>)
+declare { <vscale x 4 x i16>, <vscale x 4 x i16> } @llvm.vector.deinterleave2.nxv8i16(<vscale x 8 x i16>)
+declare <vscale x 8 x i16> @llvm.vector.interleave2.nxv8i16(<vscale x 4 x i16>, <vscale x 4 x i16>)
 
-declare { <vscale x 8 x i16>, <vscale x 8 x i16> } @llvm.experimental.vector.deinterleave2.nxv16i16(<vscale x 16 x i16>)
-declare <vscale x 16 x i16> @llvm.experimental.vector.interleave2.nxv16i16(<vscale x 8 x i16>, <vscale x 8 x i16>)
+declare { <vscale x 8 x i16>, <vscale x 8 x i16> } @llvm.vector.deinterleave2.nxv16i16(<vscale x 16 x i16>)
+declare <vscale x 16 x i16> @llvm.vector.interleave2.nxv16i16(<vscale x 8 x i16>, <vscale x 8 x i16>)
 
-declare { <vscale x 16 x i16>, <vscale x 16 x i16> } @llvm.experimental.vector.deinterleave2.nxv32i16(<vscale x 32 x i16>)
-declare <vscale x 32 x i16> @llvm.experimental.vector.interleave2.nxv32i16(<vscale x 16 x i16>, <vscale x 16 x i16>)
+declare { <vscale x 16 x i16>, <vscale x 16 x i16> } @llvm.vector.deinterleave2.nxv32i16(<vscale x 32 x i16>)
+declare <vscale x 32 x i16> @llvm.vector.interleave2.nxv32i16(<vscale x 16 x i16>, <vscale x 16 x i16>)
diff --git a/llvm/test/CodeGen/AArch64/complex-deinterleaving-i16-mul-scalable.ll b/llvm/test/CodeGen/AArch64/complex-deinterleaving-i16-mul-scalable.ll
index 07488b623b98..b4cb548f6308 100644
--- a/llvm/test/CodeGen/AArch64/complex-deinterleaving-i16-mul-scalable.ll
+++ b/llvm/test/CodeGen/AArch64/complex-deinterleaving-i16-mul-scalable.ll
@@ -26,10 +26,10 @@ define <vscale x 4 x i16> @complex_mul_v4i16(<vscale x 4 x i16> %a, <vscale x 4
 ; CHECK-NEXT:    uzp1 z0.s, z0.s, z1.s
 ; CHECK-NEXT:    ret
 entry:
-  %a.deinterleaved = tail call { <vscale x 2 x i16>, <vscale x 2 x i16> } @llvm.experimental.vector.deinterleave2.nxv4i16(<vscale x 4 x i16> %a)
+  %a.deinterleaved = tail call { <vscale x 2 x i16>, <vscale x 2 x i16> } @llvm.vector.deinterleave2.nxv4i16(<vscale x 4 x i16> %a)
   %a.real = extractvalue { <vscale x 2 x i16>, <vscale x 2 x i16> } %a.deinterleaved, 0
   %a.imag = extractvalue { <vscale x 2 x i16>, <vscale x 2 x i16> } %a.deinterleaved, 1
-  %b.deinterleaved = tail call { <vscale x 2 x i16>, <vscale x 2 x i16> } @llvm.experimental.vector.deinterleave2.nxv4i16(<vscale x 4 x i16> %b)
+  %b.deinterleaved = tail call { <vscale x 2 x i16>, <vscale x 2 x i16> } @llvm.vector.deinterleave2.nxv4i16(<vscale x 4 x i16> %b)
   %b.real = extractvalue { <vscale x 2 x i16>, <vscale x 2 x i16> } %b.deinterleaved, 0
   %b.imag = extractvalue { <vscale x 2 x i16>, <vscale x 2 x i16> } %b.deinterleaved, 1
   %0 = mul <vscale x 2 x i16> %b.imag, %a.real
@@ -38,7 +38,7 @@ entry:
   %3 = mul <vscale x 2 x i16> %b.real, %a.real
   %4 = mul <vscale x 2 x i16> %a.imag, %b.imag
   %5 = sub <vscale x 2 x i16> %3, %4
-  %interleaved.vec = tail call <vscale x 4 x i16> @llvm.experimental.vector.interleave2.nxv4i16(<vscale x 2 x i16> %5, <vscale x 2 x i16> %2)
+  %interleaved.vec = tail call <vscale x 4 x i16> @llvm.vector.interleave2.nxv4i16(<vscale x 2 x i16> %5, <vscale x 2 x i16> %2)
   ret <vscale x 4 x i16> %interleaved.vec
 }
 
@@ -52,10 +52,10 @@ define <vscale x 8 x i16> @complex_mul_v8i16(<vscale x 8 x i16> %a, <vscale x 8
 ; CHECK-NEXT:    mov z0.d, z2.d
 ; CHECK-NEXT:    ret
 entry:
-  %a.deinterleaved = tail call { <vscale x 4 x i16>, <vscale x 4 x i16> } @llvm.experimental.vector.deinterleave2.nxv8i16(<vscale x 8 x i16> %a)
+  %a.deinterleaved = tail call { <vscale x 4 x i16>, <vscale x 4 x i16> } @llvm.vector.deinterleave2.nxv8i16(<vscale x 8 x i16> %a)
   %a.real = extractvalue { <vscale x 4 x i16>, <vscale x 4 x i16> } %a.deinterleaved, 0
   %a.imag = extractvalue { <vscale x 4 x i16>, <vscale x 4 x i16> } %a.deinterleaved, 1
-  %b.deinterleaved = tail call { <vscale x 4 x i16>, <vscale x 4 x i16> } @llvm.experimental.vector.deinterleave2.nxv8i16(<vscale x 8 x i16> %b)
+  %b.deinterleaved = tail call { <vscale x 4 x i16>, <vscale x 4 x i16> } @llvm.vector.deinterleave2.nxv8i16(<vscale x 8 x i16> %b)
   %b.real = extractvalue { <vscale x 4 x i16>, <vscale x 4 x i16> } %b.deinterleaved, 0
   %b.imag = extractvalue { <vscale x 4 x i16>, <vscale x 4 x i16> } %b.deinterleaved, 1
   %0 = mul <vscale x 4 x i16> %b.imag, %a.real
@@ -64,7 +64,7 @@ entry:
   %3 = mul <vscale x 4 x i16> %b.real, %a.real
   %4 = mul <vscale x 4 x i16> %a.imag, %b.imag
   %5 = sub <vscale x 4 x i16> %3, %4
-  %interleaved.vec = tail call <vscale x 8 x i16> @llvm.experimental.vector.interleave2.nxv8i16(<vscale x 4 x i16> %5, <vscale x 4 x i16> %2)
+  %interleaved.vec = tail call <vscale x 8 x i16> @llvm.vector.interleave2.nxv8i16(<vscale x 4 x i16> %5, <vscale x 4 x i16> %2)
   ret <vscale x 8 x i16> %interleaved.vec
 }
 ; Expected to transform
@@ -81,10 +81,10 @@ define <vscale x 16 x i16> @complex_mul_v16i16(<vscale x 16 x i16> %a, <vscale x
 ; CHECK-NEXT:    mov z0.d, z5.d
 ; CHECK-NEXT:    ret
 entry:
-  %a.deinterleaved = tail call { <vscale x 8 x i16>, <vscale x 8 x i16> } @llvm.experimental.vector.deinterleave2.nxv16i16(<vscale x 16 x i16> %a)
+  %a.deinterleaved = tail call { <vscale x 8 x i16>, <vscale x 8 x i16> } @llvm.vector.deinterleave2.nxv16i16(<vscale x 16 x i16> %a)
   %a.real = extractvalue { <vscale x 8 x i16>, <vscale x 8 x i16> } %a.deinterleaved, 0
   %a.imag = extractvalue { <vscale x 8 x i16>, <vscale x 8 x i16> } %a.deinterleaved, 1
-  %b.deinterleaved = tail call { <vscale x 8 x i16>, <vscale x 8 x i16> } @llvm.experimental.vector.deinterleave2.nxv16i16(<vscale x 16 x i16> %b)
+  %b.deinterleaved = tail call { <vscale x 8 x i16>, <vscale x 8 x i16> } @llvm.vector.deinterleave2.nxv16i16(<vscale x 16 x i16> %b)
   %b.real = extractvalue { <vscale x 8 x i16>, <vscale x 8 x i16> } %b.deinterleaved, 0
   %b.imag = extractvalue { <vscale x 8 x i16>, <vscale x 8 x i16> } %b.deinterleaved, 1
   %0 = mul <vscale x 8 x i16> %b.imag, %a.real
@@ -93,7 +93,7 @@ entry:
   %3 = mul <vscale x 8 x i16> %b.real, %a.real
   %4 = mul <vscale x 8 x i16> %a.imag, %b.imag
   %5 = sub <vscale x 8 x i16> %3, %4
-  %interleaved.vec = tail call <vscale x 16 x i16> @llvm.experimental.vector.interleave2.nxv16i16(<vscale x 8 x i16> %5, <vscale x 8 x i16> %2)
+  %interleaved.vec = tail call <vscale x 16 x i16> @llvm.vector.interleave2.nxv16i16(<vscale x 8 x i16> %5, <vscale x 8 x i16> %2)
   ret <vscale x 16 x i16> %interleaved.vec
 }
 
@@ -119,10 +119,10 @@ define <vscale x 32 x i16> @complex_mul_v32i16(<vscale x 32 x i16> %a, <vscale x
 ; CHECK-NEXT:    mov z2.d, z27.d
 ; CHECK-NEXT:    ret
 entry:
-  %a.deinterleaved = tail call { <vscale x 16 x i16>, <vscale x 16 x i16> } @llvm.experimental.vector.deinterleave2.nxv32i16(<vscale x 32 x i16> %a)
+  %a.deinterleaved = tail call { <vscale x 16 x i16>, <vscale x 16 x i16> } @llvm.vector.deinterleave2.nxv32i16(<vscale x 32 x i16> %a)
   %a.real = extractvalue { <vscale x 16 x i16>, <vscale x 16 x i16> } %a.deinterleaved, 0
   %a.imag = extractvalue { <vscale x 16 x i16>, <vscale x 16 x i16> } %a.deinterleaved, 1
-  %b.deinterleaved = tail call { <vscale x 16 x i16>, <vscale x 16 x i16> } @llvm.experimental.vector.deinterleave2.nxv32i16(<vscale x 32 x i16> %b)
+  %b.deinterleaved = tail call { <vscale x 16 x i16>, <vscale x 16 x i16> } @llvm.vector.deinterleave2.nxv32i16(<vscale x 32 x i16> %b)
   %b.real = extractvalue { <vscale x 16 x i16>, <vscale x 16 x i16> } %b.deinterleaved, 0
   %b.imag = extractvalue { <vscale x 16 x i16>, <vscale x 16 x i16> } %b.deinterleaved, 1
   %0 = mul <vscale x 16 x i16> %b.imag, %a.real
@@ -131,20 +131,20 @@ entry:
   %3 = mul <vscale x 16 x i16> %b.real, %a.real
   %4 = mul <vscale x 16 x i16> %a.imag, %b.imag
   %5 = sub <vscale x 16 x i16> %3, %4
-  %interleaved.vec = tail call <vscale x 32 x i16> @llvm.experimental.vector.interleave2.nxv32i16(<vscale x 16 x i16> %5, <vscale x 16 x i16> %2)
+  %interleaved.vec = tail call <vscale x 32 x i16> @llvm.vector.interleave2.nxv32i16(<vscale x 16 x i16> %5, <vscale x 16 x i16> %2)
   ret <vscale x 32 x i16> %interleaved.vec
 }
 
-declare { <vscale x 2 x i16>, <vscale x 2 x i16> } @llvm.experimental.vector.deinterleave2.nxv4i16(<vscale x 4 x i16>)
-declare <vscale x 4 x i16> @llvm.experimental.vector.interleave2.nxv4i16(<vscale x 2 x i16>, <vscale x 2 x i16>)
+declare { <vscale x 2 x i16>, <vscale x 2 x i16> } @llvm.vector.deinterleave2.nxv4i16(<vscale x 4 x i16>)
+declare <vscale x 4 x i16> @llvm.vector.interleave2.nxv4i16(<vscale x 2 x i16>, <vscale x 2 x i16>)
 
-declare { <vscale x 4 x i16>, <vscale x 4 x i16> } @llvm.experimental.vector.deinterleave2.nxv8i16(<vscale x 8 x i16>)
-declare <vscale x 8 x i16> @llvm.experimental.vector.interleave2.nxv8i16(<vscale x 4 x i16>, <vscale x 4 x i16>)
+declare { <vscale x 4 x i16>, <vscale x 4 x i16> } @llvm.vector.deinterleave2.nxv8i16(<vscale x 8 x i16>)
+declare <vscale x 8 x i16> @llvm.vector.interleave2.nxv8i16(<vscale x 4 x i16>, <vscale x 4 x i16>)
 
-declare { <vscale x 8 x i16>, <vscale x 8 x i16> } @llvm.experimental.vector.deinterleave2.nxv16i16(<vscale x 16 x i16>)
-declare <vscale x 16 x i16> @llvm.experimental.vector.interleave2.nxv16i16(<vscale x 8 x i16>, <vscale x 8 x i16>)
+declare { <vscale x 8 x i16>, <vscale x 8 x i16> } @llvm.vector.deinterleave2.nxv16i16(<vscale x 16 x i16>)
+declare <vscale x 16 x i16> @llvm.vector.interleave2.nxv16i16(<vscale x 8 x i16>, <vscale x 8 x i16>)
 
-declare { <vscale x 16 x i16>, <vscale x 16 x i16> } @llvm.experimental.vector.deinterleave2.nxv32i16(<vscale x 32 x i16>)
-declare <vscale x 32 x i16> @llvm.experimental.vector.interleave2.nxv32i16(<vscale x 16 x i16>, <vscale x 16 x i16>)
+declare { <vscale x 16 x i16>, <vscale x 16 x i16> } @llvm.vector.deinterleave2.nxv32i16(<vscale x 32 x i16>)
+declare <vscale x 32 x i16> @llvm.vector.interleave2.nxv32i16(<vscale x 16 x i16>, <vscale x 16 x i16>)
 
 
diff --git a/llvm/test/CodeGen/AArch64/complex-deinterleaving-i32-add-scalable.ll b/llvm/test/CodeGen/AArch64/complex-deinterleaving-i32-add-scalable.ll
index 1ce480bbf3d8..458cd62269f8 100644
--- a/llvm/test/CodeGen/AArch64/complex-deinterleaving-i32-add-scalable.ll
+++ b/llvm/test/CodeGen/AArch64/complex-deinterleaving-i32-add-scalable.ll
@@ -11,15 +11,15 @@ define <vscale x 4 x i32> @complex_add_v4i32(<vscale x 4 x i32> %a, <vscale x 4
 ; CHECK-NEXT:    mov z0.d, z1.d
 ; CHECK-NEXT:    ret
 entry:
-  %a.deinterleaved = tail call { <vscale x 2 x i32>, <vscale x 2 x i32> } @llvm.experimental.vector.deinterleave2.nxv4i32(<vscale x 4 x i32> %a)
+  %a.deinterleaved = tail call { <vscale x 2 x i32>, <vscale x 2 x i32> } @llvm.vector.deinterleave2.nxv4i32(<vscale x 4 x i32> %a)
   %a.real = extractvalue { <vscale x 2 x i32>, <vscale x 2 x i32> } %a.deinterleaved, 0
   %a.imag = extractvalue { <vscale x 2 x i32>, <vscale x 2 x i32> } %a.deinterleaved, 1
-  %b.deinterleaved = tail call { <vscale x 2 x i32>, <vscale x 2 x i32> } @llvm.experimental.vector.deinterleave2.nxv4i32(<vscale x 4 x i32> %b)
+  %b.deinterleaved = tail call { <vscale x 2 x i32>, <vscale x 2 x i32> } @llvm.vector.deinterleave2.nxv4i32(<vscale x 4 x i32> %b)
   %b.real = extractvalue { <vscale x 2 x i32>, <vscale x 2 x i32> } %b.deinterleaved, 0
   %b.imag = extractvalue { <vscale x 2 x i32>, <vscale x 2 x i32> } %b.deinterleaved, 1
   %0 = sub <vscale x 2 x i32> %b.real, %a.imag
   %1 = add <vscale x 2 x i32> %b.imag, %a.real
-  %interleaved.vec = tail call <vscale x 4 x i32> @llvm.experimental.vector.interleave2.nxv4i32(<vscale x 2 x i32> %0, <vscale x 2 x i32> %1)
+  %interleaved.vec = tail call <vscale x 4 x i32> @llvm.vector.interleave2.nxv4i32(<vscale x 2 x i32> %0, <vscale x 2 x i32> %1)
   ret <vscale x 4 x i32> %interleaved.vec
 }
 
@@ -33,15 +33,15 @@ define <vscale x 8 x i32> @complex_add_v8i32(<vscale x 8 x i32> %a, <vscale x 8
 ; CHECK-NEXT:    mov z1.d, z3.d
 ; CHECK-NEXT:    ret
 entry:
-  %a.deinterleaved = tail call { <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.experimental.vector.deinterleave2.nxv8i32(<vscale x 8 x i32> %a)
+  %a.deinterleaved = tail call { <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.vector.deinterleave2.nxv8i32(<vscale x 8 x i32> %a)
   %a.real = extractvalue { <vscale x 4 x i32>, <vscale x 4 x i32> } %a.deinterleaved, 0
   %a.imag = extractvalue { <vscale x 4 x i32>, <vscale x 4 x i32> } %a.deinterleaved, 1
-  %b.deinterleaved = tail call { <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.experimental.vector.deinterleave2.nxv8i32(<vscale x 8 x i32> %b)
+  %b.deinterleaved = tail call { <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.vector.deinterleave2.nxv8i32(<vscale x 8 x i32> %b)
   %b.real = extractvalue { <vscale x 4 x i32>, <vscale x 4 x i32> } %b.deinterleaved, 0
   %b.imag = extractvalue { <vscale x 4 x i32>, <vscale x 4 x i32> } %b.deinterleaved, 1
   %0 = sub <vscale x 4 x i32> %b.real, %a.imag
   %1 = add <vscale x 4 x i32> %b.imag, %a.real
-  %interleaved.vec = tail call <vscale x 8 x i32> @llvm.experimental.vector.interleave2.nxv8i32(<vscale x 4 x i32> %0, <vscale x 4 x i32> %1)
+  %interleaved.vec = tail call <vscale x 8 x i32> @llvm.vector.interleave2.nxv8i32(<vscale x 4 x i32> %0, <vscale x 4 x i32> %1)
   ret <vscale x 8 x i32> %interleaved.vec
 }
 
@@ -59,23 +59,23 @@ define <vscale x 16 x i32> @complex_add_v16i32(<vscale x 16 x i32> %a, <vscale x
 ; CHECK-NEXT:    mov z3.d, z7.d
 ; CHECK-NEXT:    ret
 entry:
-  %a.deinterleaved = tail call { <vscale x 8 x i32>, <vscale x 8 x i32> } @llvm.experimental.vector.deinterleave2.nxv16i32(<vscale x 16 x i32> %a)
+  %a.deinterleaved = tail call { <vscale x 8 x i32>, <vscale x 8 x i32> } @llvm.vector.deinterleave2.nxv16i32(<vscale x 16 x i32> %a)
   %a.real = extractvalue { <vscale x 8 x i32>, <vscale x 8 x i32> } %a.deinterleaved, 0
   %a.imag = extractvalue { <vscale x 8 x i32>, <vscale x 8 x i32> } %a.deinterleaved, 1
-  %b.deinterleaved = tail call { <vscale x 8 x i32>, <vscale x 8 x i32> } @llvm.experimental.vector.deinterleave2.nxv16i32(<vscale x 16 x i32> %b)
+  %b.deinterleaved = tail call { <vscale x 8 x i32>, <vscale x 8 x i32> } @llvm.vector.deinterleave2.nxv16i32(<vscale x 16 x i32> %b)
   %b.real = extractvalue { <vscale x 8 x i32>, <vscale x 8 x i32> } %b.deinterleaved, 0
   %b.imag = extractvalue { <vscale x 8 x i32>, <vscale x 8 x i32> } %b.deinterleaved, 1
   %0 = sub <vscale x 8 x i32> %b.real, %a.imag
   %1 = add <vscale x 8 x i32> %b.imag, %a.real
-  %interleaved.vec = tail call <vscale x 16 x i32> @llvm.experimental.vector.interleave2.nxv16i32(<vscale x 8 x i32> %0, <vscale x 8 x i32> %1)
+  %interleaved.vec = tail call <vscale x 16 x i32> @llvm.vector.interleave2.nxv16i32(<vscale x 8 x i32> %0, <vscale x 8 x i32> %1)
   ret <vscale x 16 x i32> %interleaved.vec
 }
 
-declare { <vscale x 2 x i32>, <vscale x 2 x i32> } @llvm.experimental.vector.deinterleave2.nxv4i32(<vscale x 4 x i32>)
-declare <vscale x 4 x i32> @llvm.experimental.vector.interleave2.nxv4i32(<vscale x 2 x i32>, <vscale x 2 x i32>)
+declare { <vscale x 2 x i32>, <vscale x 2 x i32> } @llvm.vector.deinterleave2.nxv4i32(<vscale x 4 x i32>)
+declare <vscale x 4 x i32> @llvm.vector.interleave2.nxv4i32(<vscale x 2 x i32>, <vscale x 2 x i32>)
 
-declare { <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.experimental.vector.deinterleave2.nxv8i32(<vscale x 8 x i32>)
-declare <vscale x 8 x i32> @llvm.experimental.vector.interleave2.nxv8i32(<vscale x 4 x i32>, <vscale x 4 x i32>)
+declare { <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.vector.deinterleave2.nxv8i32(<vscale x 8 x i32>)
+declare <vscale x 8 x i32> @llvm.vector.interleave2.nxv8i32(<vscale x 4 x i32>, <vscale x 4 x i32>)
 
-declare { <vscale x 8 x i32>, <vscale x 8 x i32> } @llvm.experimental.vector.deinterleave2.nxv16i32(<vscale x 16 x i32>)
-declare <vscale x 16 x i32> @llvm.experimental.vector.interleave2.nxv16i32(<vscale x 8 x i32>, <vscale x 8 x i32>)
+declare { <vscale x 8 x i32>, <vscale x 8 x i32> } @llvm.vector.deinterleave2.nxv16i32(<vscale x 16 x i32>)
+declare <vscale x 16 x i32> @llvm.vector.interleave2.nxv16i32(<vscale x 8 x i32>, <vscale x 8 x i32>)
diff --git a/llvm/test/CodeGen/AArch64/complex-deinterleaving-i32-mul-scalable.ll b/llvm/test/CodeGen/AArch64/complex-deinterleaving-i32-mul-scalable.ll
index d88eef9800d7..4cfe4707b9a9 100644
--- a/llvm/test/CodeGen/AArch64/complex-deinterleaving-i32-mul-scalable.ll
+++ b/llvm/test/CodeGen/AArch64/complex-deinterleaving-i32-mul-scalable.ll
@@ -13,10 +13,10 @@ define <vscale x 4 x i32> @complex_mul_v4i32(<vscale x 4 x i32> %a, <vscale x 4
 ; CHECK-NEXT:    mov z0.d, z2.d
 ; CHECK-NEXT:    ret
 entry:
-  %a.deinterleaved = tail call { <vscale x 2 x i32>, <vscale x 2 x i32> } @llvm.experimental.vector.deinterleave2.nxv4i32(<vscale x 4 x i32> %a)
+  %a.deinterleaved = tail call { <vscale x 2 x i32>, <vscale x 2 x i32> } @llvm.vector.deinterleave2.nxv4i32(<vscale x 4 x i32> %a)
   %a.real = extractvalue { <vscale x 2 x i32>, <vscale x 2 x i32> } %a.deinterleaved, 0
   %a.imag = extractvalue { <vscale x 2 x i32>, <vscale x 2 x i32> } %a.deinterleaved, 1
-  %b.deinterleaved = tail call { <vscale x 2 x i32>, <vscale x 2 x i32> } @llvm.experimental.vector.deinterleave2.nxv4i32(<vscale x 4 x i32> %b)
+  %b.deinterleaved = tail call { <vscale x 2 x i32>, <vscale x 2 x i32> } @llvm.vector.deinterleave2.nxv4i32(<vscale x 4 x i32> %b)
   %b.real = extractvalue { <vscale x 2 x i32>, <vscale x 2 x i32> } %b.deinterleaved, 0
   %b.imag = extractvalue { <vscale x 2 x i32>, <vscale x 2 x i32> } %b.deinterleaved, 1
   %0 = mul <vscale x 2 x i32> %b.imag, %a.real
@@ -25,7 +25,7 @@ entry:
   %3 = mul <vscale x 2 x i32> %b.real, %a.real
   %4 = mul <vscale x 2 x i32> %a.imag, %b.imag
   %5 = sub <vscale x 2 x i32> %3, %4
-  %interleaved.vec = tail call <vscale x 4 x i32> @llvm.experimental.vector.interleave2.nxv4i32(<vscale x 2 x i32> %5, <vscale x 2 x i32> %2)
+  %interleaved.vec = tail call <vscale x 4 x i32> @llvm.vector.interleave2.nxv4i32(<vscale x 2 x i32> %5, <vscale x 2 x i32> %2)
   ret <vscale x 4 x i32> %interleaved.vec
 }
 
@@ -43,10 +43,10 @@ define <vscale x 8 x i32> @complex_mul_v8i32(<vscale x 8 x i32> %a, <vscale x 8
 ; CHECK-NEXT:    mov z0.d, z5.d
 ; CHECK-NEXT:    ret
 entry:
-  %a.deinterleaved = tail call { <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.experimental.vector.deinterleave2.nxv8i32(<vscale x 8 x i32> %a)
+  %a.deinterleaved = tail call { <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.vector.deinterleave2.nxv8i32(<vscale x 8 x i32> %a)
   %a.real = extractvalue { <vscale x 4 x i32>, <vscale x 4 x i32> } %a.deinterleaved, 0
   %a.imag = extractvalue { <vscale x 4 x i32>, <vscale x 4 x i32> } %a.deinterleaved, 1
-  %b.deinterleaved = tail call { <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.experimental.vector.deinterleave2.nxv8i32(<vscale x 8 x i32> %b)
+  %b.deinterleaved = tail call { <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.vector.deinterleave2.nxv8i32(<vscale x 8 x i32> %b)
   %b.real = extractvalue { <vscale x 4 x i32>, <vscale x 4 x i32> } %b.deinterleaved, 0
   %b.imag = extractvalue { <vscale x 4 x i32>, <vscale x 4 x i32> } %b.deinterleaved, 1
   %0 = mul <vscale x 4 x i32> %b.imag, %a.real
@@ -55,7 +55,7 @@ entry:
   %3 = mul <vscale x 4 x i32> %b.real, %a.real
   %4 = mul <vscale x 4 x i32> %a.imag, %b.imag
   %5 = sub <vscale x 4 x i32> %3, %4
-  %interleaved.vec = tail call <vscale x 8 x i32> @llvm.experimental.vector.interleave2.nxv8i32(<vscale x 4 x i32> %5, <vscale x 4 x i32> %2)
+  %interleaved.vec = tail call <vscale x 8 x i32> @llvm.vector.interleave2.nxv8i32(<vscale x 4 x i32> %5, <vscale x 4 x i32> %2)
   ret <vscale x 8 x i32> %interleaved.vec
 }
 
@@ -81,10 +81,10 @@ define <vscale x 16 x i32> @complex_mul_v16i32(<vscale x 16 x i32> %a, <vscale x
 ; CHECK-NEXT:    mov z2.d, z27.d
 ; CHECK-NEXT:    ret
 entry:
-  %a.deinterleaved = tail call { <vscale x 8 x i32>, <vscale x 8 x i32> } @llvm.experimental.vector.deinterleave2.nxv16i32(<vscale x 16 x i32> %a)
+  %a.deinterleaved = tail call { <vscale x 8 x i32>, <vscale x 8 x i32> } @llvm.vector.deinterleave2.nxv16i32(<vscale x 16 x i32> %a)
   %a.real = extractvalue { <vscale x 8 x i32>, <vscale x 8 x i32> } %a.deinterleaved, 0
   %a.imag = extractvalue { <vscale x 8 x i32>, <vscale x 8 x i32> } %a.deinterleaved, 1
-  %b.deinterleaved = tail call { <vscale x 8 x i32>, <vscale x 8 x i32> } @llvm.experimental.vector.deinterleave2.nxv16i32(<vscale x 16 x i32> %b)
+  %b.deinterleaved = tail call { <vscale x 8 x i32>, <vscale x 8 x i32> } @llvm.vector.deinterleave2.nxv16i32(<vscale x 16 x i32> %b)
   %b.real = extractvalue { <vscale x 8 x i32>, <vscale x 8 x i32> } %b.deinterleaved, 0
   %b.imag = extractvalue { <vscale x 8 x i32>, <vscale x 8 x i32> } %b.deinterleaved, 1
   %0 = mul <vscale x 8 x i32> %b.imag, %a.real
@@ -93,16 +93,16 @@ entry:
   %3 = mul <vscale x 8 x i32> %b.real, %a.real
   %4 = mul <vscale x 8 x i32> %a.imag, %b.imag
   %5 = sub <vscale x 8 x i32> %3, %4
-  %interleaved.vec = tail call <vscale x 16 x i32> @llvm.experimental.vector.interleave2.nxv16i32(<vscale x 8 x i32> %5, <vscale x 8 x i32> %2)
+  %interleaved.vec = tail call <vscale x 16 x i32> @llvm.vector.interleave2.nxv16i32(<vscale x 8 x i32> %5, <vscale x 8 x i32> %2)
   ret <vscale x 16 x i32> %interleaved.vec
 }
 
-declare { <vscale x 2 x i32>, <vscale x 2 x i32> } @llvm.experimental.vector.deinterleave2.nxv4i32(<vscale x 4 x i32>)
-declare <vscale x 4 x i32> @llvm.experimental.vector.interleave2.nxv4i32(<vscale x 2 x i32>, <vscale x 2 x i32>)
+declare { <vscale x 2 x i32>, <vscale x 2 x i32> } @llvm.vector.deinterleave2.nxv4i32(<vscale x 4 x i32>)
+declare <vscale x 4 x i32> @llvm.vector.interleave2.nxv4i32(<vscale x 2 x i32>, <vscale x 2 x i32>)
 
-declare { <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.experimental.vector.deinterleave2.nxv8i32(<vscale x 8 x i32>)
-declare <vscale x 8 x i32> @llvm.experimental.vector.interleave2.nxv8i32(<vscale x 4 x i32>, <vscale x 4 x i32>)
+declare { <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.vector.deinterleave2.nxv8i32(<vscale x 8 x i32>)
+declare <vscale x 8 x i32> @llvm.vector.interleave2.nxv8i32(<vscale x 4 x i32>, <vscale x 4 x i32>)
 
-declare { <vscale x 8 x i32>, <vscale x 8 x i32> } @llvm.experimental.vector.deinterleave2.nxv16i32(<vscale x 16 x i32>)
-declare <vscale x 16 x i32> @llvm.experimental.vector.interleave2.nxv16i32(<vscale x 8 x i32>, <vscale x 8 x i32>)
+declare { <vscale x 8 x i32>, <vscale x 8 x i32> } @llvm.vector.deinterleave2.nxv16i32(<vscale x 16 x i32>)
+declare <vscale x 16 x i32> @llvm.vector.interleave2.nxv16i32(<vscale x 8 x i32>, <vscale x 8 x i32>)
 
diff --git a/llvm/test/CodeGen/AArch64/complex-deinterleaving-i64-add-scalable.ll b/llvm/test/CodeGen/AArch64/complex-deinterleaving-i64-add-scalable.ll
index 0b59be9414fa..f06b55c68b7e 100644
--- a/llvm/test/CodeGen/AArch64/complex-deinterleaving-i64-add-scalable.ll
+++ b/llvm/test/CodeGen/AArch64/complex-deinterleaving-i64-add-scalable.ll
@@ -11,15 +11,15 @@ define <vscale x 2 x i64> @complex_add_v2i64(<vscale x 2 x i64> %a, <vscale x 2
 ; CHECK-NEXT:    mov z0.d, z1.d
 ; CHECK-NEXT:    ret
 entry:
-  %a.deinterleaved = tail call { <vscale x 1 x i64>, <vscale x 1 x i64> } @llvm.experimental.vector.deinterleave2.nxv2i64(<vscale x 2 x i64> %a)
+  %a.deinterleaved = tail call { <vscale x 1 x i64>, <vscale x 1 x i64> } @llvm.vector.deinterleave2.nxv2i64(<vscale x 2 x i64> %a)
   %a.real = extractvalue { <vscale x 1 x i64>, <vscale x 1 x i64> } %a.deinterleaved, 0
   %a.imag = extractvalue { <vscale x 1 x i64>, <vscale x 1 x i64> } %a.deinterleaved, 1
-  %b.deinterleaved = tail call { <vscale x 1 x i64>, <vscale x 1 x i64> } @llvm.experimental.vector.deinterleave2.nxv2i64(<vscale x 2 x i64> %b)
+  %b.deinterleaved = tail call { <vscale x 1 x i64>, <vscale x 1 x i64> } @llvm.vector.deinterleave2.nxv2i64(<vscale x 2 x i64> %b)
   %b.real = extractvalue { <vscale x 1 x i64>, <vscale x 1 x i64> } %b.deinterleaved, 0
   %b.imag = extractvalue { <vscale x 1 x i64>, <vscale x 1 x i64> } %b.deinterleaved, 1
   %0 = sub <vscale x 1 x i64> %b.real, %a.imag
   %1 = add <vscale x 1 x i64> %b.imag, %a.real
-  %interleaved.vec = tail call <vscale x 2 x i64> @llvm.experimental.vector.interleave2.nxv2i64(<vscale x 1 x i64> %0, <vscale x 1 x i64> %1)
+  %interleaved.vec = tail call <vscale x 2 x i64> @llvm.vector.interleave2.nxv2i64(<vscale x 1 x i64> %0, <vscale x 1 x i64> %1)
   ret <vscale x 2 x i64> %interleaved.vec
 }
 
@@ -33,15 +33,15 @@ define <vscale x 4 x i64> @complex_add_v4i64(<vscale x 4 x i64> %a, <vscale x 4
 ; CHECK-NEXT:    mov z1.d, z3.d
 ; CHECK-NEXT:    ret
 entry:
-  %a.deinterleaved = tail call { <vscale x 2 x i64>, <vscale x 2 x i64> } @llvm.experimental.vector.deinterleave2.nxv4i64(<vscale x 4 x i64> %a)
+  %a.deinterleaved = tail call { <vscale x 2 x i64>, <vscale x 2 x i64> } @llvm.vector.deinterleave2.nxv4i64(<vscale x 4 x i64> %a)
   %a.real = extractvalue { <vscale x 2 x i64>, <vscale x 2 x i64> } %a.deinterleaved, 0
   %a.imag = extractvalue { <vscale x 2 x i64>, <vscale x 2 x i64> } %a.deinterleaved, 1
-  %b.deinterleaved = tail call { <vscale x 2 x i64>, <vscale x 2 x i64> } @llvm.experimental.vector.deinterleave2.nxv4i64(<vscale x 4 x i64> %b)
+  %b.deinterleaved = tail call { <vscale x 2 x i64>, <vscale x 2 x i64> } @llvm.vector.deinterleave2.nxv4i64(<vscale x 4 x i64> %b)
   %b.real = extractvalue { <vscale x 2 x i64>, <vscale x 2 x i64> } %b.deinterleaved, 0
   %b.imag = extractvalue { <vscale x 2 x i64>, <vscale x 2 x i64> } %b.deinterleaved, 1
   %0 = sub <vscale x 2 x i64> %b.real, %a.imag
   %1 = add <vscale x 2 x i64> %b.imag, %a.real
-  %interleaved.vec = tail call <vscale x 4 x i64> @llvm.experimental.vector.interleave2.nxv4i64(<vscale x 2 x i64> %0, <vscale x 2 x i64> %1)
+  %interleaved.vec = tail call <vscale x 4 x i64> @llvm.vector.interleave2.nxv4i64(<vscale x 2 x i64> %0, <vscale x 2 x i64> %1)
   ret <vscale x 4 x i64> %interleaved.vec
 }
 
@@ -59,23 +59,23 @@ define <vscale x 8 x i64> @complex_add_v8i64(<vscale x 8 x i64> %a, <vscale x 8
 ; CHECK-NEXT:    mov z3.d, z7.d
 ; CHECK-NEXT:    ret
 entry:
-  %a.deinterleaved = tail call { <vscale x 4 x i64>, <vscale x 4 x i64> } @llvm.experimental.vector.deinterleave2.nxv8i64(<vscale x 8 x i64> %a)
+  %a.deinterleaved = tail call { <vscale x 4 x i64>, <vscale x 4 x i64> } @llvm.vector.deinterleave2.nxv8i64(<vscale x 8 x i64> %a)
   %a.real = extractvalue { <vscale x 4 x i64>, <vscale x 4 x i64> } %a.deinterleaved, 0
   %a.imag = extractvalue { <vscale x 4 x i64>, <vscale x 4 x i64> } %a.deinterleaved, 1
-  %b.deinterleaved = tail call { <vscale x 4 x i64>, <vscale x 4 x i64> } @llvm.experimental.vector.deinterleave2.nxv8i64(<vscale x 8 x i64> %b)
+  %b.deinterleaved = tail call { <vscale x 4 x i64>, <vscale x 4 x i64> } @llvm.vector.deinterleave2.nxv8i64(<vscale x 8 x i64> %b)
   %b.real = extractvalue { <vscale x 4 x i64>, <vscale x 4 x i64> } %b.deinterleaved, 0
   %b.imag = extractvalue { <vscale x 4 x i64>, <vscale x 4 x i64> } %b.deinterleaved, 1
   %0 = sub <vscale x 4 x i64> %b.real, %a.imag
   %1 = add <vscale x 4 x i64> %b.imag, %a.real
-  %interleaved.vec = tail call <vscale x 8 x i64> @llvm.experimental.vector.interleave2.nxv8i64(<vscale x 4 x i64> %0, <vscale x 4 x i64> %1)
+  %interleaved.vec = tail call <vscale x 8 x i64> @llvm.vector.interleave2.nxv8i64(<vscale x 4 x i64> %0, <vscale x 4 x i64> %1)
   ret <vscale x 8 x i64> %interleaved.vec
 }
 
-declare { <vscale x 1 x i64>, <vscale x 1 x i64> } @llvm.experimental.vector.deinterleave2.nxv2i64(<vscale x 2 x i64>)
-declare <vscale x 2 x i64> @llvm.experimental.vector.interleave2.nxv2i64(<vscale x 1 x i64>, <vscale x 1 x i64>)
+declare { <vscale x 1 x i64>, <vscale x 1 x i64> } @llvm.vector.deinterleave2.nxv2i64(<vscale x 2 x i64>)
+declare <vscale x 2 x i64> @llvm.vector.interleave2.nxv2i64(<vscale x 1 x i64>, <vscale x 1 x i64>)
 
-declare { <vscale x 2 x i64>, <vscale x 2 x i64> } @llvm.experimental.vector.deinterleave2.nxv4i64(<vscale x 4 x i64>)
-declare <vscale x 4 x i64> @llvm.experimental.vector.interleave2.nxv4i64(<vscale x 2 x i64>, <vscale x 2 x i64>)
+declare { <vscale x 2 x i64>, <vscale x 2 x i64> } @llvm.vector.deinterleave2.nxv4i64(<vscale x 4 x i64>)
+declare <vscale x 4 x i64> @llvm.vector.interleave2.nxv4i64(<vscale x 2 x i64>, <vscale x 2 x i64>)
 
-declare { <vscale x 4 x i64>, <vscale x 4 x i64> } @llvm.experimental.vector.deinterleave2.nxv8i64(<vscale x 8 x i64>)
-declare <vscale x 8 x i64> @llvm.experimental.vector.interleave2.nxv8i64(<vscale x 4 x i64>, <vscale x 4 x i64>)
+declare { <vscale x 4 x i64>, <vscale x 4 x i64> } @llvm.vector.deinterleave2.nxv8i64(<vscale x 8 x i64>)
+declare <vscale x 8 x i64> @llvm.vector.interleave2.nxv8i64(<vscale x 4 x i64>, <vscale x 4 x i64>)
diff --git a/llvm/test/CodeGen/AArch64/complex-deinterleaving-i64-mul-scalable.ll b/llvm/test/CodeGen/AArch64/complex-deinterleaving-i64-mul-scalable.ll
index 16e1f3e63dce..5975f3b491d4 100644
--- a/llvm/test/CodeGen/AArch64/complex-deinterleaving-i64-mul-scalable.ll
+++ b/llvm/test/CodeGen/AArch64/complex-deinterleaving-i64-mul-scalable.ll
@@ -13,10 +13,10 @@ define <vscale x 2 x i64> @complex_mul_v2i64(<vscale x 2 x i64> %a, <vscale x 2
 ; CHECK-NEXT:    mov z0.d, z2.d
 ; CHECK-NEXT:    ret
 entry:
-  %a.deinterleaved = tail call { <vscale x 1 x i64>, <vscale x 1 x i64> } @llvm.experimental.vector.deinterleave2.nxv2i64(<vscale x 2 x i64> %a)
+  %a.deinterleaved = tail call { <vscale x 1 x i64>, <vscale x 1 x i64> } @llvm.vector.deinterleave2.nxv2i64(<vscale x 2 x i64> %a)
   %a.real = extractvalue { <vscale x 1 x i64>, <vscale x 1 x i64> } %a.deinterleaved, 0
   %a.imag = extractvalue { <vscale x 1 x i64>, <vscale x 1 x i64> } %a.deinterleaved, 1
-  %b.deinterleaved = tail call { <vscale x 1 x i64>, <vscale x 1 x i64> } @llvm.experimental.vector.deinterleave2.nxv2i64(<vscale x 2 x i64> %b)
+  %b.deinterleaved = tail call { <vscale x 1 x i64>, <vscale x 1 x i64> } @llvm.vector.deinterleave2.nxv2i64(<vscale x 2 x i64> %b)
   %b.real = extractvalue { <vscale x 1 x i64>, <vscale x 1 x i64> } %b.deinterleaved, 0
   %b.imag = extractvalue { <vscale x 1 x i64>, <vscale x 1 x i64> } %b.deinterleaved, 1
   %0 = mul <vscale x 1 x i64> %b.imag, %a.real
@@ -25,7 +25,7 @@ entry:
   %3 = mul <vscale x 1 x i64> %b.real, %a.real
   %4 = mul <vscale x 1 x i64> %a.imag, %b.imag
   %5 = sub <vscale x 1 x i64> %3, %4
-  %interleaved.vec = tail call <vscale x 2 x i64> @llvm.experimental.vector.interleave2.nxv2i64(<vscale x 1 x i64> %5, <vscale x 1 x i64> %2)
+  %interleaved.vec = tail call <vscale x 2 x i64> @llvm.vector.interleave2.nxv2i64(<vscale x 1 x i64> %5, <vscale x 1 x i64> %2)
   ret <vscale x 2 x i64> %interleaved.vec
 }
 
@@ -43,10 +43,10 @@ define <vscale x 4 x i64> @complex_mul_v4i64(<vscale x 4 x i64> %a, <vscale x 4
 ; CHECK-NEXT:    mov z0.d, z5.d
 ; CHECK-NEXT:    ret
 entry:
-  %a.deinterleaved = tail call { <vscale x 2 x i64>, <vscale x 2 x i64> } @llvm.experimental.vector.deinterleave2.nxv4i64(<vscale x 4 x i64> %a)
+  %a.deinterleaved = tail call { <vscale x 2 x i64>, <vscale x 2 x i64> } @llvm.vector.deinterleave2.nxv4i64(<vscale x 4 x i64> %a)
   %a.real = extractvalue { <vscale x 2 x i64>, <vscale x 2 x i64> } %a.deinterleaved, 0
   %a.imag = extractvalue { <vscale x 2 x i64>, <vscale x 2 x i64> } %a.deinterleaved, 1
-  %b.deinterleaved = tail call { <vscale x 2 x i64>, <vscale x 2 x i64> } @llvm.experimental.vector.deinterleave2.nxv4i64(<vscale x 4 x i64> %b)
+  %b.deinterleaved = tail call { <vscale x 2 x i64>, <vscale x 2 x i64> } @llvm.vector.deinterleave2.nxv4i64(<vscale x 4 x i64> %b)
   %b.real = extractvalue { <vscale x 2 x i64>, <vscale x 2 x i64> } %b.deinterleaved, 0
   %b.imag = extractvalue { <vscale x 2 x i64>, <vscale x 2 x i64> } %b.deinterleaved, 1
   %0 = mul <vscale x 2 x i64> %b.imag, %a.real
@@ -55,7 +55,7 @@ entry:
   %3 = mul <vscale x 2 x i64> %b.real, %a.real
   %4 = mul <vscale x 2 x i64> %a.imag, %b.imag
   %5 = sub <vscale x 2 x i64> %3, %4
-  %interleaved.vec = tail call <vscale x 4 x i64> @llvm.experimental.vector.interleave2.nxv4i64(<vscale x 2 x i64> %5, <vscale x 2 x i64> %2)
+  %interleaved.vec = tail call <vscale x 4 x i64> @llvm.vector.interleave2.nxv4i64(<vscale x 2 x i64> %5, <vscale x 2 x i64> %2)
   ret <vscale x 4 x i64> %interleaved.vec
 }
 
@@ -81,10 +81,10 @@ define <vscale x 8 x i64> @complex_mul_v8i64(<vscale x 8 x i64> %a, <vscale x 8
 ; CHECK-NEXT:    mov z2.d, z27.d
 ; CHECK-NEXT:    ret
 entry:
-  %a.deinterleaved = tail call { <vscale x 4 x i64>, <vscale x 4 x i64> } @llvm.experimental.vector.deinterleave2.nxv8i64(<vscale x 8 x i64> %a)
+  %a.deinterleaved = tail call { <vscale x 4 x i64>, <vscale x 4 x i64> } @llvm.vector.deinterleave2.nxv8i64(<vscale x 8 x i64> %a)
   %a.real = extractvalue { <vscale x 4 x i64>, <vscale x 4 x i64> } %a.deinterleaved, 0
   %a.imag = extractvalue { <vscale x 4 x i64>, <vscale x 4 x i64> } %a.deinterleaved, 1
-  %b.deinterleaved = tail call { <vscale x 4 x i64>, <vscale x 4 x i64> } @llvm.experimental.vector.deinterleave2.nxv8i64(<vscale x 8 x i64> %b)
+  %b.deinterleaved = tail call { <vscale x 4 x i64>, <vscale x 4 x i64> } @llvm.vector.deinterleave2.nxv8i64(<vscale x 8 x i64> %b)
   %b.real = extractvalue { <vscale x 4 x i64>, <vscale x 4 x i64> } %b.deinterleaved, 0
   %b.imag = extractvalue { <vscale x 4 x i64>, <vscale x 4 x i64> } %b.deinterleaved, 1
   %0 = mul <vscale x 4 x i64> %b.imag, %a.real
@@ -93,7 +93,7 @@ entry:
   %3 = mul <vscale x 4 x i64> %b.real, %a.real
   %4 = mul <vscale x 4 x i64> %a.imag, %b.imag
   %5 = sub <vscale x 4 x i64> %3, %4
-  %interleaved.vec = tail call <vscale x 8 x i64> @llvm.experimental.vector.interleave2.nxv8i64(<vscale x 4 x i64> %5, <vscale x 4 x i64> %2)
+  %interleaved.vec = tail call <vscale x 8 x i64> @llvm.vector.interleave2.nxv8i64(<vscale x 4 x i64> %5, <vscale x 4 x i64> %2)
   ret <vscale x 8 x i64> %interleaved.vec
 }
 
@@ -119,11 +119,11 @@ define <vscale x 8 x i64> @complex_minus_mul_v8i64(<vscale x 8 x i64> %a, <vscal
 ; CHECK-NEXT:    mov z2.d, z27.d
 ; CHECK-NEXT:    ret
 entry:
-  %a.deinterleaved = tail call { <vscale x 4 x i64>, <vscale x 4 x i64> } @llvm.experimental.vector.deinterleave2.nxv8i64(<vscale x 8 x i64> %a)
+  %a.deinterleaved = tail call { <vscale x 4 x i64>, <vscale x 4 x i64> } @llvm.vector.deinterleave2.nxv8i64(<vscale x 8 x i64> %a)
   %a.real = extractvalue { <vscale x 4 x i64>, <vscale x 4 x i64> } %a.deinterleaved, 0
   %a.imag = extractvalue { <vscale x 4 x i64>, <vscale x 4 x i64> } %a.deinterleaved, 1
   %0 = sub <vscale x 4 x i64> zeroinitializer, %a.real
-  %b.deinterleaved = tail call { <vscale x 4 x i64>, <vscale x 4 x i64> } @llvm.experimental.vector.deinterleave2.nxv8i64(<vscale x 8 x i64> %b)
+  %b.deinterleaved = tail call { <vscale x 4 x i64>, <vscale x 4 x i64> } @llvm.vector.deinterleave2.nxv8i64(<vscale x 8 x i64> %b)
   %b.real = extractvalue { <vscale x 4 x i64>, <vscale x 4 x i64> } %b.deinterleaved, 0
   %b.imag = extractvalue { <vscale x 4 x i64>, <vscale x 4 x i64> } %b.deinterleaved, 1
   %1 = mul <vscale x 4 x i64> %b.real, %0
@@ -132,15 +132,15 @@ entry:
   %4 = mul <vscale x 4 x i64> %b.real, %a.imag
   %5 = mul <vscale x 4 x i64> %b.imag, %0
   %6 = sub <vscale x 4 x i64> %5, %4
-  %interleaved.vec = tail call <vscale x 8 x i64> @llvm.experimental.vector.interleave2.nxv8i64(<vscale x 4 x i64> %3, <vscale x 4 x i64> %6)
+  %interleaved.vec = tail call <vscale x 8 x i64> @llvm.vector.interleave2.nxv8i64(<vscale x 4 x i64> %3, <vscale x 4 x i64> %6)
   ret <vscale x 8 x i64> %interleaved.vec
 }
 
-declare { <vscale x 1 x i64>, <vscale x 1 x i64> } @llvm.experimental.vector.deinterleave2.nxv2i64(<vscale x 2 x i64>)
-declare <vscale x 2 x i64> @llvm.experimental.vector.interleave2.nxv2i64(<vscale x 1 x i64>, <vscale x 1 x i64>)
+declare { <vscale x 1 x i64>, <vscale x 1 x i64> } @llvm.vector.deinterleave2.nxv2i64(<vscale x 2 x i64>)
+declare <vscale x 2 x i64> @llvm.vector.interleave2.nxv2i64(<vscale x 1 x i64>, <vscale x 1 x i64>)
 
-declare { <vscale x 2 x i64>, <vscale x 2 x i64> } @llvm.experimental.vector.deinterleave2.nxv4i64(<vscale x 4 x i64>)
-declare <vscale x 4 x i64> @llvm.experimental.vector.interleave2.nxv4i64(<vscale x 2 x i64>, <vscale x 2 x i64>)
+declare { <vscale x 2 x i64>, <vscale x 2 x i64> } @llvm.vector.deinterleave2.nxv4i64(<vscale x 4 x i64>)
+declare <vscale x 4 x i64> @llvm.vector.interleave2.nxv4i64(<vscale x 2 x i64>, <vscale x 2 x i64>)
 
-declare { <vscale x 4 x i64>, <vscale x 4 x i64> } @llvm.experimental.vector.deinterleave2.nxv8i64(<vscale x 8 x i64>)
-declare <vscale x 8 x i64> @llvm.experimental.vector.interleave2.nxv8i64(<vscale x 4 x i64>, <vscale x 4 x i64>)
+declare { <vscale x 4 x i64>, <vscale x 4 x i64> } @llvm.vector.deinterleave2.nxv8i64(<vscale x 8 x i64>)
+declare <vscale x 8 x i64> @llvm.vector.interleave2.nxv8i64(<vscale x 4 x i64>, <vscale x 4 x i64>)
diff --git a/llvm/test/CodeGen/AArch64/complex-deinterleaving-i8-add-scalable.ll b/llvm/test/CodeGen/AArch64/complex-deinterleaving-i8-add-scalable.ll
index b631486137e6..81872c1723f2 100644
--- a/llvm/test/CodeGen/AArch64/complex-deinterleaving-i8-add-scalable.ll
+++ b/llvm/test/CodeGen/AArch64/complex-deinterleaving-i8-add-scalable.ll
@@ -22,15 +22,15 @@ define <vscale x 8 x i8> @complex_add_v8i8(<vscale x 8 x i8> %a, <vscale x 8 x i
 ; CHECK-NEXT:    uzp1 z0.h, z0.h, z2.h
 ; CHECK-NEXT:    ret
 entry:
-  %a.deinterleaved = tail call { <vscale x 4 x i8>, <vscale x 4 x i8> } @llvm.experimental.vector.deinterleave2.nxv8i8(<vscale x 8 x i8> %a)
+  %a.deinterleaved = tail call { <vscale x 4 x i8>, <vscale x 4 x i8> } @llvm.vector.deinterleave2.nxv8i8(<vscale x 8 x i8> %a)
   %a.real = extractvalue { <vscale x 4 x i8>, <vscale x 4 x i8> } %a.deinterleaved, 0
   %a.imag = extractvalue { <vscale x 4 x i8>, <vscale x 4 x i8> } %a.deinterleaved, 1
-  %b.deinterleaved = tail call { <vscale x 4 x i8>, <vscale x 4 x i8> } @llvm.experimental.vector.deinterleave2.nxv8i8(<vscale x 8 x i8> %b)
+  %b.deinterleaved = tail call { <vscale x 4 x i8>, <vscale x 4 x i8> } @llvm.vector.deinterleave2.nxv8i8(<vscale x 8 x i8> %b)
   %b.real = extractvalue { <vscale x 4 x i8>, <vscale x 4 x i8> } %b.deinterleaved, 0
   %b.imag = extractvalue { <vscale x 4 x i8>, <vscale x 4 x i8> } %b.deinterleaved, 1
   %0 = sub <vscale x 4 x i8> %b.real, %a.imag
   %1 = add <vscale x 4 x i8> %b.imag, %a.real
-  %interleaved.vec = tail call <vscale x 8 x i8> @llvm.experimental.vector.interleave2.nxv8i8(<vscale x 4 x i8> %0, <vscale x 4 x i8> %1)
+  %interleaved.vec = tail call <vscale x 8 x i8> @llvm.vector.interleave2.nxv8i8(<vscale x 4 x i8> %0, <vscale x 4 x i8> %1)
   ret <vscale x 8 x i8> %interleaved.vec
 }
 
@@ -42,15 +42,15 @@ define <vscale x 16 x i8> @complex_add_v16i8(<vscale x 16 x i8> %a, <vscale x 16
 ; CHECK-NEXT:    mov z0.d, z1.d
 ; CHECK-NEXT:    ret
 entry:
-  %a.deinterleaved = tail call { <vscale x 8 x i8>, <vscale x 8 x i8> } @llvm.experimental.vector.deinterleave2.nxv16i8(<vscale x 16 x i8> %a)
+  %a.deinterleaved = tail call { <vscale x 8 x i8>, <vscale x 8 x i8> } @llvm.vector.deinterleave2.nxv16i8(<vscale x 16 x i8> %a)
   %a.real = extractvalue { <vscale x 8 x i8>, <vscale x 8 x i8> } %a.deinterleaved, 0
   %a.imag = extractvalue { <vscale x 8 x i8>, <vscale x 8 x i8> } %a.deinterleaved, 1
-  %b.deinterleaved = tail call { <vscale x 8 x i8>, <vscale x 8 x i8> } @llvm.experimental.vector.deinterleave2.nxv16i8(<vscale x 16 x i8> %b)
+  %b.deinterleaved = tail call { <vscale x 8 x i8>, <vscale x 8 x i8> } @llvm.vector.deinterleave2.nxv16i8(<vscale x 16 x i8> %b)
   %b.real = extractvalue { <vscale x 8 x i8>, <vscale x 8 x i8> } %b.deinterleaved, 0
   %b.imag = extractvalue { <vscale x 8 x i8>, <vscale x 8 x i8> } %b.deinterleaved, 1
   %0 = sub <vscale x 8 x i8> %b.real, %a.imag
   %1 = add <vscale x 8 x i8> %b.imag, %a.real
-  %interleaved.vec = tail call <vscale x 16 x i8> @llvm.experimental.vector.interleave2.nxv16i8(<vscale x 8 x i8> %0, <vscale x 8 x i8> %1)
+  %interleaved.vec = tail call <vscale x 16 x i8> @llvm.vector.interleave2.nxv16i8(<vscale x 8 x i8> %0, <vscale x 8 x i8> %1)
   ret <vscale x 16 x i8> %interleaved.vec
 }
 
@@ -64,23 +64,23 @@ define <vscale x 32 x i8> @complex_add_v32i8(<vscale x 32 x i8> %a, <vscale x 32
 ; CHECK-NEXT:    mov z1.d, z3.d
 ; CHECK-NEXT:    ret
 entry:
-  %a.deinterleaved = tail call { <vscale x 16 x i8>, <vscale x 16 x i8> } @llvm.experimental.vector.deinterleave2.nxv32i8(<vscale x 32 x i8> %a)
+  %a.deinterleaved = tail call { <vscale x 16 x i8>, <vscale x 16 x i8> } @llvm.vector.deinterleave2.nxv32i8(<vscale x 32 x i8> %a)
   %a.real = extractvalue { <vscale x 16 x i8>, <vscale x 16 x i8> } %a.deinterleaved, 0
   %a.imag = extractvalue { <vscale x 16 x i8>, <vscale x 16 x i8> } %a.deinterleaved, 1
-  %b.deinterleaved = tail call { <vscale x 16 x i8>, <vscale x 16 x i8> } @llvm.experimental.vector.deinterleave2.nxv32i8(<vscale x 32 x i8> %b)
+  %b.deinterleaved = tail call { <vscale x 16 x i8>, <vscale x 16 x i8> } @llvm.vector.deinterleave2.nxv32i8(<vscale x 32 x i8> %b)
   %b.real = extractvalue { <vscale x 16 x i8>, <vscale x 16 x i8> } %b.deinterleaved, 0
   %b.imag = extractvalue { <vscale x 16 x i8>, <vscale x 16 x i8> } %b.deinterleaved, 1
   %0 = sub <vscale x 16 x i8> %b.real, %a.imag
   %1 = add <vscale x 16 x i8> %b.imag, %a.real
-  %interleaved.vec = tail call <vscale x 32 x i8> @llvm.experimental.vector.interleave2.nxv32i8(<vscale x 16 x i8> %0, <vscale x 16 x i8> %1)
+  %interleaved.vec = tail call <vscale x 32 x i8> @llvm.vector.interleave2.nxv32i8(<vscale x 16 x i8> %0, <vscale x 16 x i8> %1)
   ret <vscale x 32 x i8> %interleaved.vec
 }
 
-declare { <vscale x 4 x i8>, <vscale x 4 x i8> } @llvm.experimental.vector.deinterleave2.nxv8i8(<vscale x 8 x i8>)
-declare <vscale x 8 x i8> @llvm.experimental.vector.interleave2.nxv8i8(<vscale x 4 x i8>, <vscale x 4 x i8>)
+declare { <vscale x 4 x i8>, <vscale x 4 x i8> } @llvm.vector.deinterleave2.nxv8i8(<vscale x 8 x i8>)
+declare <vscale x 8 x i8> @llvm.vector.interleave2.nxv8i8(<vscale x 4 x i8>, <vscale x 4 x i8>)
 
-declare { <vscale x 8 x i8>, <vscale x 8 x i8> } @llvm.experimental.vector.deinterleave2.nxv16i8(<vscale x 16 x i8>)
-declare <vscale x 16 x i8> @llvm.experimental.vector.interleave2.nxv16i8(<vscale x 8 x i8>, <vscale x 8 x i8>)
+declare { <vscale x 8 x i8>, <vscale x 8 x i8> } @llvm.vector.deinterleave2.nxv16i8(<vscale x 16 x i8>)
+declare <vscale x 16 x i8> @llvm.vector.interleave2.nxv16i8(<vscale x 8 x i8>, <vscale x 8 x i8>)
 
-declare { <vscale x 16 x i8>, <vscale x 16 x i8> } @llvm.experimental.vector.deinterleave2.nxv32i8(<vscale x 32 x i8>)
-declare <vscale x 32 x i8> @llvm.experimental.vector.interleave2.nxv32i8(<vscale x 16 x i8>, <vscale x 16 x i8>)
+declare { <vscale x 16 x i8>, <vscale x 16 x i8> } @llvm.vector.deinterleave2.nxv32i8(<vscale x 32 x i8>)
+declare <vscale x 32 x i8> @llvm.vector.interleave2.nxv32i8(<vscale x 16 x i8>, <vscale x 16 x i8>)
diff --git a/llvm/test/CodeGen/AArch64/complex-deinterleaving-reductions-predicated-scalable.ll b/llvm/test/CodeGen/AArch64/complex-deinterleaving-reductions-predicated-scalable.ll
index 19318fdeeca7..ac2b21af29ab 100644
--- a/llvm/test/CodeGen/AArch64/complex-deinterleaving-reductions-predicated-scalable.ll
+++ b/llvm/test/CodeGen/AArch64/complex-deinterleaving-reductions-predicated-scalable.ll
@@ -69,14 +69,14 @@ vector.body:                                      ; preds = %vector.body, %entry
   %vec.phi27 = phi <vscale x 2 x double> [ zeroinitializer, %entry ], [ %16, %vector.body ]
   %scevgep = getelementptr i8, ptr %a, i64 %lsr.iv
   %scevgep34 = getelementptr i8, ptr %b, i64 %lsr.iv
-  %interleaved.mask = tail call <vscale x 4 x i1> @llvm.experimental.vector.interleave2.nxv4i1(<vscale x 2 x i1> %active.lane.mask, <vscale x 2 x i1> %active.lane.mask)
+  %interleaved.mask = tail call <vscale x 4 x i1> @llvm.vector.interleave2.nxv4i1(<vscale x 2 x i1> %active.lane.mask, <vscale x 2 x i1> %active.lane.mask)
   %wide.masked.vec = tail call <vscale x 4 x double> @llvm.masked.load.nxv4f64.p0(ptr %scevgep, i32 8, <vscale x 4 x i1> %interleaved.mask, <vscale x 4 x double> poison)
-  %strided.vec = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.experimental.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %wide.masked.vec)
+  %strided.vec = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %wide.masked.vec)
   %3 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %strided.vec, 0
   %4 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %strided.vec, 1
-  %interleaved.mask28 = tail call <vscale x 4 x i1> @llvm.experimental.vector.interleave2.nxv4i1(<vscale x 2 x i1> %active.lane.mask, <vscale x 2 x i1> %active.lane.mask)
+  %interleaved.mask28 = tail call <vscale x 4 x i1> @llvm.vector.interleave2.nxv4i1(<vscale x 2 x i1> %active.lane.mask, <vscale x 2 x i1> %active.lane.mask)
   %wide.masked.vec29 = tail call <vscale x 4 x double> @llvm.masked.load.nxv4f64.p0(ptr %scevgep34, i32 8, <vscale x 4 x i1> %interleaved.mask28, <vscale x 4 x double> poison)
-  %strided.vec30 = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.experimental.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %wide.masked.vec29)
+  %strided.vec30 = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %wide.masked.vec29)
   %5 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %strided.vec30, 0
   %6 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %strided.vec30, 1
   %7 = fmul fast <vscale x 2 x double> %6, %3
@@ -175,13 +175,13 @@ vector.body:                                      ; preds = %vector.body, %entry
   %4 = icmp ne <vscale x 2 x i32> %wide.load, zeroinitializer
   %scevgep49 = getelementptr i8, ptr %a, i64 %lsr.iv48
   %scevgep50 = getelementptr i8, ptr %b, i64 %lsr.iv48
-  %interleaved.mask = tail call <vscale x 4 x i1> @llvm.experimental.vector.interleave2.nxv4i1(<vscale x 2 x i1> %4, <vscale x 2 x i1> %4)
+  %interleaved.mask = tail call <vscale x 4 x i1> @llvm.vector.interleave2.nxv4i1(<vscale x 2 x i1> %4, <vscale x 2 x i1> %4)
   %wide.masked.vec = tail call <vscale x 4 x double> @llvm.masked.load.nxv4f64.p0(ptr %scevgep49, i32 8, <vscale x 4 x i1> %interleaved.mask, <vscale x 4 x double> poison)
-  %strided.vec = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.experimental.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %wide.masked.vec)
+  %strided.vec = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %wide.masked.vec)
   %5 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %strided.vec, 0
   %6 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %strided.vec, 1
   %wide.masked.vec32 = tail call <vscale x 4 x double> @llvm.masked.load.nxv4f64.p0(ptr %scevgep50, i32 8, <vscale x 4 x i1> %interleaved.mask, <vscale x 4 x double> poison)
-  %strided.vec33 = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.experimental.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %wide.masked.vec32)
+  %strided.vec33 = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %wide.masked.vec32)
   %7 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %strided.vec33, 0
   %8 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %strided.vec33, 1
   %9 = fmul fast <vscale x 2 x double> %8, %5
@@ -279,14 +279,14 @@ vector.body:                                      ; preds = %vector.body, %entry
   %scevgep38 = getelementptr i8, ptr %a, i64 %lsr.iv
   %scevgep39 = getelementptr i8, ptr %b, i64 %lsr.iv
   %5 = select <vscale x 2 x i1> %active.lane.mask, <vscale x 2 x i1> %4, <vscale x 2 x i1> zeroinitializer
-  %interleaved.mask = tail call <vscale x 4 x i1> @llvm.experimental.vector.interleave2.nxv4i1(<vscale x 2 x i1> %5, <vscale x 2 x i1> %5)
+  %interleaved.mask = tail call <vscale x 4 x i1> @llvm.vector.interleave2.nxv4i1(<vscale x 2 x i1> %5, <vscale x 2 x i1> %5)
   %wide.masked.vec = tail call <vscale x 4 x double> @llvm.masked.load.nxv4f64.p0(ptr %scevgep38, i32 8, <vscale x 4 x i1> %interleaved.mask, <vscale x 4 x double> poison)
-  %strided.vec = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.experimental.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %wide.masked.vec)
+  %strided.vec = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %wide.masked.vec)
   %6 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %strided.vec, 0
   %7 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %strided.vec, 1
-  %interleaved.mask31 = tail call <vscale x 4 x i1> @llvm.experimental.vector.interleave2.nxv4i1(<vscale x 2 x i1> %5, <vscale x 2 x i1> %5)
+  %interleaved.mask31 = tail call <vscale x 4 x i1> @llvm.vector.interleave2.nxv4i1(<vscale x 2 x i1> %5, <vscale x 2 x i1> %5)
   %wide.masked.vec32 = tail call <vscale x 4 x double> @llvm.masked.load.nxv4f64.p0(ptr %scevgep39, i32 8, <vscale x 4 x i1> %interleaved.mask31, <vscale x 4 x double> poison)
-  %strided.vec33 = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.experimental.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %wide.masked.vec32)
+  %strided.vec33 = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %wide.masked.vec32)
   %8 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %strided.vec33, 0
   %9 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %strided.vec33, 1
   %10 = fmul fast <vscale x 2 x double> %9, %6
@@ -320,6 +320,6 @@ declare i64 @llvm.vscale.i64()
 declare <vscale x 2 x i1> @llvm.get.active.lane.mask.nxv2i1.i64(i64, i64)
 declare <vscale x 2 x i32> @llvm.masked.load.nxv2i32.p0(ptr nocapture, i32 immarg, <vscale x 2 x i1>, <vscale x 2 x i32>)
 declare <vscale x 4 x double> @llvm.masked.load.nxv4f64.p0(ptr nocapture, i32 immarg, <vscale x 4 x i1>, <vscale x 4 x double>)
-declare <vscale x 4 x i1> @llvm.experimental.vector.interleave2.nxv4i1(<vscale x 2 x i1>, <vscale x 2 x i1>)
-declare { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.experimental.vector.deinterleave2.nxv4f64(<vscale x 4 x double>)
+declare <vscale x 4 x i1> @llvm.vector.interleave2.nxv4i1(<vscale x 2 x i1>, <vscale x 2 x i1>)
+declare { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.vector.deinterleave2.nxv4f64(<vscale x 4 x double>)
 declare double @llvm.vector.reduce.fadd.nxv2f64(double, <vscale x 2 x double>)
diff --git a/llvm/test/CodeGen/AArch64/complex-deinterleaving-reductions-scalable.ll b/llvm/test/CodeGen/AArch64/complex-deinterleaving-reductions-scalable.ll
index 664d99a3627b..af07519ad53d 100644
--- a/llvm/test/CodeGen/AArch64/complex-deinterleaving-reductions-scalable.ll
+++ b/llvm/test/CodeGen/AArch64/complex-deinterleaving-reductions-scalable.ll
@@ -17,11 +17,11 @@ define %"class.std::complex" @complex_mul_v2f64(ptr %a, ptr %b) {
 ; CHECK-NEXT:    mov z1.d, #0 // =0x0
 ; CHECK-NEXT:    cntd x9
 ; CHECK-NEXT:    ptrue p1.b
-; CHECK-NEXT:    neg x9, x9
-; CHECK-NEXT:    mov w10, #100 // =0x64
+; CHECK-NEXT:    neg x10, x9
+; CHECK-NEXT:    mov w11, #100 // =0x64
 ; CHECK-NEXT:    ptrue p0.d
 ; CHECK-NEXT:    mov x8, xzr
-; CHECK-NEXT:    and x10, x9, x10
+; CHECK-NEXT:    and x10, x10, x11
 ; CHECK-NEXT:    rdvl x11, #2
 ; CHECK-NEXT:    zip2 z0.d, z1.d, z1.d
 ; CHECK-NEXT:    zip1 z1.d, z1.d, z1.d
@@ -33,7 +33,7 @@ define %"class.std::complex" @complex_mul_v2f64(ptr %a, ptr %b) {
 ; CHECK-NEXT:    ld1d { z3.d }, p0/z, [x12, #1, mul vl]
 ; CHECK-NEXT:    ld1b { z4.b }, p1/z, [x1, x8]
 ; CHECK-NEXT:    ld1d { z5.d }, p0/z, [x13, #1, mul vl]
-; CHECK-NEXT:    adds x10, x10, x9
+; CHECK-NEXT:    subs x10, x10, x9
 ; CHECK-NEXT:    add x8, x8, x11
 ; CHECK-NEXT:    fcmla z1.d, p0/m, z4.d, z2.d, #0
 ; CHECK-NEXT:    fcmla z0.d, p0/m, z5.d, z3.d, #0
@@ -64,11 +64,11 @@ vector.body:                                      ; preds = %vector.body, %entry
   %scevgep46 = getelementptr i8, ptr %a, i64 %lsr.iv27
   %scevgep47 = getelementptr i8, ptr %b, i64 %lsr.iv27
   %wide.vec = load <vscale x 4 x double>, ptr %scevgep46, align 8
-  %3 = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.experimental.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %wide.vec)
+  %3 = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %wide.vec)
   %4 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %3, 0
   %5 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %3, 1
   %wide.vec30 = load <vscale x 4 x double>, ptr %scevgep47, align 8
-  %6 = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.experimental.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %wide.vec30)
+  %6 = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %wide.vec30)
   %7 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %6, 0
   %8 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %6, 1
   %9 = fmul fast <vscale x 2 x double> %8, %4
@@ -106,12 +106,12 @@ define %"class.std::complex" @complex_mul_nonzero_init_v2f64(ptr %a, ptr %b) {
 ; CHECK-NEXT:    cntd x9
 ; CHECK-NEXT:    fmov d2, #2.00000000
 ; CHECK-NEXT:    ptrue p0.d, vl1
-; CHECK-NEXT:    neg x9, x9
+; CHECK-NEXT:    neg x10, x9
 ; CHECK-NEXT:    ptrue p1.b
-; CHECK-NEXT:    mov w10, #100 // =0x64
+; CHECK-NEXT:    mov w11, #100 // =0x64
 ; CHECK-NEXT:    mov x8, xzr
 ; CHECK-NEXT:    sel z3.d, p0, z0.d, z1.d
-; CHECK-NEXT:    and x10, x9, x10
+; CHECK-NEXT:    and x10, x10, x11
 ; CHECK-NEXT:    rdvl x11, #2
 ; CHECK-NEXT:    mov z1.d, p0/m, z2.d
 ; CHECK-NEXT:    ptrue p0.d
@@ -125,7 +125,7 @@ define %"class.std::complex" @complex_mul_nonzero_init_v2f64(ptr %a, ptr %b) {
 ; CHECK-NEXT:    ld1d { z3.d }, p0/z, [x12, #1, mul vl]
 ; CHECK-NEXT:    ld1b { z4.b }, p1/z, [x1, x8]
 ; CHECK-NEXT:    ld1d { z5.d }, p0/z, [x13, #1, mul vl]
-; CHECK-NEXT:    adds x10, x10, x9
+; CHECK-NEXT:    subs x10, x10, x9
 ; CHECK-NEXT:    add x8, x8, x11
 ; CHECK-NEXT:    fcmla z1.d, p0/m, z4.d, z2.d, #0
 ; CHECK-NEXT:    fcmla z0.d, p0/m, z5.d, z3.d, #0
@@ -156,11 +156,11 @@ vector.body:                                      ; preds = %vector.body, %entry
   %scevgep46 = getelementptr i8, ptr %a, i64 %lsr.iv27
   %scevgep47 = getelementptr i8, ptr %b, i64 %lsr.iv27
   %wide.vec = load <vscale x 4 x double>, ptr %scevgep46, align 8
-  %3 = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.experimental.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %wide.vec)
+  %3 = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %wide.vec)
   %4 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %3, 0
   %5 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %3, 1
   %wide.vec30 = load <vscale x 4 x double>, ptr %scevgep47, align 8
-  %6 = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.experimental.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %wide.vec30)
+  %6 = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %wide.vec30)
   %7 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %6, 0
   %8 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %6, 1
   %9 = fmul fast <vscale x 2 x double> %8, %4
@@ -191,13 +191,13 @@ define %"class.std::complex" @complex_mul_v2f64_unrolled(ptr %a, ptr %b) {
 ; CHECK:       // %bb.0: // %entry
 ; CHECK-NEXT:    mov z1.d, #0 // =0x0
 ; CHECK-NEXT:    cntw x9
-; CHECK-NEXT:    mov w10, #1000 // =0x3e8
-; CHECK-NEXT:    neg x9, x9
+; CHECK-NEXT:    mov w11, #1000 // =0x3e8
+; CHECK-NEXT:    neg x10, x9
 ; CHECK-NEXT:    rdvl x12, #2
 ; CHECK-NEXT:    ptrue p1.b
 ; CHECK-NEXT:    ptrue p0.d
 ; CHECK-NEXT:    mov x8, xzr
-; CHECK-NEXT:    and x10, x9, x10
+; CHECK-NEXT:    and x10, x10, x11
 ; CHECK-NEXT:    zip2 z0.d, z1.d, z1.d
 ; CHECK-NEXT:    zip1 z1.d, z1.d, z1.d
 ; CHECK-NEXT:    add x11, x1, x12
@@ -219,7 +219,7 @@ define %"class.std::complex" @complex_mul_v2f64_unrolled(ptr %a, ptr %b) {
 ; CHECK-NEXT:    ld1d { z17.d }, p0/z, [x15, #1, mul vl]
 ; CHECK-NEXT:    ld1b { z18.b }, p1/z, [x11, x8]
 ; CHECK-NEXT:    ld1d { z19.d }, p0/z, [x17, #1, mul vl]
-; CHECK-NEXT:    adds x10, x10, x9
+; CHECK-NEXT:    subs x10, x10, x9
 ; CHECK-NEXT:    add x8, x8, x13
 ; CHECK-NEXT:    fcmla z1.d, p0/m, z7.d, z4.d, #0
 ; CHECK-NEXT:    fcmla z0.d, p0/m, z16.d, z5.d, #0
@@ -266,16 +266,16 @@ vector.body:                                      ; preds = %vector.body, %entry
   %scevgep62 = getelementptr i8, ptr %scevgep61, i64 %lsr.iv34
   %wide.vec = load <vscale x 4 x double>, ptr %scevgep57, align 8
   %wide.vec32 = load <vscale x 4 x double>, ptr %scevgep64, align 8
-  %4 = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.experimental.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %wide.vec)
-  %5 = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.experimental.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %wide.vec32)
+  %4 = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %wide.vec)
+  %5 = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %wide.vec32)
   %6 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %4, 0
   %7 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %5, 0
   %8 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %4, 1
   %9 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %5, 1
   %wide.vec34 = load <vscale x 4 x double>, ptr %scevgep58, align 8
   %wide.vec35 = load <vscale x 4 x double>, ptr %scevgep62, align 8
-  %10 = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.experimental.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %wide.vec34)
-  %11 = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.experimental.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %wide.vec35)
+  %10 = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %wide.vec34)
+  %11 = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %wide.vec35)
   %12 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %10, 0
   %13 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %11, 0
   %14 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %10, 1
@@ -375,7 +375,7 @@ vector.body:                                      ; preds = %vector.body, %entry
   %5 = add <vscale x 2 x i32> %wide.load, %vec.phi
   %6 = getelementptr inbounds %"class.std::complex", ptr %a, i64 %index
   %wide.vec = load <vscale x 4 x double>, ptr %6, align 8
-  %strided.vec = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.experimental.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %wide.vec)
+  %strided.vec = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %wide.vec)
   %7 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %strided.vec, 0
   %8 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %strided.vec, 1
   %9 = fadd fast <vscale x 2 x double> %7, %vec.phi13
@@ -396,6 +396,6 @@ middle.block:                                     ; preds = %vector.body
 
 
 declare i64 @llvm.vscale.i64()
-declare { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.experimental.vector.deinterleave2.nxv4f64(<vscale x 4 x double>)
+declare { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.vector.deinterleave2.nxv4f64(<vscale x 4 x double>)
 declare double @llvm.vector.reduce.fadd.nxv2f64(double, <vscale x 2 x double>)
 declare i32 @llvm.vector.reduce.add.nxv2i32(<vscale x 2 x i32>)
diff --git a/llvm/test/CodeGen/AArch64/complex-deinterleaving-splat-scalable.ll b/llvm/test/CodeGen/AArch64/complex-deinterleaving-splat-scalable.ll
index 17bf5ba6eb48..b4425c0c01e1 100644
--- a/llvm/test/CodeGen/AArch64/complex-deinterleaving-splat-scalable.ll
+++ b/llvm/test/CodeGen/AArch64/complex-deinterleaving-splat-scalable.ll
@@ -28,10 +28,10 @@ define <vscale x 4 x double> @complex_mul_const(<vscale x 4 x double> %a, <vscal
 ; CHECK-NEXT:    mov z1.d, z4.d
 ; CHECK-NEXT:    ret
 entry:
-  %strided.vec = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.experimental.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %a)
+  %strided.vec = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %a)
   %0 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %strided.vec, 0
   %1 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %strided.vec, 1
-  %strided.vec48 = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.experimental.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %b)
+  %strided.vec48 = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %b)
   %2 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %strided.vec48, 0
   %3 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %strided.vec48, 1
   %4 = fmul fast <vscale x 2 x double> %3, %0
@@ -46,7 +46,7 @@ entry:
   %13 = fmul fast <vscale x 2 x double> %9, splat (double 1.100000e+01)
   %14 = fmul fast <vscale x 2 x double> %6, splat (double 3.000000e+00)
   %15 = fsub fast <vscale x 2 x double> %13, %14
-  %interleaved.vec = tail call <vscale x 4 x double> @llvm.experimental.vector.interleave2.nxv4f64(<vscale x 2 x double> %15, <vscale x 2 x double> %12)
+  %interleaved.vec = tail call <vscale x 4 x double> @llvm.vector.interleave2.nxv4f64(<vscale x 2 x double> %15, <vscale x 2 x double> %12)
   ret <vscale x 4 x double> %interleaved.vec
 }
 
@@ -83,10 +83,10 @@ entry:
   %broadcast.splat = shufflevector <vscale x 2 x double> %broadcast.splatinsert, <vscale x 2 x double> poison, <vscale x 2 x i32> zeroinitializer
   %broadcast.splatinsert49 = insertelement <vscale x 2 x double> poison, double %c.coerce.fca.0.extract, i64 0
   %broadcast.splat50 = shufflevector <vscale x 2 x double> %broadcast.splatinsert49, <vscale x 2 x double> poison, <vscale x 2 x i32> zeroinitializer
-  %strided.vec = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.experimental.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %a)
+  %strided.vec = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %a)
   %0 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %strided.vec, 0
   %1 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %strided.vec, 1
-  %strided.vec48 = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.experimental.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %b)
+  %strided.vec48 = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %b)
   %2 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %strided.vec48, 0
   %3 = extractvalue { <vscale x 2 x double>, <vscale x 2 x double> } %strided.vec48, 1
   %4 = fmul fast <vscale x 2 x double> %3, %0
@@ -101,9 +101,9 @@ entry:
   %13 = fmul fast <vscale x 2 x double> %9, %broadcast.splat50
   %14 = fmul fast <vscale x 2 x double> %6, %broadcast.splat
   %15 = fsub fast <vscale x 2 x double> %13, %14
-  %interleaved.vec = tail call <vscale x 4 x double> @llvm.experimental.vector.interleave2.nxv4f64(<vscale x 2 x double> %15, <vscale x 2 x double> %12)
+  %interleaved.vec = tail call <vscale x 4 x double> @llvm.vector.interleave2.nxv4f64(<vscale x 2 x double> %15, <vscale x 2 x double> %12)
   ret <vscale x 4 x double> %interleaved.vec
 }
 
-declare { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.experimental.vector.deinterleave2.nxv4f64(<vscale x 4 x double>)
-declare <vscale x 4 x double> @llvm.experimental.vector.interleave2.nxv4f64(<vscale x 2 x double>, <vscale x 2 x double>)
+declare { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.vector.deinterleave2.nxv4f64(<vscale x 4 x double>)
+declare <vscale x 4 x double> @llvm.vector.interleave2.nxv4f64(<vscale x 2 x double>, <vscale x 2 x double>)
diff --git a/llvm/test/CodeGen/AArch64/concatbinop.ll b/llvm/test/CodeGen/AArch64/concatbinop.ll
new file mode 100644
index 000000000000..828182d18b38
--- /dev/null
+++ b/llvm/test/CodeGen/AArch64/concatbinop.ll
@@ -0,0 +1,201 @@
+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --version 4
+; RUN: llc -mtriple=aarch64 -mattr=+fullfp16 -verify-machineinstrs -o - %s | FileCheck %s
+
+
+define <8 x i16> @concat_add(<4 x i16> %a, <4 x i16> %b, <4 x i16> %c, <4 x i16> %d) {
+; CHECK-LABEL: concat_add:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    // kill: def $d1 killed $d1 def $q1
+; CHECK-NEXT:    // kill: def $d0 killed $d0 def $q0
+; CHECK-NEXT:    // kill: def $d3 killed $d3 def $q3
+; CHECK-NEXT:    // kill: def $d2 killed $d2 def $q2
+; CHECK-NEXT:    mov v1.d[1], v3.d[0]
+; CHECK-NEXT:    mov v0.d[1], v2.d[0]
+; CHECK-NEXT:    add v0.8h, v0.8h, v1.8h
+; CHECK-NEXT:    ret
+  %x = add <4 x i16> %a, %b
+  %y = add <4 x i16> %c, %d
+  %z = shufflevector <4 x i16> %x, <4 x i16> %y, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>
+  ret <8 x i16> %z
+}
+
+define <8 x i16> @concat_addtunc(<4 x i32> %a, <4 x i32> %b, <4 x i32> %c, <4 x i32> %d) {
+; CHECK-LABEL: concat_addtunc:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    add v2.4s, v2.4s, v3.4s
+; CHECK-NEXT:    add v0.4s, v0.4s, v1.4s
+; CHECK-NEXT:    uzp1 v0.8h, v0.8h, v2.8h
+; CHECK-NEXT:    ret
+  %x = add <4 x i32> %a, %b
+  %y = add <4 x i32> %c, %d
+  %xt = trunc <4 x i32> %x to <4 x i16>
+  %yt = trunc <4 x i32> %y to <4 x i16>
+  %z = shufflevector <4 x i16> %xt, <4 x i16> %yt, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>
+  ret <8 x i16> %z
+}
+
+define <8 x i16> @concat_addtunc2(<4 x i32> %a, <4 x i32> %b, <4 x i32> %c, <4 x i32> %d) {
+; CHECK-LABEL: concat_addtunc2:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    uzp1 v1.8h, v1.8h, v3.8h
+; CHECK-NEXT:    uzp1 v0.8h, v0.8h, v2.8h
+; CHECK-NEXT:    add v0.8h, v0.8h, v1.8h
+; CHECK-NEXT:    ret
+  %at = trunc <4 x i32> %a to <4 x i16>
+  %bt = trunc <4 x i32> %b to <4 x i16>
+  %ct = trunc <4 x i32> %c to <4 x i16>
+  %dt = trunc <4 x i32> %d to <4 x i16>
+  %x = add <4 x i16> %at, %bt
+  %y = add <4 x i16> %ct, %dt
+  %z = shufflevector <4 x i16> %x, <4 x i16> %y, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>
+  ret <8 x i16> %z
+}
+
+define <8 x i16> @concat_sub(<4 x i16> %a, <4 x i16> %b, <4 x i16> %c, <4 x i16> %d) {
+; CHECK-LABEL: concat_sub:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    // kill: def $d1 killed $d1 def $q1
+; CHECK-NEXT:    // kill: def $d0 killed $d0 def $q0
+; CHECK-NEXT:    // kill: def $d3 killed $d3 def $q3
+; CHECK-NEXT:    // kill: def $d2 killed $d2 def $q2
+; CHECK-NEXT:    mov v1.d[1], v3.d[0]
+; CHECK-NEXT:    mov v0.d[1], v2.d[0]
+; CHECK-NEXT:    sub v0.8h, v0.8h, v1.8h
+; CHECK-NEXT:    ret
+  %x = sub <4 x i16> %a, %b
+  %y = sub <4 x i16> %c, %d
+  %z = shufflevector <4 x i16> %x, <4 x i16> %y, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>
+  ret <8 x i16> %z
+}
+
+define <8 x i16> @concat_mul(<4 x i16> %a, <4 x i16> %b, <4 x i16> %c, <4 x i16> %d) {
+; CHECK-LABEL: concat_mul:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    // kill: def $d1 killed $d1 def $q1
+; CHECK-NEXT:    // kill: def $d0 killed $d0 def $q0
+; CHECK-NEXT:    // kill: def $d3 killed $d3 def $q3
+; CHECK-NEXT:    // kill: def $d2 killed $d2 def $q2
+; CHECK-NEXT:    mov v1.d[1], v3.d[0]
+; CHECK-NEXT:    mov v0.d[1], v2.d[0]
+; CHECK-NEXT:    mul v0.8h, v0.8h, v1.8h
+; CHECK-NEXT:    ret
+  %x = mul <4 x i16> %a, %b
+  %y = mul <4 x i16> %c, %d
+  %z = shufflevector <4 x i16> %x, <4 x i16> %y, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>
+  ret <8 x i16> %z
+}
+
+define <8 x i16> @concat_xor(<4 x i16> %a, <4 x i16> %b, <4 x i16> %c, <4 x i16> %d) {
+; CHECK-LABEL: concat_xor:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    // kill: def $d1 killed $d1 def $q1
+; CHECK-NEXT:    // kill: def $d0 killed $d0 def $q0
+; CHECK-NEXT:    // kill: def $d3 killed $d3 def $q3
+; CHECK-NEXT:    // kill: def $d2 killed $d2 def $q2
+; CHECK-NEXT:    mov v1.d[1], v3.d[0]
+; CHECK-NEXT:    mov v0.d[1], v2.d[0]
+; CHECK-NEXT:    eor v0.16b, v0.16b, v1.16b
+; CHECK-NEXT:    ret
+  %x = xor <4 x i16> %a, %b
+  %y = xor <4 x i16> %c, %d
+  %z = shufflevector <4 x i16> %x, <4 x i16> %y, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>
+  ret <8 x i16> %z
+}
+
+define <8 x half> @concat_fadd(<4 x half> %a, <4 x half> %b, <4 x half> %c, <4 x half> %d) {
+; CHECK-LABEL: concat_fadd:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    // kill: def $d1 killed $d1 def $q1
+; CHECK-NEXT:    // kill: def $d0 killed $d0 def $q0
+; CHECK-NEXT:    // kill: def $d3 killed $d3 def $q3
+; CHECK-NEXT:    // kill: def $d2 killed $d2 def $q2
+; CHECK-NEXT:    mov v1.d[1], v3.d[0]
+; CHECK-NEXT:    mov v0.d[1], v2.d[0]
+; CHECK-NEXT:    fadd v0.8h, v0.8h, v1.8h
+; CHECK-NEXT:    ret
+  %x = fadd <4 x half> %a, %b
+  %y = fadd <4 x half> %c, %d
+  %z = shufflevector <4 x half> %x, <4 x half> %y, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>
+  ret <8 x half> %z
+}
+
+define <8 x half> @concat_fmul(<4 x half> %a, <4 x half> %b, <4 x half> %c, <4 x half> %d) {
+; CHECK-LABEL: concat_fmul:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    // kill: def $d1 killed $d1 def $q1
+; CHECK-NEXT:    // kill: def $d0 killed $d0 def $q0
+; CHECK-NEXT:    // kill: def $d3 killed $d3 def $q3
+; CHECK-NEXT:    // kill: def $d2 killed $d2 def $q2
+; CHECK-NEXT:    mov v1.d[1], v3.d[0]
+; CHECK-NEXT:    mov v0.d[1], v2.d[0]
+; CHECK-NEXT:    fmul v0.8h, v0.8h, v1.8h
+; CHECK-NEXT:    ret
+  %x = fmul <4 x half> %a, %b
+  %y = fmul <4 x half> %c, %d
+  %z = shufflevector <4 x half> %x, <4 x half> %y, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>
+  ret <8 x half> %z
+}
+
+define <8 x half> @concat_min(<4 x half> %a, <4 x half> %b, <4 x half> %c, <4 x half> %d) {
+; CHECK-LABEL: concat_min:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    // kill: def $d1 killed $d1 def $q1
+; CHECK-NEXT:    // kill: def $d0 killed $d0 def $q0
+; CHECK-NEXT:    // kill: def $d3 killed $d3 def $q3
+; CHECK-NEXT:    // kill: def $d2 killed $d2 def $q2
+; CHECK-NEXT:    mov v1.d[1], v3.d[0]
+; CHECK-NEXT:    mov v0.d[1], v2.d[0]
+; CHECK-NEXT:    fminnm v0.8h, v0.8h, v1.8h
+; CHECK-NEXT:    ret
+  %x = call <4 x half> @llvm.minnum.v4f16(<4 x half> %a, <4 x half> %b)
+  %y = call <4 x half> @llvm.minnum.v4f16(<4 x half> %c, <4 x half> %d)
+  %z = shufflevector <4 x half> %x, <4 x half> %y, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>
+  ret <8 x half> %z
+}
+
+define <8 x half> @concat_minmax(<4 x half> %a, <4 x half> %b, <4 x half> %c, <4 x half> %d) {
+; CHECK-LABEL: concat_minmax:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    fmaxnm v2.4h, v2.4h, v3.4h
+; CHECK-NEXT:    fminnm v0.4h, v0.4h, v1.4h
+; CHECK-NEXT:    mov v0.d[1], v2.d[0]
+; CHECK-NEXT:    ret
+  %x = call <4 x half> @llvm.minnum.v4f16(<4 x half> %a, <4 x half> %b)
+  %y = call <4 x half> @llvm.maxnum.v4f16(<4 x half> %c, <4 x half> %d)
+  %z = shufflevector <4 x half> %x, <4 x half> %y, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>
+  ret <8 x half> %z
+}
+
+define <16 x i8> @signOf_neon(ptr nocapture noundef readonly %a, ptr nocapture noundef readonly %b) {
+; CHECK-LABEL: signOf_neon:
+; CHECK:       // %bb.0: // %entry
+; CHECK-NEXT:    ldp q1, q2, [x0]
+; CHECK-NEXT:    movi v0.16b, #1
+; CHECK-NEXT:    ldp q3, q4, [x1]
+; CHECK-NEXT:    cmhi v5.8h, v1.8h, v3.8h
+; CHECK-NEXT:    cmhi v6.8h, v2.8h, v4.8h
+; CHECK-NEXT:    cmhi v1.8h, v3.8h, v1.8h
+; CHECK-NEXT:    cmhi v2.8h, v4.8h, v2.8h
+; CHECK-NEXT:    uzp1 v3.16b, v5.16b, v6.16b
+; CHECK-NEXT:    uzp1 v1.16b, v1.16b, v2.16b
+; CHECK-NEXT:    and v0.16b, v3.16b, v0.16b
+; CHECK-NEXT:    orr v0.16b, v0.16b, v1.16b
+; CHECK-NEXT:    ret
+entry:
+  %0 = load <8 x i16>, ptr %a, align 2
+  %add.ptr = getelementptr inbounds i8, ptr %a, i64 16
+  %1 = load <8 x i16>, ptr %add.ptr, align 2
+  %2 = load <8 x i16>, ptr %b, align 2
+  %add.ptr6 = getelementptr inbounds i8, ptr %b, i64 16
+  %3 = load <8 x i16>, ptr %add.ptr6, align 2
+  %cmp.i33 = icmp ugt <8 x i16> %0, %2
+  %cmp.i31 = icmp ugt <8 x i16> %1, %3
+  %cmp.i29 = icmp ugt <8 x i16> %2, %0
+  %cmp.i = icmp ugt <8 x i16> %3, %1
+  %vmovn.i38.neg = zext <8 x i1> %cmp.i33 to <8 x i8>
+  %vmovn.i37.neg = zext <8 x i1> %cmp.i31 to <8 x i8>
+  %4 = select <8 x i1> %cmp.i29, <8 x i8> <i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1>, <8 x i8> %vmovn.i38.neg
+  %5 = select <8 x i1> %cmp.i, <8 x i8> <i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1>, <8 x i8> %vmovn.i37.neg
+  %or.i = shufflevector <8 x i8> %4, <8 x i8> %5, <16 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15>
+  ret <16 x i8> %or.i
+}
diff --git a/llvm/test/CodeGen/AArch64/fixed-vector-deinterleave.ll b/llvm/test/CodeGen/AArch64/fixed-vector-deinterleave.ll
index 2ad5623b6551..c58db8290c87 100644
--- a/llvm/test/CodeGen/AArch64/fixed-vector-deinterleave.ll
+++ b/llvm/test/CodeGen/AArch64/fixed-vector-deinterleave.ll
@@ -25,7 +25,7 @@ define {<2 x half>, <2 x half>} @vector_deinterleave_v2f16_v4f16(<4 x half> %vec
 ; CHECK-GI-NEXT:    // kill: def $d1 killed $d1 killed $q1
 ; CHECK-GI-NEXT:    fmov d0, d2
 ; CHECK-GI-NEXT:    ret
-  %retval = call {<2 x half>, <2 x half>} @llvm.experimental.vector.deinterleave2.v4f16(<4 x half> %vec)
+  %retval = call {<2 x half>, <2 x half>} @llvm.vector.deinterleave2.v4f16(<4 x half> %vec)
   ret {<2 x half>, <2 x half>}   %retval
 }
 
@@ -45,7 +45,7 @@ define {<4 x half>, <4 x half>} @vector_deinterleave_v4f16_v8f16(<8 x half> %vec
 ; CHECK-GI-NEXT:    // kill: def $d1 killed $d1 killed $q1
 ; CHECK-GI-NEXT:    fmov d0, d2
 ; CHECK-GI-NEXT:    ret
-  %retval = call {<4 x half>, <4 x half>} @llvm.experimental.vector.deinterleave2.v8f16(<8 x half> %vec)
+  %retval = call {<4 x half>, <4 x half>} @llvm.vector.deinterleave2.v8f16(<8 x half> %vec)
   ret {<4 x half>, <4 x half>}   %retval
 }
 
@@ -56,7 +56,7 @@ define {<8 x half>, <8 x half>} @vector_deinterleave_v8f16_v16f16(<16 x half> %v
 ; CHECK-NEXT:    uzp2 v1.8h, v0.8h, v1.8h
 ; CHECK-NEXT:    mov v0.16b, v2.16b
 ; CHECK-NEXT:    ret
-  %retval = call {<8 x half>, <8 x half>} @llvm.experimental.vector.deinterleave2.v16f16(<16 x half> %vec)
+  %retval = call {<8 x half>, <8 x half>} @llvm.vector.deinterleave2.v16f16(<16 x half> %vec)
   ret {<8 x half>, <8 x half>}   %retval
 }
 
@@ -76,7 +76,7 @@ define {<2 x float>, <2 x float>} @vector_deinterleave_v2f32_v4f32(<4 x float> %
 ; CHECK-GI-NEXT:    // kill: def $d1 killed $d1 killed $q1
 ; CHECK-GI-NEXT:    fmov d0, d2
 ; CHECK-GI-NEXT:    ret
-  %retval = call {<2 x float>, <2 x float>} @llvm.experimental.vector.deinterleave2.v4f32(<4 x float> %vec)
+  %retval = call {<2 x float>, <2 x float>} @llvm.vector.deinterleave2.v4f32(<4 x float> %vec)
   ret {<2 x float>, <2 x float>}   %retval
 }
 
@@ -87,7 +87,7 @@ define {<4 x float>, <4 x float>} @vector_deinterleave_v4f32_v8f32(<8 x float> %
 ; CHECK-NEXT:    uzp2 v1.4s, v0.4s, v1.4s
 ; CHECK-NEXT:    mov v0.16b, v2.16b
 ; CHECK-NEXT:    ret
-  %retval = call {<4 x float>, <4 x float>} @llvm.experimental.vector.deinterleave2.v8f32(<8 x float> %vec)
+  %retval = call {<4 x float>, <4 x float>} @llvm.vector.deinterleave2.v8f32(<8 x float> %vec)
 ret  {<4 x float>, <4 x float>}   %retval
 }
 
@@ -98,7 +98,7 @@ define {<2 x double>, <2 x double>} @vector_deinterleave_v2f64_v4f64(<4 x double
 ; CHECK-NEXT:    zip2 v1.2d, v0.2d, v1.2d
 ; CHECK-NEXT:    mov v0.16b, v2.16b
 ; CHECK-NEXT:    ret
-  %retval = call {<2 x double>, <2 x double>} @llvm.experimental.vector.deinterleave2.v4f64(<4 x double> %vec)
+  %retval = call {<2 x double>, <2 x double>} @llvm.vector.deinterleave2.v4f64(<4 x double> %vec)
   ret {<2 x double>, <2 x double>}   %retval
 }
 
@@ -111,7 +111,7 @@ define {<16 x i8>, <16 x i8>} @vector_deinterleave_v16i8_v32i8(<32 x i8> %vec) {
 ; CHECK-NEXT:    uzp2 v1.16b, v0.16b, v1.16b
 ; CHECK-NEXT:    mov v0.16b, v2.16b
 ; CHECK-NEXT:    ret
-  %retval = call {<16 x i8>, <16 x i8>} @llvm.experimental.vector.deinterleave2.v32i8(<32 x i8> %vec)
+  %retval = call {<16 x i8>, <16 x i8>} @llvm.vector.deinterleave2.v32i8(<32 x i8> %vec)
   ret {<16 x i8>, <16 x i8>}   %retval
 }
 
@@ -122,7 +122,7 @@ define {<8 x i16>, <8 x i16>} @vector_deinterleave_v8i16_v16i16(<16 x i16> %vec)
 ; CHECK-NEXT:    uzp2 v1.8h, v0.8h, v1.8h
 ; CHECK-NEXT:    mov v0.16b, v2.16b
 ; CHECK-NEXT:    ret
-  %retval = call {<8 x i16>, <8 x i16>} @llvm.experimental.vector.deinterleave2.v16i16(<16 x i16> %vec)
+  %retval = call {<8 x i16>, <8 x i16>} @llvm.vector.deinterleave2.v16i16(<16 x i16> %vec)
   ret {<8 x i16>, <8 x i16>}   %retval
 }
 
@@ -133,7 +133,7 @@ define {<4 x i32>, <4 x i32>} @vector_deinterleave_v4i32_v8i32(<8 x i32> %vec) {
 ; CHECK-NEXT:    uzp2 v1.4s, v0.4s, v1.4s
 ; CHECK-NEXT:    mov v0.16b, v2.16b
 ; CHECK-NEXT:    ret
-  %retval = call {<4 x i32>, <4 x i32>} @llvm.experimental.vector.deinterleave2.v8i32(<8 x i32> %vec)
+  %retval = call {<4 x i32>, <4 x i32>} @llvm.vector.deinterleave2.v8i32(<8 x i32> %vec)
   ret {<4 x i32>, <4 x i32>}   %retval
 }
 
@@ -144,22 +144,22 @@ define {<2 x i64>, <2 x i64>} @vector_deinterleave_v2i64_v4i64(<4 x i64> %vec) {
 ; CHECK-NEXT:    zip2 v1.2d, v0.2d, v1.2d
 ; CHECK-NEXT:    mov v0.16b, v2.16b
 ; CHECK-NEXT:    ret
-  %retval = call {<2 x i64>, <2 x i64>} @llvm.experimental.vector.deinterleave2.v4i64(<4 x i64> %vec)
+  %retval = call {<2 x i64>, <2 x i64>} @llvm.vector.deinterleave2.v4i64(<4 x i64> %vec)
   ret {<2 x i64>, <2 x i64>}   %retval
 }
 
 
 ; Floating declarations
-declare {<2 x half>,<2 x half>} @llvm.experimental.vector.deinterleave2.v4f16(<4 x half>)
-declare {<4 x half>, <4 x half>} @llvm.experimental.vector.deinterleave2.v8f16(<8 x half>)
-declare {<2 x float>, <2 x float>} @llvm.experimental.vector.deinterleave2.v4f32(<4 x float>)
-declare {<8 x half>, <8 x half>} @llvm.experimental.vector.deinterleave2.v16f16(<16 x half>)
-declare {<4 x float>, <4 x float>} @llvm.experimental.vector.deinterleave2.v8f32(<8 x float>)
-declare {<2 x double>, <2 x double>} @llvm.experimental.vector.deinterleave2.v4f64(<4 x double>)
+declare {<2 x half>,<2 x half>} @llvm.vector.deinterleave2.v4f16(<4 x half>)
+declare {<4 x half>, <4 x half>} @llvm.vector.deinterleave2.v8f16(<8 x half>)
+declare {<2 x float>, <2 x float>} @llvm.vector.deinterleave2.v4f32(<4 x float>)
+declare {<8 x half>, <8 x half>} @llvm.vector.deinterleave2.v16f16(<16 x half>)
+declare {<4 x float>, <4 x float>} @llvm.vector.deinterleave2.v8f32(<8 x float>)
+declare {<2 x double>, <2 x double>} @llvm.vector.deinterleave2.v4f64(<4 x double>)
 
 ; Integer declarations
-declare {<16 x i8>, <16 x i8>} @llvm.experimental.vector.deinterleave2.v32i8(<32 x i8>)
-declare {<8 x i16>, <8 x i16>} @llvm.experimental.vector.deinterleave2.v16i16(<16 x i16>)
-declare {<4 x i32>, <4 x i32>} @llvm.experimental.vector.deinterleave2.v8i32(<8 x i32>)
-declare {<2 x i64>, <2 x i64>} @llvm.experimental.vector.deinterleave2.v4i64(<4 x i64>)
+declare {<16 x i8>, <16 x i8>} @llvm.vector.deinterleave2.v32i8(<32 x i8>)
+declare {<8 x i16>, <8 x i16>} @llvm.vector.deinterleave2.v16i16(<16 x i16>)
+declare {<4 x i32>, <4 x i32>} @llvm.vector.deinterleave2.v8i32(<8 x i32>)
+declare {<2 x i64>, <2 x i64>} @llvm.vector.deinterleave2.v4i64(<4 x i64>)
 
diff --git a/llvm/test/CodeGen/AArch64/fixed-vector-interleave.ll b/llvm/test/CodeGen/AArch64/fixed-vector-interleave.ll
index eb81aff33e49..2e992964f598 100644
--- a/llvm/test/CodeGen/AArch64/fixed-vector-interleave.ll
+++ b/llvm/test/CodeGen/AArch64/fixed-vector-interleave.ll
@@ -7,7 +7,7 @@ define <4 x half> @interleave2_v4f16(<2 x half> %vec0, <2 x half> %vec1) {
 ; CHECK:       // %bb.0:
 ; CHECK-NEXT:    zip1 v0.4h, v0.4h, v1.4h
 ; CHECK-NEXT:    ret
-  %retval = call <4 x half> @llvm.experimental.vector.interleave2.v4f16(<2 x half> %vec0, <2 x half> %vec1)
+  %retval = call <4 x half> @llvm.vector.interleave2.v4f16(<2 x half> %vec0, <2 x half> %vec1)
   ret <4 x half> %retval
 }
 
@@ -28,7 +28,7 @@ define <8 x half> @interleave2_v8f16(<4 x half> %vec0, <4 x half> %vec1) {
 ; CHECK-GI-NEXT:    // kill: def $d1 killed $d1 def $q1
 ; CHECK-GI-NEXT:    zip1 v0.8h, v0.8h, v1.8h
 ; CHECK-GI-NEXT:    ret
-  %retval = call <8 x half> @llvm.experimental.vector.interleave2.v8f16(<4 x half> %vec0, <4 x half> %vec1)
+  %retval = call <8 x half> @llvm.vector.interleave2.v8f16(<4 x half> %vec0, <4 x half> %vec1)
   ret <8 x half> %retval
 }
 
@@ -39,7 +39,7 @@ define <16 x half> @interleave2_v16f16(<8 x half> %vec0, <8 x half> %vec1) {
 ; CHECK-NEXT:    zip2 v1.8h, v0.8h, v1.8h
 ; CHECK-NEXT:    mov v0.16b, v2.16b
 ; CHECK-NEXT:    ret
-  %retval = call <16 x half> @llvm.experimental.vector.interleave2.v16f16(<8 x half> %vec0, <8 x half> %vec1)
+  %retval = call <16 x half> @llvm.vector.interleave2.v16f16(<8 x half> %vec0, <8 x half> %vec1)
   ret <16 x half> %retval
 }
 
@@ -59,7 +59,7 @@ define <4 x float> @interleave2_v4f32(<2 x float> %vec0, <2 x float> %vec1) {
 ; CHECK-GI-NEXT:    // kill: def $d1 killed $d1 def $q1
 ; CHECK-GI-NEXT:    zip1 v0.4s, v0.4s, v1.4s
 ; CHECK-GI-NEXT:    ret
-  %retval = call <4 x float> @llvm.experimental.vector.interleave2.v4f32(<2 x float> %vec0, <2 x float> %vec1)
+  %retval = call <4 x float> @llvm.vector.interleave2.v4f32(<2 x float> %vec0, <2 x float> %vec1)
   ret <4 x float> %retval
 }
 
@@ -70,7 +70,7 @@ define <8 x float> @interleave2_v8f32(<4 x float> %vec0, <4 x float> %vec1) {
 ; CHECK-NEXT:    zip2 v1.4s, v0.4s, v1.4s
 ; CHECK-NEXT:    mov v0.16b, v2.16b
 ; CHECK-NEXT:    ret
-  %retval = call <8 x float> @llvm.experimental.vector.interleave2.v8f32(<4 x float> %vec0, <4 x float> %vec1)
+  %retval = call <8 x float> @llvm.vector.interleave2.v8f32(<4 x float> %vec0, <4 x float> %vec1)
   ret <8 x float> %retval
 }
 
@@ -81,7 +81,7 @@ define <4 x double> @interleave2_v4f64(<2 x double> %vec0, <2 x double> %vec1) {
 ; CHECK-NEXT:    zip2 v1.2d, v0.2d, v1.2d
 ; CHECK-NEXT:    mov v0.16b, v2.16b
 ; CHECK-NEXT:    ret
-  %retval = call <4 x double>@llvm.experimental.vector.interleave2.v4f64(<2 x double> %vec0, <2 x double> %vec1)
+  %retval = call <4 x double>@llvm.vector.interleave2.v4f64(<2 x double> %vec0, <2 x double> %vec1)
   ret <4 x double> %retval
 }
 
@@ -94,7 +94,7 @@ define <32 x i8> @interleave2_v32i8(<16 x i8> %vec0, <16 x i8> %vec1) {
 ; CHECK-NEXT:    zip2 v1.16b, v0.16b, v1.16b
 ; CHECK-NEXT:    mov v0.16b, v2.16b
 ; CHECK-NEXT:    ret
- %retval = call <32 x i8> @llvm.experimental.vector.interleave2.v32i8(<16 x i8> %vec0, <16 x i8> %vec1)
+ %retval = call <32 x i8> @llvm.vector.interleave2.v32i8(<16 x i8> %vec0, <16 x i8> %vec1)
   ret <32 x i8> %retval
 }
 
@@ -105,7 +105,7 @@ define <16 x i16> @interleave2_v16i16(<8 x i16> %vec0, <8 x i16> %vec1) {
 ; CHECK-NEXT:    zip2 v1.8h, v0.8h, v1.8h
 ; CHECK-NEXT:    mov v0.16b, v2.16b
 ; CHECK-NEXT:    ret
-  %retval = call <16 x i16> @llvm.experimental.vector.interleave2.v16i16(<8 x i16> %vec0, <8 x i16> %vec1)
+  %retval = call <16 x i16> @llvm.vector.interleave2.v16i16(<8 x i16> %vec0, <8 x i16> %vec1)
   ret <16 x i16> %retval
 }
 
@@ -116,7 +116,7 @@ define <8 x i32> @interleave2_v8i32(<4 x i32> %vec0, <4 x i32> %vec1) {
 ; CHECK-NEXT:    zip2 v1.4s, v0.4s, v1.4s
 ; CHECK-NEXT:    mov v0.16b, v2.16b
 ; CHECK-NEXT:    ret
-  %retval = call <8 x i32> @llvm.experimental.vector.interleave2.v8i32(<4 x i32> %vec0, <4 x i32> %vec1)
+  %retval = call <8 x i32> @llvm.vector.interleave2.v8i32(<4 x i32> %vec0, <4 x i32> %vec1)
   ret <8 x i32> %retval
 }
 
@@ -127,22 +127,22 @@ define <4 x i64> @interleave2_v4i64(<2 x i64> %vec0, <2 x i64> %vec1) {
 ; CHECK-NEXT:    zip2 v1.2d, v0.2d, v1.2d
 ; CHECK-NEXT:    mov v0.16b, v2.16b
 ; CHECK-NEXT:    ret
-  %retval = call <4 x i64> @llvm.experimental.vector.interleave2.v4i64(<2 x i64> %vec0, <2 x i64> %vec1)
+  %retval = call <4 x i64> @llvm.vector.interleave2.v4i64(<2 x i64> %vec0, <2 x i64> %vec1)
   ret <4 x i64> %retval
 }
 
 
 ; Float declarations
-declare <4 x half> @llvm.experimental.vector.interleave2.v4f16(<2 x half>, <2 x half>)
-declare <8 x half> @llvm.experimental.vector.interleave2.v8f16(<4 x half>, <4 x half>)
-declare <16 x half> @llvm.experimental.vector.interleave2.v16f16(<8 x half>, <8 x half>)
-declare <4 x float> @llvm.experimental.vector.interleave2.v4f32(<2 x float>, <2 x float>)
-declare <8 x float> @llvm.experimental.vector.interleave2.v8f32(<4 x float>, <4 x float>)
-declare <4 x double> @llvm.experimental.vector.interleave2.v4f64(<2 x double>, <2 x double>)
+declare <4 x half> @llvm.vector.interleave2.v4f16(<2 x half>, <2 x half>)
+declare <8 x half> @llvm.vector.interleave2.v8f16(<4 x half>, <4 x half>)
+declare <16 x half> @llvm.vector.interleave2.v16f16(<8 x half>, <8 x half>)
+declare <4 x float> @llvm.vector.interleave2.v4f32(<2 x float>, <2 x float>)
+declare <8 x float> @llvm.vector.interleave2.v8f32(<4 x float>, <4 x float>)
+declare <4 x double> @llvm.vector.interleave2.v4f64(<2 x double>, <2 x double>)
 
 ; Integer declarations
-declare <32 x i8> @llvm.experimental.vector.interleave2.v32i8(<16 x i8>, <16 x i8>)
-declare <16 x i16> @llvm.experimental.vector.interleave2.v16i16(<8 x i16>, <8 x i16>)
-declare <8 x i32> @llvm.experimental.vector.interleave2.v8i32(<4 x i32>, <4 x i32>)
-declare <4 x i64> @llvm.experimental.vector.interleave2.v4i64(<2 x i64>, <2 x i64>)
+declare <32 x i8> @llvm.vector.interleave2.v32i8(<16 x i8>, <16 x i8>)
+declare <16 x i16> @llvm.vector.interleave2.v16i16(<8 x i16>, <8 x i16>)
+declare <8 x i32> @llvm.vector.interleave2.v8i32(<4 x i32>, <4 x i32>)
+declare <4 x i64> @llvm.vector.interleave2.v4i64(<2 x i64>, <2 x i64>)
 
diff --git a/llvm/test/CodeGen/AArch64/fpmode.ll b/llvm/test/CodeGen/AArch64/fpmode.ll
index ebfb0696a95a..b185d9e04941 100644
--- a/llvm/test/CodeGen/AArch64/fpmode.ll
+++ b/llvm/test/CodeGen/AArch64/fpmode.ll
@@ -6,17 +6,14 @@ declare i32 @llvm.get.fpmode.i32()
 declare void @llvm.set.fpmode.i32(i32 %fpmode)
 declare void @llvm.reset.fpmode()
 
-define i32 @func_get_fpmode_soft() #0 {
-; DAG-LABEL: func_get_fpmode_soft:
+define i32 @func_get_fpmode() #0 {
+; DAG-LABEL: func_get_fpmode:
 ; DAG:       // %bb.0: // %entry
-; DAG-NEXT:    str x30, [sp, #-16]! // 8-byte Folded Spill
-; DAG-NEXT:    add x0, sp, #12
-; DAG-NEXT:    bl fegetmode
-; DAG-NEXT:    ldr w0, [sp, #12]
-; DAG-NEXT:    ldr x30, [sp], #16 // 8-byte Folded Reload
+; DAG-NEXT:    mrs x0, FPCR
+; DAG-NEXT:    // kill: def $w0 killed $w0 killed $x0
 ; DAG-NEXT:    ret
 ;
-; GIS-LABEL: func_get_fpmode_soft:
+; GIS-LABEL: func_get_fpmode:
 ; GIS:       // %bb.0: // %entry
 ; GIS-NEXT:    str x30, [sp, #-16]! // 8-byte Folded Spill
 ; GIS-NEXT:    add x0, sp, #12
@@ -29,17 +26,14 @@ entry:
   ret i32 %fpmode
 }
 
-define void @func_set_fpmode_soft(i32 %fpmode) #0 {
-; DAG-LABEL: func_set_fpmode_soft:
+define void @func_set_fpmode(i32 %fpmode) #0 {
+; DAG-LABEL: func_set_fpmode:
 ; DAG:       // %bb.0: // %entry
-; DAG-NEXT:    str x30, [sp, #-16]! // 8-byte Folded Spill
-; DAG-NEXT:    str w0, [sp, #12]
-; DAG-NEXT:    add x0, sp, #12
-; DAG-NEXT:    bl fesetmode
-; DAG-NEXT:    ldr x30, [sp], #16 // 8-byte Folded Reload
+; DAG-NEXT:    mov w8, w0
+; DAG-NEXT:    msr FPCR, x8
 ; DAG-NEXT:    ret
 ;
-; GIS-LABEL: func_set_fpmode_soft:
+; GIS-LABEL: func_set_fpmode:
 ; GIS:       // %bb.0: // %entry
 ; GIS-NEXT:    str x30, [sp, #-16]! // 8-byte Folded Spill
 ; GIS-NEXT:    str w0, [sp, #12]
@@ -52,16 +46,17 @@ entry:
   ret void
 }
 
-define void @func_reset_fpmode_soft() #0 {
-; DAG-LABEL: func_reset_fpmode_soft:
+define void @func_reset_fpmode() #0 {
+; DAG-LABEL: func_reset_fpmode:
 ; DAG:       // %bb.0: // %entry
-; DAG-NEXT:    str x30, [sp, #-16]! // 8-byte Folded Spill
-; DAG-NEXT:    mov x0, #-1 // =0xffffffffffffffff
-; DAG-NEXT:    bl fesetmode
-; DAG-NEXT:    ldr x30, [sp], #16 // 8-byte Folded Reload
+; DAG-NEXT:    mov x9, #-48904 // =0xffffffffffff40f8
+; DAG-NEXT:    mrs x8, FPCR
+; DAG-NEXT:    movk x9, #63488, lsl #16
+; DAG-NEXT:    and x8, x8, x9
+; DAG-NEXT:    msr FPCR, x8
 ; DAG-NEXT:    ret
 ;
-; GIS-LABEL: func_reset_fpmode_soft:
+; GIS-LABEL: func_reset_fpmode:
 ; GIS:       // %bb.0: // %entry
 ; GIS-NEXT:    mov x0, #-1 // =0xffffffffffffffff
 ; GIS-NEXT:    b fesetmode
@@ -70,4 +65,4 @@ entry:
   ret void
 }
 
-attributes #0 = { nounwind "use-soft-float"="true" }
+attributes #0 = { nounwind }
diff --git a/llvm/test/CodeGen/AArch64/machine-combiner-subadd2.mir b/llvm/test/CodeGen/AArch64/machine-combiner-subadd2.mir
index d1770bb25fae..0b09e8a4b5cd 100644
--- a/llvm/test/CodeGen/AArch64/machine-combiner-subadd2.mir
+++ b/llvm/test/CodeGen/AArch64/machine-combiner-subadd2.mir
@@ -237,3 +237,30 @@ body:              |
     RET_ReallyLR implicit $w0
 
 ...
+---
+# Drop nowrap flags in SUB
+
+# CHECK-LABEL: name: test8
+# CHECK:       %7:gpr64 = SUBXrr %1, %0
+# CHECK-NEXT:  %4:gpr64common = SUBXrr killed %7, killed %2
+
+name:            test8
+registers:
+  - { id: 0, class: gpr64 }
+  - { id: 1, class: gpr64 }
+  - { id: 2, class: gpr64common }
+  - { id: 3, class: gpr64 }
+  - { id: 4, class: gpr64common }
+  - { id: 5, class: gpr64 }
+body:              |
+  bb.0:
+    %1:gpr64 = COPY $x1
+    %0:gpr64 = COPY $x0
+    %2:gpr64common = ORRXri %0:gpr64, 4096
+    %3:gpr64 = ADDXrr killed %2:gpr64common, %0:gpr64
+    %4:gpr64common = nsw SUBSXrr %1:gpr64, killed %3:gpr64, implicit-def dead $nzcv
+    %5:gpr64 = SUBSXri %4:gpr64common, 0, 0, implicit-def $nzcv
+    $x0 = COPY %5:gpr64
+    RET_ReallyLR implicit $x0
+
+...
diff --git a/llvm/test/CodeGen/AArch64/mul_pow2.ll b/llvm/test/CodeGen/AArch64/mul_pow2.ll
index 90e560af4465..0c9ea51ba367 100644
--- a/llvm/test/CodeGen/AArch64/mul_pow2.ll
+++ b/llvm/test/CodeGen/AArch64/mul_pow2.ll
@@ -410,6 +410,23 @@ define i32 @test11(i32 %x) {
   ret i32 %mul
 }
 
+define i32 @test11_fast_shift(i32 %x) "target-features"="+alu-lsl-fast" {
+; CHECK-LABEL: test11_fast_shift:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    add w8, w0, w0, lsl #2
+; CHECK-NEXT:    add w0, w0, w8, lsl #1
+; CHECK-NEXT:    ret
+;
+; GISEL-LABEL: test11_fast_shift:
+; GISEL:       // %bb.0:
+; GISEL-NEXT:    mov w8, #11 // =0xb
+; GISEL-NEXT:    mul w0, w0, w8
+; GISEL-NEXT:    ret
+
+  %mul = mul nsw i32 %x, 11 ; 11 = (((1<<2) + 1) << 1) + 1
+  ret i32 %mul
+}
+
 define i32 @test12(i32 %x) {
 ; CHECK-LABEL: test12:
 ; CHECK:       // %bb.0:
@@ -545,12 +562,29 @@ define i32 @test45(i32 %x) {
   ret i32 %mul
 }
 
-; Negative test: The shift amount 4 larger than 3
+; Negative test: The shift number 5 is out of bound
+define i32 @test67_fast_shift(i32 %x) "target-features"="+alu-lsl-fast" {
+; CHECK-LABEL: test67_fast_shift:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    mov w8, #67 // =0x43
+; CHECK-NEXT:    mul w0, w0, w8
+; CHECK-NEXT:    ret
+;
+; GISEL-LABEL: test67_fast_shift:
+; GISEL:       // %bb.0:
+; GISEL-NEXT:    mov w8, #67 // =0x43
+; GISEL-NEXT:    mul w0, w0, w8
+; GISEL-NEXT:    ret
+
+  %mul = mul nsw i32 %x, 67 ; 67 = (((1<<5) + 1) << 1) + 1
+  ret i32 %mul
+}
+
 define i32 @test85_fast_shift(i32 %x) "target-features"="+alu-lsl-fast" {
 ; CHECK-LABEL: test85_fast_shift:
 ; CHECK:       // %bb.0:
-; CHECK-NEXT:    mov w8, #85 // =0x55
-; CHECK-NEXT:    mul w0, w0, w8
+; CHECK-NEXT:    add w8, w0, w0, lsl #2
+; CHECK-NEXT:    add w0, w8, w8, lsl #4
 ; CHECK-NEXT:    ret
 ;
 ; GISEL-LABEL: test85_fast_shift:
@@ -563,7 +597,25 @@ define i32 @test85_fast_shift(i32 %x) "target-features"="+alu-lsl-fast" {
   ret i32 %mul
 }
 
-; Negative test: The shift amount 5 larger than 3
+; Negative test: The shift number 5 is out of bound
+define i32 @test97_fast_shift(i32 %x) "target-features"="+alu-lsl-fast" {
+; CHECK-LABEL: test97_fast_shift:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    mov w8, #97 // =0x61
+; CHECK-NEXT:    mul w0, w0, w8
+; CHECK-NEXT:    ret
+;
+; GISEL-LABEL: test97_fast_shift:
+; GISEL:       // %bb.0:
+; GISEL-NEXT:    mov w8, #97 // =0x61
+; GISEL-NEXT:    mul w0, w0, w8
+; GISEL-NEXT:    ret
+
+  %mul = mul nsw i32 %x, 97 ; 97 = ((2 + 1) << 5) + 1
+  ret i32 %mul
+}
+
+; Negative test: The shift amount 5 larger than 4
 define i32 @test297_fast_shift(i32 %x) "target-features"="+alu-lsl-fast" {
 ; CHECK-LABEL: test297_fast_shift:
 ; CHECK:       // %bb.0:
@@ -858,9 +910,9 @@ define <4 x i32> @muladd_demand_commute(<4 x i32> %x, <4 x i32> %y) {
 ;
 ; GISEL-LABEL: muladd_demand_commute:
 ; GISEL:       // %bb.0:
-; GISEL-NEXT:    adrp x8, .LCPI49_0
+; GISEL-NEXT:    adrp x8, .LCPI52_0
 ; GISEL-NEXT:    movi v3.4s, #1, msl #16
-; GISEL-NEXT:    ldr q2, [x8, :lo12:.LCPI49_0]
+; GISEL-NEXT:    ldr q2, [x8, :lo12:.LCPI52_0]
 ; GISEL-NEXT:    mla v1.4s, v0.4s, v2.4s
 ; GISEL-NEXT:    and v0.16b, v1.16b, v3.16b
 ; GISEL-NEXT:    ret
diff --git a/llvm/test/CodeGen/AArch64/named-vector-shuffle-reverse-neon.ll b/llvm/test/CodeGen/AArch64/named-vector-shuffle-reverse-neon.ll
index 0eee19ad2adb..cff7759c72c9 100644
--- a/llvm/test/CodeGen/AArch64/named-vector-shuffle-reverse-neon.ll
+++ b/llvm/test/CodeGen/AArch64/named-vector-shuffle-reverse-neon.ll
@@ -15,7 +15,7 @@ define <16 x i8> @reverse_v16i8(<16 x i8> %a) #0 {
 ; CHECK-NEXT:    ext v0.16b, v0.16b, v0.16b, #8
 ; CHECK-NEXT:    ret
 
-  %res = call <16 x i8> @llvm.experimental.vector.reverse.v16i8(<16 x i8> %a)
+  %res = call <16 x i8> @llvm.vector.reverse.v16i8(<16 x i8> %a)
   ret <16 x i8> %res
 }
 
@@ -26,7 +26,7 @@ define <8 x i16> @reverse_v8i16(<8 x i16> %a) #0 {
 ; CHECK-NEXT:    ext v0.16b, v0.16b, v0.16b, #8
 ; CHECK-NEXT:    ret
 
-  %res = call <8 x i16> @llvm.experimental.vector.reverse.v8i16(<8 x i16> %a)
+  %res = call <8 x i16> @llvm.vector.reverse.v8i16(<8 x i16> %a)
   ret <8 x i16> %res
 }
 
@@ -35,7 +35,7 @@ define <2 x i16> @reverse_v2i16(<2 x i16> %a) #0 {
 ; CHECK:       // %bb.0:
 ; CHECK-NEXT:    rev64 v0.2s, v0.2s
 ; CHECK-NEXT:    ret
-  %res = call <2 x i16> @llvm.experimental.vector.reverse.v2i16(<2 x i16> %a)
+  %res = call <2 x i16> @llvm.vector.reverse.v2i16(<2 x i16> %a)
   ret <2 x i16> %res
 }
 
@@ -44,7 +44,7 @@ define <2 x i32> @reverse_v2i32(<2 x i32> %a) #0 {
 ; CHECK:       // %bb.0:
 ; CHECK-NEXT:    rev64 v0.2s, v0.2s
 ; CHECK-NEXT:    ret
-  %res = call <2 x i32> @llvm.experimental.vector.reverse.v2i32(<2 x i32> %a)
+  %res = call <2 x i32> @llvm.vector.reverse.v2i32(<2 x i32> %a)
   ret <2 x i32> %res
 }
 
@@ -55,7 +55,7 @@ define <4 x i32> @reverse_v4i32(<4 x i32> %a) #0 {
 ; CHECK-NEXT:    ext v0.16b, v0.16b, v0.16b, #8
 ; CHECK-NEXT:    ret
 
-  %res = call <4 x i32> @llvm.experimental.vector.reverse.v4i32(<4 x i32> %a)
+  %res = call <4 x i32> @llvm.vector.reverse.v4i32(<4 x i32> %a)
   ret <4 x i32> %res
 }
 
@@ -65,7 +65,7 @@ define <2 x i64> @reverse_v2i64(<2 x i64> %a) #0 {
 ; CHECK-NEXT:    ext v0.16b, v0.16b, v0.16b, #8
 ; CHECK-NEXT:    ret
 
-  %res = call <2 x i64> @llvm.experimental.vector.reverse.v2i64(<2 x i64> %a)
+  %res = call <2 x i64> @llvm.vector.reverse.v2i64(<2 x i64> %a)
   ret <2 x i64> %res
 }
 
@@ -76,7 +76,7 @@ define <8 x half> @reverse_v8f16(<8 x half> %a) #0 {
 ; CHECK-NEXT:    ext v0.16b, v0.16b, v0.16b, #8
 ; CHECK-NEXT:    ret
 
-  %res = call <8 x half> @llvm.experimental.vector.reverse.v8f16(<8 x half> %a)
+  %res = call <8 x half> @llvm.vector.reverse.v8f16(<8 x half> %a)
   ret <8 x half> %res
 }
 
@@ -85,7 +85,7 @@ define <2 x float> @reverse_v2f32(<2 x float> %a) #0 {
 ; CHECK:       // %bb.0:
 ; CHECK-NEXT:    rev64 v0.2s, v0.2s
 ; CHECK-NEXT:    ret
-  %res = call <2 x float> @llvm.experimental.vector.reverse.v2f32(<2 x float> %a)
+  %res = call <2 x float> @llvm.vector.reverse.v2f32(<2 x float> %a)
   ret <2 x float> %res
 }
 
@@ -96,7 +96,7 @@ define <4 x float> @reverse_v4f32(<4 x float> %a) #0 {
 ; CHECK-NEXT:    ext v0.16b, v0.16b, v0.16b, #8
 ; CHECK-NEXT:    ret
 
-  %res = call <4 x float> @llvm.experimental.vector.reverse.v4f32(<4 x float> %a)
+  %res = call <4 x float> @llvm.vector.reverse.v4f32(<4 x float> %a)
   ret <4 x float> %res
 }
 
@@ -106,7 +106,7 @@ define <2 x double> @reverse_v2f64(<2 x double> %a) #0 {
 ; CHECK-NEXT:    ext v0.16b, v0.16b, v0.16b, #8
 ; CHECK-NEXT:    ret
 
-  %res = call <2 x double> @llvm.experimental.vector.reverse.v2f64(<2 x double> %a)
+  %res = call <2 x double> @llvm.vector.reverse.v2f64(<2 x double> %a)
   ret <2 x double> %res
 }
 
@@ -117,7 +117,7 @@ define <2 x i8> @reverse_v2i8(<2 x i8> %a) #0 {
 ; CHECK-NEXT:    rev64 v0.2s, v0.2s
 ; CHECK-NEXT:    ret
 
-  %res = call <2 x i8> @llvm.experimental.vector.reverse.v2i8(<2 x i8> %a)
+  %res = call <2 x i8> @llvm.vector.reverse.v2i8(<2 x i8> %a)
   ret <2 x i8> %res
 }
 
@@ -144,7 +144,7 @@ define <8 x i32> @reverse_v8i32(<8 x i32> %a) #0 {
 ; CHECK-FASTISEL-NEXT:    add sp, sp, #16
 ; CHECK-FASTISEL-NEXT:    ret
 
-  %res = call <8 x i32> @llvm.experimental.vector.reverse.v8i32(<8 x i32> %a)
+  %res = call <8 x i32> @llvm.vector.reverse.v8i32(<8 x i32> %a)
   ret <8 x i32> %res
 }
 
@@ -182,23 +182,23 @@ define <16 x float> @reverse_v16f32(<16 x float> %a) #0 {
 ; CHECK-FASTISEL-NEXT:    add sp, sp, #32
 ; CHECK-FASTISEL-NEXT:    ret
 
-  %res = call <16 x float> @llvm.experimental.vector.reverse.v16f32(<16 x float> %a)
+  %res = call <16 x float> @llvm.vector.reverse.v16f32(<16 x float> %a)
   ret <16 x float> %res
 }
 
 
-declare <2 x i8> @llvm.experimental.vector.reverse.v2i8(<2 x i8>)
-declare <16 x i8> @llvm.experimental.vector.reverse.v16i8(<16 x i8>)
-declare <8 x i16> @llvm.experimental.vector.reverse.v8i16(<8 x i16>)
-declare <2 x i16> @llvm.experimental.vector.reverse.v2i16(<2 x i16>)
-declare <2 x i32> @llvm.experimental.vector.reverse.v2i32(<2 x i32>)
-declare <4 x i32> @llvm.experimental.vector.reverse.v4i32(<4 x i32>)
-declare <8 x i32> @llvm.experimental.vector.reverse.v8i32(<8 x i32>)
-declare <2 x i64> @llvm.experimental.vector.reverse.v2i64(<2 x i64>)
-declare <8 x half> @llvm.experimental.vector.reverse.v8f16(<8 x half>)
-declare <2 x float> @llvm.experimental.vector.reverse.v2f32(<2 x float>)
-declare <4 x float> @llvm.experimental.vector.reverse.v4f32(<4 x float>)
-declare <16 x float> @llvm.experimental.vector.reverse.v16f32(<16 x float>)
-declare <2 x double> @llvm.experimental.vector.reverse.v2f64(<2 x double>)
+declare <2 x i8> @llvm.vector.reverse.v2i8(<2 x i8>)
+declare <16 x i8> @llvm.vector.reverse.v16i8(<16 x i8>)
+declare <8 x i16> @llvm.vector.reverse.v8i16(<8 x i16>)
+declare <2 x i16> @llvm.vector.reverse.v2i16(<2 x i16>)
+declare <2 x i32> @llvm.vector.reverse.v2i32(<2 x i32>)
+declare <4 x i32> @llvm.vector.reverse.v4i32(<4 x i32>)
+declare <8 x i32> @llvm.vector.reverse.v8i32(<8 x i32>)
+declare <2 x i64> @llvm.vector.reverse.v2i64(<2 x i64>)
+declare <8 x half> @llvm.vector.reverse.v8f16(<8 x half>)
+declare <2 x float> @llvm.vector.reverse.v2f32(<2 x float>)
+declare <4 x float> @llvm.vector.reverse.v4f32(<4 x float>)
+declare <16 x float> @llvm.vector.reverse.v16f32(<16 x float>)
+declare <2 x double> @llvm.vector.reverse.v2f64(<2 x double>)
 
 attributes #0 = { nounwind "target-features"="+neon" }
diff --git a/llvm/test/CodeGen/AArch64/named-vector-shuffle-reverse-sve.ll b/llvm/test/CodeGen/AArch64/named-vector-shuffle-reverse-sve.ll
index 4d5045feca08..a84e6e7bcae8 100644
--- a/llvm/test/CodeGen/AArch64/named-vector-shuffle-reverse-sve.ll
+++ b/llvm/test/CodeGen/AArch64/named-vector-shuffle-reverse-sve.ll
@@ -14,7 +14,7 @@ define <vscale x 2 x i1> @reverse_nxv2i1(<vscale x 2 x i1> %a) #0 {
 ; CHECK-NEXT:    rev p0.d, p0.d
 ; CHECK-NEXT:    ret
 
-  %res = call <vscale x 2 x i1> @llvm.experimental.vector.reverse.nxv2i1(<vscale x 2 x i1> %a)
+  %res = call <vscale x 2 x i1> @llvm.vector.reverse.nxv2i1(<vscale x 2 x i1> %a)
   ret <vscale x 2 x i1> %res
 }
 
@@ -24,7 +24,7 @@ define <vscale x 4 x i1> @reverse_nxv4i1(<vscale x 4 x i1> %a) #0 {
 ; CHECK-NEXT:    rev p0.s, p0.s
 ; CHECK-NEXT:    ret
 
-  %res = call <vscale x 4 x i1> @llvm.experimental.vector.reverse.nxv4i1(<vscale x 4 x i1> %a)
+  %res = call <vscale x 4 x i1> @llvm.vector.reverse.nxv4i1(<vscale x 4 x i1> %a)
   ret <vscale x 4 x i1> %res
 }
 
@@ -34,7 +34,7 @@ define <vscale x 8 x i1> @reverse_nxv8i1(<vscale x 8 x i1> %a) #0 {
 ; CHECK-NEXT:    rev p0.h, p0.h
 ; CHECK-NEXT:    ret
 
-  %res = call <vscale x 8 x i1> @llvm.experimental.vector.reverse.nxv8i1(<vscale x 8 x i1> %a)
+  %res = call <vscale x 8 x i1> @llvm.vector.reverse.nxv8i1(<vscale x 8 x i1> %a)
   ret <vscale x 8 x i1> %res
 }
 
@@ -44,7 +44,7 @@ define <vscale x 16 x i1> @reverse_nxv16i1(<vscale x 16 x i1> %a) #0 {
 ; CHECK-NEXT:    rev p0.b, p0.b
 ; CHECK-NEXT:    ret
 
-  %res = call <vscale x 16 x i1> @llvm.experimental.vector.reverse.nxv16i1(<vscale x 16 x i1> %a)
+  %res = call <vscale x 16 x i1> @llvm.vector.reverse.nxv16i1(<vscale x 16 x i1> %a)
   ret <vscale x 16 x i1> %res
 }
 
@@ -70,7 +70,7 @@ define <vscale x 32 x i1> @reverse_nxv32i1(<vscale x 32 x i1> %a) #0 {
 ; CHECK-FASTISEL-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
 ; CHECK-FASTISEL-NEXT:    ret
 
-  %res = call <vscale x 32 x i1> @llvm.experimental.vector.reverse.nxv32i1(<vscale x 32 x i1> %a)
+  %res = call <vscale x 32 x i1> @llvm.vector.reverse.nxv32i1(<vscale x 32 x i1> %a)
   ret <vscale x 32 x i1> %res
 }
 
@@ -84,7 +84,7 @@ define <vscale x 16 x i8> @reverse_nxv16i8(<vscale x 16 x i8> %a) #0 {
 ; CHECK-NEXT:    rev z0.b, z0.b
 ; CHECK-NEXT:    ret
 
-  %res = call <vscale x 16 x i8> @llvm.experimental.vector.reverse.nxv16i8(<vscale x 16 x i8> %a)
+  %res = call <vscale x 16 x i8> @llvm.vector.reverse.nxv16i8(<vscale x 16 x i8> %a)
   ret <vscale x 16 x i8> %res
 }
 
@@ -94,7 +94,7 @@ define <vscale x 8 x i16> @reverse_nxv8i16(<vscale x 8 x i16> %a) #0 {
 ; CHECK-NEXT:    rev z0.h, z0.h
 ; CHECK-NEXT:    ret
 
-  %res = call <vscale x 8 x i16> @llvm.experimental.vector.reverse.nxv8i16(<vscale x 8 x i16> %a)
+  %res = call <vscale x 8 x i16> @llvm.vector.reverse.nxv8i16(<vscale x 8 x i16> %a)
   ret <vscale x 8 x i16> %res
 }
 
@@ -104,7 +104,7 @@ define <vscale x 4 x i32> @reverse_nxv4i32(<vscale x 4 x i32> %a) #0 {
 ; CHECK-NEXT:    rev z0.s, z0.s
 ; CHECK-NEXT:    ret
 
-  %res = call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> %a)
+  %res = call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> %a)
   ret <vscale x 4 x i32> %res
 }
 
@@ -114,7 +114,7 @@ define <vscale x 2 x i64> @reverse_nxv2i64(<vscale x 2 x i64> %a) #0 {
 ; CHECK-NEXT:    rev z0.d, z0.d
 ; CHECK-NEXT:    ret
 
-  %res = call <vscale x 2 x i64> @llvm.experimental.vector.reverse.nxv2i64(<vscale x 2 x i64> %a)
+  %res = call <vscale x 2 x i64> @llvm.vector.reverse.nxv2i64(<vscale x 2 x i64> %a)
   ret <vscale x 2 x i64> %res
 }
 
@@ -124,7 +124,7 @@ define <vscale x 2 x half> @reverse_nxv2f16(<vscale x 2 x half> %a) #0 {
 ; CHECK-NEXT:    rev z0.d, z0.d
 ; CHECK-NEXT:    ret
 
-  %res = call <vscale x 2 x half> @llvm.experimental.vector.reverse.nxv2f16(<vscale x 2 x half> %a)
+  %res = call <vscale x 2 x half> @llvm.vector.reverse.nxv2f16(<vscale x 2 x half> %a)
   ret <vscale x 2 x half> %res
 }
 
@@ -134,7 +134,7 @@ define <vscale x 4 x half> @reverse_nxv4f16(<vscale x 4 x half> %a) #0 {
 ; CHECK-NEXT:    rev z0.s, z0.s
 ; CHECK-NEXT:    ret
 
-  %res = call <vscale x 4 x half> @llvm.experimental.vector.reverse.nxv4f16(<vscale x 4 x half> %a)
+  %res = call <vscale x 4 x half> @llvm.vector.reverse.nxv4f16(<vscale x 4 x half> %a)
   ret <vscale x 4 x half> %res
 }
 
@@ -144,7 +144,7 @@ define <vscale x 8 x half> @reverse_nxv8f16(<vscale x 8 x half> %a) #0 {
 ; CHECK-NEXT:    rev z0.h, z0.h
 ; CHECK-NEXT:    ret
 
-  %res = call <vscale x 8 x half> @llvm.experimental.vector.reverse.nxv8f16(<vscale x 8 x half> %a)
+  %res = call <vscale x 8 x half> @llvm.vector.reverse.nxv8f16(<vscale x 8 x half> %a)
   ret <vscale x 8 x half> %res
 }
 
@@ -154,7 +154,7 @@ define <vscale x 2 x bfloat> @reverse_nxv2bf16(<vscale x 2 x bfloat> %a) #1 {
 ; CHECK-NEXT:    rev z0.d, z0.d
 ; CHECK-NEXT:    ret
 
-  %res = call <vscale x 2 x bfloat> @llvm.experimental.vector.reverse.nxv2bf16(<vscale x 2 x bfloat> %a)
+  %res = call <vscale x 2 x bfloat> @llvm.vector.reverse.nxv2bf16(<vscale x 2 x bfloat> %a)
   ret <vscale x 2 x bfloat> %res
 }
 
@@ -164,7 +164,7 @@ define <vscale x 4 x bfloat> @reverse_nxv4bf16(<vscale x 4 x bfloat> %a) #1 {
 ; CHECK-NEXT:    rev z0.s, z0.s
 ; CHECK-NEXT:    ret
 
-  %res = call <vscale x 4 x bfloat> @llvm.experimental.vector.reverse.nxv4bf16(<vscale x 4 x bfloat> %a)
+  %res = call <vscale x 4 x bfloat> @llvm.vector.reverse.nxv4bf16(<vscale x 4 x bfloat> %a)
   ret <vscale x 4 x bfloat> %res
 }
 
@@ -174,7 +174,7 @@ define <vscale x 8 x bfloat> @reverse_nxv8bf16(<vscale x 8 x bfloat> %a) #1 {
 ; CHECK-NEXT:    rev z0.h, z0.h
 ; CHECK-NEXT:    ret
 
-  %res = call <vscale x 8 x bfloat> @llvm.experimental.vector.reverse.nxv8bf16(<vscale x 8 x bfloat> %a)
+  %res = call <vscale x 8 x bfloat> @llvm.vector.reverse.nxv8bf16(<vscale x 8 x bfloat> %a)
   ret <vscale x 8 x bfloat> %res
 }
 
@@ -184,7 +184,7 @@ define <vscale x 2 x float> @reverse_nxv2f32(<vscale x 2 x float> %a) #0 {
 ; CHECK-NEXT:    rev z0.d, z0.d
 ; CHECK-NEXT:    ret
 
-  %res = call <vscale x 2 x float> @llvm.experimental.vector.reverse.nxv2f32(<vscale x 2 x float> %a)  ret <vscale x 2 x float> %res
+  %res = call <vscale x 2 x float> @llvm.vector.reverse.nxv2f32(<vscale x 2 x float> %a)  ret <vscale x 2 x float> %res
 }
 
 define <vscale x 4 x float> @reverse_nxv4f32(<vscale x 4 x float> %a) #0 {
@@ -193,7 +193,7 @@ define <vscale x 4 x float> @reverse_nxv4f32(<vscale x 4 x float> %a) #0 {
 ; CHECK-NEXT:    rev z0.s, z0.s
 ; CHECK-NEXT:    ret
 
-  %res = call <vscale x 4 x float> @llvm.experimental.vector.reverse.nxv4f32(<vscale x 4 x float> %a)  ret <vscale x 4 x float> %res
+  %res = call <vscale x 4 x float> @llvm.vector.reverse.nxv4f32(<vscale x 4 x float> %a)  ret <vscale x 4 x float> %res
 }
 
 define <vscale x 2 x double> @reverse_nxv2f64(<vscale x 2 x double> %a) #0 {
@@ -202,7 +202,7 @@ define <vscale x 2 x double> @reverse_nxv2f64(<vscale x 2 x double> %a) #0 {
 ; CHECK-NEXT:    rev z0.d, z0.d
 ; CHECK-NEXT:    ret
 
-  %res = call <vscale x 2 x double> @llvm.experimental.vector.reverse.nxv2f64(<vscale x 2 x double> %a)
+  %res = call <vscale x 2 x double> @llvm.vector.reverse.nxv2f64(<vscale x 2 x double> %a)
   ret <vscale x 2 x double> %res
 }
 
@@ -213,7 +213,7 @@ define <vscale x 2 x i8> @reverse_nxv2i8(<vscale x 2 x i8> %a) #0 {
 ; CHECK-NEXT:    rev z0.d, z0.d
 ; CHECK-NEXT:    ret
 
-  %res = call <vscale x 2 x i8> @llvm.experimental.vector.reverse.nxv2i8(<vscale x 2 x i8> %a)
+  %res = call <vscale x 2 x i8> @llvm.vector.reverse.nxv2i8(<vscale x 2 x i8> %a)
   ret <vscale x 2 x i8> %res
 }
 
@@ -239,7 +239,7 @@ define <vscale x 8 x i32> @reverse_nxv8i32(<vscale x 8 x i32> %a) #0 {
 ; CHECK-FASTISEL-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
 ; CHECK-FASTISEL-NEXT:    ret
 
-  %res = call <vscale x 8 x i32> @llvm.experimental.vector.reverse.nxv8i32(<vscale x 8 x i32> %a)
+  %res = call <vscale x 8 x i32> @llvm.vector.reverse.nxv8i32(<vscale x 8 x i32> %a)
   ret <vscale x 8 x i32> %res
 }
 
@@ -273,32 +273,32 @@ define <vscale x 16 x float> @reverse_nxv16f32(<vscale x 16 x float> %a) #0 {
 ; CHECK-FASTISEL-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
 ; CHECK-FASTISEL-NEXT:    ret
 
-  %res = call <vscale x 16 x float> @llvm.experimental.vector.reverse.nxv16f32(<vscale x 16 x float> %a)
+  %res = call <vscale x 16 x float> @llvm.vector.reverse.nxv16f32(<vscale x 16 x float> %a)
   ret <vscale x 16 x float> %res
 }
 
 
-declare <vscale x 2 x i1> @llvm.experimental.vector.reverse.nxv2i1(<vscale x 2 x i1>)
-declare <vscale x 4 x i1> @llvm.experimental.vector.reverse.nxv4i1(<vscale x 4 x i1>)
-declare <vscale x 8 x i1> @llvm.experimental.vector.reverse.nxv8i1(<vscale x 8 x i1>)
-declare <vscale x 16 x i1> @llvm.experimental.vector.reverse.nxv16i1(<vscale x 16 x i1>)
-declare <vscale x 32 x i1> @llvm.experimental.vector.reverse.nxv32i1(<vscale x 32 x i1>)
-declare <vscale x 2 x i8> @llvm.experimental.vector.reverse.nxv2i8(<vscale x 2 x i8>)
-declare <vscale x 16 x i8> @llvm.experimental.vector.reverse.nxv16i8(<vscale x 16 x i8>)
-declare <vscale x 8 x i16> @llvm.experimental.vector.reverse.nxv8i16(<vscale x 8 x i16>)
-declare <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32>)
-declare <vscale x 8 x i32> @llvm.experimental.vector.reverse.nxv8i32(<vscale x 8 x i32>)
-declare <vscale x 2 x i64> @llvm.experimental.vector.reverse.nxv2i64(<vscale x 2 x i64>)
-declare <vscale x 2 x half> @llvm.experimental.vector.reverse.nxv2f16(<vscale x 2 x half>)
-declare <vscale x 4 x half> @llvm.experimental.vector.reverse.nxv4f16(<vscale x 4 x half>)
-declare <vscale x 8 x half> @llvm.experimental.vector.reverse.nxv8f16(<vscale x 8 x half>)
-declare <vscale x 2 x bfloat> @llvm.experimental.vector.reverse.nxv2bf16(<vscale x 2 x bfloat>)
-declare <vscale x 4 x bfloat> @llvm.experimental.vector.reverse.nxv4bf16(<vscale x 4 x bfloat>)
-declare <vscale x 8 x bfloat> @llvm.experimental.vector.reverse.nxv8bf16(<vscale x 8 x bfloat>)
-declare <vscale x 2 x float> @llvm.experimental.vector.reverse.nxv2f32(<vscale x 2 x float>)
-declare <vscale x 4 x float> @llvm.experimental.vector.reverse.nxv4f32(<vscale x 4 x float>)
-declare <vscale x 16 x float> @llvm.experimental.vector.reverse.nxv16f32(<vscale x 16 x float>)
-declare <vscale x 2 x double> @llvm.experimental.vector.reverse.nxv2f64(<vscale x 2 x double>)
+declare <vscale x 2 x i1> @llvm.vector.reverse.nxv2i1(<vscale x 2 x i1>)
+declare <vscale x 4 x i1> @llvm.vector.reverse.nxv4i1(<vscale x 4 x i1>)
+declare <vscale x 8 x i1> @llvm.vector.reverse.nxv8i1(<vscale x 8 x i1>)
+declare <vscale x 16 x i1> @llvm.vector.reverse.nxv16i1(<vscale x 16 x i1>)
+declare <vscale x 32 x i1> @llvm.vector.reverse.nxv32i1(<vscale x 32 x i1>)
+declare <vscale x 2 x i8> @llvm.vector.reverse.nxv2i8(<vscale x 2 x i8>)
+declare <vscale x 16 x i8> @llvm.vector.reverse.nxv16i8(<vscale x 16 x i8>)
+declare <vscale x 8 x i16> @llvm.vector.reverse.nxv8i16(<vscale x 8 x i16>)
+declare <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32>)
+declare <vscale x 8 x i32> @llvm.vector.reverse.nxv8i32(<vscale x 8 x i32>)
+declare <vscale x 2 x i64> @llvm.vector.reverse.nxv2i64(<vscale x 2 x i64>)
+declare <vscale x 2 x half> @llvm.vector.reverse.nxv2f16(<vscale x 2 x half>)
+declare <vscale x 4 x half> @llvm.vector.reverse.nxv4f16(<vscale x 4 x half>)
+declare <vscale x 8 x half> @llvm.vector.reverse.nxv8f16(<vscale x 8 x half>)
+declare <vscale x 2 x bfloat> @llvm.vector.reverse.nxv2bf16(<vscale x 2 x bfloat>)
+declare <vscale x 4 x bfloat> @llvm.vector.reverse.nxv4bf16(<vscale x 4 x bfloat>)
+declare <vscale x 8 x bfloat> @llvm.vector.reverse.nxv8bf16(<vscale x 8 x bfloat>)
+declare <vscale x 2 x float> @llvm.vector.reverse.nxv2f32(<vscale x 2 x float>)
+declare <vscale x 4 x float> @llvm.vector.reverse.nxv4f32(<vscale x 4 x float>)
+declare <vscale x 16 x float> @llvm.vector.reverse.nxv16f32(<vscale x 16 x float>)
+declare <vscale x 2 x double> @llvm.vector.reverse.nxv2f64(<vscale x 2 x double>)
 
 
 attributes #0 = { nounwind "target-features"="+sve" }
diff --git a/llvm/test/CodeGen/AArch64/named-vector-shuffles-neon.ll b/llvm/test/CodeGen/AArch64/named-vector-shuffles-neon.ll
index 9210a5ec1c8b..f2e62bc4f3c8 100644
--- a/llvm/test/CodeGen/AArch64/named-vector-shuffles-neon.ll
+++ b/llvm/test/CodeGen/AArch64/named-vector-shuffles-neon.ll
@@ -12,7 +12,7 @@ define <16 x i8> @splice_v16i8_idx(<16 x i8> %a, <16 x i8> %b) #0 {
 ; CHECK:       // %bb.0:
 ; CHECK-NEXT:    ext v0.16b, v0.16b, v1.16b, #1
 ; CHECK-NEXT:    ret
-  %res = call <16 x i8> @llvm.experimental.vector.splice.v16i8(<16 x i8> %a, <16 x i8> %b, i32 1)
+  %res = call <16 x i8> @llvm.vector.splice.v16i8(<16 x i8> %a, <16 x i8> %b, i32 1)
   ret <16 x i8> %res
 }
 
@@ -21,7 +21,7 @@ define <2 x double> @splice_v2f64_idx(<2 x double> %a, <2 x double> %b) #0 {
 ; CHECK:       // %bb.0:
 ; CHECK-NEXT:    ext v0.16b, v0.16b, v1.16b, #8
 ; CHECK-NEXT:    ret
-  %res = call <2 x double> @llvm.experimental.vector.splice.v2f64(<2 x double> %a, <2 x double> %b, i32 1)
+  %res = call <2 x double> @llvm.vector.splice.v2f64(<2 x double> %a, <2 x double> %b, i32 1)
   ret <2 x double> %res
 }
 
@@ -31,7 +31,7 @@ define <2 x i8> @splice_v2i8_idx(<2 x i8> %a, <2 x i8> %b) #0 {
 ; CHECK:       // %bb.0:
 ; CHECK-NEXT:    ext v0.8b, v0.8b, v1.8b, #4
 ; CHECK-NEXT:    ret
-  %res = call <2 x i8> @llvm.experimental.vector.splice.v2i8(<2 x i8> %a, <2 x i8> %b, i32 1)
+  %res = call <2 x i8> @llvm.vector.splice.v2i8(<2 x i8> %a, <2 x i8> %b, i32 1)
   ret <2 x i8> %res
 }
 
@@ -42,7 +42,7 @@ define <8 x i32> @splice_v8i32_idx(<8 x i32> %a, <8 x i32> %b) #0 {
 ; CHECK-NEXT:    ext v0.16b, v1.16b, v2.16b, #4
 ; CHECK-NEXT:    ext v1.16b, v2.16b, v3.16b, #4
 ; CHECK-NEXT:    ret
-  %res = call <8 x i32> @llvm.experimental.vector.splice.v8i32(<8 x i32> %a, <8 x i32> %b, i32 5)
+  %res = call <8 x i32> @llvm.vector.splice.v8i32(<8 x i32> %a, <8 x i32> %b, i32 5)
   ret <8 x i32> %res
 }
 
@@ -56,7 +56,7 @@ define <16 x float> @splice_v16f32_idx(<16 x float> %a, <16 x float> %b) #0 {
 ; CHECK-NEXT:    ext v3.16b, v4.16b, v5.16b, #12
 ; CHECK-NEXT:    mov v2.16b, v6.16b
 ; CHECK-NEXT:    ret
-  %res = call <16 x float> @llvm.experimental.vector.splice.v16f32(<16 x float> %a, <16 x float> %b, i32 7)
+  %res = call <16 x float> @llvm.vector.splice.v16f32(<16 x float> %a, <16 x float> %b, i32 7)
   ret <16 x float> %res
 }
 
@@ -69,7 +69,7 @@ define <16 x i8> @splice_v16i8(<16 x i8> %a, <16 x i8> %b) #0 {
 ; CHECK:       // %bb.0:
 ; CHECK-NEXT:    ext v0.16b, v0.16b, v1.16b, #1
 ; CHECK-NEXT:    ret
-  %res = call <16 x i8> @llvm.experimental.vector.splice.v16i8(<16 x i8> %a, <16 x i8> %b, i32 -15)
+  %res = call <16 x i8> @llvm.vector.splice.v16i8(<16 x i8> %a, <16 x i8> %b, i32 -15)
   ret <16 x i8> %res
 }
 
@@ -78,7 +78,7 @@ define <2 x double> @splice_v2f64(<2 x double> %a, <2 x double> %b) #0 {
 ; CHECK:       // %bb.0:
 ; CHECK-NEXT:    ext v0.16b, v0.16b, v1.16b, #8
 ; CHECK-NEXT:    ret
-  %res = call <2 x double> @llvm.experimental.vector.splice.v2f64(<2 x double> %a, <2 x double> %b, i32 -1)
+  %res = call <2 x double> @llvm.vector.splice.v2f64(<2 x double> %a, <2 x double> %b, i32 -1)
   ret <2 x double> %res
 }
 
@@ -88,7 +88,7 @@ define <2 x i8> @splice_v2i8(<2 x i8> %a, <2 x i8> %b) #0 {
 ; CHECK:       // %bb.0:
 ; CHECK-NEXT:    ext v0.8b, v0.8b, v1.8b, #4
 ; CHECK-NEXT:    ret
-  %res = call <2 x i8> @llvm.experimental.vector.splice.v2i8(<2 x i8> %a, <2 x i8> %b, i32 -1)
+  %res = call <2 x i8> @llvm.vector.splice.v2i8(<2 x i8> %a, <2 x i8> %b, i32 -1)
   ret <2 x i8> %res
 }
 
@@ -99,7 +99,7 @@ define <8 x i32> @splice_v8i32(<8 x i32> %a, <8 x i32> %b) #0 {
 ; CHECK-NEXT:    ext v0.16b, v1.16b, v2.16b, #4
 ; CHECK-NEXT:    ext v1.16b, v2.16b, v3.16b, #4
 ; CHECK-NEXT:    ret
-  %res = call <8 x i32> @llvm.experimental.vector.splice.v8i32(<8 x i32> %a, <8 x i32> %b, i32 -3)
+  %res = call <8 x i32> @llvm.vector.splice.v8i32(<8 x i32> %a, <8 x i32> %b, i32 -3)
   ret <8 x i32> %res
 }
 
@@ -113,14 +113,14 @@ define <16 x float> @splice_v16f32(<16 x float> %a, <16 x float> %b) #0 {
 ; CHECK-NEXT:    ext v3.16b, v4.16b, v5.16b, #12
 ; CHECK-NEXT:    mov v2.16b, v6.16b
 ; CHECK-NEXT:    ret
-  %res = call <16 x float> @llvm.experimental.vector.splice.v16f32(<16 x float> %a, <16 x float> %b, i32 -9)
+  %res = call <16 x float> @llvm.vector.splice.v16f32(<16 x float> %a, <16 x float> %b, i32 -9)
   ret <16 x float> %res
 }
 
-declare <2 x i8> @llvm.experimental.vector.splice.v2i8(<2 x i8>, <2 x i8>, i32)
-declare <16 x i8> @llvm.experimental.vector.splice.v16i8(<16 x i8>, <16 x i8>, i32)
-declare <8 x i32> @llvm.experimental.vector.splice.v8i32(<8 x i32>, <8 x i32>, i32)
-declare <16 x float> @llvm.experimental.vector.splice.v16f32(<16 x float>, <16 x float>, i32)
-declare <2 x double> @llvm.experimental.vector.splice.v2f64(<2 x double>, <2 x double>, i32)
+declare <2 x i8> @llvm.vector.splice.v2i8(<2 x i8>, <2 x i8>, i32)
+declare <16 x i8> @llvm.vector.splice.v16i8(<16 x i8>, <16 x i8>, i32)
+declare <8 x i32> @llvm.vector.splice.v8i32(<8 x i32>, <8 x i32>, i32)
+declare <16 x float> @llvm.vector.splice.v16f32(<16 x float>, <16 x float>, i32)
+declare <2 x double> @llvm.vector.splice.v2f64(<2 x double>, <2 x double>, i32)
 
 attributes #0 = { nounwind "target-features"="+neon" }
diff --git a/llvm/test/CodeGen/AArch64/named-vector-shuffles-sve.ll b/llvm/test/CodeGen/AArch64/named-vector-shuffles-sve.ll
index fac96e07de54..f5763cd61033 100644
--- a/llvm/test/CodeGen/AArch64/named-vector-shuffles-sve.ll
+++ b/llvm/test/CodeGen/AArch64/named-vector-shuffles-sve.ll
@@ -11,7 +11,7 @@ define <vscale x 16 x i8> @splice_nxv16i8_zero_idx(<vscale x 16 x i8> %a, <vscal
 ; CHECK-LABEL: splice_nxv16i8_zero_idx:
 ; CHECK:       // %bb.0:
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 16 x i8> @llvm.experimental.vector.splice.nxv16i8(<vscale x 16 x i8> %a, <vscale x 16 x i8> %b, i32 0)
+  %res = call <vscale x 16 x i8> @llvm.vector.splice.nxv16i8(<vscale x 16 x i8> %a, <vscale x 16 x i8> %b, i32 0)
   ret <vscale x 16 x i8> %res
 }
 
@@ -20,7 +20,7 @@ define <vscale x 16 x i8> @splice_nxv16i8_first_idx(<vscale x 16 x i8> %a, <vsca
 ; CHECK:       // %bb.0:
 ; CHECK-NEXT:    ext z0.b, z0.b, z1.b, #1
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 16 x i8> @llvm.experimental.vector.splice.nxv16i8(<vscale x 16 x i8> %a, <vscale x 16 x i8> %b, i32 1)
+  %res = call <vscale x 16 x i8> @llvm.vector.splice.nxv16i8(<vscale x 16 x i8> %a, <vscale x 16 x i8> %b, i32 1)
   ret <vscale x 16 x i8> %res
 }
 
@@ -29,7 +29,7 @@ define <vscale x 16 x i8> @splice_nxv16i8_last_idx(<vscale x 16 x i8> %a, <vscal
 ; CHECK:       // %bb.0:
 ; CHECK-NEXT:    ext z0.b, z0.b, z1.b, #255
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 16 x i8> @llvm.experimental.vector.splice.nxv16i8(<vscale x 16 x i8> %a, <vscale x 16 x i8> %b, i32 255)
+  %res = call <vscale x 16 x i8> @llvm.vector.splice.nxv16i8(<vscale x 16 x i8> %a, <vscale x 16 x i8> %b, i32 255)
   ret <vscale x 16 x i8> %res
 }
 
@@ -38,7 +38,7 @@ define <vscale x 8 x i16> @splice_nxv8i16_first_idx(<vscale x 8 x i16> %a, <vsca
 ; CHECK:       // %bb.0:
 ; CHECK-NEXT:    ext z0.b, z0.b, z1.b, #2
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 8 x i16> @llvm.experimental.vector.splice.nxv8i16(<vscale x 8 x i16> %a, <vscale x 8 x i16> %b, i32 1)
+  %res = call <vscale x 8 x i16> @llvm.vector.splice.nxv8i16(<vscale x 8 x i16> %a, <vscale x 8 x i16> %b, i32 1)
   ret <vscale x 8 x i16> %res
 }
 
@@ -47,7 +47,7 @@ define <vscale x 4 x i32> @splice_nxv4i32_first_idx(<vscale x 4 x i32> %a, <vsca
 ; CHECK:       // %bb.0:
 ; CHECK-NEXT:    ext z0.b, z0.b, z1.b, #4
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 4 x i32> @llvm.experimental.vector.splice.nxv4i32(<vscale x 4 x i32> %a, <vscale x 4 x i32> %b, i32 1)
+  %res = call <vscale x 4 x i32> @llvm.vector.splice.nxv4i32(<vscale x 4 x i32> %a, <vscale x 4 x i32> %b, i32 1)
   ret <vscale x 4 x i32> %res
 }
 
@@ -56,7 +56,7 @@ define <vscale x 4 x i32> @splice_nxv4i32_last_idx(<vscale x 4 x i32> %a, <vscal
 ; CHECK:       // %bb.0:
 ; CHECK-NEXT:    ext z0.b, z0.b, z1.b, #252
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 4 x i32> @llvm.experimental.vector.splice.nxv4i32(<vscale x 4 x i32> %a, <vscale x 4 x i32> %b, i32 63)
+  %res = call <vscale x 4 x i32> @llvm.vector.splice.nxv4i32(<vscale x 4 x i32> %a, <vscale x 4 x i32> %b, i32 63)
   ret <vscale x 4 x i32> %res
 }
 
@@ -65,7 +65,7 @@ define <vscale x 2 x i64> @splice_nxv2i64_first_idx(<vscale x 2 x i64> %a, <vsca
 ; CHECK:       // %bb.0:
 ; CHECK-NEXT:    ext z0.b, z0.b, z1.b, #8
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 2 x i64> @llvm.experimental.vector.splice.nxv2i64(<vscale x 2 x i64> %a, <vscale x 2 x i64> %b, i32 1)
+  %res = call <vscale x 2 x i64> @llvm.vector.splice.nxv2i64(<vscale x 2 x i64> %a, <vscale x 2 x i64> %b, i32 1)
   ret <vscale x 2 x i64> %res
 }
 
@@ -74,7 +74,7 @@ define <vscale x 2 x i64> @splice_nxv2i64_last_idx(<vscale x 2 x i64> %a, <vscal
 ; CHECK:       // %bb.0:
 ; CHECK-NEXT:    ext z0.b, z0.b, z1.b, #248
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 2 x i64> @llvm.experimental.vector.splice.nxv2i64(<vscale x 2 x i64> %a, <vscale x 2 x i64> %b, i32 31)
+  %res = call <vscale x 2 x i64> @llvm.vector.splice.nxv2i64(<vscale x 2 x i64> %a, <vscale x 2 x i64> %b, i32 31)
   ret <vscale x 2 x i64> %res
 }
 
@@ -85,7 +85,7 @@ define <vscale x 2 x half> @splice_nxv2f16_neg_idx(<vscale x 2 x half> %a, <vsca
 ; CHECK-NEXT:    rev p0.d, p0.d
 ; CHECK-NEXT:    splice z0.d, p0, z0.d, z1.d
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 2 x half> @llvm.experimental.vector.splice.nxv2f16(<vscale x 2 x half> %a, <vscale x 2 x half> %b, i32 -1)
+  %res = call <vscale x 2 x half> @llvm.vector.splice.nxv2f16(<vscale x 2 x half> %a, <vscale x 2 x half> %b, i32 -1)
   ret <vscale x 2 x half> %res
 }
 
@@ -96,7 +96,7 @@ define <vscale x 2 x half> @splice_nxv2f16_neg2_idx(<vscale x 2 x half> %a, <vsc
 ; CHECK-NEXT:    rev p0.d, p0.d
 ; CHECK-NEXT:    splice z0.d, p0, z0.d, z1.d
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 2 x half> @llvm.experimental.vector.splice.nxv2f16(<vscale x 2 x half> %a, <vscale x 2 x half> %b, i32 -2)
+  %res = call <vscale x 2 x half> @llvm.vector.splice.nxv2f16(<vscale x 2 x half> %a, <vscale x 2 x half> %b, i32 -2)
   ret <vscale x 2 x half> %res
 }
 
@@ -105,7 +105,7 @@ define <vscale x 2 x half> @splice_nxv2f16_first_idx(<vscale x 2 x half> %a, <vs
 ; CHECK:       // %bb.0:
 ; CHECK-NEXT:    ext z0.b, z0.b, z1.b, #8
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 2 x half> @llvm.experimental.vector.splice.nxv2f16(<vscale x 2 x half> %a, <vscale x 2 x half> %b, i32 1)
+  %res = call <vscale x 2 x half> @llvm.vector.splice.nxv2f16(<vscale x 2 x half> %a, <vscale x 2 x half> %b, i32 1)
   ret <vscale x 2 x half> %res
 }
 
@@ -114,7 +114,7 @@ define <vscale x 2 x half> @splice_nxv2f16_last_idx(<vscale x 2 x half> %a, <vsc
 ; CHECK:       // %bb.0:
 ; CHECK-NEXT:    ext z0.b, z0.b, z1.b, #248
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 2 x half> @llvm.experimental.vector.splice.nxv2f16(<vscale x 2 x half> %a, <vscale x 2 x half> %b, i32 31)
+  %res = call <vscale x 2 x half> @llvm.vector.splice.nxv2f16(<vscale x 2 x half> %a, <vscale x 2 x half> %b, i32 31)
   ret <vscale x 2 x half> %res
 }
 
@@ -125,7 +125,7 @@ define <vscale x 4 x half> @splice_nxv4f16_neg_idx(<vscale x 4 x half> %a, <vsca
 ; CHECK-NEXT:    rev p0.s, p0.s
 ; CHECK-NEXT:    splice z0.s, p0, z0.s, z1.s
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 4 x half> @llvm.experimental.vector.splice.nxv4f16(<vscale x 4 x half> %a, <vscale x 4 x half> %b, i32 -1)
+  %res = call <vscale x 4 x half> @llvm.vector.splice.nxv4f16(<vscale x 4 x half> %a, <vscale x 4 x half> %b, i32 -1)
   ret <vscale x 4 x half> %res
 }
 
@@ -136,7 +136,7 @@ define <vscale x 4 x half> @splice_nxv4f16_neg3_idx(<vscale x 4 x half> %a, <vsc
 ; CHECK-NEXT:    rev p0.s, p0.s
 ; CHECK-NEXT:    splice z0.s, p0, z0.s, z1.s
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 4 x half> @llvm.experimental.vector.splice.nxv4f16(<vscale x 4 x half> %a, <vscale x 4 x half> %b, i32 -3)
+  %res = call <vscale x 4 x half> @llvm.vector.splice.nxv4f16(<vscale x 4 x half> %a, <vscale x 4 x half> %b, i32 -3)
   ret <vscale x 4 x half> %res
 }
 
@@ -145,7 +145,7 @@ define <vscale x 4 x half> @splice_nxv4f16_first_idx(<vscale x 4 x half> %a, <vs
 ; CHECK:       // %bb.0:
 ; CHECK-NEXT:    ext z0.b, z0.b, z1.b, #4
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 4 x half> @llvm.experimental.vector.splice.nxv4f16(<vscale x 4 x half> %a, <vscale x 4 x half> %b, i32 1)
+  %res = call <vscale x 4 x half> @llvm.vector.splice.nxv4f16(<vscale x 4 x half> %a, <vscale x 4 x half> %b, i32 1)
   ret <vscale x 4 x half> %res
 }
 
@@ -154,7 +154,7 @@ define <vscale x 4 x half> @splice_nxv4f16_last_idx(<vscale x 4 x half> %a, <vsc
 ; CHECK:       // %bb.0:
 ; CHECK-NEXT:    ext z0.b, z0.b, z1.b, #252
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 4 x half> @llvm.experimental.vector.splice.nxv4f16(<vscale x 4 x half> %a, <vscale x 4 x half> %b, i32 63)
+  %res = call <vscale x 4 x half> @llvm.vector.splice.nxv4f16(<vscale x 4 x half> %a, <vscale x 4 x half> %b, i32 63)
   ret <vscale x 4 x half> %res
 }
 
@@ -163,7 +163,7 @@ define <vscale x 8 x half> @splice_nxv8f16_first_idx(<vscale x 8 x half> %a, <vs
 ; CHECK:       // %bb.0:
 ; CHECK-NEXT:    ext z0.b, z0.b, z1.b, #2
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 8 x half> @llvm.experimental.vector.splice.nxv8f16(<vscale x 8 x half> %a, <vscale x 8 x half> %b, i32 1)
+  %res = call <vscale x 8 x half> @llvm.vector.splice.nxv8f16(<vscale x 8 x half> %a, <vscale x 8 x half> %b, i32 1)
   ret <vscale x 8 x half> %res
 }
 
@@ -172,7 +172,7 @@ define <vscale x 8 x half> @splice_nxv8f16_last_idx(<vscale x 8 x half> %a, <vsc
 ; CHECK:       // %bb.0:
 ; CHECK-NEXT:    ext z0.b, z0.b, z1.b, #254
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 8 x half> @llvm.experimental.vector.splice.nxv8f16(<vscale x 8 x half> %a, <vscale x 8 x half> %b, i32 127)
+  %res = call <vscale x 8 x half> @llvm.vector.splice.nxv8f16(<vscale x 8 x half> %a, <vscale x 8 x half> %b, i32 127)
   ret <vscale x 8 x half> %res
 }
 
@@ -183,7 +183,7 @@ define <vscale x 2 x float> @splice_nxv2f32_neg_idx(<vscale x 2 x float> %a, <vs
 ; CHECK-NEXT:    rev p0.d, p0.d
 ; CHECK-NEXT:    splice z0.d, p0, z0.d, z1.d
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 2 x float> @llvm.experimental.vector.splice.nxv2f32(<vscale x 2 x float> %a, <vscale x 2 x float> %b, i32 -1)
+  %res = call <vscale x 2 x float> @llvm.vector.splice.nxv2f32(<vscale x 2 x float> %a, <vscale x 2 x float> %b, i32 -1)
   ret <vscale x 2 x float> %res
 }
 
@@ -194,7 +194,7 @@ define <vscale x 2 x float> @splice_nxv2f32_neg2_idx(<vscale x 2 x float> %a, <v
 ; CHECK-NEXT:    rev p0.d, p0.d
 ; CHECK-NEXT:    splice z0.d, p0, z0.d, z1.d
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 2 x float> @llvm.experimental.vector.splice.nxv2f32(<vscale x 2 x float> %a, <vscale x 2 x float> %b, i32 -2)
+  %res = call <vscale x 2 x float> @llvm.vector.splice.nxv2f32(<vscale x 2 x float> %a, <vscale x 2 x float> %b, i32 -2)
   ret <vscale x 2 x float> %res
 }
 
@@ -203,7 +203,7 @@ define <vscale x 2 x float> @splice_nxv2f32_first_idx(<vscale x 2 x float> %a, <
 ; CHECK:       // %bb.0:
 ; CHECK-NEXT:    ext z0.b, z0.b, z1.b, #8
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 2 x float> @llvm.experimental.vector.splice.nxv2f32(<vscale x 2 x float> %a, <vscale x 2 x float> %b, i32 1)
+  %res = call <vscale x 2 x float> @llvm.vector.splice.nxv2f32(<vscale x 2 x float> %a, <vscale x 2 x float> %b, i32 1)
   ret <vscale x 2 x float> %res
 }
 
@@ -212,7 +212,7 @@ define <vscale x 2 x float> @splice_nxv2f32_last_idx(<vscale x 2 x float> %a, <v
 ; CHECK:       // %bb.0:
 ; CHECK-NEXT:    ext z0.b, z0.b, z1.b, #248
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 2 x float> @llvm.experimental.vector.splice.nxv2f32(<vscale x 2 x float> %a, <vscale x 2 x float> %b, i32 31)
+  %res = call <vscale x 2 x float> @llvm.vector.splice.nxv2f32(<vscale x 2 x float> %a, <vscale x 2 x float> %b, i32 31)
   ret <vscale x 2 x float> %res
 }
 
@@ -221,7 +221,7 @@ define <vscale x 4 x float> @splice_nxv4f32_first_idx(<vscale x 4 x float> %a, <
 ; CHECK:       // %bb.0:
 ; CHECK-NEXT:    ext z0.b, z0.b, z1.b, #4
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 4 x float> @llvm.experimental.vector.splice.nxv4f32(<vscale x 4 x float> %a, <vscale x 4 x float> %b, i32 1)
+  %res = call <vscale x 4 x float> @llvm.vector.splice.nxv4f32(<vscale x 4 x float> %a, <vscale x 4 x float> %b, i32 1)
   ret <vscale x 4 x float> %res
 }
 
@@ -230,7 +230,7 @@ define <vscale x 4 x float> @splice_nxv4f32_last_idx(<vscale x 4 x float> %a, <v
 ; CHECK:       // %bb.0:
 ; CHECK-NEXT:    ext z0.b, z0.b, z1.b, #252
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 4 x float> @llvm.experimental.vector.splice.nxv4f32(<vscale x 4 x float> %a, <vscale x 4 x float> %b, i32 63)
+  %res = call <vscale x 4 x float> @llvm.vector.splice.nxv4f32(<vscale x 4 x float> %a, <vscale x 4 x float> %b, i32 63)
   ret <vscale x 4 x float> %res
 }
 
@@ -239,7 +239,7 @@ define <vscale x 2 x double> @splice_nxv2f64_first_idx(<vscale x 2 x double> %a,
 ; CHECK:       // %bb.0:
 ; CHECK-NEXT:    ext z0.b, z0.b, z1.b, #8
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 2 x double> @llvm.experimental.vector.splice.nxv2f64(<vscale x 2 x double> %a, <vscale x 2 x double> %b, i32 1)
+  %res = call <vscale x 2 x double> @llvm.vector.splice.nxv2f64(<vscale x 2 x double> %a, <vscale x 2 x double> %b, i32 1)
   ret <vscale x 2 x double> %res
 }
 
@@ -248,7 +248,7 @@ define <vscale x 2 x double> @splice_nxv2f64_last_idx(<vscale x 2 x double> %a,
 ; CHECK:       // %bb.0:
 ; CHECK-NEXT:    ext z0.b, z0.b, z1.b, #248
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 2 x double> @llvm.experimental.vector.splice.nxv2f64(<vscale x 2 x double> %a, <vscale x 2 x double> %b, i32 31)
+  %res = call <vscale x 2 x double> @llvm.vector.splice.nxv2f64(<vscale x 2 x double> %a, <vscale x 2 x double> %b, i32 31)
   ret <vscale x 2 x double> %res
 }
 
@@ -263,7 +263,7 @@ define <vscale x 2 x i1> @splice_nxv2i1_idx(<vscale x 2 x i1> %a, <vscale x 2 x
 ; CHECK-NEXT:    and z1.d, z1.d, #0x1
 ; CHECK-NEXT:    cmpne p0.d, p0/z, z1.d, #0
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 2 x i1> @llvm.experimental.vector.splice.nxv2i1(<vscale x 2 x i1> %a, <vscale x 2 x i1> %b, i32 1)
+  %res = call <vscale x 2 x i1> @llvm.vector.splice.nxv2i1(<vscale x 2 x i1> %a, <vscale x 2 x i1> %b, i32 1)
   ret <vscale x 2 x i1> %res
 }
 
@@ -278,7 +278,7 @@ define <vscale x 4 x i1> @splice_nxv4i1_idx(<vscale x 4 x i1> %a, <vscale x 4 x
 ; CHECK-NEXT:    and z1.s, z1.s, #0x1
 ; CHECK-NEXT:    cmpne p0.s, p0/z, z1.s, #0
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 4 x i1> @llvm.experimental.vector.splice.nxv4i1(<vscale x 4 x i1> %a, <vscale x 4 x i1> %b, i32 2)
+  %res = call <vscale x 4 x i1> @llvm.vector.splice.nxv4i1(<vscale x 4 x i1> %a, <vscale x 4 x i1> %b, i32 2)
   ret <vscale x 4 x i1> %res
 }
 
@@ -293,7 +293,7 @@ define <vscale x 8 x i1> @splice_nxv8i1_idx(<vscale x 8 x i1> %a, <vscale x 8 x
 ; CHECK-NEXT:    and z1.h, z1.h, #0x1
 ; CHECK-NEXT:    cmpne p0.h, p0/z, z1.h, #0
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 8 x i1> @llvm.experimental.vector.splice.nxv8i1(<vscale x 8 x i1> %a, <vscale x 8 x i1> %b, i32 4)
+  %res = call <vscale x 8 x i1> @llvm.vector.splice.nxv8i1(<vscale x 8 x i1> %a, <vscale x 8 x i1> %b, i32 4)
   ret <vscale x 8 x i1> %res
 }
 
@@ -308,7 +308,7 @@ define <vscale x 16 x i1> @splice_nxv16i1_idx(<vscale x 16 x i1> %a, <vscale x 1
 ; CHECK-NEXT:    and z1.b, z1.b, #0x1
 ; CHECK-NEXT:    cmpne p0.b, p0/z, z1.b, #0
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 16 x i1> @llvm.experimental.vector.splice.nxv16i1(<vscale x 16 x i1> %a, <vscale x 16 x i1> %b, i32 8)
+  %res = call <vscale x 16 x i1> @llvm.vector.splice.nxv16i1(<vscale x 16 x i1> %a, <vscale x 16 x i1> %b, i32 8)
   ret <vscale x 16 x i1> %res
 }
 
@@ -318,7 +318,7 @@ define <vscale x 2 x i8> @splice_nxv2i8_idx(<vscale x 2 x i8> %a, <vscale x 2 x
 ; CHECK:       // %bb.0:
 ; CHECK-NEXT:    ext z0.b, z0.b, z1.b, #8
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 2 x i8> @llvm.experimental.vector.splice.nxv2i8(<vscale x 2 x i8> %a, <vscale x 2 x i8> %b, i32 1)
+  %res = call <vscale x 2 x i8> @llvm.vector.splice.nxv2i8(<vscale x 2 x i8> %a, <vscale x 2 x i8> %b, i32 1)
   ret <vscale x 2 x i8> %res
 }
 
@@ -340,7 +340,7 @@ define <vscale x 8 x i32> @splice_nxv8i32_idx(<vscale x 8 x i32> %a, <vscale x 8
 ; CHECK-NEXT:    addvl sp, sp, #4
 ; CHECK-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 8 x i32> @llvm.experimental.vector.splice.nxv8i32(<vscale x 8 x i32> %a, <vscale x 8 x i32> %b, i32 2)
+  %res = call <vscale x 8 x i32> @llvm.vector.splice.nxv8i32(<vscale x 8 x i32> %a, <vscale x 8 x i32> %b, i32 2)
   ret <vscale x 8 x i32> %res
 }
 
@@ -373,7 +373,7 @@ define <vscale x 16 x float> @splice_nxv16f32_16(<vscale x 16 x float> %a, <vsca
 ; CHECK-NEXT:    addvl sp, sp, #8
 ; CHECK-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 16 x float> @llvm.experimental.vector.splice.nxv16f32(<vscale x 16 x float> %a, <vscale x 16 x float> %b, i32 16)
+  %res = call <vscale x 16 x float> @llvm.vector.splice.nxv16f32(<vscale x 16 x float> %a, <vscale x 16 x float> %b, i32 16)
   ret <vscale x 16 x float> %res
 }
 
@@ -388,7 +388,7 @@ define <vscale x 16 x i8> @splice_nxv16i8(<vscale x 16 x i8> %a, <vscale x 16 x
 ; CHECK-NEXT:    rev p0.b, p0.b
 ; CHECK-NEXT:    splice z0.b, p0, z0.b, z1.b
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 16 x i8> @llvm.experimental.vector.splice.nxv16i8(<vscale x 16 x i8> %a, <vscale x 16 x i8> %b, i32 -16)
+  %res = call <vscale x 16 x i8> @llvm.vector.splice.nxv16i8(<vscale x 16 x i8> %a, <vscale x 16 x i8> %b, i32 -16)
   ret <vscale x 16 x i8> %res
 }
 
@@ -399,7 +399,7 @@ define <vscale x 16 x i8> @splice_nxv16i8_neg32(<vscale x 16 x i8> %a, <vscale x
 ; CHECK-NEXT:    rev p0.b, p0.b
 ; CHECK-NEXT:    splice z0.b, p0, z0.b, z1.b
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 16 x i8> @llvm.experimental.vector.splice.nxv16i8(<vscale x 16 x i8> %a, <vscale x 16 x i8> %b, i32 -32)
+  %res = call <vscale x 16 x i8> @llvm.vector.splice.nxv16i8(<vscale x 16 x i8> %a, <vscale x 16 x i8> %b, i32 -32)
   ret <vscale x 16 x i8> %res
 }
 
@@ -410,7 +410,7 @@ define <vscale x 16 x i8> @splice_nxv16i8_neg64(<vscale x 16 x i8> %a, <vscale x
 ; CHECK-NEXT:    rev p0.b, p0.b
 ; CHECK-NEXT:    splice z0.b, p0, z0.b, z1.b
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 16 x i8> @llvm.experimental.vector.splice.nxv16i8(<vscale x 16 x i8> %a, <vscale x 16 x i8> %b, i32 -64)
+  %res = call <vscale x 16 x i8> @llvm.vector.splice.nxv16i8(<vscale x 16 x i8> %a, <vscale x 16 x i8> %b, i32 -64)
   ret <vscale x 16 x i8> %res
 }
 
@@ -421,7 +421,7 @@ define <vscale x 16 x i8> @splice_nxv16i8_neg128(<vscale x 16 x i8> %a, <vscale
 ; CHECK-NEXT:    rev p0.b, p0.b
 ; CHECK-NEXT:    splice z0.b, p0, z0.b, z1.b
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 16 x i8> @llvm.experimental.vector.splice.nxv16i8(<vscale x 16 x i8> %a, <vscale x 16 x i8> %b, i32 -128)
+  %res = call <vscale x 16 x i8> @llvm.vector.splice.nxv16i8(<vscale x 16 x i8> %a, <vscale x 16 x i8> %b, i32 -128)
   ret <vscale x 16 x i8> %res
 }
 
@@ -432,7 +432,7 @@ define <vscale x 16 x i8> @splice_nxv16i8_neg256(<vscale x 16 x i8> %a, <vscale
 ; CHECK-NEXT:    rev p0.b, p0.b
 ; CHECK-NEXT:    splice z0.b, p0, z0.b, z1.b
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 16 x i8> @llvm.experimental.vector.splice.nxv16i8(<vscale x 16 x i8> %a, <vscale x 16 x i8> %b, i32 -256)
+  %res = call <vscale x 16 x i8> @llvm.vector.splice.nxv16i8(<vscale x 16 x i8> %a, <vscale x 16 x i8> %b, i32 -256)
   ret <vscale x 16 x i8> %res
 }
 
@@ -443,7 +443,7 @@ define <vscale x 16 x i8> @splice_nxv16i8_1(<vscale x 16 x i8> %a, <vscale x 16
 ; CHECK-NEXT:    rev p0.b, p0.b
 ; CHECK-NEXT:    splice z0.b, p0, z0.b, z1.b
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 16 x i8> @llvm.experimental.vector.splice.nxv16i8(<vscale x 16 x i8> %a, <vscale x 16 x i8> %b, i32 -1)
+  %res = call <vscale x 16 x i8> @llvm.vector.splice.nxv16i8(<vscale x 16 x i8> %a, <vscale x 16 x i8> %b, i32 -1)
   ret <vscale x 16 x i8> %res
 }
 
@@ -466,7 +466,7 @@ define <vscale x 16 x i8> @splice_nxv16i8_neg17(<vscale x 16 x i8> %a, <vscale x
 ; CHECK-NEXT:    addvl sp, sp, #2
 ; CHECK-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 16 x i8> @llvm.experimental.vector.splice.nxv16i8(<vscale x 16 x i8> %a, <vscale x 16 x i8> %b, i32 -17)
+  %res = call <vscale x 16 x i8> @llvm.vector.splice.nxv16i8(<vscale x 16 x i8> %a, <vscale x 16 x i8> %b, i32 -17)
   ret <vscale x 16 x i8> %res
 }
 
@@ -477,7 +477,7 @@ define <vscale x 8 x i16> @splice_nxv8i16(<vscale x 8 x i16> %a, <vscale x 8 x i
 ; CHECK-NEXT:    rev p0.h, p0.h
 ; CHECK-NEXT:    splice z0.h, p0, z0.h, z1.h
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 8 x i16> @llvm.experimental.vector.splice.nxv8i16(<vscale x 8 x i16> %a, <vscale x 8 x i16> %b, i32 -8)
+  %res = call <vscale x 8 x i16> @llvm.vector.splice.nxv8i16(<vscale x 8 x i16> %a, <vscale x 8 x i16> %b, i32 -8)
   ret <vscale x 8 x i16> %res
 }
 
@@ -488,7 +488,7 @@ define <vscale x 8 x i16> @splice_nxv8i16_1(<vscale x 8 x i16> %a, <vscale x 8 x
 ; CHECK-NEXT:    rev p0.h, p0.h
 ; CHECK-NEXT:    splice z0.h, p0, z0.h, z1.h
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 8 x i16> @llvm.experimental.vector.splice.nxv8i16(<vscale x 8 x i16> %a, <vscale x 8 x i16> %b, i32 -1)
+  %res = call <vscale x 8 x i16> @llvm.vector.splice.nxv8i16(<vscale x 8 x i16> %a, <vscale x 8 x i16> %b, i32 -1)
   ret <vscale x 8 x i16> %res
 }
 
@@ -511,7 +511,7 @@ define <vscale x 8 x i16> @splice_nxv8i16_neg9(<vscale x 8 x i16> %a, <vscale x
 ; CHECK-NEXT:    addvl sp, sp, #2
 ; CHECK-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 8 x i16> @llvm.experimental.vector.splice.nxv8i16(<vscale x 8 x i16> %a, <vscale x 8 x i16> %b, i32 -9)
+  %res = call <vscale x 8 x i16> @llvm.vector.splice.nxv8i16(<vscale x 8 x i16> %a, <vscale x 8 x i16> %b, i32 -9)
   ret <vscale x 8 x i16> %res
 }
 
@@ -522,7 +522,7 @@ define <vscale x 4 x i32> @splice_nxv4i32(<vscale x 4 x i32> %a, <vscale x 4 x i
 ; CHECK-NEXT:    rev p0.s, p0.s
 ; CHECK-NEXT:    splice z0.s, p0, z0.s, z1.s
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 4 x i32> @llvm.experimental.vector.splice.nxv4i32(<vscale x 4 x i32> %a, <vscale x 4 x i32> %b, i32 -4)
+  %res = call <vscale x 4 x i32> @llvm.vector.splice.nxv4i32(<vscale x 4 x i32> %a, <vscale x 4 x i32> %b, i32 -4)
   ret <vscale x 4 x i32> %res
 }
 
@@ -533,7 +533,7 @@ define <vscale x 4 x i32> @splice_nxv4i32_1(<vscale x 4 x i32> %a, <vscale x 4 x
 ; CHECK-NEXT:    rev p0.s, p0.s
 ; CHECK-NEXT:    splice z0.s, p0, z0.s, z1.s
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 4 x i32> @llvm.experimental.vector.splice.nxv4i32(<vscale x 4 x i32> %a, <vscale x 4 x i32> %b, i32 -1)
+  %res = call <vscale x 4 x i32> @llvm.vector.splice.nxv4i32(<vscale x 4 x i32> %a, <vscale x 4 x i32> %b, i32 -1)
   ret <vscale x 4 x i32> %res
 }
 
@@ -544,7 +544,7 @@ define <vscale x 4 x i32> @splice_nxv4i32_neg5(<vscale x 4 x i32> %a, <vscale x
 ; CHECK-NEXT:    rev p0.s, p0.s
 ; CHECK-NEXT:    splice z0.s, p0, z0.s, z1.s
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 4 x i32> @llvm.experimental.vector.splice.nxv4i32(<vscale x 4 x i32> %a, <vscale x 4 x i32> %b, i32 -5)
+  %res = call <vscale x 4 x i32> @llvm.vector.splice.nxv4i32(<vscale x 4 x i32> %a, <vscale x 4 x i32> %b, i32 -5)
   ret <vscale x 4 x i32> %res
 }
 
@@ -555,7 +555,7 @@ define <vscale x 2 x i64> @splice_nxv2i64(<vscale x 2 x i64> %a, <vscale x 2 x i
 ; CHECK-NEXT:    rev p0.d, p0.d
 ; CHECK-NEXT:    splice z0.d, p0, z0.d, z1.d
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 2 x i64> @llvm.experimental.vector.splice.nxv2i64(<vscale x 2 x i64> %a, <vscale x 2 x i64> %b, i32 -2)
+  %res = call <vscale x 2 x i64> @llvm.vector.splice.nxv2i64(<vscale x 2 x i64> %a, <vscale x 2 x i64> %b, i32 -2)
   ret <vscale x 2 x i64> %res
 }
 
@@ -566,7 +566,7 @@ define <vscale x 2 x i64> @splice_nxv2i64_1(<vscale x 2 x i64> %a, <vscale x 2 x
 ; CHECK-NEXT:    rev p0.d, p0.d
 ; CHECK-NEXT:    splice z0.d, p0, z0.d, z1.d
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 2 x i64> @llvm.experimental.vector.splice.nxv2i64(<vscale x 2 x i64> %a, <vscale x 2 x i64> %b, i32 -1)
+  %res = call <vscale x 2 x i64> @llvm.vector.splice.nxv2i64(<vscale x 2 x i64> %a, <vscale x 2 x i64> %b, i32 -1)
   ret <vscale x 2 x i64> %res
 }
 
@@ -577,7 +577,7 @@ define <vscale x 2 x i64> @splice_nxv2i64_neg3(<vscale x 2 x i64> %a, <vscale x
 ; CHECK-NEXT:    rev p0.d, p0.d
 ; CHECK-NEXT:    splice z0.d, p0, z0.d, z1.d
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 2 x i64> @llvm.experimental.vector.splice.nxv2i64(<vscale x 2 x i64> %a, <vscale x 2 x i64> %b, i32 -3)
+  %res = call <vscale x 2 x i64> @llvm.vector.splice.nxv2i64(<vscale x 2 x i64> %a, <vscale x 2 x i64> %b, i32 -3)
   ret <vscale x 2 x i64> %res
 }
 
@@ -588,7 +588,7 @@ define <vscale x 8 x half> @splice_nxv8f16(<vscale x 8 x half> %a, <vscale x 8 x
 ; CHECK-NEXT:    rev p0.h, p0.h
 ; CHECK-NEXT:    splice z0.h, p0, z0.h, z1.h
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 8 x half> @llvm.experimental.vector.splice.nxv8f16(<vscale x 8 x half> %a, <vscale x 8 x half> %b, i32 -8)
+  %res = call <vscale x 8 x half> @llvm.vector.splice.nxv8f16(<vscale x 8 x half> %a, <vscale x 8 x half> %b, i32 -8)
   ret <vscale x 8 x half> %res
 }
 
@@ -599,7 +599,7 @@ define <vscale x 8 x half> @splice_nxv8f16_1(<vscale x 8 x half> %a, <vscale x 8
 ; CHECK-NEXT:    rev p0.h, p0.h
 ; CHECK-NEXT:    splice z0.h, p0, z0.h, z1.h
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 8 x half> @llvm.experimental.vector.splice.nxv8f16(<vscale x 8 x half> %a, <vscale x 8 x half> %b, i32 -1)
+  %res = call <vscale x 8 x half> @llvm.vector.splice.nxv8f16(<vscale x 8 x half> %a, <vscale x 8 x half> %b, i32 -1)
   ret <vscale x 8 x half> %res
 }
 
@@ -622,7 +622,7 @@ define <vscale x 8 x half> @splice_nxv8f16_neg9(<vscale x 8 x half> %a, <vscale
 ; CHECK-NEXT:    addvl sp, sp, #2
 ; CHECK-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 8 x half> @llvm.experimental.vector.splice.nxv8f16(<vscale x 8 x half> %a, <vscale x 8 x half> %b, i32 -9)
+  %res = call <vscale x 8 x half> @llvm.vector.splice.nxv8f16(<vscale x 8 x half> %a, <vscale x 8 x half> %b, i32 -9)
   ret <vscale x 8 x half> %res
 }
 
@@ -633,7 +633,7 @@ define <vscale x 4 x float> @splice_nxv4f32(<vscale x 4 x float> %a, <vscale x 4
 ; CHECK-NEXT:    rev p0.s, p0.s
 ; CHECK-NEXT:    splice z0.s, p0, z0.s, z1.s
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 4 x float> @llvm.experimental.vector.splice.nxv4f32(<vscale x 4 x float> %a, <vscale x 4 x float> %b, i32 -4)
+  %res = call <vscale x 4 x float> @llvm.vector.splice.nxv4f32(<vscale x 4 x float> %a, <vscale x 4 x float> %b, i32 -4)
   ret <vscale x 4 x float> %res
 }
 
@@ -644,7 +644,7 @@ define <vscale x 4 x float> @splice_nxv4f32_1(<vscale x 4 x float> %a, <vscale x
 ; CHECK-NEXT:    rev p0.s, p0.s
 ; CHECK-NEXT:    splice z0.s, p0, z0.s, z1.s
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 4 x float> @llvm.experimental.vector.splice.nxv4f32(<vscale x 4 x float> %a, <vscale x 4 x float> %b, i32 -1)
+  %res = call <vscale x 4 x float> @llvm.vector.splice.nxv4f32(<vscale x 4 x float> %a, <vscale x 4 x float> %b, i32 -1)
   ret <vscale x 4 x float> %res
 }
 
@@ -655,7 +655,7 @@ define <vscale x 4 x float> @splice_nxv4f32_neg5(<vscale x 4 x float> %a, <vscal
 ; CHECK-NEXT:    rev p0.s, p0.s
 ; CHECK-NEXT:    splice z0.s, p0, z0.s, z1.s
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 4 x float> @llvm.experimental.vector.splice.nxv4f32(<vscale x 4 x float> %a, <vscale x 4 x float> %b, i32 -5)
+  %res = call <vscale x 4 x float> @llvm.vector.splice.nxv4f32(<vscale x 4 x float> %a, <vscale x 4 x float> %b, i32 -5)
   ret <vscale x 4 x float> %res
 }
 
@@ -666,7 +666,7 @@ define <vscale x 2 x double> @splice_nxv2f64(<vscale x 2 x double> %a, <vscale x
 ; CHECK-NEXT:    rev p0.d, p0.d
 ; CHECK-NEXT:    splice z0.d, p0, z0.d, z1.d
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 2 x double> @llvm.experimental.vector.splice.nxv2f64(<vscale x 2 x double> %a, <vscale x 2 x double> %b, i32 -2)
+  %res = call <vscale x 2 x double> @llvm.vector.splice.nxv2f64(<vscale x 2 x double> %a, <vscale x 2 x double> %b, i32 -2)
   ret <vscale x 2 x double> %res
 }
 
@@ -677,7 +677,7 @@ define <vscale x 2 x double> @splice_nxv2f64_1(<vscale x 2 x double> %a, <vscale
 ; CHECK-NEXT:    rev p0.d, p0.d
 ; CHECK-NEXT:    splice z0.d, p0, z0.d, z1.d
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 2 x double> @llvm.experimental.vector.splice.nxv2f64(<vscale x 2 x double> %a, <vscale x 2 x double> %b, i32 -1)
+  %res = call <vscale x 2 x double> @llvm.vector.splice.nxv2f64(<vscale x 2 x double> %a, <vscale x 2 x double> %b, i32 -1)
   ret <vscale x 2 x double> %res
 }
 
@@ -688,7 +688,7 @@ define <vscale x 2 x double> @splice_nxv2f64_neg3(<vscale x 2 x double> %a, <vsc
 ; CHECK-NEXT:    rev p0.d, p0.d
 ; CHECK-NEXT:    splice z0.d, p0, z0.d, z1.d
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 2 x double> @llvm.experimental.vector.splice.nxv2f64(<vscale x 2 x double> %a, <vscale x 2 x double> %b, i32 -3)
+  %res = call <vscale x 2 x double> @llvm.vector.splice.nxv2f64(<vscale x 2 x double> %a, <vscale x 2 x double> %b, i32 -3)
   ret <vscale x 2 x double> %res
 }
 
@@ -705,7 +705,7 @@ define <vscale x 2 x i1> @splice_nxv2i1(<vscale x 2 x i1> %a, <vscale x 2 x i1>
 ; CHECK-NEXT:    and z1.d, z1.d, #0x1
 ; CHECK-NEXT:    cmpne p0.d, p0/z, z1.d, #0
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 2 x i1> @llvm.experimental.vector.splice.nxv2i1(<vscale x 2 x i1> %a, <vscale x 2 x i1> %b, i32 -1)
+  %res = call <vscale x 2 x i1> @llvm.vector.splice.nxv2i1(<vscale x 2 x i1> %a, <vscale x 2 x i1> %b, i32 -1)
   ret <vscale x 2 x i1> %res
 }
 
@@ -722,7 +722,7 @@ define <vscale x 4 x i1> @splice_nxv4i1(<vscale x 4 x i1> %a, <vscale x 4 x i1>
 ; CHECK-NEXT:    and z1.s, z1.s, #0x1
 ; CHECK-NEXT:    cmpne p0.s, p0/z, z1.s, #0
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 4 x i1> @llvm.experimental.vector.splice.nxv4i1(<vscale x 4 x i1> %a, <vscale x 4 x i1> %b, i32 -1)
+  %res = call <vscale x 4 x i1> @llvm.vector.splice.nxv4i1(<vscale x 4 x i1> %a, <vscale x 4 x i1> %b, i32 -1)
   ret <vscale x 4 x i1> %res
 }
 
@@ -739,7 +739,7 @@ define <vscale x 8 x i1> @splice_nxv8i1(<vscale x 8 x i1> %a, <vscale x 8 x i1>
 ; CHECK-NEXT:    and z1.h, z1.h, #0x1
 ; CHECK-NEXT:    cmpne p0.h, p0/z, z1.h, #0
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 8 x i1> @llvm.experimental.vector.splice.nxv8i1(<vscale x 8 x i1> %a, <vscale x 8 x i1> %b, i32 -1)
+  %res = call <vscale x 8 x i1> @llvm.vector.splice.nxv8i1(<vscale x 8 x i1> %a, <vscale x 8 x i1> %b, i32 -1)
   ret <vscale x 8 x i1> %res
 }
 
@@ -756,7 +756,7 @@ define <vscale x 16 x i1> @splice_nxv16i1(<vscale x 16 x i1> %a, <vscale x 16 x
 ; CHECK-NEXT:    and z1.b, z1.b, #0x1
 ; CHECK-NEXT:    cmpne p0.b, p0/z, z1.b, #0
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 16 x i1> @llvm.experimental.vector.splice.nxv16i1(<vscale x 16 x i1> %a, <vscale x 16 x i1> %b, i32 -1)
+  %res = call <vscale x 16 x i1> @llvm.vector.splice.nxv16i1(<vscale x 16 x i1> %a, <vscale x 16 x i1> %b, i32 -1)
   ret <vscale x 16 x i1> %res
 }
 
@@ -768,7 +768,7 @@ define <vscale x 2 x i8> @splice_nxv2i8(<vscale x 2 x i8> %a, <vscale x 2 x i8>
 ; CHECK-NEXT:    rev p0.d, p0.d
 ; CHECK-NEXT:    splice z0.d, p0, z0.d, z1.d
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 2 x i8> @llvm.experimental.vector.splice.nxv2i8(<vscale x 2 x i8> %a, <vscale x 2 x i8> %b, i32 -2)
+  %res = call <vscale x 2 x i8> @llvm.vector.splice.nxv2i8(<vscale x 2 x i8> %a, <vscale x 2 x i8> %b, i32 -2)
   ret <vscale x 2 x i8> %res
 }
 
@@ -793,7 +793,7 @@ define <vscale x 8 x i32> @splice_nxv8i32(<vscale x 8 x i32> %a, <vscale x 8 x i
 ; CHECK-NEXT:    addvl sp, sp, #4
 ; CHECK-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 8 x i32> @llvm.experimental.vector.splice.nxv8i32(<vscale x 8 x i32> %a, <vscale x 8 x i32> %b, i32 -8)
+  %res = call <vscale x 8 x i32> @llvm.vector.splice.nxv8i32(<vscale x 8 x i32> %a, <vscale x 8 x i32> %b, i32 -8)
   ret <vscale x 8 x i32> %res
 }
 
@@ -826,26 +826,26 @@ define <vscale x 16 x float> @splice_nxv16f32_neg17(<vscale x 16 x float> %a, <v
 ; CHECK-NEXT:    addvl sp, sp, #8
 ; CHECK-NEXT:    ldr x29, [sp], #16 // 8-byte Folded Reload
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 16 x float> @llvm.experimental.vector.splice.nxv16f32(<vscale x 16 x float> %a, <vscale x 16 x float> %b, i32 -17)
+  %res = call <vscale x 16 x float> @llvm.vector.splice.nxv16f32(<vscale x 16 x float> %a, <vscale x 16 x float> %b, i32 -17)
   ret <vscale x 16 x float> %res
 }
 
-declare <vscale x 2 x i1> @llvm.experimental.vector.splice.nxv2i1(<vscale x 2 x i1>, <vscale x 2 x i1>, i32)
-declare <vscale x 4 x i1> @llvm.experimental.vector.splice.nxv4i1(<vscale x 4 x i1>, <vscale x 4 x i1>, i32)
-declare <vscale x 8 x i1> @llvm.experimental.vector.splice.nxv8i1(<vscale x 8 x i1>, <vscale x 8 x i1>, i32)
-declare <vscale x 16 x i1> @llvm.experimental.vector.splice.nxv16i1(<vscale x 16 x i1>, <vscale x 16 x i1>, i32)
-declare <vscale x 2 x i8> @llvm.experimental.vector.splice.nxv2i8(<vscale x 2 x i8>, <vscale x 2 x i8>, i32)
-declare <vscale x 16 x i8> @llvm.experimental.vector.splice.nxv16i8(<vscale x 16 x i8>, <vscale x 16 x i8>, i32)
-declare <vscale x 8 x i16> @llvm.experimental.vector.splice.nxv8i16(<vscale x 8 x i16>, <vscale x 8 x i16>, i32)
-declare <vscale x 4 x i32> @llvm.experimental.vector.splice.nxv4i32(<vscale x 4 x i32>, <vscale x 4 x i32>, i32)
-declare <vscale x 8 x i32> @llvm.experimental.vector.splice.nxv8i32(<vscale x 8 x i32>, <vscale x 8 x i32>, i32)
-declare <vscale x 2 x i64> @llvm.experimental.vector.splice.nxv2i64(<vscale x 2 x i64>, <vscale x 2 x i64>, i32)
-declare <vscale x 2 x half> @llvm.experimental.vector.splice.nxv2f16(<vscale x 2 x half>, <vscale x 2 x half>, i32)
-declare <vscale x 4 x half> @llvm.experimental.vector.splice.nxv4f16(<vscale x 4 x half>, <vscale x 4 x half>, i32)
-declare <vscale x 8 x half> @llvm.experimental.vector.splice.nxv8f16(<vscale x 8 x half>, <vscale x 8 x half>, i32)
-declare <vscale x 2 x float> @llvm.experimental.vector.splice.nxv2f32(<vscale x 2 x float>, <vscale x 2 x float>, i32)
-declare <vscale x 4 x float> @llvm.experimental.vector.splice.nxv4f32(<vscale x 4 x float>, <vscale x 4 x float>, i32)
-declare <vscale x 16 x float> @llvm.experimental.vector.splice.nxv16f32(<vscale x 16 x float>, <vscale x 16 x float>, i32)
-declare <vscale x 2 x double> @llvm.experimental.vector.splice.nxv2f64(<vscale x 2 x double>, <vscale x 2 x double>, i32)
+declare <vscale x 2 x i1> @llvm.vector.splice.nxv2i1(<vscale x 2 x i1>, <vscale x 2 x i1>, i32)
+declare <vscale x 4 x i1> @llvm.vector.splice.nxv4i1(<vscale x 4 x i1>, <vscale x 4 x i1>, i32)
+declare <vscale x 8 x i1> @llvm.vector.splice.nxv8i1(<vscale x 8 x i1>, <vscale x 8 x i1>, i32)
+declare <vscale x 16 x i1> @llvm.vector.splice.nxv16i1(<vscale x 16 x i1>, <vscale x 16 x i1>, i32)
+declare <vscale x 2 x i8> @llvm.vector.splice.nxv2i8(<vscale x 2 x i8>, <vscale x 2 x i8>, i32)
+declare <vscale x 16 x i8> @llvm.vector.splice.nxv16i8(<vscale x 16 x i8>, <vscale x 16 x i8>, i32)
+declare <vscale x 8 x i16> @llvm.vector.splice.nxv8i16(<vscale x 8 x i16>, <vscale x 8 x i16>, i32)
+declare <vscale x 4 x i32> @llvm.vector.splice.nxv4i32(<vscale x 4 x i32>, <vscale x 4 x i32>, i32)
+declare <vscale x 8 x i32> @llvm.vector.splice.nxv8i32(<vscale x 8 x i32>, <vscale x 8 x i32>, i32)
+declare <vscale x 2 x i64> @llvm.vector.splice.nxv2i64(<vscale x 2 x i64>, <vscale x 2 x i64>, i32)
+declare <vscale x 2 x half> @llvm.vector.splice.nxv2f16(<vscale x 2 x half>, <vscale x 2 x half>, i32)
+declare <vscale x 4 x half> @llvm.vector.splice.nxv4f16(<vscale x 4 x half>, <vscale x 4 x half>, i32)
+declare <vscale x 8 x half> @llvm.vector.splice.nxv8f16(<vscale x 8 x half>, <vscale x 8 x half>, i32)
+declare <vscale x 2 x float> @llvm.vector.splice.nxv2f32(<vscale x 2 x float>, <vscale x 2 x float>, i32)
+declare <vscale x 4 x float> @llvm.vector.splice.nxv4f32(<vscale x 4 x float>, <vscale x 4 x float>, i32)
+declare <vscale x 16 x float> @llvm.vector.splice.nxv16f32(<vscale x 16 x float>, <vscale x 16 x float>, i32)
+declare <vscale x 2 x double> @llvm.vector.splice.nxv2f64(<vscale x 2 x double>, <vscale x 2 x double>, i32)
 
 attributes #0 = { nounwind "target-features"="+sve" }
diff --git a/llvm/test/CodeGen/AArch64/sve-streaming-mode-fixed-length-bitcast.ll b/llvm/test/CodeGen/AArch64/sve-streaming-mode-fixed-length-bitcast.ll
index dd72c2b2bd01..e3cc74f766ee 100644
--- a/llvm/test/CodeGen/AArch64/sve-streaming-mode-fixed-length-bitcast.ll
+++ b/llvm/test/CodeGen/AArch64/sve-streaming-mode-fixed-length-bitcast.ll
@@ -60,11 +60,8 @@ define void @bitcast_v2i16(ptr %a, ptr %b) {
 ; CHECK:       // %bb.0:
 ; CHECK-NEXT:    sub sp, sp, #16
 ; CHECK-NEXT:    .cfi_def_cfa_offset 16
-; CHECK-NEXT:    ldrh w8, [x0, #2]
-; CHECK-NEXT:    str w8, [sp, #4]
-; CHECK-NEXT:    ldrh w8, [x0]
-; CHECK-NEXT:    str w8, [sp]
-; CHECK-NEXT:    ldr d0, [sp]
+; CHECK-NEXT:    ptrue p0.s, vl2
+; CHECK-NEXT:    ld1h { z0.s }, p0/z, [x0]
 ; CHECK-NEXT:    mov z1.s, z0.s[1]
 ; CHECK-NEXT:    fmov w8, s0
 ; CHECK-NEXT:    strh w8, [sp, #8]
diff --git a/llvm/test/CodeGen/AArch64/sve-streaming-mode-fixed-length-extract-vector-elt.ll b/llvm/test/CodeGen/AArch64/sve-streaming-mode-fixed-length-extract-vector-elt.ll
index 1b9bb42c8582..a752e119b2fb 100644
--- a/llvm/test/CodeGen/AArch64/sve-streaming-mode-fixed-length-extract-vector-elt.ll
+++ b/llvm/test/CodeGen/AArch64/sve-streaming-mode-fixed-length-extract-vector-elt.ll
@@ -90,8 +90,6 @@ define float @extractelement_v8f32(ptr %a) {
 define double @extractelement_v1f64(<1 x double> %op1) {
 ; CHECK-LABEL: extractelement_v1f64:
 ; CHECK:       // %bb.0:
-; CHECK-NEXT:    // kill: def $d0 killed $d0 def $z0
-; CHECK-NEXT:    // kill: def $d0 killed $d0 killed $z0
 ; CHECK-NEXT:    ret
   %r = extractelement <1 x double> %op1, i64 0
   ret double %r
diff --git a/llvm/test/CodeGen/AArch64/sve-streaming-mode-fixed-length-fp-compares.ll b/llvm/test/CodeGen/AArch64/sve-streaming-mode-fixed-length-fp-compares.ll
index e92694d1fc80..465cc179a3b9 100644
--- a/llvm/test/CodeGen/AArch64/sve-streaming-mode-fixed-length-fp-compares.ll
+++ b/llvm/test/CodeGen/AArch64/sve-streaming-mode-fixed-length-fp-compares.ll
@@ -127,11 +127,9 @@ define void @fcmp_oeq_v8f32(ptr %a, ptr %b, ptr %c) {
 define <1 x i64> @fcmp_oeq_v1f64(<1 x double> %op1, <1 x double> %op2) {
 ; CHECK-LABEL: fcmp_oeq_v1f64:
 ; CHECK:       // %bb.0:
-; CHECK-NEXT:    ptrue p0.d, vl1
-; CHECK-NEXT:    // kill: def $d1 killed $d1 def $z1
-; CHECK-NEXT:    // kill: def $d0 killed $d0 def $z0
-; CHECK-NEXT:    fcmeq p0.d, p0/z, z0.d, z1.d
-; CHECK-NEXT:    mov z0.d, p0/z, #-1 // =0xffffffffffffffff
+; CHECK-NEXT:    fcmp d0, d1
+; CHECK-NEXT:    csetm x8, eq
+; CHECK-NEXT:    mov z0.d, x8
 ; CHECK-NEXT:    // kill: def $d0 killed $d0 killed $z0
 ; CHECK-NEXT:    ret
   %cmp = fcmp oeq <1 x double> %op1, %op2
diff --git a/llvm/test/CodeGen/AArch64/sve-streaming-mode-fixed-length-fp-fma.ll b/llvm/test/CodeGen/AArch64/sve-streaming-mode-fixed-length-fp-fma.ll
index 478be9ab76dd..cbe71d715a8f 100644
--- a/llvm/test/CodeGen/AArch64/sve-streaming-mode-fixed-length-fp-fma.ll
+++ b/llvm/test/CodeGen/AArch64/sve-streaming-mode-fixed-length-fp-fma.ll
@@ -112,9 +112,6 @@ define void @fma_v8f32(ptr %a, ptr %b, ptr %c) {
 define <1 x double> @fma_v1f64(<1 x double> %op1, <1 x double> %op2, <1 x double> %op3) {
 ; CHECK-LABEL: fma_v1f64:
 ; CHECK:       // %bb.0:
-; CHECK-NEXT:    // kill: def $d0 killed $d0 def $z0
-; CHECK-NEXT:    // kill: def $d2 killed $d2 def $z2
-; CHECK-NEXT:    // kill: def $d1 killed $d1 def $z1
 ; CHECK-NEXT:    fmadd d0, d0, d1, d2
 ; CHECK-NEXT:    ret
   %mul = fmul contract <1 x double> %op1, %op2
diff --git a/llvm/test/CodeGen/AArch64/sve-streaming-mode-fixed-length-fp-minmax.ll b/llvm/test/CodeGen/AArch64/sve-streaming-mode-fixed-length-fp-minmax.ll
index 4dc034adf459..94a74763aa0e 100644
--- a/llvm/test/CodeGen/AArch64/sve-streaming-mode-fixed-length-fp-minmax.ll
+++ b/llvm/test/CodeGen/AArch64/sve-streaming-mode-fixed-length-fp-minmax.ll
@@ -99,8 +99,6 @@ define void @fmaxnm_v8f32(ptr %a, ptr %b) {
 define <1 x double> @fmaxnm_v1f64(<1 x double> %op1, <1 x double> %op2) {
 ; CHECK-LABEL: fmaxnm_v1f64:
 ; CHECK:       // %bb.0:
-; CHECK-NEXT:    // kill: def $d0 killed $d0 def $z0
-; CHECK-NEXT:    // kill: def $d1 killed $d1 def $z1
 ; CHECK-NEXT:    fmaxnm d0, d0, d1
 ; CHECK-NEXT:    ret
   %res = call <1 x double> @llvm.maxnum.v1f64(<1 x double> %op1, <1 x double> %op2)
@@ -233,8 +231,6 @@ define void @fminnm_v8f32(ptr %a, ptr %b) {
 define <1 x double> @fminnm_v1f64(<1 x double> %op1, <1 x double> %op2) {
 ; CHECK-LABEL: fminnm_v1f64:
 ; CHECK:       // %bb.0:
-; CHECK-NEXT:    // kill: def $d0 killed $d0 def $z0
-; CHECK-NEXT:    // kill: def $d1 killed $d1 def $z1
 ; CHECK-NEXT:    fminnm d0, d0, d1
 ; CHECK-NEXT:    ret
   %res = call <1 x double> @llvm.minnum.v1f64(<1 x double> %op1, <1 x double> %op2)
@@ -367,8 +363,6 @@ define void @fmax_v8f32(ptr %a, ptr %b) {
 define <1 x double> @fmax_v1f64(<1 x double> %op1, <1 x double> %op2) {
 ; CHECK-LABEL: fmax_v1f64:
 ; CHECK:       // %bb.0:
-; CHECK-NEXT:    // kill: def $d0 killed $d0 def $z0
-; CHECK-NEXT:    // kill: def $d1 killed $d1 def $z1
 ; CHECK-NEXT:    fmax d0, d0, d1
 ; CHECK-NEXT:    ret
   %res = call <1 x double> @llvm.maximum.v1f64(<1 x double> %op1, <1 x double> %op2)
@@ -501,8 +495,6 @@ define void @fmin_v8f32(ptr %a, ptr %b) {
 define <1 x double> @fmin_v1f64(<1 x double> %op1, <1 x double> %op2) {
 ; CHECK-LABEL: fmin_v1f64:
 ; CHECK:       // %bb.0:
-; CHECK-NEXT:    // kill: def $d0 killed $d0 def $z0
-; CHECK-NEXT:    // kill: def $d1 killed $d1 def $z1
 ; CHECK-NEXT:    fmin d0, d0, d1
 ; CHECK-NEXT:    ret
   %res = call <1 x double> @llvm.minimum.v1f64(<1 x double> %op1, <1 x double> %op2)
diff --git a/llvm/test/CodeGen/AArch64/sve-streaming-mode-fixed-length-fp-reduce.ll b/llvm/test/CodeGen/AArch64/sve-streaming-mode-fixed-length-fp-reduce.ll
index bd10a0e091c0..df9613a30e40 100644
--- a/llvm/test/CodeGen/AArch64/sve-streaming-mode-fixed-length-fp-reduce.ll
+++ b/llvm/test/CodeGen/AArch64/sve-streaming-mode-fixed-length-fp-reduce.ll
@@ -144,7 +144,6 @@ define float @fadda_v8f32(float %start, ptr %a) {
 define double @fadda_v1f64(double %start, <1 x double> %a) {
 ; CHECK-LABEL: fadda_v1f64:
 ; CHECK:       // %bb.0:
-; CHECK-NEXT:    // kill: def $d1 killed $d1 def $z1
 ; CHECK-NEXT:    fadd d0, d0, d1
 ; CHECK-NEXT:    ret
   %res = call double @llvm.vector.reduce.fadd.v1f64(double %start, <1 x double> %a)
@@ -263,7 +262,6 @@ define float @faddv_v8f32(float %start, ptr %a) {
 define double @faddv_v1f64(double %start, <1 x double> %a) {
 ; CHECK-LABEL: faddv_v1f64:
 ; CHECK:       // %bb.0:
-; CHECK-NEXT:    // kill: def $d1 killed $d1 def $z1
 ; CHECK-NEXT:    fadd d0, d0, d1
 ; CHECK-NEXT:    ret
   %res = call fast double @llvm.vector.reduce.fadd.v1f64(double %start, <1 x double> %a)
@@ -379,8 +377,6 @@ define float @fmaxv_v8f32(ptr %a) {
 define double @fmaxv_v1f64(<1 x double> %a) {
 ; CHECK-LABEL: fmaxv_v1f64:
 ; CHECK:       // %bb.0:
-; CHECK-NEXT:    // kill: def $d0 killed $d0 def $z0
-; CHECK-NEXT:    // kill: def $d0 killed $d0 killed $z0
 ; CHECK-NEXT:    ret
   %res = call double @llvm.vector.reduce.fmax.v1f64(<1 x double> %a)
   ret double %res
@@ -495,8 +491,6 @@ define float @fminv_v8f32(ptr %a) {
 define double @fminv_v1f64(<1 x double> %a) {
 ; CHECK-LABEL: fminv_v1f64:
 ; CHECK:       // %bb.0:
-; CHECK-NEXT:    // kill: def $d0 killed $d0 def $z0
-; CHECK-NEXT:    // kill: def $d0 killed $d0 killed $z0
 ; CHECK-NEXT:    ret
   %res = call double @llvm.vector.reduce.fmin.v1f64(<1 x double> %a)
   ret double %res
@@ -611,8 +605,6 @@ define float @fmaximumv_v8f32(ptr %a) {
 define double @fmaximumv_v1f64(<1 x double> %a) {
 ; CHECK-LABEL: fmaximumv_v1f64:
 ; CHECK:       // %bb.0:
-; CHECK-NEXT:    // kill: def $d0 killed $d0 def $z0
-; CHECK-NEXT:    // kill: def $d0 killed $d0 killed $z0
 ; CHECK-NEXT:    ret
   %res = call double @llvm.vector.reduce.fmaximum.v1f64(<1 x double> %a)
   ret double %res
@@ -727,8 +719,6 @@ define float @fminimumv_v8f32(ptr %a) {
 define double @fminimumv_v1f64(<1 x double> %a) {
 ; CHECK-LABEL: fminimumv_v1f64:
 ; CHECK:       // %bb.0:
-; CHECK-NEXT:    // kill: def $d0 killed $d0 def $z0
-; CHECK-NEXT:    // kill: def $d0 killed $d0 killed $z0
 ; CHECK-NEXT:    ret
   %res = call double @llvm.vector.reduce.fminimum.v1f64(<1 x double> %a)
   ret double %res
diff --git a/llvm/test/CodeGen/AArch64/sve-streaming-mode-fixed-length-fp-rounding.ll b/llvm/test/CodeGen/AArch64/sve-streaming-mode-fixed-length-fp-rounding.ll
index 24832d807c64..7ddc641f366c 100644
--- a/llvm/test/CodeGen/AArch64/sve-streaming-mode-fixed-length-fp-rounding.ll
+++ b/llvm/test/CodeGen/AArch64/sve-streaming-mode-fixed-length-fp-rounding.ll
@@ -101,7 +101,6 @@ define void @frintp_v8f32(ptr %a) {
 define <1 x double> @frintp_v1f64(<1 x double> %op) {
 ; CHECK-LABEL: frintp_v1f64:
 ; CHECK:       // %bb.0:
-; CHECK-NEXT:    // kill: def $d0 killed $d0 def $z0
 ; CHECK-NEXT:    frintp d0, d0
 ; CHECK-NEXT:    ret
   %res = call <1 x double> @llvm.ceil.v1f64(<1 x double> %op)
@@ -232,7 +231,6 @@ define void @frintm_v8f32(ptr %a) {
 define <1 x double> @frintm_v1f64(<1 x double> %op) {
 ; CHECK-LABEL: frintm_v1f64:
 ; CHECK:       // %bb.0:
-; CHECK-NEXT:    // kill: def $d0 killed $d0 def $z0
 ; CHECK-NEXT:    frintm d0, d0
 ; CHECK-NEXT:    ret
   %res = call <1 x double> @llvm.floor.v1f64(<1 x double> %op)
@@ -363,7 +361,6 @@ define void @frinti_v8f32(ptr %a) {
 define <1 x double> @frinti_v1f64(<1 x double> %op) {
 ; CHECK-LABEL: frinti_v1f64:
 ; CHECK:       // %bb.0:
-; CHECK-NEXT:    // kill: def $d0 killed $d0 def $z0
 ; CHECK-NEXT:    frinti d0, d0
 ; CHECK-NEXT:    ret
   %res = call <1 x double> @llvm.nearbyint.v1f64(<1 x double> %op)
@@ -494,7 +491,6 @@ define void @frintx_v8f32(ptr %a) {
 define <1 x double> @frintx_v1f64(<1 x double> %op) {
 ; CHECK-LABEL: frintx_v1f64:
 ; CHECK:       // %bb.0:
-; CHECK-NEXT:    // kill: def $d0 killed $d0 def $z0
 ; CHECK-NEXT:    frintx d0, d0
 ; CHECK-NEXT:    ret
   %res = call <1 x double> @llvm.rint.v1f64(<1 x double> %op)
@@ -625,7 +621,6 @@ define void @frinta_v8f32(ptr %a) {
 define <1 x double> @frinta_v1f64(<1 x double> %op) {
 ; CHECK-LABEL: frinta_v1f64:
 ; CHECK:       // %bb.0:
-; CHECK-NEXT:    // kill: def $d0 killed $d0 def $z0
 ; CHECK-NEXT:    frinta d0, d0
 ; CHECK-NEXT:    ret
   %res = call <1 x double> @llvm.round.v1f64(<1 x double> %op)
@@ -756,7 +751,6 @@ define void @frintn_v8f32(ptr %a) {
 define <1 x double> @frintn_v1f64(<1 x double> %op) {
 ; CHECK-LABEL: frintn_v1f64:
 ; CHECK:       // %bb.0:
-; CHECK-NEXT:    // kill: def $d0 killed $d0 def $z0
 ; CHECK-NEXT:    frintn d0, d0
 ; CHECK-NEXT:    ret
   %res = call <1 x double> @llvm.roundeven.v1f64(<1 x double> %op)
@@ -887,7 +881,6 @@ define void @frintz_v8f32(ptr %a) {
 define <1 x double> @frintz_v1f64(<1 x double> %op) {
 ; CHECK-LABEL: frintz_v1f64:
 ; CHECK:       // %bb.0:
-; CHECK-NEXT:    // kill: def $d0 killed $d0 def $z0
 ; CHECK-NEXT:    frintz d0, d0
 ; CHECK-NEXT:    ret
   %res = call <1 x double> @llvm.trunc.v1f64(<1 x double> %op)
diff --git a/llvm/test/CodeGen/AArch64/sve-streaming-mode-fixed-length-fp-select.ll b/llvm/test/CodeGen/AArch64/sve-streaming-mode-fixed-length-fp-select.ll
index 132225546fc4..7d36925fdc57 100644
--- a/llvm/test/CodeGen/AArch64/sve-streaming-mode-fixed-length-fp-select.ll
+++ b/llvm/test/CodeGen/AArch64/sve-streaming-mode-fixed-length-fp-select.ll
@@ -132,16 +132,7 @@ define <1 x double> @select_v1f64(<1 x double> %op1, <1 x double> %op2, i1 %mask
 ; CHECK-LABEL: select_v1f64:
 ; CHECK:       // %bb.0:
 ; CHECK-NEXT:    tst w0, #0x1
-; CHECK-NEXT:    // kill: def $d1 killed $d1 def $z1
-; CHECK-NEXT:    // kill: def $d0 killed $d0 def $z0
-; CHECK-NEXT:    csetm x8, ne
-; CHECK-NEXT:    mvn x9, x8
-; CHECK-NEXT:    mov z2.d, x8
-; CHECK-NEXT:    mov z3.d, x9
-; CHECK-NEXT:    and z0.d, z0.d, z2.d
-; CHECK-NEXT:    and z1.d, z1.d, z3.d
-; CHECK-NEXT:    orr z0.d, z0.d, z1.d
-; CHECK-NEXT:    // kill: def $d0 killed $d0 killed $z0
+; CHECK-NEXT:    fcsel d0, d0, d1, ne
 ; CHECK-NEXT:    ret
   %sel = select i1 %mask, <1 x double> %op1, <1 x double> %op2
   ret <1 x double> %sel
diff --git a/llvm/test/CodeGen/AArch64/sve-streaming-mode-fixed-length-fp-to-int.ll b/llvm/test/CodeGen/AArch64/sve-streaming-mode-fixed-length-fp-to-int.ll
index 58eae212d799..bf8a335a8503 100644
--- a/llvm/test/CodeGen/AArch64/sve-streaming-mode-fixed-length-fp-to-int.ll
+++ b/llvm/test/CodeGen/AArch64/sve-streaming-mode-fixed-length-fp-to-int.ll
@@ -464,7 +464,6 @@ define void @fcvtzu_v8f32_v8i64(ptr %a, ptr %b) {
 define <1 x i16> @fcvtzu_v1f64_v1i16(<1 x double> %op1) {
 ; CHECK-LABEL: fcvtzu_v1f64_v1i16:
 ; CHECK:       // %bb.0:
-; CHECK-NEXT:    // kill: def $d0 killed $d0 def $z0
 ; CHECK-NEXT:    fcvtzs w8, d0
 ; CHECK-NEXT:    mov z0.h, w8
 ; CHECK-NEXT:    // kill: def $d0 killed $d0 killed $z0
@@ -1215,7 +1214,6 @@ define void @fcvtzs_v8f32_v8i64(ptr %a, ptr %b) {
 define <1 x i16> @fcvtzs_v1f64_v1i16(<1 x double> %op1) {
 ; CHECK-LABEL: fcvtzs_v1f64_v1i16:
 ; CHECK:       // %bb.0:
-; CHECK-NEXT:    // kill: def $d0 killed $d0 def $z0
 ; CHECK-NEXT:    fcvtzs w8, d0
 ; CHECK-NEXT:    mov z0.h, w8
 ; CHECK-NEXT:    // kill: def $d0 killed $d0 killed $z0
diff --git a/llvm/test/CodeGen/AArch64/sve-streaming-mode-fixed-length-fp-vselect.ll b/llvm/test/CodeGen/AArch64/sve-streaming-mode-fixed-length-fp-vselect.ll
index 4c5a6fe2fd23..30a4f04a3d2b 100644
--- a/llvm/test/CodeGen/AArch64/sve-streaming-mode-fixed-length-fp-vselect.ll
+++ b/llvm/test/CodeGen/AArch64/sve-streaming-mode-fixed-length-fp-vselect.ll
@@ -149,16 +149,7 @@ define <1 x double> @select_v1f64(<1 x double> %op1, <1 x double> %op2, <1 x i1>
 ; CHECK-LABEL: select_v1f64:
 ; CHECK:       // %bb.0:
 ; CHECK-NEXT:    tst w0, #0x1
-; CHECK-NEXT:    // kill: def $d1 killed $d1 def $z1
-; CHECK-NEXT:    // kill: def $d0 killed $d0 def $z0
-; CHECK-NEXT:    csetm x8, ne
-; CHECK-NEXT:    mvn x9, x8
-; CHECK-NEXT:    mov z2.d, x8
-; CHECK-NEXT:    mov z3.d, x9
-; CHECK-NEXT:    and z0.d, z0.d, z2.d
-; CHECK-NEXT:    and z1.d, z1.d, z3.d
-; CHECK-NEXT:    orr z0.d, z0.d, z1.d
-; CHECK-NEXT:    // kill: def $d0 killed $d0 killed $z0
+; CHECK-NEXT:    fcsel d0, d0, d1, ne
 ; CHECK-NEXT:    ret
   %sel = select <1 x i1> %mask, <1 x double> %op1, <1 x double> %op2
   ret <1 x double> %sel
diff --git a/llvm/test/CodeGen/AArch64/sve-streaming-mode-fixed-length-ld2-alloca.ll b/llvm/test/CodeGen/AArch64/sve-streaming-mode-fixed-length-ld2-alloca.ll
index 1fc51d50b50a..efe9066f2c83 100644
--- a/llvm/test/CodeGen/AArch64/sve-streaming-mode-fixed-length-ld2-alloca.ll
+++ b/llvm/test/CodeGen/AArch64/sve-streaming-mode-fixed-length-ld2-alloca.ll
@@ -47,6 +47,7 @@ define void @alloc_v6i8(ptr %st_ptr) nounwind {
 ; CHECK-NEXT:    add x20, sp, #24
 ; CHECK-NEXT:    bl def
 ; CHECK-NEXT:    ptrue p0.b, vl3
+; CHECK-NEXT:    ptrue p1.s, vl2
 ; CHECK-NEXT:    ld2b { z0.b, z1.b }, p0/z, [x20]
 ; CHECK-NEXT:    ptrue p0.h, vl4
 ; CHECK-NEXT:    mov z2.b, z1.b[3]
@@ -63,9 +64,10 @@ define void @alloc_v6i8(ptr %st_ptr) nounwind {
 ; CHECK-NEXT:    add x8, sp, #12
 ; CHECK-NEXT:    ldr d0, [sp]
 ; CHECK-NEXT:    st1b { z0.h }, p0, [x8]
-; CHECK-NEXT:    ldrh w8, [sp, #12]
+; CHECK-NEXT:    ld1h { z0.s }, p1/z, [x8]
 ; CHECK-NEXT:    strb w9, [x19, #2]
 ; CHECK-NEXT:    ldr x30, [sp, #16] // 8-byte Folded Reload
+; CHECK-NEXT:    fmov w8, s0
 ; CHECK-NEXT:    strh w8, [x19]
 ; CHECK-NEXT:    ldp x20, x19, [sp, #32] // 16-byte Folded Reload
 ; CHECK-NEXT:    add sp, sp, #48
diff --git a/llvm/test/CodeGen/AArch64/sve-streaming-mode-fixed-length-loads.ll b/llvm/test/CodeGen/AArch64/sve-streaming-mode-fixed-length-loads.ll
index 688c39b89c0d..8ca8e6980913 100644
--- a/llvm/test/CodeGen/AArch64/sve-streaming-mode-fixed-length-loads.ll
+++ b/llvm/test/CodeGen/AArch64/sve-streaming-mode-fixed-length-loads.ll
@@ -45,14 +45,9 @@ define <32 x i8> @load_v32i8(ptr %a) {
 define <2 x i16> @load_v2i16(ptr %a) {
 ; CHECK-LABEL: load_v2i16:
 ; CHECK:       // %bb.0:
-; CHECK-NEXT:    sub sp, sp, #16
-; CHECK-NEXT:    .cfi_def_cfa_offset 16
-; CHECK-NEXT:    ldrh w8, [x0, #2]
-; CHECK-NEXT:    str w8, [sp, #12]
-; CHECK-NEXT:    ldrh w8, [x0]
-; CHECK-NEXT:    str w8, [sp, #8]
-; CHECK-NEXT:    ldr d0, [sp, #8]
-; CHECK-NEXT:    add sp, sp, #16
+; CHECK-NEXT:    ptrue p0.s, vl2
+; CHECK-NEXT:    ld1h { z0.s }, p0/z, [x0]
+; CHECK-NEXT:    // kill: def $d0 killed $d0 killed $z0
 ; CHECK-NEXT:    ret
   %load = load <2 x i16>, ptr %a
   ret <2 x i16> %load
diff --git a/llvm/test/CodeGen/AArch64/sve-streaming-mode-fixed-length-optimize-ptrue.ll b/llvm/test/CodeGen/AArch64/sve-streaming-mode-fixed-length-optimize-ptrue.ll
index 6fcb95f28333..b5adea594242 100644
--- a/llvm/test/CodeGen/AArch64/sve-streaming-mode-fixed-length-optimize-ptrue.ll
+++ b/llvm/test/CodeGen/AArch64/sve-streaming-mode-fixed-length-optimize-ptrue.ll
@@ -70,21 +70,11 @@ define void @add_v32i8(ptr %a, ptr %b) {
 define void @add_v2i16(ptr %a, ptr %b, ptr %c) {
 ; CHECK-LABEL: add_v2i16:
 ; CHECK:       // %bb.0:
-; CHECK-NEXT:    sub sp, sp, #16
-; CHECK-NEXT:    .cfi_def_cfa_offset 16
-; CHECK-NEXT:    ldrh w8, [x0, #2]
 ; CHECK-NEXT:    ptrue p0.s, vl2
-; CHECK-NEXT:    str w8, [sp, #4]
-; CHECK-NEXT:    ldrh w8, [x0]
-; CHECK-NEXT:    str w8, [sp]
-; CHECK-NEXT:    ldrh w8, [x1, #2]
-; CHECK-NEXT:    str w8, [sp, #12]
-; CHECK-NEXT:    ldrh w8, [x1]
-; CHECK-NEXT:    str w8, [sp, #8]
-; CHECK-NEXT:    ldp d0, d1, [sp]
+; CHECK-NEXT:    ld1h { z0.s }, p0/z, [x0]
+; CHECK-NEXT:    ld1h { z1.s }, p0/z, [x1]
 ; CHECK-NEXT:    add z0.s, z0.s, z1.s
 ; CHECK-NEXT:    st1h { z0.s }, p0, [x0]
-; CHECK-NEXT:    add sp, sp, #16
 ; CHECK-NEXT:    ret
   %op1 = load <2 x i16>, ptr %a
   %op2 = load <2 x i16>, ptr %b
diff --git a/llvm/test/CodeGen/AArch64/sve-streaming-mode-fixed-length-subvector.ll b/llvm/test/CodeGen/AArch64/sve-streaming-mode-fixed-length-subvector.ll
index 75bae88fc479..838db0ce8185 100644
--- a/llvm/test/CodeGen/AArch64/sve-streaming-mode-fixed-length-subvector.ll
+++ b/llvm/test/CodeGen/AArch64/sve-streaming-mode-fixed-length-subvector.ll
@@ -77,16 +77,9 @@ bb1:
 define void @subvector_v2i16(ptr %in, ptr %out) {
 ; CHECK-LABEL: subvector_v2i16:
 ; CHECK:       // %bb.0: // %bb1
-; CHECK-NEXT:    sub sp, sp, #16
-; CHECK-NEXT:    .cfi_def_cfa_offset 16
-; CHECK-NEXT:    ldrh w8, [x0, #2]
 ; CHECK-NEXT:    ptrue p0.s, vl2
-; CHECK-NEXT:    str w8, [sp, #12]
-; CHECK-NEXT:    ldrh w8, [x0]
-; CHECK-NEXT:    str w8, [sp, #8]
-; CHECK-NEXT:    ldr d0, [sp, #8]
+; CHECK-NEXT:    ld1h { z0.s }, p0/z, [x0]
 ; CHECK-NEXT:    st1h { z0.s }, p0, [x1]
-; CHECK-NEXT:    add sp, sp, #16
 ; CHECK-NEXT:    ret
   %a = load <2 x i16>, ptr %in
   br label %bb1
diff --git a/llvm/test/CodeGen/AArch64/sve-vector-deinterleave.ll b/llvm/test/CodeGen/AArch64/sve-vector-deinterleave.ll
index 9920bc6048e8..478f4a689d3c 100644
--- a/llvm/test/CodeGen/AArch64/sve-vector-deinterleave.ll
+++ b/llvm/test/CodeGen/AArch64/sve-vector-deinterleave.ll
@@ -9,7 +9,7 @@ define {<vscale x 2 x half>, <vscale x 2 x half>} @vector_deinterleave_nxv2f16_n
 ; CHECK-NEXT:    uzp1 z0.d, z2.d, z1.d
 ; CHECK-NEXT:    uzp2 z1.d, z2.d, z1.d
 ; CHECK-NEXT:    ret
-  %retval = call {<vscale x 2 x half>, <vscale x 2 x half>} @llvm.experimental.vector.deinterleave2.nxv4f16(<vscale x 4 x half> %vec)
+  %retval = call {<vscale x 2 x half>, <vscale x 2 x half>} @llvm.vector.deinterleave2.nxv4f16(<vscale x 4 x half> %vec)
   ret {<vscale x 2 x half>, <vscale x 2 x half>}   %retval
 }
 
@@ -21,7 +21,7 @@ define {<vscale x 4 x half>, <vscale x 4 x half>} @vector_deinterleave_nxv4f16_n
 ; CHECK-NEXT:    uzp1 z0.s, z2.s, z1.s
 ; CHECK-NEXT:    uzp2 z1.s, z2.s, z1.s
 ; CHECK-NEXT:    ret
-  %retval = call {<vscale x 4 x half>, <vscale x 4 x half>} @llvm.experimental.vector.deinterleave2.nxv8f16(<vscale x 8 x half> %vec)
+  %retval = call {<vscale x 4 x half>, <vscale x 4 x half>} @llvm.vector.deinterleave2.nxv8f16(<vscale x 8 x half> %vec)
   ret {<vscale x 4 x half>, <vscale x 4 x half>}   %retval
 }
 
@@ -32,7 +32,7 @@ define {<vscale x 8 x half>, <vscale x 8 x half>} @vector_deinterleave_nxv8f16_n
 ; CHECK-NEXT:    uzp2 z1.h, z0.h, z1.h
 ; CHECK-NEXT:    mov z0.d, z2.d
 ; CHECK-NEXT:    ret
-  %retval = call {<vscale x 8 x half>, <vscale x 8 x half>} @llvm.experimental.vector.deinterleave2.nxv16f16(<vscale x 16 x half> %vec)
+  %retval = call {<vscale x 8 x half>, <vscale x 8 x half>} @llvm.vector.deinterleave2.nxv16f16(<vscale x 16 x half> %vec)
   ret {<vscale x 8 x half>, <vscale x 8 x half>}   %retval
 }
 
@@ -44,7 +44,7 @@ define {<vscale x 2 x float>, <vscale x 2 x float>} @vector_deinterleave_nxv2f32
 ; CHECK-NEXT:    uzp1 z0.d, z2.d, z1.d
 ; CHECK-NEXT:    uzp2 z1.d, z2.d, z1.d
 ; CHECK-NEXT:    ret
-  %retval = call {<vscale x 2 x float>, <vscale x 2 x float>} @llvm.experimental.vector.deinterleave2.nxv4f32(<vscale x 4 x float> %vec)
+  %retval = call {<vscale x 2 x float>, <vscale x 2 x float>} @llvm.vector.deinterleave2.nxv4f32(<vscale x 4 x float> %vec)
   ret {<vscale x 2 x float>, <vscale x 2 x float>}   %retval
 }
 
@@ -55,7 +55,7 @@ define {<vscale x 4 x float>, <vscale x 4 x float>} @vector_deinterleave_nxv4f32
 ; CHECK-NEXT:    uzp2 z1.s, z0.s, z1.s
 ; CHECK-NEXT:    mov z0.d, z2.d
 ; CHECK-NEXT:    ret
-  %retval = call {<vscale x 4 x float>, <vscale x 4 x float>} @llvm.experimental.vector.deinterleave2.nxv8f32(<vscale x 8 x float> %vec)
+  %retval = call {<vscale x 4 x float>, <vscale x 4 x float>} @llvm.vector.deinterleave2.nxv8f32(<vscale x 8 x float> %vec)
 ret  {<vscale x 4 x float>, <vscale x 4 x float>}   %retval
 }
 
@@ -66,7 +66,7 @@ define {<vscale x 2 x double>, <vscale x 2 x double>} @vector_deinterleave_nxv2f
 ; CHECK-NEXT:    uzp2 z1.d, z0.d, z1.d
 ; CHECK-NEXT:    mov z0.d, z2.d
 ; CHECK-NEXT:    ret
-  %retval = call {<vscale x 2 x double>, <vscale x 2 x double>} @llvm.experimental.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %vec)
+  %retval = call {<vscale x 2 x double>, <vscale x 2 x double>} @llvm.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %vec)
   ret {<vscale x 2 x double>, <vscale x 2 x double>}   %retval
 }
 
@@ -79,7 +79,7 @@ define {<vscale x 16 x i8>, <vscale x 16 x i8>} @vector_deinterleave_nxv16i8_nxv
 ; CHECK-NEXT:    uzp2 z1.b, z0.b, z1.b
 ; CHECK-NEXT:    mov z0.d, z2.d
 ; CHECK-NEXT:    ret
-  %retval = call {<vscale x 16 x i8>, <vscale x 16 x i8>} @llvm.experimental.vector.deinterleave2.nxv32i8(<vscale x 32 x i8> %vec)
+  %retval = call {<vscale x 16 x i8>, <vscale x 16 x i8>} @llvm.vector.deinterleave2.nxv32i8(<vscale x 32 x i8> %vec)
   ret {<vscale x 16 x i8>, <vscale x 16 x i8>}   %retval
 }
 
@@ -90,7 +90,7 @@ define {<vscale x 8 x i16>, <vscale x 8 x i16>} @vector_deinterleave_nxv8i16_nxv
 ; CHECK-NEXT:    uzp2 z1.h, z0.h, z1.h
 ; CHECK-NEXT:    mov z0.d, z2.d
 ; CHECK-NEXT:    ret
-  %retval = call {<vscale x 8 x i16>, <vscale x 8 x i16>} @llvm.experimental.vector.deinterleave2.nxv16i16(<vscale x 16 x i16> %vec)
+  %retval = call {<vscale x 8 x i16>, <vscale x 8 x i16>} @llvm.vector.deinterleave2.nxv16i16(<vscale x 16 x i16> %vec)
   ret {<vscale x 8 x i16>, <vscale x 8 x i16>}   %retval
 }
 
@@ -101,7 +101,7 @@ define {<vscale x 4 x i32>, <vscale x 4 x i32>} @vector_deinterleave_nxv4i32_nxv
 ; CHECK-NEXT:    uzp2 z1.s, z0.s, z1.s
 ; CHECK-NEXT:    mov z0.d, z2.d
 ; CHECK-NEXT:    ret
-  %retval = call {<vscale x 4 x i32>, <vscale x 4 x i32>} @llvm.experimental.vector.deinterleave2.nxv8i32(<vscale x 8 x i32> %vec)
+  %retval = call {<vscale x 4 x i32>, <vscale x 4 x i32>} @llvm.vector.deinterleave2.nxv8i32(<vscale x 8 x i32> %vec)
   ret {<vscale x 4 x i32>, <vscale x 4 x i32>}   %retval
 }
 
@@ -112,7 +112,7 @@ define {<vscale x 2 x i64>, <vscale x 2 x i64>} @vector_deinterleave_nxv2i64_nxv
 ; CHECK-NEXT:    uzp2 z1.d, z0.d, z1.d
 ; CHECK-NEXT:    mov z0.d, z2.d
 ; CHECK-NEXT:    ret
-  %retval = call {<vscale x 2 x i64>, <vscale x 2 x i64>} @llvm.experimental.vector.deinterleave2.nxv4i64(<vscale x 4 x i64> %vec)
+  %retval = call {<vscale x 2 x i64>, <vscale x 2 x i64>} @llvm.vector.deinterleave2.nxv4i64(<vscale x 4 x i64> %vec)
   ret {<vscale x 2 x i64>, <vscale x 2 x i64>}   %retval
 }
 
@@ -124,7 +124,7 @@ define {<vscale x 16 x i1>, <vscale x 16 x i1>} @vector_deinterleave_nxv16i1_nxv
 ; CHECK-NEXT:    uzp2 p1.b, p0.b, p1.b
 ; CHECK-NEXT:    mov p0.b, p2.b
 ; CHECK-NEXT:    ret
-  %retval = call {<vscale x 16 x i1>, <vscale x 16 x i1>} @llvm.experimental.vector.deinterleave2.nxv32i1(<vscale x 32 x i1> %vec)
+  %retval = call {<vscale x 16 x i1>, <vscale x 16 x i1>} @llvm.vector.deinterleave2.nxv32i1(<vscale x 32 x i1> %vec)
   ret {<vscale x 16 x i1>, <vscale x 16 x i1>}   %retval
 }
 
@@ -136,7 +136,7 @@ define {<vscale x 8 x i1>, <vscale x 8 x i1>} @vector_deinterleave_nxv8i1_nxv16i
 ; CHECK-NEXT:    uzp1 p0.h, p2.h, p1.h
 ; CHECK-NEXT:    uzp2 p1.h, p2.h, p1.h
 ; CHECK-NEXT:    ret
-  %retval = call {<vscale x 8 x i1>, <vscale x 8 x i1>} @llvm.experimental.vector.deinterleave2.nxv16i1(<vscale x 16 x i1> %vec)
+  %retval = call {<vscale x 8 x i1>, <vscale x 8 x i1>} @llvm.vector.deinterleave2.nxv16i1(<vscale x 16 x i1> %vec)
   ret {<vscale x 8 x i1>, <vscale x 8 x i1>}   %retval
 }
 
@@ -148,7 +148,7 @@ define {<vscale x 4 x i1>, <vscale x 4 x i1>} @vector_deinterleave_nxv4i1_nxv8i1
 ; CHECK-NEXT:    uzp1 p0.s, p2.s, p1.s
 ; CHECK-NEXT:    uzp2 p1.s, p2.s, p1.s
 ; CHECK-NEXT:    ret
-  %retval = call {<vscale x 4 x i1>, <vscale x 4 x i1>} @llvm.experimental.vector.deinterleave2.nxv8i1(<vscale x 8 x i1> %vec)
+  %retval = call {<vscale x 4 x i1>, <vscale x 4 x i1>} @llvm.vector.deinterleave2.nxv8i1(<vscale x 8 x i1> %vec)
   ret {<vscale x 4 x i1>, <vscale x 4 x i1>}   %retval
 }
 
@@ -160,7 +160,7 @@ define {<vscale x 2 x i1>, <vscale x 2 x i1>} @vector_deinterleave_nxv2i1_nxv4i1
 ; CHECK-NEXT:    uzp1 p0.d, p2.d, p1.d
 ; CHECK-NEXT:    uzp2 p1.d, p2.d, p1.d
 ; CHECK-NEXT:    ret
-  %retval = call {<vscale x 2 x i1>, <vscale x 2 x i1>} @llvm.experimental.vector.deinterleave2.nxv4i1(<vscale x 4 x i1> %vec)
+  %retval = call {<vscale x 2 x i1>, <vscale x 2 x i1>} @llvm.vector.deinterleave2.nxv4i1(<vscale x 4 x i1> %vec)
   ret {<vscale x 2 x i1>, <vscale x 2 x i1>}   %retval
 }
 
@@ -178,7 +178,7 @@ define {<vscale x 4 x i64>, <vscale x 4 x i64>} @vector_deinterleave_nxv4i64_nxv
 ; CHECK-NEXT:    mov z1.d, z4.d
 ; CHECK-NEXT:    mov z2.d, z6.d
 ; CHECK-NEXT:    ret
-%retval = call {<vscale x 4 x i64>, <vscale x 4 x i64>} @llvm.experimental.vector.deinterleave2.nxv8i64(<vscale x 8 x i64> %vec)
+%retval = call {<vscale x 4 x i64>, <vscale x 4 x i64>} @llvm.vector.deinterleave2.nxv8i64(<vscale x 8 x i64> %vec)
 ret {<vscale x 4 x i64>, <vscale x 4 x i64>} %retval
 }
 
@@ -201,7 +201,7 @@ define {<vscale x 8 x i64>, <vscale x 8 x i64>}  @vector_deinterleave_nxv8i64_nx
 ; CHECK-NEXT:    mov z5.d, z29.d
 ; CHECK-NEXT:    mov z6.d, z30.d
 ; CHECK-NEXT:    ret
-%retval = call {<vscale x 8 x i64>, <vscale x 8 x i64>} @llvm.experimental.vector.deinterleave2.nxv16i64(<vscale x 16 x i64> %vec)
+%retval = call {<vscale x 8 x i64>, <vscale x 8 x i64>} @llvm.vector.deinterleave2.nxv16i64(<vscale x 16 x i64> %vec)
 ret {<vscale x 8 x i64>, <vscale x 8 x i64>}  %retval
 }
 
@@ -216,7 +216,7 @@ define {<vscale x 8 x i8>, <vscale x 8 x i8>} @vector_deinterleave_nxv8i8_nxv16i
 ; CHECK-NEXT:    uzp1 z0.h, z2.h, z1.h
 ; CHECK-NEXT:    uzp2 z1.h, z2.h, z1.h
 ; CHECK-NEXT:    ret
-%retval = call {<vscale x 8 x i8>, <vscale x 8 x i8>} @llvm.experimental.vector.deinterleave2.nxv16i8(<vscale x 16 x i8> %vec)
+%retval = call {<vscale x 8 x i8>, <vscale x 8 x i8>} @llvm.vector.deinterleave2.nxv16i8(<vscale x 16 x i8> %vec)
 ret {<vscale x 8 x i8>, <vscale x 8 x i8>} %retval
 }
 
@@ -228,7 +228,7 @@ define {<vscale x 4 x i16>, <vscale x 4 x i16>} @vector_deinterleave_nxv4i16_nxv
 ; CHECK-NEXT:    uzp1 z0.s, z2.s, z1.s
 ; CHECK-NEXT:    uzp2 z1.s, z2.s, z1.s
 ; CHECK-NEXT:    ret
-%retval = call {<vscale x 4 x i16>, <vscale x 4 x i16>} @llvm.experimental.vector.deinterleave2.nxv8i16(<vscale x 8 x i16> %vec)
+%retval = call {<vscale x 4 x i16>, <vscale x 4 x i16>} @llvm.vector.deinterleave2.nxv8i16(<vscale x 8 x i16> %vec)
 ret {<vscale x 4 x i16>, <vscale x 4 x i16>} %retval
 }
 
@@ -240,35 +240,35 @@ define {<vscale x 2 x i32>, <vscale x 2 x i32>} @vector_deinterleave_nxv2i32_nxv
 ; CHECK-NEXT:    uzp1 z0.d, z2.d, z1.d
 ; CHECK-NEXT:    uzp2 z1.d, z2.d, z1.d
 ; CHECK-NEXT:    ret
-%retval = call {<vscale x 2 x i32>,<vscale x 2 x i32>} @llvm.experimental.vector.deinterleave2.nxv4i32(<vscale x 4 x i32> %vec)
+%retval = call {<vscale x 2 x i32>,<vscale x 2 x i32>} @llvm.vector.deinterleave2.nxv4i32(<vscale x 4 x i32> %vec)
 ret {<vscale x 2 x i32>, <vscale x 2 x i32>} %retval
 }
 
 
 ; Floating declarations
-declare {<vscale x 2 x half>,<vscale x 2 x half>} @llvm.experimental.vector.deinterleave2.nxv4f16(<vscale x 4 x half>)
-declare {<vscale x 4 x half>, <vscale x 4 x half>} @llvm.experimental.vector.deinterleave2.nxv8f16(<vscale x 8 x half>)
-declare {<vscale x 2 x float>, <vscale x 2 x float>} @llvm.experimental.vector.deinterleave2.nxv4f32(<vscale x 4 x float>)
-declare {<vscale x 8 x half>, <vscale x 8 x half>} @llvm.experimental.vector.deinterleave2.nxv16f16(<vscale x 16 x half>)
-declare {<vscale x 4 x float>, <vscale x 4 x float>} @llvm.experimental.vector.deinterleave2.nxv8f32(<vscale x 8 x float>)
-declare {<vscale x 2 x double>, <vscale x 2 x double>} @llvm.experimental.vector.deinterleave2.nxv4f64(<vscale x 4 x double>)
+declare {<vscale x 2 x half>,<vscale x 2 x half>} @llvm.vector.deinterleave2.nxv4f16(<vscale x 4 x half>)
+declare {<vscale x 4 x half>, <vscale x 4 x half>} @llvm.vector.deinterleave2.nxv8f16(<vscale x 8 x half>)
+declare {<vscale x 2 x float>, <vscale x 2 x float>} @llvm.vector.deinterleave2.nxv4f32(<vscale x 4 x float>)
+declare {<vscale x 8 x half>, <vscale x 8 x half>} @llvm.vector.deinterleave2.nxv16f16(<vscale x 16 x half>)
+declare {<vscale x 4 x float>, <vscale x 4 x float>} @llvm.vector.deinterleave2.nxv8f32(<vscale x 8 x float>)
+declare {<vscale x 2 x double>, <vscale x 2 x double>} @llvm.vector.deinterleave2.nxv4f64(<vscale x 4 x double>)
 
 ; Integer declarations
-declare {<vscale x 16 x i8>, <vscale x 16 x i8>} @llvm.experimental.vector.deinterleave2.nxv32i8(<vscale x 32 x i8>)
-declare {<vscale x 8 x i16>, <vscale x 8 x i16>} @llvm.experimental.vector.deinterleave2.nxv16i16(<vscale x 16 x i16>)
-declare {<vscale x 4 x i32>, <vscale x 4 x i32>} @llvm.experimental.vector.deinterleave2.nxv8i32(<vscale x 8 x i32>)
-declare {<vscale x 2 x i64>, <vscale x 2 x i64>} @llvm.experimental.vector.deinterleave2.nxv4i64(<vscale x 4 x i64>)
+declare {<vscale x 16 x i8>, <vscale x 16 x i8>} @llvm.vector.deinterleave2.nxv32i8(<vscale x 32 x i8>)
+declare {<vscale x 8 x i16>, <vscale x 8 x i16>} @llvm.vector.deinterleave2.nxv16i16(<vscale x 16 x i16>)
+declare {<vscale x 4 x i32>, <vscale x 4 x i32>} @llvm.vector.deinterleave2.nxv8i32(<vscale x 8 x i32>)
+declare {<vscale x 2 x i64>, <vscale x 2 x i64>} @llvm.vector.deinterleave2.nxv4i64(<vscale x 4 x i64>)
 
 ; Predicated declarations
-declare {<vscale x 16 x i1>, <vscale x 16 x i1>} @llvm.experimental.vector.deinterleave2.nxv32i1(<vscale x 32 x i1>)
-declare {<vscale x 8 x i1>, <vscale x 8 x i1>} @llvm.experimental.vector.deinterleave2.nxv16i1(<vscale x 16 x i1>)
-declare {<vscale x 4 x i1>, <vscale x 4 x i1>} @llvm.experimental.vector.deinterleave2.nxv8i1(<vscale x 8 x i1>)
-declare {<vscale x 2 x i1>, <vscale x 2 x i1>} @llvm.experimental.vector.deinterleave2.nxv4i1(<vscale x 4 x i1>)
+declare {<vscale x 16 x i1>, <vscale x 16 x i1>} @llvm.vector.deinterleave2.nxv32i1(<vscale x 32 x i1>)
+declare {<vscale x 8 x i1>, <vscale x 8 x i1>} @llvm.vector.deinterleave2.nxv16i1(<vscale x 16 x i1>)
+declare {<vscale x 4 x i1>, <vscale x 4 x i1>} @llvm.vector.deinterleave2.nxv8i1(<vscale x 8 x i1>)
+declare {<vscale x 2 x i1>, <vscale x 2 x i1>} @llvm.vector.deinterleave2.nxv4i1(<vscale x 4 x i1>)
 
 ; Illegal size type
-declare {<vscale x 4 x i64>, <vscale x 4 x i64>} @llvm.experimental.vector.deinterleave2.nxv8i64(<vscale x 8 x i64>)
-declare {<vscale x 8 x i64>, <vscale x 8 x i64>} @llvm.experimental.vector.deinterleave2.nxv16i64(<vscale x 16 x i64>)
+declare {<vscale x 4 x i64>, <vscale x 4 x i64>} @llvm.vector.deinterleave2.nxv8i64(<vscale x 8 x i64>)
+declare {<vscale x 8 x i64>, <vscale x 8 x i64>} @llvm.vector.deinterleave2.nxv16i64(<vscale x 16 x i64>)
 
-declare {<vscale x 8 x i8>, <vscale x 8 x i8>} @llvm.experimental.vector.deinterleave2.nxv16i8(<vscale x 16 x i8>)
-declare {<vscale x 4 x i16>, <vscale x 4 x i16>} @llvm.experimental.vector.deinterleave2.nxv8i16(<vscale x 8 x i16>)
-declare {<vscale x 2 x i32>, <vscale x 2 x i32>} @llvm.experimental.vector.deinterleave2.nxv4i32(<vscale x 4 x i32>)
+declare {<vscale x 8 x i8>, <vscale x 8 x i8>} @llvm.vector.deinterleave2.nxv16i8(<vscale x 16 x i8>)
+declare {<vscale x 4 x i16>, <vscale x 4 x i16>} @llvm.vector.deinterleave2.nxv8i16(<vscale x 8 x i16>)
+declare {<vscale x 2 x i32>, <vscale x 2 x i32>} @llvm.vector.deinterleave2.nxv4i32(<vscale x 4 x i32>)
diff --git a/llvm/test/CodeGen/AArch64/sve-vector-interleave.ll b/llvm/test/CodeGen/AArch64/sve-vector-interleave.ll
index 23bf5065286e..e2c3b0abe21a 100644
--- a/llvm/test/CodeGen/AArch64/sve-vector-interleave.ll
+++ b/llvm/test/CodeGen/AArch64/sve-vector-interleave.ll
@@ -8,7 +8,7 @@ define <vscale x 4 x half> @interleave2_nxv4f16(<vscale x 2 x half> %vec0, <vsca
 ; CHECK-NEXT:    zip1 z0.d, z0.d, z1.d
 ; CHECK-NEXT:    uzp1 z0.s, z0.s, z2.s
 ; CHECK-NEXT:    ret
-  %retval = call <vscale x 4 x half> @llvm.experimental.vector.interleave2.nxv4f16(<vscale x 2 x half> %vec0, <vscale x 2 x half> %vec1)
+  %retval = call <vscale x 4 x half> @llvm.vector.interleave2.nxv4f16(<vscale x 2 x half> %vec0, <vscale x 2 x half> %vec1)
   ret  <vscale x 4 x half>   %retval
 }
 
@@ -19,7 +19,7 @@ define <vscale x 8 x half> @interleave2_nxv8f16(<vscale x 4 x half> %vec0, <vsca
 ; CHECK-NEXT:    zip1 z0.s, z0.s, z1.s
 ; CHECK-NEXT:    uzp1 z0.h, z0.h, z2.h
 ; CHECK-NEXT:    ret
-  %retval = call <vscale x 8 x half> @llvm.experimental.vector.interleave2.nxv8f16(<vscale x 4 x half> %vec0, <vscale x 4 x half> %vec1)
+  %retval = call <vscale x 8 x half> @llvm.vector.interleave2.nxv8f16(<vscale x 4 x half> %vec0, <vscale x 4 x half> %vec1)
   ret  <vscale x 8 x half>   %retval
 }
 
@@ -30,7 +30,7 @@ define <vscale x 16 x half> @interleave2_nxv16f16(<vscale x 8 x half> %vec0, <vs
 ; CHECK-NEXT:    zip2 z1.h, z0.h, z1.h
 ; CHECK-NEXT:    mov z0.d, z2.d
 ; CHECK-NEXT:    ret
-  %retval = call <vscale x 16 x half> @llvm.experimental.vector.interleave2.nxv16f16(<vscale x 8 x half> %vec0, <vscale x 8 x half> %vec1)
+  %retval = call <vscale x 16 x half> @llvm.vector.interleave2.nxv16f16(<vscale x 8 x half> %vec0, <vscale x 8 x half> %vec1)
   ret  <vscale x 16 x half>   %retval
 }
 
@@ -41,7 +41,7 @@ define <vscale x 4 x float> @interleave2_nxv4f32(<vscale x 2 x float> %vec0, <vs
 ; CHECK-NEXT:    zip1 z0.d, z0.d, z1.d
 ; CHECK-NEXT:    uzp1 z0.s, z0.s, z2.s
 ; CHECK-NEXT:    ret
-  %retval = call <vscale x 4 x float> @llvm.experimental.vector.interleave2.nxv4f32(<vscale x 2 x float> %vec0, <vscale x 2 x float> %vec1)
+  %retval = call <vscale x 4 x float> @llvm.vector.interleave2.nxv4f32(<vscale x 2 x float> %vec0, <vscale x 2 x float> %vec1)
   ret  <vscale x 4 x float>   %retval
 }
 
@@ -52,7 +52,7 @@ define <vscale x 8 x float> @interleave2_nxv8f32(<vscale x 4 x float> %vec0, <vs
 ; CHECK-NEXT:    zip2 z1.s, z0.s, z1.s
 ; CHECK-NEXT:    mov z0.d, z2.d
 ; CHECK-NEXT:    ret
-  %retval = call <vscale x 8 x float> @llvm.experimental.vector.interleave2.nxv8f32(<vscale x 4 x float> %vec0, <vscale x 4 x float> %vec1)
+  %retval = call <vscale x 8 x float> @llvm.vector.interleave2.nxv8f32(<vscale x 4 x float> %vec0, <vscale x 4 x float> %vec1)
   ret  <vscale x 8 x float>   %retval
 }
 
@@ -63,7 +63,7 @@ define <vscale x 4 x double> @interleave2_nxv4f64(<vscale x 2 x double> %vec0, <
 ; CHECK-NEXT:    zip2 z1.d, z0.d, z1.d
 ; CHECK-NEXT:    mov z0.d, z2.d
 ; CHECK-NEXT:    ret
-  %retval = call <vscale x 4 x double>@llvm.experimental.vector.interleave2.nxv4f64(<vscale x 2 x double> %vec0, <vscale x 2 x double> %vec1)
+  %retval = call <vscale x 4 x double>@llvm.vector.interleave2.nxv4f64(<vscale x 2 x double> %vec0, <vscale x 2 x double> %vec1)
   ret  <vscale x 4 x double>   %retval
 }
 
@@ -76,7 +76,7 @@ define <vscale x 32 x i8> @interleave2_nxv32i8(<vscale x 16 x i8> %vec0, <vscale
 ; CHECK-NEXT:    zip2 z1.b, z0.b, z1.b
 ; CHECK-NEXT:    mov z0.d, z2.d
 ; CHECK-NEXT:    ret
-   %retval = call <vscale x 32 x i8> @llvm.experimental.vector.interleave2.nxv32i8(<vscale x 16 x i8> %vec0, <vscale x 16 x i8> %vec1)
+   %retval = call <vscale x 32 x i8> @llvm.vector.interleave2.nxv32i8(<vscale x 16 x i8> %vec0, <vscale x 16 x i8> %vec1)
   ret  <vscale x 32 x i8>   %retval
 }
 
@@ -87,7 +87,7 @@ define <vscale x 16 x i16> @interleave2_nxv16i16(<vscale x 8 x i16> %vec0, <vsca
 ; CHECK-NEXT:    zip2 z1.h, z0.h, z1.h
 ; CHECK-NEXT:    mov z0.d, z2.d
 ; CHECK-NEXT:    ret
-  %retval = call <vscale x 16 x i16> @llvm.experimental.vector.interleave2.nxv16i16(<vscale x 8 x i16> %vec0, <vscale x 8 x i16> %vec1)
+  %retval = call <vscale x 16 x i16> @llvm.vector.interleave2.nxv16i16(<vscale x 8 x i16> %vec0, <vscale x 8 x i16> %vec1)
   ret  <vscale x 16 x i16>   %retval
 }
 
@@ -98,7 +98,7 @@ define <vscale x 8 x i32> @interleave2_nxv8i32(<vscale x 4 x i32> %vec0, <vscale
 ; CHECK-NEXT:    zip2 z1.s, z0.s, z1.s
 ; CHECK-NEXT:    mov z0.d, z2.d
 ; CHECK-NEXT:    ret
-  %retval = call <vscale x 8 x i32> @llvm.experimental.vector.interleave2.nxv8i32(<vscale x 4 x i32> %vec0, <vscale x 4 x i32> %vec1)
+  %retval = call <vscale x 8 x i32> @llvm.vector.interleave2.nxv8i32(<vscale x 4 x i32> %vec0, <vscale x 4 x i32> %vec1)
   ret  <vscale x 8 x i32>   %retval
 }
 
@@ -109,7 +109,7 @@ define <vscale x 4 x i64> @interleave2_nxv4i64(<vscale x 2 x i64> %vec0, <vscale
 ; CHECK-NEXT:    zip2 z1.d, z0.d, z1.d
 ; CHECK-NEXT:    mov z0.d, z2.d
 ; CHECK-NEXT:    ret
-  %retval = call <vscale x 4 x i64> @llvm.experimental.vector.interleave2.nxv4i64(<vscale x 2 x i64> %vec0, <vscale x 2 x i64> %vec1)
+  %retval = call <vscale x 4 x i64> @llvm.vector.interleave2.nxv4i64(<vscale x 2 x i64> %vec0, <vscale x 2 x i64> %vec1)
   ret  <vscale x 4 x i64>   %retval
 }
 
@@ -122,7 +122,7 @@ define <vscale x 32 x i1> @interleave2_nxv32i1(<vscale x 16 x i1> %vec0, <vscale
 ; CHECK-NEXT:    zip2 p1.b, p0.b, p1.b
 ; CHECK-NEXT:    mov p0.b, p2.b
 ; CHECK-NEXT:    ret
-   %retval = call <vscale x 32 x i1> @llvm.experimental.vector.interleave2.nxv32i1(<vscale x 16 x i1> %vec0, <vscale x 16 x i1> %vec1)
+   %retval = call <vscale x 32 x i1> @llvm.vector.interleave2.nxv32i1(<vscale x 16 x i1> %vec0, <vscale x 16 x i1> %vec1)
   ret  <vscale x 32 x i1>   %retval
 }
 
@@ -133,7 +133,7 @@ define <vscale x 16 x i1> @interleave2_nxv16i1(<vscale x 8 x i1> %vec0, <vscale
 ; CHECK-NEXT:    zip1 p0.h, p0.h, p1.h
 ; CHECK-NEXT:    uzp1 p0.b, p0.b, p2.b
 ; CHECK-NEXT:    ret
-  %retval = call <vscale x 16 x i1> @llvm.experimental.vector.interleave2.nxv16i1(<vscale x 8 x i1> %vec0, <vscale x 8 x i1> %vec1)
+  %retval = call <vscale x 16 x i1> @llvm.vector.interleave2.nxv16i1(<vscale x 8 x i1> %vec0, <vscale x 8 x i1> %vec1)
   ret  <vscale x 16 x i1>   %retval
 }
 
@@ -144,7 +144,7 @@ define <vscale x 8 x i1> @interleave2_nxv8i1(<vscale x 4 x i1> %vec0, <vscale x
 ; CHECK-NEXT:    zip1 p0.s, p0.s, p1.s
 ; CHECK-NEXT:    uzp1 p0.h, p0.h, p2.h
 ; CHECK-NEXT:    ret
-  %retval = call <vscale x 8 x i1> @llvm.experimental.vector.interleave2.nxv8i1(<vscale x 4 x i1> %vec0, <vscale x 4 x i1> %vec1)
+  %retval = call <vscale x 8 x i1> @llvm.vector.interleave2.nxv8i1(<vscale x 4 x i1> %vec0, <vscale x 4 x i1> %vec1)
   ret  <vscale x 8 x i1>   %retval
 }
 
@@ -155,7 +155,7 @@ define <vscale x 4 x i1> @interleave2_nxv4i1(<vscale x 2 x i1> %vec0, <vscale x
 ; CHECK-NEXT:    zip1 p0.d, p0.d, p1.d
 ; CHECK-NEXT:    uzp1 p0.s, p0.s, p2.s
 ; CHECK-NEXT:    ret
-  %retval = call <vscale x 4 x i1> @llvm.experimental.vector.interleave2.nxv4i1(<vscale x 2 x i1> %vec0, <vscale x 2 x i1> %vec1)
+  %retval = call <vscale x 4 x i1> @llvm.vector.interleave2.nxv4i1(<vscale x 2 x i1> %vec0, <vscale x 2 x i1> %vec1)
   ret  <vscale x 4 x i1>   %retval
 }
 
@@ -172,7 +172,7 @@ define <vscale x 16 x i32> @interleave2_nxv16i32(<vscale x 8 x i32> %vec0, <vsca
 ; CHECK-NEXT:    mov z1.d, z2.d
 ; CHECK-NEXT:    mov z2.d, z4.d
 ; CHECK-NEXT:    ret
-  %retval = call <vscale x 16 x i32>@llvm.experimental.vector.interleave2.nxv16i32(<vscale x 8 x i32> %vec0, <vscale x 8 x i32> %vec1)
+  %retval = call <vscale x 16 x i32>@llvm.vector.interleave2.nxv16i32(<vscale x 8 x i32> %vec0, <vscale x 8 x i32> %vec1)
   ret <vscale x 16 x i32> %retval
 }
 
@@ -187,7 +187,7 @@ define <vscale x 8 x i64> @interleave2_nxv8i64(<vscale x 4 x i64> %vec0, <vscale
 ; CHECK-NEXT:    mov z1.d, z2.d
 ; CHECK-NEXT:    mov z2.d, z4.d
 ; CHECK-NEXT:    ret
-  %retval = call <vscale x 8 x i64> @llvm.experimental.vector.interleave2.nxv8i64(<vscale x 4 x i64> %vec0, <vscale x 4 x i64> %vec1)
+  %retval = call <vscale x 8 x i64> @llvm.vector.interleave2.nxv8i64(<vscale x 4 x i64> %vec0, <vscale x 4 x i64> %vec1)
   ret <vscale x 8 x i64> %retval
 }
 
@@ -200,7 +200,7 @@ define <vscale x 16 x i8> @interleave2_nxv8i8(<vscale x 8 x i8> %vec0, <vscale x
 ; CHECK-NEXT:    zip1 z0.h, z0.h, z1.h
 ; CHECK-NEXT:    uzp1 z0.b, z0.b, z2.b
 ; CHECK-NEXT:    ret
-  %retval = call <vscale x 16 x i8> @llvm.experimental.vector.interleave2.nxv16i8(<vscale x 8 x i8> %vec0, <vscale x 8 x i8> %vec1)
+  %retval = call <vscale x 16 x i8> @llvm.vector.interleave2.nxv16i8(<vscale x 8 x i8> %vec0, <vscale x 8 x i8> %vec1)
   ret <vscale x 16 x i8> %retval
 }
 
@@ -211,7 +211,7 @@ define <vscale x 8 x i16> @interleave2_nxv4i16(<vscale x 4 x i16> %vec0, <vscale
 ; CHECK-NEXT:    zip1 z0.s, z0.s, z1.s
 ; CHECK-NEXT:    uzp1 z0.h, z0.h, z2.h
 ; CHECK-NEXT:    ret
-  %retval = call <vscale x 8 x i16> @llvm.experimental.vector.interleave2.nxv8i16(<vscale x 4 x i16> %vec0, <vscale x 4 x i16> %vec1)
+  %retval = call <vscale x 8 x i16> @llvm.vector.interleave2.nxv8i16(<vscale x 4 x i16> %vec0, <vscale x 4 x i16> %vec1)
   ret <vscale x 8 x i16> %retval
 }
 
@@ -222,34 +222,34 @@ define <vscale x 4 x i32> @interleave2_nxv2i32(<vscale x 2 x i32> %vec0, <vscale
 ; CHECK-NEXT:    zip1 z0.d, z0.d, z1.d
 ; CHECK-NEXT:    uzp1 z0.s, z0.s, z2.s
 ; CHECK-NEXT:    ret
-  %retval = call <vscale x 4 x i32> @llvm.experimental.vector.interleave2.nxv4i32(<vscale x 2 x i32> %vec0, <vscale x 2 x i32> %vec1)
+  %retval = call <vscale x 4 x i32> @llvm.vector.interleave2.nxv4i32(<vscale x 2 x i32> %vec0, <vscale x 2 x i32> %vec1)
   ret <vscale x 4 x i32> %retval
 }
 
 ; Float declarations
-declare <vscale x 4 x half> @llvm.experimental.vector.interleave2.nxv4f16(<vscale x 2 x half>, <vscale x 2 x half>)
-declare <vscale x 8 x half> @llvm.experimental.vector.interleave2.nxv8f16(<vscale x 4 x half>, <vscale x 4 x half>)
-declare <vscale x 16 x half> @llvm.experimental.vector.interleave2.nxv16f16(<vscale x 8 x half>, <vscale x 8 x half>)
-declare <vscale x 4 x float> @llvm.experimental.vector.interleave2.nxv4f32(<vscale x 2 x float>, <vscale x 2 x float>)
-declare <vscale x 8 x float> @llvm.experimental.vector.interleave2.nxv8f32(<vscale x 4 x float>, <vscale x 4 x float>)
-declare <vscale x 4 x double> @llvm.experimental.vector.interleave2.nxv4f64(<vscale x 2 x double>, <vscale x 2 x double>)
+declare <vscale x 4 x half> @llvm.vector.interleave2.nxv4f16(<vscale x 2 x half>, <vscale x 2 x half>)
+declare <vscale x 8 x half> @llvm.vector.interleave2.nxv8f16(<vscale x 4 x half>, <vscale x 4 x half>)
+declare <vscale x 16 x half> @llvm.vector.interleave2.nxv16f16(<vscale x 8 x half>, <vscale x 8 x half>)
+declare <vscale x 4 x float> @llvm.vector.interleave2.nxv4f32(<vscale x 2 x float>, <vscale x 2 x float>)
+declare <vscale x 8 x float> @llvm.vector.interleave2.nxv8f32(<vscale x 4 x float>, <vscale x 4 x float>)
+declare <vscale x 4 x double> @llvm.vector.interleave2.nxv4f64(<vscale x 2 x double>, <vscale x 2 x double>)
 
 ; Integer declarations
-declare <vscale x 32 x i8> @llvm.experimental.vector.interleave2.nxv32i8(<vscale x 16 x i8>, <vscale x 16 x i8>)
-declare <vscale x 16 x i16> @llvm.experimental.vector.interleave2.nxv16i16(<vscale x 8 x i16>, <vscale x 8 x i16>)
-declare <vscale x 8 x i32> @llvm.experimental.vector.interleave2.nxv8i32(<vscale x 4 x i32>, <vscale x 4 x i32>)
-declare <vscale x 4 x i64> @llvm.experimental.vector.interleave2.nxv4i64(<vscale x 2 x i64>, <vscale x 2 x i64>)
+declare <vscale x 32 x i8> @llvm.vector.interleave2.nxv32i8(<vscale x 16 x i8>, <vscale x 16 x i8>)
+declare <vscale x 16 x i16> @llvm.vector.interleave2.nxv16i16(<vscale x 8 x i16>, <vscale x 8 x i16>)
+declare <vscale x 8 x i32> @llvm.vector.interleave2.nxv8i32(<vscale x 4 x i32>, <vscale x 4 x i32>)
+declare <vscale x 4 x i64> @llvm.vector.interleave2.nxv4i64(<vscale x 2 x i64>, <vscale x 2 x i64>)
 
 ; Predicated
-declare <vscale x 32 x i1> @llvm.experimental.vector.interleave2.nxv32i1(<vscale x 16 x i1>, <vscale x 16 x i1>)
-declare <vscale x 16 x i1> @llvm.experimental.vector.interleave2.nxv16i1(<vscale x 8 x i1>, <vscale x 8 x i1>)
-declare <vscale x 8 x i1> @llvm.experimental.vector.interleave2.nxv8i1(<vscale x 4 x i1>, <vscale x 4 x i1>)
-declare <vscale x 4 x i1> @llvm.experimental.vector.interleave2.nxv4i1(<vscale x 2 x i1>, <vscale x 2 x i1>)
+declare <vscale x 32 x i1> @llvm.vector.interleave2.nxv32i1(<vscale x 16 x i1>, <vscale x 16 x i1>)
+declare <vscale x 16 x i1> @llvm.vector.interleave2.nxv16i1(<vscale x 8 x i1>, <vscale x 8 x i1>)
+declare <vscale x 8 x i1> @llvm.vector.interleave2.nxv8i1(<vscale x 4 x i1>, <vscale x 4 x i1>)
+declare <vscale x 4 x i1> @llvm.vector.interleave2.nxv4i1(<vscale x 2 x i1>, <vscale x 2 x i1>)
 
 ; Illegal type size
-declare <vscale x 16 x i32> @llvm.experimental.vector.interleave2.nxv16i32(<vscale x 8 x i32>, <vscale x 8 x i32>)
-declare <vscale x 8 x i64> @llvm.experimental.vector.interleave2.nxv8i64(<vscale x 4 x i64>, <vscale x 4 x i64>)
+declare <vscale x 16 x i32> @llvm.vector.interleave2.nxv16i32(<vscale x 8 x i32>, <vscale x 8 x i32>)
+declare <vscale x 8 x i64> @llvm.vector.interleave2.nxv8i64(<vscale x 4 x i64>, <vscale x 4 x i64>)
 
-declare <vscale x 16 x i8> @llvm.experimental.vector.interleave2.nxv16i8(<vscale x 8 x i8>, <vscale x 8 x i8>)
-declare <vscale x 8 x i16> @llvm.experimental.vector.interleave2.nxv8i16(<vscale x 4 x i16>, <vscale x 4 x i16>)
-declare <vscale x 4 x i32> @llvm.experimental.vector.interleave2.nxv4i32(<vscale x 2 x i32>, <vscale x 2 x i32>)
+declare <vscale x 16 x i8> @llvm.vector.interleave2.nxv16i8(<vscale x 8 x i8>, <vscale x 8 x i8>)
+declare <vscale x 8 x i16> @llvm.vector.interleave2.nxv8i16(<vscale x 4 x i16>, <vscale x 4 x i16>)
+declare <vscale x 4 x i32> @llvm.vector.interleave2.nxv4i32(<vscale x 2 x i32>, <vscale x 2 x i32>)
diff --git a/llvm/test/CodeGen/AArch64/sve-vl-arith.ll b/llvm/test/CodeGen/AArch64/sve-vl-arith.ll
index dd4294c8d3bd..de2af590acd1 100644
--- a/llvm/test/CodeGen/AArch64/sve-vl-arith.ll
+++ b/llvm/test/CodeGen/AArch64/sve-vl-arith.ll
@@ -186,8 +186,8 @@ define i64 @incd_scalar_i64(i64 %a) {
 define i64 @decb_scalar_i64(i64 %a) {
 ; NO_SCALAR_INC-LABEL: decb_scalar_i64:
 ; NO_SCALAR_INC:       // %bb.0:
-; NO_SCALAR_INC-NEXT:    rdvl x8, #-2
-; NO_SCALAR_INC-NEXT:    add x0, x0, x8
+; NO_SCALAR_INC-NEXT:    cnth x8, all, mul #4
+; NO_SCALAR_INC-NEXT:    sub x0, x0, x8
 ; NO_SCALAR_INC-NEXT:    ret
 ;
 ; CHECK-LABEL: decb_scalar_i64:
@@ -204,8 +204,7 @@ define i64 @dech_scalar_i64(i64 %a) {
 ; NO_SCALAR_INC-LABEL: dech_scalar_i64:
 ; NO_SCALAR_INC:       // %bb.0:
 ; NO_SCALAR_INC-NEXT:    cnth x8, all, mul #3
-; NO_SCALAR_INC-NEXT:    neg x8, x8
-; NO_SCALAR_INC-NEXT:    add x0, x0, x8
+; NO_SCALAR_INC-NEXT:    sub x0, x0, x8
 ; NO_SCALAR_INC-NEXT:    ret
 ;
 ; CHECK-LABEL: dech_scalar_i64:
@@ -222,8 +221,7 @@ define i64 @decw_scalar_i64(i64 %a) {
 ; NO_SCALAR_INC-LABEL: decw_scalar_i64:
 ; NO_SCALAR_INC:       // %bb.0:
 ; NO_SCALAR_INC-NEXT:    cntw x8, all, mul #3
-; NO_SCALAR_INC-NEXT:    neg x8, x8
-; NO_SCALAR_INC-NEXT:    add x0, x0, x8
+; NO_SCALAR_INC-NEXT:    sub x0, x0, x8
 ; NO_SCALAR_INC-NEXT:    ret
 ;
 ; CHECK-LABEL: decw_scalar_i64:
@@ -240,8 +238,7 @@ define i64 @decd_scalar_i64(i64 %a) {
 ; NO_SCALAR_INC-LABEL: decd_scalar_i64:
 ; NO_SCALAR_INC:       // %bb.0:
 ; NO_SCALAR_INC-NEXT:    cntd x8, all, mul #3
-; NO_SCALAR_INC-NEXT:    neg x8, x8
-; NO_SCALAR_INC-NEXT:    add x0, x0, x8
+; NO_SCALAR_INC-NEXT:    sub x0, x0, x8
 ; NO_SCALAR_INC-NEXT:    ret
 ;
 ; CHECK-LABEL: decd_scalar_i64:
@@ -345,8 +342,8 @@ define i32 @incd_scalar_i32(i32 %a) {
 define i32 @decb_scalar_i32(i32 %a) {
 ; NO_SCALAR_INC-LABEL: decb_scalar_i32:
 ; NO_SCALAR_INC:       // %bb.0:
-; NO_SCALAR_INC-NEXT:    rdvl x8, #-4
-; NO_SCALAR_INC-NEXT:    add w0, w0, w8
+; NO_SCALAR_INC-NEXT:    cnth x8, all, mul #8
+; NO_SCALAR_INC-NEXT:    sub w0, w0, w8
 ; NO_SCALAR_INC-NEXT:    ret
 ;
 ; CHECK-LABEL: decb_scalar_i32:
@@ -367,8 +364,7 @@ define i32 @dech_scalar_i32(i32 %a) {
 ; NO_SCALAR_INC-LABEL: dech_scalar_i32:
 ; NO_SCALAR_INC:       // %bb.0:
 ; NO_SCALAR_INC-NEXT:    cnth x8
-; NO_SCALAR_INC-NEXT:    neg x8, x8
-; NO_SCALAR_INC-NEXT:    add w0, w0, w8
+; NO_SCALAR_INC-NEXT:    sub w0, w0, w8
 ; NO_SCALAR_INC-NEXT:    ret
 ;
 ; CHECK-LABEL: dech_scalar_i32:
@@ -389,8 +385,7 @@ define i32 @decw_scalar_i32(i32 %a) {
 ; NO_SCALAR_INC-LABEL: decw_scalar_i32:
 ; NO_SCALAR_INC:       // %bb.0:
 ; NO_SCALAR_INC-NEXT:    cntw x8
-; NO_SCALAR_INC-NEXT:    neg x8, x8
-; NO_SCALAR_INC-NEXT:    add w0, w0, w8
+; NO_SCALAR_INC-NEXT:    sub w0, w0, w8
 ; NO_SCALAR_INC-NEXT:    ret
 ;
 ; CHECK-LABEL: decw_scalar_i32:
@@ -411,8 +406,7 @@ define i32 @decd_scalar_i32(i32 %a) {
 ; NO_SCALAR_INC-LABEL: decd_scalar_i32:
 ; NO_SCALAR_INC:       // %bb.0:
 ; NO_SCALAR_INC-NEXT:    cntd x8
-; NO_SCALAR_INC-NEXT:    neg x8, x8
-; NO_SCALAR_INC-NEXT:    add w0, w0, w8
+; NO_SCALAR_INC-NEXT:    sub w0, w0, w8
 ; NO_SCALAR_INC-NEXT:    ret
 ;
 ; CHECK-LABEL: decd_scalar_i32:
diff --git a/llvm/test/CodeGen/AArch64/sve2-bsl.ll b/llvm/test/CodeGen/AArch64/sve2-bsl.ll
index 23b2622f5f58..ef7d4abe5c5f 100644
--- a/llvm/test/CodeGen/AArch64/sve2-bsl.ll
+++ b/llvm/test/CodeGen/AArch64/sve2-bsl.ll
@@ -41,3 +41,55 @@ define <vscale x 4 x i32> @no_bsl_fold(<vscale x 4 x i32> %a, <vscale x 4 x i32>
   %c = or <vscale x 4 x i32> %1, %2
   ret <vscale x 4 x i32> %c
 }
+
+define <vscale x 16 x i8> @nbsl_i8(<vscale x 16 x i8> %a, <vscale x 16 x i8> %b) {
+; CHECK-LABEL: nbsl_i8:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    mov z2.b, #127 // =0x7f
+; CHECK-NEXT:    nbsl z0.d, z0.d, z1.d, z2.d
+; CHECK-NEXT:    ret
+  %1 = and <vscale x 16 x i8> %a, splat(i8 127)
+  %2 = and <vscale x 16 x i8> %b, splat(i8 -128)
+  %3 = or <vscale x 16 x i8> %1, %2
+  %4 = xor <vscale x 16 x i8> %3, splat(i8 -1)
+  ret <vscale x 16 x i8> %4
+}
+
+define <vscale x 8 x i16> @nbsl_i16(<vscale x 8 x i16> %a, <vscale x 8 x i16> %b) {
+; CHECK-LABEL: nbsl_i16:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    mov z2.h, #32767 // =0x7fff
+; CHECK-NEXT:    nbsl z0.d, z0.d, z1.d, z2.d
+; CHECK-NEXT:    ret
+  %1 = and <vscale x 8 x i16> %a, splat(i16 32767)
+  %2 = and <vscale x 8 x i16> %b, splat(i16 -32768)
+  %3 = or <vscale x 8 x i16> %1, %2
+  %4 = xor <vscale x 8 x i16> %3, splat(i16 -1)
+  ret <vscale x 8 x i16> %4
+}
+
+define <vscale x 4 x i32> @nbsl_i32(<vscale x 4 x i32> %a, <vscale x 4 x i32> %b) {
+; CHECK-LABEL: nbsl_i32:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    mov z2.s, #0x7fffffff
+; CHECK-NEXT:    nbsl z0.d, z0.d, z1.d, z2.d
+; CHECK-NEXT:    ret
+  %1 = and <vscale x 4 x i32> %a, splat(i32 2147483647)
+  %2 = and <vscale x 4 x i32> %b, splat(i32 -2147483648)
+  %3 = or <vscale x 4 x i32> %1, %2
+  %4 = xor <vscale x 4 x i32> %3, splat(i32 -1)
+  ret <vscale x 4 x i32> %4
+}
+
+define <vscale x 2 x i64> @nbsl_i64(<vscale x 2 x i64> %a, <vscale x 2 x i64> %b) {
+; CHECK-LABEL: nbsl_i64:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    mov z2.d, #0x7fffffffffffffff
+; CHECK-NEXT:    nbsl z0.d, z0.d, z1.d, z2.d
+; CHECK-NEXT:    ret
+  %1 = and <vscale x 2 x i64> %a, splat(i64 9223372036854775807)
+  %2 = and <vscale x 2 x i64> %b, splat(i64 -9223372036854775808)
+  %3 = or <vscale x 2 x i64> %1, %2
+  %4 = xor <vscale x 2 x i64> %3, splat(i64 -1)
+  ret <vscale x 2 x i64> %4
+}
diff --git a/llvm/test/CodeGen/AArch64/sve2-intrinsics-while-reversed.ll b/llvm/test/CodeGen/AArch64/sve2-intrinsics-while-reversed.ll
new file mode 100644
index 000000000000..5f7476397891
--- /dev/null
+++ b/llvm/test/CodeGen/AArch64/sve2-intrinsics-while-reversed.ll
@@ -0,0 +1,741 @@
+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
+; RUN: llc -mattr=+sve2 < %s | FileCheck %s
+; RUN: llc -mattr=+sme < %s | FileCheck %s
+
+target triple = "aarch64-unknown-linux-gnu"
+
+;
+; WHILEGE
+;
+
+; Cannot invert the while predicate when %a is MAX_INT.
+define <vscale x 16 x i1> @whilege_b_ww(i32 %a, i32 %b) {
+; CHECK-LABEL: whilege_b_ww:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    whilele p0.b, w1, w0
+; CHECK-NEXT:    rev p0.b, p0.b
+; CHECK-NEXT:    ret
+  %while = call <vscale x 16 x i1> @llvm.aarch64.sve.whilele.nxv16i1.i32(i32 %b, i32 %a)
+  %while.rev = call <vscale x 16 x i1> @llvm.vector.reverse.nxv16i1(<vscale x 16 x i1> %while)
+  ret <vscale x 16 x i1> %while.rev
+}
+
+; Cannot invert the while predicate when %a is MAX_INT.
+define <vscale x 16 x i1> @whilege_b_xx(i64 %a, i64 %b) {
+; CHECK-LABEL: whilege_b_xx:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    whilele p0.b, x1, x0
+; CHECK-NEXT:    rev p0.b, p0.b
+; CHECK-NEXT:    ret
+  %while = call <vscale x 16 x i1> @llvm.aarch64.sve.whilele.nxv16i1.i64(i64 %b, i64 %a)
+  %while.rev = call <vscale x 16 x i1> @llvm.vector.reverse.nxv16i1(<vscale x 16 x i1> %while)
+  ret <vscale x 16 x i1> %while.rev
+}
+
+; Cannot invert the while predicate when %a is MAX_INT.
+define <vscale x 8 x i1> @whilege_h_ww(i32 %a, i32 %b) {
+; CHECK-LABEL: whilege_h_ww:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    whilele p0.h, w1, w0
+; CHECK-NEXT:    rev p0.h, p0.h
+; CHECK-NEXT:    ret
+  %while = call <vscale x 8 x i1> @llvm.aarch64.sve.whilele.nxv8i1.i32(i32 %b, i32 %a)
+  %while.rev = call <vscale x 8 x i1> @llvm.vector.reverse.nxv8i1(<vscale x 8 x i1> %while)
+  ret <vscale x 8 x i1> %while.rev
+}
+
+; Cannot invert the while predicate when %a is MAX_INT.
+define <vscale x 8 x i1> @whilege_h_xx(i64 %a, i64 %b) {
+; CHECK-LABEL: whilege_h_xx:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    whilele p0.h, x1, x0
+; CHECK-NEXT:    rev p0.h, p0.h
+; CHECK-NEXT:    ret
+  %while = call <vscale x 8 x i1> @llvm.aarch64.sve.whilele.nxv8i1.i64(i64 %b, i64 %a)
+  %while.rev = call <vscale x 8 x i1> @llvm.vector.reverse.nxv8i1(<vscale x 8 x i1> %while)
+  ret <vscale x 8 x i1> %while.rev
+}
+
+; Cannot invert the while predicate when %a is MAX_INT.
+define <vscale x 4 x i1> @whilege_s_ww(i32 %a, i32 %b) {
+; CHECK-LABEL: whilege_s_ww:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    whilele p0.s, w1, w0
+; CHECK-NEXT:    rev p0.s, p0.s
+; CHECK-NEXT:    ret
+  %while = call <vscale x 4 x i1> @llvm.aarch64.sve.whilele.nxv4i1.i32(i32 %b, i32 %a)
+  %while.rev = call <vscale x 4 x i1> @llvm.vector.reverse.nxv4i1(<vscale x 4 x i1> %while)
+  ret <vscale x 4 x i1> %while.rev
+}
+
+; Cannot invert the while predicate when %a is MAX_INT.
+define <vscale x 4 x i1> @whilege_s_xx(i64 %a, i64 %b) {
+; CHECK-LABEL: whilege_s_xx:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    whilele p0.s, x1, x0
+; CHECK-NEXT:    rev p0.s, p0.s
+; CHECK-NEXT:    ret
+  %while = call <vscale x 4 x i1> @llvm.aarch64.sve.whilele.nxv4i1.i64(i64 %b, i64 %a)
+  %while.rev = call <vscale x 4 x i1> @llvm.vector.reverse.nxv4i1(<vscale x 4 x i1> %while)
+  ret <vscale x 4 x i1> %while.rev
+}
+
+; Cannot invert the while predicate when %a is MAX_INT.
+define <vscale x 2 x i1> @whilege_d_ww(i32 %a, i32 %b) {
+; CHECK-LABEL: whilege_d_ww:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    whilele p0.d, w1, w0
+; CHECK-NEXT:    rev p0.d, p0.d
+; CHECK-NEXT:    ret
+  %while = call <vscale x 2 x i1> @llvm.aarch64.sve.whilele.nxv2i1.i32(i32 %b, i32 %a)
+  %while.rev = call <vscale x 2 x i1> @llvm.vector.reverse.nxv2i1(<vscale x 2 x i1> %while)
+  ret <vscale x 2 x i1> %while.rev
+}
+
+; Cannot invert the while predicate when %a is MAX_INT.
+define <vscale x 2 x i1> @whilege_d_xx(i64 %a, i64 %b) {
+; CHECK-LABEL: whilege_d_xx:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    whilele p0.d, x1, x0
+; CHECK-NEXT:    rev p0.d, p0.d
+; CHECK-NEXT:    ret
+  %while = call <vscale x 2 x i1> @llvm.aarch64.sve.whilele.nxv2i1.i64(i64 %b, i64 %a)
+  %while.rev = call <vscale x 2 x i1> @llvm.vector.reverse.nxv2i1(<vscale x 2 x i1> %while)
+  ret <vscale x 2 x i1> %while.rev
+}
+
+;
+; WHILEHS
+;
+
+; Cannot invert the while predicate when %a is MAX_UINT.
+define <vscale x 16 x i1> @whilehs_b_ww(i32 %a, i32 %b) {
+; CHECK-LABEL: whilehs_b_ww:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    whilels p0.b, w1, w0
+; CHECK-NEXT:    rev p0.b, p0.b
+; CHECK-NEXT:    ret
+  %while = call <vscale x 16 x i1> @llvm.aarch64.sve.whilels.nxv16i1.i32(i32 %b, i32 %a)
+  %while.rev = call <vscale x 16 x i1> @llvm.vector.reverse.nxv16i1(<vscale x 16 x i1> %while)
+  ret <vscale x 16 x i1> %while.rev
+}
+
+; Cannot invert the while predicate when %a is MAX_UINT.
+define <vscale x 16 x i1> @whilehs_b_xx(i64 %a, i64 %b) {
+; CHECK-LABEL: whilehs_b_xx:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    whilels p0.b, x1, x0
+; CHECK-NEXT:    rev p0.b, p0.b
+; CHECK-NEXT:    ret
+  %while = call <vscale x 16 x i1> @llvm.aarch64.sve.whilels.nxv16i1.i64(i64 %b, i64 %a)
+  %while.rev = call <vscale x 16 x i1> @llvm.vector.reverse.nxv16i1(<vscale x 16 x i1> %while)
+  ret <vscale x 16 x i1> %while.rev
+}
+
+; Cannot invert the while predicate when %a is MAX_UINT.
+define <vscale x 8 x i1> @whilehs_h_ww(i32 %a, i32 %b) {
+; CHECK-LABEL: whilehs_h_ww:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    whilels p0.h, w1, w0
+; CHECK-NEXT:    rev p0.h, p0.h
+; CHECK-NEXT:    ret
+  %while = call <vscale x 8 x i1> @llvm.aarch64.sve.whilels.nxv8i1.i32(i32 %b, i32 %a)
+  %while.rev = call <vscale x 8 x i1> @llvm.vector.reverse.nxv8i1(<vscale x 8 x i1> %while)
+  ret <vscale x 8 x i1> %while.rev
+}
+
+; Cannot invert the while predicate when %a is MAX_UINT.
+define <vscale x 8 x i1> @whilehs_h_xx(i64 %a, i64 %b) {
+; CHECK-LABEL: whilehs_h_xx:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    whilels p0.h, x1, x0
+; CHECK-NEXT:    rev p0.h, p0.h
+; CHECK-NEXT:    ret
+  %while = call <vscale x 8 x i1> @llvm.aarch64.sve.whilels.nxv8i1.i64(i64 %b, i64 %a)
+  %while.rev = call <vscale x 8 x i1> @llvm.vector.reverse.nxv8i1(<vscale x 8 x i1> %while)
+  ret <vscale x 8 x i1> %while.rev
+}
+
+; Cannot invert the while predicate when %a is MAX_UINT.
+define <vscale x 4 x i1> @whilehs_s_ww(i32 %a, i32 %b) {
+; CHECK-LABEL: whilehs_s_ww:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    whilels p0.s, w1, w0
+; CHECK-NEXT:    rev p0.s, p0.s
+; CHECK-NEXT:    ret
+  %while = call <vscale x 4 x i1> @llvm.aarch64.sve.whilels.nxv4i1.i32(i32 %b, i32 %a)
+  %while.rev = call <vscale x 4 x i1> @llvm.vector.reverse.nxv4i1(<vscale x 4 x i1> %while)
+  ret <vscale x 4 x i1> %while.rev
+}
+
+; Cannot invert the while predicate when %a is MAX_UINT.
+define <vscale x 4 x i1> @whilehs_s_xx(i64 %a, i64 %b) {
+; CHECK-LABEL: whilehs_s_xx:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    whilels p0.s, x1, x0
+; CHECK-NEXT:    rev p0.s, p0.s
+; CHECK-NEXT:    ret
+  %while = call <vscale x 4 x i1> @llvm.aarch64.sve.whilels.nxv4i1.i64(i64 %b, i64 %a)
+  %while.rev = call <vscale x 4 x i1> @llvm.vector.reverse.nxv4i1(<vscale x 4 x i1> %while)
+  ret <vscale x 4 x i1> %while.rev
+}
+
+; Cannot invert the while predicate when %a is MAX_UINT.
+define <vscale x 2 x i1> @whilehs_d_ww(i32 %a, i32 %b) {
+; CHECK-LABEL: whilehs_d_ww:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    whilels p0.d, w1, w0
+; CHECK-NEXT:    rev p0.d, p0.d
+; CHECK-NEXT:    ret
+  %while = call <vscale x 2 x i1> @llvm.aarch64.sve.whilels.nxv2i1.i32(i32 %b, i32 %a)
+  %while.rev = call <vscale x 2 x i1> @llvm.vector.reverse.nxv2i1(<vscale x 2 x i1> %while)
+  ret <vscale x 2 x i1> %while.rev
+}
+
+; Cannot invert the while predicate when %a is MAX_UINT.
+define <vscale x 2 x i1> @whilehs_d_xx(i64 %a, i64 %b) {
+; CHECK-LABEL: whilehs_d_xx:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    whilels p0.d, x1, x0
+; CHECK-NEXT:    rev p0.d, p0.d
+; CHECK-NEXT:    ret
+  %while = call <vscale x 2 x i1> @llvm.aarch64.sve.whilels.nxv2i1.i64(i64 %b, i64 %a)
+  %while.rev = call <vscale x 2 x i1> @llvm.vector.reverse.nxv2i1(<vscale x 2 x i1> %while)
+  ret <vscale x 2 x i1> %while.rev
+}
+
+;
+; WHILEGT
+;
+
+define <vscale x 16 x i1> @whilegt_b_ww(i32 %a, i32 %b) {
+; CHECK-LABEL: whilegt_b_ww:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    whilegt p0.b, w0, w1
+; CHECK-NEXT:    ret
+  %while = call <vscale x 16 x i1> @llvm.aarch64.sve.whilelt.nxv16i1.i32(i32 %b, i32 %a)
+  %while.rev = call <vscale x 16 x i1> @llvm.vector.reverse.nxv16i1(<vscale x 16 x i1> %while)
+  ret <vscale x 16 x i1> %while.rev
+}
+
+define <vscale x 16 x i1> @whilegt_b_xx(i64 %a, i64 %b) {
+; CHECK-LABEL: whilegt_b_xx:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    whilegt p0.b, x0, x1
+; CHECK-NEXT:    ret
+  %while = call <vscale x 16 x i1> @llvm.aarch64.sve.whilelt.nxv16i1.i64(i64 %b, i64 %a)
+  %while.rev = call <vscale x 16 x i1> @llvm.vector.reverse.nxv16i1(<vscale x 16 x i1> %while)
+  ret <vscale x 16 x i1> %while.rev
+}
+
+define <vscale x 8 x i1> @whilegt_h_ww(i32 %a, i32 %b) {
+; CHECK-LABEL: whilegt_h_ww:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    whilegt p0.h, w0, w1
+; CHECK-NEXT:    ret
+  %while = call <vscale x 8 x i1> @llvm.aarch64.sve.whilelt.nxv8i1.i32(i32 %b, i32 %a)
+  %while.rev = call <vscale x 8 x i1> @llvm.vector.reverse.nxv8i1(<vscale x 8 x i1> %while)
+  ret <vscale x 8 x i1> %while.rev
+}
+
+define <vscale x 8 x i1> @whilegt_h_xx(i64 %a, i64 %b) {
+; CHECK-LABEL: whilegt_h_xx:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    whilegt p0.h, x0, x1
+; CHECK-NEXT:    ret
+  %while = call <vscale x 8 x i1> @llvm.aarch64.sve.whilelt.nxv8i1.i64(i64 %b, i64 %a)
+  %while.rev = call <vscale x 8 x i1> @llvm.vector.reverse.nxv8i1(<vscale x 8 x i1> %while)
+  ret <vscale x 8 x i1> %while.rev
+}
+
+define <vscale x 4 x i1> @whilegt_s_ww(i32 %a, i32 %b) {
+; CHECK-LABEL: whilegt_s_ww:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    whilegt p0.s, w0, w1
+; CHECK-NEXT:    ret
+  %while = call <vscale x 4 x i1> @llvm.aarch64.sve.whilelt.nxv4i1.i32(i32 %b, i32 %a)
+  %while.rev = call <vscale x 4 x i1> @llvm.vector.reverse.nxv4i1(<vscale x 4 x i1> %while)
+  ret <vscale x 4 x i1> %while.rev
+}
+
+define <vscale x 4 x i1> @whilegt_s_xx(i64 %a, i64 %b) {
+; CHECK-LABEL: whilegt_s_xx:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    whilegt p0.s, x0, x1
+; CHECK-NEXT:    ret
+  %while = call <vscale x 4 x i1> @llvm.aarch64.sve.whilelt.nxv4i1.i64(i64 %b, i64 %a)
+  %while.rev = call <vscale x 4 x i1> @llvm.vector.reverse.nxv4i1(<vscale x 4 x i1> %while)
+  ret <vscale x 4 x i1> %while.rev
+}
+
+define <vscale x 2 x i1> @whilegt_d_ww(i32 %a, i32 %b) {
+; CHECK-LABEL: whilegt_d_ww:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    whilegt p0.d, w0, w1
+; CHECK-NEXT:    ret
+  %while = call <vscale x 2 x i1> @llvm.aarch64.sve.whilelt.nxv2i1.i32(i32 %b, i32 %a)
+  %while.rev = call <vscale x 2 x i1> @llvm.vector.reverse.nxv2i1(<vscale x 2 x i1> %while)
+  ret <vscale x 2 x i1> %while.rev
+}
+
+define <vscale x 2 x i1> @whilegt_d_xx(i64 %a, i64 %b) {
+; CHECK-LABEL: whilegt_d_xx:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    whilegt p0.d, x0, x1
+; CHECK-NEXT:    ret
+  %while = call <vscale x 2 x i1> @llvm.aarch64.sve.whilelt.nxv2i1.i64(i64 %b, i64 %a)
+  %while.rev = call <vscale x 2 x i1> @llvm.vector.reverse.nxv2i1(<vscale x 2 x i1> %while)
+  ret <vscale x 2 x i1> %while.rev
+}
+
+;
+; WHILEHI
+;
+
+define <vscale x 16 x i1> @whilehi_b_ww(i32 %a, i32 %b) {
+; CHECK-LABEL: whilehi_b_ww:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    whilehi p0.b, w0, w1
+; CHECK-NEXT:    ret
+  %while = call <vscale x 16 x i1> @llvm.aarch64.sve.whilelo.nxv16i1.i32(i32 %b, i32 %a)
+  %while.rev = call <vscale x 16 x i1> @llvm.vector.reverse.nxv16i1(<vscale x 16 x i1> %while)
+  ret <vscale x 16 x i1> %while.rev
+}
+
+define <vscale x 16 x i1> @whilehi_b_xx(i64 %a, i64 %b) {
+; CHECK-LABEL: whilehi_b_xx:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    whilehi p0.b, x0, x1
+; CHECK-NEXT:    ret
+  %while = call <vscale x 16 x i1> @llvm.aarch64.sve.whilelo.nxv16i1.i64(i64 %b, i64 %a)
+  %while.rev = call <vscale x 16 x i1> @llvm.vector.reverse.nxv16i1(<vscale x 16 x i1> %while)
+  ret <vscale x 16 x i1> %while.rev
+}
+
+define <vscale x 8 x i1> @whilehi_h_ww(i32 %a, i32 %b) {
+; CHECK-LABEL: whilehi_h_ww:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    whilehi p0.h, w0, w1
+; CHECK-NEXT:    ret
+  %while = call <vscale x 8 x i1> @llvm.aarch64.sve.whilelo.nxv8i1.i32(i32 %b, i32 %a)
+  %while.rev = call <vscale x 8 x i1> @llvm.vector.reverse.nxv8i1(<vscale x 8 x i1> %while)
+  ret <vscale x 8 x i1> %while.rev
+}
+
+define <vscale x 8 x i1> @whilehi_h_xx(i64 %a, i64 %b) {
+; CHECK-LABEL: whilehi_h_xx:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    whilehi p0.h, x0, x1
+; CHECK-NEXT:    ret
+  %while = call <vscale x 8 x i1> @llvm.aarch64.sve.whilelo.nxv8i1.i64(i64 %b, i64 %a)
+  %while.rev = call <vscale x 8 x i1> @llvm.vector.reverse.nxv8i1(<vscale x 8 x i1> %while)
+  ret <vscale x 8 x i1> %while.rev
+}
+
+define <vscale x 4 x i1> @whilehi_s_ww(i32 %a, i32 %b) {
+; CHECK-LABEL: whilehi_s_ww:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    whilehi p0.s, w0, w1
+; CHECK-NEXT:    ret
+  %while = call <vscale x 4 x i1> @llvm.aarch64.sve.whilelo.nxv4i1.i32(i32 %b, i32 %a)
+  %while.rev = call <vscale x 4 x i1> @llvm.vector.reverse.nxv4i1(<vscale x 4 x i1> %while)
+  ret <vscale x 4 x i1> %while.rev
+}
+
+define <vscale x 4 x i1> @whilehi_s_xx(i64 %a, i64 %b) {
+; CHECK-LABEL: whilehi_s_xx:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    whilehi p0.s, x0, x1
+; CHECK-NEXT:    ret
+  %while = call <vscale x 4 x i1> @llvm.aarch64.sve.whilelo.nxv4i1.i64(i64 %b, i64 %a)
+  %while.rev = call <vscale x 4 x i1> @llvm.vector.reverse.nxv4i1(<vscale x 4 x i1> %while)
+  ret <vscale x 4 x i1> %while.rev
+}
+
+define <vscale x 2 x i1> @whilehi_d_ww(i32 %a, i32 %b) {
+; CHECK-LABEL: whilehi_d_ww:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    whilehi p0.d, w0, w1
+; CHECK-NEXT:    ret
+  %while = call <vscale x 2 x i1> @llvm.aarch64.sve.whilelo.nxv2i1.i32(i32 %b, i32 %a)
+  %while.rev = call <vscale x 2 x i1> @llvm.vector.reverse.nxv2i1(<vscale x 2 x i1> %while)
+  ret <vscale x 2 x i1> %while.rev
+}
+
+define <vscale x 2 x i1> @whilehi_d_xx(i64 %a, i64 %b) {
+; CHECK-LABEL: whilehi_d_xx:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    whilehi p0.d, x0, x1
+; CHECK-NEXT:    ret
+  %while = call <vscale x 2 x i1> @llvm.aarch64.sve.whilelo.nxv2i1.i64(i64 %b, i64 %a)
+  %while.rev = call <vscale x 2 x i1> @llvm.vector.reverse.nxv2i1(<vscale x 2 x i1> %while)
+  ret <vscale x 2 x i1> %while.rev
+}
+
+;
+; WHILELE
+;
+
+; Cannot invert the while predicate when %a is MIN_INT.
+define <vscale x 16 x i1> @whilele_b_ww(i32 %a, i32 %b) {
+; CHECK-LABEL: whilele_b_ww:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    whilege p0.b, w1, w0
+; CHECK-NEXT:    rev p0.b, p0.b
+; CHECK-NEXT:    ret
+  %while = call <vscale x 16 x i1> @llvm.aarch64.sve.whilege.nxv16i1.i32(i32 %b, i32 %a)
+  %while.rev = call <vscale x 16 x i1> @llvm.vector.reverse.nxv16i1(<vscale x 16 x i1> %while)
+  ret <vscale x 16 x i1> %while.rev
+}
+
+; Cannot invert the while predicate when %a is MIN_INT.
+define <vscale x 16 x i1> @whilele_b_xx(i64 %a, i64 %b) {
+; CHECK-LABEL: whilele_b_xx:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    whilege p0.b, x1, x0
+; CHECK-NEXT:    rev p0.b, p0.b
+; CHECK-NEXT:    ret
+  %while = call <vscale x 16 x i1> @llvm.aarch64.sve.whilege.nxv16i1.i64(i64 %b, i64 %a)
+  %while.rev = call <vscale x 16 x i1> @llvm.vector.reverse.nxv16i1(<vscale x 16 x i1> %while)
+  ret <vscale x 16 x i1> %while.rev
+}
+
+; Cannot invert the while predicate when %a is MIN_INT.
+define <vscale x 8 x i1> @whilele_h_ww(i32 %a, i32 %b) {
+; CHECK-LABEL: whilele_h_ww:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    whilege p0.h, w1, w0
+; CHECK-NEXT:    rev p0.h, p0.h
+; CHECK-NEXT:    ret
+  %while = call <vscale x 8 x i1> @llvm.aarch64.sve.whilege.nxv8i1.i32(i32 %b, i32 %a)
+  %while.rev = call <vscale x 8 x i1> @llvm.vector.reverse.nxv8i1(<vscale x 8 x i1> %while)
+  ret <vscale x 8 x i1> %while.rev
+}
+
+; Cannot invert the while predicate when %a is MIN_INT.
+define <vscale x 8 x i1> @whilele_h_xx(i64 %a, i64 %b) {
+; CHECK-LABEL: whilele_h_xx:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    whilege p0.h, x1, x0
+; CHECK-NEXT:    rev p0.h, p0.h
+; CHECK-NEXT:    ret
+  %while = call <vscale x 8 x i1> @llvm.aarch64.sve.whilege.nxv8i1.i64(i64 %b, i64 %a)
+  %while.rev = call <vscale x 8 x i1> @llvm.vector.reverse.nxv8i1(<vscale x 8 x i1> %while)
+  ret <vscale x 8 x i1> %while.rev
+}
+
+; Cannot invert the while predicate when %a is MIN_INT.
+define <vscale x 4 x i1> @whilele_s_ww(i32 %a, i32 %b) {
+; CHECK-LABEL: whilele_s_ww:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    whilege p0.s, w1, w0
+; CHECK-NEXT:    rev p0.s, p0.s
+; CHECK-NEXT:    ret
+  %while = call <vscale x 4 x i1> @llvm.aarch64.sve.whilege.nxv4i1.i32(i32 %b, i32 %a)
+  %while.rev = call <vscale x 4 x i1> @llvm.vector.reverse.nxv4i1(<vscale x 4 x i1> %while)
+  ret <vscale x 4 x i1> %while.rev
+}
+
+; Cannot invert the while predicate when %a is MIN_INT.
+define <vscale x 4 x i1> @whilele_s_xx(i64 %a, i64 %b) {
+; CHECK-LABEL: whilele_s_xx:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    whilege p0.s, x1, x0
+; CHECK-NEXT:    rev p0.s, p0.s
+; CHECK-NEXT:    ret
+  %while = call <vscale x 4 x i1> @llvm.aarch64.sve.whilege.nxv4i1.i64(i64 %b, i64 %a)
+  %while.rev = call <vscale x 4 x i1> @llvm.vector.reverse.nxv4i1(<vscale x 4 x i1> %while)
+  ret <vscale x 4 x i1> %while.rev
+}
+
+; Cannot invert the while predicate when %a is MIN_INT.
+define <vscale x 2 x i1> @whilele_d_ww(i32 %a, i32 %b) {
+; CHECK-LABEL: whilele_d_ww:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    whilege p0.d, w1, w0
+; CHECK-NEXT:    rev p0.d, p0.d
+; CHECK-NEXT:    ret
+  %while = call <vscale x 2 x i1> @llvm.aarch64.sve.whilege.nxv2i1.i32(i32 %b, i32 %a)
+  %while.rev = call <vscale x 2 x i1> @llvm.vector.reverse.nxv2i1(<vscale x 2 x i1> %while)
+  ret <vscale x 2 x i1> %while.rev
+}
+
+; Cannot invert the while predicate when %a is MIN_INT.
+define <vscale x 2 x i1> @whilele_d_xx(i64 %a, i64 %b) {
+; CHECK-LABEL: whilele_d_xx:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    whilege p0.d, x1, x0
+; CHECK-NEXT:    rev p0.d, p0.d
+; CHECK-NEXT:    ret
+  %while = call <vscale x 2 x i1> @llvm.aarch64.sve.whilege.nxv2i1.i64(i64 %b, i64 %a)
+  %while.rev = call <vscale x 2 x i1> @llvm.vector.reverse.nxv2i1(<vscale x 2 x i1> %while)
+  ret <vscale x 2 x i1> %while.rev
+}
+
+;
+; WHILELO
+;
+
+define <vscale x 16 x i1> @whilelo_b_ww(i32 %a, i32 %b) {
+; CHECK-LABEL: whilelo_b_ww:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    whilelo p0.b, w0, w1
+; CHECK-NEXT:    ret
+  %while = call <vscale x 16 x i1> @llvm.aarch64.sve.whilehi.nxv16i1.i32(i32 %b, i32 %a)
+  %while.rev = call <vscale x 16 x i1> @llvm.vector.reverse.nxv16i1(<vscale x 16 x i1> %while)
+  ret <vscale x 16 x i1> %while.rev
+}
+
+define <vscale x 16 x i1> @whilelo_b_xx(i64 %a, i64 %b) {
+; CHECK-LABEL: whilelo_b_xx:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    whilelo p0.b, x0, x1
+; CHECK-NEXT:    ret
+  %while = call <vscale x 16 x i1> @llvm.aarch64.sve.whilehi.nxv16i1.i64(i64 %b, i64 %a)
+  %while.rev = call <vscale x 16 x i1> @llvm.vector.reverse.nxv16i1(<vscale x 16 x i1> %while)
+  ret <vscale x 16 x i1> %while.rev
+}
+
+define <vscale x 8 x i1> @whilelo_h_ww(i32 %a, i32 %b) {
+; CHECK-LABEL: whilelo_h_ww:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    whilelo p0.h, w0, w1
+; CHECK-NEXT:    ret
+  %while = call <vscale x 8 x i1> @llvm.aarch64.sve.whilehi.nxv8i1.i32(i32 %b, i32 %a)
+  %while.rev = call <vscale x 8 x i1> @llvm.vector.reverse.nxv8i1(<vscale x 8 x i1> %while)
+  ret <vscale x 8 x i1> %while.rev
+}
+
+define <vscale x 8 x i1> @whilelo_h_xx(i64 %a, i64 %b) {
+; CHECK-LABEL: whilelo_h_xx:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    whilelo p0.h, x0, x1
+; CHECK-NEXT:    ret
+  %while = call <vscale x 8 x i1> @llvm.aarch64.sve.whilehi.nxv8i1.i64(i64 %b, i64 %a)
+  %while.rev = call <vscale x 8 x i1> @llvm.vector.reverse.nxv8i1(<vscale x 8 x i1> %while)
+  ret <vscale x 8 x i1> %while.rev
+}
+
+define <vscale x 4 x i1> @whilelo_s_ww(i32 %a, i32 %b) {
+; CHECK-LABEL: whilelo_s_ww:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    whilelo p0.s, w0, w1
+; CHECK-NEXT:    ret
+  %while = call <vscale x 4 x i1> @llvm.aarch64.sve.whilehi.nxv4i1.i32(i32 %b, i32 %a)
+  %while.rev = call <vscale x 4 x i1> @llvm.vector.reverse.nxv4i1(<vscale x 4 x i1> %while)
+  ret <vscale x 4 x i1> %while.rev
+}
+
+define <vscale x 4 x i1> @whilelo_s_xx(i64 %a, i64 %b) {
+; CHECK-LABEL: whilelo_s_xx:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    whilelo p0.s, x0, x1
+; CHECK-NEXT:    ret
+  %while = call <vscale x 4 x i1> @llvm.aarch64.sve.whilehi.nxv4i1.i64(i64 %b, i64 %a)
+  %while.rev = call <vscale x 4 x i1> @llvm.vector.reverse.nxv4i1(<vscale x 4 x i1> %while)
+  ret <vscale x 4 x i1> %while.rev
+}
+
+define <vscale x 2 x i1> @whilelo_d_ww(i32 %a, i32 %b) {
+; CHECK-LABEL: whilelo_d_ww:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    whilelo p0.d, w0, w1
+; CHECK-NEXT:    ret
+  %while = call <vscale x 2 x i1> @llvm.aarch64.sve.whilehi.nxv2i1.i32(i32 %b, i32 %a)
+  %while.rev = call <vscale x 2 x i1> @llvm.vector.reverse.nxv2i1(<vscale x 2 x i1> %while)
+  ret <vscale x 2 x i1> %while.rev
+}
+
+define <vscale x 2 x i1> @whilelo_d_xx(i64 %a, i64 %b) {
+; CHECK-LABEL: whilelo_d_xx:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    whilelo p0.d, x0, x1
+; CHECK-NEXT:    ret
+  %while = call <vscale x 2 x i1> @llvm.aarch64.sve.whilehi.nxv2i1.i64(i64 %b, i64 %a)
+  %while.rev = call <vscale x 2 x i1> @llvm.vector.reverse.nxv2i1(<vscale x 2 x i1> %while)
+  ret <vscale x 2 x i1> %while.rev
+}
+
+;
+; WHILELS
+;
+
+; Cannot invert the while predicate when %a is MIN_UINT.
+define <vscale x 16 x i1> @whilels_b_ww(i32 %a, i32 %b) {
+; CHECK-LABEL: whilels_b_ww:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    whilehs p0.b, w1, w0
+; CHECK-NEXT:    rev p0.b, p0.b
+; CHECK-NEXT:    ret
+  %while = call <vscale x 16 x i1> @llvm.aarch64.sve.whilehs.nxv16i1.i32(i32 %b, i32 %a)
+  %while.rev = call <vscale x 16 x i1> @llvm.vector.reverse.nxv16i1(<vscale x 16 x i1> %while)
+  ret <vscale x 16 x i1> %while.rev
+}
+
+; Cannot invert the while predicate when %a is MIN_UINT.
+define <vscale x 16 x i1> @whilels_b_xx(i64 %a, i64 %b) {
+; CHECK-LABEL: whilels_b_xx:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    whilehs p0.b, x1, x0
+; CHECK-NEXT:    rev p0.b, p0.b
+; CHECK-NEXT:    ret
+  %while = call <vscale x 16 x i1> @llvm.aarch64.sve.whilehs.nxv16i1.i64(i64 %b, i64 %a)
+  %while.rev = call <vscale x 16 x i1> @llvm.vector.reverse.nxv16i1(<vscale x 16 x i1> %while)
+  ret <vscale x 16 x i1> %while.rev
+}
+
+; Cannot invert the while predicate when %a is MIN_UINT.
+define <vscale x 8 x i1> @whilels_h_ww(i32 %a, i32 %b) {
+; CHECK-LABEL: whilels_h_ww:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    whilehs p0.h, w1, w0
+; CHECK-NEXT:    rev p0.h, p0.h
+; CHECK-NEXT:    ret
+  %while = call <vscale x 8 x i1> @llvm.aarch64.sve.whilehs.nxv8i1.i32(i32 %b, i32 %a)
+  %while.rev = call <vscale x 8 x i1> @llvm.vector.reverse.nxv8i1(<vscale x 8 x i1> %while)
+  ret <vscale x 8 x i1> %while.rev
+}
+
+; Cannot invert the while predicate when %a is MIN_UINT.
+define <vscale x 8 x i1> @whilels_h_xx(i64 %a, i64 %b) {
+; CHECK-LABEL: whilels_h_xx:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    whilehs p0.h, x1, x0
+; CHECK-NEXT:    rev p0.h, p0.h
+; CHECK-NEXT:    ret
+  %while = call <vscale x 8 x i1> @llvm.aarch64.sve.whilehs.nxv8i1.i64(i64 %b, i64 %a)
+  %while.rev = call <vscale x 8 x i1> @llvm.vector.reverse.nxv8i1(<vscale x 8 x i1> %while)
+  ret <vscale x 8 x i1> %while.rev
+}
+
+; Cannot invert the while predicate when %a is MIN_UINT.
+define <vscale x 4 x i1> @whilels_s_ww(i32 %a, i32 %b) {
+; CHECK-LABEL: whilels_s_ww:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    whilehs p0.s, w1, w0
+; CHECK-NEXT:    rev p0.s, p0.s
+; CHECK-NEXT:    ret
+  %while = call <vscale x 4 x i1> @llvm.aarch64.sve.whilehs.nxv4i1.i32(i32 %b, i32 %a)
+  %while.rev = call <vscale x 4 x i1> @llvm.vector.reverse.nxv4i1(<vscale x 4 x i1> %while)
+  ret <vscale x 4 x i1> %while.rev
+}
+
+; Cannot invert the while predicate when %a is MIN_UINT.
+define <vscale x 4 x i1> @whilels_s_xx(i64 %a, i64 %b) {
+; CHECK-LABEL: whilels_s_xx:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    whilehs p0.s, x1, x0
+; CHECK-NEXT:    rev p0.s, p0.s
+; CHECK-NEXT:    ret
+  %while = call <vscale x 4 x i1> @llvm.aarch64.sve.whilehs.nxv4i1.i64(i64 %b, i64 %a)
+  %while.rev = call <vscale x 4 x i1> @llvm.vector.reverse.nxv4i1(<vscale x 4 x i1> %while)
+  ret <vscale x 4 x i1> %while.rev
+}
+
+; Cannot invert the while predicate when %a is MIN_UINT.
+define <vscale x 2 x i1> @whilels_d_ww(i32 %a, i32 %b) {
+; CHECK-LABEL: whilels_d_ww:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    whilehs p0.d, w1, w0
+; CHECK-NEXT:    rev p0.d, p0.d
+; CHECK-NEXT:    ret
+  %while = call <vscale x 2 x i1> @llvm.aarch64.sve.whilehs.nxv2i1.i32(i32 %b, i32 %a)
+  %while.rev = call <vscale x 2 x i1> @llvm.vector.reverse.nxv2i1(<vscale x 2 x i1> %while)
+  ret <vscale x 2 x i1> %while.rev
+}
+
+; Cannot invert the while predicate when %a is MIN_UINT.
+define <vscale x 2 x i1> @whilels_d_xx(i64 %a, i64 %b) {
+; CHECK-LABEL: whilels_d_xx:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    whilehs p0.d, x1, x0
+; CHECK-NEXT:    rev p0.d, p0.d
+; CHECK-NEXT:    ret
+  %while = call <vscale x 2 x i1> @llvm.aarch64.sve.whilehs.nxv2i1.i64(i64 %b, i64 %a)
+  %while.rev = call <vscale x 2 x i1> @llvm.vector.reverse.nxv2i1(<vscale x 2 x i1> %while)
+  ret <vscale x 2 x i1> %while.rev
+}
+
+;
+; WHILELT
+;
+
+define <vscale x 16 x i1> @whilelt_b_ww(i32 %a, i32 %b) {
+; CHECK-LABEL: whilelt_b_ww:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    whilelt p0.b, w0, w1
+; CHECK-NEXT:    ret
+  %while = call <vscale x 16 x i1> @llvm.aarch64.sve.whilegt.nxv16i1.i32(i32 %b, i32 %a)
+  %while.rev = call <vscale x 16 x i1> @llvm.vector.reverse.nxv16i1(<vscale x 16 x i1> %while)
+  ret <vscale x 16 x i1> %while.rev
+}
+
+define <vscale x 16 x i1> @whilelt_b_xx(i64 %a, i64 %b) {
+; CHECK-LABEL: whilelt_b_xx:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    whilelt p0.b, x0, x1
+; CHECK-NEXT:    ret
+  %while = call <vscale x 16 x i1> @llvm.aarch64.sve.whilegt.nxv16i1.i64(i64 %b, i64 %a)
+  %while.rev = call <vscale x 16 x i1> @llvm.vector.reverse.nxv16i1(<vscale x 16 x i1> %while)
+  ret <vscale x 16 x i1> %while.rev
+}
+
+define <vscale x 8 x i1> @whilelt_h_ww(i32 %a, i32 %b) {
+; CHECK-LABEL: whilelt_h_ww:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    whilelt p0.h, w0, w1
+; CHECK-NEXT:    ret
+  %while = call <vscale x 8 x i1> @llvm.aarch64.sve.whilegt.nxv8i1.i32(i32 %b, i32 %a)
+  %while.rev = call <vscale x 8 x i1> @llvm.vector.reverse.nxv8i1(<vscale x 8 x i1> %while)
+  ret <vscale x 8 x i1> %while.rev
+}
+
+define <vscale x 8 x i1> @whilelt_h_xx(i64 %a, i64 %b) {
+; CHECK-LABEL: whilelt_h_xx:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    whilelt p0.h, x0, x1
+; CHECK-NEXT:    ret
+  %while = call <vscale x 8 x i1> @llvm.aarch64.sve.whilegt.nxv8i1.i64(i64 %b, i64 %a)
+  %while.rev = call <vscale x 8 x i1> @llvm.vector.reverse.nxv8i1(<vscale x 8 x i1> %while)
+  ret <vscale x 8 x i1> %while.rev
+}
+
+define <vscale x 4 x i1> @whilelt_s_ww(i32 %a, i32 %b) {
+; CHECK-LABEL: whilelt_s_ww:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    whilelt p0.s, w0, w1
+; CHECK-NEXT:    ret
+  %while = call <vscale x 4 x i1> @llvm.aarch64.sve.whilegt.nxv4i1.i32(i32 %b, i32 %a)
+  %while.rev = call <vscale x 4 x i1> @llvm.vector.reverse.nxv4i1(<vscale x 4 x i1> %while)
+  ret <vscale x 4 x i1> %while.rev
+}
+
+define <vscale x 4 x i1> @whilelt_s_xx(i64 %a, i64 %b) {
+; CHECK-LABEL: whilelt_s_xx:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    whilelt p0.s, x0, x1
+; CHECK-NEXT:    ret
+  %while = call <vscale x 4 x i1> @llvm.aarch64.sve.whilegt.nxv4i1.i64(i64 %b, i64 %a)
+  %while.rev = call <vscale x 4 x i1> @llvm.vector.reverse.nxv4i1(<vscale x 4 x i1> %while)
+  ret <vscale x 4 x i1> %while.rev
+}
+
+define <vscale x 2 x i1> @whilelt_d_ww(i32 %a, i32 %b) {
+; CHECK-LABEL: whilelt_d_ww:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    whilelt p0.d, w0, w1
+; CHECK-NEXT:    ret
+  %while = call <vscale x 2 x i1> @llvm.aarch64.sve.whilegt.nxv2i1.i32(i32 %b, i32 %a)
+  %while.rev = call <vscale x 2 x i1> @llvm.vector.reverse.nxv2i1(<vscale x 2 x i1> %while)
+  ret <vscale x 2 x i1> %while.rev
+}
+
+define <vscale x 2 x i1> @whilelt_d_xx(i64 %a, i64 %b) {
+; CHECK-LABEL: whilelt_d_xx:
+; CHECK:       // %bb.0:
+; CHECK-NEXT:    whilelt p0.d, x0, x1
+; CHECK-NEXT:    ret
+  %while = call <vscale x 2 x i1> @llvm.aarch64.sve.whilegt.nxv2i1.i64(i64 %b, i64 %a)
+  %while.rev = call <vscale x 2 x i1> @llvm.vector.reverse.nxv2i1(<vscale x 2 x i1> %while)
+  ret <vscale x 2 x i1> %while.rev
+}
diff --git a/llvm/test/CodeGen/AArch64/vecreduce-add.ll b/llvm/test/CodeGen/AArch64/vecreduce-add.ll
index 3254c5ebe9c6..ab7cea8dfb77 100644
--- a/llvm/test/CodeGen/AArch64/vecreduce-add.ll
+++ b/llvm/test/CodeGen/AArch64/vecreduce-add.ll
@@ -2825,10 +2825,11 @@ entry:
 define i64 @add_pair_v2i16_v2i64_zext(<2 x i16> %x, <2 x i16> %y) {
 ; CHECK-SD-LABEL: add_pair_v2i16_v2i64_zext:
 ; CHECK-SD:       // %bb.0: // %entry
-; CHECK-SD-NEXT:    movi d2, #0x00ffff0000ffff
-; CHECK-SD-NEXT:    and v0.8b, v0.8b, v2.8b
-; CHECK-SD-NEXT:    and v1.8b, v1.8b, v2.8b
+; CHECK-SD-NEXT:    // kill: def $d0 killed $d0 def $q0
+; CHECK-SD-NEXT:    // kill: def $d1 killed $d1 def $q1
+; CHECK-SD-NEXT:    movi v2.2d, #0x00ffff0000ffff
 ; CHECK-SD-NEXT:    mov v0.d[1], v1.d[0]
+; CHECK-SD-NEXT:    and v0.16b, v0.16b, v2.16b
 ; CHECK-SD-NEXT:    uaddlv d0, v0.4s
 ; CHECK-SD-NEXT:    fmov x0, d0
 ; CHECK-SD-NEXT:    ret
@@ -3578,10 +3579,11 @@ entry:
 define i64 @add_pair_v2i8_v2i64_zext(<2 x i8> %x, <2 x i8> %y) {
 ; CHECK-SD-LABEL: add_pair_v2i8_v2i64_zext:
 ; CHECK-SD:       // %bb.0: // %entry
-; CHECK-SD-NEXT:    movi d2, #0x0000ff000000ff
-; CHECK-SD-NEXT:    and v0.8b, v0.8b, v2.8b
-; CHECK-SD-NEXT:    and v1.8b, v1.8b, v2.8b
+; CHECK-SD-NEXT:    // kill: def $d0 killed $d0 def $q0
+; CHECK-SD-NEXT:    // kill: def $d1 killed $d1 def $q1
+; CHECK-SD-NEXT:    movi v2.2d, #0x0000ff000000ff
 ; CHECK-SD-NEXT:    mov v0.d[1], v1.d[0]
+; CHECK-SD-NEXT:    and v0.16b, v0.16b, v2.16b
 ; CHECK-SD-NEXT:    uaddlv d0, v0.4s
 ; CHECK-SD-NEXT:    fmov x0, d0
 ; CHECK-SD-NEXT:    ret
diff --git a/llvm/test/CodeGen/AArch64/vscale-and-sve-cnt-demandedbits.ll b/llvm/test/CodeGen/AArch64/vscale-and-sve-cnt-demandedbits.ll
index dbdab799c835..9572778484f8 100644
--- a/llvm/test/CodeGen/AArch64/vscale-and-sve-cnt-demandedbits.ll
+++ b/llvm/test/CodeGen/AArch64/vscale-and-sve-cnt-demandedbits.ll
@@ -194,7 +194,7 @@ define i32 @vscale_with_multiplier() vscale_range(1,16) {
 ; CHECK-LABEL: vscale_with_multiplier:
 ; CHECK:       // %bb.0:
 ; CHECK-NEXT:    rdvl x8, #1
-; CHECK-NEXT:    mov w9, #5
+; CHECK-NEXT:    mov w9, #5 // =0x5
 ; CHECK-NEXT:    lsr x8, x8, #4
 ; CHECK-NEXT:    mul x8, x8, x9
 ; CHECK-NEXT:    and w9, w8, #0x3f
@@ -212,7 +212,7 @@ define i32 @vscale_with_negative_multiplier() vscale_range(1,16) {
 ; CHECK-LABEL: vscale_with_negative_multiplier:
 ; CHECK:       // %bb.0:
 ; CHECK-NEXT:    rdvl x8, #1
-; CHECK-NEXT:    mov x9, #-5
+; CHECK-NEXT:    mov x9, #-5 // =0xfffffffffffffffb
 ; CHECK-NEXT:    lsr x8, x8, #4
 ; CHECK-NEXT:    mul x8, x8, x9
 ; CHECK-NEXT:    and w9, w8, #0xffffffc0
@@ -230,9 +230,9 @@ define i32 @pow2_vscale_with_negative_multiplier() vscale_range(1,16) {
 ; CHECK-LABEL: pow2_vscale_with_negative_multiplier:
 ; CHECK:       // %bb.0:
 ; CHECK-NEXT:    cntd x8
-; CHECK-NEXT:    neg x8, x8
-; CHECK-NEXT:    orr w9, w8, #0xfffffff0
-; CHECK-NEXT:    add w0, w8, w9
+; CHECK-NEXT:    neg x9, x8
+; CHECK-NEXT:    orr w9, w9, #0xfffffff0
+; CHECK-NEXT:    sub w0, w9, w8
 ; CHECK-NEXT:    ret
   %vscale = call i32 @llvm.vscale.i32()
   %mul = mul i32 %vscale, -2
diff --git a/llvm/test/CodeGen/AMDGPU/GlobalISel/legalize-trap-gfx11.mir b/llvm/test/CodeGen/AMDGPU/GlobalISel/legalize-trap-gfx11.mir
new file mode 100644
index 000000000000..ac98dca00be3
--- /dev/null
+++ b/llvm/test/CodeGen/AMDGPU/GlobalISel/legalize-trap-gfx11.mir
@@ -0,0 +1,49 @@
+# NOTE: Assertions have been autogenerated by utils/update_mir_test_checks.py UTC_ARGS: --version 2
+# RUN: llc -mtriple=amdgcn--amdhsa -mcpu=gfx1100 -o - -run-pass=legalizer %s | FileCheck -check-prefix=GFX1100 %s
+# RUN: llc -mtriple=amdgcn--amdhsa -mcpu=gfx11-generic --amdhsa-code-object-version=6 -o - -run-pass=legalizer %s | FileCheck -check-prefix=GFX1100 %s
+# RUN: llc -mtriple=amdgcn--amdhsa -mcpu=gfx1150 -o - -run-pass=legalizer %s | FileCheck -check-prefix=GFX1150 %s
+
+---
+name: test_trap
+body: |
+  bb.0:
+    ; GFX1100-LABEL: name: test_trap
+    ; GFX1100: successors: %bb.2(0x80000000)
+    ; GFX1100-NEXT: {{  $}}
+    ; GFX1100-NEXT: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 0
+    ; GFX1100-NEXT: [[C1:%[0-9]+]]:_(p1) = G_CONSTANT i64 0
+    ; GFX1100-NEXT: G_STORE [[C]](s32), [[C1]](p1) :: (store (s8), addrspace 1)
+    ; GFX1100-NEXT: S_TRAP 2
+    ; GFX1100-NEXT: [[S_SENDMSG_RTN_B32_:%[0-9]+]]:sreg_32 = S_SENDMSG_RTN_B32 128
+    ; GFX1100-NEXT: $ttmp2 = S_MOV_B32 $m0
+    ; GFX1100-NEXT: [[S_AND_B32_:%[0-9]+]]:sreg_32 = S_AND_B32 [[S_SENDMSG_RTN_B32_]], 1023, implicit-def $scc
+    ; GFX1100-NEXT: [[S_OR_B32_:%[0-9]+]]:sreg_32 = S_OR_B32 [[S_AND_B32_]], 1024, implicit-def $scc
+    ; GFX1100-NEXT: $m0 = S_MOV_B32 [[S_OR_B32_]]
+    ; GFX1100-NEXT: S_SENDMSG 1, implicit $exec, implicit $m0
+    ; GFX1100-NEXT: $m0 = S_MOV_B32 $ttmp2
+    ; GFX1100-NEXT: S_BRANCH %bb.2
+    ; GFX1100-NEXT: {{  $}}
+    ; GFX1100-NEXT: .1:
+    ; GFX1100-NEXT: successors:
+    ; GFX1100-NEXT: {{  $}}
+    ; GFX1100-NEXT: G_STORE [[C]](s32), [[C1]](p1) :: (store (s8), addrspace 1)
+    ; GFX1100-NEXT: {{  $}}
+    ; GFX1100-NEXT: .2:
+    ; GFX1100-NEXT: successors: %bb.2(0x80000000)
+    ; GFX1100-NEXT: {{  $}}
+    ; GFX1100-NEXT: S_SETHALT 5
+    ; GFX1100-NEXT: S_BRANCH %bb.2
+    ;
+    ; GFX1150-LABEL: name: test_trap
+    ; GFX1150: [[C:%[0-9]+]]:_(s32) = G_CONSTANT i32 0
+    ; GFX1150-NEXT: [[C1:%[0-9]+]]:_(p1) = G_CONSTANT i64 0
+    ; GFX1150-NEXT: G_STORE [[C]](s32), [[C1]](p1) :: (store (s8), addrspace 1)
+    ; GFX1150-NEXT: S_TRAP 2
+    ; GFX1150-NEXT: G_STORE [[C]](s32), [[C1]](p1) :: (store (s8), addrspace 1)
+    %0:_(s8) = G_CONSTANT i8 0
+    %1:_(p1) = G_CONSTANT i64 0
+    G_STORE %0, %1 :: (store 1, addrspace 1)
+    G_TRAP
+    G_STORE %0, %1 :: (store 1, addrspace 1)
+
+...
diff --git a/llvm/test/CodeGen/AMDGPU/GlobalISel/llvm.amdgcn.intersect_ray.ll b/llvm/test/CodeGen/AMDGPU/GlobalISel/llvm.amdgcn.intersect_ray.ll
index 47e476de74cf..6e96a4ddbc0b 100644
--- a/llvm/test/CodeGen/AMDGPU/GlobalISel/llvm.amdgcn.intersect_ray.ll
+++ b/llvm/test/CodeGen/AMDGPU/GlobalISel/llvm.amdgcn.intersect_ray.ll
@@ -144,40 +144,42 @@ define amdgpu_ps <4 x float> @image_bvh64_intersect_ray_a16(i64 %node_ptr, float
 define amdgpu_ps <4 x float> @image_bvh_intersect_ray_vgpr_descr(i32 %node_ptr, float %ray_extent, <3 x float> %ray_origin, <3 x float> %ray_dir, <3 x float> %ray_inv_dir, <4 x i32> %tdescr) {
 ; GFX1030-LABEL: image_bvh_intersect_ray_vgpr_descr:
 ; GFX1030:       ; %bb.0:
-; GFX1030-NEXT:    v_mov_b32_e32 v15, v0
-; GFX1030-NEXT:    v_mov_b32_e32 v16, v1
-; GFX1030-NEXT:    v_mov_b32_e32 v17, v2
-; GFX1030-NEXT:    v_mov_b32_e32 v18, v3
-; GFX1030-NEXT:    v_mov_b32_e32 v19, v4
-; GFX1030-NEXT:    v_mov_b32_e32 v20, v5
-; GFX1030-NEXT:    v_mov_b32_e32 v21, v6
-; GFX1030-NEXT:    v_mov_b32_e32 v22, v7
-; GFX1030-NEXT:    v_mov_b32_e32 v23, v8
-; GFX1030-NEXT:    v_mov_b32_e32 v24, v9
-; GFX1030-NEXT:    v_mov_b32_e32 v25, v10
+; GFX1030-NEXT:    v_mov_b32_e32 v21, v0
+; GFX1030-NEXT:    v_mov_b32_e32 v22, v1
+; GFX1030-NEXT:    v_mov_b32_e32 v23, v2
+; GFX1030-NEXT:    v_mov_b32_e32 v24, v3
+; GFX1030-NEXT:    v_mov_b32_e32 v25, v4
+; GFX1030-NEXT:    v_mov_b32_e32 v26, v5
+; GFX1030-NEXT:    v_mov_b32_e32 v27, v6
+; GFX1030-NEXT:    v_mov_b32_e32 v28, v7
+; GFX1030-NEXT:    v_mov_b32_e32 v29, v8
+; GFX1030-NEXT:    v_mov_b32_e32 v30, v9
+; GFX1030-NEXT:    v_mov_b32_e32 v31, v10
+; GFX1030-NEXT:    v_mov_b32_e32 v19, v11
+; GFX1030-NEXT:    v_mov_b32_e32 v20, v12
 ; GFX1030-NEXT:    s_mov_b32 s1, exec_lo
 ; GFX1030-NEXT:  .LBB6_1: ; =>This Inner Loop Header: Depth=1
-; GFX1030-NEXT:    v_readfirstlane_b32 s4, v11
-; GFX1030-NEXT:    v_readfirstlane_b32 s5, v12
+; GFX1030-NEXT:    v_readfirstlane_b32 s4, v19
+; GFX1030-NEXT:    v_readfirstlane_b32 s5, v20
 ; GFX1030-NEXT:    v_readfirstlane_b32 s6, v13
 ; GFX1030-NEXT:    v_readfirstlane_b32 s7, v14
-; GFX1030-NEXT:    v_cmp_eq_u64_e32 vcc_lo, s[4:5], v[11:12]
+; GFX1030-NEXT:    v_cmp_eq_u64_e32 vcc_lo, s[4:5], v[19:20]
 ; GFX1030-NEXT:    v_cmp_eq_u64_e64 s0, s[6:7], v[13:14]
 ; GFX1030-NEXT:    s_and_b32 s0, vcc_lo, s0
 ; GFX1030-NEXT:    s_and_saveexec_b32 s0, s0
-; GFX1030-NEXT:    image_bvh_intersect_ray v[0:3], v[15:25], s[4:7]
-; GFX1030-NEXT:    ; implicit-def: $vgpr11
-; GFX1030-NEXT:    ; implicit-def: $vgpr15
-; GFX1030-NEXT:    ; implicit-def: $vgpr16
-; GFX1030-NEXT:    ; implicit-def: $vgpr17
-; GFX1030-NEXT:    ; implicit-def: $vgpr18
+; GFX1030-NEXT:    image_bvh_intersect_ray v[0:3], v[21:31], s[4:7]
 ; GFX1030-NEXT:    ; implicit-def: $vgpr19
-; GFX1030-NEXT:    ; implicit-def: $vgpr20
 ; GFX1030-NEXT:    ; implicit-def: $vgpr21
 ; GFX1030-NEXT:    ; implicit-def: $vgpr22
 ; GFX1030-NEXT:    ; implicit-def: $vgpr23
 ; GFX1030-NEXT:    ; implicit-def: $vgpr24
 ; GFX1030-NEXT:    ; implicit-def: $vgpr25
+; GFX1030-NEXT:    ; implicit-def: $vgpr26
+; GFX1030-NEXT:    ; implicit-def: $vgpr27
+; GFX1030-NEXT:    ; implicit-def: $vgpr28
+; GFX1030-NEXT:    ; implicit-def: $vgpr29
+; GFX1030-NEXT:    ; implicit-def: $vgpr30
+; GFX1030-NEXT:    ; implicit-def: $vgpr31
 ; GFX1030-NEXT:    ; implicit-def: $vgpr11_vgpr12_vgpr13_vgpr14
 ; GFX1030-NEXT:    s_xor_b32 exec_lo, exec_lo, s0
 ; GFX1030-NEXT:    s_cbranch_execnz .LBB6_1
@@ -188,18 +190,20 @@ define amdgpu_ps <4 x float> @image_bvh_intersect_ray_vgpr_descr(i32 %node_ptr,
 ;
 ; GFX1013-LABEL: image_bvh_intersect_ray_vgpr_descr:
 ; GFX1013:       ; %bb.0:
+; GFX1013-NEXT:    v_mov_b32_e32 v19, v11
+; GFX1013-NEXT:    v_mov_b32_e32 v20, v12
 ; GFX1013-NEXT:    s_mov_b32 s1, exec_lo
 ; GFX1013-NEXT:  .LBB6_1: ; =>This Inner Loop Header: Depth=1
-; GFX1013-NEXT:    v_readfirstlane_b32 s4, v11
-; GFX1013-NEXT:    v_readfirstlane_b32 s5, v12
+; GFX1013-NEXT:    v_readfirstlane_b32 s4, v19
+; GFX1013-NEXT:    v_readfirstlane_b32 s5, v20
 ; GFX1013-NEXT:    v_readfirstlane_b32 s6, v13
 ; GFX1013-NEXT:    v_readfirstlane_b32 s7, v14
-; GFX1013-NEXT:    v_cmp_eq_u64_e32 vcc_lo, s[4:5], v[11:12]
+; GFX1013-NEXT:    v_cmp_eq_u64_e32 vcc_lo, s[4:5], v[19:20]
 ; GFX1013-NEXT:    v_cmp_eq_u64_e64 s0, s[6:7], v[13:14]
 ; GFX1013-NEXT:    s_and_b32 s0, vcc_lo, s0
 ; GFX1013-NEXT:    s_and_saveexec_b32 s0, s0
 ; GFX1013-NEXT:    image_bvh_intersect_ray v[15:18], v[0:10], s[4:7]
-; GFX1013-NEXT:    ; implicit-def: $vgpr11
+; GFX1013-NEXT:    ; implicit-def: $vgpr19
 ; GFX1013-NEXT:    ; implicit-def: $vgpr0_vgpr1_vgpr2_vgpr3_vgpr4_vgpr5_vgpr6_vgpr7_vgpr8_vgpr9_vgpr10
 ; GFX1013-NEXT:    ; implicit-def: $vgpr11_vgpr12_vgpr13_vgpr14
 ; GFX1013-NEXT:    s_waitcnt_depctr 0xffe3
@@ -216,25 +220,27 @@ define amdgpu_ps <4 x float> @image_bvh_intersect_ray_vgpr_descr(i32 %node_ptr,
 ;
 ; GFX11-LABEL: image_bvh_intersect_ray_vgpr_descr:
 ; GFX11:       ; %bb.0:
-; GFX11-NEXT:    v_dual_mov_b32 v18, v0 :: v_dual_mov_b32 v19, v1
+; GFX11-NEXT:    v_dual_mov_b32 v20, v0 :: v_dual_mov_b32 v21, v1
 ; GFX11-NEXT:    v_dual_mov_b32 v15, v2 :: v_dual_mov_b32 v16, v3
-; GFX11-NEXT:    v_mov_b32_e32 v17, v4
+; GFX11-NEXT:    v_dual_mov_b32 v17, v4 :: v_dual_mov_b32 v18, v11
+; GFX11-NEXT:    v_mov_b32_e32 v19, v12
 ; GFX11-NEXT:    s_mov_b32 s1, exec_lo
 ; GFX11-NEXT:  .LBB6_1: ; =>This Inner Loop Header: Depth=1
-; GFX11-NEXT:    v_readfirstlane_b32 s4, v11
-; GFX11-NEXT:    v_readfirstlane_b32 s5, v12
+; GFX11-NEXT:    s_delay_alu instid0(VALU_DEP_2) | instskip(NEXT) | instid1(VALU_DEP_2)
+; GFX11-NEXT:    v_readfirstlane_b32 s4, v18
+; GFX11-NEXT:    v_readfirstlane_b32 s5, v19
 ; GFX11-NEXT:    v_readfirstlane_b32 s6, v13
 ; GFX11-NEXT:    v_readfirstlane_b32 s7, v14
 ; GFX11-NEXT:    s_delay_alu instid0(VALU_DEP_3) | instskip(NEXT) | instid1(VALU_DEP_2)
-; GFX11-NEXT:    v_cmp_eq_u64_e32 vcc_lo, s[4:5], v[11:12]
+; GFX11-NEXT:    v_cmp_eq_u64_e32 vcc_lo, s[4:5], v[18:19]
 ; GFX11-NEXT:    v_cmp_eq_u64_e64 s0, s[6:7], v[13:14]
 ; GFX11-NEXT:    s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(SALU_CYCLE_1)
 ; GFX11-NEXT:    s_and_b32 s0, vcc_lo, s0
 ; GFX11-NEXT:    s_and_saveexec_b32 s0, s0
-; GFX11-NEXT:    image_bvh_intersect_ray v[0:3], [v18, v19, v[15:17], v[5:7], v[8:10]], s[4:7]
-; GFX11-NEXT:    ; implicit-def: $vgpr11
+; GFX11-NEXT:    image_bvh_intersect_ray v[0:3], [v20, v21, v[15:17], v[5:7], v[8:10]], s[4:7]
 ; GFX11-NEXT:    ; implicit-def: $vgpr18
-; GFX11-NEXT:    ; implicit-def: $vgpr19
+; GFX11-NEXT:    ; implicit-def: $vgpr20
+; GFX11-NEXT:    ; implicit-def: $vgpr21
 ; GFX11-NEXT:    ; implicit-def: $vgpr15_vgpr16_vgpr17
 ; GFX11-NEXT:    ; implicit-def: $vgpr5_vgpr6_vgpr7
 ; GFX11-NEXT:    ; implicit-def: $vgpr8_vgpr9_vgpr10
@@ -253,39 +259,41 @@ define amdgpu_ps <4 x float> @image_bvh_intersect_ray_vgpr_descr(i32 %node_ptr,
 define amdgpu_ps <4 x float> @image_bvh_intersect_ray_a16_vgpr_descr(i32 %node_ptr, float %ray_extent, <3 x float> %ray_origin, <3 x half> %ray_dir, <3 x half> %ray_inv_dir, <4 x i32> %tdescr) {
 ; GFX1030-LABEL: image_bvh_intersect_ray_a16_vgpr_descr:
 ; GFX1030:       ; %bb.0:
-; GFX1030-NEXT:    v_mov_b32_e32 v13, v0
-; GFX1030-NEXT:    v_mov_b32_e32 v14, v1
+; GFX1030-NEXT:    v_mov_b32_e32 v18, v0
+; GFX1030-NEXT:    v_mov_b32_e32 v19, v1
 ; GFX1030-NEXT:    v_lshrrev_b32_e32 v0, 16, v5
 ; GFX1030-NEXT:    v_and_b32_e32 v1, 0xffff, v7
-; GFX1030-NEXT:    v_mov_b32_e32 v15, v2
+; GFX1030-NEXT:    v_mov_b32_e32 v20, v2
 ; GFX1030-NEXT:    v_and_b32_e32 v2, 0xffff, v8
-; GFX1030-NEXT:    v_mov_b32_e32 v16, v3
+; GFX1030-NEXT:    v_mov_b32_e32 v21, v3
 ; GFX1030-NEXT:    v_lshlrev_b32_e32 v0, 16, v0
 ; GFX1030-NEXT:    v_lshlrev_b32_e32 v1, 16, v1
-; GFX1030-NEXT:    v_mov_b32_e32 v17, v4
-; GFX1030-NEXT:    v_alignbit_b32 v20, v2, v7, 16
+; GFX1030-NEXT:    v_mov_b32_e32 v22, v4
+; GFX1030-NEXT:    v_mov_b32_e32 v16, v9
+; GFX1030-NEXT:    v_mov_b32_e32 v17, v10
+; GFX1030-NEXT:    v_and_or_b32 v23, 0xffff, v5, v0
+; GFX1030-NEXT:    v_and_or_b32 v24, 0xffff, v6, v1
+; GFX1030-NEXT:    v_alignbit_b32 v25, v2, v7, 16
 ; GFX1030-NEXT:    s_mov_b32 s1, exec_lo
-; GFX1030-NEXT:    v_and_or_b32 v18, 0xffff, v5, v0
-; GFX1030-NEXT:    v_and_or_b32 v19, 0xffff, v6, v1
 ; GFX1030-NEXT:  .LBB7_1: ; =>This Inner Loop Header: Depth=1
-; GFX1030-NEXT:    v_readfirstlane_b32 s4, v9
-; GFX1030-NEXT:    v_readfirstlane_b32 s5, v10
+; GFX1030-NEXT:    v_readfirstlane_b32 s4, v16
+; GFX1030-NEXT:    v_readfirstlane_b32 s5, v17
 ; GFX1030-NEXT:    v_readfirstlane_b32 s6, v11
 ; GFX1030-NEXT:    v_readfirstlane_b32 s7, v12
-; GFX1030-NEXT:    v_cmp_eq_u64_e32 vcc_lo, s[4:5], v[9:10]
+; GFX1030-NEXT:    v_cmp_eq_u64_e32 vcc_lo, s[4:5], v[16:17]
 ; GFX1030-NEXT:    v_cmp_eq_u64_e64 s0, s[6:7], v[11:12]
 ; GFX1030-NEXT:    s_and_b32 s0, vcc_lo, s0
 ; GFX1030-NEXT:    s_and_saveexec_b32 s0, s0
-; GFX1030-NEXT:    image_bvh_intersect_ray v[0:3], v[13:20], s[4:7] a16
-; GFX1030-NEXT:    ; implicit-def: $vgpr9
-; GFX1030-NEXT:    ; implicit-def: $vgpr13
-; GFX1030-NEXT:    ; implicit-def: $vgpr14
-; GFX1030-NEXT:    ; implicit-def: $vgpr15
+; GFX1030-NEXT:    image_bvh_intersect_ray v[0:3], v[18:25], s[4:7] a16
 ; GFX1030-NEXT:    ; implicit-def: $vgpr16
-; GFX1030-NEXT:    ; implicit-def: $vgpr17
 ; GFX1030-NEXT:    ; implicit-def: $vgpr18
 ; GFX1030-NEXT:    ; implicit-def: $vgpr19
 ; GFX1030-NEXT:    ; implicit-def: $vgpr20
+; GFX1030-NEXT:    ; implicit-def: $vgpr21
+; GFX1030-NEXT:    ; implicit-def: $vgpr22
+; GFX1030-NEXT:    ; implicit-def: $vgpr23
+; GFX1030-NEXT:    ; implicit-def: $vgpr24
+; GFX1030-NEXT:    ; implicit-def: $vgpr25
 ; GFX1030-NEXT:    ; implicit-def: $vgpr9_vgpr10_vgpr11_vgpr12
 ; GFX1030-NEXT:    s_xor_b32 exec_lo, exec_lo, s0
 ; GFX1030-NEXT:    s_cbranch_execnz .LBB7_1
@@ -296,26 +304,28 @@ define amdgpu_ps <4 x float> @image_bvh_intersect_ray_a16_vgpr_descr(i32 %node_p
 ;
 ; GFX1013-LABEL: image_bvh_intersect_ray_a16_vgpr_descr:
 ; GFX1013:       ; %bb.0:
-; GFX1013-NEXT:    v_lshrrev_b32_e32 v13, 16, v5
-; GFX1013-NEXT:    v_and_b32_e32 v14, 0xffff, v7
+; GFX1013-NEXT:    v_mov_b32_e32 v17, v9
+; GFX1013-NEXT:    v_mov_b32_e32 v18, v10
+; GFX1013-NEXT:    v_lshrrev_b32_e32 v9, 16, v5
+; GFX1013-NEXT:    v_and_b32_e32 v10, 0xffff, v7
 ; GFX1013-NEXT:    v_and_b32_e32 v8, 0xffff, v8
 ; GFX1013-NEXT:    s_mov_b32 s1, exec_lo
-; GFX1013-NEXT:    v_lshlrev_b32_e32 v13, 16, v13
-; GFX1013-NEXT:    v_lshlrev_b32_e32 v14, 16, v14
+; GFX1013-NEXT:    v_lshlrev_b32_e32 v9, 16, v9
+; GFX1013-NEXT:    v_lshlrev_b32_e32 v10, 16, v10
 ; GFX1013-NEXT:    v_alignbit_b32 v7, v8, v7, 16
-; GFX1013-NEXT:    v_and_or_b32 v5, 0xffff, v5, v13
-; GFX1013-NEXT:    v_and_or_b32 v6, 0xffff, v6, v14
+; GFX1013-NEXT:    v_and_or_b32 v5, 0xffff, v5, v9
+; GFX1013-NEXT:    v_and_or_b32 v6, 0xffff, v6, v10
 ; GFX1013-NEXT:  .LBB7_1: ; =>This Inner Loop Header: Depth=1
-; GFX1013-NEXT:    v_readfirstlane_b32 s4, v9
-; GFX1013-NEXT:    v_readfirstlane_b32 s5, v10
+; GFX1013-NEXT:    v_readfirstlane_b32 s4, v17
+; GFX1013-NEXT:    v_readfirstlane_b32 s5, v18
 ; GFX1013-NEXT:    v_readfirstlane_b32 s6, v11
 ; GFX1013-NEXT:    v_readfirstlane_b32 s7, v12
-; GFX1013-NEXT:    v_cmp_eq_u64_e32 vcc_lo, s[4:5], v[9:10]
+; GFX1013-NEXT:    v_cmp_eq_u64_e32 vcc_lo, s[4:5], v[17:18]
 ; GFX1013-NEXT:    v_cmp_eq_u64_e64 s0, s[6:7], v[11:12]
 ; GFX1013-NEXT:    s_and_b32 s0, vcc_lo, s0
 ; GFX1013-NEXT:    s_and_saveexec_b32 s0, s0
 ; GFX1013-NEXT:    image_bvh_intersect_ray v[13:16], v[0:7], s[4:7] a16
-; GFX1013-NEXT:    ; implicit-def: $vgpr9
+; GFX1013-NEXT:    ; implicit-def: $vgpr17
 ; GFX1013-NEXT:    ; implicit-def: $vgpr0_vgpr1_vgpr2_vgpr3_vgpr4_vgpr5_vgpr6_vgpr7
 ; GFX1013-NEXT:    ; implicit-def: $vgpr9_vgpr10_vgpr11_vgpr12
 ; GFX1013-NEXT:    s_waitcnt_depctr 0xffe3
@@ -333,27 +343,28 @@ define amdgpu_ps <4 x float> @image_bvh_intersect_ray_a16_vgpr_descr(i32 %node_p
 ; GFX11-LABEL: image_bvh_intersect_ray_a16_vgpr_descr:
 ; GFX11:       ; %bb.0:
 ; GFX11-NEXT:    v_dual_mov_b32 v16, v0 :: v_dual_mov_b32 v17, v1
-; GFX11-NEXT:    v_dual_mov_b32 v15, v4 :: v_dual_and_b32 v0, 0xffff, v7
+; GFX11-NEXT:    v_dual_mov_b32 v19, v10 :: v_dual_and_b32 v0, 0xffff, v7
 ; GFX11-NEXT:    v_and_b32_e32 v1, 0xffff, v8
 ; GFX11-NEXT:    v_dual_mov_b32 v13, v2 :: v_dual_mov_b32 v14, v3
-; GFX11-NEXT:    s_mov_b32 s1, exec_lo
-; GFX11-NEXT:    s_delay_alu instid0(VALU_DEP_3)
+; GFX11-NEXT:    v_dual_mov_b32 v15, v4 :: v_dual_mov_b32 v18, v9
+; GFX11-NEXT:    s_delay_alu instid0(VALU_DEP_4)
 ; GFX11-NEXT:    v_lshl_or_b32 v4, v5, 16, v0
 ; GFX11-NEXT:    v_perm_b32 v5, v5, v7, 0x7060302
 ; GFX11-NEXT:    v_lshl_or_b32 v6, v6, 16, v1
+; GFX11-NEXT:    s_mov_b32 s1, exec_lo
 ; GFX11-NEXT:  .LBB7_1: ; =>This Inner Loop Header: Depth=1
-; GFX11-NEXT:    v_readfirstlane_b32 s4, v9
-; GFX11-NEXT:    v_readfirstlane_b32 s5, v10
+; GFX11-NEXT:    v_readfirstlane_b32 s4, v18
+; GFX11-NEXT:    v_readfirstlane_b32 s5, v19
 ; GFX11-NEXT:    v_readfirstlane_b32 s6, v11
 ; GFX11-NEXT:    v_readfirstlane_b32 s7, v12
 ; GFX11-NEXT:    s_delay_alu instid0(VALU_DEP_3) | instskip(NEXT) | instid1(VALU_DEP_2)
-; GFX11-NEXT:    v_cmp_eq_u64_e32 vcc_lo, s[4:5], v[9:10]
+; GFX11-NEXT:    v_cmp_eq_u64_e32 vcc_lo, s[4:5], v[18:19]
 ; GFX11-NEXT:    v_cmp_eq_u64_e64 s0, s[6:7], v[11:12]
 ; GFX11-NEXT:    s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(SALU_CYCLE_1)
 ; GFX11-NEXT:    s_and_b32 s0, vcc_lo, s0
 ; GFX11-NEXT:    s_and_saveexec_b32 s0, s0
 ; GFX11-NEXT:    image_bvh_intersect_ray v[0:3], [v16, v17, v[13:15], v[4:6]], s[4:7] a16
-; GFX11-NEXT:    ; implicit-def: $vgpr9
+; GFX11-NEXT:    ; implicit-def: $vgpr18
 ; GFX11-NEXT:    ; implicit-def: $vgpr16
 ; GFX11-NEXT:    ; implicit-def: $vgpr17
 ; GFX11-NEXT:    ; implicit-def: $vgpr13_vgpr14_vgpr15
@@ -373,42 +384,44 @@ define amdgpu_ps <4 x float> @image_bvh_intersect_ray_a16_vgpr_descr(i32 %node_p
 define amdgpu_ps <4 x float> @image_bvh64_intersect_ray_vgpr_descr(i64 %node_ptr, float %ray_extent, <3 x float> %ray_origin, <3 x float> %ray_dir, <3 x float> %ray_inv_dir, <4 x i32> %tdescr) {
 ; GFX1030-LABEL: image_bvh64_intersect_ray_vgpr_descr:
 ; GFX1030:       ; %bb.0:
-; GFX1030-NEXT:    v_mov_b32_e32 v16, v0
-; GFX1030-NEXT:    v_mov_b32_e32 v17, v1
-; GFX1030-NEXT:    v_mov_b32_e32 v18, v2
-; GFX1030-NEXT:    v_mov_b32_e32 v19, v3
-; GFX1030-NEXT:    v_mov_b32_e32 v20, v4
-; GFX1030-NEXT:    v_mov_b32_e32 v21, v5
-; GFX1030-NEXT:    v_mov_b32_e32 v22, v6
-; GFX1030-NEXT:    v_mov_b32_e32 v23, v7
-; GFX1030-NEXT:    v_mov_b32_e32 v24, v8
-; GFX1030-NEXT:    v_mov_b32_e32 v25, v9
-; GFX1030-NEXT:    v_mov_b32_e32 v26, v10
-; GFX1030-NEXT:    v_mov_b32_e32 v27, v11
+; GFX1030-NEXT:    v_mov_b32_e32 v22, v0
+; GFX1030-NEXT:    v_mov_b32_e32 v23, v1
+; GFX1030-NEXT:    v_mov_b32_e32 v24, v2
+; GFX1030-NEXT:    v_mov_b32_e32 v25, v3
+; GFX1030-NEXT:    v_mov_b32_e32 v26, v4
+; GFX1030-NEXT:    v_mov_b32_e32 v27, v5
+; GFX1030-NEXT:    v_mov_b32_e32 v28, v6
+; GFX1030-NEXT:    v_mov_b32_e32 v29, v7
+; GFX1030-NEXT:    v_mov_b32_e32 v30, v8
+; GFX1030-NEXT:    v_mov_b32_e32 v31, v9
+; GFX1030-NEXT:    v_mov_b32_e32 v32, v10
+; GFX1030-NEXT:    v_mov_b32_e32 v33, v11
+; GFX1030-NEXT:    v_mov_b32_e32 v20, v12
+; GFX1030-NEXT:    v_mov_b32_e32 v21, v13
 ; GFX1030-NEXT:    s_mov_b32 s1, exec_lo
 ; GFX1030-NEXT:  .LBB8_1: ; =>This Inner Loop Header: Depth=1
-; GFX1030-NEXT:    v_readfirstlane_b32 s4, v12
-; GFX1030-NEXT:    v_readfirstlane_b32 s5, v13
+; GFX1030-NEXT:    v_readfirstlane_b32 s4, v20
+; GFX1030-NEXT:    v_readfirstlane_b32 s5, v21
 ; GFX1030-NEXT:    v_readfirstlane_b32 s6, v14
 ; GFX1030-NEXT:    v_readfirstlane_b32 s7, v15
-; GFX1030-NEXT:    v_cmp_eq_u64_e32 vcc_lo, s[4:5], v[12:13]
+; GFX1030-NEXT:    v_cmp_eq_u64_e32 vcc_lo, s[4:5], v[20:21]
 ; GFX1030-NEXT:    v_cmp_eq_u64_e64 s0, s[6:7], v[14:15]
 ; GFX1030-NEXT:    s_and_b32 s0, vcc_lo, s0
 ; GFX1030-NEXT:    s_and_saveexec_b32 s0, s0
-; GFX1030-NEXT:    image_bvh64_intersect_ray v[0:3], v[16:27], s[4:7]
-; GFX1030-NEXT:    ; implicit-def: $vgpr12
-; GFX1030-NEXT:    ; implicit-def: $vgpr16
-; GFX1030-NEXT:    ; implicit-def: $vgpr17
-; GFX1030-NEXT:    ; implicit-def: $vgpr18
-; GFX1030-NEXT:    ; implicit-def: $vgpr19
+; GFX1030-NEXT:    image_bvh64_intersect_ray v[0:3], v[22:33], s[4:7]
 ; GFX1030-NEXT:    ; implicit-def: $vgpr20
-; GFX1030-NEXT:    ; implicit-def: $vgpr21
 ; GFX1030-NEXT:    ; implicit-def: $vgpr22
 ; GFX1030-NEXT:    ; implicit-def: $vgpr23
 ; GFX1030-NEXT:    ; implicit-def: $vgpr24
 ; GFX1030-NEXT:    ; implicit-def: $vgpr25
 ; GFX1030-NEXT:    ; implicit-def: $vgpr26
 ; GFX1030-NEXT:    ; implicit-def: $vgpr27
+; GFX1030-NEXT:    ; implicit-def: $vgpr28
+; GFX1030-NEXT:    ; implicit-def: $vgpr29
+; GFX1030-NEXT:    ; implicit-def: $vgpr30
+; GFX1030-NEXT:    ; implicit-def: $vgpr31
+; GFX1030-NEXT:    ; implicit-def: $vgpr32
+; GFX1030-NEXT:    ; implicit-def: $vgpr33
 ; GFX1030-NEXT:    ; implicit-def: $vgpr12_vgpr13_vgpr14_vgpr15
 ; GFX1030-NEXT:    s_xor_b32 exec_lo, exec_lo, s0
 ; GFX1030-NEXT:    s_cbranch_execnz .LBB8_1
@@ -419,18 +432,20 @@ define amdgpu_ps <4 x float> @image_bvh64_intersect_ray_vgpr_descr(i64 %node_ptr
 ;
 ; GFX1013-LABEL: image_bvh64_intersect_ray_vgpr_descr:
 ; GFX1013:       ; %bb.0:
+; GFX1013-NEXT:    v_mov_b32_e32 v20, v12
+; GFX1013-NEXT:    v_mov_b32_e32 v21, v13
 ; GFX1013-NEXT:    s_mov_b32 s1, exec_lo
 ; GFX1013-NEXT:  .LBB8_1: ; =>This Inner Loop Header: Depth=1
-; GFX1013-NEXT:    v_readfirstlane_b32 s4, v12
-; GFX1013-NEXT:    v_readfirstlane_b32 s5, v13
+; GFX1013-NEXT:    v_readfirstlane_b32 s4, v20
+; GFX1013-NEXT:    v_readfirstlane_b32 s5, v21
 ; GFX1013-NEXT:    v_readfirstlane_b32 s6, v14
 ; GFX1013-NEXT:    v_readfirstlane_b32 s7, v15
-; GFX1013-NEXT:    v_cmp_eq_u64_e32 vcc_lo, s[4:5], v[12:13]
+; GFX1013-NEXT:    v_cmp_eq_u64_e32 vcc_lo, s[4:5], v[20:21]
 ; GFX1013-NEXT:    v_cmp_eq_u64_e64 s0, s[6:7], v[14:15]
 ; GFX1013-NEXT:    s_and_b32 s0, vcc_lo, s0
 ; GFX1013-NEXT:    s_and_saveexec_b32 s0, s0
 ; GFX1013-NEXT:    image_bvh64_intersect_ray v[16:19], v[0:11], s[4:7]
-; GFX1013-NEXT:    ; implicit-def: $vgpr12
+; GFX1013-NEXT:    ; implicit-def: $vgpr20
 ; GFX1013-NEXT:    ; implicit-def: $vgpr0_vgpr1_vgpr2_vgpr3_vgpr4_vgpr5_vgpr6_vgpr7_vgpr8_vgpr9_vgpr10_vgpr11
 ; GFX1013-NEXT:    ; implicit-def: $vgpr12_vgpr13_vgpr14_vgpr15
 ; GFX1013-NEXT:    s_waitcnt_depctr 0xffe3
@@ -450,20 +465,22 @@ define amdgpu_ps <4 x float> @image_bvh64_intersect_ray_vgpr_descr(i64 %node_ptr
 ; GFX11-NEXT:    v_dual_mov_b32 v19, v0 :: v_dual_mov_b32 v20, v1
 ; GFX11-NEXT:    v_dual_mov_b32 v21, v2 :: v_dual_mov_b32 v16, v3
 ; GFX11-NEXT:    v_dual_mov_b32 v17, v4 :: v_dual_mov_b32 v18, v5
+; GFX11-NEXT:    v_dual_mov_b32 v4, v12 :: v_dual_mov_b32 v5, v13
 ; GFX11-NEXT:    s_mov_b32 s1, exec_lo
 ; GFX11-NEXT:  .LBB8_1: ; =>This Inner Loop Header: Depth=1
-; GFX11-NEXT:    v_readfirstlane_b32 s4, v12
-; GFX11-NEXT:    v_readfirstlane_b32 s5, v13
+; GFX11-NEXT:    s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_2)
+; GFX11-NEXT:    v_readfirstlane_b32 s4, v4
+; GFX11-NEXT:    v_readfirstlane_b32 s5, v5
 ; GFX11-NEXT:    v_readfirstlane_b32 s6, v14
 ; GFX11-NEXT:    v_readfirstlane_b32 s7, v15
 ; GFX11-NEXT:    s_delay_alu instid0(VALU_DEP_3) | instskip(NEXT) | instid1(VALU_DEP_2)
-; GFX11-NEXT:    v_cmp_eq_u64_e32 vcc_lo, s[4:5], v[12:13]
+; GFX11-NEXT:    v_cmp_eq_u64_e32 vcc_lo, s[4:5], v[4:5]
 ; GFX11-NEXT:    v_cmp_eq_u64_e64 s0, s[6:7], v[14:15]
 ; GFX11-NEXT:    s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(SALU_CYCLE_1)
 ; GFX11-NEXT:    s_and_b32 s0, vcc_lo, s0
 ; GFX11-NEXT:    s_and_saveexec_b32 s0, s0
 ; GFX11-NEXT:    image_bvh64_intersect_ray v[0:3], [v[19:20], v21, v[16:18], v[6:8], v[9:11]], s[4:7]
-; GFX11-NEXT:    ; implicit-def: $vgpr12
+; GFX11-NEXT:    ; implicit-def: $vgpr4
 ; GFX11-NEXT:    ; implicit-def: $vgpr19_vgpr20
 ; GFX11-NEXT:    ; implicit-def: $vgpr21
 ; GFX11-NEXT:    ; implicit-def: $vgpr16_vgpr17_vgpr18
@@ -484,41 +501,43 @@ define amdgpu_ps <4 x float> @image_bvh64_intersect_ray_vgpr_descr(i64 %node_ptr
 define amdgpu_ps <4 x float> @image_bvh64_intersect_ray_a16_vgpr_descr(i64 %node_ptr, float %ray_extent, <3 x float> %ray_origin, <3 x half> %ray_dir, <3 x half> %ray_inv_dir, <4 x i32> %tdescr) {
 ; GFX1030-LABEL: image_bvh64_intersect_ray_a16_vgpr_descr:
 ; GFX1030:       ; %bb.0:
-; GFX1030-NEXT:    v_mov_b32_e32 v14, v0
-; GFX1030-NEXT:    v_mov_b32_e32 v15, v1
+; GFX1030-NEXT:    v_mov_b32_e32 v19, v0
+; GFX1030-NEXT:    v_mov_b32_e32 v20, v1
 ; GFX1030-NEXT:    v_lshrrev_b32_e32 v0, 16, v6
 ; GFX1030-NEXT:    v_and_b32_e32 v1, 0xffff, v8
-; GFX1030-NEXT:    v_mov_b32_e32 v16, v2
+; GFX1030-NEXT:    v_mov_b32_e32 v21, v2
 ; GFX1030-NEXT:    v_and_b32_e32 v2, 0xffff, v9
-; GFX1030-NEXT:    v_mov_b32_e32 v17, v3
+; GFX1030-NEXT:    v_mov_b32_e32 v22, v3
 ; GFX1030-NEXT:    v_lshlrev_b32_e32 v0, 16, v0
 ; GFX1030-NEXT:    v_lshlrev_b32_e32 v1, 16, v1
-; GFX1030-NEXT:    v_mov_b32_e32 v18, v4
-; GFX1030-NEXT:    v_mov_b32_e32 v19, v5
-; GFX1030-NEXT:    v_alignbit_b32 v22, v2, v8, 16
-; GFX1030-NEXT:    v_and_or_b32 v20, 0xffff, v6, v0
-; GFX1030-NEXT:    v_and_or_b32 v21, 0xffff, v7, v1
+; GFX1030-NEXT:    v_mov_b32_e32 v23, v4
+; GFX1030-NEXT:    v_mov_b32_e32 v24, v5
+; GFX1030-NEXT:    v_mov_b32_e32 v17, v10
+; GFX1030-NEXT:    v_mov_b32_e32 v18, v11
+; GFX1030-NEXT:    v_and_or_b32 v25, 0xffff, v6, v0
+; GFX1030-NEXT:    v_and_or_b32 v26, 0xffff, v7, v1
+; GFX1030-NEXT:    v_alignbit_b32 v27, v2, v8, 16
 ; GFX1030-NEXT:    s_mov_b32 s1, exec_lo
 ; GFX1030-NEXT:  .LBB9_1: ; =>This Inner Loop Header: Depth=1
-; GFX1030-NEXT:    v_readfirstlane_b32 s4, v10
-; GFX1030-NEXT:    v_readfirstlane_b32 s5, v11
+; GFX1030-NEXT:    v_readfirstlane_b32 s4, v17
+; GFX1030-NEXT:    v_readfirstlane_b32 s5, v18
 ; GFX1030-NEXT:    v_readfirstlane_b32 s6, v12
 ; GFX1030-NEXT:    v_readfirstlane_b32 s7, v13
-; GFX1030-NEXT:    v_cmp_eq_u64_e32 vcc_lo, s[4:5], v[10:11]
+; GFX1030-NEXT:    v_cmp_eq_u64_e32 vcc_lo, s[4:5], v[17:18]
 ; GFX1030-NEXT:    v_cmp_eq_u64_e64 s0, s[6:7], v[12:13]
 ; GFX1030-NEXT:    s_and_b32 s0, vcc_lo, s0
 ; GFX1030-NEXT:    s_and_saveexec_b32 s0, s0
-; GFX1030-NEXT:    image_bvh64_intersect_ray v[0:3], v[14:22], s[4:7] a16
-; GFX1030-NEXT:    ; implicit-def: $vgpr10
-; GFX1030-NEXT:    ; implicit-def: $vgpr14
-; GFX1030-NEXT:    ; implicit-def: $vgpr15
-; GFX1030-NEXT:    ; implicit-def: $vgpr16
+; GFX1030-NEXT:    image_bvh64_intersect_ray v[0:3], v[19:27], s[4:7] a16
 ; GFX1030-NEXT:    ; implicit-def: $vgpr17
-; GFX1030-NEXT:    ; implicit-def: $vgpr18
 ; GFX1030-NEXT:    ; implicit-def: $vgpr19
 ; GFX1030-NEXT:    ; implicit-def: $vgpr20
 ; GFX1030-NEXT:    ; implicit-def: $vgpr21
 ; GFX1030-NEXT:    ; implicit-def: $vgpr22
+; GFX1030-NEXT:    ; implicit-def: $vgpr23
+; GFX1030-NEXT:    ; implicit-def: $vgpr24
+; GFX1030-NEXT:    ; implicit-def: $vgpr25
+; GFX1030-NEXT:    ; implicit-def: $vgpr26
+; GFX1030-NEXT:    ; implicit-def: $vgpr27
 ; GFX1030-NEXT:    ; implicit-def: $vgpr10_vgpr11_vgpr12_vgpr13
 ; GFX1030-NEXT:    s_xor_b32 exec_lo, exec_lo, s0
 ; GFX1030-NEXT:    s_cbranch_execnz .LBB9_1
@@ -529,26 +548,28 @@ define amdgpu_ps <4 x float> @image_bvh64_intersect_ray_a16_vgpr_descr(i64 %node
 ;
 ; GFX1013-LABEL: image_bvh64_intersect_ray_a16_vgpr_descr:
 ; GFX1013:       ; %bb.0:
-; GFX1013-NEXT:    v_lshrrev_b32_e32 v14, 16, v6
-; GFX1013-NEXT:    v_and_b32_e32 v15, 0xffff, v8
+; GFX1013-NEXT:    v_mov_b32_e32 v18, v10
+; GFX1013-NEXT:    v_mov_b32_e32 v19, v11
+; GFX1013-NEXT:    v_lshrrev_b32_e32 v10, 16, v6
+; GFX1013-NEXT:    v_and_b32_e32 v11, 0xffff, v8
 ; GFX1013-NEXT:    v_and_b32_e32 v9, 0xffff, v9
 ; GFX1013-NEXT:    s_mov_b32 s1, exec_lo
-; GFX1013-NEXT:    v_lshlrev_b32_e32 v14, 16, v14
-; GFX1013-NEXT:    v_lshlrev_b32_e32 v15, 16, v15
+; GFX1013-NEXT:    v_lshlrev_b32_e32 v10, 16, v10
+; GFX1013-NEXT:    v_lshlrev_b32_e32 v11, 16, v11
 ; GFX1013-NEXT:    v_alignbit_b32 v8, v9, v8, 16
-; GFX1013-NEXT:    v_and_or_b32 v6, 0xffff, v6, v14
-; GFX1013-NEXT:    v_and_or_b32 v7, 0xffff, v7, v15
+; GFX1013-NEXT:    v_and_or_b32 v6, 0xffff, v6, v10
+; GFX1013-NEXT:    v_and_or_b32 v7, 0xffff, v7, v11
 ; GFX1013-NEXT:  .LBB9_1: ; =>This Inner Loop Header: Depth=1
-; GFX1013-NEXT:    v_readfirstlane_b32 s4, v10
-; GFX1013-NEXT:    v_readfirstlane_b32 s5, v11
+; GFX1013-NEXT:    v_readfirstlane_b32 s4, v18
+; GFX1013-NEXT:    v_readfirstlane_b32 s5, v19
 ; GFX1013-NEXT:    v_readfirstlane_b32 s6, v12
 ; GFX1013-NEXT:    v_readfirstlane_b32 s7, v13
-; GFX1013-NEXT:    v_cmp_eq_u64_e32 vcc_lo, s[4:5], v[10:11]
+; GFX1013-NEXT:    v_cmp_eq_u64_e32 vcc_lo, s[4:5], v[18:19]
 ; GFX1013-NEXT:    v_cmp_eq_u64_e64 s0, s[6:7], v[12:13]
 ; GFX1013-NEXT:    s_and_b32 s0, vcc_lo, s0
 ; GFX1013-NEXT:    s_and_saveexec_b32 s0, s0
 ; GFX1013-NEXT:    image_bvh64_intersect_ray v[14:17], v[0:8], s[4:7] a16
-; GFX1013-NEXT:    ; implicit-def: $vgpr10
+; GFX1013-NEXT:    ; implicit-def: $vgpr18
 ; GFX1013-NEXT:    ; implicit-def: $vgpr0_vgpr1_vgpr2_vgpr3_vgpr4_vgpr5_vgpr6_vgpr7_vgpr8
 ; GFX1013-NEXT:    ; implicit-def: $vgpr10_vgpr11_vgpr12_vgpr13
 ; GFX1013-NEXT:    s_waitcnt_depctr 0xffe3
@@ -570,28 +591,28 @@ define amdgpu_ps <4 x float> @image_bvh64_intersect_ray_a16_vgpr_descr(i64 %node
 ; GFX11-NEXT:    v_and_b32_e32 v1, 0xffff, v9
 ; GFX11-NEXT:    v_dual_mov_b32 v19, v2 :: v_dual_mov_b32 v14, v3
 ; GFX11-NEXT:    v_dual_mov_b32 v15, v4 :: v_dual_mov_b32 v16, v5
-; GFX11-NEXT:    s_delay_alu instid0(VALU_DEP_4)
-; GFX11-NEXT:    v_lshl_or_b32 v4, v6, 16, v0
-; GFX11-NEXT:    v_perm_b32 v5, v6, v8, 0x7060302
-; GFX11-NEXT:    v_lshl_or_b32 v6, v7, 16, v1
+; GFX11-NEXT:    v_dual_mov_b32 v4, v10 :: v_dual_mov_b32 v5, v11
+; GFX11-NEXT:    v_lshl_or_b32 v20, v6, 16, v0
+; GFX11-NEXT:    v_perm_b32 v21, v6, v8, 0x7060302
+; GFX11-NEXT:    v_lshl_or_b32 v22, v7, 16, v1
 ; GFX11-NEXT:    s_mov_b32 s1, exec_lo
 ; GFX11-NEXT:  .LBB9_1: ; =>This Inner Loop Header: Depth=1
-; GFX11-NEXT:    v_readfirstlane_b32 s4, v10
-; GFX11-NEXT:    v_readfirstlane_b32 s5, v11
+; GFX11-NEXT:    v_readfirstlane_b32 s4, v4
+; GFX11-NEXT:    v_readfirstlane_b32 s5, v5
 ; GFX11-NEXT:    v_readfirstlane_b32 s6, v12
 ; GFX11-NEXT:    v_readfirstlane_b32 s7, v13
 ; GFX11-NEXT:    s_delay_alu instid0(VALU_DEP_3) | instskip(NEXT) | instid1(VALU_DEP_2)
-; GFX11-NEXT:    v_cmp_eq_u64_e32 vcc_lo, s[4:5], v[10:11]
+; GFX11-NEXT:    v_cmp_eq_u64_e32 vcc_lo, s[4:5], v[4:5]
 ; GFX11-NEXT:    v_cmp_eq_u64_e64 s0, s[6:7], v[12:13]
 ; GFX11-NEXT:    s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(SALU_CYCLE_1)
 ; GFX11-NEXT:    s_and_b32 s0, vcc_lo, s0
 ; GFX11-NEXT:    s_and_saveexec_b32 s0, s0
-; GFX11-NEXT:    image_bvh64_intersect_ray v[0:3], [v[17:18], v19, v[14:16], v[4:6]], s[4:7] a16
-; GFX11-NEXT:    ; implicit-def: $vgpr10
+; GFX11-NEXT:    image_bvh64_intersect_ray v[0:3], [v[17:18], v19, v[14:16], v[20:22]], s[4:7] a16
+; GFX11-NEXT:    ; implicit-def: $vgpr4
 ; GFX11-NEXT:    ; implicit-def: $vgpr17_vgpr18
 ; GFX11-NEXT:    ; implicit-def: $vgpr19
 ; GFX11-NEXT:    ; implicit-def: $vgpr14_vgpr15_vgpr16
-; GFX11-NEXT:    ; implicit-def: $vgpr4_vgpr5_vgpr6
+; GFX11-NEXT:    ; implicit-def: $vgpr20_vgpr21_vgpr22
 ; GFX11-NEXT:    ; implicit-def: $vgpr10_vgpr11_vgpr12_vgpr13
 ; GFX11-NEXT:    s_xor_b32 exec_lo, exec_lo, s0
 ; GFX11-NEXT:    s_cbranch_execnz .LBB9_1
diff --git a/llvm/test/CodeGen/AMDGPU/GlobalISel/mmra.ll b/llvm/test/CodeGen/AMDGPU/GlobalISel/mmra.ll
new file mode 100644
index 000000000000..71a2d3e8a530
--- /dev/null
+++ b/llvm/test/CodeGen/AMDGPU/GlobalISel/mmra.ll
@@ -0,0 +1,34 @@
+; NOTE: Assertions have been autogenerated by utils/update_mir_test_checks.py UTC_ARGS: --version 2
+; RUN: llc -global-isel -march=amdgcn -mcpu=gfx900 -stop-after=finalize-isel < %s | FileCheck %s
+
+declare void @readsMem(ptr) #0
+declare void @writesMem(ptr) #1
+
+define void @fence_loads(ptr %ptr) {
+  ; CHECK-LABEL: name: fence_loads
+  ; CHECK: bb.1 (%ir-block.0):
+  ; CHECK-NEXT:   liveins: $vgpr0, $vgpr1
+  ; CHECK-NEXT: {{  $}}
+  ; CHECK-NEXT:   [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr0
+  ; CHECK-NEXT:   [[COPY1:%[0-9]+]]:vgpr_32 = COPY $vgpr1
+  ; CHECK-NEXT:   [[REG_SEQUENCE:%[0-9]+]]:vreg_64 = REG_SEQUENCE [[COPY]], %subreg.sub0, [[COPY1]], %subreg.sub1
+  ; CHECK-NEXT:   ATOMIC_FENCE 5, 1, mmra !0
+  ; CHECK-NEXT:   [[FLAT_LOAD_UBYTE:%[0-9]+]]:vgpr_32 = FLAT_LOAD_UBYTE [[REG_SEQUENCE]], 0, 0, implicit $exec, implicit $flat_scr, mmra !1 :: (load acquire (s8) from %ir.ptr, align 4)
+  ; CHECK-NEXT:   [[S_MOV_B32_:%[0-9]+]]:sreg_32 = S_MOV_B32 1
+  ; CHECK-NEXT:   [[COPY2:%[0-9]+]]:vgpr_32 = COPY [[S_MOV_B32_]]
+  ; CHECK-NEXT:   FLAT_STORE_BYTE [[REG_SEQUENCE]], [[COPY2]], 0, 0, implicit $exec, implicit $flat_scr, mmra !2 :: (store release (s8) into %ir.ptr, align 4)
+  ; CHECK-NEXT:   SI_RETURN
+  fence release,                                        !mmra !0
+  %ld = load atomic i8, ptr %ptr acquire, align 4,      !mmra !2
+  store atomic i8 1, ptr %ptr release, align 4,         !mmra !1
+  ret void
+}
+
+; TODO: test atomicrmw, cmpxchg - current lowering doesn't work and blows up on i1 PHIs.
+
+attributes #0 = { memory(read) }
+attributes #1 = { memory(write) }
+
+!0 = !{!"foo", !"bar"}
+!1 = !{!"bux", !"baz"}
+!2 = !{!0, !1}
diff --git a/llvm/test/CodeGen/AMDGPU/amdpal-callable.ll b/llvm/test/CodeGen/AMDGPU/amdpal-callable.ll
index b7b2cb22c1b6..9d4f9434aa31 100644
--- a/llvm/test/CodeGen/AMDGPU/amdpal-callable.ll
+++ b/llvm/test/CodeGen/AMDGPU/amdpal-callable.ll
@@ -142,8 +142,8 @@ attributes #0 = { nounwind }
 
 ; GCN: amdpal.pipelines:
 ; GCN-NEXT:  - .registers:
-; GCN-NEXT:      0x2e12 (COMPUTE_PGM_RSRC1): 0xaf01ca{{$}}
-; GCN-NEXT:      0x2e13 (COMPUTE_PGM_RSRC2): 0x8001{{$}}
+; GCN-NEXT:      '0x2e12 (COMPUTE_PGM_RSRC1)': 0xaf01ca{{$}}
+; GCN-NEXT:      '0x2e13 (COMPUTE_PGM_RSRC2)': 0x8001{{$}}
 ; GCN-NEXT:    .shader_functions:
 ; GCN-NEXT:      dynamic_stack:
 ; GCN-NEXT:        .backend_stack_size: 0x10{{$}}
diff --git a/llvm/test/CodeGen/AMDGPU/amdpal-cs.ll b/llvm/test/CodeGen/AMDGPU/amdpal-cs.ll
index 98aa04f6d26e..a3fd2a942bc2 100644
--- a/llvm/test/CodeGen/AMDGPU/amdpal-cs.ll
+++ b/llvm/test/CodeGen/AMDGPU/amdpal-cs.ll
@@ -11,8 +11,8 @@
 ; GCN-NEXT:         .entry_point:    cs_amdpal
 ; GCN-NEXT:         .scratch_memory_size: 0
 ; GCN:     .registers:
-; GCN-NEXT:       0x2e12 (COMPUTE_PGM_RSRC1):
-; GCN-NEXT:       0x2e13 (COMPUTE_PGM_RSRC2):
+; GCN-NEXT:       '0x2e12 (COMPUTE_PGM_RSRC1)':
+; GCN-NEXT:       '0x2e13 (COMPUTE_PGM_RSRC2)':
 ; GCN-NEXT: ...
 ; GCN-NEXT:         .end_amdgpu_pal_metadata
 define amdgpu_cs half @cs_amdpal(half %arg0) {
diff --git a/llvm/test/CodeGen/AMDGPU/amdpal-es.ll b/llvm/test/CodeGen/AMDGPU/amdpal-es.ll
index 012b2061756b..679e0858819e 100644
--- a/llvm/test/CodeGen/AMDGPU/amdpal-es.ll
+++ b/llvm/test/CodeGen/AMDGPU/amdpal-es.ll
@@ -10,7 +10,7 @@
 ; GCN-NEXT:         .entry_point:    es_amdpal
 ; GCN-NEXT:         .scratch_memory_size: 0
 ; GCN:     .registers:
-; GCN-NEXT:       0x2cca (SPI_SHADER_PGM_RSRC1_ES): 0
+; GCN-NEXT:       '0x2cca (SPI_SHADER_PGM_RSRC1_ES)': 0
 ; GCN-NEXT: ...
 ; GCN-NEXT:         .end_amdgpu_pal_metadata
 define amdgpu_es half @es_amdpal(half %arg0) {
diff --git a/llvm/test/CodeGen/AMDGPU/amdpal-gs.ll b/llvm/test/CodeGen/AMDGPU/amdpal-gs.ll
index e2f67398d18a..75f7a1dc266d 100644
--- a/llvm/test/CodeGen/AMDGPU/amdpal-gs.ll
+++ b/llvm/test/CodeGen/AMDGPU/amdpal-gs.ll
@@ -11,7 +11,7 @@
 ; GCN-NEXT:         .entry_point:    gs_amdpal
 ; GCN-NEXT:         .scratch_memory_size: 0
 ; GCN:     .registers:
-; GCN-NEXT:       0x2c8a (SPI_SHADER_PGM_RSRC1_GS): 0
+; GCN-NEXT:       '0x2c8a (SPI_SHADER_PGM_RSRC1_GS)': 0
 ; GCN-NEXT: ...
 ; GCN-NEXT:         .end_amdgpu_pal_metadata
 define amdgpu_gs half @gs_amdpal(half %arg0) {
diff --git a/llvm/test/CodeGen/AMDGPU/amdpal-hs.ll b/llvm/test/CodeGen/AMDGPU/amdpal-hs.ll
index 9ad47c1d604f..c61578a967b6 100644
--- a/llvm/test/CodeGen/AMDGPU/amdpal-hs.ll
+++ b/llvm/test/CodeGen/AMDGPU/amdpal-hs.ll
@@ -11,7 +11,7 @@
 ; GCN-NEXT:         .entry_point:    hs_amdpal
 ; GCN-NEXT:         .scratch_memory_size: 0
 ; GCN:     .registers:
-; GCN-NEXT:       0x2d0a (SPI_SHADER_PGM_RSRC1_HS): 0
+; GCN-NEXT:       '0x2d0a (SPI_SHADER_PGM_RSRC1_HS)': 0
 ; GCN-NEXT: ...
 ; GCN-NEXT:         .end_amdgpu_pal_metadata
 define amdgpu_hs half @hs_amdpal(half %arg0) {
diff --git a/llvm/test/CodeGen/AMDGPU/amdpal-ls.ll b/llvm/test/CodeGen/AMDGPU/amdpal-ls.ll
index 8ee6f7283ce7..8162c824dc2c 100644
--- a/llvm/test/CodeGen/AMDGPU/amdpal-ls.ll
+++ b/llvm/test/CodeGen/AMDGPU/amdpal-ls.ll
@@ -10,7 +10,7 @@
 ; GCN-NEXT:         .entry_point:    ls_amdpal
 ; GCN-NEXT:         .scratch_memory_size: 0
 ; GCN:     .registers:
-; GCN-NEXT:       0x2d4a (SPI_SHADER_PGM_RSRC1_LS): 0
+; GCN-NEXT:       '0x2d4a (SPI_SHADER_PGM_RSRC1_LS)': 0
 ; GCN-NEXT: ...
 ; GCN-NEXT:         .end_amdgpu_pal_metadata
 define amdgpu_ls half @ls_amdpal(half %arg0) {
diff --git a/llvm/test/CodeGen/AMDGPU/amdpal-msgpack-cs.ll b/llvm/test/CodeGen/AMDGPU/amdpal-msgpack-cs.ll
index 0d0c70c38ace..5e21ba494df1 100644
--- a/llvm/test/CodeGen/AMDGPU/amdpal-msgpack-cs.ll
+++ b/llvm/test/CodeGen/AMDGPU/amdpal-msgpack-cs.ll
@@ -5,7 +5,7 @@
 ; amdpal compute shader: check for 0x2e12 (COMPUTE_PGM_RSRC1) in pal metadata
 ; GCN-LABEL: {{^}}cs_amdpal:
 ; GCN: .amdgpu_pal_metadata
-; GCN: 0x2e12 (COMPUTE_PGM_RSRC1)
+; GCN: '0x2e12 (COMPUTE_PGM_RSRC1)'
 define amdgpu_cs half @cs_amdpal(half %arg0) {
   %add = fadd half %arg0, 1.0
   ret half %add
diff --git a/llvm/test/CodeGen/AMDGPU/amdpal-msgpack-default.ll b/llvm/test/CodeGen/AMDGPU/amdpal-msgpack-default.ll
index b82e3ebdde4b..dc9a33ac0141 100644
--- a/llvm/test/CodeGen/AMDGPU/amdpal-msgpack-default.ll
+++ b/llvm/test/CodeGen/AMDGPU/amdpal-msgpack-default.ll
@@ -3,45 +3,45 @@
 ; RUN: llc -mtriple=amdgcn--amdpal -mcpu=gfx900 -verify-machineinstrs < %s | FileCheck -check-prefix=GFX9 -enable-var-scope %s
 
 ; amdpal compute shader: check for 0x2e12 (COMPUTE_PGM_RSRC1) in pal metadata
-; SI-DAG: 0x2e12 (COMPUTE_PGM_RSRC1): 0x2f0000{{$}}
-; VI-DAG: 0x2e12 (COMPUTE_PGM_RSRC1): 0x2f02c0{{$}}
-; GFX9-DAG: 0x2e12 (COMPUTE_PGM_RSRC1): 0x2f0000{{$}}
+; SI-DAG: '0x2e12 (COMPUTE_PGM_RSRC1)': 0x2f0000{{$}}
+; VI-DAG: '0x2e12 (COMPUTE_PGM_RSRC1)': 0x2f02c0{{$}}
+; GFX9-DAG: '0x2e12 (COMPUTE_PGM_RSRC1)': 0x2f0000{{$}}
 define amdgpu_cs half @cs_amdpal(half %arg0) {
   %add = fadd half %arg0, 1.0
   ret half %add
 }
 
 ; amdpal evaluation shader: check for 0x2cca (SPI_SHADER_PGM_RSRC1_ES) in pal metadata
-; SI-DAG: 0x2cca (SPI_SHADER_PGM_RSRC1_ES): 0x2f0000{{$}}
-; VI-DAG: 0x2cca (SPI_SHADER_PGM_RSRC1_ES): 0x2f02c0{{$}}
-; GFX9-DAG: 0x2cca (SPI_SHADER_PGM_RSRC1_ES): 0x2f0000{{$}}
+; SI-DAG: '0x2cca (SPI_SHADER_PGM_RSRC1_ES)': 0x2f0000{{$}}
+; VI-DAG: '0x2cca (SPI_SHADER_PGM_RSRC1_ES)': 0x2f02c0{{$}}
+; GFX9-DAG: '0x2cca (SPI_SHADER_PGM_RSRC1_ES)': 0x2f0000{{$}}
 define amdgpu_es half @es_amdpal(half %arg0) {
   %add = fadd half %arg0, 1.0
   ret half %add
 }
 
 ; amdpal geometry shader: check for 0x2c8a (SPI_SHADER_PGM_RSRC1_GS) in pal metadata
-; SI-DAG: 0x2c8a (SPI_SHADER_PGM_RSRC1_GS): 0x2f0000{{$}}
-; VI-DAG: 0x2c8a (SPI_SHADER_PGM_RSRC1_GS): 0x2f02c0{{$}}
-; GFX9-DAG: 0x2c8a (SPI_SHADER_PGM_RSRC1_GS): 0x2f0000{{$}}
+; SI-DAG: '0x2c8a (SPI_SHADER_PGM_RSRC1_GS)': 0x2f0000{{$}}
+; VI-DAG: '0x2c8a (SPI_SHADER_PGM_RSRC1_GS)': 0x2f02c0{{$}}
+; GFX9-DAG: '0x2c8a (SPI_SHADER_PGM_RSRC1_GS)': 0x2f0000{{$}}
 define amdgpu_gs half @gs_amdpal(half %arg0) {
   %add = fadd half %arg0, 1.0
   ret half %add
 }
 
 ; amdpal hull shader: check for 0x2d0a (SPI_SHADER_PGM_RSRC1_HS) in pal metadata
-; SI-DAG: 0x2d0a (SPI_SHADER_PGM_RSRC1_HS): 0x2f0000{{$}}
-; VI-DAG: 0x2d0a (SPI_SHADER_PGM_RSRC1_HS): 0x2f02c0{{$}}
-; GFX9-DAG: 0x2d0a (SPI_SHADER_PGM_RSRC1_HS): 0x2f0000{{$}}
+; SI-DAG: '0x2d0a (SPI_SHADER_PGM_RSRC1_HS)': 0x2f0000{{$}}
+; VI-DAG: '0x2d0a (SPI_SHADER_PGM_RSRC1_HS)': 0x2f02c0{{$}}
+; GFX9-DAG: '0x2d0a (SPI_SHADER_PGM_RSRC1_HS)': 0x2f0000{{$}}
 define amdgpu_hs half @hs_amdpal(half %arg0) {
   %add = fadd half %arg0, 1.0
   ret half %add
 }
 
 ; amdpal load shader: check for 0x2d4a (SPI_SHADER_PGM_RSRC1_LS) in pal metadata
-; SI-DAG: 0x2d4a (SPI_SHADER_PGM_RSRC1_LS): 0x2f0000{{$}}
-; VI-DAG: 0x2d4a (SPI_SHADER_PGM_RSRC1_LS): 0x2f02c0{{$}}
-; GFX9-DAG: 0x2d4a (SPI_SHADER_PGM_RSRC1_LS): 0x2f0000{{$}}
+; SI-DAG: '0x2d4a (SPI_SHADER_PGM_RSRC1_LS)': 0x2f0000{{$}}
+; VI-DAG: '0x2d4a (SPI_SHADER_PGM_RSRC1_LS)': 0x2f02c0{{$}}
+; GFX9-DAG: '0x2d4a (SPI_SHADER_PGM_RSRC1_LS)': 0x2f0000{{$}}
 define amdgpu_ls half @ls_amdpal(half %arg0) {
   %add = fadd half %arg0, 1.0
   ret half %add
@@ -49,18 +49,18 @@ define amdgpu_ls half @ls_amdpal(half %arg0) {
 
 ; amdpal pixel shader: check for 0x2c0a (SPI_SHADER_PGM_RSRC1_PS) in pal metadata
 ; below.
-; SI-DAG:           0x2c0a (SPI_SHADER_PGM_RSRC1_PS): 0x2f0000{{$}}
-; VI-DAG:           0x2c0a (SPI_SHADER_PGM_RSRC1_PS): 0x2f02c0{{$}}
-; GFX9-DAG:         0x2c0a (SPI_SHADER_PGM_RSRC1_PS): 0x2f0000{{$}}
+; SI-DAG:           '0x2c0a (SPI_SHADER_PGM_RSRC1_PS)': 0x2f0000{{$}}
+; VI-DAG:           '0x2c0a (SPI_SHADER_PGM_RSRC1_PS)': 0x2f02c0{{$}}
+; GFX9-DAG:         '0x2c0a (SPI_SHADER_PGM_RSRC1_PS)': 0x2f0000{{$}}
 define amdgpu_ps half @ps_amdpal(half %arg0) {
   %add = fadd half %arg0, 1.0
   ret half %add
 }
 
 ; amdpal vertex shader: check for 45352 (SPI_SHADER_PGM_RSRC1_VS) in pal metadata
-; SI-DAG: 0x2c4a (SPI_SHADER_PGM_RSRC1_VS): 0x2f0000{{$}}
-; VI-DAG: 0x2c4a (SPI_SHADER_PGM_RSRC1_VS): 0x2f02c0{{$}}
-; GFX9-DAG: 0x2c4a (SPI_SHADER_PGM_RSRC1_VS): 0x2f0000{{$}}
+; SI-DAG: '0x2c4a (SPI_SHADER_PGM_RSRC1_VS)': 0x2f0000{{$}}
+; VI-DAG: '0x2c4a (SPI_SHADER_PGM_RSRC1_VS)': 0x2f02c0{{$}}
+; GFX9-DAG: '0x2c4a (SPI_SHADER_PGM_RSRC1_VS)': 0x2f0000{{$}}
 define amdgpu_vs half @vs_amdpal(half %arg0) {
   %add = fadd half %arg0, 1.0
   ret half %add
@@ -75,7 +75,7 @@ define amdgpu_vs half @vs_amdpal(half %arg0) {
 ;       - 0x123456789abcdef0
 ;       - 0xfedcba9876543210
 ;     .registers:
-;       0x2c0b (SPI_SHADER_PGM_RSRC2_PS): 0x42000000
+;       '0x2c0b (SPI_SHADER_PGM_RSRC2_PS)': 0x42000000
 ; ...
 ; 	.end_amdgpu_pal_metadata
 
diff --git a/llvm/test/CodeGen/AMDGPU/amdpal-msgpack-denormal.ll b/llvm/test/CodeGen/AMDGPU/amdpal-msgpack-denormal.ll
index b86b42868005..ffce3ed08509 100644
--- a/llvm/test/CodeGen/AMDGPU/amdpal-msgpack-denormal.ll
+++ b/llvm/test/CodeGen/AMDGPU/amdpal-msgpack-denormal.ll
@@ -3,45 +3,45 @@
 ; RUN: llc -mtriple=amdgcn--amdpal -mcpu=gfx900 -verify-machineinstrs < %s | FileCheck -check-prefix=GFX9 -enable-var-scope %s
 
 ; amdpal compute shader: check for 0x2e12 (COMPUTE_PGM_RSRC1) in pal metadata
-; SI-DAG: 0x2e12 (COMPUTE_PGM_RSRC1): 0x2c0000{{$}}
-; VI-DAG: 0x2e12 (COMPUTE_PGM_RSRC1): 0x2c02c0{{$}}
-; GFX9-DAG: 0x2e12 (COMPUTE_PGM_RSRC1): 0x2c0000{{$}}
+; SI-DAG: '0x2e12 (COMPUTE_PGM_RSRC1)': 0x2c0000{{$}}
+; VI-DAG: '0x2e12 (COMPUTE_PGM_RSRC1)': 0x2c02c0{{$}}
+; GFX9-DAG: '0x2e12 (COMPUTE_PGM_RSRC1)': 0x2c0000{{$}}
 define amdgpu_cs half @cs_amdpal(half %arg0) #0 {
   %add = fadd half %arg0, 1.0
   ret half %add
 }
 
 ; amdpal evaluation shader: check for 0x2cca (SPI_SHADER_PGM_RSRC1_ES) in pal metadata
-; SI-DAG: 0x2cca (SPI_SHADER_PGM_RSRC1_ES): 0x2c0000{{$}}
-; VI-DAG: 0x2cca (SPI_SHADER_PGM_RSRC1_ES): 0x2c02c0{{$}}
-; GFX9-DAG: 0x2cca (SPI_SHADER_PGM_RSRC1_ES): 0x2c0000{{$}}
+; SI-DAG: '0x2cca (SPI_SHADER_PGM_RSRC1_ES)': 0x2c0000{{$}}
+; VI-DAG: '0x2cca (SPI_SHADER_PGM_RSRC1_ES)': 0x2c02c0{{$}}
+; GFX9-DAG: '0x2cca (SPI_SHADER_PGM_RSRC1_ES)': 0x2c0000{{$}}
 define amdgpu_es half @es_amdpal(half %arg0) #0 {
   %add = fadd half %arg0, 1.0
   ret half %add
 }
 
 ; amdpal geometry shader: check for 0x2c8a (SPI_SHADER_PGM_RSRC1_GS) in pal metadata
-; SI-DAG: 0x2c8a (SPI_SHADER_PGM_RSRC1_GS): 0x2c0000{{$}}
-; VI-DAG: 0x2c8a (SPI_SHADER_PGM_RSRC1_GS): 0x2c02c0{{$}}
-; GFX9-DAG: 0x2c8a (SPI_SHADER_PGM_RSRC1_GS): 0x2c0000{{$}}
+; SI-DAG: '0x2c8a (SPI_SHADER_PGM_RSRC1_GS)': 0x2c0000{{$}}
+; VI-DAG: '0x2c8a (SPI_SHADER_PGM_RSRC1_GS)': 0x2c02c0{{$}}
+; GFX9-DAG: '0x2c8a (SPI_SHADER_PGM_RSRC1_GS)': 0x2c0000{{$}}
 define amdgpu_gs half @gs_amdpal(half %arg0) #0 {
   %add = fadd half %arg0, 1.0
   ret half %add
 }
 
 ; amdpal hull shader: check for 0x2d0a (SPI_SHADER_PGM_RSRC1_HS) in pal metadata
-; SI-DAG: 0x2d0a (SPI_SHADER_PGM_RSRC1_HS): 0x2c0000{{$}}
-; VI-DAG: 0x2d0a (SPI_SHADER_PGM_RSRC1_HS): 0x2c02c0{{$}}
-; GFX9-DAG: 0x2d0a (SPI_SHADER_PGM_RSRC1_HS): 0x2c0000{{$}}
+; SI-DAG: '0x2d0a (SPI_SHADER_PGM_RSRC1_HS)': 0x2c0000{{$}}
+; VI-DAG: '0x2d0a (SPI_SHADER_PGM_RSRC1_HS)': 0x2c02c0{{$}}
+; GFX9-DAG: '0x2d0a (SPI_SHADER_PGM_RSRC1_HS)': 0x2c0000{{$}}
 define amdgpu_hs half @hs_amdpal(half %arg0) #0 {
   %add = fadd half %arg0, 1.0
   ret half %add
 }
 
 ; amdpal load shader: check for 0x2d4a (SPI_SHADER_PGM_RSRC1_LS) in pal metadata
-; SI-DAG: 0x2d4a (SPI_SHADER_PGM_RSRC1_LS): 0x2c0000{{$}}
-; VI-DAG: 0x2d4a (SPI_SHADER_PGM_RSRC1_LS): 0x2c02c0{{$}}
-; GFX9-DAG: 0x2d4a (SPI_SHADER_PGM_RSRC1_LS): 0x2c0000{{$}}
+; SI-DAG: '0x2d4a (SPI_SHADER_PGM_RSRC1_LS)': 0x2c0000{{$}}
+; VI-DAG: '0x2d4a (SPI_SHADER_PGM_RSRC1_LS)': 0x2c02c0{{$}}
+; GFX9-DAG: '0x2d4a (SPI_SHADER_PGM_RSRC1_LS)': 0x2c0000{{$}}
 define amdgpu_ls half @ls_amdpal(half %arg0) #0 {
   %add = fadd half %arg0, 1.0
   ret half %add
@@ -49,18 +49,18 @@ define amdgpu_ls half @ls_amdpal(half %arg0) #0 {
 
 ; amdpal pixel shader: check for 0x2c0a (SPI_SHADER_PGM_RSRC1_PS) in pal metadata
 ; below.
-; SI-DAG:           0x2c0a (SPI_SHADER_PGM_RSRC1_PS): 0x2c0000{{$}}
-; VI-DAG:           0x2c0a (SPI_SHADER_PGM_RSRC1_PS): 0x2c02c0{{$}}
-; GFX9-DAG:         0x2c0a (SPI_SHADER_PGM_RSRC1_PS): 0x2c0000{{$}}
+; SI-DAG:           '0x2c0a (SPI_SHADER_PGM_RSRC1_PS)': 0x2c0000{{$}}
+; VI-DAG:           '0x2c0a (SPI_SHADER_PGM_RSRC1_PS)': 0x2c02c0{{$}}
+; GFX9-DAG:         '0x2c0a (SPI_SHADER_PGM_RSRC1_PS)': 0x2c0000{{$}}
 define amdgpu_ps half @ps_amdpal(half %arg0) #0 {
   %add = fadd half %arg0, 1.0
   ret half %add
 }
 
 ; amdpal vertex shader: check for 45352 (SPI_SHADER_PGM_RSRC1_VS) in pal metadata
-; SI-DAG: 0x2c4a (SPI_SHADER_PGM_RSRC1_VS): 0x2c0000{{$}}
-; VI-DAG: 0x2c4a (SPI_SHADER_PGM_RSRC1_VS): 0x2c02c0{{$}}
-; GFX9-DAG: 0x2c4a (SPI_SHADER_PGM_RSRC1_VS): 0x2c0000{{$}}
+; SI-DAG: '0x2c4a (SPI_SHADER_PGM_RSRC1_VS)': 0x2c0000{{$}}
+; VI-DAG: '0x2c4a (SPI_SHADER_PGM_RSRC1_VS)': 0x2c02c0{{$}}
+; GFX9-DAG: '0x2c4a (SPI_SHADER_PGM_RSRC1_VS)': 0x2c0000{{$}}
 define amdgpu_vs half @vs_amdpal(half %arg0) #0 {
   %add = fadd half %arg0, 1.0
   ret half %add
@@ -77,7 +77,7 @@ attributes #0 = { "denormal-fp-math-f32"="preserve-sign,preserve-sign" }
 ;       - 0x123456789abcdef0
 ;       - 0xfedcba9876543210
 ;     .registers:
-;       0x2c0b (SPI_SHADER_PGM_RSRC2_PS): 0x42000000
+;       '0x2c0b (SPI_SHADER_PGM_RSRC2_PS)': 0x42000000
 ; ...
 ; 	.end_amdgpu_pal_metadata
 
diff --git a/llvm/test/CodeGen/AMDGPU/amdpal-msgpack-dx10-clamp.ll b/llvm/test/CodeGen/AMDGPU/amdpal-msgpack-dx10-clamp.ll
index b1db7aafacab..3ea3064fa743 100644
--- a/llvm/test/CodeGen/AMDGPU/amdpal-msgpack-dx10-clamp.ll
+++ b/llvm/test/CodeGen/AMDGPU/amdpal-msgpack-dx10-clamp.ll
@@ -3,45 +3,45 @@
 ; RUN: llc -mtriple=amdgcn--amdpal -mcpu=gfx900 -verify-machineinstrs < %s | FileCheck -check-prefix=GFX9 -enable-var-scope %s
 
 ; amdpal compute shader: check for 0x2e12 (COMPUTE_PGM_RSRC1) in pal metadata
-; SI-DAG: 0x2e12 (COMPUTE_PGM_RSRC1): 0xf0000{{$}}
-; VI-DAG: 0x2e12 (COMPUTE_PGM_RSRC1): 0xf02c0{{$}}
-; GFX9-DAG: 0x2e12 (COMPUTE_PGM_RSRC1): 0xf0000{{$}}
+; SI-DAG: '0x2e12 (COMPUTE_PGM_RSRC1)': 0xf0000{{$}}
+; VI-DAG: '0x2e12 (COMPUTE_PGM_RSRC1)': 0xf02c0{{$}}
+; GFX9-DAG: '0x2e12 (COMPUTE_PGM_RSRC1)': 0xf0000{{$}}
 define amdgpu_cs half @cs_amdpal(half %arg0) #0 {
   %add = fadd half %arg0, 1.0
   ret half %add
 }
 
 ; amdpal evaluation shader: check for 0x2cca (SPI_SHADER_PGM_RSRC1_ES) in pal metadata
-; SI-DAG: 0x2cca (SPI_SHADER_PGM_RSRC1_ES): 0xf0000{{$}}
-; VI-DAG: 0x2cca (SPI_SHADER_PGM_RSRC1_ES): 0xf02c0{{$}}
-; GFX9-DAG: 0x2cca (SPI_SHADER_PGM_RSRC1_ES): 0xf0000{{$}}
+; SI-DAG: '0x2cca (SPI_SHADER_PGM_RSRC1_ES)': 0xf0000{{$}}
+; VI-DAG: '0x2cca (SPI_SHADER_PGM_RSRC1_ES)': 0xf02c0{{$}}
+; GFX9-DAG: '0x2cca (SPI_SHADER_PGM_RSRC1_ES)': 0xf0000{{$}}
 define amdgpu_es half @es_amdpal(half %arg0) #0 {
   %add = fadd half %arg0, 1.0
   ret half %add
 }
 
 ; amdpal geometry shader: check for 0x2c8a (SPI_SHADER_PGM_RSRC1_GS) in pal metadata
-; SI-DAG: 0x2c8a (SPI_SHADER_PGM_RSRC1_GS): 0xf0000{{$}}
-; VI-DAG: 0x2c8a (SPI_SHADER_PGM_RSRC1_GS): 0xf02c0{{$}}
-; GFX9-DAG: 0x2c8a (SPI_SHADER_PGM_RSRC1_GS): 0xf0000{{$}}
+; SI-DAG: '0x2c8a (SPI_SHADER_PGM_RSRC1_GS)': 0xf0000{{$}}
+; VI-DAG: '0x2c8a (SPI_SHADER_PGM_RSRC1_GS)': 0xf02c0{{$}}
+; GFX9-DAG: '0x2c8a (SPI_SHADER_PGM_RSRC1_GS)': 0xf0000{{$}}
 define amdgpu_gs half @gs_amdpal(half %arg0) #0 {
   %add = fadd half %arg0, 1.0
   ret half %add
 }
 
 ; amdpal hull shader: check for 0x2d0a (SPI_SHADER_PGM_RSRC1_HS) in pal metadata
-; SI-DAG: 0x2d0a (SPI_SHADER_PGM_RSRC1_HS): 0xf0000{{$}}
-; VI-DAG: 0x2d0a (SPI_SHADER_PGM_RSRC1_HS): 0xf02c0{{$}}
-; GFX9-DAG: 0x2d0a (SPI_SHADER_PGM_RSRC1_HS): 0xf0000{{$}}
+; SI-DAG: '0x2d0a (SPI_SHADER_PGM_RSRC1_HS)': 0xf0000{{$}}
+; VI-DAG: '0x2d0a (SPI_SHADER_PGM_RSRC1_HS)': 0xf02c0{{$}}
+; GFX9-DAG: '0x2d0a (SPI_SHADER_PGM_RSRC1_HS)': 0xf0000{{$}}
 define amdgpu_hs half @hs_amdpal(half %arg0) #0 {
   %add = fadd half %arg0, 1.0
   ret half %add
 }
 
 ; amdpal load shader: check for 0x2d4a (SPI_SHADER_PGM_RSRC1_LS) in pal metadata
-; SI-DAG: 0x2d4a (SPI_SHADER_PGM_RSRC1_LS): 0xf0000{{$}}
-; VI-DAG: 0x2d4a (SPI_SHADER_PGM_RSRC1_LS): 0xf02c0{{$}}
-; GFX9-DAG: 0x2d4a (SPI_SHADER_PGM_RSRC1_LS): 0xf0000{{$}}
+; SI-DAG: '0x2d4a (SPI_SHADER_PGM_RSRC1_LS)': 0xf0000{{$}}
+; VI-DAG: '0x2d4a (SPI_SHADER_PGM_RSRC1_LS)': 0xf02c0{{$}}
+; GFX9-DAG: '0x2d4a (SPI_SHADER_PGM_RSRC1_LS)': 0xf0000{{$}}
 define amdgpu_ls half @ls_amdpal(half %arg0) #0 {
   %add = fadd half %arg0, 1.0
   ret half %add
@@ -49,18 +49,18 @@ define amdgpu_ls half @ls_amdpal(half %arg0) #0 {
 
 ; amdpal pixel shader: check for 0x2c0a (SPI_SHADER_PGM_RSRC1_PS) in pal metadata
 ; below.
-; SI-DAG:           0x2c0a (SPI_SHADER_PGM_RSRC1_PS): 0xf0000{{$}}
-; VI-DAG:           0x2c0a (SPI_SHADER_PGM_RSRC1_PS): 0xf02c0{{$}}
-; GFX9-DAG:         0x2c0a (SPI_SHADER_PGM_RSRC1_PS): 0xf0000{{$}}
+; SI-DAG:           '0x2c0a (SPI_SHADER_PGM_RSRC1_PS)': 0xf0000{{$}}
+; VI-DAG:           '0x2c0a (SPI_SHADER_PGM_RSRC1_PS)': 0xf02c0{{$}}
+; GFX9-DAG:         '0x2c0a (SPI_SHADER_PGM_RSRC1_PS)': 0xf0000{{$}}
 define amdgpu_ps half @ps_amdpal(half %arg0) #0 {
   %add = fadd half %arg0, 1.0
   ret half %add
 }
 
 ; amdpal vertex shader: check for 45352 (SPI_SHADER_PGM_RSRC1_VS) in pal metadata
-; SI-DAG: 0x2c4a (SPI_SHADER_PGM_RSRC1_VS): 0xf0000{{$}}
-; VI-DAG: 0x2c4a (SPI_SHADER_PGM_RSRC1_VS): 0xf02c0{{$}}
-; GFX9-DAG: 0x2c4a (SPI_SHADER_PGM_RSRC1_VS): 0xf0000{{$}}
+; SI-DAG: '0x2c4a (SPI_SHADER_PGM_RSRC1_VS)': 0xf0000{{$}}
+; VI-DAG: '0x2c4a (SPI_SHADER_PGM_RSRC1_VS)': 0xf02c0{{$}}
+; GFX9-DAG: '0x2c4a (SPI_SHADER_PGM_RSRC1_VS)': 0xf0000{{$}}
 define amdgpu_vs half @vs_amdpal(half %arg0) #0 {
   %add = fadd half %arg0, 1.0
   ret half %add
@@ -77,7 +77,7 @@ attributes #0 = { "amdgpu-dx10-clamp"="false" }
 ;       - 0x123456789abcdef0
 ;       - 0xfedcba9876543210
 ;     .registers:
-;       0x2c0b (SPI_SHADER_PGM_RSRC2_PS): 0x42000000
+;       '0x2c0b (SPI_SHADER_PGM_RSRC2_PS)': 0x42000000
 ; ...
 ; 	.end_amdgpu_pal_metadata
 
diff --git a/llvm/test/CodeGen/AMDGPU/amdpal-msgpack-es.ll b/llvm/test/CodeGen/AMDGPU/amdpal-msgpack-es.ll
index f97117f3d909..bcc8da6e1bf4 100644
--- a/llvm/test/CodeGen/AMDGPU/amdpal-msgpack-es.ll
+++ b/llvm/test/CodeGen/AMDGPU/amdpal-msgpack-es.ll
@@ -4,7 +4,7 @@
 ; amdpal evaluation shader: check for 0x2cca (SPI_SHADER_PGM_RSRC1_ES) in pal metadata
 ; GCN-LABEL: {{^}}es_amdpal:
 ; GCN: .amdgpu_pal_metadata
-; GCN: 0x2cca (SPI_SHADER_PGM_RSRC1_ES)
+; GCN: '0x2cca (SPI_SHADER_PGM_RSRC1_ES)'
 define amdgpu_es half @es_amdpal(half %arg0) {
   %add = fadd half %arg0, 1.0
   ret half %add
diff --git a/llvm/test/CodeGen/AMDGPU/amdpal-msgpack-gs.ll b/llvm/test/CodeGen/AMDGPU/amdpal-msgpack-gs.ll
index a32d10390b98..ef4c9cbd5006 100644
--- a/llvm/test/CodeGen/AMDGPU/amdpal-msgpack-gs.ll
+++ b/llvm/test/CodeGen/AMDGPU/amdpal-msgpack-gs.ll
@@ -5,7 +5,7 @@
 ; amdpal geometry shader: check for 0x2c8a (SPI_SHADER_PGM_RSRC1_GS) in pal metadata
 ; GCN-LABEL: {{^}}gs_amdpal:
 ; GCN: .amdgpu_pal_metadata
-; GCN: 0x2c8a (SPI_SHADER_PGM_RSRC1_GS)
+; GCN: '0x2c8a (SPI_SHADER_PGM_RSRC1_GS)'
 define amdgpu_gs half @gs_amdpal(half %arg0) {
   %add = fadd half %arg0, 1.0
   ret half %add
diff --git a/llvm/test/CodeGen/AMDGPU/amdpal-msgpack-hs.ll b/llvm/test/CodeGen/AMDGPU/amdpal-msgpack-hs.ll
index be08c93cdb31..eb814c11bceb 100644
--- a/llvm/test/CodeGen/AMDGPU/amdpal-msgpack-hs.ll
+++ b/llvm/test/CodeGen/AMDGPU/amdpal-msgpack-hs.ll
@@ -5,7 +5,7 @@
 ; amdpal hull shader: check for 0x2d0a (SPI_SHADER_PGM_RSRC1_HS) in pal metadata
 ; GCN-LABEL: {{^}}hs_amdpal:
 ; GCN: .amdgpu_pal_metadata
-; GCN: 0x2d0a (SPI_SHADER_PGM_RSRC1_HS)
+; GCN: '0x2d0a (SPI_SHADER_PGM_RSRC1_HS)'
 define amdgpu_hs half @hs_amdpal(half %arg0) {
   %add = fadd half %arg0, 1.0
   ret half %add
diff --git a/llvm/test/CodeGen/AMDGPU/amdpal-msgpack-ieee.ll b/llvm/test/CodeGen/AMDGPU/amdpal-msgpack-ieee.ll
index 95d533544c30..d4826a22db79 100644
--- a/llvm/test/CodeGen/AMDGPU/amdpal-msgpack-ieee.ll
+++ b/llvm/test/CodeGen/AMDGPU/amdpal-msgpack-ieee.ll
@@ -4,50 +4,50 @@
 ; RUN: llc -mtriple=amdgcn--amdpal -mcpu=gfx1200 -verify-machineinstrs < %s | FileCheck -check-prefix=GFX12 -enable-var-scope %s
 
 ; amdpal compute shader: check for 0x2e12 (COMPUTE_PGM_RSRC1) in pal metadata
-; SI-DAG: 0x2e12 (COMPUTE_PGM_RSRC1): 0xaf0000{{$}}
-; VI-DAG: 0x2e12 (COMPUTE_PGM_RSRC1): 0xaf02c0{{$}}
-; GFX9-DAG: 0x2e12 (COMPUTE_PGM_RSRC1): 0xaf0000{{$}}
-; GFX12-DAG: 0x2e12 (COMPUTE_PGM_RSRC1): 0x600f0000{{$}}
+; SI-DAG: '0x2e12 (COMPUTE_PGM_RSRC1)': 0xaf0000{{$}}
+; VI-DAG: '0x2e12 (COMPUTE_PGM_RSRC1)': 0xaf02c0{{$}}
+; GFX9-DAG: '0x2e12 (COMPUTE_PGM_RSRC1)': 0xaf0000{{$}}
+; GFX12-DAG: '0x2e12 (COMPUTE_PGM_RSRC1)': 0x600f0000{{$}}
 define amdgpu_cs half @cs_amdpal(half %arg0) #0 {
   %add = fadd half %arg0, 1.0
   ret half %add
 }
 
 ; amdpal evaluation shader: check for 0x2cca (SPI_SHADER_PGM_RSRC1_ES) in pal metadata
-; SI-DAG: 0x2cca (SPI_SHADER_PGM_RSRC1_ES): 0xaf0000{{$}}
-; VI-DAG: 0x2cca (SPI_SHADER_PGM_RSRC1_ES): 0xaf02c0{{$}}
-; GFX9-DAG: 0x2cca (SPI_SHADER_PGM_RSRC1_ES): 0xaf0000{{$}}
-; GFX12-DAG: 0x2cca (SPI_SHADER_PGM_RSRC1_ES): 0xf0000{{$}}
+; SI-DAG: '0x2cca (SPI_SHADER_PGM_RSRC1_ES)': 0xaf0000{{$}}
+; VI-DAG: '0x2cca (SPI_SHADER_PGM_RSRC1_ES)': 0xaf02c0{{$}}
+; GFX9-DAG: '0x2cca (SPI_SHADER_PGM_RSRC1_ES)': 0xaf0000{{$}}
+; GFX12-DAG: '0x2cca (SPI_SHADER_PGM_RSRC1_ES)': 0xf0000{{$}}
 define amdgpu_es half @es_amdpal(half %arg0) #0 {
   %add = fadd half %arg0, 1.0
   ret half %add
 }
 
 ; amdpal geometry shader: check for 0x2c8a (SPI_SHADER_PGM_RSRC1_GS) in pal metadata
-; SI-DAG: 0x2c8a (SPI_SHADER_PGM_RSRC1_GS): 0xaf0000{{$}}
-; VI-DAG: 0x2c8a (SPI_SHADER_PGM_RSRC1_GS): 0xaf02c0{{$}}
-; GFX9-DAG: 0x2c8a (SPI_SHADER_PGM_RSRC1_GS): 0xaf0000{{$}}
-; GFX12-DAG: 0x2c8a (SPI_SHADER_PGM_RSRC1_GS): 0xa0f0000{{$}}
+; SI-DAG: '0x2c8a (SPI_SHADER_PGM_RSRC1_GS)': 0xaf0000{{$}}
+; VI-DAG: '0x2c8a (SPI_SHADER_PGM_RSRC1_GS)': 0xaf02c0{{$}}
+; GFX9-DAG: '0x2c8a (SPI_SHADER_PGM_RSRC1_GS)': 0xaf0000{{$}}
+; GFX12-DAG: '0x2c8a (SPI_SHADER_PGM_RSRC1_GS)': 0xa0f0000{{$}}
 define amdgpu_gs half @gs_amdpal(half %arg0) #0 {
   %add = fadd half %arg0, 1.0
   ret half %add
 }
 
 ; amdpal hull shader: check for 0x2d0a (SPI_SHADER_PGM_RSRC1_HS) in pal metadata
-; SI-DAG: 0x2d0a (SPI_SHADER_PGM_RSRC1_HS): 0xaf0000{{$}}
-; VI-DAG: 0x2d0a (SPI_SHADER_PGM_RSRC1_HS): 0xaf02c0{{$}}
-; GFX9-DAG: 0x2d0a (SPI_SHADER_PGM_RSRC1_HS): 0xaf0000{{$}}
-; GFX12-DAG: 0x2d0a (SPI_SHADER_PGM_RSRC1_HS): 0x50f0000{{$}}
+; SI-DAG: '0x2d0a (SPI_SHADER_PGM_RSRC1_HS)': 0xaf0000{{$}}
+; VI-DAG: '0x2d0a (SPI_SHADER_PGM_RSRC1_HS)': 0xaf02c0{{$}}
+; GFX9-DAG: '0x2d0a (SPI_SHADER_PGM_RSRC1_HS)': 0xaf0000{{$}}
+; GFX12-DAG: '0x2d0a (SPI_SHADER_PGM_RSRC1_HS)': 0x50f0000{{$}}
 define amdgpu_hs half @hs_amdpal(half %arg0) #0 {
   %add = fadd half %arg0, 1.0
   ret half %add
 }
 
 ; amdpal load shader: check for 0x2d4a (SPI_SHADER_PGM_RSRC1_LS) in pal metadata
-; SI-DAG: 0x2d4a (SPI_SHADER_PGM_RSRC1_LS): 0xaf0000{{$}}
-; VI-DAG: 0x2d4a (SPI_SHADER_PGM_RSRC1_LS): 0xaf02c0{{$}}
-; GFX9-DAG: 0x2d4a (SPI_SHADER_PGM_RSRC1_LS): 0xaf0000{{$}}
-; GFX12-DAG: 0x2d4a (SPI_SHADER_PGM_RSRC1_LS): 0xf0000{{$}}
+; SI-DAG: '0x2d4a (SPI_SHADER_PGM_RSRC1_LS)': 0xaf0000{{$}}
+; VI-DAG: '0x2d4a (SPI_SHADER_PGM_RSRC1_LS)': 0xaf02c0{{$}}
+; GFX9-DAG: '0x2d4a (SPI_SHADER_PGM_RSRC1_LS)': 0xaf0000{{$}}
+; GFX12-DAG: '0x2d4a (SPI_SHADER_PGM_RSRC1_LS)': 0xf0000{{$}}
 define amdgpu_ls half @ls_amdpal(half %arg0) #0 {
   %add = fadd half %arg0, 1.0
   ret half %add
@@ -55,20 +55,20 @@ define amdgpu_ls half @ls_amdpal(half %arg0) #0 {
 
 ; amdpal pixel shader: check for 0x2c0a (SPI_SHADER_PGM_RSRC1_PS) in pal metadata
 ; below.
-; SI-DAG:           0x2c0a (SPI_SHADER_PGM_RSRC1_PS): 0xaf0000{{$}}
-; VI-DAG:           0x2c0a (SPI_SHADER_PGM_RSRC1_PS): 0xaf02c0{{$}}
-; GFX9-DAG:         0x2c0a (SPI_SHADER_PGM_RSRC1_PS): 0xaf0000{{$}}
-; GFX12-DAG:        0x2c0a (SPI_SHADER_PGM_RSRC1_PS): 0x20f0000{{$}}
+; SI-DAG:           '0x2c0a (SPI_SHADER_PGM_RSRC1_PS)': 0xaf0000{{$}}
+; VI-DAG:           '0x2c0a (SPI_SHADER_PGM_RSRC1_PS)': 0xaf02c0{{$}}
+; GFX9-DAG:         '0x2c0a (SPI_SHADER_PGM_RSRC1_PS)': 0xaf0000{{$}}
+; GFX12-DAG:        '0x2c0a (SPI_SHADER_PGM_RSRC1_PS)': 0x20f0000{{$}}
 define amdgpu_ps half @ps_amdpal(half %arg0) #0 {
   %add = fadd half %arg0, 1.0
   ret half %add
 }
 
 ; amdpal vertex shader: check for 45352 (SPI_SHADER_PGM_RSRC1_VS) in pal metadata
-; SI-DAG: 0x2c4a (SPI_SHADER_PGM_RSRC1_VS): 0xaf0000{{$}}
-; VI-DAG: 0x2c4a (SPI_SHADER_PGM_RSRC1_VS): 0xaf02c0{{$}}
-; GFX9-DAG: 0x2c4a (SPI_SHADER_PGM_RSRC1_VS): 0xaf0000{{$}}
-; GFX12-DAG: 0x2c4a (SPI_SHADER_PGM_RSRC1_VS): 0x80f0000{{$}}
+; SI-DAG: '0x2c4a (SPI_SHADER_PGM_RSRC1_VS)': 0xaf0000{{$}}
+; VI-DAG: '0x2c4a (SPI_SHADER_PGM_RSRC1_VS)': 0xaf02c0{{$}}
+; GFX9-DAG: '0x2c4a (SPI_SHADER_PGM_RSRC1_VS)': 0xaf0000{{$}}
+; GFX12-DAG: '0x2c4a (SPI_SHADER_PGM_RSRC1_VS)': 0x80f0000{{$}}
 define amdgpu_vs half @vs_amdpal(half %arg0) #0 {
   %add = fadd half %arg0, 1.0
   ret half %add
@@ -85,7 +85,7 @@ attributes #0 = { "amdgpu-ieee"="true" }
 ;       - 0x123456789abcdef0
 ;       - 0xfedcba9876543210
 ;     .registers:
-;       0x2c0b (SPI_SHADER_PGM_RSRC2_PS): 0x42000000
+;       '0x2c0b (SPI_SHADER_PGM_RSRC2_PS)': 0x42000000
 ; ...
 ; 	.end_amdgpu_pal_metadata
 
diff --git a/llvm/test/CodeGen/AMDGPU/amdpal-msgpack-ls.ll b/llvm/test/CodeGen/AMDGPU/amdpal-msgpack-ls.ll
index 46097fa20608..0d81e70b2e4f 100644
--- a/llvm/test/CodeGen/AMDGPU/amdpal-msgpack-ls.ll
+++ b/llvm/test/CodeGen/AMDGPU/amdpal-msgpack-ls.ll
@@ -4,7 +4,7 @@
 ; amdpal load shader: check for 0x2d4a (SPI_SHADER_PGM_RSRC1_LS) in pal metadata
 ; GCN-LABEL: {{^}}ls_amdpal:
 ; GCN: .amdgpu_pal_metadata
-; GCN: 0x2d4a (SPI_SHADER_PGM_RSRC1_LS)
+; GCN: '0x2d4a (SPI_SHADER_PGM_RSRC1_LS)'
 define amdgpu_ls half @ls_amdpal(half %arg0) {
   %add = fadd half %arg0, 1.0
   ret half %add
diff --git a/llvm/test/CodeGen/AMDGPU/amdpal-msgpack-ps.ll b/llvm/test/CodeGen/AMDGPU/amdpal-msgpack-ps.ll
index 9169c651f129..d31732f995b1 100644
--- a/llvm/test/CodeGen/AMDGPU/amdpal-msgpack-ps.ll
+++ b/llvm/test/CodeGen/AMDGPU/amdpal-msgpack-ps.ll
@@ -12,8 +12,8 @@
 ; GCN-NEXT:      - 0x123456789abcdef0
 ; GCN-NEXT:      - 0xfedcba9876543210
 ; GCN:         .registers:
-; GCN:           0x2c0a (SPI_SHADER_PGM_RSRC1_PS):
-; GCN:           0x2c0b (SPI_SHADER_PGM_RSRC2_PS): 0x42
+; GCN:           '0x2c0a (SPI_SHADER_PGM_RSRC1_PS)':
+; GCN:           '0x2c0b (SPI_SHADER_PGM_RSRC2_PS)': 0x42
 define amdgpu_ps half @ps_amdpal(half %arg0) {
   %add = fadd half %arg0, 1.0
   ret half %add
@@ -23,12 +23,12 @@ define amdgpu_ps half @ps_amdpal(half %arg0) {
 ;
 ; 	.amdgpu_pal_metadata
 ; ---
-; amdpal.pipelines: 
+; amdpal.pipelines:
 ;   - .internal_pipeline_hash:
 ;       - 0x123456789abcdef0
 ;       - 0xfedcba9876543210
 ;     .registers:
-;       0x2c0b (SPI_SHADER_PGM_RSRC2_PS): 0x42000000
+;       '0x2c0b (SPI_SHADER_PGM_RSRC2_PS)': 0x42000000
 ; ...
 ; 	.end_amdgpu_pal_metadata
 
diff --git a/llvm/test/CodeGen/AMDGPU/amdpal-msgpack-psenable.ll b/llvm/test/CodeGen/AMDGPU/amdpal-msgpack-psenable.ll
index d6322e2b4d3e..15b1a652077e 100644
--- a/llvm/test/CodeGen/AMDGPU/amdpal-msgpack-psenable.ll
+++ b/llvm/test/CodeGen/AMDGPU/amdpal-msgpack-psenable.ll
@@ -7,8 +7,8 @@
 ; the workaround that ensures that an interpolation mode is also set in PSEnable.
 ; GCN-LABEL: {{^}}amdpal_psenable:
 ; GCN: .amdgpu_pal_metadata
-; GCN: 0xa1b3 (SPI_PS_INPUT_ENA): 0x2
-; GCN: 0xa1b4 (SPI_PS_INPUT_ADDR): 0x2
+; GCN: '0xa1b3 (SPI_PS_INPUT_ENA)': 0x2
+; GCN: '0xa1b4 (SPI_PS_INPUT_ADDR)': 0x2
 define amdgpu_ps void @amdpal_psenable(i32 inreg, i32 inreg, i32 inreg, i32 inreg %m0, <2 x float> %pos) #6 {
   %inst23 = extractelement <2 x float> %pos, i32 0
   %inst24 = extractelement <2 x float> %pos, i32 1
diff --git a/llvm/test/CodeGen/AMDGPU/amdpal-msgpack-vs.ll b/llvm/test/CodeGen/AMDGPU/amdpal-msgpack-vs.ll
index 7c47129c28ce..42de6007f7e2 100644
--- a/llvm/test/CodeGen/AMDGPU/amdpal-msgpack-vs.ll
+++ b/llvm/test/CodeGen/AMDGPU/amdpal-msgpack-vs.ll
@@ -5,7 +5,7 @@
 ; amdpal vertex shader: check for 45352 (SPI_SHADER_PGM_RSRC1_VS) in pal metadata
 ; GCN-LABEL: {{^}}vs_amdpal:
 ; GCN: .amdgpu_pal_metadata
-; GCN: 0x2c4a (SPI_SHADER_PGM_RSRC1_VS)
+; GCN: '0x2c4a (SPI_SHADER_PGM_RSRC1_VS)'
 define amdgpu_vs half @vs_amdpal(half %arg0) {
   %add = fadd half %arg0, 1.0
   ret half %add
diff --git a/llvm/test/CodeGen/AMDGPU/amdpal-psenable.ll b/llvm/test/CodeGen/AMDGPU/amdpal-psenable.ll
index 13d2050c491f..ace21207a7eb 100644
--- a/llvm/test/CodeGen/AMDGPU/amdpal-psenable.ll
+++ b/llvm/test/CodeGen/AMDGPU/amdpal-psenable.ll
@@ -14,10 +14,10 @@
 ; GCN-NEXT:         .entry_point:    amdpal_psenable
 ; GCN-NEXT:         .scratch_memory_size: 0
 ; GCN:     .registers:
-; GCN-NEXT:       0x2c0a (SPI_SHADER_PGM_RSRC1_PS):
-; GCN-NEXT:       0x2c0b (SPI_SHADER_PGM_RSRC2_PS):
-; GCN-NEXT:       0xa1b3 (SPI_PS_INPUT_ENA): 0x2
-; GCN-NEXT:       0xa1b4 (SPI_PS_INPUT_ADDR): 0x2
+; GCN-NEXT:       '0x2c0a (SPI_SHADER_PGM_RSRC1_PS)':
+; GCN-NEXT:       '0x2c0b (SPI_SHADER_PGM_RSRC2_PS)':
+; GCN-NEXT:       '0xa1b3 (SPI_PS_INPUT_ENA)': 0x2
+; GCN-NEXT:       '0xa1b4 (SPI_PS_INPUT_ADDR)': 0x2
 ; GCN-NEXT: ...
 ; GCN-NEXT:         .end_amdgpu_pal_metadata
 define amdgpu_ps void @amdpal_psenable(i32 inreg, i32 inreg, i32 inreg, i32 inreg %m0, <2 x float> %pos) #6 {
diff --git a/llvm/test/CodeGen/AMDGPU/amdpal-usersgpr-init.ll b/llvm/test/CodeGen/AMDGPU/amdpal-usersgpr-init.ll
index 52a9d57244c2..086a126b1ddc 100644
--- a/llvm/test/CodeGen/AMDGPU/amdpal-usersgpr-init.ll
+++ b/llvm/test/CodeGen/AMDGPU/amdpal-usersgpr-init.ll
@@ -1,7 +1,7 @@
 ; RUN: llc -mtriple=amdgcn--amdpal -mcpu=gfx1100 -verify-machineinstrs < %s | FileCheck -check-prefix=GCN -enable-var-scope %s
 
 ; We want to make sure that RSRC2 is left untouched
-; GCN:       0x2e13 (COMPUTE_PGM_RSRC2): 0x78a
+; GCN:       '0x2e13 (COMPUTE_PGM_RSRC2)': 0x78a
 define amdgpu_cs half @cs_amdpal(half %arg0, half inreg %arg1) {
   %add = fadd half %arg0, 1.0
   ret half %add
@@ -9,4 +9,4 @@ define amdgpu_cs half @cs_amdpal(half %arg0, half inreg %arg1) {
 
 !amdgpu.pal.metadata.msgpack = !{!0}
 
-!0 = !{!"\82\B0amdpal.pipelines\91\89\A4.api\A6Vulkan\B0.hardware_stages\81\A3.cs\83\AB.sgpr_limith\AB.vgpr_limit\CD\01\00\AF.wavefront_size@\B7.internal_pipeline_hash\92\CF\E83\B3\C2\D1)\7FG\CF[\8A\DF\EE[\7FD,\AA.registers\8A\CD.\07\01\CD.\08\01\CD.\09\01\CD.\12\CE@,\00\00\CD.\13\CD\07\8A\CD.(\00\CD.*\CE\16\0B\22Y\CD.@\CE\10\00\00\00\CD.B\CE\10\00\00\06\CD.D\00\A8.shaders\81\A8.compute\82\B0.api_shader_hash\92\CF\D3s\A6\8D\C5x\84\D4\00\B1.hardware_mapping\91\A3.cs\B0.spill_threshold\CE\FF\FF\FF\FF\A5.type\A2Cs\B0.user_data_limit\01\AF.xgl_cache_info\82\B3.128_bit_cache_hash\92\CF\E5\A0\EB\F9}\C6\C1\13\CF\1A_\E7\F7\F2.mR\AD.llpc_version\A454.5\AEamdpal.version\92\02\03"}
\ No newline at end of file
+!0 = !{!"\82\B0amdpal.pipelines\91\89\A4.api\A6Vulkan\B0.hardware_stages\81\A3.cs\83\AB.sgpr_limith\AB.vgpr_limit\CD\01\00\AF.wavefront_size@\B7.internal_pipeline_hash\92\CF\E83\B3\C2\D1)\7FG\CF[\8A\DF\EE[\7FD,\AA.registers\8A\CD.\07\01\CD.\08\01\CD.\09\01\CD.\12\CE@,\00\00\CD.\13\CD\07\8A\CD.(\00\CD.*\CE\16\0B\22Y\CD.@\CE\10\00\00\00\CD.B\CE\10\00\00\06\CD.D\00\A8.shaders\81\A8.compute\82\B0.api_shader_hash\92\CF\D3s\A6\8D\C5x\84\D4\00\B1.hardware_mapping\91\A3.cs\B0.spill_threshold\CE\FF\FF\FF\FF\A5.type\A2Cs\B0.user_data_limit\01\AF.xgl_cache_info\82\B3.128_bit_cache_hash\92\CF\E5\A0\EB\F9}\C6\C1\13\CF\1A_\E7\F7\F2.mR\AD.llpc_version\A454.5\AEamdpal.version\92\02\03"}
diff --git a/llvm/test/CodeGen/AMDGPU/amdpal-vs.ll b/llvm/test/CodeGen/AMDGPU/amdpal-vs.ll
index ec8f698d69c2..c300ba187740 100644
--- a/llvm/test/CodeGen/AMDGPU/amdpal-vs.ll
+++ b/llvm/test/CodeGen/AMDGPU/amdpal-vs.ll
@@ -11,7 +11,7 @@
 ; GCN-NEXT:         .entry_point:    vs_amdpal
 ; GCN-NEXT:         .scratch_memory_size: 0
 ; GCN:     .registers:
-; GCN-NEXT:       0x2c4a (SPI_SHADER_PGM_RSRC1_VS): 0
+; GCN-NEXT:       '0x2c4a (SPI_SHADER_PGM_RSRC1_VS)': 0
 ; GCN-NEXT: ...
 ; GCN-NEXT:         .end_amdgpu_pal_metadata
 define amdgpu_vs half @vs_amdpal(half %arg0) {
diff --git a/llvm/test/CodeGen/AMDGPU/atomic_load_local.ll b/llvm/test/CodeGen/AMDGPU/atomic_load_local.ll
index bfd18f1b52a5..a3b6c283512f 100644
--- a/llvm/test/CodeGen/AMDGPU/atomic_load_local.ll
+++ b/llvm/test/CodeGen/AMDGPU/atomic_load_local.ll
@@ -152,3 +152,57 @@ define ptr addrspace(3) @atomic_load_monotonic_p3i8_offset(ptr addrspace(3) %ptr
   %load = load atomic ptr addrspace(3), ptr addrspace(3) %gep monotonic, align 4
   ret ptr addrspace(3) %load
 }
+
+; GCN-LABEL: {{^}}atomic_load_monotonic_f16:
+; GCN: s_waitcnt
+; GFX9-NOT: s_mov_b32 m0
+; CI-NEXT: s_mov_b32 m0
+; GCN-NEXT: ds_read_u16 v0, v0{{$}}
+; GCN-NEXT: s_waitcnt lgkmcnt(0)
+; GCN-NEXT: s_setpc_b64
+define i16 @atomic_load_monotonic_f16(ptr addrspace(3) %ptr) {
+  %load = load atomic half, ptr addrspace(3) %ptr monotonic, align 2
+  %ret = bitcast half %load to i16
+  ret i16 %ret
+}
+
+; GCN-LABEL: {{^}}atomic_load_monotonic_f16_offset:
+; GCN: s_waitcnt
+; GFX9-NOT: s_mov_b32 m0
+; CI-NEXT: s_mov_b32 m0
+; GCN-NEXT: ds_read_u16 v0, v0 offset:32{{$}}
+; GCN-NEXT: s_waitcnt lgkmcnt(0)
+; GCN-NEXT: s_setpc_b64
+define i16 @atomic_load_monotonic_f16_offset(ptr addrspace(3) %ptr) {
+  %gep = getelementptr inbounds half, ptr addrspace(3) %ptr, i32 16
+  %load = load atomic half, ptr addrspace(3) %gep monotonic, align 2
+  %ret = bitcast half %load to i16
+  ret i16 %ret
+}
+
+; GCN-LABEL: {{^}}atomic_load_monotonic_bf16:
+; GCN: s_waitcnt
+; GFX9-NOT: s_mov_b32 m0
+; CI-NEXT: s_mov_b32 m0
+; GCN-NEXT: ds_read_u16 v0, v0{{$}}
+; GCN-NEXT: s_waitcnt lgkmcnt(0)
+; GCN-NEXT: s_setpc_b64
+define i16 @atomic_load_monotonic_bf16(ptr addrspace(3) %ptr) {
+  %load = load atomic bfloat, ptr addrspace(3) %ptr monotonic, align 2
+  %ret = bitcast bfloat %load to i16
+  ret i16 %ret
+}
+
+; GCN-LABEL: {{^}}atomic_load_monotonic_bf16_offset:
+; GCN: s_waitcnt
+; GFX9-NOT: s_mov_b32 m0
+; CI-NEXT: s_mov_b32 m0
+; GCN-NEXT: ds_read_u16 v0, v0 offset:32{{$}}
+; GCN-NEXT: s_waitcnt lgkmcnt(0)
+; GCN-NEXT: s_setpc_b64
+define i16 @atomic_load_monotonic_bf16_offset(ptr addrspace(3) %ptr) {
+  %gep = getelementptr inbounds bfloat, ptr addrspace(3) %ptr, i32 16
+  %load = load atomic bfloat, ptr addrspace(3) %gep monotonic, align 2
+  %ret = bitcast bfloat %load to i16
+  ret i16 %ret
+}
diff --git a/llvm/test/CodeGen/AMDGPU/atomic_store_local.ll b/llvm/test/CodeGen/AMDGPU/atomic_store_local.ll
index 71e24c1692c7..cd1e1fb1add4 100644
--- a/llvm/test/CodeGen/AMDGPU/atomic_store_local.ll
+++ b/llvm/test/CodeGen/AMDGPU/atomic_store_local.ll
@@ -101,3 +101,56 @@ define void @atomic_store_monotonic_offset_i64(ptr addrspace(3) %ptr, i64 %val)
   ret void
 }
 
+; GCN-LABEL: {{^}}atomic_store_monotonic_f16:
+; GCN: s_waitcnt
+; GFX9-NOT: s_mov_b32 m0
+; CI-NEXT: s_mov_b32 m0
+; GCN-NEXT: ds_write_b16 v0, v1{{$}}
+; GCN-NEXT: s_waitcnt lgkmcnt(0)
+; GCN-NEXT: s_setpc_b64
+define void @atomic_store_monotonic_f16(ptr addrspace(3) %ptr, i16 %arg.val) {
+  %val = bitcast i16 %arg.val to half
+  store atomic half %val, ptr addrspace(3) %ptr monotonic, align 2
+  ret void
+}
+
+; GCN-LABEL: {{^}}atomic_store_monotonic_offset_f16:
+; GCN: s_waitcnt
+; GFX9-NOT: s_mov_b32 m0
+; CI-NEXT: s_mov_b32 m0
+; GCN-NEXT: ds_write_b16 v0, v1 offset:32{{$}}
+; GCN-NEXT: s_waitcnt lgkmcnt(0)
+; GCN-NEXT: s_setpc_b64
+define void @atomic_store_monotonic_offset_f16(ptr addrspace(3) %ptr, i16 %arg.val) {
+  %val = bitcast i16 %arg.val to half
+  %gep = getelementptr inbounds half, ptr addrspace(3) %ptr, i32 16
+  store atomic half %val, ptr addrspace(3) %gep monotonic, align 2
+  ret void
+}
+
+; GCN-LABEL: {{^}}atomic_store_monotonic_bf16:
+; GCN: s_waitcnt
+; GFX9-NOT: s_mov_b32 m0
+; CI-NEXT: s_mov_b32 m0
+; GCN-NEXT: ds_write_b16 v0, v1{{$}}
+; GCN-NEXT: s_waitcnt lgkmcnt(0)
+; GCN-NEXT: s_setpc_b64
+define void @atomic_store_monotonic_bf16(ptr addrspace(3) %ptr, i16 %arg.val) {
+  %val = bitcast i16 %arg.val to bfloat
+  store atomic bfloat %val, ptr addrspace(3) %ptr monotonic, align 2
+  ret void
+}
+
+; GCN-LABEL: {{^}}atomic_store_monotonic_offset_bf16:
+; GCN: s_waitcnt
+; GFX9-NOT: s_mov_b32 m0
+; CI-NEXT: s_mov_b32 m0
+; GCN-NEXT: ds_write_b16 v0, v1 offset:32{{$}}
+; GCN-NEXT: s_waitcnt lgkmcnt(0)
+; GCN-NEXT: s_setpc_b64
+define void @atomic_store_monotonic_offset_bf16(ptr addrspace(3) %ptr, i16 %arg.val) {
+  %val = bitcast i16 %arg.val to bfloat
+  %gep = getelementptr inbounds bfloat, ptr addrspace(3) %ptr, i32 16
+  store atomic bfloat %val, ptr addrspace(3) %gep monotonic, align 2
+  ret void
+}
diff --git a/llvm/test/CodeGen/AMDGPU/div_i128.ll b/llvm/test/CodeGen/AMDGPU/div_i128.ll
index cf99b5d80e13..b2f9bf89d9ec 100644
--- a/llvm/test/CodeGen/AMDGPU/div_i128.ll
+++ b/llvm/test/CodeGen/AMDGPU/div_i128.ll
@@ -282,21 +282,21 @@ define i128 @v_sdiv_i128_vv(i128 %lhs, i128 %rhs) {
 ; GFX9-O0-NEXT:    ; kill: def $vgpr15 killed $vgpr15 def $vgpr15_vgpr16 killed $exec
 ; GFX9-O0-NEXT:    v_mov_b32_e32 v16, v1
 ; GFX9-O0-NEXT:    v_mov_b32_e32 v9, v15
-; GFX9-O0-NEXT:    v_mov_b32_e32 v1, v16
+; GFX9-O0-NEXT:    v_mov_b32_e32 v5, v16
 ; GFX9-O0-NEXT:    v_mov_b32_e32 v10, v13
-; GFX9-O0-NEXT:    v_mov_b32_e32 v5, v14
+; GFX9-O0-NEXT:    v_mov_b32_e32 v1, v14
 ; GFX9-O0-NEXT:    v_sub_co_u32_e32 v9, vcc, v9, v4
-; GFX9-O0-NEXT:    v_subb_co_u32_e32 v1, vcc, v1, v6, vcc
-; GFX9-O0-NEXT:    v_subb_co_u32_e32 v13, vcc, v10, v4, vcc
 ; GFX9-O0-NEXT:    v_subb_co_u32_e32 v5, vcc, v5, v6, vcc
+; GFX9-O0-NEXT:    v_subb_co_u32_e32 v13, vcc, v10, v4, vcc
+; GFX9-O0-NEXT:    v_subb_co_u32_e32 v1, vcc, v1, v6, vcc
 ; GFX9-O0-NEXT:    ; implicit-def: $sgpr4
 ; GFX9-O0-NEXT:    ; implicit-def: $sgpr4
-; GFX9-O0-NEXT:    ; kill: def $vgpr13 killed $vgpr13 def $vgpr13_vgpr14 killed $exec
-; GFX9-O0-NEXT:    v_mov_b32_e32 v14, v5
+; GFX9-O0-NEXT:    ; kill: def $vgpr9 killed $vgpr9 def $vgpr9_vgpr10 killed $exec
+; GFX9-O0-NEXT:    v_mov_b32_e32 v10, v5
 ; GFX9-O0-NEXT:    ; implicit-def: $sgpr4
 ; GFX9-O0-NEXT:    ; implicit-def: $sgpr4
-; GFX9-O0-NEXT:    ; kill: def $vgpr9 killed $vgpr9 def $vgpr9_vgpr10 killed $exec
-; GFX9-O0-NEXT:    v_mov_b32_e32 v10, v1
+; GFX9-O0-NEXT:    ; kill: def $vgpr13 killed $vgpr13 def $vgpr13_vgpr14 killed $exec
+; GFX9-O0-NEXT:    v_mov_b32_e32 v14, v1
 ; GFX9-O0-NEXT:    v_mov_b32_e32 v1, v3
 ; GFX9-O0-NEXT:    v_mov_b32_e32 v5, v12
 ; GFX9-O0-NEXT:    v_xor_b32_e64 v1, v5, v1
@@ -312,21 +312,21 @@ define i128 @v_sdiv_i128_vv(i128 %lhs, i128 %rhs) {
 ; GFX9-O0-NEXT:    ; kill: def $vgpr7 killed $vgpr7 def $vgpr7_vgpr8 killed $exec
 ; GFX9-O0-NEXT:    v_mov_b32_e32 v8, v1
 ; GFX9-O0-NEXT:    v_mov_b32_e32 v1, v7
-; GFX9-O0-NEXT:    v_mov_b32_e32 v7, v8
-; GFX9-O0-NEXT:    v_mov_b32_e32 v8, v11
+; GFX9-O0-NEXT:    ; kill: def $vgpr8 killed $vgpr8 killed $vgpr7_vgpr8 killed $exec
+; GFX9-O0-NEXT:    v_mov_b32_e32 v7, v11
 ; GFX9-O0-NEXT:    v_mov_b32_e32 v2, v12
 ; GFX9-O0-NEXT:    v_sub_co_u32_e32 v1, vcc, v1, v3
-; GFX9-O0-NEXT:    v_subb_co_u32_e32 v7, vcc, v7, v5, vcc
-; GFX9-O0-NEXT:    v_subb_co_u32_e32 v11, vcc, v8, v3, vcc
-; GFX9-O0-NEXT:    v_subb_co_u32_e32 v2, vcc, v2, v5, vcc
+; GFX9-O0-NEXT:    v_subb_co_u32_e32 v8, vcc, v8, v5, vcc
+; GFX9-O0-NEXT:    v_subb_co_u32_e32 v11, vcc, v7, v3, vcc
+; GFX9-O0-NEXT:    v_subb_co_u32_e32 v7, vcc, v2, v5, vcc
 ; GFX9-O0-NEXT:    ; implicit-def: $sgpr4
 ; GFX9-O0-NEXT:    ; implicit-def: $sgpr4
-; GFX9-O0-NEXT:    ; kill: def $vgpr11 killed $vgpr11 def $vgpr11_vgpr12 killed $exec
-; GFX9-O0-NEXT:    v_mov_b32_e32 v12, v2
+; GFX9-O0-NEXT:    ; kill: def $vgpr1 killed $vgpr1 def $vgpr1_vgpr2 killed $exec
+; GFX9-O0-NEXT:    v_mov_b32_e32 v2, v8
 ; GFX9-O0-NEXT:    ; implicit-def: $sgpr4
 ; GFX9-O0-NEXT:    ; implicit-def: $sgpr4
-; GFX9-O0-NEXT:    ; kill: def $vgpr1 killed $vgpr1 def $vgpr1_vgpr2 killed $exec
-; GFX9-O0-NEXT:    v_mov_b32_e32 v2, v7
+; GFX9-O0-NEXT:    ; kill: def $vgpr11 killed $vgpr11 def $vgpr11_vgpr12 killed $exec
+; GFX9-O0-NEXT:    v_mov_b32_e32 v12, v7
 ; GFX9-O0-NEXT:    v_xor_b32_e64 v5, v5, v6
 ; GFX9-O0-NEXT:    v_xor_b32_e64 v3, v3, v4
 ; GFX9-O0-NEXT:    ; kill: def $vgpr3 killed $vgpr3 def $vgpr3_vgpr4 killed $exec
@@ -339,18 +339,26 @@ define i128 @v_sdiv_i128_vv(i128 %lhs, i128 %rhs) {
 ; GFX9-O0-NEXT:    buffer_store_dword v3, off, s[0:3], s32 offset:68 ; 4-byte Folded Spill
 ; GFX9-O0-NEXT:    s_waitcnt vmcnt(0)
 ; GFX9-O0-NEXT:    buffer_store_dword v4, off, s[0:3], s32 offset:72 ; 4-byte Folded Spill
-; GFX9-O0-NEXT:    buffer_store_dword v1, off, s[0:3], s32 offset:60 ; 4-byte Folded Spill
+; GFX9-O0-NEXT:    v_mov_b32_e32 v3, v11
+; GFX9-O0-NEXT:    v_mov_b32_e32 v4, v12
+; GFX9-O0-NEXT:    buffer_store_dword v3, off, s[0:3], s32 offset:60 ; 4-byte Folded Spill
 ; GFX9-O0-NEXT:    s_waitcnt vmcnt(0)
-; GFX9-O0-NEXT:    buffer_store_dword v2, off, s[0:3], s32 offset:64 ; 4-byte Folded Spill
-; GFX9-O0-NEXT:    buffer_store_dword v11, off, s[0:3], s32 offset:52 ; 4-byte Folded Spill
+; GFX9-O0-NEXT:    buffer_store_dword v4, off, s[0:3], s32 offset:64 ; 4-byte Folded Spill
+; GFX9-O0-NEXT:    v_mov_b32_e32 v4, v2
+; GFX9-O0-NEXT:    v_mov_b32_e32 v3, v1
+; GFX9-O0-NEXT:    buffer_store_dword v3, off, s[0:3], s32 offset:52 ; 4-byte Folded Spill
 ; GFX9-O0-NEXT:    s_waitcnt vmcnt(0)
-; GFX9-O0-NEXT:    buffer_store_dword v12, off, s[0:3], s32 offset:56 ; 4-byte Folded Spill
-; GFX9-O0-NEXT:    buffer_store_dword v9, off, s[0:3], s32 offset:44 ; 4-byte Folded Spill
+; GFX9-O0-NEXT:    buffer_store_dword v4, off, s[0:3], s32 offset:56 ; 4-byte Folded Spill
+; GFX9-O0-NEXT:    v_mov_b32_e32 v3, v13
+; GFX9-O0-NEXT:    v_mov_b32_e32 v4, v14
+; GFX9-O0-NEXT:    buffer_store_dword v3, off, s[0:3], s32 offset:44 ; 4-byte Folded Spill
 ; GFX9-O0-NEXT:    s_waitcnt vmcnt(0)
-; GFX9-O0-NEXT:    buffer_store_dword v10, off, s[0:3], s32 offset:48 ; 4-byte Folded Spill
-; GFX9-O0-NEXT:    buffer_store_dword v13, off, s[0:3], s32 offset:36 ; 4-byte Folded Spill
+; GFX9-O0-NEXT:    buffer_store_dword v4, off, s[0:3], s32 offset:48 ; 4-byte Folded Spill
+; GFX9-O0-NEXT:    v_mov_b32_e32 v3, v9
+; GFX9-O0-NEXT:    v_mov_b32_e32 v4, v10
+; GFX9-O0-NEXT:    buffer_store_dword v3, off, s[0:3], s32 offset:36 ; 4-byte Folded Spill
 ; GFX9-O0-NEXT:    s_waitcnt vmcnt(0)
-; GFX9-O0-NEXT:    buffer_store_dword v14, off, s[0:3], s32 offset:40 ; 4-byte Folded Spill
+; GFX9-O0-NEXT:    buffer_store_dword v4, off, s[0:3], s32 offset:40 ; 4-byte Folded Spill
 ; GFX9-O0-NEXT:    v_mov_b32_e32 v7, v12
 ; GFX9-O0-NEXT:    v_mov_b32_e32 v8, v2
 ; GFX9-O0-NEXT:    v_or_b32_e64 v3, v8, v7
@@ -403,7 +411,8 @@ define i128 @v_sdiv_i128_vv(i128 %lhs, i128 %rhs) {
 ; GFX9-O0-NEXT:    ; kill: def $vgpr8 killed $vgpr8 def $vgpr8_vgpr9 killed $exec
 ; GFX9-O0-NEXT:    v_mov_b32_e32 v9, v5
 ; GFX9-O0-NEXT:    v_mov_b32_e32 v5, v9
-; GFX9-O0-NEXT:    v_cmp_ne_u64_e64 s[12:13], v[11:12], s[6:7]
+; GFX9-O0-NEXT:    s_mov_b64 s[12:13], s[6:7]
+; GFX9-O0-NEXT:    v_cmp_ne_u64_e64 s[12:13], v[11:12], s[12:13]
 ; GFX9-O0-NEXT:    v_cndmask_b32_e64 v5, v5, v10, s[12:13]
 ; GFX9-O0-NEXT:    v_mov_b32_e32 v7, v6
 ; GFX9-O0-NEXT:    v_mov_b32_e32 v6, v8
@@ -439,7 +448,8 @@ define i128 @v_sdiv_i128_vv(i128 %lhs, i128 %rhs) {
 ; GFX9-O0-NEXT:    ; kill: def $vgpr11 killed $vgpr11 def $vgpr11_vgpr12 killed $exec
 ; GFX9-O0-NEXT:    v_mov_b32_e32 v12, v5
 ; GFX9-O0-NEXT:    v_mov_b32_e32 v5, v12
-; GFX9-O0-NEXT:    v_cmp_ne_u64_e64 s[8:9], v[13:14], s[6:7]
+; GFX9-O0-NEXT:    s_mov_b64 s[8:9], s[6:7]
+; GFX9-O0-NEXT:    v_cmp_ne_u64_e64 s[8:9], v[13:14], s[8:9]
 ; GFX9-O0-NEXT:    v_cndmask_b32_e64 v5, v5, v8, s[8:9]
 ; GFX9-O0-NEXT:    v_mov_b32_e32 v7, v6
 ; GFX9-O0-NEXT:    v_mov_b32_e32 v6, v11
@@ -690,10 +700,10 @@ define i128 @v_sdiv_i128_vv(i128 %lhs, i128 %rhs) {
 ; GFX9-O0-NEXT:    buffer_load_dword v28, off, s[0:3], s32 offset:260 ; 4-byte Folded Reload
 ; GFX9-O0-NEXT:    buffer_load_dword v25, off, s[0:3], s32 offset:264 ; 4-byte Folded Reload
 ; GFX9-O0-NEXT:    buffer_load_dword v26, off, s[0:3], s32 offset:268 ; 4-byte Folded Reload
-; GFX9-O0-NEXT:    buffer_load_dword v19, off, s[0:3], s32 offset:60 ; 4-byte Folded Reload
-; GFX9-O0-NEXT:    buffer_load_dword v20, off, s[0:3], s32 offset:64 ; 4-byte Folded Reload
-; GFX9-O0-NEXT:    buffer_load_dword v21, off, s[0:3], s32 offset:52 ; 4-byte Folded Reload
-; GFX9-O0-NEXT:    buffer_load_dword v22, off, s[0:3], s32 offset:56 ; 4-byte Folded Reload
+; GFX9-O0-NEXT:    buffer_load_dword v19, off, s[0:3], s32 offset:52 ; 4-byte Folded Reload
+; GFX9-O0-NEXT:    buffer_load_dword v20, off, s[0:3], s32 offset:56 ; 4-byte Folded Reload
+; GFX9-O0-NEXT:    buffer_load_dword v21, off, s[0:3], s32 offset:60 ; 4-byte Folded Reload
+; GFX9-O0-NEXT:    buffer_load_dword v22, off, s[0:3], s32 offset:64 ; 4-byte Folded Reload
 ; GFX9-O0-NEXT:    buffer_load_dword v14, off, s[0:3], s32 offset:272 ; 4-byte Folded Reload
 ; GFX9-O0-NEXT:    buffer_load_dword v15, off, s[0:3], s32 offset:276 ; 4-byte Folded Reload
 ; GFX9-O0-NEXT:    buffer_load_dword v11, off, s[0:3], s32 offset:280 ; 4-byte Folded Reload
@@ -903,14 +913,14 @@ define i128 @v_sdiv_i128_vv(i128 %lhs, i128 %rhs) {
 ; GFX9-O0-NEXT:    s_or_saveexec_b64 s[18:19], -1
 ; GFX9-O0-NEXT:    buffer_load_dword v16, off, s[0:3], s32 ; 4-byte Folded Reload
 ; GFX9-O0-NEXT:    s_mov_b64 exec, s[18:19]
-; GFX9-O0-NEXT:    buffer_load_dword v17, off, s[0:3], s32 offset:52 ; 4-byte Folded Reload
-; GFX9-O0-NEXT:    buffer_load_dword v18, off, s[0:3], s32 offset:56 ; 4-byte Folded Reload
-; GFX9-O0-NEXT:    buffer_load_dword v13, off, s[0:3], s32 offset:60 ; 4-byte Folded Reload
-; GFX9-O0-NEXT:    buffer_load_dword v14, off, s[0:3], s32 offset:64 ; 4-byte Folded Reload
-; GFX9-O0-NEXT:    buffer_load_dword v19, off, s[0:3], s32 offset:36 ; 4-byte Folded Reload
-; GFX9-O0-NEXT:    buffer_load_dword v20, off, s[0:3], s32 offset:40 ; 4-byte Folded Reload
-; GFX9-O0-NEXT:    buffer_load_dword v21, off, s[0:3], s32 offset:44 ; 4-byte Folded Reload
-; GFX9-O0-NEXT:    buffer_load_dword v22, off, s[0:3], s32 offset:48 ; 4-byte Folded Reload
+; GFX9-O0-NEXT:    buffer_load_dword v17, off, s[0:3], s32 offset:60 ; 4-byte Folded Reload
+; GFX9-O0-NEXT:    buffer_load_dword v18, off, s[0:3], s32 offset:64 ; 4-byte Folded Reload
+; GFX9-O0-NEXT:    buffer_load_dword v13, off, s[0:3], s32 offset:52 ; 4-byte Folded Reload
+; GFX9-O0-NEXT:    buffer_load_dword v14, off, s[0:3], s32 offset:56 ; 4-byte Folded Reload
+; GFX9-O0-NEXT:    buffer_load_dword v19, off, s[0:3], s32 offset:44 ; 4-byte Folded Reload
+; GFX9-O0-NEXT:    buffer_load_dword v20, off, s[0:3], s32 offset:48 ; 4-byte Folded Reload
+; GFX9-O0-NEXT:    buffer_load_dword v21, off, s[0:3], s32 offset:36 ; 4-byte Folded Reload
+; GFX9-O0-NEXT:    buffer_load_dword v22, off, s[0:3], s32 offset:40 ; 4-byte Folded Reload
 ; GFX9-O0-NEXT:    s_waitcnt vmcnt(9)
 ; GFX9-O0-NEXT:    v_mov_b32_e32 v4, v10
 ; GFX9-O0-NEXT:    s_waitcnt vmcnt(0)
@@ -1028,10 +1038,10 @@ define i128 @v_sdiv_i128_vv(i128 %lhs, i128 %rhs) {
 ; GFX9-O0-NEXT:    s_or_saveexec_b64 s[18:19], -1
 ; GFX9-O0-NEXT:    buffer_load_dword v0, off, s[0:3], s32 ; 4-byte Folded Reload
 ; GFX9-O0-NEXT:    s_mov_b64 exec, s[18:19]
-; GFX9-O0-NEXT:    buffer_load_dword v7, off, s[0:3], s32 offset:44 ; 4-byte Folded Reload
-; GFX9-O0-NEXT:    buffer_load_dword v8, off, s[0:3], s32 offset:48 ; 4-byte Folded Reload
-; GFX9-O0-NEXT:    buffer_load_dword v11, off, s[0:3], s32 offset:36 ; 4-byte Folded Reload
-; GFX9-O0-NEXT:    buffer_load_dword v12, off, s[0:3], s32 offset:40 ; 4-byte Folded Reload
+; GFX9-O0-NEXT:    buffer_load_dword v7, off, s[0:3], s32 offset:36 ; 4-byte Folded Reload
+; GFX9-O0-NEXT:    buffer_load_dword v8, off, s[0:3], s32 offset:40 ; 4-byte Folded Reload
+; GFX9-O0-NEXT:    buffer_load_dword v11, off, s[0:3], s32 offset:44 ; 4-byte Folded Reload
+; GFX9-O0-NEXT:    buffer_load_dword v12, off, s[0:3], s32 offset:48 ; 4-byte Folded Reload
 ; GFX9-O0-NEXT:    buffer_load_dword v5, off, s[0:3], s32 offset:28 ; 4-byte Folded Reload
 ; GFX9-O0-NEXT:    buffer_load_dword v6, off, s[0:3], s32 offset:32 ; 4-byte Folded Reload
 ; GFX9-O0-NEXT:    buffer_load_dword v1, off, s[0:3], s32 offset:20 ; 4-byte Folded Reload
diff --git a/llvm/test/CodeGen/AMDGPU/extra-lds-size.ll b/llvm/test/CodeGen/AMDGPU/extra-lds-size.ll
index e376c3df1ac9..96ec90b1f4d0 100644
--- a/llvm/test/CodeGen/AMDGPU/extra-lds-size.ll
+++ b/llvm/test/CodeGen/AMDGPU/extra-lds-size.ll
@@ -5,12 +5,12 @@
 
 ; Check EXTRA_LDS_SIZE in SPI_SHADER_PGM_RSRC2_PS.
 
-; GFX10-PAL: 0x2c0b (SPI_SHADER_PGM_RSRC2_PS): 0x800
+; GFX10-PAL: '0x2c0b (SPI_SHADER_PGM_RSRC2_PS)': 0x800
 
 ; GFX10-MESA: .long 45100
 ; GFX10-MESA-NEXT: .long 2048
 
-; GFX11-PAL: 0x2c0b (SPI_SHADER_PGM_RSRC2_PS): 0x400
+; GFX11-PAL: '0x2c0b (SPI_SHADER_PGM_RSRC2_PS)': 0x400
 
 ; GFX11-MESA: .long 45100
 ; GFX11-MESA-NEXT: .long 1024
diff --git a/llvm/test/CodeGen/AMDGPU/flat_atomics.ll b/llvm/test/CodeGen/AMDGPU/flat_atomics.ll
index 06ba60518adc..e44572985e6d 100644
--- a/llvm/test/CodeGen/AMDGPU/flat_atomics.ll
+++ b/llvm/test/CodeGen/AMDGPU/flat_atomics.ll
@@ -6741,6 +6741,81 @@ entry:
   ret void
 }
 
+define amdgpu_kernel void @atomic_store_bf16_offset(bfloat %in, ptr %out) {
+; GCN1-LABEL: atomic_store_bf16_offset:
+; GCN1:       ; %bb.0:
+; GCN1-NEXT:    s_load_dwordx2 s[2:3], s[0:1], 0xb
+; GCN1-NEXT:    s_load_dword s0, s[0:1], 0x9
+; GCN1-NEXT:    s_waitcnt lgkmcnt(0)
+; GCN1-NEXT:    v_mov_b32_e32 v0, s2
+; GCN1-NEXT:    v_mov_b32_e32 v1, s3
+; GCN1-NEXT:    v_mov_b32_e32 v2, s0
+; GCN1-NEXT:    flat_store_short v[0:1], v2
+; GCN1-NEXT:    s_endpgm
+;
+; GCN2-LABEL: atomic_store_bf16_offset:
+; GCN2:       ; %bb.0:
+; GCN2-NEXT:    s_load_dwordx2 s[2:3], s[0:1], 0x2c
+; GCN2-NEXT:    s_load_dword s0, s[0:1], 0x24
+; GCN2-NEXT:    s_waitcnt lgkmcnt(0)
+; GCN2-NEXT:    v_mov_b32_e32 v0, s2
+; GCN2-NEXT:    v_mov_b32_e32 v1, s3
+; GCN2-NEXT:    v_mov_b32_e32 v2, s0
+; GCN2-NEXT:    flat_store_short v[0:1], v2
+; GCN2-NEXT:    s_endpgm
+;
+; GCN3-LABEL: atomic_store_bf16_offset:
+; GCN3:       ; %bb.0:
+; GCN3-NEXT:    s_load_dwordx2 s[2:3], s[0:1], 0x2c
+; GCN3-NEXT:    s_load_dword s4, s[0:1], 0x24
+; GCN3-NEXT:    s_waitcnt lgkmcnt(0)
+; GCN3-NEXT:    v_mov_b32_e32 v0, s2
+; GCN3-NEXT:    v_mov_b32_e32 v1, s3
+; GCN3-NEXT:    v_mov_b32_e32 v2, s4
+; GCN3-NEXT:    flat_store_short v[0:1], v2
+; GCN3-NEXT:    s_endpgm
+  %gep = getelementptr bfloat, ptr %out, i64 8
+  store atomic bfloat %in, ptr %out seq_cst, align 2
+  ret void
+}
+
+define amdgpu_kernel void @atomic_store_bf16(bfloat %in, ptr %out) {
+; GCN1-LABEL: atomic_store_bf16:
+; GCN1:       ; %bb.0:
+; GCN1-NEXT:    s_load_dwordx2 s[2:3], s[0:1], 0xb
+; GCN1-NEXT:    s_load_dword s0, s[0:1], 0x9
+; GCN1-NEXT:    s_waitcnt lgkmcnt(0)
+; GCN1-NEXT:    v_mov_b32_e32 v0, s2
+; GCN1-NEXT:    v_mov_b32_e32 v1, s3
+; GCN1-NEXT:    v_mov_b32_e32 v2, s0
+; GCN1-NEXT:    flat_store_short v[0:1], v2
+; GCN1-NEXT:    s_endpgm
+;
+; GCN2-LABEL: atomic_store_bf16:
+; GCN2:       ; %bb.0:
+; GCN2-NEXT:    s_load_dwordx2 s[2:3], s[0:1], 0x2c
+; GCN2-NEXT:    s_load_dword s0, s[0:1], 0x24
+; GCN2-NEXT:    s_waitcnt lgkmcnt(0)
+; GCN2-NEXT:    v_mov_b32_e32 v0, s2
+; GCN2-NEXT:    v_mov_b32_e32 v1, s3
+; GCN2-NEXT:    v_mov_b32_e32 v2, s0
+; GCN2-NEXT:    flat_store_short v[0:1], v2
+; GCN2-NEXT:    s_endpgm
+;
+; GCN3-LABEL: atomic_store_bf16:
+; GCN3:       ; %bb.0:
+; GCN3-NEXT:    s_load_dwordx2 s[2:3], s[0:1], 0x2c
+; GCN3-NEXT:    s_load_dword s4, s[0:1], 0x24
+; GCN3-NEXT:    s_waitcnt lgkmcnt(0)
+; GCN3-NEXT:    v_mov_b32_e32 v0, s2
+; GCN3-NEXT:    v_mov_b32_e32 v1, s3
+; GCN3-NEXT:    v_mov_b32_e32 v2, s4
+; GCN3-NEXT:    flat_store_short v[0:1], v2
+; GCN3-NEXT:    s_endpgm
+  store atomic bfloat %in, ptr %out seq_cst, align 2
+  ret void
+}
+
 define amdgpu_kernel void @atomic_inc_i32_offset(ptr %out, i32 %in) {
 ; GCN1-LABEL: atomic_inc_i32_offset:
 ; GCN1:       ; %bb.0: ; %entry
@@ -7868,3 +7943,201 @@ entry:
   store i32 %val, ptr %out2
   ret void
 }
+
+define amdgpu_kernel void @atomic_load_f16_offset(ptr %in, ptr %out) {
+; GCN1-LABEL: atomic_load_f16_offset:
+; GCN1:       ; %bb.0:
+; GCN1-NEXT:    s_load_dwordx4 s[0:3], s[0:1], 0x9
+; GCN1-NEXT:    s_waitcnt lgkmcnt(0)
+; GCN1-NEXT:    s_add_u32 s0, s0, 16
+; GCN1-NEXT:    s_addc_u32 s1, s1, 0
+; GCN1-NEXT:    v_mov_b32_e32 v0, s0
+; GCN1-NEXT:    v_mov_b32_e32 v1, s1
+; GCN1-NEXT:    flat_load_ushort v2, v[0:1] glc
+; GCN1-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
+; GCN1-NEXT:    buffer_wbinvl1_vol
+; GCN1-NEXT:    v_mov_b32_e32 v0, s2
+; GCN1-NEXT:    v_mov_b32_e32 v1, s3
+; GCN1-NEXT:    flat_store_short v[0:1], v2
+; GCN1-NEXT:    s_endpgm
+;
+; GCN2-LABEL: atomic_load_f16_offset:
+; GCN2:       ; %bb.0:
+; GCN2-NEXT:    s_load_dwordx4 s[0:3], s[0:1], 0x24
+; GCN2-NEXT:    s_waitcnt lgkmcnt(0)
+; GCN2-NEXT:    s_add_u32 s0, s0, 16
+; GCN2-NEXT:    s_addc_u32 s1, s1, 0
+; GCN2-NEXT:    v_mov_b32_e32 v0, s0
+; GCN2-NEXT:    v_mov_b32_e32 v1, s1
+; GCN2-NEXT:    flat_load_ushort v2, v[0:1] glc
+; GCN2-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
+; GCN2-NEXT:    buffer_wbinvl1_vol
+; GCN2-NEXT:    v_mov_b32_e32 v0, s2
+; GCN2-NEXT:    v_mov_b32_e32 v1, s3
+; GCN2-NEXT:    flat_store_short v[0:1], v2
+; GCN2-NEXT:    s_endpgm
+;
+; GCN3-LABEL: atomic_load_f16_offset:
+; GCN3:       ; %bb.0:
+; GCN3-NEXT:    s_load_dwordx4 s[0:3], s[0:1], 0x24
+; GCN3-NEXT:    s_waitcnt lgkmcnt(0)
+; GCN3-NEXT:    v_mov_b32_e32 v0, s0
+; GCN3-NEXT:    v_mov_b32_e32 v1, s1
+; GCN3-NEXT:    flat_load_ushort v2, v[0:1] offset:16 glc
+; GCN3-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
+; GCN3-NEXT:    buffer_wbinvl1_vol
+; GCN3-NEXT:    v_mov_b32_e32 v0, s2
+; GCN3-NEXT:    v_mov_b32_e32 v1, s3
+; GCN3-NEXT:    flat_store_short v[0:1], v2
+; GCN3-NEXT:    s_endpgm
+  %gep = getelementptr half, ptr %in, i64 8
+  %val = load atomic half, ptr %gep  seq_cst, align 2
+  store half %val, ptr %out
+  ret void
+}
+
+define amdgpu_kernel void @atomic_load_f16(ptr %in, ptr %out) {
+; GCN1-LABEL: atomic_load_f16:
+; GCN1:       ; %bb.0:
+; GCN1-NEXT:    s_load_dwordx4 s[0:3], s[0:1], 0x9
+; GCN1-NEXT:    s_waitcnt lgkmcnt(0)
+; GCN1-NEXT:    v_mov_b32_e32 v0, s0
+; GCN1-NEXT:    v_mov_b32_e32 v1, s1
+; GCN1-NEXT:    flat_load_ushort v2, v[0:1] glc
+; GCN1-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
+; GCN1-NEXT:    buffer_wbinvl1_vol
+; GCN1-NEXT:    v_mov_b32_e32 v0, s2
+; GCN1-NEXT:    v_mov_b32_e32 v1, s3
+; GCN1-NEXT:    flat_store_short v[0:1], v2
+; GCN1-NEXT:    s_endpgm
+;
+; GCN2-LABEL: atomic_load_f16:
+; GCN2:       ; %bb.0:
+; GCN2-NEXT:    s_load_dwordx4 s[0:3], s[0:1], 0x24
+; GCN2-NEXT:    s_waitcnt lgkmcnt(0)
+; GCN2-NEXT:    v_mov_b32_e32 v0, s0
+; GCN2-NEXT:    v_mov_b32_e32 v1, s1
+; GCN2-NEXT:    flat_load_ushort v2, v[0:1] glc
+; GCN2-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
+; GCN2-NEXT:    buffer_wbinvl1_vol
+; GCN2-NEXT:    v_mov_b32_e32 v0, s2
+; GCN2-NEXT:    v_mov_b32_e32 v1, s3
+; GCN2-NEXT:    flat_store_short v[0:1], v2
+; GCN2-NEXT:    s_endpgm
+;
+; GCN3-LABEL: atomic_load_f16:
+; GCN3:       ; %bb.0:
+; GCN3-NEXT:    s_load_dwordx4 s[0:3], s[0:1], 0x24
+; GCN3-NEXT:    s_waitcnt lgkmcnt(0)
+; GCN3-NEXT:    v_mov_b32_e32 v0, s0
+; GCN3-NEXT:    v_mov_b32_e32 v1, s1
+; GCN3-NEXT:    flat_load_ushort v2, v[0:1] glc
+; GCN3-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
+; GCN3-NEXT:    buffer_wbinvl1_vol
+; GCN3-NEXT:    v_mov_b32_e32 v0, s2
+; GCN3-NEXT:    v_mov_b32_e32 v1, s3
+; GCN3-NEXT:    flat_store_short v[0:1], v2
+; GCN3-NEXT:    s_endpgm
+  %val = load atomic half, ptr %in seq_cst, align 2
+  store half %val, ptr %out
+  ret void
+}
+
+define amdgpu_kernel void @atomic_load_bf16_offset(ptr %in, ptr %out) {
+; GCN1-LABEL: atomic_load_bf16_offset:
+; GCN1:       ; %bb.0:
+; GCN1-NEXT:    s_load_dwordx4 s[0:3], s[0:1], 0x9
+; GCN1-NEXT:    s_waitcnt lgkmcnt(0)
+; GCN1-NEXT:    s_add_u32 s0, s0, 16
+; GCN1-NEXT:    s_addc_u32 s1, s1, 0
+; GCN1-NEXT:    v_mov_b32_e32 v0, s0
+; GCN1-NEXT:    v_mov_b32_e32 v1, s1
+; GCN1-NEXT:    flat_load_ushort v2, v[0:1] glc
+; GCN1-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
+; GCN1-NEXT:    buffer_wbinvl1_vol
+; GCN1-NEXT:    v_mov_b32_e32 v0, s2
+; GCN1-NEXT:    v_mov_b32_e32 v1, s3
+; GCN1-NEXT:    flat_store_short v[0:1], v2
+; GCN1-NEXT:    s_endpgm
+;
+; GCN2-LABEL: atomic_load_bf16_offset:
+; GCN2:       ; %bb.0:
+; GCN2-NEXT:    s_load_dwordx4 s[0:3], s[0:1], 0x24
+; GCN2-NEXT:    s_waitcnt lgkmcnt(0)
+; GCN2-NEXT:    s_add_u32 s0, s0, 16
+; GCN2-NEXT:    s_addc_u32 s1, s1, 0
+; GCN2-NEXT:    v_mov_b32_e32 v0, s0
+; GCN2-NEXT:    v_mov_b32_e32 v1, s1
+; GCN2-NEXT:    flat_load_ushort v2, v[0:1] glc
+; GCN2-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
+; GCN2-NEXT:    buffer_wbinvl1_vol
+; GCN2-NEXT:    v_mov_b32_e32 v0, s2
+; GCN2-NEXT:    v_mov_b32_e32 v1, s3
+; GCN2-NEXT:    flat_store_short v[0:1], v2
+; GCN2-NEXT:    s_endpgm
+;
+; GCN3-LABEL: atomic_load_bf16_offset:
+; GCN3:       ; %bb.0:
+; GCN3-NEXT:    s_load_dwordx4 s[0:3], s[0:1], 0x24
+; GCN3-NEXT:    s_waitcnt lgkmcnt(0)
+; GCN3-NEXT:    v_mov_b32_e32 v0, s0
+; GCN3-NEXT:    v_mov_b32_e32 v1, s1
+; GCN3-NEXT:    flat_load_ushort v2, v[0:1] offset:16 glc
+; GCN3-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
+; GCN3-NEXT:    buffer_wbinvl1_vol
+; GCN3-NEXT:    v_mov_b32_e32 v0, s2
+; GCN3-NEXT:    v_mov_b32_e32 v1, s3
+; GCN3-NEXT:    flat_store_short v[0:1], v2
+; GCN3-NEXT:    s_endpgm
+  %gep = getelementptr bfloat, ptr %in, i64 8
+  %val = load atomic bfloat, ptr %gep  seq_cst, align 2
+  store bfloat %val, ptr %out
+  ret void
+}
+
+define amdgpu_kernel void @atomic_load_bf16(ptr %in, ptr %out) {
+; GCN1-LABEL: atomic_load_bf16:
+; GCN1:       ; %bb.0:
+; GCN1-NEXT:    s_load_dwordx4 s[0:3], s[0:1], 0x9
+; GCN1-NEXT:    s_waitcnt lgkmcnt(0)
+; GCN1-NEXT:    v_mov_b32_e32 v0, s0
+; GCN1-NEXT:    v_mov_b32_e32 v1, s1
+; GCN1-NEXT:    flat_load_ushort v2, v[0:1] glc
+; GCN1-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
+; GCN1-NEXT:    buffer_wbinvl1_vol
+; GCN1-NEXT:    v_mov_b32_e32 v0, s2
+; GCN1-NEXT:    v_mov_b32_e32 v1, s3
+; GCN1-NEXT:    flat_store_short v[0:1], v2
+; GCN1-NEXT:    s_endpgm
+;
+; GCN2-LABEL: atomic_load_bf16:
+; GCN2:       ; %bb.0:
+; GCN2-NEXT:    s_load_dwordx4 s[0:3], s[0:1], 0x24
+; GCN2-NEXT:    s_waitcnt lgkmcnt(0)
+; GCN2-NEXT:    v_mov_b32_e32 v0, s0
+; GCN2-NEXT:    v_mov_b32_e32 v1, s1
+; GCN2-NEXT:    flat_load_ushort v2, v[0:1] glc
+; GCN2-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
+; GCN2-NEXT:    buffer_wbinvl1_vol
+; GCN2-NEXT:    v_mov_b32_e32 v0, s2
+; GCN2-NEXT:    v_mov_b32_e32 v1, s3
+; GCN2-NEXT:    flat_store_short v[0:1], v2
+; GCN2-NEXT:    s_endpgm
+;
+; GCN3-LABEL: atomic_load_bf16:
+; GCN3:       ; %bb.0:
+; GCN3-NEXT:    s_load_dwordx4 s[0:3], s[0:1], 0x24
+; GCN3-NEXT:    s_waitcnt lgkmcnt(0)
+; GCN3-NEXT:    v_mov_b32_e32 v0, s0
+; GCN3-NEXT:    v_mov_b32_e32 v1, s1
+; GCN3-NEXT:    flat_load_ushort v2, v[0:1] glc
+; GCN3-NEXT:    s_waitcnt vmcnt(0) lgkmcnt(0)
+; GCN3-NEXT:    buffer_wbinvl1_vol
+; GCN3-NEXT:    v_mov_b32_e32 v0, s2
+; GCN3-NEXT:    v_mov_b32_e32 v1, s3
+; GCN3-NEXT:    flat_store_short v[0:1], v2
+; GCN3-NEXT:    s_endpgm
+  %val = load atomic bfloat, ptr %in seq_cst, align 2
+  store bfloat %val, ptr %out
+  ret void
+}
diff --git a/llvm/test/CodeGen/AMDGPU/flat_atomics_i64.ll b/llvm/test/CodeGen/AMDGPU/flat_atomics_i64.ll
index 4d36fb314351..b8c8d993d389 100644
--- a/llvm/test/CodeGen/AMDGPU/flat_atomics_i64.ll
+++ b/llvm/test/CodeGen/AMDGPU/flat_atomics_i64.ll
@@ -3641,7 +3641,7 @@ define amdgpu_kernel void @atomic_xchg_f64_offset(ptr %out, double %in) {
 ; GFX12-NEXT:    v_dual_mov_b32 v2, s2 :: v_dual_mov_b32 v3, s3
 ; GFX12-NEXT:    flat_atomic_swap_b64 v[0:1], v[2:3] offset:32
 ; GFX12-NEXT:    s_wait_storecnt_dscnt 0x0
-; GFX12-NEXT:    global_inv scope:SCOPE_SYS
+; GFX12-NEXT:    global_inv scope:SCOPE_DEV
 ; GFX12-NEXT:    s_endpgm
 entry:
   %gep = getelementptr double, ptr %out, i64 4
@@ -3688,7 +3688,7 @@ define amdgpu_kernel void @atomic_xchg_pointer_offset(ptr %out, ptr %in) {
 ; GFX12-NEXT:    v_dual_mov_b32 v2, s2 :: v_dual_mov_b32 v3, s3
 ; GFX12-NEXT:    flat_atomic_swap_b64 v[0:1], v[2:3] offset:32
 ; GFX12-NEXT:    s_wait_storecnt_dscnt 0x0
-; GFX12-NEXT:    global_inv scope:SCOPE_SYS
+; GFX12-NEXT:    global_inv scope:SCOPE_DEV
 ; GFX12-NEXT:    s_endpgm
 entry:
   %gep = getelementptr ptr, ptr %out, i32 4
diff --git a/llvm/test/CodeGen/AMDGPU/fp_trunc_store_bf16.ll b/llvm/test/CodeGen/AMDGPU/fp_trunc_store_bf16.ll
new file mode 100644
index 000000000000..5aaff773689f
--- /dev/null
+++ b/llvm/test/CodeGen/AMDGPU/fp_trunc_store_bf16.ll
@@ -0,0 +1,426 @@
+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --version 4
+; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=gfx90a %s -o - | FileCheck %s
+
+define void @v2(<2 x float> %num, ptr addrspace(1) %p) {
+; CHECK-LABEL: v2:
+; CHECK:       ; %bb.0: ; %entry
+; CHECK-NEXT:    s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
+; CHECK-NEXT:    v_bfe_u32 v4, v0, 16, 1
+; CHECK-NEXT:    s_movk_i32 s4, 0x7fff
+; CHECK-NEXT:    v_add3_u32 v4, v4, v0, s4
+; CHECK-NEXT:    v_or_b32_e32 v5, 0x400000, v0
+; CHECK-NEXT:    v_cmp_u_f32_e32 vcc, v0, v0
+; CHECK-NEXT:    v_cndmask_b32_e32 v0, v4, v5, vcc
+; CHECK-NEXT:    v_bfe_u32 v4, v1, 16, 1
+; CHECK-NEXT:    v_add3_u32 v4, v4, v1, s4
+; CHECK-NEXT:    v_or_b32_e32 v5, 0x400000, v1
+; CHECK-NEXT:    v_cmp_u_f32_e32 vcc, v1, v1
+; CHECK-NEXT:    v_cndmask_b32_e32 v1, v4, v5, vcc
+; CHECK-NEXT:    s_mov_b32 s4, 0x7060302
+; CHECK-NEXT:    v_perm_b32 v0, v1, v0, s4
+; CHECK-NEXT:    global_store_dword v[2:3], v0, off
+; CHECK-NEXT:    s_waitcnt vmcnt(0)
+; CHECK-NEXT:    s_setpc_b64 s[30:31]
+entry:
+  %conv = fptrunc <2 x float> %num to <2 x bfloat>
+  store <2 x bfloat> %conv, ptr addrspace(1) %p, align 8
+  ret void
+}
+
+define void @v4(<4 x float> %num, ptr addrspace(1) %p) {
+; CHECK-LABEL: v4:
+; CHECK:       ; %bb.0: ; %entry
+; CHECK-NEXT:    s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
+; CHECK-NEXT:    v_bfe_u32 v6, v2, 16, 1
+; CHECK-NEXT:    s_movk_i32 s4, 0x7fff
+; CHECK-NEXT:    v_add3_u32 v6, v6, v2, s4
+; CHECK-NEXT:    v_or_b32_e32 v7, 0x400000, v2
+; CHECK-NEXT:    v_cmp_u_f32_e32 vcc, v2, v2
+; CHECK-NEXT:    v_cndmask_b32_e32 v2, v6, v7, vcc
+; CHECK-NEXT:    v_bfe_u32 v6, v3, 16, 1
+; CHECK-NEXT:    v_add3_u32 v6, v6, v3, s4
+; CHECK-NEXT:    v_or_b32_e32 v7, 0x400000, v3
+; CHECK-NEXT:    v_cmp_u_f32_e32 vcc, v3, v3
+; CHECK-NEXT:    v_cndmask_b32_e32 v3, v6, v7, vcc
+; CHECK-NEXT:    s_mov_b32 s5, 0x7060302
+; CHECK-NEXT:    v_perm_b32 v3, v3, v2, s5
+; CHECK-NEXT:    v_bfe_u32 v2, v0, 16, 1
+; CHECK-NEXT:    v_add3_u32 v2, v2, v0, s4
+; CHECK-NEXT:    v_or_b32_e32 v6, 0x400000, v0
+; CHECK-NEXT:    v_cmp_u_f32_e32 vcc, v0, v0
+; CHECK-NEXT:    v_cndmask_b32_e32 v0, v2, v6, vcc
+; CHECK-NEXT:    v_bfe_u32 v2, v1, 16, 1
+; CHECK-NEXT:    v_add3_u32 v2, v2, v1, s4
+; CHECK-NEXT:    v_or_b32_e32 v6, 0x400000, v1
+; CHECK-NEXT:    v_cmp_u_f32_e32 vcc, v1, v1
+; CHECK-NEXT:    v_cndmask_b32_e32 v1, v2, v6, vcc
+; CHECK-NEXT:    v_perm_b32 v2, v1, v0, s5
+; CHECK-NEXT:    global_store_dwordx2 v[4:5], v[2:3], off
+; CHECK-NEXT:    s_waitcnt vmcnt(0)
+; CHECK-NEXT:    s_setpc_b64 s[30:31]
+entry:
+  %conv = fptrunc <4 x float> %num to <4 x bfloat>
+  store <4 x bfloat> %conv, ptr addrspace(1) %p, align 8
+  ret void
+}
+
+define void @v8(<8 x float> %num, ptr addrspace(1) %p) {
+; CHECK-LABEL: v8:
+; CHECK:       ; %bb.0: ; %entry
+; CHECK-NEXT:    s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
+; CHECK-NEXT:    v_bfe_u32 v10, v6, 16, 1
+; CHECK-NEXT:    s_movk_i32 s4, 0x7fff
+; CHECK-NEXT:    v_add3_u32 v10, v10, v6, s4
+; CHECK-NEXT:    v_or_b32_e32 v11, 0x400000, v6
+; CHECK-NEXT:    v_cmp_u_f32_e32 vcc, v6, v6
+; CHECK-NEXT:    v_cndmask_b32_e32 v6, v10, v11, vcc
+; CHECK-NEXT:    v_bfe_u32 v10, v7, 16, 1
+; CHECK-NEXT:    v_add3_u32 v10, v10, v7, s4
+; CHECK-NEXT:    v_or_b32_e32 v11, 0x400000, v7
+; CHECK-NEXT:    v_cmp_u_f32_e32 vcc, v7, v7
+; CHECK-NEXT:    v_cndmask_b32_e32 v7, v10, v11, vcc
+; CHECK-NEXT:    s_mov_b32 s5, 0x7060302
+; CHECK-NEXT:    v_perm_b32 v7, v7, v6, s5
+; CHECK-NEXT:    v_bfe_u32 v6, v4, 16, 1
+; CHECK-NEXT:    v_add3_u32 v6, v6, v4, s4
+; CHECK-NEXT:    v_or_b32_e32 v10, 0x400000, v4
+; CHECK-NEXT:    v_cmp_u_f32_e32 vcc, v4, v4
+; CHECK-NEXT:    v_cndmask_b32_e32 v4, v6, v10, vcc
+; CHECK-NEXT:    v_bfe_u32 v6, v5, 16, 1
+; CHECK-NEXT:    v_add3_u32 v6, v6, v5, s4
+; CHECK-NEXT:    v_or_b32_e32 v10, 0x400000, v5
+; CHECK-NEXT:    v_cmp_u_f32_e32 vcc, v5, v5
+; CHECK-NEXT:    v_cndmask_b32_e32 v5, v6, v10, vcc
+; CHECK-NEXT:    v_perm_b32 v6, v5, v4, s5
+; CHECK-NEXT:    v_bfe_u32 v4, v2, 16, 1
+; CHECK-NEXT:    v_add3_u32 v4, v4, v2, s4
+; CHECK-NEXT:    v_or_b32_e32 v5, 0x400000, v2
+; CHECK-NEXT:    v_cmp_u_f32_e32 vcc, v2, v2
+; CHECK-NEXT:    v_cndmask_b32_e32 v2, v4, v5, vcc
+; CHECK-NEXT:    v_bfe_u32 v4, v3, 16, 1
+; CHECK-NEXT:    v_add3_u32 v4, v4, v3, s4
+; CHECK-NEXT:    v_or_b32_e32 v5, 0x400000, v3
+; CHECK-NEXT:    v_cmp_u_f32_e32 vcc, v3, v3
+; CHECK-NEXT:    v_cndmask_b32_e32 v3, v4, v5, vcc
+; CHECK-NEXT:    v_perm_b32 v5, v3, v2, s5
+; CHECK-NEXT:    v_bfe_u32 v2, v0, 16, 1
+; CHECK-NEXT:    v_add3_u32 v2, v2, v0, s4
+; CHECK-NEXT:    v_or_b32_e32 v3, 0x400000, v0
+; CHECK-NEXT:    v_cmp_u_f32_e32 vcc, v0, v0
+; CHECK-NEXT:    v_cndmask_b32_e32 v0, v2, v3, vcc
+; CHECK-NEXT:    v_bfe_u32 v2, v1, 16, 1
+; CHECK-NEXT:    v_add3_u32 v2, v2, v1, s4
+; CHECK-NEXT:    v_or_b32_e32 v3, 0x400000, v1
+; CHECK-NEXT:    v_cmp_u_f32_e32 vcc, v1, v1
+; CHECK-NEXT:    v_cndmask_b32_e32 v1, v2, v3, vcc
+; CHECK-NEXT:    v_perm_b32 v4, v1, v0, s5
+; CHECK-NEXT:    global_store_dwordx4 v[8:9], v[4:7], off
+; CHECK-NEXT:    s_waitcnt vmcnt(0)
+; CHECK-NEXT:    s_setpc_b64 s[30:31]
+entry:
+  %conv = fptrunc <8 x float> %num to <8 x bfloat>
+  store <8 x bfloat> %conv, ptr addrspace(1) %p, align 8
+  ret void
+}
+
+define void @v16(<16 x float> %num, ptr addrspace(1) %p) {
+; CHECK-LABEL: v16:
+; CHECK:       ; %bb.0: ; %entry
+; CHECK-NEXT:    s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
+; CHECK-NEXT:    v_bfe_u32 v18, v6, 16, 1
+; CHECK-NEXT:    s_movk_i32 s4, 0x7fff
+; CHECK-NEXT:    v_add3_u32 v18, v18, v6, s4
+; CHECK-NEXT:    v_or_b32_e32 v19, 0x400000, v6
+; CHECK-NEXT:    v_cmp_u_f32_e32 vcc, v6, v6
+; CHECK-NEXT:    v_cndmask_b32_e32 v6, v18, v19, vcc
+; CHECK-NEXT:    v_bfe_u32 v18, v7, 16, 1
+; CHECK-NEXT:    v_add3_u32 v18, v18, v7, s4
+; CHECK-NEXT:    v_or_b32_e32 v19, 0x400000, v7
+; CHECK-NEXT:    v_cmp_u_f32_e32 vcc, v7, v7
+; CHECK-NEXT:    v_cndmask_b32_e32 v7, v18, v19, vcc
+; CHECK-NEXT:    s_mov_b32 s5, 0x7060302
+; CHECK-NEXT:    v_perm_b32 v7, v7, v6, s5
+; CHECK-NEXT:    v_bfe_u32 v6, v4, 16, 1
+; CHECK-NEXT:    v_add3_u32 v6, v6, v4, s4
+; CHECK-NEXT:    v_or_b32_e32 v18, 0x400000, v4
+; CHECK-NEXT:    v_cmp_u_f32_e32 vcc, v4, v4
+; CHECK-NEXT:    v_cndmask_b32_e32 v4, v6, v18, vcc
+; CHECK-NEXT:    v_bfe_u32 v6, v5, 16, 1
+; CHECK-NEXT:    v_add3_u32 v6, v6, v5, s4
+; CHECK-NEXT:    v_or_b32_e32 v18, 0x400000, v5
+; CHECK-NEXT:    v_cmp_u_f32_e32 vcc, v5, v5
+; CHECK-NEXT:    v_cndmask_b32_e32 v5, v6, v18, vcc
+; CHECK-NEXT:    v_perm_b32 v6, v5, v4, s5
+; CHECK-NEXT:    v_bfe_u32 v4, v2, 16, 1
+; CHECK-NEXT:    v_add3_u32 v4, v4, v2, s4
+; CHECK-NEXT:    v_or_b32_e32 v5, 0x400000, v2
+; CHECK-NEXT:    v_cmp_u_f32_e32 vcc, v2, v2
+; CHECK-NEXT:    v_cndmask_b32_e32 v2, v4, v5, vcc
+; CHECK-NEXT:    v_bfe_u32 v4, v3, 16, 1
+; CHECK-NEXT:    v_add3_u32 v4, v4, v3, s4
+; CHECK-NEXT:    v_or_b32_e32 v5, 0x400000, v3
+; CHECK-NEXT:    v_cmp_u_f32_e32 vcc, v3, v3
+; CHECK-NEXT:    v_cndmask_b32_e32 v3, v4, v5, vcc
+; CHECK-NEXT:    v_perm_b32 v5, v3, v2, s5
+; CHECK-NEXT:    v_bfe_u32 v2, v0, 16, 1
+; CHECK-NEXT:    v_add3_u32 v2, v2, v0, s4
+; CHECK-NEXT:    v_or_b32_e32 v3, 0x400000, v0
+; CHECK-NEXT:    v_cmp_u_f32_e32 vcc, v0, v0
+; CHECK-NEXT:    v_cndmask_b32_e32 v0, v2, v3, vcc
+; CHECK-NEXT:    v_bfe_u32 v2, v1, 16, 1
+; CHECK-NEXT:    v_add3_u32 v2, v2, v1, s4
+; CHECK-NEXT:    v_or_b32_e32 v3, 0x400000, v1
+; CHECK-NEXT:    v_cmp_u_f32_e32 vcc, v1, v1
+; CHECK-NEXT:    v_cndmask_b32_e32 v1, v2, v3, vcc
+; CHECK-NEXT:    v_perm_b32 v4, v1, v0, s5
+; CHECK-NEXT:    v_bfe_u32 v0, v14, 16, 1
+; CHECK-NEXT:    v_add3_u32 v0, v0, v14, s4
+; CHECK-NEXT:    v_or_b32_e32 v1, 0x400000, v14
+; CHECK-NEXT:    v_cmp_u_f32_e32 vcc, v14, v14
+; CHECK-NEXT:    v_cndmask_b32_e32 v0, v0, v1, vcc
+; CHECK-NEXT:    v_bfe_u32 v1, v15, 16, 1
+; CHECK-NEXT:    v_add3_u32 v1, v1, v15, s4
+; CHECK-NEXT:    v_or_b32_e32 v2, 0x400000, v15
+; CHECK-NEXT:    v_cmp_u_f32_e32 vcc, v15, v15
+; CHECK-NEXT:    v_cndmask_b32_e32 v1, v1, v2, vcc
+; CHECK-NEXT:    v_perm_b32 v3, v1, v0, s5
+; CHECK-NEXT:    v_bfe_u32 v0, v12, 16, 1
+; CHECK-NEXT:    v_add3_u32 v0, v0, v12, s4
+; CHECK-NEXT:    v_or_b32_e32 v1, 0x400000, v12
+; CHECK-NEXT:    v_cmp_u_f32_e32 vcc, v12, v12
+; CHECK-NEXT:    v_cndmask_b32_e32 v0, v0, v1, vcc
+; CHECK-NEXT:    v_bfe_u32 v1, v13, 16, 1
+; CHECK-NEXT:    v_add3_u32 v1, v1, v13, s4
+; CHECK-NEXT:    v_or_b32_e32 v2, 0x400000, v13
+; CHECK-NEXT:    v_cmp_u_f32_e32 vcc, v13, v13
+; CHECK-NEXT:    v_cndmask_b32_e32 v1, v1, v2, vcc
+; CHECK-NEXT:    v_perm_b32 v2, v1, v0, s5
+; CHECK-NEXT:    v_bfe_u32 v0, v10, 16, 1
+; CHECK-NEXT:    v_add3_u32 v0, v0, v10, s4
+; CHECK-NEXT:    v_or_b32_e32 v1, 0x400000, v10
+; CHECK-NEXT:    v_cmp_u_f32_e32 vcc, v10, v10
+; CHECK-NEXT:    v_cndmask_b32_e32 v0, v0, v1, vcc
+; CHECK-NEXT:    v_bfe_u32 v1, v11, 16, 1
+; CHECK-NEXT:    v_add3_u32 v1, v1, v11, s4
+; CHECK-NEXT:    v_or_b32_e32 v10, 0x400000, v11
+; CHECK-NEXT:    v_cmp_u_f32_e32 vcc, v11, v11
+; CHECK-NEXT:    v_cndmask_b32_e32 v1, v1, v10, vcc
+; CHECK-NEXT:    v_perm_b32 v1, v1, v0, s5
+; CHECK-NEXT:    v_bfe_u32 v0, v8, 16, 1
+; CHECK-NEXT:    v_add3_u32 v0, v0, v8, s4
+; CHECK-NEXT:    v_or_b32_e32 v10, 0x400000, v8
+; CHECK-NEXT:    v_cmp_u_f32_e32 vcc, v8, v8
+; CHECK-NEXT:    v_bfe_u32 v8, v9, 16, 1
+; CHECK-NEXT:    v_cndmask_b32_e32 v0, v0, v10, vcc
+; CHECK-NEXT:    v_add3_u32 v8, v8, v9, s4
+; CHECK-NEXT:    v_or_b32_e32 v10, 0x400000, v9
+; CHECK-NEXT:    v_cmp_u_f32_e32 vcc, v9, v9
+; CHECK-NEXT:    v_cndmask_b32_e32 v8, v8, v10, vcc
+; CHECK-NEXT:    v_perm_b32 v0, v8, v0, s5
+; CHECK-NEXT:    global_store_dwordx4 v[16:17], v[0:3], off offset:16
+; CHECK-NEXT:    global_store_dwordx4 v[16:17], v[4:7], off
+; CHECK-NEXT:    s_waitcnt vmcnt(0)
+; CHECK-NEXT:    s_setpc_b64 s[30:31]
+entry:
+  %conv = fptrunc <16 x float> %num to <16 x bfloat>
+  store <16 x bfloat> %conv, ptr addrspace(1) %p, align 8
+  ret void
+}
+
+define void @v32(<32 x float> %num, ptr addrspace(1) %p) {
+; CHECK-LABEL: v32:
+; CHECK:       ; %bb.0: ; %entry
+; CHECK-NEXT:    s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
+; CHECK-NEXT:    buffer_load_dword v33, off, s[0:3], s32 offset:8
+; CHECK-NEXT:    buffer_load_dword v32, off, s[0:3], s32 offset:4
+; CHECK-NEXT:    buffer_load_dword v31, off, s[0:3], s32
+; CHECK-NEXT:    v_bfe_u32 v34, v6, 16, 1
+; CHECK-NEXT:    s_movk_i32 s4, 0x7fff
+; CHECK-NEXT:    v_add3_u32 v34, v34, v6, s4
+; CHECK-NEXT:    v_or_b32_e32 v35, 0x400000, v6
+; CHECK-NEXT:    v_cmp_u_f32_e32 vcc, v6, v6
+; CHECK-NEXT:    v_cndmask_b32_e32 v6, v34, v35, vcc
+; CHECK-NEXT:    v_bfe_u32 v34, v7, 16, 1
+; CHECK-NEXT:    v_add3_u32 v34, v34, v7, s4
+; CHECK-NEXT:    v_or_b32_e32 v35, 0x400000, v7
+; CHECK-NEXT:    v_cmp_u_f32_e32 vcc, v7, v7
+; CHECK-NEXT:    v_cndmask_b32_e32 v7, v34, v35, vcc
+; CHECK-NEXT:    s_mov_b32 s5, 0x7060302
+; CHECK-NEXT:    v_perm_b32 v7, v7, v6, s5
+; CHECK-NEXT:    v_bfe_u32 v6, v4, 16, 1
+; CHECK-NEXT:    v_add3_u32 v6, v6, v4, s4
+; CHECK-NEXT:    v_or_b32_e32 v34, 0x400000, v4
+; CHECK-NEXT:    v_cmp_u_f32_e32 vcc, v4, v4
+; CHECK-NEXT:    v_cndmask_b32_e32 v4, v6, v34, vcc
+; CHECK-NEXT:    v_bfe_u32 v6, v5, 16, 1
+; CHECK-NEXT:    v_add3_u32 v6, v6, v5, s4
+; CHECK-NEXT:    v_or_b32_e32 v34, 0x400000, v5
+; CHECK-NEXT:    v_cmp_u_f32_e32 vcc, v5, v5
+; CHECK-NEXT:    v_cndmask_b32_e32 v5, v6, v34, vcc
+; CHECK-NEXT:    v_perm_b32 v6, v5, v4, s5
+; CHECK-NEXT:    v_bfe_u32 v4, v2, 16, 1
+; CHECK-NEXT:    v_add3_u32 v4, v4, v2, s4
+; CHECK-NEXT:    v_or_b32_e32 v5, 0x400000, v2
+; CHECK-NEXT:    v_cmp_u_f32_e32 vcc, v2, v2
+; CHECK-NEXT:    v_cndmask_b32_e32 v2, v4, v5, vcc
+; CHECK-NEXT:    v_bfe_u32 v4, v3, 16, 1
+; CHECK-NEXT:    v_add3_u32 v4, v4, v3, s4
+; CHECK-NEXT:    v_or_b32_e32 v5, 0x400000, v3
+; CHECK-NEXT:    v_cmp_u_f32_e32 vcc, v3, v3
+; CHECK-NEXT:    v_cndmask_b32_e32 v3, v4, v5, vcc
+; CHECK-NEXT:    v_perm_b32 v5, v3, v2, s5
+; CHECK-NEXT:    v_bfe_u32 v2, v0, 16, 1
+; CHECK-NEXT:    v_add3_u32 v2, v2, v0, s4
+; CHECK-NEXT:    v_or_b32_e32 v3, 0x400000, v0
+; CHECK-NEXT:    v_cmp_u_f32_e32 vcc, v0, v0
+; CHECK-NEXT:    v_cndmask_b32_e32 v0, v2, v3, vcc
+; CHECK-NEXT:    v_bfe_u32 v2, v1, 16, 1
+; CHECK-NEXT:    v_add3_u32 v2, v2, v1, s4
+; CHECK-NEXT:    v_or_b32_e32 v3, 0x400000, v1
+; CHECK-NEXT:    v_cmp_u_f32_e32 vcc, v1, v1
+; CHECK-NEXT:    v_cndmask_b32_e32 v1, v2, v3, vcc
+; CHECK-NEXT:    v_perm_b32 v4, v1, v0, s5
+; CHECK-NEXT:    v_bfe_u32 v0, v14, 16, 1
+; CHECK-NEXT:    v_add3_u32 v0, v0, v14, s4
+; CHECK-NEXT:    v_or_b32_e32 v1, 0x400000, v14
+; CHECK-NEXT:    v_cmp_u_f32_e32 vcc, v14, v14
+; CHECK-NEXT:    v_cndmask_b32_e32 v0, v0, v1, vcc
+; CHECK-NEXT:    v_bfe_u32 v1, v15, 16, 1
+; CHECK-NEXT:    v_add3_u32 v1, v1, v15, s4
+; CHECK-NEXT:    v_or_b32_e32 v2, 0x400000, v15
+; CHECK-NEXT:    v_cmp_u_f32_e32 vcc, v15, v15
+; CHECK-NEXT:    v_cndmask_b32_e32 v1, v1, v2, vcc
+; CHECK-NEXT:    v_perm_b32 v3, v1, v0, s5
+; CHECK-NEXT:    v_bfe_u32 v0, v12, 16, 1
+; CHECK-NEXT:    v_add3_u32 v0, v0, v12, s4
+; CHECK-NEXT:    v_or_b32_e32 v1, 0x400000, v12
+; CHECK-NEXT:    v_cmp_u_f32_e32 vcc, v12, v12
+; CHECK-NEXT:    v_cndmask_b32_e32 v0, v0, v1, vcc
+; CHECK-NEXT:    v_bfe_u32 v1, v13, 16, 1
+; CHECK-NEXT:    v_add3_u32 v1, v1, v13, s4
+; CHECK-NEXT:    v_or_b32_e32 v2, 0x400000, v13
+; CHECK-NEXT:    v_cmp_u_f32_e32 vcc, v13, v13
+; CHECK-NEXT:    v_cndmask_b32_e32 v1, v1, v2, vcc
+; CHECK-NEXT:    v_perm_b32 v2, v1, v0, s5
+; CHECK-NEXT:    v_bfe_u32 v0, v10, 16, 1
+; CHECK-NEXT:    v_add3_u32 v0, v0, v10, s4
+; CHECK-NEXT:    v_or_b32_e32 v1, 0x400000, v10
+; CHECK-NEXT:    v_cmp_u_f32_e32 vcc, v10, v10
+; CHECK-NEXT:    v_cndmask_b32_e32 v0, v0, v1, vcc
+; CHECK-NEXT:    v_bfe_u32 v1, v11, 16, 1
+; CHECK-NEXT:    v_add3_u32 v1, v1, v11, s4
+; CHECK-NEXT:    v_or_b32_e32 v10, 0x400000, v11
+; CHECK-NEXT:    v_cmp_u_f32_e32 vcc, v11, v11
+; CHECK-NEXT:    v_cndmask_b32_e32 v1, v1, v10, vcc
+; CHECK-NEXT:    v_perm_b32 v1, v1, v0, s5
+; CHECK-NEXT:    v_bfe_u32 v0, v8, 16, 1
+; CHECK-NEXT:    v_add3_u32 v0, v0, v8, s4
+; CHECK-NEXT:    v_or_b32_e32 v10, 0x400000, v8
+; CHECK-NEXT:    v_cmp_u_f32_e32 vcc, v8, v8
+; CHECK-NEXT:    v_bfe_u32 v8, v9, 16, 1
+; CHECK-NEXT:    v_cndmask_b32_e32 v0, v0, v10, vcc
+; CHECK-NEXT:    v_add3_u32 v8, v8, v9, s4
+; CHECK-NEXT:    v_or_b32_e32 v10, 0x400000, v9
+; CHECK-NEXT:    v_cmp_u_f32_e32 vcc, v9, v9
+; CHECK-NEXT:    v_cndmask_b32_e32 v8, v8, v10, vcc
+; CHECK-NEXT:    v_perm_b32 v0, v8, v0, s5
+; CHECK-NEXT:    v_bfe_u32 v8, v22, 16, 1
+; CHECK-NEXT:    v_add3_u32 v8, v8, v22, s4
+; CHECK-NEXT:    v_or_b32_e32 v9, 0x400000, v22
+; CHECK-NEXT:    v_cmp_u_f32_e32 vcc, v22, v22
+; CHECK-NEXT:    v_cndmask_b32_e32 v8, v8, v9, vcc
+; CHECK-NEXT:    v_bfe_u32 v9, v23, 16, 1
+; CHECK-NEXT:    v_add3_u32 v9, v9, v23, s4
+; CHECK-NEXT:    v_or_b32_e32 v10, 0x400000, v23
+; CHECK-NEXT:    v_cmp_u_f32_e32 vcc, v23, v23
+; CHECK-NEXT:    v_cndmask_b32_e32 v9, v9, v10, vcc
+; CHECK-NEXT:    v_perm_b32 v11, v9, v8, s5
+; CHECK-NEXT:    v_bfe_u32 v8, v20, 16, 1
+; CHECK-NEXT:    v_add3_u32 v8, v8, v20, s4
+; CHECK-NEXT:    v_or_b32_e32 v9, 0x400000, v20
+; CHECK-NEXT:    v_cmp_u_f32_e32 vcc, v20, v20
+; CHECK-NEXT:    v_cndmask_b32_e32 v8, v8, v9, vcc
+; CHECK-NEXT:    v_bfe_u32 v9, v21, 16, 1
+; CHECK-NEXT:    v_add3_u32 v9, v9, v21, s4
+; CHECK-NEXT:    v_or_b32_e32 v10, 0x400000, v21
+; CHECK-NEXT:    v_cmp_u_f32_e32 vcc, v21, v21
+; CHECK-NEXT:    v_cndmask_b32_e32 v9, v9, v10, vcc
+; CHECK-NEXT:    v_perm_b32 v10, v9, v8, s5
+; CHECK-NEXT:    v_bfe_u32 v8, v18, 16, 1
+; CHECK-NEXT:    v_add3_u32 v8, v8, v18, s4
+; CHECK-NEXT:    v_or_b32_e32 v9, 0x400000, v18
+; CHECK-NEXT:    v_cmp_u_f32_e32 vcc, v18, v18
+; CHECK-NEXT:    v_cndmask_b32_e32 v8, v8, v9, vcc
+; CHECK-NEXT:    v_bfe_u32 v9, v19, 16, 1
+; CHECK-NEXT:    v_add3_u32 v9, v9, v19, s4
+; CHECK-NEXT:    v_or_b32_e32 v12, 0x400000, v19
+; CHECK-NEXT:    v_cmp_u_f32_e32 vcc, v19, v19
+; CHECK-NEXT:    v_cndmask_b32_e32 v9, v9, v12, vcc
+; CHECK-NEXT:    v_perm_b32 v9, v9, v8, s5
+; CHECK-NEXT:    v_bfe_u32 v8, v16, 16, 1
+; CHECK-NEXT:    v_add3_u32 v8, v8, v16, s4
+; CHECK-NEXT:    v_or_b32_e32 v12, 0x400000, v16
+; CHECK-NEXT:    v_cmp_u_f32_e32 vcc, v16, v16
+; CHECK-NEXT:    v_cndmask_b32_e32 v8, v8, v12, vcc
+; CHECK-NEXT:    v_bfe_u32 v12, v17, 16, 1
+; CHECK-NEXT:    v_add3_u32 v12, v12, v17, s4
+; CHECK-NEXT:    v_or_b32_e32 v13, 0x400000, v17
+; CHECK-NEXT:    v_cmp_u_f32_e32 vcc, v17, v17
+; CHECK-NEXT:    v_cndmask_b32_e32 v12, v12, v13, vcc
+; CHECK-NEXT:    v_perm_b32 v8, v12, v8, s5
+; CHECK-NEXT:    v_bfe_u32 v12, v30, 16, 1
+; CHECK-NEXT:    v_add3_u32 v12, v12, v30, s4
+; CHECK-NEXT:    v_or_b32_e32 v13, 0x400000, v30
+; CHECK-NEXT:    v_cmp_u_f32_e32 vcc, v30, v30
+; CHECK-NEXT:    v_cndmask_b32_e32 v12, v12, v13, vcc
+; CHECK-NEXT:    s_waitcnt vmcnt(0)
+; CHECK-NEXT:    v_bfe_u32 v13, v31, 16, 1
+; CHECK-NEXT:    v_add3_u32 v13, v13, v31, s4
+; CHECK-NEXT:    v_or_b32_e32 v14, 0x400000, v31
+; CHECK-NEXT:    v_cmp_u_f32_e32 vcc, v31, v31
+; CHECK-NEXT:    v_cndmask_b32_e32 v13, v13, v14, vcc
+; CHECK-NEXT:    v_perm_b32 v15, v13, v12, s5
+; CHECK-NEXT:    v_bfe_u32 v12, v28, 16, 1
+; CHECK-NEXT:    v_add3_u32 v12, v12, v28, s4
+; CHECK-NEXT:    v_or_b32_e32 v13, 0x400000, v28
+; CHECK-NEXT:    v_cmp_u_f32_e32 vcc, v28, v28
+; CHECK-NEXT:    v_cndmask_b32_e32 v12, v12, v13, vcc
+; CHECK-NEXT:    v_bfe_u32 v13, v29, 16, 1
+; CHECK-NEXT:    v_add3_u32 v13, v13, v29, s4
+; CHECK-NEXT:    v_or_b32_e32 v14, 0x400000, v29
+; CHECK-NEXT:    v_cmp_u_f32_e32 vcc, v29, v29
+; CHECK-NEXT:    v_cndmask_b32_e32 v13, v13, v14, vcc
+; CHECK-NEXT:    v_perm_b32 v14, v13, v12, s5
+; CHECK-NEXT:    v_bfe_u32 v12, v26, 16, 1
+; CHECK-NEXT:    v_add3_u32 v12, v12, v26, s4
+; CHECK-NEXT:    v_or_b32_e32 v13, 0x400000, v26
+; CHECK-NEXT:    v_cmp_u_f32_e32 vcc, v26, v26
+; CHECK-NEXT:    v_cndmask_b32_e32 v12, v12, v13, vcc
+; CHECK-NEXT:    v_bfe_u32 v13, v27, 16, 1
+; CHECK-NEXT:    v_add3_u32 v13, v13, v27, s4
+; CHECK-NEXT:    v_or_b32_e32 v16, 0x400000, v27
+; CHECK-NEXT:    v_cmp_u_f32_e32 vcc, v27, v27
+; CHECK-NEXT:    v_cndmask_b32_e32 v13, v13, v16, vcc
+; CHECK-NEXT:    v_perm_b32 v13, v13, v12, s5
+; CHECK-NEXT:    v_bfe_u32 v12, v24, 16, 1
+; CHECK-NEXT:    v_add3_u32 v12, v12, v24, s4
+; CHECK-NEXT:    v_or_b32_e32 v16, 0x400000, v24
+; CHECK-NEXT:    v_cmp_u_f32_e32 vcc, v24, v24
+; CHECK-NEXT:    v_cndmask_b32_e32 v12, v12, v16, vcc
+; CHECK-NEXT:    v_bfe_u32 v16, v25, 16, 1
+; CHECK-NEXT:    v_add3_u32 v16, v16, v25, s4
+; CHECK-NEXT:    v_or_b32_e32 v17, 0x400000, v25
+; CHECK-NEXT:    v_cmp_u_f32_e32 vcc, v25, v25
+; CHECK-NEXT:    v_cndmask_b32_e32 v16, v16, v17, vcc
+; CHECK-NEXT:    v_perm_b32 v12, v16, v12, s5
+; CHECK-NEXT:    global_store_dwordx4 v[32:33], v[12:15], off offset:48
+; CHECK-NEXT:    global_store_dwordx4 v[32:33], v[8:11], off offset:32
+; CHECK-NEXT:    global_store_dwordx4 v[32:33], v[0:3], off offset:16
+; CHECK-NEXT:    global_store_dwordx4 v[32:33], v[4:7], off
+; CHECK-NEXT:    s_waitcnt vmcnt(0)
+; CHECK-NEXT:    s_setpc_b64 s[30:31]
+entry:
+  %conv = fptrunc <32 x float> %num to <32 x bfloat>
+  store <32 x bfloat> %conv, ptr addrspace(1) %p, align 8
+  ret void
+}
diff --git a/llvm/test/CodeGen/AMDGPU/global_atomics.ll b/llvm/test/CodeGen/AMDGPU/global_atomics.ll
index 674d7a3c5c9b..dac3a3db7b45 100644
--- a/llvm/test/CodeGen/AMDGPU/global_atomics.ll
+++ b/llvm/test/CodeGen/AMDGPU/global_atomics.ll
@@ -6216,6 +6216,81 @@ entry:
   ret void
 }
 
+define amdgpu_kernel void @atomic_store_bf16_offset(bfloat %in, ptr addrspace(1) %out) {
+; SI-LABEL: atomic_store_bf16_offset:
+; SI:       ; %bb.0:
+; SI-NEXT:    s_load_dword s4, s[0:1], 0x9
+; SI-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0xb
+; SI-NEXT:    s_mov_b32 s3, 0xf000
+; SI-NEXT:    s_mov_b32 s2, -1
+; SI-NEXT:    s_waitcnt lgkmcnt(0)
+; SI-NEXT:    v_mov_b32_e32 v0, s4
+; SI-NEXT:    buffer_store_short v0, off, s[0:3], 0 offset:16
+; SI-NEXT:    s_endpgm
+;
+; VI-LABEL: atomic_store_bf16_offset:
+; VI:       ; %bb.0:
+; VI-NEXT:    s_load_dwordx2 s[2:3], s[0:1], 0x2c
+; VI-NEXT:    s_load_dword s4, s[0:1], 0x24
+; VI-NEXT:    s_waitcnt lgkmcnt(0)
+; VI-NEXT:    s_add_u32 s0, s2, 16
+; VI-NEXT:    s_addc_u32 s1, s3, 0
+; VI-NEXT:    v_mov_b32_e32 v0, s0
+; VI-NEXT:    v_mov_b32_e32 v1, s1
+; VI-NEXT:    v_mov_b32_e32 v2, s4
+; VI-NEXT:    flat_store_short v[0:1], v2
+; VI-NEXT:    s_endpgm
+;
+; GFX9-LABEL: atomic_store_bf16_offset:
+; GFX9:       ; %bb.0:
+; GFX9-NEXT:    s_load_dword s4, s[0:1], 0x24
+; GFX9-NEXT:    s_load_dwordx2 s[2:3], s[0:1], 0x2c
+; GFX9-NEXT:    v_mov_b32_e32 v0, 0
+; GFX9-NEXT:    s_waitcnt lgkmcnt(0)
+; GFX9-NEXT:    v_mov_b32_e32 v1, s4
+; GFX9-NEXT:    global_store_short v0, v1, s[2:3] offset:16
+; GFX9-NEXT:    s_endpgm
+  %gep = getelementptr bfloat, ptr addrspace(1) %out, i64 8
+  store atomic bfloat %in, ptr addrspace(1) %gep seq_cst, align 2
+  ret void
+}
+
+define amdgpu_kernel void @atomic_store_bf16(bfloat %in, ptr addrspace(1) %out) {
+; SI-LABEL: atomic_store_bf16:
+; SI:       ; %bb.0:
+; SI-NEXT:    s_load_dword s4, s[0:1], 0x9
+; SI-NEXT:    s_load_dwordx2 s[0:1], s[0:1], 0xb
+; SI-NEXT:    s_mov_b32 s3, 0xf000
+; SI-NEXT:    s_mov_b32 s2, -1
+; SI-NEXT:    s_waitcnt lgkmcnt(0)
+; SI-NEXT:    v_mov_b32_e32 v0, s4
+; SI-NEXT:    buffer_store_short v0, off, s[0:3], 0
+; SI-NEXT:    s_endpgm
+;
+; VI-LABEL: atomic_store_bf16:
+; VI:       ; %bb.0:
+; VI-NEXT:    s_load_dwordx2 s[2:3], s[0:1], 0x2c
+; VI-NEXT:    s_load_dword s0, s[0:1], 0x24
+; VI-NEXT:    s_waitcnt lgkmcnt(0)
+; VI-NEXT:    v_mov_b32_e32 v0, s2
+; VI-NEXT:    v_mov_b32_e32 v1, s3
+; VI-NEXT:    v_mov_b32_e32 v2, s0
+; VI-NEXT:    flat_store_short v[0:1], v2
+; VI-NEXT:    s_endpgm
+;
+; GFX9-LABEL: atomic_store_bf16:
+; GFX9:       ; %bb.0:
+; GFX9-NEXT:    s_load_dword s4, s[0:1], 0x24
+; GFX9-NEXT:    s_load_dwordx2 s[2:3], s[0:1], 0x2c
+; GFX9-NEXT:    v_mov_b32_e32 v0, 0
+; GFX9-NEXT:    s_waitcnt lgkmcnt(0)
+; GFX9-NEXT:    v_mov_b32_e32 v1, s4
+; GFX9-NEXT:    global_store_short v0, v1, s[2:3]
+; GFX9-NEXT:    s_endpgm
+  store atomic bfloat %in, ptr addrspace(1) %out seq_cst, align 2
+  ret void
+}
+
 define amdgpu_kernel void @atomic_inc_i32_offset(ptr addrspace(1) %out, i32 %in) {
 ; SI-LABEL: atomic_inc_i32_offset:
 ; SI:       ; %bb.0: ; %entry
@@ -6963,3 +7038,207 @@ entry:
   store i32 %val, ptr addrspace(1) %out2
   ret void
 }
+
+define amdgpu_kernel void @atomic_load_f16_offset(ptr addrspace(1) %in, ptr addrspace(1) %out) {
+; SI-LABEL: atomic_load_f16_offset:
+; SI:       ; %bb.0:
+; SI-NEXT:    s_load_dwordx4 s[0:3], s[0:1], 0x9
+; SI-NEXT:    s_mov_b32 s7, 0xf000
+; SI-NEXT:    s_mov_b32 s6, -1
+; SI-NEXT:    s_waitcnt lgkmcnt(0)
+; SI-NEXT:    s_mov_b32 s4, s2
+; SI-NEXT:    s_mov_b32 s5, s3
+; SI-NEXT:    s_mov_b32 s2, s6
+; SI-NEXT:    s_mov_b32 s3, s7
+; SI-NEXT:    buffer_load_ushort v0, off, s[0:3], 0 offset:16 glc
+; SI-NEXT:    s_waitcnt vmcnt(0)
+; SI-NEXT:    buffer_wbinvl1
+; SI-NEXT:    buffer_store_short v0, off, s[4:7], 0
+; SI-NEXT:    s_endpgm
+;
+; VI-LABEL: atomic_load_f16_offset:
+; VI:       ; %bb.0:
+; VI-NEXT:    s_load_dwordx4 s[0:3], s[0:1], 0x24
+; VI-NEXT:    s_mov_b32 s7, 0xf000
+; VI-NEXT:    s_mov_b32 s6, -1
+; VI-NEXT:    s_waitcnt lgkmcnt(0)
+; VI-NEXT:    s_mov_b32 s4, s2
+; VI-NEXT:    s_mov_b32 s5, s3
+; VI-NEXT:    s_mov_b32 s2, s6
+; VI-NEXT:    s_mov_b32 s3, s7
+; VI-NEXT:    buffer_load_ushort v0, off, s[0:3], 0 offset:16 glc
+; VI-NEXT:    s_waitcnt vmcnt(0)
+; VI-NEXT:    buffer_wbinvl1_vol
+; VI-NEXT:    buffer_store_short v0, off, s[4:7], 0
+; VI-NEXT:    s_endpgm
+;
+; GFX9-LABEL: atomic_load_f16_offset:
+; GFX9:       ; %bb.0:
+; GFX9-NEXT:    s_load_dwordx4 s[0:3], s[0:1], 0x24
+; GFX9-NEXT:    v_mov_b32_e32 v0, 0
+; GFX9-NEXT:    s_waitcnt lgkmcnt(0)
+; GFX9-NEXT:    global_load_ushort v1, v0, s[0:1] offset:16 glc
+; GFX9-NEXT:    s_waitcnt vmcnt(0)
+; GFX9-NEXT:    buffer_wbinvl1_vol
+; GFX9-NEXT:    global_store_short v0, v1, s[2:3]
+; GFX9-NEXT:    s_endpgm
+  %gep = getelementptr half, ptr addrspace(1) %in, i64 8
+  %val = load atomic half, ptr addrspace(1) %gep  seq_cst, align 2
+  store half %val, ptr addrspace(1) %out
+  ret void
+}
+
+define amdgpu_kernel void @atomic_load_f16_negoffset(ptr addrspace(1) %in, ptr addrspace(1) %out) {
+; SI-LABEL: atomic_load_f16_negoffset:
+; SI:       ; %bb.0:
+; SI-NEXT:    s_load_dwordx4 s[0:3], s[0:1], 0x9
+; SI-NEXT:    s_mov_b32 s7, 0xf000
+; SI-NEXT:    s_waitcnt lgkmcnt(0)
+; SI-NEXT:    s_mov_b32 s4, s2
+; SI-NEXT:    s_mov_b32 s5, s3
+; SI-NEXT:    s_mov_b32 s2, 0
+; SI-NEXT:    s_mov_b32 s3, s7
+; SI-NEXT:    v_mov_b32_e32 v0, 0xfffffe00
+; SI-NEXT:    v_mov_b32_e32 v1, -1
+; SI-NEXT:    buffer_load_ushort v0, v[0:1], s[0:3], 0 addr64 glc
+; SI-NEXT:    s_waitcnt vmcnt(0)
+; SI-NEXT:    buffer_wbinvl1
+; SI-NEXT:    s_mov_b32 s6, -1
+; SI-NEXT:    buffer_store_short v0, off, s[4:7], 0
+; SI-NEXT:    s_endpgm
+;
+; VI-LABEL: atomic_load_f16_negoffset:
+; VI:       ; %bb.0:
+; VI-NEXT:    s_load_dwordx4 s[0:3], s[0:1], 0x24
+; VI-NEXT:    s_mov_b32 s7, 0xf000
+; VI-NEXT:    s_mov_b32 s6, -1
+; VI-NEXT:    s_waitcnt lgkmcnt(0)
+; VI-NEXT:    s_add_u32 s0, s0, 0xfffffe00
+; VI-NEXT:    s_addc_u32 s1, s1, -1
+; VI-NEXT:    v_mov_b32_e32 v0, s0
+; VI-NEXT:    v_mov_b32_e32 v1, s1
+; VI-NEXT:    flat_load_ushort v0, v[0:1] glc
+; VI-NEXT:    s_waitcnt vmcnt(0)
+; VI-NEXT:    buffer_wbinvl1_vol
+; VI-NEXT:    s_mov_b32 s4, s2
+; VI-NEXT:    s_mov_b32 s5, s3
+; VI-NEXT:    buffer_store_short v0, off, s[4:7], 0
+; VI-NEXT:    s_endpgm
+;
+; GFX9-LABEL: atomic_load_f16_negoffset:
+; GFX9:       ; %bb.0:
+; GFX9-NEXT:    s_load_dwordx4 s[0:3], s[0:1], 0x24
+; GFX9-NEXT:    v_mov_b32_e32 v0, 0
+; GFX9-NEXT:    s_waitcnt lgkmcnt(0)
+; GFX9-NEXT:    global_load_ushort v1, v0, s[0:1] offset:-512 glc
+; GFX9-NEXT:    s_waitcnt vmcnt(0)
+; GFX9-NEXT:    buffer_wbinvl1_vol
+; GFX9-NEXT:    global_store_short v0, v1, s[2:3]
+; GFX9-NEXT:    s_endpgm
+  %gep = getelementptr half, ptr addrspace(1) %in, i64 -256
+  %val = load atomic half, ptr addrspace(1) %gep  seq_cst, align 2
+  store half %val, ptr addrspace(1) %out
+  ret void
+}
+
+define amdgpu_kernel void @atomic_load_bf16_offset(ptr addrspace(1) %in, ptr addrspace(1) %out) {
+; SI-LABEL: atomic_load_bf16_offset:
+; SI:       ; %bb.0:
+; SI-NEXT:    s_load_dwordx4 s[0:3], s[0:1], 0x9
+; SI-NEXT:    s_mov_b32 s7, 0xf000
+; SI-NEXT:    s_mov_b32 s6, -1
+; SI-NEXT:    s_waitcnt lgkmcnt(0)
+; SI-NEXT:    s_mov_b32 s4, s2
+; SI-NEXT:    s_mov_b32 s5, s3
+; SI-NEXT:    s_mov_b32 s2, s6
+; SI-NEXT:    s_mov_b32 s3, s7
+; SI-NEXT:    buffer_load_ushort v0, off, s[0:3], 0 offset:16 glc
+; SI-NEXT:    s_waitcnt vmcnt(0)
+; SI-NEXT:    buffer_wbinvl1
+; SI-NEXT:    buffer_store_short v0, off, s[4:7], 0
+; SI-NEXT:    s_endpgm
+;
+; VI-LABEL: atomic_load_bf16_offset:
+; VI:       ; %bb.0:
+; VI-NEXT:    s_load_dwordx4 s[0:3], s[0:1], 0x24
+; VI-NEXT:    s_mov_b32 s7, 0xf000
+; VI-NEXT:    s_mov_b32 s6, -1
+; VI-NEXT:    s_waitcnt lgkmcnt(0)
+; VI-NEXT:    s_mov_b32 s4, s2
+; VI-NEXT:    s_mov_b32 s5, s3
+; VI-NEXT:    s_mov_b32 s2, s6
+; VI-NEXT:    s_mov_b32 s3, s7
+; VI-NEXT:    buffer_load_ushort v0, off, s[0:3], 0 offset:16 glc
+; VI-NEXT:    s_waitcnt vmcnt(0)
+; VI-NEXT:    buffer_wbinvl1_vol
+; VI-NEXT:    buffer_store_short v0, off, s[4:7], 0
+; VI-NEXT:    s_endpgm
+;
+; GFX9-LABEL: atomic_load_bf16_offset:
+; GFX9:       ; %bb.0:
+; GFX9-NEXT:    s_load_dwordx4 s[0:3], s[0:1], 0x24
+; GFX9-NEXT:    v_mov_b32_e32 v0, 0
+; GFX9-NEXT:    s_waitcnt lgkmcnt(0)
+; GFX9-NEXT:    global_load_ushort v1, v0, s[0:1] offset:16 glc
+; GFX9-NEXT:    s_waitcnt vmcnt(0)
+; GFX9-NEXT:    buffer_wbinvl1_vol
+; GFX9-NEXT:    global_store_short v0, v1, s[2:3]
+; GFX9-NEXT:    s_endpgm
+  %gep = getelementptr bfloat, ptr addrspace(1) %in, i64 8
+  %val = load atomic bfloat, ptr addrspace(1) %gep  seq_cst, align 2
+  store bfloat %val, ptr addrspace(1) %out
+  ret void
+}
+
+define amdgpu_kernel void @atomic_load_bf16_negoffset(ptr addrspace(1) %in, ptr addrspace(1) %out) {
+; SI-LABEL: atomic_load_bf16_negoffset:
+; SI:       ; %bb.0:
+; SI-NEXT:    s_load_dwordx4 s[0:3], s[0:1], 0x9
+; SI-NEXT:    s_mov_b32 s7, 0xf000
+; SI-NEXT:    s_waitcnt lgkmcnt(0)
+; SI-NEXT:    s_mov_b32 s4, s2
+; SI-NEXT:    s_mov_b32 s5, s3
+; SI-NEXT:    s_mov_b32 s2, 0
+; SI-NEXT:    s_mov_b32 s3, s7
+; SI-NEXT:    v_mov_b32_e32 v0, 0xfffffe00
+; SI-NEXT:    v_mov_b32_e32 v1, -1
+; SI-NEXT:    buffer_load_ushort v0, v[0:1], s[0:3], 0 addr64 glc
+; SI-NEXT:    s_waitcnt vmcnt(0)
+; SI-NEXT:    buffer_wbinvl1
+; SI-NEXT:    s_mov_b32 s6, -1
+; SI-NEXT:    buffer_store_short v0, off, s[4:7], 0
+; SI-NEXT:    s_endpgm
+;
+; VI-LABEL: atomic_load_bf16_negoffset:
+; VI:       ; %bb.0:
+; VI-NEXT:    s_load_dwordx4 s[0:3], s[0:1], 0x24
+; VI-NEXT:    s_mov_b32 s7, 0xf000
+; VI-NEXT:    s_mov_b32 s6, -1
+; VI-NEXT:    s_waitcnt lgkmcnt(0)
+; VI-NEXT:    s_add_u32 s0, s0, 0xfffffe00
+; VI-NEXT:    s_addc_u32 s1, s1, -1
+; VI-NEXT:    v_mov_b32_e32 v0, s0
+; VI-NEXT:    v_mov_b32_e32 v1, s1
+; VI-NEXT:    flat_load_ushort v0, v[0:1] glc
+; VI-NEXT:    s_waitcnt vmcnt(0)
+; VI-NEXT:    buffer_wbinvl1_vol
+; VI-NEXT:    s_mov_b32 s4, s2
+; VI-NEXT:    s_mov_b32 s5, s3
+; VI-NEXT:    buffer_store_short v0, off, s[4:7], 0
+; VI-NEXT:    s_endpgm
+;
+; GFX9-LABEL: atomic_load_bf16_negoffset:
+; GFX9:       ; %bb.0:
+; GFX9-NEXT:    s_load_dwordx4 s[0:3], s[0:1], 0x24
+; GFX9-NEXT:    v_mov_b32_e32 v0, 0
+; GFX9-NEXT:    s_waitcnt lgkmcnt(0)
+; GFX9-NEXT:    global_load_ushort v1, v0, s[0:1] offset:-512 glc
+; GFX9-NEXT:    s_waitcnt vmcnt(0)
+; GFX9-NEXT:    buffer_wbinvl1_vol
+; GFX9-NEXT:    global_store_short v0, v1, s[2:3]
+; GFX9-NEXT:    s_endpgm
+  %gep = getelementptr bfloat, ptr addrspace(1) %in, i64 -256
+  %val = load atomic bfloat, ptr addrspace(1) %gep  seq_cst, align 2
+  store bfloat %val, ptr addrspace(1) %out
+  ret void
+}
diff --git a/llvm/test/CodeGen/AMDGPU/global_atomics_i64.ll b/llvm/test/CodeGen/AMDGPU/global_atomics_i64.ll
index 08b068f3588c..f5dbaaff9cf8 100644
--- a/llvm/test/CodeGen/AMDGPU/global_atomics_i64.ll
+++ b/llvm/test/CodeGen/AMDGPU/global_atomics_i64.ll
@@ -4570,7 +4570,7 @@ define amdgpu_kernel void @atomic_xchg_f64_offset(ptr addrspace(1) %out, double
 ; GFX12-NEXT:    v_mov_b32_e32 v0, s2
 ; GFX12-NEXT:    global_atomic_swap_b64 v2, v[0:1], s[0:1] offset:32
 ; GFX12-NEXT:    s_wait_storecnt 0x0
-; GFX12-NEXT:    global_inv scope:SCOPE_SYS
+; GFX12-NEXT:    global_inv scope:SCOPE_DEV
 ; GFX12-NEXT:    s_endpgm
 entry:
   %gep = getelementptr double, ptr addrspace(1) %out, i64 4
@@ -4625,7 +4625,7 @@ define amdgpu_kernel void @atomic_xchg_pointer_offset(ptr addrspace(1) %out, ptr
 ; GFX12-NEXT:    v_mov_b32_e32 v0, s2
 ; GFX12-NEXT:    global_atomic_swap_b64 v2, v[0:1], s[0:1] offset:32
 ; GFX12-NEXT:    s_wait_storecnt 0x0
-; GFX12-NEXT:    global_inv scope:SCOPE_SYS
+; GFX12-NEXT:    global_inv scope:SCOPE_DEV
 ; GFX12-NEXT:    s_endpgm
 entry:
   %gep = getelementptr ptr, ptr addrspace(1) %out, i64 4
diff --git a/llvm/test/CodeGen/AMDGPU/hard-clauses-img-gfx10.mir b/llvm/test/CodeGen/AMDGPU/hard-clauses-img-gfx10.mir
new file mode 100644
index 000000000000..50eea4aebd5e
--- /dev/null
+++ b/llvm/test/CodeGen/AMDGPU/hard-clauses-img-gfx10.mir
@@ -0,0 +1,34 @@
+# NOTE: Assertions have been autogenerated by utils/update_mir_test_checks.py
+# RUN: llc -mtriple=amdgcn -mcpu=gfx1010 -verify-machineinstrs -run-pass si-insert-hard-clauses %s -o - | FileCheck %s
+
+---
+name: mimg_nsa
+tracksRegLiveness: true
+body: |
+  bb.0:
+    liveins: $sgpr0_sgpr1_sgpr2_sgpr3_sgpr4_sgpr5_sgpr6_sgpr7, $sgpr8_sgpr9_sgpr10_sgpr11, $vgpr0, $vgpr1, $vgpr2, $vgpr3, $vgpr4, $vgpr5, $vgpr6, $vgpr7, $vgpr8
+    ; CHECK-LABEL: name: mimg_nsa
+    ; CHECK: liveins: $sgpr0_sgpr1_sgpr2_sgpr3_sgpr4_sgpr5_sgpr6_sgpr7, $sgpr8_sgpr9_sgpr10_sgpr11, $vgpr0, $vgpr1, $vgpr2, $vgpr3, $vgpr4, $vgpr5, $vgpr6, $vgpr7, $vgpr8
+    ; CHECK-NEXT: {{  $}}
+    ; CHECK-NEXT: $vgpr10_vgpr11_vgpr12_vgpr13 = IMAGE_SAMPLE_D_V4_V9_nsa_gfx10 $vgpr3, $vgpr8, $vgpr7, $vgpr5, $vgpr4, $vgpr6, $vgpr0, $vgpr2, $vgpr2, $sgpr0_sgpr1_sgpr2_sgpr3_sgpr4_sgpr5_sgpr6_sgpr7, $sgpr8_sgpr9_sgpr10_sgpr11, 15, 2, 0, 0, 0, 0, 0, 0, 0, implicit $exec :: (load (s128))
+    ; CHECK-NEXT: $vgpr20_vgpr21_vgpr22_vgpr23 = IMAGE_SAMPLE_D_V4_V9_nsa_gfx10 $vgpr3, $vgpr8, $vgpr7, $vgpr5, $vgpr4, $vgpr6, $vgpr0, $vgpr2, $vgpr2, $sgpr0_sgpr1_sgpr2_sgpr3_sgpr4_sgpr5_sgpr6_sgpr7, $sgpr8_sgpr9_sgpr10_sgpr11, 15, 2, 0, 0, 0, 0, 0, 0, 0, implicit $exec :: (load (s128))
+    $vgpr10_vgpr11_vgpr12_vgpr13 = IMAGE_SAMPLE_D_V4_V9_nsa_gfx10 $vgpr3, $vgpr8, $vgpr7, $vgpr5, $vgpr4, $vgpr6, $vgpr0, $vgpr2, $vgpr2, $sgpr0_sgpr1_sgpr2_sgpr3_sgpr4_sgpr5_sgpr6_sgpr7, $sgpr8_sgpr9_sgpr10_sgpr11, 15, 2, 0, 0, 0, 0, 0, 0, 0, implicit $exec  :: (load (s128))
+    $vgpr20_vgpr21_vgpr22_vgpr23 = IMAGE_SAMPLE_D_V4_V9_nsa_gfx10 $vgpr3, $vgpr8, $vgpr7, $vgpr5, $vgpr4, $vgpr6, $vgpr0, $vgpr2, $vgpr2, $sgpr0_sgpr1_sgpr2_sgpr3_sgpr4_sgpr5_sgpr6_sgpr7, $sgpr8_sgpr9_sgpr10_sgpr11, 15, 2, 0, 0, 0, 0, 0, 0, 0, implicit $exec  :: (load (s128))
+...
+
+---
+name: mimg_nsa_mixed
+tracksRegLiveness: true
+body: |
+  bb.0:
+    liveins: $sgpr0_sgpr1_sgpr2_sgpr3_sgpr4_sgpr5_sgpr6_sgpr7, $sgpr8_sgpr9_sgpr10_sgpr11, $vgpr0, $vgpr1, $vgpr2, $vgpr3, $vgpr4, $vgpr5, $vgpr6, $vgpr7, $vgpr8
+    ; CHECK-LABEL: name: mimg_nsa_mixed
+    ; CHECK: liveins: $sgpr0_sgpr1_sgpr2_sgpr3_sgpr4_sgpr5_sgpr6_sgpr7, $sgpr8_sgpr9_sgpr10_sgpr11, $vgpr0, $vgpr1, $vgpr2, $vgpr3, $vgpr4, $vgpr5, $vgpr6, $vgpr7, $vgpr8
+    ; CHECK-NEXT: {{  $}}
+    ; CHECK-NEXT: $vgpr10_vgpr11_vgpr12_vgpr13 = IMAGE_SAMPLE_D_V4_V9_nsa_gfx10 $vgpr3, $vgpr8, $vgpr7, $vgpr5, $vgpr4, $vgpr6, $vgpr0, $vgpr2, $vgpr2, $sgpr0_sgpr1_sgpr2_sgpr3_sgpr4_sgpr5_sgpr6_sgpr7, $sgpr8_sgpr9_sgpr10_sgpr11, 15, 2, 0, 0, 0, 0, 0, 0, 0, implicit $exec :: (load (s128))
+    ; CHECK-NEXT: $vgpr14 = IMAGE_SAMPLE_LZ_V1_V2_gfx10 $vgpr5_vgpr6, $sgpr0_sgpr1_sgpr2_sgpr3_sgpr4_sgpr5_sgpr6_sgpr7, $sgpr8_sgpr9_sgpr10_sgpr11, 1, 1, 0, 0, 0, 0, 0, 0, 0, implicit $exec :: (dereferenceable load (s128), addrspace 7)
+    ; CHECK-NEXT: $vgpr20_vgpr21_vgpr22_vgpr23 = IMAGE_SAMPLE_D_V4_V9_nsa_gfx10 $vgpr3, $vgpr8, $vgpr7, $vgpr5, $vgpr4, $vgpr6, $vgpr0, $vgpr2, $vgpr2, $sgpr0_sgpr1_sgpr2_sgpr3_sgpr4_sgpr5_sgpr6_sgpr7, $sgpr8_sgpr9_sgpr10_sgpr11, 15, 2, 0, 0, 0, 0, 0, 0, 0, implicit $exec :: (load (s128))
+    $vgpr10_vgpr11_vgpr12_vgpr13 = IMAGE_SAMPLE_D_V4_V9_nsa_gfx10 $vgpr3, $vgpr8, $vgpr7, $vgpr5, $vgpr4, $vgpr6, $vgpr0, $vgpr2, $vgpr2, $sgpr0_sgpr1_sgpr2_sgpr3_sgpr4_sgpr5_sgpr6_sgpr7, $sgpr8_sgpr9_sgpr10_sgpr11, 15, 2, 0, 0, 0, 0, 0, 0, 0, implicit $exec  :: (load (s128))
+    $vgpr14 = IMAGE_SAMPLE_LZ_V1_V2_gfx10 $vgpr5_vgpr6, $sgpr0_sgpr1_sgpr2_sgpr3_sgpr4_sgpr5_sgpr6_sgpr7, $sgpr8_sgpr9_sgpr10_sgpr11, 1, 1, 0, 0, 0, 0, 0, 0, 0, implicit $exec :: (dereferenceable load (s128), addrspace 7)
+    $vgpr20_vgpr21_vgpr22_vgpr23 = IMAGE_SAMPLE_D_V4_V9_nsa_gfx10 $vgpr3, $vgpr8, $vgpr7, $vgpr5, $vgpr4, $vgpr6, $vgpr0, $vgpr2, $vgpr2, $sgpr0_sgpr1_sgpr2_sgpr3_sgpr4_sgpr5_sgpr6_sgpr7, $sgpr8_sgpr9_sgpr10_sgpr11, 15, 2, 0, 0, 0, 0, 0, 0, 0, implicit $exec  :: (load (s128))
+...
diff --git a/llvm/test/CodeGen/AMDGPU/hard-clauses-img-gfx11.mir b/llvm/test/CodeGen/AMDGPU/hard-clauses-img-gfx11.mir
new file mode 100644
index 000000000000..b22de06e68a7
--- /dev/null
+++ b/llvm/test/CodeGen/AMDGPU/hard-clauses-img-gfx11.mir
@@ -0,0 +1,40 @@
+# NOTE: Assertions have been autogenerated by utils/update_mir_test_checks.py
+# RUN: llc -mtriple=amdgcn -mcpu=gfx1100 -verify-machineinstrs -run-pass si-insert-hard-clauses %s -o - | FileCheck %s
+
+---
+name: mimg_nsa
+tracksRegLiveness: true
+body: |
+  bb.0:
+    liveins: $sgpr0_sgpr1_sgpr2_sgpr3_sgpr4_sgpr5_sgpr6_sgpr7, $sgpr8_sgpr9_sgpr10_sgpr11, $vgpr0, $vgpr1, $vgpr2, $vgpr3, $vgpr4, $vgpr5, $vgpr6, $vgpr7, $vgpr8
+    ; CHECK-LABEL: name: mimg_nsa
+    ; CHECK: liveins: $sgpr0_sgpr1_sgpr2_sgpr3_sgpr4_sgpr5_sgpr6_sgpr7, $sgpr8_sgpr9_sgpr10_sgpr11, $vgpr0, $vgpr1, $vgpr2, $vgpr3, $vgpr4, $vgpr5, $vgpr6, $vgpr7, $vgpr8
+    ; CHECK-NEXT: {{  $}}
+    ; CHECK-NEXT: BUNDLE implicit-def $vgpr10_vgpr11_vgpr12, implicit-def $vgpr10, implicit-def $vgpr10_lo16, implicit-def $vgpr10_hi16, implicit-def $vgpr11, implicit-def $vgpr11_lo16, implicit-def $vgpr11_hi16, implicit-def $vgpr12, implicit-def $vgpr12_lo16, implicit-def $vgpr12_hi16, implicit-def $vgpr10_vgpr11, implicit-def $vgpr11_vgpr12, implicit-def $vgpr20_vgpr21_vgpr22, implicit-def $vgpr20, implicit-def $vgpr20_lo16, implicit-def $vgpr20_hi16, implicit-def $vgpr21, implicit-def $vgpr21_lo16, implicit-def $vgpr21_hi16, implicit-def $vgpr22, implicit-def $vgpr22_lo16, implicit-def $vgpr22_hi16, implicit-def $vgpr20_vgpr21, implicit-def $vgpr21_vgpr22, implicit $vgpr3, implicit $vgpr8, implicit $sgpr0_sgpr1_sgpr2_sgpr3_sgpr4_sgpr5_sgpr6_sgpr7, implicit $sgpr8_sgpr9_sgpr10_sgpr11, implicit $exec {
+    ; CHECK-NEXT:   S_CLAUSE 1
+    ; CHECK-NEXT:   $vgpr10_vgpr11_vgpr12 = IMAGE_SAMPLE_LZ_V3_V2_nsa_gfx11 $vgpr3, $vgpr8, $sgpr0_sgpr1_sgpr2_sgpr3_sgpr4_sgpr5_sgpr6_sgpr7, $sgpr8_sgpr9_sgpr10_sgpr11, 1, 1, -1, 0, 0, 0, 0, 0, 0, implicit $exec :: (load (s128))
+    ; CHECK-NEXT:   $vgpr20_vgpr21_vgpr22 = IMAGE_SAMPLE_LZ_V3_V2_nsa_gfx11 $vgpr3, $vgpr8, $sgpr0_sgpr1_sgpr2_sgpr3_sgpr4_sgpr5_sgpr6_sgpr7, $sgpr8_sgpr9_sgpr10_sgpr11, 14, 1, -1, 0, 0, 0, 0, 0, 0, implicit $exec :: (load (s128))
+    ; CHECK-NEXT: }
+    $vgpr10_vgpr11_vgpr12 = IMAGE_SAMPLE_LZ_V3_V2_nsa_gfx11 $vgpr3, $vgpr8, $sgpr0_sgpr1_sgpr2_sgpr3_sgpr4_sgpr5_sgpr6_sgpr7, $sgpr8_sgpr9_sgpr10_sgpr11, 1, 1, -1, 0, 0, 0, 0, 0, 0, implicit $exec  :: (load (s128))
+    $vgpr20_vgpr21_vgpr22 = IMAGE_SAMPLE_LZ_V3_V2_nsa_gfx11 $vgpr3, $vgpr8, $sgpr0_sgpr1_sgpr2_sgpr3_sgpr4_sgpr5_sgpr6_sgpr7, $sgpr8_sgpr9_sgpr10_sgpr11, 14, 1, -1, 0, 0, 0, 0, 0, 0, implicit $exec  :: (load (s128))
+...
+
+---
+name: mimg_nsa_mixed
+tracksRegLiveness: true
+body: |
+  bb.0:
+    liveins: $sgpr0_sgpr1_sgpr2_sgpr3_sgpr4_sgpr5_sgpr6_sgpr7, $sgpr8_sgpr9_sgpr10_sgpr11, $vgpr0, $vgpr1, $vgpr2, $vgpr3, $vgpr4, $vgpr5, $vgpr6, $vgpr7, $vgpr8
+    ; CHECK-LABEL: name: mimg_nsa_mixed
+    ; CHECK: liveins: $sgpr0_sgpr1_sgpr2_sgpr3_sgpr4_sgpr5_sgpr6_sgpr7, $sgpr8_sgpr9_sgpr10_sgpr11, $vgpr0, $vgpr1, $vgpr2, $vgpr3, $vgpr4, $vgpr5, $vgpr6, $vgpr7, $vgpr8
+    ; CHECK-NEXT: {{  $}}
+    ; CHECK-NEXT: BUNDLE implicit-def $vgpr10, implicit-def $vgpr10_lo16, implicit-def $vgpr10_hi16, implicit-def $vgpr14, implicit-def $vgpr14_lo16, implicit-def $vgpr14_hi16, implicit-def $vgpr20_vgpr21_vgpr22, implicit-def $vgpr20, implicit-def $vgpr20_lo16, implicit-def $vgpr20_hi16, implicit-def $vgpr21, implicit-def $vgpr21_lo16, implicit-def $vgpr21_hi16, implicit-def $vgpr22, implicit-def $vgpr22_lo16, implicit-def $vgpr22_hi16, implicit-def $vgpr20_vgpr21, implicit-def $vgpr21_vgpr22, implicit $vgpr3, implicit $vgpr8, implicit $sgpr0_sgpr1_sgpr2_sgpr3_sgpr4_sgpr5_sgpr6_sgpr7, implicit $sgpr8_sgpr9_sgpr10_sgpr11, implicit $exec, implicit $vgpr5_vgpr6 {
+    ; CHECK-NEXT:   S_CLAUSE 2
+    ; CHECK-NEXT:   $vgpr10 = IMAGE_SAMPLE_LZ_V1_V2_nsa_gfx11 $vgpr3, $vgpr8, $sgpr0_sgpr1_sgpr2_sgpr3_sgpr4_sgpr5_sgpr6_sgpr7, $sgpr8_sgpr9_sgpr10_sgpr11, 1, 1, -1, 0, 0, 0, 0, 0, 0, implicit $exec :: (load (s128))
+    ; CHECK-NEXT:   $vgpr14 = IMAGE_SAMPLE_LZ_V1_V2_gfx11 $vgpr5_vgpr6, $sgpr0_sgpr1_sgpr2_sgpr3_sgpr4_sgpr5_sgpr6_sgpr7, $sgpr8_sgpr9_sgpr10_sgpr11, 1, 1, 0, 0, 0, 0, 0, 0, 0, implicit $exec :: (dereferenceable load (s128), addrspace 7)
+    ; CHECK-NEXT:   $vgpr20_vgpr21_vgpr22 = IMAGE_SAMPLE_LZ_V3_V2_nsa_gfx11 $vgpr3, $vgpr8, $sgpr0_sgpr1_sgpr2_sgpr3_sgpr4_sgpr5_sgpr6_sgpr7, $sgpr8_sgpr9_sgpr10_sgpr11, 14, 1, -1, 0, 0, 0, 0, 0, 0, implicit $exec :: (load (s128))
+    ; CHECK-NEXT: }
+    $vgpr10 = IMAGE_SAMPLE_LZ_V1_V2_nsa_gfx11 $vgpr3, $vgpr8, $sgpr0_sgpr1_sgpr2_sgpr3_sgpr4_sgpr5_sgpr6_sgpr7, $sgpr8_sgpr9_sgpr10_sgpr11, 1, 1, -1, 0, 0, 0, 0, 0, 0, implicit $exec  :: (load (s128))
+    $vgpr14 = IMAGE_SAMPLE_LZ_V1_V2_gfx11 $vgpr5_vgpr6, $sgpr0_sgpr1_sgpr2_sgpr3_sgpr4_sgpr5_sgpr6_sgpr7, $sgpr8_sgpr9_sgpr10_sgpr11, 1, 1, 0, 0, 0, 0, 0, 0, 0, implicit $exec :: (dereferenceable load (s128), addrspace 7)
+    $vgpr20_vgpr21_vgpr22 = IMAGE_SAMPLE_LZ_V3_V2_nsa_gfx11 $vgpr3, $vgpr8, $sgpr0_sgpr1_sgpr2_sgpr3_sgpr4_sgpr5_sgpr6_sgpr7, $sgpr8_sgpr9_sgpr10_sgpr11, 14, 1, -1, 0, 0, 0, 0, 0, 0, implicit $exec  :: (load (s128))
+...
diff --git a/llvm/test/CodeGen/AMDGPU/hard-clauses-img-gfx12.mir b/llvm/test/CodeGen/AMDGPU/hard-clauses-img-gfx12.mir
new file mode 100644
index 000000000000..243a84562ab3
--- /dev/null
+++ b/llvm/test/CodeGen/AMDGPU/hard-clauses-img-gfx12.mir
@@ -0,0 +1,40 @@
+# NOTE: Assertions have been autogenerated by utils/update_mir_test_checks.py
+# RUN: llc -mtriple=amdgcn -mcpu=gfx1100 -verify-machineinstrs -run-pass si-insert-hard-clauses %s -o - | FileCheck %s
+
+---
+name: mimg
+tracksRegLiveness: true
+body: |
+  bb.0:
+    liveins: $sgpr0_sgpr1_sgpr2_sgpr3_sgpr4_sgpr5_sgpr6_sgpr7, $sgpr8_sgpr9_sgpr10_sgpr11, $vgpr0, $vgpr1, $vgpr2, $vgpr3, $vgpr4, $vgpr5, $vgpr6, $vgpr7, $vgpr8
+    ; CHECK-LABEL: name: mimg
+    ; CHECK: liveins: $sgpr0_sgpr1_sgpr2_sgpr3_sgpr4_sgpr5_sgpr6_sgpr7, $sgpr8_sgpr9_sgpr10_sgpr11, $vgpr0, $vgpr1, $vgpr2, $vgpr3, $vgpr4, $vgpr5, $vgpr6, $vgpr7, $vgpr8
+    ; CHECK-NEXT: {{  $}}
+    ; CHECK-NEXT: BUNDLE implicit-def $vgpr10_vgpr11_vgpr12, implicit-def $vgpr10, implicit-def $vgpr10_lo16, implicit-def $vgpr10_hi16, implicit-def $vgpr11, implicit-def $vgpr11_lo16, implicit-def $vgpr11_hi16, implicit-def $vgpr12, implicit-def $vgpr12_lo16, implicit-def $vgpr12_hi16, implicit-def $vgpr10_vgpr11, implicit-def $vgpr11_vgpr12, implicit-def $vgpr20_vgpr21_vgpr22, implicit-def $vgpr20, implicit-def $vgpr20_lo16, implicit-def $vgpr20_hi16, implicit-def $vgpr21, implicit-def $vgpr21_lo16, implicit-def $vgpr21_hi16, implicit-def $vgpr22, implicit-def $vgpr22_lo16, implicit-def $vgpr22_hi16, implicit-def $vgpr20_vgpr21, implicit-def $vgpr21_vgpr22, implicit $vgpr3, implicit $vgpr4, implicit $sgpr0_sgpr1_sgpr2_sgpr3_sgpr4_sgpr5_sgpr6_sgpr7, implicit $sgpr8_sgpr9_sgpr10_sgpr11, implicit $exec {
+    ; CHECK-NEXT:   S_CLAUSE 1
+    ; CHECK-NEXT:   $vgpr10_vgpr11_vgpr12 = IMAGE_SAMPLE_LZ_V3_V2_gfx12 $vgpr3, $vgpr4, $sgpr0_sgpr1_sgpr2_sgpr3_sgpr4_sgpr5_sgpr6_sgpr7, $sgpr8_sgpr9_sgpr10_sgpr11, 1, 1, -1, 0, 0, 0, 0, 0, 0, implicit $exec :: (load (s128))
+    ; CHECK-NEXT:   $vgpr20_vgpr21_vgpr22 = IMAGE_SAMPLE_LZ_V3_V2_gfx12 $vgpr3, $vgpr4, $sgpr0_sgpr1_sgpr2_sgpr3_sgpr4_sgpr5_sgpr6_sgpr7, $sgpr8_sgpr9_sgpr10_sgpr11, 14, 1, -1, 0, 0, 0, 0, 0, 0, implicit $exec :: (load (s128))
+    ; CHECK-NEXT: }
+    $vgpr10_vgpr11_vgpr12 = IMAGE_SAMPLE_LZ_V3_V2_gfx12 $vgpr3, $vgpr4, $sgpr0_sgpr1_sgpr2_sgpr3_sgpr4_sgpr5_sgpr6_sgpr7, $sgpr8_sgpr9_sgpr10_sgpr11, 1, 1, -1, 0, 0, 0, 0, 0, 0, implicit $exec  :: (load (s128))
+    $vgpr20_vgpr21_vgpr22 = IMAGE_SAMPLE_LZ_V3_V2_gfx12 $vgpr3, $vgpr4, $sgpr0_sgpr1_sgpr2_sgpr3_sgpr4_sgpr5_sgpr6_sgpr7, $sgpr8_sgpr9_sgpr10_sgpr11, 14, 1, -1, 0, 0, 0, 0, 0, 0, implicit $exec  :: (load (s128))
+...
+
+---
+name: mimg_mixed
+tracksRegLiveness: true
+body: |
+  bb.0:
+    liveins: $sgpr0_sgpr1_sgpr2_sgpr3_sgpr4_sgpr5_sgpr6_sgpr7, $sgpr8_sgpr9_sgpr10_sgpr11, $vgpr0, $vgpr1, $vgpr2, $vgpr3, $vgpr4, $vgpr5, $vgpr6, $vgpr7, $vgpr8
+    ; CHECK-LABEL: name: mimg_mixed
+    ; CHECK: liveins: $sgpr0_sgpr1_sgpr2_sgpr3_sgpr4_sgpr5_sgpr6_sgpr7, $sgpr8_sgpr9_sgpr10_sgpr11, $vgpr0, $vgpr1, $vgpr2, $vgpr3, $vgpr4, $vgpr5, $vgpr6, $vgpr7, $vgpr8
+    ; CHECK-NEXT: {{  $}}
+    ; CHECK-NEXT: BUNDLE implicit-def $vgpr10, implicit-def $vgpr10_lo16, implicit-def $vgpr10_hi16, implicit-def $vgpr14, implicit-def $vgpr14_lo16, implicit-def $vgpr14_hi16, implicit-def $vgpr20_vgpr21_vgpr22, implicit-def $vgpr20, implicit-def $vgpr20_lo16, implicit-def $vgpr20_hi16, implicit-def $vgpr21, implicit-def $vgpr21_lo16, implicit-def $vgpr21_hi16, implicit-def $vgpr22, implicit-def $vgpr22_lo16, implicit-def $vgpr22_hi16, implicit-def $vgpr20_vgpr21, implicit-def $vgpr21_vgpr22, implicit $vgpr3, implicit $vgpr4, implicit $sgpr0_sgpr1_sgpr2_sgpr3_sgpr4_sgpr5_sgpr6_sgpr7, implicit $sgpr8_sgpr9_sgpr10_sgpr11, implicit $exec, implicit $vgpr5, implicit $vgpr6 {
+    ; CHECK-NEXT:   S_CLAUSE 2
+    ; CHECK-NEXT:   $vgpr10 = IMAGE_SAMPLE_LZ_V1_V2_gfx12 $vgpr3, $vgpr4, $sgpr0_sgpr1_sgpr2_sgpr3_sgpr4_sgpr5_sgpr6_sgpr7, $sgpr8_sgpr9_sgpr10_sgpr11, 1, 1, -1, 0, 0, 0, 0, 0, 0, implicit $exec :: (load (s128))
+    ; CHECK-NEXT:   $vgpr14 = IMAGE_SAMPLE_LZ_V1_V2_gfx12 $vgpr5, $vgpr6, $sgpr0_sgpr1_sgpr2_sgpr3_sgpr4_sgpr5_sgpr6_sgpr7, $sgpr8_sgpr9_sgpr10_sgpr11, 1, 1, 0, 0, 0, 0, 0, 0, 0, implicit $exec :: (dereferenceable load (s128), addrspace 7)
+    ; CHECK-NEXT:   $vgpr20_vgpr21_vgpr22 = IMAGE_SAMPLE_LZ_V3_V2_gfx12 $vgpr3, $vgpr4, $sgpr0_sgpr1_sgpr2_sgpr3_sgpr4_sgpr5_sgpr6_sgpr7, $sgpr8_sgpr9_sgpr10_sgpr11, 14, 1, -1, 0, 0, 0, 0, 0, 0, implicit $exec :: (load (s128))
+    ; CHECK-NEXT: }
+    $vgpr10 = IMAGE_SAMPLE_LZ_V1_V2_gfx12 $vgpr3, $vgpr4, $sgpr0_sgpr1_sgpr2_sgpr3_sgpr4_sgpr5_sgpr6_sgpr7, $sgpr8_sgpr9_sgpr10_sgpr11, 1, 1, -1, 0, 0, 0, 0, 0, 0, implicit $exec  :: (load (s128))
+    $vgpr14 = IMAGE_SAMPLE_LZ_V1_V2_gfx12 $vgpr5, $vgpr6, $sgpr0_sgpr1_sgpr2_sgpr3_sgpr4_sgpr5_sgpr6_sgpr7, $sgpr8_sgpr9_sgpr10_sgpr11, 1, 1, 0, 0, 0, 0, 0, 0, 0, implicit $exec :: (dereferenceable load (s128), addrspace 7)
+    $vgpr20_vgpr21_vgpr22 = IMAGE_SAMPLE_LZ_V3_V2_gfx12 $vgpr3, $vgpr4, $sgpr0_sgpr1_sgpr2_sgpr3_sgpr4_sgpr5_sgpr6_sgpr7, $sgpr8_sgpr9_sgpr10_sgpr11, 14, 1, -1, 0, 0, 0, 0, 0, 0, implicit $exec  :: (load (s128))
+...
diff --git a/llvm/test/CodeGen/AMDGPU/hard-clauses.mir b/llvm/test/CodeGen/AMDGPU/hard-clauses.mir
index 1c6bdff51015..44b988a7121c 100644
--- a/llvm/test/CodeGen/AMDGPU/hard-clauses.mir
+++ b/llvm/test/CodeGen/AMDGPU/hard-clauses.mir
@@ -1,6 +1,7 @@
 # NOTE: Assertions have been autogenerated by utils/update_mir_test_checks.py
 # RUN: llc -mtriple=amdgcn -mcpu=gfx1010 -verify-machineinstrs -run-pass si-insert-hard-clauses %s -o - | FileCheck %s
 # RUN: llc -mtriple=amdgcn -mcpu=gfx1100 -verify-machineinstrs -run-pass si-insert-hard-clauses %s -o - | FileCheck %s -check-prefix=GFX11
+# RUN: llc -mtriple=amdgcn -mcpu=gfx1200 -verify-machineinstrs -run-pass si-insert-hard-clauses %s -o - | FileCheck %s -check-prefix=GFX12
 
 ---
 name: nop1
@@ -19,6 +20,12 @@ body: |
     ; GFX11-NEXT: {{  $}}
     ; GFX11-NEXT: $sgpr2 = S_LOAD_DWORD_IMM $sgpr0_sgpr1, 0, 0
     ; GFX11-NEXT: S_NOP 2
+    ;
+    ; GFX12-LABEL: name: nop1
+    ; GFX12: liveins: $sgpr0_sgpr1
+    ; GFX12-NEXT: {{  $}}
+    ; GFX12-NEXT: $sgpr2 = S_LOAD_DWORD_IMM $sgpr0_sgpr1, 0, 0
+    ; GFX12-NEXT: S_NOP 2
     $sgpr2 = S_LOAD_DWORD_IMM $sgpr0_sgpr1, 0, 0
     S_NOP 2
 ...
@@ -48,6 +55,16 @@ body: |
     ; GFX11-NEXT:   S_NOP 2
     ; GFX11-NEXT:   $sgpr3 = S_LOAD_DWORD_IMM $sgpr0_sgpr1, 4, 0
     ; GFX11-NEXT: }
+    ;
+    ; GFX12-LABEL: name: nop2
+    ; GFX12: liveins: $sgpr0_sgpr1
+    ; GFX12-NEXT: {{  $}}
+    ; GFX12-NEXT: BUNDLE implicit-def $sgpr2, implicit-def $sgpr2_lo16, implicit-def $sgpr2_hi16, implicit-def $sgpr3, implicit-def $sgpr3_lo16, implicit-def $sgpr3_hi16, implicit $sgpr0_sgpr1 {
+    ; GFX12-NEXT:   S_CLAUSE 2
+    ; GFX12-NEXT:   $sgpr2 = S_LOAD_DWORD_IMM $sgpr0_sgpr1, 0, 0
+    ; GFX12-NEXT:   S_NOP 2
+    ; GFX12-NEXT:   $sgpr3 = S_LOAD_DWORD_IMM $sgpr0_sgpr1, 4, 0
+    ; GFX12-NEXT: }
     $sgpr2 = S_LOAD_DWORD_IMM $sgpr0_sgpr1, 0, 0
     S_NOP 2
     $sgpr3 = S_LOAD_DWORD_IMM $sgpr0_sgpr1, 4, 0
@@ -80,6 +97,17 @@ body: |
     ; GFX11-NEXT:   $sgpr3 = S_LOAD_DWORD_IMM $sgpr0_sgpr1, 4, 0
     ; GFX11-NEXT: }
     ; GFX11-NEXT: S_NOP 2
+    ;
+    ; GFX12-LABEL: name: nop3
+    ; GFX12: liveins: $sgpr0_sgpr1
+    ; GFX12-NEXT: {{  $}}
+    ; GFX12-NEXT: BUNDLE implicit-def $sgpr2, implicit-def $sgpr2_lo16, implicit-def $sgpr2_hi16, implicit-def $sgpr3, implicit-def $sgpr3_lo16, implicit-def $sgpr3_hi16, implicit $sgpr0_sgpr1 {
+    ; GFX12-NEXT:   S_CLAUSE 2
+    ; GFX12-NEXT:   $sgpr2 = S_LOAD_DWORD_IMM $sgpr0_sgpr1, 0, 0
+    ; GFX12-NEXT:   S_NOP 2
+    ; GFX12-NEXT:   $sgpr3 = S_LOAD_DWORD_IMM $sgpr0_sgpr1, 4, 0
+    ; GFX12-NEXT: }
+    ; GFX12-NEXT: S_NOP 2
     $sgpr2 = S_LOAD_DWORD_IMM $sgpr0_sgpr1, 0, 0
     S_NOP 2
     $sgpr3 = S_LOAD_DWORD_IMM $sgpr0_sgpr1, 4, 0
@@ -274,6 +302,99 @@ body: |
     ; GFX11-NEXT:   $vgpr79 = BUFFER_LOAD_DWORD_OFFEN $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 0, 316, 0, 0, implicit $exec
     ; GFX11-NEXT:   $vgpr80 = BUFFER_LOAD_DWORD_OFFEN $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 0, 320, 0, 0, implicit $exec
     ; GFX11-NEXT: }
+    ;
+    ; GFX12-LABEL: name: long_clause
+    ; GFX12: liveins: $sgpr0_sgpr1_sgpr2_sgpr3, $vgpr0
+    ; GFX12-NEXT: {{  $}}
+    ; GFX12-NEXT: BUNDLE implicit-def $vgpr1, implicit-def $vgpr1_lo16, implicit-def $vgpr1_hi16, implicit-def $vgpr2, implicit-def $vgpr2_lo16, implicit-def $vgpr2_hi16, implicit-def $vgpr3, implicit-def $vgpr3_lo16, implicit-def $vgpr3_hi16, implicit-def $vgpr4, implicit-def $vgpr4_lo16, implicit-def $vgpr4_hi16, implicit-def $vgpr5, implicit-def $vgpr5_lo16, implicit-def $vgpr5_hi16, implicit-def $vgpr6, implicit-def $vgpr6_lo16, implicit-def $vgpr6_hi16, implicit-def $vgpr7, implicit-def $vgpr7_lo16, implicit-def $vgpr7_hi16, implicit-def $vgpr8, implicit-def $vgpr8_lo16, implicit-def $vgpr8_hi16, implicit-def $vgpr9, implicit-def $vgpr9_lo16, implicit-def $vgpr9_hi16, implicit-def $vgpr10, implicit-def $vgpr10_lo16, implicit-def $vgpr10_hi16, implicit-def $vgpr11, implicit-def $vgpr11_lo16, implicit-def $vgpr11_hi16, implicit-def $vgpr12, implicit-def $vgpr12_lo16, implicit-def $vgpr12_hi16, implicit-def $vgpr13, implicit-def $vgpr13_lo16, implicit-def $vgpr13_hi16, implicit-def $vgpr14, implicit-def $vgpr14_lo16, implicit-def $vgpr14_hi16, implicit-def $vgpr15, implicit-def $vgpr15_lo16, implicit-def $vgpr15_hi16, implicit-def $vgpr16, implicit-def $vgpr16_lo16, implicit-def $vgpr16_hi16, implicit-def $vgpr17, implicit-def $vgpr17_lo16, implicit-def $vgpr17_hi16, implicit-def $vgpr18, implicit-def $vgpr18_lo16, implicit-def $vgpr18_hi16, implicit-def $vgpr19, implicit-def $vgpr19_lo16, implicit-def $vgpr19_hi16, implicit-def $vgpr20, implicit-def $vgpr20_lo16, implicit-def $vgpr20_hi16, implicit-def $vgpr21, implicit-def $vgpr21_lo16, implicit-def $vgpr21_hi16, implicit-def $vgpr22, implicit-def $vgpr22_lo16, implicit-def $vgpr22_hi16, implicit-def $vgpr23, implicit-def $vgpr23_lo16, implicit-def $vgpr23_hi16, implicit-def $vgpr24, implicit-def $vgpr24_lo16, implicit-def $vgpr24_hi16, implicit-def $vgpr25, implicit-def $vgpr25_lo16, implicit-def $vgpr25_hi16, implicit-def $vgpr26, implicit-def $vgpr26_lo16, implicit-def $vgpr26_hi16, implicit-def $vgpr27, implicit-def $vgpr27_lo16, implicit-def $vgpr27_hi16, implicit-def $vgpr28, implicit-def $vgpr28_lo16, implicit-def $vgpr28_hi16, implicit-def $vgpr29, implicit-def $vgpr29_lo16, implicit-def $vgpr29_hi16, implicit-def $vgpr30, implicit-def $vgpr30_lo16, implicit-def $vgpr30_hi16, implicit-def $vgpr31, implicit-def $vgpr31_lo16, implicit-def $vgpr31_hi16, implicit-def $vgpr32, implicit-def $vgpr32_lo16, implicit-def $vgpr32_hi16, implicit $vgpr0, implicit $sgpr0_sgpr1_sgpr2_sgpr3, implicit $exec {
+    ; GFX12-NEXT:   S_CLAUSE 31
+    ; GFX12-NEXT:   $vgpr1 = BUFFER_LOAD_DWORD_OFFEN $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 0, 4, 0, 0, implicit $exec
+    ; GFX12-NEXT:   $vgpr2 = BUFFER_LOAD_DWORD_OFFEN $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 0, 8, 0, 0, implicit $exec
+    ; GFX12-NEXT:   $vgpr3 = BUFFER_LOAD_DWORD_OFFEN $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 0, 12, 0, 0, implicit $exec
+    ; GFX12-NEXT:   $vgpr4 = BUFFER_LOAD_DWORD_OFFEN $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 0, 16, 0, 0, implicit $exec
+    ; GFX12-NEXT:   $vgpr5 = BUFFER_LOAD_DWORD_OFFEN $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 0, 20, 0, 0, implicit $exec
+    ; GFX12-NEXT:   $vgpr6 = BUFFER_LOAD_DWORD_OFFEN $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 0, 24, 0, 0, implicit $exec
+    ; GFX12-NEXT:   $vgpr7 = BUFFER_LOAD_DWORD_OFFEN $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 0, 28, 0, 0, implicit $exec
+    ; GFX12-NEXT:   $vgpr8 = BUFFER_LOAD_DWORD_OFFEN $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 0, 32, 0, 0, implicit $exec
+    ; GFX12-NEXT:   $vgpr9 = BUFFER_LOAD_DWORD_OFFEN $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 0, 36, 0, 0, implicit $exec
+    ; GFX12-NEXT:   $vgpr10 = BUFFER_LOAD_DWORD_OFFEN $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 0, 40, 0, 0, implicit $exec
+    ; GFX12-NEXT:   $vgpr11 = BUFFER_LOAD_DWORD_OFFEN $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 0, 44, 0, 0, implicit $exec
+    ; GFX12-NEXT:   $vgpr12 = BUFFER_LOAD_DWORD_OFFEN $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 0, 48, 0, 0, implicit $exec
+    ; GFX12-NEXT:   $vgpr13 = BUFFER_LOAD_DWORD_OFFEN $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 0, 52, 0, 0, implicit $exec
+    ; GFX12-NEXT:   $vgpr14 = BUFFER_LOAD_DWORD_OFFEN $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 0, 56, 0, 0, implicit $exec
+    ; GFX12-NEXT:   $vgpr15 = BUFFER_LOAD_DWORD_OFFEN $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 0, 60, 0, 0, implicit $exec
+    ; GFX12-NEXT:   $vgpr16 = BUFFER_LOAD_DWORD_OFFEN $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 0, 64, 0, 0, implicit $exec
+    ; GFX12-NEXT:   $vgpr17 = BUFFER_LOAD_DWORD_OFFEN $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 0, 68, 0, 0, implicit $exec
+    ; GFX12-NEXT:   $vgpr18 = BUFFER_LOAD_DWORD_OFFEN $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 0, 72, 0, 0, implicit $exec
+    ; GFX12-NEXT:   $vgpr19 = BUFFER_LOAD_DWORD_OFFEN $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 0, 76, 0, 0, implicit $exec
+    ; GFX12-NEXT:   $vgpr20 = BUFFER_LOAD_DWORD_OFFEN $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 0, 80, 0, 0, implicit $exec
+    ; GFX12-NEXT:   $vgpr21 = BUFFER_LOAD_DWORD_OFFEN $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 0, 84, 0, 0, implicit $exec
+    ; GFX12-NEXT:   $vgpr22 = BUFFER_LOAD_DWORD_OFFEN $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 0, 88, 0, 0, implicit $exec
+    ; GFX12-NEXT:   $vgpr23 = BUFFER_LOAD_DWORD_OFFEN $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 0, 92, 0, 0, implicit $exec
+    ; GFX12-NEXT:   $vgpr24 = BUFFER_LOAD_DWORD_OFFEN $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 0, 96, 0, 0, implicit $exec
+    ; GFX12-NEXT:   $vgpr25 = BUFFER_LOAD_DWORD_OFFEN $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 0, 100, 0, 0, implicit $exec
+    ; GFX12-NEXT:   $vgpr26 = BUFFER_LOAD_DWORD_OFFEN $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 0, 104, 0, 0, implicit $exec
+    ; GFX12-NEXT:   $vgpr27 = BUFFER_LOAD_DWORD_OFFEN $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 0, 108, 0, 0, implicit $exec
+    ; GFX12-NEXT:   $vgpr28 = BUFFER_LOAD_DWORD_OFFEN $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 0, 112, 0, 0, implicit $exec
+    ; GFX12-NEXT:   $vgpr29 = BUFFER_LOAD_DWORD_OFFEN $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 0, 116, 0, 0, implicit $exec
+    ; GFX12-NEXT:   $vgpr30 = BUFFER_LOAD_DWORD_OFFEN $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 0, 120, 0, 0, implicit $exec
+    ; GFX12-NEXT:   $vgpr31 = BUFFER_LOAD_DWORD_OFFEN $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 0, 124, 0, 0, implicit $exec
+    ; GFX12-NEXT:   $vgpr32 = BUFFER_LOAD_DWORD_OFFEN $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 0, 128, 0, 0, implicit $exec
+    ; GFX12-NEXT: }
+    ; GFX12-NEXT: BUNDLE implicit-def $vgpr33, implicit-def $vgpr33_lo16, implicit-def $vgpr33_hi16, implicit-def $vgpr34, implicit-def $vgpr34_lo16, implicit-def $vgpr34_hi16, implicit-def $vgpr35, implicit-def $vgpr35_lo16, implicit-def $vgpr35_hi16, implicit-def $vgpr36, implicit-def $vgpr36_lo16, implicit-def $vgpr36_hi16, implicit-def $vgpr37, implicit-def $vgpr37_lo16, implicit-def $vgpr37_hi16, implicit-def $vgpr38, implicit-def $vgpr38_lo16, implicit-def $vgpr38_hi16, implicit-def $vgpr39, implicit-def $vgpr39_lo16, implicit-def $vgpr39_hi16, implicit-def $vgpr40, implicit-def $vgpr40_lo16, implicit-def $vgpr40_hi16, implicit-def $vgpr41, implicit-def $vgpr41_lo16, implicit-def $vgpr41_hi16, implicit-def $vgpr42, implicit-def $vgpr42_lo16, implicit-def $vgpr42_hi16, implicit-def $vgpr43, implicit-def $vgpr43_lo16, implicit-def $vgpr43_hi16, implicit-def $vgpr44, implicit-def $vgpr44_lo16, implicit-def $vgpr44_hi16, implicit-def $vgpr45, implicit-def $vgpr45_lo16, implicit-def $vgpr45_hi16, implicit-def $vgpr46, implicit-def $vgpr46_lo16, implicit-def $vgpr46_hi16, implicit-def $vgpr47, implicit-def $vgpr47_lo16, implicit-def $vgpr47_hi16, implicit-def $vgpr48, implicit-def $vgpr48_lo16, implicit-def $vgpr48_hi16, implicit-def $vgpr49, implicit-def $vgpr49_lo16, implicit-def $vgpr49_hi16, implicit-def $vgpr50, implicit-def $vgpr50_lo16, implicit-def $vgpr50_hi16, implicit-def $vgpr51, implicit-def $vgpr51_lo16, implicit-def $vgpr51_hi16, implicit-def $vgpr52, implicit-def $vgpr52_lo16, implicit-def $vgpr52_hi16, implicit-def $vgpr53, implicit-def $vgpr53_lo16, implicit-def $vgpr53_hi16, implicit-def $vgpr54, implicit-def $vgpr54_lo16, implicit-def $vgpr54_hi16, implicit-def $vgpr55, implicit-def $vgpr55_lo16, implicit-def $vgpr55_hi16, implicit-def $vgpr56, implicit-def $vgpr56_lo16, implicit-def $vgpr56_hi16, implicit-def $vgpr57, implicit-def $vgpr57_lo16, implicit-def $vgpr57_hi16, implicit-def $vgpr58, implicit-def $vgpr58_lo16, implicit-def $vgpr58_hi16, implicit-def $vgpr59, implicit-def $vgpr59_lo16, implicit-def $vgpr59_hi16, implicit-def $vgpr60, implicit-def $vgpr60_lo16, implicit-def $vgpr60_hi16, implicit-def $vgpr61, implicit-def $vgpr61_lo16, implicit-def $vgpr61_hi16, implicit-def $vgpr62, implicit-def $vgpr62_lo16, implicit-def $vgpr62_hi16, implicit-def $vgpr63, implicit-def $vgpr63_lo16, implicit-def $vgpr63_hi16, implicit-def $vgpr64, implicit-def $vgpr64_lo16, implicit-def $vgpr64_hi16, implicit $vgpr0, implicit $sgpr0_sgpr1_sgpr2_sgpr3, implicit $exec {
+    ; GFX12-NEXT:   S_CLAUSE 31
+    ; GFX12-NEXT:   $vgpr33 = BUFFER_LOAD_DWORD_OFFEN $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 0, 132, 0, 0, implicit $exec
+    ; GFX12-NEXT:   $vgpr34 = BUFFER_LOAD_DWORD_OFFEN $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 0, 136, 0, 0, implicit $exec
+    ; GFX12-NEXT:   $vgpr35 = BUFFER_LOAD_DWORD_OFFEN $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 0, 140, 0, 0, implicit $exec
+    ; GFX12-NEXT:   $vgpr36 = BUFFER_LOAD_DWORD_OFFEN $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 0, 144, 0, 0, implicit $exec
+    ; GFX12-NEXT:   $vgpr37 = BUFFER_LOAD_DWORD_OFFEN $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 0, 148, 0, 0, implicit $exec
+    ; GFX12-NEXT:   $vgpr38 = BUFFER_LOAD_DWORD_OFFEN $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 0, 152, 0, 0, implicit $exec
+    ; GFX12-NEXT:   $vgpr39 = BUFFER_LOAD_DWORD_OFFEN $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 0, 156, 0, 0, implicit $exec
+    ; GFX12-NEXT:   $vgpr40 = BUFFER_LOAD_DWORD_OFFEN $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 0, 160, 0, 0, implicit $exec
+    ; GFX12-NEXT:   $vgpr41 = BUFFER_LOAD_DWORD_OFFEN $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 0, 164, 0, 0, implicit $exec
+    ; GFX12-NEXT:   $vgpr42 = BUFFER_LOAD_DWORD_OFFEN $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 0, 168, 0, 0, implicit $exec
+    ; GFX12-NEXT:   $vgpr43 = BUFFER_LOAD_DWORD_OFFEN $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 0, 172, 0, 0, implicit $exec
+    ; GFX12-NEXT:   $vgpr44 = BUFFER_LOAD_DWORD_OFFEN $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 0, 176, 0, 0, implicit $exec
+    ; GFX12-NEXT:   $vgpr45 = BUFFER_LOAD_DWORD_OFFEN $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 0, 180, 0, 0, implicit $exec
+    ; GFX12-NEXT:   $vgpr46 = BUFFER_LOAD_DWORD_OFFEN $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 0, 184, 0, 0, implicit $exec
+    ; GFX12-NEXT:   $vgpr47 = BUFFER_LOAD_DWORD_OFFEN $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 0, 188, 0, 0, implicit $exec
+    ; GFX12-NEXT:   $vgpr48 = BUFFER_LOAD_DWORD_OFFEN $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 0, 192, 0, 0, implicit $exec
+    ; GFX12-NEXT:   $vgpr49 = BUFFER_LOAD_DWORD_OFFEN $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 0, 196, 0, 0, implicit $exec
+    ; GFX12-NEXT:   $vgpr50 = BUFFER_LOAD_DWORD_OFFEN $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 0, 200, 0, 0, implicit $exec
+    ; GFX12-NEXT:   $vgpr51 = BUFFER_LOAD_DWORD_OFFEN $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 0, 204, 0, 0, implicit $exec
+    ; GFX12-NEXT:   $vgpr52 = BUFFER_LOAD_DWORD_OFFEN $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 0, 208, 0, 0, implicit $exec
+    ; GFX12-NEXT:   $vgpr53 = BUFFER_LOAD_DWORD_OFFEN $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 0, 212, 0, 0, implicit $exec
+    ; GFX12-NEXT:   $vgpr54 = BUFFER_LOAD_DWORD_OFFEN $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 0, 216, 0, 0, implicit $exec
+    ; GFX12-NEXT:   $vgpr55 = BUFFER_LOAD_DWORD_OFFEN $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 0, 220, 0, 0, implicit $exec
+    ; GFX12-NEXT:   $vgpr56 = BUFFER_LOAD_DWORD_OFFEN $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 0, 224, 0, 0, implicit $exec
+    ; GFX12-NEXT:   $vgpr57 = BUFFER_LOAD_DWORD_OFFEN $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 0, 228, 0, 0, implicit $exec
+    ; GFX12-NEXT:   $vgpr58 = BUFFER_LOAD_DWORD_OFFEN $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 0, 232, 0, 0, implicit $exec
+    ; GFX12-NEXT:   $vgpr59 = BUFFER_LOAD_DWORD_OFFEN $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 0, 236, 0, 0, implicit $exec
+    ; GFX12-NEXT:   $vgpr60 = BUFFER_LOAD_DWORD_OFFEN $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 0, 240, 0, 0, implicit $exec
+    ; GFX12-NEXT:   $vgpr61 = BUFFER_LOAD_DWORD_OFFEN $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 0, 244, 0, 0, implicit $exec
+    ; GFX12-NEXT:   $vgpr62 = BUFFER_LOAD_DWORD_OFFEN $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 0, 248, 0, 0, implicit $exec
+    ; GFX12-NEXT:   $vgpr63 = BUFFER_LOAD_DWORD_OFFEN $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 0, 252, 0, 0, implicit $exec
+    ; GFX12-NEXT:   $vgpr64 = BUFFER_LOAD_DWORD_OFFEN $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 0, 256, 0, 0, implicit $exec
+    ; GFX12-NEXT: }
+    ; GFX12-NEXT: BUNDLE implicit-def $vgpr65, implicit-def $vgpr65_lo16, implicit-def $vgpr65_hi16, implicit-def $vgpr66, implicit-def $vgpr66_lo16, implicit-def $vgpr66_hi16, implicit-def $vgpr67, implicit-def $vgpr67_lo16, implicit-def $vgpr67_hi16, implicit-def $vgpr68, implicit-def $vgpr68_lo16, implicit-def $vgpr68_hi16, implicit-def $vgpr69, implicit-def $vgpr69_lo16, implicit-def $vgpr69_hi16, implicit-def $vgpr70, implicit-def $vgpr70_lo16, implicit-def $vgpr70_hi16, implicit-def $vgpr71, implicit-def $vgpr71_lo16, implicit-def $vgpr71_hi16, implicit-def $vgpr72, implicit-def $vgpr72_lo16, implicit-def $vgpr72_hi16, implicit-def $vgpr73, implicit-def $vgpr73_lo16, implicit-def $vgpr73_hi16, implicit-def $vgpr74, implicit-def $vgpr74_lo16, implicit-def $vgpr74_hi16, implicit-def $vgpr75, implicit-def $vgpr75_lo16, implicit-def $vgpr75_hi16, implicit-def $vgpr76, implicit-def $vgpr76_lo16, implicit-def $vgpr76_hi16, implicit-def $vgpr77, implicit-def $vgpr77_lo16, implicit-def $vgpr77_hi16, implicit-def $vgpr78, implicit-def $vgpr78_lo16, implicit-def $vgpr78_hi16, implicit-def $vgpr79, implicit-def $vgpr79_lo16, implicit-def $vgpr79_hi16, implicit-def $vgpr80, implicit-def $vgpr80_lo16, implicit-def $vgpr80_hi16, implicit $vgpr0, implicit $sgpr0_sgpr1_sgpr2_sgpr3, implicit $exec {
+    ; GFX12-NEXT:   S_CLAUSE 15
+    ; GFX12-NEXT:   $vgpr65 = BUFFER_LOAD_DWORD_OFFEN $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 0, 260, 0, 0, implicit $exec
+    ; GFX12-NEXT:   $vgpr66 = BUFFER_LOAD_DWORD_OFFEN $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 0, 264, 0, 0, implicit $exec
+    ; GFX12-NEXT:   $vgpr67 = BUFFER_LOAD_DWORD_OFFEN $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 0, 268, 0, 0, implicit $exec
+    ; GFX12-NEXT:   $vgpr68 = BUFFER_LOAD_DWORD_OFFEN $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 0, 272, 0, 0, implicit $exec
+    ; GFX12-NEXT:   $vgpr69 = BUFFER_LOAD_DWORD_OFFEN $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 0, 276, 0, 0, implicit $exec
+    ; GFX12-NEXT:   $vgpr70 = BUFFER_LOAD_DWORD_OFFEN $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 0, 280, 0, 0, implicit $exec
+    ; GFX12-NEXT:   $vgpr71 = BUFFER_LOAD_DWORD_OFFEN $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 0, 284, 0, 0, implicit $exec
+    ; GFX12-NEXT:   $vgpr72 = BUFFER_LOAD_DWORD_OFFEN $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 0, 288, 0, 0, implicit $exec
+    ; GFX12-NEXT:   $vgpr73 = BUFFER_LOAD_DWORD_OFFEN $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 0, 292, 0, 0, implicit $exec
+    ; GFX12-NEXT:   $vgpr74 = BUFFER_LOAD_DWORD_OFFEN $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 0, 296, 0, 0, implicit $exec
+    ; GFX12-NEXT:   $vgpr75 = BUFFER_LOAD_DWORD_OFFEN $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 0, 300, 0, 0, implicit $exec
+    ; GFX12-NEXT:   $vgpr76 = BUFFER_LOAD_DWORD_OFFEN $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 0, 304, 0, 0, implicit $exec
+    ; GFX12-NEXT:   $vgpr77 = BUFFER_LOAD_DWORD_OFFEN $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 0, 308, 0, 0, implicit $exec
+    ; GFX12-NEXT:   $vgpr78 = BUFFER_LOAD_DWORD_OFFEN $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 0, 312, 0, 0, implicit $exec
+    ; GFX12-NEXT:   $vgpr79 = BUFFER_LOAD_DWORD_OFFEN $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 0, 316, 0, 0, implicit $exec
+    ; GFX12-NEXT:   $vgpr80 = BUFFER_LOAD_DWORD_OFFEN $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 0, 320, 0, 0, implicit $exec
+    ; GFX12-NEXT: }
     $vgpr1 = BUFFER_LOAD_DWORD_OFFEN $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 0, 4, 0, 0, implicit $exec
     $vgpr2 = BUFFER_LOAD_DWORD_OFFEN $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 0, 8, 0, 0, implicit $exec
     $vgpr3 = BUFFER_LOAD_DWORD_OFFEN $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 0, 12, 0, 0, implicit $exec
@@ -357,57 +478,6 @@ body: |
 ...
 
 ---
-name: mimg_nsa
-tracksRegLiveness: true
-body: |
-  bb.0:
-    liveins: $sgpr0_sgpr1_sgpr2_sgpr3_sgpr4_sgpr5_sgpr6_sgpr7, $sgpr8_sgpr9_sgpr10_sgpr11, $vgpr0, $vgpr1, $vgpr2, $vgpr3, $vgpr4, $vgpr5, $vgpr6, $vgpr7, $vgpr8
-    ; CHECK-LABEL: name: mimg_nsa
-    ; CHECK: liveins: $sgpr0_sgpr1_sgpr2_sgpr3_sgpr4_sgpr5_sgpr6_sgpr7, $sgpr8_sgpr9_sgpr10_sgpr11, $vgpr0, $vgpr1, $vgpr2, $vgpr3, $vgpr4, $vgpr5, $vgpr6, $vgpr7, $vgpr8
-    ; CHECK-NEXT: {{  $}}
-    ; CHECK-NEXT: $vgpr10_vgpr11_vgpr12_vgpr13 = IMAGE_SAMPLE_D_V4_V9_nsa_gfx10 $vgpr3, $vgpr8, $vgpr7, $vgpr5, $vgpr4, $vgpr6, $vgpr0, $vgpr2, $vgpr2, $sgpr0_sgpr1_sgpr2_sgpr3_sgpr4_sgpr5_sgpr6_sgpr7, $sgpr8_sgpr9_sgpr10_sgpr11, 15, 2, 0, 0, 0, 0, 0, 0, 0, implicit $exec :: (load (s128))
-    ; CHECK-NEXT: $vgpr20_vgpr21_vgpr22_vgpr23 = IMAGE_SAMPLE_D_V4_V9_nsa_gfx10 $vgpr3, $vgpr8, $vgpr7, $vgpr5, $vgpr4, $vgpr6, $vgpr0, $vgpr2, $vgpr2, $sgpr0_sgpr1_sgpr2_sgpr3_sgpr4_sgpr5_sgpr6_sgpr7, $sgpr8_sgpr9_sgpr10_sgpr11, 15, 2, 0, 0, 0, 0, 0, 0, 0, implicit $exec :: (load (s128))
-    ;
-    ; GFX11-LABEL: name: mimg_nsa
-    ; GFX11: liveins: $sgpr0_sgpr1_sgpr2_sgpr3_sgpr4_sgpr5_sgpr6_sgpr7, $sgpr8_sgpr9_sgpr10_sgpr11, $vgpr0, $vgpr1, $vgpr2, $vgpr3, $vgpr4, $vgpr5, $vgpr6, $vgpr7, $vgpr8
-    ; GFX11-NEXT: {{  $}}
-    ; GFX11-NEXT: BUNDLE implicit-def $vgpr10_vgpr11_vgpr12_vgpr13, implicit-def $vgpr10, implicit-def $vgpr10_lo16, implicit-def $vgpr10_hi16, implicit-def $vgpr11, implicit-def $vgpr11_lo16, implicit-def $vgpr11_hi16, implicit-def $vgpr12, implicit-def $vgpr12_lo16, implicit-def $vgpr12_hi16, implicit-def $vgpr13, implicit-def $vgpr13_lo16, implicit-def $vgpr13_hi16, implicit-def $vgpr10_vgpr11, implicit-def $vgpr10_vgpr11_vgpr12, implicit-def $vgpr11_vgpr12, implicit-def $vgpr11_vgpr12_vgpr13, implicit-def $vgpr12_vgpr13, implicit-def $vgpr20_vgpr21_vgpr22_vgpr23, implicit-def $vgpr20, implicit-def $vgpr20_lo16, implicit-def $vgpr20_hi16, implicit-def $vgpr21, implicit-def $vgpr21_lo16, implicit-def $vgpr21_hi16, implicit-def $vgpr22, implicit-def $vgpr22_lo16, implicit-def $vgpr22_hi16, implicit-def $vgpr23, implicit-def $vgpr23_lo16, implicit-def $vgpr23_hi16, implicit-def $vgpr20_vgpr21, implicit-def $vgpr20_vgpr21_vgpr22, implicit-def $vgpr21_vgpr22, implicit-def $vgpr21_vgpr22_vgpr23, implicit-def $vgpr22_vgpr23, implicit $vgpr3, implicit $vgpr8, implicit $vgpr7, implicit $vgpr5, implicit $vgpr4, implicit $vgpr6, implicit $vgpr0, implicit $vgpr2, implicit $sgpr0_sgpr1_sgpr2_sgpr3_sgpr4_sgpr5_sgpr6_sgpr7, implicit $sgpr8_sgpr9_sgpr10_sgpr11, implicit $exec {
-    ; GFX11-NEXT:   S_CLAUSE 1
-    ; GFX11-NEXT:   $vgpr10_vgpr11_vgpr12_vgpr13 = IMAGE_SAMPLE_D_V4_V9_nsa_gfx10 $vgpr3, $vgpr8, $vgpr7, $vgpr5, $vgpr4, $vgpr6, $vgpr0, $vgpr2, $vgpr2, $sgpr0_sgpr1_sgpr2_sgpr3_sgpr4_sgpr5_sgpr6_sgpr7, $sgpr8_sgpr9_sgpr10_sgpr11, 15, 2, 0, 0, 0, 0, 0, 0, 0, implicit $exec :: (load (s128))
-    ; GFX11-NEXT:   $vgpr20_vgpr21_vgpr22_vgpr23 = IMAGE_SAMPLE_D_V4_V9_nsa_gfx10 $vgpr3, $vgpr8, $vgpr7, $vgpr5, $vgpr4, $vgpr6, $vgpr0, $vgpr2, $vgpr2, $sgpr0_sgpr1_sgpr2_sgpr3_sgpr4_sgpr5_sgpr6_sgpr7, $sgpr8_sgpr9_sgpr10_sgpr11, 15, 2, 0, 0, 0, 0, 0, 0, 0, implicit $exec :: (load (s128))
-    ; GFX11-NEXT: }
-    $vgpr10_vgpr11_vgpr12_vgpr13 = IMAGE_SAMPLE_D_V4_V9_nsa_gfx10 $vgpr3, $vgpr8, $vgpr7, $vgpr5, $vgpr4, $vgpr6, $vgpr0, $vgpr2, $vgpr2, $sgpr0_sgpr1_sgpr2_sgpr3_sgpr4_sgpr5_sgpr6_sgpr7, $sgpr8_sgpr9_sgpr10_sgpr11, 15, 2, 0, 0, 0, 0, 0, 0, 0, implicit $exec  :: (load (s128))
-    $vgpr20_vgpr21_vgpr22_vgpr23 = IMAGE_SAMPLE_D_V4_V9_nsa_gfx10 $vgpr3, $vgpr8, $vgpr7, $vgpr5, $vgpr4, $vgpr6, $vgpr0, $vgpr2, $vgpr2, $sgpr0_sgpr1_sgpr2_sgpr3_sgpr4_sgpr5_sgpr6_sgpr7, $sgpr8_sgpr9_sgpr10_sgpr11, 15, 2, 0, 0, 0, 0, 0, 0, 0, implicit $exec  :: (load (s128))
-...
-
----
-name: mimg_nsa_mixed
-tracksRegLiveness: true
-body: |
-  bb.0:
-    liveins: $sgpr0_sgpr1_sgpr2_sgpr3_sgpr4_sgpr5_sgpr6_sgpr7, $sgpr8_sgpr9_sgpr10_sgpr11, $vgpr0, $vgpr1, $vgpr2, $vgpr3, $vgpr4, $vgpr5, $vgpr6, $vgpr7, $vgpr8
-    ; CHECK-LABEL: name: mimg_nsa_mixed
-    ; CHECK: liveins: $sgpr0_sgpr1_sgpr2_sgpr3_sgpr4_sgpr5_sgpr6_sgpr7, $sgpr8_sgpr9_sgpr10_sgpr11, $vgpr0, $vgpr1, $vgpr2, $vgpr3, $vgpr4, $vgpr5, $vgpr6, $vgpr7, $vgpr8
-    ; CHECK-NEXT: {{  $}}
-    ; CHECK-NEXT: $vgpr10_vgpr11_vgpr12_vgpr13 = IMAGE_SAMPLE_D_V4_V9_nsa_gfx10 $vgpr3, $vgpr8, $vgpr7, $vgpr5, $vgpr4, $vgpr6, $vgpr0, $vgpr2, $vgpr2, $sgpr0_sgpr1_sgpr2_sgpr3_sgpr4_sgpr5_sgpr6_sgpr7, $sgpr8_sgpr9_sgpr10_sgpr11, 15, 2, 0, 0, 0, 0, 0, 0, 0, implicit $exec :: (load (s128))
-    ; CHECK-NEXT: $vgpr14 = IMAGE_SAMPLE_LZ_V1_V2_gfx10 $vgpr5_vgpr6, $sgpr0_sgpr1_sgpr2_sgpr3_sgpr4_sgpr5_sgpr6_sgpr7, $sgpr8_sgpr9_sgpr10_sgpr11, 1, 1, 0, 0, 0, 0, 0, 0, 0, implicit $exec :: (dereferenceable load (s128), addrspace 7)
-    ; CHECK-NEXT: $vgpr20_vgpr21_vgpr22_vgpr23 = IMAGE_SAMPLE_D_V4_V9_nsa_gfx10 $vgpr3, $vgpr8, $vgpr7, $vgpr5, $vgpr4, $vgpr6, $vgpr0, $vgpr2, $vgpr2, $sgpr0_sgpr1_sgpr2_sgpr3_sgpr4_sgpr5_sgpr6_sgpr7, $sgpr8_sgpr9_sgpr10_sgpr11, 15, 2, 0, 0, 0, 0, 0, 0, 0, implicit $exec :: (load (s128))
-    ;
-    ; GFX11-LABEL: name: mimg_nsa_mixed
-    ; GFX11: liveins: $sgpr0_sgpr1_sgpr2_sgpr3_sgpr4_sgpr5_sgpr6_sgpr7, $sgpr8_sgpr9_sgpr10_sgpr11, $vgpr0, $vgpr1, $vgpr2, $vgpr3, $vgpr4, $vgpr5, $vgpr6, $vgpr7, $vgpr8
-    ; GFX11-NEXT: {{  $}}
-    ; GFX11-NEXT: BUNDLE implicit-def $vgpr10_vgpr11_vgpr12_vgpr13, implicit-def $vgpr10, implicit-def $vgpr10_lo16, implicit-def $vgpr10_hi16, implicit-def $vgpr11, implicit-def $vgpr11_lo16, implicit-def $vgpr11_hi16, implicit-def $vgpr12, implicit-def $vgpr12_lo16, implicit-def $vgpr12_hi16, implicit-def $vgpr13, implicit-def $vgpr13_lo16, implicit-def $vgpr13_hi16, implicit-def $vgpr10_vgpr11, implicit-def $vgpr10_vgpr11_vgpr12, implicit-def $vgpr11_vgpr12, implicit-def $vgpr11_vgpr12_vgpr13, implicit-def $vgpr12_vgpr13, implicit-def $vgpr14, implicit-def $vgpr14_lo16, implicit-def $vgpr14_hi16, implicit-def $vgpr20_vgpr21_vgpr22_vgpr23, implicit-def $vgpr20, implicit-def $vgpr20_lo16, implicit-def $vgpr20_hi16, implicit-def $vgpr21, implicit-def $vgpr21_lo16, implicit-def $vgpr21_hi16, implicit-def $vgpr22, implicit-def $vgpr22_lo16, implicit-def $vgpr22_hi16, implicit-def $vgpr23, implicit-def $vgpr23_lo16, implicit-def $vgpr23_hi16, implicit-def $vgpr20_vgpr21, implicit-def $vgpr20_vgpr21_vgpr22, implicit-def $vgpr21_vgpr22, implicit-def $vgpr21_vgpr22_vgpr23, implicit-def $vgpr22_vgpr23, implicit $vgpr3, implicit $vgpr8, implicit $vgpr7, implicit $vgpr5, implicit $vgpr4, implicit $vgpr6, implicit $vgpr0, implicit $vgpr2, implicit $sgpr0_sgpr1_sgpr2_sgpr3_sgpr4_sgpr5_sgpr6_sgpr7, implicit $sgpr8_sgpr9_sgpr10_sgpr11, implicit $exec, implicit $vgpr5_vgpr6 {
-    ; GFX11-NEXT:   S_CLAUSE 2
-    ; GFX11-NEXT:   $vgpr10_vgpr11_vgpr12_vgpr13 = IMAGE_SAMPLE_D_V4_V9_nsa_gfx10 $vgpr3, $vgpr8, $vgpr7, $vgpr5, $vgpr4, $vgpr6, $vgpr0, $vgpr2, $vgpr2, $sgpr0_sgpr1_sgpr2_sgpr3_sgpr4_sgpr5_sgpr6_sgpr7, $sgpr8_sgpr9_sgpr10_sgpr11, 15, 2, 0, 0, 0, 0, 0, 0, 0, implicit $exec :: (load (s128))
-    ; GFX11-NEXT:   $vgpr14 = IMAGE_SAMPLE_LZ_V1_V2_gfx10 $vgpr5_vgpr6, $sgpr0_sgpr1_sgpr2_sgpr3_sgpr4_sgpr5_sgpr6_sgpr7, $sgpr8_sgpr9_sgpr10_sgpr11, 1, 1, 0, 0, 0, 0, 0, 0, 0, implicit $exec :: (dereferenceable load (s128), addrspace 7)
-    ; GFX11-NEXT:   $vgpr20_vgpr21_vgpr22_vgpr23 = IMAGE_SAMPLE_D_V4_V9_nsa_gfx10 $vgpr3, $vgpr8, $vgpr7, $vgpr5, $vgpr4, $vgpr6, $vgpr0, $vgpr2, $vgpr2, $sgpr0_sgpr1_sgpr2_sgpr3_sgpr4_sgpr5_sgpr6_sgpr7, $sgpr8_sgpr9_sgpr10_sgpr11, 15, 2, 0, 0, 0, 0, 0, 0, 0, implicit $exec :: (load (s128))
-    ; GFX11-NEXT: }
-    $vgpr10_vgpr11_vgpr12_vgpr13 = IMAGE_SAMPLE_D_V4_V9_nsa_gfx10 $vgpr3, $vgpr8, $vgpr7, $vgpr5, $vgpr4, $vgpr6, $vgpr0, $vgpr2, $vgpr2, $sgpr0_sgpr1_sgpr2_sgpr3_sgpr4_sgpr5_sgpr6_sgpr7, $sgpr8_sgpr9_sgpr10_sgpr11, 15, 2, 0, 0, 0, 0, 0, 0, 0, implicit $exec  :: (load (s128))
-    $vgpr14 = IMAGE_SAMPLE_LZ_V1_V2_gfx10 $vgpr5_vgpr6, $sgpr0_sgpr1_sgpr2_sgpr3_sgpr4_sgpr5_sgpr6_sgpr7, $sgpr8_sgpr9_sgpr10_sgpr11, 1, 1, 0, 0, 0, 0, 0, 0, 0, implicit $exec :: (dereferenceable load (s128), addrspace 7)
-    $vgpr20_vgpr21_vgpr22_vgpr23 = IMAGE_SAMPLE_D_V4_V9_nsa_gfx10 $vgpr3, $vgpr8, $vgpr7, $vgpr5, $vgpr4, $vgpr6, $vgpr0, $vgpr2, $vgpr2, $sgpr0_sgpr1_sgpr2_sgpr3_sgpr4_sgpr5_sgpr6_sgpr7, $sgpr8_sgpr9_sgpr10_sgpr11, 15, 2, 0, 0, 0, 0, 0, 0, 0, implicit $exec  :: (load (s128))
-...
-
----
 name: kill
 tracksRegLiveness: true
 body: |
@@ -432,6 +502,16 @@ body: |
     ; GFX11-NEXT:   KILL undef renamable $sgpr4
     ; GFX11-NEXT:   $sgpr3 = S_LOAD_DWORD_IMM $sgpr0_sgpr1, 4, 0
     ; GFX11-NEXT: }
+    ;
+    ; GFX12-LABEL: name: kill
+    ; GFX12: liveins: $sgpr0_sgpr1, $sgpr4
+    ; GFX12-NEXT: {{  $}}
+    ; GFX12-NEXT: BUNDLE implicit-def $sgpr2, implicit-def $sgpr2_lo16, implicit-def $sgpr2_hi16, implicit-def $sgpr3, implicit-def $sgpr3_lo16, implicit-def $sgpr3_hi16, implicit $sgpr0_sgpr1, implicit undef $sgpr4 {
+    ; GFX12-NEXT:   S_CLAUSE 1
+    ; GFX12-NEXT:   $sgpr2 = S_LOAD_DWORD_IMM $sgpr0_sgpr1, 0, 0
+    ; GFX12-NEXT:   KILL undef renamable $sgpr4
+    ; GFX12-NEXT:   $sgpr3 = S_LOAD_DWORD_IMM $sgpr0_sgpr1, 4, 0
+    ; GFX12-NEXT: }
     $sgpr2 = S_LOAD_DWORD_IMM $sgpr0_sgpr1, 0, 0
     KILL undef renamable $sgpr4
     $sgpr3 = S_LOAD_DWORD_IMM $sgpr0_sgpr1, 4, 0
@@ -464,6 +544,17 @@ body: |
     ; GFX11-NEXT:   $sgpr3 = S_LOAD_DWORD_IMM $sgpr0_sgpr1, 4, 0
     ; GFX11-NEXT: }
     ; GFX11-NEXT: KILL undef renamable $sgpr5
+    ;
+    ; GFX12-LABEL: name: kill2
+    ; GFX12: liveins: $sgpr0_sgpr1, $sgpr4, $sgpr5
+    ; GFX12-NEXT: {{  $}}
+    ; GFX12-NEXT: BUNDLE implicit-def $sgpr2, implicit-def $sgpr2_lo16, implicit-def $sgpr2_hi16, implicit-def $sgpr3, implicit-def $sgpr3_lo16, implicit-def $sgpr3_hi16, implicit $sgpr0_sgpr1, implicit undef $sgpr4 {
+    ; GFX12-NEXT:   S_CLAUSE 1
+    ; GFX12-NEXT:   $sgpr2 = S_LOAD_DWORD_IMM $sgpr0_sgpr1, 0, 0
+    ; GFX12-NEXT:   KILL undef renamable $sgpr4
+    ; GFX12-NEXT:   $sgpr3 = S_LOAD_DWORD_IMM $sgpr0_sgpr1, 4, 0
+    ; GFX12-NEXT: }
+    ; GFX12-NEXT: KILL undef renamable $sgpr5
     $sgpr2 = S_LOAD_DWORD_IMM $sgpr0_sgpr1, 0, 0
     KILL undef renamable $sgpr4
     $sgpr3 = S_LOAD_DWORD_IMM $sgpr0_sgpr1, 4, 0
@@ -490,6 +581,12 @@ body: |
     ; GFX11-NEXT: {{  $}}
     ; GFX11-NEXT: $vgpr3 = FLAT_LOAD_DWORD $vgpr0_vgpr1, 0, 0, implicit $exec, implicit $flat_scr
     ; GFX11-NEXT: $vgpr4 = FLAT_ATOMIC_ADD_RTN $vgpr0_vgpr1, $vgpr2, 4, 0, implicit $exec, implicit $flat_scr
+    ;
+    ; GFX12-LABEL: name: flat_load_atomic
+    ; GFX12: liveins: $vgpr0_vgpr1, $vgpr2
+    ; GFX12-NEXT: {{  $}}
+    ; GFX12-NEXT: $vgpr3 = FLAT_LOAD_DWORD $vgpr0_vgpr1, 0, 0, implicit $exec, implicit $flat_scr
+    ; GFX12-NEXT: $vgpr4 = FLAT_ATOMIC_ADD_RTN $vgpr0_vgpr1, $vgpr2, 4, 0, implicit $exec, implicit $flat_scr
     $vgpr3 = FLAT_LOAD_DWORD $vgpr0_vgpr1, 0, 0, implicit $exec, implicit $flat_scr
     $vgpr4 = FLAT_ATOMIC_ADD_RTN $vgpr0_vgpr1, $vgpr2, 4, 0, implicit $exec, implicit $flat_scr
 ...
@@ -514,6 +611,12 @@ body: |
     ; GFX11-NEXT: {{  $}}
     ; GFX11-NEXT: $vgpr3 = GLOBAL_LOAD_DWORD $vgpr0_vgpr1, 0, 0, implicit $exec
     ; GFX11-NEXT: $vgpr4 = GLOBAL_ATOMIC_ADD_RTN $vgpr0_vgpr1, $vgpr2, 4, 0, implicit $exec
+    ;
+    ; GFX12-LABEL: name: global_load_atomic
+    ; GFX12: liveins: $vgpr0_vgpr1, $vgpr2
+    ; GFX12-NEXT: {{  $}}
+    ; GFX12-NEXT: $vgpr3 = GLOBAL_LOAD_DWORD $vgpr0_vgpr1, 0, 0, implicit $exec
+    ; GFX12-NEXT: $vgpr4 = GLOBAL_ATOMIC_ADD_RTN $vgpr0_vgpr1, $vgpr2, 4, 0, implicit $exec
     $vgpr3 = GLOBAL_LOAD_DWORD $vgpr0_vgpr1, 0, 0, implicit $exec
     $vgpr4 = GLOBAL_ATOMIC_ADD_RTN $vgpr0_vgpr1, $vgpr2, 4, 0, implicit $exec
 ...
@@ -535,6 +638,12 @@ body: |
     ; GFX11-NEXT: {{  $}}
     ; GFX11-NEXT: $vgpr2 = FLAT_LOAD_DWORD $vgpr0_vgpr1, 0, 0, implicit $exec, implicit $flat_scr
     ; GFX11-NEXT: $vgpr3 = GLOBAL_LOAD_DWORD $vgpr0_vgpr1, 4, 0, implicit $exec, implicit $flat_scr
+    ;
+    ; GFX12-LABEL: name: flat_global_load
+    ; GFX12: liveins: $vgpr0_vgpr1
+    ; GFX12-NEXT: {{  $}}
+    ; GFX12-NEXT: $vgpr2 = FLAT_LOAD_DWORD $vgpr0_vgpr1, 0, 0, implicit $exec, implicit $flat_scr
+    ; GFX12-NEXT: $vgpr3 = GLOBAL_LOAD_DWORD $vgpr0_vgpr1, 4, 0, implicit $exec, implicit $flat_scr
     $vgpr2 = FLAT_LOAD_DWORD $vgpr0_vgpr1, 0, 0, implicit $exec, implicit $flat_scr
     $vgpr3 = GLOBAL_LOAD_DWORD $vgpr0_vgpr1, 4, 0, implicit $exec, implicit $flat_scr
 ...
@@ -559,6 +668,12 @@ body: |
     ; GFX11-NEXT: {{  $}}
     ; GFX11-NEXT: $vgpr1 = BUFFER_LOAD_DWORD_OFFSET $sgpr0_sgpr1_sgpr2_sgpr3, $sgpr4, 0, 0, 0, implicit $exec
     ; GFX11-NEXT: $vgpr0 = BUFFER_ATOMIC_ADD_OFFSET_RTN $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 4, 0, 0, implicit $exec
+    ;
+    ; GFX12-LABEL: name: buffer_load_atomic
+    ; GFX12: liveins: $sgpr0_sgpr1_sgpr2_sgpr3, $sgpr4, $vgpr0
+    ; GFX12-NEXT: {{  $}}
+    ; GFX12-NEXT: $vgpr1 = BUFFER_LOAD_DWORD_OFFSET $sgpr0_sgpr1_sgpr2_sgpr3, $sgpr4, 0, 0, 0, implicit $exec
+    ; GFX12-NEXT: $vgpr0 = BUFFER_ATOMIC_ADD_OFFSET_RTN $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 4, 0, 0, implicit $exec
     $vgpr1 = BUFFER_LOAD_DWORD_OFFSET $sgpr0_sgpr1_sgpr2_sgpr3, $sgpr4, 0, 0, 0, implicit $exec
     $vgpr0 = BUFFER_ATOMIC_ADD_OFFSET_RTN $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, 4, 0, 0, implicit $exec
 ...
@@ -580,6 +695,12 @@ body: |
     ; GFX11-NEXT: {{  $}}
     ; GFX11-NEXT: $vgpr3 = FLAT_LOAD_DWORD $vgpr0_vgpr1, 0, 0, implicit $exec, implicit $flat_scr
     ; GFX11-NEXT: FLAT_STORE_DWORD $vgpr0_vgpr1, $vgpr2, 4, 0, implicit $exec, implicit $flat_scr
+    ;
+    ; GFX12-LABEL: name: flat_load_store
+    ; GFX12: liveins: $vgpr0_vgpr1, $vgpr2
+    ; GFX12-NEXT: {{  $}}
+    ; GFX12-NEXT: $vgpr3 = FLAT_LOAD_DWORD $vgpr0_vgpr1, 0, 0, implicit $exec, implicit $flat_scr
+    ; GFX12-NEXT: FLAT_STORE_DWORD $vgpr0_vgpr1, $vgpr2, 4, 0, implicit $exec, implicit $flat_scr
     $vgpr3 = FLAT_LOAD_DWORD $vgpr0_vgpr1, 0, 0, implicit $exec, implicit $flat_scr
     FLAT_STORE_DWORD $vgpr0_vgpr1, $vgpr2, 4, 0, implicit $exec, implicit $flat_scr
 ...
@@ -601,6 +722,12 @@ body: |
     ; GFX11-NEXT: {{  $}}
     ; GFX11-NEXT: $vgpr3 = GLOBAL_LOAD_DWORD $vgpr0_vgpr1, 0, 0, implicit $exec
     ; GFX11-NEXT: GLOBAL_STORE_DWORD $vgpr0_vgpr1, $vgpr2, 4, 0, implicit $exec
+    ;
+    ; GFX12-LABEL: name: global_load_store
+    ; GFX12: liveins: $vgpr0_vgpr1, $vgpr2
+    ; GFX12-NEXT: {{  $}}
+    ; GFX12-NEXT: $vgpr3 = GLOBAL_LOAD_DWORD $vgpr0_vgpr1, 0, 0, implicit $exec
+    ; GFX12-NEXT: GLOBAL_STORE_DWORD $vgpr0_vgpr1, $vgpr2, 4, 0, implicit $exec
     $vgpr3 = GLOBAL_LOAD_DWORD $vgpr0_vgpr1, 0, 0, implicit $exec
     GLOBAL_STORE_DWORD $vgpr0_vgpr1, $vgpr2, 4, 0, implicit $exec
 ...
@@ -622,6 +749,12 @@ body: |
     ; GFX11-NEXT: {{  $}}
     ; GFX11-NEXT: $vgpr1 = BUFFER_LOAD_DWORD_OFFSET $sgpr0_sgpr1_sgpr2_sgpr3, $sgpr4, 0, 0, 0, implicit $exec
     ; GFX11-NEXT: BUFFER_STORE_DWORD_OFFSET $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, $sgpr4, 4, 0, 0, implicit $exec
+    ;
+    ; GFX12-LABEL: name: buffer_load_store
+    ; GFX12: liveins: $sgpr0_sgpr1_sgpr2_sgpr3, $sgpr4, $vgpr0
+    ; GFX12-NEXT: {{  $}}
+    ; GFX12-NEXT: $vgpr1 = BUFFER_LOAD_DWORD_OFFSET $sgpr0_sgpr1_sgpr2_sgpr3, $sgpr4, 0, 0, 0, implicit $exec
+    ; GFX12-NEXT: BUFFER_STORE_DWORD_OFFSET $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, $sgpr4, 4, 0, 0, implicit $exec
     $vgpr1 = BUFFER_LOAD_DWORD_OFFSET $sgpr0_sgpr1_sgpr2_sgpr3, $sgpr4, 0, 0, 0, implicit $exec
     BUFFER_STORE_DWORD_OFFSET $vgpr0, $sgpr0_sgpr1_sgpr2_sgpr3, $sgpr4, 4, 0, 0, implicit $exec
 ...
diff --git a/llvm/test/CodeGen/AMDGPU/live-interval-bug-in-rename-independent-subregs.mir b/llvm/test/CodeGen/AMDGPU/live-interval-bug-in-rename-independent-subregs.mir
new file mode 100644
index 000000000000..98b1b69101e5
--- /dev/null
+++ b/llvm/test/CodeGen/AMDGPU/live-interval-bug-in-rename-independent-subregs.mir
@@ -0,0 +1,208 @@
+# NOTE: Assertions have been autogenerated by utils/update_mir_test_checks.py UTC_ARGS: --version 4
+# RUN: llc -mtriple=amdgcn--amdpal -mcpu=gfx1010 -start-before=rename-independent-subregs -mattr=+wavefrontsize64 -stop-before=amdgpu-mark-last-scratch-load %s -o - | FileCheck -check-prefix=REG_ALLOC %s
+# RUN: llc -mtriple=amdgcn--amdpal -mcpu=gfx1010 -start-before=rename-independent-subregs -mattr=+wavefrontsize64 -stop-after=machine-cp %s -o - | FileCheck -check-prefix=DEAD_INST_DEL %s
+
+---
+name: _amdgpu_cs_main
+tracksRegLiveness: true
+body: |
+  ; REG_ALLOC-LABEL: name: _amdgpu_cs_main
+  ; REG_ALLOC: bb.0:
+  ; REG_ALLOC-NEXT:   successors: %bb.2(0x40000000), %bb.1(0x40000000)
+  ; REG_ALLOC-NEXT:   liveins: $sgpr0, $vgpr0, $vgpr1, $vgpr2, $vgpr3, $vgpr4, $vgpr10, $sgpr2_sgpr3, $sgpr4_sgpr5, $sgpr8_sgpr9_sgpr10_sgpr11
+  ; REG_ALLOC-NEXT: {{  $}}
+  ; REG_ALLOC-NEXT:   renamable $vgpr5_vgpr6_vgpr7_vgpr8 = BUFFER_LOAD_DWORDX4_OFFEN renamable $vgpr3, renamable $sgpr8_sgpr9_sgpr10_sgpr11, 0, 0, 0, 0, implicit $exec :: (dereferenceable load (<4 x s32>), align 1, addrspace 8)
+  ; REG_ALLOC-NEXT:   renamable $vgpr15_vgpr16_vgpr17_vgpr18 = BUFFER_LOAD_DWORDX4_OFFEN renamable $vgpr2, renamable $sgpr8_sgpr9_sgpr10_sgpr11, 0, 0, 0, 0, implicit $exec :: (dereferenceable load (<4 x s32>), align 1, addrspace 8)
+  ; REG_ALLOC-NEXT:   renamable $vgpr11_vgpr12_vgpr13_vgpr14 = BUFFER_LOAD_DWORDX4_OFFEN renamable $vgpr0, renamable $sgpr8_sgpr9_sgpr10_sgpr11, 0, 0, 0, 0, implicit $exec :: (dereferenceable load (<4 x s32>), align 1, addrspace 8)
+  ; REG_ALLOC-NEXT:   KILL killed renamable $vgpr2
+  ; REG_ALLOC-NEXT:   KILL killed renamable $vgpr0
+  ; REG_ALLOC-NEXT:   KILL killed renamable $vgpr3
+  ; REG_ALLOC-NEXT:   renamable $sgpr12 = V_READFIRSTLANE_B32 killed $vgpr5, implicit $exec
+  ; REG_ALLOC-NEXT:   renamable $vgpr4_vgpr5_vgpr6_vgpr7 = BUFFER_LOAD_DWORDX4_OFFEN killed renamable $vgpr4, killed renamable $sgpr8_sgpr9_sgpr10_sgpr11, 0, 0, 0, 0, implicit $exec :: (dereferenceable load (<4 x s32>), align 1, addrspace 8)
+  ; REG_ALLOC-NEXT:   renamable $sgpr13 = V_READFIRSTLANE_B32 killed $vgpr15, implicit $exec
+  ; REG_ALLOC-NEXT:   renamable $sgpr6_sgpr7 = V_CMP_NE_U32_e64 killed $vgpr1, 0, implicit $exec
+  ; REG_ALLOC-NEXT:   S_CMP_EQ_U64 killed renamable $sgpr12_sgpr13, killed renamable $sgpr2_sgpr3, implicit-def $scc
+  ; REG_ALLOC-NEXT:   renamable $sgpr1 = S_CSELECT_B32 1, 0, implicit $scc
+  ; REG_ALLOC-NEXT:   renamable $vgpr8 = IMPLICIT_DEF
+  ; REG_ALLOC-NEXT:   $exec = S_MOV_B64_term renamable $sgpr6_sgpr7
+  ; REG_ALLOC-NEXT:   S_CBRANCH_EXECZ %bb.1, implicit $exec
+  ; REG_ALLOC-NEXT:   S_BRANCH %bb.2
+  ; REG_ALLOC-NEXT: {{  $}}
+  ; REG_ALLOC-NEXT: bb.1:
+  ; REG_ALLOC-NEXT:   successors: %bb.3(0x40000000), %bb.5(0x40000000)
+  ; REG_ALLOC-NEXT:   liveins: $sgpr0, $vgpr8, $vgpr10, $sgpr4_sgpr5, $sgpr6_sgpr7, $vgpr11_vgpr12_vgpr13_vgpr14:0x0000000000000003, $vgpr0_vgpr1_vgpr2_vgpr3_vgpr4_vgpr5_vgpr6_vgpr7:0x0000000000000300
+  ; REG_ALLOC-NEXT: {{  $}}
+  ; REG_ALLOC-NEXT:   renamable $sgpr2_sgpr3 = S_OR_SAVEEXEC_B64 killed renamable $sgpr6_sgpr7, implicit-def $exec, implicit-def $scc, implicit $exec
+  ; REG_ALLOC-NEXT:   $exec = S_XOR_B64_term $exec, renamable $sgpr2_sgpr3, implicit-def $scc
+  ; REG_ALLOC-NEXT:   S_CBRANCH_EXECZ %bb.5, implicit $exec
+  ; REG_ALLOC-NEXT:   S_BRANCH %bb.3
+  ; REG_ALLOC-NEXT: {{  $}}
+  ; REG_ALLOC-NEXT: bb.2:
+  ; REG_ALLOC-NEXT:   successors: %bb.1(0x80000000)
+  ; REG_ALLOC-NEXT:   liveins: $sgpr0, $sgpr1, $vgpr10, $sgpr4_sgpr5, $sgpr6_sgpr7
+  ; REG_ALLOC-NEXT: {{  $}}
+  ; REG_ALLOC-NEXT:   renamable $sgpr1 = S_OR_B32 killed renamable $sgpr1, 2, implicit-def dead $scc
+  ; REG_ALLOC-NEXT:   renamable $vgpr8 = COPY killed renamable $sgpr1
+  ; REG_ALLOC-NEXT:   renamable $vgpr11_vgpr12 = IMPLICIT_DEF
+  ; REG_ALLOC-NEXT:   renamable $vgpr0_vgpr1_vgpr2_vgpr3_vgpr4_vgpr5_vgpr6_vgpr7 = IMPLICIT_DEF
+  ; REG_ALLOC-NEXT:   S_BRANCH %bb.1
+  ; REG_ALLOC-NEXT: {{  $}}
+  ; REG_ALLOC-NEXT: bb.3:
+  ; REG_ALLOC-NEXT:   successors: %bb.5(0x80000000)
+  ; REG_ALLOC-NEXT:   liveins: $sgpr0, $vgpr10, $sgpr2_sgpr3, $sgpr4_sgpr5, $vgpr11_vgpr12_vgpr13_vgpr14:0x0000000000000003, $vgpr0_vgpr1_vgpr2_vgpr3_vgpr4_vgpr5_vgpr6_vgpr7:0x0000000000000300
+  ; REG_ALLOC-NEXT: {{  $}}
+  ; REG_ALLOC-NEXT:   renamable $sgpr1 = V_READFIRSTLANE_B32 killed $vgpr11, implicit $exec
+  ; REG_ALLOC-NEXT:   renamable $sgpr6 = V_READFIRSTLANE_B32 killed $vgpr4, implicit $exec
+  ; REG_ALLOC-NEXT:   S_CMP_EQ_U32 killed renamable $sgpr6, killed renamable $sgpr1, implicit-def $scc
+  ; REG_ALLOC-NEXT:   renamable $sgpr1 = S_CSELECT_B32 1, 0, implicit $scc
+  ; REG_ALLOC-NEXT:   renamable $vgpr8 = COPY killed renamable $sgpr1
+  ; REG_ALLOC-NEXT:   S_BRANCH %bb.5
+  ; REG_ALLOC-NEXT: {{  $}}
+  ; REG_ALLOC-NEXT: bb.4:
+  ; REG_ALLOC-NEXT:   liveins: $sgpr0, $vgpr8, $vgpr10, $sgpr4_sgpr5
+  ; REG_ALLOC-NEXT: {{  $}}
+  ; REG_ALLOC-NEXT:   renamable $sgpr4_sgpr5_sgpr6_sgpr7 = S_LOAD_DWORDX4_IMM killed renamable $sgpr4_sgpr5, 0, 0 :: (invariant load (<4 x s32>), addrspace 4)
+  ; REG_ALLOC-NEXT:   renamable $vgpr0 = V_ADD_U32_e64 killed $sgpr0, killed $vgpr10, 0, implicit $exec
+  ; REG_ALLOC-NEXT:   BUFFER_STORE_DWORD_OFFEN_exact killed renamable $vgpr8, killed renamable $vgpr0, killed renamable $sgpr4_sgpr5_sgpr6_sgpr7, 0, 0, 0, 0, implicit $exec :: (dereferenceable store (s32), align 1, addrspace 8)
+  ; REG_ALLOC-NEXT:   S_ENDPGM 0
+  ; REG_ALLOC-NEXT: {{  $}}
+  ; REG_ALLOC-NEXT: bb.5:
+  ; REG_ALLOC-NEXT:   successors: %bb.4(0x80000000)
+  ; REG_ALLOC-NEXT:   liveins: $sgpr0, $vgpr8, $vgpr10, $sgpr2_sgpr3, $sgpr4_sgpr5
+  ; REG_ALLOC-NEXT: {{  $}}
+  ; REG_ALLOC-NEXT:   $exec = S_OR_B64 $exec, killed renamable $sgpr2_sgpr3, implicit-def $scc
+  ; REG_ALLOC-NEXT:   S_BRANCH %bb.4
+  ;
+  ; DEAD_INST_DEL-LABEL: name: _amdgpu_cs_main
+  ; DEAD_INST_DEL: bb.0:
+  ; DEAD_INST_DEL-NEXT:   successors: %bb.2(0x40000000), %bb.1(0x40000000)
+  ; DEAD_INST_DEL-NEXT:   liveins: $sgpr0, $vgpr0, $vgpr1, $vgpr2, $vgpr3, $vgpr4, $vgpr10, $sgpr2_sgpr3, $sgpr4_sgpr5, $sgpr8_sgpr9_sgpr10_sgpr11
+  ; DEAD_INST_DEL-NEXT: {{  $}}
+  ; DEAD_INST_DEL-NEXT:   renamable $vgpr5_vgpr6_vgpr7_vgpr8 = BUFFER_LOAD_DWORDX4_OFFEN renamable $vgpr3, renamable $sgpr8_sgpr9_sgpr10_sgpr11, 0, 0, 0, 0, implicit $exec :: (dereferenceable load (<4 x s32>), align 1, addrspace 8)
+  ; DEAD_INST_DEL-NEXT:   renamable $vgpr15_vgpr16_vgpr17_vgpr18 = BUFFER_LOAD_DWORDX4_OFFEN renamable $vgpr2, renamable $sgpr8_sgpr9_sgpr10_sgpr11, 0, 0, 0, 0, implicit $exec :: (dereferenceable load (<4 x s32>), align 1, addrspace 8)
+  ; DEAD_INST_DEL-NEXT:   renamable $vgpr11_vgpr12_vgpr13_vgpr14 = BUFFER_LOAD_DWORDX4_OFFEN renamable $vgpr0, renamable $sgpr8_sgpr9_sgpr10_sgpr11, 0, 0, 0, 0, implicit $exec :: (dereferenceable load (<4 x s32>), align 1, addrspace 8)
+  ; DEAD_INST_DEL-NEXT:   KILL killed renamable $vgpr2
+  ; DEAD_INST_DEL-NEXT:   KILL killed renamable $vgpr0
+  ; DEAD_INST_DEL-NEXT:   KILL killed renamable $vgpr3
+  ; DEAD_INST_DEL-NEXT:   renamable $sgpr12 = V_READFIRSTLANE_B32 killed $vgpr5, implicit $exec
+  ; DEAD_INST_DEL-NEXT:   renamable $vgpr4_vgpr5_vgpr6_vgpr7 = BUFFER_LOAD_DWORDX4_OFFEN killed renamable $vgpr4, killed renamable $sgpr8_sgpr9_sgpr10_sgpr11, 0, 0, 0, 0, implicit $exec :: (dereferenceable load (<4 x s32>), align 1, addrspace 8)
+  ; DEAD_INST_DEL-NEXT:   renamable $sgpr13 = V_READFIRSTLANE_B32 killed $vgpr15, implicit $exec
+  ; DEAD_INST_DEL-NEXT:   renamable $sgpr6_sgpr7 = V_CMP_NE_U32_e64 killed $vgpr1, 0, implicit $exec
+  ; DEAD_INST_DEL-NEXT:   S_CMP_EQ_U64 killed renamable $sgpr12_sgpr13, killed renamable $sgpr2_sgpr3, implicit-def $scc
+  ; DEAD_INST_DEL-NEXT:   renamable $sgpr1 = S_CSELECT_B32 1, 0, implicit $scc
+  ; DEAD_INST_DEL-NEXT:   renamable $vgpr8 = IMPLICIT_DEF
+  ; DEAD_INST_DEL-NEXT:   $exec = S_MOV_B64_term renamable $sgpr6_sgpr7
+  ; DEAD_INST_DEL-NEXT:   S_CBRANCH_EXECZ %bb.1, implicit $exec
+  ; DEAD_INST_DEL-NEXT:   S_BRANCH %bb.2
+  ; DEAD_INST_DEL-NEXT: {{  $}}
+  ; DEAD_INST_DEL-NEXT: bb.1:
+  ; DEAD_INST_DEL-NEXT:   successors: %bb.3(0x40000000), %bb.5(0x40000000)
+  ; DEAD_INST_DEL-NEXT:   liveins: $sgpr0, $vgpr8, $vgpr10, $sgpr4_sgpr5, $sgpr6_sgpr7, $vgpr11_vgpr12_vgpr13_vgpr14:0x0000000000000003, $vgpr0_vgpr1_vgpr2_vgpr3_vgpr4_vgpr5_vgpr6_vgpr7:0x0000000000000300
+  ; DEAD_INST_DEL-NEXT: {{  $}}
+  ; DEAD_INST_DEL-NEXT:   renamable $sgpr2_sgpr3 = S_OR_SAVEEXEC_B64 killed renamable $sgpr6_sgpr7, implicit-def $exec, implicit-def $scc, implicit $exec
+  ; DEAD_INST_DEL-NEXT:   $exec = S_XOR_B64_term $exec, renamable $sgpr2_sgpr3, implicit-def $scc
+  ; DEAD_INST_DEL-NEXT:   S_CBRANCH_EXECZ %bb.5, implicit $exec
+  ; DEAD_INST_DEL-NEXT:   S_BRANCH %bb.3
+  ; DEAD_INST_DEL-NEXT: {{  $}}
+  ; DEAD_INST_DEL-NEXT: bb.2:
+  ; DEAD_INST_DEL-NEXT:   successors: %bb.1(0x80000000)
+  ; DEAD_INST_DEL-NEXT:   liveins: $sgpr0, $sgpr1, $vgpr10, $sgpr4_sgpr5, $sgpr6_sgpr7
+  ; DEAD_INST_DEL-NEXT: {{  $}}
+  ; DEAD_INST_DEL-NEXT:   renamable $sgpr1 = S_OR_B32 killed renamable $sgpr1, 2, implicit-def dead $scc
+  ; DEAD_INST_DEL-NEXT:   renamable $vgpr8 = COPY killed renamable $sgpr1
+  ; DEAD_INST_DEL-NEXT:   renamable $vgpr11_vgpr12 = IMPLICIT_DEF
+  ; DEAD_INST_DEL-NEXT:   renamable $vgpr0_vgpr1_vgpr2_vgpr3_vgpr4_vgpr5_vgpr6_vgpr7 = IMPLICIT_DEF
+  ; DEAD_INST_DEL-NEXT:   S_BRANCH %bb.1
+  ; DEAD_INST_DEL-NEXT: {{  $}}
+  ; DEAD_INST_DEL-NEXT: bb.3:
+  ; DEAD_INST_DEL-NEXT:   successors: %bb.5(0x80000000)
+  ; DEAD_INST_DEL-NEXT:   liveins: $sgpr0, $vgpr10, $sgpr2_sgpr3, $sgpr4_sgpr5, $vgpr11_vgpr12_vgpr13_vgpr14:0x0000000000000003, $vgpr0_vgpr1_vgpr2_vgpr3_vgpr4_vgpr5_vgpr6_vgpr7:0x0000000000000300
+  ; DEAD_INST_DEL-NEXT: {{  $}}
+  ; DEAD_INST_DEL-NEXT:   renamable $sgpr1 = V_READFIRSTLANE_B32 killed $vgpr11, implicit $exec
+  ; DEAD_INST_DEL-NEXT:   renamable $sgpr6 = V_READFIRSTLANE_B32 killed $vgpr4, implicit $exec
+  ; DEAD_INST_DEL-NEXT:   S_CMP_EQ_U32 killed renamable $sgpr6, killed renamable $sgpr1, implicit-def $scc
+  ; DEAD_INST_DEL-NEXT:   renamable $sgpr1 = S_CSELECT_B32 1, 0, implicit $scc
+  ; DEAD_INST_DEL-NEXT:   renamable $vgpr8 = COPY killed renamable $sgpr1
+  ; DEAD_INST_DEL-NEXT:   S_BRANCH %bb.5
+  ; DEAD_INST_DEL-NEXT: {{  $}}
+  ; DEAD_INST_DEL-NEXT: bb.4:
+  ; DEAD_INST_DEL-NEXT:   liveins: $sgpr0, $vgpr8, $vgpr10, $sgpr4_sgpr5
+  ; DEAD_INST_DEL-NEXT: {{  $}}
+  ; DEAD_INST_DEL-NEXT:   renamable $sgpr4_sgpr5_sgpr6_sgpr7 = S_LOAD_DWORDX4_IMM killed renamable $sgpr4_sgpr5, 0, 0 :: (invariant load (<4 x s32>), addrspace 4)
+  ; DEAD_INST_DEL-NEXT:   renamable $vgpr0 = V_ADD_U32_e64 killed $sgpr0, killed $vgpr10, 0, implicit $exec
+  ; DEAD_INST_DEL-NEXT:   BUFFER_STORE_DWORD_OFFEN_exact killed renamable $vgpr8, killed renamable $vgpr0, killed renamable $sgpr4_sgpr5_sgpr6_sgpr7, 0, 0, 0, 0, implicit $exec :: (dereferenceable store (s32), align 1, addrspace 8)
+  ; DEAD_INST_DEL-NEXT:   S_ENDPGM 0
+  ; DEAD_INST_DEL-NEXT: {{  $}}
+  ; DEAD_INST_DEL-NEXT: bb.5:
+  ; DEAD_INST_DEL-NEXT:   successors: %bb.4(0x80000000)
+  ; DEAD_INST_DEL-NEXT:   liveins: $sgpr0, $vgpr8, $vgpr10, $sgpr2_sgpr3, $sgpr4_sgpr5
+  ; DEAD_INST_DEL-NEXT: {{  $}}
+  ; DEAD_INST_DEL-NEXT:   $exec = S_OR_B64 $exec, killed renamable $sgpr2_sgpr3, implicit-def $scc
+  ; DEAD_INST_DEL-NEXT:   S_BRANCH %bb.4
+  bb.0:
+    successors: %bb.2(0x40000000), %bb.1(0x40000000)
+    liveins: $sgpr0, $sgpr2_sgpr3, $sgpr4_sgpr5, $sgpr8_sgpr9_sgpr10_sgpr11, $vgpr0, $vgpr1, $vgpr2, $vgpr3, $vgpr4, $vgpr10
+
+    %0:sreg_32 = COPY $sgpr0
+    %1:vgpr_32 = COPY $vgpr0
+    %2:vgpr_32 = COPY $vgpr1
+    %3:vgpr_32 = COPY $vgpr2
+    %4:vgpr_32 = COPY $vgpr3
+    %5:vgpr_32 = COPY $vgpr4
+    %6:vgpr_32 = COPY $vgpr10
+    %7:sreg_64 = COPY $sgpr2_sgpr3
+    %8:sreg_64 = COPY $sgpr4_sgpr5
+    %9:sgpr_128 = COPY $sgpr8_sgpr9_sgpr10_sgpr11
+    undef %10.sub0_sub1_sub2_sub3:vreg_256 = BUFFER_LOAD_DWORDX4_OFFEN %1, %9, 0, 0, 0, 0, implicit $exec :: (dereferenceable load (<4 x s32>), align 1, addrspace 8)
+    %10.sub4_sub5_sub6_sub7:vreg_256 = BUFFER_LOAD_DWORDX4_OFFEN %5, %9, 0, 0, 0, 0, implicit $exec :: (dereferenceable load (<4 x s32>), align 1, addrspace 8)
+    %11:vreg_128 = BUFFER_LOAD_DWORDX4_OFFEN %4, %9, 0, 0, 0, 0, implicit $exec :: (dereferenceable load (<4 x s32>), align 1, addrspace 8)
+    %12:vreg_128 = BUFFER_LOAD_DWORDX4_OFFEN %3, %9, 0, 0, 0, 0, implicit $exec :: (dereferenceable load (<4 x s32>), align 1, addrspace 8)
+    undef %13.sub0:sgpr_256 = V_READFIRSTLANE_B32 %11.sub0, implicit $exec
+    %13.sub1:sgpr_256 = V_READFIRSTLANE_B32 %12.sub0, implicit $exec
+    S_CMP_EQ_U64 %13.sub0_sub1, %7, implicit-def $scc
+    %14:sreg_32 = S_CSELECT_B32 1, 0, implicit killed $scc
+    %15:sreg_64_xexec = V_CMP_NE_U32_e64 %2, 0, implicit $exec
+    %16:vgpr_32 = IMPLICIT_DEF
+    $exec = S_MOV_B64_term %15
+    S_CBRANCH_EXECZ %bb.1, implicit $exec
+    S_BRANCH %bb.2
+
+  bb.1:
+    successors: %bb.3(0x40000000), %bb.5(0x40000000)
+
+    %17:sreg_64 = S_OR_SAVEEXEC_B64 %15, implicit-def $exec, implicit-def $scc, implicit $exec
+    %18:sreg_64_xexec = S_AND_B64 $exec, %17, implicit-def $scc
+    $exec = S_XOR_B64_term $exec, %18, implicit-def $scc
+    S_CBRANCH_EXECZ %bb.5, implicit $exec
+    S_BRANCH %bb.3
+
+  bb.2:
+    successors: %bb.1(0x80000000)
+
+    %19:sreg_32 = S_OR_B32 %14, 2, implicit-def dead $scc
+    %16:vgpr_32 = COPY %19
+    undef %10.sub0_sub1:vreg_256 = IMPLICIT_DEF
+    S_BRANCH %bb.1
+
+  bb.3:
+    successors: %bb.5(0x80000000)
+
+    %20:sreg_32 = V_READFIRSTLANE_B32 %10.sub0, implicit $exec
+    %21:sreg_32 = V_READFIRSTLANE_B32 %10.sub4, implicit $exec
+    S_CMP_EQ_U32 %21, %20, implicit-def $scc
+    %22:sreg_32 = S_CSELECT_B32 1, 0, implicit killed $scc
+    %16:vgpr_32 = COPY %22
+    S_BRANCH %bb.5
+
+  bb.4:
+    %23:vgpr_32 = V_ADD_U32_e64 %0, %6, 0, implicit $exec
+    %24:sgpr_128 = S_LOAD_DWORDX4_IMM %8, 0, 0 :: (invariant load (<4 x s32>), addrspace 4)
+    BUFFER_STORE_DWORD_OFFEN_exact %16, %23, %24, 0, 0, 0, 0, implicit $exec :: (dereferenceable store (s32), align 1, addrspace 8)
+    S_ENDPGM 0
+
+  bb.5:
+    successors: %bb.4(0x80000000)
+
+    $exec = S_OR_B64 $exec, %18, implicit-def $scc
+    S_BRANCH %bb.4
+...
diff --git a/llvm/test/CodeGen/AMDGPU/llvm.amdgcn.cvt.fp8.dpp.ll b/llvm/test/CodeGen/AMDGPU/llvm.amdgcn.cvt.fp8.dpp.ll
index e21d61036375..ffedde9416bb 100644
--- a/llvm/test/CodeGen/AMDGPU/llvm.amdgcn.cvt.fp8.dpp.ll
+++ b/llvm/test/CodeGen/AMDGPU/llvm.amdgcn.cvt.fp8.dpp.ll
@@ -97,9 +97,7 @@ define amdgpu_cs void @test_cvt_sr_bf8_f32_byte0(i32 %a, i32 %r, i32 %old, ptr a
 define amdgpu_cs void @test_cvt_sr_fp8_f32_byte1(i32 %a, i32 %r, i32 %old, ptr addrspace(1) %out) {
 ; GFX12-LABEL: test_cvt_sr_fp8_f32_byte1:
 ; GFX12:       ; %bb.0:
-; GFX12-NEXT:    v_mov_b32_dpp v0, v0 quad_perm:[0,1,2,3] row_mask:0xf bank_mask:0xf bound_ctrl:1
-; GFX12-NEXT:    s_delay_alu instid0(VALU_DEP_1)
-; GFX12-NEXT:    v_cvt_sr_fp8_f32 v2, v0, v1 op_sel:[0,0,1,0]
+; GFX12-NEXT:    v_cvt_sr_fp8_f32_e64_dpp v2, v0, v1 byte_sel:1 quad_perm:[0,1,2,3] row_mask:0xf bank_mask:0xf bound_ctrl:1
 ; GFX12-NEXT:    global_store_b32 v[3:4], v2, off
 ; GFX12-NEXT:    s_nop 0
 ; GFX12-NEXT:    s_sendmsg sendmsg(MSG_DEALLOC_VGPRS)
@@ -114,9 +112,7 @@ define amdgpu_cs void @test_cvt_sr_fp8_f32_byte1(i32 %a, i32 %r, i32 %old, ptr a
 define amdgpu_cs void @test_cvt_sr_fp8_f32_byte2(i32 %a, i32 %r, i32 %old, ptr addrspace(1) %out) {
 ; GFX12-LABEL: test_cvt_sr_fp8_f32_byte2:
 ; GFX12:       ; %bb.0:
-; GFX12-NEXT:    v_mov_b32_dpp v0, v0 quad_perm:[0,1,2,3] row_mask:0xf bank_mask:0xf bound_ctrl:1
-; GFX12-NEXT:    s_delay_alu instid0(VALU_DEP_1)
-; GFX12-NEXT:    v_cvt_sr_fp8_f32 v2, v0, v1 op_sel:[0,0,0,1]
+; GFX12-NEXT:    v_cvt_sr_fp8_f32_e64_dpp v2, v0, v1 byte_sel:2 quad_perm:[0,1,2,3] row_mask:0xf bank_mask:0xf bound_ctrl:1
 ; GFX12-NEXT:    global_store_b32 v[3:4], v2, off
 ; GFX12-NEXT:    s_nop 0
 ; GFX12-NEXT:    s_sendmsg sendmsg(MSG_DEALLOC_VGPRS)
diff --git a/llvm/test/CodeGen/AMDGPU/llvm.amdgcn.cvt.fp8.ll b/llvm/test/CodeGen/AMDGPU/llvm.amdgcn.cvt.fp8.ll
index 9b8fdf901704..7662a3b78dea 100644
--- a/llvm/test/CodeGen/AMDGPU/llvm.amdgcn.cvt.fp8.ll
+++ b/llvm/test/CodeGen/AMDGPU/llvm.amdgcn.cvt.fp8.ll
@@ -385,7 +385,7 @@ define i32 @test_cvt_sr_bf8_f32_byte1(float %x, i32 %r, i32 %old) {
 ; GFX12-NEXT:    s_wait_samplecnt 0x0
 ; GFX12-NEXT:    s_wait_bvhcnt 0x0
 ; GFX12-NEXT:    s_wait_kmcnt 0x0
-; GFX12-NEXT:    v_cvt_sr_bf8_f32 v2, v0, v1 op_sel:[0,0,1,0]
+; GFX12-NEXT:    v_cvt_sr_bf8_f32 v2, v0, v1 byte_sel:1
 ; GFX12-NEXT:    s_delay_alu instid0(VALU_DEP_1)
 ; GFX12-NEXT:    v_mov_b32_e32 v0, v2
 ; GFX12-NEXT:    s_setpc_b64 s[30:31]
@@ -409,7 +409,7 @@ define i32 @test_cvt_sr_bf8_f32_byte2(float %x, i32 %r, i32 %old) {
 ; GFX12-NEXT:    s_wait_samplecnt 0x0
 ; GFX12-NEXT:    s_wait_bvhcnt 0x0
 ; GFX12-NEXT:    s_wait_kmcnt 0x0
-; GFX12-NEXT:    v_cvt_sr_bf8_f32 v2, v0, v1 op_sel:[0,0,0,1]
+; GFX12-NEXT:    v_cvt_sr_bf8_f32 v2, v0, v1 byte_sel:2
 ; GFX12-NEXT:    s_delay_alu instid0(VALU_DEP_1)
 ; GFX12-NEXT:    v_mov_b32_e32 v0, v2
 ; GFX12-NEXT:    s_setpc_b64 s[30:31]
@@ -433,7 +433,7 @@ define i32 @test_cvt_sr_bf8_f32_byte3(float %x, i32 %r, i32 %old) {
 ; GFX12-NEXT:    s_wait_samplecnt 0x0
 ; GFX12-NEXT:    s_wait_bvhcnt 0x0
 ; GFX12-NEXT:    s_wait_kmcnt 0x0
-; GFX12-NEXT:    v_cvt_sr_bf8_f32 v2, v0, v1 op_sel:[0,0,1,1]
+; GFX12-NEXT:    v_cvt_sr_bf8_f32 v2, v0, v1 byte_sel:3
 ; GFX12-NEXT:    s_delay_alu instid0(VALU_DEP_1)
 ; GFX12-NEXT:    v_mov_b32_e32 v0, v2
 ; GFX12-NEXT:    s_setpc_b64 s[30:31]
@@ -479,7 +479,7 @@ define i32 @test_cvt_sr_fp8_f32_byte1(float %x, i32 %r, i32 %old) {
 ; GFX12-NEXT:    s_wait_samplecnt 0x0
 ; GFX12-NEXT:    s_wait_bvhcnt 0x0
 ; GFX12-NEXT:    s_wait_kmcnt 0x0
-; GFX12-NEXT:    v_cvt_sr_fp8_f32 v2, v0, v1 op_sel:[0,0,1,0]
+; GFX12-NEXT:    v_cvt_sr_fp8_f32 v2, v0, v1 byte_sel:1
 ; GFX12-NEXT:    s_delay_alu instid0(VALU_DEP_1)
 ; GFX12-NEXT:    v_mov_b32_e32 v0, v2
 ; GFX12-NEXT:    s_setpc_b64 s[30:31]
@@ -503,7 +503,7 @@ define i32 @test_cvt_sr_fp8_f32_byte2(float %x, i32 %r, i32 %old) {
 ; GFX12-NEXT:    s_wait_samplecnt 0x0
 ; GFX12-NEXT:    s_wait_bvhcnt 0x0
 ; GFX12-NEXT:    s_wait_kmcnt 0x0
-; GFX12-NEXT:    v_cvt_sr_fp8_f32 v2, v0, v1 op_sel:[0,0,0,1]
+; GFX12-NEXT:    v_cvt_sr_fp8_f32 v2, v0, v1 byte_sel:2
 ; GFX12-NEXT:    s_delay_alu instid0(VALU_DEP_1)
 ; GFX12-NEXT:    v_mov_b32_e32 v0, v2
 ; GFX12-NEXT:    s_setpc_b64 s[30:31]
@@ -527,7 +527,7 @@ define i32 @test_cvt_sr_fp8_f32_byte3(float %x, i32 %r, i32 %old) {
 ; GFX12-NEXT:    s_wait_samplecnt 0x0
 ; GFX12-NEXT:    s_wait_bvhcnt 0x0
 ; GFX12-NEXT:    s_wait_kmcnt 0x0
-; GFX12-NEXT:    v_cvt_sr_fp8_f32 v2, v0, v1 op_sel:[0,0,1,1]
+; GFX12-NEXT:    v_cvt_sr_fp8_f32 v2, v0, v1 byte_sel:3
 ; GFX12-NEXT:    s_delay_alu instid0(VALU_DEP_1)
 ; GFX12-NEXT:    v_mov_b32_e32 v0, v2
 ; GFX12-NEXT:    s_setpc_b64 s[30:31]
diff --git a/llvm/test/CodeGen/AMDGPU/llvm.amdgcn.s.wait.event.ll b/llvm/test/CodeGen/AMDGPU/llvm.amdgcn.s.wait.event.ll
index 08c77148f6ae..433fefa43498 100644
--- a/llvm/test/CodeGen/AMDGPU/llvm.amdgcn.s.wait.event.ll
+++ b/llvm/test/CodeGen/AMDGPU/llvm.amdgcn.s.wait.event.ll
@@ -5,14 +5,10 @@
 
 ; GCN-LABEL: {{^}}test_wait_event:
 ; GFX11: s_wait_event 0x0
-; GFX12: s_wait_event 0x1
+; GFX12: s_wait_event 0x2
 
-define amdgpu_ps void @test_wait_event() #0 {
+define amdgpu_ps void @test_wait_event() {
 entry:
-  call void @llvm.amdgcn.s.wait.event.export.ready() #0
+  call void @llvm.amdgcn.s.wait.event.export.ready()
   ret void
 }
-
-declare void @llvm.amdgcn.s.wait.event.export.ready() #0
-
-attributes #0 = { nounwind }
diff --git a/llvm/test/CodeGen/AMDGPU/llvm.fptrunc.round.ll b/llvm/test/CodeGen/AMDGPU/llvm.fptrunc.round.ll
index a113c07b3e2b..b8c16d2ed3b2 100644
--- a/llvm/test/CodeGen/AMDGPU/llvm.fptrunc.round.ll
+++ b/llvm/test/CodeGen/AMDGPU/llvm.fptrunc.round.ll
@@ -1,52 +1,453 @@
 ; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
-; RUN: llc -mtriple=amdgcn -mcpu=gfx1030 -verify-machineinstrs < %s | FileCheck %s
-; RUN: llc -mtriple=amdgcn -mcpu=gfx1010 -verify-machineinstrs < %s | FileCheck %s
-; RUN: llc -global-isel -mtriple=amdgcn -mcpu=gfx1030 -verify-machineinstrs < %s | FileCheck %s
+; RUN: llc -global-isel=0 -mtriple=amdgcn -mcpu=gfx1030 < %s | FileCheck -check-prefixes=CHECK,SDAG %s
+; RUN: llc -global-isel=0 -mtriple=amdgcn -mcpu=gfx1010 < %s | FileCheck -check-prefixes=CHECK,SDAG %s
+; RUN: llc -global-isel -mtriple=amdgcn -mcpu=gfx1030 < %s | FileCheck -check-prefixes=CHECK,GISEL %s
 
-define amdgpu_gs void @test_fptrunc_round_upward(float %a, i32 %data0, <4 x i32> %data1, ptr addrspace(1) %out) {
-; CHECK-LABEL: test_fptrunc_round_upward:
+define amdgpu_gs half @v_fptrunc_round_f32_to_f16_upward(float %a) {
+; CHECK-LABEL: v_fptrunc_round_f32_to_f16_upward:
 ; CHECK:       ; %bb.0:
 ; CHECK-NEXT:    s_setreg_imm32_b32 hwreg(HW_REG_MODE, 2, 1), 1
 ; CHECK-NEXT:    v_cvt_f16_f32_e32 v0, v0
-; CHECK-NEXT:    global_store_short v[6:7], v0, off
-; CHECK-NEXT:    s_endpgm
-  %res = call half @llvm.fptrunc.round(float %a, metadata !"round.upward")
-  store half %res, ptr addrspace(1) %out, align 4
-  ret void
+; CHECK-NEXT:    ; return to shader part epilog
+  %res = call half @llvm.fptrunc.round.f16.f32(float %a, metadata !"round.upward")
+  ret half %res
 }
 
-define amdgpu_gs void @test_fptrunc_round_downward(float %a, i32 %data0, <4 x i32> %data1, ptr addrspace(1) %out) {
-; CHECK-LABEL: test_fptrunc_round_downward:
+define amdgpu_gs half @v_fptrunc_round_f32_to_f16_downward(float %a) {
+; CHECK-LABEL: v_fptrunc_round_f32_to_f16_downward:
 ; CHECK:       ; %bb.0:
 ; CHECK-NEXT:    s_setreg_imm32_b32 hwreg(HW_REG_MODE, 3, 1), 1
 ; CHECK-NEXT:    v_cvt_f16_f32_e32 v0, v0
-; CHECK-NEXT:    global_store_short v[6:7], v0, off
-; CHECK-NEXT:    s_endpgm
-  %res = call half @llvm.fptrunc.round(float %a, metadata !"round.downward")
-  store half %res, ptr addrspace(1) %out, align 4
-  ret void
+; CHECK-NEXT:    ; return to shader part epilog
+  %res = call half @llvm.fptrunc.round.f16.f32(float %a, metadata !"round.downward")
+  ret half %res
 }
 
-define amdgpu_gs void @test_fptrunc_round_upward_multiple_calls(float %a, float %b, i32 %data0, <4 x i32> %data1, ptr addrspace(1) %out) {
-; CHECK-LABEL: test_fptrunc_round_upward_multiple_calls:
+define amdgpu_gs void @v_fptrunc_round_f32_to_f16_upward_multiple_calls(float %a, float %b, ptr addrspace(1) %out) {
+; CHECK-LABEL: v_fptrunc_round_f32_to_f16_upward_multiple_calls:
 ; CHECK:       ; %bb.0:
 ; CHECK-NEXT:    s_setreg_imm32_b32 hwreg(HW_REG_MODE, 2, 1), 1
 ; CHECK-NEXT:    v_cvt_f16_f32_e32 v0, v0
-; CHECK-NEXT:    v_cvt_f16_f32_e32 v2, v1
+; CHECK-NEXT:    v_cvt_f16_f32_e32 v4, v1
 ; CHECK-NEXT:    s_setreg_imm32_b32 hwreg(HW_REG_MODE, 2, 2), 2
 ; CHECK-NEXT:    v_cvt_f16_f32_e32 v1, v1
 ; CHECK-NEXT:    s_setreg_imm32_b32 hwreg(HW_REG_MODE, 3, 1), 0
-; CHECK-NEXT:    v_add_f16_e32 v0, v0, v2
+; CHECK-NEXT:    v_add_f16_e32 v0, v0, v4
 ; CHECK-NEXT:    v_add_f16_e32 v0, v1, v0
-; CHECK-NEXT:    global_store_short v[7:8], v0, off
+; CHECK-NEXT:    global_store_short v[2:3], v0, off
+; CHECK-NEXT:    s_endpgm
+  %res1 = call half @llvm.fptrunc.round.f16.f32(float %a, metadata !"round.upward")
+  %res2 = call half @llvm.fptrunc.round.f16.f32(float %b, metadata !"round.upward")
+  %res3 = call half @llvm.fptrunc.round.f16.f32(float %b, metadata !"round.downward")
+  %res4 = fadd half %res1, %res2
+  %res5 = fadd half %res3, %res4
+  store half %res5, ptr addrspace(1) %out, align 4
+  ret void
+}
+
+define amdgpu_gs i32 @s_fptrunc_round_f32_to_f16_upward(float inreg %a, ptr addrspace(1) %out) {
+; CHECK-LABEL: s_fptrunc_round_f32_to_f16_upward:
+; CHECK:       ; %bb.0:
+; CHECK-NEXT:    v_mov_b32_e32 v0, s0
+; CHECK-NEXT:    s_setreg_imm32_b32 hwreg(HW_REG_MODE, 2, 1), 1
+; CHECK-NEXT:    v_cvt_f16_f32_e32 v0, v0
+; CHECK-NEXT:    v_and_b32_e32 v0, 0xffff, v0
+; CHECK-NEXT:    v_readfirstlane_b32 s0, v0
+; CHECK-NEXT:    ; return to shader part epilog
+  %res = call half @llvm.fptrunc.round.f16.f32(float %a, metadata !"round.upward")
+  %bitcast = bitcast half %res to i16
+  %ret = zext i16 %bitcast to i32
+  ret i32 %ret
+}
+
+define amdgpu_gs i32 @s_fptrunc_round_f32_to_f16_downward(float inreg %a, ptr addrspace(1) %out) {
+; CHECK-LABEL: s_fptrunc_round_f32_to_f16_downward:
+; CHECK:       ; %bb.0:
+; CHECK-NEXT:    v_mov_b32_e32 v0, s0
+; CHECK-NEXT:    s_setreg_imm32_b32 hwreg(HW_REG_MODE, 3, 1), 1
+; CHECK-NEXT:    v_cvt_f16_f32_e32 v0, v0
+; CHECK-NEXT:    v_and_b32_e32 v0, 0xffff, v0
+; CHECK-NEXT:    v_readfirstlane_b32 s0, v0
+; CHECK-NEXT:    ; return to shader part epilog
+  %res = call half @llvm.fptrunc.round.f16.f32(float %a, metadata !"round.downward")
+  %bitcast = bitcast half %res to i16
+  %ret = zext i16 %bitcast to i32
+  ret i32 %ret
+}
+
+define amdgpu_gs void @s_fptrunc_round_f32_to_f16_upward_multiple_calls(float inreg %a, float inreg %b, ptr addrspace(1) %out) {
+; CHECK-LABEL: s_fptrunc_round_f32_to_f16_upward_multiple_calls:
+; CHECK:       ; %bb.0:
+; CHECK-NEXT:    v_mov_b32_e32 v2, s0
+; CHECK-NEXT:    v_mov_b32_e32 v3, s1
+; CHECK-NEXT:    s_setreg_imm32_b32 hwreg(HW_REG_MODE, 2, 1), 1
+; CHECK-NEXT:    v_cvt_f16_f32_e32 v2, v2
+; CHECK-NEXT:    v_cvt_f16_f32_e32 v4, v3
+; CHECK-NEXT:    s_setreg_imm32_b32 hwreg(HW_REG_MODE, 2, 2), 2
+; CHECK-NEXT:    v_cvt_f16_f32_e32 v3, v3
+; CHECK-NEXT:    s_setreg_imm32_b32 hwreg(HW_REG_MODE, 3, 1), 0
+; CHECK-NEXT:    v_add_f16_e32 v2, v2, v4
+; CHECK-NEXT:    v_add_f16_e32 v2, v3, v2
+; CHECK-NEXT:    global_store_short v[0:1], v2, off
 ; CHECK-NEXT:    s_endpgm
-  %res1 = call half @llvm.fptrunc.round(float %a, metadata !"round.upward")
-  %res2 = call half @llvm.fptrunc.round(float %b, metadata !"round.upward")
-  %res3 = call half @llvm.fptrunc.round(float %b, metadata !"round.downward")
+  %res1 = call half @llvm.fptrunc.round.f16.f32(float %a, metadata !"round.upward")
+  %res2 = call half @llvm.fptrunc.round.f16.f32(float %b, metadata !"round.upward")
+  %res3 = call half @llvm.fptrunc.round.f16.f32(float %b, metadata !"round.downward")
   %res4 = fadd half %res1, %res2
   %res5 = fadd half %res3, %res4
   store half %res5, ptr addrspace(1) %out, align 4
   ret void
 }
 
-declare half @llvm.fptrunc.round(float, metadata)
+define amdgpu_gs <2 x half> @v_fptrunc_round_v2f32_to_v2f16_upward(<2 x float> %a) {
+; SDAG-LABEL: v_fptrunc_round_v2f32_to_v2f16_upward:
+; SDAG:       ; %bb.0:
+; SDAG-NEXT:    s_setreg_imm32_b32 hwreg(HW_REG_MODE, 2, 1), 1
+; SDAG-NEXT:    v_cvt_f16_f32_e32 v0, v0
+; SDAG-NEXT:    v_cvt_f16_f32_e32 v1, v1
+; SDAG-NEXT:    v_perm_b32 v0, v1, v0, 0x5040100
+; SDAG-NEXT:    ; return to shader part epilog
+;
+; GISEL-LABEL: v_fptrunc_round_v2f32_to_v2f16_upward:
+; GISEL:       ; %bb.0:
+; GISEL-NEXT:    s_setreg_imm32_b32 hwreg(HW_REG_MODE, 2, 1), 1
+; GISEL-NEXT:    v_cvt_f16_f32_e32 v0, v0
+; GISEL-NEXT:    v_cvt_f16_f32_e32 v1, v1
+; GISEL-NEXT:    v_and_b32_e32 v0, 0xffff, v0
+; GISEL-NEXT:    v_lshl_or_b32 v0, v1, 16, v0
+; GISEL-NEXT:    ; return to shader part epilog
+  %res = call <2 x half> @llvm.fptrunc.round.v2f16.v2f32(<2 x float> %a, metadata !"round.upward")
+  ret <2 x half> %res
+}
+
+define amdgpu_gs <2 x half> @v_fptrunc_round_v2f32_to_v2f16_downward(<2 x float> %a) {
+; SDAG-LABEL: v_fptrunc_round_v2f32_to_v2f16_downward:
+; SDAG:       ; %bb.0:
+; SDAG-NEXT:    s_setreg_imm32_b32 hwreg(HW_REG_MODE, 3, 1), 1
+; SDAG-NEXT:    v_cvt_f16_f32_e32 v0, v0
+; SDAG-NEXT:    v_cvt_f16_f32_e32 v1, v1
+; SDAG-NEXT:    v_perm_b32 v0, v1, v0, 0x5040100
+; SDAG-NEXT:    ; return to shader part epilog
+;
+; GISEL-LABEL: v_fptrunc_round_v2f32_to_v2f16_downward:
+; GISEL:       ; %bb.0:
+; GISEL-NEXT:    s_setreg_imm32_b32 hwreg(HW_REG_MODE, 3, 1), 1
+; GISEL-NEXT:    v_cvt_f16_f32_e32 v0, v0
+; GISEL-NEXT:    v_cvt_f16_f32_e32 v1, v1
+; GISEL-NEXT:    v_and_b32_e32 v0, 0xffff, v0
+; GISEL-NEXT:    v_lshl_or_b32 v0, v1, 16, v0
+; GISEL-NEXT:    ; return to shader part epilog
+  %res = call <2 x half> @llvm.fptrunc.round.v2f16.v2f32(<2 x float> %a, metadata !"round.downward")
+  ret <2 x half> %res
+}
+
+define amdgpu_gs void @v_fptrunc_round_v2f32_to_v2f16_upward_multiple_calls(<2 x float> %a, <2 x float> %b, ptr addrspace(1) %out) {
+; SDAG-LABEL: v_fptrunc_round_v2f32_to_v2f16_upward_multiple_calls:
+; SDAG:       ; %bb.0:
+; SDAG-NEXT:    s_setreg_imm32_b32 hwreg(HW_REG_MODE, 2, 1), 1
+; SDAG-NEXT:    v_cvt_f16_f32_e32 v0, v0
+; SDAG-NEXT:    v_cvt_f16_f32_e32 v1, v1
+; SDAG-NEXT:    v_cvt_f16_f32_e32 v6, v2
+; SDAG-NEXT:    v_cvt_f16_f32_e32 v7, v3
+; SDAG-NEXT:    v_perm_b32 v0, v1, v0, 0x5040100
+; SDAG-NEXT:    s_setreg_imm32_b32 hwreg(HW_REG_MODE, 2, 2), 2
+; SDAG-NEXT:    v_cvt_f16_f32_e32 v1, v2
+; SDAG-NEXT:    v_cvt_f16_f32_e32 v2, v3
+; SDAG-NEXT:    v_perm_b32 v3, v7, v6, 0x5040100
+; SDAG-NEXT:    v_perm_b32 v1, v2, v1, 0x5040100
+; SDAG-NEXT:    s_setreg_imm32_b32 hwreg(HW_REG_MODE, 3, 1), 0
+; SDAG-NEXT:    v_pk_add_f16 v0, v0, v3
+; SDAG-NEXT:    v_pk_add_f16 v0, v1, v0
+; SDAG-NEXT:    global_store_dword v[4:5], v0, off
+; SDAG-NEXT:    s_endpgm
+;
+; GISEL-LABEL: v_fptrunc_round_v2f32_to_v2f16_upward_multiple_calls:
+; GISEL:       ; %bb.0:
+; GISEL-NEXT:    s_setreg_imm32_b32 hwreg(HW_REG_MODE, 2, 1), 1
+; GISEL-NEXT:    v_cvt_f16_f32_e32 v0, v0
+; GISEL-NEXT:    v_cvt_f16_f32_e32 v6, v2
+; GISEL-NEXT:    v_cvt_f16_f32_e32 v1, v1
+; GISEL-NEXT:    s_setreg_imm32_b32 hwreg(HW_REG_MODE, 2, 2), 2
+; GISEL-NEXT:    v_cvt_f16_f32_e32 v2, v2
+; GISEL-NEXT:    s_setreg_imm32_b32 hwreg(HW_REG_MODE, 2, 2), 1
+; GISEL-NEXT:    v_cvt_f16_f32_e32 v7, v3
+; GISEL-NEXT:    v_and_b32_e32 v0, 0xffff, v0
+; GISEL-NEXT:    v_and_b32_e32 v6, 0xffff, v6
+; GISEL-NEXT:    s_setreg_imm32_b32 hwreg(HW_REG_MODE, 2, 2), 2
+; GISEL-NEXT:    v_cvt_f16_f32_e32 v3, v3
+; GISEL-NEXT:    v_and_b32_e32 v2, 0xffff, v2
+; GISEL-NEXT:    v_lshl_or_b32 v0, v1, 16, v0
+; GISEL-NEXT:    v_lshl_or_b32 v1, v7, 16, v6
+; GISEL-NEXT:    v_lshl_or_b32 v2, v3, 16, v2
+; GISEL-NEXT:    s_setreg_imm32_b32 hwreg(HW_REG_MODE, 3, 1), 0
+; GISEL-NEXT:    v_pk_add_f16 v0, v0, v1
+; GISEL-NEXT:    v_pk_add_f16 v0, v2, v0
+; GISEL-NEXT:    global_store_dword v[4:5], v0, off
+; GISEL-NEXT:    s_endpgm
+  %res1 = call <2 x half> @llvm.fptrunc.round.v2f16.v2f32(<2 x float> %a, metadata !"round.upward")
+  %res2 = call <2 x half> @llvm.fptrunc.round.v2f16.v2f32(<2 x float> %b, metadata !"round.upward")
+  %res3 = call <2 x half> @llvm.fptrunc.round.v2f16.v2f32(<2 x float> %b, metadata !"round.downward")
+  %res4 = fadd <2 x half> %res1, %res2
+  %res5 = fadd <2 x half> %res3, %res4
+  store <2 x half> %res5, ptr addrspace(1) %out, align 4
+  ret void
+}
+
+define amdgpu_gs <2 x i32> @s_fptrunc_round_v2f32_to_v2f16_upward(<2 x float> inreg %a, ptr addrspace(1) %out) {
+; CHECK-LABEL: s_fptrunc_round_v2f32_to_v2f16_upward:
+; CHECK:       ; %bb.0:
+; CHECK-NEXT:    v_mov_b32_e32 v0, s0
+; CHECK-NEXT:    v_mov_b32_e32 v1, s1
+; CHECK-NEXT:    s_setreg_imm32_b32 hwreg(HW_REG_MODE, 2, 1), 1
+; CHECK-NEXT:    v_cvt_f16_f32_e32 v0, v0
+; CHECK-NEXT:    v_cvt_f16_f32_e32 v1, v1
+; CHECK-NEXT:    v_and_b32_e32 v0, 0xffff, v0
+; CHECK-NEXT:    v_and_b32_e32 v1, 0xffff, v1
+; CHECK-NEXT:    v_readfirstlane_b32 s0, v0
+; CHECK-NEXT:    v_readfirstlane_b32 s1, v1
+; CHECK-NEXT:    ; return to shader part epilog
+  %res = call <2 x half> @llvm.fptrunc.round.v2f16.v2f32(<2 x float> %a, metadata !"round.upward")
+  %bitcast = bitcast <2 x half> %res to <2 x i16>
+  %ret = zext <2 x i16> %bitcast to <2 x i32>
+  ret <2 x i32> %ret
+}
+
+define amdgpu_gs <2 x i32> @s_fptrunc_round_v2f32_to_v2f16_downward(<2 x float> inreg %a, ptr addrspace(1) %out) {
+; CHECK-LABEL: s_fptrunc_round_v2f32_to_v2f16_downward:
+; CHECK:       ; %bb.0:
+; CHECK-NEXT:    v_mov_b32_e32 v0, s0
+; CHECK-NEXT:    v_mov_b32_e32 v1, s1
+; CHECK-NEXT:    s_setreg_imm32_b32 hwreg(HW_REG_MODE, 3, 1), 1
+; CHECK-NEXT:    v_cvt_f16_f32_e32 v0, v0
+; CHECK-NEXT:    v_cvt_f16_f32_e32 v1, v1
+; CHECK-NEXT:    v_and_b32_e32 v0, 0xffff, v0
+; CHECK-NEXT:    v_and_b32_e32 v1, 0xffff, v1
+; CHECK-NEXT:    v_readfirstlane_b32 s0, v0
+; CHECK-NEXT:    v_readfirstlane_b32 s1, v1
+; CHECK-NEXT:    ; return to shader part epilog
+  %res = call <2 x half> @llvm.fptrunc.round.v2f16.v2f32(<2 x float> %a, metadata !"round.downward")
+  %bitcast = bitcast <2 x half> %res to <2 x i16>
+  %ret = zext <2 x i16> %bitcast to <2 x i32>
+  ret <2 x i32> %ret
+}
+
+define amdgpu_gs void @s_fptrunc_round_v2f32_to_v2f16_upward_multiple_calls(<2 x float> inreg %a, <2 x float> inreg %b, ptr addrspace(1) %out) {
+; CHECK-LABEL: s_fptrunc_round_v2f32_to_v2f16_upward_multiple_calls:
+; CHECK:       ; %bb.0:
+; CHECK-NEXT:    v_mov_b32_e32 v2, s0
+; CHECK-NEXT:    v_mov_b32_e32 v3, s2
+; CHECK-NEXT:    v_mov_b32_e32 v4, s1
+; CHECK-NEXT:    v_mov_b32_e32 v5, s3
+; CHECK-NEXT:    s_setreg_imm32_b32 hwreg(HW_REG_MODE, 2, 1), 1
+; CHECK-NEXT:    v_cvt_f16_f32_e32 v2, v2
+; CHECK-NEXT:    v_cvt_f16_f32_e32 v6, v3
+; CHECK-NEXT:    v_cvt_f16_f32_e32 v4, v4
+; CHECK-NEXT:    v_cvt_f16_f32_e32 v7, v5
+; CHECK-NEXT:    s_setreg_imm32_b32 hwreg(HW_REG_MODE, 2, 2), 2
+; CHECK-NEXT:    v_cvt_f16_f32_e32 v3, v3
+; CHECK-NEXT:    v_and_b32_e32 v2, 0xffff, v2
+; CHECK-NEXT:    v_and_b32_e32 v6, 0xffff, v6
+; CHECK-NEXT:    v_and_b32_e32 v3, 0xffff, v3
+; CHECK-NEXT:    v_lshl_or_b32 v2, v4, 16, v2
+; CHECK-NEXT:    v_cvt_f16_f32_e32 v4, v5
+; CHECK-NEXT:    v_lshl_or_b32 v5, v7, 16, v6
+; CHECK-NEXT:    v_lshl_or_b32 v3, v4, 16, v3
+; CHECK-NEXT:    s_setreg_imm32_b32 hwreg(HW_REG_MODE, 3, 1), 0
+; CHECK-NEXT:    v_pk_add_f16 v2, v2, v5
+; CHECK-NEXT:    v_pk_add_f16 v2, v3, v2
+; CHECK-NEXT:    global_store_dword v[0:1], v2, off
+; CHECK-NEXT:    s_endpgm
+  %res1 = call <2 x half> @llvm.fptrunc.round.v2f16.v2f32(<2 x float> %a, metadata !"round.upward")
+  %res2 = call <2 x half> @llvm.fptrunc.round.v2f16.v2f32(<2 x float> %b, metadata !"round.upward")
+  %res3 = call <2 x half> @llvm.fptrunc.round.v2f16.v2f32(<2 x float> %b, metadata !"round.downward")
+  %res4 = fadd <2 x half> %res1, %res2
+  %res5 = fadd <2 x half> %res3, %res4
+  store <2 x half> %res5, ptr addrspace(1) %out, align 4
+  ret void
+}
+
+define amdgpu_gs <3 x half> @v_fptrunc_round_v3f32_to_v3f16_upward(<3 x float> %a) {
+; SDAG-LABEL: v_fptrunc_round_v3f32_to_v3f16_upward:
+; SDAG:       ; %bb.0:
+; SDAG-NEXT:    s_setreg_imm32_b32 hwreg(HW_REG_MODE, 2, 1), 1
+; SDAG-NEXT:    v_cvt_f16_f32_e32 v0, v0
+; SDAG-NEXT:    v_cvt_f16_f32_e32 v1, v1
+; SDAG-NEXT:    v_perm_b32 v0, v1, v0, 0x5040100
+; SDAG-NEXT:    v_cvt_f16_f32_e32 v1, v2
+; SDAG-NEXT:    ; return to shader part epilog
+;
+; GISEL-LABEL: v_fptrunc_round_v3f32_to_v3f16_upward:
+; GISEL:       ; %bb.0:
+; GISEL-NEXT:    s_setreg_imm32_b32 hwreg(HW_REG_MODE, 2, 1), 1
+; GISEL-NEXT:    v_cvt_f16_f32_e32 v0, v0
+; GISEL-NEXT:    v_cvt_f16_f32_e32 v1, v1
+; GISEL-NEXT:    v_and_b32_e32 v0, 0xffff, v0
+; GISEL-NEXT:    v_lshl_or_b32 v0, v1, 16, v0
+; GISEL-NEXT:    v_cvt_f16_f32_e32 v1, v2
+; GISEL-NEXT:    ; return to shader part epilog
+  %res = call <3 x half> @llvm.fptrunc.round.v3f16.v3f32(<3 x float> %a, metadata !"round.upward")
+  ret <3 x half> %res
+}
+
+define amdgpu_gs <3 x half> @v_fptrunc_round_v3f32_to_v3f16_downward(<3 x float> %a) {
+; SDAG-LABEL: v_fptrunc_round_v3f32_to_v3f16_downward:
+; SDAG:       ; %bb.0:
+; SDAG-NEXT:    s_setreg_imm32_b32 hwreg(HW_REG_MODE, 3, 1), 1
+; SDAG-NEXT:    v_cvt_f16_f32_e32 v0, v0
+; SDAG-NEXT:    v_cvt_f16_f32_e32 v1, v1
+; SDAG-NEXT:    v_perm_b32 v0, v1, v0, 0x5040100
+; SDAG-NEXT:    v_cvt_f16_f32_e32 v1, v2
+; SDAG-NEXT:    ; return to shader part epilog
+;
+; GISEL-LABEL: v_fptrunc_round_v3f32_to_v3f16_downward:
+; GISEL:       ; %bb.0:
+; GISEL-NEXT:    s_setreg_imm32_b32 hwreg(HW_REG_MODE, 3, 1), 1
+; GISEL-NEXT:    v_cvt_f16_f32_e32 v0, v0
+; GISEL-NEXT:    v_cvt_f16_f32_e32 v1, v1
+; GISEL-NEXT:    v_and_b32_e32 v0, 0xffff, v0
+; GISEL-NEXT:    v_lshl_or_b32 v0, v1, 16, v0
+; GISEL-NEXT:    v_cvt_f16_f32_e32 v1, v2
+; GISEL-NEXT:    ; return to shader part epilog
+  %res = call <3 x half> @llvm.fptrunc.round.v3f16.v3f32(<3 x float> %a, metadata !"round.downward")
+  ret <3 x half> %res
+}
+
+define amdgpu_gs <4 x half> @v_fptrunc_round_v4f32_to_v4f16_upward(<4 x float> %a) {
+; SDAG-LABEL: v_fptrunc_round_v4f32_to_v4f16_upward:
+; SDAG:       ; %bb.0:
+; SDAG-NEXT:    s_setreg_imm32_b32 hwreg(HW_REG_MODE, 2, 1), 1
+; SDAG-NEXT:    v_cvt_f16_f32_e32 v2, v2
+; SDAG-NEXT:    v_cvt_f16_f32_e32 v0, v0
+; SDAG-NEXT:    v_cvt_f16_f32_e32 v1, v1
+; SDAG-NEXT:    v_cvt_f16_f32_e32 v3, v3
+; SDAG-NEXT:    v_perm_b32 v0, v1, v0, 0x5040100
+; SDAG-NEXT:    v_perm_b32 v1, v3, v2, 0x5040100
+; SDAG-NEXT:    ; return to shader part epilog
+;
+; GISEL-LABEL: v_fptrunc_round_v4f32_to_v4f16_upward:
+; GISEL:       ; %bb.0:
+; GISEL-NEXT:    s_setreg_imm32_b32 hwreg(HW_REG_MODE, 2, 1), 1
+; GISEL-NEXT:    v_cvt_f16_f32_e32 v0, v0
+; GISEL-NEXT:    v_cvt_f16_f32_e32 v2, v2
+; GISEL-NEXT:    v_cvt_f16_f32_e32 v1, v1
+; GISEL-NEXT:    v_cvt_f16_f32_e32 v3, v3
+; GISEL-NEXT:    v_and_b32_e32 v0, 0xffff, v0
+; GISEL-NEXT:    v_and_b32_e32 v2, 0xffff, v2
+; GISEL-NEXT:    v_lshl_or_b32 v0, v1, 16, v0
+; GISEL-NEXT:    v_lshl_or_b32 v1, v3, 16, v2
+; GISEL-NEXT:    ; return to shader part epilog
+  %res = call <4 x half> @llvm.fptrunc.round.v4f16.v4f32(<4 x float> %a, metadata !"round.upward")
+  ret <4 x half> %res
+}
+
+define amdgpu_gs <4 x half> @v_fptrunc_round_v4f32_to_v4f16_downward(<4 x float> %a) {
+; SDAG-LABEL: v_fptrunc_round_v4f32_to_v4f16_downward:
+; SDAG:       ; %bb.0:
+; SDAG-NEXT:    s_setreg_imm32_b32 hwreg(HW_REG_MODE, 3, 1), 1
+; SDAG-NEXT:    v_cvt_f16_f32_e32 v2, v2
+; SDAG-NEXT:    v_cvt_f16_f32_e32 v0, v0
+; SDAG-NEXT:    v_cvt_f16_f32_e32 v1, v1
+; SDAG-NEXT:    v_cvt_f16_f32_e32 v3, v3
+; SDAG-NEXT:    v_perm_b32 v0, v1, v0, 0x5040100
+; SDAG-NEXT:    v_perm_b32 v1, v3, v2, 0x5040100
+; SDAG-NEXT:    ; return to shader part epilog
+;
+; GISEL-LABEL: v_fptrunc_round_v4f32_to_v4f16_downward:
+; GISEL:       ; %bb.0:
+; GISEL-NEXT:    s_setreg_imm32_b32 hwreg(HW_REG_MODE, 3, 1), 1
+; GISEL-NEXT:    v_cvt_f16_f32_e32 v0, v0
+; GISEL-NEXT:    v_cvt_f16_f32_e32 v2, v2
+; GISEL-NEXT:    v_cvt_f16_f32_e32 v1, v1
+; GISEL-NEXT:    v_cvt_f16_f32_e32 v3, v3
+; GISEL-NEXT:    v_and_b32_e32 v0, 0xffff, v0
+; GISEL-NEXT:    v_and_b32_e32 v2, 0xffff, v2
+; GISEL-NEXT:    v_lshl_or_b32 v0, v1, 16, v0
+; GISEL-NEXT:    v_lshl_or_b32 v1, v3, 16, v2
+; GISEL-NEXT:    ; return to shader part epilog
+  %res = call <4 x half> @llvm.fptrunc.round.v4f16.v4f32(<4 x float> %a, metadata !"round.downward")
+  ret <4 x half> %res
+}
+
+define amdgpu_gs <8 x half> @v_fptrunc_round_v8f32_to_v8f16_upward(<8 x float> %a) {
+; SDAG-LABEL: v_fptrunc_round_v8f32_to_v8f16_upward:
+; SDAG:       ; %bb.0:
+; SDAG-NEXT:    s_setreg_imm32_b32 hwreg(HW_REG_MODE, 2, 1), 1
+; SDAG-NEXT:    v_cvt_f16_f32_e32 v6, v6
+; SDAG-NEXT:    v_cvt_f16_f32_e32 v4, v4
+; SDAG-NEXT:    v_cvt_f16_f32_e32 v2, v2
+; SDAG-NEXT:    v_cvt_f16_f32_e32 v0, v0
+; SDAG-NEXT:    v_cvt_f16_f32_e32 v1, v1
+; SDAG-NEXT:    v_cvt_f16_f32_e32 v3, v3
+; SDAG-NEXT:    v_cvt_f16_f32_e32 v5, v5
+; SDAG-NEXT:    v_cvt_f16_f32_e32 v7, v7
+; SDAG-NEXT:    v_perm_b32 v0, v1, v0, 0x5040100
+; SDAG-NEXT:    v_perm_b32 v1, v3, v2, 0x5040100
+; SDAG-NEXT:    v_perm_b32 v2, v5, v4, 0x5040100
+; SDAG-NEXT:    v_perm_b32 v3, v7, v6, 0x5040100
+; SDAG-NEXT:    ; return to shader part epilog
+;
+; GISEL-LABEL: v_fptrunc_round_v8f32_to_v8f16_upward:
+; GISEL:       ; %bb.0:
+; GISEL-NEXT:    s_setreg_imm32_b32 hwreg(HW_REG_MODE, 2, 1), 1
+; GISEL-NEXT:    v_cvt_f16_f32_e32 v0, v0
+; GISEL-NEXT:    v_cvt_f16_f32_e32 v2, v2
+; GISEL-NEXT:    v_cvt_f16_f32_e32 v4, v4
+; GISEL-NEXT:    v_cvt_f16_f32_e32 v6, v6
+; GISEL-NEXT:    v_cvt_f16_f32_e32 v1, v1
+; GISEL-NEXT:    v_cvt_f16_f32_e32 v3, v3
+; GISEL-NEXT:    v_cvt_f16_f32_e32 v5, v5
+; GISEL-NEXT:    v_cvt_f16_f32_e32 v7, v7
+; GISEL-NEXT:    v_and_b32_e32 v0, 0xffff, v0
+; GISEL-NEXT:    v_and_b32_e32 v2, 0xffff, v2
+; GISEL-NEXT:    v_and_b32_e32 v4, 0xffff, v4
+; GISEL-NEXT:    v_and_b32_e32 v6, 0xffff, v6
+; GISEL-NEXT:    v_lshl_or_b32 v0, v1, 16, v0
+; GISEL-NEXT:    v_lshl_or_b32 v1, v3, 16, v2
+; GISEL-NEXT:    v_lshl_or_b32 v2, v5, 16, v4
+; GISEL-NEXT:    v_lshl_or_b32 v3, v7, 16, v6
+; GISEL-NEXT:    ; return to shader part epilog
+  %res = call <8 x half> @llvm.fptrunc.round.v8f16.v8f32(<8 x float> %a, metadata !"round.upward")
+  ret <8 x half> %res
+}
+
+define amdgpu_gs <8 x half> @v_fptrunc_round_v8f32_to_v8f16_downward(<8 x float> %a) {
+; SDAG-LABEL: v_fptrunc_round_v8f32_to_v8f16_downward:
+; SDAG:       ; %bb.0:
+; SDAG-NEXT:    s_setreg_imm32_b32 hwreg(HW_REG_MODE, 3, 1), 1
+; SDAG-NEXT:    v_cvt_f16_f32_e32 v6, v6
+; SDAG-NEXT:    v_cvt_f16_f32_e32 v4, v4
+; SDAG-NEXT:    v_cvt_f16_f32_e32 v2, v2
+; SDAG-NEXT:    v_cvt_f16_f32_e32 v0, v0
+; SDAG-NEXT:    v_cvt_f16_f32_e32 v1, v1
+; SDAG-NEXT:    v_cvt_f16_f32_e32 v3, v3
+; SDAG-NEXT:    v_cvt_f16_f32_e32 v5, v5
+; SDAG-NEXT:    v_cvt_f16_f32_e32 v7, v7
+; SDAG-NEXT:    v_perm_b32 v0, v1, v0, 0x5040100
+; SDAG-NEXT:    v_perm_b32 v1, v3, v2, 0x5040100
+; SDAG-NEXT:    v_perm_b32 v2, v5, v4, 0x5040100
+; SDAG-NEXT:    v_perm_b32 v3, v7, v6, 0x5040100
+; SDAG-NEXT:    ; return to shader part epilog
+;
+; GISEL-LABEL: v_fptrunc_round_v8f32_to_v8f16_downward:
+; GISEL:       ; %bb.0:
+; GISEL-NEXT:    s_setreg_imm32_b32 hwreg(HW_REG_MODE, 3, 1), 1
+; GISEL-NEXT:    v_cvt_f16_f32_e32 v0, v0
+; GISEL-NEXT:    v_cvt_f16_f32_e32 v2, v2
+; GISEL-NEXT:    v_cvt_f16_f32_e32 v4, v4
+; GISEL-NEXT:    v_cvt_f16_f32_e32 v6, v6
+; GISEL-NEXT:    v_cvt_f16_f32_e32 v1, v1
+; GISEL-NEXT:    v_cvt_f16_f32_e32 v3, v3
+; GISEL-NEXT:    v_cvt_f16_f32_e32 v5, v5
+; GISEL-NEXT:    v_cvt_f16_f32_e32 v7, v7
+; GISEL-NEXT:    v_and_b32_e32 v0, 0xffff, v0
+; GISEL-NEXT:    v_and_b32_e32 v2, 0xffff, v2
+; GISEL-NEXT:    v_and_b32_e32 v4, 0xffff, v4
+; GISEL-NEXT:    v_and_b32_e32 v6, 0xffff, v6
+; GISEL-NEXT:    v_lshl_or_b32 v0, v1, 16, v0
+; GISEL-NEXT:    v_lshl_or_b32 v1, v3, 16, v2
+; GISEL-NEXT:    v_lshl_or_b32 v2, v5, 16, v4
+; GISEL-NEXT:    v_lshl_or_b32 v3, v7, 16, v6
+; GISEL-NEXT:    ; return to shader part epilog
+  %res = call <8 x half> @llvm.fptrunc.round.v8f16.v8f32(<8 x float> %a, metadata !"round.downward")
+  ret <8 x half> %res
+}
diff --git a/llvm/test/CodeGen/AMDGPU/lower-work-group-id-intrinsics-pal.ll b/llvm/test/CodeGen/AMDGPU/lower-work-group-id-intrinsics-pal.ll
index cfff0a969da9..14fe4e5f48c6 100644
--- a/llvm/test/CodeGen/AMDGPU/lower-work-group-id-intrinsics-pal.ll
+++ b/llvm/test/CodeGen/AMDGPU/lower-work-group-id-intrinsics-pal.ll
@@ -1,8 +1,8 @@
 ; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
-; RUN: llc -mtriple=amdgcn-amd-hsa -mcpu=gfx900 -mattr=-architected-sgprs -global-isel=0 -verify-machineinstrs < %s | FileCheck -check-prefixes=GFX9,GFX9-SDAG %s
-; RUN: llc -mtriple=amdgcn-amd-hsa -mcpu=gfx900 -mattr=-architected-sgprs -global-isel=1 -verify-machineinstrs < %s | FileCheck -check-prefixes=GFX9,GFX9-GISEL %s
-; RUN: llc -mtriple=amdgcn-amd-hsa -mcpu=gfx900 -mattr=+architected-sgprs -global-isel=0 -verify-machineinstrs < %s | FileCheck -check-prefixes=GFX9ARCH,GFX9ARCH-SDAG %s
-; RUN: llc -mtriple=amdgcn-amd-hsa -mcpu=gfx900 -mattr=+architected-sgprs -global-isel=1 -verify-machineinstrs < %s | FileCheck -check-prefixes=GFX9ARCH,GFX9ARCH-GISEL %s
+; RUN: llc -mtriple=amdgcn-amd-hsa -mcpu=gfx900 -mattr=-architected-sgprs -global-isel=0 -verify-machineinstrs < %s | FileCheck -check-prefixes=GFX9 %s
+; RUN: llc -mtriple=amdgcn-amd-hsa -mcpu=gfx900 -mattr=-architected-sgprs -global-isel=1 -verify-machineinstrs < %s | FileCheck -check-prefixes=GFX9 %s
+; RUN: llc -mtriple=amdgcn-amd-hsa -mcpu=gfx900 -mattr=+architected-sgprs -global-isel=0 -verify-machineinstrs < %s | FileCheck -check-prefixes=GFX9ARCH-SDAG %s
+; RUN: llc -mtriple=amdgcn-amd-hsa -mcpu=gfx900 -mattr=+architected-sgprs -global-isel=1 -verify-machineinstrs < %s | FileCheck -check-prefixes=GFX9ARCH-GISEL %s
 ; RUN: llc -mtriple=amdgcn-amd-amdpal -mcpu=gfx1200 -global-isel=0 -verify-machineinstrs < %s | FileCheck -check-prefixes=GFX12,GFX12-SDAG %s
 ; RUN: llc -mtriple=amdgcn-amd-amdpal -mcpu=gfx1200 -global-isel=1 -verify-machineinstrs < %s | FileCheck -check-prefixes=GFX12,GFX12-GISEL %s
 
@@ -156,10 +156,37 @@ define amdgpu_gfx void @workgroup_ids_gfx(ptr addrspace(1) %outx, ptr addrspace(
 ; GFX9-NEXT:    s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
 ; GFX9-NEXT:    s_setpc_b64 s[30:31]
 ;
-; GFX9ARCH-LABEL: workgroup_ids_gfx:
-; GFX9ARCH:       ; %bb.0:
-; GFX9ARCH-NEXT:    s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
-; GFX9ARCH-NEXT:    s_setpc_b64 s[30:31]
+; GFX9ARCH-SDAG-LABEL: workgroup_ids_gfx:
+; GFX9ARCH-SDAG:       ; %bb.0:
+; GFX9ARCH-SDAG-NEXT:    s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
+; GFX9ARCH-SDAG-NEXT:    v_mov_b32_e32 v6, ttmp9
+; GFX9ARCH-SDAG-NEXT:    s_and_b32 s34, ttmp7, 0xffff
+; GFX9ARCH-SDAG-NEXT:    global_store_dword v[0:1], v6, off
+; GFX9ARCH-SDAG-NEXT:    s_waitcnt vmcnt(0)
+; GFX9ARCH-SDAG-NEXT:    v_mov_b32_e32 v0, s34
+; GFX9ARCH-SDAG-NEXT:    s_lshr_b32 s34, ttmp7, 16
+; GFX9ARCH-SDAG-NEXT:    global_store_dword v[2:3], v0, off
+; GFX9ARCH-SDAG-NEXT:    s_waitcnt vmcnt(0)
+; GFX9ARCH-SDAG-NEXT:    v_mov_b32_e32 v0, s34
+; GFX9ARCH-SDAG-NEXT:    global_store_dword v[4:5], v0, off
+; GFX9ARCH-SDAG-NEXT:    s_waitcnt vmcnt(0)
+; GFX9ARCH-SDAG-NEXT:    s_setpc_b64 s[30:31]
+;
+; GFX9ARCH-GISEL-LABEL: workgroup_ids_gfx:
+; GFX9ARCH-GISEL:       ; %bb.0:
+; GFX9ARCH-GISEL-NEXT:    s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
+; GFX9ARCH-GISEL-NEXT:    v_mov_b32_e32 v6, ttmp9
+; GFX9ARCH-GISEL-NEXT:    s_and_b32 s34, ttmp7, 0xffff
+; GFX9ARCH-GISEL-NEXT:    s_lshr_b32 s35, ttmp7, 16
+; GFX9ARCH-GISEL-NEXT:    global_store_dword v[0:1], v6, off
+; GFX9ARCH-GISEL-NEXT:    s_waitcnt vmcnt(0)
+; GFX9ARCH-GISEL-NEXT:    v_mov_b32_e32 v0, s34
+; GFX9ARCH-GISEL-NEXT:    global_store_dword v[2:3], v0, off
+; GFX9ARCH-GISEL-NEXT:    s_waitcnt vmcnt(0)
+; GFX9ARCH-GISEL-NEXT:    v_mov_b32_e32 v0, s35
+; GFX9ARCH-GISEL-NEXT:    global_store_dword v[4:5], v0, off
+; GFX9ARCH-GISEL-NEXT:    s_waitcnt vmcnt(0)
+; GFX9ARCH-GISEL-NEXT:    s_setpc_b64 s[30:31]
 ;
 ; GFX12-LABEL: workgroup_ids_gfx:
 ; GFX12:       ; %bb.0:
@@ -168,6 +195,18 @@ define amdgpu_gfx void @workgroup_ids_gfx(ptr addrspace(1) %outx, ptr addrspace(
 ; GFX12-NEXT:    s_wait_samplecnt 0x0
 ; GFX12-NEXT:    s_wait_bvhcnt 0x0
 ; GFX12-NEXT:    s_wait_kmcnt 0x0
+; GFX12-NEXT:    s_and_b32 s0, ttmp7, 0xffff
+; GFX12-NEXT:    s_delay_alu instid0(SALU_CYCLE_1) | instskip(SKIP_1) | instid1(SALU_CYCLE_1)
+; GFX12-NEXT:    v_dual_mov_b32 v6, ttmp9 :: v_dual_mov_b32 v7, s0
+; GFX12-NEXT:    s_lshr_b32 s1, ttmp7, 16
+; GFX12-NEXT:    v_mov_b32_e32 v8, s1
+; GFX12-NEXT:    s_wait_storecnt 0x0
+; GFX12-NEXT:    global_store_b32 v[0:1], v6, off scope:SCOPE_SYS
+; GFX12-NEXT:    s_wait_storecnt 0x0
+; GFX12-NEXT:    global_store_b32 v[2:3], v7, off scope:SCOPE_SYS
+; GFX12-NEXT:    s_wait_storecnt 0x0
+; GFX12-NEXT:    global_store_b32 v[4:5], v8, off scope:SCOPE_SYS
+; GFX12-NEXT:    s_wait_storecnt 0x0
 ; GFX12-NEXT:    s_setpc_b64 s[30:31]
   %id.x = call i32 @llvm.amdgcn.workgroup.id.x()
   %id.y = call i32 @llvm.amdgcn.workgroup.id.y()
@@ -177,11 +216,3 @@ define amdgpu_gfx void @workgroup_ids_gfx(ptr addrspace(1) %outx, ptr addrspace(
   store volatile i32 %id.z, ptr addrspace(1) %outz
   ret void
 }
-
-declare i32 @llvm.amdgcn.workgroup.id.x()
-declare i32 @llvm.amdgcn.workgroup.id.y()
-declare i32 @llvm.amdgcn.workgroup.id.z()
-declare void @llvm.amdgcn.raw.ptr.buffer.store.v3i32(<3 x i32>, ptr addrspace(8), i32, i32, i32 immarg)
-;; NOTE: These prefixes are unused and the list is autogenerated. Do not add tests below this line:
-; GFX9-GISEL: {{.*}}
-; GFX9-SDAG: {{.*}}
diff --git a/llvm/test/CodeGen/AMDGPU/mad_64_32.ll b/llvm/test/CodeGen/AMDGPU/mad_64_32.ll
index 14bcc4f994f8..400298bcff4f 100644
--- a/llvm/test/CodeGen/AMDGPU/mad_64_32.ll
+++ b/llvm/test/CodeGen/AMDGPU/mad_64_32.ll
@@ -5,6 +5,7 @@
 ; RUN: llc -mtriple=amdgcn -mcpu=gfx1100 -verify-machineinstrs < %s | FileCheck -check-prefixes=GFX11,GFX1100 %s
 ; RUN: llc -mtriple=amdgcn -mcpu=gfx1150 -verify-machineinstrs < %s | FileCheck -check-prefixes=GFX11,GFX1150 %s
 ; RUN: llc -mtriple=amdgcn -mcpu=gfx1200 -verify-machineinstrs < %s | FileCheck -check-prefixes=GFX12 %s
+; RUN: llc -mtriple=amdgcn -mcpu=gfx11-generic --amdhsa-code-object-version=6 -verify-machineinstrs < %s | FileCheck -check-prefixes=GFX11,GFX1100 %s
 
 ; On GFX11, ensure vdst and src2 do not partially overlap. Full overlap is ok.
 
diff --git a/llvm/test/CodeGen/AMDGPU/mmra.ll b/llvm/test/CodeGen/AMDGPU/mmra.ll
new file mode 100644
index 000000000000..d9b48f79739b
--- /dev/null
+++ b/llvm/test/CodeGen/AMDGPU/mmra.ll
@@ -0,0 +1,189 @@
+; NOTE: Assertions have been autogenerated by utils/update_mir_test_checks.py UTC_ARGS: --version 2
+; RUN: llc -march=amdgcn -mcpu=gfx900 -stop-after=finalize-isel < %s | FileCheck %s
+
+declare void @readsMem(ptr) #0
+declare void @writesMem(ptr) #1
+
+define void @fence_loads(ptr %ptr) {
+  ; CHECK-LABEL: name: fence_loads
+  ; CHECK: bb.0 (%ir-block.0):
+  ; CHECK-NEXT:   liveins: $vgpr0, $vgpr1
+  ; CHECK-NEXT: {{  $}}
+  ; CHECK-NEXT:   [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr1
+  ; CHECK-NEXT:   [[COPY1:%[0-9]+]]:vgpr_32 = COPY $vgpr0
+  ; CHECK-NEXT:   [[DEF:%[0-9]+]]:sgpr_32 = IMPLICIT_DEF
+  ; CHECK-NEXT:   [[DEF1:%[0-9]+]]:sgpr_32 = IMPLICIT_DEF
+  ; CHECK-NEXT:   [[REG_SEQUENCE:%[0-9]+]]:vreg_64 = REG_SEQUENCE [[COPY1]], %subreg.sub0, [[COPY]], %subreg.sub1
+  ; CHECK-NEXT:   ATOMIC_FENCE 5, 1, mmra !0
+  ; CHECK-NEXT:   [[COPY2:%[0-9]+]]:vreg_64 = COPY [[REG_SEQUENCE]], mmra !1
+  ; CHECK-NEXT:   [[FLAT_LOAD_UBYTE:%[0-9]+]]:vgpr_32 = FLAT_LOAD_UBYTE [[COPY2]], 0, 0, implicit $exec, implicit $flat_scr, mmra !1 :: (load acquire (s8) from %ir.ptr, align 4)
+  ; CHECK-NEXT:   [[S_MOV_B32_:%[0-9]+]]:sreg_32 = S_MOV_B32 1, mmra !2
+  ; CHECK-NEXT:   [[COPY3:%[0-9]+]]:vreg_64 = COPY [[REG_SEQUENCE]], mmra !2
+  ; CHECK-NEXT:   [[COPY4:%[0-9]+]]:vgpr_32 = COPY [[S_MOV_B32_]], mmra !2
+  ; CHECK-NEXT:   FLAT_STORE_BYTE [[COPY3]], killed [[COPY4]], 0, 0, implicit $exec, implicit $flat_scr, mmra !2 :: (store release (s8) into %ir.ptr, align 4)
+  ; CHECK-NEXT:   SI_RETURN
+  fence release,                                        !mmra !0
+  %ld = load atomic i8, ptr %ptr acquire, align 4,      !mmra !2
+  store atomic i8 1, ptr %ptr release, align 4,         !mmra !1
+  ret void
+}
+
+define void @atomicrmw_acq(ptr %ptr) {
+  ; CHECK-LABEL: name: atomicrmw_acq
+  ; CHECK: bb.0 (%ir-block.0):
+  ; CHECK-NEXT:   liveins: $vgpr0, $vgpr1
+  ; CHECK-NEXT: {{  $}}
+  ; CHECK-NEXT:   [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr1
+  ; CHECK-NEXT:   [[COPY1:%[0-9]+]]:vgpr_32 = COPY $vgpr0
+  ; CHECK-NEXT:   [[DEF:%[0-9]+]]:sgpr_32 = IMPLICIT_DEF
+  ; CHECK-NEXT:   [[DEF1:%[0-9]+]]:sgpr_32 = IMPLICIT_DEF
+  ; CHECK-NEXT:   [[REG_SEQUENCE:%[0-9]+]]:vreg_64 = REG_SEQUENCE [[COPY1]], %subreg.sub0, [[COPY]], %subreg.sub1
+  ; CHECK-NEXT:   [[COPY2:%[0-9]+]]:vreg_64 = COPY [[REG_SEQUENCE]], mmra !1
+  ; CHECK-NEXT:   [[FLAT_LOAD_UBYTE:%[0-9]+]]:vgpr_32 = FLAT_LOAD_UBYTE killed [[COPY2]], 0, 0, implicit $exec, implicit $flat_scr, mmra !1 :: (load acquire (s8) from %ir.ptr)
+  ; CHECK-NEXT:   SI_RETURN
+  %old.2 = atomicrmw add ptr %ptr, i8 0 acquire,        !mmra !2
+  ret void
+}
+
+define void @atomicrmw_rel(ptr %ptr) {
+  ; CHECK-LABEL: name: atomicrmw_rel
+  ; CHECK: bb.0 (%ir-block.0):
+  ; CHECK-NEXT:   successors: %bb.1(0x80000000)
+  ; CHECK-NEXT:   liveins: $vgpr0, $vgpr1
+  ; CHECK-NEXT: {{  $}}
+  ; CHECK-NEXT:   [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr1
+  ; CHECK-NEXT:   [[COPY1:%[0-9]+]]:vgpr_32 = COPY $vgpr0
+  ; CHECK-NEXT:   [[DEF:%[0-9]+]]:sgpr_32 = IMPLICIT_DEF
+  ; CHECK-NEXT:   [[DEF1:%[0-9]+]]:sgpr_32 = IMPLICIT_DEF
+  ; CHECK-NEXT:   [[REG_SEQUENCE:%[0-9]+]]:vreg_64 = REG_SEQUENCE [[COPY1]], %subreg.sub0, [[COPY]], %subreg.sub1
+  ; CHECK-NEXT:   [[COPY2:%[0-9]+]]:vgpr_32 = COPY [[REG_SEQUENCE]].sub1
+  ; CHECK-NEXT:   [[COPY3:%[0-9]+]]:vgpr_32 = COPY [[REG_SEQUENCE]].sub0
+  ; CHECK-NEXT:   [[S_MOV_B32_:%[0-9]+]]:sreg_32 = S_MOV_B32 -4
+  ; CHECK-NEXT:   [[V_AND_B32_e64_:%[0-9]+]]:vgpr_32 = V_AND_B32_e64 [[COPY3]], killed [[S_MOV_B32_]], implicit $exec
+  ; CHECK-NEXT:   [[DEF2:%[0-9]+]]:sgpr_32 = IMPLICIT_DEF
+  ; CHECK-NEXT:   [[DEF3:%[0-9]+]]:sgpr_32 = IMPLICIT_DEF
+  ; CHECK-NEXT:   [[REG_SEQUENCE1:%[0-9]+]]:vreg_64 = REG_SEQUENCE [[V_AND_B32_e64_]], %subreg.sub0, [[COPY2]], %subreg.sub1
+  ; CHECK-NEXT:   [[COPY4:%[0-9]+]]:vreg_64 = COPY [[REG_SEQUENCE1]]
+  ; CHECK-NEXT:   [[S_MOV_B32_1:%[0-9]+]]:sreg_32 = S_MOV_B32 3
+  ; CHECK-NEXT:   [[V_AND_B32_e64_1:%[0-9]+]]:vgpr_32 = V_AND_B32_e64 [[COPY3]], [[S_MOV_B32_1]], implicit $exec
+  ; CHECK-NEXT:   [[V_LSHLREV_B32_e64_:%[0-9]+]]:vgpr_32 = V_LSHLREV_B32_e64 [[S_MOV_B32_1]], killed [[V_AND_B32_e64_1]], implicit $exec
+  ; CHECK-NEXT:   [[S_MOV_B32_2:%[0-9]+]]:sreg_32 = S_MOV_B32 255
+  ; CHECK-NEXT:   [[V_LSHLREV_B32_e64_1:%[0-9]+]]:vgpr_32 = V_LSHLREV_B32_e64 killed [[V_LSHLREV_B32_e64_]], killed [[S_MOV_B32_2]], implicit $exec
+  ; CHECK-NEXT:   [[V_NOT_B32_e32_:%[0-9]+]]:vgpr_32 = V_NOT_B32_e32 [[V_LSHLREV_B32_e64_1]], implicit $exec
+  ; CHECK-NEXT:   [[COPY5:%[0-9]+]]:vreg_64 = COPY [[REG_SEQUENCE1]], mmra !2
+  ; CHECK-NEXT:   [[FLAT_LOAD_DWORD:%[0-9]+]]:vgpr_32 = FLAT_LOAD_DWORD [[COPY5]], 0, 0, implicit $exec, implicit $flat_scr, mmra !2 :: (load (s32) from %ir.AlignedAddr)
+  ; CHECK-NEXT:   [[S_MOV_B64_:%[0-9]+]]:sreg_64 = S_MOV_B64 0
+  ; CHECK-NEXT: {{  $}}
+  ; CHECK-NEXT: bb.1.atomicrmw.start:
+  ; CHECK-NEXT:   successors: %bb.2(0x04000000), %bb.1(0x7c000000)
+  ; CHECK-NEXT: {{  $}}
+  ; CHECK-NEXT:   [[PHI:%[0-9]+]]:sreg_64 = PHI [[S_MOV_B64_]], %bb.0, %7, %bb.1
+  ; CHECK-NEXT:   [[PHI1:%[0-9]+]]:vgpr_32 = PHI [[FLAT_LOAD_DWORD]], %bb.0, %6, %bb.1
+  ; CHECK-NEXT:   [[V_OR_B32_e64_:%[0-9]+]]:vgpr_32 = V_OR_B32_e64 [[V_NOT_B32_e32_]], [[V_LSHLREV_B32_e64_1]], implicit $exec
+  ; CHECK-NEXT:   [[V_AND_B32_e64_2:%[0-9]+]]:vgpr_32 = V_AND_B32_e64 [[PHI1]], killed [[V_OR_B32_e64_]], implicit $exec
+  ; CHECK-NEXT:   [[DEF4:%[0-9]+]]:sgpr_32 = IMPLICIT_DEF
+  ; CHECK-NEXT:   [[DEF5:%[0-9]+]]:sgpr_32 = IMPLICIT_DEF
+  ; CHECK-NEXT:   [[REG_SEQUENCE2:%[0-9]+]]:vreg_64 = REG_SEQUENCE [[V_AND_B32_e64_2]], %subreg.sub0, [[PHI1]], %subreg.sub1, mmra !2
+  ; CHECK-NEXT:   [[COPY6:%[0-9]+]]:vreg_64 = COPY [[REG_SEQUENCE2]], mmra !2
+  ; CHECK-NEXT:   [[FLAT_ATOMIC_CMPSWAP_RTN:%[0-9]+]]:vgpr_32 = FLAT_ATOMIC_CMPSWAP_RTN [[COPY4]], killed [[COPY6]], 0, 1, implicit $exec, implicit $flat_scr, mmra !2 :: (load store release monotonic (s32) on %ir.AlignedAddr)
+  ; CHECK-NEXT:   [[V_CMP_EQ_U32_e64_:%[0-9]+]]:sreg_64 = V_CMP_EQ_U32_e64 [[FLAT_ATOMIC_CMPSWAP_RTN]], [[PHI1]], implicit $exec, mmra !2
+  ; CHECK-NEXT:   [[SI_IF_BREAK:%[0-9]+]]:sreg_64 = SI_IF_BREAK killed [[V_CMP_EQ_U32_e64_]], [[PHI]], implicit-def dead $scc
+  ; CHECK-NEXT:   SI_LOOP [[SI_IF_BREAK]], %bb.1, implicit-def dead $exec, implicit-def dead $scc, implicit $exec
+  ; CHECK-NEXT:   S_BRANCH %bb.2
+  ; CHECK-NEXT: {{  $}}
+  ; CHECK-NEXT: bb.2.atomicrmw.end:
+  ; CHECK-NEXT:   [[PHI2:%[0-9]+]]:sreg_64 = PHI [[SI_IF_BREAK]], %bb.1
+  ; CHECK-NEXT:   SI_END_CF [[PHI2]], implicit-def dead $exec, implicit-def dead $scc, implicit $exec
+  ; CHECK-NEXT:   SI_RETURN
+  %old.2 = atomicrmw add ptr %ptr, i8 0 release,        !mmra !1
+  ret void
+}
+
+define void @cmpxchg(ptr %ptr) {
+  ; CHECK-LABEL: name: cmpxchg
+  ; CHECK: bb.0 (%ir-block.0):
+  ; CHECK-NEXT:   successors: %bb.1(0x80000000)
+  ; CHECK-NEXT:   liveins: $vgpr0, $vgpr1
+  ; CHECK-NEXT: {{  $}}
+  ; CHECK-NEXT:   [[COPY:%[0-9]+]]:vgpr_32 = COPY $vgpr1
+  ; CHECK-NEXT:   [[COPY1:%[0-9]+]]:vgpr_32 = COPY $vgpr0
+  ; CHECK-NEXT:   [[DEF:%[0-9]+]]:sgpr_32 = IMPLICIT_DEF
+  ; CHECK-NEXT:   [[DEF1:%[0-9]+]]:sgpr_32 = IMPLICIT_DEF
+  ; CHECK-NEXT:   [[REG_SEQUENCE:%[0-9]+]]:vreg_64 = REG_SEQUENCE [[COPY1]], %subreg.sub0, [[COPY]], %subreg.sub1
+  ; CHECK-NEXT:   [[COPY2:%[0-9]+]]:vgpr_32 = COPY [[REG_SEQUENCE]].sub1
+  ; CHECK-NEXT:   [[COPY3:%[0-9]+]]:vgpr_32 = COPY [[REG_SEQUENCE]].sub0
+  ; CHECK-NEXT:   [[S_MOV_B32_:%[0-9]+]]:sreg_32 = S_MOV_B32 -4
+  ; CHECK-NEXT:   [[V_AND_B32_e64_:%[0-9]+]]:vgpr_32 = V_AND_B32_e64 [[COPY3]], killed [[S_MOV_B32_]], implicit $exec
+  ; CHECK-NEXT:   [[DEF2:%[0-9]+]]:sgpr_32 = IMPLICIT_DEF
+  ; CHECK-NEXT:   [[DEF3:%[0-9]+]]:sgpr_32 = IMPLICIT_DEF
+  ; CHECK-NEXT:   [[REG_SEQUENCE1:%[0-9]+]]:vreg_64 = REG_SEQUENCE [[V_AND_B32_e64_]], %subreg.sub0, [[COPY2]], %subreg.sub1
+  ; CHECK-NEXT:   [[COPY4:%[0-9]+]]:vreg_64 = COPY [[REG_SEQUENCE1]]
+  ; CHECK-NEXT:   [[S_MOV_B32_1:%[0-9]+]]:sreg_32 = S_MOV_B32 3
+  ; CHECK-NEXT:   [[V_AND_B32_e64_1:%[0-9]+]]:vgpr_32 = V_AND_B32_e64 [[COPY3]], [[S_MOV_B32_1]], implicit $exec
+  ; CHECK-NEXT:   [[V_LSHLREV_B32_e64_:%[0-9]+]]:vgpr_32 = V_LSHLREV_B32_e64 [[S_MOV_B32_1]], killed [[V_AND_B32_e64_1]], implicit $exec
+  ; CHECK-NEXT:   [[S_MOV_B32_2:%[0-9]+]]:sreg_32 = S_MOV_B32 255
+  ; CHECK-NEXT:   [[V_LSHLREV_B32_e64_1:%[0-9]+]]:vgpr_32 = V_LSHLREV_B32_e64 [[V_LSHLREV_B32_e64_]], killed [[S_MOV_B32_2]], implicit $exec
+  ; CHECK-NEXT:   [[V_NOT_B32_e32_:%[0-9]+]]:vgpr_32 = V_NOT_B32_e32 killed [[V_LSHLREV_B32_e64_1]], implicit $exec
+  ; CHECK-NEXT:   [[S_MOV_B32_3:%[0-9]+]]:sreg_32 = S_MOV_B32 1
+  ; CHECK-NEXT:   [[V_LSHLREV_B32_e64_2:%[0-9]+]]:vgpr_32 = V_LSHLREV_B32_e64 [[V_LSHLREV_B32_e64_]], killed [[S_MOV_B32_3]], implicit $exec
+  ; CHECK-NEXT:   [[COPY5:%[0-9]+]]:vreg_64 = COPY [[REG_SEQUENCE1]], mmra !1
+  ; CHECK-NEXT:   [[FLAT_LOAD_DWORD:%[0-9]+]]:vgpr_32 = FLAT_LOAD_DWORD [[COPY5]], 0, 0, implicit $exec, implicit $flat_scr, mmra !1 :: (load (s32) from %ir.AlignedAddr)
+  ; CHECK-NEXT:   [[V_AND_B32_e64_2:%[0-9]+]]:vgpr_32 = V_AND_B32_e64 killed [[FLAT_LOAD_DWORD]], [[V_NOT_B32_e32_]], implicit $exec
+  ; CHECK-NEXT:   [[S_MOV_B64_:%[0-9]+]]:sreg_64 = S_MOV_B64 0
+  ; CHECK-NEXT:   [[DEF4:%[0-9]+]]:sreg_64 = IMPLICIT_DEF
+  ; CHECK-NEXT: {{  $}}
+  ; CHECK-NEXT: bb.1.partword.cmpxchg.loop:
+  ; CHECK-NEXT:   successors: %bb.2(0x40000000), %bb.3(0x40000000)
+  ; CHECK-NEXT: {{  $}}
+  ; CHECK-NEXT:   [[PHI:%[0-9]+]]:sreg_64 = PHI [[DEF4]], %bb.0, %12, %bb.3
+  ; CHECK-NEXT:   [[PHI1:%[0-9]+]]:sreg_64 = PHI [[S_MOV_B64_]], %bb.0, %13, %bb.3
+  ; CHECK-NEXT:   [[PHI2:%[0-9]+]]:vgpr_32 = PHI [[V_AND_B32_e64_2]], %bb.0, %11, %bb.3
+  ; CHECK-NEXT:   [[V_OR_B32_e64_:%[0-9]+]]:vgpr_32 = V_OR_B32_e64 [[PHI2]], [[V_LSHLREV_B32_e64_2]], implicit $exec
+  ; CHECK-NEXT:   [[DEF5:%[0-9]+]]:sgpr_32 = IMPLICIT_DEF
+  ; CHECK-NEXT:   [[DEF6:%[0-9]+]]:sgpr_32 = IMPLICIT_DEF
+  ; CHECK-NEXT:   [[REG_SEQUENCE2:%[0-9]+]]:vreg_64 = REG_SEQUENCE [[V_OR_B32_e64_]], %subreg.sub0, [[PHI2]], %subreg.sub1, mmra !1
+  ; CHECK-NEXT:   [[COPY6:%[0-9]+]]:vreg_64 = COPY [[REG_SEQUENCE2]], mmra !1
+  ; CHECK-NEXT:   [[FLAT_ATOMIC_CMPSWAP_RTN:%[0-9]+]]:vgpr_32 = FLAT_ATOMIC_CMPSWAP_RTN [[COPY4]], killed [[COPY6]], 0, 1, implicit $exec, implicit $flat_scr, mmra !1 :: (load store acquire acquire (s32) on %ir.AlignedAddr)
+  ; CHECK-NEXT:   [[V_CMP_NE_U32_e64_:%[0-9]+]]:sreg_64 = V_CMP_NE_U32_e64 [[FLAT_ATOMIC_CMPSWAP_RTN]], [[PHI2]], implicit $exec
+  ; CHECK-NEXT:   [[S_MOV_B64_1:%[0-9]+]]:sreg_64 = S_MOV_B64 -1
+  ; CHECK-NEXT:   [[DEF7:%[0-9]+]]:sreg_32 = IMPLICIT_DEF
+  ; CHECK-NEXT:   [[COPY7:%[0-9]+]]:vgpr_32 = COPY [[DEF7]]
+  ; CHECK-NEXT:   [[S_OR_B64_:%[0-9]+]]:sreg_64 = S_OR_B64 [[PHI]], $exec, implicit-def $scc
+  ; CHECK-NEXT:   [[SI_IF:%[0-9]+]]:sreg_64 = SI_IF killed [[V_CMP_NE_U32_e64_]], %bb.3, implicit-def dead $exec, implicit-def dead $scc, implicit $exec
+  ; CHECK-NEXT:   S_BRANCH %bb.2
+  ; CHECK-NEXT: {{  $}}
+  ; CHECK-NEXT: bb.2.partword.cmpxchg.failure:
+  ; CHECK-NEXT:   successors: %bb.3(0x80000000)
+  ; CHECK-NEXT: {{  $}}
+  ; CHECK-NEXT:   [[V_AND_B32_e64_3:%[0-9]+]]:vgpr_32 = V_AND_B32_e64 [[FLAT_ATOMIC_CMPSWAP_RTN]], [[V_NOT_B32_e32_]], implicit $exec
+  ; CHECK-NEXT:   [[V_CMP_EQ_U32_e64_:%[0-9]+]]:sreg_64 = V_CMP_EQ_U32_e64 [[PHI2]], [[V_AND_B32_e64_3]], implicit $exec
+  ; CHECK-NEXT:   [[S_ANDN2_B64_:%[0-9]+]]:sreg_64 = S_ANDN2_B64 [[S_OR_B64_]], $exec, implicit-def $scc
+  ; CHECK-NEXT:   [[S_AND_B64_:%[0-9]+]]:sreg_64 = S_AND_B64 [[V_CMP_EQ_U32_e64_]], $exec, implicit-def $scc
+  ; CHECK-NEXT:   [[S_OR_B64_1:%[0-9]+]]:sreg_64 = S_OR_B64 [[S_ANDN2_B64_]], [[S_AND_B64_]], implicit-def $scc
+  ; CHECK-NEXT: {{  $}}
+  ; CHECK-NEXT: bb.3.Flow:
+  ; CHECK-NEXT:   successors: %bb.4(0x04000000), %bb.1(0x7c000000)
+  ; CHECK-NEXT: {{  $}}
+  ; CHECK-NEXT:   [[PHI3:%[0-9]+]]:sreg_64 = PHI [[S_OR_B64_]], %bb.1, [[S_OR_B64_1]], %bb.2
+  ; CHECK-NEXT:   [[PHI4:%[0-9]+]]:vgpr_32 = PHI [[COPY7]], %bb.1, [[V_AND_B32_e64_3]], %bb.2
+  ; CHECK-NEXT:   SI_END_CF [[SI_IF]], implicit-def dead $exec, implicit-def dead $scc, implicit $exec
+  ; CHECK-NEXT:   [[COPY8:%[0-9]+]]:sreg_64 = COPY [[PHI3]]
+  ; CHECK-NEXT:   [[SI_IF_BREAK:%[0-9]+]]:sreg_64 = SI_IF_BREAK [[COPY8]], [[PHI1]], implicit-def dead $scc
+  ; CHECK-NEXT:   SI_LOOP [[SI_IF_BREAK]], %bb.1, implicit-def dead $exec, implicit-def dead $scc, implicit $exec
+  ; CHECK-NEXT:   S_BRANCH %bb.4
+  ; CHECK-NEXT: {{  $}}
+  ; CHECK-NEXT: bb.4.partword.cmpxchg.end:
+  ; CHECK-NEXT:   [[PHI5:%[0-9]+]]:sreg_64 = PHI [[SI_IF_BREAK]], %bb.3
+  ; CHECK-NEXT:   [[PHI6:%[0-9]+]]:vgpr_32 = PHI [[FLAT_ATOMIC_CMPSWAP_RTN]], %bb.3
+  ; CHECK-NEXT:   SI_END_CF [[PHI5]], implicit-def dead $exec, implicit-def dead $scc, implicit $exec
+  ; CHECK-NEXT:   SI_RETURN
+  %pair = cmpxchg ptr %ptr, i8 0, i8 1 acquire acquire, !mmra !2
+  ret void
+}
+
+attributes #0 = { memory(read) }
+attributes #1 = { memory(write) }
+
+!0 = !{!"foo", !"bar"}
+!1 = !{!"bux", !"baz"}
+!2 = !{!0, !1}
diff --git a/llvm/test/CodeGen/AMDGPU/mode-register-fpconstrain.ll b/llvm/test/CodeGen/AMDGPU/mode-register-fpconstrain.ll
new file mode 100644
index 000000000000..2403aeaa4428
--- /dev/null
+++ b/llvm/test/CodeGen/AMDGPU/mode-register-fpconstrain.ll
@@ -0,0 +1,43 @@
+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --version 4
+; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=gfx900 %s -o - | FileCheck -check-prefix=GCN %s
+
+; The si-mode-register pass is changing the default mode for FP constrained  operations.
+; It must ignore for strictfp functions.
+
+define double @ignoreStrictfp(double noundef %a, double noundef %b) #0 {
+; GCN-LABEL: ignoreStrictfp:
+; GCN:       ; %bb.0:
+; GCN-NEXT:    s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
+; GCN-NEXT:    s_setreg_imm32_b32 hwreg(HW_REG_MODE, 2, 2), 1
+; GCN-NEXT:    s_nop 1
+; GCN-NEXT:    s_setreg_imm32_b32 hwreg(HW_REG_MODE, 2, 1), 0
+; GCN-NEXT:    v_add_f64 v[0:1], v[0:1], v[2:3]
+; GCN-NEXT:    s_setpc_b64 s[30:31]
+  tail call void @llvm.amdgcn.s.setreg(i32 2177, i32 1)
+  %val = tail call double @llvm.experimental.constrained.fadd.f64(double %a, double %b, metadata !"round.dynamic", metadata !"fpexcept.strict") #0
+  ret double %val
+}
+
+define double @set_fpenv(double noundef %a, double noundef %b) #0 {
+; GCN-LABEL: set_fpenv:
+; GCN:       ; %bb.0: ; %entry
+; GCN-NEXT:    s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
+; GCN-NEXT:    s_setreg_imm32_b32 hwreg(HW_REG_MODE, 0, 23), 4
+; GCN-NEXT:    s_setreg_imm32_b32 hwreg(HW_REG_TRAPSTS, 0, 5), 0
+; GCN-NEXT:    s_nop 0
+; GCN-NEXT:    s_setreg_imm32_b32 hwreg(HW_REG_MODE, 2, 1), 0
+; GCN-NEXT:    v_add_f64 v[0:1], v[0:1], v[2:3]
+; GCN-NEXT:    s_setpc_b64 s[30:31]
+entry:
+  call void @llvm.set.fpenv.i64(i64 4)
+  %val = tail call double @llvm.experimental.constrained.fadd.f64(double %a, double %b, metadata !"round.dynamic", metadata !"fpexcept.strict") #0
+  ret double %val
+}
+
+declare void @llvm.amdgcn.s.setreg(i32 immarg, i32)
+
+declare double @llvm.experimental.constrained.fadd.f64(double, double, metadata, metadata)
+
+declare void @llvm.set.fpenv.i64(i64)
+
+attributes #0 = { strictfp }
diff --git a/llvm/test/CodeGen/AMDGPU/pal-userdata-regs.ll b/llvm/test/CodeGen/AMDGPU/pal-userdata-regs.ll
index 6d043e2b6b0a..591deda611b2 100644
--- a/llvm/test/CodeGen/AMDGPU/pal-userdata-regs.ll
+++ b/llvm/test/CodeGen/AMDGPU/pal-userdata-regs.ll
@@ -4,12 +4,12 @@
 ; full tessellation-and-geometry pipeline, compiled on gfx8 so it uses all six
 ; hardware shader types.
 
-; CHECK-DAG: 0x2c0c (SPI_SHADER_USER_DATA_PS_0): 0x10000000
-; CHECK-DAG: 0x2c4c (SPI_SHADER_USER_DATA_VS_0): 0x10000000
-; CHECK-DAG: 0x2c8c (SPI_SHADER_USER_DATA_GS_0): 0x10000000
-; CHECK-DAG: 0x2ccc (SPI_SHADER_USER_DATA_ES_0): 0x10000000
-; CHECK-DAG: 0x2d0c (SPI_SHADER_USER_DATA_HS_0): 0x10000000
-; CHECK-DAG: 0x2d4c (SPI_SHADER_USER_DATA_LS_0): 0x10000000
+; CHECK-DAG: '0x2c0c (SPI_SHADER_USER_DATA_PS_0)': 0x10000000
+; CHECK-DAG: '0x2c4c (SPI_SHADER_USER_DATA_VS_0)': 0x10000000
+; CHECK-DAG: '0x2c8c (SPI_SHADER_USER_DATA_GS_0)': 0x10000000
+; CHECK-DAG: '0x2ccc (SPI_SHADER_USER_DATA_ES_0)': 0x10000000
+; CHECK-DAG: '0x2d0c (SPI_SHADER_USER_DATA_HS_0)': 0x10000000
+; CHECK-DAG: '0x2d4c (SPI_SHADER_USER_DATA_LS_0)': 0x10000000
 
 !amdgpu.pal.metadata.msgpack = !{!0}
 
diff --git a/llvm/test/CodeGen/AMDGPU/rem_i128.ll b/llvm/test/CodeGen/AMDGPU/rem_i128.ll
index 6ba66ccf7186..b068d87c4d6f 100644
--- a/llvm/test/CodeGen/AMDGPU/rem_i128.ll
+++ b/llvm/test/CodeGen/AMDGPU/rem_i128.ll
@@ -242,130 +242,137 @@ define i128 @v_srem_i128_vv(i128 %lhs, i128 %rhs) {
 ; GFX9-O0:       ; %bb.0: ; %_udiv-special-cases
 ; GFX9-O0-NEXT:    s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
 ; GFX9-O0-NEXT:    s_xor_saveexec_b64 s[4:5], -1
-; GFX9-O0-NEXT:    buffer_store_dword v0, off, s[0:3], s32 offset:348 ; 4-byte Folded Spill
-; GFX9-O0-NEXT:    buffer_store_dword v4, off, s[0:3], s32 offset:352 ; 4-byte Folded Spill
-; GFX9-O0-NEXT:    buffer_store_dword v8, off, s[0:3], s32 offset:356 ; 4-byte Folded Spill
-; GFX9-O0-NEXT:    buffer_store_dword v16, off, s[0:3], s32 offset:360 ; 4-byte Folded Spill
+; GFX9-O0-NEXT:    buffer_store_dword v0, off, s[0:3], s32 offset:344 ; 4-byte Folded Spill
+; GFX9-O0-NEXT:    buffer_store_dword v4, off, s[0:3], s32 offset:348 ; 4-byte Folded Spill
+; GFX9-O0-NEXT:    buffer_store_dword v8, off, s[0:3], s32 offset:352 ; 4-byte Folded Spill
+; GFX9-O0-NEXT:    buffer_store_dword v16, off, s[0:3], s32 offset:356 ; 4-byte Folded Spill
 ; GFX9-O0-NEXT:    s_mov_b64 exec, s[4:5]
 ; GFX9-O0-NEXT:    ; implicit-def: $vgpr8 : SGPR spill to VGPR lane
-; GFX9-O0-NEXT:    v_mov_b32_e32 v8, v6
-; GFX9-O0-NEXT:    buffer_store_dword v4, off, s[0:3], s32 offset:120 ; 4-byte Folded Spill
-; GFX9-O0-NEXT:    buffer_store_dword v3, off, s[0:3], s32 offset:116 ; 4-byte Folded Spill
-; GFX9-O0-NEXT:    v_mov_b32_e32 v14, v2
-; GFX9-O0-NEXT:    buffer_load_dword v2, off, s[0:3], s32 offset:120 ; 4-byte Folded Reload
-; GFX9-O0-NEXT:    v_mov_b32_e32 v6, v1
-; GFX9-O0-NEXT:    buffer_load_dword v1, off, s[0:3], s32 offset:116 ; 4-byte Folded Reload
-; GFX9-O0-NEXT:    v_mov_b32_e32 v4, v0
+; GFX9-O0-NEXT:    v_mov_b32_e32 v9, v7
+; GFX9-O0-NEXT:    buffer_store_dword v5, off, s[0:3], s32 offset:116 ; 4-byte Folded Spill
+; GFX9-O0-NEXT:    v_mov_b32_e32 v8, v2
+; GFX9-O0-NEXT:    buffer_load_dword v2, off, s[0:3], s32 offset:116 ; 4-byte Folded Reload
+; GFX9-O0-NEXT:    v_mov_b32_e32 v7, v1
+; GFX9-O0-NEXT:    v_mov_b32_e32 v1, v0
 ; GFX9-O0-NEXT:    s_or_saveexec_b64 s[18:19], -1
 ; GFX9-O0-NEXT:    buffer_load_dword v0, off, s[0:3], s32 ; 4-byte Folded Reload
 ; GFX9-O0-NEXT:    s_mov_b64 exec, s[18:19]
 ; GFX9-O0-NEXT:    ; implicit-def: $sgpr4
 ; GFX9-O0-NEXT:    ; implicit-def: $sgpr4
-; GFX9-O0-NEXT:    ; kill: def $vgpr2 killed $vgpr2 def $vgpr2_vgpr3 killed $exec
-; GFX9-O0-NEXT:    v_mov_b32_e32 v3, v5
+; GFX9-O0-NEXT:    ; kill: def $vgpr4 killed $vgpr4 def $vgpr4_vgpr5 killed $exec
+; GFX9-O0-NEXT:    s_waitcnt vmcnt(1)
+; GFX9-O0-NEXT:    v_mov_b32_e32 v5, v2
 ; GFX9-O0-NEXT:    ; implicit-def: $sgpr4
 ; GFX9-O0-NEXT:    ; implicit-def: $sgpr4
-; GFX9-O0-NEXT:    ; kill: def $vgpr4 killed $vgpr4 def $vgpr4_vgpr5 killed $exec
-; GFX9-O0-NEXT:    v_mov_b32_e32 v5, v6
+; GFX9-O0-NEXT:    ; kill: def $vgpr1 killed $vgpr1 def $vgpr1_vgpr2 killed $exec
+; GFX9-O0-NEXT:    v_mov_b32_e32 v2, v7
 ; GFX9-O0-NEXT:    ; implicit-def: $sgpr4
 ; GFX9-O0-NEXT:    ; implicit-def: $sgpr4
-; GFX9-O0-NEXT:    ; kill: def $vgpr8 killed $vgpr8 def $vgpr8_vgpr9 killed $exec
-; GFX9-O0-NEXT:    v_mov_b32_e32 v9, v7
+; GFX9-O0-NEXT:    ; kill: def $vgpr6 killed $vgpr6 def $vgpr6_vgpr7 killed $exec
+; GFX9-O0-NEXT:    v_mov_b32_e32 v7, v9
 ; GFX9-O0-NEXT:    ; implicit-def: $sgpr4
 ; GFX9-O0-NEXT:    ; implicit-def: $sgpr4
-; GFX9-O0-NEXT:    ; kill: def $vgpr14 killed $vgpr14 def $vgpr14_vgpr15 killed $exec
-; GFX9-O0-NEXT:    s_waitcnt vmcnt(1)
-; GFX9-O0-NEXT:    v_mov_b32_e32 v15, v1
+; GFX9-O0-NEXT:    ; kill: def $vgpr8 killed $vgpr8 def $vgpr8_vgpr9 killed $exec
+; GFX9-O0-NEXT:    v_mov_b32_e32 v9, v3
 ; GFX9-O0-NEXT:    ; implicit-def: $sgpr4_sgpr5
 ; GFX9-O0-NEXT:    ; implicit-def: $sgpr4_sgpr5
 ; GFX9-O0-NEXT:    ; implicit-def: $sgpr4_sgpr5
 ; GFX9-O0-NEXT:    ; implicit-def: $sgpr4_sgpr5
 ; GFX9-O0-NEXT:    s_mov_b32 s4, 63
-; GFX9-O0-NEXT:    v_mov_b32_e32 v6, v14
-; GFX9-O0-NEXT:    v_mov_b32_e32 v7, v15
-; GFX9-O0-NEXT:    v_ashrrev_i64 v[12:13], s4, v[6:7]
-; GFX9-O0-NEXT:    buffer_store_dword v12, off, s[0:3], s32 offset:108 ; 4-byte Folded Spill
+; GFX9-O0-NEXT:    v_mov_b32_e32 v11, v9
+; GFX9-O0-NEXT:    v_mov_b32_e32 v10, v8
+; GFX9-O0-NEXT:    v_ashrrev_i64 v[11:12], s4, v[10:11]
+; GFX9-O0-NEXT:    buffer_store_dword v11, off, s[0:3], s32 offset:108 ; 4-byte Folded Spill
 ; GFX9-O0-NEXT:    s_waitcnt vmcnt(0)
-; GFX9-O0-NEXT:    buffer_store_dword v13, off, s[0:3], s32 offset:112 ; 4-byte Folded Spill
-; GFX9-O0-NEXT:    v_mov_b32_e32 v6, v12
-; GFX9-O0-NEXT:    v_mov_b32_e32 v7, v13
-; GFX9-O0-NEXT:    buffer_store_dword v6, off, s[0:3], s32 offset:100 ; 4-byte Folded Spill
+; GFX9-O0-NEXT:    buffer_store_dword v12, off, s[0:3], s32 offset:112 ; 4-byte Folded Spill
+; GFX9-O0-NEXT:    v_mov_b32_e32 v14, v12
+; GFX9-O0-NEXT:    v_mov_b32_e32 v13, v11
+; GFX9-O0-NEXT:    buffer_store_dword v13, off, s[0:3], s32 offset:100 ; 4-byte Folded Spill
 ; GFX9-O0-NEXT:    s_waitcnt vmcnt(0)
-; GFX9-O0-NEXT:    buffer_store_dword v7, off, s[0:3], s32 offset:104 ; 4-byte Folded Spill
-; GFX9-O0-NEXT:    v_mov_b32_e32 v6, v8
-; GFX9-O0-NEXT:    v_mov_b32_e32 v7, v9
-; GFX9-O0-NEXT:    v_ashrrev_i64 v[6:7], s4, v[6:7]
-; GFX9-O0-NEXT:    v_mov_b32_e32 v1, v15
-; GFX9-O0-NEXT:    v_mov_b32_e32 v10, v13
-; GFX9-O0-NEXT:    v_xor_b32_e64 v1, v1, v10
-; GFX9-O0-NEXT:    v_mov_b32_e32 v11, v14
-; GFX9-O0-NEXT:    v_xor_b32_e64 v13, v11, v12
+; GFX9-O0-NEXT:    buffer_store_dword v14, off, s[0:3], s32 offset:104 ; 4-byte Folded Spill
+; GFX9-O0-NEXT:    v_mov_b32_e32 v14, v7
+; GFX9-O0-NEXT:    v_mov_b32_e32 v13, v6
+; GFX9-O0-NEXT:    v_ashrrev_i64 v[15:16], s4, v[13:14]
+; GFX9-O0-NEXT:    v_mov_b32_e32 v3, v9
+; GFX9-O0-NEXT:    v_mov_b32_e32 v10, v12
+; GFX9-O0-NEXT:    v_xor_b32_e64 v3, v3, v10
+; GFX9-O0-NEXT:    ; kill: def $vgpr8 killed $vgpr8 killed $vgpr8_vgpr9 killed $exec
+; GFX9-O0-NEXT:    v_mov_b32_e32 v12, v11
+; GFX9-O0-NEXT:    v_xor_b32_e64 v13, v8, v12
 ; GFX9-O0-NEXT:    ; kill: def $vgpr13 killed $vgpr13 def $vgpr13_vgpr14 killed $exec
-; GFX9-O0-NEXT:    v_mov_b32_e32 v14, v1
-; GFX9-O0-NEXT:    v_mov_b32_e32 v1, v5
-; GFX9-O0-NEXT:    v_xor_b32_e64 v1, v1, v10
-; GFX9-O0-NEXT:    ; kill: def $vgpr4 killed $vgpr4 killed $vgpr4_vgpr5 killed $exec
-; GFX9-O0-NEXT:    v_xor_b32_e64 v15, v4, v12
-; GFX9-O0-NEXT:    ; kill: def $vgpr15 killed $vgpr15 def $vgpr15_vgpr16 killed $exec
-; GFX9-O0-NEXT:    v_mov_b32_e32 v16, v1
-; GFX9-O0-NEXT:    v_mov_b32_e32 v1, v9
-; GFX9-O0-NEXT:    v_mov_b32_e32 v4, v7
-; GFX9-O0-NEXT:    v_xor_b32_e64 v1, v1, v4
-; GFX9-O0-NEXT:    v_mov_b32_e32 v5, v8
-; GFX9-O0-NEXT:    ; kill: def $vgpr6 killed $vgpr6 killed $vgpr6_vgpr7 killed $exec
-; GFX9-O0-NEXT:    v_xor_b32_e64 v7, v5, v6
+; GFX9-O0-NEXT:    v_mov_b32_e32 v14, v3
+; GFX9-O0-NEXT:    v_mov_b32_e32 v3, v2
+; GFX9-O0-NEXT:    v_xor_b32_e64 v3, v3, v10
+; GFX9-O0-NEXT:    ; kill: def $vgpr1 killed $vgpr1 killed $vgpr1_vgpr2 killed $exec
+; GFX9-O0-NEXT:    v_xor_b32_e64 v1, v1, v12
+; GFX9-O0-NEXT:    ; kill: def $vgpr1 killed $vgpr1 def $vgpr1_vgpr2 killed $exec
+; GFX9-O0-NEXT:    v_mov_b32_e32 v2, v3
+; GFX9-O0-NEXT:    v_mov_b32_e32 v8, v7
+; GFX9-O0-NEXT:    v_mov_b32_e32 v3, v16
+; GFX9-O0-NEXT:    v_xor_b32_e64 v9, v8, v3
+; GFX9-O0-NEXT:    v_mov_b32_e32 v7, v6
+; GFX9-O0-NEXT:    v_mov_b32_e32 v6, v15
+; GFX9-O0-NEXT:    v_xor_b32_e64 v7, v7, v6
 ; GFX9-O0-NEXT:    ; kill: def $vgpr7 killed $vgpr7 def $vgpr7_vgpr8 killed $exec
-; GFX9-O0-NEXT:    v_mov_b32_e32 v8, v1
-; GFX9-O0-NEXT:    v_mov_b32_e32 v1, v3
-; GFX9-O0-NEXT:    v_xor_b32_e64 v1, v1, v4
-; GFX9-O0-NEXT:    ; kill: def $vgpr2 killed $vgpr2 killed $vgpr2_vgpr3 killed $exec
-; GFX9-O0-NEXT:    v_xor_b32_e64 v2, v2, v6
-; GFX9-O0-NEXT:    ; kill: def $vgpr2 killed $vgpr2 def $vgpr2_vgpr3 killed $exec
-; GFX9-O0-NEXT:    v_mov_b32_e32 v3, v1
-; GFX9-O0-NEXT:    v_mov_b32_e32 v9, v15
-; GFX9-O0-NEXT:    v_mov_b32_e32 v1, v16
+; GFX9-O0-NEXT:    v_mov_b32_e32 v8, v9
+; GFX9-O0-NEXT:    v_mov_b32_e32 v9, v5
+; GFX9-O0-NEXT:    v_xor_b32_e64 v9, v9, v3
+; GFX9-O0-NEXT:    ; kill: def $vgpr4 killed $vgpr4 killed $vgpr4_vgpr5 killed $exec
+; GFX9-O0-NEXT:    v_xor_b32_e64 v4, v4, v6
+; GFX9-O0-NEXT:    ; kill: def $vgpr4 killed $vgpr4 def $vgpr4_vgpr5 killed $exec
+; GFX9-O0-NEXT:    v_mov_b32_e32 v5, v9
+; GFX9-O0-NEXT:    v_mov_b32_e32 v9, v1
+; GFX9-O0-NEXT:    ; kill: def $vgpr2 killed $vgpr2 killed $vgpr1_vgpr2 killed $exec
 ; GFX9-O0-NEXT:    v_mov_b32_e32 v11, v13
-; GFX9-O0-NEXT:    v_mov_b32_e32 v5, v14
+; GFX9-O0-NEXT:    v_mov_b32_e32 v1, v14
 ; GFX9-O0-NEXT:    v_sub_co_u32_e32 v9, vcc, v9, v12
-; GFX9-O0-NEXT:    v_subb_co_u32_e32 v1, vcc, v1, v10, vcc
+; GFX9-O0-NEXT:    v_subb_co_u32_e32 v2, vcc, v2, v10, vcc
 ; GFX9-O0-NEXT:    v_subb_co_u32_e32 v13, vcc, v11, v12, vcc
-; GFX9-O0-NEXT:    v_subb_co_u32_e32 v5, vcc, v5, v10, vcc
+; GFX9-O0-NEXT:    v_subb_co_u32_e32 v1, vcc, v1, v10, vcc
 ; GFX9-O0-NEXT:    ; implicit-def: $sgpr4
 ; GFX9-O0-NEXT:    ; implicit-def: $sgpr4
-; GFX9-O0-NEXT:    ; kill: def $vgpr13 killed $vgpr13 def $vgpr13_vgpr14 killed $exec
-; GFX9-O0-NEXT:    v_mov_b32_e32 v14, v5
+; GFX9-O0-NEXT:    ; kill: def $vgpr9 killed $vgpr9 def $vgpr9_vgpr10 killed $exec
+; GFX9-O0-NEXT:    v_mov_b32_e32 v10, v2
 ; GFX9-O0-NEXT:    ; implicit-def: $sgpr4
 ; GFX9-O0-NEXT:    ; implicit-def: $sgpr4
-; GFX9-O0-NEXT:    ; kill: def $vgpr9 killed $vgpr9 def $vgpr9_vgpr10 killed $exec
-; GFX9-O0-NEXT:    v_mov_b32_e32 v10, v1
-; GFX9-O0-NEXT:    v_mov_b32_e32 v1, v2
-; GFX9-O0-NEXT:    ; kill: def $vgpr3 killed $vgpr3 killed $vgpr2_vgpr3 killed $exec
+; GFX9-O0-NEXT:    ; kill: def $vgpr13 killed $vgpr13 def $vgpr13_vgpr14 killed $exec
+; GFX9-O0-NEXT:    v_mov_b32_e32 v14, v1
+; GFX9-O0-NEXT:    v_mov_b32_e32 v1, v4
+; GFX9-O0-NEXT:    v_mov_b32_e32 v4, v5
 ; GFX9-O0-NEXT:    v_mov_b32_e32 v5, v7
 ; GFX9-O0-NEXT:    v_mov_b32_e32 v2, v8
 ; GFX9-O0-NEXT:    v_sub_co_u32_e32 v1, vcc, v1, v6
-; GFX9-O0-NEXT:    v_subb_co_u32_e32 v3, vcc, v3, v4, vcc
+; GFX9-O0-NEXT:    v_subb_co_u32_e32 v4, vcc, v4, v3, vcc
 ; GFX9-O0-NEXT:    v_subb_co_u32_e32 v11, vcc, v5, v6, vcc
-; GFX9-O0-NEXT:    v_subb_co_u32_e32 v2, vcc, v2, v4, vcc
+; GFX9-O0-NEXT:    v_subb_co_u32_e32 v3, vcc, v2, v3, vcc
 ; GFX9-O0-NEXT:    ; implicit-def: $sgpr4
 ; GFX9-O0-NEXT:    ; implicit-def: $sgpr4
-; GFX9-O0-NEXT:    ; kill: def $vgpr11 killed $vgpr11 def $vgpr11_vgpr12 killed $exec
-; GFX9-O0-NEXT:    v_mov_b32_e32 v12, v2
+; GFX9-O0-NEXT:    ; kill: def $vgpr1 killed $vgpr1 def $vgpr1_vgpr2 killed $exec
+; GFX9-O0-NEXT:    v_mov_b32_e32 v2, v4
 ; GFX9-O0-NEXT:    ; implicit-def: $sgpr4
 ; GFX9-O0-NEXT:    ; implicit-def: $sgpr4
-; GFX9-O0-NEXT:    ; kill: def $vgpr1 killed $vgpr1 def $vgpr1_vgpr2 killed $exec
-; GFX9-O0-NEXT:    v_mov_b32_e32 v2, v3
-; GFX9-O0-NEXT:    buffer_store_dword v9, off, s[0:3], s32 offset:92 ; 4-byte Folded Spill
+; GFX9-O0-NEXT:    ; kill: def $vgpr11 killed $vgpr11 def $vgpr11_vgpr12 killed $exec
+; GFX9-O0-NEXT:    v_mov_b32_e32 v12, v3
+; GFX9-O0-NEXT:    v_mov_b32_e32 v3, v13
+; GFX9-O0-NEXT:    v_mov_b32_e32 v4, v14
+; GFX9-O0-NEXT:    buffer_store_dword v3, off, s[0:3], s32 offset:92 ; 4-byte Folded Spill
 ; GFX9-O0-NEXT:    s_waitcnt vmcnt(0)
-; GFX9-O0-NEXT:    buffer_store_dword v10, off, s[0:3], s32 offset:96 ; 4-byte Folded Spill
-; GFX9-O0-NEXT:    buffer_store_dword v13, off, s[0:3], s32 offset:84 ; 4-byte Folded Spill
+; GFX9-O0-NEXT:    buffer_store_dword v4, off, s[0:3], s32 offset:96 ; 4-byte Folded Spill
+; GFX9-O0-NEXT:    v_mov_b32_e32 v3, v9
+; GFX9-O0-NEXT:    v_mov_b32_e32 v4, v10
+; GFX9-O0-NEXT:    buffer_store_dword v3, off, s[0:3], s32 offset:84 ; 4-byte Folded Spill
 ; GFX9-O0-NEXT:    s_waitcnt vmcnt(0)
-; GFX9-O0-NEXT:    buffer_store_dword v14, off, s[0:3], s32 offset:88 ; 4-byte Folded Spill
-; GFX9-O0-NEXT:    buffer_store_dword v1, off, s[0:3], s32 offset:76 ; 4-byte Folded Spill
+; GFX9-O0-NEXT:    buffer_store_dword v4, off, s[0:3], s32 offset:88 ; 4-byte Folded Spill
+; GFX9-O0-NEXT:    v_mov_b32_e32 v3, v11
+; GFX9-O0-NEXT:    v_mov_b32_e32 v4, v12
+; GFX9-O0-NEXT:    buffer_store_dword v3, off, s[0:3], s32 offset:76 ; 4-byte Folded Spill
 ; GFX9-O0-NEXT:    s_waitcnt vmcnt(0)
-; GFX9-O0-NEXT:    buffer_store_dword v2, off, s[0:3], s32 offset:80 ; 4-byte Folded Spill
-; GFX9-O0-NEXT:    buffer_store_dword v11, off, s[0:3], s32 offset:68 ; 4-byte Folded Spill
+; GFX9-O0-NEXT:    buffer_store_dword v4, off, s[0:3], s32 offset:80 ; 4-byte Folded Spill
+; GFX9-O0-NEXT:    v_mov_b32_e32 v4, v2
+; GFX9-O0-NEXT:    v_mov_b32_e32 v3, v1
+; GFX9-O0-NEXT:    buffer_store_dword v3, off, s[0:3], s32 offset:68 ; 4-byte Folded Spill
 ; GFX9-O0-NEXT:    s_waitcnt vmcnt(0)
-; GFX9-O0-NEXT:    buffer_store_dword v12, off, s[0:3], s32 offset:72 ; 4-byte Folded Spill
+; GFX9-O0-NEXT:    buffer_store_dword v4, off, s[0:3], s32 offset:72 ; 4-byte Folded Spill
 ; GFX9-O0-NEXT:    v_mov_b32_e32 v3, v11
 ; GFX9-O0-NEXT:    v_mov_b32_e32 v4, v12
 ; GFX9-O0-NEXT:    buffer_store_dword v3, off, s[0:3], s32 offset:60 ; 4-byte Folded Spill
@@ -438,7 +445,8 @@ define i128 @v_srem_i128_vv(i128 %lhs, i128 %rhs) {
 ; GFX9-O0-NEXT:    ; kill: def $vgpr8 killed $vgpr8 def $vgpr8_vgpr9 killed $exec
 ; GFX9-O0-NEXT:    v_mov_b32_e32 v9, v5
 ; GFX9-O0-NEXT:    v_mov_b32_e32 v5, v9
-; GFX9-O0-NEXT:    v_cmp_ne_u64_e64 s[12:13], v[11:12], s[6:7]
+; GFX9-O0-NEXT:    s_mov_b64 s[12:13], s[6:7]
+; GFX9-O0-NEXT:    v_cmp_ne_u64_e64 s[12:13], v[11:12], s[12:13]
 ; GFX9-O0-NEXT:    v_cndmask_b32_e64 v5, v5, v10, s[12:13]
 ; GFX9-O0-NEXT:    v_mov_b32_e32 v7, v6
 ; GFX9-O0-NEXT:    v_mov_b32_e32 v6, v8
@@ -474,7 +482,8 @@ define i128 @v_srem_i128_vv(i128 %lhs, i128 %rhs) {
 ; GFX9-O0-NEXT:    ; kill: def $vgpr11 killed $vgpr11 def $vgpr11_vgpr12 killed $exec
 ; GFX9-O0-NEXT:    v_mov_b32_e32 v12, v5
 ; GFX9-O0-NEXT:    v_mov_b32_e32 v5, v12
-; GFX9-O0-NEXT:    v_cmp_ne_u64_e64 s[8:9], v[13:14], s[6:7]
+; GFX9-O0-NEXT:    s_mov_b64 s[8:9], s[6:7]
+; GFX9-O0-NEXT:    v_cmp_ne_u64_e64 s[8:9], v[13:14], s[8:9]
 ; GFX9-O0-NEXT:    v_cndmask_b32_e64 v5, v5, v8, s[8:9]
 ; GFX9-O0-NEXT:    v_mov_b32_e32 v7, v6
 ; GFX9-O0-NEXT:    v_mov_b32_e32 v6, v11
@@ -589,27 +598,27 @@ define i128 @v_srem_i128_vv(i128 %lhs, i128 %rhs) {
 ; GFX9-O0-NEXT:    v_readlane_b32 s5, v0, 5
 ; GFX9-O0-NEXT:    s_or_b64 exec, exec, s[4:5]
 ; GFX9-O0-NEXT:  ; %bb.2: ; %Flow
-; GFX9-O0-NEXT:    buffer_load_dword v6, off, s[0:3], s32 offset:156 ; 4-byte Folded Reload
-; GFX9-O0-NEXT:    buffer_load_dword v7, off, s[0:3], s32 offset:160 ; 4-byte Folded Reload
-; GFX9-O0-NEXT:    buffer_load_dword v4, off, s[0:3], s32 offset:164 ; 4-byte Folded Reload
-; GFX9-O0-NEXT:    buffer_load_dword v5, off, s[0:3], s32 offset:168 ; 4-byte Folded Reload
-; GFX9-O0-NEXT:    buffer_load_dword v2, off, s[0:3], s32 offset:172 ; 4-byte Folded Reload
-; GFX9-O0-NEXT:    buffer_load_dword v3, off, s[0:3], s32 offset:176 ; 4-byte Folded Reload
-; GFX9-O0-NEXT:    buffer_load_dword v0, off, s[0:3], s32 offset:180 ; 4-byte Folded Reload
-; GFX9-O0-NEXT:    buffer_load_dword v1, off, s[0:3], s32 offset:184 ; 4-byte Folded Reload
+; GFX9-O0-NEXT:    buffer_load_dword v6, off, s[0:3], s32 offset:152 ; 4-byte Folded Reload
+; GFX9-O0-NEXT:    buffer_load_dword v7, off, s[0:3], s32 offset:156 ; 4-byte Folded Reload
+; GFX9-O0-NEXT:    buffer_load_dword v4, off, s[0:3], s32 offset:160 ; 4-byte Folded Reload
+; GFX9-O0-NEXT:    buffer_load_dword v5, off, s[0:3], s32 offset:164 ; 4-byte Folded Reload
+; GFX9-O0-NEXT:    buffer_load_dword v2, off, s[0:3], s32 offset:168 ; 4-byte Folded Reload
+; GFX9-O0-NEXT:    buffer_load_dword v3, off, s[0:3], s32 offset:172 ; 4-byte Folded Reload
+; GFX9-O0-NEXT:    buffer_load_dword v0, off, s[0:3], s32 offset:176 ; 4-byte Folded Reload
+; GFX9-O0-NEXT:    buffer_load_dword v1, off, s[0:3], s32 offset:180 ; 4-byte Folded Reload
 ; GFX9-O0-NEXT:    s_waitcnt vmcnt(6)
-; GFX9-O0-NEXT:    buffer_store_dword v6, off, s[0:3], s32 offset:148 ; 4-byte Folded Spill
+; GFX9-O0-NEXT:    buffer_store_dword v6, off, s[0:3], s32 offset:144 ; 4-byte Folded Spill
 ; GFX9-O0-NEXT:    s_waitcnt vmcnt(0)
-; GFX9-O0-NEXT:    buffer_store_dword v7, off, s[0:3], s32 offset:152 ; 4-byte Folded Spill
-; GFX9-O0-NEXT:    buffer_store_dword v4, off, s[0:3], s32 offset:140 ; 4-byte Folded Spill
+; GFX9-O0-NEXT:    buffer_store_dword v7, off, s[0:3], s32 offset:148 ; 4-byte Folded Spill
+; GFX9-O0-NEXT:    buffer_store_dword v4, off, s[0:3], s32 offset:136 ; 4-byte Folded Spill
 ; GFX9-O0-NEXT:    s_waitcnt vmcnt(0)
-; GFX9-O0-NEXT:    buffer_store_dword v5, off, s[0:3], s32 offset:144 ; 4-byte Folded Spill
-; GFX9-O0-NEXT:    buffer_store_dword v2, off, s[0:3], s32 offset:132 ; 4-byte Folded Spill
+; GFX9-O0-NEXT:    buffer_store_dword v5, off, s[0:3], s32 offset:140 ; 4-byte Folded Spill
+; GFX9-O0-NEXT:    buffer_store_dword v2, off, s[0:3], s32 offset:128 ; 4-byte Folded Spill
 ; GFX9-O0-NEXT:    s_waitcnt vmcnt(0)
-; GFX9-O0-NEXT:    buffer_store_dword v3, off, s[0:3], s32 offset:136 ; 4-byte Folded Spill
-; GFX9-O0-NEXT:    buffer_store_dword v0, off, s[0:3], s32 offset:124 ; 4-byte Folded Spill
+; GFX9-O0-NEXT:    buffer_store_dword v3, off, s[0:3], s32 offset:132 ; 4-byte Folded Spill
+; GFX9-O0-NEXT:    buffer_store_dword v0, off, s[0:3], s32 offset:120 ; 4-byte Folded Spill
 ; GFX9-O0-NEXT:    s_waitcnt vmcnt(0)
-; GFX9-O0-NEXT:    buffer_store_dword v1, off, s[0:3], s32 offset:128 ; 4-byte Folded Spill
+; GFX9-O0-NEXT:    buffer_store_dword v1, off, s[0:3], s32 offset:124 ; 4-byte Folded Spill
 ; GFX9-O0-NEXT:    s_branch .LBB0_5
 ; GFX9-O0-NEXT:  .LBB0_3: ; %Flow2
 ; GFX9-O0-NEXT:    s_or_saveexec_b64 s[18:19], -1
@@ -624,22 +633,22 @@ define i128 @v_srem_i128_vv(i128 %lhs, i128 %rhs) {
 ; GFX9-O0-NEXT:    buffer_load_dword v2, off, s[0:3], s32 offset:4 ; 4-byte Folded Reload
 ; GFX9-O0-NEXT:    buffer_load_dword v3, off, s[0:3], s32 offset:8 ; 4-byte Folded Reload
 ; GFX9-O0-NEXT:    s_waitcnt vmcnt(0)
-; GFX9-O0-NEXT:    buffer_store_dword v2, off, s[0:3], s32 offset:196 ; 4-byte Folded Spill
+; GFX9-O0-NEXT:    buffer_store_dword v2, off, s[0:3], s32 offset:192 ; 4-byte Folded Spill
 ; GFX9-O0-NEXT:    s_waitcnt vmcnt(0)
-; GFX9-O0-NEXT:    buffer_store_dword v3, off, s[0:3], s32 offset:200 ; 4-byte Folded Spill
-; GFX9-O0-NEXT:    buffer_store_dword v0, off, s[0:3], s32 offset:188 ; 4-byte Folded Spill
+; GFX9-O0-NEXT:    buffer_store_dword v3, off, s[0:3], s32 offset:196 ; 4-byte Folded Spill
+; GFX9-O0-NEXT:    buffer_store_dword v0, off, s[0:3], s32 offset:184 ; 4-byte Folded Spill
 ; GFX9-O0-NEXT:    s_waitcnt vmcnt(0)
-; GFX9-O0-NEXT:    buffer_store_dword v1, off, s[0:3], s32 offset:192 ; 4-byte Folded Spill
+; GFX9-O0-NEXT:    buffer_store_dword v1, off, s[0:3], s32 offset:188 ; 4-byte Folded Spill
 ; GFX9-O0-NEXT:    s_branch .LBB0_9
 ; GFX9-O0-NEXT:  .LBB0_4: ; %udiv-loop-exit
-; GFX9-O0-NEXT:    buffer_load_dword v5, off, s[0:3], s32 offset:204 ; 4-byte Folded Reload
-; GFX9-O0-NEXT:    buffer_load_dword v6, off, s[0:3], s32 offset:208 ; 4-byte Folded Reload
-; GFX9-O0-NEXT:    buffer_load_dword v7, off, s[0:3], s32 offset:212 ; 4-byte Folded Reload
-; GFX9-O0-NEXT:    buffer_load_dword v8, off, s[0:3], s32 offset:216 ; 4-byte Folded Reload
-; GFX9-O0-NEXT:    buffer_load_dword v0, off, s[0:3], s32 offset:220 ; 4-byte Folded Reload
-; GFX9-O0-NEXT:    buffer_load_dword v1, off, s[0:3], s32 offset:224 ; 4-byte Folded Reload
-; GFX9-O0-NEXT:    buffer_load_dword v9, off, s[0:3], s32 offset:228 ; 4-byte Folded Reload
-; GFX9-O0-NEXT:    buffer_load_dword v10, off, s[0:3], s32 offset:232 ; 4-byte Folded Reload
+; GFX9-O0-NEXT:    buffer_load_dword v5, off, s[0:3], s32 offset:200 ; 4-byte Folded Reload
+; GFX9-O0-NEXT:    buffer_load_dword v6, off, s[0:3], s32 offset:204 ; 4-byte Folded Reload
+; GFX9-O0-NEXT:    buffer_load_dword v7, off, s[0:3], s32 offset:208 ; 4-byte Folded Reload
+; GFX9-O0-NEXT:    buffer_load_dword v8, off, s[0:3], s32 offset:212 ; 4-byte Folded Reload
+; GFX9-O0-NEXT:    buffer_load_dword v0, off, s[0:3], s32 offset:216 ; 4-byte Folded Reload
+; GFX9-O0-NEXT:    buffer_load_dword v1, off, s[0:3], s32 offset:220 ; 4-byte Folded Reload
+; GFX9-O0-NEXT:    buffer_load_dword v9, off, s[0:3], s32 offset:224 ; 4-byte Folded Reload
+; GFX9-O0-NEXT:    buffer_load_dword v10, off, s[0:3], s32 offset:228 ; 4-byte Folded Reload
 ; GFX9-O0-NEXT:    s_mov_b32 s4, 1
 ; GFX9-O0-NEXT:    s_waitcnt vmcnt(2)
 ; GFX9-O0-NEXT:    v_lshlrev_b64 v[2:3], s4, v[0:1]
@@ -679,27 +688,27 @@ define i128 @v_srem_i128_vv(i128 %lhs, i128 %rhs) {
 ; GFX9-O0-NEXT:    v_readlane_b32 s4, v8, 6
 ; GFX9-O0-NEXT:    v_readlane_b32 s5, v8, 7
 ; GFX9-O0-NEXT:    s_or_b64 exec, exec, s[4:5]
-; GFX9-O0-NEXT:    buffer_load_dword v0, off, s[0:3], s32 offset:148 ; 4-byte Folded Reload
-; GFX9-O0-NEXT:    buffer_load_dword v1, off, s[0:3], s32 offset:152 ; 4-byte Folded Reload
-; GFX9-O0-NEXT:    buffer_load_dword v2, off, s[0:3], s32 offset:140 ; 4-byte Folded Reload
-; GFX9-O0-NEXT:    buffer_load_dword v3, off, s[0:3], s32 offset:144 ; 4-byte Folded Reload
-; GFX9-O0-NEXT:    buffer_load_dword v4, off, s[0:3], s32 offset:132 ; 4-byte Folded Reload
-; GFX9-O0-NEXT:    buffer_load_dword v5, off, s[0:3], s32 offset:136 ; 4-byte Folded Reload
-; GFX9-O0-NEXT:    buffer_load_dword v6, off, s[0:3], s32 offset:124 ; 4-byte Folded Reload
-; GFX9-O0-NEXT:    buffer_load_dword v7, off, s[0:3], s32 offset:128 ; 4-byte Folded Reload
+; GFX9-O0-NEXT:    buffer_load_dword v0, off, s[0:3], s32 offset:144 ; 4-byte Folded Reload
+; GFX9-O0-NEXT:    buffer_load_dword v1, off, s[0:3], s32 offset:148 ; 4-byte Folded Reload
+; GFX9-O0-NEXT:    buffer_load_dword v2, off, s[0:3], s32 offset:136 ; 4-byte Folded Reload
+; GFX9-O0-NEXT:    buffer_load_dword v3, off, s[0:3], s32 offset:140 ; 4-byte Folded Reload
+; GFX9-O0-NEXT:    buffer_load_dword v4, off, s[0:3], s32 offset:128 ; 4-byte Folded Reload
+; GFX9-O0-NEXT:    buffer_load_dword v5, off, s[0:3], s32 offset:132 ; 4-byte Folded Reload
+; GFX9-O0-NEXT:    buffer_load_dword v6, off, s[0:3], s32 offset:120 ; 4-byte Folded Reload
+; GFX9-O0-NEXT:    buffer_load_dword v7, off, s[0:3], s32 offset:124 ; 4-byte Folded Reload
 ; GFX9-O0-NEXT:    s_waitcnt vmcnt(0)
-; GFX9-O0-NEXT:    buffer_store_dword v6, off, s[0:3], s32 offset:212 ; 4-byte Folded Spill
+; GFX9-O0-NEXT:    buffer_store_dword v6, off, s[0:3], s32 offset:208 ; 4-byte Folded Spill
 ; GFX9-O0-NEXT:    s_waitcnt vmcnt(0)
-; GFX9-O0-NEXT:    buffer_store_dword v7, off, s[0:3], s32 offset:216 ; 4-byte Folded Spill
-; GFX9-O0-NEXT:    buffer_store_dword v4, off, s[0:3], s32 offset:204 ; 4-byte Folded Spill
+; GFX9-O0-NEXT:    buffer_store_dword v7, off, s[0:3], s32 offset:212 ; 4-byte Folded Spill
+; GFX9-O0-NEXT:    buffer_store_dword v4, off, s[0:3], s32 offset:200 ; 4-byte Folded Spill
 ; GFX9-O0-NEXT:    s_waitcnt vmcnt(0)
-; GFX9-O0-NEXT:    buffer_store_dword v5, off, s[0:3], s32 offset:208 ; 4-byte Folded Spill
-; GFX9-O0-NEXT:    buffer_store_dword v2, off, s[0:3], s32 offset:228 ; 4-byte Folded Spill
+; GFX9-O0-NEXT:    buffer_store_dword v5, off, s[0:3], s32 offset:204 ; 4-byte Folded Spill
+; GFX9-O0-NEXT:    buffer_store_dword v2, off, s[0:3], s32 offset:224 ; 4-byte Folded Spill
 ; GFX9-O0-NEXT:    s_waitcnt vmcnt(0)
-; GFX9-O0-NEXT:    buffer_store_dword v3, off, s[0:3], s32 offset:232 ; 4-byte Folded Spill
-; GFX9-O0-NEXT:    buffer_store_dword v0, off, s[0:3], s32 offset:220 ; 4-byte Folded Spill
+; GFX9-O0-NEXT:    buffer_store_dword v3, off, s[0:3], s32 offset:228 ; 4-byte Folded Spill
+; GFX9-O0-NEXT:    buffer_store_dword v0, off, s[0:3], s32 offset:216 ; 4-byte Folded Spill
 ; GFX9-O0-NEXT:    s_waitcnt vmcnt(0)
-; GFX9-O0-NEXT:    buffer_store_dword v1, off, s[0:3], s32 offset:224 ; 4-byte Folded Spill
+; GFX9-O0-NEXT:    buffer_store_dword v1, off, s[0:3], s32 offset:220 ; 4-byte Folded Spill
 ; GFX9-O0-NEXT:    s_branch .LBB0_4
 ; GFX9-O0-NEXT:  .LBB0_6: ; %udiv-do-while
 ; GFX9-O0-NEXT:    ; =>This Inner Loop Header: Depth=1
@@ -709,30 +718,30 @@ define i128 @v_srem_i128_vv(i128 %lhs, i128 %rhs) {
 ; GFX9-O0-NEXT:    s_waitcnt vmcnt(0)
 ; GFX9-O0-NEXT:    v_readlane_b32 s6, v16, 8
 ; GFX9-O0-NEXT:    v_readlane_b32 s7, v16, 9
-; GFX9-O0-NEXT:    buffer_load_dword v6, off, s[0:3], s32 offset:236 ; 4-byte Folded Reload
-; GFX9-O0-NEXT:    buffer_load_dword v7, off, s[0:3], s32 offset:240 ; 4-byte Folded Reload
-; GFX9-O0-NEXT:    buffer_load_dword v0, off, s[0:3], s32 offset:244 ; 4-byte Folded Reload
-; GFX9-O0-NEXT:    buffer_load_dword v1, off, s[0:3], s32 offset:248 ; 4-byte Folded Reload
-; GFX9-O0-NEXT:    buffer_load_dword v23, off, s[0:3], s32 offset:252 ; 4-byte Folded Reload
-; GFX9-O0-NEXT:    buffer_load_dword v24, off, s[0:3], s32 offset:256 ; 4-byte Folded Reload
-; GFX9-O0-NEXT:    buffer_load_dword v2, off, s[0:3], s32 offset:260 ; 4-byte Folded Reload
-; GFX9-O0-NEXT:    buffer_load_dword v3, off, s[0:3], s32 offset:264 ; 4-byte Folded Reload
-; GFX9-O0-NEXT:    buffer_load_dword v17, off, s[0:3], s32 offset:268 ; 4-byte Folded Reload
-; GFX9-O0-NEXT:    buffer_load_dword v18, off, s[0:3], s32 offset:272 ; 4-byte Folded Reload
-; GFX9-O0-NEXT:    buffer_load_dword v8, off, s[0:3], s32 offset:276 ; 4-byte Folded Reload
-; GFX9-O0-NEXT:    buffer_load_dword v9, off, s[0:3], s32 offset:280 ; 4-byte Folded Reload
-; GFX9-O0-NEXT:    buffer_load_dword v27, off, s[0:3], s32 offset:284 ; 4-byte Folded Reload
-; GFX9-O0-NEXT:    buffer_load_dword v28, off, s[0:3], s32 offset:288 ; 4-byte Folded Reload
-; GFX9-O0-NEXT:    buffer_load_dword v25, off, s[0:3], s32 offset:292 ; 4-byte Folded Reload
-; GFX9-O0-NEXT:    buffer_load_dword v26, off, s[0:3], s32 offset:296 ; 4-byte Folded Reload
+; GFX9-O0-NEXT:    buffer_load_dword v6, off, s[0:3], s32 offset:232 ; 4-byte Folded Reload
+; GFX9-O0-NEXT:    buffer_load_dword v7, off, s[0:3], s32 offset:236 ; 4-byte Folded Reload
+; GFX9-O0-NEXT:    buffer_load_dword v0, off, s[0:3], s32 offset:240 ; 4-byte Folded Reload
+; GFX9-O0-NEXT:    buffer_load_dword v1, off, s[0:3], s32 offset:244 ; 4-byte Folded Reload
+; GFX9-O0-NEXT:    buffer_load_dword v23, off, s[0:3], s32 offset:248 ; 4-byte Folded Reload
+; GFX9-O0-NEXT:    buffer_load_dword v24, off, s[0:3], s32 offset:252 ; 4-byte Folded Reload
+; GFX9-O0-NEXT:    buffer_load_dword v2, off, s[0:3], s32 offset:256 ; 4-byte Folded Reload
+; GFX9-O0-NEXT:    buffer_load_dword v3, off, s[0:3], s32 offset:260 ; 4-byte Folded Reload
+; GFX9-O0-NEXT:    buffer_load_dword v17, off, s[0:3], s32 offset:264 ; 4-byte Folded Reload
+; GFX9-O0-NEXT:    buffer_load_dword v18, off, s[0:3], s32 offset:268 ; 4-byte Folded Reload
+; GFX9-O0-NEXT:    buffer_load_dword v8, off, s[0:3], s32 offset:272 ; 4-byte Folded Reload
+; GFX9-O0-NEXT:    buffer_load_dword v9, off, s[0:3], s32 offset:276 ; 4-byte Folded Reload
+; GFX9-O0-NEXT:    buffer_load_dword v27, off, s[0:3], s32 offset:280 ; 4-byte Folded Reload
+; GFX9-O0-NEXT:    buffer_load_dword v28, off, s[0:3], s32 offset:284 ; 4-byte Folded Reload
+; GFX9-O0-NEXT:    buffer_load_dword v25, off, s[0:3], s32 offset:288 ; 4-byte Folded Reload
+; GFX9-O0-NEXT:    buffer_load_dword v26, off, s[0:3], s32 offset:292 ; 4-byte Folded Reload
 ; GFX9-O0-NEXT:    buffer_load_dword v19, off, s[0:3], s32 offset:52 ; 4-byte Folded Reload
 ; GFX9-O0-NEXT:    buffer_load_dword v20, off, s[0:3], s32 offset:56 ; 4-byte Folded Reload
 ; GFX9-O0-NEXT:    buffer_load_dword v21, off, s[0:3], s32 offset:60 ; 4-byte Folded Reload
 ; GFX9-O0-NEXT:    buffer_load_dword v22, off, s[0:3], s32 offset:64 ; 4-byte Folded Reload
-; GFX9-O0-NEXT:    buffer_load_dword v14, off, s[0:3], s32 offset:300 ; 4-byte Folded Reload
-; GFX9-O0-NEXT:    buffer_load_dword v15, off, s[0:3], s32 offset:304 ; 4-byte Folded Reload
-; GFX9-O0-NEXT:    buffer_load_dword v11, off, s[0:3], s32 offset:308 ; 4-byte Folded Reload
-; GFX9-O0-NEXT:    buffer_load_dword v12, off, s[0:3], s32 offset:312 ; 4-byte Folded Reload
+; GFX9-O0-NEXT:    buffer_load_dword v14, off, s[0:3], s32 offset:296 ; 4-byte Folded Reload
+; GFX9-O0-NEXT:    buffer_load_dword v15, off, s[0:3], s32 offset:300 ; 4-byte Folded Reload
+; GFX9-O0-NEXT:    buffer_load_dword v11, off, s[0:3], s32 offset:304 ; 4-byte Folded Reload
+; GFX9-O0-NEXT:    buffer_load_dword v12, off, s[0:3], s32 offset:308 ; 4-byte Folded Reload
 ; GFX9-O0-NEXT:    s_mov_b32 s4, 63
 ; GFX9-O0-NEXT:    s_waitcnt vmcnt(16)
 ; GFX9-O0-NEXT:    v_lshrrev_b64 v[29:30], s4, v[2:3]
@@ -872,24 +881,24 @@ define i128 @v_srem_i128_vv(i128 %lhs, i128 %rhs) {
 ; GFX9-O0-NEXT:    s_or_b64 s[4:5], s[4:5], s[6:7]
 ; GFX9-O0-NEXT:    v_mov_b32_e32 v18, v3
 ; GFX9-O0-NEXT:    v_mov_b32_e32 v17, v2
-; GFX9-O0-NEXT:    buffer_store_dword v17, off, s[0:3], s32 offset:156 ; 4-byte Folded Spill
+; GFX9-O0-NEXT:    buffer_store_dword v17, off, s[0:3], s32 offset:152 ; 4-byte Folded Spill
 ; GFX9-O0-NEXT:    s_waitcnt vmcnt(0)
-; GFX9-O0-NEXT:    buffer_store_dword v18, off, s[0:3], s32 offset:160 ; 4-byte Folded Spill
+; GFX9-O0-NEXT:    buffer_store_dword v18, off, s[0:3], s32 offset:156 ; 4-byte Folded Spill
 ; GFX9-O0-NEXT:    v_mov_b32_e32 v18, v1
 ; GFX9-O0-NEXT:    v_mov_b32_e32 v17, v0
-; GFX9-O0-NEXT:    buffer_store_dword v17, off, s[0:3], s32 offset:164 ; 4-byte Folded Spill
+; GFX9-O0-NEXT:    buffer_store_dword v17, off, s[0:3], s32 offset:160 ; 4-byte Folded Spill
 ; GFX9-O0-NEXT:    s_waitcnt vmcnt(0)
-; GFX9-O0-NEXT:    buffer_store_dword v18, off, s[0:3], s32 offset:168 ; 4-byte Folded Spill
+; GFX9-O0-NEXT:    buffer_store_dword v18, off, s[0:3], s32 offset:164 ; 4-byte Folded Spill
 ; GFX9-O0-NEXT:    v_mov_b32_e32 v18, v15
 ; GFX9-O0-NEXT:    v_mov_b32_e32 v17, v14
-; GFX9-O0-NEXT:    buffer_store_dword v17, off, s[0:3], s32 offset:172 ; 4-byte Folded Spill
+; GFX9-O0-NEXT:    buffer_store_dword v17, off, s[0:3], s32 offset:168 ; 4-byte Folded Spill
 ; GFX9-O0-NEXT:    s_waitcnt vmcnt(0)
-; GFX9-O0-NEXT:    buffer_store_dword v18, off, s[0:3], s32 offset:176 ; 4-byte Folded Spill
+; GFX9-O0-NEXT:    buffer_store_dword v18, off, s[0:3], s32 offset:172 ; 4-byte Folded Spill
 ; GFX9-O0-NEXT:    v_mov_b32_e32 v18, v13
 ; GFX9-O0-NEXT:    v_mov_b32_e32 v17, v12
-; GFX9-O0-NEXT:    buffer_store_dword v17, off, s[0:3], s32 offset:180 ; 4-byte Folded Spill
+; GFX9-O0-NEXT:    buffer_store_dword v17, off, s[0:3], s32 offset:176 ; 4-byte Folded Spill
 ; GFX9-O0-NEXT:    s_waitcnt vmcnt(0)
-; GFX9-O0-NEXT:    buffer_store_dword v18, off, s[0:3], s32 offset:184 ; 4-byte Folded Spill
+; GFX9-O0-NEXT:    buffer_store_dword v18, off, s[0:3], s32 offset:180 ; 4-byte Folded Spill
 ; GFX9-O0-NEXT:    s_mov_b64 s[6:7], s[4:5]
 ; GFX9-O0-NEXT:    v_writelane_b32 v16, s6, 4
 ; GFX9-O0-NEXT:    v_writelane_b32 v16, s7, 5
@@ -899,42 +908,42 @@ define i128 @v_srem_i128_vv(i128 %lhs, i128 %rhs) {
 ; GFX9-O0-NEXT:    s_or_saveexec_b64 s[18:19], -1
 ; GFX9-O0-NEXT:    buffer_store_dword v16, off, s[0:3], s32 ; 4-byte Folded Spill
 ; GFX9-O0-NEXT:    s_mov_b64 exec, s[18:19]
-; GFX9-O0-NEXT:    buffer_store_dword v14, off, s[0:3], s32 offset:292 ; 4-byte Folded Spill
+; GFX9-O0-NEXT:    buffer_store_dword v14, off, s[0:3], s32 offset:288 ; 4-byte Folded Spill
 ; GFX9-O0-NEXT:    s_waitcnt vmcnt(0)
-; GFX9-O0-NEXT:    buffer_store_dword v15, off, s[0:3], s32 offset:296 ; 4-byte Folded Spill
-; GFX9-O0-NEXT:    buffer_store_dword v12, off, s[0:3], s32 offset:284 ; 4-byte Folded Spill
+; GFX9-O0-NEXT:    buffer_store_dword v15, off, s[0:3], s32 offset:292 ; 4-byte Folded Spill
+; GFX9-O0-NEXT:    buffer_store_dword v12, off, s[0:3], s32 offset:280 ; 4-byte Folded Spill
 ; GFX9-O0-NEXT:    s_waitcnt vmcnt(0)
-; GFX9-O0-NEXT:    buffer_store_dword v13, off, s[0:3], s32 offset:288 ; 4-byte Folded Spill
-; GFX9-O0-NEXT:    buffer_store_dword v10, off, s[0:3], s32 offset:276 ; 4-byte Folded Spill
+; GFX9-O0-NEXT:    buffer_store_dword v13, off, s[0:3], s32 offset:284 ; 4-byte Folded Spill
+; GFX9-O0-NEXT:    buffer_store_dword v10, off, s[0:3], s32 offset:272 ; 4-byte Folded Spill
 ; GFX9-O0-NEXT:    s_waitcnt vmcnt(0)
-; GFX9-O0-NEXT:    buffer_store_dword v11, off, s[0:3], s32 offset:280 ; 4-byte Folded Spill
-; GFX9-O0-NEXT:    buffer_store_dword v8, off, s[0:3], s32 offset:268 ; 4-byte Folded Spill
+; GFX9-O0-NEXT:    buffer_store_dword v11, off, s[0:3], s32 offset:276 ; 4-byte Folded Spill
+; GFX9-O0-NEXT:    buffer_store_dword v8, off, s[0:3], s32 offset:264 ; 4-byte Folded Spill
 ; GFX9-O0-NEXT:    s_waitcnt vmcnt(0)
-; GFX9-O0-NEXT:    buffer_store_dword v9, off, s[0:3], s32 offset:272 ; 4-byte Folded Spill
-; GFX9-O0-NEXT:    buffer_store_dword v6, off, s[0:3], s32 offset:260 ; 4-byte Folded Spill
+; GFX9-O0-NEXT:    buffer_store_dword v9, off, s[0:3], s32 offset:268 ; 4-byte Folded Spill
+; GFX9-O0-NEXT:    buffer_store_dword v6, off, s[0:3], s32 offset:256 ; 4-byte Folded Spill
 ; GFX9-O0-NEXT:    s_waitcnt vmcnt(0)
-; GFX9-O0-NEXT:    buffer_store_dword v7, off, s[0:3], s32 offset:264 ; 4-byte Folded Spill
-; GFX9-O0-NEXT:    buffer_store_dword v4, off, s[0:3], s32 offset:252 ; 4-byte Folded Spill
+; GFX9-O0-NEXT:    buffer_store_dword v7, off, s[0:3], s32 offset:260 ; 4-byte Folded Spill
+; GFX9-O0-NEXT:    buffer_store_dword v4, off, s[0:3], s32 offset:248 ; 4-byte Folded Spill
 ; GFX9-O0-NEXT:    s_waitcnt vmcnt(0)
-; GFX9-O0-NEXT:    buffer_store_dword v5, off, s[0:3], s32 offset:256 ; 4-byte Folded Spill
-; GFX9-O0-NEXT:    buffer_store_dword v2, off, s[0:3], s32 offset:244 ; 4-byte Folded Spill
+; GFX9-O0-NEXT:    buffer_store_dword v5, off, s[0:3], s32 offset:252 ; 4-byte Folded Spill
+; GFX9-O0-NEXT:    buffer_store_dword v2, off, s[0:3], s32 offset:240 ; 4-byte Folded Spill
 ; GFX9-O0-NEXT:    s_waitcnt vmcnt(0)
-; GFX9-O0-NEXT:    buffer_store_dword v3, off, s[0:3], s32 offset:248 ; 4-byte Folded Spill
-; GFX9-O0-NEXT:    buffer_store_dword v0, off, s[0:3], s32 offset:236 ; 4-byte Folded Spill
+; GFX9-O0-NEXT:    buffer_store_dword v3, off, s[0:3], s32 offset:244 ; 4-byte Folded Spill
+; GFX9-O0-NEXT:    buffer_store_dword v0, off, s[0:3], s32 offset:232 ; 4-byte Folded Spill
 ; GFX9-O0-NEXT:    s_waitcnt vmcnt(0)
-; GFX9-O0-NEXT:    buffer_store_dword v1, off, s[0:3], s32 offset:240 ; 4-byte Folded Spill
+; GFX9-O0-NEXT:    buffer_store_dword v1, off, s[0:3], s32 offset:236 ; 4-byte Folded Spill
 ; GFX9-O0-NEXT:    s_andn2_b64 exec, exec, s[4:5]
 ; GFX9-O0-NEXT:    s_cbranch_execnz .LBB0_6
 ; GFX9-O0-NEXT:    s_branch .LBB0_1
 ; GFX9-O0-NEXT:  .LBB0_7: ; %udiv-preheader
-; GFX9-O0-NEXT:    buffer_load_dword v0, off, s[0:3], s32 offset:316 ; 4-byte Folded Reload
-; GFX9-O0-NEXT:    buffer_load_dword v1, off, s[0:3], s32 offset:320 ; 4-byte Folded Reload
-; GFX9-O0-NEXT:    buffer_load_dword v2, off, s[0:3], s32 offset:324 ; 4-byte Folded Reload
-; GFX9-O0-NEXT:    buffer_load_dword v3, off, s[0:3], s32 offset:328 ; 4-byte Folded Reload
-; GFX9-O0-NEXT:    buffer_load_dword v8, off, s[0:3], s32 offset:332 ; 4-byte Folded Reload
-; GFX9-O0-NEXT:    buffer_load_dword v9, off, s[0:3], s32 offset:336 ; 4-byte Folded Reload
-; GFX9-O0-NEXT:    buffer_load_dword v10, off, s[0:3], s32 offset:340 ; 4-byte Folded Reload
-; GFX9-O0-NEXT:    buffer_load_dword v11, off, s[0:3], s32 offset:344 ; 4-byte Folded Reload
+; GFX9-O0-NEXT:    buffer_load_dword v0, off, s[0:3], s32 offset:312 ; 4-byte Folded Reload
+; GFX9-O0-NEXT:    buffer_load_dword v1, off, s[0:3], s32 offset:316 ; 4-byte Folded Reload
+; GFX9-O0-NEXT:    buffer_load_dword v2, off, s[0:3], s32 offset:320 ; 4-byte Folded Reload
+; GFX9-O0-NEXT:    buffer_load_dword v3, off, s[0:3], s32 offset:324 ; 4-byte Folded Reload
+; GFX9-O0-NEXT:    buffer_load_dword v8, off, s[0:3], s32 offset:328 ; 4-byte Folded Reload
+; GFX9-O0-NEXT:    buffer_load_dword v9, off, s[0:3], s32 offset:332 ; 4-byte Folded Reload
+; GFX9-O0-NEXT:    buffer_load_dword v10, off, s[0:3], s32 offset:336 ; 4-byte Folded Reload
+; GFX9-O0-NEXT:    buffer_load_dword v11, off, s[0:3], s32 offset:340 ; 4-byte Folded Reload
 ; GFX9-O0-NEXT:    s_or_saveexec_b64 s[18:19], -1
 ; GFX9-O0-NEXT:    buffer_load_dword v16, off, s[0:3], s32 ; 4-byte Folded Reload
 ; GFX9-O0-NEXT:    s_mov_b64 exec, s[18:19]
@@ -1018,12 +1027,12 @@ define i128 @v_srem_i128_vv(i128 %lhs, i128 %rhs) {
 ; GFX9-O0-NEXT:    ; kill: def $vgpr12 killed $vgpr12 def $vgpr12_vgpr13 killed $exec
 ; GFX9-O0-NEXT:    v_mov_b32_e32 v13, v17
 ; GFX9-O0-NEXT:    s_mov_b64 s[8:9], s[6:7]
-; GFX9-O0-NEXT:    buffer_store_dword v14, off, s[0:3], s32 offset:300 ; 4-byte Folded Spill
+; GFX9-O0-NEXT:    buffer_store_dword v14, off, s[0:3], s32 offset:296 ; 4-byte Folded Spill
 ; GFX9-O0-NEXT:    s_waitcnt vmcnt(0)
-; GFX9-O0-NEXT:    buffer_store_dword v15, off, s[0:3], s32 offset:304 ; 4-byte Folded Spill
-; GFX9-O0-NEXT:    buffer_store_dword v12, off, s[0:3], s32 offset:308 ; 4-byte Folded Spill
+; GFX9-O0-NEXT:    buffer_store_dword v15, off, s[0:3], s32 offset:300 ; 4-byte Folded Spill
+; GFX9-O0-NEXT:    buffer_store_dword v12, off, s[0:3], s32 offset:304 ; 4-byte Folded Spill
 ; GFX9-O0-NEXT:    s_waitcnt vmcnt(0)
-; GFX9-O0-NEXT:    buffer_store_dword v13, off, s[0:3], s32 offset:312 ; 4-byte Folded Spill
+; GFX9-O0-NEXT:    buffer_store_dword v13, off, s[0:3], s32 offset:308 ; 4-byte Folded Spill
 ; GFX9-O0-NEXT:    s_mov_b64 s[4:5], s[6:7]
 ; GFX9-O0-NEXT:    v_mov_b32_e32 v15, s9
 ; GFX9-O0-NEXT:    v_mov_b32_e32 v14, s8
@@ -1034,30 +1043,30 @@ define i128 @v_srem_i128_vv(i128 %lhs, i128 %rhs) {
 ; GFX9-O0-NEXT:    s_or_saveexec_b64 s[18:19], -1
 ; GFX9-O0-NEXT:    buffer_store_dword v16, off, s[0:3], s32 ; 4-byte Folded Spill
 ; GFX9-O0-NEXT:    s_mov_b64 exec, s[18:19]
-; GFX9-O0-NEXT:    buffer_store_dword v14, off, s[0:3], s32 offset:292 ; 4-byte Folded Spill
+; GFX9-O0-NEXT:    buffer_store_dword v14, off, s[0:3], s32 offset:288 ; 4-byte Folded Spill
 ; GFX9-O0-NEXT:    s_waitcnt vmcnt(0)
-; GFX9-O0-NEXT:    buffer_store_dword v15, off, s[0:3], s32 offset:296 ; 4-byte Folded Spill
-; GFX9-O0-NEXT:    buffer_store_dword v12, off, s[0:3], s32 offset:284 ; 4-byte Folded Spill
+; GFX9-O0-NEXT:    buffer_store_dword v15, off, s[0:3], s32 offset:292 ; 4-byte Folded Spill
+; GFX9-O0-NEXT:    buffer_store_dword v12, off, s[0:3], s32 offset:280 ; 4-byte Folded Spill
 ; GFX9-O0-NEXT:    s_waitcnt vmcnt(0)
-; GFX9-O0-NEXT:    buffer_store_dword v13, off, s[0:3], s32 offset:288 ; 4-byte Folded Spill
-; GFX9-O0-NEXT:    buffer_store_dword v10, off, s[0:3], s32 offset:276 ; 4-byte Folded Spill
+; GFX9-O0-NEXT:    buffer_store_dword v13, off, s[0:3], s32 offset:284 ; 4-byte Folded Spill
+; GFX9-O0-NEXT:    buffer_store_dword v10, off, s[0:3], s32 offset:272 ; 4-byte Folded Spill
 ; GFX9-O0-NEXT:    s_waitcnt vmcnt(0)
-; GFX9-O0-NEXT:    buffer_store_dword v11, off, s[0:3], s32 offset:280 ; 4-byte Folded Spill
-; GFX9-O0-NEXT:    buffer_store_dword v8, off, s[0:3], s32 offset:268 ; 4-byte Folded Spill
+; GFX9-O0-NEXT:    buffer_store_dword v11, off, s[0:3], s32 offset:276 ; 4-byte Folded Spill
+; GFX9-O0-NEXT:    buffer_store_dword v8, off, s[0:3], s32 offset:264 ; 4-byte Folded Spill
 ; GFX9-O0-NEXT:    s_waitcnt vmcnt(0)
-; GFX9-O0-NEXT:    buffer_store_dword v9, off, s[0:3], s32 offset:272 ; 4-byte Folded Spill
-; GFX9-O0-NEXT:    buffer_store_dword v6, off, s[0:3], s32 offset:260 ; 4-byte Folded Spill
+; GFX9-O0-NEXT:    buffer_store_dword v9, off, s[0:3], s32 offset:268 ; 4-byte Folded Spill
+; GFX9-O0-NEXT:    buffer_store_dword v6, off, s[0:3], s32 offset:256 ; 4-byte Folded Spill
 ; GFX9-O0-NEXT:    s_waitcnt vmcnt(0)
-; GFX9-O0-NEXT:    buffer_store_dword v7, off, s[0:3], s32 offset:264 ; 4-byte Folded Spill
-; GFX9-O0-NEXT:    buffer_store_dword v4, off, s[0:3], s32 offset:252 ; 4-byte Folded Spill
+; GFX9-O0-NEXT:    buffer_store_dword v7, off, s[0:3], s32 offset:260 ; 4-byte Folded Spill
+; GFX9-O0-NEXT:    buffer_store_dword v4, off, s[0:3], s32 offset:248 ; 4-byte Folded Spill
 ; GFX9-O0-NEXT:    s_waitcnt vmcnt(0)
-; GFX9-O0-NEXT:    buffer_store_dword v5, off, s[0:3], s32 offset:256 ; 4-byte Folded Spill
-; GFX9-O0-NEXT:    buffer_store_dword v2, off, s[0:3], s32 offset:244 ; 4-byte Folded Spill
+; GFX9-O0-NEXT:    buffer_store_dword v5, off, s[0:3], s32 offset:252 ; 4-byte Folded Spill
+; GFX9-O0-NEXT:    buffer_store_dword v2, off, s[0:3], s32 offset:240 ; 4-byte Folded Spill
 ; GFX9-O0-NEXT:    s_waitcnt vmcnt(0)
-; GFX9-O0-NEXT:    buffer_store_dword v3, off, s[0:3], s32 offset:248 ; 4-byte Folded Spill
-; GFX9-O0-NEXT:    buffer_store_dword v0, off, s[0:3], s32 offset:236 ; 4-byte Folded Spill
+; GFX9-O0-NEXT:    buffer_store_dword v3, off, s[0:3], s32 offset:244 ; 4-byte Folded Spill
+; GFX9-O0-NEXT:    buffer_store_dword v0, off, s[0:3], s32 offset:232 ; 4-byte Folded Spill
 ; GFX9-O0-NEXT:    s_waitcnt vmcnt(0)
-; GFX9-O0-NEXT:    buffer_store_dword v1, off, s[0:3], s32 offset:240 ; 4-byte Folded Spill
+; GFX9-O0-NEXT:    buffer_store_dword v1, off, s[0:3], s32 offset:236 ; 4-byte Folded Spill
 ; GFX9-O0-NEXT:    s_branch .LBB0_6
 ; GFX9-O0-NEXT:  .LBB0_8: ; %udiv-bb1
 ; GFX9-O0-NEXT:    s_or_saveexec_b64 s[18:19], -1
@@ -1099,14 +1108,14 @@ define i128 @v_srem_i128_vv(i128 %lhs, i128 %rhs) {
 ; GFX9-O0-NEXT:    v_mov_b32_e32 v2, v3
 ; GFX9-O0-NEXT:    v_mov_b32_e32 v6, v2
 ; GFX9-O0-NEXT:    v_mov_b32_e32 v5, v1
-; GFX9-O0-NEXT:    buffer_store_dword v5, off, s[0:3], s32 offset:332 ; 4-byte Folded Spill
+; GFX9-O0-NEXT:    buffer_store_dword v5, off, s[0:3], s32 offset:328 ; 4-byte Folded Spill
 ; GFX9-O0-NEXT:    s_waitcnt vmcnt(0)
-; GFX9-O0-NEXT:    buffer_store_dword v6, off, s[0:3], s32 offset:336 ; 4-byte Folded Spill
+; GFX9-O0-NEXT:    buffer_store_dword v6, off, s[0:3], s32 offset:332 ; 4-byte Folded Spill
 ; GFX9-O0-NEXT:    v_mov_b32_e32 v5, v9
 ; GFX9-O0-NEXT:    v_mov_b32_e32 v6, v10
-; GFX9-O0-NEXT:    buffer_store_dword v5, off, s[0:3], s32 offset:340 ; 4-byte Folded Spill
+; GFX9-O0-NEXT:    buffer_store_dword v5, off, s[0:3], s32 offset:336 ; 4-byte Folded Spill
 ; GFX9-O0-NEXT:    s_waitcnt vmcnt(0)
-; GFX9-O0-NEXT:    buffer_store_dword v6, off, s[0:3], s32 offset:344 ; 4-byte Folded Spill
+; GFX9-O0-NEXT:    buffer_store_dword v6, off, s[0:3], s32 offset:340 ; 4-byte Folded Spill
 ; GFX9-O0-NEXT:    s_mov_b32 s4, 0x7f
 ; GFX9-O0-NEXT:    v_sub_u32_e64 v3, s4, v4
 ; GFX9-O0-NEXT:    v_lshlrev_b64 v[5:6], v3, v[11:12]
@@ -1152,12 +1161,12 @@ define i128 @v_srem_i128_vv(i128 %lhs, i128 %rhs) {
 ; GFX9-O0-NEXT:    ; implicit-def: $sgpr4
 ; GFX9-O0-NEXT:    ; kill: def $vgpr7 killed $vgpr7 def $vgpr7_vgpr8 killed $exec
 ; GFX9-O0-NEXT:    v_mov_b32_e32 v8, v3
-; GFX9-O0-NEXT:    buffer_store_dword v7, off, s[0:3], s32 offset:324 ; 4-byte Folded Spill
+; GFX9-O0-NEXT:    buffer_store_dword v7, off, s[0:3], s32 offset:320 ; 4-byte Folded Spill
 ; GFX9-O0-NEXT:    s_waitcnt vmcnt(0)
-; GFX9-O0-NEXT:    buffer_store_dword v8, off, s[0:3], s32 offset:328 ; 4-byte Folded Spill
-; GFX9-O0-NEXT:    buffer_store_dword v5, off, s[0:3], s32 offset:316 ; 4-byte Folded Spill
+; GFX9-O0-NEXT:    buffer_store_dword v8, off, s[0:3], s32 offset:324 ; 4-byte Folded Spill
+; GFX9-O0-NEXT:    buffer_store_dword v5, off, s[0:3], s32 offset:312 ; 4-byte Folded Spill
 ; GFX9-O0-NEXT:    s_waitcnt vmcnt(0)
-; GFX9-O0-NEXT:    buffer_store_dword v6, off, s[0:3], s32 offset:320 ; 4-byte Folded Spill
+; GFX9-O0-NEXT:    buffer_store_dword v6, off, s[0:3], s32 offset:316 ; 4-byte Folded Spill
 ; GFX9-O0-NEXT:    v_mov_b32_e32 v4, v2
 ; GFX9-O0-NEXT:    v_mov_b32_e32 v3, v10
 ; GFX9-O0-NEXT:    v_or_b32_e64 v3, v3, v4
@@ -1172,18 +1181,18 @@ define i128 @v_srem_i128_vv(i128 %lhs, i128 %rhs) {
 ; GFX9-O0-NEXT:    v_mov_b32_e32 v4, s9
 ; GFX9-O0-NEXT:    v_mov_b32_e32 v1, s6
 ; GFX9-O0-NEXT:    v_mov_b32_e32 v2, s7
-; GFX9-O0-NEXT:    buffer_store_dword v7, off, s[0:3], s32 offset:148 ; 4-byte Folded Spill
+; GFX9-O0-NEXT:    buffer_store_dword v7, off, s[0:3], s32 offset:144 ; 4-byte Folded Spill
 ; GFX9-O0-NEXT:    s_waitcnt vmcnt(0)
-; GFX9-O0-NEXT:    buffer_store_dword v8, off, s[0:3], s32 offset:152 ; 4-byte Folded Spill
-; GFX9-O0-NEXT:    buffer_store_dword v5, off, s[0:3], s32 offset:140 ; 4-byte Folded Spill
+; GFX9-O0-NEXT:    buffer_store_dword v8, off, s[0:3], s32 offset:148 ; 4-byte Folded Spill
+; GFX9-O0-NEXT:    buffer_store_dword v5, off, s[0:3], s32 offset:136 ; 4-byte Folded Spill
 ; GFX9-O0-NEXT:    s_waitcnt vmcnt(0)
-; GFX9-O0-NEXT:    buffer_store_dword v6, off, s[0:3], s32 offset:144 ; 4-byte Folded Spill
-; GFX9-O0-NEXT:    buffer_store_dword v3, off, s[0:3], s32 offset:132 ; 4-byte Folded Spill
+; GFX9-O0-NEXT:    buffer_store_dword v6, off, s[0:3], s32 offset:140 ; 4-byte Folded Spill
+; GFX9-O0-NEXT:    buffer_store_dword v3, off, s[0:3], s32 offset:128 ; 4-byte Folded Spill
 ; GFX9-O0-NEXT:    s_waitcnt vmcnt(0)
-; GFX9-O0-NEXT:    buffer_store_dword v4, off, s[0:3], s32 offset:136 ; 4-byte Folded Spill
-; GFX9-O0-NEXT:    buffer_store_dword v1, off, s[0:3], s32 offset:124 ; 4-byte Folded Spill
+; GFX9-O0-NEXT:    buffer_store_dword v4, off, s[0:3], s32 offset:132 ; 4-byte Folded Spill
+; GFX9-O0-NEXT:    buffer_store_dword v1, off, s[0:3], s32 offset:120 ; 4-byte Folded Spill
 ; GFX9-O0-NEXT:    s_waitcnt vmcnt(0)
-; GFX9-O0-NEXT:    buffer_store_dword v2, off, s[0:3], s32 offset:128 ; 4-byte Folded Spill
+; GFX9-O0-NEXT:    buffer_store_dword v2, off, s[0:3], s32 offset:124 ; 4-byte Folded Spill
 ; GFX9-O0-NEXT:    s_mov_b64 s[6:7], exec
 ; GFX9-O0-NEXT:    s_and_b64 s[4:5], s[6:7], s[4:5]
 ; GFX9-O0-NEXT:    s_xor_b64 s[6:7], s[4:5], s[6:7]
@@ -1203,18 +1212,18 @@ define i128 @v_srem_i128_vv(i128 %lhs, i128 %rhs) {
 ; GFX9-O0-NEXT:    buffer_load_dword v8, off, s[0:3], s32 offset:112 ; 4-byte Folded Reload
 ; GFX9-O0-NEXT:    buffer_load_dword v9, off, s[0:3], s32 offset:100 ; 4-byte Folded Reload
 ; GFX9-O0-NEXT:    buffer_load_dword v10, off, s[0:3], s32 offset:104 ; 4-byte Folded Reload
-; GFX9-O0-NEXT:    buffer_load_dword v13, off, s[0:3], s32 offset:84 ; 4-byte Folded Reload
-; GFX9-O0-NEXT:    buffer_load_dword v14, off, s[0:3], s32 offset:88 ; 4-byte Folded Reload
-; GFX9-O0-NEXT:    buffer_load_dword v15, off, s[0:3], s32 offset:92 ; 4-byte Folded Reload
-; GFX9-O0-NEXT:    buffer_load_dword v16, off, s[0:3], s32 offset:96 ; 4-byte Folded Reload
-; GFX9-O0-NEXT:    buffer_load_dword v19, off, s[0:3], s32 offset:76 ; 4-byte Folded Reload
-; GFX9-O0-NEXT:    buffer_load_dword v20, off, s[0:3], s32 offset:80 ; 4-byte Folded Reload
-; GFX9-O0-NEXT:    buffer_load_dword v11, off, s[0:3], s32 offset:196 ; 4-byte Folded Reload
-; GFX9-O0-NEXT:    buffer_load_dword v12, off, s[0:3], s32 offset:200 ; 4-byte Folded Reload
-; GFX9-O0-NEXT:    buffer_load_dword v5, off, s[0:3], s32 offset:188 ; 4-byte Folded Reload
-; GFX9-O0-NEXT:    buffer_load_dword v6, off, s[0:3], s32 offset:192 ; 4-byte Folded Reload
-; GFX9-O0-NEXT:    buffer_load_dword v17, off, s[0:3], s32 offset:68 ; 4-byte Folded Reload
-; GFX9-O0-NEXT:    buffer_load_dword v18, off, s[0:3], s32 offset:72 ; 4-byte Folded Reload
+; GFX9-O0-NEXT:    buffer_load_dword v13, off, s[0:3], s32 offset:92 ; 4-byte Folded Reload
+; GFX9-O0-NEXT:    buffer_load_dword v14, off, s[0:3], s32 offset:96 ; 4-byte Folded Reload
+; GFX9-O0-NEXT:    buffer_load_dword v15, off, s[0:3], s32 offset:84 ; 4-byte Folded Reload
+; GFX9-O0-NEXT:    buffer_load_dword v16, off, s[0:3], s32 offset:88 ; 4-byte Folded Reload
+; GFX9-O0-NEXT:    buffer_load_dword v19, off, s[0:3], s32 offset:68 ; 4-byte Folded Reload
+; GFX9-O0-NEXT:    buffer_load_dword v20, off, s[0:3], s32 offset:72 ; 4-byte Folded Reload
+; GFX9-O0-NEXT:    buffer_load_dword v11, off, s[0:3], s32 offset:192 ; 4-byte Folded Reload
+; GFX9-O0-NEXT:    buffer_load_dword v12, off, s[0:3], s32 offset:196 ; 4-byte Folded Reload
+; GFX9-O0-NEXT:    buffer_load_dword v5, off, s[0:3], s32 offset:184 ; 4-byte Folded Reload
+; GFX9-O0-NEXT:    buffer_load_dword v6, off, s[0:3], s32 offset:188 ; 4-byte Folded Reload
+; GFX9-O0-NEXT:    buffer_load_dword v17, off, s[0:3], s32 offset:76 ; 4-byte Folded Reload
+; GFX9-O0-NEXT:    buffer_load_dword v18, off, s[0:3], s32 offset:80 ; 4-byte Folded Reload
 ; GFX9-O0-NEXT:    s_mov_b32 s4, 32
 ; GFX9-O0-NEXT:    s_waitcnt vmcnt(2)
 ; GFX9-O0-NEXT:    v_lshrrev_b64 v[0:1], s4, v[5:6]
@@ -1486,11 +1495,11 @@ define i128 @v_srem_i128_vv(i128 %lhs, i128 %rhs) {
 ; GFX9-O0-NEXT:    v_mov_b32_e32 v3, v5
 ; GFX9-O0-NEXT:    ; kill: killed $vgpr4
 ; GFX9-O0-NEXT:    s_xor_saveexec_b64 s[4:5], -1
-; GFX9-O0-NEXT:    buffer_load_dword v0, off, s[0:3], s32 offset:348 ; 4-byte Folded Reload
+; GFX9-O0-NEXT:    buffer_load_dword v0, off, s[0:3], s32 offset:344 ; 4-byte Folded Reload
 ; GFX9-O0-NEXT:    s_nop 0
-; GFX9-O0-NEXT:    buffer_load_dword v4, off, s[0:3], s32 offset:352 ; 4-byte Folded Reload
-; GFX9-O0-NEXT:    buffer_load_dword v8, off, s[0:3], s32 offset:356 ; 4-byte Folded Reload
-; GFX9-O0-NEXT:    buffer_load_dword v16, off, s[0:3], s32 offset:360 ; 4-byte Folded Reload
+; GFX9-O0-NEXT:    buffer_load_dword v4, off, s[0:3], s32 offset:348 ; 4-byte Folded Reload
+; GFX9-O0-NEXT:    buffer_load_dword v8, off, s[0:3], s32 offset:352 ; 4-byte Folded Reload
+; GFX9-O0-NEXT:    buffer_load_dword v16, off, s[0:3], s32 offset:356 ; 4-byte Folded Reload
 ; GFX9-O0-NEXT:    s_mov_b64 exec, s[4:5]
 ; GFX9-O0-NEXT:    s_waitcnt vmcnt(0)
 ; GFX9-O0-NEXT:    s_setpc_b64 s[30:31]
diff --git a/llvm/test/CodeGen/AMDGPU/trap-abis.ll b/llvm/test/CodeGen/AMDGPU/trap-abis.ll
index 3cd6c98ef4b8..dcc5fbd142c4 100644
--- a/llvm/test/CodeGen/AMDGPU/trap-abis.ll
+++ b/llvm/test/CodeGen/AMDGPU/trap-abis.ll
@@ -3,6 +3,8 @@
 ; RUN: llc %s -o - -mtriple=amdgcn-amd-amdhsa -mcpu=gfx803 -verify-machineinstrs | FileCheck --check-prefix=HSA-TRAP-GFX803 %s
 ; RUN: llc %s -o - -mtriple=amdgcn-amd-amdhsa -mcpu=gfx900 -verify-machineinstrs | FileCheck --check-prefix=HSA-TRAP-GFX900 %s
 ; RUN: llc %s -o - -mtriple=amdgcn-amd-amdhsa -mcpu=gfx900 -mattr=-trap-handler -verify-machineinstrs | FileCheck --check-prefix=HSA-NOTRAP-GFX900 %s
+; RUN: llc %s -o - -mtriple=amdgcn-amd-amdhsa -mcpu=gfx1100 -verify-machineinstrs | FileCheck --check-prefix=HSA-TRAP-GFX1100 %s
+; RUN: llc %s -o - -O0 -mtriple=amdgcn-amd-amdhsa -mcpu=gfx1100 -verify-machineinstrs | FileCheck --check-prefix=HSA-TRAP-GFX1100-O0 %s
 
 declare void @llvm.trap() #0
 declare void @llvm.debugtrap() #1
@@ -49,6 +51,48 @@ define amdgpu_kernel void @trap(ptr addrspace(1) nocapture readonly %arg0) {
 ; HSA-NOTRAP-GFX900-NEXT:    global_store_dword v0, v1, s[0:1]
 ; HSA-NOTRAP-GFX900-NEXT:    s_waitcnt vmcnt(0)
 ; HSA-NOTRAP-GFX900-NEXT:    s_endpgm
+;
+; HSA-TRAP-GFX1100-LABEL: trap:
+; HSA-TRAP-GFX1100:       ; %bb.0:
+; HSA-TRAP-GFX1100-NEXT:    s_load_b64 s[0:1], s[0:1], 0x0
+; HSA-TRAP-GFX1100-NEXT:    v_dual_mov_b32 v0, 0 :: v_dual_mov_b32 v1, 1
+; HSA-TRAP-GFX1100-NEXT:    s_mov_b32 ttmp2, m0
+; HSA-TRAP-GFX1100-NEXT:    s_waitcnt lgkmcnt(0)
+; HSA-TRAP-GFX1100-NEXT:    global_store_b32 v0, v1, s[0:1] dlc
+; HSA-TRAP-GFX1100-NEXT:    s_waitcnt_vscnt null, 0x0
+; HSA-TRAP-GFX1100-NEXT:    s_trap 2
+; HSA-TRAP-GFX1100-NEXT:    s_sendmsg_rtn_b32 s0, sendmsg(MSG_RTN_GET_DOORBELL)
+; HSA-TRAP-GFX1100-NEXT:    s_waitcnt lgkmcnt(0)
+; HSA-TRAP-GFX1100-NEXT:    s_and_b32 s0, s0, 0x3ff
+; HSA-TRAP-GFX1100-NEXT:    s_delay_alu instid0(SALU_CYCLE_1) | instskip(NEXT) | instid1(SALU_CYCLE_1)
+; HSA-TRAP-GFX1100-NEXT:    s_bitset1_b32 s0, 10
+; HSA-TRAP-GFX1100-NEXT:    s_mov_b32 m0, s0
+; HSA-TRAP-GFX1100-NEXT:    s_sendmsg sendmsg(MSG_INTERRUPT)
+; HSA-TRAP-GFX1100-NEXT:    s_mov_b32 m0, ttmp2
+; HSA-TRAP-GFX1100-NEXT:  .LBB0_1: ; =>This Inner Loop Header: Depth=1
+; HSA-TRAP-GFX1100-NEXT:    s_sethalt 5
+; HSA-TRAP-GFX1100-NEXT:    s_branch .LBB0_1
+;
+; HSA-TRAP-GFX1100-O0-LABEL: trap:
+; HSA-TRAP-GFX1100-O0:       ; %bb.0:
+; HSA-TRAP-GFX1100-O0-NEXT:    s_load_b64 s[0:1], s[4:5], 0x0
+; HSA-TRAP-GFX1100-O0-NEXT:    v_mov_b32_e32 v0, 0
+; HSA-TRAP-GFX1100-O0-NEXT:    v_mov_b32_e32 v1, 1
+; HSA-TRAP-GFX1100-O0-NEXT:    s_waitcnt lgkmcnt(0)
+; HSA-TRAP-GFX1100-O0-NEXT:    global_store_b32 v0, v1, s[0:1] dlc
+; HSA-TRAP-GFX1100-O0-NEXT:    s_waitcnt_vscnt null, 0x0
+; HSA-TRAP-GFX1100-O0-NEXT:    s_trap 2
+; HSA-TRAP-GFX1100-O0-NEXT:    s_sendmsg_rtn_b32 s0, sendmsg(MSG_RTN_GET_DOORBELL)
+; HSA-TRAP-GFX1100-O0-NEXT:    s_mov_b32 ttmp2, m0
+; HSA-TRAP-GFX1100-O0-NEXT:    s_waitcnt lgkmcnt(0)
+; HSA-TRAP-GFX1100-O0-NEXT:    s_and_b32 s0, s0, 0x3ff
+; HSA-TRAP-GFX1100-O0-NEXT:    s_or_b32 s0, s0, 0x400
+; HSA-TRAP-GFX1100-O0-NEXT:    s_mov_b32 m0, s0
+; HSA-TRAP-GFX1100-O0-NEXT:    s_sendmsg sendmsg(MSG_INTERRUPT)
+; HSA-TRAP-GFX1100-O0-NEXT:    s_mov_b32 m0, ttmp2
+; HSA-TRAP-GFX1100-O0-NEXT:  .LBB0_1: ; =>This Inner Loop Header: Depth=1
+; HSA-TRAP-GFX1100-O0-NEXT:    s_sethalt 5
+; HSA-TRAP-GFX1100-O0-NEXT:    s_branch .LBB0_1
   store volatile i32 1, ptr addrspace(1) %arg0
   call void @llvm.trap()
   unreachable
@@ -128,6 +172,84 @@ define amdgpu_kernel void @non_entry_trap(ptr addrspace(1) nocapture readonly %a
 ; HSA-NOTRAP-GFX900-NEXT:    s_endpgm
 ; HSA-NOTRAP-GFX900-NEXT:  .LBB1_2: ; %trap
 ; HSA-NOTRAP-GFX900-NEXT:    s_endpgm
+;
+; HSA-TRAP-GFX1100-LABEL: non_entry_trap:
+; HSA-TRAP-GFX1100:       ; %bb.0: ; %entry
+; HSA-TRAP-GFX1100-NEXT:    s_load_b64 s[0:1], s[0:1], 0x0
+; HSA-TRAP-GFX1100-NEXT:    v_mov_b32_e32 v0, 0
+; HSA-TRAP-GFX1100-NEXT:    s_waitcnt lgkmcnt(0)
+; HSA-TRAP-GFX1100-NEXT:    global_load_b32 v1, v0, s[0:1] glc dlc
+; HSA-TRAP-GFX1100-NEXT:    s_waitcnt vmcnt(0)
+; HSA-TRAP-GFX1100-NEXT:    v_cmp_eq_u32_e32 vcc_lo, -1, v1
+; HSA-TRAP-GFX1100-NEXT:    s_cbranch_vccz .LBB1_2
+; HSA-TRAP-GFX1100-NEXT:  ; %bb.1: ; %ret
+; HSA-TRAP-GFX1100-NEXT:    v_mov_b32_e32 v1, 3
+; HSA-TRAP-GFX1100-NEXT:    global_store_b32 v0, v1, s[0:1] dlc
+; HSA-TRAP-GFX1100-NEXT:    s_waitcnt_vscnt null, 0x0
+; HSA-TRAP-GFX1100-NEXT:    s_nop 0
+; HSA-TRAP-GFX1100-NEXT:    s_sendmsg sendmsg(MSG_DEALLOC_VGPRS)
+; HSA-TRAP-GFX1100-NEXT:    s_endpgm
+; HSA-TRAP-GFX1100-NEXT:  .LBB1_2: ; %trap
+; HSA-TRAP-GFX1100-NEXT:    s_trap 2
+; HSA-TRAP-GFX1100-NEXT:    s_sendmsg_rtn_b32 s0, sendmsg(MSG_RTN_GET_DOORBELL)
+; HSA-TRAP-GFX1100-NEXT:    s_mov_b32 ttmp2, m0
+; HSA-TRAP-GFX1100-NEXT:    s_waitcnt lgkmcnt(0)
+; HSA-TRAP-GFX1100-NEXT:    s_and_b32 s0, s0, 0x3ff
+; HSA-TRAP-GFX1100-NEXT:    s_delay_alu instid0(SALU_CYCLE_1) | instskip(NEXT) | instid1(SALU_CYCLE_1)
+; HSA-TRAP-GFX1100-NEXT:    s_bitset1_b32 s0, 10
+; HSA-TRAP-GFX1100-NEXT:    s_mov_b32 m0, s0
+; HSA-TRAP-GFX1100-NEXT:    s_sendmsg sendmsg(MSG_INTERRUPT)
+; HSA-TRAP-GFX1100-NEXT:    s_mov_b32 m0, ttmp2
+; HSA-TRAP-GFX1100-NEXT:  .LBB1_3: ; =>This Inner Loop Header: Depth=1
+; HSA-TRAP-GFX1100-NEXT:    s_sethalt 5
+; HSA-TRAP-GFX1100-NEXT:    s_branch .LBB1_3
+;
+; HSA-TRAP-GFX1100-O0-LABEL: non_entry_trap:
+; HSA-TRAP-GFX1100-O0:       ; %bb.0: ; %entry
+; HSA-TRAP-GFX1100-O0-NEXT:    ; implicit-def: $vgpr0 : SGPR spill to VGPR lane
+; HSA-TRAP-GFX1100-O0-NEXT:    s_load_b64 s[0:1], s[4:5], 0x0
+; HSA-TRAP-GFX1100-O0-NEXT:    s_waitcnt lgkmcnt(0)
+; HSA-TRAP-GFX1100-O0-NEXT:    s_mov_b64 s[2:3], s[0:1]
+; HSA-TRAP-GFX1100-O0-NEXT:    v_writelane_b32 v0, s2, 0
+; HSA-TRAP-GFX1100-O0-NEXT:    v_writelane_b32 v0, s3, 1
+; HSA-TRAP-GFX1100-O0-NEXT:    s_or_saveexec_b32 s6, -1
+; HSA-TRAP-GFX1100-O0-NEXT:    scratch_store_b32 off, v0, off ; 4-byte Folded Spill
+; HSA-TRAP-GFX1100-O0-NEXT:    s_mov_b32 exec_lo, s6
+; HSA-TRAP-GFX1100-O0-NEXT:    v_mov_b32_e32 v0, 0
+; HSA-TRAP-GFX1100-O0-NEXT:    global_load_b32 v0, v0, s[0:1] glc dlc
+; HSA-TRAP-GFX1100-O0-NEXT:    s_waitcnt vmcnt(0)
+; HSA-TRAP-GFX1100-O0-NEXT:    s_mov_b32 s0, -1
+; HSA-TRAP-GFX1100-O0-NEXT:    ; implicit-def: $sgpr1
+; HSA-TRAP-GFX1100-O0-NEXT:    v_cmp_eq_u32_e64 s0, v0, s0
+; HSA-TRAP-GFX1100-O0-NEXT:    s_and_b32 vcc_lo, exec_lo, s0
+; HSA-TRAP-GFX1100-O0-NEXT:    s_cbranch_vccnz .LBB1_2
+; HSA-TRAP-GFX1100-O0-NEXT:  ; %bb.1: ; %trap
+; HSA-TRAP-GFX1100-O0-NEXT:    s_trap 2
+; HSA-TRAP-GFX1100-O0-NEXT:    s_sendmsg_rtn_b32 s0, sendmsg(MSG_RTN_GET_DOORBELL)
+; HSA-TRAP-GFX1100-O0-NEXT:    s_mov_b32 ttmp2, m0
+; HSA-TRAP-GFX1100-O0-NEXT:    s_waitcnt lgkmcnt(0)
+; HSA-TRAP-GFX1100-O0-NEXT:    s_and_b32 s0, s0, 0x3ff
+; HSA-TRAP-GFX1100-O0-NEXT:    s_or_b32 s0, s0, 0x400
+; HSA-TRAP-GFX1100-O0-NEXT:    s_mov_b32 m0, s0
+; HSA-TRAP-GFX1100-O0-NEXT:    s_sendmsg sendmsg(MSG_INTERRUPT)
+; HSA-TRAP-GFX1100-O0-NEXT:    s_mov_b32 m0, ttmp2
+; HSA-TRAP-GFX1100-O0-NEXT:    s_branch .LBB1_3
+; HSA-TRAP-GFX1100-O0-NEXT:  .LBB1_2: ; %ret
+; HSA-TRAP-GFX1100-O0-NEXT:    s_or_saveexec_b32 s6, -1
+; HSA-TRAP-GFX1100-O0-NEXT:    scratch_load_b32 v0, off, off ; 4-byte Folded Reload
+; HSA-TRAP-GFX1100-O0-NEXT:    s_mov_b32 exec_lo, s6
+; HSA-TRAP-GFX1100-O0-NEXT:    s_waitcnt vmcnt(0)
+; HSA-TRAP-GFX1100-O0-NEXT:    v_readlane_b32 s0, v0, 0
+; HSA-TRAP-GFX1100-O0-NEXT:    v_readlane_b32 s1, v0, 1
+; HSA-TRAP-GFX1100-O0-NEXT:    v_mov_b32_e32 v1, 0
+; HSA-TRAP-GFX1100-O0-NEXT:    v_mov_b32_e32 v2, 3
+; HSA-TRAP-GFX1100-O0-NEXT:    global_store_b32 v1, v2, s[0:1] dlc
+; HSA-TRAP-GFX1100-O0-NEXT:    s_waitcnt_vscnt null, 0x0
+; HSA-TRAP-GFX1100-O0-NEXT:    ; kill: killed $vgpr0
+; HSA-TRAP-GFX1100-O0-NEXT:    s_endpgm
+; HSA-TRAP-GFX1100-O0-NEXT:  .LBB1_3: ; =>This Inner Loop Header: Depth=1
+; HSA-TRAP-GFX1100-O0-NEXT:    s_sethalt 5
+; HSA-TRAP-GFX1100-O0-NEXT:    s_branch .LBB1_3
 entry:
   %tmp29 = load volatile i32, ptr addrspace(1) %arg0
   %cmp = icmp eq i32 %tmp29, -1
@@ -197,6 +319,21 @@ define amdgpu_kernel void @debugtrap(ptr addrspace(1) nocapture readonly %arg0)
 ; HSA-NOTRAP-GFX900-NEXT:    global_store_dword v0, v2, s[0:1]
 ; HSA-NOTRAP-GFX900-NEXT:    s_waitcnt vmcnt(0)
 ; HSA-NOTRAP-GFX900-NEXT:    s_endpgm
+;
+; HSA-TRAP-GFX1100-LABEL: debugtrap:
+; HSA-TRAP-GFX1100:       ; %bb.0:
+; HSA-TRAP-GFX1100-NEXT:    s_load_b64 s[0:1], s[0:1], 0x0
+; HSA-TRAP-GFX1100-NEXT:    v_dual_mov_b32 v0, 0 :: v_dual_mov_b32 v1, 1
+; HSA-TRAP-GFX1100-NEXT:    v_mov_b32_e32 v2, 2
+; HSA-TRAP-GFX1100-NEXT:    s_waitcnt lgkmcnt(0)
+; HSA-TRAP-GFX1100-NEXT:    global_store_b32 v0, v1, s[0:1] dlc
+; HSA-TRAP-GFX1100-NEXT:    s_waitcnt_vscnt null, 0x0
+; HSA-TRAP-GFX1100-NEXT:    s_trap 3
+; HSA-TRAP-GFX1100-NEXT:    global_store_b32 v0, v2, s[0:1] dlc
+; HSA-TRAP-GFX1100-NEXT:    s_waitcnt_vscnt null, 0x0
+; HSA-TRAP-GFX1100-NEXT:    s_nop 0
+; HSA-TRAP-GFX1100-NEXT:    s_sendmsg sendmsg(MSG_DEALLOC_VGPRS)
+; HSA-TRAP-GFX1100-NEXT:    s_endpgm
   store volatile i32 1, ptr addrspace(1) %arg0
   call void @llvm.debugtrap()
   store volatile i32 2, ptr addrspace(1) %arg0
diff --git a/llvm/test/CodeGen/AMDGPU/wave_dispatch_regs.ll b/llvm/test/CodeGen/AMDGPU/wave_dispatch_regs.ll
index e73235857728..29520cb7468c 100644
--- a/llvm/test/CodeGen/AMDGPU/wave_dispatch_regs.ll
+++ b/llvm/test/CodeGen/AMDGPU/wave_dispatch_regs.ll
@@ -21,10 +21,10 @@
 ; VI-NEXT:          .vgpr_count:     0x5
 ; GFX9-NEXT:        .vgpr_count:     0x5
 ; GCN-NEXT:     .registers:
-; SI-NEXT:        0x2e12 (COMPUTE_PGM_RSRC1): 0x{{[0-9a-f]*}}81
-; VI-NEXT:        0x2e12 (COMPUTE_PGM_RSRC1): 0x{{[0-9a-f]*}}c1
-; GFX9-NEXT:      0x2e12 (COMPUTE_PGM_RSRC1): 0x{{[0-9a-f]*}}81
-; GCN-NEXT:       0x2e13 (COMPUTE_PGM_RSRC2): 0
+; SI-NEXT:        '0x2e12 (COMPUTE_PGM_RSRC1)': 0x{{[0-9a-f]*}}81
+; VI-NEXT:        '0x2e12 (COMPUTE_PGM_RSRC1)': 0x{{[0-9a-f]*}}c1
+; GFX9-NEXT:      '0x2e12 (COMPUTE_PGM_RSRC1)': 0x{{[0-9a-f]*}}81
+; GCN-NEXT:       '0x2e13 (COMPUTE_PGM_RSRC2)': 0
 ; GCN-NEXT: ...
 ; GCN-NEXT:         .end_amdgpu_pal_metadata
 
diff --git a/llvm/test/CodeGen/ARM/minnum-maxnum-intrinsics.ll b/llvm/test/CodeGen/ARM/minnum-maxnum-intrinsics.ll
index be741f536ac7..528bfe041173 100644
--- a/llvm/test/CodeGen/ARM/minnum-maxnum-intrinsics.ll
+++ b/llvm/test/CodeGen/ARM/minnum-maxnum-intrinsics.ll
@@ -46,12 +46,10 @@ define float @fminnum32_intrinsic(float %x, float %y) {
 define float @fminnum32_nsz_intrinsic(float %x, float %y) {
 ; ARMV7-LABEL: fminnum32_nsz_intrinsic:
 ; ARMV7:       @ %bb.0:
-; ARMV7-NEXT:    vmov s0, r0
-; ARMV7-NEXT:    vmov s2, r1
-; ARMV7-NEXT:    vcmp.f32 s0, s2
-; ARMV7-NEXT:    vmrs APSR_nzcv, fpscr
-; ARMV7-NEXT:    vmovlt.f32 s2, s0
-; ARMV7-NEXT:    vmov r0, s2
+; ARMV7-NEXT:    vmov s0, r1
+; ARMV7-NEXT:    vmov s2, r0
+; ARMV7-NEXT:    vmin.f32 d0, d1, d0
+; ARMV7-NEXT:    vmov r0, s0
 ; ARMV7-NEXT:    bx lr
 ;
 ; ARMV8-LABEL: fminnum32_nsz_intrinsic:
@@ -78,9 +76,7 @@ define float @fminnum32_non_zero_intrinsic(float %x) {
 ; ARMV7:       @ %bb.0:
 ; ARMV7-NEXT:    vmov.f32 s0, #-1.000000e+00
 ; ARMV7-NEXT:    vmov s2, r0
-; ARMV7-NEXT:    vcmp.f32 s2, s0
-; ARMV7-NEXT:    vmrs APSR_nzcv, fpscr
-; ARMV7-NEXT:    vmovlt.f32 s0, s2
+; ARMV7-NEXT:    vmin.f32 d0, d1, d0
 ; ARMV7-NEXT:    vmov r0, s0
 ; ARMV7-NEXT:    bx lr
 ;
@@ -136,12 +132,10 @@ define float @fmaxnum32_intrinsic(float %x, float %y) {
 define float @fmaxnum32_nsz_intrinsic(float %x, float %y) {
 ; ARMV7-LABEL: fmaxnum32_nsz_intrinsic:
 ; ARMV7:       @ %bb.0:
-; ARMV7-NEXT:    vmov s0, r0
-; ARMV7-NEXT:    vmov s2, r1
-; ARMV7-NEXT:    vcmp.f32 s0, s2
-; ARMV7-NEXT:    vmrs APSR_nzcv, fpscr
-; ARMV7-NEXT:    vmovgt.f32 s2, s0
-; ARMV7-NEXT:    vmov r0, s2
+; ARMV7-NEXT:    vmov s0, r1
+; ARMV7-NEXT:    vmov s2, r0
+; ARMV7-NEXT:    vmax.f32 d0, d1, d0
+; ARMV7-NEXT:    vmov r0, s0
 ; ARMV7-NEXT:    bx lr
 ;
 ; ARMV8-LABEL: fmaxnum32_nsz_intrinsic:
@@ -210,9 +204,7 @@ define float @fmaxnum32_non_zero_intrinsic(float %x) {
 ; ARMV7:       @ %bb.0:
 ; ARMV7-NEXT:    vmov.f32 s0, #1.000000e+00
 ; ARMV7-NEXT:    vmov s2, r0
-; ARMV7-NEXT:    vcmp.f32 s2, s0
-; ARMV7-NEXT:    vmrs APSR_nzcv, fpscr
-; ARMV7-NEXT:    vmovgt.f32 s0, s2
+; ARMV7-NEXT:    vmax.f32 d0, d1, d0
 ; ARMV7-NEXT:    vmov r0, s0
 ; ARMV7-NEXT:    bx lr
 ;
diff --git a/llvm/test/CodeGen/BPF/CORE/offset-reloc-basic.ll b/llvm/test/CodeGen/BPF/CORE/offset-reloc-basic.ll
index 8ca8a6602737..024ed04f6e5e 100644
--- a/llvm/test/CodeGen/BPF/CORE/offset-reloc-basic.ll
+++ b/llvm/test/CodeGen/BPF/CORE/offset-reloc-basic.ll
@@ -108,8 +108,8 @@ define dso_local i32 @bpf_prog(ptr) local_unnamed_addr #0 !dbg !15 {
 ; CHECK-NEXT:        .long   0
 ; CHECK-NEXT:        .long   20
 ; CHECK-NEXT:        .long   20
-; CHECK-NEXT:        .long   124
-; CHECK-NEXT:        .long   144
+; CHECK-NEXT:        .long   108
+; CHECK-NEXT:        .long   128
 ; CHECK-NEXT:        .long   28
 ; CHECK-NEXT:        .long   8                       # FuncInfo
 
diff --git a/llvm/test/CodeGen/Hexagon/arg-copy-elison.ll b/llvm/test/CodeGen/Hexagon/arg-copy-elison.ll
new file mode 100644
index 000000000000..f0c30c301f44
--- /dev/null
+++ b/llvm/test/CodeGen/Hexagon/arg-copy-elison.ll
@@ -0,0 +1,39 @@
+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --version 4
+; RUN: llc -mtriple hexagon-- -o - %s | FileCheck %s
+
+; Reproducer for https://github.com/llvm/llvm-project/issues/89060
+;
+; Problem was a bug in argument copy elison. Given that the %alloca is
+; eliminated, the same frame index will be used for accessing %alloca and %a
+; on the fixed stack. Care must be taken when setting up
+; MachinePointerInfo/MemOperands for those accesses to either make sure that
+; we always refer to the fixed stack slot the same way (not using the
+; ir.alloca name), or make sure that we still detect that they alias each
+; other if using different kinds of MemOperands to identify the same fixed
+; stack entry.
+;
+define i32 @f(i32, i32, i32, i32, i32, i32, i32, i32, i32, i32, i32, i32, i32, i32 %q1, i32 %a, i32 %q2) {
+; CHECK-LABEL: f:
+; CHECK:         .cfi_startproc
+; CHECK-NEXT:  // %bb.0:
+; CHECK-NEXT:    {
+; CHECK-NEXT:     r0 = memw(r29+#36)
+; CHECK-NEXT:     r1 = memw(r29+#28)
+; CHECK-NEXT:    }
+; CHECK-NEXT:    {
+; CHECK-NEXT:     r0 = sub(r1,r0)
+; CHECK-NEXT:     r2 = memw(r29+#32)
+; CHECK-NEXT:     memw(r29+#32) = ##666
+; CHECK-NEXT:    }
+; CHECK-NEXT:    {
+; CHECK-NEXT:     r0 = xor(r0,r2)
+; CHECK-NEXT:     jumpr r31
+; CHECK-NEXT:    }
+  %alloca = alloca i32
+  store i32 %a, ptr %alloca     ; Should be elided.
+  store i32 666, ptr %alloca
+  %x = sub i32 %q1, %q2
+  %y = xor i32 %x, %a           ; Results in a load of %a from fixed stack.
+                                ; Using same frame index as elided %alloca.
+  ret i32 %y
+}
diff --git a/llvm/test/CodeGen/Hexagon/hexagon-copy-hoisting.mir b/llvm/test/CodeGen/Hexagon/hexagon-copy-hoisting.mir
new file mode 100644
index 000000000000..0836cac7f913
--- /dev/null
+++ b/llvm/test/CodeGen/Hexagon/hexagon-copy-hoisting.mir
@@ -0,0 +1,53 @@
+# RUN: llc -march=hexagon -run-pass hexagon-move-phicopy -o - %s | FileCheck %s
+
+# CHECK-COUNT-1: %4:intregs = COPY %1
+
+# CHECK: bb.1
+# CHECK-NOT: %4:intregs = COPY %1
+
+# CHECK: bb.2
+# CHECK-NOT: %4:intregs = COPY %1
+# CHECK: %5:intregs = COPY %0
+
+---
+name:            f0
+tracksRegLiveness: true
+registers:
+  - { id: 0, class: intregs, preferred-register: '' }
+  - { id: 1, class: intregs, preferred-register: '' }
+  - { id: 2, class: predregs, preferred-register: '' }
+  - { id: 3, class: predregs, preferred-register: '' }
+  - { id: 4, class: intregs, preferred-register: '' }
+  - { id: 5, class: intregs, preferred-register: '' }
+liveins:
+  - { reg: '$r0', virtual-reg: '%0' }
+  - { reg: '$r1', virtual-reg: '%1' }
+stack:
+  - { id: 0, offset: 0, size: 4, alignment: 8 }
+body: |
+  bb.0:
+    successors: %bb.1, %bb.2
+    liveins: $r0, $r1
+
+    %1:intregs = COPY $r1
+    %0:intregs = COPY $r0
+    %2:predregs = C2_cmpgt %0, %1
+    %3:predregs = C2_not %2
+    J2_jumpt %3, %bb.2, implicit-def dead $pc
+    J2_jump %bb.1, implicit-def dead $pc
+
+  bb.1:
+    successors: %bb.0
+
+    %4:intregs = COPY %1
+    $r1 = COPY %4
+    J2_jump %bb.0, implicit-def dead $pc
+
+  bb.2:
+    successors: %bb.0
+
+    %4:intregs = COPY %1
+    %5:intregs = COPY %0
+    $r1 = COPY %4
+    J2_jump %bb.0, implicit-def dead $pc
+...
diff --git a/llvm/test/CodeGen/LoongArch/lasx/ir-instruction/insertelement.ll b/llvm/test/CodeGen/LoongArch/lasx/ir-instruction/insertelement.ll
index 6629d3440549..25106b456d2f 100644
--- a/llvm/test/CodeGen/LoongArch/lasx/ir-instruction/insertelement.ll
+++ b/llvm/test/CodeGen/LoongArch/lasx/ir-instruction/insertelement.ll
@@ -123,10 +123,9 @@ define void @insert_32xi8_idx(ptr %src, ptr %dst, i8 %in, i32 %idx) nounwind {
 ; CHECK-NEXT:    bstrins.d $sp, $zero, 4, 0
 ; CHECK-NEXT:    xvld $xr0, $a0, 0
 ; CHECK-NEXT:    xvst $xr0, $sp, 0
-; CHECK-NEXT:    bstrpick.d $a0, $a3, 31, 0
-; CHECK-NEXT:    addi.d $a3, $sp, 0
-; CHECK-NEXT:    bstrins.d $a3, $a0, 4, 0
-; CHECK-NEXT:    st.b $a2, $a3, 0
+; CHECK-NEXT:    addi.d $a0, $sp, 0
+; CHECK-NEXT:    bstrins.d $a0, $a3, 4, 0
+; CHECK-NEXT:    st.b $a2, $a0, 0
 ; CHECK-NEXT:    xvld $xr0, $sp, 0
 ; CHECK-NEXT:    xvst $xr0, $a1, 0
 ; CHECK-NEXT:    addi.d $sp, $fp, -64
@@ -150,10 +149,9 @@ define void @insert_16xi16_idx(ptr %src, ptr %dst, i16 %in, i32 %idx) nounwind {
 ; CHECK-NEXT:    bstrins.d $sp, $zero, 4, 0
 ; CHECK-NEXT:    xvld $xr0, $a0, 0
 ; CHECK-NEXT:    xvst $xr0, $sp, 0
-; CHECK-NEXT:    bstrpick.d $a0, $a3, 31, 0
-; CHECK-NEXT:    addi.d $a3, $sp, 0
-; CHECK-NEXT:    bstrins.d $a3, $a0, 4, 1
-; CHECK-NEXT:    st.h $a2, $a3, 0
+; CHECK-NEXT:    addi.d $a0, $sp, 0
+; CHECK-NEXT:    bstrins.d $a0, $a3, 4, 1
+; CHECK-NEXT:    st.h $a2, $a0, 0
 ; CHECK-NEXT:    xvld $xr0, $sp, 0
 ; CHECK-NEXT:    xvst $xr0, $a1, 0
 ; CHECK-NEXT:    addi.d $sp, $fp, -64
@@ -177,10 +175,9 @@ define void @insert_8xi32_idx(ptr %src, ptr %dst, i32 %in, i32 %idx) nounwind {
 ; CHECK-NEXT:    bstrins.d $sp, $zero, 4, 0
 ; CHECK-NEXT:    xvld $xr0, $a0, 0
 ; CHECK-NEXT:    xvst $xr0, $sp, 0
-; CHECK-NEXT:    bstrpick.d $a0, $a3, 31, 0
-; CHECK-NEXT:    addi.d $a3, $sp, 0
-; CHECK-NEXT:    bstrins.d $a3, $a0, 4, 2
-; CHECK-NEXT:    st.w $a2, $a3, 0
+; CHECK-NEXT:    addi.d $a0, $sp, 0
+; CHECK-NEXT:    bstrins.d $a0, $a3, 4, 2
+; CHECK-NEXT:    st.w $a2, $a0, 0
 ; CHECK-NEXT:    xvld $xr0, $sp, 0
 ; CHECK-NEXT:    xvst $xr0, $a1, 0
 ; CHECK-NEXT:    addi.d $sp, $fp, -64
@@ -204,10 +201,9 @@ define void @insert_4xi64_idx(ptr %src, ptr %dst, i64 %in, i32 %idx) nounwind {
 ; CHECK-NEXT:    bstrins.d $sp, $zero, 4, 0
 ; CHECK-NEXT:    xvld $xr0, $a0, 0
 ; CHECK-NEXT:    xvst $xr0, $sp, 0
-; CHECK-NEXT:    bstrpick.d $a0, $a3, 31, 0
-; CHECK-NEXT:    addi.d $a3, $sp, 0
-; CHECK-NEXT:    bstrins.d $a3, $a0, 4, 3
-; CHECK-NEXT:    st.d $a2, $a3, 0
+; CHECK-NEXT:    addi.d $a0, $sp, 0
+; CHECK-NEXT:    bstrins.d $a0, $a3, 4, 3
+; CHECK-NEXT:    st.d $a2, $a0, 0
 ; CHECK-NEXT:    xvld $xr0, $sp, 0
 ; CHECK-NEXT:    xvst $xr0, $a1, 0
 ; CHECK-NEXT:    addi.d $sp, $fp, -64
@@ -231,10 +227,9 @@ define void @insert_8xfloat_idx(ptr %src, ptr %dst, float %in, i32 %idx) nounwin
 ; CHECK-NEXT:    bstrins.d $sp, $zero, 4, 0
 ; CHECK-NEXT:    xvld $xr1, $a0, 0
 ; CHECK-NEXT:    xvst $xr1, $sp, 0
-; CHECK-NEXT:    bstrpick.d $a0, $a2, 31, 0
-; CHECK-NEXT:    addi.d $a2, $sp, 0
-; CHECK-NEXT:    bstrins.d $a2, $a0, 4, 2
-; CHECK-NEXT:    fst.s $fa0, $a2, 0
+; CHECK-NEXT:    addi.d $a0, $sp, 0
+; CHECK-NEXT:    bstrins.d $a0, $a2, 4, 2
+; CHECK-NEXT:    fst.s $fa0, $a0, 0
 ; CHECK-NEXT:    xvld $xr0, $sp, 0
 ; CHECK-NEXT:    xvst $xr0, $a1, 0
 ; CHECK-NEXT:    addi.d $sp, $fp, -64
@@ -258,10 +253,9 @@ define void @insert_4xdouble_idx(ptr %src, ptr %dst, double %in, i32 %idx) nounw
 ; CHECK-NEXT:    bstrins.d $sp, $zero, 4, 0
 ; CHECK-NEXT:    xvld $xr1, $a0, 0
 ; CHECK-NEXT:    xvst $xr1, $sp, 0
-; CHECK-NEXT:    bstrpick.d $a0, $a2, 31, 0
-; CHECK-NEXT:    addi.d $a2, $sp, 0
-; CHECK-NEXT:    bstrins.d $a2, $a0, 4, 3
-; CHECK-NEXT:    fst.d $fa0, $a2, 0
+; CHECK-NEXT:    addi.d $a0, $sp, 0
+; CHECK-NEXT:    bstrins.d $a0, $a2, 4, 3
+; CHECK-NEXT:    fst.d $fa0, $a0, 0
 ; CHECK-NEXT:    xvld $xr0, $sp, 0
 ; CHECK-NEXT:    xvst $xr0, $a1, 0
 ; CHECK-NEXT:    addi.d $sp, $fp, -64
diff --git a/llvm/test/CodeGen/LoongArch/lsx/ir-instruction/insertelement.ll b/llvm/test/CodeGen/LoongArch/lsx/ir-instruction/insertelement.ll
index 19171b7d8ed7..7f232073ae12 100644
--- a/llvm/test/CodeGen/LoongArch/lsx/ir-instruction/insertelement.ll
+++ b/llvm/test/CodeGen/LoongArch/lsx/ir-instruction/insertelement.ll
@@ -87,10 +87,9 @@ define void @insert_16xi8_idx(ptr %src, ptr %dst, i8 %ins, i32 %idx) nounwind {
 ; CHECK-NEXT:    addi.d $sp, $sp, -16
 ; CHECK-NEXT:    vld $vr0, $a0, 0
 ; CHECK-NEXT:    vst $vr0, $sp, 0
-; CHECK-NEXT:    bstrpick.d $a0, $a3, 31, 0
-; CHECK-NEXT:    addi.d $a3, $sp, 0
-; CHECK-NEXT:    bstrins.d $a3, $a0, 3, 0
-; CHECK-NEXT:    st.b $a2, $a3, 0
+; CHECK-NEXT:    addi.d $a0, $sp, 0
+; CHECK-NEXT:    bstrins.d $a0, $a3, 3, 0
+; CHECK-NEXT:    st.b $a2, $a0, 0
 ; CHECK-NEXT:    vld $vr0, $sp, 0
 ; CHECK-NEXT:    vst $vr0, $a1, 0
 ; CHECK-NEXT:    addi.d $sp, $sp, 16
@@ -107,10 +106,9 @@ define void @insert_8xi16_idx(ptr %src, ptr %dst, i16 %ins, i32 %idx) nounwind {
 ; CHECK-NEXT:    addi.d $sp, $sp, -16
 ; CHECK-NEXT:    vld $vr0, $a0, 0
 ; CHECK-NEXT:    vst $vr0, $sp, 0
-; CHECK-NEXT:    bstrpick.d $a0, $a3, 31, 0
-; CHECK-NEXT:    addi.d $a3, $sp, 0
-; CHECK-NEXT:    bstrins.d $a3, $a0, 3, 1
-; CHECK-NEXT:    st.h $a2, $a3, 0
+; CHECK-NEXT:    addi.d $a0, $sp, 0
+; CHECK-NEXT:    bstrins.d $a0, $a3, 3, 1
+; CHECK-NEXT:    st.h $a2, $a0, 0
 ; CHECK-NEXT:    vld $vr0, $sp, 0
 ; CHECK-NEXT:    vst $vr0, $a1, 0
 ; CHECK-NEXT:    addi.d $sp, $sp, 16
@@ -127,10 +125,9 @@ define void @insert_4xi32_idx(ptr %src, ptr %dst, i32 %ins, i32 %idx) nounwind {
 ; CHECK-NEXT:    addi.d $sp, $sp, -16
 ; CHECK-NEXT:    vld $vr0, $a0, 0
 ; CHECK-NEXT:    vst $vr0, $sp, 0
-; CHECK-NEXT:    bstrpick.d $a0, $a3, 31, 0
-; CHECK-NEXT:    addi.d $a3, $sp, 0
-; CHECK-NEXT:    bstrins.d $a3, $a0, 3, 2
-; CHECK-NEXT:    st.w $a2, $a3, 0
+; CHECK-NEXT:    addi.d $a0, $sp, 0
+; CHECK-NEXT:    bstrins.d $a0, $a3, 3, 2
+; CHECK-NEXT:    st.w $a2, $a0, 0
 ; CHECK-NEXT:    vld $vr0, $sp, 0
 ; CHECK-NEXT:    vst $vr0, $a1, 0
 ; CHECK-NEXT:    addi.d $sp, $sp, 16
@@ -147,10 +144,9 @@ define void @insert_2xi64_idx(ptr %src, ptr %dst, i64 %ins, i32 %idx) nounwind {
 ; CHECK-NEXT:    addi.d $sp, $sp, -16
 ; CHECK-NEXT:    vld $vr0, $a0, 0
 ; CHECK-NEXT:    vst $vr0, $sp, 0
-; CHECK-NEXT:    bstrpick.d $a0, $a3, 31, 0
-; CHECK-NEXT:    addi.d $a3, $sp, 0
-; CHECK-NEXT:    bstrins.d $a3, $a0, 3, 3
-; CHECK-NEXT:    st.d $a2, $a3, 0
+; CHECK-NEXT:    addi.d $a0, $sp, 0
+; CHECK-NEXT:    bstrins.d $a0, $a3, 3, 3
+; CHECK-NEXT:    st.d $a2, $a0, 0
 ; CHECK-NEXT:    vld $vr0, $sp, 0
 ; CHECK-NEXT:    vst $vr0, $a1, 0
 ; CHECK-NEXT:    addi.d $sp, $sp, 16
@@ -167,10 +163,9 @@ define void @insert_4xfloat_idx(ptr %src, ptr %dst, float %ins, i32 %idx) nounwi
 ; CHECK-NEXT:    addi.d $sp, $sp, -16
 ; CHECK-NEXT:    vld $vr1, $a0, 0
 ; CHECK-NEXT:    vst $vr1, $sp, 0
-; CHECK-NEXT:    bstrpick.d $a0, $a2, 31, 0
-; CHECK-NEXT:    addi.d $a2, $sp, 0
-; CHECK-NEXT:    bstrins.d $a2, $a0, 3, 2
-; CHECK-NEXT:    fst.s $fa0, $a2, 0
+; CHECK-NEXT:    addi.d $a0, $sp, 0
+; CHECK-NEXT:    bstrins.d $a0, $a2, 3, 2
+; CHECK-NEXT:    fst.s $fa0, $a0, 0
 ; CHECK-NEXT:    vld $vr0, $sp, 0
 ; CHECK-NEXT:    vst $vr0, $a1, 0
 ; CHECK-NEXT:    addi.d $sp, $sp, 16
@@ -187,10 +182,9 @@ define void @insert_2xdouble_idx(ptr %src, ptr %dst, double %ins, i32 %idx) noun
 ; CHECK-NEXT:    addi.d $sp, $sp, -16
 ; CHECK-NEXT:    vld $vr1, $a0, 0
 ; CHECK-NEXT:    vst $vr1, $sp, 0
-; CHECK-NEXT:    bstrpick.d $a0, $a2, 31, 0
-; CHECK-NEXT:    addi.d $a2, $sp, 0
-; CHECK-NEXT:    bstrins.d $a2, $a0, 3, 3
-; CHECK-NEXT:    fst.d $fa0, $a2, 0
+; CHECK-NEXT:    addi.d $a0, $sp, 0
+; CHECK-NEXT:    bstrins.d $a0, $a2, 3, 3
+; CHECK-NEXT:    fst.d $fa0, $a0, 0
 ; CHECK-NEXT:    vld $vr0, $sp, 0
 ; CHECK-NEXT:    vst $vr0, $a1, 0
 ; CHECK-NEXT:    addi.d $sp, $sp, 16
diff --git a/llvm/test/CodeGen/LoongArch/sextw-removal.ll b/llvm/test/CodeGen/LoongArch/sextw-removal.ll
new file mode 100644
index 000000000000..6db9c1608b3c
--- /dev/null
+++ b/llvm/test/CodeGen/LoongArch/sextw-removal.ll
@@ -0,0 +1,921 @@
+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
+; RUN: llc < %s --mtriple=loongarch64 | FileCheck %s --check-prefixes=CHECK
+
+define void @test1(i32 signext %arg, i32 signext %arg1) nounwind {
+; CHECK-LABEL: test1:
+; CHECK:       # %bb.0: # %bb
+; CHECK-NEXT:    addi.d $sp, $sp, -32
+; CHECK-NEXT:    st.d $ra, $sp, 24 # 8-byte Folded Spill
+; CHECK-NEXT:    st.d $fp, $sp, 16 # 8-byte Folded Spill
+; CHECK-NEXT:    st.d $s0, $sp, 8 # 8-byte Folded Spill
+; CHECK-NEXT:    move $fp, $a1
+; CHECK-NEXT:    sra.w $s0, $a0, $a1
+; CHECK-NEXT:    .p2align 4, , 16
+; CHECK-NEXT:  .LBB0_1: # %bb2
+; CHECK-NEXT:    # =>This Inner Loop Header: Depth=1
+; CHECK-NEXT:    addi.w $a0, $s0, 0
+; CHECK-NEXT:    bl %plt(bar)
+; CHECK-NEXT:    sll.w $s0, $s0, $fp
+; CHECK-NEXT:    bnez $a0, .LBB0_1
+; CHECK-NEXT:  # %bb.2: # %bb7
+; CHECK-NEXT:    ld.d $s0, $sp, 8 # 8-byte Folded Reload
+; CHECK-NEXT:    ld.d $fp, $sp, 16 # 8-byte Folded Reload
+; CHECK-NEXT:    ld.d $ra, $sp, 24 # 8-byte Folded Reload
+; CHECK-NEXT:    addi.d $sp, $sp, 32
+; CHECK-NEXT:    ret
+bb:
+  %i = ashr i32 %arg, %arg1
+  br label %bb2
+
+bb2:                                              ; preds = %bb2, %bb
+  %i3 = phi i32 [ %i, %bb ], [ %i5, %bb2 ]
+  %i4 = tail call signext i32 @bar(i32 signext %i3)
+  %i5 = shl i32 %i3, %arg1
+  %i6 = icmp eq i32 %i4, 0
+  br i1 %i6, label %bb7, label %bb2
+
+bb7:                                              ; preds = %bb2
+  ret void
+}
+
+declare signext i32 @bar(i32 signext)
+
+define signext i32 @test2(ptr %p, i32 signext %b) nounwind {
+; CHECK-LABEL: test2:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    ld.w $a0, $a0, 0
+; CHECK-NEXT:    ori $a2, $zero, 1
+; CHECK-NEXT:    sll.w $a1, $a2, $a1
+; CHECK-NEXT:    andn $a0, $a0, $a1
+; CHECK-NEXT:    addi.w $a0, $a0, 0
+; CHECK-NEXT:    ret
+  %a = load i32, ptr %p
+  %shl = shl i32 1, %b
+  %neg = xor i32 %shl, -1
+  %and1 = and i32 %neg, %a
+  ret i32 %and1
+}
+
+define signext i32 @test3(ptr %p, i32 signext %b) nounwind {
+; CHECK-LABEL: test3:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    ld.w $a0, $a0, 0
+; CHECK-NEXT:    ori $a2, $zero, 1
+; CHECK-NEXT:    sll.w $a1, $a2, $a1
+; CHECK-NEXT:    orn $a0, $a0, $a1
+; CHECK-NEXT:    addi.w $a0, $a0, 0
+; CHECK-NEXT:    ret
+  %a = load i32, ptr %p
+  %shl = shl i32 1, %b
+  %neg = xor i32 %shl, -1
+  %and1 = or i32 %neg, %a
+  ret i32 %and1
+}
+
+define signext i32 @test4(ptr %p, i32 signext %b) nounwind {
+; CHECK-LABEL: test4:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    ld.w $a0, $a0, 0
+; CHECK-NEXT:    ori $a2, $zero, 1
+; CHECK-NEXT:    sll.w $a1, $a2, $a1
+; CHECK-NEXT:    xor $a0, $a1, $a0
+; CHECK-NEXT:    nor $a0, $a0, $zero
+; CHECK-NEXT:    addi.w $a0, $a0, 0
+; CHECK-NEXT:    ret
+  %a = load i32, ptr %p
+  %shl = shl i32 1, %b
+  %neg = xor i32 %shl, -1
+  %and1 = xor i32 %neg, %a
+  ret i32 %and1
+}
+
+define void @test5(i32 signext %arg, i32 signext %arg1) nounwind {
+; CHECK-LABEL: test5:
+; CHECK:       # %bb.0: # %bb
+; CHECK-NEXT:    addi.d $sp, $sp, -48
+; CHECK-NEXT:    st.d $ra, $sp, 40 # 8-byte Folded Spill
+; CHECK-NEXT:    st.d $fp, $sp, 32 # 8-byte Folded Spill
+; CHECK-NEXT:    st.d $s0, $sp, 24 # 8-byte Folded Spill
+; CHECK-NEXT:    st.d $s1, $sp, 16 # 8-byte Folded Spill
+; CHECK-NEXT:    st.d $s2, $sp, 8 # 8-byte Folded Spill
+; CHECK-NEXT:    sra.w $a1, $a0, $a1
+; CHECK-NEXT:    lu12i.w $a0, 349525
+; CHECK-NEXT:    ori $fp, $a0, 1365
+; CHECK-NEXT:    lu12i.w $a0, 209715
+; CHECK-NEXT:    ori $s0, $a0, 819
+; CHECK-NEXT:    lu12i.w $a0, 61680
+; CHECK-NEXT:    ori $s1, $a0, 3855
+; CHECK-NEXT:    lu12i.w $a0, 4112
+; CHECK-NEXT:    ori $s2, $a0, 257
+; CHECK-NEXT:    .p2align 4, , 16
+; CHECK-NEXT:  .LBB4_1: # %bb2
+; CHECK-NEXT:    # =>This Inner Loop Header: Depth=1
+; CHECK-NEXT:    addi.w $a0, $a1, 0
+; CHECK-NEXT:    bl %plt(bar)
+; CHECK-NEXT:    srli.d $a1, $a0, 1
+; CHECK-NEXT:    and $a1, $a1, $fp
+; CHECK-NEXT:    sub.d $a1, $a0, $a1
+; CHECK-NEXT:    and $a2, $a1, $s0
+; CHECK-NEXT:    srli.d $a1, $a1, 2
+; CHECK-NEXT:    and $a1, $a1, $s0
+; CHECK-NEXT:    add.d $a1, $a2, $a1
+; CHECK-NEXT:    srli.d $a2, $a1, 4
+; CHECK-NEXT:    add.d $a1, $a1, $a2
+; CHECK-NEXT:    and $a1, $a1, $s1
+; CHECK-NEXT:    mul.d $a1, $a1, $s2
+; CHECK-NEXT:    bstrpick.d $a1, $a1, 31, 24
+; CHECK-NEXT:    bnez $a0, .LBB4_1
+; CHECK-NEXT:  # %bb.2: # %bb7
+; CHECK-NEXT:    ld.d $s2, $sp, 8 # 8-byte Folded Reload
+; CHECK-NEXT:    ld.d $s1, $sp, 16 # 8-byte Folded Reload
+; CHECK-NEXT:    ld.d $s0, $sp, 24 # 8-byte Folded Reload
+; CHECK-NEXT:    ld.d $fp, $sp, 32 # 8-byte Folded Reload
+; CHECK-NEXT:    ld.d $ra, $sp, 40 # 8-byte Folded Reload
+; CHECK-NEXT:    addi.d $sp, $sp, 48
+; CHECK-NEXT:    ret
+bb:
+  %i = ashr i32 %arg, %arg1
+  br label %bb2
+
+bb2:                                              ; preds = %bb2, %bb
+  %i3 = phi i32 [ %i, %bb ], [ %i5, %bb2 ]
+  %i4 = tail call signext i32 @bar(i32 signext %i3)
+  %i5 = tail call i32 @llvm.ctpop.i32(i32 %i4)
+  %i6 = icmp eq i32 %i4, 0
+  br i1 %i6, label %bb7, label %bb2
+
+bb7:                                              ; preds = %bb2
+  ret void
+}
+
+declare i32 @llvm.ctpop.i32(i32)
+
+define void @test6(i32 signext %arg, i32 signext %arg1) nounwind {
+; CHECK-LABEL: test6:
+; CHECK:       # %bb.0: # %bb
+; CHECK-NEXT:    addi.d $sp, $sp, -32
+; CHECK-NEXT:    st.d $ra, $sp, 24 # 8-byte Folded Spill
+; CHECK-NEXT:    st.d $fp, $sp, 16 # 8-byte Folded Spill
+; CHECK-NEXT:    st.d $s0, $sp, 8 # 8-byte Folded Spill
+; CHECK-NEXT:    sra.w $fp, $a0, $a1
+; CHECK-NEXT:    .p2align 4, , 16
+; CHECK-NEXT:  .LBB5_1: # %bb2
+; CHECK-NEXT:    # =>This Inner Loop Header: Depth=1
+; CHECK-NEXT:    addi.w $a0, $fp, 0
+; CHECK-NEXT:    bl %plt(baz)
+; CHECK-NEXT:    bstrpick.d $s0, $a0, 31, 0
+; CHECK-NEXT:    move $a0, $s0
+; CHECK-NEXT:    bl %plt(__fixsfsi)
+; CHECK-NEXT:    move $fp, $a0
+; CHECK-NEXT:    move $a0, $s0
+; CHECK-NEXT:    move $a1, $zero
+; CHECK-NEXT:    bl %plt(__nesf2)
+; CHECK-NEXT:    bnez $a0, .LBB5_1
+; CHECK-NEXT:  # %bb.2: # %bb7
+; CHECK-NEXT:    ld.d $s0, $sp, 8 # 8-byte Folded Reload
+; CHECK-NEXT:    ld.d $fp, $sp, 16 # 8-byte Folded Reload
+; CHECK-NEXT:    ld.d $ra, $sp, 24 # 8-byte Folded Reload
+; CHECK-NEXT:    addi.d $sp, $sp, 32
+; CHECK-NEXT:    ret
+bb:
+  %i = ashr i32 %arg, %arg1
+  br label %bb2
+
+bb2:                                              ; preds = %bb2, %bb
+  %i3 = phi i32 [ %i, %bb ], [ %i5, %bb2 ]
+  %i4 = tail call float @baz(i32 signext %i3)
+  %i5 = fptosi float %i4 to i32
+  %i6 = fcmp oeq float %i4, zeroinitializer
+  br i1 %i6, label %bb7, label %bb2
+
+bb7:                                              ; preds = %bb2
+  ret void
+}
+declare float @baz(i32 signext %i3)
+
+define void @test7(i32 signext %arg, i32 signext %arg1) nounwind {
+; CHECK-LABEL: test7:
+; CHECK:       # %bb.0: # %bb
+; CHECK-NEXT:    addi.d $sp, $sp, -48
+; CHECK-NEXT:    st.d $ra, $sp, 40 # 8-byte Folded Spill
+; CHECK-NEXT:    st.d $fp, $sp, 32 # 8-byte Folded Spill
+; CHECK-NEXT:    st.d $s0, $sp, 24 # 8-byte Folded Spill
+; CHECK-NEXT:    st.d $s1, $sp, 16 # 8-byte Folded Spill
+; CHECK-NEXT:    st.d $s2, $sp, 8 # 8-byte Folded Spill
+; CHECK-NEXT:    sra.w $a0, $a0, $a1
+; CHECK-NEXT:    lu12i.w $a1, 349525
+; CHECK-NEXT:    ori $a1, $a1, 1365
+; CHECK-NEXT:    lu32i.d $a1, 349525
+; CHECK-NEXT:    lu52i.d $fp, $a1, 1365
+; CHECK-NEXT:    lu12i.w $a1, 209715
+; CHECK-NEXT:    ori $a1, $a1, 819
+; CHECK-NEXT:    lu32i.d $a1, 209715
+; CHECK-NEXT:    lu52i.d $s0, $a1, 819
+; CHECK-NEXT:    lu12i.w $a1, 61680
+; CHECK-NEXT:    ori $a1, $a1, 3855
+; CHECK-NEXT:    lu32i.d $a1, -61681
+; CHECK-NEXT:    lu52i.d $s1, $a1, 240
+; CHECK-NEXT:    lu12i.w $a1, 4112
+; CHECK-NEXT:    ori $a1, $a1, 257
+; CHECK-NEXT:    lu32i.d $a1, 65793
+; CHECK-NEXT:    lu52i.d $s2, $a1, 16
+; CHECK-NEXT:    .p2align 4, , 16
+; CHECK-NEXT:  .LBB6_1: # %bb2
+; CHECK-NEXT:    # =>This Inner Loop Header: Depth=1
+; CHECK-NEXT:    addi.w $a0, $a0, 0
+; CHECK-NEXT:    bl %plt(foo)
+; CHECK-NEXT:    srli.d $a1, $a0, 1
+; CHECK-NEXT:    and $a1, $a1, $fp
+; CHECK-NEXT:    sub.d $a0, $a0, $a1
+; CHECK-NEXT:    and $a1, $a0, $s0
+; CHECK-NEXT:    srli.d $a0, $a0, 2
+; CHECK-NEXT:    and $a0, $a0, $s0
+; CHECK-NEXT:    add.d $a0, $a1, $a0
+; CHECK-NEXT:    srli.d $a1, $a0, 4
+; CHECK-NEXT:    add.d $a0, $a0, $a1
+; CHECK-NEXT:    and $a0, $a0, $s1
+; CHECK-NEXT:    mul.d $a0, $a0, $s2
+; CHECK-NEXT:    srli.d $a0, $a0, 56
+; CHECK-NEXT:    bnez $a0, .LBB6_1
+; CHECK-NEXT:  # %bb.2: # %bb7
+; CHECK-NEXT:    ld.d $s2, $sp, 8 # 8-byte Folded Reload
+; CHECK-NEXT:    ld.d $s1, $sp, 16 # 8-byte Folded Reload
+; CHECK-NEXT:    ld.d $s0, $sp, 24 # 8-byte Folded Reload
+; CHECK-NEXT:    ld.d $fp, $sp, 32 # 8-byte Folded Reload
+; CHECK-NEXT:    ld.d $ra, $sp, 40 # 8-byte Folded Reload
+; CHECK-NEXT:    addi.d $sp, $sp, 48
+; CHECK-NEXT:    ret
+bb:
+  %i = ashr i32 %arg, %arg1
+  br label %bb2
+
+bb2:                                              ; preds = %bb2, %bb
+  %i3 = phi i32 [ %i, %bb ], [ %i6, %bb2 ]
+  %i4 = tail call signext i64 @foo(i32 signext %i3)
+  %i5 = tail call i64 @llvm.ctpop.i64(i64 %i4)
+  %i6 = trunc i64 %i5 to i32
+  %i7 = icmp eq i32 %i6, 0
+  br i1 %i7, label %bb7, label %bb2
+
+bb7:                                              ; preds = %bb2
+  ret void
+}
+
+declare i64 @llvm.ctpop.i64(i64)
+
+define void @test8(i32 signext %arg, i32 signext %arg1) nounwind {
+; CHECK-LABEL: test8:
+; CHECK:       # %bb.0: # %bb
+; CHECK-NEXT:    addi.d $sp, $sp, -16
+; CHECK-NEXT:    st.d $ra, $sp, 8 # 8-byte Folded Spill
+; CHECK-NEXT:    st.d $fp, $sp, 0 # 8-byte Folded Spill
+; CHECK-NEXT:    sra.w $a0, $a0, $a1
+; CHECK-NEXT:    addi.w $fp, $zero, -256
+; CHECK-NEXT:    .p2align 4, , 16
+; CHECK-NEXT:  .LBB7_1: # %bb2
+; CHECK-NEXT:    # =>This Inner Loop Header: Depth=1
+; CHECK-NEXT:    addi.w $a0, $a0, 0
+; CHECK-NEXT:    bl %plt(foo)
+; CHECK-NEXT:    or $a0, $a0, $fp
+; CHECK-NEXT:    bnez $a0, .LBB7_1
+; CHECK-NEXT:  # %bb.2: # %bb7
+; CHECK-NEXT:    ld.d $fp, $sp, 0 # 8-byte Folded Reload
+; CHECK-NEXT:    ld.d $ra, $sp, 8 # 8-byte Folded Reload
+; CHECK-NEXT:    addi.d $sp, $sp, 16
+; CHECK-NEXT:    ret
+bb:
+  %i = ashr i32 %arg, %arg1
+  br label %bb2
+
+bb2:                                              ; preds = %bb2, %bb
+  %i3 = phi i32 [ %i, %bb ], [ %i6, %bb2 ]
+  %i4 = tail call signext i64 @foo(i32 signext %i3)
+  %i5 = or i64 %i4, -256
+  %i6 = trunc i64 %i5 to i32
+  %i7 = icmp eq i32 %i6, 0
+  br i1 %i7, label %bb7, label %bb2
+
+bb7:                                              ; preds = %bb2
+  ret void
+}
+
+declare i64 @foo(i32 signext)
+
+define void @test9(i32 signext %arg, i32 signext %arg1) nounwind {
+; CHECK-LABEL: test9:
+; CHECK:       # %bb.0: # %bb
+; CHECK-NEXT:    addi.d $sp, $sp, -16
+; CHECK-NEXT:    st.d $ra, $sp, 8 # 8-byte Folded Spill
+; CHECK-NEXT:    st.d $fp, $sp, 0 # 8-byte Folded Spill
+; CHECK-NEXT:    sra.w $a1, $a0, $a1
+; CHECK-NEXT:    ori $fp, $zero, 254
+; CHECK-NEXT:    .p2align 4, , 16
+; CHECK-NEXT:  .LBB8_1: # %bb2
+; CHECK-NEXT:    # =>This Inner Loop Header: Depth=1
+; CHECK-NEXT:    addi.w $a0, $a1, 0
+; CHECK-NEXT:    bl %plt(bar)
+; CHECK-NEXT:    slti $a1, $a0, 255
+; CHECK-NEXT:    blt $fp, $a0, .LBB8_1
+; CHECK-NEXT:  # %bb.2: # %bb7
+; CHECK-NEXT:    ld.d $fp, $sp, 0 # 8-byte Folded Reload
+; CHECK-NEXT:    ld.d $ra, $sp, 8 # 8-byte Folded Reload
+; CHECK-NEXT:    addi.d $sp, $sp, 16
+; CHECK-NEXT:    ret
+bb:
+  %i = ashr i32 %arg, %arg1
+  br label %bb2
+
+bb2:                                              ; preds = %bb2, %bb
+  %i3 = phi i32 [ %i, %bb ], [ %i7, %bb2 ]
+  %i4 = tail call signext i32 @bar(i32 signext %i3)
+  %i5 = icmp slt i32 %i4, 255
+  %i6 = sext i1 %i5 to i32
+  %i7 = sub i32 0, %i6
+  br i1 %i5, label %bb7, label %bb2
+
+bb7:                                              ; preds = %bb2
+  ret void
+}
+
+define void @test10(i32 signext %arg, i32 signext %arg1) nounwind {
+; CHECK-LABEL: test10:
+; CHECK:       # %bb.0: # %bb
+; CHECK-NEXT:    addi.d $sp, $sp, -16
+; CHECK-NEXT:    st.d $ra, $sp, 8 # 8-byte Folded Spill
+; CHECK-NEXT:    st.d $fp, $sp, 0 # 8-byte Folded Spill
+; CHECK-NEXT:    sra.w $fp, $a0, $a1
+; CHECK-NEXT:    .p2align 4, , 16
+; CHECK-NEXT:  .LBB9_1: # %bb2
+; CHECK-NEXT:    # =>This Inner Loop Header: Depth=1
+; CHECK-NEXT:    addi.w $a0, $fp, 0
+; CHECK-NEXT:    bl %plt(baz)
+; CHECK-NEXT:    move $fp, $a0
+; CHECK-NEXT:    bstrpick.d $a0, $a0, 31, 0
+; CHECK-NEXT:    move $a1, $zero
+; CHECK-NEXT:    bl %plt(__nesf2)
+; CHECK-NEXT:    bnez $a0, .LBB9_1
+; CHECK-NEXT:  # %bb.2: # %bb7
+; CHECK-NEXT:    ld.d $fp, $sp, 0 # 8-byte Folded Reload
+; CHECK-NEXT:    ld.d $ra, $sp, 8 # 8-byte Folded Reload
+; CHECK-NEXT:    addi.d $sp, $sp, 16
+; CHECK-NEXT:    ret
+bb:
+  %i = ashr i32 %arg, %arg1
+  br label %bb2
+
+bb2:                                              ; preds = %bb2, %bb
+  %i3 = phi i32 [ %i, %bb ], [ %i5, %bb2 ]
+  %i4 = tail call float @baz(i32 signext %i3)
+  %i5 = bitcast float %i4 to i32
+  %i6 = fcmp oeq float %i4, zeroinitializer
+  br i1 %i6, label %bb7, label %bb2
+
+bb7:                                              ; preds = %bb2
+  ret void
+}
+
+define signext i32 @test11(i64 %arg1, i64 %arg2, i64 %arg3)  {
+; CHECK-LABEL: test11:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    addi.d $a2, $a2, -1
+; CHECK-NEXT:    ori $a3, $zero, 256
+; CHECK-NEXT:    .p2align 4, , 16
+; CHECK-NEXT:  .LBB10_1: # %bb2
+; CHECK-NEXT:    # =>This Inner Loop Header: Depth=1
+; CHECK-NEXT:    andi $a0, $a0, 1234
+; CHECK-NEXT:    addi.d $a2, $a2, 1
+; CHECK-NEXT:    add.d $a0, $a0, $a1
+; CHECK-NEXT:    bltu $a2, $a3, .LBB10_1
+; CHECK-NEXT:  # %bb.2: # %bb7
+; CHECK-NEXT:    addi.w $a0, $a0, 0
+; CHECK-NEXT:    ret
+entry:
+  br label %bb2
+
+bb2:                                              ; preds = %bb2, %entry
+  %i1 = phi i64 [ %arg1, %entry ], [ %i5, %bb2 ]
+  %i2 = phi i64 [ %arg3, %entry ], [ %i3, %bb2 ]
+  %i3 = add i64 %i2, 1
+  %i4 = and i64 %i1, 1234
+  %i5 = add i64 %i4, %arg2
+  %i6 = icmp ugt i64 %i2, 255
+  br i1 %i6, label %bb7, label %bb2
+
+bb7:                                              ; preds = %bb2
+  %i7 = trunc i64 %i5 to i32
+  ret i32 %i7
+}
+
+define signext i32 @test12(i64 %arg1, i64 %arg2, i64 %arg3)  {
+; CHECK-LABEL: test12:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    addi.d $a2, $a2, -1
+; CHECK-NEXT:    ori $a3, $zero, 256
+; CHECK-NEXT:    .p2align 4, , 16
+; CHECK-NEXT:  .LBB11_1: # %bb2
+; CHECK-NEXT:    # =>This Inner Loop Header: Depth=1
+; CHECK-NEXT:    xor $a0, $a0, $a1
+; CHECK-NEXT:    mul.d $a4, $a0, $a1
+; CHECK-NEXT:    add.d $a0, $a0, $a4
+; CHECK-NEXT:    and $a4, $a4, $a0
+; CHECK-NEXT:    addi.d $a2, $a2, 1
+; CHECK-NEXT:    add.d $a0, $a4, $a1
+; CHECK-NEXT:    bltu $a2, $a3, .LBB11_1
+; CHECK-NEXT:  # %bb.2: # %bb7
+; CHECK-NEXT:    addi.w $a0, $a4, 0
+; CHECK-NEXT:    ret
+entry:
+  br label %bb2
+
+bb2:                                              ; preds = %bb2, %entry
+  %i1 = phi i64 [ %arg1, %entry ], [ %i6, %bb2 ]
+  %i2 = phi i64 [ %arg3, %entry ], [ %i3, %bb2 ]
+  %i3 = add i64 %i2, 1
+  %i4 = xor i64 %i1, %arg2
+  %i5 = mul i64 %i4, %arg2
+  %i9 = add i64 %i4, %i5
+  %i8 = and i64 %i5, %i9
+  %i6 = add i64 %i8, %arg2
+  %i7 = icmp ugt i64 %i2, 255
+  br i1 %i7, label %bb7, label %bb2
+
+bb7:                                              ; preds = %bb2
+  %r = trunc i64 %i8 to i32
+  ret i32 %r
+}
+
+define signext i32 @test13(i64 %arg1, i64 %arg2, i64 %arg3)  {
+; CHECK-LABEL: test13:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    addi.d $a2, $a2, -1
+; CHECK-NEXT:    ori $a3, $zero, 256
+; CHECK-NEXT:    .p2align 4, , 16
+; CHECK-NEXT:  .LBB12_1: # %bb2
+; CHECK-NEXT:    # =>This Inner Loop Header: Depth=1
+; CHECK-NEXT:    div.d $a0, $a0, $a1
+; CHECK-NEXT:    addi.d $a2, $a2, 1
+; CHECK-NEXT:    add.d $a0, $a0, $a1
+; CHECK-NEXT:    bltu $a2, $a3, .LBB12_1
+; CHECK-NEXT:  # %bb.2: # %bb7
+; CHECK-NEXT:    addi.w $a0, $a0, 0
+; CHECK-NEXT:    ret
+entry:
+  br label %bb2
+
+bb2:                                              ; preds = %bb2, %entry
+  %i1 = phi i64 [ %arg1, %entry ], [ %i5, %bb2 ]
+  %i2 = phi i64 [ %arg3, %entry ], [ %i3, %bb2 ]
+  %i3 = add i64 %i2, 1
+  %i4 = sdiv i64 %i1, %arg2
+  %i5 = add i64 %i4, %arg2
+  %i6 = icmp ugt i64 %i2, 255
+  br i1 %i6, label %bb7, label %bb2
+
+bb7:                                              ; preds = %bb2
+  %i8 = trunc i64 %i5 to i32
+  ret i32 %i8
+}
+
+
+define signext i32 @test14(i32 signext %0, i32 signext %1) {
+; CHECK-LABEL: test14:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    ori $a2, $zero, 2
+; CHECK-NEXT:    blt $a1, $a2, .LBB13_4
+; CHECK-NEXT:  # %bb.1: # %.preheader
+; CHECK-NEXT:    ori $a3, $zero, 1
+; CHECK-NEXT:    addi.w $a2, $zero, -1
+; CHECK-NEXT:    lu32i.d $a2, 0
+; CHECK-NEXT:    ori $a4, $zero, 1000
+; CHECK-NEXT:    .p2align 4, , 16
+; CHECK-NEXT:  .LBB13_2: # =>This Inner Loop Header: Depth=1
+; CHECK-NEXT:    addi.w $a5, $a0, 0
+; CHECK-NEXT:    blt $a4, $a5, .LBB13_5
+; CHECK-NEXT:  # %bb.3: # in Loop: Header=BB13_2 Depth=1
+; CHECK-NEXT:    add.d $a0, $a3, $a0
+; CHECK-NEXT:    addi.w $a3, $a3, 1
+; CHECK-NEXT:    blt $a3, $a1, .LBB13_2
+; CHECK-NEXT:  .LBB13_4:
+; CHECK-NEXT:    addi.w $a0, $a0, 0
+; CHECK-NEXT:    ret
+; CHECK-NEXT:  .LBB13_5:
+; CHECK-NEXT:    addi.w $a0, $a2, 0
+; CHECK-NEXT:    ret
+  %3 = icmp sgt i32 %1, 1
+  br i1 %3, label %4, label %12
+
+4:                                                ; preds = %2, %8
+  %5 = phi i32 [ %10, %8 ], [ 1, %2 ]
+  %6 = phi i32 [ %9, %8 ], [ %0, %2 ]
+  %7 = icmp sgt i32 %6, 1000
+  br i1 %7, label %12, label %8
+
+8:                                                ; preds = %4
+  %9 = add nsw i32 %5, %6
+  %10 = add nuw nsw i32 %5, 1
+  %11 = icmp slt i32 %10, %1
+  br i1 %11, label %4, label %12
+
+12:                                               ; preds = %8, %4, %2
+  %13 = phi i32 [ %0, %2 ], [ -1, %4 ], [ %9, %8 ]
+  ret i32 %13
+}
+
+define signext i32 @test14b(i32 %0, i32 signext %1) {
+; CHECK-LABEL: test14b:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    ori $a2, $zero, 2
+; CHECK-NEXT:    blt $a1, $a2, .LBB14_4
+; CHECK-NEXT:  # %bb.1: # %.preheader
+; CHECK-NEXT:    ori $a3, $zero, 1
+; CHECK-NEXT:    addi.w $a2, $zero, -1
+; CHECK-NEXT:    lu32i.d $a2, 0
+; CHECK-NEXT:    ori $a4, $zero, 1000
+; CHECK-NEXT:    .p2align 4, , 16
+; CHECK-NEXT:  .LBB14_2: # =>This Inner Loop Header: Depth=1
+; CHECK-NEXT:    addi.w $a5, $a0, 0
+; CHECK-NEXT:    blt $a4, $a5, .LBB14_5
+; CHECK-NEXT:  # %bb.3: # in Loop: Header=BB14_2 Depth=1
+; CHECK-NEXT:    add.d $a0, $a3, $a0
+; CHECK-NEXT:    addi.w $a3, $a3, 1
+; CHECK-NEXT:    blt $a3, $a1, .LBB14_2
+; CHECK-NEXT:  .LBB14_4:
+; CHECK-NEXT:    addi.w $a0, $a0, 0
+; CHECK-NEXT:    ret
+; CHECK-NEXT:  .LBB14_5:
+; CHECK-NEXT:    addi.w $a0, $a2, 0
+; CHECK-NEXT:    ret
+  %3 = icmp sgt i32 %1, 1
+  br i1 %3, label %4, label %12
+
+4:                                                ; preds = %2, %8
+  %5 = phi i32 [ %10, %8 ], [ 1, %2 ]
+  %6 = phi i32 [ %9, %8 ], [ %0, %2 ]
+  %7 = icmp sgt i32 %6, 1000
+  br i1 %7, label %12, label %8
+
+8:                                                ; preds = %4
+  %9 = add nsw i32 %5, %6
+  %10 = add nuw nsw i32 %5, 1
+  %11 = icmp slt i32 %10, %1
+  br i1 %11, label %4, label %12
+
+12:                                               ; preds = %8, %4, %2
+  %13 = phi i32 [ %0, %2 ], [ -1, %4 ], [ %9, %8 ]
+  ret i32 %13
+}
+
+define signext i32 @test14c(i32 zeroext %0, i32 signext %1) {
+; CHECK-LABEL: test14c:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    ori $a2, $zero, 2
+; CHECK-NEXT:    blt $a1, $a2, .LBB15_4
+; CHECK-NEXT:  # %bb.1: # %.preheader
+; CHECK-NEXT:    ori $a3, $zero, 1
+; CHECK-NEXT:    addi.w $a2, $zero, -1
+; CHECK-NEXT:    lu32i.d $a2, 0
+; CHECK-NEXT:    ori $a4, $zero, 1000
+; CHECK-NEXT:    .p2align 4, , 16
+; CHECK-NEXT:  .LBB15_2: # =>This Inner Loop Header: Depth=1
+; CHECK-NEXT:    addi.w $a5, $a0, 0
+; CHECK-NEXT:    blt $a4, $a5, .LBB15_5
+; CHECK-NEXT:  # %bb.3: # in Loop: Header=BB15_2 Depth=1
+; CHECK-NEXT:    add.d $a0, $a3, $a0
+; CHECK-NEXT:    addi.w $a3, $a3, 1
+; CHECK-NEXT:    blt $a3, $a1, .LBB15_2
+; CHECK-NEXT:  .LBB15_4:
+; CHECK-NEXT:    addi.w $a0, $a0, 0
+; CHECK-NEXT:    ret
+; CHECK-NEXT:  .LBB15_5:
+; CHECK-NEXT:    addi.w $a0, $a2, 0
+; CHECK-NEXT:    ret
+  %3 = icmp sgt i32 %1, 1
+  br i1 %3, label %4, label %12
+
+4:                                                ; preds = %2, %8
+  %5 = phi i32 [ %10, %8 ], [ 1, %2 ]
+  %6 = phi i32 [ %9, %8 ], [ %0, %2 ]
+  %7 = icmp sgt i32 %6, 1000
+  br i1 %7, label %12, label %8
+
+8:                                                ; preds = %4
+  %9 = add nsw i32 %5, %6
+  %10 = add nuw nsw i32 %5, 1
+  %11 = icmp slt i32 %10, %1
+  br i1 %11, label %4, label %12
+
+12:                                               ; preds = %8, %4, %2
+  %13 = phi i32 [ %0, %2 ], [ -1, %4 ], [ %9, %8 ]
+  ret i32 %13
+}
+
+define signext i32 @test14d(i31 zeroext %0, i32 signext %1) {
+; CHECK-LABEL: test14d:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    ori $a2, $zero, 2
+; CHECK-NEXT:    blt $a1, $a2, .LBB16_4
+; CHECK-NEXT:  # %bb.1: # %.preheader
+; CHECK-NEXT:    ori $a3, $zero, 1
+; CHECK-NEXT:    addi.w $a2, $zero, -1
+; CHECK-NEXT:    lu32i.d $a2, 0
+; CHECK-NEXT:    ori $a4, $zero, 1000
+; CHECK-NEXT:    .p2align 4, , 16
+; CHECK-NEXT:  .LBB16_2: # =>This Inner Loop Header: Depth=1
+; CHECK-NEXT:    addi.w $a5, $a0, 0
+; CHECK-NEXT:    blt $a4, $a5, .LBB16_5
+; CHECK-NEXT:  # %bb.3: # in Loop: Header=BB16_2 Depth=1
+; CHECK-NEXT:    add.d $a0, $a3, $a0
+; CHECK-NEXT:    addi.w $a3, $a3, 1
+; CHECK-NEXT:    blt $a3, $a1, .LBB16_2
+; CHECK-NEXT:  .LBB16_4:
+; CHECK-NEXT:    addi.w $a0, $a0, 0
+; CHECK-NEXT:    ret
+; CHECK-NEXT:  .LBB16_5:
+; CHECK-NEXT:    addi.w $a0, $a2, 0
+; CHECK-NEXT:    ret
+  %zext = zext i31 %0 to i32
+  %3 = icmp sgt i32 %1, 1
+  br i1 %3, label %4, label %12
+
+4:                                                ; preds = %2, %8
+  %5 = phi i32 [ %10, %8 ], [ 1, %2 ]
+  %6 = phi i32 [ %9, %8 ], [ %zext, %2 ]
+  %7 = icmp sgt i32 %6, 1000
+  br i1 %7, label %12, label %8
+
+8:                                                ; preds = %4
+  %9 = add nsw i32 %5, %6
+  %10 = add nuw nsw i32 %5, 1
+  %11 = icmp slt i32 %10, %1
+  br i1 %11, label %4, label %12
+
+12:                                               ; preds = %8, %4, %2
+  %13 = phi i32 [ %zext, %2 ], [ -1, %4 ], [ %9, %8 ]
+  ret i32 %13
+}
+
+define signext i32 @test15(i64 %arg1, i64 %arg2, i64 %arg3, ptr %arg4)  {
+; CHECK-LABEL: test15:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    addi.d $a2, $a2, -1
+; CHECK-NEXT:    ori $a4, $zero, 256
+; CHECK-NEXT:    .p2align 4, , 16
+; CHECK-NEXT:  .LBB17_1: # %bb2
+; CHECK-NEXT:    # =>This Inner Loop Header: Depth=1
+; CHECK-NEXT:    andi $a0, $a0, 1234
+; CHECK-NEXT:    add.d $a0, $a0, $a1
+; CHECK-NEXT:    addi.d $a2, $a2, 1
+; CHECK-NEXT:    st.w $a0, $a3, 0
+; CHECK-NEXT:    bltu $a2, $a4, .LBB17_1
+; CHECK-NEXT:  # %bb.2: # %bb7
+; CHECK-NEXT:    addi.w $a0, $a0, 0
+; CHECK-NEXT:    ret
+entry:
+  br label %bb2
+
+bb2:                                              ; preds = %bb2, %entry
+  %i1 = phi i64 [ %arg1, %entry ], [ %i5, %bb2 ]
+  %i2 = phi i64 [ %arg3, %entry ], [ %i3, %bb2 ]
+  %i3 = add i64 %i2, 1
+  %i4 = and i64 %i1, 1234
+  %i5 = add i64 %i4, %arg2
+  %i8 = trunc i64 %i5 to i32
+  store i32 %i8, ptr %arg4
+  %i6 = icmp ugt i64 %i2, 255
+  br i1 %i6, label %bb7, label %bb2
+
+bb7:                                              ; preds = %bb2
+  %i7 = trunc i64 %i5 to i32
+  ret i32 %i7
+}
+
+define signext i32 @bug(i32 signext %x) {
+; CHECK-LABEL: bug:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    beqz $a0, .LBB18_2
+; CHECK-NEXT:  # %bb.1: # %if.end
+; CHECK-NEXT:    bstrpick.d $a1, $a0, 31, 16
+; CHECK-NEXT:    sltui $a1, $a1, 1
+; CHECK-NEXT:    slli.d $a2, $a0, 16
+; CHECK-NEXT:    masknez $a0, $a0, $a1
+; CHECK-NEXT:    maskeqz $a2, $a2, $a1
+; CHECK-NEXT:    or $a0, $a2, $a0
+; CHECK-NEXT:    ori $a2, $zero, 32
+; CHECK-NEXT:    masknez $a2, $a2, $a1
+; CHECK-NEXT:    ori $a3, $zero, 16
+; CHECK-NEXT:    maskeqz $a1, $a3, $a1
+; CHECK-NEXT:    or $a1, $a1, $a2
+; CHECK-NEXT:    bstrpick.d $a2, $a0, 31, 24
+; CHECK-NEXT:    sltui $a2, $a2, 1
+; CHECK-NEXT:    slli.d $a3, $a0, 8
+; CHECK-NEXT:    addi.d $a4, $a1, -8
+; CHECK-NEXT:    masknez $a0, $a0, $a2
+; CHECK-NEXT:    maskeqz $a3, $a3, $a2
+; CHECK-NEXT:    or $a0, $a3, $a0
+; CHECK-NEXT:    masknez $a1, $a1, $a2
+; CHECK-NEXT:    maskeqz $a2, $a4, $a2
+; CHECK-NEXT:    or $a1, $a2, $a1
+; CHECK-NEXT:    bstrpick.d $a2, $a0, 31, 28
+; CHECK-NEXT:    sltui $a2, $a2, 1
+; CHECK-NEXT:    slli.d $a3, $a0, 4
+; CHECK-NEXT:    addi.d $a4, $a1, -4
+; CHECK-NEXT:    masknez $a0, $a0, $a2
+; CHECK-NEXT:    maskeqz $a3, $a3, $a2
+; CHECK-NEXT:    or $a0, $a3, $a0
+; CHECK-NEXT:    masknez $a1, $a1, $a2
+; CHECK-NEXT:    maskeqz $a2, $a4, $a2
+; CHECK-NEXT:    or $a1, $a2, $a1
+; CHECK-NEXT:    bstrpick.d $a2, $a0, 31, 30
+; CHECK-NEXT:    sltui $a2, $a2, 1
+; CHECK-NEXT:    slli.d $a3, $a0, 2
+; CHECK-NEXT:    addi.d $a4, $a1, -2
+; CHECK-NEXT:    masknez $a0, $a0, $a2
+; CHECK-NEXT:    maskeqz $a3, $a3, $a2
+; CHECK-NEXT:    or $a0, $a3, $a0
+; CHECK-NEXT:    addi.w $a0, $a0, 0
+; CHECK-NEXT:    masknez $a1, $a1, $a2
+; CHECK-NEXT:    maskeqz $a2, $a4, $a2
+; CHECK-NEXT:    or $a1, $a2, $a1
+; CHECK-NEXT:    srai.d $a0, $a0, 31
+; CHECK-NEXT:    nor $a0, $a0, $zero
+; CHECK-NEXT:    add.d $a0, $a1, $a0
+; CHECK-NEXT:    addi.w $a0, $a0, 0
+; CHECK-NEXT:    ret
+; CHECK-NEXT:  .LBB18_2:
+; CHECK-NEXT:    addi.w $a0, $zero, 0
+; CHECK-NEXT:    ret
+entry:
+  %tobool.not = icmp eq i32 %x, 0
+  br i1 %tobool.not, label %cleanup, label %if.end
+
+if.end:                                           ; preds = %entry
+  %tobool1.not = icmp ult i32 %x, 65536
+  %shl = shl i32 %x, 16
+  %spec.select = select i1 %tobool1.not, i32 %shl, i32 %x
+  %spec.select43 = select i1 %tobool1.not, i32 16, i32 32
+  %tobool5.not = icmp ult i32 %spec.select, 16777216
+  %shl7 = shl i32 %spec.select, 8
+  %sub8 = add nsw i32 %spec.select43, -8
+  %x.addr.1 = select i1 %tobool5.not, i32 %shl7, i32 %spec.select
+  %r.1 = select i1 %tobool5.not, i32 %sub8, i32 %spec.select43
+  %tobool11.not = icmp ult i32 %x.addr.1, 268435456
+  %shl13 = shl i32 %x.addr.1, 4
+  %sub14 = add nsw i32 %r.1, -4
+  %x.addr.2 = select i1 %tobool11.not, i32 %shl13, i32 %x.addr.1
+  %r.2 = select i1 %tobool11.not, i32 %sub14, i32 %r.1
+  %tobool17.not = icmp ult i32 %x.addr.2, 1073741824
+  %shl19 = shl i32 %x.addr.2, 2
+  %sub20 = add nsw i32 %r.2, -2
+  %x.addr.3 = select i1 %tobool17.not, i32 %shl19, i32 %x.addr.2
+  %r.3 = select i1 %tobool17.not, i32 %sub20, i32 %r.2
+  %x.addr.3.lobit = ashr i32 %x.addr.3, 31
+  %x.addr.3.lobit.not = xor i32 %x.addr.3.lobit, -1
+  %r.4 = add nsw i32 %r.3, %x.addr.3.lobit.not
+  br label %cleanup
+
+cleanup:                                          ; preds = %entry, %if.end
+  %retval.0 = phi i32 [ %r.4, %if.end ], [ 0, %entry ]
+  ret i32 %retval.0
+}
+
+define void @test16(i32 signext %arg, i32 signext %arg1) nounwind {
+; CHECK-LABEL: test16:
+; CHECK:       # %bb.0: # %bb
+; CHECK-NEXT:    addi.d $sp, $sp, -32
+; CHECK-NEXT:    st.d $ra, $sp, 24 # 8-byte Folded Spill
+; CHECK-NEXT:    st.d $fp, $sp, 16 # 8-byte Folded Spill
+; CHECK-NEXT:    st.d $s0, $sp, 8 # 8-byte Folded Spill
+; CHECK-NEXT:    move $fp, $a1
+; CHECK-NEXT:    bl %plt(bar)
+; CHECK-NEXT:    move $s0, $a0
+; CHECK-NEXT:    .p2align 4, , 16
+; CHECK-NEXT:  .LBB19_1: # %bb2
+; CHECK-NEXT:    # =>This Inner Loop Header: Depth=1
+; CHECK-NEXT:    addi.w $a0, $s0, 0
+; CHECK-NEXT:    bl %plt(bar)
+; CHECK-NEXT:    sll.w $s0, $s0, $fp
+; CHECK-NEXT:    bnez $a0, .LBB19_1
+; CHECK-NEXT:  # %bb.2: # %bb7
+; CHECK-NEXT:    ld.d $s0, $sp, 8 # 8-byte Folded Reload
+; CHECK-NEXT:    ld.d $fp, $sp, 16 # 8-byte Folded Reload
+; CHECK-NEXT:    ld.d $ra, $sp, 24 # 8-byte Folded Reload
+; CHECK-NEXT:    addi.d $sp, $sp, 32
+; CHECK-NEXT:    ret
+bb:
+  %i = call signext i32 @bar(i32 signext %arg)
+  br label %bb2
+
+bb2:                                              ; preds = %bb2, %bb
+  %i3 = phi i32 [ %i, %bb ], [ %i5, %bb2 ]
+  %i4 = tail call signext i32 @bar(i32 signext %i3)
+  %i5 = shl i32 %i3, %arg1
+  %i6 = icmp eq i32 %i4, 0
+  br i1 %i6, label %bb7, label %bb2
+
+bb7:                                              ; preds = %bb2
+  ret void
+}
+
+define void @test17(i32 signext %arg, i32 signext %arg1) nounwind {
+; CHECK-LABEL: test17:
+; CHECK:       # %bb.0: # %bb
+; CHECK-NEXT:    addi.d $sp, $sp, -32
+; CHECK-NEXT:    st.d $ra, $sp, 24 # 8-byte Folded Spill
+; CHECK-NEXT:    st.d $fp, $sp, 16 # 8-byte Folded Spill
+; CHECK-NEXT:    st.d $s0, $sp, 8 # 8-byte Folded Spill
+; CHECK-NEXT:    move $fp, $a1
+; CHECK-NEXT:    bl %plt(bat)
+; CHECK-NEXT:    move $s0, $a0
+; CHECK-NEXT:    .p2align 4, , 16
+; CHECK-NEXT:  .LBB20_1: # %bb2
+; CHECK-NEXT:    # =>This Inner Loop Header: Depth=1
+; CHECK-NEXT:    addi.w $a0, $s0, 0
+; CHECK-NEXT:    bl %plt(bar)
+; CHECK-NEXT:    sll.w $s0, $s0, $fp
+; CHECK-NEXT:    bnez $a0, .LBB20_1
+; CHECK-NEXT:  # %bb.2: # %bb7
+; CHECK-NEXT:    ld.d $s0, $sp, 8 # 8-byte Folded Reload
+; CHECK-NEXT:    ld.d $fp, $sp, 16 # 8-byte Folded Reload
+; CHECK-NEXT:    ld.d $ra, $sp, 24 # 8-byte Folded Reload
+; CHECK-NEXT:    addi.d $sp, $sp, 32
+; CHECK-NEXT:    ret
+bb:
+  %i = call zeroext i16 @bat(i32 signext %arg)
+  %zext = zext i16 %i to i32
+  br label %bb2
+
+bb2:                                              ; preds = %bb2, %bb
+  %i3 = phi i32 [ %zext, %bb ], [ %i5, %bb2 ]
+  %i4 = tail call signext i32 @bar(i32 signext %i3)
+  %i5 = shl i32 %i3, %arg1
+  %i6 = icmp eq i32 %i4, 0
+  br i1 %i6, label %bb7, label %bb2
+
+bb7:                                              ; preds = %bb2
+  ret void
+}
+declare zeroext i16 @bat(i32 signext)
+
+define signext i32 @sextw_sh2add(i1 zeroext %0, ptr %1, i32 signext %2, i32 signext %3, i32 signext %4) {
+; CHECK-LABEL: sextw_sh2add:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    alsl.d $a2, $a2, $a3, 2
+; CHECK-NEXT:    beqz $a0, .LBB21_2
+; CHECK-NEXT:  # %bb.1:
+; CHECK-NEXT:    st.w $a2, $a1, 0
+; CHECK-NEXT:  .LBB21_2:
+; CHECK-NEXT:    add.w $a0, $a2, $a4
+; CHECK-NEXT:    ret
+  %6 = shl i32 %2, 2
+  %7 = add i32 %6, %3
+  br i1 %0, label %8, label %9
+
+8:                                                ; preds = %5
+  store i32 %7, ptr %1, align 4
+  br label %9
+
+9:                                                ; preds = %5, %8
+  %10 = add i32 %7, %4
+  ret i32 %10
+}
+
+define signext i32 @test19(i64 %arg, i1 zeroext %c1, i1 zeroext %c2, ptr %p) nounwind {
+; CHECK-LABEL: test19:
+; CHECK:       # %bb.0: # %bb
+; CHECK-NEXT:    addi.d $sp, $sp, -16
+; CHECK-NEXT:    st.d $ra, $sp, 8 # 8-byte Folded Spill
+; CHECK-NEXT:    st.d $fp, $sp, 0 # 8-byte Folded Spill
+; CHECK-NEXT:    ori $a0, $zero, 35
+; CHECK-NEXT:    lu32i.d $a0, 1
+; CHECK-NEXT:    maskeqz $fp, $a0, $a1
+; CHECK-NEXT:    st.d $fp, $a3, 0
+; CHECK-NEXT:    beqz $a2, .LBB22_2
+; CHECK-NEXT:  # %bb.1: # %bb2
+; CHECK-NEXT:    move $a0, $zero
+; CHECK-NEXT:    bl %plt(bar)
+; CHECK-NEXT:    move $fp, $a0
+; CHECK-NEXT:  .LBB22_2: # %bb7
+; CHECK-NEXT:    bl %plt(side_effect)
+; CHECK-NEXT:    addi.w $a0, $fp, 0
+; CHECK-NEXT:    ld.d $fp, $sp, 0 # 8-byte Folded Reload
+; CHECK-NEXT:    ld.d $ra, $sp, 8 # 8-byte Folded Reload
+; CHECK-NEXT:    addi.d $sp, $sp, 16
+; CHECK-NEXT:    ret
+bb:
+  %sel = select i1 %c1, i64 4294967331, i64 0
+  store i64 %sel, ptr %p, align 8
+  br i1 %c2, label %bb2, label %bb7
+
+bb2:                                              ; preds = %bb2, %bb
+  %i4 = call signext i32 @bar(i32 0)
+  %i4.sext = sext i32 %i4 to i64
+  br label %bb7
+
+bb7:                                              ; preds = %bb2
+  %phi = phi i64 [ %sel, %bb ], [ %i4.sext, %bb2 ]
+  %trunc = trunc i64 %phi to i32
+  call void @side_effect()
+  ret i32 %trunc
+}
+
+ declare void @side_effect(i64)
diff --git a/llvm/test/CodeGen/M68k/Arith/add-with-overflow.ll b/llvm/test/CodeGen/M68k/Arith/add-with-overflow.ll
index bd5e593edb33..70479b0b3ec6 100644
--- a/llvm/test/CodeGen/M68k/Arith/add-with-overflow.ll
+++ b/llvm/test/CodeGen/M68k/Arith/add-with-overflow.ll
@@ -35,7 +35,7 @@ define fastcc i1 @test6(i32 %v1, i32 %v2, ptr %X) nounwind {
 ; CHECK-NEXT:  ; %bb.1: ; %normal
 ; CHECK-NEXT:    move.l #0, (%a0)
 ; CHECK-NEXT:  .LBB1_2: ; %carry
-; CHECK-NEXT:    move.b #0, %d0
+; CHECK-NEXT:    moveq #0, %d0
 ; CHECK-NEXT:    rts
 entry:
   %t = call {i32, i1} @llvm.uadd.with.overflow.i32(i32 %v1, i32 %v2)
diff --git a/llvm/test/CodeGen/M68k/Arith/add.ll b/llvm/test/CodeGen/M68k/Arith/add.ll
index 281751e3e183..a9eb0bb815b0 100644
--- a/llvm/test/CodeGen/M68k/Arith/add.ll
+++ b/llvm/test/CodeGen/M68k/Arith/add.ll
@@ -43,7 +43,7 @@ define fastcc void @test3(ptr inreg %a) nounwind {
 ; CHECK-NEXT:    suba.l #4, %sp
 ; CHECK-NEXT:    movem.l %d2, (0,%sp) ; 8-byte Folded Spill
 ; CHECK-NEXT:    move.l (%a0), %d0
-; CHECK-NEXT:    move.l #0, %d1
+; CHECK-NEXT:    moveq #0, %d1
 ; CHECK-NEXT:    move.l #-2147483648, %d2
 ; CHECK-NEXT:    add.l (4,%a0), %d2
 ; CHECK-NEXT:    addx.l %d0, %d1
@@ -64,7 +64,7 @@ define fastcc void @test4(ptr inreg %a) nounwind {
 ; CHECK-NEXT:    suba.l #4, %sp
 ; CHECK-NEXT:    movem.l %d2, (0,%sp) ; 8-byte Folded Spill
 ; CHECK-NEXT:    move.l (%a0), %d0
-; CHECK-NEXT:    move.l #0, %d1
+; CHECK-NEXT:    moveq #0, %d1
 ; CHECK-NEXT:    move.l #128, %d2
 ; CHECK-NEXT:    add.l (4,%a0), %d2
 ; CHECK-NEXT:    addx.l %d0, %d1
diff --git a/llvm/test/CodeGen/M68k/Arith/bitwise.ll b/llvm/test/CodeGen/M68k/Arith/bitwise.ll
index 70e4dd42bfb6..74fc543a5fb8 100644
--- a/llvm/test/CodeGen/M68k/Arith/bitwise.ll
+++ b/llvm/test/CodeGen/M68k/Arith/bitwise.ll
@@ -242,7 +242,7 @@ define i64 @lshr64(i64 %a, i64 %b) nounwind {
 ; CHECK-NEXT:    add.l #-32, %d1
 ; CHECK-NEXT:    bmi .LBB18_1
 ; CHECK-NEXT:  ; %bb.2:
-; CHECK-NEXT:    move.l #0, %d0
+; CHECK-NEXT:    moveq #0, %d0
 ; CHECK-NEXT:    bra .LBB18_3
 ; CHECK-NEXT:  .LBB18_1:
 ; CHECK-NEXT:    move.l %d2, %d0
@@ -301,7 +301,7 @@ define i64 @ashr64(i64 %a, i64 %b) nounwind {
 ; CHECK-NEXT:    add.l #-32, %d3
 ; CHECK-NEXT:    bmi .LBB19_5
 ; CHECK-NEXT:  ; %bb.4:
-; CHECK-NEXT:    move.l #31, %d2
+; CHECK-NEXT:    moveq #31, %d2
 ; CHECK-NEXT:  .LBB19_5:
 ; CHECK-NEXT:    asr.l %d2, %d0
 ; CHECK-NEXT:    movem.l (0,%sp), %d2-%d3 ; 12-byte Folded Reload
@@ -322,7 +322,7 @@ define i64 @shl64(i64 %a, i64 %b) nounwind {
 ; CHECK-NEXT:    add.l #-32, %d0
 ; CHECK-NEXT:    bmi .LBB20_1
 ; CHECK-NEXT:  ; %bb.2:
-; CHECK-NEXT:    move.l #0, %d1
+; CHECK-NEXT:    moveq #0, %d1
 ; CHECK-NEXT:    bra .LBB20_3
 ; CHECK-NEXT:  .LBB20_1:
 ; CHECK-NEXT:    move.l %d2, %d1
diff --git a/llvm/test/CodeGen/M68k/Arith/divide-by-constant.ll b/llvm/test/CodeGen/M68k/Arith/divide-by-constant.ll
index 834dfe1c26f0..fcc8dd3e7662 100644
--- a/llvm/test/CodeGen/M68k/Arith/divide-by-constant.ll
+++ b/llvm/test/CodeGen/M68k/Arith/divide-by-constant.ll
@@ -40,7 +40,7 @@ define zeroext i8 @test3(i8 zeroext %x, i8 zeroext %c) {
 ; CHECK-NEXT:    move.b (11,%sp), %d0
 ; CHECK-NEXT:    and.l #255, %d0
 ; CHECK-NEXT:    muls #171, %d0
-; CHECK-NEXT:    move.w #9, %d1
+; CHECK-NEXT:    moveq #9, %d1
 ; CHECK-NEXT:    lsr.w %d1, %d0
 ; CHECK-NEXT:    and.l #65535, %d0
 ; CHECK-NEXT:    rts
@@ -58,7 +58,7 @@ define signext i16 @test4(i16 signext %x) nounwind {
 ; CHECK-NEXT:    muls #1986, %d0
 ; CHECK-NEXT:    asr.l #8, %d0
 ; CHECK-NEXT:    asr.l #8, %d0
-; CHECK-NEXT:    move.w #15, %d1
+; CHECK-NEXT:    moveq #15, %d1
 ; CHECK-NEXT:    move.w %d0, %d2
 ; CHECK-NEXT:    lsr.w %d1, %d2
 ; CHECK-NEXT:    add.w %d2, %d0
@@ -94,7 +94,7 @@ define signext i16 @test6(i16 signext %x) nounwind {
 ; CHECK-NEXT:    muls #26215, %d0
 ; CHECK-NEXT:    asr.l #8, %d0
 ; CHECK-NEXT:    asr.l #8, %d0
-; CHECK-NEXT:    move.w #15, %d1
+; CHECK-NEXT:    moveq #15, %d1
 ; CHECK-NEXT:    move.w %d0, %d2
 ; CHECK-NEXT:    lsr.w %d1, %d2
 ; CHECK-NEXT:    asr.w #2, %d0
@@ -128,7 +128,7 @@ define i8 @test8(i8 %x) nounwind {
 ; CHECK-NEXT:    lsr.b #1, %d0
 ; CHECK-NEXT:    and.l #255, %d0
 ; CHECK-NEXT:    muls #211, %d0
-; CHECK-NEXT:    move.w #13, %d1
+; CHECK-NEXT:    moveq #13, %d1
 ; CHECK-NEXT:    lsr.w %d1, %d0
 ; CHECK-NEXT:    ; kill: def $bd0 killed $bd0 killed $d0
 ; CHECK-NEXT:    rts
@@ -143,7 +143,7 @@ define i8 @test9(i8 %x) nounwind {
 ; CHECK-NEXT:    lsr.b #2, %d0
 ; CHECK-NEXT:    and.l #255, %d0
 ; CHECK-NEXT:    muls #71, %d0
-; CHECK-NEXT:    move.w #11, %d1
+; CHECK-NEXT:    moveq #11, %d1
 ; CHECK-NEXT:    lsr.w %d1, %d0
 ; CHECK-NEXT:    ; kill: def $bd0 killed $bd0 killed $d0
 ; CHECK-NEXT:    rts
@@ -156,11 +156,11 @@ define i32 @testsize1(i32 %x) minsize nounwind {
 ; CHECK:       ; %bb.0: ; %entry
 ; CHECK-NEXT:    suba.l #4, %sp
 ; CHECK-NEXT:    movem.l %d2, (0,%sp) ; 8-byte Folded Spill
-; CHECK-NEXT:    move.l #31, %d1
+; CHECK-NEXT:    moveq #31, %d1
 ; CHECK-NEXT:    move.l (8,%sp), %d0
 ; CHECK-NEXT:    move.l %d0, %d2
 ; CHECK-NEXT:    asr.l %d1, %d2
-; CHECK-NEXT:    move.l #27, %d1
+; CHECK-NEXT:    moveq #27, %d1
 ; CHECK-NEXT:    lsr.l %d1, %d2
 ; CHECK-NEXT:    add.l %d2, %d0
 ; CHECK-NEXT:    asr.l #5, %d0
diff --git a/llvm/test/CodeGen/M68k/Arith/imul.ll b/llvm/test/CodeGen/M68k/Arith/imul.ll
index f53568395c29..a1846e4d51bd 100644
--- a/llvm/test/CodeGen/M68k/Arith/imul.ll
+++ b/llvm/test/CodeGen/M68k/Arith/imul.ll
@@ -19,7 +19,7 @@ define i64 @mul4_64(i64 %A) {
 ; CHECK-NEXT:    suba.l #4, %sp
 ; CHECK-NEXT:    .cfi_def_cfa_offset -8
 ; CHECK-NEXT:    movem.l %d2, (0,%sp) ; 8-byte Folded Spill
-; CHECK-NEXT:    move.l #30, %d0
+; CHECK-NEXT:    moveq #30, %d0
 ; CHECK-NEXT:    move.l (12,%sp), %d1
 ; CHECK-NEXT:    move.l %d1, %d2
 ; CHECK-NEXT:    lsr.l %d0, %d2
@@ -38,7 +38,7 @@ define i32 @mul4096_32(i32 %A) {
 ; CHECK-LABEL: mul4096_32:
 ; CHECK:         .cfi_startproc
 ; CHECK-NEXT:  ; %bb.0:
-; CHECK-NEXT:    move.l #12, %d1
+; CHECK-NEXT:    moveq #12, %d1
 ; CHECK-NEXT:    move.l (4,%sp), %d0
 ; CHECK-NEXT:    lsl.l %d1, %d0
 ; CHECK-NEXT:    rts
@@ -53,11 +53,11 @@ define i64 @mul4096_64(i64 %A) {
 ; CHECK-NEXT:    suba.l #8, %sp
 ; CHECK-NEXT:    .cfi_def_cfa_offset -12
 ; CHECK-NEXT:    movem.l %d2-%d3, (0,%sp) ; 12-byte Folded Spill
-; CHECK-NEXT:    move.l #20, %d0
+; CHECK-NEXT:    moveq #20, %d0
 ; CHECK-NEXT:    move.l (16,%sp), %d1
 ; CHECK-NEXT:    move.l %d1, %d2
 ; CHECK-NEXT:    lsr.l %d0, %d2
-; CHECK-NEXT:    move.l #12, %d3
+; CHECK-NEXT:    moveq #12, %d3
 ; CHECK-NEXT:    move.l (12,%sp), %d0
 ; CHECK-NEXT:    lsl.l %d3, %d0
 ; CHECK-NEXT:    or.l %d2, %d0
@@ -73,7 +73,7 @@ define i32 @mulmin4096_32(i32 %A) {
 ; CHECK-LABEL: mulmin4096_32:
 ; CHECK:         .cfi_startproc
 ; CHECK-NEXT:  ; %bb.0:
-; CHECK-NEXT:    move.l #12, %d1
+; CHECK-NEXT:    moveq #12, %d1
 ; CHECK-NEXT:    move.l (4,%sp), %d0
 ; CHECK-NEXT:    lsl.l %d1, %d0
 ; CHECK-NEXT:    neg.l %d0
@@ -89,11 +89,11 @@ define i64 @mulmin4096_64(i64 %A) {
 ; CHECK-NEXT:    suba.l #8, %sp
 ; CHECK-NEXT:    .cfi_def_cfa_offset -12
 ; CHECK-NEXT:    movem.l %d2-%d3, (0,%sp) ; 12-byte Folded Spill
-; CHECK-NEXT:    move.l #20, %d0
+; CHECK-NEXT:    moveq #20, %d0
 ; CHECK-NEXT:    move.l (16,%sp), %d1
 ; CHECK-NEXT:    move.l %d1, %d2
 ; CHECK-NEXT:    lsr.l %d0, %d2
-; CHECK-NEXT:    move.l #12, %d3
+; CHECK-NEXT:    moveq #12, %d3
 ; CHECK-NEXT:    move.l (12,%sp), %d0
 ; CHECK-NEXT:    lsl.l %d3, %d0
 ; CHECK-NEXT:    or.l %d2, %d0
@@ -258,7 +258,7 @@ define i32 @mul0_32(i32 %A) {
 ; CHECK-LABEL: mul0_32:
 ; CHECK:         .cfi_startproc
 ; CHECK-NEXT:  ; %bb.0:
-; CHECK-NEXT:    move.l #0, %d0
+; CHECK-NEXT:    moveq #0, %d0
 ; CHECK-NEXT:    rts
     %mul = mul i32 %A, 0
     ret i32 %mul
diff --git a/llvm/test/CodeGen/M68k/Arith/smul-with-overflow.ll b/llvm/test/CodeGen/M68k/Arith/smul-with-overflow.ll
index 5bd4d5d48bc8..10a797f13441 100644
--- a/llvm/test/CodeGen/M68k/Arith/smul-with-overflow.ll
+++ b/llvm/test/CodeGen/M68k/Arith/smul-with-overflow.ll
@@ -24,7 +24,7 @@ entry:
 define zeroext i8 @smul_i8_no_ovf(i8 signext %a, i8 signext %b) nounwind ssp {
 ; CHECK-LABEL: smul_i8_no_ovf:
 ; CHECK:       ; %bb.0: ; %entry
-; CHECK-NEXT:    move.l #42, %d0
+; CHECK-NEXT:    moveq #42, %d0
 ; CHECK-NEXT:    rts
 entry:
   %smul = tail call { i8, i1 } @llvm.smul.with.overflow.i8(i8 %a, i8 %b)
@@ -70,7 +70,7 @@ define fastcc i1 @test1(i32 %v1, i32 %v2) nounwind {
 ; CHECK-NEXT:    lea (no,%pc), %a0
 ; CHECK-NEXT:    move.l %a0, (%sp)
 ; CHECK-NEXT:    jsr printf@PLT
-; CHECK-NEXT:    move.b #0, %d0
+; CHECK-NEXT:    moveq #0, %d0
 ; CHECK-NEXT:    adda.l #12, %sp
 ; CHECK-NEXT:    rts
 ; CHECK-NEXT:  .LBB3_1: ; %normal
@@ -78,7 +78,7 @@ define fastcc i1 @test1(i32 %v1, i32 %v2) nounwind {
 ; CHECK-NEXT:    lea (ok,%pc), %a0
 ; CHECK-NEXT:    move.l %a0, (%sp)
 ; CHECK-NEXT:    jsr printf@PLT
-; CHECK-NEXT:    move.b #1, %d0
+; CHECK-NEXT:    moveq #1, %d0
 ; CHECK-NEXT:    adda.l #12, %sp
 ; CHECK-NEXT:    rts
 entry:
@@ -108,7 +108,7 @@ define fastcc i1 @test2(i32 %v1, i32 %v2) nounwind {
 ; CHECK-NEXT:    lea (no,%pc), %a0
 ; CHECK-NEXT:    move.l %a0, (%sp)
 ; CHECK-NEXT:    jsr printf@PLT
-; CHECK-NEXT:    move.b #0, %d0
+; CHECK-NEXT:    moveq #0, %d0
 ; CHECK-NEXT:    adda.l #12, %sp
 ; CHECK-NEXT:    rts
 ; CHECK-NEXT:  .LBB4_2: ; %normal
@@ -116,7 +116,7 @@ define fastcc i1 @test2(i32 %v1, i32 %v2) nounwind {
 ; CHECK-NEXT:    lea (ok,%pc), %a0
 ; CHECK-NEXT:    move.l %a0, (%sp)
 ; CHECK-NEXT:    jsr printf@PLT
-; CHECK-NEXT:    move.b #1, %d0
+; CHECK-NEXT:    moveq #1, %d0
 ; CHECK-NEXT:    adda.l #12, %sp
 ; CHECK-NEXT:    rts
 entry:
@@ -155,7 +155,7 @@ define i32 @test4(i32 %a, i32 %b) nounwind readnone {
 ; CHECK:       ; %bb.0: ; %entry
 ; CHECK-NEXT:    move.l (8,%sp), %d0
 ; CHECK-NEXT:    add.l (4,%sp), %d0
-; CHECK-NEXT:    move.l #4, %d1
+; CHECK-NEXT:    moveq #4, %d1
 ; CHECK-NEXT:    muls.l %d1, %d0
 ; CHECK-NEXT:    rts
 entry:
diff --git a/llvm/test/CodeGen/M68k/Arith/sub-with-overflow.ll b/llvm/test/CodeGen/M68k/Arith/sub-with-overflow.ll
index 8d47c7ebf7e5..be3223156986 100644
--- a/llvm/test/CodeGen/M68k/Arith/sub-with-overflow.ll
+++ b/llvm/test/CodeGen/M68k/Arith/sub-with-overflow.ll
@@ -19,7 +19,7 @@ define i1 @func1(i32 %v1, i32 %v2) nounwind {
 ; CHECK-NEXT:    lea (no,%pc), %a0
 ; CHECK-NEXT:    move.l %a0, (%sp)
 ; CHECK-NEXT:    jsr printf@PLT
-; CHECK-NEXT:    move.b #0, %d0
+; CHECK-NEXT:    moveq #0, %d0
 ; CHECK-NEXT:    adda.l #12, %sp
 ; CHECK-NEXT:    rts
 ; CHECK-NEXT:  .LBB0_1: ; %normal
@@ -27,7 +27,7 @@ define i1 @func1(i32 %v1, i32 %v2) nounwind {
 ; CHECK-NEXT:    lea (ok,%pc), %a0
 ; CHECK-NEXT:    move.l %a0, (%sp)
 ; CHECK-NEXT:    jsr printf@PLT
-; CHECK-NEXT:    move.b #1, %d0
+; CHECK-NEXT:    moveq #1, %d0
 ; CHECK-NEXT:    adda.l #12, %sp
 ; CHECK-NEXT:    rts
 entry:
@@ -56,7 +56,7 @@ define i1 @func2(i32 %v1, i32 %v2) nounwind {
 ; CHECK-NEXT:    lea (no,%pc), %a0
 ; CHECK-NEXT:    move.l %a0, (%sp)
 ; CHECK-NEXT:    jsr printf@PLT
-; CHECK-NEXT:    move.b #0, %d0
+; CHECK-NEXT:    moveq #0, %d0
 ; CHECK-NEXT:    adda.l #12, %sp
 ; CHECK-NEXT:    rts
 ; CHECK-NEXT:  .LBB1_1: ; %normal
@@ -64,7 +64,7 @@ define i1 @func2(i32 %v1, i32 %v2) nounwind {
 ; CHECK-NEXT:    lea (ok,%pc), %a0
 ; CHECK-NEXT:    move.l %a0, (%sp)
 ; CHECK-NEXT:    jsr printf@PLT
-; CHECK-NEXT:    move.b #1, %d0
+; CHECK-NEXT:    moveq #1, %d0
 ; CHECK-NEXT:    adda.l #12, %sp
 ; CHECK-NEXT:    rts
 entry:
@@ -85,7 +85,7 @@ carry:
 define i1 @func3(i32 %x) nounwind {
 ; CHECK-LABEL: func3:
 ; CHECK:       ; %bb.0: ; %entry
-; CHECK-NEXT:    move.l #-1, %d0
+; CHECK-NEXT:    moveq #-1, %d0
 ; CHECK-NEXT:    add.l (4,%sp), %d0
 ; CHECK-NEXT:    svs %d0
 ; CHECK-NEXT:    rts
diff --git a/llvm/test/CodeGen/M68k/Arith/sub.ll b/llvm/test/CodeGen/M68k/Arith/sub.ll
index fff3601000df..16d0498b3dbb 100644
--- a/llvm/test/CodeGen/M68k/Arith/sub.ll
+++ b/llvm/test/CodeGen/M68k/Arith/sub.ll
@@ -7,7 +7,7 @@ define i32 @test1(i32 %x) {
 ; CHECK-NEXT:  ; %bb.0:
 ; CHECK-NEXT:    move.l (4,%sp), %d1
 ; CHECK-NEXT:    eori.l #31, %d1
-; CHECK-NEXT:    move.l #32, %d0
+; CHECK-NEXT:    moveq #32, %d0
 ; CHECK-NEXT:    sub.l %d1, %d0
 ; CHECK-NEXT:    rts
   %xor = xor i32 %x, 31
diff --git a/llvm/test/CodeGen/M68k/Arith/umul-with-overflow.ll b/llvm/test/CodeGen/M68k/Arith/umul-with-overflow.ll
index fd128a3e52bd..3314e65399c4 100644
--- a/llvm/test/CodeGen/M68k/Arith/umul-with-overflow.ll
+++ b/llvm/test/CodeGen/M68k/Arith/umul-with-overflow.ll
@@ -24,7 +24,7 @@ entry:
 define zeroext i8 @umul_i8_no_ovf(i8 signext %a, i8 signext %b) nounwind ssp {
 ; CHECK-LABEL: umul_i8_no_ovf:
 ; CHECK:       ; %bb.0: ; %entry
-; CHECK-NEXT:    move.l #42, %d0
+; CHECK-NEXT:    moveq #42, %d0
 ; CHECK-NEXT:    rts
 entry:
   %umul = tail call { i8, i1 } @llvm.umul.with.overflow.i8(i8 %a, i8 %b)
@@ -59,7 +59,7 @@ declare {i32, i1} @llvm.umul.with.overflow.i32(i32 %a, i32 %b)
 define i1 @a(i32 %x)  nounwind {
 ; CHECK-LABEL: a:
 ; CHECK:       ; %bb.0:
-; CHECK-NEXT:    move.l #3, %d0
+; CHECK-NEXT:    moveq #3, %d0
 ; CHECK-NEXT:    move.l (4,%sp), %d1
 ; CHECK-NEXT:    mulu.l %d0, %d1
 ; CHECK-NEXT:    svs %d0
@@ -90,7 +90,7 @@ define i32 @test3(i32 %a, i32 %b) nounwind readnone {
 ; CHECK:       ; %bb.0: ; %entry
 ; CHECK-NEXT:    move.l (8,%sp), %d0
 ; CHECK-NEXT:    add.l (4,%sp), %d0
-; CHECK-NEXT:    move.l #4, %d1
+; CHECK-NEXT:    moveq #4, %d1
 ; CHECK-NEXT:    mulu.l %d1, %d0
 ; CHECK-NEXT:    rts
 entry:
diff --git a/llvm/test/CodeGen/M68k/CConv/c-call.ll b/llvm/test/CodeGen/M68k/CConv/c-call.ll
index a9638eec6a31..badd4e31f37d 100644
--- a/llvm/test/CodeGen/M68k/CConv/c-call.ll
+++ b/llvm/test/CodeGen/M68k/CConv/c-call.ll
@@ -14,7 +14,7 @@ define i32 @test1() nounwind {
 ; CHECK-NEXT:    move.l #2, (4,%sp)
 ; CHECK-NEXT:    move.l #1, (%sp)
 ; CHECK-NEXT:    jsr (test1_callee@PLT,%pc)
-; CHECK-NEXT:    move.l #0, %d0
+; CHECK-NEXT:    moveq #0, %d0
 ; CHECK-NEXT:    adda.l #20, %sp
 ; CHECK-NEXT:    rts
 entry:
@@ -34,7 +34,7 @@ define i16 @test2() nounwind {
 ; CHECK-NEXT:    move.l #2, (4,%sp)
 ; CHECK-NEXT:    move.l #1, (%sp)
 ; CHECK-NEXT:    jsr (test2_callee@PLT,%pc)
-; CHECK-NEXT:    move.w #0, %d0
+; CHECK-NEXT:    moveq #0, %d0
 ; CHECK-NEXT:    adda.l #20, %sp
 ; CHECK-NEXT:    rts
 entry:
@@ -54,7 +54,7 @@ define i8 @test3() nounwind {
 ; CHECK-NEXT:    move.l #2, (4,%sp)
 ; CHECK-NEXT:    move.l #1, (%sp)
 ; CHECK-NEXT:    jsr (test3_callee@PLT,%pc)
-; CHECK-NEXT:    move.b #0, %d0
+; CHECK-NEXT:    moveq #0, %d0
 ; CHECK-NEXT:    adda.l #20, %sp
 ; CHECK-NEXT:    rts
 entry:
diff --git a/llvm/test/CodeGen/M68k/CConv/fastcc-call.ll b/llvm/test/CodeGen/M68k/CConv/fastcc-call.ll
index 4b0f8ed254a5..8d40ebd5228f 100644
--- a/llvm/test/CodeGen/M68k/CConv/fastcc-call.ll
+++ b/llvm/test/CodeGen/M68k/CConv/fastcc-call.ll
@@ -11,12 +11,12 @@ define i32 @foo1() nounwind uwtable {
 ; CHECK-NEXT:    suba.l #4, %sp
 ; CHECK-NEXT:    .cfi_def_cfa_offset -8
 ; CHECK-NEXT:    move.l #5, (%sp)
-; CHECK-NEXT:    move.l #1, %d0
-; CHECK-NEXT:    move.l #2, %d1
+; CHECK-NEXT:    moveq #1, %d0
+; CHECK-NEXT:    moveq #2, %d1
 ; CHECK-NEXT:    move.l #3, %a0
 ; CHECK-NEXT:    move.l #4, %a1
 ; CHECK-NEXT:    jsr (bar1@PLT,%pc)
-; CHECK-NEXT:    move.l #0, %d0
+; CHECK-NEXT:    moveq #0, %d0
 ; CHECK-NEXT:    adda.l #4, %sp
 ; CHECK-NEXT:    rts
 entry:
@@ -34,11 +34,11 @@ define i32 @foo2() nounwind uwtable {
 ; CHECK-NEXT:    suba.l #12, %sp
 ; CHECK-NEXT:    .cfi_def_cfa_offset -16
 ; CHECK-NEXT:    lea (8,%sp), %a0
-; CHECK-NEXT:    move.l #2, %d0
+; CHECK-NEXT:    moveq #2, %d0
 ; CHECK-NEXT:    lea (4,%sp), %a1
-; CHECK-NEXT:    move.l #4, %d1
+; CHECK-NEXT:    moveq #4, %d1
 ; CHECK-NEXT:    jsr (bar2@PLT,%pc)
-; CHECK-NEXT:    move.l #0, %d0
+; CHECK-NEXT:    moveq #0, %d0
 ; CHECK-NEXT:    adda.l #12, %sp
 ; CHECK-NEXT:    rts
 entry:
diff --git a/llvm/test/CodeGen/M68k/CodeModel/medium-pie-global-access.ll b/llvm/test/CodeGen/M68k/CodeModel/medium-pie-global-access.ll
index ce8f2d0a6ba7..3d398afe7dc4 100644
--- a/llvm/test/CodeGen/M68k/CodeModel/medium-pie-global-access.ll
+++ b/llvm/test/CodeGen/M68k/CodeModel/medium-pie-global-access.ll
@@ -71,7 +71,7 @@ define i32 @my_access_global_store_d() #0 {
 ; CHECK-NEXT:  ; %bb.0: ; %entry
 ; CHECK-NEXT:    move.l (d@GOTPCREL,%pc), %a0
 ; CHECK-NEXT:    move.l #2, (%a0)
-; CHECK-NEXT:    move.l #0, %d0
+; CHECK-NEXT:    moveq #0, %d0
 ; CHECK-NEXT:    rts
 entry:
  store i32 2, ptr @d, align 4
@@ -105,7 +105,7 @@ define linkonce_odr i32 @bar() comdat {
 ; CHECK-LABEL: bar:
 ; CHECK:         .cfi_startproc
 ; CHECK-NEXT:  ; %bb.0: ; %entry
-; CHECK-NEXT:    move.l #0, %d0
+; CHECK-NEXT:    moveq #0, %d0
 ; CHECK-NEXT:    rts
 entry:
  ret i32 0
diff --git a/llvm/test/CodeGen/M68k/CodeModel/small-pie-global-access.ll b/llvm/test/CodeGen/M68k/CodeModel/small-pie-global-access.ll
index 668f8a96ac6f..030f72bb3753 100644
--- a/llvm/test/CodeGen/M68k/CodeModel/small-pie-global-access.ll
+++ b/llvm/test/CodeGen/M68k/CodeModel/small-pie-global-access.ll
@@ -69,7 +69,7 @@ define i32 @my_access_global_store_d() #0 {
 ; CHECK-NEXT:  ; %bb.0: ; %entry
 ; CHECK-NEXT:    move.l (d@GOTPCREL,%pc), %a0
 ; CHECK-NEXT:    move.l #2, (%a0)
-; CHECK-NEXT:    move.l #0, %d0
+; CHECK-NEXT:    moveq #0, %d0
 ; CHECK-NEXT:    rts
 entry:
  store i32 2, ptr @d, align 4
@@ -103,7 +103,7 @@ define linkonce_odr i32 @bar() comdat {
 ; CHECK-LABEL: bar:
 ; CHECK:         .cfi_startproc
 ; CHECK-NEXT:  ; %bb.0: ; %entry
-; CHECK-NEXT:    move.l #0, %d0
+; CHECK-NEXT:    moveq #0, %d0
 ; CHECK-NEXT:    rts
 entry:
  ret i32 0
diff --git a/llvm/test/CodeGen/M68k/Control/cmp.ll b/llvm/test/CodeGen/M68k/Control/cmp.ll
index 634c08760a4e..d3a8bbb0b0c8 100644
--- a/llvm/test/CodeGen/M68k/Control/cmp.ll
+++ b/llvm/test/CodeGen/M68k/Control/cmp.ll
@@ -8,10 +8,10 @@ define i32 @test1(ptr %y) nounwind {
 ; CHECK-NEXT:    cmpi.l #0, (%a0)
 ; CHECK-NEXT:    beq .LBB0_2
 ; CHECK-NEXT:  ; %bb.1: ; %cond_false
-; CHECK-NEXT:    move.l #0, %d0
+; CHECK-NEXT:    moveq #0, %d0
 ; CHECK-NEXT:    rts
 ; CHECK-NEXT:  .LBB0_2: ; %cond_true
-; CHECK-NEXT:    move.l #1, %d0
+; CHECK-NEXT:    moveq #1, %d0
 ; CHECK-NEXT:    rts
  %tmp = load i32, ptr %y  ; <i32> [#uses=1]
  %tmp.upgrd.1 = icmp eq i32 %tmp, 0  ; <i1> [#uses=1]
@@ -33,10 +33,10 @@ define i32 @test2(ptr %y) nounwind {
 ; CHECK-NEXT:    cmpi.l #0, %d0
 ; CHECK-NEXT:    beq .LBB1_2
 ; CHECK-NEXT:  ; %bb.1: ; %cond_false
-; CHECK-NEXT:    move.l #0, %d0
+; CHECK-NEXT:    moveq #0, %d0
 ; CHECK-NEXT:    rts
 ; CHECK-NEXT:  .LBB1_2: ; %cond_true
-; CHECK-NEXT:    move.l #1, %d0
+; CHECK-NEXT:    moveq #1, %d0
 ; CHECK-NEXT:    rts
  %tmp = load i32, ptr %y  ; <i32> [#uses=1]
  %tmp1 = shl i32 %tmp, 3  ; <i32> [#uses=1]
@@ -59,10 +59,10 @@ define i8 @test2b(ptr %y) nounwind {
 ; CHECK-NEXT:    cmpi.b #0, %d0
 ; CHECK-NEXT:    beq .LBB2_2
 ; CHECK-NEXT:  ; %bb.1: ; %cond_false
-; CHECK-NEXT:    move.b #0, %d0
+; CHECK-NEXT:    moveq #0, %d0
 ; CHECK-NEXT:    rts
 ; CHECK-NEXT:  .LBB2_2: ; %cond_true
-; CHECK-NEXT:    move.b #1, %d0
+; CHECK-NEXT:    moveq #1, %d0
 ; CHECK-NEXT:    rts
  %tmp = load i8, ptr %y  ; <i8> [#uses=1]
  %tmp1 = shl i8 %tmp, 3  ; <i8> [#uses=1]
@@ -84,7 +84,7 @@ define i64 @test3(i64 %x) nounwind {
 ; CHECK-NEXT:    seq %d0
 ; CHECK-NEXT:    move.l %d0, %d1
 ; CHECK-NEXT:    and.l #255, %d1
-; CHECK-NEXT:    move.l #0, %d0
+; CHECK-NEXT:    moveq #0, %d0
 ; CHECK-NEXT:    rts
   %t = icmp eq i64 %x, 0
   %r = zext i1 %t to i64
@@ -97,7 +97,7 @@ define i64 @test4(i64 %x) nounwind {
 ; CHECK-NEXT:    suba.l #4, %sp
 ; CHECK-NEXT:    movem.l %d2, (0,%sp) ; 8-byte Folded Spill
 ; CHECK-NEXT:    move.l (8,%sp), %d1
-; CHECK-NEXT:    move.l #0, %d0
+; CHECK-NEXT:    moveq #0, %d0
 ; CHECK-NEXT:    move.l (12,%sp), %d2
 ; CHECK-NEXT:    sub.l #1, %d2
 ; CHECK-NEXT:    subx.l %d0, %d1
@@ -119,11 +119,11 @@ define i32 @test6() nounwind align 2 {
 ; CHECK-NEXT:    or.l (8,%sp), %d0
 ; CHECK-NEXT:    beq .LBB5_1
 ; CHECK-NEXT:  ; %bb.2: ; %F
-; CHECK-NEXT:    move.l #0, %d0
+; CHECK-NEXT:    moveq #0, %d0
 ; CHECK-NEXT:    adda.l #20, %sp
 ; CHECK-NEXT:    rts
 ; CHECK-NEXT:  .LBB5_1: ; %T
-; CHECK-NEXT:    move.l #1, %d0
+; CHECK-NEXT:    moveq #1, %d0
 ; CHECK-NEXT:    adda.l #20, %sp
 ; CHECK-NEXT:    rts
   %A = alloca {i64, i64}, align 8
@@ -229,7 +229,7 @@ define zeroext i1 @test15(i32 %bf.load, i32 %n) {
 ; CHECK-LABEL: test15:
 ; CHECK:         .cfi_startproc
 ; CHECK-NEXT:  ; %bb.0:
-; CHECK-NEXT:    move.l #16, %d0
+; CHECK-NEXT:    moveq #16, %d0
 ; CHECK-NEXT:    move.l (4,%sp), %d1
 ; CHECK-NEXT:    lsr.l %d0, %d1
 ; CHECK-NEXT:    move.l %d1, %d0
@@ -252,7 +252,7 @@ define i8 @test16(i16 signext %L) {
 ; CHECK-LABEL: test16:
 ; CHECK:         .cfi_startproc
 ; CHECK-NEXT:  ; %bb.0:
-; CHECK-NEXT:    move.w #15, %d1
+; CHECK-NEXT:    moveq #15, %d1
 ; CHECK-NEXT:    move.w (6,%sp), %d0
 ; CHECK-NEXT:    lsr.w %d1, %d0
 ; CHECK-NEXT:    eori.b #1, %d0
@@ -268,7 +268,7 @@ define i8 @test18(i64 %L) {
 ; CHECK-LABEL: test18:
 ; CHECK:         .cfi_startproc
 ; CHECK-NEXT:  ; %bb.0:
-; CHECK-NEXT:    move.l #31, %d1
+; CHECK-NEXT:    moveq #31, %d1
 ; CHECK-NEXT:    move.l (4,%sp), %d0
 ; CHECK-NEXT:    lsr.l %d1, %d0
 ; CHECK-NEXT:    eori.b #1, %d0
diff --git a/llvm/test/CodeGen/M68k/Control/long-setcc.ll b/llvm/test/CodeGen/M68k/Control/long-setcc.ll
index b089af5f2ae8..45a617599c1e 100644
--- a/llvm/test/CodeGen/M68k/Control/long-setcc.ll
+++ b/llvm/test/CodeGen/M68k/Control/long-setcc.ll
@@ -4,7 +4,7 @@
 define i1 @t1(i64 %x) nounwind {
 ; CHECK-LABEL: t1:
 ; CHECK:       ; %bb.0:
-; CHECK-NEXT:    move.l #31, %d1
+; CHECK-NEXT:    moveq #31, %d1
 ; CHECK-NEXT:    move.l (4,%sp), %d0
 ; CHECK-NEXT:    lsr.l %d1, %d0
 ; CHECK-NEXT:    ; kill: def $bd0 killed $bd0 killed $d0
@@ -26,7 +26,7 @@ define i1 @t2(i64 %x) nounwind {
 define i1 @t3(i32 %x) nounwind {
 ; CHECK-LABEL: t3:
 ; CHECK:       ; %bb.0:
-; CHECK-NEXT:    move.b #0, %d0
+; CHECK-NEXT:    moveq #0, %d0
 ; CHECK-NEXT:    rts
   %tmp = icmp ugt i32 %x, -1
   ret i1 %tmp
diff --git a/llvm/test/CodeGen/M68k/Control/setcc.ll b/llvm/test/CodeGen/M68k/Control/setcc.ll
index 63856e278c9e..9e03f9b90842 100644
--- a/llvm/test/CodeGen/M68k/Control/setcc.ll
+++ b/llvm/test/CodeGen/M68k/Control/setcc.ll
@@ -40,7 +40,7 @@ define fastcc i64 @t3(i64 %x) nounwind readnone ssp {
 ; CHECK:       ; %bb.0: ; %entry
 ; CHECK-NEXT:    suba.l #4, %sp
 ; CHECK-NEXT:    movem.l %d2, (0,%sp) ; 8-byte Folded Spill
-; CHECK-NEXT:    move.l #0, %d2
+; CHECK-NEXT:    moveq #0, %d2
 ; CHECK-NEXT:    sub.l #18, %d1
 ; CHECK-NEXT:    subx.l %d2, %d0
 ; CHECK-NEXT:    scs %d0
@@ -61,7 +61,7 @@ define i8 @t5(i32 %a) {
 ; CHECK-LABEL: t5:
 ; CHECK:         .cfi_startproc
 ; CHECK-NEXT:  ; %bb.0: ; %entry
-; CHECK-NEXT:    move.l #31, %d1
+; CHECK-NEXT:    moveq #31, %d1
 ; CHECK-NEXT:    move.l (4,%sp), %d0
 ; CHECK-NEXT:    lsr.l %d1, %d0
 ; CHECK-NEXT:    eori.b #1, %d0
@@ -86,7 +86,7 @@ define zeroext i1 @t6(i32 %a) {
 ; CHECK-LABEL: t6:
 ; CHECK:         .cfi_startproc
 ; CHECK-NEXT:  ; %bb.0: ; %entry
-; CHECK-NEXT:    move.l #31, %d0
+; CHECK-NEXT:    moveq #31, %d0
 ; CHECK-NEXT:    move.l (4,%sp), %d1
 ; CHECK-NEXT:    lsr.l %d0, %d1
 ; CHECK-NEXT:    eori.b #1, %d1
diff --git a/llvm/test/CodeGen/M68k/PR57660.ll b/llvm/test/CodeGen/M68k/PR57660.ll
index 184c30a33d79..bad949b08caf 100644
--- a/llvm/test/CodeGen/M68k/PR57660.ll
+++ b/llvm/test/CodeGen/M68k/PR57660.ll
@@ -7,7 +7,7 @@ define dso_local void @foo1() {
 ; CHECK-NEXT:  ; %bb.0: ; %entry
 ; CHECK-NEXT:    suba.l #2, %sp
 ; CHECK-NEXT:    .cfi_def_cfa_offset -6
-; CHECK-NEXT:    move.b #0, %d0
+; CHECK-NEXT:    moveq #0, %d0
 ; CHECK-NEXT:    move.b %d0, (0,%sp) ; 1-byte Folded Spill
 ; CHECK-NEXT:  .LBB0_1: ; %do.body
 ; CHECK-NEXT:    ; =>This Inner Loop Header: Depth=1
diff --git a/llvm/test/CodeGen/M68k/gcc_except_table.ll b/llvm/test/CodeGen/M68k/gcc_except_table.ll
index a7d2a6662724..fe0ed7861dfe 100644
--- a/llvm/test/CodeGen/M68k/gcc_except_table.ll
+++ b/llvm/test/CodeGen/M68k/gcc_except_table.ll
@@ -19,7 +19,7 @@ define i32 @foo() uwtable ssp personality ptr @__gxx_personality_v0 {
 ; CHECK-NEXT:    jsr _Z1fv@PLT
 ; CHECK-NEXT:  .Ltmp1:
 ; CHECK-NEXT:  ; %bb.1: ; %try.cont
-; CHECK-NEXT:    move.l #0, %d0
+; CHECK-NEXT:    moveq #0, %d0
 ; CHECK-NEXT:    adda.l #4, %sp
 ; CHECK-NEXT:    rts
 ; CHECK-NEXT:  .LBB0_2: ; %lpad
diff --git a/llvm/test/CodeGen/M68k/link-unlnk.ll b/llvm/test/CodeGen/M68k/link-unlnk.ll
index dfdd80e66ade..fe39a9a13494 100644
--- a/llvm/test/CodeGen/M68k/link-unlnk.ll
+++ b/llvm/test/CodeGen/M68k/link-unlnk.ll
@@ -105,7 +105,7 @@ define i32 @test_gep() {
 ; FP-NEXT:    .cfi_def_cfa_register %a6
 ; FP-NEXT:    move.l #21, (-4,%a6)
 ; FP-NEXT:    move.l #12, (-256,%a6)
-; FP-NEXT:    move.l #0, %d0
+; FP-NEXT:    moveq #0, %d0
 ; FP-NEXT:    unlk %a6
 ; FP-NEXT:    rts
 ;
@@ -116,7 +116,7 @@ define i32 @test_gep() {
 ; NO-FP-NEXT:    .cfi_def_cfa_offset -260
 ; NO-FP-NEXT:    move.l #21, (252,%sp)
 ; NO-FP-NEXT:    move.l #12, (0,%sp)
-; NO-FP-NEXT:    move.l #0, %d0
+; NO-FP-NEXT:    moveq #0, %d0
 ; NO-FP-NEXT:    adda.l #256, %sp
 ; NO-FP-NEXT:    rts
 entry:
diff --git a/llvm/test/CodeGen/Mips/atomic-min-max.ll b/llvm/test/CodeGen/Mips/atomic-min-max.ll
index 2f07d70808c1..3d3225509d1a 100644
--- a/llvm/test/CodeGen/Mips/atomic-min-max.ll
+++ b/llvm/test/CodeGen/Mips/atomic-min-max.ll
@@ -2156,8 +2156,7 @@ define i16 @test_umax_16(ptr nocapture %ptr, i16 signext %val) {
 ; MIPS32-NEXT:    # =>This Inner Loop Header: Depth=1
 ; MIPS32-NEXT:    ll $2, 0($6)
 ; MIPS32-NEXT:    srav $4, $2, $10
-; MIPS32-NEXT:    sll $4, $4, 16
-; MIPS32-NEXT:    srl $4, $4, 16
+; MIPS32-NEXT:    andi $4, $4, 65535
 ; MIPS32-NEXT:    or $1, $zero, $4
 ; MIPS32-NEXT:    sllv $4, $4, $10
 ; MIPS32-NEXT:    sltu $5, $4, $7
@@ -2695,8 +2694,7 @@ define i16 @test_umin_16(ptr nocapture %ptr, i16 signext %val) {
 ; MIPS32-NEXT:    # =>This Inner Loop Header: Depth=1
 ; MIPS32-NEXT:    ll $2, 0($6)
 ; MIPS32-NEXT:    srav $4, $2, $10
-; MIPS32-NEXT:    sll $4, $4, 16
-; MIPS32-NEXT:    srl $4, $4, 16
+; MIPS32-NEXT:    andi $4, $4, 65535
 ; MIPS32-NEXT:    or $1, $zero, $4
 ; MIPS32-NEXT:    sllv $4, $4, $10
 ; MIPS32-NEXT:    sltu $5, $4, $7
@@ -4313,8 +4311,7 @@ define i8 @test_umax_8(ptr nocapture %ptr, i8 signext %val) {
 ; MIPS32-NEXT:    # =>This Inner Loop Header: Depth=1
 ; MIPS32-NEXT:    ll $2, 0($6)
 ; MIPS32-NEXT:    srav $4, $2, $10
-; MIPS32-NEXT:    sll $4, $4, 24
-; MIPS32-NEXT:    srl $4, $4, 24
+; MIPS32-NEXT:    andi $4, $4, 255
 ; MIPS32-NEXT:    or $1, $zero, $4
 ; MIPS32-NEXT:    sllv $4, $4, $10
 ; MIPS32-NEXT:    sltu $5, $4, $7
@@ -4852,8 +4849,7 @@ define i8 @test_umin_8(ptr nocapture %ptr, i8 signext %val) {
 ; MIPS32-NEXT:    # =>This Inner Loop Header: Depth=1
 ; MIPS32-NEXT:    ll $2, 0($6)
 ; MIPS32-NEXT:    srav $4, $2, $10
-; MIPS32-NEXT:    sll $4, $4, 24
-; MIPS32-NEXT:    srl $4, $4, 24
+; MIPS32-NEXT:    andi $4, $4, 255
 ; MIPS32-NEXT:    or $1, $zero, $4
 ; MIPS32-NEXT:    sllv $4, $4, $10
 ; MIPS32-NEXT:    sltu $5, $4, $7
diff --git a/llvm/test/CodeGen/Mips/mipsr6-minmaxnum.ll b/llvm/test/CodeGen/Mips/mipsr6-minmaxnum.ll
new file mode 100644
index 000000000000..e14e89916e6d
--- /dev/null
+++ b/llvm/test/CodeGen/Mips/mipsr6-minmaxnum.ll
@@ -0,0 +1,69 @@
+; RUN: llc %s -mtriple=mipsisa32r6el-linux-gnu -o - | \
+; RUN:     FileCheck %s --check-prefix=MIPS32R6EL
+; RUN: llc %s -mtriple=mipsisa64r6el-linux-gnuabi64 -o - | \
+; RUN:     FileCheck %s --check-prefix=MIPS64R6EL
+
+define float @mins(float %x, float %y) {
+; MIPS32R6EL-LABEL:	mins
+; MIPS32R6EL:		# %bb.0:
+; MIPS32R6EL-NEXT:	jr	$ra
+; MIPS32R6EL-NEXT:	min.s	$f0, $f12, $f14
+;
+; MIPS64R6EL-LABEL:	mins
+; MIPS64R6EL:		# %bb.0:
+; MIPS64R6EL-NEXT:	jr	$ra
+; MIPS64R6EL-NEXT:	min.s	$f0, $f12, $f13
+
+  %r = tail call float @llvm.minnum.f32(float %x, float %y)
+  ret float %r
+}
+
+define float @maxs(float %x, float %y) {
+; MIPS32R6EL-LABEL:	maxs
+; MIPS32R6EL:		# %bb.0:
+; MIPS32R6EL-NEXT:	jr	$ra
+; MIPS32R6EL-NEXT:	max.s	$f0, $f12, $f14
+;
+; MIPS64R6EL-LABEL:	maxs
+; MIPS64R6EL:		# %bb.0:
+; MIPS64R6EL-NEXT:	jr	$ra
+; MIPS64R6EL-NEXT:	max.s	$f0, $f12, $f13
+
+  %r = tail call float @llvm.maxnum.f32(float %x, float %y)
+  ret float %r
+}
+
+define double @mind(double %x, double %y) {
+; MIPS32R6EL-LABEL:	mind
+; MIPS32R6EL:		# %bb.0:
+; MIPS32R6EL-NEXT:	jr	$ra
+; MIPS32R6EL-NEXT:	min.d	$f0, $f12, $f14
+;
+; MIPS64R6EL-LABEL:	mind
+; MIPS64R6EL:		# %bb.0:
+; MIPS64R6EL-NEXT:	jr	$ra
+; MIPS64R6EL-NEXT:	min.d	$f0, $f12, $f13
+
+  %r = tail call double @llvm.minnum.f64(double %x, double %y)
+  ret double %r
+}
+
+define double @maxd(double %x, double %y) {
+; MIPS32R6EL-LABEL:	maxd
+; MIPS32R6EL:		# %bb.0:
+; MIPS32R6EL-NEXT:	jr	$ra
+; MIPS32R6EL-NEXT:	max.d	$f0, $f12, $f14
+;
+; MIPS64R6EL-LABEL:	maxd
+; MIPS64R6EL:		# %bb.0:
+; MIPS64R6EL-NEXT:	jr	$ra
+; MIPS64R6EL-NEXT:	max.d	$f0, $f12, $f13
+
+  %r = tail call double @llvm.maxnum.f64(double %x, double %y)
+  ret double %r
+}
+
+declare float @llvm.minnum.f32(float, float)
+declare float @llvm.maxnum.f32(float, float)
+declare double @llvm.minnum.f64(double, double)
+declare double @llvm.maxnum.f64(double, double)
diff --git a/llvm/test/CodeGen/Mips/msa/f16-llvm-ir.ll b/llvm/test/CodeGen/Mips/msa/f16-llvm-ir.ll
index 45c7ab980edd..fe68bee408fc 100644
--- a/llvm/test/CodeGen/Mips/msa/f16-llvm-ir.ll
+++ b/llvm/test/CodeGen/Mips/msa/f16-llvm-ir.ll
@@ -2365,101 +2365,159 @@ entry:
 declare float @llvm.minnum.f32(float %Val, float %b)
 
 define void @fminnum(float %b) {
-; MIPS32-LABEL: fminnum:
-; MIPS32:       # %bb.0: # %entry
-; MIPS32-NEXT:    lui $2, %hi(_gp_disp)
-; MIPS32-NEXT:    addiu $2, $2, %lo(_gp_disp)
-; MIPS32-NEXT:    addiu $sp, $sp, -24
-; MIPS32-NEXT:    .cfi_def_cfa_offset 24
-; MIPS32-NEXT:    sw $ra, 20($sp) # 4-byte Folded Spill
-; MIPS32-NEXT:    sw $16, 16($sp) # 4-byte Folded Spill
-; MIPS32-NEXT:    .cfi_offset 31, -4
-; MIPS32-NEXT:    .cfi_offset 16, -8
-; MIPS32-NEXT:    addu $gp, $2, $25
-; MIPS32-NEXT:    mov.s $f14, $f12
-; MIPS32-NEXT:    lw $16, %got(g)($gp)
-; MIPS32-NEXT:    lh $1, 0($16)
-; MIPS32-NEXT:    fill.h $w0, $1
-; MIPS32-NEXT:    fexupr.w $w0, $w0
-; MIPS32-NEXT:    copy_s.w $1, $w0[0]
-; MIPS32-NEXT:    lw $25, %call16(fminf)($gp)
-; MIPS32-NEXT:    jalr $25
-; MIPS32-NEXT:    mtc1 $1, $f12
-; MIPS32-NEXT:    mfc1 $1, $f0
-; MIPS32-NEXT:    fill.w $w0, $1
-; MIPS32-NEXT:    fexdo.h $w0, $w0, $w0
-; MIPS32-NEXT:    copy_u.h $1, $w0[0]
-; MIPS32-NEXT:    sh $1, 0($16)
-; MIPS32-NEXT:    lw $16, 16($sp) # 4-byte Folded Reload
-; MIPS32-NEXT:    lw $ra, 20($sp) # 4-byte Folded Reload
-; MIPS32-NEXT:    jr $ra
-; MIPS32-NEXT:    addiu $sp, $sp, 24
+; MIPS32-O32-LABEL: fminnum:
+; MIPS32-O32:       # %bb.0: # %entry
+; MIPS32-O32-NEXT:    lui $2, %hi(_gp_disp)
+; MIPS32-O32-NEXT:    addiu $2, $2, %lo(_gp_disp)
+; MIPS32-O32-NEXT:    addiu $sp, $sp, -24
+; MIPS32-O32-NEXT:    .cfi_def_cfa_offset 24
+; MIPS32-O32-NEXT:    sw $ra, 20($sp) # 4-byte Folded Spill
+; MIPS32-O32-NEXT:    sw $16, 16($sp) # 4-byte Folded Spill
+; MIPS32-O32-NEXT:    .cfi_offset 31, -4
+; MIPS32-O32-NEXT:    .cfi_offset 16, -8
+; MIPS32-O32-NEXT:    addu $gp, $2, $25
+; MIPS32-O32-NEXT:    mov.s $f14, $f12
+; MIPS32-O32-NEXT:    lw $16, %got(g)($gp)
+; MIPS32-O32-NEXT:    lh $1, 0($16)
+; MIPS32-O32-NEXT:    fill.h $w0, $1
+; MIPS32-O32-NEXT:    fexupr.w $w0, $w0
+; MIPS32-O32-NEXT:    copy_s.w $1, $w0[0]
+; MIPS32-O32-NEXT:    lw $25, %call16(fminf)($gp)
+; MIPS32-O32-NEXT:    jalr $25
+; MIPS32-O32-NEXT:    mtc1 $1, $f12
+; MIPS32-O32-NEXT:    mfc1 $1, $f0
+; MIPS32-O32-NEXT:    fill.w $w0, $1
+; MIPS32-O32-NEXT:    fexdo.h $w0, $w0, $w0
+; MIPS32-O32-NEXT:    copy_u.h $1, $w0[0]
+; MIPS32-O32-NEXT:    sh $1, 0($16)
+; MIPS32-O32-NEXT:    lw $16, 16($sp) # 4-byte Folded Reload
+; MIPS32-O32-NEXT:    lw $ra, 20($sp) # 4-byte Folded Reload
+; MIPS32-O32-NEXT:    jr $ra
+; MIPS32-O32-NEXT:    addiu $sp, $sp, 24
 ;
-; MIPS64-N32-LABEL: fminnum:
-; MIPS64-N32:       # %bb.0: # %entry
-; MIPS64-N32-NEXT:    addiu $sp, $sp, -32
-; MIPS64-N32-NEXT:    .cfi_def_cfa_offset 32
-; MIPS64-N32-NEXT:    sd $ra, 24($sp) # 8-byte Folded Spill
-; MIPS64-N32-NEXT:    sd $gp, 16($sp) # 8-byte Folded Spill
-; MIPS64-N32-NEXT:    sd $16, 8($sp) # 8-byte Folded Spill
-; MIPS64-N32-NEXT:    .cfi_offset 31, -8
-; MIPS64-N32-NEXT:    .cfi_offset 28, -16
-; MIPS64-N32-NEXT:    .cfi_offset 16, -24
-; MIPS64-N32-NEXT:    lui $1, %hi(%neg(%gp_rel(fminnum)))
-; MIPS64-N32-NEXT:    addu $1, $1, $25
-; MIPS64-N32-NEXT:    addiu $gp, $1, %lo(%neg(%gp_rel(fminnum)))
-; MIPS64-N32-NEXT:    mov.s $f13, $f12
-; MIPS64-N32-NEXT:    lw $16, %got_disp(g)($gp)
-; MIPS64-N32-NEXT:    lh $1, 0($16)
-; MIPS64-N32-NEXT:    fill.h $w0, $1
-; MIPS64-N32-NEXT:    fexupr.w $w0, $w0
-; MIPS64-N32-NEXT:    copy_s.w $1, $w0[0]
-; MIPS64-N32-NEXT:    lw $25, %call16(fminf)($gp)
-; MIPS64-N32-NEXT:    jalr $25
-; MIPS64-N32-NEXT:    mtc1 $1, $f12
-; MIPS64-N32-NEXT:    mfc1 $1, $f0
-; MIPS64-N32-NEXT:    fill.w $w0, $1
-; MIPS64-N32-NEXT:    fexdo.h $w0, $w0, $w0
-; MIPS64-N32-NEXT:    copy_u.h $1, $w0[0]
-; MIPS64-N32-NEXT:    sh $1, 0($16)
-; MIPS64-N32-NEXT:    ld $16, 8($sp) # 8-byte Folded Reload
-; MIPS64-N32-NEXT:    ld $gp, 16($sp) # 8-byte Folded Reload
-; MIPS64-N32-NEXT:    ld $ra, 24($sp) # 8-byte Folded Reload
-; MIPS64-N32-NEXT:    jr $ra
-; MIPS64-N32-NEXT:    addiu $sp, $sp, 32
+; MIPS64R5-N32-LABEL: fminnum:
+; MIPS64R5-N32:       # %bb.0: # %entry
+; MIPS64R5-N32-NEXT:    addiu $sp, $sp, -32
+; MIPS64R5-N32-NEXT:    .cfi_def_cfa_offset 32
+; MIPS64R5-N32-NEXT:    sd $ra, 24($sp) # 8-byte Folded Spill
+; MIPS64R5-N32-NEXT:    sd $gp, 16($sp) # 8-byte Folded Spill
+; MIPS64R5-N32-NEXT:    sd $16, 8($sp) # 8-byte Folded Spill
+; MIPS64R5-N32-NEXT:    .cfi_offset 31, -8
+; MIPS64R5-N32-NEXT:    .cfi_offset 28, -16
+; MIPS64R5-N32-NEXT:    .cfi_offset 16, -24
+; MIPS64R5-N32-NEXT:    lui $1, %hi(%neg(%gp_rel(fminnum)))
+; MIPS64R5-N32-NEXT:    addu $1, $1, $25
+; MIPS64R5-N32-NEXT:    addiu $gp, $1, %lo(%neg(%gp_rel(fminnum)))
+; MIPS64R5-N32-NEXT:    mov.s $f13, $f12
+; MIPS64R5-N32-NEXT:    lw $16, %got_disp(g)($gp)
+; MIPS64R5-N32-NEXT:    lh $1, 0($16)
+; MIPS64R5-N32-NEXT:    fill.h $w0, $1
+; MIPS64R5-N32-NEXT:    fexupr.w $w0, $w0
+; MIPS64R5-N32-NEXT:    copy_s.w $1, $w0[0]
+; MIPS64R5-N32-NEXT:    lw $25, %call16(fminf)($gp)
+; MIPS64R5-N32-NEXT:    jalr $25
+; MIPS64R5-N32-NEXT:    mtc1 $1, $f12
+; MIPS64R5-N32-NEXT:    mfc1 $1, $f0
+; MIPS64R5-N32-NEXT:    fill.w $w0, $1
+; MIPS64R5-N32-NEXT:    fexdo.h $w0, $w0, $w0
+; MIPS64R5-N32-NEXT:    copy_u.h $1, $w0[0]
+; MIPS64R5-N32-NEXT:    sh $1, 0($16)
+; MIPS64R5-N32-NEXT:    ld $16, 8($sp) # 8-byte Folded Reload
+; MIPS64R5-N32-NEXT:    ld $gp, 16($sp) # 8-byte Folded Reload
+; MIPS64R5-N32-NEXT:    ld $ra, 24($sp) # 8-byte Folded Reload
+; MIPS64R5-N32-NEXT:    jr $ra
+; MIPS64R5-N32-NEXT:    addiu $sp, $sp, 32
+;
+; MIPS64R5-N64-LABEL: fminnum:
+; MIPS64R5-N64:       # %bb.0: # %entry
+; MIPS64R5-N64-NEXT:    daddiu $sp, $sp, -32
+; MIPS64R5-N64-NEXT:    .cfi_def_cfa_offset 32
+; MIPS64R5-N64-NEXT:    sd $ra, 24($sp) # 8-byte Folded Spill
+; MIPS64R5-N64-NEXT:    sd $gp, 16($sp) # 8-byte Folded Spill
+; MIPS64R5-N64-NEXT:    sd $16, 8($sp) # 8-byte Folded Spill
+; MIPS64R5-N64-NEXT:    .cfi_offset 31, -8
+; MIPS64R5-N64-NEXT:    .cfi_offset 28, -16
+; MIPS64R5-N64-NEXT:    .cfi_offset 16, -24
+; MIPS64R5-N64-NEXT:    lui $1, %hi(%neg(%gp_rel(fminnum)))
+; MIPS64R5-N64-NEXT:    daddu $1, $1, $25
+; MIPS64R5-N64-NEXT:    daddiu $gp, $1, %lo(%neg(%gp_rel(fminnum)))
+; MIPS64R5-N64-NEXT:    mov.s $f13, $f12
+; MIPS64R5-N64-NEXT:    ld $16, %got_disp(g)($gp)
+; MIPS64R5-N64-NEXT:    lh $1, 0($16)
+; MIPS64R5-N64-NEXT:    fill.h $w0, $1
+; MIPS64R5-N64-NEXT:    fexupr.w $w0, $w0
+; MIPS64R5-N64-NEXT:    copy_s.w $1, $w0[0]
+; MIPS64R5-N64-NEXT:    ld $25, %call16(fminf)($gp)
+; MIPS64R5-N64-NEXT:    jalr $25
+; MIPS64R5-N64-NEXT:    mtc1 $1, $f12
+; MIPS64R5-N64-NEXT:    mfc1 $1, $f0
+; MIPS64R5-N64-NEXT:    fill.w $w0, $1
+; MIPS64R5-N64-NEXT:    fexdo.h $w0, $w0, $w0
+; MIPS64R5-N64-NEXT:    copy_u.h $1, $w0[0]
+; MIPS64R5-N64-NEXT:    sh $1, 0($16)
+; MIPS64R5-N64-NEXT:    ld $16, 8($sp) # 8-byte Folded Reload
+; MIPS64R5-N64-NEXT:    ld $gp, 16($sp) # 8-byte Folded Reload
+; MIPS64R5-N64-NEXT:    ld $ra, 24($sp) # 8-byte Folded Reload
+; MIPS64R5-N64-NEXT:    jr $ra
+; MIPS64R5-N64-NEXT:    daddiu $sp, $sp, 32
+;
+; MIPSR6-O32-LABEL: fminnum:
+; MIPSR6-O32:       # %bb.0: # %entry
+; MIPSR6-O32-NEXT:    lui $2, %hi(_gp_disp)
+; MIPSR6-O32-NEXT:    addiu $2, $2, %lo(_gp_disp)
+; MIPSR6-O32-NEXT:    addu $1, $2, $25
+; MIPSR6-O32-NEXT:    lw $1, %got(g)($1)
+; MIPSR6-O32-NEXT:    lh $2, 0($1)
+; MIPSR6-O32-NEXT:    fill.h $w0, $2
+; MIPSR6-O32-NEXT:    fexupr.w $w0, $w0
+; MIPSR6-O32-NEXT:    copy_s.w $2, $w0[0]
+; MIPSR6-O32-NEXT:    mtc1 $2, $f0
+; MIPSR6-O32-NEXT:    min.s $f0, $f0, $f12
+; MIPSR6-O32-NEXT:    mfc1 $2, $f0
+; MIPSR6-O32-NEXT:    fill.w $w0, $2
+; MIPSR6-O32-NEXT:    fexdo.h $w0, $w0, $w0
+; MIPSR6-O32-NEXT:    copy_u.h $2, $w0[0]
+; MIPSR6-O32-NEXT:    jr $ra
+; MIPSR6-O32-NEXT:    sh $2, 0($1)
+;
+; MIPSR6-N32-LABEL: fminnum:
+; MIPSR6-N32:       # %bb.0: # %entry
+; MIPSR6-N32-NEXT:    lui $1, %hi(%neg(%gp_rel(fminnum)))
+; MIPSR6-N32-NEXT:    addu $1, $1, $25
+; MIPSR6-N32-NEXT:    addiu $1, $1, %lo(%neg(%gp_rel(fminnum)))
+; MIPSR6-N32-NEXT:    lw $1, %got_disp(g)($1)
+; MIPSR6-N32-NEXT:    lh $2, 0($1)
+; MIPSR6-N32-NEXT:    fill.h $w0, $2
+; MIPSR6-N32-NEXT:    fexupr.w $w0, $w0
+; MIPSR6-N32-NEXT:    copy_s.w $2, $w0[0]
+; MIPSR6-N32-NEXT:    mtc1 $2, $f0
+; MIPSR6-N32-NEXT:    min.s $f0, $f0, $f12
+; MIPSR6-N32-NEXT:    mfc1 $2, $f0
+; MIPSR6-N32-NEXT:    fill.w $w0, $2
+; MIPSR6-N32-NEXT:    fexdo.h $w0, $w0, $w0
+; MIPSR6-N32-NEXT:    copy_u.h $2, $w0[0]
+; MIPSR6-N32-NEXT:    jr $ra
+; MIPSR6-N32-NEXT:    sh $2, 0($1)
+;
+; MIPSR6-N64-LABEL: fminnum:
+; MIPSR6-N64:       # %bb.0: # %entry
+; MIPSR6-N64-NEXT:    lui $1, %hi(%neg(%gp_rel(fminnum)))
+; MIPSR6-N64-NEXT:    daddu $1, $1, $25
+; MIPSR6-N64-NEXT:    daddiu $1, $1, %lo(%neg(%gp_rel(fminnum)))
+; MIPSR6-N64-NEXT:    ld $1, %got_disp(g)($1)
+; MIPSR6-N64-NEXT:    lh $2, 0($1)
+; MIPSR6-N64-NEXT:    fill.h $w0, $2
+; MIPSR6-N64-NEXT:    fexupr.w $w0, $w0
+; MIPSR6-N64-NEXT:    copy_s.w $2, $w0[0]
+; MIPSR6-N64-NEXT:    mtc1 $2, $f0
+; MIPSR6-N64-NEXT:    min.s $f0, $f0, $f12
+; MIPSR6-N64-NEXT:    mfc1 $2, $f0
+; MIPSR6-N64-NEXT:    fill.w $w0, $2
+; MIPSR6-N64-NEXT:    fexdo.h $w0, $w0, $w0
+; MIPSR6-N64-NEXT:    copy_u.h $2, $w0[0]
+; MIPSR6-N64-NEXT:    jr $ra
+; MIPSR6-N64-NEXT:    sh $2, 0($1)
 ;
-; MIPS64-N64-LABEL: fminnum:
-; MIPS64-N64:       # %bb.0: # %entry
-; MIPS64-N64-NEXT:    daddiu $sp, $sp, -32
-; MIPS64-N64-NEXT:    .cfi_def_cfa_offset 32
-; MIPS64-N64-NEXT:    sd $ra, 24($sp) # 8-byte Folded Spill
-; MIPS64-N64-NEXT:    sd $gp, 16($sp) # 8-byte Folded Spill
-; MIPS64-N64-NEXT:    sd $16, 8($sp) # 8-byte Folded Spill
-; MIPS64-N64-NEXT:    .cfi_offset 31, -8
-; MIPS64-N64-NEXT:    .cfi_offset 28, -16
-; MIPS64-N64-NEXT:    .cfi_offset 16, -24
-; MIPS64-N64-NEXT:    lui $1, %hi(%neg(%gp_rel(fminnum)))
-; MIPS64-N64-NEXT:    daddu $1, $1, $25
-; MIPS64-N64-NEXT:    daddiu $gp, $1, %lo(%neg(%gp_rel(fminnum)))
-; MIPS64-N64-NEXT:    mov.s $f13, $f12
-; MIPS64-N64-NEXT:    ld $16, %got_disp(g)($gp)
-; MIPS64-N64-NEXT:    lh $1, 0($16)
-; MIPS64-N64-NEXT:    fill.h $w0, $1
-; MIPS64-N64-NEXT:    fexupr.w $w0, $w0
-; MIPS64-N64-NEXT:    copy_s.w $1, $w0[0]
-; MIPS64-N64-NEXT:    ld $25, %call16(fminf)($gp)
-; MIPS64-N64-NEXT:    jalr $25
-; MIPS64-N64-NEXT:    mtc1 $1, $f12
-; MIPS64-N64-NEXT:    mfc1 $1, $f0
-; MIPS64-N64-NEXT:    fill.w $w0, $1
-; MIPS64-N64-NEXT:    fexdo.h $w0, $w0, $w0
-; MIPS64-N64-NEXT:    copy_u.h $1, $w0[0]
-; MIPS64-N64-NEXT:    sh $1, 0($16)
-; MIPS64-N64-NEXT:    ld $16, 8($sp) # 8-byte Folded Reload
-; MIPS64-N64-NEXT:    ld $gp, 16($sp) # 8-byte Folded Reload
-; MIPS64-N64-NEXT:    ld $ra, 24($sp) # 8-byte Folded Reload
-; MIPS64-N64-NEXT:    jr $ra
-; MIPS64-N64-NEXT:    daddiu $sp, $sp, 32
 entry:
   %0 = load i16, ptr @g, align 2
   %1 = call float @llvm.convert.from.fp16.f32(i16 %0)
@@ -2477,101 +2535,158 @@ entry:
 declare float @llvm.maxnum.f32(float %Val, float %b)
 
 define void @fmaxnum(float %b) {
-; MIPS32-LABEL: fmaxnum:
-; MIPS32:       # %bb.0: # %entry
-; MIPS32-NEXT:    lui $2, %hi(_gp_disp)
-; MIPS32-NEXT:    addiu $2, $2, %lo(_gp_disp)
-; MIPS32-NEXT:    addiu $sp, $sp, -24
-; MIPS32-NEXT:    .cfi_def_cfa_offset 24
-; MIPS32-NEXT:    sw $ra, 20($sp) # 4-byte Folded Spill
-; MIPS32-NEXT:    sw $16, 16($sp) # 4-byte Folded Spill
-; MIPS32-NEXT:    .cfi_offset 31, -4
-; MIPS32-NEXT:    .cfi_offset 16, -8
-; MIPS32-NEXT:    addu $gp, $2, $25
-; MIPS32-NEXT:    mov.s $f14, $f12
-; MIPS32-NEXT:    lw $16, %got(g)($gp)
-; MIPS32-NEXT:    lh $1, 0($16)
-; MIPS32-NEXT:    fill.h $w0, $1
-; MIPS32-NEXT:    fexupr.w $w0, $w0
-; MIPS32-NEXT:    copy_s.w $1, $w0[0]
-; MIPS32-NEXT:    lw $25, %call16(fmaxf)($gp)
-; MIPS32-NEXT:    jalr $25
-; MIPS32-NEXT:    mtc1 $1, $f12
-; MIPS32-NEXT:    mfc1 $1, $f0
-; MIPS32-NEXT:    fill.w $w0, $1
-; MIPS32-NEXT:    fexdo.h $w0, $w0, $w0
-; MIPS32-NEXT:    copy_u.h $1, $w0[0]
-; MIPS32-NEXT:    sh $1, 0($16)
-; MIPS32-NEXT:    lw $16, 16($sp) # 4-byte Folded Reload
-; MIPS32-NEXT:    lw $ra, 20($sp) # 4-byte Folded Reload
-; MIPS32-NEXT:    jr $ra
-; MIPS32-NEXT:    addiu $sp, $sp, 24
+; MIPS32-O32-LABEL: fmaxnum:
+; MIPS32-O32:       # %bb.0: # %entry
+; MIPS32-O32-NEXT:    lui $2, %hi(_gp_disp)
+; MIPS32-O32-NEXT:    addiu $2, $2, %lo(_gp_disp)
+; MIPS32-O32-NEXT:    addiu $sp, $sp, -24
+; MIPS32-O32-NEXT:    .cfi_def_cfa_offset 24
+; MIPS32-O32-NEXT:    sw $ra, 20($sp) # 4-byte Folded Spill
+; MIPS32-O32-NEXT:    sw $16, 16($sp) # 4-byte Folded Spill
+; MIPS32-O32-NEXT:    .cfi_offset 31, -4
+; MIPS32-O32-NEXT:    .cfi_offset 16, -8
+; MIPS32-O32-NEXT:    addu $gp, $2, $25
+; MIPS32-O32-NEXT:    mov.s $f14, $f12
+; MIPS32-O32-NEXT:    lw $16, %got(g)($gp)
+; MIPS32-O32-NEXT:    lh $1, 0($16)
+; MIPS32-O32-NEXT:    fill.h $w0, $1
+; MIPS32-O32-NEXT:    fexupr.w $w0, $w0
+; MIPS32-O32-NEXT:    copy_s.w $1, $w0[0]
+; MIPS32-O32-NEXT:    lw $25, %call16(fmaxf)($gp)
+; MIPS32-O32-NEXT:    jalr $25
+; MIPS32-O32-NEXT:    mtc1 $1, $f12
+; MIPS32-O32-NEXT:    mfc1 $1, $f0
+; MIPS32-O32-NEXT:    fill.w $w0, $1
+; MIPS32-O32-NEXT:    fexdo.h $w0, $w0, $w0
+; MIPS32-O32-NEXT:    copy_u.h $1, $w0[0]
+; MIPS32-O32-NEXT:    sh $1, 0($16)
+; MIPS32-O32-NEXT:    lw $16, 16($sp) # 4-byte Folded Reload
+; MIPS32-O32-NEXT:    lw $ra, 20($sp) # 4-byte Folded Reload
+; MIPS32-O32-NEXT:    jr $ra
+; MIPS32-O32-NEXT:    addiu $sp, $sp, 24
 ;
-; MIPS64-N32-LABEL: fmaxnum:
-; MIPS64-N32:       # %bb.0: # %entry
-; MIPS64-N32-NEXT:    addiu $sp, $sp, -32
-; MIPS64-N32-NEXT:    .cfi_def_cfa_offset 32
-; MIPS64-N32-NEXT:    sd $ra, 24($sp) # 8-byte Folded Spill
-; MIPS64-N32-NEXT:    sd $gp, 16($sp) # 8-byte Folded Spill
-; MIPS64-N32-NEXT:    sd $16, 8($sp) # 8-byte Folded Spill
-; MIPS64-N32-NEXT:    .cfi_offset 31, -8
-; MIPS64-N32-NEXT:    .cfi_offset 28, -16
-; MIPS64-N32-NEXT:    .cfi_offset 16, -24
-; MIPS64-N32-NEXT:    lui $1, %hi(%neg(%gp_rel(fmaxnum)))
-; MIPS64-N32-NEXT:    addu $1, $1, $25
-; MIPS64-N32-NEXT:    addiu $gp, $1, %lo(%neg(%gp_rel(fmaxnum)))
-; MIPS64-N32-NEXT:    mov.s $f13, $f12
-; MIPS64-N32-NEXT:    lw $16, %got_disp(g)($gp)
-; MIPS64-N32-NEXT:    lh $1, 0($16)
-; MIPS64-N32-NEXT:    fill.h $w0, $1
-; MIPS64-N32-NEXT:    fexupr.w $w0, $w0
-; MIPS64-N32-NEXT:    copy_s.w $1, $w0[0]
-; MIPS64-N32-NEXT:    lw $25, %call16(fmaxf)($gp)
-; MIPS64-N32-NEXT:    jalr $25
-; MIPS64-N32-NEXT:    mtc1 $1, $f12
-; MIPS64-N32-NEXT:    mfc1 $1, $f0
-; MIPS64-N32-NEXT:    fill.w $w0, $1
-; MIPS64-N32-NEXT:    fexdo.h $w0, $w0, $w0
-; MIPS64-N32-NEXT:    copy_u.h $1, $w0[0]
-; MIPS64-N32-NEXT:    sh $1, 0($16)
-; MIPS64-N32-NEXT:    ld $16, 8($sp) # 8-byte Folded Reload
-; MIPS64-N32-NEXT:    ld $gp, 16($sp) # 8-byte Folded Reload
-; MIPS64-N32-NEXT:    ld $ra, 24($sp) # 8-byte Folded Reload
-; MIPS64-N32-NEXT:    jr $ra
-; MIPS64-N32-NEXT:    addiu $sp, $sp, 32
+; MIPS64R5-N32-LABEL: fmaxnum:
+; MIPS64R5-N32:       # %bb.0: # %entry
+; MIPS64R5-N32-NEXT:    addiu $sp, $sp, -32
+; MIPS64R5-N32-NEXT:    .cfi_def_cfa_offset 32
+; MIPS64R5-N32-NEXT:    sd $ra, 24($sp) # 8-byte Folded Spill
+; MIPS64R5-N32-NEXT:    sd $gp, 16($sp) # 8-byte Folded Spill
+; MIPS64R5-N32-NEXT:    sd $16, 8($sp) # 8-byte Folded Spill
+; MIPS64R5-N32-NEXT:    .cfi_offset 31, -8
+; MIPS64R5-N32-NEXT:    .cfi_offset 28, -16
+; MIPS64R5-N32-NEXT:    .cfi_offset 16, -24
+; MIPS64R5-N32-NEXT:    lui $1, %hi(%neg(%gp_rel(fmaxnum)))
+; MIPS64R5-N32-NEXT:    addu $1, $1, $25
+; MIPS64R5-N32-NEXT:    addiu $gp, $1, %lo(%neg(%gp_rel(fmaxnum)))
+; MIPS64R5-N32-NEXT:    mov.s $f13, $f12
+; MIPS64R5-N32-NEXT:    lw $16, %got_disp(g)($gp)
+; MIPS64R5-N32-NEXT:    lh $1, 0($16)
+; MIPS64R5-N32-NEXT:    fill.h $w0, $1
+; MIPS64R5-N32-NEXT:    fexupr.w $w0, $w0
+; MIPS64R5-N32-NEXT:    copy_s.w $1, $w0[0]
+; MIPS64R5-N32-NEXT:    lw $25, %call16(fmaxf)($gp)
+; MIPS64R5-N32-NEXT:    jalr $25
+; MIPS64R5-N32-NEXT:    mtc1 $1, $f12
+; MIPS64R5-N32-NEXT:    mfc1 $1, $f0
+; MIPS64R5-N32-NEXT:    fill.w $w0, $1
+; MIPS64R5-N32-NEXT:    fexdo.h $w0, $w0, $w0
+; MIPS64R5-N32-NEXT:    copy_u.h $1, $w0[0]
+; MIPS64R5-N32-NEXT:    sh $1, 0($16)
+; MIPS64R5-N32-NEXT:    ld $16, 8($sp) # 8-byte Folded Reload
+; MIPS64R5-N32-NEXT:    ld $gp, 16($sp) # 8-byte Folded Reload
+; MIPS64R5-N32-NEXT:    ld $ra, 24($sp) # 8-byte Folded Reload
+; MIPS64R5-N32-NEXT:    jr $ra
+; MIPS64R5-N32-NEXT:    addiu $sp, $sp, 32
 ;
-; MIPS64-N64-LABEL: fmaxnum:
-; MIPS64-N64:       # %bb.0: # %entry
-; MIPS64-N64-NEXT:    daddiu $sp, $sp, -32
-; MIPS64-N64-NEXT:    .cfi_def_cfa_offset 32
-; MIPS64-N64-NEXT:    sd $ra, 24($sp) # 8-byte Folded Spill
-; MIPS64-N64-NEXT:    sd $gp, 16($sp) # 8-byte Folded Spill
-; MIPS64-N64-NEXT:    sd $16, 8($sp) # 8-byte Folded Spill
-; MIPS64-N64-NEXT:    .cfi_offset 31, -8
-; MIPS64-N64-NEXT:    .cfi_offset 28, -16
-; MIPS64-N64-NEXT:    .cfi_offset 16, -24
-; MIPS64-N64-NEXT:    lui $1, %hi(%neg(%gp_rel(fmaxnum)))
-; MIPS64-N64-NEXT:    daddu $1, $1, $25
-; MIPS64-N64-NEXT:    daddiu $gp, $1, %lo(%neg(%gp_rel(fmaxnum)))
-; MIPS64-N64-NEXT:    mov.s $f13, $f12
-; MIPS64-N64-NEXT:    ld $16, %got_disp(g)($gp)
-; MIPS64-N64-NEXT:    lh $1, 0($16)
-; MIPS64-N64-NEXT:    fill.h $w0, $1
-; MIPS64-N64-NEXT:    fexupr.w $w0, $w0
-; MIPS64-N64-NEXT:    copy_s.w $1, $w0[0]
-; MIPS64-N64-NEXT:    ld $25, %call16(fmaxf)($gp)
-; MIPS64-N64-NEXT:    jalr $25
-; MIPS64-N64-NEXT:    mtc1 $1, $f12
-; MIPS64-N64-NEXT:    mfc1 $1, $f0
-; MIPS64-N64-NEXT:    fill.w $w0, $1
-; MIPS64-N64-NEXT:    fexdo.h $w0, $w0, $w0
-; MIPS64-N64-NEXT:    copy_u.h $1, $w0[0]
-; MIPS64-N64-NEXT:    sh $1, 0($16)
-; MIPS64-N64-NEXT:    ld $16, 8($sp) # 8-byte Folded Reload
-; MIPS64-N64-NEXT:    ld $gp, 16($sp) # 8-byte Folded Reload
-; MIPS64-N64-NEXT:    ld $ra, 24($sp) # 8-byte Folded Reload
-; MIPS64-N64-NEXT:    jr $ra
-; MIPS64-N64-NEXT:    daddiu $sp, $sp, 32
+; MIPS64R5-N64-LABEL: fmaxnum:
+; MIPS64R5-N64:       # %bb.0: # %entry
+; MIPS64R5-N64-NEXT:    daddiu $sp, $sp, -32
+; MIPS64R5-N64-NEXT:    .cfi_def_cfa_offset 32
+; MIPS64R5-N64-NEXT:    sd $ra, 24($sp) # 8-byte Folded Spill
+; MIPS64R5-N64-NEXT:    sd $gp, 16($sp) # 8-byte Folded Spill
+; MIPS64R5-N64-NEXT:    sd $16, 8($sp) # 8-byte Folded Spill
+; MIPS64R5-N64-NEXT:    .cfi_offset 31, -8
+; MIPS64R5-N64-NEXT:    .cfi_offset 28, -16
+; MIPS64R5-N64-NEXT:    .cfi_offset 16, -24
+; MIPS64R5-N64-NEXT:    lui $1, %hi(%neg(%gp_rel(fmaxnum)))
+; MIPS64R5-N64-NEXT:    daddu $1, $1, $25
+; MIPS64R5-N64-NEXT:    daddiu $gp, $1, %lo(%neg(%gp_rel(fmaxnum)))
+; MIPS64R5-N64-NEXT:    mov.s $f13, $f12
+; MIPS64R5-N64-NEXT:    ld $16, %got_disp(g)($gp)
+; MIPS64R5-N64-NEXT:    lh $1, 0($16)
+; MIPS64R5-N64-NEXT:    fill.h $w0, $1
+; MIPS64R5-N64-NEXT:    fexupr.w $w0, $w0
+; MIPS64R5-N64-NEXT:    copy_s.w $1, $w0[0]
+; MIPS64R5-N64-NEXT:    ld $25, %call16(fmaxf)($gp)
+; MIPS64R5-N64-NEXT:    jalr $25
+; MIPS64R5-N64-NEXT:    mtc1 $1, $f12
+; MIPS64R5-N64-NEXT:    mfc1 $1, $f0
+; MIPS64R5-N64-NEXT:    fill.w $w0, $1
+; MIPS64R5-N64-NEXT:    fexdo.h $w0, $w0, $w0
+; MIPS64R5-N64-NEXT:    copy_u.h $1, $w0[0]
+; MIPS64R5-N64-NEXT:    sh $1, 0($16)
+; MIPS64R5-N64-NEXT:    ld $16, 8($sp) # 8-byte Folded Reload
+; MIPS64R5-N64-NEXT:    ld $gp, 16($sp) # 8-byte Folded Reload
+; MIPS64R5-N64-NEXT:    ld $ra, 24($sp) # 8-byte Folded Reload
+; MIPS64R5-N64-NEXT:    jr $ra
+; MIPS64R5-N64-NEXT:    daddiu $sp, $sp, 32
+;
+; MIPSR6-O32-LABEL: fmaxnum:
+; MIPSR6-O32:       # %bb.0:
+; MIPSR6-O32-NEXT:    lui $2, %hi(_gp_disp)
+; MIPSR6-O32-NEXT:    addiu $2, $2, %lo(_gp_disp)
+; MIPSR6-O32-NEXT:    addu $1, $2, $25
+; MIPSR6-O32-NEXT:    lw $1, %got(g)($1)
+; MIPSR6-O32-NEXT:    lh $2, 0($1)
+; MIPSR6-O32-NEXT:    fill.h $w0, $2
+; MIPSR6-O32-NEXT:    fexupr.w $w0, $w0
+; MIPSR6-O32-NEXT:    copy_s.w $2, $w0[0]
+; MIPSR6-O32-NEXT:    mtc1 $2, $f0
+; MIPSR6-O32-NEXT:    max.s $f0, $f0, $f12
+; MIPSR6-O32-NEXT:    mfc1 $2, $f0
+; MIPSR6-O32-NEXT:    fill.w $w0, $2
+; MIPSR6-O32-NEXT:    fexdo.h $w0, $w0, $w0
+; MIPSR6-O32-NEXT:    copy_u.h $2, $w0[0]
+; MIPSR6-O32-NEXT:    jr $ra
+; MIPSR6-O32-NEXT:    sh $2, 0($1)
+;
+; MIPSR6-N32-LABEL: fmaxnum:
+; MIPSR6-N32:       # %bb.0:
+; MIPSR6-N32-NEXT:    lui $1, %hi(%neg(%gp_rel(fmaxnum)))
+; MIPSR6-N32-NEXT:    addu $1, $1, $25
+; MIPSR6-N32-NEXT:    addiu $1, $1, %lo(%neg(%gp_rel(fmaxnum)))
+; MIPSR6-N32-NEXT:    lw $1, %got_disp(g)($1)
+; MIPSR6-N32-NEXT:    lh $2, 0($1)
+; MIPSR6-N32-NEXT:    fill.h $w0, $2
+; MIPSR6-N32-NEXT:    fexupr.w $w0, $w0
+; MIPSR6-N32-NEXT:    copy_s.w $2, $w0[0]
+; MIPSR6-N32-NEXT:    mtc1 $2, $f0
+; MIPSR6-N32-NEXT:    max.s $f0, $f0, $f12
+; MIPSR6-N32-NEXT:    mfc1 $2, $f0
+; MIPSR6-N32-NEXT:    fill.w $w0, $2
+; MIPSR6-N32-NEXT:    fexdo.h $w0, $w0, $w0
+; MIPSR6-N32-NEXT:    copy_u.h $2, $w0[0]
+; MIPSR6-N32-NEXT:    jr $ra
+; MIPSR6-N32-NEXT:    sh $2, 0($1)
+;
+; MIPSR6-N64-LABEL: fmaxnum:
+; MIPSR6-N64:       # %bb.0:
+; MIPSR6-N64-NEXT:    lui $1, %hi(%neg(%gp_rel(fmaxnum)))
+; MIPSR6-N64-NEXT:    daddu $1, $1, $25
+; MIPSR6-N64-NEXT:    daddiu $1, $1, %lo(%neg(%gp_rel(fmaxnum)))
+; MIPSR6-N64-NEXT:    ld $1, %got_disp(g)($1)
+; MIPSR6-N64-NEXT:    lh $2, 0($1)
+; MIPSR6-N64-NEXT:    fill.h $w0, $2
+; MIPSR6-N64-NEXT:    fexupr.w $w0, $w0
+; MIPSR6-N64-NEXT:    copy_s.w $2, $w0[0]
+; MIPSR6-N64-NEXT:    mtc1 $2, $f0
+; MIPSR6-N64-NEXT:    max.s $f0, $f0, $f12
+; MIPSR6-N64-NEXT:    mfc1 $2, $f0
+; MIPSR6-N64-NEXT:    fill.w $w0, $2
+; MIPSR6-N64-NEXT:    fexdo.h $w0, $w0, $w0
+; MIPSR6-N64-NEXT:    copy_u.h $2, $w0[0]
+; MIPSR6-N64-NEXT:    jr $ra
+; MIPSR6-N64-NEXT:    sh $2, 0($1)
 entry:
   %0 = load i16, ptr @g, align 2
   %1 = call float @llvm.convert.from.fp16.f32(i16 %0)
diff --git a/llvm/test/CodeGen/NVPTX/rsqrt-opt.ll b/llvm/test/CodeGen/NVPTX/rsqrt-opt.ll
new file mode 100644
index 000000000000..9dda6075a23c
--- /dev/null
+++ b/llvm/test/CodeGen/NVPTX/rsqrt-opt.ll
@@ -0,0 +1,75 @@
+; RUN: llc < %s -march=nvptx64 | FileCheck %s --check-prefixes CHECK,CHECK-APPROX-OPT,CHECK-SQRT-NOOPT
+; RUN: llc < %s -march=nvptx64 -nvptx-prec-sqrtf32=0 | FileCheck %s --check-prefixes CHECK,CHECK-APPROX-OPT,CHECK-SQRT-OPT
+; RUN: llc < %s -march=nvptx64 -nvptx-rsqrt-approx-opt=0 | FileCheck %s --check-prefixes CHECK,CHECK-APPROX-NOOPT,CHECK-SQRT-NOOPT
+;
+; RUN: %if ptxas %{ llc < %s -march=nvptx64 | %ptxas-verify %}
+; RUN: %if ptxas %{ llc < %s -march=nvptx64 -nvptx-prec-sqrtf32=0 | %ptxas-verify %}
+; RUN: %if ptxas %{ llc < %s -march=nvptx64 -nvptx-rsqrt-approx-opt=0 | %ptxas-verify %}
+
+
+; CHECK-LABEL: .func{{.*}}test1
+define float @test1(float %in) local_unnamed_addr {
+; CHECK-APPROX-OPT: rsqrt.approx.f32
+; CHECK-APPROX-NOOPT: sqrt.approx.f32
+; CHECK-APPROX-NOOPT-NEXT: rcp.rn.f32
+  %sqrt = tail call float @llvm.nvvm.sqrt.approx.f(float %in)
+  %rsqrt = fdiv float 1.0, %sqrt
+  ret float %rsqrt
+}
+; CHECK-LABEL: .func{{.*}}test2
+define float @test2(float %in) local_unnamed_addr {
+; CHECK-APPROX-OPT: rsqrt.approx.ftz.f32
+; CHECK-APPROX-NOOPT: sqrt.approx.ftz.f32
+; CHECK-APPROX-NOOPT-NEXT: rcp.rn.f32
+  %sqrt = tail call float @llvm.nvvm.sqrt.approx.ftz.f(float %in)
+  %rsqrt = fdiv float 1.0, %sqrt
+  ret float %rsqrt
+}
+
+; CHECK-LABEL: .func{{.*}}test3
+define float @test3(float %in) local_unnamed_addr {
+; CHECK-SQRT-OPT: rsqrt.approx.f32
+; CHECK-SQRT-NOOPT: sqrt.rn.f32
+; CHECK-SQRT-NOOPT-NEXT: rcp.rn.f32
+  %sqrt = tail call float @llvm.nvvm.sqrt.f(float %in)
+  %rsqrt = fdiv float 1.0, %sqrt
+  ret float %rsqrt
+}
+
+; CHECK-LABEL: .func{{.*}}test4
+define float @test4(float %in) local_unnamed_addr #0 {
+; CHECK-SQRT-OPT: rsqrt.approx.ftz.f32
+; CHECK-SQRT-NOOPT: sqrt.rn.ftz.f32
+; CHECK-SQRT-NOOPT-NEXT: rcp.rn.ftz.f32
+  %sqrt = tail call float @llvm.nvvm.sqrt.f(float %in)
+  %rsqrt = fdiv float 1.0, %sqrt
+  ret float %rsqrt
+}
+
+; CHECK-LABEL: .func{{.*}}test5
+define float @test5(float %in) local_unnamed_addr {
+; CHECK-SQRT-OPT: rsqrt.approx.f32
+; CHECK-SQRT-NOOPT: sqrt.rn.f32
+; CHECK-SQRT-NOOPT-NEXT: rcp.rn.f32
+  %sqrt = tail call float @llvm.sqrt.f32(float %in)
+  %rsqrt = fdiv float 1.0, %sqrt
+  ret float %rsqrt
+}
+
+; CHECK-LABEL: .func{{.*}}test6
+define float @test6(float %in) local_unnamed_addr #0 {
+; CHECK-SQRT-OPT: rsqrt.approx.ftz.f32
+; CHECK-SQRT-NOOPT: sqrt.rn.ftz.f32
+; CHECK-SQRT-NOOPT-NEXT: rcp.rn.ftz.f32
+  %sqrt = tail call float @llvm.sqrt.f32(float %in)
+  %rsqrt = fdiv float 1.0, %sqrt
+  ret float %rsqrt
+}
+
+
+declare float @llvm.nvvm.sqrt.f(float)
+declare float @llvm.nvvm.sqrt.approx.f(float)
+declare float @llvm.nvvm.sqrt.approx.ftz.f(float)
+declare float @llvm.sqrt.f32(float)
+
+attributes #0 = { "denormal-fp-math-f32" = "preserve-sign" }
diff --git a/llvm/test/CodeGen/NVPTX/rsqrt.ll b/llvm/test/CodeGen/NVPTX/rsqrt.ll
new file mode 100644
index 000000000000..c7367245c532
--- /dev/null
+++ b/llvm/test/CodeGen/NVPTX/rsqrt.ll
@@ -0,0 +1,35 @@
+; RUN: llc < %s -march=nvptx64 | FileCheck %s
+; RUN: %if ptxas %{ llc < %s -march=nvptx64 | %ptxas-verify %}
+
+; CHECK-LABEL: .func{{.*}}test1
+define float @test1(float %in) local_unnamed_addr {
+; CHECK: rsqrt.approx.f32
+  %call = call float @llvm.nvvm.rsqrt.approx.f(float %in)
+  ret float %call
+}
+
+; CHECK-LABEL: .func{{.*}}test2
+define double @test2(double %in) local_unnamed_addr {
+; CHECK: rsqrt.approx.f64
+  %call = call double @llvm.nvvm.rsqrt.approx.d(double %in)
+  ret double %call
+}
+
+; CHECK-LABEL: .func{{.*}}test3
+define float @test3(float %in) local_unnamed_addr {
+; CHECK: rsqrt.approx.ftz.f32
+  %call = tail call float @llvm.nvvm.rsqrt.approx.ftz.f(float %in)
+  ret float %call
+}
+
+; CHECK-LABEL: .func{{.*}}test4
+define double @test4(double %in) local_unnamed_addr {
+; CHECK: rsqrt.approx.ftz.f64
+  %call = tail call double @llvm.nvvm.rsqrt.approx.ftz.d(double %in)
+  ret double %call
+}
+
+declare float @llvm.nvvm.rsqrt.approx.ftz.f(float)
+declare double @llvm.nvvm.rsqrt.approx.ftz.d(double)
+declare float @llvm.nvvm.rsqrt.approx.f(float)
+declare double @llvm.nvvm.rsqrt.approx.d(double)
diff --git a/llvm/test/CodeGen/PowerPC/fminimum-fmaximum-f128.ll b/llvm/test/CodeGen/PowerPC/fminimum-fmaximum-f128.ll
new file mode 100644
index 000000000000..6d9eb1337682
--- /dev/null
+++ b/llvm/test/CodeGen/PowerPC/fminimum-fmaximum-f128.ll
@@ -0,0 +1,97 @@
+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --version 2
+; RUN: llc -mtriple=powerpc64le-unknown-linux-gnu -mcpu=pwr9 < %s | FileCheck %s
+
+define fp128 @f128_minimum(fp128 %a, fp128 %b) {
+; CHECK-LABEL: f128_minimum:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    xscmpuqp 0, 2, 3
+; CHECK-NEXT:    vmr 4, 2
+; CHECK-NEXT:    bge 0, .LBB0_8
+; CHECK-NEXT:  # %bb.1: # %entry
+; CHECK-NEXT:    bun 0, .LBB0_9
+; CHECK-NEXT:  .LBB0_2: # %entry
+; CHECK-NEXT:    xststdcqp 0, 2, 4
+; CHECK-NEXT:    bc 4, 2, .LBB0_10
+; CHECK-NEXT:  .LBB0_3: # %entry
+; CHECK-NEXT:    xststdcqp 0, 3, 4
+; CHECK-NEXT:    bc 12, 2, .LBB0_5
+; CHECK-NEXT:  .LBB0_4: # %entry
+; CHECK-NEXT:    vmr 3, 2
+; CHECK-NEXT:  .LBB0_5: # %entry
+; CHECK-NEXT:    addis 3, 2, .LCPI0_1@toc@ha
+; CHECK-NEXT:    addi 3, 3, .LCPI0_1@toc@l
+; CHECK-NEXT:    lxv 34, 0(3)
+; CHECK-NEXT:    xscmpuqp 0, 4, 2
+; CHECK-NEXT:    beq 0, .LBB0_7
+; CHECK-NEXT:  # %bb.6: # %entry
+; CHECK-NEXT:    vmr 3, 4
+; CHECK-NEXT:  .LBB0_7: # %entry
+; CHECK-NEXT:    vmr 2, 3
+; CHECK-NEXT:    blr
+; CHECK-NEXT:  .LBB0_8: # %entry
+; CHECK-NEXT:    vmr 4, 3
+; CHECK-NEXT:    bnu 0, .LBB0_2
+; CHECK-NEXT:  .LBB0_9:
+; CHECK-NEXT:    addis 3, 2, .LCPI0_0@toc@ha
+; CHECK-NEXT:    addi 3, 3, .LCPI0_0@toc@l
+; CHECK-NEXT:    lxv 36, 0(3)
+; CHECK-NEXT:    xststdcqp 0, 2, 4
+; CHECK-NEXT:    bc 12, 2, .LBB0_3
+; CHECK-NEXT:  .LBB0_10: # %entry
+; CHECK-NEXT:    vmr 2, 4
+; CHECK-NEXT:    xststdcqp 0, 3, 4
+; CHECK-NEXT:    bc 4, 2, .LBB0_4
+; CHECK-NEXT:    b .LBB0_5
+entry:
+  %m = call fp128 @llvm.minimum.f128(fp128 %a, fp128 %b)
+  ret fp128 %m
+}
+
+define fp128 @f128_maximum(fp128 %a, fp128 %b) {
+; CHECK-LABEL: f128_maximum:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    xscmpuqp 0, 2, 3
+; CHECK-NEXT:    vmr 4, 2
+; CHECK-NEXT:    ble 0, .LBB1_8
+; CHECK-NEXT:  # %bb.1: # %entry
+; CHECK-NEXT:    bun 0, .LBB1_9
+; CHECK-NEXT:  .LBB1_2: # %entry
+; CHECK-NEXT:    xststdcqp 0, 2, 8
+; CHECK-NEXT:    bc 4, 2, .LBB1_10
+; CHECK-NEXT:  .LBB1_3: # %entry
+; CHECK-NEXT:    xststdcqp 0, 3, 8
+; CHECK-NEXT:    bc 12, 2, .LBB1_5
+; CHECK-NEXT:  .LBB1_4: # %entry
+; CHECK-NEXT:    vmr 3, 2
+; CHECK-NEXT:  .LBB1_5: # %entry
+; CHECK-NEXT:    addis 3, 2, .LCPI1_1@toc@ha
+; CHECK-NEXT:    addi 3, 3, .LCPI1_1@toc@l
+; CHECK-NEXT:    lxv 34, 0(3)
+; CHECK-NEXT:    xscmpuqp 0, 4, 2
+; CHECK-NEXT:    beq 0, .LBB1_7
+; CHECK-NEXT:  # %bb.6: # %entry
+; CHECK-NEXT:    vmr 3, 4
+; CHECK-NEXT:  .LBB1_7: # %entry
+; CHECK-NEXT:    vmr 2, 3
+; CHECK-NEXT:    blr
+; CHECK-NEXT:  .LBB1_8: # %entry
+; CHECK-NEXT:    vmr 4, 3
+; CHECK-NEXT:    bnu 0, .LBB1_2
+; CHECK-NEXT:  .LBB1_9:
+; CHECK-NEXT:    addis 3, 2, .LCPI1_0@toc@ha
+; CHECK-NEXT:    addi 3, 3, .LCPI1_0@toc@l
+; CHECK-NEXT:    lxv 36, 0(3)
+; CHECK-NEXT:    xststdcqp 0, 2, 8
+; CHECK-NEXT:    bc 12, 2, .LBB1_3
+; CHECK-NEXT:  .LBB1_10: # %entry
+; CHECK-NEXT:    vmr 2, 4
+; CHECK-NEXT:    xststdcqp 0, 3, 8
+; CHECK-NEXT:    bc 4, 2, .LBB1_4
+; CHECK-NEXT:    b .LBB1_5
+entry:
+  %m = call fp128 @llvm.maximum.f128(fp128 %a, fp128 %b)
+  ret fp128 %m
+}
+
+declare fp128 @llvm.minimum.f128(fp128, fp128)
+declare fp128 @llvm.maximum.f128(fp128, fp128)
diff --git a/llvm/test/CodeGen/PowerPC/fminimum-fmaximum.ll b/llvm/test/CodeGen/PowerPC/fminimum-fmaximum.ll
new file mode 100644
index 000000000000..c33875dbfee4
--- /dev/null
+++ b/llvm/test/CodeGen/PowerPC/fminimum-fmaximum.ll
@@ -0,0 +1,847 @@
+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --version 2
+; RUN: llc -mtriple=powerpc64le-unknown-linux-gnu -mcpu=pwr8 -mattr=-vsx < %s | FileCheck %s --check-prefix=NOVSX
+; RUN: llc -mtriple=powerpc64le-unknown-linux-gnu -mcpu=pwr8 < %s | FileCheck %s --check-prefix=VSX
+; RUN: llc -mtriple=powerpc64-ibm-aix -mcpu=pwr8 < %s | FileCheck %s --check-prefix=AIX
+
+define float @f32_minimum(float %a, float %b) {
+; NOVSX-LABEL: f32_minimum:
+; NOVSX:       # %bb.0: # %entry
+; NOVSX-NEXT:    fcmpu 0, 1, 2
+; NOVSX-NEXT:    fmr 0, 1
+; NOVSX-NEXT:    stfs 2, -8(1)
+; NOVSX-NEXT:    stfs 1, -4(1)
+; NOVSX-NEXT:    bc 12, 0, .LBB0_2
+; NOVSX-NEXT:  # %bb.1: # %entry
+; NOVSX-NEXT:    fmr 0, 2
+; NOVSX-NEXT:  .LBB0_2: # %entry
+; NOVSX-NEXT:    lwz 3, -4(1)
+; NOVSX-NEXT:    bc 4, 3, .LBB0_4
+; NOVSX-NEXT:  # %bb.3:
+; NOVSX-NEXT:    addis 4, 2, .LCPI0_0@toc@ha
+; NOVSX-NEXT:    lfs 0, .LCPI0_0@toc@l(4)
+; NOVSX-NEXT:  .LBB0_4: # %entry
+; NOVSX-NEXT:    xoris 3, 3, 32768
+; NOVSX-NEXT:    lwz 4, -8(1)
+; NOVSX-NEXT:    cmplwi 3, 0
+; NOVSX-NEXT:    bc 12, 2, .LBB0_6
+; NOVSX-NEXT:  # %bb.5: # %entry
+; NOVSX-NEXT:    fmr 1, 0
+; NOVSX-NEXT:  .LBB0_6: # %entry
+; NOVSX-NEXT:    xoris 3, 4, 32768
+; NOVSX-NEXT:    cmplwi 3, 0
+; NOVSX-NEXT:    bc 12, 2, .LBB0_8
+; NOVSX-NEXT:  # %bb.7: # %entry
+; NOVSX-NEXT:    fmr 2, 1
+; NOVSX-NEXT:  .LBB0_8: # %entry
+; NOVSX-NEXT:    addis 3, 2, .LCPI0_1@toc@ha
+; NOVSX-NEXT:    lfs 1, .LCPI0_1@toc@l(3)
+; NOVSX-NEXT:    fcmpu 0, 0, 1
+; NOVSX-NEXT:    bc 12, 2, .LBB0_10
+; NOVSX-NEXT:  # %bb.9: # %entry
+; NOVSX-NEXT:    fmr 2, 0
+; NOVSX-NEXT:  .LBB0_10: # %entry
+; NOVSX-NEXT:    fmr 1, 2
+; NOVSX-NEXT:    blr
+;
+; VSX-LABEL: f32_minimum:
+; VSX:       # %bb.0: # %entry
+; VSX-NEXT:    xscvdpspn 0, 1
+; VSX-NEXT:    fcmpu 0, 1, 2
+; VSX-NEXT:    xscvdpspn 3, 2
+; VSX-NEXT:    mffprwz 3, 0
+; VSX-NEXT:    bc 12, 3, .LBB0_2
+; VSX-NEXT:  # %bb.1: # %entry
+; VSX-NEXT:    xsmindp 0, 1, 2
+; VSX-NEXT:    b .LBB0_3
+; VSX-NEXT:  .LBB0_2:
+; VSX-NEXT:    addis 4, 2, .LCPI0_0@toc@ha
+; VSX-NEXT:    lfs 0, .LCPI0_0@toc@l(4)
+; VSX-NEXT:  .LBB0_3: # %entry
+; VSX-NEXT:    xoris 3, 3, 32768
+; VSX-NEXT:    mffprwz 4, 3
+; VSX-NEXT:    cmplwi 3, 0
+; VSX-NEXT:    bc 12, 2, .LBB0_5
+; VSX-NEXT:  # %bb.4: # %entry
+; VSX-NEXT:    fmr 1, 0
+; VSX-NEXT:  .LBB0_5: # %entry
+; VSX-NEXT:    xoris 3, 4, 32768
+; VSX-NEXT:    cmplwi 3, 0
+; VSX-NEXT:    bc 12, 2, .LBB0_7
+; VSX-NEXT:  # %bb.6: # %entry
+; VSX-NEXT:    fmr 2, 1
+; VSX-NEXT:  .LBB0_7: # %entry
+; VSX-NEXT:    xxlxor 1, 1, 1
+; VSX-NEXT:    fcmpu 0, 0, 1
+; VSX-NEXT:    bc 12, 2, .LBB0_9
+; VSX-NEXT:  # %bb.8: # %entry
+; VSX-NEXT:    fmr 2, 0
+; VSX-NEXT:  .LBB0_9: # %entry
+; VSX-NEXT:    fmr 1, 2
+; VSX-NEXT:    blr
+;
+; AIX-LABEL: f32_minimum:
+; AIX:       # %bb.0: # %entry
+; AIX-NEXT:    xscvdpspn 0, 1
+; AIX-NEXT:    fcmpu 0, 1, 2
+; AIX-NEXT:    xscvdpspn 3, 2
+; AIX-NEXT:    mffprwz 3, 0
+; AIX-NEXT:    bc 12, 3, L..BB0_2
+; AIX-NEXT:  # %bb.1: # %entry
+; AIX-NEXT:    xsmindp 0, 1, 2
+; AIX-NEXT:    b L..BB0_3
+; AIX-NEXT:  L..BB0_2:
+; AIX-NEXT:    ld 4, L..C0(2) # %const.0
+; AIX-NEXT:    lfs 0, 0(4)
+; AIX-NEXT:  L..BB0_3: # %entry
+; AIX-NEXT:    xoris 3, 3, 32768
+; AIX-NEXT:    mffprwz 4, 3
+; AIX-NEXT:    cmplwi 3, 0
+; AIX-NEXT:    bc 12, 2, L..BB0_5
+; AIX-NEXT:  # %bb.4: # %entry
+; AIX-NEXT:    fmr 1, 0
+; AIX-NEXT:  L..BB0_5: # %entry
+; AIX-NEXT:    xoris 3, 4, 32768
+; AIX-NEXT:    cmplwi 3, 0
+; AIX-NEXT:    bc 12, 2, L..BB0_7
+; AIX-NEXT:  # %bb.6: # %entry
+; AIX-NEXT:    fmr 2, 1
+; AIX-NEXT:  L..BB0_7: # %entry
+; AIX-NEXT:    xxlxor 1, 1, 1
+; AIX-NEXT:    fcmpu 0, 0, 1
+; AIX-NEXT:    bc 12, 2, L..BB0_9
+; AIX-NEXT:  # %bb.8: # %entry
+; AIX-NEXT:    fmr 2, 0
+; AIX-NEXT:  L..BB0_9: # %entry
+; AIX-NEXT:    fmr 1, 2
+; AIX-NEXT:    blr
+entry:
+  %m = call float @llvm.minimum.f32(float %a, float %b)
+  ret float %m
+}
+
+define float @f32_maximum(float %a, float %b) {
+; NOVSX-LABEL: f32_maximum:
+; NOVSX:       # %bb.0: # %entry
+; NOVSX-NEXT:    fcmpu 0, 1, 2
+; NOVSX-NEXT:    fmr 0, 1
+; NOVSX-NEXT:    stfs 2, -8(1)
+; NOVSX-NEXT:    stfs 1, -4(1)
+; NOVSX-NEXT:    bc 12, 1, .LBB1_2
+; NOVSX-NEXT:  # %bb.1: # %entry
+; NOVSX-NEXT:    fmr 0, 2
+; NOVSX-NEXT:  .LBB1_2: # %entry
+; NOVSX-NEXT:    lwz 3, -4(1)
+; NOVSX-NEXT:    bc 4, 3, .LBB1_4
+; NOVSX-NEXT:  # %bb.3:
+; NOVSX-NEXT:    addis 4, 2, .LCPI1_0@toc@ha
+; NOVSX-NEXT:    lfs 0, .LCPI1_0@toc@l(4)
+; NOVSX-NEXT:  .LBB1_4: # %entry
+; NOVSX-NEXT:    cmpwi 3, 0
+; NOVSX-NEXT:    lwz 4, -8(1)
+; NOVSX-NEXT:    bc 12, 2, .LBB1_6
+; NOVSX-NEXT:  # %bb.5: # %entry
+; NOVSX-NEXT:    fmr 1, 0
+; NOVSX-NEXT:  .LBB1_6: # %entry
+; NOVSX-NEXT:    cmpwi 4, 0
+; NOVSX-NEXT:    bc 12, 2, .LBB1_8
+; NOVSX-NEXT:  # %bb.7: # %entry
+; NOVSX-NEXT:    fmr 2, 1
+; NOVSX-NEXT:  .LBB1_8: # %entry
+; NOVSX-NEXT:    addis 3, 2, .LCPI1_1@toc@ha
+; NOVSX-NEXT:    lfs 1, .LCPI1_1@toc@l(3)
+; NOVSX-NEXT:    fcmpu 0, 0, 1
+; NOVSX-NEXT:    bc 12, 2, .LBB1_10
+; NOVSX-NEXT:  # %bb.9: # %entry
+; NOVSX-NEXT:    fmr 2, 0
+; NOVSX-NEXT:  .LBB1_10: # %entry
+; NOVSX-NEXT:    fmr 1, 2
+; NOVSX-NEXT:    blr
+;
+; VSX-LABEL: f32_maximum:
+; VSX:       # %bb.0: # %entry
+; VSX-NEXT:    xscvdpspn 0, 1
+; VSX-NEXT:    fcmpu 0, 1, 2
+; VSX-NEXT:    xscvdpspn 3, 2
+; VSX-NEXT:    mffprwz 3, 0
+; VSX-NEXT:    bc 12, 3, .LBB1_2
+; VSX-NEXT:  # %bb.1: # %entry
+; VSX-NEXT:    xsmaxdp 0, 1, 2
+; VSX-NEXT:    b .LBB1_3
+; VSX-NEXT:  .LBB1_2:
+; VSX-NEXT:    addis 4, 2, .LCPI1_0@toc@ha
+; VSX-NEXT:    lfs 0, .LCPI1_0@toc@l(4)
+; VSX-NEXT:  .LBB1_3: # %entry
+; VSX-NEXT:    mffprwz 4, 3
+; VSX-NEXT:    cmpwi 3, 0
+; VSX-NEXT:    bc 12, 2, .LBB1_5
+; VSX-NEXT:  # %bb.4: # %entry
+; VSX-NEXT:    fmr 1, 0
+; VSX-NEXT:  .LBB1_5: # %entry
+; VSX-NEXT:    cmpwi 4, 0
+; VSX-NEXT:    bc 12, 2, .LBB1_7
+; VSX-NEXT:  # %bb.6: # %entry
+; VSX-NEXT:    fmr 2, 1
+; VSX-NEXT:  .LBB1_7: # %entry
+; VSX-NEXT:    xxlxor 1, 1, 1
+; VSX-NEXT:    fcmpu 0, 0, 1
+; VSX-NEXT:    bc 12, 2, .LBB1_9
+; VSX-NEXT:  # %bb.8: # %entry
+; VSX-NEXT:    fmr 2, 0
+; VSX-NEXT:  .LBB1_9: # %entry
+; VSX-NEXT:    fmr 1, 2
+; VSX-NEXT:    blr
+;
+; AIX-LABEL: f32_maximum:
+; AIX:       # %bb.0: # %entry
+; AIX-NEXT:    xscvdpspn 0, 1
+; AIX-NEXT:    fcmpu 0, 1, 2
+; AIX-NEXT:    xscvdpspn 3, 2
+; AIX-NEXT:    mffprwz 3, 0
+; AIX-NEXT:    bc 12, 3, L..BB1_2
+; AIX-NEXT:  # %bb.1: # %entry
+; AIX-NEXT:    xsmaxdp 0, 1, 2
+; AIX-NEXT:    b L..BB1_3
+; AIX-NEXT:  L..BB1_2:
+; AIX-NEXT:    ld 4, L..C1(2) # %const.0
+; AIX-NEXT:    lfs 0, 0(4)
+; AIX-NEXT:  L..BB1_3: # %entry
+; AIX-NEXT:    mffprwz 4, 3
+; AIX-NEXT:    cmpwi 3, 0
+; AIX-NEXT:    bc 12, 2, L..BB1_5
+; AIX-NEXT:  # %bb.4: # %entry
+; AIX-NEXT:    fmr 1, 0
+; AIX-NEXT:  L..BB1_5: # %entry
+; AIX-NEXT:    cmpwi 4, 0
+; AIX-NEXT:    bc 12, 2, L..BB1_7
+; AIX-NEXT:  # %bb.6: # %entry
+; AIX-NEXT:    fmr 2, 1
+; AIX-NEXT:  L..BB1_7: # %entry
+; AIX-NEXT:    xxlxor 1, 1, 1
+; AIX-NEXT:    fcmpu 0, 0, 1
+; AIX-NEXT:    bc 12, 2, L..BB1_9
+; AIX-NEXT:  # %bb.8: # %entry
+; AIX-NEXT:    fmr 2, 0
+; AIX-NEXT:  L..BB1_9: # %entry
+; AIX-NEXT:    fmr 1, 2
+; AIX-NEXT:    blr
+entry:
+  %m = call float @llvm.maximum.f32(float %a, float %b)
+  ret float %m
+}
+
+define double @f64_minimum(double %a, double %b) {
+; NOVSX-LABEL: f64_minimum:
+; NOVSX:       # %bb.0: # %entry
+; NOVSX-NEXT:    fcmpu 0, 1, 2
+; NOVSX-NEXT:    fmr 0, 1
+; NOVSX-NEXT:    stfd 2, -16(1)
+; NOVSX-NEXT:    stfd 1, -8(1)
+; NOVSX-NEXT:    bc 12, 0, .LBB2_2
+; NOVSX-NEXT:  # %bb.1: # %entry
+; NOVSX-NEXT:    fmr 0, 2
+; NOVSX-NEXT:  .LBB2_2: # %entry
+; NOVSX-NEXT:    ld 3, -8(1)
+; NOVSX-NEXT:    bc 4, 3, .LBB2_4
+; NOVSX-NEXT:  # %bb.3:
+; NOVSX-NEXT:    addis 4, 2, .LCPI2_0@toc@ha
+; NOVSX-NEXT:    lfs 0, .LCPI2_0@toc@l(4)
+; NOVSX-NEXT:  .LBB2_4: # %entry
+; NOVSX-NEXT:    li 5, 1
+; NOVSX-NEXT:    ld 4, -16(1)
+; NOVSX-NEXT:    rldic 5, 5, 63, 0
+; NOVSX-NEXT:    cmpd 3, 5
+; NOVSX-NEXT:    bc 12, 2, .LBB2_6
+; NOVSX-NEXT:  # %bb.5: # %entry
+; NOVSX-NEXT:    fmr 1, 0
+; NOVSX-NEXT:  .LBB2_6: # %entry
+; NOVSX-NEXT:    cmpd 4, 5
+; NOVSX-NEXT:    bc 12, 2, .LBB2_8
+; NOVSX-NEXT:  # %bb.7: # %entry
+; NOVSX-NEXT:    fmr 2, 1
+; NOVSX-NEXT:  .LBB2_8: # %entry
+; NOVSX-NEXT:    addis 3, 2, .LCPI2_1@toc@ha
+; NOVSX-NEXT:    lfs 1, .LCPI2_1@toc@l(3)
+; NOVSX-NEXT:    fcmpu 0, 0, 1
+; NOVSX-NEXT:    bc 12, 2, .LBB2_10
+; NOVSX-NEXT:  # %bb.9: # %entry
+; NOVSX-NEXT:    fmr 2, 0
+; NOVSX-NEXT:  .LBB2_10: # %entry
+; NOVSX-NEXT:    fmr 1, 2
+; NOVSX-NEXT:    blr
+;
+; VSX-LABEL: f64_minimum:
+; VSX:       # %bb.0: # %entry
+; VSX-NEXT:    fcmpu 0, 1, 2
+; VSX-NEXT:    mffprd 3, 1
+; VSX-NEXT:    bc 12, 3, .LBB2_2
+; VSX-NEXT:  # %bb.1: # %entry
+; VSX-NEXT:    xsmindp 0, 1, 2
+; VSX-NEXT:    b .LBB2_3
+; VSX-NEXT:  .LBB2_2:
+; VSX-NEXT:    addis 4, 2, .LCPI2_0@toc@ha
+; VSX-NEXT:    lfs 0, .LCPI2_0@toc@l(4)
+; VSX-NEXT:  .LBB2_3: # %entry
+; VSX-NEXT:    li 5, 1
+; VSX-NEXT:    mffprd 4, 2
+; VSX-NEXT:    rldic 5, 5, 63, 0
+; VSX-NEXT:    cmpd 3, 5
+; VSX-NEXT:    bc 12, 2, .LBB2_5
+; VSX-NEXT:  # %bb.4: # %entry
+; VSX-NEXT:    fmr 1, 0
+; VSX-NEXT:  .LBB2_5: # %entry
+; VSX-NEXT:    cmpd 4, 5
+; VSX-NEXT:    bc 12, 2, .LBB2_7
+; VSX-NEXT:  # %bb.6: # %entry
+; VSX-NEXT:    fmr 2, 1
+; VSX-NEXT:  .LBB2_7: # %entry
+; VSX-NEXT:    xxlxor 1, 1, 1
+; VSX-NEXT:    fcmpu 0, 0, 1
+; VSX-NEXT:    bc 12, 2, .LBB2_9
+; VSX-NEXT:  # %bb.8: # %entry
+; VSX-NEXT:    fmr 2, 0
+; VSX-NEXT:  .LBB2_9: # %entry
+; VSX-NEXT:    fmr 1, 2
+; VSX-NEXT:    blr
+;
+; AIX-LABEL: f64_minimum:
+; AIX:       # %bb.0: # %entry
+; AIX-NEXT:    fcmpu 0, 1, 2
+; AIX-NEXT:    mffprd 3, 1
+; AIX-NEXT:    bc 12, 3, L..BB2_2
+; AIX-NEXT:  # %bb.1: # %entry
+; AIX-NEXT:    xsmindp 0, 1, 2
+; AIX-NEXT:    b L..BB2_3
+; AIX-NEXT:  L..BB2_2:
+; AIX-NEXT:    ld 4, L..C2(2) # %const.0
+; AIX-NEXT:    lfs 0, 0(4)
+; AIX-NEXT:  L..BB2_3: # %entry
+; AIX-NEXT:    li 5, 1
+; AIX-NEXT:    mffprd 4, 2
+; AIX-NEXT:    rldic 5, 5, 63, 0
+; AIX-NEXT:    cmpd 3, 5
+; AIX-NEXT:    bc 12, 2, L..BB2_5
+; AIX-NEXT:  # %bb.4: # %entry
+; AIX-NEXT:    fmr 1, 0
+; AIX-NEXT:  L..BB2_5: # %entry
+; AIX-NEXT:    cmpd 4, 5
+; AIX-NEXT:    bc 12, 2, L..BB2_7
+; AIX-NEXT:  # %bb.6: # %entry
+; AIX-NEXT:    fmr 2, 1
+; AIX-NEXT:  L..BB2_7: # %entry
+; AIX-NEXT:    xxlxor 1, 1, 1
+; AIX-NEXT:    fcmpu 0, 0, 1
+; AIX-NEXT:    bc 12, 2, L..BB2_9
+; AIX-NEXT:  # %bb.8: # %entry
+; AIX-NEXT:    fmr 2, 0
+; AIX-NEXT:  L..BB2_9: # %entry
+; AIX-NEXT:    fmr 1, 2
+; AIX-NEXT:    blr
+entry:
+  %m = call double @llvm.minimum.f64(double %a, double %b)
+  ret double %m
+}
+
+define double @f64_maximum(double %a, double %b) {
+; NOVSX-LABEL: f64_maximum:
+; NOVSX:       # %bb.0: # %entry
+; NOVSX-NEXT:    fcmpu 0, 1, 2
+; NOVSX-NEXT:    fmr 0, 1
+; NOVSX-NEXT:    stfd 2, -16(1)
+; NOVSX-NEXT:    stfd 1, -8(1)
+; NOVSX-NEXT:    bc 12, 1, .LBB3_2
+; NOVSX-NEXT:  # %bb.1: # %entry
+; NOVSX-NEXT:    fmr 0, 2
+; NOVSX-NEXT:  .LBB3_2: # %entry
+; NOVSX-NEXT:    ld 3, -8(1)
+; NOVSX-NEXT:    bc 4, 3, .LBB3_4
+; NOVSX-NEXT:  # %bb.3:
+; NOVSX-NEXT:    addis 4, 2, .LCPI3_0@toc@ha
+; NOVSX-NEXT:    lfs 0, .LCPI3_0@toc@l(4)
+; NOVSX-NEXT:  .LBB3_4: # %entry
+; NOVSX-NEXT:    cmpdi 3, 0
+; NOVSX-NEXT:    ld 4, -16(1)
+; NOVSX-NEXT:    bc 12, 2, .LBB3_6
+; NOVSX-NEXT:  # %bb.5: # %entry
+; NOVSX-NEXT:    fmr 1, 0
+; NOVSX-NEXT:  .LBB3_6: # %entry
+; NOVSX-NEXT:    cmpdi 4, 0
+; NOVSX-NEXT:    bc 12, 2, .LBB3_8
+; NOVSX-NEXT:  # %bb.7: # %entry
+; NOVSX-NEXT:    fmr 2, 1
+; NOVSX-NEXT:  .LBB3_8: # %entry
+; NOVSX-NEXT:    addis 3, 2, .LCPI3_1@toc@ha
+; NOVSX-NEXT:    lfs 1, .LCPI3_1@toc@l(3)
+; NOVSX-NEXT:    fcmpu 0, 0, 1
+; NOVSX-NEXT:    bc 12, 2, .LBB3_10
+; NOVSX-NEXT:  # %bb.9: # %entry
+; NOVSX-NEXT:    fmr 2, 0
+; NOVSX-NEXT:  .LBB3_10: # %entry
+; NOVSX-NEXT:    fmr 1, 2
+; NOVSX-NEXT:    blr
+;
+; VSX-LABEL: f64_maximum:
+; VSX:       # %bb.0: # %entry
+; VSX-NEXT:    fcmpu 0, 1, 2
+; VSX-NEXT:    mffprd 3, 1
+; VSX-NEXT:    bc 12, 3, .LBB3_2
+; VSX-NEXT:  # %bb.1: # %entry
+; VSX-NEXT:    xsmaxdp 0, 1, 2
+; VSX-NEXT:    b .LBB3_3
+; VSX-NEXT:  .LBB3_2:
+; VSX-NEXT:    addis 4, 2, .LCPI3_0@toc@ha
+; VSX-NEXT:    lfs 0, .LCPI3_0@toc@l(4)
+; VSX-NEXT:  .LBB3_3: # %entry
+; VSX-NEXT:    mffprd 4, 2
+; VSX-NEXT:    cmpdi 3, 0
+; VSX-NEXT:    bc 12, 2, .LBB3_5
+; VSX-NEXT:  # %bb.4: # %entry
+; VSX-NEXT:    fmr 1, 0
+; VSX-NEXT:  .LBB3_5: # %entry
+; VSX-NEXT:    cmpdi 4, 0
+; VSX-NEXT:    bc 12, 2, .LBB3_7
+; VSX-NEXT:  # %bb.6: # %entry
+; VSX-NEXT:    fmr 2, 1
+; VSX-NEXT:  .LBB3_7: # %entry
+; VSX-NEXT:    xxlxor 1, 1, 1
+; VSX-NEXT:    fcmpu 0, 0, 1
+; VSX-NEXT:    bc 12, 2, .LBB3_9
+; VSX-NEXT:  # %bb.8: # %entry
+; VSX-NEXT:    fmr 2, 0
+; VSX-NEXT:  .LBB3_9: # %entry
+; VSX-NEXT:    fmr 1, 2
+; VSX-NEXT:    blr
+;
+; AIX-LABEL: f64_maximum:
+; AIX:       # %bb.0: # %entry
+; AIX-NEXT:    fcmpu 0, 1, 2
+; AIX-NEXT:    mffprd 3, 1
+; AIX-NEXT:    bc 12, 3, L..BB3_2
+; AIX-NEXT:  # %bb.1: # %entry
+; AIX-NEXT:    xsmaxdp 0, 1, 2
+; AIX-NEXT:    b L..BB3_3
+; AIX-NEXT:  L..BB3_2:
+; AIX-NEXT:    ld 4, L..C3(2) # %const.0
+; AIX-NEXT:    lfs 0, 0(4)
+; AIX-NEXT:  L..BB3_3: # %entry
+; AIX-NEXT:    mffprd 4, 2
+; AIX-NEXT:    cmpdi 3, 0
+; AIX-NEXT:    bc 12, 2, L..BB3_5
+; AIX-NEXT:  # %bb.4: # %entry
+; AIX-NEXT:    fmr 1, 0
+; AIX-NEXT:  L..BB3_5: # %entry
+; AIX-NEXT:    cmpdi 4, 0
+; AIX-NEXT:    bc 12, 2, L..BB3_7
+; AIX-NEXT:  # %bb.6: # %entry
+; AIX-NEXT:    fmr 2, 1
+; AIX-NEXT:  L..BB3_7: # %entry
+; AIX-NEXT:    xxlxor 1, 1, 1
+; AIX-NEXT:    fcmpu 0, 0, 1
+; AIX-NEXT:    bc 12, 2, L..BB3_9
+; AIX-NEXT:  # %bb.8: # %entry
+; AIX-NEXT:    fmr 2, 0
+; AIX-NEXT:  L..BB3_9: # %entry
+; AIX-NEXT:    fmr 1, 2
+; AIX-NEXT:    blr
+entry:
+  %m = call double @llvm.maximum.f64(double %a, double %b)
+  ret double %m
+}
+
+define <4 x float> @v4f32_minimum(<4 x float> %a, <4 x float> %b) {
+; NOVSX-LABEL: v4f32_minimum:
+; NOVSX:       # %bb.0: # %entry
+; NOVSX-NEXT:    vcmpeqfp 0, 3, 3
+; NOVSX-NEXT:    vcmpeqfp 1, 2, 2
+; NOVSX-NEXT:    addis 3, 2, .LCPI4_0@toc@ha
+; NOVSX-NEXT:    addi 3, 3, .LCPI4_0@toc@l
+; NOVSX-NEXT:    vnot 0, 0
+; NOVSX-NEXT:    vnot 1, 1
+; NOVSX-NEXT:    vspltisb 4, -1
+; NOVSX-NEXT:    vcmpgtfp 5, 3, 2
+; NOVSX-NEXT:    vslw 4, 4, 4
+; NOVSX-NEXT:    vor 0, 1, 0
+; NOVSX-NEXT:    lvx 1, 0, 3
+; NOVSX-NEXT:    vsel 5, 3, 2, 5
+; NOVSX-NEXT:    vsel 5, 5, 1, 0
+; NOVSX-NEXT:    vcmpequw 0, 2, 4
+; NOVSX-NEXT:    vcmpequw 4, 3, 4
+; NOVSX-NEXT:    vsel 2, 5, 2, 0
+; NOVSX-NEXT:    vsel 2, 2, 3, 4
+; NOVSX-NEXT:    vxor 3, 3, 3
+; NOVSX-NEXT:    vcmpeqfp 3, 5, 3
+; NOVSX-NEXT:    vsel 2, 5, 2, 3
+; NOVSX-NEXT:    blr
+;
+; VSX-LABEL: v4f32_minimum:
+; VSX:       # %bb.0: # %entry
+; VSX-NEXT:    xvcmpeqsp 1, 35, 35
+; VSX-NEXT:    xvcmpeqsp 2, 34, 34
+; VSX-NEXT:    addis 3, 2, .LCPI4_0@toc@ha
+; VSX-NEXT:    xxleqv 36, 36, 36
+; VSX-NEXT:    xvminsp 0, 34, 35
+; VSX-NEXT:    vslw 4, 4, 4
+; VSX-NEXT:    addi 3, 3, .LCPI4_0@toc@l
+; VSX-NEXT:    xxlnor 1, 1, 1
+; VSX-NEXT:    xxlnor 2, 2, 2
+; VSX-NEXT:    vcmpequw 5, 2, 4
+; VSX-NEXT:    xxlor 1, 2, 1
+; VSX-NEXT:    lxvd2x 2, 0, 3
+; VSX-NEXT:    xxsel 0, 0, 2, 1
+; VSX-NEXT:    xxlxor 2, 2, 2
+; VSX-NEXT:    xvcmpeqsp 2, 0, 2
+; VSX-NEXT:    xxsel 1, 0, 34, 37
+; VSX-NEXT:    vcmpequw 2, 3, 4
+; VSX-NEXT:    xxsel 1, 1, 35, 34
+; VSX-NEXT:    xxsel 34, 0, 1, 2
+; VSX-NEXT:    blr
+;
+; AIX-LABEL: v4f32_minimum:
+; AIX:       # %bb.0: # %entry
+; AIX-NEXT:    xvcmpeqsp 1, 35, 35
+; AIX-NEXT:    xvcmpeqsp 2, 34, 34
+; AIX-NEXT:    ld 3, L..C4(2) # %const.0
+; AIX-NEXT:    xxleqv 36, 36, 36
+; AIX-NEXT:    xvminsp 0, 34, 35
+; AIX-NEXT:    vslw 4, 4, 4
+; AIX-NEXT:    xxlnor 1, 1, 1
+; AIX-NEXT:    xxlnor 2, 2, 2
+; AIX-NEXT:    vcmpequw 5, 2, 4
+; AIX-NEXT:    xxlor 1, 2, 1
+; AIX-NEXT:    lxvw4x 2, 0, 3
+; AIX-NEXT:    xxsel 0, 0, 2, 1
+; AIX-NEXT:    xxlxor 2, 2, 2
+; AIX-NEXT:    xvcmpeqsp 2, 0, 2
+; AIX-NEXT:    xxsel 1, 0, 34, 37
+; AIX-NEXT:    vcmpequw 2, 3, 4
+; AIX-NEXT:    xxsel 1, 1, 35, 34
+; AIX-NEXT:    xxsel 34, 0, 1, 2
+; AIX-NEXT:    blr
+entry:
+  %m = call <4 x float> @llvm.minimum.v4f32(<4 x float> %a, <4 x float> %b)
+  ret <4 x float> %m
+}
+
+define <4 x float> @v4f32_maximum(<4 x float> %a, <4 x float> %b) {
+; NOVSX-LABEL: v4f32_maximum:
+; NOVSX:       # %bb.0: # %entry
+; NOVSX-NEXT:    vcmpeqfp 5, 3, 3
+; NOVSX-NEXT:    vcmpeqfp 0, 2, 2
+; NOVSX-NEXT:    addis 3, 2, .LCPI5_0@toc@ha
+; NOVSX-NEXT:    addi 3, 3, .LCPI5_0@toc@l
+; NOVSX-NEXT:    vnot 5, 5
+; NOVSX-NEXT:    vnot 0, 0
+; NOVSX-NEXT:    vcmpgtfp 4, 2, 3
+; NOVSX-NEXT:    vor 5, 0, 5
+; NOVSX-NEXT:    lvx 0, 0, 3
+; NOVSX-NEXT:    vsel 4, 3, 2, 4
+; NOVSX-NEXT:    vsel 4, 4, 0, 5
+; NOVSX-NEXT:    vxor 5, 5, 5
+; NOVSX-NEXT:    vcmpequw 0, 2, 5
+; NOVSX-NEXT:    vsel 2, 4, 2, 0
+; NOVSX-NEXT:    vcmpequw 0, 3, 5
+; NOVSX-NEXT:    vsel 2, 2, 3, 0
+; NOVSX-NEXT:    vcmpeqfp 3, 4, 5
+; NOVSX-NEXT:    vsel 2, 4, 2, 3
+; NOVSX-NEXT:    blr
+;
+; VSX-LABEL: v4f32_maximum:
+; VSX:       # %bb.0: # %entry
+; VSX-NEXT:    xvcmpeqsp 1, 35, 35
+; VSX-NEXT:    xvcmpeqsp 2, 34, 34
+; VSX-NEXT:    addis 3, 2, .LCPI5_0@toc@ha
+; VSX-NEXT:    addi 3, 3, .LCPI5_0@toc@l
+; VSX-NEXT:    xxlnor 1, 1, 1
+; VSX-NEXT:    xxlnor 2, 2, 2
+; VSX-NEXT:    xvmaxsp 0, 34, 35
+; VSX-NEXT:    xxlxor 36, 36, 36
+; VSX-NEXT:    vcmpequw 5, 2, 4
+; VSX-NEXT:    xxlor 1, 2, 1
+; VSX-NEXT:    lxvd2x 2, 0, 3
+; VSX-NEXT:    xxsel 0, 0, 2, 1
+; VSX-NEXT:    xvcmpeqsp 2, 0, 36
+; VSX-NEXT:    xxsel 1, 0, 34, 37
+; VSX-NEXT:    vcmpequw 2, 3, 4
+; VSX-NEXT:    xxsel 1, 1, 35, 34
+; VSX-NEXT:    xxsel 34, 0, 1, 2
+; VSX-NEXT:    blr
+;
+; AIX-LABEL: v4f32_maximum:
+; AIX:       # %bb.0: # %entry
+; AIX-NEXT:    xvcmpeqsp 1, 35, 35
+; AIX-NEXT:    xvcmpeqsp 2, 34, 34
+; AIX-NEXT:    ld 3, L..C5(2) # %const.0
+; AIX-NEXT:    xvmaxsp 0, 34, 35
+; AIX-NEXT:    xxlxor 36, 36, 36
+; AIX-NEXT:    xxlnor 1, 1, 1
+; AIX-NEXT:    xxlnor 2, 2, 2
+; AIX-NEXT:    vcmpequw 5, 2, 4
+; AIX-NEXT:    xxlor 1, 2, 1
+; AIX-NEXT:    lxvw4x 2, 0, 3
+; AIX-NEXT:    xxsel 0, 0, 2, 1
+; AIX-NEXT:    xvcmpeqsp 2, 0, 36
+; AIX-NEXT:    xxsel 1, 0, 34, 37
+; AIX-NEXT:    vcmpequw 2, 3, 4
+; AIX-NEXT:    xxsel 1, 1, 35, 34
+; AIX-NEXT:    xxsel 34, 0, 1, 2
+; AIX-NEXT:    blr
+entry:
+  %m = call <4 x float> @llvm.maximum.v4f32(<4 x float> %a, <4 x float> %b)
+  ret <4 x float> %m
+}
+
+define <2 x double> @v2f64_minimum(<2 x double> %a, <2 x double> %b) {
+; NOVSX-LABEL: v2f64_minimum:
+; NOVSX:       # %bb.0: # %entry
+; NOVSX-NEXT:    fcmpu 0, 1, 3
+; NOVSX-NEXT:    fmr 6, 1
+; NOVSX-NEXT:    stfd 4, -16(1)
+; NOVSX-NEXT:    stfd 2, -8(1)
+; NOVSX-NEXT:    stfd 3, -32(1)
+; NOVSX-NEXT:    stfd 1, -24(1)
+; NOVSX-NEXT:    bc 12, 0, .LBB6_2
+; NOVSX-NEXT:  # %bb.1: # %entry
+; NOVSX-NEXT:    fmr 6, 3
+; NOVSX-NEXT:  .LBB6_2: # %entry
+; NOVSX-NEXT:    addis 3, 2, .LCPI6_0@toc@ha
+; NOVSX-NEXT:    ld 4, -24(1)
+; NOVSX-NEXT:    lfs 0, .LCPI6_0@toc@l(3)
+; NOVSX-NEXT:    fmr 5, 0
+; NOVSX-NEXT:    bc 12, 3, .LBB6_4
+; NOVSX-NEXT:  # %bb.3: # %entry
+; NOVSX-NEXT:    fmr 5, 6
+; NOVSX-NEXT:  .LBB6_4: # %entry
+; NOVSX-NEXT:    li 3, 1
+; NOVSX-NEXT:    ld 5, -32(1)
+; NOVSX-NEXT:    rldic 3, 3, 63, 0
+; NOVSX-NEXT:    cmpd 4, 3
+; NOVSX-NEXT:    bc 12, 2, .LBB6_6
+; NOVSX-NEXT:  # %bb.5: # %entry
+; NOVSX-NEXT:    fmr 1, 5
+; NOVSX-NEXT:  .LBB6_6: # %entry
+; NOVSX-NEXT:    cmpd 5, 3
+; NOVSX-NEXT:    bc 12, 2, .LBB6_8
+; NOVSX-NEXT:  # %bb.7: # %entry
+; NOVSX-NEXT:    fmr 3, 1
+; NOVSX-NEXT:  .LBB6_8: # %entry
+; NOVSX-NEXT:    addis 4, 2, .LCPI6_1@toc@ha
+; NOVSX-NEXT:    lfs 1, .LCPI6_1@toc@l(4)
+; NOVSX-NEXT:    fcmpu 0, 5, 1
+; NOVSX-NEXT:    bc 12, 2, .LBB6_10
+; NOVSX-NEXT:  # %bb.9: # %entry
+; NOVSX-NEXT:    fmr 3, 5
+; NOVSX-NEXT:  .LBB6_10: # %entry
+; NOVSX-NEXT:    fcmpu 0, 2, 4
+; NOVSX-NEXT:    fmr 5, 2
+; NOVSX-NEXT:    bc 12, 0, .LBB6_12
+; NOVSX-NEXT:  # %bb.11: # %entry
+; NOVSX-NEXT:    fmr 5, 4
+; NOVSX-NEXT:  .LBB6_12: # %entry
+; NOVSX-NEXT:    ld 5, -8(1)
+; NOVSX-NEXT:    bc 12, 3, .LBB6_14
+; NOVSX-NEXT:  # %bb.13: # %entry
+; NOVSX-NEXT:    fmr 0, 5
+; NOVSX-NEXT:  .LBB6_14: # %entry
+; NOVSX-NEXT:    cmpd 5, 3
+; NOVSX-NEXT:    ld 4, -16(1)
+; NOVSX-NEXT:    bc 4, 2, .LBB6_19
+; NOVSX-NEXT:  # %bb.15: # %entry
+; NOVSX-NEXT:    cmpd 4, 3
+; NOVSX-NEXT:    bc 4, 2, .LBB6_20
+; NOVSX-NEXT:  .LBB6_16: # %entry
+; NOVSX-NEXT:    fcmpu 0, 0, 1
+; NOVSX-NEXT:    bc 12, 2, .LBB6_18
+; NOVSX-NEXT:  .LBB6_17: # %entry
+; NOVSX-NEXT:    fmr 4, 0
+; NOVSX-NEXT:  .LBB6_18: # %entry
+; NOVSX-NEXT:    fmr 1, 3
+; NOVSX-NEXT:    fmr 2, 4
+; NOVSX-NEXT:    blr
+; NOVSX-NEXT:  .LBB6_19: # %entry
+; NOVSX-NEXT:    fmr 2, 0
+; NOVSX-NEXT:    cmpd 4, 3
+; NOVSX-NEXT:    bc 12, 2, .LBB6_16
+; NOVSX-NEXT:  .LBB6_20: # %entry
+; NOVSX-NEXT:    fmr 4, 2
+; NOVSX-NEXT:    fcmpu 0, 0, 1
+; NOVSX-NEXT:    bc 4, 2, .LBB6_17
+; NOVSX-NEXT:    b .LBB6_18
+;
+; VSX-LABEL: v2f64_minimum:
+; VSX:       # %bb.0: # %entry
+; VSX-NEXT:    addis 3, 2, .LCPI6_0@toc@ha
+; VSX-NEXT:    xvcmpeqdp 36, 35, 35
+; VSX-NEXT:    xvcmpeqdp 37, 34, 34
+; VSX-NEXT:    addi 3, 3, .LCPI6_0@toc@l
+; VSX-NEXT:    xxlnor 36, 36, 36
+; VSX-NEXT:    xxlnor 37, 37, 37
+; VSX-NEXT:    xvmindp 0, 34, 35
+; VSX-NEXT:    lxvd2x 2, 0, 3
+; VSX-NEXT:    addis 3, 2, .LCPI6_1@toc@ha
+; VSX-NEXT:    xxlor 1, 37, 36
+; VSX-NEXT:    addi 3, 3, .LCPI6_1@toc@l
+; VSX-NEXT:    lxvd2x 36, 0, 3
+; VSX-NEXT:    vcmpequd 5, 2, 4
+; VSX-NEXT:    xxsel 0, 0, 2, 1
+; VSX-NEXT:    xxlxor 2, 2, 2
+; VSX-NEXT:    xxsel 1, 0, 34, 37
+; VSX-NEXT:    vcmpequd 2, 3, 4
+; VSX-NEXT:    xxsel 1, 1, 35, 34
+; VSX-NEXT:    xvcmpeqdp 34, 0, 2
+; VSX-NEXT:    xxsel 34, 0, 1, 34
+; VSX-NEXT:    blr
+;
+; AIX-LABEL: v2f64_minimum:
+; AIX:       # %bb.0: # %entry
+; AIX-NEXT:    ld 3, L..C6(2) # %const.0
+; AIX-NEXT:    xvcmpeqdp 36, 35, 35
+; AIX-NEXT:    xvcmpeqdp 37, 34, 34
+; AIX-NEXT:    lxvd2x 2, 0, 3
+; AIX-NEXT:    ld 3, L..C7(2) # %const.1
+; AIX-NEXT:    xxlnor 36, 36, 36
+; AIX-NEXT:    xxlnor 37, 37, 37
+; AIX-NEXT:    xvmindp 0, 34, 35
+; AIX-NEXT:    xxlor 1, 37, 36
+; AIX-NEXT:    lxvd2x 36, 0, 3
+; AIX-NEXT:    vcmpequd 5, 2, 4
+; AIX-NEXT:    xxsel 0, 0, 2, 1
+; AIX-NEXT:    xxlxor 2, 2, 2
+; AIX-NEXT:    xxsel 1, 0, 34, 37
+; AIX-NEXT:    vcmpequd 2, 3, 4
+; AIX-NEXT:    xxsel 1, 1, 35, 34
+; AIX-NEXT:    xvcmpeqdp 34, 0, 2
+; AIX-NEXT:    xxsel 34, 0, 1, 34
+; AIX-NEXT:    blr
+entry:
+  %m = call <2 x double> @llvm.minimum.v2f64(<2 x double> %a, <2 x double> %b)
+  ret <2 x double> %m
+}
+
+define <2 x double> @v2f64_maximum(<2 x double> %a, <2 x double> %b) {
+; NOVSX-LABEL: v2f64_maximum:
+; NOVSX:       # %bb.0: # %entry
+; NOVSX-NEXT:    fcmpu 0, 1, 3
+; NOVSX-NEXT:    fmr 6, 1
+; NOVSX-NEXT:    stfd 4, -16(1)
+; NOVSX-NEXT:    stfd 2, -8(1)
+; NOVSX-NEXT:    stfd 3, -32(1)
+; NOVSX-NEXT:    stfd 1, -24(1)
+; NOVSX-NEXT:    bc 12, 1, .LBB7_2
+; NOVSX-NEXT:  # %bb.1: # %entry
+; NOVSX-NEXT:    fmr 6, 3
+; NOVSX-NEXT:  .LBB7_2: # %entry
+; NOVSX-NEXT:    addis 4, 2, .LCPI7_0@toc@ha
+; NOVSX-NEXT:    ld 3, -24(1)
+; NOVSX-NEXT:    lfs 0, .LCPI7_0@toc@l(4)
+; NOVSX-NEXT:    fmr 5, 0
+; NOVSX-NEXT:    bc 12, 3, .LBB7_4
+; NOVSX-NEXT:  # %bb.3: # %entry
+; NOVSX-NEXT:    fmr 5, 6
+; NOVSX-NEXT:  .LBB7_4: # %entry
+; NOVSX-NEXT:    cmpdi 3, 0
+; NOVSX-NEXT:    ld 4, -32(1)
+; NOVSX-NEXT:    bc 12, 2, .LBB7_6
+; NOVSX-NEXT:  # %bb.5: # %entry
+; NOVSX-NEXT:    fmr 1, 5
+; NOVSX-NEXT:  .LBB7_6: # %entry
+; NOVSX-NEXT:    cmpdi 4, 0
+; NOVSX-NEXT:    bc 12, 2, .LBB7_8
+; NOVSX-NEXT:  # %bb.7: # %entry
+; NOVSX-NEXT:    fmr 3, 1
+; NOVSX-NEXT:  .LBB7_8: # %entry
+; NOVSX-NEXT:    addis 3, 2, .LCPI7_1@toc@ha
+; NOVSX-NEXT:    lfs 1, .LCPI7_1@toc@l(3)
+; NOVSX-NEXT:    fcmpu 0, 5, 1
+; NOVSX-NEXT:    bc 12, 2, .LBB7_10
+; NOVSX-NEXT:  # %bb.9: # %entry
+; NOVSX-NEXT:    fmr 3, 5
+; NOVSX-NEXT:  .LBB7_10: # %entry
+; NOVSX-NEXT:    fcmpu 0, 2, 4
+; NOVSX-NEXT:    fmr 5, 2
+; NOVSX-NEXT:    bc 12, 1, .LBB7_12
+; NOVSX-NEXT:  # %bb.11: # %entry
+; NOVSX-NEXT:    fmr 5, 4
+; NOVSX-NEXT:  .LBB7_12: # %entry
+; NOVSX-NEXT:    ld 4, -8(1)
+; NOVSX-NEXT:    bc 12, 3, .LBB7_14
+; NOVSX-NEXT:  # %bb.13: # %entry
+; NOVSX-NEXT:    fmr 0, 5
+; NOVSX-NEXT:  .LBB7_14: # %entry
+; NOVSX-NEXT:    cmpdi 4, 0
+; NOVSX-NEXT:    ld 3, -16(1)
+; NOVSX-NEXT:    bc 4, 2, .LBB7_19
+; NOVSX-NEXT:  # %bb.15: # %entry
+; NOVSX-NEXT:    cmpdi 3, 0
+; NOVSX-NEXT:    bc 4, 2, .LBB7_20
+; NOVSX-NEXT:  .LBB7_16: # %entry
+; NOVSX-NEXT:    fcmpu 0, 0, 1
+; NOVSX-NEXT:    bc 12, 2, .LBB7_18
+; NOVSX-NEXT:  .LBB7_17: # %entry
+; NOVSX-NEXT:    fmr 4, 0
+; NOVSX-NEXT:  .LBB7_18: # %entry
+; NOVSX-NEXT:    fmr 1, 3
+; NOVSX-NEXT:    fmr 2, 4
+; NOVSX-NEXT:    blr
+; NOVSX-NEXT:  .LBB7_19: # %entry
+; NOVSX-NEXT:    fmr 2, 0
+; NOVSX-NEXT:    cmpdi 3, 0
+; NOVSX-NEXT:    bc 12, 2, .LBB7_16
+; NOVSX-NEXT:  .LBB7_20: # %entry
+; NOVSX-NEXT:    fmr 4, 2
+; NOVSX-NEXT:    fcmpu 0, 0, 1
+; NOVSX-NEXT:    bc 4, 2, .LBB7_17
+; NOVSX-NEXT:    b .LBB7_18
+;
+; VSX-LABEL: v2f64_maximum:
+; VSX:       # %bb.0: # %entry
+; VSX-NEXT:    addis 3, 2, .LCPI7_0@toc@ha
+; VSX-NEXT:    xvcmpeqdp 36, 35, 35
+; VSX-NEXT:    xvcmpeqdp 37, 34, 34
+; VSX-NEXT:    addi 3, 3, .LCPI7_0@toc@l
+; VSX-NEXT:    xxlnor 36, 36, 36
+; VSX-NEXT:    xxlnor 37, 37, 37
+; VSX-NEXT:    xvmaxdp 0, 34, 35
+; VSX-NEXT:    lxvd2x 2, 0, 3
+; VSX-NEXT:    xxlor 1, 37, 36
+; VSX-NEXT:    xxlxor 36, 36, 36
+; VSX-NEXT:    vcmpequd 5, 2, 4
+; VSX-NEXT:    xxsel 0, 0, 2, 1
+; VSX-NEXT:    xxsel 1, 0, 34, 37
+; VSX-NEXT:    vcmpequd 2, 3, 4
+; VSX-NEXT:    xxsel 1, 1, 35, 34
+; VSX-NEXT:    xvcmpeqdp 34, 0, 36
+; VSX-NEXT:    xxsel 34, 0, 1, 34
+; VSX-NEXT:    blr
+;
+; AIX-LABEL: v2f64_maximum:
+; AIX:       # %bb.0: # %entry
+; AIX-NEXT:    ld 3, L..C8(2) # %const.0
+; AIX-NEXT:    xvcmpeqdp 36, 35, 35
+; AIX-NEXT:    xvcmpeqdp 37, 34, 34
+; AIX-NEXT:    lxvd2x 2, 0, 3
+; AIX-NEXT:    xxlnor 36, 36, 36
+; AIX-NEXT:    xxlnor 37, 37, 37
+; AIX-NEXT:    xvmaxdp 0, 34, 35
+; AIX-NEXT:    xxlor 1, 37, 36
+; AIX-NEXT:    xxlxor 36, 36, 36
+; AIX-NEXT:    vcmpequd 5, 2, 4
+; AIX-NEXT:    xxsel 0, 0, 2, 1
+; AIX-NEXT:    xxsel 1, 0, 34, 37
+; AIX-NEXT:    vcmpequd 2, 3, 4
+; AIX-NEXT:    xxsel 1, 1, 35, 34
+; AIX-NEXT:    xvcmpeqdp 34, 0, 36
+; AIX-NEXT:    xxsel 34, 0, 1, 34
+; AIX-NEXT:    blr
+entry:
+  %m = call <2 x double> @llvm.maximum.v2f64(<2 x double> %a, <2 x double> %b)
+  ret <2 x double> %m
+}
+
+declare float @llvm.maximum.f32(float, float)
+declare double @llvm.maximum.f64(double, double)
+declare <4 x float> @llvm.maximum.v4f32(<4 x float>, <4 x float>)
+declare <2 x double> @llvm.maximum.v2f64(<2 x double>, <2 x double>)
+
+declare float @llvm.minimum.f32(float, float)
+declare double @llvm.minimum.f64(double, double)
+declare <4 x float> @llvm.minimum.v4f32(<4 x float>, <4 x float>)
+declare <2 x double> @llvm.minimum.v2f64(<2 x double>, <2 x double>)
diff --git a/llvm/test/CodeGen/PowerPC/legalize-vaarg.ll b/llvm/test/CodeGen/PowerPC/legalize-vaarg.ll
index 8980049969da..b7f8b8af2472 100644
--- a/llvm/test/CodeGen/PowerPC/legalize-vaarg.ll
+++ b/llvm/test/CodeGen/PowerPC/legalize-vaarg.ll
@@ -1,7 +1,6 @@
 ; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
 ;RUN: llc < %s --mtriple=powerpc64-unknown-linux-gnu -mattr=+altivec | FileCheck %s -check-prefix=BE
 ;RUN: llc < %s --mtriple=powerpc64le-unknown-linux-gnu -mattr=+altivec | FileCheck %s -check-prefix=LE
-;RUN: llc < %s --mtriple=powerpc64-unknown-linux-gnu -mattr=+altivec -ppc-gather-alias-max-depth=0 | FileCheck %s -check-prefix=FORWARD
 
 define <8 x i32> @test_large_vec_vaarg(i32 %n, ...) {
 ; BE-LABEL: test_large_vec_vaarg:
@@ -36,22 +35,6 @@ define <8 x i32> @test_large_vec_vaarg(i32 %n, ...) {
 ; LE-NEXT:    lxvd2x 0, 0, 3
 ; LE-NEXT:    xxswapd 35, 0
 ; LE-NEXT:    blr
-;
-; FORWARD-LABEL: test_large_vec_vaarg:
-; FORWARD:       # %bb.0:
-; FORWARD-NEXT:    ld 3, -8(1)
-; FORWARD-NEXT:    addi 3, 3, 15
-; FORWARD-NEXT:    rldicr 3, 3, 0, 59
-; FORWARD-NEXT:    addi 4, 3, 16
-; FORWARD-NEXT:    std 4, -8(1)
-; FORWARD-NEXT:    ld 4, -8(1)
-; FORWARD-NEXT:    lvx 2, 0, 3
-; FORWARD-NEXT:    addi 4, 4, 15
-; FORWARD-NEXT:    rldicr 3, 4, 0, 59
-; FORWARD-NEXT:    addi 4, 3, 16
-; FORWARD-NEXT:    std 4, -8(1)
-; FORWARD-NEXT:    lvx 3, 0, 3
-; FORWARD-NEXT:    blr
   %args = alloca ptr, align 4
   %x = va_arg ptr %args, <8 x i32>
   ret <8 x i32> %x
diff --git a/llvm/test/CodeGen/PowerPC/peephole-counter-XToI.mir b/llvm/test/CodeGen/PowerPC/peephole-counter-XToI.mir
new file mode 100644
index 000000000000..d8f2b08adaf2
--- /dev/null
+++ b/llvm/test/CodeGen/PowerPC/peephole-counter-XToI.mir
@@ -0,0 +1,85 @@
+# NOTE: Assertions have been autogenerated by utils/update_mir_test_checks.py UTC_ARGS: --version 4
+# REQUIRES: asserts
+# RUN: llc -mtriple=powerpc64le-unknown-linux-gnu -verify-machineinstrs \
+# RUN:   -run-pass ppc-mi-peepholes %s -o - | FileCheck %s --check-prefix=ALL
+# RUN: llc -mtriple=powerpc64le-unknown-linux-gnu -verify-machineinstrs \
+# RUN:   -run-pass ppc-mi-peepholes %s -o - -debug-counter=ppc-xtoi-peephole-skip=0,ppc-xtoi-peephole-count=8 \
+# RUN:   | FileCheck %s --check-prefix=ALL
+# RUN: llc -mtriple=powerpc64le-unknown-linux-gnu -verify-machineinstrs \
+# RUN:   -run-pass ppc-mi-peepholes %s -o - -debug-counter=ppc-xtoi-peephole-skip=3,ppc-xtoi-peephole-count=2 \
+# RUN:   | FileCheck %s --check-prefix=ONE-FIRSTSTORE
+# RUN: llc -mtriple=powerpc64le-unknown-linux-gnu -verify-machineinstrs \
+# RUN:   -run-pass ppc-mi-peepholes %s -o - -debug-counter=ppc-xtoi-peephole-skip=5,ppc-xtoi-peephole-count=2 \
+# RUN:   | FileCheck %s --check-prefix=ONE-SECONDSTORE
+# RUN: llc -mtriple=powerpc64le-unknown-linux-gnu -verify-machineinstrs \
+# RUN:   -run-pass ppc-mi-peepholes %s -o - -debug-counter=ppc-xtoi-peephole-skip=3,ppc-xtoi-peephole-count=4 \
+# RUN:   | FileCheck %s --check-prefix=TWO
+
+---
+name: foldDForm
+tracksRegLiveness: true
+body:             |
+  bb.0.entry:
+    liveins: $x3
+
+    ; ALL-LABEL: name: foldDForm
+    ; ALL: liveins: $x3
+    ; ALL-NEXT: {{  $}}
+    ; ALL-NEXT: [[COPY:%[0-9]+]]:g8rc_and_g8rc_nox0 = COPY killed $x3
+    ; ALL-NEXT: dead [[ADDI8_:%[0-9]+]]:g8rc_and_g8rc_nox0 = ADDI8 [[COPY]], 144
+    ; ALL-NEXT: [[LI8_:%[0-9]+]]:g8rc = LI8 0
+    ; ALL-NEXT: STD [[LI8_]], 160, [[COPY]]
+    ; ALL-NEXT: dead [[ADDI8_1:%[0-9]+]]:g8rc_and_g8rc_nox0 = ADDI8 [[COPY]], 160
+    ; ALL-NEXT: STD [[LI8_]], 176, [[COPY]]
+    ; ALL-NEXT: dead [[ADDI8_2:%[0-9]+]]:g8rc_and_g8rc_nox0 = ADDI8 [[COPY]], 176
+    ; ALL-NEXT: STD killed [[LI8_]], 192, killed [[COPY]]
+    ; ALL-NEXT: BLR8 implicit $lr8, implicit $rm
+    ;
+    ; ONE-FIRSTSTORE-LABEL: name: foldDForm
+    ; ONE-FIRSTSTORE: liveins: $x3
+    ; ONE-FIRSTSTORE-NEXT: {{  $}}
+    ; ONE-FIRSTSTORE-NEXT: [[COPY:%[0-9]+]]:g8rc_and_g8rc_nox0 = COPY killed $x3
+    ; ONE-FIRSTSTORE-NEXT: dead [[ADDI8_:%[0-9]+]]:g8rc_and_g8rc_nox0 = ADDI8 [[COPY]], 144
+    ; ONE-FIRSTSTORE-NEXT: [[LI8_:%[0-9]+]]:g8rc = LI8 0
+    ; ONE-FIRSTSTORE-NEXT: STD [[LI8_]], 160, [[COPY]]
+    ; ONE-FIRSTSTORE-NEXT: [[ADDI8_1:%[0-9]+]]:g8rc_and_g8rc_nox0 = ADDI8 [[COPY]], 160
+    ; ONE-FIRSTSTORE-NEXT: STD [[LI8_]], 16, killed [[ADDI8_1]]
+    ; ONE-FIRSTSTORE-NEXT: [[ADDI8_2:%[0-9]+]]:g8rc_and_g8rc_nox0 = ADDI8 killed [[COPY]], 176
+    ; ONE-FIRSTSTORE-NEXT: STD killed [[LI8_]], 16, killed [[ADDI8_2]]
+    ; ONE-FIRSTSTORE-NEXT: BLR8 implicit $lr8, implicit $rm
+    ;
+    ; ONE-SECONDSTORE-LABEL: name: foldDForm
+    ; ONE-SECONDSTORE: liveins: $x3
+    ; ONE-SECONDSTORE-NEXT: {{  $}}
+    ; ONE-SECONDSTORE-NEXT: [[COPY:%[0-9]+]]:g8rc_and_g8rc_nox0 = COPY killed $x3
+    ; ONE-SECONDSTORE-NEXT: [[ADDI8_:%[0-9]+]]:g8rc_and_g8rc_nox0 = ADDI8 [[COPY]], 144
+    ; ONE-SECONDSTORE-NEXT: [[LI8_:%[0-9]+]]:g8rc = LI8 0
+    ; ONE-SECONDSTORE-NEXT: STD [[LI8_]], 16, killed [[ADDI8_]]
+    ; ONE-SECONDSTORE-NEXT: dead [[ADDI8_1:%[0-9]+]]:g8rc_and_g8rc_nox0 = ADDI8 [[COPY]], 160
+    ; ONE-SECONDSTORE-NEXT: STD [[LI8_]], 176, [[COPY]]
+    ; ONE-SECONDSTORE-NEXT: [[ADDI8_2:%[0-9]+]]:g8rc_and_g8rc_nox0 = ADDI8 killed [[COPY]], 176
+    ; ONE-SECONDSTORE-NEXT: STD killed [[LI8_]], 16, killed [[ADDI8_2]]
+    ; ONE-SECONDSTORE-NEXT: BLR8 implicit $lr8, implicit $rm
+    ;
+    ; TWO-LABEL: name: foldDForm
+    ; TWO: liveins: $x3
+    ; TWO-NEXT: {{  $}}
+    ; TWO-NEXT: [[COPY:%[0-9]+]]:g8rc_and_g8rc_nox0 = COPY killed $x3
+    ; TWO-NEXT: dead [[ADDI8_:%[0-9]+]]:g8rc_and_g8rc_nox0 = ADDI8 [[COPY]], 144
+    ; TWO-NEXT: [[LI8_:%[0-9]+]]:g8rc = LI8 0
+    ; TWO-NEXT: STD [[LI8_]], 160, [[COPY]]
+    ; TWO-NEXT: dead [[ADDI8_1:%[0-9]+]]:g8rc_and_g8rc_nox0 = ADDI8 [[COPY]], 160
+    ; TWO-NEXT: STD [[LI8_]], 176, [[COPY]]
+    ; TWO-NEXT: [[ADDI8_2:%[0-9]+]]:g8rc_and_g8rc_nox0 = ADDI8 killed [[COPY]], 176
+    ; TWO-NEXT: STD killed [[LI8_]], 16, killed [[ADDI8_2]]
+    ; TWO-NEXT: BLR8 implicit $lr8, implicit $rm
+    %0:g8rc_and_g8rc_nox0 = COPY $x3
+    %1:g8rc_and_g8rc_nox0 = ADDI8 %0:g8rc_and_g8rc_nox0, 144
+    %2:g8rc = LI8 0
+    STD %2:g8rc, 16, %1:g8rc_and_g8rc_nox0
+    %3:g8rc_and_g8rc_nox0 = ADDI8 %0:g8rc_and_g8rc_nox0, 160
+    STD %2:g8rc, 16, %3:g8rc_and_g8rc_nox0
+    %4:g8rc_and_g8rc_nox0 = ADDI8 %0:g8rc_and_g8rc_nox0, 176
+    STD killed %2:g8rc, 16, %4:g8rc_and_g8rc_nox0
+    BLR8 implicit $lr8, implicit $rm
+...
diff --git a/llvm/test/CodeGen/PowerPC/peephole-counter-perOp.mir b/llvm/test/CodeGen/PowerPC/peephole-counter-perOp.mir
new file mode 100644
index 000000000000..cf3ff291e26c
--- /dev/null
+++ b/llvm/test/CodeGen/PowerPC/peephole-counter-perOp.mir
@@ -0,0 +1,74 @@
+# NOTE: Assertions have been autogenerated by utils/update_mir_test_checks.py UTC_ARGS: --version 4
+# REQUIRES: asserts
+# RUN: llc -mtriple=powerpc64le-unknown-linux-gnu -verify-machineinstrs \
+# RUN:   -run-pass ppc-mi-peepholes %s -o - | FileCheck %s --check-prefix=ALL
+# RUN: llc -mtriple=powerpc64le-unknown-linux-gnu -verify-machineinstrs \
+# RUN:   -run-pass ppc-mi-peepholes %s -o - -debug-counter=ppc-per-op-peephole-skip=0,ppc-per-op-peephole-count=6 \
+# RUN:   | FileCheck %s --check-prefix=ALL
+# RUN: llc -mtriple=powerpc64le-unknown-linux-gnu -verify-machineinstrs \
+# RUN:   -run-pass ppc-mi-peepholes %s -o - -debug-counter=ppc-per-op-peephole-skip=3,ppc-per-op-peephole-count=1 \
+# RUN:   | FileCheck %s --check-prefix=ONE-FIRST-RLWINM
+# RUN: llc -mtriple=powerpc64le-unknown-linux-gnu -verify-machineinstrs \
+# RUN:   -run-pass ppc-mi-peepholes %s -o - -debug-counter=ppc-per-op-peephole-skip=4,ppc-per-op-peephole-count=1 \
+# RUN:   | FileCheck %s --check-prefix=ONE-SECOND-RLWINM
+# RUN: llc -mtriple=powerpc64le-unknown-linux-gnu -verify-machineinstrs \
+# RUN:   -run-pass ppc-mi-peepholes %s -o - -debug-counter=ppc-per-op-peephole-skip=3,ppc-per-op-peephole-count=2 \
+# RUN:   | FileCheck %s --check-prefix=TWO
+
+---
+name: testFoldRLWINM
+#CHECK: name: testFoldRLWINM
+tracksRegLiveness: true
+body: |
+  bb.0.entry:
+    liveins: $x3
+    ; ALL-LABEL: name: testFoldRLWINM
+    ; ALL: liveins: $x3
+    ; ALL-NEXT: {{  $}}
+    ; ALL-NEXT: [[COPY:%[0-9]+]]:g8rc = COPY killed $x3
+    ; ALL-NEXT: [[COPY1:%[0-9]+]]:gprc = COPY killed [[COPY]].sub_32
+    ; ALL-NEXT: dead [[RLWINM:%[0-9]+]]:gprc = RLWINM [[COPY1]], 14, 0, 12
+    ; ALL-NEXT: dead [[RLWINM1:%[0-9]+]]:gprc = RLWINM [[COPY1]], 14, 0, 11
+    ; ALL-NEXT: dead [[RLWINM2:%[0-9]+]]:gprc = RLWINM killed [[COPY1]], 14, 0, 10
+    ; ALL-NEXT: BLR8 implicit $lr8, implicit $rm
+    ;
+    ; ONE-FIRST-RLWINM-LABEL: name: testFoldRLWINM
+    ; ONE-FIRST-RLWINM: liveins: $x3
+    ; ONE-FIRST-RLWINM-NEXT: {{  $}}
+    ; ONE-FIRST-RLWINM-NEXT: [[COPY:%[0-9]+]]:g8rc = COPY killed $x3
+    ; ONE-FIRST-RLWINM-NEXT: [[COPY1:%[0-9]+]]:gprc = COPY killed [[COPY]].sub_32
+    ; ONE-FIRST-RLWINM-NEXT: [[RLWINM:%[0-9]+]]:gprc = RLWINM [[COPY1]], 27, 5, 31
+    ; ONE-FIRST-RLWINM-NEXT: dead [[RLWINM1:%[0-9]+]]:gprc = RLWINM killed [[COPY1]], 14, 0, 12
+    ; ONE-FIRST-RLWINM-NEXT: dead [[RLWINM2:%[0-9]+]]:gprc = RLWINM [[RLWINM]], 19, 0, 11
+    ; ONE-FIRST-RLWINM-NEXT: dead [[RLWINM3:%[0-9]+]]:gprc = RLWINM killed [[RLWINM]], 19, 0, 10
+    ; ONE-FIRST-RLWINM-NEXT: BLR8 implicit $lr8, implicit $rm
+    ;
+    ; ONE-SECOND-RLWINM-LABEL: name: testFoldRLWINM
+    ; ONE-SECOND-RLWINM: liveins: $x3
+    ; ONE-SECOND-RLWINM-NEXT: {{  $}}
+    ; ONE-SECOND-RLWINM-NEXT: [[COPY:%[0-9]+]]:g8rc = COPY killed $x3
+    ; ONE-SECOND-RLWINM-NEXT: [[COPY1:%[0-9]+]]:gprc = COPY killed [[COPY]].sub_32
+    ; ONE-SECOND-RLWINM-NEXT: [[RLWINM:%[0-9]+]]:gprc = RLWINM [[COPY1]], 27, 5, 31
+    ; ONE-SECOND-RLWINM-NEXT: dead [[RLWINM1:%[0-9]+]]:gprc = RLWINM [[RLWINM]], 19, 0, 12
+    ; ONE-SECOND-RLWINM-NEXT: dead [[RLWINM2:%[0-9]+]]:gprc = RLWINM killed [[COPY1]], 14, 0, 11
+    ; ONE-SECOND-RLWINM-NEXT: dead [[RLWINM3:%[0-9]+]]:gprc = RLWINM killed [[RLWINM]], 19, 0, 10
+    ; ONE-SECOND-RLWINM-NEXT: BLR8 implicit $lr8, implicit $rm
+    ;
+    ; TWO-LABEL: name: testFoldRLWINM
+    ; TWO: liveins: $x3
+    ; TWO-NEXT: {{  $}}
+    ; TWO-NEXT: [[COPY:%[0-9]+]]:g8rc = COPY killed $x3
+    ; TWO-NEXT: [[COPY1:%[0-9]+]]:gprc = COPY killed [[COPY]].sub_32
+    ; TWO-NEXT: [[RLWINM:%[0-9]+]]:gprc = RLWINM [[COPY1]], 27, 5, 31
+    ; TWO-NEXT: dead [[RLWINM1:%[0-9]+]]:gprc = RLWINM [[COPY1]], 14, 0, 12
+    ; TWO-NEXT: dead [[RLWINM2:%[0-9]+]]:gprc = RLWINM killed [[COPY1]], 14, 0, 11
+    ; TWO-NEXT: dead [[RLWINM3:%[0-9]+]]:gprc = RLWINM killed [[RLWINM]], 19, 0, 10
+    ; TWO-NEXT: BLR8 implicit $lr8, implicit $rm
+    %0:g8rc = COPY $x3
+    %1:gprc = COPY %0.sub_32:g8rc
+    %2:gprc = RLWINM %1:gprc, 27, 5, 31
+    %3:gprc = RLWINM %2:gprc, 19, 0, 12
+    %4:gprc = RLWINM %2:gprc, 19, 0, 11
+    %5:gprc = RLWINM %2:gprc, 19, 0, 10
+    BLR8 implicit $lr8, implicit $rm
+...
diff --git a/llvm/test/CodeGen/RISCV/O0-pipeline.ll b/llvm/test/CodeGen/RISCV/O0-pipeline.ll
index faf37545e1a1..56bd4bd0c08f 100644
--- a/llvm/test/CodeGen/RISCV/O0-pipeline.ll
+++ b/llvm/test/CodeGen/RISCV/O0-pipeline.ll
@@ -47,6 +47,10 @@
 ; CHECK-NEXT:       Eliminate PHI nodes for register allocation
 ; CHECK-NEXT:       Two-Address instruction pass
 ; CHECK-NEXT:       Fast Register Allocator
+; CHECK-NEXT:       MachineDominator Tree Construction
+; CHECK-NEXT:       Slot index numbering
+; CHECK-NEXT:       Live Interval Analysis
+; CHECK-NEXT:       RISC-V Coalesce VSETVLI pass
 ; CHECK-NEXT:       Fast Register Allocator
 ; CHECK-NEXT:       Remove Redundant DEBUG_VALUE analysis
 ; CHECK-NEXT:       Fixup Statepoint Caller Saved
diff --git a/llvm/test/CodeGen/RISCV/O3-pipeline.ll b/llvm/test/CodeGen/RISCV/O3-pipeline.ll
index 90472f246918..4121d1110911 100644
--- a/llvm/test/CodeGen/RISCV/O3-pipeline.ll
+++ b/llvm/test/CodeGen/RISCV/O3-pipeline.ll
@@ -143,6 +143,7 @@
 ; CHECK-NEXT:       Machine Optimization Remark Emitter
 ; CHECK-NEXT:       Greedy Register Allocator
 ; CHECK-NEXT:       Virtual Register Rewriter
+; CHECK-NEXT:       RISC-V Coalesce VSETVLI pass
 ; CHECK-NEXT:       Virtual Register Map
 ; CHECK-NEXT:       Live Register Matrix
 ; CHECK-NEXT:       Greedy Register Allocator
diff --git a/llvm/test/CodeGen/RISCV/addimm-mulimm.ll b/llvm/test/CodeGen/RISCV/addimm-mulimm.ll
index 48fa69e10456..8fb251a75bd1 100644
--- a/llvm/test/CodeGen/RISCV/addimm-mulimm.ll
+++ b/llvm/test/CodeGen/RISCV/addimm-mulimm.ll
@@ -10,15 +10,17 @@
 define i32 @add_mul_combine_accept_a1(i32 %x) {
 ; RV32IMB-LABEL: add_mul_combine_accept_a1:
 ; RV32IMB:       # %bb.0:
-; RV32IMB-NEXT:    li a1, 29
-; RV32IMB-NEXT:    mul a0, a0, a1
+; RV32IMB-NEXT:    sh1add a1, a0, a0
+; RV32IMB-NEXT:    slli a0, a0, 5
+; RV32IMB-NEXT:    sub a0, a0, a1
 ; RV32IMB-NEXT:    addi a0, a0, 1073
 ; RV32IMB-NEXT:    ret
 ;
 ; RV64IMB-LABEL: add_mul_combine_accept_a1:
 ; RV64IMB:       # %bb.0:
-; RV64IMB-NEXT:    li a1, 29
-; RV64IMB-NEXT:    mul a0, a0, a1
+; RV64IMB-NEXT:    sh1add a1, a0, a0
+; RV64IMB-NEXT:    slli a0, a0, 5
+; RV64IMB-NEXT:    subw a0, a0, a1
 ; RV64IMB-NEXT:    addiw a0, a0, 1073
 ; RV64IMB-NEXT:    ret
   %tmp0 = add i32 %x, 37
@@ -29,15 +31,17 @@ define i32 @add_mul_combine_accept_a1(i32 %x) {
 define signext i32 @add_mul_combine_accept_a2(i32 signext %x) {
 ; RV32IMB-LABEL: add_mul_combine_accept_a2:
 ; RV32IMB:       # %bb.0:
-; RV32IMB-NEXT:    li a1, 29
-; RV32IMB-NEXT:    mul a0, a0, a1
+; RV32IMB-NEXT:    sh1add a1, a0, a0
+; RV32IMB-NEXT:    slli a0, a0, 5
+; RV32IMB-NEXT:    sub a0, a0, a1
 ; RV32IMB-NEXT:    addi a0, a0, 1073
 ; RV32IMB-NEXT:    ret
 ;
 ; RV64IMB-LABEL: add_mul_combine_accept_a2:
 ; RV64IMB:       # %bb.0:
-; RV64IMB-NEXT:    li a1, 29
-; RV64IMB-NEXT:    mul a0, a0, a1
+; RV64IMB-NEXT:    sh1add a1, a0, a0
+; RV64IMB-NEXT:    slli a0, a0, 5
+; RV64IMB-NEXT:    subw a0, a0, a1
 ; RV64IMB-NEXT:    addiw a0, a0, 1073
 ; RV64IMB-NEXT:    ret
   %tmp0 = add i32 %x, 37
@@ -49,10 +53,14 @@ define i64 @add_mul_combine_accept_a3(i64 %x) {
 ; RV32IMB-LABEL: add_mul_combine_accept_a3:
 ; RV32IMB:       # %bb.0:
 ; RV32IMB-NEXT:    li a2, 29
-; RV32IMB-NEXT:    mul a1, a1, a2
-; RV32IMB-NEXT:    mulhu a3, a0, a2
-; RV32IMB-NEXT:    add a1, a3, a1
-; RV32IMB-NEXT:    mul a2, a0, a2
+; RV32IMB-NEXT:    mulhu a2, a0, a2
+; RV32IMB-NEXT:    sh1add a3, a1, a1
+; RV32IMB-NEXT:    slli a1, a1, 5
+; RV32IMB-NEXT:    sub a1, a1, a3
+; RV32IMB-NEXT:    add a1, a2, a1
+; RV32IMB-NEXT:    sh1add a2, a0, a0
+; RV32IMB-NEXT:    slli a0, a0, 5
+; RV32IMB-NEXT:    sub a2, a0, a2
 ; RV32IMB-NEXT:    addi a0, a2, 1073
 ; RV32IMB-NEXT:    sltu a2, a0, a2
 ; RV32IMB-NEXT:    add a1, a1, a2
@@ -60,8 +68,9 @@ define i64 @add_mul_combine_accept_a3(i64 %x) {
 ;
 ; RV64IMB-LABEL: add_mul_combine_accept_a3:
 ; RV64IMB:       # %bb.0:
-; RV64IMB-NEXT:    li a1, 29
-; RV64IMB-NEXT:    mul a0, a0, a1
+; RV64IMB-NEXT:    sh1add a1, a0, a0
+; RV64IMB-NEXT:    slli a0, a0, 5
+; RV64IMB-NEXT:    sub a0, a0, a1
 ; RV64IMB-NEXT:    addi a0, a0, 1073
 ; RV64IMB-NEXT:    ret
   %tmp0 = add i64 %x, 37
@@ -72,8 +81,9 @@ define i64 @add_mul_combine_accept_a3(i64 %x) {
 define i32 @add_mul_combine_accept_b1(i32 %x) {
 ; RV32IMB-LABEL: add_mul_combine_accept_b1:
 ; RV32IMB:       # %bb.0:
-; RV32IMB-NEXT:    li a1, 23
-; RV32IMB-NEXT:    mul a0, a0, a1
+; RV32IMB-NEXT:    sh3add a1, a0, a0
+; RV32IMB-NEXT:    slli a0, a0, 5
+; RV32IMB-NEXT:    sub a0, a0, a1
 ; RV32IMB-NEXT:    lui a1, 50
 ; RV32IMB-NEXT:    addi a1, a1, 1119
 ; RV32IMB-NEXT:    add a0, a0, a1
@@ -81,8 +91,9 @@ define i32 @add_mul_combine_accept_b1(i32 %x) {
 ;
 ; RV64IMB-LABEL: add_mul_combine_accept_b1:
 ; RV64IMB:       # %bb.0:
-; RV64IMB-NEXT:    li a1, 23
-; RV64IMB-NEXT:    mul a0, a0, a1
+; RV64IMB-NEXT:    sh3add a1, a0, a0
+; RV64IMB-NEXT:    slli a0, a0, 5
+; RV64IMB-NEXT:    subw a0, a0, a1
 ; RV64IMB-NEXT:    lui a1, 50
 ; RV64IMB-NEXT:    addi a1, a1, 1119
 ; RV64IMB-NEXT:    addw a0, a0, a1
@@ -95,8 +106,9 @@ define i32 @add_mul_combine_accept_b1(i32 %x) {
 define signext i32 @add_mul_combine_accept_b2(i32 signext %x) {
 ; RV32IMB-LABEL: add_mul_combine_accept_b2:
 ; RV32IMB:       # %bb.0:
-; RV32IMB-NEXT:    li a1, 23
-; RV32IMB-NEXT:    mul a0, a0, a1
+; RV32IMB-NEXT:    sh3add a1, a0, a0
+; RV32IMB-NEXT:    slli a0, a0, 5
+; RV32IMB-NEXT:    sub a0, a0, a1
 ; RV32IMB-NEXT:    lui a1, 50
 ; RV32IMB-NEXT:    addi a1, a1, 1119
 ; RV32IMB-NEXT:    add a0, a0, a1
@@ -104,8 +116,9 @@ define signext i32 @add_mul_combine_accept_b2(i32 signext %x) {
 ;
 ; RV64IMB-LABEL: add_mul_combine_accept_b2:
 ; RV64IMB:       # %bb.0:
-; RV64IMB-NEXT:    li a1, 23
-; RV64IMB-NEXT:    mul a0, a0, a1
+; RV64IMB-NEXT:    sh3add a1, a0, a0
+; RV64IMB-NEXT:    slli a0, a0, 5
+; RV64IMB-NEXT:    subw a0, a0, a1
 ; RV64IMB-NEXT:    lui a1, 50
 ; RV64IMB-NEXT:    addi a1, a1, 1119
 ; RV64IMB-NEXT:    addw a0, a0, a1
@@ -119,10 +132,14 @@ define i64 @add_mul_combine_accept_b3(i64 %x) {
 ; RV32IMB-LABEL: add_mul_combine_accept_b3:
 ; RV32IMB:       # %bb.0:
 ; RV32IMB-NEXT:    li a2, 23
-; RV32IMB-NEXT:    mul a1, a1, a2
-; RV32IMB-NEXT:    mulhu a3, a0, a2
-; RV32IMB-NEXT:    add a1, a3, a1
-; RV32IMB-NEXT:    mul a2, a0, a2
+; RV32IMB-NEXT:    mulhu a2, a0, a2
+; RV32IMB-NEXT:    sh3add a3, a1, a1
+; RV32IMB-NEXT:    slli a1, a1, 5
+; RV32IMB-NEXT:    sub a1, a1, a3
+; RV32IMB-NEXT:    add a1, a2, a1
+; RV32IMB-NEXT:    sh3add a2, a0, a0
+; RV32IMB-NEXT:    slli a0, a0, 5
+; RV32IMB-NEXT:    sub a2, a0, a2
 ; RV32IMB-NEXT:    lui a0, 50
 ; RV32IMB-NEXT:    addi a0, a0, 1119
 ; RV32IMB-NEXT:    add a0, a2, a0
@@ -132,8 +149,9 @@ define i64 @add_mul_combine_accept_b3(i64 %x) {
 ;
 ; RV64IMB-LABEL: add_mul_combine_accept_b3:
 ; RV64IMB:       # %bb.0:
-; RV64IMB-NEXT:    li a1, 23
-; RV64IMB-NEXT:    mul a0, a0, a1
+; RV64IMB-NEXT:    sh3add a1, a0, a0
+; RV64IMB-NEXT:    slli a0, a0, 5
+; RV64IMB-NEXT:    sub a0, a0, a1
 ; RV64IMB-NEXT:    lui a1, 50
 ; RV64IMB-NEXT:    addiw a1, a1, 1119
 ; RV64IMB-NEXT:    add a0, a0, a1
@@ -147,15 +165,17 @@ define i32 @add_mul_combine_reject_a1(i32 %x) {
 ; RV32IMB-LABEL: add_mul_combine_reject_a1:
 ; RV32IMB:       # %bb.0:
 ; RV32IMB-NEXT:    addi a0, a0, 1971
-; RV32IMB-NEXT:    li a1, 29
-; RV32IMB-NEXT:    mul a0, a0, a1
+; RV32IMB-NEXT:    sh1add a1, a0, a0
+; RV32IMB-NEXT:    slli a0, a0, 5
+; RV32IMB-NEXT:    sub a0, a0, a1
 ; RV32IMB-NEXT:    ret
 ;
 ; RV64IMB-LABEL: add_mul_combine_reject_a1:
 ; RV64IMB:       # %bb.0:
 ; RV64IMB-NEXT:    addi a0, a0, 1971
-; RV64IMB-NEXT:    li a1, 29
-; RV64IMB-NEXT:    mulw a0, a0, a1
+; RV64IMB-NEXT:    sh1add a1, a0, a0
+; RV64IMB-NEXT:    slli a0, a0, 5
+; RV64IMB-NEXT:    subw a0, a0, a1
 ; RV64IMB-NEXT:    ret
   %tmp0 = add i32 %x, 1971
   %tmp1 = mul i32 %tmp0, 29
@@ -166,15 +186,17 @@ define signext i32 @add_mul_combine_reject_a2(i32 signext %x) {
 ; RV32IMB-LABEL: add_mul_combine_reject_a2:
 ; RV32IMB:       # %bb.0:
 ; RV32IMB-NEXT:    addi a0, a0, 1971
-; RV32IMB-NEXT:    li a1, 29
-; RV32IMB-NEXT:    mul a0, a0, a1
+; RV32IMB-NEXT:    sh1add a1, a0, a0
+; RV32IMB-NEXT:    slli a0, a0, 5
+; RV32IMB-NEXT:    sub a0, a0, a1
 ; RV32IMB-NEXT:    ret
 ;
 ; RV64IMB-LABEL: add_mul_combine_reject_a2:
 ; RV64IMB:       # %bb.0:
 ; RV64IMB-NEXT:    addi a0, a0, 1971
-; RV64IMB-NEXT:    li a1, 29
-; RV64IMB-NEXT:    mulw a0, a0, a1
+; RV64IMB-NEXT:    sh1add a1, a0, a0
+; RV64IMB-NEXT:    slli a0, a0, 5
+; RV64IMB-NEXT:    subw a0, a0, a1
 ; RV64IMB-NEXT:    ret
   %tmp0 = add i32 %x, 1971
   %tmp1 = mul i32 %tmp0, 29
@@ -185,10 +207,14 @@ define i64 @add_mul_combine_reject_a3(i64 %x) {
 ; RV32IMB-LABEL: add_mul_combine_reject_a3:
 ; RV32IMB:       # %bb.0:
 ; RV32IMB-NEXT:    li a2, 29
-; RV32IMB-NEXT:    mul a1, a1, a2
-; RV32IMB-NEXT:    mulhu a3, a0, a2
-; RV32IMB-NEXT:    add a1, a3, a1
-; RV32IMB-NEXT:    mul a2, a0, a2
+; RV32IMB-NEXT:    mulhu a2, a0, a2
+; RV32IMB-NEXT:    sh1add a3, a1, a1
+; RV32IMB-NEXT:    slli a1, a1, 5
+; RV32IMB-NEXT:    sub a1, a1, a3
+; RV32IMB-NEXT:    add a1, a2, a1
+; RV32IMB-NEXT:    sh1add a2, a0, a0
+; RV32IMB-NEXT:    slli a0, a0, 5
+; RV32IMB-NEXT:    sub a2, a0, a2
 ; RV32IMB-NEXT:    lui a0, 14
 ; RV32IMB-NEXT:    addi a0, a0, -185
 ; RV32IMB-NEXT:    add a0, a2, a0
@@ -199,8 +225,9 @@ define i64 @add_mul_combine_reject_a3(i64 %x) {
 ; RV64IMB-LABEL: add_mul_combine_reject_a3:
 ; RV64IMB:       # %bb.0:
 ; RV64IMB-NEXT:    addi a0, a0, 1971
-; RV64IMB-NEXT:    li a1, 29
-; RV64IMB-NEXT:    mul a0, a0, a1
+; RV64IMB-NEXT:    sh1add a1, a0, a0
+; RV64IMB-NEXT:    slli a0, a0, 5
+; RV64IMB-NEXT:    sub a0, a0, a1
 ; RV64IMB-NEXT:    ret
   %tmp0 = add i64 %x, 1971
   %tmp1 = mul i64 %tmp0, 29
@@ -251,10 +278,12 @@ define i64 @add_mul_combine_reject_c3(i64 %x) {
 ; RV32IMB-LABEL: add_mul_combine_reject_c3:
 ; RV32IMB:       # %bb.0:
 ; RV32IMB-NEXT:    li a2, 73
-; RV32IMB-NEXT:    mul a1, a1, a2
-; RV32IMB-NEXT:    mulhu a3, a0, a2
-; RV32IMB-NEXT:    add a1, a3, a1
-; RV32IMB-NEXT:    mul a2, a0, a2
+; RV32IMB-NEXT:    mulhu a2, a0, a2
+; RV32IMB-NEXT:    sh3add a3, a1, a1
+; RV32IMB-NEXT:    sh3add a1, a3, a1
+; RV32IMB-NEXT:    add a1, a2, a1
+; RV32IMB-NEXT:    sh3add a2, a0, a0
+; RV32IMB-NEXT:    sh3add a2, a2, a0
 ; RV32IMB-NEXT:    lui a0, 18
 ; RV32IMB-NEXT:    addi a0, a0, -728
 ; RV32IMB-NEXT:    add a0, a2, a0
@@ -343,15 +372,17 @@ define i32 @add_mul_combine_reject_e1(i32 %x) {
 ; RV32IMB-LABEL: add_mul_combine_reject_e1:
 ; RV32IMB:       # %bb.0:
 ; RV32IMB-NEXT:    addi a0, a0, 1971
-; RV32IMB-NEXT:    li a1, 29
-; RV32IMB-NEXT:    mul a0, a0, a1
+; RV32IMB-NEXT:    sh1add a1, a0, a0
+; RV32IMB-NEXT:    slli a0, a0, 5
+; RV32IMB-NEXT:    sub a0, a0, a1
 ; RV32IMB-NEXT:    ret
 ;
 ; RV64IMB-LABEL: add_mul_combine_reject_e1:
 ; RV64IMB:       # %bb.0:
 ; RV64IMB-NEXT:    addi a0, a0, 1971
-; RV64IMB-NEXT:    li a1, 29
-; RV64IMB-NEXT:    mulw a0, a0, a1
+; RV64IMB-NEXT:    sh1add a1, a0, a0
+; RV64IMB-NEXT:    slli a0, a0, 5
+; RV64IMB-NEXT:    subw a0, a0, a1
 ; RV64IMB-NEXT:    ret
   %tmp0 = mul i32 %x, 29
   %tmp1 = add i32 %tmp0, 57159
@@ -362,15 +393,17 @@ define signext i32 @add_mul_combine_reject_e2(i32 signext %x) {
 ; RV32IMB-LABEL: add_mul_combine_reject_e2:
 ; RV32IMB:       # %bb.0:
 ; RV32IMB-NEXT:    addi a0, a0, 1971
-; RV32IMB-NEXT:    li a1, 29
-; RV32IMB-NEXT:    mul a0, a0, a1
+; RV32IMB-NEXT:    sh1add a1, a0, a0
+; RV32IMB-NEXT:    slli a0, a0, 5
+; RV32IMB-NEXT:    sub a0, a0, a1
 ; RV32IMB-NEXT:    ret
 ;
 ; RV64IMB-LABEL: add_mul_combine_reject_e2:
 ; RV64IMB:       # %bb.0:
 ; RV64IMB-NEXT:    addi a0, a0, 1971
-; RV64IMB-NEXT:    li a1, 29
-; RV64IMB-NEXT:    mulw a0, a0, a1
+; RV64IMB-NEXT:    sh1add a1, a0, a0
+; RV64IMB-NEXT:    slli a0, a0, 5
+; RV64IMB-NEXT:    subw a0, a0, a1
 ; RV64IMB-NEXT:    ret
   %tmp0 = mul i32 %x, 29
   %tmp1 = add i32 %tmp0, 57159
@@ -381,10 +414,14 @@ define i64 @add_mul_combine_reject_e3(i64 %x) {
 ; RV32IMB-LABEL: add_mul_combine_reject_e3:
 ; RV32IMB:       # %bb.0:
 ; RV32IMB-NEXT:    li a2, 29
-; RV32IMB-NEXT:    mul a1, a1, a2
-; RV32IMB-NEXT:    mulhu a3, a0, a2
-; RV32IMB-NEXT:    add a1, a3, a1
-; RV32IMB-NEXT:    mul a2, a0, a2
+; RV32IMB-NEXT:    mulhu a2, a0, a2
+; RV32IMB-NEXT:    sh1add a3, a1, a1
+; RV32IMB-NEXT:    slli a1, a1, 5
+; RV32IMB-NEXT:    sub a1, a1, a3
+; RV32IMB-NEXT:    add a1, a2, a1
+; RV32IMB-NEXT:    sh1add a2, a0, a0
+; RV32IMB-NEXT:    slli a0, a0, 5
+; RV32IMB-NEXT:    sub a2, a0, a2
 ; RV32IMB-NEXT:    lui a0, 14
 ; RV32IMB-NEXT:    addi a0, a0, -185
 ; RV32IMB-NEXT:    add a0, a2, a0
@@ -395,8 +432,9 @@ define i64 @add_mul_combine_reject_e3(i64 %x) {
 ; RV64IMB-LABEL: add_mul_combine_reject_e3:
 ; RV64IMB:       # %bb.0:
 ; RV64IMB-NEXT:    addi a0, a0, 1971
-; RV64IMB-NEXT:    li a1, 29
-; RV64IMB-NEXT:    mul a0, a0, a1
+; RV64IMB-NEXT:    sh1add a1, a0, a0
+; RV64IMB-NEXT:    slli a0, a0, 5
+; RV64IMB-NEXT:    sub a0, a0, a1
 ; RV64IMB-NEXT:    ret
   %tmp0 = mul i64 %x, 29
   %tmp1 = add i64 %tmp0, 57159
@@ -407,16 +445,18 @@ define i32 @add_mul_combine_reject_f1(i32 %x) {
 ; RV32IMB-LABEL: add_mul_combine_reject_f1:
 ; RV32IMB:       # %bb.0:
 ; RV32IMB-NEXT:    addi a0, a0, 1972
-; RV32IMB-NEXT:    li a1, 29
-; RV32IMB-NEXT:    mul a0, a0, a1
+; RV32IMB-NEXT:    sh1add a1, a0, a0
+; RV32IMB-NEXT:    slli a0, a0, 5
+; RV32IMB-NEXT:    sub a0, a0, a1
 ; RV32IMB-NEXT:    addi a0, a0, 11
 ; RV32IMB-NEXT:    ret
 ;
 ; RV64IMB-LABEL: add_mul_combine_reject_f1:
 ; RV64IMB:       # %bb.0:
 ; RV64IMB-NEXT:    addi a0, a0, 1972
-; RV64IMB-NEXT:    li a1, 29
-; RV64IMB-NEXT:    mul a0, a0, a1
+; RV64IMB-NEXT:    sh1add a1, a0, a0
+; RV64IMB-NEXT:    slli a0, a0, 5
+; RV64IMB-NEXT:    subw a0, a0, a1
 ; RV64IMB-NEXT:    addiw a0, a0, 11
 ; RV64IMB-NEXT:    ret
   %tmp0 = mul i32 %x, 29
@@ -428,16 +468,18 @@ define signext i32 @add_mul_combine_reject_f2(i32 signext %x) {
 ; RV32IMB-LABEL: add_mul_combine_reject_f2:
 ; RV32IMB:       # %bb.0:
 ; RV32IMB-NEXT:    addi a0, a0, 1972
-; RV32IMB-NEXT:    li a1, 29
-; RV32IMB-NEXT:    mul a0, a0, a1
+; RV32IMB-NEXT:    sh1add a1, a0, a0
+; RV32IMB-NEXT:    slli a0, a0, 5
+; RV32IMB-NEXT:    sub a0, a0, a1
 ; RV32IMB-NEXT:    addi a0, a0, 11
 ; RV32IMB-NEXT:    ret
 ;
 ; RV64IMB-LABEL: add_mul_combine_reject_f2:
 ; RV64IMB:       # %bb.0:
 ; RV64IMB-NEXT:    addi a0, a0, 1972
-; RV64IMB-NEXT:    li a1, 29
-; RV64IMB-NEXT:    mul a0, a0, a1
+; RV64IMB-NEXT:    sh1add a1, a0, a0
+; RV64IMB-NEXT:    slli a0, a0, 5
+; RV64IMB-NEXT:    subw a0, a0, a1
 ; RV64IMB-NEXT:    addiw a0, a0, 11
 ; RV64IMB-NEXT:    ret
   %tmp0 = mul i32 %x, 29
@@ -449,10 +491,14 @@ define i64 @add_mul_combine_reject_f3(i64 %x) {
 ; RV32IMB-LABEL: add_mul_combine_reject_f3:
 ; RV32IMB:       # %bb.0:
 ; RV32IMB-NEXT:    li a2, 29
-; RV32IMB-NEXT:    mul a1, a1, a2
-; RV32IMB-NEXT:    mulhu a3, a0, a2
-; RV32IMB-NEXT:    add a1, a3, a1
-; RV32IMB-NEXT:    mul a2, a0, a2
+; RV32IMB-NEXT:    mulhu a2, a0, a2
+; RV32IMB-NEXT:    sh1add a3, a1, a1
+; RV32IMB-NEXT:    slli a1, a1, 5
+; RV32IMB-NEXT:    sub a1, a1, a3
+; RV32IMB-NEXT:    add a1, a2, a1
+; RV32IMB-NEXT:    sh1add a2, a0, a0
+; RV32IMB-NEXT:    slli a0, a0, 5
+; RV32IMB-NEXT:    sub a2, a0, a2
 ; RV32IMB-NEXT:    lui a0, 14
 ; RV32IMB-NEXT:    addi a0, a0, -145
 ; RV32IMB-NEXT:    add a0, a2, a0
@@ -463,8 +509,9 @@ define i64 @add_mul_combine_reject_f3(i64 %x) {
 ; RV64IMB-LABEL: add_mul_combine_reject_f3:
 ; RV64IMB:       # %bb.0:
 ; RV64IMB-NEXT:    addi a0, a0, 1972
-; RV64IMB-NEXT:    li a1, 29
-; RV64IMB-NEXT:    mul a0, a0, a1
+; RV64IMB-NEXT:    sh1add a1, a0, a0
+; RV64IMB-NEXT:    slli a0, a0, 5
+; RV64IMB-NEXT:    sub a0, a0, a1
 ; RV64IMB-NEXT:    addi a0, a0, 11
 ; RV64IMB-NEXT:    ret
   %tmp0 = mul i64 %x, 29
@@ -518,10 +565,12 @@ define i64 @add_mul_combine_reject_g3(i64 %x) {
 ; RV32IMB-LABEL: add_mul_combine_reject_g3:
 ; RV32IMB:       # %bb.0:
 ; RV32IMB-NEXT:    li a2, 73
-; RV32IMB-NEXT:    mul a1, a1, a2
-; RV32IMB-NEXT:    mulhu a3, a0, a2
-; RV32IMB-NEXT:    add a1, a3, a1
-; RV32IMB-NEXT:    mul a2, a0, a2
+; RV32IMB-NEXT:    mulhu a2, a0, a2
+; RV32IMB-NEXT:    sh3add a3, a1, a1
+; RV32IMB-NEXT:    sh3add a1, a3, a1
+; RV32IMB-NEXT:    add a1, a2, a1
+; RV32IMB-NEXT:    sh3add a2, a0, a0
+; RV32IMB-NEXT:    sh3add a2, a2, a0
 ; RV32IMB-NEXT:    lui a0, 2
 ; RV32IMB-NEXT:    addi a0, a0, -882
 ; RV32IMB-NEXT:    add a0, a2, a0
diff --git a/llvm/test/CodeGen/RISCV/alu64.ll b/llvm/test/CodeGen/RISCV/alu64.ll
index d2ee80e6aa95..f032756e007b 100644
--- a/llvm/test/CodeGen/RISCV/alu64.ll
+++ b/llvm/test/CodeGen/RISCV/alu64.ll
@@ -57,8 +57,8 @@ define i64 @sltiu(i64 %a) nounwind {
 ;
 ; RV32I-LABEL: sltiu:
 ; RV32I:       # %bb.0:
-; RV32I-NEXT:    seqz a1, a1
 ; RV32I-NEXT:    sltiu a0, a0, 3
+; RV32I-NEXT:    seqz a1, a1
 ; RV32I-NEXT:    and a0, a1, a0
 ; RV32I-NEXT:    li a1, 0
 ; RV32I-NEXT:    ret
diff --git a/llvm/test/CodeGen/RISCV/atomic-rmw.ll b/llvm/test/CodeGen/RISCV/atomic-rmw.ll
index fe5300174061..4871622ec236 100644
--- a/llvm/test/CodeGen/RISCV/atomic-rmw.ll
+++ b/llvm/test/CodeGen/RISCV/atomic-rmw.ll
@@ -2950,7 +2950,7 @@ define i8 @atomicrmw_and_i8_monotonic(ptr %a, i8 %b) nounwind {
 ; RV32IA-NEXT:    not a3, a3
 ; RV32IA-NEXT:    andi a1, a1, 255
 ; RV32IA-NEXT:    sll a1, a1, a0
-; RV32IA-NEXT:    or a1, a3, a1
+; RV32IA-NEXT:    or a1, a1, a3
 ; RV32IA-NEXT:    amoand.w a1, a1, (a2)
 ; RV32IA-NEXT:    srl a0, a1, a0
 ; RV32IA-NEXT:    ret
@@ -2974,7 +2974,7 @@ define i8 @atomicrmw_and_i8_monotonic(ptr %a, i8 %b) nounwind {
 ; RV64IA-NOZACAS-NEXT:    not a3, a3
 ; RV64IA-NOZACAS-NEXT:    andi a1, a1, 255
 ; RV64IA-NOZACAS-NEXT:    sllw a1, a1, a0
-; RV64IA-NOZACAS-NEXT:    or a1, a3, a1
+; RV64IA-NOZACAS-NEXT:    or a1, a1, a3
 ; RV64IA-NOZACAS-NEXT:    amoand.w a1, a1, (a2)
 ; RV64IA-NOZACAS-NEXT:    srlw a0, a1, a0
 ; RV64IA-NOZACAS-NEXT:    ret
@@ -2988,7 +2988,7 @@ define i8 @atomicrmw_and_i8_monotonic(ptr %a, i8 %b) nounwind {
 ; RV64IA-ZACAS-NEXT:    not a3, a3
 ; RV64IA-ZACAS-NEXT:    andi a1, a1, 255
 ; RV64IA-ZACAS-NEXT:    sllw a1, a1, a0
-; RV64IA-ZACAS-NEXT:    or a1, a3, a1
+; RV64IA-ZACAS-NEXT:    or a1, a1, a3
 ; RV64IA-ZACAS-NEXT:    amoand.w a1, a1, (a2)
 ; RV64IA-ZACAS-NEXT:    srlw a0, a1, a0
 ; RV64IA-ZACAS-NEXT:    ret
@@ -3026,7 +3026,7 @@ define i8 @atomicrmw_and_i8_acquire(ptr %a, i8 %b) nounwind {
 ; RV32IA-WMO-NEXT:    not a3, a3
 ; RV32IA-WMO-NEXT:    andi a1, a1, 255
 ; RV32IA-WMO-NEXT:    sll a1, a1, a0
-; RV32IA-WMO-NEXT:    or a1, a3, a1
+; RV32IA-WMO-NEXT:    or a1, a1, a3
 ; RV32IA-WMO-NEXT:    amoand.w.aq a1, a1, (a2)
 ; RV32IA-WMO-NEXT:    srl a0, a1, a0
 ; RV32IA-WMO-NEXT:    ret
@@ -3040,7 +3040,7 @@ define i8 @atomicrmw_and_i8_acquire(ptr %a, i8 %b) nounwind {
 ; RV32IA-TSO-NEXT:    not a3, a3
 ; RV32IA-TSO-NEXT:    andi a1, a1, 255
 ; RV32IA-TSO-NEXT:    sll a1, a1, a0
-; RV32IA-TSO-NEXT:    or a1, a3, a1
+; RV32IA-TSO-NEXT:    or a1, a1, a3
 ; RV32IA-TSO-NEXT:    amoand.w a1, a1, (a2)
 ; RV32IA-TSO-NEXT:    srl a0, a1, a0
 ; RV32IA-TSO-NEXT:    ret
@@ -3064,7 +3064,7 @@ define i8 @atomicrmw_and_i8_acquire(ptr %a, i8 %b) nounwind {
 ; RV64IA-WMO-NOZACAS-NEXT:    not a3, a3
 ; RV64IA-WMO-NOZACAS-NEXT:    andi a1, a1, 255
 ; RV64IA-WMO-NOZACAS-NEXT:    sllw a1, a1, a0
-; RV64IA-WMO-NOZACAS-NEXT:    or a1, a3, a1
+; RV64IA-WMO-NOZACAS-NEXT:    or a1, a1, a3
 ; RV64IA-WMO-NOZACAS-NEXT:    amoand.w.aq a1, a1, (a2)
 ; RV64IA-WMO-NOZACAS-NEXT:    srlw a0, a1, a0
 ; RV64IA-WMO-NOZACAS-NEXT:    ret
@@ -3078,7 +3078,7 @@ define i8 @atomicrmw_and_i8_acquire(ptr %a, i8 %b) nounwind {
 ; RV64IA-TSO-NOZACAS-NEXT:    not a3, a3
 ; RV64IA-TSO-NOZACAS-NEXT:    andi a1, a1, 255
 ; RV64IA-TSO-NOZACAS-NEXT:    sllw a1, a1, a0
-; RV64IA-TSO-NOZACAS-NEXT:    or a1, a3, a1
+; RV64IA-TSO-NOZACAS-NEXT:    or a1, a1, a3
 ; RV64IA-TSO-NOZACAS-NEXT:    amoand.w a1, a1, (a2)
 ; RV64IA-TSO-NOZACAS-NEXT:    srlw a0, a1, a0
 ; RV64IA-TSO-NOZACAS-NEXT:    ret
@@ -3092,7 +3092,7 @@ define i8 @atomicrmw_and_i8_acquire(ptr %a, i8 %b) nounwind {
 ; RV64IA-WMO-ZACAS-NEXT:    not a3, a3
 ; RV64IA-WMO-ZACAS-NEXT:    andi a1, a1, 255
 ; RV64IA-WMO-ZACAS-NEXT:    sllw a1, a1, a0
-; RV64IA-WMO-ZACAS-NEXT:    or a1, a3, a1
+; RV64IA-WMO-ZACAS-NEXT:    or a1, a1, a3
 ; RV64IA-WMO-ZACAS-NEXT:    amoand.w.aq a1, a1, (a2)
 ; RV64IA-WMO-ZACAS-NEXT:    srlw a0, a1, a0
 ; RV64IA-WMO-ZACAS-NEXT:    ret
@@ -3106,7 +3106,7 @@ define i8 @atomicrmw_and_i8_acquire(ptr %a, i8 %b) nounwind {
 ; RV64IA-TSO-ZACAS-NEXT:    not a3, a3
 ; RV64IA-TSO-ZACAS-NEXT:    andi a1, a1, 255
 ; RV64IA-TSO-ZACAS-NEXT:    sllw a1, a1, a0
-; RV64IA-TSO-ZACAS-NEXT:    or a1, a3, a1
+; RV64IA-TSO-ZACAS-NEXT:    or a1, a1, a3
 ; RV64IA-TSO-ZACAS-NEXT:    amoand.w a1, a1, (a2)
 ; RV64IA-TSO-ZACAS-NEXT:    srlw a0, a1, a0
 ; RV64IA-TSO-ZACAS-NEXT:    ret
@@ -3144,7 +3144,7 @@ define i8 @atomicrmw_and_i8_release(ptr %a, i8 %b) nounwind {
 ; RV32IA-WMO-NEXT:    not a3, a3
 ; RV32IA-WMO-NEXT:    andi a1, a1, 255
 ; RV32IA-WMO-NEXT:    sll a1, a1, a0
-; RV32IA-WMO-NEXT:    or a1, a3, a1
+; RV32IA-WMO-NEXT:    or a1, a1, a3
 ; RV32IA-WMO-NEXT:    amoand.w.rl a1, a1, (a2)
 ; RV32IA-WMO-NEXT:    srl a0, a1, a0
 ; RV32IA-WMO-NEXT:    ret
@@ -3158,7 +3158,7 @@ define i8 @atomicrmw_and_i8_release(ptr %a, i8 %b) nounwind {
 ; RV32IA-TSO-NEXT:    not a3, a3
 ; RV32IA-TSO-NEXT:    andi a1, a1, 255
 ; RV32IA-TSO-NEXT:    sll a1, a1, a0
-; RV32IA-TSO-NEXT:    or a1, a3, a1
+; RV32IA-TSO-NEXT:    or a1, a1, a3
 ; RV32IA-TSO-NEXT:    amoand.w a1, a1, (a2)
 ; RV32IA-TSO-NEXT:    srl a0, a1, a0
 ; RV32IA-TSO-NEXT:    ret
@@ -3182,7 +3182,7 @@ define i8 @atomicrmw_and_i8_release(ptr %a, i8 %b) nounwind {
 ; RV64IA-WMO-NOZACAS-NEXT:    not a3, a3
 ; RV64IA-WMO-NOZACAS-NEXT:    andi a1, a1, 255
 ; RV64IA-WMO-NOZACAS-NEXT:    sllw a1, a1, a0
-; RV64IA-WMO-NOZACAS-NEXT:    or a1, a3, a1
+; RV64IA-WMO-NOZACAS-NEXT:    or a1, a1, a3
 ; RV64IA-WMO-NOZACAS-NEXT:    amoand.w.rl a1, a1, (a2)
 ; RV64IA-WMO-NOZACAS-NEXT:    srlw a0, a1, a0
 ; RV64IA-WMO-NOZACAS-NEXT:    ret
@@ -3196,7 +3196,7 @@ define i8 @atomicrmw_and_i8_release(ptr %a, i8 %b) nounwind {
 ; RV64IA-TSO-NOZACAS-NEXT:    not a3, a3
 ; RV64IA-TSO-NOZACAS-NEXT:    andi a1, a1, 255
 ; RV64IA-TSO-NOZACAS-NEXT:    sllw a1, a1, a0
-; RV64IA-TSO-NOZACAS-NEXT:    or a1, a3, a1
+; RV64IA-TSO-NOZACAS-NEXT:    or a1, a1, a3
 ; RV64IA-TSO-NOZACAS-NEXT:    amoand.w a1, a1, (a2)
 ; RV64IA-TSO-NOZACAS-NEXT:    srlw a0, a1, a0
 ; RV64IA-TSO-NOZACAS-NEXT:    ret
@@ -3210,7 +3210,7 @@ define i8 @atomicrmw_and_i8_release(ptr %a, i8 %b) nounwind {
 ; RV64IA-WMO-ZACAS-NEXT:    not a3, a3
 ; RV64IA-WMO-ZACAS-NEXT:    andi a1, a1, 255
 ; RV64IA-WMO-ZACAS-NEXT:    sllw a1, a1, a0
-; RV64IA-WMO-ZACAS-NEXT:    or a1, a3, a1
+; RV64IA-WMO-ZACAS-NEXT:    or a1, a1, a3
 ; RV64IA-WMO-ZACAS-NEXT:    amoand.w.rl a1, a1, (a2)
 ; RV64IA-WMO-ZACAS-NEXT:    srlw a0, a1, a0
 ; RV64IA-WMO-ZACAS-NEXT:    ret
@@ -3224,7 +3224,7 @@ define i8 @atomicrmw_and_i8_release(ptr %a, i8 %b) nounwind {
 ; RV64IA-TSO-ZACAS-NEXT:    not a3, a3
 ; RV64IA-TSO-ZACAS-NEXT:    andi a1, a1, 255
 ; RV64IA-TSO-ZACAS-NEXT:    sllw a1, a1, a0
-; RV64IA-TSO-ZACAS-NEXT:    or a1, a3, a1
+; RV64IA-TSO-ZACAS-NEXT:    or a1, a1, a3
 ; RV64IA-TSO-ZACAS-NEXT:    amoand.w a1, a1, (a2)
 ; RV64IA-TSO-ZACAS-NEXT:    srlw a0, a1, a0
 ; RV64IA-TSO-ZACAS-NEXT:    ret
@@ -3262,7 +3262,7 @@ define i8 @atomicrmw_and_i8_acq_rel(ptr %a, i8 %b) nounwind {
 ; RV32IA-WMO-NEXT:    not a3, a3
 ; RV32IA-WMO-NEXT:    andi a1, a1, 255
 ; RV32IA-WMO-NEXT:    sll a1, a1, a0
-; RV32IA-WMO-NEXT:    or a1, a3, a1
+; RV32IA-WMO-NEXT:    or a1, a1, a3
 ; RV32IA-WMO-NEXT:    amoand.w.aqrl a1, a1, (a2)
 ; RV32IA-WMO-NEXT:    srl a0, a1, a0
 ; RV32IA-WMO-NEXT:    ret
@@ -3276,7 +3276,7 @@ define i8 @atomicrmw_and_i8_acq_rel(ptr %a, i8 %b) nounwind {
 ; RV32IA-TSO-NEXT:    not a3, a3
 ; RV32IA-TSO-NEXT:    andi a1, a1, 255
 ; RV32IA-TSO-NEXT:    sll a1, a1, a0
-; RV32IA-TSO-NEXT:    or a1, a3, a1
+; RV32IA-TSO-NEXT:    or a1, a1, a3
 ; RV32IA-TSO-NEXT:    amoand.w a1, a1, (a2)
 ; RV32IA-TSO-NEXT:    srl a0, a1, a0
 ; RV32IA-TSO-NEXT:    ret
@@ -3300,7 +3300,7 @@ define i8 @atomicrmw_and_i8_acq_rel(ptr %a, i8 %b) nounwind {
 ; RV64IA-WMO-NOZACAS-NEXT:    not a3, a3
 ; RV64IA-WMO-NOZACAS-NEXT:    andi a1, a1, 255
 ; RV64IA-WMO-NOZACAS-NEXT:    sllw a1, a1, a0
-; RV64IA-WMO-NOZACAS-NEXT:    or a1, a3, a1
+; RV64IA-WMO-NOZACAS-NEXT:    or a1, a1, a3
 ; RV64IA-WMO-NOZACAS-NEXT:    amoand.w.aqrl a1, a1, (a2)
 ; RV64IA-WMO-NOZACAS-NEXT:    srlw a0, a1, a0
 ; RV64IA-WMO-NOZACAS-NEXT:    ret
@@ -3314,7 +3314,7 @@ define i8 @atomicrmw_and_i8_acq_rel(ptr %a, i8 %b) nounwind {
 ; RV64IA-TSO-NOZACAS-NEXT:    not a3, a3
 ; RV64IA-TSO-NOZACAS-NEXT:    andi a1, a1, 255
 ; RV64IA-TSO-NOZACAS-NEXT:    sllw a1, a1, a0
-; RV64IA-TSO-NOZACAS-NEXT:    or a1, a3, a1
+; RV64IA-TSO-NOZACAS-NEXT:    or a1, a1, a3
 ; RV64IA-TSO-NOZACAS-NEXT:    amoand.w a1, a1, (a2)
 ; RV64IA-TSO-NOZACAS-NEXT:    srlw a0, a1, a0
 ; RV64IA-TSO-NOZACAS-NEXT:    ret
@@ -3328,7 +3328,7 @@ define i8 @atomicrmw_and_i8_acq_rel(ptr %a, i8 %b) nounwind {
 ; RV64IA-WMO-ZACAS-NEXT:    not a3, a3
 ; RV64IA-WMO-ZACAS-NEXT:    andi a1, a1, 255
 ; RV64IA-WMO-ZACAS-NEXT:    sllw a1, a1, a0
-; RV64IA-WMO-ZACAS-NEXT:    or a1, a3, a1
+; RV64IA-WMO-ZACAS-NEXT:    or a1, a1, a3
 ; RV64IA-WMO-ZACAS-NEXT:    amoand.w.aqrl a1, a1, (a2)
 ; RV64IA-WMO-ZACAS-NEXT:    srlw a0, a1, a0
 ; RV64IA-WMO-ZACAS-NEXT:    ret
@@ -3342,7 +3342,7 @@ define i8 @atomicrmw_and_i8_acq_rel(ptr %a, i8 %b) nounwind {
 ; RV64IA-TSO-ZACAS-NEXT:    not a3, a3
 ; RV64IA-TSO-ZACAS-NEXT:    andi a1, a1, 255
 ; RV64IA-TSO-ZACAS-NEXT:    sllw a1, a1, a0
-; RV64IA-TSO-ZACAS-NEXT:    or a1, a3, a1
+; RV64IA-TSO-ZACAS-NEXT:    or a1, a1, a3
 ; RV64IA-TSO-ZACAS-NEXT:    amoand.w a1, a1, (a2)
 ; RV64IA-TSO-ZACAS-NEXT:    srlw a0, a1, a0
 ; RV64IA-TSO-ZACAS-NEXT:    ret
@@ -3380,7 +3380,7 @@ define i8 @atomicrmw_and_i8_seq_cst(ptr %a, i8 %b) nounwind {
 ; RV32IA-WMO-NEXT:    not a3, a3
 ; RV32IA-WMO-NEXT:    andi a1, a1, 255
 ; RV32IA-WMO-NEXT:    sll a1, a1, a0
-; RV32IA-WMO-NEXT:    or a1, a3, a1
+; RV32IA-WMO-NEXT:    or a1, a1, a3
 ; RV32IA-WMO-NEXT:    amoand.w.aqrl a1, a1, (a2)
 ; RV32IA-WMO-NEXT:    srl a0, a1, a0
 ; RV32IA-WMO-NEXT:    ret
@@ -3394,7 +3394,7 @@ define i8 @atomicrmw_and_i8_seq_cst(ptr %a, i8 %b) nounwind {
 ; RV32IA-TSO-NEXT:    not a3, a3
 ; RV32IA-TSO-NEXT:    andi a1, a1, 255
 ; RV32IA-TSO-NEXT:    sll a1, a1, a0
-; RV32IA-TSO-NEXT:    or a1, a3, a1
+; RV32IA-TSO-NEXT:    or a1, a1, a3
 ; RV32IA-TSO-NEXT:    amoand.w a1, a1, (a2)
 ; RV32IA-TSO-NEXT:    srl a0, a1, a0
 ; RV32IA-TSO-NEXT:    ret
@@ -3418,7 +3418,7 @@ define i8 @atomicrmw_and_i8_seq_cst(ptr %a, i8 %b) nounwind {
 ; RV64IA-WMO-NOZACAS-NEXT:    not a3, a3
 ; RV64IA-WMO-NOZACAS-NEXT:    andi a1, a1, 255
 ; RV64IA-WMO-NOZACAS-NEXT:    sllw a1, a1, a0
-; RV64IA-WMO-NOZACAS-NEXT:    or a1, a3, a1
+; RV64IA-WMO-NOZACAS-NEXT:    or a1, a1, a3
 ; RV64IA-WMO-NOZACAS-NEXT:    amoand.w.aqrl a1, a1, (a2)
 ; RV64IA-WMO-NOZACAS-NEXT:    srlw a0, a1, a0
 ; RV64IA-WMO-NOZACAS-NEXT:    ret
@@ -3432,7 +3432,7 @@ define i8 @atomicrmw_and_i8_seq_cst(ptr %a, i8 %b) nounwind {
 ; RV64IA-TSO-NOZACAS-NEXT:    not a3, a3
 ; RV64IA-TSO-NOZACAS-NEXT:    andi a1, a1, 255
 ; RV64IA-TSO-NOZACAS-NEXT:    sllw a1, a1, a0
-; RV64IA-TSO-NOZACAS-NEXT:    or a1, a3, a1
+; RV64IA-TSO-NOZACAS-NEXT:    or a1, a1, a3
 ; RV64IA-TSO-NOZACAS-NEXT:    amoand.w a1, a1, (a2)
 ; RV64IA-TSO-NOZACAS-NEXT:    srlw a0, a1, a0
 ; RV64IA-TSO-NOZACAS-NEXT:    ret
@@ -3446,7 +3446,7 @@ define i8 @atomicrmw_and_i8_seq_cst(ptr %a, i8 %b) nounwind {
 ; RV64IA-WMO-ZACAS-NEXT:    not a3, a3
 ; RV64IA-WMO-ZACAS-NEXT:    andi a1, a1, 255
 ; RV64IA-WMO-ZACAS-NEXT:    sllw a1, a1, a0
-; RV64IA-WMO-ZACAS-NEXT:    or a1, a3, a1
+; RV64IA-WMO-ZACAS-NEXT:    or a1, a1, a3
 ; RV64IA-WMO-ZACAS-NEXT:    amoand.w.aqrl a1, a1, (a2)
 ; RV64IA-WMO-ZACAS-NEXT:    srlw a0, a1, a0
 ; RV64IA-WMO-ZACAS-NEXT:    ret
@@ -3460,7 +3460,7 @@ define i8 @atomicrmw_and_i8_seq_cst(ptr %a, i8 %b) nounwind {
 ; RV64IA-TSO-ZACAS-NEXT:    not a3, a3
 ; RV64IA-TSO-ZACAS-NEXT:    andi a1, a1, 255
 ; RV64IA-TSO-ZACAS-NEXT:    sllw a1, a1, a0
-; RV64IA-TSO-ZACAS-NEXT:    or a1, a3, a1
+; RV64IA-TSO-ZACAS-NEXT:    or a1, a1, a3
 ; RV64IA-TSO-ZACAS-NEXT:    amoand.w a1, a1, (a2)
 ; RV64IA-TSO-ZACAS-NEXT:    srlw a0, a1, a0
 ; RV64IA-TSO-ZACAS-NEXT:    ret
@@ -12863,7 +12863,7 @@ define i16 @atomicrmw_and_i16_monotonic(ptr %a, i16 %b) nounwind {
 ; RV32IA-NEXT:    not a4, a4
 ; RV32IA-NEXT:    and a1, a1, a3
 ; RV32IA-NEXT:    sll a1, a1, a0
-; RV32IA-NEXT:    or a1, a4, a1
+; RV32IA-NEXT:    or a1, a1, a4
 ; RV32IA-NEXT:    amoand.w a1, a1, (a2)
 ; RV32IA-NEXT:    srl a0, a1, a0
 ; RV32IA-NEXT:    ret
@@ -12888,7 +12888,7 @@ define i16 @atomicrmw_and_i16_monotonic(ptr %a, i16 %b) nounwind {
 ; RV64IA-NOZACAS-NEXT:    not a4, a4
 ; RV64IA-NOZACAS-NEXT:    and a1, a1, a3
 ; RV64IA-NOZACAS-NEXT:    sllw a1, a1, a0
-; RV64IA-NOZACAS-NEXT:    or a1, a4, a1
+; RV64IA-NOZACAS-NEXT:    or a1, a1, a4
 ; RV64IA-NOZACAS-NEXT:    amoand.w a1, a1, (a2)
 ; RV64IA-NOZACAS-NEXT:    srlw a0, a1, a0
 ; RV64IA-NOZACAS-NEXT:    ret
@@ -12903,7 +12903,7 @@ define i16 @atomicrmw_and_i16_monotonic(ptr %a, i16 %b) nounwind {
 ; RV64IA-ZACAS-NEXT:    not a4, a4
 ; RV64IA-ZACAS-NEXT:    and a1, a1, a3
 ; RV64IA-ZACAS-NEXT:    sllw a1, a1, a0
-; RV64IA-ZACAS-NEXT:    or a1, a4, a1
+; RV64IA-ZACAS-NEXT:    or a1, a1, a4
 ; RV64IA-ZACAS-NEXT:    amoand.w a1, a1, (a2)
 ; RV64IA-ZACAS-NEXT:    srlw a0, a1, a0
 ; RV64IA-ZACAS-NEXT:    ret
@@ -12942,7 +12942,7 @@ define i16 @atomicrmw_and_i16_acquire(ptr %a, i16 %b) nounwind {
 ; RV32IA-WMO-NEXT:    not a4, a4
 ; RV32IA-WMO-NEXT:    and a1, a1, a3
 ; RV32IA-WMO-NEXT:    sll a1, a1, a0
-; RV32IA-WMO-NEXT:    or a1, a4, a1
+; RV32IA-WMO-NEXT:    or a1, a1, a4
 ; RV32IA-WMO-NEXT:    amoand.w.aq a1, a1, (a2)
 ; RV32IA-WMO-NEXT:    srl a0, a1, a0
 ; RV32IA-WMO-NEXT:    ret
@@ -12957,7 +12957,7 @@ define i16 @atomicrmw_and_i16_acquire(ptr %a, i16 %b) nounwind {
 ; RV32IA-TSO-NEXT:    not a4, a4
 ; RV32IA-TSO-NEXT:    and a1, a1, a3
 ; RV32IA-TSO-NEXT:    sll a1, a1, a0
-; RV32IA-TSO-NEXT:    or a1, a4, a1
+; RV32IA-TSO-NEXT:    or a1, a1, a4
 ; RV32IA-TSO-NEXT:    amoand.w a1, a1, (a2)
 ; RV32IA-TSO-NEXT:    srl a0, a1, a0
 ; RV32IA-TSO-NEXT:    ret
@@ -12982,7 +12982,7 @@ define i16 @atomicrmw_and_i16_acquire(ptr %a, i16 %b) nounwind {
 ; RV64IA-WMO-NOZACAS-NEXT:    not a4, a4
 ; RV64IA-WMO-NOZACAS-NEXT:    and a1, a1, a3
 ; RV64IA-WMO-NOZACAS-NEXT:    sllw a1, a1, a0
-; RV64IA-WMO-NOZACAS-NEXT:    or a1, a4, a1
+; RV64IA-WMO-NOZACAS-NEXT:    or a1, a1, a4
 ; RV64IA-WMO-NOZACAS-NEXT:    amoand.w.aq a1, a1, (a2)
 ; RV64IA-WMO-NOZACAS-NEXT:    srlw a0, a1, a0
 ; RV64IA-WMO-NOZACAS-NEXT:    ret
@@ -12997,7 +12997,7 @@ define i16 @atomicrmw_and_i16_acquire(ptr %a, i16 %b) nounwind {
 ; RV64IA-TSO-NOZACAS-NEXT:    not a4, a4
 ; RV64IA-TSO-NOZACAS-NEXT:    and a1, a1, a3
 ; RV64IA-TSO-NOZACAS-NEXT:    sllw a1, a1, a0
-; RV64IA-TSO-NOZACAS-NEXT:    or a1, a4, a1
+; RV64IA-TSO-NOZACAS-NEXT:    or a1, a1, a4
 ; RV64IA-TSO-NOZACAS-NEXT:    amoand.w a1, a1, (a2)
 ; RV64IA-TSO-NOZACAS-NEXT:    srlw a0, a1, a0
 ; RV64IA-TSO-NOZACAS-NEXT:    ret
@@ -13012,7 +13012,7 @@ define i16 @atomicrmw_and_i16_acquire(ptr %a, i16 %b) nounwind {
 ; RV64IA-WMO-ZACAS-NEXT:    not a4, a4
 ; RV64IA-WMO-ZACAS-NEXT:    and a1, a1, a3
 ; RV64IA-WMO-ZACAS-NEXT:    sllw a1, a1, a0
-; RV64IA-WMO-ZACAS-NEXT:    or a1, a4, a1
+; RV64IA-WMO-ZACAS-NEXT:    or a1, a1, a4
 ; RV64IA-WMO-ZACAS-NEXT:    amoand.w.aq a1, a1, (a2)
 ; RV64IA-WMO-ZACAS-NEXT:    srlw a0, a1, a0
 ; RV64IA-WMO-ZACAS-NEXT:    ret
@@ -13027,7 +13027,7 @@ define i16 @atomicrmw_and_i16_acquire(ptr %a, i16 %b) nounwind {
 ; RV64IA-TSO-ZACAS-NEXT:    not a4, a4
 ; RV64IA-TSO-ZACAS-NEXT:    and a1, a1, a3
 ; RV64IA-TSO-ZACAS-NEXT:    sllw a1, a1, a0
-; RV64IA-TSO-ZACAS-NEXT:    or a1, a4, a1
+; RV64IA-TSO-ZACAS-NEXT:    or a1, a1, a4
 ; RV64IA-TSO-ZACAS-NEXT:    amoand.w a1, a1, (a2)
 ; RV64IA-TSO-ZACAS-NEXT:    srlw a0, a1, a0
 ; RV64IA-TSO-ZACAS-NEXT:    ret
@@ -13066,7 +13066,7 @@ define i16 @atomicrmw_and_i16_release(ptr %a, i16 %b) nounwind {
 ; RV32IA-WMO-NEXT:    not a4, a4
 ; RV32IA-WMO-NEXT:    and a1, a1, a3
 ; RV32IA-WMO-NEXT:    sll a1, a1, a0
-; RV32IA-WMO-NEXT:    or a1, a4, a1
+; RV32IA-WMO-NEXT:    or a1, a1, a4
 ; RV32IA-WMO-NEXT:    amoand.w.rl a1, a1, (a2)
 ; RV32IA-WMO-NEXT:    srl a0, a1, a0
 ; RV32IA-WMO-NEXT:    ret
@@ -13081,7 +13081,7 @@ define i16 @atomicrmw_and_i16_release(ptr %a, i16 %b) nounwind {
 ; RV32IA-TSO-NEXT:    not a4, a4
 ; RV32IA-TSO-NEXT:    and a1, a1, a3
 ; RV32IA-TSO-NEXT:    sll a1, a1, a0
-; RV32IA-TSO-NEXT:    or a1, a4, a1
+; RV32IA-TSO-NEXT:    or a1, a1, a4
 ; RV32IA-TSO-NEXT:    amoand.w a1, a1, (a2)
 ; RV32IA-TSO-NEXT:    srl a0, a1, a0
 ; RV32IA-TSO-NEXT:    ret
@@ -13106,7 +13106,7 @@ define i16 @atomicrmw_and_i16_release(ptr %a, i16 %b) nounwind {
 ; RV64IA-WMO-NOZACAS-NEXT:    not a4, a4
 ; RV64IA-WMO-NOZACAS-NEXT:    and a1, a1, a3
 ; RV64IA-WMO-NOZACAS-NEXT:    sllw a1, a1, a0
-; RV64IA-WMO-NOZACAS-NEXT:    or a1, a4, a1
+; RV64IA-WMO-NOZACAS-NEXT:    or a1, a1, a4
 ; RV64IA-WMO-NOZACAS-NEXT:    amoand.w.rl a1, a1, (a2)
 ; RV64IA-WMO-NOZACAS-NEXT:    srlw a0, a1, a0
 ; RV64IA-WMO-NOZACAS-NEXT:    ret
@@ -13121,7 +13121,7 @@ define i16 @atomicrmw_and_i16_release(ptr %a, i16 %b) nounwind {
 ; RV64IA-TSO-NOZACAS-NEXT:    not a4, a4
 ; RV64IA-TSO-NOZACAS-NEXT:    and a1, a1, a3
 ; RV64IA-TSO-NOZACAS-NEXT:    sllw a1, a1, a0
-; RV64IA-TSO-NOZACAS-NEXT:    or a1, a4, a1
+; RV64IA-TSO-NOZACAS-NEXT:    or a1, a1, a4
 ; RV64IA-TSO-NOZACAS-NEXT:    amoand.w a1, a1, (a2)
 ; RV64IA-TSO-NOZACAS-NEXT:    srlw a0, a1, a0
 ; RV64IA-TSO-NOZACAS-NEXT:    ret
@@ -13136,7 +13136,7 @@ define i16 @atomicrmw_and_i16_release(ptr %a, i16 %b) nounwind {
 ; RV64IA-WMO-ZACAS-NEXT:    not a4, a4
 ; RV64IA-WMO-ZACAS-NEXT:    and a1, a1, a3
 ; RV64IA-WMO-ZACAS-NEXT:    sllw a1, a1, a0
-; RV64IA-WMO-ZACAS-NEXT:    or a1, a4, a1
+; RV64IA-WMO-ZACAS-NEXT:    or a1, a1, a4
 ; RV64IA-WMO-ZACAS-NEXT:    amoand.w.rl a1, a1, (a2)
 ; RV64IA-WMO-ZACAS-NEXT:    srlw a0, a1, a0
 ; RV64IA-WMO-ZACAS-NEXT:    ret
@@ -13151,7 +13151,7 @@ define i16 @atomicrmw_and_i16_release(ptr %a, i16 %b) nounwind {
 ; RV64IA-TSO-ZACAS-NEXT:    not a4, a4
 ; RV64IA-TSO-ZACAS-NEXT:    and a1, a1, a3
 ; RV64IA-TSO-ZACAS-NEXT:    sllw a1, a1, a0
-; RV64IA-TSO-ZACAS-NEXT:    or a1, a4, a1
+; RV64IA-TSO-ZACAS-NEXT:    or a1, a1, a4
 ; RV64IA-TSO-ZACAS-NEXT:    amoand.w a1, a1, (a2)
 ; RV64IA-TSO-ZACAS-NEXT:    srlw a0, a1, a0
 ; RV64IA-TSO-ZACAS-NEXT:    ret
@@ -13190,7 +13190,7 @@ define i16 @atomicrmw_and_i16_acq_rel(ptr %a, i16 %b) nounwind {
 ; RV32IA-WMO-NEXT:    not a4, a4
 ; RV32IA-WMO-NEXT:    and a1, a1, a3
 ; RV32IA-WMO-NEXT:    sll a1, a1, a0
-; RV32IA-WMO-NEXT:    or a1, a4, a1
+; RV32IA-WMO-NEXT:    or a1, a1, a4
 ; RV32IA-WMO-NEXT:    amoand.w.aqrl a1, a1, (a2)
 ; RV32IA-WMO-NEXT:    srl a0, a1, a0
 ; RV32IA-WMO-NEXT:    ret
@@ -13205,7 +13205,7 @@ define i16 @atomicrmw_and_i16_acq_rel(ptr %a, i16 %b) nounwind {
 ; RV32IA-TSO-NEXT:    not a4, a4
 ; RV32IA-TSO-NEXT:    and a1, a1, a3
 ; RV32IA-TSO-NEXT:    sll a1, a1, a0
-; RV32IA-TSO-NEXT:    or a1, a4, a1
+; RV32IA-TSO-NEXT:    or a1, a1, a4
 ; RV32IA-TSO-NEXT:    amoand.w a1, a1, (a2)
 ; RV32IA-TSO-NEXT:    srl a0, a1, a0
 ; RV32IA-TSO-NEXT:    ret
@@ -13230,7 +13230,7 @@ define i16 @atomicrmw_and_i16_acq_rel(ptr %a, i16 %b) nounwind {
 ; RV64IA-WMO-NOZACAS-NEXT:    not a4, a4
 ; RV64IA-WMO-NOZACAS-NEXT:    and a1, a1, a3
 ; RV64IA-WMO-NOZACAS-NEXT:    sllw a1, a1, a0
-; RV64IA-WMO-NOZACAS-NEXT:    or a1, a4, a1
+; RV64IA-WMO-NOZACAS-NEXT:    or a1, a1, a4
 ; RV64IA-WMO-NOZACAS-NEXT:    amoand.w.aqrl a1, a1, (a2)
 ; RV64IA-WMO-NOZACAS-NEXT:    srlw a0, a1, a0
 ; RV64IA-WMO-NOZACAS-NEXT:    ret
@@ -13245,7 +13245,7 @@ define i16 @atomicrmw_and_i16_acq_rel(ptr %a, i16 %b) nounwind {
 ; RV64IA-TSO-NOZACAS-NEXT:    not a4, a4
 ; RV64IA-TSO-NOZACAS-NEXT:    and a1, a1, a3
 ; RV64IA-TSO-NOZACAS-NEXT:    sllw a1, a1, a0
-; RV64IA-TSO-NOZACAS-NEXT:    or a1, a4, a1
+; RV64IA-TSO-NOZACAS-NEXT:    or a1, a1, a4
 ; RV64IA-TSO-NOZACAS-NEXT:    amoand.w a1, a1, (a2)
 ; RV64IA-TSO-NOZACAS-NEXT:    srlw a0, a1, a0
 ; RV64IA-TSO-NOZACAS-NEXT:    ret
@@ -13260,7 +13260,7 @@ define i16 @atomicrmw_and_i16_acq_rel(ptr %a, i16 %b) nounwind {
 ; RV64IA-WMO-ZACAS-NEXT:    not a4, a4
 ; RV64IA-WMO-ZACAS-NEXT:    and a1, a1, a3
 ; RV64IA-WMO-ZACAS-NEXT:    sllw a1, a1, a0
-; RV64IA-WMO-ZACAS-NEXT:    or a1, a4, a1
+; RV64IA-WMO-ZACAS-NEXT:    or a1, a1, a4
 ; RV64IA-WMO-ZACAS-NEXT:    amoand.w.aqrl a1, a1, (a2)
 ; RV64IA-WMO-ZACAS-NEXT:    srlw a0, a1, a0
 ; RV64IA-WMO-ZACAS-NEXT:    ret
@@ -13275,7 +13275,7 @@ define i16 @atomicrmw_and_i16_acq_rel(ptr %a, i16 %b) nounwind {
 ; RV64IA-TSO-ZACAS-NEXT:    not a4, a4
 ; RV64IA-TSO-ZACAS-NEXT:    and a1, a1, a3
 ; RV64IA-TSO-ZACAS-NEXT:    sllw a1, a1, a0
-; RV64IA-TSO-ZACAS-NEXT:    or a1, a4, a1
+; RV64IA-TSO-ZACAS-NEXT:    or a1, a1, a4
 ; RV64IA-TSO-ZACAS-NEXT:    amoand.w a1, a1, (a2)
 ; RV64IA-TSO-ZACAS-NEXT:    srlw a0, a1, a0
 ; RV64IA-TSO-ZACAS-NEXT:    ret
@@ -13314,7 +13314,7 @@ define i16 @atomicrmw_and_i16_seq_cst(ptr %a, i16 %b) nounwind {
 ; RV32IA-WMO-NEXT:    not a4, a4
 ; RV32IA-WMO-NEXT:    and a1, a1, a3
 ; RV32IA-WMO-NEXT:    sll a1, a1, a0
-; RV32IA-WMO-NEXT:    or a1, a4, a1
+; RV32IA-WMO-NEXT:    or a1, a1, a4
 ; RV32IA-WMO-NEXT:    amoand.w.aqrl a1, a1, (a2)
 ; RV32IA-WMO-NEXT:    srl a0, a1, a0
 ; RV32IA-WMO-NEXT:    ret
@@ -13329,7 +13329,7 @@ define i16 @atomicrmw_and_i16_seq_cst(ptr %a, i16 %b) nounwind {
 ; RV32IA-TSO-NEXT:    not a4, a4
 ; RV32IA-TSO-NEXT:    and a1, a1, a3
 ; RV32IA-TSO-NEXT:    sll a1, a1, a0
-; RV32IA-TSO-NEXT:    or a1, a4, a1
+; RV32IA-TSO-NEXT:    or a1, a1, a4
 ; RV32IA-TSO-NEXT:    amoand.w a1, a1, (a2)
 ; RV32IA-TSO-NEXT:    srl a0, a1, a0
 ; RV32IA-TSO-NEXT:    ret
@@ -13354,7 +13354,7 @@ define i16 @atomicrmw_and_i16_seq_cst(ptr %a, i16 %b) nounwind {
 ; RV64IA-WMO-NOZACAS-NEXT:    not a4, a4
 ; RV64IA-WMO-NOZACAS-NEXT:    and a1, a1, a3
 ; RV64IA-WMO-NOZACAS-NEXT:    sllw a1, a1, a0
-; RV64IA-WMO-NOZACAS-NEXT:    or a1, a4, a1
+; RV64IA-WMO-NOZACAS-NEXT:    or a1, a1, a4
 ; RV64IA-WMO-NOZACAS-NEXT:    amoand.w.aqrl a1, a1, (a2)
 ; RV64IA-WMO-NOZACAS-NEXT:    srlw a0, a1, a0
 ; RV64IA-WMO-NOZACAS-NEXT:    ret
@@ -13369,7 +13369,7 @@ define i16 @atomicrmw_and_i16_seq_cst(ptr %a, i16 %b) nounwind {
 ; RV64IA-TSO-NOZACAS-NEXT:    not a4, a4
 ; RV64IA-TSO-NOZACAS-NEXT:    and a1, a1, a3
 ; RV64IA-TSO-NOZACAS-NEXT:    sllw a1, a1, a0
-; RV64IA-TSO-NOZACAS-NEXT:    or a1, a4, a1
+; RV64IA-TSO-NOZACAS-NEXT:    or a1, a1, a4
 ; RV64IA-TSO-NOZACAS-NEXT:    amoand.w a1, a1, (a2)
 ; RV64IA-TSO-NOZACAS-NEXT:    srlw a0, a1, a0
 ; RV64IA-TSO-NOZACAS-NEXT:    ret
@@ -13384,7 +13384,7 @@ define i16 @atomicrmw_and_i16_seq_cst(ptr %a, i16 %b) nounwind {
 ; RV64IA-WMO-ZACAS-NEXT:    not a4, a4
 ; RV64IA-WMO-ZACAS-NEXT:    and a1, a1, a3
 ; RV64IA-WMO-ZACAS-NEXT:    sllw a1, a1, a0
-; RV64IA-WMO-ZACAS-NEXT:    or a1, a4, a1
+; RV64IA-WMO-ZACAS-NEXT:    or a1, a1, a4
 ; RV64IA-WMO-ZACAS-NEXT:    amoand.w.aqrl a1, a1, (a2)
 ; RV64IA-WMO-ZACAS-NEXT:    srlw a0, a1, a0
 ; RV64IA-WMO-ZACAS-NEXT:    ret
@@ -13399,7 +13399,7 @@ define i16 @atomicrmw_and_i16_seq_cst(ptr %a, i16 %b) nounwind {
 ; RV64IA-TSO-ZACAS-NEXT:    not a4, a4
 ; RV64IA-TSO-ZACAS-NEXT:    and a1, a1, a3
 ; RV64IA-TSO-ZACAS-NEXT:    sllw a1, a1, a0
-; RV64IA-TSO-ZACAS-NEXT:    or a1, a4, a1
+; RV64IA-TSO-ZACAS-NEXT:    or a1, a1, a4
 ; RV64IA-TSO-ZACAS-NEXT:    amoand.w a1, a1, (a2)
 ; RV64IA-TSO-ZACAS-NEXT:    srlw a0, a1, a0
 ; RV64IA-TSO-ZACAS-NEXT:    ret
diff --git a/llvm/test/CodeGen/RISCV/atomic-signext.ll b/llvm/test/CodeGen/RISCV/atomic-signext.ll
index bdf3b28d2d52..f7268f628812 100644
--- a/llvm/test/CodeGen/RISCV/atomic-signext.ll
+++ b/llvm/test/CodeGen/RISCV/atomic-signext.ll
@@ -362,7 +362,7 @@ define signext i8 @atomicrmw_and_i8_monotonic(ptr %a, i8 %b) nounwind {
 ; RV32IA-NEXT:    not a3, a3
 ; RV32IA-NEXT:    andi a1, a1, 255
 ; RV32IA-NEXT:    sll a1, a1, a0
-; RV32IA-NEXT:    or a1, a3, a1
+; RV32IA-NEXT:    or a1, a1, a3
 ; RV32IA-NEXT:    amoand.w a1, a1, (a2)
 ; RV32IA-NEXT:    srl a0, a1, a0
 ; RV32IA-NEXT:    slli a0, a0, 24
@@ -390,7 +390,7 @@ define signext i8 @atomicrmw_and_i8_monotonic(ptr %a, i8 %b) nounwind {
 ; RV64IA-NEXT:    not a3, a3
 ; RV64IA-NEXT:    andi a1, a1, 255
 ; RV64IA-NEXT:    sllw a1, a1, a0
-; RV64IA-NEXT:    or a1, a3, a1
+; RV64IA-NEXT:    or a1, a1, a3
 ; RV64IA-NEXT:    amoand.w a1, a1, (a2)
 ; RV64IA-NEXT:    srlw a0, a1, a0
 ; RV64IA-NEXT:    slli a0, a0, 56
@@ -1403,7 +1403,7 @@ define signext i16 @atomicrmw_and_i16_monotonic(ptr %a, i16 %b) nounwind {
 ; RV32IA-NEXT:    not a4, a4
 ; RV32IA-NEXT:    and a1, a1, a3
 ; RV32IA-NEXT:    sll a1, a1, a0
-; RV32IA-NEXT:    or a1, a4, a1
+; RV32IA-NEXT:    or a1, a1, a4
 ; RV32IA-NEXT:    amoand.w a1, a1, (a2)
 ; RV32IA-NEXT:    srl a0, a1, a0
 ; RV32IA-NEXT:    slli a0, a0, 16
@@ -1432,7 +1432,7 @@ define signext i16 @atomicrmw_and_i16_monotonic(ptr %a, i16 %b) nounwind {
 ; RV64IA-NEXT:    not a4, a4
 ; RV64IA-NEXT:    and a1, a1, a3
 ; RV64IA-NEXT:    sllw a1, a1, a0
-; RV64IA-NEXT:    or a1, a4, a1
+; RV64IA-NEXT:    or a1, a1, a4
 ; RV64IA-NEXT:    amoand.w a1, a1, (a2)
 ; RV64IA-NEXT:    srlw a0, a1, a0
 ; RV64IA-NEXT:    slli a0, a0, 48
diff --git a/llvm/test/CodeGen/RISCV/atomicrmw-uinc-udec-wrap.ll b/llvm/test/CodeGen/RISCV/atomicrmw-uinc-udec-wrap.ll
index f96e1bad2e38..a5a2ae79966c 100644
--- a/llvm/test/CodeGen/RISCV/atomicrmw-uinc-udec-wrap.ll
+++ b/llvm/test/CodeGen/RISCV/atomicrmw-uinc-udec-wrap.ll
@@ -372,10 +372,10 @@ define i32 @atomicrmw_uinc_wrap_i32(ptr %ptr, i32 %val) {
 ; RV32IA-NEXT:    # =>This Loop Header: Depth=1
 ; RV32IA-NEXT:    # Child Loop BB2_3 Depth 2
 ; RV32IA-NEXT:    mv a3, a2
-; RV32IA-NEXT:    addi a4, a2, 1
-; RV32IA-NEXT:    sltu a2, a2, a1
-; RV32IA-NEXT:    neg a2, a2
-; RV32IA-NEXT:    and a4, a2, a4
+; RV32IA-NEXT:    addi a2, a2, 1
+; RV32IA-NEXT:    sltu a4, a3, a1
+; RV32IA-NEXT:    neg a4, a4
+; RV32IA-NEXT:    and a4, a4, a2
 ; RV32IA-NEXT:  .LBB2_3: # %atomicrmw.start
 ; RV32IA-NEXT:    # Parent Loop BB2_1 Depth=1
 ; RV32IA-NEXT:    # => This Inner Loop Header: Depth=2
@@ -607,10 +607,10 @@ define i64 @atomicrmw_uinc_wrap_i64(ptr %ptr, i64 %val) {
 ; RV64IA-NEXT:    # =>This Loop Header: Depth=1
 ; RV64IA-NEXT:    # Child Loop BB3_3 Depth 2
 ; RV64IA-NEXT:    mv a3, a2
-; RV64IA-NEXT:    addi a4, a2, 1
-; RV64IA-NEXT:    sltu a2, a2, a1
-; RV64IA-NEXT:    neg a2, a2
-; RV64IA-NEXT:    and a4, a2, a4
+; RV64IA-NEXT:    addi a2, a2, 1
+; RV64IA-NEXT:    sltu a4, a3, a1
+; RV64IA-NEXT:    neg a4, a4
+; RV64IA-NEXT:    and a4, a4, a2
 ; RV64IA-NEXT:  .LBB3_3: # %atomicrmw.start
 ; RV64IA-NEXT:    # Parent Loop BB3_1 Depth=1
 ; RV64IA-NEXT:    # => This Inner Loop Header: Depth=2
diff --git a/llvm/test/CodeGen/RISCV/attributes.ll b/llvm/test/CodeGen/RISCV/attributes.ll
index 080783fdeec0..7bd3440c9dc0 100644
--- a/llvm/test/CodeGen/RISCV/attributes.ll
+++ b/llvm/test/CodeGen/RISCV/attributes.ll
@@ -112,12 +112,12 @@
 ; RUN: llc -mtriple=riscv32 -mattr=+experimental-zvfbfmin %s -o - | FileCheck --check-prefixes=CHECK,RV32ZVFBFMIN %s
 ; RUN: llc -mtriple=riscv32 -mattr=+experimental-zvfbfwma %s -o - | FileCheck --check-prefixes=CHECK,RV32ZVFBFWMA %s
 ; RUN: llc -mtriple=riscv32 -mattr=+experimental-zaamo %s -o - | FileCheck --check-prefix=RV32ZAAMO %s
-; RUN: llc -mtriple=riscv32 -mattr=+zacas %s -o - | FileCheck --check-prefix=RV32ZACAS %s
+; RUN: llc -mtriple=riscv32 -mattr=+a,zacas %s -o - | FileCheck --check-prefix=RV32ZACAS %s
 ; RUN: llc -mtriple=riscv32 -mattr=+experimental-zalasr %s -o - | FileCheck --check-prefix=RV32ZALASR %s
 ; RUN: llc -mtriple=riscv32 -mattr=+experimental-zalrsc %s -o - | FileCheck --check-prefix=RV32ZALRSC %s
 ; RUN: llc -mtriple=riscv32 -mattr=+zama16b %s -o - | FileCheck --check-prefixes=CHECK,RV32ZAMA16B %s
 ; RUN: llc -mtriple=riscv32 -mattr=+experimental-zicfilp %s -o - | FileCheck --check-prefix=RV32ZICFILP %s
-; RUN: llc -mtriple=riscv32 -mattr=+experimental-zabha %s -o - | FileCheck --check-prefix=RV32ZABHA %s
+; RUN: llc -mtriple=riscv32 -mattr=+a,+experimental-zabha %s -o - | FileCheck --check-prefix=RV32ZABHA %s
 ; RUN: llc -mtriple=riscv32 -mattr=+experimental-ssnpm  %s -o - | FileCheck --check-prefix=RV32SSNPM %s
 ; RUN: llc -mtriple=riscv32 -mattr=+experimental-smnpm  %s -o - | FileCheck --check-prefix=RV32SMNPM %s
 ; RUN: llc -mtriple=riscv32 -mattr=+experimental-smmpm %s -o - | FileCheck --check-prefix=RV32SMMPM %s
@@ -244,11 +244,11 @@
 ; RUN: llc -mtriple=riscv64 -mattr=+experimental-zvfbfmin %s -o - | FileCheck --check-prefixes=CHECK,RV64ZVFBFMIN %s
 ; RUN: llc -mtriple=riscv64 -mattr=+experimental-zvfbfwma %s -o - | FileCheck --check-prefixes=CHECK,RV64ZVFBFWMA %s
 ; RUN: llc -mtriple=riscv64 -mattr=+experimental-zaamo %s -o - | FileCheck --check-prefix=RV64ZAAMO %s
-; RUN: llc -mtriple=riscv64 -mattr=+zacas %s -o - | FileCheck --check-prefix=RV64ZACAS %s
+; RUN: llc -mtriple=riscv64 -mattr=+a,zacas %s -o - | FileCheck --check-prefix=RV64ZACAS %s
 ; RUN: llc -mtriple=riscv64 -mattr=+experimental-zalasr %s -o - | FileCheck --check-prefix=RV64ZALASR %s
 ; RUN: llc -mtriple=riscv64 -mattr=+experimental-zalrsc %s -o - | FileCheck --check-prefix=RV64ZALRSC %s
 ; RUN: llc -mtriple=riscv64 -mattr=+experimental-zicfilp %s -o - | FileCheck --check-prefix=RV64ZICFILP %s
-; RUN: llc -mtriple=riscv64 -mattr=+experimental-zabha %s -o - | FileCheck --check-prefix=RV64ZABHA %s
+; RUN: llc -mtriple=riscv64 -mattr=+a,+experimental-zabha %s -o - | FileCheck --check-prefix=RV64ZABHA %s
 ; RUN: llc -mtriple=riscv64 -mattr=+experimental-ssnpm  %s -o - | FileCheck --check-prefix=RV64SSNPM %s
 ; RUN: llc -mtriple=riscv64 -mattr=+experimental-smnpm  %s -o - | FileCheck --check-prefix=RV64SMNPM %s
 ; RUN: llc -mtriple=riscv64 -mattr=+experimental-smmpm %s -o - | FileCheck --check-prefix=RV64SMMPM %s
@@ -256,6 +256,19 @@
 ; RUN: llc -mtriple=riscv64 -mattr=+experimental-supm %s -o - | FileCheck --check-prefix=RV64SUPM %s
 ; RUN: llc -mtriple=riscv64 -mattr=+experimental-ssqosid %s -o - | FileCheck --check-prefix=RV64SSQOSID %s
 
+; Tests for profile features.
+; RUN: llc -mtriple=riscv32 -mattr=+rvi20u32 %s -o - | FileCheck --check-prefix=RVI20U32 %s
+; RUN: llc -mtriple=riscv64 -mattr=+rvi20u64 %s -o - | FileCheck --check-prefix=RVI20U64 %s
+; RUN: llc -mtriple=riscv64 -mattr=+rva20u64 %s -o - | FileCheck --check-prefix=RVA20U64 %s
+; RUN: llc -mtriple=riscv64 -mattr=+rva20s64 %s -o - | FileCheck --check-prefix=RVA20S64 %s
+; RUN: llc -mtriple=riscv64 -mattr=+rva22u64 %s -o - | FileCheck --check-prefix=RVA22U64 %s
+; RUN: llc -mtriple=riscv64 -mattr=+rva22s64 %s -o - | FileCheck --check-prefix=RVA22S64 %s
+; RUN: llc -mtriple=riscv64 -mattr=+rva23u64 %s -o - | FileCheck --check-prefix=RVA23U64 %s
+; RUN: llc -mtriple=riscv64 -mattr=+rva23s64 %s -o - | FileCheck --check-prefix=RVA23S64 %s
+; RUN: llc -mtriple=riscv64 -mattr=+rvb23u64 %s -o - | FileCheck --check-prefix=RVB23U64 %s
+; RUN: llc -mtriple=riscv64 -mattr=+rvb23s64 %s -o - | FileCheck --check-prefix=RVB23S64 %s
+; RUN: llc -mtriple=riscv32 -mattr=+rvm23u32 %s -o - | FileCheck --check-prefix=RVM23U32 %s
+
 ; CHECK: .attribute 4, 16
 
 ; RV32M: .attribute 5, "rv32i2p1_m2p0"
@@ -326,7 +339,7 @@
 ; RV32XSFVFWMACCQQQ: .attribute 5, "rv32i2p1_f2p2_zicsr2p0_zve32f1p0_zve32x1p0_zvfbfmin1p0_zvl32b1p0_xsfvfwmaccqqq1p0"
 ; RV32XTHEADCMO: .attribute 5, "rv32i2p1_xtheadcmo1p0"
 ; RV32XTHEADCONDMOV: .attribute 5, "rv32i2p1_xtheadcondmov1p0"
-; RV32XTHEADFMEMIDX: .attribute 5, "rv32i2p1_f2p2_zicsr2p0_xtheadfmemidx1p0"
+; RV32XTHEADFMEMIDX: .attribute 5, "rv32i2p1_xtheadfmemidx1p0"
 ; RV32XTHEADMAC: .attribute 5, "rv32i2p1_xtheadmac1p0"
 ; RV32XTHEADMEMIDX: .attribute 5, "rv32i2p1_xtheadmemidx1p0"
 ; RV32XTHEADMEMPAIR: .attribute 5, "rv32i2p1_xtheadmempair1p0"
@@ -452,7 +465,7 @@
 ; RV64XTHEADBS: .attribute 5, "rv64i2p1_xtheadbs1p0"
 ; RV64XTHEADCMO: .attribute 5, "rv64i2p1_xtheadcmo1p0"
 ; RV64XTHEADCONDMOV: .attribute 5, "rv64i2p1_xtheadcondmov1p0"
-; RV64XTHEADFMEMIDX: .attribute 5, "rv64i2p1_f2p2_zicsr2p0_xtheadfmemidx1p0"
+; RV64XTHEADFMEMIDX: .attribute 5, "rv64i2p1_xtheadfmemidx1p0"
 ; RV64XTHEADMAC: .attribute 5, "rv64i2p1_xtheadmac1p0"
 ; RV64XTHEADMEMIDX: .attribute 5, "rv64i2p1_xtheadmemidx1p0"
 ; RV64XTHEADMEMPAIR: .attribute 5, "rv64i2p1_xtheadmempair1p0"
@@ -512,6 +525,18 @@
 ; RV64SUPM: .attribute 5, "rv64i2p1_supm0p8"
 ; RV64SSQOSID: .attribute 5, "rv64i2p1_ssqosid1p0"
 
+; RVI20U32: .attribute 5, "rv32i2p1"
+; RVI20U64: .attribute 5, "rv64i2p1"
+; RVA20U64: .attribute 5, "rv64i2p1_m2p0_a2p1_f2p2_d2p2_c2p0_ziccamoa1p0_ziccif1p0_zicclsm1p0_ziccrse1p0_zicntr2p0_zicsr2p0_za128rs1p0"
+; RVA20S64: .attribute 5, "rv64i2p1_m2p0_a2p1_f2p2_d2p2_c2p0_ziccamoa1p0_ziccif1p0_zicclsm1p0_ziccrse1p0_zicntr2p0_zicsr2p0_zifencei2p0_za128rs1p0_ssccptr1p0_sstvala1p0_sstvecd1p0_svade1p0_svbare1p0"
+; RVA22U64: .attribute 5, "rv64i2p1_m2p0_a2p1_f2p2_d2p2_c2p0_zic64b1p0_zicbom1p0_zicbop1p0_zicboz1p0_ziccamoa1p0_ziccif1p0_zicclsm1p0_ziccrse1p0_zicntr2p0_zicsr2p0_zihintpause2p0_zihpm2p0_za64rs1p0_zfhmin1p0_zba1p0_zbb1p0_zbs1p0_zkt1p0"
+; RVA22S64: .attribute 5, "rv64i2p1_m2p0_a2p1_f2p2_d2p2_c2p0_zic64b1p0_zicbom1p0_zicbop1p0_zicboz1p0_ziccamoa1p0_ziccif1p0_zicclsm1p0_ziccrse1p0_zicntr2p0_zicsr2p0_zifencei2p0_zihintpause2p0_zihpm2p0_za64rs1p0_zfhmin1p0_zba1p0_zbb1p0_zbs1p0_zkt1p0_ssccptr1p0_sscounterenw1p0_sstvala1p0_sstvecd1p0_svade1p0_svbare1p0_svinval1p0_svpbmt1p0"
+; RVA23U64: .attribute 5, "rv64i2p1_m2p0_a2p1_f2p2_d2p2_c2p0_v1p0_zic64b1p0_zicbom1p0_zicbop1p0_zicboz1p0_ziccamoa1p0_ziccif1p0_zicclsm1p0_ziccrse1p0_zicntr2p0_zicond1p0_zicsr2p0_zihintntl1p0_zihintpause2p0_zihpm2p0_zimop1p0_za64rs1p0_zawrs1p0_zfa1p0_zfhmin1p0_zca1p0_zcb1p0_zcmop1p0_zba1p0_zbb1p0_zbs1p0_zkt1p0_zvbb1p0_zve32f1p0_zve32x1p0_zve64d1p0_zve64f1p0_zve64x1p0_zvfhmin1p0_zvkb1p0_zvkt1p0_zvl128b1p0_zvl32b1p0_zvl64b1p0"
+; RVA23S64: .attribute 5, "rv64i2p1_m2p0_a2p1_f2p2_d2p2_c2p0_v1p0_h1p0_zic64b1p0_zicbom1p0_zicbop1p0_zicboz1p0_ziccamoa1p0_ziccif1p0_zicclsm1p0_ziccrse1p0_zicntr2p0_zicond1p0_zicsr2p0_zifencei2p0_zihintntl1p0_zihintpause2p0_zihpm2p0_zimop1p0_za64rs1p0_zawrs1p0_zfa1p0_zfhmin1p0_zca1p0_zcb1p0_zcmop1p0_zba1p0_zbb1p0_zbs1p0_zkt1p0_zvbb1p0_zve32f1p0_zve32x1p0_zve64d1p0_zve64f1p0_zve64x1p0_zvfhmin1p0_zvkb1p0_zvkt1p0_zvl128b1p0_zvl32b1p0_zvl64b1p0_shcounterenw1p0_shgatpa1p0_shtvala1p0_shvsatpa1p0_shvstvala1p0_shvstvecd1p0_ssccptr1p0_sscofpmf1p0_sscounterenw1p0_ssnpm0p8_ssstateen1p0_sstc1p0_sstvala1p0_sstvecd1p0_ssu64xl1p0_svade1p0_svbare1p0_svinval1p0_svnapot1p0_svpbmt1p0"
+; RVB23U64: .attribute 5, "rv64i2p1_m2p0_a2p1_f2p2_d2p2_c2p0_zic64b1p0_zicbom1p0_zicbop1p0_zicboz1p0_ziccamoa1p0_ziccif1p0_zicclsm1p0_ziccrse1p0_zicntr2p0_zicond1p0_zicsr2p0_zihintntl1p0_zihintpause2p0_zihpm2p0_zimop1p0_za64rs1p0_zawrs1p0_zfa1p0_zca1p0_zcb1p0_zcmop1p0_zba1p0_zbb1p0_zbs1p0_zkt1p0"
+; RVB23S64: .attribute 5, "rv64i2p1_m2p0_a2p1_f2p2_d2p2_c2p0_zic64b1p0_zicbom1p0_zicbop1p0_zicboz1p0_ziccamoa1p0_ziccif1p0_zicclsm1p0_ziccrse1p0_zicntr2p0_zicond1p0_zicsr2p0_zifencei2p0_zihintntl1p0_zihintpause2p0_zihpm2p0_zimop1p0_za64rs1p0_zawrs1p0_zfa1p0_zca1p0_zcb1p0_zcmop1p0_zba1p0_zbb1p0_zbs1p0_zkt1p0_ssccptr1p0_sscofpmf1p0_sscounterenw1p0_sstc1p0_sstvala1p0_sstvecd1p0_ssu64xl1p0_svade1p0_svbare1p0_svinval1p0_svnapot1p0_svpbmt1p0"
+; RVM23U32: .attribute 5, "rv32i2p1_m2p0_zicbop1p0_zicond1p0_zicsr2p0_zihintntl1p0_zihintpause2p0_zimop1p0_zca1p0_zcb1p0_zce1p0_zcmop1p0_zcmp1p0_zcmt1p0_zba1p0_zbb1p0_zbs1p0"
+
 define i32 @addi(i32 %a) {
   %1 = add i32 %a, 1
   ret i32 %1
diff --git a/llvm/test/CodeGen/RISCV/bfloat-convert.ll b/llvm/test/CodeGen/RISCV/bfloat-convert.ll
index 9e2b0b5c3cbb..770dcccee882 100644
--- a/llvm/test/CodeGen/RISCV/bfloat-convert.ll
+++ b/llvm/test/CodeGen/RISCV/bfloat-convert.ll
@@ -456,92 +456,80 @@ define i64 @fcvt_l_bf16(bfloat %a) nounwind {
 define i64 @fcvt_l_bf16_sat(bfloat %a) nounwind {
 ; RV32IZFBFMIN-LABEL: fcvt_l_bf16_sat:
 ; RV32IZFBFMIN:       # %bb.0: # %start
-; RV32IZFBFMIN-NEXT:    addi sp, sp, -32
-; RV32IZFBFMIN-NEXT:    sw ra, 28(sp) # 4-byte Folded Spill
-; RV32IZFBFMIN-NEXT:    sw s0, 24(sp) # 4-byte Folded Spill
-; RV32IZFBFMIN-NEXT:    sw s1, 20(sp) # 4-byte Folded Spill
-; RV32IZFBFMIN-NEXT:    sw s2, 16(sp) # 4-byte Folded Spill
-; RV32IZFBFMIN-NEXT:    sw s3, 12(sp) # 4-byte Folded Spill
-; RV32IZFBFMIN-NEXT:    fsw fs0, 8(sp) # 4-byte Folded Spill
-; RV32IZFBFMIN-NEXT:    lui a0, %hi(.LCPI10_0)
-; RV32IZFBFMIN-NEXT:    flw fa5, %lo(.LCPI10_0)(a0)
+; RV32IZFBFMIN-NEXT:    addi sp, sp, -16
+; RV32IZFBFMIN-NEXT:    sw ra, 12(sp) # 4-byte Folded Spill
+; RV32IZFBFMIN-NEXT:    sw s0, 8(sp) # 4-byte Folded Spill
+; RV32IZFBFMIN-NEXT:    fsw fs0, 4(sp) # 4-byte Folded Spill
 ; RV32IZFBFMIN-NEXT:    fcvt.s.bf16 fs0, fa0
-; RV32IZFBFMIN-NEXT:    flt.s s0, fa5, fs0
-; RV32IZFBFMIN-NEXT:    neg s1, s0
 ; RV32IZFBFMIN-NEXT:    lui a0, 913408
 ; RV32IZFBFMIN-NEXT:    fmv.w.x fa5, a0
-; RV32IZFBFMIN-NEXT:    fle.s s2, fa5, fs0
-; RV32IZFBFMIN-NEXT:    neg s3, s2
+; RV32IZFBFMIN-NEXT:    fle.s s0, fa5, fs0
 ; RV32IZFBFMIN-NEXT:    fmv.s fa0, fs0
 ; RV32IZFBFMIN-NEXT:    call __fixsfdi
-; RV32IZFBFMIN-NEXT:    and a0, s3, a0
-; RV32IZFBFMIN-NEXT:    or a0, s1, a0
-; RV32IZFBFMIN-NEXT:    feq.s a2, fs0, fs0
-; RV32IZFBFMIN-NEXT:    neg a2, a2
 ; RV32IZFBFMIN-NEXT:    lui a4, 524288
-; RV32IZFBFMIN-NEXT:    lui a3, 524288
-; RV32IZFBFMIN-NEXT:    beqz s2, .LBB10_2
+; RV32IZFBFMIN-NEXT:    lui a2, 524288
+; RV32IZFBFMIN-NEXT:    beqz s0, .LBB10_2
 ; RV32IZFBFMIN-NEXT:  # %bb.1: # %start
-; RV32IZFBFMIN-NEXT:    mv a3, a1
+; RV32IZFBFMIN-NEXT:    mv a2, a1
 ; RV32IZFBFMIN-NEXT:  .LBB10_2: # %start
-; RV32IZFBFMIN-NEXT:    and a0, a2, a0
-; RV32IZFBFMIN-NEXT:    beqz s0, .LBB10_4
+; RV32IZFBFMIN-NEXT:    lui a1, %hi(.LCPI10_0)
+; RV32IZFBFMIN-NEXT:    flw fa5, %lo(.LCPI10_0)(a1)
+; RV32IZFBFMIN-NEXT:    flt.s a3, fa5, fs0
+; RV32IZFBFMIN-NEXT:    beqz a3, .LBB10_4
 ; RV32IZFBFMIN-NEXT:  # %bb.3:
-; RV32IZFBFMIN-NEXT:    addi a3, a4, -1
+; RV32IZFBFMIN-NEXT:    addi a2, a4, -1
 ; RV32IZFBFMIN-NEXT:  .LBB10_4: # %start
-; RV32IZFBFMIN-NEXT:    and a1, a2, a3
-; RV32IZFBFMIN-NEXT:    lw ra, 28(sp) # 4-byte Folded Reload
-; RV32IZFBFMIN-NEXT:    lw s0, 24(sp) # 4-byte Folded Reload
-; RV32IZFBFMIN-NEXT:    lw s1, 20(sp) # 4-byte Folded Reload
-; RV32IZFBFMIN-NEXT:    lw s2, 16(sp) # 4-byte Folded Reload
-; RV32IZFBFMIN-NEXT:    lw s3, 12(sp) # 4-byte Folded Reload
-; RV32IZFBFMIN-NEXT:    flw fs0, 8(sp) # 4-byte Folded Reload
-; RV32IZFBFMIN-NEXT:    addi sp, sp, 32
+; RV32IZFBFMIN-NEXT:    feq.s a1, fs0, fs0
+; RV32IZFBFMIN-NEXT:    neg a4, a1
+; RV32IZFBFMIN-NEXT:    and a1, a4, a2
+; RV32IZFBFMIN-NEXT:    neg a2, a3
+; RV32IZFBFMIN-NEXT:    neg a3, s0
+; RV32IZFBFMIN-NEXT:    and a0, a3, a0
+; RV32IZFBFMIN-NEXT:    or a0, a2, a0
+; RV32IZFBFMIN-NEXT:    and a0, a4, a0
+; RV32IZFBFMIN-NEXT:    lw ra, 12(sp) # 4-byte Folded Reload
+; RV32IZFBFMIN-NEXT:    lw s0, 8(sp) # 4-byte Folded Reload
+; RV32IZFBFMIN-NEXT:    flw fs0, 4(sp) # 4-byte Folded Reload
+; RV32IZFBFMIN-NEXT:    addi sp, sp, 16
 ; RV32IZFBFMIN-NEXT:    ret
 ;
 ; R32IDZFBFMIN-LABEL: fcvt_l_bf16_sat:
 ; R32IDZFBFMIN:       # %bb.0: # %start
-; R32IDZFBFMIN-NEXT:    addi sp, sp, -32
-; R32IDZFBFMIN-NEXT:    sw ra, 28(sp) # 4-byte Folded Spill
-; R32IDZFBFMIN-NEXT:    sw s0, 24(sp) # 4-byte Folded Spill
-; R32IDZFBFMIN-NEXT:    sw s1, 20(sp) # 4-byte Folded Spill
-; R32IDZFBFMIN-NEXT:    sw s2, 16(sp) # 4-byte Folded Spill
-; R32IDZFBFMIN-NEXT:    sw s3, 12(sp) # 4-byte Folded Spill
+; R32IDZFBFMIN-NEXT:    addi sp, sp, -16
+; R32IDZFBFMIN-NEXT:    sw ra, 12(sp) # 4-byte Folded Spill
+; R32IDZFBFMIN-NEXT:    sw s0, 8(sp) # 4-byte Folded Spill
 ; R32IDZFBFMIN-NEXT:    fsd fs0, 0(sp) # 8-byte Folded Spill
-; R32IDZFBFMIN-NEXT:    lui a0, %hi(.LCPI10_0)
-; R32IDZFBFMIN-NEXT:    flw fa5, %lo(.LCPI10_0)(a0)
 ; R32IDZFBFMIN-NEXT:    fcvt.s.bf16 fs0, fa0
-; R32IDZFBFMIN-NEXT:    flt.s s0, fa5, fs0
-; R32IDZFBFMIN-NEXT:    neg s1, s0
 ; R32IDZFBFMIN-NEXT:    lui a0, 913408
 ; R32IDZFBFMIN-NEXT:    fmv.w.x fa5, a0
-; R32IDZFBFMIN-NEXT:    fle.s s2, fa5, fs0
-; R32IDZFBFMIN-NEXT:    neg s3, s2
+; R32IDZFBFMIN-NEXT:    fle.s s0, fa5, fs0
 ; R32IDZFBFMIN-NEXT:    fmv.s fa0, fs0
 ; R32IDZFBFMIN-NEXT:    call __fixsfdi
-; R32IDZFBFMIN-NEXT:    and a0, s3, a0
-; R32IDZFBFMIN-NEXT:    or a0, s1, a0
-; R32IDZFBFMIN-NEXT:    feq.s a2, fs0, fs0
-; R32IDZFBFMIN-NEXT:    neg a2, a2
 ; R32IDZFBFMIN-NEXT:    lui a4, 524288
-; R32IDZFBFMIN-NEXT:    lui a3, 524288
-; R32IDZFBFMIN-NEXT:    beqz s2, .LBB10_2
+; R32IDZFBFMIN-NEXT:    lui a2, 524288
+; R32IDZFBFMIN-NEXT:    beqz s0, .LBB10_2
 ; R32IDZFBFMIN-NEXT:  # %bb.1: # %start
-; R32IDZFBFMIN-NEXT:    mv a3, a1
+; R32IDZFBFMIN-NEXT:    mv a2, a1
 ; R32IDZFBFMIN-NEXT:  .LBB10_2: # %start
-; R32IDZFBFMIN-NEXT:    and a0, a2, a0
-; R32IDZFBFMIN-NEXT:    beqz s0, .LBB10_4
+; R32IDZFBFMIN-NEXT:    lui a1, %hi(.LCPI10_0)
+; R32IDZFBFMIN-NEXT:    flw fa5, %lo(.LCPI10_0)(a1)
+; R32IDZFBFMIN-NEXT:    flt.s a3, fa5, fs0
+; R32IDZFBFMIN-NEXT:    beqz a3, .LBB10_4
 ; R32IDZFBFMIN-NEXT:  # %bb.3:
-; R32IDZFBFMIN-NEXT:    addi a3, a4, -1
+; R32IDZFBFMIN-NEXT:    addi a2, a4, -1
 ; R32IDZFBFMIN-NEXT:  .LBB10_4: # %start
-; R32IDZFBFMIN-NEXT:    and a1, a2, a3
-; R32IDZFBFMIN-NEXT:    lw ra, 28(sp) # 4-byte Folded Reload
-; R32IDZFBFMIN-NEXT:    lw s0, 24(sp) # 4-byte Folded Reload
-; R32IDZFBFMIN-NEXT:    lw s1, 20(sp) # 4-byte Folded Reload
-; R32IDZFBFMIN-NEXT:    lw s2, 16(sp) # 4-byte Folded Reload
-; R32IDZFBFMIN-NEXT:    lw s3, 12(sp) # 4-byte Folded Reload
+; R32IDZFBFMIN-NEXT:    feq.s a1, fs0, fs0
+; R32IDZFBFMIN-NEXT:    neg a4, a1
+; R32IDZFBFMIN-NEXT:    and a1, a4, a2
+; R32IDZFBFMIN-NEXT:    neg a2, a3
+; R32IDZFBFMIN-NEXT:    neg a3, s0
+; R32IDZFBFMIN-NEXT:    and a0, a3, a0
+; R32IDZFBFMIN-NEXT:    or a0, a2, a0
+; R32IDZFBFMIN-NEXT:    and a0, a4, a0
+; R32IDZFBFMIN-NEXT:    lw ra, 12(sp) # 4-byte Folded Reload
+; R32IDZFBFMIN-NEXT:    lw s0, 8(sp) # 4-byte Folded Reload
 ; R32IDZFBFMIN-NEXT:    fld fs0, 0(sp) # 8-byte Folded Reload
-; R32IDZFBFMIN-NEXT:    addi sp, sp, 32
+; R32IDZFBFMIN-NEXT:    addi sp, sp, 16
 ; R32IDZFBFMIN-NEXT:    ret
 ;
 ; RV32ID-LABEL: fcvt_l_bf16_sat:
diff --git a/llvm/test/CodeGen/RISCV/bitreverse-shift.ll b/llvm/test/CodeGen/RISCV/bitreverse-shift.ll
index f29b16991726..b0281ba7d238 100644
--- a/llvm/test/CodeGen/RISCV/bitreverse-shift.ll
+++ b/llvm/test/CodeGen/RISCV/bitreverse-shift.ll
@@ -4,7 +4,7 @@
 ; RUN: llc -mtriple=riscv64 -mattr=+zbkb -verify-machineinstrs < %s \
 ; RUN:   | FileCheck %s -check-prefixes=RV64ZBKB
 
-; TODO: These tests can be optmised
+; TODO: These tests can be optimised
 ;       fold (bitreverse(srl (bitreverse c), x)) -> (shl c, x)
 ;       fold (bitreverse(shl (bitreverse c), x)) -> (srl c, x)
 
diff --git a/llvm/test/CodeGen/RISCV/double-convert.ll b/llvm/test/CodeGen/RISCV/double-convert.ll
index c147d6ec6d9b..6024a29da33d 100644
--- a/llvm/test/CodeGen/RISCV/double-convert.ll
+++ b/llvm/test/CodeGen/RISCV/double-convert.ll
@@ -692,28 +692,27 @@ define i64 @fcvt_l_d_sat(double %a) nounwind {
 ; RV32IFD-NEXT:    fmv.d fs0, fa0
 ; RV32IFD-NEXT:    fle.d s0, fa5, fa0
 ; RV32IFD-NEXT:    call __fixdfdi
-; RV32IFD-NEXT:    lui a3, 524288
-; RV32IFD-NEXT:    li a4, 1
+; RV32IFD-NEXT:    lui a4, 524288
 ; RV32IFD-NEXT:    lui a2, 524288
-; RV32IFD-NEXT:    bne s0, a4, .LBB12_2
+; RV32IFD-NEXT:    beqz s0, .LBB12_2
 ; RV32IFD-NEXT:  # %bb.1: # %start
 ; RV32IFD-NEXT:    mv a2, a1
 ; RV32IFD-NEXT:  .LBB12_2: # %start
 ; RV32IFD-NEXT:    lui a1, %hi(.LCPI12_1)
 ; RV32IFD-NEXT:    fld fa5, %lo(.LCPI12_1)(a1)
-; RV32IFD-NEXT:    flt.d a4, fa5, fs0
-; RV32IFD-NEXT:    beqz a4, .LBB12_4
+; RV32IFD-NEXT:    flt.d a3, fa5, fs0
+; RV32IFD-NEXT:    beqz a3, .LBB12_4
 ; RV32IFD-NEXT:  # %bb.3:
-; RV32IFD-NEXT:    addi a2, a3, -1
+; RV32IFD-NEXT:    addi a2, a4, -1
 ; RV32IFD-NEXT:  .LBB12_4: # %start
 ; RV32IFD-NEXT:    feq.d a1, fs0, fs0
-; RV32IFD-NEXT:    neg a3, a1
-; RV32IFD-NEXT:    and a1, a3, a2
-; RV32IFD-NEXT:    neg a2, a4
-; RV32IFD-NEXT:    neg a4, s0
-; RV32IFD-NEXT:    and a0, a4, a0
-; RV32IFD-NEXT:    or a0, a2, a0
+; RV32IFD-NEXT:    neg a4, a1
+; RV32IFD-NEXT:    and a1, a4, a2
+; RV32IFD-NEXT:    neg a2, a3
+; RV32IFD-NEXT:    neg a3, s0
 ; RV32IFD-NEXT:    and a0, a3, a0
+; RV32IFD-NEXT:    or a0, a2, a0
+; RV32IFD-NEXT:    and a0, a4, a0
 ; RV32IFD-NEXT:    lw ra, 12(sp) # 4-byte Folded Reload
 ; RV32IFD-NEXT:    lw s0, 8(sp) # 4-byte Folded Reload
 ; RV32IFD-NEXT:    fld fs0, 0(sp) # 8-byte Folded Reload
@@ -790,33 +789,32 @@ define i64 @fcvt_l_d_sat(double %a) nounwind {
 ; RV32I-NEXT:    sw s3, 12(sp) # 4-byte Folded Spill
 ; RV32I-NEXT:    sw s4, 8(sp) # 4-byte Folded Spill
 ; RV32I-NEXT:    sw s5, 4(sp) # 4-byte Folded Spill
-; RV32I-NEXT:    sw s6, 0(sp) # 4-byte Folded Spill
 ; RV32I-NEXT:    mv s0, a1
 ; RV32I-NEXT:    mv s1, a0
+; RV32I-NEXT:    lui a3, 278016
+; RV32I-NEXT:    addi a3, a3, -1
+; RV32I-NEXT:    li a2, -1
+; RV32I-NEXT:    call __gtdf2
+; RV32I-NEXT:    mv s2, a0
 ; RV32I-NEXT:    lui a3, 802304
+; RV32I-NEXT:    mv a0, s1
+; RV32I-NEXT:    mv a1, s0
 ; RV32I-NEXT:    li a2, 0
 ; RV32I-NEXT:    call __gedf2
-; RV32I-NEXT:    mv s2, a0
+; RV32I-NEXT:    mv s3, a0
 ; RV32I-NEXT:    mv a0, s1
 ; RV32I-NEXT:    mv a1, s0
 ; RV32I-NEXT:    call __fixdfdi
-; RV32I-NEXT:    mv s3, a0
-; RV32I-NEXT:    mv s4, a1
-; RV32I-NEXT:    lui s6, 524288
-; RV32I-NEXT:    bgez s2, .LBB12_2
+; RV32I-NEXT:    mv s4, a0
+; RV32I-NEXT:    mv s5, a1
+; RV32I-NEXT:    lui a0, 524288
+; RV32I-NEXT:    bgez s3, .LBB12_2
 ; RV32I-NEXT:  # %bb.1: # %start
-; RV32I-NEXT:    lui s4, 524288
+; RV32I-NEXT:    lui s5, 524288
 ; RV32I-NEXT:  .LBB12_2: # %start
-; RV32I-NEXT:    lui a3, 278016
-; RV32I-NEXT:    addi a3, a3, -1
-; RV32I-NEXT:    li a2, -1
-; RV32I-NEXT:    mv a0, s1
-; RV32I-NEXT:    mv a1, s0
-; RV32I-NEXT:    call __gtdf2
-; RV32I-NEXT:    mv s5, a0
-; RV32I-NEXT:    blez a0, .LBB12_4
+; RV32I-NEXT:    blez s2, .LBB12_4
 ; RV32I-NEXT:  # %bb.3: # %start
-; RV32I-NEXT:    addi s4, s6, -1
+; RV32I-NEXT:    addi s5, a0, -1
 ; RV32I-NEXT:  .LBB12_4: # %start
 ; RV32I-NEXT:    mv a0, s1
 ; RV32I-NEXT:    mv a1, s0
@@ -825,11 +823,11 @@ define i64 @fcvt_l_d_sat(double %a) nounwind {
 ; RV32I-NEXT:    call __unorddf2
 ; RV32I-NEXT:    snez a0, a0
 ; RV32I-NEXT:    addi a0, a0, -1
-; RV32I-NEXT:    and a1, a0, s4
-; RV32I-NEXT:    slti a2, s2, 0
+; RV32I-NEXT:    and a1, a0, s5
+; RV32I-NEXT:    slti a2, s3, 0
 ; RV32I-NEXT:    addi a2, a2, -1
-; RV32I-NEXT:    and a2, a2, s3
-; RV32I-NEXT:    sgtz a3, s5
+; RV32I-NEXT:    and a2, a2, s4
+; RV32I-NEXT:    sgtz a3, s2
 ; RV32I-NEXT:    neg a3, a3
 ; RV32I-NEXT:    or a2, a3, a2
 ; RV32I-NEXT:    and a0, a0, a2
@@ -840,7 +838,6 @@ define i64 @fcvt_l_d_sat(double %a) nounwind {
 ; RV32I-NEXT:    lw s3, 12(sp) # 4-byte Folded Reload
 ; RV32I-NEXT:    lw s4, 8(sp) # 4-byte Folded Reload
 ; RV32I-NEXT:    lw s5, 4(sp) # 4-byte Folded Reload
-; RV32I-NEXT:    lw s6, 0(sp) # 4-byte Folded Reload
 ; RV32I-NEXT:    addi sp, sp, 32
 ; RV32I-NEXT:    ret
 ;
@@ -949,22 +946,23 @@ define i64 @fcvt_lu_d_sat(double %a) nounwind {
 ; RV32IFD-NEXT:    addi sp, sp, -16
 ; RV32IFD-NEXT:    sw ra, 12(sp) # 4-byte Folded Spill
 ; RV32IFD-NEXT:    sw s0, 8(sp) # 4-byte Folded Spill
-; RV32IFD-NEXT:    sw s1, 4(sp) # 4-byte Folded Spill
-; RV32IFD-NEXT:    lui a0, %hi(.LCPI14_0)
-; RV32IFD-NEXT:    fld fa5, %lo(.LCPI14_0)(a0)
-; RV32IFD-NEXT:    flt.d a0, fa5, fa0
-; RV32IFD-NEXT:    neg s0, a0
+; RV32IFD-NEXT:    fsd fs0, 0(sp) # 8-byte Folded Spill
+; RV32IFD-NEXT:    fmv.d fs0, fa0
 ; RV32IFD-NEXT:    fcvt.d.w fa5, zero
 ; RV32IFD-NEXT:    fle.d a0, fa5, fa0
-; RV32IFD-NEXT:    neg s1, a0
+; RV32IFD-NEXT:    neg s0, a0
 ; RV32IFD-NEXT:    call __fixunsdfdi
-; RV32IFD-NEXT:    and a0, s1, a0
-; RV32IFD-NEXT:    or a0, s0, a0
-; RV32IFD-NEXT:    and a1, s1, a1
-; RV32IFD-NEXT:    or a1, s0, a1
+; RV32IFD-NEXT:    lui a2, %hi(.LCPI14_0)
+; RV32IFD-NEXT:    fld fa5, %lo(.LCPI14_0)(a2)
+; RV32IFD-NEXT:    and a0, s0, a0
+; RV32IFD-NEXT:    flt.d a2, fa5, fs0
+; RV32IFD-NEXT:    neg a2, a2
+; RV32IFD-NEXT:    or a0, a2, a0
+; RV32IFD-NEXT:    and a1, s0, a1
+; RV32IFD-NEXT:    or a1, a2, a1
 ; RV32IFD-NEXT:    lw ra, 12(sp) # 4-byte Folded Reload
 ; RV32IFD-NEXT:    lw s0, 8(sp) # 4-byte Folded Reload
-; RV32IFD-NEXT:    lw s1, 4(sp) # 4-byte Folded Reload
+; RV32IFD-NEXT:    fld fs0, 0(sp) # 8-byte Folded Reload
 ; RV32IFD-NEXT:    addi sp, sp, 16
 ; RV32IFD-NEXT:    ret
 ;
@@ -983,27 +981,24 @@ define i64 @fcvt_lu_d_sat(double %a) nounwind {
 ; RV32IZFINXZDINX-NEXT:    sw ra, 12(sp) # 4-byte Folded Spill
 ; RV32IZFINXZDINX-NEXT:    sw s0, 8(sp) # 4-byte Folded Spill
 ; RV32IZFINXZDINX-NEXT:    sw s1, 4(sp) # 4-byte Folded Spill
-; RV32IZFINXZDINX-NEXT:    sw s2, 0(sp) # 4-byte Folded Spill
 ; RV32IZFINXZDINX-NEXT:    mv s1, a1
-; RV32IZFINXZDINX-NEXT:    fcvt.d.w a2, zero
 ; RV32IZFINXZDINX-NEXT:    mv s0, a0
-; RV32IZFINXZDINX-NEXT:    fle.d a0, a2, s0
-; RV32IZFINXZDINX-NEXT:    neg s2, a0
-; RV32IZFINXZDINX-NEXT:    mv a0, s0
 ; RV32IZFINXZDINX-NEXT:    call __fixunsdfdi
-; RV32IZFINXZDINX-NEXT:    lui a2, %hi(.LCPI14_0)
-; RV32IZFINXZDINX-NEXT:    lw a3, %lo(.LCPI14_0+4)(a2)
-; RV32IZFINXZDINX-NEXT:    lw a2, %lo(.LCPI14_0)(a2)
-; RV32IZFINXZDINX-NEXT:    and a0, s2, a0
-; RV32IZFINXZDINX-NEXT:    flt.d a2, a2, s0
+; RV32IZFINXZDINX-NEXT:    fcvt.d.w a2, zero
+; RV32IZFINXZDINX-NEXT:    lui a4, %hi(.LCPI14_0)
+; RV32IZFINXZDINX-NEXT:    lw a5, %lo(.LCPI14_0+4)(a4)
+; RV32IZFINXZDINX-NEXT:    lw a4, %lo(.LCPI14_0)(a4)
+; RV32IZFINXZDINX-NEXT:    fle.d a2, a2, s0
 ; RV32IZFINXZDINX-NEXT:    neg a2, a2
-; RV32IZFINXZDINX-NEXT:    or a0, a2, a0
-; RV32IZFINXZDINX-NEXT:    and a1, s2, a1
-; RV32IZFINXZDINX-NEXT:    or a1, a2, a1
+; RV32IZFINXZDINX-NEXT:    and a0, a2, a0
+; RV32IZFINXZDINX-NEXT:    flt.d a3, a4, s0
+; RV32IZFINXZDINX-NEXT:    neg a3, a3
+; RV32IZFINXZDINX-NEXT:    or a0, a3, a0
+; RV32IZFINXZDINX-NEXT:    and a1, a2, a1
+; RV32IZFINXZDINX-NEXT:    or a1, a3, a1
 ; RV32IZFINXZDINX-NEXT:    lw ra, 12(sp) # 4-byte Folded Reload
 ; RV32IZFINXZDINX-NEXT:    lw s0, 8(sp) # 4-byte Folded Reload
 ; RV32IZFINXZDINX-NEXT:    lw s1, 4(sp) # 4-byte Folded Reload
-; RV32IZFINXZDINX-NEXT:    lw s2, 0(sp) # 4-byte Folded Reload
 ; RV32IZFINXZDINX-NEXT:    addi sp, sp, 16
 ; RV32IZFINXZDINX-NEXT:    ret
 ;
diff --git a/llvm/test/CodeGen/RISCV/double-round-conv-sat.ll b/llvm/test/CodeGen/RISCV/double-round-conv-sat.ll
index f1c56b320b76..927eee2e9e54 100644
--- a/llvm/test/CodeGen/RISCV/double-round-conv-sat.ll
+++ b/llvm/test/CodeGen/RISCV/double-round-conv-sat.ll
@@ -53,28 +53,27 @@ define i64 @test_floor_si64(double %x) nounwind {
 ; RV32IFD-NEXT:    fmv.d fs0, fa0
 ; RV32IFD-NEXT:    fle.d s0, fa5, fa0
 ; RV32IFD-NEXT:    call __fixdfdi
-; RV32IFD-NEXT:    lui a3, 524288
-; RV32IFD-NEXT:    li a4, 1
+; RV32IFD-NEXT:    lui a4, 524288
 ; RV32IFD-NEXT:    lui a2, 524288
-; RV32IFD-NEXT:    bne s0, a4, .LBB1_2
+; RV32IFD-NEXT:    beqz s0, .LBB1_2
 ; RV32IFD-NEXT:  # %bb.1:
 ; RV32IFD-NEXT:    mv a2, a1
 ; RV32IFD-NEXT:  .LBB1_2:
 ; RV32IFD-NEXT:    lui a1, %hi(.LCPI1_1)
 ; RV32IFD-NEXT:    fld fa5, %lo(.LCPI1_1)(a1)
-; RV32IFD-NEXT:    flt.d a4, fa5, fs0
-; RV32IFD-NEXT:    beqz a4, .LBB1_4
+; RV32IFD-NEXT:    flt.d a3, fa5, fs0
+; RV32IFD-NEXT:    beqz a3, .LBB1_4
 ; RV32IFD-NEXT:  # %bb.3:
-; RV32IFD-NEXT:    addi a2, a3, -1
+; RV32IFD-NEXT:    addi a2, a4, -1
 ; RV32IFD-NEXT:  .LBB1_4:
 ; RV32IFD-NEXT:    feq.d a1, fs0, fs0
-; RV32IFD-NEXT:    neg a3, a1
-; RV32IFD-NEXT:    and a1, a3, a2
-; RV32IFD-NEXT:    neg a2, a4
-; RV32IFD-NEXT:    neg a4, s0
-; RV32IFD-NEXT:    and a0, a4, a0
-; RV32IFD-NEXT:    or a0, a2, a0
+; RV32IFD-NEXT:    neg a4, a1
+; RV32IFD-NEXT:    and a1, a4, a2
+; RV32IFD-NEXT:    neg a2, a3
+; RV32IFD-NEXT:    neg a3, s0
 ; RV32IFD-NEXT:    and a0, a3, a0
+; RV32IFD-NEXT:    or a0, a2, a0
+; RV32IFD-NEXT:    and a0, a4, a0
 ; RV32IFD-NEXT:    lw ra, 12(sp) # 4-byte Folded Reload
 ; RV32IFD-NEXT:    lw s0, 8(sp) # 4-byte Folded Reload
 ; RV32IFD-NEXT:    fld fs0, 0(sp) # 8-byte Folded Reload
@@ -299,28 +298,27 @@ define i64 @test_ceil_si64(double %x) nounwind {
 ; RV32IFD-NEXT:    fmv.d fs0, fa0
 ; RV32IFD-NEXT:    fle.d s0, fa5, fa0
 ; RV32IFD-NEXT:    call __fixdfdi
-; RV32IFD-NEXT:    lui a3, 524288
-; RV32IFD-NEXT:    li a4, 1
+; RV32IFD-NEXT:    lui a4, 524288
 ; RV32IFD-NEXT:    lui a2, 524288
-; RV32IFD-NEXT:    bne s0, a4, .LBB5_2
+; RV32IFD-NEXT:    beqz s0, .LBB5_2
 ; RV32IFD-NEXT:  # %bb.1:
 ; RV32IFD-NEXT:    mv a2, a1
 ; RV32IFD-NEXT:  .LBB5_2:
 ; RV32IFD-NEXT:    lui a1, %hi(.LCPI5_1)
 ; RV32IFD-NEXT:    fld fa5, %lo(.LCPI5_1)(a1)
-; RV32IFD-NEXT:    flt.d a4, fa5, fs0
-; RV32IFD-NEXT:    beqz a4, .LBB5_4
+; RV32IFD-NEXT:    flt.d a3, fa5, fs0
+; RV32IFD-NEXT:    beqz a3, .LBB5_4
 ; RV32IFD-NEXT:  # %bb.3:
-; RV32IFD-NEXT:    addi a2, a3, -1
+; RV32IFD-NEXT:    addi a2, a4, -1
 ; RV32IFD-NEXT:  .LBB5_4:
 ; RV32IFD-NEXT:    feq.d a1, fs0, fs0
-; RV32IFD-NEXT:    neg a3, a1
-; RV32IFD-NEXT:    and a1, a3, a2
-; RV32IFD-NEXT:    neg a2, a4
-; RV32IFD-NEXT:    neg a4, s0
-; RV32IFD-NEXT:    and a0, a4, a0
-; RV32IFD-NEXT:    or a0, a2, a0
+; RV32IFD-NEXT:    neg a4, a1
+; RV32IFD-NEXT:    and a1, a4, a2
+; RV32IFD-NEXT:    neg a2, a3
+; RV32IFD-NEXT:    neg a3, s0
 ; RV32IFD-NEXT:    and a0, a3, a0
+; RV32IFD-NEXT:    or a0, a2, a0
+; RV32IFD-NEXT:    and a0, a4, a0
 ; RV32IFD-NEXT:    lw ra, 12(sp) # 4-byte Folded Reload
 ; RV32IFD-NEXT:    lw s0, 8(sp) # 4-byte Folded Reload
 ; RV32IFD-NEXT:    fld fs0, 0(sp) # 8-byte Folded Reload
@@ -545,28 +543,27 @@ define i64 @test_trunc_si64(double %x) nounwind {
 ; RV32IFD-NEXT:    fmv.d fs0, fa0
 ; RV32IFD-NEXT:    fle.d s0, fa5, fa0
 ; RV32IFD-NEXT:    call __fixdfdi
-; RV32IFD-NEXT:    lui a3, 524288
-; RV32IFD-NEXT:    li a4, 1
+; RV32IFD-NEXT:    lui a4, 524288
 ; RV32IFD-NEXT:    lui a2, 524288
-; RV32IFD-NEXT:    bne s0, a4, .LBB9_2
+; RV32IFD-NEXT:    beqz s0, .LBB9_2
 ; RV32IFD-NEXT:  # %bb.1:
 ; RV32IFD-NEXT:    mv a2, a1
 ; RV32IFD-NEXT:  .LBB9_2:
 ; RV32IFD-NEXT:    lui a1, %hi(.LCPI9_1)
 ; RV32IFD-NEXT:    fld fa5, %lo(.LCPI9_1)(a1)
-; RV32IFD-NEXT:    flt.d a4, fa5, fs0
-; RV32IFD-NEXT:    beqz a4, .LBB9_4
+; RV32IFD-NEXT:    flt.d a3, fa5, fs0
+; RV32IFD-NEXT:    beqz a3, .LBB9_4
 ; RV32IFD-NEXT:  # %bb.3:
-; RV32IFD-NEXT:    addi a2, a3, -1
+; RV32IFD-NEXT:    addi a2, a4, -1
 ; RV32IFD-NEXT:  .LBB9_4:
 ; RV32IFD-NEXT:    feq.d a1, fs0, fs0
-; RV32IFD-NEXT:    neg a3, a1
-; RV32IFD-NEXT:    and a1, a3, a2
-; RV32IFD-NEXT:    neg a2, a4
-; RV32IFD-NEXT:    neg a4, s0
-; RV32IFD-NEXT:    and a0, a4, a0
-; RV32IFD-NEXT:    or a0, a2, a0
+; RV32IFD-NEXT:    neg a4, a1
+; RV32IFD-NEXT:    and a1, a4, a2
+; RV32IFD-NEXT:    neg a2, a3
+; RV32IFD-NEXT:    neg a3, s0
 ; RV32IFD-NEXT:    and a0, a3, a0
+; RV32IFD-NEXT:    or a0, a2, a0
+; RV32IFD-NEXT:    and a0, a4, a0
 ; RV32IFD-NEXT:    lw ra, 12(sp) # 4-byte Folded Reload
 ; RV32IFD-NEXT:    lw s0, 8(sp) # 4-byte Folded Reload
 ; RV32IFD-NEXT:    fld fs0, 0(sp) # 8-byte Folded Reload
@@ -791,28 +788,27 @@ define i64 @test_round_si64(double %x) nounwind {
 ; RV32IFD-NEXT:    fmv.d fs0, fa0
 ; RV32IFD-NEXT:    fle.d s0, fa5, fa0
 ; RV32IFD-NEXT:    call __fixdfdi
-; RV32IFD-NEXT:    lui a3, 524288
-; RV32IFD-NEXT:    li a4, 1
+; RV32IFD-NEXT:    lui a4, 524288
 ; RV32IFD-NEXT:    lui a2, 524288
-; RV32IFD-NEXT:    bne s0, a4, .LBB13_2
+; RV32IFD-NEXT:    beqz s0, .LBB13_2
 ; RV32IFD-NEXT:  # %bb.1:
 ; RV32IFD-NEXT:    mv a2, a1
 ; RV32IFD-NEXT:  .LBB13_2:
 ; RV32IFD-NEXT:    lui a1, %hi(.LCPI13_1)
 ; RV32IFD-NEXT:    fld fa5, %lo(.LCPI13_1)(a1)
-; RV32IFD-NEXT:    flt.d a4, fa5, fs0
-; RV32IFD-NEXT:    beqz a4, .LBB13_4
+; RV32IFD-NEXT:    flt.d a3, fa5, fs0
+; RV32IFD-NEXT:    beqz a3, .LBB13_4
 ; RV32IFD-NEXT:  # %bb.3:
-; RV32IFD-NEXT:    addi a2, a3, -1
+; RV32IFD-NEXT:    addi a2, a4, -1
 ; RV32IFD-NEXT:  .LBB13_4:
 ; RV32IFD-NEXT:    feq.d a1, fs0, fs0
-; RV32IFD-NEXT:    neg a3, a1
-; RV32IFD-NEXT:    and a1, a3, a2
-; RV32IFD-NEXT:    neg a2, a4
-; RV32IFD-NEXT:    neg a4, s0
-; RV32IFD-NEXT:    and a0, a4, a0
-; RV32IFD-NEXT:    or a0, a2, a0
+; RV32IFD-NEXT:    neg a4, a1
+; RV32IFD-NEXT:    and a1, a4, a2
+; RV32IFD-NEXT:    neg a2, a3
+; RV32IFD-NEXT:    neg a3, s0
 ; RV32IFD-NEXT:    and a0, a3, a0
+; RV32IFD-NEXT:    or a0, a2, a0
+; RV32IFD-NEXT:    and a0, a4, a0
 ; RV32IFD-NEXT:    lw ra, 12(sp) # 4-byte Folded Reload
 ; RV32IFD-NEXT:    lw s0, 8(sp) # 4-byte Folded Reload
 ; RV32IFD-NEXT:    fld fs0, 0(sp) # 8-byte Folded Reload
@@ -1037,28 +1033,27 @@ define i64 @test_roundeven_si64(double %x) nounwind {
 ; RV32IFD-NEXT:    fmv.d fs0, fa0
 ; RV32IFD-NEXT:    fle.d s0, fa5, fa0
 ; RV32IFD-NEXT:    call __fixdfdi
-; RV32IFD-NEXT:    lui a3, 524288
-; RV32IFD-NEXT:    li a4, 1
+; RV32IFD-NEXT:    lui a4, 524288
 ; RV32IFD-NEXT:    lui a2, 524288
-; RV32IFD-NEXT:    bne s0, a4, .LBB17_2
+; RV32IFD-NEXT:    beqz s0, .LBB17_2
 ; RV32IFD-NEXT:  # %bb.1:
 ; RV32IFD-NEXT:    mv a2, a1
 ; RV32IFD-NEXT:  .LBB17_2:
 ; RV32IFD-NEXT:    lui a1, %hi(.LCPI17_1)
 ; RV32IFD-NEXT:    fld fa5, %lo(.LCPI17_1)(a1)
-; RV32IFD-NEXT:    flt.d a4, fa5, fs0
-; RV32IFD-NEXT:    beqz a4, .LBB17_4
+; RV32IFD-NEXT:    flt.d a3, fa5, fs0
+; RV32IFD-NEXT:    beqz a3, .LBB17_4
 ; RV32IFD-NEXT:  # %bb.3:
-; RV32IFD-NEXT:    addi a2, a3, -1
+; RV32IFD-NEXT:    addi a2, a4, -1
 ; RV32IFD-NEXT:  .LBB17_4:
 ; RV32IFD-NEXT:    feq.d a1, fs0, fs0
-; RV32IFD-NEXT:    neg a3, a1
-; RV32IFD-NEXT:    and a1, a3, a2
-; RV32IFD-NEXT:    neg a2, a4
-; RV32IFD-NEXT:    neg a4, s0
-; RV32IFD-NEXT:    and a0, a4, a0
-; RV32IFD-NEXT:    or a0, a2, a0
+; RV32IFD-NEXT:    neg a4, a1
+; RV32IFD-NEXT:    and a1, a4, a2
+; RV32IFD-NEXT:    neg a2, a3
+; RV32IFD-NEXT:    neg a3, s0
 ; RV32IFD-NEXT:    and a0, a3, a0
+; RV32IFD-NEXT:    or a0, a2, a0
+; RV32IFD-NEXT:    and a0, a4, a0
 ; RV32IFD-NEXT:    lw ra, 12(sp) # 4-byte Folded Reload
 ; RV32IFD-NEXT:    lw s0, 8(sp) # 4-byte Folded Reload
 ; RV32IFD-NEXT:    fld fs0, 0(sp) # 8-byte Folded Reload
@@ -1283,28 +1278,27 @@ define i64 @test_rint_si64(double %x) nounwind {
 ; RV32IFD-NEXT:    fmv.d fs0, fa0
 ; RV32IFD-NEXT:    fle.d s0, fa5, fa0
 ; RV32IFD-NEXT:    call __fixdfdi
-; RV32IFD-NEXT:    lui a3, 524288
-; RV32IFD-NEXT:    li a4, 1
+; RV32IFD-NEXT:    lui a4, 524288
 ; RV32IFD-NEXT:    lui a2, 524288
-; RV32IFD-NEXT:    bne s0, a4, .LBB21_2
+; RV32IFD-NEXT:    beqz s0, .LBB21_2
 ; RV32IFD-NEXT:  # %bb.1:
 ; RV32IFD-NEXT:    mv a2, a1
 ; RV32IFD-NEXT:  .LBB21_2:
 ; RV32IFD-NEXT:    lui a1, %hi(.LCPI21_1)
 ; RV32IFD-NEXT:    fld fa5, %lo(.LCPI21_1)(a1)
-; RV32IFD-NEXT:    flt.d a4, fa5, fs0
-; RV32IFD-NEXT:    beqz a4, .LBB21_4
+; RV32IFD-NEXT:    flt.d a3, fa5, fs0
+; RV32IFD-NEXT:    beqz a3, .LBB21_4
 ; RV32IFD-NEXT:  # %bb.3:
-; RV32IFD-NEXT:    addi a2, a3, -1
+; RV32IFD-NEXT:    addi a2, a4, -1
 ; RV32IFD-NEXT:  .LBB21_4:
 ; RV32IFD-NEXT:    feq.d a1, fs0, fs0
-; RV32IFD-NEXT:    neg a3, a1
-; RV32IFD-NEXT:    and a1, a3, a2
-; RV32IFD-NEXT:    neg a2, a4
-; RV32IFD-NEXT:    neg a4, s0
-; RV32IFD-NEXT:    and a0, a4, a0
-; RV32IFD-NEXT:    or a0, a2, a0
+; RV32IFD-NEXT:    neg a4, a1
+; RV32IFD-NEXT:    and a1, a4, a2
+; RV32IFD-NEXT:    neg a2, a3
+; RV32IFD-NEXT:    neg a3, s0
 ; RV32IFD-NEXT:    and a0, a3, a0
+; RV32IFD-NEXT:    or a0, a2, a0
+; RV32IFD-NEXT:    and a0, a4, a0
 ; RV32IFD-NEXT:    lw ra, 12(sp) # 4-byte Folded Reload
 ; RV32IFD-NEXT:    lw s0, 8(sp) # 4-byte Folded Reload
 ; RV32IFD-NEXT:    fld fs0, 0(sp) # 8-byte Folded Reload
diff --git a/llvm/test/CodeGen/RISCV/fixups-diff.ll b/llvm/test/CodeGen/RISCV/fixups-diff.ll
index cc1c87b1fe37..84a7d18ed150 100644
--- a/llvm/test/CodeGen/RISCV/fixups-diff.ll
+++ b/llvm/test/CodeGen/RISCV/fixups-diff.ll
@@ -27,7 +27,7 @@ entry:
 ; CHECK:      }
 
 ; CHECK:      Section {{.*}} .rela.eh_frame {
-; CHECK-NEXT:   0x1C R_RISCV_32_PCREL <null> 0x0
+; CHECK-NEXT:   0x1C R_RISCV_32_PCREL .L0  0x0
 ; CHECK-NEXT: }
 
 !llvm.dbg.cu = !{!0}
diff --git a/llvm/test/CodeGen/RISCV/float-convert.ll b/llvm/test/CodeGen/RISCV/float-convert.ll
index 653b64ec7304..7eabd3f5f227 100644
--- a/llvm/test/CodeGen/RISCV/float-convert.ll
+++ b/llvm/test/CodeGen/RISCV/float-convert.ll
@@ -275,26 +275,24 @@ define i32 @fcvt_wu_s_sat(float %a) nounwind {
 ; RV32I-NEXT:    sw ra, 12(sp) # 4-byte Folded Spill
 ; RV32I-NEXT:    sw s0, 8(sp) # 4-byte Folded Spill
 ; RV32I-NEXT:    sw s1, 4(sp) # 4-byte Folded Spill
-; RV32I-NEXT:    sw s2, 0(sp) # 4-byte Folded Spill
 ; RV32I-NEXT:    mv s0, a0
-; RV32I-NEXT:    lui a1, 325632
-; RV32I-NEXT:    addi a1, a1, -1
-; RV32I-NEXT:    call __gtsf2
-; RV32I-NEXT:    sgtz a0, a0
-; RV32I-NEXT:    neg s1, a0
-; RV32I-NEXT:    mv a0, s0
 ; RV32I-NEXT:    li a1, 0
 ; RV32I-NEXT:    call __gesf2
 ; RV32I-NEXT:    slti a0, a0, 0
-; RV32I-NEXT:    addi s2, a0, -1
+; RV32I-NEXT:    addi s1, a0, -1
 ; RV32I-NEXT:    mv a0, s0
 ; RV32I-NEXT:    call __fixunssfsi
-; RV32I-NEXT:    and a0, s2, a0
-; RV32I-NEXT:    or a0, s1, a0
+; RV32I-NEXT:    and s1, s1, a0
+; RV32I-NEXT:    lui a1, 325632
+; RV32I-NEXT:    addi a1, a1, -1
+; RV32I-NEXT:    mv a0, s0
+; RV32I-NEXT:    call __gtsf2
+; RV32I-NEXT:    sgtz a0, a0
+; RV32I-NEXT:    neg a0, a0
+; RV32I-NEXT:    or a0, a0, s1
 ; RV32I-NEXT:    lw ra, 12(sp) # 4-byte Folded Reload
 ; RV32I-NEXT:    lw s0, 8(sp) # 4-byte Folded Reload
 ; RV32I-NEXT:    lw s1, 4(sp) # 4-byte Folded Reload
-; RV32I-NEXT:    lw s2, 0(sp) # 4-byte Folded Reload
 ; RV32I-NEXT:    addi sp, sp, 16
 ; RV32I-NEXT:    ret
 ;
@@ -618,38 +616,36 @@ define i64 @fcvt_l_s_sat(float %a) nounwind {
 ; RV32IF-NEXT:    addi sp, sp, -16
 ; RV32IF-NEXT:    sw ra, 12(sp) # 4-byte Folded Spill
 ; RV32IF-NEXT:    sw s0, 8(sp) # 4-byte Folded Spill
-; RV32IF-NEXT:    sw s1, 4(sp) # 4-byte Folded Spill
-; RV32IF-NEXT:    fsw fs0, 0(sp) # 4-byte Folded Spill
+; RV32IF-NEXT:    fsw fs0, 4(sp) # 4-byte Folded Spill
 ; RV32IF-NEXT:    fmv.s fs0, fa0
 ; RV32IF-NEXT:    lui a0, 913408
 ; RV32IF-NEXT:    fmv.w.x fa5, a0
 ; RV32IF-NEXT:    fle.s s0, fa5, fa0
-; RV32IF-NEXT:    neg s1, s0
 ; RV32IF-NEXT:    call __fixsfdi
-; RV32IF-NEXT:    lui a2, %hi(.LCPI12_0)
-; RV32IF-NEXT:    flw fa5, %lo(.LCPI12_0)(a2)
-; RV32IF-NEXT:    and a0, s1, a0
-; RV32IF-NEXT:    flt.s a3, fa5, fs0
-; RV32IF-NEXT:    neg a2, a3
-; RV32IF-NEXT:    or a0, a2, a0
-; RV32IF-NEXT:    feq.s a2, fs0, fs0
-; RV32IF-NEXT:    neg a2, a2
-; RV32IF-NEXT:    lui a5, 524288
 ; RV32IF-NEXT:    lui a4, 524288
+; RV32IF-NEXT:    lui a2, 524288
 ; RV32IF-NEXT:    beqz s0, .LBB12_2
 ; RV32IF-NEXT:  # %bb.1: # %start
-; RV32IF-NEXT:    mv a4, a1
+; RV32IF-NEXT:    mv a2, a1
 ; RV32IF-NEXT:  .LBB12_2: # %start
-; RV32IF-NEXT:    and a0, a2, a0
+; RV32IF-NEXT:    lui a1, %hi(.LCPI12_0)
+; RV32IF-NEXT:    flw fa5, %lo(.LCPI12_0)(a1)
+; RV32IF-NEXT:    flt.s a3, fa5, fs0
 ; RV32IF-NEXT:    beqz a3, .LBB12_4
 ; RV32IF-NEXT:  # %bb.3:
-; RV32IF-NEXT:    addi a4, a5, -1
+; RV32IF-NEXT:    addi a2, a4, -1
 ; RV32IF-NEXT:  .LBB12_4: # %start
-; RV32IF-NEXT:    and a1, a2, a4
+; RV32IF-NEXT:    feq.s a1, fs0, fs0
+; RV32IF-NEXT:    neg a4, a1
+; RV32IF-NEXT:    and a1, a4, a2
+; RV32IF-NEXT:    neg a2, s0
+; RV32IF-NEXT:    and a0, a2, a0
+; RV32IF-NEXT:    neg a2, a3
+; RV32IF-NEXT:    or a0, a2, a0
+; RV32IF-NEXT:    and a0, a4, a0
 ; RV32IF-NEXT:    lw ra, 12(sp) # 4-byte Folded Reload
 ; RV32IF-NEXT:    lw s0, 8(sp) # 4-byte Folded Reload
-; RV32IF-NEXT:    lw s1, 4(sp) # 4-byte Folded Reload
-; RV32IF-NEXT:    flw fs0, 0(sp) # 4-byte Folded Reload
+; RV32IF-NEXT:    flw fs0, 4(sp) # 4-byte Folded Reload
 ; RV32IF-NEXT:    addi sp, sp, 16
 ; RV32IF-NEXT:    ret
 ;
@@ -867,22 +863,23 @@ define i64 @fcvt_lu_s_sat(float %a) nounwind {
 ; RV32IF-NEXT:    addi sp, sp, -16
 ; RV32IF-NEXT:    sw ra, 12(sp) # 4-byte Folded Spill
 ; RV32IF-NEXT:    sw s0, 8(sp) # 4-byte Folded Spill
-; RV32IF-NEXT:    sw s1, 4(sp) # 4-byte Folded Spill
-; RV32IF-NEXT:    lui a0, %hi(.LCPI14_0)
-; RV32IF-NEXT:    flw fa5, %lo(.LCPI14_0)(a0)
-; RV32IF-NEXT:    flt.s a0, fa5, fa0
-; RV32IF-NEXT:    neg s0, a0
+; RV32IF-NEXT:    fsw fs0, 4(sp) # 4-byte Folded Spill
+; RV32IF-NEXT:    fmv.s fs0, fa0
 ; RV32IF-NEXT:    fmv.w.x fa5, zero
 ; RV32IF-NEXT:    fle.s a0, fa5, fa0
-; RV32IF-NEXT:    neg s1, a0
+; RV32IF-NEXT:    neg s0, a0
 ; RV32IF-NEXT:    call __fixunssfdi
-; RV32IF-NEXT:    and a0, s1, a0
-; RV32IF-NEXT:    or a0, s0, a0
-; RV32IF-NEXT:    and a1, s1, a1
-; RV32IF-NEXT:    or a1, s0, a1
+; RV32IF-NEXT:    lui a2, %hi(.LCPI14_0)
+; RV32IF-NEXT:    flw fa5, %lo(.LCPI14_0)(a2)
+; RV32IF-NEXT:    and a0, s0, a0
+; RV32IF-NEXT:    flt.s a2, fa5, fs0
+; RV32IF-NEXT:    neg a2, a2
+; RV32IF-NEXT:    or a0, a2, a0
+; RV32IF-NEXT:    and a1, s0, a1
+; RV32IF-NEXT:    or a1, a2, a1
 ; RV32IF-NEXT:    lw ra, 12(sp) # 4-byte Folded Reload
 ; RV32IF-NEXT:    lw s0, 8(sp) # 4-byte Folded Reload
-; RV32IF-NEXT:    lw s1, 4(sp) # 4-byte Folded Reload
+; RV32IF-NEXT:    flw fs0, 4(sp) # 4-byte Folded Reload
 ; RV32IF-NEXT:    addi sp, sp, 16
 ; RV32IF-NEXT:    ret
 ;
@@ -901,17 +898,19 @@ define i64 @fcvt_lu_s_sat(float %a) nounwind {
 ; RV32IZFINX-NEXT:    sw ra, 12(sp) # 4-byte Folded Spill
 ; RV32IZFINX-NEXT:    sw s0, 8(sp) # 4-byte Folded Spill
 ; RV32IZFINX-NEXT:    sw s1, 4(sp) # 4-byte Folded Spill
-; RV32IZFINX-NEXT:    lui a1, %hi(.LCPI14_0)
-; RV32IZFINX-NEXT:    lw a1, %lo(.LCPI14_0)(a1)
-; RV32IZFINX-NEXT:    flt.s a1, a1, a0
-; RV32IZFINX-NEXT:    neg s0, a1
-; RV32IZFINX-NEXT:    fle.s a1, zero, a0
-; RV32IZFINX-NEXT:    neg s1, a1
+; RV32IZFINX-NEXT:    mv s0, a0
+; RV32IZFINX-NEXT:    fle.s a0, zero, a0
+; RV32IZFINX-NEXT:    neg s1, a0
+; RV32IZFINX-NEXT:    mv a0, s0
 ; RV32IZFINX-NEXT:    call __fixunssfdi
+; RV32IZFINX-NEXT:    lui a2, %hi(.LCPI14_0)
+; RV32IZFINX-NEXT:    lw a2, %lo(.LCPI14_0)(a2)
 ; RV32IZFINX-NEXT:    and a0, s1, a0
-; RV32IZFINX-NEXT:    or a0, s0, a0
+; RV32IZFINX-NEXT:    flt.s a2, a2, s0
+; RV32IZFINX-NEXT:    neg a2, a2
+; RV32IZFINX-NEXT:    or a0, a2, a0
 ; RV32IZFINX-NEXT:    and a1, s1, a1
-; RV32IZFINX-NEXT:    or a1, s0, a1
+; RV32IZFINX-NEXT:    or a1, a2, a1
 ; RV32IZFINX-NEXT:    lw ra, 12(sp) # 4-byte Folded Reload
 ; RV32IZFINX-NEXT:    lw s0, 8(sp) # 4-byte Folded Reload
 ; RV32IZFINX-NEXT:    lw s1, 4(sp) # 4-byte Folded Reload
@@ -929,33 +928,36 @@ define i64 @fcvt_lu_s_sat(float %a) nounwind {
 ;
 ; RV32I-LABEL: fcvt_lu_s_sat:
 ; RV32I:       # %bb.0: # %start
-; RV32I-NEXT:    addi sp, sp, -16
-; RV32I-NEXT:    sw ra, 12(sp) # 4-byte Folded Spill
-; RV32I-NEXT:    sw s0, 8(sp) # 4-byte Folded Spill
-; RV32I-NEXT:    sw s1, 4(sp) # 4-byte Folded Spill
-; RV32I-NEXT:    sw s2, 0(sp) # 4-byte Folded Spill
+; RV32I-NEXT:    addi sp, sp, -32
+; RV32I-NEXT:    sw ra, 28(sp) # 4-byte Folded Spill
+; RV32I-NEXT:    sw s0, 24(sp) # 4-byte Folded Spill
+; RV32I-NEXT:    sw s1, 20(sp) # 4-byte Folded Spill
+; RV32I-NEXT:    sw s2, 16(sp) # 4-byte Folded Spill
+; RV32I-NEXT:    sw s3, 12(sp) # 4-byte Folded Spill
 ; RV32I-NEXT:    mv s0, a0
-; RV32I-NEXT:    lui a1, 391168
-; RV32I-NEXT:    addi a1, a1, -1
-; RV32I-NEXT:    call __gtsf2
-; RV32I-NEXT:    sgtz a0, a0
-; RV32I-NEXT:    neg s1, a0
-; RV32I-NEXT:    mv a0, s0
 ; RV32I-NEXT:    li a1, 0
 ; RV32I-NEXT:    call __gesf2
 ; RV32I-NEXT:    slti a0, a0, 0
 ; RV32I-NEXT:    addi s2, a0, -1
 ; RV32I-NEXT:    mv a0, s0
 ; RV32I-NEXT:    call __fixunssfdi
-; RV32I-NEXT:    and a0, s2, a0
-; RV32I-NEXT:    or a0, s1, a0
-; RV32I-NEXT:    and a1, s2, a1
-; RV32I-NEXT:    or a1, s1, a1
-; RV32I-NEXT:    lw ra, 12(sp) # 4-byte Folded Reload
-; RV32I-NEXT:    lw s0, 8(sp) # 4-byte Folded Reload
-; RV32I-NEXT:    lw s1, 4(sp) # 4-byte Folded Reload
-; RV32I-NEXT:    lw s2, 0(sp) # 4-byte Folded Reload
-; RV32I-NEXT:    addi sp, sp, 16
+; RV32I-NEXT:    mv s1, a1
+; RV32I-NEXT:    and s3, s2, a0
+; RV32I-NEXT:    lui a1, 391168
+; RV32I-NEXT:    addi a1, a1, -1
+; RV32I-NEXT:    mv a0, s0
+; RV32I-NEXT:    call __gtsf2
+; RV32I-NEXT:    sgtz a0, a0
+; RV32I-NEXT:    neg a1, a0
+; RV32I-NEXT:    or a0, a1, s3
+; RV32I-NEXT:    and a2, s2, s1
+; RV32I-NEXT:    or a1, a1, a2
+; RV32I-NEXT:    lw ra, 28(sp) # 4-byte Folded Reload
+; RV32I-NEXT:    lw s0, 24(sp) # 4-byte Folded Reload
+; RV32I-NEXT:    lw s1, 20(sp) # 4-byte Folded Reload
+; RV32I-NEXT:    lw s2, 16(sp) # 4-byte Folded Reload
+; RV32I-NEXT:    lw s3, 12(sp) # 4-byte Folded Reload
+; RV32I-NEXT:    addi sp, sp, 32
 ; RV32I-NEXT:    ret
 ;
 ; RV64I-LABEL: fcvt_lu_s_sat:
@@ -2089,26 +2091,24 @@ define zeroext i32 @fcvt_wu_s_sat_zext(float %a) nounwind {
 ; RV32I-NEXT:    sw ra, 12(sp) # 4-byte Folded Spill
 ; RV32I-NEXT:    sw s0, 8(sp) # 4-byte Folded Spill
 ; RV32I-NEXT:    sw s1, 4(sp) # 4-byte Folded Spill
-; RV32I-NEXT:    sw s2, 0(sp) # 4-byte Folded Spill
 ; RV32I-NEXT:    mv s0, a0
-; RV32I-NEXT:    lui a1, 325632
-; RV32I-NEXT:    addi a1, a1, -1
-; RV32I-NEXT:    call __gtsf2
-; RV32I-NEXT:    sgtz a0, a0
-; RV32I-NEXT:    neg s1, a0
-; RV32I-NEXT:    mv a0, s0
 ; RV32I-NEXT:    li a1, 0
 ; RV32I-NEXT:    call __gesf2
 ; RV32I-NEXT:    slti a0, a0, 0
-; RV32I-NEXT:    addi s2, a0, -1
+; RV32I-NEXT:    addi s1, a0, -1
 ; RV32I-NEXT:    mv a0, s0
 ; RV32I-NEXT:    call __fixunssfsi
-; RV32I-NEXT:    and a0, s2, a0
-; RV32I-NEXT:    or a0, s1, a0
+; RV32I-NEXT:    and s1, s1, a0
+; RV32I-NEXT:    lui a1, 325632
+; RV32I-NEXT:    addi a1, a1, -1
+; RV32I-NEXT:    mv a0, s0
+; RV32I-NEXT:    call __gtsf2
+; RV32I-NEXT:    sgtz a0, a0
+; RV32I-NEXT:    neg a0, a0
+; RV32I-NEXT:    or a0, a0, s1
 ; RV32I-NEXT:    lw ra, 12(sp) # 4-byte Folded Reload
 ; RV32I-NEXT:    lw s0, 8(sp) # 4-byte Folded Reload
 ; RV32I-NEXT:    lw s1, 4(sp) # 4-byte Folded Reload
-; RV32I-NEXT:    lw s2, 0(sp) # 4-byte Folded Reload
 ; RV32I-NEXT:    addi sp, sp, 16
 ; RV32I-NEXT:    ret
 ;
diff --git a/llvm/test/CodeGen/RISCV/float-round-conv-sat.ll b/llvm/test/CodeGen/RISCV/float-round-conv-sat.ll
index 4f747c278da0..5e99c7eb9056 100644
--- a/llvm/test/CodeGen/RISCV/float-round-conv-sat.ll
+++ b/llvm/test/CodeGen/RISCV/float-round-conv-sat.ll
@@ -37,8 +37,7 @@ define i64 @test_floor_si64(float %x) nounwind {
 ; RV32IF-NEXT:    addi sp, sp, -16
 ; RV32IF-NEXT:    sw ra, 12(sp) # 4-byte Folded Spill
 ; RV32IF-NEXT:    sw s0, 8(sp) # 4-byte Folded Spill
-; RV32IF-NEXT:    sw s1, 4(sp) # 4-byte Folded Spill
-; RV32IF-NEXT:    fsw fs0, 0(sp) # 4-byte Folded Spill
+; RV32IF-NEXT:    fsw fs0, 4(sp) # 4-byte Folded Spill
 ; RV32IF-NEXT:    fmv.s fs0, fa0
 ; RV32IF-NEXT:    lui a0, 307200
 ; RV32IF-NEXT:    fmv.w.x fa5, a0
@@ -53,33 +52,32 @@ define i64 @test_floor_si64(float %x) nounwind {
 ; RV32IF-NEXT:    lui a0, 913408
 ; RV32IF-NEXT:    fmv.w.x fa5, a0
 ; RV32IF-NEXT:    fle.s s0, fa5, fs0
-; RV32IF-NEXT:    neg s1, s0
 ; RV32IF-NEXT:    fmv.s fa0, fs0
 ; RV32IF-NEXT:    call __fixsfdi
-; RV32IF-NEXT:    lui a2, %hi(.LCPI1_0)
-; RV32IF-NEXT:    flw fa5, %lo(.LCPI1_0)(a2)
-; RV32IF-NEXT:    and a0, s1, a0
-; RV32IF-NEXT:    flt.s a3, fa5, fs0
-; RV32IF-NEXT:    neg a2, a3
-; RV32IF-NEXT:    or a0, a2, a0
-; RV32IF-NEXT:    feq.s a2, fs0, fs0
-; RV32IF-NEXT:    neg a2, a2
-; RV32IF-NEXT:    lui a5, 524288
 ; RV32IF-NEXT:    lui a4, 524288
+; RV32IF-NEXT:    lui a2, 524288
 ; RV32IF-NEXT:    beqz s0, .LBB1_4
 ; RV32IF-NEXT:  # %bb.3:
-; RV32IF-NEXT:    mv a4, a1
+; RV32IF-NEXT:    mv a2, a1
 ; RV32IF-NEXT:  .LBB1_4:
-; RV32IF-NEXT:    and a0, a2, a0
+; RV32IF-NEXT:    lui a1, %hi(.LCPI1_0)
+; RV32IF-NEXT:    flw fa5, %lo(.LCPI1_0)(a1)
+; RV32IF-NEXT:    flt.s a3, fa5, fs0
 ; RV32IF-NEXT:    beqz a3, .LBB1_6
 ; RV32IF-NEXT:  # %bb.5:
-; RV32IF-NEXT:    addi a4, a5, -1
+; RV32IF-NEXT:    addi a2, a4, -1
 ; RV32IF-NEXT:  .LBB1_6:
-; RV32IF-NEXT:    and a1, a2, a4
+; RV32IF-NEXT:    feq.s a1, fs0, fs0
+; RV32IF-NEXT:    neg a4, a1
+; RV32IF-NEXT:    and a1, a4, a2
+; RV32IF-NEXT:    neg a2, s0
+; RV32IF-NEXT:    and a0, a2, a0
+; RV32IF-NEXT:    neg a2, a3
+; RV32IF-NEXT:    or a0, a2, a0
+; RV32IF-NEXT:    and a0, a4, a0
 ; RV32IF-NEXT:    lw ra, 12(sp) # 4-byte Folded Reload
 ; RV32IF-NEXT:    lw s0, 8(sp) # 4-byte Folded Reload
-; RV32IF-NEXT:    lw s1, 4(sp) # 4-byte Folded Reload
-; RV32IF-NEXT:    flw fs0, 0(sp) # 4-byte Folded Reload
+; RV32IF-NEXT:    flw fs0, 4(sp) # 4-byte Folded Reload
 ; RV32IF-NEXT:    addi sp, sp, 16
 ; RV32IF-NEXT:    ret
 ;
@@ -117,23 +115,23 @@ define i64 @test_floor_si64(float %x) nounwind {
 ; RV32IZFINX-NEXT:    lui a2, %hi(.LCPI1_0)
 ; RV32IZFINX-NEXT:    lw a2, %lo(.LCPI1_0)(a2)
 ; RV32IZFINX-NEXT:    and a0, s2, a0
-; RV32IZFINX-NEXT:    flt.s a3, a2, s0
-; RV32IZFINX-NEXT:    neg a2, a3
+; RV32IZFINX-NEXT:    flt.s a4, a2, s0
+; RV32IZFINX-NEXT:    neg a2, a4
 ; RV32IZFINX-NEXT:    or a0, a2, a0
 ; RV32IZFINX-NEXT:    feq.s a2, s0, s0
 ; RV32IZFINX-NEXT:    neg a2, a2
 ; RV32IZFINX-NEXT:    lui a5, 524288
-; RV32IZFINX-NEXT:    lui a4, 524288
+; RV32IZFINX-NEXT:    lui a3, 524288
 ; RV32IZFINX-NEXT:    beqz s1, .LBB1_4
 ; RV32IZFINX-NEXT:  # %bb.3:
-; RV32IZFINX-NEXT:    mv a4, a1
+; RV32IZFINX-NEXT:    mv a3, a1
 ; RV32IZFINX-NEXT:  .LBB1_4:
 ; RV32IZFINX-NEXT:    and a0, a2, a0
-; RV32IZFINX-NEXT:    beqz a3, .LBB1_6
+; RV32IZFINX-NEXT:    beqz a4, .LBB1_6
 ; RV32IZFINX-NEXT:  # %bb.5:
-; RV32IZFINX-NEXT:    addi a4, a5, -1
+; RV32IZFINX-NEXT:    addi a3, a5, -1
 ; RV32IZFINX-NEXT:  .LBB1_6:
-; RV32IZFINX-NEXT:    and a1, a2, a4
+; RV32IZFINX-NEXT:    and a1, a2, a3
 ; RV32IZFINX-NEXT:    lw ra, 12(sp) # 4-byte Folded Reload
 ; RV32IZFINX-NEXT:    lw s0, 8(sp) # 4-byte Folded Reload
 ; RV32IZFINX-NEXT:    lw s1, 4(sp) # 4-byte Folded Reload
@@ -299,8 +297,7 @@ define i64 @test_ceil_si64(float %x) nounwind {
 ; RV32IF-NEXT:    addi sp, sp, -16
 ; RV32IF-NEXT:    sw ra, 12(sp) # 4-byte Folded Spill
 ; RV32IF-NEXT:    sw s0, 8(sp) # 4-byte Folded Spill
-; RV32IF-NEXT:    sw s1, 4(sp) # 4-byte Folded Spill
-; RV32IF-NEXT:    fsw fs0, 0(sp) # 4-byte Folded Spill
+; RV32IF-NEXT:    fsw fs0, 4(sp) # 4-byte Folded Spill
 ; RV32IF-NEXT:    fmv.s fs0, fa0
 ; RV32IF-NEXT:    lui a0, 307200
 ; RV32IF-NEXT:    fmv.w.x fa5, a0
@@ -315,33 +312,32 @@ define i64 @test_ceil_si64(float %x) nounwind {
 ; RV32IF-NEXT:    lui a0, 913408
 ; RV32IF-NEXT:    fmv.w.x fa5, a0
 ; RV32IF-NEXT:    fle.s s0, fa5, fs0
-; RV32IF-NEXT:    neg s1, s0
 ; RV32IF-NEXT:    fmv.s fa0, fs0
 ; RV32IF-NEXT:    call __fixsfdi
-; RV32IF-NEXT:    lui a2, %hi(.LCPI5_0)
-; RV32IF-NEXT:    flw fa5, %lo(.LCPI5_0)(a2)
-; RV32IF-NEXT:    and a0, s1, a0
-; RV32IF-NEXT:    flt.s a3, fa5, fs0
-; RV32IF-NEXT:    neg a2, a3
-; RV32IF-NEXT:    or a0, a2, a0
-; RV32IF-NEXT:    feq.s a2, fs0, fs0
-; RV32IF-NEXT:    neg a2, a2
-; RV32IF-NEXT:    lui a5, 524288
 ; RV32IF-NEXT:    lui a4, 524288
+; RV32IF-NEXT:    lui a2, 524288
 ; RV32IF-NEXT:    beqz s0, .LBB5_4
 ; RV32IF-NEXT:  # %bb.3:
-; RV32IF-NEXT:    mv a4, a1
+; RV32IF-NEXT:    mv a2, a1
 ; RV32IF-NEXT:  .LBB5_4:
-; RV32IF-NEXT:    and a0, a2, a0
+; RV32IF-NEXT:    lui a1, %hi(.LCPI5_0)
+; RV32IF-NEXT:    flw fa5, %lo(.LCPI5_0)(a1)
+; RV32IF-NEXT:    flt.s a3, fa5, fs0
 ; RV32IF-NEXT:    beqz a3, .LBB5_6
 ; RV32IF-NEXT:  # %bb.5:
-; RV32IF-NEXT:    addi a4, a5, -1
+; RV32IF-NEXT:    addi a2, a4, -1
 ; RV32IF-NEXT:  .LBB5_6:
-; RV32IF-NEXT:    and a1, a2, a4
+; RV32IF-NEXT:    feq.s a1, fs0, fs0
+; RV32IF-NEXT:    neg a4, a1
+; RV32IF-NEXT:    and a1, a4, a2
+; RV32IF-NEXT:    neg a2, s0
+; RV32IF-NEXT:    and a0, a2, a0
+; RV32IF-NEXT:    neg a2, a3
+; RV32IF-NEXT:    or a0, a2, a0
+; RV32IF-NEXT:    and a0, a4, a0
 ; RV32IF-NEXT:    lw ra, 12(sp) # 4-byte Folded Reload
 ; RV32IF-NEXT:    lw s0, 8(sp) # 4-byte Folded Reload
-; RV32IF-NEXT:    lw s1, 4(sp) # 4-byte Folded Reload
-; RV32IF-NEXT:    flw fs0, 0(sp) # 4-byte Folded Reload
+; RV32IF-NEXT:    flw fs0, 4(sp) # 4-byte Folded Reload
 ; RV32IF-NEXT:    addi sp, sp, 16
 ; RV32IF-NEXT:    ret
 ;
@@ -379,23 +375,23 @@ define i64 @test_ceil_si64(float %x) nounwind {
 ; RV32IZFINX-NEXT:    lui a2, %hi(.LCPI5_0)
 ; RV32IZFINX-NEXT:    lw a2, %lo(.LCPI5_0)(a2)
 ; RV32IZFINX-NEXT:    and a0, s2, a0
-; RV32IZFINX-NEXT:    flt.s a3, a2, s0
-; RV32IZFINX-NEXT:    neg a2, a3
+; RV32IZFINX-NEXT:    flt.s a4, a2, s0
+; RV32IZFINX-NEXT:    neg a2, a4
 ; RV32IZFINX-NEXT:    or a0, a2, a0
 ; RV32IZFINX-NEXT:    feq.s a2, s0, s0
 ; RV32IZFINX-NEXT:    neg a2, a2
 ; RV32IZFINX-NEXT:    lui a5, 524288
-; RV32IZFINX-NEXT:    lui a4, 524288
+; RV32IZFINX-NEXT:    lui a3, 524288
 ; RV32IZFINX-NEXT:    beqz s1, .LBB5_4
 ; RV32IZFINX-NEXT:  # %bb.3:
-; RV32IZFINX-NEXT:    mv a4, a1
+; RV32IZFINX-NEXT:    mv a3, a1
 ; RV32IZFINX-NEXT:  .LBB5_4:
 ; RV32IZFINX-NEXT:    and a0, a2, a0
-; RV32IZFINX-NEXT:    beqz a3, .LBB5_6
+; RV32IZFINX-NEXT:    beqz a4, .LBB5_6
 ; RV32IZFINX-NEXT:  # %bb.5:
-; RV32IZFINX-NEXT:    addi a4, a5, -1
+; RV32IZFINX-NEXT:    addi a3, a5, -1
 ; RV32IZFINX-NEXT:  .LBB5_6:
-; RV32IZFINX-NEXT:    and a1, a2, a4
+; RV32IZFINX-NEXT:    and a1, a2, a3
 ; RV32IZFINX-NEXT:    lw ra, 12(sp) # 4-byte Folded Reload
 ; RV32IZFINX-NEXT:    lw s0, 8(sp) # 4-byte Folded Reload
 ; RV32IZFINX-NEXT:    lw s1, 4(sp) # 4-byte Folded Reload
@@ -561,8 +557,7 @@ define i64 @test_trunc_si64(float %x) nounwind {
 ; RV32IF-NEXT:    addi sp, sp, -16
 ; RV32IF-NEXT:    sw ra, 12(sp) # 4-byte Folded Spill
 ; RV32IF-NEXT:    sw s0, 8(sp) # 4-byte Folded Spill
-; RV32IF-NEXT:    sw s1, 4(sp) # 4-byte Folded Spill
-; RV32IF-NEXT:    fsw fs0, 0(sp) # 4-byte Folded Spill
+; RV32IF-NEXT:    fsw fs0, 4(sp) # 4-byte Folded Spill
 ; RV32IF-NEXT:    fmv.s fs0, fa0
 ; RV32IF-NEXT:    lui a0, 307200
 ; RV32IF-NEXT:    fmv.w.x fa5, a0
@@ -577,33 +572,32 @@ define i64 @test_trunc_si64(float %x) nounwind {
 ; RV32IF-NEXT:    lui a0, 913408
 ; RV32IF-NEXT:    fmv.w.x fa5, a0
 ; RV32IF-NEXT:    fle.s s0, fa5, fs0
-; RV32IF-NEXT:    neg s1, s0
 ; RV32IF-NEXT:    fmv.s fa0, fs0
 ; RV32IF-NEXT:    call __fixsfdi
-; RV32IF-NEXT:    lui a2, %hi(.LCPI9_0)
-; RV32IF-NEXT:    flw fa5, %lo(.LCPI9_0)(a2)
-; RV32IF-NEXT:    and a0, s1, a0
-; RV32IF-NEXT:    flt.s a3, fa5, fs0
-; RV32IF-NEXT:    neg a2, a3
-; RV32IF-NEXT:    or a0, a2, a0
-; RV32IF-NEXT:    feq.s a2, fs0, fs0
-; RV32IF-NEXT:    neg a2, a2
-; RV32IF-NEXT:    lui a5, 524288
 ; RV32IF-NEXT:    lui a4, 524288
+; RV32IF-NEXT:    lui a2, 524288
 ; RV32IF-NEXT:    beqz s0, .LBB9_4
 ; RV32IF-NEXT:  # %bb.3:
-; RV32IF-NEXT:    mv a4, a1
+; RV32IF-NEXT:    mv a2, a1
 ; RV32IF-NEXT:  .LBB9_4:
-; RV32IF-NEXT:    and a0, a2, a0
+; RV32IF-NEXT:    lui a1, %hi(.LCPI9_0)
+; RV32IF-NEXT:    flw fa5, %lo(.LCPI9_0)(a1)
+; RV32IF-NEXT:    flt.s a3, fa5, fs0
 ; RV32IF-NEXT:    beqz a3, .LBB9_6
 ; RV32IF-NEXT:  # %bb.5:
-; RV32IF-NEXT:    addi a4, a5, -1
+; RV32IF-NEXT:    addi a2, a4, -1
 ; RV32IF-NEXT:  .LBB9_6:
-; RV32IF-NEXT:    and a1, a2, a4
+; RV32IF-NEXT:    feq.s a1, fs0, fs0
+; RV32IF-NEXT:    neg a4, a1
+; RV32IF-NEXT:    and a1, a4, a2
+; RV32IF-NEXT:    neg a2, s0
+; RV32IF-NEXT:    and a0, a2, a0
+; RV32IF-NEXT:    neg a2, a3
+; RV32IF-NEXT:    or a0, a2, a0
+; RV32IF-NEXT:    and a0, a4, a0
 ; RV32IF-NEXT:    lw ra, 12(sp) # 4-byte Folded Reload
 ; RV32IF-NEXT:    lw s0, 8(sp) # 4-byte Folded Reload
-; RV32IF-NEXT:    lw s1, 4(sp) # 4-byte Folded Reload
-; RV32IF-NEXT:    flw fs0, 0(sp) # 4-byte Folded Reload
+; RV32IF-NEXT:    flw fs0, 4(sp) # 4-byte Folded Reload
 ; RV32IF-NEXT:    addi sp, sp, 16
 ; RV32IF-NEXT:    ret
 ;
@@ -641,23 +635,23 @@ define i64 @test_trunc_si64(float %x) nounwind {
 ; RV32IZFINX-NEXT:    lui a2, %hi(.LCPI9_0)
 ; RV32IZFINX-NEXT:    lw a2, %lo(.LCPI9_0)(a2)
 ; RV32IZFINX-NEXT:    and a0, s2, a0
-; RV32IZFINX-NEXT:    flt.s a3, a2, s0
-; RV32IZFINX-NEXT:    neg a2, a3
+; RV32IZFINX-NEXT:    flt.s a4, a2, s0
+; RV32IZFINX-NEXT:    neg a2, a4
 ; RV32IZFINX-NEXT:    or a0, a2, a0
 ; RV32IZFINX-NEXT:    feq.s a2, s0, s0
 ; RV32IZFINX-NEXT:    neg a2, a2
 ; RV32IZFINX-NEXT:    lui a5, 524288
-; RV32IZFINX-NEXT:    lui a4, 524288
+; RV32IZFINX-NEXT:    lui a3, 524288
 ; RV32IZFINX-NEXT:    beqz s1, .LBB9_4
 ; RV32IZFINX-NEXT:  # %bb.3:
-; RV32IZFINX-NEXT:    mv a4, a1
+; RV32IZFINX-NEXT:    mv a3, a1
 ; RV32IZFINX-NEXT:  .LBB9_4:
 ; RV32IZFINX-NEXT:    and a0, a2, a0
-; RV32IZFINX-NEXT:    beqz a3, .LBB9_6
+; RV32IZFINX-NEXT:    beqz a4, .LBB9_6
 ; RV32IZFINX-NEXT:  # %bb.5:
-; RV32IZFINX-NEXT:    addi a4, a5, -1
+; RV32IZFINX-NEXT:    addi a3, a5, -1
 ; RV32IZFINX-NEXT:  .LBB9_6:
-; RV32IZFINX-NEXT:    and a1, a2, a4
+; RV32IZFINX-NEXT:    and a1, a2, a3
 ; RV32IZFINX-NEXT:    lw ra, 12(sp) # 4-byte Folded Reload
 ; RV32IZFINX-NEXT:    lw s0, 8(sp) # 4-byte Folded Reload
 ; RV32IZFINX-NEXT:    lw s1, 4(sp) # 4-byte Folded Reload
@@ -823,8 +817,7 @@ define i64 @test_round_si64(float %x) nounwind {
 ; RV32IF-NEXT:    addi sp, sp, -16
 ; RV32IF-NEXT:    sw ra, 12(sp) # 4-byte Folded Spill
 ; RV32IF-NEXT:    sw s0, 8(sp) # 4-byte Folded Spill
-; RV32IF-NEXT:    sw s1, 4(sp) # 4-byte Folded Spill
-; RV32IF-NEXT:    fsw fs0, 0(sp) # 4-byte Folded Spill
+; RV32IF-NEXT:    fsw fs0, 4(sp) # 4-byte Folded Spill
 ; RV32IF-NEXT:    fmv.s fs0, fa0
 ; RV32IF-NEXT:    lui a0, 307200
 ; RV32IF-NEXT:    fmv.w.x fa5, a0
@@ -839,33 +832,32 @@ define i64 @test_round_si64(float %x) nounwind {
 ; RV32IF-NEXT:    lui a0, 913408
 ; RV32IF-NEXT:    fmv.w.x fa5, a0
 ; RV32IF-NEXT:    fle.s s0, fa5, fs0
-; RV32IF-NEXT:    neg s1, s0
 ; RV32IF-NEXT:    fmv.s fa0, fs0
 ; RV32IF-NEXT:    call __fixsfdi
-; RV32IF-NEXT:    lui a2, %hi(.LCPI13_0)
-; RV32IF-NEXT:    flw fa5, %lo(.LCPI13_0)(a2)
-; RV32IF-NEXT:    and a0, s1, a0
-; RV32IF-NEXT:    flt.s a3, fa5, fs0
-; RV32IF-NEXT:    neg a2, a3
-; RV32IF-NEXT:    or a0, a2, a0
-; RV32IF-NEXT:    feq.s a2, fs0, fs0
-; RV32IF-NEXT:    neg a2, a2
-; RV32IF-NEXT:    lui a5, 524288
 ; RV32IF-NEXT:    lui a4, 524288
+; RV32IF-NEXT:    lui a2, 524288
 ; RV32IF-NEXT:    beqz s0, .LBB13_4
 ; RV32IF-NEXT:  # %bb.3:
-; RV32IF-NEXT:    mv a4, a1
+; RV32IF-NEXT:    mv a2, a1
 ; RV32IF-NEXT:  .LBB13_4:
-; RV32IF-NEXT:    and a0, a2, a0
+; RV32IF-NEXT:    lui a1, %hi(.LCPI13_0)
+; RV32IF-NEXT:    flw fa5, %lo(.LCPI13_0)(a1)
+; RV32IF-NEXT:    flt.s a3, fa5, fs0
 ; RV32IF-NEXT:    beqz a3, .LBB13_6
 ; RV32IF-NEXT:  # %bb.5:
-; RV32IF-NEXT:    addi a4, a5, -1
+; RV32IF-NEXT:    addi a2, a4, -1
 ; RV32IF-NEXT:  .LBB13_6:
-; RV32IF-NEXT:    and a1, a2, a4
+; RV32IF-NEXT:    feq.s a1, fs0, fs0
+; RV32IF-NEXT:    neg a4, a1
+; RV32IF-NEXT:    and a1, a4, a2
+; RV32IF-NEXT:    neg a2, s0
+; RV32IF-NEXT:    and a0, a2, a0
+; RV32IF-NEXT:    neg a2, a3
+; RV32IF-NEXT:    or a0, a2, a0
+; RV32IF-NEXT:    and a0, a4, a0
 ; RV32IF-NEXT:    lw ra, 12(sp) # 4-byte Folded Reload
 ; RV32IF-NEXT:    lw s0, 8(sp) # 4-byte Folded Reload
-; RV32IF-NEXT:    lw s1, 4(sp) # 4-byte Folded Reload
-; RV32IF-NEXT:    flw fs0, 0(sp) # 4-byte Folded Reload
+; RV32IF-NEXT:    flw fs0, 4(sp) # 4-byte Folded Reload
 ; RV32IF-NEXT:    addi sp, sp, 16
 ; RV32IF-NEXT:    ret
 ;
@@ -903,23 +895,23 @@ define i64 @test_round_si64(float %x) nounwind {
 ; RV32IZFINX-NEXT:    lui a2, %hi(.LCPI13_0)
 ; RV32IZFINX-NEXT:    lw a2, %lo(.LCPI13_0)(a2)
 ; RV32IZFINX-NEXT:    and a0, s2, a0
-; RV32IZFINX-NEXT:    flt.s a3, a2, s0
-; RV32IZFINX-NEXT:    neg a2, a3
+; RV32IZFINX-NEXT:    flt.s a4, a2, s0
+; RV32IZFINX-NEXT:    neg a2, a4
 ; RV32IZFINX-NEXT:    or a0, a2, a0
 ; RV32IZFINX-NEXT:    feq.s a2, s0, s0
 ; RV32IZFINX-NEXT:    neg a2, a2
 ; RV32IZFINX-NEXT:    lui a5, 524288
-; RV32IZFINX-NEXT:    lui a4, 524288
+; RV32IZFINX-NEXT:    lui a3, 524288
 ; RV32IZFINX-NEXT:    beqz s1, .LBB13_4
 ; RV32IZFINX-NEXT:  # %bb.3:
-; RV32IZFINX-NEXT:    mv a4, a1
+; RV32IZFINX-NEXT:    mv a3, a1
 ; RV32IZFINX-NEXT:  .LBB13_4:
 ; RV32IZFINX-NEXT:    and a0, a2, a0
-; RV32IZFINX-NEXT:    beqz a3, .LBB13_6
+; RV32IZFINX-NEXT:    beqz a4, .LBB13_6
 ; RV32IZFINX-NEXT:  # %bb.5:
-; RV32IZFINX-NEXT:    addi a4, a5, -1
+; RV32IZFINX-NEXT:    addi a3, a5, -1
 ; RV32IZFINX-NEXT:  .LBB13_6:
-; RV32IZFINX-NEXT:    and a1, a2, a4
+; RV32IZFINX-NEXT:    and a1, a2, a3
 ; RV32IZFINX-NEXT:    lw ra, 12(sp) # 4-byte Folded Reload
 ; RV32IZFINX-NEXT:    lw s0, 8(sp) # 4-byte Folded Reload
 ; RV32IZFINX-NEXT:    lw s1, 4(sp) # 4-byte Folded Reload
@@ -1085,8 +1077,7 @@ define i64 @test_roundeven_si64(float %x) nounwind {
 ; RV32IF-NEXT:    addi sp, sp, -16
 ; RV32IF-NEXT:    sw ra, 12(sp) # 4-byte Folded Spill
 ; RV32IF-NEXT:    sw s0, 8(sp) # 4-byte Folded Spill
-; RV32IF-NEXT:    sw s1, 4(sp) # 4-byte Folded Spill
-; RV32IF-NEXT:    fsw fs0, 0(sp) # 4-byte Folded Spill
+; RV32IF-NEXT:    fsw fs0, 4(sp) # 4-byte Folded Spill
 ; RV32IF-NEXT:    fmv.s fs0, fa0
 ; RV32IF-NEXT:    lui a0, 307200
 ; RV32IF-NEXT:    fmv.w.x fa5, a0
@@ -1101,33 +1092,32 @@ define i64 @test_roundeven_si64(float %x) nounwind {
 ; RV32IF-NEXT:    lui a0, 913408
 ; RV32IF-NEXT:    fmv.w.x fa5, a0
 ; RV32IF-NEXT:    fle.s s0, fa5, fs0
-; RV32IF-NEXT:    neg s1, s0
 ; RV32IF-NEXT:    fmv.s fa0, fs0
 ; RV32IF-NEXT:    call __fixsfdi
-; RV32IF-NEXT:    lui a2, %hi(.LCPI17_0)
-; RV32IF-NEXT:    flw fa5, %lo(.LCPI17_0)(a2)
-; RV32IF-NEXT:    and a0, s1, a0
-; RV32IF-NEXT:    flt.s a3, fa5, fs0
-; RV32IF-NEXT:    neg a2, a3
-; RV32IF-NEXT:    or a0, a2, a0
-; RV32IF-NEXT:    feq.s a2, fs0, fs0
-; RV32IF-NEXT:    neg a2, a2
-; RV32IF-NEXT:    lui a5, 524288
 ; RV32IF-NEXT:    lui a4, 524288
+; RV32IF-NEXT:    lui a2, 524288
 ; RV32IF-NEXT:    beqz s0, .LBB17_4
 ; RV32IF-NEXT:  # %bb.3:
-; RV32IF-NEXT:    mv a4, a1
+; RV32IF-NEXT:    mv a2, a1
 ; RV32IF-NEXT:  .LBB17_4:
-; RV32IF-NEXT:    and a0, a2, a0
+; RV32IF-NEXT:    lui a1, %hi(.LCPI17_0)
+; RV32IF-NEXT:    flw fa5, %lo(.LCPI17_0)(a1)
+; RV32IF-NEXT:    flt.s a3, fa5, fs0
 ; RV32IF-NEXT:    beqz a3, .LBB17_6
 ; RV32IF-NEXT:  # %bb.5:
-; RV32IF-NEXT:    addi a4, a5, -1
+; RV32IF-NEXT:    addi a2, a4, -1
 ; RV32IF-NEXT:  .LBB17_6:
-; RV32IF-NEXT:    and a1, a2, a4
+; RV32IF-NEXT:    feq.s a1, fs0, fs0
+; RV32IF-NEXT:    neg a4, a1
+; RV32IF-NEXT:    and a1, a4, a2
+; RV32IF-NEXT:    neg a2, s0
+; RV32IF-NEXT:    and a0, a2, a0
+; RV32IF-NEXT:    neg a2, a3
+; RV32IF-NEXT:    or a0, a2, a0
+; RV32IF-NEXT:    and a0, a4, a0
 ; RV32IF-NEXT:    lw ra, 12(sp) # 4-byte Folded Reload
 ; RV32IF-NEXT:    lw s0, 8(sp) # 4-byte Folded Reload
-; RV32IF-NEXT:    lw s1, 4(sp) # 4-byte Folded Reload
-; RV32IF-NEXT:    flw fs0, 0(sp) # 4-byte Folded Reload
+; RV32IF-NEXT:    flw fs0, 4(sp) # 4-byte Folded Reload
 ; RV32IF-NEXT:    addi sp, sp, 16
 ; RV32IF-NEXT:    ret
 ;
@@ -1165,23 +1155,23 @@ define i64 @test_roundeven_si64(float %x) nounwind {
 ; RV32IZFINX-NEXT:    lui a2, %hi(.LCPI17_0)
 ; RV32IZFINX-NEXT:    lw a2, %lo(.LCPI17_0)(a2)
 ; RV32IZFINX-NEXT:    and a0, s2, a0
-; RV32IZFINX-NEXT:    flt.s a3, a2, s0
-; RV32IZFINX-NEXT:    neg a2, a3
+; RV32IZFINX-NEXT:    flt.s a4, a2, s0
+; RV32IZFINX-NEXT:    neg a2, a4
 ; RV32IZFINX-NEXT:    or a0, a2, a0
 ; RV32IZFINX-NEXT:    feq.s a2, s0, s0
 ; RV32IZFINX-NEXT:    neg a2, a2
 ; RV32IZFINX-NEXT:    lui a5, 524288
-; RV32IZFINX-NEXT:    lui a4, 524288
+; RV32IZFINX-NEXT:    lui a3, 524288
 ; RV32IZFINX-NEXT:    beqz s1, .LBB17_4
 ; RV32IZFINX-NEXT:  # %bb.3:
-; RV32IZFINX-NEXT:    mv a4, a1
+; RV32IZFINX-NEXT:    mv a3, a1
 ; RV32IZFINX-NEXT:  .LBB17_4:
 ; RV32IZFINX-NEXT:    and a0, a2, a0
-; RV32IZFINX-NEXT:    beqz a3, .LBB17_6
+; RV32IZFINX-NEXT:    beqz a4, .LBB17_6
 ; RV32IZFINX-NEXT:  # %bb.5:
-; RV32IZFINX-NEXT:    addi a4, a5, -1
+; RV32IZFINX-NEXT:    addi a3, a5, -1
 ; RV32IZFINX-NEXT:  .LBB17_6:
-; RV32IZFINX-NEXT:    and a1, a2, a4
+; RV32IZFINX-NEXT:    and a1, a2, a3
 ; RV32IZFINX-NEXT:    lw ra, 12(sp) # 4-byte Folded Reload
 ; RV32IZFINX-NEXT:    lw s0, 8(sp) # 4-byte Folded Reload
 ; RV32IZFINX-NEXT:    lw s1, 4(sp) # 4-byte Folded Reload
@@ -1347,8 +1337,7 @@ define i64 @test_rint_si64(float %x) nounwind {
 ; RV32IF-NEXT:    addi sp, sp, -16
 ; RV32IF-NEXT:    sw ra, 12(sp) # 4-byte Folded Spill
 ; RV32IF-NEXT:    sw s0, 8(sp) # 4-byte Folded Spill
-; RV32IF-NEXT:    sw s1, 4(sp) # 4-byte Folded Spill
-; RV32IF-NEXT:    fsw fs0, 0(sp) # 4-byte Folded Spill
+; RV32IF-NEXT:    fsw fs0, 4(sp) # 4-byte Folded Spill
 ; RV32IF-NEXT:    fmv.s fs0, fa0
 ; RV32IF-NEXT:    lui a0, 307200
 ; RV32IF-NEXT:    fmv.w.x fa5, a0
@@ -1363,33 +1352,32 @@ define i64 @test_rint_si64(float %x) nounwind {
 ; RV32IF-NEXT:    lui a0, 913408
 ; RV32IF-NEXT:    fmv.w.x fa5, a0
 ; RV32IF-NEXT:    fle.s s0, fa5, fs0
-; RV32IF-NEXT:    neg s1, s0
 ; RV32IF-NEXT:    fmv.s fa0, fs0
 ; RV32IF-NEXT:    call __fixsfdi
-; RV32IF-NEXT:    lui a2, %hi(.LCPI21_0)
-; RV32IF-NEXT:    flw fa5, %lo(.LCPI21_0)(a2)
-; RV32IF-NEXT:    and a0, s1, a0
-; RV32IF-NEXT:    flt.s a3, fa5, fs0
-; RV32IF-NEXT:    neg a2, a3
-; RV32IF-NEXT:    or a0, a2, a0
-; RV32IF-NEXT:    feq.s a2, fs0, fs0
-; RV32IF-NEXT:    neg a2, a2
-; RV32IF-NEXT:    lui a5, 524288
 ; RV32IF-NEXT:    lui a4, 524288
+; RV32IF-NEXT:    lui a2, 524288
 ; RV32IF-NEXT:    beqz s0, .LBB21_4
 ; RV32IF-NEXT:  # %bb.3:
-; RV32IF-NEXT:    mv a4, a1
+; RV32IF-NEXT:    mv a2, a1
 ; RV32IF-NEXT:  .LBB21_4:
-; RV32IF-NEXT:    and a0, a2, a0
+; RV32IF-NEXT:    lui a1, %hi(.LCPI21_0)
+; RV32IF-NEXT:    flw fa5, %lo(.LCPI21_0)(a1)
+; RV32IF-NEXT:    flt.s a3, fa5, fs0
 ; RV32IF-NEXT:    beqz a3, .LBB21_6
 ; RV32IF-NEXT:  # %bb.5:
-; RV32IF-NEXT:    addi a4, a5, -1
+; RV32IF-NEXT:    addi a2, a4, -1
 ; RV32IF-NEXT:  .LBB21_6:
-; RV32IF-NEXT:    and a1, a2, a4
+; RV32IF-NEXT:    feq.s a1, fs0, fs0
+; RV32IF-NEXT:    neg a4, a1
+; RV32IF-NEXT:    and a1, a4, a2
+; RV32IF-NEXT:    neg a2, s0
+; RV32IF-NEXT:    and a0, a2, a0
+; RV32IF-NEXT:    neg a2, a3
+; RV32IF-NEXT:    or a0, a2, a0
+; RV32IF-NEXT:    and a0, a4, a0
 ; RV32IF-NEXT:    lw ra, 12(sp) # 4-byte Folded Reload
 ; RV32IF-NEXT:    lw s0, 8(sp) # 4-byte Folded Reload
-; RV32IF-NEXT:    lw s1, 4(sp) # 4-byte Folded Reload
-; RV32IF-NEXT:    flw fs0, 0(sp) # 4-byte Folded Reload
+; RV32IF-NEXT:    flw fs0, 4(sp) # 4-byte Folded Reload
 ; RV32IF-NEXT:    addi sp, sp, 16
 ; RV32IF-NEXT:    ret
 ;
@@ -1427,23 +1415,23 @@ define i64 @test_rint_si64(float %x) nounwind {
 ; RV32IZFINX-NEXT:    lui a2, %hi(.LCPI21_0)
 ; RV32IZFINX-NEXT:    lw a2, %lo(.LCPI21_0)(a2)
 ; RV32IZFINX-NEXT:    and a0, s2, a0
-; RV32IZFINX-NEXT:    flt.s a3, a2, s0
-; RV32IZFINX-NEXT:    neg a2, a3
+; RV32IZFINX-NEXT:    flt.s a4, a2, s0
+; RV32IZFINX-NEXT:    neg a2, a4
 ; RV32IZFINX-NEXT:    or a0, a2, a0
 ; RV32IZFINX-NEXT:    feq.s a2, s0, s0
 ; RV32IZFINX-NEXT:    neg a2, a2
 ; RV32IZFINX-NEXT:    lui a5, 524288
-; RV32IZFINX-NEXT:    lui a4, 524288
+; RV32IZFINX-NEXT:    lui a3, 524288
 ; RV32IZFINX-NEXT:    beqz s1, .LBB21_4
 ; RV32IZFINX-NEXT:  # %bb.3:
-; RV32IZFINX-NEXT:    mv a4, a1
+; RV32IZFINX-NEXT:    mv a3, a1
 ; RV32IZFINX-NEXT:  .LBB21_4:
 ; RV32IZFINX-NEXT:    and a0, a2, a0
-; RV32IZFINX-NEXT:    beqz a3, .LBB21_6
+; RV32IZFINX-NEXT:    beqz a4, .LBB21_6
 ; RV32IZFINX-NEXT:  # %bb.5:
-; RV32IZFINX-NEXT:    addi a4, a5, -1
+; RV32IZFINX-NEXT:    addi a3, a5, -1
 ; RV32IZFINX-NEXT:  .LBB21_6:
-; RV32IZFINX-NEXT:    and a1, a2, a4
+; RV32IZFINX-NEXT:    and a1, a2, a3
 ; RV32IZFINX-NEXT:    lw ra, 12(sp) # 4-byte Folded Reload
 ; RV32IZFINX-NEXT:    lw s0, 8(sp) # 4-byte Folded Reload
 ; RV32IZFINX-NEXT:    lw s1, 4(sp) # 4-byte Folded Reload
diff --git a/llvm/test/CodeGen/RISCV/forced-atomics.ll b/llvm/test/CodeGen/RISCV/forced-atomics.ll
index c303690aadff..f6a53a9d76dd 100644
--- a/llvm/test/CodeGen/RISCV/forced-atomics.ll
+++ b/llvm/test/CodeGen/RISCV/forced-atomics.ll
@@ -3567,8 +3567,8 @@ define i64 @rmw64_umax_seq_cst(ptr %p) nounwind {
 ; RV32-NEXT:    # in Loop: Header=BB51_2 Depth=1
 ; RV32-NEXT:    neg a3, a0
 ; RV32-NEXT:    and a3, a3, a1
-; RV32-NEXT:    sw a1, 4(sp)
 ; RV32-NEXT:    sw a4, 0(sp)
+; RV32-NEXT:    sw a1, 4(sp)
 ; RV32-NEXT:    mv a1, sp
 ; RV32-NEXT:    li a4, 5
 ; RV32-NEXT:    li a5, 5
@@ -3659,8 +3659,8 @@ define i64 @rmw64_umin_seq_cst(ptr %p) nounwind {
 ; RV32-NEXT:    # in Loop: Header=BB52_2 Depth=1
 ; RV32-NEXT:    neg a3, a0
 ; RV32-NEXT:    and a3, a3, a1
-; RV32-NEXT:    sw a1, 4(sp)
 ; RV32-NEXT:    sw a4, 0(sp)
+; RV32-NEXT:    sw a1, 4(sp)
 ; RV32-NEXT:    mv a1, sp
 ; RV32-NEXT:    li a4, 5
 ; RV32-NEXT:    li a5, 5
diff --git a/llvm/test/CodeGen/RISCV/fpclamptosat.ll b/llvm/test/CodeGen/RISCV/fpclamptosat.ll
index 06ab813faf02..deb5a6d4013d 100644
--- a/llvm/test/CodeGen/RISCV/fpclamptosat.ll
+++ b/llvm/test/CodeGen/RISCV/fpclamptosat.ll
@@ -114,8 +114,8 @@ define i32 @utest_f64i32(double %x) {
 ; RV32IF-NEXT:    sw ra, 12(sp) # 4-byte Folded Spill
 ; RV32IF-NEXT:    .cfi_offset ra, -4
 ; RV32IF-NEXT:    call __fixunsdfdi
-; RV32IF-NEXT:    seqz a1, a1
 ; RV32IF-NEXT:    sltiu a2, a0, -1
+; RV32IF-NEXT:    seqz a1, a1
 ; RV32IF-NEXT:    and a1, a1, a2
 ; RV32IF-NEXT:    addi a1, a1, -1
 ; RV32IF-NEXT:    or a0, a1, a0
@@ -429,8 +429,8 @@ define i32 @utesth_f16i32(half %x) {
 ; RV32-NEXT:    .cfi_offset ra, -4
 ; RV32-NEXT:    call __extendhfsf2
 ; RV32-NEXT:    call __fixunssfdi
-; RV32-NEXT:    seqz a1, a1
 ; RV32-NEXT:    sltiu a2, a0, -1
+; RV32-NEXT:    seqz a1, a1
 ; RV32-NEXT:    and a1, a1, a2
 ; RV32-NEXT:    addi a1, a1, -1
 ; RV32-NEXT:    or a0, a1, a0
diff --git a/llvm/test/CodeGen/RISCV/half-convert.ll b/llvm/test/CodeGen/RISCV/half-convert.ll
index 277749c75bbb..31fb6e2ee9c8 100644
--- a/llvm/test/CodeGen/RISCV/half-convert.ll
+++ b/llvm/test/CodeGen/RISCV/half-convert.ll
@@ -2145,47 +2145,41 @@ define i64 @fcvt_l_h(half %a) nounwind {
 define i64 @fcvt_l_h_sat(half %a) nounwind {
 ; RV32IZFH-LABEL: fcvt_l_h_sat:
 ; RV32IZFH:       # %bb.0: # %start
-; RV32IZFH-NEXT:    addi sp, sp, -32
-; RV32IZFH-NEXT:    sw ra, 28(sp) # 4-byte Folded Spill
-; RV32IZFH-NEXT:    sw s0, 24(sp) # 4-byte Folded Spill
-; RV32IZFH-NEXT:    sw s1, 20(sp) # 4-byte Folded Spill
-; RV32IZFH-NEXT:    sw s2, 16(sp) # 4-byte Folded Spill
-; RV32IZFH-NEXT:    sw s3, 12(sp) # 4-byte Folded Spill
-; RV32IZFH-NEXT:    fsw fs0, 8(sp) # 4-byte Folded Spill
-; RV32IZFH-NEXT:    lui a0, %hi(.LCPI10_0)
-; RV32IZFH-NEXT:    flw fa5, %lo(.LCPI10_0)(a0)
+; RV32IZFH-NEXT:    addi sp, sp, -16
+; RV32IZFH-NEXT:    sw ra, 12(sp) # 4-byte Folded Spill
+; RV32IZFH-NEXT:    sw s0, 8(sp) # 4-byte Folded Spill
+; RV32IZFH-NEXT:    fsw fs0, 4(sp) # 4-byte Folded Spill
 ; RV32IZFH-NEXT:    fcvt.s.h fs0, fa0
-; RV32IZFH-NEXT:    flt.s s0, fa5, fs0
-; RV32IZFH-NEXT:    neg s1, s0
 ; RV32IZFH-NEXT:    lui a0, 913408
 ; RV32IZFH-NEXT:    fmv.w.x fa5, a0
-; RV32IZFH-NEXT:    fle.s s2, fa5, fs0
-; RV32IZFH-NEXT:    neg s3, s2
+; RV32IZFH-NEXT:    fle.s s0, fa5, fs0
 ; RV32IZFH-NEXT:    fmv.s fa0, fs0
 ; RV32IZFH-NEXT:    call __fixsfdi
-; RV32IZFH-NEXT:    and a0, s3, a0
-; RV32IZFH-NEXT:    or a0, s1, a0
-; RV32IZFH-NEXT:    feq.s a2, fs0, fs0
-; RV32IZFH-NEXT:    neg a2, a2
 ; RV32IZFH-NEXT:    lui a4, 524288
-; RV32IZFH-NEXT:    lui a3, 524288
-; RV32IZFH-NEXT:    beqz s2, .LBB10_2
+; RV32IZFH-NEXT:    lui a2, 524288
+; RV32IZFH-NEXT:    beqz s0, .LBB10_2
 ; RV32IZFH-NEXT:  # %bb.1: # %start
-; RV32IZFH-NEXT:    mv a3, a1
+; RV32IZFH-NEXT:    mv a2, a1
 ; RV32IZFH-NEXT:  .LBB10_2: # %start
-; RV32IZFH-NEXT:    and a0, a2, a0
-; RV32IZFH-NEXT:    beqz s0, .LBB10_4
+; RV32IZFH-NEXT:    lui a1, %hi(.LCPI10_0)
+; RV32IZFH-NEXT:    flw fa5, %lo(.LCPI10_0)(a1)
+; RV32IZFH-NEXT:    flt.s a3, fa5, fs0
+; RV32IZFH-NEXT:    beqz a3, .LBB10_4
 ; RV32IZFH-NEXT:  # %bb.3:
-; RV32IZFH-NEXT:    addi a3, a4, -1
+; RV32IZFH-NEXT:    addi a2, a4, -1
 ; RV32IZFH-NEXT:  .LBB10_4: # %start
-; RV32IZFH-NEXT:    and a1, a2, a3
-; RV32IZFH-NEXT:    lw ra, 28(sp) # 4-byte Folded Reload
-; RV32IZFH-NEXT:    lw s0, 24(sp) # 4-byte Folded Reload
-; RV32IZFH-NEXT:    lw s1, 20(sp) # 4-byte Folded Reload
-; RV32IZFH-NEXT:    lw s2, 16(sp) # 4-byte Folded Reload
-; RV32IZFH-NEXT:    lw s3, 12(sp) # 4-byte Folded Reload
-; RV32IZFH-NEXT:    flw fs0, 8(sp) # 4-byte Folded Reload
-; RV32IZFH-NEXT:    addi sp, sp, 32
+; RV32IZFH-NEXT:    feq.s a1, fs0, fs0
+; RV32IZFH-NEXT:    neg a4, a1
+; RV32IZFH-NEXT:    and a1, a4, a2
+; RV32IZFH-NEXT:    neg a2, a3
+; RV32IZFH-NEXT:    neg a3, s0
+; RV32IZFH-NEXT:    and a0, a3, a0
+; RV32IZFH-NEXT:    or a0, a2, a0
+; RV32IZFH-NEXT:    and a0, a4, a0
+; RV32IZFH-NEXT:    lw ra, 12(sp) # 4-byte Folded Reload
+; RV32IZFH-NEXT:    lw s0, 8(sp) # 4-byte Folded Reload
+; RV32IZFH-NEXT:    flw fs0, 4(sp) # 4-byte Folded Reload
+; RV32IZFH-NEXT:    addi sp, sp, 16
 ; RV32IZFH-NEXT:    ret
 ;
 ; RV64IZFH-LABEL: fcvt_l_h_sat:
@@ -2199,47 +2193,41 @@ define i64 @fcvt_l_h_sat(half %a) nounwind {
 ;
 ; RV32IDZFH-LABEL: fcvt_l_h_sat:
 ; RV32IDZFH:       # %bb.0: # %start
-; RV32IDZFH-NEXT:    addi sp, sp, -32
-; RV32IDZFH-NEXT:    sw ra, 28(sp) # 4-byte Folded Spill
-; RV32IDZFH-NEXT:    sw s0, 24(sp) # 4-byte Folded Spill
-; RV32IDZFH-NEXT:    sw s1, 20(sp) # 4-byte Folded Spill
-; RV32IDZFH-NEXT:    sw s2, 16(sp) # 4-byte Folded Spill
-; RV32IDZFH-NEXT:    sw s3, 12(sp) # 4-byte Folded Spill
+; RV32IDZFH-NEXT:    addi sp, sp, -16
+; RV32IDZFH-NEXT:    sw ra, 12(sp) # 4-byte Folded Spill
+; RV32IDZFH-NEXT:    sw s0, 8(sp) # 4-byte Folded Spill
 ; RV32IDZFH-NEXT:    fsd fs0, 0(sp) # 8-byte Folded Spill
-; RV32IDZFH-NEXT:    lui a0, %hi(.LCPI10_0)
-; RV32IDZFH-NEXT:    flw fa5, %lo(.LCPI10_0)(a0)
 ; RV32IDZFH-NEXT:    fcvt.s.h fs0, fa0
-; RV32IDZFH-NEXT:    flt.s s0, fa5, fs0
-; RV32IDZFH-NEXT:    neg s1, s0
 ; RV32IDZFH-NEXT:    lui a0, 913408
 ; RV32IDZFH-NEXT:    fmv.w.x fa5, a0
-; RV32IDZFH-NEXT:    fle.s s2, fa5, fs0
-; RV32IDZFH-NEXT:    neg s3, s2
+; RV32IDZFH-NEXT:    fle.s s0, fa5, fs0
 ; RV32IDZFH-NEXT:    fmv.s fa0, fs0
 ; RV32IDZFH-NEXT:    call __fixsfdi
-; RV32IDZFH-NEXT:    and a0, s3, a0
-; RV32IDZFH-NEXT:    or a0, s1, a0
-; RV32IDZFH-NEXT:    feq.s a2, fs0, fs0
-; RV32IDZFH-NEXT:    neg a2, a2
 ; RV32IDZFH-NEXT:    lui a4, 524288
-; RV32IDZFH-NEXT:    lui a3, 524288
-; RV32IDZFH-NEXT:    beqz s2, .LBB10_2
+; RV32IDZFH-NEXT:    lui a2, 524288
+; RV32IDZFH-NEXT:    beqz s0, .LBB10_2
 ; RV32IDZFH-NEXT:  # %bb.1: # %start
-; RV32IDZFH-NEXT:    mv a3, a1
+; RV32IDZFH-NEXT:    mv a2, a1
 ; RV32IDZFH-NEXT:  .LBB10_2: # %start
-; RV32IDZFH-NEXT:    and a0, a2, a0
-; RV32IDZFH-NEXT:    beqz s0, .LBB10_4
+; RV32IDZFH-NEXT:    lui a1, %hi(.LCPI10_0)
+; RV32IDZFH-NEXT:    flw fa5, %lo(.LCPI10_0)(a1)
+; RV32IDZFH-NEXT:    flt.s a3, fa5, fs0
+; RV32IDZFH-NEXT:    beqz a3, .LBB10_4
 ; RV32IDZFH-NEXT:  # %bb.3:
-; RV32IDZFH-NEXT:    addi a3, a4, -1
+; RV32IDZFH-NEXT:    addi a2, a4, -1
 ; RV32IDZFH-NEXT:  .LBB10_4: # %start
-; RV32IDZFH-NEXT:    and a1, a2, a3
-; RV32IDZFH-NEXT:    lw ra, 28(sp) # 4-byte Folded Reload
-; RV32IDZFH-NEXT:    lw s0, 24(sp) # 4-byte Folded Reload
-; RV32IDZFH-NEXT:    lw s1, 20(sp) # 4-byte Folded Reload
-; RV32IDZFH-NEXT:    lw s2, 16(sp) # 4-byte Folded Reload
-; RV32IDZFH-NEXT:    lw s3, 12(sp) # 4-byte Folded Reload
+; RV32IDZFH-NEXT:    feq.s a1, fs0, fs0
+; RV32IDZFH-NEXT:    neg a4, a1
+; RV32IDZFH-NEXT:    and a1, a4, a2
+; RV32IDZFH-NEXT:    neg a2, a3
+; RV32IDZFH-NEXT:    neg a3, s0
+; RV32IDZFH-NEXT:    and a0, a3, a0
+; RV32IDZFH-NEXT:    or a0, a2, a0
+; RV32IDZFH-NEXT:    and a0, a4, a0
+; RV32IDZFH-NEXT:    lw ra, 12(sp) # 4-byte Folded Reload
+; RV32IDZFH-NEXT:    lw s0, 8(sp) # 4-byte Folded Reload
 ; RV32IDZFH-NEXT:    fld fs0, 0(sp) # 8-byte Folded Reload
-; RV32IDZFH-NEXT:    addi sp, sp, 32
+; RV32IDZFH-NEXT:    addi sp, sp, 16
 ; RV32IDZFH-NEXT:    ret
 ;
 ; RV64IDZFH-LABEL: fcvt_l_h_sat:
@@ -2515,47 +2503,41 @@ define i64 @fcvt_l_h_sat(half %a) nounwind {
 ;
 ; RV32ID-LABEL: fcvt_l_h_sat:
 ; RV32ID:       # %bb.0: # %start
-; RV32ID-NEXT:    addi sp, sp, -32
-; RV32ID-NEXT:    sw ra, 28(sp) # 4-byte Folded Spill
-; RV32ID-NEXT:    sw s0, 24(sp) # 4-byte Folded Spill
-; RV32ID-NEXT:    sw s1, 20(sp) # 4-byte Folded Spill
-; RV32ID-NEXT:    sw s2, 16(sp) # 4-byte Folded Spill
-; RV32ID-NEXT:    sw s3, 12(sp) # 4-byte Folded Spill
+; RV32ID-NEXT:    addi sp, sp, -16
+; RV32ID-NEXT:    sw ra, 12(sp) # 4-byte Folded Spill
+; RV32ID-NEXT:    sw s0, 8(sp) # 4-byte Folded Spill
 ; RV32ID-NEXT:    fsd fs0, 0(sp) # 8-byte Folded Spill
 ; RV32ID-NEXT:    call __extendhfsf2
-; RV32ID-NEXT:    lui a0, %hi(.LCPI10_0)
-; RV32ID-NEXT:    flw fa5, %lo(.LCPI10_0)(a0)
 ; RV32ID-NEXT:    fmv.s fs0, fa0
-; RV32ID-NEXT:    flt.s s0, fa5, fa0
-; RV32ID-NEXT:    neg s1, s0
 ; RV32ID-NEXT:    lui a0, 913408
 ; RV32ID-NEXT:    fmv.w.x fa5, a0
-; RV32ID-NEXT:    fle.s s2, fa5, fa0
-; RV32ID-NEXT:    neg s3, s2
+; RV32ID-NEXT:    fle.s s0, fa5, fa0
 ; RV32ID-NEXT:    call __fixsfdi
-; RV32ID-NEXT:    and a0, s3, a0
-; RV32ID-NEXT:    or a0, s1, a0
-; RV32ID-NEXT:    feq.s a2, fs0, fs0
-; RV32ID-NEXT:    neg a2, a2
 ; RV32ID-NEXT:    lui a4, 524288
-; RV32ID-NEXT:    lui a3, 524288
-; RV32ID-NEXT:    beqz s2, .LBB10_2
+; RV32ID-NEXT:    lui a2, 524288
+; RV32ID-NEXT:    beqz s0, .LBB10_2
 ; RV32ID-NEXT:  # %bb.1: # %start
-; RV32ID-NEXT:    mv a3, a1
+; RV32ID-NEXT:    mv a2, a1
 ; RV32ID-NEXT:  .LBB10_2: # %start
-; RV32ID-NEXT:    and a0, a2, a0
-; RV32ID-NEXT:    beqz s0, .LBB10_4
+; RV32ID-NEXT:    lui a1, %hi(.LCPI10_0)
+; RV32ID-NEXT:    flw fa5, %lo(.LCPI10_0)(a1)
+; RV32ID-NEXT:    flt.s a3, fa5, fs0
+; RV32ID-NEXT:    beqz a3, .LBB10_4
 ; RV32ID-NEXT:  # %bb.3:
-; RV32ID-NEXT:    addi a3, a4, -1
+; RV32ID-NEXT:    addi a2, a4, -1
 ; RV32ID-NEXT:  .LBB10_4: # %start
-; RV32ID-NEXT:    and a1, a2, a3
-; RV32ID-NEXT:    lw ra, 28(sp) # 4-byte Folded Reload
-; RV32ID-NEXT:    lw s0, 24(sp) # 4-byte Folded Reload
-; RV32ID-NEXT:    lw s1, 20(sp) # 4-byte Folded Reload
-; RV32ID-NEXT:    lw s2, 16(sp) # 4-byte Folded Reload
-; RV32ID-NEXT:    lw s3, 12(sp) # 4-byte Folded Reload
+; RV32ID-NEXT:    feq.s a1, fs0, fs0
+; RV32ID-NEXT:    neg a4, a1
+; RV32ID-NEXT:    and a1, a4, a2
+; RV32ID-NEXT:    neg a2, s0
+; RV32ID-NEXT:    and a0, a2, a0
+; RV32ID-NEXT:    neg a2, a3
+; RV32ID-NEXT:    or a0, a2, a0
+; RV32ID-NEXT:    and a0, a4, a0
+; RV32ID-NEXT:    lw ra, 12(sp) # 4-byte Folded Reload
+; RV32ID-NEXT:    lw s0, 8(sp) # 4-byte Folded Reload
 ; RV32ID-NEXT:    fld fs0, 0(sp) # 8-byte Folded Reload
-; RV32ID-NEXT:    addi sp, sp, 32
+; RV32ID-NEXT:    addi sp, sp, 16
 ; RV32ID-NEXT:    ret
 ;
 ; RV64ID-LABEL: fcvt_l_h_sat:
@@ -2574,47 +2556,41 @@ define i64 @fcvt_l_h_sat(half %a) nounwind {
 ;
 ; RV32IFZFHMIN-LABEL: fcvt_l_h_sat:
 ; RV32IFZFHMIN:       # %bb.0: # %start
-; RV32IFZFHMIN-NEXT:    addi sp, sp, -32
-; RV32IFZFHMIN-NEXT:    sw ra, 28(sp) # 4-byte Folded Spill
-; RV32IFZFHMIN-NEXT:    sw s0, 24(sp) # 4-byte Folded Spill
-; RV32IFZFHMIN-NEXT:    sw s1, 20(sp) # 4-byte Folded Spill
-; RV32IFZFHMIN-NEXT:    sw s2, 16(sp) # 4-byte Folded Spill
-; RV32IFZFHMIN-NEXT:    sw s3, 12(sp) # 4-byte Folded Spill
-; RV32IFZFHMIN-NEXT:    fsw fs0, 8(sp) # 4-byte Folded Spill
-; RV32IFZFHMIN-NEXT:    lui a0, %hi(.LCPI10_0)
-; RV32IFZFHMIN-NEXT:    flw fa5, %lo(.LCPI10_0)(a0)
+; RV32IFZFHMIN-NEXT:    addi sp, sp, -16
+; RV32IFZFHMIN-NEXT:    sw ra, 12(sp) # 4-byte Folded Spill
+; RV32IFZFHMIN-NEXT:    sw s0, 8(sp) # 4-byte Folded Spill
+; RV32IFZFHMIN-NEXT:    fsw fs0, 4(sp) # 4-byte Folded Spill
 ; RV32IFZFHMIN-NEXT:    fcvt.s.h fs0, fa0
-; RV32IFZFHMIN-NEXT:    flt.s s0, fa5, fs0
-; RV32IFZFHMIN-NEXT:    neg s1, s0
 ; RV32IFZFHMIN-NEXT:    lui a0, 913408
 ; RV32IFZFHMIN-NEXT:    fmv.w.x fa5, a0
-; RV32IFZFHMIN-NEXT:    fle.s s2, fa5, fs0
-; RV32IFZFHMIN-NEXT:    neg s3, s2
+; RV32IFZFHMIN-NEXT:    fle.s s0, fa5, fs0
 ; RV32IFZFHMIN-NEXT:    fmv.s fa0, fs0
 ; RV32IFZFHMIN-NEXT:    call __fixsfdi
-; RV32IFZFHMIN-NEXT:    and a0, s3, a0
-; RV32IFZFHMIN-NEXT:    or a0, s1, a0
-; RV32IFZFHMIN-NEXT:    feq.s a2, fs0, fs0
-; RV32IFZFHMIN-NEXT:    neg a2, a2
 ; RV32IFZFHMIN-NEXT:    lui a4, 524288
-; RV32IFZFHMIN-NEXT:    lui a3, 524288
-; RV32IFZFHMIN-NEXT:    beqz s2, .LBB10_2
+; RV32IFZFHMIN-NEXT:    lui a2, 524288
+; RV32IFZFHMIN-NEXT:    beqz s0, .LBB10_2
 ; RV32IFZFHMIN-NEXT:  # %bb.1: # %start
-; RV32IFZFHMIN-NEXT:    mv a3, a1
+; RV32IFZFHMIN-NEXT:    mv a2, a1
 ; RV32IFZFHMIN-NEXT:  .LBB10_2: # %start
-; RV32IFZFHMIN-NEXT:    and a0, a2, a0
-; RV32IFZFHMIN-NEXT:    beqz s0, .LBB10_4
+; RV32IFZFHMIN-NEXT:    lui a1, %hi(.LCPI10_0)
+; RV32IFZFHMIN-NEXT:    flw fa5, %lo(.LCPI10_0)(a1)
+; RV32IFZFHMIN-NEXT:    flt.s a3, fa5, fs0
+; RV32IFZFHMIN-NEXT:    beqz a3, .LBB10_4
 ; RV32IFZFHMIN-NEXT:  # %bb.3:
-; RV32IFZFHMIN-NEXT:    addi a3, a4, -1
+; RV32IFZFHMIN-NEXT:    addi a2, a4, -1
 ; RV32IFZFHMIN-NEXT:  .LBB10_4: # %start
-; RV32IFZFHMIN-NEXT:    and a1, a2, a3
-; RV32IFZFHMIN-NEXT:    lw ra, 28(sp) # 4-byte Folded Reload
-; RV32IFZFHMIN-NEXT:    lw s0, 24(sp) # 4-byte Folded Reload
-; RV32IFZFHMIN-NEXT:    lw s1, 20(sp) # 4-byte Folded Reload
-; RV32IFZFHMIN-NEXT:    lw s2, 16(sp) # 4-byte Folded Reload
-; RV32IFZFHMIN-NEXT:    lw s3, 12(sp) # 4-byte Folded Reload
-; RV32IFZFHMIN-NEXT:    flw fs0, 8(sp) # 4-byte Folded Reload
-; RV32IFZFHMIN-NEXT:    addi sp, sp, 32
+; RV32IFZFHMIN-NEXT:    feq.s a1, fs0, fs0
+; RV32IFZFHMIN-NEXT:    neg a4, a1
+; RV32IFZFHMIN-NEXT:    and a1, a4, a2
+; RV32IFZFHMIN-NEXT:    neg a2, a3
+; RV32IFZFHMIN-NEXT:    neg a3, s0
+; RV32IFZFHMIN-NEXT:    and a0, a3, a0
+; RV32IFZFHMIN-NEXT:    or a0, a2, a0
+; RV32IFZFHMIN-NEXT:    and a0, a4, a0
+; RV32IFZFHMIN-NEXT:    lw ra, 12(sp) # 4-byte Folded Reload
+; RV32IFZFHMIN-NEXT:    lw s0, 8(sp) # 4-byte Folded Reload
+; RV32IFZFHMIN-NEXT:    flw fs0, 4(sp) # 4-byte Folded Reload
+; RV32IFZFHMIN-NEXT:    addi sp, sp, 16
 ; RV32IFZFHMIN-NEXT:    ret
 ;
 ; CHECK64-IZFHMIN-LABEL: fcvt_l_h_sat:
@@ -2629,47 +2605,41 @@ define i64 @fcvt_l_h_sat(half %a) nounwind {
 ;
 ; RV32IDZFHMIN-LABEL: fcvt_l_h_sat:
 ; RV32IDZFHMIN:       # %bb.0: # %start
-; RV32IDZFHMIN-NEXT:    addi sp, sp, -32
-; RV32IDZFHMIN-NEXT:    sw ra, 28(sp) # 4-byte Folded Spill
-; RV32IDZFHMIN-NEXT:    sw s0, 24(sp) # 4-byte Folded Spill
-; RV32IDZFHMIN-NEXT:    sw s1, 20(sp) # 4-byte Folded Spill
-; RV32IDZFHMIN-NEXT:    sw s2, 16(sp) # 4-byte Folded Spill
-; RV32IDZFHMIN-NEXT:    sw s3, 12(sp) # 4-byte Folded Spill
+; RV32IDZFHMIN-NEXT:    addi sp, sp, -16
+; RV32IDZFHMIN-NEXT:    sw ra, 12(sp) # 4-byte Folded Spill
+; RV32IDZFHMIN-NEXT:    sw s0, 8(sp) # 4-byte Folded Spill
 ; RV32IDZFHMIN-NEXT:    fsd fs0, 0(sp) # 8-byte Folded Spill
-; RV32IDZFHMIN-NEXT:    lui a0, %hi(.LCPI10_0)
-; RV32IDZFHMIN-NEXT:    flw fa5, %lo(.LCPI10_0)(a0)
 ; RV32IDZFHMIN-NEXT:    fcvt.s.h fs0, fa0
-; RV32IDZFHMIN-NEXT:    flt.s s0, fa5, fs0
-; RV32IDZFHMIN-NEXT:    neg s1, s0
 ; RV32IDZFHMIN-NEXT:    lui a0, 913408
 ; RV32IDZFHMIN-NEXT:    fmv.w.x fa5, a0
-; RV32IDZFHMIN-NEXT:    fle.s s2, fa5, fs0
-; RV32IDZFHMIN-NEXT:    neg s3, s2
+; RV32IDZFHMIN-NEXT:    fle.s s0, fa5, fs0
 ; RV32IDZFHMIN-NEXT:    fmv.s fa0, fs0
 ; RV32IDZFHMIN-NEXT:    call __fixsfdi
-; RV32IDZFHMIN-NEXT:    and a0, s3, a0
-; RV32IDZFHMIN-NEXT:    or a0, s1, a0
-; RV32IDZFHMIN-NEXT:    feq.s a2, fs0, fs0
-; RV32IDZFHMIN-NEXT:    neg a2, a2
 ; RV32IDZFHMIN-NEXT:    lui a4, 524288
-; RV32IDZFHMIN-NEXT:    lui a3, 524288
-; RV32IDZFHMIN-NEXT:    beqz s2, .LBB10_2
+; RV32IDZFHMIN-NEXT:    lui a2, 524288
+; RV32IDZFHMIN-NEXT:    beqz s0, .LBB10_2
 ; RV32IDZFHMIN-NEXT:  # %bb.1: # %start
-; RV32IDZFHMIN-NEXT:    mv a3, a1
+; RV32IDZFHMIN-NEXT:    mv a2, a1
 ; RV32IDZFHMIN-NEXT:  .LBB10_2: # %start
-; RV32IDZFHMIN-NEXT:    and a0, a2, a0
-; RV32IDZFHMIN-NEXT:    beqz s0, .LBB10_4
+; RV32IDZFHMIN-NEXT:    lui a1, %hi(.LCPI10_0)
+; RV32IDZFHMIN-NEXT:    flw fa5, %lo(.LCPI10_0)(a1)
+; RV32IDZFHMIN-NEXT:    flt.s a3, fa5, fs0
+; RV32IDZFHMIN-NEXT:    beqz a3, .LBB10_4
 ; RV32IDZFHMIN-NEXT:  # %bb.3:
-; RV32IDZFHMIN-NEXT:    addi a3, a4, -1
+; RV32IDZFHMIN-NEXT:    addi a2, a4, -1
 ; RV32IDZFHMIN-NEXT:  .LBB10_4: # %start
-; RV32IDZFHMIN-NEXT:    and a1, a2, a3
-; RV32IDZFHMIN-NEXT:    lw ra, 28(sp) # 4-byte Folded Reload
-; RV32IDZFHMIN-NEXT:    lw s0, 24(sp) # 4-byte Folded Reload
-; RV32IDZFHMIN-NEXT:    lw s1, 20(sp) # 4-byte Folded Reload
-; RV32IDZFHMIN-NEXT:    lw s2, 16(sp) # 4-byte Folded Reload
-; RV32IDZFHMIN-NEXT:    lw s3, 12(sp) # 4-byte Folded Reload
+; RV32IDZFHMIN-NEXT:    feq.s a1, fs0, fs0
+; RV32IDZFHMIN-NEXT:    neg a4, a1
+; RV32IDZFHMIN-NEXT:    and a1, a4, a2
+; RV32IDZFHMIN-NEXT:    neg a2, a3
+; RV32IDZFHMIN-NEXT:    neg a3, s0
+; RV32IDZFHMIN-NEXT:    and a0, a3, a0
+; RV32IDZFHMIN-NEXT:    or a0, a2, a0
+; RV32IDZFHMIN-NEXT:    and a0, a4, a0
+; RV32IDZFHMIN-NEXT:    lw ra, 12(sp) # 4-byte Folded Reload
+; RV32IDZFHMIN-NEXT:    lw s0, 8(sp) # 4-byte Folded Reload
 ; RV32IDZFHMIN-NEXT:    fld fs0, 0(sp) # 8-byte Folded Reload
-; RV32IDZFHMIN-NEXT:    addi sp, sp, 32
+; RV32IDZFHMIN-NEXT:    addi sp, sp, 16
 ; RV32IDZFHMIN-NEXT:    ret
 ;
 ; CHECK32-IZHINXMIN-LABEL: fcvt_l_h_sat:
diff --git a/llvm/test/CodeGen/RISCV/half-round-conv-sat.ll b/llvm/test/CodeGen/RISCV/half-round-conv-sat.ll
index 9c95210bfa7c..04a8a66f4459 100644
--- a/llvm/test/CodeGen/RISCV/half-round-conv-sat.ll
+++ b/llvm/test/CodeGen/RISCV/half-round-conv-sat.ll
@@ -108,40 +108,38 @@ define i64 @test_floor_si64(half %x) nounwind {
 ; RV32IZFH-NEXT:    addi sp, sp, -16
 ; RV32IZFH-NEXT:    sw ra, 12(sp) # 4-byte Folded Spill
 ; RV32IZFH-NEXT:    sw s0, 8(sp) # 4-byte Folded Spill
-; RV32IZFH-NEXT:    sw s1, 4(sp) # 4-byte Folded Spill
-; RV32IZFH-NEXT:    fsw fs0, 0(sp) # 4-byte Folded Spill
+; RV32IZFH-NEXT:    fsw fs0, 4(sp) # 4-byte Folded Spill
 ; RV32IZFH-NEXT:    fcvt.s.h fs0, fa0
 ; RV32IZFH-NEXT:    lui a0, 913408
 ; RV32IZFH-NEXT:    fmv.w.x fa5, a0
 ; RV32IZFH-NEXT:    fle.s s0, fa5, fs0
-; RV32IZFH-NEXT:    neg s1, s0
 ; RV32IZFH-NEXT:    fmv.s fa0, fs0
 ; RV32IZFH-NEXT:    call __fixsfdi
-; RV32IZFH-NEXT:    lui a2, %hi(.LCPI1_1)
-; RV32IZFH-NEXT:    flw fa5, %lo(.LCPI1_1)(a2)
-; RV32IZFH-NEXT:    and a0, s1, a0
-; RV32IZFH-NEXT:    flt.s a3, fa5, fs0
-; RV32IZFH-NEXT:    neg a2, a3
-; RV32IZFH-NEXT:    or a0, a2, a0
-; RV32IZFH-NEXT:    feq.s a2, fs0, fs0
-; RV32IZFH-NEXT:    neg a2, a2
-; RV32IZFH-NEXT:    lui a5, 524288
 ; RV32IZFH-NEXT:    lui a4, 524288
+; RV32IZFH-NEXT:    lui a2, 524288
 ; RV32IZFH-NEXT:    beqz s0, .LBB1_4
 ; RV32IZFH-NEXT:  # %bb.3:
-; RV32IZFH-NEXT:    mv a4, a1
+; RV32IZFH-NEXT:    mv a2, a1
 ; RV32IZFH-NEXT:  .LBB1_4:
+; RV32IZFH-NEXT:    lui a1, %hi(.LCPI1_1)
+; RV32IZFH-NEXT:    flw fa5, %lo(.LCPI1_1)(a1)
+; RV32IZFH-NEXT:    flt.s a3, fa5, fs0
+; RV32IZFH-NEXT:    beqz a3, .LBB1_6
+; RV32IZFH-NEXT:  # %bb.5:
+; RV32IZFH-NEXT:    addi a2, a4, -1
+; RV32IZFH-NEXT:  .LBB1_6:
+; RV32IZFH-NEXT:    feq.s a1, fs0, fs0
+; RV32IZFH-NEXT:    neg a4, a1
+; RV32IZFH-NEXT:    and a1, a4, a2
+; RV32IZFH-NEXT:    neg a2, s0
 ; RV32IZFH-NEXT:    and a0, a2, a0
+; RV32IZFH-NEXT:    neg a2, a3
+; RV32IZFH-NEXT:    or a0, a2, a0
+; RV32IZFH-NEXT:    and a0, a4, a0
 ; RV32IZFH-NEXT:    lw ra, 12(sp) # 4-byte Folded Reload
 ; RV32IZFH-NEXT:    lw s0, 8(sp) # 4-byte Folded Reload
-; RV32IZFH-NEXT:    lw s1, 4(sp) # 4-byte Folded Reload
-; RV32IZFH-NEXT:    flw fs0, 0(sp) # 4-byte Folded Reload
+; RV32IZFH-NEXT:    flw fs0, 4(sp) # 4-byte Folded Reload
 ; RV32IZFH-NEXT:    addi sp, sp, 16
-; RV32IZFH-NEXT:    beqz a3, .LBB1_6
-; RV32IZFH-NEXT:  # %bb.5:
-; RV32IZFH-NEXT:    addi a4, a5, -1
-; RV32IZFH-NEXT:  .LBB1_6:
-; RV32IZFH-NEXT:    and a1, a2, a4
 ; RV32IZFH-NEXT:    ret
 ;
 ; RV64IZFH-LABEL: test_floor_si64:
@@ -179,16 +177,16 @@ define i64 @test_floor_si64(half %x) nounwind {
 ; RV32IZHINX-NEXT:    lui a2, %hi(.LCPI1_1)
 ; RV32IZHINX-NEXT:    lw a2, %lo(.LCPI1_1)(a2)
 ; RV32IZHINX-NEXT:    and a0, s2, a0
-; RV32IZHINX-NEXT:    flt.s a3, a2, s0
-; RV32IZHINX-NEXT:    neg a2, a3
+; RV32IZHINX-NEXT:    flt.s a4, a2, s0
+; RV32IZHINX-NEXT:    neg a2, a4
 ; RV32IZHINX-NEXT:    or a0, a2, a0
 ; RV32IZHINX-NEXT:    feq.s a2, s0, s0
 ; RV32IZHINX-NEXT:    neg a2, a2
 ; RV32IZHINX-NEXT:    lui a5, 524288
-; RV32IZHINX-NEXT:    lui a4, 524288
+; RV32IZHINX-NEXT:    lui a3, 524288
 ; RV32IZHINX-NEXT:    beqz s1, .LBB1_4
 ; RV32IZHINX-NEXT:  # %bb.3:
-; RV32IZHINX-NEXT:    mv a4, a1
+; RV32IZHINX-NEXT:    mv a3, a1
 ; RV32IZHINX-NEXT:  .LBB1_4:
 ; RV32IZHINX-NEXT:    and a0, a2, a0
 ; RV32IZHINX-NEXT:    lw ra, 12(sp) # 4-byte Folded Reload
@@ -196,11 +194,11 @@ define i64 @test_floor_si64(half %x) nounwind {
 ; RV32IZHINX-NEXT:    lw s1, 4(sp) # 4-byte Folded Reload
 ; RV32IZHINX-NEXT:    lw s2, 0(sp) # 4-byte Folded Reload
 ; RV32IZHINX-NEXT:    addi sp, sp, 16
-; RV32IZHINX-NEXT:    beqz a3, .LBB1_6
+; RV32IZHINX-NEXT:    beqz a4, .LBB1_6
 ; RV32IZHINX-NEXT:  # %bb.5:
-; RV32IZHINX-NEXT:    addi a4, a5, -1
+; RV32IZHINX-NEXT:    addi a3, a5, -1
 ; RV32IZHINX-NEXT:  .LBB1_6:
-; RV32IZHINX-NEXT:    and a1, a2, a4
+; RV32IZHINX-NEXT:    and a1, a2, a3
 ; RV32IZHINX-NEXT:    ret
 ;
 ; RV64IZHINX-LABEL: test_floor_si64:
@@ -238,41 +236,39 @@ define i64 @test_floor_si64(half %x) nounwind {
 ; RV32IZFHMIN-NEXT:    addi sp, sp, -16
 ; RV32IZFHMIN-NEXT:    sw ra, 12(sp) # 4-byte Folded Spill
 ; RV32IZFHMIN-NEXT:    sw s0, 8(sp) # 4-byte Folded Spill
-; RV32IZFHMIN-NEXT:    sw s1, 4(sp) # 4-byte Folded Spill
-; RV32IZFHMIN-NEXT:    fsw fs0, 0(sp) # 4-byte Folded Spill
+; RV32IZFHMIN-NEXT:    fsw fs0, 4(sp) # 4-byte Folded Spill
 ; RV32IZFHMIN-NEXT:    fcvt.h.s fa5, fa5
 ; RV32IZFHMIN-NEXT:    fcvt.s.h fs0, fa5
 ; RV32IZFHMIN-NEXT:    lui a0, 913408
 ; RV32IZFHMIN-NEXT:    fmv.w.x fa5, a0
 ; RV32IZFHMIN-NEXT:    fle.s s0, fa5, fs0
-; RV32IZFHMIN-NEXT:    neg s1, s0
 ; RV32IZFHMIN-NEXT:    fmv.s fa0, fs0
 ; RV32IZFHMIN-NEXT:    call __fixsfdi
-; RV32IZFHMIN-NEXT:    lui a2, %hi(.LCPI1_0)
-; RV32IZFHMIN-NEXT:    flw fa5, %lo(.LCPI1_0)(a2)
-; RV32IZFHMIN-NEXT:    and a0, s1, a0
-; RV32IZFHMIN-NEXT:    flt.s a3, fa5, fs0
-; RV32IZFHMIN-NEXT:    neg a2, a3
-; RV32IZFHMIN-NEXT:    or a0, a2, a0
-; RV32IZFHMIN-NEXT:    feq.s a2, fs0, fs0
-; RV32IZFHMIN-NEXT:    neg a2, a2
-; RV32IZFHMIN-NEXT:    lui a5, 524288
 ; RV32IZFHMIN-NEXT:    lui a4, 524288
+; RV32IZFHMIN-NEXT:    lui a2, 524288
 ; RV32IZFHMIN-NEXT:    beqz s0, .LBB1_4
 ; RV32IZFHMIN-NEXT:  # %bb.3:
-; RV32IZFHMIN-NEXT:    mv a4, a1
+; RV32IZFHMIN-NEXT:    mv a2, a1
 ; RV32IZFHMIN-NEXT:  .LBB1_4:
+; RV32IZFHMIN-NEXT:    lui a1, %hi(.LCPI1_0)
+; RV32IZFHMIN-NEXT:    flw fa5, %lo(.LCPI1_0)(a1)
+; RV32IZFHMIN-NEXT:    flt.s a3, fa5, fs0
+; RV32IZFHMIN-NEXT:    beqz a3, .LBB1_6
+; RV32IZFHMIN-NEXT:  # %bb.5:
+; RV32IZFHMIN-NEXT:    addi a2, a4, -1
+; RV32IZFHMIN-NEXT:  .LBB1_6:
+; RV32IZFHMIN-NEXT:    feq.s a1, fs0, fs0
+; RV32IZFHMIN-NEXT:    neg a4, a1
+; RV32IZFHMIN-NEXT:    and a1, a4, a2
+; RV32IZFHMIN-NEXT:    neg a2, s0
 ; RV32IZFHMIN-NEXT:    and a0, a2, a0
+; RV32IZFHMIN-NEXT:    neg a2, a3
+; RV32IZFHMIN-NEXT:    or a0, a2, a0
+; RV32IZFHMIN-NEXT:    and a0, a4, a0
 ; RV32IZFHMIN-NEXT:    lw ra, 12(sp) # 4-byte Folded Reload
 ; RV32IZFHMIN-NEXT:    lw s0, 8(sp) # 4-byte Folded Reload
-; RV32IZFHMIN-NEXT:    lw s1, 4(sp) # 4-byte Folded Reload
-; RV32IZFHMIN-NEXT:    flw fs0, 0(sp) # 4-byte Folded Reload
+; RV32IZFHMIN-NEXT:    flw fs0, 4(sp) # 4-byte Folded Reload
 ; RV32IZFHMIN-NEXT:    addi sp, sp, 16
-; RV32IZFHMIN-NEXT:    beqz a3, .LBB1_6
-; RV32IZFHMIN-NEXT:  # %bb.5:
-; RV32IZFHMIN-NEXT:    addi a4, a5, -1
-; RV32IZFHMIN-NEXT:  .LBB1_6:
-; RV32IZFHMIN-NEXT:    and a1, a2, a4
 ; RV32IZFHMIN-NEXT:    ret
 ;
 ; RV64IZFHMIN-LABEL: test_floor_si64:
@@ -324,16 +320,16 @@ define i64 @test_floor_si64(half %x) nounwind {
 ; RV32IZHINXMIN-NEXT:    lui a2, %hi(.LCPI1_0)
 ; RV32IZHINXMIN-NEXT:    lw a2, %lo(.LCPI1_0)(a2)
 ; RV32IZHINXMIN-NEXT:    and a0, s2, a0
-; RV32IZHINXMIN-NEXT:    flt.s a3, a2, s0
-; RV32IZHINXMIN-NEXT:    neg a2, a3
+; RV32IZHINXMIN-NEXT:    flt.s a4, a2, s0
+; RV32IZHINXMIN-NEXT:    neg a2, a4
 ; RV32IZHINXMIN-NEXT:    or a0, a2, a0
 ; RV32IZHINXMIN-NEXT:    feq.s a2, s0, s0
 ; RV32IZHINXMIN-NEXT:    neg a2, a2
 ; RV32IZHINXMIN-NEXT:    lui a5, 524288
-; RV32IZHINXMIN-NEXT:    lui a4, 524288
+; RV32IZHINXMIN-NEXT:    lui a3, 524288
 ; RV32IZHINXMIN-NEXT:    beqz s1, .LBB1_4
 ; RV32IZHINXMIN-NEXT:  # %bb.3:
-; RV32IZHINXMIN-NEXT:    mv a4, a1
+; RV32IZHINXMIN-NEXT:    mv a3, a1
 ; RV32IZHINXMIN-NEXT:  .LBB1_4:
 ; RV32IZHINXMIN-NEXT:    and a0, a2, a0
 ; RV32IZHINXMIN-NEXT:    lw ra, 12(sp) # 4-byte Folded Reload
@@ -341,11 +337,11 @@ define i64 @test_floor_si64(half %x) nounwind {
 ; RV32IZHINXMIN-NEXT:    lw s1, 4(sp) # 4-byte Folded Reload
 ; RV32IZHINXMIN-NEXT:    lw s2, 0(sp) # 4-byte Folded Reload
 ; RV32IZHINXMIN-NEXT:    addi sp, sp, 16
-; RV32IZHINXMIN-NEXT:    beqz a3, .LBB1_6
+; RV32IZHINXMIN-NEXT:    beqz a4, .LBB1_6
 ; RV32IZHINXMIN-NEXT:  # %bb.5:
-; RV32IZHINXMIN-NEXT:    addi a4, a5, -1
+; RV32IZHINXMIN-NEXT:    addi a3, a5, -1
 ; RV32IZHINXMIN-NEXT:  .LBB1_6:
-; RV32IZHINXMIN-NEXT:    and a1, a2, a4
+; RV32IZHINXMIN-NEXT:    and a1, a2, a3
 ; RV32IZHINXMIN-NEXT:    ret
 ;
 ; RV64IZHINXMIN-LABEL: test_floor_si64:
@@ -824,40 +820,38 @@ define i64 @test_ceil_si64(half %x) nounwind {
 ; RV32IZFH-NEXT:    addi sp, sp, -16
 ; RV32IZFH-NEXT:    sw ra, 12(sp) # 4-byte Folded Spill
 ; RV32IZFH-NEXT:    sw s0, 8(sp) # 4-byte Folded Spill
-; RV32IZFH-NEXT:    sw s1, 4(sp) # 4-byte Folded Spill
-; RV32IZFH-NEXT:    fsw fs0, 0(sp) # 4-byte Folded Spill
+; RV32IZFH-NEXT:    fsw fs0, 4(sp) # 4-byte Folded Spill
 ; RV32IZFH-NEXT:    fcvt.s.h fs0, fa0
 ; RV32IZFH-NEXT:    lui a0, 913408
 ; RV32IZFH-NEXT:    fmv.w.x fa5, a0
 ; RV32IZFH-NEXT:    fle.s s0, fa5, fs0
-; RV32IZFH-NEXT:    neg s1, s0
 ; RV32IZFH-NEXT:    fmv.s fa0, fs0
 ; RV32IZFH-NEXT:    call __fixsfdi
-; RV32IZFH-NEXT:    lui a2, %hi(.LCPI5_1)
-; RV32IZFH-NEXT:    flw fa5, %lo(.LCPI5_1)(a2)
-; RV32IZFH-NEXT:    and a0, s1, a0
-; RV32IZFH-NEXT:    flt.s a3, fa5, fs0
-; RV32IZFH-NEXT:    neg a2, a3
-; RV32IZFH-NEXT:    or a0, a2, a0
-; RV32IZFH-NEXT:    feq.s a2, fs0, fs0
-; RV32IZFH-NEXT:    neg a2, a2
-; RV32IZFH-NEXT:    lui a5, 524288
 ; RV32IZFH-NEXT:    lui a4, 524288
+; RV32IZFH-NEXT:    lui a2, 524288
 ; RV32IZFH-NEXT:    beqz s0, .LBB5_4
 ; RV32IZFH-NEXT:  # %bb.3:
-; RV32IZFH-NEXT:    mv a4, a1
+; RV32IZFH-NEXT:    mv a2, a1
 ; RV32IZFH-NEXT:  .LBB5_4:
+; RV32IZFH-NEXT:    lui a1, %hi(.LCPI5_1)
+; RV32IZFH-NEXT:    flw fa5, %lo(.LCPI5_1)(a1)
+; RV32IZFH-NEXT:    flt.s a3, fa5, fs0
+; RV32IZFH-NEXT:    beqz a3, .LBB5_6
+; RV32IZFH-NEXT:  # %bb.5:
+; RV32IZFH-NEXT:    addi a2, a4, -1
+; RV32IZFH-NEXT:  .LBB5_6:
+; RV32IZFH-NEXT:    feq.s a1, fs0, fs0
+; RV32IZFH-NEXT:    neg a4, a1
+; RV32IZFH-NEXT:    and a1, a4, a2
+; RV32IZFH-NEXT:    neg a2, s0
 ; RV32IZFH-NEXT:    and a0, a2, a0
+; RV32IZFH-NEXT:    neg a2, a3
+; RV32IZFH-NEXT:    or a0, a2, a0
+; RV32IZFH-NEXT:    and a0, a4, a0
 ; RV32IZFH-NEXT:    lw ra, 12(sp) # 4-byte Folded Reload
 ; RV32IZFH-NEXT:    lw s0, 8(sp) # 4-byte Folded Reload
-; RV32IZFH-NEXT:    lw s1, 4(sp) # 4-byte Folded Reload
-; RV32IZFH-NEXT:    flw fs0, 0(sp) # 4-byte Folded Reload
+; RV32IZFH-NEXT:    flw fs0, 4(sp) # 4-byte Folded Reload
 ; RV32IZFH-NEXT:    addi sp, sp, 16
-; RV32IZFH-NEXT:    beqz a3, .LBB5_6
-; RV32IZFH-NEXT:  # %bb.5:
-; RV32IZFH-NEXT:    addi a4, a5, -1
-; RV32IZFH-NEXT:  .LBB5_6:
-; RV32IZFH-NEXT:    and a1, a2, a4
 ; RV32IZFH-NEXT:    ret
 ;
 ; RV64IZFH-LABEL: test_ceil_si64:
@@ -895,16 +889,16 @@ define i64 @test_ceil_si64(half %x) nounwind {
 ; RV32IZHINX-NEXT:    lui a2, %hi(.LCPI5_1)
 ; RV32IZHINX-NEXT:    lw a2, %lo(.LCPI5_1)(a2)
 ; RV32IZHINX-NEXT:    and a0, s2, a0
-; RV32IZHINX-NEXT:    flt.s a3, a2, s0
-; RV32IZHINX-NEXT:    neg a2, a3
+; RV32IZHINX-NEXT:    flt.s a4, a2, s0
+; RV32IZHINX-NEXT:    neg a2, a4
 ; RV32IZHINX-NEXT:    or a0, a2, a0
 ; RV32IZHINX-NEXT:    feq.s a2, s0, s0
 ; RV32IZHINX-NEXT:    neg a2, a2
 ; RV32IZHINX-NEXT:    lui a5, 524288
-; RV32IZHINX-NEXT:    lui a4, 524288
+; RV32IZHINX-NEXT:    lui a3, 524288
 ; RV32IZHINX-NEXT:    beqz s1, .LBB5_4
 ; RV32IZHINX-NEXT:  # %bb.3:
-; RV32IZHINX-NEXT:    mv a4, a1
+; RV32IZHINX-NEXT:    mv a3, a1
 ; RV32IZHINX-NEXT:  .LBB5_4:
 ; RV32IZHINX-NEXT:    and a0, a2, a0
 ; RV32IZHINX-NEXT:    lw ra, 12(sp) # 4-byte Folded Reload
@@ -912,11 +906,11 @@ define i64 @test_ceil_si64(half %x) nounwind {
 ; RV32IZHINX-NEXT:    lw s1, 4(sp) # 4-byte Folded Reload
 ; RV32IZHINX-NEXT:    lw s2, 0(sp) # 4-byte Folded Reload
 ; RV32IZHINX-NEXT:    addi sp, sp, 16
-; RV32IZHINX-NEXT:    beqz a3, .LBB5_6
+; RV32IZHINX-NEXT:    beqz a4, .LBB5_6
 ; RV32IZHINX-NEXT:  # %bb.5:
-; RV32IZHINX-NEXT:    addi a4, a5, -1
+; RV32IZHINX-NEXT:    addi a3, a5, -1
 ; RV32IZHINX-NEXT:  .LBB5_6:
-; RV32IZHINX-NEXT:    and a1, a2, a4
+; RV32IZHINX-NEXT:    and a1, a2, a3
 ; RV32IZHINX-NEXT:    ret
 ;
 ; RV64IZHINX-LABEL: test_ceil_si64:
@@ -954,41 +948,39 @@ define i64 @test_ceil_si64(half %x) nounwind {
 ; RV32IZFHMIN-NEXT:    addi sp, sp, -16
 ; RV32IZFHMIN-NEXT:    sw ra, 12(sp) # 4-byte Folded Spill
 ; RV32IZFHMIN-NEXT:    sw s0, 8(sp) # 4-byte Folded Spill
-; RV32IZFHMIN-NEXT:    sw s1, 4(sp) # 4-byte Folded Spill
-; RV32IZFHMIN-NEXT:    fsw fs0, 0(sp) # 4-byte Folded Spill
+; RV32IZFHMIN-NEXT:    fsw fs0, 4(sp) # 4-byte Folded Spill
 ; RV32IZFHMIN-NEXT:    fcvt.h.s fa5, fa5
 ; RV32IZFHMIN-NEXT:    fcvt.s.h fs0, fa5
 ; RV32IZFHMIN-NEXT:    lui a0, 913408
 ; RV32IZFHMIN-NEXT:    fmv.w.x fa5, a0
 ; RV32IZFHMIN-NEXT:    fle.s s0, fa5, fs0
-; RV32IZFHMIN-NEXT:    neg s1, s0
 ; RV32IZFHMIN-NEXT:    fmv.s fa0, fs0
 ; RV32IZFHMIN-NEXT:    call __fixsfdi
-; RV32IZFHMIN-NEXT:    lui a2, %hi(.LCPI5_0)
-; RV32IZFHMIN-NEXT:    flw fa5, %lo(.LCPI5_0)(a2)
-; RV32IZFHMIN-NEXT:    and a0, s1, a0
-; RV32IZFHMIN-NEXT:    flt.s a3, fa5, fs0
-; RV32IZFHMIN-NEXT:    neg a2, a3
-; RV32IZFHMIN-NEXT:    or a0, a2, a0
-; RV32IZFHMIN-NEXT:    feq.s a2, fs0, fs0
-; RV32IZFHMIN-NEXT:    neg a2, a2
-; RV32IZFHMIN-NEXT:    lui a5, 524288
 ; RV32IZFHMIN-NEXT:    lui a4, 524288
+; RV32IZFHMIN-NEXT:    lui a2, 524288
 ; RV32IZFHMIN-NEXT:    beqz s0, .LBB5_4
 ; RV32IZFHMIN-NEXT:  # %bb.3:
-; RV32IZFHMIN-NEXT:    mv a4, a1
+; RV32IZFHMIN-NEXT:    mv a2, a1
 ; RV32IZFHMIN-NEXT:  .LBB5_4:
+; RV32IZFHMIN-NEXT:    lui a1, %hi(.LCPI5_0)
+; RV32IZFHMIN-NEXT:    flw fa5, %lo(.LCPI5_0)(a1)
+; RV32IZFHMIN-NEXT:    flt.s a3, fa5, fs0
+; RV32IZFHMIN-NEXT:    beqz a3, .LBB5_6
+; RV32IZFHMIN-NEXT:  # %bb.5:
+; RV32IZFHMIN-NEXT:    addi a2, a4, -1
+; RV32IZFHMIN-NEXT:  .LBB5_6:
+; RV32IZFHMIN-NEXT:    feq.s a1, fs0, fs0
+; RV32IZFHMIN-NEXT:    neg a4, a1
+; RV32IZFHMIN-NEXT:    and a1, a4, a2
+; RV32IZFHMIN-NEXT:    neg a2, s0
 ; RV32IZFHMIN-NEXT:    and a0, a2, a0
+; RV32IZFHMIN-NEXT:    neg a2, a3
+; RV32IZFHMIN-NEXT:    or a0, a2, a0
+; RV32IZFHMIN-NEXT:    and a0, a4, a0
 ; RV32IZFHMIN-NEXT:    lw ra, 12(sp) # 4-byte Folded Reload
 ; RV32IZFHMIN-NEXT:    lw s0, 8(sp) # 4-byte Folded Reload
-; RV32IZFHMIN-NEXT:    lw s1, 4(sp) # 4-byte Folded Reload
-; RV32IZFHMIN-NEXT:    flw fs0, 0(sp) # 4-byte Folded Reload
+; RV32IZFHMIN-NEXT:    flw fs0, 4(sp) # 4-byte Folded Reload
 ; RV32IZFHMIN-NEXT:    addi sp, sp, 16
-; RV32IZFHMIN-NEXT:    beqz a3, .LBB5_6
-; RV32IZFHMIN-NEXT:  # %bb.5:
-; RV32IZFHMIN-NEXT:    addi a4, a5, -1
-; RV32IZFHMIN-NEXT:  .LBB5_6:
-; RV32IZFHMIN-NEXT:    and a1, a2, a4
 ; RV32IZFHMIN-NEXT:    ret
 ;
 ; RV64IZFHMIN-LABEL: test_ceil_si64:
@@ -1040,16 +1032,16 @@ define i64 @test_ceil_si64(half %x) nounwind {
 ; RV32IZHINXMIN-NEXT:    lui a2, %hi(.LCPI5_0)
 ; RV32IZHINXMIN-NEXT:    lw a2, %lo(.LCPI5_0)(a2)
 ; RV32IZHINXMIN-NEXT:    and a0, s2, a0
-; RV32IZHINXMIN-NEXT:    flt.s a3, a2, s0
-; RV32IZHINXMIN-NEXT:    neg a2, a3
+; RV32IZHINXMIN-NEXT:    flt.s a4, a2, s0
+; RV32IZHINXMIN-NEXT:    neg a2, a4
 ; RV32IZHINXMIN-NEXT:    or a0, a2, a0
 ; RV32IZHINXMIN-NEXT:    feq.s a2, s0, s0
 ; RV32IZHINXMIN-NEXT:    neg a2, a2
 ; RV32IZHINXMIN-NEXT:    lui a5, 524288
-; RV32IZHINXMIN-NEXT:    lui a4, 524288
+; RV32IZHINXMIN-NEXT:    lui a3, 524288
 ; RV32IZHINXMIN-NEXT:    beqz s1, .LBB5_4
 ; RV32IZHINXMIN-NEXT:  # %bb.3:
-; RV32IZHINXMIN-NEXT:    mv a4, a1
+; RV32IZHINXMIN-NEXT:    mv a3, a1
 ; RV32IZHINXMIN-NEXT:  .LBB5_4:
 ; RV32IZHINXMIN-NEXT:    and a0, a2, a0
 ; RV32IZHINXMIN-NEXT:    lw ra, 12(sp) # 4-byte Folded Reload
@@ -1057,11 +1049,11 @@ define i64 @test_ceil_si64(half %x) nounwind {
 ; RV32IZHINXMIN-NEXT:    lw s1, 4(sp) # 4-byte Folded Reload
 ; RV32IZHINXMIN-NEXT:    lw s2, 0(sp) # 4-byte Folded Reload
 ; RV32IZHINXMIN-NEXT:    addi sp, sp, 16
-; RV32IZHINXMIN-NEXT:    beqz a3, .LBB5_6
+; RV32IZHINXMIN-NEXT:    beqz a4, .LBB5_6
 ; RV32IZHINXMIN-NEXT:  # %bb.5:
-; RV32IZHINXMIN-NEXT:    addi a4, a5, -1
+; RV32IZHINXMIN-NEXT:    addi a3, a5, -1
 ; RV32IZHINXMIN-NEXT:  .LBB5_6:
-; RV32IZHINXMIN-NEXT:    and a1, a2, a4
+; RV32IZHINXMIN-NEXT:    and a1, a2, a3
 ; RV32IZHINXMIN-NEXT:    ret
 ;
 ; RV64IZHINXMIN-LABEL: test_ceil_si64:
@@ -1540,40 +1532,38 @@ define i64 @test_trunc_si64(half %x) nounwind {
 ; RV32IZFH-NEXT:    addi sp, sp, -16
 ; RV32IZFH-NEXT:    sw ra, 12(sp) # 4-byte Folded Spill
 ; RV32IZFH-NEXT:    sw s0, 8(sp) # 4-byte Folded Spill
-; RV32IZFH-NEXT:    sw s1, 4(sp) # 4-byte Folded Spill
-; RV32IZFH-NEXT:    fsw fs0, 0(sp) # 4-byte Folded Spill
+; RV32IZFH-NEXT:    fsw fs0, 4(sp) # 4-byte Folded Spill
 ; RV32IZFH-NEXT:    fcvt.s.h fs0, fa0
 ; RV32IZFH-NEXT:    lui a0, 913408
 ; RV32IZFH-NEXT:    fmv.w.x fa5, a0
 ; RV32IZFH-NEXT:    fle.s s0, fa5, fs0
-; RV32IZFH-NEXT:    neg s1, s0
 ; RV32IZFH-NEXT:    fmv.s fa0, fs0
 ; RV32IZFH-NEXT:    call __fixsfdi
-; RV32IZFH-NEXT:    lui a2, %hi(.LCPI9_1)
-; RV32IZFH-NEXT:    flw fa5, %lo(.LCPI9_1)(a2)
-; RV32IZFH-NEXT:    and a0, s1, a0
-; RV32IZFH-NEXT:    flt.s a3, fa5, fs0
-; RV32IZFH-NEXT:    neg a2, a3
-; RV32IZFH-NEXT:    or a0, a2, a0
-; RV32IZFH-NEXT:    feq.s a2, fs0, fs0
-; RV32IZFH-NEXT:    neg a2, a2
-; RV32IZFH-NEXT:    lui a5, 524288
 ; RV32IZFH-NEXT:    lui a4, 524288
+; RV32IZFH-NEXT:    lui a2, 524288
 ; RV32IZFH-NEXT:    beqz s0, .LBB9_4
 ; RV32IZFH-NEXT:  # %bb.3:
-; RV32IZFH-NEXT:    mv a4, a1
+; RV32IZFH-NEXT:    mv a2, a1
 ; RV32IZFH-NEXT:  .LBB9_4:
+; RV32IZFH-NEXT:    lui a1, %hi(.LCPI9_1)
+; RV32IZFH-NEXT:    flw fa5, %lo(.LCPI9_1)(a1)
+; RV32IZFH-NEXT:    flt.s a3, fa5, fs0
+; RV32IZFH-NEXT:    beqz a3, .LBB9_6
+; RV32IZFH-NEXT:  # %bb.5:
+; RV32IZFH-NEXT:    addi a2, a4, -1
+; RV32IZFH-NEXT:  .LBB9_6:
+; RV32IZFH-NEXT:    feq.s a1, fs0, fs0
+; RV32IZFH-NEXT:    neg a4, a1
+; RV32IZFH-NEXT:    and a1, a4, a2
+; RV32IZFH-NEXT:    neg a2, s0
 ; RV32IZFH-NEXT:    and a0, a2, a0
+; RV32IZFH-NEXT:    neg a2, a3
+; RV32IZFH-NEXT:    or a0, a2, a0
+; RV32IZFH-NEXT:    and a0, a4, a0
 ; RV32IZFH-NEXT:    lw ra, 12(sp) # 4-byte Folded Reload
 ; RV32IZFH-NEXT:    lw s0, 8(sp) # 4-byte Folded Reload
-; RV32IZFH-NEXT:    lw s1, 4(sp) # 4-byte Folded Reload
-; RV32IZFH-NEXT:    flw fs0, 0(sp) # 4-byte Folded Reload
+; RV32IZFH-NEXT:    flw fs0, 4(sp) # 4-byte Folded Reload
 ; RV32IZFH-NEXT:    addi sp, sp, 16
-; RV32IZFH-NEXT:    beqz a3, .LBB9_6
-; RV32IZFH-NEXT:  # %bb.5:
-; RV32IZFH-NEXT:    addi a4, a5, -1
-; RV32IZFH-NEXT:  .LBB9_6:
-; RV32IZFH-NEXT:    and a1, a2, a4
 ; RV32IZFH-NEXT:    ret
 ;
 ; RV64IZFH-LABEL: test_trunc_si64:
@@ -1611,16 +1601,16 @@ define i64 @test_trunc_si64(half %x) nounwind {
 ; RV32IZHINX-NEXT:    lui a2, %hi(.LCPI9_1)
 ; RV32IZHINX-NEXT:    lw a2, %lo(.LCPI9_1)(a2)
 ; RV32IZHINX-NEXT:    and a0, s2, a0
-; RV32IZHINX-NEXT:    flt.s a3, a2, s0
-; RV32IZHINX-NEXT:    neg a2, a3
+; RV32IZHINX-NEXT:    flt.s a4, a2, s0
+; RV32IZHINX-NEXT:    neg a2, a4
 ; RV32IZHINX-NEXT:    or a0, a2, a0
 ; RV32IZHINX-NEXT:    feq.s a2, s0, s0
 ; RV32IZHINX-NEXT:    neg a2, a2
 ; RV32IZHINX-NEXT:    lui a5, 524288
-; RV32IZHINX-NEXT:    lui a4, 524288
+; RV32IZHINX-NEXT:    lui a3, 524288
 ; RV32IZHINX-NEXT:    beqz s1, .LBB9_4
 ; RV32IZHINX-NEXT:  # %bb.3:
-; RV32IZHINX-NEXT:    mv a4, a1
+; RV32IZHINX-NEXT:    mv a3, a1
 ; RV32IZHINX-NEXT:  .LBB9_4:
 ; RV32IZHINX-NEXT:    and a0, a2, a0
 ; RV32IZHINX-NEXT:    lw ra, 12(sp) # 4-byte Folded Reload
@@ -1628,11 +1618,11 @@ define i64 @test_trunc_si64(half %x) nounwind {
 ; RV32IZHINX-NEXT:    lw s1, 4(sp) # 4-byte Folded Reload
 ; RV32IZHINX-NEXT:    lw s2, 0(sp) # 4-byte Folded Reload
 ; RV32IZHINX-NEXT:    addi sp, sp, 16
-; RV32IZHINX-NEXT:    beqz a3, .LBB9_6
+; RV32IZHINX-NEXT:    beqz a4, .LBB9_6
 ; RV32IZHINX-NEXT:  # %bb.5:
-; RV32IZHINX-NEXT:    addi a4, a5, -1
+; RV32IZHINX-NEXT:    addi a3, a5, -1
 ; RV32IZHINX-NEXT:  .LBB9_6:
-; RV32IZHINX-NEXT:    and a1, a2, a4
+; RV32IZHINX-NEXT:    and a1, a2, a3
 ; RV32IZHINX-NEXT:    ret
 ;
 ; RV64IZHINX-LABEL: test_trunc_si64:
@@ -1670,41 +1660,39 @@ define i64 @test_trunc_si64(half %x) nounwind {
 ; RV32IZFHMIN-NEXT:    addi sp, sp, -16
 ; RV32IZFHMIN-NEXT:    sw ra, 12(sp) # 4-byte Folded Spill
 ; RV32IZFHMIN-NEXT:    sw s0, 8(sp) # 4-byte Folded Spill
-; RV32IZFHMIN-NEXT:    sw s1, 4(sp) # 4-byte Folded Spill
-; RV32IZFHMIN-NEXT:    fsw fs0, 0(sp) # 4-byte Folded Spill
+; RV32IZFHMIN-NEXT:    fsw fs0, 4(sp) # 4-byte Folded Spill
 ; RV32IZFHMIN-NEXT:    fcvt.h.s fa5, fa5
 ; RV32IZFHMIN-NEXT:    fcvt.s.h fs0, fa5
 ; RV32IZFHMIN-NEXT:    lui a0, 913408
 ; RV32IZFHMIN-NEXT:    fmv.w.x fa5, a0
 ; RV32IZFHMIN-NEXT:    fle.s s0, fa5, fs0
-; RV32IZFHMIN-NEXT:    neg s1, s0
 ; RV32IZFHMIN-NEXT:    fmv.s fa0, fs0
 ; RV32IZFHMIN-NEXT:    call __fixsfdi
-; RV32IZFHMIN-NEXT:    lui a2, %hi(.LCPI9_0)
-; RV32IZFHMIN-NEXT:    flw fa5, %lo(.LCPI9_0)(a2)
-; RV32IZFHMIN-NEXT:    and a0, s1, a0
-; RV32IZFHMIN-NEXT:    flt.s a3, fa5, fs0
-; RV32IZFHMIN-NEXT:    neg a2, a3
-; RV32IZFHMIN-NEXT:    or a0, a2, a0
-; RV32IZFHMIN-NEXT:    feq.s a2, fs0, fs0
-; RV32IZFHMIN-NEXT:    neg a2, a2
-; RV32IZFHMIN-NEXT:    lui a5, 524288
 ; RV32IZFHMIN-NEXT:    lui a4, 524288
+; RV32IZFHMIN-NEXT:    lui a2, 524288
 ; RV32IZFHMIN-NEXT:    beqz s0, .LBB9_4
 ; RV32IZFHMIN-NEXT:  # %bb.3:
-; RV32IZFHMIN-NEXT:    mv a4, a1
+; RV32IZFHMIN-NEXT:    mv a2, a1
 ; RV32IZFHMIN-NEXT:  .LBB9_4:
+; RV32IZFHMIN-NEXT:    lui a1, %hi(.LCPI9_0)
+; RV32IZFHMIN-NEXT:    flw fa5, %lo(.LCPI9_0)(a1)
+; RV32IZFHMIN-NEXT:    flt.s a3, fa5, fs0
+; RV32IZFHMIN-NEXT:    beqz a3, .LBB9_6
+; RV32IZFHMIN-NEXT:  # %bb.5:
+; RV32IZFHMIN-NEXT:    addi a2, a4, -1
+; RV32IZFHMIN-NEXT:  .LBB9_6:
+; RV32IZFHMIN-NEXT:    feq.s a1, fs0, fs0
+; RV32IZFHMIN-NEXT:    neg a4, a1
+; RV32IZFHMIN-NEXT:    and a1, a4, a2
+; RV32IZFHMIN-NEXT:    neg a2, s0
 ; RV32IZFHMIN-NEXT:    and a0, a2, a0
+; RV32IZFHMIN-NEXT:    neg a2, a3
+; RV32IZFHMIN-NEXT:    or a0, a2, a0
+; RV32IZFHMIN-NEXT:    and a0, a4, a0
 ; RV32IZFHMIN-NEXT:    lw ra, 12(sp) # 4-byte Folded Reload
 ; RV32IZFHMIN-NEXT:    lw s0, 8(sp) # 4-byte Folded Reload
-; RV32IZFHMIN-NEXT:    lw s1, 4(sp) # 4-byte Folded Reload
-; RV32IZFHMIN-NEXT:    flw fs0, 0(sp) # 4-byte Folded Reload
+; RV32IZFHMIN-NEXT:    flw fs0, 4(sp) # 4-byte Folded Reload
 ; RV32IZFHMIN-NEXT:    addi sp, sp, 16
-; RV32IZFHMIN-NEXT:    beqz a3, .LBB9_6
-; RV32IZFHMIN-NEXT:  # %bb.5:
-; RV32IZFHMIN-NEXT:    addi a4, a5, -1
-; RV32IZFHMIN-NEXT:  .LBB9_6:
-; RV32IZFHMIN-NEXT:    and a1, a2, a4
 ; RV32IZFHMIN-NEXT:    ret
 ;
 ; RV64IZFHMIN-LABEL: test_trunc_si64:
@@ -1756,16 +1744,16 @@ define i64 @test_trunc_si64(half %x) nounwind {
 ; RV32IZHINXMIN-NEXT:    lui a2, %hi(.LCPI9_0)
 ; RV32IZHINXMIN-NEXT:    lw a2, %lo(.LCPI9_0)(a2)
 ; RV32IZHINXMIN-NEXT:    and a0, s2, a0
-; RV32IZHINXMIN-NEXT:    flt.s a3, a2, s0
-; RV32IZHINXMIN-NEXT:    neg a2, a3
+; RV32IZHINXMIN-NEXT:    flt.s a4, a2, s0
+; RV32IZHINXMIN-NEXT:    neg a2, a4
 ; RV32IZHINXMIN-NEXT:    or a0, a2, a0
 ; RV32IZHINXMIN-NEXT:    feq.s a2, s0, s0
 ; RV32IZHINXMIN-NEXT:    neg a2, a2
 ; RV32IZHINXMIN-NEXT:    lui a5, 524288
-; RV32IZHINXMIN-NEXT:    lui a4, 524288
+; RV32IZHINXMIN-NEXT:    lui a3, 524288
 ; RV32IZHINXMIN-NEXT:    beqz s1, .LBB9_4
 ; RV32IZHINXMIN-NEXT:  # %bb.3:
-; RV32IZHINXMIN-NEXT:    mv a4, a1
+; RV32IZHINXMIN-NEXT:    mv a3, a1
 ; RV32IZHINXMIN-NEXT:  .LBB9_4:
 ; RV32IZHINXMIN-NEXT:    and a0, a2, a0
 ; RV32IZHINXMIN-NEXT:    lw ra, 12(sp) # 4-byte Folded Reload
@@ -1773,11 +1761,11 @@ define i64 @test_trunc_si64(half %x) nounwind {
 ; RV32IZHINXMIN-NEXT:    lw s1, 4(sp) # 4-byte Folded Reload
 ; RV32IZHINXMIN-NEXT:    lw s2, 0(sp) # 4-byte Folded Reload
 ; RV32IZHINXMIN-NEXT:    addi sp, sp, 16
-; RV32IZHINXMIN-NEXT:    beqz a3, .LBB9_6
+; RV32IZHINXMIN-NEXT:    beqz a4, .LBB9_6
 ; RV32IZHINXMIN-NEXT:  # %bb.5:
-; RV32IZHINXMIN-NEXT:    addi a4, a5, -1
+; RV32IZHINXMIN-NEXT:    addi a3, a5, -1
 ; RV32IZHINXMIN-NEXT:  .LBB9_6:
-; RV32IZHINXMIN-NEXT:    and a1, a2, a4
+; RV32IZHINXMIN-NEXT:    and a1, a2, a3
 ; RV32IZHINXMIN-NEXT:    ret
 ;
 ; RV64IZHINXMIN-LABEL: test_trunc_si64:
@@ -2256,40 +2244,38 @@ define i64 @test_round_si64(half %x) nounwind {
 ; RV32IZFH-NEXT:    addi sp, sp, -16
 ; RV32IZFH-NEXT:    sw ra, 12(sp) # 4-byte Folded Spill
 ; RV32IZFH-NEXT:    sw s0, 8(sp) # 4-byte Folded Spill
-; RV32IZFH-NEXT:    sw s1, 4(sp) # 4-byte Folded Spill
-; RV32IZFH-NEXT:    fsw fs0, 0(sp) # 4-byte Folded Spill
+; RV32IZFH-NEXT:    fsw fs0, 4(sp) # 4-byte Folded Spill
 ; RV32IZFH-NEXT:    fcvt.s.h fs0, fa0
 ; RV32IZFH-NEXT:    lui a0, 913408
 ; RV32IZFH-NEXT:    fmv.w.x fa5, a0
 ; RV32IZFH-NEXT:    fle.s s0, fa5, fs0
-; RV32IZFH-NEXT:    neg s1, s0
 ; RV32IZFH-NEXT:    fmv.s fa0, fs0
 ; RV32IZFH-NEXT:    call __fixsfdi
-; RV32IZFH-NEXT:    lui a2, %hi(.LCPI13_1)
-; RV32IZFH-NEXT:    flw fa5, %lo(.LCPI13_1)(a2)
-; RV32IZFH-NEXT:    and a0, s1, a0
-; RV32IZFH-NEXT:    flt.s a3, fa5, fs0
-; RV32IZFH-NEXT:    neg a2, a3
-; RV32IZFH-NEXT:    or a0, a2, a0
-; RV32IZFH-NEXT:    feq.s a2, fs0, fs0
-; RV32IZFH-NEXT:    neg a2, a2
-; RV32IZFH-NEXT:    lui a5, 524288
 ; RV32IZFH-NEXT:    lui a4, 524288
+; RV32IZFH-NEXT:    lui a2, 524288
 ; RV32IZFH-NEXT:    beqz s0, .LBB13_4
 ; RV32IZFH-NEXT:  # %bb.3:
-; RV32IZFH-NEXT:    mv a4, a1
+; RV32IZFH-NEXT:    mv a2, a1
 ; RV32IZFH-NEXT:  .LBB13_4:
+; RV32IZFH-NEXT:    lui a1, %hi(.LCPI13_1)
+; RV32IZFH-NEXT:    flw fa5, %lo(.LCPI13_1)(a1)
+; RV32IZFH-NEXT:    flt.s a3, fa5, fs0
+; RV32IZFH-NEXT:    beqz a3, .LBB13_6
+; RV32IZFH-NEXT:  # %bb.5:
+; RV32IZFH-NEXT:    addi a2, a4, -1
+; RV32IZFH-NEXT:  .LBB13_6:
+; RV32IZFH-NEXT:    feq.s a1, fs0, fs0
+; RV32IZFH-NEXT:    neg a4, a1
+; RV32IZFH-NEXT:    and a1, a4, a2
+; RV32IZFH-NEXT:    neg a2, s0
 ; RV32IZFH-NEXT:    and a0, a2, a0
+; RV32IZFH-NEXT:    neg a2, a3
+; RV32IZFH-NEXT:    or a0, a2, a0
+; RV32IZFH-NEXT:    and a0, a4, a0
 ; RV32IZFH-NEXT:    lw ra, 12(sp) # 4-byte Folded Reload
 ; RV32IZFH-NEXT:    lw s0, 8(sp) # 4-byte Folded Reload
-; RV32IZFH-NEXT:    lw s1, 4(sp) # 4-byte Folded Reload
-; RV32IZFH-NEXT:    flw fs0, 0(sp) # 4-byte Folded Reload
+; RV32IZFH-NEXT:    flw fs0, 4(sp) # 4-byte Folded Reload
 ; RV32IZFH-NEXT:    addi sp, sp, 16
-; RV32IZFH-NEXT:    beqz a3, .LBB13_6
-; RV32IZFH-NEXT:  # %bb.5:
-; RV32IZFH-NEXT:    addi a4, a5, -1
-; RV32IZFH-NEXT:  .LBB13_6:
-; RV32IZFH-NEXT:    and a1, a2, a4
 ; RV32IZFH-NEXT:    ret
 ;
 ; RV64IZFH-LABEL: test_round_si64:
@@ -2327,16 +2313,16 @@ define i64 @test_round_si64(half %x) nounwind {
 ; RV32IZHINX-NEXT:    lui a2, %hi(.LCPI13_1)
 ; RV32IZHINX-NEXT:    lw a2, %lo(.LCPI13_1)(a2)
 ; RV32IZHINX-NEXT:    and a0, s2, a0
-; RV32IZHINX-NEXT:    flt.s a3, a2, s0
-; RV32IZHINX-NEXT:    neg a2, a3
+; RV32IZHINX-NEXT:    flt.s a4, a2, s0
+; RV32IZHINX-NEXT:    neg a2, a4
 ; RV32IZHINX-NEXT:    or a0, a2, a0
 ; RV32IZHINX-NEXT:    feq.s a2, s0, s0
 ; RV32IZHINX-NEXT:    neg a2, a2
 ; RV32IZHINX-NEXT:    lui a5, 524288
-; RV32IZHINX-NEXT:    lui a4, 524288
+; RV32IZHINX-NEXT:    lui a3, 524288
 ; RV32IZHINX-NEXT:    beqz s1, .LBB13_4
 ; RV32IZHINX-NEXT:  # %bb.3:
-; RV32IZHINX-NEXT:    mv a4, a1
+; RV32IZHINX-NEXT:    mv a3, a1
 ; RV32IZHINX-NEXT:  .LBB13_4:
 ; RV32IZHINX-NEXT:    and a0, a2, a0
 ; RV32IZHINX-NEXT:    lw ra, 12(sp) # 4-byte Folded Reload
@@ -2344,11 +2330,11 @@ define i64 @test_round_si64(half %x) nounwind {
 ; RV32IZHINX-NEXT:    lw s1, 4(sp) # 4-byte Folded Reload
 ; RV32IZHINX-NEXT:    lw s2, 0(sp) # 4-byte Folded Reload
 ; RV32IZHINX-NEXT:    addi sp, sp, 16
-; RV32IZHINX-NEXT:    beqz a3, .LBB13_6
+; RV32IZHINX-NEXT:    beqz a4, .LBB13_6
 ; RV32IZHINX-NEXT:  # %bb.5:
-; RV32IZHINX-NEXT:    addi a4, a5, -1
+; RV32IZHINX-NEXT:    addi a3, a5, -1
 ; RV32IZHINX-NEXT:  .LBB13_6:
-; RV32IZHINX-NEXT:    and a1, a2, a4
+; RV32IZHINX-NEXT:    and a1, a2, a3
 ; RV32IZHINX-NEXT:    ret
 ;
 ; RV64IZHINX-LABEL: test_round_si64:
@@ -2386,41 +2372,39 @@ define i64 @test_round_si64(half %x) nounwind {
 ; RV32IZFHMIN-NEXT:    addi sp, sp, -16
 ; RV32IZFHMIN-NEXT:    sw ra, 12(sp) # 4-byte Folded Spill
 ; RV32IZFHMIN-NEXT:    sw s0, 8(sp) # 4-byte Folded Spill
-; RV32IZFHMIN-NEXT:    sw s1, 4(sp) # 4-byte Folded Spill
-; RV32IZFHMIN-NEXT:    fsw fs0, 0(sp) # 4-byte Folded Spill
+; RV32IZFHMIN-NEXT:    fsw fs0, 4(sp) # 4-byte Folded Spill
 ; RV32IZFHMIN-NEXT:    fcvt.h.s fa5, fa5
 ; RV32IZFHMIN-NEXT:    fcvt.s.h fs0, fa5
 ; RV32IZFHMIN-NEXT:    lui a0, 913408
 ; RV32IZFHMIN-NEXT:    fmv.w.x fa5, a0
 ; RV32IZFHMIN-NEXT:    fle.s s0, fa5, fs0
-; RV32IZFHMIN-NEXT:    neg s1, s0
 ; RV32IZFHMIN-NEXT:    fmv.s fa0, fs0
 ; RV32IZFHMIN-NEXT:    call __fixsfdi
-; RV32IZFHMIN-NEXT:    lui a2, %hi(.LCPI13_0)
-; RV32IZFHMIN-NEXT:    flw fa5, %lo(.LCPI13_0)(a2)
-; RV32IZFHMIN-NEXT:    and a0, s1, a0
-; RV32IZFHMIN-NEXT:    flt.s a3, fa5, fs0
-; RV32IZFHMIN-NEXT:    neg a2, a3
-; RV32IZFHMIN-NEXT:    or a0, a2, a0
-; RV32IZFHMIN-NEXT:    feq.s a2, fs0, fs0
-; RV32IZFHMIN-NEXT:    neg a2, a2
-; RV32IZFHMIN-NEXT:    lui a5, 524288
 ; RV32IZFHMIN-NEXT:    lui a4, 524288
+; RV32IZFHMIN-NEXT:    lui a2, 524288
 ; RV32IZFHMIN-NEXT:    beqz s0, .LBB13_4
 ; RV32IZFHMIN-NEXT:  # %bb.3:
-; RV32IZFHMIN-NEXT:    mv a4, a1
+; RV32IZFHMIN-NEXT:    mv a2, a1
 ; RV32IZFHMIN-NEXT:  .LBB13_4:
+; RV32IZFHMIN-NEXT:    lui a1, %hi(.LCPI13_0)
+; RV32IZFHMIN-NEXT:    flw fa5, %lo(.LCPI13_0)(a1)
+; RV32IZFHMIN-NEXT:    flt.s a3, fa5, fs0
+; RV32IZFHMIN-NEXT:    beqz a3, .LBB13_6
+; RV32IZFHMIN-NEXT:  # %bb.5:
+; RV32IZFHMIN-NEXT:    addi a2, a4, -1
+; RV32IZFHMIN-NEXT:  .LBB13_6:
+; RV32IZFHMIN-NEXT:    feq.s a1, fs0, fs0
+; RV32IZFHMIN-NEXT:    neg a4, a1
+; RV32IZFHMIN-NEXT:    and a1, a4, a2
+; RV32IZFHMIN-NEXT:    neg a2, s0
 ; RV32IZFHMIN-NEXT:    and a0, a2, a0
+; RV32IZFHMIN-NEXT:    neg a2, a3
+; RV32IZFHMIN-NEXT:    or a0, a2, a0
+; RV32IZFHMIN-NEXT:    and a0, a4, a0
 ; RV32IZFHMIN-NEXT:    lw ra, 12(sp) # 4-byte Folded Reload
 ; RV32IZFHMIN-NEXT:    lw s0, 8(sp) # 4-byte Folded Reload
-; RV32IZFHMIN-NEXT:    lw s1, 4(sp) # 4-byte Folded Reload
-; RV32IZFHMIN-NEXT:    flw fs0, 0(sp) # 4-byte Folded Reload
+; RV32IZFHMIN-NEXT:    flw fs0, 4(sp) # 4-byte Folded Reload
 ; RV32IZFHMIN-NEXT:    addi sp, sp, 16
-; RV32IZFHMIN-NEXT:    beqz a3, .LBB13_6
-; RV32IZFHMIN-NEXT:  # %bb.5:
-; RV32IZFHMIN-NEXT:    addi a4, a5, -1
-; RV32IZFHMIN-NEXT:  .LBB13_6:
-; RV32IZFHMIN-NEXT:    and a1, a2, a4
 ; RV32IZFHMIN-NEXT:    ret
 ;
 ; RV64IZFHMIN-LABEL: test_round_si64:
@@ -2472,16 +2456,16 @@ define i64 @test_round_si64(half %x) nounwind {
 ; RV32IZHINXMIN-NEXT:    lui a2, %hi(.LCPI13_0)
 ; RV32IZHINXMIN-NEXT:    lw a2, %lo(.LCPI13_0)(a2)
 ; RV32IZHINXMIN-NEXT:    and a0, s2, a0
-; RV32IZHINXMIN-NEXT:    flt.s a3, a2, s0
-; RV32IZHINXMIN-NEXT:    neg a2, a3
+; RV32IZHINXMIN-NEXT:    flt.s a4, a2, s0
+; RV32IZHINXMIN-NEXT:    neg a2, a4
 ; RV32IZHINXMIN-NEXT:    or a0, a2, a0
 ; RV32IZHINXMIN-NEXT:    feq.s a2, s0, s0
 ; RV32IZHINXMIN-NEXT:    neg a2, a2
 ; RV32IZHINXMIN-NEXT:    lui a5, 524288
-; RV32IZHINXMIN-NEXT:    lui a4, 524288
+; RV32IZHINXMIN-NEXT:    lui a3, 524288
 ; RV32IZHINXMIN-NEXT:    beqz s1, .LBB13_4
 ; RV32IZHINXMIN-NEXT:  # %bb.3:
-; RV32IZHINXMIN-NEXT:    mv a4, a1
+; RV32IZHINXMIN-NEXT:    mv a3, a1
 ; RV32IZHINXMIN-NEXT:  .LBB13_4:
 ; RV32IZHINXMIN-NEXT:    and a0, a2, a0
 ; RV32IZHINXMIN-NEXT:    lw ra, 12(sp) # 4-byte Folded Reload
@@ -2489,11 +2473,11 @@ define i64 @test_round_si64(half %x) nounwind {
 ; RV32IZHINXMIN-NEXT:    lw s1, 4(sp) # 4-byte Folded Reload
 ; RV32IZHINXMIN-NEXT:    lw s2, 0(sp) # 4-byte Folded Reload
 ; RV32IZHINXMIN-NEXT:    addi sp, sp, 16
-; RV32IZHINXMIN-NEXT:    beqz a3, .LBB13_6
+; RV32IZHINXMIN-NEXT:    beqz a4, .LBB13_6
 ; RV32IZHINXMIN-NEXT:  # %bb.5:
-; RV32IZHINXMIN-NEXT:    addi a4, a5, -1
+; RV32IZHINXMIN-NEXT:    addi a3, a5, -1
 ; RV32IZHINXMIN-NEXT:  .LBB13_6:
-; RV32IZHINXMIN-NEXT:    and a1, a2, a4
+; RV32IZHINXMIN-NEXT:    and a1, a2, a3
 ; RV32IZHINXMIN-NEXT:    ret
 ;
 ; RV64IZHINXMIN-LABEL: test_round_si64:
@@ -2972,40 +2956,38 @@ define i64 @test_roundeven_si64(half %x) nounwind {
 ; RV32IZFH-NEXT:    addi sp, sp, -16
 ; RV32IZFH-NEXT:    sw ra, 12(sp) # 4-byte Folded Spill
 ; RV32IZFH-NEXT:    sw s0, 8(sp) # 4-byte Folded Spill
-; RV32IZFH-NEXT:    sw s1, 4(sp) # 4-byte Folded Spill
-; RV32IZFH-NEXT:    fsw fs0, 0(sp) # 4-byte Folded Spill
+; RV32IZFH-NEXT:    fsw fs0, 4(sp) # 4-byte Folded Spill
 ; RV32IZFH-NEXT:    fcvt.s.h fs0, fa0
 ; RV32IZFH-NEXT:    lui a0, 913408
 ; RV32IZFH-NEXT:    fmv.w.x fa5, a0
 ; RV32IZFH-NEXT:    fle.s s0, fa5, fs0
-; RV32IZFH-NEXT:    neg s1, s0
 ; RV32IZFH-NEXT:    fmv.s fa0, fs0
 ; RV32IZFH-NEXT:    call __fixsfdi
-; RV32IZFH-NEXT:    lui a2, %hi(.LCPI17_1)
-; RV32IZFH-NEXT:    flw fa5, %lo(.LCPI17_1)(a2)
-; RV32IZFH-NEXT:    and a0, s1, a0
-; RV32IZFH-NEXT:    flt.s a3, fa5, fs0
-; RV32IZFH-NEXT:    neg a2, a3
-; RV32IZFH-NEXT:    or a0, a2, a0
-; RV32IZFH-NEXT:    feq.s a2, fs0, fs0
-; RV32IZFH-NEXT:    neg a2, a2
-; RV32IZFH-NEXT:    lui a5, 524288
 ; RV32IZFH-NEXT:    lui a4, 524288
+; RV32IZFH-NEXT:    lui a2, 524288
 ; RV32IZFH-NEXT:    beqz s0, .LBB17_4
 ; RV32IZFH-NEXT:  # %bb.3:
-; RV32IZFH-NEXT:    mv a4, a1
+; RV32IZFH-NEXT:    mv a2, a1
 ; RV32IZFH-NEXT:  .LBB17_4:
+; RV32IZFH-NEXT:    lui a1, %hi(.LCPI17_1)
+; RV32IZFH-NEXT:    flw fa5, %lo(.LCPI17_1)(a1)
+; RV32IZFH-NEXT:    flt.s a3, fa5, fs0
+; RV32IZFH-NEXT:    beqz a3, .LBB17_6
+; RV32IZFH-NEXT:  # %bb.5:
+; RV32IZFH-NEXT:    addi a2, a4, -1
+; RV32IZFH-NEXT:  .LBB17_6:
+; RV32IZFH-NEXT:    feq.s a1, fs0, fs0
+; RV32IZFH-NEXT:    neg a4, a1
+; RV32IZFH-NEXT:    and a1, a4, a2
+; RV32IZFH-NEXT:    neg a2, s0
 ; RV32IZFH-NEXT:    and a0, a2, a0
+; RV32IZFH-NEXT:    neg a2, a3
+; RV32IZFH-NEXT:    or a0, a2, a0
+; RV32IZFH-NEXT:    and a0, a4, a0
 ; RV32IZFH-NEXT:    lw ra, 12(sp) # 4-byte Folded Reload
 ; RV32IZFH-NEXT:    lw s0, 8(sp) # 4-byte Folded Reload
-; RV32IZFH-NEXT:    lw s1, 4(sp) # 4-byte Folded Reload
-; RV32IZFH-NEXT:    flw fs0, 0(sp) # 4-byte Folded Reload
+; RV32IZFH-NEXT:    flw fs0, 4(sp) # 4-byte Folded Reload
 ; RV32IZFH-NEXT:    addi sp, sp, 16
-; RV32IZFH-NEXT:    beqz a3, .LBB17_6
-; RV32IZFH-NEXT:  # %bb.5:
-; RV32IZFH-NEXT:    addi a4, a5, -1
-; RV32IZFH-NEXT:  .LBB17_6:
-; RV32IZFH-NEXT:    and a1, a2, a4
 ; RV32IZFH-NEXT:    ret
 ;
 ; RV64IZFH-LABEL: test_roundeven_si64:
@@ -3043,16 +3025,16 @@ define i64 @test_roundeven_si64(half %x) nounwind {
 ; RV32IZHINX-NEXT:    lui a2, %hi(.LCPI17_1)
 ; RV32IZHINX-NEXT:    lw a2, %lo(.LCPI17_1)(a2)
 ; RV32IZHINX-NEXT:    and a0, s2, a0
-; RV32IZHINX-NEXT:    flt.s a3, a2, s0
-; RV32IZHINX-NEXT:    neg a2, a3
+; RV32IZHINX-NEXT:    flt.s a4, a2, s0
+; RV32IZHINX-NEXT:    neg a2, a4
 ; RV32IZHINX-NEXT:    or a0, a2, a0
 ; RV32IZHINX-NEXT:    feq.s a2, s0, s0
 ; RV32IZHINX-NEXT:    neg a2, a2
 ; RV32IZHINX-NEXT:    lui a5, 524288
-; RV32IZHINX-NEXT:    lui a4, 524288
+; RV32IZHINX-NEXT:    lui a3, 524288
 ; RV32IZHINX-NEXT:    beqz s1, .LBB17_4
 ; RV32IZHINX-NEXT:  # %bb.3:
-; RV32IZHINX-NEXT:    mv a4, a1
+; RV32IZHINX-NEXT:    mv a3, a1
 ; RV32IZHINX-NEXT:  .LBB17_4:
 ; RV32IZHINX-NEXT:    and a0, a2, a0
 ; RV32IZHINX-NEXT:    lw ra, 12(sp) # 4-byte Folded Reload
@@ -3060,11 +3042,11 @@ define i64 @test_roundeven_si64(half %x) nounwind {
 ; RV32IZHINX-NEXT:    lw s1, 4(sp) # 4-byte Folded Reload
 ; RV32IZHINX-NEXT:    lw s2, 0(sp) # 4-byte Folded Reload
 ; RV32IZHINX-NEXT:    addi sp, sp, 16
-; RV32IZHINX-NEXT:    beqz a3, .LBB17_6
+; RV32IZHINX-NEXT:    beqz a4, .LBB17_6
 ; RV32IZHINX-NEXT:  # %bb.5:
-; RV32IZHINX-NEXT:    addi a4, a5, -1
+; RV32IZHINX-NEXT:    addi a3, a5, -1
 ; RV32IZHINX-NEXT:  .LBB17_6:
-; RV32IZHINX-NEXT:    and a1, a2, a4
+; RV32IZHINX-NEXT:    and a1, a2, a3
 ; RV32IZHINX-NEXT:    ret
 ;
 ; RV64IZHINX-LABEL: test_roundeven_si64:
@@ -3102,41 +3084,39 @@ define i64 @test_roundeven_si64(half %x) nounwind {
 ; RV32IZFHMIN-NEXT:    addi sp, sp, -16
 ; RV32IZFHMIN-NEXT:    sw ra, 12(sp) # 4-byte Folded Spill
 ; RV32IZFHMIN-NEXT:    sw s0, 8(sp) # 4-byte Folded Spill
-; RV32IZFHMIN-NEXT:    sw s1, 4(sp) # 4-byte Folded Spill
-; RV32IZFHMIN-NEXT:    fsw fs0, 0(sp) # 4-byte Folded Spill
+; RV32IZFHMIN-NEXT:    fsw fs0, 4(sp) # 4-byte Folded Spill
 ; RV32IZFHMIN-NEXT:    fcvt.h.s fa5, fa5
 ; RV32IZFHMIN-NEXT:    fcvt.s.h fs0, fa5
 ; RV32IZFHMIN-NEXT:    lui a0, 913408
 ; RV32IZFHMIN-NEXT:    fmv.w.x fa5, a0
 ; RV32IZFHMIN-NEXT:    fle.s s0, fa5, fs0
-; RV32IZFHMIN-NEXT:    neg s1, s0
 ; RV32IZFHMIN-NEXT:    fmv.s fa0, fs0
 ; RV32IZFHMIN-NEXT:    call __fixsfdi
-; RV32IZFHMIN-NEXT:    lui a2, %hi(.LCPI17_0)
-; RV32IZFHMIN-NEXT:    flw fa5, %lo(.LCPI17_0)(a2)
-; RV32IZFHMIN-NEXT:    and a0, s1, a0
-; RV32IZFHMIN-NEXT:    flt.s a3, fa5, fs0
-; RV32IZFHMIN-NEXT:    neg a2, a3
-; RV32IZFHMIN-NEXT:    or a0, a2, a0
-; RV32IZFHMIN-NEXT:    feq.s a2, fs0, fs0
-; RV32IZFHMIN-NEXT:    neg a2, a2
-; RV32IZFHMIN-NEXT:    lui a5, 524288
 ; RV32IZFHMIN-NEXT:    lui a4, 524288
+; RV32IZFHMIN-NEXT:    lui a2, 524288
 ; RV32IZFHMIN-NEXT:    beqz s0, .LBB17_4
 ; RV32IZFHMIN-NEXT:  # %bb.3:
-; RV32IZFHMIN-NEXT:    mv a4, a1
+; RV32IZFHMIN-NEXT:    mv a2, a1
 ; RV32IZFHMIN-NEXT:  .LBB17_4:
+; RV32IZFHMIN-NEXT:    lui a1, %hi(.LCPI17_0)
+; RV32IZFHMIN-NEXT:    flw fa5, %lo(.LCPI17_0)(a1)
+; RV32IZFHMIN-NEXT:    flt.s a3, fa5, fs0
+; RV32IZFHMIN-NEXT:    beqz a3, .LBB17_6
+; RV32IZFHMIN-NEXT:  # %bb.5:
+; RV32IZFHMIN-NEXT:    addi a2, a4, -1
+; RV32IZFHMIN-NEXT:  .LBB17_6:
+; RV32IZFHMIN-NEXT:    feq.s a1, fs0, fs0
+; RV32IZFHMIN-NEXT:    neg a4, a1
+; RV32IZFHMIN-NEXT:    and a1, a4, a2
+; RV32IZFHMIN-NEXT:    neg a2, s0
 ; RV32IZFHMIN-NEXT:    and a0, a2, a0
+; RV32IZFHMIN-NEXT:    neg a2, a3
+; RV32IZFHMIN-NEXT:    or a0, a2, a0
+; RV32IZFHMIN-NEXT:    and a0, a4, a0
 ; RV32IZFHMIN-NEXT:    lw ra, 12(sp) # 4-byte Folded Reload
 ; RV32IZFHMIN-NEXT:    lw s0, 8(sp) # 4-byte Folded Reload
-; RV32IZFHMIN-NEXT:    lw s1, 4(sp) # 4-byte Folded Reload
-; RV32IZFHMIN-NEXT:    flw fs0, 0(sp) # 4-byte Folded Reload
+; RV32IZFHMIN-NEXT:    flw fs0, 4(sp) # 4-byte Folded Reload
 ; RV32IZFHMIN-NEXT:    addi sp, sp, 16
-; RV32IZFHMIN-NEXT:    beqz a3, .LBB17_6
-; RV32IZFHMIN-NEXT:  # %bb.5:
-; RV32IZFHMIN-NEXT:    addi a4, a5, -1
-; RV32IZFHMIN-NEXT:  .LBB17_6:
-; RV32IZFHMIN-NEXT:    and a1, a2, a4
 ; RV32IZFHMIN-NEXT:    ret
 ;
 ; RV64IZFHMIN-LABEL: test_roundeven_si64:
@@ -3188,16 +3168,16 @@ define i64 @test_roundeven_si64(half %x) nounwind {
 ; RV32IZHINXMIN-NEXT:    lui a2, %hi(.LCPI17_0)
 ; RV32IZHINXMIN-NEXT:    lw a2, %lo(.LCPI17_0)(a2)
 ; RV32IZHINXMIN-NEXT:    and a0, s2, a0
-; RV32IZHINXMIN-NEXT:    flt.s a3, a2, s0
-; RV32IZHINXMIN-NEXT:    neg a2, a3
+; RV32IZHINXMIN-NEXT:    flt.s a4, a2, s0
+; RV32IZHINXMIN-NEXT:    neg a2, a4
 ; RV32IZHINXMIN-NEXT:    or a0, a2, a0
 ; RV32IZHINXMIN-NEXT:    feq.s a2, s0, s0
 ; RV32IZHINXMIN-NEXT:    neg a2, a2
 ; RV32IZHINXMIN-NEXT:    lui a5, 524288
-; RV32IZHINXMIN-NEXT:    lui a4, 524288
+; RV32IZHINXMIN-NEXT:    lui a3, 524288
 ; RV32IZHINXMIN-NEXT:    beqz s1, .LBB17_4
 ; RV32IZHINXMIN-NEXT:  # %bb.3:
-; RV32IZHINXMIN-NEXT:    mv a4, a1
+; RV32IZHINXMIN-NEXT:    mv a3, a1
 ; RV32IZHINXMIN-NEXT:  .LBB17_4:
 ; RV32IZHINXMIN-NEXT:    and a0, a2, a0
 ; RV32IZHINXMIN-NEXT:    lw ra, 12(sp) # 4-byte Folded Reload
@@ -3205,11 +3185,11 @@ define i64 @test_roundeven_si64(half %x) nounwind {
 ; RV32IZHINXMIN-NEXT:    lw s1, 4(sp) # 4-byte Folded Reload
 ; RV32IZHINXMIN-NEXT:    lw s2, 0(sp) # 4-byte Folded Reload
 ; RV32IZHINXMIN-NEXT:    addi sp, sp, 16
-; RV32IZHINXMIN-NEXT:    beqz a3, .LBB17_6
+; RV32IZHINXMIN-NEXT:    beqz a4, .LBB17_6
 ; RV32IZHINXMIN-NEXT:  # %bb.5:
-; RV32IZHINXMIN-NEXT:    addi a4, a5, -1
+; RV32IZHINXMIN-NEXT:    addi a3, a5, -1
 ; RV32IZHINXMIN-NEXT:  .LBB17_6:
-; RV32IZHINXMIN-NEXT:    and a1, a2, a4
+; RV32IZHINXMIN-NEXT:    and a1, a2, a3
 ; RV32IZHINXMIN-NEXT:    ret
 ;
 ; RV64IZHINXMIN-LABEL: test_roundeven_si64:
@@ -3688,40 +3668,38 @@ define i64 @test_rint_si64(half %x) nounwind {
 ; RV32IZFH-NEXT:    addi sp, sp, -16
 ; RV32IZFH-NEXT:    sw ra, 12(sp) # 4-byte Folded Spill
 ; RV32IZFH-NEXT:    sw s0, 8(sp) # 4-byte Folded Spill
-; RV32IZFH-NEXT:    sw s1, 4(sp) # 4-byte Folded Spill
-; RV32IZFH-NEXT:    fsw fs0, 0(sp) # 4-byte Folded Spill
+; RV32IZFH-NEXT:    fsw fs0, 4(sp) # 4-byte Folded Spill
 ; RV32IZFH-NEXT:    fcvt.s.h fs0, fa0
 ; RV32IZFH-NEXT:    lui a0, 913408
 ; RV32IZFH-NEXT:    fmv.w.x fa5, a0
 ; RV32IZFH-NEXT:    fle.s s0, fa5, fs0
-; RV32IZFH-NEXT:    neg s1, s0
 ; RV32IZFH-NEXT:    fmv.s fa0, fs0
 ; RV32IZFH-NEXT:    call __fixsfdi
-; RV32IZFH-NEXT:    lui a2, %hi(.LCPI21_1)
-; RV32IZFH-NEXT:    flw fa5, %lo(.LCPI21_1)(a2)
-; RV32IZFH-NEXT:    and a0, s1, a0
-; RV32IZFH-NEXT:    flt.s a3, fa5, fs0
-; RV32IZFH-NEXT:    neg a2, a3
-; RV32IZFH-NEXT:    or a0, a2, a0
-; RV32IZFH-NEXT:    feq.s a2, fs0, fs0
-; RV32IZFH-NEXT:    neg a2, a2
-; RV32IZFH-NEXT:    lui a5, 524288
 ; RV32IZFH-NEXT:    lui a4, 524288
+; RV32IZFH-NEXT:    lui a2, 524288
 ; RV32IZFH-NEXT:    beqz s0, .LBB21_4
 ; RV32IZFH-NEXT:  # %bb.3:
-; RV32IZFH-NEXT:    mv a4, a1
+; RV32IZFH-NEXT:    mv a2, a1
 ; RV32IZFH-NEXT:  .LBB21_4:
+; RV32IZFH-NEXT:    lui a1, %hi(.LCPI21_1)
+; RV32IZFH-NEXT:    flw fa5, %lo(.LCPI21_1)(a1)
+; RV32IZFH-NEXT:    flt.s a3, fa5, fs0
+; RV32IZFH-NEXT:    beqz a3, .LBB21_6
+; RV32IZFH-NEXT:  # %bb.5:
+; RV32IZFH-NEXT:    addi a2, a4, -1
+; RV32IZFH-NEXT:  .LBB21_6:
+; RV32IZFH-NEXT:    feq.s a1, fs0, fs0
+; RV32IZFH-NEXT:    neg a4, a1
+; RV32IZFH-NEXT:    and a1, a4, a2
+; RV32IZFH-NEXT:    neg a2, s0
 ; RV32IZFH-NEXT:    and a0, a2, a0
+; RV32IZFH-NEXT:    neg a2, a3
+; RV32IZFH-NEXT:    or a0, a2, a0
+; RV32IZFH-NEXT:    and a0, a4, a0
 ; RV32IZFH-NEXT:    lw ra, 12(sp) # 4-byte Folded Reload
 ; RV32IZFH-NEXT:    lw s0, 8(sp) # 4-byte Folded Reload
-; RV32IZFH-NEXT:    lw s1, 4(sp) # 4-byte Folded Reload
-; RV32IZFH-NEXT:    flw fs0, 0(sp) # 4-byte Folded Reload
+; RV32IZFH-NEXT:    flw fs0, 4(sp) # 4-byte Folded Reload
 ; RV32IZFH-NEXT:    addi sp, sp, 16
-; RV32IZFH-NEXT:    beqz a3, .LBB21_6
-; RV32IZFH-NEXT:  # %bb.5:
-; RV32IZFH-NEXT:    addi a4, a5, -1
-; RV32IZFH-NEXT:  .LBB21_6:
-; RV32IZFH-NEXT:    and a1, a2, a4
 ; RV32IZFH-NEXT:    ret
 ;
 ; RV64IZFH-LABEL: test_rint_si64:
@@ -3759,16 +3737,16 @@ define i64 @test_rint_si64(half %x) nounwind {
 ; RV32IZHINX-NEXT:    lui a2, %hi(.LCPI21_1)
 ; RV32IZHINX-NEXT:    lw a2, %lo(.LCPI21_1)(a2)
 ; RV32IZHINX-NEXT:    and a0, s2, a0
-; RV32IZHINX-NEXT:    flt.s a3, a2, s0
-; RV32IZHINX-NEXT:    neg a2, a3
+; RV32IZHINX-NEXT:    flt.s a4, a2, s0
+; RV32IZHINX-NEXT:    neg a2, a4
 ; RV32IZHINX-NEXT:    or a0, a2, a0
 ; RV32IZHINX-NEXT:    feq.s a2, s0, s0
 ; RV32IZHINX-NEXT:    neg a2, a2
 ; RV32IZHINX-NEXT:    lui a5, 524288
-; RV32IZHINX-NEXT:    lui a4, 524288
+; RV32IZHINX-NEXT:    lui a3, 524288
 ; RV32IZHINX-NEXT:    beqz s1, .LBB21_4
 ; RV32IZHINX-NEXT:  # %bb.3:
-; RV32IZHINX-NEXT:    mv a4, a1
+; RV32IZHINX-NEXT:    mv a3, a1
 ; RV32IZHINX-NEXT:  .LBB21_4:
 ; RV32IZHINX-NEXT:    and a0, a2, a0
 ; RV32IZHINX-NEXT:    lw ra, 12(sp) # 4-byte Folded Reload
@@ -3776,11 +3754,11 @@ define i64 @test_rint_si64(half %x) nounwind {
 ; RV32IZHINX-NEXT:    lw s1, 4(sp) # 4-byte Folded Reload
 ; RV32IZHINX-NEXT:    lw s2, 0(sp) # 4-byte Folded Reload
 ; RV32IZHINX-NEXT:    addi sp, sp, 16
-; RV32IZHINX-NEXT:    beqz a3, .LBB21_6
+; RV32IZHINX-NEXT:    beqz a4, .LBB21_6
 ; RV32IZHINX-NEXT:  # %bb.5:
-; RV32IZHINX-NEXT:    addi a4, a5, -1
+; RV32IZHINX-NEXT:    addi a3, a5, -1
 ; RV32IZHINX-NEXT:  .LBB21_6:
-; RV32IZHINX-NEXT:    and a1, a2, a4
+; RV32IZHINX-NEXT:    and a1, a2, a3
 ; RV32IZHINX-NEXT:    ret
 ;
 ; RV64IZHINX-LABEL: test_rint_si64:
@@ -3818,41 +3796,39 @@ define i64 @test_rint_si64(half %x) nounwind {
 ; RV32IZFHMIN-NEXT:    addi sp, sp, -16
 ; RV32IZFHMIN-NEXT:    sw ra, 12(sp) # 4-byte Folded Spill
 ; RV32IZFHMIN-NEXT:    sw s0, 8(sp) # 4-byte Folded Spill
-; RV32IZFHMIN-NEXT:    sw s1, 4(sp) # 4-byte Folded Spill
-; RV32IZFHMIN-NEXT:    fsw fs0, 0(sp) # 4-byte Folded Spill
+; RV32IZFHMIN-NEXT:    fsw fs0, 4(sp) # 4-byte Folded Spill
 ; RV32IZFHMIN-NEXT:    fcvt.h.s fa5, fa5
 ; RV32IZFHMIN-NEXT:    fcvt.s.h fs0, fa5
 ; RV32IZFHMIN-NEXT:    lui a0, 913408
 ; RV32IZFHMIN-NEXT:    fmv.w.x fa5, a0
 ; RV32IZFHMIN-NEXT:    fle.s s0, fa5, fs0
-; RV32IZFHMIN-NEXT:    neg s1, s0
 ; RV32IZFHMIN-NEXT:    fmv.s fa0, fs0
 ; RV32IZFHMIN-NEXT:    call __fixsfdi
-; RV32IZFHMIN-NEXT:    lui a2, %hi(.LCPI21_0)
-; RV32IZFHMIN-NEXT:    flw fa5, %lo(.LCPI21_0)(a2)
-; RV32IZFHMIN-NEXT:    and a0, s1, a0
-; RV32IZFHMIN-NEXT:    flt.s a3, fa5, fs0
-; RV32IZFHMIN-NEXT:    neg a2, a3
-; RV32IZFHMIN-NEXT:    or a0, a2, a0
-; RV32IZFHMIN-NEXT:    feq.s a2, fs0, fs0
-; RV32IZFHMIN-NEXT:    neg a2, a2
-; RV32IZFHMIN-NEXT:    lui a5, 524288
 ; RV32IZFHMIN-NEXT:    lui a4, 524288
+; RV32IZFHMIN-NEXT:    lui a2, 524288
 ; RV32IZFHMIN-NEXT:    beqz s0, .LBB21_4
 ; RV32IZFHMIN-NEXT:  # %bb.3:
-; RV32IZFHMIN-NEXT:    mv a4, a1
+; RV32IZFHMIN-NEXT:    mv a2, a1
 ; RV32IZFHMIN-NEXT:  .LBB21_4:
+; RV32IZFHMIN-NEXT:    lui a1, %hi(.LCPI21_0)
+; RV32IZFHMIN-NEXT:    flw fa5, %lo(.LCPI21_0)(a1)
+; RV32IZFHMIN-NEXT:    flt.s a3, fa5, fs0
+; RV32IZFHMIN-NEXT:    beqz a3, .LBB21_6
+; RV32IZFHMIN-NEXT:  # %bb.5:
+; RV32IZFHMIN-NEXT:    addi a2, a4, -1
+; RV32IZFHMIN-NEXT:  .LBB21_6:
+; RV32IZFHMIN-NEXT:    feq.s a1, fs0, fs0
+; RV32IZFHMIN-NEXT:    neg a4, a1
+; RV32IZFHMIN-NEXT:    and a1, a4, a2
+; RV32IZFHMIN-NEXT:    neg a2, s0
 ; RV32IZFHMIN-NEXT:    and a0, a2, a0
+; RV32IZFHMIN-NEXT:    neg a2, a3
+; RV32IZFHMIN-NEXT:    or a0, a2, a0
+; RV32IZFHMIN-NEXT:    and a0, a4, a0
 ; RV32IZFHMIN-NEXT:    lw ra, 12(sp) # 4-byte Folded Reload
 ; RV32IZFHMIN-NEXT:    lw s0, 8(sp) # 4-byte Folded Reload
-; RV32IZFHMIN-NEXT:    lw s1, 4(sp) # 4-byte Folded Reload
-; RV32IZFHMIN-NEXT:    flw fs0, 0(sp) # 4-byte Folded Reload
+; RV32IZFHMIN-NEXT:    flw fs0, 4(sp) # 4-byte Folded Reload
 ; RV32IZFHMIN-NEXT:    addi sp, sp, 16
-; RV32IZFHMIN-NEXT:    beqz a3, .LBB21_6
-; RV32IZFHMIN-NEXT:  # %bb.5:
-; RV32IZFHMIN-NEXT:    addi a4, a5, -1
-; RV32IZFHMIN-NEXT:  .LBB21_6:
-; RV32IZFHMIN-NEXT:    and a1, a2, a4
 ; RV32IZFHMIN-NEXT:    ret
 ;
 ; RV64IZFHMIN-LABEL: test_rint_si64:
@@ -3904,16 +3880,16 @@ define i64 @test_rint_si64(half %x) nounwind {
 ; RV32IZHINXMIN-NEXT:    lui a2, %hi(.LCPI21_0)
 ; RV32IZHINXMIN-NEXT:    lw a2, %lo(.LCPI21_0)(a2)
 ; RV32IZHINXMIN-NEXT:    and a0, s2, a0
-; RV32IZHINXMIN-NEXT:    flt.s a3, a2, s0
-; RV32IZHINXMIN-NEXT:    neg a2, a3
+; RV32IZHINXMIN-NEXT:    flt.s a4, a2, s0
+; RV32IZHINXMIN-NEXT:    neg a2, a4
 ; RV32IZHINXMIN-NEXT:    or a0, a2, a0
 ; RV32IZHINXMIN-NEXT:    feq.s a2, s0, s0
 ; RV32IZHINXMIN-NEXT:    neg a2, a2
 ; RV32IZHINXMIN-NEXT:    lui a5, 524288
-; RV32IZHINXMIN-NEXT:    lui a4, 524288
+; RV32IZHINXMIN-NEXT:    lui a3, 524288
 ; RV32IZHINXMIN-NEXT:    beqz s1, .LBB21_4
 ; RV32IZHINXMIN-NEXT:  # %bb.3:
-; RV32IZHINXMIN-NEXT:    mv a4, a1
+; RV32IZHINXMIN-NEXT:    mv a3, a1
 ; RV32IZHINXMIN-NEXT:  .LBB21_4:
 ; RV32IZHINXMIN-NEXT:    and a0, a2, a0
 ; RV32IZHINXMIN-NEXT:    lw ra, 12(sp) # 4-byte Folded Reload
@@ -3921,11 +3897,11 @@ define i64 @test_rint_si64(half %x) nounwind {
 ; RV32IZHINXMIN-NEXT:    lw s1, 4(sp) # 4-byte Folded Reload
 ; RV32IZHINXMIN-NEXT:    lw s2, 0(sp) # 4-byte Folded Reload
 ; RV32IZHINXMIN-NEXT:    addi sp, sp, 16
-; RV32IZHINXMIN-NEXT:    beqz a3, .LBB21_6
+; RV32IZHINXMIN-NEXT:    beqz a4, .LBB21_6
 ; RV32IZHINXMIN-NEXT:  # %bb.5:
-; RV32IZHINXMIN-NEXT:    addi a4, a5, -1
+; RV32IZHINXMIN-NEXT:    addi a3, a5, -1
 ; RV32IZHINXMIN-NEXT:  .LBB21_6:
-; RV32IZHINXMIN-NEXT:    and a1, a2, a4
+; RV32IZHINXMIN-NEXT:    and a1, a2, a3
 ; RV32IZHINXMIN-NEXT:    ret
 ;
 ; RV64IZHINXMIN-LABEL: test_rint_si64:
diff --git a/llvm/test/CodeGen/RISCV/iabs.ll b/llvm/test/CodeGen/RISCV/iabs.ll
index 98c886333d69..a0c85ab4dca7 100644
--- a/llvm/test/CodeGen/RISCV/iabs.ll
+++ b/llvm/test/CodeGen/RISCV/iabs.ll
@@ -630,8 +630,8 @@ define void @zext16_abs8(i8 %x, ptr %p) {
 ; RV32I-LABEL: zext16_abs8:
 ; RV32I:       # %bb.0:
 ; RV32I-NEXT:    slli a0, a0, 24
-; RV32I-NEXT:    srai a0, a0, 24
 ; RV32I-NEXT:    srai a2, a0, 31
+; RV32I-NEXT:    srai a0, a0, 24
 ; RV32I-NEXT:    xor a0, a0, a2
 ; RV32I-NEXT:    sub a0, a0, a2
 ; RV32I-NEXT:    sh a0, 0(a1)
@@ -648,8 +648,8 @@ define void @zext16_abs8(i8 %x, ptr %p) {
 ; RV64I-LABEL: zext16_abs8:
 ; RV64I:       # %bb.0:
 ; RV64I-NEXT:    slli a0, a0, 56
-; RV64I-NEXT:    srai a0, a0, 56
 ; RV64I-NEXT:    srai a2, a0, 63
+; RV64I-NEXT:    srai a0, a0, 56
 ; RV64I-NEXT:    xor a0, a0, a2
 ; RV64I-NEXT:    subw a0, a0, a2
 ; RV64I-NEXT:    sh a0, 0(a1)
diff --git a/llvm/test/CodeGen/RISCV/pr84653_pr85190.ll b/llvm/test/CodeGen/RISCV/pr84653_pr85190.ll
new file mode 100644
index 000000000000..3fa494e1a57d
--- /dev/null
+++ b/llvm/test/CodeGen/RISCV/pr84653_pr85190.ll
@@ -0,0 +1,95 @@
+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --version 4
+; RUN: llc < %s -mtriple=riscv64 | FileCheck %s --check-prefixes=CHECK-NOZBB
+; RUN: llc < %s -mtriple=riscv64 -mattr=+zbb | FileCheck %s --check-prefixes=CHECK-ZBB
+
+; This test case miscompiled for ZBB (DAGCombiner turned a SELECT into a more
+; poisonous AND operation).
+define i1 @pr84653(i32 %x) {
+; CHECK-NOZBB-LABEL: pr84653:
+; CHECK-NOZBB:       # %bb.0:
+; CHECK-NOZBB-NEXT:    sext.w a1, a0
+; CHECK-NOZBB-NEXT:    sgtz a2, a1
+; CHECK-NOZBB-NEXT:    lui a3, 524288
+; CHECK-NOZBB-NEXT:    addi a3, a3, -1
+; CHECK-NOZBB-NEXT:    xor a0, a0, a3
+; CHECK-NOZBB-NEXT:    sext.w a0, a0
+; CHECK-NOZBB-NEXT:    slt a0, a0, a1
+; CHECK-NOZBB-NEXT:    and a0, a2, a0
+; CHECK-NOZBB-NEXT:    ret
+;
+; CHECK-ZBB-LABEL: pr84653:
+; CHECK-ZBB:       # %bb.0:
+; CHECK-ZBB-NEXT:    sext.w a1, a0
+; CHECK-ZBB-NEXT:    lui a2, 524288
+; CHECK-ZBB-NEXT:    addi a2, a2, -1
+; CHECK-ZBB-NEXT:    xor a0, a0, a2
+; CHECK-ZBB-NEXT:    sext.w a0, a0
+; CHECK-ZBB-NEXT:    max a0, a0, zero
+; CHECK-ZBB-NEXT:    slt a0, a0, a1
+; CHECK-ZBB-NEXT:    ret
+  %cmp1 = icmp sgt i32 %x, 0
+  %sub = sub nsw i32 2147483647, %x  ; 0x7fffffff
+  %cmp2 = icmp sgt i32 %x, %sub
+  %r = select i1 %cmp1, i1 %cmp2, i1 false
+  ret i1 %r
+}
+
+; This test case miscompiled for ZBB (DAGCombiner turned a SELECT into a more
+; poisonous AND operation).
+define i1 @pr85190(i64 %a) {
+; CHECK-NOZBB-LABEL: pr85190:
+; CHECK-NOZBB:       # %bb.0:
+; CHECK-NOZBB-NEXT:    ori a1, a0, 7
+; CHECK-NOZBB-NEXT:    slti a2, a0, 0
+; CHECK-NOZBB-NEXT:    li a3, -1
+; CHECK-NOZBB-NEXT:    slli a3, a3, 63
+; CHECK-NOZBB-NEXT:    sub a3, a3, a1
+; CHECK-NOZBB-NEXT:    slt a0, a0, a3
+; CHECK-NOZBB-NEXT:    and a0, a2, a0
+; CHECK-NOZBB-NEXT:    ret
+;
+; CHECK-ZBB-LABEL: pr85190:
+; CHECK-ZBB:       # %bb.0:
+; CHECK-ZBB-NEXT:    ori a1, a0, 7
+; CHECK-ZBB-NEXT:    li a2, -1
+; CHECK-ZBB-NEXT:    slli a2, a2, 63
+; CHECK-ZBB-NEXT:    sub a2, a2, a1
+; CHECK-ZBB-NEXT:    slt a0, a0, a2
+; CHECK-ZBB-NEXT:    ret
+  %or = or i64 %a, 7
+  %cmp1 = icmp slt i64 %a, 0
+  %sub = sub nsw i64 -9223372036854775808, %or  ; 0x8000000000000000
+  %cmp2 = icmp sgt i64 %sub, %a
+  %res = select i1 %cmp1, i1 %cmp2, i1 false
+  ret i1 %res
+}
+
+define i1 @select_to_or(i32 %x) {
+; CHECK-NOZBB-LABEL: select_to_or:
+; CHECK-NOZBB:       # %bb.0:
+; CHECK-NOZBB-NEXT:    sext.w a1, a0
+; CHECK-NOZBB-NEXT:    sgtz a2, a1
+; CHECK-NOZBB-NEXT:    lui a3, 524288
+; CHECK-NOZBB-NEXT:    addi a3, a3, -1
+; CHECK-NOZBB-NEXT:    xor a0, a0, a3
+; CHECK-NOZBB-NEXT:    sext.w a0, a0
+; CHECK-NOZBB-NEXT:    slt a0, a0, a1
+; CHECK-NOZBB-NEXT:    or a0, a2, a0
+; CHECK-NOZBB-NEXT:    ret
+;
+; CHECK-ZBB-LABEL: select_to_or:
+; CHECK-ZBB:       # %bb.0:
+; CHECK-ZBB-NEXT:    sext.w a1, a0
+; CHECK-ZBB-NEXT:    lui a2, 524288
+; CHECK-ZBB-NEXT:    addi a2, a2, -1
+; CHECK-ZBB-NEXT:    xor a0, a0, a2
+; CHECK-ZBB-NEXT:    sext.w a0, a0
+; CHECK-ZBB-NEXT:    min a0, a0, zero
+; CHECK-ZBB-NEXT:    slt a0, a0, a1
+; CHECK-ZBB-NEXT:    ret
+  %cmp1 = icmp sgt i32 %x, 0
+  %sub = sub nsw i32 2147483647, %x  ; 0x7fffffff
+  %cmp2 = icmp sgt i32 %x, %sub
+  %r = select i1 %cmp1, i1 true, i1 %cmp2
+  ret i1 %r
+}
diff --git a/llvm/test/CodeGen/RISCV/pr89833.ll b/llvm/test/CodeGen/RISCV/pr89833.ll
new file mode 100644
index 000000000000..54a985040e75
--- /dev/null
+++ b/llvm/test/CodeGen/RISCV/pr89833.ll
@@ -0,0 +1,16 @@
+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
+; RUN: llc < %s -mtriple=riscv64 -mattr=+v | FileCheck %s
+
+declare void @llvm.riscv.masked.strided.store.nxv16i8.p0.i64(<vscale x 16 x i8>, ptr, i64, <vscale x 16 x i1>)
+
+define void @test(<vscale x 16 x i16> %value, <vscale x 16 x i1> %mask) {
+; CHECK-LABEL: test:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli a0, zero, e8, m2, ta, ma
+; CHECK-NEXT:    vnsrl.wi v12, v8, 0
+; CHECK-NEXT:    vse8.v v12, (zero), v0.t
+; CHECK-NEXT:    ret
+  %trunc = trunc <vscale x 16 x i16> %value to <vscale x 16 x i8>
+  call void @llvm.riscv.masked.strided.store.nxv16i8.p0.i64(<vscale x 16 x i8> %trunc, ptr null, i64 1, <vscale x 16 x i1> %mask)
+  ret void
+}
diff --git a/llvm/test/CodeGen/RISCV/rv32zba.ll b/llvm/test/CodeGen/RISCV/rv32zba.ll
index a78f823d3184..2a72c1288f65 100644
--- a/llvm/test/CodeGen/RISCV/rv32zba.ll
+++ b/llvm/test/CodeGen/RISCV/rv32zba.ll
@@ -407,8 +407,8 @@ define i32 @mul25(i32 %a) {
 ;
 ; RV32ZBA-LABEL: mul25:
 ; RV32ZBA:       # %bb.0:
-; RV32ZBA-NEXT:    sh1add a1, a0, a0
-; RV32ZBA-NEXT:    sh3add a0, a1, a0
+; RV32ZBA-NEXT:    sh2add a0, a0, a0
+; RV32ZBA-NEXT:    sh2add a0, a0, a0
 ; RV32ZBA-NEXT:    ret
   %c = mul i32 %a, 25
   ret i32 %c
@@ -455,8 +455,8 @@ define i32 @mul27(i32 %a) {
 ;
 ; RV32ZBA-LABEL: mul27:
 ; RV32ZBA:       # %bb.0:
-; RV32ZBA-NEXT:    sh3add a0, a0, a0
 ; RV32ZBA-NEXT:    sh1add a0, a0, a0
+; RV32ZBA-NEXT:    sh3add a0, a0, a0
 ; RV32ZBA-NEXT:    ret
   %c = mul i32 %a, 27
   ret i32 %c
@@ -471,8 +471,8 @@ define i32 @mul45(i32 %a) {
 ;
 ; RV32ZBA-LABEL: mul45:
 ; RV32ZBA:       # %bb.0:
-; RV32ZBA-NEXT:    sh3add a0, a0, a0
 ; RV32ZBA-NEXT:    sh2add a0, a0, a0
+; RV32ZBA-NEXT:    sh3add a0, a0, a0
 ; RV32ZBA-NEXT:    ret
   %c = mul i32 %a, 45
   ret i32 %c
diff --git a/llvm/test/CodeGen/RISCV/rv64-legal-i32/rv64zba.ll b/llvm/test/CodeGen/RISCV/rv64-legal-i32/rv64zba.ll
index ee9b73ca82f2..c3ae40124ba0 100644
--- a/llvm/test/CodeGen/RISCV/rv64-legal-i32/rv64zba.ll
+++ b/llvm/test/CodeGen/RISCV/rv64-legal-i32/rv64zba.ll
@@ -634,7 +634,6 @@ define i64 @zext_mul288(i32 signext %a) {
 }
 
 ; We can't use slli.uw becaues the shift amount is more than 31.
-; FIXME: The zext.w is unneeded.
 define i64 @zext_mul12884901888(i32 signext %a) {
 ; RV64I-LABEL: zext_mul12884901888:
 ; RV64I:       # %bb.0:
@@ -647,7 +646,6 @@ define i64 @zext_mul12884901888(i32 signext %a) {
 ;
 ; RV64ZBA-LABEL: zext_mul12884901888:
 ; RV64ZBA:       # %bb.0:
-; RV64ZBA-NEXT:    andi a0, a0, -1
 ; RV64ZBA-NEXT:    sh1add a0, a0, a0
 ; RV64ZBA-NEXT:    slli a0, a0, 32
 ; RV64ZBA-NEXT:    ret
@@ -657,7 +655,6 @@ define i64 @zext_mul12884901888(i32 signext %a) {
 }
 
 ; We can't use slli.uw becaues the shift amount is more than 31.
-; FIXME: The zext.w is unneeded.
 define i64 @zext_mul21474836480(i32 signext %a) {
 ; RV64I-LABEL: zext_mul21474836480:
 ; RV64I:       # %bb.0:
@@ -670,7 +667,6 @@ define i64 @zext_mul21474836480(i32 signext %a) {
 ;
 ; RV64ZBA-LABEL: zext_mul21474836480:
 ; RV64ZBA:       # %bb.0:
-; RV64ZBA-NEXT:    andi a0, a0, -1
 ; RV64ZBA-NEXT:    sh2add a0, a0, a0
 ; RV64ZBA-NEXT:    slli a0, a0, 32
 ; RV64ZBA-NEXT:    ret
@@ -680,7 +676,6 @@ define i64 @zext_mul21474836480(i32 signext %a) {
 }
 
 ; We can't use slli.uw becaues the shift amount is more than 31.
-; FIXME: The zext.w is unneeded.
 define i64 @zext_mul38654705664(i32 signext %a) {
 ; RV64I-LABEL: zext_mul38654705664:
 ; RV64I:       # %bb.0:
@@ -693,7 +688,6 @@ define i64 @zext_mul38654705664(i32 signext %a) {
 ;
 ; RV64ZBA-LABEL: zext_mul38654705664:
 ; RV64ZBA:       # %bb.0:
-; RV64ZBA-NEXT:    andi a0, a0, -1
 ; RV64ZBA-NEXT:    sh3add a0, a0, a0
 ; RV64ZBA-NEXT:    slli a0, a0, 32
 ; RV64ZBA-NEXT:    ret
@@ -963,8 +957,8 @@ define i64 @mul25(i64 %a) {
 ;
 ; RV64ZBA-LABEL: mul25:
 ; RV64ZBA:       # %bb.0:
-; RV64ZBA-NEXT:    sh1add a1, a0, a0
-; RV64ZBA-NEXT:    sh3add a0, a1, a0
+; RV64ZBA-NEXT:    sh2add a0, a0, a0
+; RV64ZBA-NEXT:    sh2add a0, a0, a0
 ; RV64ZBA-NEXT:    ret
   %c = mul i64 %a, 25
   ret i64 %c
@@ -1011,8 +1005,8 @@ define i64 @mul27(i64 %a) {
 ;
 ; RV64ZBA-LABEL: mul27:
 ; RV64ZBA:       # %bb.0:
-; RV64ZBA-NEXT:    sh3add a0, a0, a0
 ; RV64ZBA-NEXT:    sh1add a0, a0, a0
+; RV64ZBA-NEXT:    sh3add a0, a0, a0
 ; RV64ZBA-NEXT:    ret
   %c = mul i64 %a, 27
   ret i64 %c
@@ -1027,8 +1021,8 @@ define i64 @mul45(i64 %a) {
 ;
 ; RV64ZBA-LABEL: mul45:
 ; RV64ZBA:       # %bb.0:
-; RV64ZBA-NEXT:    sh3add a0, a0, a0
 ; RV64ZBA-NEXT:    sh2add a0, a0, a0
+; RV64ZBA-NEXT:    sh3add a0, a0, a0
 ; RV64ZBA-NEXT:    ret
   %c = mul i64 %a, 45
   ret i64 %c
diff --git a/llvm/test/CodeGen/RISCV/rv64-legal-i32/rv64zbb-zbkb.ll b/llvm/test/CodeGen/RISCV/rv64-legal-i32/rv64zbb-zbkb.ll
index 39a5b9b0f367..c98ad4592a66 100644
--- a/llvm/test/CodeGen/RISCV/rv64-legal-i32/rv64zbb-zbkb.ll
+++ b/llvm/test/CodeGen/RISCV/rv64-legal-i32/rv64zbb-zbkb.ll
@@ -102,11 +102,10 @@ declare i32 @llvm.fshl.i32(i32, i32, i32)
 define signext i32 @rol_i32(i32 signext %a, i32 signext %b) nounwind {
 ; RV64I-LABEL: rol_i32:
 ; RV64I:       # %bb.0:
-; RV64I-NEXT:    andi a2, a1, -1
-; RV64I-NEXT:    sllw a1, a0, a1
-; RV64I-NEXT:    negw a2, a2
-; RV64I-NEXT:    srlw a0, a0, a2
-; RV64I-NEXT:    or a0, a1, a0
+; RV64I-NEXT:    sllw a2, a0, a1
+; RV64I-NEXT:    negw a1, a1
+; RV64I-NEXT:    srlw a0, a0, a1
+; RV64I-NEXT:    or a0, a2, a0
 ; RV64I-NEXT:    ret
 ;
 ; RV64ZBB-ZBKB-LABEL: rol_i32:
@@ -121,11 +120,10 @@ define signext i32 @rol_i32(i32 signext %a, i32 signext %b) nounwind {
 define void @rol_i32_nosext(i32 signext %a, i32 signext %b, ptr %x) nounwind {
 ; RV64I-LABEL: rol_i32_nosext:
 ; RV64I:       # %bb.0:
-; RV64I-NEXT:    andi a3, a1, -1
-; RV64I-NEXT:    sllw a1, a0, a1
-; RV64I-NEXT:    negw a3, a3
-; RV64I-NEXT:    srlw a0, a0, a3
-; RV64I-NEXT:    or a0, a1, a0
+; RV64I-NEXT:    sllw a3, a0, a1
+; RV64I-NEXT:    negw a1, a1
+; RV64I-NEXT:    srlw a0, a0, a1
+; RV64I-NEXT:    or a0, a3, a0
 ; RV64I-NEXT:    sw a0, 0(a2)
 ; RV64I-NEXT:    ret
 ;
@@ -142,12 +140,11 @@ define void @rol_i32_nosext(i32 signext %a, i32 signext %b, ptr %x) nounwind {
 define signext i32 @rol_i32_neg_constant_rhs(i32 signext %a) nounwind {
 ; RV64I-LABEL: rol_i32_neg_constant_rhs:
 ; RV64I:       # %bb.0:
-; RV64I-NEXT:    andi a1, a0, -1
-; RV64I-NEXT:    li a2, -2
-; RV64I-NEXT:    sllw a0, a2, a0
-; RV64I-NEXT:    negw a1, a1
-; RV64I-NEXT:    srlw a1, a2, a1
-; RV64I-NEXT:    or a0, a0, a1
+; RV64I-NEXT:    li a1, -2
+; RV64I-NEXT:    sllw a2, a1, a0
+; RV64I-NEXT:    negw a0, a0
+; RV64I-NEXT:    srlw a0, a1, a0
+; RV64I-NEXT:    or a0, a2, a0
 ; RV64I-NEXT:    ret
 ;
 ; RV64ZBB-ZBKB-LABEL: rol_i32_neg_constant_rhs:
@@ -183,11 +180,10 @@ declare i32 @llvm.fshr.i32(i32, i32, i32)
 define signext i32 @ror_i32(i32 signext %a, i32 signext %b) nounwind {
 ; RV64I-LABEL: ror_i32:
 ; RV64I:       # %bb.0:
-; RV64I-NEXT:    andi a2, a1, -1
-; RV64I-NEXT:    srlw a1, a0, a1
-; RV64I-NEXT:    negw a2, a2
-; RV64I-NEXT:    sllw a0, a0, a2
-; RV64I-NEXT:    or a0, a1, a0
+; RV64I-NEXT:    srlw a2, a0, a1
+; RV64I-NEXT:    negw a1, a1
+; RV64I-NEXT:    sllw a0, a0, a1
+; RV64I-NEXT:    or a0, a2, a0
 ; RV64I-NEXT:    ret
 ;
 ; RV64ZBB-ZBKB-LABEL: ror_i32:
@@ -202,11 +198,10 @@ define signext i32 @ror_i32(i32 signext %a, i32 signext %b) nounwind {
 define void @ror_i32_nosext(i32 signext %a, i32 signext %b, ptr %x) nounwind {
 ; RV64I-LABEL: ror_i32_nosext:
 ; RV64I:       # %bb.0:
-; RV64I-NEXT:    andi a3, a1, -1
-; RV64I-NEXT:    srlw a1, a0, a1
-; RV64I-NEXT:    negw a3, a3
-; RV64I-NEXT:    sllw a0, a0, a3
-; RV64I-NEXT:    or a0, a1, a0
+; RV64I-NEXT:    srlw a3, a0, a1
+; RV64I-NEXT:    negw a1, a1
+; RV64I-NEXT:    sllw a0, a0, a1
+; RV64I-NEXT:    or a0, a3, a0
 ; RV64I-NEXT:    sw a0, 0(a2)
 ; RV64I-NEXT:    ret
 ;
@@ -223,12 +218,11 @@ define void @ror_i32_nosext(i32 signext %a, i32 signext %b, ptr %x) nounwind {
 define signext i32 @ror_i32_neg_constant_rhs(i32 signext %a) nounwind {
 ; RV64I-LABEL: ror_i32_neg_constant_rhs:
 ; RV64I:       # %bb.0:
-; RV64I-NEXT:    andi a1, a0, -1
-; RV64I-NEXT:    li a2, -2
-; RV64I-NEXT:    srlw a0, a2, a0
-; RV64I-NEXT:    negw a1, a1
-; RV64I-NEXT:    sllw a1, a2, a1
-; RV64I-NEXT:    or a0, a0, a1
+; RV64I-NEXT:    li a1, -2
+; RV64I-NEXT:    srlw a2, a1, a0
+; RV64I-NEXT:    negw a0, a0
+; RV64I-NEXT:    sllw a0, a1, a0
+; RV64I-NEXT:    or a0, a2, a0
 ; RV64I-NEXT:    ret
 ;
 ; RV64ZBB-ZBKB-LABEL: ror_i32_neg_constant_rhs:
diff --git a/llvm/test/CodeGen/RISCV/rv64-legal-i32/xaluo.ll b/llvm/test/CodeGen/RISCV/rv64-legal-i32/xaluo.ll
index 3c1b76818781..1c794a1bd168 100644
--- a/llvm/test/CodeGen/RISCV/rv64-legal-i32/xaluo.ll
+++ b/llvm/test/CodeGen/RISCV/rv64-legal-i32/xaluo.ll
@@ -731,12 +731,13 @@ define zeroext i1 @smulo2.i64(i64 %v1, ptr %res) {
 ; RV64ZBA-LABEL: smulo2.i64:
 ; RV64ZBA:       # %bb.0: # %entry
 ; RV64ZBA-NEXT:    li a2, 13
-; RV64ZBA-NEXT:    mulh a3, a0, a2
-; RV64ZBA-NEXT:    mul a2, a0, a2
-; RV64ZBA-NEXT:    srai a0, a2, 63
-; RV64ZBA-NEXT:    xor a0, a3, a0
+; RV64ZBA-NEXT:    mulh a2, a0, a2
+; RV64ZBA-NEXT:    sh1add a3, a0, a0
+; RV64ZBA-NEXT:    sh2add a3, a3, a0
+; RV64ZBA-NEXT:    srai a0, a3, 63
+; RV64ZBA-NEXT:    xor a0, a2, a0
 ; RV64ZBA-NEXT:    snez a0, a0
-; RV64ZBA-NEXT:    sd a2, 0(a1)
+; RV64ZBA-NEXT:    sd a3, 0(a1)
 ; RV64ZBA-NEXT:    ret
 ;
 ; RV64ZICOND-LABEL: smulo2.i64:
@@ -809,9 +810,9 @@ define zeroext i1 @umulo2.i32(i32 signext %v1, ptr %res) {
 ;
 ; RV64ZBA-LABEL: umulo2.i32:
 ; RV64ZBA:       # %bb.0: # %entry
-; RV64ZBA-NEXT:    zext.w a0, a0
-; RV64ZBA-NEXT:    sh1add a2, a0, a0
-; RV64ZBA-NEXT:    sh2add a2, a2, a0
+; RV64ZBA-NEXT:    zext.w a2, a0
+; RV64ZBA-NEXT:    sh1add.uw a0, a0, a2
+; RV64ZBA-NEXT:    sh2add a2, a0, a2
 ; RV64ZBA-NEXT:    srli a0, a2, 32
 ; RV64ZBA-NEXT:    snez a0, a0
 ; RV64ZBA-NEXT:    sw a2, 0(a1)
@@ -925,10 +926,11 @@ define zeroext i1 @umulo2.i64(i64 %v1, ptr %res) {
 ;
 ; RV64ZBA-LABEL: umulo2.i64:
 ; RV64ZBA:       # %bb.0: # %entry
-; RV64ZBA-NEXT:    li a3, 13
-; RV64ZBA-NEXT:    mulhu a2, a0, a3
+; RV64ZBA-NEXT:    li a2, 13
+; RV64ZBA-NEXT:    mulhu a2, a0, a2
 ; RV64ZBA-NEXT:    snez a2, a2
-; RV64ZBA-NEXT:    mul a0, a0, a3
+; RV64ZBA-NEXT:    sh1add a3, a0, a0
+; RV64ZBA-NEXT:    sh2add a0, a3, a0
 ; RV64ZBA-NEXT:    sd a0, 0(a1)
 ; RV64ZBA-NEXT:    mv a0, a2
 ; RV64ZBA-NEXT:    ret
diff --git a/llvm/test/CodeGen/RISCV/rv64i-demanded-bits.ll b/llvm/test/CodeGen/RISCV/rv64i-demanded-bits.ll
index 6a441e2b9f67..b8c43289bdfe 100644
--- a/llvm/test/CodeGen/RISCV/rv64i-demanded-bits.ll
+++ b/llvm/test/CodeGen/RISCV/rv64i-demanded-bits.ll
@@ -192,3 +192,17 @@ entry:
   %or = or i32 %and, 255
   ret i32 %or
 }
+
+define i64 @and_allones(i32 signext %x) {
+; CHECK-LABEL: and_allones:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    addi a0, a0, -1
+; CHECK-NEXT:    li a1, 1
+; CHECK-NEXT:    sll a0, a1, a0
+; CHECK-NEXT:    ret
+entry:
+  %y = zext i32 %x to i64
+  %shamt = add nsw i64 %y, -1
+  %ret = shl i64 1, %shamt
+  ret i64 %ret
+}
diff --git a/llvm/test/CodeGen/RISCV/rv64xtheadba.ll b/llvm/test/CodeGen/RISCV/rv64xtheadba.ll
index 1450c86c76d0..7754128370d6 100644
--- a/llvm/test/CodeGen/RISCV/rv64xtheadba.ll
+++ b/llvm/test/CodeGen/RISCV/rv64xtheadba.ll
@@ -252,6 +252,183 @@ define i64 @addmul72(i64 %a, i64 %b) {
   ret i64 %d
 }
 
+define i64 @mul11(i64 %a) {
+; RV64I-LABEL: mul11:
+; RV64I:       # %bb.0:
+; RV64I-NEXT:    li a1, 11
+; RV64I-NEXT:    mul a0, a0, a1
+; RV64I-NEXT:    ret
+;
+; RV64XTHEADBA-LABEL: mul11:
+; RV64XTHEADBA:       # %bb.0:
+; RV64XTHEADBA-NEXT:    th.addsl a1, a0, a0, 2
+; RV64XTHEADBA-NEXT:    th.addsl a0, a0, a1, 1
+; RV64XTHEADBA-NEXT:    ret
+  %c = mul i64 %a, 11
+  ret i64 %c
+}
+
+define i64 @mul19(i64 %a) {
+; RV64I-LABEL: mul19:
+; RV64I:       # %bb.0:
+; RV64I-NEXT:    li a1, 19
+; RV64I-NEXT:    mul a0, a0, a1
+; RV64I-NEXT:    ret
+;
+; RV64XTHEADBA-LABEL: mul19:
+; RV64XTHEADBA:       # %bb.0:
+; RV64XTHEADBA-NEXT:    th.addsl a1, a0, a0, 3
+; RV64XTHEADBA-NEXT:    th.addsl a0, a0, a1, 1
+; RV64XTHEADBA-NEXT:    ret
+  %c = mul i64 %a, 19
+  ret i64 %c
+}
+
+define i64 @mul13(i64 %a) {
+; RV64I-LABEL: mul13:
+; RV64I:       # %bb.0:
+; RV64I-NEXT:    li a1, 13
+; RV64I-NEXT:    mul a0, a0, a1
+; RV64I-NEXT:    ret
+;
+; RV64XTHEADBA-LABEL: mul13:
+; RV64XTHEADBA:       # %bb.0:
+; RV64XTHEADBA-NEXT:    th.addsl a1, a0, a0, 1
+; RV64XTHEADBA-NEXT:    th.addsl a0, a0, a1, 2
+; RV64XTHEADBA-NEXT:    ret
+  %c = mul i64 %a, 13
+  ret i64 %c
+}
+
+define i64 @mul21(i64 %a) {
+; RV64I-LABEL: mul21:
+; RV64I:       # %bb.0:
+; RV64I-NEXT:    li a1, 21
+; RV64I-NEXT:    mul a0, a0, a1
+; RV64I-NEXT:    ret
+;
+; RV64XTHEADBA-LABEL: mul21:
+; RV64XTHEADBA:       # %bb.0:
+; RV64XTHEADBA-NEXT:    th.addsl a1, a0, a0, 2
+; RV64XTHEADBA-NEXT:    th.addsl a0, a0, a1, 2
+; RV64XTHEADBA-NEXT:    ret
+  %c = mul i64 %a, 21
+  ret i64 %c
+}
+
+define i64 @mul37(i64 %a) {
+; RV64I-LABEL: mul37:
+; RV64I:       # %bb.0:
+; RV64I-NEXT:    li a1, 37
+; RV64I-NEXT:    mul a0, a0, a1
+; RV64I-NEXT:    ret
+;
+; RV64XTHEADBA-LABEL: mul37:
+; RV64XTHEADBA:       # %bb.0:
+; RV64XTHEADBA-NEXT:    th.addsl a1, a0, a0, 3
+; RV64XTHEADBA-NEXT:    th.addsl a0, a0, a1, 2
+; RV64XTHEADBA-NEXT:    ret
+  %c = mul i64 %a, 37
+  ret i64 %c
+}
+
+define i64 @mul25(i64 %a) {
+; RV64I-LABEL: mul25:
+; RV64I:       # %bb.0:
+; RV64I-NEXT:    li a1, 25
+; RV64I-NEXT:    mul a0, a0, a1
+; RV64I-NEXT:    ret
+;
+; RV64XTHEADBA-LABEL: mul25:
+; RV64XTHEADBA:       # %bb.0:
+; RV64XTHEADBA-NEXT:    th.addsl a0, a0, a0, 2
+; RV64XTHEADBA-NEXT:    th.addsl a0, a0, a0, 2
+; RV64XTHEADBA-NEXT:    ret
+  %c = mul i64 %a, 25
+  ret i64 %c
+}
+
+define i64 @mul41(i64 %a) {
+; RV64I-LABEL: mul41:
+; RV64I:       # %bb.0:
+; RV64I-NEXT:    li a1, 41
+; RV64I-NEXT:    mul a0, a0, a1
+; RV64I-NEXT:    ret
+;
+; RV64XTHEADBA-LABEL: mul41:
+; RV64XTHEADBA:       # %bb.0:
+; RV64XTHEADBA-NEXT:    th.addsl a1, a0, a0, 2
+; RV64XTHEADBA-NEXT:    th.addsl a0, a0, a1, 3
+; RV64XTHEADBA-NEXT:    ret
+  %c = mul i64 %a, 41
+  ret i64 %c
+}
+
+define i64 @mul73(i64 %a) {
+; RV64I-LABEL: mul73:
+; RV64I:       # %bb.0:
+; RV64I-NEXT:    li a1, 73
+; RV64I-NEXT:    mul a0, a0, a1
+; RV64I-NEXT:    ret
+;
+; RV64XTHEADBA-LABEL: mul73:
+; RV64XTHEADBA:       # %bb.0:
+; RV64XTHEADBA-NEXT:    th.addsl a1, a0, a0, 3
+; RV64XTHEADBA-NEXT:    th.addsl a0, a0, a1, 3
+; RV64XTHEADBA-NEXT:    ret
+  %c = mul i64 %a, 73
+  ret i64 %c
+}
+
+define i64 @mul27(i64 %a) {
+; RV64I-LABEL: mul27:
+; RV64I:       # %bb.0:
+; RV64I-NEXT:    li a1, 27
+; RV64I-NEXT:    mul a0, a0, a1
+; RV64I-NEXT:    ret
+;
+; RV64XTHEADBA-LABEL: mul27:
+; RV64XTHEADBA:       # %bb.0:
+; RV64XTHEADBA-NEXT:    th.addsl a0, a0, a0, 1
+; RV64XTHEADBA-NEXT:    th.addsl a0, a0, a0, 3
+; RV64XTHEADBA-NEXT:    ret
+  %c = mul i64 %a, 27
+  ret i64 %c
+}
+
+define i64 @mul45(i64 %a) {
+; RV64I-LABEL: mul45:
+; RV64I:       # %bb.0:
+; RV64I-NEXT:    li a1, 45
+; RV64I-NEXT:    mul a0, a0, a1
+; RV64I-NEXT:    ret
+;
+; RV64XTHEADBA-LABEL: mul45:
+; RV64XTHEADBA:       # %bb.0:
+; RV64XTHEADBA-NEXT:    th.addsl a0, a0, a0, 2
+; RV64XTHEADBA-NEXT:    th.addsl a0, a0, a0, 3
+; RV64XTHEADBA-NEXT:    ret
+  %c = mul i64 %a, 45
+  ret i64 %c
+}
+
+define i64 @mul81(i64 %a) {
+; RV64I-LABEL: mul81:
+; RV64I:       # %bb.0:
+; RV64I-NEXT:    li a1, 81
+; RV64I-NEXT:    mul a0, a0, a1
+; RV64I-NEXT:    ret
+;
+; RV64XTHEADBA-LABEL: mul81:
+; RV64XTHEADBA:       # %bb.0:
+; RV64XTHEADBA-NEXT:    th.addsl a0, a0, a0, 3
+; RV64XTHEADBA-NEXT:    th.addsl a0, a0, a0, 3
+; RV64XTHEADBA-NEXT:    ret
+  %c = mul i64 %a, 81
+  ret i64 %c
+}
+
+
 define i64 @mul96(i64 %a) {
 ; RV64I-LABEL: mul96:
 ; RV64I:       # %bb.0:
diff --git a/llvm/test/CodeGen/RISCV/rv64zba.ll b/llvm/test/CodeGen/RISCV/rv64zba.ll
index ccb23fc2bbfa..817e2b7d0bd9 100644
--- a/llvm/test/CodeGen/RISCV/rv64zba.ll
+++ b/llvm/test/CodeGen/RISCV/rv64zba.ll
@@ -4,7 +4,9 @@
 ; RUN: llc -mtriple=riscv64 -mattr=+m,+zba -verify-machineinstrs < %s \
 ; RUN:   | FileCheck %s -check-prefixes=CHECK,RV64ZBA,RV64ZBANOZBB
 ; RUN: llc -mtriple=riscv64 -mattr=+m,+zba,+zbb -verify-machineinstrs < %s \
-; RUN:   | FileCheck %s -check-prefixes=CHECK,RV64ZBA,RV64ZBAZBB
+; RUN:   | FileCheck %s -check-prefixes=CHECK,RV64ZBA,RV64ZBAZBB,RV64ZBAZBBNOZBS
+; RUN: llc -mtriple=riscv64 -mattr=+m,+zba,+zbb,+zbs -verify-machineinstrs < %s \
+; RUN:   | FileCheck %s -check-prefixes=CHECK,RV64ZBA,RV64ZBAZBB,RV64ZBAZBBZBS
 
 define i64 @slliuw(i64 %a) nounwind {
 ; RV64I-LABEL: slliuw:
@@ -641,31 +643,52 @@ define i64 @mul96(i64 %a) {
 }
 
 define i64 @mul119(i64 %a) {
-; CHECK-LABEL: mul119:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    li a1, 119
-; CHECK-NEXT:    mul a0, a0, a1
-; CHECK-NEXT:    ret
+; RV64I-LABEL: mul119:
+; RV64I:       # %bb.0:
+; RV64I-NEXT:    li a1, 119
+; RV64I-NEXT:    mul a0, a0, a1
+; RV64I-NEXT:    ret
+;
+; RV64ZBA-LABEL: mul119:
+; RV64ZBA:       # %bb.0:
+; RV64ZBA-NEXT:    sh3add a1, a0, a0
+; RV64ZBA-NEXT:    slli a0, a0, 7
+; RV64ZBA-NEXT:    sub a0, a0, a1
+; RV64ZBA-NEXT:    ret
   %c = mul i64 %a, 119
   ret i64 %c
 }
 
 define i64 @mul123(i64 %a) {
-; CHECK-LABEL: mul123:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    li a1, 123
-; CHECK-NEXT:    mul a0, a0, a1
-; CHECK-NEXT:    ret
+; RV64I-LABEL: mul123:
+; RV64I:       # %bb.0:
+; RV64I-NEXT:    li a1, 123
+; RV64I-NEXT:    mul a0, a0, a1
+; RV64I-NEXT:    ret
+;
+; RV64ZBA-LABEL: mul123:
+; RV64ZBA:       # %bb.0:
+; RV64ZBA-NEXT:    sh2add a1, a0, a0
+; RV64ZBA-NEXT:    slli a0, a0, 7
+; RV64ZBA-NEXT:    sub a0, a0, a1
+; RV64ZBA-NEXT:    ret
   %c = mul i64 %a, 123
   ret i64 %c
 }
 
 define i64 @mul125(i64 %a) {
-; CHECK-LABEL: mul125:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    li a1, 125
-; CHECK-NEXT:    mul a0, a0, a1
-; CHECK-NEXT:    ret
+; RV64I-LABEL: mul125:
+; RV64I:       # %bb.0:
+; RV64I-NEXT:    li a1, 125
+; RV64I-NEXT:    mul a0, a0, a1
+; RV64I-NEXT:    ret
+;
+; RV64ZBA-LABEL: mul125:
+; RV64ZBA:       # %bb.0:
+; RV64ZBA-NEXT:    sh1add a1, a0, a0
+; RV64ZBA-NEXT:    slli a0, a0, 7
+; RV64ZBA-NEXT:    sub a0, a0, a1
+; RV64ZBA-NEXT:    ret
   %c = mul i64 %a, 125
   ret i64 %c
 }
@@ -753,6 +776,25 @@ define i64 @mul288(i64 %a) {
   ret i64 %c
 }
 
+define i64 @zext_mul68(i32 signext %a) {
+; RV64I-LABEL: zext_mul68:
+; RV64I:       # %bb.0:
+; RV64I-NEXT:    li a1, 17
+; RV64I-NEXT:    slli a1, a1, 34
+; RV64I-NEXT:    slli a0, a0, 32
+; RV64I-NEXT:    mulhu a0, a0, a1
+; RV64I-NEXT:    ret
+;
+; RV64ZBA-LABEL: zext_mul68:
+; RV64ZBA:       # %bb.0:
+; RV64ZBA-NEXT:    slli.uw a1, a0, 6
+; RV64ZBA-NEXT:    sh2add.uw a0, a0, a1
+; RV64ZBA-NEXT:    ret
+  %b = zext i32 %a to i64
+  %c = mul i64 %b, 68
+  ret i64 %c
+}
+
 define i64 @zext_mul96(i32 signext %a) {
 ; RV64I-LABEL: zext_mul96:
 ; RV64I:       # %bb.0:
@@ -811,7 +853,6 @@ define i64 @zext_mul288(i32 signext %a) {
 }
 
 ; We can't use slli.uw becaues the shift amount is more than 31.
-; FIXME: The zext.w is unneeded.
 define i64 @zext_mul12884901888(i32 signext %a) {
 ; RV64I-LABEL: zext_mul12884901888:
 ; RV64I:       # %bb.0:
@@ -824,7 +865,6 @@ define i64 @zext_mul12884901888(i32 signext %a) {
 ;
 ; RV64ZBA-LABEL: zext_mul12884901888:
 ; RV64ZBA:       # %bb.0:
-; RV64ZBA-NEXT:    andi a0, a0, -1
 ; RV64ZBA-NEXT:    sh1add a0, a0, a0
 ; RV64ZBA-NEXT:    slli a0, a0, 32
 ; RV64ZBA-NEXT:    ret
@@ -834,7 +874,6 @@ define i64 @zext_mul12884901888(i32 signext %a) {
 }
 
 ; We can't use slli.uw becaues the shift amount is more than 31.
-; FIXME: The zext.w is unneeded.
 define i64 @zext_mul21474836480(i32 signext %a) {
 ; RV64I-LABEL: zext_mul21474836480:
 ; RV64I:       # %bb.0:
@@ -847,7 +886,6 @@ define i64 @zext_mul21474836480(i32 signext %a) {
 ;
 ; RV64ZBA-LABEL: zext_mul21474836480:
 ; RV64ZBA:       # %bb.0:
-; RV64ZBA-NEXT:    andi a0, a0, -1
 ; RV64ZBA-NEXT:    sh2add a0, a0, a0
 ; RV64ZBA-NEXT:    slli a0, a0, 32
 ; RV64ZBA-NEXT:    ret
@@ -857,7 +895,6 @@ define i64 @zext_mul21474836480(i32 signext %a) {
 }
 
 ; We can't use slli.uw becaues the shift amount is more than 31.
-; FIXME: The zext.w is unneeded.
 define i64 @zext_mul38654705664(i32 signext %a) {
 ; RV64I-LABEL: zext_mul38654705664:
 ; RV64I:       # %bb.0:
@@ -870,7 +907,6 @@ define i64 @zext_mul38654705664(i32 signext %a) {
 ;
 ; RV64ZBA-LABEL: zext_mul38654705664:
 ; RV64ZBA:       # %bb.0:
-; RV64ZBA-NEXT:    andi a0, a0, -1
 ; RV64ZBA-NEXT:    sh3add a0, a0, a0
 ; RV64ZBA-NEXT:    slli a0, a0, 32
 ; RV64ZBA-NEXT:    ret
@@ -1140,8 +1176,8 @@ define i64 @mul25(i64 %a) {
 ;
 ; RV64ZBA-LABEL: mul25:
 ; RV64ZBA:       # %bb.0:
-; RV64ZBA-NEXT:    sh1add a1, a0, a0
-; RV64ZBA-NEXT:    sh3add a0, a1, a0
+; RV64ZBA-NEXT:    sh2add a0, a0, a0
+; RV64ZBA-NEXT:    sh2add a0, a0, a0
 ; RV64ZBA-NEXT:    ret
   %c = mul i64 %a, 25
   ret i64 %c
@@ -1188,8 +1224,8 @@ define i64 @mul27(i64 %a) {
 ;
 ; RV64ZBA-LABEL: mul27:
 ; RV64ZBA:       # %bb.0:
-; RV64ZBA-NEXT:    sh3add a0, a0, a0
 ; RV64ZBA-NEXT:    sh1add a0, a0, a0
+; RV64ZBA-NEXT:    sh3add a0, a0, a0
 ; RV64ZBA-NEXT:    ret
   %c = mul i64 %a, 27
   ret i64 %c
@@ -1204,8 +1240,8 @@ define i64 @mul45(i64 %a) {
 ;
 ; RV64ZBA-LABEL: mul45:
 ; RV64ZBA:       # %bb.0:
-; RV64ZBA-NEXT:    sh3add a0, a0, a0
 ; RV64ZBA-NEXT:    sh2add a0, a0, a0
+; RV64ZBA-NEXT:    sh3add a0, a0, a0
 ; RV64ZBA-NEXT:    ret
   %c = mul i64 %a, 45
   ret i64 %c
@@ -2699,3 +2735,121 @@ define i64 @mul_neg8(i64 %a) {
   %c = mul i64 %a, -8
   ret i64 %c
 }
+
+define i64 @bext_mul12(i32 %1, i32 %2) {
+; RV64I-LABEL: bext_mul12:
+; RV64I:       # %bb.0: # %entry
+; RV64I-NEXT:    srlw a0, a0, a1
+; RV64I-NEXT:    andi a0, a0, 1
+; RV64I-NEXT:    li a1, 12
+; RV64I-NEXT:    mul a0, a0, a1
+; RV64I-NEXT:    ret
+;
+; RV64ZBANOZBB-LABEL: bext_mul12:
+; RV64ZBANOZBB:       # %bb.0: # %entry
+; RV64ZBANOZBB-NEXT:    srlw a0, a0, a1
+; RV64ZBANOZBB-NEXT:    andi a0, a0, 1
+; RV64ZBANOZBB-NEXT:    sh1add a0, a0, a0
+; RV64ZBANOZBB-NEXT:    slli a0, a0, 2
+; RV64ZBANOZBB-NEXT:    ret
+;
+; RV64ZBAZBBNOZBS-LABEL: bext_mul12:
+; RV64ZBAZBBNOZBS:       # %bb.0: # %entry
+; RV64ZBAZBBNOZBS-NEXT:    srlw a0, a0, a1
+; RV64ZBAZBBNOZBS-NEXT:    andi a0, a0, 1
+; RV64ZBAZBBNOZBS-NEXT:    sh1add a0, a0, a0
+; RV64ZBAZBBNOZBS-NEXT:    slli a0, a0, 2
+; RV64ZBAZBBNOZBS-NEXT:    ret
+;
+; RV64ZBAZBBZBS-LABEL: bext_mul12:
+; RV64ZBAZBBZBS:       # %bb.0: # %entry
+; RV64ZBAZBBZBS-NEXT:    bext a0, a0, a1
+; RV64ZBAZBBZBS-NEXT:    sh1add a0, a0, a0
+; RV64ZBAZBBZBS-NEXT:    slli a0, a0, 2
+; RV64ZBAZBBZBS-NEXT:    ret
+entry:
+  %3 = lshr i32 %1, %2
+  %4 = and i32 %3, 1
+  %5 = zext nneg i32 %4 to i64
+  %6 = mul i64 %5, 12
+  ret i64 %6
+}
+
+define i64 @bext_mul45(i32 %1, i32 %2) {
+; RV64I-LABEL: bext_mul45:
+; RV64I:       # %bb.0: # %entry
+; RV64I-NEXT:    srlw a0, a0, a1
+; RV64I-NEXT:    andi a0, a0, 1
+; RV64I-NEXT:    li a1, 45
+; RV64I-NEXT:    mul a0, a0, a1
+; RV64I-NEXT:    ret
+;
+; RV64ZBANOZBB-LABEL: bext_mul45:
+; RV64ZBANOZBB:       # %bb.0: # %entry
+; RV64ZBANOZBB-NEXT:    srlw a0, a0, a1
+; RV64ZBANOZBB-NEXT:    andi a0, a0, 1
+; RV64ZBANOZBB-NEXT:    sh2add a0, a0, a0
+; RV64ZBANOZBB-NEXT:    sh3add a0, a0, a0
+; RV64ZBANOZBB-NEXT:    ret
+;
+; RV64ZBAZBBNOZBS-LABEL: bext_mul45:
+; RV64ZBAZBBNOZBS:       # %bb.0: # %entry
+; RV64ZBAZBBNOZBS-NEXT:    srlw a0, a0, a1
+; RV64ZBAZBBNOZBS-NEXT:    andi a0, a0, 1
+; RV64ZBAZBBNOZBS-NEXT:    sh2add a0, a0, a0
+; RV64ZBAZBBNOZBS-NEXT:    sh3add a0, a0, a0
+; RV64ZBAZBBNOZBS-NEXT:    ret
+;
+; RV64ZBAZBBZBS-LABEL: bext_mul45:
+; RV64ZBAZBBZBS:       # %bb.0: # %entry
+; RV64ZBAZBBZBS-NEXT:    bext a0, a0, a1
+; RV64ZBAZBBZBS-NEXT:    sh2add a0, a0, a0
+; RV64ZBAZBBZBS-NEXT:    sh3add a0, a0, a0
+; RV64ZBAZBBZBS-NEXT:    ret
+entry:
+  %3 = lshr i32 %1, %2
+  %4 = and i32 %3, 1
+  %5 = zext nneg i32 %4 to i64
+  %6 = mul i64 %5, 45
+  ret i64 %6
+}
+
+define i64 @bext_mul132(i32 %1, i32 %2) {
+; RV64I-LABEL: bext_mul132:
+; RV64I:       # %bb.0: # %entry
+; RV64I-NEXT:    srlw a0, a0, a1
+; RV64I-NEXT:    andi a0, a0, 1
+; RV64I-NEXT:    li a1, 132
+; RV64I-NEXT:    mul a0, a0, a1
+; RV64I-NEXT:    ret
+;
+; RV64ZBANOZBB-LABEL: bext_mul132:
+; RV64ZBANOZBB:       # %bb.0: # %entry
+; RV64ZBANOZBB-NEXT:    srlw a0, a0, a1
+; RV64ZBANOZBB-NEXT:    andi a0, a0, 1
+; RV64ZBANOZBB-NEXT:    slli a1, a0, 7
+; RV64ZBANOZBB-NEXT:    sh2add a0, a0, a1
+; RV64ZBANOZBB-NEXT:    ret
+;
+; RV64ZBAZBBNOZBS-LABEL: bext_mul132:
+; RV64ZBAZBBNOZBS:       # %bb.0: # %entry
+; RV64ZBAZBBNOZBS-NEXT:    srlw a0, a0, a1
+; RV64ZBAZBBNOZBS-NEXT:    andi a0, a0, 1
+; RV64ZBAZBBNOZBS-NEXT:    slli a1, a0, 7
+; RV64ZBAZBBNOZBS-NEXT:    sh2add a0, a0, a1
+; RV64ZBAZBBNOZBS-NEXT:    ret
+;
+; RV64ZBAZBBZBS-LABEL: bext_mul132:
+; RV64ZBAZBBZBS:       # %bb.0: # %entry
+; RV64ZBAZBBZBS-NEXT:    bext a0, a0, a1
+; RV64ZBAZBBZBS-NEXT:    slli a1, a0, 7
+; RV64ZBAZBBZBS-NEXT:    sh2add a0, a0, a1
+; RV64ZBAZBBZBS-NEXT:    ret
+entry:
+  %3 = lshr i32 %1, %2
+  %4 = and i32 %3, 1
+  %5 = zext nneg i32 %4 to i64
+  %6 = mul i64 %5, 132
+  ret i64 %6
+}
+
diff --git a/llvm/test/CodeGen/RISCV/rvv/commutable.ll b/llvm/test/CodeGen/RISCV/rvv/commutable.ll
new file mode 100644
index 000000000000..06a6327d3892
--- /dev/null
+++ b/llvm/test/CodeGen/RISCV/rvv/commutable.ll
@@ -0,0 +1,824 @@
+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
+; RUN: sed 's/iXLen/i32/g' %s | llc -mtriple=riscv32 -mattr=+f,+d,+zvfh,+v \
+; RUN:   -verify-machineinstrs | FileCheck %s
+; RUN: sed 's/iXLen/i64/g' %s | llc -mtriple=riscv64 -mattr=+f,+d,+zvfh,+v \
+; RUN:   -verify-machineinstrs | FileCheck %s
+
+; vadd.vv
+declare <vscale x 1 x i64> @llvm.riscv.vadd.nxv1i64.nxv1i64(<vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, iXLen);
+define <vscale x 1 x i64> @commutable_vadd_vv(<vscale x 1 x i64> %0, <vscale x 1 x i64> %1, iXLen %2) nounwind {
+; CHECK-LABEL: commutable_vadd_vv:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
+; CHECK-NEXT:    vadd.vv v8, v8, v9
+; CHECK-NEXT:    vsetvli a0, zero, e64, m1, ta, ma
+; CHECK-NEXT:    vadd.vv v8, v8, v8
+; CHECK-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vadd.nxv1i64.nxv1i64(<vscale x 1 x i64> undef, <vscale x 1 x i64> %0, <vscale x 1 x i64> %1, iXLen %2)
+  %b = call <vscale x 1 x i64> @llvm.riscv.vadd.nxv1i64.nxv1i64(<vscale x 1 x i64> undef, <vscale x 1 x i64> %1, <vscale x 1 x i64> %0, iXLen %2)
+  %ret = add <vscale x 1 x i64> %a, %b
+  ret <vscale x 1 x i64> %ret
+}
+
+declare <vscale x 1 x i64> @llvm.riscv.vadd.mask.nxv1i64.nxv1i64(<vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i1>, iXLen, iXLen);
+define <vscale x 1 x i64> @commutable_vadd_vv_masked(<vscale x 1 x i64> %0, <vscale x 1 x i64> %1, <vscale x 1 x i1> %mask, iXLen %2) {
+; CHECK-LABEL: commutable_vadd_vv_masked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
+; CHECK-NEXT:    vadd.vv v10, v8, v9, v0.t
+; CHECK-NEXT:    vadd.vv v8, v8, v9, v0.t
+; CHECK-NEXT:    vsetvli a0, zero, e64, m1, ta, ma
+; CHECK-NEXT:    vadd.vv v8, v10, v8
+; CHECK-NEXT:    ret
+  %a = call <vscale x 1 x i64> @llvm.riscv.vadd.mask.nxv1i64.nxv1i64(<vscale x 1 x i64> undef, <vscale x 1 x i64> %0, <vscale x 1 x i64> %1, <vscale x 1 x i1> %mask, iXLen %2, iXLen 1)
+  %b = call <vscale x 1 x i64> @llvm.riscv.vadd.mask.nxv1i64.nxv1i64(<vscale x 1 x i64> undef, <vscale x 1 x i64> %1, <vscale x 1 x i64> %0, <vscale x 1 x i1> %mask, iXLen %2, iXLen 1)
+  %ret = add <vscale x 1 x i64> %a, %b
+  ret <vscale x 1 x i64> %ret
+}
+
+; vand.vv
+declare <vscale x 1 x i64> @llvm.riscv.vand.nxv1i64.nxv1i64(<vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, iXLen);
+define <vscale x 1 x i64> @commutable_vand_vv(<vscale x 1 x i64> %0, <vscale x 1 x i64> %1, iXLen %2) nounwind {
+; CHECK-LABEL: commutable_vand_vv:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
+; CHECK-NEXT:    vand.vv v8, v8, v9
+; CHECK-NEXT:    vsetvli a0, zero, e64, m1, ta, ma
+; CHECK-NEXT:    vadd.vv v8, v8, v8
+; CHECK-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vand.nxv1i64.nxv1i64(<vscale x 1 x i64> undef, <vscale x 1 x i64> %0, <vscale x 1 x i64> %1, iXLen %2)
+  %b = call <vscale x 1 x i64> @llvm.riscv.vand.nxv1i64.nxv1i64(<vscale x 1 x i64> undef, <vscale x 1 x i64> %1, <vscale x 1 x i64> %0, iXLen %2)
+  %ret = add <vscale x 1 x i64> %a, %b
+  ret <vscale x 1 x i64> %ret
+}
+
+declare <vscale x 1 x i64> @llvm.riscv.vand.mask.nxv1i64.nxv1i64(<vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i1>, iXLen, iXLen);
+define <vscale x 1 x i64> @commutable_vand_vv_masked(<vscale x 1 x i64> %0, <vscale x 1 x i64> %1, <vscale x 1 x i1> %mask, iXLen %2) {
+; CHECK-LABEL: commutable_vand_vv_masked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
+; CHECK-NEXT:    vand.vv v10, v8, v9, v0.t
+; CHECK-NEXT:    vand.vv v8, v8, v9, v0.t
+; CHECK-NEXT:    vsetvli a0, zero, e64, m1, ta, ma
+; CHECK-NEXT:    vadd.vv v8, v10, v8
+; CHECK-NEXT:    ret
+  %a = call <vscale x 1 x i64> @llvm.riscv.vand.mask.nxv1i64.nxv1i64(<vscale x 1 x i64> undef, <vscale x 1 x i64> %0, <vscale x 1 x i64> %1, <vscale x 1 x i1> %mask, iXLen %2, iXLen 1)
+  %b = call <vscale x 1 x i64> @llvm.riscv.vand.mask.nxv1i64.nxv1i64(<vscale x 1 x i64> undef, <vscale x 1 x i64> %1, <vscale x 1 x i64> %0, <vscale x 1 x i1> %mask, iXLen %2, iXLen 1)
+  %ret = add <vscale x 1 x i64> %a, %b
+  ret <vscale x 1 x i64> %ret
+}
+
+; vor.vv
+declare <vscale x 1 x i64> @llvm.riscv.vor.nxv1i64.nxv1i64(<vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, iXLen);
+define <vscale x 1 x i64> @commutable_vor_vv(<vscale x 1 x i64> %0, <vscale x 1 x i64> %1, iXLen %2) nounwind {
+; CHECK-LABEL: commutable_vor_vv:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
+; CHECK-NEXT:    vor.vv v8, v8, v9
+; CHECK-NEXT:    vsetvli a0, zero, e64, m1, ta, ma
+; CHECK-NEXT:    vadd.vv v8, v8, v8
+; CHECK-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vor.nxv1i64.nxv1i64(<vscale x 1 x i64> undef, <vscale x 1 x i64> %0, <vscale x 1 x i64> %1, iXLen %2)
+  %b = call <vscale x 1 x i64> @llvm.riscv.vor.nxv1i64.nxv1i64(<vscale x 1 x i64> undef, <vscale x 1 x i64> %1, <vscale x 1 x i64> %0, iXLen %2)
+  %ret = add <vscale x 1 x i64> %a, %b
+  ret <vscale x 1 x i64> %ret
+}
+
+declare <vscale x 1 x i64> @llvm.riscv.vor.mask.nxv1i64.nxv1i64(<vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i1>, iXLen, iXLen);
+define <vscale x 1 x i64> @commutable_vor_vv_masked(<vscale x 1 x i64> %0, <vscale x 1 x i64> %1, <vscale x 1 x i1> %mask, iXLen %2) {
+; CHECK-LABEL: commutable_vor_vv_masked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
+; CHECK-NEXT:    vor.vv v10, v8, v9, v0.t
+; CHECK-NEXT:    vor.vv v8, v8, v9, v0.t
+; CHECK-NEXT:    vsetvli a0, zero, e64, m1, ta, ma
+; CHECK-NEXT:    vadd.vv v8, v10, v8
+; CHECK-NEXT:    ret
+  %a = call <vscale x 1 x i64> @llvm.riscv.vor.mask.nxv1i64.nxv1i64(<vscale x 1 x i64> undef, <vscale x 1 x i64> %0, <vscale x 1 x i64> %1, <vscale x 1 x i1> %mask, iXLen %2, iXLen 1)
+  %b = call <vscale x 1 x i64> @llvm.riscv.vor.mask.nxv1i64.nxv1i64(<vscale x 1 x i64> undef, <vscale x 1 x i64> %1, <vscale x 1 x i64> %0, <vscale x 1 x i1> %mask, iXLen %2, iXLen 1)
+  %ret = add <vscale x 1 x i64> %a, %b
+  ret <vscale x 1 x i64> %ret
+}
+
+; vxor.vv
+declare <vscale x 1 x i64> @llvm.riscv.vxor.nxv1i64.nxv1i64(<vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, iXLen);
+define <vscale x 1 x i64> @commutable_vxor_vv(<vscale x 1 x i64> %0, <vscale x 1 x i64> %1, iXLen %2) nounwind {
+; CHECK-LABEL: commutable_vxor_vv:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
+; CHECK-NEXT:    vxor.vv v8, v8, v9
+; CHECK-NEXT:    vsetvli a0, zero, e64, m1, ta, ma
+; CHECK-NEXT:    vadd.vv v8, v8, v8
+; CHECK-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vxor.nxv1i64.nxv1i64(<vscale x 1 x i64> undef, <vscale x 1 x i64> %0, <vscale x 1 x i64> %1, iXLen %2)
+  %b = call <vscale x 1 x i64> @llvm.riscv.vxor.nxv1i64.nxv1i64(<vscale x 1 x i64> undef, <vscale x 1 x i64> %1, <vscale x 1 x i64> %0, iXLen %2)
+  %ret = add <vscale x 1 x i64> %a, %b
+  ret <vscale x 1 x i64> %ret
+}
+
+declare <vscale x 1 x i64> @llvm.riscv.vxor.mask.nxv1i64.nxv1i64(<vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i1>, iXLen, iXLen);
+define <vscale x 1 x i64> @commutable_vxor_vv_masked(<vscale x 1 x i64> %0, <vscale x 1 x i64> %1, <vscale x 1 x i1> %mask, iXLen %2) {
+; CHECK-LABEL: commutable_vxor_vv_masked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
+; CHECK-NEXT:    vxor.vv v10, v8, v9, v0.t
+; CHECK-NEXT:    vxor.vv v8, v8, v9, v0.t
+; CHECK-NEXT:    vsetvli a0, zero, e64, m1, ta, ma
+; CHECK-NEXT:    vadd.vv v8, v10, v8
+; CHECK-NEXT:    ret
+  %a = call <vscale x 1 x i64> @llvm.riscv.vxor.mask.nxv1i64.nxv1i64(<vscale x 1 x i64> undef, <vscale x 1 x i64> %0, <vscale x 1 x i64> %1, <vscale x 1 x i1> %mask, iXLen %2, iXLen 1)
+  %b = call <vscale x 1 x i64> @llvm.riscv.vxor.mask.nxv1i64.nxv1i64(<vscale x 1 x i64> undef, <vscale x 1 x i64> %1, <vscale x 1 x i64> %0, <vscale x 1 x i1> %mask, iXLen %2, iXLen 1)
+  %ret = add <vscale x 1 x i64> %a, %b
+  ret <vscale x 1 x i64> %ret
+}
+
+; vmseq.vv
+declare <vscale x 1 x i1> @llvm.riscv.vmseq.nxv1i64(<vscale x 1 x i64>, <vscale x 1 x i64>, iXLen);
+define <vscale x 1 x i1> @commutable_vmseq_vv(<vscale x 1 x i64> %0, <vscale x 1 x i64> %1, iXLen %2) nounwind {
+; CHECK-LABEL: commutable_vmseq_vv:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
+; CHECK-NEXT:    vmseq.vv v8, v8, v9
+; CHECK-NEXT:    vsetvli a0, zero, e8, mf8, ta, ma
+; CHECK-NEXT:    vmxor.mm v0, v8, v8
+; CHECK-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i1> @llvm.riscv.vmseq.nxv1i64(<vscale x 1 x i64> %0, <vscale x 1 x i64> %1, iXLen %2)
+  %b = call <vscale x 1 x i1> @llvm.riscv.vmseq.nxv1i64(<vscale x 1 x i64> %1, <vscale x 1 x i64> %0, iXLen %2)
+  %ret = add <vscale x 1 x i1> %a, %b
+  ret <vscale x 1 x i1> %ret
+}
+
+declare <vscale x 1 x i1> @llvm.riscv.vmseq.mask.nxv1i64(<vscale x 1 x i1>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i1>, iXLen);
+define <vscale x 1 x i1> @commutable_vmseq_vv_masked(<vscale x 1 x i64> %0, <vscale x 1 x i64> %1, <vscale x 1 x i1> %mask, iXLen %2) {
+; CHECK-LABEL: commutable_vmseq_vv_masked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
+; CHECK-NEXT:    vmseq.vv v10, v8, v9, v0.t
+; CHECK-NEXT:    vmseq.vv v8, v8, v9, v0.t
+; CHECK-NEXT:    vsetvli a0, zero, e8, mf8, ta, ma
+; CHECK-NEXT:    vmxor.mm v0, v10, v8
+; CHECK-NEXT:    ret
+  %a = call <vscale x 1 x i1> @llvm.riscv.vmseq.mask.nxv1i64(<vscale x 1 x i1> undef, <vscale x 1 x i64> %0, <vscale x 1 x i64> %1, <vscale x 1 x i1> %mask, iXLen %2)
+  %b = call <vscale x 1 x i1> @llvm.riscv.vmseq.mask.nxv1i64(<vscale x 1 x i1> undef, <vscale x 1 x i64> %1, <vscale x 1 x i64> %0, <vscale x 1 x i1> %mask, iXLen %2)
+  %ret = add <vscale x 1 x i1> %a, %b
+  ret <vscale x 1 x i1> %ret
+}
+
+; vmsne.vv
+declare <vscale x 1 x i1> @llvm.riscv.vmsne.nxv1i64(<vscale x 1 x i64>, <vscale x 1 x i64>, iXLen);
+define <vscale x 1 x i1> @commutable_vmsne_vv(<vscale x 1 x i64> %0, <vscale x 1 x i64> %1, iXLen %2) nounwind {
+; CHECK-LABEL: commutable_vmsne_vv:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
+; CHECK-NEXT:    vmsne.vv v8, v8, v9
+; CHECK-NEXT:    vsetvli a0, zero, e8, mf8, ta, ma
+; CHECK-NEXT:    vmxor.mm v0, v8, v8
+; CHECK-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i1> @llvm.riscv.vmsne.nxv1i64(<vscale x 1 x i64> %0, <vscale x 1 x i64> %1, iXLen %2)
+  %b = call <vscale x 1 x i1> @llvm.riscv.vmsne.nxv1i64(<vscale x 1 x i64> %1, <vscale x 1 x i64> %0, iXLen %2)
+  %ret = add <vscale x 1 x i1> %a, %b
+  ret <vscale x 1 x i1> %ret
+}
+
+declare <vscale x 1 x i1> @llvm.riscv.vmsne.mask.nxv1i64(<vscale x 1 x i1>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i1>, iXLen);
+define <vscale x 1 x i1> @commutable_vmsne_vv_masked(<vscale x 1 x i64> %0, <vscale x 1 x i64> %1, <vscale x 1 x i1> %mask, iXLen %2) {
+; CHECK-LABEL: commutable_vmsne_vv_masked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
+; CHECK-NEXT:    vmsne.vv v10, v8, v9, v0.t
+; CHECK-NEXT:    vmsne.vv v8, v8, v9, v0.t
+; CHECK-NEXT:    vsetvli a0, zero, e8, mf8, ta, ma
+; CHECK-NEXT:    vmxor.mm v0, v10, v8
+; CHECK-NEXT:    ret
+  %a = call <vscale x 1 x i1> @llvm.riscv.vmsne.mask.nxv1i64(<vscale x 1 x i1> undef, <vscale x 1 x i64> %0, <vscale x 1 x i64> %1, <vscale x 1 x i1> %mask, iXLen %2)
+  %b = call <vscale x 1 x i1> @llvm.riscv.vmsne.mask.nxv1i64(<vscale x 1 x i1> undef, <vscale x 1 x i64> %1, <vscale x 1 x i64> %0, <vscale x 1 x i1> %mask, iXLen %2)
+  %ret = add <vscale x 1 x i1> %a, %b
+  ret <vscale x 1 x i1> %ret
+}
+
+; vmin.vv
+declare <vscale x 1 x i64> @llvm.riscv.vmin.nxv1i64.nxv1i64(<vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, iXLen);
+define <vscale x 1 x i64> @commutable_vmin_vv(<vscale x 1 x i64> %0, <vscale x 1 x i64> %1, iXLen %2) nounwind {
+; CHECK-LABEL: commutable_vmin_vv:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
+; CHECK-NEXT:    vmin.vv v8, v8, v9
+; CHECK-NEXT:    vsetvli a0, zero, e64, m1, ta, ma
+; CHECK-NEXT:    vadd.vv v8, v8, v8
+; CHECK-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vmin.nxv1i64.nxv1i64(<vscale x 1 x i64> undef, <vscale x 1 x i64> %0, <vscale x 1 x i64> %1, iXLen %2)
+  %b = call <vscale x 1 x i64> @llvm.riscv.vmin.nxv1i64.nxv1i64(<vscale x 1 x i64> undef, <vscale x 1 x i64> %1, <vscale x 1 x i64> %0, iXLen %2)
+  %ret = add <vscale x 1 x i64> %a, %b
+  ret <vscale x 1 x i64> %ret
+}
+
+declare <vscale x 1 x i64> @llvm.riscv.vmin.mask.nxv1i64.nxv1i64(<vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i1>, iXLen, iXLen);
+define <vscale x 1 x i64> @commutable_vmin_vv_masked(<vscale x 1 x i64> %0, <vscale x 1 x i64> %1, <vscale x 1 x i1> %mask, iXLen %2) {
+; CHECK-LABEL: commutable_vmin_vv_masked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
+; CHECK-NEXT:    vmin.vv v10, v8, v9, v0.t
+; CHECK-NEXT:    vmin.vv v8, v8, v9, v0.t
+; CHECK-NEXT:    vsetvli a0, zero, e64, m1, ta, ma
+; CHECK-NEXT:    vadd.vv v8, v10, v8
+; CHECK-NEXT:    ret
+  %a = call <vscale x 1 x i64> @llvm.riscv.vmin.mask.nxv1i64.nxv1i64(<vscale x 1 x i64> undef, <vscale x 1 x i64> %0, <vscale x 1 x i64> %1, <vscale x 1 x i1> %mask, iXLen %2, iXLen 1)
+  %b = call <vscale x 1 x i64> @llvm.riscv.vmin.mask.nxv1i64.nxv1i64(<vscale x 1 x i64> undef, <vscale x 1 x i64> %1, <vscale x 1 x i64> %0, <vscale x 1 x i1> %mask, iXLen %2, iXLen 1)
+  %ret = add <vscale x 1 x i64> %a, %b
+  ret <vscale x 1 x i64> %ret
+}
+
+; vminu.vv
+declare <vscale x 1 x i64> @llvm.riscv.vminu.nxv1i64.nxv1i64(<vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, iXLen);
+define <vscale x 1 x i64> @commutable_vminu_vv(<vscale x 1 x i64> %0, <vscale x 1 x i64> %1, iXLen %2) nounwind {
+; CHECK-LABEL: commutable_vminu_vv:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
+; CHECK-NEXT:    vminu.vv v8, v8, v9
+; CHECK-NEXT:    vsetvli a0, zero, e64, m1, ta, ma
+; CHECK-NEXT:    vadd.vv v8, v8, v8
+; CHECK-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vminu.nxv1i64.nxv1i64(<vscale x 1 x i64> undef, <vscale x 1 x i64> %0, <vscale x 1 x i64> %1, iXLen %2)
+  %b = call <vscale x 1 x i64> @llvm.riscv.vminu.nxv1i64.nxv1i64(<vscale x 1 x i64> undef, <vscale x 1 x i64> %1, <vscale x 1 x i64> %0, iXLen %2)
+  %ret = add <vscale x 1 x i64> %a, %b
+  ret <vscale x 1 x i64> %ret
+}
+
+declare <vscale x 1 x i64> @llvm.riscv.vminu.mask.nxv1i64.nxv1i64(<vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i1>, iXLen, iXLen);
+define <vscale x 1 x i64> @commutable_vminu_vv_masked(<vscale x 1 x i64> %0, <vscale x 1 x i64> %1, <vscale x 1 x i1> %mask, iXLen %2) {
+; CHECK-LABEL: commutable_vminu_vv_masked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
+; CHECK-NEXT:    vminu.vv v10, v8, v9, v0.t
+; CHECK-NEXT:    vminu.vv v8, v8, v9, v0.t
+; CHECK-NEXT:    vsetvli a0, zero, e64, m1, ta, ma
+; CHECK-NEXT:    vadd.vv v8, v10, v8
+; CHECK-NEXT:    ret
+  %a = call <vscale x 1 x i64> @llvm.riscv.vminu.mask.nxv1i64.nxv1i64(<vscale x 1 x i64> undef, <vscale x 1 x i64> %0, <vscale x 1 x i64> %1, <vscale x 1 x i1> %mask, iXLen %2, iXLen 1)
+  %b = call <vscale x 1 x i64> @llvm.riscv.vminu.mask.nxv1i64.nxv1i64(<vscale x 1 x i64> undef, <vscale x 1 x i64> %1, <vscale x 1 x i64> %0, <vscale x 1 x i1> %mask, iXLen %2, iXLen 1)
+  %ret = add <vscale x 1 x i64> %a, %b
+  ret <vscale x 1 x i64> %ret
+}
+
+; vmax.vv
+declare <vscale x 1 x i64> @llvm.riscv.vmax.nxv1i64.nxv1i64(<vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, iXLen);
+define <vscale x 1 x i64> @commutable_vmax_vv(<vscale x 1 x i64> %0, <vscale x 1 x i64> %1, iXLen %2) nounwind {
+; CHECK-LABEL: commutable_vmax_vv:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
+; CHECK-NEXT:    vmax.vv v8, v8, v9
+; CHECK-NEXT:    vsetvli a0, zero, e64, m1, ta, ma
+; CHECK-NEXT:    vadd.vv v8, v8, v8
+; CHECK-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vmax.nxv1i64.nxv1i64(<vscale x 1 x i64> undef, <vscale x 1 x i64> %0, <vscale x 1 x i64> %1, iXLen %2)
+  %b = call <vscale x 1 x i64> @llvm.riscv.vmax.nxv1i64.nxv1i64(<vscale x 1 x i64> undef, <vscale x 1 x i64> %1, <vscale x 1 x i64> %0, iXLen %2)
+  %ret = add <vscale x 1 x i64> %a, %b
+  ret <vscale x 1 x i64> %ret
+}
+
+declare <vscale x 1 x i64> @llvm.riscv.vmax.mask.nxv1i64.nxv1i64(<vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i1>, iXLen, iXLen);
+define <vscale x 1 x i64> @commutable_vmax_vv_masked(<vscale x 1 x i64> %0, <vscale x 1 x i64> %1, <vscale x 1 x i1> %mask, iXLen %2) {
+; CHECK-LABEL: commutable_vmax_vv_masked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
+; CHECK-NEXT:    vmax.vv v10, v8, v9, v0.t
+; CHECK-NEXT:    vmax.vv v8, v8, v9, v0.t
+; CHECK-NEXT:    vsetvli a0, zero, e64, m1, ta, ma
+; CHECK-NEXT:    vadd.vv v8, v10, v8
+; CHECK-NEXT:    ret
+  %a = call <vscale x 1 x i64> @llvm.riscv.vmax.mask.nxv1i64.nxv1i64(<vscale x 1 x i64> undef, <vscale x 1 x i64> %0, <vscale x 1 x i64> %1, <vscale x 1 x i1> %mask, iXLen %2, iXLen 1)
+  %b = call <vscale x 1 x i64> @llvm.riscv.vmax.mask.nxv1i64.nxv1i64(<vscale x 1 x i64> undef, <vscale x 1 x i64> %1, <vscale x 1 x i64> %0, <vscale x 1 x i1> %mask, iXLen %2, iXLen 1)
+  %ret = add <vscale x 1 x i64> %a, %b
+  ret <vscale x 1 x i64> %ret
+}
+
+; vmaxu.vv
+declare <vscale x 1 x i64> @llvm.riscv.vmaxu.nxv1i64.nxv1i64(<vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, iXLen);
+define <vscale x 1 x i64> @commutable_vmaxu_vv(<vscale x 1 x i64> %0, <vscale x 1 x i64> %1, iXLen %2) nounwind {
+; CHECK-LABEL: commutable_vmaxu_vv:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
+; CHECK-NEXT:    vmaxu.vv v8, v8, v9
+; CHECK-NEXT:    vsetvli a0, zero, e64, m1, ta, ma
+; CHECK-NEXT:    vadd.vv v8, v8, v8
+; CHECK-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vmaxu.nxv1i64.nxv1i64(<vscale x 1 x i64> undef, <vscale x 1 x i64> %0, <vscale x 1 x i64> %1, iXLen %2)
+  %b = call <vscale x 1 x i64> @llvm.riscv.vmaxu.nxv1i64.nxv1i64(<vscale x 1 x i64> undef, <vscale x 1 x i64> %1, <vscale x 1 x i64> %0, iXLen %2)
+  %ret = add <vscale x 1 x i64> %a, %b
+  ret <vscale x 1 x i64> %ret
+}
+
+declare <vscale x 1 x i64> @llvm.riscv.vmaxu.mask.nxv1i64.nxv1i64(<vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i1>, iXLen, iXLen);
+define <vscale x 1 x i64> @commutable_vmaxu_vv_masked(<vscale x 1 x i64> %0, <vscale x 1 x i64> %1, <vscale x 1 x i1> %mask, iXLen %2) {
+; CHECK-LABEL: commutable_vmaxu_vv_masked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
+; CHECK-NEXT:    vmaxu.vv v10, v8, v9, v0.t
+; CHECK-NEXT:    vmaxu.vv v8, v8, v9, v0.t
+; CHECK-NEXT:    vsetvli a0, zero, e64, m1, ta, ma
+; CHECK-NEXT:    vadd.vv v8, v10, v8
+; CHECK-NEXT:    ret
+  %a = call <vscale x 1 x i64> @llvm.riscv.vmaxu.mask.nxv1i64.nxv1i64(<vscale x 1 x i64> undef, <vscale x 1 x i64> %0, <vscale x 1 x i64> %1, <vscale x 1 x i1> %mask, iXLen %2, iXLen 1)
+  %b = call <vscale x 1 x i64> @llvm.riscv.vmaxu.mask.nxv1i64.nxv1i64(<vscale x 1 x i64> undef, <vscale x 1 x i64> %1, <vscale x 1 x i64> %0, <vscale x 1 x i1> %mask, iXLen %2, iXLen 1)
+  %ret = add <vscale x 1 x i64> %a, %b
+  ret <vscale x 1 x i64> %ret
+}
+
+; vmul.vv
+declare <vscale x 1 x i64> @llvm.riscv.vmul.nxv1i64.nxv1i64(<vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, iXLen);
+define <vscale x 1 x i64> @commutable_vmul_vv(<vscale x 1 x i64> %0, <vscale x 1 x i64> %1, iXLen %2) nounwind {
+; CHECK-LABEL: commutable_vmul_vv:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
+; CHECK-NEXT:    vmul.vv v8, v8, v9
+; CHECK-NEXT:    vsetvli a0, zero, e64, m1, ta, ma
+; CHECK-NEXT:    vadd.vv v8, v8, v8
+; CHECK-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vmul.nxv1i64.nxv1i64(<vscale x 1 x i64> undef, <vscale x 1 x i64> %0, <vscale x 1 x i64> %1, iXLen %2)
+  %b = call <vscale x 1 x i64> @llvm.riscv.vmul.nxv1i64.nxv1i64(<vscale x 1 x i64> undef, <vscale x 1 x i64> %1, <vscale x 1 x i64> %0, iXLen %2)
+  %ret = add <vscale x 1 x i64> %a, %b
+  ret <vscale x 1 x i64> %ret
+}
+
+declare <vscale x 1 x i64> @llvm.riscv.vmul.mask.nxv1i64.nxv1i64(<vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i1>, iXLen, iXLen);
+define <vscale x 1 x i64> @commutable_vmul_vv_masked(<vscale x 1 x i64> %0, <vscale x 1 x i64> %1, <vscale x 1 x i1> %mask, iXLen %2) {
+; CHECK-LABEL: commutable_vmul_vv_masked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
+; CHECK-NEXT:    vmul.vv v10, v8, v9, v0.t
+; CHECK-NEXT:    vmul.vv v8, v8, v9, v0.t
+; CHECK-NEXT:    vsetvli a0, zero, e64, m1, ta, ma
+; CHECK-NEXT:    vadd.vv v8, v10, v8
+; CHECK-NEXT:    ret
+  %a = call <vscale x 1 x i64> @llvm.riscv.vmul.mask.nxv1i64.nxv1i64(<vscale x 1 x i64> undef, <vscale x 1 x i64> %0, <vscale x 1 x i64> %1, <vscale x 1 x i1> %mask, iXLen %2, iXLen 1)
+  %b = call <vscale x 1 x i64> @llvm.riscv.vmul.mask.nxv1i64.nxv1i64(<vscale x 1 x i64> undef, <vscale x 1 x i64> %1, <vscale x 1 x i64> %0, <vscale x 1 x i1> %mask, iXLen %2, iXLen 1)
+  %ret = add <vscale x 1 x i64> %a, %b
+  ret <vscale x 1 x i64> %ret
+}
+
+; vmulh.vv
+declare <vscale x 1 x i64> @llvm.riscv.vmulh.nxv1i64.nxv1i64(<vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, iXLen);
+define <vscale x 1 x i64> @commutable_vmulh_vv(<vscale x 1 x i64> %0, <vscale x 1 x i64> %1, iXLen %2) nounwind {
+; CHECK-LABEL: commutable_vmulh_vv:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
+; CHECK-NEXT:    vmulh.vv v8, v8, v9
+; CHECK-NEXT:    vsetvli a0, zero, e64, m1, ta, ma
+; CHECK-NEXT:    vadd.vv v8, v8, v8
+; CHECK-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vmulh.nxv1i64.nxv1i64(<vscale x 1 x i64> undef, <vscale x 1 x i64> %0, <vscale x 1 x i64> %1, iXLen %2)
+  %b = call <vscale x 1 x i64> @llvm.riscv.vmulh.nxv1i64.nxv1i64(<vscale x 1 x i64> undef, <vscale x 1 x i64> %1, <vscale x 1 x i64> %0, iXLen %2)
+  %ret = add <vscale x 1 x i64> %a, %b
+  ret <vscale x 1 x i64> %ret
+}
+
+declare <vscale x 1 x i64> @llvm.riscv.vmulh.mask.nxv1i64.nxv1i64(<vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i1>, iXLen, iXLen);
+define <vscale x 1 x i64> @commutable_vmulh_vv_masked(<vscale x 1 x i64> %0, <vscale x 1 x i64> %1, <vscale x 1 x i1> %mask, iXLen %2) {
+; CHECK-LABEL: commutable_vmulh_vv_masked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
+; CHECK-NEXT:    vmulh.vv v10, v8, v9, v0.t
+; CHECK-NEXT:    vmulh.vv v8, v8, v9, v0.t
+; CHECK-NEXT:    vsetvli a0, zero, e64, m1, ta, ma
+; CHECK-NEXT:    vadd.vv v8, v10, v8
+; CHECK-NEXT:    ret
+  %a = call <vscale x 1 x i64> @llvm.riscv.vmulh.mask.nxv1i64.nxv1i64(<vscale x 1 x i64> undef, <vscale x 1 x i64> %0, <vscale x 1 x i64> %1, <vscale x 1 x i1> %mask, iXLen %2, iXLen 1)
+  %b = call <vscale x 1 x i64> @llvm.riscv.vmulh.mask.nxv1i64.nxv1i64(<vscale x 1 x i64> undef, <vscale x 1 x i64> %1, <vscale x 1 x i64> %0, <vscale x 1 x i1> %mask, iXLen %2, iXLen 1)
+  %ret = add <vscale x 1 x i64> %a, %b
+  ret <vscale x 1 x i64> %ret
+}
+
+; vmulhu.vv
+declare <vscale x 1 x i64> @llvm.riscv.vmulhu.nxv1i64.nxv1i64(<vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, iXLen);
+define <vscale x 1 x i64> @commutable_vmulhu_vv(<vscale x 1 x i64> %0, <vscale x 1 x i64> %1, iXLen %2) nounwind {
+; CHECK-LABEL: commutable_vmulhu_vv:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
+; CHECK-NEXT:    vmulhu.vv v8, v8, v9
+; CHECK-NEXT:    vsetvli a0, zero, e64, m1, ta, ma
+; CHECK-NEXT:    vadd.vv v8, v8, v8
+; CHECK-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vmulhu.nxv1i64.nxv1i64(<vscale x 1 x i64> undef, <vscale x 1 x i64> %0, <vscale x 1 x i64> %1, iXLen %2)
+  %b = call <vscale x 1 x i64> @llvm.riscv.vmulhu.nxv1i64.nxv1i64(<vscale x 1 x i64> undef, <vscale x 1 x i64> %1, <vscale x 1 x i64> %0, iXLen %2)
+  %ret = add <vscale x 1 x i64> %a, %b
+  ret <vscale x 1 x i64> %ret
+}
+
+declare <vscale x 1 x i64> @llvm.riscv.vmulhu.mask.nxv1i64.nxv1i64(<vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i1>, iXLen, iXLen);
+define <vscale x 1 x i64> @commutable_vmulhu_vv_masked(<vscale x 1 x i64> %0, <vscale x 1 x i64> %1, <vscale x 1 x i1> %mask, iXLen %2) {
+; CHECK-LABEL: commutable_vmulhu_vv_masked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
+; CHECK-NEXT:    vmulhu.vv v10, v8, v9, v0.t
+; CHECK-NEXT:    vmulhu.vv v8, v8, v9, v0.t
+; CHECK-NEXT:    vsetvli a0, zero, e64, m1, ta, ma
+; CHECK-NEXT:    vadd.vv v8, v10, v8
+; CHECK-NEXT:    ret
+  %a = call <vscale x 1 x i64> @llvm.riscv.vmulhu.mask.nxv1i64.nxv1i64(<vscale x 1 x i64> undef, <vscale x 1 x i64> %0, <vscale x 1 x i64> %1, <vscale x 1 x i1> %mask, iXLen %2, iXLen 1)
+  %b = call <vscale x 1 x i64> @llvm.riscv.vmulhu.mask.nxv1i64.nxv1i64(<vscale x 1 x i64> undef, <vscale x 1 x i64> %1, <vscale x 1 x i64> %0, <vscale x 1 x i1> %mask, iXLen %2, iXLen 1)
+  %ret = add <vscale x 1 x i64> %a, %b
+  ret <vscale x 1 x i64> %ret
+}
+
+; vwadd.vv
+declare <vscale x 1 x i64> @llvm.riscv.vwadd.nxv1i64.nxv1i32.nxv1i32(<vscale x 1 x i64>, <vscale x 1 x i32>, <vscale x 1 x i32>, iXLen);
+define <vscale x 1 x i64> @commutable_vwadd_vv(<vscale x 1 x i32> %0, <vscale x 1 x i32> %1, iXLen %2) nounwind {
+; CHECK-LABEL: commutable_vwadd_vv:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli zero, a0, e32, mf2, ta, ma
+; CHECK-NEXT:    vwadd.vv v10, v8, v9
+; CHECK-NEXT:    vsetvli a0, zero, e64, m1, ta, ma
+; CHECK-NEXT:    vadd.vv v8, v10, v10
+; CHECK-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vwadd.nxv1i64.nxv1i32.nxv1i32(<vscale x 1 x i64> undef, <vscale x 1 x i32> %0, <vscale x 1 x i32> %1, iXLen %2)
+  %b = call <vscale x 1 x i64> @llvm.riscv.vwadd.nxv1i64.nxv1i32.nxv1i32(<vscale x 1 x i64> undef, <vscale x 1 x i32> %1, <vscale x 1 x i32> %0, iXLen %2)
+  %ret = add <vscale x 1 x i64> %a, %b
+  ret <vscale x 1 x i64> %ret
+}
+
+declare <vscale x 1 x i64> @llvm.riscv.vwadd.mask.nxv1i64.nxv1i32.nxv1i32(<vscale x 1 x i64>, <vscale x 1 x i32>, <vscale x 1 x i32>, <vscale x 1 x i1>, iXLen, iXLen);
+define <vscale x 1 x i64> @commutable_vwadd_vv_masked(<vscale x 1 x i32> %0, <vscale x 1 x i32> %1, <vscale x 1 x i1> %mask, iXLen %2) {
+; CHECK-LABEL: commutable_vwadd_vv_masked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e32, mf2, ta, ma
+; CHECK-NEXT:    vwadd.vv v10, v8, v9, v0.t
+; CHECK-NEXT:    vwadd.vv v11, v8, v9, v0.t
+; CHECK-NEXT:    vsetvli a0, zero, e64, m1, ta, ma
+; CHECK-NEXT:    vadd.vv v8, v10, v11
+; CHECK-NEXT:    ret
+  %a = call <vscale x 1 x i64> @llvm.riscv.vwadd.mask.nxv1i64.nxv1i32.nxv1i32(<vscale x 1 x i64> undef, <vscale x 1 x i32> %0, <vscale x 1 x i32> %1, <vscale x 1 x i1> %mask, iXLen %2, iXLen 1)
+  %b = call <vscale x 1 x i64> @llvm.riscv.vwadd.mask.nxv1i64.nxv1i32.nxv1i32(<vscale x 1 x i64> undef, <vscale x 1 x i32> %1, <vscale x 1 x i32> %0, <vscale x 1 x i1> %mask, iXLen %2, iXLen 1)
+  %ret = add <vscale x 1 x i64> %a, %b
+  ret <vscale x 1 x i64> %ret
+}
+
+; vwaddu.vv
+declare <vscale x 1 x i64> @llvm.riscv.vwaddu.nxv1i64.nxv1i32.nxv1i32(<vscale x 1 x i64>, <vscale x 1 x i32>, <vscale x 1 x i32>, iXLen);
+define <vscale x 1 x i64> @commutable_vwaddu_vv(<vscale x 1 x i32> %0, <vscale x 1 x i32> %1, iXLen %2) nounwind {
+; CHECK-LABEL: commutable_vwaddu_vv:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli zero, a0, e32, mf2, ta, ma
+; CHECK-NEXT:    vwaddu.vv v10, v8, v9
+; CHECK-NEXT:    vsetvli a0, zero, e64, m1, ta, ma
+; CHECK-NEXT:    vadd.vv v8, v10, v10
+; CHECK-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vwaddu.nxv1i64.nxv1i32.nxv1i32(<vscale x 1 x i64> undef, <vscale x 1 x i32> %0, <vscale x 1 x i32> %1, iXLen %2)
+  %b = call <vscale x 1 x i64> @llvm.riscv.vwaddu.nxv1i64.nxv1i32.nxv1i32(<vscale x 1 x i64> undef, <vscale x 1 x i32> %1, <vscale x 1 x i32> %0, iXLen %2)
+  %ret = add <vscale x 1 x i64> %a, %b
+  ret <vscale x 1 x i64> %ret
+}
+
+declare <vscale x 1 x i64> @llvm.riscv.vwaddu.mask.nxv1i64.nxv1i32.nxv1i32(<vscale x 1 x i64>, <vscale x 1 x i32>, <vscale x 1 x i32>, <vscale x 1 x i1>, iXLen, iXLen);
+define <vscale x 1 x i64> @commutable_vwaddu_vv_masked(<vscale x 1 x i32> %0, <vscale x 1 x i32> %1, <vscale x 1 x i1> %mask, iXLen %2) {
+; CHECK-LABEL: commutable_vwaddu_vv_masked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e32, mf2, ta, ma
+; CHECK-NEXT:    vwaddu.vv v10, v8, v9, v0.t
+; CHECK-NEXT:    vwaddu.vv v11, v8, v9, v0.t
+; CHECK-NEXT:    vsetvli a0, zero, e64, m1, ta, ma
+; CHECK-NEXT:    vadd.vv v8, v10, v11
+; CHECK-NEXT:    ret
+  %a = call <vscale x 1 x i64> @llvm.riscv.vwaddu.mask.nxv1i64.nxv1i32.nxv1i32(<vscale x 1 x i64> undef, <vscale x 1 x i32> %0, <vscale x 1 x i32> %1, <vscale x 1 x i1> %mask, iXLen %2, iXLen 1)
+  %b = call <vscale x 1 x i64> @llvm.riscv.vwaddu.mask.nxv1i64.nxv1i32.nxv1i32(<vscale x 1 x i64> undef, <vscale x 1 x i32> %1, <vscale x 1 x i32> %0, <vscale x 1 x i1> %mask, iXLen %2, iXLen 1)
+  %ret = add <vscale x 1 x i64> %a, %b
+  ret <vscale x 1 x i64> %ret
+}
+
+; vwmul.vv
+declare <vscale x 1 x i64> @llvm.riscv.vwmul.nxv1i64.nxv1i32.nxv1i32(<vscale x 1 x i64>, <vscale x 1 x i32>, <vscale x 1 x i32>, iXLen);
+define <vscale x 1 x i64> @commutable_vwmul_vv(<vscale x 1 x i32> %0, <vscale x 1 x i32> %1, iXLen %2) nounwind {
+; CHECK-LABEL: commutable_vwmul_vv:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli zero, a0, e32, mf2, ta, ma
+; CHECK-NEXT:    vwmul.vv v10, v8, v9
+; CHECK-NEXT:    vsetvli a0, zero, e64, m1, ta, ma
+; CHECK-NEXT:    vadd.vv v8, v10, v10
+; CHECK-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vwmul.nxv1i64.nxv1i32.nxv1i32(<vscale x 1 x i64> undef, <vscale x 1 x i32> %0, <vscale x 1 x i32> %1, iXLen %2)
+  %b = call <vscale x 1 x i64> @llvm.riscv.vwmul.nxv1i64.nxv1i32.nxv1i32(<vscale x 1 x i64> undef, <vscale x 1 x i32> %1, <vscale x 1 x i32> %0, iXLen %2)
+  %ret = add <vscale x 1 x i64> %a, %b
+  ret <vscale x 1 x i64> %ret
+}
+
+declare <vscale x 1 x i64> @llvm.riscv.vwmul.mask.nxv1i64.nxv1i32.nxv1i32(<vscale x 1 x i64>, <vscale x 1 x i32>, <vscale x 1 x i32>, <vscale x 1 x i1>, iXLen, iXLen);
+define <vscale x 1 x i64> @commutable_vwmul_vv_masked(<vscale x 1 x i32> %0, <vscale x 1 x i32> %1, <vscale x 1 x i1> %mask, iXLen %2) {
+; CHECK-LABEL: commutable_vwmul_vv_masked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e32, mf2, ta, ma
+; CHECK-NEXT:    vwmul.vv v10, v8, v9, v0.t
+; CHECK-NEXT:    vwmul.vv v11, v8, v9, v0.t
+; CHECK-NEXT:    vsetvli a0, zero, e64, m1, ta, ma
+; CHECK-NEXT:    vadd.vv v8, v10, v11
+; CHECK-NEXT:    ret
+  %a = call <vscale x 1 x i64> @llvm.riscv.vwmul.mask.nxv1i64.nxv1i32.nxv1i32(<vscale x 1 x i64> undef, <vscale x 1 x i32> %0, <vscale x 1 x i32> %1, <vscale x 1 x i1> %mask, iXLen %2, iXLen 1)
+  %b = call <vscale x 1 x i64> @llvm.riscv.vwmul.mask.nxv1i64.nxv1i32.nxv1i32(<vscale x 1 x i64> undef, <vscale x 1 x i32> %1, <vscale x 1 x i32> %0, <vscale x 1 x i1> %mask, iXLen %2, iXLen 1)
+  %ret = add <vscale x 1 x i64> %a, %b
+  ret <vscale x 1 x i64> %ret
+}
+
+; vwmulu.vv
+declare <vscale x 1 x i64> @llvm.riscv.vwmulu.nxv1i64.nxv1i32.nxv1i32(<vscale x 1 x i64>, <vscale x 1 x i32>, <vscale x 1 x i32>, iXLen);
+define <vscale x 1 x i64> @commutable_vwmulu_vv(<vscale x 1 x i32> %0, <vscale x 1 x i32> %1, iXLen %2) nounwind {
+; CHECK-LABEL: commutable_vwmulu_vv:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli zero, a0, e32, mf2, ta, ma
+; CHECK-NEXT:    vwmulu.vv v10, v8, v9
+; CHECK-NEXT:    vsetvli a0, zero, e64, m1, ta, ma
+; CHECK-NEXT:    vadd.vv v8, v10, v10
+; CHECK-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vwmulu.nxv1i64.nxv1i32.nxv1i32(<vscale x 1 x i64> undef, <vscale x 1 x i32> %0, <vscale x 1 x i32> %1, iXLen %2)
+  %b = call <vscale x 1 x i64> @llvm.riscv.vwmulu.nxv1i64.nxv1i32.nxv1i32(<vscale x 1 x i64> undef, <vscale x 1 x i32> %1, <vscale x 1 x i32> %0, iXLen %2)
+  %ret = add <vscale x 1 x i64> %a, %b
+  ret <vscale x 1 x i64> %ret
+}
+
+declare <vscale x 1 x i64> @llvm.riscv.vwmulu.mask.nxv1i64.nxv1i32.nxv1i32(<vscale x 1 x i64>, <vscale x 1 x i32>, <vscale x 1 x i32>, <vscale x 1 x i1>, iXLen, iXLen);
+define <vscale x 1 x i64> @commutable_vwmulu_vv_masked(<vscale x 1 x i32> %0, <vscale x 1 x i32> %1, <vscale x 1 x i1> %mask, iXLen %2) {
+; CHECK-LABEL: commutable_vwmulu_vv_masked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e32, mf2, ta, ma
+; CHECK-NEXT:    vwmulu.vv v10, v8, v9, v0.t
+; CHECK-NEXT:    vwmulu.vv v11, v8, v9, v0.t
+; CHECK-NEXT:    vsetvli a0, zero, e64, m1, ta, ma
+; CHECK-NEXT:    vadd.vv v8, v10, v11
+; CHECK-NEXT:    ret
+  %a = call <vscale x 1 x i64> @llvm.riscv.vwmulu.mask.nxv1i64.nxv1i32.nxv1i32(<vscale x 1 x i64> undef, <vscale x 1 x i32> %0, <vscale x 1 x i32> %1, <vscale x 1 x i1> %mask, iXLen %2, iXLen 1)
+  %b = call <vscale x 1 x i64> @llvm.riscv.vwmulu.mask.nxv1i64.nxv1i32.nxv1i32(<vscale x 1 x i64> undef, <vscale x 1 x i32> %1, <vscale x 1 x i32> %0, <vscale x 1 x i1> %mask, iXLen %2, iXLen 1)
+  %ret = add <vscale x 1 x i64> %a, %b
+  ret <vscale x 1 x i64> %ret
+}
+
+; vwmacc.vv
+declare <vscale x 1 x i64> @llvm.riscv.vwmacc.nxv1i64.nxv1i32(<vscale x 1 x i64>, <vscale x 1 x i32>, <vscale x 1 x i32>, iXLen, iXLen);
+define <vscale x 1 x i64> @commutable_vwmacc_vv(<vscale x 1 x i32> %0, <vscale x 1 x i32> %1, iXLen %2) nounwind {
+; CHECK-LABEL: commutable_vwmacc_vv:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli zero, a0, e32, mf2, ta, ma
+; CHECK-NEXT:    vwmacc.vv v10, v8, v9
+; CHECK-NEXT:    vsetvli a0, zero, e64, m1, ta, ma
+; CHECK-NEXT:    vadd.vv v8, v10, v10
+; CHECK-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vwmacc.nxv1i64.nxv1i32(<vscale x 1 x i64> undef, <vscale x 1 x i32> %0, <vscale x 1 x i32> %1, iXLen %2, iXLen 1)
+  %b = call <vscale x 1 x i64> @llvm.riscv.vwmacc.nxv1i64.nxv1i32(<vscale x 1 x i64> undef, <vscale x 1 x i32> %1, <vscale x 1 x i32> %0, iXLen %2, iXLen 1)
+  %ret = add <vscale x 1 x i64> %a, %b
+  ret <vscale x 1 x i64> %ret
+}
+
+declare <vscale x 1 x i64> @llvm.riscv.vwmacc.mask.nxv1i64.nxv1i32(<vscale x 1 x i64>, <vscale x 1 x i32>, <vscale x 1 x i32>, <vscale x 1 x i1>, iXLen, iXLen);
+define <vscale x 1 x i64> @commutable_vwmacc_vv_masked(<vscale x 1 x i32> %0, <vscale x 1 x i32> %1, <vscale x 1 x i1> %mask, iXLen %2) {
+; CHECK-LABEL: commutable_vwmacc_vv_masked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e32, mf2, ta, ma
+; CHECK-NEXT:    vwmacc.vv v10, v8, v9, v0.t
+; CHECK-NEXT:    vwmacc.vv v11, v9, v8, v0.t
+; CHECK-NEXT:    vsetvli a0, zero, e64, m1, ta, ma
+; CHECK-NEXT:    vadd.vv v8, v10, v11
+; CHECK-NEXT:    ret
+  %a = call <vscale x 1 x i64> @llvm.riscv.vwmacc.mask.nxv1i64.nxv1i32(<vscale x 1 x i64> undef, <vscale x 1 x i32> %0, <vscale x 1 x i32> %1, <vscale x 1 x i1> %mask, iXLen %2, iXLen 1)
+  %b = call <vscale x 1 x i64> @llvm.riscv.vwmacc.mask.nxv1i64.nxv1i32(<vscale x 1 x i64> undef, <vscale x 1 x i32> %1, <vscale x 1 x i32> %0, <vscale x 1 x i1> %mask, iXLen %2, iXLen 1)
+  %ret = add <vscale x 1 x i64> %a, %b
+  ret <vscale x 1 x i64> %ret
+}
+
+; vwmaccu.vv
+declare <vscale x 1 x i64> @llvm.riscv.vwmaccu.nxv1i64.nxv1i32(<vscale x 1 x i64>, <vscale x 1 x i32>, <vscale x 1 x i32>, iXLen, iXLen);
+define <vscale x 1 x i64> @commutable_vwmaccu_vv(<vscale x 1 x i32> %0, <vscale x 1 x i32> %1, iXLen %2) nounwind {
+; CHECK-LABEL: commutable_vwmaccu_vv:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli zero, a0, e32, mf2, ta, ma
+; CHECK-NEXT:    vwmaccu.vv v10, v8, v9
+; CHECK-NEXT:    vsetvli a0, zero, e64, m1, ta, ma
+; CHECK-NEXT:    vadd.vv v8, v10, v10
+; CHECK-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vwmaccu.nxv1i64.nxv1i32(<vscale x 1 x i64> undef, <vscale x 1 x i32> %0, <vscale x 1 x i32> %1, iXLen %2, iXLen 1)
+  %b = call <vscale x 1 x i64> @llvm.riscv.vwmaccu.nxv1i64.nxv1i32(<vscale x 1 x i64> undef, <vscale x 1 x i32> %1, <vscale x 1 x i32> %0, iXLen %2, iXLen 1)
+  %ret = add <vscale x 1 x i64> %a, %b
+  ret <vscale x 1 x i64> %ret
+}
+
+declare <vscale x 1 x i64> @llvm.riscv.vwmaccu.mask.nxv1i64.nxv1i32(<vscale x 1 x i64>, <vscale x 1 x i32>, <vscale x 1 x i32>, <vscale x 1 x i1>, iXLen, iXLen);
+define <vscale x 1 x i64> @commutable_vwmaccu_vv_masked(<vscale x 1 x i32> %0, <vscale x 1 x i32> %1, <vscale x 1 x i1> %mask, iXLen %2) {
+; CHECK-LABEL: commutable_vwmaccu_vv_masked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e32, mf2, ta, ma
+; CHECK-NEXT:    vwmaccu.vv v10, v8, v9, v0.t
+; CHECK-NEXT:    vwmaccu.vv v11, v9, v8, v0.t
+; CHECK-NEXT:    vsetvli a0, zero, e64, m1, ta, ma
+; CHECK-NEXT:    vadd.vv v8, v10, v11
+; CHECK-NEXT:    ret
+  %a = call <vscale x 1 x i64> @llvm.riscv.vwmaccu.mask.nxv1i64.nxv1i32(<vscale x 1 x i64> undef, <vscale x 1 x i32> %0, <vscale x 1 x i32> %1, <vscale x 1 x i1> %mask, iXLen %2, iXLen 1)
+  %b = call <vscale x 1 x i64> @llvm.riscv.vwmaccu.mask.nxv1i64.nxv1i32(<vscale x 1 x i64> undef, <vscale x 1 x i32> %1, <vscale x 1 x i32> %0, <vscale x 1 x i1> %mask, iXLen %2, iXLen 1)
+  %ret = add <vscale x 1 x i64> %a, %b
+  ret <vscale x 1 x i64> %ret
+}
+
+; vadc.vvm
+declare <vscale x 1 x i64> @llvm.riscv.vadc.nxv1i64.nxv1i64(<vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i1>, iXLen);
+define <vscale x 1 x i64> @commutable_vadc_vv(<vscale x 1 x i64> %0, <vscale x 1 x i64> %1, <vscale x 1 x i1> %mask, iXLen %2) nounwind {
+; CHECK-LABEL: commutable_vadc_vv:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
+; CHECK-NEXT:    vadc.vvm v10, v8, v9, v0
+; CHECK-NEXT:    vadc.vvm v8, v8, v9, v0
+; CHECK-NEXT:    vsetvli a0, zero, e64, m1, ta, ma
+; CHECK-NEXT:    vadd.vv v8, v10, v8
+; CHECK-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vadc.nxv1i64.nxv1i64(<vscale x 1 x i64> undef, <vscale x 1 x i64> %0, <vscale x 1 x i64> %1, <vscale x 1 x i1> %mask, iXLen %2)
+  %b = call <vscale x 1 x i64> @llvm.riscv.vadc.nxv1i64.nxv1i64(<vscale x 1 x i64> undef, <vscale x 1 x i64> %1, <vscale x 1 x i64> %0, <vscale x 1 x i1> %mask, iXLen %2)
+  %ret = add <vscale x 1 x i64> %a, %b
+  ret <vscale x 1 x i64> %ret
+}
+
+; vsadd.vv
+declare <vscale x 1 x i64> @llvm.riscv.vsadd.nxv1i64.nxv1i64(<vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, iXLen);
+define <vscale x 1 x i64> @commutable_vsadd_vv(<vscale x 1 x i64> %0, <vscale x 1 x i64> %1, iXLen %2) nounwind {
+; CHECK-LABEL: commutable_vsadd_vv:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
+; CHECK-NEXT:    vsadd.vv v10, v8, v9
+; CHECK-NEXT:    vsadd.vv v8, v9, v8
+; CHECK-NEXT:    vsetvli a0, zero, e64, m1, ta, ma
+; CHECK-NEXT:    vadd.vv v8, v10, v8
+; CHECK-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vsadd.nxv1i64.nxv1i64(<vscale x 1 x i64> undef, <vscale x 1 x i64> %0, <vscale x 1 x i64> %1, iXLen %2)
+  %b = call <vscale x 1 x i64> @llvm.riscv.vsadd.nxv1i64.nxv1i64(<vscale x 1 x i64> undef, <vscale x 1 x i64> %1, <vscale x 1 x i64> %0, iXLen %2)
+  %ret = add <vscale x 1 x i64> %a, %b
+  ret <vscale x 1 x i64> %ret
+}
+
+declare <vscale x 1 x i64> @llvm.riscv.vsadd.mask.nxv1i64.nxv1i64(<vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i1>, iXLen, iXLen);
+define <vscale x 1 x i64> @commutable_vsadd_vv_masked(<vscale x 1 x i64> %0, <vscale x 1 x i64> %1, <vscale x 1 x i1> %mask, iXLen %2) {
+; CHECK-LABEL: commutable_vsadd_vv_masked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
+; CHECK-NEXT:    vsadd.vv v10, v8, v9, v0.t
+; CHECK-NEXT:    vsadd.vv v8, v9, v8, v0.t
+; CHECK-NEXT:    vsetvli a0, zero, e64, m1, ta, ma
+; CHECK-NEXT:    vadd.vv v8, v10, v8
+; CHECK-NEXT:    ret
+  %a = call <vscale x 1 x i64> @llvm.riscv.vsadd.mask.nxv1i64.nxv1i64(<vscale x 1 x i64> undef, <vscale x 1 x i64> %0, <vscale x 1 x i64> %1, <vscale x 1 x i1> %mask, iXLen %2, iXLen 1)
+  %b = call <vscale x 1 x i64> @llvm.riscv.vsadd.mask.nxv1i64.nxv1i64(<vscale x 1 x i64> undef, <vscale x 1 x i64> %1, <vscale x 1 x i64> %0, <vscale x 1 x i1> %mask, iXLen %2, iXLen 1)
+  %ret = add <vscale x 1 x i64> %a, %b
+  ret <vscale x 1 x i64> %ret
+}
+
+; vsaddu.vv
+declare <vscale x 1 x i64> @llvm.riscv.vsaddu.nxv1i64.nxv1i64(<vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, iXLen);
+define <vscale x 1 x i64> @commutable_vsaddu_vv(<vscale x 1 x i64> %0, <vscale x 1 x i64> %1, iXLen %2) nounwind {
+; CHECK-LABEL: commutable_vsaddu_vv:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
+; CHECK-NEXT:    vsaddu.vv v10, v8, v9
+; CHECK-NEXT:    vsaddu.vv v8, v9, v8
+; CHECK-NEXT:    vsetvli a0, zero, e64, m1, ta, ma
+; CHECK-NEXT:    vadd.vv v8, v10, v8
+; CHECK-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vsaddu.nxv1i64.nxv1i64(<vscale x 1 x i64> undef, <vscale x 1 x i64> %0, <vscale x 1 x i64> %1, iXLen %2)
+  %b = call <vscale x 1 x i64> @llvm.riscv.vsaddu.nxv1i64.nxv1i64(<vscale x 1 x i64> undef, <vscale x 1 x i64> %1, <vscale x 1 x i64> %0, iXLen %2)
+  %ret = add <vscale x 1 x i64> %a, %b
+  ret <vscale x 1 x i64> %ret
+}
+
+declare <vscale x 1 x i64> @llvm.riscv.vsaddu.mask.nxv1i64.nxv1i64(<vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i1>, iXLen, iXLen);
+define <vscale x 1 x i64> @commutable_vsaddu_vv_masked(<vscale x 1 x i64> %0, <vscale x 1 x i64> %1, <vscale x 1 x i1> %mask, iXLen %2) {
+; CHECK-LABEL: commutable_vsaddu_vv_masked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
+; CHECK-NEXT:    vsaddu.vv v10, v8, v9, v0.t
+; CHECK-NEXT:    vsaddu.vv v8, v9, v8, v0.t
+; CHECK-NEXT:    vsetvli a0, zero, e64, m1, ta, ma
+; CHECK-NEXT:    vadd.vv v8, v10, v8
+; CHECK-NEXT:    ret
+  %a = call <vscale x 1 x i64> @llvm.riscv.vsaddu.mask.nxv1i64.nxv1i64(<vscale x 1 x i64> undef, <vscale x 1 x i64> %0, <vscale x 1 x i64> %1, <vscale x 1 x i1> %mask, iXLen %2, iXLen 1)
+  %b = call <vscale x 1 x i64> @llvm.riscv.vsaddu.mask.nxv1i64.nxv1i64(<vscale x 1 x i64> undef, <vscale x 1 x i64> %1, <vscale x 1 x i64> %0, <vscale x 1 x i1> %mask, iXLen %2, iXLen 1)
+  %ret = add <vscale x 1 x i64> %a, %b
+  ret <vscale x 1 x i64> %ret
+}
+
+; vaadd.vv
+declare <vscale x 1 x i64> @llvm.riscv.vaadd.nxv1i64.nxv1i64(<vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, iXLen, iXLen);
+define <vscale x 1 x i64> @commutable_vaadd_vv(<vscale x 1 x i64> %0, <vscale x 1 x i64> %1, iXLen %2) nounwind {
+; CHECK-LABEL: commutable_vaadd_vv:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
+; CHECK-NEXT:    csrwi vxrm, 0
+; CHECK-NEXT:    vaadd.vv v8, v8, v9
+; CHECK-NEXT:    vsetvli a0, zero, e64, m1, ta, ma
+; CHECK-NEXT:    vadd.vv v8, v8, v8
+; CHECK-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vaadd.nxv1i64.nxv1i64(<vscale x 1 x i64> undef, <vscale x 1 x i64> %0, <vscale x 1 x i64> %1, iXLen 0, iXLen %2)
+  %b = call <vscale x 1 x i64> @llvm.riscv.vaadd.nxv1i64.nxv1i64(<vscale x 1 x i64> undef, <vscale x 1 x i64> %1, <vscale x 1 x i64> %0, iXLen 0, iXLen %2)
+  %ret = add <vscale x 1 x i64> %a, %b
+  ret <vscale x 1 x i64> %ret
+}
+
+declare <vscale x 1 x i64> @llvm.riscv.vaadd.mask.nxv1i64.nxv1i64(<vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i1>, iXLen, iXLen, iXLen);
+define <vscale x 1 x i64> @commutable_vaadd_vv_masked(<vscale x 1 x i64> %0, <vscale x 1 x i64> %1, <vscale x 1 x i1> %mask, iXLen %2) {
+; CHECK-LABEL: commutable_vaadd_vv_masked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
+; CHECK-NEXT:    csrwi vxrm, 0
+; CHECK-NEXT:    vaadd.vv v10, v8, v9, v0.t
+; CHECK-NEXT:    vaadd.vv v8, v8, v9, v0.t
+; CHECK-NEXT:    vsetvli a0, zero, e64, m1, ta, ma
+; CHECK-NEXT:    vadd.vv v8, v10, v8
+; CHECK-NEXT:    ret
+  %a = call <vscale x 1 x i64> @llvm.riscv.vaadd.mask.nxv1i64.nxv1i64(<vscale x 1 x i64> undef, <vscale x 1 x i64> %0, <vscale x 1 x i64> %1, <vscale x 1 x i1> %mask, iXLen 0, iXLen %2, iXLen 1)
+  %b = call <vscale x 1 x i64> @llvm.riscv.vaadd.mask.nxv1i64.nxv1i64(<vscale x 1 x i64> undef, <vscale x 1 x i64> %1, <vscale x 1 x i64> %0, <vscale x 1 x i1> %mask, iXLen 0, iXLen %2, iXLen 1)
+  %ret = add <vscale x 1 x i64> %a, %b
+  ret <vscale x 1 x i64> %ret
+}
+
+; vaaddu.vv
+declare <vscale x 1 x i64> @llvm.riscv.vaaddu.nxv1i64.nxv1i64(<vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, iXLen, iXLen);
+define <vscale x 1 x i64> @commutable_vaaddu_vv(<vscale x 1 x i64> %0, <vscale x 1 x i64> %1, iXLen %2) nounwind {
+; CHECK-LABEL: commutable_vaaddu_vv:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
+; CHECK-NEXT:    csrwi vxrm, 0
+; CHECK-NEXT:    vaaddu.vv v8, v8, v9
+; CHECK-NEXT:    vsetvli a0, zero, e64, m1, ta, ma
+; CHECK-NEXT:    vadd.vv v8, v8, v8
+; CHECK-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vaaddu.nxv1i64.nxv1i64(<vscale x 1 x i64> undef, <vscale x 1 x i64> %0, <vscale x 1 x i64> %1, iXLen 0, iXLen %2)
+  %b = call <vscale x 1 x i64> @llvm.riscv.vaaddu.nxv1i64.nxv1i64(<vscale x 1 x i64> undef, <vscale x 1 x i64> %1, <vscale x 1 x i64> %0, iXLen 0, iXLen %2)
+  %ret = add <vscale x 1 x i64> %a, %b
+  ret <vscale x 1 x i64> %ret
+}
+
+declare <vscale x 1 x i64> @llvm.riscv.vaaddu.mask.nxv1i64.nxv1i64(<vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i1>, iXLen, iXLen, iXLen);
+define <vscale x 1 x i64> @commutable_vaaddu_vv_masked(<vscale x 1 x i64> %0, <vscale x 1 x i64> %1, <vscale x 1 x i1> %mask, iXLen %2) {
+; CHECK-LABEL: commutable_vaaddu_vv_masked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
+; CHECK-NEXT:    csrwi vxrm, 0
+; CHECK-NEXT:    vaaddu.vv v10, v8, v9, v0.t
+; CHECK-NEXT:    vaaddu.vv v8, v8, v9, v0.t
+; CHECK-NEXT:    vsetvli a0, zero, e64, m1, ta, ma
+; CHECK-NEXT:    vadd.vv v8, v10, v8
+; CHECK-NEXT:    ret
+  %a = call <vscale x 1 x i64> @llvm.riscv.vaaddu.mask.nxv1i64.nxv1i64(<vscale x 1 x i64> undef, <vscale x 1 x i64> %0, <vscale x 1 x i64> %1, <vscale x 1 x i1> %mask, iXLen 0, iXLen %2, iXLen 1)
+  %b = call <vscale x 1 x i64> @llvm.riscv.vaaddu.mask.nxv1i64.nxv1i64(<vscale x 1 x i64> undef, <vscale x 1 x i64> %1, <vscale x 1 x i64> %0, <vscale x 1 x i1> %mask, iXLen 0, iXLen %2, iXLen 1)
+  %ret = add <vscale x 1 x i64> %a, %b
+  ret <vscale x 1 x i64> %ret
+}
+
+; vsmul.vv
+declare <vscale x 1 x i64> @llvm.riscv.vsmul.nxv1i64.nxv1i64(<vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, iXLen, iXLen);
+define <vscale x 1 x i64> @commutable_vsmul_vv(<vscale x 1 x i64> %0, <vscale x 1 x i64> %1, iXLen %2) nounwind {
+; CHECK-LABEL: commutable_vsmul_vv:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
+; CHECK-NEXT:    csrwi vxrm, 0
+; CHECK-NEXT:    vsmul.vv v10, v8, v9
+; CHECK-NEXT:    vsmul.vv v8, v9, v8
+; CHECK-NEXT:    vsetvli a0, zero, e64, m1, ta, ma
+; CHECK-NEXT:    vadd.vv v8, v10, v8
+; CHECK-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i64> @llvm.riscv.vsmul.nxv1i64.nxv1i64(<vscale x 1 x i64> undef, <vscale x 1 x i64> %0, <vscale x 1 x i64> %1, iXLen 0, iXLen %2)
+  %b = call <vscale x 1 x i64> @llvm.riscv.vsmul.nxv1i64.nxv1i64(<vscale x 1 x i64> undef, <vscale x 1 x i64> %1, <vscale x 1 x i64> %0, iXLen 0, iXLen %2)
+  %ret = add <vscale x 1 x i64> %a, %b
+  ret <vscale x 1 x i64> %ret
+}
+
+declare <vscale x 1 x i64> @llvm.riscv.vsmul.mask.nxv1i64.nxv1i64(<vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i1>, iXLen, iXLen, iXLen);
+define <vscale x 1 x i64> @commutable_vsmul_vv_masked(<vscale x 1 x i64> %0, <vscale x 1 x i64> %1, <vscale x 1 x i1> %mask, iXLen %2) {
+; CHECK-LABEL: commutable_vsmul_vv_masked:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetvli zero, a0, e64, m1, ta, ma
+; CHECK-NEXT:    csrwi vxrm, 0
+; CHECK-NEXT:    vsmul.vv v10, v8, v9, v0.t
+; CHECK-NEXT:    vsmul.vv v8, v9, v8, v0.t
+; CHECK-NEXT:    vsetvli a0, zero, e64, m1, ta, ma
+; CHECK-NEXT:    vadd.vv v8, v10, v8
+; CHECK-NEXT:    ret
+  %a = call <vscale x 1 x i64> @llvm.riscv.vsmul.mask.nxv1i64.nxv1i64(<vscale x 1 x i64> undef, <vscale x 1 x i64> %0, <vscale x 1 x i64> %1, <vscale x 1 x i1> %mask, iXLen 0, iXLen %2, iXLen 1)
+  %b = call <vscale x 1 x i64> @llvm.riscv.vsmul.mask.nxv1i64.nxv1i64(<vscale x 1 x i64> undef, <vscale x 1 x i64> %1, <vscale x 1 x i64> %0, <vscale x 1 x i1> %mask, iXLen 0, iXLen %2, iXLen 1)
+  %ret = add <vscale x 1 x i64> %a, %b
+  ret <vscale x 1 x i64> %ret
+}
diff --git a/llvm/test/CodeGen/RISCV/rvv/fixed-vectors-deinterleave-load.ll b/llvm/test/CodeGen/RISCV/rvv/fixed-vectors-deinterleave-load.ll
index 83edd49bc963..1587f770f87c 100644
--- a/llvm/test/CodeGen/RISCV/rvv/fixed-vectors-deinterleave-load.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/fixed-vectors-deinterleave-load.ll
@@ -35,7 +35,7 @@ define {<16 x i1>, <16 x i1>} @vector_deinterleave_load_v16i1_v32i1(ptr %p) {
 ; CHECK-NEXT:    vmv.v.v v0, v9
 ; CHECK-NEXT:    ret
   %vec = load <32 x i1>, ptr %p
-  %retval = call {<16 x i1>, <16 x i1>} @llvm.experimental.vector.deinterleave2.v32i1(<32 x i1> %vec)
+  %retval = call {<16 x i1>, <16 x i1>} @llvm.vector.deinterleave2.v32i1(<32 x i1> %vec)
   ret {<16 x i1>, <16 x i1>} %retval
 }
 
@@ -46,7 +46,7 @@ define {<16 x i8>, <16 x i8>} @vector_deinterleave_load_v16i8_v32i8(ptr %p) {
 ; CHECK-NEXT:    vlseg2e8.v v8, (a0)
 ; CHECK-NEXT:    ret
   %vec = load <32 x i8>, ptr %p
-  %retval = call {<16 x i8>, <16 x i8>} @llvm.experimental.vector.deinterleave2.v32i8(<32 x i8> %vec)
+  %retval = call {<16 x i8>, <16 x i8>} @llvm.vector.deinterleave2.v32i8(<32 x i8> %vec)
   ret {<16 x i8>, <16 x i8>} %retval
 }
 
@@ -62,7 +62,7 @@ define {<8 x i16>, <8 x i16>} @vector_deinterleave_load_v8i16_v16i16_align1(ptr
 ; CHECK-NEXT:    vnsrl.wi v9, v10, 16
 ; CHECK-NEXT:    ret
   %vec = load <16 x i16>, ptr %p, align 1
-  %retval = call {<8 x i16>, <8 x i16>} @llvm.experimental.vector.deinterleave2.v16i16(<16 x i16> %vec)
+  %retval = call {<8 x i16>, <8 x i16>} @llvm.vector.deinterleave2.v16i16(<16 x i16> %vec)
   ret {<8 x i16>, <8 x i16>} %retval
 }
 
@@ -73,7 +73,7 @@ define {<8 x i16>, <8 x i16>} @vector_deinterleave_load_v8i16_v16i16(ptr %p) {
 ; CHECK-NEXT:    vlseg2e16.v v8, (a0)
 ; CHECK-NEXT:    ret
   %vec = load <16 x i16>, ptr %p
-  %retval = call {<8 x i16>, <8 x i16>} @llvm.experimental.vector.deinterleave2.v16i16(<16 x i16> %vec)
+  %retval = call {<8 x i16>, <8 x i16>} @llvm.vector.deinterleave2.v16i16(<16 x i16> %vec)
   ret {<8 x i16>, <8 x i16>} %retval
 }
 
@@ -84,7 +84,7 @@ define {<4 x i32>, <4 x i32>} @vector_deinterleave_load_v4i32_vv8i32(ptr %p) {
 ; CHECK-NEXT:    vlseg2e32.v v8, (a0)
 ; CHECK-NEXT:    ret
   %vec = load <8 x i32>, ptr %p
-  %retval = call {<4 x i32>, <4 x i32>} @llvm.experimental.vector.deinterleave2.v8i32(<8 x i32> %vec)
+  %retval = call {<4 x i32>, <4 x i32>} @llvm.vector.deinterleave2.v8i32(<8 x i32> %vec)
   ret {<4 x i32>, <4 x i32>} %retval
 }
 
@@ -95,15 +95,15 @@ define {<2 x i64>, <2 x i64>} @vector_deinterleave_load_v2i64_v4i64(ptr %p) {
 ; CHECK-NEXT:    vlseg2e64.v v8, (a0)
 ; CHECK-NEXT:    ret
   %vec = load <4 x i64>, ptr %p
-  %retval = call {<2 x i64>, <2 x i64>} @llvm.experimental.vector.deinterleave2.v4i64(<4 x i64> %vec)
+  %retval = call {<2 x i64>, <2 x i64>} @llvm.vector.deinterleave2.v4i64(<4 x i64> %vec)
   ret {<2 x i64>, <2 x i64>} %retval
 }
 
-declare {<16 x i1>, <16 x i1>} @llvm.experimental.vector.deinterleave2.v32i1(<32 x i1>)
-declare {<16 x i8>, <16 x i8>} @llvm.experimental.vector.deinterleave2.v32i8(<32 x i8>)
-declare {<8 x i16>, <8 x i16>} @llvm.experimental.vector.deinterleave2.v16i16(<16 x i16>)
-declare {<4 x i32>, <4 x i32>} @llvm.experimental.vector.deinterleave2.v8i32(<8 x i32>)
-declare {<2 x i64>, <2 x i64>} @llvm.experimental.vector.deinterleave2.v4i64(<4 x i64>)
+declare {<16 x i1>, <16 x i1>} @llvm.vector.deinterleave2.v32i1(<32 x i1>)
+declare {<16 x i8>, <16 x i8>} @llvm.vector.deinterleave2.v32i8(<32 x i8>)
+declare {<8 x i16>, <8 x i16>} @llvm.vector.deinterleave2.v16i16(<16 x i16>)
+declare {<4 x i32>, <4 x i32>} @llvm.vector.deinterleave2.v8i32(<8 x i32>)
+declare {<2 x i64>, <2 x i64>} @llvm.vector.deinterleave2.v4i64(<4 x i64>)
 
 ; Floats
 
@@ -114,7 +114,7 @@ define {<2 x half>, <2 x half>} @vector_deinterleave_load_v2f16_v4f16(ptr %p) {
 ; CHECK-NEXT:    vlseg2e16.v v8, (a0)
 ; CHECK-NEXT:    ret
   %vec = load <4 x half>, ptr %p
-  %retval = call {<2 x half>, <2 x half>} @llvm.experimental.vector.deinterleave2.v4f16(<4 x half> %vec)
+  %retval = call {<2 x half>, <2 x half>} @llvm.vector.deinterleave2.v4f16(<4 x half> %vec)
   ret {<2 x half>, <2 x half>} %retval
 }
 
@@ -125,7 +125,7 @@ define {<4 x half>, <4 x half>} @vector_deinterleave_load_v4f16_v8f16(ptr %p) {
 ; CHECK-NEXT:    vlseg2e16.v v8, (a0)
 ; CHECK-NEXT:    ret
   %vec = load <8 x half>, ptr %p
-  %retval = call {<4 x half>, <4 x half>} @llvm.experimental.vector.deinterleave2.v8f16(<8 x half> %vec)
+  %retval = call {<4 x half>, <4 x half>} @llvm.vector.deinterleave2.v8f16(<8 x half> %vec)
   ret {<4 x half>, <4 x half>} %retval
 }
 
@@ -136,7 +136,7 @@ define {<2 x float>, <2 x float>} @vector_deinterleave_load_v2f32_v4f32(ptr %p)
 ; CHECK-NEXT:    vlseg2e32.v v8, (a0)
 ; CHECK-NEXT:    ret
   %vec = load <4 x float>, ptr %p
-  %retval = call {<2 x float>, <2 x float>} @llvm.experimental.vector.deinterleave2.v4f32(<4 x float> %vec)
+  %retval = call {<2 x float>, <2 x float>} @llvm.vector.deinterleave2.v4f32(<4 x float> %vec)
   ret {<2 x float>, <2 x float>} %retval
 }
 
@@ -147,7 +147,7 @@ define {<8 x half>, <8 x half>} @vector_deinterleave_load_v8f16_v16f16(ptr %p) {
 ; CHECK-NEXT:    vlseg2e16.v v8, (a0)
 ; CHECK-NEXT:    ret
   %vec = load <16 x half>, ptr %p
-  %retval = call {<8 x half>, <8 x half>} @llvm.experimental.vector.deinterleave2.v16f16(<16 x half> %vec)
+  %retval = call {<8 x half>, <8 x half>} @llvm.vector.deinterleave2.v16f16(<16 x half> %vec)
   ret {<8 x half>, <8 x half>} %retval
 }
 
@@ -158,7 +158,7 @@ define {<4 x float>, <4 x float>} @vector_deinterleave_load_v4f32_v8f32(ptr %p)
 ; CHECK-NEXT:    vlseg2e32.v v8, (a0)
 ; CHECK-NEXT:    ret
   %vec = load <8 x float>, ptr %p
-  %retval = call {<4 x float>, <4 x float>} @llvm.experimental.vector.deinterleave2.v8f32(<8 x float> %vec)
+  %retval = call {<4 x float>, <4 x float>} @llvm.vector.deinterleave2.v8f32(<8 x float> %vec)
   ret  {<4 x float>, <4 x float>} %retval
 }
 
@@ -169,13 +169,13 @@ define {<2 x double>, <2 x double>} @vector_deinterleave_load_v2f64_v4f64(ptr %p
 ; CHECK-NEXT:    vlseg2e64.v v8, (a0)
 ; CHECK-NEXT:    ret
   %vec = load <4 x double>, ptr %p
-  %retval = call {<2 x double>, <2 x double>} @llvm.experimental.vector.deinterleave2.v4f64(<4 x double> %vec)
+  %retval = call {<2 x double>, <2 x double>} @llvm.vector.deinterleave2.v4f64(<4 x double> %vec)
   ret {<2 x double>, <2 x double>} %retval
 }
 
-declare {<2 x half>,<2 x half>} @llvm.experimental.vector.deinterleave2.v4f16(<4 x half>)
-declare {<4 x half>, <4 x half>} @llvm.experimental.vector.deinterleave2.v8f16(<8 x half>)
-declare {<2 x float>, <2 x float>} @llvm.experimental.vector.deinterleave2.v4f32(<4 x float>)
-declare {<8 x half>, <8 x half>} @llvm.experimental.vector.deinterleave2.v16f16(<16 x half>)
-declare {<4 x float>, <4 x float>} @llvm.experimental.vector.deinterleave2.v8f32(<8 x float>)
-declare {<2 x double>, <2 x double>} @llvm.experimental.vector.deinterleave2.v4f64(<4 x double>)
+declare {<2 x half>,<2 x half>} @llvm.vector.deinterleave2.v4f16(<4 x half>)
+declare {<4 x half>, <4 x half>} @llvm.vector.deinterleave2.v8f16(<8 x half>)
+declare {<2 x float>, <2 x float>} @llvm.vector.deinterleave2.v4f32(<4 x float>)
+declare {<8 x half>, <8 x half>} @llvm.vector.deinterleave2.v16f16(<16 x half>)
+declare {<4 x float>, <4 x float>} @llvm.vector.deinterleave2.v8f32(<8 x float>)
+declare {<2 x double>, <2 x double>} @llvm.vector.deinterleave2.v4f64(<4 x double>)
diff --git a/llvm/test/CodeGen/RISCV/rvv/fixed-vectors-fp-buildvec.ll b/llvm/test/CodeGen/RISCV/rvv/fixed-vectors-fp-buildvec.ll
index 8e214e405478..9e83efd35195 100644
--- a/llvm/test/CodeGen/RISCV/rvv/fixed-vectors-fp-buildvec.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/fixed-vectors-fp-buildvec.ll
@@ -1407,8 +1407,8 @@ define <8 x float> @buildvec_v8f32_zvl256(float %e0, float %e1, float %e2, float
 ; CHECK-NEXT:    vfmv.v.f v8, fa4
 ; CHECK-NEXT:    vfslide1down.vf v8, v8, fa5
 ; CHECK-NEXT:    vfslide1down.vf v8, v8, fa6
-; CHECK-NEXT:    vmv.v.i v0, 15
 ; CHECK-NEXT:    vfslide1down.vf v8, v8, fa7
+; CHECK-NEXT:    vmv.v.i v0, 15
 ; CHECK-NEXT:    vslidedown.vi v8, v9, 4, v0.t
 ; CHECK-NEXT:    ret
   %v0 = insertelement <8 x float> poison, float %e0, i64 0
@@ -1458,8 +1458,8 @@ define <8 x double> @buildvec_v8f64_zvl512(double %e0, double %e1, double %e2, d
 ; CHECK-NEXT:    vfmv.v.f v8, fa4
 ; CHECK-NEXT:    vfslide1down.vf v8, v8, fa5
 ; CHECK-NEXT:    vfslide1down.vf v8, v8, fa6
-; CHECK-NEXT:    vmv.v.i v0, 15
 ; CHECK-NEXT:    vfslide1down.vf v8, v8, fa7
+; CHECK-NEXT:    vmv.v.i v0, 15
 ; CHECK-NEXT:    vslidedown.vi v8, v9, 4, v0.t
 ; CHECK-NEXT:    ret
   %v0 = insertelement <8 x double> poison, double %e0, i64 0
diff --git a/llvm/test/CodeGen/RISCV/rvv/fixed-vectors-fp-interleave.ll b/llvm/test/CodeGen/RISCV/rvv/fixed-vectors-fp-interleave.ll
index 6bfd0ac93267..ed152e64a91e 100644
--- a/llvm/test/CodeGen/RISCV/rvv/fixed-vectors-fp-interleave.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/fixed-vectors-fp-interleave.ll
@@ -57,8 +57,8 @@ define <4 x double> @interleave_v2f64(<2 x double> %x, <2 x double> %y) {
 ; RV32-V512-NEXT:    vid.v v10
 ; RV32-V512-NEXT:    vsrl.vi v11, v10, 1
 ; RV32-V512-NEXT:    vsetvli zero, zero, e64, m1, ta, mu
-; RV32-V512-NEXT:    vmv.v.i v0, 10
 ; RV32-V512-NEXT:    vrgatherei16.vv v10, v8, v11
+; RV32-V512-NEXT:    vmv.v.i v0, 10
 ; RV32-V512-NEXT:    vrgatherei16.vv v10, v9, v11, v0.t
 ; RV32-V512-NEXT:    vmv.v.v v8, v10
 ; RV32-V512-NEXT:    ret
@@ -68,8 +68,8 @@ define <4 x double> @interleave_v2f64(<2 x double> %x, <2 x double> %y) {
 ; RV64-V512-NEXT:    vsetivli zero, 4, e64, m1, ta, mu
 ; RV64-V512-NEXT:    vid.v v10
 ; RV64-V512-NEXT:    vsrl.vi v11, v10, 1
-; RV64-V512-NEXT:    vmv.v.i v0, 10
 ; RV64-V512-NEXT:    vrgather.vv v10, v8, v11
+; RV64-V512-NEXT:    vmv.v.i v0, 10
 ; RV64-V512-NEXT:    vrgather.vv v10, v9, v11, v0.t
 ; RV64-V512-NEXT:    vmv.v.v v8, v10
 ; RV64-V512-NEXT:    ret
diff --git a/llvm/test/CodeGen/RISCV/rvv/fixed-vectors-fp2i-sat.ll b/llvm/test/CodeGen/RISCV/rvv/fixed-vectors-fp2i-sat.ll
index 85b849045e8c..a8e4af2d7368 100644
--- a/llvm/test/CodeGen/RISCV/rvv/fixed-vectors-fp2i-sat.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/fixed-vectors-fp2i-sat.ll
@@ -395,8 +395,8 @@ define void @fp2si_v8f64_v8i8(ptr %x, ptr %y) {
 ; RV32-NEXT:    fmin.d fa5, fa5, fa4
 ; RV32-NEXT:    fcvt.w.d a2, fa5, rtz
 ; RV32-NEXT:    and a0, a0, a2
-; RV32-NEXT:    vmv.v.i v0, 15
 ; RV32-NEXT:    vslide1down.vx v9, v9, a0
+; RV32-NEXT:    vmv.v.i v0, 15
 ; RV32-NEXT:    vslidedown.vi v9, v8, 4, v0.t
 ; RV32-NEXT:    vse8.v v9, (a1)
 ; RV32-NEXT:    addi sp, s0, -128
@@ -496,8 +496,8 @@ define void @fp2si_v8f64_v8i8(ptr %x, ptr %y) {
 ; RV64-NEXT:    fmin.d fa5, fa5, fa4
 ; RV64-NEXT:    fcvt.l.d a2, fa5, rtz
 ; RV64-NEXT:    and a0, a0, a2
-; RV64-NEXT:    vmv.v.i v0, 15
 ; RV64-NEXT:    vslide1down.vx v9, v9, a0
+; RV64-NEXT:    vmv.v.i v0, 15
 ; RV64-NEXT:    vslidedown.vi v9, v8, 4, v0.t
 ; RV64-NEXT:    vse8.v v9, (a1)
 ; RV64-NEXT:    addi sp, s0, -128
@@ -580,8 +580,8 @@ define void @fp2ui_v8f64_v8i8(ptr %x, ptr %y) {
 ; RV32-NEXT:    fmax.d fa4, fa4, fa3
 ; RV32-NEXT:    fmin.d fa5, fa4, fa5
 ; RV32-NEXT:    fcvt.wu.d a0, fa5, rtz
-; RV32-NEXT:    vmv.v.i v0, 15
 ; RV32-NEXT:    vslide1down.vx v9, v9, a0
+; RV32-NEXT:    vmv.v.i v0, 15
 ; RV32-NEXT:    vslidedown.vi v9, v8, 4, v0.t
 ; RV32-NEXT:    vse8.v v9, (a1)
 ; RV32-NEXT:    addi sp, s0, -128
@@ -656,8 +656,8 @@ define void @fp2ui_v8f64_v8i8(ptr %x, ptr %y) {
 ; RV64-NEXT:    fmax.d fa4, fa4, fa3
 ; RV64-NEXT:    fmin.d fa5, fa4, fa5
 ; RV64-NEXT:    fcvt.lu.d a0, fa5, rtz
-; RV64-NEXT:    vmv.v.i v0, 15
 ; RV64-NEXT:    vslide1down.vx v9, v9, a0
+; RV64-NEXT:    vmv.v.i v0, 15
 ; RV64-NEXT:    vslidedown.vi v9, v8, 4, v0.t
 ; RV64-NEXT:    vse8.v v9, (a1)
 ; RV64-NEXT:    addi sp, s0, -128
diff --git a/llvm/test/CodeGen/RISCV/rvv/fixed-vectors-int-interleave.ll b/llvm/test/CodeGen/RISCV/rvv/fixed-vectors-int-interleave.ll
index 6da83644413b..40ff8b50d99d 100644
--- a/llvm/test/CodeGen/RISCV/rvv/fixed-vectors-int-interleave.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/fixed-vectors-int-interleave.ll
@@ -70,8 +70,8 @@ define <4 x i64> @interleave_v2i64(<2 x i64> %x, <2 x i64> %y) {
 ; RV32-V512-NEXT:    vid.v v10
 ; RV32-V512-NEXT:    vsrl.vi v11, v10, 1
 ; RV32-V512-NEXT:    vsetvli zero, zero, e64, m1, ta, mu
-; RV32-V512-NEXT:    vmv.v.i v0, 10
 ; RV32-V512-NEXT:    vrgatherei16.vv v10, v8, v11
+; RV32-V512-NEXT:    vmv.v.i v0, 10
 ; RV32-V512-NEXT:    vrgatherei16.vv v10, v9, v11, v0.t
 ; RV32-V512-NEXT:    vmv.v.v v8, v10
 ; RV32-V512-NEXT:    ret
@@ -81,8 +81,8 @@ define <4 x i64> @interleave_v2i64(<2 x i64> %x, <2 x i64> %y) {
 ; RV64-V512-NEXT:    vsetivli zero, 4, e64, m1, ta, mu
 ; RV64-V512-NEXT:    vid.v v10
 ; RV64-V512-NEXT:    vsrl.vi v11, v10, 1
-; RV64-V512-NEXT:    vmv.v.i v0, 10
 ; RV64-V512-NEXT:    vrgather.vv v10, v8, v11
+; RV64-V512-NEXT:    vmv.v.i v0, 10
 ; RV64-V512-NEXT:    vrgather.vv v10, v9, v11, v0.t
 ; RV64-V512-NEXT:    vmv.v.v v8, v10
 ; RV64-V512-NEXT:    ret
@@ -195,8 +195,8 @@ define <4 x i32> @interleave_v4i32_offset_1(<4 x i32> %x, <4 x i32> %y) {
 ; V128-NEXT:    vsetivli zero, 4, e32, m1, ta, mu
 ; V128-NEXT:    vid.v v8
 ; V128-NEXT:    vsrl.vi v8, v8, 1
-; V128-NEXT:    vmv.v.i v0, 10
 ; V128-NEXT:    vadd.vi v8, v8, 1
+; V128-NEXT:    vmv.v.i v0, 10
 ; V128-NEXT:    vrgather.vv v10, v9, v8, v0.t
 ; V128-NEXT:    vmv.v.v v8, v10
 ; V128-NEXT:    ret
@@ -210,8 +210,8 @@ define <4 x i32> @interleave_v4i32_offset_1(<4 x i32> %x, <4 x i32> %y) {
 ; V512-NEXT:    vsetivli zero, 4, e32, mf2, ta, mu
 ; V512-NEXT:    vid.v v8
 ; V512-NEXT:    vsrl.vi v8, v8, 1
-; V512-NEXT:    vmv.v.i v0, 10
 ; V512-NEXT:    vadd.vi v8, v8, 1
+; V512-NEXT:    vmv.v.i v0, 10
 ; V512-NEXT:    vrgather.vv v10, v9, v8, v0.t
 ; V512-NEXT:    vmv1r.v v8, v10
 ; V512-NEXT:    ret
diff --git a/llvm/test/CodeGen/RISCV/rvv/fixed-vectors-int-shuffles.ll b/llvm/test/CodeGen/RISCV/rvv/fixed-vectors-int-shuffles.ll
index 0e8d9cf03066..58af6ac246d1 100644
--- a/llvm/test/CodeGen/RISCV/rvv/fixed-vectors-int-shuffles.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/fixed-vectors-int-shuffles.ll
@@ -89,8 +89,8 @@ define <4 x i16> @vrgather_shuffle_vv_v4i16(<4 x i16> %x, <4 x i16> %y) {
 ; CHECK-NEXT:    addi a0, a0, %lo(.LCPI6_0)
 ; CHECK-NEXT:    vsetivli zero, 4, e16, mf2, ta, mu
 ; CHECK-NEXT:    vle16.v v11, (a0)
-; CHECK-NEXT:    vmv.v.i v0, 8
 ; CHECK-NEXT:    vrgather.vv v10, v8, v11
+; CHECK-NEXT:    vmv.v.i v0, 8
 ; CHECK-NEXT:    vrgather.vi v10, v9, 1, v0.t
 ; CHECK-NEXT:    vmv1r.v v8, v10
 ; CHECK-NEXT:    ret
@@ -162,16 +162,16 @@ define <8 x i64> @vrgather_shuffle_vv_v8i64(<8 x i64> %x, <8 x i64> %y) {
 ; RV32:       # %bb.0:
 ; RV32-NEXT:    vsetivli zero, 8, e16, m1, ta, ma
 ; RV32-NEXT:    vmv.v.i v16, 2
-; RV32-NEXT:    li a0, 5
-; RV32-NEXT:    vslide1down.vx v20, v16, a0
 ; RV32-NEXT:    lui a0, %hi(.LCPI11_0)
 ; RV32-NEXT:    addi a0, a0, %lo(.LCPI11_0)
-; RV32-NEXT:    vle16.v v21, (a0)
+; RV32-NEXT:    vle16.v v20, (a0)
+; RV32-NEXT:    li a0, 5
+; RV32-NEXT:    vslide1down.vx v21, v16, a0
 ; RV32-NEXT:    vsetvli zero, zero, e64, m4, ta, mu
+; RV32-NEXT:    vrgatherei16.vv v16, v8, v20
 ; RV32-NEXT:    li a0, 164
 ; RV32-NEXT:    vmv.s.x v0, a0
-; RV32-NEXT:    vrgatherei16.vv v16, v8, v21
-; RV32-NEXT:    vrgatherei16.vv v16, v12, v20, v0.t
+; RV32-NEXT:    vrgatherei16.vv v16, v12, v21, v0.t
 ; RV32-NEXT:    vmv.v.v v8, v16
 ; RV32-NEXT:    ret
 ;
@@ -210,13 +210,13 @@ define <8 x i64> @vrgather_shuffle_xv_v8i64(<8 x i64> %x) {
 ; RV32-NEXT:    vsetivli zero, 8, e64, m4, ta, mu
 ; RV32-NEXT:    vle16.v v16, (a0)
 ; RV32-NEXT:    vmv.v.i v20, -1
+; RV32-NEXT:    vrgatherei16.vv v12, v20, v16
 ; RV32-NEXT:    lui a0, %hi(.LCPI12_1)
 ; RV32-NEXT:    addi a0, a0, %lo(.LCPI12_1)
-; RV32-NEXT:    vle16.v v17, (a0)
+; RV32-NEXT:    vle16.v v16, (a0)
 ; RV32-NEXT:    li a0, 113
 ; RV32-NEXT:    vmv.s.x v0, a0
-; RV32-NEXT:    vrgatherei16.vv v12, v20, v16
-; RV32-NEXT:    vrgatherei16.vv v12, v8, v17, v0.t
+; RV32-NEXT:    vrgatherei16.vv v12, v8, v16, v0.t
 ; RV32-NEXT:    vmv.v.v v8, v12
 ; RV32-NEXT:    ret
 ;
@@ -368,9 +368,9 @@ define <8 x i8> @splat_ve2_we0(<8 x i8> %v, <8 x i8> %w) {
 ; CHECK-LABEL: splat_ve2_we0:
 ; CHECK:       # %bb.0:
 ; CHECK-NEXT:    vsetivli zero, 8, e8, mf2, ta, mu
+; CHECK-NEXT:    vrgather.vi v10, v8, 2
 ; CHECK-NEXT:    li a0, 66
 ; CHECK-NEXT:    vmv.s.x v0, a0
-; CHECK-NEXT:    vrgather.vi v10, v8, 2
 ; CHECK-NEXT:    vrgather.vi v10, v9, 0, v0.t
 ; CHECK-NEXT:    vmv1r.v v8, v10
 ; CHECK-NEXT:    ret
@@ -387,9 +387,9 @@ define <8 x i8> @splat_ve2_we0_ins_i0ve4(<8 x i8> %v, <8 x i8> %w) {
 ; CHECK-NEXT:    vsetvli zero, zero, e8, mf2, tu, ma
 ; CHECK-NEXT:    vmv.s.x v11, a0
 ; CHECK-NEXT:    vsetvli zero, zero, e8, mf2, ta, mu
+; CHECK-NEXT:    vrgather.vv v10, v8, v11
 ; CHECK-NEXT:    li a0, 66
 ; CHECK-NEXT:    vmv.s.x v0, a0
-; CHECK-NEXT:    vrgather.vv v10, v8, v11
 ; CHECK-NEXT:    vrgather.vi v10, v9, 0, v0.t
 ; CHECK-NEXT:    vmv1r.v v8, v10
 ; CHECK-NEXT:    ret
@@ -422,9 +422,9 @@ define <8 x i8> @splat_ve2_we0_ins_i2ve4(<8 x i8> %v, <8 x i8> %w) {
 ; CHECK-NEXT:    vsetivli zero, 2, e32, mf2, ta, ma
 ; CHECK-NEXT:    vmv.v.x v11, a0
 ; CHECK-NEXT:    vsetivli zero, 8, e8, mf2, ta, mu
+; CHECK-NEXT:    vrgather.vv v10, v8, v11
 ; CHECK-NEXT:    li a0, 66
 ; CHECK-NEXT:    vmv.s.x v0, a0
-; CHECK-NEXT:    vrgather.vv v10, v8, v11
 ; CHECK-NEXT:    vrgather.vi v10, v9, 0, v0.t
 ; CHECK-NEXT:    vmv1r.v v8, v10
 ; CHECK-NEXT:    ret
@@ -441,9 +441,9 @@ define <8 x i8> @splat_ve2_we0_ins_i2we4(<8 x i8> %v, <8 x i8> %w) {
 ; CHECK-NEXT:    vsetivli zero, 3, e8, mf2, tu, ma
 ; CHECK-NEXT:    vslideup.vi v11, v10, 2
 ; CHECK-NEXT:    vsetivli zero, 8, e8, mf2, ta, mu
+; CHECK-NEXT:    vrgather.vi v10, v8, 2
 ; CHECK-NEXT:    li a0, 70
 ; CHECK-NEXT:    vmv.s.x v0, a0
-; CHECK-NEXT:    vrgather.vi v10, v8, 2
 ; CHECK-NEXT:    vrgather.vv v10, v9, v11, v0.t
 ; CHECK-NEXT:    vmv1r.v v8, v10
 ; CHECK-NEXT:    ret
@@ -464,9 +464,9 @@ define <8 x i8> @splat_ve2_we0_ins_i2ve4_i5we6(<8 x i8> %v, <8 x i8> %w) {
 ; CHECK-NEXT:    vsetivli zero, 2, e32, mf2, ta, ma
 ; CHECK-NEXT:    vmv.v.x v12, a0
 ; CHECK-NEXT:    vsetivli zero, 8, e8, mf2, ta, mu
+; CHECK-NEXT:    vrgather.vv v10, v8, v12
 ; CHECK-NEXT:    li a0, 98
 ; CHECK-NEXT:    vmv.s.x v0, a0
-; CHECK-NEXT:    vrgather.vv v10, v8, v12
 ; CHECK-NEXT:    vrgather.vv v10, v9, v11, v0.t
 ; CHECK-NEXT:    vmv1r.v v8, v10
 ; CHECK-NEXT:    ret
@@ -681,9 +681,9 @@ define <8 x i8> @merge_non_contiguous_slideup_slidedown(<8 x i8> %v, <8 x i8> %w
 ; CHECK-LABEL: merge_non_contiguous_slideup_slidedown:
 ; CHECK:       # %bb.0:
 ; CHECK-NEXT:    vsetivli zero, 8, e8, mf2, ta, mu
+; CHECK-NEXT:    vslidedown.vi v8, v8, 2
 ; CHECK-NEXT:    li a0, 234
 ; CHECK-NEXT:    vmv.s.x v0, a0
-; CHECK-NEXT:    vslidedown.vi v8, v8, 2
 ; CHECK-NEXT:    vslideup.vi v8, v9, 1, v0.t
 ; CHECK-NEXT:    ret
   %res = shufflevector <8 x i8> %v, <8 x i8> %w, <8 x i32> <i32 2, i32 8, i32 4, i32 10, i32 6, i32 12, i32 13, i32 14>
@@ -695,12 +695,12 @@ define <8 x i8> @unmergable(<8 x i8> %v, <8 x i8> %w) {
 ; CHECK-LABEL: unmergable:
 ; CHECK:       # %bb.0:
 ; CHECK-NEXT:    vsetivli zero, 8, e8, mf2, ta, mu
+; CHECK-NEXT:    vslidedown.vi v8, v8, 2
 ; CHECK-NEXT:    lui a0, %hi(.LCPI46_0)
 ; CHECK-NEXT:    addi a0, a0, %lo(.LCPI46_0)
 ; CHECK-NEXT:    vle8.v v10, (a0)
 ; CHECK-NEXT:    li a0, 234
 ; CHECK-NEXT:    vmv.s.x v0, a0
-; CHECK-NEXT:    vslidedown.vi v8, v8, 2
 ; CHECK-NEXT:    vrgather.vv v8, v9, v10, v0.t
 ; CHECK-NEXT:    ret
   %res = shufflevector <8 x i8> %v, <8 x i8> %w, <8 x i32> <i32 2, i32 9, i32 4, i32 11, i32 6, i32 13, i32 8, i32 15>
diff --git a/llvm/test/CodeGen/RISCV/rvv/fixed-vectors-interleave-store.ll b/llvm/test/CodeGen/RISCV/rvv/fixed-vectors-interleave-store.ll
index 9161cedd58e3..8de9cc25ae09 100644
--- a/llvm/test/CodeGen/RISCV/rvv/fixed-vectors-interleave-store.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/fixed-vectors-interleave-store.ll
@@ -23,7 +23,7 @@ define void @vector_interleave_store_v32i1_v16i1(<16 x i1> %a, <16 x i1> %b, ptr
 ; CHECK-NEXT:    vmsne.vi v8, v12, 0
 ; CHECK-NEXT:    vsm.v v8, (a0)
 ; CHECK-NEXT:    ret
-  %res = call <32 x i1> @llvm.experimental.vector.interleave2.v32i1(<16 x i1> %a, <16 x i1> %b)
+  %res = call <32 x i1> @llvm.vector.interleave2.v32i1(<16 x i1> %a, <16 x i1> %b)
   store <32 x i1> %res, ptr %p
   ret void
 }
@@ -40,7 +40,7 @@ define void @vector_interleave_store_v16i16_v8i16_align1(<8 x i16> %a, <8 x i16>
 ; CHECK-NEXT:    vsetvli zero, a1, e8, m2, ta, ma
 ; CHECK-NEXT:    vse8.v v10, (a0)
 ; CHECK-NEXT:    ret
-  %res = call <16 x i16> @llvm.experimental.vector.interleave2.v16i16(<8 x i16> %a, <8 x i16> %b)
+  %res = call <16 x i16> @llvm.vector.interleave2.v16i16(<8 x i16> %a, <8 x i16> %b)
   store <16 x i16> %res, ptr %p, align 1
   ret void
 }
@@ -51,7 +51,7 @@ define void @vector_interleave_store_v16i16_v8i16(<8 x i16> %a, <8 x i16> %b, pt
 ; CHECK-NEXT:    vsetivli zero, 8, e16, m1, ta, ma
 ; CHECK-NEXT:    vsseg2e16.v v8, (a0)
 ; CHECK-NEXT:    ret
-  %res = call <16 x i16> @llvm.experimental.vector.interleave2.v16i16(<8 x i16> %a, <8 x i16> %b)
+  %res = call <16 x i16> @llvm.vector.interleave2.v16i16(<8 x i16> %a, <8 x i16> %b)
   store <16 x i16> %res, ptr %p
   ret void
 }
@@ -62,7 +62,7 @@ define void @vector_interleave_store_v8i32_v4i32(<4 x i32> %a, <4 x i32> %b, ptr
 ; CHECK-NEXT:    vsetivli zero, 4, e32, m1, ta, ma
 ; CHECK-NEXT:    vsseg2e32.v v8, (a0)
 ; CHECK-NEXT:    ret
-  %res = call <8 x i32> @llvm.experimental.vector.interleave2.v8i32(<4 x i32> %a, <4 x i32> %b)
+  %res = call <8 x i32> @llvm.vector.interleave2.v8i32(<4 x i32> %a, <4 x i32> %b)
   store <8 x i32> %res, ptr %p
   ret void
 }
@@ -73,15 +73,15 @@ define void @vector_interleave_store_v4i64_v2i64(<2 x i64> %a, <2 x i64> %b, ptr
 ; CHECK-NEXT:    vsetivli zero, 2, e64, m1, ta, ma
 ; CHECK-NEXT:    vsseg2e64.v v8, (a0)
 ; CHECK-NEXT:    ret
-  %res = call <4 x i64> @llvm.experimental.vector.interleave2.v4i64(<2 x i64> %a, <2 x i64> %b)
+  %res = call <4 x i64> @llvm.vector.interleave2.v4i64(<2 x i64> %a, <2 x i64> %b)
   store <4 x i64> %res, ptr %p
   ret void
 }
 
-declare <32 x i1> @llvm.experimental.vector.interleave2.v32i1(<16 x i1>, <16 x i1>)
-declare <16 x i16> @llvm.experimental.vector.interleave2.v16i16(<8 x i16>, <8 x i16>)
-declare <8 x i32> @llvm.experimental.vector.interleave2.v8i32(<4 x i32>, <4 x i32>)
-declare <4 x i64> @llvm.experimental.vector.interleave2.v4i64(<2 x i64>, <2 x i64>)
+declare <32 x i1> @llvm.vector.interleave2.v32i1(<16 x i1>, <16 x i1>)
+declare <16 x i16> @llvm.vector.interleave2.v16i16(<8 x i16>, <8 x i16>)
+declare <8 x i32> @llvm.vector.interleave2.v8i32(<4 x i32>, <4 x i32>)
+declare <4 x i64> @llvm.vector.interleave2.v4i64(<2 x i64>, <2 x i64>)
 
 ; Floats
 
@@ -91,7 +91,7 @@ define void @vector_interleave_store_v4f16_v2f16(<2 x half> %a, <2 x half> %b, p
 ; CHECK-NEXT:    vsetivli zero, 2, e16, mf4, ta, ma
 ; CHECK-NEXT:    vsseg2e16.v v8, (a0)
 ; CHECK-NEXT:    ret
-  %res = call <4 x half> @llvm.experimental.vector.interleave2.v4f16(<2 x half> %a, <2 x half> %b)
+  %res = call <4 x half> @llvm.vector.interleave2.v4f16(<2 x half> %a, <2 x half> %b)
   store <4 x half> %res, ptr %p
   ret void
 }
@@ -102,7 +102,7 @@ define void @vector_interleave_store_v8f16_v4f16(<4 x half> %a, <4 x half> %b, p
 ; CHECK-NEXT:    vsetivli zero, 4, e16, mf2, ta, ma
 ; CHECK-NEXT:    vsseg2e16.v v8, (a0)
 ; CHECK-NEXT:    ret
-  %res = call <8 x half> @llvm.experimental.vector.interleave2.v8f16(<4 x half> %a, <4 x half> %b)
+  %res = call <8 x half> @llvm.vector.interleave2.v8f16(<4 x half> %a, <4 x half> %b)
   store <8 x half> %res, ptr %p
   ret void
 }
@@ -113,7 +113,7 @@ define void @vector_interleave_store_v4f32_v2f32(<2 x float> %a, <2 x float> %b,
 ; CHECK-NEXT:    vsetivli zero, 2, e32, mf2, ta, ma
 ; CHECK-NEXT:    vsseg2e32.v v8, (a0)
 ; CHECK-NEXT:    ret
-  %res = call <4 x float> @llvm.experimental.vector.interleave2.v4f32(<2 x float> %a, <2 x float> %b)
+  %res = call <4 x float> @llvm.vector.interleave2.v4f32(<2 x float> %a, <2 x float> %b)
   store <4 x float> %res, ptr %p
   ret void
 }
@@ -124,7 +124,7 @@ define void @vector_interleave_store_v16f16_v8f16(<8 x half> %a, <8 x half> %b,
 ; CHECK-NEXT:    vsetivli zero, 8, e16, m1, ta, ma
 ; CHECK-NEXT:    vsseg2e16.v v8, (a0)
 ; CHECK-NEXT:    ret
-  %res = call <16 x half> @llvm.experimental.vector.interleave2.v16f16(<8 x half> %a, <8 x half> %b)
+  %res = call <16 x half> @llvm.vector.interleave2.v16f16(<8 x half> %a, <8 x half> %b)
   store <16 x half> %res, ptr %p
   ret void
 }
@@ -135,7 +135,7 @@ define void @vector_interleave_store_v8f32_v4f32(<4 x float> %a, <4 x float> %b,
 ; CHECK-NEXT:    vsetivli zero, 4, e32, m1, ta, ma
 ; CHECK-NEXT:    vsseg2e32.v v8, (a0)
 ; CHECK-NEXT:    ret
-  %res = call <8 x float> @llvm.experimental.vector.interleave2.v8f32(<4 x float> %a, <4 x float> %b)
+  %res = call <8 x float> @llvm.vector.interleave2.v8f32(<4 x float> %a, <4 x float> %b)
   store <8 x float> %res, ptr %p
   ret void
 }
@@ -146,15 +146,15 @@ define void @vector_interleave_store_v4f64_v2f64(<2 x double> %a, <2 x double> %
 ; CHECK-NEXT:    vsetivli zero, 2, e64, m1, ta, ma
 ; CHECK-NEXT:    vsseg2e64.v v8, (a0)
 ; CHECK-NEXT:    ret
-  %res = call <4 x double> @llvm.experimental.vector.interleave2.v4f64(<2 x double> %a, <2 x double> %b)
+  %res = call <4 x double> @llvm.vector.interleave2.v4f64(<2 x double> %a, <2 x double> %b)
   store <4 x double> %res, ptr %p
   ret void
 }
 
 
-declare <4 x half> @llvm.experimental.vector.interleave2.v4f16(<2 x half>, <2 x half>)
-declare <8 x half> @llvm.experimental.vector.interleave2.v8f16(<4 x half>, <4 x half>)
-declare <4 x float> @llvm.experimental.vector.interleave2.v4f32(<2 x float>, <2 x float>)
-declare <16 x half> @llvm.experimental.vector.interleave2.v16f16(<8 x half>, <8 x half>)
-declare <8 x float> @llvm.experimental.vector.interleave2.v8f32(<4 x float>, <4 x float>)
-declare <4 x double> @llvm.experimental.vector.interleave2.v4f64(<2 x double>, <2 x double>)
+declare <4 x half> @llvm.vector.interleave2.v4f16(<2 x half>, <2 x half>)
+declare <8 x half> @llvm.vector.interleave2.v8f16(<4 x half>, <4 x half>)
+declare <4 x float> @llvm.vector.interleave2.v4f32(<2 x float>, <2 x float>)
+declare <16 x half> @llvm.vector.interleave2.v16f16(<8 x half>, <8 x half>)
+declare <8 x float> @llvm.vector.interleave2.v8f32(<4 x float>, <4 x float>)
+declare <4 x double> @llvm.vector.interleave2.v4f64(<2 x double>, <2 x double>)
diff --git a/llvm/test/CodeGen/RISCV/rvv/fixed-vectors-interleaved-access.ll b/llvm/test/CodeGen/RISCV/rvv/fixed-vectors-interleaved-access.ll
index f98cb343a2ab..99364264de82 100644
--- a/llvm/test/CodeGen/RISCV/rvv/fixed-vectors-interleaved-access.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/fixed-vectors-interleaved-access.ll
@@ -159,16 +159,17 @@ define {<8 x i64>, <8 x i64>, <8 x i64>, <8 x i64>, <8 x i64>, <8 x i64>} @load_
 ; RV32-NEXT:    addi sp, sp, -16
 ; RV32-NEXT:    .cfi_def_cfa_offset 16
 ; RV32-NEXT:    csrr a2, vlenb
-; RV32-NEXT:    li a3, 56
+; RV32-NEXT:    li a3, 54
 ; RV32-NEXT:    mul a2, a2, a3
 ; RV32-NEXT:    sub sp, sp, a2
-; RV32-NEXT:    .cfi_escape 0x0f, 0x0d, 0x72, 0x00, 0x11, 0x10, 0x22, 0x11, 0x38, 0x92, 0xa2, 0x38, 0x00, 0x1e, 0x22 # sp + 16 + 56 * vlenb
+; RV32-NEXT:    .cfi_escape 0x0f, 0x0d, 0x72, 0x00, 0x11, 0x10, 0x22, 0x11, 0x36, 0x92, 0xa2, 0x38, 0x00, 0x1e, 0x22 # sp + 16 + 54 * vlenb
 ; RV32-NEXT:    addi a3, a1, 256
 ; RV32-NEXT:    li a2, 32
 ; RV32-NEXT:    vsetvli zero, a2, e32, m8, ta, ma
 ; RV32-NEXT:    vle32.v v16, (a3)
 ; RV32-NEXT:    csrr a3, vlenb
-; RV32-NEXT:    slli a3, a3, 5
+; RV32-NEXT:    li a4, 21
+; RV32-NEXT:    mul a3, a3, a4
 ; RV32-NEXT:    add a3, sp, a3
 ; RV32-NEXT:    addi a3, a3, 16
 ; RV32-NEXT:    vs8r.v v16, (a3) # Unknown-size Folded Spill
@@ -176,31 +177,30 @@ define {<8 x i64>, <8 x i64>, <8 x i64>, <8 x i64>, <8 x i64>, <8 x i64>} @load_
 ; RV32-NEXT:    vsetivli zero, 16, e32, m4, ta, ma
 ; RV32-NEXT:    vslideup.vi v8, v16, 4
 ; RV32-NEXT:    csrr a4, vlenb
-; RV32-NEXT:    slli a4, a4, 4
+; RV32-NEXT:    slli a5, a4, 3
+; RV32-NEXT:    add a4, a5, a4
 ; RV32-NEXT:    add a4, sp, a4
 ; RV32-NEXT:    addi a4, a4, 16
 ; RV32-NEXT:    vs4r.v v8, (a4) # Unknown-size Folded Spill
 ; RV32-NEXT:    lui a4, 12
-; RV32-NEXT:    vmv.s.x v3, a4
+; RV32-NEXT:    vmv.s.x v0, a4
+; RV32-NEXT:    csrr a4, vlenb
+; RV32-NEXT:    add a4, sp, a4
+; RV32-NEXT:    addi a4, a4, 16
+; RV32-NEXT:    vs1r.v v0, (a4) # Unknown-size Folded Spill
 ; RV32-NEXT:    vsetivli zero, 16, e32, m8, ta, ma
 ; RV32-NEXT:    vslidedown.vi v16, v16, 16
 ; RV32-NEXT:    csrr a4, vlenb
-; RV32-NEXT:    li a5, 24
+; RV32-NEXT:    li a5, 37
 ; RV32-NEXT:    mul a4, a4, a5
 ; RV32-NEXT:    add a4, sp, a4
 ; RV32-NEXT:    addi a4, a4, 16
 ; RV32-NEXT:    vs8r.v v16, (a4) # Unknown-size Folded Spill
 ; RV32-NEXT:    vsetivli zero, 16, e32, m4, ta, mu
-; RV32-NEXT:    vmv1r.v v0, v3
-; RV32-NEXT:    csrr a4, vlenb
-; RV32-NEXT:    slli a4, a4, 2
-; RV32-NEXT:    add a4, sp, a4
-; RV32-NEXT:    addi a4, a4, 16
-; RV32-NEXT:    vs1r.v v3, (a4) # Unknown-size Folded Spill
 ; RV32-NEXT:    vslideup.vi v8, v16, 10, v0.t
 ; RV32-NEXT:    csrr a4, vlenb
-; RV32-NEXT:    li a5, 20
-; RV32-NEXT:    mul a4, a4, a5
+; RV32-NEXT:    slli a5, a4, 4
+; RV32-NEXT:    add a4, a5, a4
 ; RV32-NEXT:    add a4, sp, a4
 ; RV32-NEXT:    addi a4, a4, 16
 ; RV32-NEXT:    vs4r.v v8, (a4) # Unknown-size Folded Spill
@@ -209,71 +209,82 @@ define {<8 x i64>, <8 x i64>, <8 x i64>, <8 x i64>, <8 x i64>, <8 x i64>} @load_
 ; RV32-NEXT:    vsetvli zero, a2, e32, m8, ta, mu
 ; RV32-NEXT:    vle16.v v8, (a4)
 ; RV32-NEXT:    csrr a4, vlenb
-; RV32-NEXT:    slli a4, a4, 3
+; RV32-NEXT:    li a5, 13
+; RV32-NEXT:    mul a4, a4, a5
 ; RV32-NEXT:    add a4, sp, a4
 ; RV32-NEXT:    addi a4, a4, 16
 ; RV32-NEXT:    vs4r.v v8, (a4) # Unknown-size Folded Spill
-; RV32-NEXT:    lui a4, %hi(.LCPI6_1)
-; RV32-NEXT:    addi a4, a4, %lo(.LCPI6_1)
-; RV32-NEXT:    lui a5, 1
-; RV32-NEXT:    vle16.v v8, (a4)
-; RV32-NEXT:    addi a4, sp, 16
-; RV32-NEXT:    vs4r.v v8, (a4) # Unknown-size Folded Spill
-; RV32-NEXT:    vle32.v v16, (a1)
+; RV32-NEXT:    vle32.v v24, (a1)
 ; RV32-NEXT:    csrr a1, vlenb
-; RV32-NEXT:    li a4, 40
+; RV32-NEXT:    li a4, 45
 ; RV32-NEXT:    mul a1, a1, a4
 ; RV32-NEXT:    add a1, sp, a1
 ; RV32-NEXT:    addi a1, a1, 16
-; RV32-NEXT:    vs8r.v v16, (a1) # Unknown-size Folded Spill
-; RV32-NEXT:    vle32.v v24, (a3)
+; RV32-NEXT:    vs8r.v v24, (a1) # Unknown-size Folded Spill
+; RV32-NEXT:    lui a1, %hi(.LCPI6_1)
+; RV32-NEXT:    addi a1, a1, %lo(.LCPI6_1)
+; RV32-NEXT:    lui a4, 1
+; RV32-NEXT:    addi a4, a4, -64
+; RV32-NEXT:    vle16.v v8, (a1)
 ; RV32-NEXT:    csrr a1, vlenb
-; RV32-NEXT:    li a3, 48
-; RV32-NEXT:    mul a1, a1, a3
+; RV32-NEXT:    slli a5, a1, 2
+; RV32-NEXT:    add a1, a5, a1
 ; RV32-NEXT:    add a1, sp, a1
 ; RV32-NEXT:    addi a1, a1, 16
-; RV32-NEXT:    vs8r.v v24, (a1) # Unknown-size Folded Spill
-; RV32-NEXT:    addi a1, a5, -64
-; RV32-NEXT:    vmv.s.x v0, a1
+; RV32-NEXT:    vs4r.v v8, (a1) # Unknown-size Folded Spill
+; RV32-NEXT:    vle32.v v16, (a3)
 ; RV32-NEXT:    csrr a1, vlenb
-; RV32-NEXT:    li a3, 12
+; RV32-NEXT:    li a3, 29
 ; RV32-NEXT:    mul a1, a1, a3
 ; RV32-NEXT:    add a1, sp, a1
 ; RV32-NEXT:    addi a1, a1, 16
-; RV32-NEXT:    vs1r.v v0, (a1) # Unknown-size Folded Spill
+; RV32-NEXT:    vs8r.v v16, (a1) # Unknown-size Folded Spill
+; RV32-NEXT:    vmv.s.x v2, a4
 ; RV32-NEXT:    csrr a1, vlenb
-; RV32-NEXT:    slli a1, a1, 3
+; RV32-NEXT:    li a3, 13
+; RV32-NEXT:    mul a1, a1, a3
 ; RV32-NEXT:    add a1, sp, a1
 ; RV32-NEXT:    addi a1, a1, 16
 ; RV32-NEXT:    vl4r.v v4, (a1) # Unknown-size Folded Reload
-; RV32-NEXT:    vrgatherei16.vv v8, v16, v4
-; RV32-NEXT:    addi a1, sp, 16
-; RV32-NEXT:    vl4r.v v16, (a1) # Unknown-size Folded Reload
-; RV32-NEXT:    vrgatherei16.vv v8, v24, v16, v0.t
+; RV32-NEXT:    vrgatherei16.vv v8, v24, v4
+; RV32-NEXT:    vmv1r.v v0, v2
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    slli a3, a1, 2
+; RV32-NEXT:    add a1, a3, a1
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    addi a1, a1, 16
+; RV32-NEXT:    vl4r.v v24, (a1) # Unknown-size Folded Reload
+; RV32-NEXT:    vrgatherei16.vv v8, v16, v24, v0.t
 ; RV32-NEXT:    vsetivli zero, 12, e32, m4, tu, ma
 ; RV32-NEXT:    csrr a1, vlenb
-; RV32-NEXT:    li a3, 20
-; RV32-NEXT:    mul a1, a1, a3
+; RV32-NEXT:    slli a3, a1, 4
+; RV32-NEXT:    add a1, a3, a1
 ; RV32-NEXT:    add a1, sp, a1
 ; RV32-NEXT:    addi a1, a1, 16
 ; RV32-NEXT:    vl4r.v v12, (a1) # Unknown-size Folded Reload
 ; RV32-NEXT:    vmv.v.v v12, v8
 ; RV32-NEXT:    csrr a1, vlenb
-; RV32-NEXT:    li a3, 20
-; RV32-NEXT:    mul a1, a1, a3
+; RV32-NEXT:    slli a3, a1, 4
+; RV32-NEXT:    add a1, a3, a1
 ; RV32-NEXT:    add a1, sp, a1
 ; RV32-NEXT:    addi a1, a1, 16
 ; RV32-NEXT:    vs4r.v v12, (a1) # Unknown-size Folded Spill
 ; RV32-NEXT:    vsetivli zero, 16, e32, m4, ta, mu
 ; RV32-NEXT:    csrr a1, vlenb
-; RV32-NEXT:    slli a1, a1, 5
+; RV32-NEXT:    li a3, 21
+; RV32-NEXT:    mul a1, a1, a3
 ; RV32-NEXT:    add a1, sp, a1
 ; RV32-NEXT:    addi a1, a1, 16
-; RV32-NEXT:    vl8r.v v16, (a1) # Unknown-size Folded Reload
+; RV32-NEXT:    vl8r.v v8, (a1) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv4r.v v16, v8
 ; RV32-NEXT:    vslideup.vi v8, v16, 2
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    addi a1, a1, 16
+; RV32-NEXT:    vl1r.v v3, (a1) # Unknown-size Folded Reload
 ; RV32-NEXT:    vmv1r.v v0, v3
 ; RV32-NEXT:    csrr a1, vlenb
-; RV32-NEXT:    li a3, 24
+; RV32-NEXT:    li a3, 37
 ; RV32-NEXT:    mul a1, a1, a3
 ; RV32-NEXT:    add a1, sp, a1
 ; RV32-NEXT:    addi a1, a1, 16
@@ -283,36 +294,45 @@ define {<8 x i64>, <8 x i64>, <8 x i64>, <8 x i64>, <8 x i64>, <8 x i64>} @load_
 ; RV32-NEXT:    lui a1, %hi(.LCPI6_2)
 ; RV32-NEXT:    addi a1, a1, %lo(.LCPI6_2)
 ; RV32-NEXT:    vsetvli zero, a2, e32, m8, ta, mu
-; RV32-NEXT:    lui a3, %hi(.LCPI6_3)
-; RV32-NEXT:    addi a3, a3, %lo(.LCPI6_3)
-; RV32-NEXT:    vle16.v v24, (a1)
-; RV32-NEXT:    vle16.v v8, (a3)
+; RV32-NEXT:    vle16.v v8, (a1)
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    li a3, 13
+; RV32-NEXT:    mul a1, a1, a3
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    addi a1, a1, 16
+; RV32-NEXT:    vs4r.v v8, (a1) # Unknown-size Folded Spill
+; RV32-NEXT:    lui a1, %hi(.LCPI6_3)
+; RV32-NEXT:    addi a1, a1, %lo(.LCPI6_3)
+; RV32-NEXT:    vle16.v v8, (a1)
 ; RV32-NEXT:    csrr a1, vlenb
-; RV32-NEXT:    slli a1, a1, 3
+; RV32-NEXT:    slli a3, a1, 2
+; RV32-NEXT:    add a1, a3, a1
 ; RV32-NEXT:    add a1, sp, a1
 ; RV32-NEXT:    addi a1, a1, 16
 ; RV32-NEXT:    vs4r.v v8, (a1) # Unknown-size Folded Spill
 ; RV32-NEXT:    csrr a1, vlenb
-; RV32-NEXT:    li a3, 40
+; RV32-NEXT:    li a3, 45
 ; RV32-NEXT:    mul a1, a1, a3
 ; RV32-NEXT:    add a1, sp, a1
 ; RV32-NEXT:    addi a1, a1, 16
-; RV32-NEXT:    vl8r.v v0, (a1) # Unknown-size Folded Reload
-; RV32-NEXT:    vrgatherei16.vv v8, v0, v24
+; RV32-NEXT:    vl8r.v v24, (a1) # Unknown-size Folded Reload
 ; RV32-NEXT:    csrr a1, vlenb
-; RV32-NEXT:    li a3, 12
+; RV32-NEXT:    li a3, 13
 ; RV32-NEXT:    mul a1, a1, a3
 ; RV32-NEXT:    add a1, sp, a1
 ; RV32-NEXT:    addi a1, a1, 16
-; RV32-NEXT:    vl1r.v v0, (a1) # Unknown-size Folded Reload
+; RV32-NEXT:    vl4r.v v4, (a1) # Unknown-size Folded Reload
+; RV32-NEXT:    vrgatherei16.vv v8, v24, v4
+; RV32-NEXT:    vmv1r.v v0, v2
 ; RV32-NEXT:    csrr a1, vlenb
-; RV32-NEXT:    li a3, 48
+; RV32-NEXT:    li a3, 29
 ; RV32-NEXT:    mul a1, a1, a3
 ; RV32-NEXT:    add a1, sp, a1
 ; RV32-NEXT:    addi a1, a1, 16
 ; RV32-NEXT:    vl8r.v v24, (a1) # Unknown-size Folded Reload
 ; RV32-NEXT:    csrr a1, vlenb
-; RV32-NEXT:    slli a1, a1, 3
+; RV32-NEXT:    slli a3, a1, 2
+; RV32-NEXT:    add a1, a3, a1
 ; RV32-NEXT:    add a1, sp, a1
 ; RV32-NEXT:    addi a1, a1, 16
 ; RV32-NEXT:    vl4r.v v4, (a1) # Unknown-size Folded Reload
@@ -320,8 +340,8 @@ define {<8 x i64>, <8 x i64>, <8 x i64>, <8 x i64>, <8 x i64>, <8 x i64>} @load_
 ; RV32-NEXT:    vsetivli zero, 12, e32, m4, tu, ma
 ; RV32-NEXT:    vmv.v.v v20, v8
 ; RV32-NEXT:    csrr a1, vlenb
-; RV32-NEXT:    li a3, 12
-; RV32-NEXT:    mul a1, a1, a3
+; RV32-NEXT:    slli a3, a1, 2
+; RV32-NEXT:    add a1, a3, a1
 ; RV32-NEXT:    add a1, sp, a1
 ; RV32-NEXT:    addi a1, a1, 16
 ; RV32-NEXT:    vs4r.v v20, (a1) # Unknown-size Folded Spill
@@ -330,171 +350,178 @@ define {<8 x i64>, <8 x i64>, <8 x i64>, <8 x i64>, <8 x i64>, <8 x i64>} @load_
 ; RV32-NEXT:    vsetivli zero, 16, e32, m4, ta, mu
 ; RV32-NEXT:    vle16.v v8, (a1)
 ; RV32-NEXT:    csrr a1, vlenb
-; RV32-NEXT:    slli a1, a1, 5
+; RV32-NEXT:    li a3, 21
+; RV32-NEXT:    mul a1, a1, a3
 ; RV32-NEXT:    add a1, sp, a1
 ; RV32-NEXT:    addi a1, a1, 16
 ; RV32-NEXT:    vl8r.v v24, (a1) # Unknown-size Folded Reload
 ; RV32-NEXT:    vrgatherei16.vv v12, v24, v8
+; RV32-NEXT:    vmv1r.v v0, v3
+; RV32-NEXT:    vslideup.vi v12, v16, 6, v0.t
 ; RV32-NEXT:    csrr a1, vlenb
-; RV32-NEXT:    slli a1, a1, 2
+; RV32-NEXT:    li a3, 13
+; RV32-NEXT:    mul a1, a1, a3
 ; RV32-NEXT:    add a1, sp, a1
 ; RV32-NEXT:    addi a1, a1, 16
-; RV32-NEXT:    vl1r.v v3, (a1) # Unknown-size Folded Reload
-; RV32-NEXT:    vmv1r.v v0, v3
-; RV32-NEXT:    vslideup.vi v12, v16, 6, v0.t
-; RV32-NEXT:    vmv.v.v v4, v12
+; RV32-NEXT:    vs4r.v v12, (a1) # Unknown-size Folded Spill
 ; RV32-NEXT:    lui a1, %hi(.LCPI6_5)
 ; RV32-NEXT:    addi a1, a1, %lo(.LCPI6_5)
 ; RV32-NEXT:    vsetvli zero, a2, e32, m8, ta, mu
-; RV32-NEXT:    lui a3, %hi(.LCPI6_6)
-; RV32-NEXT:    addi a3, a3, %lo(.LCPI6_6)
 ; RV32-NEXT:    vle16.v v24, (a1)
-; RV32-NEXT:    vle16.v v8, (a3)
-; RV32-NEXT:    csrr a1, vlenb
-; RV32-NEXT:    slli a1, a1, 3
-; RV32-NEXT:    add a1, sp, a1
-; RV32-NEXT:    addi a1, a1, 16
-; RV32-NEXT:    vs4r.v v8, (a1) # Unknown-size Folded Spill
-; RV32-NEXT:    li a1, 960
-; RV32-NEXT:    vmv.s.x v2, a1
+; RV32-NEXT:    lui a1, %hi(.LCPI6_6)
+; RV32-NEXT:    addi a1, a1, %lo(.LCPI6_6)
+; RV32-NEXT:    li a3, 960
+; RV32-NEXT:    vle16.v v4, (a1)
+; RV32-NEXT:    vmv.s.x v0, a3
+; RV32-NEXT:    addi a1, sp, 16
+; RV32-NEXT:    vs1r.v v0, (a1) # Unknown-size Folded Spill
 ; RV32-NEXT:    csrr a1, vlenb
-; RV32-NEXT:    li a3, 40
+; RV32-NEXT:    li a3, 45
 ; RV32-NEXT:    mul a1, a1, a3
 ; RV32-NEXT:    add a1, sp, a1
 ; RV32-NEXT:    addi a1, a1, 16
 ; RV32-NEXT:    vl8r.v v16, (a1) # Unknown-size Folded Reload
 ; RV32-NEXT:    vrgatherei16.vv v8, v16, v24
-; RV32-NEXT:    vmv1r.v v0, v2
 ; RV32-NEXT:    csrr a1, vlenb
-; RV32-NEXT:    li a3, 48
+; RV32-NEXT:    li a3, 29
 ; RV32-NEXT:    mul a1, a1, a3
 ; RV32-NEXT:    add a1, sp, a1
 ; RV32-NEXT:    addi a1, a1, 16
 ; RV32-NEXT:    vl8r.v v24, (a1) # Unknown-size Folded Reload
+; RV32-NEXT:    vrgatherei16.vv v8, v24, v4, v0.t
+; RV32-NEXT:    vsetivli zero, 10, e32, m4, tu, ma
 ; RV32-NEXT:    csrr a1, vlenb
-; RV32-NEXT:    slli a1, a1, 3
+; RV32-NEXT:    li a3, 13
+; RV32-NEXT:    mul a1, a1, a3
 ; RV32-NEXT:    add a1, sp, a1
 ; RV32-NEXT:    addi a1, a1, 16
-; RV32-NEXT:    vl4r.v v16, (a1) # Unknown-size Folded Reload
-; RV32-NEXT:    vrgatherei16.vv v8, v24, v16, v0.t
-; RV32-NEXT:    vsetivli zero, 10, e32, m4, tu, ma
-; RV32-NEXT:    vmv.v.v v4, v8
+; RV32-NEXT:    vl4r.v v12, (a1) # Unknown-size Folded Reload
+; RV32-NEXT:    vmv.v.v v12, v8
 ; RV32-NEXT:    csrr a1, vlenb
-; RV32-NEXT:    slli a1, a1, 3
+; RV32-NEXT:    li a3, 13
+; RV32-NEXT:    mul a1, a1, a3
 ; RV32-NEXT:    add a1, sp, a1
 ; RV32-NEXT:    addi a1, a1, 16
-; RV32-NEXT:    vs4r.v v4, (a1) # Unknown-size Folded Spill
+; RV32-NEXT:    vs4r.v v12, (a1) # Unknown-size Folded Spill
 ; RV32-NEXT:    lui a1, %hi(.LCPI6_7)
 ; RV32-NEXT:    addi a1, a1, %lo(.LCPI6_7)
 ; RV32-NEXT:    vsetivli zero, 16, e32, m4, ta, mu
 ; RV32-NEXT:    vle16.v v8, (a1)
 ; RV32-NEXT:    csrr a1, vlenb
-; RV32-NEXT:    slli a1, a1, 5
+; RV32-NEXT:    li a3, 21
+; RV32-NEXT:    mul a1, a1, a3
 ; RV32-NEXT:    add a1, sp, a1
 ; RV32-NEXT:    addi a1, a1, 16
-; RV32-NEXT:    vl8r.v v24, (a1) # Unknown-size Folded Reload
-; RV32-NEXT:    vrgatherei16.vv v28, v24, v8
+; RV32-NEXT:    vl8r.v v16, (a1) # Unknown-size Folded Reload
+; RV32-NEXT:    vrgatherei16.vv v4, v16, v8
 ; RV32-NEXT:    vmv1r.v v0, v3
 ; RV32-NEXT:    csrr a1, vlenb
-; RV32-NEXT:    li a3, 24
+; RV32-NEXT:    li a3, 37
 ; RV32-NEXT:    mul a1, a1, a3
 ; RV32-NEXT:    add a1, sp, a1
 ; RV32-NEXT:    addi a1, a1, 16
 ; RV32-NEXT:    vl8r.v v8, (a1) # Unknown-size Folded Reload
-; RV32-NEXT:    vslideup.vi v28, v8, 4, v0.t
-; RV32-NEXT:    vmv.v.v v4, v28
+; RV32-NEXT:    vslideup.vi v4, v8, 4, v0.t
 ; RV32-NEXT:    lui a1, %hi(.LCPI6_8)
 ; RV32-NEXT:    addi a1, a1, %lo(.LCPI6_8)
 ; RV32-NEXT:    vsetvli zero, a2, e32, m8, ta, mu
-; RV32-NEXT:    lui a3, %hi(.LCPI6_9)
-; RV32-NEXT:    addi a3, a3, %lo(.LCPI6_9)
-; RV32-NEXT:    vle16.v v28, (a1)
-; RV32-NEXT:    vle16.v v24, (a3)
+; RV32-NEXT:    vle16.v v0, (a1)
+; RV32-NEXT:    lui a1, %hi(.LCPI6_9)
+; RV32-NEXT:    addi a1, a1, %lo(.LCPI6_9)
+; RV32-NEXT:    vle16.v v8, (a1)
 ; RV32-NEXT:    csrr a1, vlenb
-; RV32-NEXT:    li a3, 40
-; RV32-NEXT:    mul a1, a1, a3
 ; RV32-NEXT:    add a1, sp, a1
 ; RV32-NEXT:    addi a1, a1, 16
-; RV32-NEXT:    vl8r.v v16, (a1) # Unknown-size Folded Reload
-; RV32-NEXT:    vrgatherei16.vv v8, v16, v28
-; RV32-NEXT:    vmv1r.v v0, v2
+; RV32-NEXT:    vs4r.v v8, (a1) # Unknown-size Folded Spill
 ; RV32-NEXT:    csrr a1, vlenb
-; RV32-NEXT:    li a3, 48
+; RV32-NEXT:    li a3, 45
 ; RV32-NEXT:    mul a1, a1, a3
 ; RV32-NEXT:    add a1, sp, a1
 ; RV32-NEXT:    addi a1, a1, 16
 ; RV32-NEXT:    vl8r.v v16, (a1) # Unknown-size Folded Reload
-; RV32-NEXT:    vrgatherei16.vv v8, v16, v24, v0.t
+; RV32-NEXT:    vrgatherei16.vv v8, v16, v0
+; RV32-NEXT:    addi a1, sp, 16
+; RV32-NEXT:    vl1r.v v0, (a1) # Unknown-size Folded Reload
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    addi a1, a1, 16
+; RV32-NEXT:    vl4r.v v16, (a1) # Unknown-size Folded Reload
+; RV32-NEXT:    vrgatherei16.vv v8, v24, v16, v0.t
 ; RV32-NEXT:    vsetivli zero, 10, e32, m4, tu, ma
 ; RV32-NEXT:    vmv.v.v v4, v8
 ; RV32-NEXT:    csrr a1, vlenb
-; RV32-NEXT:    slli a1, a1, 2
 ; RV32-NEXT:    add a1, sp, a1
 ; RV32-NEXT:    addi a1, a1, 16
 ; RV32-NEXT:    vs4r.v v4, (a1) # Unknown-size Folded Spill
 ; RV32-NEXT:    vsetivli zero, 16, e32, m4, ta, mu
-; RV32-NEXT:    lui a1, %hi(.LCPI6_10)
-; RV32-NEXT:    addi a1, a1, %lo(.LCPI6_10)
-; RV32-NEXT:    vle16.v v4, (a1)
-; RV32-NEXT:    lui a1, 15
-; RV32-NEXT:    vmv.s.x v6, a1
 ; RV32-NEXT:    csrr a1, vlenb
-; RV32-NEXT:    slli a1, a1, 5
+; RV32-NEXT:    li a3, 21
+; RV32-NEXT:    mul a1, a1, a3
 ; RV32-NEXT:    add a1, sp, a1
 ; RV32-NEXT:    addi a1, a1, 16
-; RV32-NEXT:    vl8r.v v24, (a1) # Unknown-size Folded Reload
-; RV32-NEXT:    vslideup.vi v28, v24, 6
-; RV32-NEXT:    vmv1r.v v0, v6
+; RV32-NEXT:    vl8r.v v8, (a1) # Unknown-size Folded Reload
+; RV32-NEXT:    vslideup.vi v12, v8, 6
+; RV32-NEXT:    lui a1, %hi(.LCPI6_10)
+; RV32-NEXT:    addi a1, a1, %lo(.LCPI6_10)
+; RV32-NEXT:    vle16.v v8, (a1)
+; RV32-NEXT:    lui a1, 15
+; RV32-NEXT:    vmv.s.x v24, a1
+; RV32-NEXT:    vmv1r.v v0, v24
 ; RV32-NEXT:    csrr a1, vlenb
-; RV32-NEXT:    li a3, 24
+; RV32-NEXT:    li a3, 37
 ; RV32-NEXT:    mul a1, a1, a3
 ; RV32-NEXT:    add a1, sp, a1
 ; RV32-NEXT:    addi a1, a1, 16
-; RV32-NEXT:    vl8r.v v8, (a1) # Unknown-size Folded Reload
-; RV32-NEXT:    vrgatherei16.vv v28, v8, v4, v0.t
+; RV32-NEXT:    vl8r.v v16, (a1) # Unknown-size Folded Reload
+; RV32-NEXT:    vrgatherei16.vv v12, v16, v8, v0.t
+; RV32-NEXT:    vmv.v.v v28, v12
 ; RV32-NEXT:    lui a1, %hi(.LCPI6_11)
 ; RV32-NEXT:    addi a1, a1, %lo(.LCPI6_11)
 ; RV32-NEXT:    vsetvli zero, a2, e32, m8, ta, mu
-; RV32-NEXT:    lui a3, %hi(.LCPI6_12)
-; RV32-NEXT:    addi a3, a3, %lo(.LCPI6_12)
 ; RV32-NEXT:    vle16.v v0, (a1)
-; RV32-NEXT:    vle16.v v24, (a3)
-; RV32-NEXT:    li a1, 1008
-; RV32-NEXT:    vmv.s.x v7, a1
-; RV32-NEXT:    csrr a1, vlenb
-; RV32-NEXT:    slli a1, a1, 5
-; RV32-NEXT:    add a1, sp, a1
-; RV32-NEXT:    addi a1, a1, 16
-; RV32-NEXT:    vs1r.v v7, (a1) # Unknown-size Folded Spill
+; RV32-NEXT:    lui a1, %hi(.LCPI6_12)
+; RV32-NEXT:    addi a1, a1, %lo(.LCPI6_12)
+; RV32-NEXT:    li a3, 1008
+; RV32-NEXT:    vle16.v v4, (a1)
+; RV32-NEXT:    vmv.s.x v25, a3
+; RV32-NEXT:    addi a1, sp, 16
+; RV32-NEXT:    vs1r.v v25, (a1) # Unknown-size Folded Spill
 ; RV32-NEXT:    csrr a1, vlenb
-; RV32-NEXT:    li a3, 40
+; RV32-NEXT:    li a3, 45
 ; RV32-NEXT:    mul a1, a1, a3
 ; RV32-NEXT:    add a1, sp, a1
 ; RV32-NEXT:    addi a1, a1, 16
 ; RV32-NEXT:    vl8r.v v16, (a1) # Unknown-size Folded Reload
 ; RV32-NEXT:    vrgatherei16.vv v8, v16, v0
-; RV32-NEXT:    vmv1r.v v0, v7
+; RV32-NEXT:    vmv1r.v v0, v25
 ; RV32-NEXT:    csrr a1, vlenb
-; RV32-NEXT:    li a3, 48
+; RV32-NEXT:    li a3, 29
 ; RV32-NEXT:    mul a1, a1, a3
 ; RV32-NEXT:    add a1, sp, a1
 ; RV32-NEXT:    addi a1, a1, 16
 ; RV32-NEXT:    vl8r.v v16, (a1) # Unknown-size Folded Reload
-; RV32-NEXT:    vrgatherei16.vv v8, v16, v24, v0.t
+; RV32-NEXT:    vrgatherei16.vv v8, v16, v4, v0.t
 ; RV32-NEXT:    vsetivli zero, 10, e32, m4, tu, ma
 ; RV32-NEXT:    vmv.v.v v28, v8
+; RV32-NEXT:    csrr a1, vlenb
+; RV32-NEXT:    li a3, 21
+; RV32-NEXT:    mul a1, a1, a3
+; RV32-NEXT:    add a1, sp, a1
+; RV32-NEXT:    addi a1, a1, 16
+; RV32-NEXT:    vs4r.v v28, (a1) # Unknown-size Folded Spill
 ; RV32-NEXT:    lui a1, %hi(.LCPI6_13)
 ; RV32-NEXT:    addi a1, a1, %lo(.LCPI6_13)
 ; RV32-NEXT:    vsetivli zero, 16, e32, m4, ta, mu
 ; RV32-NEXT:    vle16.v v8, (a1)
-; RV32-NEXT:    vmv1r.v v0, v6
+; RV32-NEXT:    vmv1r.v v0, v24
 ; RV32-NEXT:    csrr a1, vlenb
-; RV32-NEXT:    slli a1, a1, 4
+; RV32-NEXT:    slli a3, a1, 3
+; RV32-NEXT:    add a1, a3, a1
 ; RV32-NEXT:    add a1, sp, a1
 ; RV32-NEXT:    addi a1, a1, 16
 ; RV32-NEXT:    vl4r.v v24, (a1) # Unknown-size Folded Reload
 ; RV32-NEXT:    csrr a1, vlenb
-; RV32-NEXT:    li a3, 24
+; RV32-NEXT:    li a3, 37
 ; RV32-NEXT:    mul a1, a1, a3
 ; RV32-NEXT:    add a1, sp, a1
 ; RV32-NEXT:    addi a1, a1, 16
@@ -503,79 +530,70 @@ define {<8 x i64>, <8 x i64>, <8 x i64>, <8 x i64>, <8 x i64>, <8 x i64>} @load_
 ; RV32-NEXT:    lui a1, %hi(.LCPI6_14)
 ; RV32-NEXT:    addi a1, a1, %lo(.LCPI6_14)
 ; RV32-NEXT:    vsetvli zero, a2, e32, m8, ta, mu
-; RV32-NEXT:    lui a2, %hi(.LCPI6_15)
-; RV32-NEXT:    addi a2, a2, %lo(.LCPI6_15)
 ; RV32-NEXT:    vle16.v v16, (a1)
-; RV32-NEXT:    vle16.v v8, (a2)
-; RV32-NEXT:    csrr a1, vlenb
-; RV32-NEXT:    li a2, 24
-; RV32-NEXT:    mul a1, a1, a2
-; RV32-NEXT:    add a1, sp, a1
-; RV32-NEXT:    addi a1, a1, 16
-; RV32-NEXT:    vs4r.v v8, (a1) # Unknown-size Folded Spill
+; RV32-NEXT:    lui a1, %hi(.LCPI6_15)
+; RV32-NEXT:    addi a1, a1, %lo(.LCPI6_15)
+; RV32-NEXT:    vle16.v v28, (a1)
 ; RV32-NEXT:    csrr a1, vlenb
-; RV32-NEXT:    li a2, 40
+; RV32-NEXT:    li a2, 45
 ; RV32-NEXT:    mul a1, a1, a2
 ; RV32-NEXT:    add a1, sp, a1
 ; RV32-NEXT:    addi a1, a1, 16
 ; RV32-NEXT:    vl8r.v v0, (a1) # Unknown-size Folded Reload
 ; RV32-NEXT:    vrgatherei16.vv v8, v0, v16
-; RV32-NEXT:    csrr a1, vlenb
-; RV32-NEXT:    slli a1, a1, 5
-; RV32-NEXT:    add a1, sp, a1
-; RV32-NEXT:    addi a1, a1, 16
+; RV32-NEXT:    addi a1, sp, 16
 ; RV32-NEXT:    vl1r.v v0, (a1) # Unknown-size Folded Reload
 ; RV32-NEXT:    csrr a1, vlenb
-; RV32-NEXT:    li a2, 48
+; RV32-NEXT:    li a2, 29
 ; RV32-NEXT:    mul a1, a1, a2
 ; RV32-NEXT:    add a1, sp, a1
 ; RV32-NEXT:    addi a1, a1, 16
 ; RV32-NEXT:    vl8r.v v16, (a1) # Unknown-size Folded Reload
-; RV32-NEXT:    csrr a1, vlenb
-; RV32-NEXT:    li a2, 24
-; RV32-NEXT:    mul a1, a1, a2
-; RV32-NEXT:    add a1, sp, a1
-; RV32-NEXT:    addi a1, a1, 16
-; RV32-NEXT:    vl4r.v v4, (a1) # Unknown-size Folded Reload
-; RV32-NEXT:    vrgatherei16.vv v8, v16, v4, v0.t
+; RV32-NEXT:    vrgatherei16.vv v8, v16, v28, v0.t
 ; RV32-NEXT:    vsetivli zero, 10, e32, m4, tu, ma
 ; RV32-NEXT:    vmv.v.v v24, v8
 ; RV32-NEXT:    addi a1, a0, 320
 ; RV32-NEXT:    vsetivli zero, 16, e32, m4, ta, ma
 ; RV32-NEXT:    vse32.v v24, (a1)
 ; RV32-NEXT:    addi a1, a0, 256
-; RV32-NEXT:    vse32.v v28, (a1)
+; RV32-NEXT:    csrr a2, vlenb
+; RV32-NEXT:    li a3, 21
+; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    add a2, sp, a2
+; RV32-NEXT:    addi a2, a2, 16
+; RV32-NEXT:    vl4r.v v8, (a2) # Unknown-size Folded Reload
+; RV32-NEXT:    vse32.v v8, (a1)
 ; RV32-NEXT:    addi a1, a0, 192
 ; RV32-NEXT:    csrr a2, vlenb
-; RV32-NEXT:    slli a2, a2, 2
 ; RV32-NEXT:    add a2, sp, a2
 ; RV32-NEXT:    addi a2, a2, 16
 ; RV32-NEXT:    vl4r.v v8, (a2) # Unknown-size Folded Reload
 ; RV32-NEXT:    vse32.v v8, (a1)
 ; RV32-NEXT:    addi a1, a0, 128
 ; RV32-NEXT:    csrr a2, vlenb
-; RV32-NEXT:    slli a2, a2, 3
+; RV32-NEXT:    li a3, 13
+; RV32-NEXT:    mul a2, a2, a3
 ; RV32-NEXT:    add a2, sp, a2
 ; RV32-NEXT:    addi a2, a2, 16
 ; RV32-NEXT:    vl4r.v v8, (a2) # Unknown-size Folded Reload
 ; RV32-NEXT:    vse32.v v8, (a1)
 ; RV32-NEXT:    addi a1, a0, 64
 ; RV32-NEXT:    csrr a2, vlenb
-; RV32-NEXT:    li a3, 12
-; RV32-NEXT:    mul a2, a2, a3
+; RV32-NEXT:    slli a3, a2, 2
+; RV32-NEXT:    add a2, a3, a2
 ; RV32-NEXT:    add a2, sp, a2
 ; RV32-NEXT:    addi a2, a2, 16
 ; RV32-NEXT:    vl4r.v v8, (a2) # Unknown-size Folded Reload
 ; RV32-NEXT:    vse32.v v8, (a1)
 ; RV32-NEXT:    csrr a1, vlenb
-; RV32-NEXT:    li a2, 20
-; RV32-NEXT:    mul a1, a1, a2
+; RV32-NEXT:    slli a2, a1, 4
+; RV32-NEXT:    add a1, a2, a1
 ; RV32-NEXT:    add a1, sp, a1
 ; RV32-NEXT:    addi a1, a1, 16
 ; RV32-NEXT:    vl4r.v v8, (a1) # Unknown-size Folded Reload
 ; RV32-NEXT:    vse32.v v8, (a0)
 ; RV32-NEXT:    csrr a0, vlenb
-; RV32-NEXT:    li a1, 56
+; RV32-NEXT:    li a1, 54
 ; RV32-NEXT:    mul a0, a0, a1
 ; RV32-NEXT:    add sp, sp, a0
 ; RV32-NEXT:    addi sp, sp, 16
@@ -586,320 +604,324 @@ define {<8 x i64>, <8 x i64>, <8 x i64>, <8 x i64>, <8 x i64>, <8 x i64>} @load_
 ; RV64-NEXT:    addi sp, sp, -16
 ; RV64-NEXT:    .cfi_def_cfa_offset 16
 ; RV64-NEXT:    csrr a2, vlenb
-; RV64-NEXT:    li a3, 52
+; RV64-NEXT:    li a3, 56
 ; RV64-NEXT:    mul a2, a2, a3
 ; RV64-NEXT:    sub sp, sp, a2
-; RV64-NEXT:    .cfi_escape 0x0f, 0x0d, 0x72, 0x00, 0x11, 0x10, 0x22, 0x11, 0x34, 0x92, 0xa2, 0x38, 0x00, 0x1e, 0x22 # sp + 16 + 52 * vlenb
+; RV64-NEXT:    .cfi_escape 0x0f, 0x0d, 0x72, 0x00, 0x11, 0x10, 0x22, 0x11, 0x38, 0x92, 0xa2, 0x38, 0x00, 0x1e, 0x22 # sp + 16 + 56 * vlenb
 ; RV64-NEXT:    vsetivli zero, 16, e64, m8, ta, ma
 ; RV64-NEXT:    addi a2, a1, 256
 ; RV64-NEXT:    vle64.v v16, (a2)
 ; RV64-NEXT:    csrr a2, vlenb
-; RV64-NEXT:    li a3, 27
-; RV64-NEXT:    mul a2, a2, a3
+; RV64-NEXT:    slli a2, a2, 5
 ; RV64-NEXT:    add a2, sp, a2
 ; RV64-NEXT:    addi a2, a2, 16
 ; RV64-NEXT:    vs8r.v v16, (a2) # Unknown-size Folded Spill
 ; RV64-NEXT:    addi a2, a1, 128
 ; RV64-NEXT:    vle64.v v8, (a2)
 ; RV64-NEXT:    csrr a2, vlenb
-; RV64-NEXT:    li a3, 35
+; RV64-NEXT:    li a3, 40
 ; RV64-NEXT:    mul a2, a2, a3
 ; RV64-NEXT:    add a2, sp, a2
 ; RV64-NEXT:    addi a2, a2, 16
 ; RV64-NEXT:    vs8r.v v8, (a2) # Unknown-size Folded Spill
-; RV64-NEXT:    vle64.v v8, (a1)
-; RV64-NEXT:    csrr a1, vlenb
-; RV64-NEXT:    li a2, 43
-; RV64-NEXT:    mul a1, a1, a2
-; RV64-NEXT:    add a1, sp, a1
-; RV64-NEXT:    addi a1, a1, 16
-; RV64-NEXT:    vs8r.v v8, (a1) # Unknown-size Folded Spill
+; RV64-NEXT:    vle64.v v24, (a1)
 ; RV64-NEXT:    vsetivli zero, 8, e64, m4, ta, ma
 ; RV64-NEXT:    vrgather.vi v8, v16, 4
 ; RV64-NEXT:    li a1, 128
-; RV64-NEXT:    vmv.s.x v0, a1
+; RV64-NEXT:    vmv.s.x v4, a1
 ; RV64-NEXT:    vsetivli zero, 8, e64, m8, ta, ma
-; RV64-NEXT:    vslidedown.vi v24, v16, 8
-; RV64-NEXT:    vsetivli zero, 8, e64, m4, ta, mu
-; RV64-NEXT:    vmv1r.v v28, v0
+; RV64-NEXT:    vslidedown.vi v16, v16, 8
 ; RV64-NEXT:    csrr a1, vlenb
-; RV64-NEXT:    slli a2, a1, 1
-; RV64-NEXT:    add a1, a2, a1
+; RV64-NEXT:    li a2, 24
+; RV64-NEXT:    mul a1, a1, a2
 ; RV64-NEXT:    add a1, sp, a1
 ; RV64-NEXT:    addi a1, a1, 16
-; RV64-NEXT:    vs1r.v v0, (a1) # Unknown-size Folded Spill
-; RV64-NEXT:    vrgather.vi v8, v24, 2, v0.t
+; RV64-NEXT:    vs8r.v v16, (a1) # Unknown-size Folded Spill
+; RV64-NEXT:    vsetivli zero, 8, e64, m4, ta, mu
+; RV64-NEXT:    vmv1r.v v0, v4
 ; RV64-NEXT:    csrr a1, vlenb
-; RV64-NEXT:    li a2, 19
+; RV64-NEXT:    li a2, 20
 ; RV64-NEXT:    mul a1, a1, a2
 ; RV64-NEXT:    add a1, sp, a1
 ; RV64-NEXT:    addi a1, a1, 16
-; RV64-NEXT:    vs8r.v v24, (a1) # Unknown-size Folded Spill
-; RV64-NEXT:    vmv.v.v v4, v8
+; RV64-NEXT:    vs1r.v v4, (a1) # Unknown-size Folded Spill
+; RV64-NEXT:    vrgather.vi v8, v16, 2, v0.t
+; RV64-NEXT:    vmv.v.v v20, v8
 ; RV64-NEXT:    vsetivli zero, 16, e16, m2, ta, ma
 ; RV64-NEXT:    li a1, 6
 ; RV64-NEXT:    vid.v v8
-; RV64-NEXT:    vmul.vx v2, v8, a1
+; RV64-NEXT:    vmul.vx v6, v8, a1
 ; RV64-NEXT:    vsetvli zero, zero, e64, m8, ta, ma
+; RV64-NEXT:    vrgatherei16.vv v8, v24, v6
 ; RV64-NEXT:    csrr a1, vlenb
-; RV64-NEXT:    li a2, 43
+; RV64-NEXT:    li a2, 48
 ; RV64-NEXT:    mul a1, a1, a2
 ; RV64-NEXT:    add a1, sp, a1
 ; RV64-NEXT:    addi a1, a1, 16
-; RV64-NEXT:    vl8r.v v16, (a1) # Unknown-size Folded Reload
-; RV64-NEXT:    vrgatherei16.vv v8, v16, v2
+; RV64-NEXT:    vs8r.v v24, (a1) # Unknown-size Folded Spill
 ; RV64-NEXT:    vsetvli zero, zero, e16, m2, ta, ma
 ; RV64-NEXT:    li a1, 56
-; RV64-NEXT:    vmv.s.x v1, a1
-; RV64-NEXT:    vadd.vi v30, v2, -16
+; RV64-NEXT:    vmv.s.x v5, a1
+; RV64-NEXT:    vadd.vi v16, v6, -16
 ; RV64-NEXT:    vsetvli zero, zero, e64, m8, ta, mu
-; RV64-NEXT:    vmv1r.v v0, v1
+; RV64-NEXT:    vmv1r.v v0, v5
 ; RV64-NEXT:    csrr a1, vlenb
-; RV64-NEXT:    li a2, 35
+; RV64-NEXT:    li a2, 40
 ; RV64-NEXT:    mul a1, a1, a2
 ; RV64-NEXT:    add a1, sp, a1
 ; RV64-NEXT:    addi a1, a1, 16
-; RV64-NEXT:    vl8r.v v16, (a1) # Unknown-size Folded Reload
-; RV64-NEXT:    vrgatherei16.vv v8, v16, v30, v0.t
+; RV64-NEXT:    vl8r.v v24, (a1) # Unknown-size Folded Reload
+; RV64-NEXT:    vrgatherei16.vv v8, v24, v16, v0.t
 ; RV64-NEXT:    vsetivli zero, 6, e64, m4, tu, ma
-; RV64-NEXT:    vmv.v.v v4, v8
+; RV64-NEXT:    vmv.v.v v20, v8
 ; RV64-NEXT:    csrr a1, vlenb
-; RV64-NEXT:    slli a2, a1, 4
-; RV64-NEXT:    sub a1, a2, a1
+; RV64-NEXT:    slli a1, a1, 4
 ; RV64-NEXT:    add a1, sp, a1
 ; RV64-NEXT:    addi a1, a1, 16
-; RV64-NEXT:    vs4r.v v4, (a1) # Unknown-size Folded Spill
+; RV64-NEXT:    vs4r.v v20, (a1) # Unknown-size Folded Spill
 ; RV64-NEXT:    vsetivli zero, 8, e64, m4, ta, mu
 ; RV64-NEXT:    csrr a1, vlenb
-; RV64-NEXT:    li a2, 27
-; RV64-NEXT:    mul a1, a1, a2
+; RV64-NEXT:    slli a1, a1, 5
 ; RV64-NEXT:    add a1, sp, a1
 ; RV64-NEXT:    addi a1, a1, 16
 ; RV64-NEXT:    vl8r.v v16, (a1) # Unknown-size Folded Reload
-; RV64-NEXT:    vrgather.vi v4, v16, 5
-; RV64-NEXT:    vmv1r.v v0, v28
-; RV64-NEXT:    vrgather.vi v4, v24, 3, v0.t
-; RV64-NEXT:    vsetivli zero, 16, e16, m2, ta, ma
-; RV64-NEXT:    addi a1, sp, 16
-; RV64-NEXT:    vs2r.v v2, (a1) # Unknown-size Folded Spill
-; RV64-NEXT:    vadd.vi v16, v2, 1
-; RV64-NEXT:    vsetvli zero, zero, e64, m8, ta, ma
+; RV64-NEXT:    vrgather.vi v24, v16, 5
+; RV64-NEXT:    vmv1r.v v0, v4
 ; RV64-NEXT:    csrr a1, vlenb
-; RV64-NEXT:    li a2, 43
+; RV64-NEXT:    li a2, 24
 ; RV64-NEXT:    mul a1, a1, a2
 ; RV64-NEXT:    add a1, sp, a1
 ; RV64-NEXT:    addi a1, a1, 16
-; RV64-NEXT:    vl8r.v v24, (a1) # Unknown-size Folded Reload
-; RV64-NEXT:    vrgatherei16.vv v8, v24, v16
-; RV64-NEXT:    vsetvli zero, zero, e16, m2, ta, ma
-; RV64-NEXT:    vadd.vi v16, v2, -15
+; RV64-NEXT:    vl8r.v v16, (a1) # Unknown-size Folded Reload
+; RV64-NEXT:    vrgather.vi v24, v16, 3, v0.t
+; RV64-NEXT:    vsetivli zero, 16, e16, m2, ta, ma
+; RV64-NEXT:    vadd.vi v28, v6, 1
+; RV64-NEXT:    vsetvli zero, zero, e64, m8, ta, ma
 ; RV64-NEXT:    csrr a1, vlenb
-; RV64-NEXT:    li a2, 11
+; RV64-NEXT:    li a2, 48
 ; RV64-NEXT:    mul a1, a1, a2
 ; RV64-NEXT:    add a1, sp, a1
 ; RV64-NEXT:    addi a1, a1, 16
-; RV64-NEXT:    vs2r.v v16, (a1) # Unknown-size Folded Spill
+; RV64-NEXT:    vl8r.v v16, (a1) # Unknown-size Folded Reload
+; RV64-NEXT:    vrgatherei16.vv v8, v16, v28
+; RV64-NEXT:    vsetvli zero, zero, e16, m2, ta, ma
+; RV64-NEXT:    vadd.vi v28, v6, -15
 ; RV64-NEXT:    vsetvli zero, zero, e64, m8, ta, mu
-; RV64-NEXT:    vmv1r.v v0, v1
+; RV64-NEXT:    vmv1r.v v0, v5
 ; RV64-NEXT:    csrr a1, vlenb
-; RV64-NEXT:    li a2, 35
+; RV64-NEXT:    li a2, 40
 ; RV64-NEXT:    mul a1, a1, a2
 ; RV64-NEXT:    add a1, sp, a1
 ; RV64-NEXT:    addi a1, a1, 16
 ; RV64-NEXT:    vl8r.v v16, (a1) # Unknown-size Folded Reload
+; RV64-NEXT:    vrgatherei16.vv v8, v16, v28, v0.t
+; RV64-NEXT:    vsetivli zero, 6, e64, m4, tu, ma
+; RV64-NEXT:    vmv.v.v v24, v8
 ; RV64-NEXT:    csrr a1, vlenb
-; RV64-NEXT:    li a2, 11
+; RV64-NEXT:    li a2, 12
 ; RV64-NEXT:    mul a1, a1, a2
 ; RV64-NEXT:    add a1, sp, a1
 ; RV64-NEXT:    addi a1, a1, 16
-; RV64-NEXT:    vl2r.v v2, (a1) # Unknown-size Folded Reload
-; RV64-NEXT:    vrgatherei16.vv v8, v16, v2, v0.t
-; RV64-NEXT:    vsetivli zero, 6, e64, m4, tu, ma
-; RV64-NEXT:    vmv.v.v v4, v8
+; RV64-NEXT:    vs4r.v v24, (a1) # Unknown-size Folded Spill
+; RV64-NEXT:    vsetivli zero, 16, e16, m2, ta, ma
+; RV64-NEXT:    vmv2r.v v26, v6
+; RV64-NEXT:    vadd.vi v24, v6, 2
+; RV64-NEXT:    vsetvli zero, zero, e64, m8, ta, ma
 ; RV64-NEXT:    csrr a1, vlenb
-; RV64-NEXT:    li a2, 11
+; RV64-NEXT:    li a2, 48
 ; RV64-NEXT:    mul a1, a1, a2
 ; RV64-NEXT:    add a1, sp, a1
 ; RV64-NEXT:    addi a1, a1, 16
-; RV64-NEXT:    vs4r.v v4, (a1) # Unknown-size Folded Spill
-; RV64-NEXT:    vsetivli zero, 16, e16, m2, ta, ma
-; RV64-NEXT:    addi a1, sp, 16
-; RV64-NEXT:    vl2r.v v2, (a1) # Unknown-size Folded Reload
-; RV64-NEXT:    vadd.vi v6, v2, 2
-; RV64-NEXT:    vsetvli zero, zero, e64, m8, ta, ma
-; RV64-NEXT:    vrgatherei16.vv v8, v24, v6
+; RV64-NEXT:    vl8r.v v0, (a1) # Unknown-size Folded Reload
+; RV64-NEXT:    vrgatherei16.vv v8, v0, v24
 ; RV64-NEXT:    vsetvli zero, zero, e16, m2, ta, ma
 ; RV64-NEXT:    li a1, 24
-; RV64-NEXT:    vmv.s.x v7, a1
-; RV64-NEXT:    vadd.vi v26, v2, -14
+; RV64-NEXT:    vmv.s.x v0, a1
+; RV64-NEXT:    csrr a1, vlenb
+; RV64-NEXT:    slli a1, a1, 1
+; RV64-NEXT:    add a1, sp, a1
+; RV64-NEXT:    addi a1, a1, 16
+; RV64-NEXT:    vs1r.v v0, (a1) # Unknown-size Folded Spill
+; RV64-NEXT:    vadd.vi v24, v26, -14
+; RV64-NEXT:    vmv2r.v v6, v26
 ; RV64-NEXT:    vsetvli zero, zero, e64, m8, ta, mu
-; RV64-NEXT:    vmv1r.v v0, v7
-; RV64-NEXT:    vrgatherei16.vv v8, v16, v26, v0.t
+; RV64-NEXT:    vrgatherei16.vv v8, v16, v24, v0.t
 ; RV64-NEXT:    vsetivli zero, 4, e32, m1, ta, ma
 ; RV64-NEXT:    vmv.v.i v12, 6
 ; RV64-NEXT:    vsetivli zero, 8, e64, m4, ta, mu
 ; RV64-NEXT:    csrr a1, vlenb
-; RV64-NEXT:    li a2, 27
-; RV64-NEXT:    mul a1, a1, a2
+; RV64-NEXT:    slli a1, a1, 5
 ; RV64-NEXT:    add a1, sp, a1
 ; RV64-NEXT:    addi a1, a1, 16
 ; RV64-NEXT:    vl8r.v v24, (a1) # Unknown-size Folded Reload
-; RV64-NEXT:    vrgatherei16.vv v16, v24, v12
+; RV64-NEXT:    vrgatherei16.vv v20, v24, v12
 ; RV64-NEXT:    csrr a1, vlenb
-; RV64-NEXT:    slli a2, a1, 1
-; RV64-NEXT:    add a1, a2, a1
+; RV64-NEXT:    li a2, 20
+; RV64-NEXT:    mul a1, a1, a2
 ; RV64-NEXT:    add a1, sp, a1
 ; RV64-NEXT:    addi a1, a1, 16
-; RV64-NEXT:    vl1r.v v6, (a1) # Unknown-size Folded Reload
-; RV64-NEXT:    vmv1r.v v0, v6
+; RV64-NEXT:    vl1r.v v0, (a1) # Unknown-size Folded Reload
 ; RV64-NEXT:    csrr a1, vlenb
-; RV64-NEXT:    li a2, 19
+; RV64-NEXT:    li a2, 24
 ; RV64-NEXT:    mul a1, a1, a2
 ; RV64-NEXT:    add a1, sp, a1
 ; RV64-NEXT:    addi a1, a1, 16
 ; RV64-NEXT:    vl8r.v v24, (a1) # Unknown-size Folded Reload
-; RV64-NEXT:    vrgather.vi v16, v24, 4, v0.t
+; RV64-NEXT:    vrgather.vi v20, v24, 4, v0.t
 ; RV64-NEXT:    vsetivli zero, 5, e64, m4, tu, ma
-; RV64-NEXT:    vmv.v.v v16, v8
+; RV64-NEXT:    vmv.v.v v20, v8
 ; RV64-NEXT:    csrr a1, vlenb
-; RV64-NEXT:    slli a2, a1, 3
-; RV64-NEXT:    sub a1, a2, a1
+; RV64-NEXT:    slli a1, a1, 3
 ; RV64-NEXT:    add a1, sp, a1
 ; RV64-NEXT:    addi a1, a1, 16
-; RV64-NEXT:    vs4r.v v16, (a1) # Unknown-size Folded Spill
+; RV64-NEXT:    vs4r.v v20, (a1) # Unknown-size Folded Spill
 ; RV64-NEXT:    vsetivli zero, 16, e16, m2, ta, ma
-; RV64-NEXT:    vadd.vi v28, v2, 3
+; RV64-NEXT:    vmv2r.v v10, v6
+; RV64-NEXT:    csrr a1, vlenb
+; RV64-NEXT:    li a2, 6
+; RV64-NEXT:    mul a1, a1, a2
+; RV64-NEXT:    add a1, sp, a1
+; RV64-NEXT:    addi a1, a1, 16
+; RV64-NEXT:    vs2r.v v6, (a1) # Unknown-size Folded Spill
+; RV64-NEXT:    vadd.vi v8, v6, 3
 ; RV64-NEXT:    vsetvli zero, zero, e64, m8, ta, ma
 ; RV64-NEXT:    csrr a1, vlenb
-; RV64-NEXT:    li a2, 43
+; RV64-NEXT:    li a2, 48
 ; RV64-NEXT:    mul a1, a1, a2
 ; RV64-NEXT:    add a1, sp, a1
 ; RV64-NEXT:    addi a1, a1, 16
-; RV64-NEXT:    vl8r.v v16, (a1) # Unknown-size Folded Reload
-; RV64-NEXT:    vrgatherei16.vv v8, v16, v28
+; RV64-NEXT:    vl8r.v v0, (a1) # Unknown-size Folded Reload
+; RV64-NEXT:    vrgatherei16.vv v16, v0, v8
 ; RV64-NEXT:    vsetvli zero, zero, e16, m2, ta, ma
-; RV64-NEXT:    vadd.vi v28, v2, -13
+; RV64-NEXT:    vadd.vi v28, v10, -13
 ; RV64-NEXT:    vsetvli zero, zero, e64, m8, ta, mu
-; RV64-NEXT:    vmv1r.v v0, v7
 ; RV64-NEXT:    csrr a1, vlenb
-; RV64-NEXT:    li a2, 35
+; RV64-NEXT:    slli a1, a1, 1
+; RV64-NEXT:    add a1, sp, a1
+; RV64-NEXT:    addi a1, a1, 16
+; RV64-NEXT:    vl1r.v v0, (a1) # Unknown-size Folded Reload
+; RV64-NEXT:    csrr a1, vlenb
+; RV64-NEXT:    li a2, 40
 ; RV64-NEXT:    mul a1, a1, a2
 ; RV64-NEXT:    add a1, sp, a1
 ; RV64-NEXT:    addi a1, a1, 16
-; RV64-NEXT:    vl8r.v v16, (a1) # Unknown-size Folded Reload
-; RV64-NEXT:    vrgatherei16.vv v8, v16, v28, v0.t
+; RV64-NEXT:    vl8r.v v8, (a1) # Unknown-size Folded Reload
+; RV64-NEXT:    vrgatherei16.vv v16, v8, v28, v0.t
 ; RV64-NEXT:    lui a1, 16
 ; RV64-NEXT:    addi a1, a1, 7
 ; RV64-NEXT:    vsetivli zero, 4, e32, m1, ta, ma
 ; RV64-NEXT:    vmv.v.x v12, a1
 ; RV64-NEXT:    vsetivli zero, 8, e64, m4, ta, mu
 ; RV64-NEXT:    csrr a1, vlenb
-; RV64-NEXT:    li a2, 27
-; RV64-NEXT:    mul a1, a1, a2
+; RV64-NEXT:    slli a1, a1, 5
 ; RV64-NEXT:    add a1, sp, a1
 ; RV64-NEXT:    addi a1, a1, 16
-; RV64-NEXT:    vl8r.v v24, (a1) # Unknown-size Folded Reload
-; RV64-NEXT:    vrgatherei16.vv v16, v24, v12
-; RV64-NEXT:    vmv1r.v v0, v6
+; RV64-NEXT:    vl8r.v v0, (a1) # Unknown-size Folded Reload
+; RV64-NEXT:    vmv4r.v v8, v0
+; RV64-NEXT:    vrgatherei16.vv v20, v0, v12
 ; RV64-NEXT:    csrr a1, vlenb
-; RV64-NEXT:    li a2, 19
+; RV64-NEXT:    li a2, 20
 ; RV64-NEXT:    mul a1, a1, a2
 ; RV64-NEXT:    add a1, sp, a1
 ; RV64-NEXT:    addi a1, a1, 16
-; RV64-NEXT:    vl8r.v v24, (a1) # Unknown-size Folded Reload
-; RV64-NEXT:    vrgather.vi v16, v24, 5, v0.t
+; RV64-NEXT:    vl1r.v v0, (a1) # Unknown-size Folded Reload
+; RV64-NEXT:    vrgather.vi v20, v24, 5, v0.t
 ; RV64-NEXT:    vsetivli zero, 5, e64, m4, tu, ma
-; RV64-NEXT:    vmv.v.v v16, v8
+; RV64-NEXT:    vmv.v.v v20, v16
 ; RV64-NEXT:    csrr a1, vlenb
-; RV64-NEXT:    slli a2, a1, 1
-; RV64-NEXT:    add a1, a2, a1
+; RV64-NEXT:    li a2, 20
+; RV64-NEXT:    mul a1, a1, a2
 ; RV64-NEXT:    add a1, sp, a1
 ; RV64-NEXT:    addi a1, a1, 16
-; RV64-NEXT:    vs4r.v v16, (a1) # Unknown-size Folded Spill
+; RV64-NEXT:    vs4r.v v20, (a1) # Unknown-size Folded Spill
 ; RV64-NEXT:    lui a1, 96
 ; RV64-NEXT:    vsetivli zero, 4, e32, m1, ta, ma
-; RV64-NEXT:    vmv.v.x v8, a1
+; RV64-NEXT:    vmv.v.x v12, a1
 ; RV64-NEXT:    vsetivli zero, 8, e64, m4, ta, mu
 ; RV64-NEXT:    li a1, 192
 ; RV64-NEXT:    vmv.s.x v0, a1
 ; RV64-NEXT:    csrr a1, vlenb
-; RV64-NEXT:    slli a1, a1, 1
 ; RV64-NEXT:    add a1, sp, a1
 ; RV64-NEXT:    addi a1, a1, 16
 ; RV64-NEXT:    vs1r.v v0, (a1) # Unknown-size Folded Spill
+; RV64-NEXT:    vrgather.vi v28, v8, 2
+; RV64-NEXT:    vrgatherei16.vv v28, v24, v12, v0.t
+; RV64-NEXT:    vsetivli zero, 16, e16, m2, ta, ma
 ; RV64-NEXT:    csrr a1, vlenb
-; RV64-NEXT:    li a2, 27
+; RV64-NEXT:    li a2, 6
 ; RV64-NEXT:    mul a1, a1, a2
 ; RV64-NEXT:    add a1, sp, a1
 ; RV64-NEXT:    addi a1, a1, 16
-; RV64-NEXT:    vl8r.v v16, (a1) # Unknown-size Folded Reload
-; RV64-NEXT:    vrgather.vi v4, v16, 2
-; RV64-NEXT:    vrgatherei16.vv v4, v24, v8, v0.t
-; RV64-NEXT:    vsetivli zero, 16, e16, m2, ta, ma
-; RV64-NEXT:    vadd.vi v16, v2, 4
+; RV64-NEXT:    vl2r.v v24, (a1) # Unknown-size Folded Reload
+; RV64-NEXT:    vadd.vi v16, v24, 4
 ; RV64-NEXT:    vsetvli zero, zero, e64, m8, ta, ma
 ; RV64-NEXT:    csrr a1, vlenb
-; RV64-NEXT:    li a2, 43
+; RV64-NEXT:    li a2, 48
 ; RV64-NEXT:    mul a1, a1, a2
 ; RV64-NEXT:    add a1, sp, a1
 ; RV64-NEXT:    addi a1, a1, 16
-; RV64-NEXT:    vl8r.v v24, (a1) # Unknown-size Folded Reload
-; RV64-NEXT:    vrgatherei16.vv v8, v24, v16
+; RV64-NEXT:    vl8r.v v0, (a1) # Unknown-size Folded Reload
+; RV64-NEXT:    vrgatherei16.vv v8, v0, v16
 ; RV64-NEXT:    vsetvli zero, zero, e16, m2, ta, ma
 ; RV64-NEXT:    li a1, 28
-; RV64-NEXT:    vmv.s.x v1, a1
-; RV64-NEXT:    vadd.vi v16, v2, -12
+; RV64-NEXT:    vmv.s.x v0, a1
+; RV64-NEXT:    addi a1, sp, 16
+; RV64-NEXT:    vs1r.v v0, (a1) # Unknown-size Folded Spill
+; RV64-NEXT:    vadd.vi v26, v24, -12
 ; RV64-NEXT:    vsetvli zero, zero, e64, m8, ta, mu
-; RV64-NEXT:    vmv1r.v v0, v1
 ; RV64-NEXT:    csrr a1, vlenb
-; RV64-NEXT:    li a2, 35
+; RV64-NEXT:    li a2, 40
 ; RV64-NEXT:    mul a1, a1, a2
 ; RV64-NEXT:    add a1, sp, a1
 ; RV64-NEXT:    addi a1, a1, 16
-; RV64-NEXT:    vl8r.v v24, (a1) # Unknown-size Folded Reload
-; RV64-NEXT:    vrgatherei16.vv v8, v24, v16, v0.t
+; RV64-NEXT:    vl8r.v v16, (a1) # Unknown-size Folded Reload
+; RV64-NEXT:    vrgatherei16.vv v8, v16, v26, v0.t
 ; RV64-NEXT:    vsetivli zero, 5, e64, m4, tu, ma
-; RV64-NEXT:    vmv.v.v v4, v8
+; RV64-NEXT:    vmv.v.v v28, v8
+; RV64-NEXT:    csrr a1, vlenb
+; RV64-NEXT:    slli a1, a1, 1
+; RV64-NEXT:    add a1, sp, a1
+; RV64-NEXT:    addi a1, a1, 16
+; RV64-NEXT:    vs4r.v v28, (a1) # Unknown-size Folded Spill
 ; RV64-NEXT:    lui a1, 112
 ; RV64-NEXT:    addi a1, a1, 1
 ; RV64-NEXT:    vsetivli zero, 4, e32, m1, ta, ma
 ; RV64-NEXT:    vmv.v.x v12, a1
 ; RV64-NEXT:    vsetivli zero, 8, e64, m4, ta, mu
 ; RV64-NEXT:    csrr a1, vlenb
-; RV64-NEXT:    li a2, 27
-; RV64-NEXT:    mul a1, a1, a2
+; RV64-NEXT:    slli a1, a1, 5
 ; RV64-NEXT:    add a1, sp, a1
 ; RV64-NEXT:    addi a1, a1, 16
 ; RV64-NEXT:    vl8r.v v16, (a1) # Unknown-size Folded Reload
 ; RV64-NEXT:    vrgather.vi v8, v16, 3
 ; RV64-NEXT:    csrr a1, vlenb
-; RV64-NEXT:    slli a1, a1, 1
 ; RV64-NEXT:    add a1, sp, a1
 ; RV64-NEXT:    addi a1, a1, 16
 ; RV64-NEXT:    vl1r.v v0, (a1) # Unknown-size Folded Reload
 ; RV64-NEXT:    csrr a1, vlenb
-; RV64-NEXT:    li a2, 19
+; RV64-NEXT:    li a2, 24
 ; RV64-NEXT:    mul a1, a1, a2
 ; RV64-NEXT:    add a1, sp, a1
 ; RV64-NEXT:    addi a1, a1, 16
 ; RV64-NEXT:    vl8r.v v16, (a1) # Unknown-size Folded Reload
 ; RV64-NEXT:    vrgatherei16.vv v8, v16, v12, v0.t
 ; RV64-NEXT:    vsetivli zero, 16, e16, m2, ta, ma
-; RV64-NEXT:    vadd.vi v12, v2, 5
+; RV64-NEXT:    vadd.vi v12, v24, 5
 ; RV64-NEXT:    vsetvli zero, zero, e64, m8, ta, ma
 ; RV64-NEXT:    csrr a1, vlenb
-; RV64-NEXT:    li a2, 43
+; RV64-NEXT:    li a2, 48
 ; RV64-NEXT:    mul a1, a1, a2
 ; RV64-NEXT:    add a1, sp, a1
 ; RV64-NEXT:    addi a1, a1, 16
-; RV64-NEXT:    vl8r.v v24, (a1) # Unknown-size Folded Reload
-; RV64-NEXT:    vrgatherei16.vv v16, v24, v12
+; RV64-NEXT:    vl8r.v v0, (a1) # Unknown-size Folded Reload
+; RV64-NEXT:    vrgatherei16.vv v16, v0, v12
 ; RV64-NEXT:    vsetvli zero, zero, e16, m2, ta, ma
-; RV64-NEXT:    vadd.vi v12, v2, -11
+; RV64-NEXT:    vadd.vi v12, v24, -11
 ; RV64-NEXT:    vsetvli zero, zero, e64, m8, ta, mu
-; RV64-NEXT:    vmv1r.v v0, v1
+; RV64-NEXT:    addi a1, sp, 16
+; RV64-NEXT:    vl1r.v v0, (a1) # Unknown-size Folded Reload
 ; RV64-NEXT:    csrr a1, vlenb
-; RV64-NEXT:    li a2, 35
+; RV64-NEXT:    li a2, 40
 ; RV64-NEXT:    mul a1, a1, a2
 ; RV64-NEXT:    add a1, sp, a1
 ; RV64-NEXT:    addi a1, a1, 16
@@ -911,40 +933,43 @@ define {<8 x i64>, <8 x i64>, <8 x i64>, <8 x i64>, <8 x i64>, <8 x i64>} @load_
 ; RV64-NEXT:    vsetivli zero, 8, e64, m4, ta, ma
 ; RV64-NEXT:    vse64.v v8, (a1)
 ; RV64-NEXT:    addi a1, a0, 256
-; RV64-NEXT:    vse64.v v4, (a1)
+; RV64-NEXT:    csrr a2, vlenb
+; RV64-NEXT:    slli a2, a2, 1
+; RV64-NEXT:    add a2, sp, a2
+; RV64-NEXT:    addi a2, a2, 16
+; RV64-NEXT:    vl4r.v v8, (a2) # Unknown-size Folded Reload
+; RV64-NEXT:    vse64.v v8, (a1)
 ; RV64-NEXT:    addi a1, a0, 192
 ; RV64-NEXT:    csrr a2, vlenb
-; RV64-NEXT:    slli a3, a2, 1
-; RV64-NEXT:    add a2, a3, a2
+; RV64-NEXT:    li a3, 20
+; RV64-NEXT:    mul a2, a2, a3
 ; RV64-NEXT:    add a2, sp, a2
 ; RV64-NEXT:    addi a2, a2, 16
 ; RV64-NEXT:    vl4r.v v8, (a2) # Unknown-size Folded Reload
 ; RV64-NEXT:    vse64.v v8, (a1)
 ; RV64-NEXT:    addi a1, a0, 128
 ; RV64-NEXT:    csrr a2, vlenb
-; RV64-NEXT:    slli a3, a2, 3
-; RV64-NEXT:    sub a2, a3, a2
+; RV64-NEXT:    slli a2, a2, 3
 ; RV64-NEXT:    add a2, sp, a2
 ; RV64-NEXT:    addi a2, a2, 16
 ; RV64-NEXT:    vl4r.v v8, (a2) # Unknown-size Folded Reload
 ; RV64-NEXT:    vse64.v v8, (a1)
 ; RV64-NEXT:    addi a1, a0, 64
 ; RV64-NEXT:    csrr a2, vlenb
-; RV64-NEXT:    li a3, 11
+; RV64-NEXT:    li a3, 12
 ; RV64-NEXT:    mul a2, a2, a3
 ; RV64-NEXT:    add a2, sp, a2
 ; RV64-NEXT:    addi a2, a2, 16
 ; RV64-NEXT:    vl4r.v v8, (a2) # Unknown-size Folded Reload
 ; RV64-NEXT:    vse64.v v8, (a1)
 ; RV64-NEXT:    csrr a1, vlenb
-; RV64-NEXT:    slli a2, a1, 4
-; RV64-NEXT:    sub a1, a2, a1
+; RV64-NEXT:    slli a1, a1, 4
 ; RV64-NEXT:    add a1, sp, a1
 ; RV64-NEXT:    addi a1, a1, 16
 ; RV64-NEXT:    vl4r.v v8, (a1) # Unknown-size Folded Reload
 ; RV64-NEXT:    vse64.v v8, (a0)
 ; RV64-NEXT:    csrr a0, vlenb
-; RV64-NEXT:    li a1, 52
+; RV64-NEXT:    li a1, 56
 ; RV64-NEXT:    mul a0, a0, a1
 ; RV64-NEXT:    add sp, sp, a0
 ; RV64-NEXT:    addi sp, sp, 16
diff --git a/llvm/test/CodeGen/RISCV/rvv/fixed-vectors-mask-buildvec.ll b/llvm/test/CodeGen/RISCV/rvv/fixed-vectors-mask-buildvec.ll
index c295fed2c28c..023d707f07bf 100644
--- a/llvm/test/CodeGen/RISCV/rvv/fixed-vectors-mask-buildvec.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/fixed-vectors-mask-buildvec.ll
@@ -286,8 +286,8 @@ define <8 x i1> @buildvec_mask_nonconst_v8i1_2(i1 %x, i1 %y, i1 %z, i1 %w) {
 ; CHECK-NEXT:    vslide1down.vx v9, v9, a1
 ; CHECK-NEXT:    vslide1down.vx v8, v8, a3
 ; CHECK-NEXT:    vslide1down.vx v8, v8, zero
-; CHECK-NEXT:    vmv.v.i v0, 15
 ; CHECK-NEXT:    vslide1down.vx v8, v8, a2
+; CHECK-NEXT:    vmv.v.i v0, 15
 ; CHECK-NEXT:    vslidedown.vi v8, v9, 4, v0.t
 ; CHECK-NEXT:    vand.vi v8, v8, 1
 ; CHECK-NEXT:    vmsne.vi v0, v8, 0
@@ -303,8 +303,8 @@ define <8 x i1> @buildvec_mask_nonconst_v8i1_2(i1 %x, i1 %y, i1 %z, i1 %w) {
 ; ZVE32F-NEXT:    vslide1down.vx v9, v9, a1
 ; ZVE32F-NEXT:    vslide1down.vx v8, v8, a3
 ; ZVE32F-NEXT:    vslide1down.vx v8, v8, zero
-; ZVE32F-NEXT:    vmv.v.i v0, 15
 ; ZVE32F-NEXT:    vslide1down.vx v8, v8, a2
+; ZVE32F-NEXT:    vmv.v.i v0, 15
 ; ZVE32F-NEXT:    vslidedown.vi v8, v9, 4, v0.t
 ; ZVE32F-NEXT:    vand.vi v8, v8, 1
 ; ZVE32F-NEXT:    vmsne.vi v0, v8, 0
@@ -331,8 +331,8 @@ define <8 x i1> @buildvec_mask_optsize_nonconst_v8i1_2(i1 %x, i1 %y, i1 %z, i1 %
 ; CHECK-NEXT:    vslide1down.vx v9, v9, a1
 ; CHECK-NEXT:    vslide1down.vx v8, v8, a3
 ; CHECK-NEXT:    vslide1down.vx v8, v8, zero
-; CHECK-NEXT:    vmv.v.i v0, 15
 ; CHECK-NEXT:    vslide1down.vx v8, v8, a2
+; CHECK-NEXT:    vmv.v.i v0, 15
 ; CHECK-NEXT:    vslidedown.vi v8, v9, 4, v0.t
 ; CHECK-NEXT:    vand.vi v8, v8, 1
 ; CHECK-NEXT:    vmsne.vi v0, v8, 0
@@ -348,8 +348,8 @@ define <8 x i1> @buildvec_mask_optsize_nonconst_v8i1_2(i1 %x, i1 %y, i1 %z, i1 %
 ; ZVE32F-NEXT:    vslide1down.vx v9, v9, a1
 ; ZVE32F-NEXT:    vslide1down.vx v8, v8, a3
 ; ZVE32F-NEXT:    vslide1down.vx v8, v8, zero
-; ZVE32F-NEXT:    vmv.v.i v0, 15
 ; ZVE32F-NEXT:    vslide1down.vx v8, v8, a2
+; ZVE32F-NEXT:    vmv.v.i v0, 15
 ; ZVE32F-NEXT:    vslidedown.vi v8, v9, 4, v0.t
 ; ZVE32F-NEXT:    vand.vi v8, v8, 1
 ; ZVE32F-NEXT:    vmsne.vi v0, v8, 0
@@ -375,8 +375,8 @@ define <8 x i1> @buildvec_mask_optsize_nonconst_v8i1(i1 %x, i1 %y) optsize {
 ; CHECK-NEXT:    vslide1down.vx v9, v9, a1
 ; CHECK-NEXT:    vslide1down.vx v8, v8, a1
 ; CHECK-NEXT:    vslide1down.vx v8, v8, a1
-; CHECK-NEXT:    vmv.v.i v0, 15
 ; CHECK-NEXT:    vslide1down.vx v8, v8, a1
+; CHECK-NEXT:    vmv.v.i v0, 15
 ; CHECK-NEXT:    vslidedown.vi v8, v9, 4, v0.t
 ; CHECK-NEXT:    vand.vi v8, v8, 1
 ; CHECK-NEXT:    vmsne.vi v0, v8, 0
@@ -391,8 +391,8 @@ define <8 x i1> @buildvec_mask_optsize_nonconst_v8i1(i1 %x, i1 %y) optsize {
 ; ZVE32F-NEXT:    vslide1down.vx v9, v9, a1
 ; ZVE32F-NEXT:    vslide1down.vx v8, v8, a1
 ; ZVE32F-NEXT:    vslide1down.vx v8, v8, a1
-; ZVE32F-NEXT:    vmv.v.i v0, 15
 ; ZVE32F-NEXT:    vslide1down.vx v8, v8, a1
+; ZVE32F-NEXT:    vmv.v.i v0, 15
 ; ZVE32F-NEXT:    vslidedown.vi v8, v9, 4, v0.t
 ; ZVE32F-NEXT:    vand.vi v8, v8, 1
 ; ZVE32F-NEXT:    vmsne.vi v0, v8, 0
diff --git a/llvm/test/CodeGen/RISCV/rvv/fixed-vectors-mask-splat.ll b/llvm/test/CodeGen/RISCV/rvv/fixed-vectors-mask-splat.ll
index 4f7b885d998e..7fc442c88d10 100644
--- a/llvm/test/CodeGen/RISCV/rvv/fixed-vectors-mask-splat.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/fixed-vectors-mask-splat.ll
@@ -25,10 +25,10 @@ define void @splat_v1i1(ptr %x, i1 %y) {
 ; CHECK-LABEL: splat_v1i1:
 ; CHECK:       # %bb.0:
 ; CHECK-NEXT:    vsetivli zero, 1, e8, mf8, ta, ma
-; CHECK-NEXT:    andi a1, a1, 1
-; CHECK-NEXT:    vmv.s.x v8, a1
-; CHECK-NEXT:    vmsne.vi v0, v8, 0
 ; CHECK-NEXT:    vmv.s.x v8, zero
+; CHECK-NEXT:    andi a1, a1, 1
+; CHECK-NEXT:    vmv.s.x v9, a1
+; CHECK-NEXT:    vmsne.vi v0, v9, 0
 ; CHECK-NEXT:    vmerge.vim v8, v8, 1, v0
 ; CHECK-NEXT:    vsetivli zero, 8, e8, mf2, ta, ma
 ; CHECK-NEXT:    vmv.v.i v9, 0
diff --git a/llvm/test/CodeGen/RISCV/rvv/fixed-vectors-masked-gather.ll b/llvm/test/CodeGen/RISCV/rvv/fixed-vectors-masked-gather.ll
index 9fbc22221f99..539a8403c935 100644
--- a/llvm/test/CodeGen/RISCV/rvv/fixed-vectors-masked-gather.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/fixed-vectors-masked-gather.ll
@@ -12728,8 +12728,8 @@ define <8 x i16> @mgather_strided_unaligned(ptr %base) {
 ; RV32-NEXT:    vmv.v.x v8, a3
 ; RV32-NEXT:    vslide1down.vx v8, v8, a5
 ; RV32-NEXT:    vslide1down.vx v8, v8, a6
-; RV32-NEXT:    vmv.v.i v0, 15
 ; RV32-NEXT:    vslide1down.vx v8, v8, a7
+; RV32-NEXT:    vmv.v.i v0, 15
 ; RV32-NEXT:    vslidedown.vi v8, v9, 4, v0.t
 ; RV32-NEXT:    ret
 ;
@@ -12803,8 +12803,8 @@ define <8 x i16> @mgather_strided_unaligned(ptr %base) {
 ; RV64V-NEXT:    vmv.v.x v8, a3
 ; RV64V-NEXT:    vslide1down.vx v8, v8, a5
 ; RV64V-NEXT:    vslide1down.vx v8, v8, a6
-; RV64V-NEXT:    vmv.v.i v0, 15
 ; RV64V-NEXT:    vslide1down.vx v8, v8, a7
+; RV64V-NEXT:    vmv.v.i v0, 15
 ; RV64V-NEXT:    vslidedown.vi v8, v9, 4, v0.t
 ; RV64V-NEXT:    addi sp, s0, -128
 ; RV64V-NEXT:    ld ra, 120(sp) # 8-byte Folded Reload
@@ -12854,8 +12854,8 @@ define <8 x i16> @mgather_strided_unaligned(ptr %base) {
 ; RV64ZVE32F-NEXT:    vmv.v.x v8, a5
 ; RV64ZVE32F-NEXT:    vslide1down.vx v8, v8, a6
 ; RV64ZVE32F-NEXT:    vslide1down.vx v8, v8, a7
-; RV64ZVE32F-NEXT:    vmv.v.i v0, 15
 ; RV64ZVE32F-NEXT:    vslide1down.vx v8, v8, a0
+; RV64ZVE32F-NEXT:    vmv.v.i v0, 15
 ; RV64ZVE32F-NEXT:    vslidedown.vi v8, v9, 4, v0.t
 ; RV64ZVE32F-NEXT:    ret
   %ptrs = getelementptr inbounds i16, ptr %base, <8 x i32>  <i32 0, i32 2, i32 4, i32 6, i32 8, i32 10, i32 12, i32 14>
@@ -12896,8 +12896,8 @@ define <8 x i16> @mgather_strided_2xSEW(ptr %base) {
 ; RV64ZVE32F-NEXT:    vslide1down.vx v10, v8, a3
 ; RV64ZVE32F-NEXT:    vslide1down.vx v8, v9, a4
 ; RV64ZVE32F-NEXT:    vslide1down.vx v8, v8, a5
-; RV64ZVE32F-NEXT:    vmv.v.i v0, 15
 ; RV64ZVE32F-NEXT:    vslide1down.vx v8, v8, a6
+; RV64ZVE32F-NEXT:    vmv.v.i v0, 15
 ; RV64ZVE32F-NEXT:    vslidedown.vi v8, v10, 4, v0.t
 ; RV64ZVE32F-NEXT:    ret
   %ptrs = getelementptr inbounds i16, ptr %base, <8 x i32>  <i32 0, i32 1, i32 4, i32 5, i32 8, i32 9, i32 12, i32 13>
@@ -12941,8 +12941,8 @@ define <8 x i16> @mgather_strided_2xSEW_with_offset(ptr %base) {
 ; RV64ZVE32F-NEXT:    vslide1down.vx v10, v8, a4
 ; RV64ZVE32F-NEXT:    vslide1down.vx v8, v9, a5
 ; RV64ZVE32F-NEXT:    vslide1down.vx v8, v8, a6
-; RV64ZVE32F-NEXT:    vmv.v.i v0, 15
 ; RV64ZVE32F-NEXT:    vslide1down.vx v8, v8, a7
+; RV64ZVE32F-NEXT:    vmv.v.i v0, 15
 ; RV64ZVE32F-NEXT:    vslidedown.vi v8, v10, 4, v0.t
 ; RV64ZVE32F-NEXT:    ret
   %ptrs = getelementptr inbounds i16, ptr %base, <8 x i64>  <i64 2, i64 3, i64 6, i64 7, i64 10, i64 11, i64 14, i64 15>
@@ -12986,8 +12986,8 @@ define <8 x i16> @mgather_reverse_unit_strided_2xSEW(ptr %base) {
 ; RV64ZVE32F-NEXT:    vslide1down.vx v10, v8, a4
 ; RV64ZVE32F-NEXT:    vslide1down.vx v8, v9, a5
 ; RV64ZVE32F-NEXT:    vslide1down.vx v8, v8, a6
-; RV64ZVE32F-NEXT:    vmv.v.i v0, 15
 ; RV64ZVE32F-NEXT:    vslide1down.vx v8, v8, a7
+; RV64ZVE32F-NEXT:    vmv.v.i v0, 15
 ; RV64ZVE32F-NEXT:    vslidedown.vi v8, v10, 4, v0.t
 ; RV64ZVE32F-NEXT:    ret
   %ptrs = getelementptr inbounds i16, ptr %base, <8 x i64>  <i64 14, i64 15, i64 12, i64 13, i64 10, i64 11, i64 8, i64 9>
@@ -13031,8 +13031,8 @@ define <8 x i16> @mgather_reverse_strided_2xSEW(ptr %base) {
 ; RV64ZVE32F-NEXT:    vslide1down.vx v10, v8, a4
 ; RV64ZVE32F-NEXT:    vslide1down.vx v8, v9, a5
 ; RV64ZVE32F-NEXT:    vslide1down.vx v8, v8, a6
-; RV64ZVE32F-NEXT:    vmv.v.i v0, 15
 ; RV64ZVE32F-NEXT:    vslide1down.vx v8, v8, a7
+; RV64ZVE32F-NEXT:    vmv.v.i v0, 15
 ; RV64ZVE32F-NEXT:    vslidedown.vi v8, v10, 4, v0.t
 ; RV64ZVE32F-NEXT:    ret
   %ptrs = getelementptr inbounds i16, ptr %base, <8 x i64>  <i64 14, i64 15, i64 10, i64 11, i64 6, i64 7, i64 2, i64 3>
@@ -13074,8 +13074,8 @@ define <8 x i16> @mgather_gather_2xSEW(ptr %base) {
 ; RV64ZVE32F-NEXT:    vslide1down.vx v10, v8, a3
 ; RV64ZVE32F-NEXT:    vslide1down.vx v8, v9, a4
 ; RV64ZVE32F-NEXT:    vslide1down.vx v8, v8, a5
-; RV64ZVE32F-NEXT:    vmv.v.i v0, 15
 ; RV64ZVE32F-NEXT:    vslide1down.vx v8, v8, a6
+; RV64ZVE32F-NEXT:    vmv.v.i v0, 15
 ; RV64ZVE32F-NEXT:    vslidedown.vi v8, v10, 4, v0.t
 ; RV64ZVE32F-NEXT:    ret
   %ptrs = getelementptr inbounds i16, ptr %base, <8 x i32>  <i32 0, i32 1, i32 8, i32 9, i32 4, i32 5, i32 2, i32 3>
@@ -13120,8 +13120,8 @@ define <8 x i16> @mgather_gather_2xSEW_unaligned(ptr %base) {
 ; RV64ZVE32F-NEXT:    vslide1down.vx v10, v8, a3
 ; RV64ZVE32F-NEXT:    vslide1down.vx v8, v9, a4
 ; RV64ZVE32F-NEXT:    vslide1down.vx v8, v8, a5
-; RV64ZVE32F-NEXT:    vmv.v.i v0, 15
 ; RV64ZVE32F-NEXT:    vslide1down.vx v8, v8, a6
+; RV64ZVE32F-NEXT:    vmv.v.i v0, 15
 ; RV64ZVE32F-NEXT:    vslidedown.vi v8, v10, 4, v0.t
 ; RV64ZVE32F-NEXT:    ret
   %ptrs = getelementptr inbounds i16, ptr %base, <8 x i32>  <i32 0, i32 1, i32 9, i32 10, i32 4, i32 5, i32 2, i32 3>
@@ -13167,8 +13167,8 @@ define <8 x i16> @mgather_gather_2xSEW_unaligned2(ptr %base) {
 ; RV64ZVE32F-NEXT:    vslide1down.vx v10, v8, a4
 ; RV64ZVE32F-NEXT:    vslide1down.vx v8, v9, a5
 ; RV64ZVE32F-NEXT:    vslide1down.vx v8, v8, a2
-; RV64ZVE32F-NEXT:    vmv.v.i v0, 15
 ; RV64ZVE32F-NEXT:    vslide1down.vx v8, v8, a6
+; RV64ZVE32F-NEXT:    vmv.v.i v0, 15
 ; RV64ZVE32F-NEXT:    vslidedown.vi v8, v10, 4, v0.t
 ; RV64ZVE32F-NEXT:    ret
   %ptrs = getelementptr inbounds i16, ptr %base, <8 x i32>  <i32 1, i32 2, i32 9, i32 10, i32 4, i32 5, i32 2, i32 3>
@@ -13217,8 +13217,8 @@ define <8 x i16> @mgather_gather_4xSEW(ptr %base) {
 ; RV64ZVE32F-NEXT:    vslide1down.vx v10, v8, a3
 ; RV64ZVE32F-NEXT:    vslide1down.vx v8, v9, a4
 ; RV64ZVE32F-NEXT:    vslide1down.vx v8, v8, a5
-; RV64ZVE32F-NEXT:    vmv.v.i v0, 15
 ; RV64ZVE32F-NEXT:    vslide1down.vx v8, v8, a6
+; RV64ZVE32F-NEXT:    vmv.v.i v0, 15
 ; RV64ZVE32F-NEXT:    vslidedown.vi v8, v10, 4, v0.t
 ; RV64ZVE32F-NEXT:    ret
   %ptrs = getelementptr inbounds i16, ptr %base, <8 x i32>  <i32 0, i32 1, i32 2, i32 3, i32 8, i32 9, i32 10, i32 11>
@@ -13264,8 +13264,8 @@ define <8 x i16> @mgather_gather_4xSEW_partial_align(ptr %base) {
 ; RV64ZVE32F-NEXT:    vslide1down.vx v10, v8, a3
 ; RV64ZVE32F-NEXT:    vslide1down.vx v8, v9, a4
 ; RV64ZVE32F-NEXT:    vslide1down.vx v8, v8, a5
-; RV64ZVE32F-NEXT:    vmv.v.i v0, 15
 ; RV64ZVE32F-NEXT:    vslide1down.vx v8, v8, a6
+; RV64ZVE32F-NEXT:    vmv.v.i v0, 15
 ; RV64ZVE32F-NEXT:    vslidedown.vi v8, v10, 4, v0.t
 ; RV64ZVE32F-NEXT:    ret
   %ptrs = getelementptr inbounds i16, ptr %base, <8 x i32>  <i32 0, i32 1, i32 2, i32 3, i32 8, i32 9, i32 10, i32 11>
@@ -13320,8 +13320,8 @@ define <8 x i16> @mgather_shuffle_rotate(ptr %base) {
 ; RV64ZVE32F-NEXT:    vslide1down.vx v8, v8, a6
 ; RV64ZVE32F-NEXT:    vslide1down.vx v9, v9, a1
 ; RV64ZVE32F-NEXT:    vslide1down.vx v9, v9, a2
-; RV64ZVE32F-NEXT:    vmv.v.i v0, 15
 ; RV64ZVE32F-NEXT:    vslide1down.vx v9, v9, a3
+; RV64ZVE32F-NEXT:    vmv.v.i v0, 15
 ; RV64ZVE32F-NEXT:    vslidedown.vi v8, v9, 4, v0.t
 ; RV64ZVE32F-NEXT:    ret
   %ptrs = getelementptr inbounds i16, ptr %base, <8 x i64>  <i64 4, i64 5, i64 6, i64 7, i64 0, i64 1, i64 2, i64 3>
@@ -13367,8 +13367,8 @@ define <8 x i16> @mgather_shuffle_vrgather(ptr %base) {
 ; RV64ZVE32F-NEXT:    vslide1down.vx v10, v8, a3
 ; RV64ZVE32F-NEXT:    vslide1down.vx v8, v9, a4
 ; RV64ZVE32F-NEXT:    vslide1down.vx v8, v8, a5
-; RV64ZVE32F-NEXT:    vmv.v.i v0, 15
 ; RV64ZVE32F-NEXT:    vslide1down.vx v8, v8, a6
+; RV64ZVE32F-NEXT:    vmv.v.i v0, 15
 ; RV64ZVE32F-NEXT:    vslidedown.vi v8, v10, 4, v0.t
 ; RV64ZVE32F-NEXT:    ret
   %ptrs = getelementptr inbounds i16, ptr %base, <8 x i64>  <i64 0, i64 2, i64 3, i64 1, i64 4, i64 5, i64 6, i64 7>
diff --git a/llvm/test/CodeGen/RISCV/rvv/fixed-vectors-reduction-formation.ll b/llvm/test/CodeGen/RISCV/rvv/fixed-vectors-reduction-formation.ll
index 2a0ec47a3de0..5f456c782431 100644
--- a/llvm/test/CodeGen/RISCV/rvv/fixed-vectors-reduction-formation.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/fixed-vectors-reduction-formation.ll
@@ -221,10 +221,10 @@ define i32 @reduce_sum_16xi32_prefix7(ptr %p) {
 ; CHECK-LABEL: reduce_sum_16xi32_prefix7:
 ; CHECK:       # %bb.0:
 ; CHECK-NEXT:    vsetivli zero, 8, e32, m2, ta, ma
-; CHECK-NEXT:    vle32.v v8, (a0)
-; CHECK-NEXT:    vmv.s.x v10, zero
-; CHECK-NEXT:    vslideup.vi v8, v10, 7
-; CHECK-NEXT:    vredsum.vs v8, v8, v10
+; CHECK-NEXT:    vmv.s.x v8, zero
+; CHECK-NEXT:    vle32.v v10, (a0)
+; CHECK-NEXT:    vslideup.vi v10, v8, 7
+; CHECK-NEXT:    vredsum.vs v8, v10, v8
 ; CHECK-NEXT:    vmv.x.s a0, v8
 ; CHECK-NEXT:    ret
   %v = load <16 x i32>, ptr %p, align 256
@@ -248,9 +248,9 @@ define i32 @reduce_sum_16xi32_prefix8(ptr %p) {
 ; CHECK-LABEL: reduce_sum_16xi32_prefix8:
 ; CHECK:       # %bb.0:
 ; CHECK-NEXT:    vsetivli zero, 8, e32, m2, ta, ma
-; CHECK-NEXT:    vle32.v v8, (a0)
-; CHECK-NEXT:    vmv.s.x v10, zero
-; CHECK-NEXT:    vredsum.vs v8, v8, v10
+; CHECK-NEXT:    vmv.s.x v8, zero
+; CHECK-NEXT:    vle32.v v10, (a0)
+; CHECK-NEXT:    vredsum.vs v8, v10, v8
 ; CHECK-NEXT:    vmv.x.s a0, v8
 ; CHECK-NEXT:    ret
   %v = load <16 x i32>, ptr %p, align 256
@@ -670,15 +670,15 @@ define i32 @reduce_smax_16xi32_prefix5(ptr %p) {
 ; CHECK:       # %bb.0:
 ; CHECK-NEXT:    lui a1, 524288
 ; CHECK-NEXT:    vsetivli zero, 8, e32, m2, ta, ma
-; CHECK-NEXT:    vle32.v v8, (a0)
-; CHECK-NEXT:    vmv.s.x v10, a1
+; CHECK-NEXT:    vmv.s.x v8, a1
+; CHECK-NEXT:    vle32.v v10, (a0)
 ; CHECK-NEXT:    vsetivli zero, 6, e32, m2, tu, ma
-; CHECK-NEXT:    vslideup.vi v8, v10, 5
+; CHECK-NEXT:    vslideup.vi v10, v8, 5
 ; CHECK-NEXT:    vsetivli zero, 7, e32, m2, tu, ma
-; CHECK-NEXT:    vslideup.vi v8, v10, 6
+; CHECK-NEXT:    vslideup.vi v10, v8, 6
 ; CHECK-NEXT:    vsetivli zero, 8, e32, m2, ta, ma
-; CHECK-NEXT:    vslideup.vi v8, v10, 7
-; CHECK-NEXT:    vredmax.vs v8, v8, v8
+; CHECK-NEXT:    vslideup.vi v10, v8, 7
+; CHECK-NEXT:    vredmax.vs v8, v10, v10
 ; CHECK-NEXT:    vmv.x.s a0, v8
 ; CHECK-NEXT:    ret
   %v = load <16 x i32>, ptr %p, align 256
@@ -715,15 +715,15 @@ define i32 @reduce_smin_16xi32_prefix5(ptr %p) {
 ; CHECK-NEXT:    lui a1, 524288
 ; CHECK-NEXT:    addi a1, a1, -1
 ; CHECK-NEXT:    vsetivli zero, 8, e32, m2, ta, ma
-; CHECK-NEXT:    vle32.v v8, (a0)
-; CHECK-NEXT:    vmv.s.x v10, a1
+; CHECK-NEXT:    vmv.s.x v8, a1
+; CHECK-NEXT:    vle32.v v10, (a0)
 ; CHECK-NEXT:    vsetivli zero, 6, e32, m2, tu, ma
-; CHECK-NEXT:    vslideup.vi v8, v10, 5
+; CHECK-NEXT:    vslideup.vi v10, v8, 5
 ; CHECK-NEXT:    vsetivli zero, 7, e32, m2, tu, ma
-; CHECK-NEXT:    vslideup.vi v8, v10, 6
+; CHECK-NEXT:    vslideup.vi v10, v8, 6
 ; CHECK-NEXT:    vsetivli zero, 8, e32, m2, ta, ma
-; CHECK-NEXT:    vslideup.vi v8, v10, 7
-; CHECK-NEXT:    vredmin.vs v8, v8, v8
+; CHECK-NEXT:    vslideup.vi v10, v8, 7
+; CHECK-NEXT:    vredmin.vs v8, v10, v10
 ; CHECK-NEXT:    vmv.x.s a0, v8
 ; CHECK-NEXT:    ret
   %v = load <16 x i32>, ptr %p, align 256
@@ -830,9 +830,9 @@ define float @reduce_fadd_16xf32_prefix2(ptr %p) {
 ; CHECK-LABEL: reduce_fadd_16xf32_prefix2:
 ; CHECK:       # %bb.0:
 ; CHECK-NEXT:    vsetivli zero, 2, e32, mf2, ta, ma
-; CHECK-NEXT:    vle32.v v8, (a0)
-; CHECK-NEXT:    vmv.s.x v9, zero
-; CHECK-NEXT:    vfredusum.vs v8, v8, v9
+; CHECK-NEXT:    vmv.s.x v8, zero
+; CHECK-NEXT:    vle32.v v9, (a0)
+; CHECK-NEXT:    vfredusum.vs v8, v9, v8
 ; CHECK-NEXT:    vfmv.f.s fa0, v8
 ; CHECK-NEXT:    ret
   %v = load <16 x float>, ptr %p, align 256
@@ -847,15 +847,15 @@ define float @reduce_fadd_16xi32_prefix5(ptr %p) {
 ; CHECK:       # %bb.0:
 ; CHECK-NEXT:    lui a1, 524288
 ; CHECK-NEXT:    vsetivli zero, 8, e32, m2, ta, ma
-; CHECK-NEXT:    vle32.v v8, (a0)
-; CHECK-NEXT:    vmv.s.x v10, a1
+; CHECK-NEXT:    vmv.s.x v8, a1
+; CHECK-NEXT:    vle32.v v10, (a0)
 ; CHECK-NEXT:    vsetivli zero, 6, e32, m2, tu, ma
-; CHECK-NEXT:    vslideup.vi v8, v10, 5
+; CHECK-NEXT:    vslideup.vi v10, v8, 5
 ; CHECK-NEXT:    vsetivli zero, 7, e32, m2, tu, ma
-; CHECK-NEXT:    vslideup.vi v8, v10, 6
+; CHECK-NEXT:    vslideup.vi v10, v8, 6
 ; CHECK-NEXT:    vsetivli zero, 8, e32, m2, ta, ma
-; CHECK-NEXT:    vslideup.vi v8, v10, 7
-; CHECK-NEXT:    vfredusum.vs v8, v8, v10
+; CHECK-NEXT:    vslideup.vi v10, v8, 7
+; CHECK-NEXT:    vfredusum.vs v8, v10, v8
 ; CHECK-NEXT:    vfmv.f.s fa0, v8
 ; CHECK-NEXT:    ret
   %v = load <16 x float>, ptr %p, align 256
diff --git a/llvm/test/CodeGen/RISCV/rvv/fixed-vectors-reduction-int-vp.ll b/llvm/test/CodeGen/RISCV/rvv/fixed-vectors-reduction-int-vp.ll
index 2495178ea762..02a989a96996 100644
--- a/llvm/test/CodeGen/RISCV/rvv/fixed-vectors-reduction-int-vp.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/fixed-vectors-reduction-int-vp.ll
@@ -574,24 +574,14 @@ define signext i32 @vpreduce_add_v2i32(i32 signext %s, <2 x i32> %v, <2 x i1> %m
 declare i32 @llvm.vp.reduce.umax.v2i32(i32, <2 x i32>, <2 x i1>, i32)
 
 define signext i32 @vpreduce_umax_v2i32(i32 signext %s, <2 x i32> %v, <2 x i1> %m, i32 zeroext %evl) {
-; RV32-LABEL: vpreduce_umax_v2i32:
-; RV32:       # %bb.0:
-; RV32-NEXT:    vsetivli zero, 1, e32, m1, ta, ma
-; RV32-NEXT:    vmv.s.x v9, a0
-; RV32-NEXT:    vsetvli zero, a1, e32, mf2, ta, ma
-; RV32-NEXT:    vredmaxu.vs v9, v8, v9, v0.t
-; RV32-NEXT:    vmv.x.s a0, v9
-; RV32-NEXT:    ret
-;
-; RV64-LABEL: vpreduce_umax_v2i32:
-; RV64:       # %bb.0:
-; RV64-NEXT:    andi a0, a0, -1
-; RV64-NEXT:    vsetivli zero, 1, e32, m1, ta, ma
-; RV64-NEXT:    vmv.s.x v9, a0
-; RV64-NEXT:    vsetvli zero, a1, e32, mf2, ta, ma
-; RV64-NEXT:    vredmaxu.vs v9, v8, v9, v0.t
-; RV64-NEXT:    vmv.x.s a0, v9
-; RV64-NEXT:    ret
+; CHECK-LABEL: vpreduce_umax_v2i32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli zero, 1, e32, m1, ta, ma
+; CHECK-NEXT:    vmv.s.x v9, a0
+; CHECK-NEXT:    vsetvli zero, a1, e32, mf2, ta, ma
+; CHECK-NEXT:    vredmaxu.vs v9, v8, v9, v0.t
+; CHECK-NEXT:    vmv.x.s a0, v9
+; CHECK-NEXT:    ret
   %r = call i32 @llvm.vp.reduce.umax.v2i32(i32 %s, <2 x i32> %v, <2 x i1> %m, i32 %evl)
   ret i32 %r
 }
@@ -614,24 +604,14 @@ define signext i32 @vpreduce_smax_v2i32(i32 signext %s, <2 x i32> %v, <2 x i1> %
 declare i32 @llvm.vp.reduce.umin.v2i32(i32, <2 x i32>, <2 x i1>, i32)
 
 define signext i32 @vpreduce_umin_v2i32(i32 signext %s, <2 x i32> %v, <2 x i1> %m, i32 zeroext %evl) {
-; RV32-LABEL: vpreduce_umin_v2i32:
-; RV32:       # %bb.0:
-; RV32-NEXT:    vsetivli zero, 1, e32, m1, ta, ma
-; RV32-NEXT:    vmv.s.x v9, a0
-; RV32-NEXT:    vsetvli zero, a1, e32, mf2, ta, ma
-; RV32-NEXT:    vredminu.vs v9, v8, v9, v0.t
-; RV32-NEXT:    vmv.x.s a0, v9
-; RV32-NEXT:    ret
-;
-; RV64-LABEL: vpreduce_umin_v2i32:
-; RV64:       # %bb.0:
-; RV64-NEXT:    andi a0, a0, -1
-; RV64-NEXT:    vsetivli zero, 1, e32, m1, ta, ma
-; RV64-NEXT:    vmv.s.x v9, a0
-; RV64-NEXT:    vsetvli zero, a1, e32, mf2, ta, ma
-; RV64-NEXT:    vredminu.vs v9, v8, v9, v0.t
-; RV64-NEXT:    vmv.x.s a0, v9
-; RV64-NEXT:    ret
+; CHECK-LABEL: vpreduce_umin_v2i32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli zero, 1, e32, m1, ta, ma
+; CHECK-NEXT:    vmv.s.x v9, a0
+; CHECK-NEXT:    vsetvli zero, a1, e32, mf2, ta, ma
+; CHECK-NEXT:    vredminu.vs v9, v8, v9, v0.t
+; CHECK-NEXT:    vmv.x.s a0, v9
+; CHECK-NEXT:    ret
   %r = call i32 @llvm.vp.reduce.umin.v2i32(i32 %s, <2 x i32> %v, <2 x i1> %m, i32 %evl)
   ret i32 %r
 }
@@ -714,24 +694,14 @@ define signext i32 @vpreduce_add_v4i32(i32 signext %s, <4 x i32> %v, <4 x i1> %m
 declare i32 @llvm.vp.reduce.umax.v4i32(i32, <4 x i32>, <4 x i1>, i32)
 
 define signext i32 @vpreduce_umax_v4i32(i32 signext %s, <4 x i32> %v, <4 x i1> %m, i32 zeroext %evl) {
-; RV32-LABEL: vpreduce_umax_v4i32:
-; RV32:       # %bb.0:
-; RV32-NEXT:    vsetivli zero, 1, e32, m1, ta, ma
-; RV32-NEXT:    vmv.s.x v9, a0
-; RV32-NEXT:    vsetvli zero, a1, e32, m1, ta, ma
-; RV32-NEXT:    vredmaxu.vs v9, v8, v9, v0.t
-; RV32-NEXT:    vmv.x.s a0, v9
-; RV32-NEXT:    ret
-;
-; RV64-LABEL: vpreduce_umax_v4i32:
-; RV64:       # %bb.0:
-; RV64-NEXT:    andi a0, a0, -1
-; RV64-NEXT:    vsetivli zero, 1, e32, m1, ta, ma
-; RV64-NEXT:    vmv.s.x v9, a0
-; RV64-NEXT:    vsetvli zero, a1, e32, m1, ta, ma
-; RV64-NEXT:    vredmaxu.vs v9, v8, v9, v0.t
-; RV64-NEXT:    vmv.x.s a0, v9
-; RV64-NEXT:    ret
+; CHECK-LABEL: vpreduce_umax_v4i32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli zero, 1, e32, m1, ta, ma
+; CHECK-NEXT:    vmv.s.x v9, a0
+; CHECK-NEXT:    vsetvli zero, a1, e32, m1, ta, ma
+; CHECK-NEXT:    vredmaxu.vs v9, v8, v9, v0.t
+; CHECK-NEXT:    vmv.x.s a0, v9
+; CHECK-NEXT:    ret
   %r = call i32 @llvm.vp.reduce.umax.v4i32(i32 %s, <4 x i32> %v, <4 x i1> %m, i32 %evl)
   ret i32 %r
 }
@@ -754,24 +724,14 @@ define signext i32 @vpreduce_smax_v4i32(i32 signext %s, <4 x i32> %v, <4 x i1> %
 declare i32 @llvm.vp.reduce.umin.v4i32(i32, <4 x i32>, <4 x i1>, i32)
 
 define signext i32 @vpreduce_umin_v4i32(i32 signext %s, <4 x i32> %v, <4 x i1> %m, i32 zeroext %evl) {
-; RV32-LABEL: vpreduce_umin_v4i32:
-; RV32:       # %bb.0:
-; RV32-NEXT:    vsetivli zero, 1, e32, m1, ta, ma
-; RV32-NEXT:    vmv.s.x v9, a0
-; RV32-NEXT:    vsetvli zero, a1, e32, m1, ta, ma
-; RV32-NEXT:    vredminu.vs v9, v8, v9, v0.t
-; RV32-NEXT:    vmv.x.s a0, v9
-; RV32-NEXT:    ret
-;
-; RV64-LABEL: vpreduce_umin_v4i32:
-; RV64:       # %bb.0:
-; RV64-NEXT:    andi a0, a0, -1
-; RV64-NEXT:    vsetivli zero, 1, e32, m1, ta, ma
-; RV64-NEXT:    vmv.s.x v9, a0
-; RV64-NEXT:    vsetvli zero, a1, e32, m1, ta, ma
-; RV64-NEXT:    vredminu.vs v9, v8, v9, v0.t
-; RV64-NEXT:    vmv.x.s a0, v9
-; RV64-NEXT:    ret
+; CHECK-LABEL: vpreduce_umin_v4i32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli zero, 1, e32, m1, ta, ma
+; CHECK-NEXT:    vmv.s.x v9, a0
+; CHECK-NEXT:    vsetvli zero, a1, e32, m1, ta, ma
+; CHECK-NEXT:    vredminu.vs v9, v8, v9, v0.t
+; CHECK-NEXT:    vmv.x.s a0, v9
+; CHECK-NEXT:    ret
   %r = call i32 @llvm.vp.reduce.umin.v4i32(i32 %s, <4 x i32> %v, <4 x i1> %m, i32 %evl)
   ret i32 %r
 }
diff --git a/llvm/test/CodeGen/RISCV/rvv/fixed-vectors-shuffle-reverse-bitrotate.ll b/llvm/test/CodeGen/RISCV/rvv/fixed-vectors-shuffle-reverse-bitrotate.ll
index d4c0477408fd..a81f740f1739 100644
--- a/llvm/test/CodeGen/RISCV/rvv/fixed-vectors-shuffle-reverse-bitrotate.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/fixed-vectors-shuffle-reverse-bitrotate.ll
@@ -16,8 +16,8 @@ define <256 x i1> @reverse_v256i1(<256 x i1> %a) {
 ; CHECK-NEXT:    vrgather.vv v12, v8, v10
 ; CHECK-NEXT:    vmsne.vi v0, v12, 0
 ; CHECK-NEXT:    ret
-  %res = call <256 x i1> @llvm.experimental.vector.reverse.v256i1(<256 x i1> %a)
+  %res = call <256 x i1> @llvm.vector.reverse.v256i1(<256 x i1> %a)
   ret <256 x i1> %res
 }
 
-declare <256 x i1> @llvm.experimental.vector.reverse.v256i1(<256 x i1>)
+declare <256 x i1> @llvm.vector.reverse.v256i1(<256 x i1>)
diff --git a/llvm/test/CodeGen/RISCV/rvv/fixed-vectors-shuffle-reverse.ll b/llvm/test/CodeGen/RISCV/rvv/fixed-vectors-shuffle-reverse.ll
index 8f9f1c2729fc..47d7baade8b4 100644
--- a/llvm/test/CodeGen/RISCV/rvv/fixed-vectors-shuffle-reverse.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/fixed-vectors-shuffle-reverse.ll
@@ -29,7 +29,7 @@ define <2 x i1> @reverse_v2i1(<2 x i1> %a) {
 ; ZVBB-NEXT:    vbrev.v v8, v0
 ; ZVBB-NEXT:    vsrl.vi v0, v8, 6
 ; ZVBB-NEXT:    ret
-  %res = call <2 x i1> @llvm.experimental.vector.reverse.v2i1(<2 x i1> %a)
+  %res = call <2 x i1> @llvm.vector.reverse.v2i1(<2 x i1> %a)
   ret <2 x i1> %res
 }
 
@@ -51,7 +51,7 @@ define <4 x i1> @reverse_v4i1(<4 x i1> %a) {
 ; ZVBB-NEXT:    vbrev.v v8, v0
 ; ZVBB-NEXT:    vsrl.vi v0, v8, 4
 ; ZVBB-NEXT:    ret
-  %res = call <4 x i1> @llvm.experimental.vector.reverse.v4i1(<4 x i1> %a)
+  %res = call <4 x i1> @llvm.vector.reverse.v4i1(<4 x i1> %a)
   ret <4 x i1> %res
 }
 
@@ -72,7 +72,7 @@ define <8 x i1> @reverse_v8i1(<8 x i1> %a) {
 ; ZVBB-NEXT:    vsetivli zero, 1, e8, mf8, ta, ma
 ; ZVBB-NEXT:    vbrev.v v0, v0
 ; ZVBB-NEXT:    ret
-  %res = call <8 x i1> @llvm.experimental.vector.reverse.v8i1(<8 x i1> %a)
+  %res = call <8 x i1> @llvm.vector.reverse.v8i1(<8 x i1> %a)
   ret <8 x i1> %res
 }
 
@@ -93,7 +93,7 @@ define <16 x i1> @reverse_v16i1(<16 x i1> %a) {
 ; ZVBB-NEXT:    vsetivli zero, 1, e16, mf4, ta, ma
 ; ZVBB-NEXT:    vbrev.v v0, v0
 ; ZVBB-NEXT:    ret
-  %res = call <16 x i1> @llvm.experimental.vector.reverse.v16i1(<16 x i1> %a)
+  %res = call <16 x i1> @llvm.vector.reverse.v16i1(<16 x i1> %a)
   ret <16 x i1> %res
 }
 
@@ -116,7 +116,7 @@ define <32 x i1> @reverse_v32i1(<32 x i1> %a) {
 ; ZVBB-NEXT:    vsetivli zero, 1, e32, mf2, ta, ma
 ; ZVBB-NEXT:    vbrev.v v0, v0
 ; ZVBB-NEXT:    ret
-  %res = call <32 x i1> @llvm.experimental.vector.reverse.v32i1(<32 x i1> %a)
+  %res = call <32 x i1> @llvm.vector.reverse.v32i1(<32 x i1> %a)
   ret <32 x i1> %res
 }
 
@@ -139,7 +139,7 @@ define <64 x i1> @reverse_v64i1(<64 x i1> %a) {
 ; ZVBB-NEXT:    vsetivli zero, 1, e64, m1, ta, ma
 ; ZVBB-NEXT:    vbrev.v v0, v0
 ; ZVBB-NEXT:    ret
-  %res = call <64 x i1> @llvm.experimental.vector.reverse.v64i1(<64 x i1> %a)
+  %res = call <64 x i1> @llvm.vector.reverse.v64i1(<64 x i1> %a)
   ret <64 x i1> %res
 }
 
@@ -156,7 +156,7 @@ define <128 x i1> @reverse_v128i1(<128 x i1> %a) {
 ; CHECK-NEXT:    vrgather.vv v24, v16, v8
 ; CHECK-NEXT:    vmsne.vi v0, v24, 0
 ; CHECK-NEXT:    ret
-  %res = call <128 x i1> @llvm.experimental.vector.reverse.v128i1(<128 x i1> %a)
+  %res = call <128 x i1> @llvm.vector.reverse.v128i1(<128 x i1> %a)
   ret <128 x i1> %res
 }
 
@@ -164,7 +164,7 @@ define <1 x i8> @reverse_v1i8(<1 x i8> %a) {
 ; CHECK-LABEL: reverse_v1i8:
 ; CHECK:       # %bb.0:
 ; CHECK-NEXT:    ret
-  %res = call <1 x i8> @llvm.experimental.vector.reverse.v1i8(<1 x i8> %a)
+  %res = call <1 x i8> @llvm.vector.reverse.v1i8(<1 x i8> %a)
   ret <1 x i8> %res
 }
 
@@ -182,7 +182,7 @@ define <2 x i8> @reverse_v2i8(<2 x i8> %a) {
 ; ZVBB-NEXT:    vsetivli zero, 1, e16, mf4, ta, ma
 ; ZVBB-NEXT:    vrev8.v v8, v8
 ; ZVBB-NEXT:    ret
-  %res = call <2 x i8> @llvm.experimental.vector.reverse.v2i8(<2 x i8> %a)
+  %res = call <2 x i8> @llvm.vector.reverse.v2i8(<2 x i8> %a)
   ret <2 x i8> %res
 }
 
@@ -195,7 +195,7 @@ define <4 x i8> @reverse_v4i8(<4 x i8> %a) {
 ; CHECK-NEXT:    vrgather.vv v9, v8, v10
 ; CHECK-NEXT:    vmv1r.v v8, v9
 ; CHECK-NEXT:    ret
-  %res = call <4 x i8> @llvm.experimental.vector.reverse.v4i8(<4 x i8> %a)
+  %res = call <4 x i8> @llvm.vector.reverse.v4i8(<4 x i8> %a)
   ret <4 x i8> %res
 }
 
@@ -208,7 +208,7 @@ define <8 x i8> @reverse_v8i8(<8 x i8> %a) {
 ; CHECK-NEXT:    vrgather.vv v9, v8, v10
 ; CHECK-NEXT:    vmv1r.v v8, v9
 ; CHECK-NEXT:    ret
-  %res = call <8 x i8> @llvm.experimental.vector.reverse.v8i8(<8 x i8> %a)
+  %res = call <8 x i8> @llvm.vector.reverse.v8i8(<8 x i8> %a)
   ret <8 x i8> %res
 }
 
@@ -221,7 +221,7 @@ define <16 x i8> @reverse_v16i8(<16 x i8> %a) {
 ; CHECK-NEXT:    vrgather.vv v9, v8, v10
 ; CHECK-NEXT:    vmv.v.v v8, v9
 ; CHECK-NEXT:    ret
-  %res = call <16 x i8> @llvm.experimental.vector.reverse.v16i8(<16 x i8> %a)
+  %res = call <16 x i8> @llvm.vector.reverse.v16i8(<16 x i8> %a)
   ret <16 x i8> %res
 }
 
@@ -236,7 +236,7 @@ define <32 x i8> @reverse_v32i8(<32 x i8> %a) {
 ; CHECK-NEXT:    vrgather.vv v10, v8, v12
 ; CHECK-NEXT:    vmv.v.v v8, v10
 ; CHECK-NEXT:    ret
-  %res = call <32 x i8> @llvm.experimental.vector.reverse.v32i8(<32 x i8> %a)
+  %res = call <32 x i8> @llvm.vector.reverse.v32i8(<32 x i8> %a)
   ret <32 x i8> %res
 }
 
@@ -251,7 +251,7 @@ define <64 x i8> @reverse_v64i8(<64 x i8> %a) {
 ; CHECK-NEXT:    vrgather.vv v12, v8, v16
 ; CHECK-NEXT:    vmv.v.v v8, v12
 ; CHECK-NEXT:    ret
-  %res = call <64 x i8> @llvm.experimental.vector.reverse.v64i8(<64 x i8> %a)
+  %res = call <64 x i8> @llvm.vector.reverse.v64i8(<64 x i8> %a)
   ret <64 x i8> %res
 }
 
@@ -259,7 +259,7 @@ define <1 x i16> @reverse_v1i16(<1 x i16> %a) {
 ; CHECK-LABEL: reverse_v1i16:
 ; CHECK:       # %bb.0:
 ; CHECK-NEXT:    ret
-  %res = call <1 x i16> @llvm.experimental.vector.reverse.v1i16(<1 x i16> %a)
+  %res = call <1 x i16> @llvm.vector.reverse.v1i16(<1 x i16> %a)
   ret <1 x i16> %res
 }
 
@@ -277,7 +277,7 @@ define <2 x i16> @reverse_v2i16(<2 x i16> %a) {
 ; ZVBB-NEXT:    vsetivli zero, 1, e32, mf2, ta, ma
 ; ZVBB-NEXT:    vror.vi v8, v8, 16
 ; ZVBB-NEXT:    ret
-  %res = call <2 x i16> @llvm.experimental.vector.reverse.v2i16(<2 x i16> %a)
+  %res = call <2 x i16> @llvm.vector.reverse.v2i16(<2 x i16> %a)
   ret <2 x i16> %res
 }
 
@@ -290,7 +290,7 @@ define <4 x i16> @reverse_v4i16(<4 x i16> %a) {
 ; CHECK-NEXT:    vrgather.vv v9, v8, v10
 ; CHECK-NEXT:    vmv1r.v v8, v9
 ; CHECK-NEXT:    ret
-  %res = call <4 x i16> @llvm.experimental.vector.reverse.v4i16(<4 x i16> %a)
+  %res = call <4 x i16> @llvm.vector.reverse.v4i16(<4 x i16> %a)
   ret <4 x i16> %res
 }
 
@@ -303,7 +303,7 @@ define <8 x i16> @reverse_v8i16(<8 x i16> %a) {
 ; CHECK-NEXT:    vrgather.vv v9, v8, v10
 ; CHECK-NEXT:    vmv.v.v v8, v9
 ; CHECK-NEXT:    ret
-  %res = call <8 x i16> @llvm.experimental.vector.reverse.v8i16(<8 x i16> %a)
+  %res = call <8 x i16> @llvm.vector.reverse.v8i16(<8 x i16> %a)
   ret <8 x i16> %res
 }
 
@@ -316,7 +316,7 @@ define <16 x i16> @reverse_v16i16(<16 x i16> %a) {
 ; CHECK-NEXT:    vrgather.vv v10, v8, v12
 ; CHECK-NEXT:    vmv.v.v v8, v10
 ; CHECK-NEXT:    ret
-  %res = call <16 x i16> @llvm.experimental.vector.reverse.v16i16(<16 x i16> %a)
+  %res = call <16 x i16> @llvm.vector.reverse.v16i16(<16 x i16> %a)
   ret <16 x i16> %res
 }
 
@@ -332,7 +332,7 @@ define <32 x i16> @reverse_v32i16(<32 x i16> %a) {
 ; CHECK-NEXT:    vrgather.vv v12, v8, v16
 ; CHECK-NEXT:    vmv.v.v v8, v12
 ; CHECK-NEXT:    ret
-  %res = call <32 x i16> @llvm.experimental.vector.reverse.v32i16(<32 x i16> %a)
+  %res = call <32 x i16> @llvm.vector.reverse.v32i16(<32 x i16> %a)
   ret <32 x i16> %res
 }
 
@@ -340,7 +340,7 @@ define <1 x i32> @reverse_v1i32(<1 x i32> %a) {
 ; CHECK-LABEL: reverse_v1i32:
 ; CHECK:       # %bb.0:
 ; CHECK-NEXT:    ret
-  %res = call <1 x i32> @llvm.experimental.vector.reverse.v1i32(<1 x i32> %a)
+  %res = call <1 x i32> @llvm.vector.reverse.v1i32(<1 x i32> %a)
   ret <1 x i32> %res
 }
 
@@ -358,7 +358,7 @@ define <2 x i32> @reverse_v2i32(<2 x i32> %a) {
 ; ZVBB-NEXT:    vsetivli zero, 1, e64, m1, ta, ma
 ; ZVBB-NEXT:    vror.vi v8, v8, 32
 ; ZVBB-NEXT:    ret
-  %res = call <2 x i32> @llvm.experimental.vector.reverse.v2i32(<2 x i32> %a)
+  %res = call <2 x i32> @llvm.vector.reverse.v2i32(<2 x i32> %a)
   ret <2 x i32> %res
 }
 
@@ -371,7 +371,7 @@ define <4 x i32> @reverse_v4i32(<4 x i32> %a) {
 ; CHECK-NEXT:    vrgather.vv v9, v8, v10
 ; CHECK-NEXT:    vmv.v.v v8, v9
 ; CHECK-NEXT:    ret
-  %res = call <4 x i32> @llvm.experimental.vector.reverse.v4i32(<4 x i32> %a)
+  %res = call <4 x i32> @llvm.vector.reverse.v4i32(<4 x i32> %a)
   ret <4 x i32> %res
 }
 
@@ -385,7 +385,7 @@ define <8 x i32> @reverse_v8i32(<8 x i32> %a) {
 ; CHECK-NEXT:    vrgatherei16.vv v10, v8, v12
 ; CHECK-NEXT:    vmv.v.v v8, v10
 ; CHECK-NEXT:    ret
-  %res = call <8 x i32> @llvm.experimental.vector.reverse.v8i32(<8 x i32> %a)
+  %res = call <8 x i32> @llvm.vector.reverse.v8i32(<8 x i32> %a)
   ret <8 x i32> %res
 }
 
@@ -399,7 +399,7 @@ define <16 x i32> @reverse_v16i32(<16 x i32> %a) {
 ; CHECK-NEXT:    vrgatherei16.vv v12, v8, v16
 ; CHECK-NEXT:    vmv.v.v v8, v12
 ; CHECK-NEXT:    ret
-  %res = call <16 x i32> @llvm.experimental.vector.reverse.v16i32(<16 x i32> %a)
+  %res = call <16 x i32> @llvm.vector.reverse.v16i32(<16 x i32> %a)
   ret <16 x i32> %res
 }
 
@@ -407,7 +407,7 @@ define <1 x i64> @reverse_v1i64(<1 x i64> %a) {
 ; CHECK-LABEL: reverse_v1i64:
 ; CHECK:       # %bb.0:
 ; CHECK-NEXT:    ret
-  %res = call <1 x i64> @llvm.experimental.vector.reverse.v1i64(<1 x i64> %a)
+  %res = call <1 x i64> @llvm.vector.reverse.v1i64(<1 x i64> %a)
   ret <1 x i64> %res
 }
 
@@ -419,7 +419,7 @@ define <2 x i64> @reverse_v2i64(<2 x i64> %a) {
 ; CHECK-NEXT:    vslideup.vi v9, v8, 1
 ; CHECK-NEXT:    vmv.v.v v8, v9
 ; CHECK-NEXT:    ret
-  %res = call <2 x i64> @llvm.experimental.vector.reverse.v2i64(<2 x i64> %a)
+  %res = call <2 x i64> @llvm.vector.reverse.v2i64(<2 x i64> %a)
   ret <2 x i64> %res
 }
 
@@ -433,7 +433,7 @@ define <4 x i64> @reverse_v4i64(<4 x i64> %a) {
 ; CHECK-NEXT:    vrgatherei16.vv v10, v8, v12
 ; CHECK-NEXT:    vmv.v.v v8, v10
 ; CHECK-NEXT:    ret
-  %res = call <4 x i64> @llvm.experimental.vector.reverse.v4i64(<4 x i64> %a)
+  %res = call <4 x i64> @llvm.vector.reverse.v4i64(<4 x i64> %a)
   ret <4 x i64> %res
 }
 
@@ -447,7 +447,7 @@ define <8 x i64> @reverse_v8i64(<8 x i64> %a) {
 ; CHECK-NEXT:    vrgatherei16.vv v12, v8, v16
 ; CHECK-NEXT:    vmv.v.v v8, v12
 ; CHECK-NEXT:    ret
-  %res = call <8 x i64> @llvm.experimental.vector.reverse.v8i64(<8 x i64> %a)
+  %res = call <8 x i64> @llvm.vector.reverse.v8i64(<8 x i64> %a)
   ret <8 x i64> %res
 }
 
@@ -456,7 +456,7 @@ define <1 x half> @reverse_v1f16(<1 x half> %a) {
 ; CHECK-LABEL: reverse_v1f16:
 ; CHECK:       # %bb.0:
 ; CHECK-NEXT:    ret
-  %res = call <1 x half> @llvm.experimental.vector.reverse.v1f16(<1 x half> %a)
+  %res = call <1 x half> @llvm.vector.reverse.v1f16(<1 x half> %a)
   ret <1 x half> %res
 }
 
@@ -474,7 +474,7 @@ define <2 x half> @reverse_v2f16(<2 x half> %a) {
 ; ZVBB-NEXT:    vsetivli zero, 1, e32, mf2, ta, ma
 ; ZVBB-NEXT:    vror.vi v8, v8, 16
 ; ZVBB-NEXT:    ret
-  %res = call <2 x half> @llvm.experimental.vector.reverse.v2f16(<2 x half> %a)
+  %res = call <2 x half> @llvm.vector.reverse.v2f16(<2 x half> %a)
   ret <2 x half> %res
 }
 
@@ -487,7 +487,7 @@ define <4 x half> @reverse_v4f16(<4 x half> %a) {
 ; CHECK-NEXT:    vrgather.vv v9, v8, v10
 ; CHECK-NEXT:    vmv1r.v v8, v9
 ; CHECK-NEXT:    ret
-  %res = call <4 x half> @llvm.experimental.vector.reverse.v4f16(<4 x half> %a)
+  %res = call <4 x half> @llvm.vector.reverse.v4f16(<4 x half> %a)
   ret <4 x half> %res
 }
 
@@ -500,7 +500,7 @@ define <8 x half> @reverse_v8f16(<8 x half> %a) {
 ; CHECK-NEXT:    vrgather.vv v9, v8, v10
 ; CHECK-NEXT:    vmv.v.v v8, v9
 ; CHECK-NEXT:    ret
-  %res = call <8 x half> @llvm.experimental.vector.reverse.v8f16(<8 x half> %a)
+  %res = call <8 x half> @llvm.vector.reverse.v8f16(<8 x half> %a)
   ret <8 x half> %res
 }
 
@@ -513,7 +513,7 @@ define <16 x half> @reverse_v16f16(<16 x half> %a) {
 ; CHECK-NEXT:    vrgather.vv v10, v8, v12
 ; CHECK-NEXT:    vmv.v.v v8, v10
 ; CHECK-NEXT:    ret
-  %res = call <16 x half> @llvm.experimental.vector.reverse.v16f16(<16 x half> %a)
+  %res = call <16 x half> @llvm.vector.reverse.v16f16(<16 x half> %a)
   ret <16 x half> %res
 }
 
@@ -529,7 +529,7 @@ define <32 x half> @reverse_v32f16(<32 x half> %a) {
 ; CHECK-NEXT:    vrgather.vv v12, v8, v16
 ; CHECK-NEXT:    vmv.v.v v8, v12
 ; CHECK-NEXT:    ret
-  %res = call <32 x half> @llvm.experimental.vector.reverse.v32f16(<32 x half> %a)
+  %res = call <32 x half> @llvm.vector.reverse.v32f16(<32 x half> %a)
   ret <32 x half> %res
 }
 
@@ -537,7 +537,7 @@ define <1 x float> @reverse_v1f32(<1 x float> %a) {
 ; CHECK-LABEL: reverse_v1f32:
 ; CHECK:       # %bb.0:
 ; CHECK-NEXT:    ret
-  %res = call <1 x float> @llvm.experimental.vector.reverse.v1f32(<1 x float> %a)
+  %res = call <1 x float> @llvm.vector.reverse.v1f32(<1 x float> %a)
   ret <1 x float> %res
 }
 
@@ -555,7 +555,7 @@ define <2 x float> @reverse_v2f32(<2 x float> %a) {
 ; ZVBB-NEXT:    vsetivli zero, 1, e64, m1, ta, ma
 ; ZVBB-NEXT:    vror.vi v8, v8, 32
 ; ZVBB-NEXT:    ret
-  %res = call <2 x float> @llvm.experimental.vector.reverse.v2f32(<2 x float> %a)
+  %res = call <2 x float> @llvm.vector.reverse.v2f32(<2 x float> %a)
   ret <2 x float> %res
 }
 
@@ -568,7 +568,7 @@ define <4 x float> @reverse_v4f32(<4 x float> %a) {
 ; CHECK-NEXT:    vrgather.vv v9, v8, v10
 ; CHECK-NEXT:    vmv.v.v v8, v9
 ; CHECK-NEXT:    ret
-  %res = call <4 x float> @llvm.experimental.vector.reverse.v4f32(<4 x float> %a)
+  %res = call <4 x float> @llvm.vector.reverse.v4f32(<4 x float> %a)
   ret <4 x float> %res
 }
 
@@ -582,7 +582,7 @@ define <8 x float> @reverse_v8f32(<8 x float> %a) {
 ; CHECK-NEXT:    vrgatherei16.vv v10, v8, v12
 ; CHECK-NEXT:    vmv.v.v v8, v10
 ; CHECK-NEXT:    ret
-  %res = call <8 x float> @llvm.experimental.vector.reverse.v8f32(<8 x float> %a)
+  %res = call <8 x float> @llvm.vector.reverse.v8f32(<8 x float> %a)
   ret <8 x float> %res
 }
 
@@ -596,7 +596,7 @@ define <16 x float> @reverse_v16f32(<16 x float> %a) {
 ; CHECK-NEXT:    vrgatherei16.vv v12, v8, v16
 ; CHECK-NEXT:    vmv.v.v v8, v12
 ; CHECK-NEXT:    ret
-  %res = call <16 x float> @llvm.experimental.vector.reverse.v16f32(<16 x float> %a)
+  %res = call <16 x float> @llvm.vector.reverse.v16f32(<16 x float> %a)
   ret <16 x float> %res
 }
 
@@ -604,7 +604,7 @@ define <1 x double> @reverse_v1f64(<1 x double> %a) {
 ; CHECK-LABEL: reverse_v1f64:
 ; CHECK:       # %bb.0:
 ; CHECK-NEXT:    ret
-  %res = call <1 x double> @llvm.experimental.vector.reverse.v1f64(<1 x double> %a)
+  %res = call <1 x double> @llvm.vector.reverse.v1f64(<1 x double> %a)
   ret <1 x double> %res
 }
 
@@ -616,7 +616,7 @@ define <2 x double> @reverse_v2f64(<2 x double> %a) {
 ; CHECK-NEXT:    vslideup.vi v9, v8, 1
 ; CHECK-NEXT:    vmv.v.v v8, v9
 ; CHECK-NEXT:    ret
-  %res = call <2 x double> @llvm.experimental.vector.reverse.v2f64(<2 x double> %a)
+  %res = call <2 x double> @llvm.vector.reverse.v2f64(<2 x double> %a)
   ret <2 x double> %res
 }
 
@@ -630,7 +630,7 @@ define <4 x double> @reverse_v4f64(<4 x double> %a) {
 ; CHECK-NEXT:    vrgatherei16.vv v10, v8, v12
 ; CHECK-NEXT:    vmv.v.v v8, v10
 ; CHECK-NEXT:    ret
-  %res = call <4 x double> @llvm.experimental.vector.reverse.v4f64(<4 x double> %a)
+  %res = call <4 x double> @llvm.vector.reverse.v4f64(<4 x double> %a)
   ret <4 x double> %res
 }
 
@@ -644,7 +644,7 @@ define <8 x double> @reverse_v8f64(<8 x double> %a) {
 ; CHECK-NEXT:    vrgatherei16.vv v12, v8, v16
 ; CHECK-NEXT:    vmv.v.v v8, v12
 ; CHECK-NEXT:    ret
-  %res = call <8 x double> @llvm.experimental.vector.reverse.v8f64(<8 x double> %a)
+  %res = call <8 x double> @llvm.vector.reverse.v8f64(<8 x double> %a)
   ret <8 x double> %res
 }
 
@@ -729,7 +729,7 @@ define <3 x i64> @reverse_v3i64(<3 x i64> %a) {
 ; RV64-ZVBB-NEXT:    vrgatherei16.vv v10, v8, v12
 ; RV64-ZVBB-NEXT:    vmv.v.v v8, v10
 ; RV64-ZVBB-NEXT:    ret
-  %res = call <3 x i64> @llvm.experimental.vector.reverse.v3i64(<3 x i64> %a)
+  %res = call <3 x i64> @llvm.vector.reverse.v3i64(<3 x i64> %a)
   ret <3 x i64> %res
 }
 
@@ -813,7 +813,7 @@ define <6 x i64> @reverse_v6i64(<6 x i64> %a) {
 ; RV64-ZVBB-NEXT:    vrgatherei16.vv v12, v8, v16
 ; RV64-ZVBB-NEXT:    vmv.v.v v8, v12
 ; RV64-ZVBB-NEXT:    ret
-  %res = call <6 x i64> @llvm.experimental.vector.reverse.v6i64(<6 x i64> %a)
+  %res = call <6 x i64> @llvm.vector.reverse.v6i64(<6 x i64> %a)
   ret <6 x i64> %res
 }
 
@@ -901,54 +901,54 @@ define <12 x i64> @reverse_v12i64(<12 x i64> %a) {
 ; RV64-ZVBB-NEXT:    vrgatherei16.vv v16, v8, v24
 ; RV64-ZVBB-NEXT:    vmv.v.v v8, v16
 ; RV64-ZVBB-NEXT:    ret
-  %res = call <12 x i64> @llvm.experimental.vector.reverse.v12i64(<12 x i64> %a)
+  %res = call <12 x i64> @llvm.vector.reverse.v12i64(<12 x i64> %a)
   ret <12 x i64> %res
 }
 
-declare <2 x i1> @llvm.experimental.vector.reverse.v2i1(<2 x i1>)
-declare <4 x i1> @llvm.experimental.vector.reverse.v4i1(<4 x i1>)
-declare <8 x i1> @llvm.experimental.vector.reverse.v8i1(<8 x i1>)
-declare <16 x i1> @llvm.experimental.vector.reverse.v16i1(<16 x i1>)
-declare <32 x i1> @llvm.experimental.vector.reverse.v32i1(<32 x i1>)
-declare <64 x i1> @llvm.experimental.vector.reverse.v64i1(<64 x i1>)
-declare <128 x i1> @llvm.experimental.vector.reverse.v128i1(<128 x i1>)
-declare <1 x i8> @llvm.experimental.vector.reverse.v1i8(<1 x i8>)
-declare <2 x i8> @llvm.experimental.vector.reverse.v2i8(<2 x i8>)
-declare <4 x i8> @llvm.experimental.vector.reverse.v4i8(<4 x i8>)
-declare <8 x i8> @llvm.experimental.vector.reverse.v8i8(<8 x i8>)
-declare <16 x i8> @llvm.experimental.vector.reverse.v16i8(<16 x i8>)
-declare <32 x i8> @llvm.experimental.vector.reverse.v32i8(<32 x i8>)
-declare <64 x i8> @llvm.experimental.vector.reverse.v64i8(<64 x i8>)
-declare <1 x i16> @llvm.experimental.vector.reverse.v1i16(<1 x i16>)
-declare <2 x i16> @llvm.experimental.vector.reverse.v2i16(<2 x i16>)
-declare <4 x i16> @llvm.experimental.vector.reverse.v4i16(<4 x i16>)
-declare <8 x i16> @llvm.experimental.vector.reverse.v8i16(<8 x i16>)
-declare <16 x i16> @llvm.experimental.vector.reverse.v16i16(<16 x i16>)
-declare <32 x i16> @llvm.experimental.vector.reverse.v32i16(<32 x i16>)
-declare <1 x i32> @llvm.experimental.vector.reverse.v1i32(<1 x i32>)
-declare <2 x i32> @llvm.experimental.vector.reverse.v2i32(<2 x i32>)
-declare <4 x i32> @llvm.experimental.vector.reverse.v4i32(<4 x i32>)
-declare <8 x i32> @llvm.experimental.vector.reverse.v8i32(<8 x i32>)
-declare <16 x i32> @llvm.experimental.vector.reverse.v16i32(<16 x i32>)
-declare <1 x i64> @llvm.experimental.vector.reverse.v1i64(<1 x i64>)
-declare <2 x i64> @llvm.experimental.vector.reverse.v2i64(<2 x i64>)
-declare <4 x i64> @llvm.experimental.vector.reverse.v4i64(<4 x i64>)
-declare <8 x i64> @llvm.experimental.vector.reverse.v8i64(<8 x i64>)
-declare <1 x half> @llvm.experimental.vector.reverse.v1f16(<1 x half>)
-declare <2 x half> @llvm.experimental.vector.reverse.v2f16(<2 x half>)
-declare <4 x half> @llvm.experimental.vector.reverse.v4f16(<4 x half>)
-declare <8 x half> @llvm.experimental.vector.reverse.v8f16(<8 x half>)
-declare <16 x half> @llvm.experimental.vector.reverse.v16f16(<16 x half>)
-declare <32 x half> @llvm.experimental.vector.reverse.v32f16(<32 x half>)
-declare <1 x float> @llvm.experimental.vector.reverse.v1f32(<1 x float>)
-declare <2 x float> @llvm.experimental.vector.reverse.v2f32(<2 x float>)
-declare <4 x float> @llvm.experimental.vector.reverse.v4f32(<4 x float>)
-declare <8 x float> @llvm.experimental.vector.reverse.v8f32(<8 x float>)
-declare <16 x float> @llvm.experimental.vector.reverse.v16f32(<16 x float>)
-declare <1 x double> @llvm.experimental.vector.reverse.v1f64(<1 x double>)
-declare <2 x double> @llvm.experimental.vector.reverse.v2f64(<2 x double>)
-declare <4 x double> @llvm.experimental.vector.reverse.v4f64(<4 x double>)
-declare <8 x double> @llvm.experimental.vector.reverse.v8f64(<8 x double>)
-declare <3 x i64> @llvm.experimental.vector.reverse.v3i64(<3 x i64>)
-declare <6 x i64> @llvm.experimental.vector.reverse.v6i64(<6 x i64>)
-declare <12 x i64> @llvm.experimental.vector.reverse.v12i64(<12 x i64>)
+declare <2 x i1> @llvm.vector.reverse.v2i1(<2 x i1>)
+declare <4 x i1> @llvm.vector.reverse.v4i1(<4 x i1>)
+declare <8 x i1> @llvm.vector.reverse.v8i1(<8 x i1>)
+declare <16 x i1> @llvm.vector.reverse.v16i1(<16 x i1>)
+declare <32 x i1> @llvm.vector.reverse.v32i1(<32 x i1>)
+declare <64 x i1> @llvm.vector.reverse.v64i1(<64 x i1>)
+declare <128 x i1> @llvm.vector.reverse.v128i1(<128 x i1>)
+declare <1 x i8> @llvm.vector.reverse.v1i8(<1 x i8>)
+declare <2 x i8> @llvm.vector.reverse.v2i8(<2 x i8>)
+declare <4 x i8> @llvm.vector.reverse.v4i8(<4 x i8>)
+declare <8 x i8> @llvm.vector.reverse.v8i8(<8 x i8>)
+declare <16 x i8> @llvm.vector.reverse.v16i8(<16 x i8>)
+declare <32 x i8> @llvm.vector.reverse.v32i8(<32 x i8>)
+declare <64 x i8> @llvm.vector.reverse.v64i8(<64 x i8>)
+declare <1 x i16> @llvm.vector.reverse.v1i16(<1 x i16>)
+declare <2 x i16> @llvm.vector.reverse.v2i16(<2 x i16>)
+declare <4 x i16> @llvm.vector.reverse.v4i16(<4 x i16>)
+declare <8 x i16> @llvm.vector.reverse.v8i16(<8 x i16>)
+declare <16 x i16> @llvm.vector.reverse.v16i16(<16 x i16>)
+declare <32 x i16> @llvm.vector.reverse.v32i16(<32 x i16>)
+declare <1 x i32> @llvm.vector.reverse.v1i32(<1 x i32>)
+declare <2 x i32> @llvm.vector.reverse.v2i32(<2 x i32>)
+declare <4 x i32> @llvm.vector.reverse.v4i32(<4 x i32>)
+declare <8 x i32> @llvm.vector.reverse.v8i32(<8 x i32>)
+declare <16 x i32> @llvm.vector.reverse.v16i32(<16 x i32>)
+declare <1 x i64> @llvm.vector.reverse.v1i64(<1 x i64>)
+declare <2 x i64> @llvm.vector.reverse.v2i64(<2 x i64>)
+declare <4 x i64> @llvm.vector.reverse.v4i64(<4 x i64>)
+declare <8 x i64> @llvm.vector.reverse.v8i64(<8 x i64>)
+declare <1 x half> @llvm.vector.reverse.v1f16(<1 x half>)
+declare <2 x half> @llvm.vector.reverse.v2f16(<2 x half>)
+declare <4 x half> @llvm.vector.reverse.v4f16(<4 x half>)
+declare <8 x half> @llvm.vector.reverse.v8f16(<8 x half>)
+declare <16 x half> @llvm.vector.reverse.v16f16(<16 x half>)
+declare <32 x half> @llvm.vector.reverse.v32f16(<32 x half>)
+declare <1 x float> @llvm.vector.reverse.v1f32(<1 x float>)
+declare <2 x float> @llvm.vector.reverse.v2f32(<2 x float>)
+declare <4 x float> @llvm.vector.reverse.v4f32(<4 x float>)
+declare <8 x float> @llvm.vector.reverse.v8f32(<8 x float>)
+declare <16 x float> @llvm.vector.reverse.v16f32(<16 x float>)
+declare <1 x double> @llvm.vector.reverse.v1f64(<1 x double>)
+declare <2 x double> @llvm.vector.reverse.v2f64(<2 x double>)
+declare <4 x double> @llvm.vector.reverse.v4f64(<4 x double>)
+declare <8 x double> @llvm.vector.reverse.v8f64(<8 x double>)
+declare <3 x i64> @llvm.vector.reverse.v3i64(<3 x i64>)
+declare <6 x i64> @llvm.vector.reverse.v6i64(<6 x i64>)
+declare <12 x i64> @llvm.vector.reverse.v12i64(<12 x i64>)
diff --git a/llvm/test/CodeGen/RISCV/rvv/fpclamptosat_vec.ll b/llvm/test/CodeGen/RISCV/rvv/fpclamptosat_vec.ll
index b3bda5973eb8..a6b2d3141f22 100644
--- a/llvm/test/CodeGen/RISCV/rvv/fpclamptosat_vec.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/fpclamptosat_vec.ll
@@ -2190,65 +2190,66 @@ define <2 x i64> @stest_f64i64(<2 x double> %x) {
 ; CHECK-NOV-NEXT:    .cfi_offset s0, -16
 ; CHECK-NOV-NEXT:    .cfi_offset s1, -24
 ; CHECK-NOV-NEXT:    .cfi_offset fs0, -32
-; CHECK-NOV-NEXT:    fmv.d fs0, fa0
-; CHECK-NOV-NEXT:    fmv.d fa0, fa1
+; CHECK-NOV-NEXT:    fmv.d fs0, fa1
 ; CHECK-NOV-NEXT:    call __fixdfti
 ; CHECK-NOV-NEXT:    mv s0, a0
 ; CHECK-NOV-NEXT:    mv s1, a1
 ; CHECK-NOV-NEXT:    fmv.d fa0, fs0
 ; CHECK-NOV-NEXT:    call __fixdfti
-; CHECK-NOV-NEXT:    li a2, -1
-; CHECK-NOV-NEXT:    srli a3, a2, 1
-; CHECK-NOV-NEXT:    beqz s1, .LBB18_3
+; CHECK-NOV-NEXT:    mv a2, a0
+; CHECK-NOV-NEXT:    li a0, -1
+; CHECK-NOV-NEXT:    srli a3, a0, 1
+; CHECK-NOV-NEXT:    beqz a1, .LBB18_3
 ; CHECK-NOV-NEXT:  # %bb.1: # %entry
-; CHECK-NOV-NEXT:    slti a4, s1, 0
-; CHECK-NOV-NEXT:    bnez a1, .LBB18_4
+; CHECK-NOV-NEXT:    slti a4, a1, 0
+; CHECK-NOV-NEXT:    bnez s1, .LBB18_4
 ; CHECK-NOV-NEXT:  .LBB18_2:
-; CHECK-NOV-NEXT:    sltu a5, a0, a3
+; CHECK-NOV-NEXT:    sltu a5, s0, a3
 ; CHECK-NOV-NEXT:    beqz a5, .LBB18_5
 ; CHECK-NOV-NEXT:    j .LBB18_6
 ; CHECK-NOV-NEXT:  .LBB18_3:
-; CHECK-NOV-NEXT:    sltu a4, s0, a3
-; CHECK-NOV-NEXT:    beqz a1, .LBB18_2
+; CHECK-NOV-NEXT:    sltu a4, a2, a3
+; CHECK-NOV-NEXT:    beqz s1, .LBB18_2
 ; CHECK-NOV-NEXT:  .LBB18_4: # %entry
-; CHECK-NOV-NEXT:    slti a5, a1, 0
+; CHECK-NOV-NEXT:    slti a5, s1, 0
 ; CHECK-NOV-NEXT:    bnez a5, .LBB18_6
 ; CHECK-NOV-NEXT:  .LBB18_5: # %entry
-; CHECK-NOV-NEXT:    mv a0, a3
+; CHECK-NOV-NEXT:    mv s0, a3
 ; CHECK-NOV-NEXT:  .LBB18_6: # %entry
 ; CHECK-NOV-NEXT:    neg a6, a5
 ; CHECK-NOV-NEXT:    neg a5, a4
-; CHECK-NOV-NEXT:    and a5, a5, s1
+; CHECK-NOV-NEXT:    and a5, a5, a1
 ; CHECK-NOV-NEXT:    bnez a4, .LBB18_8
 ; CHECK-NOV-NEXT:  # %bb.7: # %entry
-; CHECK-NOV-NEXT:    mv s0, a3
+; CHECK-NOV-NEXT:    mv a2, a3
 ; CHECK-NOV-NEXT:  .LBB18_8: # %entry
-; CHECK-NOV-NEXT:    and a4, a6, a1
-; CHECK-NOV-NEXT:    slli a1, a2, 63
-; CHECK-NOV-NEXT:    beq a5, a2, .LBB18_11
+; CHECK-NOV-NEXT:    and a4, a6, s1
+; CHECK-NOV-NEXT:    slli a1, a0, 63
+; CHECK-NOV-NEXT:    beq a5, a0, .LBB18_11
 ; CHECK-NOV-NEXT:  # %bb.9: # %entry
 ; CHECK-NOV-NEXT:    slti a3, a5, 0
 ; CHECK-NOV-NEXT:    xori a3, a3, 1
-; CHECK-NOV-NEXT:    bne a4, a2, .LBB18_12
+; CHECK-NOV-NEXT:    bne a4, a0, .LBB18_12
 ; CHECK-NOV-NEXT:  .LBB18_10:
-; CHECK-NOV-NEXT:    sltu a2, a1, a0
-; CHECK-NOV-NEXT:    beqz a2, .LBB18_13
+; CHECK-NOV-NEXT:    sltu a0, a1, s0
+; CHECK-NOV-NEXT:    beqz a0, .LBB18_13
 ; CHECK-NOV-NEXT:    j .LBB18_14
 ; CHECK-NOV-NEXT:  .LBB18_11:
-; CHECK-NOV-NEXT:    sltu a3, a1, s0
-; CHECK-NOV-NEXT:    beq a4, a2, .LBB18_10
+; CHECK-NOV-NEXT:    sltu a3, a1, a2
+; CHECK-NOV-NEXT:    beq a4, a0, .LBB18_10
 ; CHECK-NOV-NEXT:  .LBB18_12: # %entry
-; CHECK-NOV-NEXT:    slti a2, a4, 0
-; CHECK-NOV-NEXT:    xori a2, a2, 1
-; CHECK-NOV-NEXT:    bnez a2, .LBB18_14
+; CHECK-NOV-NEXT:    slti a0, a4, 0
+; CHECK-NOV-NEXT:    xori a0, a0, 1
+; CHECK-NOV-NEXT:    bnez a0, .LBB18_14
 ; CHECK-NOV-NEXT:  .LBB18_13: # %entry
-; CHECK-NOV-NEXT:    mv a0, a1
+; CHECK-NOV-NEXT:    mv s0, a1
 ; CHECK-NOV-NEXT:  .LBB18_14: # %entry
 ; CHECK-NOV-NEXT:    bnez a3, .LBB18_16
 ; CHECK-NOV-NEXT:  # %bb.15: # %entry
-; CHECK-NOV-NEXT:    mv s0, a1
+; CHECK-NOV-NEXT:    mv a2, a1
 ; CHECK-NOV-NEXT:  .LBB18_16: # %entry
-; CHECK-NOV-NEXT:    mv a1, s0
+; CHECK-NOV-NEXT:    mv a0, s0
+; CHECK-NOV-NEXT:    mv a1, a2
 ; CHECK-NOV-NEXT:    ld ra, 24(sp) # 8-byte Folded Reload
 ; CHECK-NOV-NEXT:    ld s0, 16(sp) # 8-byte Folded Reload
 ; CHECK-NOV-NEXT:    ld s1, 8(sp) # 8-byte Folded Reload
@@ -2273,43 +2274,43 @@ define <2 x i64> @stest_f64i64(<2 x double> %x) {
 ; CHECK-V-NEXT:    addi a0, sp, 32
 ; CHECK-V-NEXT:    vs1r.v v8, (a0) # Unknown-size Folded Spill
 ; CHECK-V-NEXT:    vsetivli zero, 1, e64, m1, ta, ma
-; CHECK-V-NEXT:    vfmv.f.s fa0, v8
+; CHECK-V-NEXT:    vslidedown.vi v9, v8, 1
+; CHECK-V-NEXT:    vfmv.f.s fa0, v9
 ; CHECK-V-NEXT:    call __fixdfti
 ; CHECK-V-NEXT:    mv s0, a0
 ; CHECK-V-NEXT:    mv s1, a1
 ; CHECK-V-NEXT:    vsetivli zero, 1, e64, m1, ta, ma
 ; CHECK-V-NEXT:    addi a0, sp, 32
 ; CHECK-V-NEXT:    vl1r.v v8, (a0) # Unknown-size Folded Reload
-; CHECK-V-NEXT:    vslidedown.vi v8, v8, 1
 ; CHECK-V-NEXT:    vfmv.f.s fa0, v8
 ; CHECK-V-NEXT:    call __fixdfti
 ; CHECK-V-NEXT:    li a2, -1
 ; CHECK-V-NEXT:    srli a3, a2, 1
-; CHECK-V-NEXT:    beqz s1, .LBB18_3
+; CHECK-V-NEXT:    beqz a1, .LBB18_3
 ; CHECK-V-NEXT:  # %bb.1: # %entry
-; CHECK-V-NEXT:    slti a4, s1, 0
-; CHECK-V-NEXT:    bnez a1, .LBB18_4
+; CHECK-V-NEXT:    slti a4, a1, 0
+; CHECK-V-NEXT:    bnez s1, .LBB18_4
 ; CHECK-V-NEXT:  .LBB18_2:
-; CHECK-V-NEXT:    sltu a5, a0, a3
+; CHECK-V-NEXT:    sltu a5, s0, a3
 ; CHECK-V-NEXT:    beqz a5, .LBB18_5
 ; CHECK-V-NEXT:    j .LBB18_6
 ; CHECK-V-NEXT:  .LBB18_3:
-; CHECK-V-NEXT:    sltu a4, s0, a3
-; CHECK-V-NEXT:    beqz a1, .LBB18_2
+; CHECK-V-NEXT:    sltu a4, a0, a3
+; CHECK-V-NEXT:    beqz s1, .LBB18_2
 ; CHECK-V-NEXT:  .LBB18_4: # %entry
-; CHECK-V-NEXT:    slti a5, a1, 0
+; CHECK-V-NEXT:    slti a5, s1, 0
 ; CHECK-V-NEXT:    bnez a5, .LBB18_6
 ; CHECK-V-NEXT:  .LBB18_5: # %entry
-; CHECK-V-NEXT:    mv a0, a3
+; CHECK-V-NEXT:    mv s0, a3
 ; CHECK-V-NEXT:  .LBB18_6: # %entry
 ; CHECK-V-NEXT:    neg a6, a5
 ; CHECK-V-NEXT:    neg a5, a4
-; CHECK-V-NEXT:    and a5, a5, s1
+; CHECK-V-NEXT:    and a5, a5, a1
 ; CHECK-V-NEXT:    bnez a4, .LBB18_8
 ; CHECK-V-NEXT:  # %bb.7: # %entry
-; CHECK-V-NEXT:    mv s0, a3
+; CHECK-V-NEXT:    mv a0, a3
 ; CHECK-V-NEXT:  .LBB18_8: # %entry
-; CHECK-V-NEXT:    and a4, a6, a1
+; CHECK-V-NEXT:    and a4, a6, s1
 ; CHECK-V-NEXT:    slli a1, a2, 63
 ; CHECK-V-NEXT:    beq a5, a2, .LBB18_11
 ; CHECK-V-NEXT:  # %bb.9: # %entry
@@ -2317,26 +2318,26 @@ define <2 x i64> @stest_f64i64(<2 x double> %x) {
 ; CHECK-V-NEXT:    xori a3, a3, 1
 ; CHECK-V-NEXT:    bne a4, a2, .LBB18_12
 ; CHECK-V-NEXT:  .LBB18_10:
-; CHECK-V-NEXT:    sltu a2, a1, a0
+; CHECK-V-NEXT:    sltu a2, a1, s0
 ; CHECK-V-NEXT:    beqz a2, .LBB18_13
 ; CHECK-V-NEXT:    j .LBB18_14
 ; CHECK-V-NEXT:  .LBB18_11:
-; CHECK-V-NEXT:    sltu a3, a1, s0
+; CHECK-V-NEXT:    sltu a3, a1, a0
 ; CHECK-V-NEXT:    beq a4, a2, .LBB18_10
 ; CHECK-V-NEXT:  .LBB18_12: # %entry
 ; CHECK-V-NEXT:    slti a2, a4, 0
 ; CHECK-V-NEXT:    xori a2, a2, 1
 ; CHECK-V-NEXT:    bnez a2, .LBB18_14
 ; CHECK-V-NEXT:  .LBB18_13: # %entry
-; CHECK-V-NEXT:    mv a0, a1
+; CHECK-V-NEXT:    mv s0, a1
 ; CHECK-V-NEXT:  .LBB18_14: # %entry
 ; CHECK-V-NEXT:    bnez a3, .LBB18_16
 ; CHECK-V-NEXT:  # %bb.15: # %entry
-; CHECK-V-NEXT:    mv s0, a1
+; CHECK-V-NEXT:    mv a0, a1
 ; CHECK-V-NEXT:  .LBB18_16: # %entry
 ; CHECK-V-NEXT:    vsetivli zero, 2, e64, m1, ta, ma
-; CHECK-V-NEXT:    vmv.s.x v8, s0
-; CHECK-V-NEXT:    vmv.s.x v9, a0
+; CHECK-V-NEXT:    vmv.s.x v8, a0
+; CHECK-V-NEXT:    vmv.s.x v9, s0
 ; CHECK-V-NEXT:    vslideup.vi v8, v9, 1
 ; CHECK-V-NEXT:    csrr a0, vlenb
 ; CHECK-V-NEXT:    slli a0, a0, 1
@@ -2369,19 +2370,19 @@ define <2 x i64> @utest_f64i64(<2 x double> %x) {
 ; CHECK-NOV-NEXT:    .cfi_offset s0, -16
 ; CHECK-NOV-NEXT:    .cfi_offset s1, -24
 ; CHECK-NOV-NEXT:    .cfi_offset fs0, -32
-; CHECK-NOV-NEXT:    fmv.d fs0, fa0
-; CHECK-NOV-NEXT:    fmv.d fa0, fa1
+; CHECK-NOV-NEXT:    fmv.d fs0, fa1
 ; CHECK-NOV-NEXT:    call __fixunsdfti
 ; CHECK-NOV-NEXT:    mv s0, a0
 ; CHECK-NOV-NEXT:    mv s1, a1
 ; CHECK-NOV-NEXT:    fmv.d fa0, fs0
 ; CHECK-NOV-NEXT:    call __fixunsdfti
-; CHECK-NOV-NEXT:    snez a2, s1
 ; CHECK-NOV-NEXT:    snez a1, a1
+; CHECK-NOV-NEXT:    snez a2, s1
+; CHECK-NOV-NEXT:    addi a2, a2, -1
+; CHECK-NOV-NEXT:    and a2, a2, s0
 ; CHECK-NOV-NEXT:    addi a1, a1, -1
-; CHECK-NOV-NEXT:    and a0, a1, a0
-; CHECK-NOV-NEXT:    addi a1, a2, -1
-; CHECK-NOV-NEXT:    and a1, a1, s0
+; CHECK-NOV-NEXT:    and a1, a1, a0
+; CHECK-NOV-NEXT:    mv a0, a2
 ; CHECK-NOV-NEXT:    ld ra, 24(sp) # 8-byte Folded Reload
 ; CHECK-NOV-NEXT:    ld s0, 16(sp) # 8-byte Folded Reload
 ; CHECK-NOV-NEXT:    ld s1, 8(sp) # 8-byte Folded Reload
@@ -2406,25 +2407,25 @@ define <2 x i64> @utest_f64i64(<2 x double> %x) {
 ; CHECK-V-NEXT:    addi a0, sp, 32
 ; CHECK-V-NEXT:    vs1r.v v8, (a0) # Unknown-size Folded Spill
 ; CHECK-V-NEXT:    vsetivli zero, 1, e64, m1, ta, ma
-; CHECK-V-NEXT:    vfmv.f.s fa0, v8
+; CHECK-V-NEXT:    vslidedown.vi v9, v8, 1
+; CHECK-V-NEXT:    vfmv.f.s fa0, v9
 ; CHECK-V-NEXT:    call __fixunsdfti
 ; CHECK-V-NEXT:    mv s0, a0
 ; CHECK-V-NEXT:    mv s1, a1
 ; CHECK-V-NEXT:    vsetivli zero, 1, e64, m1, ta, ma
 ; CHECK-V-NEXT:    addi a0, sp, 32
 ; CHECK-V-NEXT:    vl1r.v v8, (a0) # Unknown-size Folded Reload
-; CHECK-V-NEXT:    vslidedown.vi v8, v8, 1
 ; CHECK-V-NEXT:    vfmv.f.s fa0, v8
 ; CHECK-V-NEXT:    call __fixunsdfti
-; CHECK-V-NEXT:    snez a2, s1
 ; CHECK-V-NEXT:    snez a1, a1
-; CHECK-V-NEXT:    addi a1, a1, -1
-; CHECK-V-NEXT:    and a0, a1, a0
+; CHECK-V-NEXT:    snez a2, s1
 ; CHECK-V-NEXT:    addi a2, a2, -1
 ; CHECK-V-NEXT:    and a2, a2, s0
+; CHECK-V-NEXT:    addi a1, a1, -1
+; CHECK-V-NEXT:    and a0, a1, a0
 ; CHECK-V-NEXT:    vsetivli zero, 2, e64, m1, ta, ma
-; CHECK-V-NEXT:    vmv.s.x v8, a2
-; CHECK-V-NEXT:    vmv.s.x v9, a0
+; CHECK-V-NEXT:    vmv.s.x v8, a0
+; CHECK-V-NEXT:    vmv.s.x v9, a2
 ; CHECK-V-NEXT:    vslideup.vi v8, v9, 1
 ; CHECK-V-NEXT:    csrr a0, vlenb
 ; CHECK-V-NEXT:    slli a0, a0, 1
@@ -2466,32 +2467,32 @@ define <2 x i64> @ustest_f64i64(<2 x double> %x) {
 ; CHECK-NOV-NEXT:  # %bb.1: # %entry
 ; CHECK-NOV-NEXT:    li a2, 1
 ; CHECK-NOV-NEXT:  .LBB20_2: # %entry
-; CHECK-NOV-NEXT:    slti a4, s1, 1
 ; CHECK-NOV-NEXT:    slti a3, a1, 1
+; CHECK-NOV-NEXT:    slti a4, s1, 1
 ; CHECK-NOV-NEXT:    blez a1, .LBB20_4
 ; CHECK-NOV-NEXT:  # %bb.3: # %entry
 ; CHECK-NOV-NEXT:    li a1, 1
 ; CHECK-NOV-NEXT:  .LBB20_4: # %entry
+; CHECK-NOV-NEXT:    neg a4, a4
 ; CHECK-NOV-NEXT:    neg a3, a3
 ; CHECK-NOV-NEXT:    and a3, a3, a0
-; CHECK-NOV-NEXT:    neg a0, a4
 ; CHECK-NOV-NEXT:    beqz a1, .LBB20_7
 ; CHECK-NOV-NEXT:  # %bb.5: # %entry
 ; CHECK-NOV-NEXT:    sgtz a1, a1
-; CHECK-NOV-NEXT:    and a0, a0, s0
+; CHECK-NOV-NEXT:    and a4, a4, s0
 ; CHECK-NOV-NEXT:    bnez a2, .LBB20_8
 ; CHECK-NOV-NEXT:  .LBB20_6:
-; CHECK-NOV-NEXT:    snez a2, a0
+; CHECK-NOV-NEXT:    snez a0, a4
 ; CHECK-NOV-NEXT:    j .LBB20_9
 ; CHECK-NOV-NEXT:  .LBB20_7:
 ; CHECK-NOV-NEXT:    snez a1, a3
-; CHECK-NOV-NEXT:    and a0, a0, s0
+; CHECK-NOV-NEXT:    and a4, a4, s0
 ; CHECK-NOV-NEXT:    beqz a2, .LBB20_6
 ; CHECK-NOV-NEXT:  .LBB20_8: # %entry
-; CHECK-NOV-NEXT:    sgtz a2, a2
+; CHECK-NOV-NEXT:    sgtz a0, a2
 ; CHECK-NOV-NEXT:  .LBB20_9: # %entry
-; CHECK-NOV-NEXT:    neg a2, a2
-; CHECK-NOV-NEXT:    and a0, a2, a0
+; CHECK-NOV-NEXT:    neg a0, a0
+; CHECK-NOV-NEXT:    and a0, a0, a4
 ; CHECK-NOV-NEXT:    neg a1, a1
 ; CHECK-NOV-NEXT:    and a1, a1, a3
 ; CHECK-NOV-NEXT:    ld ra, 24(sp) # 8-byte Folded Reload
@@ -2533,15 +2534,15 @@ define <2 x i64> @ustest_f64i64(<2 x double> %x) {
 ; CHECK-V-NEXT:  # %bb.1: # %entry
 ; CHECK-V-NEXT:    li a2, 1
 ; CHECK-V-NEXT:  .LBB20_2: # %entry
-; CHECK-V-NEXT:    slti a3, s1, 1
 ; CHECK-V-NEXT:    slti a4, a1, 1
+; CHECK-V-NEXT:    slti a3, s1, 1
 ; CHECK-V-NEXT:    blez a1, .LBB20_4
 ; CHECK-V-NEXT:  # %bb.3: # %entry
 ; CHECK-V-NEXT:    li a1, 1
 ; CHECK-V-NEXT:  .LBB20_4: # %entry
+; CHECK-V-NEXT:    neg a3, a3
 ; CHECK-V-NEXT:    neg a4, a4
 ; CHECK-V-NEXT:    and a0, a4, a0
-; CHECK-V-NEXT:    neg a3, a3
 ; CHECK-V-NEXT:    beqz a1, .LBB20_7
 ; CHECK-V-NEXT:  # %bb.5: # %entry
 ; CHECK-V-NEXT:    sgtz a1, a1
@@ -2596,65 +2597,66 @@ define <2 x i64> @stest_f32i64(<2 x float> %x) {
 ; CHECK-NOV-NEXT:    .cfi_offset s0, -16
 ; CHECK-NOV-NEXT:    .cfi_offset s1, -24
 ; CHECK-NOV-NEXT:    .cfi_offset fs0, -32
-; CHECK-NOV-NEXT:    fmv.s fs0, fa0
-; CHECK-NOV-NEXT:    fmv.s fa0, fa1
+; CHECK-NOV-NEXT:    fmv.s fs0, fa1
 ; CHECK-NOV-NEXT:    call __fixsfti
 ; CHECK-NOV-NEXT:    mv s0, a0
 ; CHECK-NOV-NEXT:    mv s1, a1
 ; CHECK-NOV-NEXT:    fmv.s fa0, fs0
 ; CHECK-NOV-NEXT:    call __fixsfti
-; CHECK-NOV-NEXT:    li a2, -1
-; CHECK-NOV-NEXT:    srli a3, a2, 1
-; CHECK-NOV-NEXT:    beqz s1, .LBB21_3
+; CHECK-NOV-NEXT:    mv a2, a0
+; CHECK-NOV-NEXT:    li a0, -1
+; CHECK-NOV-NEXT:    srli a3, a0, 1
+; CHECK-NOV-NEXT:    beqz a1, .LBB21_3
 ; CHECK-NOV-NEXT:  # %bb.1: # %entry
-; CHECK-NOV-NEXT:    slti a4, s1, 0
-; CHECK-NOV-NEXT:    bnez a1, .LBB21_4
+; CHECK-NOV-NEXT:    slti a4, a1, 0
+; CHECK-NOV-NEXT:    bnez s1, .LBB21_4
 ; CHECK-NOV-NEXT:  .LBB21_2:
-; CHECK-NOV-NEXT:    sltu a5, a0, a3
+; CHECK-NOV-NEXT:    sltu a5, s0, a3
 ; CHECK-NOV-NEXT:    beqz a5, .LBB21_5
 ; CHECK-NOV-NEXT:    j .LBB21_6
 ; CHECK-NOV-NEXT:  .LBB21_3:
-; CHECK-NOV-NEXT:    sltu a4, s0, a3
-; CHECK-NOV-NEXT:    beqz a1, .LBB21_2
+; CHECK-NOV-NEXT:    sltu a4, a2, a3
+; CHECK-NOV-NEXT:    beqz s1, .LBB21_2
 ; CHECK-NOV-NEXT:  .LBB21_4: # %entry
-; CHECK-NOV-NEXT:    slti a5, a1, 0
+; CHECK-NOV-NEXT:    slti a5, s1, 0
 ; CHECK-NOV-NEXT:    bnez a5, .LBB21_6
 ; CHECK-NOV-NEXT:  .LBB21_5: # %entry
-; CHECK-NOV-NEXT:    mv a0, a3
+; CHECK-NOV-NEXT:    mv s0, a3
 ; CHECK-NOV-NEXT:  .LBB21_6: # %entry
 ; CHECK-NOV-NEXT:    neg a6, a5
 ; CHECK-NOV-NEXT:    neg a5, a4
-; CHECK-NOV-NEXT:    and a5, a5, s1
+; CHECK-NOV-NEXT:    and a5, a5, a1
 ; CHECK-NOV-NEXT:    bnez a4, .LBB21_8
 ; CHECK-NOV-NEXT:  # %bb.7: # %entry
-; CHECK-NOV-NEXT:    mv s0, a3
+; CHECK-NOV-NEXT:    mv a2, a3
 ; CHECK-NOV-NEXT:  .LBB21_8: # %entry
-; CHECK-NOV-NEXT:    and a4, a6, a1
-; CHECK-NOV-NEXT:    slli a1, a2, 63
-; CHECK-NOV-NEXT:    beq a5, a2, .LBB21_11
+; CHECK-NOV-NEXT:    and a4, a6, s1
+; CHECK-NOV-NEXT:    slli a1, a0, 63
+; CHECK-NOV-NEXT:    beq a5, a0, .LBB21_11
 ; CHECK-NOV-NEXT:  # %bb.9: # %entry
 ; CHECK-NOV-NEXT:    slti a3, a5, 0
 ; CHECK-NOV-NEXT:    xori a3, a3, 1
-; CHECK-NOV-NEXT:    bne a4, a2, .LBB21_12
+; CHECK-NOV-NEXT:    bne a4, a0, .LBB21_12
 ; CHECK-NOV-NEXT:  .LBB21_10:
-; CHECK-NOV-NEXT:    sltu a2, a1, a0
-; CHECK-NOV-NEXT:    beqz a2, .LBB21_13
+; CHECK-NOV-NEXT:    sltu a0, a1, s0
+; CHECK-NOV-NEXT:    beqz a0, .LBB21_13
 ; CHECK-NOV-NEXT:    j .LBB21_14
 ; CHECK-NOV-NEXT:  .LBB21_11:
-; CHECK-NOV-NEXT:    sltu a3, a1, s0
-; CHECK-NOV-NEXT:    beq a4, a2, .LBB21_10
+; CHECK-NOV-NEXT:    sltu a3, a1, a2
+; CHECK-NOV-NEXT:    beq a4, a0, .LBB21_10
 ; CHECK-NOV-NEXT:  .LBB21_12: # %entry
-; CHECK-NOV-NEXT:    slti a2, a4, 0
-; CHECK-NOV-NEXT:    xori a2, a2, 1
-; CHECK-NOV-NEXT:    bnez a2, .LBB21_14
+; CHECK-NOV-NEXT:    slti a0, a4, 0
+; CHECK-NOV-NEXT:    xori a0, a0, 1
+; CHECK-NOV-NEXT:    bnez a0, .LBB21_14
 ; CHECK-NOV-NEXT:  .LBB21_13: # %entry
-; CHECK-NOV-NEXT:    mv a0, a1
+; CHECK-NOV-NEXT:    mv s0, a1
 ; CHECK-NOV-NEXT:  .LBB21_14: # %entry
 ; CHECK-NOV-NEXT:    bnez a3, .LBB21_16
 ; CHECK-NOV-NEXT:  # %bb.15: # %entry
-; CHECK-NOV-NEXT:    mv s0, a1
+; CHECK-NOV-NEXT:    mv a2, a1
 ; CHECK-NOV-NEXT:  .LBB21_16: # %entry
-; CHECK-NOV-NEXT:    mv a1, s0
+; CHECK-NOV-NEXT:    mv a0, s0
+; CHECK-NOV-NEXT:    mv a1, a2
 ; CHECK-NOV-NEXT:    ld ra, 24(sp) # 8-byte Folded Reload
 ; CHECK-NOV-NEXT:    ld s0, 16(sp) # 8-byte Folded Reload
 ; CHECK-NOV-NEXT:    ld s1, 8(sp) # 8-byte Folded Reload
@@ -2679,43 +2681,43 @@ define <2 x i64> @stest_f32i64(<2 x float> %x) {
 ; CHECK-V-NEXT:    addi a0, sp, 32
 ; CHECK-V-NEXT:    vs1r.v v8, (a0) # Unknown-size Folded Spill
 ; CHECK-V-NEXT:    vsetivli zero, 1, e32, mf2, ta, ma
-; CHECK-V-NEXT:    vfmv.f.s fa0, v8
+; CHECK-V-NEXT:    vslidedown.vi v9, v8, 1
+; CHECK-V-NEXT:    vfmv.f.s fa0, v9
 ; CHECK-V-NEXT:    call __fixsfti
 ; CHECK-V-NEXT:    mv s0, a0
 ; CHECK-V-NEXT:    mv s1, a1
 ; CHECK-V-NEXT:    vsetivli zero, 1, e32, mf2, ta, ma
 ; CHECK-V-NEXT:    addi a0, sp, 32
 ; CHECK-V-NEXT:    vl1r.v v8, (a0) # Unknown-size Folded Reload
-; CHECK-V-NEXT:    vslidedown.vi v8, v8, 1
 ; CHECK-V-NEXT:    vfmv.f.s fa0, v8
 ; CHECK-V-NEXT:    call __fixsfti
 ; CHECK-V-NEXT:    li a2, -1
 ; CHECK-V-NEXT:    srli a3, a2, 1
-; CHECK-V-NEXT:    beqz s1, .LBB21_3
+; CHECK-V-NEXT:    beqz a1, .LBB21_3
 ; CHECK-V-NEXT:  # %bb.1: # %entry
-; CHECK-V-NEXT:    slti a4, s1, 0
-; CHECK-V-NEXT:    bnez a1, .LBB21_4
+; CHECK-V-NEXT:    slti a4, a1, 0
+; CHECK-V-NEXT:    bnez s1, .LBB21_4
 ; CHECK-V-NEXT:  .LBB21_2:
-; CHECK-V-NEXT:    sltu a5, a0, a3
+; CHECK-V-NEXT:    sltu a5, s0, a3
 ; CHECK-V-NEXT:    beqz a5, .LBB21_5
 ; CHECK-V-NEXT:    j .LBB21_6
 ; CHECK-V-NEXT:  .LBB21_3:
-; CHECK-V-NEXT:    sltu a4, s0, a3
-; CHECK-V-NEXT:    beqz a1, .LBB21_2
+; CHECK-V-NEXT:    sltu a4, a0, a3
+; CHECK-V-NEXT:    beqz s1, .LBB21_2
 ; CHECK-V-NEXT:  .LBB21_4: # %entry
-; CHECK-V-NEXT:    slti a5, a1, 0
+; CHECK-V-NEXT:    slti a5, s1, 0
 ; CHECK-V-NEXT:    bnez a5, .LBB21_6
 ; CHECK-V-NEXT:  .LBB21_5: # %entry
-; CHECK-V-NEXT:    mv a0, a3
+; CHECK-V-NEXT:    mv s0, a3
 ; CHECK-V-NEXT:  .LBB21_6: # %entry
 ; CHECK-V-NEXT:    neg a6, a5
 ; CHECK-V-NEXT:    neg a5, a4
-; CHECK-V-NEXT:    and a5, a5, s1
+; CHECK-V-NEXT:    and a5, a5, a1
 ; CHECK-V-NEXT:    bnez a4, .LBB21_8
 ; CHECK-V-NEXT:  # %bb.7: # %entry
-; CHECK-V-NEXT:    mv s0, a3
+; CHECK-V-NEXT:    mv a0, a3
 ; CHECK-V-NEXT:  .LBB21_8: # %entry
-; CHECK-V-NEXT:    and a4, a6, a1
+; CHECK-V-NEXT:    and a4, a6, s1
 ; CHECK-V-NEXT:    slli a1, a2, 63
 ; CHECK-V-NEXT:    beq a5, a2, .LBB21_11
 ; CHECK-V-NEXT:  # %bb.9: # %entry
@@ -2723,26 +2725,26 @@ define <2 x i64> @stest_f32i64(<2 x float> %x) {
 ; CHECK-V-NEXT:    xori a3, a3, 1
 ; CHECK-V-NEXT:    bne a4, a2, .LBB21_12
 ; CHECK-V-NEXT:  .LBB21_10:
-; CHECK-V-NEXT:    sltu a2, a1, a0
+; CHECK-V-NEXT:    sltu a2, a1, s0
 ; CHECK-V-NEXT:    beqz a2, .LBB21_13
 ; CHECK-V-NEXT:    j .LBB21_14
 ; CHECK-V-NEXT:  .LBB21_11:
-; CHECK-V-NEXT:    sltu a3, a1, s0
+; CHECK-V-NEXT:    sltu a3, a1, a0
 ; CHECK-V-NEXT:    beq a4, a2, .LBB21_10
 ; CHECK-V-NEXT:  .LBB21_12: # %entry
 ; CHECK-V-NEXT:    slti a2, a4, 0
 ; CHECK-V-NEXT:    xori a2, a2, 1
 ; CHECK-V-NEXT:    bnez a2, .LBB21_14
 ; CHECK-V-NEXT:  .LBB21_13: # %entry
-; CHECK-V-NEXT:    mv a0, a1
+; CHECK-V-NEXT:    mv s0, a1
 ; CHECK-V-NEXT:  .LBB21_14: # %entry
 ; CHECK-V-NEXT:    bnez a3, .LBB21_16
 ; CHECK-V-NEXT:  # %bb.15: # %entry
-; CHECK-V-NEXT:    mv s0, a1
+; CHECK-V-NEXT:    mv a0, a1
 ; CHECK-V-NEXT:  .LBB21_16: # %entry
 ; CHECK-V-NEXT:    vsetivli zero, 2, e64, m1, ta, ma
-; CHECK-V-NEXT:    vmv.s.x v8, s0
-; CHECK-V-NEXT:    vmv.s.x v9, a0
+; CHECK-V-NEXT:    vmv.s.x v8, a0
+; CHECK-V-NEXT:    vmv.s.x v9, s0
 ; CHECK-V-NEXT:    vslideup.vi v8, v9, 1
 ; CHECK-V-NEXT:    csrr a0, vlenb
 ; CHECK-V-NEXT:    slli a0, a0, 1
@@ -2775,19 +2777,19 @@ define <2 x i64> @utest_f32i64(<2 x float> %x) {
 ; CHECK-NOV-NEXT:    .cfi_offset s0, -16
 ; CHECK-NOV-NEXT:    .cfi_offset s1, -24
 ; CHECK-NOV-NEXT:    .cfi_offset fs0, -32
-; CHECK-NOV-NEXT:    fmv.s fs0, fa0
-; CHECK-NOV-NEXT:    fmv.s fa0, fa1
+; CHECK-NOV-NEXT:    fmv.s fs0, fa1
 ; CHECK-NOV-NEXT:    call __fixunssfti
 ; CHECK-NOV-NEXT:    mv s0, a0
 ; CHECK-NOV-NEXT:    mv s1, a1
 ; CHECK-NOV-NEXT:    fmv.s fa0, fs0
 ; CHECK-NOV-NEXT:    call __fixunssfti
-; CHECK-NOV-NEXT:    snez a2, s1
 ; CHECK-NOV-NEXT:    snez a1, a1
+; CHECK-NOV-NEXT:    snez a2, s1
+; CHECK-NOV-NEXT:    addi a2, a2, -1
+; CHECK-NOV-NEXT:    and a2, a2, s0
 ; CHECK-NOV-NEXT:    addi a1, a1, -1
-; CHECK-NOV-NEXT:    and a0, a1, a0
-; CHECK-NOV-NEXT:    addi a1, a2, -1
-; CHECK-NOV-NEXT:    and a1, a1, s0
+; CHECK-NOV-NEXT:    and a1, a1, a0
+; CHECK-NOV-NEXT:    mv a0, a2
 ; CHECK-NOV-NEXT:    ld ra, 24(sp) # 8-byte Folded Reload
 ; CHECK-NOV-NEXT:    ld s0, 16(sp) # 8-byte Folded Reload
 ; CHECK-NOV-NEXT:    ld s1, 8(sp) # 8-byte Folded Reload
@@ -2812,25 +2814,25 @@ define <2 x i64> @utest_f32i64(<2 x float> %x) {
 ; CHECK-V-NEXT:    addi a0, sp, 32
 ; CHECK-V-NEXT:    vs1r.v v8, (a0) # Unknown-size Folded Spill
 ; CHECK-V-NEXT:    vsetivli zero, 1, e32, mf2, ta, ma
-; CHECK-V-NEXT:    vfmv.f.s fa0, v8
+; CHECK-V-NEXT:    vslidedown.vi v9, v8, 1
+; CHECK-V-NEXT:    vfmv.f.s fa0, v9
 ; CHECK-V-NEXT:    call __fixunssfti
 ; CHECK-V-NEXT:    mv s0, a0
 ; CHECK-V-NEXT:    mv s1, a1
 ; CHECK-V-NEXT:    vsetivli zero, 1, e32, mf2, ta, ma
 ; CHECK-V-NEXT:    addi a0, sp, 32
 ; CHECK-V-NEXT:    vl1r.v v8, (a0) # Unknown-size Folded Reload
-; CHECK-V-NEXT:    vslidedown.vi v8, v8, 1
 ; CHECK-V-NEXT:    vfmv.f.s fa0, v8
 ; CHECK-V-NEXT:    call __fixunssfti
-; CHECK-V-NEXT:    snez a2, s1
 ; CHECK-V-NEXT:    snez a1, a1
-; CHECK-V-NEXT:    addi a1, a1, -1
-; CHECK-V-NEXT:    and a0, a1, a0
+; CHECK-V-NEXT:    snez a2, s1
 ; CHECK-V-NEXT:    addi a2, a2, -1
 ; CHECK-V-NEXT:    and a2, a2, s0
+; CHECK-V-NEXT:    addi a1, a1, -1
+; CHECK-V-NEXT:    and a0, a1, a0
 ; CHECK-V-NEXT:    vsetivli zero, 2, e64, m1, ta, ma
-; CHECK-V-NEXT:    vmv.s.x v8, a2
-; CHECK-V-NEXT:    vmv.s.x v9, a0
+; CHECK-V-NEXT:    vmv.s.x v8, a0
+; CHECK-V-NEXT:    vmv.s.x v9, a2
 ; CHECK-V-NEXT:    vslideup.vi v8, v9, 1
 ; CHECK-V-NEXT:    csrr a0, vlenb
 ; CHECK-V-NEXT:    slli a0, a0, 1
@@ -2872,32 +2874,32 @@ define <2 x i64> @ustest_f32i64(<2 x float> %x) {
 ; CHECK-NOV-NEXT:  # %bb.1: # %entry
 ; CHECK-NOV-NEXT:    li a2, 1
 ; CHECK-NOV-NEXT:  .LBB23_2: # %entry
-; CHECK-NOV-NEXT:    slti a4, s1, 1
 ; CHECK-NOV-NEXT:    slti a3, a1, 1
+; CHECK-NOV-NEXT:    slti a4, s1, 1
 ; CHECK-NOV-NEXT:    blez a1, .LBB23_4
 ; CHECK-NOV-NEXT:  # %bb.3: # %entry
 ; CHECK-NOV-NEXT:    li a1, 1
 ; CHECK-NOV-NEXT:  .LBB23_4: # %entry
+; CHECK-NOV-NEXT:    neg a4, a4
 ; CHECK-NOV-NEXT:    neg a3, a3
 ; CHECK-NOV-NEXT:    and a3, a3, a0
-; CHECK-NOV-NEXT:    neg a0, a4
 ; CHECK-NOV-NEXT:    beqz a1, .LBB23_7
 ; CHECK-NOV-NEXT:  # %bb.5: # %entry
 ; CHECK-NOV-NEXT:    sgtz a1, a1
-; CHECK-NOV-NEXT:    and a0, a0, s0
+; CHECK-NOV-NEXT:    and a4, a4, s0
 ; CHECK-NOV-NEXT:    bnez a2, .LBB23_8
 ; CHECK-NOV-NEXT:  .LBB23_6:
-; CHECK-NOV-NEXT:    snez a2, a0
+; CHECK-NOV-NEXT:    snez a0, a4
 ; CHECK-NOV-NEXT:    j .LBB23_9
 ; CHECK-NOV-NEXT:  .LBB23_7:
 ; CHECK-NOV-NEXT:    snez a1, a3
-; CHECK-NOV-NEXT:    and a0, a0, s0
+; CHECK-NOV-NEXT:    and a4, a4, s0
 ; CHECK-NOV-NEXT:    beqz a2, .LBB23_6
 ; CHECK-NOV-NEXT:  .LBB23_8: # %entry
-; CHECK-NOV-NEXT:    sgtz a2, a2
+; CHECK-NOV-NEXT:    sgtz a0, a2
 ; CHECK-NOV-NEXT:  .LBB23_9: # %entry
-; CHECK-NOV-NEXT:    neg a2, a2
-; CHECK-NOV-NEXT:    and a0, a2, a0
+; CHECK-NOV-NEXT:    neg a0, a0
+; CHECK-NOV-NEXT:    and a0, a0, a4
 ; CHECK-NOV-NEXT:    neg a1, a1
 ; CHECK-NOV-NEXT:    and a1, a1, a3
 ; CHECK-NOV-NEXT:    ld ra, 24(sp) # 8-byte Folded Reload
@@ -2939,15 +2941,15 @@ define <2 x i64> @ustest_f32i64(<2 x float> %x) {
 ; CHECK-V-NEXT:  # %bb.1: # %entry
 ; CHECK-V-NEXT:    li a2, 1
 ; CHECK-V-NEXT:  .LBB23_2: # %entry
-; CHECK-V-NEXT:    slti a3, s1, 1
 ; CHECK-V-NEXT:    slti a4, a1, 1
+; CHECK-V-NEXT:    slti a3, s1, 1
 ; CHECK-V-NEXT:    blez a1, .LBB23_4
 ; CHECK-V-NEXT:  # %bb.3: # %entry
 ; CHECK-V-NEXT:    li a1, 1
 ; CHECK-V-NEXT:  .LBB23_4: # %entry
+; CHECK-V-NEXT:    neg a3, a3
 ; CHECK-V-NEXT:    neg a4, a4
 ; CHECK-V-NEXT:    and a0, a4, a0
-; CHECK-V-NEXT:    neg a3, a3
 ; CHECK-V-NEXT:    beqz a1, .LBB23_7
 ; CHECK-V-NEXT:  # %bb.5: # %entry
 ; CHECK-V-NEXT:    sgtz a1, a1
@@ -3002,8 +3004,8 @@ define <2 x i64> @stest_f16i64(<2 x half> %x) {
 ; CHECK-NOV-NEXT:    .cfi_offset s0, -16
 ; CHECK-NOV-NEXT:    .cfi_offset s1, -24
 ; CHECK-NOV-NEXT:    .cfi_offset s2, -32
-; CHECK-NOV-NEXT:    mv s2, a0
-; CHECK-NOV-NEXT:    fmv.w.x fa0, a1
+; CHECK-NOV-NEXT:    mv s2, a1
+; CHECK-NOV-NEXT:    fmv.w.x fa0, a0
 ; CHECK-NOV-NEXT:    call __extendhfsf2
 ; CHECK-NOV-NEXT:    call __fixsfti
 ; CHECK-NOV-NEXT:    mv s0, a0
@@ -3011,58 +3013,60 @@ define <2 x i64> @stest_f16i64(<2 x half> %x) {
 ; CHECK-NOV-NEXT:    fmv.w.x fa0, s2
 ; CHECK-NOV-NEXT:    call __extendhfsf2
 ; CHECK-NOV-NEXT:    call __fixsfti
-; CHECK-NOV-NEXT:    li a2, -1
-; CHECK-NOV-NEXT:    srli a3, a2, 1
-; CHECK-NOV-NEXT:    beqz s1, .LBB24_3
+; CHECK-NOV-NEXT:    mv a2, a0
+; CHECK-NOV-NEXT:    li a0, -1
+; CHECK-NOV-NEXT:    srli a3, a0, 1
+; CHECK-NOV-NEXT:    beqz a1, .LBB24_3
 ; CHECK-NOV-NEXT:  # %bb.1: # %entry
-; CHECK-NOV-NEXT:    slti a4, s1, 0
-; CHECK-NOV-NEXT:    bnez a1, .LBB24_4
+; CHECK-NOV-NEXT:    slti a4, a1, 0
+; CHECK-NOV-NEXT:    bnez s1, .LBB24_4
 ; CHECK-NOV-NEXT:  .LBB24_2:
-; CHECK-NOV-NEXT:    sltu a5, a0, a3
+; CHECK-NOV-NEXT:    sltu a5, s0, a3
 ; CHECK-NOV-NEXT:    beqz a5, .LBB24_5
 ; CHECK-NOV-NEXT:    j .LBB24_6
 ; CHECK-NOV-NEXT:  .LBB24_3:
-; CHECK-NOV-NEXT:    sltu a4, s0, a3
-; CHECK-NOV-NEXT:    beqz a1, .LBB24_2
+; CHECK-NOV-NEXT:    sltu a4, a2, a3
+; CHECK-NOV-NEXT:    beqz s1, .LBB24_2
 ; CHECK-NOV-NEXT:  .LBB24_4: # %entry
-; CHECK-NOV-NEXT:    slti a5, a1, 0
+; CHECK-NOV-NEXT:    slti a5, s1, 0
 ; CHECK-NOV-NEXT:    bnez a5, .LBB24_6
 ; CHECK-NOV-NEXT:  .LBB24_5: # %entry
-; CHECK-NOV-NEXT:    mv a0, a3
+; CHECK-NOV-NEXT:    mv s0, a3
 ; CHECK-NOV-NEXT:  .LBB24_6: # %entry
 ; CHECK-NOV-NEXT:    neg a6, a5
 ; CHECK-NOV-NEXT:    neg a5, a4
-; CHECK-NOV-NEXT:    and a5, a5, s1
+; CHECK-NOV-NEXT:    and a5, a5, a1
 ; CHECK-NOV-NEXT:    bnez a4, .LBB24_8
 ; CHECK-NOV-NEXT:  # %bb.7: # %entry
-; CHECK-NOV-NEXT:    mv s0, a3
+; CHECK-NOV-NEXT:    mv a2, a3
 ; CHECK-NOV-NEXT:  .LBB24_8: # %entry
-; CHECK-NOV-NEXT:    and a4, a6, a1
-; CHECK-NOV-NEXT:    slli a1, a2, 63
-; CHECK-NOV-NEXT:    beq a5, a2, .LBB24_11
+; CHECK-NOV-NEXT:    and a4, a6, s1
+; CHECK-NOV-NEXT:    slli a1, a0, 63
+; CHECK-NOV-NEXT:    beq a5, a0, .LBB24_11
 ; CHECK-NOV-NEXT:  # %bb.9: # %entry
 ; CHECK-NOV-NEXT:    slti a3, a5, 0
 ; CHECK-NOV-NEXT:    xori a3, a3, 1
-; CHECK-NOV-NEXT:    bne a4, a2, .LBB24_12
+; CHECK-NOV-NEXT:    bne a4, a0, .LBB24_12
 ; CHECK-NOV-NEXT:  .LBB24_10:
-; CHECK-NOV-NEXT:    sltu a2, a1, a0
-; CHECK-NOV-NEXT:    beqz a2, .LBB24_13
+; CHECK-NOV-NEXT:    sltu a0, a1, s0
+; CHECK-NOV-NEXT:    beqz a0, .LBB24_13
 ; CHECK-NOV-NEXT:    j .LBB24_14
 ; CHECK-NOV-NEXT:  .LBB24_11:
-; CHECK-NOV-NEXT:    sltu a3, a1, s0
-; CHECK-NOV-NEXT:    beq a4, a2, .LBB24_10
+; CHECK-NOV-NEXT:    sltu a3, a1, a2
+; CHECK-NOV-NEXT:    beq a4, a0, .LBB24_10
 ; CHECK-NOV-NEXT:  .LBB24_12: # %entry
-; CHECK-NOV-NEXT:    slti a2, a4, 0
-; CHECK-NOV-NEXT:    xori a2, a2, 1
-; CHECK-NOV-NEXT:    bnez a2, .LBB24_14
+; CHECK-NOV-NEXT:    slti a0, a4, 0
+; CHECK-NOV-NEXT:    xori a0, a0, 1
+; CHECK-NOV-NEXT:    bnez a0, .LBB24_14
 ; CHECK-NOV-NEXT:  .LBB24_13: # %entry
-; CHECK-NOV-NEXT:    mv a0, a1
+; CHECK-NOV-NEXT:    mv s0, a1
 ; CHECK-NOV-NEXT:  .LBB24_14: # %entry
 ; CHECK-NOV-NEXT:    bnez a3, .LBB24_16
 ; CHECK-NOV-NEXT:  # %bb.15: # %entry
-; CHECK-NOV-NEXT:    mv s0, a1
+; CHECK-NOV-NEXT:    mv a2, a1
 ; CHECK-NOV-NEXT:  .LBB24_16: # %entry
-; CHECK-NOV-NEXT:    mv a1, s0
+; CHECK-NOV-NEXT:    mv a0, s0
+; CHECK-NOV-NEXT:    mv a1, a2
 ; CHECK-NOV-NEXT:    ld ra, 24(sp) # 8-byte Folded Reload
 ; CHECK-NOV-NEXT:    ld s0, 16(sp) # 8-byte Folded Reload
 ; CHECK-NOV-NEXT:    ld s1, 8(sp) # 8-byte Folded Reload
@@ -3082,8 +3086,8 @@ define <2 x i64> @stest_f16i64(<2 x half> %x) {
 ; CHECK-V-NEXT:    .cfi_offset s0, -16
 ; CHECK-V-NEXT:    .cfi_offset s1, -24
 ; CHECK-V-NEXT:    .cfi_offset s2, -32
-; CHECK-V-NEXT:    mv s2, a0
-; CHECK-V-NEXT:    fmv.w.x fa0, a1
+; CHECK-V-NEXT:    mv s2, a1
+; CHECK-V-NEXT:    fmv.w.x fa0, a0
 ; CHECK-V-NEXT:    call __extendhfsf2
 ; CHECK-V-NEXT:    call __fixsfti
 ; CHECK-V-NEXT:    mv s0, a0
@@ -3093,31 +3097,31 @@ define <2 x i64> @stest_f16i64(<2 x half> %x) {
 ; CHECK-V-NEXT:    call __fixsfti
 ; CHECK-V-NEXT:    li a2, -1
 ; CHECK-V-NEXT:    srli a3, a2, 1
-; CHECK-V-NEXT:    beqz s1, .LBB24_3
+; CHECK-V-NEXT:    beqz a1, .LBB24_3
 ; CHECK-V-NEXT:  # %bb.1: # %entry
-; CHECK-V-NEXT:    slti a4, s1, 0
-; CHECK-V-NEXT:    bnez a1, .LBB24_4
+; CHECK-V-NEXT:    slti a4, a1, 0
+; CHECK-V-NEXT:    bnez s1, .LBB24_4
 ; CHECK-V-NEXT:  .LBB24_2:
-; CHECK-V-NEXT:    sltu a5, a0, a3
+; CHECK-V-NEXT:    sltu a5, s0, a3
 ; CHECK-V-NEXT:    beqz a5, .LBB24_5
 ; CHECK-V-NEXT:    j .LBB24_6
 ; CHECK-V-NEXT:  .LBB24_3:
-; CHECK-V-NEXT:    sltu a4, s0, a3
-; CHECK-V-NEXT:    beqz a1, .LBB24_2
+; CHECK-V-NEXT:    sltu a4, a0, a3
+; CHECK-V-NEXT:    beqz s1, .LBB24_2
 ; CHECK-V-NEXT:  .LBB24_4: # %entry
-; CHECK-V-NEXT:    slti a5, a1, 0
+; CHECK-V-NEXT:    slti a5, s1, 0
 ; CHECK-V-NEXT:    bnez a5, .LBB24_6
 ; CHECK-V-NEXT:  .LBB24_5: # %entry
-; CHECK-V-NEXT:    mv a0, a3
+; CHECK-V-NEXT:    mv s0, a3
 ; CHECK-V-NEXT:  .LBB24_6: # %entry
 ; CHECK-V-NEXT:    neg a6, a5
 ; CHECK-V-NEXT:    neg a5, a4
-; CHECK-V-NEXT:    and a5, a5, s1
+; CHECK-V-NEXT:    and a5, a5, a1
 ; CHECK-V-NEXT:    bnez a4, .LBB24_8
 ; CHECK-V-NEXT:  # %bb.7: # %entry
-; CHECK-V-NEXT:    mv s0, a3
+; CHECK-V-NEXT:    mv a0, a3
 ; CHECK-V-NEXT:  .LBB24_8: # %entry
-; CHECK-V-NEXT:    and a4, a6, a1
+; CHECK-V-NEXT:    and a4, a6, s1
 ; CHECK-V-NEXT:    slli a1, a2, 63
 ; CHECK-V-NEXT:    beq a5, a2, .LBB24_11
 ; CHECK-V-NEXT:  # %bb.9: # %entry
@@ -3125,26 +3129,26 @@ define <2 x i64> @stest_f16i64(<2 x half> %x) {
 ; CHECK-V-NEXT:    xori a3, a3, 1
 ; CHECK-V-NEXT:    bne a4, a2, .LBB24_12
 ; CHECK-V-NEXT:  .LBB24_10:
-; CHECK-V-NEXT:    sltu a2, a1, a0
+; CHECK-V-NEXT:    sltu a2, a1, s0
 ; CHECK-V-NEXT:    beqz a2, .LBB24_13
 ; CHECK-V-NEXT:    j .LBB24_14
 ; CHECK-V-NEXT:  .LBB24_11:
-; CHECK-V-NEXT:    sltu a3, a1, s0
+; CHECK-V-NEXT:    sltu a3, a1, a0
 ; CHECK-V-NEXT:    beq a4, a2, .LBB24_10
 ; CHECK-V-NEXT:  .LBB24_12: # %entry
 ; CHECK-V-NEXT:    slti a2, a4, 0
 ; CHECK-V-NEXT:    xori a2, a2, 1
 ; CHECK-V-NEXT:    bnez a2, .LBB24_14
 ; CHECK-V-NEXT:  .LBB24_13: # %entry
-; CHECK-V-NEXT:    mv a0, a1
+; CHECK-V-NEXT:    mv s0, a1
 ; CHECK-V-NEXT:  .LBB24_14: # %entry
 ; CHECK-V-NEXT:    bnez a3, .LBB24_16
 ; CHECK-V-NEXT:  # %bb.15: # %entry
-; CHECK-V-NEXT:    mv s0, a1
+; CHECK-V-NEXT:    mv a0, a1
 ; CHECK-V-NEXT:  .LBB24_16: # %entry
 ; CHECK-V-NEXT:    vsetivli zero, 2, e64, m1, ta, ma
-; CHECK-V-NEXT:    vmv.s.x v9, s0
-; CHECK-V-NEXT:    vmv.s.x v8, a0
+; CHECK-V-NEXT:    vmv.s.x v9, a0
+; CHECK-V-NEXT:    vmv.s.x v8, s0
 ; CHECK-V-NEXT:    vslideup.vi v8, v9, 1
 ; CHECK-V-NEXT:    ld ra, 24(sp) # 8-byte Folded Reload
 ; CHECK-V-NEXT:    ld s0, 16(sp) # 8-byte Folded Reload
@@ -3175,8 +3179,8 @@ define <2 x i64> @utesth_f16i64(<2 x half> %x) {
 ; CHECK-NOV-NEXT:    .cfi_offset s0, -16
 ; CHECK-NOV-NEXT:    .cfi_offset s1, -24
 ; CHECK-NOV-NEXT:    .cfi_offset s2, -32
-; CHECK-NOV-NEXT:    mv s0, a0
-; CHECK-NOV-NEXT:    fmv.w.x fa0, a1
+; CHECK-NOV-NEXT:    mv s0, a1
+; CHECK-NOV-NEXT:    fmv.w.x fa0, a0
 ; CHECK-NOV-NEXT:    call __extendhfsf2
 ; CHECK-NOV-NEXT:    call __fixunssfti
 ; CHECK-NOV-NEXT:    mv s1, a0
@@ -3184,12 +3188,13 @@ define <2 x i64> @utesth_f16i64(<2 x half> %x) {
 ; CHECK-NOV-NEXT:    fmv.w.x fa0, s0
 ; CHECK-NOV-NEXT:    call __extendhfsf2
 ; CHECK-NOV-NEXT:    call __fixunssfti
-; CHECK-NOV-NEXT:    snez a2, s2
 ; CHECK-NOV-NEXT:    snez a1, a1
+; CHECK-NOV-NEXT:    snez a2, s2
+; CHECK-NOV-NEXT:    addi a2, a2, -1
+; CHECK-NOV-NEXT:    and a2, a2, s1
 ; CHECK-NOV-NEXT:    addi a1, a1, -1
-; CHECK-NOV-NEXT:    and a0, a1, a0
-; CHECK-NOV-NEXT:    addi a1, a2, -1
-; CHECK-NOV-NEXT:    and a1, a1, s1
+; CHECK-NOV-NEXT:    and a1, a1, a0
+; CHECK-NOV-NEXT:    mv a0, a2
 ; CHECK-NOV-NEXT:    ld ra, 24(sp) # 8-byte Folded Reload
 ; CHECK-NOV-NEXT:    ld s0, 16(sp) # 8-byte Folded Reload
 ; CHECK-NOV-NEXT:    ld s1, 8(sp) # 8-byte Folded Reload
@@ -3209,8 +3214,8 @@ define <2 x i64> @utesth_f16i64(<2 x half> %x) {
 ; CHECK-V-NEXT:    .cfi_offset s0, -16
 ; CHECK-V-NEXT:    .cfi_offset s1, -24
 ; CHECK-V-NEXT:    .cfi_offset s2, -32
-; CHECK-V-NEXT:    mv s0, a0
-; CHECK-V-NEXT:    fmv.w.x fa0, a1
+; CHECK-V-NEXT:    mv s0, a1
+; CHECK-V-NEXT:    fmv.w.x fa0, a0
 ; CHECK-V-NEXT:    call __extendhfsf2
 ; CHECK-V-NEXT:    call __fixunssfti
 ; CHECK-V-NEXT:    mv s1, a0
@@ -3218,15 +3223,15 @@ define <2 x i64> @utesth_f16i64(<2 x half> %x) {
 ; CHECK-V-NEXT:    fmv.w.x fa0, s0
 ; CHECK-V-NEXT:    call __extendhfsf2
 ; CHECK-V-NEXT:    call __fixunssfti
-; CHECK-V-NEXT:    snez a2, s2
 ; CHECK-V-NEXT:    snez a1, a1
-; CHECK-V-NEXT:    addi a1, a1, -1
-; CHECK-V-NEXT:    and a0, a1, a0
+; CHECK-V-NEXT:    snez a2, s2
 ; CHECK-V-NEXT:    addi a2, a2, -1
 ; CHECK-V-NEXT:    and a2, a2, s1
+; CHECK-V-NEXT:    addi a1, a1, -1
+; CHECK-V-NEXT:    and a0, a1, a0
 ; CHECK-V-NEXT:    vsetivli zero, 2, e64, m1, ta, ma
-; CHECK-V-NEXT:    vmv.s.x v9, a2
-; CHECK-V-NEXT:    vmv.s.x v8, a0
+; CHECK-V-NEXT:    vmv.s.x v9, a0
+; CHECK-V-NEXT:    vmv.s.x v8, a2
 ; CHECK-V-NEXT:    vslideup.vi v8, v9, 1
 ; CHECK-V-NEXT:    ld ra, 24(sp) # 8-byte Folded Reload
 ; CHECK-V-NEXT:    ld s0, 16(sp) # 8-byte Folded Reload
@@ -3269,32 +3274,32 @@ define <2 x i64> @ustest_f16i64(<2 x half> %x) {
 ; CHECK-NOV-NEXT:  # %bb.1: # %entry
 ; CHECK-NOV-NEXT:    li a2, 1
 ; CHECK-NOV-NEXT:  .LBB26_2: # %entry
-; CHECK-NOV-NEXT:    slti a4, s1, 1
 ; CHECK-NOV-NEXT:    slti a3, a1, 1
+; CHECK-NOV-NEXT:    slti a4, s1, 1
 ; CHECK-NOV-NEXT:    blez a1, .LBB26_4
 ; CHECK-NOV-NEXT:  # %bb.3: # %entry
 ; CHECK-NOV-NEXT:    li a1, 1
 ; CHECK-NOV-NEXT:  .LBB26_4: # %entry
+; CHECK-NOV-NEXT:    neg a4, a4
 ; CHECK-NOV-NEXT:    neg a3, a3
 ; CHECK-NOV-NEXT:    and a3, a3, a0
-; CHECK-NOV-NEXT:    neg a0, a4
 ; CHECK-NOV-NEXT:    beqz a1, .LBB26_7
 ; CHECK-NOV-NEXT:  # %bb.5: # %entry
 ; CHECK-NOV-NEXT:    sgtz a1, a1
-; CHECK-NOV-NEXT:    and a0, a0, s0
+; CHECK-NOV-NEXT:    and a4, a4, s0
 ; CHECK-NOV-NEXT:    bnez a2, .LBB26_8
 ; CHECK-NOV-NEXT:  .LBB26_6:
-; CHECK-NOV-NEXT:    snez a2, a0
+; CHECK-NOV-NEXT:    snez a0, a4
 ; CHECK-NOV-NEXT:    j .LBB26_9
 ; CHECK-NOV-NEXT:  .LBB26_7:
 ; CHECK-NOV-NEXT:    snez a1, a3
-; CHECK-NOV-NEXT:    and a0, a0, s0
+; CHECK-NOV-NEXT:    and a4, a4, s0
 ; CHECK-NOV-NEXT:    beqz a2, .LBB26_6
 ; CHECK-NOV-NEXT:  .LBB26_8: # %entry
-; CHECK-NOV-NEXT:    sgtz a2, a2
+; CHECK-NOV-NEXT:    sgtz a0, a2
 ; CHECK-NOV-NEXT:  .LBB26_9: # %entry
-; CHECK-NOV-NEXT:    neg a2, a2
-; CHECK-NOV-NEXT:    and a0, a2, a0
+; CHECK-NOV-NEXT:    neg a0, a0
+; CHECK-NOV-NEXT:    and a0, a0, a4
 ; CHECK-NOV-NEXT:    neg a1, a1
 ; CHECK-NOV-NEXT:    and a1, a1, a3
 ; CHECK-NOV-NEXT:    ld ra, 24(sp) # 8-byte Folded Reload
@@ -3330,15 +3335,15 @@ define <2 x i64> @ustest_f16i64(<2 x half> %x) {
 ; CHECK-V-NEXT:  # %bb.1: # %entry
 ; CHECK-V-NEXT:    li a2, 1
 ; CHECK-V-NEXT:  .LBB26_2: # %entry
-; CHECK-V-NEXT:    slti a3, s1, 1
 ; CHECK-V-NEXT:    slti a4, a1, 1
+; CHECK-V-NEXT:    slti a3, s1, 1
 ; CHECK-V-NEXT:    blez a1, .LBB26_4
 ; CHECK-V-NEXT:  # %bb.3: # %entry
 ; CHECK-V-NEXT:    li a1, 1
 ; CHECK-V-NEXT:  .LBB26_4: # %entry
+; CHECK-V-NEXT:    neg a3, a3
 ; CHECK-V-NEXT:    neg a4, a4
 ; CHECK-V-NEXT:    and a0, a4, a0
-; CHECK-V-NEXT:    neg a3, a3
 ; CHECK-V-NEXT:    beqz a1, .LBB26_7
 ; CHECK-V-NEXT:  # %bb.5: # %entry
 ; CHECK-V-NEXT:    sgtz a1, a1
@@ -5811,15 +5816,15 @@ define <2 x i64> @ustest_f64i64_mm(<2 x double> %x) {
 ; CHECK-NOV-NEXT:    mv s1, a1
 ; CHECK-NOV-NEXT:    fmv.d fa0, fs0
 ; CHECK-NOV-NEXT:    call __fixdfti
-; CHECK-NOV-NEXT:    mv a2, s1
-; CHECK-NOV-NEXT:    mv a3, a1
+; CHECK-NOV-NEXT:    mv a2, a1
 ; CHECK-NOV-NEXT:    blez a1, .LBB47_2
 ; CHECK-NOV-NEXT:  # %bb.1: # %entry
-; CHECK-NOV-NEXT:    li a3, 1
+; CHECK-NOV-NEXT:    li a2, 1
 ; CHECK-NOV-NEXT:  .LBB47_2: # %entry
-; CHECK-NOV-NEXT:    blez a2, .LBB47_4
+; CHECK-NOV-NEXT:    mv a3, s1
+; CHECK-NOV-NEXT:    blez s1, .LBB47_4
 ; CHECK-NOV-NEXT:  # %bb.3: # %entry
-; CHECK-NOV-NEXT:    li a2, 1
+; CHECK-NOV-NEXT:    li a3, 1
 ; CHECK-NOV-NEXT:  .LBB47_4: # %entry
 ; CHECK-NOV-NEXT:    slti a1, a1, 1
 ; CHECK-NOV-NEXT:    neg a1, a1
@@ -5827,11 +5832,11 @@ define <2 x i64> @ustest_f64i64_mm(<2 x double> %x) {
 ; CHECK-NOV-NEXT:    slti a0, s1, 1
 ; CHECK-NOV-NEXT:    neg a0, a0
 ; CHECK-NOV-NEXT:    and a0, a0, s0
+; CHECK-NOV-NEXT:    slti a3, a3, 0
+; CHECK-NOV-NEXT:    addi a3, a3, -1
+; CHECK-NOV-NEXT:    and a0, a3, a0
 ; CHECK-NOV-NEXT:    slti a2, a2, 0
 ; CHECK-NOV-NEXT:    addi a2, a2, -1
-; CHECK-NOV-NEXT:    and a0, a2, a0
-; CHECK-NOV-NEXT:    slti a2, a3, 0
-; CHECK-NOV-NEXT:    addi a2, a2, -1
 ; CHECK-NOV-NEXT:    and a1, a2, a1
 ; CHECK-NOV-NEXT:    ld ra, 24(sp) # 8-byte Folded Reload
 ; CHECK-NOV-NEXT:    ld s0, 16(sp) # 8-byte Folded Reload
@@ -5867,15 +5872,15 @@ define <2 x i64> @ustest_f64i64_mm(<2 x double> %x) {
 ; CHECK-V-NEXT:    vl1r.v v8, (a0) # Unknown-size Folded Reload
 ; CHECK-V-NEXT:    vfmv.f.s fa0, v8
 ; CHECK-V-NEXT:    call __fixdfti
-; CHECK-V-NEXT:    mv a2, s1
-; CHECK-V-NEXT:    mv a3, a1
+; CHECK-V-NEXT:    mv a2, a1
 ; CHECK-V-NEXT:    blez a1, .LBB47_2
 ; CHECK-V-NEXT:  # %bb.1: # %entry
-; CHECK-V-NEXT:    li a3, 1
+; CHECK-V-NEXT:    li a2, 1
 ; CHECK-V-NEXT:  .LBB47_2: # %entry
-; CHECK-V-NEXT:    blez a2, .LBB47_4
+; CHECK-V-NEXT:    mv a3, s1
+; CHECK-V-NEXT:    blez s1, .LBB47_4
 ; CHECK-V-NEXT:  # %bb.3: # %entry
-; CHECK-V-NEXT:    li a2, 1
+; CHECK-V-NEXT:    li a3, 1
 ; CHECK-V-NEXT:  .LBB47_4: # %entry
 ; CHECK-V-NEXT:    slti a1, a1, 1
 ; CHECK-V-NEXT:    neg a1, a1
@@ -5883,11 +5888,11 @@ define <2 x i64> @ustest_f64i64_mm(<2 x double> %x) {
 ; CHECK-V-NEXT:    slti a1, s1, 1
 ; CHECK-V-NEXT:    neg a1, a1
 ; CHECK-V-NEXT:    and a1, a1, s0
+; CHECK-V-NEXT:    slti a3, a3, 0
+; CHECK-V-NEXT:    addi a3, a3, -1
+; CHECK-V-NEXT:    and a1, a3, a1
 ; CHECK-V-NEXT:    slti a2, a2, 0
 ; CHECK-V-NEXT:    addi a2, a2, -1
-; CHECK-V-NEXT:    and a1, a2, a1
-; CHECK-V-NEXT:    slti a2, a3, 0
-; CHECK-V-NEXT:    addi a2, a2, -1
 ; CHECK-V-NEXT:    and a0, a2, a0
 ; CHECK-V-NEXT:    vsetivli zero, 2, e64, m1, ta, ma
 ; CHECK-V-NEXT:    vmv.s.x v8, a0
@@ -6197,15 +6202,15 @@ define <2 x i64> @ustest_f32i64_mm(<2 x float> %x) {
 ; CHECK-NOV-NEXT:    mv s1, a1
 ; CHECK-NOV-NEXT:    fmv.s fa0, fs0
 ; CHECK-NOV-NEXT:    call __fixsfti
-; CHECK-NOV-NEXT:    mv a2, s1
-; CHECK-NOV-NEXT:    mv a3, a1
+; CHECK-NOV-NEXT:    mv a2, a1
 ; CHECK-NOV-NEXT:    blez a1, .LBB50_2
 ; CHECK-NOV-NEXT:  # %bb.1: # %entry
-; CHECK-NOV-NEXT:    li a3, 1
+; CHECK-NOV-NEXT:    li a2, 1
 ; CHECK-NOV-NEXT:  .LBB50_2: # %entry
-; CHECK-NOV-NEXT:    blez a2, .LBB50_4
+; CHECK-NOV-NEXT:    mv a3, s1
+; CHECK-NOV-NEXT:    blez s1, .LBB50_4
 ; CHECK-NOV-NEXT:  # %bb.3: # %entry
-; CHECK-NOV-NEXT:    li a2, 1
+; CHECK-NOV-NEXT:    li a3, 1
 ; CHECK-NOV-NEXT:  .LBB50_4: # %entry
 ; CHECK-NOV-NEXT:    slti a1, a1, 1
 ; CHECK-NOV-NEXT:    neg a1, a1
@@ -6213,11 +6218,11 @@ define <2 x i64> @ustest_f32i64_mm(<2 x float> %x) {
 ; CHECK-NOV-NEXT:    slti a0, s1, 1
 ; CHECK-NOV-NEXT:    neg a0, a0
 ; CHECK-NOV-NEXT:    and a0, a0, s0
+; CHECK-NOV-NEXT:    slti a3, a3, 0
+; CHECK-NOV-NEXT:    addi a3, a3, -1
+; CHECK-NOV-NEXT:    and a0, a3, a0
 ; CHECK-NOV-NEXT:    slti a2, a2, 0
 ; CHECK-NOV-NEXT:    addi a2, a2, -1
-; CHECK-NOV-NEXT:    and a0, a2, a0
-; CHECK-NOV-NEXT:    slti a2, a3, 0
-; CHECK-NOV-NEXT:    addi a2, a2, -1
 ; CHECK-NOV-NEXT:    and a1, a2, a1
 ; CHECK-NOV-NEXT:    ld ra, 24(sp) # 8-byte Folded Reload
 ; CHECK-NOV-NEXT:    ld s0, 16(sp) # 8-byte Folded Reload
@@ -6253,15 +6258,15 @@ define <2 x i64> @ustest_f32i64_mm(<2 x float> %x) {
 ; CHECK-V-NEXT:    vl1r.v v8, (a0) # Unknown-size Folded Reload
 ; CHECK-V-NEXT:    vfmv.f.s fa0, v8
 ; CHECK-V-NEXT:    call __fixsfti
-; CHECK-V-NEXT:    mv a2, s1
-; CHECK-V-NEXT:    mv a3, a1
+; CHECK-V-NEXT:    mv a2, a1
 ; CHECK-V-NEXT:    blez a1, .LBB50_2
 ; CHECK-V-NEXT:  # %bb.1: # %entry
-; CHECK-V-NEXT:    li a3, 1
+; CHECK-V-NEXT:    li a2, 1
 ; CHECK-V-NEXT:  .LBB50_2: # %entry
-; CHECK-V-NEXT:    blez a2, .LBB50_4
+; CHECK-V-NEXT:    mv a3, s1
+; CHECK-V-NEXT:    blez s1, .LBB50_4
 ; CHECK-V-NEXT:  # %bb.3: # %entry
-; CHECK-V-NEXT:    li a2, 1
+; CHECK-V-NEXT:    li a3, 1
 ; CHECK-V-NEXT:  .LBB50_4: # %entry
 ; CHECK-V-NEXT:    slti a1, a1, 1
 ; CHECK-V-NEXT:    neg a1, a1
@@ -6269,11 +6274,11 @@ define <2 x i64> @ustest_f32i64_mm(<2 x float> %x) {
 ; CHECK-V-NEXT:    slti a1, s1, 1
 ; CHECK-V-NEXT:    neg a1, a1
 ; CHECK-V-NEXT:    and a1, a1, s0
+; CHECK-V-NEXT:    slti a3, a3, 0
+; CHECK-V-NEXT:    addi a3, a3, -1
+; CHECK-V-NEXT:    and a1, a3, a1
 ; CHECK-V-NEXT:    slti a2, a2, 0
 ; CHECK-V-NEXT:    addi a2, a2, -1
-; CHECK-V-NEXT:    and a1, a2, a1
-; CHECK-V-NEXT:    slti a2, a3, 0
-; CHECK-V-NEXT:    addi a2, a2, -1
 ; CHECK-V-NEXT:    and a0, a2, a0
 ; CHECK-V-NEXT:    vsetivli zero, 2, e64, m1, ta, ma
 ; CHECK-V-NEXT:    vmv.s.x v8, a0
@@ -6575,15 +6580,15 @@ define <2 x i64> @ustest_f16i64_mm(<2 x half> %x) {
 ; CHECK-NOV-NEXT:    fmv.w.x fa0, s2
 ; CHECK-NOV-NEXT:    call __extendhfsf2
 ; CHECK-NOV-NEXT:    call __fixsfti
-; CHECK-NOV-NEXT:    mv a2, s1
-; CHECK-NOV-NEXT:    mv a3, a1
+; CHECK-NOV-NEXT:    mv a2, a1
 ; CHECK-NOV-NEXT:    blez a1, .LBB53_2
 ; CHECK-NOV-NEXT:  # %bb.1: # %entry
-; CHECK-NOV-NEXT:    li a3, 1
+; CHECK-NOV-NEXT:    li a2, 1
 ; CHECK-NOV-NEXT:  .LBB53_2: # %entry
-; CHECK-NOV-NEXT:    blez a2, .LBB53_4
+; CHECK-NOV-NEXT:    mv a3, s1
+; CHECK-NOV-NEXT:    blez s1, .LBB53_4
 ; CHECK-NOV-NEXT:  # %bb.3: # %entry
-; CHECK-NOV-NEXT:    li a2, 1
+; CHECK-NOV-NEXT:    li a3, 1
 ; CHECK-NOV-NEXT:  .LBB53_4: # %entry
 ; CHECK-NOV-NEXT:    slti a1, a1, 1
 ; CHECK-NOV-NEXT:    neg a1, a1
@@ -6591,11 +6596,11 @@ define <2 x i64> @ustest_f16i64_mm(<2 x half> %x) {
 ; CHECK-NOV-NEXT:    slti a0, s1, 1
 ; CHECK-NOV-NEXT:    neg a0, a0
 ; CHECK-NOV-NEXT:    and a0, a0, s0
+; CHECK-NOV-NEXT:    slti a3, a3, 0
+; CHECK-NOV-NEXT:    addi a3, a3, -1
+; CHECK-NOV-NEXT:    and a0, a3, a0
 ; CHECK-NOV-NEXT:    slti a2, a2, 0
 ; CHECK-NOV-NEXT:    addi a2, a2, -1
-; CHECK-NOV-NEXT:    and a0, a2, a0
-; CHECK-NOV-NEXT:    slti a2, a3, 0
-; CHECK-NOV-NEXT:    addi a2, a2, -1
 ; CHECK-NOV-NEXT:    and a1, a2, a1
 ; CHECK-NOV-NEXT:    ld ra, 24(sp) # 8-byte Folded Reload
 ; CHECK-NOV-NEXT:    ld s0, 16(sp) # 8-byte Folded Reload
@@ -6625,15 +6630,15 @@ define <2 x i64> @ustest_f16i64_mm(<2 x half> %x) {
 ; CHECK-V-NEXT:    fmv.w.x fa0, s2
 ; CHECK-V-NEXT:    call __extendhfsf2
 ; CHECK-V-NEXT:    call __fixsfti
-; CHECK-V-NEXT:    mv a2, s1
-; CHECK-V-NEXT:    mv a3, a1
+; CHECK-V-NEXT:    mv a2, a1
 ; CHECK-V-NEXT:    blez a1, .LBB53_2
 ; CHECK-V-NEXT:  # %bb.1: # %entry
-; CHECK-V-NEXT:    li a3, 1
+; CHECK-V-NEXT:    li a2, 1
 ; CHECK-V-NEXT:  .LBB53_2: # %entry
-; CHECK-V-NEXT:    blez a2, .LBB53_4
+; CHECK-V-NEXT:    mv a3, s1
+; CHECK-V-NEXT:    blez s1, .LBB53_4
 ; CHECK-V-NEXT:  # %bb.3: # %entry
-; CHECK-V-NEXT:    li a2, 1
+; CHECK-V-NEXT:    li a3, 1
 ; CHECK-V-NEXT:  .LBB53_4: # %entry
 ; CHECK-V-NEXT:    slti a1, a1, 1
 ; CHECK-V-NEXT:    neg a1, a1
@@ -6641,11 +6646,11 @@ define <2 x i64> @ustest_f16i64_mm(<2 x half> %x) {
 ; CHECK-V-NEXT:    slti a1, s1, 1
 ; CHECK-V-NEXT:    neg a1, a1
 ; CHECK-V-NEXT:    and a1, a1, s0
+; CHECK-V-NEXT:    slti a3, a3, 0
+; CHECK-V-NEXT:    addi a3, a3, -1
+; CHECK-V-NEXT:    and a1, a3, a1
 ; CHECK-V-NEXT:    slti a2, a2, 0
 ; CHECK-V-NEXT:    addi a2, a2, -1
-; CHECK-V-NEXT:    and a1, a2, a1
-; CHECK-V-NEXT:    slti a2, a3, 0
-; CHECK-V-NEXT:    addi a2, a2, -1
 ; CHECK-V-NEXT:    and a0, a2, a0
 ; CHECK-V-NEXT:    vsetivli zero, 2, e64, m1, ta, ma
 ; CHECK-V-NEXT:    vmv.s.x v9, a0
diff --git a/llvm/test/CodeGen/RISCV/rvv/named-vector-shuffle-reverse.ll b/llvm/test/CodeGen/RISCV/rvv/named-vector-shuffle-reverse.ll
index 4e08f401ca4e..96094eea631b 100644
--- a/llvm/test/CodeGen/RISCV/rvv/named-vector-shuffle-reverse.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/named-vector-shuffle-reverse.ll
@@ -104,7 +104,7 @@ define <vscale x 2 x i1> @reverse_nxv2i1(<vscale x 2 x i1> %a) {
 ; RV64-BITS-512-NEXT:    vand.vi v8, v10, 1
 ; RV64-BITS-512-NEXT:    vmsne.vi v0, v8, 0
 ; RV64-BITS-512-NEXT:    ret
-  %res = call <vscale x 2 x i1> @llvm.experimental.vector.reverse.nxv2i1(<vscale x 2 x i1> %a)
+  %res = call <vscale x 2 x i1> @llvm.vector.reverse.nxv2i1(<vscale x 2 x i1> %a)
   ret <vscale x 2 x i1> %res
 }
 
@@ -202,7 +202,7 @@ define <vscale x 4 x i1> @reverse_nxv4i1(<vscale x 4 x i1> %a) {
 ; RV64-BITS-512-NEXT:    vand.vi v8, v10, 1
 ; RV64-BITS-512-NEXT:    vmsne.vi v0, v8, 0
 ; RV64-BITS-512-NEXT:    ret
-  %res = call <vscale x 4 x i1> @llvm.experimental.vector.reverse.nxv4i1(<vscale x 4 x i1> %a)
+  %res = call <vscale x 4 x i1> @llvm.vector.reverse.nxv4i1(<vscale x 4 x i1> %a)
   ret <vscale x 4 x i1> %res
 }
 
@@ -294,7 +294,7 @@ define <vscale x 8 x i1> @reverse_nxv8i1(<vscale x 8 x i1> %a) {
 ; RV64-BITS-512-NEXT:    vand.vi v8, v10, 1
 ; RV64-BITS-512-NEXT:    vmsne.vi v0, v8, 0
 ; RV64-BITS-512-NEXT:    ret
-  %res = call <vscale x 8 x i1> @llvm.experimental.vector.reverse.nxv8i1(<vscale x 8 x i1> %a)
+  %res = call <vscale x 8 x i1> @llvm.vector.reverse.nxv8i1(<vscale x 8 x i1> %a)
   ret <vscale x 8 x i1> %res
 }
 
@@ -392,7 +392,7 @@ define <vscale x 16 x i1> @reverse_nxv16i1(<vscale x 16 x i1> %a) {
 ; RV64-BITS-512-NEXT:    vand.vi v8, v12, 1
 ; RV64-BITS-512-NEXT:    vmsne.vi v0, v8, 0
 ; RV64-BITS-512-NEXT:    ret
-  %res = call <vscale x 16 x i1> @llvm.experimental.vector.reverse.nxv16i1(<vscale x 16 x i1> %a)
+  %res = call <vscale x 16 x i1> @llvm.vector.reverse.nxv16i1(<vscale x 16 x i1> %a)
   ret <vscale x 16 x i1> %res
 }
 
@@ -490,7 +490,7 @@ define <vscale x 32 x i1> @reverse_nxv32i1(<vscale x 32 x i1> %a) {
 ; RV64-BITS-512-NEXT:    vand.vi v8, v16, 1
 ; RV64-BITS-512-NEXT:    vmsne.vi v0, v8, 0
 ; RV64-BITS-512-NEXT:    ret
-  %res = call <vscale x 32 x i1> @llvm.experimental.vector.reverse.nxv32i1(<vscale x 32 x i1> %a)
+  %res = call <vscale x 32 x i1> @llvm.vector.reverse.nxv32i1(<vscale x 32 x i1> %a)
   ret <vscale x 32 x i1> %res
 }
 
@@ -600,7 +600,7 @@ define <vscale x 64 x i1> @reverse_nxv64i1(<vscale x 64 x i1> %a) {
 ; RV64-BITS-512-NEXT:    vand.vi v8, v24, 1
 ; RV64-BITS-512-NEXT:    vmsne.vi v0, v8, 0
 ; RV64-BITS-512-NEXT:    ret
-  %res = call <vscale x 64 x i1> @llvm.experimental.vector.reverse.nxv64i1(<vscale x 64 x i1> %a)
+  %res = call <vscale x 64 x i1> @llvm.vector.reverse.nxv64i1(<vscale x 64 x i1> %a)
   ret <vscale x 64 x i1> %res
 }
 
@@ -682,7 +682,7 @@ define <vscale x 1 x i8> @reverse_nxv1i8(<vscale x 1 x i8> %a) {
 ; RV64-BITS-512-NEXT:    vrgather.vv v9, v8, v10
 ; RV64-BITS-512-NEXT:    vmv1r.v v8, v9
 ; RV64-BITS-512-NEXT:    ret
-  %res = call <vscale x 1 x i8> @llvm.experimental.vector.reverse.nxv1i8(<vscale x 1 x i8> %a)
+  %res = call <vscale x 1 x i8> @llvm.vector.reverse.nxv1i8(<vscale x 1 x i8> %a)
   ret <vscale x 1 x i8> %res
 }
 
@@ -760,7 +760,7 @@ define <vscale x 2 x i8> @reverse_nxv2i8(<vscale x 2 x i8> %a) {
 ; RV64-BITS-512-NEXT:    vrgather.vv v9, v8, v10
 ; RV64-BITS-512-NEXT:    vmv1r.v v8, v9
 ; RV64-BITS-512-NEXT:    ret
-  %res = call <vscale x 2 x i8> @llvm.experimental.vector.reverse.nxv2i8(<vscale x 2 x i8> %a)
+  %res = call <vscale x 2 x i8> @llvm.vector.reverse.nxv2i8(<vscale x 2 x i8> %a)
   ret <vscale x 2 x i8> %res
 }
 
@@ -838,7 +838,7 @@ define <vscale x 4 x i8> @reverse_nxv4i8(<vscale x 4 x i8> %a) {
 ; RV64-BITS-512-NEXT:    vrgather.vv v9, v8, v10
 ; RV64-BITS-512-NEXT:    vmv1r.v v8, v9
 ; RV64-BITS-512-NEXT:    ret
-  %res = call <vscale x 4 x i8> @llvm.experimental.vector.reverse.nxv4i8(<vscale x 4 x i8> %a)
+  %res = call <vscale x 4 x i8> @llvm.vector.reverse.nxv4i8(<vscale x 4 x i8> %a)
   ret <vscale x 4 x i8> %res
 }
 
@@ -910,7 +910,7 @@ define <vscale x 8 x i8> @reverse_nxv8i8(<vscale x 8 x i8> %a) {
 ; RV64-BITS-512-NEXT:    vrgather.vv v9, v8, v10
 ; RV64-BITS-512-NEXT:    vmv.v.v v8, v9
 ; RV64-BITS-512-NEXT:    ret
-  %res = call <vscale x 8 x i8> @llvm.experimental.vector.reverse.nxv8i8(<vscale x 8 x i8> %a)
+  %res = call <vscale x 8 x i8> @llvm.vector.reverse.nxv8i8(<vscale x 8 x i8> %a)
   ret <vscale x 8 x i8> %res
 }
 
@@ -988,7 +988,7 @@ define <vscale x 16 x i8> @reverse_nxv16i8(<vscale x 16 x i8> %a) {
 ; RV64-BITS-512-NEXT:    vrgather.vv v10, v8, v12
 ; RV64-BITS-512-NEXT:    vmv.v.v v8, v10
 ; RV64-BITS-512-NEXT:    ret
-  %res = call <vscale x 16 x i8> @llvm.experimental.vector.reverse.nxv16i8(<vscale x 16 x i8> %a)
+  %res = call <vscale x 16 x i8> @llvm.vector.reverse.nxv16i8(<vscale x 16 x i8> %a)
   ret <vscale x 16 x i8> %res
 }
 
@@ -1066,7 +1066,7 @@ define <vscale x 32 x i8> @reverse_nxv32i8(<vscale x 32 x i8> %a) {
 ; RV64-BITS-512-NEXT:    vrgather.vv v12, v8, v16
 ; RV64-BITS-512-NEXT:    vmv.v.v v8, v12
 ; RV64-BITS-512-NEXT:    ret
-  %res = call <vscale x 32 x i8> @llvm.experimental.vector.reverse.nxv32i8(<vscale x 32 x i8> %a)
+  %res = call <vscale x 32 x i8> @llvm.vector.reverse.nxv32i8(<vscale x 32 x i8> %a)
   ret <vscale x 32 x i8> %res
 }
 
@@ -1148,7 +1148,7 @@ define <vscale x 64 x i8> @reverse_nxv64i8(<vscale x 64 x i8> %a) {
 ; RV64-BITS-512-NEXT:    vrgather.vv v16, v12, v24
 ; RV64-BITS-512-NEXT:    vmv8r.v v8, v16
 ; RV64-BITS-512-NEXT:    ret
-  %res = call <vscale x 64 x i8> @llvm.experimental.vector.reverse.nxv64i8(<vscale x 64 x i8> %a)
+  %res = call <vscale x 64 x i8> @llvm.vector.reverse.nxv64i8(<vscale x 64 x i8> %a)
   ret <vscale x 64 x i8> %res
 }
 
@@ -1164,7 +1164,7 @@ define <vscale x 1 x i16> @reverse_nxv1i16(<vscale x 1 x i16> %a) {
 ; CHECK-NEXT:    vrgather.vv v9, v8, v10
 ; CHECK-NEXT:    vmv1r.v v8, v9
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 1 x i16> @llvm.experimental.vector.reverse.nxv1i16(<vscale x 1 x i16> %a)
+  %res = call <vscale x 1 x i16> @llvm.vector.reverse.nxv1i16(<vscale x 1 x i16> %a)
   ret <vscale x 1 x i16> %res
 }
 
@@ -1180,7 +1180,7 @@ define <vscale x 2 x i16> @reverse_nxv2i16(<vscale x 2 x i16> %a) {
 ; CHECK-NEXT:    vrgather.vv v9, v8, v10
 ; CHECK-NEXT:    vmv1r.v v8, v9
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 2 x i16> @llvm.experimental.vector.reverse.nxv2i16(<vscale x 2 x i16> %a)
+  %res = call <vscale x 2 x i16> @llvm.vector.reverse.nxv2i16(<vscale x 2 x i16> %a)
   ret <vscale x 2 x i16> %res
 }
 
@@ -1196,7 +1196,7 @@ define <vscale x 4 x i16> @reverse_nxv4i16(<vscale x 4 x i16> %a) {
 ; CHECK-NEXT:    vrgather.vv v9, v8, v10
 ; CHECK-NEXT:    vmv.v.v v8, v9
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 4 x i16> @llvm.experimental.vector.reverse.nxv4i16(<vscale x 4 x i16> %a)
+  %res = call <vscale x 4 x i16> @llvm.vector.reverse.nxv4i16(<vscale x 4 x i16> %a)
   ret <vscale x 4 x i16> %res
 }
 
@@ -1211,7 +1211,7 @@ define <vscale x 8 x i16> @reverse_nxv8i16(<vscale x 8 x i16> %a) {
 ; CHECK-NEXT:    vrgather.vv v10, v8, v12
 ; CHECK-NEXT:    vmv.v.v v8, v10
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 8 x i16> @llvm.experimental.vector.reverse.nxv8i16(<vscale x 8 x i16> %a)
+  %res = call <vscale x 8 x i16> @llvm.vector.reverse.nxv8i16(<vscale x 8 x i16> %a)
   ret <vscale x 8 x i16> %res
 }
 
@@ -1227,7 +1227,7 @@ define <vscale x 16 x i16> @reverse_nxv16i16(<vscale x 16 x i16> %a) {
 ; CHECK-NEXT:    vrgather.vv v12, v8, v16
 ; CHECK-NEXT:    vmv.v.v v8, v12
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 16 x i16> @llvm.experimental.vector.reverse.nxv16i16(<vscale x 16 x i16> %a)
+  %res = call <vscale x 16 x i16> @llvm.vector.reverse.nxv16i16(<vscale x 16 x i16> %a)
   ret <vscale x 16 x i16> %res
 }
 
@@ -1243,7 +1243,7 @@ define <vscale x 32 x i16> @reverse_nxv32i16(<vscale x 32 x i16> %a) {
 ; CHECK-NEXT:    vrgather.vv v16, v8, v24
 ; CHECK-NEXT:    vmv.v.v v8, v16
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 32 x i16> @llvm.experimental.vector.reverse.nxv32i16(<vscale x 32 x i16> %a)
+  %res = call <vscale x 32 x i16> @llvm.vector.reverse.nxv32i16(<vscale x 32 x i16> %a)
   ret <vscale x 32 x i16> %res
 }
 
@@ -1259,7 +1259,7 @@ define <vscale x 1 x i32> @reverse_nxv1i32(<vscale x 1 x i32> %a) {
 ; CHECK-NEXT:    vrgather.vv v9, v8, v10
 ; CHECK-NEXT:    vmv1r.v v8, v9
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 1 x i32> @llvm.experimental.vector.reverse.nxv1i32(<vscale x 1 x i32> %a)
+  %res = call <vscale x 1 x i32> @llvm.vector.reverse.nxv1i32(<vscale x 1 x i32> %a)
   ret <vscale x 1 x i32> %res
 }
 
@@ -1275,7 +1275,7 @@ define <vscale x 2 x i32> @reverse_nxv2i32(<vscale x 2 x i32> %a) {
 ; CHECK-NEXT:    vrgather.vv v9, v8, v10
 ; CHECK-NEXT:    vmv.v.v v8, v9
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 2 x i32> @llvm.experimental.vector.reverse.nxv2i32(<vscale x 2 x i32> %a)
+  %res = call <vscale x 2 x i32> @llvm.vector.reverse.nxv2i32(<vscale x 2 x i32> %a)
   ret <vscale x 2 x i32> %res
 }
 
@@ -1291,7 +1291,7 @@ define <vscale x 4 x i32> @reverse_nxv4i32(<vscale x 4 x i32> %a) {
 ; CHECK-NEXT:    vrgather.vv v10, v8, v12
 ; CHECK-NEXT:    vmv.v.v v8, v10
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> %a)
+  %res = call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> %a)
   ret <vscale x 4 x i32> %res
 }
 
@@ -1306,7 +1306,7 @@ define <vscale x 8 x i32> @reverse_nxv8i32(<vscale x 8 x i32> %a) {
 ; CHECK-NEXT:    vrgather.vv v12, v8, v16
 ; CHECK-NEXT:    vmv.v.v v8, v12
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 8 x i32> @llvm.experimental.vector.reverse.nxv8i32(<vscale x 8 x i32> %a)
+  %res = call <vscale x 8 x i32> @llvm.vector.reverse.nxv8i32(<vscale x 8 x i32> %a)
   ret <vscale x 8 x i32> %res
 }
 
@@ -1322,7 +1322,7 @@ define <vscale x 16 x i32> @reverse_nxv16i32(<vscale x 16 x i32> %a) {
 ; CHECK-NEXT:    vrgather.vv v16, v8, v24
 ; CHECK-NEXT:    vmv.v.v v8, v16
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 16 x i32> @llvm.experimental.vector.reverse.nxv16i32(<vscale x 16 x i32> %a)
+  %res = call <vscale x 16 x i32> @llvm.vector.reverse.nxv16i32(<vscale x 16 x i32> %a)
   ret <vscale x 16 x i32> %res
 }
 
@@ -1338,7 +1338,7 @@ define <vscale x 1 x i64> @reverse_nxv1i64(<vscale x 1 x i64> %a) {
 ; CHECK-NEXT:    vrgather.vv v9, v8, v10
 ; CHECK-NEXT:    vmv.v.v v8, v9
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 1 x i64> @llvm.experimental.vector.reverse.nxv1i64(<vscale x 1 x i64> %a)
+  %res = call <vscale x 1 x i64> @llvm.vector.reverse.nxv1i64(<vscale x 1 x i64> %a)
   ret <vscale x 1 x i64> %res
 }
 
@@ -1354,7 +1354,7 @@ define <vscale x 2 x i64> @reverse_nxv2i64(<vscale x 2 x i64> %a) {
 ; CHECK-NEXT:    vrgather.vv v10, v8, v12
 ; CHECK-NEXT:    vmv.v.v v8, v10
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 2 x i64> @llvm.experimental.vector.reverse.nxv2i64(<vscale x 2 x i64> %a)
+  %res = call <vscale x 2 x i64> @llvm.vector.reverse.nxv2i64(<vscale x 2 x i64> %a)
   ret <vscale x 2 x i64> %res
 }
 
@@ -1370,7 +1370,7 @@ define <vscale x 4 x i64> @reverse_nxv4i64(<vscale x 4 x i64> %a) {
 ; CHECK-NEXT:    vrgather.vv v12, v8, v16
 ; CHECK-NEXT:    vmv.v.v v8, v12
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 4 x i64> @llvm.experimental.vector.reverse.nxv4i64(<vscale x 4 x i64> %a)
+  %res = call <vscale x 4 x i64> @llvm.vector.reverse.nxv4i64(<vscale x 4 x i64> %a)
   ret <vscale x 4 x i64> %res
 }
 
@@ -1385,7 +1385,7 @@ define <vscale x 8 x i64> @reverse_nxv8i64(<vscale x 8 x i64> %a) {
 ; CHECK-NEXT:    vrgather.vv v16, v8, v24
 ; CHECK-NEXT:    vmv.v.v v8, v16
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 8 x i64> @llvm.experimental.vector.reverse.nxv8i64(<vscale x 8 x i64> %a)
+  %res = call <vscale x 8 x i64> @llvm.vector.reverse.nxv8i64(<vscale x 8 x i64> %a)
   ret <vscale x 8 x i64> %res
 }
 
@@ -1405,7 +1405,7 @@ define <vscale x 1 x half> @reverse_nxv1f16(<vscale x 1 x half> %a) {
 ; CHECK-NEXT:    vrgather.vv v9, v8, v10
 ; CHECK-NEXT:    vmv1r.v v8, v9
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 1 x half> @llvm.experimental.vector.reverse.nxv1f16(<vscale x 1 x half> %a)
+  %res = call <vscale x 1 x half> @llvm.vector.reverse.nxv1f16(<vscale x 1 x half> %a)
   ret <vscale x 1 x half> %res
 }
 
@@ -1421,7 +1421,7 @@ define <vscale x 2 x half> @reverse_nxv2f16(<vscale x 2 x half> %a) {
 ; CHECK-NEXT:    vrgather.vv v9, v8, v10
 ; CHECK-NEXT:    vmv1r.v v8, v9
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 2 x half> @llvm.experimental.vector.reverse.nxv2f16(<vscale x 2 x half> %a)
+  %res = call <vscale x 2 x half> @llvm.vector.reverse.nxv2f16(<vscale x 2 x half> %a)
   ret <vscale x 2 x half> %res
 }
 
@@ -1437,7 +1437,7 @@ define <vscale x 4 x half> @reverse_nxv4f16(<vscale x 4 x half> %a) {
 ; CHECK-NEXT:    vrgather.vv v9, v8, v10
 ; CHECK-NEXT:    vmv.v.v v8, v9
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 4 x half> @llvm.experimental.vector.reverse.nxv4f16(<vscale x 4 x half> %a)
+  %res = call <vscale x 4 x half> @llvm.vector.reverse.nxv4f16(<vscale x 4 x half> %a)
   ret <vscale x 4 x half> %res
 }
 
@@ -1452,7 +1452,7 @@ define <vscale x 8 x half> @reverse_nxv8f16(<vscale x 8 x half> %a) {
 ; CHECK-NEXT:    vrgather.vv v10, v8, v12
 ; CHECK-NEXT:    vmv.v.v v8, v10
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 8 x half> @llvm.experimental.vector.reverse.nxv8f16(<vscale x 8 x half> %a)
+  %res = call <vscale x 8 x half> @llvm.vector.reverse.nxv8f16(<vscale x 8 x half> %a)
   ret <vscale x 8 x half> %res
 }
 
@@ -1468,7 +1468,7 @@ define <vscale x 16 x half> @reverse_nxv16f16(<vscale x 16 x half> %a) {
 ; CHECK-NEXT:    vrgather.vv v12, v8, v16
 ; CHECK-NEXT:    vmv.v.v v8, v12
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 16 x half> @llvm.experimental.vector.reverse.nxv16f16(<vscale x 16 x half> %a)
+  %res = call <vscale x 16 x half> @llvm.vector.reverse.nxv16f16(<vscale x 16 x half> %a)
   ret <vscale x 16 x half> %res
 }
 
@@ -1484,7 +1484,7 @@ define <vscale x 32 x half> @reverse_nxv32f16(<vscale x 32 x half> %a) {
 ; CHECK-NEXT:    vrgather.vv v16, v8, v24
 ; CHECK-NEXT:    vmv.v.v v8, v16
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 32 x half> @llvm.experimental.vector.reverse.nxv32f16(<vscale x 32 x half> %a)
+  %res = call <vscale x 32 x half> @llvm.vector.reverse.nxv32f16(<vscale x 32 x half> %a)
   ret <vscale x 32 x half> %res
 }
 
@@ -1500,7 +1500,7 @@ define <vscale x 1 x float> @reverse_nxv1f32(<vscale x 1 x float> %a) {
 ; CHECK-NEXT:    vrgather.vv v9, v8, v10
 ; CHECK-NEXT:    vmv1r.v v8, v9
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 1 x float> @llvm.experimental.vector.reverse.nxv1f32(<vscale x 1 x float> %a)
+  %res = call <vscale x 1 x float> @llvm.vector.reverse.nxv1f32(<vscale x 1 x float> %a)
   ret <vscale x 1 x float> %res
 }
 
@@ -1516,7 +1516,7 @@ define <vscale x 2 x float> @reverse_nxv2f32(<vscale x 2 x float> %a) {
 ; CHECK-NEXT:    vrgather.vv v9, v8, v10
 ; CHECK-NEXT:    vmv.v.v v8, v9
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 2 x float> @llvm.experimental.vector.reverse.nxv2f32(<vscale x 2 x float> %a)
+  %res = call <vscale x 2 x float> @llvm.vector.reverse.nxv2f32(<vscale x 2 x float> %a)
   ret <vscale x 2 x float> %res
 }
 
@@ -1532,7 +1532,7 @@ define <vscale x 4 x float> @reverse_nxv4f32(<vscale x 4 x float> %a) {
 ; CHECK-NEXT:    vrgather.vv v10, v8, v12
 ; CHECK-NEXT:    vmv.v.v v8, v10
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 4 x float> @llvm.experimental.vector.reverse.nxv4f32(<vscale x 4 x float> %a)
+  %res = call <vscale x 4 x float> @llvm.vector.reverse.nxv4f32(<vscale x 4 x float> %a)
   ret <vscale x 4 x float> %res
 }
 
@@ -1547,7 +1547,7 @@ define <vscale x 8 x float> @reverse_nxv8f32(<vscale x 8 x float> %a) {
 ; CHECK-NEXT:    vrgather.vv v12, v8, v16
 ; CHECK-NEXT:    vmv.v.v v8, v12
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 8 x float> @llvm.experimental.vector.reverse.nxv8f32(<vscale x 8 x float> %a)
+  %res = call <vscale x 8 x float> @llvm.vector.reverse.nxv8f32(<vscale x 8 x float> %a)
   ret <vscale x 8 x float> %res
 }
 
@@ -1563,7 +1563,7 @@ define <vscale x 16 x float> @reverse_nxv16f32(<vscale x 16 x float> %a) {
 ; CHECK-NEXT:    vrgather.vv v16, v8, v24
 ; CHECK-NEXT:    vmv.v.v v8, v16
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 16 x float> @llvm.experimental.vector.reverse.nxv16f32(<vscale x 16 x float> %a)
+  %res = call <vscale x 16 x float> @llvm.vector.reverse.nxv16f32(<vscale x 16 x float> %a)
   ret <vscale x 16 x float> %res
 }
 
@@ -1579,7 +1579,7 @@ define <vscale x 1 x double> @reverse_nxv1f64(<vscale x 1 x double> %a) {
 ; CHECK-NEXT:    vrgather.vv v9, v8, v10
 ; CHECK-NEXT:    vmv.v.v v8, v9
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 1 x double> @llvm.experimental.vector.reverse.nxv1f64(<vscale x 1 x double> %a)
+  %res = call <vscale x 1 x double> @llvm.vector.reverse.nxv1f64(<vscale x 1 x double> %a)
   ret <vscale x 1 x double> %res
 }
 
@@ -1595,7 +1595,7 @@ define <vscale x 2 x double> @reverse_nxv2f64(<vscale x 2 x double> %a) {
 ; CHECK-NEXT:    vrgather.vv v10, v8, v12
 ; CHECK-NEXT:    vmv.v.v v8, v10
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 2 x double> @llvm.experimental.vector.reverse.nxv2f64(<vscale x 2 x double> %a)
+  %res = call <vscale x 2 x double> @llvm.vector.reverse.nxv2f64(<vscale x 2 x double> %a)
   ret <vscale x 2 x double> %res
 }
 
@@ -1611,7 +1611,7 @@ define <vscale x 4 x double> @reverse_nxv4f64(<vscale x 4 x double> %a) {
 ; CHECK-NEXT:    vrgather.vv v12, v8, v16
 ; CHECK-NEXT:    vmv.v.v v8, v12
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 4 x double> @llvm.experimental.vector.reverse.nxv4f64(<vscale x 4 x double> %a)
+  %res = call <vscale x 4 x double> @llvm.vector.reverse.nxv4f64(<vscale x 4 x double> %a)
   ret <vscale x 4 x double> %res
 }
 
@@ -1626,7 +1626,7 @@ define <vscale x 8 x double> @reverse_nxv8f64(<vscale x 8 x double> %a) {
 ; CHECK-NEXT:    vrgather.vv v16, v8, v24
 ; CHECK-NEXT:    vmv.v.v v8, v16
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 8 x double> @llvm.experimental.vector.reverse.nxv8f64(<vscale x 8 x double> %a)
+  %res = call <vscale x 8 x double> @llvm.vector.reverse.nxv8f64(<vscale x 8 x double> %a)
   ret <vscale x 8 x double> %res
 }
 
@@ -1646,7 +1646,7 @@ define <vscale x 3 x i64> @reverse_nxv3i64(<vscale x 3 x i64> %a) {
 ; CHECK-NEXT:    vmv1r.v v9, v18
 ; CHECK-NEXT:    vmv1r.v v10, v19
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 3 x i64> @llvm.experimental.vector.reverse.nxv3i64(<vscale x 3 x i64> %a)
+  %res = call <vscale x 3 x i64> @llvm.vector.reverse.nxv3i64(<vscale x 3 x i64> %a)
   ret <vscale x 3 x i64> %res
 }
 
@@ -1663,7 +1663,7 @@ define <vscale x 6 x i64> @reverse_nxv6i64(<vscale x 6 x i64> %a) {
 ; CHECK-NEXT:    vmv2r.v v10, v28
 ; CHECK-NEXT:    vmv2r.v v12, v30
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 6 x i64> @llvm.experimental.vector.reverse.nxv6i64(<vscale x 6 x i64> %a)
+  %res = call <vscale x 6 x i64> @llvm.vector.reverse.nxv6i64(<vscale x 6 x i64> %a)
   ret <vscale x 6 x i64> %res
 }
 
@@ -1739,53 +1739,53 @@ define <vscale x 12 x i64> @reverse_nxv12i64(<vscale x 12 x i64> %a) {
 ; RV64-NEXT:    ld s0, 64(sp) # 8-byte Folded Reload
 ; RV64-NEXT:    addi sp, sp, 80
 ; RV64-NEXT:    ret
-  %res = call <vscale x 12 x i64> @llvm.experimental.vector.reverse.nxv12i64(<vscale x 12 x i64> %a)
+  %res = call <vscale x 12 x i64> @llvm.vector.reverse.nxv12i64(<vscale x 12 x i64> %a)
   ret <vscale x 12 x i64> %res
 }
 
-declare <vscale x 2 x i1> @llvm.experimental.vector.reverse.nxv2i1(<vscale x 2 x i1>)
-declare <vscale x 4 x i1> @llvm.experimental.vector.reverse.nxv4i1(<vscale x 4 x i1>)
-declare <vscale x 8 x i1> @llvm.experimental.vector.reverse.nxv8i1(<vscale x 8 x i1>)
-declare <vscale x 16 x i1> @llvm.experimental.vector.reverse.nxv16i1(<vscale x 16 x i1>)
-declare <vscale x 32 x i1> @llvm.experimental.vector.reverse.nxv32i1(<vscale x 32 x i1>)
-declare <vscale x 64 x i1> @llvm.experimental.vector.reverse.nxv64i1(<vscale x 64 x i1>)
-declare <vscale x 1 x i8> @llvm.experimental.vector.reverse.nxv1i8(<vscale x 1 x i8>)
-declare <vscale x 2 x i8> @llvm.experimental.vector.reverse.nxv2i8(<vscale x 2 x i8>)
-declare <vscale x 4 x i8> @llvm.experimental.vector.reverse.nxv4i8(<vscale x 4 x i8>)
-declare <vscale x 8 x i8> @llvm.experimental.vector.reverse.nxv8i8(<vscale x 8 x i8>)
-declare <vscale x 16 x i8> @llvm.experimental.vector.reverse.nxv16i8(<vscale x 16 x i8>)
-declare <vscale x 32 x i8> @llvm.experimental.vector.reverse.nxv32i8(<vscale x 32 x i8>)
-declare <vscale x 64 x i8> @llvm.experimental.vector.reverse.nxv64i8(<vscale x 64 x i8>)
-declare <vscale x 1 x i16> @llvm.experimental.vector.reverse.nxv1i16(<vscale x 1 x i16>)
-declare <vscale x 2 x i16> @llvm.experimental.vector.reverse.nxv2i16(<vscale x 2 x i16>)
-declare <vscale x 4 x i16> @llvm.experimental.vector.reverse.nxv4i16(<vscale x 4 x i16>)
-declare <vscale x 8 x i16> @llvm.experimental.vector.reverse.nxv8i16(<vscale x 8 x i16>)
-declare <vscale x 16 x i16> @llvm.experimental.vector.reverse.nxv16i16(<vscale x 16 x i16>)
-declare <vscale x 32 x i16> @llvm.experimental.vector.reverse.nxv32i16(<vscale x 32 x i16>)
-declare <vscale x 1 x i32> @llvm.experimental.vector.reverse.nxv1i32(<vscale x 1 x i32>)
-declare <vscale x 2 x i32> @llvm.experimental.vector.reverse.nxv2i32(<vscale x 2 x i32>)
-declare <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32>)
-declare <vscale x 8 x i32> @llvm.experimental.vector.reverse.nxv8i32(<vscale x 8 x i32>)
-declare <vscale x 16 x i32> @llvm.experimental.vector.reverse.nxv16i32(<vscale x 16 x i32>)
-declare <vscale x 1 x i64> @llvm.experimental.vector.reverse.nxv1i64(<vscale x 1 x i64>)
-declare <vscale x 2 x i64> @llvm.experimental.vector.reverse.nxv2i64(<vscale x 2 x i64>)
-declare <vscale x 4 x i64> @llvm.experimental.vector.reverse.nxv4i64(<vscale x 4 x i64>)
-declare <vscale x 8 x i64> @llvm.experimental.vector.reverse.nxv8i64(<vscale x 8 x i64>)
-declare <vscale x 1 x half> @llvm.experimental.vector.reverse.nxv1f16(<vscale x 1 x half>)
-declare <vscale x 2 x half> @llvm.experimental.vector.reverse.nxv2f16(<vscale x 2 x half>)
-declare <vscale x 4 x half> @llvm.experimental.vector.reverse.nxv4f16(<vscale x 4 x half>)
-declare <vscale x 8 x half> @llvm.experimental.vector.reverse.nxv8f16(<vscale x 8 x half>)
-declare <vscale x 16 x half> @llvm.experimental.vector.reverse.nxv16f16(<vscale x 16 x half>)
-declare <vscale x 32 x half> @llvm.experimental.vector.reverse.nxv32f16(<vscale x 32 x half>)
-declare <vscale x 1 x float> @llvm.experimental.vector.reverse.nxv1f32(<vscale x 1 x float>)
-declare <vscale x 2 x float> @llvm.experimental.vector.reverse.nxv2f32(<vscale x 2 x float>)
-declare <vscale x 4 x float> @llvm.experimental.vector.reverse.nxv4f32(<vscale x 4 x float>)
-declare <vscale x 8 x float> @llvm.experimental.vector.reverse.nxv8f32(<vscale x 8 x float>)
-declare <vscale x 16 x float> @llvm.experimental.vector.reverse.nxv16f32(<vscale x 16 x float>)
-declare <vscale x 1 x double> @llvm.experimental.vector.reverse.nxv1f64(<vscale x 1 x double>)
-declare <vscale x 2 x double> @llvm.experimental.vector.reverse.nxv2f64(<vscale x 2 x double>)
-declare <vscale x 4 x double> @llvm.experimental.vector.reverse.nxv4f64(<vscale x 4 x double>)
-declare <vscale x 8 x double> @llvm.experimental.vector.reverse.nxv8f64(<vscale x 8 x double>)
-declare <vscale x 3 x i64> @llvm.experimental.vector.reverse.nxv3i64(<vscale x 3 x i64>)
-declare <vscale x 6 x i64> @llvm.experimental.vector.reverse.nxv6i64(<vscale x 6 x i64>)
-declare <vscale x 12 x i64> @llvm.experimental.vector.reverse.nxv12i64(<vscale x 12 x i64>)
+declare <vscale x 2 x i1> @llvm.vector.reverse.nxv2i1(<vscale x 2 x i1>)
+declare <vscale x 4 x i1> @llvm.vector.reverse.nxv4i1(<vscale x 4 x i1>)
+declare <vscale x 8 x i1> @llvm.vector.reverse.nxv8i1(<vscale x 8 x i1>)
+declare <vscale x 16 x i1> @llvm.vector.reverse.nxv16i1(<vscale x 16 x i1>)
+declare <vscale x 32 x i1> @llvm.vector.reverse.nxv32i1(<vscale x 32 x i1>)
+declare <vscale x 64 x i1> @llvm.vector.reverse.nxv64i1(<vscale x 64 x i1>)
+declare <vscale x 1 x i8> @llvm.vector.reverse.nxv1i8(<vscale x 1 x i8>)
+declare <vscale x 2 x i8> @llvm.vector.reverse.nxv2i8(<vscale x 2 x i8>)
+declare <vscale x 4 x i8> @llvm.vector.reverse.nxv4i8(<vscale x 4 x i8>)
+declare <vscale x 8 x i8> @llvm.vector.reverse.nxv8i8(<vscale x 8 x i8>)
+declare <vscale x 16 x i8> @llvm.vector.reverse.nxv16i8(<vscale x 16 x i8>)
+declare <vscale x 32 x i8> @llvm.vector.reverse.nxv32i8(<vscale x 32 x i8>)
+declare <vscale x 64 x i8> @llvm.vector.reverse.nxv64i8(<vscale x 64 x i8>)
+declare <vscale x 1 x i16> @llvm.vector.reverse.nxv1i16(<vscale x 1 x i16>)
+declare <vscale x 2 x i16> @llvm.vector.reverse.nxv2i16(<vscale x 2 x i16>)
+declare <vscale x 4 x i16> @llvm.vector.reverse.nxv4i16(<vscale x 4 x i16>)
+declare <vscale x 8 x i16> @llvm.vector.reverse.nxv8i16(<vscale x 8 x i16>)
+declare <vscale x 16 x i16> @llvm.vector.reverse.nxv16i16(<vscale x 16 x i16>)
+declare <vscale x 32 x i16> @llvm.vector.reverse.nxv32i16(<vscale x 32 x i16>)
+declare <vscale x 1 x i32> @llvm.vector.reverse.nxv1i32(<vscale x 1 x i32>)
+declare <vscale x 2 x i32> @llvm.vector.reverse.nxv2i32(<vscale x 2 x i32>)
+declare <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32>)
+declare <vscale x 8 x i32> @llvm.vector.reverse.nxv8i32(<vscale x 8 x i32>)
+declare <vscale x 16 x i32> @llvm.vector.reverse.nxv16i32(<vscale x 16 x i32>)
+declare <vscale x 1 x i64> @llvm.vector.reverse.nxv1i64(<vscale x 1 x i64>)
+declare <vscale x 2 x i64> @llvm.vector.reverse.nxv2i64(<vscale x 2 x i64>)
+declare <vscale x 4 x i64> @llvm.vector.reverse.nxv4i64(<vscale x 4 x i64>)
+declare <vscale x 8 x i64> @llvm.vector.reverse.nxv8i64(<vscale x 8 x i64>)
+declare <vscale x 1 x half> @llvm.vector.reverse.nxv1f16(<vscale x 1 x half>)
+declare <vscale x 2 x half> @llvm.vector.reverse.nxv2f16(<vscale x 2 x half>)
+declare <vscale x 4 x half> @llvm.vector.reverse.nxv4f16(<vscale x 4 x half>)
+declare <vscale x 8 x half> @llvm.vector.reverse.nxv8f16(<vscale x 8 x half>)
+declare <vscale x 16 x half> @llvm.vector.reverse.nxv16f16(<vscale x 16 x half>)
+declare <vscale x 32 x half> @llvm.vector.reverse.nxv32f16(<vscale x 32 x half>)
+declare <vscale x 1 x float> @llvm.vector.reverse.nxv1f32(<vscale x 1 x float>)
+declare <vscale x 2 x float> @llvm.vector.reverse.nxv2f32(<vscale x 2 x float>)
+declare <vscale x 4 x float> @llvm.vector.reverse.nxv4f32(<vscale x 4 x float>)
+declare <vscale x 8 x float> @llvm.vector.reverse.nxv8f32(<vscale x 8 x float>)
+declare <vscale x 16 x float> @llvm.vector.reverse.nxv16f32(<vscale x 16 x float>)
+declare <vscale x 1 x double> @llvm.vector.reverse.nxv1f64(<vscale x 1 x double>)
+declare <vscale x 2 x double> @llvm.vector.reverse.nxv2f64(<vscale x 2 x double>)
+declare <vscale x 4 x double> @llvm.vector.reverse.nxv4f64(<vscale x 4 x double>)
+declare <vscale x 8 x double> @llvm.vector.reverse.nxv8f64(<vscale x 8 x double>)
+declare <vscale x 3 x i64> @llvm.vector.reverse.nxv3i64(<vscale x 3 x i64>)
+declare <vscale x 6 x i64> @llvm.vector.reverse.nxv6i64(<vscale x 6 x i64>)
+declare <vscale x 12 x i64> @llvm.vector.reverse.nxv12i64(<vscale x 12 x i64>)
diff --git a/llvm/test/CodeGen/RISCV/rvv/shuffle-reverse.ll b/llvm/test/CodeGen/RISCV/rvv/shuffle-reverse.ll
index 032d32109933..ab7da9e0faf2 100644
--- a/llvm/test/CodeGen/RISCV/rvv/shuffle-reverse.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/shuffle-reverse.ll
@@ -49,8 +49,8 @@ define <8 x i8> @v4i8_2(<4 x i8> %a, <4 x i8> %b) {
 ; CHECK-NEXT:    vid.v v11
 ; CHECK-NEXT:    vrsub.vi v12, v11, 7
 ; CHECK-NEXT:    vrgather.vv v10, v8, v12
-; CHECK-NEXT:    vmv.v.i v0, 15
 ; CHECK-NEXT:    vrsub.vi v8, v11, 3
+; CHECK-NEXT:    vmv.v.i v0, 15
 ; CHECK-NEXT:    vrgather.vv v10, v9, v8, v0.t
 ; CHECK-NEXT:    vmv1r.v v8, v10
 ; CHECK-NEXT:    ret
@@ -174,8 +174,8 @@ define <8 x i16> @v4i16_2(<4 x i16> %a, <4 x i16> %b) {
 ; CHECK-NEXT:    vid.v v11
 ; CHECK-NEXT:    vrsub.vi v12, v11, 7
 ; CHECK-NEXT:    vrgather.vv v10, v8, v12
-; CHECK-NEXT:    vmv.v.i v0, 15
 ; CHECK-NEXT:    vrsub.vi v8, v11, 3
+; CHECK-NEXT:    vmv.v.i v0, 15
 ; CHECK-NEXT:    vrgather.vv v10, v9, v8, v0.t
 ; CHECK-NEXT:    vmv.v.v v8, v10
 ; CHECK-NEXT:    ret
@@ -492,8 +492,8 @@ define <8 x half> @v4f16_2(<4 x half> %a, <4 x half> %b) {
 ; CHECK-NEXT:    vid.v v11
 ; CHECK-NEXT:    vrsub.vi v12, v11, 7
 ; CHECK-NEXT:    vrgather.vv v10, v8, v12
-; CHECK-NEXT:    vmv.v.i v0, 15
 ; CHECK-NEXT:    vrsub.vi v8, v11, 3
+; CHECK-NEXT:    vmv.v.i v0, 15
 ; CHECK-NEXT:    vrgather.vv v10, v9, v8, v0.t
 ; CHECK-NEXT:    vmv.v.v v8, v10
 ; CHECK-NEXT:    ret
diff --git a/llvm/test/CodeGen/RISCV/rvv/vector-deinterleave-fixed.ll b/llvm/test/CodeGen/RISCV/rvv/vector-deinterleave-fixed.ll
index f3c70ed78c74..d02fe5b205f7 100644
--- a/llvm/test/CodeGen/RISCV/rvv/vector-deinterleave-fixed.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/vector-deinterleave-fixed.ll
@@ -31,7 +31,7 @@ define {<16 x i1>, <16 x i1>} @vector_deinterleave_v16i1_v32i1(<32 x i1> %vec) {
 ; CHECK-NEXT:    vmsne.vi v8, v13, 0
 ; CHECK-NEXT:    vmv.v.v v0, v9
 ; CHECK-NEXT:    ret
-%retval = call {<16 x i1>, <16 x i1>} @llvm.experimental.vector.deinterleave2.v32i1(<32 x i1> %vec)
+%retval = call {<16 x i1>, <16 x i1>} @llvm.vector.deinterleave2.v32i1(<32 x i1> %vec)
 ret {<16 x i1>, <16 x i1>} %retval
 }
 
@@ -44,7 +44,7 @@ define {<16 x i8>, <16 x i8>} @vector_deinterleave_v16i8_v32i8(<32 x i8> %vec) {
 ; CHECK-NEXT:    vmv.v.v v8, v10
 ; CHECK-NEXT:    vmv.v.v v9, v11
 ; CHECK-NEXT:    ret
-%retval = call {<16 x i8>, <16 x i8>} @llvm.experimental.vector.deinterleave2.v32i8(<32 x i8> %vec)
+%retval = call {<16 x i8>, <16 x i8>} @llvm.vector.deinterleave2.v32i8(<32 x i8> %vec)
 ret {<16 x i8>, <16 x i8>} %retval
 }
 
@@ -57,7 +57,7 @@ define {<8 x i16>, <8 x i16>} @vector_deinterleave_v8i16_v16i16(<16 x i16> %vec)
 ; CHECK-NEXT:    vmv.v.v v8, v10
 ; CHECK-NEXT:    vmv.v.v v9, v11
 ; CHECK-NEXT:    ret
-%retval = call {<8 x i16>, <8 x i16>} @llvm.experimental.vector.deinterleave2.v16i16(<16 x i16> %vec)
+%retval = call {<8 x i16>, <8 x i16>} @llvm.vector.deinterleave2.v16i16(<16 x i16> %vec)
 ret {<8 x i16>, <8 x i16>} %retval
 }
 
@@ -71,7 +71,7 @@ define {<4 x i32>, <4 x i32>} @vector_deinterleave_v4i32_vv8i32(<8 x i32> %vec)
 ; CHECK-NEXT:    vmv.v.v v8, v11
 ; CHECK-NEXT:    vmv.v.v v9, v10
 ; CHECK-NEXT:    ret
-%retval = call {<4 x i32>, <4 x i32>} @llvm.experimental.vector.deinterleave2.v8i32(<8 x i32> %vec)
+%retval = call {<4 x i32>, <4 x i32>} @llvm.vector.deinterleave2.v8i32(<8 x i32> %vec)
 ret {<4 x i32>, <4 x i32>} %retval
 }
 
@@ -87,15 +87,15 @@ define {<2 x i64>, <2 x i64>} @vector_deinterleave_v2i64_v4i64(<4 x i64> %vec) {
 ; CHECK-NEXT:    vmerge.vvm v9, v9, v10, v0
 ; CHECK-NEXT:    vslideup.vi v8, v10, 1
 ; CHECK-NEXT:    ret
-%retval = call {<2 x i64>, <2 x i64>} @llvm.experimental.vector.deinterleave2.v4i64(<4 x i64> %vec)
+%retval = call {<2 x i64>, <2 x i64>} @llvm.vector.deinterleave2.v4i64(<4 x i64> %vec)
 ret {<2 x i64>, <2 x i64>} %retval
 }
 
-declare {<16 x i1>, <16 x i1>} @llvm.experimental.vector.deinterleave2.v32i1(<32 x i1>)
-declare {<16 x i8>, <16 x i8>} @llvm.experimental.vector.deinterleave2.v32i8(<32 x i8>)
-declare {<8 x i16>, <8 x i16>} @llvm.experimental.vector.deinterleave2.v16i16(<16 x i16>)
-declare {<4 x i32>, <4 x i32>} @llvm.experimental.vector.deinterleave2.v8i32(<8 x i32>)
-declare {<2 x i64>, <2 x i64>} @llvm.experimental.vector.deinterleave2.v4i64(<4 x i64>)
+declare {<16 x i1>, <16 x i1>} @llvm.vector.deinterleave2.v32i1(<32 x i1>)
+declare {<16 x i8>, <16 x i8>} @llvm.vector.deinterleave2.v32i8(<32 x i8>)
+declare {<8 x i16>, <8 x i16>} @llvm.vector.deinterleave2.v16i16(<16 x i16>)
+declare {<4 x i32>, <4 x i32>} @llvm.vector.deinterleave2.v8i32(<8 x i32>)
+declare {<2 x i64>, <2 x i64>} @llvm.vector.deinterleave2.v4i64(<4 x i64>)
 
 ; Floats
 
@@ -107,7 +107,7 @@ define {<2 x half>, <2 x half>} @vector_deinterleave_v2f16_v4f16(<4 x half> %vec
 ; CHECK-NEXT:    vnsrl.wi v9, v8, 16
 ; CHECK-NEXT:    vmv1r.v v8, v10
 ; CHECK-NEXT:    ret
-%retval = call {<2 x half>, <2 x half>} @llvm.experimental.vector.deinterleave2.v4f16(<4 x half> %vec)
+%retval = call {<2 x half>, <2 x half>} @llvm.vector.deinterleave2.v4f16(<4 x half> %vec)
 ret {<2 x half>, <2 x half>} %retval
 }
 
@@ -119,7 +119,7 @@ define {<4 x half>, <4 x half>} @vector_deinterleave_v4f16_v8f16(<8 x half> %vec
 ; CHECK-NEXT:    vnsrl.wi v9, v8, 16
 ; CHECK-NEXT:    vmv1r.v v8, v10
 ; CHECK-NEXT:    ret
-%retval = call {<4 x half>, <4 x half>} @llvm.experimental.vector.deinterleave2.v8f16(<8 x half> %vec)
+%retval = call {<4 x half>, <4 x half>} @llvm.vector.deinterleave2.v8f16(<8 x half> %vec)
 ret {<4 x half>, <4 x half>} %retval
 }
 
@@ -131,7 +131,7 @@ define {<2 x float>, <2 x float>} @vector_deinterleave_v2f32_v4f32(<4 x float> %
 ; CHECK-NEXT:    vnsrl.wx v9, v8, a0
 ; CHECK-NEXT:    vnsrl.wi v8, v8, 0
 ; CHECK-NEXT:    ret
-%retval = call {<2 x float>, <2 x float>} @llvm.experimental.vector.deinterleave2.v4f32(<4 x float> %vec)
+%retval = call {<2 x float>, <2 x float>} @llvm.vector.deinterleave2.v4f32(<4 x float> %vec)
 ret {<2 x float>, <2 x float>} %retval
 }
 
@@ -144,7 +144,7 @@ define {<8 x half>, <8 x half>} @vector_deinterleave_v8f16_v16f16(<16 x half> %v
 ; CHECK-NEXT:    vmv.v.v v8, v10
 ; CHECK-NEXT:    vmv.v.v v9, v11
 ; CHECK-NEXT:    ret
-%retval = call {<8 x half>, <8 x half>} @llvm.experimental.vector.deinterleave2.v16f16(<16 x half> %vec)
+%retval = call {<8 x half>, <8 x half>} @llvm.vector.deinterleave2.v16f16(<16 x half> %vec)
 ret {<8 x half>, <8 x half>} %retval
 }
 
@@ -158,7 +158,7 @@ define {<4 x float>, <4 x float>} @vector_deinterleave_v4f32_v8f32(<8 x float> %
 ; CHECK-NEXT:    vmv.v.v v8, v11
 ; CHECK-NEXT:    vmv.v.v v9, v10
 ; CHECK-NEXT:    ret
-%retval = call {<4 x float>, <4 x float>} @llvm.experimental.vector.deinterleave2.v8f32(<8 x float> %vec)
+%retval = call {<4 x float>, <4 x float>} @llvm.vector.deinterleave2.v8f32(<8 x float> %vec)
 ret  {<4 x float>, <4 x float>} %retval
 }
 
@@ -174,13 +174,13 @@ define {<2 x double>, <2 x double>} @vector_deinterleave_v2f64_v4f64(<4 x double
 ; CHECK-NEXT:    vmerge.vvm v9, v9, v10, v0
 ; CHECK-NEXT:    vslideup.vi v8, v10, 1
 ; CHECK-NEXT:    ret
-%retval = call {<2 x double>, <2 x double>} @llvm.experimental.vector.deinterleave2.v4f64(<4 x double> %vec)
+%retval = call {<2 x double>, <2 x double>} @llvm.vector.deinterleave2.v4f64(<4 x double> %vec)
 ret {<2 x double>, <2 x double>} %retval
 }
 
-declare {<2 x half>,<2 x half>} @llvm.experimental.vector.deinterleave2.v4f16(<4 x half>)
-declare {<4 x half>, <4 x half>} @llvm.experimental.vector.deinterleave2.v8f16(<8 x half>)
-declare {<2 x float>, <2 x float>} @llvm.experimental.vector.deinterleave2.v4f32(<4 x float>)
-declare {<8 x half>, <8 x half>} @llvm.experimental.vector.deinterleave2.v16f16(<16 x half>)
-declare {<4 x float>, <4 x float>} @llvm.experimental.vector.deinterleave2.v8f32(<8 x float>)
-declare {<2 x double>, <2 x double>} @llvm.experimental.vector.deinterleave2.v4f64(<4 x double>)
+declare {<2 x half>,<2 x half>} @llvm.vector.deinterleave2.v4f16(<4 x half>)
+declare {<4 x half>, <4 x half>} @llvm.vector.deinterleave2.v8f16(<8 x half>)
+declare {<2 x float>, <2 x float>} @llvm.vector.deinterleave2.v4f32(<4 x float>)
+declare {<8 x half>, <8 x half>} @llvm.vector.deinterleave2.v16f16(<16 x half>)
+declare {<4 x float>, <4 x float>} @llvm.vector.deinterleave2.v8f32(<8 x float>)
+declare {<2 x double>, <2 x double>} @llvm.vector.deinterleave2.v4f64(<4 x double>)
diff --git a/llvm/test/CodeGen/RISCV/rvv/vector-deinterleave-load.ll b/llvm/test/CodeGen/RISCV/rvv/vector-deinterleave-load.ll
index 6a712080fda7..8f4ff37fffb0 100644
--- a/llvm/test/CodeGen/RISCV/rvv/vector-deinterleave-load.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/vector-deinterleave-load.ll
@@ -24,7 +24,7 @@ define {<vscale x 16 x i1>, <vscale x 16 x i1>} @vector_deinterleave_load_nxv16i
 ; CHECK-NEXT:    vmsne.vi v9, v10, 0
 ; CHECK-NEXT:    ret
   %vec = load <vscale x 32 x i1>, ptr %p
-  %retval = call {<vscale x 16 x i1>, <vscale x 16 x i1>} @llvm.experimental.vector.deinterleave2.nxv32i1(<vscale x 32 x i1> %vec)
+  %retval = call {<vscale x 16 x i1>, <vscale x 16 x i1>} @llvm.vector.deinterleave2.nxv32i1(<vscale x 32 x i1> %vec)
   ret {<vscale x 16 x i1>, <vscale x 16 x i1>} %retval
 }
 
@@ -35,7 +35,7 @@ define {<vscale x 16 x i8>, <vscale x 16 x i8>} @vector_deinterleave_load_nxv16i
 ; CHECK-NEXT:    vlseg2e8.v v8, (a0)
 ; CHECK-NEXT:    ret
   %vec = load <vscale x 32 x i8>, ptr %p
-  %retval = call {<vscale x 16 x i8>, <vscale x 16 x i8>} @llvm.experimental.vector.deinterleave2.nxv32i8(<vscale x 32 x i8> %vec)
+  %retval = call {<vscale x 16 x i8>, <vscale x 16 x i8>} @llvm.vector.deinterleave2.nxv32i8(<vscale x 32 x i8> %vec)
   ret {<vscale x 16 x i8>, <vscale x 16 x i8>} %retval
 }
 
@@ -49,7 +49,7 @@ define {<vscale x 8 x i16>, <vscale x 8 x i16>} @vector_deinterleave_load_nxv8i1
 ; CHECK-NEXT:    vnsrl.wi v10, v12, 16
 ; CHECK-NEXT:    ret
   %vec = load <vscale x 16 x i16>, ptr %p, align 1
-  %retval = call {<vscale x 8 x i16>, <vscale x 8 x i16>} @llvm.experimental.vector.deinterleave2.nxv16i16(<vscale x 16 x i16> %vec)
+  %retval = call {<vscale x 8 x i16>, <vscale x 8 x i16>} @llvm.vector.deinterleave2.nxv16i16(<vscale x 16 x i16> %vec)
   ret {<vscale x 8 x i16>, <vscale x 8 x i16>} %retval
 }
 
@@ -60,7 +60,7 @@ define {<vscale x 8 x i16>, <vscale x 8 x i16>} @vector_deinterleave_load_nxv8i1
 ; CHECK-NEXT:    vlseg2e16.v v8, (a0)
 ; CHECK-NEXT:    ret
   %vec = load <vscale x 16 x i16>, ptr %p
-  %retval = call {<vscale x 8 x i16>, <vscale x 8 x i16>} @llvm.experimental.vector.deinterleave2.nxv16i16(<vscale x 16 x i16> %vec)
+  %retval = call {<vscale x 8 x i16>, <vscale x 8 x i16>} @llvm.vector.deinterleave2.nxv16i16(<vscale x 16 x i16> %vec)
   ret {<vscale x 8 x i16>, <vscale x 8 x i16>} %retval
 }
 
@@ -71,7 +71,7 @@ define {<vscale x 4 x i32>, <vscale x 4 x i32>} @vector_deinterleave_load_nxv4i3
 ; CHECK-NEXT:    vlseg2e32.v v8, (a0)
 ; CHECK-NEXT:    ret
   %vec = load <vscale x 8 x i32>, ptr %p
-  %retval = call {<vscale x 4 x i32>, <vscale x 4 x i32>} @llvm.experimental.vector.deinterleave2.nxv8i32(<vscale x 8 x i32> %vec)
+  %retval = call {<vscale x 4 x i32>, <vscale x 4 x i32>} @llvm.vector.deinterleave2.nxv8i32(<vscale x 8 x i32> %vec)
   ret {<vscale x 4 x i32>, <vscale x 4 x i32>} %retval
 }
 
@@ -82,7 +82,7 @@ define {<vscale x 2 x i64>, <vscale x 2 x i64>} @vector_deinterleave_load_nxv2i6
 ; CHECK-NEXT:    vlseg2e64.v v8, (a0)
 ; CHECK-NEXT:    ret
   %vec = load <vscale x 4 x i64>, ptr %p
-  %retval = call {<vscale x 2 x i64>, <vscale x 2 x i64>} @llvm.experimental.vector.deinterleave2.nxv4i64(<vscale x 4 x i64> %vec)
+  %retval = call {<vscale x 2 x i64>, <vscale x 2 x i64>} @llvm.vector.deinterleave2.nxv4i64(<vscale x 4 x i64> %vec)
   ret {<vscale x 2 x i64>, <vscale x 2 x i64>} %retval
 }
 
@@ -93,7 +93,7 @@ define {<vscale x 4 x i64>, <vscale x 4 x i64>} @vector_deinterleave_load_nxv4i6
 ; CHECK-NEXT:    vlseg2e64.v v8, (a0)
 ; CHECK-NEXT:    ret
   %vec = load <vscale x 8 x i64>, ptr %p
-  %retval = call {<vscale x 4 x i64>, <vscale x 4 x i64>} @llvm.experimental.vector.deinterleave2.nxv8i64(<vscale x 8 x i64> %vec)
+  %retval = call {<vscale x 4 x i64>, <vscale x 4 x i64>} @llvm.vector.deinterleave2.nxv8i64(<vscale x 8 x i64> %vec)
   ret {<vscale x 4 x i64>, <vscale x 4 x i64>} %retval
 }
 
@@ -171,17 +171,17 @@ define {<vscale x 8 x i64>, <vscale x 8 x i64>} @vector_deinterleave_load_nxv8i6
 ; CHECK-NEXT:    addi sp, sp, 16
 ; CHECK-NEXT:    ret
   %vec = load <vscale x 16 x i64>, ptr %p
-  %retval = call {<vscale x 8 x i64>, <vscale x 8 x i64>} @llvm.experimental.vector.deinterleave2.nxv16i64(<vscale x 16 x i64> %vec)
+  %retval = call {<vscale x 8 x i64>, <vscale x 8 x i64>} @llvm.vector.deinterleave2.nxv16i64(<vscale x 16 x i64> %vec)
   ret {<vscale x 8 x i64>, <vscale x 8 x i64>} %retval
 }
 
-declare {<vscale x 16 x i1>, <vscale x 16 x i1>} @llvm.experimental.vector.deinterleave2.nxv32i1(<vscale x 32 x i1>)
-declare {<vscale x 16 x i8>, <vscale x 16 x i8>} @llvm.experimental.vector.deinterleave2.nxv32i8(<vscale x 32 x i8>)
-declare {<vscale x 8 x i16>, <vscale x 8 x i16>} @llvm.experimental.vector.deinterleave2.nxv16i16(<vscale x 16 x i16>)
-declare {<vscale x 4 x i32>, <vscale x 4 x i32>} @llvm.experimental.vector.deinterleave2.nxv8i32(<vscale x 8 x i32>)
-declare {<vscale x 2 x i64>, <vscale x 2 x i64>} @llvm.experimental.vector.deinterleave2.nxv4i64(<vscale x 4 x i64>)
-declare {<vscale x 4 x i64>, <vscale x 4 x i64>} @llvm.experimental.vector.deinterleave2.nxv8i64(<vscale x 8 x i64>)
-declare {<vscale x 8 x i64>, <vscale x 8 x i64>} @llvm.experimental.vector.deinterleave2.nxv16i64(<vscale x 16 x i64>)
+declare {<vscale x 16 x i1>, <vscale x 16 x i1>} @llvm.vector.deinterleave2.nxv32i1(<vscale x 32 x i1>)
+declare {<vscale x 16 x i8>, <vscale x 16 x i8>} @llvm.vector.deinterleave2.nxv32i8(<vscale x 32 x i8>)
+declare {<vscale x 8 x i16>, <vscale x 8 x i16>} @llvm.vector.deinterleave2.nxv16i16(<vscale x 16 x i16>)
+declare {<vscale x 4 x i32>, <vscale x 4 x i32>} @llvm.vector.deinterleave2.nxv8i32(<vscale x 8 x i32>)
+declare {<vscale x 2 x i64>, <vscale x 2 x i64>} @llvm.vector.deinterleave2.nxv4i64(<vscale x 4 x i64>)
+declare {<vscale x 4 x i64>, <vscale x 4 x i64>} @llvm.vector.deinterleave2.nxv8i64(<vscale x 8 x i64>)
+declare {<vscale x 8 x i64>, <vscale x 8 x i64>} @llvm.vector.deinterleave2.nxv16i64(<vscale x 16 x i64>)
 
 ; Floats
 
@@ -192,7 +192,7 @@ define {<vscale x 2 x half>, <vscale x 2 x half>} @vector_deinterleave_load_nxv2
 ; CHECK-NEXT:    vlseg2e16.v v8, (a0)
 ; CHECK-NEXT:    ret
   %vec = load <vscale x 4 x half>, ptr %p
-  %retval = call {<vscale x 2 x half>, <vscale x 2 x half>} @llvm.experimental.vector.deinterleave2.nxv4f16(<vscale x 4 x half> %vec)
+  %retval = call {<vscale x 2 x half>, <vscale x 2 x half>} @llvm.vector.deinterleave2.nxv4f16(<vscale x 4 x half> %vec)
   ret {<vscale x 2 x half>, <vscale x 2 x half>} %retval
 }
 
@@ -203,7 +203,7 @@ define {<vscale x 4 x half>, <vscale x 4 x half>} @vector_deinterleave_load_nxv4
 ; CHECK-NEXT:    vlseg2e16.v v8, (a0)
 ; CHECK-NEXT:    ret
   %vec = load <vscale x 8 x half>, ptr %p
-  %retval = call {<vscale x 4 x half>, <vscale x 4 x half>} @llvm.experimental.vector.deinterleave2.nxv8f16(<vscale x 8 x half> %vec)
+  %retval = call {<vscale x 4 x half>, <vscale x 4 x half>} @llvm.vector.deinterleave2.nxv8f16(<vscale x 8 x half> %vec)
   ret {<vscale x 4 x half>, <vscale x 4 x half>} %retval
 }
 
@@ -214,7 +214,7 @@ define {<vscale x 2 x float>, <vscale x 2 x float>} @vector_deinterleave_load_nx
 ; CHECK-NEXT:    vlseg2e32.v v8, (a0)
 ; CHECK-NEXT:    ret
   %vec = load <vscale x 4 x float>, ptr %p
-  %retval = call {<vscale x 2 x float>, <vscale x 2 x float>} @llvm.experimental.vector.deinterleave2.nxv4f32(<vscale x 4 x float> %vec)
+  %retval = call {<vscale x 2 x float>, <vscale x 2 x float>} @llvm.vector.deinterleave2.nxv4f32(<vscale x 4 x float> %vec)
   ret {<vscale x 2 x float>, <vscale x 2 x float>} %retval
 }
 
@@ -225,7 +225,7 @@ define {<vscale x 8 x half>, <vscale x 8 x half>} @vector_deinterleave_load_nxv8
 ; CHECK-NEXT:    vlseg2e16.v v8, (a0)
 ; CHECK-NEXT:    ret
   %vec = load <vscale x 16 x half>, ptr %p
-  %retval = call {<vscale x 8 x half>, <vscale x 8 x half>} @llvm.experimental.vector.deinterleave2.nxv16f16(<vscale x 16 x half> %vec)
+  %retval = call {<vscale x 8 x half>, <vscale x 8 x half>} @llvm.vector.deinterleave2.nxv16f16(<vscale x 16 x half> %vec)
   ret {<vscale x 8 x half>, <vscale x 8 x half>} %retval
 }
 
@@ -236,7 +236,7 @@ define {<vscale x 4 x float>, <vscale x 4 x float>} @vector_deinterleave_load_nx
 ; CHECK-NEXT:    vlseg2e32.v v8, (a0)
 ; CHECK-NEXT:    ret
   %vec = load <vscale x 8 x float>, ptr %p
-  %retval = call {<vscale x 4 x float>, <vscale x 4 x float>} @llvm.experimental.vector.deinterleave2.nxv8f32(<vscale x 8 x float> %vec)
+  %retval = call {<vscale x 4 x float>, <vscale x 4 x float>} @llvm.vector.deinterleave2.nxv8f32(<vscale x 8 x float> %vec)
   ret  {<vscale x 4 x float>, <vscale x 4 x float>} %retval
 }
 
@@ -247,13 +247,13 @@ define {<vscale x 2 x double>, <vscale x 2 x double>} @vector_deinterleave_load_
 ; CHECK-NEXT:    vlseg2e64.v v8, (a0)
 ; CHECK-NEXT:    ret
   %vec = load <vscale x 4 x double>, ptr %p
-  %retval = call {<vscale x 2 x double>, <vscale x 2 x double>} @llvm.experimental.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %vec)
+  %retval = call {<vscale x 2 x double>, <vscale x 2 x double>} @llvm.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %vec)
   ret {<vscale x 2 x double>, <vscale x 2 x double>} %retval
 }
 
-declare {<vscale x 2 x half>,<vscale x 2 x half>} @llvm.experimental.vector.deinterleave2.nxv4f16(<vscale x 4 x half>)
-declare {<vscale x 4 x half>, <vscale x 4 x half>} @llvm.experimental.vector.deinterleave2.nxv8f16(<vscale x 8 x half>)
-declare {<vscale x 2 x float>, <vscale x 2 x float>} @llvm.experimental.vector.deinterleave2.nxv4f32(<vscale x 4 x float>)
-declare {<vscale x 8 x half>, <vscale x 8 x half>} @llvm.experimental.vector.deinterleave2.nxv16f16(<vscale x 16 x half>)
-declare {<vscale x 4 x float>, <vscale x 4 x float>} @llvm.experimental.vector.deinterleave2.nxv8f32(<vscale x 8 x float>)
-declare {<vscale x 2 x double>, <vscale x 2 x double>} @llvm.experimental.vector.deinterleave2.nxv4f64(<vscale x 4 x double>)
+declare {<vscale x 2 x half>,<vscale x 2 x half>} @llvm.vector.deinterleave2.nxv4f16(<vscale x 4 x half>)
+declare {<vscale x 4 x half>, <vscale x 4 x half>} @llvm.vector.deinterleave2.nxv8f16(<vscale x 8 x half>)
+declare {<vscale x 2 x float>, <vscale x 2 x float>} @llvm.vector.deinterleave2.nxv4f32(<vscale x 4 x float>)
+declare {<vscale x 8 x half>, <vscale x 8 x half>} @llvm.vector.deinterleave2.nxv16f16(<vscale x 16 x half>)
+declare {<vscale x 4 x float>, <vscale x 4 x float>} @llvm.vector.deinterleave2.nxv8f32(<vscale x 8 x float>)
+declare {<vscale x 2 x double>, <vscale x 2 x double>} @llvm.vector.deinterleave2.nxv4f64(<vscale x 4 x double>)
diff --git a/llvm/test/CodeGen/RISCV/rvv/vector-deinterleave.ll b/llvm/test/CodeGen/RISCV/rvv/vector-deinterleave.ll
index d98597fabcd9..7797577362c9 100644
--- a/llvm/test/CodeGen/RISCV/rvv/vector-deinterleave.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/vector-deinterleave.ll
@@ -21,7 +21,7 @@ define {<vscale x 16 x i1>, <vscale x 16 x i1>} @vector_deinterleave_nxv16i1_nxv
 ; CHECK-NEXT:    vnsrl.wi v10, v12, 8
 ; CHECK-NEXT:    vmsne.vi v9, v10, 0
 ; CHECK-NEXT:    ret
-%retval = call {<vscale x 16 x i1>, <vscale x 16 x i1>} @llvm.experimental.vector.deinterleave2.nxv32i1(<vscale x 32 x i1> %vec)
+%retval = call {<vscale x 16 x i1>, <vscale x 16 x i1>} @llvm.vector.deinterleave2.nxv32i1(<vscale x 32 x i1> %vec)
 ret {<vscale x 16 x i1>, <vscale x 16 x i1>} %retval
 }
 
@@ -34,7 +34,7 @@ define {<vscale x 16 x i8>, <vscale x 16 x i8>} @vector_deinterleave_nxv16i8_nxv
 ; CHECK-NEXT:    vmv.v.v v8, v12
 ; CHECK-NEXT:    vmv.v.v v10, v14
 ; CHECK-NEXT:    ret
-%retval = call {<vscale x 16 x i8>, <vscale x 16 x i8>} @llvm.experimental.vector.deinterleave2.nxv32i8(<vscale x 32 x i8> %vec)
+%retval = call {<vscale x 16 x i8>, <vscale x 16 x i8>} @llvm.vector.deinterleave2.nxv32i8(<vscale x 32 x i8> %vec)
 ret {<vscale x 16 x i8>, <vscale x 16 x i8>} %retval
 }
 
@@ -47,7 +47,7 @@ define {<vscale x 8 x i16>, <vscale x 8 x i16>} @vector_deinterleave_nxv8i16_nxv
 ; CHECK-NEXT:    vmv.v.v v8, v12
 ; CHECK-NEXT:    vmv.v.v v10, v14
 ; CHECK-NEXT:    ret
-%retval = call {<vscale x 8 x i16>, <vscale x 8 x i16>} @llvm.experimental.vector.deinterleave2.nxv16i16(<vscale x 16 x i16> %vec)
+%retval = call {<vscale x 8 x i16>, <vscale x 8 x i16>} @llvm.vector.deinterleave2.nxv16i16(<vscale x 16 x i16> %vec)
 ret {<vscale x 8 x i16>, <vscale x 8 x i16>} %retval
 }
 
@@ -61,7 +61,7 @@ define {<vscale x 4 x i32>, <vscale x 4 x i32>} @vector_deinterleave_nxv4i32_nxv
 ; CHECK-NEXT:    vmv.v.v v8, v14
 ; CHECK-NEXT:    vmv.v.v v10, v12
 ; CHECK-NEXT:    ret
-%retval = call {<vscale x 4 x i32>, <vscale x 4 x i32>} @llvm.experimental.vector.deinterleave2.nxv8i32(<vscale x 8 x i32> %vec)
+%retval = call {<vscale x 4 x i32>, <vscale x 4 x i32>} @llvm.vector.deinterleave2.nxv8i32(<vscale x 8 x i32> %vec)
 ret {<vscale x 4 x i32>, <vscale x 4 x i32>} %retval
 }
 
@@ -77,15 +77,15 @@ define {<vscale x 2 x i64>, <vscale x 2 x i64>} @vector_deinterleave_nxv2i64_nxv
 ; CHECK-NEXT:    vmv2r.v v8, v12
 ; CHECK-NEXT:    vmv2r.v v10, v20
 ; CHECK-NEXT:    ret
-%retval = call {<vscale x 2 x i64>, <vscale x 2 x i64>} @llvm.experimental.vector.deinterleave2.nxv4i64(<vscale x 4 x i64> %vec)
+%retval = call {<vscale x 2 x i64>, <vscale x 2 x i64>} @llvm.vector.deinterleave2.nxv4i64(<vscale x 4 x i64> %vec)
 ret {<vscale x 2 x i64>, <vscale x 2 x i64>} %retval
 }
 
-declare {<vscale x 16 x i1>, <vscale x 16 x i1>} @llvm.experimental.vector.deinterleave2.nxv32i1(<vscale x 32 x i1>)
-declare {<vscale x 16 x i8>, <vscale x 16 x i8>} @llvm.experimental.vector.deinterleave2.nxv32i8(<vscale x 32 x i8>)
-declare {<vscale x 8 x i16>, <vscale x 8 x i16>} @llvm.experimental.vector.deinterleave2.nxv16i16(<vscale x 16 x i16>)
-declare {<vscale x 4 x i32>, <vscale x 4 x i32>} @llvm.experimental.vector.deinterleave2.nxv8i32(<vscale x 8 x i32>)
-declare {<vscale x 2 x i64>, <vscale x 2 x i64>} @llvm.experimental.vector.deinterleave2.nxv4i64(<vscale x 4 x i64>)
+declare {<vscale x 16 x i1>, <vscale x 16 x i1>} @llvm.vector.deinterleave2.nxv32i1(<vscale x 32 x i1>)
+declare {<vscale x 16 x i8>, <vscale x 16 x i8>} @llvm.vector.deinterleave2.nxv32i8(<vscale x 32 x i8>)
+declare {<vscale x 8 x i16>, <vscale x 8 x i16>} @llvm.vector.deinterleave2.nxv16i16(<vscale x 16 x i16>)
+declare {<vscale x 4 x i32>, <vscale x 4 x i32>} @llvm.vector.deinterleave2.nxv8i32(<vscale x 8 x i32>)
+declare {<vscale x 2 x i64>, <vscale x 2 x i64>} @llvm.vector.deinterleave2.nxv4i64(<vscale x 4 x i64>)
 
 define {<vscale x 64 x i1>, <vscale x 64 x i1>} @vector_deinterleave_nxv64i1_nxv128i1(<vscale x 128 x i1> %vec) {
 ; CHECK-LABEL: vector_deinterleave_nxv64i1_nxv128i1:
@@ -110,7 +110,7 @@ define {<vscale x 64 x i1>, <vscale x 64 x i1>} @vector_deinterleave_nxv64i1_nxv
 ; CHECK-NEXT:    vmsne.vi v9, v24, 0
 ; CHECK-NEXT:    vmv1r.v v8, v7
 ; CHECK-NEXT:    ret
-%retval = call {<vscale x 64 x i1>, <vscale x 64 x i1>} @llvm.experimental.vector.deinterleave2.nxv128i1(<vscale x 128 x i1> %vec)
+%retval = call {<vscale x 64 x i1>, <vscale x 64 x i1>} @llvm.vector.deinterleave2.nxv128i1(<vscale x 128 x i1> %vec)
 ret {<vscale x 64 x i1>, <vscale x 64 x i1>} %retval
 }
 
@@ -125,7 +125,7 @@ define {<vscale x 64 x i8>, <vscale x 64 x i8>} @vector_deinterleave_nxv64i8_nxv
 ; CHECK-NEXT:    vnsrl.wi v4, v16, 8
 ; CHECK-NEXT:    vmv8r.v v16, v0
 ; CHECK-NEXT:    ret
-%retval = call {<vscale x 64 x i8>, <vscale x 64 x i8>} @llvm.experimental.vector.deinterleave2.nxv128i8(<vscale x 128 x i8> %vec)
+%retval = call {<vscale x 64 x i8>, <vscale x 64 x i8>} @llvm.vector.deinterleave2.nxv128i8(<vscale x 128 x i8> %vec)
 ret {<vscale x 64 x i8>, <vscale x 64 x i8>} %retval
 }
 
@@ -140,7 +140,7 @@ define {<vscale x 32 x i16>, <vscale x 32 x i16>} @vector_deinterleave_nxv32i16_
 ; CHECK-NEXT:    vnsrl.wi v4, v16, 16
 ; CHECK-NEXT:    vmv8r.v v16, v0
 ; CHECK-NEXT:    ret
-%retval = call {<vscale x 32 x i16>, <vscale x 32 x i16>} @llvm.experimental.vector.deinterleave2.nxv64i16(<vscale x 64 x i16> %vec)
+%retval = call {<vscale x 32 x i16>, <vscale x 32 x i16>} @llvm.vector.deinterleave2.nxv64i16(<vscale x 64 x i16> %vec)
 ret {<vscale x 32 x i16>, <vscale x 32 x i16>} %retval
 }
 
@@ -156,7 +156,7 @@ define {<vscale x 16 x i32>, <vscale x 16 x i32>} @vector_deinterleave_nxv16i32_
 ; CHECK-NEXT:    vnsrl.wi v4, v24, 0
 ; CHECK-NEXT:    vmv8r.v v8, v0
 ; CHECK-NEXT:    ret
-%retval = call {<vscale x 16 x i32>, <vscale x 16 x i32>} @llvm.experimental.vector.deinterleave2.nxv32i32(<vscale x 32 x i32> %vec)
+%retval = call {<vscale x 16 x i32>, <vscale x 16 x i32>} @llvm.vector.deinterleave2.nxv32i32(<vscale x 32 x i32> %vec)
 ret {<vscale x 16 x i32>, <vscale x 16 x i32>} %retval
 }
 
@@ -229,15 +229,15 @@ define {<vscale x 8 x i64>, <vscale x 8 x i64>} @vector_deinterleave_nxv8i64_nxv
 ; CHECK-NEXT:    add sp, sp, a0
 ; CHECK-NEXT:    addi sp, sp, 16
 ; CHECK-NEXT:    ret
-%retval = call {<vscale x 8 x i64>, <vscale x 8 x i64>} @llvm.experimental.vector.deinterleave2.nxv16i64(<vscale x 16 x i64> %vec)
+%retval = call {<vscale x 8 x i64>, <vscale x 8 x i64>} @llvm.vector.deinterleave2.nxv16i64(<vscale x 16 x i64> %vec)
 ret {<vscale x 8 x i64>, <vscale x 8 x i64>} %retval
 }
 
-declare {<vscale x 64 x i1>, <vscale x 64 x i1>} @llvm.experimental.vector.deinterleave2.nxv128i1(<vscale x 128 x i1>)
-declare {<vscale x 64 x i8>, <vscale x 64 x i8>} @llvm.experimental.vector.deinterleave2.nxv128i8(<vscale x 128 x i8>)
-declare {<vscale x 32 x i16>, <vscale x 32 x i16>} @llvm.experimental.vector.deinterleave2.nxv64i16(<vscale x 64 x i16>)
-declare {<vscale x 16 x i32>, <vscale x 16 x i32>} @llvm.experimental.vector.deinterleave2.nxv32i32(<vscale x 32 x i32>)
-declare {<vscale x 8 x i64>, <vscale x 8 x i64>} @llvm.experimental.vector.deinterleave2.nxv16i64(<vscale x 16 x i64>)
+declare {<vscale x 64 x i1>, <vscale x 64 x i1>} @llvm.vector.deinterleave2.nxv128i1(<vscale x 128 x i1>)
+declare {<vscale x 64 x i8>, <vscale x 64 x i8>} @llvm.vector.deinterleave2.nxv128i8(<vscale x 128 x i8>)
+declare {<vscale x 32 x i16>, <vscale x 32 x i16>} @llvm.vector.deinterleave2.nxv64i16(<vscale x 64 x i16>)
+declare {<vscale x 16 x i32>, <vscale x 16 x i32>} @llvm.vector.deinterleave2.nxv32i32(<vscale x 32 x i32>)
+declare {<vscale x 8 x i64>, <vscale x 8 x i64>} @llvm.vector.deinterleave2.nxv16i64(<vscale x 16 x i64>)
 
 ; Floats
 
@@ -249,7 +249,7 @@ define {<vscale x 2 x half>, <vscale x 2 x half>} @vector_deinterleave_nxv2f16_n
 ; CHECK-NEXT:    vnsrl.wi v9, v8, 16
 ; CHECK-NEXT:    vmv1r.v v8, v10
 ; CHECK-NEXT:    ret
-%retval = call {<vscale x 2 x half>, <vscale x 2 x half>} @llvm.experimental.vector.deinterleave2.nxv4f16(<vscale x 4 x half> %vec)
+%retval = call {<vscale x 2 x half>, <vscale x 2 x half>} @llvm.vector.deinterleave2.nxv4f16(<vscale x 4 x half> %vec)
 ret {<vscale x 2 x half>, <vscale x 2 x half>} %retval
 }
 
@@ -262,7 +262,7 @@ define {<vscale x 4 x half>, <vscale x 4 x half>} @vector_deinterleave_nxv4f16_n
 ; CHECK-NEXT:    vmv.v.v v8, v10
 ; CHECK-NEXT:    vmv.v.v v9, v11
 ; CHECK-NEXT:    ret
-%retval = call {<vscale x 4 x half>, <vscale x 4 x half>} @llvm.experimental.vector.deinterleave2.nxv8f16(<vscale x 8 x half> %vec)
+%retval = call {<vscale x 4 x half>, <vscale x 4 x half>} @llvm.vector.deinterleave2.nxv8f16(<vscale x 8 x half> %vec)
 ret {<vscale x 4 x half>, <vscale x 4 x half>} %retval
 }
 
@@ -276,7 +276,7 @@ define {<vscale x 2 x float>, <vscale x 2 x float>} @vector_deinterleave_nxv2f32
 ; CHECK-NEXT:    vmv.v.v v8, v11
 ; CHECK-NEXT:    vmv.v.v v9, v10
 ; CHECK-NEXT:    ret
-%retval = call {<vscale x 2 x float>, <vscale x 2 x float>} @llvm.experimental.vector.deinterleave2.nxv4f32(<vscale x 4 x float> %vec)
+%retval = call {<vscale x 2 x float>, <vscale x 2 x float>} @llvm.vector.deinterleave2.nxv4f32(<vscale x 4 x float> %vec)
 ret {<vscale x 2 x float>, <vscale x 2 x float>} %retval
 }
 
@@ -289,7 +289,7 @@ define {<vscale x 8 x half>, <vscale x 8 x half>} @vector_deinterleave_nxv8f16_n
 ; CHECK-NEXT:    vmv.v.v v8, v12
 ; CHECK-NEXT:    vmv.v.v v10, v14
 ; CHECK-NEXT:    ret
-%retval = call {<vscale x 8 x half>, <vscale x 8 x half>} @llvm.experimental.vector.deinterleave2.nxv16f16(<vscale x 16 x half> %vec)
+%retval = call {<vscale x 8 x half>, <vscale x 8 x half>} @llvm.vector.deinterleave2.nxv16f16(<vscale x 16 x half> %vec)
 ret {<vscale x 8 x half>, <vscale x 8 x half>} %retval
 }
 
@@ -303,7 +303,7 @@ define {<vscale x 4 x float>, <vscale x 4 x float>} @vector_deinterleave_nxv4f32
 ; CHECK-NEXT:    vmv.v.v v8, v14
 ; CHECK-NEXT:    vmv.v.v v10, v12
 ; CHECK-NEXT:    ret
-%retval = call {<vscale x 4 x float>, <vscale x 4 x float>} @llvm.experimental.vector.deinterleave2.nxv8f32(<vscale x 8 x float> %vec)
+%retval = call {<vscale x 4 x float>, <vscale x 4 x float>} @llvm.vector.deinterleave2.nxv8f32(<vscale x 8 x float> %vec)
 ret  {<vscale x 4 x float>, <vscale x 4 x float>} %retval
 }
 
@@ -319,16 +319,16 @@ define {<vscale x 2 x double>, <vscale x 2 x double>} @vector_deinterleave_nxv2f
 ; CHECK-NEXT:    vmv2r.v v8, v12
 ; CHECK-NEXT:    vmv2r.v v10, v20
 ; CHECK-NEXT:    ret
-%retval = call {<vscale x 2 x double>, <vscale x 2 x double>} @llvm.experimental.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %vec)
+%retval = call {<vscale x 2 x double>, <vscale x 2 x double>} @llvm.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %vec)
 ret {<vscale x 2 x double>, <vscale x 2 x double>} %retval
 }
 
-declare {<vscale x 2 x half>,<vscale x 2 x half>} @llvm.experimental.vector.deinterleave2.nxv4f16(<vscale x 4 x half>)
-declare {<vscale x 4 x half>, <vscale x 4 x half>} @llvm.experimental.vector.deinterleave2.nxv8f16(<vscale x 8 x half>)
-declare {<vscale x 2 x float>, <vscale x 2 x float>} @llvm.experimental.vector.deinterleave2.nxv4f32(<vscale x 4 x float>)
-declare {<vscale x 8 x half>, <vscale x 8 x half>} @llvm.experimental.vector.deinterleave2.nxv16f16(<vscale x 16 x half>)
-declare {<vscale x 4 x float>, <vscale x 4 x float>} @llvm.experimental.vector.deinterleave2.nxv8f32(<vscale x 8 x float>)
-declare {<vscale x 2 x double>, <vscale x 2 x double>} @llvm.experimental.vector.deinterleave2.nxv4f64(<vscale x 4 x double>)
+declare {<vscale x 2 x half>,<vscale x 2 x half>} @llvm.vector.deinterleave2.nxv4f16(<vscale x 4 x half>)
+declare {<vscale x 4 x half>, <vscale x 4 x half>} @llvm.vector.deinterleave2.nxv8f16(<vscale x 8 x half>)
+declare {<vscale x 2 x float>, <vscale x 2 x float>} @llvm.vector.deinterleave2.nxv4f32(<vscale x 4 x float>)
+declare {<vscale x 8 x half>, <vscale x 8 x half>} @llvm.vector.deinterleave2.nxv16f16(<vscale x 16 x half>)
+declare {<vscale x 4 x float>, <vscale x 4 x float>} @llvm.vector.deinterleave2.nxv8f32(<vscale x 8 x float>)
+declare {<vscale x 2 x double>, <vscale x 2 x double>} @llvm.vector.deinterleave2.nxv4f64(<vscale x 4 x double>)
 
 define {<vscale x 32 x half>, <vscale x 32 x half>} @vector_deinterleave_nxv32f16_nxv64f16(<vscale x 64 x half> %vec) {
 ; CHECK-LABEL: vector_deinterleave_nxv32f16_nxv64f16:
@@ -341,7 +341,7 @@ define {<vscale x 32 x half>, <vscale x 32 x half>} @vector_deinterleave_nxv32f1
 ; CHECK-NEXT:    vnsrl.wi v4, v16, 16
 ; CHECK-NEXT:    vmv8r.v v16, v0
 ; CHECK-NEXT:    ret
-%retval = call {<vscale x 32 x half>, <vscale x 32 x half>} @llvm.experimental.vector.deinterleave2.nxv64f16(<vscale x 64 x half> %vec)
+%retval = call {<vscale x 32 x half>, <vscale x 32 x half>} @llvm.vector.deinterleave2.nxv64f16(<vscale x 64 x half> %vec)
 ret {<vscale x 32 x half>, <vscale x 32 x half>} %retval
 }
 
@@ -357,7 +357,7 @@ define {<vscale x 16 x float>, <vscale x 16 x float>} @vector_deinterleave_nxv16
 ; CHECK-NEXT:    vnsrl.wi v4, v24, 0
 ; CHECK-NEXT:    vmv8r.v v8, v0
 ; CHECK-NEXT:    ret
-%retval = call {<vscale x 16 x float>, <vscale x 16 x float>} @llvm.experimental.vector.deinterleave2.nxv32f32(<vscale x 32 x float> %vec)
+%retval = call {<vscale x 16 x float>, <vscale x 16 x float>} @llvm.vector.deinterleave2.nxv32f32(<vscale x 32 x float> %vec)
 ret  {<vscale x 16 x float>, <vscale x 16 x float>} %retval
 }
 
@@ -430,10 +430,10 @@ define {<vscale x 8 x double>, <vscale x 8 x double>} @vector_deinterleave_nxv8f
 ; CHECK-NEXT:    add sp, sp, a0
 ; CHECK-NEXT:    addi sp, sp, 16
 ; CHECK-NEXT:    ret
-%retval = call {<vscale x 8 x double>, <vscale x 8 x double>} @llvm.experimental.vector.deinterleave2.nxv16f64(<vscale x 16 x double> %vec)
+%retval = call {<vscale x 8 x double>, <vscale x 8 x double>} @llvm.vector.deinterleave2.nxv16f64(<vscale x 16 x double> %vec)
 ret {<vscale x 8 x double>, <vscale x 8 x double>} %retval
 }
 
-declare {<vscale x 32 x half>, <vscale x 32 x half>} @llvm.experimental.vector.deinterleave2.nxv64f16(<vscale x 64 x half>)
-declare {<vscale x 16 x float>, <vscale x 16 x float>} @llvm.experimental.vector.deinterleave2.nxv32f32(<vscale x 32 x float>)
-declare {<vscale x 8 x double>, <vscale x 8 x double>} @llvm.experimental.vector.deinterleave2.nxv16f64(<vscale x 16 x double>)
+declare {<vscale x 32 x half>, <vscale x 32 x half>} @llvm.vector.deinterleave2.nxv64f16(<vscale x 64 x half>)
+declare {<vscale x 16 x float>, <vscale x 16 x float>} @llvm.vector.deinterleave2.nxv32f32(<vscale x 32 x float>)
+declare {<vscale x 8 x double>, <vscale x 8 x double>} @llvm.vector.deinterleave2.nxv16f64(<vscale x 16 x double>)
diff --git a/llvm/test/CodeGen/RISCV/rvv/vector-interleave-fixed.ll b/llvm/test/CodeGen/RISCV/rvv/vector-interleave-fixed.ll
index 6ebe8e095469..99872c199a1e 100644
--- a/llvm/test/CodeGen/RISCV/rvv/vector-interleave-fixed.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/vector-interleave-fixed.ll
@@ -41,7 +41,7 @@ define <32 x i1> @vector_interleave_v32i1_v16i1(<16 x i1> %a, <16 x i1> %b) {
 ; ZVBB-NEXT:    vsetvli zero, a0, e8, m2, ta, ma
 ; ZVBB-NEXT:    vmsne.vi v0, v12, 0
 ; ZVBB-NEXT:    ret
-	   %res = call <32 x i1> @llvm.experimental.vector.interleave2.v32i1(<16 x i1> %a, <16 x i1> %b)
+	   %res = call <32 x i1> @llvm.vector.interleave2.v32i1(<16 x i1> %a, <16 x i1> %b)
 	   ret <32 x i1> %res
 }
 
@@ -62,7 +62,7 @@ define <16 x i16> @vector_interleave_v16i16_v8i16(<8 x i16> %a, <8 x i16> %b) {
 ; ZVBB-NEXT:    vwaddu.wv v10, v10, v8
 ; ZVBB-NEXT:    vmv2r.v v8, v10
 ; ZVBB-NEXT:    ret
-	   %res = call <16 x i16> @llvm.experimental.vector.interleave2.v16i16(<8 x i16> %a, <8 x i16> %b)
+	   %res = call <16 x i16> @llvm.vector.interleave2.v16i16(<8 x i16> %a, <8 x i16> %b)
 	   ret <16 x i16> %res
 }
 
@@ -84,7 +84,7 @@ define <8 x i32> @vector_interleave_v8i32_v4i32(<4 x i32> %a, <4 x i32> %b) {
 ; ZVBB-NEXT:    vwaddu.wv v10, v10, v8
 ; ZVBB-NEXT:    vmv2r.v v8, v10
 ; ZVBB-NEXT:    ret
-	   %res = call <8 x i32> @llvm.experimental.vector.interleave2.v8i32(<4 x i32> %a, <4 x i32> %b)
+	   %res = call <8 x i32> @llvm.vector.interleave2.v8i32(<4 x i32> %a, <4 x i32> %b)
 	   ret <8 x i32> %res
 }
 
@@ -118,14 +118,14 @@ define <4 x i64> @vector_interleave_v4i64_v2i64(<2 x i64> %a, <2 x i64> %b) {
 ; ZVBB-NEXT:    vrgatherei16.vv v10, v8, v12
 ; ZVBB-NEXT:    vmv.v.v v8, v10
 ; ZVBB-NEXT:    ret
-	   %res = call <4 x i64> @llvm.experimental.vector.interleave2.v4i64(<2 x i64> %a, <2 x i64> %b)
+	   %res = call <4 x i64> @llvm.vector.interleave2.v4i64(<2 x i64> %a, <2 x i64> %b)
 	   ret <4 x i64> %res
 }
 
-declare <32 x i1> @llvm.experimental.vector.interleave2.v32i1(<16 x i1>, <16 x i1>)
-declare <16 x i16> @llvm.experimental.vector.interleave2.v16i16(<8 x i16>, <8 x i16>)
-declare <8 x i32> @llvm.experimental.vector.interleave2.v8i32(<4 x i32>, <4 x i32>)
-declare <4 x i64> @llvm.experimental.vector.interleave2.v4i64(<2 x i64>, <2 x i64>)
+declare <32 x i1> @llvm.vector.interleave2.v32i1(<16 x i1>, <16 x i1>)
+declare <16 x i16> @llvm.vector.interleave2.v16i16(<8 x i16>, <8 x i16>)
+declare <8 x i32> @llvm.vector.interleave2.v8i32(<4 x i32>, <4 x i32>)
+declare <4 x i64> @llvm.vector.interleave2.v4i64(<2 x i64>, <2 x i64>)
 
 ; Floats
 
@@ -146,7 +146,7 @@ define <4 x half> @vector_interleave_v4f16_v2f16(<2 x half> %a, <2 x half> %b) {
 ; ZVBB-NEXT:    vwaddu.wv v10, v10, v8
 ; ZVBB-NEXT:    vmv1r.v v8, v10
 ; ZVBB-NEXT:    ret
-	   %res = call <4 x half> @llvm.experimental.vector.interleave2.v4f16(<2 x half> %a, <2 x half> %b)
+	   %res = call <4 x half> @llvm.vector.interleave2.v4f16(<2 x half> %a, <2 x half> %b)
 	   ret <4 x half> %res
 }
 
@@ -167,7 +167,7 @@ define <8 x half> @vector_interleave_v8f16_v4f16(<4 x half> %a, <4 x half> %b) {
 ; ZVBB-NEXT:    vwaddu.wv v10, v10, v8
 ; ZVBB-NEXT:    vmv1r.v v8, v10
 ; ZVBB-NEXT:    ret
-	   %res = call <8 x half> @llvm.experimental.vector.interleave2.v8f16(<4 x half> %a, <4 x half> %b)
+	   %res = call <8 x half> @llvm.vector.interleave2.v8f16(<4 x half> %a, <4 x half> %b)
 	   ret <8 x half> %res
 }
 
@@ -189,7 +189,7 @@ define <4 x float> @vector_interleave_v4f32_v2f32(<2 x float> %a, <2 x float> %b
 ; ZVBB-NEXT:    vwaddu.wv v10, v10, v8
 ; ZVBB-NEXT:    vmv1r.v v8, v10
 ; ZVBB-NEXT:    ret
-	   %res = call <4 x float> @llvm.experimental.vector.interleave2.v4f32(<2 x float> %a, <2 x float> %b)
+	   %res = call <4 x float> @llvm.vector.interleave2.v4f32(<2 x float> %a, <2 x float> %b)
 	   ret <4 x float> %res
 }
 
@@ -210,7 +210,7 @@ define <16 x half> @vector_interleave_v16f16_v8f16(<8 x half> %a, <8 x half> %b)
 ; ZVBB-NEXT:    vwaddu.wv v10, v10, v8
 ; ZVBB-NEXT:    vmv2r.v v8, v10
 ; ZVBB-NEXT:    ret
-	   %res = call <16 x half> @llvm.experimental.vector.interleave2.v16f16(<8 x half> %a, <8 x half> %b)
+	   %res = call <16 x half> @llvm.vector.interleave2.v16f16(<8 x half> %a, <8 x half> %b)
 	   ret <16 x half> %res
 }
 
@@ -232,7 +232,7 @@ define <8 x float> @vector_interleave_v8f32_v4f32(<4 x float> %a, <4 x float> %b
 ; ZVBB-NEXT:    vwaddu.wv v10, v10, v8
 ; ZVBB-NEXT:    vmv2r.v v8, v10
 ; ZVBB-NEXT:    ret
-	   %res = call <8 x float> @llvm.experimental.vector.interleave2.v8f32(<4 x float> %a, <4 x float> %b)
+	   %res = call <8 x float> @llvm.vector.interleave2.v8f32(<4 x float> %a, <4 x float> %b)
 	   ret <8 x float> %res
 }
 
@@ -266,17 +266,17 @@ define <4 x double> @vector_interleave_v4f64_v2f64(<2 x double> %a, <2 x double>
 ; ZVBB-NEXT:    vrgatherei16.vv v10, v8, v12
 ; ZVBB-NEXT:    vmv.v.v v8, v10
 ; ZVBB-NEXT:    ret
-	   %res = call <4 x double> @llvm.experimental.vector.interleave2.v4f64(<2 x double> %a, <2 x double> %b)
+	   %res = call <4 x double> @llvm.vector.interleave2.v4f64(<2 x double> %a, <2 x double> %b)
 	   ret <4 x double> %res
 }
 
 
-declare <4 x half> @llvm.experimental.vector.interleave2.v4f16(<2 x half>, <2 x half>)
-declare <8 x half> @llvm.experimental.vector.interleave2.v8f16(<4 x half>, <4 x half>)
-declare <4 x float> @llvm.experimental.vector.interleave2.v4f32(<2 x float>, <2 x float>)
-declare <16 x half> @llvm.experimental.vector.interleave2.v16f16(<8 x half>, <8 x half>)
-declare <8 x float> @llvm.experimental.vector.interleave2.v8f32(<4 x float>, <4 x float>)
-declare <4 x double> @llvm.experimental.vector.interleave2.v4f64(<2 x double>, <2 x double>)
+declare <4 x half> @llvm.vector.interleave2.v4f16(<2 x half>, <2 x half>)
+declare <8 x half> @llvm.vector.interleave2.v8f16(<4 x half>, <4 x half>)
+declare <4 x float> @llvm.vector.interleave2.v4f32(<2 x float>, <2 x float>)
+declare <16 x half> @llvm.vector.interleave2.v16f16(<8 x half>, <8 x half>)
+declare <8 x float> @llvm.vector.interleave2.v8f32(<4 x float>, <4 x float>)
+declare <4 x double> @llvm.vector.interleave2.v4f64(<2 x double>, <2 x double>)
 ;; NOTE: These prefixes are unused and the list is autogenerated. Do not add tests below this line:
 ; RV32: {{.*}}
 ; RV64: {{.*}}
diff --git a/llvm/test/CodeGen/RISCV/rvv/vector-interleave-store.ll b/llvm/test/CodeGen/RISCV/rvv/vector-interleave-store.ll
index 9a5e86d61c26..7ade47e60bc6 100644
--- a/llvm/test/CodeGen/RISCV/rvv/vector-interleave-store.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/vector-interleave-store.ll
@@ -27,7 +27,7 @@ define void @vector_interleave_store_nxv32i1_nxv16i1(<vscale x 16 x i1> %a, <vsc
 ; CHECK-NEXT:    vsetvli a1, zero, e8, m4, ta, ma
 ; CHECK-NEXT:    vsm.v v9, (a0)
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 32 x i1> @llvm.experimental.vector.interleave2.nxv32i1(<vscale x 16 x i1> %a, <vscale x 16 x i1> %b)
+  %res = call <vscale x 32 x i1> @llvm.vector.interleave2.nxv32i1(<vscale x 16 x i1> %a, <vscale x 16 x i1> %b)
   store <vscale x 32 x i1> %res, ptr %p
   ret void
 }
@@ -42,7 +42,7 @@ define void @vector_interleave_store_nxv16i16_nxv8i16_align1(<vscale x 8 x i16>
 ; CHECK-NEXT:    vwmaccu.vx v12, a1, v10
 ; CHECK-NEXT:    vs4r.v v12, (a0)
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 16 x i16> @llvm.experimental.vector.interleave2.nxv16i16(<vscale x 8 x i16> %a, <vscale x 8 x i16> %b)
+  %res = call <vscale x 16 x i16> @llvm.vector.interleave2.nxv16i16(<vscale x 8 x i16> %a, <vscale x 8 x i16> %b)
   store <vscale x 16 x i16> %res, ptr %p, align 1
   ret void
 }
@@ -53,7 +53,7 @@ define void @vector_interleave_store_nxv16i16_nxv8i16(<vscale x 8 x i16> %a, <vs
 ; CHECK-NEXT:    vsetvli a1, zero, e16, m2, ta, ma
 ; CHECK-NEXT:    vsseg2e16.v v8, (a0)
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 16 x i16> @llvm.experimental.vector.interleave2.nxv16i16(<vscale x 8 x i16> %a, <vscale x 8 x i16> %b)
+  %res = call <vscale x 16 x i16> @llvm.vector.interleave2.nxv16i16(<vscale x 8 x i16> %a, <vscale x 8 x i16> %b)
   store <vscale x 16 x i16> %res, ptr %p
   ret void
 }
@@ -64,7 +64,7 @@ define void @vector_interleave_store_nxv8i32_nxv4i32(<vscale x 4 x i32> %a, <vsc
 ; CHECK-NEXT:    vsetvli a1, zero, e32, m2, ta, ma
 ; CHECK-NEXT:    vsseg2e32.v v8, (a0)
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 8 x i32> @llvm.experimental.vector.interleave2.nxv8i32(<vscale x 4 x i32> %a, <vscale x 4 x i32> %b)
+  %res = call <vscale x 8 x i32> @llvm.vector.interleave2.nxv8i32(<vscale x 4 x i32> %a, <vscale x 4 x i32> %b)
   store <vscale x 8 x i32> %res, ptr %p
   ret void
 }
@@ -75,7 +75,7 @@ define void @vector_interleave_store_nxv4i64_nxv2i64(<vscale x 2 x i64> %a, <vsc
 ; CHECK-NEXT:    vsetvli a1, zero, e64, m2, ta, ma
 ; CHECK-NEXT:    vsseg2e64.v v8, (a0)
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 4 x i64> @llvm.experimental.vector.interleave2.nxv4i64(<vscale x 2 x i64> %a, <vscale x 2 x i64> %b)
+  %res = call <vscale x 4 x i64> @llvm.vector.interleave2.nxv4i64(<vscale x 2 x i64> %a, <vscale x 2 x i64> %b)
   store <vscale x 4 x i64> %res, ptr %p
   ret void
 }
@@ -86,7 +86,7 @@ define void @vector_interleave_store_nxv8i64_nxv4i64(<vscale x 4 x i64> %a, <vsc
 ; CHECK-NEXT:    vsetvli a1, zero, e64, m4, ta, ma
 ; CHECK-NEXT:    vsseg2e64.v v8, (a0)
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 8 x i64> @llvm.experimental.vector.interleave2.nxv8i64(<vscale x 4 x i64> %a, <vscale x 4 x i64> %b)
+  %res = call <vscale x 8 x i64> @llvm.vector.interleave2.nxv8i64(<vscale x 4 x i64> %a, <vscale x 4 x i64> %b)
   store <vscale x 8 x i64> %res, ptr %p
   ret void
 }
@@ -107,14 +107,14 @@ define void @vector_interleave_store_nxv16i64_nxv8i64(<vscale x 8 x i64> %a, <vs
 ; CHECK-NEXT:    srli a2, a1, 1
 ; CHECK-NEXT:    vsetvli a3, zero, e16, m2, ta, mu
 ; CHECK-NEXT:    vid.v v24
-; CHECK-NEXT:    vand.vi v26, v24, 1
-; CHECK-NEXT:    vmsne.vi v28, v26, 0
-; CHECK-NEXT:    vsrl.vi v24, v24, 1
+; CHECK-NEXT:    vsrl.vi v26, v24, 1
+; CHECK-NEXT:    vand.vi v24, v24, 1
+; CHECK-NEXT:    vmsne.vi v28, v24, 0
 ; CHECK-NEXT:    vmv1r.v v0, v28
-; CHECK-NEXT:    vadd.vx v24, v24, a2, v0.t
+; CHECK-NEXT:    vadd.vx v26, v26, a2, v0.t
 ; CHECK-NEXT:    vmv4r.v v12, v16
 ; CHECK-NEXT:    vsetvli zero, zero, e64, m8, ta, ma
-; CHECK-NEXT:    vrgatherei16.vv v0, v8, v24
+; CHECK-NEXT:    vrgatherei16.vv v0, v8, v26
 ; CHECK-NEXT:    csrr a2, vlenb
 ; CHECK-NEXT:    slli a2, a2, 3
 ; CHECK-NEXT:    add a2, sp, a2
@@ -123,7 +123,7 @@ define void @vector_interleave_store_nxv16i64_nxv8i64(<vscale x 8 x i64> %a, <vs
 ; CHECK-NEXT:    addi a2, sp, 16
 ; CHECK-NEXT:    vl8r.v v8, (a2) # Unknown-size Folded Reload
 ; CHECK-NEXT:    vmv4r.v v16, v12
-; CHECK-NEXT:    vrgatherei16.vv v8, v16, v24
+; CHECK-NEXT:    vrgatherei16.vv v8, v16, v26
 ; CHECK-NEXT:    slli a1, a1, 3
 ; CHECK-NEXT:    add a1, a0, a1
 ; CHECK-NEXT:    vs8r.v v8, (a1)
@@ -138,17 +138,17 @@ define void @vector_interleave_store_nxv16i64_nxv8i64(<vscale x 8 x i64> %a, <vs
 ; CHECK-NEXT:    add sp, sp, a0
 ; CHECK-NEXT:    addi sp, sp, 16
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 16 x i64> @llvm.experimental.vector.interleave2.nxv16i64(<vscale x 8 x i64> %a, <vscale x 8 x i64> %b)
+  %res = call <vscale x 16 x i64> @llvm.vector.interleave2.nxv16i64(<vscale x 8 x i64> %a, <vscale x 8 x i64> %b)
   store <vscale x 16 x i64> %res, ptr %p
   ret void
 }
 
-declare <vscale x 32 x i1> @llvm.experimental.vector.interleave2.nxv32i1(<vscale x 16 x i1>, <vscale x 16 x i1>)
-declare <vscale x 16 x i16> @llvm.experimental.vector.interleave2.nxv16i16(<vscale x 8 x i16>, <vscale x 8 x i16>)
-declare <vscale x 8 x i32> @llvm.experimental.vector.interleave2.nxv8i32(<vscale x 4 x i32>, <vscale x 4 x i32>)
-declare <vscale x 4 x i64> @llvm.experimental.vector.interleave2.nxv4i64(<vscale x 2 x i64>, <vscale x 2 x i64>)
-declare <vscale x 8 x i64> @llvm.experimental.vector.interleave2.nxv8i64(<vscale x 4 x i64>, <vscale x 4 x i64>)
-declare <vscale x 16 x i64> @llvm.experimental.vector.interleave2.nxv16i64(<vscale x 8 x i64>, <vscale x 8 x i64>)
+declare <vscale x 32 x i1> @llvm.vector.interleave2.nxv32i1(<vscale x 16 x i1>, <vscale x 16 x i1>)
+declare <vscale x 16 x i16> @llvm.vector.interleave2.nxv16i16(<vscale x 8 x i16>, <vscale x 8 x i16>)
+declare <vscale x 8 x i32> @llvm.vector.interleave2.nxv8i32(<vscale x 4 x i32>, <vscale x 4 x i32>)
+declare <vscale x 4 x i64> @llvm.vector.interleave2.nxv4i64(<vscale x 2 x i64>, <vscale x 2 x i64>)
+declare <vscale x 8 x i64> @llvm.vector.interleave2.nxv8i64(<vscale x 4 x i64>, <vscale x 4 x i64>)
+declare <vscale x 16 x i64> @llvm.vector.interleave2.nxv16i64(<vscale x 8 x i64>, <vscale x 8 x i64>)
 
 ; Floats
 
@@ -158,7 +158,7 @@ define void @vector_interleave_store_nxv4f16_nxv2f16(<vscale x 2 x half> %a, <vs
 ; CHECK-NEXT:    vsetvli a1, zero, e16, mf2, ta, ma
 ; CHECK-NEXT:    vsseg2e16.v v8, (a0)
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 4 x half> @llvm.experimental.vector.interleave2.nxv4f16(<vscale x 2 x half> %a, <vscale x 2 x half> %b)
+  %res = call <vscale x 4 x half> @llvm.vector.interleave2.nxv4f16(<vscale x 2 x half> %a, <vscale x 2 x half> %b)
   store <vscale x 4 x half> %res, ptr %p
   ret void
 }
@@ -169,7 +169,7 @@ define void @vector_interleave_store_nxv8f16_nxv4f16(<vscale x 4 x half> %a, <vs
 ; CHECK-NEXT:    vsetvli a1, zero, e16, m1, ta, ma
 ; CHECK-NEXT:    vsseg2e16.v v8, (a0)
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 8 x half> @llvm.experimental.vector.interleave2.nxv8f16(<vscale x 4 x half> %a, <vscale x 4 x half> %b)
+  %res = call <vscale x 8 x half> @llvm.vector.interleave2.nxv8f16(<vscale x 4 x half> %a, <vscale x 4 x half> %b)
   store <vscale x 8 x half> %res, ptr %p
   ret void
 }
@@ -180,7 +180,7 @@ define void @vector_interleave_store_nxv4f32_nxv2f32(<vscale x 2 x float> %a, <v
 ; CHECK-NEXT:    vsetvli a1, zero, e32, m1, ta, ma
 ; CHECK-NEXT:    vsseg2e32.v v8, (a0)
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 4 x float> @llvm.experimental.vector.interleave2.nxv4f32(<vscale x 2 x float> %a, <vscale x 2 x float> %b)
+  %res = call <vscale x 4 x float> @llvm.vector.interleave2.nxv4f32(<vscale x 2 x float> %a, <vscale x 2 x float> %b)
   store <vscale x 4 x float> %res, ptr %p
   ret void
 }
@@ -191,7 +191,7 @@ define void @vector_interleave_store_nxv16f16_nxv8f16(<vscale x 8 x half> %a, <v
 ; CHECK-NEXT:    vsetvli a1, zero, e16, m2, ta, ma
 ; CHECK-NEXT:    vsseg2e16.v v8, (a0)
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 16 x half> @llvm.experimental.vector.interleave2.nxv16f16(<vscale x 8 x half> %a, <vscale x 8 x half> %b)
+  %res = call <vscale x 16 x half> @llvm.vector.interleave2.nxv16f16(<vscale x 8 x half> %a, <vscale x 8 x half> %b)
   store <vscale x 16 x half> %res, ptr %p
   ret void
 }
@@ -202,7 +202,7 @@ define void @vector_interleave_store_nxv8f32_nxv4f32(<vscale x 4 x float> %a, <v
 ; CHECK-NEXT:    vsetvli a1, zero, e32, m2, ta, ma
 ; CHECK-NEXT:    vsseg2e32.v v8, (a0)
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 8 x float> @llvm.experimental.vector.interleave2.nxv8f32(<vscale x 4 x float> %a, <vscale x 4 x float> %b)
+  %res = call <vscale x 8 x float> @llvm.vector.interleave2.nxv8f32(<vscale x 4 x float> %a, <vscale x 4 x float> %b)
   store <vscale x 8 x float> %res, ptr %p
   ret void
 }
@@ -213,15 +213,15 @@ define void @vector_interleave_store_nxv4f64_nxv2f64(<vscale x 2 x double> %a, <
 ; CHECK-NEXT:    vsetvli a1, zero, e64, m2, ta, ma
 ; CHECK-NEXT:    vsseg2e64.v v8, (a0)
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 4 x double> @llvm.experimental.vector.interleave2.nxv4f64(<vscale x 2 x double> %a, <vscale x 2 x double> %b)
+  %res = call <vscale x 4 x double> @llvm.vector.interleave2.nxv4f64(<vscale x 2 x double> %a, <vscale x 2 x double> %b)
   store <vscale x 4 x double> %res, ptr %p
   ret void
 }
 
 
-declare <vscale x 4 x half> @llvm.experimental.vector.interleave2.nxv4f16(<vscale x 2 x half>, <vscale x 2 x half>)
-declare <vscale x 8 x half> @llvm.experimental.vector.interleave2.nxv8f16(<vscale x 4 x half>, <vscale x 4 x half>)
-declare <vscale x 4 x float> @llvm.experimental.vector.interleave2.nxv4f32(<vscale x 2 x float>, <vscale x 2 x float>)
-declare <vscale x 16 x half> @llvm.experimental.vector.interleave2.nxv16f16(<vscale x 8 x half>, <vscale x 8 x half>)
-declare <vscale x 8 x float> @llvm.experimental.vector.interleave2.nxv8f32(<vscale x 4 x float>, <vscale x 4 x float>)
-declare <vscale x 4 x double> @llvm.experimental.vector.interleave2.nxv4f64(<vscale x 2 x double>, <vscale x 2 x double>)
+declare <vscale x 4 x half> @llvm.vector.interleave2.nxv4f16(<vscale x 2 x half>, <vscale x 2 x half>)
+declare <vscale x 8 x half> @llvm.vector.interleave2.nxv8f16(<vscale x 4 x half>, <vscale x 4 x half>)
+declare <vscale x 4 x float> @llvm.vector.interleave2.nxv4f32(<vscale x 2 x float>, <vscale x 2 x float>)
+declare <vscale x 16 x half> @llvm.vector.interleave2.nxv16f16(<vscale x 8 x half>, <vscale x 8 x half>)
+declare <vscale x 8 x float> @llvm.vector.interleave2.nxv8f32(<vscale x 4 x float>, <vscale x 4 x float>)
+declare <vscale x 4 x double> @llvm.vector.interleave2.nxv4f64(<vscale x 2 x double>, <vscale x 2 x double>)
diff --git a/llvm/test/CodeGen/RISCV/rvv/vector-interleave.ll b/llvm/test/CodeGen/RISCV/rvv/vector-interleave.ll
index 4b6ad0f27214..a7e0ad6ee5f4 100644
--- a/llvm/test/CodeGen/RISCV/rvv/vector-interleave.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/vector-interleave.ll
@@ -47,7 +47,7 @@ define <vscale x 32 x i1> @vector_interleave_nxv32i1_nxv16i1(<vscale x 16 x i1>
 ; ZVBB-NEXT:    vsetvli zero, a1, e8, mf2, ta, ma
 ; ZVBB-NEXT:    vslideup.vx v0, v8, a0
 ; ZVBB-NEXT:    ret
-  %res = call <vscale x 32 x i1> @llvm.experimental.vector.interleave2.nxv32i1(<vscale x 16 x i1> %a, <vscale x 16 x i1> %b)
+  %res = call <vscale x 32 x i1> @llvm.vector.interleave2.nxv32i1(<vscale x 16 x i1> %a, <vscale x 16 x i1> %b)
   ret <vscale x 32 x i1> %res
 }
 
@@ -68,7 +68,7 @@ define <vscale x 32 x i8> @vector_interleave_nxv32i8_nxv16i8(<vscale x 16 x i8>
 ; ZVBB-NEXT:    vwaddu.wv v12, v12, v8
 ; ZVBB-NEXT:    vmv4r.v v8, v12
 ; ZVBB-NEXT:    ret
-  %res = call <vscale x 32 x i8> @llvm.experimental.vector.interleave2.nxv32i8(<vscale x 16 x i8> %a, <vscale x 16 x i8> %b)
+  %res = call <vscale x 32 x i8> @llvm.vector.interleave2.nxv32i8(<vscale x 16 x i8> %a, <vscale x 16 x i8> %b)
   ret <vscale x 32 x i8> %res
 }
 
@@ -89,7 +89,7 @@ define <vscale x 16 x i16> @vector_interleave_nxv16i16_nxv8i16(<vscale x 8 x i16
 ; ZVBB-NEXT:    vwaddu.wv v12, v12, v8
 ; ZVBB-NEXT:    vmv4r.v v8, v12
 ; ZVBB-NEXT:    ret
-  %res = call <vscale x 16 x i16> @llvm.experimental.vector.interleave2.nxv16i16(<vscale x 8 x i16> %a, <vscale x 8 x i16> %b)
+  %res = call <vscale x 16 x i16> @llvm.vector.interleave2.nxv16i16(<vscale x 8 x i16> %a, <vscale x 8 x i16> %b)
   ret <vscale x 16 x i16> %res
 }
 
@@ -111,7 +111,7 @@ define <vscale x 8 x i32> @vector_interleave_nxv8i32_nxv4i32(<vscale x 4 x i32>
 ; ZVBB-NEXT:    vwaddu.wv v12, v12, v8
 ; ZVBB-NEXT:    vmv4r.v v8, v12
 ; ZVBB-NEXT:    ret
-  %res = call <vscale x 8 x i32> @llvm.experimental.vector.interleave2.nxv8i32(<vscale x 4 x i32> %a, <vscale x 4 x i32> %b)
+  %res = call <vscale x 8 x i32> @llvm.vector.interleave2.nxv8i32(<vscale x 4 x i32> %a, <vscale x 4 x i32> %b)
   ret <vscale x 8 x i32> %res
 }
 
@@ -122,9 +122,9 @@ define <vscale x 4 x i64> @vector_interleave_nxv4i64_nxv2i64(<vscale x 2 x i64>
 ; CHECK-NEXT:    srli a0, a0, 2
 ; CHECK-NEXT:    vsetvli a1, zero, e16, m1, ta, mu
 ; CHECK-NEXT:    vid.v v12
-; CHECK-NEXT:    vand.vi v13, v12, 1
-; CHECK-NEXT:    vmsne.vi v0, v13, 0
 ; CHECK-NEXT:    vsrl.vi v16, v12, 1
+; CHECK-NEXT:    vand.vi v12, v12, 1
+; CHECK-NEXT:    vmsne.vi v0, v12, 0
 ; CHECK-NEXT:    vadd.vx v16, v16, a0, v0.t
 ; CHECK-NEXT:    vsetvli zero, zero, e64, m4, ta, ma
 ; CHECK-NEXT:    vrgatherei16.vv v12, v8, v16
@@ -137,23 +137,23 @@ define <vscale x 4 x i64> @vector_interleave_nxv4i64_nxv2i64(<vscale x 2 x i64>
 ; ZVBB-NEXT:    srli a0, a0, 2
 ; ZVBB-NEXT:    vsetvli a1, zero, e16, m1, ta, mu
 ; ZVBB-NEXT:    vid.v v12
-; ZVBB-NEXT:    vand.vi v13, v12, 1
-; ZVBB-NEXT:    vmsne.vi v0, v13, 0
 ; ZVBB-NEXT:    vsrl.vi v16, v12, 1
+; ZVBB-NEXT:    vand.vi v12, v12, 1
+; ZVBB-NEXT:    vmsne.vi v0, v12, 0
 ; ZVBB-NEXT:    vadd.vx v16, v16, a0, v0.t
 ; ZVBB-NEXT:    vsetvli zero, zero, e64, m4, ta, ma
 ; ZVBB-NEXT:    vrgatherei16.vv v12, v8, v16
 ; ZVBB-NEXT:    vmv.v.v v8, v12
 ; ZVBB-NEXT:    ret
-  %res = call <vscale x 4 x i64> @llvm.experimental.vector.interleave2.nxv4i64(<vscale x 2 x i64> %a, <vscale x 2 x i64> %b)
+  %res = call <vscale x 4 x i64> @llvm.vector.interleave2.nxv4i64(<vscale x 2 x i64> %a, <vscale x 2 x i64> %b)
   ret <vscale x 4 x i64> %res
 }
 
-declare <vscale x 32 x i1> @llvm.experimental.vector.interleave2.nxv32i1(<vscale x 16 x i1>, <vscale x 16 x i1>)
-declare <vscale x 32 x i8> @llvm.experimental.vector.interleave2.nxv32i8(<vscale x 16 x i8>, <vscale x 16 x i8>)
-declare <vscale x 16 x i16> @llvm.experimental.vector.interleave2.nxv16i16(<vscale x 8 x i16>, <vscale x 8 x i16>)
-declare <vscale x 8 x i32> @llvm.experimental.vector.interleave2.nxv8i32(<vscale x 4 x i32>, <vscale x 4 x i32>)
-declare <vscale x 4 x i64> @llvm.experimental.vector.interleave2.nxv4i64(<vscale x 2 x i64>, <vscale x 2 x i64>)
+declare <vscale x 32 x i1> @llvm.vector.interleave2.nxv32i1(<vscale x 16 x i1>, <vscale x 16 x i1>)
+declare <vscale x 32 x i8> @llvm.vector.interleave2.nxv32i8(<vscale x 16 x i8>, <vscale x 16 x i8>)
+declare <vscale x 16 x i16> @llvm.vector.interleave2.nxv16i16(<vscale x 8 x i16>, <vscale x 8 x i16>)
+declare <vscale x 8 x i32> @llvm.vector.interleave2.nxv8i32(<vscale x 4 x i32>, <vscale x 4 x i32>)
+declare <vscale x 4 x i64> @llvm.vector.interleave2.nxv4i64(<vscale x 2 x i64>, <vscale x 2 x i64>)
 
 define <vscale x 128 x i1> @vector_interleave_nxv128i1_nxv64i1(<vscale x 64 x i1> %a, <vscale x 64 x i1> %b) {
 ; CHECK-LABEL: vector_interleave_nxv128i1_nxv64i1:
@@ -196,7 +196,7 @@ define <vscale x 128 x i1> @vector_interleave_nxv128i1_nxv64i1(<vscale x 64 x i1
 ; ZVBB-NEXT:    vsetvli a0, zero, e8, m8, ta, ma
 ; ZVBB-NEXT:    vmsne.vi v8, v24, 0
 ; ZVBB-NEXT:    ret
-  %res = call <vscale x 128 x i1> @llvm.experimental.vector.interleave2.nxv128i1(<vscale x 64 x i1> %a, <vscale x 64 x i1> %b)
+  %res = call <vscale x 128 x i1> @llvm.vector.interleave2.nxv128i1(<vscale x 64 x i1> %a, <vscale x 64 x i1> %b)
   ret <vscale x 128 x i1> %res
 }
 
@@ -223,7 +223,7 @@ define <vscale x 128 x i8> @vector_interleave_nxv128i8_nxv64i8(<vscale x 64 x i8
 ; ZVBB-NEXT:    vwaddu.wv v0, v0, v28
 ; ZVBB-NEXT:    vmv8r.v v16, v0
 ; ZVBB-NEXT:    ret
-  %res = call <vscale x 128 x i8> @llvm.experimental.vector.interleave2.nxv128i8(<vscale x 64 x i8> %a, <vscale x 64 x i8> %b)
+  %res = call <vscale x 128 x i8> @llvm.vector.interleave2.nxv128i8(<vscale x 64 x i8> %a, <vscale x 64 x i8> %b)
   ret <vscale x 128 x i8> %res
 }
 
@@ -250,7 +250,7 @@ define <vscale x 64 x i16> @vector_interleave_nxv64i16_nxv32i16(<vscale x 32 x i
 ; ZVBB-NEXT:    vwaddu.wv v0, v0, v28
 ; ZVBB-NEXT:    vmv8r.v v16, v0
 ; ZVBB-NEXT:    ret
-  %res = call <vscale x 64 x i16> @llvm.experimental.vector.interleave2.nxv64i16(<vscale x 32 x i16> %a, <vscale x 32 x i16> %b)
+  %res = call <vscale x 64 x i16> @llvm.vector.interleave2.nxv64i16(<vscale x 32 x i16> %a, <vscale x 32 x i16> %b)
   ret <vscale x 64 x i16> %res
 }
 
@@ -278,7 +278,7 @@ define <vscale x 32 x i32> @vector_interleave_nxv32i32_nxv16i32(<vscale x 16 x i
 ; ZVBB-NEXT:    vmv8r.v v8, v24
 ; ZVBB-NEXT:    vmv8r.v v16, v0
 ; ZVBB-NEXT:    ret
-  %res = call <vscale x 32 x i32> @llvm.experimental.vector.interleave2.nxv32i32(<vscale x 16 x i32> %a, <vscale x 16 x i32> %b)
+  %res = call <vscale x 32 x i32> @llvm.vector.interleave2.nxv32i32(<vscale x 16 x i32> %a, <vscale x 16 x i32> %b)
   ret <vscale x 32 x i32> %res
 }
 
@@ -288,32 +288,44 @@ define <vscale x 16 x i64> @vector_interleave_nxv16i64_nxv8i64(<vscale x 8 x i64
 ; CHECK-NEXT:    addi sp, sp, -16
 ; CHECK-NEXT:    .cfi_def_cfa_offset 16
 ; CHECK-NEXT:    csrr a0, vlenb
-; CHECK-NEXT:    slli a0, a0, 3
+; CHECK-NEXT:    slli a0, a0, 4
 ; CHECK-NEXT:    sub sp, sp, a0
-; CHECK-NEXT:    .cfi_escape 0x0f, 0x0d, 0x72, 0x00, 0x11, 0x10, 0x22, 0x11, 0x08, 0x92, 0xa2, 0x38, 0x00, 0x1e, 0x22 # sp + 16 + 8 * vlenb
-; CHECK-NEXT:    vmv8r.v v0, v8
+; CHECK-NEXT:    .cfi_escape 0x0f, 0x0d, 0x72, 0x00, 0x11, 0x10, 0x22, 0x11, 0x10, 0x92, 0xa2, 0x38, 0x00, 0x1e, 0x22 # sp + 16 + 16 * vlenb
+; CHECK-NEXT:    csrr a0, vlenb
+; CHECK-NEXT:    slli a0, a0, 3
+; CHECK-NEXT:    add a0, sp, a0
+; CHECK-NEXT:    addi a0, a0, 16
+; CHECK-NEXT:    vs8r.v v8, (a0) # Unknown-size Folded Spill
 ; CHECK-NEXT:    csrr a0, vlenb
 ; CHECK-NEXT:    srli a0, a0, 1
 ; CHECK-NEXT:    vsetvli a1, zero, e16, m2, ta, mu
 ; CHECK-NEXT:    vid.v v24
-; CHECK-NEXT:    vand.vi v26, v24, 1
-; CHECK-NEXT:    vmsne.vi v10, v26, 0
-; CHECK-NEXT:    vsrl.vi v8, v24, 1
-; CHECK-NEXT:    vmv8r.v v24, v0
-; CHECK-NEXT:    vmv4r.v v12, v4
-; CHECK-NEXT:    vmv1r.v v0, v10
-; CHECK-NEXT:    vadd.vx v8, v8, a0, v0.t
+; CHECK-NEXT:    vsrl.vi v6, v24, 1
+; CHECK-NEXT:    vand.vi v8, v24, 1
+; CHECK-NEXT:    vmsne.vi v0, v8, 0
+; CHECK-NEXT:    csrr a1, vlenb
+; CHECK-NEXT:    slli a1, a1, 3
+; CHECK-NEXT:    add a1, sp, a1
+; CHECK-NEXT:    addi a1, a1, 16
+; CHECK-NEXT:    vl8r.v v24, (a1) # Unknown-size Folded Reload
+; CHECK-NEXT:    vadd.vx v6, v6, a0, v0.t
 ; CHECK-NEXT:    vmv4r.v v28, v16
 ; CHECK-NEXT:    vsetvli zero, zero, e64, m8, ta, ma
-; CHECK-NEXT:    vrgatherei16.vv v0, v24, v8
+; CHECK-NEXT:    vrgatherei16.vv v8, v24, v6
 ; CHECK-NEXT:    addi a0, sp, 16
-; CHECK-NEXT:    vs8r.v v0, (a0) # Unknown-size Folded Spill
+; CHECK-NEXT:    vs8r.v v8, (a0) # Unknown-size Folded Spill
+; CHECK-NEXT:    csrr a0, vlenb
+; CHECK-NEXT:    slli a0, a0, 3
+; CHECK-NEXT:    add a0, sp, a0
+; CHECK-NEXT:    addi a0, a0, 16
+; CHECK-NEXT:    vl8r.v v8, (a0) # Unknown-size Folded Reload
 ; CHECK-NEXT:    vmv4r.v v16, v12
-; CHECK-NEXT:    vrgatherei16.vv v24, v16, v8
+; CHECK-NEXT:    vrgatherei16.vv v24, v16, v6
+; CHECK-NEXT:    addi a0, sp, 16
 ; CHECK-NEXT:    vl8r.v v8, (a0) # Unknown-size Folded Reload
 ; CHECK-NEXT:    vmv.v.v v16, v24
 ; CHECK-NEXT:    csrr a0, vlenb
-; CHECK-NEXT:    slli a0, a0, 3
+; CHECK-NEXT:    slli a0, a0, 4
 ; CHECK-NEXT:    add sp, sp, a0
 ; CHECK-NEXT:    addi sp, sp, 16
 ; CHECK-NEXT:    ret
@@ -323,44 +335,56 @@ define <vscale x 16 x i64> @vector_interleave_nxv16i64_nxv8i64(<vscale x 8 x i64
 ; ZVBB-NEXT:    addi sp, sp, -16
 ; ZVBB-NEXT:    .cfi_def_cfa_offset 16
 ; ZVBB-NEXT:    csrr a0, vlenb
-; ZVBB-NEXT:    slli a0, a0, 3
+; ZVBB-NEXT:    slli a0, a0, 4
 ; ZVBB-NEXT:    sub sp, sp, a0
-; ZVBB-NEXT:    .cfi_escape 0x0f, 0x0d, 0x72, 0x00, 0x11, 0x10, 0x22, 0x11, 0x08, 0x92, 0xa2, 0x38, 0x00, 0x1e, 0x22 # sp + 16 + 8 * vlenb
-; ZVBB-NEXT:    vmv8r.v v0, v8
+; ZVBB-NEXT:    .cfi_escape 0x0f, 0x0d, 0x72, 0x00, 0x11, 0x10, 0x22, 0x11, 0x10, 0x92, 0xa2, 0x38, 0x00, 0x1e, 0x22 # sp + 16 + 16 * vlenb
+; ZVBB-NEXT:    csrr a0, vlenb
+; ZVBB-NEXT:    slli a0, a0, 3
+; ZVBB-NEXT:    add a0, sp, a0
+; ZVBB-NEXT:    addi a0, a0, 16
+; ZVBB-NEXT:    vs8r.v v8, (a0) # Unknown-size Folded Spill
 ; ZVBB-NEXT:    csrr a0, vlenb
 ; ZVBB-NEXT:    srli a0, a0, 1
 ; ZVBB-NEXT:    vsetvli a1, zero, e16, m2, ta, mu
 ; ZVBB-NEXT:    vid.v v24
-; ZVBB-NEXT:    vand.vi v26, v24, 1
-; ZVBB-NEXT:    vmsne.vi v10, v26, 0
-; ZVBB-NEXT:    vsrl.vi v8, v24, 1
-; ZVBB-NEXT:    vmv8r.v v24, v0
-; ZVBB-NEXT:    vmv4r.v v12, v4
-; ZVBB-NEXT:    vmv1r.v v0, v10
-; ZVBB-NEXT:    vadd.vx v8, v8, a0, v0.t
+; ZVBB-NEXT:    vsrl.vi v6, v24, 1
+; ZVBB-NEXT:    vand.vi v8, v24, 1
+; ZVBB-NEXT:    vmsne.vi v0, v8, 0
+; ZVBB-NEXT:    csrr a1, vlenb
+; ZVBB-NEXT:    slli a1, a1, 3
+; ZVBB-NEXT:    add a1, sp, a1
+; ZVBB-NEXT:    addi a1, a1, 16
+; ZVBB-NEXT:    vl8r.v v24, (a1) # Unknown-size Folded Reload
+; ZVBB-NEXT:    vadd.vx v6, v6, a0, v0.t
 ; ZVBB-NEXT:    vmv4r.v v28, v16
 ; ZVBB-NEXT:    vsetvli zero, zero, e64, m8, ta, ma
-; ZVBB-NEXT:    vrgatherei16.vv v0, v24, v8
+; ZVBB-NEXT:    vrgatherei16.vv v8, v24, v6
 ; ZVBB-NEXT:    addi a0, sp, 16
-; ZVBB-NEXT:    vs8r.v v0, (a0) # Unknown-size Folded Spill
+; ZVBB-NEXT:    vs8r.v v8, (a0) # Unknown-size Folded Spill
+; ZVBB-NEXT:    csrr a0, vlenb
+; ZVBB-NEXT:    slli a0, a0, 3
+; ZVBB-NEXT:    add a0, sp, a0
+; ZVBB-NEXT:    addi a0, a0, 16
+; ZVBB-NEXT:    vl8r.v v8, (a0) # Unknown-size Folded Reload
 ; ZVBB-NEXT:    vmv4r.v v16, v12
-; ZVBB-NEXT:    vrgatherei16.vv v24, v16, v8
+; ZVBB-NEXT:    vrgatherei16.vv v24, v16, v6
+; ZVBB-NEXT:    addi a0, sp, 16
 ; ZVBB-NEXT:    vl8r.v v8, (a0) # Unknown-size Folded Reload
 ; ZVBB-NEXT:    vmv.v.v v16, v24
 ; ZVBB-NEXT:    csrr a0, vlenb
-; ZVBB-NEXT:    slli a0, a0, 3
+; ZVBB-NEXT:    slli a0, a0, 4
 ; ZVBB-NEXT:    add sp, sp, a0
 ; ZVBB-NEXT:    addi sp, sp, 16
 ; ZVBB-NEXT:    ret
-  %res = call <vscale x 16 x i64> @llvm.experimental.vector.interleave2.nxv16i64(<vscale x 8 x i64> %a, <vscale x 8 x i64> %b)
+  %res = call <vscale x 16 x i64> @llvm.vector.interleave2.nxv16i64(<vscale x 8 x i64> %a, <vscale x 8 x i64> %b)
   ret <vscale x 16 x i64> %res
 }
 
-declare <vscale x 128 x i1> @llvm.experimental.vector.interleave2.nxv128i1(<vscale x 64 x i1>, <vscale x 64 x i1>)
-declare <vscale x 128 x i8> @llvm.experimental.vector.interleave2.nxv128i8(<vscale x 64 x i8>, <vscale x 64 x i8>)
-declare <vscale x 64 x i16> @llvm.experimental.vector.interleave2.nxv64i16(<vscale x 32 x i16>, <vscale x 32 x i16>)
-declare <vscale x 32 x i32> @llvm.experimental.vector.interleave2.nxv32i32(<vscale x 16 x i32>, <vscale x 16 x i32>)
-declare <vscale x 16 x i64> @llvm.experimental.vector.interleave2.nxv16i64(<vscale x 8 x i64>, <vscale x 8 x i64>)
+declare <vscale x 128 x i1> @llvm.vector.interleave2.nxv128i1(<vscale x 64 x i1>, <vscale x 64 x i1>)
+declare <vscale x 128 x i8> @llvm.vector.interleave2.nxv128i8(<vscale x 64 x i8>, <vscale x 64 x i8>)
+declare <vscale x 64 x i16> @llvm.vector.interleave2.nxv64i16(<vscale x 32 x i16>, <vscale x 32 x i16>)
+declare <vscale x 32 x i32> @llvm.vector.interleave2.nxv32i32(<vscale x 16 x i32>, <vscale x 16 x i32>)
+declare <vscale x 16 x i64> @llvm.vector.interleave2.nxv16i64(<vscale x 8 x i64>, <vscale x 8 x i64>)
 
 ; Floats
 
@@ -395,7 +419,7 @@ define <vscale x 4 x half> @vector_interleave_nxv4f16_nxv2f16(<vscale x 2 x half
 ; ZVBB-NEXT:    vslideup.vx v10, v8, a0
 ; ZVBB-NEXT:    vmv.v.v v8, v10
 ; ZVBB-NEXT:    ret
-  %res = call <vscale x 4 x half> @llvm.experimental.vector.interleave2.nxv4f16(<vscale x 2 x half> %a, <vscale x 2 x half> %b)
+  %res = call <vscale x 4 x half> @llvm.vector.interleave2.nxv4f16(<vscale x 2 x half> %a, <vscale x 2 x half> %b)
   ret <vscale x 4 x half> %res
 }
 
@@ -416,7 +440,7 @@ define <vscale x 8 x half> @vector_interleave_nxv8f16_nxv4f16(<vscale x 4 x half
 ; ZVBB-NEXT:    vwaddu.wv v10, v10, v8
 ; ZVBB-NEXT:    vmv2r.v v8, v10
 ; ZVBB-NEXT:    ret
-  %res = call <vscale x 8 x half> @llvm.experimental.vector.interleave2.nxv8f16(<vscale x 4 x half> %a, <vscale x 4 x half> %b)
+  %res = call <vscale x 8 x half> @llvm.vector.interleave2.nxv8f16(<vscale x 4 x half> %a, <vscale x 4 x half> %b)
   ret <vscale x 8 x half> %res
 }
 
@@ -438,7 +462,7 @@ define <vscale x 4 x float> @vector_interleave_nxv4f32_nxv2f32(<vscale x 2 x flo
 ; ZVBB-NEXT:    vwaddu.wv v10, v10, v8
 ; ZVBB-NEXT:    vmv2r.v v8, v10
 ; ZVBB-NEXT:    ret
-  %res = call <vscale x 4 x float> @llvm.experimental.vector.interleave2.nxv4f32(<vscale x 2 x float> %a, <vscale x 2 x float> %b)
+  %res = call <vscale x 4 x float> @llvm.vector.interleave2.nxv4f32(<vscale x 2 x float> %a, <vscale x 2 x float> %b)
   ret <vscale x 4 x float> %res
 }
 
@@ -459,7 +483,7 @@ define <vscale x 16 x half> @vector_interleave_nxv16f16_nxv8f16(<vscale x 8 x ha
 ; ZVBB-NEXT:    vwaddu.wv v12, v12, v8
 ; ZVBB-NEXT:    vmv4r.v v8, v12
 ; ZVBB-NEXT:    ret
-  %res = call <vscale x 16 x half> @llvm.experimental.vector.interleave2.nxv16f16(<vscale x 8 x half> %a, <vscale x 8 x half> %b)
+  %res = call <vscale x 16 x half> @llvm.vector.interleave2.nxv16f16(<vscale x 8 x half> %a, <vscale x 8 x half> %b)
   ret <vscale x 16 x half> %res
 }
 
@@ -481,7 +505,7 @@ define <vscale x 8 x float> @vector_interleave_nxv8f32_nxv4f32(<vscale x 4 x flo
 ; ZVBB-NEXT:    vwaddu.wv v12, v12, v8
 ; ZVBB-NEXT:    vmv4r.v v8, v12
 ; ZVBB-NEXT:    ret
-  %res = call <vscale x 8 x float> @llvm.experimental.vector.interleave2.nxv8f32(<vscale x 4 x float> %a, <vscale x 4 x float> %b)
+  %res = call <vscale x 8 x float> @llvm.vector.interleave2.nxv8f32(<vscale x 4 x float> %a, <vscale x 4 x float> %b)
   ret <vscale x 8 x float> %res
 }
 
@@ -492,9 +516,9 @@ define <vscale x 4 x double> @vector_interleave_nxv4f64_nxv2f64(<vscale x 2 x do
 ; CHECK-NEXT:    srli a0, a0, 2
 ; CHECK-NEXT:    vsetvli a1, zero, e16, m1, ta, mu
 ; CHECK-NEXT:    vid.v v12
-; CHECK-NEXT:    vand.vi v13, v12, 1
-; CHECK-NEXT:    vmsne.vi v0, v13, 0
 ; CHECK-NEXT:    vsrl.vi v16, v12, 1
+; CHECK-NEXT:    vand.vi v12, v12, 1
+; CHECK-NEXT:    vmsne.vi v0, v12, 0
 ; CHECK-NEXT:    vadd.vx v16, v16, a0, v0.t
 ; CHECK-NEXT:    vsetvli zero, zero, e64, m4, ta, ma
 ; CHECK-NEXT:    vrgatherei16.vv v12, v8, v16
@@ -507,25 +531,25 @@ define <vscale x 4 x double> @vector_interleave_nxv4f64_nxv2f64(<vscale x 2 x do
 ; ZVBB-NEXT:    srli a0, a0, 2
 ; ZVBB-NEXT:    vsetvli a1, zero, e16, m1, ta, mu
 ; ZVBB-NEXT:    vid.v v12
-; ZVBB-NEXT:    vand.vi v13, v12, 1
-; ZVBB-NEXT:    vmsne.vi v0, v13, 0
 ; ZVBB-NEXT:    vsrl.vi v16, v12, 1
+; ZVBB-NEXT:    vand.vi v12, v12, 1
+; ZVBB-NEXT:    vmsne.vi v0, v12, 0
 ; ZVBB-NEXT:    vadd.vx v16, v16, a0, v0.t
 ; ZVBB-NEXT:    vsetvli zero, zero, e64, m4, ta, ma
 ; ZVBB-NEXT:    vrgatherei16.vv v12, v8, v16
 ; ZVBB-NEXT:    vmv.v.v v8, v12
 ; ZVBB-NEXT:    ret
-  %res = call <vscale x 4 x double> @llvm.experimental.vector.interleave2.nxv4f64(<vscale x 2 x double> %a, <vscale x 2 x double> %b)
+  %res = call <vscale x 4 x double> @llvm.vector.interleave2.nxv4f64(<vscale x 2 x double> %a, <vscale x 2 x double> %b)
   ret <vscale x 4 x double> %res
 }
 
 
-declare <vscale x 4 x half> @llvm.experimental.vector.interleave2.nxv4f16(<vscale x 2 x half>, <vscale x 2 x half>)
-declare <vscale x 8 x half> @llvm.experimental.vector.interleave2.nxv8f16(<vscale x 4 x half>, <vscale x 4 x half>)
-declare <vscale x 4 x float> @llvm.experimental.vector.interleave2.nxv4f32(<vscale x 2 x float>, <vscale x 2 x float>)
-declare <vscale x 16 x half> @llvm.experimental.vector.interleave2.nxv16f16(<vscale x 8 x half>, <vscale x 8 x half>)
-declare <vscale x 8 x float> @llvm.experimental.vector.interleave2.nxv8f32(<vscale x 4 x float>, <vscale x 4 x float>)
-declare <vscale x 4 x double> @llvm.experimental.vector.interleave2.nxv4f64(<vscale x 2 x double>, <vscale x 2 x double>)
+declare <vscale x 4 x half> @llvm.vector.interleave2.nxv4f16(<vscale x 2 x half>, <vscale x 2 x half>)
+declare <vscale x 8 x half> @llvm.vector.interleave2.nxv8f16(<vscale x 4 x half>, <vscale x 4 x half>)
+declare <vscale x 4 x float> @llvm.vector.interleave2.nxv4f32(<vscale x 2 x float>, <vscale x 2 x float>)
+declare <vscale x 16 x half> @llvm.vector.interleave2.nxv16f16(<vscale x 8 x half>, <vscale x 8 x half>)
+declare <vscale x 8 x float> @llvm.vector.interleave2.nxv8f32(<vscale x 4 x float>, <vscale x 4 x float>)
+declare <vscale x 4 x double> @llvm.vector.interleave2.nxv4f64(<vscale x 2 x double>, <vscale x 2 x double>)
 
 define <vscale x 64 x half> @vector_interleave_nxv64f16_nxv32f16(<vscale x 32 x half> %a, <vscale x 32 x half> %b) {
 ; CHECK-LABEL: vector_interleave_nxv64f16_nxv32f16:
@@ -550,7 +574,7 @@ define <vscale x 64 x half> @vector_interleave_nxv64f16_nxv32f16(<vscale x 32 x
 ; ZVBB-NEXT:    vwaddu.wv v0, v0, v28
 ; ZVBB-NEXT:    vmv8r.v v16, v0
 ; ZVBB-NEXT:    ret
-  %res = call <vscale x 64 x half> @llvm.experimental.vector.interleave2.nxv64f16(<vscale x 32 x half> %a, <vscale x 32 x half> %b)
+  %res = call <vscale x 64 x half> @llvm.vector.interleave2.nxv64f16(<vscale x 32 x half> %a, <vscale x 32 x half> %b)
   ret <vscale x 64 x half> %res
 }
 
@@ -578,7 +602,7 @@ define <vscale x 32 x float> @vector_interleave_nxv32f32_nxv16f32(<vscale x 16 x
 ; ZVBB-NEXT:    vmv8r.v v8, v24
 ; ZVBB-NEXT:    vmv8r.v v16, v0
 ; ZVBB-NEXT:    ret
-  %res = call <vscale x 32 x float> @llvm.experimental.vector.interleave2.nxv32f32(<vscale x 16 x float> %a, <vscale x 16 x float> %b)
+  %res = call <vscale x 32 x float> @llvm.vector.interleave2.nxv32f32(<vscale x 16 x float> %a, <vscale x 16 x float> %b)
   ret <vscale x 32 x float> %res
 }
 
@@ -588,32 +612,44 @@ define <vscale x 16 x double> @vector_interleave_nxv16f64_nxv8f64(<vscale x 8 x
 ; CHECK-NEXT:    addi sp, sp, -16
 ; CHECK-NEXT:    .cfi_def_cfa_offset 16
 ; CHECK-NEXT:    csrr a0, vlenb
-; CHECK-NEXT:    slli a0, a0, 3
+; CHECK-NEXT:    slli a0, a0, 4
 ; CHECK-NEXT:    sub sp, sp, a0
-; CHECK-NEXT:    .cfi_escape 0x0f, 0x0d, 0x72, 0x00, 0x11, 0x10, 0x22, 0x11, 0x08, 0x92, 0xa2, 0x38, 0x00, 0x1e, 0x22 # sp + 16 + 8 * vlenb
-; CHECK-NEXT:    vmv8r.v v0, v8
+; CHECK-NEXT:    .cfi_escape 0x0f, 0x0d, 0x72, 0x00, 0x11, 0x10, 0x22, 0x11, 0x10, 0x92, 0xa2, 0x38, 0x00, 0x1e, 0x22 # sp + 16 + 16 * vlenb
+; CHECK-NEXT:    csrr a0, vlenb
+; CHECK-NEXT:    slli a0, a0, 3
+; CHECK-NEXT:    add a0, sp, a0
+; CHECK-NEXT:    addi a0, a0, 16
+; CHECK-NEXT:    vs8r.v v8, (a0) # Unknown-size Folded Spill
 ; CHECK-NEXT:    csrr a0, vlenb
 ; CHECK-NEXT:    srli a0, a0, 1
 ; CHECK-NEXT:    vsetvli a1, zero, e16, m2, ta, mu
 ; CHECK-NEXT:    vid.v v24
-; CHECK-NEXT:    vand.vi v26, v24, 1
-; CHECK-NEXT:    vmsne.vi v10, v26, 0
-; CHECK-NEXT:    vsrl.vi v8, v24, 1
-; CHECK-NEXT:    vmv8r.v v24, v0
-; CHECK-NEXT:    vmv4r.v v12, v4
-; CHECK-NEXT:    vmv1r.v v0, v10
-; CHECK-NEXT:    vadd.vx v8, v8, a0, v0.t
+; CHECK-NEXT:    vsrl.vi v6, v24, 1
+; CHECK-NEXT:    vand.vi v8, v24, 1
+; CHECK-NEXT:    vmsne.vi v0, v8, 0
+; CHECK-NEXT:    csrr a1, vlenb
+; CHECK-NEXT:    slli a1, a1, 3
+; CHECK-NEXT:    add a1, sp, a1
+; CHECK-NEXT:    addi a1, a1, 16
+; CHECK-NEXT:    vl8r.v v24, (a1) # Unknown-size Folded Reload
+; CHECK-NEXT:    vadd.vx v6, v6, a0, v0.t
 ; CHECK-NEXT:    vmv4r.v v28, v16
 ; CHECK-NEXT:    vsetvli zero, zero, e64, m8, ta, ma
-; CHECK-NEXT:    vrgatherei16.vv v0, v24, v8
+; CHECK-NEXT:    vrgatherei16.vv v8, v24, v6
 ; CHECK-NEXT:    addi a0, sp, 16
-; CHECK-NEXT:    vs8r.v v0, (a0) # Unknown-size Folded Spill
+; CHECK-NEXT:    vs8r.v v8, (a0) # Unknown-size Folded Spill
+; CHECK-NEXT:    csrr a0, vlenb
+; CHECK-NEXT:    slli a0, a0, 3
+; CHECK-NEXT:    add a0, sp, a0
+; CHECK-NEXT:    addi a0, a0, 16
+; CHECK-NEXT:    vl8r.v v8, (a0) # Unknown-size Folded Reload
 ; CHECK-NEXT:    vmv4r.v v16, v12
-; CHECK-NEXT:    vrgatherei16.vv v24, v16, v8
+; CHECK-NEXT:    vrgatherei16.vv v24, v16, v6
+; CHECK-NEXT:    addi a0, sp, 16
 ; CHECK-NEXT:    vl8r.v v8, (a0) # Unknown-size Folded Reload
 ; CHECK-NEXT:    vmv.v.v v16, v24
 ; CHECK-NEXT:    csrr a0, vlenb
-; CHECK-NEXT:    slli a0, a0, 3
+; CHECK-NEXT:    slli a0, a0, 4
 ; CHECK-NEXT:    add sp, sp, a0
 ; CHECK-NEXT:    addi sp, sp, 16
 ; CHECK-NEXT:    ret
@@ -623,36 +659,48 @@ define <vscale x 16 x double> @vector_interleave_nxv16f64_nxv8f64(<vscale x 8 x
 ; ZVBB-NEXT:    addi sp, sp, -16
 ; ZVBB-NEXT:    .cfi_def_cfa_offset 16
 ; ZVBB-NEXT:    csrr a0, vlenb
-; ZVBB-NEXT:    slli a0, a0, 3
+; ZVBB-NEXT:    slli a0, a0, 4
 ; ZVBB-NEXT:    sub sp, sp, a0
-; ZVBB-NEXT:    .cfi_escape 0x0f, 0x0d, 0x72, 0x00, 0x11, 0x10, 0x22, 0x11, 0x08, 0x92, 0xa2, 0x38, 0x00, 0x1e, 0x22 # sp + 16 + 8 * vlenb
-; ZVBB-NEXT:    vmv8r.v v0, v8
+; ZVBB-NEXT:    .cfi_escape 0x0f, 0x0d, 0x72, 0x00, 0x11, 0x10, 0x22, 0x11, 0x10, 0x92, 0xa2, 0x38, 0x00, 0x1e, 0x22 # sp + 16 + 16 * vlenb
+; ZVBB-NEXT:    csrr a0, vlenb
+; ZVBB-NEXT:    slli a0, a0, 3
+; ZVBB-NEXT:    add a0, sp, a0
+; ZVBB-NEXT:    addi a0, a0, 16
+; ZVBB-NEXT:    vs8r.v v8, (a0) # Unknown-size Folded Spill
 ; ZVBB-NEXT:    csrr a0, vlenb
 ; ZVBB-NEXT:    srli a0, a0, 1
 ; ZVBB-NEXT:    vsetvli a1, zero, e16, m2, ta, mu
 ; ZVBB-NEXT:    vid.v v24
-; ZVBB-NEXT:    vand.vi v26, v24, 1
-; ZVBB-NEXT:    vmsne.vi v10, v26, 0
-; ZVBB-NEXT:    vsrl.vi v8, v24, 1
-; ZVBB-NEXT:    vmv8r.v v24, v0
-; ZVBB-NEXT:    vmv4r.v v12, v4
-; ZVBB-NEXT:    vmv1r.v v0, v10
-; ZVBB-NEXT:    vadd.vx v8, v8, a0, v0.t
+; ZVBB-NEXT:    vsrl.vi v6, v24, 1
+; ZVBB-NEXT:    vand.vi v8, v24, 1
+; ZVBB-NEXT:    vmsne.vi v0, v8, 0
+; ZVBB-NEXT:    csrr a1, vlenb
+; ZVBB-NEXT:    slli a1, a1, 3
+; ZVBB-NEXT:    add a1, sp, a1
+; ZVBB-NEXT:    addi a1, a1, 16
+; ZVBB-NEXT:    vl8r.v v24, (a1) # Unknown-size Folded Reload
+; ZVBB-NEXT:    vadd.vx v6, v6, a0, v0.t
 ; ZVBB-NEXT:    vmv4r.v v28, v16
 ; ZVBB-NEXT:    vsetvli zero, zero, e64, m8, ta, ma
-; ZVBB-NEXT:    vrgatherei16.vv v0, v24, v8
+; ZVBB-NEXT:    vrgatherei16.vv v8, v24, v6
 ; ZVBB-NEXT:    addi a0, sp, 16
-; ZVBB-NEXT:    vs8r.v v0, (a0) # Unknown-size Folded Spill
+; ZVBB-NEXT:    vs8r.v v8, (a0) # Unknown-size Folded Spill
+; ZVBB-NEXT:    csrr a0, vlenb
+; ZVBB-NEXT:    slli a0, a0, 3
+; ZVBB-NEXT:    add a0, sp, a0
+; ZVBB-NEXT:    addi a0, a0, 16
+; ZVBB-NEXT:    vl8r.v v8, (a0) # Unknown-size Folded Reload
 ; ZVBB-NEXT:    vmv4r.v v16, v12
-; ZVBB-NEXT:    vrgatherei16.vv v24, v16, v8
+; ZVBB-NEXT:    vrgatherei16.vv v24, v16, v6
+; ZVBB-NEXT:    addi a0, sp, 16
 ; ZVBB-NEXT:    vl8r.v v8, (a0) # Unknown-size Folded Reload
 ; ZVBB-NEXT:    vmv.v.v v16, v24
 ; ZVBB-NEXT:    csrr a0, vlenb
-; ZVBB-NEXT:    slli a0, a0, 3
+; ZVBB-NEXT:    slli a0, a0, 4
 ; ZVBB-NEXT:    add sp, sp, a0
 ; ZVBB-NEXT:    addi sp, sp, 16
 ; ZVBB-NEXT:    ret
-  %res = call <vscale x 16 x double> @llvm.experimental.vector.interleave2.nxv16f64(<vscale x 8 x double> %a, <vscale x 8 x double> %b)
+  %res = call <vscale x 16 x double> @llvm.vector.interleave2.nxv16f64(<vscale x 8 x double> %a, <vscale x 8 x double> %b)
   ret <vscale x 16 x double> %res
 }
 
@@ -670,7 +718,7 @@ define <vscale x 8 x i32> @vector_interleave_nxv8i32_nxv4i32_poison(<vscale x 4
 ; ZVBB-NEXT:    vzext.vf2 v12, v8
 ; ZVBB-NEXT:    vmv.v.v v8, v12
 ; ZVBB-NEXT:    ret
-  %res = call <vscale x 8 x i32> @llvm.experimental.vector.interleave2.nxv8i32(<vscale x 4 x i32> %a, <vscale x 4 x i32> poison)
+  %res = call <vscale x 8 x i32> @llvm.vector.interleave2.nxv8i32(<vscale x 4 x i32> %a, <vscale x 4 x i32> poison)
   ret <vscale x 8 x i32> %res
 }
 
@@ -690,10 +738,10 @@ define <vscale x 8 x i32> @vector_interleave_nxv8i32_nxv4i32_poison2(<vscale x 4
 ; ZVBB-NEXT:    vwsll.vx v12, v8, a0
 ; ZVBB-NEXT:    vmv4r.v v8, v12
 ; ZVBB-NEXT:    ret
-  %res = call <vscale x 8 x i32> @llvm.experimental.vector.interleave2.nxv8i32(<vscale x 4 x i32> poison, <vscale x 4 x i32> %a)
+  %res = call <vscale x 8 x i32> @llvm.vector.interleave2.nxv8i32(<vscale x 4 x i32> poison, <vscale x 4 x i32> %a)
   ret <vscale x 8 x i32> %res
 }
 
-declare <vscale x 64 x half> @llvm.experimental.vector.interleave2.nxv64f16(<vscale x 32 x half>, <vscale x 32 x half>)
-declare <vscale x 32 x float> @llvm.experimental.vector.interleave2.nxv32f32(<vscale x 16 x float>, <vscale x 16 x float>)
-declare <vscale x 16 x double> @llvm.experimental.vector.interleave2.nxv16f64(<vscale x 8 x double>, <vscale x 8 x double>)
+declare <vscale x 64 x half> @llvm.vector.interleave2.nxv64f16(<vscale x 32 x half>, <vscale x 32 x half>)
+declare <vscale x 32 x float> @llvm.vector.interleave2.nxv32f32(<vscale x 16 x float>, <vscale x 16 x float>)
+declare <vscale x 16 x double> @llvm.vector.interleave2.nxv16f64(<vscale x 8 x double>, <vscale x 8 x double>)
diff --git a/llvm/test/CodeGen/RISCV/rvv/vector-reassociations.ll b/llvm/test/CodeGen/RISCV/rvv/vector-reassociations.ll
new file mode 100644
index 000000000000..6435c1c14e06
--- /dev/null
+++ b/llvm/test/CodeGen/RISCV/rvv/vector-reassociations.ll
@@ -0,0 +1,253 @@
+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --version 4
+; RUN: llc -mtriple=riscv32 -mattr='+v' -O3 %s -o - | FileCheck %s
+
+declare <vscale x 1 x i8> @llvm.riscv.vadd.nxv1i8.nxv1i8(
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  i32)
+
+declare <vscale x 1 x i8> @llvm.riscv.vadd.mask.nxv1i8.nxv1i8(
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i1>,
+  i32, i32)
+
+declare <vscale x 1 x i8> @llvm.riscv.vsub.nxv1i8.nxv1i8(
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  i32)
+
+declare <vscale x 1 x i8> @llvm.riscv.vmul.nxv1i8.nxv1i8(
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  <vscale x 1 x i8>,
+  i32)
+
+define <vscale x 1 x i8> @simple_vadd_vv(<vscale x 1 x i8> %0, <vscale x 1 x i8> %1, i32 %2) nounwind {
+; CHECK-LABEL: simple_vadd_vv:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli zero, a0, e8, mf8, ta, ma
+; CHECK-NEXT:    vadd.vv v9, v8, v9
+; CHECK-NEXT:    vadd.vv v8, v8, v8
+; CHECK-NEXT:    vadd.vv v8, v8, v9
+; CHECK-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i8> @llvm.riscv.vadd.nxv1i8.nxv1i8(
+    <vscale x 1 x i8> undef,
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i8> %1,
+    i32 %2)
+
+  %b = call <vscale x 1 x i8> @llvm.riscv.vadd.nxv1i8.nxv1i8(
+    <vscale x 1 x i8> undef,
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i8> %a,
+    i32 %2)
+
+  %c = call <vscale x 1 x i8> @llvm.riscv.vadd.nxv1i8.nxv1i8(
+    <vscale x 1 x i8> undef,
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i8> %b,
+    i32 %2)
+
+  ret <vscale x 1 x i8> %c
+}
+
+define <vscale x 1 x i8> @simple_vadd_vsub_vv(<vscale x 1 x i8> %0, <vscale x 1 x i8> %1, i32 %2) nounwind {
+; CHECK-LABEL: simple_vadd_vsub_vv:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli zero, a0, e8, mf8, ta, ma
+; CHECK-NEXT:    vsub.vv v9, v8, v9
+; CHECK-NEXT:    vadd.vv v8, v8, v8
+; CHECK-NEXT:    vadd.vv v8, v8, v9
+; CHECK-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i8> @llvm.riscv.vsub.nxv1i8.nxv1i8(
+    <vscale x 1 x i8> undef,
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i8> %1,
+    i32 %2)
+
+  %b = call <vscale x 1 x i8> @llvm.riscv.vadd.nxv1i8.nxv1i8(
+    <vscale x 1 x i8> undef,
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i8> %a,
+    i32 %2)
+
+  %c = call <vscale x 1 x i8> @llvm.riscv.vadd.nxv1i8.nxv1i8(
+    <vscale x 1 x i8> undef,
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i8> %b,
+    i32 %2)
+
+  ret <vscale x 1 x i8> %c
+}
+
+define <vscale x 1 x i8> @simple_vmul_vv(<vscale x 1 x i8> %0, <vscale x 1 x i8> %1, i32 %2) nounwind {
+; CHECK-LABEL: simple_vmul_vv:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli zero, a0, e8, mf8, ta, ma
+; CHECK-NEXT:    vmul.vv v9, v8, v9
+; CHECK-NEXT:    vmul.vv v8, v8, v8
+; CHECK-NEXT:    vmul.vv v8, v8, v9
+; CHECK-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i8> @llvm.riscv.vmul.nxv1i8.nxv1i8(
+    <vscale x 1 x i8> undef,
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i8> %1,
+    i32 %2)
+
+  %b = call <vscale x 1 x i8> @llvm.riscv.vmul.nxv1i8.nxv1i8(
+    <vscale x 1 x i8> undef,
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i8> %a,
+    i32 %2)
+
+  %c = call <vscale x 1 x i8> @llvm.riscv.vmul.nxv1i8.nxv1i8(
+    <vscale x 1 x i8> undef,
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i8> %b,
+    i32 %2)
+
+  ret <vscale x 1 x i8> %c
+}
+
+; With passthru and masks.
+define <vscale x 1 x i8> @vadd_vv_passthru(<vscale x 1 x i8> %0, <vscale x 1 x i8> %1, i32 %2) nounwind {
+; CHECK-LABEL: vadd_vv_passthru:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli zero, a0, e8, mf8, tu, ma
+; CHECK-NEXT:    vmv1r.v v10, v8
+; CHECK-NEXT:    vadd.vv v10, v8, v9
+; CHECK-NEXT:    vmv1r.v v9, v8
+; CHECK-NEXT:    vadd.vv v9, v8, v8
+; CHECK-NEXT:    vadd.vv v8, v9, v10
+; CHECK-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i8> @llvm.riscv.vadd.nxv1i8.nxv1i8(
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i8> %1,
+    i32 %2)
+
+  %b = call <vscale x 1 x i8> @llvm.riscv.vadd.nxv1i8.nxv1i8(
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i8> %a,
+    i32 %2)
+
+  %c = call <vscale x 1 x i8> @llvm.riscv.vadd.nxv1i8.nxv1i8(
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i8> %b,
+    i32 %2)
+
+  ret <vscale x 1 x i8> %c
+}
+
+define <vscale x 1 x i8> @vadd_vv_passthru_negative(<vscale x 1 x i8> %0, <vscale x 1 x i8> %1, i32 %2) nounwind {
+; CHECK-LABEL: vadd_vv_passthru_negative:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli zero, a0, e8, mf8, tu, ma
+; CHECK-NEXT:    vmv1r.v v10, v8
+; CHECK-NEXT:    vadd.vv v10, v8, v9
+; CHECK-NEXT:    vadd.vv v9, v8, v10
+; CHECK-NEXT:    vadd.vv v8, v8, v9
+; CHECK-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i8> @llvm.riscv.vadd.nxv1i8.nxv1i8(
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i8> %1,
+    i32 %2)
+
+  %b = call <vscale x 1 x i8> @llvm.riscv.vadd.nxv1i8.nxv1i8(
+    <vscale x 1 x i8> %1,
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i8> %a,
+    i32 %2)
+
+  %c = call <vscale x 1 x i8> @llvm.riscv.vadd.nxv1i8.nxv1i8(
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i8> %b,
+    i32 %2)
+
+  ret <vscale x 1 x i8> %c
+}
+
+define <vscale x 1 x i8> @vadd_vv_mask(<vscale x 1 x i8> %0, <vscale x 1 x i8> %1, i32 %2, <vscale x 1 x i1> %m) nounwind {
+; CHECK-LABEL: vadd_vv_mask:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli zero, a0, e8, mf8, ta, mu
+; CHECK-NEXT:    vmv1r.v v10, v8
+; CHECK-NEXT:    vadd.vv v10, v8, v9, v0.t
+; CHECK-NEXT:    vmv1r.v v9, v8
+; CHECK-NEXT:    vadd.vv v9, v8, v8, v0.t
+; CHECK-NEXT:    vadd.vv v8, v9, v10, v0.t
+; CHECK-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i8> @llvm.riscv.vadd.mask.nxv1i8.nxv1i8(
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i8> %1,
+    <vscale x 1 x i1> %m,
+    i32 %2, i32 1)
+
+  %b = call <vscale x 1 x i8> @llvm.riscv.vadd.mask.nxv1i8.nxv1i8(
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i8> %a,
+    <vscale x 1 x i1> %m,
+    i32 %2, i32 1)
+
+  %c = call <vscale x 1 x i8> @llvm.riscv.vadd.mask.nxv1i8.nxv1i8(
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i8> %b,
+    <vscale x 1 x i1> %m,
+    i32 %2, i32 1)
+
+  ret <vscale x 1 x i8> %c
+}
+
+define <vscale x 1 x i8> @vadd_vv_mask_negative(<vscale x 1 x i8> %0, <vscale x 1 x i8> %1, i32 %2, <vscale x 1 x i1> %m, <vscale x 1 x i1> %m2) nounwind {
+; CHECK-LABEL: vadd_vv_mask_negative:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli zero, a0, e8, mf8, ta, mu
+; CHECK-NEXT:    vmv1r.v v11, v8
+; CHECK-NEXT:    vadd.vv v11, v8, v9, v0.t
+; CHECK-NEXT:    vmv1r.v v9, v8
+; CHECK-NEXT:    vadd.vv v9, v8, v11, v0.t
+; CHECK-NEXT:    vmv1r.v v0, v10
+; CHECK-NEXT:    vadd.vv v8, v8, v9, v0.t
+; CHECK-NEXT:    ret
+entry:
+  %a = call <vscale x 1 x i8> @llvm.riscv.vadd.mask.nxv1i8.nxv1i8(
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i8> %1,
+    <vscale x 1 x i1> %m,
+    i32 %2, i32 1)
+
+  %b = call <vscale x 1 x i8> @llvm.riscv.vadd.mask.nxv1i8.nxv1i8(
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i8> %a,
+    <vscale x 1 x i1> %m,
+    i32 %2, i32 1)
+
+  %c = call <vscale x 1 x i8> @llvm.riscv.vadd.mask.nxv1i8.nxv1i8(
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i8> %0,
+    <vscale x 1 x i8> %b,
+    <vscale x 1 x i1> %m2,
+    i32 %2, i32 1)
+
+  ret <vscale x 1 x i8> %c
+}
+
diff --git a/llvm/test/CodeGen/RISCV/rvv/vector-splice.ll b/llvm/test/CodeGen/RISCV/rvv/vector-splice.ll
index c98242437f62..be56db52e349 100644
--- a/llvm/test/CodeGen/RISCV/rvv/vector-splice.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/vector-splice.ll
@@ -4,7 +4,7 @@
 
 ; Tests assume VLEN=128 or vscale_range_min=2.
 
-declare <vscale x 1 x i1> @llvm.experimental.vector.splice.nxv1i1(<vscale x 1 x i1>, <vscale x 1 x i1>, i32)
+declare <vscale x 1 x i1> @llvm.vector.splice.nxv1i1(<vscale x 1 x i1>, <vscale x 1 x i1>, i32)
 
 define <vscale x 1 x i1> @splice_nxv1i1_offset_negone(<vscale x 1 x i1> %a, <vscale x 1 x i1> %b) #0 {
 ; CHECK-LABEL: splice_nxv1i1_offset_negone:
@@ -24,7 +24,7 @@ define <vscale x 1 x i1> @splice_nxv1i1_offset_negone(<vscale x 1 x i1> %a, <vsc
 ; CHECK-NEXT:    vand.vi v8, v9, 1
 ; CHECK-NEXT:    vmsne.vi v0, v8, 0
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 1 x i1> @llvm.experimental.vector.splice.nxv1i1(<vscale x 1 x i1> %a, <vscale x 1 x i1> %b, i32 -1)
+  %res = call <vscale x 1 x i1> @llvm.vector.splice.nxv1i1(<vscale x 1 x i1> %a, <vscale x 1 x i1> %b, i32 -1)
   ret <vscale x 1 x i1> %res
 }
 
@@ -48,11 +48,11 @@ define <vscale x 1 x i1> @splice_nxv1i1_offset_max(<vscale x 1 x i1> %a, <vscale
 ; CHECK-NEXT:    vand.vi v8, v9, 1
 ; CHECK-NEXT:    vmsne.vi v0, v8, 0
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 1 x i1> @llvm.experimental.vector.splice.nxv1i1(<vscale x 1 x i1> %a, <vscale x 1 x i1> %b, i32 1)
+  %res = call <vscale x 1 x i1> @llvm.vector.splice.nxv1i1(<vscale x 1 x i1> %a, <vscale x 1 x i1> %b, i32 1)
   ret <vscale x 1 x i1> %res
 }
 
-declare <vscale x 2 x i1> @llvm.experimental.vector.splice.nxv2i1(<vscale x 2 x i1>, <vscale x 2 x i1>, i32)
+declare <vscale x 2 x i1> @llvm.vector.splice.nxv2i1(<vscale x 2 x i1>, <vscale x 2 x i1>, i32)
 
 define <vscale x 2 x i1> @splice_nxv2i1_offset_negone(<vscale x 2 x i1> %a, <vscale x 2 x i1> %b) #0 {
 ; CHECK-LABEL: splice_nxv2i1_offset_negone:
@@ -72,7 +72,7 @@ define <vscale x 2 x i1> @splice_nxv2i1_offset_negone(<vscale x 2 x i1> %a, <vsc
 ; CHECK-NEXT:    vand.vi v8, v9, 1
 ; CHECK-NEXT:    vmsne.vi v0, v8, 0
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 2 x i1> @llvm.experimental.vector.splice.nxv2i1(<vscale x 2 x i1> %a, <vscale x 2 x i1> %b, i32 -1)
+  %res = call <vscale x 2 x i1> @llvm.vector.splice.nxv2i1(<vscale x 2 x i1> %a, <vscale x 2 x i1> %b, i32 -1)
   ret <vscale x 2 x i1> %res
 }
 
@@ -96,11 +96,11 @@ define <vscale x 2 x i1> @splice_nxv2i1_offset_max(<vscale x 2 x i1> %a, <vscale
 ; CHECK-NEXT:    vand.vi v8, v9, 1
 ; CHECK-NEXT:    vmsne.vi v0, v8, 0
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 2 x i1> @llvm.experimental.vector.splice.nxv2i1(<vscale x 2 x i1> %a, <vscale x 2 x i1> %b, i32 3)
+  %res = call <vscale x 2 x i1> @llvm.vector.splice.nxv2i1(<vscale x 2 x i1> %a, <vscale x 2 x i1> %b, i32 3)
   ret <vscale x 2 x i1> %res
 }
 
-declare <vscale x 4 x i1> @llvm.experimental.vector.splice.nxv4i1(<vscale x 4 x i1>, <vscale x 4 x i1>, i32)
+declare <vscale x 4 x i1> @llvm.vector.splice.nxv4i1(<vscale x 4 x i1>, <vscale x 4 x i1>, i32)
 
 define <vscale x 4 x i1> @splice_nxv4i1_offset_negone(<vscale x 4 x i1> %a, <vscale x 4 x i1> %b) #0 {
 ; CHECK-LABEL: splice_nxv4i1_offset_negone:
@@ -120,7 +120,7 @@ define <vscale x 4 x i1> @splice_nxv4i1_offset_negone(<vscale x 4 x i1> %a, <vsc
 ; CHECK-NEXT:    vand.vi v8, v9, 1
 ; CHECK-NEXT:    vmsne.vi v0, v8, 0
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 4 x i1> @llvm.experimental.vector.splice.nxv4i1(<vscale x 4 x i1> %a, <vscale x 4 x i1> %b, i32 -1)
+  %res = call <vscale x 4 x i1> @llvm.vector.splice.nxv4i1(<vscale x 4 x i1> %a, <vscale x 4 x i1> %b, i32 -1)
   ret <vscale x 4 x i1> %res
 }
 
@@ -144,11 +144,11 @@ define <vscale x 4 x i1> @splice_nxv4i1_offset_max(<vscale x 4 x i1> %a, <vscale
 ; CHECK-NEXT:    vand.vi v8, v9, 1
 ; CHECK-NEXT:    vmsne.vi v0, v8, 0
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 4 x i1> @llvm.experimental.vector.splice.nxv4i1(<vscale x 4 x i1> %a, <vscale x 4 x i1> %b, i32 7)
+  %res = call <vscale x 4 x i1> @llvm.vector.splice.nxv4i1(<vscale x 4 x i1> %a, <vscale x 4 x i1> %b, i32 7)
   ret <vscale x 4 x i1> %res
 }
 
-declare <vscale x 8 x i1> @llvm.experimental.vector.splice.nxv8i1(<vscale x 8 x i1>, <vscale x 8 x i1>, i32)
+declare <vscale x 8 x i1> @llvm.vector.splice.nxv8i1(<vscale x 8 x i1>, <vscale x 8 x i1>, i32)
 
 define <vscale x 8 x i1> @splice_nxv8i1_offset_negone(<vscale x 8 x i1> %a, <vscale x 8 x i1> %b) #0 {
 ; CHECK-LABEL: splice_nxv8i1_offset_negone:
@@ -167,7 +167,7 @@ define <vscale x 8 x i1> @splice_nxv8i1_offset_negone(<vscale x 8 x i1> %a, <vsc
 ; CHECK-NEXT:    vand.vi v8, v9, 1
 ; CHECK-NEXT:    vmsne.vi v0, v8, 0
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 8 x i1> @llvm.experimental.vector.splice.nxv8i1(<vscale x 8 x i1> %a, <vscale x 8 x i1> %b, i32 -1)
+  %res = call <vscale x 8 x i1> @llvm.vector.splice.nxv8i1(<vscale x 8 x i1> %a, <vscale x 8 x i1> %b, i32 -1)
   ret <vscale x 8 x i1> %res
 }
 
@@ -190,11 +190,11 @@ define <vscale x 8 x i1> @splice_nxv8i1_offset_max(<vscale x 8 x i1> %a, <vscale
 ; CHECK-NEXT:    vand.vi v8, v9, 1
 ; CHECK-NEXT:    vmsne.vi v0, v8, 0
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 8 x i1> @llvm.experimental.vector.splice.nxv8i1(<vscale x 8 x i1> %a, <vscale x 8 x i1> %b, i32 15)
+  %res = call <vscale x 8 x i1> @llvm.vector.splice.nxv8i1(<vscale x 8 x i1> %a, <vscale x 8 x i1> %b, i32 15)
   ret <vscale x 8 x i1> %res
 }
 
-declare <vscale x 16 x i1> @llvm.experimental.vector.splice.nxv16i1(<vscale x 16 x i1>, <vscale x 16 x i1>, i32)
+declare <vscale x 16 x i1> @llvm.vector.splice.nxv16i1(<vscale x 16 x i1>, <vscale x 16 x i1>, i32)
 
 define <vscale x 16 x i1> @splice_nxv16i1_offset_negone(<vscale x 16 x i1> %a, <vscale x 16 x i1> %b) #0 {
 ; CHECK-LABEL: splice_nxv16i1_offset_negone:
@@ -216,7 +216,7 @@ define <vscale x 16 x i1> @splice_nxv16i1_offset_negone(<vscale x 16 x i1> %a, <
 ; CHECK-NEXT:    vand.vi v8, v8, 1
 ; CHECK-NEXT:    vmsne.vi v0, v8, 0
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 16 x i1> @llvm.experimental.vector.splice.nxv16i1(<vscale x 16 x i1> %a, <vscale x 16 x i1> %b, i32 -1)
+  %res = call <vscale x 16 x i1> @llvm.vector.splice.nxv16i1(<vscale x 16 x i1> %a, <vscale x 16 x i1> %b, i32 -1)
   ret <vscale x 16 x i1> %res
 }
 
@@ -240,11 +240,11 @@ define <vscale x 16 x i1> @splice_nxv16i1_offset_max(<vscale x 16 x i1> %a, <vsc
 ; CHECK-NEXT:    vand.vi v8, v8, 1
 ; CHECK-NEXT:    vmsne.vi v0, v8, 0
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 16 x i1> @llvm.experimental.vector.splice.nxv16i1(<vscale x 16 x i1> %a, <vscale x 16 x i1> %b, i32 31)
+  %res = call <vscale x 16 x i1> @llvm.vector.splice.nxv16i1(<vscale x 16 x i1> %a, <vscale x 16 x i1> %b, i32 31)
   ret <vscale x 16 x i1> %res
 }
 
-declare <vscale x 32 x i1> @llvm.experimental.vector.splice.nxv32i1(<vscale x 32 x i1>, <vscale x 32 x i1>, i32)
+declare <vscale x 32 x i1> @llvm.vector.splice.nxv32i1(<vscale x 32 x i1>, <vscale x 32 x i1>, i32)
 
 define <vscale x 32 x i1> @splice_nxv32i1_offset_negone(<vscale x 32 x i1> %a, <vscale x 32 x i1> %b) #0 {
 ; CHECK-LABEL: splice_nxv32i1_offset_negone:
@@ -266,7 +266,7 @@ define <vscale x 32 x i1> @splice_nxv32i1_offset_negone(<vscale x 32 x i1> %a, <
 ; CHECK-NEXT:    vand.vi v8, v8, 1
 ; CHECK-NEXT:    vmsne.vi v0, v8, 0
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 32 x i1> @llvm.experimental.vector.splice.nxv32i1(<vscale x 32 x i1> %a, <vscale x 32 x i1> %b, i32 -1)
+  %res = call <vscale x 32 x i1> @llvm.vector.splice.nxv32i1(<vscale x 32 x i1> %a, <vscale x 32 x i1> %b, i32 -1)
   ret <vscale x 32 x i1> %res
 }
 
@@ -289,11 +289,11 @@ define <vscale x 32 x i1> @splice_nxv32i1_offset_max(<vscale x 32 x i1> %a, <vsc
 ; CHECK-NEXT:    vand.vi v8, v16, 1
 ; CHECK-NEXT:    vmsne.vi v0, v8, 0
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 32 x i1> @llvm.experimental.vector.splice.nxv32i1(<vscale x 32 x i1> %a, <vscale x 32 x i1> %b, i32 63)
+  %res = call <vscale x 32 x i1> @llvm.vector.splice.nxv32i1(<vscale x 32 x i1> %a, <vscale x 32 x i1> %b, i32 63)
   ret <vscale x 32 x i1> %res
 }
 
-declare <vscale x 64 x i1> @llvm.experimental.vector.splice.nxv64i1(<vscale x 64 x i1>, <vscale x 64 x i1>, i32)
+declare <vscale x 64 x i1> @llvm.vector.splice.nxv64i1(<vscale x 64 x i1>, <vscale x 64 x i1>, i32)
 
 define <vscale x 64 x i1> @splice_nxv64i1_offset_negone(<vscale x 64 x i1> %a, <vscale x 64 x i1> %b) #0 {
 ; CHECK-LABEL: splice_nxv64i1_offset_negone:
@@ -315,7 +315,7 @@ define <vscale x 64 x i1> @splice_nxv64i1_offset_negone(<vscale x 64 x i1> %a, <
 ; CHECK-NEXT:    vand.vi v8, v8, 1
 ; CHECK-NEXT:    vmsne.vi v0, v8, 0
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 64 x i1> @llvm.experimental.vector.splice.nxv64i1(<vscale x 64 x i1> %a, <vscale x 64 x i1> %b, i32 -1)
+  %res = call <vscale x 64 x i1> @llvm.vector.splice.nxv64i1(<vscale x 64 x i1> %a, <vscale x 64 x i1> %b, i32 -1)
   ret <vscale x 64 x i1> %res
 }
 
@@ -338,17 +338,17 @@ define <vscale x 64 x i1> @splice_nxv64i1_offset_max(<vscale x 64 x i1> %a, <vsc
 ; CHECK-NEXT:    vand.vi v8, v24, 1
 ; CHECK-NEXT:    vmsne.vi v0, v8, 0
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 64 x i1> @llvm.experimental.vector.splice.nxv64i1(<vscale x 64 x i1> %a, <vscale x 64 x i1> %b, i32 127)
+  %res = call <vscale x 64 x i1> @llvm.vector.splice.nxv64i1(<vscale x 64 x i1> %a, <vscale x 64 x i1> %b, i32 127)
   ret <vscale x 64 x i1> %res
 }
 
-declare <vscale x 1 x i8> @llvm.experimental.vector.splice.nxv1i8(<vscale x 1 x i8>, <vscale x 1 x i8>, i32)
+declare <vscale x 1 x i8> @llvm.vector.splice.nxv1i8(<vscale x 1 x i8>, <vscale x 1 x i8>, i32)
 
 define <vscale x 1 x i8> @splice_nxv1i8_offset_zero(<vscale x 1 x i8> %a, <vscale x 1 x i8> %b) #0 {
 ; CHECK-LABEL: splice_nxv1i8_offset_zero:
 ; CHECK:       # %bb.0:
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 1 x i8> @llvm.experimental.vector.splice.nxv1i8(<vscale x 1 x i8> %a, <vscale x 1 x i8> %b, i32 0)
+  %res = call <vscale x 1 x i8> @llvm.vector.splice.nxv1i8(<vscale x 1 x i8> %a, <vscale x 1 x i8> %b, i32 0)
   ret <vscale x 1 x i8> %res
 }
 
@@ -363,7 +363,7 @@ define <vscale x 1 x i8> @splice_nxv1i8_offset_negone(<vscale x 1 x i8> %a, <vsc
 ; CHECK-NEXT:    vsetvli a0, zero, e8, mf8, ta, ma
 ; CHECK-NEXT:    vslideup.vi v8, v9, 1
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 1 x i8> @llvm.experimental.vector.splice.nxv1i8(<vscale x 1 x i8> %a, <vscale x 1 x i8> %b, i32 -1)
+  %res = call <vscale x 1 x i8> @llvm.vector.splice.nxv1i8(<vscale x 1 x i8> %a, <vscale x 1 x i8> %b, i32 -1)
   ret <vscale x 1 x i8> %res
 }
 
@@ -378,7 +378,7 @@ define <vscale x 1 x i8> @splice_nxv1i8_offset_min(<vscale x 1 x i8> %a, <vscale
 ; CHECK-NEXT:    vsetvli a0, zero, e8, mf8, ta, ma
 ; CHECK-NEXT:    vslideup.vi v8, v9, 2
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 1 x i8> @llvm.experimental.vector.splice.nxv1i8(<vscale x 1 x i8> %a, <vscale x 1 x i8> %b, i32 -2)
+  %res = call <vscale x 1 x i8> @llvm.vector.splice.nxv1i8(<vscale x 1 x i8> %a, <vscale x 1 x i8> %b, i32 -2)
   ret <vscale x 1 x i8> %res
 }
 
@@ -393,17 +393,17 @@ define <vscale x 1 x i8> @splice_nxv1i8_offset_max(<vscale x 1 x i8> %a, <vscale
 ; CHECK-NEXT:    vsetvli a1, zero, e8, mf8, ta, ma
 ; CHECK-NEXT:    vslideup.vx v8, v9, a0
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 1 x i8> @llvm.experimental.vector.splice.nxv1i8(<vscale x 1 x i8> %a, <vscale x 1 x i8> %b, i32 1)
+  %res = call <vscale x 1 x i8> @llvm.vector.splice.nxv1i8(<vscale x 1 x i8> %a, <vscale x 1 x i8> %b, i32 1)
   ret <vscale x 1 x i8> %res
 }
 
-declare <vscale x 2 x i8> @llvm.experimental.vector.splice.nxv2i8(<vscale x 2 x i8>, <vscale x 2 x i8>, i32)
+declare <vscale x 2 x i8> @llvm.vector.splice.nxv2i8(<vscale x 2 x i8>, <vscale x 2 x i8>, i32)
 
 define <vscale x 2 x i8> @splice_nxv2i8_offset_zero(<vscale x 2 x i8> %a, <vscale x 2 x i8> %b) #0 {
 ; CHECK-LABEL: splice_nxv2i8_offset_zero:
 ; CHECK:       # %bb.0:
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 2 x i8> @llvm.experimental.vector.splice.nxv2i8(<vscale x 2 x i8> %a, <vscale x 2 x i8> %b, i32 0)
+  %res = call <vscale x 2 x i8> @llvm.vector.splice.nxv2i8(<vscale x 2 x i8> %a, <vscale x 2 x i8> %b, i32 0)
   ret <vscale x 2 x i8> %res
 }
 
@@ -418,7 +418,7 @@ define <vscale x 2 x i8> @splice_nxv2i8_offset_negone(<vscale x 2 x i8> %a, <vsc
 ; CHECK-NEXT:    vsetvli a0, zero, e8, mf4, ta, ma
 ; CHECK-NEXT:    vslideup.vi v8, v9, 1
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 2 x i8> @llvm.experimental.vector.splice.nxv2i8(<vscale x 2 x i8> %a, <vscale x 2 x i8> %b, i32 -1)
+  %res = call <vscale x 2 x i8> @llvm.vector.splice.nxv2i8(<vscale x 2 x i8> %a, <vscale x 2 x i8> %b, i32 -1)
   ret <vscale x 2 x i8> %res
 }
 
@@ -433,7 +433,7 @@ define <vscale x 2 x i8> @splice_nxv2i8_offset_min(<vscale x 2 x i8> %a, <vscale
 ; CHECK-NEXT:    vsetvli a0, zero, e8, mf4, ta, ma
 ; CHECK-NEXT:    vslideup.vi v8, v9, 4
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 2 x i8> @llvm.experimental.vector.splice.nxv2i8(<vscale x 2 x i8> %a, <vscale x 2 x i8> %b, i32 -4)
+  %res = call <vscale x 2 x i8> @llvm.vector.splice.nxv2i8(<vscale x 2 x i8> %a, <vscale x 2 x i8> %b, i32 -4)
   ret <vscale x 2 x i8> %res
 }
 
@@ -448,17 +448,17 @@ define <vscale x 2 x i8> @splice_nxv2i8_offset_max(<vscale x 2 x i8> %a, <vscale
 ; CHECK-NEXT:    vsetvli a1, zero, e8, mf4, ta, ma
 ; CHECK-NEXT:    vslideup.vx v8, v9, a0
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 2 x i8> @llvm.experimental.vector.splice.nxv2i8(<vscale x 2 x i8> %a, <vscale x 2 x i8> %b, i32 3)
+  %res = call <vscale x 2 x i8> @llvm.vector.splice.nxv2i8(<vscale x 2 x i8> %a, <vscale x 2 x i8> %b, i32 3)
   ret <vscale x 2 x i8> %res
 }
 
-declare <vscale x 4 x i8> @llvm.experimental.vector.splice.nxv4i8(<vscale x 4 x i8>, <vscale x 4 x i8>, i32)
+declare <vscale x 4 x i8> @llvm.vector.splice.nxv4i8(<vscale x 4 x i8>, <vscale x 4 x i8>, i32)
 
 define <vscale x 4 x i8> @splice_nxv4i8_offset_zero(<vscale x 4 x i8> %a, <vscale x 4 x i8> %b) #0 {
 ; CHECK-LABEL: splice_nxv4i8_offset_zero:
 ; CHECK:       # %bb.0:
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 4 x i8> @llvm.experimental.vector.splice.nxv4i8(<vscale x 4 x i8> %a, <vscale x 4 x i8> %b, i32 0)
+  %res = call <vscale x 4 x i8> @llvm.vector.splice.nxv4i8(<vscale x 4 x i8> %a, <vscale x 4 x i8> %b, i32 0)
   ret <vscale x 4 x i8> %res
 }
 
@@ -473,7 +473,7 @@ define <vscale x 4 x i8> @splice_nxv4i8_offset_negone(<vscale x 4 x i8> %a, <vsc
 ; CHECK-NEXT:    vsetvli a0, zero, e8, mf2, ta, ma
 ; CHECK-NEXT:    vslideup.vi v8, v9, 1
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 4 x i8> @llvm.experimental.vector.splice.nxv4i8(<vscale x 4 x i8> %a, <vscale x 4 x i8> %b, i32 -1)
+  %res = call <vscale x 4 x i8> @llvm.vector.splice.nxv4i8(<vscale x 4 x i8> %a, <vscale x 4 x i8> %b, i32 -1)
   ret <vscale x 4 x i8> %res
 }
 
@@ -488,7 +488,7 @@ define <vscale x 4 x i8> @splice_nxv4i8_offset_min(<vscale x 4 x i8> %a, <vscale
 ; CHECK-NEXT:    vsetvli a0, zero, e8, mf2, ta, ma
 ; CHECK-NEXT:    vslideup.vi v8, v9, 8
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 4 x i8> @llvm.experimental.vector.splice.nxv4i8(<vscale x 4 x i8> %a, <vscale x 4 x i8> %b, i32 -8)
+  %res = call <vscale x 4 x i8> @llvm.vector.splice.nxv4i8(<vscale x 4 x i8> %a, <vscale x 4 x i8> %b, i32 -8)
   ret <vscale x 4 x i8> %res
 }
 
@@ -503,17 +503,17 @@ define <vscale x 4 x i8> @splice_nxv4i8_offset_max(<vscale x 4 x i8> %a, <vscale
 ; CHECK-NEXT:    vsetvli a1, zero, e8, mf2, ta, ma
 ; CHECK-NEXT:    vslideup.vx v8, v9, a0
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 4 x i8> @llvm.experimental.vector.splice.nxv4i8(<vscale x 4 x i8> %a, <vscale x 4 x i8> %b, i32 7)
+  %res = call <vscale x 4 x i8> @llvm.vector.splice.nxv4i8(<vscale x 4 x i8> %a, <vscale x 4 x i8> %b, i32 7)
   ret <vscale x 4 x i8> %res
 }
 
-declare <vscale x 8 x i8> @llvm.experimental.vector.splice.nxv8i8(<vscale x 8 x i8>, <vscale x 8 x i8>, i32)
+declare <vscale x 8 x i8> @llvm.vector.splice.nxv8i8(<vscale x 8 x i8>, <vscale x 8 x i8>, i32)
 
 define <vscale x 8 x i8> @splice_nxv8i8_offset_zero(<vscale x 8 x i8> %a, <vscale x 8 x i8> %b) #0 {
 ; CHECK-LABEL: splice_nxv8i8_offset_zero:
 ; CHECK:       # %bb.0:
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 8 x i8> @llvm.experimental.vector.splice.nxv8i8(<vscale x 8 x i8> %a, <vscale x 8 x i8> %b, i32 0)
+  %res = call <vscale x 8 x i8> @llvm.vector.splice.nxv8i8(<vscale x 8 x i8> %a, <vscale x 8 x i8> %b, i32 0)
   ret <vscale x 8 x i8> %res
 }
 
@@ -527,7 +527,7 @@ define <vscale x 8 x i8> @splice_nxv8i8_offset_negone(<vscale x 8 x i8> %a, <vsc
 ; CHECK-NEXT:    vsetvli a0, zero, e8, m1, ta, ma
 ; CHECK-NEXT:    vslideup.vi v8, v9, 1
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 8 x i8> @llvm.experimental.vector.splice.nxv8i8(<vscale x 8 x i8> %a, <vscale x 8 x i8> %b, i32 -1)
+  %res = call <vscale x 8 x i8> @llvm.vector.splice.nxv8i8(<vscale x 8 x i8> %a, <vscale x 8 x i8> %b, i32 -1)
   ret <vscale x 8 x i8> %res
 }
 
@@ -541,7 +541,7 @@ define <vscale x 8 x i8> @splice_nxv8i8_offset_min(<vscale x 8 x i8> %a, <vscale
 ; CHECK-NEXT:    vsetvli a0, zero, e8, m1, ta, ma
 ; CHECK-NEXT:    vslideup.vi v8, v9, 16
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 8 x i8> @llvm.experimental.vector.splice.nxv8i8(<vscale x 8 x i8> %a, <vscale x 8 x i8> %b, i32 -16)
+  %res = call <vscale x 8 x i8> @llvm.vector.splice.nxv8i8(<vscale x 8 x i8> %a, <vscale x 8 x i8> %b, i32 -16)
   ret <vscale x 8 x i8> %res
 }
 
@@ -555,17 +555,17 @@ define <vscale x 8 x i8> @splice_nxv8i8_offset_max(<vscale x 8 x i8> %a, <vscale
 ; CHECK-NEXT:    vsetvli a1, zero, e8, m1, ta, ma
 ; CHECK-NEXT:    vslideup.vx v8, v9, a0
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 8 x i8> @llvm.experimental.vector.splice.nxv8i8(<vscale x 8 x i8> %a, <vscale x 8 x i8> %b, i32 15)
+  %res = call <vscale x 8 x i8> @llvm.vector.splice.nxv8i8(<vscale x 8 x i8> %a, <vscale x 8 x i8> %b, i32 15)
   ret <vscale x 8 x i8> %res
 }
 
-declare <vscale x 16 x i8> @llvm.experimental.vector.splice.nxv16i8(<vscale x 16 x i8>, <vscale x 16 x i8>, i32)
+declare <vscale x 16 x i8> @llvm.vector.splice.nxv16i8(<vscale x 16 x i8>, <vscale x 16 x i8>, i32)
 
 define <vscale x 16 x i8> @splice_nxv16i8_offset_zero(<vscale x 16 x i8> %a, <vscale x 16 x i8> %b) #0 {
 ; CHECK-LABEL: splice_nxv16i8_offset_zero:
 ; CHECK:       # %bb.0:
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 16 x i8> @llvm.experimental.vector.splice.nxv16i8(<vscale x 16 x i8> %a, <vscale x 16 x i8> %b, i32 0)
+  %res = call <vscale x 16 x i8> @llvm.vector.splice.nxv16i8(<vscale x 16 x i8> %a, <vscale x 16 x i8> %b, i32 0)
   ret <vscale x 16 x i8> %res
 }
 
@@ -580,7 +580,7 @@ define <vscale x 16 x i8> @splice_nxv16i8_offset_negone(<vscale x 16 x i8> %a, <
 ; CHECK-NEXT:    vsetvli a0, zero, e8, m2, ta, ma
 ; CHECK-NEXT:    vslideup.vi v8, v10, 1
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 16 x i8> @llvm.experimental.vector.splice.nxv16i8(<vscale x 16 x i8> %a, <vscale x 16 x i8> %b, i32 -1)
+  %res = call <vscale x 16 x i8> @llvm.vector.splice.nxv16i8(<vscale x 16 x i8> %a, <vscale x 16 x i8> %b, i32 -1)
   ret <vscale x 16 x i8> %res
 }
 
@@ -596,7 +596,7 @@ define <vscale x 16 x i8> @splice_nxv16i8_offset_min(<vscale x 16 x i8> %a, <vsc
 ; CHECK-NEXT:    vsetvli a0, zero, e8, m2, ta, ma
 ; CHECK-NEXT:    vslideup.vx v8, v10, a1
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 16 x i8> @llvm.experimental.vector.splice.nxv16i8(<vscale x 16 x i8> %a, <vscale x 16 x i8> %b, i32 -32)
+  %res = call <vscale x 16 x i8> @llvm.vector.splice.nxv16i8(<vscale x 16 x i8> %a, <vscale x 16 x i8> %b, i32 -32)
   ret <vscale x 16 x i8> %res
 }
 
@@ -611,17 +611,17 @@ define <vscale x 16 x i8> @splice_nxv16i8_offset_max(<vscale x 16 x i8> %a, <vsc
 ; CHECK-NEXT:    vsetvli a1, zero, e8, m2, ta, ma
 ; CHECK-NEXT:    vslideup.vx v8, v10, a0
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 16 x i8> @llvm.experimental.vector.splice.nxv16i8(<vscale x 16 x i8> %a, <vscale x 16 x i8> %b, i32 31)
+  %res = call <vscale x 16 x i8> @llvm.vector.splice.nxv16i8(<vscale x 16 x i8> %a, <vscale x 16 x i8> %b, i32 31)
   ret <vscale x 16 x i8> %res
 }
 
-declare <vscale x 32 x i8> @llvm.experimental.vector.splice.nxv32i8(<vscale x 32 x i8>, <vscale x 32 x i8>, i32)
+declare <vscale x 32 x i8> @llvm.vector.splice.nxv32i8(<vscale x 32 x i8>, <vscale x 32 x i8>, i32)
 
 define <vscale x 32 x i8> @splice_nxv32i8_offset_zero(<vscale x 32 x i8> %a, <vscale x 32 x i8> %b) #0 {
 ; CHECK-LABEL: splice_nxv32i8_offset_zero:
 ; CHECK:       # %bb.0:
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 32 x i8> @llvm.experimental.vector.splice.nxv32i8(<vscale x 32 x i8> %a, <vscale x 32 x i8> %b, i32 0)
+  %res = call <vscale x 32 x i8> @llvm.vector.splice.nxv32i8(<vscale x 32 x i8> %a, <vscale x 32 x i8> %b, i32 0)
   ret <vscale x 32 x i8> %res
 }
 
@@ -636,7 +636,7 @@ define <vscale x 32 x i8> @splice_nxv32i8_offset_negone(<vscale x 32 x i8> %a, <
 ; CHECK-NEXT:    vsetvli a0, zero, e8, m4, ta, ma
 ; CHECK-NEXT:    vslideup.vi v8, v12, 1
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 32 x i8> @llvm.experimental.vector.splice.nxv32i8(<vscale x 32 x i8> %a, <vscale x 32 x i8> %b, i32 -1)
+  %res = call <vscale x 32 x i8> @llvm.vector.splice.nxv32i8(<vscale x 32 x i8> %a, <vscale x 32 x i8> %b, i32 -1)
   ret <vscale x 32 x i8> %res
 }
 
@@ -652,7 +652,7 @@ define <vscale x 32 x i8> @splice_nxv32i8_offset_min(<vscale x 32 x i8> %a, <vsc
 ; CHECK-NEXT:    vsetvli a0, zero, e8, m4, ta, ma
 ; CHECK-NEXT:    vslideup.vx v8, v12, a1
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 32 x i8> @llvm.experimental.vector.splice.nxv32i8(<vscale x 32 x i8> %a, <vscale x 32 x i8> %b, i32 -64)
+  %res = call <vscale x 32 x i8> @llvm.vector.splice.nxv32i8(<vscale x 32 x i8> %a, <vscale x 32 x i8> %b, i32 -64)
   ret <vscale x 32 x i8> %res
 }
 
@@ -668,17 +668,17 @@ define <vscale x 32 x i8> @splice_nxv32i8_offset_max(<vscale x 32 x i8> %a, <vsc
 ; CHECK-NEXT:    vsetvli a1, zero, e8, m4, ta, ma
 ; CHECK-NEXT:    vslideup.vx v8, v12, a0
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 32 x i8> @llvm.experimental.vector.splice.nxv32i8(<vscale x 32 x i8> %a, <vscale x 32 x i8> %b, i32 63)
+  %res = call <vscale x 32 x i8> @llvm.vector.splice.nxv32i8(<vscale x 32 x i8> %a, <vscale x 32 x i8> %b, i32 63)
   ret <vscale x 32 x i8> %res
 }
 
-declare <vscale x 64 x i8> @llvm.experimental.vector.splice.nxv64i8(<vscale x 64 x i8>, <vscale x 64 x i8>, i32)
+declare <vscale x 64 x i8> @llvm.vector.splice.nxv64i8(<vscale x 64 x i8>, <vscale x 64 x i8>, i32)
 
 define <vscale x 64 x i8> @splice_nxv64i8_offset_zero(<vscale x 64 x i8> %a, <vscale x 64 x i8> %b) #0 {
 ; CHECK-LABEL: splice_nxv64i8_offset_zero:
 ; CHECK:       # %bb.0:
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 64 x i8> @llvm.experimental.vector.splice.nxv64i8(<vscale x 64 x i8> %a, <vscale x 64 x i8> %b, i32 0)
+  %res = call <vscale x 64 x i8> @llvm.vector.splice.nxv64i8(<vscale x 64 x i8> %a, <vscale x 64 x i8> %b, i32 0)
   ret <vscale x 64 x i8> %res
 }
 
@@ -693,7 +693,7 @@ define <vscale x 64 x i8> @splice_nxv64i8_offset_negone(<vscale x 64 x i8> %a, <
 ; CHECK-NEXT:    vsetvli a0, zero, e8, m8, ta, ma
 ; CHECK-NEXT:    vslideup.vi v8, v16, 1
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 64 x i8> @llvm.experimental.vector.splice.nxv64i8(<vscale x 64 x i8> %a, <vscale x 64 x i8> %b, i32 -1)
+  %res = call <vscale x 64 x i8> @llvm.vector.splice.nxv64i8(<vscale x 64 x i8> %a, <vscale x 64 x i8> %b, i32 -1)
   ret <vscale x 64 x i8> %res
 }
 
@@ -709,7 +709,7 @@ define <vscale x 64 x i8> @splice_nxv64i8_offset_min(<vscale x 64 x i8> %a, <vsc
 ; CHECK-NEXT:    vsetvli a0, zero, e8, m8, ta, ma
 ; CHECK-NEXT:    vslideup.vx v8, v16, a1
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 64 x i8> @llvm.experimental.vector.splice.nxv64i8(<vscale x 64 x i8> %a, <vscale x 64 x i8> %b, i32 -128)
+  %res = call <vscale x 64 x i8> @llvm.vector.splice.nxv64i8(<vscale x 64 x i8> %a, <vscale x 64 x i8> %b, i32 -128)
   ret <vscale x 64 x i8> %res
 }
 
@@ -725,17 +725,17 @@ define <vscale x 64 x i8> @splice_nxv64i8_offset_max(<vscale x 64 x i8> %a, <vsc
 ; CHECK-NEXT:    vsetvli a1, zero, e8, m8, ta, ma
 ; CHECK-NEXT:    vslideup.vx v8, v16, a0
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 64 x i8> @llvm.experimental.vector.splice.nxv64i8(<vscale x 64 x i8> %a, <vscale x 64 x i8> %b, i32 127)
+  %res = call <vscale x 64 x i8> @llvm.vector.splice.nxv64i8(<vscale x 64 x i8> %a, <vscale x 64 x i8> %b, i32 127)
   ret <vscale x 64 x i8> %res
 }
 
-declare <vscale x 1 x i16> @llvm.experimental.vector.splice.nxv1i16(<vscale x 1 x i16>, <vscale x 1 x i16>, i32)
+declare <vscale x 1 x i16> @llvm.vector.splice.nxv1i16(<vscale x 1 x i16>, <vscale x 1 x i16>, i32)
 
 define <vscale x 1 x i16> @splice_nxv1i16_offset_zero(<vscale x 1 x i16> %a, <vscale x 1 x i16> %b) #0 {
 ; CHECK-LABEL: splice_nxv1i16_offset_zero:
 ; CHECK:       # %bb.0:
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 1 x i16> @llvm.experimental.vector.splice.nxv1i16(<vscale x 1 x i16> %a, <vscale x 1 x i16> %b, i32 0)
+  %res = call <vscale x 1 x i16> @llvm.vector.splice.nxv1i16(<vscale x 1 x i16> %a, <vscale x 1 x i16> %b, i32 0)
   ret <vscale x 1 x i16> %res
 }
 
@@ -750,7 +750,7 @@ define <vscale x 1 x i16> @splice_nxv1i16_offset_negone(<vscale x 1 x i16> %a, <
 ; CHECK-NEXT:    vsetvli a0, zero, e16, mf4, ta, ma
 ; CHECK-NEXT:    vslideup.vi v8, v9, 1
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 1 x i16> @llvm.experimental.vector.splice.nxv1i16(<vscale x 1 x i16> %a, <vscale x 1 x i16> %b, i32 -1)
+  %res = call <vscale x 1 x i16> @llvm.vector.splice.nxv1i16(<vscale x 1 x i16> %a, <vscale x 1 x i16> %b, i32 -1)
   ret <vscale x 1 x i16> %res
 }
 
@@ -765,7 +765,7 @@ define <vscale x 1 x i16> @splice_nxv1i16_offset_min(<vscale x 1 x i16> %a, <vsc
 ; CHECK-NEXT:    vsetvli a0, zero, e16, mf4, ta, ma
 ; CHECK-NEXT:    vslideup.vi v8, v9, 2
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 1 x i16> @llvm.experimental.vector.splice.nxv1i16(<vscale x 1 x i16> %a, <vscale x 1 x i16> %b, i32 -2)
+  %res = call <vscale x 1 x i16> @llvm.vector.splice.nxv1i16(<vscale x 1 x i16> %a, <vscale x 1 x i16> %b, i32 -2)
   ret <vscale x 1 x i16> %res
 }
 
@@ -780,17 +780,17 @@ define <vscale x 1 x i16> @splice_nxv1i16_offset_max(<vscale x 1 x i16> %a, <vsc
 ; CHECK-NEXT:    vsetvli a1, zero, e16, mf4, ta, ma
 ; CHECK-NEXT:    vslideup.vx v8, v9, a0
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 1 x i16> @llvm.experimental.vector.splice.nxv1i16(<vscale x 1 x i16> %a, <vscale x 1 x i16> %b, i32 1)
+  %res = call <vscale x 1 x i16> @llvm.vector.splice.nxv1i16(<vscale x 1 x i16> %a, <vscale x 1 x i16> %b, i32 1)
   ret <vscale x 1 x i16> %res
 }
 
-declare <vscale x 2 x i16> @llvm.experimental.vector.splice.nxv2i16(<vscale x 2 x i16>, <vscale x 2 x i16>, i32)
+declare <vscale x 2 x i16> @llvm.vector.splice.nxv2i16(<vscale x 2 x i16>, <vscale x 2 x i16>, i32)
 
 define <vscale x 2 x i16> @splice_nxv2i16_offset_zero(<vscale x 2 x i16> %a, <vscale x 2 x i16> %b) #0 {
 ; CHECK-LABEL: splice_nxv2i16_offset_zero:
 ; CHECK:       # %bb.0:
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 2 x i16> @llvm.experimental.vector.splice.nxv2i16(<vscale x 2 x i16> %a, <vscale x 2 x i16> %b, i32 0)
+  %res = call <vscale x 2 x i16> @llvm.vector.splice.nxv2i16(<vscale x 2 x i16> %a, <vscale x 2 x i16> %b, i32 0)
   ret <vscale x 2 x i16> %res
 }
 
@@ -805,7 +805,7 @@ define <vscale x 2 x i16> @splice_nxv2i16_offset_negone(<vscale x 2 x i16> %a, <
 ; CHECK-NEXT:    vsetvli a0, zero, e16, mf2, ta, ma
 ; CHECK-NEXT:    vslideup.vi v8, v9, 1
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 2 x i16> @llvm.experimental.vector.splice.nxv2i16(<vscale x 2 x i16> %a, <vscale x 2 x i16> %b, i32 -1)
+  %res = call <vscale x 2 x i16> @llvm.vector.splice.nxv2i16(<vscale x 2 x i16> %a, <vscale x 2 x i16> %b, i32 -1)
   ret <vscale x 2 x i16> %res
 }
 
@@ -820,7 +820,7 @@ define <vscale x 2 x i16> @splice_nxv2i16_offset_min(<vscale x 2 x i16> %a, <vsc
 ; CHECK-NEXT:    vsetvli a0, zero, e16, mf2, ta, ma
 ; CHECK-NEXT:    vslideup.vi v8, v9, 4
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 2 x i16> @llvm.experimental.vector.splice.nxv2i16(<vscale x 2 x i16> %a, <vscale x 2 x i16> %b, i32 -4)
+  %res = call <vscale x 2 x i16> @llvm.vector.splice.nxv2i16(<vscale x 2 x i16> %a, <vscale x 2 x i16> %b, i32 -4)
   ret <vscale x 2 x i16> %res
 }
 
@@ -835,17 +835,17 @@ define <vscale x 2 x i16> @splice_nxv2i16_offset_max(<vscale x 2 x i16> %a, <vsc
 ; CHECK-NEXT:    vsetvli a1, zero, e16, mf2, ta, ma
 ; CHECK-NEXT:    vslideup.vx v8, v9, a0
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 2 x i16> @llvm.experimental.vector.splice.nxv2i16(<vscale x 2 x i16> %a, <vscale x 2 x i16> %b, i32 3)
+  %res = call <vscale x 2 x i16> @llvm.vector.splice.nxv2i16(<vscale x 2 x i16> %a, <vscale x 2 x i16> %b, i32 3)
   ret <vscale x 2 x i16> %res
 }
 
-declare <vscale x 4 x i16> @llvm.experimental.vector.splice.nxv4i16(<vscale x 4 x i16>, <vscale x 4 x i16>, i32)
+declare <vscale x 4 x i16> @llvm.vector.splice.nxv4i16(<vscale x 4 x i16>, <vscale x 4 x i16>, i32)
 
 define <vscale x 4 x i16> @splice_nxv4i16_offset_zero(<vscale x 4 x i16> %a, <vscale x 4 x i16> %b) #0 {
 ; CHECK-LABEL: splice_nxv4i16_offset_zero:
 ; CHECK:       # %bb.0:
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 4 x i16> @llvm.experimental.vector.splice.nxv4i16(<vscale x 4 x i16> %a, <vscale x 4 x i16> %b, i32 0)
+  %res = call <vscale x 4 x i16> @llvm.vector.splice.nxv4i16(<vscale x 4 x i16> %a, <vscale x 4 x i16> %b, i32 0)
   ret <vscale x 4 x i16> %res
 }
 
@@ -860,7 +860,7 @@ define <vscale x 4 x i16> @splice_nxv4i16_offset_negone(<vscale x 4 x i16> %a, <
 ; CHECK-NEXT:    vsetvli a0, zero, e16, m1, ta, ma
 ; CHECK-NEXT:    vslideup.vi v8, v9, 1
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 4 x i16> @llvm.experimental.vector.splice.nxv4i16(<vscale x 4 x i16> %a, <vscale x 4 x i16> %b, i32 -1)
+  %res = call <vscale x 4 x i16> @llvm.vector.splice.nxv4i16(<vscale x 4 x i16> %a, <vscale x 4 x i16> %b, i32 -1)
   ret <vscale x 4 x i16> %res
 }
 
@@ -875,7 +875,7 @@ define <vscale x 4 x i16> @splice_nxv4i16_offset_min(<vscale x 4 x i16> %a, <vsc
 ; CHECK-NEXT:    vsetvli a0, zero, e16, m1, ta, ma
 ; CHECK-NEXT:    vslideup.vi v8, v9, 8
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 4 x i16> @llvm.experimental.vector.splice.nxv4i16(<vscale x 4 x i16> %a, <vscale x 4 x i16> %b, i32 -8)
+  %res = call <vscale x 4 x i16> @llvm.vector.splice.nxv4i16(<vscale x 4 x i16> %a, <vscale x 4 x i16> %b, i32 -8)
   ret <vscale x 4 x i16> %res
 }
 
@@ -890,17 +890,17 @@ define <vscale x 4 x i16> @splice_nxv4i16_offset_max(<vscale x 4 x i16> %a, <vsc
 ; CHECK-NEXT:    vsetvli a1, zero, e16, m1, ta, ma
 ; CHECK-NEXT:    vslideup.vx v8, v9, a0
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 4 x i16> @llvm.experimental.vector.splice.nxv4i16(<vscale x 4 x i16> %a, <vscale x 4 x i16> %b, i32 7)
+  %res = call <vscale x 4 x i16> @llvm.vector.splice.nxv4i16(<vscale x 4 x i16> %a, <vscale x 4 x i16> %b, i32 7)
   ret <vscale x 4 x i16> %res
 }
 
-declare <vscale x 8 x i16> @llvm.experimental.vector.splice.nxv8i16(<vscale x 8 x i16>, <vscale x 8 x i16>, i32)
+declare <vscale x 8 x i16> @llvm.vector.splice.nxv8i16(<vscale x 8 x i16>, <vscale x 8 x i16>, i32)
 
 define <vscale x 8 x i16> @splice_nxv8i16_offset_zero(<vscale x 8 x i16> %a, <vscale x 8 x i16> %b) #0 {
 ; CHECK-LABEL: splice_nxv8i16_offset_zero:
 ; CHECK:       # %bb.0:
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 8 x i16> @llvm.experimental.vector.splice.nxv8i16(<vscale x 8 x i16> %a, <vscale x 8 x i16> %b, i32 0)
+  %res = call <vscale x 8 x i16> @llvm.vector.splice.nxv8i16(<vscale x 8 x i16> %a, <vscale x 8 x i16> %b, i32 0)
   ret <vscale x 8 x i16> %res
 }
 
@@ -914,7 +914,7 @@ define <vscale x 8 x i16> @splice_nxv8i16_offset_negone(<vscale x 8 x i16> %a, <
 ; CHECK-NEXT:    vsetvli a0, zero, e16, m2, ta, ma
 ; CHECK-NEXT:    vslideup.vi v8, v10, 1
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 8 x i16> @llvm.experimental.vector.splice.nxv8i16(<vscale x 8 x i16> %a, <vscale x 8 x i16> %b, i32 -1)
+  %res = call <vscale x 8 x i16> @llvm.vector.splice.nxv8i16(<vscale x 8 x i16> %a, <vscale x 8 x i16> %b, i32 -1)
   ret <vscale x 8 x i16> %res
 }
 
@@ -928,7 +928,7 @@ define <vscale x 8 x i16> @splice_nxv8i16_offset_min(<vscale x 8 x i16> %a, <vsc
 ; CHECK-NEXT:    vsetvli a0, zero, e16, m2, ta, ma
 ; CHECK-NEXT:    vslideup.vi v8, v10, 16
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 8 x i16> @llvm.experimental.vector.splice.nxv8i16(<vscale x 8 x i16> %a, <vscale x 8 x i16> %b, i32 -16)
+  %res = call <vscale x 8 x i16> @llvm.vector.splice.nxv8i16(<vscale x 8 x i16> %a, <vscale x 8 x i16> %b, i32 -16)
   ret <vscale x 8 x i16> %res
 }
 
@@ -942,17 +942,17 @@ define <vscale x 8 x i16> @splice_nxv8i16_offset_max(<vscale x 8 x i16> %a, <vsc
 ; CHECK-NEXT:    vsetvli a1, zero, e16, m2, ta, ma
 ; CHECK-NEXT:    vslideup.vx v8, v10, a0
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 8 x i16> @llvm.experimental.vector.splice.nxv8i16(<vscale x 8 x i16> %a, <vscale x 8 x i16> %b, i32 15)
+  %res = call <vscale x 8 x i16> @llvm.vector.splice.nxv8i16(<vscale x 8 x i16> %a, <vscale x 8 x i16> %b, i32 15)
   ret <vscale x 8 x i16> %res
 }
 
-declare <vscale x 16 x i16> @llvm.experimental.vector.splice.nxv16i16(<vscale x 16 x i16>, <vscale x 16 x i16>, i32)
+declare <vscale x 16 x i16> @llvm.vector.splice.nxv16i16(<vscale x 16 x i16>, <vscale x 16 x i16>, i32)
 
 define <vscale x 16 x i16> @splice_nxv16i16_offset_zero(<vscale x 16 x i16> %a, <vscale x 16 x i16> %b) #0 {
 ; CHECK-LABEL: splice_nxv16i16_offset_zero:
 ; CHECK:       # %bb.0:
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 16 x i16> @llvm.experimental.vector.splice.nxv16i16(<vscale x 16 x i16> %a, <vscale x 16 x i16> %b, i32 0)
+  %res = call <vscale x 16 x i16> @llvm.vector.splice.nxv16i16(<vscale x 16 x i16> %a, <vscale x 16 x i16> %b, i32 0)
   ret <vscale x 16 x i16> %res
 }
 
@@ -967,7 +967,7 @@ define <vscale x 16 x i16> @splice_nxv16i16_offset_negone(<vscale x 16 x i16> %a
 ; CHECK-NEXT:    vsetvli a0, zero, e16, m4, ta, ma
 ; CHECK-NEXT:    vslideup.vi v8, v12, 1
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 16 x i16> @llvm.experimental.vector.splice.nxv16i16(<vscale x 16 x i16> %a, <vscale x 16 x i16> %b, i32 -1)
+  %res = call <vscale x 16 x i16> @llvm.vector.splice.nxv16i16(<vscale x 16 x i16> %a, <vscale x 16 x i16> %b, i32 -1)
   ret <vscale x 16 x i16> %res
 }
 
@@ -983,7 +983,7 @@ define <vscale x 16 x i16> @splice_nxv16i16_offset_min(<vscale x 16 x i16> %a, <
 ; CHECK-NEXT:    vsetvli a0, zero, e16, m4, ta, ma
 ; CHECK-NEXT:    vslideup.vx v8, v12, a1
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 16 x i16> @llvm.experimental.vector.splice.nxv16i16(<vscale x 16 x i16> %a, <vscale x 16 x i16> %b, i32 -32)
+  %res = call <vscale x 16 x i16> @llvm.vector.splice.nxv16i16(<vscale x 16 x i16> %a, <vscale x 16 x i16> %b, i32 -32)
   ret <vscale x 16 x i16> %res
 }
 
@@ -998,17 +998,17 @@ define <vscale x 16 x i16> @splice_nxv16i16_offset_max(<vscale x 16 x i16> %a, <
 ; CHECK-NEXT:    vsetvli a1, zero, e16, m4, ta, ma
 ; CHECK-NEXT:    vslideup.vx v8, v12, a0
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 16 x i16> @llvm.experimental.vector.splice.nxv16i16(<vscale x 16 x i16> %a, <vscale x 16 x i16> %b, i32 31)
+  %res = call <vscale x 16 x i16> @llvm.vector.splice.nxv16i16(<vscale x 16 x i16> %a, <vscale x 16 x i16> %b, i32 31)
   ret <vscale x 16 x i16> %res
 }
 
-declare <vscale x 32 x i16> @llvm.experimental.vector.splice.nxv32i16(<vscale x 32 x i16>, <vscale x 32 x i16>, i32)
+declare <vscale x 32 x i16> @llvm.vector.splice.nxv32i16(<vscale x 32 x i16>, <vscale x 32 x i16>, i32)
 
 define <vscale x 32 x i16> @splice_nxv32i16_offset_zero(<vscale x 32 x i16> %a, <vscale x 32 x i16> %b) #0 {
 ; CHECK-LABEL: splice_nxv32i16_offset_zero:
 ; CHECK:       # %bb.0:
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 32 x i16> @llvm.experimental.vector.splice.nxv32i16(<vscale x 32 x i16> %a, <vscale x 32 x i16> %b, i32 0)
+  %res = call <vscale x 32 x i16> @llvm.vector.splice.nxv32i16(<vscale x 32 x i16> %a, <vscale x 32 x i16> %b, i32 0)
   ret <vscale x 32 x i16> %res
 }
 
@@ -1023,7 +1023,7 @@ define <vscale x 32 x i16> @splice_nxv32i16_offset_negone(<vscale x 32 x i16> %a
 ; CHECK-NEXT:    vsetvli a0, zero, e16, m8, ta, ma
 ; CHECK-NEXT:    vslideup.vi v8, v16, 1
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 32 x i16> @llvm.experimental.vector.splice.nxv32i16(<vscale x 32 x i16> %a, <vscale x 32 x i16> %b, i32 -1)
+  %res = call <vscale x 32 x i16> @llvm.vector.splice.nxv32i16(<vscale x 32 x i16> %a, <vscale x 32 x i16> %b, i32 -1)
   ret <vscale x 32 x i16> %res
 }
 
@@ -1039,7 +1039,7 @@ define <vscale x 32 x i16> @splice_nxv32i16_offset_min(<vscale x 32 x i16> %a, <
 ; CHECK-NEXT:    vsetvli a0, zero, e16, m8, ta, ma
 ; CHECK-NEXT:    vslideup.vx v8, v16, a1
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 32 x i16> @llvm.experimental.vector.splice.nxv32i16(<vscale x 32 x i16> %a, <vscale x 32 x i16> %b, i32 -64)
+  %res = call <vscale x 32 x i16> @llvm.vector.splice.nxv32i16(<vscale x 32 x i16> %a, <vscale x 32 x i16> %b, i32 -64)
   ret <vscale x 32 x i16> %res
 }
 
@@ -1055,17 +1055,17 @@ define <vscale x 32 x i16> @splice_nxv32i16_offset_max(<vscale x 32 x i16> %a, <
 ; CHECK-NEXT:    vsetvli a1, zero, e16, m8, ta, ma
 ; CHECK-NEXT:    vslideup.vx v8, v16, a0
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 32 x i16> @llvm.experimental.vector.splice.nxv32i16(<vscale x 32 x i16> %a, <vscale x 32 x i16> %b, i32 63)
+  %res = call <vscale x 32 x i16> @llvm.vector.splice.nxv32i16(<vscale x 32 x i16> %a, <vscale x 32 x i16> %b, i32 63)
   ret <vscale x 32 x i16> %res
 }
 
-declare <vscale x 1 x i32> @llvm.experimental.vector.splice.nxv1i32(<vscale x 1 x i32>, <vscale x 1 x i32>, i32)
+declare <vscale x 1 x i32> @llvm.vector.splice.nxv1i32(<vscale x 1 x i32>, <vscale x 1 x i32>, i32)
 
 define <vscale x 1 x i32> @splice_nxv1i32_offset_zero(<vscale x 1 x i32> %a, <vscale x 1 x i32> %b) #0 {
 ; CHECK-LABEL: splice_nxv1i32_offset_zero:
 ; CHECK:       # %bb.0:
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 1 x i32> @llvm.experimental.vector.splice.nxv1i32(<vscale x 1 x i32> %a, <vscale x 1 x i32> %b, i32 0)
+  %res = call <vscale x 1 x i32> @llvm.vector.splice.nxv1i32(<vscale x 1 x i32> %a, <vscale x 1 x i32> %b, i32 0)
   ret <vscale x 1 x i32> %res
 }
 
@@ -1080,7 +1080,7 @@ define <vscale x 1 x i32> @splice_nxv1i32_offset_negone(<vscale x 1 x i32> %a, <
 ; CHECK-NEXT:    vsetvli a0, zero, e32, mf2, ta, ma
 ; CHECK-NEXT:    vslideup.vi v8, v9, 1
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 1 x i32> @llvm.experimental.vector.splice.nxv1i32(<vscale x 1 x i32> %a, <vscale x 1 x i32> %b, i32 -1)
+  %res = call <vscale x 1 x i32> @llvm.vector.splice.nxv1i32(<vscale x 1 x i32> %a, <vscale x 1 x i32> %b, i32 -1)
   ret <vscale x 1 x i32> %res
 }
 
@@ -1095,7 +1095,7 @@ define <vscale x 1 x i32> @splice_nxv1i32_offset_min(<vscale x 1 x i32> %a, <vsc
 ; CHECK-NEXT:    vsetvli a0, zero, e32, mf2, ta, ma
 ; CHECK-NEXT:    vslideup.vi v8, v9, 2
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 1 x i32> @llvm.experimental.vector.splice.nxv1i32(<vscale x 1 x i32> %a, <vscale x 1 x i32> %b, i32 -2)
+  %res = call <vscale x 1 x i32> @llvm.vector.splice.nxv1i32(<vscale x 1 x i32> %a, <vscale x 1 x i32> %b, i32 -2)
   ret <vscale x 1 x i32> %res
 }
 
@@ -1110,17 +1110,17 @@ define <vscale x 1 x i32> @splice_nxv1i32_offset_max(<vscale x 1 x i32> %a, <vsc
 ; CHECK-NEXT:    vsetvli a1, zero, e32, mf2, ta, ma
 ; CHECK-NEXT:    vslideup.vx v8, v9, a0
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 1 x i32> @llvm.experimental.vector.splice.nxv1i32(<vscale x 1 x i32> %a, <vscale x 1 x i32> %b, i32 1)
+  %res = call <vscale x 1 x i32> @llvm.vector.splice.nxv1i32(<vscale x 1 x i32> %a, <vscale x 1 x i32> %b, i32 1)
   ret <vscale x 1 x i32> %res
 }
 
-declare <vscale x 2 x i32> @llvm.experimental.vector.splice.nxv2i32(<vscale x 2 x i32>, <vscale x 2 x i32>, i32)
+declare <vscale x 2 x i32> @llvm.vector.splice.nxv2i32(<vscale x 2 x i32>, <vscale x 2 x i32>, i32)
 
 define <vscale x 2 x i32> @splice_nxv2i32_offset_zero(<vscale x 2 x i32> %a, <vscale x 2 x i32> %b) #0 {
 ; CHECK-LABEL: splice_nxv2i32_offset_zero:
 ; CHECK:       # %bb.0:
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 2 x i32> @llvm.experimental.vector.splice.nxv2i32(<vscale x 2 x i32> %a, <vscale x 2 x i32> %b, i32 0)
+  %res = call <vscale x 2 x i32> @llvm.vector.splice.nxv2i32(<vscale x 2 x i32> %a, <vscale x 2 x i32> %b, i32 0)
   ret <vscale x 2 x i32> %res
 }
 
@@ -1135,7 +1135,7 @@ define <vscale x 2 x i32> @splice_nxv2i32_offset_negone(<vscale x 2 x i32> %a, <
 ; CHECK-NEXT:    vsetvli a0, zero, e32, m1, ta, ma
 ; CHECK-NEXT:    vslideup.vi v8, v9, 1
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 2 x i32> @llvm.experimental.vector.splice.nxv2i32(<vscale x 2 x i32> %a, <vscale x 2 x i32> %b, i32 -1)
+  %res = call <vscale x 2 x i32> @llvm.vector.splice.nxv2i32(<vscale x 2 x i32> %a, <vscale x 2 x i32> %b, i32 -1)
   ret <vscale x 2 x i32> %res
 }
 
@@ -1150,7 +1150,7 @@ define <vscale x 2 x i32> @splice_nxv2i32_offset_min(<vscale x 2 x i32> %a, <vsc
 ; CHECK-NEXT:    vsetvli a0, zero, e32, m1, ta, ma
 ; CHECK-NEXT:    vslideup.vi v8, v9, 4
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 2 x i32> @llvm.experimental.vector.splice.nxv2i32(<vscale x 2 x i32> %a, <vscale x 2 x i32> %b, i32 -4)
+  %res = call <vscale x 2 x i32> @llvm.vector.splice.nxv2i32(<vscale x 2 x i32> %a, <vscale x 2 x i32> %b, i32 -4)
   ret <vscale x 2 x i32> %res
 }
 
@@ -1165,17 +1165,17 @@ define <vscale x 2 x i32> @splice_nxv2i32_offset_max(<vscale x 2 x i32> %a, <vsc
 ; CHECK-NEXT:    vsetvli a1, zero, e32, m1, ta, ma
 ; CHECK-NEXT:    vslideup.vx v8, v9, a0
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 2 x i32> @llvm.experimental.vector.splice.nxv2i32(<vscale x 2 x i32> %a, <vscale x 2 x i32> %b, i32 3)
+  %res = call <vscale x 2 x i32> @llvm.vector.splice.nxv2i32(<vscale x 2 x i32> %a, <vscale x 2 x i32> %b, i32 3)
   ret <vscale x 2 x i32> %res
 }
 
-declare <vscale x 4 x i32> @llvm.experimental.vector.splice.nxv4i32(<vscale x 4 x i32>, <vscale x 4 x i32>, i32)
+declare <vscale x 4 x i32> @llvm.vector.splice.nxv4i32(<vscale x 4 x i32>, <vscale x 4 x i32>, i32)
 
 define <vscale x 4 x i32> @splice_nxv4i32_offset_zero(<vscale x 4 x i32> %a, <vscale x 4 x i32> %b) #0 {
 ; CHECK-LABEL: splice_nxv4i32_offset_zero:
 ; CHECK:       # %bb.0:
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 4 x i32> @llvm.experimental.vector.splice.nxv4i32(<vscale x 4 x i32> %a, <vscale x 4 x i32> %b, i32 0)
+  %res = call <vscale x 4 x i32> @llvm.vector.splice.nxv4i32(<vscale x 4 x i32> %a, <vscale x 4 x i32> %b, i32 0)
   ret <vscale x 4 x i32> %res
 }
 
@@ -1190,7 +1190,7 @@ define <vscale x 4 x i32> @splice_nxv4i32_offset_negone(<vscale x 4 x i32> %a, <
 ; CHECK-NEXT:    vsetvli a0, zero, e32, m2, ta, ma
 ; CHECK-NEXT:    vslideup.vi v8, v10, 1
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 4 x i32> @llvm.experimental.vector.splice.nxv4i32(<vscale x 4 x i32> %a, <vscale x 4 x i32> %b, i32 -1)
+  %res = call <vscale x 4 x i32> @llvm.vector.splice.nxv4i32(<vscale x 4 x i32> %a, <vscale x 4 x i32> %b, i32 -1)
   ret <vscale x 4 x i32> %res
 }
 
@@ -1205,7 +1205,7 @@ define <vscale x 4 x i32> @splice_nxv4i32_offset_min(<vscale x 4 x i32> %a, <vsc
 ; CHECK-NEXT:    vsetvli a0, zero, e32, m2, ta, ma
 ; CHECK-NEXT:    vslideup.vi v8, v10, 8
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 4 x i32> @llvm.experimental.vector.splice.nxv4i32(<vscale x 4 x i32> %a, <vscale x 4 x i32> %b, i32 -8)
+  %res = call <vscale x 4 x i32> @llvm.vector.splice.nxv4i32(<vscale x 4 x i32> %a, <vscale x 4 x i32> %b, i32 -8)
   ret <vscale x 4 x i32> %res
 }
 
@@ -1220,17 +1220,17 @@ define <vscale x 4 x i32> @splice_nxv4i32_offset_max(<vscale x 4 x i32> %a, <vsc
 ; CHECK-NEXT:    vsetvli a1, zero, e32, m2, ta, ma
 ; CHECK-NEXT:    vslideup.vx v8, v10, a0
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 4 x i32> @llvm.experimental.vector.splice.nxv4i32(<vscale x 4 x i32> %a, <vscale x 4 x i32> %b, i32 7)
+  %res = call <vscale x 4 x i32> @llvm.vector.splice.nxv4i32(<vscale x 4 x i32> %a, <vscale x 4 x i32> %b, i32 7)
   ret <vscale x 4 x i32> %res
 }
 
-declare <vscale x 8 x i32> @llvm.experimental.vector.splice.nxv8i32(<vscale x 8 x i32>, <vscale x 8 x i32>, i32)
+declare <vscale x 8 x i32> @llvm.vector.splice.nxv8i32(<vscale x 8 x i32>, <vscale x 8 x i32>, i32)
 
 define <vscale x 8 x i32> @splice_nxv8i32_offset_zero(<vscale x 8 x i32> %a, <vscale x 8 x i32> %b) #0 {
 ; CHECK-LABEL: splice_nxv8i32_offset_zero:
 ; CHECK:       # %bb.0:
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 8 x i32> @llvm.experimental.vector.splice.nxv8i32(<vscale x 8 x i32> %a, <vscale x 8 x i32> %b, i32 0)
+  %res = call <vscale x 8 x i32> @llvm.vector.splice.nxv8i32(<vscale x 8 x i32> %a, <vscale x 8 x i32> %b, i32 0)
   ret <vscale x 8 x i32> %res
 }
 
@@ -1244,7 +1244,7 @@ define <vscale x 8 x i32> @splice_nxv8i32_offset_negone(<vscale x 8 x i32> %a, <
 ; CHECK-NEXT:    vsetvli a0, zero, e32, m4, ta, ma
 ; CHECK-NEXT:    vslideup.vi v8, v12, 1
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 8 x i32> @llvm.experimental.vector.splice.nxv8i32(<vscale x 8 x i32> %a, <vscale x 8 x i32> %b, i32 -1)
+  %res = call <vscale x 8 x i32> @llvm.vector.splice.nxv8i32(<vscale x 8 x i32> %a, <vscale x 8 x i32> %b, i32 -1)
   ret <vscale x 8 x i32> %res
 }
 
@@ -1258,7 +1258,7 @@ define <vscale x 8 x i32> @splice_nxv8i32_offset_min(<vscale x 8 x i32> %a, <vsc
 ; CHECK-NEXT:    vsetvli a0, zero, e32, m4, ta, ma
 ; CHECK-NEXT:    vslideup.vi v8, v12, 16
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 8 x i32> @llvm.experimental.vector.splice.nxv8i32(<vscale x 8 x i32> %a, <vscale x 8 x i32> %b, i32 -16)
+  %res = call <vscale x 8 x i32> @llvm.vector.splice.nxv8i32(<vscale x 8 x i32> %a, <vscale x 8 x i32> %b, i32 -16)
   ret <vscale x 8 x i32> %res
 }
 
@@ -1272,17 +1272,17 @@ define <vscale x 8 x i32> @splice_nxv8i32_offset_max(<vscale x 8 x i32> %a, <vsc
 ; CHECK-NEXT:    vsetvli a1, zero, e32, m4, ta, ma
 ; CHECK-NEXT:    vslideup.vx v8, v12, a0
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 8 x i32> @llvm.experimental.vector.splice.nxv8i32(<vscale x 8 x i32> %a, <vscale x 8 x i32> %b, i32 15)
+  %res = call <vscale x 8 x i32> @llvm.vector.splice.nxv8i32(<vscale x 8 x i32> %a, <vscale x 8 x i32> %b, i32 15)
   ret <vscale x 8 x i32> %res
 }
 
-declare <vscale x 16 x i32> @llvm.experimental.vector.splice.nxv16i32(<vscale x 16 x i32>, <vscale x 16 x i32>, i32)
+declare <vscale x 16 x i32> @llvm.vector.splice.nxv16i32(<vscale x 16 x i32>, <vscale x 16 x i32>, i32)
 
 define <vscale x 16 x i32> @splice_nxv16i32_offset_zero(<vscale x 16 x i32> %a, <vscale x 16 x i32> %b) #0 {
 ; CHECK-LABEL: splice_nxv16i32_offset_zero:
 ; CHECK:       # %bb.0:
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 16 x i32> @llvm.experimental.vector.splice.nxv16i32(<vscale x 16 x i32> %a, <vscale x 16 x i32> %b, i32 0)
+  %res = call <vscale x 16 x i32> @llvm.vector.splice.nxv16i32(<vscale x 16 x i32> %a, <vscale x 16 x i32> %b, i32 0)
   ret <vscale x 16 x i32> %res
 }
 
@@ -1297,7 +1297,7 @@ define <vscale x 16 x i32> @splice_nxv16i32_offset_negone(<vscale x 16 x i32> %a
 ; CHECK-NEXT:    vsetvli a0, zero, e32, m8, ta, ma
 ; CHECK-NEXT:    vslideup.vi v8, v16, 1
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 16 x i32> @llvm.experimental.vector.splice.nxv16i32(<vscale x 16 x i32> %a, <vscale x 16 x i32> %b, i32 -1)
+  %res = call <vscale x 16 x i32> @llvm.vector.splice.nxv16i32(<vscale x 16 x i32> %a, <vscale x 16 x i32> %b, i32 -1)
   ret <vscale x 16 x i32> %res
 }
 
@@ -1313,7 +1313,7 @@ define <vscale x 16 x i32> @splice_nxv16i32_offset_min(<vscale x 16 x i32> %a, <
 ; CHECK-NEXT:    vsetvli a0, zero, e32, m8, ta, ma
 ; CHECK-NEXT:    vslideup.vx v8, v16, a1
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 16 x i32> @llvm.experimental.vector.splice.nxv16i32(<vscale x 16 x i32> %a, <vscale x 16 x i32> %b, i32 -32)
+  %res = call <vscale x 16 x i32> @llvm.vector.splice.nxv16i32(<vscale x 16 x i32> %a, <vscale x 16 x i32> %b, i32 -32)
   ret <vscale x 16 x i32> %res
 }
 
@@ -1328,17 +1328,17 @@ define <vscale x 16 x i32> @splice_nxv16i32_offset_max(<vscale x 16 x i32> %a, <
 ; CHECK-NEXT:    vsetvli a1, zero, e32, m8, ta, ma
 ; CHECK-NEXT:    vslideup.vx v8, v16, a0
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 16 x i32> @llvm.experimental.vector.splice.nxv16i32(<vscale x 16 x i32> %a, <vscale x 16 x i32> %b, i32 31)
+  %res = call <vscale x 16 x i32> @llvm.vector.splice.nxv16i32(<vscale x 16 x i32> %a, <vscale x 16 x i32> %b, i32 31)
   ret <vscale x 16 x i32> %res
 }
 
-declare <vscale x 1 x i64> @llvm.experimental.vector.splice.nxv1i64(<vscale x 1 x i64>, <vscale x 1 x i64>, i32)
+declare <vscale x 1 x i64> @llvm.vector.splice.nxv1i64(<vscale x 1 x i64>, <vscale x 1 x i64>, i32)
 
 define <vscale x 1 x i64> @splice_nxv1i64_offset_zero(<vscale x 1 x i64> %a, <vscale x 1 x i64> %b) #0 {
 ; CHECK-LABEL: splice_nxv1i64_offset_zero:
 ; CHECK:       # %bb.0:
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 1 x i64> @llvm.experimental.vector.splice.nxv1i64(<vscale x 1 x i64> %a, <vscale x 1 x i64> %b, i32 0)
+  %res = call <vscale x 1 x i64> @llvm.vector.splice.nxv1i64(<vscale x 1 x i64> %a, <vscale x 1 x i64> %b, i32 0)
   ret <vscale x 1 x i64> %res
 }
 
@@ -1353,7 +1353,7 @@ define <vscale x 1 x i64> @splice_nxv1i64_offset_negone(<vscale x 1 x i64> %a, <
 ; CHECK-NEXT:    vsetvli a0, zero, e64, m1, ta, ma
 ; CHECK-NEXT:    vslideup.vi v8, v9, 1
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 1 x i64> @llvm.experimental.vector.splice.nxv1i64(<vscale x 1 x i64> %a, <vscale x 1 x i64> %b, i32 -1)
+  %res = call <vscale x 1 x i64> @llvm.vector.splice.nxv1i64(<vscale x 1 x i64> %a, <vscale x 1 x i64> %b, i32 -1)
   ret <vscale x 1 x i64> %res
 }
 
@@ -1368,7 +1368,7 @@ define <vscale x 1 x i64> @splice_nxv1i64_offset_min(<vscale x 1 x i64> %a, <vsc
 ; CHECK-NEXT:    vsetvli a0, zero, e64, m1, ta, ma
 ; CHECK-NEXT:    vslideup.vi v8, v9, 2
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 1 x i64> @llvm.experimental.vector.splice.nxv1i64(<vscale x 1 x i64> %a, <vscale x 1 x i64> %b, i32 -2)
+  %res = call <vscale x 1 x i64> @llvm.vector.splice.nxv1i64(<vscale x 1 x i64> %a, <vscale x 1 x i64> %b, i32 -2)
   ret <vscale x 1 x i64> %res
 }
 
@@ -1383,17 +1383,17 @@ define <vscale x 1 x i64> @splice_nxv1i64_offset_max(<vscale x 1 x i64> %a, <vsc
 ; CHECK-NEXT:    vsetvli a1, zero, e64, m1, ta, ma
 ; CHECK-NEXT:    vslideup.vx v8, v9, a0
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 1 x i64> @llvm.experimental.vector.splice.nxv1i64(<vscale x 1 x i64> %a, <vscale x 1 x i64> %b, i32 1)
+  %res = call <vscale x 1 x i64> @llvm.vector.splice.nxv1i64(<vscale x 1 x i64> %a, <vscale x 1 x i64> %b, i32 1)
   ret <vscale x 1 x i64> %res
 }
 
-declare <vscale x 2 x i64> @llvm.experimental.vector.splice.nxv2i64(<vscale x 2 x i64>, <vscale x 2 x i64>, i32)
+declare <vscale x 2 x i64> @llvm.vector.splice.nxv2i64(<vscale x 2 x i64>, <vscale x 2 x i64>, i32)
 
 define <vscale x 2 x i64> @splice_nxv2i64_offset_zero(<vscale x 2 x i64> %a, <vscale x 2 x i64> %b) #0 {
 ; CHECK-LABEL: splice_nxv2i64_offset_zero:
 ; CHECK:       # %bb.0:
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 2 x i64> @llvm.experimental.vector.splice.nxv2i64(<vscale x 2 x i64> %a, <vscale x 2 x i64> %b, i32 0)
+  %res = call <vscale x 2 x i64> @llvm.vector.splice.nxv2i64(<vscale x 2 x i64> %a, <vscale x 2 x i64> %b, i32 0)
   ret <vscale x 2 x i64> %res
 }
 
@@ -1408,7 +1408,7 @@ define <vscale x 2 x i64> @splice_nxv2i64_offset_negone(<vscale x 2 x i64> %a, <
 ; CHECK-NEXT:    vsetvli a0, zero, e64, m2, ta, ma
 ; CHECK-NEXT:    vslideup.vi v8, v10, 1
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 2 x i64> @llvm.experimental.vector.splice.nxv2i64(<vscale x 2 x i64> %a, <vscale x 2 x i64> %b, i32 -1)
+  %res = call <vscale x 2 x i64> @llvm.vector.splice.nxv2i64(<vscale x 2 x i64> %a, <vscale x 2 x i64> %b, i32 -1)
   ret <vscale x 2 x i64> %res
 }
 
@@ -1423,7 +1423,7 @@ define <vscale x 2 x i64> @splice_nxv2i64_offset_min(<vscale x 2 x i64> %a, <vsc
 ; CHECK-NEXT:    vsetvli a0, zero, e64, m2, ta, ma
 ; CHECK-NEXT:    vslideup.vi v8, v10, 4
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 2 x i64> @llvm.experimental.vector.splice.nxv2i64(<vscale x 2 x i64> %a, <vscale x 2 x i64> %b, i32 -4)
+  %res = call <vscale x 2 x i64> @llvm.vector.splice.nxv2i64(<vscale x 2 x i64> %a, <vscale x 2 x i64> %b, i32 -4)
   ret <vscale x 2 x i64> %res
 }
 
@@ -1438,17 +1438,17 @@ define <vscale x 2 x i64> @splice_nxv2i64_offset_max(<vscale x 2 x i64> %a, <vsc
 ; CHECK-NEXT:    vsetvli a1, zero, e64, m2, ta, ma
 ; CHECK-NEXT:    vslideup.vx v8, v10, a0
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 2 x i64> @llvm.experimental.vector.splice.nxv2i64(<vscale x 2 x i64> %a, <vscale x 2 x i64> %b, i32 3)
+  %res = call <vscale x 2 x i64> @llvm.vector.splice.nxv2i64(<vscale x 2 x i64> %a, <vscale x 2 x i64> %b, i32 3)
   ret <vscale x 2 x i64> %res
 }
 
-declare <vscale x 4 x i64> @llvm.experimental.vector.splice.nxv4i64(<vscale x 4 x i64>, <vscale x 4 x i64>, i32)
+declare <vscale x 4 x i64> @llvm.vector.splice.nxv4i64(<vscale x 4 x i64>, <vscale x 4 x i64>, i32)
 
 define <vscale x 4 x i64> @splice_nxv4i64_offset_zero(<vscale x 4 x i64> %a, <vscale x 4 x i64> %b) #0 {
 ; CHECK-LABEL: splice_nxv4i64_offset_zero:
 ; CHECK:       # %bb.0:
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 4 x i64> @llvm.experimental.vector.splice.nxv4i64(<vscale x 4 x i64> %a, <vscale x 4 x i64> %b, i32 0)
+  %res = call <vscale x 4 x i64> @llvm.vector.splice.nxv4i64(<vscale x 4 x i64> %a, <vscale x 4 x i64> %b, i32 0)
   ret <vscale x 4 x i64> %res
 }
 
@@ -1463,7 +1463,7 @@ define <vscale x 4 x i64> @splice_nxv4i64_offset_negone(<vscale x 4 x i64> %a, <
 ; CHECK-NEXT:    vsetvli a0, zero, e64, m4, ta, ma
 ; CHECK-NEXT:    vslideup.vi v8, v12, 1
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 4 x i64> @llvm.experimental.vector.splice.nxv4i64(<vscale x 4 x i64> %a, <vscale x 4 x i64> %b, i32 -1)
+  %res = call <vscale x 4 x i64> @llvm.vector.splice.nxv4i64(<vscale x 4 x i64> %a, <vscale x 4 x i64> %b, i32 -1)
   ret <vscale x 4 x i64> %res
 }
 
@@ -1478,7 +1478,7 @@ define <vscale x 4 x i64> @splice_nxv4i64_offset_min(<vscale x 4 x i64> %a, <vsc
 ; CHECK-NEXT:    vsetvli a0, zero, e64, m4, ta, ma
 ; CHECK-NEXT:    vslideup.vi v8, v12, 8
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 4 x i64> @llvm.experimental.vector.splice.nxv4i64(<vscale x 4 x i64> %a, <vscale x 4 x i64> %b, i32 -8)
+  %res = call <vscale x 4 x i64> @llvm.vector.splice.nxv4i64(<vscale x 4 x i64> %a, <vscale x 4 x i64> %b, i32 -8)
   ret <vscale x 4 x i64> %res
 }
 
@@ -1493,17 +1493,17 @@ define <vscale x 4 x i64> @splice_nxv4i64_offset_max(<vscale x 4 x i64> %a, <vsc
 ; CHECK-NEXT:    vsetvli a1, zero, e64, m4, ta, ma
 ; CHECK-NEXT:    vslideup.vx v8, v12, a0
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 4 x i64> @llvm.experimental.vector.splice.nxv4i64(<vscale x 4 x i64> %a, <vscale x 4 x i64> %b, i32 7)
+  %res = call <vscale x 4 x i64> @llvm.vector.splice.nxv4i64(<vscale x 4 x i64> %a, <vscale x 4 x i64> %b, i32 7)
   ret <vscale x 4 x i64> %res
 }
 
-declare <vscale x 8 x i64> @llvm.experimental.vector.splice.nxv8i64(<vscale x 8 x i64>, <vscale x 8 x i64>, i32)
+declare <vscale x 8 x i64> @llvm.vector.splice.nxv8i64(<vscale x 8 x i64>, <vscale x 8 x i64>, i32)
 
 define <vscale x 8 x i64> @splice_nxv8i64_offset_zero(<vscale x 8 x i64> %a, <vscale x 8 x i64> %b) #0 {
 ; CHECK-LABEL: splice_nxv8i64_offset_zero:
 ; CHECK:       # %bb.0:
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 8 x i64> @llvm.experimental.vector.splice.nxv8i64(<vscale x 8 x i64> %a, <vscale x 8 x i64> %b, i32 0)
+  %res = call <vscale x 8 x i64> @llvm.vector.splice.nxv8i64(<vscale x 8 x i64> %a, <vscale x 8 x i64> %b, i32 0)
   ret <vscale x 8 x i64> %res
 }
 
@@ -1517,7 +1517,7 @@ define <vscale x 8 x i64> @splice_nxv8i64_offset_negone(<vscale x 8 x i64> %a, <
 ; CHECK-NEXT:    vsetvli a0, zero, e64, m8, ta, ma
 ; CHECK-NEXT:    vslideup.vi v8, v16, 1
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 8 x i64> @llvm.experimental.vector.splice.nxv8i64(<vscale x 8 x i64> %a, <vscale x 8 x i64> %b, i32 -1)
+  %res = call <vscale x 8 x i64> @llvm.vector.splice.nxv8i64(<vscale x 8 x i64> %a, <vscale x 8 x i64> %b, i32 -1)
   ret <vscale x 8 x i64> %res
 }
 
@@ -1531,7 +1531,7 @@ define <vscale x 8 x i64> @splice_nxv8i64_offset_min(<vscale x 8 x i64> %a, <vsc
 ; CHECK-NEXT:    vsetvli a0, zero, e64, m8, ta, ma
 ; CHECK-NEXT:    vslideup.vi v8, v16, 16
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 8 x i64> @llvm.experimental.vector.splice.nxv8i64(<vscale x 8 x i64> %a, <vscale x 8 x i64> %b, i32 -16)
+  %res = call <vscale x 8 x i64> @llvm.vector.splice.nxv8i64(<vscale x 8 x i64> %a, <vscale x 8 x i64> %b, i32 -16)
   ret <vscale x 8 x i64> %res
 }
 
@@ -1545,17 +1545,17 @@ define <vscale x 8 x i64> @splice_nxv8i64_offset_max(<vscale x 8 x i64> %a, <vsc
 ; CHECK-NEXT:    vsetvli a1, zero, e64, m8, ta, ma
 ; CHECK-NEXT:    vslideup.vx v8, v16, a0
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 8 x i64> @llvm.experimental.vector.splice.nxv8i64(<vscale x 8 x i64> %a, <vscale x 8 x i64> %b, i32 15)
+  %res = call <vscale x 8 x i64> @llvm.vector.splice.nxv8i64(<vscale x 8 x i64> %a, <vscale x 8 x i64> %b, i32 15)
   ret <vscale x 8 x i64> %res
 }
 
-declare <vscale x 1 x half> @llvm.experimental.vector.splice.nxv1f16(<vscale x 1 x half>, <vscale x 1 x half>, i32)
+declare <vscale x 1 x half> @llvm.vector.splice.nxv1f16(<vscale x 1 x half>, <vscale x 1 x half>, i32)
 
 define <vscale x 1 x half> @splice_nxv1f16_offset_zero(<vscale x 1 x half> %a, <vscale x 1 x half> %b) #0 {
 ; CHECK-LABEL: splice_nxv1f16_offset_zero:
 ; CHECK:       # %bb.0:
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 1 x half> @llvm.experimental.vector.splice.nxv1f16(<vscale x 1 x half> %a, <vscale x 1 x half> %b, i32 0)
+  %res = call <vscale x 1 x half> @llvm.vector.splice.nxv1f16(<vscale x 1 x half> %a, <vscale x 1 x half> %b, i32 0)
   ret <vscale x 1 x half> %res
 }
 
@@ -1570,7 +1570,7 @@ define <vscale x 1 x half> @splice_nxv1f16_offset_negone(<vscale x 1 x half> %a,
 ; CHECK-NEXT:    vsetvli a0, zero, e16, mf4, ta, ma
 ; CHECK-NEXT:    vslideup.vi v8, v9, 1
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 1 x half> @llvm.experimental.vector.splice.nxv1f16(<vscale x 1 x half> %a, <vscale x 1 x half> %b, i32 -1)
+  %res = call <vscale x 1 x half> @llvm.vector.splice.nxv1f16(<vscale x 1 x half> %a, <vscale x 1 x half> %b, i32 -1)
   ret <vscale x 1 x half> %res
 }
 
@@ -1585,7 +1585,7 @@ define <vscale x 1 x half> @splice_nxv1f16_offset_min(<vscale x 1 x half> %a, <v
 ; CHECK-NEXT:    vsetvli a0, zero, e16, mf4, ta, ma
 ; CHECK-NEXT:    vslideup.vi v8, v9, 2
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 1 x half> @llvm.experimental.vector.splice.nxv1f16(<vscale x 1 x half> %a, <vscale x 1 x half> %b, i32 -2)
+  %res = call <vscale x 1 x half> @llvm.vector.splice.nxv1f16(<vscale x 1 x half> %a, <vscale x 1 x half> %b, i32 -2)
   ret <vscale x 1 x half> %res
 }
 
@@ -1600,17 +1600,17 @@ define <vscale x 1 x half> @splice_nxv1f16_offset_max(<vscale x 1 x half> %a, <v
 ; CHECK-NEXT:    vsetvli a1, zero, e16, mf4, ta, ma
 ; CHECK-NEXT:    vslideup.vx v8, v9, a0
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 1 x half> @llvm.experimental.vector.splice.nxv1f16(<vscale x 1 x half> %a, <vscale x 1 x half> %b, i32 1)
+  %res = call <vscale x 1 x half> @llvm.vector.splice.nxv1f16(<vscale x 1 x half> %a, <vscale x 1 x half> %b, i32 1)
   ret <vscale x 1 x half> %res
 }
 
-declare <vscale x 2 x half> @llvm.experimental.vector.splice.nxv2f16(<vscale x 2 x half>, <vscale x 2 x half>, i32)
+declare <vscale x 2 x half> @llvm.vector.splice.nxv2f16(<vscale x 2 x half>, <vscale x 2 x half>, i32)
 
 define <vscale x 2 x half> @splice_nxv2f16_offset_zero(<vscale x 2 x half> %a, <vscale x 2 x half> %b) #0 {
 ; CHECK-LABEL: splice_nxv2f16_offset_zero:
 ; CHECK:       # %bb.0:
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 2 x half> @llvm.experimental.vector.splice.nxv2f16(<vscale x 2 x half> %a, <vscale x 2 x half> %b, i32 0)
+  %res = call <vscale x 2 x half> @llvm.vector.splice.nxv2f16(<vscale x 2 x half> %a, <vscale x 2 x half> %b, i32 0)
   ret <vscale x 2 x half> %res
 }
 
@@ -1625,7 +1625,7 @@ define <vscale x 2 x half> @splice_nxv2f16_offset_negone(<vscale x 2 x half> %a,
 ; CHECK-NEXT:    vsetvli a0, zero, e16, mf2, ta, ma
 ; CHECK-NEXT:    vslideup.vi v8, v9, 1
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 2 x half> @llvm.experimental.vector.splice.nxv2f16(<vscale x 2 x half> %a, <vscale x 2 x half> %b, i32 -1)
+  %res = call <vscale x 2 x half> @llvm.vector.splice.nxv2f16(<vscale x 2 x half> %a, <vscale x 2 x half> %b, i32 -1)
   ret <vscale x 2 x half> %res
 }
 
@@ -1640,7 +1640,7 @@ define <vscale x 2 x half> @splice_nxv2f16_offset_min(<vscale x 2 x half> %a, <v
 ; CHECK-NEXT:    vsetvli a0, zero, e16, mf2, ta, ma
 ; CHECK-NEXT:    vslideup.vi v8, v9, 4
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 2 x half> @llvm.experimental.vector.splice.nxv2f16(<vscale x 2 x half> %a, <vscale x 2 x half> %b, i32 -4)
+  %res = call <vscale x 2 x half> @llvm.vector.splice.nxv2f16(<vscale x 2 x half> %a, <vscale x 2 x half> %b, i32 -4)
   ret <vscale x 2 x half> %res
 }
 
@@ -1655,17 +1655,17 @@ define <vscale x 2 x half> @splice_nxv2f16_offset_max(<vscale x 2 x half> %a, <v
 ; CHECK-NEXT:    vsetvli a1, zero, e16, mf2, ta, ma
 ; CHECK-NEXT:    vslideup.vx v8, v9, a0
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 2 x half> @llvm.experimental.vector.splice.nxv2f16(<vscale x 2 x half> %a, <vscale x 2 x half> %b, i32 3)
+  %res = call <vscale x 2 x half> @llvm.vector.splice.nxv2f16(<vscale x 2 x half> %a, <vscale x 2 x half> %b, i32 3)
   ret <vscale x 2 x half> %res
 }
 
-declare <vscale x 4 x half> @llvm.experimental.vector.splice.nxv4f16(<vscale x 4 x half>, <vscale x 4 x half>, i32)
+declare <vscale x 4 x half> @llvm.vector.splice.nxv4f16(<vscale x 4 x half>, <vscale x 4 x half>, i32)
 
 define <vscale x 4 x half> @splice_nxv4f16_offset_zero(<vscale x 4 x half> %a, <vscale x 4 x half> %b) #0 {
 ; CHECK-LABEL: splice_nxv4f16_offset_zero:
 ; CHECK:       # %bb.0:
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 4 x half> @llvm.experimental.vector.splice.nxv4f16(<vscale x 4 x half> %a, <vscale x 4 x half> %b, i32 0)
+  %res = call <vscale x 4 x half> @llvm.vector.splice.nxv4f16(<vscale x 4 x half> %a, <vscale x 4 x half> %b, i32 0)
   ret <vscale x 4 x half> %res
 }
 
@@ -1680,7 +1680,7 @@ define <vscale x 4 x half> @splice_nxv4f16_offset_negone(<vscale x 4 x half> %a,
 ; CHECK-NEXT:    vsetvli a0, zero, e16, m1, ta, ma
 ; CHECK-NEXT:    vslideup.vi v8, v9, 1
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 4 x half> @llvm.experimental.vector.splice.nxv4f16(<vscale x 4 x half> %a, <vscale x 4 x half> %b, i32 -1)
+  %res = call <vscale x 4 x half> @llvm.vector.splice.nxv4f16(<vscale x 4 x half> %a, <vscale x 4 x half> %b, i32 -1)
   ret <vscale x 4 x half> %res
 }
 
@@ -1695,7 +1695,7 @@ define <vscale x 4 x half> @splice_nxv4f16_offset_min(<vscale x 4 x half> %a, <v
 ; CHECK-NEXT:    vsetvli a0, zero, e16, m1, ta, ma
 ; CHECK-NEXT:    vslideup.vi v8, v9, 8
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 4 x half> @llvm.experimental.vector.splice.nxv4f16(<vscale x 4 x half> %a, <vscale x 4 x half> %b, i32 -8)
+  %res = call <vscale x 4 x half> @llvm.vector.splice.nxv4f16(<vscale x 4 x half> %a, <vscale x 4 x half> %b, i32 -8)
   ret <vscale x 4 x half> %res
 }
 
@@ -1710,17 +1710,17 @@ define <vscale x 4 x half> @splice_nxv4f16_offset_max(<vscale x 4 x half> %a, <v
 ; CHECK-NEXT:    vsetvli a1, zero, e16, m1, ta, ma
 ; CHECK-NEXT:    vslideup.vx v8, v9, a0
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 4 x half> @llvm.experimental.vector.splice.nxv4f16(<vscale x 4 x half> %a, <vscale x 4 x half> %b, i32 7)
+  %res = call <vscale x 4 x half> @llvm.vector.splice.nxv4f16(<vscale x 4 x half> %a, <vscale x 4 x half> %b, i32 7)
   ret <vscale x 4 x half> %res
 }
 
-declare <vscale x 8 x half> @llvm.experimental.vector.splice.nxv8f16(<vscale x 8 x half>, <vscale x 8 x half>, i32)
+declare <vscale x 8 x half> @llvm.vector.splice.nxv8f16(<vscale x 8 x half>, <vscale x 8 x half>, i32)
 
 define <vscale x 8 x half> @splice_nxv8f16_offset_zero(<vscale x 8 x half> %a, <vscale x 8 x half> %b) #0 {
 ; CHECK-LABEL: splice_nxv8f16_offset_zero:
 ; CHECK:       # %bb.0:
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 8 x half> @llvm.experimental.vector.splice.nxv8f16(<vscale x 8 x half> %a, <vscale x 8 x half> %b, i32 0)
+  %res = call <vscale x 8 x half> @llvm.vector.splice.nxv8f16(<vscale x 8 x half> %a, <vscale x 8 x half> %b, i32 0)
   ret <vscale x 8 x half> %res
 }
 
@@ -1734,7 +1734,7 @@ define <vscale x 8 x half> @splice_nxv8f16_offset_negone(<vscale x 8 x half> %a,
 ; CHECK-NEXT:    vsetvli a0, zero, e16, m2, ta, ma
 ; CHECK-NEXT:    vslideup.vi v8, v10, 1
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 8 x half> @llvm.experimental.vector.splice.nxv8f16(<vscale x 8 x half> %a, <vscale x 8 x half> %b, i32 -1)
+  %res = call <vscale x 8 x half> @llvm.vector.splice.nxv8f16(<vscale x 8 x half> %a, <vscale x 8 x half> %b, i32 -1)
   ret <vscale x 8 x half> %res
 }
 
@@ -1748,7 +1748,7 @@ define <vscale x 8 x half> @splice_nxv8f16_offset_min(<vscale x 8 x half> %a, <v
 ; CHECK-NEXT:    vsetvli a0, zero, e16, m2, ta, ma
 ; CHECK-NEXT:    vslideup.vi v8, v10, 16
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 8 x half> @llvm.experimental.vector.splice.nxv8f16(<vscale x 8 x half> %a, <vscale x 8 x half> %b, i32 -16)
+  %res = call <vscale x 8 x half> @llvm.vector.splice.nxv8f16(<vscale x 8 x half> %a, <vscale x 8 x half> %b, i32 -16)
   ret <vscale x 8 x half> %res
 }
 
@@ -1762,17 +1762,17 @@ define <vscale x 8 x half> @splice_nxv8f16_offset_max(<vscale x 8 x half> %a, <v
 ; CHECK-NEXT:    vsetvli a1, zero, e16, m2, ta, ma
 ; CHECK-NEXT:    vslideup.vx v8, v10, a0
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 8 x half> @llvm.experimental.vector.splice.nxv8f16(<vscale x 8 x half> %a, <vscale x 8 x half> %b, i32 15)
+  %res = call <vscale x 8 x half> @llvm.vector.splice.nxv8f16(<vscale x 8 x half> %a, <vscale x 8 x half> %b, i32 15)
   ret <vscale x 8 x half> %res
 }
 
-declare <vscale x 16 x half> @llvm.experimental.vector.splice.nxv16f16(<vscale x 16 x half>, <vscale x 16 x half>, i32)
+declare <vscale x 16 x half> @llvm.vector.splice.nxv16f16(<vscale x 16 x half>, <vscale x 16 x half>, i32)
 
 define <vscale x 16 x half> @splice_nxv16f16_offset_zero(<vscale x 16 x half> %a, <vscale x 16 x half> %b) #0 {
 ; CHECK-LABEL: splice_nxv16f16_offset_zero:
 ; CHECK:       # %bb.0:
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 16 x half> @llvm.experimental.vector.splice.nxv16f16(<vscale x 16 x half> %a, <vscale x 16 x half> %b, i32 0)
+  %res = call <vscale x 16 x half> @llvm.vector.splice.nxv16f16(<vscale x 16 x half> %a, <vscale x 16 x half> %b, i32 0)
   ret <vscale x 16 x half> %res
 }
 
@@ -1787,7 +1787,7 @@ define <vscale x 16 x half> @splice_nxv16f16_offset_negone(<vscale x 16 x half>
 ; CHECK-NEXT:    vsetvli a0, zero, e16, m4, ta, ma
 ; CHECK-NEXT:    vslideup.vi v8, v12, 1
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 16 x half> @llvm.experimental.vector.splice.nxv16f16(<vscale x 16 x half> %a, <vscale x 16 x half> %b, i32 -1)
+  %res = call <vscale x 16 x half> @llvm.vector.splice.nxv16f16(<vscale x 16 x half> %a, <vscale x 16 x half> %b, i32 -1)
   ret <vscale x 16 x half> %res
 }
 
@@ -1803,7 +1803,7 @@ define <vscale x 16 x half> @splice_nxv16f16_offset_min(<vscale x 16 x half> %a,
 ; CHECK-NEXT:    vsetvli a0, zero, e16, m4, ta, ma
 ; CHECK-NEXT:    vslideup.vx v8, v12, a1
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 16 x half> @llvm.experimental.vector.splice.nxv16f16(<vscale x 16 x half> %a, <vscale x 16 x half> %b, i32 -32)
+  %res = call <vscale x 16 x half> @llvm.vector.splice.nxv16f16(<vscale x 16 x half> %a, <vscale x 16 x half> %b, i32 -32)
   ret <vscale x 16 x half> %res
 }
 
@@ -1818,17 +1818,17 @@ define <vscale x 16 x half> @splice_nxv16f16_offset_max(<vscale x 16 x half> %a,
 ; CHECK-NEXT:    vsetvli a1, zero, e16, m4, ta, ma
 ; CHECK-NEXT:    vslideup.vx v8, v12, a0
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 16 x half> @llvm.experimental.vector.splice.nxv16f16(<vscale x 16 x half> %a, <vscale x 16 x half> %b, i32 31)
+  %res = call <vscale x 16 x half> @llvm.vector.splice.nxv16f16(<vscale x 16 x half> %a, <vscale x 16 x half> %b, i32 31)
   ret <vscale x 16 x half> %res
 }
 
-declare <vscale x 32 x half> @llvm.experimental.vector.splice.nxv32f16(<vscale x 32 x half>, <vscale x 32 x half>, i32)
+declare <vscale x 32 x half> @llvm.vector.splice.nxv32f16(<vscale x 32 x half>, <vscale x 32 x half>, i32)
 
 define <vscale x 32 x half> @splice_nxv32f16_offset_zero(<vscale x 32 x half> %a, <vscale x 32 x half> %b) #0 {
 ; CHECK-LABEL: splice_nxv32f16_offset_zero:
 ; CHECK:       # %bb.0:
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 32 x half> @llvm.experimental.vector.splice.nxv32f16(<vscale x 32 x half> %a, <vscale x 32 x half> %b, i32 0)
+  %res = call <vscale x 32 x half> @llvm.vector.splice.nxv32f16(<vscale x 32 x half> %a, <vscale x 32 x half> %b, i32 0)
   ret <vscale x 32 x half> %res
 }
 
@@ -1843,7 +1843,7 @@ define <vscale x 32 x half> @splice_nxv32f16_offset_negone(<vscale x 32 x half>
 ; CHECK-NEXT:    vsetvli a0, zero, e16, m8, ta, ma
 ; CHECK-NEXT:    vslideup.vi v8, v16, 1
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 32 x half> @llvm.experimental.vector.splice.nxv32f16(<vscale x 32 x half> %a, <vscale x 32 x half> %b, i32 -1)
+  %res = call <vscale x 32 x half> @llvm.vector.splice.nxv32f16(<vscale x 32 x half> %a, <vscale x 32 x half> %b, i32 -1)
   ret <vscale x 32 x half> %res
 }
 
@@ -1859,7 +1859,7 @@ define <vscale x 32 x half> @splice_nxv32f16_offset_min(<vscale x 32 x half> %a,
 ; CHECK-NEXT:    vsetvli a0, zero, e16, m8, ta, ma
 ; CHECK-NEXT:    vslideup.vx v8, v16, a1
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 32 x half> @llvm.experimental.vector.splice.nxv32f16(<vscale x 32 x half> %a, <vscale x 32 x half> %b, i32 -64)
+  %res = call <vscale x 32 x half> @llvm.vector.splice.nxv32f16(<vscale x 32 x half> %a, <vscale x 32 x half> %b, i32 -64)
   ret <vscale x 32 x half> %res
 }
 
@@ -1875,17 +1875,17 @@ define <vscale x 32 x half> @splice_nxv32f16_offset_max(<vscale x 32 x half> %a,
 ; CHECK-NEXT:    vsetvli a1, zero, e16, m8, ta, ma
 ; CHECK-NEXT:    vslideup.vx v8, v16, a0
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 32 x half> @llvm.experimental.vector.splice.nxv32f16(<vscale x 32 x half> %a, <vscale x 32 x half> %b, i32 63)
+  %res = call <vscale x 32 x half> @llvm.vector.splice.nxv32f16(<vscale x 32 x half> %a, <vscale x 32 x half> %b, i32 63)
   ret <vscale x 32 x half> %res
 }
 
-declare <vscale x 1 x float> @llvm.experimental.vector.splice.nxv1f32(<vscale x 1 x float>, <vscale x 1 x float>, i32)
+declare <vscale x 1 x float> @llvm.vector.splice.nxv1f32(<vscale x 1 x float>, <vscale x 1 x float>, i32)
 
 define <vscale x 1 x float> @splice_nxv1f32_offset_zero(<vscale x 1 x float> %a, <vscale x 1 x float> %b) #0 {
 ; CHECK-LABEL: splice_nxv1f32_offset_zero:
 ; CHECK:       # %bb.0:
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 1 x float> @llvm.experimental.vector.splice.nxv1f32(<vscale x 1 x float> %a, <vscale x 1 x float> %b, i32 0)
+  %res = call <vscale x 1 x float> @llvm.vector.splice.nxv1f32(<vscale x 1 x float> %a, <vscale x 1 x float> %b, i32 0)
   ret <vscale x 1 x float> %res
 }
 
@@ -1900,7 +1900,7 @@ define <vscale x 1 x float> @splice_nxv1f32_offset_negone(<vscale x 1 x float> %
 ; CHECK-NEXT:    vsetvli a0, zero, e32, mf2, ta, ma
 ; CHECK-NEXT:    vslideup.vi v8, v9, 1
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 1 x float> @llvm.experimental.vector.splice.nxv1f32(<vscale x 1 x float> %a, <vscale x 1 x float> %b, i32 -1)
+  %res = call <vscale x 1 x float> @llvm.vector.splice.nxv1f32(<vscale x 1 x float> %a, <vscale x 1 x float> %b, i32 -1)
   ret <vscale x 1 x float> %res
 }
 
@@ -1915,7 +1915,7 @@ define <vscale x 1 x float> @splice_nxv1f32_offset_min(<vscale x 1 x float> %a,
 ; CHECK-NEXT:    vsetvli a0, zero, e32, mf2, ta, ma
 ; CHECK-NEXT:    vslideup.vi v8, v9, 2
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 1 x float> @llvm.experimental.vector.splice.nxv1f32(<vscale x 1 x float> %a, <vscale x 1 x float> %b, i32 -2)
+  %res = call <vscale x 1 x float> @llvm.vector.splice.nxv1f32(<vscale x 1 x float> %a, <vscale x 1 x float> %b, i32 -2)
   ret <vscale x 1 x float> %res
 }
 
@@ -1930,17 +1930,17 @@ define <vscale x 1 x float> @splice_nxv1f32_offset_max(<vscale x 1 x float> %a,
 ; CHECK-NEXT:    vsetvli a1, zero, e32, mf2, ta, ma
 ; CHECK-NEXT:    vslideup.vx v8, v9, a0
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 1 x float> @llvm.experimental.vector.splice.nxv1f32(<vscale x 1 x float> %a, <vscale x 1 x float> %b, i32 1)
+  %res = call <vscale x 1 x float> @llvm.vector.splice.nxv1f32(<vscale x 1 x float> %a, <vscale x 1 x float> %b, i32 1)
   ret <vscale x 1 x float> %res
 }
 
-declare <vscale x 2 x float> @llvm.experimental.vector.splice.nxv2f32(<vscale x 2 x float>, <vscale x 2 x float>, i32)
+declare <vscale x 2 x float> @llvm.vector.splice.nxv2f32(<vscale x 2 x float>, <vscale x 2 x float>, i32)
 
 define <vscale x 2 x float> @splice_nxv2f32_offset_zero(<vscale x 2 x float> %a, <vscale x 2 x float> %b) #0 {
 ; CHECK-LABEL: splice_nxv2f32_offset_zero:
 ; CHECK:       # %bb.0:
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 2 x float> @llvm.experimental.vector.splice.nxv2f32(<vscale x 2 x float> %a, <vscale x 2 x float> %b, i32 0)
+  %res = call <vscale x 2 x float> @llvm.vector.splice.nxv2f32(<vscale x 2 x float> %a, <vscale x 2 x float> %b, i32 0)
   ret <vscale x 2 x float> %res
 }
 
@@ -1955,7 +1955,7 @@ define <vscale x 2 x float> @splice_nxv2f32_offset_negone(<vscale x 2 x float> %
 ; CHECK-NEXT:    vsetvli a0, zero, e32, m1, ta, ma
 ; CHECK-NEXT:    vslideup.vi v8, v9, 1
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 2 x float> @llvm.experimental.vector.splice.nxv2f32(<vscale x 2 x float> %a, <vscale x 2 x float> %b, i32 -1)
+  %res = call <vscale x 2 x float> @llvm.vector.splice.nxv2f32(<vscale x 2 x float> %a, <vscale x 2 x float> %b, i32 -1)
   ret <vscale x 2 x float> %res
 }
 
@@ -1970,7 +1970,7 @@ define <vscale x 2 x float> @splice_nxv2f32_offset_min(<vscale x 2 x float> %a,
 ; CHECK-NEXT:    vsetvli a0, zero, e32, m1, ta, ma
 ; CHECK-NEXT:    vslideup.vi v8, v9, 4
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 2 x float> @llvm.experimental.vector.splice.nxv2f32(<vscale x 2 x float> %a, <vscale x 2 x float> %b, i32 -4)
+  %res = call <vscale x 2 x float> @llvm.vector.splice.nxv2f32(<vscale x 2 x float> %a, <vscale x 2 x float> %b, i32 -4)
   ret <vscale x 2 x float> %res
 }
 
@@ -1985,17 +1985,17 @@ define <vscale x 2 x float> @splice_nxv2f32_offset_max(<vscale x 2 x float> %a,
 ; CHECK-NEXT:    vsetvli a1, zero, e32, m1, ta, ma
 ; CHECK-NEXT:    vslideup.vx v8, v9, a0
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 2 x float> @llvm.experimental.vector.splice.nxv2f32(<vscale x 2 x float> %a, <vscale x 2 x float> %b, i32 3)
+  %res = call <vscale x 2 x float> @llvm.vector.splice.nxv2f32(<vscale x 2 x float> %a, <vscale x 2 x float> %b, i32 3)
   ret <vscale x 2 x float> %res
 }
 
-declare <vscale x 4 x float> @llvm.experimental.vector.splice.nxv4f32(<vscale x 4 x float>, <vscale x 4 x float>, i32)
+declare <vscale x 4 x float> @llvm.vector.splice.nxv4f32(<vscale x 4 x float>, <vscale x 4 x float>, i32)
 
 define <vscale x 4 x float> @splice_nxv4f32_offset_zero(<vscale x 4 x float> %a, <vscale x 4 x float> %b) #0 {
 ; CHECK-LABEL: splice_nxv4f32_offset_zero:
 ; CHECK:       # %bb.0:
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 4 x float> @llvm.experimental.vector.splice.nxv4f32(<vscale x 4 x float> %a, <vscale x 4 x float> %b, i32 0)
+  %res = call <vscale x 4 x float> @llvm.vector.splice.nxv4f32(<vscale x 4 x float> %a, <vscale x 4 x float> %b, i32 0)
   ret <vscale x 4 x float> %res
 }
 
@@ -2010,7 +2010,7 @@ define <vscale x 4 x float> @splice_nxv4f32_offset_negone(<vscale x 4 x float> %
 ; CHECK-NEXT:    vsetvli a0, zero, e32, m2, ta, ma
 ; CHECK-NEXT:    vslideup.vi v8, v10, 1
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 4 x float> @llvm.experimental.vector.splice.nxv4f32(<vscale x 4 x float> %a, <vscale x 4 x float> %b, i32 -1)
+  %res = call <vscale x 4 x float> @llvm.vector.splice.nxv4f32(<vscale x 4 x float> %a, <vscale x 4 x float> %b, i32 -1)
   ret <vscale x 4 x float> %res
 }
 
@@ -2025,7 +2025,7 @@ define <vscale x 4 x float> @splice_nxv4f32_offset_min(<vscale x 4 x float> %a,
 ; CHECK-NEXT:    vsetvli a0, zero, e32, m2, ta, ma
 ; CHECK-NEXT:    vslideup.vi v8, v10, 8
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 4 x float> @llvm.experimental.vector.splice.nxv4f32(<vscale x 4 x float> %a, <vscale x 4 x float> %b, i32 -8)
+  %res = call <vscale x 4 x float> @llvm.vector.splice.nxv4f32(<vscale x 4 x float> %a, <vscale x 4 x float> %b, i32 -8)
   ret <vscale x 4 x float> %res
 }
 
@@ -2040,17 +2040,17 @@ define <vscale x 4 x float> @splice_nxv4f32_offset_max(<vscale x 4 x float> %a,
 ; CHECK-NEXT:    vsetvli a1, zero, e32, m2, ta, ma
 ; CHECK-NEXT:    vslideup.vx v8, v10, a0
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 4 x float> @llvm.experimental.vector.splice.nxv4f32(<vscale x 4 x float> %a, <vscale x 4 x float> %b, i32 7)
+  %res = call <vscale x 4 x float> @llvm.vector.splice.nxv4f32(<vscale x 4 x float> %a, <vscale x 4 x float> %b, i32 7)
   ret <vscale x 4 x float> %res
 }
 
-declare <vscale x 8 x float> @llvm.experimental.vector.splice.nxv8f32(<vscale x 8 x float>, <vscale x 8 x float>, i32)
+declare <vscale x 8 x float> @llvm.vector.splice.nxv8f32(<vscale x 8 x float>, <vscale x 8 x float>, i32)
 
 define <vscale x 8 x float> @splice_nxv8f32_offset_zero(<vscale x 8 x float> %a, <vscale x 8 x float> %b) #0 {
 ; CHECK-LABEL: splice_nxv8f32_offset_zero:
 ; CHECK:       # %bb.0:
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 8 x float> @llvm.experimental.vector.splice.nxv8f32(<vscale x 8 x float> %a, <vscale x 8 x float> %b, i32 0)
+  %res = call <vscale x 8 x float> @llvm.vector.splice.nxv8f32(<vscale x 8 x float> %a, <vscale x 8 x float> %b, i32 0)
   ret <vscale x 8 x float> %res
 }
 
@@ -2064,7 +2064,7 @@ define <vscale x 8 x float> @splice_nxv8f32_offset_negone(<vscale x 8 x float> %
 ; CHECK-NEXT:    vsetvli a0, zero, e32, m4, ta, ma
 ; CHECK-NEXT:    vslideup.vi v8, v12, 1
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 8 x float> @llvm.experimental.vector.splice.nxv8f32(<vscale x 8 x float> %a, <vscale x 8 x float> %b, i32 -1)
+  %res = call <vscale x 8 x float> @llvm.vector.splice.nxv8f32(<vscale x 8 x float> %a, <vscale x 8 x float> %b, i32 -1)
   ret <vscale x 8 x float> %res
 }
 
@@ -2078,7 +2078,7 @@ define <vscale x 8 x float> @splice_nxv8f32_offset_min(<vscale x 8 x float> %a,
 ; CHECK-NEXT:    vsetvli a0, zero, e32, m4, ta, ma
 ; CHECK-NEXT:    vslideup.vi v8, v12, 16
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 8 x float> @llvm.experimental.vector.splice.nxv8f32(<vscale x 8 x float> %a, <vscale x 8 x float> %b, i32 -16)
+  %res = call <vscale x 8 x float> @llvm.vector.splice.nxv8f32(<vscale x 8 x float> %a, <vscale x 8 x float> %b, i32 -16)
   ret <vscale x 8 x float> %res
 }
 
@@ -2092,17 +2092,17 @@ define <vscale x 8 x float> @splice_nxv8f32_offset_max(<vscale x 8 x float> %a,
 ; CHECK-NEXT:    vsetvli a1, zero, e32, m4, ta, ma
 ; CHECK-NEXT:    vslideup.vx v8, v12, a0
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 8 x float> @llvm.experimental.vector.splice.nxv8f32(<vscale x 8 x float> %a, <vscale x 8 x float> %b, i32 15)
+  %res = call <vscale x 8 x float> @llvm.vector.splice.nxv8f32(<vscale x 8 x float> %a, <vscale x 8 x float> %b, i32 15)
   ret <vscale x 8 x float> %res
 }
 
-declare <vscale x 16 x float> @llvm.experimental.vector.splice.nxv16f32(<vscale x 16 x float>, <vscale x 16 x float>, i32)
+declare <vscale x 16 x float> @llvm.vector.splice.nxv16f32(<vscale x 16 x float>, <vscale x 16 x float>, i32)
 
 define <vscale x 16 x float> @splice_nxv16f32_offset_zero(<vscale x 16 x float> %a, <vscale x 16 x float> %b) #0 {
 ; CHECK-LABEL: splice_nxv16f32_offset_zero:
 ; CHECK:       # %bb.0:
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 16 x float> @llvm.experimental.vector.splice.nxv16f32(<vscale x 16 x float> %a, <vscale x 16 x float> %b, i32 0)
+  %res = call <vscale x 16 x float> @llvm.vector.splice.nxv16f32(<vscale x 16 x float> %a, <vscale x 16 x float> %b, i32 0)
   ret <vscale x 16 x float> %res
 }
 
@@ -2117,7 +2117,7 @@ define <vscale x 16 x float> @splice_nxv16f32_offset_negone(<vscale x 16 x float
 ; CHECK-NEXT:    vsetvli a0, zero, e32, m8, ta, ma
 ; CHECK-NEXT:    vslideup.vi v8, v16, 1
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 16 x float> @llvm.experimental.vector.splice.nxv16f32(<vscale x 16 x float> %a, <vscale x 16 x float> %b, i32 -1)
+  %res = call <vscale x 16 x float> @llvm.vector.splice.nxv16f32(<vscale x 16 x float> %a, <vscale x 16 x float> %b, i32 -1)
   ret <vscale x 16 x float> %res
 }
 
@@ -2133,7 +2133,7 @@ define <vscale x 16 x float> @splice_nxv16f32_offset_min(<vscale x 16 x float> %
 ; CHECK-NEXT:    vsetvli a0, zero, e32, m8, ta, ma
 ; CHECK-NEXT:    vslideup.vx v8, v16, a1
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 16 x float> @llvm.experimental.vector.splice.nxv16f32(<vscale x 16 x float> %a, <vscale x 16 x float> %b, i32 -32)
+  %res = call <vscale x 16 x float> @llvm.vector.splice.nxv16f32(<vscale x 16 x float> %a, <vscale x 16 x float> %b, i32 -32)
   ret <vscale x 16 x float> %res
 }
 
@@ -2148,17 +2148,17 @@ define <vscale x 16 x float> @splice_nxv16f32_offset_max(<vscale x 16 x float> %
 ; CHECK-NEXT:    vsetvli a1, zero, e32, m8, ta, ma
 ; CHECK-NEXT:    vslideup.vx v8, v16, a0
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 16 x float> @llvm.experimental.vector.splice.nxv16f32(<vscale x 16 x float> %a, <vscale x 16 x float> %b, i32 31)
+  %res = call <vscale x 16 x float> @llvm.vector.splice.nxv16f32(<vscale x 16 x float> %a, <vscale x 16 x float> %b, i32 31)
   ret <vscale x 16 x float> %res
 }
 
-declare <vscale x 1 x double> @llvm.experimental.vector.splice.nxv1f64(<vscale x 1 x double>, <vscale x 1 x double>, i32)
+declare <vscale x 1 x double> @llvm.vector.splice.nxv1f64(<vscale x 1 x double>, <vscale x 1 x double>, i32)
 
 define <vscale x 1 x double> @splice_nxv1f64_offset_zero(<vscale x 1 x double> %a, <vscale x 1 x double> %b) #0 {
 ; CHECK-LABEL: splice_nxv1f64_offset_zero:
 ; CHECK:       # %bb.0:
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 1 x double> @llvm.experimental.vector.splice.nxv1f64(<vscale x 1 x double> %a, <vscale x 1 x double> %b, i32 0)
+  %res = call <vscale x 1 x double> @llvm.vector.splice.nxv1f64(<vscale x 1 x double> %a, <vscale x 1 x double> %b, i32 0)
   ret <vscale x 1 x double> %res
 }
 
@@ -2173,7 +2173,7 @@ define <vscale x 1 x double> @splice_nxv1f64_offset_negone(<vscale x 1 x double>
 ; CHECK-NEXT:    vsetvli a0, zero, e64, m1, ta, ma
 ; CHECK-NEXT:    vslideup.vi v8, v9, 1
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 1 x double> @llvm.experimental.vector.splice.nxv1f64(<vscale x 1 x double> %a, <vscale x 1 x double> %b, i32 -1)
+  %res = call <vscale x 1 x double> @llvm.vector.splice.nxv1f64(<vscale x 1 x double> %a, <vscale x 1 x double> %b, i32 -1)
   ret <vscale x 1 x double> %res
 }
 
@@ -2188,7 +2188,7 @@ define <vscale x 1 x double> @splice_nxv1f64_offset_min(<vscale x 1 x double> %a
 ; CHECK-NEXT:    vsetvli a0, zero, e64, m1, ta, ma
 ; CHECK-NEXT:    vslideup.vi v8, v9, 2
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 1 x double> @llvm.experimental.vector.splice.nxv1f64(<vscale x 1 x double> %a, <vscale x 1 x double> %b, i32 -2)
+  %res = call <vscale x 1 x double> @llvm.vector.splice.nxv1f64(<vscale x 1 x double> %a, <vscale x 1 x double> %b, i32 -2)
   ret <vscale x 1 x double> %res
 }
 
@@ -2203,17 +2203,17 @@ define <vscale x 1 x double> @splice_nxv1f64_offset_max(<vscale x 1 x double> %a
 ; CHECK-NEXT:    vsetvli a1, zero, e64, m1, ta, ma
 ; CHECK-NEXT:    vslideup.vx v8, v9, a0
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 1 x double> @llvm.experimental.vector.splice.nxv1f64(<vscale x 1 x double> %a, <vscale x 1 x double> %b, i32 1)
+  %res = call <vscale x 1 x double> @llvm.vector.splice.nxv1f64(<vscale x 1 x double> %a, <vscale x 1 x double> %b, i32 1)
   ret <vscale x 1 x double> %res
 }
 
-declare <vscale x 2 x double> @llvm.experimental.vector.splice.nxv2f64(<vscale x 2 x double>, <vscale x 2 x double>, i32)
+declare <vscale x 2 x double> @llvm.vector.splice.nxv2f64(<vscale x 2 x double>, <vscale x 2 x double>, i32)
 
 define <vscale x 2 x double> @splice_nxv2f64_offset_zero(<vscale x 2 x double> %a, <vscale x 2 x double> %b) #0 {
 ; CHECK-LABEL: splice_nxv2f64_offset_zero:
 ; CHECK:       # %bb.0:
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 2 x double> @llvm.experimental.vector.splice.nxv2f64(<vscale x 2 x double> %a, <vscale x 2 x double> %b, i32 0)
+  %res = call <vscale x 2 x double> @llvm.vector.splice.nxv2f64(<vscale x 2 x double> %a, <vscale x 2 x double> %b, i32 0)
   ret <vscale x 2 x double> %res
 }
 
@@ -2228,7 +2228,7 @@ define <vscale x 2 x double> @splice_nxv2f64_offset_negone(<vscale x 2 x double>
 ; CHECK-NEXT:    vsetvli a0, zero, e64, m2, ta, ma
 ; CHECK-NEXT:    vslideup.vi v8, v10, 1
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 2 x double> @llvm.experimental.vector.splice.nxv2f64(<vscale x 2 x double> %a, <vscale x 2 x double> %b, i32 -1)
+  %res = call <vscale x 2 x double> @llvm.vector.splice.nxv2f64(<vscale x 2 x double> %a, <vscale x 2 x double> %b, i32 -1)
   ret <vscale x 2 x double> %res
 }
 
@@ -2243,7 +2243,7 @@ define <vscale x 2 x double> @splice_nxv2f64_offset_min(<vscale x 2 x double> %a
 ; CHECK-NEXT:    vsetvli a0, zero, e64, m2, ta, ma
 ; CHECK-NEXT:    vslideup.vi v8, v10, 4
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 2 x double> @llvm.experimental.vector.splice.nxv2f64(<vscale x 2 x double> %a, <vscale x 2 x double> %b, i32 -4)
+  %res = call <vscale x 2 x double> @llvm.vector.splice.nxv2f64(<vscale x 2 x double> %a, <vscale x 2 x double> %b, i32 -4)
   ret <vscale x 2 x double> %res
 }
 
@@ -2258,17 +2258,17 @@ define <vscale x 2 x double> @splice_nxv2f64_offset_max(<vscale x 2 x double> %a
 ; CHECK-NEXT:    vsetvli a1, zero, e64, m2, ta, ma
 ; CHECK-NEXT:    vslideup.vx v8, v10, a0
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 2 x double> @llvm.experimental.vector.splice.nxv2f64(<vscale x 2 x double> %a, <vscale x 2 x double> %b, i32 3)
+  %res = call <vscale x 2 x double> @llvm.vector.splice.nxv2f64(<vscale x 2 x double> %a, <vscale x 2 x double> %b, i32 3)
   ret <vscale x 2 x double> %res
 }
 
-declare <vscale x 4 x double> @llvm.experimental.vector.splice.nxv4f64(<vscale x 4 x double>, <vscale x 4 x double>, i32)
+declare <vscale x 4 x double> @llvm.vector.splice.nxv4f64(<vscale x 4 x double>, <vscale x 4 x double>, i32)
 
 define <vscale x 4 x double> @splice_nxv4f64_offset_zero(<vscale x 4 x double> %a, <vscale x 4 x double> %b) #0 {
 ; CHECK-LABEL: splice_nxv4f64_offset_zero:
 ; CHECK:       # %bb.0:
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 4 x double> @llvm.experimental.vector.splice.nxv4f64(<vscale x 4 x double> %a, <vscale x 4 x double> %b, i32 0)
+  %res = call <vscale x 4 x double> @llvm.vector.splice.nxv4f64(<vscale x 4 x double> %a, <vscale x 4 x double> %b, i32 0)
   ret <vscale x 4 x double> %res
 }
 
@@ -2283,7 +2283,7 @@ define <vscale x 4 x double> @splice_nxv4f64_offset_negone(<vscale x 4 x double>
 ; CHECK-NEXT:    vsetvli a0, zero, e64, m4, ta, ma
 ; CHECK-NEXT:    vslideup.vi v8, v12, 1
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 4 x double> @llvm.experimental.vector.splice.nxv4f64(<vscale x 4 x double> %a, <vscale x 4 x double> %b, i32 -1)
+  %res = call <vscale x 4 x double> @llvm.vector.splice.nxv4f64(<vscale x 4 x double> %a, <vscale x 4 x double> %b, i32 -1)
   ret <vscale x 4 x double> %res
 }
 
@@ -2298,7 +2298,7 @@ define <vscale x 4 x double> @splice_nxv4f64_offset_min(<vscale x 4 x double> %a
 ; CHECK-NEXT:    vsetvli a0, zero, e64, m4, ta, ma
 ; CHECK-NEXT:    vslideup.vi v8, v12, 8
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 4 x double> @llvm.experimental.vector.splice.nxv4f64(<vscale x 4 x double> %a, <vscale x 4 x double> %b, i32 -8)
+  %res = call <vscale x 4 x double> @llvm.vector.splice.nxv4f64(<vscale x 4 x double> %a, <vscale x 4 x double> %b, i32 -8)
   ret <vscale x 4 x double> %res
 }
 
@@ -2313,17 +2313,17 @@ define <vscale x 4 x double> @splice_nxv4f64_offset_max(<vscale x 4 x double> %a
 ; CHECK-NEXT:    vsetvli a1, zero, e64, m4, ta, ma
 ; CHECK-NEXT:    vslideup.vx v8, v12, a0
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 4 x double> @llvm.experimental.vector.splice.nxv4f64(<vscale x 4 x double> %a, <vscale x 4 x double> %b, i32 7)
+  %res = call <vscale x 4 x double> @llvm.vector.splice.nxv4f64(<vscale x 4 x double> %a, <vscale x 4 x double> %b, i32 7)
   ret <vscale x 4 x double> %res
 }
 
-declare <vscale x 8 x double> @llvm.experimental.vector.splice.nxv8f64(<vscale x 8 x double>, <vscale x 8 x double>, i32)
+declare <vscale x 8 x double> @llvm.vector.splice.nxv8f64(<vscale x 8 x double>, <vscale x 8 x double>, i32)
 
 define <vscale x 8 x double> @splice_nxv8f64_offset_zero(<vscale x 8 x double> %a, <vscale x 8 x double> %b) #0 {
 ; CHECK-LABEL: splice_nxv8f64_offset_zero:
 ; CHECK:       # %bb.0:
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 8 x double> @llvm.experimental.vector.splice.nxv8f64(<vscale x 8 x double> %a, <vscale x 8 x double> %b, i32 0)
+  %res = call <vscale x 8 x double> @llvm.vector.splice.nxv8f64(<vscale x 8 x double> %a, <vscale x 8 x double> %b, i32 0)
   ret <vscale x 8 x double> %res
 }
 
@@ -2337,7 +2337,7 @@ define <vscale x 8 x double> @splice_nxv8f64_offset_negone(<vscale x 8 x double>
 ; CHECK-NEXT:    vsetvli a0, zero, e64, m8, ta, ma
 ; CHECK-NEXT:    vslideup.vi v8, v16, 1
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 8 x double> @llvm.experimental.vector.splice.nxv8f64(<vscale x 8 x double> %a, <vscale x 8 x double> %b, i32 -1)
+  %res = call <vscale x 8 x double> @llvm.vector.splice.nxv8f64(<vscale x 8 x double> %a, <vscale x 8 x double> %b, i32 -1)
   ret <vscale x 8 x double> %res
 }
 
@@ -2351,7 +2351,7 @@ define <vscale x 8 x double> @splice_nxv8f64_offset_min(<vscale x 8 x double> %a
 ; CHECK-NEXT:    vsetvli a0, zero, e64, m8, ta, ma
 ; CHECK-NEXT:    vslideup.vi v8, v16, 16
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 8 x double> @llvm.experimental.vector.splice.nxv8f64(<vscale x 8 x double> %a, <vscale x 8 x double> %b, i32 -16)
+  %res = call <vscale x 8 x double> @llvm.vector.splice.nxv8f64(<vscale x 8 x double> %a, <vscale x 8 x double> %b, i32 -16)
   ret <vscale x 8 x double> %res
 }
 
@@ -2365,7 +2365,7 @@ define <vscale x 8 x double> @splice_nxv8f64_offset_max(<vscale x 8 x double> %a
 ; CHECK-NEXT:    vsetvli a1, zero, e64, m8, ta, ma
 ; CHECK-NEXT:    vslideup.vx v8, v16, a0
 ; CHECK-NEXT:    ret
-  %res = call <vscale x 8 x double> @llvm.experimental.vector.splice.nxv8f64(<vscale x 8 x double> %a, <vscale x 8 x double> %b, i32 15)
+  %res = call <vscale x 8 x double> @llvm.vector.splice.nxv8f64(<vscale x 8 x double> %a, <vscale x 8 x double> %b, i32 15)
   ret <vscale x 8 x double> %res
 }
 
diff --git a/llvm/test/CodeGen/RISCV/rvv/vmfeq.ll b/llvm/test/CodeGen/RISCV/rvv/vmfeq.ll
index ffeb399291e1..e7184921d87a 100644
--- a/llvm/test/CodeGen/RISCV/rvv/vmfeq.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/vmfeq.ll
@@ -34,10 +34,9 @@ declare <vscale x 1 x i1> @llvm.riscv.vmfeq.mask.nxv1f16(
 define <vscale x 1 x i1> @intrinsic_vmfeq_mask_vv_nxv1f16_nxv1f16(<vscale x 1 x i1> %0, <vscale x 1 x half> %1, <vscale x 1 x half> %2, <vscale x 1 x half> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmfeq_mask_vv_nxv1f16_nxv1f16:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
-; CHECK-NEXT:    vmfeq.vv v8, v8, v9
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv1r.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
+; CHECK-NEXT:    vmfeq.vv v0, v8, v9
 ; CHECK-NEXT:    vmfeq.vv v11, v9, v10, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v11
 ; CHECK-NEXT:    ret
@@ -86,10 +85,9 @@ declare <vscale x 2 x i1> @llvm.riscv.vmfeq.mask.nxv2f16(
 define <vscale x 2 x i1> @intrinsic_vmfeq_mask_vv_nxv2f16_nxv2f16(<vscale x 2 x i1> %0, <vscale x 2 x half> %1, <vscale x 2 x half> %2, <vscale x 2 x half> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmfeq_mask_vv_nxv2f16_nxv2f16:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e16, mf2, ta, mu
-; CHECK-NEXT:    vmfeq.vv v8, v8, v9
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv1r.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e16, mf2, ta, mu
+; CHECK-NEXT:    vmfeq.vv v0, v8, v9
 ; CHECK-NEXT:    vmfeq.vv v11, v9, v10, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v11
 ; CHECK-NEXT:    ret
@@ -138,10 +136,9 @@ declare <vscale x 4 x i1> @llvm.riscv.vmfeq.mask.nxv4f16(
 define <vscale x 4 x i1> @intrinsic_vmfeq_mask_vv_nxv4f16_nxv4f16(<vscale x 4 x i1> %0, <vscale x 4 x half> %1, <vscale x 4 x half> %2, <vscale x 4 x half> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmfeq_mask_vv_nxv4f16_nxv4f16:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e16, m1, ta, mu
-; CHECK-NEXT:    vmfeq.vv v8, v8, v9
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv.v.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e16, m1, ta, mu
+; CHECK-NEXT:    vmfeq.vv v0, v8, v9
 ; CHECK-NEXT:    vmfeq.vv v11, v9, v10, v0.t
 ; CHECK-NEXT:    vmv.v.v v0, v11
 ; CHECK-NEXT:    ret
@@ -190,12 +187,11 @@ declare <vscale x 8 x i1> @llvm.riscv.vmfeq.mask.nxv8f16(
 define <vscale x 8 x i1> @intrinsic_vmfeq_mask_vv_nxv8f16_nxv8f16(<vscale x 8 x i1> %0, <vscale x 8 x half> %1, <vscale x 8 x half> %2, <vscale x 8 x half> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmfeq_mask_vv_nxv8f16_nxv8f16:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v14, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e16, m2, ta, mu
-; CHECK-NEXT:    vmfeq.vv v14, v8, v10
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmfeq.vv v0, v8, v10
+; CHECK-NEXT:    vmfeq.vv v14, v10, v12, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v14
-; CHECK-NEXT:    vmfeq.vv v8, v10, v12, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 8 x i1> @llvm.riscv.vmfeq.nxv8f16(
@@ -242,12 +238,11 @@ declare <vscale x 16 x i1> @llvm.riscv.vmfeq.mask.nxv16f16(
 define <vscale x 16 x i1> @intrinsic_vmfeq_mask_vv_nxv16f16_nxv16f16(<vscale x 16 x i1> %0, <vscale x 16 x half> %1, <vscale x 16 x half> %2, <vscale x 16 x half> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmfeq_mask_vv_nxv16f16_nxv16f16:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v20, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e16, m4, ta, mu
-; CHECK-NEXT:    vmfeq.vv v20, v8, v12
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmfeq.vv v0, v8, v12
+; CHECK-NEXT:    vmfeq.vv v20, v12, v16, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v20
-; CHECK-NEXT:    vmfeq.vv v8, v12, v16, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 16 x i1> @llvm.riscv.vmfeq.nxv16f16(
@@ -294,10 +289,9 @@ declare <vscale x 1 x i1> @llvm.riscv.vmfeq.mask.nxv1f32(
 define <vscale x 1 x i1> @intrinsic_vmfeq_mask_vv_nxv1f32_nxv1f32(<vscale x 1 x i1> %0, <vscale x 1 x float> %1, <vscale x 1 x float> %2, <vscale x 1 x float> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmfeq_mask_vv_nxv1f32_nxv1f32:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e32, mf2, ta, mu
-; CHECK-NEXT:    vmfeq.vv v8, v8, v9
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv1r.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e32, mf2, ta, mu
+; CHECK-NEXT:    vmfeq.vv v0, v8, v9
 ; CHECK-NEXT:    vmfeq.vv v11, v9, v10, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v11
 ; CHECK-NEXT:    ret
@@ -346,10 +340,9 @@ declare <vscale x 2 x i1> @llvm.riscv.vmfeq.mask.nxv2f32(
 define <vscale x 2 x i1> @intrinsic_vmfeq_mask_vv_nxv2f32_nxv2f32(<vscale x 2 x i1> %0, <vscale x 2 x float> %1, <vscale x 2 x float> %2, <vscale x 2 x float> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmfeq_mask_vv_nxv2f32_nxv2f32:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e32, m1, ta, mu
-; CHECK-NEXT:    vmfeq.vv v8, v8, v9
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv.v.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e32, m1, ta, mu
+; CHECK-NEXT:    vmfeq.vv v0, v8, v9
 ; CHECK-NEXT:    vmfeq.vv v11, v9, v10, v0.t
 ; CHECK-NEXT:    vmv.v.v v0, v11
 ; CHECK-NEXT:    ret
@@ -398,12 +391,11 @@ declare <vscale x 4 x i1> @llvm.riscv.vmfeq.mask.nxv4f32(
 define <vscale x 4 x i1> @intrinsic_vmfeq_mask_vv_nxv4f32_nxv4f32(<vscale x 4 x i1> %0, <vscale x 4 x float> %1, <vscale x 4 x float> %2, <vscale x 4 x float> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmfeq_mask_vv_nxv4f32_nxv4f32:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v14, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e32, m2, ta, mu
-; CHECK-NEXT:    vmfeq.vv v14, v8, v10
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmfeq.vv v0, v8, v10
+; CHECK-NEXT:    vmfeq.vv v14, v10, v12, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v14
-; CHECK-NEXT:    vmfeq.vv v8, v10, v12, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 4 x i1> @llvm.riscv.vmfeq.nxv4f32(
@@ -450,12 +442,11 @@ declare <vscale x 8 x i1> @llvm.riscv.vmfeq.mask.nxv8f32(
 define <vscale x 8 x i1> @intrinsic_vmfeq_mask_vv_nxv8f32_nxv8f32(<vscale x 8 x i1> %0, <vscale x 8 x float> %1, <vscale x 8 x float> %2, <vscale x 8 x float> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmfeq_mask_vv_nxv8f32_nxv8f32:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v20, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e32, m4, ta, mu
-; CHECK-NEXT:    vmfeq.vv v20, v8, v12
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmfeq.vv v0, v8, v12
+; CHECK-NEXT:    vmfeq.vv v20, v12, v16, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v20
-; CHECK-NEXT:    vmfeq.vv v8, v12, v16, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 8 x i1> @llvm.riscv.vmfeq.nxv8f32(
@@ -502,10 +493,9 @@ declare <vscale x 1 x i1> @llvm.riscv.vmfeq.mask.nxv1f64(
 define <vscale x 1 x i1> @intrinsic_vmfeq_mask_vv_nxv1f64_nxv1f64(<vscale x 1 x i1> %0, <vscale x 1 x double> %1, <vscale x 1 x double> %2, <vscale x 1 x double> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmfeq_mask_vv_nxv1f64_nxv1f64:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e64, m1, ta, mu
-; CHECK-NEXT:    vmfeq.vv v8, v8, v9
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv.v.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e64, m1, ta, mu
+; CHECK-NEXT:    vmfeq.vv v0, v8, v9
 ; CHECK-NEXT:    vmfeq.vv v11, v9, v10, v0.t
 ; CHECK-NEXT:    vmv.v.v v0, v11
 ; CHECK-NEXT:    ret
@@ -554,12 +544,11 @@ declare <vscale x 2 x i1> @llvm.riscv.vmfeq.mask.nxv2f64(
 define <vscale x 2 x i1> @intrinsic_vmfeq_mask_vv_nxv2f64_nxv2f64(<vscale x 2 x i1> %0, <vscale x 2 x double> %1, <vscale x 2 x double> %2, <vscale x 2 x double> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmfeq_mask_vv_nxv2f64_nxv2f64:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v14, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e64, m2, ta, mu
-; CHECK-NEXT:    vmfeq.vv v14, v8, v10
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmfeq.vv v0, v8, v10
+; CHECK-NEXT:    vmfeq.vv v14, v10, v12, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v14
-; CHECK-NEXT:    vmfeq.vv v8, v10, v12, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 2 x i1> @llvm.riscv.vmfeq.nxv2f64(
@@ -606,12 +595,11 @@ declare <vscale x 4 x i1> @llvm.riscv.vmfeq.mask.nxv4f64(
 define <vscale x 4 x i1> @intrinsic_vmfeq_mask_vv_nxv4f64_nxv4f64(<vscale x 4 x i1> %0, <vscale x 4 x double> %1, <vscale x 4 x double> %2, <vscale x 4 x double> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmfeq_mask_vv_nxv4f64_nxv4f64:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v20, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e64, m4, ta, mu
-; CHECK-NEXT:    vmfeq.vv v20, v8, v12
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmfeq.vv v0, v8, v12
+; CHECK-NEXT:    vmfeq.vv v20, v12, v16, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v20
-; CHECK-NEXT:    vmfeq.vv v8, v12, v16, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 4 x i1> @llvm.riscv.vmfeq.nxv4f64(
diff --git a/llvm/test/CodeGen/RISCV/rvv/vmfge.ll b/llvm/test/CodeGen/RISCV/rvv/vmfge.ll
index 993b50a1c81c..a6dad9eaa4f3 100644
--- a/llvm/test/CodeGen/RISCV/rvv/vmfge.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/vmfge.ll
@@ -34,10 +34,9 @@ declare <vscale x 1 x i1> @llvm.riscv.vmfge.mask.nxv1f16(
 define <vscale x 1 x i1> @intrinsic_vmfge_mask_vv_nxv1f16_nxv1f16(<vscale x 1 x i1> %0, <vscale x 1 x half> %1, <vscale x 1 x half> %2, <vscale x 1 x half> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmfge_mask_vv_nxv1f16_nxv1f16:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
-; CHECK-NEXT:    vmfle.vv v8, v9, v8
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv1r.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
+; CHECK-NEXT:    vmfle.vv v0, v9, v8
 ; CHECK-NEXT:    vmfle.vv v11, v10, v9, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v11
 ; CHECK-NEXT:    ret
@@ -86,10 +85,9 @@ declare <vscale x 2 x i1> @llvm.riscv.vmfge.mask.nxv2f16(
 define <vscale x 2 x i1> @intrinsic_vmfge_mask_vv_nxv2f16_nxv2f16(<vscale x 2 x i1> %0, <vscale x 2 x half> %1, <vscale x 2 x half> %2, <vscale x 2 x half> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmfge_mask_vv_nxv2f16_nxv2f16:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e16, mf2, ta, mu
-; CHECK-NEXT:    vmfle.vv v8, v9, v8
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv1r.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e16, mf2, ta, mu
+; CHECK-NEXT:    vmfle.vv v0, v9, v8
 ; CHECK-NEXT:    vmfle.vv v11, v10, v9, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v11
 ; CHECK-NEXT:    ret
@@ -138,10 +136,9 @@ declare <vscale x 4 x i1> @llvm.riscv.vmfge.mask.nxv4f16(
 define <vscale x 4 x i1> @intrinsic_vmfge_mask_vv_nxv4f16_nxv4f16(<vscale x 4 x i1> %0, <vscale x 4 x half> %1, <vscale x 4 x half> %2, <vscale x 4 x half> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmfge_mask_vv_nxv4f16_nxv4f16:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e16, m1, ta, mu
-; CHECK-NEXT:    vmfle.vv v8, v9, v8
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv.v.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e16, m1, ta, mu
+; CHECK-NEXT:    vmfle.vv v0, v9, v8
 ; CHECK-NEXT:    vmfle.vv v11, v10, v9, v0.t
 ; CHECK-NEXT:    vmv.v.v v0, v11
 ; CHECK-NEXT:    ret
@@ -190,12 +187,11 @@ declare <vscale x 8 x i1> @llvm.riscv.vmfge.mask.nxv8f16(
 define <vscale x 8 x i1> @intrinsic_vmfge_mask_vv_nxv8f16_nxv8f16(<vscale x 8 x i1> %0, <vscale x 8 x half> %1, <vscale x 8 x half> %2, <vscale x 8 x half> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmfge_mask_vv_nxv8f16_nxv8f16:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v14, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e16, m2, ta, mu
-; CHECK-NEXT:    vmfle.vv v14, v10, v8
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmfle.vv v0, v10, v8
+; CHECK-NEXT:    vmfle.vv v14, v12, v10, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v14
-; CHECK-NEXT:    vmfle.vv v8, v12, v10, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 8 x i1> @llvm.riscv.vmfge.nxv8f16(
@@ -242,12 +238,11 @@ declare <vscale x 16 x i1> @llvm.riscv.vmfge.mask.nxv16f16(
 define <vscale x 16 x i1> @intrinsic_vmfge_mask_vv_nxv16f16_nxv16f16(<vscale x 16 x i1> %0, <vscale x 16 x half> %1, <vscale x 16 x half> %2, <vscale x 16 x half> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmfge_mask_vv_nxv16f16_nxv16f16:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v20, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e16, m4, ta, mu
-; CHECK-NEXT:    vmfle.vv v20, v12, v8
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmfle.vv v0, v12, v8
+; CHECK-NEXT:    vmfle.vv v20, v16, v12, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v20
-; CHECK-NEXT:    vmfle.vv v8, v16, v12, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 16 x i1> @llvm.riscv.vmfge.nxv16f16(
@@ -294,10 +289,9 @@ declare <vscale x 1 x i1> @llvm.riscv.vmfge.mask.nxv1f32(
 define <vscale x 1 x i1> @intrinsic_vmfge_mask_vv_nxv1f32_nxv1f32(<vscale x 1 x i1> %0, <vscale x 1 x float> %1, <vscale x 1 x float> %2, <vscale x 1 x float> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmfge_mask_vv_nxv1f32_nxv1f32:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e32, mf2, ta, mu
-; CHECK-NEXT:    vmfle.vv v8, v9, v8
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv1r.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e32, mf2, ta, mu
+; CHECK-NEXT:    vmfle.vv v0, v9, v8
 ; CHECK-NEXT:    vmfle.vv v11, v10, v9, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v11
 ; CHECK-NEXT:    ret
@@ -346,10 +340,9 @@ declare <vscale x 2 x i1> @llvm.riscv.vmfge.mask.nxv2f32(
 define <vscale x 2 x i1> @intrinsic_vmfge_mask_vv_nxv2f32_nxv2f32(<vscale x 2 x i1> %0, <vscale x 2 x float> %1, <vscale x 2 x float> %2, <vscale x 2 x float> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmfge_mask_vv_nxv2f32_nxv2f32:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e32, m1, ta, mu
-; CHECK-NEXT:    vmfle.vv v8, v9, v8
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv.v.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e32, m1, ta, mu
+; CHECK-NEXT:    vmfle.vv v0, v9, v8
 ; CHECK-NEXT:    vmfle.vv v11, v10, v9, v0.t
 ; CHECK-NEXT:    vmv.v.v v0, v11
 ; CHECK-NEXT:    ret
@@ -398,12 +391,11 @@ declare <vscale x 4 x i1> @llvm.riscv.vmfge.mask.nxv4f32(
 define <vscale x 4 x i1> @intrinsic_vmfge_mask_vv_nxv4f32_nxv4f32(<vscale x 4 x i1> %0, <vscale x 4 x float> %1, <vscale x 4 x float> %2, <vscale x 4 x float> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmfge_mask_vv_nxv4f32_nxv4f32:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v14, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e32, m2, ta, mu
-; CHECK-NEXT:    vmfle.vv v14, v10, v8
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmfle.vv v0, v10, v8
+; CHECK-NEXT:    vmfle.vv v14, v12, v10, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v14
-; CHECK-NEXT:    vmfle.vv v8, v12, v10, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 4 x i1> @llvm.riscv.vmfge.nxv4f32(
@@ -450,12 +442,11 @@ declare <vscale x 8 x i1> @llvm.riscv.vmfge.mask.nxv8f32(
 define <vscale x 8 x i1> @intrinsic_vmfge_mask_vv_nxv8f32_nxv8f32(<vscale x 8 x i1> %0, <vscale x 8 x float> %1, <vscale x 8 x float> %2, <vscale x 8 x float> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmfge_mask_vv_nxv8f32_nxv8f32:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v20, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e32, m4, ta, mu
-; CHECK-NEXT:    vmfle.vv v20, v12, v8
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmfle.vv v0, v12, v8
+; CHECK-NEXT:    vmfle.vv v20, v16, v12, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v20
-; CHECK-NEXT:    vmfle.vv v8, v16, v12, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 8 x i1> @llvm.riscv.vmfge.nxv8f32(
@@ -502,10 +493,9 @@ declare <vscale x 1 x i1> @llvm.riscv.vmfge.mask.nxv1f64(
 define <vscale x 1 x i1> @intrinsic_vmfge_mask_vv_nxv1f64_nxv1f64(<vscale x 1 x i1> %0, <vscale x 1 x double> %1, <vscale x 1 x double> %2, <vscale x 1 x double> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmfge_mask_vv_nxv1f64_nxv1f64:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e64, m1, ta, mu
-; CHECK-NEXT:    vmfle.vv v8, v9, v8
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv.v.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e64, m1, ta, mu
+; CHECK-NEXT:    vmfle.vv v0, v9, v8
 ; CHECK-NEXT:    vmfle.vv v11, v10, v9, v0.t
 ; CHECK-NEXT:    vmv.v.v v0, v11
 ; CHECK-NEXT:    ret
@@ -554,12 +544,11 @@ declare <vscale x 2 x i1> @llvm.riscv.vmfge.mask.nxv2f64(
 define <vscale x 2 x i1> @intrinsic_vmfge_mask_vv_nxv2f64_nxv2f64(<vscale x 2 x i1> %0, <vscale x 2 x double> %1, <vscale x 2 x double> %2, <vscale x 2 x double> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmfge_mask_vv_nxv2f64_nxv2f64:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v14, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e64, m2, ta, mu
-; CHECK-NEXT:    vmfle.vv v14, v10, v8
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmfle.vv v0, v10, v8
+; CHECK-NEXT:    vmfle.vv v14, v12, v10, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v14
-; CHECK-NEXT:    vmfle.vv v8, v12, v10, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 2 x i1> @llvm.riscv.vmfge.nxv2f64(
@@ -606,12 +595,11 @@ declare <vscale x 4 x i1> @llvm.riscv.vmfge.mask.nxv4f64(
 define <vscale x 4 x i1> @intrinsic_vmfge_mask_vv_nxv4f64_nxv4f64(<vscale x 4 x i1> %0, <vscale x 4 x double> %1, <vscale x 4 x double> %2, <vscale x 4 x double> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmfge_mask_vv_nxv4f64_nxv4f64:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v20, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e64, m4, ta, mu
-; CHECK-NEXT:    vmfle.vv v20, v12, v8
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmfle.vv v0, v12, v8
+; CHECK-NEXT:    vmfle.vv v20, v16, v12, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v20
-; CHECK-NEXT:    vmfle.vv v8, v16, v12, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 4 x i1> @llvm.riscv.vmfge.nxv4f64(
diff --git a/llvm/test/CodeGen/RISCV/rvv/vmfgt.ll b/llvm/test/CodeGen/RISCV/rvv/vmfgt.ll
index 427f0eb28e7d..f643a4036381 100644
--- a/llvm/test/CodeGen/RISCV/rvv/vmfgt.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/vmfgt.ll
@@ -34,10 +34,9 @@ declare <vscale x 1 x i1> @llvm.riscv.vmfgt.mask.nxv1f16(
 define <vscale x 1 x i1> @intrinsic_vmfgt_mask_vv_nxv1f16_nxv1f16(<vscale x 1 x i1> %0, <vscale x 1 x half> %1, <vscale x 1 x half> %2, <vscale x 1 x half> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmfgt_mask_vv_nxv1f16_nxv1f16:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
-; CHECK-NEXT:    vmflt.vv v8, v9, v8
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv1r.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
+; CHECK-NEXT:    vmflt.vv v0, v9, v8
 ; CHECK-NEXT:    vmflt.vv v11, v10, v9, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v11
 ; CHECK-NEXT:    ret
@@ -86,10 +85,9 @@ declare <vscale x 2 x i1> @llvm.riscv.vmfgt.mask.nxv2f16(
 define <vscale x 2 x i1> @intrinsic_vmfgt_mask_vv_nxv2f16_nxv2f16(<vscale x 2 x i1> %0, <vscale x 2 x half> %1, <vscale x 2 x half> %2, <vscale x 2 x half> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmfgt_mask_vv_nxv2f16_nxv2f16:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e16, mf2, ta, mu
-; CHECK-NEXT:    vmflt.vv v8, v9, v8
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv1r.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e16, mf2, ta, mu
+; CHECK-NEXT:    vmflt.vv v0, v9, v8
 ; CHECK-NEXT:    vmflt.vv v11, v10, v9, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v11
 ; CHECK-NEXT:    ret
@@ -138,10 +136,9 @@ declare <vscale x 4 x i1> @llvm.riscv.vmfgt.mask.nxv4f16(
 define <vscale x 4 x i1> @intrinsic_vmfgt_mask_vv_nxv4f16_nxv4f16(<vscale x 4 x i1> %0, <vscale x 4 x half> %1, <vscale x 4 x half> %2, <vscale x 4 x half> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmfgt_mask_vv_nxv4f16_nxv4f16:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e16, m1, ta, mu
-; CHECK-NEXT:    vmflt.vv v8, v9, v8
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv.v.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e16, m1, ta, mu
+; CHECK-NEXT:    vmflt.vv v0, v9, v8
 ; CHECK-NEXT:    vmflt.vv v11, v10, v9, v0.t
 ; CHECK-NEXT:    vmv.v.v v0, v11
 ; CHECK-NEXT:    ret
@@ -190,12 +187,11 @@ declare <vscale x 8 x i1> @llvm.riscv.vmfgt.mask.nxv8f16(
 define <vscale x 8 x i1> @intrinsic_vmfgt_mask_vv_nxv8f16_nxv8f16(<vscale x 8 x i1> %0, <vscale x 8 x half> %1, <vscale x 8 x half> %2, <vscale x 8 x half> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmfgt_mask_vv_nxv8f16_nxv8f16:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v14, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e16, m2, ta, mu
-; CHECK-NEXT:    vmflt.vv v14, v10, v8
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmflt.vv v0, v10, v8
+; CHECK-NEXT:    vmflt.vv v14, v12, v10, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v14
-; CHECK-NEXT:    vmflt.vv v8, v12, v10, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 8 x i1> @llvm.riscv.vmfgt.nxv8f16(
@@ -242,12 +238,11 @@ declare <vscale x 16 x i1> @llvm.riscv.vmfgt.mask.nxv16f16(
 define <vscale x 16 x i1> @intrinsic_vmfgt_mask_vv_nxv16f16_nxv16f16(<vscale x 16 x i1> %0, <vscale x 16 x half> %1, <vscale x 16 x half> %2, <vscale x 16 x half> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmfgt_mask_vv_nxv16f16_nxv16f16:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v20, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e16, m4, ta, mu
-; CHECK-NEXT:    vmflt.vv v20, v12, v8
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmflt.vv v0, v12, v8
+; CHECK-NEXT:    vmflt.vv v20, v16, v12, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v20
-; CHECK-NEXT:    vmflt.vv v8, v16, v12, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 16 x i1> @llvm.riscv.vmfgt.nxv16f16(
@@ -294,10 +289,9 @@ declare <vscale x 1 x i1> @llvm.riscv.vmfgt.mask.nxv1f32(
 define <vscale x 1 x i1> @intrinsic_vmfgt_mask_vv_nxv1f32_nxv1f32(<vscale x 1 x i1> %0, <vscale x 1 x float> %1, <vscale x 1 x float> %2, <vscale x 1 x float> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmfgt_mask_vv_nxv1f32_nxv1f32:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e32, mf2, ta, mu
-; CHECK-NEXT:    vmflt.vv v8, v9, v8
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv1r.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e32, mf2, ta, mu
+; CHECK-NEXT:    vmflt.vv v0, v9, v8
 ; CHECK-NEXT:    vmflt.vv v11, v10, v9, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v11
 ; CHECK-NEXT:    ret
@@ -346,10 +340,9 @@ declare <vscale x 2 x i1> @llvm.riscv.vmfgt.mask.nxv2f32(
 define <vscale x 2 x i1> @intrinsic_vmfgt_mask_vv_nxv2f32_nxv2f32(<vscale x 2 x i1> %0, <vscale x 2 x float> %1, <vscale x 2 x float> %2, <vscale x 2 x float> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmfgt_mask_vv_nxv2f32_nxv2f32:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e32, m1, ta, mu
-; CHECK-NEXT:    vmflt.vv v8, v9, v8
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv.v.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e32, m1, ta, mu
+; CHECK-NEXT:    vmflt.vv v0, v9, v8
 ; CHECK-NEXT:    vmflt.vv v11, v10, v9, v0.t
 ; CHECK-NEXT:    vmv.v.v v0, v11
 ; CHECK-NEXT:    ret
@@ -398,12 +391,11 @@ declare <vscale x 4 x i1> @llvm.riscv.vmfgt.mask.nxv4f32(
 define <vscale x 4 x i1> @intrinsic_vmfgt_mask_vv_nxv4f32_nxv4f32(<vscale x 4 x i1> %0, <vscale x 4 x float> %1, <vscale x 4 x float> %2, <vscale x 4 x float> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmfgt_mask_vv_nxv4f32_nxv4f32:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v14, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e32, m2, ta, mu
-; CHECK-NEXT:    vmflt.vv v14, v10, v8
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmflt.vv v0, v10, v8
+; CHECK-NEXT:    vmflt.vv v14, v12, v10, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v14
-; CHECK-NEXT:    vmflt.vv v8, v12, v10, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 4 x i1> @llvm.riscv.vmfgt.nxv4f32(
@@ -450,12 +442,11 @@ declare <vscale x 8 x i1> @llvm.riscv.vmfgt.mask.nxv8f32(
 define <vscale x 8 x i1> @intrinsic_vmfgt_mask_vv_nxv8f32_nxv8f32(<vscale x 8 x i1> %0, <vscale x 8 x float> %1, <vscale x 8 x float> %2, <vscale x 8 x float> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmfgt_mask_vv_nxv8f32_nxv8f32:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v20, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e32, m4, ta, mu
-; CHECK-NEXT:    vmflt.vv v20, v12, v8
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmflt.vv v0, v12, v8
+; CHECK-NEXT:    vmflt.vv v20, v16, v12, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v20
-; CHECK-NEXT:    vmflt.vv v8, v16, v12, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 8 x i1> @llvm.riscv.vmfgt.nxv8f32(
@@ -502,10 +493,9 @@ declare <vscale x 1 x i1> @llvm.riscv.vmfgt.mask.nxv1f64(
 define <vscale x 1 x i1> @intrinsic_vmfgt_mask_vv_nxv1f64_nxv1f64(<vscale x 1 x i1> %0, <vscale x 1 x double> %1, <vscale x 1 x double> %2, <vscale x 1 x double> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmfgt_mask_vv_nxv1f64_nxv1f64:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e64, m1, ta, mu
-; CHECK-NEXT:    vmflt.vv v8, v9, v8
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv.v.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e64, m1, ta, mu
+; CHECK-NEXT:    vmflt.vv v0, v9, v8
 ; CHECK-NEXT:    vmflt.vv v11, v10, v9, v0.t
 ; CHECK-NEXT:    vmv.v.v v0, v11
 ; CHECK-NEXT:    ret
@@ -554,12 +544,11 @@ declare <vscale x 2 x i1> @llvm.riscv.vmfgt.mask.nxv2f64(
 define <vscale x 2 x i1> @intrinsic_vmfgt_mask_vv_nxv2f64_nxv2f64(<vscale x 2 x i1> %0, <vscale x 2 x double> %1, <vscale x 2 x double> %2, <vscale x 2 x double> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmfgt_mask_vv_nxv2f64_nxv2f64:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v14, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e64, m2, ta, mu
-; CHECK-NEXT:    vmflt.vv v14, v10, v8
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmflt.vv v0, v10, v8
+; CHECK-NEXT:    vmflt.vv v14, v12, v10, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v14
-; CHECK-NEXT:    vmflt.vv v8, v12, v10, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 2 x i1> @llvm.riscv.vmfgt.nxv2f64(
@@ -606,12 +595,11 @@ declare <vscale x 4 x i1> @llvm.riscv.vmfgt.mask.nxv4f64(
 define <vscale x 4 x i1> @intrinsic_vmfgt_mask_vv_nxv4f64_nxv4f64(<vscale x 4 x i1> %0, <vscale x 4 x double> %1, <vscale x 4 x double> %2, <vscale x 4 x double> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmfgt_mask_vv_nxv4f64_nxv4f64:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v20, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e64, m4, ta, mu
-; CHECK-NEXT:    vmflt.vv v20, v12, v8
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmflt.vv v0, v12, v8
+; CHECK-NEXT:    vmflt.vv v20, v16, v12, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v20
-; CHECK-NEXT:    vmflt.vv v8, v16, v12, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 4 x i1> @llvm.riscv.vmfgt.nxv4f64(
diff --git a/llvm/test/CodeGen/RISCV/rvv/vmfle.ll b/llvm/test/CodeGen/RISCV/rvv/vmfle.ll
index e5327632fc04..6c52364c1fbd 100644
--- a/llvm/test/CodeGen/RISCV/rvv/vmfle.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/vmfle.ll
@@ -34,10 +34,9 @@ declare <vscale x 1 x i1> @llvm.riscv.vmfle.mask.nxv1f16(
 define <vscale x 1 x i1> @intrinsic_vmfle_mask_vv_nxv1f16_nxv1f16(<vscale x 1 x i1> %0, <vscale x 1 x half> %1, <vscale x 1 x half> %2, <vscale x 1 x half> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmfle_mask_vv_nxv1f16_nxv1f16:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
-; CHECK-NEXT:    vmfle.vv v8, v8, v9
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv1r.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
+; CHECK-NEXT:    vmfle.vv v0, v8, v9
 ; CHECK-NEXT:    vmfle.vv v11, v9, v10, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v11
 ; CHECK-NEXT:    ret
@@ -86,10 +85,9 @@ declare <vscale x 2 x i1> @llvm.riscv.vmfle.mask.nxv2f16(
 define <vscale x 2 x i1> @intrinsic_vmfle_mask_vv_nxv2f16_nxv2f16(<vscale x 2 x i1> %0, <vscale x 2 x half> %1, <vscale x 2 x half> %2, <vscale x 2 x half> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmfle_mask_vv_nxv2f16_nxv2f16:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e16, mf2, ta, mu
-; CHECK-NEXT:    vmfle.vv v8, v8, v9
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv1r.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e16, mf2, ta, mu
+; CHECK-NEXT:    vmfle.vv v0, v8, v9
 ; CHECK-NEXT:    vmfle.vv v11, v9, v10, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v11
 ; CHECK-NEXT:    ret
@@ -138,10 +136,9 @@ declare <vscale x 4 x i1> @llvm.riscv.vmfle.mask.nxv4f16(
 define <vscale x 4 x i1> @intrinsic_vmfle_mask_vv_nxv4f16_nxv4f16(<vscale x 4 x i1> %0, <vscale x 4 x half> %1, <vscale x 4 x half> %2, <vscale x 4 x half> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmfle_mask_vv_nxv4f16_nxv4f16:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e16, m1, ta, mu
-; CHECK-NEXT:    vmfle.vv v8, v8, v9
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv.v.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e16, m1, ta, mu
+; CHECK-NEXT:    vmfle.vv v0, v8, v9
 ; CHECK-NEXT:    vmfle.vv v11, v9, v10, v0.t
 ; CHECK-NEXT:    vmv.v.v v0, v11
 ; CHECK-NEXT:    ret
@@ -190,12 +187,11 @@ declare <vscale x 8 x i1> @llvm.riscv.vmfle.mask.nxv8f16(
 define <vscale x 8 x i1> @intrinsic_vmfle_mask_vv_nxv8f16_nxv8f16(<vscale x 8 x i1> %0, <vscale x 8 x half> %1, <vscale x 8 x half> %2, <vscale x 8 x half> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmfle_mask_vv_nxv8f16_nxv8f16:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v14, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e16, m2, ta, mu
-; CHECK-NEXT:    vmfle.vv v14, v8, v10
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmfle.vv v0, v8, v10
+; CHECK-NEXT:    vmfle.vv v14, v10, v12, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v14
-; CHECK-NEXT:    vmfle.vv v8, v10, v12, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 8 x i1> @llvm.riscv.vmfle.nxv8f16(
@@ -242,12 +238,11 @@ declare <vscale x 16 x i1> @llvm.riscv.vmfle.mask.nxv16f16(
 define <vscale x 16 x i1> @intrinsic_vmfle_mask_vv_nxv16f16_nxv16f16(<vscale x 16 x i1> %0, <vscale x 16 x half> %1, <vscale x 16 x half> %2, <vscale x 16 x half> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmfle_mask_vv_nxv16f16_nxv16f16:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v20, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e16, m4, ta, mu
-; CHECK-NEXT:    vmfle.vv v20, v8, v12
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmfle.vv v0, v8, v12
+; CHECK-NEXT:    vmfle.vv v20, v12, v16, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v20
-; CHECK-NEXT:    vmfle.vv v8, v12, v16, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 16 x i1> @llvm.riscv.vmfle.nxv16f16(
@@ -294,10 +289,9 @@ declare <vscale x 1 x i1> @llvm.riscv.vmfle.mask.nxv1f32(
 define <vscale x 1 x i1> @intrinsic_vmfle_mask_vv_nxv1f32_nxv1f32(<vscale x 1 x i1> %0, <vscale x 1 x float> %1, <vscale x 1 x float> %2, <vscale x 1 x float> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmfle_mask_vv_nxv1f32_nxv1f32:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e32, mf2, ta, mu
-; CHECK-NEXT:    vmfle.vv v8, v8, v9
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv1r.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e32, mf2, ta, mu
+; CHECK-NEXT:    vmfle.vv v0, v8, v9
 ; CHECK-NEXT:    vmfle.vv v11, v9, v10, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v11
 ; CHECK-NEXT:    ret
@@ -346,10 +340,9 @@ declare <vscale x 2 x i1> @llvm.riscv.vmfle.mask.nxv2f32(
 define <vscale x 2 x i1> @intrinsic_vmfle_mask_vv_nxv2f32_nxv2f32(<vscale x 2 x i1> %0, <vscale x 2 x float> %1, <vscale x 2 x float> %2, <vscale x 2 x float> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmfle_mask_vv_nxv2f32_nxv2f32:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e32, m1, ta, mu
-; CHECK-NEXT:    vmfle.vv v8, v8, v9
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv.v.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e32, m1, ta, mu
+; CHECK-NEXT:    vmfle.vv v0, v8, v9
 ; CHECK-NEXT:    vmfle.vv v11, v9, v10, v0.t
 ; CHECK-NEXT:    vmv.v.v v0, v11
 ; CHECK-NEXT:    ret
@@ -398,12 +391,11 @@ declare <vscale x 4 x i1> @llvm.riscv.vmfle.mask.nxv4f32(
 define <vscale x 4 x i1> @intrinsic_vmfle_mask_vv_nxv4f32_nxv4f32(<vscale x 4 x i1> %0, <vscale x 4 x float> %1, <vscale x 4 x float> %2, <vscale x 4 x float> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmfle_mask_vv_nxv4f32_nxv4f32:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v14, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e32, m2, ta, mu
-; CHECK-NEXT:    vmfle.vv v14, v8, v10
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmfle.vv v0, v8, v10
+; CHECK-NEXT:    vmfle.vv v14, v10, v12, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v14
-; CHECK-NEXT:    vmfle.vv v8, v10, v12, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 4 x i1> @llvm.riscv.vmfle.nxv4f32(
@@ -450,12 +442,11 @@ declare <vscale x 8 x i1> @llvm.riscv.vmfle.mask.nxv8f32(
 define <vscale x 8 x i1> @intrinsic_vmfle_mask_vv_nxv8f32_nxv8f32(<vscale x 8 x i1> %0, <vscale x 8 x float> %1, <vscale x 8 x float> %2, <vscale x 8 x float> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmfle_mask_vv_nxv8f32_nxv8f32:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v20, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e32, m4, ta, mu
-; CHECK-NEXT:    vmfle.vv v20, v8, v12
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmfle.vv v0, v8, v12
+; CHECK-NEXT:    vmfle.vv v20, v12, v16, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v20
-; CHECK-NEXT:    vmfle.vv v8, v12, v16, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 8 x i1> @llvm.riscv.vmfle.nxv8f32(
@@ -502,10 +493,9 @@ declare <vscale x 1 x i1> @llvm.riscv.vmfle.mask.nxv1f64(
 define <vscale x 1 x i1> @intrinsic_vmfle_mask_vv_nxv1f64_nxv1f64(<vscale x 1 x i1> %0, <vscale x 1 x double> %1, <vscale x 1 x double> %2, <vscale x 1 x double> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmfle_mask_vv_nxv1f64_nxv1f64:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e64, m1, ta, mu
-; CHECK-NEXT:    vmfle.vv v8, v8, v9
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv.v.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e64, m1, ta, mu
+; CHECK-NEXT:    vmfle.vv v0, v8, v9
 ; CHECK-NEXT:    vmfle.vv v11, v9, v10, v0.t
 ; CHECK-NEXT:    vmv.v.v v0, v11
 ; CHECK-NEXT:    ret
@@ -554,12 +544,11 @@ declare <vscale x 2 x i1> @llvm.riscv.vmfle.mask.nxv2f64(
 define <vscale x 2 x i1> @intrinsic_vmfle_mask_vv_nxv2f64_nxv2f64(<vscale x 2 x i1> %0, <vscale x 2 x double> %1, <vscale x 2 x double> %2, <vscale x 2 x double> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmfle_mask_vv_nxv2f64_nxv2f64:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v14, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e64, m2, ta, mu
-; CHECK-NEXT:    vmfle.vv v14, v8, v10
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmfle.vv v0, v8, v10
+; CHECK-NEXT:    vmfle.vv v14, v10, v12, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v14
-; CHECK-NEXT:    vmfle.vv v8, v10, v12, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 2 x i1> @llvm.riscv.vmfle.nxv2f64(
@@ -606,12 +595,11 @@ declare <vscale x 4 x i1> @llvm.riscv.vmfle.mask.nxv4f64(
 define <vscale x 4 x i1> @intrinsic_vmfle_mask_vv_nxv4f64_nxv4f64(<vscale x 4 x i1> %0, <vscale x 4 x double> %1, <vscale x 4 x double> %2, <vscale x 4 x double> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmfle_mask_vv_nxv4f64_nxv4f64:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v20, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e64, m4, ta, mu
-; CHECK-NEXT:    vmfle.vv v20, v8, v12
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmfle.vv v0, v8, v12
+; CHECK-NEXT:    vmfle.vv v20, v12, v16, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v20
-; CHECK-NEXT:    vmfle.vv v8, v12, v16, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 4 x i1> @llvm.riscv.vmfle.nxv4f64(
diff --git a/llvm/test/CodeGen/RISCV/rvv/vmflt.ll b/llvm/test/CodeGen/RISCV/rvv/vmflt.ll
index 64f257e355ce..37a9c6b081a1 100644
--- a/llvm/test/CodeGen/RISCV/rvv/vmflt.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/vmflt.ll
@@ -34,10 +34,9 @@ declare <vscale x 1 x i1> @llvm.riscv.vmflt.mask.nxv1f16(
 define <vscale x 1 x i1> @intrinsic_vmflt_mask_vv_nxv1f16_nxv1f16(<vscale x 1 x i1> %0, <vscale x 1 x half> %1, <vscale x 1 x half> %2, <vscale x 1 x half> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmflt_mask_vv_nxv1f16_nxv1f16:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
-; CHECK-NEXT:    vmflt.vv v8, v8, v9
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv1r.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
+; CHECK-NEXT:    vmflt.vv v0, v8, v9
 ; CHECK-NEXT:    vmflt.vv v11, v9, v10, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v11
 ; CHECK-NEXT:    ret
@@ -86,10 +85,9 @@ declare <vscale x 2 x i1> @llvm.riscv.vmflt.mask.nxv2f16(
 define <vscale x 2 x i1> @intrinsic_vmflt_mask_vv_nxv2f16_nxv2f16(<vscale x 2 x i1> %0, <vscale x 2 x half> %1, <vscale x 2 x half> %2, <vscale x 2 x half> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmflt_mask_vv_nxv2f16_nxv2f16:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e16, mf2, ta, mu
-; CHECK-NEXT:    vmflt.vv v8, v8, v9
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv1r.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e16, mf2, ta, mu
+; CHECK-NEXT:    vmflt.vv v0, v8, v9
 ; CHECK-NEXT:    vmflt.vv v11, v9, v10, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v11
 ; CHECK-NEXT:    ret
@@ -138,10 +136,9 @@ declare <vscale x 4 x i1> @llvm.riscv.vmflt.mask.nxv4f16(
 define <vscale x 4 x i1> @intrinsic_vmflt_mask_vv_nxv4f16_nxv4f16(<vscale x 4 x i1> %0, <vscale x 4 x half> %1, <vscale x 4 x half> %2, <vscale x 4 x half> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmflt_mask_vv_nxv4f16_nxv4f16:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e16, m1, ta, mu
-; CHECK-NEXT:    vmflt.vv v8, v8, v9
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv.v.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e16, m1, ta, mu
+; CHECK-NEXT:    vmflt.vv v0, v8, v9
 ; CHECK-NEXT:    vmflt.vv v11, v9, v10, v0.t
 ; CHECK-NEXT:    vmv.v.v v0, v11
 ; CHECK-NEXT:    ret
@@ -190,12 +187,11 @@ declare <vscale x 8 x i1> @llvm.riscv.vmflt.mask.nxv8f16(
 define <vscale x 8 x i1> @intrinsic_vmflt_mask_vv_nxv8f16_nxv8f16(<vscale x 8 x i1> %0, <vscale x 8 x half> %1, <vscale x 8 x half> %2, <vscale x 8 x half> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmflt_mask_vv_nxv8f16_nxv8f16:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v14, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e16, m2, ta, mu
-; CHECK-NEXT:    vmflt.vv v14, v8, v10
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmflt.vv v0, v8, v10
+; CHECK-NEXT:    vmflt.vv v14, v10, v12, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v14
-; CHECK-NEXT:    vmflt.vv v8, v10, v12, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 8 x i1> @llvm.riscv.vmflt.nxv8f16(
@@ -242,12 +238,11 @@ declare <vscale x 16 x i1> @llvm.riscv.vmflt.mask.nxv16f16(
 define <vscale x 16 x i1> @intrinsic_vmflt_mask_vv_nxv16f16_nxv16f16(<vscale x 16 x i1> %0, <vscale x 16 x half> %1, <vscale x 16 x half> %2, <vscale x 16 x half> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmflt_mask_vv_nxv16f16_nxv16f16:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v20, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e16, m4, ta, mu
-; CHECK-NEXT:    vmflt.vv v20, v8, v12
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmflt.vv v0, v8, v12
+; CHECK-NEXT:    vmflt.vv v20, v12, v16, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v20
-; CHECK-NEXT:    vmflt.vv v8, v12, v16, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 16 x i1> @llvm.riscv.vmflt.nxv16f16(
@@ -294,10 +289,9 @@ declare <vscale x 1 x i1> @llvm.riscv.vmflt.mask.nxv1f32(
 define <vscale x 1 x i1> @intrinsic_vmflt_mask_vv_nxv1f32_nxv1f32(<vscale x 1 x i1> %0, <vscale x 1 x float> %1, <vscale x 1 x float> %2, <vscale x 1 x float> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmflt_mask_vv_nxv1f32_nxv1f32:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e32, mf2, ta, mu
-; CHECK-NEXT:    vmflt.vv v8, v8, v9
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv1r.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e32, mf2, ta, mu
+; CHECK-NEXT:    vmflt.vv v0, v8, v9
 ; CHECK-NEXT:    vmflt.vv v11, v9, v10, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v11
 ; CHECK-NEXT:    ret
@@ -346,10 +340,9 @@ declare <vscale x 2 x i1> @llvm.riscv.vmflt.mask.nxv2f32(
 define <vscale x 2 x i1> @intrinsic_vmflt_mask_vv_nxv2f32_nxv2f32(<vscale x 2 x i1> %0, <vscale x 2 x float> %1, <vscale x 2 x float> %2, <vscale x 2 x float> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmflt_mask_vv_nxv2f32_nxv2f32:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e32, m1, ta, mu
-; CHECK-NEXT:    vmflt.vv v8, v8, v9
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv.v.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e32, m1, ta, mu
+; CHECK-NEXT:    vmflt.vv v0, v8, v9
 ; CHECK-NEXT:    vmflt.vv v11, v9, v10, v0.t
 ; CHECK-NEXT:    vmv.v.v v0, v11
 ; CHECK-NEXT:    ret
@@ -398,12 +391,11 @@ declare <vscale x 4 x i1> @llvm.riscv.vmflt.mask.nxv4f32(
 define <vscale x 4 x i1> @intrinsic_vmflt_mask_vv_nxv4f32_nxv4f32(<vscale x 4 x i1> %0, <vscale x 4 x float> %1, <vscale x 4 x float> %2, <vscale x 4 x float> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmflt_mask_vv_nxv4f32_nxv4f32:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v14, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e32, m2, ta, mu
-; CHECK-NEXT:    vmflt.vv v14, v8, v10
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmflt.vv v0, v8, v10
+; CHECK-NEXT:    vmflt.vv v14, v10, v12, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v14
-; CHECK-NEXT:    vmflt.vv v8, v10, v12, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 4 x i1> @llvm.riscv.vmflt.nxv4f32(
@@ -450,12 +442,11 @@ declare <vscale x 8 x i1> @llvm.riscv.vmflt.mask.nxv8f32(
 define <vscale x 8 x i1> @intrinsic_vmflt_mask_vv_nxv8f32_nxv8f32(<vscale x 8 x i1> %0, <vscale x 8 x float> %1, <vscale x 8 x float> %2, <vscale x 8 x float> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmflt_mask_vv_nxv8f32_nxv8f32:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v20, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e32, m4, ta, mu
-; CHECK-NEXT:    vmflt.vv v20, v8, v12
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmflt.vv v0, v8, v12
+; CHECK-NEXT:    vmflt.vv v20, v12, v16, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v20
-; CHECK-NEXT:    vmflt.vv v8, v12, v16, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 8 x i1> @llvm.riscv.vmflt.nxv8f32(
@@ -502,10 +493,9 @@ declare <vscale x 1 x i1> @llvm.riscv.vmflt.mask.nxv1f64(
 define <vscale x 1 x i1> @intrinsic_vmflt_mask_vv_nxv1f64_nxv1f64(<vscale x 1 x i1> %0, <vscale x 1 x double> %1, <vscale x 1 x double> %2, <vscale x 1 x double> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmflt_mask_vv_nxv1f64_nxv1f64:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e64, m1, ta, mu
-; CHECK-NEXT:    vmflt.vv v8, v8, v9
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv.v.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e64, m1, ta, mu
+; CHECK-NEXT:    vmflt.vv v0, v8, v9
 ; CHECK-NEXT:    vmflt.vv v11, v9, v10, v0.t
 ; CHECK-NEXT:    vmv.v.v v0, v11
 ; CHECK-NEXT:    ret
@@ -554,12 +544,11 @@ declare <vscale x 2 x i1> @llvm.riscv.vmflt.mask.nxv2f64(
 define <vscale x 2 x i1> @intrinsic_vmflt_mask_vv_nxv2f64_nxv2f64(<vscale x 2 x i1> %0, <vscale x 2 x double> %1, <vscale x 2 x double> %2, <vscale x 2 x double> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmflt_mask_vv_nxv2f64_nxv2f64:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v14, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e64, m2, ta, mu
-; CHECK-NEXT:    vmflt.vv v14, v8, v10
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmflt.vv v0, v8, v10
+; CHECK-NEXT:    vmflt.vv v14, v10, v12, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v14
-; CHECK-NEXT:    vmflt.vv v8, v10, v12, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 2 x i1> @llvm.riscv.vmflt.nxv2f64(
@@ -606,12 +595,11 @@ declare <vscale x 4 x i1> @llvm.riscv.vmflt.mask.nxv4f64(
 define <vscale x 4 x i1> @intrinsic_vmflt_mask_vv_nxv4f64_nxv4f64(<vscale x 4 x i1> %0, <vscale x 4 x double> %1, <vscale x 4 x double> %2, <vscale x 4 x double> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmflt_mask_vv_nxv4f64_nxv4f64:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v20, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e64, m4, ta, mu
-; CHECK-NEXT:    vmflt.vv v20, v8, v12
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmflt.vv v0, v8, v12
+; CHECK-NEXT:    vmflt.vv v20, v12, v16, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v20
-; CHECK-NEXT:    vmflt.vv v8, v12, v16, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 4 x i1> @llvm.riscv.vmflt.nxv4f64(
diff --git a/llvm/test/CodeGen/RISCV/rvv/vmfne.ll b/llvm/test/CodeGen/RISCV/rvv/vmfne.ll
index 6f6a2a5e8783..5defce42091e 100644
--- a/llvm/test/CodeGen/RISCV/rvv/vmfne.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/vmfne.ll
@@ -34,10 +34,9 @@ declare <vscale x 1 x i1> @llvm.riscv.vmfne.mask.nxv1f16(
 define <vscale x 1 x i1> @intrinsic_vmfne_mask_vv_nxv1f16_nxv1f16(<vscale x 1 x i1> %0, <vscale x 1 x half> %1, <vscale x 1 x half> %2, <vscale x 1 x half> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmfne_mask_vv_nxv1f16_nxv1f16:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
-; CHECK-NEXT:    vmfne.vv v8, v8, v9
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv1r.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
+; CHECK-NEXT:    vmfne.vv v0, v8, v9
 ; CHECK-NEXT:    vmfne.vv v11, v9, v10, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v11
 ; CHECK-NEXT:    ret
@@ -86,10 +85,9 @@ declare <vscale x 2 x i1> @llvm.riscv.vmfne.mask.nxv2f16(
 define <vscale x 2 x i1> @intrinsic_vmfne_mask_vv_nxv2f16_nxv2f16(<vscale x 2 x i1> %0, <vscale x 2 x half> %1, <vscale x 2 x half> %2, <vscale x 2 x half> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmfne_mask_vv_nxv2f16_nxv2f16:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e16, mf2, ta, mu
-; CHECK-NEXT:    vmfne.vv v8, v8, v9
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv1r.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e16, mf2, ta, mu
+; CHECK-NEXT:    vmfne.vv v0, v8, v9
 ; CHECK-NEXT:    vmfne.vv v11, v9, v10, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v11
 ; CHECK-NEXT:    ret
@@ -138,10 +136,9 @@ declare <vscale x 4 x i1> @llvm.riscv.vmfne.mask.nxv4f16(
 define <vscale x 4 x i1> @intrinsic_vmfne_mask_vv_nxv4f16_nxv4f16(<vscale x 4 x i1> %0, <vscale x 4 x half> %1, <vscale x 4 x half> %2, <vscale x 4 x half> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmfne_mask_vv_nxv4f16_nxv4f16:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e16, m1, ta, mu
-; CHECK-NEXT:    vmfne.vv v8, v8, v9
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv.v.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e16, m1, ta, mu
+; CHECK-NEXT:    vmfne.vv v0, v8, v9
 ; CHECK-NEXT:    vmfne.vv v11, v9, v10, v0.t
 ; CHECK-NEXT:    vmv.v.v v0, v11
 ; CHECK-NEXT:    ret
@@ -190,12 +187,11 @@ declare <vscale x 8 x i1> @llvm.riscv.vmfne.mask.nxv8f16(
 define <vscale x 8 x i1> @intrinsic_vmfne_mask_vv_nxv8f16_nxv8f16(<vscale x 8 x i1> %0, <vscale x 8 x half> %1, <vscale x 8 x half> %2, <vscale x 8 x half> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmfne_mask_vv_nxv8f16_nxv8f16:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v14, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e16, m2, ta, mu
-; CHECK-NEXT:    vmfne.vv v14, v8, v10
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmfne.vv v0, v8, v10
+; CHECK-NEXT:    vmfne.vv v14, v10, v12, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v14
-; CHECK-NEXT:    vmfne.vv v8, v10, v12, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 8 x i1> @llvm.riscv.vmfne.nxv8f16(
@@ -242,12 +238,11 @@ declare <vscale x 16 x i1> @llvm.riscv.vmfne.mask.nxv16f16(
 define <vscale x 16 x i1> @intrinsic_vmfne_mask_vv_nxv16f16_nxv16f16(<vscale x 16 x i1> %0, <vscale x 16 x half> %1, <vscale x 16 x half> %2, <vscale x 16 x half> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmfne_mask_vv_nxv16f16_nxv16f16:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v20, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e16, m4, ta, mu
-; CHECK-NEXT:    vmfne.vv v20, v8, v12
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmfne.vv v0, v8, v12
+; CHECK-NEXT:    vmfne.vv v20, v12, v16, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v20
-; CHECK-NEXT:    vmfne.vv v8, v12, v16, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 16 x i1> @llvm.riscv.vmfne.nxv16f16(
@@ -294,10 +289,9 @@ declare <vscale x 1 x i1> @llvm.riscv.vmfne.mask.nxv1f32(
 define <vscale x 1 x i1> @intrinsic_vmfne_mask_vv_nxv1f32_nxv1f32(<vscale x 1 x i1> %0, <vscale x 1 x float> %1, <vscale x 1 x float> %2, <vscale x 1 x float> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmfne_mask_vv_nxv1f32_nxv1f32:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e32, mf2, ta, mu
-; CHECK-NEXT:    vmfne.vv v8, v8, v9
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv1r.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e32, mf2, ta, mu
+; CHECK-NEXT:    vmfne.vv v0, v8, v9
 ; CHECK-NEXT:    vmfne.vv v11, v9, v10, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v11
 ; CHECK-NEXT:    ret
@@ -346,10 +340,9 @@ declare <vscale x 2 x i1> @llvm.riscv.vmfne.mask.nxv2f32(
 define <vscale x 2 x i1> @intrinsic_vmfne_mask_vv_nxv2f32_nxv2f32(<vscale x 2 x i1> %0, <vscale x 2 x float> %1, <vscale x 2 x float> %2, <vscale x 2 x float> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmfne_mask_vv_nxv2f32_nxv2f32:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e32, m1, ta, mu
-; CHECK-NEXT:    vmfne.vv v8, v8, v9
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv.v.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e32, m1, ta, mu
+; CHECK-NEXT:    vmfne.vv v0, v8, v9
 ; CHECK-NEXT:    vmfne.vv v11, v9, v10, v0.t
 ; CHECK-NEXT:    vmv.v.v v0, v11
 ; CHECK-NEXT:    ret
@@ -398,12 +391,11 @@ declare <vscale x 4 x i1> @llvm.riscv.vmfne.mask.nxv4f32(
 define <vscale x 4 x i1> @intrinsic_vmfne_mask_vv_nxv4f32_nxv4f32(<vscale x 4 x i1> %0, <vscale x 4 x float> %1, <vscale x 4 x float> %2, <vscale x 4 x float> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmfne_mask_vv_nxv4f32_nxv4f32:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v14, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e32, m2, ta, mu
-; CHECK-NEXT:    vmfne.vv v14, v8, v10
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmfne.vv v0, v8, v10
+; CHECK-NEXT:    vmfne.vv v14, v10, v12, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v14
-; CHECK-NEXT:    vmfne.vv v8, v10, v12, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 4 x i1> @llvm.riscv.vmfne.nxv4f32(
@@ -450,12 +442,11 @@ declare <vscale x 8 x i1> @llvm.riscv.vmfne.mask.nxv8f32(
 define <vscale x 8 x i1> @intrinsic_vmfne_mask_vv_nxv8f32_nxv8f32(<vscale x 8 x i1> %0, <vscale x 8 x float> %1, <vscale x 8 x float> %2, <vscale x 8 x float> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmfne_mask_vv_nxv8f32_nxv8f32:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v20, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e32, m4, ta, mu
-; CHECK-NEXT:    vmfne.vv v20, v8, v12
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmfne.vv v0, v8, v12
+; CHECK-NEXT:    vmfne.vv v20, v12, v16, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v20
-; CHECK-NEXT:    vmfne.vv v8, v12, v16, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 8 x i1> @llvm.riscv.vmfne.nxv8f32(
@@ -502,10 +493,9 @@ declare <vscale x 1 x i1> @llvm.riscv.vmfne.mask.nxv1f64(
 define <vscale x 1 x i1> @intrinsic_vmfne_mask_vv_nxv1f64_nxv1f64(<vscale x 1 x i1> %0, <vscale x 1 x double> %1, <vscale x 1 x double> %2, <vscale x 1 x double> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmfne_mask_vv_nxv1f64_nxv1f64:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e64, m1, ta, mu
-; CHECK-NEXT:    vmfne.vv v8, v8, v9
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv.v.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e64, m1, ta, mu
+; CHECK-NEXT:    vmfne.vv v0, v8, v9
 ; CHECK-NEXT:    vmfne.vv v11, v9, v10, v0.t
 ; CHECK-NEXT:    vmv.v.v v0, v11
 ; CHECK-NEXT:    ret
@@ -554,12 +544,11 @@ declare <vscale x 2 x i1> @llvm.riscv.vmfne.mask.nxv2f64(
 define <vscale x 2 x i1> @intrinsic_vmfne_mask_vv_nxv2f64_nxv2f64(<vscale x 2 x i1> %0, <vscale x 2 x double> %1, <vscale x 2 x double> %2, <vscale x 2 x double> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmfne_mask_vv_nxv2f64_nxv2f64:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v14, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e64, m2, ta, mu
-; CHECK-NEXT:    vmfne.vv v14, v8, v10
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmfne.vv v0, v8, v10
+; CHECK-NEXT:    vmfne.vv v14, v10, v12, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v14
-; CHECK-NEXT:    vmfne.vv v8, v10, v12, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 2 x i1> @llvm.riscv.vmfne.nxv2f64(
@@ -606,12 +595,11 @@ declare <vscale x 4 x i1> @llvm.riscv.vmfne.mask.nxv4f64(
 define <vscale x 4 x i1> @intrinsic_vmfne_mask_vv_nxv4f64_nxv4f64(<vscale x 4 x i1> %0, <vscale x 4 x double> %1, <vscale x 4 x double> %2, <vscale x 4 x double> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmfne_mask_vv_nxv4f64_nxv4f64:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v20, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e64, m4, ta, mu
-; CHECK-NEXT:    vmfne.vv v20, v8, v12
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmfne.vv v0, v8, v12
+; CHECK-NEXT:    vmfne.vv v20, v12, v16, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v20
-; CHECK-NEXT:    vmfne.vv v8, v12, v16, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 4 x i1> @llvm.riscv.vmfne.nxv4f64(
diff --git a/llvm/test/CodeGen/RISCV/rvv/vmseq.ll b/llvm/test/CodeGen/RISCV/rvv/vmseq.ll
index da1c751b5663..cc6c1f585bb7 100644
--- a/llvm/test/CodeGen/RISCV/rvv/vmseq.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/vmseq.ll
@@ -34,10 +34,9 @@ declare <vscale x 1 x i1> @llvm.riscv.vmseq.mask.nxv1i8(
 define <vscale x 1 x i1> @intrinsic_vmseq_mask_vv_nxv1i8_nxv1i8(<vscale x 1 x i1> %0, <vscale x 1 x i8> %1, <vscale x 1 x i8> %2, <vscale x 1 x i8> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmseq_mask_vv_nxv1i8_nxv1i8:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e8, mf8, ta, mu
-; CHECK-NEXT:    vmseq.vv v8, v8, v9
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv1r.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e8, mf8, ta, mu
+; CHECK-NEXT:    vmseq.vv v0, v8, v9
 ; CHECK-NEXT:    vmseq.vv v11, v9, v10, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v11
 ; CHECK-NEXT:    ret
@@ -86,10 +85,9 @@ declare <vscale x 2 x i1> @llvm.riscv.vmseq.mask.nxv2i8(
 define <vscale x 2 x i1> @intrinsic_vmseq_mask_vv_nxv2i8_nxv2i8(<vscale x 2 x i1> %0, <vscale x 2 x i8> %1, <vscale x 2 x i8> %2, <vscale x 2 x i8> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmseq_mask_vv_nxv2i8_nxv2i8:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e8, mf4, ta, mu
-; CHECK-NEXT:    vmseq.vv v8, v8, v9
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv1r.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e8, mf4, ta, mu
+; CHECK-NEXT:    vmseq.vv v0, v8, v9
 ; CHECK-NEXT:    vmseq.vv v11, v9, v10, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v11
 ; CHECK-NEXT:    ret
@@ -138,10 +136,9 @@ declare <vscale x 4 x i1> @llvm.riscv.vmseq.mask.nxv4i8(
 define <vscale x 4 x i1> @intrinsic_vmseq_mask_vv_nxv4i8_nxv4i8(<vscale x 4 x i1> %0, <vscale x 4 x i8> %1, <vscale x 4 x i8> %2, <vscale x 4 x i8> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmseq_mask_vv_nxv4i8_nxv4i8:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e8, mf2, ta, mu
-; CHECK-NEXT:    vmseq.vv v8, v8, v9
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv1r.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e8, mf2, ta, mu
+; CHECK-NEXT:    vmseq.vv v0, v8, v9
 ; CHECK-NEXT:    vmseq.vv v11, v9, v10, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v11
 ; CHECK-NEXT:    ret
@@ -190,10 +187,9 @@ declare <vscale x 8 x i1> @llvm.riscv.vmseq.mask.nxv8i8(
 define <vscale x 8 x i1> @intrinsic_vmseq_mask_vv_nxv8i8_nxv8i8(<vscale x 8 x i1> %0, <vscale x 8 x i8> %1, <vscale x 8 x i8> %2, <vscale x 8 x i8> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmseq_mask_vv_nxv8i8_nxv8i8:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e8, m1, ta, mu
-; CHECK-NEXT:    vmseq.vv v8, v8, v9
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv.v.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e8, m1, ta, mu
+; CHECK-NEXT:    vmseq.vv v0, v8, v9
 ; CHECK-NEXT:    vmseq.vv v11, v9, v10, v0.t
 ; CHECK-NEXT:    vmv.v.v v0, v11
 ; CHECK-NEXT:    ret
@@ -242,12 +238,11 @@ declare <vscale x 16 x i1> @llvm.riscv.vmseq.mask.nxv16i8(
 define <vscale x 16 x i1> @intrinsic_vmseq_mask_vv_nxv16i8_nxv16i8(<vscale x 16 x i1> %0, <vscale x 16 x i8> %1, <vscale x 16 x i8> %2, <vscale x 16 x i8> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmseq_mask_vv_nxv16i8_nxv16i8:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v14, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e8, m2, ta, mu
-; CHECK-NEXT:    vmseq.vv v14, v8, v10
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmseq.vv v0, v8, v10
+; CHECK-NEXT:    vmseq.vv v14, v10, v12, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v14
-; CHECK-NEXT:    vmseq.vv v8, v10, v12, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 16 x i1> @llvm.riscv.vmseq.nxv16i8(
@@ -294,12 +289,11 @@ declare <vscale x 32 x i1> @llvm.riscv.vmseq.mask.nxv32i8(
 define <vscale x 32 x i1> @intrinsic_vmseq_mask_vv_nxv32i8_nxv32i8(<vscale x 32 x i1> %0, <vscale x 32 x i8> %1, <vscale x 32 x i8> %2, <vscale x 32 x i8> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmseq_mask_vv_nxv32i8_nxv32i8:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v20, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e8, m4, ta, mu
-; CHECK-NEXT:    vmseq.vv v20, v8, v12
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmseq.vv v0, v8, v12
+; CHECK-NEXT:    vmseq.vv v20, v12, v16, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v20
-; CHECK-NEXT:    vmseq.vv v8, v12, v16, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 32 x i1> @llvm.riscv.vmseq.nxv32i8(
@@ -346,10 +340,9 @@ declare <vscale x 1 x i1> @llvm.riscv.vmseq.mask.nxv1i16(
 define <vscale x 1 x i1> @intrinsic_vmseq_mask_vv_nxv1i16_nxv1i16(<vscale x 1 x i1> %0, <vscale x 1 x i16> %1, <vscale x 1 x i16> %2, <vscale x 1 x i16> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmseq_mask_vv_nxv1i16_nxv1i16:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
-; CHECK-NEXT:    vmseq.vv v8, v8, v9
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv1r.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
+; CHECK-NEXT:    vmseq.vv v0, v8, v9
 ; CHECK-NEXT:    vmseq.vv v11, v9, v10, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v11
 ; CHECK-NEXT:    ret
@@ -398,10 +391,9 @@ declare <vscale x 2 x i1> @llvm.riscv.vmseq.mask.nxv2i16(
 define <vscale x 2 x i1> @intrinsic_vmseq_mask_vv_nxv2i16_nxv2i16(<vscale x 2 x i1> %0, <vscale x 2 x i16> %1, <vscale x 2 x i16> %2, <vscale x 2 x i16> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmseq_mask_vv_nxv2i16_nxv2i16:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e16, mf2, ta, mu
-; CHECK-NEXT:    vmseq.vv v8, v8, v9
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv1r.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e16, mf2, ta, mu
+; CHECK-NEXT:    vmseq.vv v0, v8, v9
 ; CHECK-NEXT:    vmseq.vv v11, v9, v10, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v11
 ; CHECK-NEXT:    ret
@@ -450,10 +442,9 @@ declare <vscale x 4 x i1> @llvm.riscv.vmseq.mask.nxv4i16(
 define <vscale x 4 x i1> @intrinsic_vmseq_mask_vv_nxv4i16_nxv4i16(<vscale x 4 x i1> %0, <vscale x 4 x i16> %1, <vscale x 4 x i16> %2, <vscale x 4 x i16> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmseq_mask_vv_nxv4i16_nxv4i16:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e16, m1, ta, mu
-; CHECK-NEXT:    vmseq.vv v8, v8, v9
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv.v.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e16, m1, ta, mu
+; CHECK-NEXT:    vmseq.vv v0, v8, v9
 ; CHECK-NEXT:    vmseq.vv v11, v9, v10, v0.t
 ; CHECK-NEXT:    vmv.v.v v0, v11
 ; CHECK-NEXT:    ret
@@ -502,12 +493,11 @@ declare <vscale x 8 x i1> @llvm.riscv.vmseq.mask.nxv8i16(
 define <vscale x 8 x i1> @intrinsic_vmseq_mask_vv_nxv8i16_nxv8i16(<vscale x 8 x i1> %0, <vscale x 8 x i16> %1, <vscale x 8 x i16> %2, <vscale x 8 x i16> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmseq_mask_vv_nxv8i16_nxv8i16:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v14, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e16, m2, ta, mu
-; CHECK-NEXT:    vmseq.vv v14, v8, v10
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmseq.vv v0, v8, v10
+; CHECK-NEXT:    vmseq.vv v14, v10, v12, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v14
-; CHECK-NEXT:    vmseq.vv v8, v10, v12, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 8 x i1> @llvm.riscv.vmseq.nxv8i16(
@@ -554,12 +544,11 @@ declare <vscale x 16 x i1> @llvm.riscv.vmseq.mask.nxv16i16(
 define <vscale x 16 x i1> @intrinsic_vmseq_mask_vv_nxv16i16_nxv16i16(<vscale x 16 x i1> %0, <vscale x 16 x i16> %1, <vscale x 16 x i16> %2, <vscale x 16 x i16> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmseq_mask_vv_nxv16i16_nxv16i16:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v20, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e16, m4, ta, mu
-; CHECK-NEXT:    vmseq.vv v20, v8, v12
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmseq.vv v0, v8, v12
+; CHECK-NEXT:    vmseq.vv v20, v12, v16, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v20
-; CHECK-NEXT:    vmseq.vv v8, v12, v16, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 16 x i1> @llvm.riscv.vmseq.nxv16i16(
@@ -606,10 +595,9 @@ declare <vscale x 1 x i1> @llvm.riscv.vmseq.mask.nxv1i32(
 define <vscale x 1 x i1> @intrinsic_vmseq_mask_vv_nxv1i32_nxv1i32(<vscale x 1 x i1> %0, <vscale x 1 x i32> %1, <vscale x 1 x i32> %2, <vscale x 1 x i32> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmseq_mask_vv_nxv1i32_nxv1i32:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e32, mf2, ta, mu
-; CHECK-NEXT:    vmseq.vv v8, v8, v9
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv1r.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e32, mf2, ta, mu
+; CHECK-NEXT:    vmseq.vv v0, v8, v9
 ; CHECK-NEXT:    vmseq.vv v11, v9, v10, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v11
 ; CHECK-NEXT:    ret
@@ -658,10 +646,9 @@ declare <vscale x 2 x i1> @llvm.riscv.vmseq.mask.nxv2i32(
 define <vscale x 2 x i1> @intrinsic_vmseq_mask_vv_nxv2i32_nxv2i32(<vscale x 2 x i1> %0, <vscale x 2 x i32> %1, <vscale x 2 x i32> %2, <vscale x 2 x i32> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmseq_mask_vv_nxv2i32_nxv2i32:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e32, m1, ta, mu
-; CHECK-NEXT:    vmseq.vv v8, v8, v9
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv.v.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e32, m1, ta, mu
+; CHECK-NEXT:    vmseq.vv v0, v8, v9
 ; CHECK-NEXT:    vmseq.vv v11, v9, v10, v0.t
 ; CHECK-NEXT:    vmv.v.v v0, v11
 ; CHECK-NEXT:    ret
@@ -710,12 +697,11 @@ declare <vscale x 4 x i1> @llvm.riscv.vmseq.mask.nxv4i32(
 define <vscale x 4 x i1> @intrinsic_vmseq_mask_vv_nxv4i32_nxv4i32(<vscale x 4 x i1> %0, <vscale x 4 x i32> %1, <vscale x 4 x i32> %2, <vscale x 4 x i32> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmseq_mask_vv_nxv4i32_nxv4i32:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v14, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e32, m2, ta, mu
-; CHECK-NEXT:    vmseq.vv v14, v8, v10
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmseq.vv v0, v8, v10
+; CHECK-NEXT:    vmseq.vv v14, v10, v12, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v14
-; CHECK-NEXT:    vmseq.vv v8, v10, v12, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 4 x i1> @llvm.riscv.vmseq.nxv4i32(
@@ -762,12 +748,11 @@ declare <vscale x 8 x i1> @llvm.riscv.vmseq.mask.nxv8i32(
 define <vscale x 8 x i1> @intrinsic_vmseq_mask_vv_nxv8i32_nxv8i32(<vscale x 8 x i1> %0, <vscale x 8 x i32> %1, <vscale x 8 x i32> %2, <vscale x 8 x i32> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmseq_mask_vv_nxv8i32_nxv8i32:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v20, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e32, m4, ta, mu
-; CHECK-NEXT:    vmseq.vv v20, v8, v12
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmseq.vv v0, v8, v12
+; CHECK-NEXT:    vmseq.vv v20, v12, v16, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v20
-; CHECK-NEXT:    vmseq.vv v8, v12, v16, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 8 x i1> @llvm.riscv.vmseq.nxv8i32(
@@ -814,10 +799,9 @@ declare <vscale x 1 x i1> @llvm.riscv.vmseq.mask.nxv1i64(
 define <vscale x 1 x i1> @intrinsic_vmseq_mask_vv_nxv1i64_nxv1i64(<vscale x 1 x i1> %0, <vscale x 1 x i64> %1, <vscale x 1 x i64> %2, <vscale x 1 x i64> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmseq_mask_vv_nxv1i64_nxv1i64:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e64, m1, ta, mu
-; CHECK-NEXT:    vmseq.vv v8, v8, v9
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv.v.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e64, m1, ta, mu
+; CHECK-NEXT:    vmseq.vv v0, v8, v9
 ; CHECK-NEXT:    vmseq.vv v11, v9, v10, v0.t
 ; CHECK-NEXT:    vmv.v.v v0, v11
 ; CHECK-NEXT:    ret
@@ -866,12 +850,11 @@ declare <vscale x 2 x i1> @llvm.riscv.vmseq.mask.nxv2i64(
 define <vscale x 2 x i1> @intrinsic_vmseq_mask_vv_nxv2i64_nxv2i64(<vscale x 2 x i1> %0, <vscale x 2 x i64> %1, <vscale x 2 x i64> %2, <vscale x 2 x i64> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmseq_mask_vv_nxv2i64_nxv2i64:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v14, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e64, m2, ta, mu
-; CHECK-NEXT:    vmseq.vv v14, v8, v10
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmseq.vv v0, v8, v10
+; CHECK-NEXT:    vmseq.vv v14, v10, v12, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v14
-; CHECK-NEXT:    vmseq.vv v8, v10, v12, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 2 x i1> @llvm.riscv.vmseq.nxv2i64(
@@ -918,12 +901,11 @@ declare <vscale x 4 x i1> @llvm.riscv.vmseq.mask.nxv4i64(
 define <vscale x 4 x i1> @intrinsic_vmseq_mask_vv_nxv4i64_nxv4i64(<vscale x 4 x i1> %0, <vscale x 4 x i64> %1, <vscale x 4 x i64> %2, <vscale x 4 x i64> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmseq_mask_vv_nxv4i64_nxv4i64:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v20, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e64, m4, ta, mu
-; CHECK-NEXT:    vmseq.vv v20, v8, v12
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmseq.vv v0, v8, v12
+; CHECK-NEXT:    vmseq.vv v20, v12, v16, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v20
-; CHECK-NEXT:    vmseq.vv v8, v12, v16, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 4 x i1> @llvm.riscv.vmseq.nxv4i64(
diff --git a/llvm/test/CodeGen/RISCV/rvv/vmsge.ll b/llvm/test/CodeGen/RISCV/rvv/vmsge.ll
index 502fb9b24148..c8f9b60a3f2d 100644
--- a/llvm/test/CodeGen/RISCV/rvv/vmsge.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/vmsge.ll
@@ -34,10 +34,9 @@ declare <vscale x 1 x i1> @llvm.riscv.vmsge.mask.nxv1i8(
 define <vscale x 1 x i1> @intrinsic_vmsge_mask_vv_nxv1i8_nxv1i8(<vscale x 1 x i1> %0, <vscale x 1 x i8> %1, <vscale x 1 x i8> %2, <vscale x 1 x i8> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsge_mask_vv_nxv1i8_nxv1i8:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e8, mf8, ta, mu
-; CHECK-NEXT:    vmsle.vv v8, v9, v8
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv1r.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e8, mf8, ta, mu
+; CHECK-NEXT:    vmsle.vv v0, v9, v8
 ; CHECK-NEXT:    vmsle.vv v11, v10, v9, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v11
 ; CHECK-NEXT:    ret
@@ -86,10 +85,9 @@ declare <vscale x 2 x i1> @llvm.riscv.vmsge.mask.nxv2i8(
 define <vscale x 2 x i1> @intrinsic_vmsge_mask_vv_nxv2i8_nxv2i8(<vscale x 2 x i1> %0, <vscale x 2 x i8> %1, <vscale x 2 x i8> %2, <vscale x 2 x i8> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsge_mask_vv_nxv2i8_nxv2i8:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e8, mf4, ta, mu
-; CHECK-NEXT:    vmsle.vv v8, v9, v8
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv1r.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e8, mf4, ta, mu
+; CHECK-NEXT:    vmsle.vv v0, v9, v8
 ; CHECK-NEXT:    vmsle.vv v11, v10, v9, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v11
 ; CHECK-NEXT:    ret
@@ -138,10 +136,9 @@ declare <vscale x 4 x i1> @llvm.riscv.vmsge.mask.nxv4i8(
 define <vscale x 4 x i1> @intrinsic_vmsge_mask_vv_nxv4i8_nxv4i8(<vscale x 4 x i1> %0, <vscale x 4 x i8> %1, <vscale x 4 x i8> %2, <vscale x 4 x i8> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsge_mask_vv_nxv4i8_nxv4i8:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e8, mf2, ta, mu
-; CHECK-NEXT:    vmsle.vv v8, v9, v8
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv1r.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e8, mf2, ta, mu
+; CHECK-NEXT:    vmsle.vv v0, v9, v8
 ; CHECK-NEXT:    vmsle.vv v11, v10, v9, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v11
 ; CHECK-NEXT:    ret
@@ -190,10 +187,9 @@ declare <vscale x 8 x i1> @llvm.riscv.vmsge.mask.nxv8i8(
 define <vscale x 8 x i1> @intrinsic_vmsge_mask_vv_nxv8i8_nxv8i8(<vscale x 8 x i1> %0, <vscale x 8 x i8> %1, <vscale x 8 x i8> %2, <vscale x 8 x i8> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsge_mask_vv_nxv8i8_nxv8i8:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e8, m1, ta, mu
-; CHECK-NEXT:    vmsle.vv v8, v9, v8
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv.v.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e8, m1, ta, mu
+; CHECK-NEXT:    vmsle.vv v0, v9, v8
 ; CHECK-NEXT:    vmsle.vv v11, v10, v9, v0.t
 ; CHECK-NEXT:    vmv.v.v v0, v11
 ; CHECK-NEXT:    ret
@@ -242,12 +238,11 @@ declare <vscale x 16 x i1> @llvm.riscv.vmsge.mask.nxv16i8(
 define <vscale x 16 x i1> @intrinsic_vmsge_mask_vv_nxv16i8_nxv16i8(<vscale x 16 x i1> %0, <vscale x 16 x i8> %1, <vscale x 16 x i8> %2, <vscale x 16 x i8> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsge_mask_vv_nxv16i8_nxv16i8:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v14, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e8, m2, ta, mu
-; CHECK-NEXT:    vmsle.vv v14, v10, v8
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmsle.vv v0, v10, v8
+; CHECK-NEXT:    vmsle.vv v14, v12, v10, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v14
-; CHECK-NEXT:    vmsle.vv v8, v12, v10, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 16 x i1> @llvm.riscv.vmsge.nxv16i8(
@@ -294,12 +289,11 @@ declare <vscale x 32 x i1> @llvm.riscv.vmsge.mask.nxv32i8(
 define <vscale x 32 x i1> @intrinsic_vmsge_mask_vv_nxv32i8_nxv32i8(<vscale x 32 x i1> %0, <vscale x 32 x i8> %1, <vscale x 32 x i8> %2, <vscale x 32 x i8> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsge_mask_vv_nxv32i8_nxv32i8:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v20, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e8, m4, ta, mu
-; CHECK-NEXT:    vmsle.vv v20, v12, v8
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmsle.vv v0, v12, v8
+; CHECK-NEXT:    vmsle.vv v20, v16, v12, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v20
-; CHECK-NEXT:    vmsle.vv v8, v16, v12, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 32 x i1> @llvm.riscv.vmsge.nxv32i8(
@@ -346,10 +340,9 @@ declare <vscale x 1 x i1> @llvm.riscv.vmsge.mask.nxv1i16(
 define <vscale x 1 x i1> @intrinsic_vmsge_mask_vv_nxv1i16_nxv1i16(<vscale x 1 x i1> %0, <vscale x 1 x i16> %1, <vscale x 1 x i16> %2, <vscale x 1 x i16> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsge_mask_vv_nxv1i16_nxv1i16:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
-; CHECK-NEXT:    vmsle.vv v8, v9, v8
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv1r.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
+; CHECK-NEXT:    vmsle.vv v0, v9, v8
 ; CHECK-NEXT:    vmsle.vv v11, v10, v9, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v11
 ; CHECK-NEXT:    ret
@@ -398,10 +391,9 @@ declare <vscale x 2 x i1> @llvm.riscv.vmsge.mask.nxv2i16(
 define <vscale x 2 x i1> @intrinsic_vmsge_mask_vv_nxv2i16_nxv2i16(<vscale x 2 x i1> %0, <vscale x 2 x i16> %1, <vscale x 2 x i16> %2, <vscale x 2 x i16> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsge_mask_vv_nxv2i16_nxv2i16:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e16, mf2, ta, mu
-; CHECK-NEXT:    vmsle.vv v8, v9, v8
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv1r.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e16, mf2, ta, mu
+; CHECK-NEXT:    vmsle.vv v0, v9, v8
 ; CHECK-NEXT:    vmsle.vv v11, v10, v9, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v11
 ; CHECK-NEXT:    ret
@@ -450,10 +442,9 @@ declare <vscale x 4 x i1> @llvm.riscv.vmsge.mask.nxv4i16(
 define <vscale x 4 x i1> @intrinsic_vmsge_mask_vv_nxv4i16_nxv4i16(<vscale x 4 x i1> %0, <vscale x 4 x i16> %1, <vscale x 4 x i16> %2, <vscale x 4 x i16> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsge_mask_vv_nxv4i16_nxv4i16:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e16, m1, ta, mu
-; CHECK-NEXT:    vmsle.vv v8, v9, v8
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv.v.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e16, m1, ta, mu
+; CHECK-NEXT:    vmsle.vv v0, v9, v8
 ; CHECK-NEXT:    vmsle.vv v11, v10, v9, v0.t
 ; CHECK-NEXT:    vmv.v.v v0, v11
 ; CHECK-NEXT:    ret
@@ -502,12 +493,11 @@ declare <vscale x 8 x i1> @llvm.riscv.vmsge.mask.nxv8i16(
 define <vscale x 8 x i1> @intrinsic_vmsge_mask_vv_nxv8i16_nxv8i16(<vscale x 8 x i1> %0, <vscale x 8 x i16> %1, <vscale x 8 x i16> %2, <vscale x 8 x i16> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsge_mask_vv_nxv8i16_nxv8i16:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v14, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e16, m2, ta, mu
-; CHECK-NEXT:    vmsle.vv v14, v10, v8
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmsle.vv v0, v10, v8
+; CHECK-NEXT:    vmsle.vv v14, v12, v10, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v14
-; CHECK-NEXT:    vmsle.vv v8, v12, v10, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 8 x i1> @llvm.riscv.vmsge.nxv8i16(
@@ -554,12 +544,11 @@ declare <vscale x 16 x i1> @llvm.riscv.vmsge.mask.nxv16i16(
 define <vscale x 16 x i1> @intrinsic_vmsge_mask_vv_nxv16i16_nxv16i16(<vscale x 16 x i1> %0, <vscale x 16 x i16> %1, <vscale x 16 x i16> %2, <vscale x 16 x i16> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsge_mask_vv_nxv16i16_nxv16i16:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v20, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e16, m4, ta, mu
-; CHECK-NEXT:    vmsle.vv v20, v12, v8
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmsle.vv v0, v12, v8
+; CHECK-NEXT:    vmsle.vv v20, v16, v12, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v20
-; CHECK-NEXT:    vmsle.vv v8, v16, v12, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 16 x i1> @llvm.riscv.vmsge.nxv16i16(
@@ -606,10 +595,9 @@ declare <vscale x 1 x i1> @llvm.riscv.vmsge.mask.nxv1i32(
 define <vscale x 1 x i1> @intrinsic_vmsge_mask_vv_nxv1i32_nxv1i32(<vscale x 1 x i1> %0, <vscale x 1 x i32> %1, <vscale x 1 x i32> %2, <vscale x 1 x i32> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsge_mask_vv_nxv1i32_nxv1i32:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e32, mf2, ta, mu
-; CHECK-NEXT:    vmsle.vv v8, v9, v8
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv1r.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e32, mf2, ta, mu
+; CHECK-NEXT:    vmsle.vv v0, v9, v8
 ; CHECK-NEXT:    vmsle.vv v11, v10, v9, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v11
 ; CHECK-NEXT:    ret
@@ -658,10 +646,9 @@ declare <vscale x 2 x i1> @llvm.riscv.vmsge.mask.nxv2i32(
 define <vscale x 2 x i1> @intrinsic_vmsge_mask_vv_nxv2i32_nxv2i32(<vscale x 2 x i1> %0, <vscale x 2 x i32> %1, <vscale x 2 x i32> %2, <vscale x 2 x i32> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsge_mask_vv_nxv2i32_nxv2i32:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e32, m1, ta, mu
-; CHECK-NEXT:    vmsle.vv v8, v9, v8
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv.v.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e32, m1, ta, mu
+; CHECK-NEXT:    vmsle.vv v0, v9, v8
 ; CHECK-NEXT:    vmsle.vv v11, v10, v9, v0.t
 ; CHECK-NEXT:    vmv.v.v v0, v11
 ; CHECK-NEXT:    ret
@@ -710,12 +697,11 @@ declare <vscale x 4 x i1> @llvm.riscv.vmsge.mask.nxv4i32(
 define <vscale x 4 x i1> @intrinsic_vmsge_mask_vv_nxv4i32_nxv4i32(<vscale x 4 x i1> %0, <vscale x 4 x i32> %1, <vscale x 4 x i32> %2, <vscale x 4 x i32> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsge_mask_vv_nxv4i32_nxv4i32:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v14, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e32, m2, ta, mu
-; CHECK-NEXT:    vmsle.vv v14, v10, v8
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmsle.vv v0, v10, v8
+; CHECK-NEXT:    vmsle.vv v14, v12, v10, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v14
-; CHECK-NEXT:    vmsle.vv v8, v12, v10, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 4 x i1> @llvm.riscv.vmsge.nxv4i32(
@@ -762,12 +748,11 @@ declare <vscale x 8 x i1> @llvm.riscv.vmsge.mask.nxv8i32(
 define <vscale x 8 x i1> @intrinsic_vmsge_mask_vv_nxv8i32_nxv8i32(<vscale x 8 x i1> %0, <vscale x 8 x i32> %1, <vscale x 8 x i32> %2, <vscale x 8 x i32> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsge_mask_vv_nxv8i32_nxv8i32:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v20, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e32, m4, ta, mu
-; CHECK-NEXT:    vmsle.vv v20, v12, v8
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmsle.vv v0, v12, v8
+; CHECK-NEXT:    vmsle.vv v20, v16, v12, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v20
-; CHECK-NEXT:    vmsle.vv v8, v16, v12, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 8 x i1> @llvm.riscv.vmsge.nxv8i32(
@@ -814,10 +799,9 @@ declare <vscale x 1 x i1> @llvm.riscv.vmsge.mask.nxv1i64(
 define <vscale x 1 x i1> @intrinsic_vmsge_mask_vv_nxv1i64_nxv1i64(<vscale x 1 x i1> %0, <vscale x 1 x i64> %1, <vscale x 1 x i64> %2, <vscale x 1 x i64> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsge_mask_vv_nxv1i64_nxv1i64:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e64, m1, ta, mu
-; CHECK-NEXT:    vmsle.vv v8, v9, v8
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv.v.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e64, m1, ta, mu
+; CHECK-NEXT:    vmsle.vv v0, v9, v8
 ; CHECK-NEXT:    vmsle.vv v11, v10, v9, v0.t
 ; CHECK-NEXT:    vmv.v.v v0, v11
 ; CHECK-NEXT:    ret
@@ -866,12 +850,11 @@ declare <vscale x 2 x i1> @llvm.riscv.vmsge.mask.nxv2i64(
 define <vscale x 2 x i1> @intrinsic_vmsge_mask_vv_nxv2i64_nxv2i64(<vscale x 2 x i1> %0, <vscale x 2 x i64> %1, <vscale x 2 x i64> %2, <vscale x 2 x i64> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsge_mask_vv_nxv2i64_nxv2i64:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v14, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e64, m2, ta, mu
-; CHECK-NEXT:    vmsle.vv v14, v10, v8
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmsle.vv v0, v10, v8
+; CHECK-NEXT:    vmsle.vv v14, v12, v10, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v14
-; CHECK-NEXT:    vmsle.vv v8, v12, v10, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 2 x i1> @llvm.riscv.vmsge.nxv2i64(
@@ -918,12 +901,11 @@ declare <vscale x 4 x i1> @llvm.riscv.vmsge.mask.nxv4i64(
 define <vscale x 4 x i1> @intrinsic_vmsge_mask_vv_nxv4i64_nxv4i64(<vscale x 4 x i1> %0, <vscale x 4 x i64> %1, <vscale x 4 x i64> %2, <vscale x 4 x i64> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsge_mask_vv_nxv4i64_nxv4i64:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v20, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e64, m4, ta, mu
-; CHECK-NEXT:    vmsle.vv v20, v12, v8
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmsle.vv v0, v12, v8
+; CHECK-NEXT:    vmsle.vv v20, v16, v12, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v20
-; CHECK-NEXT:    vmsle.vv v8, v16, v12, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 4 x i1> @llvm.riscv.vmsge.nxv4i64(
diff --git a/llvm/test/CodeGen/RISCV/rvv/vmsgeu.ll b/llvm/test/CodeGen/RISCV/rvv/vmsgeu.ll
index 9410a99d8142..b6c6d9e90f61 100644
--- a/llvm/test/CodeGen/RISCV/rvv/vmsgeu.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/vmsgeu.ll
@@ -34,10 +34,9 @@ declare <vscale x 1 x i1> @llvm.riscv.vmsgeu.mask.nxv1i8(
 define <vscale x 1 x i1> @intrinsic_vmsgeu_mask_vv_nxv1i8_nxv1i8(<vscale x 1 x i1> %0, <vscale x 1 x i8> %1, <vscale x 1 x i8> %2, <vscale x 1 x i8> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsgeu_mask_vv_nxv1i8_nxv1i8:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e8, mf8, ta, mu
-; CHECK-NEXT:    vmsleu.vv v8, v9, v8
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv1r.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e8, mf8, ta, mu
+; CHECK-NEXT:    vmsleu.vv v0, v9, v8
 ; CHECK-NEXT:    vmsleu.vv v11, v10, v9, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v11
 ; CHECK-NEXT:    ret
@@ -86,10 +85,9 @@ declare <vscale x 2 x i1> @llvm.riscv.vmsgeu.mask.nxv2i8(
 define <vscale x 2 x i1> @intrinsic_vmsgeu_mask_vv_nxv2i8_nxv2i8(<vscale x 2 x i1> %0, <vscale x 2 x i8> %1, <vscale x 2 x i8> %2, <vscale x 2 x i8> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsgeu_mask_vv_nxv2i8_nxv2i8:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e8, mf4, ta, mu
-; CHECK-NEXT:    vmsleu.vv v8, v9, v8
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv1r.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e8, mf4, ta, mu
+; CHECK-NEXT:    vmsleu.vv v0, v9, v8
 ; CHECK-NEXT:    vmsleu.vv v11, v10, v9, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v11
 ; CHECK-NEXT:    ret
@@ -138,10 +136,9 @@ declare <vscale x 4 x i1> @llvm.riscv.vmsgeu.mask.nxv4i8(
 define <vscale x 4 x i1> @intrinsic_vmsgeu_mask_vv_nxv4i8_nxv4i8(<vscale x 4 x i1> %0, <vscale x 4 x i8> %1, <vscale x 4 x i8> %2, <vscale x 4 x i8> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsgeu_mask_vv_nxv4i8_nxv4i8:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e8, mf2, ta, mu
-; CHECK-NEXT:    vmsleu.vv v8, v9, v8
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv1r.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e8, mf2, ta, mu
+; CHECK-NEXT:    vmsleu.vv v0, v9, v8
 ; CHECK-NEXT:    vmsleu.vv v11, v10, v9, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v11
 ; CHECK-NEXT:    ret
@@ -190,10 +187,9 @@ declare <vscale x 8 x i1> @llvm.riscv.vmsgeu.mask.nxv8i8(
 define <vscale x 8 x i1> @intrinsic_vmsgeu_mask_vv_nxv8i8_nxv8i8(<vscale x 8 x i1> %0, <vscale x 8 x i8> %1, <vscale x 8 x i8> %2, <vscale x 8 x i8> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsgeu_mask_vv_nxv8i8_nxv8i8:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e8, m1, ta, mu
-; CHECK-NEXT:    vmsleu.vv v8, v9, v8
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv.v.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e8, m1, ta, mu
+; CHECK-NEXT:    vmsleu.vv v0, v9, v8
 ; CHECK-NEXT:    vmsleu.vv v11, v10, v9, v0.t
 ; CHECK-NEXT:    vmv.v.v v0, v11
 ; CHECK-NEXT:    ret
@@ -242,12 +238,11 @@ declare <vscale x 16 x i1> @llvm.riscv.vmsgeu.mask.nxv16i8(
 define <vscale x 16 x i1> @intrinsic_vmsgeu_mask_vv_nxv16i8_nxv16i8(<vscale x 16 x i1> %0, <vscale x 16 x i8> %1, <vscale x 16 x i8> %2, <vscale x 16 x i8> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsgeu_mask_vv_nxv16i8_nxv16i8:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v14, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e8, m2, ta, mu
-; CHECK-NEXT:    vmsleu.vv v14, v10, v8
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmsleu.vv v0, v10, v8
+; CHECK-NEXT:    vmsleu.vv v14, v12, v10, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v14
-; CHECK-NEXT:    vmsleu.vv v8, v12, v10, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 16 x i1> @llvm.riscv.vmsgeu.nxv16i8(
@@ -294,12 +289,11 @@ declare <vscale x 32 x i1> @llvm.riscv.vmsgeu.mask.nxv32i8(
 define <vscale x 32 x i1> @intrinsic_vmsgeu_mask_vv_nxv32i8_nxv32i8(<vscale x 32 x i1> %0, <vscale x 32 x i8> %1, <vscale x 32 x i8> %2, <vscale x 32 x i8> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsgeu_mask_vv_nxv32i8_nxv32i8:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v20, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e8, m4, ta, mu
-; CHECK-NEXT:    vmsleu.vv v20, v12, v8
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmsleu.vv v0, v12, v8
+; CHECK-NEXT:    vmsleu.vv v20, v16, v12, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v20
-; CHECK-NEXT:    vmsleu.vv v8, v16, v12, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 32 x i1> @llvm.riscv.vmsgeu.nxv32i8(
@@ -346,10 +340,9 @@ declare <vscale x 1 x i1> @llvm.riscv.vmsgeu.mask.nxv1i16(
 define <vscale x 1 x i1> @intrinsic_vmsgeu_mask_vv_nxv1i16_nxv1i16(<vscale x 1 x i1> %0, <vscale x 1 x i16> %1, <vscale x 1 x i16> %2, <vscale x 1 x i16> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsgeu_mask_vv_nxv1i16_nxv1i16:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
-; CHECK-NEXT:    vmsleu.vv v8, v9, v8
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv1r.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
+; CHECK-NEXT:    vmsleu.vv v0, v9, v8
 ; CHECK-NEXT:    vmsleu.vv v11, v10, v9, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v11
 ; CHECK-NEXT:    ret
@@ -398,10 +391,9 @@ declare <vscale x 2 x i1> @llvm.riscv.vmsgeu.mask.nxv2i16(
 define <vscale x 2 x i1> @intrinsic_vmsgeu_mask_vv_nxv2i16_nxv2i16(<vscale x 2 x i1> %0, <vscale x 2 x i16> %1, <vscale x 2 x i16> %2, <vscale x 2 x i16> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsgeu_mask_vv_nxv2i16_nxv2i16:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e16, mf2, ta, mu
-; CHECK-NEXT:    vmsleu.vv v8, v9, v8
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv1r.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e16, mf2, ta, mu
+; CHECK-NEXT:    vmsleu.vv v0, v9, v8
 ; CHECK-NEXT:    vmsleu.vv v11, v10, v9, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v11
 ; CHECK-NEXT:    ret
@@ -450,10 +442,9 @@ declare <vscale x 4 x i1> @llvm.riscv.vmsgeu.mask.nxv4i16(
 define <vscale x 4 x i1> @intrinsic_vmsgeu_mask_vv_nxv4i16_nxv4i16(<vscale x 4 x i1> %0, <vscale x 4 x i16> %1, <vscale x 4 x i16> %2, <vscale x 4 x i16> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsgeu_mask_vv_nxv4i16_nxv4i16:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e16, m1, ta, mu
-; CHECK-NEXT:    vmsleu.vv v8, v9, v8
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv.v.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e16, m1, ta, mu
+; CHECK-NEXT:    vmsleu.vv v0, v9, v8
 ; CHECK-NEXT:    vmsleu.vv v11, v10, v9, v0.t
 ; CHECK-NEXT:    vmv.v.v v0, v11
 ; CHECK-NEXT:    ret
@@ -502,12 +493,11 @@ declare <vscale x 8 x i1> @llvm.riscv.vmsgeu.mask.nxv8i16(
 define <vscale x 8 x i1> @intrinsic_vmsgeu_mask_vv_nxv8i16_nxv8i16(<vscale x 8 x i1> %0, <vscale x 8 x i16> %1, <vscale x 8 x i16> %2, <vscale x 8 x i16> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsgeu_mask_vv_nxv8i16_nxv8i16:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v14, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e16, m2, ta, mu
-; CHECK-NEXT:    vmsleu.vv v14, v10, v8
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmsleu.vv v0, v10, v8
+; CHECK-NEXT:    vmsleu.vv v14, v12, v10, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v14
-; CHECK-NEXT:    vmsleu.vv v8, v12, v10, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 8 x i1> @llvm.riscv.vmsgeu.nxv8i16(
@@ -554,12 +544,11 @@ declare <vscale x 16 x i1> @llvm.riscv.vmsgeu.mask.nxv16i16(
 define <vscale x 16 x i1> @intrinsic_vmsgeu_mask_vv_nxv16i16_nxv16i16(<vscale x 16 x i1> %0, <vscale x 16 x i16> %1, <vscale x 16 x i16> %2, <vscale x 16 x i16> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsgeu_mask_vv_nxv16i16_nxv16i16:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v20, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e16, m4, ta, mu
-; CHECK-NEXT:    vmsleu.vv v20, v12, v8
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmsleu.vv v0, v12, v8
+; CHECK-NEXT:    vmsleu.vv v20, v16, v12, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v20
-; CHECK-NEXT:    vmsleu.vv v8, v16, v12, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 16 x i1> @llvm.riscv.vmsgeu.nxv16i16(
@@ -606,10 +595,9 @@ declare <vscale x 1 x i1> @llvm.riscv.vmsgeu.mask.nxv1i32(
 define <vscale x 1 x i1> @intrinsic_vmsgeu_mask_vv_nxv1i32_nxv1i32(<vscale x 1 x i1> %0, <vscale x 1 x i32> %1, <vscale x 1 x i32> %2, <vscale x 1 x i32> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsgeu_mask_vv_nxv1i32_nxv1i32:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e32, mf2, ta, mu
-; CHECK-NEXT:    vmsleu.vv v8, v9, v8
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv1r.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e32, mf2, ta, mu
+; CHECK-NEXT:    vmsleu.vv v0, v9, v8
 ; CHECK-NEXT:    vmsleu.vv v11, v10, v9, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v11
 ; CHECK-NEXT:    ret
@@ -658,10 +646,9 @@ declare <vscale x 2 x i1> @llvm.riscv.vmsgeu.mask.nxv2i32(
 define <vscale x 2 x i1> @intrinsic_vmsgeu_mask_vv_nxv2i32_nxv2i32(<vscale x 2 x i1> %0, <vscale x 2 x i32> %1, <vscale x 2 x i32> %2, <vscale x 2 x i32> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsgeu_mask_vv_nxv2i32_nxv2i32:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e32, m1, ta, mu
-; CHECK-NEXT:    vmsleu.vv v8, v9, v8
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv.v.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e32, m1, ta, mu
+; CHECK-NEXT:    vmsleu.vv v0, v9, v8
 ; CHECK-NEXT:    vmsleu.vv v11, v10, v9, v0.t
 ; CHECK-NEXT:    vmv.v.v v0, v11
 ; CHECK-NEXT:    ret
@@ -710,12 +697,11 @@ declare <vscale x 4 x i1> @llvm.riscv.vmsgeu.mask.nxv4i32(
 define <vscale x 4 x i1> @intrinsic_vmsgeu_mask_vv_nxv4i32_nxv4i32(<vscale x 4 x i1> %0, <vscale x 4 x i32> %1, <vscale x 4 x i32> %2, <vscale x 4 x i32> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsgeu_mask_vv_nxv4i32_nxv4i32:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v14, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e32, m2, ta, mu
-; CHECK-NEXT:    vmsleu.vv v14, v10, v8
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmsleu.vv v0, v10, v8
+; CHECK-NEXT:    vmsleu.vv v14, v12, v10, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v14
-; CHECK-NEXT:    vmsleu.vv v8, v12, v10, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 4 x i1> @llvm.riscv.vmsgeu.nxv4i32(
@@ -762,12 +748,11 @@ declare <vscale x 8 x i1> @llvm.riscv.vmsgeu.mask.nxv8i32(
 define <vscale x 8 x i1> @intrinsic_vmsgeu_mask_vv_nxv8i32_nxv8i32(<vscale x 8 x i1> %0, <vscale x 8 x i32> %1, <vscale x 8 x i32> %2, <vscale x 8 x i32> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsgeu_mask_vv_nxv8i32_nxv8i32:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v20, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e32, m4, ta, mu
-; CHECK-NEXT:    vmsleu.vv v20, v12, v8
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmsleu.vv v0, v12, v8
+; CHECK-NEXT:    vmsleu.vv v20, v16, v12, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v20
-; CHECK-NEXT:    vmsleu.vv v8, v16, v12, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 8 x i1> @llvm.riscv.vmsgeu.nxv8i32(
@@ -814,10 +799,9 @@ declare <vscale x 1 x i1> @llvm.riscv.vmsgeu.mask.nxv1i64(
 define <vscale x 1 x i1> @intrinsic_vmsgeu_mask_vv_nxv1i64_nxv1i64(<vscale x 1 x i1> %0, <vscale x 1 x i64> %1, <vscale x 1 x i64> %2, <vscale x 1 x i64> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsgeu_mask_vv_nxv1i64_nxv1i64:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e64, m1, ta, mu
-; CHECK-NEXT:    vmsleu.vv v8, v9, v8
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv.v.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e64, m1, ta, mu
+; CHECK-NEXT:    vmsleu.vv v0, v9, v8
 ; CHECK-NEXT:    vmsleu.vv v11, v10, v9, v0.t
 ; CHECK-NEXT:    vmv.v.v v0, v11
 ; CHECK-NEXT:    ret
@@ -866,12 +850,11 @@ declare <vscale x 2 x i1> @llvm.riscv.vmsgeu.mask.nxv2i64(
 define <vscale x 2 x i1> @intrinsic_vmsgeu_mask_vv_nxv2i64_nxv2i64(<vscale x 2 x i1> %0, <vscale x 2 x i64> %1, <vscale x 2 x i64> %2, <vscale x 2 x i64> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsgeu_mask_vv_nxv2i64_nxv2i64:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v14, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e64, m2, ta, mu
-; CHECK-NEXT:    vmsleu.vv v14, v10, v8
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmsleu.vv v0, v10, v8
+; CHECK-NEXT:    vmsleu.vv v14, v12, v10, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v14
-; CHECK-NEXT:    vmsleu.vv v8, v12, v10, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 2 x i1> @llvm.riscv.vmsgeu.nxv2i64(
@@ -918,12 +901,11 @@ declare <vscale x 4 x i1> @llvm.riscv.vmsgeu.mask.nxv4i64(
 define <vscale x 4 x i1> @intrinsic_vmsgeu_mask_vv_nxv4i64_nxv4i64(<vscale x 4 x i1> %0, <vscale x 4 x i64> %1, <vscale x 4 x i64> %2, <vscale x 4 x i64> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsgeu_mask_vv_nxv4i64_nxv4i64:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v20, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e64, m4, ta, mu
-; CHECK-NEXT:    vmsleu.vv v20, v12, v8
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmsleu.vv v0, v12, v8
+; CHECK-NEXT:    vmsleu.vv v20, v16, v12, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v20
-; CHECK-NEXT:    vmsleu.vv v8, v16, v12, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 4 x i1> @llvm.riscv.vmsgeu.nxv4i64(
diff --git a/llvm/test/CodeGen/RISCV/rvv/vmsgt.ll b/llvm/test/CodeGen/RISCV/rvv/vmsgt.ll
index b7a676e7f2dd..dfd7096a65eb 100644
--- a/llvm/test/CodeGen/RISCV/rvv/vmsgt.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/vmsgt.ll
@@ -34,10 +34,9 @@ declare <vscale x 1 x i1> @llvm.riscv.vmsgt.mask.nxv1i8(
 define <vscale x 1 x i1> @intrinsic_vmsgt_mask_vv_nxv1i8_nxv1i8(<vscale x 1 x i1> %0, <vscale x 1 x i8> %1, <vscale x 1 x i8> %2, <vscale x 1 x i8> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsgt_mask_vv_nxv1i8_nxv1i8:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e8, mf8, ta, mu
-; CHECK-NEXT:    vmslt.vv v8, v9, v8
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv1r.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e8, mf8, ta, mu
+; CHECK-NEXT:    vmslt.vv v0, v9, v8
 ; CHECK-NEXT:    vmslt.vv v11, v10, v9, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v11
 ; CHECK-NEXT:    ret
@@ -86,10 +85,9 @@ declare <vscale x 2 x i1> @llvm.riscv.vmsgt.mask.nxv2i8(
 define <vscale x 2 x i1> @intrinsic_vmsgt_mask_vv_nxv2i8_nxv2i8(<vscale x 2 x i1> %0, <vscale x 2 x i8> %1, <vscale x 2 x i8> %2, <vscale x 2 x i8> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsgt_mask_vv_nxv2i8_nxv2i8:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e8, mf4, ta, mu
-; CHECK-NEXT:    vmslt.vv v8, v9, v8
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv1r.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e8, mf4, ta, mu
+; CHECK-NEXT:    vmslt.vv v0, v9, v8
 ; CHECK-NEXT:    vmslt.vv v11, v10, v9, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v11
 ; CHECK-NEXT:    ret
@@ -138,10 +136,9 @@ declare <vscale x 4 x i1> @llvm.riscv.vmsgt.mask.nxv4i8(
 define <vscale x 4 x i1> @intrinsic_vmsgt_mask_vv_nxv4i8_nxv4i8(<vscale x 4 x i1> %0, <vscale x 4 x i8> %1, <vscale x 4 x i8> %2, <vscale x 4 x i8> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsgt_mask_vv_nxv4i8_nxv4i8:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e8, mf2, ta, mu
-; CHECK-NEXT:    vmslt.vv v8, v9, v8
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv1r.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e8, mf2, ta, mu
+; CHECK-NEXT:    vmslt.vv v0, v9, v8
 ; CHECK-NEXT:    vmslt.vv v11, v10, v9, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v11
 ; CHECK-NEXT:    ret
@@ -190,10 +187,9 @@ declare <vscale x 8 x i1> @llvm.riscv.vmsgt.mask.nxv8i8(
 define <vscale x 8 x i1> @intrinsic_vmsgt_mask_vv_nxv8i8_nxv8i8(<vscale x 8 x i1> %0, <vscale x 8 x i8> %1, <vscale x 8 x i8> %2, <vscale x 8 x i8> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsgt_mask_vv_nxv8i8_nxv8i8:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e8, m1, ta, mu
-; CHECK-NEXT:    vmslt.vv v8, v9, v8
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv.v.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e8, m1, ta, mu
+; CHECK-NEXT:    vmslt.vv v0, v9, v8
 ; CHECK-NEXT:    vmslt.vv v11, v10, v9, v0.t
 ; CHECK-NEXT:    vmv.v.v v0, v11
 ; CHECK-NEXT:    ret
@@ -242,12 +238,11 @@ declare <vscale x 16 x i1> @llvm.riscv.vmsgt.mask.nxv16i8(
 define <vscale x 16 x i1> @intrinsic_vmsgt_mask_vv_nxv16i8_nxv16i8(<vscale x 16 x i1> %0, <vscale x 16 x i8> %1, <vscale x 16 x i8> %2, <vscale x 16 x i8> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsgt_mask_vv_nxv16i8_nxv16i8:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v14, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e8, m2, ta, mu
-; CHECK-NEXT:    vmslt.vv v14, v10, v8
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmslt.vv v0, v10, v8
+; CHECK-NEXT:    vmslt.vv v14, v12, v10, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v14
-; CHECK-NEXT:    vmslt.vv v8, v12, v10, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 16 x i1> @llvm.riscv.vmsgt.nxv16i8(
@@ -294,12 +289,11 @@ declare <vscale x 32 x i1> @llvm.riscv.vmsgt.mask.nxv32i8(
 define <vscale x 32 x i1> @intrinsic_vmsgt_mask_vv_nxv32i8_nxv32i8(<vscale x 32 x i1> %0, <vscale x 32 x i8> %1, <vscale x 32 x i8> %2, <vscale x 32 x i8> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsgt_mask_vv_nxv32i8_nxv32i8:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v20, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e8, m4, ta, mu
-; CHECK-NEXT:    vmslt.vv v20, v12, v8
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmslt.vv v0, v12, v8
+; CHECK-NEXT:    vmslt.vv v20, v16, v12, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v20
-; CHECK-NEXT:    vmslt.vv v8, v16, v12, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 32 x i1> @llvm.riscv.vmsgt.nxv32i8(
@@ -346,10 +340,9 @@ declare <vscale x 1 x i1> @llvm.riscv.vmsgt.mask.nxv1i16(
 define <vscale x 1 x i1> @intrinsic_vmsgt_mask_vv_nxv1i16_nxv1i16(<vscale x 1 x i1> %0, <vscale x 1 x i16> %1, <vscale x 1 x i16> %2, <vscale x 1 x i16> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsgt_mask_vv_nxv1i16_nxv1i16:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
-; CHECK-NEXT:    vmslt.vv v8, v9, v8
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv1r.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
+; CHECK-NEXT:    vmslt.vv v0, v9, v8
 ; CHECK-NEXT:    vmslt.vv v11, v10, v9, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v11
 ; CHECK-NEXT:    ret
@@ -398,10 +391,9 @@ declare <vscale x 2 x i1> @llvm.riscv.vmsgt.mask.nxv2i16(
 define <vscale x 2 x i1> @intrinsic_vmsgt_mask_vv_nxv2i16_nxv2i16(<vscale x 2 x i1> %0, <vscale x 2 x i16> %1, <vscale x 2 x i16> %2, <vscale x 2 x i16> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsgt_mask_vv_nxv2i16_nxv2i16:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e16, mf2, ta, mu
-; CHECK-NEXT:    vmslt.vv v8, v9, v8
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv1r.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e16, mf2, ta, mu
+; CHECK-NEXT:    vmslt.vv v0, v9, v8
 ; CHECK-NEXT:    vmslt.vv v11, v10, v9, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v11
 ; CHECK-NEXT:    ret
@@ -450,10 +442,9 @@ declare <vscale x 4 x i1> @llvm.riscv.vmsgt.mask.nxv4i16(
 define <vscale x 4 x i1> @intrinsic_vmsgt_mask_vv_nxv4i16_nxv4i16(<vscale x 4 x i1> %0, <vscale x 4 x i16> %1, <vscale x 4 x i16> %2, <vscale x 4 x i16> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsgt_mask_vv_nxv4i16_nxv4i16:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e16, m1, ta, mu
-; CHECK-NEXT:    vmslt.vv v8, v9, v8
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv.v.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e16, m1, ta, mu
+; CHECK-NEXT:    vmslt.vv v0, v9, v8
 ; CHECK-NEXT:    vmslt.vv v11, v10, v9, v0.t
 ; CHECK-NEXT:    vmv.v.v v0, v11
 ; CHECK-NEXT:    ret
@@ -502,12 +493,11 @@ declare <vscale x 8 x i1> @llvm.riscv.vmsgt.mask.nxv8i16(
 define <vscale x 8 x i1> @intrinsic_vmsgt_mask_vv_nxv8i16_nxv8i16(<vscale x 8 x i1> %0, <vscale x 8 x i16> %1, <vscale x 8 x i16> %2, <vscale x 8 x i16> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsgt_mask_vv_nxv8i16_nxv8i16:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v14, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e16, m2, ta, mu
-; CHECK-NEXT:    vmslt.vv v14, v10, v8
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmslt.vv v0, v10, v8
+; CHECK-NEXT:    vmslt.vv v14, v12, v10, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v14
-; CHECK-NEXT:    vmslt.vv v8, v12, v10, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 8 x i1> @llvm.riscv.vmsgt.nxv8i16(
@@ -554,12 +544,11 @@ declare <vscale x 16 x i1> @llvm.riscv.vmsgt.mask.nxv16i16(
 define <vscale x 16 x i1> @intrinsic_vmsgt_mask_vv_nxv16i16_nxv16i16(<vscale x 16 x i1> %0, <vscale x 16 x i16> %1, <vscale x 16 x i16> %2, <vscale x 16 x i16> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsgt_mask_vv_nxv16i16_nxv16i16:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v20, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e16, m4, ta, mu
-; CHECK-NEXT:    vmslt.vv v20, v12, v8
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmslt.vv v0, v12, v8
+; CHECK-NEXT:    vmslt.vv v20, v16, v12, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v20
-; CHECK-NEXT:    vmslt.vv v8, v16, v12, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 16 x i1> @llvm.riscv.vmsgt.nxv16i16(
@@ -606,10 +595,9 @@ declare <vscale x 1 x i1> @llvm.riscv.vmsgt.mask.nxv1i32(
 define <vscale x 1 x i1> @intrinsic_vmsgt_mask_vv_nxv1i32_nxv1i32(<vscale x 1 x i1> %0, <vscale x 1 x i32> %1, <vscale x 1 x i32> %2, <vscale x 1 x i32> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsgt_mask_vv_nxv1i32_nxv1i32:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e32, mf2, ta, mu
-; CHECK-NEXT:    vmslt.vv v8, v9, v8
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv1r.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e32, mf2, ta, mu
+; CHECK-NEXT:    vmslt.vv v0, v9, v8
 ; CHECK-NEXT:    vmslt.vv v11, v10, v9, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v11
 ; CHECK-NEXT:    ret
@@ -658,10 +646,9 @@ declare <vscale x 2 x i1> @llvm.riscv.vmsgt.mask.nxv2i32(
 define <vscale x 2 x i1> @intrinsic_vmsgt_mask_vv_nxv2i32_nxv2i32(<vscale x 2 x i1> %0, <vscale x 2 x i32> %1, <vscale x 2 x i32> %2, <vscale x 2 x i32> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsgt_mask_vv_nxv2i32_nxv2i32:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e32, m1, ta, mu
-; CHECK-NEXT:    vmslt.vv v8, v9, v8
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv.v.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e32, m1, ta, mu
+; CHECK-NEXT:    vmslt.vv v0, v9, v8
 ; CHECK-NEXT:    vmslt.vv v11, v10, v9, v0.t
 ; CHECK-NEXT:    vmv.v.v v0, v11
 ; CHECK-NEXT:    ret
@@ -710,12 +697,11 @@ declare <vscale x 4 x i1> @llvm.riscv.vmsgt.mask.nxv4i32(
 define <vscale x 4 x i1> @intrinsic_vmsgt_mask_vv_nxv4i32_nxv4i32(<vscale x 4 x i1> %0, <vscale x 4 x i32> %1, <vscale x 4 x i32> %2, <vscale x 4 x i32> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsgt_mask_vv_nxv4i32_nxv4i32:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v14, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e32, m2, ta, mu
-; CHECK-NEXT:    vmslt.vv v14, v10, v8
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmslt.vv v0, v10, v8
+; CHECK-NEXT:    vmslt.vv v14, v12, v10, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v14
-; CHECK-NEXT:    vmslt.vv v8, v12, v10, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 4 x i1> @llvm.riscv.vmsgt.nxv4i32(
@@ -762,12 +748,11 @@ declare <vscale x 8 x i1> @llvm.riscv.vmsgt.mask.nxv8i32(
 define <vscale x 8 x i1> @intrinsic_vmsgt_mask_vv_nxv8i32_nxv8i32(<vscale x 8 x i1> %0, <vscale x 8 x i32> %1, <vscale x 8 x i32> %2, <vscale x 8 x i32> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsgt_mask_vv_nxv8i32_nxv8i32:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v20, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e32, m4, ta, mu
-; CHECK-NEXT:    vmslt.vv v20, v12, v8
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmslt.vv v0, v12, v8
+; CHECK-NEXT:    vmslt.vv v20, v16, v12, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v20
-; CHECK-NEXT:    vmslt.vv v8, v16, v12, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 8 x i1> @llvm.riscv.vmsgt.nxv8i32(
@@ -814,10 +799,9 @@ declare <vscale x 1 x i1> @llvm.riscv.vmsgt.mask.nxv1i64(
 define <vscale x 1 x i1> @intrinsic_vmsgt_mask_vv_nxv1i64_nxv1i64(<vscale x 1 x i1> %0, <vscale x 1 x i64> %1, <vscale x 1 x i64> %2, <vscale x 1 x i64> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsgt_mask_vv_nxv1i64_nxv1i64:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e64, m1, ta, mu
-; CHECK-NEXT:    vmslt.vv v8, v9, v8
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv.v.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e64, m1, ta, mu
+; CHECK-NEXT:    vmslt.vv v0, v9, v8
 ; CHECK-NEXT:    vmslt.vv v11, v10, v9, v0.t
 ; CHECK-NEXT:    vmv.v.v v0, v11
 ; CHECK-NEXT:    ret
@@ -866,12 +850,11 @@ declare <vscale x 2 x i1> @llvm.riscv.vmsgt.mask.nxv2i64(
 define <vscale x 2 x i1> @intrinsic_vmsgt_mask_vv_nxv2i64_nxv2i64(<vscale x 2 x i1> %0, <vscale x 2 x i64> %1, <vscale x 2 x i64> %2, <vscale x 2 x i64> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsgt_mask_vv_nxv2i64_nxv2i64:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v14, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e64, m2, ta, mu
-; CHECK-NEXT:    vmslt.vv v14, v10, v8
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmslt.vv v0, v10, v8
+; CHECK-NEXT:    vmslt.vv v14, v12, v10, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v14
-; CHECK-NEXT:    vmslt.vv v8, v12, v10, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 2 x i1> @llvm.riscv.vmsgt.nxv2i64(
@@ -918,12 +901,11 @@ declare <vscale x 4 x i1> @llvm.riscv.vmsgt.mask.nxv4i64(
 define <vscale x 4 x i1> @intrinsic_vmsgt_mask_vv_nxv4i64_nxv4i64(<vscale x 4 x i1> %0, <vscale x 4 x i64> %1, <vscale x 4 x i64> %2, <vscale x 4 x i64> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsgt_mask_vv_nxv4i64_nxv4i64:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v20, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e64, m4, ta, mu
-; CHECK-NEXT:    vmslt.vv v20, v12, v8
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmslt.vv v0, v12, v8
+; CHECK-NEXT:    vmslt.vv v20, v16, v12, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v20
-; CHECK-NEXT:    vmslt.vv v8, v16, v12, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 4 x i1> @llvm.riscv.vmsgt.nxv4i64(
diff --git a/llvm/test/CodeGen/RISCV/rvv/vmsgtu.ll b/llvm/test/CodeGen/RISCV/rvv/vmsgtu.ll
index 88a632de067a..8826be03bbeb 100644
--- a/llvm/test/CodeGen/RISCV/rvv/vmsgtu.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/vmsgtu.ll
@@ -34,10 +34,9 @@ declare <vscale x 1 x i1> @llvm.riscv.vmsgtu.mask.nxv1i8(
 define <vscale x 1 x i1> @intrinsic_vmsgtu_mask_vv_nxv1i8_nxv1i8(<vscale x 1 x i1> %0, <vscale x 1 x i8> %1, <vscale x 1 x i8> %2, <vscale x 1 x i8> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsgtu_mask_vv_nxv1i8_nxv1i8:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e8, mf8, ta, mu
-; CHECK-NEXT:    vmsltu.vv v8, v9, v8
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv1r.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e8, mf8, ta, mu
+; CHECK-NEXT:    vmsltu.vv v0, v9, v8
 ; CHECK-NEXT:    vmsltu.vv v11, v10, v9, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v11
 ; CHECK-NEXT:    ret
@@ -86,10 +85,9 @@ declare <vscale x 2 x i1> @llvm.riscv.vmsgtu.mask.nxv2i8(
 define <vscale x 2 x i1> @intrinsic_vmsgtu_mask_vv_nxv2i8_nxv2i8(<vscale x 2 x i1> %0, <vscale x 2 x i8> %1, <vscale x 2 x i8> %2, <vscale x 2 x i8> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsgtu_mask_vv_nxv2i8_nxv2i8:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e8, mf4, ta, mu
-; CHECK-NEXT:    vmsltu.vv v8, v9, v8
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv1r.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e8, mf4, ta, mu
+; CHECK-NEXT:    vmsltu.vv v0, v9, v8
 ; CHECK-NEXT:    vmsltu.vv v11, v10, v9, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v11
 ; CHECK-NEXT:    ret
@@ -138,10 +136,9 @@ declare <vscale x 4 x i1> @llvm.riscv.vmsgtu.mask.nxv4i8(
 define <vscale x 4 x i1> @intrinsic_vmsgtu_mask_vv_nxv4i8_nxv4i8(<vscale x 4 x i1> %0, <vscale x 4 x i8> %1, <vscale x 4 x i8> %2, <vscale x 4 x i8> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsgtu_mask_vv_nxv4i8_nxv4i8:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e8, mf2, ta, mu
-; CHECK-NEXT:    vmsltu.vv v8, v9, v8
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv1r.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e8, mf2, ta, mu
+; CHECK-NEXT:    vmsltu.vv v0, v9, v8
 ; CHECK-NEXT:    vmsltu.vv v11, v10, v9, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v11
 ; CHECK-NEXT:    ret
@@ -190,10 +187,9 @@ declare <vscale x 8 x i1> @llvm.riscv.vmsgtu.mask.nxv8i8(
 define <vscale x 8 x i1> @intrinsic_vmsgtu_mask_vv_nxv8i8_nxv8i8(<vscale x 8 x i1> %0, <vscale x 8 x i8> %1, <vscale x 8 x i8> %2, <vscale x 8 x i8> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsgtu_mask_vv_nxv8i8_nxv8i8:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e8, m1, ta, mu
-; CHECK-NEXT:    vmsltu.vv v8, v9, v8
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv.v.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e8, m1, ta, mu
+; CHECK-NEXT:    vmsltu.vv v0, v9, v8
 ; CHECK-NEXT:    vmsltu.vv v11, v10, v9, v0.t
 ; CHECK-NEXT:    vmv.v.v v0, v11
 ; CHECK-NEXT:    ret
@@ -242,12 +238,11 @@ declare <vscale x 16 x i1> @llvm.riscv.vmsgtu.mask.nxv16i8(
 define <vscale x 16 x i1> @intrinsic_vmsgtu_mask_vv_nxv16i8_nxv16i8(<vscale x 16 x i1> %0, <vscale x 16 x i8> %1, <vscale x 16 x i8> %2, <vscale x 16 x i8> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsgtu_mask_vv_nxv16i8_nxv16i8:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v14, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e8, m2, ta, mu
-; CHECK-NEXT:    vmsltu.vv v14, v10, v8
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmsltu.vv v0, v10, v8
+; CHECK-NEXT:    vmsltu.vv v14, v12, v10, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v14
-; CHECK-NEXT:    vmsltu.vv v8, v12, v10, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 16 x i1> @llvm.riscv.vmsgtu.nxv16i8(
@@ -294,12 +289,11 @@ declare <vscale x 32 x i1> @llvm.riscv.vmsgtu.mask.nxv32i8(
 define <vscale x 32 x i1> @intrinsic_vmsgtu_mask_vv_nxv32i8_nxv32i8(<vscale x 32 x i1> %0, <vscale x 32 x i8> %1, <vscale x 32 x i8> %2, <vscale x 32 x i8> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsgtu_mask_vv_nxv32i8_nxv32i8:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v20, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e8, m4, ta, mu
-; CHECK-NEXT:    vmsltu.vv v20, v12, v8
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmsltu.vv v0, v12, v8
+; CHECK-NEXT:    vmsltu.vv v20, v16, v12, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v20
-; CHECK-NEXT:    vmsltu.vv v8, v16, v12, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 32 x i1> @llvm.riscv.vmsgtu.nxv32i8(
@@ -346,10 +340,9 @@ declare <vscale x 1 x i1> @llvm.riscv.vmsgtu.mask.nxv1i16(
 define <vscale x 1 x i1> @intrinsic_vmsgtu_mask_vv_nxv1i16_nxv1i16(<vscale x 1 x i1> %0, <vscale x 1 x i16> %1, <vscale x 1 x i16> %2, <vscale x 1 x i16> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsgtu_mask_vv_nxv1i16_nxv1i16:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
-; CHECK-NEXT:    vmsltu.vv v8, v9, v8
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv1r.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
+; CHECK-NEXT:    vmsltu.vv v0, v9, v8
 ; CHECK-NEXT:    vmsltu.vv v11, v10, v9, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v11
 ; CHECK-NEXT:    ret
@@ -398,10 +391,9 @@ declare <vscale x 2 x i1> @llvm.riscv.vmsgtu.mask.nxv2i16(
 define <vscale x 2 x i1> @intrinsic_vmsgtu_mask_vv_nxv2i16_nxv2i16(<vscale x 2 x i1> %0, <vscale x 2 x i16> %1, <vscale x 2 x i16> %2, <vscale x 2 x i16> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsgtu_mask_vv_nxv2i16_nxv2i16:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e16, mf2, ta, mu
-; CHECK-NEXT:    vmsltu.vv v8, v9, v8
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv1r.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e16, mf2, ta, mu
+; CHECK-NEXT:    vmsltu.vv v0, v9, v8
 ; CHECK-NEXT:    vmsltu.vv v11, v10, v9, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v11
 ; CHECK-NEXT:    ret
@@ -450,10 +442,9 @@ declare <vscale x 4 x i1> @llvm.riscv.vmsgtu.mask.nxv4i16(
 define <vscale x 4 x i1> @intrinsic_vmsgtu_mask_vv_nxv4i16_nxv4i16(<vscale x 4 x i1> %0, <vscale x 4 x i16> %1, <vscale x 4 x i16> %2, <vscale x 4 x i16> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsgtu_mask_vv_nxv4i16_nxv4i16:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e16, m1, ta, mu
-; CHECK-NEXT:    vmsltu.vv v8, v9, v8
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv.v.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e16, m1, ta, mu
+; CHECK-NEXT:    vmsltu.vv v0, v9, v8
 ; CHECK-NEXT:    vmsltu.vv v11, v10, v9, v0.t
 ; CHECK-NEXT:    vmv.v.v v0, v11
 ; CHECK-NEXT:    ret
@@ -502,12 +493,11 @@ declare <vscale x 8 x i1> @llvm.riscv.vmsgtu.mask.nxv8i16(
 define <vscale x 8 x i1> @intrinsic_vmsgtu_mask_vv_nxv8i16_nxv8i16(<vscale x 8 x i1> %0, <vscale x 8 x i16> %1, <vscale x 8 x i16> %2, <vscale x 8 x i16> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsgtu_mask_vv_nxv8i16_nxv8i16:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v14, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e16, m2, ta, mu
-; CHECK-NEXT:    vmsltu.vv v14, v10, v8
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmsltu.vv v0, v10, v8
+; CHECK-NEXT:    vmsltu.vv v14, v12, v10, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v14
-; CHECK-NEXT:    vmsltu.vv v8, v12, v10, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 8 x i1> @llvm.riscv.vmsgtu.nxv8i16(
@@ -554,12 +544,11 @@ declare <vscale x 16 x i1> @llvm.riscv.vmsgtu.mask.nxv16i16(
 define <vscale x 16 x i1> @intrinsic_vmsgtu_mask_vv_nxv16i16_nxv16i16(<vscale x 16 x i1> %0, <vscale x 16 x i16> %1, <vscale x 16 x i16> %2, <vscale x 16 x i16> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsgtu_mask_vv_nxv16i16_nxv16i16:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v20, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e16, m4, ta, mu
-; CHECK-NEXT:    vmsltu.vv v20, v12, v8
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmsltu.vv v0, v12, v8
+; CHECK-NEXT:    vmsltu.vv v20, v16, v12, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v20
-; CHECK-NEXT:    vmsltu.vv v8, v16, v12, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 16 x i1> @llvm.riscv.vmsgtu.nxv16i16(
@@ -606,10 +595,9 @@ declare <vscale x 1 x i1> @llvm.riscv.vmsgtu.mask.nxv1i32(
 define <vscale x 1 x i1> @intrinsic_vmsgtu_mask_vv_nxv1i32_nxv1i32(<vscale x 1 x i1> %0, <vscale x 1 x i32> %1, <vscale x 1 x i32> %2, <vscale x 1 x i32> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsgtu_mask_vv_nxv1i32_nxv1i32:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e32, mf2, ta, mu
-; CHECK-NEXT:    vmsltu.vv v8, v9, v8
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv1r.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e32, mf2, ta, mu
+; CHECK-NEXT:    vmsltu.vv v0, v9, v8
 ; CHECK-NEXT:    vmsltu.vv v11, v10, v9, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v11
 ; CHECK-NEXT:    ret
@@ -658,10 +646,9 @@ declare <vscale x 2 x i1> @llvm.riscv.vmsgtu.mask.nxv2i32(
 define <vscale x 2 x i1> @intrinsic_vmsgtu_mask_vv_nxv2i32_nxv2i32(<vscale x 2 x i1> %0, <vscale x 2 x i32> %1, <vscale x 2 x i32> %2, <vscale x 2 x i32> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsgtu_mask_vv_nxv2i32_nxv2i32:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e32, m1, ta, mu
-; CHECK-NEXT:    vmsltu.vv v8, v9, v8
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv.v.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e32, m1, ta, mu
+; CHECK-NEXT:    vmsltu.vv v0, v9, v8
 ; CHECK-NEXT:    vmsltu.vv v11, v10, v9, v0.t
 ; CHECK-NEXT:    vmv.v.v v0, v11
 ; CHECK-NEXT:    ret
@@ -710,12 +697,11 @@ declare <vscale x 4 x i1> @llvm.riscv.vmsgtu.mask.nxv4i32(
 define <vscale x 4 x i1> @intrinsic_vmsgtu_mask_vv_nxv4i32_nxv4i32(<vscale x 4 x i1> %0, <vscale x 4 x i32> %1, <vscale x 4 x i32> %2, <vscale x 4 x i32> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsgtu_mask_vv_nxv4i32_nxv4i32:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v14, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e32, m2, ta, mu
-; CHECK-NEXT:    vmsltu.vv v14, v10, v8
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmsltu.vv v0, v10, v8
+; CHECK-NEXT:    vmsltu.vv v14, v12, v10, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v14
-; CHECK-NEXT:    vmsltu.vv v8, v12, v10, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 4 x i1> @llvm.riscv.vmsgtu.nxv4i32(
@@ -762,12 +748,11 @@ declare <vscale x 8 x i1> @llvm.riscv.vmsgtu.mask.nxv8i32(
 define <vscale x 8 x i1> @intrinsic_vmsgtu_mask_vv_nxv8i32_nxv8i32(<vscale x 8 x i1> %0, <vscale x 8 x i32> %1, <vscale x 8 x i32> %2, <vscale x 8 x i32> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsgtu_mask_vv_nxv8i32_nxv8i32:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v20, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e32, m4, ta, mu
-; CHECK-NEXT:    vmsltu.vv v20, v12, v8
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmsltu.vv v0, v12, v8
+; CHECK-NEXT:    vmsltu.vv v20, v16, v12, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v20
-; CHECK-NEXT:    vmsltu.vv v8, v16, v12, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 8 x i1> @llvm.riscv.vmsgtu.nxv8i32(
@@ -814,10 +799,9 @@ declare <vscale x 1 x i1> @llvm.riscv.vmsgtu.mask.nxv1i64(
 define <vscale x 1 x i1> @intrinsic_vmsgtu_mask_vv_nxv1i64_nxv1i64(<vscale x 1 x i1> %0, <vscale x 1 x i64> %1, <vscale x 1 x i64> %2, <vscale x 1 x i64> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsgtu_mask_vv_nxv1i64_nxv1i64:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e64, m1, ta, mu
-; CHECK-NEXT:    vmsltu.vv v8, v9, v8
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv.v.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e64, m1, ta, mu
+; CHECK-NEXT:    vmsltu.vv v0, v9, v8
 ; CHECK-NEXT:    vmsltu.vv v11, v10, v9, v0.t
 ; CHECK-NEXT:    vmv.v.v v0, v11
 ; CHECK-NEXT:    ret
@@ -866,12 +850,11 @@ declare <vscale x 2 x i1> @llvm.riscv.vmsgtu.mask.nxv2i64(
 define <vscale x 2 x i1> @intrinsic_vmsgtu_mask_vv_nxv2i64_nxv2i64(<vscale x 2 x i1> %0, <vscale x 2 x i64> %1, <vscale x 2 x i64> %2, <vscale x 2 x i64> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsgtu_mask_vv_nxv2i64_nxv2i64:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v14, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e64, m2, ta, mu
-; CHECK-NEXT:    vmsltu.vv v14, v10, v8
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmsltu.vv v0, v10, v8
+; CHECK-NEXT:    vmsltu.vv v14, v12, v10, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v14
-; CHECK-NEXT:    vmsltu.vv v8, v12, v10, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 2 x i1> @llvm.riscv.vmsgtu.nxv2i64(
@@ -918,12 +901,11 @@ declare <vscale x 4 x i1> @llvm.riscv.vmsgtu.mask.nxv4i64(
 define <vscale x 4 x i1> @intrinsic_vmsgtu_mask_vv_nxv4i64_nxv4i64(<vscale x 4 x i1> %0, <vscale x 4 x i64> %1, <vscale x 4 x i64> %2, <vscale x 4 x i64> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsgtu_mask_vv_nxv4i64_nxv4i64:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v20, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e64, m4, ta, mu
-; CHECK-NEXT:    vmsltu.vv v20, v12, v8
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmsltu.vv v0, v12, v8
+; CHECK-NEXT:    vmsltu.vv v20, v16, v12, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v20
-; CHECK-NEXT:    vmsltu.vv v8, v16, v12, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 4 x i1> @llvm.riscv.vmsgtu.nxv4i64(
diff --git a/llvm/test/CodeGen/RISCV/rvv/vmsle.ll b/llvm/test/CodeGen/RISCV/rvv/vmsle.ll
index 2248ba03adfe..5d5a28edbfe1 100644
--- a/llvm/test/CodeGen/RISCV/rvv/vmsle.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/vmsle.ll
@@ -34,10 +34,9 @@ declare <vscale x 1 x i1> @llvm.riscv.vmsle.mask.nxv1i8(
 define <vscale x 1 x i1> @intrinsic_vmsle_mask_vv_nxv1i8_nxv1i8(<vscale x 1 x i1> %0, <vscale x 1 x i8> %1, <vscale x 1 x i8> %2, <vscale x 1 x i8> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsle_mask_vv_nxv1i8_nxv1i8:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e8, mf8, ta, mu
-; CHECK-NEXT:    vmsle.vv v8, v8, v9
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv1r.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e8, mf8, ta, mu
+; CHECK-NEXT:    vmsle.vv v0, v8, v9
 ; CHECK-NEXT:    vmsle.vv v11, v9, v10, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v11
 ; CHECK-NEXT:    ret
@@ -86,10 +85,9 @@ declare <vscale x 2 x i1> @llvm.riscv.vmsle.mask.nxv2i8(
 define <vscale x 2 x i1> @intrinsic_vmsle_mask_vv_nxv2i8_nxv2i8(<vscale x 2 x i1> %0, <vscale x 2 x i8> %1, <vscale x 2 x i8> %2, <vscale x 2 x i8> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsle_mask_vv_nxv2i8_nxv2i8:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e8, mf4, ta, mu
-; CHECK-NEXT:    vmsle.vv v8, v8, v9
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv1r.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e8, mf4, ta, mu
+; CHECK-NEXT:    vmsle.vv v0, v8, v9
 ; CHECK-NEXT:    vmsle.vv v11, v9, v10, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v11
 ; CHECK-NEXT:    ret
@@ -138,10 +136,9 @@ declare <vscale x 4 x i1> @llvm.riscv.vmsle.mask.nxv4i8(
 define <vscale x 4 x i1> @intrinsic_vmsle_mask_vv_nxv4i8_nxv4i8(<vscale x 4 x i1> %0, <vscale x 4 x i8> %1, <vscale x 4 x i8> %2, <vscale x 4 x i8> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsle_mask_vv_nxv4i8_nxv4i8:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e8, mf2, ta, mu
-; CHECK-NEXT:    vmsle.vv v8, v8, v9
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv1r.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e8, mf2, ta, mu
+; CHECK-NEXT:    vmsle.vv v0, v8, v9
 ; CHECK-NEXT:    vmsle.vv v11, v9, v10, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v11
 ; CHECK-NEXT:    ret
@@ -190,10 +187,9 @@ declare <vscale x 8 x i1> @llvm.riscv.vmsle.mask.nxv8i8(
 define <vscale x 8 x i1> @intrinsic_vmsle_mask_vv_nxv8i8_nxv8i8(<vscale x 8 x i1> %0, <vscale x 8 x i8> %1, <vscale x 8 x i8> %2, <vscale x 8 x i8> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsle_mask_vv_nxv8i8_nxv8i8:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e8, m1, ta, mu
-; CHECK-NEXT:    vmsle.vv v8, v8, v9
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv.v.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e8, m1, ta, mu
+; CHECK-NEXT:    vmsle.vv v0, v8, v9
 ; CHECK-NEXT:    vmsle.vv v11, v9, v10, v0.t
 ; CHECK-NEXT:    vmv.v.v v0, v11
 ; CHECK-NEXT:    ret
@@ -242,12 +238,11 @@ declare <vscale x 16 x i1> @llvm.riscv.vmsle.mask.nxv16i8(
 define <vscale x 16 x i1> @intrinsic_vmsle_mask_vv_nxv16i8_nxv16i8(<vscale x 16 x i1> %0, <vscale x 16 x i8> %1, <vscale x 16 x i8> %2, <vscale x 16 x i8> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsle_mask_vv_nxv16i8_nxv16i8:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v14, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e8, m2, ta, mu
-; CHECK-NEXT:    vmsle.vv v14, v8, v10
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmsle.vv v0, v8, v10
+; CHECK-NEXT:    vmsle.vv v14, v10, v12, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v14
-; CHECK-NEXT:    vmsle.vv v8, v10, v12, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 16 x i1> @llvm.riscv.vmsle.nxv16i8(
@@ -294,12 +289,11 @@ declare <vscale x 32 x i1> @llvm.riscv.vmsle.mask.nxv32i8(
 define <vscale x 32 x i1> @intrinsic_vmsle_mask_vv_nxv32i8_nxv32i8(<vscale x 32 x i1> %0, <vscale x 32 x i8> %1, <vscale x 32 x i8> %2, <vscale x 32 x i8> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsle_mask_vv_nxv32i8_nxv32i8:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v20, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e8, m4, ta, mu
-; CHECK-NEXT:    vmsle.vv v20, v8, v12
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmsle.vv v0, v8, v12
+; CHECK-NEXT:    vmsle.vv v20, v12, v16, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v20
-; CHECK-NEXT:    vmsle.vv v8, v12, v16, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 32 x i1> @llvm.riscv.vmsle.nxv32i8(
@@ -346,10 +340,9 @@ declare <vscale x 1 x i1> @llvm.riscv.vmsle.mask.nxv1i16(
 define <vscale x 1 x i1> @intrinsic_vmsle_mask_vv_nxv1i16_nxv1i16(<vscale x 1 x i1> %0, <vscale x 1 x i16> %1, <vscale x 1 x i16> %2, <vscale x 1 x i16> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsle_mask_vv_nxv1i16_nxv1i16:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
-; CHECK-NEXT:    vmsle.vv v8, v8, v9
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv1r.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
+; CHECK-NEXT:    vmsle.vv v0, v8, v9
 ; CHECK-NEXT:    vmsle.vv v11, v9, v10, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v11
 ; CHECK-NEXT:    ret
@@ -398,10 +391,9 @@ declare <vscale x 2 x i1> @llvm.riscv.vmsle.mask.nxv2i16(
 define <vscale x 2 x i1> @intrinsic_vmsle_mask_vv_nxv2i16_nxv2i16(<vscale x 2 x i1> %0, <vscale x 2 x i16> %1, <vscale x 2 x i16> %2, <vscale x 2 x i16> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsle_mask_vv_nxv2i16_nxv2i16:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e16, mf2, ta, mu
-; CHECK-NEXT:    vmsle.vv v8, v8, v9
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv1r.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e16, mf2, ta, mu
+; CHECK-NEXT:    vmsle.vv v0, v8, v9
 ; CHECK-NEXT:    vmsle.vv v11, v9, v10, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v11
 ; CHECK-NEXT:    ret
@@ -450,10 +442,9 @@ declare <vscale x 4 x i1> @llvm.riscv.vmsle.mask.nxv4i16(
 define <vscale x 4 x i1> @intrinsic_vmsle_mask_vv_nxv4i16_nxv4i16(<vscale x 4 x i1> %0, <vscale x 4 x i16> %1, <vscale x 4 x i16> %2, <vscale x 4 x i16> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsle_mask_vv_nxv4i16_nxv4i16:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e16, m1, ta, mu
-; CHECK-NEXT:    vmsle.vv v8, v8, v9
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv.v.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e16, m1, ta, mu
+; CHECK-NEXT:    vmsle.vv v0, v8, v9
 ; CHECK-NEXT:    vmsle.vv v11, v9, v10, v0.t
 ; CHECK-NEXT:    vmv.v.v v0, v11
 ; CHECK-NEXT:    ret
@@ -502,12 +493,11 @@ declare <vscale x 8 x i1> @llvm.riscv.vmsle.mask.nxv8i16(
 define <vscale x 8 x i1> @intrinsic_vmsle_mask_vv_nxv8i16_nxv8i16(<vscale x 8 x i1> %0, <vscale x 8 x i16> %1, <vscale x 8 x i16> %2, <vscale x 8 x i16> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsle_mask_vv_nxv8i16_nxv8i16:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v14, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e16, m2, ta, mu
-; CHECK-NEXT:    vmsle.vv v14, v8, v10
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmsle.vv v0, v8, v10
+; CHECK-NEXT:    vmsle.vv v14, v10, v12, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v14
-; CHECK-NEXT:    vmsle.vv v8, v10, v12, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 8 x i1> @llvm.riscv.vmsle.nxv8i16(
@@ -554,12 +544,11 @@ declare <vscale x 16 x i1> @llvm.riscv.vmsle.mask.nxv16i16(
 define <vscale x 16 x i1> @intrinsic_vmsle_mask_vv_nxv16i16_nxv16i16(<vscale x 16 x i1> %0, <vscale x 16 x i16> %1, <vscale x 16 x i16> %2, <vscale x 16 x i16> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsle_mask_vv_nxv16i16_nxv16i16:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v20, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e16, m4, ta, mu
-; CHECK-NEXT:    vmsle.vv v20, v8, v12
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmsle.vv v0, v8, v12
+; CHECK-NEXT:    vmsle.vv v20, v12, v16, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v20
-; CHECK-NEXT:    vmsle.vv v8, v12, v16, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 16 x i1> @llvm.riscv.vmsle.nxv16i16(
@@ -606,10 +595,9 @@ declare <vscale x 1 x i1> @llvm.riscv.vmsle.mask.nxv1i32(
 define <vscale x 1 x i1> @intrinsic_vmsle_mask_vv_nxv1i32_nxv1i32(<vscale x 1 x i1> %0, <vscale x 1 x i32> %1, <vscale x 1 x i32> %2, <vscale x 1 x i32> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsle_mask_vv_nxv1i32_nxv1i32:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e32, mf2, ta, mu
-; CHECK-NEXT:    vmsle.vv v8, v8, v9
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv1r.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e32, mf2, ta, mu
+; CHECK-NEXT:    vmsle.vv v0, v8, v9
 ; CHECK-NEXT:    vmsle.vv v11, v9, v10, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v11
 ; CHECK-NEXT:    ret
@@ -658,10 +646,9 @@ declare <vscale x 2 x i1> @llvm.riscv.vmsle.mask.nxv2i32(
 define <vscale x 2 x i1> @intrinsic_vmsle_mask_vv_nxv2i32_nxv2i32(<vscale x 2 x i1> %0, <vscale x 2 x i32> %1, <vscale x 2 x i32> %2, <vscale x 2 x i32> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsle_mask_vv_nxv2i32_nxv2i32:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e32, m1, ta, mu
-; CHECK-NEXT:    vmsle.vv v8, v8, v9
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv.v.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e32, m1, ta, mu
+; CHECK-NEXT:    vmsle.vv v0, v8, v9
 ; CHECK-NEXT:    vmsle.vv v11, v9, v10, v0.t
 ; CHECK-NEXT:    vmv.v.v v0, v11
 ; CHECK-NEXT:    ret
@@ -710,12 +697,11 @@ declare <vscale x 4 x i1> @llvm.riscv.vmsle.mask.nxv4i32(
 define <vscale x 4 x i1> @intrinsic_vmsle_mask_vv_nxv4i32_nxv4i32(<vscale x 4 x i1> %0, <vscale x 4 x i32> %1, <vscale x 4 x i32> %2, <vscale x 4 x i32> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsle_mask_vv_nxv4i32_nxv4i32:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v14, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e32, m2, ta, mu
-; CHECK-NEXT:    vmsle.vv v14, v8, v10
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmsle.vv v0, v8, v10
+; CHECK-NEXT:    vmsle.vv v14, v10, v12, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v14
-; CHECK-NEXT:    vmsle.vv v8, v10, v12, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 4 x i1> @llvm.riscv.vmsle.nxv4i32(
@@ -762,12 +748,11 @@ declare <vscale x 8 x i1> @llvm.riscv.vmsle.mask.nxv8i32(
 define <vscale x 8 x i1> @intrinsic_vmsle_mask_vv_nxv8i32_nxv8i32(<vscale x 8 x i1> %0, <vscale x 8 x i32> %1, <vscale x 8 x i32> %2, <vscale x 8 x i32> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsle_mask_vv_nxv8i32_nxv8i32:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v20, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e32, m4, ta, mu
-; CHECK-NEXT:    vmsle.vv v20, v8, v12
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmsle.vv v0, v8, v12
+; CHECK-NEXT:    vmsle.vv v20, v12, v16, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v20
-; CHECK-NEXT:    vmsle.vv v8, v12, v16, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 8 x i1> @llvm.riscv.vmsle.nxv8i32(
@@ -814,10 +799,9 @@ declare <vscale x 1 x i1> @llvm.riscv.vmsle.mask.nxv1i64(
 define <vscale x 1 x i1> @intrinsic_vmsle_mask_vv_nxv1i64_nxv1i64(<vscale x 1 x i1> %0, <vscale x 1 x i64> %1, <vscale x 1 x i64> %2, <vscale x 1 x i64> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsle_mask_vv_nxv1i64_nxv1i64:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e64, m1, ta, mu
-; CHECK-NEXT:    vmsle.vv v8, v8, v9
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv.v.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e64, m1, ta, mu
+; CHECK-NEXT:    vmsle.vv v0, v8, v9
 ; CHECK-NEXT:    vmsle.vv v11, v9, v10, v0.t
 ; CHECK-NEXT:    vmv.v.v v0, v11
 ; CHECK-NEXT:    ret
@@ -866,12 +850,11 @@ declare <vscale x 2 x i1> @llvm.riscv.vmsle.mask.nxv2i64(
 define <vscale x 2 x i1> @intrinsic_vmsle_mask_vv_nxv2i64_nxv2i64(<vscale x 2 x i1> %0, <vscale x 2 x i64> %1, <vscale x 2 x i64> %2, <vscale x 2 x i64> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsle_mask_vv_nxv2i64_nxv2i64:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v14, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e64, m2, ta, mu
-; CHECK-NEXT:    vmsle.vv v14, v8, v10
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmsle.vv v0, v8, v10
+; CHECK-NEXT:    vmsle.vv v14, v10, v12, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v14
-; CHECK-NEXT:    vmsle.vv v8, v10, v12, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 2 x i1> @llvm.riscv.vmsle.nxv2i64(
@@ -918,12 +901,11 @@ declare <vscale x 4 x i1> @llvm.riscv.vmsle.mask.nxv4i64(
 define <vscale x 4 x i1> @intrinsic_vmsle_mask_vv_nxv4i64_nxv4i64(<vscale x 4 x i1> %0, <vscale x 4 x i64> %1, <vscale x 4 x i64> %2, <vscale x 4 x i64> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsle_mask_vv_nxv4i64_nxv4i64:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v20, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e64, m4, ta, mu
-; CHECK-NEXT:    vmsle.vv v20, v8, v12
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmsle.vv v0, v8, v12
+; CHECK-NEXT:    vmsle.vv v20, v12, v16, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v20
-; CHECK-NEXT:    vmsle.vv v8, v12, v16, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 4 x i1> @llvm.riscv.vmsle.nxv4i64(
diff --git a/llvm/test/CodeGen/RISCV/rvv/vmsleu.ll b/llvm/test/CodeGen/RISCV/rvv/vmsleu.ll
index 57bae83b25e0..c58ac2d07183 100644
--- a/llvm/test/CodeGen/RISCV/rvv/vmsleu.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/vmsleu.ll
@@ -34,10 +34,9 @@ declare <vscale x 1 x i1> @llvm.riscv.vmsleu.mask.nxv1i8(
 define <vscale x 1 x i1> @intrinsic_vmsleu_mask_vv_nxv1i8_nxv1i8(<vscale x 1 x i1> %0, <vscale x 1 x i8> %1, <vscale x 1 x i8> %2, <vscale x 1 x i8> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsleu_mask_vv_nxv1i8_nxv1i8:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e8, mf8, ta, mu
-; CHECK-NEXT:    vmsleu.vv v8, v8, v9
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv1r.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e8, mf8, ta, mu
+; CHECK-NEXT:    vmsleu.vv v0, v8, v9
 ; CHECK-NEXT:    vmsleu.vv v11, v9, v10, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v11
 ; CHECK-NEXT:    ret
@@ -86,10 +85,9 @@ declare <vscale x 2 x i1> @llvm.riscv.vmsleu.mask.nxv2i8(
 define <vscale x 2 x i1> @intrinsic_vmsleu_mask_vv_nxv2i8_nxv2i8(<vscale x 2 x i1> %0, <vscale x 2 x i8> %1, <vscale x 2 x i8> %2, <vscale x 2 x i8> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsleu_mask_vv_nxv2i8_nxv2i8:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e8, mf4, ta, mu
-; CHECK-NEXT:    vmsleu.vv v8, v8, v9
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv1r.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e8, mf4, ta, mu
+; CHECK-NEXT:    vmsleu.vv v0, v8, v9
 ; CHECK-NEXT:    vmsleu.vv v11, v9, v10, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v11
 ; CHECK-NEXT:    ret
@@ -138,10 +136,9 @@ declare <vscale x 4 x i1> @llvm.riscv.vmsleu.mask.nxv4i8(
 define <vscale x 4 x i1> @intrinsic_vmsleu_mask_vv_nxv4i8_nxv4i8(<vscale x 4 x i1> %0, <vscale x 4 x i8> %1, <vscale x 4 x i8> %2, <vscale x 4 x i8> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsleu_mask_vv_nxv4i8_nxv4i8:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e8, mf2, ta, mu
-; CHECK-NEXT:    vmsleu.vv v8, v8, v9
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv1r.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e8, mf2, ta, mu
+; CHECK-NEXT:    vmsleu.vv v0, v8, v9
 ; CHECK-NEXT:    vmsleu.vv v11, v9, v10, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v11
 ; CHECK-NEXT:    ret
@@ -190,10 +187,9 @@ declare <vscale x 8 x i1> @llvm.riscv.vmsleu.mask.nxv8i8(
 define <vscale x 8 x i1> @intrinsic_vmsleu_mask_vv_nxv8i8_nxv8i8(<vscale x 8 x i1> %0, <vscale x 8 x i8> %1, <vscale x 8 x i8> %2, <vscale x 8 x i8> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsleu_mask_vv_nxv8i8_nxv8i8:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e8, m1, ta, mu
-; CHECK-NEXT:    vmsleu.vv v8, v8, v9
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv.v.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e8, m1, ta, mu
+; CHECK-NEXT:    vmsleu.vv v0, v8, v9
 ; CHECK-NEXT:    vmsleu.vv v11, v9, v10, v0.t
 ; CHECK-NEXT:    vmv.v.v v0, v11
 ; CHECK-NEXT:    ret
@@ -242,12 +238,11 @@ declare <vscale x 16 x i1> @llvm.riscv.vmsleu.mask.nxv16i8(
 define <vscale x 16 x i1> @intrinsic_vmsleu_mask_vv_nxv16i8_nxv16i8(<vscale x 16 x i1> %0, <vscale x 16 x i8> %1, <vscale x 16 x i8> %2, <vscale x 16 x i8> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsleu_mask_vv_nxv16i8_nxv16i8:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v14, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e8, m2, ta, mu
-; CHECK-NEXT:    vmsleu.vv v14, v8, v10
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmsleu.vv v0, v8, v10
+; CHECK-NEXT:    vmsleu.vv v14, v10, v12, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v14
-; CHECK-NEXT:    vmsleu.vv v8, v10, v12, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 16 x i1> @llvm.riscv.vmsleu.nxv16i8(
@@ -294,12 +289,11 @@ declare <vscale x 32 x i1> @llvm.riscv.vmsleu.mask.nxv32i8(
 define <vscale x 32 x i1> @intrinsic_vmsleu_mask_vv_nxv32i8_nxv32i8(<vscale x 32 x i1> %0, <vscale x 32 x i8> %1, <vscale x 32 x i8> %2, <vscale x 32 x i8> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsleu_mask_vv_nxv32i8_nxv32i8:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v20, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e8, m4, ta, mu
-; CHECK-NEXT:    vmsleu.vv v20, v8, v12
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmsleu.vv v0, v8, v12
+; CHECK-NEXT:    vmsleu.vv v20, v12, v16, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v20
-; CHECK-NEXT:    vmsleu.vv v8, v12, v16, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 32 x i1> @llvm.riscv.vmsleu.nxv32i8(
@@ -346,10 +340,9 @@ declare <vscale x 1 x i1> @llvm.riscv.vmsleu.mask.nxv1i16(
 define <vscale x 1 x i1> @intrinsic_vmsleu_mask_vv_nxv1i16_nxv1i16(<vscale x 1 x i1> %0, <vscale x 1 x i16> %1, <vscale x 1 x i16> %2, <vscale x 1 x i16> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsleu_mask_vv_nxv1i16_nxv1i16:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
-; CHECK-NEXT:    vmsleu.vv v8, v8, v9
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv1r.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
+; CHECK-NEXT:    vmsleu.vv v0, v8, v9
 ; CHECK-NEXT:    vmsleu.vv v11, v9, v10, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v11
 ; CHECK-NEXT:    ret
@@ -398,10 +391,9 @@ declare <vscale x 2 x i1> @llvm.riscv.vmsleu.mask.nxv2i16(
 define <vscale x 2 x i1> @intrinsic_vmsleu_mask_vv_nxv2i16_nxv2i16(<vscale x 2 x i1> %0, <vscale x 2 x i16> %1, <vscale x 2 x i16> %2, <vscale x 2 x i16> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsleu_mask_vv_nxv2i16_nxv2i16:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e16, mf2, ta, mu
-; CHECK-NEXT:    vmsleu.vv v8, v8, v9
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv1r.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e16, mf2, ta, mu
+; CHECK-NEXT:    vmsleu.vv v0, v8, v9
 ; CHECK-NEXT:    vmsleu.vv v11, v9, v10, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v11
 ; CHECK-NEXT:    ret
@@ -450,10 +442,9 @@ declare <vscale x 4 x i1> @llvm.riscv.vmsleu.mask.nxv4i16(
 define <vscale x 4 x i1> @intrinsic_vmsleu_mask_vv_nxv4i16_nxv4i16(<vscale x 4 x i1> %0, <vscale x 4 x i16> %1, <vscale x 4 x i16> %2, <vscale x 4 x i16> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsleu_mask_vv_nxv4i16_nxv4i16:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e16, m1, ta, mu
-; CHECK-NEXT:    vmsleu.vv v8, v8, v9
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv.v.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e16, m1, ta, mu
+; CHECK-NEXT:    vmsleu.vv v0, v8, v9
 ; CHECK-NEXT:    vmsleu.vv v11, v9, v10, v0.t
 ; CHECK-NEXT:    vmv.v.v v0, v11
 ; CHECK-NEXT:    ret
@@ -502,12 +493,11 @@ declare <vscale x 8 x i1> @llvm.riscv.vmsleu.mask.nxv8i16(
 define <vscale x 8 x i1> @intrinsic_vmsleu_mask_vv_nxv8i16_nxv8i16(<vscale x 8 x i1> %0, <vscale x 8 x i16> %1, <vscale x 8 x i16> %2, <vscale x 8 x i16> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsleu_mask_vv_nxv8i16_nxv8i16:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v14, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e16, m2, ta, mu
-; CHECK-NEXT:    vmsleu.vv v14, v8, v10
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmsleu.vv v0, v8, v10
+; CHECK-NEXT:    vmsleu.vv v14, v10, v12, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v14
-; CHECK-NEXT:    vmsleu.vv v8, v10, v12, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 8 x i1> @llvm.riscv.vmsleu.nxv8i16(
@@ -554,12 +544,11 @@ declare <vscale x 16 x i1> @llvm.riscv.vmsleu.mask.nxv16i16(
 define <vscale x 16 x i1> @intrinsic_vmsleu_mask_vv_nxv16i16_nxv16i16(<vscale x 16 x i1> %0, <vscale x 16 x i16> %1, <vscale x 16 x i16> %2, <vscale x 16 x i16> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsleu_mask_vv_nxv16i16_nxv16i16:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v20, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e16, m4, ta, mu
-; CHECK-NEXT:    vmsleu.vv v20, v8, v12
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmsleu.vv v0, v8, v12
+; CHECK-NEXT:    vmsleu.vv v20, v12, v16, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v20
-; CHECK-NEXT:    vmsleu.vv v8, v12, v16, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 16 x i1> @llvm.riscv.vmsleu.nxv16i16(
@@ -606,10 +595,9 @@ declare <vscale x 1 x i1> @llvm.riscv.vmsleu.mask.nxv1i32(
 define <vscale x 1 x i1> @intrinsic_vmsleu_mask_vv_nxv1i32_nxv1i32(<vscale x 1 x i1> %0, <vscale x 1 x i32> %1, <vscale x 1 x i32> %2, <vscale x 1 x i32> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsleu_mask_vv_nxv1i32_nxv1i32:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e32, mf2, ta, mu
-; CHECK-NEXT:    vmsleu.vv v8, v8, v9
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv1r.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e32, mf2, ta, mu
+; CHECK-NEXT:    vmsleu.vv v0, v8, v9
 ; CHECK-NEXT:    vmsleu.vv v11, v9, v10, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v11
 ; CHECK-NEXT:    ret
@@ -658,10 +646,9 @@ declare <vscale x 2 x i1> @llvm.riscv.vmsleu.mask.nxv2i32(
 define <vscale x 2 x i1> @intrinsic_vmsleu_mask_vv_nxv2i32_nxv2i32(<vscale x 2 x i1> %0, <vscale x 2 x i32> %1, <vscale x 2 x i32> %2, <vscale x 2 x i32> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsleu_mask_vv_nxv2i32_nxv2i32:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e32, m1, ta, mu
-; CHECK-NEXT:    vmsleu.vv v8, v8, v9
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv.v.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e32, m1, ta, mu
+; CHECK-NEXT:    vmsleu.vv v0, v8, v9
 ; CHECK-NEXT:    vmsleu.vv v11, v9, v10, v0.t
 ; CHECK-NEXT:    vmv.v.v v0, v11
 ; CHECK-NEXT:    ret
@@ -710,12 +697,11 @@ declare <vscale x 4 x i1> @llvm.riscv.vmsleu.mask.nxv4i32(
 define <vscale x 4 x i1> @intrinsic_vmsleu_mask_vv_nxv4i32_nxv4i32(<vscale x 4 x i1> %0, <vscale x 4 x i32> %1, <vscale x 4 x i32> %2, <vscale x 4 x i32> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsleu_mask_vv_nxv4i32_nxv4i32:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v14, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e32, m2, ta, mu
-; CHECK-NEXT:    vmsleu.vv v14, v8, v10
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmsleu.vv v0, v8, v10
+; CHECK-NEXT:    vmsleu.vv v14, v10, v12, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v14
-; CHECK-NEXT:    vmsleu.vv v8, v10, v12, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 4 x i1> @llvm.riscv.vmsleu.nxv4i32(
@@ -762,12 +748,11 @@ declare <vscale x 8 x i1> @llvm.riscv.vmsleu.mask.nxv8i32(
 define <vscale x 8 x i1> @intrinsic_vmsleu_mask_vv_nxv8i32_nxv8i32(<vscale x 8 x i1> %0, <vscale x 8 x i32> %1, <vscale x 8 x i32> %2, <vscale x 8 x i32> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsleu_mask_vv_nxv8i32_nxv8i32:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v20, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e32, m4, ta, mu
-; CHECK-NEXT:    vmsleu.vv v20, v8, v12
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmsleu.vv v0, v8, v12
+; CHECK-NEXT:    vmsleu.vv v20, v12, v16, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v20
-; CHECK-NEXT:    vmsleu.vv v8, v12, v16, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 8 x i1> @llvm.riscv.vmsleu.nxv8i32(
@@ -814,10 +799,9 @@ declare <vscale x 1 x i1> @llvm.riscv.vmsleu.mask.nxv1i64(
 define <vscale x 1 x i1> @intrinsic_vmsleu_mask_vv_nxv1i64_nxv1i64(<vscale x 1 x i1> %0, <vscale x 1 x i64> %1, <vscale x 1 x i64> %2, <vscale x 1 x i64> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsleu_mask_vv_nxv1i64_nxv1i64:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e64, m1, ta, mu
-; CHECK-NEXT:    vmsleu.vv v8, v8, v9
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv.v.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e64, m1, ta, mu
+; CHECK-NEXT:    vmsleu.vv v0, v8, v9
 ; CHECK-NEXT:    vmsleu.vv v11, v9, v10, v0.t
 ; CHECK-NEXT:    vmv.v.v v0, v11
 ; CHECK-NEXT:    ret
@@ -866,12 +850,11 @@ declare <vscale x 2 x i1> @llvm.riscv.vmsleu.mask.nxv2i64(
 define <vscale x 2 x i1> @intrinsic_vmsleu_mask_vv_nxv2i64_nxv2i64(<vscale x 2 x i1> %0, <vscale x 2 x i64> %1, <vscale x 2 x i64> %2, <vscale x 2 x i64> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsleu_mask_vv_nxv2i64_nxv2i64:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v14, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e64, m2, ta, mu
-; CHECK-NEXT:    vmsleu.vv v14, v8, v10
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmsleu.vv v0, v8, v10
+; CHECK-NEXT:    vmsleu.vv v14, v10, v12, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v14
-; CHECK-NEXT:    vmsleu.vv v8, v10, v12, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 2 x i1> @llvm.riscv.vmsleu.nxv2i64(
@@ -918,12 +901,11 @@ declare <vscale x 4 x i1> @llvm.riscv.vmsleu.mask.nxv4i64(
 define <vscale x 4 x i1> @intrinsic_vmsleu_mask_vv_nxv4i64_nxv4i64(<vscale x 4 x i1> %0, <vscale x 4 x i64> %1, <vscale x 4 x i64> %2, <vscale x 4 x i64> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsleu_mask_vv_nxv4i64_nxv4i64:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v20, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e64, m4, ta, mu
-; CHECK-NEXT:    vmsleu.vv v20, v8, v12
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmsleu.vv v0, v8, v12
+; CHECK-NEXT:    vmsleu.vv v20, v12, v16, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v20
-; CHECK-NEXT:    vmsleu.vv v8, v12, v16, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 4 x i1> @llvm.riscv.vmsleu.nxv4i64(
diff --git a/llvm/test/CodeGen/RISCV/rvv/vmslt.ll b/llvm/test/CodeGen/RISCV/rvv/vmslt.ll
index 6783f7feb624..6c6e580b043d 100644
--- a/llvm/test/CodeGen/RISCV/rvv/vmslt.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/vmslt.ll
@@ -34,10 +34,9 @@ declare <vscale x 1 x i1> @llvm.riscv.vmslt.mask.nxv1i8(
 define <vscale x 1 x i1> @intrinsic_vmslt_mask_vv_nxv1i8_nxv1i8(<vscale x 1 x i1> %0, <vscale x 1 x i8> %1, <vscale x 1 x i8> %2, <vscale x 1 x i8> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmslt_mask_vv_nxv1i8_nxv1i8:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e8, mf8, ta, mu
-; CHECK-NEXT:    vmslt.vv v8, v8, v9
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv1r.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e8, mf8, ta, mu
+; CHECK-NEXT:    vmslt.vv v0, v8, v9
 ; CHECK-NEXT:    vmslt.vv v11, v9, v10, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v11
 ; CHECK-NEXT:    ret
@@ -86,10 +85,9 @@ declare <vscale x 2 x i1> @llvm.riscv.vmslt.mask.nxv2i8(
 define <vscale x 2 x i1> @intrinsic_vmslt_mask_vv_nxv2i8_nxv2i8(<vscale x 2 x i1> %0, <vscale x 2 x i8> %1, <vscale x 2 x i8> %2, <vscale x 2 x i8> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmslt_mask_vv_nxv2i8_nxv2i8:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e8, mf4, ta, mu
-; CHECK-NEXT:    vmslt.vv v8, v8, v9
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv1r.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e8, mf4, ta, mu
+; CHECK-NEXT:    vmslt.vv v0, v8, v9
 ; CHECK-NEXT:    vmslt.vv v11, v9, v10, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v11
 ; CHECK-NEXT:    ret
@@ -138,10 +136,9 @@ declare <vscale x 4 x i1> @llvm.riscv.vmslt.mask.nxv4i8(
 define <vscale x 4 x i1> @intrinsic_vmslt_mask_vv_nxv4i8_nxv4i8(<vscale x 4 x i1> %0, <vscale x 4 x i8> %1, <vscale x 4 x i8> %2, <vscale x 4 x i8> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmslt_mask_vv_nxv4i8_nxv4i8:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e8, mf2, ta, mu
-; CHECK-NEXT:    vmslt.vv v8, v8, v9
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv1r.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e8, mf2, ta, mu
+; CHECK-NEXT:    vmslt.vv v0, v8, v9
 ; CHECK-NEXT:    vmslt.vv v11, v9, v10, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v11
 ; CHECK-NEXT:    ret
@@ -190,10 +187,9 @@ declare <vscale x 8 x i1> @llvm.riscv.vmslt.mask.nxv8i8(
 define <vscale x 8 x i1> @intrinsic_vmslt_mask_vv_nxv8i8_nxv8i8(<vscale x 8 x i1> %0, <vscale x 8 x i8> %1, <vscale x 8 x i8> %2, <vscale x 8 x i8> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmslt_mask_vv_nxv8i8_nxv8i8:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e8, m1, ta, mu
-; CHECK-NEXT:    vmslt.vv v8, v8, v9
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv.v.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e8, m1, ta, mu
+; CHECK-NEXT:    vmslt.vv v0, v8, v9
 ; CHECK-NEXT:    vmslt.vv v11, v9, v10, v0.t
 ; CHECK-NEXT:    vmv.v.v v0, v11
 ; CHECK-NEXT:    ret
@@ -242,12 +238,11 @@ declare <vscale x 16 x i1> @llvm.riscv.vmslt.mask.nxv16i8(
 define <vscale x 16 x i1> @intrinsic_vmslt_mask_vv_nxv16i8_nxv16i8(<vscale x 16 x i1> %0, <vscale x 16 x i8> %1, <vscale x 16 x i8> %2, <vscale x 16 x i8> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmslt_mask_vv_nxv16i8_nxv16i8:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v14, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e8, m2, ta, mu
-; CHECK-NEXT:    vmslt.vv v14, v8, v10
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmslt.vv v0, v8, v10
+; CHECK-NEXT:    vmslt.vv v14, v10, v12, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v14
-; CHECK-NEXT:    vmslt.vv v8, v10, v12, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 16 x i1> @llvm.riscv.vmslt.nxv16i8(
@@ -294,12 +289,11 @@ declare <vscale x 32 x i1> @llvm.riscv.vmslt.mask.nxv32i8(
 define <vscale x 32 x i1> @intrinsic_vmslt_mask_vv_nxv32i8_nxv32i8(<vscale x 32 x i1> %0, <vscale x 32 x i8> %1, <vscale x 32 x i8> %2, <vscale x 32 x i8> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmslt_mask_vv_nxv32i8_nxv32i8:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v20, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e8, m4, ta, mu
-; CHECK-NEXT:    vmslt.vv v20, v8, v12
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmslt.vv v0, v8, v12
+; CHECK-NEXT:    vmslt.vv v20, v12, v16, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v20
-; CHECK-NEXT:    vmslt.vv v8, v12, v16, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 32 x i1> @llvm.riscv.vmslt.nxv32i8(
@@ -346,10 +340,9 @@ declare <vscale x 1 x i1> @llvm.riscv.vmslt.mask.nxv1i16(
 define <vscale x 1 x i1> @intrinsic_vmslt_mask_vv_nxv1i16_nxv1i16(<vscale x 1 x i1> %0, <vscale x 1 x i16> %1, <vscale x 1 x i16> %2, <vscale x 1 x i16> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmslt_mask_vv_nxv1i16_nxv1i16:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
-; CHECK-NEXT:    vmslt.vv v8, v8, v9
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv1r.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
+; CHECK-NEXT:    vmslt.vv v0, v8, v9
 ; CHECK-NEXT:    vmslt.vv v11, v9, v10, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v11
 ; CHECK-NEXT:    ret
@@ -398,10 +391,9 @@ declare <vscale x 2 x i1> @llvm.riscv.vmslt.mask.nxv2i16(
 define <vscale x 2 x i1> @intrinsic_vmslt_mask_vv_nxv2i16_nxv2i16(<vscale x 2 x i1> %0, <vscale x 2 x i16> %1, <vscale x 2 x i16> %2, <vscale x 2 x i16> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmslt_mask_vv_nxv2i16_nxv2i16:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e16, mf2, ta, mu
-; CHECK-NEXT:    vmslt.vv v8, v8, v9
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv1r.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e16, mf2, ta, mu
+; CHECK-NEXT:    vmslt.vv v0, v8, v9
 ; CHECK-NEXT:    vmslt.vv v11, v9, v10, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v11
 ; CHECK-NEXT:    ret
@@ -450,10 +442,9 @@ declare <vscale x 4 x i1> @llvm.riscv.vmslt.mask.nxv4i16(
 define <vscale x 4 x i1> @intrinsic_vmslt_mask_vv_nxv4i16_nxv4i16(<vscale x 4 x i1> %0, <vscale x 4 x i16> %1, <vscale x 4 x i16> %2, <vscale x 4 x i16> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmslt_mask_vv_nxv4i16_nxv4i16:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e16, m1, ta, mu
-; CHECK-NEXT:    vmslt.vv v8, v8, v9
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv.v.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e16, m1, ta, mu
+; CHECK-NEXT:    vmslt.vv v0, v8, v9
 ; CHECK-NEXT:    vmslt.vv v11, v9, v10, v0.t
 ; CHECK-NEXT:    vmv.v.v v0, v11
 ; CHECK-NEXT:    ret
@@ -502,12 +493,11 @@ declare <vscale x 8 x i1> @llvm.riscv.vmslt.mask.nxv8i16(
 define <vscale x 8 x i1> @intrinsic_vmslt_mask_vv_nxv8i16_nxv8i16(<vscale x 8 x i1> %0, <vscale x 8 x i16> %1, <vscale x 8 x i16> %2, <vscale x 8 x i16> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmslt_mask_vv_nxv8i16_nxv8i16:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v14, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e16, m2, ta, mu
-; CHECK-NEXT:    vmslt.vv v14, v8, v10
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmslt.vv v0, v8, v10
+; CHECK-NEXT:    vmslt.vv v14, v10, v12, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v14
-; CHECK-NEXT:    vmslt.vv v8, v10, v12, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 8 x i1> @llvm.riscv.vmslt.nxv8i16(
@@ -554,12 +544,11 @@ declare <vscale x 16 x i1> @llvm.riscv.vmslt.mask.nxv16i16(
 define <vscale x 16 x i1> @intrinsic_vmslt_mask_vv_nxv16i16_nxv16i16(<vscale x 16 x i1> %0, <vscale x 16 x i16> %1, <vscale x 16 x i16> %2, <vscale x 16 x i16> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmslt_mask_vv_nxv16i16_nxv16i16:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v20, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e16, m4, ta, mu
-; CHECK-NEXT:    vmslt.vv v20, v8, v12
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmslt.vv v0, v8, v12
+; CHECK-NEXT:    vmslt.vv v20, v12, v16, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v20
-; CHECK-NEXT:    vmslt.vv v8, v12, v16, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 16 x i1> @llvm.riscv.vmslt.nxv16i16(
@@ -606,10 +595,9 @@ declare <vscale x 1 x i1> @llvm.riscv.vmslt.mask.nxv1i32(
 define <vscale x 1 x i1> @intrinsic_vmslt_mask_vv_nxv1i32_nxv1i32(<vscale x 1 x i1> %0, <vscale x 1 x i32> %1, <vscale x 1 x i32> %2, <vscale x 1 x i32> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmslt_mask_vv_nxv1i32_nxv1i32:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e32, mf2, ta, mu
-; CHECK-NEXT:    vmslt.vv v8, v8, v9
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv1r.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e32, mf2, ta, mu
+; CHECK-NEXT:    vmslt.vv v0, v8, v9
 ; CHECK-NEXT:    vmslt.vv v11, v9, v10, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v11
 ; CHECK-NEXT:    ret
@@ -658,10 +646,9 @@ declare <vscale x 2 x i1> @llvm.riscv.vmslt.mask.nxv2i32(
 define <vscale x 2 x i1> @intrinsic_vmslt_mask_vv_nxv2i32_nxv2i32(<vscale x 2 x i1> %0, <vscale x 2 x i32> %1, <vscale x 2 x i32> %2, <vscale x 2 x i32> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmslt_mask_vv_nxv2i32_nxv2i32:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e32, m1, ta, mu
-; CHECK-NEXT:    vmslt.vv v8, v8, v9
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv.v.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e32, m1, ta, mu
+; CHECK-NEXT:    vmslt.vv v0, v8, v9
 ; CHECK-NEXT:    vmslt.vv v11, v9, v10, v0.t
 ; CHECK-NEXT:    vmv.v.v v0, v11
 ; CHECK-NEXT:    ret
@@ -710,12 +697,11 @@ declare <vscale x 4 x i1> @llvm.riscv.vmslt.mask.nxv4i32(
 define <vscale x 4 x i1> @intrinsic_vmslt_mask_vv_nxv4i32_nxv4i32(<vscale x 4 x i1> %0, <vscale x 4 x i32> %1, <vscale x 4 x i32> %2, <vscale x 4 x i32> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmslt_mask_vv_nxv4i32_nxv4i32:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v14, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e32, m2, ta, mu
-; CHECK-NEXT:    vmslt.vv v14, v8, v10
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmslt.vv v0, v8, v10
+; CHECK-NEXT:    vmslt.vv v14, v10, v12, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v14
-; CHECK-NEXT:    vmslt.vv v8, v10, v12, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 4 x i1> @llvm.riscv.vmslt.nxv4i32(
@@ -762,12 +748,11 @@ declare <vscale x 8 x i1> @llvm.riscv.vmslt.mask.nxv8i32(
 define <vscale x 8 x i1> @intrinsic_vmslt_mask_vv_nxv8i32_nxv8i32(<vscale x 8 x i1> %0, <vscale x 8 x i32> %1, <vscale x 8 x i32> %2, <vscale x 8 x i32> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmslt_mask_vv_nxv8i32_nxv8i32:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v20, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e32, m4, ta, mu
-; CHECK-NEXT:    vmslt.vv v20, v8, v12
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmslt.vv v0, v8, v12
+; CHECK-NEXT:    vmslt.vv v20, v12, v16, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v20
-; CHECK-NEXT:    vmslt.vv v8, v12, v16, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 8 x i1> @llvm.riscv.vmslt.nxv8i32(
@@ -814,10 +799,9 @@ declare <vscale x 1 x i1> @llvm.riscv.vmslt.mask.nxv1i64(
 define <vscale x 1 x i1> @intrinsic_vmslt_mask_vv_nxv1i64_nxv1i64(<vscale x 1 x i1> %0, <vscale x 1 x i64> %1, <vscale x 1 x i64> %2, <vscale x 1 x i64> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmslt_mask_vv_nxv1i64_nxv1i64:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e64, m1, ta, mu
-; CHECK-NEXT:    vmslt.vv v8, v8, v9
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv.v.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e64, m1, ta, mu
+; CHECK-NEXT:    vmslt.vv v0, v8, v9
 ; CHECK-NEXT:    vmslt.vv v11, v9, v10, v0.t
 ; CHECK-NEXT:    vmv.v.v v0, v11
 ; CHECK-NEXT:    ret
@@ -866,12 +850,11 @@ declare <vscale x 2 x i1> @llvm.riscv.vmslt.mask.nxv2i64(
 define <vscale x 2 x i1> @intrinsic_vmslt_mask_vv_nxv2i64_nxv2i64(<vscale x 2 x i1> %0, <vscale x 2 x i64> %1, <vscale x 2 x i64> %2, <vscale x 2 x i64> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmslt_mask_vv_nxv2i64_nxv2i64:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v14, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e64, m2, ta, mu
-; CHECK-NEXT:    vmslt.vv v14, v8, v10
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmslt.vv v0, v8, v10
+; CHECK-NEXT:    vmslt.vv v14, v10, v12, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v14
-; CHECK-NEXT:    vmslt.vv v8, v10, v12, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 2 x i1> @llvm.riscv.vmslt.nxv2i64(
@@ -918,12 +901,11 @@ declare <vscale x 4 x i1> @llvm.riscv.vmslt.mask.nxv4i64(
 define <vscale x 4 x i1> @intrinsic_vmslt_mask_vv_nxv4i64_nxv4i64(<vscale x 4 x i1> %0, <vscale x 4 x i64> %1, <vscale x 4 x i64> %2, <vscale x 4 x i64> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmslt_mask_vv_nxv4i64_nxv4i64:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v20, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e64, m4, ta, mu
-; CHECK-NEXT:    vmslt.vv v20, v8, v12
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmslt.vv v0, v8, v12
+; CHECK-NEXT:    vmslt.vv v20, v12, v16, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v20
-; CHECK-NEXT:    vmslt.vv v8, v12, v16, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 4 x i1> @llvm.riscv.vmslt.nxv4i64(
diff --git a/llvm/test/CodeGen/RISCV/rvv/vmsltu.ll b/llvm/test/CodeGen/RISCV/rvv/vmsltu.ll
index b082b735a020..76f3e449ab58 100644
--- a/llvm/test/CodeGen/RISCV/rvv/vmsltu.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/vmsltu.ll
@@ -34,10 +34,9 @@ declare <vscale x 1 x i1> @llvm.riscv.vmsltu.mask.nxv1i8(
 define <vscale x 1 x i1> @intrinsic_vmsltu_mask_vv_nxv1i8_nxv1i8(<vscale x 1 x i1> %0, <vscale x 1 x i8> %1, <vscale x 1 x i8> %2, <vscale x 1 x i8> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsltu_mask_vv_nxv1i8_nxv1i8:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e8, mf8, ta, mu
-; CHECK-NEXT:    vmsltu.vv v8, v8, v9
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv1r.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e8, mf8, ta, mu
+; CHECK-NEXT:    vmsltu.vv v0, v8, v9
 ; CHECK-NEXT:    vmsltu.vv v11, v9, v10, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v11
 ; CHECK-NEXT:    ret
@@ -86,10 +85,9 @@ declare <vscale x 2 x i1> @llvm.riscv.vmsltu.mask.nxv2i8(
 define <vscale x 2 x i1> @intrinsic_vmsltu_mask_vv_nxv2i8_nxv2i8(<vscale x 2 x i1> %0, <vscale x 2 x i8> %1, <vscale x 2 x i8> %2, <vscale x 2 x i8> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsltu_mask_vv_nxv2i8_nxv2i8:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e8, mf4, ta, mu
-; CHECK-NEXT:    vmsltu.vv v8, v8, v9
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv1r.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e8, mf4, ta, mu
+; CHECK-NEXT:    vmsltu.vv v0, v8, v9
 ; CHECK-NEXT:    vmsltu.vv v11, v9, v10, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v11
 ; CHECK-NEXT:    ret
@@ -138,10 +136,9 @@ declare <vscale x 4 x i1> @llvm.riscv.vmsltu.mask.nxv4i8(
 define <vscale x 4 x i1> @intrinsic_vmsltu_mask_vv_nxv4i8_nxv4i8(<vscale x 4 x i1> %0, <vscale x 4 x i8> %1, <vscale x 4 x i8> %2, <vscale x 4 x i8> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsltu_mask_vv_nxv4i8_nxv4i8:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e8, mf2, ta, mu
-; CHECK-NEXT:    vmsltu.vv v8, v8, v9
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv1r.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e8, mf2, ta, mu
+; CHECK-NEXT:    vmsltu.vv v0, v8, v9
 ; CHECK-NEXT:    vmsltu.vv v11, v9, v10, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v11
 ; CHECK-NEXT:    ret
@@ -190,10 +187,9 @@ declare <vscale x 8 x i1> @llvm.riscv.vmsltu.mask.nxv8i8(
 define <vscale x 8 x i1> @intrinsic_vmsltu_mask_vv_nxv8i8_nxv8i8(<vscale x 8 x i1> %0, <vscale x 8 x i8> %1, <vscale x 8 x i8> %2, <vscale x 8 x i8> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsltu_mask_vv_nxv8i8_nxv8i8:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e8, m1, ta, mu
-; CHECK-NEXT:    vmsltu.vv v8, v8, v9
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv.v.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e8, m1, ta, mu
+; CHECK-NEXT:    vmsltu.vv v0, v8, v9
 ; CHECK-NEXT:    vmsltu.vv v11, v9, v10, v0.t
 ; CHECK-NEXT:    vmv.v.v v0, v11
 ; CHECK-NEXT:    ret
@@ -242,12 +238,11 @@ declare <vscale x 16 x i1> @llvm.riscv.vmsltu.mask.nxv16i8(
 define <vscale x 16 x i1> @intrinsic_vmsltu_mask_vv_nxv16i8_nxv16i8(<vscale x 16 x i1> %0, <vscale x 16 x i8> %1, <vscale x 16 x i8> %2, <vscale x 16 x i8> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsltu_mask_vv_nxv16i8_nxv16i8:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v14, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e8, m2, ta, mu
-; CHECK-NEXT:    vmsltu.vv v14, v8, v10
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmsltu.vv v0, v8, v10
+; CHECK-NEXT:    vmsltu.vv v14, v10, v12, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v14
-; CHECK-NEXT:    vmsltu.vv v8, v10, v12, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 16 x i1> @llvm.riscv.vmsltu.nxv16i8(
@@ -294,12 +289,11 @@ declare <vscale x 32 x i1> @llvm.riscv.vmsltu.mask.nxv32i8(
 define <vscale x 32 x i1> @intrinsic_vmsltu_mask_vv_nxv32i8_nxv32i8(<vscale x 32 x i1> %0, <vscale x 32 x i8> %1, <vscale x 32 x i8> %2, <vscale x 32 x i8> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsltu_mask_vv_nxv32i8_nxv32i8:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v20, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e8, m4, ta, mu
-; CHECK-NEXT:    vmsltu.vv v20, v8, v12
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmsltu.vv v0, v8, v12
+; CHECK-NEXT:    vmsltu.vv v20, v12, v16, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v20
-; CHECK-NEXT:    vmsltu.vv v8, v12, v16, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 32 x i1> @llvm.riscv.vmsltu.nxv32i8(
@@ -346,10 +340,9 @@ declare <vscale x 1 x i1> @llvm.riscv.vmsltu.mask.nxv1i16(
 define <vscale x 1 x i1> @intrinsic_vmsltu_mask_vv_nxv1i16_nxv1i16(<vscale x 1 x i1> %0, <vscale x 1 x i16> %1, <vscale x 1 x i16> %2, <vscale x 1 x i16> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsltu_mask_vv_nxv1i16_nxv1i16:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
-; CHECK-NEXT:    vmsltu.vv v8, v8, v9
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv1r.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
+; CHECK-NEXT:    vmsltu.vv v0, v8, v9
 ; CHECK-NEXT:    vmsltu.vv v11, v9, v10, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v11
 ; CHECK-NEXT:    ret
@@ -398,10 +391,9 @@ declare <vscale x 2 x i1> @llvm.riscv.vmsltu.mask.nxv2i16(
 define <vscale x 2 x i1> @intrinsic_vmsltu_mask_vv_nxv2i16_nxv2i16(<vscale x 2 x i1> %0, <vscale x 2 x i16> %1, <vscale x 2 x i16> %2, <vscale x 2 x i16> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsltu_mask_vv_nxv2i16_nxv2i16:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e16, mf2, ta, mu
-; CHECK-NEXT:    vmsltu.vv v8, v8, v9
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv1r.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e16, mf2, ta, mu
+; CHECK-NEXT:    vmsltu.vv v0, v8, v9
 ; CHECK-NEXT:    vmsltu.vv v11, v9, v10, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v11
 ; CHECK-NEXT:    ret
@@ -450,10 +442,9 @@ declare <vscale x 4 x i1> @llvm.riscv.vmsltu.mask.nxv4i16(
 define <vscale x 4 x i1> @intrinsic_vmsltu_mask_vv_nxv4i16_nxv4i16(<vscale x 4 x i1> %0, <vscale x 4 x i16> %1, <vscale x 4 x i16> %2, <vscale x 4 x i16> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsltu_mask_vv_nxv4i16_nxv4i16:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e16, m1, ta, mu
-; CHECK-NEXT:    vmsltu.vv v8, v8, v9
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv.v.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e16, m1, ta, mu
+; CHECK-NEXT:    vmsltu.vv v0, v8, v9
 ; CHECK-NEXT:    vmsltu.vv v11, v9, v10, v0.t
 ; CHECK-NEXT:    vmv.v.v v0, v11
 ; CHECK-NEXT:    ret
@@ -502,12 +493,11 @@ declare <vscale x 8 x i1> @llvm.riscv.vmsltu.mask.nxv8i16(
 define <vscale x 8 x i1> @intrinsic_vmsltu_mask_vv_nxv8i16_nxv8i16(<vscale x 8 x i1> %0, <vscale x 8 x i16> %1, <vscale x 8 x i16> %2, <vscale x 8 x i16> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsltu_mask_vv_nxv8i16_nxv8i16:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v14, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e16, m2, ta, mu
-; CHECK-NEXT:    vmsltu.vv v14, v8, v10
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmsltu.vv v0, v8, v10
+; CHECK-NEXT:    vmsltu.vv v14, v10, v12, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v14
-; CHECK-NEXT:    vmsltu.vv v8, v10, v12, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 8 x i1> @llvm.riscv.vmsltu.nxv8i16(
@@ -554,12 +544,11 @@ declare <vscale x 16 x i1> @llvm.riscv.vmsltu.mask.nxv16i16(
 define <vscale x 16 x i1> @intrinsic_vmsltu_mask_vv_nxv16i16_nxv16i16(<vscale x 16 x i1> %0, <vscale x 16 x i16> %1, <vscale x 16 x i16> %2, <vscale x 16 x i16> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsltu_mask_vv_nxv16i16_nxv16i16:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v20, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e16, m4, ta, mu
-; CHECK-NEXT:    vmsltu.vv v20, v8, v12
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmsltu.vv v0, v8, v12
+; CHECK-NEXT:    vmsltu.vv v20, v12, v16, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v20
-; CHECK-NEXT:    vmsltu.vv v8, v12, v16, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 16 x i1> @llvm.riscv.vmsltu.nxv16i16(
@@ -606,10 +595,9 @@ declare <vscale x 1 x i1> @llvm.riscv.vmsltu.mask.nxv1i32(
 define <vscale x 1 x i1> @intrinsic_vmsltu_mask_vv_nxv1i32_nxv1i32(<vscale x 1 x i1> %0, <vscale x 1 x i32> %1, <vscale x 1 x i32> %2, <vscale x 1 x i32> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsltu_mask_vv_nxv1i32_nxv1i32:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e32, mf2, ta, mu
-; CHECK-NEXT:    vmsltu.vv v8, v8, v9
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv1r.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e32, mf2, ta, mu
+; CHECK-NEXT:    vmsltu.vv v0, v8, v9
 ; CHECK-NEXT:    vmsltu.vv v11, v9, v10, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v11
 ; CHECK-NEXT:    ret
@@ -658,10 +646,9 @@ declare <vscale x 2 x i1> @llvm.riscv.vmsltu.mask.nxv2i32(
 define <vscale x 2 x i1> @intrinsic_vmsltu_mask_vv_nxv2i32_nxv2i32(<vscale x 2 x i1> %0, <vscale x 2 x i32> %1, <vscale x 2 x i32> %2, <vscale x 2 x i32> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsltu_mask_vv_nxv2i32_nxv2i32:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e32, m1, ta, mu
-; CHECK-NEXT:    vmsltu.vv v8, v8, v9
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv.v.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e32, m1, ta, mu
+; CHECK-NEXT:    vmsltu.vv v0, v8, v9
 ; CHECK-NEXT:    vmsltu.vv v11, v9, v10, v0.t
 ; CHECK-NEXT:    vmv.v.v v0, v11
 ; CHECK-NEXT:    ret
@@ -710,12 +697,11 @@ declare <vscale x 4 x i1> @llvm.riscv.vmsltu.mask.nxv4i32(
 define <vscale x 4 x i1> @intrinsic_vmsltu_mask_vv_nxv4i32_nxv4i32(<vscale x 4 x i1> %0, <vscale x 4 x i32> %1, <vscale x 4 x i32> %2, <vscale x 4 x i32> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsltu_mask_vv_nxv4i32_nxv4i32:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v14, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e32, m2, ta, mu
-; CHECK-NEXT:    vmsltu.vv v14, v8, v10
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmsltu.vv v0, v8, v10
+; CHECK-NEXT:    vmsltu.vv v14, v10, v12, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v14
-; CHECK-NEXT:    vmsltu.vv v8, v10, v12, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 4 x i1> @llvm.riscv.vmsltu.nxv4i32(
@@ -762,12 +748,11 @@ declare <vscale x 8 x i1> @llvm.riscv.vmsltu.mask.nxv8i32(
 define <vscale x 8 x i1> @intrinsic_vmsltu_mask_vv_nxv8i32_nxv8i32(<vscale x 8 x i1> %0, <vscale x 8 x i32> %1, <vscale x 8 x i32> %2, <vscale x 8 x i32> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsltu_mask_vv_nxv8i32_nxv8i32:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v20, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e32, m4, ta, mu
-; CHECK-NEXT:    vmsltu.vv v20, v8, v12
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmsltu.vv v0, v8, v12
+; CHECK-NEXT:    vmsltu.vv v20, v12, v16, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v20
-; CHECK-NEXT:    vmsltu.vv v8, v12, v16, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 8 x i1> @llvm.riscv.vmsltu.nxv8i32(
@@ -814,10 +799,9 @@ declare <vscale x 1 x i1> @llvm.riscv.vmsltu.mask.nxv1i64(
 define <vscale x 1 x i1> @intrinsic_vmsltu_mask_vv_nxv1i64_nxv1i64(<vscale x 1 x i1> %0, <vscale x 1 x i64> %1, <vscale x 1 x i64> %2, <vscale x 1 x i64> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsltu_mask_vv_nxv1i64_nxv1i64:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e64, m1, ta, mu
-; CHECK-NEXT:    vmsltu.vv v8, v8, v9
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv.v.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e64, m1, ta, mu
+; CHECK-NEXT:    vmsltu.vv v0, v8, v9
 ; CHECK-NEXT:    vmsltu.vv v11, v9, v10, v0.t
 ; CHECK-NEXT:    vmv.v.v v0, v11
 ; CHECK-NEXT:    ret
@@ -866,12 +850,11 @@ declare <vscale x 2 x i1> @llvm.riscv.vmsltu.mask.nxv2i64(
 define <vscale x 2 x i1> @intrinsic_vmsltu_mask_vv_nxv2i64_nxv2i64(<vscale x 2 x i1> %0, <vscale x 2 x i64> %1, <vscale x 2 x i64> %2, <vscale x 2 x i64> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsltu_mask_vv_nxv2i64_nxv2i64:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v14, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e64, m2, ta, mu
-; CHECK-NEXT:    vmsltu.vv v14, v8, v10
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmsltu.vv v0, v8, v10
+; CHECK-NEXT:    vmsltu.vv v14, v10, v12, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v14
-; CHECK-NEXT:    vmsltu.vv v8, v10, v12, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 2 x i1> @llvm.riscv.vmsltu.nxv2i64(
@@ -918,12 +901,11 @@ declare <vscale x 4 x i1> @llvm.riscv.vmsltu.mask.nxv4i64(
 define <vscale x 4 x i1> @intrinsic_vmsltu_mask_vv_nxv4i64_nxv4i64(<vscale x 4 x i1> %0, <vscale x 4 x i64> %1, <vscale x 4 x i64> %2, <vscale x 4 x i64> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsltu_mask_vv_nxv4i64_nxv4i64:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v20, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e64, m4, ta, mu
-; CHECK-NEXT:    vmsltu.vv v20, v8, v12
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmsltu.vv v0, v8, v12
+; CHECK-NEXT:    vmsltu.vv v20, v12, v16, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v20
-; CHECK-NEXT:    vmsltu.vv v8, v12, v16, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 4 x i1> @llvm.riscv.vmsltu.nxv4i64(
diff --git a/llvm/test/CodeGen/RISCV/rvv/vmsne.ll b/llvm/test/CodeGen/RISCV/rvv/vmsne.ll
index bb4575e5d72c..161c1bc4314f 100644
--- a/llvm/test/CodeGen/RISCV/rvv/vmsne.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/vmsne.ll
@@ -34,10 +34,9 @@ declare <vscale x 1 x i1> @llvm.riscv.vmsne.mask.nxv1i8(
 define <vscale x 1 x i1> @intrinsic_vmsne_mask_vv_nxv1i8_nxv1i8(<vscale x 1 x i1> %0, <vscale x 1 x i8> %1, <vscale x 1 x i8> %2, <vscale x 1 x i8> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsne_mask_vv_nxv1i8_nxv1i8:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e8, mf8, ta, mu
-; CHECK-NEXT:    vmsne.vv v8, v8, v9
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv1r.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e8, mf8, ta, mu
+; CHECK-NEXT:    vmsne.vv v0, v8, v9
 ; CHECK-NEXT:    vmsne.vv v11, v9, v10, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v11
 ; CHECK-NEXT:    ret
@@ -86,10 +85,9 @@ declare <vscale x 2 x i1> @llvm.riscv.vmsne.mask.nxv2i8(
 define <vscale x 2 x i1> @intrinsic_vmsne_mask_vv_nxv2i8_nxv2i8(<vscale x 2 x i1> %0, <vscale x 2 x i8> %1, <vscale x 2 x i8> %2, <vscale x 2 x i8> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsne_mask_vv_nxv2i8_nxv2i8:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e8, mf4, ta, mu
-; CHECK-NEXT:    vmsne.vv v8, v8, v9
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv1r.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e8, mf4, ta, mu
+; CHECK-NEXT:    vmsne.vv v0, v8, v9
 ; CHECK-NEXT:    vmsne.vv v11, v9, v10, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v11
 ; CHECK-NEXT:    ret
@@ -138,10 +136,9 @@ declare <vscale x 4 x i1> @llvm.riscv.vmsne.mask.nxv4i8(
 define <vscale x 4 x i1> @intrinsic_vmsne_mask_vv_nxv4i8_nxv4i8(<vscale x 4 x i1> %0, <vscale x 4 x i8> %1, <vscale x 4 x i8> %2, <vscale x 4 x i8> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsne_mask_vv_nxv4i8_nxv4i8:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e8, mf2, ta, mu
-; CHECK-NEXT:    vmsne.vv v8, v8, v9
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv1r.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e8, mf2, ta, mu
+; CHECK-NEXT:    vmsne.vv v0, v8, v9
 ; CHECK-NEXT:    vmsne.vv v11, v9, v10, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v11
 ; CHECK-NEXT:    ret
@@ -190,10 +187,9 @@ declare <vscale x 8 x i1> @llvm.riscv.vmsne.mask.nxv8i8(
 define <vscale x 8 x i1> @intrinsic_vmsne_mask_vv_nxv8i8_nxv8i8(<vscale x 8 x i1> %0, <vscale x 8 x i8> %1, <vscale x 8 x i8> %2, <vscale x 8 x i8> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsne_mask_vv_nxv8i8_nxv8i8:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e8, m1, ta, mu
-; CHECK-NEXT:    vmsne.vv v8, v8, v9
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv.v.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e8, m1, ta, mu
+; CHECK-NEXT:    vmsne.vv v0, v8, v9
 ; CHECK-NEXT:    vmsne.vv v11, v9, v10, v0.t
 ; CHECK-NEXT:    vmv.v.v v0, v11
 ; CHECK-NEXT:    ret
@@ -242,12 +238,11 @@ declare <vscale x 16 x i1> @llvm.riscv.vmsne.mask.nxv16i8(
 define <vscale x 16 x i1> @intrinsic_vmsne_mask_vv_nxv16i8_nxv16i8(<vscale x 16 x i1> %0, <vscale x 16 x i8> %1, <vscale x 16 x i8> %2, <vscale x 16 x i8> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsne_mask_vv_nxv16i8_nxv16i8:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v14, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e8, m2, ta, mu
-; CHECK-NEXT:    vmsne.vv v14, v8, v10
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmsne.vv v0, v8, v10
+; CHECK-NEXT:    vmsne.vv v14, v10, v12, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v14
-; CHECK-NEXT:    vmsne.vv v8, v10, v12, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 16 x i1> @llvm.riscv.vmsne.nxv16i8(
@@ -294,12 +289,11 @@ declare <vscale x 32 x i1> @llvm.riscv.vmsne.mask.nxv32i8(
 define <vscale x 32 x i1> @intrinsic_vmsne_mask_vv_nxv32i8_nxv32i8(<vscale x 32 x i1> %0, <vscale x 32 x i8> %1, <vscale x 32 x i8> %2, <vscale x 32 x i8> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsne_mask_vv_nxv32i8_nxv32i8:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v20, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e8, m4, ta, mu
-; CHECK-NEXT:    vmsne.vv v20, v8, v12
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmsne.vv v0, v8, v12
+; CHECK-NEXT:    vmsne.vv v20, v12, v16, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v20
-; CHECK-NEXT:    vmsne.vv v8, v12, v16, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 32 x i1> @llvm.riscv.vmsne.nxv32i8(
@@ -346,10 +340,9 @@ declare <vscale x 1 x i1> @llvm.riscv.vmsne.mask.nxv1i16(
 define <vscale x 1 x i1> @intrinsic_vmsne_mask_vv_nxv1i16_nxv1i16(<vscale x 1 x i1> %0, <vscale x 1 x i16> %1, <vscale x 1 x i16> %2, <vscale x 1 x i16> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsne_mask_vv_nxv1i16_nxv1i16:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
-; CHECK-NEXT:    vmsne.vv v8, v8, v9
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv1r.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e16, mf4, ta, mu
+; CHECK-NEXT:    vmsne.vv v0, v8, v9
 ; CHECK-NEXT:    vmsne.vv v11, v9, v10, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v11
 ; CHECK-NEXT:    ret
@@ -398,10 +391,9 @@ declare <vscale x 2 x i1> @llvm.riscv.vmsne.mask.nxv2i16(
 define <vscale x 2 x i1> @intrinsic_vmsne_mask_vv_nxv2i16_nxv2i16(<vscale x 2 x i1> %0, <vscale x 2 x i16> %1, <vscale x 2 x i16> %2, <vscale x 2 x i16> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsne_mask_vv_nxv2i16_nxv2i16:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e16, mf2, ta, mu
-; CHECK-NEXT:    vmsne.vv v8, v8, v9
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv1r.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e16, mf2, ta, mu
+; CHECK-NEXT:    vmsne.vv v0, v8, v9
 ; CHECK-NEXT:    vmsne.vv v11, v9, v10, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v11
 ; CHECK-NEXT:    ret
@@ -450,10 +442,9 @@ declare <vscale x 4 x i1> @llvm.riscv.vmsne.mask.nxv4i16(
 define <vscale x 4 x i1> @intrinsic_vmsne_mask_vv_nxv4i16_nxv4i16(<vscale x 4 x i1> %0, <vscale x 4 x i16> %1, <vscale x 4 x i16> %2, <vscale x 4 x i16> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsne_mask_vv_nxv4i16_nxv4i16:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e16, m1, ta, mu
-; CHECK-NEXT:    vmsne.vv v8, v8, v9
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv.v.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e16, m1, ta, mu
+; CHECK-NEXT:    vmsne.vv v0, v8, v9
 ; CHECK-NEXT:    vmsne.vv v11, v9, v10, v0.t
 ; CHECK-NEXT:    vmv.v.v v0, v11
 ; CHECK-NEXT:    ret
@@ -502,12 +493,11 @@ declare <vscale x 8 x i1> @llvm.riscv.vmsne.mask.nxv8i16(
 define <vscale x 8 x i1> @intrinsic_vmsne_mask_vv_nxv8i16_nxv8i16(<vscale x 8 x i1> %0, <vscale x 8 x i16> %1, <vscale x 8 x i16> %2, <vscale x 8 x i16> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsne_mask_vv_nxv8i16_nxv8i16:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v14, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e16, m2, ta, mu
-; CHECK-NEXT:    vmsne.vv v14, v8, v10
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmsne.vv v0, v8, v10
+; CHECK-NEXT:    vmsne.vv v14, v10, v12, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v14
-; CHECK-NEXT:    vmsne.vv v8, v10, v12, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 8 x i1> @llvm.riscv.vmsne.nxv8i16(
@@ -554,12 +544,11 @@ declare <vscale x 16 x i1> @llvm.riscv.vmsne.mask.nxv16i16(
 define <vscale x 16 x i1> @intrinsic_vmsne_mask_vv_nxv16i16_nxv16i16(<vscale x 16 x i1> %0, <vscale x 16 x i16> %1, <vscale x 16 x i16> %2, <vscale x 16 x i16> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsne_mask_vv_nxv16i16_nxv16i16:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v20, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e16, m4, ta, mu
-; CHECK-NEXT:    vmsne.vv v20, v8, v12
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmsne.vv v0, v8, v12
+; CHECK-NEXT:    vmsne.vv v20, v12, v16, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v20
-; CHECK-NEXT:    vmsne.vv v8, v12, v16, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 16 x i1> @llvm.riscv.vmsne.nxv16i16(
@@ -606,10 +595,9 @@ declare <vscale x 1 x i1> @llvm.riscv.vmsne.mask.nxv1i32(
 define <vscale x 1 x i1> @intrinsic_vmsne_mask_vv_nxv1i32_nxv1i32(<vscale x 1 x i1> %0, <vscale x 1 x i32> %1, <vscale x 1 x i32> %2, <vscale x 1 x i32> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsne_mask_vv_nxv1i32_nxv1i32:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e32, mf2, ta, mu
-; CHECK-NEXT:    vmsne.vv v8, v8, v9
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv1r.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e32, mf2, ta, mu
+; CHECK-NEXT:    vmsne.vv v0, v8, v9
 ; CHECK-NEXT:    vmsne.vv v11, v9, v10, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v11
 ; CHECK-NEXT:    ret
@@ -658,10 +646,9 @@ declare <vscale x 2 x i1> @llvm.riscv.vmsne.mask.nxv2i32(
 define <vscale x 2 x i1> @intrinsic_vmsne_mask_vv_nxv2i32_nxv2i32(<vscale x 2 x i1> %0, <vscale x 2 x i32> %1, <vscale x 2 x i32> %2, <vscale x 2 x i32> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsne_mask_vv_nxv2i32_nxv2i32:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e32, m1, ta, mu
-; CHECK-NEXT:    vmsne.vv v8, v8, v9
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv.v.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e32, m1, ta, mu
+; CHECK-NEXT:    vmsne.vv v0, v8, v9
 ; CHECK-NEXT:    vmsne.vv v11, v9, v10, v0.t
 ; CHECK-NEXT:    vmv.v.v v0, v11
 ; CHECK-NEXT:    ret
@@ -710,12 +697,11 @@ declare <vscale x 4 x i1> @llvm.riscv.vmsne.mask.nxv4i32(
 define <vscale x 4 x i1> @intrinsic_vmsne_mask_vv_nxv4i32_nxv4i32(<vscale x 4 x i1> %0, <vscale x 4 x i32> %1, <vscale x 4 x i32> %2, <vscale x 4 x i32> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsne_mask_vv_nxv4i32_nxv4i32:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v14, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e32, m2, ta, mu
-; CHECK-NEXT:    vmsne.vv v14, v8, v10
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmsne.vv v0, v8, v10
+; CHECK-NEXT:    vmsne.vv v14, v10, v12, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v14
-; CHECK-NEXT:    vmsne.vv v8, v10, v12, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 4 x i1> @llvm.riscv.vmsne.nxv4i32(
@@ -762,12 +748,11 @@ declare <vscale x 8 x i1> @llvm.riscv.vmsne.mask.nxv8i32(
 define <vscale x 8 x i1> @intrinsic_vmsne_mask_vv_nxv8i32_nxv8i32(<vscale x 8 x i1> %0, <vscale x 8 x i32> %1, <vscale x 8 x i32> %2, <vscale x 8 x i32> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsne_mask_vv_nxv8i32_nxv8i32:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v20, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e32, m4, ta, mu
-; CHECK-NEXT:    vmsne.vv v20, v8, v12
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmsne.vv v0, v8, v12
+; CHECK-NEXT:    vmsne.vv v20, v12, v16, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v20
-; CHECK-NEXT:    vmsne.vv v8, v12, v16, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 8 x i1> @llvm.riscv.vmsne.nxv8i32(
@@ -814,10 +799,9 @@ declare <vscale x 1 x i1> @llvm.riscv.vmsne.mask.nxv1i64(
 define <vscale x 1 x i1> @intrinsic_vmsne_mask_vv_nxv1i64_nxv1i64(<vscale x 1 x i1> %0, <vscale x 1 x i64> %1, <vscale x 1 x i64> %2, <vscale x 1 x i64> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsne_mask_vv_nxv1i64_nxv1i64:
 ; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    vsetvli zero, a0, e64, m1, ta, mu
-; CHECK-NEXT:    vmsne.vv v8, v8, v9
 ; CHECK-NEXT:    vmv1r.v v11, v0
-; CHECK-NEXT:    vmv.v.v v0, v8
+; CHECK-NEXT:    vsetvli zero, a0, e64, m1, ta, mu
+; CHECK-NEXT:    vmsne.vv v0, v8, v9
 ; CHECK-NEXT:    vmsne.vv v11, v9, v10, v0.t
 ; CHECK-NEXT:    vmv.v.v v0, v11
 ; CHECK-NEXT:    ret
@@ -866,12 +850,11 @@ declare <vscale x 2 x i1> @llvm.riscv.vmsne.mask.nxv2i64(
 define <vscale x 2 x i1> @intrinsic_vmsne_mask_vv_nxv2i64_nxv2i64(<vscale x 2 x i1> %0, <vscale x 2 x i64> %1, <vscale x 2 x i64> %2, <vscale x 2 x i64> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsne_mask_vv_nxv2i64_nxv2i64:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v14, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e64, m2, ta, mu
-; CHECK-NEXT:    vmsne.vv v14, v8, v10
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmsne.vv v0, v8, v10
+; CHECK-NEXT:    vmsne.vv v14, v10, v12, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v14
-; CHECK-NEXT:    vmsne.vv v8, v10, v12, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 2 x i1> @llvm.riscv.vmsne.nxv2i64(
@@ -918,12 +901,11 @@ declare <vscale x 4 x i1> @llvm.riscv.vmsne.mask.nxv4i64(
 define <vscale x 4 x i1> @intrinsic_vmsne_mask_vv_nxv4i64_nxv4i64(<vscale x 4 x i1> %0, <vscale x 4 x i64> %1, <vscale x 4 x i64> %2, <vscale x 4 x i64> %3, iXLen %4) nounwind {
 ; CHECK-LABEL: intrinsic_vmsne_mask_vv_nxv4i64_nxv4i64:
 ; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vmv1r.v v20, v0
 ; CHECK-NEXT:    vsetvli zero, a0, e64, m4, ta, mu
-; CHECK-NEXT:    vmsne.vv v20, v8, v12
-; CHECK-NEXT:    vmv1r.v v8, v0
+; CHECK-NEXT:    vmsne.vv v0, v8, v12
+; CHECK-NEXT:    vmsne.vv v20, v12, v16, v0.t
 ; CHECK-NEXT:    vmv1r.v v0, v20
-; CHECK-NEXT:    vmsne.vv v8, v12, v16, v0.t
-; CHECK-NEXT:    vmv1r.v v0, v8
 ; CHECK-NEXT:    ret
 entry:
   %mask = call <vscale x 4 x i1> @llvm.riscv.vmsne.nxv4i64(
diff --git a/llvm/test/CodeGen/RISCV/rvv/vreductions-int-vp.ll b/llvm/test/CodeGen/RISCV/rvv/vreductions-int-vp.ll
index 618e169e1f96..7bcf37b1af3c 100644
--- a/llvm/test/CodeGen/RISCV/rvv/vreductions-int-vp.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/vreductions-int-vp.ll
@@ -841,24 +841,14 @@ define signext i32 @vpreduce_add_nxv1i32(i32 signext %s, <vscale x 1 x i32> %v,
 declare i32 @llvm.vp.reduce.umax.nxv1i32(i32, <vscale x 1 x i32>, <vscale x 1 x i1>, i32)
 
 define signext i32 @vpreduce_umax_nxv1i32(i32 signext %s, <vscale x 1 x i32> %v, <vscale x 1 x i1> %m, i32 zeroext %evl) {
-; RV32-LABEL: vpreduce_umax_nxv1i32:
-; RV32:       # %bb.0:
-; RV32-NEXT:    vsetivli zero, 1, e32, m1, ta, ma
-; RV32-NEXT:    vmv.s.x v9, a0
-; RV32-NEXT:    vsetvli zero, a1, e32, mf2, ta, ma
-; RV32-NEXT:    vredmaxu.vs v9, v8, v9, v0.t
-; RV32-NEXT:    vmv.x.s a0, v9
-; RV32-NEXT:    ret
-;
-; RV64-LABEL: vpreduce_umax_nxv1i32:
-; RV64:       # %bb.0:
-; RV64-NEXT:    andi a0, a0, -1
-; RV64-NEXT:    vsetivli zero, 1, e32, m1, ta, ma
-; RV64-NEXT:    vmv.s.x v9, a0
-; RV64-NEXT:    vsetvli zero, a1, e32, mf2, ta, ma
-; RV64-NEXT:    vredmaxu.vs v9, v8, v9, v0.t
-; RV64-NEXT:    vmv.x.s a0, v9
-; RV64-NEXT:    ret
+; CHECK-LABEL: vpreduce_umax_nxv1i32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli zero, 1, e32, m1, ta, ma
+; CHECK-NEXT:    vmv.s.x v9, a0
+; CHECK-NEXT:    vsetvli zero, a1, e32, mf2, ta, ma
+; CHECK-NEXT:    vredmaxu.vs v9, v8, v9, v0.t
+; CHECK-NEXT:    vmv.x.s a0, v9
+; CHECK-NEXT:    ret
   %r = call i32 @llvm.vp.reduce.umax.nxv1i32(i32 %s, <vscale x 1 x i32> %v, <vscale x 1 x i1> %m, i32 %evl)
   ret i32 %r
 }
@@ -881,24 +871,14 @@ define signext i32 @vpreduce_smax_nxv1i32(i32 signext %s, <vscale x 1 x i32> %v,
 declare i32 @llvm.vp.reduce.umin.nxv1i32(i32, <vscale x 1 x i32>, <vscale x 1 x i1>, i32)
 
 define signext i32 @vpreduce_umin_nxv1i32(i32 signext %s, <vscale x 1 x i32> %v, <vscale x 1 x i1> %m, i32 zeroext %evl) {
-; RV32-LABEL: vpreduce_umin_nxv1i32:
-; RV32:       # %bb.0:
-; RV32-NEXT:    vsetivli zero, 1, e32, m1, ta, ma
-; RV32-NEXT:    vmv.s.x v9, a0
-; RV32-NEXT:    vsetvli zero, a1, e32, mf2, ta, ma
-; RV32-NEXT:    vredminu.vs v9, v8, v9, v0.t
-; RV32-NEXT:    vmv.x.s a0, v9
-; RV32-NEXT:    ret
-;
-; RV64-LABEL: vpreduce_umin_nxv1i32:
-; RV64:       # %bb.0:
-; RV64-NEXT:    andi a0, a0, -1
-; RV64-NEXT:    vsetivli zero, 1, e32, m1, ta, ma
-; RV64-NEXT:    vmv.s.x v9, a0
-; RV64-NEXT:    vsetvli zero, a1, e32, mf2, ta, ma
-; RV64-NEXT:    vredminu.vs v9, v8, v9, v0.t
-; RV64-NEXT:    vmv.x.s a0, v9
-; RV64-NEXT:    ret
+; CHECK-LABEL: vpreduce_umin_nxv1i32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli zero, 1, e32, m1, ta, ma
+; CHECK-NEXT:    vmv.s.x v9, a0
+; CHECK-NEXT:    vsetvli zero, a1, e32, mf2, ta, ma
+; CHECK-NEXT:    vredminu.vs v9, v8, v9, v0.t
+; CHECK-NEXT:    vmv.x.s a0, v9
+; CHECK-NEXT:    ret
   %r = call i32 @llvm.vp.reduce.umin.nxv1i32(i32 %s, <vscale x 1 x i32> %v, <vscale x 1 x i1> %m, i32 %evl)
   ret i32 %r
 }
@@ -981,24 +961,14 @@ define signext i32 @vpreduce_add_nxv2i32(i32 signext %s, <vscale x 2 x i32> %v,
 declare i32 @llvm.vp.reduce.umax.nxv2i32(i32, <vscale x 2 x i32>, <vscale x 2 x i1>, i32)
 
 define signext i32 @vpreduce_umax_nxv2i32(i32 signext %s, <vscale x 2 x i32> %v, <vscale x 2 x i1> %m, i32 zeroext %evl) {
-; RV32-LABEL: vpreduce_umax_nxv2i32:
-; RV32:       # %bb.0:
-; RV32-NEXT:    vsetivli zero, 1, e32, m1, ta, ma
-; RV32-NEXT:    vmv.s.x v9, a0
-; RV32-NEXT:    vsetvli zero, a1, e32, m1, ta, ma
-; RV32-NEXT:    vredmaxu.vs v9, v8, v9, v0.t
-; RV32-NEXT:    vmv.x.s a0, v9
-; RV32-NEXT:    ret
-;
-; RV64-LABEL: vpreduce_umax_nxv2i32:
-; RV64:       # %bb.0:
-; RV64-NEXT:    andi a0, a0, -1
-; RV64-NEXT:    vsetivli zero, 1, e32, m1, ta, ma
-; RV64-NEXT:    vmv.s.x v9, a0
-; RV64-NEXT:    vsetvli zero, a1, e32, m1, ta, ma
-; RV64-NEXT:    vredmaxu.vs v9, v8, v9, v0.t
-; RV64-NEXT:    vmv.x.s a0, v9
-; RV64-NEXT:    ret
+; CHECK-LABEL: vpreduce_umax_nxv2i32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli zero, 1, e32, m1, ta, ma
+; CHECK-NEXT:    vmv.s.x v9, a0
+; CHECK-NEXT:    vsetvli zero, a1, e32, m1, ta, ma
+; CHECK-NEXT:    vredmaxu.vs v9, v8, v9, v0.t
+; CHECK-NEXT:    vmv.x.s a0, v9
+; CHECK-NEXT:    ret
   %r = call i32 @llvm.vp.reduce.umax.nxv2i32(i32 %s, <vscale x 2 x i32> %v, <vscale x 2 x i1> %m, i32 %evl)
   ret i32 %r
 }
@@ -1021,24 +991,14 @@ define signext i32 @vpreduce_smax_nxv2i32(i32 signext %s, <vscale x 2 x i32> %v,
 declare i32 @llvm.vp.reduce.umin.nxv2i32(i32, <vscale x 2 x i32>, <vscale x 2 x i1>, i32)
 
 define signext i32 @vpreduce_umin_nxv2i32(i32 signext %s, <vscale x 2 x i32> %v, <vscale x 2 x i1> %m, i32 zeroext %evl) {
-; RV32-LABEL: vpreduce_umin_nxv2i32:
-; RV32:       # %bb.0:
-; RV32-NEXT:    vsetivli zero, 1, e32, m1, ta, ma
-; RV32-NEXT:    vmv.s.x v9, a0
-; RV32-NEXT:    vsetvli zero, a1, e32, m1, ta, ma
-; RV32-NEXT:    vredminu.vs v9, v8, v9, v0.t
-; RV32-NEXT:    vmv.x.s a0, v9
-; RV32-NEXT:    ret
-;
-; RV64-LABEL: vpreduce_umin_nxv2i32:
-; RV64:       # %bb.0:
-; RV64-NEXT:    andi a0, a0, -1
-; RV64-NEXT:    vsetivli zero, 1, e32, m1, ta, ma
-; RV64-NEXT:    vmv.s.x v9, a0
-; RV64-NEXT:    vsetvli zero, a1, e32, m1, ta, ma
-; RV64-NEXT:    vredminu.vs v9, v8, v9, v0.t
-; RV64-NEXT:    vmv.x.s a0, v9
-; RV64-NEXT:    ret
+; CHECK-LABEL: vpreduce_umin_nxv2i32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli zero, 1, e32, m1, ta, ma
+; CHECK-NEXT:    vmv.s.x v9, a0
+; CHECK-NEXT:    vsetvli zero, a1, e32, m1, ta, ma
+; CHECK-NEXT:    vredminu.vs v9, v8, v9, v0.t
+; CHECK-NEXT:    vmv.x.s a0, v9
+; CHECK-NEXT:    ret
   %r = call i32 @llvm.vp.reduce.umin.nxv2i32(i32 %s, <vscale x 2 x i32> %v, <vscale x 2 x i1> %m, i32 %evl)
   ret i32 %r
 }
@@ -1121,24 +1081,14 @@ define signext i32 @vpreduce_add_nxv4i32(i32 signext %s, <vscale x 4 x i32> %v,
 declare i32 @llvm.vp.reduce.umax.nxv4i32(i32, <vscale x 4 x i32>, <vscale x 4 x i1>, i32)
 
 define signext i32 @vpreduce_umax_nxv4i32(i32 signext %s, <vscale x 4 x i32> %v, <vscale x 4 x i1> %m, i32 zeroext %evl) {
-; RV32-LABEL: vpreduce_umax_nxv4i32:
-; RV32:       # %bb.0:
-; RV32-NEXT:    vsetivli zero, 1, e32, m1, ta, ma
-; RV32-NEXT:    vmv.s.x v10, a0
-; RV32-NEXT:    vsetvli zero, a1, e32, m2, ta, ma
-; RV32-NEXT:    vredmaxu.vs v10, v8, v10, v0.t
-; RV32-NEXT:    vmv.x.s a0, v10
-; RV32-NEXT:    ret
-;
-; RV64-LABEL: vpreduce_umax_nxv4i32:
-; RV64:       # %bb.0:
-; RV64-NEXT:    andi a0, a0, -1
-; RV64-NEXT:    vsetivli zero, 1, e32, m1, ta, ma
-; RV64-NEXT:    vmv.s.x v10, a0
-; RV64-NEXT:    vsetvli zero, a1, e32, m2, ta, ma
-; RV64-NEXT:    vredmaxu.vs v10, v8, v10, v0.t
-; RV64-NEXT:    vmv.x.s a0, v10
-; RV64-NEXT:    ret
+; CHECK-LABEL: vpreduce_umax_nxv4i32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli zero, 1, e32, m1, ta, ma
+; CHECK-NEXT:    vmv.s.x v10, a0
+; CHECK-NEXT:    vsetvli zero, a1, e32, m2, ta, ma
+; CHECK-NEXT:    vredmaxu.vs v10, v8, v10, v0.t
+; CHECK-NEXT:    vmv.x.s a0, v10
+; CHECK-NEXT:    ret
   %r = call i32 @llvm.vp.reduce.umax.nxv4i32(i32 %s, <vscale x 4 x i32> %v, <vscale x 4 x i1> %m, i32 %evl)
   ret i32 %r
 }
@@ -1146,56 +1096,30 @@ define signext i32 @vpreduce_umax_nxv4i32(i32 signext %s, <vscale x 4 x i32> %v,
 declare i32 @llvm.vp.reduce.umax.nxv32i32(i32, <vscale x 32 x i32>, <vscale x 32 x i1>, i32)
 
 define signext i32 @vpreduce_umax_nxv32i32(i32 signext %s, <vscale x 32 x i32> %v, <vscale x 32 x i1> %m, i32 zeroext %evl) {
-; RV32-LABEL: vpreduce_umax_nxv32i32:
-; RV32:       # %bb.0:
-; RV32-NEXT:    csrr a3, vlenb
-; RV32-NEXT:    srli a2, a3, 2
-; RV32-NEXT:    vsetvli a4, zero, e8, mf2, ta, ma
-; RV32-NEXT:    vslidedown.vx v24, v0, a2
-; RV32-NEXT:    slli a3, a3, 1
-; RV32-NEXT:    sub a2, a1, a3
-; RV32-NEXT:    sltu a4, a1, a2
-; RV32-NEXT:    addi a4, a4, -1
-; RV32-NEXT:    and a2, a4, a2
-; RV32-NEXT:    bltu a1, a3, .LBB67_2
-; RV32-NEXT:  # %bb.1:
-; RV32-NEXT:    mv a1, a3
-; RV32-NEXT:  .LBB67_2:
-; RV32-NEXT:    vsetvli zero, zero, e32, m2, ta, ma
-; RV32-NEXT:    vmv.s.x v25, a0
-; RV32-NEXT:    vsetvli zero, a1, e32, m8, ta, ma
-; RV32-NEXT:    vredmaxu.vs v25, v8, v25, v0.t
-; RV32-NEXT:    vsetvli zero, a2, e32, m8, ta, ma
-; RV32-NEXT:    vmv1r.v v0, v24
-; RV32-NEXT:    vredmaxu.vs v25, v16, v25, v0.t
-; RV32-NEXT:    vmv.x.s a0, v25
-; RV32-NEXT:    ret
-;
-; RV64-LABEL: vpreduce_umax_nxv32i32:
-; RV64:       # %bb.0:
-; RV64-NEXT:    csrr a3, vlenb
-; RV64-NEXT:    srli a2, a3, 2
-; RV64-NEXT:    vsetvli a4, zero, e8, mf2, ta, ma
-; RV64-NEXT:    vslidedown.vx v24, v0, a2
-; RV64-NEXT:    andi a2, a0, -1
-; RV64-NEXT:    slli a3, a3, 1
-; RV64-NEXT:    sub a0, a1, a3
-; RV64-NEXT:    sltu a4, a1, a0
-; RV64-NEXT:    addi a4, a4, -1
-; RV64-NEXT:    and a0, a4, a0
-; RV64-NEXT:    bltu a1, a3, .LBB67_2
-; RV64-NEXT:  # %bb.1:
-; RV64-NEXT:    mv a1, a3
-; RV64-NEXT:  .LBB67_2:
-; RV64-NEXT:    vsetvli zero, zero, e32, m2, ta, ma
-; RV64-NEXT:    vmv.s.x v25, a2
-; RV64-NEXT:    vsetvli zero, a1, e32, m8, ta, ma
-; RV64-NEXT:    vredmaxu.vs v25, v8, v25, v0.t
-; RV64-NEXT:    vsetvli zero, a0, e32, m8, ta, ma
-; RV64-NEXT:    vmv1r.v v0, v24
-; RV64-NEXT:    vredmaxu.vs v25, v16, v25, v0.t
-; RV64-NEXT:    vmv.x.s a0, v25
-; RV64-NEXT:    ret
+; CHECK-LABEL: vpreduce_umax_nxv32i32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    csrr a3, vlenb
+; CHECK-NEXT:    srli a2, a3, 2
+; CHECK-NEXT:    vsetvli a4, zero, e8, mf2, ta, ma
+; CHECK-NEXT:    vslidedown.vx v24, v0, a2
+; CHECK-NEXT:    slli a3, a3, 1
+; CHECK-NEXT:    sub a2, a1, a3
+; CHECK-NEXT:    sltu a4, a1, a2
+; CHECK-NEXT:    addi a4, a4, -1
+; CHECK-NEXT:    and a2, a4, a2
+; CHECK-NEXT:    bltu a1, a3, .LBB67_2
+; CHECK-NEXT:  # %bb.1:
+; CHECK-NEXT:    mv a1, a3
+; CHECK-NEXT:  .LBB67_2:
+; CHECK-NEXT:    vsetvli zero, zero, e32, m2, ta, ma
+; CHECK-NEXT:    vmv.s.x v25, a0
+; CHECK-NEXT:    vsetvli zero, a1, e32, m8, ta, ma
+; CHECK-NEXT:    vredmaxu.vs v25, v8, v25, v0.t
+; CHECK-NEXT:    vsetvli zero, a2, e32, m8, ta, ma
+; CHECK-NEXT:    vmv1r.v v0, v24
+; CHECK-NEXT:    vredmaxu.vs v25, v16, v25, v0.t
+; CHECK-NEXT:    vmv.x.s a0, v25
+; CHECK-NEXT:    ret
   %r = call i32 @llvm.vp.reduce.umax.nxv32i32(i32 %s, <vscale x 32 x i32> %v, <vscale x 32 x i1> %m, i32 %evl)
   ret i32 %r
 }
@@ -1218,24 +1142,14 @@ define signext i32 @vpreduce_smax_nxv4i32(i32 signext %s, <vscale x 4 x i32> %v,
 declare i32 @llvm.vp.reduce.umin.nxv4i32(i32, <vscale x 4 x i32>, <vscale x 4 x i1>, i32)
 
 define signext i32 @vpreduce_umin_nxv4i32(i32 signext %s, <vscale x 4 x i32> %v, <vscale x 4 x i1> %m, i32 zeroext %evl) {
-; RV32-LABEL: vpreduce_umin_nxv4i32:
-; RV32:       # %bb.0:
-; RV32-NEXT:    vsetivli zero, 1, e32, m1, ta, ma
-; RV32-NEXT:    vmv.s.x v10, a0
-; RV32-NEXT:    vsetvli zero, a1, e32, m2, ta, ma
-; RV32-NEXT:    vredminu.vs v10, v8, v10, v0.t
-; RV32-NEXT:    vmv.x.s a0, v10
-; RV32-NEXT:    ret
-;
-; RV64-LABEL: vpreduce_umin_nxv4i32:
-; RV64:       # %bb.0:
-; RV64-NEXT:    andi a0, a0, -1
-; RV64-NEXT:    vsetivli zero, 1, e32, m1, ta, ma
-; RV64-NEXT:    vmv.s.x v10, a0
-; RV64-NEXT:    vsetvli zero, a1, e32, m2, ta, ma
-; RV64-NEXT:    vredminu.vs v10, v8, v10, v0.t
-; RV64-NEXT:    vmv.x.s a0, v10
-; RV64-NEXT:    ret
+; CHECK-LABEL: vpreduce_umin_nxv4i32:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    vsetivli zero, 1, e32, m1, ta, ma
+; CHECK-NEXT:    vmv.s.x v10, a0
+; CHECK-NEXT:    vsetvli zero, a1, e32, m2, ta, ma
+; CHECK-NEXT:    vredminu.vs v10, v8, v10, v0.t
+; CHECK-NEXT:    vmv.x.s a0, v10
+; CHECK-NEXT:    ret
   %r = call i32 @llvm.vp.reduce.umin.nxv4i32(i32 %s, <vscale x 4 x i32> %v, <vscale x 4 x i1> %m, i32 %evl)
   ret i32 %r
 }
diff --git a/llvm/test/CodeGen/RISCV/rvv/vsetvli-insert-crossbb.ll b/llvm/test/CodeGen/RISCV/rvv/vsetvli-insert-crossbb.ll
index 621445fb2dc5..4ff2fc7a5fff 100644
--- a/llvm/test/CodeGen/RISCV/rvv/vsetvli-insert-crossbb.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/vsetvli-insert-crossbb.ll
@@ -487,6 +487,54 @@ for.end:                                          ; preds = %for.body, %entry
   ret void
 }
 
+define void @saxpy_vec_demanded_fields(i64 %n, float %a, ptr nocapture readonly %x, ptr nocapture %y) {
+; CHECK-LABEL: saxpy_vec_demanded_fields:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    vsetvli a3, a0, e32, m8, ta, ma
+; CHECK-NEXT:    beqz a3, .LBB9_2
+; CHECK-NEXT:  .LBB9_1: # %for.body
+; CHECK-NEXT:    # =>This Inner Loop Header: Depth=1
+; CHECK-NEXT:    vsetvli zero, a3, e32, m8, ta, ma
+; CHECK-NEXT:    vle32.v v8, (a1)
+; CHECK-NEXT:    vle32.v v16, (a2)
+; CHECK-NEXT:    slli a4, a3, 2
+; CHECK-NEXT:    add a1, a1, a4
+; CHECK-NEXT:    vsetvli zero, zero, e32, m8, tu, ma
+; CHECK-NEXT:    vfmacc.vf v16, fa0, v8
+; CHECK-NEXT:    vse32.v v16, (a2)
+; CHECK-NEXT:    sub a0, a0, a3
+; CHECK-NEXT:    vsetvli a3, a0, e16, m4, ta, ma
+; CHECK-NEXT:    add a2, a2, a4
+; CHECK-NEXT:    bnez a3, .LBB9_1
+; CHECK-NEXT:  .LBB9_2: # %for.end
+; CHECK-NEXT:    ret
+entry:
+  %0 = tail call i64 @llvm.riscv.vsetvli.i64(i64 %n, i64 2, i64 3)
+  %cmp.not13 = icmp eq i64 %0, 0
+  br i1 %cmp.not13, label %for.end, label %for.body
+
+for.body:                                         ; preds = %for.body, %entry
+  %1 = phi i64 [ %7, %for.body ], [ %0, %entry ]
+  %n.addr.016 = phi i64 [ %sub, %for.body ], [ %n, %entry ]
+  %x.addr.015 = phi ptr [ %add.ptr, %for.body ], [ %x, %entry ]
+  %y.addr.014 = phi ptr [ %add.ptr1, %for.body ], [ %y, %entry ]
+  %2 = bitcast ptr %x.addr.015 to ptr
+  %3 = tail call <vscale x 16 x float> @llvm.riscv.vle.nxv16f32.i64(<vscale x 16 x float> undef, ptr %2, i64 %1)
+  %add.ptr = getelementptr inbounds float, ptr %x.addr.015, i64 %1
+  %4 = bitcast ptr %y.addr.014 to ptr
+  %5 = tail call <vscale x 16 x float> @llvm.riscv.vle.nxv16f32.i64(<vscale x 16 x float> undef, ptr %4, i64 %1)
+  %6 = tail call <vscale x 16 x float> @llvm.riscv.vfmacc.nxv16f32.f32.i64(<vscale x 16 x float> %5, float %a, <vscale x 16 x float> %3, i64 7, i64 %1, i64 0)
+  tail call void @llvm.riscv.vse.nxv16f32.i64(<vscale x 16 x float> %6, ptr %4, i64 %1)
+  %add.ptr1 = getelementptr inbounds float, ptr %y.addr.014, i64 %1
+  %sub = sub i64 %n.addr.016, %1
+  %7 = tail call i64 @llvm.riscv.vsetvli.i64(i64 %sub, i64 1, i64 2)
+  %cmp.not = icmp eq i64 %7, 0
+  br i1 %cmp.not, label %for.end, label %for.body
+
+for.end:                                          ; preds = %for.body, %entry
+  ret void
+}
+
 declare i64 @llvm.riscv.vsetvli.i64(i64, i64 immarg, i64 immarg)
 declare <vscale x 16 x float> @llvm.riscv.vle.nxv16f32.i64(<vscale x 16 x float>, ptr nocapture, i64)
 declare <vscale x 16 x float> @llvm.riscv.vfmacc.nxv16f32.f32.i64(<vscale x 16 x float>, float, <vscale x 16 x float>, i64, i64, i64)
@@ -501,12 +549,12 @@ define <vscale x 2 x i32> @test_vsetvli_x0_x0(ptr %x, ptr %y, <vscale x 2 x i32>
 ; CHECK-NEXT:    vsetvli zero, a2, e32, m1, ta, ma
 ; CHECK-NEXT:    vle32.v v9, (a0)
 ; CHECK-NEXT:    andi a3, a3, 1
-; CHECK-NEXT:    beqz a3, .LBB9_2
+; CHECK-NEXT:    beqz a3, .LBB10_2
 ; CHECK-NEXT:  # %bb.1: # %if
 ; CHECK-NEXT:    vle16.v v10, (a1)
 ; CHECK-NEXT:    vsetvli zero, zero, e16, mf2, ta, ma
 ; CHECK-NEXT:    vwcvt.x.x.v v8, v10
-; CHECK-NEXT:  .LBB9_2: # %if.end
+; CHECK-NEXT:  .LBB10_2: # %if.end
 ; CHECK-NEXT:    vsetvli zero, zero, e32, m1, ta, ma
 ; CHECK-NEXT:    vadd.vv v8, v9, v8
 ; CHECK-NEXT:    ret
@@ -540,19 +588,19 @@ define <vscale x 2 x i32> @test_vsetvli_x0_x0_2(ptr %x, ptr %y, ptr %z, i64 %vl,
 ; CHECK-NEXT:    vsetvli zero, a3, e32, m1, ta, ma
 ; CHECK-NEXT:    vle32.v v9, (a0)
 ; CHECK-NEXT:    andi a4, a4, 1
-; CHECK-NEXT:    beqz a4, .LBB10_2
+; CHECK-NEXT:    beqz a4, .LBB11_2
 ; CHECK-NEXT:  # %bb.1: # %if
 ; CHECK-NEXT:    vle16.v v10, (a1)
 ; CHECK-NEXT:    vsetvli zero, zero, e16, mf2, ta, ma
 ; CHECK-NEXT:    vwadd.wv v9, v9, v10
-; CHECK-NEXT:  .LBB10_2: # %if.end
+; CHECK-NEXT:  .LBB11_2: # %if.end
 ; CHECK-NEXT:    andi a5, a5, 1
-; CHECK-NEXT:    beqz a5, .LBB10_4
+; CHECK-NEXT:    beqz a5, .LBB11_4
 ; CHECK-NEXT:  # %bb.3: # %if2
 ; CHECK-NEXT:    vsetvli zero, zero, e16, mf2, ta, ma
 ; CHECK-NEXT:    vle16.v v10, (a2)
 ; CHECK-NEXT:    vwadd.wv v9, v9, v10
-; CHECK-NEXT:  .LBB10_4: # %if2.end
+; CHECK-NEXT:  .LBB11_4: # %if2.end
 ; CHECK-NEXT:    vsetvli zero, zero, e32, m1, ta, ma
 ; CHECK-NEXT:    vadd.vv v8, v9, v8
 ; CHECK-NEXT:    ret
@@ -586,11 +634,11 @@ define void @vlmax(i64 %N, ptr %c, ptr %a, ptr %b) {
 ; CHECK-LABEL: vlmax:
 ; CHECK:       # %bb.0: # %entry
 ; CHECK-NEXT:    vsetvli a6, zero, e64, m1, ta, ma
-; CHECK-NEXT:    blez a0, .LBB11_3
+; CHECK-NEXT:    blez a0, .LBB12_3
 ; CHECK-NEXT:  # %bb.1: # %for.body.preheader
 ; CHECK-NEXT:    li a5, 0
 ; CHECK-NEXT:    slli a4, a6, 3
-; CHECK-NEXT:  .LBB11_2: # %for.body
+; CHECK-NEXT:  .LBB12_2: # %for.body
 ; CHECK-NEXT:    # =>This Inner Loop Header: Depth=1
 ; CHECK-NEXT:    vle64.v v8, (a2)
 ; CHECK-NEXT:    vle64.v v9, (a3)
@@ -600,8 +648,8 @@ define void @vlmax(i64 %N, ptr %c, ptr %a, ptr %b) {
 ; CHECK-NEXT:    add a1, a1, a4
 ; CHECK-NEXT:    add a3, a3, a4
 ; CHECK-NEXT:    add a2, a2, a4
-; CHECK-NEXT:    blt a5, a0, .LBB11_2
-; CHECK-NEXT:  .LBB11_3: # %for.end
+; CHECK-NEXT:    blt a5, a0, .LBB12_2
+; CHECK-NEXT:  .LBB12_3: # %for.end
 ; CHECK-NEXT:    ret
 entry:
   %0 = tail call i64 @llvm.riscv.vsetvlimax.i64(i64 3, i64 0)
@@ -633,18 +681,18 @@ define void @vector_init_vlmax(i64 %N, ptr %c) {
 ; CHECK-LABEL: vector_init_vlmax:
 ; CHECK:       # %bb.0: # %entry
 ; CHECK-NEXT:    vsetvli a2, zero, e64, m1, ta, ma
-; CHECK-NEXT:    blez a0, .LBB12_3
+; CHECK-NEXT:    blez a0, .LBB13_3
 ; CHECK-NEXT:  # %bb.1: # %for.body.preheader
 ; CHECK-NEXT:    li a3, 0
 ; CHECK-NEXT:    slli a4, a2, 3
 ; CHECK-NEXT:    vmv.v.i v8, 0
-; CHECK-NEXT:  .LBB12_2: # %for.body
+; CHECK-NEXT:  .LBB13_2: # %for.body
 ; CHECK-NEXT:    # =>This Inner Loop Header: Depth=1
 ; CHECK-NEXT:    vse64.v v8, (a1)
 ; CHECK-NEXT:    add a3, a3, a2
 ; CHECK-NEXT:    add a1, a1, a4
-; CHECK-NEXT:    blt a3, a0, .LBB12_2
-; CHECK-NEXT:  .LBB12_3: # %for.end
+; CHECK-NEXT:    blt a3, a0, .LBB13_2
+; CHECK-NEXT:  .LBB13_3: # %for.end
 ; CHECK-NEXT:    ret
 entry:
   %0 = tail call i64 @llvm.riscv.vsetvlimax.i64(i64 3, i64 0)
@@ -669,20 +717,20 @@ define void @vector_init_vsetvli_N(i64 %N, ptr %c) {
 ; CHECK-LABEL: vector_init_vsetvli_N:
 ; CHECK:       # %bb.0: # %entry
 ; CHECK-NEXT:    vsetvli a2, a0, e64, m1, ta, ma
-; CHECK-NEXT:    blez a0, .LBB13_3
+; CHECK-NEXT:    blez a0, .LBB14_3
 ; CHECK-NEXT:  # %bb.1: # %for.body.preheader
 ; CHECK-NEXT:    li a3, 0
 ; CHECK-NEXT:    slli a4, a2, 3
 ; CHECK-NEXT:    vsetvli a5, zero, e64, m1, ta, ma
 ; CHECK-NEXT:    vmv.v.i v8, 0
-; CHECK-NEXT:  .LBB13_2: # %for.body
+; CHECK-NEXT:  .LBB14_2: # %for.body
 ; CHECK-NEXT:    # =>This Inner Loop Header: Depth=1
 ; CHECK-NEXT:    vsetvli zero, a2, e64, m1, ta, ma
 ; CHECK-NEXT:    vse64.v v8, (a1)
 ; CHECK-NEXT:    add a3, a3, a2
 ; CHECK-NEXT:    add a1, a1, a4
-; CHECK-NEXT:    blt a3, a0, .LBB13_2
-; CHECK-NEXT:  .LBB13_3: # %for.end
+; CHECK-NEXT:    blt a3, a0, .LBB14_2
+; CHECK-NEXT:  .LBB14_3: # %for.end
 ; CHECK-NEXT:    ret
 entry:
   %0 = tail call i64 @llvm.riscv.vsetvli(i64 %N, i64 3, i64 0)
@@ -711,13 +759,13 @@ define void @vector_init_vsetvli_fv(i64 %N, ptr %c) {
 ; CHECK-NEXT:    slli a4, a3, 3
 ; CHECK-NEXT:    vsetvli a5, zero, e64, m1, ta, ma
 ; CHECK-NEXT:    vmv.v.i v8, 0
-; CHECK-NEXT:  .LBB14_1: # %for.body
+; CHECK-NEXT:  .LBB15_1: # %for.body
 ; CHECK-NEXT:    # =>This Inner Loop Header: Depth=1
 ; CHECK-NEXT:    vsetivli zero, 4, e64, m1, ta, ma
 ; CHECK-NEXT:    vse64.v v8, (a1)
 ; CHECK-NEXT:    add a2, a2, a3
 ; CHECK-NEXT:    add a1, a1, a4
-; CHECK-NEXT:    blt a2, a0, .LBB14_1
+; CHECK-NEXT:    blt a2, a0, .LBB15_1
 ; CHECK-NEXT:  # %bb.2: # %for.end
 ; CHECK-NEXT:    ret
 entry:
@@ -745,13 +793,13 @@ define void @vector_init_vsetvli_fv2(i64 %N, ptr %c) {
 ; CHECK-NEXT:    li a2, 0
 ; CHECK-NEXT:    vsetvli a3, zero, e64, m1, ta, ma
 ; CHECK-NEXT:    vmv.v.i v8, 0
-; CHECK-NEXT:  .LBB15_1: # %for.body
+; CHECK-NEXT:  .LBB16_1: # %for.body
 ; CHECK-NEXT:    # =>This Inner Loop Header: Depth=1
 ; CHECK-NEXT:    vsetivli zero, 4, e64, m1, ta, ma
 ; CHECK-NEXT:    vse64.v v8, (a1)
 ; CHECK-NEXT:    addi a2, a2, 4
 ; CHECK-NEXT:    addi a1, a1, 32
-; CHECK-NEXT:    blt a2, a0, .LBB15_1
+; CHECK-NEXT:    blt a2, a0, .LBB16_1
 ; CHECK-NEXT:  # %bb.2: # %for.end
 ; CHECK-NEXT:    ret
 entry:
@@ -779,13 +827,13 @@ define void @vector_init_vsetvli_fv3(i64 %N, ptr %c) {
 ; CHECK-NEXT:    li a2, 0
 ; CHECK-NEXT:    vsetvli a3, zero, e64, m1, ta, ma
 ; CHECK-NEXT:    vmv.v.i v8, 0
-; CHECK-NEXT:  .LBB16_1: # %for.body
+; CHECK-NEXT:  .LBB17_1: # %for.body
 ; CHECK-NEXT:    # =>This Inner Loop Header: Depth=1
 ; CHECK-NEXT:    vsetivli zero, 4, e64, m1, ta, ma
 ; CHECK-NEXT:    vse64.v v8, (a1)
 ; CHECK-NEXT:    addi a2, a2, 4
 ; CHECK-NEXT:    addi a1, a1, 32
-; CHECK-NEXT:    blt a2, a0, .LBB16_1
+; CHECK-NEXT:    blt a2, a0, .LBB17_1
 ; CHECK-NEXT:  # %bb.2: # %for.end
 ; CHECK-NEXT:    ret
 entry:
@@ -861,10 +909,10 @@ define <vscale x 1 x double> @compat_store_consistency(i1 %cond, <vscale x 1 x d
 ; CHECK-NEXT:    vsetvli a3, zero, e64, m1, ta, ma
 ; CHECK-NEXT:    vfadd.vv v8, v8, v9
 ; CHECK-NEXT:    vs1r.v v8, (a1)
-; CHECK-NEXT:    beqz a0, .LBB19_2
+; CHECK-NEXT:    beqz a0, .LBB20_2
 ; CHECK-NEXT:  # %bb.1: # %if.then
 ; CHECK-NEXT:    vse32.v v10, (a2)
-; CHECK-NEXT:  .LBB19_2: # %if.end
+; CHECK-NEXT:  .LBB20_2: # %if.end
 ; CHECK-NEXT:    ret
 entry:
   %res = fadd <vscale x 1 x double> %a, %b
@@ -886,16 +934,16 @@ define <vscale x 2 x i32> @test_ratio_only_vmv_s_x(ptr %x, ptr %y, i1 %cond) nou
 ; CHECK-LABEL: test_ratio_only_vmv_s_x:
 ; CHECK:       # %bb.0: # %entry
 ; CHECK-NEXT:    andi a2, a2, 1
-; CHECK-NEXT:    beqz a2, .LBB20_2
+; CHECK-NEXT:    beqz a2, .LBB21_2
 ; CHECK-NEXT:  # %bb.1: # %if
 ; CHECK-NEXT:    vsetivli zero, 2, e16, mf2, ta, ma
 ; CHECK-NEXT:    vle16.v v9, (a1)
 ; CHECK-NEXT:    vwcvt.x.x.v v8, v9
-; CHECK-NEXT:    j .LBB20_3
-; CHECK-NEXT:  .LBB20_2:
+; CHECK-NEXT:    j .LBB21_3
+; CHECK-NEXT:  .LBB21_2:
 ; CHECK-NEXT:    vsetivli zero, 2, e32, m1, ta, ma
 ; CHECK-NEXT:    vle32.v v8, (a0)
-; CHECK-NEXT:  .LBB20_3: # %if.end
+; CHECK-NEXT:  .LBB21_3: # %if.end
 ; CHECK-NEXT:    vsetvli zero, zero, e32, m1, tu, ma
 ; CHECK-NEXT:    vmv.s.x v8, zero
 ; CHECK-NEXT:    ret
@@ -918,16 +966,16 @@ define <vscale x 2 x i32> @test_ratio_only_vmv_s_x2(ptr %x, ptr %y, i1 %cond) no
 ; CHECK-LABEL: test_ratio_only_vmv_s_x2:
 ; CHECK:       # %bb.0: # %entry
 ; CHECK-NEXT:    andi a2, a2, 1
-; CHECK-NEXT:    beqz a2, .LBB21_2
+; CHECK-NEXT:    beqz a2, .LBB22_2
 ; CHECK-NEXT:  # %bb.1: # %if
 ; CHECK-NEXT:    vsetivli zero, 2, e32, m1, ta, ma
 ; CHECK-NEXT:    vle32.v v8, (a0)
-; CHECK-NEXT:    j .LBB21_3
-; CHECK-NEXT:  .LBB21_2:
+; CHECK-NEXT:    j .LBB22_3
+; CHECK-NEXT:  .LBB22_2:
 ; CHECK-NEXT:    vsetivli zero, 2, e16, mf2, ta, ma
 ; CHECK-NEXT:    vle16.v v9, (a1)
 ; CHECK-NEXT:    vwcvt.x.x.v v8, v9
-; CHECK-NEXT:  .LBB21_3: # %if.end
+; CHECK-NEXT:  .LBB22_3: # %if.end
 ; CHECK-NEXT:    vsetvli zero, zero, e32, m1, tu, ma
 ; CHECK-NEXT:    vmv.s.x v8, zero
 ; CHECK-NEXT:    ret
@@ -953,13 +1001,13 @@ define void @pre_over_vle(ptr %A) {
 ; CHECK:       # %bb.0: # %entry
 ; CHECK-NEXT:    addi a1, a0, 800
 ; CHECK-NEXT:    vsetivli zero, 2, e32, mf2, ta, ma
-; CHECK-NEXT:  .LBB22_1: # %vector.body
+; CHECK-NEXT:  .LBB23_1: # %vector.body
 ; CHECK-NEXT:    # =>This Inner Loop Header: Depth=1
 ; CHECK-NEXT:    vle8.v v8, (a0)
 ; CHECK-NEXT:    vsext.vf4 v9, v8
 ; CHECK-NEXT:    vse32.v v9, (a0)
 ; CHECK-NEXT:    addi a0, a0, 8
-; CHECK-NEXT:    bne a0, a1, .LBB22_1
+; CHECK-NEXT:    bne a0, a1, .LBB23_1
 ; CHECK-NEXT:  # %bb.2: # %exit
 ; CHECK-NEXT:    ret
 entry:
diff --git a/llvm/test/CodeGen/RISCV/rvv/vsetvli-insert-crossbb.mir b/llvm/test/CodeGen/RISCV/rvv/vsetvli-insert-crossbb.mir
index fdcce72a01eb..16c4a1a0a89e 100644
--- a/llvm/test/CodeGen/RISCV/rvv/vsetvli-insert-crossbb.mir
+++ b/llvm/test/CodeGen/RISCV/rvv/vsetvli-insert-crossbb.mir
@@ -130,6 +130,10 @@
     ret void
   }
 
+  define void @pre_undemanded_vl() {
+    ret void
+  }
+
   declare i32 @llvm.vector.reduce.add.v4i32(<4 x i32>)
 
   declare <vscale x 1 x i64> @llvm.riscv.vadd.nxv1i64.nxv1i64.i64(<vscale x 1 x i64>, <vscale x 1 x i64>, <vscale x 1 x i64>, i64) #1
@@ -517,10 +521,10 @@ body:             |
   ; CHECK-NEXT:   [[COPY1:%[0-9]+]]:gpr = COPY $x10
   ; CHECK-NEXT:   [[DEF:%[0-9]+]]:gpr = IMPLICIT_DEF
   ; CHECK-NEXT:   %pt:vr = IMPLICIT_DEF
-  ; CHECK-NEXT:   dead [[PseudoVSETVLIX0_:%[0-9]+]]:gpr = PseudoVSETVLIX0 $x0, 223 /* e64, mf2, ta, ma */, implicit-def $vl, implicit-def $vtype
+  ; CHECK-NEXT:   dead [[PseudoVSETVLIX0_:%[0-9]+]]:gpr = PseudoVSETVLIX0 killed $x0, 223 /* e64, mf2, ta, ma */, implicit-def $vl, implicit-def $vtype
   ; CHECK-NEXT:   [[PseudoVID_V_MF2_:%[0-9]+]]:vr = PseudoVID_V_MF2 %pt, -1, 6 /* e64 */, 0 /* tu, mu */, implicit $vl, implicit $vtype
   ; CHECK-NEXT:   %pt2:vr = IMPLICIT_DEF
-  ; CHECK-NEXT:   dead [[PseudoVSETVLIX0_1:%[0-9]+]]:gpr = PseudoVSETVLIX0 $x0, 215 /* e32, mf2, ta, ma */, implicit-def $vl, implicit-def $vtype
+  ; CHECK-NEXT:   dead [[PseudoVSETVLIX0_1:%[0-9]+]]:gpr = PseudoVSETVLIX0 killed $x0, 215 /* e32, mf2, ta, ma */, implicit-def $vl, implicit-def $vtype
   ; CHECK-NEXT:   [[PseudoVMV_V_I_MF2_:%[0-9]+]]:vrnov0 = PseudoVMV_V_I_MF2 %pt2, 0, -1, 5 /* e32 */, 0 /* tu, mu */, implicit $vl, implicit $vtype
   ; CHECK-NEXT: {{  $}}
   ; CHECK-NEXT: bb.1:
@@ -609,7 +613,7 @@ body:             |
   ; CHECK-NEXT:   [[SRLI:%[0-9]+]]:gpr = SRLI [[PseudoReadVLENB]], 3
   ; CHECK-NEXT:   [[COPY1:%[0-9]+]]:gpr = COPY $x11
   ; CHECK-NEXT:   %pt:vr = IMPLICIT_DEF
-  ; CHECK-NEXT:   dead [[PseudoVSETVLIX0_:%[0-9]+]]:gpr = PseudoVSETVLIX0 $x0, 216 /* e64, m1, ta, ma */, implicit-def $vl, implicit-def $vtype
+  ; CHECK-NEXT:   dead [[PseudoVSETVLIX0_:%[0-9]+]]:gpr = PseudoVSETVLIX0 killed $x0, 216 /* e64, m1, ta, ma */, implicit-def $vl, implicit-def $vtype
   ; CHECK-NEXT:   [[PseudoVID_V_M1_:%[0-9]+]]:vr = PseudoVID_V_M1 %pt, -1, 6 /* e64 */, 0 /* tu, mu */, implicit $vl, implicit $vtype
   ; CHECK-NEXT:   [[COPY2:%[0-9]+]]:gpr = COPY $x0
   ; CHECK-NEXT: {{  $}}
@@ -681,7 +685,7 @@ body:             |
   ; CHECK-NEXT:   [[SRLI:%[0-9]+]]:gpr = SRLI [[PseudoReadVLENB]], 3
   ; CHECK-NEXT:   [[COPY1:%[0-9]+]]:gpr = COPY $x11
   ; CHECK-NEXT:   %pt:vr = IMPLICIT_DEF
-  ; CHECK-NEXT:   dead [[PseudoVSETVLIX0_:%[0-9]+]]:gpr = PseudoVSETVLIX0 $x0, 216 /* e64, m1, ta, ma */, implicit-def $vl, implicit-def $vtype
+  ; CHECK-NEXT:   dead [[PseudoVSETVLIX0_:%[0-9]+]]:gpr = PseudoVSETVLIX0 killed $x0, 216 /* e64, m1, ta, ma */, implicit-def $vl, implicit-def $vtype
   ; CHECK-NEXT:   [[PseudoVID_V_M1_:%[0-9]+]]:vr = PseudoVID_V_M1 %pt, -1, 6 /* e64 */, 3 /* ta, ma */, implicit $vl, implicit $vtype
   ; CHECK-NEXT:   [[COPY2:%[0-9]+]]:gpr = COPY $x0
   ; CHECK-NEXT: {{  $}}
@@ -866,7 +870,7 @@ body:             |
   ; CHECK-NEXT:   %t3:vr = COPY $v2
   ; CHECK-NEXT:   %t4:vr = COPY $v3
   ; CHECK-NEXT:   %t5:vrnov0 = COPY $v1
-  ; CHECK-NEXT:   dead [[PseudoVSETVLIX0_:%[0-9]+]]:gpr = PseudoVSETVLIX0 $x0, 216 /* e64, m1, ta, ma */, implicit-def $vl, implicit-def $vtype
+  ; CHECK-NEXT:   dead [[PseudoVSETVLIX0_:%[0-9]+]]:gpr = PseudoVSETVLIX0 killed $x0, 216 /* e64, m1, ta, ma */, implicit-def $vl, implicit-def $vtype
   ; CHECK-NEXT:   %t6:vr = PseudoVMSEQ_VI_M1 %t1, 0, -1, 6 /* e64 */, implicit $vl, implicit $vtype
   ; CHECK-NEXT:   PseudoBR %bb.1
   ; CHECK-NEXT: {{  $}}
@@ -949,7 +953,7 @@ body:             |
   ; CHECK-NEXT:   %vlenb:gpr = PseudoReadVLENB
   ; CHECK-NEXT:   %inc:gpr = SRLI killed %vlenb, 3
   ; CHECK-NEXT:   %pt:vr = IMPLICIT_DEF
-  ; CHECK-NEXT:   dead [[PseudoVSETVLIX0_:%[0-9]+]]:gpr = PseudoVSETVLIX0 $x0, 216 /* e64, m1, ta, ma */, implicit-def $vl, implicit-def $vtype
+  ; CHECK-NEXT:   dead [[PseudoVSETVLIX0_:%[0-9]+]]:gpr = PseudoVSETVLIX0 killed $x0, 216 /* e64, m1, ta, ma */, implicit-def $vl, implicit-def $vtype
   ; CHECK-NEXT:   [[PseudoVID_V_M1_:%[0-9]+]]:vr = PseudoVID_V_M1 %pt, -1, 6 /* e64 */, 0 /* tu, mu */, implicit $vl, implicit $vtype
   ; CHECK-NEXT:   [[COPY3:%[0-9]+]]:gpr = COPY $x0
   ; CHECK-NEXT:   PseudoBR %bb.1
@@ -1041,3 +1045,12 @@ body:             |
     PseudoRET
 
 ...
+---
+name: pre_undemanded_vl
+body: |
+  bb.0:
+    PseudoBR %bb.1
+  bb.1:
+    %x:gpr = PseudoVMV_X_S undef $noreg, 6
+    PseudoBR %bb.1
+...
diff --git a/llvm/test/CodeGen/RISCV/rvv/vsetvli-insert.mir b/llvm/test/CodeGen/RISCV/rvv/vsetvli-insert.mir
index e8620c848f8d..e567897aa868 100644
--- a/llvm/test/CodeGen/RISCV/rvv/vsetvli-insert.mir
+++ b/llvm/test/CodeGen/RISCV/rvv/vsetvli-insert.mir
@@ -1,6 +1,6 @@
 # NOTE: Assertions have been autogenerated by utils/update_mir_test_checks.py
 # RUN: llc %s -o - -mtriple=riscv64 -mattr=v \
-# RUN:     -run-pass=riscv-insert-vsetvli | FileCheck %s
+# RUN:     -run-pass=riscv-insert-vsetvli,riscv-coalesce-vsetvli | FileCheck %s
 
 --- |
   source_filename = "vsetvli-insert.ll"
@@ -166,7 +166,7 @@ body:             |
     ; CHECK-NEXT: dead $x0 = PseudoVSETVLI [[COPY]], 216 /* e64, m1, ta, ma */, implicit-def $vl, implicit-def $vtype
     ; CHECK-NEXT: [[PseudoVLE64_V_M1_:%[0-9]+]]:vr = PseudoVLE64_V_M1 %pt, [[COPY2]], $noreg, 6 /* e64 */, 0 /* tu, mu */, implicit $vl, implicit $vtype
     ; CHECK-NEXT: %pt2:vr = IMPLICIT_DEF
-    ; CHECK-NEXT: [[PseudoVADD_VV_M1_:%[0-9]+]]:vr = PseudoVADD_VV_M1 %pt2, killed [[PseudoVLE64_V_M1_]], [[COPY1]], $noreg, 6 /* e64 */, 0 /* tu, mu */, implicit $vl, implicit $vtype
+    ; CHECK-NEXT: [[PseudoVADD_VV_M1_:%[0-9]+]]:vr = PseudoVADD_VV_M1 %pt2, [[PseudoVLE64_V_M1_]], [[COPY1]], $noreg, 6 /* e64 */, 0 /* tu, mu */, implicit $vl, implicit $vtype
     ; CHECK-NEXT: $v8 = COPY [[PseudoVADD_VV_M1_]]
     ; CHECK-NEXT: PseudoRET implicit $v8
     %2:gprnox0 = COPY $x11
@@ -208,7 +208,7 @@ body:             |
     ; CHECK-NEXT: dead $x0 = PseudoVSETVLI [[COPY]], 216 /* e64, m1, ta, ma */, implicit-def $vl, implicit-def $vtype
     ; CHECK-NEXT: [[PseudoVLE32_V_MF2_:%[0-9]+]]:vr = PseudoVLE32_V_MF2 %pt, [[COPY1]], $noreg, 5 /* e32 */, 0 /* tu, mu */, implicit $vl, implicit $vtype
     ; CHECK-NEXT: %dead:vr = IMPLICIT_DEF
-    ; CHECK-NEXT: early-clobber %3:vr = PseudoVZEXT_VF2_M1 %dead, killed [[PseudoVLE32_V_MF2_]], $noreg, 6 /* e64 */, 0 /* tu, mu */, implicit $vl, implicit $vtype
+    ; CHECK-NEXT: early-clobber %3:vr = PseudoVZEXT_VF2_M1 %dead, [[PseudoVLE32_V_MF2_]], $noreg, 6 /* e64 */, 0 /* tu, mu */, implicit $vl, implicit $vtype
     ; CHECK-NEXT: $v8 = COPY %3
     ; CHECK-NEXT: PseudoRET implicit $v8
     %1:gprnox0 = COPY $x11
@@ -282,8 +282,8 @@ body:             |
     ; CHECK-NEXT: [[PseudoVLE64_V_M1_:%[0-9]+]]:vr = PseudoVLE64_V_M1 %pt, [[COPY1]], 2, 6 /* e64 */, 0 /* tu, mu */, implicit $vl, implicit $vtype :: (load (s128) from %ir.x)
     ; CHECK-NEXT: [[PseudoVLE64_V_M1_1:%[0-9]+]]:vr = PseudoVLE64_V_M1 %pt2, [[COPY]], 2, 6 /* e64 */, 0 /* tu, mu */, implicit $vl, implicit $vtype :: (load (s128) from %ir.y)
     ; CHECK-NEXT: %pt3:vr = IMPLICIT_DEF
-    ; CHECK-NEXT: [[PseudoVADD_VV_M1_:%[0-9]+]]:vr = PseudoVADD_VV_M1 %pt3, killed [[PseudoVLE64_V_M1_]], killed [[PseudoVLE64_V_M1_1]], 2, 6 /* e64 */, 0 /* tu, mu */, implicit $vl, implicit $vtype
-    ; CHECK-NEXT: PseudoVSE64_V_M1 killed [[PseudoVADD_VV_M1_]], [[COPY1]], 2, 6 /* e64 */, implicit $vl, implicit $vtype :: (store (s128) into %ir.x)
+    ; CHECK-NEXT: [[PseudoVADD_VV_M1_:%[0-9]+]]:vr = PseudoVADD_VV_M1 %pt3, [[PseudoVLE64_V_M1_]], [[PseudoVLE64_V_M1_1]], 2, 6 /* e64 */, 0 /* tu, mu */, implicit $vl, implicit $vtype
+    ; CHECK-NEXT: PseudoVSE64_V_M1 [[PseudoVADD_VV_M1_]], [[COPY1]], 2, 6 /* e64 */, implicit $vl, implicit $vtype :: (store (s128) into %ir.x)
     ; CHECK-NEXT: PseudoRET
     %1:gpr = COPY $x11
     %0:gpr = COPY $x10
@@ -324,12 +324,12 @@ body:             |
     ; CHECK-NEXT: %pt:vr = IMPLICIT_DEF
     ; CHECK-NEXT: dead $x0 = PseudoVSETIVLI 2, 216 /* e64, m1, ta, ma */, implicit-def $vl, implicit-def $vtype
     ; CHECK-NEXT: [[PseudoVLE64_V_M1_:%[0-9]+]]:vr = PseudoVLE64_V_M1 %pt, [[COPY]], 2, 6 /* e64 */, 0 /* tu, mu */, implicit $vl, implicit $vtype :: (load (s128) from %ir.x)
-    ; CHECK-NEXT: dead [[PseudoVSETVLIX0_:%[0-9]+]]:gpr = PseudoVSETVLIX0 $x0, 216 /* e64, m1, ta, ma */, implicit-def $vl, implicit-def $vtype
+    ; CHECK-NEXT: dead [[PseudoVSETVLIX0_:%[0-9]+]]:gpr = PseudoVSETVLIX0 killed $x0, 216 /* e64, m1, ta, ma */, implicit-def $vl, implicit-def $vtype
     ; CHECK-NEXT: [[PseudoVMV_V_I_M1_:%[0-9]+]]:vr = PseudoVMV_V_I_M1 $noreg, 0, -1, 6 /* e64 */, 0 /* tu, mu */, implicit $vl, implicit $vtype
     ; CHECK-NEXT: [[DEF:%[0-9]+]]:vr = IMPLICIT_DEF
     ; CHECK-NEXT: dead $x0 = PseudoVSETIVLI 2, 216 /* e64, m1, ta, ma */, implicit-def $vl, implicit-def $vtype
-    ; CHECK-NEXT: [[PseudoVREDSUM_VS_M1_E8_:%[0-9]+]]:vr = PseudoVREDSUM_VS_M1_E8 [[DEF]], killed [[PseudoVLE64_V_M1_]], killed [[PseudoVMV_V_I_M1_]], 2, 6 /* e64 */, 1 /* ta, mu */, implicit $vl, implicit $vtype
-    ; CHECK-NEXT: [[PseudoVMV_X_S:%[0-9]+]]:gpr = PseudoVMV_X_S killed [[PseudoVREDSUM_VS_M1_E8_]], 6 /* e64 */, implicit $vtype
+    ; CHECK-NEXT: [[PseudoVREDSUM_VS_M1_E8_:%[0-9]+]]:vr = PseudoVREDSUM_VS_M1_E8 [[DEF]], [[PseudoVLE64_V_M1_]], [[PseudoVMV_V_I_M1_]], 2, 6 /* e64 */, 1 /* ta, mu */, implicit $vl, implicit $vtype
+    ; CHECK-NEXT: [[PseudoVMV_X_S:%[0-9]+]]:gpr = PseudoVMV_X_S [[PseudoVREDSUM_VS_M1_E8_]], 6 /* e64 */, implicit $vtype
     ; CHECK-NEXT: $x10 = COPY [[PseudoVMV_X_S]]
     ; CHECK-NEXT: PseudoRET implicit $x10
     %0:gpr = COPY $x10
@@ -370,7 +370,7 @@ body:             |
     ; CHECK-NEXT: [[COPY:%[0-9]+]]:gprnox0 = COPY $x10
     ; CHECK-NEXT: [[COPY1:%[0-9]+]]:vr = COPY $v9
     ; CHECK-NEXT: [[COPY2:%[0-9]+]]:vr = COPY $v8
-    ; CHECK-NEXT: [[PseudoVSETVLI:%[0-9]+]]:gprnox0 = PseudoVSETVLI [[COPY]], 88 /* e64, m1, ta, mu */, implicit-def $vl, implicit-def $vtype
+    ; CHECK-NEXT: dead [[PseudoVSETVLI:%[0-9]+]]:gprnox0 = PseudoVSETVLI [[COPY]], 88 /* e64, m1, ta, mu */, implicit-def $vl, implicit-def $vtype
     ; CHECK-NEXT: %pt:vr = IMPLICIT_DEF
     ; CHECK-NEXT: [[PseudoVADD_VV_M1_:%[0-9]+]]:vr = PseudoVADD_VV_M1 %pt, [[COPY2]], [[COPY1]], $noreg, 6 /* e64 */, 0 /* tu, mu */, implicit $vl, implicit $vtype
     ; CHECK-NEXT: $v8 = COPY [[PseudoVADD_VV_M1_]]
@@ -418,7 +418,7 @@ body:             |
     ; CHECK-NEXT: INLINEASM &"", 1 /* sideeffect attdialect */
     ; CHECK-NEXT: %pt2:vr = IMPLICIT_DEF
     ; CHECK-NEXT: dead $x0 = PseudoVSETVLI [[COPY]], 216 /* e64, m1, ta, ma */, implicit-def $vl, implicit-def $vtype
-    ; CHECK-NEXT: [[PseudoVADD_VV_M1_:%[0-9]+]]:vr = PseudoVADD_VV_M1 %pt2, killed [[PseudoVLE64_V_M1_]], [[COPY1]], $noreg, 6 /* e64 */, 0 /* tu, mu */, implicit $vl, implicit $vtype
+    ; CHECK-NEXT: [[PseudoVADD_VV_M1_:%[0-9]+]]:vr = PseudoVADD_VV_M1 %pt2, [[PseudoVLE64_V_M1_]], [[COPY1]], $noreg, 6 /* e64 */, 0 /* tu, mu */, implicit $vl, implicit $vtype
     ; CHECK-NEXT: $v8 = COPY [[PseudoVADD_VV_M1_]]
     ; CHECK-NEXT: PseudoRET implicit $v8
     %2:gprnox0 = COPY $x11
@@ -445,9 +445,9 @@ body:             |
     ; CHECK-NEXT: {{  $}}
     ; CHECK-NEXT: %pt:vrm2 = IMPLICIT_DEF
     ; CHECK-NEXT: dead $x0 = PseudoVSETIVLI 4, 217 /* e64, m2, ta, ma */, implicit-def $vl, implicit-def $vtype
-    ; CHECK-NEXT: [[PseudoVID_V_M2_:%[0-9]+]]:vrm2 = PseudoVID_V_M2 %pt, 4, 6 /* e64 */, 3 /* ta, ma */, implicit $vl, implicit $vtype
+    ; CHECK-NEXT: dead [[PseudoVID_V_M2_:%[0-9]+]]:vrm2 = PseudoVID_V_M2 %pt, 4, 6 /* e64 */, 3 /* ta, ma */, implicit $vl, implicit $vtype
     ; CHECK-NEXT: dead $x0 = PseudoVSETVLIX0 killed $x0, 198 /* e8, mf4, ta, ma */, implicit-def $vl, implicit-def $vtype, implicit $vl
-    ; CHECK-NEXT: [[PseudoVMV_V_I_MF4_:%[0-9]+]]:vr = PseudoVMV_V_I_MF4 $noreg, 0, 4, 3 /* e8 */, 0 /* tu, mu */, implicit $vl, implicit $vtype
+    ; CHECK-NEXT: dead [[PseudoVMV_V_I_MF4_:%[0-9]+]]:vr = PseudoVMV_V_I_MF4 $noreg, 0, 4, 3 /* e8 */, 0 /* tu, mu */, implicit $vl, implicit $vtype
     ; CHECK-NEXT: PseudoRET
     %pt:vrm2 = IMPLICIT_DEF
     %0:vrm2 = PseudoVID_V_M2 %pt, 4, 6, 3
@@ -467,14 +467,14 @@ body:             |
   ; CHECK-NEXT: {{  $}}
   ; CHECK-NEXT:   %cond:gpr = COPY $x10
   ; CHECK-NEXT:   dead $x0 = PseudoVSETIVLI 2, 215 /* e32, mf2, ta, ma */, implicit-def $vl, implicit-def $vtype
-  ; CHECK-NEXT:   [[PseudoVMV_V_I_MF2_:%[0-9]+]]:vr = PseudoVMV_V_I_MF2 $noreg, 1, 2, 5 /* e32 */, 0 /* tu, mu */, implicit $vl, implicit $vtype
+  ; CHECK-NEXT:   dead [[PseudoVMV_V_I_MF2_:%[0-9]+]]:vr = PseudoVMV_V_I_MF2 $noreg, 1, 2, 5 /* e32 */, 0 /* tu, mu */, implicit $vl, implicit $vtype
   ; CHECK-NEXT:   BEQ %cond, $x0, %bb.2
   ; CHECK-NEXT: {{  $}}
   ; CHECK-NEXT: bb.1:
   ; CHECK-NEXT:   successors: %bb.2(0x80000000)
   ; CHECK-NEXT: {{  $}}
   ; CHECK-NEXT:   dead $x0 = PseudoVSETVLIX0 killed $x0, 216 /* e64, m1, ta, ma */, implicit-def $vl, implicit-def $vtype, implicit $vl
-  ; CHECK-NEXT:   [[PseudoVMV_V_I_M1_:%[0-9]+]]:vr = PseudoVMV_V_I_M1 $noreg, 1, 2, 6 /* e64 */, 0 /* tu, mu */, implicit $vl, implicit $vtype
+  ; CHECK-NEXT:   dead [[PseudoVMV_V_I_M1_:%[0-9]+]]:vr = PseudoVMV_V_I_M1 $noreg, 1, 2, 6 /* e64 */, 0 /* tu, mu */, implicit $vl, implicit $vtype
   ; CHECK-NEXT: {{  $}}
   ; CHECK-NEXT: bb.2:
   ; CHECK-NEXT:   successors: %bb.4(0x40000000), %bb.3(0x40000000)
@@ -488,8 +488,8 @@ body:             |
   ; CHECK-NEXT: {{  $}}
   ; CHECK-NEXT: bb.4:
   ; CHECK-NEXT:   $x0 = PseudoVSETIVLI 2, 215 /* e32, mf2, ta, ma */, implicit-def $vl, implicit-def $vtype
-  ; CHECK-NEXT:   [[PseudoVMV_X_S:%[0-9]+]]:gpr = PseudoVMV_X_S $noreg, 5 /* e32 */, implicit $vtype
-  ; CHECK-NEXT:   [[PseudoVMV_V_I_MF2_1:%[0-9]+]]:vr = PseudoVMV_V_I_MF2 $noreg, 1, 2, 5 /* e32 */, 0 /* tu, mu */, implicit $vl, implicit $vtype
+  ; CHECK-NEXT:   dead [[PseudoVMV_X_S:%[0-9]+]]:gpr = PseudoVMV_X_S $noreg, 5 /* e32 */, implicit $vtype
+  ; CHECK-NEXT:   dead [[PseudoVMV_V_I_MF2_1:%[0-9]+]]:vr = PseudoVMV_V_I_MF2 $noreg, 1, 2, 5 /* e32 */, 0 /* tu, mu */, implicit $vl, implicit $vtype
   ; CHECK-NEXT:   PseudoRET
   bb.0:
     liveins: $x10
@@ -517,10 +517,10 @@ body:             |
     ; CHECK: liveins: $x1
     ; CHECK-NEXT: {{  $}}
     ; CHECK-NEXT: dead $x0 = PseudoVSETIVLI 3, 216 /* e64, m1, ta, ma */, implicit-def $vl, implicit-def $vtype
-    ; CHECK-NEXT: [[COPY:%[0-9]+]]:gpr = COPY $vtype
+    ; CHECK-NEXT: dead [[COPY:%[0-9]+]]:gpr = COPY $vtype
     ; CHECK-NEXT: $vl = COPY $x1
     ; CHECK-NEXT: dead $x0 = PseudoVSETIVLI 3, 216 /* e64, m1, ta, ma */, implicit-def $vl, implicit-def $vtype
-    ; CHECK-NEXT: [[PseudoVADD_VV_M1_:%[0-9]+]]:vr = PseudoVADD_VV_M1 $noreg, $noreg, $noreg, 3, 6 /* e64 */, 0 /* tu, mu */, implicit $vl, implicit $vtype
+    ; CHECK-NEXT: dead [[PseudoVADD_VV_M1_:%[0-9]+]]:vr = PseudoVADD_VV_M1 $noreg, $noreg, $noreg, 3, 6 /* e64 */, 0 /* tu, mu */, implicit $vl, implicit $vtype
     ; CHECK-NEXT: PseudoRET
     dead $x0 = PseudoVSETIVLI 3, 216, implicit-def $vl, implicit-def $vtype
     %1:gpr = COPY $vtype
diff --git a/llvm/test/CodeGen/RISCV/rvv/vwadd-sdnode.ll b/llvm/test/CodeGen/RISCV/rvv/vwadd-sdnode.ll
index 21ddf1a6e114..d70f619c3601 100644
--- a/llvm/test/CodeGen/RISCV/rvv/vwadd-sdnode.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/vwadd-sdnode.ll
@@ -1484,7 +1484,6 @@ define <vscale x 8 x i64> @vwadd_vx_splat_zext(<vscale x 8 x i32> %va, i32 %b) {
 ;
 ; RV64-LABEL: vwadd_vx_splat_zext:
 ; RV64:       # %bb.0:
-; RV64-NEXT:    andi a0, a0, -1
 ; RV64-NEXT:    vsetvli a1, zero, e32, m4, ta, ma
 ; RV64-NEXT:    vwaddu.vx v16, v8, a0
 ; RV64-NEXT:    vmv8r.v v8, v16
diff --git a/llvm/test/CodeGen/RISCV/rvv/zvlsseg-spill.mir b/llvm/test/CodeGen/RISCV/rvv/zvlsseg-spill.mir
index d0b76e7e4535..fcd852f1210d 100644
--- a/llvm/test/CodeGen/RISCV/rvv/zvlsseg-spill.mir
+++ b/llvm/test/CodeGen/RISCV/rvv/zvlsseg-spill.mir
@@ -27,7 +27,7 @@ body: |
     ; CHECK-NEXT: $x12 = frame-setup SLLI killed $x12, 3
     ; CHECK-NEXT: $x2 = frame-setup SUB $x2, killed $x12
     ; CHECK-NEXT: frame-setup CFI_INSTRUCTION escape 0x0f, 0x0d, 0x72, 0x00, 0x11, 0x10, 0x22, 0x11, 0x08, 0x92, 0xa2, 0x38, 0x00, 0x1e, 0x22
-    ; CHECK-NEXT: dead $x0 = PseudoVSETVLI killed renamable $x11, 152 /* e64, m1, tu, ma */, implicit-def $vl, implicit-def $vtype
+    ; CHECK-NEXT: dead $x0 = PseudoVSETVLI killed renamable $x11, 216 /* e64, m1, ta, ma */, implicit-def $vl, implicit-def $vtype
     ; CHECK-NEXT: $v0_v1_v2_v3_v4_v5_v6 = PseudoVLSEG7E64_V_M1 undef $v0_v1_v2_v3_v4_v5_v6, renamable $x10, $noreg, 6 /* e64 */, 0 /* tu, mu */, implicit $vl, implicit $vtype
     ; CHECK-NEXT: $x11 = ADDI $x2, 16
     ; CHECK-NEXT: $x12 = PseudoReadVLENB
diff --git a/llvm/test/CodeGen/RISCV/select.ll b/llvm/test/CodeGen/RISCV/select.ll
index e07e52091e9e..ffbbe31412ed 100644
--- a/llvm/test/CodeGen/RISCV/select.ll
+++ b/llvm/test/CodeGen/RISCV/select.ll
@@ -1858,3 +1858,113 @@ define i32 @select_cst6(i1 zeroext %cond) {
   %ret = select i1 %cond, i32 2049, i32 2047
   ret i32 %ret
 }
+
+@select_redundant_czero_eqz_data = global i32 0, align 4
+
+define void @select_redundant_czero_eqz1(ptr %0, ptr %1) {
+; RV32IM-LABEL: select_redundant_czero_eqz1:
+; RV32IM:       # %bb.0: # %entry
+; RV32IM-NEXT:    bnez a0, .LBB49_2
+; RV32IM-NEXT:  # %bb.1:
+; RV32IM-NEXT:    lui a0, %hi(select_redundant_czero_eqz_data)
+; RV32IM-NEXT:    addi a0, a0, %lo(select_redundant_czero_eqz_data)
+; RV32IM-NEXT:  .LBB49_2: # %entry
+; RV32IM-NEXT:    sw a0, 0(a1)
+; RV32IM-NEXT:    ret
+;
+; RV64IM-LABEL: select_redundant_czero_eqz1:
+; RV64IM:       # %bb.0: # %entry
+; RV64IM-NEXT:    bnez a0, .LBB49_2
+; RV64IM-NEXT:  # %bb.1:
+; RV64IM-NEXT:    lui a0, %hi(select_redundant_czero_eqz_data)
+; RV64IM-NEXT:    addi a0, a0, %lo(select_redundant_czero_eqz_data)
+; RV64IM-NEXT:  .LBB49_2: # %entry
+; RV64IM-NEXT:    sd a0, 0(a1)
+; RV64IM-NEXT:    ret
+;
+; RV64IMXVTCONDOPS-LABEL: select_redundant_czero_eqz1:
+; RV64IMXVTCONDOPS:       # %bb.0: # %entry
+; RV64IMXVTCONDOPS-NEXT:    lui a2, %hi(select_redundant_czero_eqz_data)
+; RV64IMXVTCONDOPS-NEXT:    addi a2, a2, %lo(select_redundant_czero_eqz_data)
+; RV64IMXVTCONDOPS-NEXT:    vt.maskcn a2, a2, a0
+; RV64IMXVTCONDOPS-NEXT:    or a0, a2, a0
+; RV64IMXVTCONDOPS-NEXT:    sd a0, 0(a1)
+; RV64IMXVTCONDOPS-NEXT:    ret
+;
+; RV32IMZICOND-LABEL: select_redundant_czero_eqz1:
+; RV32IMZICOND:       # %bb.0: # %entry
+; RV32IMZICOND-NEXT:    lui a2, %hi(select_redundant_czero_eqz_data)
+; RV32IMZICOND-NEXT:    addi a2, a2, %lo(select_redundant_czero_eqz_data)
+; RV32IMZICOND-NEXT:    czero.nez a2, a2, a0
+; RV32IMZICOND-NEXT:    or a0, a2, a0
+; RV32IMZICOND-NEXT:    sw a0, 0(a1)
+; RV32IMZICOND-NEXT:    ret
+;
+; RV64IMZICOND-LABEL: select_redundant_czero_eqz1:
+; RV64IMZICOND:       # %bb.0: # %entry
+; RV64IMZICOND-NEXT:    lui a2, %hi(select_redundant_czero_eqz_data)
+; RV64IMZICOND-NEXT:    addi a2, a2, %lo(select_redundant_czero_eqz_data)
+; RV64IMZICOND-NEXT:    czero.nez a2, a2, a0
+; RV64IMZICOND-NEXT:    or a0, a2, a0
+; RV64IMZICOND-NEXT:    sd a0, 0(a1)
+; RV64IMZICOND-NEXT:    ret
+entry:
+  %3 = icmp eq ptr %0, null
+  %4 = select i1 %3, ptr @select_redundant_czero_eqz_data, ptr %0
+  store ptr %4, ptr %1, align 8
+  ret void
+}
+
+define void @select_redundant_czero_eqz2(ptr %0, ptr %1) {
+; RV32IM-LABEL: select_redundant_czero_eqz2:
+; RV32IM:       # %bb.0: # %entry
+; RV32IM-NEXT:    bnez a0, .LBB50_2
+; RV32IM-NEXT:  # %bb.1: # %entry
+; RV32IM-NEXT:    lui a0, %hi(select_redundant_czero_eqz_data)
+; RV32IM-NEXT:    addi a0, a0, %lo(select_redundant_czero_eqz_data)
+; RV32IM-NEXT:  .LBB50_2: # %entry
+; RV32IM-NEXT:    sw a0, 0(a1)
+; RV32IM-NEXT:    ret
+;
+; RV64IM-LABEL: select_redundant_czero_eqz2:
+; RV64IM:       # %bb.0: # %entry
+; RV64IM-NEXT:    bnez a0, .LBB50_2
+; RV64IM-NEXT:  # %bb.1: # %entry
+; RV64IM-NEXT:    lui a0, %hi(select_redundant_czero_eqz_data)
+; RV64IM-NEXT:    addi a0, a0, %lo(select_redundant_czero_eqz_data)
+; RV64IM-NEXT:  .LBB50_2: # %entry
+; RV64IM-NEXT:    sd a0, 0(a1)
+; RV64IM-NEXT:    ret
+;
+; RV64IMXVTCONDOPS-LABEL: select_redundant_czero_eqz2:
+; RV64IMXVTCONDOPS:       # %bb.0: # %entry
+; RV64IMXVTCONDOPS-NEXT:    lui a2, %hi(select_redundant_czero_eqz_data)
+; RV64IMXVTCONDOPS-NEXT:    addi a2, a2, %lo(select_redundant_czero_eqz_data)
+; RV64IMXVTCONDOPS-NEXT:    vt.maskcn a2, a2, a0
+; RV64IMXVTCONDOPS-NEXT:    or a0, a0, a2
+; RV64IMXVTCONDOPS-NEXT:    sd a0, 0(a1)
+; RV64IMXVTCONDOPS-NEXT:    ret
+;
+; RV32IMZICOND-LABEL: select_redundant_czero_eqz2:
+; RV32IMZICOND:       # %bb.0: # %entry
+; RV32IMZICOND-NEXT:    lui a2, %hi(select_redundant_czero_eqz_data)
+; RV32IMZICOND-NEXT:    addi a2, a2, %lo(select_redundant_czero_eqz_data)
+; RV32IMZICOND-NEXT:    czero.nez a2, a2, a0
+; RV32IMZICOND-NEXT:    or a0, a0, a2
+; RV32IMZICOND-NEXT:    sw a0, 0(a1)
+; RV32IMZICOND-NEXT:    ret
+;
+; RV64IMZICOND-LABEL: select_redundant_czero_eqz2:
+; RV64IMZICOND:       # %bb.0: # %entry
+; RV64IMZICOND-NEXT:    lui a2, %hi(select_redundant_czero_eqz_data)
+; RV64IMZICOND-NEXT:    addi a2, a2, %lo(select_redundant_czero_eqz_data)
+; RV64IMZICOND-NEXT:    czero.nez a2, a2, a0
+; RV64IMZICOND-NEXT:    or a0, a0, a2
+; RV64IMZICOND-NEXT:    sd a0, 0(a1)
+; RV64IMZICOND-NEXT:    ret
+entry:
+  %3 = icmp ne ptr %0, null
+  %4 = select i1 %3, ptr %0, ptr @select_redundant_czero_eqz_data
+  store ptr %4, ptr %1, align 8
+  ret void
+}
diff --git a/llvm/test/CodeGen/RISCV/xaluo.ll b/llvm/test/CodeGen/RISCV/xaluo.ll
index ac67c0769f70..b1efe53290e8 100644
--- a/llvm/test/CodeGen/RISCV/xaluo.ll
+++ b/llvm/test/CodeGen/RISCV/xaluo.ll
@@ -1268,12 +1268,13 @@ define zeroext i1 @smulo2.i32(i32 signext %v1, ptr %res) {
 ; RV32ZBA-LABEL: smulo2.i32:
 ; RV32ZBA:       # %bb.0: # %entry
 ; RV32ZBA-NEXT:    li a2, 13
-; RV32ZBA-NEXT:    mulh a3, a0, a2
-; RV32ZBA-NEXT:    mul a2, a0, a2
-; RV32ZBA-NEXT:    srai a0, a2, 31
-; RV32ZBA-NEXT:    xor a0, a3, a0
+; RV32ZBA-NEXT:    mulh a2, a0, a2
+; RV32ZBA-NEXT:    sh1add a3, a0, a0
+; RV32ZBA-NEXT:    sh2add a3, a3, a0
+; RV32ZBA-NEXT:    srai a0, a3, 31
+; RV32ZBA-NEXT:    xor a0, a2, a0
 ; RV32ZBA-NEXT:    snez a0, a0
-; RV32ZBA-NEXT:    sw a2, 0(a1)
+; RV32ZBA-NEXT:    sw a3, 0(a1)
 ; RV32ZBA-NEXT:    ret
 ;
 ; RV64ZBA-LABEL: smulo2.i32:
@@ -1577,13 +1578,15 @@ define zeroext i1 @smulo2.i64(i64 %v1, ptr %res) {
 ; RV32ZBA:       # %bb.0: # %entry
 ; RV32ZBA-NEXT:    li a3, 13
 ; RV32ZBA-NEXT:    mulhu a4, a0, a3
-; RV32ZBA-NEXT:    mul a5, a1, a3
+; RV32ZBA-NEXT:    sh1add a5, a1, a1
+; RV32ZBA-NEXT:    sh2add a5, a5, a1
 ; RV32ZBA-NEXT:    add a4, a5, a4
 ; RV32ZBA-NEXT:    sltu a5, a4, a5
 ; RV32ZBA-NEXT:    mulhu a6, a1, a3
 ; RV32ZBA-NEXT:    add a5, a6, a5
 ; RV32ZBA-NEXT:    srai a1, a1, 31
-; RV32ZBA-NEXT:    mul a6, a1, a3
+; RV32ZBA-NEXT:    sh1add a6, a1, a1
+; RV32ZBA-NEXT:    sh2add a6, a6, a1
 ; RV32ZBA-NEXT:    add a6, a5, a6
 ; RV32ZBA-NEXT:    srai a7, a4, 31
 ; RV32ZBA-NEXT:    xor t0, a6, a7
@@ -1593,7 +1596,8 @@ define zeroext i1 @smulo2.i64(i64 %v1, ptr %res) {
 ; RV32ZBA-NEXT:    xor a1, a1, a7
 ; RV32ZBA-NEXT:    or a1, t0, a1
 ; RV32ZBA-NEXT:    snez a1, a1
-; RV32ZBA-NEXT:    mul a0, a0, a3
+; RV32ZBA-NEXT:    sh1add a3, a0, a0
+; RV32ZBA-NEXT:    sh2add a0, a3, a0
 ; RV32ZBA-NEXT:    sw a0, 0(a2)
 ; RV32ZBA-NEXT:    sw a4, 4(a2)
 ; RV32ZBA-NEXT:    mv a0, a1
@@ -1602,12 +1606,13 @@ define zeroext i1 @smulo2.i64(i64 %v1, ptr %res) {
 ; RV64ZBA-LABEL: smulo2.i64:
 ; RV64ZBA:       # %bb.0: # %entry
 ; RV64ZBA-NEXT:    li a2, 13
-; RV64ZBA-NEXT:    mulh a3, a0, a2
-; RV64ZBA-NEXT:    mul a2, a0, a2
-; RV64ZBA-NEXT:    srai a0, a2, 63
-; RV64ZBA-NEXT:    xor a0, a3, a0
+; RV64ZBA-NEXT:    mulh a2, a0, a2
+; RV64ZBA-NEXT:    sh1add a3, a0, a0
+; RV64ZBA-NEXT:    sh2add a3, a3, a0
+; RV64ZBA-NEXT:    srai a0, a3, 63
+; RV64ZBA-NEXT:    xor a0, a2, a0
 ; RV64ZBA-NEXT:    snez a0, a0
-; RV64ZBA-NEXT:    sd a2, 0(a1)
+; RV64ZBA-NEXT:    sd a3, 0(a1)
 ; RV64ZBA-NEXT:    ret
 ;
 ; RV32ZICOND-LABEL: smulo2.i64:
@@ -1743,19 +1748,20 @@ define zeroext i1 @umulo2.i32(i32 signext %v1, ptr %res) {
 ;
 ; RV32ZBA-LABEL: umulo2.i32:
 ; RV32ZBA:       # %bb.0: # %entry
-; RV32ZBA-NEXT:    li a3, 13
-; RV32ZBA-NEXT:    mulhu a2, a0, a3
+; RV32ZBA-NEXT:    li a2, 13
+; RV32ZBA-NEXT:    mulhu a2, a0, a2
 ; RV32ZBA-NEXT:    snez a2, a2
-; RV32ZBA-NEXT:    mul a0, a0, a3
+; RV32ZBA-NEXT:    sh1add a3, a0, a0
+; RV32ZBA-NEXT:    sh2add a0, a3, a0
 ; RV32ZBA-NEXT:    sw a0, 0(a1)
 ; RV32ZBA-NEXT:    mv a0, a2
 ; RV32ZBA-NEXT:    ret
 ;
 ; RV64ZBA-LABEL: umulo2.i32:
 ; RV64ZBA:       # %bb.0: # %entry
-; RV64ZBA-NEXT:    zext.w a0, a0
-; RV64ZBA-NEXT:    sh1add a2, a0, a0
-; RV64ZBA-NEXT:    sh2add a2, a2, a0
+; RV64ZBA-NEXT:    zext.w a2, a0
+; RV64ZBA-NEXT:    sh1add.uw a0, a0, a2
+; RV64ZBA-NEXT:    sh2add a2, a0, a2
 ; RV64ZBA-NEXT:    srli a0, a2, 32
 ; RV64ZBA-NEXT:    snez a0, a0
 ; RV64ZBA-NEXT:    sw a2, 0(a1)
@@ -1995,25 +2001,28 @@ define zeroext i1 @umulo2.i64(i64 %v1, ptr %res) {
 ; RV32ZBA-LABEL: umulo2.i64:
 ; RV32ZBA:       # %bb.0: # %entry
 ; RV32ZBA-NEXT:    li a3, 13
-; RV32ZBA-NEXT:    mul a4, a1, a3
-; RV32ZBA-NEXT:    mulhu a5, a0, a3
-; RV32ZBA-NEXT:    add a4, a5, a4
-; RV32ZBA-NEXT:    sltu a5, a4, a5
+; RV32ZBA-NEXT:    mulhu a4, a0, a3
+; RV32ZBA-NEXT:    sh1add a5, a1, a1
+; RV32ZBA-NEXT:    sh2add a5, a5, a1
+; RV32ZBA-NEXT:    add a5, a4, a5
+; RV32ZBA-NEXT:    sltu a4, a5, a4
 ; RV32ZBA-NEXT:    mulhu a1, a1, a3
 ; RV32ZBA-NEXT:    snez a1, a1
-; RV32ZBA-NEXT:    or a1, a1, a5
-; RV32ZBA-NEXT:    mul a0, a0, a3
+; RV32ZBA-NEXT:    or a1, a1, a4
+; RV32ZBA-NEXT:    sh1add a3, a0, a0
+; RV32ZBA-NEXT:    sh2add a0, a3, a0
 ; RV32ZBA-NEXT:    sw a0, 0(a2)
-; RV32ZBA-NEXT:    sw a4, 4(a2)
+; RV32ZBA-NEXT:    sw a5, 4(a2)
 ; RV32ZBA-NEXT:    mv a0, a1
 ; RV32ZBA-NEXT:    ret
 ;
 ; RV64ZBA-LABEL: umulo2.i64:
 ; RV64ZBA:       # %bb.0: # %entry
-; RV64ZBA-NEXT:    li a3, 13
-; RV64ZBA-NEXT:    mulhu a2, a0, a3
+; RV64ZBA-NEXT:    li a2, 13
+; RV64ZBA-NEXT:    mulhu a2, a0, a2
 ; RV64ZBA-NEXT:    snez a2, a2
-; RV64ZBA-NEXT:    mul a0, a0, a3
+; RV64ZBA-NEXT:    sh1add a3, a0, a0
+; RV64ZBA-NEXT:    sh2add a0, a3, a0
 ; RV64ZBA-NEXT:    sd a0, 0(a1)
 ; RV64ZBA-NEXT:    mv a0, a2
 ; RV64ZBA-NEXT:    ret
diff --git a/llvm/test/CodeGen/SPIRV/pointers/argument-ptr-to-struct.ll b/llvm/test/CodeGen/SPIRV/pointers/argument-ptr-to-struct.ll
new file mode 100644
index 000000000000..ac72ec28c37d
--- /dev/null
+++ b/llvm/test/CodeGen/SPIRV/pointers/argument-ptr-to-struct.ll
@@ -0,0 +1,37 @@
+; RUN: llc -O0 -mtriple=spirv64-unknown-unknown %s -o - | FileCheck %s
+; RUN: %if spirv-tools %{ llc -O0 -mtriple=spirv64-unknown-unknown %s -o - -filetype=obj | spirv-val %}
+
+; CHECK-DAG: %[[#VOID:]] = OpTypeVoid
+; CHECK-DAG: %[[#FLOAT:]] = OpTypeFloat 32
+; CHECK-DAG: %[[#UCHAR:]] = OpTypeInt 8 0
+; CHECK-DAG: %[[#UINT:]] = OpTypeInt 32 0
+; CHECK-DAG: %[[#STRUCT_S:]] = OpTypeStruct %[[#FLOAT]] %[[#UCHAR]] %[[#UINT]]
+; CHECK-DAG: %[[#PTR_STRUCT_S:]] = OpTypePointer Function %[[#STRUCT_S]]
+; CHECK-DAG: %[[#FUNC_TYPE_K:]] = OpTypeFunction %[[#VOID]] %[[#PTR_STRUCT_S]]
+; CHECK-DAG: %[[#FUNC_TYPE_H:]] = OpTypeFunction %[[#UINT]] %[[#PTR_STRUCT_S]]
+
+; CHECK: %[[#]] = OpFunction %[[#VOID]] None %[[#FUNC_TYPE_K]]
+; CHECK: %[[#]] = OpFunctionParameter %[[#PTR_STRUCT_S]]
+
+; CHECK: %[[#]] = OpFunction %[[#UINT]] None %[[#FUNC_TYPE_H]]
+; CHECK: %[[#]] = OpFunctionParameter %[[#PTR_STRUCT_S]]
+
+%struct.s = type { float, i8, i32 }
+
+define spir_kernel void @k(ptr noundef byval(%struct.s) align 4 %x) {
+entry:
+  %c = getelementptr inbounds %struct.s, ptr %x, i32 0, i32 2
+  %l = load i32, ptr %c, align 4
+  %add = add nsw i32 %l, 1
+  %c1 = getelementptr inbounds %struct.s, ptr %x, i32 0, i32 2
+  store i32 %add, ptr %c1, align 4
+  ret void
+}
+
+define spir_func i32 @h(ptr noundef byval(%struct.s) align 4 %x) {
+entry:
+  %c = getelementptr inbounds %struct.s, ptr %x, i32 0, i32 2
+  %l = load i32, ptr %c, align 4
+  %add = add nsw i32 %l, 1
+  ret i32 %add
+}
diff --git a/llvm/test/CodeGen/SPIRV/pointers/global-ptrtoint.ll b/llvm/test/CodeGen/SPIRV/pointers/global-ptrtoint.ll
new file mode 100644
index 000000000000..d0c64b4353ec
--- /dev/null
+++ b/llvm/test/CodeGen/SPIRV/pointers/global-ptrtoint.ll
@@ -0,0 +1,28 @@
+; This test is to check that correct virtual register type is created after ptrtoint.
+
+; RUN: llc -O0 -mtriple=spirv64-unknown-unknown %s -o - | FileCheck %s
+; RUN: %if spirv-tools %{ llc -O0 -mtriple=spirv64-unknown-unknown %s -o - -filetype=obj | spirv-val %}
+
+; RUN: llc -O0 -mtriple=spirv32-unknown-unknown %s -o - | FileCheck %s
+; RUN: %if spirv-tools %{ llc -O0 -mtriple=spirv32-unknown-unknown %s -o - -filetype=obj | spirv-val %}
+
+; CHECK: OpName %[[GlobalValue:.*]] "dev_global"
+; CHECK-DAG: %[[TyI64:.*]] = OpTypeInt 64 0
+; CHECK-DAG: %[[TyStruct:.*]] = OpTypeStruct %[[TyI64]] %[[TyI64]]
+; CHECK-DAG: %[[Const128:.*]] = OpConstant %[[TyI64]] 128
+; CHECK-DAG: %[[GlobalValue]] = OpVariable
+; CHECK-DAG: %[[PtrToInt:.*]] = OpSpecConstantOp %[[TyI64]] 117 %12
+; TODO: The following bitcast line looks unneeded and we may expect it to be removed in future
+; CHECK-DAG: %[[UseGlobalValue:.*]] = OpSpecConstantOp %[[TyI64]] 124 %[[PtrToInt]]
+; CHECK-DAG: %[[ConstComposite:.*]] = OpConstantComposite %[[TyStruct]] %[[Const128]] %[[UseGlobalValue]]
+; CHECK-DAG: %[[TyPtrStruct:.*]] = OpTypePointer CrossWorkgroup %[[TyStruct]]
+; CHECK: OpVariable %[[TyPtrStruct]] CrossWorkgroup %[[ConstComposite]]
+; CHECK: OpFunction
+
+@dev_global = addrspace(1) global [2 x i32] zeroinitializer
+@__AsanDeviceGlobalMetadata = addrspace(1) global { i64, i64 } { i64 128, i64 ptrtoint (ptr addrspace(1) @dev_global to i64) }
+
+define void @foo() {
+entry:
+  ret void
+}
diff --git a/llvm/test/CodeGen/SPIRV/pointers/variables-storage-class.ll b/llvm/test/CodeGen/SPIRV/pointers/variables-storage-class.ll
new file mode 100644
index 000000000000..034feed72dc7
--- /dev/null
+++ b/llvm/test/CodeGen/SPIRV/pointers/variables-storage-class.ll
@@ -0,0 +1,19 @@
+; RUN: llc -O0 -mtriple=spirv64-unknown-unknown %s -o - | FileCheck %s
+; RUN: %if spirv-tools %{ llc -O0 -mtriple=spirv64-unknown-unknown %s -o - -filetype=obj | spirv-val %}
+
+@Ptr = addrspace(1) global ptr addrspace(1) null
+@Init = private addrspace(2) constant i32 123
+
+; CHECK-DAG: %[[#PTR:]] = OpVariable %[[#]] UniformConstant %[[#]]
+; CHECK-DAG: %[[#INIT:]] = OpVariable %[[#]] CrossWorkgroup %[[#]]
+
+; CHECK: %[[#]] = OpLoad %[[#]] %[[#INIT]] Aligned 8
+; CHECK: OpCopyMemorySized %[[#]] %[[#PTR]] %[[#]] Aligned 4
+
+define spir_kernel void @Foo() {
+  %l = load ptr addrspace(1), ptr addrspace(1) @Ptr, align 8
+  call void @llvm.memcpy.p1.p2.i64(ptr addrspace(1) align 4 %l, ptr addrspace(2) align 1 @Init, i64 4, i1 false)
+  ret void
+}
+
+declare void @llvm.memcpy.p1.p2.i64(ptr addrspace(1) noalias nocapture writeonly, ptr addrspace(2) noalias nocapture readonly, i64, i1 immarg)
diff --git a/llvm/test/CodeGen/SPIRV/printf.ll b/llvm/test/CodeGen/SPIRV/printf.ll
new file mode 100644
index 000000000000..483fc1f244e5
--- /dev/null
+++ b/llvm/test/CodeGen/SPIRV/printf.ll
@@ -0,0 +1,40 @@
+; RUN: llc -O0 -mtriple=spirv64-unknown-unknown %s -o - | FileCheck %s
+; RUN: %if spirv-tools %{ llc -O0 -mtriple=spirv64-unknown-unknown %s -o - -filetype=obj | spirv-val %}
+
+; RUN: llc -O0 -mtriple=spirv32-unknown-unknown %s -o - | FileCheck %s
+; RUN: %if spirv-tools %{ llc -O0 -mtriple=spirv32-unknown-unknown %s -o - -filetype=obj | spirv-val %}
+
+; CHECK: %[[#ExtImport:]] = OpExtInstImport "OpenCL.std"
+; CHECK: %[[#Char:]] = OpTypeInt 8 0
+; CHECK: %[[#CharPtr:]] = OpTypePointer UniformConstant %[[#Char]]
+; CHECK: %[[#GV:]] = OpVariable %[[#]] UniformConstant %[[#]]
+; CHECK: OpFunction
+; CHECK: %[[#Arg1:]] = OpFunctionParameter
+; CHECK: %[[#Arg2:]] = OpFunctionParameter
+; CHECK: %[[#CastedGV:]] = OpBitcast %[[#CharPtr]] %[[#GV]]
+; CHECK-NEXT: OpExtInst %[[#]] %[[#ExtImport]] printf %[[#CastedGV]] %[[#ArgConst:]]
+; CHECK-NEXT: OpExtInst %[[#]] %[[#ExtImport]] printf %[[#CastedGV]] %[[#ArgConst]]
+; CHECK-NEXT: OpExtInst %[[#]] %[[#ExtImport]] printf %[[#Arg1]] %[[#ArgConst:]]
+; CHECK-NEXT: OpExtInst %[[#]] %[[#ExtImport]] printf %[[#Arg1]] %[[#ArgConst]]
+; CHECK-NEXT: %[[#CastedArg2:]] = OpBitcast %[[#CharPtr]] %[[#Arg2]]
+; CHECK-NEXT: OpExtInst %[[#]] %[[#ExtImport]] printf %[[#CastedArg2]] %[[#ArgConst]]
+; CHECK-NEXT: OpExtInst %[[#]] %[[#ExtImport]] printf %[[#CastedArg2]] %[[#ArgConst]]
+; CHECK: OpFunctionEnd
+
+%struct = type { [6 x i8] }
+
+@FmtStr = internal addrspace(2) constant [6 x i8] c"c=%c\0A\00", align 1
+
+define spir_kernel void @foo(ptr addrspace(2) %_arg_fmt1, ptr addrspace(2) byval(%struct) %_arg_fmt2) {
+entry:
+  %r1 = tail call spir_func i32 (ptr addrspace(2), ...) @_Z6printfPU3AS2Kcz(ptr addrspace(2) @FmtStr, i8 signext 97)
+  %r2 = tail call spir_func i32 (ptr addrspace(2), ...) @_Z18__spirv_ocl_printfPU3AS2Kcz(ptr addrspace(2) @FmtStr, i8 signext 97)
+  %r3 = tail call spir_func i32 (ptr addrspace(2), ...) @_Z6printfPU3AS2Kcz(ptr addrspace(2) %_arg_fmt1, i8 signext 97)
+  %r4 = tail call spir_func i32 (ptr addrspace(2), ...) @_Z18__spirv_ocl_printfPU3AS2Kcz(ptr addrspace(2) %_arg_fmt1, i8 signext 97)
+  %r5 = tail call spir_func i32 (ptr addrspace(2), ...) @_Z6printfPU3AS2Kcz(ptr addrspace(2) %_arg_fmt2, i8 signext 97)
+  %r6 = tail call spir_func i32 (ptr addrspace(2), ...) @_Z18__spirv_ocl_printfPU3AS2Kcz(ptr addrspace(2) %_arg_fmt2, i8 signext 97)
+  ret void
+}
+
+declare dso_local spir_func i32 @_Z6printfPU3AS2Kcz(ptr addrspace(2), ...)
+declare dso_local spir_func i32 @_Z18__spirv_ocl_printfPU3AS2Kcz(ptr addrspace(2), ...)
diff --git a/llvm/test/CodeGen/SPIRV/transcoding/memcpy-zext.ll b/llvm/test/CodeGen/SPIRV/transcoding/memcpy-zext.ll
index ea0197548a81..89fa93b4fcda 100644
--- a/llvm/test/CodeGen/SPIRV/transcoding/memcpy-zext.ll
+++ b/llvm/test/CodeGen/SPIRV/transcoding/memcpy-zext.ll
@@ -3,8 +3,7 @@
 ; RUN: llc -O0 -mtriple=spirv64-unknown-unknown %s -o - | FileCheck %s --check-prefixes=CHECK,CHECK-64
 ; RUN: %if spirv-tools %{ llc -O0 -mtriple=spirv64-unknown-unknown %s -o - -filetype=obj | spirv-val %}
 
-; CHECK-64-DAG: %[[#i64:]] = OpTypeInt 64 0
-
+; CHECK-DAG:    %[[#i64:]] = OpTypeInt 64 0
 ; CHECK-DAG:    %[[#i8:]] = OpTypeInt 8 0
 ; CHECK-DAG:    %[[#i32:]] = OpTypeInt 32 0
 ; CHECK-DAG:    %[[#one:]] = OpConstant %[[#i32]] 1
@@ -14,19 +13,28 @@
 ; CHECK-DAG:    %[[#test_arr_init:]] = OpConstantComposite %[[#i32x3]] %[[#one]] %[[#two]] %[[#three]]
 ; CHECK-DAG:    %[[#szconst1024:]] = OpConstant %[[#i32]] 1024
 ; CHECK-DAG:    %[[#szconst42:]] = OpConstant %[[#i8]] 42
+; CHECK-DAG:    %[[#szconst123:]] = OpConstant %[[#i64]] 123
 ; CHECK-DAG:    %[[#const_i32x3_ptr:]] = OpTypePointer UniformConstant %[[#i32x3]]
 ; CHECK-DAG:    %[[#test_arr:]] = OpVariable %[[#const_i32x3_ptr]] UniformConstant %[[#test_arr_init]]
 ; CHECK-DAG:    %[[#i32x3_ptr:]] = OpTypePointer Function %[[#i32x3]]
 ; CHECK:        %[[#arr:]] = OpVariable %[[#i32x3_ptr]] Function
 
 ; CHECK-32:     OpCopyMemorySized %[[#arr]] %[[#test_arr]] %[[#szconst1024]]
-; CHECK-64:     %[[#szconstext1024:]] = OpUConvert %[[#i64:]] %[[#szconst1024:]]
-; CHECK-64:     OpCopyMemorySized %[[#arr]] %[[#test_arr]] %[[#szconstext1024]]
-
 ; CHECK-32:     %[[#szconstext42:]] = OpUConvert %[[#i32:]] %[[#szconst42:]]
 ; CHECK-32:     OpCopyMemorySized %[[#arr]] %[[#test_arr]] %[[#szconstext42]]
+; CHECK-32:     OpCopyMemorySized %[[#arr]] %[[#test_arr]] %[[#szconst123]]
+
+; If/when Backend stoped rewrite actual reg size of i8/i16/i32/i64 with i32,
+; i32 = G_TRUNC i64 would appear for the 32-bit target, switching the following
+; TODO patterns instead of the last line above.
+; TODO:         %[[#szconstext123:]] = OpUConvert %[[#i32:]] %[[#szconst123:]]
+; TODO:         OpCopyMemorySized %[[#arr]] %[[#test_arr]] %[[#szconst123]]
+
+; CHECK-64:     %[[#szconstext1024:]] = OpUConvert %[[#i64:]] %[[#szconst1024:]]
+; CHECK-64:     OpCopyMemorySized %[[#arr]] %[[#test_arr]] %[[#szconstext1024]]
 ; CHECK-64:     %[[#szconstext42:]] = OpUConvert %[[#i64:]] %[[#szconst42:]]
 ; CHECK-64:     OpCopyMemorySized %[[#arr]] %[[#test_arr]] %[[#szconstext42]]
+; CHECK-64:     OpCopyMemorySized %[[#arr]] %[[#test_arr]] %[[#szconst123]]
 
 @__const.test.arr = private unnamed_addr addrspace(2) constant [3 x i32] [i32 1, i32 2, i32 3]
 
@@ -36,8 +44,10 @@ entry:
   %dest = bitcast ptr %arr to ptr
   call void @llvm.memcpy.p0.p2.i32(ptr align 4 %dest, ptr addrspace(2) align 4 @__const.test.arr, i32 1024, i1 false)
   call void @llvm.memcpy.p0.p2.i8(ptr align 4 %dest, ptr addrspace(2) align 4 @__const.test.arr, i8 42, i1 false)
+  call void @llvm.memcpy.p0.p2.i64(ptr align 4 %dest, ptr addrspace(2) align 4 @__const.test.arr, i64 123, i1 false)
   ret void
 }
 
 declare void @llvm.memcpy.p0.p2.i32(ptr nocapture writeonly, ptr addrspace(2) nocapture readonly, i32, i1)
 declare void @llvm.memcpy.p0.p2.i8(ptr nocapture writeonly, ptr addrspace(2) nocapture readonly, i8, i1)
+declare void @llvm.memcpy.p0.p2.i64(ptr nocapture writeonly, ptr addrspace(2) nocapture readonly, i64, i1)
diff --git a/llvm/test/CodeGen/SPIRV/transcoding/spirv-event-null.ll b/llvm/test/CodeGen/SPIRV/transcoding/spirv-event-null.ll
new file mode 100644
index 000000000000..fe0d96f2773e
--- /dev/null
+++ b/llvm/test/CodeGen/SPIRV/transcoding/spirv-event-null.ll
@@ -0,0 +1,33 @@
+; RUN: llc -O0 -mtriple=spirv32-unknown-unknown %s -o - | FileCheck %s
+; RUN: %if spirv-tools %{ llc -O0 -mtriple=spirv32-unknown-unknown %s -o - -filetype=obj | spirv-val %}
+
+; RUN: llc -O0 -mtriple=spirv64-unknown-unknown %s -o - | FileCheck %s
+; RUN: %if spirv-tools %{ llc -O0 -mtriple=spirv64-unknown-unknown %s -o - -filetype=obj | spirv-val %}
+
+; CHECK-DAG: %[[#TyEvent:]] = OpTypeEvent
+; CHECK-DAG: %[[#TyStruct:]] = OpTypeStruct %[[#TyEvent]]
+; CHECK-DAG: %[[#ConstEvent:]] = OpConstantNull %[[#TyEvent]]
+; CHECK-DAG: %[[#TyEventPtr:]] = OpTypePointer Function %[[#TyEvent]]
+; CHECK-DAG: %[[#TyStructPtr:]] = OpTypePointer Function %[[#TyStruct]]
+; CHECK: OpFunction
+; CHECK: OpFunctionParameter
+; CHECK: %[[#Src:]] = OpFunctionParameter
+; CHECK: OpVariable %[[#TyStructPtr]] Function
+; CHECK: %[[#EventVar:]] = OpVariable %[[#TyEventPtr]] Function
+; CHECK: %[[#Dest:]] = OpInBoundsPtrAccessChain
+; CHECK: %[[#CopyRes:]] = OpGroupAsyncCopy %[[#TyEvent]] %[[#]] %[[#Dest]] %[[#Src]] %[[#]] %[[#]] %[[#ConstEvent]]
+; CHECK: OpStore %[[#EventVar]] %[[#CopyRes]]
+
+%"class.sycl::_V1::device_event" = type { target("spirv.Event") }
+
+define spir_kernel void @foo(ptr addrspace(1) %_arg_out_ptr, ptr addrspace(3) noundef %_arg_local_acc) {
+entry:
+  %var = alloca %"class.sycl::_V1::device_event"
+  %dev_event.i.sroa.0 = alloca target("spirv.Event")
+  %add.ptr.i26 = getelementptr inbounds i32, ptr addrspace(1) %_arg_out_ptr, i64 0
+  %call3.i = tail call spir_func target("spirv.Event") @_Z22__spirv_GroupAsyncCopyjPU3AS1iPU3AS3Kimm9ocl_event(i32 2, ptr addrspace(1) %add.ptr.i26, ptr addrspace(3) %_arg_local_acc, i64 16, i64 10, target("spirv.Event") zeroinitializer)
+  store target("spirv.Event") %call3.i, ptr %dev_event.i.sroa.0
+  ret void
+}
+
+declare dso_local spir_func target("spirv.Event") @_Z22__spirv_GroupAsyncCopyjPU3AS1iPU3AS3Kimm9ocl_event(i32, ptr addrspace(1), ptr addrspace(3), i64, i64, target("spirv.Event"))
diff --git a/llvm/test/CodeGen/SPIRV/types/or-i1.ll b/llvm/test/CodeGen/SPIRV/types/or-i1.ll
new file mode 100644
index 000000000000..32dc258d8555
--- /dev/null
+++ b/llvm/test/CodeGen/SPIRV/types/or-i1.ll
@@ -0,0 +1,17 @@
+; RUN: llc -O0 -mtriple=spirv64-unknown-unknown %s -o - | FileCheck %s
+; RUN: %if spirv-tools %{ llc -O0 -mtriple=spirv64-unknown-unknown %s -o - -filetype=obj | spirv-val %}
+
+; CHECK-DAG: %[[#BOOL:]] = OpTypeBool
+; CHECK: %[[#BOOL:]] = OpLogicalOr %[[#BOOL]] %[[#]] %[[#]]
+
+define spir_kernel void @foo(
+  ptr addrspace(1) nocapture noundef writeonly %Dst,
+  i32 %a, i32 %b) local_unnamed_addr #0 {
+entry:
+  %a1 = trunc i32 %a to i1
+  %b1 = trunc i32 %b to i1
+  %ab1 = or i1 %a1, %b1
+  %ab32 = zext i1 %ab1 to i32
+  store i32 %ab32, ptr addrspace(1) %Dst
+  ret void
+}
diff --git a/llvm/test/CodeGen/SystemZ/atomic-load-08.ll b/llvm/test/CodeGen/SystemZ/atomic-load-08.ll
index 4d914e3ea0e1..83050ef87591 100644
--- a/llvm/test/CodeGen/SystemZ/atomic-load-08.ll
+++ b/llvm/test/CodeGen/SystemZ/atomic-load-08.ll
@@ -2,8 +2,8 @@
 ; loads with a bitcast, and this test case gets converted into that form as
 ; well by the AtomicExpand pass.
 ;
-; RUN: llc < %s -mtriple=s390x-linux-gnu | FileCheck %s
-; RUN: llc < %s -mtriple=s390x-linux-gnu -mcpu=z13 | FileCheck %s
+; RUN: llc < %s -mtriple=s390x-linux-gnu | FileCheck -check-prefixes=CHECK,BASE %s
+; RUN: llc < %s -mtriple=s390x-linux-gnu -mcpu=z13 | FileCheck -check-prefixes=CHECK,Z13 %s
 
 define void @f1(ptr %ret, ptr %src) {
 ; CHECK-LABEL: f1:
@@ -17,6 +17,34 @@ define void @f1(ptr %ret, ptr %src) {
   ret void
 }
 
+define void @f1_fpuse(ptr %ret, ptr %src) {
+; CHECK-LABEL: f1_fpuse:
+; CHECK:       # %bb.0:
+; BASE-NEXT: aghi	%r15, -176
+; BASE-NEXT: .cfi_def_cfa_offset 336
+
+; CHECK-NEXT:	lpq	%r0, 0(%r3)
+
+; BASE-NEXT: stg %r1, 168(%r15)
+; BASE-NEXT: stg %r0, 160(%r15)
+; BASE-NEXT: ld	%f0, 160(%r15)
+; BASE-NEXT: ld	%f2, 168(%r15)
+
+; Z13-NEXT: vlvgp %v0, %r0, %r1
+; Z13-NEXT: vrepg %v2, %v0, 1
+
+; CHECK-NEXT:	axbr	%f0, %f0
+; CHECK-NEXT:	std	%f0, 0(%r2)
+; CHECK-NEXT:	std	%f2, 8(%r2)
+; BASE-NEXT:	aghi	%r15, 176
+; CHECK-NEXT:	br	%r14
+
+  %val = load atomic fp128, ptr %src seq_cst, align 16
+  %use = fadd fp128 %val, %val
+  store fp128 %use, ptr %ret, align 8
+  ret void
+}
+
 define void @f2(ptr %ret, ptr %src) {
 ; CHECK-LABEL: f2:
 ; CHECK: brasl %r14, __atomic_load@PLT
@@ -24,3 +52,19 @@ define void @f2(ptr %ret, ptr %src) {
   store fp128 %val, ptr %ret, align 8
   ret void
 }
+
+define void @f2_fpuse(ptr %ret, ptr %src) {
+; CHECK-LABEL: f2_fpuse:
+; CHECK: brasl %r14, __atomic_load@PLT
+; CHECK-NEXT:   ld	%f0, 160(%r15)
+; CHECK-NEXT:	ld	%f2, 168(%r15)
+; CHECK-NEXT:	axbr	%f0, %f0
+; CHECK-NEXT:	std	%f0, 0(%r13)
+; CHECK-NEXT:	std	%f2, 8(%r13)
+; CHECK-NEXT:	lmg	%r13, %r15, 280(%r15)
+; CHECK-NEXT:	br	%r14
+  %val = load atomic fp128, ptr %src seq_cst, align 8
+  %use = fadd fp128 %val, %val
+  store fp128 %use, ptr %ret, align 8
+  ret void
+}
diff --git a/llvm/test/CodeGen/SystemZ/atomic-store-08.ll b/llvm/test/CodeGen/SystemZ/atomic-store-08.ll
index f7f4f4d967db..4d1693477f01 100644
--- a/llvm/test/CodeGen/SystemZ/atomic-store-08.ll
+++ b/llvm/test/CodeGen/SystemZ/atomic-store-08.ll
@@ -1,8 +1,8 @@
 ; Test long double atomic stores. The atomic store is converted to i128 by
 ; the AtomicExpand pass.
 ;
-; RUN: llc < %s -mtriple=s390x-linux-gnu | FileCheck %s
-; RUN: llc < %s -mtriple=s390x-linux-gnu -mcpu=z13 | FileCheck %s
+; RUN: llc < %s -mtriple=s390x-linux-gnu | FileCheck -check-prefixes=CHECK,BASE %s
+; xUN: llc < %s -mtriple=s390x-linux-gnu -mcpu=z13 | FileCheck -check-prefixes=CHECK,Z13 %s
 
 define void @f1(ptr %dst, ptr %src) {
 ; CHECK-LABEL: f1:
@@ -17,6 +17,29 @@ define void @f1(ptr %dst, ptr %src) {
   ret void
 }
 
+define void @f1_fpsrc(ptr %dst, ptr %src) {
+; CHECK-LABEL: f1_fpsrc:
+; CHECK:       # %bb.0:
+; CHECK-NEXT: ld	%f0, 0(%r3)
+; CHECK-NEXT: ld	%f2, 8(%r3)
+; CHECK-NEXT: axbr	%f0, %f0
+
+; BASE-NEXT: lgdr	%r1, %f2
+; BASE-NEXT: lgdr	%r0, %f0
+
+; Z13-NEXT: vmrhg	%v0, %v0, %v2
+; Z13-NEXT: vlgvg	%r1, %v0, 1
+; Z13-NEXT: vlgvg	%r0, %v0, 0
+
+; CHECK-NEXT: stpq	%r0, 0(%r2)
+; CHECK-NEXT: bcr	15, %r0
+; CHECK-NEXT: br	%r14
+  %val = load fp128, ptr %src, align 8
+  %add = fadd fp128 %val, %val
+  store atomic fp128 %add, ptr %dst seq_cst, align 16
+  ret void
+}
+
 define void @f2(ptr %dst, ptr %src) {
 ; CHECK-LABEL: f2:
 ; CHECK: brasl %r14, __atomic_store@PLT
@@ -24,3 +47,27 @@ define void @f2(ptr %dst, ptr %src) {
   store atomic fp128 %val, ptr %dst seq_cst, align 8
   ret void
 }
+
+define void @f2_fpuse(ptr %dst, ptr %src) {
+; CHECK-LABEL: f2_fpuse:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:	stmg	%r14, %r15, 112(%r15)
+; CHECK-NEXT:	.cfi_offset %r14, -48
+; CHECK-NEXT:	.cfi_offset %r15, -40
+; CHECK-NEXT:	aghi	%r15, -176
+; CHECK-NEXT:	.cfi_def_cfa_offset 336
+; CHECK-NEXT:	ld	%f0, 0(%r3)
+; CHECK-NEXT:	ld	%f2, 8(%r3)
+; CHECK-NEXT:	lgr	%r3, %r2
+; CHECK-NEXT:	axbr	%f0, %f0
+; CHECK-NEXT:	la	%r4, 160(%r15)
+; CHECK-NEXT:	lghi	%r2, 16
+; CHECK-NEXT:	lhi	%r5, 5
+; CHECK-NEXT:	std	%f0, 160(%r15)
+; CHECK-NEXT:	std	%f2, 168(%r15)
+; CHECK-NEXT: brasl %r14, __atomic_store@PLT
+  %val = load fp128, ptr %src, align 8
+  %add = fadd fp128 %val, %val
+  store atomic fp128 %add, ptr %dst seq_cst, align 8
+  ret void
+}
diff --git a/llvm/test/CodeGen/SystemZ/atomicrmw-xchg-07.ll b/llvm/test/CodeGen/SystemZ/atomicrmw-xchg-07.ll
index 18fa89e6ca6c..f5d8dc092a7e 100644
--- a/llvm/test/CodeGen/SystemZ/atomicrmw-xchg-07.ll
+++ b/llvm/test/CodeGen/SystemZ/atomicrmw-xchg-07.ll
@@ -26,3 +26,45 @@ define void @f1(ptr align 16 %ret, ptr align 16 %src, ptr align 16 %b) {
   store fp128 %res, ptr %ret, align 16
   ret void
 }
+
+define void @f1_fpuse(ptr align 16 %ret, ptr align 16 %src, ptr align 16 %b) {
+; CHECK-LABEL: f1_fpuse:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    stmg %r12, %r15, 96(%r15)
+; CHECK-NEXT:    .cfi_offset %r12, -64
+; CHECK-NEXT:    .cfi_offset %r13, -56
+; CHECK-NEXT:    .cfi_offset %r15, -40
+; CHECK-NEXT:    aghi %r15, -176
+; CHECK-NEXT:    .cfi_def_cfa_offset 336
+; CHECK-NEXT:    ld %f0, 0(%r4)
+; CHECK-NEXT:    ld %f2, 8(%r4)
+; CHECK-NEXT:    lg %r0, 8(%r3)
+; CHECK-NEXT:    lg %r1, 0(%r3)
+; CHECK-NEXT:    axbr %f0, %f0
+; CHECK-NEXT:    lgdr %r5, %f2
+; CHECK-NEXT:    lgdr %r4, %f0
+; CHECK-NEXT:  .LBB1_1: # %atomicrmw.start
+; CHECK-NEXT:    # =>This Inner Loop Header: Depth=1
+; CHECK-NEXT:    lgr %r12, %r1
+; CHECK-NEXT:    lgr %r13, %r0
+; CHECK-NEXT:    cdsg %r12, %r4, 0(%r3)
+; CHECK-NEXT:    lgr %r0, %r13
+; CHECK-NEXT:    lgr %r1, %r12
+; CHECK-NEXT:    jl .LBB1_1
+; CHECK-NEXT:  # %bb.2: # %atomicrmw.end
+; CHECK-NEXT:    stg %r1, 160(%r15)
+; CHECK-NEXT:    stg %r0, 168(%r15)
+; CHECK-NEXT:    ld %f0, 160(%r15)
+; CHECK-NEXT:    ld %f2, 168(%r15)
+; CHECK-NEXT:    axbr %f0, %f0
+; CHECK-NEXT:    std %f0, 0(%r2)
+; CHECK-NEXT:    std %f2, 8(%r2)
+; CHECK-NEXT:    lmg %r12, %r15, 272(%r15)
+; CHECK-NEXT:    br %r14
+  %val = load fp128, ptr %b, align 16
+  %add.src = fadd fp128 %val, %val
+  %res = atomicrmw xchg ptr %src, fp128 %add.src seq_cst
+  %res.x2 = fadd fp128 %res, %res
+  store fp128 %res.x2, ptr %ret, align 16
+  ret void
+}
diff --git a/llvm/test/CodeGen/SystemZ/zos-no-eh-label.ll b/llvm/test/CodeGen/SystemZ/zos-no-eh-label.ll
new file mode 100644
index 000000000000..b28f0dd17599
--- /dev/null
+++ b/llvm/test/CodeGen/SystemZ/zos-no-eh-label.ll
@@ -0,0 +1,12 @@
+; RUN: llc -mtriple s390x-ibm-zos < %s | FileCheck %s
+
+define signext i32 @_Z9computeitv() personality ptr @__zos_cxx_personality_v2 {
+  ret i32 0
+}
+
+declare i32 @__zos_cxx_personality_v2(...)
+
+; The personality function is unused, therefore check that it is not referenced.
+; There should also be no exception table.
+; CHECK-NOT: __zos_cxx_personality_v2
+; CHECK-NOT: GCC_except_table
diff --git a/llvm/test/CodeGen/WebAssembly/lower-em-ehsjlj-multi-return.ll b/llvm/test/CodeGen/WebAssembly/lower-em-ehsjlj-multi-return.ll
index 4f33439db770..ae7ad64ffe57 100644
--- a/llvm/test/CodeGen/WebAssembly/lower-em-ehsjlj-multi-return.ll
+++ b/llvm/test/CodeGen/WebAssembly/lower-em-ehsjlj-multi-return.ll
@@ -1,5 +1,5 @@
-; RUN: not --crash llc < %s -enable-emscripten-cxx-exceptions -mattr=+multivalue 2>&1 | FileCheck %s --check-prefix=EH
-; RUN: not --crash llc < %s -enable-emscripten-sjlj -mattr=+multivalue 2>&1 | FileCheck %s --check-prefix=SJLJ
+; RUN: not --crash llc < %s -enable-emscripten-cxx-exceptions -mattr=+multivalue -target-abi=experimental-mv 2>&1 | FileCheck %s --check-prefix=EH
+; RUN: not --crash llc < %s -enable-emscripten-sjlj -mattr=+multivalue -target-abi=experimental-mv 2>&1 | FileCheck %s --check-prefix=SJLJ
 
 ; Currently multivalue returning functions are not supported in Emscripten EH /
 ; SjLj. Make sure they error out.
diff --git a/llvm/test/CodeGen/WebAssembly/multivalue-dont-move-def-past-use.mir b/llvm/test/CodeGen/WebAssembly/multivalue-dont-move-def-past-use.mir
index 4b4661b14466..eb9dfa9dfa60 100644
--- a/llvm/test/CodeGen/WebAssembly/multivalue-dont-move-def-past-use.mir
+++ b/llvm/test/CodeGen/WebAssembly/multivalue-dont-move-def-past-use.mir
@@ -1,5 +1,5 @@
 # NOTE: Assertions have been autogenerated by utils/update_mir_test_checks.py
-# RUN: llc -mtriple=wasm32-unknown-unknown -mattr=+multivalue -run-pass=wasm-reg-stackify -verify-machineinstrs %s -o - | FileCheck %s
+# RUN: llc -mtriple=wasm32-unknown-unknown -mattr=+multivalue -target-abi=experimental-mv -run-pass=wasm-reg-stackify -verify-machineinstrs %s -o - | FileCheck %s
 
 --- |
   target datalayout = "e-m:e-p:32:32-p10:8:8-p20:8:8-i64:64-n32:64-S128-ni:1:10:20"
diff --git a/llvm/test/CodeGen/WebAssembly/multivalue-stackify.ll b/llvm/test/CodeGen/WebAssembly/multivalue-stackify.ll
index 52a8c686824d..0b5a304589aa 100644
--- a/llvm/test/CodeGen/WebAssembly/multivalue-stackify.ll
+++ b/llvm/test/CodeGen/WebAssembly/multivalue-stackify.ll
@@ -1,7 +1,7 @@
 ; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
 ; NOTE: Test functions have been generated by multivalue-stackify.py.
 
-; RUN: llc < %s -verify-machineinstrs -mattr=+multivalue | FileCheck %s
+; RUN: llc < %s -verify-machineinstrs -mattr=+multivalue -target-abi=experimental-mv | FileCheck %s
 
 ; Test that the multivalue stackification works
 
diff --git a/llvm/test/CodeGen/WebAssembly/multivalue.ll b/llvm/test/CodeGen/WebAssembly/multivalue.ll
index 675009c8f3e5..5001db7e57a1 100644
--- a/llvm/test/CodeGen/WebAssembly/multivalue.ll
+++ b/llvm/test/CodeGen/WebAssembly/multivalue.ll
@@ -1,7 +1,8 @@
-; RUN: llc < %s -asm-verbose=false -verify-machineinstrs -mcpu=mvp -mattr=+multivalue,+tail-call | FileCheck %s
-; RUN: llc < %s -asm-verbose=false -verify-machineinstrs -mcpu=mvp -mattr=+reference-types,+multivalue,+tail-call | FileCheck --check-prefix REF %s
-; RUN: llc < %s -asm-verbose=false -verify-machineinstrs -disable-wasm-fallthrough-return-opt -wasm-disable-explicit-locals -wasm-keep-registers -mcpu=mvp -mattr=+multivalue,+tail-call | FileCheck %s --check-prefix REGS
-; RUN: llc < %s --filetype=obj -mcpu=mvp -mattr=+multivalue,+tail-call | obj2yaml | FileCheck %s --check-prefix OBJ
+; RUN: llc < %s -asm-verbose=false -verify-machineinstrs -mcpu=mvp -mattr=+multivalue,+tail-call -target-abi=experimental-mv | FileCheck %s
+; RUN: llc < %s -asm-verbose=false -verify-machineinstrs -mcpu=mvp -mattr=+reference-types,+multivalue,+tail-call -target-abi=experimental-mv | FileCheck --check-prefix REF %s
+; RUN: llc < %s -asm-verbose=false -verify-machineinstrs -disable-wasm-fallthrough-return-opt -wasm-disable-explicit-locals -wasm-keep-registers -mcpu=mvp -mattr=+multivalue,+tail-call -target-abi=experimental-mv | FileCheck %s --check-prefix REGS
+; RUN: llc < %s --filetype=obj -mcpu=mvp -mattr=+multivalue,+tail-call -target-abi=experimental-mv | obj2yaml | FileCheck %s --check-prefix OBJ
+; RUN: llc < %s -asm-verbose=false -verify-machineinstrs -mcpu=mvp -mattr=+multivalue,+tail-call | FileCheck %s --check-prefix NO-MULTIVALUE
 
 ; Test that the multivalue calls, returns, function types, and block
 ; types work as expected.
@@ -19,6 +20,7 @@ declare void @use_i64(i64)
 ; CHECK-NEXT: i32.const 42{{$}}
 ; CHECK-NEXT: i64.const 42{{$}}
 ; CHECK-NEXT: end_function{{$}}
+; NO-MULTIVALUE-NOT: .functype pair_const () -> (i32, i64)
 define %pair @pair_const() {
   ret %pair { i32 42, i64 42 }
 }
diff --git a/llvm/test/CodeGen/WebAssembly/multivalue_libcall.ll b/llvm/test/CodeGen/WebAssembly/multivalue_libcall.ll
index 47c5ae7b457d..2958b115df9d 100644
--- a/llvm/test/CodeGen/WebAssembly/multivalue_libcall.ll
+++ b/llvm/test/CodeGen/WebAssembly/multivalue_libcall.ll
@@ -1,5 +1,5 @@
 ; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --version 2
-; RUN: llc < %s -verify-machineinstrs -mcpu=mvp -mattr=+multivalue | FileCheck %s --check-prefix=MULTIVALUE
+; RUN: llc < %s -verify-machineinstrs -mcpu=mvp -mattr=+multivalue -target-abi=experimental-mv | FileCheck %s --check-prefix=MULTIVALUE
 ; RUN: llc < %s -verify-machineinstrs -mcpu=mvp | FileCheck %s --check-prefix=NO_MULTIVALUE
 
 ; Test libcall signatures when multivalue is enabled and disabled
diff --git a/llvm/test/CodeGen/WebAssembly/ref-type-mem2local.ll b/llvm/test/CodeGen/WebAssembly/ref-type-mem2local.ll
index 4b32a0945e2e..911e5bb516a2 100644
--- a/llvm/test/CodeGen/WebAssembly/ref-type-mem2local.ll
+++ b/llvm/test/CodeGen/WebAssembly/ref-type-mem2local.ll
@@ -1,4 +1,5 @@
 ; RUN: llc < %s -mattr=+reference-types -stop-after=wasm-ref-type-mem2local | FileCheck %s
+; RUN: llc < %s -stop-after=wasm-ref-type-mem2local | FileCheck %s --check-prefix=ATTR
 
 target triple = "wasm32-unknown-unknown"
 
@@ -51,7 +52,40 @@ entry:
   %i32.loaded = load i32, ptr %alloc.i32
   call void @take_i32(i32 %i32.loaded)
   ; CHECK: %alloc.i32 = alloca i32, align 4{{$}}
-  ; CHECK-NOT: addrspace(1)
+  ; CHECK-NOT: alloca i32 {{.*}} addrspace(1)
 
   ret void
 }
+
+; The same function as test_ref_type_mem2local, but here +reference-types is
+; given in the function attribute.
+; Reference type allocas should be moved to addrspace(1)
+; ATTR-LABEL: @test_ref_type_mem2local_func_attr
+define void @test_ref_type_mem2local_func_attr() #0 {
+entry:
+  %alloc.externref = alloca %externref, align 1
+  %eref = call %externref @get_externref()
+  store %externref %eref, ptr %alloc.externref, align 1
+  %eref.loaded = load %externref, ptr %alloc.externref, align 1
+  call void @take_externref(%externref %eref.loaded)
+  ; ATTR:      %alloc.externref.var = alloca ptr addrspace(10), align 1, addrspace(1)
+  ; ATTR-NEXT: %eref = call ptr addrspace(10) @get_externref()
+  ; ATTR-NEXT: store ptr addrspace(10) %eref, ptr addrspace(1) %alloc.externref.var, align 1
+  ; ATTR-NEXT: %eref.loaded = load ptr addrspace(10), ptr addrspace(1) %alloc.externref.var, align 1
+  ; ATTR-NEXT: call void @take_externref(ptr addrspace(10) %eref.loaded)
+
+  %alloc.funcref = alloca %funcref, align 1
+  %fref = call %funcref @get_funcref()
+  store %funcref %fref, ptr %alloc.funcref, align 1
+  %fref.loaded = load %funcref, ptr %alloc.funcref, align 1
+  call void @take_funcref(%funcref %fref.loaded)
+  ; ATTR-NEXT: %alloc.funcref.var = alloca ptr addrspace(20), align 1, addrspace(1)
+  ; ATTR-NEXT: %fref = call ptr addrspace(20) @get_funcref()
+  ; ATTR-NEXT: store ptr addrspace(20) %fref, ptr addrspace(1) %alloc.funcref.var, align 1
+  ; ATTR-NEXT: %fref.loaded = load ptr addrspace(20), ptr addrspace(1) %alloc.funcref.var, align 1
+  ; ATTR-NEXT: call void @take_funcref(ptr addrspace(20) %fref.loaded)
+
+  ret void
+}
+
+attributes #0 = { "target-features"="+reference-types" }
diff --git a/llvm/test/CodeGen/WebAssembly/unreachable.ll b/llvm/test/CodeGen/WebAssembly/unreachable.ll
index 5368c2ba5b8d..ccac31a9af4a 100644
--- a/llvm/test/CodeGen/WebAssembly/unreachable.ll
+++ b/llvm/test/CodeGen/WebAssembly/unreachable.ll
@@ -30,7 +30,6 @@ define void @trap_ret_void() {
 ; CHECK:         .functype trap_ret_void () -> ()
 ; CHECK-NEXT:  # %bb.0:
 ; CHECK-NEXT:    unreachable
-; CHECK-NEXT:    # fallthrough-return
 ; CHECK-NEXT:    end_function
   call void @llvm.trap()
   ret void
@@ -54,7 +53,6 @@ define void @trap_unreacheable() {
 ; CHECK:         .functype trap_unreacheable () -> ()
 ; CHECK-NEXT:  # %bb.0:
 ; CHECK-NEXT:    unreachable
-; CHECK-NEXT:    unreachable
 ; CHECK-NEXT:    end_function
   call void @llvm.trap()
   unreachable
@@ -94,3 +92,12 @@ define i32 @missing_ret_noreturn_unreachable() {
   call void @ext_never_return()
   unreachable
 }
+
+define i32 @no_crash_for_other_instruction_after_trap(ptr %p, i32 %b) {
+; CHECK-LABEL: no_crash_for_other_instruction_after_trap:
+; CHECK:      unreachable
+; CHECK-NEXT: end_function
+  %a = load i32, ptr %p
+  call void @llvm.trap()
+  ret i32 %a
+}
diff --git a/llvm/test/CodeGen/X86/AMX/amx-lower-tile-copy.ll b/llvm/test/CodeGen/X86/AMX/amx-lower-tile-copy.ll
index 4686361ad2fc..a0085afbaf02 100644
--- a/llvm/test/CodeGen/X86/AMX/amx-lower-tile-copy.ll
+++ b/llvm/test/CodeGen/X86/AMX/amx-lower-tile-copy.ll
@@ -44,12 +44,8 @@ define dso_local void @test1(ptr%buf) nounwind {
 ; CHECK-NEXT:    tileloadd 3024(%rsp,%rax), %tmm3 # 1024-byte Folded Reload
 ; CHECK-NEXT:    tileloadd (%rbx,%r15), %tmm0
 ; CHECK-NEXT:    tileloadd (%rbx,%r15), %tmm1
-; CHECK-NEXT:    # implicit-def: $rax
-; CHECK-NEXT:    movq %rax, {{[-0-9]+}}(%r{{[sb]}}p) # 8-byte Spill
-; CHECK-NEXT:    movabsq $64, %rax
 ; CHECK-NEXT:    tilestored %tmm3, 1024(%rsp,%rax) # 1024-byte Folded Spill
 ; CHECK-NEXT:    tileloadd {{[-0-9]+}}(%r{{[sb]}}p), %tmm2 # 1024-byte Folded Reload
-; CHECK-NEXT:    movq {{[-0-9]+}}(%r{{[sb]}}p), %rax # 8-byte Reload
 ; CHECK-NEXT:    tdpbssd %tmm1, %tmm0, %tmm2
 ; CHECK-NEXT:    tilestored %tmm2, (%rbx,%r15)
 ; CHECK-NEXT:    incl %r14d
@@ -111,16 +107,10 @@ define dso_local void @test1(ptr%buf) nounwind {
 ; EGPR-NEXT:    # EVEX TO VEX Compression encoding: [0xc4,0xe2,0x7b,0x4b,0x9c,0x04,0xd0,0x0b,0x00,0x00]
 ; EGPR-NEXT:    tileloadd (%rbx,%r15), %tmm0 # EVEX TO VEX Compression encoding: [0xc4,0xa2,0x7b,0x4b,0x04,0x3b]
 ; EGPR-NEXT:    tileloadd (%rbx,%r15), %tmm1 # EVEX TO VEX Compression encoding: [0xc4,0xa2,0x7b,0x4b,0x0c,0x3b]
-; EGPR-NEXT:    # implicit-def: $rax
-; EGPR-NEXT:    movq %rax, {{[-0-9]+}}(%r{{[sb]}}p) # 8-byte Spill
-; EGPR-NEXT:    # encoding: [0x48,0x89,0x84,0x24,0xb8,0x03,0x00,0x00]
-; EGPR-NEXT:    movabsq $64, %rax # encoding: [0x48,0xb8,0x40,0x00,0x00,0x00,0x00,0x00,0x00,0x00]
 ; EGPR-NEXT:    tilestored %tmm3, 1024(%rsp,%rax) # 1024-byte Folded Spill
 ; EGPR-NEXT:    # EVEX TO VEX Compression encoding: [0xc4,0xe2,0x7a,0x4b,0x9c,0x04,0x00,0x04,0x00,0x00]
 ; EGPR-NEXT:    tileloadd {{[-0-9]+}}(%r{{[sb]}}p), %tmm2 # 1024-byte Folded Reload
 ; EGPR-NEXT:    # EVEX TO VEX Compression encoding: [0xc4,0xe2,0x7b,0x4b,0x94,0x24,0x00,0x04,0x00,0x00]
-; EGPR-NEXT:    movq {{[-0-9]+}}(%r{{[sb]}}p), %rax # 8-byte Reload
-; EGPR-NEXT:    # encoding: [0x48,0x8b,0x84,0x24,0xb8,0x03,0x00,0x00]
 ; EGPR-NEXT:    tdpbssd %tmm1, %tmm0, %tmm2 # encoding: [0xc4,0xe2,0x73,0x5e,0xd0]
 ; EGPR-NEXT:    tilestored %tmm2, (%rbx,%r15) # EVEX TO VEX Compression encoding: [0xc4,0xa2,0x7a,0x4b,0x14,0x3b]
 ; EGPR-NEXT:    incl %r14d # encoding: [0x41,0xff,0xc6]
diff --git a/llvm/test/CodeGen/X86/abdu-vector-128.ll b/llvm/test/CodeGen/X86/abdu-vector-128.ll
index dd180b67e492..0c33e8973c2d 100644
--- a/llvm/test/CodeGen/X86/abdu-vector-128.ll
+++ b/llvm/test/CodeGen/X86/abdu-vector-128.ll
@@ -715,43 +715,41 @@ define <2 x i64> @abd_cmp_v2i64_multiuse_cmp(<2 x i64> %a, <2 x i64> %b) nounwin
 ;
 ; SSE42-LABEL: abd_cmp_v2i64_multiuse_cmp:
 ; SSE42:       # %bb.0:
-; SSE42-NEXT:    movdqa {{.*#+}} xmm2 = [9223372036854775808,9223372036854775808]
+; SSE42-NEXT:    movdqa %xmm0, %xmm2
+; SSE42-NEXT:    psubq %xmm1, %xmm2
 ; SSE42-NEXT:    movdqa %xmm1, %xmm3
-; SSE42-NEXT:    pxor %xmm2, %xmm3
-; SSE42-NEXT:    pxor %xmm0, %xmm2
-; SSE42-NEXT:    pcmpgtq %xmm3, %xmm2
-; SSE42-NEXT:    movdqa %xmm0, %xmm3
-; SSE42-NEXT:    psubq %xmm1, %xmm3
-; SSE42-NEXT:    psubq %xmm0, %xmm1
-; SSE42-NEXT:    movdqa %xmm2, %xmm0
-; SSE42-NEXT:    blendvpd %xmm0, %xmm3, %xmm1
-; SSE42-NEXT:    paddq %xmm1, %xmm2
-; SSE42-NEXT:    movdqa %xmm2, %xmm0
+; SSE42-NEXT:    psubq %xmm0, %xmm3
+; SSE42-NEXT:    movdqa {{.*#+}} xmm4 = [9223372036854775808,9223372036854775808]
+; SSE42-NEXT:    pxor %xmm4, %xmm1
+; SSE42-NEXT:    pxor %xmm4, %xmm0
+; SSE42-NEXT:    pcmpgtq %xmm1, %xmm0
+; SSE42-NEXT:    blendvpd %xmm0, %xmm2, %xmm3
+; SSE42-NEXT:    paddq %xmm3, %xmm0
 ; SSE42-NEXT:    retq
 ;
 ; AVX1-LABEL: abd_cmp_v2i64_multiuse_cmp:
 ; AVX1:       # %bb.0:
-; AVX1-NEXT:    vmovddup {{.*#+}} xmm2 = [9223372036854775808,9223372036854775808]
-; AVX1-NEXT:    # xmm2 = mem[0,0]
-; AVX1-NEXT:    vpxor %xmm2, %xmm1, %xmm3
-; AVX1-NEXT:    vpxor %xmm2, %xmm0, %xmm2
-; AVX1-NEXT:    vpcmpgtq %xmm3, %xmm2, %xmm2
-; AVX1-NEXT:    vpsubq %xmm1, %xmm0, %xmm3
-; AVX1-NEXT:    vpsubq %xmm0, %xmm1, %xmm0
-; AVX1-NEXT:    vblendvpd %xmm2, %xmm3, %xmm0, %xmm0
-; AVX1-NEXT:    vpaddq %xmm0, %xmm2, %xmm0
+; AVX1-NEXT:    vpsubq %xmm1, %xmm0, %xmm2
+; AVX1-NEXT:    vpsubq %xmm0, %xmm1, %xmm3
+; AVX1-NEXT:    vmovddup {{.*#+}} xmm4 = [9223372036854775808,9223372036854775808]
+; AVX1-NEXT:    # xmm4 = mem[0,0]
+; AVX1-NEXT:    vpxor %xmm4, %xmm1, %xmm1
+; AVX1-NEXT:    vpxor %xmm4, %xmm0, %xmm0
+; AVX1-NEXT:    vpcmpgtq %xmm1, %xmm0, %xmm0
+; AVX1-NEXT:    vblendvpd %xmm0, %xmm2, %xmm3, %xmm1
+; AVX1-NEXT:    vpaddq %xmm1, %xmm0, %xmm0
 ; AVX1-NEXT:    retq
 ;
 ; AVX2-LABEL: abd_cmp_v2i64_multiuse_cmp:
 ; AVX2:       # %bb.0:
-; AVX2-NEXT:    vpbroadcastq {{.*#+}} xmm2 = [9223372036854775808,9223372036854775808]
-; AVX2-NEXT:    vpxor %xmm2, %xmm1, %xmm3
-; AVX2-NEXT:    vpxor %xmm2, %xmm0, %xmm2
-; AVX2-NEXT:    vpcmpgtq %xmm3, %xmm2, %xmm2
-; AVX2-NEXT:    vpsubq %xmm1, %xmm0, %xmm3
-; AVX2-NEXT:    vpsubq %xmm0, %xmm1, %xmm0
-; AVX2-NEXT:    vblendvpd %xmm2, %xmm3, %xmm0, %xmm0
-; AVX2-NEXT:    vpaddq %xmm0, %xmm2, %xmm0
+; AVX2-NEXT:    vpsubq %xmm1, %xmm0, %xmm2
+; AVX2-NEXT:    vpsubq %xmm0, %xmm1, %xmm3
+; AVX2-NEXT:    vpbroadcastq {{.*#+}} xmm4 = [9223372036854775808,9223372036854775808]
+; AVX2-NEXT:    vpxor %xmm4, %xmm1, %xmm1
+; AVX2-NEXT:    vpxor %xmm4, %xmm0, %xmm0
+; AVX2-NEXT:    vpcmpgtq %xmm1, %xmm0, %xmm0
+; AVX2-NEXT:    vblendvpd %xmm0, %xmm2, %xmm3, %xmm1
+; AVX2-NEXT:    vpaddq %xmm1, %xmm0, %xmm0
 ; AVX2-NEXT:    retq
 ;
 ; AVX512-LABEL: abd_cmp_v2i64_multiuse_cmp:
diff --git a/llvm/test/CodeGen/X86/apx/kmov-postrapseudos.ll b/llvm/test/CodeGen/X86/apx/kmov-postrapseudos.ll
index 017024c173c3..b2cb2c3e04b3 100644
--- a/llvm/test/CodeGen/X86/apx/kmov-postrapseudos.ll
+++ b/llvm/test/CodeGen/X86/apx/kmov-postrapseudos.ll
@@ -52,10 +52,7 @@ alloca_21:
 define i32 @kmovrk_1(<4 x ptr> %arg) {
 ; AVX512-LABEL: kmovrk_1:
 ; AVX512:       # %bb.0: # %bb
-; AVX512-NEXT:    # kill: def $ymm0 killed $ymm0 def $zmm0
-; AVX512-NEXT:    vptestmq %zmm0, %zmm0, %k0 # encoding: [0x62,0xf2,0xfd,0x48,0x27,0xc0]
-; AVX512-NEXT:    kmovw %k0, %eax # EVEX TO VEX Compression encoding: [0xc5,0xf8,0x93,0xc0]
-; AVX512-NEXT:    testb $15, %al # encoding: [0xa8,0x0f]
+; AVX512-NEXT:    vptest %ymm0, %ymm0 # encoding: [0xc4,0xe2,0x7d,0x17,0xc0]
 ; AVX512-NEXT:    jne .LBB2_1 # encoding: [0x75,A]
 ; AVX512-NEXT:    # fixup A - offset: 1, value: .LBB2_1-1, kind: FK_PCRel_1
 ; AVX512-NEXT:  # %bb.2: # %bb3
@@ -66,10 +63,7 @@ define i32 @kmovrk_1(<4 x ptr> %arg) {
 ;
 ; AVX512BW-LABEL: kmovrk_1:
 ; AVX512BW:       # %bb.0: # %bb
-; AVX512BW-NEXT:    # kill: def $ymm0 killed $ymm0 def $zmm0
-; AVX512BW-NEXT:    vptestmq %zmm0, %zmm0, %k0 # encoding: [0x62,0xf2,0xfd,0x48,0x27,0xc0]
-; AVX512BW-NEXT:    kmovd %k0, %eax # EVEX TO VEX Compression encoding: [0xc5,0xfb,0x93,0xc0]
-; AVX512BW-NEXT:    testb $15, %al # encoding: [0xa8,0x0f]
+; AVX512BW-NEXT:    vptest %ymm0, %ymm0 # encoding: [0xc4,0xe2,0x7d,0x17,0xc0]
 ; AVX512BW-NEXT:    jne .LBB2_1 # encoding: [0x75,A]
 ; AVX512BW-NEXT:    # fixup A - offset: 1, value: .LBB2_1-1, kind: FK_PCRel_1
 ; AVX512BW-NEXT:  # %bb.2: # %bb3
diff --git a/llvm/test/CodeGen/X86/avgceils.ll b/llvm/test/CodeGen/X86/avgceils.ll
index 4529ea275df9..f44f98c2a41a 100644
--- a/llvm/test/CodeGen/X86/avgceils.ll
+++ b/llvm/test/CodeGen/X86/avgceils.ll
@@ -9,7 +9,7 @@
 ; 128-bit vectors
 ;
 
-define <16 x i8> @test_fixed_v16i8(<16 x i8> %a0, <16 x i8> %a1) {
+define <16 x i8> @test_fixed_v16i8(<16 x i8> %a0, <16 x i8> %a1) nounwind {
 ; SSE-LABEL: test_fixed_v16i8:
 ; SSE:       # %bb.0:
 ; SSE-NEXT:    movdqa %xmm0, %xmm2
@@ -65,7 +65,7 @@ define <16 x i8> @test_fixed_v16i8(<16 x i8> %a0, <16 x i8> %a1) {
   ret <16 x i8> %res
 }
 
-define <16 x i8> @test_ext_v16i8(<16 x i8> %a0, <16 x i8> %a1) {
+define <16 x i8> @test_ext_v16i8(<16 x i8> %a0, <16 x i8> %a1) nounwind {
 ; SSE2-LABEL: test_ext_v16i8:
 ; SSE2:       # %bb.0:
 ; SSE2-NEXT:    punpckhbw {{.*#+}} xmm2 = xmm2[8],xmm0[8],xmm2[9],xmm0[9],xmm2[10],xmm0[10],xmm2[11],xmm0[11],xmm2[12],xmm0[12],xmm2[13],xmm0[13],xmm2[14],xmm0[14],xmm2[15],xmm0[15]
@@ -165,7 +165,7 @@ define <16 x i8> @test_ext_v16i8(<16 x i8> %a0, <16 x i8> %a1) {
   ret <16 x i8> %res
 }
 
-define <8 x i16> @test_fixed_v8i16(<8 x i16> %a0, <8 x i16> %a1) {
+define <8 x i16> @test_fixed_v8i16(<8 x i16> %a0, <8 x i16> %a1) nounwind {
 ; SSE-LABEL: test_fixed_v8i16:
 ; SSE:       # %bb.0:
 ; SSE-NEXT:    movdqa %xmm0, %xmm2
@@ -190,7 +190,7 @@ define <8 x i16> @test_fixed_v8i16(<8 x i16> %a0, <8 x i16> %a1) {
   ret <8 x i16> %res
 }
 
-define <8 x i16> @test_ext_v8i16(<8 x i16> %a0, <8 x i16> %a1) {
+define <8 x i16> @test_ext_v8i16(<8 x i16> %a0, <8 x i16> %a1) nounwind {
 ; SSE2-LABEL: test_ext_v8i16:
 ; SSE2:       # %bb.0:
 ; SSE2-NEXT:    punpcklwd {{.*#+}} xmm2 = xmm2[0],xmm0[0],xmm2[1],xmm0[1],xmm2[2],xmm0[2],xmm2[3],xmm0[3]
@@ -289,7 +289,7 @@ define <8 x i16> @test_ext_v8i16(<8 x i16> %a0, <8 x i16> %a1) {
   ret <8 x i16> %res
 }
 
-define <4 x i32> @test_fixed_v4i32(<4 x i32> %a0, <4 x i32> %a1) {
+define <4 x i32> @test_fixed_v4i32(<4 x i32> %a0, <4 x i32> %a1) nounwind {
 ; SSE-LABEL: test_fixed_v4i32:
 ; SSE:       # %bb.0:
 ; SSE-NEXT:    movdqa %xmm0, %xmm2
@@ -314,7 +314,7 @@ define <4 x i32> @test_fixed_v4i32(<4 x i32> %a0, <4 x i32> %a1) {
   ret <4 x i32> %res
 }
 
-define <4 x i32> @test_ext_v4i32(<4 x i32> %a0, <4 x i32> %a1) {
+define <4 x i32> @test_ext_v4i32(<4 x i32> %a0, <4 x i32> %a1) nounwind {
 ; SSE2-LABEL: test_ext_v4i32:
 ; SSE2:       # %bb.0:
 ; SSE2-NEXT:    pshufd {{.*#+}} xmm2 = xmm0[2,3,2,3]
@@ -410,7 +410,7 @@ define <4 x i32> @test_ext_v4i32(<4 x i32> %a0, <4 x i32> %a1) {
   ret <4 x i32> %res
 }
 
-define <2 x i64> @test_fixed_v2i64(<2 x i64> %a0, <2 x i64> %a1) {
+define <2 x i64> @test_fixed_v2i64(<2 x i64> %a0, <2 x i64> %a1) nounwind {
 ; SSE2-LABEL: test_fixed_v2i64:
 ; SSE2:       # %bb.0:
 ; SSE2-NEXT:    movdqa %xmm0, %xmm2
@@ -472,7 +472,7 @@ define <2 x i64> @test_fixed_v2i64(<2 x i64> %a0, <2 x i64> %a1) {
   ret <2 x i64> %res
 }
 
-define <2 x i64> @test_ext_v2i64(<2 x i64> %a0, <2 x i64> %a1) {
+define <2 x i64> @test_ext_v2i64(<2 x i64> %a0, <2 x i64> %a1) nounwind {
 ; SSE2-LABEL: test_ext_v2i64:
 ; SSE2:       # %bb.0:
 ; SSE2-NEXT:    movq %xmm0, %rax
@@ -574,7 +574,7 @@ define <2 x i64> @test_ext_v2i64(<2 x i64> %a0, <2 x i64> %a1) {
 ; 256-bit vectors
 ;
 
-define <32 x i8> @test_fixed_v32i8(<32 x i8> %a0, <32 x i8> %a1) {
+define <32 x i8> @test_fixed_v32i8(<32 x i8> %a0, <32 x i8> %a1) nounwind {
 ; SSE-LABEL: test_fixed_v32i8:
 ; SSE:       # %bb.0:
 ; SSE-NEXT:    movdqa %xmm1, %xmm4
@@ -649,7 +649,7 @@ define <32 x i8> @test_fixed_v32i8(<32 x i8> %a0, <32 x i8> %a1) {
   ret <32 x i8> %res
 }
 
-define <32 x i8> @test_ext_v32i8(<32 x i8> %a0, <32 x i8> %a1) {
+define <32 x i8> @test_ext_v32i8(<32 x i8> %a0, <32 x i8> %a1) nounwind {
 ; SSE2-LABEL: test_ext_v32i8:
 ; SSE2:       # %bb.0:
 ; SSE2-NEXT:    punpckhbw {{.*#+}} xmm5 = xmm5[8],xmm0[8],xmm5[9],xmm0[9],xmm5[10],xmm0[10],xmm5[11],xmm0[11],xmm5[12],xmm0[12],xmm5[13],xmm0[13],xmm5[14],xmm0[14],xmm5[15],xmm0[15]
@@ -806,7 +806,7 @@ define <32 x i8> @test_ext_v32i8(<32 x i8> %a0, <32 x i8> %a1) {
   ret <32 x i8> %res
 }
 
-define <16 x i16> @test_fixed_v16i16(<16 x i16> %a0, <16 x i16> %a1) {
+define <16 x i16> @test_fixed_v16i16(<16 x i16> %a0, <16 x i16> %a1) nounwind {
 ; SSE-LABEL: test_fixed_v16i16:
 ; SSE:       # %bb.0:
 ; SSE-NEXT:    movdqa %xmm1, %xmm4
@@ -858,7 +858,7 @@ define <16 x i16> @test_fixed_v16i16(<16 x i16> %a0, <16 x i16> %a1) {
   ret <16 x i16> %res
 }
 
-define <16 x i16> @test_ext_v16i16(<16 x i16> %a0, <16 x i16> %a1) {
+define <16 x i16> @test_ext_v16i16(<16 x i16> %a0, <16 x i16> %a1) nounwind {
 ; SSE2-LABEL: test_ext_v16i16:
 ; SSE2:       # %bb.0:
 ; SSE2-NEXT:    punpcklwd {{.*#+}} xmm4 = xmm4[0],xmm1[0],xmm4[1],xmm1[1],xmm4[2],xmm1[2],xmm4[3],xmm1[3]
@@ -1014,7 +1014,7 @@ define <16 x i16> @test_ext_v16i16(<16 x i16> %a0, <16 x i16> %a1) {
   ret <16 x i16> %res
 }
 
-define <8 x i32> @test_fixed_v8i32(<8 x i32> %a0, <8 x i32> %a1) {
+define <8 x i32> @test_fixed_v8i32(<8 x i32> %a0, <8 x i32> %a1) nounwind {
 ; SSE-LABEL: test_fixed_v8i32:
 ; SSE:       # %bb.0:
 ; SSE-NEXT:    movdqa %xmm1, %xmm4
@@ -1066,7 +1066,7 @@ define <8 x i32> @test_fixed_v8i32(<8 x i32> %a0, <8 x i32> %a1) {
   ret <8 x i32> %res
 }
 
-define <8 x i32> @test_ext_v8i32(<8 x i32> %a0, <8 x i32> %a1) {
+define <8 x i32> @test_ext_v8i32(<8 x i32> %a0, <8 x i32> %a1) nounwind {
 ; SSE2-LABEL: test_ext_v8i32:
 ; SSE2:       # %bb.0:
 ; SSE2-NEXT:    pshufd {{.*#+}} xmm6 = xmm0[2,3,2,3]
@@ -1218,7 +1218,7 @@ define <8 x i32> @test_ext_v8i32(<8 x i32> %a0, <8 x i32> %a1) {
   ret <8 x i32> %res
 }
 
-define <4 x i64> @test_fixed_v4i64(<4 x i64> %a0, <4 x i64> %a1) {
+define <4 x i64> @test_fixed_v4i64(<4 x i64> %a0, <4 x i64> %a1) nounwind {
 ; SSE2-LABEL: test_fixed_v4i64:
 ; SSE2:       # %bb.0:
 ; SSE2-NEXT:    movdqa %xmm1, %xmm4
@@ -1306,27 +1306,15 @@ define <4 x i64> @test_fixed_v4i64(<4 x i64> %a0, <4 x i64> %a1) {
   ret <4 x i64> %res
 }
 
-define <4 x i64> @test_ext_v4i64(<4 x i64> %a0, <4 x i64> %a1) {
+define <4 x i64> @test_ext_v4i64(<4 x i64> %a0, <4 x i64> %a1) nounwind {
 ; SSE2-LABEL: test_ext_v4i64:
 ; SSE2:       # %bb.0:
 ; SSE2-NEXT:    pushq %rbp
-; SSE2-NEXT:    .cfi_def_cfa_offset 16
 ; SSE2-NEXT:    pushq %r15
-; SSE2-NEXT:    .cfi_def_cfa_offset 24
 ; SSE2-NEXT:    pushq %r14
-; SSE2-NEXT:    .cfi_def_cfa_offset 32
 ; SSE2-NEXT:    pushq %r13
-; SSE2-NEXT:    .cfi_def_cfa_offset 40
 ; SSE2-NEXT:    pushq %r12
-; SSE2-NEXT:    .cfi_def_cfa_offset 48
 ; SSE2-NEXT:    pushq %rbx
-; SSE2-NEXT:    .cfi_def_cfa_offset 56
-; SSE2-NEXT:    .cfi_offset %rbx, -56
-; SSE2-NEXT:    .cfi_offset %r12, -48
-; SSE2-NEXT:    .cfi_offset %r13, -40
-; SSE2-NEXT:    .cfi_offset %r14, -32
-; SSE2-NEXT:    .cfi_offset %r15, -24
-; SSE2-NEXT:    .cfi_offset %rbp, -16
 ; SSE2-NEXT:    movq %xmm0, %r11
 ; SSE2-NEXT:    movq %r11, %r12
 ; SSE2-NEXT:    sarq $63, %r12
@@ -1382,39 +1370,21 @@ define <4 x i64> @test_ext_v4i64(<4 x i64> %a0, <4 x i64> %a1) {
 ; SSE2-NEXT:    punpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm2[0]
 ; SSE2-NEXT:    punpcklqdq {{.*#+}} xmm1 = xmm1[0],xmm3[0]
 ; SSE2-NEXT:    popq %rbx
-; SSE2-NEXT:    .cfi_def_cfa_offset 48
 ; SSE2-NEXT:    popq %r12
-; SSE2-NEXT:    .cfi_def_cfa_offset 40
 ; SSE2-NEXT:    popq %r13
-; SSE2-NEXT:    .cfi_def_cfa_offset 32
 ; SSE2-NEXT:    popq %r14
-; SSE2-NEXT:    .cfi_def_cfa_offset 24
 ; SSE2-NEXT:    popq %r15
-; SSE2-NEXT:    .cfi_def_cfa_offset 16
 ; SSE2-NEXT:    popq %rbp
-; SSE2-NEXT:    .cfi_def_cfa_offset 8
 ; SSE2-NEXT:    retq
 ;
 ; SSE4-LABEL: test_ext_v4i64:
 ; SSE4:       # %bb.0:
 ; SSE4-NEXT:    pushq %rbp
-; SSE4-NEXT:    .cfi_def_cfa_offset 16
 ; SSE4-NEXT:    pushq %r15
-; SSE4-NEXT:    .cfi_def_cfa_offset 24
 ; SSE4-NEXT:    pushq %r14
-; SSE4-NEXT:    .cfi_def_cfa_offset 32
 ; SSE4-NEXT:    pushq %r13
-; SSE4-NEXT:    .cfi_def_cfa_offset 40
 ; SSE4-NEXT:    pushq %r12
-; SSE4-NEXT:    .cfi_def_cfa_offset 48
 ; SSE4-NEXT:    pushq %rbx
-; SSE4-NEXT:    .cfi_def_cfa_offset 56
-; SSE4-NEXT:    .cfi_offset %rbx, -56
-; SSE4-NEXT:    .cfi_offset %r12, -48
-; SSE4-NEXT:    .cfi_offset %r13, -40
-; SSE4-NEXT:    .cfi_offset %r14, -32
-; SSE4-NEXT:    .cfi_offset %r15, -24
-; SSE4-NEXT:    .cfi_offset %rbp, -16
 ; SSE4-NEXT:    pextrq $1, %xmm0, %r11
 ; SSE4-NEXT:    movq %r11, %r12
 ; SSE4-NEXT:    sarq $63, %r12
@@ -1466,39 +1436,21 @@ define <4 x i64> @test_ext_v4i64(<4 x i64> %a0, <4 x i64> %a1) {
 ; SSE4-NEXT:    punpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm2[0]
 ; SSE4-NEXT:    punpcklqdq {{.*#+}} xmm1 = xmm1[0],xmm3[0]
 ; SSE4-NEXT:    popq %rbx
-; SSE4-NEXT:    .cfi_def_cfa_offset 48
 ; SSE4-NEXT:    popq %r12
-; SSE4-NEXT:    .cfi_def_cfa_offset 40
 ; SSE4-NEXT:    popq %r13
-; SSE4-NEXT:    .cfi_def_cfa_offset 32
 ; SSE4-NEXT:    popq %r14
-; SSE4-NEXT:    .cfi_def_cfa_offset 24
 ; SSE4-NEXT:    popq %r15
-; SSE4-NEXT:    .cfi_def_cfa_offset 16
 ; SSE4-NEXT:    popq %rbp
-; SSE4-NEXT:    .cfi_def_cfa_offset 8
 ; SSE4-NEXT:    retq
 ;
 ; AVX1-LABEL: test_ext_v4i64:
 ; AVX1:       # %bb.0:
 ; AVX1-NEXT:    pushq %rbp
-; AVX1-NEXT:    .cfi_def_cfa_offset 16
 ; AVX1-NEXT:    pushq %r15
-; AVX1-NEXT:    .cfi_def_cfa_offset 24
 ; AVX1-NEXT:    pushq %r14
-; AVX1-NEXT:    .cfi_def_cfa_offset 32
 ; AVX1-NEXT:    pushq %r13
-; AVX1-NEXT:    .cfi_def_cfa_offset 40
 ; AVX1-NEXT:    pushq %r12
-; AVX1-NEXT:    .cfi_def_cfa_offset 48
 ; AVX1-NEXT:    pushq %rbx
-; AVX1-NEXT:    .cfi_def_cfa_offset 56
-; AVX1-NEXT:    .cfi_offset %rbx, -56
-; AVX1-NEXT:    .cfi_offset %r12, -48
-; AVX1-NEXT:    .cfi_offset %r13, -40
-; AVX1-NEXT:    .cfi_offset %r14, -32
-; AVX1-NEXT:    .cfi_offset %r15, -24
-; AVX1-NEXT:    .cfi_offset %rbp, -16
 ; AVX1-NEXT:    vextractf128 $1, %ymm0, %xmm2
 ; AVX1-NEXT:    vpextrq $1, %xmm2, %r11
 ; AVX1-NEXT:    movq %r11, %r12
@@ -1553,39 +1505,21 @@ define <4 x i64> @test_ext_v4i64(<4 x i64> %a0, <4 x i64> %a1) {
 ; AVX1-NEXT:    vpunpcklqdq {{.*#+}} xmm1 = xmm3[0],xmm2[0]
 ; AVX1-NEXT:    vinsertf128 $1, %xmm0, %ymm1, %ymm0
 ; AVX1-NEXT:    popq %rbx
-; AVX1-NEXT:    .cfi_def_cfa_offset 48
 ; AVX1-NEXT:    popq %r12
-; AVX1-NEXT:    .cfi_def_cfa_offset 40
 ; AVX1-NEXT:    popq %r13
-; AVX1-NEXT:    .cfi_def_cfa_offset 32
 ; AVX1-NEXT:    popq %r14
-; AVX1-NEXT:    .cfi_def_cfa_offset 24
 ; AVX1-NEXT:    popq %r15
-; AVX1-NEXT:    .cfi_def_cfa_offset 16
 ; AVX1-NEXT:    popq %rbp
-; AVX1-NEXT:    .cfi_def_cfa_offset 8
 ; AVX1-NEXT:    retq
 ;
 ; AVX2-LABEL: test_ext_v4i64:
 ; AVX2:       # %bb.0:
 ; AVX2-NEXT:    pushq %rbp
-; AVX2-NEXT:    .cfi_def_cfa_offset 16
 ; AVX2-NEXT:    pushq %r15
-; AVX2-NEXT:    .cfi_def_cfa_offset 24
 ; AVX2-NEXT:    pushq %r14
-; AVX2-NEXT:    .cfi_def_cfa_offset 32
 ; AVX2-NEXT:    pushq %r13
-; AVX2-NEXT:    .cfi_def_cfa_offset 40
 ; AVX2-NEXT:    pushq %r12
-; AVX2-NEXT:    .cfi_def_cfa_offset 48
 ; AVX2-NEXT:    pushq %rbx
-; AVX2-NEXT:    .cfi_def_cfa_offset 56
-; AVX2-NEXT:    .cfi_offset %rbx, -56
-; AVX2-NEXT:    .cfi_offset %r12, -48
-; AVX2-NEXT:    .cfi_offset %r13, -40
-; AVX2-NEXT:    .cfi_offset %r14, -32
-; AVX2-NEXT:    .cfi_offset %r15, -24
-; AVX2-NEXT:    .cfi_offset %rbp, -16
 ; AVX2-NEXT:    vextracti128 $1, %ymm0, %xmm2
 ; AVX2-NEXT:    vpextrq $1, %xmm2, %r11
 ; AVX2-NEXT:    movq %r11, %r12
@@ -1640,39 +1574,21 @@ define <4 x i64> @test_ext_v4i64(<4 x i64> %a0, <4 x i64> %a1) {
 ; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm1 = xmm3[0],xmm2[0]
 ; AVX2-NEXT:    vinserti128 $1, %xmm0, %ymm1, %ymm0
 ; AVX2-NEXT:    popq %rbx
-; AVX2-NEXT:    .cfi_def_cfa_offset 48
 ; AVX2-NEXT:    popq %r12
-; AVX2-NEXT:    .cfi_def_cfa_offset 40
 ; AVX2-NEXT:    popq %r13
-; AVX2-NEXT:    .cfi_def_cfa_offset 32
 ; AVX2-NEXT:    popq %r14
-; AVX2-NEXT:    .cfi_def_cfa_offset 24
 ; AVX2-NEXT:    popq %r15
-; AVX2-NEXT:    .cfi_def_cfa_offset 16
 ; AVX2-NEXT:    popq %rbp
-; AVX2-NEXT:    .cfi_def_cfa_offset 8
 ; AVX2-NEXT:    retq
 ;
 ; AVX512-LABEL: test_ext_v4i64:
 ; AVX512:       # %bb.0:
 ; AVX512-NEXT:    pushq %rbp
-; AVX512-NEXT:    .cfi_def_cfa_offset 16
 ; AVX512-NEXT:    pushq %r15
-; AVX512-NEXT:    .cfi_def_cfa_offset 24
 ; AVX512-NEXT:    pushq %r14
-; AVX512-NEXT:    .cfi_def_cfa_offset 32
 ; AVX512-NEXT:    pushq %r13
-; AVX512-NEXT:    .cfi_def_cfa_offset 40
 ; AVX512-NEXT:    pushq %r12
-; AVX512-NEXT:    .cfi_def_cfa_offset 48
 ; AVX512-NEXT:    pushq %rbx
-; AVX512-NEXT:    .cfi_def_cfa_offset 56
-; AVX512-NEXT:    .cfi_offset %rbx, -56
-; AVX512-NEXT:    .cfi_offset %r12, -48
-; AVX512-NEXT:    .cfi_offset %r13, -40
-; AVX512-NEXT:    .cfi_offset %r14, -32
-; AVX512-NEXT:    .cfi_offset %r15, -24
-; AVX512-NEXT:    .cfi_offset %rbp, -16
 ; AVX512-NEXT:    vextracti128 $1, %ymm0, %xmm2
 ; AVX512-NEXT:    vpextrq $1, %xmm2, %r11
 ; AVX512-NEXT:    movq %r11, %r12
@@ -1727,17 +1643,11 @@ define <4 x i64> @test_ext_v4i64(<4 x i64> %a0, <4 x i64> %a1) {
 ; AVX512-NEXT:    vpunpcklqdq {{.*#+}} xmm1 = xmm3[0],xmm2[0]
 ; AVX512-NEXT:    vinserti128 $1, %xmm0, %ymm1, %ymm0
 ; AVX512-NEXT:    popq %rbx
-; AVX512-NEXT:    .cfi_def_cfa_offset 48
 ; AVX512-NEXT:    popq %r12
-; AVX512-NEXT:    .cfi_def_cfa_offset 40
 ; AVX512-NEXT:    popq %r13
-; AVX512-NEXT:    .cfi_def_cfa_offset 32
 ; AVX512-NEXT:    popq %r14
-; AVX512-NEXT:    .cfi_def_cfa_offset 24
 ; AVX512-NEXT:    popq %r15
-; AVX512-NEXT:    .cfi_def_cfa_offset 16
 ; AVX512-NEXT:    popq %rbp
-; AVX512-NEXT:    .cfi_def_cfa_offset 8
 ; AVX512-NEXT:    retq
   %x0 = sext <4 x i64> %a0 to <4 x i128>
   %x1 = sext <4 x i64> %a1 to <4 x i128>
@@ -1752,7 +1662,7 @@ define <4 x i64> @test_ext_v4i64(<4 x i64> %a0, <4 x i64> %a1) {
 ; 512-bit vectors
 ;
 
-define <64 x i8> @test_fixed_v64i8(<64 x i8> %a0, <64 x i8> %a1) {
+define <64 x i8> @test_fixed_v64i8(<64 x i8> %a0, <64 x i8> %a1) nounwind {
 ; SSE-LABEL: test_fixed_v64i8:
 ; SSE:       # %bb.0:
 ; SSE-NEXT:    movdqa %xmm3, %xmm11
@@ -1864,7 +1774,7 @@ define <64 x i8> @test_fixed_v64i8(<64 x i8> %a0, <64 x i8> %a1) {
   ret <64 x i8> %res
 }
 
-define <64 x i8> @test_ext_v64i8(<64 x i8> %a0, <64 x i8> %a1) {
+define <64 x i8> @test_ext_v64i8(<64 x i8> %a0, <64 x i8> %a1) nounwind {
 ; SSE2-LABEL: test_ext_v64i8:
 ; SSE2:       # %bb.0:
 ; SSE2-NEXT:    movdqa %xmm3, %xmm8
@@ -2144,7 +2054,7 @@ define <64 x i8> @test_ext_v64i8(<64 x i8> %a0, <64 x i8> %a1) {
   ret <64 x i8> %res
 }
 
-define <32 x i16> @test_fixed_v32i16(<32 x i16> %a0, <32 x i16> %a1) {
+define <32 x i16> @test_fixed_v32i16(<32 x i16> %a0, <32 x i16> %a1) nounwind {
 ; SSE-LABEL: test_fixed_v32i16:
 ; SSE:       # %bb.0:
 ; SSE-NEXT:    movdqa %xmm3, %xmm8
@@ -2220,7 +2130,7 @@ define <32 x i16> @test_fixed_v32i16(<32 x i16> %a0, <32 x i16> %a1) {
   ret <32 x i16> %res
 }
 
-define <32 x i16> @test_ext_v32i16(<32 x i16> %a0, <32 x i16> %a1) {
+define <32 x i16> @test_ext_v32i16(<32 x i16> %a0, <32 x i16> %a1) nounwind {
 ; SSE2-LABEL: test_ext_v32i16:
 ; SSE2:       # %bb.0:
 ; SSE2-NEXT:    punpcklwd {{.*#+}} xmm13 = xmm13[0],xmm3[0],xmm13[1],xmm3[1],xmm13[2],xmm3[2],xmm13[3],xmm3[3]
@@ -2498,7 +2408,7 @@ define <32 x i16> @test_ext_v32i16(<32 x i16> %a0, <32 x i16> %a1) {
   ret <32 x i16> %res
 }
 
-define <16 x i32> @test_fixed_v16i32(<16 x i32> %a0, <16 x i32> %a1) {
+define <16 x i32> @test_fixed_v16i32(<16 x i32> %a0, <16 x i32> %a1) nounwind {
 ; SSE-LABEL: test_fixed_v16i32:
 ; SSE:       # %bb.0:
 ; SSE-NEXT:    movdqa %xmm3, %xmm8
@@ -2574,7 +2484,7 @@ define <16 x i32> @test_fixed_v16i32(<16 x i32> %a0, <16 x i32> %a1) {
   ret <16 x i32> %res
 }
 
-define <16 x i32> @test_ext_v16i32(<16 x i32> %a0, <16 x i32> %a1) {
+define <16 x i32> @test_ext_v16i32(<16 x i32> %a0, <16 x i32> %a1) nounwind {
 ; SSE2-LABEL: test_ext_v16i32:
 ; SSE2:       # %bb.0:
 ; SSE2-NEXT:    pshufd {{.*#+}} xmm13 = xmm0[2,3,2,3]
@@ -2848,7 +2758,7 @@ define <16 x i32> @test_ext_v16i32(<16 x i32> %a0, <16 x i32> %a1) {
   ret <16 x i32> %res
 }
 
-define <8 x i64> @test_fixed_v8i64(<8 x i64> %a0, <8 x i64> %a1) {
+define <8 x i64> @test_fixed_v8i64(<8 x i64> %a0, <8 x i64> %a1) nounwind {
 ; SSE2-LABEL: test_fixed_v8i64:
 ; SSE2:       # %bb.0:
 ; SSE2-NEXT:    movdqa %xmm3, %xmm8
@@ -2985,29 +2895,16 @@ define <8 x i64> @test_fixed_v8i64(<8 x i64> %a0, <8 x i64> %a1) {
   ret <8 x i64> %res
 }
 
-define <8 x i64> @test_ext_v8i64(<8 x i64> %a0, <8 x i64> %a1) {
+define <8 x i64> @test_ext_v8i64(<8 x i64> %a0, <8 x i64> %a1) nounwind {
 ; SSE2-LABEL: test_ext_v8i64:
 ; SSE2:       # %bb.0:
 ; SSE2-NEXT:    pushq %rbp
-; SSE2-NEXT:    .cfi_def_cfa_offset 16
 ; SSE2-NEXT:    pushq %r15
-; SSE2-NEXT:    .cfi_def_cfa_offset 24
 ; SSE2-NEXT:    pushq %r14
-; SSE2-NEXT:    .cfi_def_cfa_offset 32
 ; SSE2-NEXT:    pushq %r13
-; SSE2-NEXT:    .cfi_def_cfa_offset 40
 ; SSE2-NEXT:    pushq %r12
-; SSE2-NEXT:    .cfi_def_cfa_offset 48
 ; SSE2-NEXT:    pushq %rbx
-; SSE2-NEXT:    .cfi_def_cfa_offset 56
 ; SSE2-NEXT:    pushq %rax
-; SSE2-NEXT:    .cfi_def_cfa_offset 64
-; SSE2-NEXT:    .cfi_offset %rbx, -56
-; SSE2-NEXT:    .cfi_offset %r12, -48
-; SSE2-NEXT:    .cfi_offset %r13, -40
-; SSE2-NEXT:    .cfi_offset %r14, -32
-; SSE2-NEXT:    .cfi_offset %r15, -24
-; SSE2-NEXT:    .cfi_offset %rbp, -16
 ; SSE2-NEXT:    movq %xmm0, %rax
 ; SSE2-NEXT:    movq %rax, {{[-0-9]+}}(%r{{[sb]}}p) # 8-byte Spill
 ; SSE2-NEXT:    sarq $63, %rax
@@ -3137,43 +3034,23 @@ define <8 x i64> @test_ext_v8i64(<8 x i64> %a0, <8 x i64> %a1) {
 ; SSE2-NEXT:    punpcklqdq {{.*#+}} xmm2 = xmm2[0],xmm6[0]
 ; SSE2-NEXT:    punpcklqdq {{.*#+}} xmm3 = xmm3[0],xmm7[0]
 ; SSE2-NEXT:    addq $8, %rsp
-; SSE2-NEXT:    .cfi_def_cfa_offset 56
 ; SSE2-NEXT:    popq %rbx
-; SSE2-NEXT:    .cfi_def_cfa_offset 48
 ; SSE2-NEXT:    popq %r12
-; SSE2-NEXT:    .cfi_def_cfa_offset 40
 ; SSE2-NEXT:    popq %r13
-; SSE2-NEXT:    .cfi_def_cfa_offset 32
 ; SSE2-NEXT:    popq %r14
-; SSE2-NEXT:    .cfi_def_cfa_offset 24
 ; SSE2-NEXT:    popq %r15
-; SSE2-NEXT:    .cfi_def_cfa_offset 16
 ; SSE2-NEXT:    popq %rbp
-; SSE2-NEXT:    .cfi_def_cfa_offset 8
 ; SSE2-NEXT:    retq
 ;
 ; SSE4-LABEL: test_ext_v8i64:
 ; SSE4:       # %bb.0:
 ; SSE4-NEXT:    pushq %rbp
-; SSE4-NEXT:    .cfi_def_cfa_offset 16
 ; SSE4-NEXT:    pushq %r15
-; SSE4-NEXT:    .cfi_def_cfa_offset 24
 ; SSE4-NEXT:    pushq %r14
-; SSE4-NEXT:    .cfi_def_cfa_offset 32
 ; SSE4-NEXT:    pushq %r13
-; SSE4-NEXT:    .cfi_def_cfa_offset 40
 ; SSE4-NEXT:    pushq %r12
-; SSE4-NEXT:    .cfi_def_cfa_offset 48
 ; SSE4-NEXT:    pushq %rbx
-; SSE4-NEXT:    .cfi_def_cfa_offset 56
 ; SSE4-NEXT:    subq $16, %rsp
-; SSE4-NEXT:    .cfi_def_cfa_offset 72
-; SSE4-NEXT:    .cfi_offset %rbx, -56
-; SSE4-NEXT:    .cfi_offset %r12, -48
-; SSE4-NEXT:    .cfi_offset %r13, -40
-; SSE4-NEXT:    .cfi_offset %r14, -32
-; SSE4-NEXT:    .cfi_offset %r15, -24
-; SSE4-NEXT:    .cfi_offset %rbp, -16
 ; SSE4-NEXT:    pextrq $1, %xmm0, %rax
 ; SSE4-NEXT:    movq %rax, {{[-0-9]+}}(%r{{[sb]}}p) # 8-byte Spill
 ; SSE4-NEXT:    sarq $63, %rax
@@ -3301,43 +3178,23 @@ define <8 x i64> @test_ext_v8i64(<8 x i64> %a0, <8 x i64> %a1) {
 ; SSE4-NEXT:    punpcklqdq {{.*#+}} xmm2 = xmm2[0],xmm6[0]
 ; SSE4-NEXT:    punpcklqdq {{.*#+}} xmm3 = xmm3[0],xmm7[0]
 ; SSE4-NEXT:    addq $16, %rsp
-; SSE4-NEXT:    .cfi_def_cfa_offset 56
 ; SSE4-NEXT:    popq %rbx
-; SSE4-NEXT:    .cfi_def_cfa_offset 48
 ; SSE4-NEXT:    popq %r12
-; SSE4-NEXT:    .cfi_def_cfa_offset 40
 ; SSE4-NEXT:    popq %r13
-; SSE4-NEXT:    .cfi_def_cfa_offset 32
 ; SSE4-NEXT:    popq %r14
-; SSE4-NEXT:    .cfi_def_cfa_offset 24
 ; SSE4-NEXT:    popq %r15
-; SSE4-NEXT:    .cfi_def_cfa_offset 16
 ; SSE4-NEXT:    popq %rbp
-; SSE4-NEXT:    .cfi_def_cfa_offset 8
 ; SSE4-NEXT:    retq
 ;
 ; AVX1-LABEL: test_ext_v8i64:
 ; AVX1:       # %bb.0:
 ; AVX1-NEXT:    pushq %rbp
-; AVX1-NEXT:    .cfi_def_cfa_offset 16
 ; AVX1-NEXT:    pushq %r15
-; AVX1-NEXT:    .cfi_def_cfa_offset 24
 ; AVX1-NEXT:    pushq %r14
-; AVX1-NEXT:    .cfi_def_cfa_offset 32
 ; AVX1-NEXT:    pushq %r13
-; AVX1-NEXT:    .cfi_def_cfa_offset 40
 ; AVX1-NEXT:    pushq %r12
-; AVX1-NEXT:    .cfi_def_cfa_offset 48
 ; AVX1-NEXT:    pushq %rbx
-; AVX1-NEXT:    .cfi_def_cfa_offset 56
 ; AVX1-NEXT:    pushq %rax
-; AVX1-NEXT:    .cfi_def_cfa_offset 64
-; AVX1-NEXT:    .cfi_offset %rbx, -56
-; AVX1-NEXT:    .cfi_offset %r12, -48
-; AVX1-NEXT:    .cfi_offset %r13, -40
-; AVX1-NEXT:    .cfi_offset %r14, -32
-; AVX1-NEXT:    .cfi_offset %r15, -24
-; AVX1-NEXT:    .cfi_offset %rbp, -16
 ; AVX1-NEXT:    vextractf128 $1, %ymm0, %xmm4
 ; AVX1-NEXT:    vpextrq $1, %xmm4, %rax
 ; AVX1-NEXT:    movq %rax, {{[-0-9]+}}(%r{{[sb]}}p) # 8-byte Spill
@@ -3465,43 +3322,23 @@ define <8 x i64> @test_ext_v8i64(<8 x i64> %a0, <8 x i64> %a1) {
 ; AVX1-NEXT:    vpunpcklqdq {{.*#+}} xmm2 = xmm7[0],xmm6[0]
 ; AVX1-NEXT:    vinsertf128 $1, %xmm1, %ymm2, %ymm1
 ; AVX1-NEXT:    addq $8, %rsp
-; AVX1-NEXT:    .cfi_def_cfa_offset 56
 ; AVX1-NEXT:    popq %rbx
-; AVX1-NEXT:    .cfi_def_cfa_offset 48
 ; AVX1-NEXT:    popq %r12
-; AVX1-NEXT:    .cfi_def_cfa_offset 40
 ; AVX1-NEXT:    popq %r13
-; AVX1-NEXT:    .cfi_def_cfa_offset 32
 ; AVX1-NEXT:    popq %r14
-; AVX1-NEXT:    .cfi_def_cfa_offset 24
 ; AVX1-NEXT:    popq %r15
-; AVX1-NEXT:    .cfi_def_cfa_offset 16
 ; AVX1-NEXT:    popq %rbp
-; AVX1-NEXT:    .cfi_def_cfa_offset 8
 ; AVX1-NEXT:    retq
 ;
 ; AVX2-LABEL: test_ext_v8i64:
 ; AVX2:       # %bb.0:
 ; AVX2-NEXT:    pushq %rbp
-; AVX2-NEXT:    .cfi_def_cfa_offset 16
 ; AVX2-NEXT:    pushq %r15
-; AVX2-NEXT:    .cfi_def_cfa_offset 24
 ; AVX2-NEXT:    pushq %r14
-; AVX2-NEXT:    .cfi_def_cfa_offset 32
 ; AVX2-NEXT:    pushq %r13
-; AVX2-NEXT:    .cfi_def_cfa_offset 40
 ; AVX2-NEXT:    pushq %r12
-; AVX2-NEXT:    .cfi_def_cfa_offset 48
 ; AVX2-NEXT:    pushq %rbx
-; AVX2-NEXT:    .cfi_def_cfa_offset 56
 ; AVX2-NEXT:    pushq %rax
-; AVX2-NEXT:    .cfi_def_cfa_offset 64
-; AVX2-NEXT:    .cfi_offset %rbx, -56
-; AVX2-NEXT:    .cfi_offset %r12, -48
-; AVX2-NEXT:    .cfi_offset %r13, -40
-; AVX2-NEXT:    .cfi_offset %r14, -32
-; AVX2-NEXT:    .cfi_offset %r15, -24
-; AVX2-NEXT:    .cfi_offset %rbp, -16
 ; AVX2-NEXT:    vextracti128 $1, %ymm0, %xmm4
 ; AVX2-NEXT:    vpextrq $1, %xmm4, %rax
 ; AVX2-NEXT:    movq %rax, {{[-0-9]+}}(%r{{[sb]}}p) # 8-byte Spill
@@ -3629,43 +3466,23 @@ define <8 x i64> @test_ext_v8i64(<8 x i64> %a0, <8 x i64> %a1) {
 ; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm2 = xmm7[0],xmm6[0]
 ; AVX2-NEXT:    vinserti128 $1, %xmm1, %ymm2, %ymm1
 ; AVX2-NEXT:    addq $8, %rsp
-; AVX2-NEXT:    .cfi_def_cfa_offset 56
 ; AVX2-NEXT:    popq %rbx
-; AVX2-NEXT:    .cfi_def_cfa_offset 48
 ; AVX2-NEXT:    popq %r12
-; AVX2-NEXT:    .cfi_def_cfa_offset 40
 ; AVX2-NEXT:    popq %r13
-; AVX2-NEXT:    .cfi_def_cfa_offset 32
 ; AVX2-NEXT:    popq %r14
-; AVX2-NEXT:    .cfi_def_cfa_offset 24
 ; AVX2-NEXT:    popq %r15
-; AVX2-NEXT:    .cfi_def_cfa_offset 16
 ; AVX2-NEXT:    popq %rbp
-; AVX2-NEXT:    .cfi_def_cfa_offset 8
 ; AVX2-NEXT:    retq
 ;
 ; AVX512-LABEL: test_ext_v8i64:
 ; AVX512:       # %bb.0:
 ; AVX512-NEXT:    pushq %rbp
-; AVX512-NEXT:    .cfi_def_cfa_offset 16
 ; AVX512-NEXT:    pushq %r15
-; AVX512-NEXT:    .cfi_def_cfa_offset 24
 ; AVX512-NEXT:    pushq %r14
-; AVX512-NEXT:    .cfi_def_cfa_offset 32
 ; AVX512-NEXT:    pushq %r13
-; AVX512-NEXT:    .cfi_def_cfa_offset 40
 ; AVX512-NEXT:    pushq %r12
-; AVX512-NEXT:    .cfi_def_cfa_offset 48
 ; AVX512-NEXT:    pushq %rbx
-; AVX512-NEXT:    .cfi_def_cfa_offset 56
 ; AVX512-NEXT:    pushq %rax
-; AVX512-NEXT:    .cfi_def_cfa_offset 64
-; AVX512-NEXT:    .cfi_offset %rbx, -56
-; AVX512-NEXT:    .cfi_offset %r12, -48
-; AVX512-NEXT:    .cfi_offset %r13, -40
-; AVX512-NEXT:    .cfi_offset %r14, -32
-; AVX512-NEXT:    .cfi_offset %r15, -24
-; AVX512-NEXT:    .cfi_offset %rbp, -16
 ; AVX512-NEXT:    vextracti64x4 $1, %zmm0, %ymm2
 ; AVX512-NEXT:    vextracti128 $1, %ymm2, %xmm3
 ; AVX512-NEXT:    vpextrq $1, %xmm3, %rax
@@ -3796,19 +3613,12 @@ define <8 x i64> @test_ext_v8i64(<8 x i64> %a0, <8 x i64> %a1) {
 ; AVX512-NEXT:    vinserti128 $1, %xmm1, %ymm2, %ymm1
 ; AVX512-NEXT:    vinserti64x4 $1, %ymm0, %zmm1, %zmm0
 ; AVX512-NEXT:    addq $8, %rsp
-; AVX512-NEXT:    .cfi_def_cfa_offset 56
 ; AVX512-NEXT:    popq %rbx
-; AVX512-NEXT:    .cfi_def_cfa_offset 48
 ; AVX512-NEXT:    popq %r12
-; AVX512-NEXT:    .cfi_def_cfa_offset 40
 ; AVX512-NEXT:    popq %r13
-; AVX512-NEXT:    .cfi_def_cfa_offset 32
 ; AVX512-NEXT:    popq %r14
-; AVX512-NEXT:    .cfi_def_cfa_offset 24
 ; AVX512-NEXT:    popq %r15
-; AVX512-NEXT:    .cfi_def_cfa_offset 16
 ; AVX512-NEXT:    popq %rbp
-; AVX512-NEXT:    .cfi_def_cfa_offset 8
 ; AVX512-NEXT:    retq
   %x0 = sext <8 x i64> %a0 to <8 x i128>
   %x1 = sext <8 x i64> %a1 to <8 x i128>
diff --git a/llvm/test/CodeGen/X86/avgceilu.ll b/llvm/test/CodeGen/X86/avgceilu.ll
index dee1a5a720f9..d34894cc0fbb 100644
--- a/llvm/test/CodeGen/X86/avgceilu.ll
+++ b/llvm/test/CodeGen/X86/avgceilu.ll
@@ -9,7 +9,7 @@
 ; 128-bit vectors
 ;
 
-define <16 x i8> @test_fixed_v16i8(<16 x i8> %a0, <16 x i8> %a1) {
+define <16 x i8> @test_fixed_v16i8(<16 x i8> %a0, <16 x i8> %a1) nounwind {
 ; SSE-LABEL: test_fixed_v16i8:
 ; SSE:       # %bb.0:
 ; SSE-NEXT:    pavgb %xmm1, %xmm0
@@ -26,7 +26,7 @@ define <16 x i8> @test_fixed_v16i8(<16 x i8> %a0, <16 x i8> %a1) {
   ret <16 x i8> %res
 }
 
-define <16 x i8> @test_ext_v16i8(<16 x i8> %a0, <16 x i8> %a1) {
+define <16 x i8> @test_ext_v16i8(<16 x i8> %a0, <16 x i8> %a1) nounwind {
 ; SSE-LABEL: test_ext_v16i8:
 ; SSE:       # %bb.0:
 ; SSE-NEXT:    pavgb %xmm1, %xmm0
@@ -45,7 +45,7 @@ define <16 x i8> @test_ext_v16i8(<16 x i8> %a0, <16 x i8> %a1) {
   ret <16 x i8> %res
 }
 
-define <8 x i16> @test_fixed_v8i16(<8 x i16> %a0, <8 x i16> %a1) {
+define <8 x i16> @test_fixed_v8i16(<8 x i16> %a0, <8 x i16> %a1) nounwind {
 ; SSE-LABEL: test_fixed_v8i16:
 ; SSE:       # %bb.0:
 ; SSE-NEXT:    pavgw %xmm1, %xmm0
@@ -62,7 +62,7 @@ define <8 x i16> @test_fixed_v8i16(<8 x i16> %a0, <8 x i16> %a1) {
   ret <8 x i16> %res
 }
 
-define <8 x i16> @test_ext_v8i16(<8 x i16> %a0, <8 x i16> %a1) {
+define <8 x i16> @test_ext_v8i16(<8 x i16> %a0, <8 x i16> %a1) nounwind {
 ; SSE-LABEL: test_ext_v8i16:
 ; SSE:       # %bb.0:
 ; SSE-NEXT:    pavgw %xmm1, %xmm0
@@ -81,7 +81,7 @@ define <8 x i16> @test_ext_v8i16(<8 x i16> %a0, <8 x i16> %a1) {
   ret <8 x i16> %res
 }
 
-define <4 x i32> @test_fixed_v4i32(<4 x i32> %a0, <4 x i32> %a1) {
+define <4 x i32> @test_fixed_v4i32(<4 x i32> %a0, <4 x i32> %a1) nounwind {
 ; SSE-LABEL: test_fixed_v4i32:
 ; SSE:       # %bb.0:
 ; SSE-NEXT:    movdqa %xmm0, %xmm2
@@ -106,7 +106,7 @@ define <4 x i32> @test_fixed_v4i32(<4 x i32> %a0, <4 x i32> %a1) {
   ret <4 x i32> %res
 }
 
-define <4 x i32> @test_ext_v4i32(<4 x i32> %a0, <4 x i32> %a1) {
+define <4 x i32> @test_ext_v4i32(<4 x i32> %a0, <4 x i32> %a1) nounwind {
 ; SSE2-LABEL: test_ext_v4i32:
 ; SSE2:       # %bb.0:
 ; SSE2-NEXT:    pxor %xmm3, %xmm3
@@ -195,7 +195,7 @@ define <4 x i32> @test_ext_v4i32(<4 x i32> %a0, <4 x i32> %a1) {
   ret <4 x i32> %res
 }
 
-define <2 x i64> @test_fixed_v2i64(<2 x i64> %a0, <2 x i64> %a1) {
+define <2 x i64> @test_fixed_v2i64(<2 x i64> %a0, <2 x i64> %a1) nounwind {
 ; SSE-LABEL: test_fixed_v2i64:
 ; SSE:       # %bb.0:
 ; SSE-NEXT:    movdqa %xmm0, %xmm2
@@ -220,7 +220,7 @@ define <2 x i64> @test_fixed_v2i64(<2 x i64> %a0, <2 x i64> %a1) {
   ret <2 x i64> %res
 }
 
-define <2 x i64> @test_ext_v2i64(<2 x i64> %a0, <2 x i64> %a1) {
+define <2 x i64> @test_ext_v2i64(<2 x i64> %a0, <2 x i64> %a1) nounwind {
 ; SSE2-LABEL: test_ext_v2i64:
 ; SSE2:       # %bb.0:
 ; SSE2-NEXT:    pshufd {{.*#+}} xmm2 = xmm0[2,3,2,3]
@@ -310,7 +310,7 @@ define <2 x i64> @test_ext_v2i64(<2 x i64> %a0, <2 x i64> %a1) {
 ; 256-bit vectors
 ;
 
-define <32 x i8> @test_fixed_v32i8(<32 x i8> %a0, <32 x i8> %a1) {
+define <32 x i8> @test_fixed_v32i8(<32 x i8> %a0, <32 x i8> %a1) nounwind {
 ; SSE-LABEL: test_fixed_v32i8:
 ; SSE:       # %bb.0:
 ; SSE-NEXT:    pavgb %xmm2, %xmm0
@@ -342,7 +342,7 @@ define <32 x i8> @test_fixed_v32i8(<32 x i8> %a0, <32 x i8> %a1) {
   ret <32 x i8> %res
 }
 
-define <32 x i8> @test_ext_v32i8(<32 x i8> %a0, <32 x i8> %a1) {
+define <32 x i8> @test_ext_v32i8(<32 x i8> %a0, <32 x i8> %a1) nounwind {
 ; SSE-LABEL: test_ext_v32i8:
 ; SSE:       # %bb.0:
 ; SSE-NEXT:    pavgb %xmm2, %xmm0
@@ -376,7 +376,7 @@ define <32 x i8> @test_ext_v32i8(<32 x i8> %a0, <32 x i8> %a1) {
   ret <32 x i8> %res
 }
 
-define <16 x i16> @test_fixed_v16i16(<16 x i16> %a0, <16 x i16> %a1) {
+define <16 x i16> @test_fixed_v16i16(<16 x i16> %a0, <16 x i16> %a1) nounwind {
 ; SSE-LABEL: test_fixed_v16i16:
 ; SSE:       # %bb.0:
 ; SSE-NEXT:    pavgw %xmm2, %xmm0
@@ -408,7 +408,7 @@ define <16 x i16> @test_fixed_v16i16(<16 x i16> %a0, <16 x i16> %a1) {
   ret <16 x i16> %res
 }
 
-define <16 x i16> @test_ext_v16i16(<16 x i16> %a0, <16 x i16> %a1) {
+define <16 x i16> @test_ext_v16i16(<16 x i16> %a0, <16 x i16> %a1) nounwind {
 ; SSE-LABEL: test_ext_v16i16:
 ; SSE:       # %bb.0:
 ; SSE-NEXT:    pavgw %xmm2, %xmm0
@@ -442,7 +442,7 @@ define <16 x i16> @test_ext_v16i16(<16 x i16> %a0, <16 x i16> %a1) {
   ret <16 x i16> %res
 }
 
-define <8 x i32> @test_fixed_v8i32(<8 x i32> %a0, <8 x i32> %a1) {
+define <8 x i32> @test_fixed_v8i32(<8 x i32> %a0, <8 x i32> %a1) nounwind {
 ; SSE-LABEL: test_fixed_v8i32:
 ; SSE:       # %bb.0:
 ; SSE-NEXT:    movdqa %xmm1, %xmm4
@@ -494,7 +494,7 @@ define <8 x i32> @test_fixed_v8i32(<8 x i32> %a0, <8 x i32> %a1) {
   ret <8 x i32> %res
 }
 
-define <8 x i32> @test_ext_v8i32(<8 x i32> %a0, <8 x i32> %a1) {
+define <8 x i32> @test_ext_v8i32(<8 x i32> %a0, <8 x i32> %a1) nounwind {
 ; SSE2-LABEL: test_ext_v8i32:
 ; SSE2:       # %bb.0:
 ; SSE2-NEXT:    movdqa %xmm0, %xmm4
@@ -629,7 +629,7 @@ define <8 x i32> @test_ext_v8i32(<8 x i32> %a0, <8 x i32> %a1) {
   ret <8 x i32> %res
 }
 
-define <4 x i64> @test_fixed_v4i64(<4 x i64> %a0, <4 x i64> %a1) {
+define <4 x i64> @test_fixed_v4i64(<4 x i64> %a0, <4 x i64> %a1) nounwind {
 ; SSE-LABEL: test_fixed_v4i64:
 ; SSE:       # %bb.0:
 ; SSE-NEXT:    movdqa %xmm1, %xmm4
@@ -681,7 +681,7 @@ define <4 x i64> @test_fixed_v4i64(<4 x i64> %a0, <4 x i64> %a1) {
   ret <4 x i64> %res
 }
 
-define <4 x i64> @test_ext_v4i64(<4 x i64> %a0, <4 x i64> %a1) {
+define <4 x i64> @test_ext_v4i64(<4 x i64> %a0, <4 x i64> %a1) nounwind {
 ; SSE2-LABEL: test_ext_v4i64:
 ; SSE2:       # %bb.0:
 ; SSE2-NEXT:    pshufd {{.*#+}} xmm4 = xmm1[2,3,2,3]
@@ -937,7 +937,7 @@ define <4 x i64> @test_ext_v4i64(<4 x i64> %a0, <4 x i64> %a1) {
 ; 512-bit vectors
 ;
 
-define <64 x i8> @test_fixed_v64i8(<64 x i8> %a0, <64 x i8> %a1) {
+define <64 x i8> @test_fixed_v64i8(<64 x i8> %a0, <64 x i8> %a1) nounwind {
 ; SSE-LABEL: test_fixed_v64i8:
 ; SSE:       # %bb.0:
 ; SSE-NEXT:    pavgb %xmm4, %xmm0
@@ -977,7 +977,7 @@ define <64 x i8> @test_fixed_v64i8(<64 x i8> %a0, <64 x i8> %a1) {
   ret <64 x i8> %res
 }
 
-define <64 x i8> @test_ext_v64i8(<64 x i8> %a0, <64 x i8> %a1) {
+define <64 x i8> @test_ext_v64i8(<64 x i8> %a0, <64 x i8> %a1) nounwind {
 ; SSE-LABEL: test_ext_v64i8:
 ; SSE:       # %bb.0:
 ; SSE-NEXT:    pavgb %xmm4, %xmm0
@@ -1019,7 +1019,7 @@ define <64 x i8> @test_ext_v64i8(<64 x i8> %a0, <64 x i8> %a1) {
   ret <64 x i8> %res
 }
 
-define <32 x i16> @test_fixed_v32i16(<32 x i16> %a0, <32 x i16> %a1) {
+define <32 x i16> @test_fixed_v32i16(<32 x i16> %a0, <32 x i16> %a1) nounwind {
 ; SSE-LABEL: test_fixed_v32i16:
 ; SSE:       # %bb.0:
 ; SSE-NEXT:    pavgw %xmm4, %xmm0
@@ -1059,7 +1059,7 @@ define <32 x i16> @test_fixed_v32i16(<32 x i16> %a0, <32 x i16> %a1) {
   ret <32 x i16> %res
 }
 
-define <32 x i16> @test_ext_v32i16(<32 x i16> %a0, <32 x i16> %a1) {
+define <32 x i16> @test_ext_v32i16(<32 x i16> %a0, <32 x i16> %a1) nounwind {
 ; SSE-LABEL: test_ext_v32i16:
 ; SSE:       # %bb.0:
 ; SSE-NEXT:    pavgw %xmm4, %xmm0
@@ -1101,7 +1101,7 @@ define <32 x i16> @test_ext_v32i16(<32 x i16> %a0, <32 x i16> %a1) {
   ret <32 x i16> %res
 }
 
-define <16 x i32> @test_fixed_v16i32(<16 x i32> %a0, <16 x i32> %a1) {
+define <16 x i32> @test_fixed_v16i32(<16 x i32> %a0, <16 x i32> %a1) nounwind {
 ; SSE-LABEL: test_fixed_v16i32:
 ; SSE:       # %bb.0:
 ; SSE-NEXT:    movdqa %xmm3, %xmm8
@@ -1177,7 +1177,7 @@ define <16 x i32> @test_fixed_v16i32(<16 x i32> %a0, <16 x i32> %a1) {
   ret <16 x i32> %res
 }
 
-define <16 x i32> @test_ext_v16i32(<16 x i32> %a0, <16 x i32> %a1) {
+define <16 x i32> @test_ext_v16i32(<16 x i32> %a0, <16 x i32> %a1) nounwind {
 ; SSE2-LABEL: test_ext_v16i32:
 ; SSE2:       # %bb.0:
 ; SSE2-NEXT:    movdqa %xmm2, %xmm8
@@ -1413,7 +1413,7 @@ define <16 x i32> @test_ext_v16i32(<16 x i32> %a0, <16 x i32> %a1) {
   ret <16 x i32> %res
 }
 
-define <8 x i64> @test_fixed_v8i64(<8 x i64> %a0, <8 x i64> %a1) {
+define <8 x i64> @test_fixed_v8i64(<8 x i64> %a0, <8 x i64> %a1) nounwind {
 ; SSE-LABEL: test_fixed_v8i64:
 ; SSE:       # %bb.0:
 ; SSE-NEXT:    movdqa %xmm3, %xmm8
@@ -1489,27 +1489,15 @@ define <8 x i64> @test_fixed_v8i64(<8 x i64> %a0, <8 x i64> %a1) {
   ret <8 x i64> %res
 }
 
-define <8 x i64> @test_ext_v8i64(<8 x i64> %a0, <8 x i64> %a1) {
+define <8 x i64> @test_ext_v8i64(<8 x i64> %a0, <8 x i64> %a1) nounwind {
 ; SSE2-LABEL: test_ext_v8i64:
 ; SSE2:       # %bb.0:
 ; SSE2-NEXT:    pushq %rbp
-; SSE2-NEXT:    .cfi_def_cfa_offset 16
 ; SSE2-NEXT:    pushq %r15
-; SSE2-NEXT:    .cfi_def_cfa_offset 24
 ; SSE2-NEXT:    pushq %r14
-; SSE2-NEXT:    .cfi_def_cfa_offset 32
 ; SSE2-NEXT:    pushq %r13
-; SSE2-NEXT:    .cfi_def_cfa_offset 40
 ; SSE2-NEXT:    pushq %r12
-; SSE2-NEXT:    .cfi_def_cfa_offset 48
 ; SSE2-NEXT:    pushq %rbx
-; SSE2-NEXT:    .cfi_def_cfa_offset 56
-; SSE2-NEXT:    .cfi_offset %rbx, -56
-; SSE2-NEXT:    .cfi_offset %r12, -48
-; SSE2-NEXT:    .cfi_offset %r13, -40
-; SSE2-NEXT:    .cfi_offset %r14, -32
-; SSE2-NEXT:    .cfi_offset %r15, -24
-; SSE2-NEXT:    .cfi_offset %rbp, -16
 ; SSE2-NEXT:    pshufd {{.*#+}} xmm8 = xmm3[2,3,2,3]
 ; SSE2-NEXT:    movq %xmm8, %rcx
 ; SSE2-NEXT:    movq %rcx, {{[-0-9]+}}(%r{{[sb]}}p) # 8-byte Spill
@@ -1617,39 +1605,21 @@ define <8 x i64> @test_ext_v8i64(<8 x i64> %a0, <8 x i64> %a1) {
 ; SSE2-NEXT:    punpcklqdq {{.*#+}} xmm2 = xmm2[0],xmm6[0]
 ; SSE2-NEXT:    punpcklqdq {{.*#+}} xmm3 = xmm3[0],xmm7[0]
 ; SSE2-NEXT:    popq %rbx
-; SSE2-NEXT:    .cfi_def_cfa_offset 48
 ; SSE2-NEXT:    popq %r12
-; SSE2-NEXT:    .cfi_def_cfa_offset 40
 ; SSE2-NEXT:    popq %r13
-; SSE2-NEXT:    .cfi_def_cfa_offset 32
 ; SSE2-NEXT:    popq %r14
-; SSE2-NEXT:    .cfi_def_cfa_offset 24
 ; SSE2-NEXT:    popq %r15
-; SSE2-NEXT:    .cfi_def_cfa_offset 16
 ; SSE2-NEXT:    popq %rbp
-; SSE2-NEXT:    .cfi_def_cfa_offset 8
 ; SSE2-NEXT:    retq
 ;
 ; SSE4-LABEL: test_ext_v8i64:
 ; SSE4:       # %bb.0:
 ; SSE4-NEXT:    pushq %rbp
-; SSE4-NEXT:    .cfi_def_cfa_offset 16
 ; SSE4-NEXT:    pushq %r15
-; SSE4-NEXT:    .cfi_def_cfa_offset 24
 ; SSE4-NEXT:    pushq %r14
-; SSE4-NEXT:    .cfi_def_cfa_offset 32
 ; SSE4-NEXT:    pushq %r13
-; SSE4-NEXT:    .cfi_def_cfa_offset 40
 ; SSE4-NEXT:    pushq %r12
-; SSE4-NEXT:    .cfi_def_cfa_offset 48
 ; SSE4-NEXT:    pushq %rbx
-; SSE4-NEXT:    .cfi_def_cfa_offset 56
-; SSE4-NEXT:    .cfi_offset %rbx, -56
-; SSE4-NEXT:    .cfi_offset %r12, -48
-; SSE4-NEXT:    .cfi_offset %r13, -40
-; SSE4-NEXT:    .cfi_offset %r14, -32
-; SSE4-NEXT:    .cfi_offset %r15, -24
-; SSE4-NEXT:    .cfi_offset %rbp, -16
 ; SSE4-NEXT:    movq %xmm3, %rcx
 ; SSE4-NEXT:    movq %rcx, {{[-0-9]+}}(%r{{[sb]}}p) # 8-byte Spill
 ; SSE4-NEXT:    movq %xmm7, %rdx
@@ -1747,39 +1717,21 @@ define <8 x i64> @test_ext_v8i64(<8 x i64> %a0, <8 x i64> %a1) {
 ; SSE4-NEXT:    punpcklqdq {{.*#+}} xmm2 = xmm2[0],xmm6[0]
 ; SSE4-NEXT:    punpcklqdq {{.*#+}} xmm3 = xmm3[0],xmm7[0]
 ; SSE4-NEXT:    popq %rbx
-; SSE4-NEXT:    .cfi_def_cfa_offset 48
 ; SSE4-NEXT:    popq %r12
-; SSE4-NEXT:    .cfi_def_cfa_offset 40
 ; SSE4-NEXT:    popq %r13
-; SSE4-NEXT:    .cfi_def_cfa_offset 32
 ; SSE4-NEXT:    popq %r14
-; SSE4-NEXT:    .cfi_def_cfa_offset 24
 ; SSE4-NEXT:    popq %r15
-; SSE4-NEXT:    .cfi_def_cfa_offset 16
 ; SSE4-NEXT:    popq %rbp
-; SSE4-NEXT:    .cfi_def_cfa_offset 8
 ; SSE4-NEXT:    retq
 ;
 ; AVX1-LABEL: test_ext_v8i64:
 ; AVX1:       # %bb.0:
 ; AVX1-NEXT:    pushq %rbp
-; AVX1-NEXT:    .cfi_def_cfa_offset 16
 ; AVX1-NEXT:    pushq %r15
-; AVX1-NEXT:    .cfi_def_cfa_offset 24
 ; AVX1-NEXT:    pushq %r14
-; AVX1-NEXT:    .cfi_def_cfa_offset 32
 ; AVX1-NEXT:    pushq %r13
-; AVX1-NEXT:    .cfi_def_cfa_offset 40
 ; AVX1-NEXT:    pushq %r12
-; AVX1-NEXT:    .cfi_def_cfa_offset 48
 ; AVX1-NEXT:    pushq %rbx
-; AVX1-NEXT:    .cfi_def_cfa_offset 56
-; AVX1-NEXT:    .cfi_offset %rbx, -56
-; AVX1-NEXT:    .cfi_offset %r12, -48
-; AVX1-NEXT:    .cfi_offset %r13, -40
-; AVX1-NEXT:    .cfi_offset %r14, -32
-; AVX1-NEXT:    .cfi_offset %r15, -24
-; AVX1-NEXT:    .cfi_offset %rbp, -16
 ; AVX1-NEXT:    vmovq %xmm1, %rcx
 ; AVX1-NEXT:    movq %rcx, {{[-0-9]+}}(%r{{[sb]}}p) # 8-byte Spill
 ; AVX1-NEXT:    vmovq %xmm3, %rdx
@@ -1885,39 +1837,21 @@ define <8 x i64> @test_ext_v8i64(<8 x i64> %a0, <8 x i64> %a1) {
 ; AVX1-NEXT:    vpunpcklqdq {{.*#+}} xmm2 = xmm7[0],xmm6[0]
 ; AVX1-NEXT:    vinsertf128 $1, %xmm1, %ymm2, %ymm1
 ; AVX1-NEXT:    popq %rbx
-; AVX1-NEXT:    .cfi_def_cfa_offset 48
 ; AVX1-NEXT:    popq %r12
-; AVX1-NEXT:    .cfi_def_cfa_offset 40
 ; AVX1-NEXT:    popq %r13
-; AVX1-NEXT:    .cfi_def_cfa_offset 32
 ; AVX1-NEXT:    popq %r14
-; AVX1-NEXT:    .cfi_def_cfa_offset 24
 ; AVX1-NEXT:    popq %r15
-; AVX1-NEXT:    .cfi_def_cfa_offset 16
 ; AVX1-NEXT:    popq %rbp
-; AVX1-NEXT:    .cfi_def_cfa_offset 8
 ; AVX1-NEXT:    retq
 ;
 ; AVX2-LABEL: test_ext_v8i64:
 ; AVX2:       # %bb.0:
 ; AVX2-NEXT:    pushq %rbp
-; AVX2-NEXT:    .cfi_def_cfa_offset 16
 ; AVX2-NEXT:    pushq %r15
-; AVX2-NEXT:    .cfi_def_cfa_offset 24
 ; AVX2-NEXT:    pushq %r14
-; AVX2-NEXT:    .cfi_def_cfa_offset 32
 ; AVX2-NEXT:    pushq %r13
-; AVX2-NEXT:    .cfi_def_cfa_offset 40
 ; AVX2-NEXT:    pushq %r12
-; AVX2-NEXT:    .cfi_def_cfa_offset 48
 ; AVX2-NEXT:    pushq %rbx
-; AVX2-NEXT:    .cfi_def_cfa_offset 56
-; AVX2-NEXT:    .cfi_offset %rbx, -56
-; AVX2-NEXT:    .cfi_offset %r12, -48
-; AVX2-NEXT:    .cfi_offset %r13, -40
-; AVX2-NEXT:    .cfi_offset %r14, -32
-; AVX2-NEXT:    .cfi_offset %r15, -24
-; AVX2-NEXT:    .cfi_offset %rbp, -16
 ; AVX2-NEXT:    vmovq %xmm1, %rcx
 ; AVX2-NEXT:    movq %rcx, {{[-0-9]+}}(%r{{[sb]}}p) # 8-byte Spill
 ; AVX2-NEXT:    vmovq %xmm3, %rdx
@@ -2023,39 +1957,21 @@ define <8 x i64> @test_ext_v8i64(<8 x i64> %a0, <8 x i64> %a1) {
 ; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm2 = xmm7[0],xmm6[0]
 ; AVX2-NEXT:    vinserti128 $1, %xmm1, %ymm2, %ymm1
 ; AVX2-NEXT:    popq %rbx
-; AVX2-NEXT:    .cfi_def_cfa_offset 48
 ; AVX2-NEXT:    popq %r12
-; AVX2-NEXT:    .cfi_def_cfa_offset 40
 ; AVX2-NEXT:    popq %r13
-; AVX2-NEXT:    .cfi_def_cfa_offset 32
 ; AVX2-NEXT:    popq %r14
-; AVX2-NEXT:    .cfi_def_cfa_offset 24
 ; AVX2-NEXT:    popq %r15
-; AVX2-NEXT:    .cfi_def_cfa_offset 16
 ; AVX2-NEXT:    popq %rbp
-; AVX2-NEXT:    .cfi_def_cfa_offset 8
 ; AVX2-NEXT:    retq
 ;
 ; AVX512-LABEL: test_ext_v8i64:
 ; AVX512:       # %bb.0:
 ; AVX512-NEXT:    pushq %rbp
-; AVX512-NEXT:    .cfi_def_cfa_offset 16
 ; AVX512-NEXT:    pushq %r15
-; AVX512-NEXT:    .cfi_def_cfa_offset 24
 ; AVX512-NEXT:    pushq %r14
-; AVX512-NEXT:    .cfi_def_cfa_offset 32
 ; AVX512-NEXT:    pushq %r13
-; AVX512-NEXT:    .cfi_def_cfa_offset 40
 ; AVX512-NEXT:    pushq %r12
-; AVX512-NEXT:    .cfi_def_cfa_offset 48
 ; AVX512-NEXT:    pushq %rbx
-; AVX512-NEXT:    .cfi_def_cfa_offset 56
-; AVX512-NEXT:    .cfi_offset %rbx, -56
-; AVX512-NEXT:    .cfi_offset %r12, -48
-; AVX512-NEXT:    .cfi_offset %r13, -40
-; AVX512-NEXT:    .cfi_offset %r14, -32
-; AVX512-NEXT:    .cfi_offset %r15, -24
-; AVX512-NEXT:    .cfi_offset %rbp, -16
 ; AVX512-NEXT:    vmovq %xmm0, %rcx
 ; AVX512-NEXT:    movq %rcx, {{[-0-9]+}}(%r{{[sb]}}p) # 8-byte Spill
 ; AVX512-NEXT:    vmovq %xmm1, %rdx
@@ -2164,17 +2080,11 @@ define <8 x i64> @test_ext_v8i64(<8 x i64> %a0, <8 x i64> %a1) {
 ; AVX512-NEXT:    vinserti128 $1, %xmm1, %ymm2, %ymm1
 ; AVX512-NEXT:    vinserti64x4 $1, %ymm0, %zmm1, %zmm0
 ; AVX512-NEXT:    popq %rbx
-; AVX512-NEXT:    .cfi_def_cfa_offset 48
 ; AVX512-NEXT:    popq %r12
-; AVX512-NEXT:    .cfi_def_cfa_offset 40
 ; AVX512-NEXT:    popq %r13
-; AVX512-NEXT:    .cfi_def_cfa_offset 32
 ; AVX512-NEXT:    popq %r14
-; AVX512-NEXT:    .cfi_def_cfa_offset 24
 ; AVX512-NEXT:    popq %r15
-; AVX512-NEXT:    .cfi_def_cfa_offset 16
 ; AVX512-NEXT:    popq %rbp
-; AVX512-NEXT:    .cfi_def_cfa_offset 8
 ; AVX512-NEXT:    retq
   %x0 = zext <8 x i64> %a0 to <8 x i128>
   %x1 = zext <8 x i64> %a1 to <8 x i128>
diff --git a/llvm/test/CodeGen/X86/avgfloors.ll b/llvm/test/CodeGen/X86/avgfloors.ll
index a3864ab4bb44..efee831a15c7 100644
--- a/llvm/test/CodeGen/X86/avgfloors.ll
+++ b/llvm/test/CodeGen/X86/avgfloors.ll
@@ -9,7 +9,7 @@
 ; 128-bit vectors
 ;
 
-define <16 x i8> @test_fixed_v16i8(<16 x i8> %a0, <16 x i8> %a1) {
+define <16 x i8> @test_fixed_v16i8(<16 x i8> %a0, <16 x i8> %a1) nounwind {
 ; SSE-LABEL: test_fixed_v16i8:
 ; SSE:       # %bb.0:
 ; SSE-NEXT:    movdqa %xmm0, %xmm2
@@ -64,7 +64,7 @@ define <16 x i8> @test_fixed_v16i8(<16 x i8> %a0, <16 x i8> %a1) {
   ret <16 x i8> %res
 }
 
-define <16 x i8> @test_ext_v16i8(<16 x i8> %a0, <16 x i8> %a1) {
+define <16 x i8> @test_ext_v16i8(<16 x i8> %a0, <16 x i8> %a1) nounwind {
 ; SSE2-LABEL: test_ext_v16i8:
 ; SSE2:       # %bb.0:
 ; SSE2-NEXT:    punpcklbw {{.*#+}} xmm2 = xmm2[0],xmm0[0],xmm2[1],xmm0[1],xmm2[2],xmm0[2],xmm2[3],xmm0[3],xmm2[4],xmm0[4],xmm2[5],xmm0[5],xmm2[6],xmm0[6],xmm2[7],xmm0[7]
@@ -150,7 +150,7 @@ define <16 x i8> @test_ext_v16i8(<16 x i8> %a0, <16 x i8> %a1) {
   ret <16 x i8> %res
 }
 
-define <8 x i16> @test_fixed_v8i16(<8 x i16> %a0, <8 x i16> %a1) {
+define <8 x i16> @test_fixed_v8i16(<8 x i16> %a0, <8 x i16> %a1) nounwind {
 ; SSE-LABEL: test_fixed_v8i16:
 ; SSE:       # %bb.0:
 ; SSE-NEXT:    movdqa %xmm0, %xmm2
@@ -174,7 +174,7 @@ define <8 x i16> @test_fixed_v8i16(<8 x i16> %a0, <8 x i16> %a1) {
   ret <8 x i16> %res
 }
 
-define <8 x i16> @test_ext_v8i16(<8 x i16> %a0, <8 x i16> %a1) {
+define <8 x i16> @test_ext_v8i16(<8 x i16> %a0, <8 x i16> %a1) nounwind {
 ; SSE2-LABEL: test_ext_v8i16:
 ; SSE2:       # %bb.0:
 ; SSE2-NEXT:    punpckhwd {{.*#+}} xmm2 = xmm2[4],xmm0[4],xmm2[5],xmm0[5],xmm2[6],xmm0[6],xmm2[7],xmm0[7]
@@ -259,7 +259,7 @@ define <8 x i16> @test_ext_v8i16(<8 x i16> %a0, <8 x i16> %a1) {
   ret <8 x i16> %res
 }
 
-define <4 x i32> @test_fixed_v4i32(<4 x i32> %a0, <4 x i32> %a1) {
+define <4 x i32> @test_fixed_v4i32(<4 x i32> %a0, <4 x i32> %a1) nounwind {
 ; SSE-LABEL: test_fixed_v4i32:
 ; SSE:       # %bb.0:
 ; SSE-NEXT:    movdqa %xmm0, %xmm2
@@ -283,7 +283,7 @@ define <4 x i32> @test_fixed_v4i32(<4 x i32> %a0, <4 x i32> %a1) {
   ret <4 x i32> %res
 }
 
-define <4 x i32> @test_ext_v4i32(<4 x i32> %a0, <4 x i32> %a1) {
+define <4 x i32> @test_ext_v4i32(<4 x i32> %a0, <4 x i32> %a1) nounwind {
 ; SSE2-LABEL: test_ext_v4i32:
 ; SSE2:       # %bb.0:
 ; SSE2-NEXT:    pxor %xmm2, %xmm2
@@ -365,7 +365,7 @@ define <4 x i32> @test_ext_v4i32(<4 x i32> %a0, <4 x i32> %a1) {
   ret <4 x i32> %res
 }
 
-define <2 x i64> @test_fixed_v2i64(<2 x i64> %a0, <2 x i64> %a1) {
+define <2 x i64> @test_fixed_v2i64(<2 x i64> %a0, <2 x i64> %a1) nounwind {
 ; SSE2-LABEL: test_fixed_v2i64:
 ; SSE2:       # %bb.0:
 ; SSE2-NEXT:    movdqa %xmm0, %xmm2
@@ -425,7 +425,7 @@ define <2 x i64> @test_fixed_v2i64(<2 x i64> %a0, <2 x i64> %a1) {
   ret <2 x i64> %res
 }
 
-define <2 x i64> @test_ext_v2i64(<2 x i64> %a0, <2 x i64> %a1) {
+define <2 x i64> @test_ext_v2i64(<2 x i64> %a0, <2 x i64> %a1) nounwind {
 ; SSE2-LABEL: test_ext_v2i64:
 ; SSE2:       # %bb.0:
 ; SSE2-NEXT:    pshufd {{.*#+}} xmm2 = xmm0[2,3,2,3]
@@ -514,7 +514,7 @@ define <2 x i64> @test_ext_v2i64(<2 x i64> %a0, <2 x i64> %a1) {
 ; 256-bit vectors
 ;
 
-define <32 x i8> @test_fixed_v32i8(<32 x i8> %a0, <32 x i8> %a1) {
+define <32 x i8> @test_fixed_v32i8(<32 x i8> %a0, <32 x i8> %a1) nounwind {
 ; SSE-LABEL: test_fixed_v32i8:
 ; SSE:       # %bb.0:
 ; SSE-NEXT:    movdqa %xmm1, %xmm4
@@ -587,7 +587,7 @@ define <32 x i8> @test_fixed_v32i8(<32 x i8> %a0, <32 x i8> %a1) {
   ret <32 x i8> %res
 }
 
-define <32 x i8> @test_ext_v32i8(<32 x i8> %a0, <32 x i8> %a1) {
+define <32 x i8> @test_ext_v32i8(<32 x i8> %a0, <32 x i8> %a1) nounwind {
 ; SSE2-LABEL: test_ext_v32i8:
 ; SSE2:       # %bb.0:
 ; SSE2-NEXT:    punpcklbw {{.*#+}} xmm4 = xmm4[0],xmm1[0],xmm4[1],xmm1[1],xmm4[2],xmm1[2],xmm4[3],xmm1[3],xmm4[4],xmm1[4],xmm4[5],xmm1[5],xmm4[6],xmm1[6],xmm4[7],xmm1[7]
@@ -723,7 +723,7 @@ define <32 x i8> @test_ext_v32i8(<32 x i8> %a0, <32 x i8> %a1) {
   ret <32 x i8> %res
 }
 
-define <16 x i16> @test_fixed_v16i16(<16 x i16> %a0, <16 x i16> %a1) {
+define <16 x i16> @test_fixed_v16i16(<16 x i16> %a0, <16 x i16> %a1) nounwind {
 ; SSE-LABEL: test_fixed_v16i16:
 ; SSE:       # %bb.0:
 ; SSE-NEXT:    movdqa %xmm1, %xmm4
@@ -773,7 +773,7 @@ define <16 x i16> @test_fixed_v16i16(<16 x i16> %a0, <16 x i16> %a1) {
   ret <16 x i16> %res
 }
 
-define <16 x i16> @test_ext_v16i16(<16 x i16> %a0, <16 x i16> %a1) {
+define <16 x i16> @test_ext_v16i16(<16 x i16> %a0, <16 x i16> %a1) nounwind {
 ; SSE2-LABEL: test_ext_v16i16:
 ; SSE2:       # %bb.0:
 ; SSE2-NEXT:    punpckhwd {{.*#+}} xmm4 = xmm4[4],xmm0[4],xmm4[5],xmm0[5],xmm4[6],xmm0[6],xmm4[7],xmm0[7]
@@ -908,7 +908,7 @@ define <16 x i16> @test_ext_v16i16(<16 x i16> %a0, <16 x i16> %a1) {
   ret <16 x i16> %res
 }
 
-define <8 x i32> @test_fixed_v8i32(<8 x i32> %a0, <8 x i32> %a1) {
+define <8 x i32> @test_fixed_v8i32(<8 x i32> %a0, <8 x i32> %a1) nounwind {
 ; SSE-LABEL: test_fixed_v8i32:
 ; SSE:       # %bb.0:
 ; SSE-NEXT:    movdqa %xmm1, %xmm4
@@ -958,7 +958,7 @@ define <8 x i32> @test_fixed_v8i32(<8 x i32> %a0, <8 x i32> %a1) {
   ret <8 x i32> %res
 }
 
-define <8 x i32> @test_ext_v8i32(<8 x i32> %a0, <8 x i32> %a1) {
+define <8 x i32> @test_ext_v8i32(<8 x i32> %a0, <8 x i32> %a1) nounwind {
 ; SSE2-LABEL: test_ext_v8i32:
 ; SSE2:       # %bb.0:
 ; SSE2-NEXT:    pxor %xmm4, %xmm4
@@ -1089,7 +1089,7 @@ define <8 x i32> @test_ext_v8i32(<8 x i32> %a0, <8 x i32> %a1) {
   ret <8 x i32> %res
 }
 
-define <4 x i64> @test_fixed_v4i64(<4 x i64> %a0, <4 x i64> %a1) {
+define <4 x i64> @test_fixed_v4i64(<4 x i64> %a0, <4 x i64> %a1) nounwind {
 ; SSE2-LABEL: test_fixed_v4i64:
 ; SSE2:       # %bb.0:
 ; SSE2-NEXT:    movdqa %xmm1, %xmm4
@@ -1173,27 +1173,15 @@ define <4 x i64> @test_fixed_v4i64(<4 x i64> %a0, <4 x i64> %a1) {
   ret <4 x i64> %res
 }
 
-define <4 x i64> @test_ext_v4i64(<4 x i64> %a0, <4 x i64> %a1) {
+define <4 x i64> @test_ext_v4i64(<4 x i64> %a0, <4 x i64> %a1) nounwind {
 ; SSE2-LABEL: test_ext_v4i64:
 ; SSE2:       # %bb.0:
 ; SSE2-NEXT:    pushq %rbp
-; SSE2-NEXT:    .cfi_def_cfa_offset 16
 ; SSE2-NEXT:    pushq %r15
-; SSE2-NEXT:    .cfi_def_cfa_offset 24
 ; SSE2-NEXT:    pushq %r14
-; SSE2-NEXT:    .cfi_def_cfa_offset 32
 ; SSE2-NEXT:    pushq %r13
-; SSE2-NEXT:    .cfi_def_cfa_offset 40
 ; SSE2-NEXT:    pushq %r12
-; SSE2-NEXT:    .cfi_def_cfa_offset 48
 ; SSE2-NEXT:    pushq %rbx
-; SSE2-NEXT:    .cfi_def_cfa_offset 56
-; SSE2-NEXT:    .cfi_offset %rbx, -56
-; SSE2-NEXT:    .cfi_offset %r12, -48
-; SSE2-NEXT:    .cfi_offset %r13, -40
-; SSE2-NEXT:    .cfi_offset %r14, -32
-; SSE2-NEXT:    .cfi_offset %r15, -24
-; SSE2-NEXT:    .cfi_offset %rbp, -16
 ; SSE2-NEXT:    pshufd {{.*#+}} xmm4 = xmm1[2,3,2,3]
 ; SSE2-NEXT:    movq %xmm4, %rdx
 ; SSE2-NEXT:    movq %rdx, %r14
@@ -1241,39 +1229,21 @@ define <4 x i64> @test_ext_v4i64(<4 x i64> %a0, <4 x i64> %a1) {
 ; SSE2-NEXT:    punpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm2[0]
 ; SSE2-NEXT:    punpcklqdq {{.*#+}} xmm1 = xmm1[0],xmm3[0]
 ; SSE2-NEXT:    popq %rbx
-; SSE2-NEXT:    .cfi_def_cfa_offset 48
 ; SSE2-NEXT:    popq %r12
-; SSE2-NEXT:    .cfi_def_cfa_offset 40
 ; SSE2-NEXT:    popq %r13
-; SSE2-NEXT:    .cfi_def_cfa_offset 32
 ; SSE2-NEXT:    popq %r14
-; SSE2-NEXT:    .cfi_def_cfa_offset 24
 ; SSE2-NEXT:    popq %r15
-; SSE2-NEXT:    .cfi_def_cfa_offset 16
 ; SSE2-NEXT:    popq %rbp
-; SSE2-NEXT:    .cfi_def_cfa_offset 8
 ; SSE2-NEXT:    retq
 ;
 ; SSE4-LABEL: test_ext_v4i64:
 ; SSE4:       # %bb.0:
 ; SSE4-NEXT:    pushq %rbp
-; SSE4-NEXT:    .cfi_def_cfa_offset 16
 ; SSE4-NEXT:    pushq %r15
-; SSE4-NEXT:    .cfi_def_cfa_offset 24
 ; SSE4-NEXT:    pushq %r14
-; SSE4-NEXT:    .cfi_def_cfa_offset 32
 ; SSE4-NEXT:    pushq %r13
-; SSE4-NEXT:    .cfi_def_cfa_offset 40
 ; SSE4-NEXT:    pushq %r12
-; SSE4-NEXT:    .cfi_def_cfa_offset 48
 ; SSE4-NEXT:    pushq %rbx
-; SSE4-NEXT:    .cfi_def_cfa_offset 56
-; SSE4-NEXT:    .cfi_offset %rbx, -56
-; SSE4-NEXT:    .cfi_offset %r12, -48
-; SSE4-NEXT:    .cfi_offset %r13, -40
-; SSE4-NEXT:    .cfi_offset %r14, -32
-; SSE4-NEXT:    .cfi_offset %r15, -24
-; SSE4-NEXT:    .cfi_offset %rbp, -16
 ; SSE4-NEXT:    movq %xmm1, %rdi
 ; SSE4-NEXT:    movq %rdi, %r14
 ; SSE4-NEXT:    sarq $63, %r14
@@ -1317,39 +1287,21 @@ define <4 x i64> @test_ext_v4i64(<4 x i64> %a0, <4 x i64> %a1) {
 ; SSE4-NEXT:    punpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm2[0]
 ; SSE4-NEXT:    punpcklqdq {{.*#+}} xmm1 = xmm1[0],xmm3[0]
 ; SSE4-NEXT:    popq %rbx
-; SSE4-NEXT:    .cfi_def_cfa_offset 48
 ; SSE4-NEXT:    popq %r12
-; SSE4-NEXT:    .cfi_def_cfa_offset 40
 ; SSE4-NEXT:    popq %r13
-; SSE4-NEXT:    .cfi_def_cfa_offset 32
 ; SSE4-NEXT:    popq %r14
-; SSE4-NEXT:    .cfi_def_cfa_offset 24
 ; SSE4-NEXT:    popq %r15
-; SSE4-NEXT:    .cfi_def_cfa_offset 16
 ; SSE4-NEXT:    popq %rbp
-; SSE4-NEXT:    .cfi_def_cfa_offset 8
 ; SSE4-NEXT:    retq
 ;
 ; AVX1-LABEL: test_ext_v4i64:
 ; AVX1:       # %bb.0:
 ; AVX1-NEXT:    pushq %rbp
-; AVX1-NEXT:    .cfi_def_cfa_offset 16
 ; AVX1-NEXT:    pushq %r15
-; AVX1-NEXT:    .cfi_def_cfa_offset 24
 ; AVX1-NEXT:    pushq %r14
-; AVX1-NEXT:    .cfi_def_cfa_offset 32
 ; AVX1-NEXT:    pushq %r13
-; AVX1-NEXT:    .cfi_def_cfa_offset 40
 ; AVX1-NEXT:    pushq %r12
-; AVX1-NEXT:    .cfi_def_cfa_offset 48
 ; AVX1-NEXT:    pushq %rbx
-; AVX1-NEXT:    .cfi_def_cfa_offset 56
-; AVX1-NEXT:    .cfi_offset %rbx, -56
-; AVX1-NEXT:    .cfi_offset %r12, -48
-; AVX1-NEXT:    .cfi_offset %r13, -40
-; AVX1-NEXT:    .cfi_offset %r14, -32
-; AVX1-NEXT:    .cfi_offset %r15, -24
-; AVX1-NEXT:    .cfi_offset %rbp, -16
 ; AVX1-NEXT:    vmovq %xmm0, %rdx
 ; AVX1-NEXT:    movq %rdx, %r14
 ; AVX1-NEXT:    sarq $63, %r14
@@ -1396,39 +1348,21 @@ define <4 x i64> @test_ext_v4i64(<4 x i64> %a0, <4 x i64> %a1) {
 ; AVX1-NEXT:    vpunpcklqdq {{.*#+}} xmm1 = xmm3[0],xmm2[0]
 ; AVX1-NEXT:    vinsertf128 $1, %xmm0, %ymm1, %ymm0
 ; AVX1-NEXT:    popq %rbx
-; AVX1-NEXT:    .cfi_def_cfa_offset 48
 ; AVX1-NEXT:    popq %r12
-; AVX1-NEXT:    .cfi_def_cfa_offset 40
 ; AVX1-NEXT:    popq %r13
-; AVX1-NEXT:    .cfi_def_cfa_offset 32
 ; AVX1-NEXT:    popq %r14
-; AVX1-NEXT:    .cfi_def_cfa_offset 24
 ; AVX1-NEXT:    popq %r15
-; AVX1-NEXT:    .cfi_def_cfa_offset 16
 ; AVX1-NEXT:    popq %rbp
-; AVX1-NEXT:    .cfi_def_cfa_offset 8
 ; AVX1-NEXT:    retq
 ;
 ; AVX2-LABEL: test_ext_v4i64:
 ; AVX2:       # %bb.0:
 ; AVX2-NEXT:    pushq %rbp
-; AVX2-NEXT:    .cfi_def_cfa_offset 16
 ; AVX2-NEXT:    pushq %r15
-; AVX2-NEXT:    .cfi_def_cfa_offset 24
 ; AVX2-NEXT:    pushq %r14
-; AVX2-NEXT:    .cfi_def_cfa_offset 32
 ; AVX2-NEXT:    pushq %r13
-; AVX2-NEXT:    .cfi_def_cfa_offset 40
 ; AVX2-NEXT:    pushq %r12
-; AVX2-NEXT:    .cfi_def_cfa_offset 48
 ; AVX2-NEXT:    pushq %rbx
-; AVX2-NEXT:    .cfi_def_cfa_offset 56
-; AVX2-NEXT:    .cfi_offset %rbx, -56
-; AVX2-NEXT:    .cfi_offset %r12, -48
-; AVX2-NEXT:    .cfi_offset %r13, -40
-; AVX2-NEXT:    .cfi_offset %r14, -32
-; AVX2-NEXT:    .cfi_offset %r15, -24
-; AVX2-NEXT:    .cfi_offset %rbp, -16
 ; AVX2-NEXT:    vmovq %xmm0, %rdx
 ; AVX2-NEXT:    movq %rdx, %r14
 ; AVX2-NEXT:    sarq $63, %r14
@@ -1475,39 +1409,21 @@ define <4 x i64> @test_ext_v4i64(<4 x i64> %a0, <4 x i64> %a1) {
 ; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm1 = xmm3[0],xmm2[0]
 ; AVX2-NEXT:    vinserti128 $1, %xmm0, %ymm1, %ymm0
 ; AVX2-NEXT:    popq %rbx
-; AVX2-NEXT:    .cfi_def_cfa_offset 48
 ; AVX2-NEXT:    popq %r12
-; AVX2-NEXT:    .cfi_def_cfa_offset 40
 ; AVX2-NEXT:    popq %r13
-; AVX2-NEXT:    .cfi_def_cfa_offset 32
 ; AVX2-NEXT:    popq %r14
-; AVX2-NEXT:    .cfi_def_cfa_offset 24
 ; AVX2-NEXT:    popq %r15
-; AVX2-NEXT:    .cfi_def_cfa_offset 16
 ; AVX2-NEXT:    popq %rbp
-; AVX2-NEXT:    .cfi_def_cfa_offset 8
 ; AVX2-NEXT:    retq
 ;
 ; AVX512-LABEL: test_ext_v4i64:
 ; AVX512:       # %bb.0:
 ; AVX512-NEXT:    pushq %rbp
-; AVX512-NEXT:    .cfi_def_cfa_offset 16
 ; AVX512-NEXT:    pushq %r15
-; AVX512-NEXT:    .cfi_def_cfa_offset 24
 ; AVX512-NEXT:    pushq %r14
-; AVX512-NEXT:    .cfi_def_cfa_offset 32
 ; AVX512-NEXT:    pushq %r13
-; AVX512-NEXT:    .cfi_def_cfa_offset 40
 ; AVX512-NEXT:    pushq %r12
-; AVX512-NEXT:    .cfi_def_cfa_offset 48
 ; AVX512-NEXT:    pushq %rbx
-; AVX512-NEXT:    .cfi_def_cfa_offset 56
-; AVX512-NEXT:    .cfi_offset %rbx, -56
-; AVX512-NEXT:    .cfi_offset %r12, -48
-; AVX512-NEXT:    .cfi_offset %r13, -40
-; AVX512-NEXT:    .cfi_offset %r14, -32
-; AVX512-NEXT:    .cfi_offset %r15, -24
-; AVX512-NEXT:    .cfi_offset %rbp, -16
 ; AVX512-NEXT:    vmovq %xmm0, %rdx
 ; AVX512-NEXT:    movq %rdx, %r14
 ; AVX512-NEXT:    sarq $63, %r14
@@ -1554,17 +1470,11 @@ define <4 x i64> @test_ext_v4i64(<4 x i64> %a0, <4 x i64> %a1) {
 ; AVX512-NEXT:    vpunpcklqdq {{.*#+}} xmm1 = xmm3[0],xmm2[0]
 ; AVX512-NEXT:    vinserti128 $1, %xmm0, %ymm1, %ymm0
 ; AVX512-NEXT:    popq %rbx
-; AVX512-NEXT:    .cfi_def_cfa_offset 48
 ; AVX512-NEXT:    popq %r12
-; AVX512-NEXT:    .cfi_def_cfa_offset 40
 ; AVX512-NEXT:    popq %r13
-; AVX512-NEXT:    .cfi_def_cfa_offset 32
 ; AVX512-NEXT:    popq %r14
-; AVX512-NEXT:    .cfi_def_cfa_offset 24
 ; AVX512-NEXT:    popq %r15
-; AVX512-NEXT:    .cfi_def_cfa_offset 16
 ; AVX512-NEXT:    popq %rbp
-; AVX512-NEXT:    .cfi_def_cfa_offset 8
 ; AVX512-NEXT:    retq
   %x0 = sext <4 x i64> %a0 to <4 x i128>
   %x1 = sext <4 x i64> %a1 to <4 x i128>
@@ -1578,7 +1488,7 @@ define <4 x i64> @test_ext_v4i64(<4 x i64> %a0, <4 x i64> %a1) {
 ; 512-bit vectors
 ;
 
-define <64 x i8> @test_fixed_v64i8(<64 x i8> %a0, <64 x i8> %a1) {
+define <64 x i8> @test_fixed_v64i8(<64 x i8> %a0, <64 x i8> %a1) nounwind {
 ; SSE-LABEL: test_fixed_v64i8:
 ; SSE:       # %bb.0:
 ; SSE-NEXT:    movdqa %xmm3, %xmm10
@@ -1690,7 +1600,7 @@ define <64 x i8> @test_fixed_v64i8(<64 x i8> %a0, <64 x i8> %a1) {
   ret <64 x i8> %res
 }
 
-define <64 x i8> @test_ext_v64i8(<64 x i8> %a0, <64 x i8> %a1) {
+define <64 x i8> @test_ext_v64i8(<64 x i8> %a0, <64 x i8> %a1) nounwind {
 ; SSE2-LABEL: test_ext_v64i8:
 ; SSE2:       # %bb.0:
 ; SSE2-NEXT:    punpcklbw {{.*#+}} xmm13 = xmm13[0],xmm3[0],xmm13[1],xmm3[1],xmm13[2],xmm3[2],xmm13[3],xmm3[3],xmm13[4],xmm3[4],xmm13[5],xmm3[5],xmm13[6],xmm3[6],xmm13[7],xmm3[7]
@@ -1934,7 +1844,7 @@ define <64 x i8> @test_ext_v64i8(<64 x i8> %a0, <64 x i8> %a1) {
   ret <64 x i8> %res
 }
 
-define <32 x i16> @test_fixed_v32i16(<32 x i16> %a0, <32 x i16> %a1) {
+define <32 x i16> @test_fixed_v32i16(<32 x i16> %a0, <32 x i16> %a1) nounwind {
 ; SSE-LABEL: test_fixed_v32i16:
 ; SSE:       # %bb.0:
 ; SSE-NEXT:    movdqa %xmm3, %xmm8
@@ -2007,7 +1917,7 @@ define <32 x i16> @test_fixed_v32i16(<32 x i16> %a0, <32 x i16> %a1) {
   ret <32 x i16> %res
 }
 
-define <32 x i16> @test_ext_v32i16(<32 x i16> %a0, <32 x i16> %a1) {
+define <32 x i16> @test_ext_v32i16(<32 x i16> %a0, <32 x i16> %a1) nounwind {
 ; SSE2-LABEL: test_ext_v32i16:
 ; SSE2:       # %bb.0:
 ; SSE2-NEXT:    movdqa %xmm3, %xmm9
@@ -2251,7 +2161,7 @@ define <32 x i16> @test_ext_v32i16(<32 x i16> %a0, <32 x i16> %a1) {
   ret <32 x i16> %res
 }
 
-define <16 x i32> @test_fixed_v16i32(<16 x i32> %a0, <16 x i32> %a1) {
+define <16 x i32> @test_fixed_v16i32(<16 x i32> %a0, <16 x i32> %a1) nounwind {
 ; SSE-LABEL: test_fixed_v16i32:
 ; SSE:       # %bb.0:
 ; SSE-NEXT:    movdqa %xmm3, %xmm8
@@ -2324,7 +2234,7 @@ define <16 x i32> @test_fixed_v16i32(<16 x i32> %a0, <16 x i32> %a1) {
   ret <16 x i32> %res
 }
 
-define <16 x i32> @test_ext_v16i32(<16 x i32> %a0, <16 x i32> %a1) {
+define <16 x i32> @test_ext_v16i32(<16 x i32> %a0, <16 x i32> %a1) nounwind {
 ; SSE2-LABEL: test_ext_v16i32:
 ; SSE2:       # %bb.0:
 ; SSE2-NEXT:    pxor %xmm8, %xmm8
@@ -2561,7 +2471,7 @@ define <16 x i32> @test_ext_v16i32(<16 x i32> %a0, <16 x i32> %a1) {
   ret <16 x i32> %res
 }
 
-define <8 x i64> @test_fixed_v8i64(<8 x i64> %a0, <8 x i64> %a1) {
+define <8 x i64> @test_fixed_v8i64(<8 x i64> %a0, <8 x i64> %a1) nounwind {
 ; SSE2-LABEL: test_fixed_v8i64:
 ; SSE2:       # %bb.0:
 ; SSE2-NEXT:    movdqa %xmm3, %xmm11
@@ -2698,29 +2608,16 @@ define <8 x i64> @test_fixed_v8i64(<8 x i64> %a0, <8 x i64> %a1) {
   ret <8 x i64> %res
 }
 
-define <8 x i64> @test_ext_v8i64(<8 x i64> %a0, <8 x i64> %a1) {
+define <8 x i64> @test_ext_v8i64(<8 x i64> %a0, <8 x i64> %a1) nounwind {
 ; SSE2-LABEL: test_ext_v8i64:
 ; SSE2:       # %bb.0:
 ; SSE2-NEXT:    pushq %rbp
-; SSE2-NEXT:    .cfi_def_cfa_offset 16
 ; SSE2-NEXT:    pushq %r15
-; SSE2-NEXT:    .cfi_def_cfa_offset 24
 ; SSE2-NEXT:    pushq %r14
-; SSE2-NEXT:    .cfi_def_cfa_offset 32
 ; SSE2-NEXT:    pushq %r13
-; SSE2-NEXT:    .cfi_def_cfa_offset 40
 ; SSE2-NEXT:    pushq %r12
-; SSE2-NEXT:    .cfi_def_cfa_offset 48
 ; SSE2-NEXT:    pushq %rbx
-; SSE2-NEXT:    .cfi_def_cfa_offset 56
 ; SSE2-NEXT:    pushq %rax
-; SSE2-NEXT:    .cfi_def_cfa_offset 64
-; SSE2-NEXT:    .cfi_offset %rbx, -56
-; SSE2-NEXT:    .cfi_offset %r12, -48
-; SSE2-NEXT:    .cfi_offset %r13, -40
-; SSE2-NEXT:    .cfi_offset %r14, -32
-; SSE2-NEXT:    .cfi_offset %r15, -24
-; SSE2-NEXT:    .cfi_offset %rbp, -16
 ; SSE2-NEXT:    pshufd {{.*#+}} xmm8 = xmm3[2,3,2,3]
 ; SSE2-NEXT:    movq %xmm8, %rax
 ; SSE2-NEXT:    movq %rax, {{[-0-9]+}}(%r{{[sb]}}p) # 8-byte Spill
@@ -2832,43 +2729,23 @@ define <8 x i64> @test_ext_v8i64(<8 x i64> %a0, <8 x i64> %a1) {
 ; SSE2-NEXT:    punpcklqdq {{.*#+}} xmm2 = xmm2[0],xmm6[0]
 ; SSE2-NEXT:    punpcklqdq {{.*#+}} xmm3 = xmm3[0],xmm7[0]
 ; SSE2-NEXT:    addq $8, %rsp
-; SSE2-NEXT:    .cfi_def_cfa_offset 56
 ; SSE2-NEXT:    popq %rbx
-; SSE2-NEXT:    .cfi_def_cfa_offset 48
 ; SSE2-NEXT:    popq %r12
-; SSE2-NEXT:    .cfi_def_cfa_offset 40
 ; SSE2-NEXT:    popq %r13
-; SSE2-NEXT:    .cfi_def_cfa_offset 32
 ; SSE2-NEXT:    popq %r14
-; SSE2-NEXT:    .cfi_def_cfa_offset 24
 ; SSE2-NEXT:    popq %r15
-; SSE2-NEXT:    .cfi_def_cfa_offset 16
 ; SSE2-NEXT:    popq %rbp
-; SSE2-NEXT:    .cfi_def_cfa_offset 8
 ; SSE2-NEXT:    retq
 ;
 ; SSE4-LABEL: test_ext_v8i64:
 ; SSE4:       # %bb.0:
 ; SSE4-NEXT:    pushq %rbp
-; SSE4-NEXT:    .cfi_def_cfa_offset 16
 ; SSE4-NEXT:    pushq %r15
-; SSE4-NEXT:    .cfi_def_cfa_offset 24
 ; SSE4-NEXT:    pushq %r14
-; SSE4-NEXT:    .cfi_def_cfa_offset 32
 ; SSE4-NEXT:    pushq %r13
-; SSE4-NEXT:    .cfi_def_cfa_offset 40
 ; SSE4-NEXT:    pushq %r12
-; SSE4-NEXT:    .cfi_def_cfa_offset 48
 ; SSE4-NEXT:    pushq %rbx
-; SSE4-NEXT:    .cfi_def_cfa_offset 56
 ; SSE4-NEXT:    pushq %rax
-; SSE4-NEXT:    .cfi_def_cfa_offset 64
-; SSE4-NEXT:    .cfi_offset %rbx, -56
-; SSE4-NEXT:    .cfi_offset %r12, -48
-; SSE4-NEXT:    .cfi_offset %r13, -40
-; SSE4-NEXT:    .cfi_offset %r14, -32
-; SSE4-NEXT:    .cfi_offset %r15, -24
-; SSE4-NEXT:    .cfi_offset %rbp, -16
 ; SSE4-NEXT:    movq %xmm3, %rax
 ; SSE4-NEXT:    movq %rax, {{[-0-9]+}}(%r{{[sb]}}p) # 8-byte Spill
 ; SSE4-NEXT:    movq %rax, %rcx
@@ -2972,43 +2849,23 @@ define <8 x i64> @test_ext_v8i64(<8 x i64> %a0, <8 x i64> %a1) {
 ; SSE4-NEXT:    punpcklqdq {{.*#+}} xmm2 = xmm2[0],xmm6[0]
 ; SSE4-NEXT:    punpcklqdq {{.*#+}} xmm3 = xmm3[0],xmm7[0]
 ; SSE4-NEXT:    addq $8, %rsp
-; SSE4-NEXT:    .cfi_def_cfa_offset 56
 ; SSE4-NEXT:    popq %rbx
-; SSE4-NEXT:    .cfi_def_cfa_offset 48
 ; SSE4-NEXT:    popq %r12
-; SSE4-NEXT:    .cfi_def_cfa_offset 40
 ; SSE4-NEXT:    popq %r13
-; SSE4-NEXT:    .cfi_def_cfa_offset 32
 ; SSE4-NEXT:    popq %r14
-; SSE4-NEXT:    .cfi_def_cfa_offset 24
 ; SSE4-NEXT:    popq %r15
-; SSE4-NEXT:    .cfi_def_cfa_offset 16
 ; SSE4-NEXT:    popq %rbp
-; SSE4-NEXT:    .cfi_def_cfa_offset 8
 ; SSE4-NEXT:    retq
 ;
 ; AVX1-LABEL: test_ext_v8i64:
 ; AVX1:       # %bb.0:
 ; AVX1-NEXT:    pushq %rbp
-; AVX1-NEXT:    .cfi_def_cfa_offset 16
 ; AVX1-NEXT:    pushq %r15
-; AVX1-NEXT:    .cfi_def_cfa_offset 24
 ; AVX1-NEXT:    pushq %r14
-; AVX1-NEXT:    .cfi_def_cfa_offset 32
 ; AVX1-NEXT:    pushq %r13
-; AVX1-NEXT:    .cfi_def_cfa_offset 40
 ; AVX1-NEXT:    pushq %r12
-; AVX1-NEXT:    .cfi_def_cfa_offset 48
 ; AVX1-NEXT:    pushq %rbx
-; AVX1-NEXT:    .cfi_def_cfa_offset 56
 ; AVX1-NEXT:    pushq %rax
-; AVX1-NEXT:    .cfi_def_cfa_offset 64
-; AVX1-NEXT:    .cfi_offset %rbx, -56
-; AVX1-NEXT:    .cfi_offset %r12, -48
-; AVX1-NEXT:    .cfi_offset %r13, -40
-; AVX1-NEXT:    .cfi_offset %r14, -32
-; AVX1-NEXT:    .cfi_offset %r15, -24
-; AVX1-NEXT:    .cfi_offset %rbp, -16
 ; AVX1-NEXT:    vmovq %xmm1, %rax
 ; AVX1-NEXT:    movq %rax, {{[-0-9]+}}(%r{{[sb]}}p) # 8-byte Spill
 ; AVX1-NEXT:    movq %rax, %rcx
@@ -3118,43 +2975,23 @@ define <8 x i64> @test_ext_v8i64(<8 x i64> %a0, <8 x i64> %a1) {
 ; AVX1-NEXT:    vpunpcklqdq {{.*#+}} xmm2 = xmm7[0],xmm6[0]
 ; AVX1-NEXT:    vinsertf128 $1, %xmm1, %ymm2, %ymm1
 ; AVX1-NEXT:    addq $8, %rsp
-; AVX1-NEXT:    .cfi_def_cfa_offset 56
 ; AVX1-NEXT:    popq %rbx
-; AVX1-NEXT:    .cfi_def_cfa_offset 48
 ; AVX1-NEXT:    popq %r12
-; AVX1-NEXT:    .cfi_def_cfa_offset 40
 ; AVX1-NEXT:    popq %r13
-; AVX1-NEXT:    .cfi_def_cfa_offset 32
 ; AVX1-NEXT:    popq %r14
-; AVX1-NEXT:    .cfi_def_cfa_offset 24
 ; AVX1-NEXT:    popq %r15
-; AVX1-NEXT:    .cfi_def_cfa_offset 16
 ; AVX1-NEXT:    popq %rbp
-; AVX1-NEXT:    .cfi_def_cfa_offset 8
 ; AVX1-NEXT:    retq
 ;
 ; AVX2-LABEL: test_ext_v8i64:
 ; AVX2:       # %bb.0:
 ; AVX2-NEXT:    pushq %rbp
-; AVX2-NEXT:    .cfi_def_cfa_offset 16
 ; AVX2-NEXT:    pushq %r15
-; AVX2-NEXT:    .cfi_def_cfa_offset 24
 ; AVX2-NEXT:    pushq %r14
-; AVX2-NEXT:    .cfi_def_cfa_offset 32
 ; AVX2-NEXT:    pushq %r13
-; AVX2-NEXT:    .cfi_def_cfa_offset 40
 ; AVX2-NEXT:    pushq %r12
-; AVX2-NEXT:    .cfi_def_cfa_offset 48
 ; AVX2-NEXT:    pushq %rbx
-; AVX2-NEXT:    .cfi_def_cfa_offset 56
 ; AVX2-NEXT:    pushq %rax
-; AVX2-NEXT:    .cfi_def_cfa_offset 64
-; AVX2-NEXT:    .cfi_offset %rbx, -56
-; AVX2-NEXT:    .cfi_offset %r12, -48
-; AVX2-NEXT:    .cfi_offset %r13, -40
-; AVX2-NEXT:    .cfi_offset %r14, -32
-; AVX2-NEXT:    .cfi_offset %r15, -24
-; AVX2-NEXT:    .cfi_offset %rbp, -16
 ; AVX2-NEXT:    vmovq %xmm1, %rax
 ; AVX2-NEXT:    movq %rax, {{[-0-9]+}}(%r{{[sb]}}p) # 8-byte Spill
 ; AVX2-NEXT:    movq %rax, %rcx
@@ -3264,43 +3101,23 @@ define <8 x i64> @test_ext_v8i64(<8 x i64> %a0, <8 x i64> %a1) {
 ; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm2 = xmm7[0],xmm6[0]
 ; AVX2-NEXT:    vinserti128 $1, %xmm1, %ymm2, %ymm1
 ; AVX2-NEXT:    addq $8, %rsp
-; AVX2-NEXT:    .cfi_def_cfa_offset 56
 ; AVX2-NEXT:    popq %rbx
-; AVX2-NEXT:    .cfi_def_cfa_offset 48
 ; AVX2-NEXT:    popq %r12
-; AVX2-NEXT:    .cfi_def_cfa_offset 40
 ; AVX2-NEXT:    popq %r13
-; AVX2-NEXT:    .cfi_def_cfa_offset 32
 ; AVX2-NEXT:    popq %r14
-; AVX2-NEXT:    .cfi_def_cfa_offset 24
 ; AVX2-NEXT:    popq %r15
-; AVX2-NEXT:    .cfi_def_cfa_offset 16
 ; AVX2-NEXT:    popq %rbp
-; AVX2-NEXT:    .cfi_def_cfa_offset 8
 ; AVX2-NEXT:    retq
 ;
 ; AVX512-LABEL: test_ext_v8i64:
 ; AVX512:       # %bb.0:
 ; AVX512-NEXT:    pushq %rbp
-; AVX512-NEXT:    .cfi_def_cfa_offset 16
 ; AVX512-NEXT:    pushq %r15
-; AVX512-NEXT:    .cfi_def_cfa_offset 24
 ; AVX512-NEXT:    pushq %r14
-; AVX512-NEXT:    .cfi_def_cfa_offset 32
 ; AVX512-NEXT:    pushq %r13
-; AVX512-NEXT:    .cfi_def_cfa_offset 40
 ; AVX512-NEXT:    pushq %r12
-; AVX512-NEXT:    .cfi_def_cfa_offset 48
 ; AVX512-NEXT:    pushq %rbx
-; AVX512-NEXT:    .cfi_def_cfa_offset 56
 ; AVX512-NEXT:    pushq %rax
-; AVX512-NEXT:    .cfi_def_cfa_offset 64
-; AVX512-NEXT:    .cfi_offset %rbx, -56
-; AVX512-NEXT:    .cfi_offset %r12, -48
-; AVX512-NEXT:    .cfi_offset %r13, -40
-; AVX512-NEXT:    .cfi_offset %r14, -32
-; AVX512-NEXT:    .cfi_offset %r15, -24
-; AVX512-NEXT:    .cfi_offset %rbp, -16
 ; AVX512-NEXT:    vmovq %xmm0, %rax
 ; AVX512-NEXT:    movq %rax, {{[-0-9]+}}(%r{{[sb]}}p) # 8-byte Spill
 ; AVX512-NEXT:    movq %rax, %rcx
@@ -3413,19 +3230,12 @@ define <8 x i64> @test_ext_v8i64(<8 x i64> %a0, <8 x i64> %a1) {
 ; AVX512-NEXT:    vinserti128 $1, %xmm1, %ymm2, %ymm1
 ; AVX512-NEXT:    vinserti64x4 $1, %ymm0, %zmm1, %zmm0
 ; AVX512-NEXT:    addq $8, %rsp
-; AVX512-NEXT:    .cfi_def_cfa_offset 56
 ; AVX512-NEXT:    popq %rbx
-; AVX512-NEXT:    .cfi_def_cfa_offset 48
 ; AVX512-NEXT:    popq %r12
-; AVX512-NEXT:    .cfi_def_cfa_offset 40
 ; AVX512-NEXT:    popq %r13
-; AVX512-NEXT:    .cfi_def_cfa_offset 32
 ; AVX512-NEXT:    popq %r14
-; AVX512-NEXT:    .cfi_def_cfa_offset 24
 ; AVX512-NEXT:    popq %r15
-; AVX512-NEXT:    .cfi_def_cfa_offset 16
 ; AVX512-NEXT:    popq %rbp
-; AVX512-NEXT:    .cfi_def_cfa_offset 8
 ; AVX512-NEXT:    retq
   %x0 = sext <8 x i64> %a0 to <8 x i128>
   %x1 = sext <8 x i64> %a1 to <8 x i128>
diff --git a/llvm/test/CodeGen/X86/avgflooru.ll b/llvm/test/CodeGen/X86/avgflooru.ll
index e07c1f55991e..000457c5ab1e 100644
--- a/llvm/test/CodeGen/X86/avgflooru.ll
+++ b/llvm/test/CodeGen/X86/avgflooru.ll
@@ -9,7 +9,7 @@
 ; 128-bit vectors
 ;
 
-define <16 x i8> @test_fixed_v16i8(<16 x i8> %a0, <16 x i8> %a1) {
+define <16 x i8> @test_fixed_v16i8(<16 x i8> %a0, <16 x i8> %a1) nounwind {
 ; SSE-LABEL: test_fixed_v16i8:
 ; SSE:       # %bb.0:
 ; SSE-NEXT:    movdqa %xmm0, %xmm2
@@ -53,7 +53,7 @@ define <16 x i8> @test_fixed_v16i8(<16 x i8> %a0, <16 x i8> %a1) {
   ret <16 x i8> %res
 }
 
-define <16 x i8> @test_ext_v16i8(<16 x i8> %a0, <16 x i8> %a1) {
+define <16 x i8> @test_ext_v16i8(<16 x i8> %a0, <16 x i8> %a1) nounwind {
 ; SSE2-LABEL: test_ext_v16i8:
 ; SSE2:       # %bb.0:
 ; SSE2-NEXT:    pxor %xmm2, %xmm2
@@ -127,7 +127,7 @@ define <16 x i8> @test_ext_v16i8(<16 x i8> %a0, <16 x i8> %a1) {
   ret <16 x i8> %res
 }
 
-define <8 x i16> @test_fixed_v8i16(<8 x i16> %a0, <8 x i16> %a1) {
+define <8 x i16> @test_fixed_v8i16(<8 x i16> %a0, <8 x i16> %a1) nounwind {
 ; SSE-LABEL: test_fixed_v8i16:
 ; SSE:       # %bb.0:
 ; SSE-NEXT:    movdqa %xmm0, %xmm2
@@ -151,7 +151,7 @@ define <8 x i16> @test_fixed_v8i16(<8 x i16> %a0, <8 x i16> %a1) {
   ret <8 x i16> %res
 }
 
-define <8 x i16> @test_ext_v8i16(<8 x i16> %a0, <8 x i16> %a1) {
+define <8 x i16> @test_ext_v8i16(<8 x i16> %a0, <8 x i16> %a1) nounwind {
 ; SSE2-LABEL: test_ext_v8i16:
 ; SSE2:       # %bb.0:
 ; SSE2-NEXT:    pxor %xmm2, %xmm2
@@ -227,7 +227,7 @@ define <8 x i16> @test_ext_v8i16(<8 x i16> %a0, <8 x i16> %a1) {
   ret <8 x i16> %res
 }
 
-define <4 x i32> @test_fixed_v4i32(<4 x i32> %a0, <4 x i32> %a1) {
+define <4 x i32> @test_fixed_v4i32(<4 x i32> %a0, <4 x i32> %a1) nounwind {
 ; SSE-LABEL: test_fixed_v4i32:
 ; SSE:       # %bb.0:
 ; SSE-NEXT:    movdqa %xmm0, %xmm2
@@ -251,7 +251,7 @@ define <4 x i32> @test_fixed_v4i32(<4 x i32> %a0, <4 x i32> %a1) {
   ret <4 x i32> %res
 }
 
-define <4 x i32> @test_ext_v4i32(<4 x i32> %a0, <4 x i32> %a1) {
+define <4 x i32> @test_ext_v4i32(<4 x i32> %a0, <4 x i32> %a1) nounwind {
 ; SSE2-LABEL: test_ext_v4i32:
 ; SSE2:       # %bb.0:
 ; SSE2-NEXT:    pxor %xmm2, %xmm2
@@ -325,7 +325,7 @@ define <4 x i32> @test_ext_v4i32(<4 x i32> %a0, <4 x i32> %a1) {
   ret <4 x i32> %res
 }
 
-define <2 x i64> @test_fixed_v2i64(<2 x i64> %a0, <2 x i64> %a1) {
+define <2 x i64> @test_fixed_v2i64(<2 x i64> %a0, <2 x i64> %a1) nounwind {
 ; SSE-LABEL: test_fixed_v2i64:
 ; SSE:       # %bb.0:
 ; SSE-NEXT:    movdqa %xmm0, %xmm2
@@ -349,7 +349,7 @@ define <2 x i64> @test_fixed_v2i64(<2 x i64> %a0, <2 x i64> %a1) {
   ret <2 x i64> %res
 }
 
-define <2 x i64> @test_ext_v2i64(<2 x i64> %a0, <2 x i64> %a1) {
+define <2 x i64> @test_ext_v2i64(<2 x i64> %a0, <2 x i64> %a1) nounwind {
 ; SSE2-LABEL: test_ext_v2i64:
 ; SSE2:       # %bb.0:
 ; SSE2-NEXT:    pshufd {{.*#+}} xmm2 = xmm0[2,3,2,3]
@@ -458,7 +458,7 @@ define <2 x i64> @test_ext_v2i64(<2 x i64> %a0, <2 x i64> %a1) {
 ; 256-bit vectors
 ;
 
-define <32 x i8> @test_fixed_v32i8(<32 x i8> %a0, <32 x i8> %a1) {
+define <32 x i8> @test_fixed_v32i8(<32 x i8> %a0, <32 x i8> %a1) nounwind {
 ; SSE-LABEL: test_fixed_v32i8:
 ; SSE:       # %bb.0:
 ; SSE-NEXT:    movdqa %xmm1, %xmm4
@@ -516,7 +516,7 @@ define <32 x i8> @test_fixed_v32i8(<32 x i8> %a0, <32 x i8> %a1) {
   ret <32 x i8> %res
 }
 
-define <32 x i8> @test_ext_v32i8(<32 x i8> %a0, <32 x i8> %a1) {
+define <32 x i8> @test_ext_v32i8(<32 x i8> %a0, <32 x i8> %a1) nounwind {
 ; SSE2-LABEL: test_ext_v32i8:
 ; SSE2:       # %bb.0:
 ; SSE2-NEXT:    pxor %xmm4, %xmm4
@@ -627,7 +627,7 @@ define <32 x i8> @test_ext_v32i8(<32 x i8> %a0, <32 x i8> %a1) {
   ret <32 x i8> %res
 }
 
-define <16 x i16> @test_fixed_v16i16(<16 x i16> %a0, <16 x i16> %a1) {
+define <16 x i16> @test_fixed_v16i16(<16 x i16> %a0, <16 x i16> %a1) nounwind {
 ; SSE-LABEL: test_fixed_v16i16:
 ; SSE:       # %bb.0:
 ; SSE-NEXT:    movdqa %xmm1, %xmm4
@@ -677,7 +677,7 @@ define <16 x i16> @test_fixed_v16i16(<16 x i16> %a0, <16 x i16> %a1) {
   ret <16 x i16> %res
 }
 
-define <16 x i16> @test_ext_v16i16(<16 x i16> %a0, <16 x i16> %a1) {
+define <16 x i16> @test_ext_v16i16(<16 x i16> %a0, <16 x i16> %a1) nounwind {
 ; SSE2-LABEL: test_ext_v16i16:
 ; SSE2:       # %bb.0:
 ; SSE2-NEXT:    pxor %xmm4, %xmm4
@@ -792,7 +792,7 @@ define <16 x i16> @test_ext_v16i16(<16 x i16> %a0, <16 x i16> %a1) {
   ret <16 x i16> %res
 }
 
-define <8 x i32> @test_fixed_v8i32(<8 x i32> %a0, <8 x i32> %a1) {
+define <8 x i32> @test_fixed_v8i32(<8 x i32> %a0, <8 x i32> %a1) nounwind {
 ; SSE-LABEL: test_fixed_v8i32:
 ; SSE:       # %bb.0:
 ; SSE-NEXT:    movdqa %xmm1, %xmm4
@@ -842,7 +842,7 @@ define <8 x i32> @test_fixed_v8i32(<8 x i32> %a0, <8 x i32> %a1) {
   ret <8 x i32> %res
 }
 
-define <8 x i32> @test_ext_v8i32(<8 x i32> %a0, <8 x i32> %a1) {
+define <8 x i32> @test_ext_v8i32(<8 x i32> %a0, <8 x i32> %a1) nounwind {
 ; SSE2-LABEL: test_ext_v8i32:
 ; SSE2:       # %bb.0:
 ; SSE2-NEXT:    pxor %xmm4, %xmm4
@@ -954,7 +954,7 @@ define <8 x i32> @test_ext_v8i32(<8 x i32> %a0, <8 x i32> %a1) {
   ret <8 x i32> %res
 }
 
-define <4 x i64> @test_fixed_v4i64(<4 x i64> %a0, <4 x i64> %a1) {
+define <4 x i64> @test_fixed_v4i64(<4 x i64> %a0, <4 x i64> %a1) nounwind {
 ; SSE-LABEL: test_fixed_v4i64:
 ; SSE:       # %bb.0:
 ; SSE-NEXT:    movdqa %xmm1, %xmm4
@@ -1004,7 +1004,7 @@ define <4 x i64> @test_fixed_v4i64(<4 x i64> %a0, <4 x i64> %a1) {
   ret <4 x i64> %res
 }
 
-define <4 x i64> @test_ext_v4i64(<4 x i64> %a0, <4 x i64> %a1) {
+define <4 x i64> @test_ext_v4i64(<4 x i64> %a0, <4 x i64> %a1) nounwind {
 ; SSE2-LABEL: test_ext_v4i64:
 ; SSE2:       # %bb.0:
 ; SSE2-NEXT:    pshufd {{.*#+}} xmm4 = xmm1[2,3,2,3]
@@ -1199,7 +1199,7 @@ define <4 x i64> @test_ext_v4i64(<4 x i64> %a0, <4 x i64> %a1) {
 ; 512-bit vectors
 ;
 
-define <64 x i8> @test_fixed_v64i8(<64 x i8> %a0, <64 x i8> %a1) {
+define <64 x i8> @test_fixed_v64i8(<64 x i8> %a0, <64 x i8> %a1) nounwind {
 ; SSE-LABEL: test_fixed_v64i8:
 ; SSE:       # %bb.0:
 ; SSE-NEXT:    movdqa %xmm3, %xmm9
@@ -1286,7 +1286,7 @@ define <64 x i8> @test_fixed_v64i8(<64 x i8> %a0, <64 x i8> %a1) {
   ret <64 x i8> %res
 }
 
-define <64 x i8> @test_ext_v64i8(<64 x i8> %a0, <64 x i8> %a1) {
+define <64 x i8> @test_ext_v64i8(<64 x i8> %a0, <64 x i8> %a1) nounwind {
 ; SSE2-LABEL: test_ext_v64i8:
 ; SSE2:       # %bb.0:
 ; SSE2-NEXT:    pxor %xmm8, %xmm8
@@ -1481,7 +1481,7 @@ define <64 x i8> @test_ext_v64i8(<64 x i8> %a0, <64 x i8> %a1) {
   ret <64 x i8> %res
 }
 
-define <32 x i16> @test_fixed_v32i16(<32 x i16> %a0, <32 x i16> %a1) {
+define <32 x i16> @test_fixed_v32i16(<32 x i16> %a0, <32 x i16> %a1) nounwind {
 ; SSE-LABEL: test_fixed_v32i16:
 ; SSE:       # %bb.0:
 ; SSE-NEXT:    movdqa %xmm3, %xmm8
@@ -1554,7 +1554,7 @@ define <32 x i16> @test_fixed_v32i16(<32 x i16> %a0, <32 x i16> %a1) {
   ret <32 x i16> %res
 }
 
-define <32 x i16> @test_ext_v32i16(<32 x i16> %a0, <32 x i16> %a1) {
+define <32 x i16> @test_ext_v32i16(<32 x i16> %a0, <32 x i16> %a1) nounwind {
 ; SSE2-LABEL: test_ext_v32i16:
 ; SSE2:       # %bb.0:
 ; SSE2-NEXT:    pxor %xmm8, %xmm8
@@ -1757,7 +1757,7 @@ define <32 x i16> @test_ext_v32i16(<32 x i16> %a0, <32 x i16> %a1) {
   ret <32 x i16> %res
 }
 
-define <16 x i32> @test_fixed_v16i32(<16 x i32> %a0, <16 x i32> %a1) {
+define <16 x i32> @test_fixed_v16i32(<16 x i32> %a0, <16 x i32> %a1) nounwind {
 ; SSE-LABEL: test_fixed_v16i32:
 ; SSE:       # %bb.0:
 ; SSE-NEXT:    movdqa %xmm3, %xmm8
@@ -1830,7 +1830,7 @@ define <16 x i32> @test_fixed_v16i32(<16 x i32> %a0, <16 x i32> %a1) {
   ret <16 x i32> %res
 }
 
-define <16 x i32> @test_ext_v16i32(<16 x i32> %a0, <16 x i32> %a1) {
+define <16 x i32> @test_ext_v16i32(<16 x i32> %a0, <16 x i32> %a1) nounwind {
 ; SSE2-LABEL: test_ext_v16i32:
 ; SSE2:       # %bb.0:
 ; SSE2-NEXT:    pxor %xmm8, %xmm8
@@ -2027,7 +2027,7 @@ define <16 x i32> @test_ext_v16i32(<16 x i32> %a0, <16 x i32> %a1) {
   ret <16 x i32> %res
 }
 
-define <8 x i64> @test_fixed_v8i64(<8 x i64> %a0, <8 x i64> %a1) {
+define <8 x i64> @test_fixed_v8i64(<8 x i64> %a0, <8 x i64> %a1) nounwind {
 ; SSE-LABEL: test_fixed_v8i64:
 ; SSE:       # %bb.0:
 ; SSE-NEXT:    movdqa %xmm3, %xmm8
@@ -2100,27 +2100,15 @@ define <8 x i64> @test_fixed_v8i64(<8 x i64> %a0, <8 x i64> %a1) {
   ret <8 x i64> %res
 }
 
-define <8 x i64> @test_ext_v8i64(<8 x i64> %a0, <8 x i64> %a1) {
+define <8 x i64> @test_ext_v8i64(<8 x i64> %a0, <8 x i64> %a1) nounwind {
 ; SSE2-LABEL: test_ext_v8i64:
 ; SSE2:       # %bb.0:
 ; SSE2-NEXT:    pushq %rbp
-; SSE2-NEXT:    .cfi_def_cfa_offset 16
 ; SSE2-NEXT:    pushq %r15
-; SSE2-NEXT:    .cfi_def_cfa_offset 24
 ; SSE2-NEXT:    pushq %r14
-; SSE2-NEXT:    .cfi_def_cfa_offset 32
 ; SSE2-NEXT:    pushq %r13
-; SSE2-NEXT:    .cfi_def_cfa_offset 40
 ; SSE2-NEXT:    pushq %r12
-; SSE2-NEXT:    .cfi_def_cfa_offset 48
 ; SSE2-NEXT:    pushq %rbx
-; SSE2-NEXT:    .cfi_def_cfa_offset 56
-; SSE2-NEXT:    .cfi_offset %rbx, -56
-; SSE2-NEXT:    .cfi_offset %r12, -48
-; SSE2-NEXT:    .cfi_offset %r13, -40
-; SSE2-NEXT:    .cfi_offset %r14, -32
-; SSE2-NEXT:    .cfi_offset %r15, -24
-; SSE2-NEXT:    .cfi_offset %rbp, -16
 ; SSE2-NEXT:    pshufd {{.*#+}} xmm8 = xmm3[2,3,2,3]
 ; SSE2-NEXT:    movq %xmm3, %rbx
 ; SSE2-NEXT:    pshufd {{.*#+}} xmm3 = xmm2[2,3,2,3]
@@ -2194,39 +2182,21 @@ define <8 x i64> @test_ext_v8i64(<8 x i64> %a0, <8 x i64> %a1) {
 ; SSE2-NEXT:    punpcklqdq {{.*#+}} xmm2 = xmm2[0],xmm6[0]
 ; SSE2-NEXT:    punpcklqdq {{.*#+}} xmm3 = xmm3[0],xmm7[0]
 ; SSE2-NEXT:    popq %rbx
-; SSE2-NEXT:    .cfi_def_cfa_offset 48
 ; SSE2-NEXT:    popq %r12
-; SSE2-NEXT:    .cfi_def_cfa_offset 40
 ; SSE2-NEXT:    popq %r13
-; SSE2-NEXT:    .cfi_def_cfa_offset 32
 ; SSE2-NEXT:    popq %r14
-; SSE2-NEXT:    .cfi_def_cfa_offset 24
 ; SSE2-NEXT:    popq %r15
-; SSE2-NEXT:    .cfi_def_cfa_offset 16
 ; SSE2-NEXT:    popq %rbp
-; SSE2-NEXT:    .cfi_def_cfa_offset 8
 ; SSE2-NEXT:    retq
 ;
 ; SSE4-LABEL: test_ext_v8i64:
 ; SSE4:       # %bb.0:
 ; SSE4-NEXT:    pushq %rbp
-; SSE4-NEXT:    .cfi_def_cfa_offset 16
 ; SSE4-NEXT:    pushq %r15
-; SSE4-NEXT:    .cfi_def_cfa_offset 24
 ; SSE4-NEXT:    pushq %r14
-; SSE4-NEXT:    .cfi_def_cfa_offset 32
 ; SSE4-NEXT:    pushq %r13
-; SSE4-NEXT:    .cfi_def_cfa_offset 40
 ; SSE4-NEXT:    pushq %r12
-; SSE4-NEXT:    .cfi_def_cfa_offset 48
 ; SSE4-NEXT:    pushq %rbx
-; SSE4-NEXT:    .cfi_def_cfa_offset 56
-; SSE4-NEXT:    .cfi_offset %rbx, -56
-; SSE4-NEXT:    .cfi_offset %r12, -48
-; SSE4-NEXT:    .cfi_offset %r13, -40
-; SSE4-NEXT:    .cfi_offset %r14, -32
-; SSE4-NEXT:    .cfi_offset %r15, -24
-; SSE4-NEXT:    .cfi_offset %rbp, -16
 ; SSE4-NEXT:    pextrq $1, %xmm3, %r14
 ; SSE4-NEXT:    movq %xmm2, %r13
 ; SSE4-NEXT:    pextrq $1, %xmm2, %rbp
@@ -2292,39 +2262,21 @@ define <8 x i64> @test_ext_v8i64(<8 x i64> %a0, <8 x i64> %a1) {
 ; SSE4-NEXT:    punpcklqdq {{.*#+}} xmm2 = xmm2[0],xmm6[0]
 ; SSE4-NEXT:    punpcklqdq {{.*#+}} xmm3 = xmm3[0],xmm7[0]
 ; SSE4-NEXT:    popq %rbx
-; SSE4-NEXT:    .cfi_def_cfa_offset 48
 ; SSE4-NEXT:    popq %r12
-; SSE4-NEXT:    .cfi_def_cfa_offset 40
 ; SSE4-NEXT:    popq %r13
-; SSE4-NEXT:    .cfi_def_cfa_offset 32
 ; SSE4-NEXT:    popq %r14
-; SSE4-NEXT:    .cfi_def_cfa_offset 24
 ; SSE4-NEXT:    popq %r15
-; SSE4-NEXT:    .cfi_def_cfa_offset 16
 ; SSE4-NEXT:    popq %rbp
-; SSE4-NEXT:    .cfi_def_cfa_offset 8
 ; SSE4-NEXT:    retq
 ;
 ; AVX1-LABEL: test_ext_v8i64:
 ; AVX1:       # %bb.0:
 ; AVX1-NEXT:    pushq %rbp
-; AVX1-NEXT:    .cfi_def_cfa_offset 16
 ; AVX1-NEXT:    pushq %r15
-; AVX1-NEXT:    .cfi_def_cfa_offset 24
 ; AVX1-NEXT:    pushq %r14
-; AVX1-NEXT:    .cfi_def_cfa_offset 32
 ; AVX1-NEXT:    pushq %r13
-; AVX1-NEXT:    .cfi_def_cfa_offset 40
 ; AVX1-NEXT:    pushq %r12
-; AVX1-NEXT:    .cfi_def_cfa_offset 48
 ; AVX1-NEXT:    pushq %rbx
-; AVX1-NEXT:    .cfi_def_cfa_offset 56
-; AVX1-NEXT:    .cfi_offset %rbx, -56
-; AVX1-NEXT:    .cfi_offset %r12, -48
-; AVX1-NEXT:    .cfi_offset %r13, -40
-; AVX1-NEXT:    .cfi_offset %r14, -32
-; AVX1-NEXT:    .cfi_offset %r15, -24
-; AVX1-NEXT:    .cfi_offset %rbp, -16
 ; AVX1-NEXT:    vpextrq $1, %xmm1, %rbx
 ; AVX1-NEXT:    vextractf128 $1, %ymm1, %xmm4
 ; AVX1-NEXT:    vmovq %xmm4, %r15
@@ -2396,39 +2348,21 @@ define <8 x i64> @test_ext_v8i64(<8 x i64> %a0, <8 x i64> %a1) {
 ; AVX1-NEXT:    vpunpcklqdq {{.*#+}} xmm2 = xmm7[0],xmm6[0]
 ; AVX1-NEXT:    vinsertf128 $1, %xmm1, %ymm2, %ymm1
 ; AVX1-NEXT:    popq %rbx
-; AVX1-NEXT:    .cfi_def_cfa_offset 48
 ; AVX1-NEXT:    popq %r12
-; AVX1-NEXT:    .cfi_def_cfa_offset 40
 ; AVX1-NEXT:    popq %r13
-; AVX1-NEXT:    .cfi_def_cfa_offset 32
 ; AVX1-NEXT:    popq %r14
-; AVX1-NEXT:    .cfi_def_cfa_offset 24
 ; AVX1-NEXT:    popq %r15
-; AVX1-NEXT:    .cfi_def_cfa_offset 16
 ; AVX1-NEXT:    popq %rbp
-; AVX1-NEXT:    .cfi_def_cfa_offset 8
 ; AVX1-NEXT:    retq
 ;
 ; AVX2-LABEL: test_ext_v8i64:
 ; AVX2:       # %bb.0:
 ; AVX2-NEXT:    pushq %rbp
-; AVX2-NEXT:    .cfi_def_cfa_offset 16
 ; AVX2-NEXT:    pushq %r15
-; AVX2-NEXT:    .cfi_def_cfa_offset 24
 ; AVX2-NEXT:    pushq %r14
-; AVX2-NEXT:    .cfi_def_cfa_offset 32
 ; AVX2-NEXT:    pushq %r13
-; AVX2-NEXT:    .cfi_def_cfa_offset 40
 ; AVX2-NEXT:    pushq %r12
-; AVX2-NEXT:    .cfi_def_cfa_offset 48
 ; AVX2-NEXT:    pushq %rbx
-; AVX2-NEXT:    .cfi_def_cfa_offset 56
-; AVX2-NEXT:    .cfi_offset %rbx, -56
-; AVX2-NEXT:    .cfi_offset %r12, -48
-; AVX2-NEXT:    .cfi_offset %r13, -40
-; AVX2-NEXT:    .cfi_offset %r14, -32
-; AVX2-NEXT:    .cfi_offset %r15, -24
-; AVX2-NEXT:    .cfi_offset %rbp, -16
 ; AVX2-NEXT:    vpextrq $1, %xmm1, %rbx
 ; AVX2-NEXT:    vextracti128 $1, %ymm1, %xmm4
 ; AVX2-NEXT:    vmovq %xmm4, %r15
@@ -2500,39 +2434,21 @@ define <8 x i64> @test_ext_v8i64(<8 x i64> %a0, <8 x i64> %a1) {
 ; AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm2 = xmm7[0],xmm6[0]
 ; AVX2-NEXT:    vinserti128 $1, %xmm1, %ymm2, %ymm1
 ; AVX2-NEXT:    popq %rbx
-; AVX2-NEXT:    .cfi_def_cfa_offset 48
 ; AVX2-NEXT:    popq %r12
-; AVX2-NEXT:    .cfi_def_cfa_offset 40
 ; AVX2-NEXT:    popq %r13
-; AVX2-NEXT:    .cfi_def_cfa_offset 32
 ; AVX2-NEXT:    popq %r14
-; AVX2-NEXT:    .cfi_def_cfa_offset 24
 ; AVX2-NEXT:    popq %r15
-; AVX2-NEXT:    .cfi_def_cfa_offset 16
 ; AVX2-NEXT:    popq %rbp
-; AVX2-NEXT:    .cfi_def_cfa_offset 8
 ; AVX2-NEXT:    retq
 ;
 ; AVX512-LABEL: test_ext_v8i64:
 ; AVX512:       # %bb.0:
 ; AVX512-NEXT:    pushq %rbp
-; AVX512-NEXT:    .cfi_def_cfa_offset 16
 ; AVX512-NEXT:    pushq %r15
-; AVX512-NEXT:    .cfi_def_cfa_offset 24
 ; AVX512-NEXT:    pushq %r14
-; AVX512-NEXT:    .cfi_def_cfa_offset 32
 ; AVX512-NEXT:    pushq %r13
-; AVX512-NEXT:    .cfi_def_cfa_offset 40
 ; AVX512-NEXT:    pushq %r12
-; AVX512-NEXT:    .cfi_def_cfa_offset 48
 ; AVX512-NEXT:    pushq %rbx
-; AVX512-NEXT:    .cfi_def_cfa_offset 56
-; AVX512-NEXT:    .cfi_offset %rbx, -56
-; AVX512-NEXT:    .cfi_offset %r12, -48
-; AVX512-NEXT:    .cfi_offset %r13, -40
-; AVX512-NEXT:    .cfi_offset %r14, -32
-; AVX512-NEXT:    .cfi_offset %r15, -24
-; AVX512-NEXT:    .cfi_offset %rbp, -16
 ; AVX512-NEXT:    vpextrq $1, %xmm0, %r10
 ; AVX512-NEXT:    vextracti128 $1, %ymm0, %xmm2
 ; AVX512-NEXT:    vpextrq $1, %xmm2, %r13
@@ -2607,17 +2523,11 @@ define <8 x i64> @test_ext_v8i64(<8 x i64> %a0, <8 x i64> %a1) {
 ; AVX512-NEXT:    vinserti128 $1, %xmm1, %ymm2, %ymm1
 ; AVX512-NEXT:    vinserti64x4 $1, %ymm0, %zmm1, %zmm0
 ; AVX512-NEXT:    popq %rbx
-; AVX512-NEXT:    .cfi_def_cfa_offset 48
 ; AVX512-NEXT:    popq %r12
-; AVX512-NEXT:    .cfi_def_cfa_offset 40
 ; AVX512-NEXT:    popq %r13
-; AVX512-NEXT:    .cfi_def_cfa_offset 32
 ; AVX512-NEXT:    popq %r14
-; AVX512-NEXT:    .cfi_def_cfa_offset 24
 ; AVX512-NEXT:    popq %r15
-; AVX512-NEXT:    .cfi_def_cfa_offset 16
 ; AVX512-NEXT:    popq %rbp
-; AVX512-NEXT:    .cfi_def_cfa_offset 8
 ; AVX512-NEXT:    retq
   %x0 = zext <8 x i64> %a0 to <8 x i128>
   %x1 = zext <8 x i64> %a1 to <8 x i128>
diff --git a/llvm/test/CodeGen/X86/avx512-broadcast-arith.ll b/llvm/test/CodeGen/X86/avx512-broadcast-arith.ll
index 13d1265a249d..7e48b3719cf0 100644
--- a/llvm/test/CodeGen/X86/avx512-broadcast-arith.ll
+++ b/llvm/test/CodeGen/X86/avx512-broadcast-arith.ll
@@ -30,13 +30,13 @@ define <64 x i8> @add_v64i8_broadcasts(<64 x i8> %a0, i64 %a1, i8 %a2) {
 ; AVX512F-NEXT:    vinserti128 $1, %xmm4, %ymm3, %ymm3
 ; AVX512F-NEXT:    vinserti64x4 $1, %ymm2, %zmm3, %zmm2
 ; AVX512F-NEXT:    vpternlogq $216, %zmm2, %zmm1, %zmm0
-; AVX512F-NEXT:    vpaddb %ymm1, %ymm0, %ymm3
-; AVX512F-NEXT:    vextracti64x4 $1, %zmm0, %ymm0
+; AVX512F-NEXT:    vextracti64x4 $1, %zmm0, %ymm3
+; AVX512F-NEXT:    vpaddb %ymm1, %ymm3, %ymm3
 ; AVX512F-NEXT:    vpaddb %ymm1, %ymm0, %ymm0
-; AVX512F-NEXT:    vinserti64x4 $1, %ymm0, %zmm3, %zmm4
+; AVX512F-NEXT:    vinserti64x4 $1, %ymm3, %zmm0, %zmm4
+; AVX512F-NEXT:    vpaddb %ymm1, %ymm3, %ymm3
 ; AVX512F-NEXT:    vpaddb %ymm1, %ymm0, %ymm0
-; AVX512F-NEXT:    vpaddb %ymm1, %ymm3, %ymm1
-; AVX512F-NEXT:    vinserti64x4 $1, %ymm0, %zmm1, %zmm0
+; AVX512F-NEXT:    vinserti64x4 $1, %ymm3, %zmm0, %zmm0
 ; AVX512F-NEXT:    vpternlogq $226, %zmm4, %zmm2, %zmm0
 ; AVX512F-NEXT:    retq
 ;
diff --git a/llvm/test/CodeGen/X86/avx512bwvl-arith.ll b/llvm/test/CodeGen/X86/avx512bwvl-arith.ll
index 4988fc35b10e..33819c9e0102 100644
--- a/llvm/test/CodeGen/X86/avx512bwvl-arith.ll
+++ b/llvm/test/CodeGen/X86/avx512bwvl-arith.ll
@@ -1,6 +1,6 @@
 ; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
-; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx512bw,+avx512vl | FileCheck %s
-; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx512bw,+avx512vl,-evex512 | FileCheck %s
+; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx512bw,+avx512vl | FileCheck %s --check-prefixes=CHECK,EVEX512
+; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx512bw,+avx512vl,-evex512 | FileCheck %s --check-prefixes=CHECK,EVEX256
 
 ; 256-bit
 
@@ -236,3 +236,34 @@ define <8 x i16> @vpmullw128_test(<8 x i16> %i, <8 x i16> %j) {
   ret <8 x i16> %x
 }
 
+define i16 @PR90356(<16 x i1> %a) {
+; EVEX512-LABEL: PR90356:
+; EVEX512:       # %bb.0:
+; EVEX512-NEXT:    vpsllw $7, %xmm0, %xmm0
+; EVEX512-NEXT:    vpmovb2m %xmm0, %k1
+; EVEX512-NEXT:    vpternlogd $255, %zmm0, %zmm0, %zmm0 {%k1} {z}
+; EVEX512-NEXT:    movb $63, %al
+; EVEX512-NEXT:    kmovd %eax, %k1
+; EVEX512-NEXT:    vpexpandq %zmm0, %zmm0 {%k1} {z}
+; EVEX512-NEXT:    vptestmd %zmm0, %zmm0, %k0
+; EVEX512-NEXT:    kmovd %k0, %eax
+; EVEX512-NEXT:    # kill: def $ax killed $ax killed $eax
+; EVEX512-NEXT:    vzeroupper
+; EVEX512-NEXT:    retq
+;
+; EVEX256-LABEL: PR90356:
+; EVEX256:       # %bb.0:
+; EVEX256-NEXT:    vpsllw $7, %xmm0, %xmm0
+; EVEX256-NEXT:    vpmovb2m %xmm0, %k0
+; EVEX256-NEXT:    vpmovm2w %k0, %ymm0
+; EVEX256-NEXT:    vpxor %xmm1, %xmm1, %xmm1
+; EVEX256-NEXT:    vpblendd {{.*#+}} ymm0 = ymm0[0,1,2,3,4,5],ymm1[6,7]
+; EVEX256-NEXT:    vpmovw2m %ymm0, %k0
+; EVEX256-NEXT:    kmovd %k0, %eax
+; EVEX256-NEXT:    # kill: def $ax killed $ax killed $eax
+; EVEX256-NEXT:    vzeroupper
+; EVEX256-NEXT:    retq
+  %1 = shufflevector <16 x i1> %a, <16 x i1> zeroinitializer, <16 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 8, i32 9, i32 10, i32 11, i32 28, i32 29, i32 30, i32 31>
+  %2 = bitcast <16 x i1> %1 to i16
+  ret i16 %2
+}
diff --git a/llvm/test/CodeGen/X86/cmp.ll b/llvm/test/CodeGen/X86/cmp.ll
index 30e52f063075..402da547613c 100644
--- a/llvm/test/CodeGen/X86/cmp.ll
+++ b/llvm/test/CodeGen/X86/cmp.ll
@@ -1,6 +1,6 @@
 ; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
-; RUN: llc < %s -mtriple=x86_64-unknown-unknown -show-mc-encoding | FileCheck %s
-; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+ndd -show-mc-encoding | FileCheck --check-prefix=NDD %s
+; RUN: llc < %s -mtriple=x86_64-unknown-unknown -show-mc-encoding | FileCheck %s --check-prefixes=CHECK,NO-NDD
+; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+ndd -show-mc-encoding | FileCheck --check-prefixes=CHECK,NDD %s
 
 @d = dso_local global i8 0, align 1
 @d64 = dso_local global i64 0
@@ -17,18 +17,6 @@ define i32 @test1(i32 %X, ptr %y) nounwind {
 ; CHECK-NEXT:  .LBB0_2: # %ReturnBlock
 ; CHECK-NEXT:    xorl %eax, %eax # encoding: [0x31,0xc0]
 ; CHECK-NEXT:    retq # encoding: [0xc3]
-;
-; NDD-LABEL: test1:
-; NDD:       # %bb.0: # %entry
-; NDD-NEXT:    cmpl $0, (%rsi) # encoding: [0x83,0x3e,0x00]
-; NDD-NEXT:    je .LBB0_2 # encoding: [0x74,A]
-; NDD-NEXT:    # fixup A - offset: 1, value: .LBB0_2-1, kind: FK_PCRel_1
-; NDD-NEXT:  # %bb.1: # %cond_true
-; NDD-NEXT:    movl $1, %eax # encoding: [0xb8,0x01,0x00,0x00,0x00]
-; NDD-NEXT:    retq # encoding: [0xc3]
-; NDD-NEXT:  .LBB0_2: # %ReturnBlock
-; NDD-NEXT:    xorl %eax, %eax # encoding: [0x31,0xc0]
-; NDD-NEXT:    retq # encoding: [0xc3]
 entry:
   %tmp = load i32, ptr %y
   %tmp.upgrd.1 = icmp eq i32 %tmp, 0
@@ -54,19 +42,6 @@ define i32 @test2(i32 %X, ptr %y) nounwind {
 ; CHECK-NEXT:  .LBB1_2: # %ReturnBlock
 ; CHECK-NEXT:    xorl %eax, %eax # encoding: [0x31,0xc0]
 ; CHECK-NEXT:    retq # encoding: [0xc3]
-;
-; NDD-LABEL: test2:
-; NDD:       # %bb.0: # %entry
-; NDD-NEXT:    testl $536870911, (%rsi) # encoding: [0xf7,0x06,0xff,0xff,0xff,0x1f]
-; NDD-NEXT:    # imm = 0x1FFFFFFF
-; NDD-NEXT:    je .LBB1_2 # encoding: [0x74,A]
-; NDD-NEXT:    # fixup A - offset: 1, value: .LBB1_2-1, kind: FK_PCRel_1
-; NDD-NEXT:  # %bb.1: # %cond_true
-; NDD-NEXT:    movl $1, %eax # encoding: [0xb8,0x01,0x00,0x00,0x00]
-; NDD-NEXT:    retq # encoding: [0xc3]
-; NDD-NEXT:  .LBB1_2: # %ReturnBlock
-; NDD-NEXT:    xorl %eax, %eax # encoding: [0x31,0xc0]
-; NDD-NEXT:    retq # encoding: [0xc3]
 entry:
   %tmp = load i32, ptr %y
   %tmp1 = shl i32 %tmp, 3
@@ -92,18 +67,6 @@ define i8 @test2b(i8 %X, ptr %y) nounwind {
 ; CHECK-NEXT:  .LBB2_2: # %ReturnBlock
 ; CHECK-NEXT:    xorl %eax, %eax # encoding: [0x31,0xc0]
 ; CHECK-NEXT:    retq # encoding: [0xc3]
-;
-; NDD-LABEL: test2b:
-; NDD:       # %bb.0: # %entry
-; NDD-NEXT:    testb $31, (%rsi) # encoding: [0xf6,0x06,0x1f]
-; NDD-NEXT:    je .LBB2_2 # encoding: [0x74,A]
-; NDD-NEXT:    # fixup A - offset: 1, value: .LBB2_2-1, kind: FK_PCRel_1
-; NDD-NEXT:  # %bb.1: # %cond_true
-; NDD-NEXT:    movb $1, %al # encoding: [0xb0,0x01]
-; NDD-NEXT:    retq # encoding: [0xc3]
-; NDD-NEXT:  .LBB2_2: # %ReturnBlock
-; NDD-NEXT:    xorl %eax, %eax # encoding: [0x31,0xc0]
-; NDD-NEXT:    retq # encoding: [0xc3]
 entry:
   %tmp = load i8, ptr %y
   %tmp1 = shl i8 %tmp, 3
@@ -124,13 +87,6 @@ define i64 @test3(i64 %x) nounwind {
 ; CHECK-NEXT:    testq %rdi, %rdi # encoding: [0x48,0x85,0xff]
 ; CHECK-NEXT:    sete %al # encoding: [0x0f,0x94,0xc0]
 ; CHECK-NEXT:    retq # encoding: [0xc3]
-;
-; NDD-LABEL: test3:
-; NDD:       # %bb.0: # %entry
-; NDD-NEXT:    xorl %eax, %eax # encoding: [0x31,0xc0]
-; NDD-NEXT:    testq %rdi, %rdi # encoding: [0x48,0x85,0xff]
-; NDD-NEXT:    sete %al # encoding: [0x0f,0x94,0xc0]
-; NDD-NEXT:    retq # encoding: [0xc3]
 entry:
   %t = icmp eq i64 %x, 0
   %r = zext i1 %t to i64
@@ -144,13 +100,6 @@ define i64 @test4(i64 %x) nounwind {
 ; CHECK-NEXT:    testq %rdi, %rdi # encoding: [0x48,0x85,0xff]
 ; CHECK-NEXT:    setle %al # encoding: [0x0f,0x9e,0xc0]
 ; CHECK-NEXT:    retq # encoding: [0xc3]
-;
-; NDD-LABEL: test4:
-; NDD:       # %bb.0:
-; NDD-NEXT:    xorl %eax, %eax # encoding: [0x31,0xc0]
-; NDD-NEXT:    testq %rdi, %rdi # encoding: [0x48,0x85,0xff]
-; NDD-NEXT:    setle %al # encoding: [0x0f,0x9e,0xc0]
-; NDD-NEXT:    retq # encoding: [0xc3]
   %t = icmp slt i64 %x, 1
   %r = zext i1 %t to i64
   ret i64 %r
@@ -176,26 +125,6 @@ define i32 @test5(double %A) nounwind {
 ; CHECK-NEXT:    jmp foo@PLT # TAILCALL
 ; CHECK-NEXT:    # encoding: [0xeb,A]
 ; CHECK-NEXT:    # fixup A - offset: 1, value: foo@PLT-1, kind: FK_PCRel_1
-;
-; NDD-LABEL: test5:
-; NDD:       # %bb.0: # %entry
-; NDD-NEXT:    ucomisd {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0 # encoding: [0x66,0x0f,0x2e,0x05,A,A,A,A]
-; NDD-NEXT:    # fixup A - offset: 4, value: {{\.?LCPI[0-9]+_[0-9]+}}-4, kind: reloc_riprel_4byte
-; NDD-NEXT:    ja .LBB5_3 # encoding: [0x77,A]
-; NDD-NEXT:    # fixup A - offset: 1, value: .LBB5_3-1, kind: FK_PCRel_1
-; NDD-NEXT:  # %bb.1: # %entry
-; NDD-NEXT:    ucomisd {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0 # encoding: [0x66,0x0f,0x2e,0x05,A,A,A,A]
-; NDD-NEXT:    # fixup A - offset: 4, value: {{\.?LCPI[0-9]+_[0-9]+}}-4, kind: reloc_riprel_4byte
-; NDD-NEXT:    jb .LBB5_3 # encoding: [0x72,A]
-; NDD-NEXT:    # fixup A - offset: 1, value: .LBB5_3-1, kind: FK_PCRel_1
-; NDD-NEXT:  # %bb.2: # %bb12
-; NDD-NEXT:    movl $32, %eax # encoding: [0xb8,0x20,0x00,0x00,0x00]
-; NDD-NEXT:    retq # encoding: [0xc3]
-; NDD-NEXT:  .LBB5_3: # %bb8
-; NDD-NEXT:    xorl %eax, %eax # encoding: [0x31,0xc0]
-; NDD-NEXT:    jmp foo@PLT # TAILCALL
-; NDD-NEXT:    # encoding: [0xeb,A]
-; NDD-NEXT:    # fixup A - offset: 1, value: foo@PLT-1, kind: FK_PCRel_1
 entry:
   %tmp2 = fcmp ogt double %A, 1.500000e+02
   %tmp5 = fcmp ult double %A, 7.500000e+01
@@ -224,18 +153,6 @@ define i32 @test6() nounwind align 2 {
 ; CHECK-NEXT:  .LBB6_1: # %T
 ; CHECK-NEXT:    movl $1, %eax # encoding: [0xb8,0x01,0x00,0x00,0x00]
 ; CHECK-NEXT:    retq # encoding: [0xc3]
-;
-; NDD-LABEL: test6:
-; NDD:       # %bb.0: # %entry
-; NDD-NEXT:    cmpq $0, -{{[0-9]+}}(%rsp) # encoding: [0x48,0x83,0x7c,0x24,0xf8,0x00]
-; NDD-NEXT:    je .LBB6_1 # encoding: [0x74,A]
-; NDD-NEXT:    # fixup A - offset: 1, value: .LBB6_1-1, kind: FK_PCRel_1
-; NDD-NEXT:  # %bb.2: # %F
-; NDD-NEXT:    xorl %eax, %eax # encoding: [0x31,0xc0]
-; NDD-NEXT:    retq # encoding: [0xc3]
-; NDD-NEXT:  .LBB6_1: # %T
-; NDD-NEXT:    movl $1, %eax # encoding: [0xb8,0x01,0x00,0x00,0x00]
-; NDD-NEXT:    retq # encoding: [0xc3]
 entry:
   %A = alloca { i64, i64 }, align 8
   %B = getelementptr inbounds { i64, i64 }, ptr %A, i64 0, i32 1
@@ -251,12 +168,12 @@ F:
 }
 
 define i32 @test7(i64 %res) nounwind {
-; CHECK-LABEL: test7:
-; CHECK:       # %bb.0: # %entry
-; CHECK-NEXT:    xorl %eax, %eax # encoding: [0x31,0xc0]
-; CHECK-NEXT:    shrq $32, %rdi # encoding: [0x48,0xc1,0xef,0x20]
-; CHECK-NEXT:    sete %al # encoding: [0x0f,0x94,0xc0]
-; CHECK-NEXT:    retq # encoding: [0xc3]
+; NO-NDD-LABEL: test7:
+; NO-NDD:       # %bb.0: # %entry
+; NO-NDD-NEXT:    xorl %eax, %eax # encoding: [0x31,0xc0]
+; NO-NDD-NEXT:    shrq $32, %rdi # encoding: [0x48,0xc1,0xef,0x20]
+; NO-NDD-NEXT:    sete %al # encoding: [0x0f,0x94,0xc0]
+; NO-NDD-NEXT:    retq # encoding: [0xc3]
 ;
 ; NDD-LABEL: test7:
 ; NDD:       # %bb.0: # %entry
@@ -271,13 +188,13 @@ entry:
 }
 
 define i32 @test8(i64 %res) nounwind {
-; CHECK-LABEL: test8:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    shrq $32, %rdi # encoding: [0x48,0xc1,0xef,0x20]
-; CHECK-NEXT:    xorl %eax, %eax # encoding: [0x31,0xc0]
-; CHECK-NEXT:    cmpl $3, %edi # encoding: [0x83,0xff,0x03]
-; CHECK-NEXT:    setb %al # encoding: [0x0f,0x92,0xc0]
-; CHECK-NEXT:    retq # encoding: [0xc3]
+; NO-NDD-LABEL: test8:
+; NO-NDD:       # %bb.0:
+; NO-NDD-NEXT:    shrq $32, %rdi # encoding: [0x48,0xc1,0xef,0x20]
+; NO-NDD-NEXT:    xorl %eax, %eax # encoding: [0x31,0xc0]
+; NO-NDD-NEXT:    cmpl $3, %edi # encoding: [0x83,0xff,0x03]
+; NO-NDD-NEXT:    setb %al # encoding: [0x0f,0x92,0xc0]
+; NO-NDD-NEXT:    retq # encoding: [0xc3]
 ;
 ; NDD-LABEL: test8:
 ; NDD:       # %bb.0:
@@ -292,12 +209,12 @@ define i32 @test8(i64 %res) nounwind {
 }
 
 define i32 @test9(i64 %res) nounwind {
-; CHECK-LABEL: test9:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    xorl %eax, %eax # encoding: [0x31,0xc0]
-; CHECK-NEXT:    shrq $33, %rdi # encoding: [0x48,0xc1,0xef,0x21]
-; CHECK-NEXT:    sete %al # encoding: [0x0f,0x94,0xc0]
-; CHECK-NEXT:    retq # encoding: [0xc3]
+; NO-NDD-LABEL: test9:
+; NO-NDD:       # %bb.0:
+; NO-NDD-NEXT:    xorl %eax, %eax # encoding: [0x31,0xc0]
+; NO-NDD-NEXT:    shrq $33, %rdi # encoding: [0x48,0xc1,0xef,0x21]
+; NO-NDD-NEXT:    sete %al # encoding: [0x0f,0x94,0xc0]
+; NO-NDD-NEXT:    retq # encoding: [0xc3]
 ;
 ; NDD-LABEL: test9:
 ; NDD:       # %bb.0:
@@ -311,12 +228,12 @@ define i32 @test9(i64 %res) nounwind {
 }
 
 define i32 @test10(i64 %res) nounwind {
-; CHECK-LABEL: test10:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    xorl %eax, %eax # encoding: [0x31,0xc0]
-; CHECK-NEXT:    shrq $32, %rdi # encoding: [0x48,0xc1,0xef,0x20]
-; CHECK-NEXT:    setne %al # encoding: [0x0f,0x95,0xc0]
-; CHECK-NEXT:    retq # encoding: [0xc3]
+; NO-NDD-LABEL: test10:
+; NO-NDD:       # %bb.0:
+; NO-NDD-NEXT:    xorl %eax, %eax # encoding: [0x31,0xc0]
+; NO-NDD-NEXT:    shrq $32, %rdi # encoding: [0x48,0xc1,0xef,0x20]
+; NO-NDD-NEXT:    setne %al # encoding: [0x0f,0x95,0xc0]
+; NO-NDD-NEXT:    retq # encoding: [0xc3]
 ;
 ; NDD-LABEL: test10:
 ; NDD:       # %bb.0:
@@ -330,13 +247,13 @@ define i32 @test10(i64 %res) nounwind {
 }
 
 define i32 @test11(i64 %l) nounwind {
-; CHECK-LABEL: test11:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    shrq $47, %rdi # encoding: [0x48,0xc1,0xef,0x2f]
-; CHECK-NEXT:    xorl %eax, %eax # encoding: [0x31,0xc0]
-; CHECK-NEXT:    cmpl $1, %edi # encoding: [0x83,0xff,0x01]
-; CHECK-NEXT:    sete %al # encoding: [0x0f,0x94,0xc0]
-; CHECK-NEXT:    retq # encoding: [0xc3]
+; NO-NDD-LABEL: test11:
+; NO-NDD:       # %bb.0:
+; NO-NDD-NEXT:    shrq $47, %rdi # encoding: [0x48,0xc1,0xef,0x2f]
+; NO-NDD-NEXT:    xorl %eax, %eax # encoding: [0x31,0xc0]
+; NO-NDD-NEXT:    cmpl $1, %edi # encoding: [0x83,0xff,0x01]
+; NO-NDD-NEXT:    sete %al # encoding: [0x0f,0x94,0xc0]
+; NO-NDD-NEXT:    retq # encoding: [0xc3]
 ;
 ; NDD-LABEL: test11:
 ; NDD:       # %bb.0:
@@ -372,27 +289,6 @@ define i32 @test12() ssp uwtable {
 ; CHECK-NEXT:    popq %rcx # encoding: [0x59]
 ; CHECK-NEXT:    .cfi_def_cfa_offset 8
 ; CHECK-NEXT:    retq # encoding: [0xc3]
-;
-; NDD-LABEL: test12:
-; NDD:       # %bb.0: # %entry
-; NDD-NEXT:    pushq %rax # encoding: [0x50]
-; NDD-NEXT:    .cfi_def_cfa_offset 16
-; NDD-NEXT:    callq test12b@PLT # encoding: [0xe8,A,A,A,A]
-; NDD-NEXT:    # fixup A - offset: 1, value: test12b@PLT-4, kind: FK_PCRel_4
-; NDD-NEXT:    testb %al, %al # encoding: [0x84,0xc0]
-; NDD-NEXT:    je .LBB12_2 # encoding: [0x74,A]
-; NDD-NEXT:    # fixup A - offset: 1, value: .LBB12_2-1, kind: FK_PCRel_1
-; NDD-NEXT:  # %bb.1: # %T
-; NDD-NEXT:    movl $1, %eax # encoding: [0xb8,0x01,0x00,0x00,0x00]
-; NDD-NEXT:    popq %rcx # encoding: [0x59]
-; NDD-NEXT:    .cfi_def_cfa_offset 8
-; NDD-NEXT:    retq # encoding: [0xc3]
-; NDD-NEXT:  .LBB12_2: # %F
-; NDD-NEXT:    .cfi_def_cfa_offset 16
-; NDD-NEXT:    movl $2, %eax # encoding: [0xb8,0x02,0x00,0x00,0x00]
-; NDD-NEXT:    popq %rcx # encoding: [0x59]
-; NDD-NEXT:    .cfi_def_cfa_offset 8
-; NDD-NEXT:    retq # encoding: [0xc3]
 entry:
   %tmp1 = call zeroext i1 @test12b()
   br i1 %tmp1, label %T, label %F
@@ -407,12 +303,12 @@ F:
 declare zeroext i1 @test12b()
 
 define i32 @test13(i32 %mask, i32 %base, i32 %intra) {
-; CHECK-LABEL: test13:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    movl %esi, %eax # encoding: [0x89,0xf0]
-; CHECK-NEXT:    testb $8, %dil # encoding: [0x40,0xf6,0xc7,0x08]
-; CHECK-NEXT:    cmovnel %edx, %eax # encoding: [0x0f,0x45,0xc2]
-; CHECK-NEXT:    retq # encoding: [0xc3]
+; NO-NDD-LABEL: test13:
+; NO-NDD:       # %bb.0:
+; NO-NDD-NEXT:    movl %esi, %eax # encoding: [0x89,0xf0]
+; NO-NDD-NEXT:    testb $8, %dil # encoding: [0x40,0xf6,0xc7,0x08]
+; NO-NDD-NEXT:    cmovnel %edx, %eax # encoding: [0x0f,0x45,0xc2]
+; NO-NDD-NEXT:    retq # encoding: [0xc3]
 ;
 ; NDD-LABEL: test13:
 ; NDD:       # %bb.0:
@@ -426,12 +322,12 @@ define i32 @test13(i32 %mask, i32 %base, i32 %intra) {
 }
 
 define i32 @test14(i32 %mask, i32 %base, i32 %intra) {
-; CHECK-LABEL: test14:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    movl %esi, %eax # encoding: [0x89,0xf0]
-; CHECK-NEXT:    shrl $7, %edi # encoding: [0xc1,0xef,0x07]
-; CHECK-NEXT:    cmovnsl %edx, %eax # encoding: [0x0f,0x49,0xc2]
-; CHECK-NEXT:    retq # encoding: [0xc3]
+; NO-NDD-LABEL: test14:
+; NO-NDD:       # %bb.0:
+; NO-NDD-NEXT:    movl %esi, %eax # encoding: [0x89,0xf0]
+; NO-NDD-NEXT:    shrl $7, %edi # encoding: [0xc1,0xef,0x07]
+; NO-NDD-NEXT:    cmovnsl %edx, %eax # encoding: [0x0f,0x49,0xc2]
+; NO-NDD-NEXT:    retq # encoding: [0xc3]
 ;
 ; NDD-LABEL: test14:
 ; NDD:       # %bb.0:
@@ -446,14 +342,14 @@ define i32 @test14(i32 %mask, i32 %base, i32 %intra) {
 
 ; PR19964
 define zeroext i1 @test15(i32 %bf.load, i32 %n) {
-; CHECK-LABEL: test15:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    shrl $16, %edi # encoding: [0xc1,0xef,0x10]
-; CHECK-NEXT:    sete %cl # encoding: [0x0f,0x94,0xc1]
-; CHECK-NEXT:    cmpl %esi, %edi # encoding: [0x39,0xf7]
-; CHECK-NEXT:    setae %al # encoding: [0x0f,0x93,0xc0]
-; CHECK-NEXT:    orb %cl, %al # encoding: [0x08,0xc8]
-; CHECK-NEXT:    retq # encoding: [0xc3]
+; NO-NDD-LABEL: test15:
+; NO-NDD:       # %bb.0:
+; NO-NDD-NEXT:    shrl $16, %edi # encoding: [0xc1,0xef,0x10]
+; NO-NDD-NEXT:    sete %cl # encoding: [0x0f,0x94,0xc1]
+; NO-NDD-NEXT:    cmpl %esi, %edi # encoding: [0x39,0xf7]
+; NO-NDD-NEXT:    setae %al # encoding: [0x0f,0x93,0xc0]
+; NO-NDD-NEXT:    orb %cl, %al # encoding: [0x08,0xc8]
+; NO-NDD-NEXT:    retq # encoding: [0xc3]
 ;
 ; NDD-LABEL: test15:
 ; NDD:       # %bb.0:
@@ -476,12 +372,6 @@ define i8 @signbit_i16(i16 signext %L) {
 ; CHECK-NEXT:    testw %di, %di # encoding: [0x66,0x85,0xff]
 ; CHECK-NEXT:    setns %al # encoding: [0x0f,0x99,0xc0]
 ; CHECK-NEXT:    retq # encoding: [0xc3]
-;
-; NDD-LABEL: signbit_i16:
-; NDD:       # %bb.0:
-; NDD-NEXT:    testw %di, %di # encoding: [0x66,0x85,0xff]
-; NDD-NEXT:    setns %al # encoding: [0x0f,0x99,0xc0]
-; NDD-NEXT:    retq # encoding: [0xc3]
   %lshr = lshr i16 %L, 15
   %trunc = trunc i16 %lshr to i8
   %not = xor i8 %trunc, 1
@@ -494,12 +384,6 @@ define i8 @signbit_i32(i32 %L) {
 ; CHECK-NEXT:    testl %edi, %edi # encoding: [0x85,0xff]
 ; CHECK-NEXT:    setns %al # encoding: [0x0f,0x99,0xc0]
 ; CHECK-NEXT:    retq # encoding: [0xc3]
-;
-; NDD-LABEL: signbit_i32:
-; NDD:       # %bb.0:
-; NDD-NEXT:    testl %edi, %edi # encoding: [0x85,0xff]
-; NDD-NEXT:    setns %al # encoding: [0x0f,0x99,0xc0]
-; NDD-NEXT:    retq # encoding: [0xc3]
   %lshr = lshr i32 %L, 31
   %trunc = trunc i32 %lshr to i8
   %not = xor i8 %trunc, 1
@@ -512,12 +396,6 @@ define i8 @signbit_i64(i64 %L) {
 ; CHECK-NEXT:    testq %rdi, %rdi # encoding: [0x48,0x85,0xff]
 ; CHECK-NEXT:    setns %al # encoding: [0x0f,0x99,0xc0]
 ; CHECK-NEXT:    retq # encoding: [0xc3]
-;
-; NDD-LABEL: signbit_i64:
-; NDD:       # %bb.0:
-; NDD-NEXT:    testq %rdi, %rdi # encoding: [0x48,0x85,0xff]
-; NDD-NEXT:    setns %al # encoding: [0x0f,0x99,0xc0]
-; NDD-NEXT:    retq # encoding: [0xc3]
   %lshr = lshr i64 %L, 63
   %trunc = trunc i64 %lshr to i8
   %not = xor i8 %trunc, 1
@@ -530,12 +408,6 @@ define zeroext i1 @signbit_i32_i1(i32 %L) {
 ; CHECK-NEXT:    testl %edi, %edi # encoding: [0x85,0xff]
 ; CHECK-NEXT:    setns %al # encoding: [0x0f,0x99,0xc0]
 ; CHECK-NEXT:    retq # encoding: [0xc3]
-;
-; NDD-LABEL: signbit_i32_i1:
-; NDD:       # %bb.0:
-; NDD-NEXT:    testl %edi, %edi # encoding: [0x85,0xff]
-; NDD-NEXT:    setns %al # encoding: [0x0f,0x99,0xc0]
-; NDD-NEXT:    retq # encoding: [0xc3]
   %lshr = lshr i32 %L, 31
   %trunc = trunc i32 %lshr to i1
   %not = xor i1 %trunc, true
@@ -544,20 +416,20 @@ define zeroext i1 @signbit_i32_i1(i32 %L) {
 
 ; This test failed due to incorrect handling of "shift + icmp" sequence
 define void @test20(i32 %bf.load, i8 %x1, ptr %b_addr) {
-; CHECK-LABEL: test20:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    xorl %eax, %eax # encoding: [0x31,0xc0]
-; CHECK-NEXT:    testl $16777215, %edi # encoding: [0xf7,0xc7,0xff,0xff,0xff,0x00]
-; CHECK-NEXT:    # imm = 0xFFFFFF
-; CHECK-NEXT:    setne %al # encoding: [0x0f,0x95,0xc0]
-; CHECK-NEXT:    movzbl %sil, %ecx # encoding: [0x40,0x0f,0xb6,0xce]
-; CHECK-NEXT:    addl %eax, %ecx # encoding: [0x01,0xc1]
-; CHECK-NEXT:    setne (%rdx) # encoding: [0x0f,0x95,0x02]
-; CHECK-NEXT:    testl $16777215, %edi # encoding: [0xf7,0xc7,0xff,0xff,0xff,0x00]
-; CHECK-NEXT:    # imm = 0xFFFFFF
-; CHECK-NEXT:    setne d(%rip) # encoding: [0x0f,0x95,0x05,A,A,A,A]
-; CHECK-NEXT:    # fixup A - offset: 3, value: d-4, kind: reloc_riprel_4byte
-; CHECK-NEXT:    retq # encoding: [0xc3]
+; NO-NDD-LABEL: test20:
+; NO-NDD:       # %bb.0:
+; NO-NDD-NEXT:    xorl %eax, %eax # encoding: [0x31,0xc0]
+; NO-NDD-NEXT:    testl $16777215, %edi # encoding: [0xf7,0xc7,0xff,0xff,0xff,0x00]
+; NO-NDD-NEXT:    # imm = 0xFFFFFF
+; NO-NDD-NEXT:    setne %al # encoding: [0x0f,0x95,0xc0]
+; NO-NDD-NEXT:    movzbl %sil, %ecx # encoding: [0x40,0x0f,0xb6,0xce]
+; NO-NDD-NEXT:    addl %eax, %ecx # encoding: [0x01,0xc1]
+; NO-NDD-NEXT:    setne (%rdx) # encoding: [0x0f,0x95,0x02]
+; NO-NDD-NEXT:    testl $16777215, %edi # encoding: [0xf7,0xc7,0xff,0xff,0xff,0x00]
+; NO-NDD-NEXT:    # imm = 0xFFFFFF
+; NO-NDD-NEXT:    setne d(%rip) # encoding: [0x0f,0x95,0x05,A,A,A,A]
+; NO-NDD-NEXT:    # fixup A - offset: 3, value: d-4, kind: reloc_riprel_4byte
+; NO-NDD-NEXT:    retq # encoding: [0xc3]
 ;
 ; NDD-LABEL: test20:
 ; NDD:       # %bb.0:
@@ -593,11 +465,6 @@ define i32 @highmask_i64_simplify(i64 %val) {
 ; CHECK:       # %bb.0:
 ; CHECK-NEXT:    xorl %eax, %eax # encoding: [0x31,0xc0]
 ; CHECK-NEXT:    retq # encoding: [0xc3]
-;
-; NDD-LABEL: highmask_i64_simplify:
-; NDD:       # %bb.0:
-; NDD-NEXT:    xorl %eax, %eax # encoding: [0x31,0xc0]
-; NDD-NEXT:    retq # encoding: [0xc3]
   %and = and i64 %val, -2199023255552
   %cmp = icmp ult i64 %and, 0
   %ret = zext i1 %cmp to i32
@@ -605,12 +472,12 @@ define i32 @highmask_i64_simplify(i64 %val) {
 }
 
 define i32 @highmask_i64_mask64(i64 %val) {
-; CHECK-LABEL: highmask_i64_mask64:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    xorl %eax, %eax # encoding: [0x31,0xc0]
-; CHECK-NEXT:    shrq $41, %rdi # encoding: [0x48,0xc1,0xef,0x29]
-; CHECK-NEXT:    setne %al # encoding: [0x0f,0x95,0xc0]
-; CHECK-NEXT:    retq # encoding: [0xc3]
+; NO-NDD-LABEL: highmask_i64_mask64:
+; NO-NDD:       # %bb.0:
+; NO-NDD-NEXT:    xorl %eax, %eax # encoding: [0x31,0xc0]
+; NO-NDD-NEXT:    shrq $41, %rdi # encoding: [0x48,0xc1,0xef,0x29]
+; NO-NDD-NEXT:    setne %al # encoding: [0x0f,0x95,0xc0]
+; NO-NDD-NEXT:    retq # encoding: [0xc3]
 ;
 ; NDD-LABEL: highmask_i64_mask64:
 ; NDD:       # %bb.0:
@@ -625,14 +492,14 @@ define i32 @highmask_i64_mask64(i64 %val) {
 }
 
 define i64 @highmask_i64_mask64_extra_use(i64 %val) nounwind {
-; CHECK-LABEL: highmask_i64_mask64_extra_use:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    xorl %eax, %eax # encoding: [0x31,0xc0]
-; CHECK-NEXT:    movq %rdi, %rcx # encoding: [0x48,0x89,0xf9]
-; CHECK-NEXT:    shrq $41, %rcx # encoding: [0x48,0xc1,0xe9,0x29]
-; CHECK-NEXT:    setne %al # encoding: [0x0f,0x95,0xc0]
-; CHECK-NEXT:    imulq %rdi, %rax # encoding: [0x48,0x0f,0xaf,0xc7]
-; CHECK-NEXT:    retq # encoding: [0xc3]
+; NO-NDD-LABEL: highmask_i64_mask64_extra_use:
+; NO-NDD:       # %bb.0:
+; NO-NDD-NEXT:    xorl %eax, %eax # encoding: [0x31,0xc0]
+; NO-NDD-NEXT:    movq %rdi, %rcx # encoding: [0x48,0x89,0xf9]
+; NO-NDD-NEXT:    shrq $41, %rcx # encoding: [0x48,0xc1,0xe9,0x29]
+; NO-NDD-NEXT:    setne %al # encoding: [0x0f,0x95,0xc0]
+; NO-NDD-NEXT:    imulq %rdi, %rax # encoding: [0x48,0x0f,0xaf,0xc7]
+; NO-NDD-NEXT:    retq # encoding: [0xc3]
 ;
 ; NDD-LABEL: highmask_i64_mask64_extra_use:
 ; NDD:       # %bb.0:
@@ -649,12 +516,12 @@ define i64 @highmask_i64_mask64_extra_use(i64 %val) nounwind {
 }
 
 define i32 @highmask_i64_mask32(i64 %val) {
-; CHECK-LABEL: highmask_i64_mask32:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    xorl %eax, %eax # encoding: [0x31,0xc0]
-; CHECK-NEXT:    shrq $20, %rdi # encoding: [0x48,0xc1,0xef,0x14]
-; CHECK-NEXT:    sete %al # encoding: [0x0f,0x94,0xc0]
-; CHECK-NEXT:    retq # encoding: [0xc3]
+; NO-NDD-LABEL: highmask_i64_mask32:
+; NO-NDD:       # %bb.0:
+; NO-NDD-NEXT:    xorl %eax, %eax # encoding: [0x31,0xc0]
+; NO-NDD-NEXT:    shrq $20, %rdi # encoding: [0x48,0xc1,0xef,0x14]
+; NO-NDD-NEXT:    sete %al # encoding: [0x0f,0x94,0xc0]
+; NO-NDD-NEXT:    retq # encoding: [0xc3]
 ;
 ; NDD-LABEL: highmask_i64_mask32:
 ; NDD:       # %bb.0:
@@ -669,14 +536,14 @@ define i32 @highmask_i64_mask32(i64 %val) {
 }
 
 define i64 @highmask_i64_mask32_extra_use(i64 %val) nounwind {
-; CHECK-LABEL: highmask_i64_mask32_extra_use:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    xorl %eax, %eax # encoding: [0x31,0xc0]
-; CHECK-NEXT:    testq $-1048576, %rdi # encoding: [0x48,0xf7,0xc7,0x00,0x00,0xf0,0xff]
-; CHECK-NEXT:    # imm = 0xFFF00000
-; CHECK-NEXT:    sete %al # encoding: [0x0f,0x94,0xc0]
-; CHECK-NEXT:    imulq %rdi, %rax # encoding: [0x48,0x0f,0xaf,0xc7]
-; CHECK-NEXT:    retq # encoding: [0xc3]
+; NO-NDD-LABEL: highmask_i64_mask32_extra_use:
+; NO-NDD:       # %bb.0:
+; NO-NDD-NEXT:    xorl %eax, %eax # encoding: [0x31,0xc0]
+; NO-NDD-NEXT:    testq $-1048576, %rdi # encoding: [0x48,0xf7,0xc7,0x00,0x00,0xf0,0xff]
+; NO-NDD-NEXT:    # imm = 0xFFF00000
+; NO-NDD-NEXT:    sete %al # encoding: [0x0f,0x94,0xc0]
+; NO-NDD-NEXT:    imulq %rdi, %rax # encoding: [0x48,0x0f,0xaf,0xc7]
+; NO-NDD-NEXT:    retq # encoding: [0xc3]
 ;
 ; NDD-LABEL: highmask_i64_mask32_extra_use:
 ; NDD:       # %bb.0:
@@ -700,13 +567,6 @@ define i32 @highmask_i64_mask8(i64 %val) {
 ; CHECK-NEXT:    testq $-16, %rdi # encoding: [0x48,0xf7,0xc7,0xf0,0xff,0xff,0xff]
 ; CHECK-NEXT:    setne %al # encoding: [0x0f,0x95,0xc0]
 ; CHECK-NEXT:    retq # encoding: [0xc3]
-;
-; NDD-LABEL: highmask_i64_mask8:
-; NDD:       # %bb.0:
-; NDD-NEXT:    xorl %eax, %eax # encoding: [0x31,0xc0]
-; NDD-NEXT:    testq $-16, %rdi # encoding: [0x48,0xf7,0xc7,0xf0,0xff,0xff,0xff]
-; NDD-NEXT:    setne %al # encoding: [0x0f,0x95,0xc0]
-; NDD-NEXT:    retq # encoding: [0xc3]
   %and = and i64 %val, -16
   %cmp = icmp ne i64 %and, 0
   %ret = zext i1 %cmp to i32
@@ -714,12 +574,12 @@ define i32 @highmask_i64_mask8(i64 %val) {
 }
 
 define i32 @lowmask_i64_mask64(i64 %val) {
-; CHECK-LABEL: lowmask_i64_mask64:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    xorl %eax, %eax # encoding: [0x31,0xc0]
-; CHECK-NEXT:    shlq $16, %rdi # encoding: [0x48,0xc1,0xe7,0x10]
-; CHECK-NEXT:    sete %al # encoding: [0x0f,0x94,0xc0]
-; CHECK-NEXT:    retq # encoding: [0xc3]
+; NO-NDD-LABEL: lowmask_i64_mask64:
+; NO-NDD:       # %bb.0:
+; NO-NDD-NEXT:    xorl %eax, %eax # encoding: [0x31,0xc0]
+; NO-NDD-NEXT:    shlq $16, %rdi # encoding: [0x48,0xc1,0xe7,0x10]
+; NO-NDD-NEXT:    sete %al # encoding: [0x0f,0x94,0xc0]
+; NO-NDD-NEXT:    retq # encoding: [0xc3]
 ;
 ; NDD-LABEL: lowmask_i64_mask64:
 ; NDD:       # %bb.0:
@@ -734,14 +594,14 @@ define i32 @lowmask_i64_mask64(i64 %val) {
 }
 
 define i64 @lowmask_i64_mask64_extra_use(i64 %val) nounwind {
-; CHECK-LABEL: lowmask_i64_mask64_extra_use:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    xorl %eax, %eax # encoding: [0x31,0xc0]
-; CHECK-NEXT:    movq %rdi, %rcx # encoding: [0x48,0x89,0xf9]
-; CHECK-NEXT:    shlq $16, %rcx # encoding: [0x48,0xc1,0xe1,0x10]
-; CHECK-NEXT:    sete %al # encoding: [0x0f,0x94,0xc0]
-; CHECK-NEXT:    imulq %rdi, %rax # encoding: [0x48,0x0f,0xaf,0xc7]
-; CHECK-NEXT:    retq # encoding: [0xc3]
+; NO-NDD-LABEL: lowmask_i64_mask64_extra_use:
+; NO-NDD:       # %bb.0:
+; NO-NDD-NEXT:    xorl %eax, %eax # encoding: [0x31,0xc0]
+; NO-NDD-NEXT:    movq %rdi, %rcx # encoding: [0x48,0x89,0xf9]
+; NO-NDD-NEXT:    shlq $16, %rcx # encoding: [0x48,0xc1,0xe1,0x10]
+; NO-NDD-NEXT:    sete %al # encoding: [0x0f,0x94,0xc0]
+; NO-NDD-NEXT:    imulq %rdi, %rax # encoding: [0x48,0x0f,0xaf,0xc7]
+; NO-NDD-NEXT:    retq # encoding: [0xc3]
 ;
 ; NDD-LABEL: lowmask_i64_mask64_extra_use:
 ; NDD:       # %bb.0:
@@ -758,12 +618,12 @@ define i64 @lowmask_i64_mask64_extra_use(i64 %val) nounwind {
 }
 
 define i32 @lowmask_i64_mask32(i64 %val) {
-; CHECK-LABEL: lowmask_i64_mask32:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    xorl %eax, %eax # encoding: [0x31,0xc0]
-; CHECK-NEXT:    shlq $44, %rdi # encoding: [0x48,0xc1,0xe7,0x2c]
-; CHECK-NEXT:    setne %al # encoding: [0x0f,0x95,0xc0]
-; CHECK-NEXT:    retq # encoding: [0xc3]
+; NO-NDD-LABEL: lowmask_i64_mask32:
+; NO-NDD:       # %bb.0:
+; NO-NDD-NEXT:    xorl %eax, %eax # encoding: [0x31,0xc0]
+; NO-NDD-NEXT:    shlq $44, %rdi # encoding: [0x48,0xc1,0xe7,0x2c]
+; NO-NDD-NEXT:    setne %al # encoding: [0x0f,0x95,0xc0]
+; NO-NDD-NEXT:    retq # encoding: [0xc3]
 ;
 ; NDD-LABEL: lowmask_i64_mask32:
 ; NDD:       # %bb.0:
@@ -778,14 +638,14 @@ define i32 @lowmask_i64_mask32(i64 %val) {
 }
 
 define i64 @lowmask_i64_mask32_extra_use(i64 %val) nounwind {
-; CHECK-LABEL: lowmask_i64_mask32_extra_use:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    xorl %eax, %eax # encoding: [0x31,0xc0]
-; CHECK-NEXT:    testl $1048575, %edi # encoding: [0xf7,0xc7,0xff,0xff,0x0f,0x00]
-; CHECK-NEXT:    # imm = 0xFFFFF
-; CHECK-NEXT:    setne %al # encoding: [0x0f,0x95,0xc0]
-; CHECK-NEXT:    imulq %rdi, %rax # encoding: [0x48,0x0f,0xaf,0xc7]
-; CHECK-NEXT:    retq # encoding: [0xc3]
+; NO-NDD-LABEL: lowmask_i64_mask32_extra_use:
+; NO-NDD:       # %bb.0:
+; NO-NDD-NEXT:    xorl %eax, %eax # encoding: [0x31,0xc0]
+; NO-NDD-NEXT:    testl $1048575, %edi # encoding: [0xf7,0xc7,0xff,0xff,0x0f,0x00]
+; NO-NDD-NEXT:    # imm = 0xFFFFF
+; NO-NDD-NEXT:    setne %al # encoding: [0x0f,0x95,0xc0]
+; NO-NDD-NEXT:    imulq %rdi, %rax # encoding: [0x48,0x0f,0xaf,0xc7]
+; NO-NDD-NEXT:    retq # encoding: [0xc3]
 ;
 ; NDD-LABEL: lowmask_i64_mask32_extra_use:
 ; NDD:       # %bb.0:
@@ -809,13 +669,6 @@ define i32 @lowmask_i64_mask8(i64 %val) {
 ; CHECK-NEXT:    testb $31, %dil # encoding: [0x40,0xf6,0xc7,0x1f]
 ; CHECK-NEXT:    sete %al # encoding: [0x0f,0x94,0xc0]
 ; CHECK-NEXT:    retq # encoding: [0xc3]
-;
-; NDD-LABEL: lowmask_i64_mask8:
-; NDD:       # %bb.0:
-; NDD-NEXT:    xorl %eax, %eax # encoding: [0x31,0xc0]
-; NDD-NEXT:    testb $31, %dil # encoding: [0x40,0xf6,0xc7,0x1f]
-; NDD-NEXT:    sete %al # encoding: [0x0f,0x94,0xc0]
-; NDD-NEXT:    retq # encoding: [0xc3]
   %and = and i64 %val, 31
   %cmp = icmp eq i64 %and, 0
   %ret = zext i1 %cmp to i32
@@ -830,14 +683,6 @@ define i32 @highmask_i32_mask32(i32 %val) {
 ; CHECK-NEXT:    # imm = 0xFFF00000
 ; CHECK-NEXT:    setne %al # encoding: [0x0f,0x95,0xc0]
 ; CHECK-NEXT:    retq # encoding: [0xc3]
-;
-; NDD-LABEL: highmask_i32_mask32:
-; NDD:       # %bb.0:
-; NDD-NEXT:    xorl %eax, %eax # encoding: [0x31,0xc0]
-; NDD-NEXT:    testl $-1048576, %edi # encoding: [0xf7,0xc7,0x00,0x00,0xf0,0xff]
-; NDD-NEXT:    # imm = 0xFFF00000
-; NDD-NEXT:    setne %al # encoding: [0x0f,0x95,0xc0]
-; NDD-NEXT:    retq # encoding: [0xc3]
   %and = and i32 %val, -1048576
   %cmp = icmp ne i32 %and, 0
   %ret = zext i1 %cmp to i32
@@ -851,13 +696,6 @@ define i32 @highmask_i32_mask8(i32 %val) {
 ; CHECK-NEXT:    testl $-16, %edi # encoding: [0xf7,0xc7,0xf0,0xff,0xff,0xff]
 ; CHECK-NEXT:    sete %al # encoding: [0x0f,0x94,0xc0]
 ; CHECK-NEXT:    retq # encoding: [0xc3]
-;
-; NDD-LABEL: highmask_i32_mask8:
-; NDD:       # %bb.0:
-; NDD-NEXT:    xorl %eax, %eax # encoding: [0x31,0xc0]
-; NDD-NEXT:    testl $-16, %edi # encoding: [0xf7,0xc7,0xf0,0xff,0xff,0xff]
-; NDD-NEXT:    sete %al # encoding: [0x0f,0x94,0xc0]
-; NDD-NEXT:    retq # encoding: [0xc3]
   %and = and i32 %val, -16
   %cmp = icmp eq i32 %and, 0
   %ret = zext i1 %cmp to i32
@@ -872,14 +710,6 @@ define i32 @lowmask_i32_mask32(i32 %val) {
 ; CHECK-NEXT:    # imm = 0xFFFFF
 ; CHECK-NEXT:    sete %al # encoding: [0x0f,0x94,0xc0]
 ; CHECK-NEXT:    retq # encoding: [0xc3]
-;
-; NDD-LABEL: lowmask_i32_mask32:
-; NDD:       # %bb.0:
-; NDD-NEXT:    xorl %eax, %eax # encoding: [0x31,0xc0]
-; NDD-NEXT:    testl $1048575, %edi # encoding: [0xf7,0xc7,0xff,0xff,0x0f,0x00]
-; NDD-NEXT:    # imm = 0xFFFFF
-; NDD-NEXT:    sete %al # encoding: [0x0f,0x94,0xc0]
-; NDD-NEXT:    retq # encoding: [0xc3]
   %and = and i32 %val, 1048575
   %cmp = icmp eq i32 %and, 0
   %ret = zext i1 %cmp to i32
@@ -893,13 +723,6 @@ define i32 @lowmask_i32_mask8(i32 %val) {
 ; CHECK-NEXT:    testb $31, %dil # encoding: [0x40,0xf6,0xc7,0x1f]
 ; CHECK-NEXT:    setne %al # encoding: [0x0f,0x95,0xc0]
 ; CHECK-NEXT:    retq # encoding: [0xc3]
-;
-; NDD-LABEL: lowmask_i32_mask8:
-; NDD:       # %bb.0:
-; NDD-NEXT:    xorl %eax, %eax # encoding: [0x31,0xc0]
-; NDD-NEXT:    testb $31, %dil # encoding: [0x40,0xf6,0xc7,0x1f]
-; NDD-NEXT:    setne %al # encoding: [0x0f,0x95,0xc0]
-; NDD-NEXT:    retq # encoding: [0xc3]
   %and = and i32 %val, 31
   %cmp = icmp ne i32 %and, 0
   %ret = zext i1 %cmp to i32
@@ -907,12 +730,12 @@ define i32 @lowmask_i32_mask8(i32 %val) {
 }
 
 define i1 @shifted_mask64_testb(i64 %a) {
-; CHECK-LABEL: shifted_mask64_testb:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    shrq $50, %rdi # encoding: [0x48,0xc1,0xef,0x32]
-; CHECK-NEXT:    testb %dil, %dil # encoding: [0x40,0x84,0xff]
-; CHECK-NEXT:    setne %al # encoding: [0x0f,0x95,0xc0]
-; CHECK-NEXT:    retq # encoding: [0xc3]
+; NO-NDD-LABEL: shifted_mask64_testb:
+; NO-NDD:       # %bb.0:
+; NO-NDD-NEXT:    shrq $50, %rdi # encoding: [0x48,0xc1,0xef,0x32]
+; NO-NDD-NEXT:    testb %dil, %dil # encoding: [0x40,0x84,0xff]
+; NO-NDD-NEXT:    setne %al # encoding: [0x0f,0x95,0xc0]
+; NO-NDD-NEXT:    retq # encoding: [0xc3]
 ;
 ; NDD-LABEL: shifted_mask64_testb:
 ; NDD:       # %bb.0:
@@ -926,12 +749,12 @@ define i1 @shifted_mask64_testb(i64 %a) {
 }
 
 define i1 @shifted_mask64_testw(i64 %a) {
-; CHECK-LABEL: shifted_mask64_testw:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    shrq $33, %rdi # encoding: [0x48,0xc1,0xef,0x21]
-; CHECK-NEXT:    testw %di, %di # encoding: [0x66,0x85,0xff]
-; CHECK-NEXT:    setne %al # encoding: [0x0f,0x95,0xc0]
-; CHECK-NEXT:    retq # encoding: [0xc3]
+; NO-NDD-LABEL: shifted_mask64_testw:
+; NO-NDD:       # %bb.0:
+; NO-NDD-NEXT:    shrq $33, %rdi # encoding: [0x48,0xc1,0xef,0x21]
+; NO-NDD-NEXT:    testw %di, %di # encoding: [0x66,0x85,0xff]
+; NO-NDD-NEXT:    setne %al # encoding: [0x0f,0x95,0xc0]
+; NO-NDD-NEXT:    retq # encoding: [0xc3]
 ;
 ; NDD-LABEL: shifted_mask64_testw:
 ; NDD:       # %bb.0:
@@ -945,12 +768,12 @@ define i1 @shifted_mask64_testw(i64 %a) {
 }
 
 define i1 @shifted_mask64_testl(i64 %a) {
-; CHECK-LABEL: shifted_mask64_testl:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    shrq $7, %rdi # encoding: [0x48,0xc1,0xef,0x07]
-; CHECK-NEXT:    testl %edi, %edi # encoding: [0x85,0xff]
-; CHECK-NEXT:    sete %al # encoding: [0x0f,0x94,0xc0]
-; CHECK-NEXT:    retq # encoding: [0xc3]
+; NO-NDD-LABEL: shifted_mask64_testl:
+; NO-NDD:       # %bb.0:
+; NO-NDD-NEXT:    shrq $7, %rdi # encoding: [0x48,0xc1,0xef,0x07]
+; NO-NDD-NEXT:    testl %edi, %edi # encoding: [0x85,0xff]
+; NO-NDD-NEXT:    sete %al # encoding: [0x0f,0x94,0xc0]
+; NO-NDD-NEXT:    retq # encoding: [0xc3]
 ;
 ; NDD-LABEL: shifted_mask64_testl:
 ; NDD:       # %bb.0:
@@ -964,15 +787,15 @@ define i1 @shifted_mask64_testl(i64 %a) {
 }
 
 define i1 @shifted_mask64_extra_use_const(i64 %a) {
-; CHECK-LABEL: shifted_mask64_extra_use_const:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    movabsq $287104476244869120, %rcx # encoding: [0x48,0xb9,0x00,0x00,0x00,0x00,0x00,0x00,0xfc,0x03]
-; CHECK-NEXT:    # imm = 0x3FC000000000000
-; CHECK-NEXT:    testq %rcx, %rdi # encoding: [0x48,0x85,0xcf]
-; CHECK-NEXT:    setne %al # encoding: [0x0f,0x95,0xc0]
-; CHECK-NEXT:    movq %rcx, d64(%rip) # encoding: [0x48,0x89,0x0d,A,A,A,A]
-; CHECK-NEXT:    # fixup A - offset: 3, value: d64-4, kind: reloc_riprel_4byte
-; CHECK-NEXT:    retq # encoding: [0xc3]
+; NO-NDD-LABEL: shifted_mask64_extra_use_const:
+; NO-NDD:       # %bb.0:
+; NO-NDD-NEXT:    movabsq $287104476244869120, %rcx # encoding: [0x48,0xb9,0x00,0x00,0x00,0x00,0x00,0x00,0xfc,0x03]
+; NO-NDD-NEXT:    # imm = 0x3FC000000000000
+; NO-NDD-NEXT:    testq %rcx, %rdi # encoding: [0x48,0x85,0xcf]
+; NO-NDD-NEXT:    setne %al # encoding: [0x0f,0x95,0xc0]
+; NO-NDD-NEXT:    movq %rcx, d64(%rip) # encoding: [0x48,0x89,0x0d,A,A,A,A]
+; NO-NDD-NEXT:    # fixup A - offset: 3, value: d64-4, kind: reloc_riprel_4byte
+; NO-NDD-NEXT:    retq # encoding: [0xc3]
 ;
 ; NDD-LABEL: shifted_mask64_extra_use_const:
 ; NDD:       # %bb.0:
@@ -990,15 +813,15 @@ define i1 @shifted_mask64_extra_use_const(i64 %a) {
 }
 
 define i1 @shifted_mask64_extra_use_and(i64 %a) {
-; CHECK-LABEL: shifted_mask64_extra_use_and:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    movabsq $287104476244869120, %rcx # encoding: [0x48,0xb9,0x00,0x00,0x00,0x00,0x00,0x00,0xfc,0x03]
-; CHECK-NEXT:    # imm = 0x3FC000000000000
-; CHECK-NEXT:    andq %rdi, %rcx # encoding: [0x48,0x21,0xf9]
-; CHECK-NEXT:    setne %al # encoding: [0x0f,0x95,0xc0]
-; CHECK-NEXT:    movq %rcx, d64(%rip) # encoding: [0x48,0x89,0x0d,A,A,A,A]
-; CHECK-NEXT:    # fixup A - offset: 3, value: d64-4, kind: reloc_riprel_4byte
-; CHECK-NEXT:    retq # encoding: [0xc3]
+; NO-NDD-LABEL: shifted_mask64_extra_use_and:
+; NO-NDD:       # %bb.0:
+; NO-NDD-NEXT:    movabsq $287104476244869120, %rcx # encoding: [0x48,0xb9,0x00,0x00,0x00,0x00,0x00,0x00,0xfc,0x03]
+; NO-NDD-NEXT:    # imm = 0x3FC000000000000
+; NO-NDD-NEXT:    andq %rdi, %rcx # encoding: [0x48,0x21,0xf9]
+; NO-NDD-NEXT:    setne %al # encoding: [0x0f,0x95,0xc0]
+; NO-NDD-NEXT:    movq %rcx, d64(%rip) # encoding: [0x48,0x89,0x0d,A,A,A,A]
+; NO-NDD-NEXT:    # fixup A - offset: 3, value: d64-4, kind: reloc_riprel_4byte
+; NO-NDD-NEXT:    retq # encoding: [0xc3]
 ;
 ; NDD-LABEL: shifted_mask64_extra_use_and:
 ; NDD:       # %bb.0:
@@ -1022,13 +845,6 @@ define i1 @shifted_mask32_testl_immediate(i64 %a) {
 ; CHECK-NEXT:    # imm = 0x3FC0000
 ; CHECK-NEXT:    setne %al # encoding: [0x0f,0x95,0xc0]
 ; CHECK-NEXT:    retq # encoding: [0xc3]
-;
-; NDD-LABEL: shifted_mask32_testl_immediate:
-; NDD:       # %bb.0:
-; NDD-NEXT:    testl $66846720, %edi # encoding: [0xf7,0xc7,0x00,0x00,0xfc,0x03]
-; NDD-NEXT:    # imm = 0x3FC0000
-; NDD-NEXT:    setne %al # encoding: [0x0f,0x95,0xc0]
-; NDD-NEXT:    retq # encoding: [0xc3]
   %v0 = and i64 %a, 66846720  ; 0xff << 18
   %v1 = icmp ne i64 %v0, 0
   ret i1 %v1
@@ -1044,16 +860,6 @@ define i1 @shifted_mask32_extra_use_const(i64 %a) {
 ; CHECK-NEXT:    # fixup A - offset: 3, value: d64-8, kind: reloc_riprel_4byte
 ; CHECK-NEXT:    # imm = 0x3FC0000
 ; CHECK-NEXT:    retq # encoding: [0xc3]
-;
-; NDD-LABEL: shifted_mask32_extra_use_const:
-; NDD:       # %bb.0:
-; NDD-NEXT:    testl $66846720, %edi # encoding: [0xf7,0xc7,0x00,0x00,0xfc,0x03]
-; NDD-NEXT:    # imm = 0x3FC0000
-; NDD-NEXT:    setne %al # encoding: [0x0f,0x95,0xc0]
-; NDD-NEXT:    movq $66846720, d64(%rip) # encoding: [0x48,0xc7,0x05,A,A,A,A,0x00,0x00,0xfc,0x03]
-; NDD-NEXT:    # fixup A - offset: 3, value: d64-8, kind: reloc_riprel_4byte
-; NDD-NEXT:    # imm = 0x3FC0000
-; NDD-NEXT:    retq # encoding: [0xc3]
   %v0 = and i64 %a, 66846720  ; 0xff << 18
   %v1 = icmp ne i64 %v0, 0
   store i64 66846720, ptr @d64
@@ -1061,14 +867,14 @@ define i1 @shifted_mask32_extra_use_const(i64 %a) {
 }
 
 define i1 @shifted_mask32_extra_use_and(i64 %a) {
-; CHECK-LABEL: shifted_mask32_extra_use_and:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    andq $66846720, %rdi # encoding: [0x48,0x81,0xe7,0x00,0x00,0xfc,0x03]
-; CHECK-NEXT:    # imm = 0x3FC0000
-; CHECK-NEXT:    setne %al # encoding: [0x0f,0x95,0xc0]
-; CHECK-NEXT:    movq %rdi, d64(%rip) # encoding: [0x48,0x89,0x3d,A,A,A,A]
-; CHECK-NEXT:    # fixup A - offset: 3, value: d64-4, kind: reloc_riprel_4byte
-; CHECK-NEXT:    retq # encoding: [0xc3]
+; NO-NDD-LABEL: shifted_mask32_extra_use_and:
+; NO-NDD:       # %bb.0:
+; NO-NDD-NEXT:    andq $66846720, %rdi # encoding: [0x48,0x81,0xe7,0x00,0x00,0xfc,0x03]
+; NO-NDD-NEXT:    # imm = 0x3FC0000
+; NO-NDD-NEXT:    setne %al # encoding: [0x0f,0x95,0xc0]
+; NO-NDD-NEXT:    movq %rdi, d64(%rip) # encoding: [0x48,0x89,0x3d,A,A,A,A]
+; NO-NDD-NEXT:    # fixup A - offset: 3, value: d64-4, kind: reloc_riprel_4byte
+; NO-NDD-NEXT:    retq # encoding: [0xc3]
 ;
 ; NDD-LABEL: shifted_mask32_extra_use_and:
 ; NDD:       # %bb.0:
@@ -1085,14 +891,14 @@ define i1 @shifted_mask32_extra_use_and(i64 %a) {
 }
 
 define { i64, i64 } @pr39968(i64, i64, i32) {
-; CHECK-LABEL: pr39968:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    xorl %eax, %eax # encoding: [0x31,0xc0]
-; CHECK-NEXT:    testb $64, %dl # encoding: [0xf6,0xc2,0x40]
-; CHECK-NEXT:    cmovneq %rdi, %rsi # encoding: [0x48,0x0f,0x45,0xf7]
-; CHECK-NEXT:    cmovneq %rdi, %rax # encoding: [0x48,0x0f,0x45,0xc7]
-; CHECK-NEXT:    movq %rsi, %rdx # encoding: [0x48,0x89,0xf2]
-; CHECK-NEXT:    retq # encoding: [0xc3]
+; NO-NDD-LABEL: pr39968:
+; NO-NDD:       # %bb.0:
+; NO-NDD-NEXT:    xorl %eax, %eax # encoding: [0x31,0xc0]
+; NO-NDD-NEXT:    testb $64, %dl # encoding: [0xf6,0xc2,0x40]
+; NO-NDD-NEXT:    cmovneq %rdi, %rsi # encoding: [0x48,0x0f,0x45,0xf7]
+; NO-NDD-NEXT:    cmovneq %rdi, %rax # encoding: [0x48,0x0f,0x45,0xc7]
+; NO-NDD-NEXT:    movq %rsi, %rdx # encoding: [0x48,0x89,0xf2]
+; NO-NDD-NEXT:    retq # encoding: [0xc3]
 ;
 ; NDD-LABEL: pr39968:
 ; NDD:       # %bb.0:
@@ -1124,18 +930,6 @@ define i32 @pr42189(i16 signext %c) {
 ; CHECK-NEXT:    jmp g@PLT # TAILCALL
 ; CHECK-NEXT:    # encoding: [0xeb,A]
 ; CHECK-NEXT:    # fixup A - offset: 1, value: g@PLT-1, kind: FK_PCRel_1
-;
-; NDD-LABEL: pr42189:
-; NDD:       # %bb.0: # %entry
-; NDD-NEXT:    cmpl $32767, %edi # encoding: [0x81,0xff,0xff,0x7f,0x00,0x00]
-; NDD-NEXT:    # imm = 0x7FFF
-; NDD-NEXT:    jne f@PLT # TAILCALL
-; NDD-NEXT:    # encoding: [0x75,A]
-; NDD-NEXT:    # fixup A - offset: 1, value: f@PLT-1, kind: FK_PCRel_1
-; NDD-NEXT:  # %bb.1: # %if.then
-; NDD-NEXT:    jmp g@PLT # TAILCALL
-; NDD-NEXT:    # encoding: [0xeb,A]
-; NDD-NEXT:    # fixup A - offset: 1, value: g@PLT-1, kind: FK_PCRel_1
 entry:
   %cmp = icmp eq i16 %c, 32767
   br i1 %cmp, label %if.then, label %if.end
@@ -1160,12 +954,12 @@ declare i32 @f()
 ; The store makes sure the chain result of the load is used which used to
 ; prevent the post isel peephole from catching this.
 define i1 @fold_test_and_with_chain(ptr %x, ptr %y, i32 %z) {
-; CHECK-LABEL: fold_test_and_with_chain:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    testl %edx, (%rdi) # encoding: [0x85,0x17]
-; CHECK-NEXT:    sete %al # encoding: [0x0f,0x94,0xc0]
-; CHECK-NEXT:    movl %edx, (%rsi) # encoding: [0x89,0x16]
-; CHECK-NEXT:    retq # encoding: [0xc3]
+; NO-NDD-LABEL: fold_test_and_with_chain:
+; NO-NDD:       # %bb.0:
+; NO-NDD-NEXT:    testl %edx, (%rdi) # encoding: [0x85,0x17]
+; NO-NDD-NEXT:    sete %al # encoding: [0x0f,0x94,0xc0]
+; NO-NDD-NEXT:    movl %edx, (%rsi) # encoding: [0x89,0x16]
+; NO-NDD-NEXT:    retq # encoding: [0xc3]
 ;
 ; NDD-LABEL: fold_test_and_with_chain:
 ; NDD:       # %bb.0:
diff --git a/llvm/test/CodeGen/X86/combine-mul.ll b/llvm/test/CodeGen/X86/combine-mul.ll
index 8d2bb77a9e1a..5d7bf4a2c978 100644
--- a/llvm/test/CodeGen/X86/combine-mul.ll
+++ b/llvm/test/CodeGen/X86/combine-mul.ll
@@ -80,13 +80,13 @@ define <4 x i32> @combine_vec_mul_pow2b(<4 x i32> %x) {
 define <4 x i64> @combine_vec_mul_pow2c(<4 x i64> %x) {
 ; SSE-LABEL: combine_vec_mul_pow2c:
 ; SSE:       # %bb.0:
+; SSE-NEXT:    movdqa %xmm0, %xmm2
+; SSE-NEXT:    paddq %xmm0, %xmm2
+; SSE-NEXT:    pblendw {{.*#+}} xmm0 = xmm0[0,1,2,3],xmm2[4,5,6,7]
 ; SSE-NEXT:    movdqa %xmm1, %xmm2
 ; SSE-NEXT:    psllq $4, %xmm2
 ; SSE-NEXT:    psllq $2, %xmm1
 ; SSE-NEXT:    pblendw {{.*#+}} xmm1 = xmm1[0,1,2,3],xmm2[4,5,6,7]
-; SSE-NEXT:    movdqa %xmm0, %xmm2
-; SSE-NEXT:    paddq %xmm0, %xmm2
-; SSE-NEXT:    pblendw {{.*#+}} xmm0 = xmm0[0,1,2,3],xmm2[4,5,6,7]
 ; SSE-NEXT:    retq
 ;
 ; AVX-LABEL: combine_vec_mul_pow2c:
@@ -399,14 +399,12 @@ define i64 @combine_mul_self_demandedbits(i64 %x) {
 ; SSE:       # %bb.0:
 ; SSE-NEXT:    movq %rdi, %rax
 ; SSE-NEXT:    imulq %rdi, %rax
-; SSE-NEXT:    andq $-3, %rax
 ; SSE-NEXT:    retq
 ;
 ; AVX-LABEL: combine_mul_self_demandedbits:
 ; AVX:       # %bb.0:
 ; AVX-NEXT:    movq %rdi, %rax
 ; AVX-NEXT:    imulq %rdi, %rax
-; AVX-NEXT:    andq $-3, %rax
 ; AVX-NEXT:    retq
   %1 = mul i64 %x, %x
   %2 = and i64 %1, -3
diff --git a/llvm/test/CodeGen/X86/combine-or-shuffle.ll b/llvm/test/CodeGen/X86/combine-or-shuffle.ll
new file mode 100644
index 000000000000..175d21a4f706
--- /dev/null
+++ b/llvm/test/CodeGen/X86/combine-or-shuffle.ll
@@ -0,0 +1,862 @@
+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
+; RUN: llc < %s -mtriple=x86_64-unknown-linux-gnu -mattr=+sse    | FileCheck %s -check-prefixes=SSE,SSE2
+; RUN: llc < %s -mtriple=x86_64-unknown-linux-gnu -mattr=+sse4.1 | FileCheck %s -check-prefixes=SSE,SSE4
+; RUN: llc < %s -mtriple=x86_64-unknown-linux-gnu -mattr=+avx  | FileCheck %s -check-prefixes=AVX,AVX1
+; RUN: llc < %s -mtriple=x86_64-unknown-linux-gnu -mattr=+avx2 | FileCheck %s -check-prefixes=AVX,AVX2
+; RUN: llc < %s -mtriple=x86_64-unknown-linux-gnu -mcpu=x86-64-v4 | FileCheck %s -check-prefixes=AVX,AVX512
+
+; Verify that each of the following test cases is folded into a single
+; instruction which performs a blend operation.
+
+define <2 x i64> @test1(<2 x i64> %a, <2 x i64> %b) {
+; SSE2-LABEL: test1:
+; SSE2:       # %bb.0:
+; SSE2-NEXT:    shufps {{.*#+}} xmm0 = xmm0[0,1],xmm1[2,3]
+; SSE2-NEXT:    retq
+;
+; SSE4-LABEL: test1:
+; SSE4:       # %bb.0:
+; SSE4-NEXT:    blendps {{.*#+}} xmm0 = xmm0[0,1],xmm1[2,3]
+; SSE4-NEXT:    retq
+;
+; AVX-LABEL: test1:
+; AVX:       # %bb.0:
+; AVX-NEXT:    vblendps {{.*#+}} xmm0 = xmm0[0,1],xmm1[2,3]
+; AVX-NEXT:    retq
+  %shuf1 = shufflevector <2 x i64> %a, <2 x i64> zeroinitializer, <2 x i32><i32 0, i32 2>
+  %shuf2 = shufflevector <2 x i64> %b, <2 x i64> zeroinitializer, <2 x i32><i32 2, i32 1>
+  %or = or <2 x i64> %shuf1, %shuf2
+  ret <2 x i64> %or
+}
+
+
+define <4 x i32> @test2(<4 x i32> %a, <4 x i32> %b) {
+; SSE2-LABEL: test2:
+; SSE2:       # %bb.0:
+; SSE2-NEXT:    movsd {{.*#+}} xmm0 = xmm1[0],xmm0[1]
+; SSE2-NEXT:    retq
+;
+; SSE4-LABEL: test2:
+; SSE4:       # %bb.0:
+; SSE4-NEXT:    blendps {{.*#+}} xmm0 = xmm1[0,1],xmm0[2,3]
+; SSE4-NEXT:    retq
+;
+; AVX-LABEL: test2:
+; AVX:       # %bb.0:
+; AVX-NEXT:    vblendps {{.*#+}} xmm0 = xmm1[0,1],xmm0[2,3]
+; AVX-NEXT:    retq
+  %shuf1 = shufflevector <4 x i32> %a, <4 x i32> zeroinitializer, <4 x i32><i32 4, i32 4, i32 2, i32 3>
+  %shuf2 = shufflevector <4 x i32> %b, <4 x i32> zeroinitializer, <4 x i32><i32 0, i32 1, i32 4, i32 4>
+  %or = or <4 x i32> %shuf1, %shuf2
+  ret <4 x i32> %or
+}
+
+
+define <2 x i64> @test3(<2 x i64> %a, <2 x i64> %b) {
+; SSE2-LABEL: test3:
+; SSE2:       # %bb.0:
+; SSE2-NEXT:    movsd {{.*#+}} xmm0 = xmm1[0],xmm0[1]
+; SSE2-NEXT:    retq
+;
+; SSE4-LABEL: test3:
+; SSE4:       # %bb.0:
+; SSE4-NEXT:    blendps {{.*#+}} xmm0 = xmm1[0,1],xmm0[2,3]
+; SSE4-NEXT:    retq
+;
+; AVX-LABEL: test3:
+; AVX:       # %bb.0:
+; AVX-NEXT:    vblendps {{.*#+}} xmm0 = xmm1[0,1],xmm0[2,3]
+; AVX-NEXT:    retq
+  %shuf1 = shufflevector <2 x i64> %a, <2 x i64> zeroinitializer, <2 x i32><i32 2, i32 1>
+  %shuf2 = shufflevector <2 x i64> %b, <2 x i64> zeroinitializer, <2 x i32><i32 0, i32 2>
+  %or = or <2 x i64> %shuf1, %shuf2
+  ret <2 x i64> %or
+}
+
+
+define <4 x i32> @test4(<4 x i32> %a, <4 x i32> %b) {
+; SSE2-LABEL: test4:
+; SSE2:       # %bb.0:
+; SSE2-NEXT:    movss {{.*#+}} xmm1 = xmm0[0],xmm1[1,2,3]
+; SSE2-NEXT:    movaps %xmm1, %xmm0
+; SSE2-NEXT:    retq
+;
+; SSE4-LABEL: test4:
+; SSE4:       # %bb.0:
+; SSE4-NEXT:    blendps {{.*#+}} xmm0 = xmm0[0],xmm1[1,2,3]
+; SSE4-NEXT:    retq
+;
+; AVX-LABEL: test4:
+; AVX:       # %bb.0:
+; AVX-NEXT:    vblendps {{.*#+}} xmm0 = xmm0[0],xmm1[1,2,3]
+; AVX-NEXT:    retq
+  %shuf1 = shufflevector <4 x i32> %a, <4 x i32> zeroinitializer, <4 x i32><i32 0, i32 4, i32 4, i32 4>
+  %shuf2 = shufflevector <4 x i32> %b, <4 x i32> zeroinitializer, <4 x i32><i32 4, i32 1, i32 2, i32 3>
+  %or = or <4 x i32> %shuf1, %shuf2
+  ret <4 x i32> %or
+}
+
+
+define <4 x i32> @test5(<4 x i32> %a, <4 x i32> %b) {
+; SSE2-LABEL: test5:
+; SSE2:       # %bb.0:
+; SSE2-NEXT:    movss {{.*#+}} xmm0 = xmm1[0],xmm0[1,2,3]
+; SSE2-NEXT:    retq
+;
+; SSE4-LABEL: test5:
+; SSE4:       # %bb.0:
+; SSE4-NEXT:    blendps {{.*#+}} xmm0 = xmm1[0],xmm0[1,2,3]
+; SSE4-NEXT:    retq
+;
+; AVX-LABEL: test5:
+; AVX:       # %bb.0:
+; AVX-NEXT:    vblendps {{.*#+}} xmm0 = xmm1[0],xmm0[1,2,3]
+; AVX-NEXT:    retq
+  %shuf1 = shufflevector <4 x i32> %a, <4 x i32> zeroinitializer, <4 x i32><i32 4, i32 1, i32 2, i32 3>
+  %shuf2 = shufflevector <4 x i32> %b, <4 x i32> zeroinitializer, <4 x i32><i32 0, i32 4, i32 4, i32 4>
+  %or = or <4 x i32> %shuf1, %shuf2
+  ret <4 x i32> %or
+}
+
+
+define <4 x i32> @test6(<4 x i32> %a, <4 x i32> %b) {
+; SSE2-LABEL: test6:
+; SSE2:       # %bb.0:
+; SSE2-NEXT:    shufps {{.*#+}} xmm0 = xmm0[0,1],xmm1[2,3]
+; SSE2-NEXT:    retq
+;
+; SSE4-LABEL: test6:
+; SSE4:       # %bb.0:
+; SSE4-NEXT:    blendps {{.*#+}} xmm0 = xmm0[0,1],xmm1[2,3]
+; SSE4-NEXT:    retq
+;
+; AVX-LABEL: test6:
+; AVX:       # %bb.0:
+; AVX-NEXT:    vblendps {{.*#+}} xmm0 = xmm0[0,1],xmm1[2,3]
+; AVX-NEXT:    retq
+  %shuf1 = shufflevector <4 x i32> %a, <4 x i32> zeroinitializer, <4 x i32><i32 0, i32 1, i32 4, i32 4>
+  %shuf2 = shufflevector <4 x i32> %b, <4 x i32> zeroinitializer, <4 x i32><i32 4, i32 4, i32 2, i32 3>
+  %or = or <4 x i32> %shuf1, %shuf2
+  ret <4 x i32> %or
+}
+
+
+define <4 x i32> @test7(<4 x i32> %a, <4 x i32> %b) {
+; SSE2-LABEL: test7:
+; SSE2:       # %bb.0:
+; SSE2-NEXT:    shufps {{.*#+}} xmm0 = xmm0[0,1],xmm1[2,3]
+; SSE2-NEXT:    retq
+;
+; SSE4-LABEL: test7:
+; SSE4:       # %bb.0:
+; SSE4-NEXT:    blendps {{.*#+}} xmm0 = xmm0[0,1],xmm1[2,3]
+; SSE4-NEXT:    retq
+;
+; AVX-LABEL: test7:
+; AVX:       # %bb.0:
+; AVX-NEXT:    vblendps {{.*#+}} xmm0 = xmm0[0,1],xmm1[2,3]
+; AVX-NEXT:    retq
+  %and1 = and <4 x i32> %a, <i32 -1, i32 -1, i32 0, i32 0>
+  %and2 = and <4 x i32> %b, <i32 0, i32 0, i32 -1, i32 -1>
+  %or = or <4 x i32> %and1, %and2
+  ret <4 x i32> %or
+}
+
+
+define <2 x i64> @test8(<2 x i64> %a, <2 x i64> %b) {
+; SSE2-LABEL: test8:
+; SSE2:       # %bb.0:
+; SSE2-NEXT:    shufps {{.*#+}} xmm0 = xmm0[0,1],xmm1[2,3]
+; SSE2-NEXT:    retq
+;
+; SSE4-LABEL: test8:
+; SSE4:       # %bb.0:
+; SSE4-NEXT:    blendps {{.*#+}} xmm0 = xmm0[0,1],xmm1[2,3]
+; SSE4-NEXT:    retq
+;
+; AVX-LABEL: test8:
+; AVX:       # %bb.0:
+; AVX-NEXT:    vblendps {{.*#+}} xmm0 = xmm0[0,1],xmm1[2,3]
+; AVX-NEXT:    retq
+  %and1 = and <2 x i64> %a, <i64 -1, i64 0>
+  %and2 = and <2 x i64> %b, <i64 0, i64 -1>
+  %or = or <2 x i64> %and1, %and2
+  ret <2 x i64> %or
+}
+
+
+define <4 x i32> @test9(<4 x i32> %a, <4 x i32> %b) {
+; SSE2-LABEL: test9:
+; SSE2:       # %bb.0:
+; SSE2-NEXT:    movsd {{.*#+}} xmm0 = xmm1[0],xmm0[1]
+; SSE2-NEXT:    retq
+;
+; SSE4-LABEL: test9:
+; SSE4:       # %bb.0:
+; SSE4-NEXT:    blendps {{.*#+}} xmm0 = xmm1[0,1],xmm0[2,3]
+; SSE4-NEXT:    retq
+;
+; AVX-LABEL: test9:
+; AVX:       # %bb.0:
+; AVX-NEXT:    vblendps {{.*#+}} xmm0 = xmm1[0,1],xmm0[2,3]
+; AVX-NEXT:    retq
+  %and1 = and <4 x i32> %a, <i32 0, i32 0, i32 -1, i32 -1>
+  %and2 = and <4 x i32> %b, <i32 -1, i32 -1, i32 0, i32 0>
+  %or = or <4 x i32> %and1, %and2
+  ret <4 x i32> %or
+}
+
+
+define <2 x i64> @test10(<2 x i64> %a, <2 x i64> %b) {
+; SSE2-LABEL: test10:
+; SSE2:       # %bb.0:
+; SSE2-NEXT:    movsd {{.*#+}} xmm0 = xmm1[0],xmm0[1]
+; SSE2-NEXT:    retq
+;
+; SSE4-LABEL: test10:
+; SSE4:       # %bb.0:
+; SSE4-NEXT:    blendps {{.*#+}} xmm0 = xmm1[0,1],xmm0[2,3]
+; SSE4-NEXT:    retq
+;
+; AVX-LABEL: test10:
+; AVX:       # %bb.0:
+; AVX-NEXT:    vblendps {{.*#+}} xmm0 = xmm1[0,1],xmm0[2,3]
+; AVX-NEXT:    retq
+  %and1 = and <2 x i64> %a, <i64 0, i64 -1>
+  %and2 = and <2 x i64> %b, <i64 -1, i64 0>
+  %or = or <2 x i64> %and1, %and2
+  ret <2 x i64> %or
+}
+
+
+define <4 x i32> @test11(<4 x i32> %a, <4 x i32> %b) {
+; SSE2-LABEL: test11:
+; SSE2:       # %bb.0:
+; SSE2-NEXT:    movss {{.*#+}} xmm1 = xmm0[0],xmm1[1,2,3]
+; SSE2-NEXT:    movaps %xmm1, %xmm0
+; SSE2-NEXT:    retq
+;
+; SSE4-LABEL: test11:
+; SSE4:       # %bb.0:
+; SSE4-NEXT:    blendps {{.*#+}} xmm0 = xmm0[0],xmm1[1,2,3]
+; SSE4-NEXT:    retq
+;
+; AVX-LABEL: test11:
+; AVX:       # %bb.0:
+; AVX-NEXT:    vblendps {{.*#+}} xmm0 = xmm0[0],xmm1[1,2,3]
+; AVX-NEXT:    retq
+  %and1 = and <4 x i32> %a, <i32 -1, i32 0, i32 0, i32 0>
+  %and2 = and <4 x i32> %b, <i32 0, i32 -1, i32 -1, i32 -1>
+  %or = or <4 x i32> %and1, %and2
+  ret <4 x i32> %or
+}
+
+
+define <4 x i32> @test12(<4 x i32> %a, <4 x i32> %b) {
+; SSE2-LABEL: test12:
+; SSE2:       # %bb.0:
+; SSE2-NEXT:    movss {{.*#+}} xmm0 = xmm1[0],xmm0[1,2,3]
+; SSE2-NEXT:    retq
+;
+; SSE4-LABEL: test12:
+; SSE4:       # %bb.0:
+; SSE4-NEXT:    blendps {{.*#+}} xmm0 = xmm1[0],xmm0[1,2,3]
+; SSE4-NEXT:    retq
+;
+; AVX-LABEL: test12:
+; AVX:       # %bb.0:
+; AVX-NEXT:    vblendps {{.*#+}} xmm0 = xmm1[0],xmm0[1,2,3]
+; AVX-NEXT:    retq
+  %and1 = and <4 x i32> %a, <i32 0, i32 -1, i32 -1, i32 -1>
+  %and2 = and <4 x i32> %b, <i32 -1, i32 0, i32 0, i32 0>
+  %or = or <4 x i32> %and1, %and2
+  ret <4 x i32> %or
+}
+
+
+; Verify that the following test cases are folded into single shuffles.
+
+define <4 x i32> @test13(<4 x i32> %a, <4 x i32> %b) {
+; SSE-LABEL: test13:
+; SSE:       # %bb.0:
+; SSE-NEXT:    shufps {{.*#+}} xmm0 = xmm0[1,1],xmm1[2,3]
+; SSE-NEXT:    retq
+;
+; AVX-LABEL: test13:
+; AVX:       # %bb.0:
+; AVX-NEXT:    vshufps {{.*#+}} xmm0 = xmm0[1,1],xmm1[2,3]
+; AVX-NEXT:    retq
+  %shuf1 = shufflevector <4 x i32> %a, <4 x i32> zeroinitializer, <4 x i32><i32 1, i32 1, i32 4, i32 4>
+  %shuf2 = shufflevector <4 x i32> %b, <4 x i32> zeroinitializer, <4 x i32><i32 4, i32 4, i32 2, i32 3>
+  %or = or <4 x i32> %shuf1, %shuf2
+  ret <4 x i32> %or
+}
+
+
+define <2 x i64> @test14(<2 x i64> %a, <2 x i64> %b) {
+; SSE-LABEL: test14:
+; SSE:       # %bb.0:
+; SSE-NEXT:    movlhps {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; SSE-NEXT:    retq
+;
+; AVX-LABEL: test14:
+; AVX:       # %bb.0:
+; AVX-NEXT:    vmovlhps {{.*#+}} xmm0 = xmm0[0],xmm1[0]
+; AVX-NEXT:    retq
+  %shuf1 = shufflevector <2 x i64> %a, <2 x i64> zeroinitializer, <2 x i32><i32 0, i32 2>
+  %shuf2 = shufflevector <2 x i64> %b, <2 x i64> zeroinitializer, <2 x i32><i32 2, i32 0>
+  %or = or <2 x i64> %shuf1, %shuf2
+  ret <2 x i64> %or
+}
+
+
+define <4 x i32> @test15(<4 x i32> %a, <4 x i32> %b) {
+; SSE-LABEL: test15:
+; SSE:       # %bb.0:
+; SSE-NEXT:    shufps {{.*#+}} xmm1 = xmm1[2,1],xmm0[2,1]
+; SSE-NEXT:    movaps %xmm1, %xmm0
+; SSE-NEXT:    retq
+;
+; AVX-LABEL: test15:
+; AVX:       # %bb.0:
+; AVX-NEXT:    vshufps {{.*#+}} xmm0 = xmm1[2,1],xmm0[2,1]
+; AVX-NEXT:    retq
+  %shuf1 = shufflevector <4 x i32> %a, <4 x i32> zeroinitializer, <4 x i32><i32 4, i32 4, i32 2, i32 1>
+  %shuf2 = shufflevector <4 x i32> %b, <4 x i32> zeroinitializer, <4 x i32><i32 2, i32 1, i32 4, i32 4>
+  %or = or <4 x i32> %shuf1, %shuf2
+  ret <4 x i32> %or
+}
+
+
+define <2 x i64> @test16(<2 x i64> %a, <2 x i64> %b) {
+; SSE-LABEL: test16:
+; SSE:       # %bb.0:
+; SSE-NEXT:    movlhps {{.*#+}} xmm1 = xmm1[0],xmm0[0]
+; SSE-NEXT:    movaps %xmm1, %xmm0
+; SSE-NEXT:    retq
+;
+; AVX-LABEL: test16:
+; AVX:       # %bb.0:
+; AVX-NEXT:    vmovlhps {{.*#+}} xmm0 = xmm1[0],xmm0[0]
+; AVX-NEXT:    retq
+  %shuf1 = shufflevector <2 x i64> %a, <2 x i64> zeroinitializer, <2 x i32><i32 2, i32 0>
+  %shuf2 = shufflevector <2 x i64> %b, <2 x i64> zeroinitializer, <2 x i32><i32 0, i32 2>
+  %or = or <2 x i64> %shuf1, %shuf2
+  ret <2 x i64> %or
+}
+
+
+; Verify that the dag-combiner does not fold a OR of two shuffles into a single
+; shuffle instruction when the shuffle indexes are not compatible.
+
+define <4 x i32> @test17(<4 x i32> %a, <4 x i32> %b) {
+; SSE-LABEL: test17:
+; SSE:       # %bb.0:
+; SSE-NEXT:    psllq $32, %xmm0
+; SSE-NEXT:    movq {{.*#+}} xmm1 = xmm1[0],zero
+; SSE-NEXT:    por %xmm1, %xmm0
+; SSE-NEXT:    retq
+;
+; AVX-LABEL: test17:
+; AVX:       # %bb.0:
+; AVX-NEXT:    vpsllq $32, %xmm0, %xmm0
+; AVX-NEXT:    vmovq {{.*#+}} xmm1 = xmm1[0],zero
+; AVX-NEXT:    vpor %xmm1, %xmm0, %xmm0
+; AVX-NEXT:    retq
+  %shuf1 = shufflevector <4 x i32> %a, <4 x i32> zeroinitializer, <4 x i32><i32 4, i32 0, i32 4, i32 2>
+  %shuf2 = shufflevector <4 x i32> %b, <4 x i32> zeroinitializer, <4 x i32><i32 0, i32 1, i32 4, i32 4>
+  %or = or <4 x i32> %shuf1, %shuf2
+  ret <4 x i32> %or
+}
+
+
+define <4 x i32> @test18(<4 x i32> %a, <4 x i32> %b) {
+; SSE2-LABEL: test18:
+; SSE2:       # %bb.0:
+; SSE2-NEXT:    xorps %xmm2, %xmm2
+; SSE2-NEXT:    xorps %xmm3, %xmm3
+; SSE2-NEXT:    movss {{.*#+}} xmm3 = xmm0[0],xmm3[1,2,3]
+; SSE2-NEXT:    pshufd {{.*#+}} xmm0 = xmm3[1,0,1,1]
+; SSE2-NEXT:    movss {{.*#+}} xmm2 = xmm1[0],xmm2[1,2,3]
+; SSE2-NEXT:    orps %xmm0, %xmm2
+; SSE2-NEXT:    movaps %xmm2, %xmm0
+; SSE2-NEXT:    retq
+;
+; SSE4-LABEL: test18:
+; SSE4:       # %bb.0:
+; SSE4-NEXT:    pxor %xmm2, %xmm2
+; SSE4-NEXT:    pblendw {{.*#+}} xmm0 = xmm0[0,1],xmm2[2,3,4,5,6,7]
+; SSE4-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[1,0,1,1]
+; SSE4-NEXT:    pblendw {{.*#+}} xmm2 = xmm1[0,1],xmm2[2,3,4,5,6,7]
+; SSE4-NEXT:    por %xmm0, %xmm2
+; SSE4-NEXT:    movdqa %xmm2, %xmm0
+; SSE4-NEXT:    retq
+;
+; AVX1-LABEL: test18:
+; AVX1:       # %bb.0:
+; AVX1-NEXT:    vxorps %xmm2, %xmm2, %xmm2
+; AVX1-NEXT:    vblendps {{.*#+}} xmm0 = xmm0[0],xmm2[1,2,3]
+; AVX1-NEXT:    vshufps {{.*#+}} xmm0 = xmm0[1,0,1,1]
+; AVX1-NEXT:    vblendps {{.*#+}} xmm1 = xmm1[0],xmm2[1,2,3]
+; AVX1-NEXT:    vorps %xmm1, %xmm0, %xmm0
+; AVX1-NEXT:    retq
+;
+; AVX2-LABEL: test18:
+; AVX2:       # %bb.0:
+; AVX2-NEXT:    vxorps %xmm2, %xmm2, %xmm2
+; AVX2-NEXT:    vblendps {{.*#+}} xmm0 = xmm0[0],xmm2[1,2,3]
+; AVX2-NEXT:    vshufps {{.*#+}} xmm0 = xmm0[1,0,1,1]
+; AVX2-NEXT:    vblendps {{.*#+}} xmm1 = xmm1[0],xmm2[1,2,3]
+; AVX2-NEXT:    vorps %xmm1, %xmm0, %xmm0
+; AVX2-NEXT:    retq
+;
+; AVX512-LABEL: test18:
+; AVX512:       # %bb.0:
+; AVX512-NEXT:    vpshufb {{.*#+}} xmm0 = zero,zero,zero,zero,xmm0[0,1,2,3],zero,zero,zero,zero,zero,zero,zero,zero
+; AVX512-NEXT:    vpxor %xmm2, %xmm2, %xmm2
+; AVX512-NEXT:    vpblendw {{.*#+}} xmm1 = xmm1[0,1],xmm2[2,3,4,5,6,7]
+; AVX512-NEXT:    vpor %xmm1, %xmm0, %xmm0
+; AVX512-NEXT:    retq
+  %shuf1 = shufflevector <4 x i32> %a, <4 x i32> zeroinitializer, <4 x i32><i32 4, i32 0, i32 4, i32 4>
+  %shuf2 = shufflevector <4 x i32> %b, <4 x i32> zeroinitializer, <4 x i32><i32 0, i32 4, i32 4, i32 4>
+  %or = or <4 x i32> %shuf1, %shuf2
+  ret <4 x i32> %or
+}
+
+
+define <4 x i32> @test19(<4 x i32> %a, <4 x i32> %b) {
+; SSE2-LABEL: test19:
+; SSE2:       # %bb.0:
+; SSE2-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[0,3,2,3]
+; SSE2-NEXT:    pxor %xmm2, %xmm2
+; SSE2-NEXT:    punpckldq {{.*#+}} xmm2 = xmm2[0],xmm0[0],xmm2[1],xmm0[1]
+; SSE2-NEXT:    movdqa %xmm1, %xmm0
+; SSE2-NEXT:    pslldq {{.*#+}} xmm0 = zero,zero,zero,zero,zero,zero,zero,zero,xmm0[0,1,2,3,4,5,6,7]
+; SSE2-NEXT:    shufps {{.*#+}} xmm0 = xmm0[2,0],xmm1[2,2]
+; SSE2-NEXT:    orps %xmm2, %xmm0
+; SSE2-NEXT:    retq
+;
+; SSE4-LABEL: test19:
+; SSE4:       # %bb.0:
+; SSE4-NEXT:    pshufd {{.*#+}} xmm2 = xmm0[0,0,2,3]
+; SSE4-NEXT:    pxor %xmm3, %xmm3
+; SSE4-NEXT:    pblendw {{.*#+}} xmm2 = xmm3[0,1],xmm2[2,3],xmm3[4,5],xmm2[6,7]
+; SSE4-NEXT:    pshufd {{.*#+}} xmm0 = xmm1[0,1,2,2]
+; SSE4-NEXT:    pblendw {{.*#+}} xmm0 = xmm0[0,1],xmm3[2,3],xmm0[4,5,6,7]
+; SSE4-NEXT:    por %xmm2, %xmm0
+; SSE4-NEXT:    retq
+;
+; AVX1-LABEL: test19:
+; AVX1:       # %bb.0:
+; AVX1-NEXT:    vshufps {{.*#+}} xmm0 = xmm0[0,0,2,3]
+; AVX1-NEXT:    vxorps %xmm2, %xmm2, %xmm2
+; AVX1-NEXT:    vblendps {{.*#+}} xmm0 = xmm2[0],xmm0[1],xmm2[2],xmm0[3]
+; AVX1-NEXT:    vshufps {{.*#+}} xmm1 = xmm1[0,1,2,2]
+; AVX1-NEXT:    vblendps {{.*#+}} xmm1 = xmm1[0],xmm2[1],xmm1[2,3]
+; AVX1-NEXT:    vorps %xmm1, %xmm0, %xmm0
+; AVX1-NEXT:    retq
+;
+; AVX2-LABEL: test19:
+; AVX2:       # %bb.0:
+; AVX2-NEXT:    vshufps {{.*#+}} xmm0 = xmm0[0,0,2,3]
+; AVX2-NEXT:    vxorps %xmm2, %xmm2, %xmm2
+; AVX2-NEXT:    vblendps {{.*#+}} xmm0 = xmm2[0],xmm0[1],xmm2[2],xmm0[3]
+; AVX2-NEXT:    vshufps {{.*#+}} xmm1 = xmm1[0,1,2,2]
+; AVX2-NEXT:    vblendps {{.*#+}} xmm1 = xmm1[0],xmm2[1],xmm1[2,3]
+; AVX2-NEXT:    vorps %xmm1, %xmm0, %xmm0
+; AVX2-NEXT:    retq
+;
+; AVX512-LABEL: test19:
+; AVX512:       # %bb.0:
+; AVX512-NEXT:    vpshufb {{.*#+}} xmm0 = zero,zero,zero,zero,xmm0[0,1,2,3],zero,zero,zero,zero,xmm0[12,13,14,15]
+; AVX512-NEXT:    vpshufb {{.*#+}} xmm1 = xmm1[0,1,2,3],zero,zero,zero,zero,xmm1[8,9,10,11,8,9,10,11]
+; AVX512-NEXT:    vpor %xmm1, %xmm0, %xmm0
+; AVX512-NEXT:    retq
+  %shuf1 = shufflevector <4 x i32> %a, <4 x i32> zeroinitializer, <4 x i32><i32 4, i32 0, i32 4, i32 3>
+  %shuf2 = shufflevector <4 x i32> %b, <4 x i32> zeroinitializer, <4 x i32><i32 0, i32 4, i32 2, i32 2>
+  %or = or <4 x i32> %shuf1, %shuf2
+  ret <4 x i32> %or
+}
+
+
+define <2 x i64> @test20(<2 x i64> %a, <2 x i64> %b) {
+; SSE-LABEL: test20:
+; SSE:       # %bb.0:
+; SSE-NEXT:    por %xmm1, %xmm0
+; SSE-NEXT:    movq {{.*#+}} xmm0 = xmm0[0],zero
+; SSE-NEXT:    retq
+;
+; AVX-LABEL: test20:
+; AVX:       # %bb.0:
+; AVX-NEXT:    vpor %xmm1, %xmm0, %xmm0
+; AVX-NEXT:    vmovq {{.*#+}} xmm0 = xmm0[0],zero
+; AVX-NEXT:    retq
+  %shuf1 = shufflevector <2 x i64> %a, <2 x i64> zeroinitializer, <2 x i32><i32 0, i32 2>
+  %shuf2 = shufflevector <2 x i64> %b, <2 x i64> zeroinitializer, <2 x i32><i32 0, i32 2>
+  %or = or <2 x i64> %shuf1, %shuf2
+  ret <2 x i64> %or
+}
+
+
+define <2 x i64> @test21(<2 x i64> %a, <2 x i64> %b) {
+; SSE-LABEL: test21:
+; SSE:       # %bb.0:
+; SSE-NEXT:    por %xmm1, %xmm0
+; SSE-NEXT:    pslldq {{.*#+}} xmm0 = zero,zero,zero,zero,zero,zero,zero,zero,xmm0[0,1,2,3,4,5,6,7]
+; SSE-NEXT:    retq
+;
+; AVX1-LABEL: test21:
+; AVX1:       # %bb.0:
+; AVX1-NEXT:    vpor %xmm1, %xmm0, %xmm0
+; AVX1-NEXT:    vpslldq {{.*#+}} xmm0 = zero,zero,zero,zero,zero,zero,zero,zero,xmm0[0,1,2,3,4,5,6,7]
+; AVX1-NEXT:    retq
+;
+; AVX2-LABEL: test21:
+; AVX2:       # %bb.0:
+; AVX2-NEXT:    vpor %xmm1, %xmm0, %xmm0
+; AVX2-NEXT:    vpslldq {{.*#+}} xmm0 = zero,zero,zero,zero,zero,zero,zero,zero,xmm0[0,1,2,3,4,5,6,7]
+; AVX2-NEXT:    retq
+;
+; AVX512-LABEL: test21:
+; AVX512:       # %bb.0:
+; AVX512-NEXT:    vorpd %xmm1, %xmm0, %xmm0
+; AVX512-NEXT:    vpslldq {{.*#+}} xmm0 = zero,zero,zero,zero,zero,zero,zero,zero,xmm0[0,1,2,3,4,5,6,7]
+; AVX512-NEXT:    retq
+  %shuf1 = shufflevector <2 x i64> %a, <2 x i64> zeroinitializer, <2 x i32><i32 2, i32 0>
+  %shuf2 = shufflevector <2 x i64> %b, <2 x i64> zeroinitializer, <2 x i32><i32 2, i32 0>
+  %or = or <2 x i64> %shuf1, %shuf2
+  ret <2 x i64> %or
+}
+
+
+; Verify that the dag-combiner keeps the correct domain for float/double vectors
+; bitcast to use the mask-or blend combine.
+
+define <2 x double> @test22(<2 x double> %a0, <2 x double> %a1) {
+; SSE2-LABEL: test22:
+; SSE2:       # %bb.0:
+; SSE2-NEXT:    movsd {{.*#+}} xmm0 = xmm1[0],xmm0[1]
+; SSE2-NEXT:    retq
+;
+; SSE4-LABEL: test22:
+; SSE4:       # %bb.0:
+; SSE4-NEXT:    blendps {{.*#+}} xmm0 = xmm1[0,1],xmm0[2,3]
+; SSE4-NEXT:    retq
+;
+; AVX-LABEL: test22:
+; AVX:       # %bb.0:
+; AVX-NEXT:    vblendps {{.*#+}} xmm0 = xmm1[0,1],xmm0[2,3]
+; AVX-NEXT:    retq
+  %bc1 = bitcast <2 x double> %a0 to <2 x i64>
+  %bc2 = bitcast <2 x double> %a1 to <2 x i64>
+  %and1 = and <2 x i64> %bc1, <i64 0, i64 -1>
+  %and2 = and <2 x i64> %bc2, <i64 -1, i64 0>
+  %or = or <2 x i64> %and1, %and2
+  %bc3 = bitcast <2 x i64> %or to <2 x double>
+  ret <2 x double> %bc3
+}
+
+
+define <4 x float> @test23(<4 x float> %a0, <4 x float> %a1) {
+; SSE2-LABEL: test23:
+; SSE2:       # %bb.0:
+; SSE2-NEXT:    shufps {{.*#+}} xmm0 = xmm0[1,2],xmm1[0,3]
+; SSE2-NEXT:    shufps {{.*#+}} xmm0 = xmm0[2,0,1,3]
+; SSE2-NEXT:    retq
+;
+; SSE4-LABEL: test23:
+; SSE4:       # %bb.0:
+; SSE4-NEXT:    blendps {{.*#+}} xmm0 = xmm1[0],xmm0[1,2],xmm1[3]
+; SSE4-NEXT:    retq
+;
+; AVX-LABEL: test23:
+; AVX:       # %bb.0:
+; AVX-NEXT:    vblendps {{.*#+}} xmm0 = xmm1[0],xmm0[1,2],xmm1[3]
+; AVX-NEXT:    retq
+  %bc1 = bitcast <4 x float> %a0 to <4 x i32>
+  %bc2 = bitcast <4 x float> %a1 to <4 x i32>
+  %and1 = and <4 x i32> %bc1, <i32 0, i32 -1, i32 -1, i32 0>
+  %and2 = and <4 x i32> %bc2, <i32 -1, i32 0, i32 0, i32 -1>
+  %or = or <4 x i32> %and1, %and2
+  %bc3 = bitcast <4 x i32> %or to <4 x float>
+  ret <4 x float> %bc3
+}
+
+
+define <4 x float> @test24(<4 x float> %a0, <4 x float> %a1) {
+; SSE2-LABEL: test24:
+; SSE2:       # %bb.0:
+; SSE2-NEXT:    movsd {{.*#+}} xmm0 = xmm1[0],xmm0[1]
+; SSE2-NEXT:    retq
+;
+; SSE4-LABEL: test24:
+; SSE4:       # %bb.0:
+; SSE4-NEXT:    blendps {{.*#+}} xmm0 = xmm1[0,1],xmm0[2,3]
+; SSE4-NEXT:    retq
+;
+; AVX-LABEL: test24:
+; AVX:       # %bb.0:
+; AVX-NEXT:    vblendps {{.*#+}} xmm0 = xmm1[0,1],xmm0[2,3]
+; AVX-NEXT:    retq
+  %bc1 = bitcast <4 x float> %a0 to <2 x i64>
+  %bc2 = bitcast <4 x float> %a1 to <2 x i64>
+  %and1 = and <2 x i64> %bc1, <i64 0, i64 -1>
+  %and2 = and <2 x i64> %bc2, <i64 -1, i64 0>
+  %or = or <2 x i64> %and1, %and2
+  %bc3 = bitcast <2 x i64> %or to <4 x float>
+  ret <4 x float> %bc3
+}
+
+
+define <4 x float> @test25(<4 x float> %a0) {
+; SSE2-LABEL: test25:
+; SSE2:       # %bb.0:
+; SSE2-NEXT:    shufps {{.*#+}} xmm0 = xmm0[1,2],mem[0,3]
+; SSE2-NEXT:    shufps {{.*#+}} xmm0 = xmm0[2,0,1,3]
+; SSE2-NEXT:    retq
+;
+; SSE4-LABEL: test25:
+; SSE4:       # %bb.0:
+; SSE4-NEXT:    blendps {{.*#+}} xmm0 = mem[0],xmm0[1,2],mem[3]
+; SSE4-NEXT:    retq
+;
+; AVX1-LABEL: test25:
+; AVX1:       # %bb.0:
+; AVX1-NEXT:    vblendps {{.*#+}} xmm0 = mem[0],xmm0[1,2],mem[3]
+; AVX1-NEXT:    retq
+;
+; AVX2-LABEL: test25:
+; AVX2:       # %bb.0:
+; AVX2-NEXT:    vbroadcastss {{.*#+}} xmm1 = [1.0E+0,1.0E+0,1.0E+0,1.0E+0]
+; AVX2-NEXT:    vblendps {{.*#+}} xmm0 = xmm1[0],xmm0[1,2],xmm1[3]
+; AVX2-NEXT:    retq
+;
+; AVX512-LABEL: test25:
+; AVX512:       # %bb.0:
+; AVX512-NEXT:    vbroadcastss {{.*#+}} xmm1 = [1.0E+0,1.0E+0,1.0E+0,1.0E+0]
+; AVX512-NEXT:    vblendps {{.*#+}} xmm0 = xmm1[0],xmm0[1,2],xmm1[3]
+; AVX512-NEXT:    retq
+  %bc1 = bitcast <4 x float> %a0 to <4 x i32>
+  %bc2 = bitcast <4 x float> <float 1.0, float 1.0, float 1.0, float 1.0> to <4 x i32>
+  %and1 = and <4 x i32> %bc1, <i32 0, i32 -1, i32 -1, i32 0>
+  %and2 = and <4 x i32> %bc2, <i32 -1, i32 0, i32 0, i32 -1>
+  %or = or <4 x i32> %and1, %and2
+  %bc3 = bitcast <4 x i32> %or to <4 x float>
+  ret <4 x float> %bc3
+}
+
+
+; Verify that the DAGCombiner doesn't crash in the attempt to check if a shuffle
+; with illegal type has a legal mask. Method 'isShuffleMaskLegal' only knows how to
+; handle legal vector value types.
+define <4 x i8> @test_crash(<4 x i8> %a, <4 x i8> %b) {
+; SSE2-LABEL: test_crash:
+; SSE2:       # %bb.0:
+; SSE2-NEXT:    movaps {{.*#+}} xmm2 = [65535,0,65535,65535,65535,65535,65535,65535]
+; SSE2-NEXT:    andps %xmm2, %xmm1
+; SSE2-NEXT:    andnps %xmm0, %xmm2
+; SSE2-NEXT:    orps %xmm1, %xmm2
+; SSE2-NEXT:    movaps %xmm2, %xmm0
+; SSE2-NEXT:    retq
+;
+; SSE4-LABEL: test_crash:
+; SSE4:       # %bb.0:
+; SSE4-NEXT:    pblendw {{.*#+}} xmm0 = xmm1[0],xmm0[1],xmm1[2,3,4,5,6,7]
+; SSE4-NEXT:    retq
+;
+; AVX-LABEL: test_crash:
+; AVX:       # %bb.0:
+; AVX-NEXT:    vpblendw {{.*#+}} xmm0 = xmm1[0],xmm0[1],xmm1[2,3,4,5,6,7]
+; AVX-NEXT:    retq
+  %shuf1 = shufflevector <4 x i8> %a, <4 x i8> zeroinitializer, <4 x i32><i32 4, i32 4, i32 2, i32 3>
+  %shuf2 = shufflevector <4 x i8> %b, <4 x i8> zeroinitializer, <4 x i32><i32 0, i32 1, i32 4, i32 4>
+  %or = or <4 x i8> %shuf1, %shuf2
+  ret <4 x i8> %or
+}
+
+; Verify that we can fold regardless of which operand is the zeroinitializer
+
+define <4 x i32> @test2b(<4 x i32> %a, <4 x i32> %b) {
+; SSE2-LABEL: test2b:
+; SSE2:       # %bb.0:
+; SSE2-NEXT:    movsd {{.*#+}} xmm0 = xmm1[0],xmm0[1]
+; SSE2-NEXT:    retq
+;
+; SSE4-LABEL: test2b:
+; SSE4:       # %bb.0:
+; SSE4-NEXT:    blendps {{.*#+}} xmm0 = xmm1[0,1],xmm0[2,3]
+; SSE4-NEXT:    retq
+;
+; AVX-LABEL: test2b:
+; AVX:       # %bb.0:
+; AVX-NEXT:    vblendps {{.*#+}} xmm0 = xmm1[0,1],xmm0[2,3]
+; AVX-NEXT:    retq
+  %shuf1 = shufflevector <4 x i32> zeroinitializer, <4 x i32> %a, <4 x i32><i32 0, i32 0, i32 6, i32 7>
+  %shuf2 = shufflevector <4 x i32> %b, <4 x i32> zeroinitializer, <4 x i32><i32 0, i32 1, i32 4, i32 4>
+  %or = or <4 x i32> %shuf1, %shuf2
+  ret <4 x i32> %or
+}
+
+define <4 x i32> @test2c(<4 x i32> %a, <4 x i32> %b) {
+; SSE2-LABEL: test2c:
+; SSE2:       # %bb.0:
+; SSE2-NEXT:    movsd {{.*#+}} xmm0 = xmm1[0],xmm0[1]
+; SSE2-NEXT:    retq
+;
+; SSE4-LABEL: test2c:
+; SSE4:       # %bb.0:
+; SSE4-NEXT:    blendps {{.*#+}} xmm0 = xmm1[0,1],xmm0[2,3]
+; SSE4-NEXT:    retq
+;
+; AVX-LABEL: test2c:
+; AVX:       # %bb.0:
+; AVX-NEXT:    vblendps {{.*#+}} xmm0 = xmm1[0,1],xmm0[2,3]
+; AVX-NEXT:    retq
+  %shuf1 = shufflevector <4 x i32> zeroinitializer, <4 x i32> %a, <4 x i32><i32 0, i32 0, i32 6, i32 7>
+  %shuf2 = shufflevector <4 x i32> zeroinitializer, <4 x i32> %b, <4 x i32><i32 4, i32 5, i32 0, i32 0>
+  %or = or <4 x i32> %shuf1, %shuf2
+  ret <4 x i32> %or
+}
+
+
+define <4 x i32> @test2d(<4 x i32> %a, <4 x i32> %b) {
+; SSE2-LABEL: test2d:
+; SSE2:       # %bb.0:
+; SSE2-NEXT:    movsd {{.*#+}} xmm0 = xmm1[0],xmm0[1]
+; SSE2-NEXT:    retq
+;
+; SSE4-LABEL: test2d:
+; SSE4:       # %bb.0:
+; SSE4-NEXT:    blendps {{.*#+}} xmm0 = xmm1[0,1],xmm0[2,3]
+; SSE4-NEXT:    retq
+;
+; AVX-LABEL: test2d:
+; AVX:       # %bb.0:
+; AVX-NEXT:    vblendps {{.*#+}} xmm0 = xmm1[0,1],xmm0[2,3]
+; AVX-NEXT:    retq
+  %shuf1 = shufflevector <4 x i32> %a, <4 x i32> zeroinitializer, <4 x i32><i32 4, i32 4, i32 2, i32 3>
+  %shuf2 = shufflevector <4 x i32> zeroinitializer, <4 x i32> %b, <4 x i32><i32 4, i32 5, i32 0, i32 0>
+  %or = or <4 x i32> %shuf1, %shuf2
+  ret <4 x i32> %or
+}
+
+; Make sure we can have an undef where an index pointing to the zero vector should be
+
+define <4 x i32> @test2e(<4 x i32> %a, <4 x i32> %b) {
+; SSE2-LABEL: test2e:
+; SSE2:       # %bb.0:
+; SSE2-NEXT:    movsd {{.*#+}} xmm0 = xmm1[0],xmm0[1]
+; SSE2-NEXT:    retq
+;
+; SSE4-LABEL: test2e:
+; SSE4:       # %bb.0:
+; SSE4-NEXT:    blendps {{.*#+}} xmm0 = xmm1[0,1],xmm0[2,3]
+; SSE4-NEXT:    retq
+;
+; AVX-LABEL: test2e:
+; AVX:       # %bb.0:
+; AVX-NEXT:    vblendps {{.*#+}} xmm0 = xmm1[0,1],xmm0[2,3]
+; AVX-NEXT:    retq
+  %shuf1 = shufflevector <4 x i32> %a, <4 x i32> <i32 0, i32 undef, i32 undef, i32 undef>, <4 x i32><i32 undef, i32 4, i32 2, i32 3>
+  %shuf2 = shufflevector <4 x i32> %b, <4 x i32> <i32 0, i32 undef, i32 undef, i32 undef>, <4 x i32><i32 0, i32 1, i32 4, i32 4>
+  %or = or <4 x i32> %shuf1, %shuf2
+  ret <4 x i32> %or
+}
+
+define <4 x i32> @test2f(<4 x i32> %a, <4 x i32> %b) {
+; SSE2-LABEL: test2f:
+; SSE2:       # %bb.0:
+; SSE2-NEXT:    movsd {{.*#+}} xmm0 = xmm1[0],xmm0[1]
+; SSE2-NEXT:    retq
+;
+; SSE4-LABEL: test2f:
+; SSE4:       # %bb.0:
+; SSE4-NEXT:    blendps {{.*#+}} xmm0 = xmm1[0,1],xmm0[2,3]
+; SSE4-NEXT:    retq
+;
+; AVX-LABEL: test2f:
+; AVX:       # %bb.0:
+; AVX-NEXT:    vblendps {{.*#+}} xmm0 = xmm1[0,1],xmm0[2,3]
+; AVX-NEXT:    retq
+  %shuf1 = shufflevector <4 x i32> %a, <4 x i32> <i32 0, i32 undef, i32 undef, i32 undef>, <4 x i32><i32 4, i32 4, i32 2, i32 3>
+  %shuf2 = shufflevector <4 x i32> %b, <4 x i32> <i32 0, i32 undef, i32 undef, i32 undef>, <4 x i32><i32 undef, i32 1, i32 4, i32 4>
+  %or = or <4 x i32> %shuf1, %shuf2
+  ret <4 x i32> %or
+}
+
+; (or (and X, c1), c2) -> (and (or X, c2), c1|c2) iff (c1 & c2) != 0
+
+define <2 x i64> @or_and_v2i64(<2 x i64> %a0) {
+; SSE-LABEL: or_and_v2i64:
+; SSE:       # %bb.0:
+; SSE-NEXT:    orps {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0
+; SSE-NEXT:    andps {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0
+; SSE-NEXT:    retq
+;
+; AVX1-LABEL: or_and_v2i64:
+; AVX1:       # %bb.0:
+; AVX1-NEXT:    vorps {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0, %xmm0
+; AVX1-NEXT:    vandps {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0, %xmm0
+; AVX1-NEXT:    retq
+;
+; AVX2-LABEL: or_and_v2i64:
+; AVX2:       # %bb.0:
+; AVX2-NEXT:    vorps {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0, %xmm0
+; AVX2-NEXT:    vandps {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0, %xmm0
+; AVX2-NEXT:    retq
+;
+; AVX512-LABEL: or_and_v2i64:
+; AVX512:       # %bb.0:
+; AVX512-NEXT:    vpbroadcastq {{.*#+}} xmm1 = [7,7]
+; AVX512-NEXT:    vpternlogq $200, {{\.?LCPI[0-9]+_[0-9]+}}(%rip){1to2}, %xmm1, %xmm0
+; AVX512-NEXT:    retq
+  %1 = and <2 x i64> %a0, <i64 7, i64 7>
+  %2 = or <2 x i64> %1, <i64 3, i64 3>
+  ret <2 x i64> %2
+}
+
+define <4 x i32> @or_and_v4i32(<4 x i32> %a0) {
+; SSE-LABEL: or_and_v4i32:
+; SSE:       # %bb.0:
+; SSE-NEXT:    orps {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0
+; SSE-NEXT:    andps {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0
+; SSE-NEXT:    retq
+;
+; AVX1-LABEL: or_and_v4i32:
+; AVX1:       # %bb.0:
+; AVX1-NEXT:    vorps {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0, %xmm0
+; AVX1-NEXT:    vandps {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0, %xmm0
+; AVX1-NEXT:    retq
+;
+; AVX2-LABEL: or_and_v4i32:
+; AVX2:       # %bb.0:
+; AVX2-NEXT:    vorps {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0, %xmm0
+; AVX2-NEXT:    vandps {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0, %xmm0
+; AVX2-NEXT:    retq
+;
+; AVX512-LABEL: or_and_v4i32:
+; AVX512:       # %bb.0:
+; AVX512-NEXT:    vpmovsxbd {{.*#+}} xmm1 = [3,3,15,7]
+; AVX512-NEXT:    vpternlogd $200, {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm1, %xmm0
+; AVX512-NEXT:    retq
+  %1 = and <4 x i32> %a0, <i32 1, i32 3, i32 5, i32 7>
+  %2 = or <4 x i32> %1, <i32 3, i32 2, i32 15, i32 2>
+  ret <4 x i32> %2
+}
+
+; If all masked bits are going to be set, that's a constant fold.
+
+define <4 x i32> @or_and_v4i32_fold(<4 x i32> %a0) {
+; SSE-LABEL: or_and_v4i32_fold:
+; SSE:       # %bb.0:
+; SSE-NEXT:    movaps {{.*#+}} xmm0 = [3,3,3,3]
+; SSE-NEXT:    retq
+;
+; AVX-LABEL: or_and_v4i32_fold:
+; AVX:       # %bb.0:
+; AVX-NEXT:    vbroadcastss {{.*#+}} xmm0 = [3,3,3,3]
+; AVX-NEXT:    retq
+  %1 = and <4 x i32> %a0, <i32 1, i32 1, i32 1, i32 1>
+  %2 = or <4 x i32> %1, <i32 3, i32 3, i32 3, i32 3>
+  ret <4 x i32> %2
+}
diff --git a/llvm/test/CodeGen/X86/combine-or.ll b/llvm/test/CodeGen/X86/combine-or.ll
index 5188de543f72..3b2102f46a29 100644
--- a/llvm/test/CodeGen/X86/combine-or.ll
+++ b/llvm/test/CodeGen/X86/combine-or.ll
@@ -1,6 +1,8 @@
 ; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
-; RUN: llc < %s -mtriple=x86_64-unknown-linux-gnu -mcpu=corei7 | FileCheck %s -check-prefixes=CHECK,CHECK-LV
-; RUN: llc < %s -mtriple=x86_64-unknown-linux-gnu -mcpu=corei7 -early-live-intervals | FileCheck %s -check-prefixes=CHECK,CHECK-LIS
+; RUN: llc < %s -mtriple=x86_64-unknown-linux-gnu -mcpu=corei7 | FileCheck %s -check-prefixes=CHECK,SSE
+; RUN: llc < %s -mtriple=x86_64-unknown-linux-gnu -mcpu=corei7 -early-live-intervals | FileCheck %s -check-prefixes=CHECK,SSE
+; RUN: llc < %s -mtriple=x86_64-unknown-linux-gnu -mcpu=corei7-avx | FileCheck %s -check-prefixes=CHECK,AVX,AVX1
+; RUN: llc < %s -mtriple=x86_64-unknown-linux-gnu -mcpu=x86-64-v3 | FileCheck %s -check-prefixes=CHECK,AVX,AVX2
 
 define i32 @or_self(i32 %x) {
 ; CHECK-LABEL: or_self:
@@ -19,472 +21,34 @@ define <4 x i32> @or_self_vec(<4 x i32> %x) {
   ret <4 x i32> %or
 }
 
-; Verify that each of the following test cases is folded into a single
-; instruction which performs a blend operation.
-
-define <2 x i64> @test1(<2 x i64> %a, <2 x i64> %b) {
-; CHECK-LABEL: test1:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    blendps {{.*#+}} xmm0 = xmm0[0,1],xmm1[2,3]
-; CHECK-NEXT:    retq
-  %shuf1 = shufflevector <2 x i64> %a, <2 x i64> zeroinitializer, <2 x i32><i32 0, i32 2>
-  %shuf2 = shufflevector <2 x i64> %b, <2 x i64> zeroinitializer, <2 x i32><i32 2, i32 1>
-  %or = or <2 x i64> %shuf1, %shuf2
-  ret <2 x i64> %or
-}
-
-
-define <4 x i32> @test2(<4 x i32> %a, <4 x i32> %b) {
-; CHECK-LABEL: test2:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    blendps {{.*#+}} xmm0 = xmm1[0,1],xmm0[2,3]
-; CHECK-NEXT:    retq
-  %shuf1 = shufflevector <4 x i32> %a, <4 x i32> zeroinitializer, <4 x i32><i32 4, i32 4, i32 2, i32 3>
-  %shuf2 = shufflevector <4 x i32> %b, <4 x i32> zeroinitializer, <4 x i32><i32 0, i32 1, i32 4, i32 4>
-  %or = or <4 x i32> %shuf1, %shuf2
-  ret <4 x i32> %or
-}
-
-
-define <2 x i64> @test3(<2 x i64> %a, <2 x i64> %b) {
-; CHECK-LABEL: test3:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    blendps {{.*#+}} xmm0 = xmm1[0,1],xmm0[2,3]
-; CHECK-NEXT:    retq
-  %shuf1 = shufflevector <2 x i64> %a, <2 x i64> zeroinitializer, <2 x i32><i32 2, i32 1>
-  %shuf2 = shufflevector <2 x i64> %b, <2 x i64> zeroinitializer, <2 x i32><i32 0, i32 2>
-  %or = or <2 x i64> %shuf1, %shuf2
-  ret <2 x i64> %or
-}
-
-
-define <4 x i32> @test4(<4 x i32> %a, <4 x i32> %b) {
-; CHECK-LABEL: test4:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    blendps {{.*#+}} xmm0 = xmm0[0],xmm1[1,2,3]
-; CHECK-NEXT:    retq
-  %shuf1 = shufflevector <4 x i32> %a, <4 x i32> zeroinitializer, <4 x i32><i32 0, i32 4, i32 4, i32 4>
-  %shuf2 = shufflevector <4 x i32> %b, <4 x i32> zeroinitializer, <4 x i32><i32 4, i32 1, i32 2, i32 3>
-  %or = or <4 x i32> %shuf1, %shuf2
-  ret <4 x i32> %or
-}
-
-
-define <4 x i32> @test5(<4 x i32> %a, <4 x i32> %b) {
-; CHECK-LABEL: test5:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    blendps {{.*#+}} xmm0 = xmm1[0],xmm0[1,2,3]
-; CHECK-NEXT:    retq
-  %shuf1 = shufflevector <4 x i32> %a, <4 x i32> zeroinitializer, <4 x i32><i32 4, i32 1, i32 2, i32 3>
-  %shuf2 = shufflevector <4 x i32> %b, <4 x i32> zeroinitializer, <4 x i32><i32 0, i32 4, i32 4, i32 4>
-  %or = or <4 x i32> %shuf1, %shuf2
-  ret <4 x i32> %or
-}
-
-
-define <4 x i32> @test6(<4 x i32> %a, <4 x i32> %b) {
-; CHECK-LABEL: test6:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    blendps {{.*#+}} xmm0 = xmm0[0,1],xmm1[2,3]
-; CHECK-NEXT:    retq
-  %shuf1 = shufflevector <4 x i32> %a, <4 x i32> zeroinitializer, <4 x i32><i32 0, i32 1, i32 4, i32 4>
-  %shuf2 = shufflevector <4 x i32> %b, <4 x i32> zeroinitializer, <4 x i32><i32 4, i32 4, i32 2, i32 3>
-  %or = or <4 x i32> %shuf1, %shuf2
-  ret <4 x i32> %or
-}
-
-
-define <4 x i32> @test7(<4 x i32> %a, <4 x i32> %b) {
-; CHECK-LABEL: test7:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    blendps {{.*#+}} xmm0 = xmm0[0,1],xmm1[2,3]
-; CHECK-NEXT:    retq
-  %and1 = and <4 x i32> %a, <i32 -1, i32 -1, i32 0, i32 0>
-  %and2 = and <4 x i32> %b, <i32 0, i32 0, i32 -1, i32 -1>
-  %or = or <4 x i32> %and1, %and2
-  ret <4 x i32> %or
-}
-
-
-define <2 x i64> @test8(<2 x i64> %a, <2 x i64> %b) {
-; CHECK-LABEL: test8:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    blendps {{.*#+}} xmm0 = xmm0[0,1],xmm1[2,3]
-; CHECK-NEXT:    retq
-  %and1 = and <2 x i64> %a, <i64 -1, i64 0>
-  %and2 = and <2 x i64> %b, <i64 0, i64 -1>
-  %or = or <2 x i64> %and1, %and2
-  ret <2 x i64> %or
-}
-
-
-define <4 x i32> @test9(<4 x i32> %a, <4 x i32> %b) {
-; CHECK-LABEL: test9:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    blendps {{.*#+}} xmm0 = xmm1[0,1],xmm0[2,3]
-; CHECK-NEXT:    retq
-  %and1 = and <4 x i32> %a, <i32 0, i32 0, i32 -1, i32 -1>
-  %and2 = and <4 x i32> %b, <i32 -1, i32 -1, i32 0, i32 0>
-  %or = or <4 x i32> %and1, %and2
-  ret <4 x i32> %or
-}
-
-
-define <2 x i64> @test10(<2 x i64> %a, <2 x i64> %b) {
-; CHECK-LABEL: test10:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    blendps {{.*#+}} xmm0 = xmm1[0,1],xmm0[2,3]
-; CHECK-NEXT:    retq
-  %and1 = and <2 x i64> %a, <i64 0, i64 -1>
-  %and2 = and <2 x i64> %b, <i64 -1, i64 0>
-  %or = or <2 x i64> %and1, %and2
-  ret <2 x i64> %or
-}
-
-
-define <4 x i32> @test11(<4 x i32> %a, <4 x i32> %b) {
-; CHECK-LABEL: test11:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    blendps {{.*#+}} xmm0 = xmm0[0],xmm1[1,2,3]
-; CHECK-NEXT:    retq
-  %and1 = and <4 x i32> %a, <i32 -1, i32 0, i32 0, i32 0>
-  %and2 = and <4 x i32> %b, <i32 0, i32 -1, i32 -1, i32 -1>
-  %or = or <4 x i32> %and1, %and2
-  ret <4 x i32> %or
-}
-
-
-define <4 x i32> @test12(<4 x i32> %a, <4 x i32> %b) {
-; CHECK-LABEL: test12:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    blendps {{.*#+}} xmm0 = xmm1[0],xmm0[1,2,3]
-; CHECK-NEXT:    retq
-  %and1 = and <4 x i32> %a, <i32 0, i32 -1, i32 -1, i32 -1>
-  %and2 = and <4 x i32> %b, <i32 -1, i32 0, i32 0, i32 0>
-  %or = or <4 x i32> %and1, %and2
-  ret <4 x i32> %or
-}
-
-
-; Verify that the following test cases are folded into single shuffles.
-
-define <4 x i32> @test13(<4 x i32> %a, <4 x i32> %b) {
-; CHECK-LABEL: test13:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    shufps {{.*#+}} xmm0 = xmm0[1,1],xmm1[2,3]
-; CHECK-NEXT:    retq
-  %shuf1 = shufflevector <4 x i32> %a, <4 x i32> zeroinitializer, <4 x i32><i32 1, i32 1, i32 4, i32 4>
-  %shuf2 = shufflevector <4 x i32> %b, <4 x i32> zeroinitializer, <4 x i32><i32 4, i32 4, i32 2, i32 3>
-  %or = or <4 x i32> %shuf1, %shuf2
-  ret <4 x i32> %or
-}
-
-
-define <2 x i64> @test14(<2 x i64> %a, <2 x i64> %b) {
-; CHECK-LABEL: test14:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    movlhps {{.*#+}} xmm0 = xmm0[0],xmm1[0]
-; CHECK-NEXT:    retq
-  %shuf1 = shufflevector <2 x i64> %a, <2 x i64> zeroinitializer, <2 x i32><i32 0, i32 2>
-  %shuf2 = shufflevector <2 x i64> %b, <2 x i64> zeroinitializer, <2 x i32><i32 2, i32 0>
-  %or = or <2 x i64> %shuf1, %shuf2
-  ret <2 x i64> %or
-}
-
-
-define <4 x i32> @test15(<4 x i32> %a, <4 x i32> %b) {
-; CHECK-LABEL: test15:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    shufps {{.*#+}} xmm1 = xmm1[2,1],xmm0[2,1]
-; CHECK-NEXT:    movaps %xmm1, %xmm0
-; CHECK-NEXT:    retq
-  %shuf1 = shufflevector <4 x i32> %a, <4 x i32> zeroinitializer, <4 x i32><i32 4, i32 4, i32 2, i32 1>
-  %shuf2 = shufflevector <4 x i32> %b, <4 x i32> zeroinitializer, <4 x i32><i32 2, i32 1, i32 4, i32 4>
-  %or = or <4 x i32> %shuf1, %shuf2
-  ret <4 x i32> %or
-}
-
-
-define <2 x i64> @test16(<2 x i64> %a, <2 x i64> %b) {
-; CHECK-LABEL: test16:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    movlhps {{.*#+}} xmm1 = xmm1[0],xmm0[0]
-; CHECK-NEXT:    movaps %xmm1, %xmm0
-; CHECK-NEXT:    retq
-  %shuf1 = shufflevector <2 x i64> %a, <2 x i64> zeroinitializer, <2 x i32><i32 2, i32 0>
-  %shuf2 = shufflevector <2 x i64> %b, <2 x i64> zeroinitializer, <2 x i32><i32 0, i32 2>
-  %or = or <2 x i64> %shuf1, %shuf2
-  ret <2 x i64> %or
-}
-
-
-; Verify that the dag-combiner does not fold a OR of two shuffles into a single
-; shuffle instruction when the shuffle indexes are not compatible.
-
-define <4 x i32> @test17(<4 x i32> %a, <4 x i32> %b) {
-; CHECK-LABEL: test17:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    psllq $32, %xmm0
-; CHECK-NEXT:    movq {{.*#+}} xmm1 = xmm1[0],zero
-; CHECK-NEXT:    por %xmm1, %xmm0
-; CHECK-NEXT:    retq
-  %shuf1 = shufflevector <4 x i32> %a, <4 x i32> zeroinitializer, <4 x i32><i32 4, i32 0, i32 4, i32 2>
-  %shuf2 = shufflevector <4 x i32> %b, <4 x i32> zeroinitializer, <4 x i32><i32 0, i32 1, i32 4, i32 4>
-  %or = or <4 x i32> %shuf1, %shuf2
-  ret <4 x i32> %or
-}
-
-
-define <4 x i32> @test18(<4 x i32> %a, <4 x i32> %b) {
-; CHECK-LABEL: test18:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    pxor %xmm2, %xmm2
-; CHECK-NEXT:    pblendw {{.*#+}} xmm0 = xmm0[0,1],xmm2[2,3,4,5,6,7]
-; CHECK-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[1,0,1,1]
-; CHECK-NEXT:    pblendw {{.*#+}} xmm2 = xmm1[0,1],xmm2[2,3,4,5,6,7]
-; CHECK-NEXT:    por %xmm0, %xmm2
-; CHECK-NEXT:    movdqa %xmm2, %xmm0
-; CHECK-NEXT:    retq
-  %shuf1 = shufflevector <4 x i32> %a, <4 x i32> zeroinitializer, <4 x i32><i32 4, i32 0, i32 4, i32 4>
-  %shuf2 = shufflevector <4 x i32> %b, <4 x i32> zeroinitializer, <4 x i32><i32 0, i32 4, i32 4, i32 4>
-  %or = or <4 x i32> %shuf1, %shuf2
-  ret <4 x i32> %or
-}
-
-
-define <4 x i32> @test19(<4 x i32> %a, <4 x i32> %b) {
-; CHECK-LABEL: test19:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    pshufd {{.*#+}} xmm2 = xmm0[0,0,2,3]
-; CHECK-NEXT:    pxor %xmm3, %xmm3
-; CHECK-NEXT:    pblendw {{.*#+}} xmm2 = xmm3[0,1],xmm2[2,3],xmm3[4,5],xmm2[6,7]
-; CHECK-NEXT:    pshufd {{.*#+}} xmm0 = xmm1[0,1,2,2]
-; CHECK-NEXT:    pblendw {{.*#+}} xmm0 = xmm0[0,1],xmm3[2,3],xmm0[4,5,6,7]
-; CHECK-NEXT:    por %xmm2, %xmm0
-; CHECK-NEXT:    retq
-  %shuf1 = shufflevector <4 x i32> %a, <4 x i32> zeroinitializer, <4 x i32><i32 4, i32 0, i32 4, i32 3>
-  %shuf2 = shufflevector <4 x i32> %b, <4 x i32> zeroinitializer, <4 x i32><i32 0, i32 4, i32 2, i32 2>
-  %or = or <4 x i32> %shuf1, %shuf2
-  ret <4 x i32> %or
-}
-
-
-define <2 x i64> @test20(<2 x i64> %a, <2 x i64> %b) {
-; CHECK-LABEL: test20:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    por %xmm1, %xmm0
-; CHECK-NEXT:    movq {{.*#+}} xmm0 = xmm0[0],zero
-; CHECK-NEXT:    retq
-  %shuf1 = shufflevector <2 x i64> %a, <2 x i64> zeroinitializer, <2 x i32><i32 0, i32 2>
-  %shuf2 = shufflevector <2 x i64> %b, <2 x i64> zeroinitializer, <2 x i32><i32 0, i32 2>
-  %or = or <2 x i64> %shuf1, %shuf2
-  ret <2 x i64> %or
-}
-
-
-define <2 x i64> @test21(<2 x i64> %a, <2 x i64> %b) {
-; CHECK-LABEL: test21:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    por %xmm1, %xmm0
-; CHECK-NEXT:    pslldq {{.*#+}} xmm0 = zero,zero,zero,zero,zero,zero,zero,zero,xmm0[0,1,2,3,4,5,6,7]
-; CHECK-NEXT:    retq
-  %shuf1 = shufflevector <2 x i64> %a, <2 x i64> zeroinitializer, <2 x i32><i32 2, i32 0>
-  %shuf2 = shufflevector <2 x i64> %b, <2 x i64> zeroinitializer, <2 x i32><i32 2, i32 0>
-  %or = or <2 x i64> %shuf1, %shuf2
-  ret <2 x i64> %or
-}
-
-
-; Verify that the dag-combiner keeps the correct domain for float/double vectors
-; bitcast to use the mask-or blend combine.
-
-define <2 x double> @test22(<2 x double> %a0, <2 x double> %a1) {
-; CHECK-LABEL: test22:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    blendps {{.*#+}} xmm0 = xmm1[0,1],xmm0[2,3]
-; CHECK-NEXT:    retq
-  %bc1 = bitcast <2 x double> %a0 to <2 x i64>
-  %bc2 = bitcast <2 x double> %a1 to <2 x i64>
-  %and1 = and <2 x i64> %bc1, <i64 0, i64 -1>
-  %and2 = and <2 x i64> %bc2, <i64 -1, i64 0>
-  %or = or <2 x i64> %and1, %and2
-  %bc3 = bitcast <2 x i64> %or to <2 x double>
-  ret <2 x double> %bc3
-}
-
-
-define <4 x float> @test23(<4 x float> %a0, <4 x float> %a1) {
-; CHECK-LABEL: test23:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    blendps {{.*#+}} xmm0 = xmm1[0],xmm0[1,2],xmm1[3]
-; CHECK-NEXT:    retq
-  %bc1 = bitcast <4 x float> %a0 to <4 x i32>
-  %bc2 = bitcast <4 x float> %a1 to <4 x i32>
-  %and1 = and <4 x i32> %bc1, <i32 0, i32 -1, i32 -1, i32 0>
-  %and2 = and <4 x i32> %bc2, <i32 -1, i32 0, i32 0, i32 -1>
-  %or = or <4 x i32> %and1, %and2
-  %bc3 = bitcast <4 x i32> %or to <4 x float>
-  ret <4 x float> %bc3
-}
-
-
-define <4 x float> @test24(<4 x float> %a0, <4 x float> %a1) {
-; CHECK-LABEL: test24:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    blendps {{.*#+}} xmm0 = xmm1[0,1],xmm0[2,3]
-; CHECK-NEXT:    retq
-  %bc1 = bitcast <4 x float> %a0 to <2 x i64>
-  %bc2 = bitcast <4 x float> %a1 to <2 x i64>
-  %and1 = and <2 x i64> %bc1, <i64 0, i64 -1>
-  %and2 = and <2 x i64> %bc2, <i64 -1, i64 0>
-  %or = or <2 x i64> %and1, %and2
-  %bc3 = bitcast <2 x i64> %or to <4 x float>
-  ret <4 x float> %bc3
-}
-
-
-define <4 x float> @test25(<4 x float> %a0) {
-; CHECK-LABEL: test25:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    blendps {{.*#+}} xmm0 = mem[0],xmm0[1,2],mem[3]
-; CHECK-NEXT:    retq
-  %bc1 = bitcast <4 x float> %a0 to <4 x i32>
-  %bc2 = bitcast <4 x float> <float 1.0, float 1.0, float 1.0, float 1.0> to <4 x i32>
-  %and1 = and <4 x i32> %bc1, <i32 0, i32 -1, i32 -1, i32 0>
-  %and2 = and <4 x i32> %bc2, <i32 -1, i32 0, i32 0, i32 -1>
-  %or = or <4 x i32> %and1, %and2
-  %bc3 = bitcast <4 x i32> %or to <4 x float>
-  ret <4 x float> %bc3
-}
-
-
-; Verify that the DAGCombiner doesn't crash in the attempt to check if a shuffle
-; with illegal type has a legal mask. Method 'isShuffleMaskLegal' only knows how to
-; handle legal vector value types.
-define <4 x i8> @test_crash(<4 x i8> %a, <4 x i8> %b) {
-; CHECK-LABEL: test_crash:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    pblendw {{.*#+}} xmm0 = xmm1[0],xmm0[1],xmm1[2,3,4,5,6,7]
-; CHECK-NEXT:    retq
-  %shuf1 = shufflevector <4 x i8> %a, <4 x i8> zeroinitializer, <4 x i32><i32 4, i32 4, i32 2, i32 3>
-  %shuf2 = shufflevector <4 x i8> %b, <4 x i8> zeroinitializer, <4 x i32><i32 0, i32 1, i32 4, i32 4>
-  %or = or <4 x i8> %shuf1, %shuf2
-  ret <4 x i8> %or
-}
-
-; Verify that we can fold regardless of which operand is the zeroinitializer
-
-define <4 x i32> @test2b(<4 x i32> %a, <4 x i32> %b) {
-; CHECK-LABEL: test2b:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    blendps {{.*#+}} xmm0 = xmm1[0,1],xmm0[2,3]
-; CHECK-NEXT:    retq
-  %shuf1 = shufflevector <4 x i32> zeroinitializer, <4 x i32> %a, <4 x i32><i32 0, i32 0, i32 6, i32 7>
-  %shuf2 = shufflevector <4 x i32> %b, <4 x i32> zeroinitializer, <4 x i32><i32 0, i32 1, i32 4, i32 4>
-  %or = or <4 x i32> %shuf1, %shuf2
-  ret <4 x i32> %or
-}
-
-define <4 x i32> @test2c(<4 x i32> %a, <4 x i32> %b) {
-; CHECK-LABEL: test2c:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    blendps {{.*#+}} xmm0 = xmm1[0,1],xmm0[2,3]
-; CHECK-NEXT:    retq
-  %shuf1 = shufflevector <4 x i32> zeroinitializer, <4 x i32> %a, <4 x i32><i32 0, i32 0, i32 6, i32 7>
-  %shuf2 = shufflevector <4 x i32> zeroinitializer, <4 x i32> %b, <4 x i32><i32 4, i32 5, i32 0, i32 0>
-  %or = or <4 x i32> %shuf1, %shuf2
-  ret <4 x i32> %or
-}
-
-
-define <4 x i32> @test2d(<4 x i32> %a, <4 x i32> %b) {
-; CHECK-LABEL: test2d:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    blendps {{.*#+}} xmm0 = xmm1[0,1],xmm0[2,3]
-; CHECK-NEXT:    retq
-  %shuf1 = shufflevector <4 x i32> %a, <4 x i32> zeroinitializer, <4 x i32><i32 4, i32 4, i32 2, i32 3>
-  %shuf2 = shufflevector <4 x i32> zeroinitializer, <4 x i32> %b, <4 x i32><i32 4, i32 5, i32 0, i32 0>
-  %or = or <4 x i32> %shuf1, %shuf2
-  ret <4 x i32> %or
-}
-
-; Make sure we can have an undef where an index pointing to the zero vector should be
-
-define <4 x i32> @test2e(<4 x i32> %a, <4 x i32> %b) {
-; CHECK-LABEL: test2e:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    blendps {{.*#+}} xmm0 = xmm1[0,1],xmm0[2,3]
-; CHECK-NEXT:    retq
-  %shuf1 = shufflevector <4 x i32> %a, <4 x i32> <i32 0, i32 undef, i32 undef, i32 undef>, <4 x i32><i32 undef, i32 4, i32 2, i32 3>
-  %shuf2 = shufflevector <4 x i32> %b, <4 x i32> <i32 0, i32 undef, i32 undef, i32 undef>, <4 x i32><i32 0, i32 1, i32 4, i32 4>
-  %or = or <4 x i32> %shuf1, %shuf2
-  ret <4 x i32> %or
-}
-
-define <4 x i32> @test2f(<4 x i32> %a, <4 x i32> %b) {
-; CHECK-LABEL: test2f:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    blendps {{.*#+}} xmm0 = xmm1[0,1],xmm0[2,3]
-; CHECK-NEXT:    retq
-  %shuf1 = shufflevector <4 x i32> %a, <4 x i32> <i32 0, i32 undef, i32 undef, i32 undef>, <4 x i32><i32 4, i32 4, i32 2, i32 3>
-  %shuf2 = shufflevector <4 x i32> %b, <4 x i32> <i32 0, i32 undef, i32 undef, i32 undef>, <4 x i32><i32 undef, i32 1, i32 4, i32 4>
-  %or = or <4 x i32> %shuf1, %shuf2
-  ret <4 x i32> %or
-}
-
-; (or (and X, c1), c2) -> (and (or X, c2), c1|c2) iff (c1 & c2) != 0
-
-define <2 x i64> @or_and_v2i64(<2 x i64> %a0) {
-; CHECK-LABEL: or_and_v2i64:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    orps {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0
-; CHECK-NEXT:    andps {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0
-; CHECK-NEXT:    retq
-  %1 = and <2 x i64> %a0, <i64 7, i64 7>
-  %2 = or <2 x i64> %1, <i64 3, i64 3>
-  ret <2 x i64> %2
-}
-
-define <4 x i32> @or_and_v4i32(<4 x i32> %a0) {
-; CHECK-LABEL: or_and_v4i32:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    orps {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0
-; CHECK-NEXT:    andps {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0
-; CHECK-NEXT:    retq
-  %1 = and <4 x i32> %a0, <i32 1, i32 3, i32 5, i32 7>
-  %2 = or <4 x i32> %1, <i32 3, i32 2, i32 15, i32 2>
-  ret <4 x i32> %2
-}
-
-; If all masked bits are going to be set, that's a constant fold.
-
-define <4 x i32> @or_and_v4i32_fold(<4 x i32> %a0) {
-; CHECK-LABEL: or_and_v4i32_fold:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    movaps {{.*#+}} xmm0 = [3,3,3,3]
-; CHECK-NEXT:    retq
-  %1 = and <4 x i32> %a0, <i32 1, i32 1, i32 1, i32 1>
-  %2 = or <4 x i32> %1, <i32 3, i32 3, i32 3, i32 3>
-  ret <4 x i32> %2
-}
-
 ; fold (or x, c) -> c iff (x & ~c) == 0
 
 define <2 x i64> @or_zext_v2i32(<2 x i32> %a0) {
-; CHECK-LABEL: or_zext_v2i32:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    movaps {{.*#+}} xmm0 = [4294967295,4294967295]
-; CHECK-NEXT:    retq
+; SSE-LABEL: or_zext_v2i32:
+; SSE:       # %bb.0:
+; SSE-NEXT:    movaps {{.*#+}} xmm0 = [4294967295,4294967295]
+; SSE-NEXT:    retq
+;
+; AVX-LABEL: or_zext_v2i32:
+; AVX:       # %bb.0:
+; AVX-NEXT:    vmovddup {{.*#+}} xmm0 = [4294967295,4294967295]
+; AVX-NEXT:    # xmm0 = mem[0,0]
+; AVX-NEXT:    retq
   %1 = zext <2 x i32> %a0 to <2 x i64>
   %2 = or <2 x i64> %1, <i64 4294967295, i64 4294967295>
   ret <2 x i64> %2
 }
 
 define <4 x i32> @or_zext_v4i16(<4 x i16> %a0) {
-; CHECK-LABEL: or_zext_v4i16:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    movaps {{.*#+}} xmm0 = [65535,65535,65535,65535]
-; CHECK-NEXT:    retq
+; SSE-LABEL: or_zext_v4i16:
+; SSE:       # %bb.0:
+; SSE-NEXT:    movaps {{.*#+}} xmm0 = [65535,65535,65535,65535]
+; SSE-NEXT:    retq
+;
+; AVX-LABEL: or_zext_v4i16:
+; AVX:       # %bb.0:
+; AVX-NEXT:    vbroadcastss {{.*#+}} xmm0 = [65535,65535,65535,65535]
+; AVX-NEXT:    retq
   %1 = zext <4 x i16> %a0 to <4 x i32>
   %2 = or <4 x i32> %1, <i32 65535, i32 65535, i32 65535, i32 65535>
   ret <4 x i32> %2
@@ -522,12 +86,19 @@ define i64 @or_and_and_commute_i64(i64 %x, i64 %y) {
 }
 
 define <4 x i32> @or_and_and_v4i32(<4 x i32> %x, <4 x i32> %y) {
-; CHECK-LABEL: or_and_and_v4i32:
-; CHECK:       # %bb.0:
-; CHECK-NEXT:    andps {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm1
-; CHECK-NEXT:    andps {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0
-; CHECK-NEXT:    orps %xmm1, %xmm0
-; CHECK-NEXT:    retq
+; SSE-LABEL: or_and_and_v4i32:
+; SSE:       # %bb.0:
+; SSE-NEXT:    andps {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm1
+; SSE-NEXT:    andps {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0
+; SSE-NEXT:    orps %xmm1, %xmm0
+; SSE-NEXT:    retq
+;
+; AVX-LABEL: or_and_and_v4i32:
+; AVX:       # %bb.0:
+; AVX-NEXT:    vandps {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm1, %xmm1
+; AVX-NEXT:    vandps {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0, %xmm0
+; AVX-NEXT:    vorps %xmm1, %xmm0, %xmm0
+; AVX-NEXT:    retq
   %xy = or <4 x i32> %x, %y
   %mx = and <4 x i32> %x, <i32 2, i32 4, i32 8, i32 16>
   %mxy = and <4 x i32> %xy, <i32 1, i32 -1, i32 -5, i32 -25>
@@ -611,7 +182,106 @@ define i32 @or_and_multiuse_and_multiuse_i32(i32 %x, i32 %y) nounwind {
   ret i32 %r
 }
 
+define i64 @or_build_pair_not(i32 %a0, i32 %a1) {
+; CHECK-LABEL: or_build_pair_not:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    # kill: def $esi killed $esi def $rsi
+; CHECK-NEXT:    shlq $32, %rsi
+; CHECK-NEXT:    movl %edi, %eax
+; CHECK-NEXT:    orq %rsi, %rax
+; CHECK-NEXT:    notq %rax
+; CHECK-NEXT:    retq
+  %n0 = xor i32 %a0, -1
+  %n1 = xor i32 %a1, -1
+  %x0 = zext i32 %n0 to i64
+  %x1 = zext i32 %n1 to i64
+  %hi = shl i64 %x1, 32
+  %r = or i64 %hi, %x0
+  ret i64 %r
+}
+
+define i64 @PR89533(<64 x i8> %a0) {
+; SSE-LABEL: PR89533:
+; SSE:       # %bb.0:
+; SSE-NEXT:    movdqa {{.*#+}} xmm4 = [95,95,95,95,95,95,95,95,95,95,95,95,95,95,95,95]
+; SSE-NEXT:    pcmpeqb %xmm4, %xmm0
+; SSE-NEXT:    pmovmskb %xmm0, %eax
+; SSE-NEXT:    xorl $65535, %eax # imm = 0xFFFF
+; SSE-NEXT:    pcmpeqb %xmm4, %xmm1
+; SSE-NEXT:    pmovmskb %xmm1, %ecx
+; SSE-NEXT:    notl %ecx
+; SSE-NEXT:    shll $16, %ecx
+; SSE-NEXT:    orl %eax, %ecx
+; SSE-NEXT:    pcmpeqb %xmm4, %xmm2
+; SSE-NEXT:    pmovmskb %xmm2, %edx
+; SSE-NEXT:    xorl $65535, %edx # imm = 0xFFFF
+; SSE-NEXT:    pcmpeqb %xmm4, %xmm3
+; SSE-NEXT:    pmovmskb %xmm3, %eax
+; SSE-NEXT:    notl %eax
+; SSE-NEXT:    shll $16, %eax
+; SSE-NEXT:    orl %edx, %eax
+; SSE-NEXT:    shlq $32, %rax
+; SSE-NEXT:    orq %rcx, %rax
+; SSE-NEXT:    je .LBB11_2
+; SSE-NEXT:  # %bb.1: # %cond.false
+; SSE-NEXT:    rep bsfq %rax, %rax
+; SSE-NEXT:    retq
+; SSE-NEXT:  .LBB11_2: # %cond.end
+; SSE-NEXT:    movl $64, %eax
+; SSE-NEXT:    retq
+;
+; AVX1-LABEL: PR89533:
+; AVX1:       # %bb.0:
+; AVX1-NEXT:    vbroadcastss {{.*#+}} xmm2 = [95,95,95,95,95,95,95,95,95,95,95,95,95,95,95,95]
+; AVX1-NEXT:    vpcmpeqb %xmm2, %xmm0, %xmm3
+; AVX1-NEXT:    vpmovmskb %xmm3, %eax
+; AVX1-NEXT:    xorl $65535, %eax # imm = 0xFFFF
+; AVX1-NEXT:    vextractf128 $1, %ymm0, %xmm0
+; AVX1-NEXT:    vpcmpeqb %xmm2, %xmm0, %xmm0
+; AVX1-NEXT:    vpmovmskb %xmm0, %ecx
+; AVX1-NEXT:    notl %ecx
+; AVX1-NEXT:    shll $16, %ecx
+; AVX1-NEXT:    orl %eax, %ecx
+; AVX1-NEXT:    vpcmpeqb %xmm2, %xmm1, %xmm0
+; AVX1-NEXT:    vpmovmskb %xmm0, %edx
+; AVX1-NEXT:    xorl $65535, %edx # imm = 0xFFFF
+; AVX1-NEXT:    vextractf128 $1, %ymm1, %xmm0
+; AVX1-NEXT:    vpcmpeqb %xmm2, %xmm0, %xmm0
+; AVX1-NEXT:    vpmovmskb %xmm0, %eax
+; AVX1-NEXT:    notl %eax
+; AVX1-NEXT:    shll $16, %eax
+; AVX1-NEXT:    orl %edx, %eax
+; AVX1-NEXT:    shlq $32, %rax
+; AVX1-NEXT:    orq %rcx, %rax
+; AVX1-NEXT:    je .LBB11_2
+; AVX1-NEXT:  # %bb.1: # %cond.false
+; AVX1-NEXT:    rep bsfq %rax, %rax
+; AVX1-NEXT:    vzeroupper
+; AVX1-NEXT:    retq
+; AVX1-NEXT:  .LBB11_2: # %cond.end
+; AVX1-NEXT:    movl $64, %eax
+; AVX1-NEXT:    vzeroupper
+; AVX1-NEXT:    retq
+;
+; AVX2-LABEL: PR89533:
+; AVX2:       # %bb.0:
+; AVX2-NEXT:    vpbroadcastb {{.*#+}} ymm2 = [95,95,95,95,95,95,95,95,95,95,95,95,95,95,95,95,95,95,95,95,95,95,95,95,95,95,95,95,95,95,95,95]
+; AVX2-NEXT:    vpcmpeqb %ymm2, %ymm0, %ymm0
+; AVX2-NEXT:    vpmovmskb %ymm0, %eax
+; AVX2-NEXT:    vpcmpeqb %ymm2, %ymm1, %ymm0
+; AVX2-NEXT:    vpmovmskb %ymm0, %ecx
+; AVX2-NEXT:    shlq $32, %rcx
+; AVX2-NEXT:    orq %rax, %rcx
+; AVX2-NEXT:    notq %rcx
+; AVX2-NEXT:    xorl %eax, %eax
+; AVX2-NEXT:    tzcntq %rcx, %rax
+; AVX2-NEXT:    vzeroupper
+; AVX2-NEXT:    retq
+  %cmp = icmp ne <64 x i8> %a0, <i8 95, i8 95, i8 95, i8 95, i8 95, i8 95, i8 95, i8 95, i8 95, i8 95, i8 95, i8 95, i8 95, i8 95, i8 95, i8 95, i8 95, i8 95, i8 95, i8 95, i8 95, i8 95, i8 95, i8 95, i8 95, i8 95, i8 95, i8 95, i8 95, i8 95, i8 95, i8 95, i8 95, i8 95, i8 95, i8 95, i8 95, i8 95, i8 95, i8 95, i8 95, i8 95, i8 95, i8 95, i8 95, i8 95, i8 95, i8 95, i8 95, i8 95, i8 95, i8 95, i8 95, i8 95, i8 95, i8 95, i8 95, i8 95, i8 95, i8 95, i8 95, i8 95, i8 95, i8 95>
+  %mask = bitcast <64 x i1> %cmp to i64
+  %tz = tail call i64 @llvm.cttz.i64(i64 %mask, i1 false)
+  ret i64 %tz
+}
+
 declare void @use_i32(i32)
-;; NOTE: These prefixes are unused and the list is autogenerated. Do not add tests below this line:
-; CHECK-LIS: {{.*}}
-; CHECK-LV: {{.*}}
+
diff --git a/llvm/test/CodeGen/X86/div-rem-pair-recomposition-signed.ll b/llvm/test/CodeGen/X86/div-rem-pair-recomposition-signed.ll
index e12ca56023a7..33cc8e96f663 100644
--- a/llvm/test/CodeGen/X86/div-rem-pair-recomposition-signed.ll
+++ b/llvm/test/CodeGen/X86/div-rem-pair-recomposition-signed.ll
@@ -182,101 +182,101 @@ define i128 @scalar_i128(i128 %x, i128 %y, ptr %divdst) nounwind {
 ; X86-NEXT:    movl {{[0-9]+}}(%esp), %ecx
 ; X86-NEXT:    movl %ecx, %eax
 ; X86-NEXT:    sarl $31, %eax
-; X86-NEXT:    movl %edx, %edi
-; X86-NEXT:    sarl $31, %edi
+; X86-NEXT:    movl %edx, %ebx
+; X86-NEXT:    sarl $31, %ebx
 ; X86-NEXT:    movl %eax, %esi
 ; X86-NEXT:    xorl %ecx, %esi
 ; X86-NEXT:    movl %esi, %ebp
 ; X86-NEXT:    movl %eax, %ecx
 ; X86-NEXT:    xorl {{[0-9]+}}(%esp), %ecx
-; X86-NEXT:    movl %ecx, %ebx
+; X86-NEXT:    movl %ecx, %edi
 ; X86-NEXT:    movl %eax, %ecx
 ; X86-NEXT:    xorl {{[0-9]+}}(%esp), %ecx
 ; X86-NEXT:    movl %eax, %esi
 ; X86-NEXT:    xorl {{[0-9]+}}(%esp), %esi
 ; X86-NEXT:    subl %eax, %esi
-; X86-NEXT:    movl %esi, (%esp) # 4-byte Spill
+; X86-NEXT:    movl %esi, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
 ; X86-NEXT:    sbbl %eax, %ecx
 ; X86-NEXT:    movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
-; X86-NEXT:    sbbl %eax, %ebx
-; X86-NEXT:    movl %ebx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT:    sbbl %eax, %edi
+; X86-NEXT:    movl %edi, (%esp) # 4-byte Spill
 ; X86-NEXT:    sbbl %eax, %ebp
 ; X86-NEXT:    movl %ebp, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
-; X86-NEXT:    movl %edi, %esi
+; X86-NEXT:    movl %ebx, %esi
 ; X86-NEXT:    xorl %edx, %esi
-; X86-NEXT:    movl %edi, %edx
+; X86-NEXT:    movl %ebx, %edx
 ; X86-NEXT:    xorl {{[0-9]+}}(%esp), %edx
-; X86-NEXT:    movl %edi, %ebx
-; X86-NEXT:    xorl {{[0-9]+}}(%esp), %ebx
-; X86-NEXT:    movl %edi, %ebp
+; X86-NEXT:    movl %ebx, %edi
+; X86-NEXT:    xorl {{[0-9]+}}(%esp), %edi
+; X86-NEXT:    movl %ebx, %ebp
 ; X86-NEXT:    xorl {{[0-9]+}}(%esp), %ebp
-; X86-NEXT:    subl %edi, %ebp
-; X86-NEXT:    sbbl %edi, %ebx
-; X86-NEXT:    sbbl %edi, %edx
-; X86-NEXT:    sbbl %edi, %esi
-; X86-NEXT:    xorl %eax, %edi
-; X86-NEXT:    movl %edi, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
-; X86-NEXT:    movl %ebx, %eax
+; X86-NEXT:    subl %ebx, %ebp
+; X86-NEXT:    sbbl %ebx, %edi
+; X86-NEXT:    sbbl %ebx, %edx
+; X86-NEXT:    sbbl %ebx, %esi
+; X86-NEXT:    xorl %eax, %ebx
+; X86-NEXT:    movl %ebx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT:    movl %edi, %eax
 ; X86-NEXT:    orl %esi, %eax
 ; X86-NEXT:    movl %ebp, %ecx
 ; X86-NEXT:    orl %edx, %ecx
-; X86-NEXT:    movl %edx, %edi
+; X86-NEXT:    movl %edx, %ebx
 ; X86-NEXT:    orl %eax, %ecx
 ; X86-NEXT:    sete %cl
 ; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
 ; X86-NEXT:    orl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Folded Reload
-; X86-NEXT:    movl (%esp), %edx # 4-byte Reload
-; X86-NEXT:    orl {{[-0-9]+}}(%e{{[sb]}}p), %edx # 4-byte Folded Reload
+; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %edx # 4-byte Reload
+; X86-NEXT:    orl (%esp), %edx # 4-byte Folded Reload
 ; X86-NEXT:    orl %eax, %edx
 ; X86-NEXT:    sete %al
 ; X86-NEXT:    orb %cl, %al
 ; X86-NEXT:    movb %al, {{[-0-9]+}}(%e{{[sb]}}p) # 1-byte Spill
 ; X86-NEXT:    bsrl %esi, %edx
 ; X86-NEXT:    xorl $31, %edx
-; X86-NEXT:    bsrl %edi, %ecx
+; X86-NEXT:    bsrl %ebx, %ecx
 ; X86-NEXT:    xorl $31, %ecx
 ; X86-NEXT:    addl $32, %ecx
 ; X86-NEXT:    testl %esi, %esi
 ; X86-NEXT:    cmovnel %edx, %ecx
-; X86-NEXT:    bsrl %ebx, %edx
+; X86-NEXT:    bsrl %edi, %edx
 ; X86-NEXT:    xorl $31, %edx
 ; X86-NEXT:    movl %ebp, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
 ; X86-NEXT:    bsrl %ebp, %ebp
 ; X86-NEXT:    xorl $31, %ebp
 ; X86-NEXT:    addl $32, %ebp
-; X86-NEXT:    movl %ebx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
-; X86-NEXT:    testl %ebx, %ebx
+; X86-NEXT:    movl %edi, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT:    testl %edi, %edi
 ; X86-NEXT:    cmovnel %edx, %ebp
 ; X86-NEXT:    addl $64, %ebp
-; X86-NEXT:    movl %edi, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT:    movl %ebx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
 ; X86-NEXT:    movl %esi, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
-; X86-NEXT:    orl %esi, %edi
+; X86-NEXT:    orl %esi, %ebx
 ; X86-NEXT:    cmovnel %ecx, %ebp
-; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %edi # 4-byte Reload
-; X86-NEXT:    bsrl %edi, %edx
+; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %ebx # 4-byte Reload
+; X86-NEXT:    bsrl %ebx, %edx
 ; X86-NEXT:    xorl $31, %edx
-; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT:    movl (%esp), %eax # 4-byte Reload
 ; X86-NEXT:    bsrl %eax, %ecx
 ; X86-NEXT:    xorl $31, %ecx
 ; X86-NEXT:    addl $32, %ecx
-; X86-NEXT:    testl %edi, %edi
+; X86-NEXT:    testl %ebx, %ebx
 ; X86-NEXT:    cmovnel %edx, %ecx
-; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %ebx # 4-byte Reload
-; X86-NEXT:    bsrl %ebx, %esi
+; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %edi # 4-byte Reload
+; X86-NEXT:    bsrl %edi, %esi
 ; X86-NEXT:    xorl $31, %esi
-; X86-NEXT:    bsrl (%esp), %edx # 4-byte Folded Reload
+; X86-NEXT:    bsrl {{[-0-9]+}}(%e{{[sb]}}p), %edx # 4-byte Folded Reload
 ; X86-NEXT:    xorl $31, %edx
 ; X86-NEXT:    addl $32, %edx
-; X86-NEXT:    testl %ebx, %ebx
+; X86-NEXT:    testl %edi, %edi
 ; X86-NEXT:    cmovnel %esi, %edx
 ; X86-NEXT:    addl $64, %edx
 ; X86-NEXT:    movl %eax, %esi
-; X86-NEXT:    orl %edi, %esi
+; X86-NEXT:    orl %ebx, %esi
 ; X86-NEXT:    cmovnel %ecx, %edx
 ; X86-NEXT:    xorl %esi, %esi
 ; X86-NEXT:    subl %edx, %ebp
-; X86-NEXT:    movl $0, %ebx
-; X86-NEXT:    sbbl %ebx, %ebx
+; X86-NEXT:    movl $0, %edi
+; X86-NEXT:    sbbl %edi, %edi
 ; X86-NEXT:    movl $0, %edx
 ; X86-NEXT:    sbbl %edx, %edx
 ; X86-NEXT:    movl $0, %eax
@@ -284,40 +284,40 @@ define i128 @scalar_i128(i128 %x, i128 %y, ptr %divdst) nounwind {
 ; X86-NEXT:    movl $127, %ecx
 ; X86-NEXT:    movl %ebp, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
 ; X86-NEXT:    cmpl %ebp, %ecx
+; X86-NEXT:    movl %edx, %ebp
 ; X86-NEXT:    movl $0, %ecx
-; X86-NEXT:    sbbl %ebx, %ecx
+; X86-NEXT:    sbbl %edi, %ecx
 ; X86-NEXT:    movl $0, %ecx
-; X86-NEXT:    movl %edx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
 ; X86-NEXT:    sbbl %edx, %ecx
 ; X86-NEXT:    movl $0, %ecx
 ; X86-NEXT:    movl %eax, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
 ; X86-NEXT:    sbbl %eax, %ecx
 ; X86-NEXT:    setb %cl
 ; X86-NEXT:    orb {{[-0-9]+}}(%e{{[sb]}}p), %cl # 1-byte Folded Reload
-; X86-NEXT:    cmovnel %esi, %edi
-; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %edx # 4-byte Reload
+; X86-NEXT:    movl %ebx, %edx
 ; X86-NEXT:    cmovnel %esi, %edx
+; X86-NEXT:    movl (%esp), %ebx # 4-byte Reload
+; X86-NEXT:    cmovnel %esi, %ebx
 ; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
 ; X86-NEXT:    cmovnel %esi, %eax
-; X86-NEXT:    cmovel (%esp), %esi # 4-byte Folded Reload
-; X86-NEXT:    movl {{[0-9]+}}(%esp), %ebp
+; X86-NEXT:    cmovel {{[-0-9]+}}(%e{{[sb]}}p), %esi # 4-byte Folded Reload
 ; X86-NEXT:    jne .LBB4_8
 ; X86-NEXT:  # %bb.1: # %_udiv-special-cases
-; X86-NEXT:    movl %ebx, %ecx
-; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %ebx # 4-byte Reload
-; X86-NEXT:    xorl $127, %ebx
-; X86-NEXT:    orl {{[-0-9]+}}(%e{{[sb]}}p), %ebx # 4-byte Folded Reload
+; X86-NEXT:    movl %edi, %ecx
+; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %edi # 4-byte Reload
+; X86-NEXT:    xorl $127, %edi
+; X86-NEXT:    orl %ebp, %edi
 ; X86-NEXT:    movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
 ; X86-NEXT:    orl {{[-0-9]+}}(%e{{[sb]}}p), %ecx # 4-byte Folded Reload
-; X86-NEXT:    orl %ebx, %ecx
+; X86-NEXT:    orl %edi, %ecx
 ; X86-NEXT:    je .LBB4_8
 ; X86-NEXT:  # %bb.2: # %udiv-bb1
-; X86-NEXT:    movl (%esp), %eax # 4-byte Reload
-; X86-NEXT:    movl %eax, {{[0-9]+}}(%esp)
 ; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
 ; X86-NEXT:    movl %eax, {{[0-9]+}}(%esp)
 ; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
 ; X86-NEXT:    movl %eax, {{[0-9]+}}(%esp)
+; X86-NEXT:    movl (%esp), %eax # 4-byte Reload
+; X86-NEXT:    movl %eax, {{[0-9]+}}(%esp)
 ; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
 ; X86-NEXT:    movl %eax, {{[0-9]+}}(%esp)
 ; X86-NEXT:    movl $0, {{[0-9]+}}(%esp)
@@ -332,234 +332,233 @@ define i128 @scalar_i128(i128 %x, i128 %y, ptr %divdst) nounwind {
 ; X86-NEXT:    shrb $3, %al
 ; X86-NEXT:    andb $15, %al
 ; X86-NEXT:    negb %al
-; X86-NEXT:    movsbl %al, %ebx
-; X86-NEXT:    movl 144(%esp,%ebx), %edx
-; X86-NEXT:    movl 148(%esp,%ebx), %edi
+; X86-NEXT:    movsbl %al, %edi
+; X86-NEXT:    movl 144(%esp,%edi), %edx
+; X86-NEXT:    movl 148(%esp,%edi), %esi
 ; X86-NEXT:    movb %ch, %cl
-; X86-NEXT:    shldl %cl, %edx, %edi
+; X86-NEXT:    shldl %cl, %edx, %esi
+; X86-NEXT:    movl %esi, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
 ; X86-NEXT:    shll %cl, %edx
 ; X86-NEXT:    notb %cl
-; X86-NEXT:    movl 140(%esp,%ebx), %eax
-; X86-NEXT:    movl %eax, %esi
-; X86-NEXT:    shrl %esi
-; X86-NEXT:    shrl %cl, %esi
-; X86-NEXT:    orl %edx, %esi
-; X86-NEXT:    movl %esi, %edx
-; X86-NEXT:    movl 136(%esp,%ebx), %esi
+; X86-NEXT:    movl 140(%esp,%edi), %eax
+; X86-NEXT:    movl %eax, %ebx
+; X86-NEXT:    shrl %ebx
+; X86-NEXT:    shrl %cl, %ebx
+; X86-NEXT:    orl %edx, %ebx
+; X86-NEXT:    movl 136(%esp,%edi), %edx
 ; X86-NEXT:    movb %ch, %cl
-; X86-NEXT:    shldl %cl, %esi, %eax
-; X86-NEXT:    shll %cl, %esi
-; X86-NEXT:    movl %esi, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT:    shldl %cl, %edx, %eax
+; X86-NEXT:    shll %cl, %edx
+; X86-NEXT:    movl %edx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
 ; X86-NEXT:    addl $1, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Folded Spill
 ; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %ecx # 4-byte Reload
 ; X86-NEXT:    adcl $0, %ecx
-; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %ebx # 4-byte Reload
-; X86-NEXT:    adcl $0, %ebx
-; X86-NEXT:    adcl $0, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Folded Spill
+; X86-NEXT:    adcl $0, %ebp
+; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %edx # 4-byte Reload
+; X86-NEXT:    adcl $0, %edx
 ; X86-NEXT:    jae .LBB4_3
 ; X86-NEXT:  # %bb.6:
-; X86-NEXT:    xorl %ebx, %ebx
-; X86-NEXT:    xorl %esi, %esi
+; X86-NEXT:    xorl %edi, %edi
+; X86-NEXT:    xorl %ecx, %ecx
+; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %edx # 4-byte Reload
 ; X86-NEXT:    jmp .LBB4_7
 ; X86-NEXT:  .LBB4_3: # %udiv-preheader
-; X86-NEXT:    movl (%esp), %esi # 4-byte Reload
-; X86-NEXT:    movl %esi, {{[0-9]+}}(%esp)
+; X86-NEXT:    movl %ecx, %esi
+; X86-NEXT:    movl %esi, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
 ; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %esi # 4-byte Reload
 ; X86-NEXT:    movl %esi, {{[0-9]+}}(%esp)
 ; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %esi # 4-byte Reload
 ; X86-NEXT:    movl %esi, {{[0-9]+}}(%esp)
+; X86-NEXT:    movl (%esp), %esi # 4-byte Reload
+; X86-NEXT:    movl %esi, {{[0-9]+}}(%esp)
 ; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %esi # 4-byte Reload
 ; X86-NEXT:    movl %esi, {{[0-9]+}}(%esp)
 ; X86-NEXT:    movl $0, {{[0-9]+}}(%esp)
 ; X86-NEXT:    movl $0, {{[0-9]+}}(%esp)
 ; X86-NEXT:    movl $0, {{[0-9]+}}(%esp)
 ; X86-NEXT:    movl $0, {{[0-9]+}}(%esp)
-; X86-NEXT:    movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
-; X86-NEXT:    movl %ebx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
-; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %ebx # 4-byte Reload
-; X86-NEXT:    movb %bl, %ch
+; X86-NEXT:    movl %edx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %edx # 4-byte Reload
+; X86-NEXT:    movl %ebp, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT:    movb %dl, %ch
 ; X86-NEXT:    andb $7, %ch
-; X86-NEXT:    movb %bl, %cl
+; X86-NEXT:    movb %dl, %cl
 ; X86-NEXT:    shrb $3, %cl
 ; X86-NEXT:    andb $15, %cl
-; X86-NEXT:    movzbl %cl, %ebp
-; X86-NEXT:    movl 100(%esp,%ebp), %esi
-; X86-NEXT:    movl %esi, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
-; X86-NEXT:    movl %eax, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
-; X86-NEXT:    movl 96(%esp,%ebp), %ebx
-; X86-NEXT:    movl %ebp, %eax
-; X86-NEXT:    movl %edx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
-; X86-NEXT:    movl %ebx, %edx
-; X86-NEXT:    movb %ch, %cl
-; X86-NEXT:    shrdl %cl, %esi, %edx
-; X86-NEXT:    movl 88(%esp,%ebp), %ebp
-; X86-NEXT:    movl 92(%esp,%eax), %esi
-; X86-NEXT:    movl %esi, %eax
-; X86-NEXT:    shrl %cl, %eax
-; X86-NEXT:    notb %cl
-; X86-NEXT:    addl %ebx, %ebx
-; X86-NEXT:    shll %cl, %ebx
-; X86-NEXT:    orl %eax, %ebx
+; X86-NEXT:    movzbl %cl, %edx
+; X86-NEXT:    movl 100(%esp,%edx), %esi
+; X86-NEXT:    movl %esi, (%esp) # 4-byte Spill
+; X86-NEXT:    movl 96(%esp,%edx), %edi
 ; X86-NEXT:    movl %ebx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT:    movl %edi, %ebp
 ; X86-NEXT:    movb %ch, %cl
-; X86-NEXT:    shrl %cl, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Folded Spill
 ; X86-NEXT:    shrdl %cl, %esi, %ebp
-; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
-; X86-NEXT:    addl $-1, %eax
-; X86-NEXT:    movl %eax, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
-; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
-; X86-NEXT:    adcl $-1, %eax
-; X86-NEXT:    movl %eax, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
-; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
-; X86-NEXT:    adcl $-1, %eax
-; X86-NEXT:    movl %eax, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT:    movl 88(%esp,%edx), %ebx
+; X86-NEXT:    movl 92(%esp,%edx), %esi
+; X86-NEXT:    movl %esi, %edx
+; X86-NEXT:    shrl %cl, %edx
+; X86-NEXT:    notb %cl
+; X86-NEXT:    addl %edi, %edi
+; X86-NEXT:    shll %cl, %edi
+; X86-NEXT:    orl %edx, %edi
+; X86-NEXT:    movl %edi, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT:    movb %ch, %cl
+; X86-NEXT:    shrl %cl, (%esp) # 4-byte Folded Spill
+; X86-NEXT:    shrdl %cl, %esi, %ebx
+; X86-NEXT:    movl %ebx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
 ; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %ecx # 4-byte Reload
-; X86-NEXT:    movl %ecx, %eax
-; X86-NEXT:    adcl $-1, %eax
-; X86-NEXT:    movl %eax, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
-; X86-NEXT:    xorl %esi, %esi
+; X86-NEXT:    addl $-1, %ecx
+; X86-NEXT:    movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %ecx # 4-byte Reload
+; X86-NEXT:    adcl $-1, %ecx
+; X86-NEXT:    movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %ecx # 4-byte Reload
+; X86-NEXT:    adcl $-1, %ecx
+; X86-NEXT:    movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %edx # 4-byte Reload
+; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %ecx # 4-byte Reload
+; X86-NEXT:    adcl $-1, %ecx
+; X86-NEXT:    movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT:    movl $0, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Folded Spill
 ; X86-NEXT:    movl $0, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Folded Spill
-; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %ebx # 4-byte Reload
 ; X86-NEXT:    .p2align 4, 0x90
 ; X86-NEXT:  .LBB4_4: # %udiv-do-while
 ; X86-NEXT:    # =>This Inner Loop Header: Depth=1
-; X86-NEXT:    movl %ebx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
 ; X86-NEXT:    movl %edx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
-; X86-NEXT:    movl %esi, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
-; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %ebx # 4-byte Reload
-; X86-NEXT:    shldl $1, %edx, %ebx
+; X86-NEXT:    movl %ebp, %esi
+; X86-NEXT:    shldl $1, %ebp, (%esp) # 4-byte Folded Spill
+; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %ebp # 4-byte Reload
+; X86-NEXT:    shldl $1, %ebp, %esi
 ; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %edx # 4-byte Reload
-; X86-NEXT:    shldl $1, %edx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Folded Spill
-; X86-NEXT:    shldl $1, %ebp, %edx
-; X86-NEXT:    shldl $1, %edi, %ebp
-; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
-; X86-NEXT:    shldl $1, %eax, %edi
-; X86-NEXT:    orl %esi, %edi
-; X86-NEXT:    movl %edi, (%esp) # 4-byte Spill
-; X86-NEXT:    movl %ecx, %edi
+; X86-NEXT:    shldl $1, %edx, %ebp
+; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %edi # 4-byte Reload
+; X86-NEXT:    shldl $1, %edi, %edx
 ; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %ecx # 4-byte Reload
-; X86-NEXT:    shldl $1, %ecx, %eax
-; X86-NEXT:    orl %esi, %eax
-; X86-NEXT:    movl %eax, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
-; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT:    shldl $1, %ecx, %edi
+; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %ebx # 4-byte Reload
+; X86-NEXT:    orl %ebx, %edi
+; X86-NEXT:    movl %edi, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
 ; X86-NEXT:    shldl $1, %eax, %ecx
-; X86-NEXT:    orl %esi, %ecx
+; X86-NEXT:    orl %ebx, %ecx
 ; X86-NEXT:    movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
-; X86-NEXT:    addl %eax, %eax
-; X86-NEXT:    orl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Folded Reload
+; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %ecx # 4-byte Reload
+; X86-NEXT:    shldl $1, %ecx, %eax
+; X86-NEXT:    orl %ebx, %eax
 ; X86-NEXT:    movl %eax, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
-; X86-NEXT:    cmpl %ebp, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Folded Reload
+; X86-NEXT:    addl %ecx, %ecx
+; X86-NEXT:    orl {{[-0-9]+}}(%e{{[sb]}}p), %ecx # 4-byte Folded Reload
+; X86-NEXT:    movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT:    cmpl %edx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Folded Reload
 ; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %ecx # 4-byte Reload
-; X86-NEXT:    sbbl %edx, %ecx
+; X86-NEXT:    sbbl %ebp, %ecx
 ; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %ecx # 4-byte Reload
-; X86-NEXT:    sbbl {{[-0-9]+}}(%e{{[sb]}}p), %ecx # 4-byte Folded Reload
+; X86-NEXT:    sbbl %esi, %ecx
 ; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %ecx # 4-byte Reload
-; X86-NEXT:    sbbl %ebx, %ecx
+; X86-NEXT:    sbbl (%esp), %ecx # 4-byte Folded Reload
 ; X86-NEXT:    sarl $31, %ecx
 ; X86-NEXT:    movl %ecx, %eax
 ; X86-NEXT:    andl $1, %eax
 ; X86-NEXT:    movl %eax, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
-; X86-NEXT:    movl %ecx, %esi
-; X86-NEXT:    andl %edi, %esi
+; X86-NEXT:    movl %ecx, %ebx
+; X86-NEXT:    andl {{[-0-9]+}}(%e{{[sb]}}p), %ebx # 4-byte Folded Reload
 ; X86-NEXT:    movl %ecx, %edi
 ; X86-NEXT:    andl {{[-0-9]+}}(%e{{[sb]}}p), %edi # 4-byte Folded Reload
 ; X86-NEXT:    movl %ecx, %eax
 ; X86-NEXT:    andl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Folded Reload
 ; X86-NEXT:    andl {{[-0-9]+}}(%e{{[sb]}}p), %ecx # 4-byte Folded Reload
-; X86-NEXT:    subl %ecx, %ebp
-; X86-NEXT:    sbbl %eax, %edx
+; X86-NEXT:    subl %ecx, %edx
 ; X86-NEXT:    movl %edx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
 ; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %edx # 4-byte Reload
-; X86-NEXT:    sbbl %edi, %edx
-; X86-NEXT:    movl (%esp), %edi # 4-byte Reload
-; X86-NEXT:    sbbl %esi, %ebx
-; X86-NEXT:    movl %ebx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
-; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %ebx # 4-byte Reload
+; X86-NEXT:    sbbl %eax, %ebp
+; X86-NEXT:    movl %ebp, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT:    sbbl %edi, %esi
+; X86-NEXT:    movl %esi, %ebp
+; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
+; X86-NEXT:    sbbl %ebx, (%esp) # 4-byte Folded Spill
 ; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %ecx # 4-byte Reload
 ; X86-NEXT:    addl $-1, %ecx
-; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
-; X86-NEXT:    adcl $-1, %eax
+; X86-NEXT:    adcl $-1, %edx
+; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %ebx # 4-byte Reload
+; X86-NEXT:    adcl $-1, %ebx
 ; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %esi # 4-byte Reload
 ; X86-NEXT:    adcl $-1, %esi
-; X86-NEXT:    adcl $-1, %ebx
-; X86-NEXT:    movl %eax, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
-; X86-NEXT:    orl %ebx, %eax
-; X86-NEXT:    movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT:    movl %edx, %edi
 ; X86-NEXT:    movl %esi, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
-; X86-NEXT:    orl %esi, %ecx
-; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %esi # 4-byte Reload
-; X86-NEXT:    orl %eax, %ecx
-; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %ecx # 4-byte Reload
+; X86-NEXT:    orl %esi, %edi
+; X86-NEXT:    movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT:    movl %ebx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT:    orl %ebx, %ecx
+; X86-NEXT:    orl %edi, %ecx
 ; X86-NEXT:    jne .LBB4_4
 ; X86-NEXT:  # %bb.5:
-; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %edx # 4-byte Reload
-; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
-; X86-NEXT:    movl {{[0-9]+}}(%esp), %ebp
 ; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %ebx # 4-byte Reload
+; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %edx # 4-byte Reload
+; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %ecx # 4-byte Reload
+; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %edi # 4-byte Reload
 ; X86-NEXT:  .LBB4_7: # %udiv-loop-exit
-; X86-NEXT:    shldl $1, %edx, %edi
-; X86-NEXT:    orl %esi, %edi
-; X86-NEXT:    shldl $1, %eax, %edx
-; X86-NEXT:    orl %esi, %edx
-; X86-NEXT:    movl %esi, %ecx
+; X86-NEXT:    shldl $1, %ebx, %edx
+; X86-NEXT:    orl %ecx, %edx
+; X86-NEXT:    shldl $1, %eax, %ebx
+; X86-NEXT:    orl %ecx, %ebx
 ; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %esi # 4-byte Reload
 ; X86-NEXT:    shldl $1, %esi, %eax
 ; X86-NEXT:    orl %ecx, %eax
 ; X86-NEXT:    addl %esi, %esi
-; X86-NEXT:    orl %ebx, %esi
+; X86-NEXT:    orl %edi, %esi
 ; X86-NEXT:  .LBB4_8: # %udiv-end
 ; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %ecx # 4-byte Reload
-; X86-NEXT:    xorl %ecx, %edi
 ; X86-NEXT:    xorl %ecx, %edx
+; X86-NEXT:    xorl %ecx, %ebx
 ; X86-NEXT:    xorl %ecx, %eax
 ; X86-NEXT:    xorl %ecx, %esi
 ; X86-NEXT:    subl %ecx, %esi
 ; X86-NEXT:    movl %esi, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
 ; X86-NEXT:    sbbl %ecx, %eax
 ; X86-NEXT:    movl %eax, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT:    sbbl %ecx, %ebx
 ; X86-NEXT:    sbbl %ecx, %edx
-; X86-NEXT:    sbbl %ecx, %edi
-; X86-NEXT:    movl %edi, (%esp) # 4-byte Spill
-; X86-NEXT:    movl %esi, (%ebp)
-; X86-NEXT:    movl %eax, 4(%ebp)
-; X86-NEXT:    movl %edx, 8(%ebp)
-; X86-NEXT:    movl %edi, 12(%ebp)
-; X86-NEXT:    movl {{[0-9]+}}(%esp), %edi
+; X86-NEXT:    movl %edx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT:    movl {{[0-9]+}}(%esp), %ecx
+; X86-NEXT:    movl %esi, (%ecx)
+; X86-NEXT:    movl %eax, 4(%ecx)
+; X86-NEXT:    movl %ebx, 8(%ecx)
+; X86-NEXT:    movl %edx, 12(%ecx)
+; X86-NEXT:    movl {{[0-9]+}}(%esp), %ecx
 ; X86-NEXT:    movl {{[0-9]+}}(%esp), %ebp
+; X86-NEXT:    movl %ebx, %edi
+; X86-NEXT:    mull %ecx
 ; X86-NEXT:    movl %edx, %ebx
-; X86-NEXT:    mull %edi
-; X86-NEXT:    movl %edx, %ecx
 ; X86-NEXT:    movl %eax, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
 ; X86-NEXT:    movl %esi, %eax
-; X86-NEXT:    mull %edi
-; X86-NEXT:    movl %eax, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
-; X86-NEXT:    movl %edx, %edi
-; X86-NEXT:    addl {{[-0-9]+}}(%e{{[sb]}}p), %edi # 4-byte Folded Reload
-; X86-NEXT:    adcl $0, %ecx
+; X86-NEXT:    mull %ecx
+; X86-NEXT:    movl %eax, (%esp) # 4-byte Spill
+; X86-NEXT:    movl %edx, %ecx
+; X86-NEXT:    addl {{[-0-9]+}}(%e{{[sb]}}p), %ecx # 4-byte Folded Reload
+; X86-NEXT:    adcl $0, %ebx
 ; X86-NEXT:    movl %esi, %eax
 ; X86-NEXT:    movl {{[0-9]+}}(%esp), %esi
 ; X86-NEXT:    mull %esi
-; X86-NEXT:    addl %edi, %eax
+; X86-NEXT:    addl %ecx, %eax
 ; X86-NEXT:    movl %eax, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
-; X86-NEXT:    adcl %ecx, %edx
-; X86-NEXT:    movl %edx, %edi
+; X86-NEXT:    adcl %ebx, %edx
+; X86-NEXT:    movl %edx, %ebx
 ; X86-NEXT:    setb %cl
 ; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %esi # 4-byte Reload
 ; X86-NEXT:    movl %esi, %eax
 ; X86-NEXT:    mull {{[0-9]+}}(%esp)
-; X86-NEXT:    addl %edi, %eax
+; X86-NEXT:    addl %ebx, %eax
 ; X86-NEXT:    movl %eax, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
 ; X86-NEXT:    movzbl %cl, %eax
 ; X86-NEXT:    adcl %eax, %edx
 ; X86-NEXT:    movl %edx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %ecx # 4-byte Reload
 ; X86-NEXT:    movl {{[0-9]+}}(%esp), %eax
-; X86-NEXT:    movl (%esp), %ecx # 4-byte Reload
 ; X86-NEXT:    imull %eax, %ecx
-; X86-NEXT:    mull %ebx
-; X86-NEXT:    movl %eax, (%esp) # 4-byte Spill
-; X86-NEXT:    imull {{[0-9]+}}(%esp), %ebx
-; X86-NEXT:    addl %edx, %ebx
-; X86-NEXT:    addl %ecx, %ebx
+; X86-NEXT:    mull %edi
+; X86-NEXT:    movl %eax, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT:    imull {{[0-9]+}}(%esp), %edi
+; X86-NEXT:    addl %edx, %edi
+; X86-NEXT:    addl %ecx, %edi
 ; X86-NEXT:    movl {{[0-9]+}}(%esp), %eax
 ; X86-NEXT:    movl %eax, %ecx
 ; X86-NEXT:    imull %esi, %ecx
@@ -568,12 +567,12 @@ define i128 @scalar_i128(i128 %x, i128 %y, ptr %divdst) nounwind {
 ; X86-NEXT:    mull %edx
 ; X86-NEXT:    addl %edx, %ebp
 ; X86-NEXT:    addl %ecx, %ebp
-; X86-NEXT:    addl (%esp), %eax # 4-byte Folded Reload
-; X86-NEXT:    adcl %ebx, %ebp
+; X86-NEXT:    addl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Folded Reload
+; X86-NEXT:    adcl %edi, %ebp
 ; X86-NEXT:    addl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Folded Reload
 ; X86-NEXT:    adcl {{[-0-9]+}}(%e{{[sb]}}p), %ebp # 4-byte Folded Reload
 ; X86-NEXT:    movl {{[0-9]+}}(%esp), %edx
-; X86-NEXT:    subl {{[-0-9]+}}(%e{{[sb]}}p), %edx # 4-byte Folded Reload
+; X86-NEXT:    subl (%esp), %edx # 4-byte Folded Reload
 ; X86-NEXT:    movl {{[0-9]+}}(%esp), %ecx
 ; X86-NEXT:    sbbl {{[-0-9]+}}(%e{{[sb]}}p), %ecx # 4-byte Folded Reload
 ; X86-NEXT:    movl {{[0-9]+}}(%esp), %esi
diff --git a/llvm/test/CodeGen/X86/fold-masked-merge.ll b/llvm/test/CodeGen/X86/fold-masked-merge.ll
index 135494ac25f8..b2614c5fe049 100644
--- a/llvm/test/CodeGen/X86/fold-masked-merge.ll
+++ b/llvm/test/CodeGen/X86/fold-masked-merge.ll
@@ -56,9 +56,7 @@ define i8 @masked_merge2(i8 %a0, i8 %a1, i8 %a2) {
 ; NOBMI-LABEL: masked_merge2:
 ; NOBMI:       # %bb.0:
 ; NOBMI-NEXT:    movl %esi, %eax
-; NOBMI-NEXT:    xorb %sil, %al
-; NOBMI-NEXT:    andb %dil, %al
-; NOBMI-NEXT:    xorb %sil, %al
+; NOBMI-NEXT:    # kill: def $al killed $al killed $eax
 ; NOBMI-NEXT:    retq
 ;
 ; BMI-LABEL: masked_merge2:
diff --git a/llvm/test/CodeGen/X86/freeze-binary.ll b/llvm/test/CodeGen/X86/freeze-binary.ll
index b212e9438e1b..c79da37988e4 100644
--- a/llvm/test/CodeGen/X86/freeze-binary.ll
+++ b/llvm/test/CodeGen/X86/freeze-binary.ll
@@ -459,8 +459,7 @@ define i32 @freeze_ashr(i32 %a0) nounwind {
 ; X64-LABEL: freeze_ashr:
 ; X64:       # %bb.0:
 ; X64-NEXT:    movl %edi, %eax
-; X64-NEXT:    sarl $3, %eax
-; X64-NEXT:    sarl $3, %eax
+; X64-NEXT:    sarl $6, %eax
 ; X64-NEXT:    retq
   %x = ashr i32 %a0, 3
   %y = freeze i32 %x
@@ -531,30 +530,12 @@ define i32 @freeze_ashr_outofrange(i32 %a0) nounwind {
 define <8 x i16> @freeze_ashr_vec(<8 x i16> %a0) nounwind {
 ; X86-LABEL: freeze_ashr_vec:
 ; X86:       # %bb.0:
-; X86-NEXT:    movdqa %xmm0, %xmm2
-; X86-NEXT:    psraw $1, %xmm2
-; X86-NEXT:    movdqa {{.*#+}} xmm1 = [65535,0,65535,0,65535,0,65535,0]
-; X86-NEXT:    movdqa %xmm1, %xmm3
-; X86-NEXT:    pandn %xmm2, %xmm3
-; X86-NEXT:    psraw $3, %xmm0
-; X86-NEXT:    pand %xmm1, %xmm0
-; X86-NEXT:    por %xmm3, %xmm0
-; X86-NEXT:    movdqa %xmm0, %xmm2
-; X86-NEXT:    psraw $3, %xmm2
-; X86-NEXT:    psraw $1, %xmm0
-; X86-NEXT:    pand %xmm1, %xmm0
-; X86-NEXT:    pandn %xmm2, %xmm1
-; X86-NEXT:    por %xmm1, %xmm0
+; X86-NEXT:    psraw $4, %xmm0
 ; X86-NEXT:    retl
 ;
 ; X64-LABEL: freeze_ashr_vec:
 ; X64:       # %bb.0:
-; X64-NEXT:    vpsraw $1, %xmm0, %xmm1
-; X64-NEXT:    vpsraw $3, %xmm0, %xmm0
-; X64-NEXT:    vpblendw {{.*#+}} xmm0 = xmm0[0],xmm1[1],xmm0[2],xmm1[3],xmm0[4],xmm1[5],xmm0[6],xmm1[7]
-; X64-NEXT:    vpsraw $3, %xmm0, %xmm1
-; X64-NEXT:    vpsraw $1, %xmm0, %xmm0
-; X64-NEXT:    vpblendw {{.*#+}} xmm0 = xmm0[0],xmm1[1],xmm0[2],xmm1[3],xmm0[4],xmm1[5],xmm0[6],xmm1[7]
+; X64-NEXT:    vpsraw $4, %xmm0, %xmm0
 ; X64-NEXT:    retq
   %x = ashr <8 x i16> %a0, <i16 3, i16 1, i16 3, i16 1, i16 3, i16 1, i16 3, i16 1>
   %y = freeze <8 x i16> %x
@@ -592,8 +573,7 @@ define i32 @freeze_lshr(i32 %a0) nounwind {
 ; X64-LABEL: freeze_lshr:
 ; X64:       # %bb.0:
 ; X64-NEXT:    movl %edi, %eax
-; X64-NEXT:    shrl $2, %eax
-; X64-NEXT:    shrl %eax
+; X64-NEXT:    shrl $3, %eax
 ; X64-NEXT:    retq
   %x = lshr i32 %a0, 2
   %y = freeze i32 %x
@@ -664,30 +644,12 @@ define i32 @freeze_lshr_outofrange(i32 %a0) nounwind {
 define <8 x i16> @freeze_lshr_vec(<8 x i16> %a0) nounwind {
 ; X86-LABEL: freeze_lshr_vec:
 ; X86:       # %bb.0:
-; X86-NEXT:    movdqa %xmm0, %xmm2
-; X86-NEXT:    psrlw $1, %xmm2
-; X86-NEXT:    movdqa {{.*#+}} xmm1 = [65535,0,65535,0,65535,0,65535,0]
-; X86-NEXT:    movdqa %xmm1, %xmm3
-; X86-NEXT:    pandn %xmm2, %xmm3
-; X86-NEXT:    psrlw $2, %xmm0
-; X86-NEXT:    pand %xmm1, %xmm0
-; X86-NEXT:    por %xmm3, %xmm0
-; X86-NEXT:    movdqa %xmm0, %xmm2
-; X86-NEXT:    psrlw $2, %xmm2
-; X86-NEXT:    psrlw $1, %xmm0
-; X86-NEXT:    pand %xmm1, %xmm0
-; X86-NEXT:    pandn %xmm2, %xmm1
-; X86-NEXT:    por %xmm1, %xmm0
+; X86-NEXT:    psrlw $3, %xmm0
 ; X86-NEXT:    retl
 ;
 ; X64-LABEL: freeze_lshr_vec:
 ; X64:       # %bb.0:
-; X64-NEXT:    vpsrlw $1, %xmm0, %xmm1
-; X64-NEXT:    vpsrlw $2, %xmm0, %xmm0
-; X64-NEXT:    vpblendw {{.*#+}} xmm0 = xmm0[0],xmm1[1],xmm0[2],xmm1[3],xmm0[4],xmm1[5],xmm0[6],xmm1[7]
-; X64-NEXT:    vpsrlw $2, %xmm0, %xmm1
-; X64-NEXT:    vpsrlw $1, %xmm0, %xmm0
-; X64-NEXT:    vpblendw {{.*#+}} xmm0 = xmm0[0],xmm1[1],xmm0[2],xmm1[3],xmm0[4],xmm1[5],xmm0[6],xmm1[7]
+; X64-NEXT:    vpsrlw $3, %xmm0, %xmm0
 ; X64-NEXT:    retq
   %x = lshr <8 x i16> %a0, <i16 2, i16 1, i16 2, i16 1, i16 2, i16 1, i16 2, i16 1>
   %y = freeze <8 x i16> %x
diff --git a/llvm/test/CodeGen/X86/freeze-combine.ll b/llvm/test/CodeGen/X86/freeze-combine.ll
index b037a6d9a1b9..1cfb8627a4dd 100644
--- a/llvm/test/CodeGen/X86/freeze-combine.ll
+++ b/llvm/test/CodeGen/X86/freeze-combine.ll
@@ -3,9 +3,9 @@
 define i32 @const() {
   ; CHECK-LABEL: name: const
   ; CHECK: bb.0 (%ir-block.0):
-  ; CHECK:   [[MOV32ri:%[0-9]+]]:gr32 = MOV32ri 1
-  ; CHECK:   $eax = COPY [[MOV32ri]]
-  ; CHECK:   RET 0, $eax
+  ; CHECK-NEXT:   [[MOV32ri:%[0-9]+]]:gr32 = MOV32ri 1
+  ; CHECK-NEXT:   $eax = COPY [[MOV32ri]]
+  ; CHECK-NEXT:   RET 0, $eax
   %y = freeze i32 1
   ret i32 %y
 }
@@ -13,11 +13,11 @@ define i32 @const() {
 define i32 @fold(i32 %x) {
   ; CHECK-LABEL: name: fold
   ; CHECK: bb.0 (%ir-block.0):
-  ; CHECK:   liveins: $edi
-  ; CHECK:   [[COPY:%[0-9]+]]:gr32 = COPY $edi
-  ; CHECK:   [[COPY1:%[0-9]+]]:gr32 = COPY [[COPY]]
-  ; CHECK:   $eax = COPY [[COPY1]]
-  ; CHECK:   RET 0, $eax
+  ; CHECK-NEXT:   liveins: $edi
+  ; CHECK-NEXT: {{  $}}
+  ; CHECK-NEXT:   [[COPY:%[0-9]+]]:gr32 = COPY $edi
+  ; CHECK-NEXT:   $eax = COPY [[COPY]]
+  ; CHECK-NEXT:   RET 0, $eax
   %y = freeze i32 %x
   %z = freeze i32 %y
   ret i32 %z
diff --git a/llvm/test/CodeGen/X86/freeze-vector.ll b/llvm/test/CodeGen/X86/freeze-vector.ll
index ee7f4aea02c0..fe240286462e 100644
--- a/llvm/test/CodeGen/X86/freeze-vector.ll
+++ b/llvm/test/CodeGen/X86/freeze-vector.ll
@@ -672,3 +672,23 @@ define void @pr59677(i32 %x, ptr %out) nounwind {
   ret void
 }
 declare <4 x float> @llvm.sin.v4f32(<4 x float>)
+
+; Test that we can eliminate freeze by changing the BUILD_VECTOR to a splat
+; zero vector.
+define void @freeze_buildvector_not_simple_type(ptr %dst) nounwind {
+; X86-LABEL: freeze_buildvector_not_simple_type:
+; X86:       # %bb.0:
+; X86-NEXT:    movl {{[0-9]+}}(%esp), %eax
+; X86-NEXT:    movb $0, 4(%eax)
+; X86-NEXT:    movl $0, (%eax)
+; X86-NEXT:    retl
+;
+; X64-LABEL: freeze_buildvector_not_simple_type:
+; X64:       # %bb.0:
+; X64-NEXT:    movb $0, 4(%rdi)
+; X64-NEXT:    movl $0, (%rdi)
+; X64-NEXT:    retq
+  %i0 = freeze <5 x i8> <i8 poison, i8 0, i8 0, i8 undef, i8 0>
+  store <5 x i8> %i0, ptr %dst
+  ret void
+}
diff --git a/llvm/test/CodeGen/X86/gfni-funnel-shifts.ll b/llvm/test/CodeGen/X86/gfni-funnel-shifts.ll
index 0c341dc63a9e..afe0ebb9dcb4 100644
--- a/llvm/test/CodeGen/X86/gfni-funnel-shifts.ll
+++ b/llvm/test/CodeGen/X86/gfni-funnel-shifts.ll
@@ -522,17 +522,17 @@ declare <16 x i8> @llvm.fshl.v16i8(<16 x i8>, <16 x i8>, <16 x i8>)
 define <16 x i8> @splatconstant_fshr_v16i8(<16 x i8> %a, <16 x i8> %b) nounwind {
 ; GFNISSE-LABEL: splatconstant_fshr_v16i8:
 ; GFNISSE:       # %bb.0:
+; GFNISSE-NEXT:    paddb %xmm0, %xmm0
 ; GFNISSE-NEXT:    psrlw $7, %xmm1
 ; GFNISSE-NEXT:    pand {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm1
-; GFNISSE-NEXT:    paddb %xmm0, %xmm0
 ; GFNISSE-NEXT:    por %xmm1, %xmm0
 ; GFNISSE-NEXT:    retq
 ;
 ; GFNIAVX1OR2-LABEL: splatconstant_fshr_v16i8:
 ; GFNIAVX1OR2:       # %bb.0:
+; GFNIAVX1OR2-NEXT:    vpaddb %xmm0, %xmm0, %xmm0
 ; GFNIAVX1OR2-NEXT:    vpsrlw $7, %xmm1, %xmm1
 ; GFNIAVX1OR2-NEXT:    vpand {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm1, %xmm1
-; GFNIAVX1OR2-NEXT:    vpaddb %xmm0, %xmm0, %xmm0
 ; GFNIAVX1OR2-NEXT:    vpor %xmm1, %xmm0, %xmm0
 ; GFNIAVX1OR2-NEXT:    retq
 ;
diff --git a/llvm/test/CodeGen/X86/gfni-rotates.ll b/llvm/test/CodeGen/X86/gfni-rotates.ll
index 7ab8300b269a..96aff5b2af31 100644
--- a/llvm/test/CodeGen/X86/gfni-rotates.ll
+++ b/llvm/test/CodeGen/X86/gfni-rotates.ll
@@ -421,18 +421,18 @@ define <16 x i8> @splatconstant_rotr_v16i8(<16 x i8> %a) nounwind {
 ; GFNISSE-LABEL: splatconstant_rotr_v16i8:
 ; GFNISSE:       # %bb.0:
 ; GFNISSE-NEXT:    movdqa %xmm0, %xmm1
-; GFNISSE-NEXT:    psrlw $7, %xmm1
-; GFNISSE-NEXT:    pand {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm1
-; GFNISSE-NEXT:    paddb %xmm0, %xmm0
+; GFNISSE-NEXT:    paddb %xmm0, %xmm1
+; GFNISSE-NEXT:    psrlw $7, %xmm0
+; GFNISSE-NEXT:    pand {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0
 ; GFNISSE-NEXT:    por %xmm1, %xmm0
 ; GFNISSE-NEXT:    retq
 ;
 ; GFNIAVX1OR2-LABEL: splatconstant_rotr_v16i8:
 ; GFNIAVX1OR2:       # %bb.0:
-; GFNIAVX1OR2-NEXT:    vpsrlw $7, %xmm0, %xmm1
-; GFNIAVX1OR2-NEXT:    vpand {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm1, %xmm1
-; GFNIAVX1OR2-NEXT:    vpaddb %xmm0, %xmm0, %xmm0
-; GFNIAVX1OR2-NEXT:    vpor %xmm1, %xmm0, %xmm0
+; GFNIAVX1OR2-NEXT:    vpaddb %xmm0, %xmm0, %xmm1
+; GFNIAVX1OR2-NEXT:    vpsrlw $7, %xmm0, %xmm0
+; GFNIAVX1OR2-NEXT:    vpand {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0, %xmm0
+; GFNIAVX1OR2-NEXT:    vpor %xmm0, %xmm1, %xmm0
 ; GFNIAVX1OR2-NEXT:    retq
 ;
 ; GFNIAVX512-LABEL: splatconstant_rotr_v16i8:
diff --git a/llvm/test/CodeGen/X86/known-never-zero.ll b/llvm/test/CodeGen/X86/known-never-zero.ll
index 39d02f9112f4..2f780e3c6fe1 100644
--- a/llvm/test/CodeGen/X86/known-never-zero.ll
+++ b/llvm/test/CodeGen/X86/known-never-zero.ll
@@ -676,12 +676,13 @@ define i32 @rotr_known_nonzero(i32 %xx, i32 %y) {
 ; X64:       # %bb.0:
 ; X64-NEXT:    movl %esi, %ecx
 ; X64-NEXT:    orl $256, %edi # imm = 0x100
+; X64-NEXT:    movl %edi, %eax
 ; X64-NEXT:    # kill: def $cl killed $cl killed $ecx
-; X64-NEXT:    rorl %cl, %edi
+; X64-NEXT:    rorl %cl, %eax
 ; X64-NEXT:    testl %edi, %edi
 ; X64-NEXT:    je .LBB22_1
 ; X64-NEXT:  # %bb.2: # %cond.false
-; X64-NEXT:    rep bsfl %edi, %eax
+; X64-NEXT:    rep bsfl %eax, %eax
 ; X64-NEXT:    retq
 ; X64-NEXT:  .LBB22_1:
 ; X64-NEXT:    movl $32, %eax
@@ -713,12 +714,13 @@ define i32 @rotr_maybe_zero(i32 %x, i32 %y) {
 ; X64-LABEL: rotr_maybe_zero:
 ; X64:       # %bb.0:
 ; X64-NEXT:    movl %esi, %ecx
+; X64-NEXT:    movl %edi, %eax
 ; X64-NEXT:    # kill: def $cl killed $cl killed $ecx
-; X64-NEXT:    rorl %cl, %edi
+; X64-NEXT:    rorl %cl, %eax
 ; X64-NEXT:    testl %edi, %edi
 ; X64-NEXT:    je .LBB23_1
 ; X64-NEXT:  # %bb.2: # %cond.false
-; X64-NEXT:    rep bsfl %edi, %eax
+; X64-NEXT:    rep bsfl %eax, %eax
 ; X64-NEXT:    retq
 ; X64-NEXT:  .LBB23_1:
 ; X64-NEXT:    movl $32, %eax
@@ -773,12 +775,13 @@ define i32 @rotr_with_fshr_maybe_zero(i32 %x, i32 %y) {
 ; X64-LABEL: rotr_with_fshr_maybe_zero:
 ; X64:       # %bb.0:
 ; X64-NEXT:    movl %esi, %ecx
+; X64-NEXT:    movl %edi, %eax
 ; X64-NEXT:    # kill: def $cl killed $cl killed $ecx
-; X64-NEXT:    rorl %cl, %edi
+; X64-NEXT:    rorl %cl, %eax
 ; X64-NEXT:    testl %edi, %edi
 ; X64-NEXT:    je .LBB25_1
 ; X64-NEXT:  # %bb.2: # %cond.false
-; X64-NEXT:    rep bsfl %edi, %eax
+; X64-NEXT:    rep bsfl %eax, %eax
 ; X64-NEXT:    retq
 ; X64-NEXT:  .LBB25_1:
 ; X64-NEXT:    movl $32, %eax
@@ -808,12 +811,13 @@ define i32 @rotl_known_nonzero(i32 %xx, i32 %y) {
 ; X64:       # %bb.0:
 ; X64-NEXT:    movl %esi, %ecx
 ; X64-NEXT:    orl $256, %edi # imm = 0x100
+; X64-NEXT:    movl %edi, %eax
 ; X64-NEXT:    # kill: def $cl killed $cl killed $ecx
-; X64-NEXT:    roll %cl, %edi
+; X64-NEXT:    roll %cl, %eax
 ; X64-NEXT:    testl %edi, %edi
 ; X64-NEXT:    je .LBB26_1
 ; X64-NEXT:  # %bb.2: # %cond.false
-; X64-NEXT:    rep bsfl %edi, %eax
+; X64-NEXT:    rep bsfl %eax, %eax
 ; X64-NEXT:    retq
 ; X64-NEXT:  .LBB26_1:
 ; X64-NEXT:    movl $32, %eax
@@ -845,12 +849,13 @@ define i32 @rotl_maybe_zero(i32 %x, i32 %y) {
 ; X64-LABEL: rotl_maybe_zero:
 ; X64:       # %bb.0:
 ; X64-NEXT:    movl %esi, %ecx
+; X64-NEXT:    movl %edi, %eax
 ; X64-NEXT:    # kill: def $cl killed $cl killed $ecx
-; X64-NEXT:    roll %cl, %edi
+; X64-NEXT:    roll %cl, %eax
 ; X64-NEXT:    testl %edi, %edi
 ; X64-NEXT:    je .LBB27_1
 ; X64-NEXT:  # %bb.2: # %cond.false
-; X64-NEXT:    rep bsfl %edi, %eax
+; X64-NEXT:    rep bsfl %eax, %eax
 ; X64-NEXT:    retq
 ; X64-NEXT:  .LBB27_1:
 ; X64-NEXT:    movl $32, %eax
@@ -905,12 +910,13 @@ define i32 @rotl_with_fshl_maybe_zero(i32 %x, i32 %y) {
 ; X64-LABEL: rotl_with_fshl_maybe_zero:
 ; X64:       # %bb.0:
 ; X64-NEXT:    movl %esi, %ecx
+; X64-NEXT:    movl %edi, %eax
 ; X64-NEXT:    # kill: def $cl killed $cl killed $ecx
-; X64-NEXT:    roll %cl, %edi
+; X64-NEXT:    roll %cl, %eax
 ; X64-NEXT:    testl %edi, %edi
 ; X64-NEXT:    je .LBB29_1
 ; X64-NEXT:  # %bb.2: # %cond.false
-; X64-NEXT:    rep bsfl %edi, %eax
+; X64-NEXT:    rep bsfl %eax, %eax
 ; X64-NEXT:    retq
 ; X64-NEXT:  .LBB29_1:
 ; X64-NEXT:    movl $32, %eax
diff --git a/llvm/test/CodeGen/X86/memcmp-minsize-x32.ll b/llvm/test/CodeGen/X86/memcmp-minsize-x32.ll
index ae1320f8b086..200a8184d4bd 100644
--- a/llvm/test/CodeGen/X86/memcmp-minsize-x32.ll
+++ b/llvm/test/CodeGen/X86/memcmp-minsize-x32.ll
@@ -1,6 +1,6 @@
 ; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
-; RUN: llc < %s -mtriple=i686-unknown-unknown -mattr=cmov | FileCheck %s --check-prefix=X86 --check-prefix=X86-NOSSE
-; RUN: llc < %s -mtriple=i686-unknown-unknown -mattr=+sse2 | FileCheck %s --check-prefix=X86 --check-prefix=X86-SSE2
+; RUN: llc < %s -mtriple=i686-unknown-unknown -mattr=cmov | FileCheck %s --check-prefixes=X86,X86-NOSSE
+; RUN: llc < %s -mtriple=i686-unknown-unknown -mattr=+sse2 | FileCheck %s --check-prefixes=X86,X86-SSE2
 
 ; This tests codegen time inlining/optimization of memcmp
 ; rdar://6480398
diff --git a/llvm/test/CodeGen/X86/memcmp-minsize.ll b/llvm/test/CodeGen/X86/memcmp-minsize.ll
index 544d1c49f26b..9c20f3e0cdef 100644
--- a/llvm/test/CodeGen/X86/memcmp-minsize.ll
+++ b/llvm/test/CodeGen/X86/memcmp-minsize.ll
@@ -1,7 +1,7 @@
 ; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
-; RUN: llc < %s -mtriple=x86_64-unknown-unknown | FileCheck %s --check-prefix=X64 --check-prefix=X64-SSE2
-; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=avx | FileCheck %s --check-prefix=X64 --check-prefix=X64-AVX --check-prefix=X64-AVX1
-; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=avx2 | FileCheck %s --check-prefix=X64 --check-prefix=X64-AVX --check-prefix=X64-AVX2
+; RUN: llc < %s -mtriple=x86_64-unknown-unknown | FileCheck %s --check-prefixes=X64,X64-SSE2
+; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=avx | FileCheck %s --check-prefixes=X64,X64-AVX,X64-AVX1
+; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=avx2 | FileCheck %s --check-prefixes=X64,X64-AVX,X64-AVX2
 
 ; This tests codegen time inlining/optimization of memcmp
 ; rdar://6480398
diff --git a/llvm/test/CodeGen/X86/memcmp-optsize-x32.ll b/llvm/test/CodeGen/X86/memcmp-optsize-x32.ll
index 762691151f4b..3db6ae8b76b2 100644
--- a/llvm/test/CodeGen/X86/memcmp-optsize-x32.ll
+++ b/llvm/test/CodeGen/X86/memcmp-optsize-x32.ll
@@ -1,6 +1,6 @@
 ; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
-; RUN: llc < %s -mtriple=i686-unknown-unknown -mattr=cmov | FileCheck %s --check-prefix=X86 --check-prefix=X86-NOSSE
-; RUN: llc < %s -mtriple=i686-unknown-unknown -mattr=+sse2 | FileCheck %s --check-prefix=X86 --check-prefix=X86-SSE2
+; RUN: llc < %s -mtriple=i686-unknown-unknown -mattr=cmov | FileCheck %s --check-prefixes=X86,X86-NOSSE
+; RUN: llc < %s -mtriple=i686-unknown-unknown -mattr=+sse2 | FileCheck %s --check-prefixes=X86,X86-SSE2
 
 ; This tests codegen time inlining/optimization of memcmp
 ; rdar://6480398
diff --git a/llvm/test/CodeGen/X86/memcmp-optsize.ll b/llvm/test/CodeGen/X86/memcmp-optsize.ll
index c0c7b98d471c..edd61641ad2a 100644
--- a/llvm/test/CodeGen/X86/memcmp-optsize.ll
+++ b/llvm/test/CodeGen/X86/memcmp-optsize.ll
@@ -1,7 +1,7 @@
 ; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
-; RUN: llc < %s -mtriple=x86_64-unknown-unknown | FileCheck %s --check-prefix=X64 --check-prefix=X64-SSE2
-; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=avx | FileCheck %s --check-prefix=X64 --check-prefix=X64-AVX --check-prefix=X64-AVX1
-; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=avx2 | FileCheck %s --check-prefix=X64 --check-prefix=X64-AVX --check-prefix=X64-AVX2
+; RUN: llc < %s -mtriple=x86_64-unknown-unknown | FileCheck %s --check-prefixes=X64,X64-SSE2
+; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=avx | FileCheck %s --check-prefixes=X64,X64-AVX,X64-AVX1
+; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=avx2 | FileCheck %s --check-prefixes=X64,X64-AVX,X64-AVX2
 
 ; This tests codegen time inlining/optimization of memcmp
 ; rdar://6480398
diff --git a/llvm/test/CodeGen/X86/memcmp-pgso-x32.ll b/llvm/test/CodeGen/X86/memcmp-pgso-x32.ll
index cb45fd3ebb90..1c301da26bea 100644
--- a/llvm/test/CodeGen/X86/memcmp-pgso-x32.ll
+++ b/llvm/test/CodeGen/X86/memcmp-pgso-x32.ll
@@ -1,6 +1,6 @@
 ; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
-; RUN: llc < %s -mtriple=i686-unknown-unknown -mattr=cmov | FileCheck %s --check-prefix=X86 --check-prefix=X86-NOSSE
-; RUN: llc < %s -mtriple=i686-unknown-unknown -mattr=+sse2 | FileCheck %s --check-prefix=X86 --check-prefix=X86-SSE2
+; RUN: llc < %s -mtriple=i686-unknown-unknown -mattr=cmov | FileCheck %s --check-prefixes=X86,X86-NOSSE
+; RUN: llc < %s -mtriple=i686-unknown-unknown -mattr=+sse2 | FileCheck %s --check-prefixes=X86,X86-SSE2
 
 ; This tests codegen time inlining/optimization of memcmp
 ; rdar://6480398
diff --git a/llvm/test/CodeGen/X86/memcmp-pgso.ll b/llvm/test/CodeGen/X86/memcmp-pgso.ll
index 720344a22e43..1ee3317b9c96 100644
--- a/llvm/test/CodeGen/X86/memcmp-pgso.ll
+++ b/llvm/test/CodeGen/X86/memcmp-pgso.ll
@@ -1,7 +1,7 @@
 ; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
-; RUN: llc < %s -mtriple=x86_64-unknown-unknown | FileCheck %s --check-prefix=X64 --check-prefix=X64-SSE2
-; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=avx | FileCheck %s --check-prefix=X64 --check-prefix=X64-AVX --check-prefix=X64-AVX1
-; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=avx2 | FileCheck %s --check-prefix=X64 --check-prefix=X64-AVX --check-prefix=X64-AVX2
+; RUN: llc < %s -mtriple=x86_64-unknown-unknown | FileCheck %s --check-prefixes=X64,X64-SSE2
+; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=avx | FileCheck %s --check-prefixes=X64,X64-AVX,X64-AVX1
+; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=avx2 | FileCheck %s --check-prefixes=X64,X64-AVX,X64-AVX2
 
 ; This tests codegen time inlining/optimization of memcmp
 ; rdar://6480398
diff --git a/llvm/test/CodeGen/X86/midpoint-int-vec-128.ll b/llvm/test/CodeGen/X86/midpoint-int-vec-128.ll
index d3cced3233ea..5a6375e08bca 100644
--- a/llvm/test/CodeGen/X86/midpoint-int-vec-128.ll
+++ b/llvm/test/CodeGen/X86/midpoint-int-vec-128.ll
@@ -893,27 +893,26 @@ define <2 x i64> @vec128_i64_signed_reg_reg(<2 x i64> %a1, <2 x i64> %a2) nounwi
 ; SSE41-NEXT:    pcmpeqd %xmm3, %xmm0
 ; SSE41-NEXT:    pshufd {{.*#+}} xmm3 = xmm0[1,1,3,3]
 ; SSE41-NEXT:    pand %xmm5, %xmm3
-; SSE41-NEXT:    pshufd {{.*#+}} xmm6 = xmm4[1,1,3,3]
-; SSE41-NEXT:    por %xmm3, %xmm6
-; SSE41-NEXT:    por {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm6
-; SSE41-NEXT:    pand %xmm5, %xmm0
-; SSE41-NEXT:    por %xmm4, %xmm0
-; SSE41-NEXT:    movdqa %xmm2, %xmm3
-; SSE41-NEXT:    psubq %xmm1, %xmm3
+; SSE41-NEXT:    pshufd {{.*#+}} xmm0 = xmm4[1,1,3,3]
+; SSE41-NEXT:    por %xmm3, %xmm0
+; SSE41-NEXT:    pmovsxbq {{.*#+}} xmm3 = [1,1]
+; SSE41-NEXT:    por %xmm0, %xmm3
+; SSE41-NEXT:    movdqa %xmm2, %xmm4
+; SSE41-NEXT:    psubq %xmm1, %xmm4
 ; SSE41-NEXT:    psubq %xmm2, %xmm1
-; SSE41-NEXT:    blendvpd %xmm0, %xmm3, %xmm1
+; SSE41-NEXT:    blendvpd %xmm0, %xmm4, %xmm1
 ; SSE41-NEXT:    movapd %xmm1, %xmm0
 ; SSE41-NEXT:    psrlq $1, %xmm0
 ; SSE41-NEXT:    psrlq $33, %xmm1
-; SSE41-NEXT:    pmuludq %xmm6, %xmm1
-; SSE41-NEXT:    movdqa %xmm6, %xmm3
-; SSE41-NEXT:    psrlq $32, %xmm3
-; SSE41-NEXT:    pmuludq %xmm0, %xmm3
-; SSE41-NEXT:    paddq %xmm1, %xmm3
-; SSE41-NEXT:    psllq $32, %xmm3
-; SSE41-NEXT:    pmuludq %xmm6, %xmm0
+; SSE41-NEXT:    pmuludq %xmm3, %xmm1
+; SSE41-NEXT:    movdqa %xmm3, %xmm4
+; SSE41-NEXT:    psrlq $32, %xmm4
+; SSE41-NEXT:    pmuludq %xmm0, %xmm4
+; SSE41-NEXT:    paddq %xmm1, %xmm4
+; SSE41-NEXT:    psllq $32, %xmm4
+; SSE41-NEXT:    pmuludq %xmm3, %xmm0
 ; SSE41-NEXT:    paddq %xmm2, %xmm0
-; SSE41-NEXT:    paddq %xmm3, %xmm0
+; SSE41-NEXT:    paddq %xmm4, %xmm0
 ; SSE41-NEXT:    retq
 ;
 ; AVX-LABEL: vec128_i64_signed_reg_reg:
@@ -1077,27 +1076,26 @@ define <2 x i64> @vec128_i64_unsigned_reg_reg(<2 x i64> %a1, <2 x i64> %a2) noun
 ; SSE41-NEXT:    pcmpeqd %xmm3, %xmm0
 ; SSE41-NEXT:    pshufd {{.*#+}} xmm3 = xmm0[1,1,3,3]
 ; SSE41-NEXT:    pand %xmm5, %xmm3
-; SSE41-NEXT:    pshufd {{.*#+}} xmm6 = xmm4[1,1,3,3]
-; SSE41-NEXT:    por %xmm3, %xmm6
-; SSE41-NEXT:    por {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm6
-; SSE41-NEXT:    pand %xmm5, %xmm0
-; SSE41-NEXT:    por %xmm4, %xmm0
-; SSE41-NEXT:    movdqa %xmm2, %xmm3
-; SSE41-NEXT:    psubq %xmm1, %xmm3
+; SSE41-NEXT:    pshufd {{.*#+}} xmm0 = xmm4[1,1,3,3]
+; SSE41-NEXT:    por %xmm3, %xmm0
+; SSE41-NEXT:    pmovsxbq {{.*#+}} xmm3 = [1,1]
+; SSE41-NEXT:    por %xmm0, %xmm3
+; SSE41-NEXT:    movdqa %xmm2, %xmm4
+; SSE41-NEXT:    psubq %xmm1, %xmm4
 ; SSE41-NEXT:    psubq %xmm2, %xmm1
-; SSE41-NEXT:    blendvpd %xmm0, %xmm3, %xmm1
+; SSE41-NEXT:    blendvpd %xmm0, %xmm4, %xmm1
 ; SSE41-NEXT:    movapd %xmm1, %xmm0
 ; SSE41-NEXT:    psrlq $1, %xmm0
 ; SSE41-NEXT:    psrlq $33, %xmm1
-; SSE41-NEXT:    pmuludq %xmm6, %xmm1
-; SSE41-NEXT:    movdqa %xmm6, %xmm3
-; SSE41-NEXT:    psrlq $32, %xmm3
-; SSE41-NEXT:    pmuludq %xmm0, %xmm3
-; SSE41-NEXT:    paddq %xmm1, %xmm3
-; SSE41-NEXT:    psllq $32, %xmm3
-; SSE41-NEXT:    pmuludq %xmm6, %xmm0
+; SSE41-NEXT:    pmuludq %xmm3, %xmm1
+; SSE41-NEXT:    movdqa %xmm3, %xmm4
+; SSE41-NEXT:    psrlq $32, %xmm4
+; SSE41-NEXT:    pmuludq %xmm0, %xmm4
+; SSE41-NEXT:    paddq %xmm1, %xmm4
+; SSE41-NEXT:    psllq $32, %xmm4
+; SSE41-NEXT:    pmuludq %xmm3, %xmm0
 ; SSE41-NEXT:    paddq %xmm2, %xmm0
-; SSE41-NEXT:    paddq %xmm3, %xmm0
+; SSE41-NEXT:    paddq %xmm4, %xmm0
 ; SSE41-NEXT:    retq
 ;
 ; AVX1-LABEL: vec128_i64_unsigned_reg_reg:
@@ -1993,14 +1991,14 @@ define <8 x i16> @vec128_i16_unsigned_reg_reg(<8 x i16> %a1, <8 x i16> %a2) noun
 ;
 ; AVX512VL-FALLBACK-LABEL: vec128_i16_unsigned_reg_reg:
 ; AVX512VL-FALLBACK:       # %bb.0:
-; AVX512VL-FALLBACK-NEXT:    vpminuw %xmm1, %xmm0, %xmm2
-; AVX512VL-FALLBACK-NEXT:    vpmaxuw %xmm1, %xmm0, %xmm1
-; AVX512VL-FALLBACK-NEXT:    vpsubw %xmm2, %xmm1, %xmm1
-; AVX512VL-FALLBACK-NEXT:    vpsrlw $1, %xmm1, %xmm1
-; AVX512VL-FALLBACK-NEXT:    vpcmpeqw %xmm2, %xmm0, %xmm2
-; AVX512VL-FALLBACK-NEXT:    vpternlogq $15, %xmm2, %xmm2, %xmm2
-; AVX512VL-FALLBACK-NEXT:    vpxor %xmm2, %xmm1, %xmm1
-; AVX512VL-FALLBACK-NEXT:    vpsubw %xmm2, %xmm1, %xmm1
+; AVX512VL-FALLBACK-NEXT:    vpmaxuw %xmm1, %xmm0, %xmm2
+; AVX512VL-FALLBACK-NEXT:    vpminuw %xmm1, %xmm0, %xmm1
+; AVX512VL-FALLBACK-NEXT:    vpsubw %xmm1, %xmm2, %xmm2
+; AVX512VL-FALLBACK-NEXT:    vpsrlw $1, %xmm2, %xmm2
+; AVX512VL-FALLBACK-NEXT:    vpcmpeqw %xmm1, %xmm0, %xmm1
+; AVX512VL-FALLBACK-NEXT:    vpternlogq $15, %xmm1, %xmm1, %xmm1
+; AVX512VL-FALLBACK-NEXT:    vpxor %xmm1, %xmm2, %xmm2
+; AVX512VL-FALLBACK-NEXT:    vpsubw %xmm1, %xmm2, %xmm1
 ; AVX512VL-FALLBACK-NEXT:    vpaddw %xmm0, %xmm1, %xmm0
 ; AVX512VL-FALLBACK-NEXT:    retq
 ;
@@ -2786,14 +2784,14 @@ define <16 x i8> @vec128_i8_unsigned_reg_reg(<16 x i8> %a1, <16 x i8> %a2) nounw
 ;
 ; AVX512VL-FALLBACK-LABEL: vec128_i8_unsigned_reg_reg:
 ; AVX512VL-FALLBACK:       # %bb.0:
-; AVX512VL-FALLBACK-NEXT:    vpminub %xmm1, %xmm0, %xmm2
-; AVX512VL-FALLBACK-NEXT:    vpmaxub %xmm1, %xmm0, %xmm1
-; AVX512VL-FALLBACK-NEXT:    vpsubb %xmm2, %xmm1, %xmm1
-; AVX512VL-FALLBACK-NEXT:    vpsrlw $1, %xmm1, %xmm1
-; AVX512VL-FALLBACK-NEXT:    vpcmpeqb %xmm2, %xmm0, %xmm2
-; AVX512VL-FALLBACK-NEXT:    vpternlogq $15, %xmm2, %xmm2, %xmm2
-; AVX512VL-FALLBACK-NEXT:    vpternlogd $108, {{\.?LCPI[0-9]+_[0-9]+}}(%rip){1to4}, %xmm2, %xmm1
-; AVX512VL-FALLBACK-NEXT:    vpsubb %xmm2, %xmm1, %xmm1
+; AVX512VL-FALLBACK-NEXT:    vpmaxub %xmm1, %xmm0, %xmm2
+; AVX512VL-FALLBACK-NEXT:    vpminub %xmm1, %xmm0, %xmm1
+; AVX512VL-FALLBACK-NEXT:    vpsubb %xmm1, %xmm2, %xmm2
+; AVX512VL-FALLBACK-NEXT:    vpsrlw $1, %xmm2, %xmm2
+; AVX512VL-FALLBACK-NEXT:    vpcmpeqb %xmm1, %xmm0, %xmm1
+; AVX512VL-FALLBACK-NEXT:    vpternlogq $15, %xmm1, %xmm1, %xmm1
+; AVX512VL-FALLBACK-NEXT:    vpternlogd $108, {{\.?LCPI[0-9]+_[0-9]+}}(%rip){1to4}, %xmm1, %xmm2
+; AVX512VL-FALLBACK-NEXT:    vpsubb %xmm1, %xmm2, %xmm1
 ; AVX512VL-FALLBACK-NEXT:    vpaddb %xmm0, %xmm1, %xmm0
 ; AVX512VL-FALLBACK-NEXT:    retq
 ;
diff --git a/llvm/test/CodeGen/X86/midpoint-int-vec-256.ll b/llvm/test/CodeGen/X86/midpoint-int-vec-256.ll
index cc08396ae8c7..e880a1acc9e8 100644
--- a/llvm/test/CodeGen/X86/midpoint-int-vec-256.ll
+++ b/llvm/test/CodeGen/X86/midpoint-int-vec-256.ll
@@ -1445,14 +1445,14 @@ define <16 x i16> @vec256_i16_unsigned_reg_reg(<16 x i16> %a1, <16 x i16> %a2) n
 ;
 ; AVX512VL-FALLBACK-LABEL: vec256_i16_unsigned_reg_reg:
 ; AVX512VL-FALLBACK:       # %bb.0:
-; AVX512VL-FALLBACK-NEXT:    vpminuw %ymm1, %ymm0, %ymm2
-; AVX512VL-FALLBACK-NEXT:    vpmaxuw %ymm1, %ymm0, %ymm1
-; AVX512VL-FALLBACK-NEXT:    vpsubw %ymm2, %ymm1, %ymm1
-; AVX512VL-FALLBACK-NEXT:    vpsrlw $1, %ymm1, %ymm1
-; AVX512VL-FALLBACK-NEXT:    vpcmpeqw %ymm2, %ymm0, %ymm2
-; AVX512VL-FALLBACK-NEXT:    vpternlogq $15, %ymm2, %ymm2, %ymm2
-; AVX512VL-FALLBACK-NEXT:    vpxor %ymm2, %ymm1, %ymm1
-; AVX512VL-FALLBACK-NEXT:    vpsubw %ymm2, %ymm1, %ymm1
+; AVX512VL-FALLBACK-NEXT:    vpmaxuw %ymm1, %ymm0, %ymm2
+; AVX512VL-FALLBACK-NEXT:    vpminuw %ymm1, %ymm0, %ymm1
+; AVX512VL-FALLBACK-NEXT:    vpsubw %ymm1, %ymm2, %ymm2
+; AVX512VL-FALLBACK-NEXT:    vpsrlw $1, %ymm2, %ymm2
+; AVX512VL-FALLBACK-NEXT:    vpcmpeqw %ymm1, %ymm0, %ymm1
+; AVX512VL-FALLBACK-NEXT:    vpternlogq $15, %ymm1, %ymm1, %ymm1
+; AVX512VL-FALLBACK-NEXT:    vpxor %ymm1, %ymm2, %ymm2
+; AVX512VL-FALLBACK-NEXT:    vpsubw %ymm1, %ymm2, %ymm1
 ; AVX512VL-FALLBACK-NEXT:    vpaddw %ymm0, %ymm1, %ymm0
 ; AVX512VL-FALLBACK-NEXT:    retq
 ;
@@ -2210,14 +2210,14 @@ define <32 x i8> @vec256_i8_unsigned_reg_reg(<32 x i8> %a1, <32 x i8> %a2) nounw
 ;
 ; AVX512VL-FALLBACK-LABEL: vec256_i8_unsigned_reg_reg:
 ; AVX512VL-FALLBACK:       # %bb.0:
-; AVX512VL-FALLBACK-NEXT:    vpminub %ymm1, %ymm0, %ymm2
-; AVX512VL-FALLBACK-NEXT:    vpmaxub %ymm1, %ymm0, %ymm1
-; AVX512VL-FALLBACK-NEXT:    vpsubb %ymm2, %ymm1, %ymm1
-; AVX512VL-FALLBACK-NEXT:    vpsrlw $1, %ymm1, %ymm1
-; AVX512VL-FALLBACK-NEXT:    vpcmpeqb %ymm2, %ymm0, %ymm2
-; AVX512VL-FALLBACK-NEXT:    vpternlogq $15, %ymm2, %ymm2, %ymm2
-; AVX512VL-FALLBACK-NEXT:    vpternlogd $108, {{\.?LCPI[0-9]+_[0-9]+}}(%rip){1to8}, %ymm2, %ymm1
-; AVX512VL-FALLBACK-NEXT:    vpsubb %ymm2, %ymm1, %ymm1
+; AVX512VL-FALLBACK-NEXT:    vpmaxub %ymm1, %ymm0, %ymm2
+; AVX512VL-FALLBACK-NEXT:    vpminub %ymm1, %ymm0, %ymm1
+; AVX512VL-FALLBACK-NEXT:    vpsubb %ymm1, %ymm2, %ymm2
+; AVX512VL-FALLBACK-NEXT:    vpsrlw $1, %ymm2, %ymm2
+; AVX512VL-FALLBACK-NEXT:    vpcmpeqb %ymm1, %ymm0, %ymm1
+; AVX512VL-FALLBACK-NEXT:    vpternlogq $15, %ymm1, %ymm1, %ymm1
+; AVX512VL-FALLBACK-NEXT:    vpternlogd $108, {{\.?LCPI[0-9]+_[0-9]+}}(%rip){1to8}, %ymm1, %ymm2
+; AVX512VL-FALLBACK-NEXT:    vpsubb %ymm1, %ymm2, %ymm1
 ; AVX512VL-FALLBACK-NEXT:    vpaddb %ymm0, %ymm1, %ymm0
 ; AVX512VL-FALLBACK-NEXT:    retq
 ;
diff --git a/llvm/test/CodeGen/X86/midpoint-int-vec-512.ll b/llvm/test/CodeGen/X86/midpoint-int-vec-512.ll
index 2fdf6ef224ca..366dad1612b4 100644
--- a/llvm/test/CodeGen/X86/midpoint-int-vec-512.ll
+++ b/llvm/test/CodeGen/X86/midpoint-int-vec-512.ll
@@ -684,22 +684,21 @@ define <64 x i8> @vec512_i8_signed_reg_reg(<64 x i8> %a1, <64 x i8> %a2) nounwin
 ; AVX512F-NEXT:    vpcmpgtb %ymm2, %ymm3, %ymm4
 ; AVX512F-NEXT:    vpcmpgtb %ymm1, %ymm0, %ymm5
 ; AVX512F-NEXT:    vinserti64x4 $1, %ymm4, %zmm5, %zmm4
-; AVX512F-NEXT:    vpminsb %ymm2, %ymm3, %ymm5
-; AVX512F-NEXT:    vpmaxsb %ymm2, %ymm3, %ymm2
-; AVX512F-NEXT:    vpsubb %ymm5, %ymm2, %ymm2
 ; AVX512F-NEXT:    vpminsb %ymm1, %ymm0, %ymm5
 ; AVX512F-NEXT:    vpmaxsb %ymm1, %ymm0, %ymm1
 ; AVX512F-NEXT:    vpsubb %ymm5, %ymm1, %ymm1
-; AVX512F-NEXT:    vpsrlw $1, %ymm1, %ymm1
+; AVX512F-NEXT:    vpminsb %ymm2, %ymm3, %ymm5
+; AVX512F-NEXT:    vpmaxsb %ymm2, %ymm3, %ymm2
+; AVX512F-NEXT:    vpsubb %ymm5, %ymm2, %ymm2
 ; AVX512F-NEXT:    vpsrlw $1, %ymm2, %ymm2
+; AVX512F-NEXT:    vpbroadcastb {{.*#+}} ymm5 = [127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127]
+; AVX512F-NEXT:    vpand %ymm5, %ymm2, %ymm2
+; AVX512F-NEXT:    vpsrlw $1, %ymm1, %ymm1
+; AVX512F-NEXT:    vpand %ymm5, %ymm1, %ymm1
 ; AVX512F-NEXT:    vinserti64x4 $1, %ymm2, %zmm1, %zmm5
-; AVX512F-NEXT:    vpbroadcastd {{.*#+}} zmm6 = [127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127]
-; AVX512F-NEXT:    vpandq %zmm6, %zmm5, %zmm5
-; AVX512F-NEXT:    vpand %ymm6, %ymm2, %ymm2
-; AVX512F-NEXT:    vpxor %xmm7, %xmm7, %xmm7
-; AVX512F-NEXT:    vpsubb %ymm2, %ymm7, %ymm2
-; AVX512F-NEXT:    vpand %ymm6, %ymm1, %ymm1
-; AVX512F-NEXT:    vpsubb %ymm1, %ymm7, %ymm1
+; AVX512F-NEXT:    vpxor %xmm6, %xmm6, %xmm6
+; AVX512F-NEXT:    vpsubb %ymm2, %ymm6, %ymm2
+; AVX512F-NEXT:    vpsubb %ymm1, %ymm6, %ymm1
 ; AVX512F-NEXT:    vinserti64x4 $1, %ymm2, %zmm1, %zmm1
 ; AVX512F-NEXT:    vpternlogq $226, %zmm5, %zmm4, %zmm1
 ; AVX512F-NEXT:    vextracti64x4 $1, %zmm1, %ymm2
@@ -715,22 +714,21 @@ define <64 x i8> @vec512_i8_signed_reg_reg(<64 x i8> %a1, <64 x i8> %a2) nounwin
 ; AVX512VL-FALLBACK-NEXT:    vpcmpgtb %ymm2, %ymm3, %ymm4
 ; AVX512VL-FALLBACK-NEXT:    vpcmpgtb %ymm1, %ymm0, %ymm5
 ; AVX512VL-FALLBACK-NEXT:    vinserti64x4 $1, %ymm4, %zmm5, %zmm4
-; AVX512VL-FALLBACK-NEXT:    vpminsb %ymm2, %ymm3, %ymm5
-; AVX512VL-FALLBACK-NEXT:    vpmaxsb %ymm2, %ymm3, %ymm2
-; AVX512VL-FALLBACK-NEXT:    vpsubb %ymm5, %ymm2, %ymm2
 ; AVX512VL-FALLBACK-NEXT:    vpminsb %ymm1, %ymm0, %ymm5
 ; AVX512VL-FALLBACK-NEXT:    vpmaxsb %ymm1, %ymm0, %ymm1
 ; AVX512VL-FALLBACK-NEXT:    vpsubb %ymm5, %ymm1, %ymm1
-; AVX512VL-FALLBACK-NEXT:    vpsrlw $1, %ymm1, %ymm1
+; AVX512VL-FALLBACK-NEXT:    vpminsb %ymm2, %ymm3, %ymm5
+; AVX512VL-FALLBACK-NEXT:    vpmaxsb %ymm2, %ymm3, %ymm2
+; AVX512VL-FALLBACK-NEXT:    vpsubb %ymm5, %ymm2, %ymm2
 ; AVX512VL-FALLBACK-NEXT:    vpsrlw $1, %ymm2, %ymm2
+; AVX512VL-FALLBACK-NEXT:    vpbroadcastd {{.*#+}} ymm5 = [127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127]
+; AVX512VL-FALLBACK-NEXT:    vpand %ymm5, %ymm2, %ymm2
+; AVX512VL-FALLBACK-NEXT:    vpsrlw $1, %ymm1, %ymm1
+; AVX512VL-FALLBACK-NEXT:    vpand %ymm5, %ymm1, %ymm1
 ; AVX512VL-FALLBACK-NEXT:    vinserti64x4 $1, %ymm2, %zmm1, %zmm5
-; AVX512VL-FALLBACK-NEXT:    vpbroadcastd {{.*#+}} zmm6 = [127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127]
-; AVX512VL-FALLBACK-NEXT:    vpandq %zmm6, %zmm5, %zmm5
-; AVX512VL-FALLBACK-NEXT:    vpand %ymm6, %ymm2, %ymm2
-; AVX512VL-FALLBACK-NEXT:    vpxor %xmm7, %xmm7, %xmm7
-; AVX512VL-FALLBACK-NEXT:    vpsubb %ymm2, %ymm7, %ymm2
-; AVX512VL-FALLBACK-NEXT:    vpand %ymm6, %ymm1, %ymm1
-; AVX512VL-FALLBACK-NEXT:    vpsubb %ymm1, %ymm7, %ymm1
+; AVX512VL-FALLBACK-NEXT:    vpxor %xmm6, %xmm6, %xmm6
+; AVX512VL-FALLBACK-NEXT:    vpsubb %ymm2, %ymm6, %ymm2
+; AVX512VL-FALLBACK-NEXT:    vpsubb %ymm1, %ymm6, %ymm1
 ; AVX512VL-FALLBACK-NEXT:    vinserti64x4 $1, %ymm2, %zmm1, %zmm1
 ; AVX512VL-FALLBACK-NEXT:    vpternlogq $226, %zmm5, %zmm4, %zmm1
 ; AVX512VL-FALLBACK-NEXT:    vextracti64x4 $1, %zmm1, %ymm2
@@ -772,20 +770,19 @@ define <64 x i8> @vec512_i8_unsigned_reg_reg(<64 x i8> %a1, <64 x i8> %a2) nounw
 ; AVX512F-NEXT:    vpminub %ymm1, %ymm0, %ymm6
 ; AVX512F-NEXT:    vpcmpeqb %ymm6, %ymm0, %ymm7
 ; AVX512F-NEXT:    vinserti64x4 $1, %ymm5, %zmm7, %zmm5
-; AVX512F-NEXT:    vpmaxub %ymm2, %ymm3, %ymm2
-; AVX512F-NEXT:    vpsubb %ymm4, %ymm2, %ymm2
 ; AVX512F-NEXT:    vpmaxub %ymm1, %ymm0, %ymm1
 ; AVX512F-NEXT:    vpsubb %ymm6, %ymm1, %ymm1
-; AVX512F-NEXT:    vpsrlw $1, %ymm1, %ymm1
+; AVX512F-NEXT:    vpmaxub %ymm2, %ymm3, %ymm2
+; AVX512F-NEXT:    vpsubb %ymm4, %ymm2, %ymm2
 ; AVX512F-NEXT:    vpsrlw $1, %ymm2, %ymm2
+; AVX512F-NEXT:    vpbroadcastb {{.*#+}} ymm4 = [127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127]
+; AVX512F-NEXT:    vpand %ymm4, %ymm2, %ymm2
+; AVX512F-NEXT:    vpsrlw $1, %ymm1, %ymm1
+; AVX512F-NEXT:    vpand %ymm4, %ymm1, %ymm1
 ; AVX512F-NEXT:    vinserti64x4 $1, %ymm2, %zmm1, %zmm4
-; AVX512F-NEXT:    vpbroadcastd {{.*#+}} zmm6 = [127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127]
-; AVX512F-NEXT:    vpandq %zmm6, %zmm4, %zmm4
-; AVX512F-NEXT:    vpand %ymm6, %ymm2, %ymm2
-; AVX512F-NEXT:    vpxor %xmm7, %xmm7, %xmm7
-; AVX512F-NEXT:    vpsubb %ymm2, %ymm7, %ymm2
-; AVX512F-NEXT:    vpand %ymm6, %ymm1, %ymm1
-; AVX512F-NEXT:    vpsubb %ymm1, %ymm7, %ymm1
+; AVX512F-NEXT:    vpxor %xmm6, %xmm6, %xmm6
+; AVX512F-NEXT:    vpsubb %ymm2, %ymm6, %ymm2
+; AVX512F-NEXT:    vpsubb %ymm1, %ymm6, %ymm1
 ; AVX512F-NEXT:    vinserti64x4 $1, %ymm2, %zmm1, %zmm1
 ; AVX512F-NEXT:    vpternlogq $184, %zmm4, %zmm5, %zmm1
 ; AVX512F-NEXT:    vextracti64x4 $1, %zmm1, %ymm2
@@ -803,20 +800,19 @@ define <64 x i8> @vec512_i8_unsigned_reg_reg(<64 x i8> %a1, <64 x i8> %a2) nounw
 ; AVX512VL-FALLBACK-NEXT:    vpminub %ymm1, %ymm0, %ymm6
 ; AVX512VL-FALLBACK-NEXT:    vpcmpeqb %ymm6, %ymm0, %ymm7
 ; AVX512VL-FALLBACK-NEXT:    vinserti64x4 $1, %ymm5, %zmm7, %zmm5
-; AVX512VL-FALLBACK-NEXT:    vpmaxub %ymm2, %ymm3, %ymm2
-; AVX512VL-FALLBACK-NEXT:    vpsubb %ymm4, %ymm2, %ymm2
 ; AVX512VL-FALLBACK-NEXT:    vpmaxub %ymm1, %ymm0, %ymm1
 ; AVX512VL-FALLBACK-NEXT:    vpsubb %ymm6, %ymm1, %ymm1
-; AVX512VL-FALLBACK-NEXT:    vpsrlw $1, %ymm1, %ymm1
+; AVX512VL-FALLBACK-NEXT:    vpmaxub %ymm2, %ymm3, %ymm2
+; AVX512VL-FALLBACK-NEXT:    vpsubb %ymm4, %ymm2, %ymm2
 ; AVX512VL-FALLBACK-NEXT:    vpsrlw $1, %ymm2, %ymm2
+; AVX512VL-FALLBACK-NEXT:    vpbroadcastd {{.*#+}} ymm4 = [127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127]
+; AVX512VL-FALLBACK-NEXT:    vpand %ymm4, %ymm2, %ymm2
+; AVX512VL-FALLBACK-NEXT:    vpsrlw $1, %ymm1, %ymm1
+; AVX512VL-FALLBACK-NEXT:    vpand %ymm4, %ymm1, %ymm1
 ; AVX512VL-FALLBACK-NEXT:    vinserti64x4 $1, %ymm2, %zmm1, %zmm4
-; AVX512VL-FALLBACK-NEXT:    vpbroadcastd {{.*#+}} zmm6 = [127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127]
-; AVX512VL-FALLBACK-NEXT:    vpandq %zmm6, %zmm4, %zmm4
-; AVX512VL-FALLBACK-NEXT:    vpand %ymm6, %ymm2, %ymm2
-; AVX512VL-FALLBACK-NEXT:    vpxor %xmm7, %xmm7, %xmm7
-; AVX512VL-FALLBACK-NEXT:    vpsubb %ymm2, %ymm7, %ymm2
-; AVX512VL-FALLBACK-NEXT:    vpand %ymm6, %ymm1, %ymm1
-; AVX512VL-FALLBACK-NEXT:    vpsubb %ymm1, %ymm7, %ymm1
+; AVX512VL-FALLBACK-NEXT:    vpxor %xmm6, %xmm6, %xmm6
+; AVX512VL-FALLBACK-NEXT:    vpsubb %ymm2, %ymm6, %ymm2
+; AVX512VL-FALLBACK-NEXT:    vpsubb %ymm1, %ymm6, %ymm1
 ; AVX512VL-FALLBACK-NEXT:    vinserti64x4 $1, %ymm2, %zmm1, %zmm1
 ; AVX512VL-FALLBACK-NEXT:    vpternlogq $184, %zmm4, %zmm5, %zmm1
 ; AVX512VL-FALLBACK-NEXT:    vextracti64x4 $1, %zmm1, %ymm2
diff --git a/llvm/test/CodeGen/X86/named-vector-shuffle-reverse.ll b/llvm/test/CodeGen/X86/named-vector-shuffle-reverse.ll
index f0917be88744..2a5e834f0ac7 100644
--- a/llvm/test/CodeGen/X86/named-vector-shuffle-reverse.ll
+++ b/llvm/test/CodeGen/X86/named-vector-shuffle-reverse.ll
@@ -23,7 +23,7 @@ define <16 x i8> @reverse_v16i8(<16 x i8> %a) #0 {
 ; CHECK-NEXT:    packuswb %xmm2, %xmm0
 ; CHECK-NEXT:    retq
 
-  %res = call <16 x i8> @llvm.experimental.vector.reverse.v16i8(<16 x i8> %a)
+  %res = call <16 x i8> @llvm.vector.reverse.v16i8(<16 x i8> %a)
   ret <16 x i8> %res
 }
 
@@ -34,7 +34,7 @@ define <8 x i16> @reverse_v8i16(<8 x i16> %a) #0 {
 ; CHECK-NEXT:    pshuflw {{.*#+}} xmm0 = xmm0[3,2,1,0,4,5,6,7]
 ; CHECK-NEXT:    pshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,7,6,5,4]
 ; CHECK-NEXT:    retq
-  %res = call <8 x i16> @llvm.experimental.vector.reverse.v8i16(<8 x i16> %a)
+  %res = call <8 x i16> @llvm.vector.reverse.v8i16(<8 x i16> %a)
   ret <8 x i16> %res
 }
 
@@ -43,7 +43,7 @@ define <4 x i32> @reverse_v4i32(<4 x i32> %a) #0 {
 ; CHECK:       # %bb.0:
 ; CHECK-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[3,2,1,0]
 ; CHECK-NEXT:    retq
-  %res = call <4 x i32> @llvm.experimental.vector.reverse.v4i32(<4 x i32> %a)
+  %res = call <4 x i32> @llvm.vector.reverse.v4i32(<4 x i32> %a)
   ret <4 x i32> %res
 }
 
@@ -52,7 +52,7 @@ define <2 x i64> @reverse_v2i64(<2 x i64> %a) #0 {
 ; CHECK:       # %bb.0:
 ; CHECK-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[2,3,0,1]
 ; CHECK-NEXT:    retq
-  %res = call <2 x i64> @llvm.experimental.vector.reverse.v2i64(<2 x i64> %a)
+  %res = call <2 x i64> @llvm.vector.reverse.v2i64(<2 x i64> %a)
   ret <2 x i64> %res
 }
 
@@ -61,7 +61,7 @@ define <4 x float> @reverse_v4f32(<4 x float> %a) #0 {
 ; CHECK:       # %bb.0:
 ; CHECK-NEXT:    shufps {{.*#+}} xmm0 = xmm0[3,2,1,0]
 ; CHECK-NEXT:    retq
-  %res = call <4 x float> @llvm.experimental.vector.reverse.v4f32(<4 x float> %a)
+  %res = call <4 x float> @llvm.vector.reverse.v4f32(<4 x float> %a)
   ret <4 x float> %res
 }
 
@@ -70,7 +70,7 @@ define <2 x double> @reverse_v2f64(<2 x double> %a) #0 {
 ; CHECK:       # %bb.0:
 ; CHECK-NEXT:    shufps {{.*#+}} xmm0 = xmm0[2,3,0,1]
 ; CHECK-NEXT:    retq
-  %res = call <2 x double> @llvm.experimental.vector.reverse.v2f64(<2 x double> %a)
+  %res = call <2 x double> @llvm.vector.reverse.v2f64(<2 x double> %a)
   ret <2 x double> %res
 }
 
@@ -83,7 +83,7 @@ define <2 x i8> @reverse_v2i8(<2 x i8> %a) #0 {
 ; CHECK-NEXT:    psllw $8, %xmm0
 ; CHECK-NEXT:    por %xmm1, %xmm0
 ; CHECK-NEXT:    retq
-  %res = call <2 x i8> @llvm.experimental.vector.reverse.v2i8(<2 x i8> %a)
+  %res = call <2 x i8> @llvm.vector.reverse.v2i8(<2 x i8> %a)
   ret <2 x i8> %res
 }
 
@@ -95,7 +95,7 @@ define <8 x i32> @reverse_v8i32(<8 x i32> %a) #0 {
 ; CHECK-NEXT:    pshufd {{.*#+}} xmm1 = xmm0[3,2,1,0]
 ; CHECK-NEXT:    movdqa %xmm2, %xmm0
 ; CHECK-NEXT:    retq
-  %res = call <8 x i32> @llvm.experimental.vector.reverse.v8i32(<8 x i32> %a)
+  %res = call <8 x i32> @llvm.vector.reverse.v8i32(<8 x i32> %a)
   ret <8 x i32> %res
 }
 
@@ -115,20 +115,20 @@ define <16 x float> @reverse_v16f32(<16 x float> %a) #0 {
 ; CHECK-NEXT:    movaps %xmm5, %xmm3
 ; CHECK-NEXT:    retq
 
-  %res = call <16 x float> @llvm.experimental.vector.reverse.v16f32(<16 x float> %a)
+  %res = call <16 x float> @llvm.vector.reverse.v16f32(<16 x float> %a)
   ret <16 x float> %res
 }
 
 
-declare <2 x i8> @llvm.experimental.vector.reverse.v2i8(<2 x i8>)
-declare <16 x i8> @llvm.experimental.vector.reverse.v16i8(<16 x i8>)
-declare <8 x i16> @llvm.experimental.vector.reverse.v8i16(<8 x i16>)
-declare <4 x i32> @llvm.experimental.vector.reverse.v4i32(<4 x i32>)
-declare <8 x i32> @llvm.experimental.vector.reverse.v8i32(<8 x i32>)
-declare <2 x i64> @llvm.experimental.vector.reverse.v2i64(<2 x i64>)
-declare <8 x half> @llvm.experimental.vector.reverse.v8f16(<8 x half>)
-declare <4 x float> @llvm.experimental.vector.reverse.v4f32(<4 x float>)
-declare <16 x float> @llvm.experimental.vector.reverse.v16f32(<16 x float>)
-declare <2 x double> @llvm.experimental.vector.reverse.v2f64(<2 x double>)
+declare <2 x i8> @llvm.vector.reverse.v2i8(<2 x i8>)
+declare <16 x i8> @llvm.vector.reverse.v16i8(<16 x i8>)
+declare <8 x i16> @llvm.vector.reverse.v8i16(<8 x i16>)
+declare <4 x i32> @llvm.vector.reverse.v4i32(<4 x i32>)
+declare <8 x i32> @llvm.vector.reverse.v8i32(<8 x i32>)
+declare <2 x i64> @llvm.vector.reverse.v2i64(<2 x i64>)
+declare <8 x half> @llvm.vector.reverse.v8f16(<8 x half>)
+declare <4 x float> @llvm.vector.reverse.v4f32(<4 x float>)
+declare <16 x float> @llvm.vector.reverse.v16f32(<16 x float>)
+declare <2 x double> @llvm.vector.reverse.v2f64(<2 x double>)
 
 attributes #0 = { nounwind }
diff --git a/llvm/test/CodeGen/X86/pr34592.ll b/llvm/test/CodeGen/X86/pr34592.ll
index 9887dac76e20..23de746ecb35 100644
--- a/llvm/test/CodeGen/X86/pr34592.ll
+++ b/llvm/test/CodeGen/X86/pr34592.ll
@@ -1,51 +1,76 @@
 ; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
-; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=avx2 -O0 | FileCheck %s
+; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=avx2 -O0 | FileCheck %s --check-prefixes=CHECK-O0
+; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=avx2 -O3 | FileCheck %s --check-prefixes=CHECK-O3
 
-define <16 x i64> @pluto(<16 x i64> %arg, <16 x i64> %arg1, <16 x i64> %arg2, <16 x i64> %arg3, <16 x i64> %arg4) {
-; CHECK-LABEL: pluto:
-; CHECK:       # %bb.0: # %bb
-; CHECK-NEXT:    pushq %rbp
-; CHECK-NEXT:    .cfi_def_cfa_offset 16
-; CHECK-NEXT:    .cfi_offset %rbp, -16
-; CHECK-NEXT:    movq %rsp, %rbp
-; CHECK-NEXT:    .cfi_def_cfa_register %rbp
-; CHECK-NEXT:    andq $-32, %rsp
-; CHECK-NEXT:    subq $32, %rsp
-; CHECK-NEXT:    vmovaps %ymm4, %ymm10
-; CHECK-NEXT:    vmovaps %ymm3, %ymm9
-; CHECK-NEXT:    vmovaps %ymm1, %ymm8
-; CHECK-NEXT:    vmovaps 240(%rbp), %ymm4
-; CHECK-NEXT:    vmovaps 208(%rbp), %ymm3
-; CHECK-NEXT:    vmovaps 176(%rbp), %ymm1
-; CHECK-NEXT:    vmovaps 144(%rbp), %ymm1
-; CHECK-NEXT:    vmovaps 112(%rbp), %ymm11
-; CHECK-NEXT:    vmovaps 80(%rbp), %ymm11
-; CHECK-NEXT:    vmovaps 48(%rbp), %ymm11
-; CHECK-NEXT:    vmovaps 16(%rbp), %ymm11
-; CHECK-NEXT:    vpblendd {{.*#+}} ymm0 = ymm6[0,1,2,3,4,5],ymm2[6,7]
-; CHECK-NEXT:    vpunpcklqdq {{.*#+}} ymm1 = ymm1[0],ymm3[0],ymm1[2],ymm3[2]
-; CHECK-NEXT:    vpermq {{.*#+}} ymm1 = ymm1[0,2,1,3]
-; CHECK-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[3,1,2,1]
-; CHECK-NEXT:    vpblendd {{.*#+}} ymm0 = ymm0[0,1],ymm1[2,3,4,5],ymm0[6,7]
-; CHECK-NEXT:    vperm2i128 {{.*#+}} ymm1 = ymm7[2,3],ymm6[0,1]
-; CHECK-NEXT:    vxorps %xmm2, %xmm2, %xmm2
-; CHECK-NEXT:    vpblendd {{.*#+}} ymm2 = ymm1[0,1],ymm2[2,3],ymm1[4,5,6,7]
-; CHECK-NEXT:    vpunpcklqdq {{.*#+}} ymm1 = ymm7[0],ymm5[0],ymm7[2],ymm5[2]
-; CHECK-NEXT:    vpermq {{.*#+}} ymm1 = ymm1[2,1,2,3]
-; CHECK-NEXT:    vpermq {{.*#+}} ymm4 = ymm4[1,1,1,1]
-; CHECK-NEXT:    vpblendd {{.*#+}} ymm1 = ymm1[0,1],ymm4[2,3,4,5],ymm1[6,7]
-; CHECK-NEXT:    vmovaps %xmm3, %xmm4
-; CHECK-NEXT:    vmovaps %xmm7, %xmm3
-; CHECK-NEXT:    vpblendd {{.*#+}} xmm4 = xmm3[0,1],xmm4[2,3]
-; CHECK-NEXT:    # implicit-def: $ymm3
-; CHECK-NEXT:    vmovaps %xmm4, %xmm3
-; CHECK-NEXT:    vpermq {{.*#+}} ymm4 = ymm3[0,0,1,3]
-; CHECK-NEXT:    vpslldq {{.*#+}} ymm3 = zero,zero,zero,zero,zero,zero,zero,zero,ymm5[0,1,2,3,4,5,6,7],zero,zero,zero,zero,zero,zero,zero,zero,ymm5[16,17,18,19,20,21,22,23]
-; CHECK-NEXT:    vpblendd {{.*#+}} ymm3 = ymm3[0,1],ymm4[2,3,4,5],ymm3[6,7]
-; CHECK-NEXT:    movq %rbp, %rsp
-; CHECK-NEXT:    popq %rbp
-; CHECK-NEXT:    .cfi_def_cfa %rsp, 8
-; CHECK-NEXT:    retq
+define <16 x i64> @pluto(<16 x i64> %arg, <16 x i64> %arg1, <16 x i64> %arg2, <16 x i64> %arg3, <16 x i64> %arg4) nounwind {
+; CHECK-O0-LABEL: pluto:
+; CHECK-O0:       # %bb.0: # %bb
+; CHECK-O0-NEXT:    pushq %rbp
+; CHECK-O0-NEXT:    movq %rsp, %rbp
+; CHECK-O0-NEXT:    andq $-32, %rsp
+; CHECK-O0-NEXT:    subq $32, %rsp
+; CHECK-O0-NEXT:    vmovaps %ymm4, %ymm10
+; CHECK-O0-NEXT:    vmovaps %ymm3, %ymm9
+; CHECK-O0-NEXT:    vmovaps %ymm1, %ymm8
+; CHECK-O0-NEXT:    vmovaps 240(%rbp), %ymm4
+; CHECK-O0-NEXT:    vmovaps 208(%rbp), %ymm3
+; CHECK-O0-NEXT:    vmovaps 176(%rbp), %ymm1
+; CHECK-O0-NEXT:    vmovaps 144(%rbp), %ymm1
+; CHECK-O0-NEXT:    vmovaps 112(%rbp), %ymm11
+; CHECK-O0-NEXT:    vmovaps 80(%rbp), %ymm11
+; CHECK-O0-NEXT:    vmovaps 48(%rbp), %ymm11
+; CHECK-O0-NEXT:    vmovaps 16(%rbp), %ymm11
+; CHECK-O0-NEXT:    vpblendd {{.*#+}} ymm0 = ymm6[0,1,2,3,4,5],ymm2[6,7]
+; CHECK-O0-NEXT:    vpunpcklqdq {{.*#+}} ymm1 = ymm1[0],ymm3[0],ymm1[2],ymm3[2]
+; CHECK-O0-NEXT:    vpermq {{.*#+}} ymm1 = ymm1[0,2,1,3]
+; CHECK-O0-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[3,1,2,1]
+; CHECK-O0-NEXT:    vpblendd {{.*#+}} ymm0 = ymm0[0,1],ymm1[2,3,4,5],ymm0[6,7]
+; CHECK-O0-NEXT:    vperm2i128 {{.*#+}} ymm1 = ymm7[2,3],ymm6[0,1]
+; CHECK-O0-NEXT:    vxorps %xmm2, %xmm2, %xmm2
+; CHECK-O0-NEXT:    vpblendd {{.*#+}} ymm2 = ymm1[0,1],ymm2[2,3],ymm1[4,5,6,7]
+; CHECK-O0-NEXT:    vpunpcklqdq {{.*#+}} ymm1 = ymm7[0],ymm5[0],ymm7[2],ymm5[2]
+; CHECK-O0-NEXT:    vpermq {{.*#+}} ymm1 = ymm1[2,1,2,3]
+; CHECK-O0-NEXT:    vpermq {{.*#+}} ymm4 = ymm4[1,1,1,1]
+; CHECK-O0-NEXT:    vpblendd {{.*#+}} ymm1 = ymm1[0,1],ymm4[2,3,4,5],ymm1[6,7]
+; CHECK-O0-NEXT:    vmovaps %xmm3, %xmm4
+; CHECK-O0-NEXT:    vmovaps %xmm7, %xmm3
+; CHECK-O0-NEXT:    vpblendd {{.*#+}} xmm4 = xmm3[0,1],xmm4[2,3]
+; CHECK-O0-NEXT:    # implicit-def: $ymm3
+; CHECK-O0-NEXT:    vmovaps %xmm4, %xmm3
+; CHECK-O0-NEXT:    vpermq {{.*#+}} ymm4 = ymm3[0,0,1,3]
+; CHECK-O0-NEXT:    vpslldq {{.*#+}} ymm3 = zero,zero,zero,zero,zero,zero,zero,zero,ymm5[0,1,2,3,4,5,6,7],zero,zero,zero,zero,zero,zero,zero,zero,ymm5[16,17,18,19,20,21,22,23]
+; CHECK-O0-NEXT:    vpblendd {{.*#+}} ymm3 = ymm3[0,1],ymm4[2,3,4,5],ymm3[6,7]
+; CHECK-O0-NEXT:    movq %rbp, %rsp
+; CHECK-O0-NEXT:    popq %rbp
+; CHECK-O0-NEXT:    retq
+;
+; CHECK-O3-LABEL: pluto:
+; CHECK-O3:       # %bb.0: # %bb
+; CHECK-O3-NEXT:    pushq %rbp
+; CHECK-O3-NEXT:    movq %rsp, %rbp
+; CHECK-O3-NEXT:    andq $-32, %rsp
+; CHECK-O3-NEXT:    subq $32, %rsp
+; CHECK-O3-NEXT:    vmovdqa 208(%rbp), %ymm3
+; CHECK-O3-NEXT:    vmovdqa 144(%rbp), %ymm0
+; CHECK-O3-NEXT:    vpblendd {{.*#+}} ymm1 = ymm6[0,1,2,3,4,5],ymm2[6,7]
+; CHECK-O3-NEXT:    vpunpcklqdq {{.*#+}} ymm0 = ymm0[0],ymm3[0],ymm0[2],ymm3[2]
+; CHECK-O3-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,2,1,3]
+; CHECK-O3-NEXT:    vpermq {{.*#+}} ymm1 = ymm1[3,1,2,1]
+; CHECK-O3-NEXT:    vpblendd {{.*#+}} ymm0 = ymm1[0,1],ymm0[2,3,4,5],ymm1[6,7]
+; CHECK-O3-NEXT:    vperm2i128 {{.*#+}} ymm1 = ymm7[2,3],ymm6[0,1]
+; CHECK-O3-NEXT:    vpxor %xmm2, %xmm2, %xmm2
+; CHECK-O3-NEXT:    vpblendd {{.*#+}} ymm2 = ymm1[0,1],ymm2[2,3],ymm1[4,5,6,7]
+; CHECK-O3-NEXT:    vpunpcklqdq {{.*#+}} ymm1 = ymm7[0],ymm5[0],ymm7[2],ymm5[2]
+; CHECK-O3-NEXT:    vpermq {{.*#+}} ymm1 = ymm1[2,1,2,3]
+; CHECK-O3-NEXT:    vpbroadcastq 248(%rbp), %ymm4
+; CHECK-O3-NEXT:    vpblendd {{.*#+}} ymm1 = ymm1[0,1],ymm4[2,3,4,5],ymm1[6,7]
+; CHECK-O3-NEXT:    vpblendd {{.*#+}} xmm3 = xmm7[0,1],xmm3[2,3]
+; CHECK-O3-NEXT:    vpermq {{.*#+}} ymm3 = ymm3[0,0,1,3]
+; CHECK-O3-NEXT:    vpslldq {{.*#+}} ymm4 = zero,zero,zero,zero,zero,zero,zero,zero,ymm5[0,1,2,3,4,5,6,7],zero,zero,zero,zero,zero,zero,zero,zero,ymm5[16,17,18,19,20,21,22,23]
+; CHECK-O3-NEXT:    vpblendd {{.*#+}} ymm3 = ymm4[0,1],ymm3[2,3,4,5],ymm4[6,7]
+; CHECK-O3-NEXT:    movq %rbp, %rsp
+; CHECK-O3-NEXT:    popq %rbp
+; CHECK-O3-NEXT:    retq
 bb:
   %tmp = select <16 x i1> <i1 false, i1 false, i1 false, i1 false, i1 false, i1 false, i1 false, i1 false, i1 false, i1 false, i1 true, i1 true, i1 false, i1 false, i1 false, i1 false>, <16 x i64> %arg, <16 x i64> %arg1
   %tmp5 = select <16 x i1> <i1 true, i1 false, i1 false, i1 true, i1 true, i1 false, i1 false, i1 true, i1 false, i1 true, i1 false, i1 false, i1 false, i1 false, i1 false, i1 false>, <16 x i64> %arg2, <16 x i64> zeroinitializer
diff --git a/llvm/test/CodeGen/X86/pr38539.ll b/llvm/test/CodeGen/X86/pr38539.ll
index 04aff9b7d2e5..ace78b38d53e 100644
--- a/llvm/test/CodeGen/X86/pr38539.ll
+++ b/llvm/test/CodeGen/X86/pr38539.ll
@@ -22,7 +22,7 @@ define void @f() nounwind {
 ; X86-NEXT:    pushl %edi
 ; X86-NEXT:    pushl %esi
 ; X86-NEXT:    andl $-16, %esp
-; X86-NEXT:    subl $176, %esp
+; X86-NEXT:    subl $160, %esp
 ; X86-NEXT:    movzbl {{[0-9]+}}(%esp), %edx
 ; X86-NEXT:    movl {{[0-9]+}}(%esp), %esi
 ; X86-NEXT:    movl {{[0-9]+}}(%esp), %edi
@@ -47,55 +47,54 @@ define void @f() nounwind {
 ; X86-NEXT:    testl %edi, %edi
 ; X86-NEXT:    jne .LBB0_1
 ; X86-NEXT:  # %bb.2: # %BB_udiv-special-cases
-; X86-NEXT:    bsrl %esi, %ecx
-; X86-NEXT:    xorl $31, %ecx
-; X86-NEXT:    addl $32, %ecx
+; X86-NEXT:    bsrl %esi, %eax
+; X86-NEXT:    xorl $31, %eax
+; X86-NEXT:    addl $32, %eax
 ; X86-NEXT:    jmp .LBB0_3
 ; X86-NEXT:  .LBB0_1:
-; X86-NEXT:    bsrl %edi, %ecx
-; X86-NEXT:    xorl $31, %ecx
+; X86-NEXT:    bsrl %edi, %eax
+; X86-NEXT:    xorl $31, %eax
 ; X86-NEXT:  .LBB0_3: # %BB_udiv-special-cases
-; X86-NEXT:    xorl %eax, %eax
+; X86-NEXT:    xorl %ecx, %ecx
 ; X86-NEXT:    testl %edx, %edx
-; X86-NEXT:    movl %edi, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT:    movl %esi, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
 ; X86-NEXT:    jne .LBB0_4
 ; X86-NEXT:  # %bb.5: # %BB_udiv-special-cases
-; X86-NEXT:    addl $64, %ecx
+; X86-NEXT:    addl $64, %eax
 ; X86-NEXT:    jmp .LBB0_6
 ; X86-NEXT:  .LBB0_4:
-; X86-NEXT:    bsrl %edx, %ecx
-; X86-NEXT:    xorl $31, %ecx
-; X86-NEXT:    addl $32, %ecx
+; X86-NEXT:    bsrl %edx, %eax
+; X86-NEXT:    xorl $31, %eax
+; X86-NEXT:    addl $32, %eax
 ; X86-NEXT:  .LBB0_6: # %BB_udiv-special-cases
-; X86-NEXT:    subl $62, %ecx
+; X86-NEXT:    subl $62, %eax
 ; X86-NEXT:    movl $0, %ebx
 ; X86-NEXT:    sbbl %ebx, %ebx
-; X86-NEXT:    sbbl %eax, %eax
-; X86-NEXT:    addl $-66, %ecx
+; X86-NEXT:    sbbl %ecx, %ecx
+; X86-NEXT:    addl $-66, %eax
 ; X86-NEXT:    adcl $-1, %ebx
-; X86-NEXT:    adcl $3, %eax
-; X86-NEXT:    movl %eax, %edi
-; X86-NEXT:    movb $1, %al
-; X86-NEXT:    testb %al, %al
+; X86-NEXT:    adcl $3, %ecx
+; X86-NEXT:    movl %ecx, %esi
+; X86-NEXT:    movb $1, %cl
+; X86-NEXT:    testb %cl, %cl
 ; X86-NEXT:    jne .LBB0_11
 ; X86-NEXT:  # %bb.7: # %BB_udiv-special-cases
-; X86-NEXT:    andl $3, %edi
-; X86-NEXT:    movl %ecx, %eax
-; X86-NEXT:    xorl $65, %eax
-; X86-NEXT:    movl %edi, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
-; X86-NEXT:    orl %edi, %eax
-; X86-NEXT:    orl %ebx, %eax
+; X86-NEXT:    andl $3, %esi
+; X86-NEXT:    movl %eax, %ecx
+; X86-NEXT:    xorl $65, %ecx
+; X86-NEXT:    movl %esi, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT:    orl %esi, %ecx
+; X86-NEXT:    orl %ebx, %ecx
 ; X86-NEXT:    je .LBB0_11
 ; X86-NEXT:  # %bb.8: # %udiv-bb1
-; X86-NEXT:    movl %esi, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
-; X86-NEXT:    movl %ecx, %eax
+; X86-NEXT:    movl %eax, %ecx
 ; X86-NEXT:    addl $1, %ecx
 ; X86-NEXT:    movl %ecx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
 ; X86-NEXT:    adcl $0, %ebx
-; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %esi # 4-byte Reload
-; X86-NEXT:    adcl $0, %esi
-; X86-NEXT:    andl $3, %esi
-; X86-NEXT:    movl %esi, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT:    movl %ebx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %ebx # 4-byte Reload
+; X86-NEXT:    adcl $0, %ebx
+; X86-NEXT:    andl $3, %ebx
 ; X86-NEXT:    movb $65, %cl
 ; X86-NEXT:    subb %al, %cl
 ; X86-NEXT:    movb %cl, %ch
@@ -112,29 +111,31 @@ define void @f() nounwind {
 ; X86-NEXT:    movl $0, {{[0-9]+}}(%esp)
 ; X86-NEXT:    movl $0, {{[0-9]+}}(%esp)
 ; X86-NEXT:    movl $0, {{[0-9]+}}(%esp)
-; X86-NEXT:    movl %edx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
-; X86-NEXT:    movl 136(%esp,%eax), %edx
+; X86-NEXT:    movl %edi, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT:    movl 120(%esp,%eax), %edi
 ; X86-NEXT:    movb %ch, %cl
-; X86-NEXT:    shll %cl, %edx
+; X86-NEXT:    shll %cl, %edi
 ; X86-NEXT:    notb %cl
-; X86-NEXT:    movl 128(%esp,%eax), %edi
-; X86-NEXT:    movl 132(%esp,%eax), %esi
-; X86-NEXT:    movl %esi, %eax
+; X86-NEXT:    movl 112(%esp,%eax), %esi
+; X86-NEXT:    movl %edx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT:    movl 116(%esp,%eax), %edx
+; X86-NEXT:    movl %edx, %eax
 ; X86-NEXT:    shrl %eax
 ; X86-NEXT:    shrl %cl, %eax
 ; X86-NEXT:    movb %ch, %cl
-; X86-NEXT:    shldl %cl, %edi, %esi
+; X86-NEXT:    shldl %cl, %esi, %edx
+; X86-NEXT:    movl %edx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT:    shll %cl, %esi
 ; X86-NEXT:    movl %esi, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
-; X86-NEXT:    shll %cl, %edi
-; X86-NEXT:    movl %edi, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
 ; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %ecx # 4-byte Reload
-; X86-NEXT:    orl {{[-0-9]+}}(%e{{[sb]}}p), %ecx # 4-byte Folded Reload
+; X86-NEXT:    movl %ebx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
 ; X86-NEXT:    orl %ebx, %ecx
+; X86-NEXT:    orl {{[-0-9]+}}(%e{{[sb]}}p), %ecx # 4-byte Folded Reload
 ; X86-NEXT:    je .LBB0_11
 ; X86-NEXT:  # %bb.9: # %udiv-preheader
-; X86-NEXT:    orl %eax, %edx
-; X86-NEXT:    andl $3, %edx
-; X86-NEXT:    movl %edx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT:    orl %eax, %edi
+; X86-NEXT:    andl $3, %edi
+; X86-NEXT:    movl %edi, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
 ; X86-NEXT:    movl $0, {{[0-9]+}}(%esp)
 ; X86-NEXT:    movl $0, {{[0-9]+}}(%esp)
 ; X86-NEXT:    movl $0, {{[0-9]+}}(%esp)
@@ -149,20 +150,20 @@ define void @f() nounwind {
 ; X86-NEXT:    # kill: def $al killed $al killed $eax
 ; X86-NEXT:    shrb $3, %al
 ; X86-NEXT:    andb $15, %al
-; X86-NEXT:    movzbl %al, %esi
-; X86-NEXT:    movl 80(%esp,%esi), %edx
-; X86-NEXT:    movl 84(%esp,%esi), %eax
-; X86-NEXT:    movl %eax, %edi
+; X86-NEXT:    movzbl %al, %eax
+; X86-NEXT:    movl 64(%esp,%eax), %edi
+; X86-NEXT:    movl 68(%esp,%eax), %edx
+; X86-NEXT:    movl %edx, %esi
 ; X86-NEXT:    movb %ch, %cl
-; X86-NEXT:    shrl %cl, %edi
+; X86-NEXT:    shrl %cl, %esi
 ; X86-NEXT:    notb %cl
-; X86-NEXT:    movl 88(%esp,%esi), %esi
-; X86-NEXT:    addl %esi, %esi
-; X86-NEXT:    shll %cl, %esi
-; X86-NEXT:    orl %edi, %esi
+; X86-NEXT:    movl 72(%esp,%eax), %ebx
+; X86-NEXT:    addl %ebx, %ebx
+; X86-NEXT:    shll %cl, %ebx
+; X86-NEXT:    orl %esi, %ebx
 ; X86-NEXT:    movb %ch, %cl
-; X86-NEXT:    shrdl %cl, %eax, %edx
-; X86-NEXT:    movl %edx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT:    shrdl %cl, %edx, %edi
+; X86-NEXT:    movl %edi, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
 ; X86-NEXT:    movl $0, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Folded Spill
 ; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
 ; X86-NEXT:    addl $-1, %eax
@@ -179,63 +180,62 @@ define void @f() nounwind {
 ; X86-NEXT:    .p2align 4, 0x90
 ; X86-NEXT:  .LBB0_10: # %udiv-do-while
 ; X86-NEXT:    # =>This Inner Loop Header: Depth=1
-; X86-NEXT:    movl %ebx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
-; X86-NEXT:    shldl $1, %esi, %ecx
-; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %edx # 4-byte Reload
-; X86-NEXT:    shldl $1, %edx, %esi
-; X86-NEXT:    movl %esi, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT:    movl %ebx, %edi
+; X86-NEXT:    shldl $1, %ebx, %ecx
 ; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %ebx # 4-byte Reload
-; X86-NEXT:    movl %ebx, %eax
-; X86-NEXT:    andl $2, %eax
-; X86-NEXT:    shrl %eax
-; X86-NEXT:    leal (%eax,%edx,2), %edx
-; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %edi # 4-byte Reload
-; X86-NEXT:    shldl $1, %edi, %ebx
+; X86-NEXT:    shldl $1, %ebx, %edi
 ; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %esi # 4-byte Reload
-; X86-NEXT:    orl %esi, %ebx
+; X86-NEXT:    movl %esi, %edx
+; X86-NEXT:    andl $2, %edx
+; X86-NEXT:    shrl %edx
+; X86-NEXT:    leal (%edx,%ebx,2), %ebx
+; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %edx # 4-byte Reload
+; X86-NEXT:    shldl $1, %edx, %esi
+; X86-NEXT:    orl {{[-0-9]+}}(%e{{[sb]}}p), %esi # 4-byte Folded Reload
 ; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
-; X86-NEXT:    shldl $1, %eax, %edi
-; X86-NEXT:    orl %esi, %edi
-; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %esi # 4-byte Reload
-; X86-NEXT:    movl %edi, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT:    shldl $1, %eax, %edx
+; X86-NEXT:    orl {{[-0-9]+}}(%e{{[sb]}}p), %edx # 4-byte Folded Reload
+; X86-NEXT:    movl %edx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
 ; X86-NEXT:    addl %eax, %eax
 ; X86-NEXT:    orl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Folded Reload
 ; X86-NEXT:    movl %eax, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
-; X86-NEXT:    andl $3, %ebx
-; X86-NEXT:    movl %ebx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
-; X86-NEXT:    cmpl %edx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Folded Reload
-; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
-; X86-NEXT:    sbbl %esi, %eax
-; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %ebx # 4-byte Reload
-; X86-NEXT:    sbbl %ecx, %ebx
-; X86-NEXT:    shll $30, %ebx
-; X86-NEXT:    movl %ebx, %eax
-; X86-NEXT:    sarl $30, %eax
-; X86-NEXT:    sarl $31, %ebx
-; X86-NEXT:    shrdl $1, %ebx, %eax
-; X86-NEXT:    movl %eax, %edi
-; X86-NEXT:    andl $1, %edi
-; X86-NEXT:    movl %edi, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT:    andl $3, %esi
+; X86-NEXT:    movl %esi, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT:    cmpl %ebx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Folded Reload
+; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %edx # 4-byte Reload
+; X86-NEXT:    sbbl %edi, %edx
+; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %esi # 4-byte Reload
+; X86-NEXT:    sbbl %ecx, %esi
+; X86-NEXT:    shll $30, %esi
+; X86-NEXT:    movl %esi, %edx
+; X86-NEXT:    sarl $30, %edx
+; X86-NEXT:    sarl $31, %esi
+; X86-NEXT:    shrdl $1, %esi, %edx
+; X86-NEXT:    movl %edx, %eax
+; X86-NEXT:    andl $1, %eax
+; X86-NEXT:    movl %eax, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT:    andl {{[-0-9]+}}(%e{{[sb]}}p), %edx # 4-byte Folded Reload
+; X86-NEXT:    movl %esi, %eax
 ; X86-NEXT:    andl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Folded Reload
-; X86-NEXT:    movl %ebx, %edi
-; X86-NEXT:    andl {{[-0-9]+}}(%e{{[sb]}}p), %edi # 4-byte Folded Reload
-; X86-NEXT:    andl {{[-0-9]+}}(%e{{[sb]}}p), %ebx # 4-byte Folded Reload
-; X86-NEXT:    subl %eax, %edx
-; X86-NEXT:    movl %edx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
-; X86-NEXT:    sbbl %ebx, %esi
-; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %ebx # 4-byte Reload
-; X86-NEXT:    sbbl %edi, %ecx
+; X86-NEXT:    andl {{[-0-9]+}}(%e{{[sb]}}p), %esi # 4-byte Folded Reload
+; X86-NEXT:    subl %edx, %ebx
+; X86-NEXT:    movl %ebx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT:    sbbl %esi, %edi
+; X86-NEXT:    movl %edi, %ebx
+; X86-NEXT:    sbbl %eax, %ecx
 ; X86-NEXT:    andl $3, %ecx
 ; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %eax # 4-byte Reload
 ; X86-NEXT:    addl $-1, %eax
-; X86-NEXT:    adcl $-1, %ebx
-; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %edi # 4-byte Reload
-; X86-NEXT:    adcl $3, %edi
-; X86-NEXT:    andl $3, %edi
+; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %edx # 4-byte Reload
+; X86-NEXT:    adcl $-1, %edx
+; X86-NEXT:    movl {{[-0-9]+}}(%e{{[sb]}}p), %esi # 4-byte Reload
+; X86-NEXT:    adcl $3, %esi
+; X86-NEXT:    andl $3, %esi
 ; X86-NEXT:    movl %eax, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
-; X86-NEXT:    movl %edi, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
-; X86-NEXT:    orl %edi, %eax
-; X86-NEXT:    orl %ebx, %eax
+; X86-NEXT:    movl %esi, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT:    orl %esi, %eax
+; X86-NEXT:    movl %edx, {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Spill
+; X86-NEXT:    orl %edx, %eax
 ; X86-NEXT:    jne .LBB0_10
 ; X86-NEXT:  .LBB0_11: # %udiv-end
 ; X86-NEXT:    cmpb $0, {{[-0-9]+}}(%e{{[sb]}}p) # 1-byte Folded Reload
diff --git a/llvm/test/CodeGen/X86/pr62286.ll b/llvm/test/CodeGen/X86/pr62286.ll
index 782c84408f25..1b13cee628df 100644
--- a/llvm/test/CodeGen/X86/pr62286.ll
+++ b/llvm/test/CodeGen/X86/pr62286.ll
@@ -8,21 +8,20 @@ define i64 @PR62286(i32 %a) {
 ; SSE-LABEL: PR62286:
 ; SSE:       # %bb.0:
 ; SSE-NEXT:    movd %edi, %xmm0
-; SSE-NEXT:    pshufd {{.*#+}} xmm1 = xmm0[0,1,1,0]
-; SSE-NEXT:    paddd %xmm1, %xmm1
+; SSE-NEXT:    pshufd {{.*#+}} xmm1 = xmm0[1,0,1,0]
+; SSE-NEXT:    paddd %xmm0, %xmm0
 ; SSE-NEXT:    pxor %xmm2, %xmm2
-; SSE-NEXT:    punpckldq {{.*#+}} xmm1 = xmm1[0],xmm2[0],xmm1[1],xmm2[1]
-; SSE-NEXT:    pxor %xmm3, %xmm3
-; SSE-NEXT:    pcmpgtd %xmm1, %xmm3
-; SSE-NEXT:    punpckldq {{.*#+}} xmm1 = xmm1[0],xmm3[0],xmm1[1],xmm3[1]
-; SSE-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[1,0,1,0]
-; SSE-NEXT:    pand {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0
-; SSE-NEXT:    pcmpgtd %xmm0, %xmm2
 ; SSE-NEXT:    punpckldq {{.*#+}} xmm0 = xmm0[0],xmm2[0],xmm0[1],xmm2[1]
-; SSE-NEXT:    paddq %xmm1, %xmm0
-; SSE-NEXT:    pshufd {{.*#+}} xmm1 = xmm0[2,3,2,3]
+; SSE-NEXT:    pxor %xmm3, %xmm3
+; SSE-NEXT:    pcmpgtd %xmm0, %xmm3
+; SSE-NEXT:    punpckldq {{.*#+}} xmm0 = xmm0[0],xmm3[0],xmm0[1],xmm3[1]
+; SSE-NEXT:    pand {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm1
+; SSE-NEXT:    pcmpgtd %xmm1, %xmm2
+; SSE-NEXT:    punpckldq {{.*#+}} xmm1 = xmm1[0],xmm2[0],xmm1[1],xmm2[1]
 ; SSE-NEXT:    paddq %xmm0, %xmm1
-; SSE-NEXT:    movq %xmm1, %rax
+; SSE-NEXT:    pshufd {{.*#+}} xmm0 = xmm1[2,3,2,3]
+; SSE-NEXT:    paddq %xmm1, %xmm0
+; SSE-NEXT:    movq %xmm0, %rax
 ; SSE-NEXT:    retq
 ;
 ; AVX1-LABEL: PR62286:
@@ -47,10 +46,10 @@ define i64 @PR62286(i32 %a) {
 ; AVX2-LABEL: PR62286:
 ; AVX2:       # %bb.0:
 ; AVX2-NEXT:    vmovd %edi, %xmm0
-; AVX2-NEXT:    vpslldq {{.*#+}} xmm1 = zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,xmm0[0,1,2,3]
-; AVX2-NEXT:    vpaddd %xmm0, %xmm0, %xmm0
-; AVX2-NEXT:    vpand {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm1, %xmm1
-; AVX2-NEXT:    vpor %xmm0, %xmm1, %xmm0
+; AVX2-NEXT:    vpaddd %xmm0, %xmm0, %xmm1
+; AVX2-NEXT:    vpslldq {{.*#+}} xmm0 = zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,zero,xmm0[0,1,2,3]
+; AVX2-NEXT:    vpand {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0, %xmm0
+; AVX2-NEXT:    vpblendd {{.*#+}} xmm0 = xmm1[0],xmm0[1,2,3]
 ; AVX2-NEXT:    vpmovsxdq %xmm0, %ymm0
 ; AVX2-NEXT:    vextracti128 $1, %ymm0, %xmm1
 ; AVX2-NEXT:    vpaddq %xmm1, %xmm0, %xmm0
diff --git a/llvm/test/CodeGen/X86/scheduler-backtracking.ll b/llvm/test/CodeGen/X86/scheduler-backtracking.ll
index a9f3e8b22fb6..785b97d8c240 100644
--- a/llvm/test/CodeGen/X86/scheduler-backtracking.ll
+++ b/llvm/test/CodeGen/X86/scheduler-backtracking.ll
@@ -14,7 +14,6 @@ define i256 @test1(i256 %a) nounwind {
 ; ILP:       # %bb.0:
 ; ILP-NEXT:    movq %rdi, %rax
 ; ILP-NEXT:    leal (%rsi,%rsi), %ecx
-; ILP-NEXT:    addb $3, %cl
 ; ILP-NEXT:    movq $0, -{{[0-9]+}}(%rsp)
 ; ILP-NEXT:    movq $0, -{{[0-9]+}}(%rsp)
 ; ILP-NEXT:    movq $0, -{{[0-9]+}}(%rsp)
@@ -23,6 +22,7 @@ define i256 @test1(i256 %a) nounwind {
 ; ILP-NEXT:    movq $0, -{{[0-9]+}}(%rsp)
 ; ILP-NEXT:    movq $0, -{{[0-9]+}}(%rsp)
 ; ILP-NEXT:    movq $0, -{{[0-9]+}}(%rsp)
+; ILP-NEXT:    addb $3, %cl
 ; ILP-NEXT:    movl %ecx, %edx
 ; ILP-NEXT:    shrb $3, %dl
 ; ILP-NEXT:    andb $7, %cl
diff --git a/llvm/test/CodeGen/X86/sdiv_fix_sat.ll b/llvm/test/CodeGen/X86/sdiv_fix_sat.ll
index 31297a06f809..a1cabb433d87 100644
--- a/llvm/test/CodeGen/X86/sdiv_fix_sat.ll
+++ b/llvm/test/CodeGen/X86/sdiv_fix_sat.ll
@@ -563,18 +563,20 @@ define <4 x i32> @vec(<4 x i32> %x, <4 x i32> %y) nounwind {
 ; X64-NEXT:    subq $120, %rsp
 ; X64-NEXT:    movdqa %xmm1, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
 ; X64-NEXT:    movdqa %xmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
-; X64-NEXT:    pshufd {{.*#+}} xmm3 = xmm0[2,1,3,3]
-; X64-NEXT:    psllq $32, %xmm3
+; X64-NEXT:    pxor %xmm3, %xmm3
+; X64-NEXT:    punpckhdq {{.*#+}} xmm3 = xmm3[2],xmm0[2],xmm3[3],xmm0[3]
 ; X64-NEXT:    pshufd {{.*#+}} xmm2 = xmm3[1,3,2,3]
 ; X64-NEXT:    psrad $31, %xmm2
 ; X64-NEXT:    psrlq $31, %xmm3
 ; X64-NEXT:    pshufd {{.*#+}} xmm0 = xmm3[0,2,2,3]
 ; X64-NEXT:    punpckldq {{.*#+}} xmm0 = xmm0[0],xmm2[0],xmm0[1],xmm2[1]
 ; X64-NEXT:    movdqa %xmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
-; X64-NEXT:    movq %xmm0, %rbx
-; X64-NEXT:    movq %rbx, %r13
-; X64-NEXT:    sarq $63, %r13
-; X64-NEXT:    shldq $31, %rbx, %r13
+; X64-NEXT:    movq %xmm0, %rbp
+; X64-NEXT:    movq %rbp, %r14
+; X64-NEXT:    sarq $63, %r14
+; X64-NEXT:    shldq $31, %rbp, %r14
+; X64-NEXT:    movq %rbp, %r15
+; X64-NEXT:    shlq $31, %r15
 ; X64-NEXT:    pshufd {{.*#+}} xmm1 = xmm1[2,3,2,3]
 ; X64-NEXT:    pxor %xmm0, %xmm0
 ; X64-NEXT:    pcmpgtd %xmm1, %xmm0
@@ -582,113 +584,113 @@ define <4 x i32> @vec(<4 x i32> %x, <4 x i32> %y) nounwind {
 ; X64-NEXT:    movdqa %xmm1, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
 ; X64-NEXT:    movq %xmm1, %rdx
 ; X64-NEXT:    movq %rdx, {{[-0-9]+}}(%r{{[sb]}}p) # 8-byte Spill
-; X64-NEXT:    movq %rdx, %r15
-; X64-NEXT:    sarq $63, %r15
-; X64-NEXT:    movq %rbx, %r12
-; X64-NEXT:    shlq $31, %r12
-; X64-NEXT:    movq %r12, %rdi
-; X64-NEXT:    movq %r13, %rsi
-; X64-NEXT:    movq %r15, %rcx
+; X64-NEXT:    movq %rdx, %rbx
+; X64-NEXT:    sarq $63, %rbx
+; X64-NEXT:    movq %r15, %rdi
+; X64-NEXT:    movq %r14, %rsi
+; X64-NEXT:    movq %rbx, %rcx
 ; X64-NEXT:    callq __divti3@PLT
-; X64-NEXT:    movq %rax, %rbp
+; X64-NEXT:    movq %rax, %r13
 ; X64-NEXT:    movq %rax, {{[-0-9]+}}(%r{{[sb]}}p) # 8-byte Spill
-; X64-NEXT:    movq %rdx, %r14
+; X64-NEXT:    movq %rdx, %r12
 ; X64-NEXT:    movq %rdx, {{[-0-9]+}}(%r{{[sb]}}p) # 8-byte Spill
-; X64-NEXT:    subq $1, %rbp
-; X64-NEXT:    sbbq $0, %r14
-; X64-NEXT:    shrq $63, %rbx
-; X64-NEXT:    xorl %r15d, %ebx
-; X64-NEXT:    movq %r12, %rdi
-; X64-NEXT:    movq %r13, %rsi
+; X64-NEXT:    subq $1, %r13
+; X64-NEXT:    sbbq $0, %r12
+; X64-NEXT:    movq %r15, %rdi
+; X64-NEXT:    movq %r14, %rsi
 ; X64-NEXT:    movq {{[-0-9]+}}(%r{{[sb]}}p), %rdx # 8-byte Reload
-; X64-NEXT:    movq %r15, %rcx
+; X64-NEXT:    movq %rbx, %rcx
 ; X64-NEXT:    callq __modti3@PLT
 ; X64-NEXT:    orq %rax, %rdx
 ; X64-NEXT:    setne %al
+; X64-NEXT:    shrq $63, %rbp
+; X64-NEXT:    xorl %ebp, %ebx
 ; X64-NEXT:    testb %bl, %al
-; X64-NEXT:    cmoveq {{[-0-9]+}}(%r{{[sb]}}p), %r14 # 8-byte Folded Reload
-; X64-NEXT:    cmoveq {{[-0-9]+}}(%r{{[sb]}}p), %rbp # 8-byte Folded Reload
+; X64-NEXT:    cmoveq {{[-0-9]+}}(%r{{[sb]}}p), %r13 # 8-byte Folded Reload
+; X64-NEXT:    cmoveq {{[-0-9]+}}(%r{{[sb]}}p), %r12 # 8-byte Folded Reload
 ; X64-NEXT:    xorl %ecx, %ecx
 ; X64-NEXT:    movl $4294967295, %edx # imm = 0xFFFFFFFF
-; X64-NEXT:    cmpq %rdx, %rbp
-; X64-NEXT:    movq %r14, %rax
+; X64-NEXT:    cmpq %rdx, %r13
+; X64-NEXT:    movq %r12, %rax
 ; X64-NEXT:    sbbq $0, %rax
-; X64-NEXT:    cmovgeq %rcx, %r14
-; X64-NEXT:    cmovgeq %rdx, %rbp
+; X64-NEXT:    cmovgeq %rdx, %r13
+; X64-NEXT:    cmovgeq %rcx, %r12
 ; X64-NEXT:    movabsq $-4294967296, %rcx # imm = 0xFFFFFFFF00000000
-; X64-NEXT:    cmpq %rbp, %rcx
+; X64-NEXT:    cmpq %r13, %rcx
 ; X64-NEXT:    movq $-1, {{[-0-9]+}}(%r{{[sb]}}p) # 8-byte Folded Spill
 ; X64-NEXT:    movq $-1, %rax
-; X64-NEXT:    sbbq %r14, %rax
-; X64-NEXT:    cmovgeq %rcx, %rbp
-; X64-NEXT:    movq %rbp, %xmm0
+; X64-NEXT:    sbbq %r12, %rax
+; X64-NEXT:    cmovgeq %rcx, %r13
+; X64-NEXT:    movq %r13, %xmm0
 ; X64-NEXT:    movdqa %xmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
 ; X64-NEXT:    pshufd $238, {{[-0-9]+}}(%r{{[sb]}}p), %xmm0 # 16-byte Folded Reload
 ; X64-NEXT:    # xmm0 = mem[2,3,2,3]
-; X64-NEXT:    movq %xmm0, %rbx
-; X64-NEXT:    movq %rbx, %r13
-; X64-NEXT:    sarq $63, %r13
-; X64-NEXT:    shldq $31, %rbx, %r13
+; X64-NEXT:    movq %xmm0, %rbp
+; X64-NEXT:    movq %rbp, %r14
+; X64-NEXT:    sarq $63, %r14
+; X64-NEXT:    shldq $31, %rbp, %r14
+; X64-NEXT:    movq %rbp, %r15
+; X64-NEXT:    shlq $31, %r15
 ; X64-NEXT:    pshufd $238, {{[-0-9]+}}(%r{{[sb]}}p), %xmm0 # 16-byte Folded Reload
 ; X64-NEXT:    # xmm0 = mem[2,3,2,3]
 ; X64-NEXT:    movq %xmm0, %rdx
 ; X64-NEXT:    movq %rdx, {{[-0-9]+}}(%r{{[sb]}}p) # 8-byte Spill
-; X64-NEXT:    movq %rdx, %r15
-; X64-NEXT:    sarq $63, %r15
-; X64-NEXT:    movq %rbx, %r12
-; X64-NEXT:    shlq $31, %r12
-; X64-NEXT:    movq %r12, %rdi
-; X64-NEXT:    movq %r13, %rsi
-; X64-NEXT:    movq %r15, %rcx
+; X64-NEXT:    movq %rdx, %rbx
+; X64-NEXT:    sarq $63, %rbx
+; X64-NEXT:    movq %r15, %rdi
+; X64-NEXT:    movq %r14, %rsi
+; X64-NEXT:    movq %rbx, %rcx
 ; X64-NEXT:    callq __divti3@PLT
-; X64-NEXT:    movq %rax, %rbp
+; X64-NEXT:    movq %rax, %r13
 ; X64-NEXT:    movq %rax, {{[-0-9]+}}(%r{{[sb]}}p) # 8-byte Spill
-; X64-NEXT:    movq %rdx, %r14
+; X64-NEXT:    movq %rdx, %r12
 ; X64-NEXT:    movq %rdx, {{[-0-9]+}}(%r{{[sb]}}p) # 8-byte Spill
-; X64-NEXT:    subq $1, %rbp
-; X64-NEXT:    sbbq $0, %r14
-; X64-NEXT:    shrq $63, %rbx
-; X64-NEXT:    xorl %r15d, %ebx
-; X64-NEXT:    movq %r12, %rdi
-; X64-NEXT:    movq %r13, %rsi
+; X64-NEXT:    subq $1, %r13
+; X64-NEXT:    sbbq $0, %r12
+; X64-NEXT:    movq %r15, %rdi
+; X64-NEXT:    movq %r14, %rsi
 ; X64-NEXT:    movq {{[-0-9]+}}(%r{{[sb]}}p), %rdx # 8-byte Reload
-; X64-NEXT:    movq %r15, %rcx
+; X64-NEXT:    movq %rbx, %rcx
 ; X64-NEXT:    callq __modti3@PLT
 ; X64-NEXT:    orq %rax, %rdx
 ; X64-NEXT:    setne %al
+; X64-NEXT:    shrq $63, %rbp
+; X64-NEXT:    xorl %ebp, %ebx
 ; X64-NEXT:    testb %bl, %al
-; X64-NEXT:    cmoveq {{[-0-9]+}}(%r{{[sb]}}p), %r14 # 8-byte Folded Reload
-; X64-NEXT:    cmoveq {{[-0-9]+}}(%r{{[sb]}}p), %rbp # 8-byte Folded Reload
+; X64-NEXT:    cmoveq {{[-0-9]+}}(%r{{[sb]}}p), %r13 # 8-byte Folded Reload
+; X64-NEXT:    cmoveq {{[-0-9]+}}(%r{{[sb]}}p), %r12 # 8-byte Folded Reload
 ; X64-NEXT:    movl $4294967295, %ecx # imm = 0xFFFFFFFF
-; X64-NEXT:    cmpq %rcx, %rbp
-; X64-NEXT:    movq %r14, %rax
+; X64-NEXT:    cmpq %rcx, %r13
+; X64-NEXT:    movq %r12, %rax
 ; X64-NEXT:    sbbq $0, %rax
+; X64-NEXT:    cmovgeq %rcx, %r13
 ; X64-NEXT:    movl $0, %eax
-; X64-NEXT:    cmovgeq %rax, %r14
-; X64-NEXT:    cmovgeq %rcx, %rbp
+; X64-NEXT:    cmovgeq %rax, %r12
 ; X64-NEXT:    movabsq $-4294967296, %rcx # imm = 0xFFFFFFFF00000000
-; X64-NEXT:    cmpq %rbp, %rcx
+; X64-NEXT:    cmpq %r13, %rcx
 ; X64-NEXT:    movq $-1, %rax
-; X64-NEXT:    sbbq %r14, %rax
-; X64-NEXT:    cmovgeq %rcx, %rbp
-; X64-NEXT:    movq %rbp, %xmm0
+; X64-NEXT:    sbbq %r12, %rax
+; X64-NEXT:    cmovgeq %rcx, %r13
+; X64-NEXT:    movq %r13, %xmm0
 ; X64-NEXT:    movdqa {{[-0-9]+}}(%r{{[sb]}}p), %xmm1 # 16-byte Reload
 ; X64-NEXT:    punpcklqdq {{.*#+}} xmm1 = xmm1[0],xmm0[0]
 ; X64-NEXT:    psrlq $1, %xmm1
 ; X64-NEXT:    movdqa %xmm1, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
-; X64-NEXT:    pshufd $212, {{[-0-9]+}}(%r{{[sb]}}p), %xmm0 # 16-byte Folded Reload
-; X64-NEXT:    # xmm0 = mem[0,1,1,3]
-; X64-NEXT:    psllq $32, %xmm0
+; X64-NEXT:    pxor %xmm0, %xmm0
+; X64-NEXT:    punpckldq {{[-0-9]+}}(%r{{[sb]}}p), %xmm0 # 16-byte Folded Reload
+; X64-NEXT:    # xmm0 = xmm0[0],mem[0],xmm0[1],mem[1]
 ; X64-NEXT:    pshufd {{.*#+}} xmm1 = xmm0[1,3,2,3]
 ; X64-NEXT:    psrad $31, %xmm1
 ; X64-NEXT:    psrlq $31, %xmm0
 ; X64-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[0,2,2,3]
 ; X64-NEXT:    punpckldq {{.*#+}} xmm0 = xmm0[0],xmm1[0],xmm0[1],xmm1[1]
 ; X64-NEXT:    movdqa %xmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
-; X64-NEXT:    movq %xmm0, %rbx
-; X64-NEXT:    movq %rbx, %r13
-; X64-NEXT:    sarq $63, %r13
-; X64-NEXT:    shldq $31, %rbx, %r13
+; X64-NEXT:    movq %xmm0, %rbp
+; X64-NEXT:    movq %rbp, %r14
+; X64-NEXT:    sarq $63, %r14
+; X64-NEXT:    shldq $31, %rbp, %r14
+; X64-NEXT:    movq %rbp, %r15
+; X64-NEXT:    shlq $31, %r15
 ; X64-NEXT:    movdqa {{[-0-9]+}}(%r{{[sb]}}p), %xmm0 # 16-byte Reload
 ; X64-NEXT:    pxor %xmm1, %xmm1
 ; X64-NEXT:    pcmpgtd %xmm0, %xmm1
@@ -696,94 +698,92 @@ define <4 x i32> @vec(<4 x i32> %x, <4 x i32> %y) nounwind {
 ; X64-NEXT:    movdqa %xmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
 ; X64-NEXT:    movq %xmm0, %rdx
 ; X64-NEXT:    movq %rdx, {{[-0-9]+}}(%r{{[sb]}}p) # 8-byte Spill
-; X64-NEXT:    movq %rdx, %r15
-; X64-NEXT:    sarq $63, %r15
-; X64-NEXT:    movq %rbx, %r12
-; X64-NEXT:    shlq $31, %r12
-; X64-NEXT:    movq %r12, %rdi
-; X64-NEXT:    movq %r13, %rsi
-; X64-NEXT:    movq %r15, %rcx
+; X64-NEXT:    movq %rdx, %rbx
+; X64-NEXT:    sarq $63, %rbx
+; X64-NEXT:    movq %r15, %rdi
+; X64-NEXT:    movq %r14, %rsi
+; X64-NEXT:    movq %rbx, %rcx
 ; X64-NEXT:    callq __divti3@PLT
-; X64-NEXT:    movq %rax, %rbp
+; X64-NEXT:    movq %rax, %r13
 ; X64-NEXT:    movq %rax, {{[-0-9]+}}(%r{{[sb]}}p) # 8-byte Spill
-; X64-NEXT:    movq %rdx, %r14
+; X64-NEXT:    movq %rdx, %r12
 ; X64-NEXT:    movq %rdx, {{[-0-9]+}}(%r{{[sb]}}p) # 8-byte Spill
-; X64-NEXT:    subq $1, %rbp
-; X64-NEXT:    sbbq $0, %r14
-; X64-NEXT:    shrq $63, %rbx
-; X64-NEXT:    xorl %r15d, %ebx
-; X64-NEXT:    movq %r12, %rdi
-; X64-NEXT:    movq %r13, %rsi
+; X64-NEXT:    subq $1, %r13
+; X64-NEXT:    sbbq $0, %r12
+; X64-NEXT:    movq %r15, %rdi
+; X64-NEXT:    movq %r14, %rsi
 ; X64-NEXT:    movq {{[-0-9]+}}(%r{{[sb]}}p), %rdx # 8-byte Reload
-; X64-NEXT:    movq %r15, %rcx
+; X64-NEXT:    movq %rbx, %rcx
 ; X64-NEXT:    callq __modti3@PLT
 ; X64-NEXT:    orq %rax, %rdx
 ; X64-NEXT:    setne %al
+; X64-NEXT:    shrq $63, %rbp
+; X64-NEXT:    xorl %ebp, %ebx
 ; X64-NEXT:    testb %bl, %al
-; X64-NEXT:    cmoveq {{[-0-9]+}}(%r{{[sb]}}p), %r14 # 8-byte Folded Reload
-; X64-NEXT:    cmoveq {{[-0-9]+}}(%r{{[sb]}}p), %rbp # 8-byte Folded Reload
+; X64-NEXT:    cmoveq {{[-0-9]+}}(%r{{[sb]}}p), %r13 # 8-byte Folded Reload
+; X64-NEXT:    cmoveq {{[-0-9]+}}(%r{{[sb]}}p), %r12 # 8-byte Folded Reload
 ; X64-NEXT:    movl $4294967295, %ecx # imm = 0xFFFFFFFF
-; X64-NEXT:    cmpq %rcx, %rbp
-; X64-NEXT:    movq %r14, %rax
+; X64-NEXT:    cmpq %rcx, %r13
+; X64-NEXT:    movq %r12, %rax
 ; X64-NEXT:    sbbq $0, %rax
+; X64-NEXT:    cmovgeq %rcx, %r13
 ; X64-NEXT:    movl $0, %eax
-; X64-NEXT:    cmovgeq %rax, %r14
-; X64-NEXT:    cmovgeq %rcx, %rbp
+; X64-NEXT:    cmovgeq %rax, %r12
 ; X64-NEXT:    movabsq $-4294967296, %rcx # imm = 0xFFFFFFFF00000000
-; X64-NEXT:    cmpq %rbp, %rcx
+; X64-NEXT:    cmpq %r13, %rcx
 ; X64-NEXT:    movq $-1, %rax
-; X64-NEXT:    sbbq %r14, %rax
-; X64-NEXT:    cmovgeq %rcx, %rbp
-; X64-NEXT:    movq %rbp, %xmm0
+; X64-NEXT:    sbbq %r12, %rax
+; X64-NEXT:    cmovgeq %rcx, %r13
+; X64-NEXT:    movq %r13, %xmm0
 ; X64-NEXT:    movdqa %xmm0, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
 ; X64-NEXT:    pshufd $238, {{[-0-9]+}}(%r{{[sb]}}p), %xmm0 # 16-byte Folded Reload
 ; X64-NEXT:    # xmm0 = mem[2,3,2,3]
-; X64-NEXT:    movq %xmm0, %rbx
-; X64-NEXT:    movq %rbx, %r13
-; X64-NEXT:    sarq $63, %r13
-; X64-NEXT:    shldq $31, %rbx, %r13
+; X64-NEXT:    movq %xmm0, %rbp
+; X64-NEXT:    movq %rbp, %r14
+; X64-NEXT:    sarq $63, %r14
+; X64-NEXT:    shldq $31, %rbp, %r14
+; X64-NEXT:    movq %rbp, %r15
+; X64-NEXT:    shlq $31, %r15
 ; X64-NEXT:    pshufd $238, {{[-0-9]+}}(%r{{[sb]}}p), %xmm0 # 16-byte Folded Reload
 ; X64-NEXT:    # xmm0 = mem[2,3,2,3]
 ; X64-NEXT:    movq %xmm0, %rdx
 ; X64-NEXT:    movq %rdx, {{[-0-9]+}}(%r{{[sb]}}p) # 8-byte Spill
-; X64-NEXT:    movq %rdx, %r15
-; X64-NEXT:    sarq $63, %r15
-; X64-NEXT:    movq %rbx, %r12
-; X64-NEXT:    shlq $31, %r12
-; X64-NEXT:    movq %r12, %rdi
-; X64-NEXT:    movq %r13, %rsi
-; X64-NEXT:    movq %r15, %rcx
+; X64-NEXT:    movq %rdx, %rbx
+; X64-NEXT:    sarq $63, %rbx
+; X64-NEXT:    movq %r15, %rdi
+; X64-NEXT:    movq %r14, %rsi
+; X64-NEXT:    movq %rbx, %rcx
 ; X64-NEXT:    callq __divti3@PLT
-; X64-NEXT:    movq %rax, %rbp
+; X64-NEXT:    movq %rax, %r13
 ; X64-NEXT:    movq %rax, {{[-0-9]+}}(%r{{[sb]}}p) # 8-byte Spill
-; X64-NEXT:    movq %rdx, %r14
+; X64-NEXT:    movq %rdx, %r12
 ; X64-NEXT:    movq %rdx, {{[-0-9]+}}(%r{{[sb]}}p) # 8-byte Spill
-; X64-NEXT:    subq $1, %rbp
-; X64-NEXT:    sbbq $0, %r14
-; X64-NEXT:    shrq $63, %rbx
-; X64-NEXT:    xorl %r15d, %ebx
-; X64-NEXT:    movq %r12, %rdi
-; X64-NEXT:    movq %r13, %rsi
+; X64-NEXT:    subq $1, %r13
+; X64-NEXT:    sbbq $0, %r12
+; X64-NEXT:    movq %r15, %rdi
+; X64-NEXT:    movq %r14, %rsi
 ; X64-NEXT:    movq {{[-0-9]+}}(%r{{[sb]}}p), %rdx # 8-byte Reload
-; X64-NEXT:    movq %r15, %rcx
+; X64-NEXT:    movq %rbx, %rcx
 ; X64-NEXT:    callq __modti3@PLT
 ; X64-NEXT:    orq %rax, %rdx
 ; X64-NEXT:    setne %al
+; X64-NEXT:    shrq $63, %rbp
+; X64-NEXT:    xorl %ebp, %ebx
 ; X64-NEXT:    testb %bl, %al
-; X64-NEXT:    cmoveq {{[-0-9]+}}(%r{{[sb]}}p), %r14 # 8-byte Folded Reload
-; X64-NEXT:    cmoveq {{[-0-9]+}}(%r{{[sb]}}p), %rbp # 8-byte Folded Reload
+; X64-NEXT:    cmoveq {{[-0-9]+}}(%r{{[sb]}}p), %r13 # 8-byte Folded Reload
+; X64-NEXT:    cmoveq {{[-0-9]+}}(%r{{[sb]}}p), %r12 # 8-byte Folded Reload
 ; X64-NEXT:    movl $4294967295, %ecx # imm = 0xFFFFFFFF
-; X64-NEXT:    cmpq %rcx, %rbp
-; X64-NEXT:    movq %r14, %rax
+; X64-NEXT:    cmpq %rcx, %r13
+; X64-NEXT:    movq %r12, %rax
 ; X64-NEXT:    sbbq $0, %rax
+; X64-NEXT:    cmovgeq %rcx, %r13
 ; X64-NEXT:    movl $0, %eax
-; X64-NEXT:    cmovgeq %rax, %r14
-; X64-NEXT:    cmovgeq %rcx, %rbp
+; X64-NEXT:    cmovgeq %rax, %r12
 ; X64-NEXT:    movabsq $-4294967296, %rax # imm = 0xFFFFFFFF00000000
-; X64-NEXT:    cmpq %rbp, %rax
-; X64-NEXT:    sbbq %r14, {{[-0-9]+}}(%r{{[sb]}}p) # 8-byte Folded Spill
-; X64-NEXT:    cmovgeq %rax, %rbp
-; X64-NEXT:    movq %rbp, %xmm1
+; X64-NEXT:    cmpq %r13, %rax
+; X64-NEXT:    sbbq %r12, {{[-0-9]+}}(%r{{[sb]}}p) # 8-byte Folded Spill
+; X64-NEXT:    cmovgeq %rax, %r13
+; X64-NEXT:    movq %r13, %xmm1
 ; X64-NEXT:    movdqa {{[-0-9]+}}(%r{{[sb]}}p), %xmm0 # 16-byte Reload
 ; X64-NEXT:    punpcklqdq {{.*#+}} xmm0 = xmm0[0],xmm1[0]
 ; X64-NEXT:    psrlq $1, %xmm0
diff --git a/llvm/test/CodeGen/X86/setcc-non-simple-type.ll b/llvm/test/CodeGen/X86/setcc-non-simple-type.ll
index 97c3c2040b29..a80d8d8cd01b 100644
--- a/llvm/test/CodeGen/X86/setcc-non-simple-type.ll
+++ b/llvm/test/CodeGen/X86/setcc-non-simple-type.ll
@@ -46,7 +46,6 @@ define void @failing(ptr %0, ptr %1) nounwind {
 ; CHECK-NEXT:    movq 24(%rsi), %rcx
 ; CHECK-NEXT:    movq 32(%rsi), %rdx
 ; CHECK-NEXT:    movdqa {{.*#+}} xmm0 = [0,0,0,0,0,0,0,0,1,0,0,0,0,0,0,0]
-; CHECK-NEXT:    xorl %esi, %esi
 ; CHECK-NEXT:    movdqa {{.*#+}} xmm1 = [1,1]
 ; CHECK-NEXT:    movdqa {{.*#+}} xmm2 = [2,2]
 ; CHECK-NEXT:    .p2align 4, 0x90
@@ -54,39 +53,45 @@ define void @failing(ptr %0, ptr %1) nounwind {
 ; CHECK-NEXT:    # =>This Loop Header: Depth=1
 ; CHECK-NEXT:    # Child Loop BB0_2 Depth 2
 ; CHECK-NEXT:    xorpd %xmm3, %xmm3
-; CHECK-NEXT:    movq $-1024, %rdi # imm = 0xFC00
+; CHECK-NEXT:    movq $-1024, %rsi # imm = 0xFC00
 ; CHECK-NEXT:    movdqa %xmm0, %xmm4
 ; CHECK-NEXT:    .p2align 4, 0x90
 ; CHECK-NEXT:  .LBB0_2: # %vector.body
 ; CHECK-NEXT:    # Parent Loop BB0_1 Depth=1
 ; CHECK-NEXT:    # => This Inner Loop Header: Depth=2
-; CHECK-NEXT:    cmpq 1024(%rdx,%rdi), %rsi
-; CHECK-NEXT:    movq %rcx, %r8
-; CHECK-NEXT:    sbbq 1032(%rdx,%rdi), %r8
-; CHECK-NEXT:    setge %r8b
-; CHECK-NEXT:    movzbl %r8b, %r8d
-; CHECK-NEXT:    andl $1, %r8d
+; CHECK-NEXT:    movdqu 1024(%rdx,%rsi), %xmm5
+; CHECK-NEXT:    movdqu 1040(%rdx,%rsi), %xmm6
+; CHECK-NEXT:    movq %xmm5, %rdi
+; CHECK-NEXT:    movq %xmm6, %r8
+; CHECK-NEXT:    pshufd {{.*#+}} xmm5 = xmm5[2,3,2,3]
+; CHECK-NEXT:    movq %xmm5, %r9
+; CHECK-NEXT:    pshufd {{.*#+}} xmm5 = xmm6[2,3,2,3]
+; CHECK-NEXT:    movq %xmm5, %r10
 ; CHECK-NEXT:    negq %r8
-; CHECK-NEXT:    movq %r8, %xmm5
-; CHECK-NEXT:    cmpq 1040(%rdx,%rdi), %rsi
 ; CHECK-NEXT:    movq %rcx, %r8
-; CHECK-NEXT:    sbbq 1048(%rdx,%rdi), %r8
+; CHECK-NEXT:    sbbq %r10, %r8
 ; CHECK-NEXT:    setge %r8b
 ; CHECK-NEXT:    movzbl %r8b, %r8d
-; CHECK-NEXT:    andl $1, %r8d
 ; CHECK-NEXT:    negq %r8
-; CHECK-NEXT:    movq %r8, %xmm6
-; CHECK-NEXT:    punpcklqdq {{.*#+}} xmm5 = xmm5[0],xmm6[0]
-; CHECK-NEXT:    movdqa %xmm1, %xmm6
-; CHECK-NEXT:    psllq %xmm4, %xmm6
+; CHECK-NEXT:    movq %r8, %xmm5
+; CHECK-NEXT:    negq %rdi
+; CHECK-NEXT:    movq %rcx, %rdi
+; CHECK-NEXT:    sbbq %r9, %rdi
+; CHECK-NEXT:    setge %dil
+; CHECK-NEXT:    movzbl %dil, %edi
+; CHECK-NEXT:    negq %rdi
+; CHECK-NEXT:    movq %rdi, %xmm6
+; CHECK-NEXT:    punpcklqdq {{.*#+}} xmm6 = xmm6[0],xmm5[0]
+; CHECK-NEXT:    movdqa %xmm1, %xmm5
+; CHECK-NEXT:    psllq %xmm4, %xmm5
 ; CHECK-NEXT:    pshufd {{.*#+}} xmm7 = xmm4[2,3,2,3]
 ; CHECK-NEXT:    movdqa %xmm1, %xmm8
 ; CHECK-NEXT:    psllq %xmm7, %xmm8
-; CHECK-NEXT:    movsd {{.*#+}} xmm8 = xmm6[0],xmm8[1]
-; CHECK-NEXT:    andpd %xmm5, %xmm8
+; CHECK-NEXT:    movsd {{.*#+}} xmm8 = xmm5[0],xmm8[1]
+; CHECK-NEXT:    andpd %xmm6, %xmm8
 ; CHECK-NEXT:    orpd %xmm8, %xmm3
 ; CHECK-NEXT:    paddq %xmm2, %xmm4
-; CHECK-NEXT:    addq $32, %rdi
+; CHECK-NEXT:    addq $32, %rsi
 ; CHECK-NEXT:    jne .LBB0_2
 ; CHECK-NEXT:  # %bb.3: # %middle.block
 ; CHECK-NEXT:    # in Loop: Header=BB0_1 Depth=1
@@ -101,7 +106,6 @@ define void @failing(ptr %0, ptr %1) nounwind {
 ; CHECK-AVX2-NEXT:    movq 24(%rsi), %rcx
 ; CHECK-AVX2-NEXT:    movq 32(%rsi), %rdx
 ; CHECK-AVX2-NEXT:    vpmovsxbq {{.*#+}} xmm0 = [0,1]
-; CHECK-AVX2-NEXT:    xorl %esi, %esi
 ; CHECK-AVX2-NEXT:    vpmovsxbq {{.*#+}} xmm1 = [1,1]
 ; CHECK-AVX2-NEXT:    vpmovsxbq {{.*#+}} xmm2 = [2,2]
 ; CHECK-AVX2-NEXT:    .p2align 4, 0x90
@@ -109,34 +113,40 @@ define void @failing(ptr %0, ptr %1) nounwind {
 ; CHECK-AVX2-NEXT:    # =>This Loop Header: Depth=1
 ; CHECK-AVX2-NEXT:    # Child Loop BB0_2 Depth 2
 ; CHECK-AVX2-NEXT:    vpxor %xmm3, %xmm3, %xmm3
-; CHECK-AVX2-NEXT:    movq $-1024, %rdi # imm = 0xFC00
+; CHECK-AVX2-NEXT:    movq $-1024, %rsi # imm = 0xFC00
 ; CHECK-AVX2-NEXT:    vmovdqa %xmm0, %xmm4
 ; CHECK-AVX2-NEXT:    .p2align 4, 0x90
 ; CHECK-AVX2-NEXT:  .LBB0_2: # %vector.body
 ; CHECK-AVX2-NEXT:    # Parent Loop BB0_1 Depth=1
 ; CHECK-AVX2-NEXT:    # => This Inner Loop Header: Depth=2
-; CHECK-AVX2-NEXT:    cmpq 1024(%rdx,%rdi), %rsi
-; CHECK-AVX2-NEXT:    movq %rcx, %r8
-; CHECK-AVX2-NEXT:    sbbq 1032(%rdx,%rdi), %r8
+; CHECK-AVX2-NEXT:    vmovdqu 1024(%rdx,%rsi), %xmm5
+; CHECK-AVX2-NEXT:    vmovdqu 1040(%rdx,%rsi), %xmm6
+; CHECK-AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm7 = xmm5[0],xmm6[0]
+; CHECK-AVX2-NEXT:    vpunpckhqdq {{.*#+}} xmm5 = xmm5[1],xmm6[1]
+; CHECK-AVX2-NEXT:    vmovq %xmm5, %rdi
+; CHECK-AVX2-NEXT:    vpextrq $1, %xmm5, %r8
+; CHECK-AVX2-NEXT:    vmovq %xmm7, %r9
+; CHECK-AVX2-NEXT:    vpextrq $1, %xmm7, %r10
+; CHECK-AVX2-NEXT:    negq %r10
+; CHECK-AVX2-NEXT:    movq %rcx, %r10
+; CHECK-AVX2-NEXT:    sbbq %r8, %r10
 ; CHECK-AVX2-NEXT:    setge %r8b
 ; CHECK-AVX2-NEXT:    movzbl %r8b, %r8d
-; CHECK-AVX2-NEXT:    andl $1, %r8d
 ; CHECK-AVX2-NEXT:    negq %r8
 ; CHECK-AVX2-NEXT:    vmovq %r8, %xmm5
-; CHECK-AVX2-NEXT:    cmpq 1040(%rdx,%rdi), %rsi
+; CHECK-AVX2-NEXT:    negq %r9
 ; CHECK-AVX2-NEXT:    movq %rcx, %r8
-; CHECK-AVX2-NEXT:    sbbq 1048(%rdx,%rdi), %r8
-; CHECK-AVX2-NEXT:    setge %r8b
-; CHECK-AVX2-NEXT:    movzbl %r8b, %r8d
-; CHECK-AVX2-NEXT:    andl $1, %r8d
-; CHECK-AVX2-NEXT:    negq %r8
-; CHECK-AVX2-NEXT:    vmovq %r8, %xmm6
-; CHECK-AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm5 = xmm5[0],xmm6[0]
+; CHECK-AVX2-NEXT:    sbbq %rdi, %r8
+; CHECK-AVX2-NEXT:    setge %dil
+; CHECK-AVX2-NEXT:    movzbl %dil, %edi
+; CHECK-AVX2-NEXT:    negq %rdi
+; CHECK-AVX2-NEXT:    vmovq %rdi, %xmm6
+; CHECK-AVX2-NEXT:    vpunpcklqdq {{.*#+}} xmm5 = xmm6[0],xmm5[0]
 ; CHECK-AVX2-NEXT:    vpsllvq %xmm4, %xmm1, %xmm6
 ; CHECK-AVX2-NEXT:    vpand %xmm6, %xmm5, %xmm5
 ; CHECK-AVX2-NEXT:    vpor %xmm3, %xmm5, %xmm3
 ; CHECK-AVX2-NEXT:    vpaddq %xmm2, %xmm4, %xmm4
-; CHECK-AVX2-NEXT:    addq $32, %rdi
+; CHECK-AVX2-NEXT:    addq $32, %rsi
 ; CHECK-AVX2-NEXT:    jne .LBB0_2
 ; CHECK-AVX2-NEXT:  # %bb.3: # %middle.block
 ; CHECK-AVX2-NEXT:    # in Loop: Header=BB0_1 Depth=1
diff --git a/llvm/test/CodeGen/X86/shift-combine.ll b/llvm/test/CodeGen/X86/shift-combine.ll
index bb0fd9c68afb..30c3d53dd37c 100644
--- a/llvm/test/CodeGen/X86/shift-combine.ll
+++ b/llvm/test/CodeGen/X86/shift-combine.ll
@@ -787,3 +787,38 @@ define <4 x i32> @or_tree_with_mismatching_shifts_vec_i32(<4 x i32> %a, <4 x i32
   %r = or <4 x i32> %or.ab, %or.cd
   ret <4 x i32> %r
 }
+
+; Reproducer for a DAGCombiner::combineShiftOfShiftedLogic bug. DAGCombiner
+; need to check that the sum of the shift amounts fits in i8, which is the
+; legal type used to described X86 shift amounts. Verify that we do not try to
+; create a shift with 130+160 as shift amount, and verify that the stored
+; value do not depend on %a1.
+define void @combineShiftOfShiftedLogic(i128 %a1, i32 %a2, ptr %p) {
+; X86-LABEL: combineShiftOfShiftedLogic:
+; X86:       # %bb.0:
+; X86-NEXT:    movl {{[0-9]+}}(%esp), %eax
+; X86-NEXT:    movl {{[0-9]+}}(%esp), %ecx
+; X86-NEXT:    movl %eax, 20(%ecx)
+; X86-NEXT:    movl $0, 16(%ecx)
+; X86-NEXT:    movl $0, 12(%ecx)
+; X86-NEXT:    movl $0, 8(%ecx)
+; X86-NEXT:    movl $0, 4(%ecx)
+; X86-NEXT:    movl $0, (%ecx)
+; X86-NEXT:    retl
+;
+; X64-LABEL: combineShiftOfShiftedLogic:
+; X64:       # %bb.0:
+; X64-NEXT:    # kill: def $edx killed $edx def $rdx
+; X64-NEXT:    shlq $32, %rdx
+; X64-NEXT:    movq %rdx, 16(%rcx)
+; X64-NEXT:    movq $0, 8(%rcx)
+; X64-NEXT:    movq $0, (%rcx)
+; X64-NEXT:    retq
+  %zext1 = zext i128 %a1 to i192
+  %zext2 = zext i32 %a2 to i192
+  %shl = shl i192 %zext1, 130
+  %or = or i192 %shl, %zext2
+  %res = shl i192 %or, 160
+  store i192 %res, ptr %p, align 8
+  ret void
+}
diff --git a/llvm/test/CodeGen/X86/subreg-to-reg-1.ll b/llvm/test/CodeGen/X86/subreg-to-reg-1.ll
index 8acdb6176f57..49bff9e075e5 100644
--- a/llvm/test/CodeGen/X86/subreg-to-reg-1.ll
+++ b/llvm/test/CodeGen/X86/subreg-to-reg-1.ll
@@ -1,14 +1,17 @@
+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --version 4
 ; RUN: llc < %s -mtriple=x86_64-- | FileCheck %s
 
-; CHECK:     {{leal	.*[)], %e.*}}
-; CHECK-NOT: {{leal	.*[)], %e.*}}
-
 ; Don't eliminate or coalesce away the explicit zero-extension!
 ; This is currently using an leal because of a 3-addressification detail,
 ; though this isn't necessary; The point of this test is to make sure
 ; a 32-bit add is used.
 
 define i64 @foo(i64 %a) nounwind {
+; CHECK-LABEL: foo:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    leal -1(%rdi), %eax
+; CHECK-NEXT:    incq %rax
+; CHECK-NEXT:    retq
   %b = add i64 %a, 4294967295
   %c = and i64 %b, 4294967295
   %d = add i64 %c, 1
diff --git a/llvm/test/CodeGen/X86/subreg-to-reg-3.ll b/llvm/test/CodeGen/X86/subreg-to-reg-3.ll
index db9d0d12c3d7..2bd5ca1716af 100644
--- a/llvm/test/CodeGen/X86/subreg-to-reg-3.ll
+++ b/llvm/test/CodeGen/X86/subreg-to-reg-3.ll
@@ -1,10 +1,14 @@
+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --version 4
 ; RUN: llc < %s -mtriple=x86_64-- | FileCheck %s
 
-; CHECK: imull
-
 ; Don't eliminate or coalesce away the explicit zero-extension!
 
 define i64 @foo(i64 %a) {
+; CHECK-LABEL: foo:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    imull $7823, %edi, %eax # imm = 0x1E8F
+; CHECK-NEXT:    incq %rax
+; CHECK-NEXT:    retq
   %b = mul i64 %a, 7823
   %c = and i64 %b, 4294967295
   %d = add i64 %c, 1
diff --git a/llvm/test/CodeGen/X86/subreg-to-reg-6.ll b/llvm/test/CodeGen/X86/subreg-to-reg-6.ll
index 7a6f78fac368..f0dc17b55661 100644
--- a/llvm/test/CodeGen/X86/subreg-to-reg-6.ll
+++ b/llvm/test/CodeGen/X86/subreg-to-reg-6.ll
@@ -1,6 +1,18 @@
-; RUN: llc < %s -mtriple=x86_64--
+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --version 4
+; RUN: llc < %s -mtriple=x86_64-- | FileCheck %s
 
 define i64 @foo() nounwind {
+; CHECK-LABEL: foo:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    cmpl $12, 0
+; CHECK-NEXT:    je .LBB0_1
+; CHECK-NEXT:  # %bb.2: # %bb65
+; CHECK-NEXT:    xorl %ecx, %ecx
+; CHECK-NEXT:    #APP
+; CHECK-NEXT:    #NO_APP
+; CHECK-NEXT:    xorl %eax, %eax
+; CHECK-NEXT:    retq
+; CHECK-NEXT:  .LBB0_1: # %bb56
 entry:
 	%t0 = load i32, ptr null, align 8
 	switch i32 %t0, label %bb65 [
@@ -22,6 +34,14 @@ bb65:
 }
 
 define i64 @bar(i64 %t0) nounwind {
+; CHECK-LABEL: bar:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    movq %rdi, %rax
+; CHECK-NEXT:    xorl %ecx, %ecx
+; CHECK-NEXT:    #APP
+; CHECK-NEXT:    #NO_APP
+; CHECK-NEXT:    negl %eax
+; CHECK-NEXT:    retq
 	call void asm "", "{cx}"(i64 0) nounwind
 	%t1 = sub i64 0, %t0
 	%t2 = and i64 %t1, 4294967295
diff --git a/llvm/test/CodeGen/X86/subvectorwise-store-of-vector-splat.ll b/llvm/test/CodeGen/X86/subvectorwise-store-of-vector-splat.ll
index e0f438eb7cc8..ae66c5420638 100644
--- a/llvm/test/CodeGen/X86/subvectorwise-store-of-vector-splat.ll
+++ b/llvm/test/CodeGen/X86/subvectorwise-store-of-vector-splat.ll
@@ -3060,12 +3060,7 @@ define void @vec384_v3i32(ptr %in.subvec.ptr, ptr %out.subvec.ptr, ptr %out.vec.
 ; SCALAR:       # %bb.0:
 ; SCALAR-NEXT:    movl 8(%rdi), %eax
 ; SCALAR-NEXT:    movq (%rdi), %rcx
-; SCALAR-NEXT:    movq %rcx, %rdi
-; SCALAR-NEXT:    shrq $32, %rdi
-; SCALAR-NEXT:    notl %edi
-; SCALAR-NEXT:    shlq $32, %rdi
-; SCALAR-NEXT:    notl %ecx
-; SCALAR-NEXT:    orq %rdi, %rcx
+; SCALAR-NEXT:    notq %rcx
 ; SCALAR-NEXT:    notl %eax
 ; SCALAR-NEXT:    movl %eax, 8(%rsi)
 ; SCALAR-NEXT:    movq %rcx, (%rsi)
@@ -3196,12 +3191,7 @@ define void @vec384_v3f32(ptr %in.subvec.ptr, ptr %out.subvec.ptr, ptr %out.vec.
 ; SCALAR:       # %bb.0:
 ; SCALAR-NEXT:    movl 8(%rdi), %eax
 ; SCALAR-NEXT:    movq (%rdi), %rcx
-; SCALAR-NEXT:    movq %rcx, %rdi
-; SCALAR-NEXT:    shrq $32, %rdi
-; SCALAR-NEXT:    notl %edi
-; SCALAR-NEXT:    shlq $32, %rdi
-; SCALAR-NEXT:    notl %ecx
-; SCALAR-NEXT:    orq %rdi, %rcx
+; SCALAR-NEXT:    notq %rcx
 ; SCALAR-NEXT:    notl %eax
 ; SCALAR-NEXT:    movl %eax, 8(%rsi)
 ; SCALAR-NEXT:    movq %rcx, (%rsi)
@@ -4216,25 +4206,10 @@ define void @vec384_v6i32(ptr %in.subvec.ptr, ptr %out.subvec.ptr, ptr %out.vec.
 ; SCALAR:       # %bb.0:
 ; SCALAR-NEXT:    movq (%rdi), %rax
 ; SCALAR-NEXT:    movq 8(%rdi), %rcx
-; SCALAR-NEXT:    movq %rax, %r8
-; SCALAR-NEXT:    shrq $32, %r8
-; SCALAR-NEXT:    movq %rcx, %r9
-; SCALAR-NEXT:    shrq $32, %r9
 ; SCALAR-NEXT:    movq 16(%rdi), %rdi
-; SCALAR-NEXT:    movq %rdi, %r10
-; SCALAR-NEXT:    shrq $32, %r10
-; SCALAR-NEXT:    notl %r10d
-; SCALAR-NEXT:    shlq $32, %r10
-; SCALAR-NEXT:    notl %edi
-; SCALAR-NEXT:    orq %r10, %rdi
-; SCALAR-NEXT:    notl %r9d
-; SCALAR-NEXT:    shlq $32, %r9
-; SCALAR-NEXT:    notl %ecx
-; SCALAR-NEXT:    orq %r9, %rcx
-; SCALAR-NEXT:    notl %r8d
-; SCALAR-NEXT:    shlq $32, %r8
-; SCALAR-NEXT:    notl %eax
-; SCALAR-NEXT:    orq %r8, %rax
+; SCALAR-NEXT:    notq %rdi
+; SCALAR-NEXT:    notq %rcx
+; SCALAR-NEXT:    notq %rax
 ; SCALAR-NEXT:    movq %rax, (%rsi)
 ; SCALAR-NEXT:    movq %rcx, 8(%rsi)
 ; SCALAR-NEXT:    movq %rdi, 16(%rsi)
@@ -4303,25 +4278,10 @@ define void @vec384_v6f32(ptr %in.subvec.ptr, ptr %out.subvec.ptr, ptr %out.vec.
 ; SCALAR:       # %bb.0:
 ; SCALAR-NEXT:    movq (%rdi), %rax
 ; SCALAR-NEXT:    movq 8(%rdi), %rcx
-; SCALAR-NEXT:    movq %rax, %r8
-; SCALAR-NEXT:    shrq $32, %r8
-; SCALAR-NEXT:    movq %rcx, %r9
-; SCALAR-NEXT:    shrq $32, %r9
 ; SCALAR-NEXT:    movq 16(%rdi), %rdi
-; SCALAR-NEXT:    movq %rdi, %r10
-; SCALAR-NEXT:    shrq $32, %r10
-; SCALAR-NEXT:    notl %r10d
-; SCALAR-NEXT:    shlq $32, %r10
-; SCALAR-NEXT:    notl %edi
-; SCALAR-NEXT:    orq %r10, %rdi
-; SCALAR-NEXT:    notl %r9d
-; SCALAR-NEXT:    shlq $32, %r9
-; SCALAR-NEXT:    notl %ecx
-; SCALAR-NEXT:    orq %r9, %rcx
-; SCALAR-NEXT:    notl %r8d
-; SCALAR-NEXT:    shlq $32, %r8
-; SCALAR-NEXT:    notl %eax
-; SCALAR-NEXT:    orq %r8, %rax
+; SCALAR-NEXT:    notq %rdi
+; SCALAR-NEXT:    notq %rcx
+; SCALAR-NEXT:    notq %rax
 ; SCALAR-NEXT:    movq %rax, (%rsi)
 ; SCALAR-NEXT:    movq %rcx, 8(%rsi)
 ; SCALAR-NEXT:    movq %rdi, 16(%rsi)
diff --git a/llvm/test/CodeGen/X86/vec_saddo.ll b/llvm/test/CodeGen/X86/vec_saddo.ll
index cee30f5fe5da..460c5fe11f82 100644
--- a/llvm/test/CodeGen/X86/vec_saddo.ll
+++ b/llvm/test/CodeGen/X86/vec_saddo.ll
@@ -1045,16 +1045,12 @@ define <4 x i32> @saddo_v4i1(<4 x i1> %a0, <4 x i1> %a1, ptr %p2) nounwind {
 ;
 ; AVX512-LABEL: saddo_v4i1:
 ; AVX512:       # %bb.0:
+; AVX512-NEXT:    vpxor %xmm1, %xmm0, %xmm2
+; AVX512-NEXT:    vpslld $31, %xmm2, %xmm2
+; AVX512-NEXT:    vptestmd %xmm2, %xmm2, %k0
+; AVX512-NEXT:    vpand %xmm1, %xmm0, %xmm0
 ; AVX512-NEXT:    vpslld $31, %xmm0, %xmm0
-; AVX512-NEXT:    vptestmd %xmm0, %xmm0, %k0
-; AVX512-NEXT:    vpslld $31, %xmm1, %xmm0
-; AVX512-NEXT:    vptestmd %xmm0, %xmm0, %k1
-; AVX512-NEXT:    kxorw %k1, %k0, %k2
-; AVX512-NEXT:    kandw %k1, %k0, %k1
-; AVX512-NEXT:    vpcmpeqd %xmm0, %xmm0, %xmm0
-; AVX512-NEXT:    vmovdqa32 %xmm0, %xmm0 {%k1} {z}
-; AVX512-NEXT:    kshiftlw $12, %k2, %k0
-; AVX512-NEXT:    kshiftrw $12, %k0, %k0
+; AVX512-NEXT:    vpsrad $31, %xmm0, %xmm0
 ; AVX512-NEXT:    kmovd %k0, %eax
 ; AVX512-NEXT:    movb %al, (%rdi)
 ; AVX512-NEXT:    retq
diff --git a/llvm/test/CodeGen/X86/vec_ssubo.ll b/llvm/test/CodeGen/X86/vec_ssubo.ll
index 64ed08104885..d06993da6365 100644
--- a/llvm/test/CodeGen/X86/vec_ssubo.ll
+++ b/llvm/test/CodeGen/X86/vec_ssubo.ll
@@ -1062,16 +1062,12 @@ define <4 x i32> @ssubo_v4i1(<4 x i1> %a0, <4 x i1> %a1, ptr %p2) nounwind {
 ;
 ; AVX512-LABEL: ssubo_v4i1:
 ; AVX512:       # %bb.0:
+; AVX512-NEXT:    vpxor %xmm1, %xmm0, %xmm2
+; AVX512-NEXT:    vpslld $31, %xmm2, %xmm2
+; AVX512-NEXT:    vptestmd %xmm2, %xmm2, %k0
+; AVX512-NEXT:    vpandn %xmm1, %xmm0, %xmm0
 ; AVX512-NEXT:    vpslld $31, %xmm0, %xmm0
-; AVX512-NEXT:    vptestmd %xmm0, %xmm0, %k0
-; AVX512-NEXT:    vpslld $31, %xmm1, %xmm1
-; AVX512-NEXT:    vptestmd %xmm1, %xmm1, %k1
-; AVX512-NEXT:    kxorw %k1, %k0, %k0
-; AVX512-NEXT:    vptestnmd %xmm0, %xmm0, %k1 {%k1}
-; AVX512-NEXT:    vpcmpeqd %xmm0, %xmm0, %xmm0
-; AVX512-NEXT:    vmovdqa32 %xmm0, %xmm0 {%k1} {z}
-; AVX512-NEXT:    kshiftlw $12, %k0, %k0
-; AVX512-NEXT:    kshiftrw $12, %k0, %k0
+; AVX512-NEXT:    vpsrad $31, %xmm0, %xmm0
 ; AVX512-NEXT:    kmovd %k0, %eax
 ; AVX512-NEXT:    movb %al, (%rdi)
 ; AVX512-NEXT:    retq
diff --git a/llvm/test/CodeGen/X86/vec_uaddo.ll b/llvm/test/CodeGen/X86/vec_uaddo.ll
index 950e943bd902..bac118095331 100644
--- a/llvm/test/CodeGen/X86/vec_uaddo.ll
+++ b/llvm/test/CodeGen/X86/vec_uaddo.ll
@@ -1098,16 +1098,12 @@ define <4 x i32> @uaddo_v4i1(<4 x i1> %a0, <4 x i1> %a1, ptr %p2) nounwind {
 ;
 ; AVX512-LABEL: uaddo_v4i1:
 ; AVX512:       # %bb.0:
+; AVX512-NEXT:    vpxor %xmm1, %xmm0, %xmm2
+; AVX512-NEXT:    vpslld $31, %xmm2, %xmm2
+; AVX512-NEXT:    vptestmd %xmm2, %xmm2, %k0
+; AVX512-NEXT:    vpand %xmm1, %xmm0, %xmm0
 ; AVX512-NEXT:    vpslld $31, %xmm0, %xmm0
-; AVX512-NEXT:    vptestmd %xmm0, %xmm0, %k0
-; AVX512-NEXT:    vpslld $31, %xmm1, %xmm0
-; AVX512-NEXT:    vptestmd %xmm0, %xmm0, %k1
-; AVX512-NEXT:    kxorw %k1, %k0, %k2
-; AVX512-NEXT:    kandw %k1, %k0, %k1
-; AVX512-NEXT:    vpcmpeqd %xmm0, %xmm0, %xmm0
-; AVX512-NEXT:    vmovdqa32 %xmm0, %xmm0 {%k1} {z}
-; AVX512-NEXT:    kshiftlw $12, %k2, %k0
-; AVX512-NEXT:    kshiftrw $12, %k0, %k0
+; AVX512-NEXT:    vpsrad $31, %xmm0, %xmm0
 ; AVX512-NEXT:    kmovd %k0, %eax
 ; AVX512-NEXT:    movb %al, (%rdi)
 ; AVX512-NEXT:    retq
diff --git a/llvm/test/CodeGen/X86/vec_usubo.ll b/llvm/test/CodeGen/X86/vec_usubo.ll
index 7de972770d8d..ab75ada72f25 100644
--- a/llvm/test/CodeGen/X86/vec_usubo.ll
+++ b/llvm/test/CodeGen/X86/vec_usubo.ll
@@ -1145,16 +1145,12 @@ define <4 x i32> @usubo_v4i1(<4 x i1> %a0, <4 x i1> %a1, ptr %p2) nounwind {
 ;
 ; AVX512-LABEL: usubo_v4i1:
 ; AVX512:       # %bb.0:
+; AVX512-NEXT:    vpxor %xmm1, %xmm0, %xmm2
+; AVX512-NEXT:    vpslld $31, %xmm2, %xmm2
+; AVX512-NEXT:    vptestmd %xmm2, %xmm2, %k0
+; AVX512-NEXT:    vpandn %xmm1, %xmm0, %xmm0
 ; AVX512-NEXT:    vpslld $31, %xmm0, %xmm0
-; AVX512-NEXT:    vptestmd %xmm0, %xmm0, %k0
-; AVX512-NEXT:    vpslld $31, %xmm1, %xmm1
-; AVX512-NEXT:    vptestmd %xmm1, %xmm1, %k1
-; AVX512-NEXT:    kxorw %k1, %k0, %k0
-; AVX512-NEXT:    vptestnmd %xmm0, %xmm0, %k1 {%k1}
-; AVX512-NEXT:    vpcmpeqd %xmm0, %xmm0, %xmm0
-; AVX512-NEXT:    vmovdqa32 %xmm0, %xmm0 {%k1} {z}
-; AVX512-NEXT:    kshiftlw $12, %k0, %k0
-; AVX512-NEXT:    kshiftrw $12, %k0, %k0
+; AVX512-NEXT:    vpsrad $31, %xmm0, %xmm0
 ; AVX512-NEXT:    kmovd %k0, %eax
 ; AVX512-NEXT:    movb %al, (%rdi)
 ; AVX512-NEXT:    retq
diff --git a/llvm/test/CodeGen/X86/vector-bo-select.ll b/llvm/test/CodeGen/X86/vector-bo-select.ll
index 78797b9acc2e..27aaad6353ed 100644
--- a/llvm/test/CodeGen/X86/vector-bo-select.ll
+++ b/llvm/test/CodeGen/X86/vector-bo-select.ll
@@ -3137,11 +3137,11 @@ define <8 x i64> @mul_v8i64_cast_cond(i8 noundef zeroext %pb, <8 x i64> noundef
 ; AVX512-LABEL: mul_v8i64_cast_cond:
 ; AVX512:       # %bb.0:
 ; AVX512-NEXT:    kmovw %edi, %k1
-; AVX512-NEXT:    vpsrlq $32, %zmm1, %zmm2
-; AVX512-NEXT:    vpmuludq %zmm2, %zmm0, %zmm2
-; AVX512-NEXT:    vpsrlq $32, %zmm0, %zmm3
-; AVX512-NEXT:    vpmuludq %zmm1, %zmm3, %zmm3
-; AVX512-NEXT:    vpaddq %zmm3, %zmm2, %zmm2
+; AVX512-NEXT:    vpsrlq $32, %zmm0, %zmm2
+; AVX512-NEXT:    vpmuludq %zmm1, %zmm2, %zmm2
+; AVX512-NEXT:    vpsrlq $32, %zmm1, %zmm3
+; AVX512-NEXT:    vpmuludq %zmm3, %zmm0, %zmm3
+; AVX512-NEXT:    vpaddq %zmm2, %zmm3, %zmm2
 ; AVX512-NEXT:    vpsllq $32, %zmm2, %zmm2
 ; AVX512-NEXT:    vpmuludq %zmm1, %zmm0, %zmm1
 ; AVX512-NEXT:    vpaddq %zmm2, %zmm1, %zmm0 {%k1}
diff --git a/llvm/test/CodeGen/X86/vector-fshr-128.ll b/llvm/test/CodeGen/X86/vector-fshr-128.ll
index b839452725a9..3aaa9268a8d8 100644
--- a/llvm/test/CodeGen/X86/vector-fshr-128.ll
+++ b/llvm/test/CodeGen/X86/vector-fshr-128.ll
@@ -58,12 +58,12 @@ define <2 x i64> @var_funnnel_v2i64(<2 x i64> %x, <2 x i64> %y, <2 x i64> %amt)
 ; SSE41-NEXT:    psrlq %xmm4, %xmm1
 ; SSE41-NEXT:    pblendw {{.*#+}} xmm5 = xmm5[0,1,2,3],xmm1[4,5,6,7]
 ; SSE41-NEXT:    pandn %xmm3, %xmm2
-; SSE41-NEXT:    pshufd {{.*#+}} xmm1 = xmm2[2,3,2,3]
 ; SSE41-NEXT:    paddq %xmm0, %xmm0
-; SSE41-NEXT:    movdqa %xmm0, %xmm3
-; SSE41-NEXT:    psllq %xmm1, %xmm3
+; SSE41-NEXT:    movdqa %xmm0, %xmm1
+; SSE41-NEXT:    psllq %xmm2, %xmm1
+; SSE41-NEXT:    pshufd {{.*#+}} xmm2 = xmm2[2,3,2,3]
 ; SSE41-NEXT:    psllq %xmm2, %xmm0
-; SSE41-NEXT:    pblendw {{.*#+}} xmm0 = xmm0[0,1,2,3],xmm3[4,5,6,7]
+; SSE41-NEXT:    pblendw {{.*#+}} xmm0 = xmm1[0,1,2,3],xmm0[4,5,6,7]
 ; SSE41-NEXT:    por %xmm5, %xmm0
 ; SSE41-NEXT:    retq
 ;
@@ -76,11 +76,11 @@ define <2 x i64> @var_funnnel_v2i64(<2 x i64> %x, <2 x i64> %y, <2 x i64> %amt)
 ; AVX1-NEXT:    vpsrlq %xmm4, %xmm1, %xmm1
 ; AVX1-NEXT:    vpblendw {{.*#+}} xmm1 = xmm5[0,1,2,3],xmm1[4,5,6,7]
 ; AVX1-NEXT:    vpandn %xmm3, %xmm2, %xmm2
-; AVX1-NEXT:    vpshufd {{.*#+}} xmm3 = xmm2[2,3,2,3]
 ; AVX1-NEXT:    vpaddq %xmm0, %xmm0, %xmm0
-; AVX1-NEXT:    vpsllq %xmm3, %xmm0, %xmm3
+; AVX1-NEXT:    vpsllq %xmm2, %xmm0, %xmm3
+; AVX1-NEXT:    vpshufd {{.*#+}} xmm2 = xmm2[2,3,2,3]
 ; AVX1-NEXT:    vpsllq %xmm2, %xmm0, %xmm0
-; AVX1-NEXT:    vpblendw {{.*#+}} xmm0 = xmm0[0,1,2,3],xmm3[4,5,6,7]
+; AVX1-NEXT:    vpblendw {{.*#+}} xmm0 = xmm3[0,1,2,3],xmm0[4,5,6,7]
 ; AVX1-NEXT:    vpor %xmm1, %xmm0, %xmm0
 ; AVX1-NEXT:    retq
 ;
@@ -158,13 +158,13 @@ define <2 x i64> @var_funnnel_v2i64(<2 x i64> %x, <2 x i64> %y, <2 x i64> %amt)
 ; XOPAVX1-LABEL: var_funnnel_v2i64:
 ; XOPAVX1:       # %bb.0:
 ; XOPAVX1-NEXT:    vpmovsxbq {{.*#+}} xmm3 = [63,63]
-; XOPAVX1-NEXT:    vpand %xmm3, %xmm2, %xmm4
-; XOPAVX1-NEXT:    vpxor %xmm5, %xmm5, %xmm5
-; XOPAVX1-NEXT:    vpsubq %xmm4, %xmm5, %xmm4
-; XOPAVX1-NEXT:    vpshlq %xmm4, %xmm1, %xmm1
-; XOPAVX1-NEXT:    vpandn %xmm3, %xmm2, %xmm2
+; XOPAVX1-NEXT:    vpandn %xmm3, %xmm2, %xmm4
 ; XOPAVX1-NEXT:    vpaddq %xmm0, %xmm0, %xmm0
-; XOPAVX1-NEXT:    vpshlq %xmm2, %xmm0, %xmm0
+; XOPAVX1-NEXT:    vpshlq %xmm4, %xmm0, %xmm0
+; XOPAVX1-NEXT:    vpand %xmm3, %xmm2, %xmm2
+; XOPAVX1-NEXT:    vpxor %xmm3, %xmm3, %xmm3
+; XOPAVX1-NEXT:    vpsubq %xmm2, %xmm3, %xmm2
+; XOPAVX1-NEXT:    vpshlq %xmm2, %xmm1, %xmm1
 ; XOPAVX1-NEXT:    vpor %xmm1, %xmm0, %xmm0
 ; XOPAVX1-NEXT:    retq
 ;
@@ -366,13 +366,13 @@ define <4 x i32> @var_funnnel_v4i32(<4 x i32> %x, <4 x i32> %y, <4 x i32> %amt)
 ; XOPAVX1-LABEL: var_funnnel_v4i32:
 ; XOPAVX1:       # %bb.0:
 ; XOPAVX1-NEXT:    vbroadcastss {{.*#+}} xmm3 = [31,31,31,31]
-; XOPAVX1-NEXT:    vpand %xmm3, %xmm2, %xmm4
-; XOPAVX1-NEXT:    vpxor %xmm5, %xmm5, %xmm5
-; XOPAVX1-NEXT:    vpsubd %xmm4, %xmm5, %xmm4
-; XOPAVX1-NEXT:    vpshld %xmm4, %xmm1, %xmm1
-; XOPAVX1-NEXT:    vpandn %xmm3, %xmm2, %xmm2
+; XOPAVX1-NEXT:    vpandn %xmm3, %xmm2, %xmm4
 ; XOPAVX1-NEXT:    vpaddd %xmm0, %xmm0, %xmm0
-; XOPAVX1-NEXT:    vpshld %xmm2, %xmm0, %xmm0
+; XOPAVX1-NEXT:    vpshld %xmm4, %xmm0, %xmm0
+; XOPAVX1-NEXT:    vpand %xmm3, %xmm2, %xmm2
+; XOPAVX1-NEXT:    vpxor %xmm3, %xmm3, %xmm3
+; XOPAVX1-NEXT:    vpsubd %xmm2, %xmm3, %xmm2
+; XOPAVX1-NEXT:    vpshld %xmm2, %xmm1, %xmm1
 ; XOPAVX1-NEXT:    vpor %xmm1, %xmm0, %xmm0
 ; XOPAVX1-NEXT:    retq
 ;
@@ -646,26 +646,26 @@ define <8 x i16> @var_funnnel_v8i16(<8 x i16> %x, <8 x i16> %y, <8 x i16> %amt)
 ; XOPAVX1-LABEL: var_funnnel_v8i16:
 ; XOPAVX1:       # %bb.0:
 ; XOPAVX1-NEXT:    vbroadcastss {{.*#+}} xmm3 = [15,15,15,15,15,15,15,15]
-; XOPAVX1-NEXT:    vpand %xmm3, %xmm2, %xmm4
-; XOPAVX1-NEXT:    vpxor %xmm5, %xmm5, %xmm5
-; XOPAVX1-NEXT:    vpsubw %xmm4, %xmm5, %xmm4
-; XOPAVX1-NEXT:    vpshlw %xmm4, %xmm1, %xmm1
-; XOPAVX1-NEXT:    vpandn %xmm3, %xmm2, %xmm2
+; XOPAVX1-NEXT:    vpandn %xmm3, %xmm2, %xmm4
 ; XOPAVX1-NEXT:    vpaddw %xmm0, %xmm0, %xmm0
-; XOPAVX1-NEXT:    vpshlw %xmm2, %xmm0, %xmm0
+; XOPAVX1-NEXT:    vpshlw %xmm4, %xmm0, %xmm0
+; XOPAVX1-NEXT:    vpand %xmm3, %xmm2, %xmm2
+; XOPAVX1-NEXT:    vpxor %xmm3, %xmm3, %xmm3
+; XOPAVX1-NEXT:    vpsubw %xmm2, %xmm3, %xmm2
+; XOPAVX1-NEXT:    vpshlw %xmm2, %xmm1, %xmm1
 ; XOPAVX1-NEXT:    vpor %xmm1, %xmm0, %xmm0
 ; XOPAVX1-NEXT:    retq
 ;
 ; XOPAVX2-LABEL: var_funnnel_v8i16:
 ; XOPAVX2:       # %bb.0:
 ; XOPAVX2-NEXT:    vpbroadcastw {{.*#+}} xmm3 = [15,15,15,15,15,15,15,15]
-; XOPAVX2-NEXT:    vpand %xmm3, %xmm2, %xmm4
-; XOPAVX2-NEXT:    vpxor %xmm5, %xmm5, %xmm5
-; XOPAVX2-NEXT:    vpsubw %xmm4, %xmm5, %xmm4
-; XOPAVX2-NEXT:    vpshlw %xmm4, %xmm1, %xmm1
-; XOPAVX2-NEXT:    vpandn %xmm3, %xmm2, %xmm2
+; XOPAVX2-NEXT:    vpandn %xmm3, %xmm2, %xmm4
 ; XOPAVX2-NEXT:    vpaddw %xmm0, %xmm0, %xmm0
-; XOPAVX2-NEXT:    vpshlw %xmm2, %xmm0, %xmm0
+; XOPAVX2-NEXT:    vpshlw %xmm4, %xmm0, %xmm0
+; XOPAVX2-NEXT:    vpand %xmm3, %xmm2, %xmm2
+; XOPAVX2-NEXT:    vpxor %xmm3, %xmm3, %xmm3
+; XOPAVX2-NEXT:    vpsubw %xmm2, %xmm3, %xmm2
+; XOPAVX2-NEXT:    vpshlw %xmm2, %xmm1, %xmm1
 ; XOPAVX2-NEXT:    vpor %xmm1, %xmm0, %xmm0
 ; XOPAVX2-NEXT:    retq
 ;
@@ -995,26 +995,26 @@ define <16 x i8> @var_funnnel_v16i8(<16 x i8> %x, <16 x i8> %y, <16 x i8> %amt)
 ; XOPAVX1-LABEL: var_funnnel_v16i8:
 ; XOPAVX1:       # %bb.0:
 ; XOPAVX1-NEXT:    vbroadcastss {{.*#+}} xmm3 = [7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7]
-; XOPAVX1-NEXT:    vpand %xmm3, %xmm2, %xmm4
-; XOPAVX1-NEXT:    vpxor %xmm5, %xmm5, %xmm5
-; XOPAVX1-NEXT:    vpsubb %xmm4, %xmm5, %xmm4
-; XOPAVX1-NEXT:    vpshlb %xmm4, %xmm1, %xmm1
-; XOPAVX1-NEXT:    vpandn %xmm3, %xmm2, %xmm2
+; XOPAVX1-NEXT:    vpandn %xmm3, %xmm2, %xmm4
 ; XOPAVX1-NEXT:    vpaddb %xmm0, %xmm0, %xmm0
-; XOPAVX1-NEXT:    vpshlb %xmm2, %xmm0, %xmm0
+; XOPAVX1-NEXT:    vpshlb %xmm4, %xmm0, %xmm0
+; XOPAVX1-NEXT:    vpand %xmm3, %xmm2, %xmm2
+; XOPAVX1-NEXT:    vpxor %xmm3, %xmm3, %xmm3
+; XOPAVX1-NEXT:    vpsubb %xmm2, %xmm3, %xmm2
+; XOPAVX1-NEXT:    vpshlb %xmm2, %xmm1, %xmm1
 ; XOPAVX1-NEXT:    vpor %xmm1, %xmm0, %xmm0
 ; XOPAVX1-NEXT:    retq
 ;
 ; XOPAVX2-LABEL: var_funnnel_v16i8:
 ; XOPAVX2:       # %bb.0:
 ; XOPAVX2-NEXT:    vpbroadcastb {{.*#+}} xmm3 = [7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7]
-; XOPAVX2-NEXT:    vpand %xmm3, %xmm2, %xmm4
-; XOPAVX2-NEXT:    vpxor %xmm5, %xmm5, %xmm5
-; XOPAVX2-NEXT:    vpsubb %xmm4, %xmm5, %xmm4
-; XOPAVX2-NEXT:    vpshlb %xmm4, %xmm1, %xmm1
-; XOPAVX2-NEXT:    vpandn %xmm3, %xmm2, %xmm2
+; XOPAVX2-NEXT:    vpandn %xmm3, %xmm2, %xmm4
 ; XOPAVX2-NEXT:    vpaddb %xmm0, %xmm0, %xmm0
-; XOPAVX2-NEXT:    vpshlb %xmm2, %xmm0, %xmm0
+; XOPAVX2-NEXT:    vpshlb %xmm4, %xmm0, %xmm0
+; XOPAVX2-NEXT:    vpand %xmm3, %xmm2, %xmm2
+; XOPAVX2-NEXT:    vpxor %xmm3, %xmm3, %xmm3
+; XOPAVX2-NEXT:    vpsubb %xmm2, %xmm3, %xmm2
+; XOPAVX2-NEXT:    vpshlb %xmm2, %xmm1, %xmm1
 ; XOPAVX2-NEXT:    vpor %xmm1, %xmm0, %xmm0
 ; XOPAVX2-NEXT:    retq
 ;
diff --git a/llvm/test/CodeGen/X86/vector-fshr-256.ll b/llvm/test/CodeGen/X86/vector-fshr-256.ll
index 7b6b0ea83c7e..fc65f759f5fb 100644
--- a/llvm/test/CodeGen/X86/vector-fshr-256.ll
+++ b/llvm/test/CodeGen/X86/vector-fshr-256.ll
@@ -486,22 +486,22 @@ define <16 x i16> @var_funnnel_v16i16(<16 x i16> %x, <16 x i16> %y, <16 x i16> %
 ; XOPAVX2-LABEL: var_funnnel_v16i16:
 ; XOPAVX2:       # %bb.0:
 ; XOPAVX2-NEXT:    vpbroadcastw {{.*#+}} ymm3 = [15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15]
-; XOPAVX2-NEXT:    vpand %ymm3, %ymm2, %ymm4
+; XOPAVX2-NEXT:    vpandn %ymm3, %ymm2, %ymm4
 ; XOPAVX2-NEXT:    vextracti128 $1, %ymm4, %xmm5
-; XOPAVX2-NEXT:    vpxor %xmm6, %xmm6, %xmm6
-; XOPAVX2-NEXT:    vpsubw %xmm5, %xmm6, %xmm5
-; XOPAVX2-NEXT:    vextracti128 $1, %ymm1, %xmm7
-; XOPAVX2-NEXT:    vpshlw %xmm5, %xmm7, %xmm5
-; XOPAVX2-NEXT:    vpsubw %xmm4, %xmm6, %xmm4
-; XOPAVX2-NEXT:    vpshlw %xmm4, %xmm1, %xmm1
-; XOPAVX2-NEXT:    vinserti128 $1, %xmm5, %ymm1, %ymm1
-; XOPAVX2-NEXT:    vpandn %ymm3, %ymm2, %ymm2
-; XOPAVX2-NEXT:    vextracti128 $1, %ymm2, %xmm3
 ; XOPAVX2-NEXT:    vpaddw %ymm0, %ymm0, %ymm0
-; XOPAVX2-NEXT:    vextracti128 $1, %ymm0, %xmm4
-; XOPAVX2-NEXT:    vpshlw %xmm3, %xmm4, %xmm3
-; XOPAVX2-NEXT:    vpshlw %xmm2, %xmm0, %xmm0
-; XOPAVX2-NEXT:    vinserti128 $1, %xmm3, %ymm0, %ymm0
+; XOPAVX2-NEXT:    vextracti128 $1, %ymm0, %xmm6
+; XOPAVX2-NEXT:    vpshlw %xmm5, %xmm6, %xmm5
+; XOPAVX2-NEXT:    vpshlw %xmm4, %xmm0, %xmm0
+; XOPAVX2-NEXT:    vinserti128 $1, %xmm5, %ymm0, %ymm0
+; XOPAVX2-NEXT:    vpand %ymm3, %ymm2, %ymm2
+; XOPAVX2-NEXT:    vextracti128 $1, %ymm2, %xmm3
+; XOPAVX2-NEXT:    vpxor %xmm4, %xmm4, %xmm4
+; XOPAVX2-NEXT:    vpsubw %xmm3, %xmm4, %xmm3
+; XOPAVX2-NEXT:    vextracti128 $1, %ymm1, %xmm5
+; XOPAVX2-NEXT:    vpshlw %xmm3, %xmm5, %xmm3
+; XOPAVX2-NEXT:    vpsubw %xmm2, %xmm4, %xmm2
+; XOPAVX2-NEXT:    vpshlw %xmm2, %xmm1, %xmm1
+; XOPAVX2-NEXT:    vinserti128 $1, %xmm3, %ymm1, %ymm1
 ; XOPAVX2-NEXT:    vpor %ymm1, %ymm0, %ymm0
 ; XOPAVX2-NEXT:    retq
   %res = call <16 x i16> @llvm.fshr.v16i16(<16 x i16> %x, <16 x i16> %y, <16 x i16> %amt)
diff --git a/llvm/test/CodeGen/X86/vector-fshr-sub128.ll b/llvm/test/CodeGen/X86/vector-fshr-sub128.ll
index 0426c48aecfc..a6067a960fc0 100644
--- a/llvm/test/CodeGen/X86/vector-fshr-sub128.ll
+++ b/llvm/test/CodeGen/X86/vector-fshr-sub128.ll
@@ -185,13 +185,13 @@ define <2 x i32> @var_funnnel_v2i32(<2 x i32> %x, <2 x i32> %y, <2 x i32> %amt)
 ; XOPAVX1-LABEL: var_funnnel_v2i32:
 ; XOPAVX1:       # %bb.0:
 ; XOPAVX1-NEXT:    vbroadcastss {{.*#+}} xmm3 = [31,31,31,31]
-; XOPAVX1-NEXT:    vpand %xmm3, %xmm2, %xmm4
-; XOPAVX1-NEXT:    vpxor %xmm5, %xmm5, %xmm5
-; XOPAVX1-NEXT:    vpsubd %xmm4, %xmm5, %xmm4
-; XOPAVX1-NEXT:    vpshld %xmm4, %xmm1, %xmm1
-; XOPAVX1-NEXT:    vpandn %xmm3, %xmm2, %xmm2
+; XOPAVX1-NEXT:    vpandn %xmm3, %xmm2, %xmm4
 ; XOPAVX1-NEXT:    vpaddd %xmm0, %xmm0, %xmm0
-; XOPAVX1-NEXT:    vpshld %xmm2, %xmm0, %xmm0
+; XOPAVX1-NEXT:    vpshld %xmm4, %xmm0, %xmm0
+; XOPAVX1-NEXT:    vpand %xmm3, %xmm2, %xmm2
+; XOPAVX1-NEXT:    vpxor %xmm3, %xmm3, %xmm3
+; XOPAVX1-NEXT:    vpsubd %xmm2, %xmm3, %xmm2
+; XOPAVX1-NEXT:    vpshld %xmm2, %xmm1, %xmm1
 ; XOPAVX1-NEXT:    vpor %xmm1, %xmm0, %xmm0
 ; XOPAVX1-NEXT:    retq
 ;
diff --git a/llvm/test/CodeGen/X86/vector-shift-shl-128.ll b/llvm/test/CodeGen/X86/vector-shift-shl-128.ll
index c54da38ef10c..75baba5f35f7 100644
--- a/llvm/test/CodeGen/X86/vector-shift-shl-128.ll
+++ b/llvm/test/CodeGen/X86/vector-shift-shl-128.ll
@@ -927,9 +927,9 @@ define <2 x i64> @constant_shift_v2i64(<2 x i64> %a) nounwind {
 ; SSE2-LABEL: constant_shift_v2i64:
 ; SSE2:       # %bb.0:
 ; SSE2-NEXT:    movdqa %xmm0, %xmm1
-; SSE2-NEXT:    psllq $7, %xmm1
-; SSE2-NEXT:    paddq %xmm0, %xmm0
-; SSE2-NEXT:    shufps {{.*#+}} xmm0 = xmm0[0,1],xmm1[2,3]
+; SSE2-NEXT:    paddq %xmm0, %xmm1
+; SSE2-NEXT:    psllq $7, %xmm0
+; SSE2-NEXT:    movsd {{.*#+}} xmm0 = xmm1[0],xmm0[1]
 ; SSE2-NEXT:    retq
 ;
 ; SSE41-LABEL: constant_shift_v2i64:
@@ -975,9 +975,9 @@ define <2 x i64> @constant_shift_v2i64(<2 x i64> %a) nounwind {
 ; X86-SSE-LABEL: constant_shift_v2i64:
 ; X86-SSE:       # %bb.0:
 ; X86-SSE-NEXT:    movdqa %xmm0, %xmm1
-; X86-SSE-NEXT:    psllq $7, %xmm1
-; X86-SSE-NEXT:    paddq %xmm0, %xmm0
-; X86-SSE-NEXT:    shufps {{.*#+}} xmm0 = xmm0[0,1],xmm1[2,3]
+; X86-SSE-NEXT:    paddq %xmm0, %xmm1
+; X86-SSE-NEXT:    psllq $7, %xmm0
+; X86-SSE-NEXT:    movsd {{.*#+}} xmm0 = xmm1[0],xmm0[1]
 ; X86-SSE-NEXT:    retl
   %shift = shl <2 x i64> %a, <i64 1, i64 7>
   ret <2 x i64> %shift
diff --git a/llvm/test/CodeGen/Xtensa/call.ll b/llvm/test/CodeGen/Xtensa/call.ll
new file mode 100644
index 000000000000..24c7c4f558e1
--- /dev/null
+++ b/llvm/test/CodeGen/Xtensa/call.ll
@@ -0,0 +1,49 @@
+; RUN: llc --mtriple=xtensa < %s | FileCheck %s
+
+declare i32 @external_function(i32)
+
+define i32 @test_call_external(i32 %a) nounwind {
+; CHECK-LABEL: test_call_external:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:  s32i  a0, a1, 0
+; CHECK-NEXT:  l32r  a8, .LCPI0_0
+; CHECK-NEXT:  callx0  a8
+; CHECK-NEXT:  l32i  a0, a1, 0
+; CHECK-NEXT:  ret
+  %1 = call i32 @external_function(i32 %a)
+  ret i32 %1
+}
+
+define i32 @defined_function(i32 %a) nounwind {
+; CHECK-LABEL: defined_function:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:  addi  a2, a2, 1
+; CHECK-NEXT:  ret
+  %1 = add i32 %a, 1
+  ret i32 %1
+}
+
+define i32 @test_call_defined(i32 %a) nounwind {
+; CHECK-LABEL: test_call_defined:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:  s32i  a0, a1, 0
+; CHECK-NEXT:  l32r  a8, .LCPI2_0
+; CHECK-NEXT:  callx0  a8
+; CHECK-NEXT:  l32i  a0, a1, 0
+; CHECK-NEXT:  ret
+  %1 = call i32 @defined_function(i32 %a) nounwind
+  ret i32 %1
+}
+
+define i32 @test_call_indirect(ptr %a, i32 %b) nounwind {
+; CHECK-LABEL: test_call_indirect:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:  s32i  a0, a1, 0
+; CHECK-NEXT:  or  a8, a2, a2
+; CHECK-NEXT:  or  a2, a3, a3
+; CHECK-NEXT:  callx0  a8
+; CHECK-NEXT:  l32i  a0, a1, 0
+; CHECK-NEXT:  ret
+  %1 = call i32 %a(i32 %b)
+  ret i32 %1
+}
diff --git a/llvm/test/CodeGen/Xtensa/calling-conv.ll b/llvm/test/CodeGen/Xtensa/calling-conv.ll
new file mode 100644
index 000000000000..41ae4220145c
--- /dev/null
+++ b/llvm/test/CodeGen/Xtensa/calling-conv.ll
@@ -0,0 +1,78 @@
+; RUN: llc -mtriple=xtensa -O1 -verify-machineinstrs < %s \
+; RUN:   | FileCheck %s -check-prefix=XTENSA
+
+; Check placement of first 6 arguments in registers and 7th argument on stack
+define dso_local i32 @test1(i32 noundef %0, i32 noundef %1, i32 noundef %2, i32 noundef %3, i32 noundef %4, i32 noundef %5, ptr nocapture noundef readonly byval(i32) align 4 %6) {
+; XTENSA-LABEL: @test1
+; XTENSA:  add	 a8, a7, a2
+; XTENSA:  l32i a9, a1, 0
+; XTENSA:  add  a2, a8, a9
+; XTENSA:  ret
+  %8 = load i32, ptr %6, align 4
+  %9 = add nsw i32 %5, %0
+  %10 = add nsw i32 %9, %8
+  ret i32 %10
+}
+
+; Check placement of second i64 argument in registers
+define dso_local i32 @test2(i32 noundef %0, i64 noundef %1, i32 noundef %2) {
+; XTENSA-LABEL: @test2
+; XTENSA:  add	 a8, a6, a2
+; XTENSA:  add	 a2, a8, a4
+; XTENSA:  ret
+  %4 = trunc i64 %1 to i32
+  %5 = add nsw i32 %2, %0
+  %6 = add nsw i32 %5, %4
+  ret i32 %6
+}
+
+; Check placement of first argument typeof i8 in register
+define dso_local i32 @test3(i8 noundef signext %0, i64 noundef %1, i32 noundef %2) {
+; XTENSA-LABEL: @test3
+; XTENSA:  add  a8, a2, a6
+; XTENSA:  add  a2, a8, a4
+; XTENSA:  ret
+  %4 = trunc i64 %1 to i32
+  %5 = sext i8 %0 to i32
+  %6 = add nsw i32 %5, %2
+  %7 = add nsw i32 %6, %4
+  ret i32 %7
+}
+
+; Check placement of 4th argument typeof i64 on stack
+define dso_local i32 @test4(i8 noundef signext %0, i64 noundef %1, i32 noundef %2, ptr nocapture noundef readonly byval(i64) align 8 %3) {
+; XTENSA-LABEL: @test4
+; XTENSA: add  a8, a2, a6
+; XTENSA: add  a8, a8, a4
+; XTENSA: l32i a9, a1, 0
+; XTENSA: add  a2, a8, a9
+; XTENSA: ret
+  %5 = load i64, ptr %3, align 8
+  %6 = trunc i64 %1 to i32
+  %7 = trunc i64 %5 to i32
+  %8 = sext i8 %0 to i32
+  %9 = add nsw i32 %8, %2
+  %10 = add nsw i32 %9, %6
+  %11 = add nsw i32 %10, %7
+  ret i32 %11
+}
+
+; Check placement of 128 bit structure on registers
+define dso_local i32 @test5([4 x i32] %0, i32 noundef %1) {
+; XTENSA-LABEL: @test5
+; XTENSA: add  a2, a2, a6
+; XTENSA: ret
+  %3 = extractvalue [4 x i32] %0, 0
+  %4 = add nsw i32 %3, %1
+  ret i32 %4
+}
+
+; Check placement of 128 bit structure on stack
+define dso_local i32 @test6(i32 noundef %0, [4 x i32] %1) {
+; XTENSA-LABEL: @test6
+; XTENSA: add a2, a3, a2
+; XTENSA: ret
+  %3 = extractvalue [4 x i32] %1, 0
+  %4 = add nsw i32 %3, %0
+  ret i32 %4
+}
diff --git a/llvm/test/CodeGen/Xtensa/constantpool.ll b/llvm/test/CodeGen/Xtensa/constantpool.ll
new file mode 100644
index 000000000000..9b380d2c37b9
--- /dev/null
+++ b/llvm/test/CodeGen/Xtensa/constantpool.ll
@@ -0,0 +1,28 @@
+; RUN: llc -mtriple=xtensa -verify-machineinstrs < %s \
+; RUN:   | FileCheck %s
+
+; Test placement of the i32,i64, float and double constants in constantpool
+
+define dso_local i32 @const_i32() #0 {
+; CHECK: .literal_position
+; CHECK-NEXT: .literal .LCPI0_0, 74565
+; CHECK-LABEL: const_i32:
+; CHECK: l32r a2, .LCPI0_0
+  %1 = alloca i32, align 4
+  store i32 74565, ptr %1, align 4
+  %2 = load i32, ptr %1, align 4
+  ret i32 %2
+}
+
+define dso_local i64 @const_int64() #0 {
+; CHECK: .literal_position
+; CHECK-NEXT: .literal .LCPI1_0, 305419896
+; CHECK-NEXT: .literal .LCPI1_1, -1859959449
+; CHECK-LABEL: const_int64:
+; CHECK: l32r a3, .LCPI1_0
+; CHECK: l32r a2, .LCPI1_1
+  %1 = alloca i64, align 8
+  store i64 1311768467302729063, ptr %1, align 8
+  %2 = load i64, ptr %1, align 8
+  ret i64 %2
+}
diff --git a/llvm/test/CodeGen/Xtensa/stack-access.ll b/llvm/test/CodeGen/Xtensa/stack-access.ll
new file mode 100644
index 000000000000..1590d24f228f
--- /dev/null
+++ b/llvm/test/CodeGen/Xtensa/stack-access.ll
@@ -0,0 +1,35 @@
+; RUN: llc -mtriple=xtensa -O0 -verify-machineinstrs < %s \
+; RUN:   | FileCheck %s -check-prefix=XTENSA
+
+define i8 @loadi8(i8 %a) {
+; XTENSA-LABEL: loadi8:
+; XTENSA: s8i	a2, a1, 3
+; XTENSA: l8ui	a2, a1, 3
+; XTENSA: ret
+  %b = alloca i8, align 1
+  store i8 %a, ptr %b, align 1
+  %1 = load i8, ptr %b, align 1
+  ret i8 %1
+}
+
+define i16 @loadi16(i16 %a) {
+; XTENSA-LABEL: loadi16:
+; XTENSA: s16i	a2, a1, 2
+; XTENSA: l16ui	a2, a1, 2
+; XTENSA: ret
+	%b = alloca i16, align 2
+  store i16 %a, ptr %b, align 2
+  %1 = load i16, ptr %b, align 2
+	ret i16 %1
+}
+
+define i32 @loadi32(i32 %a) {
+; XTENSA-LABEL: loadi32:
+; XTENSA: s32i	a2, a1, 0
+; XTENSA: l32i	a2, a1, 0
+; XTENSA: ret
+	%b = alloca i32, align 4
+  store i32 %a, ptr %b, align 4
+  %1 = load i32, ptr %b, align 4
+	ret i32 %1
+}
diff --git a/llvm/test/DebugInfo/LoongArch/dwarf-loongarch-relocs.ll b/llvm/test/DebugInfo/LoongArch/dwarf-loongarch-relocs.ll
index d6a1d8d6e136..d28836d56037 100644
--- a/llvm/test/DebugInfo/LoongArch/dwarf-loongarch-relocs.ll
+++ b/llvm/test/DebugInfo/LoongArch/dwarf-loongarch-relocs.ll
@@ -18,21 +18,21 @@
 ; RELOCS-BOTH:         Section ({{.*}}) .rela.debug_frame {
 ; RELOCS-NORL-NEXT:      0x1C R_LARCH_32 .debug_frame 0x0
 ; RELOCS-NORL-NEXT:      0x20 R_LARCH_64 .text 0x0
-; RELOCS-ENRL-NEXT:      0x1C R_LARCH_32 <null> 0x0
-; RELOCS-ENRL-NEXT:      0x20 R_LARCH_64 <null> 0x0
-; RELOCS-ENRL-NEXT:      0x28 R_LARCH_ADD64 <null> 0x0
-; RELOCS-ENRL-NEXT:      0x28 R_LARCH_SUB64 <null> 0x0
-; RELOCS-ENRL-NEXT:      0x3F R_LARCH_ADD6 <null> 0x0
-; RELOCS-ENRL-NEXT:      0x3F R_LARCH_SUB6 <null> 0x0
+; RELOCS-ENRL-NEXT:      0x1C R_LARCH_32 .L0  0x0
+; RELOCS-ENRL-NEXT:      0x20 R_LARCH_64 .L0  0x0
+; RELOCS-ENRL-NEXT:      0x28 R_LARCH_ADD64 .L0  0x0
+; RELOCS-ENRL-NEXT:      0x28 R_LARCH_SUB64 .L0  0x0
+; RELOCS-ENRL-NEXT:      0x3F R_LARCH_ADD6 .L0  0x0
+; RELOCS-ENRL-NEXT:      0x3F R_LARCH_SUB6 .L0  0x0
 ; RELOCS-BOTH-NEXT:    }
 ; RELOCS-BOTH:         Section ({{.*}}) .rela.debug_line {
 ; RELOCS-BOTH-NEXT:      0x22 R_LARCH_32 .debug_line_str 0x0
 ; RELOCS-BOTH-NEXT:      0x31 R_LARCH_32 .debug_line_str 0x2
 ; RELOCS-BOTH-NEXT:      0x46 R_LARCH_32 .debug_line_str 0x1B
 ; RELOCS-NORL-NEXT:      0x4F R_LARCH_64 .text 0x0
-; RELOCS-ENRL-NEXT:      0x4F R_LARCH_64 <null> 0x0
-; RELOCS-ENRL-NEXT:      0x5F R_LARCH_ADD16 <null> 0x0
-; RELOCS-ENRL-NEXT:      0x5F R_LARCH_SUB16 <null> 0x0
+; RELOCS-ENRL-NEXT:      0x4F R_LARCH_64 .L0  0x0
+; RELOCS-ENRL-NEXT:      0x5F R_LARCH_ADD16 .L0  0x0
+; RELOCS-ENRL-NEXT:      0x5F R_LARCH_SUB16 .L0  0x0
 ; RELOCS-BOTH-NEXT:    }
 ; RELOCS-BOTH-NEXT:  ]
 
diff --git a/llvm/test/DebugInfo/RISCV/dwarf-riscv-relocs.ll b/llvm/test/DebugInfo/RISCV/dwarf-riscv-relocs.ll
index e5de1713f4e0..99594b5e01e9 100644
--- a/llvm/test/DebugInfo/RISCV/dwarf-riscv-relocs.ll
+++ b/llvm/test/DebugInfo/RISCV/dwarf-riscv-relocs.ll
@@ -6,14 +6,14 @@
 
 ; Check that we actually have relocations, otherwise this is kind of pointless.
 ; READOBJ-RELOCS:  Section ({{.*}}) .rela.debug_info {
-; READOBJ-RELOCS:    0x1B R_RISCV_ADD32 <null> 0x0
-; READOBJ-RELOCS-NEXT:    0x1B R_RISCV_SUB32 <null> 0x0
+; READOBJ-RELOCS:    0x1B R_RISCV_ADD32 .L0  0x0
+; READOBJ-RELOCS-NEXT:    0x1B R_RISCV_SUB32 .L0  0x0
 ; READOBJ-RELOCS:  Section ({{.*}}) .rela.debug_frame {
-; READOBJ-RELOCS:    0x20 R_RISCV_ADD32 <null> 0x0
-; READOBJ-RELOCS-NEXT:    0x20 R_RISCV_SUB32 <null> 0x0
+; READOBJ-RELOCS:    0x20 R_RISCV_ADD32 .L0  0x0
+; READOBJ-RELOCS-NEXT:    0x20 R_RISCV_SUB32 .L0  0x0
 ; READOBJ-RELOCS:  Section ({{.*}}) .rela.debug_line {
-; READOBJ-RELOCS:    0x5A R_RISCV_ADD16 <null> 0x0
-; READOBJ-RELOCS-NEXT:    0x5A R_RISCV_SUB16 <null> 0x0
+; READOBJ-RELOCS:    0x5A R_RISCV_ADD16 .L0  0x0
+; READOBJ-RELOCS-NEXT:    0x5A R_RISCV_SUB16 .L0  0x0
 
 ; Check that we can print the source, even with relocations.
 ; OBJDUMP-SOURCE: Disassembly of section .text:
diff --git a/llvm/test/DebugInfo/RISCV/relax-debug-frame.ll b/llvm/test/DebugInfo/RISCV/relax-debug-frame.ll
index f655a7c0a7ef..ffef0ec23406 100644
--- a/llvm/test/DebugInfo/RISCV/relax-debug-frame.ll
+++ b/llvm/test/DebugInfo/RISCV/relax-debug-frame.ll
@@ -4,11 +4,11 @@
 ; RUN:     | FileCheck -check-prefix=RELAX-DWARFDUMP %s
 ;
 ; RELAX:      Section ({{.*}}) .rela.eh_frame {
-; RELAX-NEXT:   0x1C R_RISCV_32_PCREL <null> 0x0
-; RELAX-NEXT:   0x30 R_RISCV_32_PCREL <null> 0x0
-; RELAX-NEXT:   0x44 R_RISCV_32_PCREL <null> 0x0
-; RELAX-NEXT:   0x48 R_RISCV_ADD32 <null> 0x0
-; RELAX-NEXT:   0x48 R_RISCV_SUB32 <null> 0x0
+; RELAX-NEXT:   0x1C R_RISCV_32_PCREL .L0  0x0
+; RELAX-NEXT:   0x30 R_RISCV_32_PCREL .L0  0x0
+; RELAX-NEXT:   0x44 R_RISCV_32_PCREL .L0  0x0
+; RELAX-NEXT:   0x48 R_RISCV_ADD32 .L0  0x0
+; RELAX-NEXT:   0x48 R_RISCV_SUB32 .L0  0x0
 ; RELAX-NEXT:  }
 
 ; RELAX-DWARFDUMP-NOT: error: failed to compute relocation
diff --git a/llvm/test/DebugInfo/Symbolize/ELF/riscv-empty-name-symbol.s b/llvm/test/DebugInfo/Symbolize/ELF/riscv-temporary-symbol.s
index 1e0fa8a30618..0b54f104ab95 100644
--- a/llvm/test/DebugInfo/Symbolize/ELF/riscv-empty-name-symbol.s
+++ b/llvm/test/DebugInfo/Symbolize/ELF/riscv-temporary-symbol.s
@@ -1,10 +1,11 @@
 # REQUIRES: riscv-registered-target
-## Ignore empty name symbols.
+## Ignore .L0 symbols that are generated by LLVM integrated assembler and GNU
+## assembler for .debug_line/.eh_frame related assembler directives.
 
 # RUN: llvm-mc -filetype=obj -triple=riscv64 %s -o %t
 # RUN: llvm-readelf -s %t | FileCheck %s --check-prefix=SYM
 
-# SYM: 0000000000000004  0 NOTYPE LOCAL  DEFAULT [[#]] {{$}}
+# SYM: 0000000000000004  0 NOTYPE LOCAL  DEFAULT [[#]] .L0 {{$}}
 # SYM: 0000000000000000  0 NOTYPE GLOBAL DEFAULT [[#]] foo
 
 ## Make sure we test at an address larger than or equal to an empty name symbol.
diff --git a/llvm/test/DebugInfo/X86/call-origin-linkage-names.ll b/llvm/test/DebugInfo/X86/call-origin-linkage-names.ll
new file mode 100644
index 000000000000..c8ca3075c0ff
--- /dev/null
+++ b/llvm/test/DebugInfo/X86/call-origin-linkage-names.ll
@@ -0,0 +1,96 @@
+; RUN: llc %s --filetype=obj -o - -dwarf-linkage-names=Abstract -add-linkage-names-to-declaration-call-origins=false \
+; RUN: | llvm-dwarfdump - | FileCheck %s --check-prefixes=COMMON,DISABLE --implicit-check-not=DW_AT_linkage_name
+; RUN: llc %s --filetype=obj -o - -dwarf-linkage-names=Abstract -add-linkage-names-to-declaration-call-origins=true \
+; RUN: | llvm-dwarfdump - | FileCheck %s --check-prefixes=COMMON,ENABLE --implicit-check-not=DW_AT_linkage_name
+
+;; Check that -add-linkage-names-to-declaration-call-origins controls whether
+;; linkage names are added to declarations referenced by DW_AT_call_origin
+;; attributes.
+;;
+;; $ cat test.cpp
+;; void a();
+;; __attribute__((optnone))
+;; void b() {}
+;; void c();
+;; extern "C" {
+;;   void d();
+;; }
+;;
+;; void e() {
+;;   a(); //< Reference declaration DIE (add linkage name).
+;;   b(); //< Reference definition DIE (don't add linkage name).
+;;   c(); //< Reference definition DIE (don't add linkage name).
+;;   d(); //< Reference declaration DIE, but there's no linkage name.
+;; }
+;;
+;; __attribute__((optnone))
+;; void c() {}
+;; $ clang++ -emit-llvm -S -O1 -g
+
+; COMMON:       DW_TAG_call_site
+; ENABLE-NEXT:    DW_AT_call_origin    (0x[[a:[a-z0-9]+]] "_Z1av")
+; DISABLE-NEXT:   DW_AT_call_origin    (0x[[a:[a-z0-9]+]] "a")
+; COMMON:       DW_TAG_call_site
+; COMMON-NEXT:    DW_AT_call_origin     (0x[[b:[a-z0-9]+]] "b")
+; COMMON:       DW_TAG_call_site
+; COMMON-NEXT:    DW_AT_call_origin     (0x[[c:[a-z0-9]+]] "c")
+; COMMON:       DW_TAG_call_site
+; COMMON-NEXT:    DW_AT_call_origin     (0x[[d:[a-z0-9]+]] "d")
+
+; COMMON: 0x[[a]]: DW_TAG_subprogram
+; COMMON:   DW_AT_name  ("a")
+; ENABLE:   DW_AT_linkage_name  ("_Z1av")
+; COMMON: 0x[[b]]: DW_TAG_subprogram
+; COMMON:   DW_AT_name  ("b")
+; COMMON: 0x[[c]]: DW_TAG_subprogram
+; COMMON:   DW_AT_name  ("c")
+; COMMON: 0x[[d]]: DW_TAG_subprogram
+; COMMON:   DW_AT_name  ("d")
+
+target triple = "x86_64-unknown-linux-gnu"
+
+define dso_local void @_Z1ev() local_unnamed_addr !dbg !13 {
+entry:
+  tail call void @_Z1av(), !dbg !14
+  tail call void @_Z1bv(), !dbg !15
+  tail call void @_Z1cv(), !dbg !16
+  tail call void @d(), !dbg !17
+  ret void, !dbg !18
+}
+
+define dso_local void @_Z1bv() local_unnamed_addr !dbg !9 {
+entry:
+  ret void, !dbg !12
+}
+
+declare !dbg !19 void @_Z1av() local_unnamed_addr
+
+define dso_local void @_Z1cv() local_unnamed_addr !dbg !20 {
+entry:
+  ret void, !dbg !21
+}
+declare !dbg !22 void @d() local_unnamed_addr
+
+!llvm.dbg.cu = !{!0}
+!llvm.module.flags = !{!2, !3}
+!llvm.ident = !{!8}
+
+!0 = distinct !DICompileUnit(language: DW_LANG_C_plus_plus_14, file: !1, producer: "clang version 19.0.0git", isOptimized: true, runtimeVersion: 0, emissionKind: FullDebug, splitDebugInlining: false, nameTableKind: None)
+!1 = !DIFile(filename: "test.cpp", directory: "/")
+!2 = !{i32 7, !"Dwarf Version", i32 5}
+!3 = !{i32 2, !"Debug Info Version", i32 3}
+!8 = !{!"clang version 19.0.0"}
+!9 = distinct !DISubprogram(name: "b", linkageName: "_Z1bv", scope: !1, file: !1, line: 3, type: !10, scopeLine: 3, flags: DIFlagPrototyped | DIFlagAllCallsDescribed, spFlags: DISPFlagDefinition | DISPFlagOptimized, unit: !0)
+!10 = !DISubroutineType(types: !11)
+!11 = !{null}
+!12 = !DILocation(line: 3, column: 11, scope: !9)
+!13 = distinct !DISubprogram(name: "e", linkageName: "_Z1ev", scope: !1, file: !1, line: 9, type: !10, scopeLine: 9, flags: DIFlagPrototyped | DIFlagAllCallsDescribed, spFlags: DISPFlagDefinition | DISPFlagOptimized, unit: !0)
+!14 = !DILocation(line: 10, column: 3, scope: !13)
+!15 = !DILocation(line: 11, column: 3, scope: !13)
+!16 = !DILocation(line: 12, column: 3, scope: !13)
+!17 = !DILocation(line: 13, column: 3, scope: !13)
+!18 = !DILocation(line: 14, column: 1, scope: !13)
+!19 = !DISubprogram(name: "a", linkageName: "_Z1av", scope: !1, file: !1, line: 1, type: !10, flags: DIFlagPrototyped, spFlags: DISPFlagOptimized)
+!20 = distinct !DISubprogram(name: "c", linkageName: "_Z1cv", scope: !1, file: !1, line: 17, type: !10, scopeLine: 17, flags: DIFlagPrototyped | DIFlagAllCallsDescribed, spFlags: DISPFlagDefinition | DISPFlagOptimized, unit: !0)
+!21 = !DILocation(line: 17, column: 11, scope: !20)
+!22 = !DISubprogram(name: "d", scope: !1, file: !1, line: 6, type: !10, flags: DIFlagPrototyped, spFlags: DISPFlagOptimized)
diff --git a/llvm/test/DebugInfo/X86/invalid-cu-abbrev-contribution-dwp.s b/llvm/test/DebugInfo/X86/invalid-cu-abbrev-contribution-dwp.s
new file mode 100644
index 000000000000..6cc539cf8594
--- /dev/null
+++ b/llvm/test/DebugInfo/X86/invalid-cu-abbrev-contribution-dwp.s
@@ -0,0 +1,58 @@
+# RUN: llvm-mc -triple x86_64-unknown-linux %s -filetype=obj -o - | \
+# RUN:   llvm-dwarfdump -debug-info - 2>&1 | FileCheck %s
+
+# CHECK: warning: DWARF package unit at offset 0x00000000 missing abbreviation column
+
+    .section .debug_abbrev.dwo, "e", @progbits
+.LAbbrBegin:
+    .uleb128 1                  # Abbreviation Code
+    .uleb128 17                 # DW_TAG_compile_unit
+    .byte 0                     # DW_CHILDREN_no
+    .uleb128 3                  # DW_AT_name
+    .uleb128 8                  # DW_FORM_string
+    .uleb128 0x2131             # DW_AT_GNU_dwo_id
+    .uleb128 7                  # DW_FORM_data8
+    .byte 0                     # EOM(1)
+    .byte 0                     # EOM(2)
+    .byte 0                     # EOM(3)
+.LAbbrEnd:
+
+    .section .debug_info.dwo, "e", @progbits
+.LCUBegin:
+    .long .LCUEnd-.LCUVersion   # Length
+.LCUVersion:
+    .short 4                    # Version
+    .long 0                     # Abbrev offset
+    .byte 4                     # Address size
+    .uleb128 1                  # Abbrev [1] DW_TAG_compile_unit
+    .asciz "a.c"                # DW_AT_name
+    .quad 0x1100001122222222    # DW_AT_GNU_dwo_id
+.LCUEnd:
+
+    .section .debug_cu_index, "", @progbits
+## Header:
+    .short 2                    # Version
+    .space 2                    # Padding
+    .long 1                     # Section count (Invalid, should be 2)
+    .long 1                     # Unit count
+    .long 4                     # Slot count
+## Hash Table of Signatures:
+    .quad 0
+    .quad 0
+    .quad 0x1100001122222222
+    .quad 0
+## Parallel Table of Indexes:
+    .long 0
+    .long 0
+    .long 1
+    .long 0
+## Table of Section Offsets:
+## Row 0:
+    .long 1                     # DW_SECT_INFO
+#    .long 3                     # DW_SECT_ABBREV (Intentionally omitted)
+## Row 1:
+    .long .LCUBegin-.debug_info.dwo     # Offset in .debug_info.dwo
+#    .long .LAbbrBegin-.debug_abbrev.dwo # Offset in .debug_abbrev.dwo (Intentionally omitted)
+## Table of Section Sizes:
+    .long .LCUEnd-.LCUBegin           # Size of the contribution in .debug_info.dwo
+    .long .LAbbrEnd-.LAbbrBegin       # Size of the contribution in .debug_abbrev.dwo (Intentionally omitted)
diff --git a/llvm/test/DebugInfo/X86/invalid-cu-abbrev-offset-dwp.s b/llvm/test/DebugInfo/X86/invalid-cu-abbrev-offset-dwp.s
new file mode 100644
index 000000000000..b9e769e3ed4a
--- /dev/null
+++ b/llvm/test/DebugInfo/X86/invalid-cu-abbrev-offset-dwp.s
@@ -0,0 +1,58 @@
+# RUN: llvm-mc -triple x86_64-unknown-linux %s -filetype=obj -o - | \
+# RUN:   llvm-dwarfdump -debug-info - 2>&1 | FileCheck %s
+
+# CHECK: warning: DWARF package unit at offset 0x00000000 has a non-zero abbreviation offset
+
+    .section .debug_abbrev.dwo, "e", @progbits
+.LAbbrBegin:
+    .uleb128 1                  # Abbreviation Code
+    .uleb128 17                 # DW_TAG_compile_unit
+    .byte 0                     # DW_CHILDREN_no
+    .uleb128 3                  # DW_AT_name
+    .uleb128 8                  # DW_FORM_string
+    .uleb128 0x2131             # DW_AT_GNU_dwo_id
+    .uleb128 7                  # DW_FORM_data8
+    .byte 0                     # EOM(1)
+    .byte 0                     # EOM(2)
+    .byte 0                     # EOM(3)
+.LAbbrEnd:
+
+    .section .debug_info.dwo, "e", @progbits
+.LCUBegin:
+    .long .LCUEnd-.LCUVersion   # Length
+.LCUVersion:
+    .short 4                    # Version
+    .long 1                     # Abbrev offset (Invalid, should be 0)
+    .byte 4                     # Address size
+    .uleb128 1                  # Abbrev [1] DW_TAG_compile_unit
+    .asciz "a.c"                # DW_AT_name
+    .quad 0x1100001122222222    # DW_AT_GNU_dwo_id
+.LCUEnd:
+
+    .section .debug_cu_index, "", @progbits
+## Header:
+    .short 2                    # Version
+    .space 2                    # Padding
+    .long 2                     # Section count
+    .long 1                     # Unit count
+    .long 4                     # Slot count
+## Hash Table of Signatures:
+    .quad 0
+    .quad 0
+    .quad 0x1100001122222222
+    .quad 0
+## Parallel Table of Indexes:
+    .long 0
+    .long 0
+    .long 1
+    .long 0
+## Table of Section Offsets:
+## Row 0:
+    .long 1                     # DW_SECT_INFO
+    .long 3                     # DW_SECT_ABBREV
+## Row 1:
+    .long .LCUBegin-.debug_info.dwo     # Offset in .debug_info.dwo
+    .long .LAbbrBegin-.debug_abbrev.dwo # Offset in .debug_abbrev.dwo
+## Table of Section Sizes:
+    .long .LCUEnd-.LCUBegin
+    .long .LAbbrEnd-.LAbbrBegin
diff --git a/llvm/test/DebugInfo/X86/invalid-cu-length-dwp.s b/llvm/test/DebugInfo/X86/invalid-cu-length-dwp.s
index 299bea1bf1a7..d67416736093 100644
--- a/llvm/test/DebugInfo/X86/invalid-cu-length-dwp.s
+++ b/llvm/test/DebugInfo/X86/invalid-cu-length-dwp.s
@@ -1,9 +1,11 @@
 # RUN: llvm-mc -triple x86_64-unknown-linux %s -filetype=obj -o - | \
-# RUN:   llvm-dwarfdump -debug-info -
+# RUN:   llvm-dwarfdump -debug-info - 2>&1 | FileCheck %s
 
 ## llvm-dwarfdump used to crash with this input because of an invalid size
 ## of the compilation unit contribution in the .debug_cu_index section.
 
+# CHECK: warning: DWARF package unit at offset 0x00000000 has an inconsistent index (expected: 23, actual: 24)
+
     .section .debug_abbrev.dwo, "e", @progbits
 .LAbbrBegin:
     .uleb128 1                  # Abbreviation Code
diff --git a/llvm/test/ExecutionEngine/JITLink/RISCV/anonymous_symbol.s b/llvm/test/ExecutionEngine/JITLink/RISCV/anonymous_symbol.s
index a5038022dfe0..e7114e4d643c 100644
--- a/llvm/test/ExecutionEngine/JITLink/RISCV/anonymous_symbol.s
+++ b/llvm/test/ExecutionEngine/JITLink/RISCV/anonymous_symbol.s
@@ -7,7 +7,7 @@
 # the section start and section end. So that by relocating these symbol, the section length
 # can be calculated.
 #
-# CHECK: Creating defined graph symbol for ELF symbol ""
+# CHECK: Creating defined graph symbol for ELF symbol ".L0 "
 # CHECK: Creating defined graph symbol for ELF symbol "main"
         .text
         .globl main
diff --git a/llvm/test/Instrumentation/HWAddressSanitizer/basic.ll b/llvm/test/Instrumentation/HWAddressSanitizer/basic.ll
index 5dededaccaad..4212293f4254 100644
--- a/llvm/test/Instrumentation/HWAddressSanitizer/basic.ll
+++ b/llvm/test/Instrumentation/HWAddressSanitizer/basic.ll
@@ -9,8 +9,6 @@
 ; RUN: opt < %s -passes=hwasan -hwasan-recover=0 -hwasan-mapping-offset=0 -S | FileCheck %s --check-prefixes=ABORT-ZERO-BASED-SHADOW
 ; RUN: opt < %s -passes=hwasan -hwasan-recover=1 -hwasan-mapping-offset=0 -S | FileCheck %s --check-prefixes=RECOVER-ZERO-BASED-SHADOW
 
-; CHECK: @llvm.used = appending global [1 x ptr] [ptr @hwasan.module_ctor]
-; CHECK: @llvm.global_ctors = appending global [1 x { i32, ptr, ptr }] [{ i32, ptr, ptr } { i32 0, ptr @hwasan.module_ctor, ptr @hwasan.module_ctor }]
 
 target datalayout = "e-m:e-i8:8:32-i16:16:32-i64:64-i128:128-n32:64-S128"
 target triple = "aarch64--linux-android10000"
@@ -101,7 +99,7 @@ define i8 @test_load8(ptr %a) sanitize_hwaddress {
 ; ABORT-ZERO-BASED-SHADOW-SAME: (ptr [[A:%.*]]) #[[ATTR0:[0-9]+]] {
 ; ABORT-ZERO-BASED-SHADOW-NEXT:  entry:
 ; ABORT-ZERO-BASED-SHADOW-NEXT:    [[DOTHWASAN_SHADOW:%.*]] = call ptr asm "", "=r,0"(ptr null)
-; ABORT-ZERO-BASED-SHADOW-NEXT:    call void @llvm.hwasan.check.memaccess.shortgranules(ptr [[DOTHWASAN_SHADOW]], ptr [[A]], i32 0)
+; ABORT-ZERO-BASED-SHADOW-NEXT:    call void @llvm.hwasan.check.memaccess.shortgranules.fixedshadow(ptr [[A]], i32 0, i64 0)
 ; ABORT-ZERO-BASED-SHADOW-NEXT:    [[B:%.*]] = load i8, ptr [[A]], align 4
 ; ABORT-ZERO-BASED-SHADOW-NEXT:    ret i8 [[B]]
 ;
@@ -233,7 +231,7 @@ define i16 @test_load16(ptr %a) sanitize_hwaddress {
 ; ABORT-ZERO-BASED-SHADOW-SAME: (ptr [[A:%.*]]) #[[ATTR0]] {
 ; ABORT-ZERO-BASED-SHADOW-NEXT:  entry:
 ; ABORT-ZERO-BASED-SHADOW-NEXT:    [[DOTHWASAN_SHADOW:%.*]] = call ptr asm "", "=r,0"(ptr null)
-; ABORT-ZERO-BASED-SHADOW-NEXT:    call void @llvm.hwasan.check.memaccess.shortgranules(ptr [[DOTHWASAN_SHADOW]], ptr [[A]], i32 1)
+; ABORT-ZERO-BASED-SHADOW-NEXT:    call void @llvm.hwasan.check.memaccess.shortgranules.fixedshadow(ptr [[A]], i32 1, i64 0)
 ; ABORT-ZERO-BASED-SHADOW-NEXT:    [[B:%.*]] = load i16, ptr [[A]], align 4
 ; ABORT-ZERO-BASED-SHADOW-NEXT:    ret i16 [[B]]
 ;
@@ -365,7 +363,7 @@ define i32 @test_load32(ptr %a) sanitize_hwaddress {
 ; ABORT-ZERO-BASED-SHADOW-SAME: (ptr [[A:%.*]]) #[[ATTR0]] {
 ; ABORT-ZERO-BASED-SHADOW-NEXT:  entry:
 ; ABORT-ZERO-BASED-SHADOW-NEXT:    [[DOTHWASAN_SHADOW:%.*]] = call ptr asm "", "=r,0"(ptr null)
-; ABORT-ZERO-BASED-SHADOW-NEXT:    call void @llvm.hwasan.check.memaccess.shortgranules(ptr [[DOTHWASAN_SHADOW]], ptr [[A]], i32 2)
+; ABORT-ZERO-BASED-SHADOW-NEXT:    call void @llvm.hwasan.check.memaccess.shortgranules.fixedshadow(ptr [[A]], i32 2, i64 0)
 ; ABORT-ZERO-BASED-SHADOW-NEXT:    [[B:%.*]] = load i32, ptr [[A]], align 4
 ; ABORT-ZERO-BASED-SHADOW-NEXT:    ret i32 [[B]]
 ;
@@ -497,7 +495,7 @@ define i64 @test_load64(ptr %a) sanitize_hwaddress {
 ; ABORT-ZERO-BASED-SHADOW-SAME: (ptr [[A:%.*]]) #[[ATTR0]] {
 ; ABORT-ZERO-BASED-SHADOW-NEXT:  entry:
 ; ABORT-ZERO-BASED-SHADOW-NEXT:    [[DOTHWASAN_SHADOW:%.*]] = call ptr asm "", "=r,0"(ptr null)
-; ABORT-ZERO-BASED-SHADOW-NEXT:    call void @llvm.hwasan.check.memaccess.shortgranules(ptr [[DOTHWASAN_SHADOW]], ptr [[A]], i32 3)
+; ABORT-ZERO-BASED-SHADOW-NEXT:    call void @llvm.hwasan.check.memaccess.shortgranules.fixedshadow(ptr [[A]], i32 3, i64 0)
 ; ABORT-ZERO-BASED-SHADOW-NEXT:    [[B:%.*]] = load i64, ptr [[A]], align 8
 ; ABORT-ZERO-BASED-SHADOW-NEXT:    ret i64 [[B]]
 ;
@@ -629,7 +627,7 @@ define i128 @test_load128(ptr %a) sanitize_hwaddress {
 ; ABORT-ZERO-BASED-SHADOW-SAME: (ptr [[A:%.*]]) #[[ATTR0]] {
 ; ABORT-ZERO-BASED-SHADOW-NEXT:  entry:
 ; ABORT-ZERO-BASED-SHADOW-NEXT:    [[DOTHWASAN_SHADOW:%.*]] = call ptr asm "", "=r,0"(ptr null)
-; ABORT-ZERO-BASED-SHADOW-NEXT:    call void @llvm.hwasan.check.memaccess.shortgranules(ptr [[DOTHWASAN_SHADOW]], ptr [[A]], i32 4)
+; ABORT-ZERO-BASED-SHADOW-NEXT:    call void @llvm.hwasan.check.memaccess.shortgranules.fixedshadow(ptr [[A]], i32 4, i64 0)
 ; ABORT-ZERO-BASED-SHADOW-NEXT:    [[B:%.*]] = load i128, ptr [[A]], align 16
 ; ABORT-ZERO-BASED-SHADOW-NEXT:    ret i128 [[B]]
 ;
@@ -830,7 +828,7 @@ define void @test_store8(ptr %a, i8 %b) sanitize_hwaddress {
 ; ABORT-ZERO-BASED-SHADOW-SAME: (ptr [[A:%.*]], i8 [[B:%.*]]) #[[ATTR0]] {
 ; ABORT-ZERO-BASED-SHADOW-NEXT:  entry:
 ; ABORT-ZERO-BASED-SHADOW-NEXT:    [[DOTHWASAN_SHADOW:%.*]] = call ptr asm "", "=r,0"(ptr null)
-; ABORT-ZERO-BASED-SHADOW-NEXT:    call void @llvm.hwasan.check.memaccess.shortgranules(ptr [[DOTHWASAN_SHADOW]], ptr [[A]], i32 16)
+; ABORT-ZERO-BASED-SHADOW-NEXT:    call void @llvm.hwasan.check.memaccess.shortgranules.fixedshadow(ptr [[A]], i32 16, i64 0)
 ; ABORT-ZERO-BASED-SHADOW-NEXT:    store i8 [[B]], ptr [[A]], align 4
 ; ABORT-ZERO-BASED-SHADOW-NEXT:    ret void
 ;
@@ -962,7 +960,7 @@ define void @test_store16(ptr %a, i16 %b) sanitize_hwaddress {
 ; ABORT-ZERO-BASED-SHADOW-SAME: (ptr [[A:%.*]], i16 [[B:%.*]]) #[[ATTR0]] {
 ; ABORT-ZERO-BASED-SHADOW-NEXT:  entry:
 ; ABORT-ZERO-BASED-SHADOW-NEXT:    [[DOTHWASAN_SHADOW:%.*]] = call ptr asm "", "=r,0"(ptr null)
-; ABORT-ZERO-BASED-SHADOW-NEXT:    call void @llvm.hwasan.check.memaccess.shortgranules(ptr [[DOTHWASAN_SHADOW]], ptr [[A]], i32 17)
+; ABORT-ZERO-BASED-SHADOW-NEXT:    call void @llvm.hwasan.check.memaccess.shortgranules.fixedshadow(ptr [[A]], i32 17, i64 0)
 ; ABORT-ZERO-BASED-SHADOW-NEXT:    store i16 [[B]], ptr [[A]], align 4
 ; ABORT-ZERO-BASED-SHADOW-NEXT:    ret void
 ;
@@ -1094,7 +1092,7 @@ define void @test_store32(ptr %a, i32 %b) sanitize_hwaddress {
 ; ABORT-ZERO-BASED-SHADOW-SAME: (ptr [[A:%.*]], i32 [[B:%.*]]) #[[ATTR0]] {
 ; ABORT-ZERO-BASED-SHADOW-NEXT:  entry:
 ; ABORT-ZERO-BASED-SHADOW-NEXT:    [[DOTHWASAN_SHADOW:%.*]] = call ptr asm "", "=r,0"(ptr null)
-; ABORT-ZERO-BASED-SHADOW-NEXT:    call void @llvm.hwasan.check.memaccess.shortgranules(ptr [[DOTHWASAN_SHADOW]], ptr [[A]], i32 18)
+; ABORT-ZERO-BASED-SHADOW-NEXT:    call void @llvm.hwasan.check.memaccess.shortgranules.fixedshadow(ptr [[A]], i32 18, i64 0)
 ; ABORT-ZERO-BASED-SHADOW-NEXT:    store i32 [[B]], ptr [[A]], align 4
 ; ABORT-ZERO-BASED-SHADOW-NEXT:    ret void
 ;
@@ -1226,7 +1224,7 @@ define void @test_store64(ptr %a, i64 %b) sanitize_hwaddress {
 ; ABORT-ZERO-BASED-SHADOW-SAME: (ptr [[A:%.*]], i64 [[B:%.*]]) #[[ATTR0]] {
 ; ABORT-ZERO-BASED-SHADOW-NEXT:  entry:
 ; ABORT-ZERO-BASED-SHADOW-NEXT:    [[DOTHWASAN_SHADOW:%.*]] = call ptr asm "", "=r,0"(ptr null)
-; ABORT-ZERO-BASED-SHADOW-NEXT:    call void @llvm.hwasan.check.memaccess.shortgranules(ptr [[DOTHWASAN_SHADOW]], ptr [[A]], i32 19)
+; ABORT-ZERO-BASED-SHADOW-NEXT:    call void @llvm.hwasan.check.memaccess.shortgranules.fixedshadow(ptr [[A]], i32 19, i64 0)
 ; ABORT-ZERO-BASED-SHADOW-NEXT:    store i64 [[B]], ptr [[A]], align 8
 ; ABORT-ZERO-BASED-SHADOW-NEXT:    ret void
 ;
@@ -1358,7 +1356,7 @@ define void @test_store128(ptr %a, i128 %b) sanitize_hwaddress {
 ; ABORT-ZERO-BASED-SHADOW-SAME: (ptr [[A:%.*]], i128 [[B:%.*]]) #[[ATTR0]] {
 ; ABORT-ZERO-BASED-SHADOW-NEXT:  entry:
 ; ABORT-ZERO-BASED-SHADOW-NEXT:    [[DOTHWASAN_SHADOW:%.*]] = call ptr asm "", "=r,0"(ptr null)
-; ABORT-ZERO-BASED-SHADOW-NEXT:    call void @llvm.hwasan.check.memaccess.shortgranules(ptr [[DOTHWASAN_SHADOW]], ptr [[A]], i32 20)
+; ABORT-ZERO-BASED-SHADOW-NEXT:    call void @llvm.hwasan.check.memaccess.shortgranules.fixedshadow(ptr [[A]], i32 20, i64 0)
 ; ABORT-ZERO-BASED-SHADOW-NEXT:    store i128 [[B]], ptr [[A]], align 16
 ; ABORT-ZERO-BASED-SHADOW-NEXT:    ret void
 ;
diff --git a/llvm/test/Instrumentation/HWAddressSanitizer/fixed-shadow.ll b/llvm/test/Instrumentation/HWAddressSanitizer/fixed-shadow.ll
new file mode 100644
index 000000000000..980189c5607f
--- /dev/null
+++ b/llvm/test/Instrumentation/HWAddressSanitizer/fixed-shadow.ll
@@ -0,0 +1,229 @@
+; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --version 2
+; Test basic address sanitizer instrumentation.
+;
+; RUN: opt < %s -passes=hwasan -hwasan-mapping-offset=4398046511104 -S | FileCheck %s
+
+
+target datalayout = "e-m:e-i8:8:32-i16:16:32-i64:64-i128:128-n32:64-S128"
+target triple = "aarch64--linux-android9001"
+
+define i8 @test_load8(ptr %a) sanitize_hwaddress {
+; CHECK-LABEL: define i8 @test_load8
+; CHECK-SAME: (ptr [[A:%.*]]) #[[ATTR0:[0-9]+]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[DOTHWASAN_SHADOW:%.*]] = call ptr asm "", "=r,0"(ptr inttoptr (i64 4398046511104 to ptr))
+; CHECK-NEXT:    call void @llvm.hwasan.check.memaccess.shortgranules.fixedshadow(ptr [[A]], i32 0, i64 4398046511104)
+; CHECK-NEXT:    [[B:%.*]] = load i8, ptr [[A]], align 4
+; CHECK-NEXT:    ret i8 [[B]]
+;
+entry:
+  %b = load i8, ptr %a, align 4
+  ret i8 %b
+}
+
+define i16 @test_load16(ptr %a) sanitize_hwaddress {
+; CHECK-LABEL: define i16 @test_load16
+; CHECK-SAME: (ptr [[A:%.*]]) #[[ATTR0]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[DOTHWASAN_SHADOW:%.*]] = call ptr asm "", "=r,0"(ptr inttoptr (i64 4398046511104 to ptr))
+; CHECK-NEXT:    call void @llvm.hwasan.check.memaccess.shortgranules.fixedshadow(ptr [[A]], i32 1, i64 4398046511104)
+; CHECK-NEXT:    [[B:%.*]] = load i16, ptr [[A]], align 4
+; CHECK-NEXT:    ret i16 [[B]]
+;
+entry:
+  %b = load i16, ptr %a, align 4
+  ret i16 %b
+}
+
+define i32 @test_load32(ptr %a) sanitize_hwaddress {
+; CHECK-LABEL: define i32 @test_load32
+; CHECK-SAME: (ptr [[A:%.*]]) #[[ATTR0]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[DOTHWASAN_SHADOW:%.*]] = call ptr asm "", "=r,0"(ptr inttoptr (i64 4398046511104 to ptr))
+; CHECK-NEXT:    call void @llvm.hwasan.check.memaccess.shortgranules.fixedshadow(ptr [[A]], i32 2, i64 4398046511104)
+; CHECK-NEXT:    [[B:%.*]] = load i32, ptr [[A]], align 4
+; CHECK-NEXT:    ret i32 [[B]]
+;
+entry:
+  %b = load i32, ptr %a, align 4
+  ret i32 %b
+}
+
+define i64 @test_load64(ptr %a) sanitize_hwaddress {
+; CHECK-LABEL: define i64 @test_load64
+; CHECK-SAME: (ptr [[A:%.*]]) #[[ATTR0]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[DOTHWASAN_SHADOW:%.*]] = call ptr asm "", "=r,0"(ptr inttoptr (i64 4398046511104 to ptr))
+; CHECK-NEXT:    call void @llvm.hwasan.check.memaccess.shortgranules.fixedshadow(ptr [[A]], i32 3, i64 4398046511104)
+; CHECK-NEXT:    [[B:%.*]] = load i64, ptr [[A]], align 8
+; CHECK-NEXT:    ret i64 [[B]]
+;
+entry:
+  %b = load i64, ptr %a, align 8
+  ret i64 %b
+}
+
+define i128 @test_load128(ptr %a) sanitize_hwaddress {
+; CHECK-LABEL: define i128 @test_load128
+; CHECK-SAME: (ptr [[A:%.*]]) #[[ATTR0]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[DOTHWASAN_SHADOW:%.*]] = call ptr asm "", "=r,0"(ptr inttoptr (i64 4398046511104 to ptr))
+; CHECK-NEXT:    call void @llvm.hwasan.check.memaccess.shortgranules.fixedshadow(ptr [[A]], i32 4, i64 4398046511104)
+; CHECK-NEXT:    [[B:%.*]] = load i128, ptr [[A]], align 16
+; CHECK-NEXT:    ret i128 [[B]]
+;
+entry:
+  %b = load i128, ptr %a, align 16
+  ret i128 %b
+}
+
+define i40 @test_load40(ptr %a) sanitize_hwaddress {
+; CHECK-LABEL: define i40 @test_load40
+; CHECK-SAME: (ptr [[A:%.*]]) #[[ATTR0]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[DOTHWASAN_SHADOW:%.*]] = call ptr asm "", "=r,0"(ptr inttoptr (i64 4398046511104 to ptr))
+; CHECK-NEXT:    [[TMP0:%.*]] = ptrtoint ptr [[A]] to i64
+; CHECK-NEXT:    call void @__hwasan_loadN(i64 [[TMP0]], i64 5)
+; CHECK-NEXT:    [[B:%.*]] = load i40, ptr [[A]], align 4
+; CHECK-NEXT:    ret i40 [[B]]
+;
+entry:
+  %b = load i40, ptr %a, align 4
+  ret i40 %b
+}
+
+define void @test_store8(ptr %a, i8 %b) sanitize_hwaddress {
+; CHECK-LABEL: define void @test_store8
+; CHECK-SAME: (ptr [[A:%.*]], i8 [[B:%.*]]) #[[ATTR0]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[DOTHWASAN_SHADOW:%.*]] = call ptr asm "", "=r,0"(ptr inttoptr (i64 4398046511104 to ptr))
+; CHECK-NEXT:    call void @llvm.hwasan.check.memaccess.shortgranules.fixedshadow(ptr [[A]], i32 16, i64 4398046511104)
+; CHECK-NEXT:    store i8 [[B]], ptr [[A]], align 4
+; CHECK-NEXT:    ret void
+;
+entry:
+  store i8 %b, ptr %a, align 4
+  ret void
+}
+
+define void @test_store16(ptr %a, i16 %b) sanitize_hwaddress {
+; CHECK-LABEL: define void @test_store16
+; CHECK-SAME: (ptr [[A:%.*]], i16 [[B:%.*]]) #[[ATTR0]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[DOTHWASAN_SHADOW:%.*]] = call ptr asm "", "=r,0"(ptr inttoptr (i64 4398046511104 to ptr))
+; CHECK-NEXT:    call void @llvm.hwasan.check.memaccess.shortgranules.fixedshadow(ptr [[A]], i32 17, i64 4398046511104)
+; CHECK-NEXT:    store i16 [[B]], ptr [[A]], align 4
+; CHECK-NEXT:    ret void
+;
+entry:
+  store i16 %b, ptr %a, align 4
+  ret void
+}
+
+define void @test_store32(ptr %a, i32 %b) sanitize_hwaddress {
+; CHECK-LABEL: define void @test_store32
+; CHECK-SAME: (ptr [[A:%.*]], i32 [[B:%.*]]) #[[ATTR0]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[DOTHWASAN_SHADOW:%.*]] = call ptr asm "", "=r,0"(ptr inttoptr (i64 4398046511104 to ptr))
+; CHECK-NEXT:    call void @llvm.hwasan.check.memaccess.shortgranules.fixedshadow(ptr [[A]], i32 18, i64 4398046511104)
+; CHECK-NEXT:    store i32 [[B]], ptr [[A]], align 4
+; CHECK-NEXT:    ret void
+;
+entry:
+  store i32 %b, ptr %a, align 4
+  ret void
+}
+
+define void @test_store64(ptr %a, i64 %b) sanitize_hwaddress {
+; CHECK-LABEL: define void @test_store64
+; CHECK-SAME: (ptr [[A:%.*]], i64 [[B:%.*]]) #[[ATTR0]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[DOTHWASAN_SHADOW:%.*]] = call ptr asm "", "=r,0"(ptr inttoptr (i64 4398046511104 to ptr))
+; CHECK-NEXT:    call void @llvm.hwasan.check.memaccess.shortgranules.fixedshadow(ptr [[A]], i32 19, i64 4398046511104)
+; CHECK-NEXT:    store i64 [[B]], ptr [[A]], align 8
+; CHECK-NEXT:    ret void
+;
+entry:
+  store i64 %b, ptr %a, align 8
+  ret void
+}
+
+define void @test_store128(ptr %a, i128 %b) sanitize_hwaddress {
+; CHECK-LABEL: define void @test_store128
+; CHECK-SAME: (ptr [[A:%.*]], i128 [[B:%.*]]) #[[ATTR0]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[DOTHWASAN_SHADOW:%.*]] = call ptr asm "", "=r,0"(ptr inttoptr (i64 4398046511104 to ptr))
+; CHECK-NEXT:    call void @llvm.hwasan.check.memaccess.shortgranules.fixedshadow(ptr [[A]], i32 20, i64 4398046511104)
+; CHECK-NEXT:    store i128 [[B]], ptr [[A]], align 16
+; CHECK-NEXT:    ret void
+;
+entry:
+  store i128 %b, ptr %a, align 16
+  ret void
+}
+
+define void @test_store40(ptr %a, i40 %b) sanitize_hwaddress {
+; CHECK-LABEL: define void @test_store40
+; CHECK-SAME: (ptr [[A:%.*]], i40 [[B:%.*]]) #[[ATTR0]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[DOTHWASAN_SHADOW:%.*]] = call ptr asm "", "=r,0"(ptr inttoptr (i64 4398046511104 to ptr))
+; CHECK-NEXT:    [[TMP0:%.*]] = ptrtoint ptr [[A]] to i64
+; CHECK-NEXT:    call void @__hwasan_storeN(i64 [[TMP0]], i64 5)
+; CHECK-NEXT:    store i40 [[B]], ptr [[A]], align 4
+; CHECK-NEXT:    ret void
+;
+entry:
+  store i40 %b, ptr %a, align 4
+  ret void
+}
+
+define void @test_store_unaligned(ptr %a, i64 %b) sanitize_hwaddress {
+; CHECK-LABEL: define void @test_store_unaligned
+; CHECK-SAME: (ptr [[A:%.*]], i64 [[B:%.*]]) #[[ATTR0]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[DOTHWASAN_SHADOW:%.*]] = call ptr asm "", "=r,0"(ptr inttoptr (i64 4398046511104 to ptr))
+; CHECK-NEXT:    [[TMP0:%.*]] = ptrtoint ptr [[A]] to i64
+; CHECK-NEXT:    call void @__hwasan_storeN(i64 [[TMP0]], i64 8)
+; CHECK-NEXT:    store i64 [[B]], ptr [[A]], align 4
+; CHECK-NEXT:    ret void
+;
+entry:
+  store i64 %b, ptr %a, align 4
+  ret void
+}
+
+define i8 @test_load_noattr(ptr %a) {
+; CHECK-LABEL: define i8 @test_load_noattr
+; CHECK-SAME: (ptr [[A:%.*]]) {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[B:%.*]] = load i8, ptr [[A]], align 4
+; CHECK-NEXT:    ret i8 [[B]]
+;
+entry:
+  %b = load i8, ptr %a, align 4
+  ret i8 %b
+}
+
+define i8 @test_load_notmyattr(ptr %a) sanitize_address {
+; CHECK-LABEL: define i8 @test_load_notmyattr
+; CHECK-SAME: (ptr [[A:%.*]]) #[[ATTR1:[0-9]+]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[B:%.*]] = load i8, ptr [[A]], align 4
+; CHECK-NEXT:    ret i8 [[B]]
+;
+entry:
+  %b = load i8, ptr %a, align 4
+  ret i8 %b
+}
+
+define i8 @test_load_addrspace(ptr addrspace(256) %a) sanitize_hwaddress {
+; CHECK-LABEL: define i8 @test_load_addrspace
+; CHECK-SAME: (ptr addrspace(256) [[A:%.*]]) #[[ATTR0]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[B:%.*]] = load i8, ptr addrspace(256) [[A]], align 4
+; CHECK-NEXT:    ret i8 [[B]]
+;
+entry:
+  %b = load i8, ptr addrspace(256) %a, align 4
+  ret i8 %b
+}
diff --git a/llvm/test/Instrumentation/HWAddressSanitizer/prologue.ll b/llvm/test/Instrumentation/HWAddressSanitizer/prologue.ll
index f8e0364ba19f..49f0bf739cb6 100644
--- a/llvm/test/Instrumentation/HWAddressSanitizer/prologue.ll
+++ b/llvm/test/Instrumentation/HWAddressSanitizer/prologue.ll
@@ -69,7 +69,7 @@ define i32 @test_load(ptr %a) sanitize_hwaddress {
 ; FUCHSIA-SAME: (ptr [[A:%.*]]) #[[ATTR0:[0-9]+]] {
 ; FUCHSIA-NEXT:  entry:
 ; FUCHSIA-NEXT:    [[DOTHWASAN_SHADOW:%.*]] = call ptr asm "", "=r,0"(ptr null)
-; FUCHSIA-NEXT:    call void @llvm.hwasan.check.memaccess.shortgranules(ptr [[DOTHWASAN_SHADOW]], ptr [[A]], i32 2)
+; FUCHSIA-NEXT:    call void @llvm.hwasan.check.memaccess.shortgranules.fixedshadow(ptr [[A]], i32 2, i64 0)
 ; FUCHSIA-NEXT:    [[X:%.*]] = load i32, ptr [[A]], align 4
 ; FUCHSIA-NEXT:    ret i32 [[X]]
 ;
@@ -77,7 +77,7 @@ define i32 @test_load(ptr %a) sanitize_hwaddress {
 ; FUCHSIA-LIBCALL-SAME: (ptr [[A:%.*]]) #[[ATTR0:[0-9]+]] {
 ; FUCHSIA-LIBCALL-NEXT:  entry:
 ; FUCHSIA-LIBCALL-NEXT:    [[DOTHWASAN_SHADOW:%.*]] = call ptr asm "", "=r,0"(ptr null)
-; FUCHSIA-LIBCALL-NEXT:    call void @llvm.hwasan.check.memaccess.shortgranules(ptr [[DOTHWASAN_SHADOW]], ptr [[A]], i32 2)
+; FUCHSIA-LIBCALL-NEXT:    call void @llvm.hwasan.check.memaccess.shortgranules.fixedshadow(ptr [[A]], i32 2, i64 0)
 ; FUCHSIA-LIBCALL-NEXT:    [[X:%.*]] = load i32, ptr [[A]], align 4
 ; FUCHSIA-LIBCALL-NEXT:    ret i32 [[X]]
 ;
diff --git a/llvm/test/Instrumentation/InstrProfiling/Coro/coro-split-musttail.ll b/llvm/test/Instrumentation/InstrProfiling/Coro/coro-split-musttail.ll
deleted file mode 100644
index a7321833d748..000000000000
--- a/llvm/test/Instrumentation/InstrProfiling/Coro/coro-split-musttail.ll
+++ /dev/null
@@ -1,63 +0,0 @@
-; Tests that instrumentation doesn't interfere with lowering (coro-split).
-; It should convert coro.resume followed by a suspend to a musttail call.
-
-; RUN: opt < %s -passes='pgo-instr-gen,cgscc(coro-split),simplifycfg,early-cse' -S | FileCheck %s
-
-define void @f() #0 {
-entry:
-  %id = call token @llvm.coro.id(i32 0, ptr null, ptr null, ptr null)
-  %alloc = call ptr @malloc(i64 16) #3
-  %vFrame = call noalias nonnull ptr @llvm.coro.begin(token %id, ptr %alloc)
-
-  %save = call token @llvm.coro.save(ptr null)
-  %addr1 = call ptr @llvm.coro.subfn.addr(ptr null, i8 0)
-  call fastcc void %addr1(ptr null)
-
-  %suspend = call i8 @llvm.coro.suspend(token %save, i1 false)
-  switch i8 %suspend, label %exit [
-    i8 0, label %await.ready
-    i8 1, label %exit
-  ]
-await.ready:
-  %save2 = call token @llvm.coro.save(ptr null)
-  %addr2 = call ptr @llvm.coro.subfn.addr(ptr null, i8 0)
-  call fastcc void %addr2(ptr null)
-
-  %suspend2 = call i8 @llvm.coro.suspend(token %save2, i1 false)
-  switch i8 %suspend2, label %exit [
-    i8 0, label %exit
-    i8 1, label %exit
-  ]
-exit:
-  call i1 @llvm.coro.end(ptr null, i1 false, token none)
-  ret void
-}
-
-; Verify that in the initial function resume is not marked with musttail.
-; CHECK-LABEL: @f(
-; CHECK: %[[addr1:.+]] = call ptr @llvm.coro.subfn.addr(ptr null, i8 0)
-; CHECK-NOT: musttail call fastcc void %[[addr1]](ptr null)
-
-; Verify that in the resume part resume call is marked with musttail.
-; CHECK-LABEL: @f.resume(
-; CHECK: %[[addr2:.+]] = call ptr @llvm.coro.subfn.addr(ptr null, i8 0)
-; CHECK: call void @llvm.instrprof
-; CHECK-NEXT: musttail call fastcc void %[[addr2]](ptr null)
-; CHECK-NEXT: ret void
-
-declare token @llvm.coro.id(i32, ptr readnone, ptr nocapture readonly, ptr) #1
-declare i1 @llvm.coro.alloc(token) #2
-declare i64 @llvm.coro.size.i64() #3
-declare ptr @llvm.coro.begin(token, ptr writeonly) #2
-declare token @llvm.coro.save(ptr) #2
-declare ptr @llvm.coro.frame() #3
-declare i8 @llvm.coro.suspend(token, i1) #2
-declare ptr @llvm.coro.free(token, ptr nocapture readonly) #1
-declare i1 @llvm.coro.end(ptr, i1, token) #2
-declare ptr @llvm.coro.subfn.addr(ptr nocapture readonly, i8) #1
-declare ptr @malloc(i64)
-
-attributes #0 = { presplitcoroutine }
-attributes #1 = { argmemonly nounwind readonly }
-attributes #2 = { nounwind }
-attributes #3 = { nounwind readnone }
diff --git a/llvm/test/Instrumentation/InstrProfiling/Coro/coro-split-musttail1.ll b/llvm/test/Instrumentation/InstrProfiling/Coro/coro-split-musttail1.ll
deleted file mode 100644
index 6098dee9a580..000000000000
--- a/llvm/test/Instrumentation/InstrProfiling/Coro/coro-split-musttail1.ll
+++ /dev/null
@@ -1,97 +0,0 @@
-; Tests that instrumentation doesn't interfere with lowering (coro-split).
-; It should convert coro.resume followed by a suspend to a musttail call.
-
-; RUN: opt < %s -passes='pgo-instr-gen,cgscc(coro-split),simplifycfg,early-cse' -S | FileCheck %s
-
-define void @f() #0 {
-entry:
-  %id = call token @llvm.coro.id(i32 0, ptr null, ptr null, ptr null)
-  %alloc = call ptr @malloc(i64 16) #3
-  %vFrame = call noalias nonnull ptr @llvm.coro.begin(token %id, ptr %alloc)
-
-  %save = call token @llvm.coro.save(ptr null)
-  %addr1 = call ptr @llvm.coro.subfn.addr(ptr null, i8 0)
-  call fastcc void %addr1(ptr null)
-
-  %suspend = call i8 @llvm.coro.suspend(token %save, i1 false)
-  switch i8 %suspend, label %exit [
-    i8 0, label %await.suspend
-    i8 1, label %exit
-  ]
-await.suspend:
-  %save2 = call token @llvm.coro.save(ptr null)
-  %br0 = call i8 @switch_result()
-  switch i8 %br0, label %unreach [
-    i8 0, label %await.resume3
-    i8 1, label %await.resume1
-    i8 2, label %await.resume2
-  ]
-await.resume1:
-  %hdl = call ptr @g()
-  %addr2 = call ptr @llvm.coro.subfn.addr(ptr %hdl, i8 0)
-  call fastcc void %addr2(ptr %hdl)
-  br label %final.suspend
-await.resume2:
-  %hdl2 = call ptr @h()
-  %addr3 = call ptr @llvm.coro.subfn.addr(ptr %hdl2, i8 0)
-  call fastcc void %addr3(ptr %hdl2)
-  br label %final.suspend
-await.resume3:
-  %addr4 = call ptr @llvm.coro.subfn.addr(ptr null, i8 0)
-  call fastcc void %addr4(ptr null)
-  br label %final.suspend
-final.suspend:
-  %suspend2 = call i8 @llvm.coro.suspend(token %save2, i1 false)
-  switch i8 %suspend2, label %exit [
-    i8 0, label %pre.exit
-    i8 1, label %exit
-  ]
-pre.exit:
-  br label %exit
-exit:
-  call i1 @llvm.coro.end(ptr null, i1 false, token none)
-  ret void
-unreach:
-  unreachable
-}
-
-; Verify that in the initial function resume is not marked with musttail.
-; CHECK-LABEL: @f(
-; CHECK: %[[addr1:.+]] = call ptr @llvm.coro.subfn.addr(ptr null, i8 0)
-; CHECK-NOT: musttail call fastcc void %[[addr1]](ptr null)
-
-; Verify that in the resume part resume call is marked with musttail.
-; CHECK-LABEL: @f.resume(
-; CHECK: %[[hdl:.+]] = call ptr @g()
-; CHECK-NEXT: %[[addr2:.+]] = call ptr @llvm.coro.subfn.addr(ptr %[[hdl]], i8 0)
-; CHECK: musttail call fastcc void %[[addr2]](ptr %[[hdl]])
-; CHECK-NEXT: ret void
-; CHECK: %[[hdl2:.+]] = call ptr @h()
-; CHECK-NEXT: %[[addr3:.+]] = call ptr @llvm.coro.subfn.addr(ptr %[[hdl2]], i8 0)
-; CHECK: musttail call fastcc void %[[addr3]](ptr %[[hdl2]])
-; CHECK-NEXT: ret void
-; CHECK: %[[addr4:.+]] = call ptr @llvm.coro.subfn.addr(ptr null, i8 0)
-; CHECK: musttail call fastcc void %[[addr4]](ptr null)
-; CHECK-NEXT: ret void
-
-
-
-declare token @llvm.coro.id(i32, ptr readnone, ptr nocapture readonly, ptr) #1
-declare i1 @llvm.coro.alloc(token) #2
-declare i64 @llvm.coro.size.i64() #3
-declare ptr @llvm.coro.begin(token, ptr writeonly) #2
-declare token @llvm.coro.save(ptr) #2
-declare ptr @llvm.coro.frame() #3
-declare i8 @llvm.coro.suspend(token, i1) #2
-declare ptr @llvm.coro.free(token, ptr nocapture readonly) #1
-declare i1 @llvm.coro.end(ptr, i1, token) #2
-declare ptr @llvm.coro.subfn.addr(ptr nocapture readonly, i8) #1
-declare ptr @malloc(i64)
-declare i8 @switch_result()
-declare ptr @g()
-declare ptr @h()
-
-attributes #0 = { presplitcoroutine }
-attributes #1 = { argmemonly nounwind readonly }
-attributes #2 = { nounwind }
-attributes #3 = { nounwind readnone }
diff --git a/llvm/test/Instrumentation/InstrProfiling/Coro/coro-split-musttail10.ll b/llvm/test/Instrumentation/InstrProfiling/Coro/coro-split-musttail10.ll
deleted file mode 100644
index f43b10ebf42e..000000000000
--- a/llvm/test/Instrumentation/InstrProfiling/Coro/coro-split-musttail10.ll
+++ /dev/null
@@ -1,55 +0,0 @@
-; Tests that instrumentation doesn't interfere with lowering (coro-split).
-; It should convert coro.resume followed by a suspend to a musttail call.
-
-; RUN: opt < %s -passes='pgo-instr-gen,cgscc(coro-split),simplifycfg,early-cse' -S | FileCheck %s
-
-target triple = "wasm64-unknown-unknown"
-
-define void @f() #0 {
-entry:
-  %id = call token @llvm.coro.id(i32 0, ptr null, ptr null, ptr null)
-  %alloc = call ptr @malloc(i64 16) #3
-  %vFrame = call noalias nonnull ptr @llvm.coro.begin(token %id, ptr %alloc)
-
-  %save = call token @llvm.coro.save(ptr null)
-  %addr1 = call ptr @llvm.coro.subfn.addr(ptr null, i8 0)
-  call fastcc void %addr1(ptr null)
-
-  %suspend = call i8 @llvm.coro.suspend(token %save, i1 false)
-  switch i8 %suspend, label %exit [
-    i8 0, label %await.ready
-    i8 1, label %exit
-  ]
-await.ready:
-  %save2 = call token @llvm.coro.save(ptr null)
-  %addr2 = call ptr @llvm.coro.subfn.addr(ptr null, i8 0)
-  call fastcc void %addr2(ptr null)
-
-  %suspend2 = call i8 @llvm.coro.suspend(token %save2, i1 false)
-  switch i8 %suspend2, label %exit [
-    i8 0, label %exit
-    i8 1, label %exit
-  ]
-exit:
-  call i1 @llvm.coro.end(ptr null, i1 false, token none)
-  ret void
-}
-
-; CHECK: musttail call
-
-declare token @llvm.coro.id(i32, ptr readnone, ptr nocapture readonly, ptr) #1
-declare i1 @llvm.coro.alloc(token) #2
-declare i64 @llvm.coro.size.i64() #3
-declare ptr @llvm.coro.begin(token, ptr writeonly) #2
-declare token @llvm.coro.save(ptr) #2
-declare ptr @llvm.coro.frame() #3
-declare i8 @llvm.coro.suspend(token, i1) #2
-declare ptr @llvm.coro.free(token, ptr nocapture readonly) #1
-declare i1 @llvm.coro.end(ptr, i1, token) #2
-declare ptr @llvm.coro.subfn.addr(ptr nocapture readonly, i8) #1
-declare ptr @malloc(i64)
-
-attributes #0 = { presplitcoroutine "target-features"="+tail-call" }
-attributes #1 = { argmemonly nounwind readonly }
-attributes #2 = { nounwind }
-attributes #3 = { nounwind readnone }
diff --git a/llvm/test/Instrumentation/InstrProfiling/Coro/coro-split-musttail11.ll b/llvm/test/Instrumentation/InstrProfiling/Coro/coro-split-musttail11.ll
deleted file mode 100644
index fc5bb9a1b20b..000000000000
--- a/llvm/test/Instrumentation/InstrProfiling/Coro/coro-split-musttail11.ll
+++ /dev/null
@@ -1,55 +0,0 @@
-; Tests that instrumentation doesn't interfere with lowering (coro-split).
-; It should convert coro.resume followed by a suspend to a musttail call.
-
-; RUN: opt < %s -passes='pgo-instr-gen,cgscc(coro-split),simplifycfg,early-cse' -S | FileCheck %s
-
-target triple = "wasm32-unknown-unknown"
-
-define void @f() #0 {
-entry:
-  %id = call token @llvm.coro.id(i32 0, ptr null, ptr null, ptr null)
-  %alloc = call ptr @malloc(i64 16) #3
-  %vFrame = call noalias nonnull ptr @llvm.coro.begin(token %id, ptr %alloc)
-
-  %save = call token @llvm.coro.save(ptr null)
-  %addr1 = call ptr @llvm.coro.subfn.addr(ptr null, i8 0)
-  call fastcc void %addr1(ptr null)
-
-  %suspend = call i8 @llvm.coro.suspend(token %save, i1 false)
-  switch i8 %suspend, label %exit [
-    i8 0, label %await.ready
-    i8 1, label %exit
-  ]
-await.ready:
-  %save2 = call token @llvm.coro.save(ptr null)
-  %addr2 = call ptr @llvm.coro.subfn.addr(ptr null, i8 0)
-  call fastcc void %addr2(ptr null)
-
-  %suspend2 = call i8 @llvm.coro.suspend(token %save2, i1 false)
-  switch i8 %suspend2, label %exit [
-    i8 0, label %exit
-    i8 1, label %exit
-  ]
-exit:
-  call i1 @llvm.coro.end(ptr null, i1 false, token none)
-  ret void
-}
-
-; CHECK: musttail call
-
-declare token @llvm.coro.id(i32, ptr readnone, ptr nocapture readonly, ptr) #1
-declare i1 @llvm.coro.alloc(token) #2
-declare i64 @llvm.coro.size.i64() #3
-declare ptr @llvm.coro.begin(token, ptr writeonly) #2
-declare token @llvm.coro.save(ptr) #2
-declare ptr @llvm.coro.frame() #3
-declare i8 @llvm.coro.suspend(token, i1) #2
-declare ptr @llvm.coro.free(token, ptr nocapture readonly) #1
-declare i1 @llvm.coro.end(ptr, i1, token) #2
-declare ptr @llvm.coro.subfn.addr(ptr nocapture readonly, i8) #1
-declare ptr @malloc(i64)
-
-attributes #0 = { presplitcoroutine "target-features"="+tail-call" }
-attributes #1 = { argmemonly nounwind readonly }
-attributes #2 = { nounwind }
-attributes #3 = { nounwind readnone }
diff --git a/llvm/test/Instrumentation/InstrProfiling/Coro/coro-split-musttail12.ll b/llvm/test/Instrumentation/InstrProfiling/Coro/coro-split-musttail12.ll
deleted file mode 100644
index 634d0106a2e6..000000000000
--- a/llvm/test/Instrumentation/InstrProfiling/Coro/coro-split-musttail12.ll
+++ /dev/null
@@ -1,85 +0,0 @@
-; Tests that instrumentation doesn't interfere with lowering (coro-split).
-; It should convert coro.resume followed by a suspend to a musttail call.
-
-; RUN: opt < %s -passes='pgo-instr-gen,cgscc(coro-split),simplifycfg,early-cse' -S | FileCheck %s
-
-declare void @fakeresume1(ptr)
-declare void @print()
-
-define void @f(i1 %cond) #0 {
-entry:
-  %id = call token @llvm.coro.id(i32 0, ptr null, ptr null, ptr null)
-  %alloc = call ptr @malloc(i64 16) #3
-  %vFrame = call noalias nonnull ptr @llvm.coro.begin(token %id, ptr %alloc)
-
-  %save = call token @llvm.coro.save(ptr null)
-
-  %init_suspend = call i8 @llvm.coro.suspend(token %save, i1 false)
-  switch i8 %init_suspend, label %coro.end [
-    i8 0, label %await.ready
-    i8 1, label %coro.end
-  ]
-await.ready:
-  %save2 = call token @llvm.coro.save(ptr null)
-  br i1 %cond, label %then, label %else
-
-then:
-  call fastcc void @fakeresume1(ptr align 8 null)
-  br label %merge
-
-else:
-  br label %merge
-
-merge:
-  %v0 = phi i1 [0, %then], [1, %else]
-  br label %compare
-
-compare:
-  %cond.cmp = icmp eq i1 %v0, 0
-  br i1 %cond.cmp, label %ready, label %prepare
-
-prepare:
-  call void @print()
-  br label %ready
-
-ready:
-  %suspend = call i8 @llvm.coro.suspend(token %save2, i1 true)
-  %switch = icmp ult i8 %suspend, 2
-  br i1 %switch, label %cleanup, label %coro.end
-
-cleanup:
-  %free.handle = call ptr @llvm.coro.free(token %id, ptr %vFrame)
-  %.not = icmp eq ptr %free.handle, null
-  br i1 %.not, label %coro.end, label %coro.free
-
-coro.free:
-  call void @delete(ptr nonnull %free.handle) #2
-  br label %coro.end
-
-coro.end:
-  call i1 @llvm.coro.end(ptr null, i1 false, token none)
-  ret void
-}
-
-; CHECK-LABEL: @f.resume(
-; CHECK-NOT:      }
-; CHECK:          call void @print()
-
-
-declare token @llvm.coro.id(i32, ptr readnone, ptr nocapture readonly, ptr) #1
-declare i1 @llvm.coro.alloc(token) #2
-declare i64 @llvm.coro.size.i64() #3
-declare ptr @llvm.coro.begin(token, ptr writeonly) #2
-declare token @llvm.coro.save(ptr) #2
-declare ptr @llvm.coro.frame() #3
-declare i8 @llvm.coro.suspend(token, i1) #2
-declare ptr @llvm.coro.free(token, ptr nocapture readonly) #1
-declare i1 @llvm.coro.end(ptr, i1, token) #2
-declare ptr @llvm.coro.subfn.addr(ptr nocapture readonly, i8) #1
-declare ptr @malloc(i64)
-declare void @delete(ptr nonnull) #2
-
-attributes #0 = { presplitcoroutine }
-attributes #1 = { argmemonly nounwind readonly }
-attributes #2 = { nounwind }
-attributes #3 = { nounwind readnone }
diff --git a/llvm/test/Instrumentation/InstrProfiling/Coro/coro-split-musttail13.ll b/llvm/test/Instrumentation/InstrProfiling/Coro/coro-split-musttail13.ll
deleted file mode 100644
index 2f9a14c90107..000000000000
--- a/llvm/test/Instrumentation/InstrProfiling/Coro/coro-split-musttail13.ll
+++ /dev/null
@@ -1,76 +0,0 @@
-; Tests that instrumentation doesn't interfere with lowering (coro-split).
-; It should convert coro.resume followed by a suspend to a musttail call.
-
-; RUN: opt < %s -passes='pgo-instr-gen,cgscc(coro-split),simplifycfg,early-cse' -S | FileCheck %s
-
-declare void @fakeresume1(ptr)
-declare void @may_throw(ptr)
-declare void @print()
-
-define void @f(i1 %cond) #0 personality i32 3 {
-entry:
-  %id = call token @llvm.coro.id(i32 0, ptr null, ptr null, ptr null)
-  %alloc = call ptr @malloc(i64 16) #3
-  %vFrame = call noalias nonnull ptr @llvm.coro.begin(token %id, ptr %alloc)
-
-  %save = call token @llvm.coro.save(ptr null)
-
-  %init_suspend = call i8 @llvm.coro.suspend(token %save, i1 false)
-  switch i8 %init_suspend, label %coro.end [
-    i8 0, label %await.ready
-    i8 1, label %coro.end
-  ]
-await.ready:
-  call fastcc void @fakeresume1(ptr align 8 null)
-  invoke void @may_throw(ptr null)
-    to label %ready unwind label %lpad
-
-ready:
-  %save2 = call token @llvm.coro.save(ptr null)
-  %suspend = call i8 @llvm.coro.suspend(token %save2, i1 true)
-  %switch = icmp ult i8 %suspend, 2
-  br i1 %switch, label %cleanup, label %coro.end
-
-cleanup:
-  %free.handle = call ptr @llvm.coro.free(token %id, ptr %vFrame)
-  %.not = icmp eq ptr %free.handle, null
-  br i1 %.not, label %coro.end, label %coro.free
-
-lpad:
-  %lpval = landingpad { ptr, i32 }
-     cleanup
-
-  %need.resume = call i1 @llvm.coro.end(ptr null, i1 true, token none)
-  resume { ptr, i32 } %lpval
-
-coro.free:
-  call void @delete(ptr nonnull %free.handle) #2
-  br label %coro.end
-
-coro.end:
-  call i1 @llvm.coro.end(ptr null, i1 false, token none)
-  ret void
-}
-
-; CHECK-LABEL: @f.resume(
-; CHECK-NOT:          musttail call fastcc void @fakeresume1(
-; CHECK:     }
-
-
-declare token @llvm.coro.id(i32, ptr readnone, ptr nocapture readonly, ptr) #1
-declare i1 @llvm.coro.alloc(token) #2
-declare i64 @llvm.coro.size.i64() #3
-declare ptr @llvm.coro.begin(token, ptr writeonly) #2
-declare token @llvm.coro.save(ptr) #2
-declare ptr @llvm.coro.frame() #3
-declare i8 @llvm.coro.suspend(token, i1) #2
-declare ptr @llvm.coro.free(token, ptr nocapture readonly) #1
-declare i1 @llvm.coro.end(ptr, i1, token) #2
-declare ptr @llvm.coro.subfn.addr(ptr nocapture readonly, i8) #1
-declare ptr @malloc(i64)
-declare void @delete(ptr nonnull) #2
-
-attributes #0 = { presplitcoroutine }
-attributes #1 = { argmemonly nounwind readonly }
-attributes #2 = { nounwind }
-attributes #3 = { nounwind readnone }
diff --git a/llvm/test/Instrumentation/InstrProfiling/Coro/coro-split-musttail2.ll b/llvm/test/Instrumentation/InstrProfiling/Coro/coro-split-musttail2.ll
deleted file mode 100644
index 61b61a200e70..000000000000
--- a/llvm/test/Instrumentation/InstrProfiling/Coro/coro-split-musttail2.ll
+++ /dev/null
@@ -1,68 +0,0 @@
-; Tests that instrumentation doesn't interfere with lowering (coro-split).
-; It should convert coro.resume followed by a suspend to a musttail call.
-
-; RUN: opt < %s -passes='pgo-instr-gen,cgscc(coro-split),simplifycfg,early-cse' -S | FileCheck %s
-
-define void @fakeresume1(ptr)  {
-entry:
-  ret void;
-}
-
-define void @fakeresume2(ptr align 8)  {
-entry:
-  ret void;
-}
-
-define void @g() #0 {
-entry:
-  %id = call token @llvm.coro.id(i32 0, ptr null, ptr null, ptr null)
-  %alloc = call ptr @malloc(i64 16) #3
-  %vFrame = call noalias nonnull ptr @llvm.coro.begin(token %id, ptr %alloc)
-
-  %save = call token @llvm.coro.save(ptr null)
-  call fastcc void @fakeresume1(ptr null)
-
-  %suspend = call i8 @llvm.coro.suspend(token %save, i1 false)
-  switch i8 %suspend, label %exit [
-    i8 0, label %await.ready
-    i8 1, label %exit
-  ]
-await.ready:
-  %save2 = call token @llvm.coro.save(ptr null)
-  call fastcc void @fakeresume2(ptr align 8 null)
-
-  %suspend2 = call i8 @llvm.coro.suspend(token %save2, i1 false)
-  switch i8 %suspend2, label %exit [
-    i8 0, label %exit
-    i8 1, label %exit
-  ]
-exit:
-  call i1 @llvm.coro.end(ptr null, i1 false, token none)
-  ret void
-}
-
-; Verify that in the initial function resume is not marked with musttail.
-; CHECK-LABEL: @g(
-; CHECK-NOT: musttail call fastcc void @fakeresume1(ptr null)
-
-; Verify that in the resume part resume call is marked with musttail.
-; CHECK-LABEL: @g.resume(
-; CHECK: musttail call fastcc void @fakeresume2(ptr align 8 null)
-; CHECK-NEXT: ret void
-
-declare token @llvm.coro.id(i32, ptr readnone, ptr nocapture readonly, ptr) #1
-declare i1 @llvm.coro.alloc(token) #2
-declare i64 @llvm.coro.size.i64() #3
-declare ptr @llvm.coro.begin(token, ptr writeonly) #2
-declare token @llvm.coro.save(ptr) #2
-declare ptr @llvm.coro.frame() #3
-declare i8 @llvm.coro.suspend(token, i1) #2
-declare ptr @llvm.coro.free(token, ptr nocapture readonly) #1
-declare i1 @llvm.coro.end(ptr, i1, token) #2
-declare ptr @llvm.coro.subfn.addr(ptr nocapture readonly, i8) #1
-declare ptr @malloc(i64)
-
-attributes #0 = { presplitcoroutine }
-attributes #1 = { argmemonly nounwind readonly }
-attributes #2 = { nounwind }
-attributes #3 = { nounwind readnone }
diff --git a/llvm/test/Instrumentation/InstrProfiling/Coro/coro-split-musttail3.ll b/llvm/test/Instrumentation/InstrProfiling/Coro/coro-split-musttail3.ll
deleted file mode 100644
index 82176b8085e6..000000000000
--- a/llvm/test/Instrumentation/InstrProfiling/Coro/coro-split-musttail3.ll
+++ /dev/null
@@ -1,91 +0,0 @@
-; Tests that instrumentation doesn't interfere with lowering (coro-split).
-; It should convert coro.resume followed by a suspend to a musttail call.
-
-; RUN: opt < %s -passes='pgo-instr-gen,cgscc(coro-split),simplifycfg,early-cse' -S | FileCheck %s
-
-define void @f() #0 {
-entry:
-  %id = call token @llvm.coro.id(i32 0, ptr null, ptr null, ptr null)
-  %alloc = call ptr @malloc(i64 16) #3
-  %vFrame = call noalias nonnull ptr @llvm.coro.begin(token %id, ptr %alloc)
-
-  %save = call token @llvm.coro.save(ptr null)
-  %addr1 = call ptr @llvm.coro.subfn.addr(ptr null, i8 0)
-  call fastcc void %addr1(ptr null)
-
-  %suspend = call i8 @llvm.coro.suspend(token %save, i1 false)
-  %cmp = icmp eq i8 %suspend, 0
-  br i1 %cmp, label %await.suspend, label %exit
-await.suspend:
-  %save2 = call token @llvm.coro.save(ptr null)
-  %br0 = call i8 @switch_result()
-  switch i8 %br0, label %unreach [
-    i8 0, label %await.resume3
-    i8 1, label %await.resume1
-    i8 2, label %await.resume2
-  ]
-await.resume1:
-  %hdl = call ptr @g()
-  %addr2 = call ptr @llvm.coro.subfn.addr(ptr %hdl, i8 0)
-  call fastcc void %addr2(ptr %hdl)
-  br label %final.suspend
-await.resume2:
-  %hdl2 = call ptr @h()
-  %addr3 = call ptr @llvm.coro.subfn.addr(ptr %hdl2, i8 0)
-  call fastcc void %addr3(ptr %hdl2)
-  br label %final.suspend
-await.resume3:
-  %addr4 = call ptr @llvm.coro.subfn.addr(ptr null, i8 0)
-  call fastcc void %addr4(ptr null)
-  br label %final.suspend
-final.suspend:
-  %suspend2 = call i8 @llvm.coro.suspend(token %save2, i1 false)
-  %cmp2 = icmp eq i8 %suspend2, 0
-  br i1 %cmp2, label %pre.exit, label %exit
-pre.exit:
-  br label %exit
-exit:
-  call i1 @llvm.coro.end(ptr null, i1 false, token none)
-  ret void
-unreach:
-  unreachable
-}
-
-; Verify that in the initial function resume is not marked with musttail.
-; CHECK-LABEL: @f(
-; CHECK: %[[addr1:.+]] = call ptr @llvm.coro.subfn.addr(ptr null, i8 0)
-; CHECK-NOT: musttail call fastcc void %[[addr1]](ptr null)
-
-; Verify that in the resume part resume call is marked with musttail.
-; CHECK-LABEL: @f.resume(
-; CHECK: %[[hdl:.+]] = call ptr @g()
-; CHECK-NEXT: %[[addr2:.+]] = call ptr @llvm.coro.subfn.addr(ptr %[[hdl]], i8 0)
-; CHECK: musttail call fastcc void %[[addr2]](ptr %[[hdl]])
-; CHECK-NEXT: ret void
-; CHECK: %[[hdl2:.+]] = call ptr @h()
-; CHECK-NEXT: %[[addr3:.+]] = call ptr @llvm.coro.subfn.addr(ptr %[[hdl2]], i8 0)
-; CHECK: musttail call fastcc void %[[addr3]](ptr %[[hdl2]])
-; CHECK-NEXT: ret void
-; CHECK: %[[addr4:.+]] = call ptr @llvm.coro.subfn.addr(ptr null, i8 0)
-; CHECK: musttail call fastcc void %[[addr4]](ptr null)
-; CHECK-NEXT: ret void
-
-declare token @llvm.coro.id(i32, ptr readnone, ptr nocapture readonly, ptr) #1
-declare i1 @llvm.coro.alloc(token) #2
-declare i64 @llvm.coro.size.i64() #3
-declare ptr @llvm.coro.begin(token, ptr writeonly) #2
-declare token @llvm.coro.save(ptr) #2
-declare ptr @llvm.coro.frame() #3
-declare i8 @llvm.coro.suspend(token, i1) #2
-declare ptr @llvm.coro.free(token, ptr nocapture readonly) #1
-declare i1 @llvm.coro.end(ptr, i1, token) #2
-declare ptr @llvm.coro.subfn.addr(ptr nocapture readonly, i8) #1
-declare ptr @malloc(i64)
-declare i8 @switch_result()
-declare ptr @g()
-declare ptr @h()
-
-attributes #0 = { presplitcoroutine }
-attributes #1 = { argmemonly nounwind readonly }
-attributes #2 = { nounwind }
-attributes #3 = { nounwind readnone }
diff --git a/llvm/test/Instrumentation/InstrProfiling/Coro/coro-split-musttail4.ll b/llvm/test/Instrumentation/InstrProfiling/Coro/coro-split-musttail4.ll
deleted file mode 100644
index be70fc4b51f1..000000000000
--- a/llvm/test/Instrumentation/InstrProfiling/Coro/coro-split-musttail4.ll
+++ /dev/null
@@ -1,66 +0,0 @@
-; Tests that instrumentation doesn't interfere with lowering (coro-split).
-; It should convert coro.resume followed by a suspend to a musttail call.
-
-; RUN: opt < %s -passes='pgo-instr-gen,cgscc(coro-split),simplifycfg,early-cse' -S | FileCheck %s
-
-define void @fakeresume1(ptr)  {
-entry:
-  ret void;
-}
-
-define void @f() #0 {
-entry:
-  %id = call token @llvm.coro.id(i32 0, ptr null, ptr null, ptr null)
-  %alloc = call ptr @malloc(i64 16) #3
-  %vFrame = call noalias nonnull ptr @llvm.coro.begin(token %id, ptr %alloc)
-
-  %save = call token @llvm.coro.save(ptr null)
-
-  %init_suspend = call i8 @llvm.coro.suspend(token %save, i1 false)
-  switch i8 %init_suspend, label %coro.end [
-    i8 0, label %await.ready
-    i8 1, label %coro.end
-  ]
-await.ready:
-  %save2 = call token @llvm.coro.save(ptr null)
-
-  call fastcc void @fakeresume1(ptr align 8 null)
-  %suspend = call i8 @llvm.coro.suspend(token %save2, i1 true)
-  %switch = icmp ult i8 %suspend, 2
-  br i1 %switch, label %cleanup, label %coro.end
-
-cleanup:
-  %free.handle = call ptr @llvm.coro.free(token %id, ptr %vFrame)
-  %.not = icmp eq ptr %free.handle, null
-  br i1 %.not, label %coro.end, label %coro.free
-
-coro.free:
-  call void @delete(ptr nonnull %free.handle) #2
-  br label %coro.end
-
-coro.end:
-  call i1 @llvm.coro.end(ptr null, i1 false, token none)
-  ret void
-}
-
-; CHECK-LABEL: @f.resume(
-; CHECK:          musttail call fastcc void @fakeresume1(
-; CHECK-NEXT:     ret void
-
-declare token @llvm.coro.id(i32, ptr readnone, ptr nocapture readonly, ptr) #1
-declare i1 @llvm.coro.alloc(token) #2
-declare i64 @llvm.coro.size.i64() #3
-declare ptr @llvm.coro.begin(token, ptr writeonly) #2
-declare token @llvm.coro.save(ptr) #2
-declare ptr @llvm.coro.frame() #3
-declare i8 @llvm.coro.suspend(token, i1) #2
-declare ptr @llvm.coro.free(token, ptr nocapture readonly) #1
-declare i1 @llvm.coro.end(ptr, i1, token) #2
-declare ptr @llvm.coro.subfn.addr(ptr nocapture readonly, i8) #1
-declare ptr @malloc(i64)
-declare void @delete(ptr nonnull) #2
-
-attributes #0 = { presplitcoroutine }
-attributes #1 = { argmemonly nounwind readonly }
-attributes #2 = { nounwind }
-attributes #3 = { nounwind readnone }
diff --git a/llvm/test/Instrumentation/InstrProfiling/Coro/coro-split-musttail5.ll b/llvm/test/Instrumentation/InstrProfiling/Coro/coro-split-musttail5.ll
deleted file mode 100644
index 3e5bddd8e131..000000000000
--- a/llvm/test/Instrumentation/InstrProfiling/Coro/coro-split-musttail5.ll
+++ /dev/null
@@ -1,63 +0,0 @@
-; Tests that instrumentation doesn't interfere with lowering (coro-split).
-; It should convert coro.resume followed by a suspend to a musttail call.
-
-; RUN: opt < %s -passes='pgo-instr-gen,cgscc(coro-split),simplifycfg,early-cse' -S | FileCheck %s
-
-declare void @fakeresume1(ptr align 8)
-
-define void @g() #0 {
-entry:
-  %id = call token @llvm.coro.id(i32 0, ptr null, ptr null, ptr null)
-  %alloc = call ptr @malloc(i64 16) #3
-  %alloc.var = alloca i8
-  call void @llvm.lifetime.start.p0(i64 1, ptr %alloc.var)
-  %vFrame = call noalias nonnull ptr @llvm.coro.begin(token %id, ptr %alloc)
-
-  %save = call token @llvm.coro.save(ptr null)
-  %suspend = call i8 @llvm.coro.suspend(token %save, i1 false)
-
-  switch i8 %suspend, label %exit [
-    i8 0, label %await.suspend
-    i8 1, label %exit
-  ]
-await.suspend:
-  %save2 = call token @llvm.coro.save(ptr null)
-  call fastcc void @fakeresume1(ptr align 8 null)
-  %suspend2 = call i8 @llvm.coro.suspend(token %save2, i1 false)
-  switch i8 %suspend2, label %exit [
-    i8 0, label %await.ready
-    i8 1, label %exit
-  ]
-await.ready:
-  call void @consume(ptr %alloc.var)
-  call void @llvm.lifetime.end.p0(i64 1, ptr %alloc.var)
-  br label %exit
-exit:
-  call i1 @llvm.coro.end(ptr null, i1 false, token none)
-  ret void
-}
-
-; Verify that in the resume part resume call is marked with musttail.
-; CHECK-LABEL: @g.resume(
-; CHECK:          musttail call fastcc void @fakeresume1(ptr align 8 null)
-; CHECK-NEXT:     ret void
-
-declare token @llvm.coro.id(i32, ptr readnone, ptr nocapture readonly, ptr) #1
-declare i1 @llvm.coro.alloc(token) #2
-declare i64 @llvm.coro.size.i64() #3
-declare ptr @llvm.coro.begin(token, ptr writeonly) #2
-declare token @llvm.coro.save(ptr) #2
-declare ptr @llvm.coro.frame() #3
-declare i8 @llvm.coro.suspend(token, i1) #2
-declare ptr @llvm.coro.free(token, ptr nocapture readonly) #1
-declare i1 @llvm.coro.end(ptr, i1, token) #2
-declare ptr @llvm.coro.subfn.addr(ptr nocapture readonly, i8) #1
-declare ptr @malloc(i64)
-declare void @consume(ptr)
-declare void @llvm.lifetime.start.p0(i64, ptr nocapture)
-declare void @llvm.lifetime.end.p0(i64, ptr nocapture)
-
-attributes #0 = { presplitcoroutine }
-attributes #1 = { argmemonly nounwind readonly }
-attributes #2 = { nounwind }
-attributes #3 = { nounwind readnone }
diff --git a/llvm/test/Instrumentation/InstrProfiling/Coro/coro-split-musttail6.ll b/llvm/test/Instrumentation/InstrProfiling/Coro/coro-split-musttail6.ll
deleted file mode 100644
index 4359d5305d4d..000000000000
--- a/llvm/test/Instrumentation/InstrProfiling/Coro/coro-split-musttail6.ll
+++ /dev/null
@@ -1,112 +0,0 @@
-; Tests that instrumentation doesn't interfere with lowering (coro-split).
-; It should convert coro.resume followed by a suspend to a musttail call.
-
-; RUN: opt < %s -passes='pgo-instr-gen,cgscc(coro-split),simplifycfg,early-cse' -S | FileCheck %s
-
-declare void @fakeresume1(ptr align 8)
-
-define void @g() #0 {
-entry:
-  %id = call token @llvm.coro.id(i32 0, ptr null, ptr null, ptr null)
-  %alloc = call ptr @malloc(i64 16) #3
-  %alloc.var = alloca i64
-  call void @llvm.lifetime.start.p0(i64 1, ptr %alloc.var)
-  %vFrame = call noalias nonnull ptr @llvm.coro.begin(token %id, ptr %alloc)
-
-  %save = call token @llvm.coro.save(ptr null)
-  %suspend = call i8 @llvm.coro.suspend(token %save, i1 false)
-
-  switch i8 %suspend, label %exit [
-    i8 0, label %await.suspend
-    i8 1, label %exit
-  ]
-await.suspend:
-  %save2 = call token @llvm.coro.save(ptr null)
-  call fastcc void @fakeresume1(ptr align 8 null)
-  %suspend2 = call i8 @llvm.coro.suspend(token %save2, i1 false)
-  switch i8 %suspend2, label %exit [
-    i8 0, label %await.ready
-    i8 1, label %exit
-  ]
-await.ready:
-  call void @consume(ptr %alloc.var)
-  call void @llvm.lifetime.end.p0(i64 1, ptr %alloc.var)
-  br label %exit
-exit:
-  call i1 @llvm.coro.end(ptr null, i1 false, token none)
-  ret void
-}
-
-; Verify that in the resume part resume call is marked with musttail.
-; CHECK-LABEL: @g.resume(
-; CHECK:      musttail call fastcc void @fakeresume1(ptr align 8 null)
-; CHECK-NEXT: ret void
-
-; It has a cleanup bb.
-define void @f() #0 {
-entry:
-  %id = call token @llvm.coro.id(i32 0, ptr null, ptr null, ptr null)
-  %alloc = call ptr @malloc(i64 16) #3
-  %alloc.var = alloca i64
-  call void @llvm.lifetime.start.p0(i64 1, ptr %alloc.var)
-  %vFrame = call noalias nonnull ptr @llvm.coro.begin(token %id, ptr %alloc)
-
-  %save = call token @llvm.coro.save(ptr null)
-  %suspend = call i8 @llvm.coro.suspend(token %save, i1 false)
-
-  switch i8 %suspend, label %exit [
-    i8 0, label %await.suspend
-    i8 1, label %exit
-  ]
-await.suspend:
-  %save2 = call token @llvm.coro.save(ptr null)
-  call fastcc void @fakeresume1(ptr align 8 null)
-  %suspend2 = call i8 @llvm.coro.suspend(token %save2, i1 false)
-  switch i8 %suspend2, label %exit [
-    i8 0, label %await.ready
-    i8 1, label %cleanup
-  ]
-await.ready:
-  call void @consume(ptr %alloc.var)
-  call void @llvm.lifetime.end.p0(i64 1, ptr %alloc.var)
-  br label %exit
-
-cleanup:
-  %free.handle = call ptr @llvm.coro.free(token %id, ptr %vFrame)
-  %.not = icmp eq ptr %free.handle, null
-  br i1 %.not, label %exit, label %coro.free
-
-coro.free:
-  call void @delete(ptr nonnull %free.handle) #2
-  br label %exit
-
-exit:
-  call i1 @llvm.coro.end(ptr null, i1 false, token none)
-  ret void
-}
-
-; Verify that in the resume part resume call is marked with musttail.
-; CHECK-LABEL: @f.resume(
-; CHECK:      musttail call fastcc void @fakeresume1(ptr align 8 null)
-; CHECK-NEXT: ret void
-
-declare token @llvm.coro.id(i32, ptr readnone, ptr nocapture readonly, ptr) #1
-declare i1 @llvm.coro.alloc(token) #2
-declare i64 @llvm.coro.size.i64() #3
-declare ptr @llvm.coro.begin(token, ptr writeonly) #2
-declare token @llvm.coro.save(ptr) #2
-declare ptr @llvm.coro.frame() #3
-declare i8 @llvm.coro.suspend(token, i1) #2
-declare ptr @llvm.coro.free(token, ptr nocapture readonly) #1
-declare i1 @llvm.coro.end(ptr, i1, token) #2
-declare ptr @llvm.coro.subfn.addr(ptr nocapture readonly, i8) #1
-declare ptr @malloc(i64)
-declare void @delete(ptr nonnull) #2
-declare void @consume(ptr)
-declare void @llvm.lifetime.start.p0(i64, ptr nocapture)
-declare void @llvm.lifetime.end.p0(i64, ptr nocapture)
-
-attributes #0 = { presplitcoroutine }
-attributes #1 = { argmemonly nounwind readonly }
-attributes #2 = { nounwind }
-attributes #3 = { nounwind readnone }
diff --git a/llvm/test/Instrumentation/InstrProfiling/Coro/coro-split-musttail7.ll b/llvm/test/Instrumentation/InstrProfiling/Coro/coro-split-musttail7.ll
deleted file mode 100644
index 2a14be0f9218..000000000000
--- a/llvm/test/Instrumentation/InstrProfiling/Coro/coro-split-musttail7.ll
+++ /dev/null
@@ -1,115 +0,0 @@
-; Tests that instrumentation doesn't interfere with lowering (coro-split).
-; It should convert coro.resume followed by a suspend to a musttail call.
-
-; The difference between this and coro-split-musttail5.ll and coro-split-musttail5.ll
-; is that this contains dead instruction generated during the transformation,
-; which makes the optimization harder.
-; RUN: opt < %s -passes='pgo-instr-gen,cgscc(coro-split),simplifycfg,early-cse' -S | FileCheck %s
-
-declare void @fakeresume1(ptr align 8)
-
-define void @g() #0 {
-entry:
-  %id = call token @llvm.coro.id(i32 0, ptr null, ptr null, ptr null)
-  %alloc = call ptr @malloc(i64 16) #3
-  %alloc.var = alloca i64
-  call void @llvm.lifetime.start.p0(i64 1, ptr %alloc.var)
-  %vFrame = call noalias nonnull ptr @llvm.coro.begin(token %id, ptr %alloc)
-
-  %save = call token @llvm.coro.save(ptr null)
-  %suspend = call i8 @llvm.coro.suspend(token %save, i1 false)
-
-  switch i8 %suspend, label %exit [
-    i8 0, label %await.suspend
-    i8 1, label %exit
-  ]
-await.suspend:
-  %save2 = call token @llvm.coro.save(ptr null)
-  call fastcc void @fakeresume1(ptr align 8 null)
-  %suspend2 = call i8 @llvm.coro.suspend(token %save2, i1 false)
-  switch i8 %suspend2, label %exit [
-    i8 0, label %await.ready
-    i8 1, label %exit
-  ]
-await.ready:
-  call void @consume(ptr %alloc.var)
-  call void @llvm.lifetime.end.p0(i64 1, ptr %alloc.var)
-  br label %exit
-exit:
-  call i1 @llvm.coro.end(ptr null, i1 false, token none)
-  ret void
-}
-
-; Verify that in the resume part resume call is marked with musttail.
-; CHECK-LABEL: @g.resume(
-; CHECK:         musttail call fastcc void @fakeresume1(ptr align 8 null)
-; CHECK-NEXT:    ret void
-
-; It has a cleanup bb.
-define void @f() #0 {
-entry:
-  %id = call token @llvm.coro.id(i32 0, ptr null, ptr null, ptr null)
-  %alloc = call ptr @malloc(i64 16) #3
-  %alloc.var = alloca i64
-  call void @llvm.lifetime.start.p0(i64 1, ptr %alloc.var)
-  %vFrame = call noalias nonnull ptr @llvm.coro.begin(token %id, ptr %alloc)
-
-  %save = call token @llvm.coro.save(ptr null)
-  %suspend = call i8 @llvm.coro.suspend(token %save, i1 false)
-
-  switch i8 %suspend, label %exit [
-    i8 0, label %await.suspend
-    i8 1, label %exit
-  ]
-await.suspend:
-  %save2 = call token @llvm.coro.save(ptr null)
-  call fastcc void @fakeresume1(ptr align 8 null)
-  %suspend2 = call i8 @llvm.coro.suspend(token %save2, i1 false)
-  switch i8 %suspend2, label %exit [
-    i8 0, label %await.ready
-    i8 1, label %cleanup
-  ]
-await.ready:
-  call void @consume(ptr %alloc.var)
-  call void @llvm.lifetime.end.p0(i64 1, ptr %alloc.var)
-  br label %exit
-
-cleanup:
-  %free.handle = call ptr @llvm.coro.free(token %id, ptr %vFrame)
-  %.not = icmp eq ptr %free.handle, null
-  br i1 %.not, label %exit, label %coro.free
-
-coro.free:
-  call void @delete(ptr nonnull %free.handle) #2
-  br label %exit
-
-exit:
-  call i1 @llvm.coro.end(ptr null, i1 false, token none)
-  ret void
-}
-
-; Verify that in the resume part resume call is marked with musttail.
-; CHECK-LABEL: @f.resume(
-; CHECK:         musttail call fastcc void @fakeresume1(ptr align 8 null)
-; CHECK-NEXT:    ret void
-
-declare token @llvm.coro.id(i32, ptr readnone, ptr nocapture readonly, ptr) #1
-declare i1 @llvm.coro.alloc(token) #2
-declare i64 @llvm.coro.size.i64() #3
-declare ptr @llvm.coro.begin(token, ptr writeonly) #2
-declare token @llvm.coro.save(ptr) #2
-declare ptr @llvm.coro.frame() #3
-declare i8 @llvm.coro.suspend(token, i1) #2
-declare ptr @llvm.coro.free(token, ptr nocapture readonly) #1
-declare i1 @llvm.coro.end(ptr, i1, token) #2
-declare ptr @llvm.coro.subfn.addr(ptr nocapture readonly, i8) #1
-declare ptr @malloc(i64)
-declare void @delete(ptr nonnull) #2
-declare void @consume(ptr)
-declare void @llvm.lifetime.start.p0(i64, ptr nocapture)
-declare void @llvm.lifetime.end.p0(i64, ptr nocapture)
-
-attributes #0 = { presplitcoroutine }
-attributes #1 = { argmemonly nounwind readonly }
-attributes #2 = { nounwind }
-attributes #3 = { nounwind readnone }
diff --git a/llvm/test/Instrumentation/MemorySanitizer/X86/mmx-intrinsics.ll b/llvm/test/Instrumentation/MemorySanitizer/X86/mmx-intrinsics.ll
new file mode 100644
index 000000000000..39b2b6225d8b
--- /dev/null
+++ b/llvm/test/Instrumentation/MemorySanitizer/X86/mmx-intrinsics.ll
@@ -0,0 +1,3617 @@
+; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --version 4
+; RUN: opt %s -S -passes=msan 2>&1 | FileCheck %s
+
+target datalayout = "e-m:o-p270:32:32-p271:32:32-p272:64:64-i64:64-f80:128-n8:16:32:64-S128"
+target triple = "x86_64-unknown-linux-gnu"
+
+declare x86_mmx @llvm.x86.ssse3.phadd.w(x86_mmx, x86_mmx) nounwind readnone
+
+define i64 @test1(<1 x i64> %a, <1 x i64> %b) #0 {
+; CHECK-LABEL: define i64 @test1(
+; CHECK-SAME: <1 x i64> [[A:%.*]], <1 x i64> [[B:%.*]]) #[[ATTR1:[0-9]+]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP9:%.*]] = load <1 x i64>, ptr inttoptr (i64 add (i64 ptrtoint (ptr @__msan_param_tls to i64), i64 8) to ptr), align 8
+; CHECK-NEXT:    [[TMP10:%.*]] = load <1 x i64>, ptr @__msan_param_tls, align 8
+; CHECK-NEXT:    call void @llvm.donothing()
+; CHECK-NEXT:    [[TMP12:%.*]] = bitcast <1 x i64> [[TMP9]] to <4 x i16>
+; CHECK-NEXT:    [[TMP0:%.*]] = bitcast <1 x i64> [[B]] to <4 x i16>
+; CHECK-NEXT:    [[TMP13:%.*]] = bitcast <1 x i64> [[TMP10]] to <4 x i16>
+; CHECK-NEXT:    [[TMP1:%.*]] = bitcast <1 x i64> [[A]] to <4 x i16>
+; CHECK-NEXT:    [[TMP16:%.*]] = bitcast <4 x i16> [[TMP13]] to i64
+; CHECK-NEXT:    [[TMP2:%.*]] = bitcast <4 x i16> [[TMP1]] to x86_mmx
+; CHECK-NEXT:    [[TMP8:%.*]] = bitcast <4 x i16> [[TMP12]] to i64
+; CHECK-NEXT:    [[TMP3:%.*]] = bitcast <4 x i16> [[TMP0]] to x86_mmx
+; CHECK-NEXT:    [[_MSPROP:%.*]] = or i64 [[TMP16]], [[TMP8]]
+; CHECK-NEXT:    [[TMP4:%.*]] = tail call x86_mmx @llvm.x86.ssse3.phadd.w(x86_mmx [[TMP2]], x86_mmx [[TMP3]]) #[[ATTR5:[0-9]+]]
+; CHECK-NEXT:    [[TMP11:%.*]] = bitcast i64 [[_MSPROP]] to <4 x i16>
+; CHECK-NEXT:    [[TMP5:%.*]] = bitcast x86_mmx [[TMP4]] to <4 x i16>
+; CHECK-NEXT:    [[TMP6:%.*]] = bitcast <4 x i16> [[TMP11]] to <1 x i64>
+; CHECK-NEXT:    [[TMP14:%.*]] = bitcast <4 x i16> [[TMP5]] to <1 x i64>
+; CHECK-NEXT:    [[TMP7:%.*]] = extractelement <1 x i64> [[TMP6]], i32 0
+; CHECK-NEXT:    [[TMP15:%.*]] = extractelement <1 x i64> [[TMP14]], i32 0
+; CHECK-NEXT:    store i64 [[TMP7]], ptr @__msan_retval_tls, align 8
+; CHECK-NEXT:    ret i64 [[TMP15]]
+;
+entry:
+  %0 = bitcast <1 x i64> %b to <4 x i16>
+  %1 = bitcast <1 x i64> %a to <4 x i16>
+  %2 = bitcast <4 x i16> %1 to x86_mmx
+  %3 = bitcast <4 x i16> %0 to x86_mmx
+  %4 = tail call x86_mmx @llvm.x86.ssse3.phadd.w(x86_mmx %2, x86_mmx %3) nounwind readnone
+  %5 = bitcast x86_mmx %4 to <4 x i16>
+  %6 = bitcast <4 x i16> %5 to <1 x i64>
+  %7 = extractelement <1 x i64> %6, i32 0
+  ret i64 %7
+}
+
+declare x86_mmx @llvm.x86.mmx.pcmpgt.d(x86_mmx, x86_mmx) nounwind readnone
+
+define i64 @test88(<1 x i64> %a, <1 x i64> %b) #0 {
+; CHECK-LABEL: define i64 @test88(
+; CHECK-SAME: <1 x i64> [[A:%.*]], <1 x i64> [[B:%.*]]) #[[ATTR1]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP8:%.*]] = load <1 x i64>, ptr inttoptr (i64 add (i64 ptrtoint (ptr @__msan_param_tls to i64), i64 8) to ptr), align 8
+; CHECK-NEXT:    [[TMP10:%.*]] = load <1 x i64>, ptr @__msan_param_tls, align 8
+; CHECK-NEXT:    call void @llvm.donothing()
+; CHECK-NEXT:    [[TMP11:%.*]] = bitcast <1 x i64> [[TMP8]] to <2 x i32>
+; CHECK-NEXT:    [[TMP0:%.*]] = bitcast <1 x i64> [[B]] to <2 x i32>
+; CHECK-NEXT:    [[TMP14:%.*]] = bitcast <1 x i64> [[TMP10]] to <2 x i32>
+; CHECK-NEXT:    [[TMP1:%.*]] = bitcast <1 x i64> [[A]] to <2 x i32>
+; CHECK-NEXT:    [[TMP6:%.*]] = bitcast <2 x i32> [[TMP14]] to i64
+; CHECK-NEXT:    [[MMX_VAR_I:%.*]] = bitcast <2 x i32> [[TMP1]] to x86_mmx
+; CHECK-NEXT:    [[TMP7:%.*]] = bitcast <2 x i32> [[TMP11]] to i64
+; CHECK-NEXT:    [[MMX_VAR1_I:%.*]] = bitcast <2 x i32> [[TMP0]] to x86_mmx
+; CHECK-NEXT:    [[_MSPROP:%.*]] = or i64 [[TMP6]], [[TMP7]]
+; CHECK-NEXT:    [[TMP2:%.*]] = tail call x86_mmx @llvm.x86.mmx.pcmpgt.d(x86_mmx [[MMX_VAR_I]], x86_mmx [[MMX_VAR1_I]]) #[[ATTR2:[0-9]+]]
+; CHECK-NEXT:    [[TMP9:%.*]] = bitcast i64 [[_MSPROP]] to <2 x i32>
+; CHECK-NEXT:    [[TMP3:%.*]] = bitcast x86_mmx [[TMP2]] to <2 x i32>
+; CHECK-NEXT:    [[TMP4:%.*]] = bitcast <2 x i32> [[TMP9]] to <1 x i64>
+; CHECK-NEXT:    [[TMP12:%.*]] = bitcast <2 x i32> [[TMP3]] to <1 x i64>
+; CHECK-NEXT:    [[TMP5:%.*]] = extractelement <1 x i64> [[TMP4]], i32 0
+; CHECK-NEXT:    [[TMP13:%.*]] = extractelement <1 x i64> [[TMP12]], i32 0
+; CHECK-NEXT:    store i64 [[TMP5]], ptr @__msan_retval_tls, align 8
+; CHECK-NEXT:    ret i64 [[TMP13]]
+;
+entry:
+  %0 = bitcast <1 x i64> %b to <2 x i32>
+  %1 = bitcast <1 x i64> %a to <2 x i32>
+  %mmx_var.i = bitcast <2 x i32> %1 to x86_mmx
+  %mmx_var1.i = bitcast <2 x i32> %0 to x86_mmx
+  %2 = tail call x86_mmx @llvm.x86.mmx.pcmpgt.d(x86_mmx %mmx_var.i, x86_mmx %mmx_var1.i) nounwind
+  %3 = bitcast x86_mmx %2 to <2 x i32>
+  %4 = bitcast <2 x i32> %3 to <1 x i64>
+  %5 = extractelement <1 x i64> %4, i32 0
+  ret i64 %5
+}
+
+declare x86_mmx @llvm.x86.mmx.pcmpgt.w(x86_mmx, x86_mmx) nounwind readnone
+
+define i64 @test87(<1 x i64> %a, <1 x i64> %b) #0 {
+; CHECK-LABEL: define i64 @test87(
+; CHECK-SAME: <1 x i64> [[A:%.*]], <1 x i64> [[B:%.*]]) #[[ATTR1]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP8:%.*]] = load <1 x i64>, ptr inttoptr (i64 add (i64 ptrtoint (ptr @__msan_param_tls to i64), i64 8) to ptr), align 8
+; CHECK-NEXT:    [[TMP10:%.*]] = load <1 x i64>, ptr @__msan_param_tls, align 8
+; CHECK-NEXT:    call void @llvm.donothing()
+; CHECK-NEXT:    [[TMP11:%.*]] = bitcast <1 x i64> [[TMP8]] to <4 x i16>
+; CHECK-NEXT:    [[TMP0:%.*]] = bitcast <1 x i64> [[B]] to <4 x i16>
+; CHECK-NEXT:    [[TMP14:%.*]] = bitcast <1 x i64> [[TMP10]] to <4 x i16>
+; CHECK-NEXT:    [[TMP1:%.*]] = bitcast <1 x i64> [[A]] to <4 x i16>
+; CHECK-NEXT:    [[TMP6:%.*]] = bitcast <4 x i16> [[TMP14]] to i64
+; CHECK-NEXT:    [[MMX_VAR_I:%.*]] = bitcast <4 x i16> [[TMP1]] to x86_mmx
+; CHECK-NEXT:    [[TMP7:%.*]] = bitcast <4 x i16> [[TMP11]] to i64
+; CHECK-NEXT:    [[MMX_VAR1_I:%.*]] = bitcast <4 x i16> [[TMP0]] to x86_mmx
+; CHECK-NEXT:    [[_MSPROP:%.*]] = or i64 [[TMP6]], [[TMP7]]
+; CHECK-NEXT:    [[TMP2:%.*]] = tail call x86_mmx @llvm.x86.mmx.pcmpgt.w(x86_mmx [[MMX_VAR_I]], x86_mmx [[MMX_VAR1_I]]) #[[ATTR2]]
+; CHECK-NEXT:    [[TMP9:%.*]] = bitcast i64 [[_MSPROP]] to <4 x i16>
+; CHECK-NEXT:    [[TMP3:%.*]] = bitcast x86_mmx [[TMP2]] to <4 x i16>
+; CHECK-NEXT:    [[TMP4:%.*]] = bitcast <4 x i16> [[TMP9]] to <1 x i64>
+; CHECK-NEXT:    [[TMP12:%.*]] = bitcast <4 x i16> [[TMP3]] to <1 x i64>
+; CHECK-NEXT:    [[TMP5:%.*]] = extractelement <1 x i64> [[TMP4]], i32 0
+; CHECK-NEXT:    [[TMP13:%.*]] = extractelement <1 x i64> [[TMP12]], i32 0
+; CHECK-NEXT:    store i64 [[TMP5]], ptr @__msan_retval_tls, align 8
+; CHECK-NEXT:    ret i64 [[TMP13]]
+;
+entry:
+  %0 = bitcast <1 x i64> %b to <4 x i16>
+  %1 = bitcast <1 x i64> %a to <4 x i16>
+  %mmx_var.i = bitcast <4 x i16> %1 to x86_mmx
+  %mmx_var1.i = bitcast <4 x i16> %0 to x86_mmx
+  %2 = tail call x86_mmx @llvm.x86.mmx.pcmpgt.w(x86_mmx %mmx_var.i, x86_mmx %mmx_var1.i) nounwind
+  %3 = bitcast x86_mmx %2 to <4 x i16>
+  %4 = bitcast <4 x i16> %3 to <1 x i64>
+  %5 = extractelement <1 x i64> %4, i32 0
+  ret i64 %5
+}
+
+declare x86_mmx @llvm.x86.mmx.pcmpgt.b(x86_mmx, x86_mmx) nounwind readnone
+
+define i64 @test86(<1 x i64> %a, <1 x i64> %b) #0 {
+; CHECK-LABEL: define i64 @test86(
+; CHECK-SAME: <1 x i64> [[A:%.*]], <1 x i64> [[B:%.*]]) #[[ATTR1]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP8:%.*]] = load <1 x i64>, ptr inttoptr (i64 add (i64 ptrtoint (ptr @__msan_param_tls to i64), i64 8) to ptr), align 8
+; CHECK-NEXT:    [[TMP10:%.*]] = load <1 x i64>, ptr @__msan_param_tls, align 8
+; CHECK-NEXT:    call void @llvm.donothing()
+; CHECK-NEXT:    [[TMP11:%.*]] = bitcast <1 x i64> [[TMP8]] to <8 x i8>
+; CHECK-NEXT:    [[TMP0:%.*]] = bitcast <1 x i64> [[B]] to <8 x i8>
+; CHECK-NEXT:    [[TMP14:%.*]] = bitcast <1 x i64> [[TMP10]] to <8 x i8>
+; CHECK-NEXT:    [[TMP1:%.*]] = bitcast <1 x i64> [[A]] to <8 x i8>
+; CHECK-NEXT:    [[TMP6:%.*]] = bitcast <8 x i8> [[TMP14]] to i64
+; CHECK-NEXT:    [[MMX_VAR_I:%.*]] = bitcast <8 x i8> [[TMP1]] to x86_mmx
+; CHECK-NEXT:    [[TMP7:%.*]] = bitcast <8 x i8> [[TMP11]] to i64
+; CHECK-NEXT:    [[MMX_VAR1_I:%.*]] = bitcast <8 x i8> [[TMP0]] to x86_mmx
+; CHECK-NEXT:    [[_MSPROP:%.*]] = or i64 [[TMP6]], [[TMP7]]
+; CHECK-NEXT:    [[TMP2:%.*]] = tail call x86_mmx @llvm.x86.mmx.pcmpgt.b(x86_mmx [[MMX_VAR_I]], x86_mmx [[MMX_VAR1_I]]) #[[ATTR2]]
+; CHECK-NEXT:    [[TMP9:%.*]] = bitcast i64 [[_MSPROP]] to <8 x i8>
+; CHECK-NEXT:    [[TMP3:%.*]] = bitcast x86_mmx [[TMP2]] to <8 x i8>
+; CHECK-NEXT:    [[TMP4:%.*]] = bitcast <8 x i8> [[TMP9]] to <1 x i64>
+; CHECK-NEXT:    [[TMP12:%.*]] = bitcast <8 x i8> [[TMP3]] to <1 x i64>
+; CHECK-NEXT:    [[TMP5:%.*]] = extractelement <1 x i64> [[TMP4]], i32 0
+; CHECK-NEXT:    [[TMP13:%.*]] = extractelement <1 x i64> [[TMP12]], i32 0
+; CHECK-NEXT:    store i64 [[TMP5]], ptr @__msan_retval_tls, align 8
+; CHECK-NEXT:    ret i64 [[TMP13]]
+;
+entry:
+  %0 = bitcast <1 x i64> %b to <8 x i8>
+  %1 = bitcast <1 x i64> %a to <8 x i8>
+  %mmx_var.i = bitcast <8 x i8> %1 to x86_mmx
+  %mmx_var1.i = bitcast <8 x i8> %0 to x86_mmx
+  %2 = tail call x86_mmx @llvm.x86.mmx.pcmpgt.b(x86_mmx %mmx_var.i, x86_mmx %mmx_var1.i) nounwind
+  %3 = bitcast x86_mmx %2 to <8 x i8>
+  %4 = bitcast <8 x i8> %3 to <1 x i64>
+  %5 = extractelement <1 x i64> %4, i32 0
+  ret i64 %5
+}
+
+declare x86_mmx @llvm.x86.mmx.pcmpeq.d(x86_mmx, x86_mmx) nounwind readnone
+
+define i64 @test85(<1 x i64> %a, <1 x i64> %b) #0 {
+; CHECK-LABEL: define i64 @test85(
+; CHECK-SAME: <1 x i64> [[A:%.*]], <1 x i64> [[B:%.*]]) #[[ATTR1]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP8:%.*]] = load <1 x i64>, ptr inttoptr (i64 add (i64 ptrtoint (ptr @__msan_param_tls to i64), i64 8) to ptr), align 8
+; CHECK-NEXT:    [[TMP10:%.*]] = load <1 x i64>, ptr @__msan_param_tls, align 8
+; CHECK-NEXT:    call void @llvm.donothing()
+; CHECK-NEXT:    [[TMP11:%.*]] = bitcast <1 x i64> [[TMP8]] to <2 x i32>
+; CHECK-NEXT:    [[TMP0:%.*]] = bitcast <1 x i64> [[B]] to <2 x i32>
+; CHECK-NEXT:    [[TMP14:%.*]] = bitcast <1 x i64> [[TMP10]] to <2 x i32>
+; CHECK-NEXT:    [[TMP1:%.*]] = bitcast <1 x i64> [[A]] to <2 x i32>
+; CHECK-NEXT:    [[TMP6:%.*]] = bitcast <2 x i32> [[TMP14]] to i64
+; CHECK-NEXT:    [[MMX_VAR_I:%.*]] = bitcast <2 x i32> [[TMP1]] to x86_mmx
+; CHECK-NEXT:    [[TMP7:%.*]] = bitcast <2 x i32> [[TMP11]] to i64
+; CHECK-NEXT:    [[MMX_VAR1_I:%.*]] = bitcast <2 x i32> [[TMP0]] to x86_mmx
+; CHECK-NEXT:    [[_MSPROP:%.*]] = or i64 [[TMP6]], [[TMP7]]
+; CHECK-NEXT:    [[TMP2:%.*]] = tail call x86_mmx @llvm.x86.mmx.pcmpeq.d(x86_mmx [[MMX_VAR_I]], x86_mmx [[MMX_VAR1_I]]) #[[ATTR2]]
+; CHECK-NEXT:    [[TMP9:%.*]] = bitcast i64 [[_MSPROP]] to <2 x i32>
+; CHECK-NEXT:    [[TMP3:%.*]] = bitcast x86_mmx [[TMP2]] to <2 x i32>
+; CHECK-NEXT:    [[TMP4:%.*]] = bitcast <2 x i32> [[TMP9]] to <1 x i64>
+; CHECK-NEXT:    [[TMP12:%.*]] = bitcast <2 x i32> [[TMP3]] to <1 x i64>
+; CHECK-NEXT:    [[TMP5:%.*]] = extractelement <1 x i64> [[TMP4]], i32 0
+; CHECK-NEXT:    [[TMP13:%.*]] = extractelement <1 x i64> [[TMP12]], i32 0
+; CHECK-NEXT:    store i64 [[TMP5]], ptr @__msan_retval_tls, align 8
+; CHECK-NEXT:    ret i64 [[TMP13]]
+;
+entry:
+  %0 = bitcast <1 x i64> %b to <2 x i32>
+  %1 = bitcast <1 x i64> %a to <2 x i32>
+  %mmx_var.i = bitcast <2 x i32> %1 to x86_mmx
+  %mmx_var1.i = bitcast <2 x i32> %0 to x86_mmx
+  %2 = tail call x86_mmx @llvm.x86.mmx.pcmpeq.d(x86_mmx %mmx_var.i, x86_mmx %mmx_var1.i) nounwind
+  %3 = bitcast x86_mmx %2 to <2 x i32>
+  %4 = bitcast <2 x i32> %3 to <1 x i64>
+  %5 = extractelement <1 x i64> %4, i32 0
+  ret i64 %5
+}
+
+declare x86_mmx @llvm.x86.mmx.pcmpeq.w(x86_mmx, x86_mmx) nounwind readnone
+
+define i64 @test84(<1 x i64> %a, <1 x i64> %b) #0 {
+; CHECK-LABEL: define i64 @test84(
+; CHECK-SAME: <1 x i64> [[A:%.*]], <1 x i64> [[B:%.*]]) #[[ATTR1]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP8:%.*]] = load <1 x i64>, ptr inttoptr (i64 add (i64 ptrtoint (ptr @__msan_param_tls to i64), i64 8) to ptr), align 8
+; CHECK-NEXT:    [[TMP10:%.*]] = load <1 x i64>, ptr @__msan_param_tls, align 8
+; CHECK-NEXT:    call void @llvm.donothing()
+; CHECK-NEXT:    [[TMP11:%.*]] = bitcast <1 x i64> [[TMP8]] to <4 x i16>
+; CHECK-NEXT:    [[TMP0:%.*]] = bitcast <1 x i64> [[B]] to <4 x i16>
+; CHECK-NEXT:    [[TMP14:%.*]] = bitcast <1 x i64> [[TMP10]] to <4 x i16>
+; CHECK-NEXT:    [[TMP1:%.*]] = bitcast <1 x i64> [[A]] to <4 x i16>
+; CHECK-NEXT:    [[TMP6:%.*]] = bitcast <4 x i16> [[TMP14]] to i64
+; CHECK-NEXT:    [[MMX_VAR_I:%.*]] = bitcast <4 x i16> [[TMP1]] to x86_mmx
+; CHECK-NEXT:    [[TMP7:%.*]] = bitcast <4 x i16> [[TMP11]] to i64
+; CHECK-NEXT:    [[MMX_VAR1_I:%.*]] = bitcast <4 x i16> [[TMP0]] to x86_mmx
+; CHECK-NEXT:    [[_MSPROP:%.*]] = or i64 [[TMP6]], [[TMP7]]
+; CHECK-NEXT:    [[TMP2:%.*]] = tail call x86_mmx @llvm.x86.mmx.pcmpeq.w(x86_mmx [[MMX_VAR_I]], x86_mmx [[MMX_VAR1_I]]) #[[ATTR2]]
+; CHECK-NEXT:    [[TMP9:%.*]] = bitcast i64 [[_MSPROP]] to <4 x i16>
+; CHECK-NEXT:    [[TMP3:%.*]] = bitcast x86_mmx [[TMP2]] to <4 x i16>
+; CHECK-NEXT:    [[TMP4:%.*]] = bitcast <4 x i16> [[TMP9]] to <1 x i64>
+; CHECK-NEXT:    [[TMP12:%.*]] = bitcast <4 x i16> [[TMP3]] to <1 x i64>
+; CHECK-NEXT:    [[TMP5:%.*]] = extractelement <1 x i64> [[TMP4]], i32 0
+; CHECK-NEXT:    [[TMP13:%.*]] = extractelement <1 x i64> [[TMP12]], i32 0
+; CHECK-NEXT:    store i64 [[TMP5]], ptr @__msan_retval_tls, align 8
+; CHECK-NEXT:    ret i64 [[TMP13]]
+;
+entry:
+  %0 = bitcast <1 x i64> %b to <4 x i16>
+  %1 = bitcast <1 x i64> %a to <4 x i16>
+  %mmx_var.i = bitcast <4 x i16> %1 to x86_mmx
+  %mmx_var1.i = bitcast <4 x i16> %0 to x86_mmx
+  %2 = tail call x86_mmx @llvm.x86.mmx.pcmpeq.w(x86_mmx %mmx_var.i, x86_mmx %mmx_var1.i) nounwind
+  %3 = bitcast x86_mmx %2 to <4 x i16>
+  %4 = bitcast <4 x i16> %3 to <1 x i64>
+  %5 = extractelement <1 x i64> %4, i32 0
+  ret i64 %5
+}
+
+declare x86_mmx @llvm.x86.mmx.pcmpeq.b(x86_mmx, x86_mmx) nounwind readnone
+
+define i64 @test83(<1 x i64> %a, <1 x i64> %b) #0 {
+; CHECK-LABEL: define i64 @test83(
+; CHECK-SAME: <1 x i64> [[A:%.*]], <1 x i64> [[B:%.*]]) #[[ATTR1]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP8:%.*]] = load <1 x i64>, ptr inttoptr (i64 add (i64 ptrtoint (ptr @__msan_param_tls to i64), i64 8) to ptr), align 8
+; CHECK-NEXT:    [[TMP10:%.*]] = load <1 x i64>, ptr @__msan_param_tls, align 8
+; CHECK-NEXT:    call void @llvm.donothing()
+; CHECK-NEXT:    [[TMP11:%.*]] = bitcast <1 x i64> [[TMP8]] to <8 x i8>
+; CHECK-NEXT:    [[TMP0:%.*]] = bitcast <1 x i64> [[B]] to <8 x i8>
+; CHECK-NEXT:    [[TMP14:%.*]] = bitcast <1 x i64> [[TMP10]] to <8 x i8>
+; CHECK-NEXT:    [[TMP1:%.*]] = bitcast <1 x i64> [[A]] to <8 x i8>
+; CHECK-NEXT:    [[TMP6:%.*]] = bitcast <8 x i8> [[TMP14]] to i64
+; CHECK-NEXT:    [[MMX_VAR_I:%.*]] = bitcast <8 x i8> [[TMP1]] to x86_mmx
+; CHECK-NEXT:    [[TMP7:%.*]] = bitcast <8 x i8> [[TMP11]] to i64
+; CHECK-NEXT:    [[MMX_VAR1_I:%.*]] = bitcast <8 x i8> [[TMP0]] to x86_mmx
+; CHECK-NEXT:    [[_MSPROP:%.*]] = or i64 [[TMP6]], [[TMP7]]
+; CHECK-NEXT:    [[TMP2:%.*]] = tail call x86_mmx @llvm.x86.mmx.pcmpeq.b(x86_mmx [[MMX_VAR_I]], x86_mmx [[MMX_VAR1_I]]) #[[ATTR2]]
+; CHECK-NEXT:    [[TMP9:%.*]] = bitcast i64 [[_MSPROP]] to <8 x i8>
+; CHECK-NEXT:    [[TMP3:%.*]] = bitcast x86_mmx [[TMP2]] to <8 x i8>
+; CHECK-NEXT:    [[TMP4:%.*]] = bitcast <8 x i8> [[TMP9]] to <1 x i64>
+; CHECK-NEXT:    [[TMP12:%.*]] = bitcast <8 x i8> [[TMP3]] to <1 x i64>
+; CHECK-NEXT:    [[TMP5:%.*]] = extractelement <1 x i64> [[TMP4]], i32 0
+; CHECK-NEXT:    [[TMP13:%.*]] = extractelement <1 x i64> [[TMP12]], i32 0
+; CHECK-NEXT:    store i64 [[TMP5]], ptr @__msan_retval_tls, align 8
+; CHECK-NEXT:    ret i64 [[TMP13]]
+;
+entry:
+  %0 = bitcast <1 x i64> %b to <8 x i8>
+  %1 = bitcast <1 x i64> %a to <8 x i8>
+  %mmx_var.i = bitcast <8 x i8> %1 to x86_mmx
+  %mmx_var1.i = bitcast <8 x i8> %0 to x86_mmx
+  %2 = tail call x86_mmx @llvm.x86.mmx.pcmpeq.b(x86_mmx %mmx_var.i, x86_mmx %mmx_var1.i) nounwind
+  %3 = bitcast x86_mmx %2 to <8 x i8>
+  %4 = bitcast <8 x i8> %3 to <1 x i64>
+  %5 = extractelement <1 x i64> %4, i32 0
+  ret i64 %5
+}
+
+declare x86_mmx @llvm.x86.mmx.punpckldq(x86_mmx, x86_mmx) nounwind readnone
+
+define i64 @test82(<1 x i64> %a, <1 x i64> %b) #0 {
+; CHECK-LABEL: define i64 @test82(
+; CHECK-SAME: <1 x i64> [[A:%.*]], <1 x i64> [[B:%.*]]) #[[ATTR1]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP8:%.*]] = load <1 x i64>, ptr inttoptr (i64 add (i64 ptrtoint (ptr @__msan_param_tls to i64), i64 8) to ptr), align 8
+; CHECK-NEXT:    [[TMP10:%.*]] = load <1 x i64>, ptr @__msan_param_tls, align 8
+; CHECK-NEXT:    call void @llvm.donothing()
+; CHECK-NEXT:    [[TMP11:%.*]] = bitcast <1 x i64> [[TMP8]] to <2 x i32>
+; CHECK-NEXT:    [[TMP0:%.*]] = bitcast <1 x i64> [[B]] to <2 x i32>
+; CHECK-NEXT:    [[TMP14:%.*]] = bitcast <1 x i64> [[TMP10]] to <2 x i32>
+; CHECK-NEXT:    [[TMP1:%.*]] = bitcast <1 x i64> [[A]] to <2 x i32>
+; CHECK-NEXT:    [[TMP6:%.*]] = bitcast <2 x i32> [[TMP14]] to i64
+; CHECK-NEXT:    [[MMX_VAR_I:%.*]] = bitcast <2 x i32> [[TMP1]] to x86_mmx
+; CHECK-NEXT:    [[TMP7:%.*]] = bitcast <2 x i32> [[TMP11]] to i64
+; CHECK-NEXT:    [[MMX_VAR1_I:%.*]] = bitcast <2 x i32> [[TMP0]] to x86_mmx
+; CHECK-NEXT:    [[_MSPROP:%.*]] = or i64 [[TMP6]], [[TMP7]]
+; CHECK-NEXT:    [[TMP2:%.*]] = tail call x86_mmx @llvm.x86.mmx.punpckldq(x86_mmx [[MMX_VAR_I]], x86_mmx [[MMX_VAR1_I]]) #[[ATTR2]]
+; CHECK-NEXT:    [[TMP9:%.*]] = bitcast i64 [[_MSPROP]] to <2 x i32>
+; CHECK-NEXT:    [[TMP3:%.*]] = bitcast x86_mmx [[TMP2]] to <2 x i32>
+; CHECK-NEXT:    [[TMP4:%.*]] = bitcast <2 x i32> [[TMP9]] to <1 x i64>
+; CHECK-NEXT:    [[TMP12:%.*]] = bitcast <2 x i32> [[TMP3]] to <1 x i64>
+; CHECK-NEXT:    [[TMP5:%.*]] = extractelement <1 x i64> [[TMP4]], i32 0
+; CHECK-NEXT:    [[TMP13:%.*]] = extractelement <1 x i64> [[TMP12]], i32 0
+; CHECK-NEXT:    store i64 [[TMP5]], ptr @__msan_retval_tls, align 8
+; CHECK-NEXT:    ret i64 [[TMP13]]
+;
+entry:
+  %0 = bitcast <1 x i64> %b to <2 x i32>
+  %1 = bitcast <1 x i64> %a to <2 x i32>
+  %mmx_var.i = bitcast <2 x i32> %1 to x86_mmx
+  %mmx_var1.i = bitcast <2 x i32> %0 to x86_mmx
+  %2 = tail call x86_mmx @llvm.x86.mmx.punpckldq(x86_mmx %mmx_var.i, x86_mmx %mmx_var1.i) nounwind
+  %3 = bitcast x86_mmx %2 to <2 x i32>
+  %4 = bitcast <2 x i32> %3 to <1 x i64>
+  %5 = extractelement <1 x i64> %4, i32 0
+  ret i64 %5
+}
+
+declare x86_mmx @llvm.x86.mmx.punpcklwd(x86_mmx, x86_mmx) nounwind readnone
+
+define i64 @test81(<1 x i64> %a, <1 x i64> %b) #0 {
+; CHECK-LABEL: define i64 @test81(
+; CHECK-SAME: <1 x i64> [[A:%.*]], <1 x i64> [[B:%.*]]) #[[ATTR1]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP8:%.*]] = load <1 x i64>, ptr inttoptr (i64 add (i64 ptrtoint (ptr @__msan_param_tls to i64), i64 8) to ptr), align 8
+; CHECK-NEXT:    [[TMP10:%.*]] = load <1 x i64>, ptr @__msan_param_tls, align 8
+; CHECK-NEXT:    call void @llvm.donothing()
+; CHECK-NEXT:    [[TMP11:%.*]] = bitcast <1 x i64> [[TMP8]] to <4 x i16>
+; CHECK-NEXT:    [[TMP0:%.*]] = bitcast <1 x i64> [[B]] to <4 x i16>
+; CHECK-NEXT:    [[TMP14:%.*]] = bitcast <1 x i64> [[TMP10]] to <4 x i16>
+; CHECK-NEXT:    [[TMP1:%.*]] = bitcast <1 x i64> [[A]] to <4 x i16>
+; CHECK-NEXT:    [[TMP6:%.*]] = bitcast <4 x i16> [[TMP14]] to i64
+; CHECK-NEXT:    [[MMX_VAR_I:%.*]] = bitcast <4 x i16> [[TMP1]] to x86_mmx
+; CHECK-NEXT:    [[TMP7:%.*]] = bitcast <4 x i16> [[TMP11]] to i64
+; CHECK-NEXT:    [[MMX_VAR1_I:%.*]] = bitcast <4 x i16> [[TMP0]] to x86_mmx
+; CHECK-NEXT:    [[_MSPROP:%.*]] = or i64 [[TMP6]], [[TMP7]]
+; CHECK-NEXT:    [[TMP2:%.*]] = tail call x86_mmx @llvm.x86.mmx.punpcklwd(x86_mmx [[MMX_VAR_I]], x86_mmx [[MMX_VAR1_I]]) #[[ATTR2]]
+; CHECK-NEXT:    [[TMP9:%.*]] = bitcast i64 [[_MSPROP]] to <4 x i16>
+; CHECK-NEXT:    [[TMP3:%.*]] = bitcast x86_mmx [[TMP2]] to <4 x i16>
+; CHECK-NEXT:    [[TMP4:%.*]] = bitcast <4 x i16> [[TMP9]] to <1 x i64>
+; CHECK-NEXT:    [[TMP12:%.*]] = bitcast <4 x i16> [[TMP3]] to <1 x i64>
+; CHECK-NEXT:    [[TMP5:%.*]] = extractelement <1 x i64> [[TMP4]], i32 0
+; CHECK-NEXT:    [[TMP13:%.*]] = extractelement <1 x i64> [[TMP12]], i32 0
+; CHECK-NEXT:    store i64 [[TMP5]], ptr @__msan_retval_tls, align 8
+; CHECK-NEXT:    ret i64 [[TMP13]]
+;
+entry:
+  %0 = bitcast <1 x i64> %b to <4 x i16>
+  %1 = bitcast <1 x i64> %a to <4 x i16>
+  %mmx_var.i = bitcast <4 x i16> %1 to x86_mmx
+  %mmx_var1.i = bitcast <4 x i16> %0 to x86_mmx
+  %2 = tail call x86_mmx @llvm.x86.mmx.punpcklwd(x86_mmx %mmx_var.i, x86_mmx %mmx_var1.i) nounwind
+  %3 = bitcast x86_mmx %2 to <4 x i16>
+  %4 = bitcast <4 x i16> %3 to <1 x i64>
+  %5 = extractelement <1 x i64> %4, i32 0
+  ret i64 %5
+}
+
+declare x86_mmx @llvm.x86.mmx.punpcklbw(x86_mmx, x86_mmx) nounwind readnone
+
+define i64 @test80(<1 x i64> %a, <1 x i64> %b) #0 {
+; CHECK-LABEL: define i64 @test80(
+; CHECK-SAME: <1 x i64> [[A:%.*]], <1 x i64> [[B:%.*]]) #[[ATTR1]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP8:%.*]] = load <1 x i64>, ptr inttoptr (i64 add (i64 ptrtoint (ptr @__msan_param_tls to i64), i64 8) to ptr), align 8
+; CHECK-NEXT:    [[TMP10:%.*]] = load <1 x i64>, ptr @__msan_param_tls, align 8
+; CHECK-NEXT:    call void @llvm.donothing()
+; CHECK-NEXT:    [[TMP11:%.*]] = bitcast <1 x i64> [[TMP8]] to <8 x i8>
+; CHECK-NEXT:    [[TMP0:%.*]] = bitcast <1 x i64> [[B]] to <8 x i8>
+; CHECK-NEXT:    [[TMP14:%.*]] = bitcast <1 x i64> [[TMP10]] to <8 x i8>
+; CHECK-NEXT:    [[TMP1:%.*]] = bitcast <1 x i64> [[A]] to <8 x i8>
+; CHECK-NEXT:    [[TMP6:%.*]] = bitcast <8 x i8> [[TMP14]] to i64
+; CHECK-NEXT:    [[MMX_VAR_I:%.*]] = bitcast <8 x i8> [[TMP1]] to x86_mmx
+; CHECK-NEXT:    [[TMP7:%.*]] = bitcast <8 x i8> [[TMP11]] to i64
+; CHECK-NEXT:    [[MMX_VAR1_I:%.*]] = bitcast <8 x i8> [[TMP0]] to x86_mmx
+; CHECK-NEXT:    [[_MSPROP:%.*]] = or i64 [[TMP6]], [[TMP7]]
+; CHECK-NEXT:    [[TMP2:%.*]] = tail call x86_mmx @llvm.x86.mmx.punpcklbw(x86_mmx [[MMX_VAR_I]], x86_mmx [[MMX_VAR1_I]]) #[[ATTR2]]
+; CHECK-NEXT:    [[TMP9:%.*]] = bitcast i64 [[_MSPROP]] to <8 x i8>
+; CHECK-NEXT:    [[TMP3:%.*]] = bitcast x86_mmx [[TMP2]] to <8 x i8>
+; CHECK-NEXT:    [[TMP4:%.*]] = bitcast <8 x i8> [[TMP9]] to <1 x i64>
+; CHECK-NEXT:    [[TMP12:%.*]] = bitcast <8 x i8> [[TMP3]] to <1 x i64>
+; CHECK-NEXT:    [[TMP5:%.*]] = extractelement <1 x i64> [[TMP4]], i32 0
+; CHECK-NEXT:    [[TMP13:%.*]] = extractelement <1 x i64> [[TMP12]], i32 0
+; CHECK-NEXT:    store i64 [[TMP5]], ptr @__msan_retval_tls, align 8
+; CHECK-NEXT:    ret i64 [[TMP13]]
+;
+entry:
+  %0 = bitcast <1 x i64> %b to <8 x i8>
+  %1 = bitcast <1 x i64> %a to <8 x i8>
+  %mmx_var.i = bitcast <8 x i8> %1 to x86_mmx
+  %mmx_var1.i = bitcast <8 x i8> %0 to x86_mmx
+  %2 = tail call x86_mmx @llvm.x86.mmx.punpcklbw(x86_mmx %mmx_var.i, x86_mmx %mmx_var1.i) nounwind
+  %3 = bitcast x86_mmx %2 to <8 x i8>
+  %4 = bitcast <8 x i8> %3 to <1 x i64>
+  %5 = extractelement <1 x i64> %4, i32 0
+  ret i64 %5
+}
+
+declare x86_mmx @llvm.x86.mmx.punpckhdq(x86_mmx, x86_mmx) nounwind readnone
+
+define i64 @test79(<1 x i64> %a, <1 x i64> %b) #0 {
+; CHECK-LABEL: define i64 @test79(
+; CHECK-SAME: <1 x i64> [[A:%.*]], <1 x i64> [[B:%.*]]) #[[ATTR1]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP8:%.*]] = load <1 x i64>, ptr inttoptr (i64 add (i64 ptrtoint (ptr @__msan_param_tls to i64), i64 8) to ptr), align 8
+; CHECK-NEXT:    [[TMP10:%.*]] = load <1 x i64>, ptr @__msan_param_tls, align 8
+; CHECK-NEXT:    call void @llvm.donothing()
+; CHECK-NEXT:    [[TMP11:%.*]] = bitcast <1 x i64> [[TMP8]] to <2 x i32>
+; CHECK-NEXT:    [[TMP0:%.*]] = bitcast <1 x i64> [[B]] to <2 x i32>
+; CHECK-NEXT:    [[TMP14:%.*]] = bitcast <1 x i64> [[TMP10]] to <2 x i32>
+; CHECK-NEXT:    [[TMP1:%.*]] = bitcast <1 x i64> [[A]] to <2 x i32>
+; CHECK-NEXT:    [[TMP6:%.*]] = bitcast <2 x i32> [[TMP14]] to i64
+; CHECK-NEXT:    [[MMX_VAR_I:%.*]] = bitcast <2 x i32> [[TMP1]] to x86_mmx
+; CHECK-NEXT:    [[TMP7:%.*]] = bitcast <2 x i32> [[TMP11]] to i64
+; CHECK-NEXT:    [[MMX_VAR1_I:%.*]] = bitcast <2 x i32> [[TMP0]] to x86_mmx
+; CHECK-NEXT:    [[_MSPROP:%.*]] = or i64 [[TMP6]], [[TMP7]]
+; CHECK-NEXT:    [[TMP2:%.*]] = tail call x86_mmx @llvm.x86.mmx.punpckhdq(x86_mmx [[MMX_VAR_I]], x86_mmx [[MMX_VAR1_I]]) #[[ATTR2]]
+; CHECK-NEXT:    [[TMP9:%.*]] = bitcast i64 [[_MSPROP]] to <2 x i32>
+; CHECK-NEXT:    [[TMP3:%.*]] = bitcast x86_mmx [[TMP2]] to <2 x i32>
+; CHECK-NEXT:    [[TMP4:%.*]] = bitcast <2 x i32> [[TMP9]] to <1 x i64>
+; CHECK-NEXT:    [[TMP12:%.*]] = bitcast <2 x i32> [[TMP3]] to <1 x i64>
+; CHECK-NEXT:    [[TMP5:%.*]] = extractelement <1 x i64> [[TMP4]], i32 0
+; CHECK-NEXT:    [[TMP13:%.*]] = extractelement <1 x i64> [[TMP12]], i32 0
+; CHECK-NEXT:    store i64 [[TMP5]], ptr @__msan_retval_tls, align 8
+; CHECK-NEXT:    ret i64 [[TMP13]]
+;
+entry:
+  %0 = bitcast <1 x i64> %b to <2 x i32>
+  %1 = bitcast <1 x i64> %a to <2 x i32>
+  %mmx_var.i = bitcast <2 x i32> %1 to x86_mmx
+  %mmx_var1.i = bitcast <2 x i32> %0 to x86_mmx
+  %2 = tail call x86_mmx @llvm.x86.mmx.punpckhdq(x86_mmx %mmx_var.i, x86_mmx %mmx_var1.i) nounwind
+  %3 = bitcast x86_mmx %2 to <2 x i32>
+  %4 = bitcast <2 x i32> %3 to <1 x i64>
+  %5 = extractelement <1 x i64> %4, i32 0
+  ret i64 %5
+}
+
+declare x86_mmx @llvm.x86.mmx.punpckhwd(x86_mmx, x86_mmx) nounwind readnone
+
+define i64 @test78(<1 x i64> %a, <1 x i64> %b) #0 {
+; CHECK-LABEL: define i64 @test78(
+; CHECK-SAME: <1 x i64> [[A:%.*]], <1 x i64> [[B:%.*]]) #[[ATTR1]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP8:%.*]] = load <1 x i64>, ptr inttoptr (i64 add (i64 ptrtoint (ptr @__msan_param_tls to i64), i64 8) to ptr), align 8
+; CHECK-NEXT:    [[TMP10:%.*]] = load <1 x i64>, ptr @__msan_param_tls, align 8
+; CHECK-NEXT:    call void @llvm.donothing()
+; CHECK-NEXT:    [[TMP11:%.*]] = bitcast <1 x i64> [[TMP8]] to <4 x i16>
+; CHECK-NEXT:    [[TMP0:%.*]] = bitcast <1 x i64> [[B]] to <4 x i16>
+; CHECK-NEXT:    [[TMP14:%.*]] = bitcast <1 x i64> [[TMP10]] to <4 x i16>
+; CHECK-NEXT:    [[TMP1:%.*]] = bitcast <1 x i64> [[A]] to <4 x i16>
+; CHECK-NEXT:    [[TMP6:%.*]] = bitcast <4 x i16> [[TMP14]] to i64
+; CHECK-NEXT:    [[MMX_VAR_I:%.*]] = bitcast <4 x i16> [[TMP1]] to x86_mmx
+; CHECK-NEXT:    [[TMP7:%.*]] = bitcast <4 x i16> [[TMP11]] to i64
+; CHECK-NEXT:    [[MMX_VAR1_I:%.*]] = bitcast <4 x i16> [[TMP0]] to x86_mmx
+; CHECK-NEXT:    [[_MSPROP:%.*]] = or i64 [[TMP6]], [[TMP7]]
+; CHECK-NEXT:    [[TMP2:%.*]] = tail call x86_mmx @llvm.x86.mmx.punpckhwd(x86_mmx [[MMX_VAR_I]], x86_mmx [[MMX_VAR1_I]]) #[[ATTR2]]
+; CHECK-NEXT:    [[TMP9:%.*]] = bitcast i64 [[_MSPROP]] to <4 x i16>
+; CHECK-NEXT:    [[TMP3:%.*]] = bitcast x86_mmx [[TMP2]] to <4 x i16>
+; CHECK-NEXT:    [[TMP4:%.*]] = bitcast <4 x i16> [[TMP9]] to <1 x i64>
+; CHECK-NEXT:    [[TMP12:%.*]] = bitcast <4 x i16> [[TMP3]] to <1 x i64>
+; CHECK-NEXT:    [[TMP5:%.*]] = extractelement <1 x i64> [[TMP4]], i32 0
+; CHECK-NEXT:    [[TMP13:%.*]] = extractelement <1 x i64> [[TMP12]], i32 0
+; CHECK-NEXT:    store i64 [[TMP5]], ptr @__msan_retval_tls, align 8
+; CHECK-NEXT:    ret i64 [[TMP13]]
+;
+entry:
+  %0 = bitcast <1 x i64> %b to <4 x i16>
+  %1 = bitcast <1 x i64> %a to <4 x i16>
+  %mmx_var.i = bitcast <4 x i16> %1 to x86_mmx
+  %mmx_var1.i = bitcast <4 x i16> %0 to x86_mmx
+  %2 = tail call x86_mmx @llvm.x86.mmx.punpckhwd(x86_mmx %mmx_var.i, x86_mmx %mmx_var1.i) nounwind
+  %3 = bitcast x86_mmx %2 to <4 x i16>
+  %4 = bitcast <4 x i16> %3 to <1 x i64>
+  %5 = extractelement <1 x i64> %4, i32 0
+  ret i64 %5
+}
+
+declare x86_mmx @llvm.x86.mmx.punpckhbw(x86_mmx, x86_mmx) nounwind readnone
+
+define i64 @test77(<1 x i64> %a, <1 x i64> %b) #0 {
+; CHECK-LABEL: define i64 @test77(
+; CHECK-SAME: <1 x i64> [[A:%.*]], <1 x i64> [[B:%.*]]) #[[ATTR1]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP8:%.*]] = load <1 x i64>, ptr inttoptr (i64 add (i64 ptrtoint (ptr @__msan_param_tls to i64), i64 8) to ptr), align 8
+; CHECK-NEXT:    [[TMP10:%.*]] = load <1 x i64>, ptr @__msan_param_tls, align 8
+; CHECK-NEXT:    call void @llvm.donothing()
+; CHECK-NEXT:    [[TMP11:%.*]] = bitcast <1 x i64> [[TMP8]] to <8 x i8>
+; CHECK-NEXT:    [[TMP0:%.*]] = bitcast <1 x i64> [[B]] to <8 x i8>
+; CHECK-NEXT:    [[TMP14:%.*]] = bitcast <1 x i64> [[TMP10]] to <8 x i8>
+; CHECK-NEXT:    [[TMP1:%.*]] = bitcast <1 x i64> [[A]] to <8 x i8>
+; CHECK-NEXT:    [[TMP6:%.*]] = bitcast <8 x i8> [[TMP14]] to i64
+; CHECK-NEXT:    [[MMX_VAR_I:%.*]] = bitcast <8 x i8> [[TMP1]] to x86_mmx
+; CHECK-NEXT:    [[TMP7:%.*]] = bitcast <8 x i8> [[TMP11]] to i64
+; CHECK-NEXT:    [[MMX_VAR1_I:%.*]] = bitcast <8 x i8> [[TMP0]] to x86_mmx
+; CHECK-NEXT:    [[_MSPROP:%.*]] = or i64 [[TMP6]], [[TMP7]]
+; CHECK-NEXT:    [[TMP2:%.*]] = tail call x86_mmx @llvm.x86.mmx.punpckhbw(x86_mmx [[MMX_VAR_I]], x86_mmx [[MMX_VAR1_I]]) #[[ATTR2]]
+; CHECK-NEXT:    [[TMP9:%.*]] = bitcast i64 [[_MSPROP]] to <8 x i8>
+; CHECK-NEXT:    [[TMP3:%.*]] = bitcast x86_mmx [[TMP2]] to <8 x i8>
+; CHECK-NEXT:    [[TMP4:%.*]] = bitcast <8 x i8> [[TMP9]] to <1 x i64>
+; CHECK-NEXT:    [[TMP12:%.*]] = bitcast <8 x i8> [[TMP3]] to <1 x i64>
+; CHECK-NEXT:    [[TMP5:%.*]] = extractelement <1 x i64> [[TMP4]], i32 0
+; CHECK-NEXT:    [[TMP13:%.*]] = extractelement <1 x i64> [[TMP12]], i32 0
+; CHECK-NEXT:    store i64 [[TMP5]], ptr @__msan_retval_tls, align 8
+; CHECK-NEXT:    ret i64 [[TMP13]]
+;
+entry:
+  %0 = bitcast <1 x i64> %b to <8 x i8>
+  %1 = bitcast <1 x i64> %a to <8 x i8>
+  %mmx_var.i = bitcast <8 x i8> %1 to x86_mmx
+  %mmx_var1.i = bitcast <8 x i8> %0 to x86_mmx
+  %2 = tail call x86_mmx @llvm.x86.mmx.punpckhbw(x86_mmx %mmx_var.i, x86_mmx %mmx_var1.i) nounwind
+  %3 = bitcast x86_mmx %2 to <8 x i8>
+  %4 = bitcast <8 x i8> %3 to <1 x i64>
+  %5 = extractelement <1 x i64> %4, i32 0
+  ret i64 %5
+}
+
+declare x86_mmx @llvm.x86.mmx.packuswb(x86_mmx, x86_mmx) nounwind readnone
+
+define i64 @test76(<1 x i64> %a, <1 x i64> %b) #0 {
+; CHECK-LABEL: define i64 @test76(
+; CHECK-SAME: <1 x i64> [[A:%.*]], <1 x i64> [[B:%.*]]) #[[ATTR1]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP16:%.*]] = load <1 x i64>, ptr inttoptr (i64 add (i64 ptrtoint (ptr @__msan_param_tls to i64), i64 8) to ptr), align 8
+; CHECK-NEXT:    [[TMP17:%.*]] = load <1 x i64>, ptr @__msan_param_tls, align 8
+; CHECK-NEXT:    call void @llvm.donothing()
+; CHECK-NEXT:    [[TMP19:%.*]] = bitcast <1 x i64> [[TMP16]] to <4 x i16>
+; CHECK-NEXT:    [[TMP0:%.*]] = bitcast <1 x i64> [[B]] to <4 x i16>
+; CHECK-NEXT:    [[TMP20:%.*]] = bitcast <1 x i64> [[TMP17]] to <4 x i16>
+; CHECK-NEXT:    [[TMP1:%.*]] = bitcast <1 x i64> [[A]] to <4 x i16>
+; CHECK-NEXT:    [[TMP23:%.*]] = bitcast <4 x i16> [[TMP20]] to i64
+; CHECK-NEXT:    [[MMX_VAR_I:%.*]] = bitcast <4 x i16> [[TMP1]] to x86_mmx
+; CHECK-NEXT:    [[TMP7:%.*]] = bitcast <4 x i16> [[TMP19]] to i64
+; CHECK-NEXT:    [[MMX_VAR1_I:%.*]] = bitcast <4 x i16> [[TMP0]] to x86_mmx
+; CHECK-NEXT:    [[TMP8:%.*]] = bitcast i64 [[TMP23]] to <4 x i16>
+; CHECK-NEXT:    [[TMP9:%.*]] = bitcast i64 [[TMP7]] to <4 x i16>
+; CHECK-NEXT:    [[TMP10:%.*]] = icmp ne <4 x i16> [[TMP8]], zeroinitializer
+; CHECK-NEXT:    [[TMP11:%.*]] = sext <4 x i1> [[TMP10]] to <4 x i16>
+; CHECK-NEXT:    [[TMP12:%.*]] = icmp ne <4 x i16> [[TMP9]], zeroinitializer
+; CHECK-NEXT:    [[TMP13:%.*]] = sext <4 x i1> [[TMP12]] to <4 x i16>
+; CHECK-NEXT:    [[TMP14:%.*]] = bitcast <4 x i16> [[TMP11]] to x86_mmx
+; CHECK-NEXT:    [[TMP15:%.*]] = bitcast <4 x i16> [[TMP13]] to x86_mmx
+; CHECK-NEXT:    [[_MSPROP_VECTOR_PACK:%.*]] = call x86_mmx @llvm.x86.mmx.packsswb(x86_mmx [[TMP14]], x86_mmx [[TMP15]])
+; CHECK-NEXT:    [[TMP2:%.*]] = bitcast x86_mmx [[_MSPROP_VECTOR_PACK]] to i64
+; CHECK-NEXT:    [[TMP3:%.*]] = tail call x86_mmx @llvm.x86.mmx.packuswb(x86_mmx [[MMX_VAR_I]], x86_mmx [[MMX_VAR1_I]]) #[[ATTR2]]
+; CHECK-NEXT:    [[TMP18:%.*]] = bitcast i64 [[TMP2]] to <8 x i8>
+; CHECK-NEXT:    [[TMP4:%.*]] = bitcast x86_mmx [[TMP3]] to <8 x i8>
+; CHECK-NEXT:    [[TMP5:%.*]] = bitcast <8 x i8> [[TMP18]] to <1 x i64>
+; CHECK-NEXT:    [[TMP21:%.*]] = bitcast <8 x i8> [[TMP4]] to <1 x i64>
+; CHECK-NEXT:    [[TMP6:%.*]] = extractelement <1 x i64> [[TMP5]], i32 0
+; CHECK-NEXT:    [[TMP22:%.*]] = extractelement <1 x i64> [[TMP21]], i32 0
+; CHECK-NEXT:    store i64 [[TMP6]], ptr @__msan_retval_tls, align 8
+; CHECK-NEXT:    ret i64 [[TMP22]]
+;
+entry:
+  %0 = bitcast <1 x i64> %b to <4 x i16>
+  %1 = bitcast <1 x i64> %a to <4 x i16>
+  %mmx_var.i = bitcast <4 x i16> %1 to x86_mmx
+  %mmx_var1.i = bitcast <4 x i16> %0 to x86_mmx
+  %2 = tail call x86_mmx @llvm.x86.mmx.packuswb(x86_mmx %mmx_var.i, x86_mmx %mmx_var1.i) nounwind
+  %3 = bitcast x86_mmx %2 to <8 x i8>
+  %4 = bitcast <8 x i8> %3 to <1 x i64>
+  %5 = extractelement <1 x i64> %4, i32 0
+  ret i64 %5
+}
+
+declare x86_mmx @llvm.x86.mmx.packssdw(x86_mmx, x86_mmx) nounwind readnone
+
+define i64 @test75(<1 x i64> %a, <1 x i64> %b) #0 {
+; CHECK-LABEL: define i64 @test75(
+; CHECK-SAME: <1 x i64> [[A:%.*]], <1 x i64> [[B:%.*]]) #[[ATTR1]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP16:%.*]] = load <1 x i64>, ptr inttoptr (i64 add (i64 ptrtoint (ptr @__msan_param_tls to i64), i64 8) to ptr), align 8
+; CHECK-NEXT:    [[TMP17:%.*]] = load <1 x i64>, ptr @__msan_param_tls, align 8
+; CHECK-NEXT:    call void @llvm.donothing()
+; CHECK-NEXT:    [[TMP19:%.*]] = bitcast <1 x i64> [[TMP16]] to <2 x i32>
+; CHECK-NEXT:    [[TMP0:%.*]] = bitcast <1 x i64> [[B]] to <2 x i32>
+; CHECK-NEXT:    [[TMP20:%.*]] = bitcast <1 x i64> [[TMP17]] to <2 x i32>
+; CHECK-NEXT:    [[TMP1:%.*]] = bitcast <1 x i64> [[A]] to <2 x i32>
+; CHECK-NEXT:    [[TMP23:%.*]] = bitcast <2 x i32> [[TMP20]] to i64
+; CHECK-NEXT:    [[MMX_VAR_I:%.*]] = bitcast <2 x i32> [[TMP1]] to x86_mmx
+; CHECK-NEXT:    [[TMP7:%.*]] = bitcast <2 x i32> [[TMP19]] to i64
+; CHECK-NEXT:    [[MMX_VAR1_I:%.*]] = bitcast <2 x i32> [[TMP0]] to x86_mmx
+; CHECK-NEXT:    [[TMP8:%.*]] = bitcast i64 [[TMP23]] to <2 x i32>
+; CHECK-NEXT:    [[TMP9:%.*]] = bitcast i64 [[TMP7]] to <2 x i32>
+; CHECK-NEXT:    [[TMP10:%.*]] = icmp ne <2 x i32> [[TMP8]], zeroinitializer
+; CHECK-NEXT:    [[TMP11:%.*]] = sext <2 x i1> [[TMP10]] to <2 x i32>
+; CHECK-NEXT:    [[TMP12:%.*]] = icmp ne <2 x i32> [[TMP9]], zeroinitializer
+; CHECK-NEXT:    [[TMP13:%.*]] = sext <2 x i1> [[TMP12]] to <2 x i32>
+; CHECK-NEXT:    [[TMP14:%.*]] = bitcast <2 x i32> [[TMP11]] to x86_mmx
+; CHECK-NEXT:    [[TMP15:%.*]] = bitcast <2 x i32> [[TMP13]] to x86_mmx
+; CHECK-NEXT:    [[_MSPROP_VECTOR_PACK:%.*]] = call x86_mmx @llvm.x86.mmx.packssdw(x86_mmx [[TMP14]], x86_mmx [[TMP15]])
+; CHECK-NEXT:    [[TMP2:%.*]] = bitcast x86_mmx [[_MSPROP_VECTOR_PACK]] to i64
+; CHECK-NEXT:    [[TMP3:%.*]] = tail call x86_mmx @llvm.x86.mmx.packssdw(x86_mmx [[MMX_VAR_I]], x86_mmx [[MMX_VAR1_I]]) #[[ATTR2]]
+; CHECK-NEXT:    [[TMP18:%.*]] = bitcast i64 [[TMP2]] to <4 x i16>
+; CHECK-NEXT:    [[TMP4:%.*]] = bitcast x86_mmx [[TMP3]] to <4 x i16>
+; CHECK-NEXT:    [[TMP5:%.*]] = bitcast <4 x i16> [[TMP18]] to <1 x i64>
+; CHECK-NEXT:    [[TMP21:%.*]] = bitcast <4 x i16> [[TMP4]] to <1 x i64>
+; CHECK-NEXT:    [[TMP6:%.*]] = extractelement <1 x i64> [[TMP5]], i32 0
+; CHECK-NEXT:    [[TMP22:%.*]] = extractelement <1 x i64> [[TMP21]], i32 0
+; CHECK-NEXT:    store i64 [[TMP6]], ptr @__msan_retval_tls, align 8
+; CHECK-NEXT:    ret i64 [[TMP22]]
+;
+entry:
+  %0 = bitcast <1 x i64> %b to <2 x i32>
+  %1 = bitcast <1 x i64> %a to <2 x i32>
+  %mmx_var.i = bitcast <2 x i32> %1 to x86_mmx
+  %mmx_var1.i = bitcast <2 x i32> %0 to x86_mmx
+  %2 = tail call x86_mmx @llvm.x86.mmx.packssdw(x86_mmx %mmx_var.i, x86_mmx %mmx_var1.i) nounwind
+  %3 = bitcast x86_mmx %2 to <4 x i16>
+  %4 = bitcast <4 x i16> %3 to <1 x i64>
+  %5 = extractelement <1 x i64> %4, i32 0
+  ret i64 %5
+}
+
+declare x86_mmx @llvm.x86.mmx.packsswb(x86_mmx, x86_mmx) nounwind readnone
+
+define i64 @test74(<1 x i64> %a, <1 x i64> %b) #0 {
+; CHECK-LABEL: define i64 @test74(
+; CHECK-SAME: <1 x i64> [[A:%.*]], <1 x i64> [[B:%.*]]) #[[ATTR1]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP16:%.*]] = load <1 x i64>, ptr inttoptr (i64 add (i64 ptrtoint (ptr @__msan_param_tls to i64), i64 8) to ptr), align 8
+; CHECK-NEXT:    [[TMP17:%.*]] = load <1 x i64>, ptr @__msan_param_tls, align 8
+; CHECK-NEXT:    call void @llvm.donothing()
+; CHECK-NEXT:    [[TMP19:%.*]] = bitcast <1 x i64> [[TMP16]] to <4 x i16>
+; CHECK-NEXT:    [[TMP0:%.*]] = bitcast <1 x i64> [[B]] to <4 x i16>
+; CHECK-NEXT:    [[TMP20:%.*]] = bitcast <1 x i64> [[TMP17]] to <4 x i16>
+; CHECK-NEXT:    [[TMP1:%.*]] = bitcast <1 x i64> [[A]] to <4 x i16>
+; CHECK-NEXT:    [[TMP23:%.*]] = bitcast <4 x i16> [[TMP20]] to i64
+; CHECK-NEXT:    [[MMX_VAR_I:%.*]] = bitcast <4 x i16> [[TMP1]] to x86_mmx
+; CHECK-NEXT:    [[TMP7:%.*]] = bitcast <4 x i16> [[TMP19]] to i64
+; CHECK-NEXT:    [[MMX_VAR1_I:%.*]] = bitcast <4 x i16> [[TMP0]] to x86_mmx
+; CHECK-NEXT:    [[TMP8:%.*]] = bitcast i64 [[TMP23]] to <4 x i16>
+; CHECK-NEXT:    [[TMP9:%.*]] = bitcast i64 [[TMP7]] to <4 x i16>
+; CHECK-NEXT:    [[TMP10:%.*]] = icmp ne <4 x i16> [[TMP8]], zeroinitializer
+; CHECK-NEXT:    [[TMP11:%.*]] = sext <4 x i1> [[TMP10]] to <4 x i16>
+; CHECK-NEXT:    [[TMP12:%.*]] = icmp ne <4 x i16> [[TMP9]], zeroinitializer
+; CHECK-NEXT:    [[TMP13:%.*]] = sext <4 x i1> [[TMP12]] to <4 x i16>
+; CHECK-NEXT:    [[TMP14:%.*]] = bitcast <4 x i16> [[TMP11]] to x86_mmx
+; CHECK-NEXT:    [[TMP15:%.*]] = bitcast <4 x i16> [[TMP13]] to x86_mmx
+; CHECK-NEXT:    [[_MSPROP_VECTOR_PACK:%.*]] = call x86_mmx @llvm.x86.mmx.packsswb(x86_mmx [[TMP14]], x86_mmx [[TMP15]])
+; CHECK-NEXT:    [[TMP2:%.*]] = bitcast x86_mmx [[_MSPROP_VECTOR_PACK]] to i64
+; CHECK-NEXT:    [[TMP3:%.*]] = tail call x86_mmx @llvm.x86.mmx.packsswb(x86_mmx [[MMX_VAR_I]], x86_mmx [[MMX_VAR1_I]]) #[[ATTR2]]
+; CHECK-NEXT:    [[TMP18:%.*]] = bitcast i64 [[TMP2]] to <8 x i8>
+; CHECK-NEXT:    [[TMP4:%.*]] = bitcast x86_mmx [[TMP3]] to <8 x i8>
+; CHECK-NEXT:    [[TMP5:%.*]] = bitcast <8 x i8> [[TMP18]] to <1 x i64>
+; CHECK-NEXT:    [[TMP21:%.*]] = bitcast <8 x i8> [[TMP4]] to <1 x i64>
+; CHECK-NEXT:    [[TMP6:%.*]] = extractelement <1 x i64> [[TMP5]], i32 0
+; CHECK-NEXT:    [[TMP22:%.*]] = extractelement <1 x i64> [[TMP21]], i32 0
+; CHECK-NEXT:    store i64 [[TMP6]], ptr @__msan_retval_tls, align 8
+; CHECK-NEXT:    ret i64 [[TMP22]]
+;
+entry:
+  %0 = bitcast <1 x i64> %b to <4 x i16>
+  %1 = bitcast <1 x i64> %a to <4 x i16>
+  %mmx_var.i = bitcast <4 x i16> %1 to x86_mmx
+  %mmx_var1.i = bitcast <4 x i16> %0 to x86_mmx
+  %2 = tail call x86_mmx @llvm.x86.mmx.packsswb(x86_mmx %mmx_var.i, x86_mmx %mmx_var1.i) nounwind
+  %3 = bitcast x86_mmx %2 to <8 x i8>
+  %4 = bitcast <8 x i8> %3 to <1 x i64>
+  %5 = extractelement <1 x i64> %4, i32 0
+  ret i64 %5
+}
+
+declare x86_mmx @llvm.x86.mmx.psrai.d(x86_mmx, i32) nounwind readnone
+
+define i64 @test73(<1 x i64> %a) #0 {
+; CHECK-LABEL: define i64 @test73(
+; CHECK-SAME: <1 x i64> [[A:%.*]]) #[[ATTR1]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP7:%.*]] = load <1 x i64>, ptr @__msan_param_tls, align 8
+; CHECK-NEXT:    call void @llvm.donothing()
+; CHECK-NEXT:    [[TMP8:%.*]] = bitcast <1 x i64> [[TMP7]] to <2 x i32>
+; CHECK-NEXT:    [[TMP0:%.*]] = bitcast <1 x i64> [[A]] to <2 x i32>
+; CHECK-NEXT:    [[TMP9:%.*]] = bitcast <2 x i32> [[TMP8]] to i64
+; CHECK-NEXT:    [[MMX_VAR_I:%.*]] = bitcast <2 x i32> [[TMP0]] to x86_mmx
+; CHECK-NEXT:    [[TMP10:%.*]] = bitcast i64 [[TMP9]] to x86_mmx
+; CHECK-NEXT:    [[TMP1:%.*]] = call x86_mmx @llvm.x86.mmx.psrai.d(x86_mmx [[TMP10]], i32 3)
+; CHECK-NEXT:    [[TMP2:%.*]] = bitcast x86_mmx [[TMP1]] to i64
+; CHECK-NEXT:    [[TMP11:%.*]] = or i64 [[TMP2]], 0
+; CHECK-NEXT:    [[TMP3:%.*]] = tail call x86_mmx @llvm.x86.mmx.psrai.d(x86_mmx [[MMX_VAR_I]], i32 3) #[[ATTR2]]
+; CHECK-NEXT:    [[TMP14:%.*]] = bitcast i64 [[TMP11]] to <2 x i32>
+; CHECK-NEXT:    [[TMP4:%.*]] = bitcast x86_mmx [[TMP3]] to <2 x i32>
+; CHECK-NEXT:    [[TMP5:%.*]] = bitcast <2 x i32> [[TMP14]] to <1 x i64>
+; CHECK-NEXT:    [[TMP12:%.*]] = bitcast <2 x i32> [[TMP4]] to <1 x i64>
+; CHECK-NEXT:    [[TMP6:%.*]] = extractelement <1 x i64> [[TMP5]], i32 0
+; CHECK-NEXT:    [[TMP13:%.*]] = extractelement <1 x i64> [[TMP12]], i32 0
+; CHECK-NEXT:    store i64 [[TMP6]], ptr @__msan_retval_tls, align 8
+; CHECK-NEXT:    ret i64 [[TMP13]]
+;
+entry:
+  %0 = bitcast <1 x i64> %a to <2 x i32>
+  %mmx_var.i = bitcast <2 x i32> %0 to x86_mmx
+  %1 = tail call x86_mmx @llvm.x86.mmx.psrai.d(x86_mmx %mmx_var.i, i32 3) nounwind
+  %2 = bitcast x86_mmx %1 to <2 x i32>
+  %3 = bitcast <2 x i32> %2 to <1 x i64>
+  %4 = extractelement <1 x i64> %3, i32 0
+  ret i64 %4
+}
+
+declare x86_mmx @llvm.x86.mmx.psrai.w(x86_mmx, i32) nounwind readnone
+
+define i64 @test72(<1 x i64> %a) #0 {
+; CHECK-LABEL: define i64 @test72(
+; CHECK-SAME: <1 x i64> [[A:%.*]]) #[[ATTR1]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP7:%.*]] = load <1 x i64>, ptr @__msan_param_tls, align 8
+; CHECK-NEXT:    call void @llvm.donothing()
+; CHECK-NEXT:    [[TMP8:%.*]] = bitcast <1 x i64> [[TMP7]] to <4 x i16>
+; CHECK-NEXT:    [[TMP0:%.*]] = bitcast <1 x i64> [[A]] to <4 x i16>
+; CHECK-NEXT:    [[TMP9:%.*]] = bitcast <4 x i16> [[TMP8]] to i64
+; CHECK-NEXT:    [[MMX_VAR_I:%.*]] = bitcast <4 x i16> [[TMP0]] to x86_mmx
+; CHECK-NEXT:    [[TMP10:%.*]] = bitcast i64 [[TMP9]] to x86_mmx
+; CHECK-NEXT:    [[TMP1:%.*]] = call x86_mmx @llvm.x86.mmx.psrai.w(x86_mmx [[TMP10]], i32 3)
+; CHECK-NEXT:    [[TMP2:%.*]] = bitcast x86_mmx [[TMP1]] to i64
+; CHECK-NEXT:    [[TMP11:%.*]] = or i64 [[TMP2]], 0
+; CHECK-NEXT:    [[TMP3:%.*]] = tail call x86_mmx @llvm.x86.mmx.psrai.w(x86_mmx [[MMX_VAR_I]], i32 3) #[[ATTR2]]
+; CHECK-NEXT:    [[TMP14:%.*]] = bitcast i64 [[TMP11]] to <4 x i16>
+; CHECK-NEXT:    [[TMP4:%.*]] = bitcast x86_mmx [[TMP3]] to <4 x i16>
+; CHECK-NEXT:    [[TMP5:%.*]] = bitcast <4 x i16> [[TMP14]] to <1 x i64>
+; CHECK-NEXT:    [[TMP12:%.*]] = bitcast <4 x i16> [[TMP4]] to <1 x i64>
+; CHECK-NEXT:    [[TMP6:%.*]] = extractelement <1 x i64> [[TMP5]], i32 0
+; CHECK-NEXT:    [[TMP13:%.*]] = extractelement <1 x i64> [[TMP12]], i32 0
+; CHECK-NEXT:    store i64 [[TMP6]], ptr @__msan_retval_tls, align 8
+; CHECK-NEXT:    ret i64 [[TMP13]]
+;
+entry:
+  %0 = bitcast <1 x i64> %a to <4 x i16>
+  %mmx_var.i = bitcast <4 x i16> %0 to x86_mmx
+  %1 = tail call x86_mmx @llvm.x86.mmx.psrai.w(x86_mmx %mmx_var.i, i32 3) nounwind
+  %2 = bitcast x86_mmx %1 to <4 x i16>
+  %3 = bitcast <4 x i16> %2 to <1 x i64>
+  %4 = extractelement <1 x i64> %3, i32 0
+  ret i64 %4
+}
+
+define i64 @test72_2(<1 x i64> %a) #0 {
+; CHECK-LABEL: define i64 @test72_2(
+; CHECK-SAME: <1 x i64> [[A:%.*]]) #[[ATTR1]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP7:%.*]] = load <1 x i64>, ptr @__msan_param_tls, align 8
+; CHECK-NEXT:    call void @llvm.donothing()
+; CHECK-NEXT:    [[TMP8:%.*]] = bitcast <1 x i64> [[TMP7]] to <4 x i16>
+; CHECK-NEXT:    [[TMP0:%.*]] = bitcast <1 x i64> [[A]] to <4 x i16>
+; CHECK-NEXT:    [[TMP9:%.*]] = bitcast <4 x i16> [[TMP8]] to i64
+; CHECK-NEXT:    [[MMX_VAR_I:%.*]] = bitcast <4 x i16> [[TMP0]] to x86_mmx
+; CHECK-NEXT:    [[TMP10:%.*]] = bitcast i64 [[TMP9]] to x86_mmx
+; CHECK-NEXT:    [[TMP1:%.*]] = call x86_mmx @llvm.x86.mmx.psrai.w(x86_mmx [[TMP10]], i32 0)
+; CHECK-NEXT:    [[TMP2:%.*]] = bitcast x86_mmx [[TMP1]] to i64
+; CHECK-NEXT:    [[TMP11:%.*]] = or i64 [[TMP2]], 0
+; CHECK-NEXT:    [[TMP3:%.*]] = tail call x86_mmx @llvm.x86.mmx.psrai.w(x86_mmx [[MMX_VAR_I]], i32 0) #[[ATTR2]]
+; CHECK-NEXT:    [[TMP14:%.*]] = bitcast i64 [[TMP11]] to <4 x i16>
+; CHECK-NEXT:    [[TMP4:%.*]] = bitcast x86_mmx [[TMP3]] to <4 x i16>
+; CHECK-NEXT:    [[TMP5:%.*]] = bitcast <4 x i16> [[TMP14]] to <1 x i64>
+; CHECK-NEXT:    [[TMP12:%.*]] = bitcast <4 x i16> [[TMP4]] to <1 x i64>
+; CHECK-NEXT:    [[TMP6:%.*]] = extractelement <1 x i64> [[TMP5]], i32 0
+; CHECK-NEXT:    [[TMP13:%.*]] = extractelement <1 x i64> [[TMP12]], i32 0
+; CHECK-NEXT:    store i64 [[TMP6]], ptr @__msan_retval_tls, align 8
+; CHECK-NEXT:    ret i64 [[TMP13]]
+;
+entry:
+  %0 = bitcast <1 x i64> %a to <4 x i16>
+  %mmx_var.i = bitcast <4 x i16> %0 to x86_mmx
+  %1 = tail call x86_mmx @llvm.x86.mmx.psrai.w(x86_mmx %mmx_var.i, i32 0) nounwind
+  %2 = bitcast x86_mmx %1 to <4 x i16>
+  %3 = bitcast <4 x i16> %2 to <1 x i64>
+  %4 = extractelement <1 x i64> %3, i32 0
+  ret i64 %4
+}
+
+declare x86_mmx @llvm.x86.mmx.psrli.q(x86_mmx, i32) nounwind readnone
+
+define i64 @test71(<1 x i64> %a) #0 {
+; CHECK-LABEL: define i64 @test71(
+; CHECK-SAME: <1 x i64> [[A:%.*]]) #[[ATTR1]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP5:%.*]] = load <1 x i64>, ptr @__msan_param_tls, align 8
+; CHECK-NEXT:    call void @llvm.donothing()
+; CHECK-NEXT:    [[_MSPROP:%.*]] = extractelement <1 x i64> [[TMP5]], i32 0
+; CHECK-NEXT:    [[TMP0:%.*]] = extractelement <1 x i64> [[A]], i32 0
+; CHECK-NEXT:    [[MMX_VAR_I:%.*]] = bitcast i64 [[TMP0]] to x86_mmx
+; CHECK-NEXT:    [[TMP6:%.*]] = bitcast i64 [[_MSPROP]] to x86_mmx
+; CHECK-NEXT:    [[TMP1:%.*]] = call x86_mmx @llvm.x86.mmx.psrli.q(x86_mmx [[TMP6]], i32 3)
+; CHECK-NEXT:    [[TMP2:%.*]] = bitcast x86_mmx [[TMP1]] to i64
+; CHECK-NEXT:    [[TMP7:%.*]] = or i64 [[TMP2]], 0
+; CHECK-NEXT:    [[TMP3:%.*]] = tail call x86_mmx @llvm.x86.mmx.psrli.q(x86_mmx [[MMX_VAR_I]], i32 3) #[[ATTR2]]
+; CHECK-NEXT:    [[TMP4:%.*]] = bitcast x86_mmx [[TMP3]] to i64
+; CHECK-NEXT:    store i64 [[TMP7]], ptr @__msan_retval_tls, align 8
+; CHECK-NEXT:    ret i64 [[TMP4]]
+;
+entry:
+  %0 = extractelement <1 x i64> %a, i32 0
+  %mmx_var.i = bitcast i64 %0 to x86_mmx
+  %1 = tail call x86_mmx @llvm.x86.mmx.psrli.q(x86_mmx %mmx_var.i, i32 3) nounwind
+  %2 = bitcast x86_mmx %1 to i64
+  ret i64 %2
+}
+
+declare x86_mmx @llvm.x86.mmx.psrli.d(x86_mmx, i32) nounwind readnone
+
+define i64 @test70(<1 x i64> %a) #0 {
+; CHECK-LABEL: define i64 @test70(
+; CHECK-SAME: <1 x i64> [[A:%.*]]) #[[ATTR1]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP7:%.*]] = load <1 x i64>, ptr @__msan_param_tls, align 8
+; CHECK-NEXT:    call void @llvm.donothing()
+; CHECK-NEXT:    [[TMP8:%.*]] = bitcast <1 x i64> [[TMP7]] to <2 x i32>
+; CHECK-NEXT:    [[TMP0:%.*]] = bitcast <1 x i64> [[A]] to <2 x i32>
+; CHECK-NEXT:    [[TMP9:%.*]] = bitcast <2 x i32> [[TMP8]] to i64
+; CHECK-NEXT:    [[MMX_VAR_I:%.*]] = bitcast <2 x i32> [[TMP0]] to x86_mmx
+; CHECK-NEXT:    [[TMP10:%.*]] = bitcast i64 [[TMP9]] to x86_mmx
+; CHECK-NEXT:    [[TMP1:%.*]] = call x86_mmx @llvm.x86.mmx.psrli.d(x86_mmx [[TMP10]], i32 3)
+; CHECK-NEXT:    [[TMP2:%.*]] = bitcast x86_mmx [[TMP1]] to i64
+; CHECK-NEXT:    [[TMP11:%.*]] = or i64 [[TMP2]], 0
+; CHECK-NEXT:    [[TMP3:%.*]] = tail call x86_mmx @llvm.x86.mmx.psrli.d(x86_mmx [[MMX_VAR_I]], i32 3) #[[ATTR2]]
+; CHECK-NEXT:    [[TMP14:%.*]] = bitcast i64 [[TMP11]] to <2 x i32>
+; CHECK-NEXT:    [[TMP4:%.*]] = bitcast x86_mmx [[TMP3]] to <2 x i32>
+; CHECK-NEXT:    [[TMP5:%.*]] = bitcast <2 x i32> [[TMP14]] to <1 x i64>
+; CHECK-NEXT:    [[TMP12:%.*]] = bitcast <2 x i32> [[TMP4]] to <1 x i64>
+; CHECK-NEXT:    [[TMP6:%.*]] = extractelement <1 x i64> [[TMP5]], i32 0
+; CHECK-NEXT:    [[TMP13:%.*]] = extractelement <1 x i64> [[TMP12]], i32 0
+; CHECK-NEXT:    store i64 [[TMP6]], ptr @__msan_retval_tls, align 8
+; CHECK-NEXT:    ret i64 [[TMP13]]
+;
+entry:
+  %0 = bitcast <1 x i64> %a to <2 x i32>
+  %mmx_var.i = bitcast <2 x i32> %0 to x86_mmx
+  %1 = tail call x86_mmx @llvm.x86.mmx.psrli.d(x86_mmx %mmx_var.i, i32 3) nounwind
+  %2 = bitcast x86_mmx %1 to <2 x i32>
+  %3 = bitcast <2 x i32> %2 to <1 x i64>
+  %4 = extractelement <1 x i64> %3, i32 0
+  ret i64 %4
+}
+
+define i64 @test70_2(<1 x i64> %a) #0 {
+; CHECK-LABEL: define i64 @test70_2(
+; CHECK-SAME: <1 x i64> [[A:%.*]]) #[[ATTR1]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP7:%.*]] = load <1 x i64>, ptr @__msan_param_tls, align 8
+; CHECK-NEXT:    call void @llvm.donothing()
+; CHECK-NEXT:    [[TMP8:%.*]] = bitcast <1 x i64> [[TMP7]] to <2 x i32>
+; CHECK-NEXT:    [[TMP0:%.*]] = bitcast <1 x i64> [[A]] to <2 x i32>
+; CHECK-NEXT:    [[TMP9:%.*]] = bitcast <2 x i32> [[TMP8]] to i64
+; CHECK-NEXT:    [[MMX_VAR_I:%.*]] = bitcast <2 x i32> [[TMP0]] to x86_mmx
+; CHECK-NEXT:    [[TMP10:%.*]] = bitcast i64 [[TMP9]] to x86_mmx
+; CHECK-NEXT:    [[TMP1:%.*]] = call x86_mmx @llvm.x86.mmx.psrli.d(x86_mmx [[TMP10]], i32 0)
+; CHECK-NEXT:    [[TMP2:%.*]] = bitcast x86_mmx [[TMP1]] to i64
+; CHECK-NEXT:    [[TMP11:%.*]] = or i64 [[TMP2]], 0
+; CHECK-NEXT:    [[TMP3:%.*]] = tail call x86_mmx @llvm.x86.mmx.psrli.d(x86_mmx [[MMX_VAR_I]], i32 0) #[[ATTR2]]
+; CHECK-NEXT:    [[TMP14:%.*]] = bitcast i64 [[TMP11]] to <2 x i32>
+; CHECK-NEXT:    [[TMP4:%.*]] = bitcast x86_mmx [[TMP3]] to <2 x i32>
+; CHECK-NEXT:    [[TMP5:%.*]] = bitcast <2 x i32> [[TMP14]] to <1 x i64>
+; CHECK-NEXT:    [[TMP12:%.*]] = bitcast <2 x i32> [[TMP4]] to <1 x i64>
+; CHECK-NEXT:    [[TMP6:%.*]] = extractelement <1 x i64> [[TMP5]], i32 0
+; CHECK-NEXT:    [[TMP13:%.*]] = extractelement <1 x i64> [[TMP12]], i32 0
+; CHECK-NEXT:    store i64 [[TMP6]], ptr @__msan_retval_tls, align 8
+; CHECK-NEXT:    ret i64 [[TMP13]]
+;
+entry:
+  %0 = bitcast <1 x i64> %a to <2 x i32>
+  %mmx_var.i = bitcast <2 x i32> %0 to x86_mmx
+  %1 = tail call x86_mmx @llvm.x86.mmx.psrli.d(x86_mmx %mmx_var.i, i32 0) nounwind
+  %2 = bitcast x86_mmx %1 to <2 x i32>
+  %3 = bitcast <2 x i32> %2 to <1 x i64>
+  %4 = extractelement <1 x i64> %3, i32 0
+  ret i64 %4
+}
+
+declare x86_mmx @llvm.x86.mmx.psrli.w(x86_mmx, i32) nounwind readnone
+
+define i64 @test69(<1 x i64> %a) #0 {
+; CHECK-LABEL: define i64 @test69(
+; CHECK-SAME: <1 x i64> [[A:%.*]]) #[[ATTR1]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP7:%.*]] = load <1 x i64>, ptr @__msan_param_tls, align 8
+; CHECK-NEXT:    call void @llvm.donothing()
+; CHECK-NEXT:    [[TMP8:%.*]] = bitcast <1 x i64> [[TMP7]] to <4 x i16>
+; CHECK-NEXT:    [[TMP0:%.*]] = bitcast <1 x i64> [[A]] to <4 x i16>
+; CHECK-NEXT:    [[TMP9:%.*]] = bitcast <4 x i16> [[TMP8]] to i64
+; CHECK-NEXT:    [[MMX_VAR_I:%.*]] = bitcast <4 x i16> [[TMP0]] to x86_mmx
+; CHECK-NEXT:    [[TMP10:%.*]] = bitcast i64 [[TMP9]] to x86_mmx
+; CHECK-NEXT:    [[TMP1:%.*]] = call x86_mmx @llvm.x86.mmx.psrli.w(x86_mmx [[TMP10]], i32 3)
+; CHECK-NEXT:    [[TMP2:%.*]] = bitcast x86_mmx [[TMP1]] to i64
+; CHECK-NEXT:    [[TMP11:%.*]] = or i64 [[TMP2]], 0
+; CHECK-NEXT:    [[TMP3:%.*]] = tail call x86_mmx @llvm.x86.mmx.psrli.w(x86_mmx [[MMX_VAR_I]], i32 3) #[[ATTR2]]
+; CHECK-NEXT:    [[TMP14:%.*]] = bitcast i64 [[TMP11]] to <4 x i16>
+; CHECK-NEXT:    [[TMP4:%.*]] = bitcast x86_mmx [[TMP3]] to <4 x i16>
+; CHECK-NEXT:    [[TMP5:%.*]] = bitcast <4 x i16> [[TMP14]] to <1 x i64>
+; CHECK-NEXT:    [[TMP12:%.*]] = bitcast <4 x i16> [[TMP4]] to <1 x i64>
+; CHECK-NEXT:    [[TMP6:%.*]] = extractelement <1 x i64> [[TMP5]], i32 0
+; CHECK-NEXT:    [[TMP13:%.*]] = extractelement <1 x i64> [[TMP12]], i32 0
+; CHECK-NEXT:    store i64 [[TMP6]], ptr @__msan_retval_tls, align 8
+; CHECK-NEXT:    ret i64 [[TMP13]]
+;
+entry:
+  %0 = bitcast <1 x i64> %a to <4 x i16>
+  %mmx_var.i = bitcast <4 x i16> %0 to x86_mmx
+  %1 = tail call x86_mmx @llvm.x86.mmx.psrli.w(x86_mmx %mmx_var.i, i32 3) nounwind
+  %2 = bitcast x86_mmx %1 to <4 x i16>
+  %3 = bitcast <4 x i16> %2 to <1 x i64>
+  %4 = extractelement <1 x i64> %3, i32 0
+  ret i64 %4
+}
+
+declare x86_mmx @llvm.x86.mmx.pslli.q(x86_mmx, i32) nounwind readnone
+
+define i64 @test68(<1 x i64> %a) #0 {
+; CHECK-LABEL: define i64 @test68(
+; CHECK-SAME: <1 x i64> [[A:%.*]]) #[[ATTR1]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP5:%.*]] = load <1 x i64>, ptr @__msan_param_tls, align 8
+; CHECK-NEXT:    call void @llvm.donothing()
+; CHECK-NEXT:    [[_MSPROP:%.*]] = extractelement <1 x i64> [[TMP5]], i32 0
+; CHECK-NEXT:    [[TMP0:%.*]] = extractelement <1 x i64> [[A]], i32 0
+; CHECK-NEXT:    [[MMX_VAR_I:%.*]] = bitcast i64 [[TMP0]] to x86_mmx
+; CHECK-NEXT:    [[TMP6:%.*]] = bitcast i64 [[_MSPROP]] to x86_mmx
+; CHECK-NEXT:    [[TMP1:%.*]] = call x86_mmx @llvm.x86.mmx.pslli.q(x86_mmx [[TMP6]], i32 3)
+; CHECK-NEXT:    [[TMP2:%.*]] = bitcast x86_mmx [[TMP1]] to i64
+; CHECK-NEXT:    [[TMP7:%.*]] = or i64 [[TMP2]], 0
+; CHECK-NEXT:    [[TMP3:%.*]] = tail call x86_mmx @llvm.x86.mmx.pslli.q(x86_mmx [[MMX_VAR_I]], i32 3) #[[ATTR2]]
+; CHECK-NEXT:    [[TMP4:%.*]] = bitcast x86_mmx [[TMP3]] to i64
+; CHECK-NEXT:    store i64 [[TMP7]], ptr @__msan_retval_tls, align 8
+; CHECK-NEXT:    ret i64 [[TMP4]]
+;
+entry:
+  %0 = extractelement <1 x i64> %a, i32 0
+  %mmx_var.i = bitcast i64 %0 to x86_mmx
+  %1 = tail call x86_mmx @llvm.x86.mmx.pslli.q(x86_mmx %mmx_var.i, i32 3) nounwind
+  %2 = bitcast x86_mmx %1 to i64
+  ret i64 %2
+}
+
+declare x86_mmx @llvm.x86.mmx.pslli.d(x86_mmx, i32) nounwind readnone
+
+define i64 @test67(<1 x i64> %a) #0 {
+; CHECK-LABEL: define i64 @test67(
+; CHECK-SAME: <1 x i64> [[A:%.*]]) #[[ATTR1]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP7:%.*]] = load <1 x i64>, ptr @__msan_param_tls, align 8
+; CHECK-NEXT:    call void @llvm.donothing()
+; CHECK-NEXT:    [[TMP8:%.*]] = bitcast <1 x i64> [[TMP7]] to <2 x i32>
+; CHECK-NEXT:    [[TMP0:%.*]] = bitcast <1 x i64> [[A]] to <2 x i32>
+; CHECK-NEXT:    [[TMP9:%.*]] = bitcast <2 x i32> [[TMP8]] to i64
+; CHECK-NEXT:    [[MMX_VAR_I:%.*]] = bitcast <2 x i32> [[TMP0]] to x86_mmx
+; CHECK-NEXT:    [[TMP10:%.*]] = bitcast i64 [[TMP9]] to x86_mmx
+; CHECK-NEXT:    [[TMP1:%.*]] = call x86_mmx @llvm.x86.mmx.pslli.d(x86_mmx [[TMP10]], i32 3)
+; CHECK-NEXT:    [[TMP2:%.*]] = bitcast x86_mmx [[TMP1]] to i64
+; CHECK-NEXT:    [[TMP11:%.*]] = or i64 [[TMP2]], 0
+; CHECK-NEXT:    [[TMP3:%.*]] = tail call x86_mmx @llvm.x86.mmx.pslli.d(x86_mmx [[MMX_VAR_I]], i32 3) #[[ATTR2]]
+; CHECK-NEXT:    [[TMP14:%.*]] = bitcast i64 [[TMP11]] to <2 x i32>
+; CHECK-NEXT:    [[TMP4:%.*]] = bitcast x86_mmx [[TMP3]] to <2 x i32>
+; CHECK-NEXT:    [[TMP5:%.*]] = bitcast <2 x i32> [[TMP14]] to <1 x i64>
+; CHECK-NEXT:    [[TMP12:%.*]] = bitcast <2 x i32> [[TMP4]] to <1 x i64>
+; CHECK-NEXT:    [[TMP6:%.*]] = extractelement <1 x i64> [[TMP5]], i32 0
+; CHECK-NEXT:    [[TMP13:%.*]] = extractelement <1 x i64> [[TMP12]], i32 0
+; CHECK-NEXT:    store i64 [[TMP6]], ptr @__msan_retval_tls, align 8
+; CHECK-NEXT:    ret i64 [[TMP13]]
+;
+entry:
+  %0 = bitcast <1 x i64> %a to <2 x i32>
+  %mmx_var.i = bitcast <2 x i32> %0 to x86_mmx
+  %1 = tail call x86_mmx @llvm.x86.mmx.pslli.d(x86_mmx %mmx_var.i, i32 3) nounwind
+  %2 = bitcast x86_mmx %1 to <2 x i32>
+  %3 = bitcast <2 x i32> %2 to <1 x i64>
+  %4 = extractelement <1 x i64> %3, i32 0
+  ret i64 %4
+}
+
+declare x86_mmx @llvm.x86.mmx.pslli.w(x86_mmx, i32) nounwind readnone
+
+define i64 @test66(<1 x i64> %a) #0 {
+; CHECK-LABEL: define i64 @test66(
+; CHECK-SAME: <1 x i64> [[A:%.*]]) #[[ATTR1]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP7:%.*]] = load <1 x i64>, ptr @__msan_param_tls, align 8
+; CHECK-NEXT:    call void @llvm.donothing()
+; CHECK-NEXT:    [[TMP8:%.*]] = bitcast <1 x i64> [[TMP7]] to <4 x i16>
+; CHECK-NEXT:    [[TMP0:%.*]] = bitcast <1 x i64> [[A]] to <4 x i16>
+; CHECK-NEXT:    [[TMP9:%.*]] = bitcast <4 x i16> [[TMP8]] to i64
+; CHECK-NEXT:    [[MMX_VAR_I:%.*]] = bitcast <4 x i16> [[TMP0]] to x86_mmx
+; CHECK-NEXT:    [[TMP10:%.*]] = bitcast i64 [[TMP9]] to x86_mmx
+; CHECK-NEXT:    [[TMP1:%.*]] = call x86_mmx @llvm.x86.mmx.pslli.w(x86_mmx [[TMP10]], i32 3)
+; CHECK-NEXT:    [[TMP2:%.*]] = bitcast x86_mmx [[TMP1]] to i64
+; CHECK-NEXT:    [[TMP11:%.*]] = or i64 [[TMP2]], 0
+; CHECK-NEXT:    [[TMP3:%.*]] = tail call x86_mmx @llvm.x86.mmx.pslli.w(x86_mmx [[MMX_VAR_I]], i32 3) #[[ATTR2]]
+; CHECK-NEXT:    [[TMP14:%.*]] = bitcast i64 [[TMP11]] to <4 x i16>
+; CHECK-NEXT:    [[TMP4:%.*]] = bitcast x86_mmx [[TMP3]] to <4 x i16>
+; CHECK-NEXT:    [[TMP5:%.*]] = bitcast <4 x i16> [[TMP14]] to <1 x i64>
+; CHECK-NEXT:    [[TMP12:%.*]] = bitcast <4 x i16> [[TMP4]] to <1 x i64>
+; CHECK-NEXT:    [[TMP6:%.*]] = extractelement <1 x i64> [[TMP5]], i32 0
+; CHECK-NEXT:    [[TMP13:%.*]] = extractelement <1 x i64> [[TMP12]], i32 0
+; CHECK-NEXT:    store i64 [[TMP6]], ptr @__msan_retval_tls, align 8
+; CHECK-NEXT:    ret i64 [[TMP13]]
+;
+entry:
+  %0 = bitcast <1 x i64> %a to <4 x i16>
+  %mmx_var.i = bitcast <4 x i16> %0 to x86_mmx
+  %1 = tail call x86_mmx @llvm.x86.mmx.pslli.w(x86_mmx %mmx_var.i, i32 3) nounwind
+  %2 = bitcast x86_mmx %1 to <4 x i16>
+  %3 = bitcast <4 x i16> %2 to <1 x i64>
+  %4 = extractelement <1 x i64> %3, i32 0
+  ret i64 %4
+}
+
+define i64 @test66_2(<1 x i64> %a) #0 {
+; CHECK-LABEL: define i64 @test66_2(
+; CHECK-SAME: <1 x i64> [[A:%.*]]) #[[ATTR1]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP7:%.*]] = load <1 x i64>, ptr @__msan_param_tls, align 8
+; CHECK-NEXT:    call void @llvm.donothing()
+; CHECK-NEXT:    [[TMP8:%.*]] = bitcast <1 x i64> [[TMP7]] to <4 x i16>
+; CHECK-NEXT:    [[TMP0:%.*]] = bitcast <1 x i64> [[A]] to <4 x i16>
+; CHECK-NEXT:    [[TMP9:%.*]] = bitcast <4 x i16> [[TMP8]] to i64
+; CHECK-NEXT:    [[MMX_VAR_I:%.*]] = bitcast <4 x i16> [[TMP0]] to x86_mmx
+; CHECK-NEXT:    [[TMP10:%.*]] = bitcast i64 [[TMP9]] to x86_mmx
+; CHECK-NEXT:    [[TMP1:%.*]] = call x86_mmx @llvm.x86.mmx.pslli.w(x86_mmx [[TMP10]], i32 0)
+; CHECK-NEXT:    [[TMP2:%.*]] = bitcast x86_mmx [[TMP1]] to i64
+; CHECK-NEXT:    [[TMP11:%.*]] = or i64 [[TMP2]], 0
+; CHECK-NEXT:    [[TMP3:%.*]] = tail call x86_mmx @llvm.x86.mmx.pslli.w(x86_mmx [[MMX_VAR_I]], i32 0) #[[ATTR2]]
+; CHECK-NEXT:    [[TMP14:%.*]] = bitcast i64 [[TMP11]] to <4 x i16>
+; CHECK-NEXT:    [[TMP4:%.*]] = bitcast x86_mmx [[TMP3]] to <4 x i16>
+; CHECK-NEXT:    [[TMP5:%.*]] = bitcast <4 x i16> [[TMP14]] to <1 x i64>
+; CHECK-NEXT:    [[TMP12:%.*]] = bitcast <4 x i16> [[TMP4]] to <1 x i64>
+; CHECK-NEXT:    [[TMP6:%.*]] = extractelement <1 x i64> [[TMP5]], i32 0
+; CHECK-NEXT:    [[TMP13:%.*]] = extractelement <1 x i64> [[TMP12]], i32 0
+; CHECK-NEXT:    store i64 [[TMP6]], ptr @__msan_retval_tls, align 8
+; CHECK-NEXT:    ret i64 [[TMP13]]
+;
+entry:
+  %0 = bitcast <1 x i64> %a to <4 x i16>
+  %mmx_var.i = bitcast <4 x i16> %0 to x86_mmx
+  %1 = tail call x86_mmx @llvm.x86.mmx.pslli.w(x86_mmx %mmx_var.i, i32 0) nounwind
+  %2 = bitcast x86_mmx %1 to <4 x i16>
+  %3 = bitcast <4 x i16> %2 to <1 x i64>
+  %4 = extractelement <1 x i64> %3, i32 0
+  ret i64 %4
+}
+
+declare x86_mmx @llvm.x86.mmx.psra.d(x86_mmx, x86_mmx) nounwind readnone
+
+define i64 @test65(<1 x i64> %a, <1 x i64> %b) #0 {
+; CHECK-LABEL: define i64 @test65(
+; CHECK-SAME: <1 x i64> [[A:%.*]], <1 x i64> [[B:%.*]]) #[[ATTR1]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP9:%.*]] = load <1 x i64>, ptr @__msan_param_tls, align 8
+; CHECK-NEXT:    [[TMP10:%.*]] = load <1 x i64>, ptr inttoptr (i64 add (i64 ptrtoint (ptr @__msan_param_tls to i64), i64 8) to ptr), align 8
+; CHECK-NEXT:    call void @llvm.donothing()
+; CHECK-NEXT:    [[TMP11:%.*]] = bitcast <1 x i64> [[TMP9]] to <2 x i32>
+; CHECK-NEXT:    [[TMP0:%.*]] = bitcast <1 x i64> [[A]] to <2 x i32>
+; CHECK-NEXT:    [[TMP12:%.*]] = bitcast <2 x i32> [[TMP11]] to i64
+; CHECK-NEXT:    [[MMX_VAR_I:%.*]] = bitcast <2 x i32> [[TMP0]] to x86_mmx
+; CHECK-NEXT:    [[_MSPROP:%.*]] = extractelement <1 x i64> [[TMP10]], i32 0
+; CHECK-NEXT:    [[TMP1:%.*]] = extractelement <1 x i64> [[B]], i32 0
+; CHECK-NEXT:    [[MMX_VAR1_I:%.*]] = bitcast i64 [[TMP1]] to x86_mmx
+; CHECK-NEXT:    [[TMP13:%.*]] = icmp ne i64 [[_MSPROP]], 0
+; CHECK-NEXT:    [[TMP14:%.*]] = sext i1 [[TMP13]] to i64
+; CHECK-NEXT:    [[TMP8:%.*]] = bitcast i64 [[TMP12]] to x86_mmx
+; CHECK-NEXT:    [[TMP2:%.*]] = call x86_mmx @llvm.x86.mmx.psra.d(x86_mmx [[TMP8]], x86_mmx [[MMX_VAR1_I]])
+; CHECK-NEXT:    [[TMP3:%.*]] = bitcast x86_mmx [[TMP2]] to i64
+; CHECK-NEXT:    [[TMP15:%.*]] = or i64 [[TMP3]], [[TMP14]]
+; CHECK-NEXT:    [[TMP4:%.*]] = tail call x86_mmx @llvm.x86.mmx.psra.d(x86_mmx [[MMX_VAR_I]], x86_mmx [[MMX_VAR1_I]]) #[[ATTR2]]
+; CHECK-NEXT:    [[TMP18:%.*]] = bitcast i64 [[TMP15]] to <2 x i32>
+; CHECK-NEXT:    [[TMP5:%.*]] = bitcast x86_mmx [[TMP4]] to <2 x i32>
+; CHECK-NEXT:    [[TMP6:%.*]] = bitcast <2 x i32> [[TMP18]] to <1 x i64>
+; CHECK-NEXT:    [[TMP16:%.*]] = bitcast <2 x i32> [[TMP5]] to <1 x i64>
+; CHECK-NEXT:    [[TMP7:%.*]] = extractelement <1 x i64> [[TMP6]], i32 0
+; CHECK-NEXT:    [[TMP17:%.*]] = extractelement <1 x i64> [[TMP16]], i32 0
+; CHECK-NEXT:    store i64 [[TMP7]], ptr @__msan_retval_tls, align 8
+; CHECK-NEXT:    ret i64 [[TMP17]]
+;
+entry:
+  %0 = bitcast <1 x i64> %a to <2 x i32>
+  %mmx_var.i = bitcast <2 x i32> %0 to x86_mmx
+  %1 = extractelement <1 x i64> %b, i32 0
+  %mmx_var1.i = bitcast i64 %1 to x86_mmx
+  %2 = tail call x86_mmx @llvm.x86.mmx.psra.d(x86_mmx %mmx_var.i, x86_mmx %mmx_var1.i) nounwind
+  %3 = bitcast x86_mmx %2 to <2 x i32>
+  %4 = bitcast <2 x i32> %3 to <1 x i64>
+  %5 = extractelement <1 x i64> %4, i32 0
+  ret i64 %5
+}
+
+declare x86_mmx @llvm.x86.mmx.psra.w(x86_mmx, x86_mmx) nounwind readnone
+
+define i64 @test64(<1 x i64> %a, <1 x i64> %b) #0 {
+; CHECK-LABEL: define i64 @test64(
+; CHECK-SAME: <1 x i64> [[A:%.*]], <1 x i64> [[B:%.*]]) #[[ATTR1]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP9:%.*]] = load <1 x i64>, ptr @__msan_param_tls, align 8
+; CHECK-NEXT:    [[TMP10:%.*]] = load <1 x i64>, ptr inttoptr (i64 add (i64 ptrtoint (ptr @__msan_param_tls to i64), i64 8) to ptr), align 8
+; CHECK-NEXT:    call void @llvm.donothing()
+; CHECK-NEXT:    [[TMP11:%.*]] = bitcast <1 x i64> [[TMP9]] to <4 x i16>
+; CHECK-NEXT:    [[TMP0:%.*]] = bitcast <1 x i64> [[A]] to <4 x i16>
+; CHECK-NEXT:    [[TMP12:%.*]] = bitcast <4 x i16> [[TMP11]] to i64
+; CHECK-NEXT:    [[MMX_VAR_I:%.*]] = bitcast <4 x i16> [[TMP0]] to x86_mmx
+; CHECK-NEXT:    [[_MSPROP:%.*]] = extractelement <1 x i64> [[TMP10]], i32 0
+; CHECK-NEXT:    [[TMP1:%.*]] = extractelement <1 x i64> [[B]], i32 0
+; CHECK-NEXT:    [[MMX_VAR1_I:%.*]] = bitcast i64 [[TMP1]] to x86_mmx
+; CHECK-NEXT:    [[TMP13:%.*]] = icmp ne i64 [[_MSPROP]], 0
+; CHECK-NEXT:    [[TMP14:%.*]] = sext i1 [[TMP13]] to i64
+; CHECK-NEXT:    [[TMP8:%.*]] = bitcast i64 [[TMP12]] to x86_mmx
+; CHECK-NEXT:    [[TMP2:%.*]] = call x86_mmx @llvm.x86.mmx.psra.w(x86_mmx [[TMP8]], x86_mmx [[MMX_VAR1_I]])
+; CHECK-NEXT:    [[TMP3:%.*]] = bitcast x86_mmx [[TMP2]] to i64
+; CHECK-NEXT:    [[TMP15:%.*]] = or i64 [[TMP3]], [[TMP14]]
+; CHECK-NEXT:    [[TMP4:%.*]] = tail call x86_mmx @llvm.x86.mmx.psra.w(x86_mmx [[MMX_VAR_I]], x86_mmx [[MMX_VAR1_I]]) #[[ATTR2]]
+; CHECK-NEXT:    [[TMP18:%.*]] = bitcast i64 [[TMP15]] to <4 x i16>
+; CHECK-NEXT:    [[TMP5:%.*]] = bitcast x86_mmx [[TMP4]] to <4 x i16>
+; CHECK-NEXT:    [[TMP6:%.*]] = bitcast <4 x i16> [[TMP18]] to <1 x i64>
+; CHECK-NEXT:    [[TMP16:%.*]] = bitcast <4 x i16> [[TMP5]] to <1 x i64>
+; CHECK-NEXT:    [[TMP7:%.*]] = extractelement <1 x i64> [[TMP6]], i32 0
+; CHECK-NEXT:    [[TMP17:%.*]] = extractelement <1 x i64> [[TMP16]], i32 0
+; CHECK-NEXT:    store i64 [[TMP7]], ptr @__msan_retval_tls, align 8
+; CHECK-NEXT:    ret i64 [[TMP17]]
+;
+entry:
+  %0 = bitcast <1 x i64> %a to <4 x i16>
+  %mmx_var.i = bitcast <4 x i16> %0 to x86_mmx
+  %1 = extractelement <1 x i64> %b, i32 0
+  %mmx_var1.i = bitcast i64 %1 to x86_mmx
+  %2 = tail call x86_mmx @llvm.x86.mmx.psra.w(x86_mmx %mmx_var.i, x86_mmx %mmx_var1.i) nounwind
+  %3 = bitcast x86_mmx %2 to <4 x i16>
+  %4 = bitcast <4 x i16> %3 to <1 x i64>
+  %5 = extractelement <1 x i64> %4, i32 0
+  ret i64 %5
+}
+
+declare x86_mmx @llvm.x86.mmx.psrl.q(x86_mmx, x86_mmx) nounwind readnone
+
+define i64 @test63(<1 x i64> %a, <1 x i64> %b) #0 {
+; CHECK-LABEL: define i64 @test63(
+; CHECK-SAME: <1 x i64> [[A:%.*]], <1 x i64> [[B:%.*]]) #[[ATTR1]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP7:%.*]] = load <1 x i64>, ptr @__msan_param_tls, align 8
+; CHECK-NEXT:    [[TMP8:%.*]] = load <1 x i64>, ptr inttoptr (i64 add (i64 ptrtoint (ptr @__msan_param_tls to i64), i64 8) to ptr), align 8
+; CHECK-NEXT:    call void @llvm.donothing()
+; CHECK-NEXT:    [[_MSPROP:%.*]] = extractelement <1 x i64> [[TMP7]], i32 0
+; CHECK-NEXT:    [[TMP0:%.*]] = extractelement <1 x i64> [[A]], i32 0
+; CHECK-NEXT:    [[MMX_VAR_I:%.*]] = bitcast i64 [[TMP0]] to x86_mmx
+; CHECK-NEXT:    [[_MSPROP1:%.*]] = extractelement <1 x i64> [[TMP8]], i32 0
+; CHECK-NEXT:    [[TMP1:%.*]] = extractelement <1 x i64> [[B]], i32 0
+; CHECK-NEXT:    [[MMX_VAR1_I:%.*]] = bitcast i64 [[TMP1]] to x86_mmx
+; CHECK-NEXT:    [[TMP9:%.*]] = icmp ne i64 [[_MSPROP1]], 0
+; CHECK-NEXT:    [[TMP10:%.*]] = sext i1 [[TMP9]] to i64
+; CHECK-NEXT:    [[TMP6:%.*]] = bitcast i64 [[_MSPROP]] to x86_mmx
+; CHECK-NEXT:    [[TMP2:%.*]] = call x86_mmx @llvm.x86.mmx.psrl.q(x86_mmx [[TMP6]], x86_mmx [[MMX_VAR1_I]])
+; CHECK-NEXT:    [[TMP3:%.*]] = bitcast x86_mmx [[TMP2]] to i64
+; CHECK-NEXT:    [[TMP11:%.*]] = or i64 [[TMP3]], [[TMP10]]
+; CHECK-NEXT:    [[TMP4:%.*]] = tail call x86_mmx @llvm.x86.mmx.psrl.q(x86_mmx [[MMX_VAR_I]], x86_mmx [[MMX_VAR1_I]]) #[[ATTR2]]
+; CHECK-NEXT:    [[TMP5:%.*]] = bitcast x86_mmx [[TMP4]] to i64
+; CHECK-NEXT:    store i64 [[TMP11]], ptr @__msan_retval_tls, align 8
+; CHECK-NEXT:    ret i64 [[TMP5]]
+;
+entry:
+  %0 = extractelement <1 x i64> %a, i32 0
+  %mmx_var.i = bitcast i64 %0 to x86_mmx
+  %1 = extractelement <1 x i64> %b, i32 0
+  %mmx_var1.i = bitcast i64 %1 to x86_mmx
+  %2 = tail call x86_mmx @llvm.x86.mmx.psrl.q(x86_mmx %mmx_var.i, x86_mmx %mmx_var1.i) nounwind
+  %3 = bitcast x86_mmx %2 to i64
+  ret i64 %3
+}
+
+declare x86_mmx @llvm.x86.mmx.psrl.d(x86_mmx, x86_mmx) nounwind readnone
+
+define i64 @test62(<1 x i64> %a, <1 x i64> %b) #0 {
+; CHECK-LABEL: define i64 @test62(
+; CHECK-SAME: <1 x i64> [[A:%.*]], <1 x i64> [[B:%.*]]) #[[ATTR1]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP9:%.*]] = load <1 x i64>, ptr @__msan_param_tls, align 8
+; CHECK-NEXT:    [[TMP10:%.*]] = load <1 x i64>, ptr inttoptr (i64 add (i64 ptrtoint (ptr @__msan_param_tls to i64), i64 8) to ptr), align 8
+; CHECK-NEXT:    call void @llvm.donothing()
+; CHECK-NEXT:    [[TMP11:%.*]] = bitcast <1 x i64> [[TMP9]] to <2 x i32>
+; CHECK-NEXT:    [[TMP0:%.*]] = bitcast <1 x i64> [[A]] to <2 x i32>
+; CHECK-NEXT:    [[TMP12:%.*]] = bitcast <2 x i32> [[TMP11]] to i64
+; CHECK-NEXT:    [[MMX_VAR_I:%.*]] = bitcast <2 x i32> [[TMP0]] to x86_mmx
+; CHECK-NEXT:    [[_MSPROP:%.*]] = extractelement <1 x i64> [[TMP10]], i32 0
+; CHECK-NEXT:    [[TMP1:%.*]] = extractelement <1 x i64> [[B]], i32 0
+; CHECK-NEXT:    [[MMX_VAR1_I:%.*]] = bitcast i64 [[TMP1]] to x86_mmx
+; CHECK-NEXT:    [[TMP13:%.*]] = icmp ne i64 [[_MSPROP]], 0
+; CHECK-NEXT:    [[TMP14:%.*]] = sext i1 [[TMP13]] to i64
+; CHECK-NEXT:    [[TMP8:%.*]] = bitcast i64 [[TMP12]] to x86_mmx
+; CHECK-NEXT:    [[TMP2:%.*]] = call x86_mmx @llvm.x86.mmx.psrl.d(x86_mmx [[TMP8]], x86_mmx [[MMX_VAR1_I]])
+; CHECK-NEXT:    [[TMP3:%.*]] = bitcast x86_mmx [[TMP2]] to i64
+; CHECK-NEXT:    [[TMP15:%.*]] = or i64 [[TMP3]], [[TMP14]]
+; CHECK-NEXT:    [[TMP4:%.*]] = tail call x86_mmx @llvm.x86.mmx.psrl.d(x86_mmx [[MMX_VAR_I]], x86_mmx [[MMX_VAR1_I]]) #[[ATTR2]]
+; CHECK-NEXT:    [[TMP18:%.*]] = bitcast i64 [[TMP15]] to <2 x i32>
+; CHECK-NEXT:    [[TMP5:%.*]] = bitcast x86_mmx [[TMP4]] to <2 x i32>
+; CHECK-NEXT:    [[TMP6:%.*]] = bitcast <2 x i32> [[TMP18]] to <1 x i64>
+; CHECK-NEXT:    [[TMP16:%.*]] = bitcast <2 x i32> [[TMP5]] to <1 x i64>
+; CHECK-NEXT:    [[TMP7:%.*]] = extractelement <1 x i64> [[TMP6]], i32 0
+; CHECK-NEXT:    [[TMP17:%.*]] = extractelement <1 x i64> [[TMP16]], i32 0
+; CHECK-NEXT:    store i64 [[TMP7]], ptr @__msan_retval_tls, align 8
+; CHECK-NEXT:    ret i64 [[TMP17]]
+;
+entry:
+  %0 = bitcast <1 x i64> %a to <2 x i32>
+  %mmx_var.i = bitcast <2 x i32> %0 to x86_mmx
+  %1 = extractelement <1 x i64> %b, i32 0
+  %mmx_var1.i = bitcast i64 %1 to x86_mmx
+  %2 = tail call x86_mmx @llvm.x86.mmx.psrl.d(x86_mmx %mmx_var.i, x86_mmx %mmx_var1.i) nounwind
+  %3 = bitcast x86_mmx %2 to <2 x i32>
+  %4 = bitcast <2 x i32> %3 to <1 x i64>
+  %5 = extractelement <1 x i64> %4, i32 0
+  ret i64 %5
+}
+
+declare x86_mmx @llvm.x86.mmx.psrl.w(x86_mmx, x86_mmx) nounwind readnone
+
+define i64 @test61(<1 x i64> %a, <1 x i64> %b) #0 {
+; CHECK-LABEL: define i64 @test61(
+; CHECK-SAME: <1 x i64> [[A:%.*]], <1 x i64> [[B:%.*]]) #[[ATTR1]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP9:%.*]] = load <1 x i64>, ptr @__msan_param_tls, align 8
+; CHECK-NEXT:    [[TMP10:%.*]] = load <1 x i64>, ptr inttoptr (i64 add (i64 ptrtoint (ptr @__msan_param_tls to i64), i64 8) to ptr), align 8
+; CHECK-NEXT:    call void @llvm.donothing()
+; CHECK-NEXT:    [[TMP11:%.*]] = bitcast <1 x i64> [[TMP9]] to <4 x i16>
+; CHECK-NEXT:    [[TMP0:%.*]] = bitcast <1 x i64> [[A]] to <4 x i16>
+; CHECK-NEXT:    [[TMP12:%.*]] = bitcast <4 x i16> [[TMP11]] to i64
+; CHECK-NEXT:    [[MMX_VAR_I:%.*]] = bitcast <4 x i16> [[TMP0]] to x86_mmx
+; CHECK-NEXT:    [[_MSPROP:%.*]] = extractelement <1 x i64> [[TMP10]], i32 0
+; CHECK-NEXT:    [[TMP1:%.*]] = extractelement <1 x i64> [[B]], i32 0
+; CHECK-NEXT:    [[MMX_VAR1_I:%.*]] = bitcast i64 [[TMP1]] to x86_mmx
+; CHECK-NEXT:    [[TMP13:%.*]] = icmp ne i64 [[_MSPROP]], 0
+; CHECK-NEXT:    [[TMP14:%.*]] = sext i1 [[TMP13]] to i64
+; CHECK-NEXT:    [[TMP8:%.*]] = bitcast i64 [[TMP12]] to x86_mmx
+; CHECK-NEXT:    [[TMP2:%.*]] = call x86_mmx @llvm.x86.mmx.psrl.w(x86_mmx [[TMP8]], x86_mmx [[MMX_VAR1_I]])
+; CHECK-NEXT:    [[TMP3:%.*]] = bitcast x86_mmx [[TMP2]] to i64
+; CHECK-NEXT:    [[TMP15:%.*]] = or i64 [[TMP3]], [[TMP14]]
+; CHECK-NEXT:    [[TMP4:%.*]] = tail call x86_mmx @llvm.x86.mmx.psrl.w(x86_mmx [[MMX_VAR_I]], x86_mmx [[MMX_VAR1_I]]) #[[ATTR2]]
+; CHECK-NEXT:    [[TMP18:%.*]] = bitcast i64 [[TMP15]] to <4 x i16>
+; CHECK-NEXT:    [[TMP5:%.*]] = bitcast x86_mmx [[TMP4]] to <4 x i16>
+; CHECK-NEXT:    [[TMP6:%.*]] = bitcast <4 x i16> [[TMP18]] to <1 x i64>
+; CHECK-NEXT:    [[TMP16:%.*]] = bitcast <4 x i16> [[TMP5]] to <1 x i64>
+; CHECK-NEXT:    [[TMP7:%.*]] = extractelement <1 x i64> [[TMP6]], i32 0
+; CHECK-NEXT:    [[TMP17:%.*]] = extractelement <1 x i64> [[TMP16]], i32 0
+; CHECK-NEXT:    store i64 [[TMP7]], ptr @__msan_retval_tls, align 8
+; CHECK-NEXT:    ret i64 [[TMP17]]
+;
+entry:
+  %0 = bitcast <1 x i64> %a to <4 x i16>
+  %mmx_var.i = bitcast <4 x i16> %0 to x86_mmx
+  %1 = extractelement <1 x i64> %b, i32 0
+  %mmx_var1.i = bitcast i64 %1 to x86_mmx
+  %2 = tail call x86_mmx @llvm.x86.mmx.psrl.w(x86_mmx %mmx_var.i, x86_mmx %mmx_var1.i) nounwind
+  %3 = bitcast x86_mmx %2 to <4 x i16>
+  %4 = bitcast <4 x i16> %3 to <1 x i64>
+  %5 = extractelement <1 x i64> %4, i32 0
+  ret i64 %5
+}
+
+declare x86_mmx @llvm.x86.mmx.psll.q(x86_mmx, x86_mmx) nounwind readnone
+
+define i64 @test60(<1 x i64> %a, <1 x i64> %b) #0 {
+; CHECK-LABEL: define i64 @test60(
+; CHECK-SAME: <1 x i64> [[A:%.*]], <1 x i64> [[B:%.*]]) #[[ATTR1]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP7:%.*]] = load <1 x i64>, ptr @__msan_param_tls, align 8
+; CHECK-NEXT:    [[TMP8:%.*]] = load <1 x i64>, ptr inttoptr (i64 add (i64 ptrtoint (ptr @__msan_param_tls to i64), i64 8) to ptr), align 8
+; CHECK-NEXT:    call void @llvm.donothing()
+; CHECK-NEXT:    [[_MSPROP:%.*]] = extractelement <1 x i64> [[TMP7]], i32 0
+; CHECK-NEXT:    [[TMP0:%.*]] = extractelement <1 x i64> [[A]], i32 0
+; CHECK-NEXT:    [[MMX_VAR_I:%.*]] = bitcast i64 [[TMP0]] to x86_mmx
+; CHECK-NEXT:    [[_MSPROP1:%.*]] = extractelement <1 x i64> [[TMP8]], i32 0
+; CHECK-NEXT:    [[TMP1:%.*]] = extractelement <1 x i64> [[B]], i32 0
+; CHECK-NEXT:    [[MMX_VAR1_I:%.*]] = bitcast i64 [[TMP1]] to x86_mmx
+; CHECK-NEXT:    [[TMP9:%.*]] = icmp ne i64 [[_MSPROP1]], 0
+; CHECK-NEXT:    [[TMP10:%.*]] = sext i1 [[TMP9]] to i64
+; CHECK-NEXT:    [[TMP6:%.*]] = bitcast i64 [[_MSPROP]] to x86_mmx
+; CHECK-NEXT:    [[TMP2:%.*]] = call x86_mmx @llvm.x86.mmx.psll.q(x86_mmx [[TMP6]], x86_mmx [[MMX_VAR1_I]])
+; CHECK-NEXT:    [[TMP3:%.*]] = bitcast x86_mmx [[TMP2]] to i64
+; CHECK-NEXT:    [[TMP11:%.*]] = or i64 [[TMP3]], [[TMP10]]
+; CHECK-NEXT:    [[TMP4:%.*]] = tail call x86_mmx @llvm.x86.mmx.psll.q(x86_mmx [[MMX_VAR_I]], x86_mmx [[MMX_VAR1_I]]) #[[ATTR2]]
+; CHECK-NEXT:    [[TMP5:%.*]] = bitcast x86_mmx [[TMP4]] to i64
+; CHECK-NEXT:    store i64 [[TMP11]], ptr @__msan_retval_tls, align 8
+; CHECK-NEXT:    ret i64 [[TMP5]]
+;
+entry:
+  %0 = extractelement <1 x i64> %a, i32 0
+  %mmx_var.i = bitcast i64 %0 to x86_mmx
+  %1 = extractelement <1 x i64> %b, i32 0
+  %mmx_var1.i = bitcast i64 %1 to x86_mmx
+  %2 = tail call x86_mmx @llvm.x86.mmx.psll.q(x86_mmx %mmx_var.i, x86_mmx %mmx_var1.i) nounwind
+  %3 = bitcast x86_mmx %2 to i64
+  ret i64 %3
+}
+
+declare x86_mmx @llvm.x86.mmx.psll.d(x86_mmx, x86_mmx) nounwind readnone
+
+define i64 @test59(<1 x i64> %a, <1 x i64> %b) #0 {
+; CHECK-LABEL: define i64 @test59(
+; CHECK-SAME: <1 x i64> [[A:%.*]], <1 x i64> [[B:%.*]]) #[[ATTR1]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP9:%.*]] = load <1 x i64>, ptr @__msan_param_tls, align 8
+; CHECK-NEXT:    [[TMP10:%.*]] = load <1 x i64>, ptr inttoptr (i64 add (i64 ptrtoint (ptr @__msan_param_tls to i64), i64 8) to ptr), align 8
+; CHECK-NEXT:    call void @llvm.donothing()
+; CHECK-NEXT:    [[TMP11:%.*]] = bitcast <1 x i64> [[TMP9]] to <2 x i32>
+; CHECK-NEXT:    [[TMP0:%.*]] = bitcast <1 x i64> [[A]] to <2 x i32>
+; CHECK-NEXT:    [[TMP12:%.*]] = bitcast <2 x i32> [[TMP11]] to i64
+; CHECK-NEXT:    [[MMX_VAR_I:%.*]] = bitcast <2 x i32> [[TMP0]] to x86_mmx
+; CHECK-NEXT:    [[_MSPROP:%.*]] = extractelement <1 x i64> [[TMP10]], i32 0
+; CHECK-NEXT:    [[TMP1:%.*]] = extractelement <1 x i64> [[B]], i32 0
+; CHECK-NEXT:    [[MMX_VAR1_I:%.*]] = bitcast i64 [[TMP1]] to x86_mmx
+; CHECK-NEXT:    [[TMP13:%.*]] = icmp ne i64 [[_MSPROP]], 0
+; CHECK-NEXT:    [[TMP14:%.*]] = sext i1 [[TMP13]] to i64
+; CHECK-NEXT:    [[TMP8:%.*]] = bitcast i64 [[TMP12]] to x86_mmx
+; CHECK-NEXT:    [[TMP2:%.*]] = call x86_mmx @llvm.x86.mmx.psll.d(x86_mmx [[TMP8]], x86_mmx [[MMX_VAR1_I]])
+; CHECK-NEXT:    [[TMP3:%.*]] = bitcast x86_mmx [[TMP2]] to i64
+; CHECK-NEXT:    [[TMP15:%.*]] = or i64 [[TMP3]], [[TMP14]]
+; CHECK-NEXT:    [[TMP4:%.*]] = tail call x86_mmx @llvm.x86.mmx.psll.d(x86_mmx [[MMX_VAR_I]], x86_mmx [[MMX_VAR1_I]]) #[[ATTR2]]
+; CHECK-NEXT:    [[TMP18:%.*]] = bitcast i64 [[TMP15]] to <2 x i32>
+; CHECK-NEXT:    [[TMP5:%.*]] = bitcast x86_mmx [[TMP4]] to <2 x i32>
+; CHECK-NEXT:    [[TMP6:%.*]] = bitcast <2 x i32> [[TMP18]] to <1 x i64>
+; CHECK-NEXT:    [[TMP16:%.*]] = bitcast <2 x i32> [[TMP5]] to <1 x i64>
+; CHECK-NEXT:    [[TMP7:%.*]] = extractelement <1 x i64> [[TMP6]], i32 0
+; CHECK-NEXT:    [[TMP17:%.*]] = extractelement <1 x i64> [[TMP16]], i32 0
+; CHECK-NEXT:    store i64 [[TMP7]], ptr @__msan_retval_tls, align 8
+; CHECK-NEXT:    ret i64 [[TMP17]]
+;
+entry:
+  %0 = bitcast <1 x i64> %a to <2 x i32>
+  %mmx_var.i = bitcast <2 x i32> %0 to x86_mmx
+  %1 = extractelement <1 x i64> %b, i32 0
+  %mmx_var1.i = bitcast i64 %1 to x86_mmx
+  %2 = tail call x86_mmx @llvm.x86.mmx.psll.d(x86_mmx %mmx_var.i, x86_mmx %mmx_var1.i) nounwind
+  %3 = bitcast x86_mmx %2 to <2 x i32>
+  %4 = bitcast <2 x i32> %3 to <1 x i64>
+  %5 = extractelement <1 x i64> %4, i32 0
+  ret i64 %5
+}
+
+declare x86_mmx @llvm.x86.mmx.psll.w(x86_mmx, x86_mmx) nounwind readnone
+
+define i64 @test58(<1 x i64> %a, <1 x i64> %b) #0 {
+; CHECK-LABEL: define i64 @test58(
+; CHECK-SAME: <1 x i64> [[A:%.*]], <1 x i64> [[B:%.*]]) #[[ATTR1]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP9:%.*]] = load <1 x i64>, ptr @__msan_param_tls, align 8
+; CHECK-NEXT:    [[TMP10:%.*]] = load <1 x i64>, ptr inttoptr (i64 add (i64 ptrtoint (ptr @__msan_param_tls to i64), i64 8) to ptr), align 8
+; CHECK-NEXT:    call void @llvm.donothing()
+; CHECK-NEXT:    [[TMP11:%.*]] = bitcast <1 x i64> [[TMP9]] to <4 x i16>
+; CHECK-NEXT:    [[TMP0:%.*]] = bitcast <1 x i64> [[A]] to <4 x i16>
+; CHECK-NEXT:    [[TMP12:%.*]] = bitcast <4 x i16> [[TMP11]] to i64
+; CHECK-NEXT:    [[MMX_VAR_I:%.*]] = bitcast <4 x i16> [[TMP0]] to x86_mmx
+; CHECK-NEXT:    [[_MSPROP:%.*]] = extractelement <1 x i64> [[TMP10]], i32 0
+; CHECK-NEXT:    [[TMP1:%.*]] = extractelement <1 x i64> [[B]], i32 0
+; CHECK-NEXT:    [[MMX_VAR1_I:%.*]] = bitcast i64 [[TMP1]] to x86_mmx
+; CHECK-NEXT:    [[TMP13:%.*]] = icmp ne i64 [[_MSPROP]], 0
+; CHECK-NEXT:    [[TMP14:%.*]] = sext i1 [[TMP13]] to i64
+; CHECK-NEXT:    [[TMP8:%.*]] = bitcast i64 [[TMP12]] to x86_mmx
+; CHECK-NEXT:    [[TMP2:%.*]] = call x86_mmx @llvm.x86.mmx.psll.w(x86_mmx [[TMP8]], x86_mmx [[MMX_VAR1_I]])
+; CHECK-NEXT:    [[TMP3:%.*]] = bitcast x86_mmx [[TMP2]] to i64
+; CHECK-NEXT:    [[TMP15:%.*]] = or i64 [[TMP3]], [[TMP14]]
+; CHECK-NEXT:    [[TMP4:%.*]] = tail call x86_mmx @llvm.x86.mmx.psll.w(x86_mmx [[MMX_VAR_I]], x86_mmx [[MMX_VAR1_I]]) #[[ATTR2]]
+; CHECK-NEXT:    [[TMP18:%.*]] = bitcast i64 [[TMP15]] to <4 x i16>
+; CHECK-NEXT:    [[TMP5:%.*]] = bitcast x86_mmx [[TMP4]] to <4 x i16>
+; CHECK-NEXT:    [[TMP6:%.*]] = bitcast <4 x i16> [[TMP18]] to <1 x i64>
+; CHECK-NEXT:    [[TMP16:%.*]] = bitcast <4 x i16> [[TMP5]] to <1 x i64>
+; CHECK-NEXT:    [[TMP7:%.*]] = extractelement <1 x i64> [[TMP6]], i32 0
+; CHECK-NEXT:    [[TMP17:%.*]] = extractelement <1 x i64> [[TMP16]], i32 0
+; CHECK-NEXT:    store i64 [[TMP7]], ptr @__msan_retval_tls, align 8
+; CHECK-NEXT:    ret i64 [[TMP17]]
+;
+entry:
+  %0 = bitcast <1 x i64> %a to <4 x i16>
+  %mmx_var.i = bitcast <4 x i16> %0 to x86_mmx
+  %1 = extractelement <1 x i64> %b, i32 0
+  %mmx_var1.i = bitcast i64 %1 to x86_mmx
+  %2 = tail call x86_mmx @llvm.x86.mmx.psll.w(x86_mmx %mmx_var.i, x86_mmx %mmx_var1.i) nounwind
+  %3 = bitcast x86_mmx %2 to <4 x i16>
+  %4 = bitcast <4 x i16> %3 to <1 x i64>
+  %5 = extractelement <1 x i64> %4, i32 0
+  ret i64 %5
+}
+
+declare x86_mmx @llvm.x86.mmx.pxor(x86_mmx, x86_mmx) nounwind readnone
+
+define i64 @test56(<1 x i64> %a, <1 x i64> %b) #0 {
+; CHECK-LABEL: define i64 @test56(
+; CHECK-SAME: <1 x i64> [[A:%.*]], <1 x i64> [[B:%.*]]) #[[ATTR1]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP8:%.*]] = load <1 x i64>, ptr inttoptr (i64 add (i64 ptrtoint (ptr @__msan_param_tls to i64), i64 8) to ptr), align 8
+; CHECK-NEXT:    [[TMP10:%.*]] = load <1 x i64>, ptr @__msan_param_tls, align 8
+; CHECK-NEXT:    call void @llvm.donothing()
+; CHECK-NEXT:    [[TMP11:%.*]] = bitcast <1 x i64> [[TMP8]] to <2 x i32>
+; CHECK-NEXT:    [[TMP0:%.*]] = bitcast <1 x i64> [[B]] to <2 x i32>
+; CHECK-NEXT:    [[TMP14:%.*]] = bitcast <1 x i64> [[TMP10]] to <2 x i32>
+; CHECK-NEXT:    [[TMP1:%.*]] = bitcast <1 x i64> [[A]] to <2 x i32>
+; CHECK-NEXT:    [[TMP6:%.*]] = bitcast <2 x i32> [[TMP14]] to i64
+; CHECK-NEXT:    [[MMX_VAR_I:%.*]] = bitcast <2 x i32> [[TMP1]] to x86_mmx
+; CHECK-NEXT:    [[TMP7:%.*]] = bitcast <2 x i32> [[TMP11]] to i64
+; CHECK-NEXT:    [[MMX_VAR1_I:%.*]] = bitcast <2 x i32> [[TMP0]] to x86_mmx
+; CHECK-NEXT:    [[_MSPROP:%.*]] = or i64 [[TMP6]], [[TMP7]]
+; CHECK-NEXT:    [[TMP2:%.*]] = tail call x86_mmx @llvm.x86.mmx.pxor(x86_mmx [[MMX_VAR_I]], x86_mmx [[MMX_VAR1_I]]) #[[ATTR2]]
+; CHECK-NEXT:    [[TMP9:%.*]] = bitcast i64 [[_MSPROP]] to <2 x i32>
+; CHECK-NEXT:    [[TMP3:%.*]] = bitcast x86_mmx [[TMP2]] to <2 x i32>
+; CHECK-NEXT:    [[TMP4:%.*]] = bitcast <2 x i32> [[TMP9]] to <1 x i64>
+; CHECK-NEXT:    [[TMP12:%.*]] = bitcast <2 x i32> [[TMP3]] to <1 x i64>
+; CHECK-NEXT:    [[TMP5:%.*]] = extractelement <1 x i64> [[TMP4]], i32 0
+; CHECK-NEXT:    [[TMP13:%.*]] = extractelement <1 x i64> [[TMP12]], i32 0
+; CHECK-NEXT:    store i64 [[TMP5]], ptr @__msan_retval_tls, align 8
+; CHECK-NEXT:    ret i64 [[TMP13]]
+;
+entry:
+  %0 = bitcast <1 x i64> %b to <2 x i32>
+  %1 = bitcast <1 x i64> %a to <2 x i32>
+  %mmx_var.i = bitcast <2 x i32> %1 to x86_mmx
+  %mmx_var1.i = bitcast <2 x i32> %0 to x86_mmx
+  %2 = tail call x86_mmx @llvm.x86.mmx.pxor(x86_mmx %mmx_var.i, x86_mmx %mmx_var1.i) nounwind
+  %3 = bitcast x86_mmx %2 to <2 x i32>
+  %4 = bitcast <2 x i32> %3 to <1 x i64>
+  %5 = extractelement <1 x i64> %4, i32 0
+  ret i64 %5
+}
+
+declare x86_mmx @llvm.x86.mmx.por(x86_mmx, x86_mmx) nounwind readnone
+
+define i64 @test55(<1 x i64> %a, <1 x i64> %b) #0 {
+; CHECK-LABEL: define i64 @test55(
+; CHECK-SAME: <1 x i64> [[A:%.*]], <1 x i64> [[B:%.*]]) #[[ATTR1]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP8:%.*]] = load <1 x i64>, ptr inttoptr (i64 add (i64 ptrtoint (ptr @__msan_param_tls to i64), i64 8) to ptr), align 8
+; CHECK-NEXT:    [[TMP10:%.*]] = load <1 x i64>, ptr @__msan_param_tls, align 8
+; CHECK-NEXT:    call void @llvm.donothing()
+; CHECK-NEXT:    [[TMP11:%.*]] = bitcast <1 x i64> [[TMP8]] to <2 x i32>
+; CHECK-NEXT:    [[TMP0:%.*]] = bitcast <1 x i64> [[B]] to <2 x i32>
+; CHECK-NEXT:    [[TMP14:%.*]] = bitcast <1 x i64> [[TMP10]] to <2 x i32>
+; CHECK-NEXT:    [[TMP1:%.*]] = bitcast <1 x i64> [[A]] to <2 x i32>
+; CHECK-NEXT:    [[TMP6:%.*]] = bitcast <2 x i32> [[TMP14]] to i64
+; CHECK-NEXT:    [[MMX_VAR_I:%.*]] = bitcast <2 x i32> [[TMP1]] to x86_mmx
+; CHECK-NEXT:    [[TMP7:%.*]] = bitcast <2 x i32> [[TMP11]] to i64
+; CHECK-NEXT:    [[MMX_VAR1_I:%.*]] = bitcast <2 x i32> [[TMP0]] to x86_mmx
+; CHECK-NEXT:    [[_MSPROP:%.*]] = or i64 [[TMP6]], [[TMP7]]
+; CHECK-NEXT:    [[TMP2:%.*]] = tail call x86_mmx @llvm.x86.mmx.por(x86_mmx [[MMX_VAR_I]], x86_mmx [[MMX_VAR1_I]]) #[[ATTR2]]
+; CHECK-NEXT:    [[TMP9:%.*]] = bitcast i64 [[_MSPROP]] to <2 x i32>
+; CHECK-NEXT:    [[TMP3:%.*]] = bitcast x86_mmx [[TMP2]] to <2 x i32>
+; CHECK-NEXT:    [[TMP4:%.*]] = bitcast <2 x i32> [[TMP9]] to <1 x i64>
+; CHECK-NEXT:    [[TMP12:%.*]] = bitcast <2 x i32> [[TMP3]] to <1 x i64>
+; CHECK-NEXT:    [[TMP5:%.*]] = extractelement <1 x i64> [[TMP4]], i32 0
+; CHECK-NEXT:    [[TMP13:%.*]] = extractelement <1 x i64> [[TMP12]], i32 0
+; CHECK-NEXT:    store i64 [[TMP5]], ptr @__msan_retval_tls, align 8
+; CHECK-NEXT:    ret i64 [[TMP13]]
+;
+entry:
+  %0 = bitcast <1 x i64> %b to <2 x i32>
+  %1 = bitcast <1 x i64> %a to <2 x i32>
+  %mmx_var.i = bitcast <2 x i32> %1 to x86_mmx
+  %mmx_var1.i = bitcast <2 x i32> %0 to x86_mmx
+  %2 = tail call x86_mmx @llvm.x86.mmx.por(x86_mmx %mmx_var.i, x86_mmx %mmx_var1.i) nounwind
+  %3 = bitcast x86_mmx %2 to <2 x i32>
+  %4 = bitcast <2 x i32> %3 to <1 x i64>
+  %5 = extractelement <1 x i64> %4, i32 0
+  ret i64 %5
+}
+
+declare x86_mmx @llvm.x86.mmx.pandn(x86_mmx, x86_mmx) nounwind readnone
+
+define i64 @test54(<1 x i64> %a, <1 x i64> %b) #0 {
+; CHECK-LABEL: define i64 @test54(
+; CHECK-SAME: <1 x i64> [[A:%.*]], <1 x i64> [[B:%.*]]) #[[ATTR1]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP8:%.*]] = load <1 x i64>, ptr inttoptr (i64 add (i64 ptrtoint (ptr @__msan_param_tls to i64), i64 8) to ptr), align 8
+; CHECK-NEXT:    [[TMP10:%.*]] = load <1 x i64>, ptr @__msan_param_tls, align 8
+; CHECK-NEXT:    call void @llvm.donothing()
+; CHECK-NEXT:    [[TMP11:%.*]] = bitcast <1 x i64> [[TMP8]] to <2 x i32>
+; CHECK-NEXT:    [[TMP0:%.*]] = bitcast <1 x i64> [[B]] to <2 x i32>
+; CHECK-NEXT:    [[TMP14:%.*]] = bitcast <1 x i64> [[TMP10]] to <2 x i32>
+; CHECK-NEXT:    [[TMP1:%.*]] = bitcast <1 x i64> [[A]] to <2 x i32>
+; CHECK-NEXT:    [[TMP6:%.*]] = bitcast <2 x i32> [[TMP14]] to i64
+; CHECK-NEXT:    [[MMX_VAR_I:%.*]] = bitcast <2 x i32> [[TMP1]] to x86_mmx
+; CHECK-NEXT:    [[TMP7:%.*]] = bitcast <2 x i32> [[TMP11]] to i64
+; CHECK-NEXT:    [[MMX_VAR1_I:%.*]] = bitcast <2 x i32> [[TMP0]] to x86_mmx
+; CHECK-NEXT:    [[_MSPROP:%.*]] = or i64 [[TMP6]], [[TMP7]]
+; CHECK-NEXT:    [[TMP2:%.*]] = tail call x86_mmx @llvm.x86.mmx.pandn(x86_mmx [[MMX_VAR_I]], x86_mmx [[MMX_VAR1_I]]) #[[ATTR2]]
+; CHECK-NEXT:    [[TMP9:%.*]] = bitcast i64 [[_MSPROP]] to <2 x i32>
+; CHECK-NEXT:    [[TMP3:%.*]] = bitcast x86_mmx [[TMP2]] to <2 x i32>
+; CHECK-NEXT:    [[TMP4:%.*]] = bitcast <2 x i32> [[TMP9]] to <1 x i64>
+; CHECK-NEXT:    [[TMP12:%.*]] = bitcast <2 x i32> [[TMP3]] to <1 x i64>
+; CHECK-NEXT:    [[TMP5:%.*]] = extractelement <1 x i64> [[TMP4]], i32 0
+; CHECK-NEXT:    [[TMP13:%.*]] = extractelement <1 x i64> [[TMP12]], i32 0
+; CHECK-NEXT:    store i64 [[TMP5]], ptr @__msan_retval_tls, align 8
+; CHECK-NEXT:    ret i64 [[TMP13]]
+;
+entry:
+  %0 = bitcast <1 x i64> %b to <2 x i32>
+  %1 = bitcast <1 x i64> %a to <2 x i32>
+  %mmx_var.i = bitcast <2 x i32> %1 to x86_mmx
+  %mmx_var1.i = bitcast <2 x i32> %0 to x86_mmx
+  %2 = tail call x86_mmx @llvm.x86.mmx.pandn(x86_mmx %mmx_var.i, x86_mmx %mmx_var1.i) nounwind
+  %3 = bitcast x86_mmx %2 to <2 x i32>
+  %4 = bitcast <2 x i32> %3 to <1 x i64>
+  %5 = extractelement <1 x i64> %4, i32 0
+  ret i64 %5
+}
+
+declare x86_mmx @llvm.x86.mmx.pand(x86_mmx, x86_mmx) nounwind readnone
+
+define i64 @test53(<1 x i64> %a, <1 x i64> %b) #0 {
+; CHECK-LABEL: define i64 @test53(
+; CHECK-SAME: <1 x i64> [[A:%.*]], <1 x i64> [[B:%.*]]) #[[ATTR1]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP8:%.*]] = load <1 x i64>, ptr inttoptr (i64 add (i64 ptrtoint (ptr @__msan_param_tls to i64), i64 8) to ptr), align 8
+; CHECK-NEXT:    [[TMP10:%.*]] = load <1 x i64>, ptr @__msan_param_tls, align 8
+; CHECK-NEXT:    call void @llvm.donothing()
+; CHECK-NEXT:    [[TMP11:%.*]] = bitcast <1 x i64> [[TMP8]] to <2 x i32>
+; CHECK-NEXT:    [[TMP0:%.*]] = bitcast <1 x i64> [[B]] to <2 x i32>
+; CHECK-NEXT:    [[TMP14:%.*]] = bitcast <1 x i64> [[TMP10]] to <2 x i32>
+; CHECK-NEXT:    [[TMP1:%.*]] = bitcast <1 x i64> [[A]] to <2 x i32>
+; CHECK-NEXT:    [[TMP6:%.*]] = bitcast <2 x i32> [[TMP14]] to i64
+; CHECK-NEXT:    [[MMX_VAR_I:%.*]] = bitcast <2 x i32> [[TMP1]] to x86_mmx
+; CHECK-NEXT:    [[TMP7:%.*]] = bitcast <2 x i32> [[TMP11]] to i64
+; CHECK-NEXT:    [[MMX_VAR1_I:%.*]] = bitcast <2 x i32> [[TMP0]] to x86_mmx
+; CHECK-NEXT:    [[_MSPROP:%.*]] = or i64 [[TMP6]], [[TMP7]]
+; CHECK-NEXT:    [[TMP2:%.*]] = tail call x86_mmx @llvm.x86.mmx.pand(x86_mmx [[MMX_VAR_I]], x86_mmx [[MMX_VAR1_I]]) #[[ATTR2]]
+; CHECK-NEXT:    [[TMP9:%.*]] = bitcast i64 [[_MSPROP]] to <2 x i32>
+; CHECK-NEXT:    [[TMP3:%.*]] = bitcast x86_mmx [[TMP2]] to <2 x i32>
+; CHECK-NEXT:    [[TMP4:%.*]] = bitcast <2 x i32> [[TMP9]] to <1 x i64>
+; CHECK-NEXT:    [[TMP12:%.*]] = bitcast <2 x i32> [[TMP3]] to <1 x i64>
+; CHECK-NEXT:    [[TMP5:%.*]] = extractelement <1 x i64> [[TMP4]], i32 0
+; CHECK-NEXT:    [[TMP13:%.*]] = extractelement <1 x i64> [[TMP12]], i32 0
+; CHECK-NEXT:    store i64 [[TMP5]], ptr @__msan_retval_tls, align 8
+; CHECK-NEXT:    ret i64 [[TMP13]]
+;
+entry:
+  %0 = bitcast <1 x i64> %b to <2 x i32>
+  %1 = bitcast <1 x i64> %a to <2 x i32>
+  %mmx_var.i = bitcast <2 x i32> %1 to x86_mmx
+  %mmx_var1.i = bitcast <2 x i32> %0 to x86_mmx
+  %2 = tail call x86_mmx @llvm.x86.mmx.pand(x86_mmx %mmx_var.i, x86_mmx %mmx_var1.i) nounwind
+  %3 = bitcast x86_mmx %2 to <2 x i32>
+  %4 = bitcast <2 x i32> %3 to <1 x i64>
+  %5 = extractelement <1 x i64> %4, i32 0
+  ret i64 %5
+}
+
+declare x86_mmx @llvm.x86.mmx.pmull.w(x86_mmx, x86_mmx) nounwind readnone
+
+define i64 @test52(<1 x i64> %a, <1 x i64> %b) #0 {
+; CHECK-LABEL: define i64 @test52(
+; CHECK-SAME: <1 x i64> [[A:%.*]], <1 x i64> [[B:%.*]]) #[[ATTR1]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP8:%.*]] = load <1 x i64>, ptr inttoptr (i64 add (i64 ptrtoint (ptr @__msan_param_tls to i64), i64 8) to ptr), align 8
+; CHECK-NEXT:    [[TMP10:%.*]] = load <1 x i64>, ptr @__msan_param_tls, align 8
+; CHECK-NEXT:    call void @llvm.donothing()
+; CHECK-NEXT:    [[TMP11:%.*]] = bitcast <1 x i64> [[TMP8]] to <4 x i16>
+; CHECK-NEXT:    [[TMP0:%.*]] = bitcast <1 x i64> [[B]] to <4 x i16>
+; CHECK-NEXT:    [[TMP14:%.*]] = bitcast <1 x i64> [[TMP10]] to <4 x i16>
+; CHECK-NEXT:    [[TMP1:%.*]] = bitcast <1 x i64> [[A]] to <4 x i16>
+; CHECK-NEXT:    [[TMP6:%.*]] = bitcast <4 x i16> [[TMP14]] to i64
+; CHECK-NEXT:    [[MMX_VAR_I:%.*]] = bitcast <4 x i16> [[TMP1]] to x86_mmx
+; CHECK-NEXT:    [[TMP7:%.*]] = bitcast <4 x i16> [[TMP11]] to i64
+; CHECK-NEXT:    [[MMX_VAR1_I:%.*]] = bitcast <4 x i16> [[TMP0]] to x86_mmx
+; CHECK-NEXT:    [[_MSPROP:%.*]] = or i64 [[TMP6]], [[TMP7]]
+; CHECK-NEXT:    [[TMP2:%.*]] = tail call x86_mmx @llvm.x86.mmx.pmull.w(x86_mmx [[MMX_VAR_I]], x86_mmx [[MMX_VAR1_I]]) #[[ATTR2]]
+; CHECK-NEXT:    [[TMP9:%.*]] = bitcast i64 [[_MSPROP]] to <4 x i16>
+; CHECK-NEXT:    [[TMP3:%.*]] = bitcast x86_mmx [[TMP2]] to <4 x i16>
+; CHECK-NEXT:    [[TMP4:%.*]] = bitcast <4 x i16> [[TMP9]] to <1 x i64>
+; CHECK-NEXT:    [[TMP12:%.*]] = bitcast <4 x i16> [[TMP3]] to <1 x i64>
+; CHECK-NEXT:    [[TMP5:%.*]] = extractelement <1 x i64> [[TMP4]], i32 0
+; CHECK-NEXT:    [[TMP13:%.*]] = extractelement <1 x i64> [[TMP12]], i32 0
+; CHECK-NEXT:    store i64 [[TMP5]], ptr @__msan_retval_tls, align 8
+; CHECK-NEXT:    ret i64 [[TMP13]]
+;
+entry:
+  %0 = bitcast <1 x i64> %b to <4 x i16>
+  %1 = bitcast <1 x i64> %a to <4 x i16>
+  %mmx_var.i = bitcast <4 x i16> %1 to x86_mmx
+  %mmx_var1.i = bitcast <4 x i16> %0 to x86_mmx
+  %2 = tail call x86_mmx @llvm.x86.mmx.pmull.w(x86_mmx %mmx_var.i, x86_mmx %mmx_var1.i) nounwind
+  %3 = bitcast x86_mmx %2 to <4 x i16>
+  %4 = bitcast <4 x i16> %3 to <1 x i64>
+  %5 = extractelement <1 x i64> %4, i32 0
+  ret i64 %5
+}
+
+define i64 @test51(<1 x i64> %a, <1 x i64> %b) #0 {
+; CHECK-LABEL: define i64 @test51(
+; CHECK-SAME: <1 x i64> [[A:%.*]], <1 x i64> [[B:%.*]]) #[[ATTR1]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP8:%.*]] = load <1 x i64>, ptr inttoptr (i64 add (i64 ptrtoint (ptr @__msan_param_tls to i64), i64 8) to ptr), align 8
+; CHECK-NEXT:    [[TMP10:%.*]] = load <1 x i64>, ptr @__msan_param_tls, align 8
+; CHECK-NEXT:    call void @llvm.donothing()
+; CHECK-NEXT:    [[TMP11:%.*]] = bitcast <1 x i64> [[TMP8]] to <4 x i16>
+; CHECK-NEXT:    [[TMP0:%.*]] = bitcast <1 x i64> [[B]] to <4 x i16>
+; CHECK-NEXT:    [[TMP14:%.*]] = bitcast <1 x i64> [[TMP10]] to <4 x i16>
+; CHECK-NEXT:    [[TMP1:%.*]] = bitcast <1 x i64> [[A]] to <4 x i16>
+; CHECK-NEXT:    [[TMP6:%.*]] = bitcast <4 x i16> [[TMP14]] to i64
+; CHECK-NEXT:    [[MMX_VAR_I:%.*]] = bitcast <4 x i16> [[TMP1]] to x86_mmx
+; CHECK-NEXT:    [[TMP7:%.*]] = bitcast <4 x i16> [[TMP11]] to i64
+; CHECK-NEXT:    [[MMX_VAR1_I:%.*]] = bitcast <4 x i16> [[TMP0]] to x86_mmx
+; CHECK-NEXT:    [[_MSPROP:%.*]] = or i64 [[TMP6]], [[TMP7]]
+; CHECK-NEXT:    [[TMP2:%.*]] = tail call x86_mmx @llvm.x86.mmx.pmull.w(x86_mmx [[MMX_VAR_I]], x86_mmx [[MMX_VAR1_I]]) #[[ATTR2]]
+; CHECK-NEXT:    [[TMP9:%.*]] = bitcast i64 [[_MSPROP]] to <4 x i16>
+; CHECK-NEXT:    [[TMP3:%.*]] = bitcast x86_mmx [[TMP2]] to <4 x i16>
+; CHECK-NEXT:    [[TMP4:%.*]] = bitcast <4 x i16> [[TMP9]] to <1 x i64>
+; CHECK-NEXT:    [[TMP12:%.*]] = bitcast <4 x i16> [[TMP3]] to <1 x i64>
+; CHECK-NEXT:    [[TMP5:%.*]] = extractelement <1 x i64> [[TMP4]], i32 0
+; CHECK-NEXT:    [[TMP13:%.*]] = extractelement <1 x i64> [[TMP12]], i32 0
+; CHECK-NEXT:    store i64 [[TMP5]], ptr @__msan_retval_tls, align 8
+; CHECK-NEXT:    ret i64 [[TMP13]]
+;
+entry:
+  %0 = bitcast <1 x i64> %b to <4 x i16>
+  %1 = bitcast <1 x i64> %a to <4 x i16>
+  %mmx_var.i = bitcast <4 x i16> %1 to x86_mmx
+  %mmx_var1.i = bitcast <4 x i16> %0 to x86_mmx
+  %2 = tail call x86_mmx @llvm.x86.mmx.pmull.w(x86_mmx %mmx_var.i, x86_mmx %mmx_var1.i) nounwind
+  %3 = bitcast x86_mmx %2 to <4 x i16>
+  %4 = bitcast <4 x i16> %3 to <1 x i64>
+  %5 = extractelement <1 x i64> %4, i32 0
+  ret i64 %5
+}
+
+declare x86_mmx @llvm.x86.mmx.pmulh.w(x86_mmx, x86_mmx) nounwind readnone
+
+define i64 @test50(<1 x i64> %a, <1 x i64> %b) #0 {
+; CHECK-LABEL: define i64 @test50(
+; CHECK-SAME: <1 x i64> [[A:%.*]], <1 x i64> [[B:%.*]]) #[[ATTR1]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP8:%.*]] = load <1 x i64>, ptr inttoptr (i64 add (i64 ptrtoint (ptr @__msan_param_tls to i64), i64 8) to ptr), align 8
+; CHECK-NEXT:    [[TMP10:%.*]] = load <1 x i64>, ptr @__msan_param_tls, align 8
+; CHECK-NEXT:    call void @llvm.donothing()
+; CHECK-NEXT:    [[TMP11:%.*]] = bitcast <1 x i64> [[TMP8]] to <4 x i16>
+; CHECK-NEXT:    [[TMP0:%.*]] = bitcast <1 x i64> [[B]] to <4 x i16>
+; CHECK-NEXT:    [[TMP14:%.*]] = bitcast <1 x i64> [[TMP10]] to <4 x i16>
+; CHECK-NEXT:    [[TMP1:%.*]] = bitcast <1 x i64> [[A]] to <4 x i16>
+; CHECK-NEXT:    [[TMP6:%.*]] = bitcast <4 x i16> [[TMP14]] to i64
+; CHECK-NEXT:    [[MMX_VAR_I:%.*]] = bitcast <4 x i16> [[TMP1]] to x86_mmx
+; CHECK-NEXT:    [[TMP7:%.*]] = bitcast <4 x i16> [[TMP11]] to i64
+; CHECK-NEXT:    [[MMX_VAR1_I:%.*]] = bitcast <4 x i16> [[TMP0]] to x86_mmx
+; CHECK-NEXT:    [[_MSPROP:%.*]] = or i64 [[TMP6]], [[TMP7]]
+; CHECK-NEXT:    [[TMP2:%.*]] = tail call x86_mmx @llvm.x86.mmx.pmulh.w(x86_mmx [[MMX_VAR_I]], x86_mmx [[MMX_VAR1_I]]) #[[ATTR2]]
+; CHECK-NEXT:    [[TMP9:%.*]] = bitcast i64 [[_MSPROP]] to <4 x i16>
+; CHECK-NEXT:    [[TMP3:%.*]] = bitcast x86_mmx [[TMP2]] to <4 x i16>
+; CHECK-NEXT:    [[TMP4:%.*]] = bitcast <4 x i16> [[TMP9]] to <1 x i64>
+; CHECK-NEXT:    [[TMP12:%.*]] = bitcast <4 x i16> [[TMP3]] to <1 x i64>
+; CHECK-NEXT:    [[TMP5:%.*]] = extractelement <1 x i64> [[TMP4]], i32 0
+; CHECK-NEXT:    [[TMP13:%.*]] = extractelement <1 x i64> [[TMP12]], i32 0
+; CHECK-NEXT:    store i64 [[TMP5]], ptr @__msan_retval_tls, align 8
+; CHECK-NEXT:    ret i64 [[TMP13]]
+;
+entry:
+  %0 = bitcast <1 x i64> %b to <4 x i16>
+  %1 = bitcast <1 x i64> %a to <4 x i16>
+  %mmx_var.i = bitcast <4 x i16> %1 to x86_mmx
+  %mmx_var1.i = bitcast <4 x i16> %0 to x86_mmx
+  %2 = tail call x86_mmx @llvm.x86.mmx.pmulh.w(x86_mmx %mmx_var.i, x86_mmx %mmx_var1.i) nounwind
+  %3 = bitcast x86_mmx %2 to <4 x i16>
+  %4 = bitcast <4 x i16> %3 to <1 x i64>
+  %5 = extractelement <1 x i64> %4, i32 0
+  ret i64 %5
+}
+
+declare x86_mmx @llvm.x86.mmx.pmadd.wd(x86_mmx, x86_mmx) nounwind readnone
+
+define i64 @test49(<1 x i64> %a, <1 x i64> %b) #0 {
+; CHECK-LABEL: define i64 @test49(
+; CHECK-SAME: <1 x i64> [[A:%.*]], <1 x i64> [[B:%.*]]) #[[ATTR1]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP13:%.*]] = load <1 x i64>, ptr inttoptr (i64 add (i64 ptrtoint (ptr @__msan_param_tls to i64), i64 8) to ptr), align 8
+; CHECK-NEXT:    [[TMP15:%.*]] = load <1 x i64>, ptr @__msan_param_tls, align 8
+; CHECK-NEXT:    call void @llvm.donothing()
+; CHECK-NEXT:    [[TMP16:%.*]] = bitcast <1 x i64> [[TMP13]] to <4 x i16>
+; CHECK-NEXT:    [[TMP0:%.*]] = bitcast <1 x i64> [[B]] to <4 x i16>
+; CHECK-NEXT:    [[TMP19:%.*]] = bitcast <1 x i64> [[TMP15]] to <4 x i16>
+; CHECK-NEXT:    [[TMP1:%.*]] = bitcast <1 x i64> [[A]] to <4 x i16>
+; CHECK-NEXT:    [[TMP6:%.*]] = bitcast <4 x i16> [[TMP19]] to i64
+; CHECK-NEXT:    [[MMX_VAR_I:%.*]] = bitcast <4 x i16> [[TMP1]] to x86_mmx
+; CHECK-NEXT:    [[TMP7:%.*]] = bitcast <4 x i16> [[TMP16]] to i64
+; CHECK-NEXT:    [[MMX_VAR1_I:%.*]] = bitcast <4 x i16> [[TMP0]] to x86_mmx
+; CHECK-NEXT:    [[TMP8:%.*]] = or i64 [[TMP6]], [[TMP7]]
+; CHECK-NEXT:    [[TMP9:%.*]] = bitcast i64 [[TMP8]] to <2 x i32>
+; CHECK-NEXT:    [[TMP10:%.*]] = icmp ne <2 x i32> [[TMP9]], zeroinitializer
+; CHECK-NEXT:    [[TMP11:%.*]] = sext <2 x i1> [[TMP10]] to <2 x i32>
+; CHECK-NEXT:    [[TMP12:%.*]] = bitcast <2 x i32> [[TMP11]] to i64
+; CHECK-NEXT:    [[TMP2:%.*]] = tail call x86_mmx @llvm.x86.mmx.pmadd.wd(x86_mmx [[MMX_VAR_I]], x86_mmx [[MMX_VAR1_I]]) #[[ATTR2]]
+; CHECK-NEXT:    [[TMP14:%.*]] = bitcast i64 [[TMP12]] to <2 x i32>
+; CHECK-NEXT:    [[TMP3:%.*]] = bitcast x86_mmx [[TMP2]] to <2 x i32>
+; CHECK-NEXT:    [[TMP4:%.*]] = bitcast <2 x i32> [[TMP14]] to <1 x i64>
+; CHECK-NEXT:    [[TMP17:%.*]] = bitcast <2 x i32> [[TMP3]] to <1 x i64>
+; CHECK-NEXT:    [[TMP5:%.*]] = extractelement <1 x i64> [[TMP4]], i32 0
+; CHECK-NEXT:    [[TMP18:%.*]] = extractelement <1 x i64> [[TMP17]], i32 0
+; CHECK-NEXT:    store i64 [[TMP5]], ptr @__msan_retval_tls, align 8
+; CHECK-NEXT:    ret i64 [[TMP18]]
+;
+entry:
+  %0 = bitcast <1 x i64> %b to <4 x i16>
+  %1 = bitcast <1 x i64> %a to <4 x i16>
+  %mmx_var.i = bitcast <4 x i16> %1 to x86_mmx
+  %mmx_var1.i = bitcast <4 x i16> %0 to x86_mmx
+  %2 = tail call x86_mmx @llvm.x86.mmx.pmadd.wd(x86_mmx %mmx_var.i, x86_mmx %mmx_var1.i) nounwind
+  %3 = bitcast x86_mmx %2 to <2 x i32>
+  %4 = bitcast <2 x i32> %3 to <1 x i64>
+  %5 = extractelement <1 x i64> %4, i32 0
+  ret i64 %5
+}
+
+declare x86_mmx @llvm.x86.mmx.psubus.w(x86_mmx, x86_mmx) nounwind readnone
+
+define i64 @test48(<1 x i64> %a, <1 x i64> %b) #0 {
+; CHECK-LABEL: define i64 @test48(
+; CHECK-SAME: <1 x i64> [[A:%.*]], <1 x i64> [[B:%.*]]) #[[ATTR1]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP8:%.*]] = load <1 x i64>, ptr inttoptr (i64 add (i64 ptrtoint (ptr @__msan_param_tls to i64), i64 8) to ptr), align 8
+; CHECK-NEXT:    [[TMP10:%.*]] = load <1 x i64>, ptr @__msan_param_tls, align 8
+; CHECK-NEXT:    call void @llvm.donothing()
+; CHECK-NEXT:    [[TMP11:%.*]] = bitcast <1 x i64> [[TMP8]] to <4 x i16>
+; CHECK-NEXT:    [[TMP0:%.*]] = bitcast <1 x i64> [[B]] to <4 x i16>
+; CHECK-NEXT:    [[TMP14:%.*]] = bitcast <1 x i64> [[TMP10]] to <4 x i16>
+; CHECK-NEXT:    [[TMP1:%.*]] = bitcast <1 x i64> [[A]] to <4 x i16>
+; CHECK-NEXT:    [[TMP6:%.*]] = bitcast <4 x i16> [[TMP14]] to i64
+; CHECK-NEXT:    [[MMX_VAR_I:%.*]] = bitcast <4 x i16> [[TMP1]] to x86_mmx
+; CHECK-NEXT:    [[TMP7:%.*]] = bitcast <4 x i16> [[TMP11]] to i64
+; CHECK-NEXT:    [[MMX_VAR1_I:%.*]] = bitcast <4 x i16> [[TMP0]] to x86_mmx
+; CHECK-NEXT:    [[_MSPROP:%.*]] = or i64 [[TMP6]], [[TMP7]]
+; CHECK-NEXT:    [[TMP2:%.*]] = tail call x86_mmx @llvm.x86.mmx.psubus.w(x86_mmx [[MMX_VAR_I]], x86_mmx [[MMX_VAR1_I]]) #[[ATTR2]]
+; CHECK-NEXT:    [[TMP9:%.*]] = bitcast i64 [[_MSPROP]] to <4 x i16>
+; CHECK-NEXT:    [[TMP3:%.*]] = bitcast x86_mmx [[TMP2]] to <4 x i16>
+; CHECK-NEXT:    [[TMP4:%.*]] = bitcast <4 x i16> [[TMP9]] to <1 x i64>
+; CHECK-NEXT:    [[TMP12:%.*]] = bitcast <4 x i16> [[TMP3]] to <1 x i64>
+; CHECK-NEXT:    [[TMP5:%.*]] = extractelement <1 x i64> [[TMP4]], i32 0
+; CHECK-NEXT:    [[TMP13:%.*]] = extractelement <1 x i64> [[TMP12]], i32 0
+; CHECK-NEXT:    store i64 [[TMP5]], ptr @__msan_retval_tls, align 8
+; CHECK-NEXT:    ret i64 [[TMP13]]
+;
+entry:
+  %0 = bitcast <1 x i64> %b to <4 x i16>
+  %1 = bitcast <1 x i64> %a to <4 x i16>
+  %mmx_var.i = bitcast <4 x i16> %1 to x86_mmx
+  %mmx_var1.i = bitcast <4 x i16> %0 to x86_mmx
+  %2 = tail call x86_mmx @llvm.x86.mmx.psubus.w(x86_mmx %mmx_var.i, x86_mmx %mmx_var1.i) nounwind
+  %3 = bitcast x86_mmx %2 to <4 x i16>
+  %4 = bitcast <4 x i16> %3 to <1 x i64>
+  %5 = extractelement <1 x i64> %4, i32 0
+  ret i64 %5
+}
+
+declare x86_mmx @llvm.x86.mmx.psubus.b(x86_mmx, x86_mmx) nounwind readnone
+
+define i64 @test47(<1 x i64> %a, <1 x i64> %b) #0 {
+; CHECK-LABEL: define i64 @test47(
+; CHECK-SAME: <1 x i64> [[A:%.*]], <1 x i64> [[B:%.*]]) #[[ATTR1]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP8:%.*]] = load <1 x i64>, ptr inttoptr (i64 add (i64 ptrtoint (ptr @__msan_param_tls to i64), i64 8) to ptr), align 8
+; CHECK-NEXT:    [[TMP10:%.*]] = load <1 x i64>, ptr @__msan_param_tls, align 8
+; CHECK-NEXT:    call void @llvm.donothing()
+; CHECK-NEXT:    [[TMP11:%.*]] = bitcast <1 x i64> [[TMP8]] to <8 x i8>
+; CHECK-NEXT:    [[TMP0:%.*]] = bitcast <1 x i64> [[B]] to <8 x i8>
+; CHECK-NEXT:    [[TMP14:%.*]] = bitcast <1 x i64> [[TMP10]] to <8 x i8>
+; CHECK-NEXT:    [[TMP1:%.*]] = bitcast <1 x i64> [[A]] to <8 x i8>
+; CHECK-NEXT:    [[TMP6:%.*]] = bitcast <8 x i8> [[TMP14]] to i64
+; CHECK-NEXT:    [[MMX_VAR_I:%.*]] = bitcast <8 x i8> [[TMP1]] to x86_mmx
+; CHECK-NEXT:    [[TMP7:%.*]] = bitcast <8 x i8> [[TMP11]] to i64
+; CHECK-NEXT:    [[MMX_VAR1_I:%.*]] = bitcast <8 x i8> [[TMP0]] to x86_mmx
+; CHECK-NEXT:    [[_MSPROP:%.*]] = or i64 [[TMP6]], [[TMP7]]
+; CHECK-NEXT:    [[TMP2:%.*]] = tail call x86_mmx @llvm.x86.mmx.psubus.b(x86_mmx [[MMX_VAR_I]], x86_mmx [[MMX_VAR1_I]]) #[[ATTR2]]
+; CHECK-NEXT:    [[TMP9:%.*]] = bitcast i64 [[_MSPROP]] to <8 x i8>
+; CHECK-NEXT:    [[TMP3:%.*]] = bitcast x86_mmx [[TMP2]] to <8 x i8>
+; CHECK-NEXT:    [[TMP4:%.*]] = bitcast <8 x i8> [[TMP9]] to <1 x i64>
+; CHECK-NEXT:    [[TMP12:%.*]] = bitcast <8 x i8> [[TMP3]] to <1 x i64>
+; CHECK-NEXT:    [[TMP5:%.*]] = extractelement <1 x i64> [[TMP4]], i32 0
+; CHECK-NEXT:    [[TMP13:%.*]] = extractelement <1 x i64> [[TMP12]], i32 0
+; CHECK-NEXT:    store i64 [[TMP5]], ptr @__msan_retval_tls, align 8
+; CHECK-NEXT:    ret i64 [[TMP13]]
+;
+entry:
+  %0 = bitcast <1 x i64> %b to <8 x i8>
+  %1 = bitcast <1 x i64> %a to <8 x i8>
+  %mmx_var.i = bitcast <8 x i8> %1 to x86_mmx
+  %mmx_var1.i = bitcast <8 x i8> %0 to x86_mmx
+  %2 = tail call x86_mmx @llvm.x86.mmx.psubus.b(x86_mmx %mmx_var.i, x86_mmx %mmx_var1.i) nounwind
+  %3 = bitcast x86_mmx %2 to <8 x i8>
+  %4 = bitcast <8 x i8> %3 to <1 x i64>
+  %5 = extractelement <1 x i64> %4, i32 0
+  ret i64 %5
+}
+
+declare x86_mmx @llvm.x86.mmx.psubs.w(x86_mmx, x86_mmx) nounwind readnone
+
+define i64 @test46(<1 x i64> %a, <1 x i64> %b) #0 {
+; CHECK-LABEL: define i64 @test46(
+; CHECK-SAME: <1 x i64> [[A:%.*]], <1 x i64> [[B:%.*]]) #[[ATTR1]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP8:%.*]] = load <1 x i64>, ptr inttoptr (i64 add (i64 ptrtoint (ptr @__msan_param_tls to i64), i64 8) to ptr), align 8
+; CHECK-NEXT:    [[TMP10:%.*]] = load <1 x i64>, ptr @__msan_param_tls, align 8
+; CHECK-NEXT:    call void @llvm.donothing()
+; CHECK-NEXT:    [[TMP11:%.*]] = bitcast <1 x i64> [[TMP8]] to <4 x i16>
+; CHECK-NEXT:    [[TMP0:%.*]] = bitcast <1 x i64> [[B]] to <4 x i16>
+; CHECK-NEXT:    [[TMP14:%.*]] = bitcast <1 x i64> [[TMP10]] to <4 x i16>
+; CHECK-NEXT:    [[TMP1:%.*]] = bitcast <1 x i64> [[A]] to <4 x i16>
+; CHECK-NEXT:    [[TMP6:%.*]] = bitcast <4 x i16> [[TMP14]] to i64
+; CHECK-NEXT:    [[MMX_VAR_I:%.*]] = bitcast <4 x i16> [[TMP1]] to x86_mmx
+; CHECK-NEXT:    [[TMP7:%.*]] = bitcast <4 x i16> [[TMP11]] to i64
+; CHECK-NEXT:    [[MMX_VAR1_I:%.*]] = bitcast <4 x i16> [[TMP0]] to x86_mmx
+; CHECK-NEXT:    [[_MSPROP:%.*]] = or i64 [[TMP6]], [[TMP7]]
+; CHECK-NEXT:    [[TMP2:%.*]] = tail call x86_mmx @llvm.x86.mmx.psubs.w(x86_mmx [[MMX_VAR_I]], x86_mmx [[MMX_VAR1_I]]) #[[ATTR2]]
+; CHECK-NEXT:    [[TMP9:%.*]] = bitcast i64 [[_MSPROP]] to <4 x i16>
+; CHECK-NEXT:    [[TMP3:%.*]] = bitcast x86_mmx [[TMP2]] to <4 x i16>
+; CHECK-NEXT:    [[TMP4:%.*]] = bitcast <4 x i16> [[TMP9]] to <1 x i64>
+; CHECK-NEXT:    [[TMP12:%.*]] = bitcast <4 x i16> [[TMP3]] to <1 x i64>
+; CHECK-NEXT:    [[TMP5:%.*]] = extractelement <1 x i64> [[TMP4]], i32 0
+; CHECK-NEXT:    [[TMP13:%.*]] = extractelement <1 x i64> [[TMP12]], i32 0
+; CHECK-NEXT:    store i64 [[TMP5]], ptr @__msan_retval_tls, align 8
+; CHECK-NEXT:    ret i64 [[TMP13]]
+;
+entry:
+  %0 = bitcast <1 x i64> %b to <4 x i16>
+  %1 = bitcast <1 x i64> %a to <4 x i16>
+  %mmx_var.i = bitcast <4 x i16> %1 to x86_mmx
+  %mmx_var1.i = bitcast <4 x i16> %0 to x86_mmx
+  %2 = tail call x86_mmx @llvm.x86.mmx.psubs.w(x86_mmx %mmx_var.i, x86_mmx %mmx_var1.i) nounwind
+  %3 = bitcast x86_mmx %2 to <4 x i16>
+  %4 = bitcast <4 x i16> %3 to <1 x i64>
+  %5 = extractelement <1 x i64> %4, i32 0
+  ret i64 %5
+}
+
+declare x86_mmx @llvm.x86.mmx.psubs.b(x86_mmx, x86_mmx) nounwind readnone
+
+define i64 @test45(<1 x i64> %a, <1 x i64> %b) #0 {
+; CHECK-LABEL: define i64 @test45(
+; CHECK-SAME: <1 x i64> [[A:%.*]], <1 x i64> [[B:%.*]]) #[[ATTR1]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP8:%.*]] = load <1 x i64>, ptr inttoptr (i64 add (i64 ptrtoint (ptr @__msan_param_tls to i64), i64 8) to ptr), align 8
+; CHECK-NEXT:    [[TMP10:%.*]] = load <1 x i64>, ptr @__msan_param_tls, align 8
+; CHECK-NEXT:    call void @llvm.donothing()
+; CHECK-NEXT:    [[TMP11:%.*]] = bitcast <1 x i64> [[TMP8]] to <8 x i8>
+; CHECK-NEXT:    [[TMP0:%.*]] = bitcast <1 x i64> [[B]] to <8 x i8>
+; CHECK-NEXT:    [[TMP14:%.*]] = bitcast <1 x i64> [[TMP10]] to <8 x i8>
+; CHECK-NEXT:    [[TMP1:%.*]] = bitcast <1 x i64> [[A]] to <8 x i8>
+; CHECK-NEXT:    [[TMP6:%.*]] = bitcast <8 x i8> [[TMP14]] to i64
+; CHECK-NEXT:    [[MMX_VAR_I:%.*]] = bitcast <8 x i8> [[TMP1]] to x86_mmx
+; CHECK-NEXT:    [[TMP7:%.*]] = bitcast <8 x i8> [[TMP11]] to i64
+; CHECK-NEXT:    [[MMX_VAR1_I:%.*]] = bitcast <8 x i8> [[TMP0]] to x86_mmx
+; CHECK-NEXT:    [[_MSPROP:%.*]] = or i64 [[TMP6]], [[TMP7]]
+; CHECK-NEXT:    [[TMP2:%.*]] = tail call x86_mmx @llvm.x86.mmx.psubs.b(x86_mmx [[MMX_VAR_I]], x86_mmx [[MMX_VAR1_I]]) #[[ATTR2]]
+; CHECK-NEXT:    [[TMP9:%.*]] = bitcast i64 [[_MSPROP]] to <8 x i8>
+; CHECK-NEXT:    [[TMP3:%.*]] = bitcast x86_mmx [[TMP2]] to <8 x i8>
+; CHECK-NEXT:    [[TMP4:%.*]] = bitcast <8 x i8> [[TMP9]] to <1 x i64>
+; CHECK-NEXT:    [[TMP12:%.*]] = bitcast <8 x i8> [[TMP3]] to <1 x i64>
+; CHECK-NEXT:    [[TMP5:%.*]] = extractelement <1 x i64> [[TMP4]], i32 0
+; CHECK-NEXT:    [[TMP13:%.*]] = extractelement <1 x i64> [[TMP12]], i32 0
+; CHECK-NEXT:    store i64 [[TMP5]], ptr @__msan_retval_tls, align 8
+; CHECK-NEXT:    ret i64 [[TMP13]]
+;
+entry:
+  %0 = bitcast <1 x i64> %b to <8 x i8>
+  %1 = bitcast <1 x i64> %a to <8 x i8>
+  %mmx_var.i = bitcast <8 x i8> %1 to x86_mmx
+  %mmx_var1.i = bitcast <8 x i8> %0 to x86_mmx
+  %2 = tail call x86_mmx @llvm.x86.mmx.psubs.b(x86_mmx %mmx_var.i, x86_mmx %mmx_var1.i) nounwind
+  %3 = bitcast x86_mmx %2 to <8 x i8>
+  %4 = bitcast <8 x i8> %3 to <1 x i64>
+  %5 = extractelement <1 x i64> %4, i32 0
+  ret i64 %5
+}
+
+define i64 @test44(<1 x i64> %a, <1 x i64> %b) #0 {
+; CHECK-LABEL: define i64 @test44(
+; CHECK-SAME: <1 x i64> [[A:%.*]], <1 x i64> [[B:%.*]]) #[[ATTR1]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP4:%.*]] = load <1 x i64>, ptr @__msan_param_tls, align 8
+; CHECK-NEXT:    [[TMP5:%.*]] = load <1 x i64>, ptr inttoptr (i64 add (i64 ptrtoint (ptr @__msan_param_tls to i64), i64 8) to ptr), align 8
+; CHECK-NEXT:    call void @llvm.donothing()
+; CHECK-NEXT:    [[_MSPROP:%.*]] = extractelement <1 x i64> [[TMP4]], i32 0
+; CHECK-NEXT:    [[TMP0:%.*]] = extractelement <1 x i64> [[A]], i32 0
+; CHECK-NEXT:    [[MMX_VAR:%.*]] = bitcast i64 [[TMP0]] to x86_mmx
+; CHECK-NEXT:    [[_MSPROP1:%.*]] = extractelement <1 x i64> [[TMP5]], i32 0
+; CHECK-NEXT:    [[TMP1:%.*]] = extractelement <1 x i64> [[B]], i32 0
+; CHECK-NEXT:    [[MMX_VAR1:%.*]] = bitcast i64 [[TMP1]] to x86_mmx
+; CHECK-NEXT:    [[_MSPROP2:%.*]] = or i64 [[_MSPROP]], [[_MSPROP1]]
+; CHECK-NEXT:    [[TMP2:%.*]] = tail call x86_mmx @llvm.x86.mmx.psub.q(x86_mmx [[MMX_VAR]], x86_mmx [[MMX_VAR1]])
+; CHECK-NEXT:    [[TMP3:%.*]] = bitcast x86_mmx [[TMP2]] to i64
+; CHECK-NEXT:    store i64 [[_MSPROP2]], ptr @__msan_retval_tls, align 8
+; CHECK-NEXT:    ret i64 [[TMP3]]
+;
+entry:
+  %0 = extractelement <1 x i64> %a, i32 0
+  %mmx_var = bitcast i64 %0 to x86_mmx
+  %1 = extractelement <1 x i64> %b, i32 0
+  %mmx_var1 = bitcast i64 %1 to x86_mmx
+  %2 = tail call x86_mmx @llvm.x86.mmx.psub.q(x86_mmx %mmx_var, x86_mmx %mmx_var1)
+  %3 = bitcast x86_mmx %2 to i64
+  ret i64 %3
+}
+
+declare x86_mmx @llvm.x86.mmx.psub.q(x86_mmx, x86_mmx) nounwind readnone
+
+declare x86_mmx @llvm.x86.mmx.psub.d(x86_mmx, x86_mmx) nounwind readnone
+
+define i64 @test43(<1 x i64> %a, <1 x i64> %b) #0 {
+; CHECK-LABEL: define i64 @test43(
+; CHECK-SAME: <1 x i64> [[A:%.*]], <1 x i64> [[B:%.*]]) #[[ATTR1]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP8:%.*]] = load <1 x i64>, ptr inttoptr (i64 add (i64 ptrtoint (ptr @__msan_param_tls to i64), i64 8) to ptr), align 8
+; CHECK-NEXT:    [[TMP10:%.*]] = load <1 x i64>, ptr @__msan_param_tls, align 8
+; CHECK-NEXT:    call void @llvm.donothing()
+; CHECK-NEXT:    [[TMP11:%.*]] = bitcast <1 x i64> [[TMP8]] to <2 x i32>
+; CHECK-NEXT:    [[TMP0:%.*]] = bitcast <1 x i64> [[B]] to <2 x i32>
+; CHECK-NEXT:    [[TMP14:%.*]] = bitcast <1 x i64> [[TMP10]] to <2 x i32>
+; CHECK-NEXT:    [[TMP1:%.*]] = bitcast <1 x i64> [[A]] to <2 x i32>
+; CHECK-NEXT:    [[TMP6:%.*]] = bitcast <2 x i32> [[TMP14]] to i64
+; CHECK-NEXT:    [[MMX_VAR_I:%.*]] = bitcast <2 x i32> [[TMP1]] to x86_mmx
+; CHECK-NEXT:    [[TMP7:%.*]] = bitcast <2 x i32> [[TMP11]] to i64
+; CHECK-NEXT:    [[MMX_VAR1_I:%.*]] = bitcast <2 x i32> [[TMP0]] to x86_mmx
+; CHECK-NEXT:    [[_MSPROP:%.*]] = or i64 [[TMP6]], [[TMP7]]
+; CHECK-NEXT:    [[TMP2:%.*]] = tail call x86_mmx @llvm.x86.mmx.psub.d(x86_mmx [[MMX_VAR_I]], x86_mmx [[MMX_VAR1_I]]) #[[ATTR2]]
+; CHECK-NEXT:    [[TMP9:%.*]] = bitcast i64 [[_MSPROP]] to <2 x i32>
+; CHECK-NEXT:    [[TMP3:%.*]] = bitcast x86_mmx [[TMP2]] to <2 x i32>
+; CHECK-NEXT:    [[TMP4:%.*]] = bitcast <2 x i32> [[TMP9]] to <1 x i64>
+; CHECK-NEXT:    [[TMP12:%.*]] = bitcast <2 x i32> [[TMP3]] to <1 x i64>
+; CHECK-NEXT:    [[TMP5:%.*]] = extractelement <1 x i64> [[TMP4]], i32 0
+; CHECK-NEXT:    [[TMP13:%.*]] = extractelement <1 x i64> [[TMP12]], i32 0
+; CHECK-NEXT:    store i64 [[TMP5]], ptr @__msan_retval_tls, align 8
+; CHECK-NEXT:    ret i64 [[TMP13]]
+;
+entry:
+  %0 = bitcast <1 x i64> %b to <2 x i32>
+  %1 = bitcast <1 x i64> %a to <2 x i32>
+  %mmx_var.i = bitcast <2 x i32> %1 to x86_mmx
+  %mmx_var1.i = bitcast <2 x i32> %0 to x86_mmx
+  %2 = tail call x86_mmx @llvm.x86.mmx.psub.d(x86_mmx %mmx_var.i, x86_mmx %mmx_var1.i) nounwind
+  %3 = bitcast x86_mmx %2 to <2 x i32>
+  %4 = bitcast <2 x i32> %3 to <1 x i64>
+  %5 = extractelement <1 x i64> %4, i32 0
+  ret i64 %5
+}
+
+declare x86_mmx @llvm.x86.mmx.psub.w(x86_mmx, x86_mmx) nounwind readnone
+
+define i64 @test42(<1 x i64> %a, <1 x i64> %b) #0 {
+; CHECK-LABEL: define i64 @test42(
+; CHECK-SAME: <1 x i64> [[A:%.*]], <1 x i64> [[B:%.*]]) #[[ATTR1]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP8:%.*]] = load <1 x i64>, ptr inttoptr (i64 add (i64 ptrtoint (ptr @__msan_param_tls to i64), i64 8) to ptr), align 8
+; CHECK-NEXT:    [[TMP10:%.*]] = load <1 x i64>, ptr @__msan_param_tls, align 8
+; CHECK-NEXT:    call void @llvm.donothing()
+; CHECK-NEXT:    [[TMP11:%.*]] = bitcast <1 x i64> [[TMP8]] to <4 x i16>
+; CHECK-NEXT:    [[TMP0:%.*]] = bitcast <1 x i64> [[B]] to <4 x i16>
+; CHECK-NEXT:    [[TMP14:%.*]] = bitcast <1 x i64> [[TMP10]] to <4 x i16>
+; CHECK-NEXT:    [[TMP1:%.*]] = bitcast <1 x i64> [[A]] to <4 x i16>
+; CHECK-NEXT:    [[TMP6:%.*]] = bitcast <4 x i16> [[TMP14]] to i64
+; CHECK-NEXT:    [[MMX_VAR_I:%.*]] = bitcast <4 x i16> [[TMP1]] to x86_mmx
+; CHECK-NEXT:    [[TMP7:%.*]] = bitcast <4 x i16> [[TMP11]] to i64
+; CHECK-NEXT:    [[MMX_VAR1_I:%.*]] = bitcast <4 x i16> [[TMP0]] to x86_mmx
+; CHECK-NEXT:    [[_MSPROP:%.*]] = or i64 [[TMP6]], [[TMP7]]
+; CHECK-NEXT:    [[TMP2:%.*]] = tail call x86_mmx @llvm.x86.mmx.psub.w(x86_mmx [[MMX_VAR_I]], x86_mmx [[MMX_VAR1_I]]) #[[ATTR2]]
+; CHECK-NEXT:    [[TMP9:%.*]] = bitcast i64 [[_MSPROP]] to <4 x i16>
+; CHECK-NEXT:    [[TMP3:%.*]] = bitcast x86_mmx [[TMP2]] to <4 x i16>
+; CHECK-NEXT:    [[TMP4:%.*]] = bitcast <4 x i16> [[TMP9]] to <1 x i64>
+; CHECK-NEXT:    [[TMP12:%.*]] = bitcast <4 x i16> [[TMP3]] to <1 x i64>
+; CHECK-NEXT:    [[TMP5:%.*]] = extractelement <1 x i64> [[TMP4]], i32 0
+; CHECK-NEXT:    [[TMP13:%.*]] = extractelement <1 x i64> [[TMP12]], i32 0
+; CHECK-NEXT:    store i64 [[TMP5]], ptr @__msan_retval_tls, align 8
+; CHECK-NEXT:    ret i64 [[TMP13]]
+;
+entry:
+  %0 = bitcast <1 x i64> %b to <4 x i16>
+  %1 = bitcast <1 x i64> %a to <4 x i16>
+  %mmx_var.i = bitcast <4 x i16> %1 to x86_mmx
+  %mmx_var1.i = bitcast <4 x i16> %0 to x86_mmx
+  %2 = tail call x86_mmx @llvm.x86.mmx.psub.w(x86_mmx %mmx_var.i, x86_mmx %mmx_var1.i) nounwind
+  %3 = bitcast x86_mmx %2 to <4 x i16>
+  %4 = bitcast <4 x i16> %3 to <1 x i64>
+  %5 = extractelement <1 x i64> %4, i32 0
+  ret i64 %5
+}
+
+declare x86_mmx @llvm.x86.mmx.psub.b(x86_mmx, x86_mmx) nounwind readnone
+
+define i64 @test41(<1 x i64> %a, <1 x i64> %b) #0 {
+; CHECK-LABEL: define i64 @test41(
+; CHECK-SAME: <1 x i64> [[A:%.*]], <1 x i64> [[B:%.*]]) #[[ATTR1]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP8:%.*]] = load <1 x i64>, ptr inttoptr (i64 add (i64 ptrtoint (ptr @__msan_param_tls to i64), i64 8) to ptr), align 8
+; CHECK-NEXT:    [[TMP10:%.*]] = load <1 x i64>, ptr @__msan_param_tls, align 8
+; CHECK-NEXT:    call void @llvm.donothing()
+; CHECK-NEXT:    [[TMP11:%.*]] = bitcast <1 x i64> [[TMP8]] to <8 x i8>
+; CHECK-NEXT:    [[TMP0:%.*]] = bitcast <1 x i64> [[B]] to <8 x i8>
+; CHECK-NEXT:    [[TMP14:%.*]] = bitcast <1 x i64> [[TMP10]] to <8 x i8>
+; CHECK-NEXT:    [[TMP1:%.*]] = bitcast <1 x i64> [[A]] to <8 x i8>
+; CHECK-NEXT:    [[TMP6:%.*]] = bitcast <8 x i8> [[TMP14]] to i64
+; CHECK-NEXT:    [[MMX_VAR_I:%.*]] = bitcast <8 x i8> [[TMP1]] to x86_mmx
+; CHECK-NEXT:    [[TMP7:%.*]] = bitcast <8 x i8> [[TMP11]] to i64
+; CHECK-NEXT:    [[MMX_VAR1_I:%.*]] = bitcast <8 x i8> [[TMP0]] to x86_mmx
+; CHECK-NEXT:    [[_MSPROP:%.*]] = or i64 [[TMP6]], [[TMP7]]
+; CHECK-NEXT:    [[TMP2:%.*]] = tail call x86_mmx @llvm.x86.mmx.psub.b(x86_mmx [[MMX_VAR_I]], x86_mmx [[MMX_VAR1_I]]) #[[ATTR2]]
+; CHECK-NEXT:    [[TMP9:%.*]] = bitcast i64 [[_MSPROP]] to <8 x i8>
+; CHECK-NEXT:    [[TMP3:%.*]] = bitcast x86_mmx [[TMP2]] to <8 x i8>
+; CHECK-NEXT:    [[TMP4:%.*]] = bitcast <8 x i8> [[TMP9]] to <1 x i64>
+; CHECK-NEXT:    [[TMP12:%.*]] = bitcast <8 x i8> [[TMP3]] to <1 x i64>
+; CHECK-NEXT:    [[TMP5:%.*]] = extractelement <1 x i64> [[TMP4]], i32 0
+; CHECK-NEXT:    [[TMP13:%.*]] = extractelement <1 x i64> [[TMP12]], i32 0
+; CHECK-NEXT:    store i64 [[TMP5]], ptr @__msan_retval_tls, align 8
+; CHECK-NEXT:    ret i64 [[TMP13]]
+;
+entry:
+  %0 = bitcast <1 x i64> %b to <8 x i8>
+  %1 = bitcast <1 x i64> %a to <8 x i8>
+  %mmx_var.i = bitcast <8 x i8> %1 to x86_mmx
+  %mmx_var1.i = bitcast <8 x i8> %0 to x86_mmx
+  %2 = tail call x86_mmx @llvm.x86.mmx.psub.b(x86_mmx %mmx_var.i, x86_mmx %mmx_var1.i) nounwind
+  %3 = bitcast x86_mmx %2 to <8 x i8>
+  %4 = bitcast <8 x i8> %3 to <1 x i64>
+  %5 = extractelement <1 x i64> %4, i32 0
+  ret i64 %5
+}
+
+declare x86_mmx @llvm.x86.mmx.paddus.w(x86_mmx, x86_mmx) nounwind readnone
+
+define i64 @test40(<1 x i64> %a, <1 x i64> %b) #0 {
+; CHECK-LABEL: define i64 @test40(
+; CHECK-SAME: <1 x i64> [[A:%.*]], <1 x i64> [[B:%.*]]) #[[ATTR1]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP8:%.*]] = load <1 x i64>, ptr inttoptr (i64 add (i64 ptrtoint (ptr @__msan_param_tls to i64), i64 8) to ptr), align 8
+; CHECK-NEXT:    [[TMP10:%.*]] = load <1 x i64>, ptr @__msan_param_tls, align 8
+; CHECK-NEXT:    call void @llvm.donothing()
+; CHECK-NEXT:    [[TMP11:%.*]] = bitcast <1 x i64> [[TMP8]] to <4 x i16>
+; CHECK-NEXT:    [[TMP0:%.*]] = bitcast <1 x i64> [[B]] to <4 x i16>
+; CHECK-NEXT:    [[TMP14:%.*]] = bitcast <1 x i64> [[TMP10]] to <4 x i16>
+; CHECK-NEXT:    [[TMP1:%.*]] = bitcast <1 x i64> [[A]] to <4 x i16>
+; CHECK-NEXT:    [[TMP6:%.*]] = bitcast <4 x i16> [[TMP14]] to i64
+; CHECK-NEXT:    [[MMX_VAR_I:%.*]] = bitcast <4 x i16> [[TMP1]] to x86_mmx
+; CHECK-NEXT:    [[TMP7:%.*]] = bitcast <4 x i16> [[TMP11]] to i64
+; CHECK-NEXT:    [[MMX_VAR1_I:%.*]] = bitcast <4 x i16> [[TMP0]] to x86_mmx
+; CHECK-NEXT:    [[_MSPROP:%.*]] = or i64 [[TMP6]], [[TMP7]]
+; CHECK-NEXT:    [[TMP2:%.*]] = tail call x86_mmx @llvm.x86.mmx.paddus.w(x86_mmx [[MMX_VAR_I]], x86_mmx [[MMX_VAR1_I]]) #[[ATTR2]]
+; CHECK-NEXT:    [[TMP9:%.*]] = bitcast i64 [[_MSPROP]] to <4 x i16>
+; CHECK-NEXT:    [[TMP3:%.*]] = bitcast x86_mmx [[TMP2]] to <4 x i16>
+; CHECK-NEXT:    [[TMP4:%.*]] = bitcast <4 x i16> [[TMP9]] to <1 x i64>
+; CHECK-NEXT:    [[TMP12:%.*]] = bitcast <4 x i16> [[TMP3]] to <1 x i64>
+; CHECK-NEXT:    [[TMP5:%.*]] = extractelement <1 x i64> [[TMP4]], i32 0
+; CHECK-NEXT:    [[TMP13:%.*]] = extractelement <1 x i64> [[TMP12]], i32 0
+; CHECK-NEXT:    store i64 [[TMP5]], ptr @__msan_retval_tls, align 8
+; CHECK-NEXT:    ret i64 [[TMP13]]
+;
+entry:
+  %0 = bitcast <1 x i64> %b to <4 x i16>
+  %1 = bitcast <1 x i64> %a to <4 x i16>
+  %mmx_var.i = bitcast <4 x i16> %1 to x86_mmx
+  %mmx_var1.i = bitcast <4 x i16> %0 to x86_mmx
+  %2 = tail call x86_mmx @llvm.x86.mmx.paddus.w(x86_mmx %mmx_var.i, x86_mmx %mmx_var1.i) nounwind
+  %3 = bitcast x86_mmx %2 to <4 x i16>
+  %4 = bitcast <4 x i16> %3 to <1 x i64>
+  %5 = extractelement <1 x i64> %4, i32 0
+  ret i64 %5
+}
+
+declare x86_mmx @llvm.x86.mmx.paddus.b(x86_mmx, x86_mmx) nounwind readnone
+
+define i64 @test39(<1 x i64> %a, <1 x i64> %b) #0 {
+; CHECK-LABEL: define i64 @test39(
+; CHECK-SAME: <1 x i64> [[A:%.*]], <1 x i64> [[B:%.*]]) #[[ATTR1]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP8:%.*]] = load <1 x i64>, ptr inttoptr (i64 add (i64 ptrtoint (ptr @__msan_param_tls to i64), i64 8) to ptr), align 8
+; CHECK-NEXT:    [[TMP10:%.*]] = load <1 x i64>, ptr @__msan_param_tls, align 8
+; CHECK-NEXT:    call void @llvm.donothing()
+; CHECK-NEXT:    [[TMP11:%.*]] = bitcast <1 x i64> [[TMP8]] to <8 x i8>
+; CHECK-NEXT:    [[TMP0:%.*]] = bitcast <1 x i64> [[B]] to <8 x i8>
+; CHECK-NEXT:    [[TMP14:%.*]] = bitcast <1 x i64> [[TMP10]] to <8 x i8>
+; CHECK-NEXT:    [[TMP1:%.*]] = bitcast <1 x i64> [[A]] to <8 x i8>
+; CHECK-NEXT:    [[TMP6:%.*]] = bitcast <8 x i8> [[TMP14]] to i64
+; CHECK-NEXT:    [[MMX_VAR_I:%.*]] = bitcast <8 x i8> [[TMP1]] to x86_mmx
+; CHECK-NEXT:    [[TMP7:%.*]] = bitcast <8 x i8> [[TMP11]] to i64
+; CHECK-NEXT:    [[MMX_VAR1_I:%.*]] = bitcast <8 x i8> [[TMP0]] to x86_mmx
+; CHECK-NEXT:    [[_MSPROP:%.*]] = or i64 [[TMP6]], [[TMP7]]
+; CHECK-NEXT:    [[TMP2:%.*]] = tail call x86_mmx @llvm.x86.mmx.paddus.b(x86_mmx [[MMX_VAR_I]], x86_mmx [[MMX_VAR1_I]]) #[[ATTR2]]
+; CHECK-NEXT:    [[TMP9:%.*]] = bitcast i64 [[_MSPROP]] to <8 x i8>
+; CHECK-NEXT:    [[TMP3:%.*]] = bitcast x86_mmx [[TMP2]] to <8 x i8>
+; CHECK-NEXT:    [[TMP4:%.*]] = bitcast <8 x i8> [[TMP9]] to <1 x i64>
+; CHECK-NEXT:    [[TMP12:%.*]] = bitcast <8 x i8> [[TMP3]] to <1 x i64>
+; CHECK-NEXT:    [[TMP5:%.*]] = extractelement <1 x i64> [[TMP4]], i32 0
+; CHECK-NEXT:    [[TMP13:%.*]] = extractelement <1 x i64> [[TMP12]], i32 0
+; CHECK-NEXT:    store i64 [[TMP5]], ptr @__msan_retval_tls, align 8
+; CHECK-NEXT:    ret i64 [[TMP13]]
+;
+entry:
+  %0 = bitcast <1 x i64> %b to <8 x i8>
+  %1 = bitcast <1 x i64> %a to <8 x i8>
+  %mmx_var.i = bitcast <8 x i8> %1 to x86_mmx
+  %mmx_var1.i = bitcast <8 x i8> %0 to x86_mmx
+  %2 = tail call x86_mmx @llvm.x86.mmx.paddus.b(x86_mmx %mmx_var.i, x86_mmx %mmx_var1.i) nounwind
+  %3 = bitcast x86_mmx %2 to <8 x i8>
+  %4 = bitcast <8 x i8> %3 to <1 x i64>
+  %5 = extractelement <1 x i64> %4, i32 0
+  ret i64 %5
+}
+
+declare x86_mmx @llvm.x86.mmx.padds.w(x86_mmx, x86_mmx) nounwind readnone
+
+define i64 @test38(<1 x i64> %a, <1 x i64> %b) #0 {
+; CHECK-LABEL: define i64 @test38(
+; CHECK-SAME: <1 x i64> [[A:%.*]], <1 x i64> [[B:%.*]]) #[[ATTR1]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP8:%.*]] = load <1 x i64>, ptr inttoptr (i64 add (i64 ptrtoint (ptr @__msan_param_tls to i64), i64 8) to ptr), align 8
+; CHECK-NEXT:    [[TMP10:%.*]] = load <1 x i64>, ptr @__msan_param_tls, align 8
+; CHECK-NEXT:    call void @llvm.donothing()
+; CHECK-NEXT:    [[TMP11:%.*]] = bitcast <1 x i64> [[TMP8]] to <4 x i16>
+; CHECK-NEXT:    [[TMP0:%.*]] = bitcast <1 x i64> [[B]] to <4 x i16>
+; CHECK-NEXT:    [[TMP14:%.*]] = bitcast <1 x i64> [[TMP10]] to <4 x i16>
+; CHECK-NEXT:    [[TMP1:%.*]] = bitcast <1 x i64> [[A]] to <4 x i16>
+; CHECK-NEXT:    [[TMP6:%.*]] = bitcast <4 x i16> [[TMP14]] to i64
+; CHECK-NEXT:    [[MMX_VAR_I:%.*]] = bitcast <4 x i16> [[TMP1]] to x86_mmx
+; CHECK-NEXT:    [[TMP7:%.*]] = bitcast <4 x i16> [[TMP11]] to i64
+; CHECK-NEXT:    [[MMX_VAR1_I:%.*]] = bitcast <4 x i16> [[TMP0]] to x86_mmx
+; CHECK-NEXT:    [[_MSPROP:%.*]] = or i64 [[TMP6]], [[TMP7]]
+; CHECK-NEXT:    [[TMP2:%.*]] = tail call x86_mmx @llvm.x86.mmx.padds.w(x86_mmx [[MMX_VAR_I]], x86_mmx [[MMX_VAR1_I]]) #[[ATTR2]]
+; CHECK-NEXT:    [[TMP9:%.*]] = bitcast i64 [[_MSPROP]] to <4 x i16>
+; CHECK-NEXT:    [[TMP3:%.*]] = bitcast x86_mmx [[TMP2]] to <4 x i16>
+; CHECK-NEXT:    [[TMP4:%.*]] = bitcast <4 x i16> [[TMP9]] to <1 x i64>
+; CHECK-NEXT:    [[TMP12:%.*]] = bitcast <4 x i16> [[TMP3]] to <1 x i64>
+; CHECK-NEXT:    [[TMP5:%.*]] = extractelement <1 x i64> [[TMP4]], i32 0
+; CHECK-NEXT:    [[TMP13:%.*]] = extractelement <1 x i64> [[TMP12]], i32 0
+; CHECK-NEXT:    store i64 [[TMP5]], ptr @__msan_retval_tls, align 8
+; CHECK-NEXT:    ret i64 [[TMP13]]
+;
+entry:
+  %0 = bitcast <1 x i64> %b to <4 x i16>
+  %1 = bitcast <1 x i64> %a to <4 x i16>
+  %mmx_var.i = bitcast <4 x i16> %1 to x86_mmx
+  %mmx_var1.i = bitcast <4 x i16> %0 to x86_mmx
+  %2 = tail call x86_mmx @llvm.x86.mmx.padds.w(x86_mmx %mmx_var.i, x86_mmx %mmx_var1.i) nounwind
+  %3 = bitcast x86_mmx %2 to <4 x i16>
+  %4 = bitcast <4 x i16> %3 to <1 x i64>
+  %5 = extractelement <1 x i64> %4, i32 0
+  ret i64 %5
+}
+
+declare x86_mmx @llvm.x86.mmx.padds.b(x86_mmx, x86_mmx) nounwind readnone
+
+define i64 @test37(<1 x i64> %a, <1 x i64> %b) #0 {
+; CHECK-LABEL: define i64 @test37(
+; CHECK-SAME: <1 x i64> [[A:%.*]], <1 x i64> [[B:%.*]]) #[[ATTR1]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP8:%.*]] = load <1 x i64>, ptr inttoptr (i64 add (i64 ptrtoint (ptr @__msan_param_tls to i64), i64 8) to ptr), align 8
+; CHECK-NEXT:    [[TMP10:%.*]] = load <1 x i64>, ptr @__msan_param_tls, align 8
+; CHECK-NEXT:    call void @llvm.donothing()
+; CHECK-NEXT:    [[TMP11:%.*]] = bitcast <1 x i64> [[TMP8]] to <8 x i8>
+; CHECK-NEXT:    [[TMP0:%.*]] = bitcast <1 x i64> [[B]] to <8 x i8>
+; CHECK-NEXT:    [[TMP14:%.*]] = bitcast <1 x i64> [[TMP10]] to <8 x i8>
+; CHECK-NEXT:    [[TMP1:%.*]] = bitcast <1 x i64> [[A]] to <8 x i8>
+; CHECK-NEXT:    [[TMP6:%.*]] = bitcast <8 x i8> [[TMP14]] to i64
+; CHECK-NEXT:    [[MMX_VAR_I:%.*]] = bitcast <8 x i8> [[TMP1]] to x86_mmx
+; CHECK-NEXT:    [[TMP7:%.*]] = bitcast <8 x i8> [[TMP11]] to i64
+; CHECK-NEXT:    [[MMX_VAR1_I:%.*]] = bitcast <8 x i8> [[TMP0]] to x86_mmx
+; CHECK-NEXT:    [[_MSPROP:%.*]] = or i64 [[TMP6]], [[TMP7]]
+; CHECK-NEXT:    [[TMP2:%.*]] = tail call x86_mmx @llvm.x86.mmx.padds.b(x86_mmx [[MMX_VAR_I]], x86_mmx [[MMX_VAR1_I]]) #[[ATTR2]]
+; CHECK-NEXT:    [[TMP9:%.*]] = bitcast i64 [[_MSPROP]] to <8 x i8>
+; CHECK-NEXT:    [[TMP3:%.*]] = bitcast x86_mmx [[TMP2]] to <8 x i8>
+; CHECK-NEXT:    [[TMP4:%.*]] = bitcast <8 x i8> [[TMP9]] to <1 x i64>
+; CHECK-NEXT:    [[TMP12:%.*]] = bitcast <8 x i8> [[TMP3]] to <1 x i64>
+; CHECK-NEXT:    [[TMP5:%.*]] = extractelement <1 x i64> [[TMP4]], i32 0
+; CHECK-NEXT:    [[TMP13:%.*]] = extractelement <1 x i64> [[TMP12]], i32 0
+; CHECK-NEXT:    store i64 [[TMP5]], ptr @__msan_retval_tls, align 8
+; CHECK-NEXT:    ret i64 [[TMP13]]
+;
+entry:
+  %0 = bitcast <1 x i64> %b to <8 x i8>
+  %1 = bitcast <1 x i64> %a to <8 x i8>
+  %mmx_var.i = bitcast <8 x i8> %1 to x86_mmx
+  %mmx_var1.i = bitcast <8 x i8> %0 to x86_mmx
+  %2 = tail call x86_mmx @llvm.x86.mmx.padds.b(x86_mmx %mmx_var.i, x86_mmx %mmx_var1.i) nounwind
+  %3 = bitcast x86_mmx %2 to <8 x i8>
+  %4 = bitcast <8 x i8> %3 to <1 x i64>
+  %5 = extractelement <1 x i64> %4, i32 0
+  ret i64 %5
+}
+
+declare x86_mmx @llvm.x86.mmx.padd.q(x86_mmx, x86_mmx) nounwind readnone
+
+define i64 @test36(<1 x i64> %a, <1 x i64> %b) #0 {
+; CHECK-LABEL: define i64 @test36(
+; CHECK-SAME: <1 x i64> [[A:%.*]], <1 x i64> [[B:%.*]]) #[[ATTR1]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP4:%.*]] = load <1 x i64>, ptr @__msan_param_tls, align 8
+; CHECK-NEXT:    [[TMP5:%.*]] = load <1 x i64>, ptr inttoptr (i64 add (i64 ptrtoint (ptr @__msan_param_tls to i64), i64 8) to ptr), align 8
+; CHECK-NEXT:    call void @llvm.donothing()
+; CHECK-NEXT:    [[_MSPROP:%.*]] = extractelement <1 x i64> [[TMP4]], i32 0
+; CHECK-NEXT:    [[TMP0:%.*]] = extractelement <1 x i64> [[A]], i32 0
+; CHECK-NEXT:    [[MMX_VAR:%.*]] = bitcast i64 [[TMP0]] to x86_mmx
+; CHECK-NEXT:    [[_MSPROP1:%.*]] = extractelement <1 x i64> [[TMP5]], i32 0
+; CHECK-NEXT:    [[TMP1:%.*]] = extractelement <1 x i64> [[B]], i32 0
+; CHECK-NEXT:    [[MMX_VAR1:%.*]] = bitcast i64 [[TMP1]] to x86_mmx
+; CHECK-NEXT:    [[_MSPROP2:%.*]] = or i64 [[_MSPROP]], [[_MSPROP1]]
+; CHECK-NEXT:    [[TMP2:%.*]] = tail call x86_mmx @llvm.x86.mmx.padd.q(x86_mmx [[MMX_VAR]], x86_mmx [[MMX_VAR1]])
+; CHECK-NEXT:    [[TMP3:%.*]] = bitcast x86_mmx [[TMP2]] to i64
+; CHECK-NEXT:    store i64 [[_MSPROP2]], ptr @__msan_retval_tls, align 8
+; CHECK-NEXT:    ret i64 [[TMP3]]
+;
+entry:
+  %0 = extractelement <1 x i64> %a, i32 0
+  %mmx_var = bitcast i64 %0 to x86_mmx
+  %1 = extractelement <1 x i64> %b, i32 0
+  %mmx_var1 = bitcast i64 %1 to x86_mmx
+  %2 = tail call x86_mmx @llvm.x86.mmx.padd.q(x86_mmx %mmx_var, x86_mmx %mmx_var1)
+  %3 = bitcast x86_mmx %2 to i64
+  ret i64 %3
+}
+
+declare x86_mmx @llvm.x86.mmx.padd.d(x86_mmx, x86_mmx) nounwind readnone
+
+define i64 @test35(<1 x i64> %a, <1 x i64> %b) #0 {
+; CHECK-LABEL: define i64 @test35(
+; CHECK-SAME: <1 x i64> [[A:%.*]], <1 x i64> [[B:%.*]]) #[[ATTR1]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP8:%.*]] = load <1 x i64>, ptr inttoptr (i64 add (i64 ptrtoint (ptr @__msan_param_tls to i64), i64 8) to ptr), align 8
+; CHECK-NEXT:    [[TMP10:%.*]] = load <1 x i64>, ptr @__msan_param_tls, align 8
+; CHECK-NEXT:    call void @llvm.donothing()
+; CHECK-NEXT:    [[TMP11:%.*]] = bitcast <1 x i64> [[TMP8]] to <2 x i32>
+; CHECK-NEXT:    [[TMP0:%.*]] = bitcast <1 x i64> [[B]] to <2 x i32>
+; CHECK-NEXT:    [[TMP14:%.*]] = bitcast <1 x i64> [[TMP10]] to <2 x i32>
+; CHECK-NEXT:    [[TMP1:%.*]] = bitcast <1 x i64> [[A]] to <2 x i32>
+; CHECK-NEXT:    [[TMP6:%.*]] = bitcast <2 x i32> [[TMP14]] to i64
+; CHECK-NEXT:    [[MMX_VAR_I:%.*]] = bitcast <2 x i32> [[TMP1]] to x86_mmx
+; CHECK-NEXT:    [[TMP7:%.*]] = bitcast <2 x i32> [[TMP11]] to i64
+; CHECK-NEXT:    [[MMX_VAR1_I:%.*]] = bitcast <2 x i32> [[TMP0]] to x86_mmx
+; CHECK-NEXT:    [[_MSPROP:%.*]] = or i64 [[TMP6]], [[TMP7]]
+; CHECK-NEXT:    [[TMP2:%.*]] = tail call x86_mmx @llvm.x86.mmx.padd.d(x86_mmx [[MMX_VAR_I]], x86_mmx [[MMX_VAR1_I]]) #[[ATTR2]]
+; CHECK-NEXT:    [[TMP9:%.*]] = bitcast i64 [[_MSPROP]] to <2 x i32>
+; CHECK-NEXT:    [[TMP3:%.*]] = bitcast x86_mmx [[TMP2]] to <2 x i32>
+; CHECK-NEXT:    [[TMP4:%.*]] = bitcast <2 x i32> [[TMP9]] to <1 x i64>
+; CHECK-NEXT:    [[TMP12:%.*]] = bitcast <2 x i32> [[TMP3]] to <1 x i64>
+; CHECK-NEXT:    [[TMP5:%.*]] = extractelement <1 x i64> [[TMP4]], i32 0
+; CHECK-NEXT:    [[TMP13:%.*]] = extractelement <1 x i64> [[TMP12]], i32 0
+; CHECK-NEXT:    store i64 [[TMP5]], ptr @__msan_retval_tls, align 8
+; CHECK-NEXT:    ret i64 [[TMP13]]
+;
+entry:
+  %0 = bitcast <1 x i64> %b to <2 x i32>
+  %1 = bitcast <1 x i64> %a to <2 x i32>
+  %mmx_var.i = bitcast <2 x i32> %1 to x86_mmx
+  %mmx_var1.i = bitcast <2 x i32> %0 to x86_mmx
+  %2 = tail call x86_mmx @llvm.x86.mmx.padd.d(x86_mmx %mmx_var.i, x86_mmx %mmx_var1.i) nounwind
+  %3 = bitcast x86_mmx %2 to <2 x i32>
+  %4 = bitcast <2 x i32> %3 to <1 x i64>
+  %5 = extractelement <1 x i64> %4, i32 0
+  ret i64 %5
+}
+
+declare x86_mmx @llvm.x86.mmx.padd.w(x86_mmx, x86_mmx) nounwind readnone
+
+define i64 @test34(<1 x i64> %a, <1 x i64> %b) #0 {
+; CHECK-LABEL: define i64 @test34(
+; CHECK-SAME: <1 x i64> [[A:%.*]], <1 x i64> [[B:%.*]]) #[[ATTR1]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP8:%.*]] = load <1 x i64>, ptr inttoptr (i64 add (i64 ptrtoint (ptr @__msan_param_tls to i64), i64 8) to ptr), align 8
+; CHECK-NEXT:    [[TMP10:%.*]] = load <1 x i64>, ptr @__msan_param_tls, align 8
+; CHECK-NEXT:    call void @llvm.donothing()
+; CHECK-NEXT:    [[TMP11:%.*]] = bitcast <1 x i64> [[TMP8]] to <4 x i16>
+; CHECK-NEXT:    [[TMP0:%.*]] = bitcast <1 x i64> [[B]] to <4 x i16>
+; CHECK-NEXT:    [[TMP14:%.*]] = bitcast <1 x i64> [[TMP10]] to <4 x i16>
+; CHECK-NEXT:    [[TMP1:%.*]] = bitcast <1 x i64> [[A]] to <4 x i16>
+; CHECK-NEXT:    [[TMP6:%.*]] = bitcast <4 x i16> [[TMP14]] to i64
+; CHECK-NEXT:    [[MMX_VAR_I:%.*]] = bitcast <4 x i16> [[TMP1]] to x86_mmx
+; CHECK-NEXT:    [[TMP7:%.*]] = bitcast <4 x i16> [[TMP11]] to i64
+; CHECK-NEXT:    [[MMX_VAR1_I:%.*]] = bitcast <4 x i16> [[TMP0]] to x86_mmx
+; CHECK-NEXT:    [[_MSPROP:%.*]] = or i64 [[TMP6]], [[TMP7]]
+; CHECK-NEXT:    [[TMP2:%.*]] = tail call x86_mmx @llvm.x86.mmx.padd.w(x86_mmx [[MMX_VAR_I]], x86_mmx [[MMX_VAR1_I]]) #[[ATTR2]]
+; CHECK-NEXT:    [[TMP9:%.*]] = bitcast i64 [[_MSPROP]] to <4 x i16>
+; CHECK-NEXT:    [[TMP3:%.*]] = bitcast x86_mmx [[TMP2]] to <4 x i16>
+; CHECK-NEXT:    [[TMP4:%.*]] = bitcast <4 x i16> [[TMP9]] to <1 x i64>
+; CHECK-NEXT:    [[TMP12:%.*]] = bitcast <4 x i16> [[TMP3]] to <1 x i64>
+; CHECK-NEXT:    [[TMP5:%.*]] = extractelement <1 x i64> [[TMP4]], i32 0
+; CHECK-NEXT:    [[TMP13:%.*]] = extractelement <1 x i64> [[TMP12]], i32 0
+; CHECK-NEXT:    store i64 [[TMP5]], ptr @__msan_retval_tls, align 8
+; CHECK-NEXT:    ret i64 [[TMP13]]
+;
+entry:
+  %0 = bitcast <1 x i64> %b to <4 x i16>
+  %1 = bitcast <1 x i64> %a to <4 x i16>
+  %mmx_var.i = bitcast <4 x i16> %1 to x86_mmx
+  %mmx_var1.i = bitcast <4 x i16> %0 to x86_mmx
+  %2 = tail call x86_mmx @llvm.x86.mmx.padd.w(x86_mmx %mmx_var.i, x86_mmx %mmx_var1.i) nounwind
+  %3 = bitcast x86_mmx %2 to <4 x i16>
+  %4 = bitcast <4 x i16> %3 to <1 x i64>
+  %5 = extractelement <1 x i64> %4, i32 0
+  ret i64 %5
+}
+
+declare x86_mmx @llvm.x86.mmx.padd.b(x86_mmx, x86_mmx) nounwind readnone
+
+define i64 @test33(<1 x i64> %a, <1 x i64> %b) #0 {
+; CHECK-LABEL: define i64 @test33(
+; CHECK-SAME: <1 x i64> [[A:%.*]], <1 x i64> [[B:%.*]]) #[[ATTR1]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP8:%.*]] = load <1 x i64>, ptr inttoptr (i64 add (i64 ptrtoint (ptr @__msan_param_tls to i64), i64 8) to ptr), align 8
+; CHECK-NEXT:    [[TMP10:%.*]] = load <1 x i64>, ptr @__msan_param_tls, align 8
+; CHECK-NEXT:    call void @llvm.donothing()
+; CHECK-NEXT:    [[TMP11:%.*]] = bitcast <1 x i64> [[TMP8]] to <8 x i8>
+; CHECK-NEXT:    [[TMP0:%.*]] = bitcast <1 x i64> [[B]] to <8 x i8>
+; CHECK-NEXT:    [[TMP14:%.*]] = bitcast <1 x i64> [[TMP10]] to <8 x i8>
+; CHECK-NEXT:    [[TMP1:%.*]] = bitcast <1 x i64> [[A]] to <8 x i8>
+; CHECK-NEXT:    [[TMP6:%.*]] = bitcast <8 x i8> [[TMP14]] to i64
+; CHECK-NEXT:    [[MMX_VAR_I:%.*]] = bitcast <8 x i8> [[TMP1]] to x86_mmx
+; CHECK-NEXT:    [[TMP7:%.*]] = bitcast <8 x i8> [[TMP11]] to i64
+; CHECK-NEXT:    [[MMX_VAR1_I:%.*]] = bitcast <8 x i8> [[TMP0]] to x86_mmx
+; CHECK-NEXT:    [[_MSPROP:%.*]] = or i64 [[TMP6]], [[TMP7]]
+; CHECK-NEXT:    [[TMP2:%.*]] = tail call x86_mmx @llvm.x86.mmx.padd.b(x86_mmx [[MMX_VAR_I]], x86_mmx [[MMX_VAR1_I]]) #[[ATTR2]]
+; CHECK-NEXT:    [[TMP9:%.*]] = bitcast i64 [[_MSPROP]] to <8 x i8>
+; CHECK-NEXT:    [[TMP3:%.*]] = bitcast x86_mmx [[TMP2]] to <8 x i8>
+; CHECK-NEXT:    [[TMP4:%.*]] = bitcast <8 x i8> [[TMP9]] to <1 x i64>
+; CHECK-NEXT:    [[TMP12:%.*]] = bitcast <8 x i8> [[TMP3]] to <1 x i64>
+; CHECK-NEXT:    [[TMP5:%.*]] = extractelement <1 x i64> [[TMP4]], i32 0
+; CHECK-NEXT:    [[TMP13:%.*]] = extractelement <1 x i64> [[TMP12]], i32 0
+; CHECK-NEXT:    store i64 [[TMP5]], ptr @__msan_retval_tls, align 8
+; CHECK-NEXT:    ret i64 [[TMP13]]
+;
+entry:
+  %0 = bitcast <1 x i64> %b to <8 x i8>
+  %1 = bitcast <1 x i64> %a to <8 x i8>
+  %mmx_var.i = bitcast <8 x i8> %1 to x86_mmx
+  %mmx_var1.i = bitcast <8 x i8> %0 to x86_mmx
+  %2 = tail call x86_mmx @llvm.x86.mmx.padd.b(x86_mmx %mmx_var.i, x86_mmx %mmx_var1.i) nounwind
+  %3 = bitcast x86_mmx %2 to <8 x i8>
+  %4 = bitcast <8 x i8> %3 to <1 x i64>
+  %5 = extractelement <1 x i64> %4, i32 0
+  ret i64 %5
+}
+
+declare x86_mmx @llvm.x86.mmx.psad.bw(x86_mmx, x86_mmx) nounwind readnone
+
+define i64 @test32(<1 x i64> %a, <1 x i64> %b) #0 {
+; CHECK-LABEL: define i64 @test32(
+; CHECK-SAME: <1 x i64> [[A:%.*]], <1 x i64> [[B:%.*]]) #[[ATTR1]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP5:%.*]] = load <1 x i64>, ptr inttoptr (i64 add (i64 ptrtoint (ptr @__msan_param_tls to i64), i64 8) to ptr), align 8
+; CHECK-NEXT:    [[TMP12:%.*]] = load <1 x i64>, ptr @__msan_param_tls, align 8
+; CHECK-NEXT:    call void @llvm.donothing()
+; CHECK-NEXT:    [[TMP13:%.*]] = bitcast <1 x i64> [[TMP5]] to <8 x i8>
+; CHECK-NEXT:    [[TMP0:%.*]] = bitcast <1 x i64> [[B]] to <8 x i8>
+; CHECK-NEXT:    [[TMP4:%.*]] = bitcast <1 x i64> [[TMP12]] to <8 x i8>
+; CHECK-NEXT:    [[TMP1:%.*]] = bitcast <1 x i64> [[A]] to <8 x i8>
+; CHECK-NEXT:    [[TMP6:%.*]] = bitcast <8 x i8> [[TMP4]] to i64
+; CHECK-NEXT:    [[MMX_VAR_I:%.*]] = bitcast <8 x i8> [[TMP1]] to x86_mmx
+; CHECK-NEXT:    [[TMP7:%.*]] = bitcast <8 x i8> [[TMP13]] to i64
+; CHECK-NEXT:    [[MMX_VAR1_I:%.*]] = bitcast <8 x i8> [[TMP0]] to x86_mmx
+; CHECK-NEXT:    [[TMP8:%.*]] = or i64 [[TMP6]], [[TMP7]]
+; CHECK-NEXT:    [[TMP9:%.*]] = icmp ne i64 [[TMP8]], 0
+; CHECK-NEXT:    [[TMP10:%.*]] = sext i1 [[TMP9]] to i64
+; CHECK-NEXT:    [[TMP11:%.*]] = lshr i64 [[TMP10]], 48
+; CHECK-NEXT:    [[TMP2:%.*]] = tail call x86_mmx @llvm.x86.mmx.psad.bw(x86_mmx [[MMX_VAR_I]], x86_mmx [[MMX_VAR1_I]]) #[[ATTR2]]
+; CHECK-NEXT:    [[TMP3:%.*]] = bitcast x86_mmx [[TMP2]] to i64
+; CHECK-NEXT:    store i64 [[TMP11]], ptr @__msan_retval_tls, align 8
+; CHECK-NEXT:    ret i64 [[TMP3]]
+;
+entry:
+  %0 = bitcast <1 x i64> %b to <8 x i8>
+  %1 = bitcast <1 x i64> %a to <8 x i8>
+  %mmx_var.i = bitcast <8 x i8> %1 to x86_mmx
+  %mmx_var1.i = bitcast <8 x i8> %0 to x86_mmx
+  %2 = tail call x86_mmx @llvm.x86.mmx.psad.bw(x86_mmx %mmx_var.i, x86_mmx %mmx_var1.i) nounwind
+  %3 = bitcast x86_mmx %2 to i64
+  ret i64 %3
+}
+
+declare x86_mmx @llvm.x86.mmx.pmins.w(x86_mmx, x86_mmx) nounwind readnone
+
+define i64 @test31(<1 x i64> %a, <1 x i64> %b) #0 {
+; CHECK-LABEL: define i64 @test31(
+; CHECK-SAME: <1 x i64> [[A:%.*]], <1 x i64> [[B:%.*]]) #[[ATTR1]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP8:%.*]] = load <1 x i64>, ptr inttoptr (i64 add (i64 ptrtoint (ptr @__msan_param_tls to i64), i64 8) to ptr), align 8
+; CHECK-NEXT:    [[TMP10:%.*]] = load <1 x i64>, ptr @__msan_param_tls, align 8
+; CHECK-NEXT:    call void @llvm.donothing()
+; CHECK-NEXT:    [[TMP11:%.*]] = bitcast <1 x i64> [[TMP8]] to <4 x i16>
+; CHECK-NEXT:    [[TMP0:%.*]] = bitcast <1 x i64> [[B]] to <4 x i16>
+; CHECK-NEXT:    [[TMP14:%.*]] = bitcast <1 x i64> [[TMP10]] to <4 x i16>
+; CHECK-NEXT:    [[TMP1:%.*]] = bitcast <1 x i64> [[A]] to <4 x i16>
+; CHECK-NEXT:    [[TMP6:%.*]] = bitcast <4 x i16> [[TMP14]] to i64
+; CHECK-NEXT:    [[MMX_VAR_I:%.*]] = bitcast <4 x i16> [[TMP1]] to x86_mmx
+; CHECK-NEXT:    [[TMP7:%.*]] = bitcast <4 x i16> [[TMP11]] to i64
+; CHECK-NEXT:    [[MMX_VAR1_I:%.*]] = bitcast <4 x i16> [[TMP0]] to x86_mmx
+; CHECK-NEXT:    [[_MSPROP:%.*]] = or i64 [[TMP6]], [[TMP7]]
+; CHECK-NEXT:    [[TMP2:%.*]] = tail call x86_mmx @llvm.x86.mmx.pmins.w(x86_mmx [[MMX_VAR_I]], x86_mmx [[MMX_VAR1_I]]) #[[ATTR2]]
+; CHECK-NEXT:    [[TMP9:%.*]] = bitcast i64 [[_MSPROP]] to <4 x i16>
+; CHECK-NEXT:    [[TMP3:%.*]] = bitcast x86_mmx [[TMP2]] to <4 x i16>
+; CHECK-NEXT:    [[TMP4:%.*]] = bitcast <4 x i16> [[TMP9]] to <1 x i64>
+; CHECK-NEXT:    [[TMP12:%.*]] = bitcast <4 x i16> [[TMP3]] to <1 x i64>
+; CHECK-NEXT:    [[TMP5:%.*]] = extractelement <1 x i64> [[TMP4]], i32 0
+; CHECK-NEXT:    [[TMP13:%.*]] = extractelement <1 x i64> [[TMP12]], i32 0
+; CHECK-NEXT:    store i64 [[TMP5]], ptr @__msan_retval_tls, align 8
+; CHECK-NEXT:    ret i64 [[TMP13]]
+;
+entry:
+  %0 = bitcast <1 x i64> %b to <4 x i16>
+  %1 = bitcast <1 x i64> %a to <4 x i16>
+  %mmx_var.i = bitcast <4 x i16> %1 to x86_mmx
+  %mmx_var1.i = bitcast <4 x i16> %0 to x86_mmx
+  %2 = tail call x86_mmx @llvm.x86.mmx.pmins.w(x86_mmx %mmx_var.i, x86_mmx %mmx_var1.i) nounwind
+  %3 = bitcast x86_mmx %2 to <4 x i16>
+  %4 = bitcast <4 x i16> %3 to <1 x i64>
+  %5 = extractelement <1 x i64> %4, i32 0
+  ret i64 %5
+}
+
+declare x86_mmx @llvm.x86.mmx.pminu.b(x86_mmx, x86_mmx) nounwind readnone
+
+define i64 @test30(<1 x i64> %a, <1 x i64> %b) #0 {
+; CHECK-LABEL: define i64 @test30(
+; CHECK-SAME: <1 x i64> [[A:%.*]], <1 x i64> [[B:%.*]]) #[[ATTR1]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP8:%.*]] = load <1 x i64>, ptr inttoptr (i64 add (i64 ptrtoint (ptr @__msan_param_tls to i64), i64 8) to ptr), align 8
+; CHECK-NEXT:    [[TMP10:%.*]] = load <1 x i64>, ptr @__msan_param_tls, align 8
+; CHECK-NEXT:    call void @llvm.donothing()
+; CHECK-NEXT:    [[TMP11:%.*]] = bitcast <1 x i64> [[TMP8]] to <8 x i8>
+; CHECK-NEXT:    [[TMP0:%.*]] = bitcast <1 x i64> [[B]] to <8 x i8>
+; CHECK-NEXT:    [[TMP14:%.*]] = bitcast <1 x i64> [[TMP10]] to <8 x i8>
+; CHECK-NEXT:    [[TMP1:%.*]] = bitcast <1 x i64> [[A]] to <8 x i8>
+; CHECK-NEXT:    [[TMP6:%.*]] = bitcast <8 x i8> [[TMP14]] to i64
+; CHECK-NEXT:    [[MMX_VAR_I:%.*]] = bitcast <8 x i8> [[TMP1]] to x86_mmx
+; CHECK-NEXT:    [[TMP7:%.*]] = bitcast <8 x i8> [[TMP11]] to i64
+; CHECK-NEXT:    [[MMX_VAR1_I:%.*]] = bitcast <8 x i8> [[TMP0]] to x86_mmx
+; CHECK-NEXT:    [[_MSPROP:%.*]] = or i64 [[TMP6]], [[TMP7]]
+; CHECK-NEXT:    [[TMP2:%.*]] = tail call x86_mmx @llvm.x86.mmx.pminu.b(x86_mmx [[MMX_VAR_I]], x86_mmx [[MMX_VAR1_I]]) #[[ATTR2]]
+; CHECK-NEXT:    [[TMP9:%.*]] = bitcast i64 [[_MSPROP]] to <8 x i8>
+; CHECK-NEXT:    [[TMP3:%.*]] = bitcast x86_mmx [[TMP2]] to <8 x i8>
+; CHECK-NEXT:    [[TMP4:%.*]] = bitcast <8 x i8> [[TMP9]] to <1 x i64>
+; CHECK-NEXT:    [[TMP12:%.*]] = bitcast <8 x i8> [[TMP3]] to <1 x i64>
+; CHECK-NEXT:    [[TMP5:%.*]] = extractelement <1 x i64> [[TMP4]], i32 0
+; CHECK-NEXT:    [[TMP13:%.*]] = extractelement <1 x i64> [[TMP12]], i32 0
+; CHECK-NEXT:    store i64 [[TMP5]], ptr @__msan_retval_tls, align 8
+; CHECK-NEXT:    ret i64 [[TMP13]]
+;
+entry:
+  %0 = bitcast <1 x i64> %b to <8 x i8>
+  %1 = bitcast <1 x i64> %a to <8 x i8>
+  %mmx_var.i = bitcast <8 x i8> %1 to x86_mmx
+  %mmx_var1.i = bitcast <8 x i8> %0 to x86_mmx
+  %2 = tail call x86_mmx @llvm.x86.mmx.pminu.b(x86_mmx %mmx_var.i, x86_mmx %mmx_var1.i) nounwind
+  %3 = bitcast x86_mmx %2 to <8 x i8>
+  %4 = bitcast <8 x i8> %3 to <1 x i64>
+  %5 = extractelement <1 x i64> %4, i32 0
+  ret i64 %5
+}
+
+declare x86_mmx @llvm.x86.mmx.pmaxs.w(x86_mmx, x86_mmx) nounwind readnone
+
+define i64 @test29(<1 x i64> %a, <1 x i64> %b) #0 {
+; CHECK-LABEL: define i64 @test29(
+; CHECK-SAME: <1 x i64> [[A:%.*]], <1 x i64> [[B:%.*]]) #[[ATTR1]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP8:%.*]] = load <1 x i64>, ptr inttoptr (i64 add (i64 ptrtoint (ptr @__msan_param_tls to i64), i64 8) to ptr), align 8
+; CHECK-NEXT:    [[TMP10:%.*]] = load <1 x i64>, ptr @__msan_param_tls, align 8
+; CHECK-NEXT:    call void @llvm.donothing()
+; CHECK-NEXT:    [[TMP11:%.*]] = bitcast <1 x i64> [[TMP8]] to <4 x i16>
+; CHECK-NEXT:    [[TMP0:%.*]] = bitcast <1 x i64> [[B]] to <4 x i16>
+; CHECK-NEXT:    [[TMP14:%.*]] = bitcast <1 x i64> [[TMP10]] to <4 x i16>
+; CHECK-NEXT:    [[TMP1:%.*]] = bitcast <1 x i64> [[A]] to <4 x i16>
+; CHECK-NEXT:    [[TMP6:%.*]] = bitcast <4 x i16> [[TMP14]] to i64
+; CHECK-NEXT:    [[MMX_VAR_I:%.*]] = bitcast <4 x i16> [[TMP1]] to x86_mmx
+; CHECK-NEXT:    [[TMP7:%.*]] = bitcast <4 x i16> [[TMP11]] to i64
+; CHECK-NEXT:    [[MMX_VAR1_I:%.*]] = bitcast <4 x i16> [[TMP0]] to x86_mmx
+; CHECK-NEXT:    [[_MSPROP:%.*]] = or i64 [[TMP6]], [[TMP7]]
+; CHECK-NEXT:    [[TMP2:%.*]] = tail call x86_mmx @llvm.x86.mmx.pmaxs.w(x86_mmx [[MMX_VAR_I]], x86_mmx [[MMX_VAR1_I]]) #[[ATTR2]]
+; CHECK-NEXT:    [[TMP9:%.*]] = bitcast i64 [[_MSPROP]] to <4 x i16>
+; CHECK-NEXT:    [[TMP3:%.*]] = bitcast x86_mmx [[TMP2]] to <4 x i16>
+; CHECK-NEXT:    [[TMP4:%.*]] = bitcast <4 x i16> [[TMP9]] to <1 x i64>
+; CHECK-NEXT:    [[TMP12:%.*]] = bitcast <4 x i16> [[TMP3]] to <1 x i64>
+; CHECK-NEXT:    [[TMP5:%.*]] = extractelement <1 x i64> [[TMP4]], i32 0
+; CHECK-NEXT:    [[TMP13:%.*]] = extractelement <1 x i64> [[TMP12]], i32 0
+; CHECK-NEXT:    store i64 [[TMP5]], ptr @__msan_retval_tls, align 8
+; CHECK-NEXT:    ret i64 [[TMP13]]
+;
+entry:
+  %0 = bitcast <1 x i64> %b to <4 x i16>
+  %1 = bitcast <1 x i64> %a to <4 x i16>
+  %mmx_var.i = bitcast <4 x i16> %1 to x86_mmx
+  %mmx_var1.i = bitcast <4 x i16> %0 to x86_mmx
+  %2 = tail call x86_mmx @llvm.x86.mmx.pmaxs.w(x86_mmx %mmx_var.i, x86_mmx %mmx_var1.i) nounwind
+  %3 = bitcast x86_mmx %2 to <4 x i16>
+  %4 = bitcast <4 x i16> %3 to <1 x i64>
+  %5 = extractelement <1 x i64> %4, i32 0
+  ret i64 %5
+}
+
+declare x86_mmx @llvm.x86.mmx.pmaxu.b(x86_mmx, x86_mmx) nounwind readnone
+
+define i64 @test28(<1 x i64> %a, <1 x i64> %b) #0 {
+; CHECK-LABEL: define i64 @test28(
+; CHECK-SAME: <1 x i64> [[A:%.*]], <1 x i64> [[B:%.*]]) #[[ATTR1]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP8:%.*]] = load <1 x i64>, ptr inttoptr (i64 add (i64 ptrtoint (ptr @__msan_param_tls to i64), i64 8) to ptr), align 8
+; CHECK-NEXT:    [[TMP10:%.*]] = load <1 x i64>, ptr @__msan_param_tls, align 8
+; CHECK-NEXT:    call void @llvm.donothing()
+; CHECK-NEXT:    [[TMP11:%.*]] = bitcast <1 x i64> [[TMP8]] to <8 x i8>
+; CHECK-NEXT:    [[TMP0:%.*]] = bitcast <1 x i64> [[B]] to <8 x i8>
+; CHECK-NEXT:    [[TMP14:%.*]] = bitcast <1 x i64> [[TMP10]] to <8 x i8>
+; CHECK-NEXT:    [[TMP1:%.*]] = bitcast <1 x i64> [[A]] to <8 x i8>
+; CHECK-NEXT:    [[TMP6:%.*]] = bitcast <8 x i8> [[TMP14]] to i64
+; CHECK-NEXT:    [[MMX_VAR_I:%.*]] = bitcast <8 x i8> [[TMP1]] to x86_mmx
+; CHECK-NEXT:    [[TMP7:%.*]] = bitcast <8 x i8> [[TMP11]] to i64
+; CHECK-NEXT:    [[MMX_VAR1_I:%.*]] = bitcast <8 x i8> [[TMP0]] to x86_mmx
+; CHECK-NEXT:    [[_MSPROP:%.*]] = or i64 [[TMP6]], [[TMP7]]
+; CHECK-NEXT:    [[TMP2:%.*]] = tail call x86_mmx @llvm.x86.mmx.pmaxu.b(x86_mmx [[MMX_VAR_I]], x86_mmx [[MMX_VAR1_I]]) #[[ATTR2]]
+; CHECK-NEXT:    [[TMP9:%.*]] = bitcast i64 [[_MSPROP]] to <8 x i8>
+; CHECK-NEXT:    [[TMP3:%.*]] = bitcast x86_mmx [[TMP2]] to <8 x i8>
+; CHECK-NEXT:    [[TMP4:%.*]] = bitcast <8 x i8> [[TMP9]] to <1 x i64>
+; CHECK-NEXT:    [[TMP12:%.*]] = bitcast <8 x i8> [[TMP3]] to <1 x i64>
+; CHECK-NEXT:    [[TMP5:%.*]] = extractelement <1 x i64> [[TMP4]], i32 0
+; CHECK-NEXT:    [[TMP13:%.*]] = extractelement <1 x i64> [[TMP12]], i32 0
+; CHECK-NEXT:    store i64 [[TMP5]], ptr @__msan_retval_tls, align 8
+; CHECK-NEXT:    ret i64 [[TMP13]]
+;
+entry:
+  %0 = bitcast <1 x i64> %b to <8 x i8>
+  %1 = bitcast <1 x i64> %a to <8 x i8>
+  %mmx_var.i = bitcast <8 x i8> %1 to x86_mmx
+  %mmx_var1.i = bitcast <8 x i8> %0 to x86_mmx
+  %2 = tail call x86_mmx @llvm.x86.mmx.pmaxu.b(x86_mmx %mmx_var.i, x86_mmx %mmx_var1.i) nounwind
+  %3 = bitcast x86_mmx %2 to <8 x i8>
+  %4 = bitcast <8 x i8> %3 to <1 x i64>
+  %5 = extractelement <1 x i64> %4, i32 0
+  ret i64 %5
+}
+
+declare x86_mmx @llvm.x86.mmx.pavg.w(x86_mmx, x86_mmx) nounwind readnone
+
+define i64 @test27(<1 x i64> %a, <1 x i64> %b) #0 {
+; CHECK-LABEL: define i64 @test27(
+; CHECK-SAME: <1 x i64> [[A:%.*]], <1 x i64> [[B:%.*]]) #[[ATTR1]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP8:%.*]] = load <1 x i64>, ptr inttoptr (i64 add (i64 ptrtoint (ptr @__msan_param_tls to i64), i64 8) to ptr), align 8
+; CHECK-NEXT:    [[TMP10:%.*]] = load <1 x i64>, ptr @__msan_param_tls, align 8
+; CHECK-NEXT:    call void @llvm.donothing()
+; CHECK-NEXT:    [[TMP11:%.*]] = bitcast <1 x i64> [[TMP8]] to <4 x i16>
+; CHECK-NEXT:    [[TMP0:%.*]] = bitcast <1 x i64> [[B]] to <4 x i16>
+; CHECK-NEXT:    [[TMP14:%.*]] = bitcast <1 x i64> [[TMP10]] to <4 x i16>
+; CHECK-NEXT:    [[TMP1:%.*]] = bitcast <1 x i64> [[A]] to <4 x i16>
+; CHECK-NEXT:    [[TMP6:%.*]] = bitcast <4 x i16> [[TMP14]] to i64
+; CHECK-NEXT:    [[MMX_VAR_I:%.*]] = bitcast <4 x i16> [[TMP1]] to x86_mmx
+; CHECK-NEXT:    [[TMP7:%.*]] = bitcast <4 x i16> [[TMP11]] to i64
+; CHECK-NEXT:    [[MMX_VAR1_I:%.*]] = bitcast <4 x i16> [[TMP0]] to x86_mmx
+; CHECK-NEXT:    [[_MSPROP:%.*]] = or i64 [[TMP6]], [[TMP7]]
+; CHECK-NEXT:    [[TMP2:%.*]] = tail call x86_mmx @llvm.x86.mmx.pavg.w(x86_mmx [[MMX_VAR_I]], x86_mmx [[MMX_VAR1_I]]) #[[ATTR2]]
+; CHECK-NEXT:    [[TMP9:%.*]] = bitcast i64 [[_MSPROP]] to <4 x i16>
+; CHECK-NEXT:    [[TMP3:%.*]] = bitcast x86_mmx [[TMP2]] to <4 x i16>
+; CHECK-NEXT:    [[TMP4:%.*]] = bitcast <4 x i16> [[TMP9]] to <1 x i64>
+; CHECK-NEXT:    [[TMP12:%.*]] = bitcast <4 x i16> [[TMP3]] to <1 x i64>
+; CHECK-NEXT:    [[TMP5:%.*]] = extractelement <1 x i64> [[TMP4]], i32 0
+; CHECK-NEXT:    [[TMP13:%.*]] = extractelement <1 x i64> [[TMP12]], i32 0
+; CHECK-NEXT:    store i64 [[TMP5]], ptr @__msan_retval_tls, align 8
+; CHECK-NEXT:    ret i64 [[TMP13]]
+;
+entry:
+  %0 = bitcast <1 x i64> %b to <4 x i16>
+  %1 = bitcast <1 x i64> %a to <4 x i16>
+  %mmx_var.i = bitcast <4 x i16> %1 to x86_mmx
+  %mmx_var1.i = bitcast <4 x i16> %0 to x86_mmx
+  %2 = tail call x86_mmx @llvm.x86.mmx.pavg.w(x86_mmx %mmx_var.i, x86_mmx %mmx_var1.i) nounwind
+  %3 = bitcast x86_mmx %2 to <4 x i16>
+  %4 = bitcast <4 x i16> %3 to <1 x i64>
+  %5 = extractelement <1 x i64> %4, i32 0
+  ret i64 %5
+}
+
+declare x86_mmx @llvm.x86.mmx.pavg.b(x86_mmx, x86_mmx) nounwind readnone
+
+define i64 @test26(<1 x i64> %a, <1 x i64> %b) #0 {
+; CHECK-LABEL: define i64 @test26(
+; CHECK-SAME: <1 x i64> [[A:%.*]], <1 x i64> [[B:%.*]]) #[[ATTR1]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP8:%.*]] = load <1 x i64>, ptr inttoptr (i64 add (i64 ptrtoint (ptr @__msan_param_tls to i64), i64 8) to ptr), align 8
+; CHECK-NEXT:    [[TMP10:%.*]] = load <1 x i64>, ptr @__msan_param_tls, align 8
+; CHECK-NEXT:    call void @llvm.donothing()
+; CHECK-NEXT:    [[TMP11:%.*]] = bitcast <1 x i64> [[TMP8]] to <8 x i8>
+; CHECK-NEXT:    [[TMP0:%.*]] = bitcast <1 x i64> [[B]] to <8 x i8>
+; CHECK-NEXT:    [[TMP14:%.*]] = bitcast <1 x i64> [[TMP10]] to <8 x i8>
+; CHECK-NEXT:    [[TMP1:%.*]] = bitcast <1 x i64> [[A]] to <8 x i8>
+; CHECK-NEXT:    [[TMP6:%.*]] = bitcast <8 x i8> [[TMP14]] to i64
+; CHECK-NEXT:    [[MMX_VAR_I:%.*]] = bitcast <8 x i8> [[TMP1]] to x86_mmx
+; CHECK-NEXT:    [[TMP7:%.*]] = bitcast <8 x i8> [[TMP11]] to i64
+; CHECK-NEXT:    [[MMX_VAR1_I:%.*]] = bitcast <8 x i8> [[TMP0]] to x86_mmx
+; CHECK-NEXT:    [[_MSPROP:%.*]] = or i64 [[TMP6]], [[TMP7]]
+; CHECK-NEXT:    [[TMP2:%.*]] = tail call x86_mmx @llvm.x86.mmx.pavg.b(x86_mmx [[MMX_VAR_I]], x86_mmx [[MMX_VAR1_I]]) #[[ATTR2]]
+; CHECK-NEXT:    [[TMP9:%.*]] = bitcast i64 [[_MSPROP]] to <8 x i8>
+; CHECK-NEXT:    [[TMP3:%.*]] = bitcast x86_mmx [[TMP2]] to <8 x i8>
+; CHECK-NEXT:    [[TMP4:%.*]] = bitcast <8 x i8> [[TMP9]] to <1 x i64>
+; CHECK-NEXT:    [[TMP12:%.*]] = bitcast <8 x i8> [[TMP3]] to <1 x i64>
+; CHECK-NEXT:    [[TMP5:%.*]] = extractelement <1 x i64> [[TMP4]], i32 0
+; CHECK-NEXT:    [[TMP13:%.*]] = extractelement <1 x i64> [[TMP12]], i32 0
+; CHECK-NEXT:    store i64 [[TMP5]], ptr @__msan_retval_tls, align 8
+; CHECK-NEXT:    ret i64 [[TMP13]]
+;
+entry:
+  %0 = bitcast <1 x i64> %b to <8 x i8>
+  %1 = bitcast <1 x i64> %a to <8 x i8>
+  %mmx_var.i = bitcast <8 x i8> %1 to x86_mmx
+  %mmx_var1.i = bitcast <8 x i8> %0 to x86_mmx
+  %2 = tail call x86_mmx @llvm.x86.mmx.pavg.b(x86_mmx %mmx_var.i, x86_mmx %mmx_var1.i) nounwind
+  %3 = bitcast x86_mmx %2 to <8 x i8>
+  %4 = bitcast <8 x i8> %3 to <1 x i64>
+  %5 = extractelement <1 x i64> %4, i32 0
+  ret i64 %5
+}
+
+declare void @llvm.x86.mmx.movnt.dq(ptr, x86_mmx) nounwind
+
+define void @test25(ptr %p, <1 x i64> %a) nounwind optsize ssp #0 {
+; CHECK-LABEL: define void @test25(
+; CHECK-SAME: ptr [[P:%.*]], <1 x i64> [[A:%.*]]) #[[ATTR3:[0-9]+]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP2:%.*]] = load <1 x i64>, ptr inttoptr (i64 add (i64 ptrtoint (ptr @__msan_param_tls to i64), i64 8) to ptr), align 8
+; CHECK-NEXT:    [[TMP1:%.*]] = load i64, ptr @__msan_param_tls, align 8
+; CHECK-NEXT:    call void @llvm.donothing()
+; CHECK-NEXT:    [[_MSPROP:%.*]] = extractelement <1 x i64> [[TMP2]], i32 0
+; CHECK-NEXT:    [[TMP0:%.*]] = extractelement <1 x i64> [[A]], i32 0
+; CHECK-NEXT:    [[MMX_VAR_I:%.*]] = bitcast i64 [[TMP0]] to x86_mmx
+; CHECK-NEXT:    [[_MSCMP:%.*]] = icmp ne i64 [[TMP1]], 0
+; CHECK-NEXT:    [[_MSCMP1:%.*]] = icmp ne i64 [[_MSPROP]], 0
+; CHECK-NEXT:    [[_MSOR:%.*]] = or i1 [[_MSCMP]], [[_MSCMP1]]
+; CHECK-NEXT:    br i1 [[_MSOR]], label [[TMP3:%.*]], label [[TMP4:%.*]], !prof [[PROF0:![0-9]+]]
+; CHECK:       3:
+; CHECK-NEXT:    call void @__msan_warning_noreturn() #[[ATTR6:[0-9]+]]
+; CHECK-NEXT:    unreachable
+; CHECK:       4:
+; CHECK-NEXT:    tail call void @llvm.x86.mmx.movnt.dq(ptr [[P]], x86_mmx [[MMX_VAR_I]]) #[[ATTR2]]
+; CHECK-NEXT:    ret void
+;
+entry:
+  %0 = extractelement <1 x i64> %a, i32 0
+  %mmx_var.i = bitcast i64 %0 to x86_mmx
+  tail call void @llvm.x86.mmx.movnt.dq(ptr %p, x86_mmx %mmx_var.i) nounwind
+  ret void
+}
+
+declare i32 @llvm.x86.mmx.pmovmskb(x86_mmx) nounwind readnone
+
+define i32 @test24(<1 x i64> %a) #0 {
+; CHECK-LABEL: define i32 @test24(
+; CHECK-SAME: <1 x i64> [[A:%.*]]) #[[ATTR1]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP2:%.*]] = load <1 x i64>, ptr @__msan_param_tls, align 8
+; CHECK-NEXT:    call void @llvm.donothing()
+; CHECK-NEXT:    [[TMP6:%.*]] = bitcast <1 x i64> [[TMP2]] to <8 x i8>
+; CHECK-NEXT:    [[TMP0:%.*]] = bitcast <1 x i64> [[A]] to <8 x i8>
+; CHECK-NEXT:    [[TMP3:%.*]] = bitcast <8 x i8> [[TMP6]] to i64
+; CHECK-NEXT:    [[MMX_VAR_I:%.*]] = bitcast <8 x i8> [[TMP0]] to x86_mmx
+; CHECK-NEXT:    [[_MSCMP:%.*]] = icmp ne i64 [[TMP3]], 0
+; CHECK-NEXT:    br i1 [[_MSCMP]], label [[TMP4:%.*]], label [[TMP5:%.*]], !prof [[PROF0]]
+; CHECK:       4:
+; CHECK-NEXT:    call void @__msan_warning_noreturn() #[[ATTR6]]
+; CHECK-NEXT:    unreachable
+; CHECK:       5:
+; CHECK-NEXT:    [[TMP1:%.*]] = tail call i32 @llvm.x86.mmx.pmovmskb(x86_mmx [[MMX_VAR_I]]) #[[ATTR2]]
+; CHECK-NEXT:    store i32 0, ptr @__msan_retval_tls, align 8
+; CHECK-NEXT:    ret i32 [[TMP1]]
+;
+entry:
+  %0 = bitcast <1 x i64> %a to <8 x i8>
+  %mmx_var.i = bitcast <8 x i8> %0 to x86_mmx
+  %1 = tail call i32 @llvm.x86.mmx.pmovmskb(x86_mmx %mmx_var.i) nounwind
+  ret i32 %1
+}
+
+declare void @llvm.x86.mmx.maskmovq(x86_mmx, x86_mmx, ptr) nounwind
+
+define void @test23(<1 x i64> %d, <1 x i64> %n, ptr %p) nounwind optsize ssp #0 {
+; CHECK-LABEL: define void @test23(
+; CHECK-SAME: <1 x i64> [[D:%.*]], <1 x i64> [[N:%.*]], ptr [[P:%.*]]) #[[ATTR3]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP4:%.*]] = load <1 x i64>, ptr inttoptr (i64 add (i64 ptrtoint (ptr @__msan_param_tls to i64), i64 8) to ptr), align 8
+; CHECK-NEXT:    [[TMP6:%.*]] = load <1 x i64>, ptr @__msan_param_tls, align 8
+; CHECK-NEXT:    [[TMP2:%.*]] = load i64, ptr inttoptr (i64 add (i64 ptrtoint (ptr @__msan_param_tls to i64), i64 16) to ptr), align 8
+; CHECK-NEXT:    call void @llvm.donothing()
+; CHECK-NEXT:    [[TMP3:%.*]] = bitcast <1 x i64> [[TMP4]] to <8 x i8>
+; CHECK-NEXT:    [[TMP0:%.*]] = bitcast <1 x i64> [[N]] to <8 x i8>
+; CHECK-NEXT:    [[TMP5:%.*]] = bitcast <1 x i64> [[TMP6]] to <8 x i8>
+; CHECK-NEXT:    [[TMP1:%.*]] = bitcast <1 x i64> [[D]] to <8 x i8>
+; CHECK-NEXT:    [[TMP7:%.*]] = bitcast <8 x i8> [[TMP5]] to i64
+; CHECK-NEXT:    [[MMX_VAR_I:%.*]] = bitcast <8 x i8> [[TMP1]] to x86_mmx
+; CHECK-NEXT:    [[TMP8:%.*]] = bitcast <8 x i8> [[TMP3]] to i64
+; CHECK-NEXT:    [[MMX_VAR1_I:%.*]] = bitcast <8 x i8> [[TMP0]] to x86_mmx
+; CHECK-NEXT:    [[_MSCMP:%.*]] = icmp ne i64 [[TMP7]], 0
+; CHECK-NEXT:    [[_MSCMP1:%.*]] = icmp ne i64 [[TMP8]], 0
+; CHECK-NEXT:    [[_MSOR:%.*]] = or i1 [[_MSCMP]], [[_MSCMP1]]
+; CHECK-NEXT:    [[_MSCMP2:%.*]] = icmp ne i64 [[TMP2]], 0
+; CHECK-NEXT:    [[_MSOR3:%.*]] = or i1 [[_MSOR]], [[_MSCMP2]]
+; CHECK-NEXT:    br i1 [[_MSOR3]], label [[TMP9:%.*]], label [[TMP10:%.*]], !prof [[PROF0]]
+; CHECK:       9:
+; CHECK-NEXT:    call void @__msan_warning_noreturn() #[[ATTR6]]
+; CHECK-NEXT:    unreachable
+; CHECK:       10:
+; CHECK-NEXT:    tail call void @llvm.x86.mmx.maskmovq(x86_mmx [[MMX_VAR_I]], x86_mmx [[MMX_VAR1_I]], ptr [[P]]) #[[ATTR2]]
+; CHECK-NEXT:    ret void
+;
+entry:
+  %0 = bitcast <1 x i64> %n to <8 x i8>
+  %1 = bitcast <1 x i64> %d to <8 x i8>
+  %mmx_var.i = bitcast <8 x i8> %1 to x86_mmx
+  %mmx_var1.i = bitcast <8 x i8> %0 to x86_mmx
+  tail call void @llvm.x86.mmx.maskmovq(x86_mmx %mmx_var.i, x86_mmx %mmx_var1.i, ptr %p) nounwind
+  ret void
+}
+
+declare x86_mmx @llvm.x86.mmx.pmulhu.w(x86_mmx, x86_mmx) nounwind readnone
+
+define i64 @test22(<1 x i64> %a, <1 x i64> %b) #0 {
+; CHECK-LABEL: define i64 @test22(
+; CHECK-SAME: <1 x i64> [[A:%.*]], <1 x i64> [[B:%.*]]) #[[ATTR1]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP8:%.*]] = load <1 x i64>, ptr inttoptr (i64 add (i64 ptrtoint (ptr @__msan_param_tls to i64), i64 8) to ptr), align 8
+; CHECK-NEXT:    [[TMP10:%.*]] = load <1 x i64>, ptr @__msan_param_tls, align 8
+; CHECK-NEXT:    call void @llvm.donothing()
+; CHECK-NEXT:    [[TMP11:%.*]] = bitcast <1 x i64> [[TMP8]] to <4 x i16>
+; CHECK-NEXT:    [[TMP0:%.*]] = bitcast <1 x i64> [[B]] to <4 x i16>
+; CHECK-NEXT:    [[TMP14:%.*]] = bitcast <1 x i64> [[TMP10]] to <4 x i16>
+; CHECK-NEXT:    [[TMP1:%.*]] = bitcast <1 x i64> [[A]] to <4 x i16>
+; CHECK-NEXT:    [[TMP6:%.*]] = bitcast <4 x i16> [[TMP14]] to i64
+; CHECK-NEXT:    [[MMX_VAR_I:%.*]] = bitcast <4 x i16> [[TMP1]] to x86_mmx
+; CHECK-NEXT:    [[TMP7:%.*]] = bitcast <4 x i16> [[TMP11]] to i64
+; CHECK-NEXT:    [[MMX_VAR1_I:%.*]] = bitcast <4 x i16> [[TMP0]] to x86_mmx
+; CHECK-NEXT:    [[_MSPROP:%.*]] = or i64 [[TMP6]], [[TMP7]]
+; CHECK-NEXT:    [[TMP2:%.*]] = tail call x86_mmx @llvm.x86.mmx.pmulhu.w(x86_mmx [[MMX_VAR_I]], x86_mmx [[MMX_VAR1_I]]) #[[ATTR2]]
+; CHECK-NEXT:    [[TMP9:%.*]] = bitcast i64 [[_MSPROP]] to <4 x i16>
+; CHECK-NEXT:    [[TMP3:%.*]] = bitcast x86_mmx [[TMP2]] to <4 x i16>
+; CHECK-NEXT:    [[TMP4:%.*]] = bitcast <4 x i16> [[TMP9]] to <1 x i64>
+; CHECK-NEXT:    [[TMP12:%.*]] = bitcast <4 x i16> [[TMP3]] to <1 x i64>
+; CHECK-NEXT:    [[TMP5:%.*]] = extractelement <1 x i64> [[TMP4]], i32 0
+; CHECK-NEXT:    [[TMP13:%.*]] = extractelement <1 x i64> [[TMP12]], i32 0
+; CHECK-NEXT:    store i64 [[TMP5]], ptr @__msan_retval_tls, align 8
+; CHECK-NEXT:    ret i64 [[TMP13]]
+;
+entry:
+  %0 = bitcast <1 x i64> %b to <4 x i16>
+  %1 = bitcast <1 x i64> %a to <4 x i16>
+  %mmx_var.i = bitcast <4 x i16> %1 to x86_mmx
+  %mmx_var1.i = bitcast <4 x i16> %0 to x86_mmx
+  %2 = tail call x86_mmx @llvm.x86.mmx.pmulhu.w(x86_mmx %mmx_var.i, x86_mmx %mmx_var1.i) nounwind
+  %3 = bitcast x86_mmx %2 to <4 x i16>
+  %4 = bitcast <4 x i16> %3 to <1 x i64>
+  %5 = extractelement <1 x i64> %4, i32 0
+  ret i64 %5
+}
+
+declare x86_mmx @llvm.x86.sse.pshuf.w(x86_mmx, i8) nounwind readnone
+
+define i64 @test21(<1 x i64> %a) #0 {
+; CHECK-LABEL: define i64 @test21(
+; CHECK-SAME: <1 x i64> [[A:%.*]]) #[[ATTR1]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP7:%.*]] = load <1 x i64>, ptr @__msan_param_tls, align 8
+; CHECK-NEXT:    call void @llvm.donothing()
+; CHECK-NEXT:    [[TMP8:%.*]] = bitcast <1 x i64> [[TMP7]] to <4 x i16>
+; CHECK-NEXT:    [[TMP0:%.*]] = bitcast <1 x i64> [[A]] to <4 x i16>
+; CHECK-NEXT:    [[TMP9:%.*]] = bitcast <4 x i16> [[TMP8]] to i64
+; CHECK-NEXT:    [[TMP1:%.*]] = bitcast <4 x i16> [[TMP0]] to x86_mmx
+; CHECK-NEXT:    [[_MSCMP:%.*]] = icmp ne i64 [[TMP9]], 0
+; CHECK-NEXT:    br i1 [[_MSCMP]], label [[TMP10:%.*]], label [[TMP6:%.*]], !prof [[PROF0]]
+; CHECK:       5:
+; CHECK-NEXT:    call void @__msan_warning_noreturn() #[[ATTR6]]
+; CHECK-NEXT:    unreachable
+; CHECK:       6:
+; CHECK-NEXT:    [[TMP2:%.*]] = tail call x86_mmx @llvm.x86.sse.pshuf.w(x86_mmx [[TMP1]], i8 3) #[[ATTR5]]
+; CHECK-NEXT:    [[TMP3:%.*]] = bitcast x86_mmx [[TMP2]] to <4 x i16>
+; CHECK-NEXT:    [[TMP4:%.*]] = bitcast <4 x i16> [[TMP3]] to <1 x i64>
+; CHECK-NEXT:    [[TMP5:%.*]] = extractelement <1 x i64> [[TMP4]], i32 0
+; CHECK-NEXT:    store i64 0, ptr @__msan_retval_tls, align 8
+; CHECK-NEXT:    ret i64 [[TMP5]]
+;
+entry:
+  %0 = bitcast <1 x i64> %a to <4 x i16>
+  %1 = bitcast <4 x i16> %0 to x86_mmx
+  %2 = tail call x86_mmx @llvm.x86.sse.pshuf.w(x86_mmx %1, i8 3) nounwind readnone
+  %3 = bitcast x86_mmx %2 to <4 x i16>
+  %4 = bitcast <4 x i16> %3 to <1 x i64>
+  %5 = extractelement <1 x i64> %4, i32 0
+  ret i64 %5
+}
+
+define i32 @test21_2(<1 x i64> %a) #0 {
+; CHECK-LABEL: define i32 @test21_2(
+; CHECK-SAME: <1 x i64> [[A:%.*]]) #[[ATTR1]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP7:%.*]] = load <1 x i64>, ptr @__msan_param_tls, align 8
+; CHECK-NEXT:    call void @llvm.donothing()
+; CHECK-NEXT:    [[TMP8:%.*]] = bitcast <1 x i64> [[TMP7]] to <4 x i16>
+; CHECK-NEXT:    [[TMP0:%.*]] = bitcast <1 x i64> [[A]] to <4 x i16>
+; CHECK-NEXT:    [[TMP9:%.*]] = bitcast <4 x i16> [[TMP8]] to i64
+; CHECK-NEXT:    [[TMP1:%.*]] = bitcast <4 x i16> [[TMP0]] to x86_mmx
+; CHECK-NEXT:    [[_MSCMP:%.*]] = icmp ne i64 [[TMP9]], 0
+; CHECK-NEXT:    br i1 [[_MSCMP]], label [[TMP10:%.*]], label [[TMP6:%.*]], !prof [[PROF0]]
+; CHECK:       5:
+; CHECK-NEXT:    call void @__msan_warning_noreturn() #[[ATTR6]]
+; CHECK-NEXT:    unreachable
+; CHECK:       6:
+; CHECK-NEXT:    [[TMP2:%.*]] = tail call x86_mmx @llvm.x86.sse.pshuf.w(x86_mmx [[TMP1]], i8 3) #[[ATTR5]]
+; CHECK-NEXT:    [[TMP3:%.*]] = bitcast x86_mmx [[TMP2]] to <4 x i16>
+; CHECK-NEXT:    [[TMP4:%.*]] = bitcast <4 x i16> [[TMP3]] to <2 x i32>
+; CHECK-NEXT:    [[TMP5:%.*]] = extractelement <2 x i32> [[TMP4]], i32 0
+; CHECK-NEXT:    store i32 0, ptr @__msan_retval_tls, align 8
+; CHECK-NEXT:    ret i32 [[TMP5]]
+;
+entry:
+  %0 = bitcast <1 x i64> %a to <4 x i16>
+  %1 = bitcast <4 x i16> %0 to x86_mmx
+  %2 = tail call x86_mmx @llvm.x86.sse.pshuf.w(x86_mmx %1, i8 3) nounwind readnone
+  %3 = bitcast x86_mmx %2 to <4 x i16>
+  %4 = bitcast <4 x i16> %3 to <2 x i32>
+  %5 = extractelement <2 x i32> %4, i32 0
+  ret i32 %5
+}
+
+declare x86_mmx @llvm.x86.mmx.pmulu.dq(x86_mmx, x86_mmx) nounwind readnone
+
+define i64 @test20(<1 x i64> %a, <1 x i64> %b) #0 {
+; CHECK-LABEL: define i64 @test20(
+; CHECK-SAME: <1 x i64> [[A:%.*]], <1 x i64> [[B:%.*]]) #[[ATTR1]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP5:%.*]] = load <1 x i64>, ptr inttoptr (i64 add (i64 ptrtoint (ptr @__msan_param_tls to i64), i64 8) to ptr), align 8
+; CHECK-NEXT:    [[TMP8:%.*]] = load <1 x i64>, ptr @__msan_param_tls, align 8
+; CHECK-NEXT:    call void @llvm.donothing()
+; CHECK-NEXT:    [[TMP9:%.*]] = bitcast <1 x i64> [[TMP5]] to <2 x i32>
+; CHECK-NEXT:    [[TMP0:%.*]] = bitcast <1 x i64> [[B]] to <2 x i32>
+; CHECK-NEXT:    [[TMP4:%.*]] = bitcast <1 x i64> [[TMP8]] to <2 x i32>
+; CHECK-NEXT:    [[TMP1:%.*]] = bitcast <1 x i64> [[A]] to <2 x i32>
+; CHECK-NEXT:    [[TMP6:%.*]] = bitcast <2 x i32> [[TMP4]] to i64
+; CHECK-NEXT:    [[MMX_VAR_I:%.*]] = bitcast <2 x i32> [[TMP1]] to x86_mmx
+; CHECK-NEXT:    [[TMP7:%.*]] = bitcast <2 x i32> [[TMP9]] to i64
+; CHECK-NEXT:    [[MMX_VAR1_I:%.*]] = bitcast <2 x i32> [[TMP0]] to x86_mmx
+; CHECK-NEXT:    [[_MSPROP:%.*]] = or i64 [[TMP6]], [[TMP7]]
+; CHECK-NEXT:    [[TMP2:%.*]] = tail call x86_mmx @llvm.x86.mmx.pmulu.dq(x86_mmx [[MMX_VAR_I]], x86_mmx [[MMX_VAR1_I]]) #[[ATTR2]]
+; CHECK-NEXT:    [[TMP3:%.*]] = bitcast x86_mmx [[TMP2]] to i64
+; CHECK-NEXT:    store i64 [[_MSPROP]], ptr @__msan_retval_tls, align 8
+; CHECK-NEXT:    ret i64 [[TMP3]]
+;
+entry:
+  %0 = bitcast <1 x i64> %b to <2 x i32>
+  %1 = bitcast <1 x i64> %a to <2 x i32>
+  %mmx_var.i = bitcast <2 x i32> %1 to x86_mmx
+  %mmx_var1.i = bitcast <2 x i32> %0 to x86_mmx
+  %2 = tail call x86_mmx @llvm.x86.mmx.pmulu.dq(x86_mmx %mmx_var.i, x86_mmx %mmx_var1.i) nounwind
+  %3 = bitcast x86_mmx %2 to i64
+  ret i64 %3
+}
+
+declare <2 x double> @llvm.x86.sse.cvtpi2pd(x86_mmx) nounwind readnone
+
+define <2 x double> @test19(<1 x i64> %a) #0 {
+; CHECK-LABEL: define <2 x double> @test19(
+; CHECK-SAME: <1 x i64> [[A:%.*]]) #[[ATTR1]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP4:%.*]] = load <1 x i64>, ptr @__msan_param_tls, align 8
+; CHECK-NEXT:    call void @llvm.donothing()
+; CHECK-NEXT:    [[TMP7:%.*]] = bitcast <1 x i64> [[TMP4]] to <2 x i32>
+; CHECK-NEXT:    [[TMP0:%.*]] = bitcast <1 x i64> [[A]] to <2 x i32>
+; CHECK-NEXT:    [[TMP3:%.*]] = bitcast <2 x i32> [[TMP7]] to i64
+; CHECK-NEXT:    [[TMP1:%.*]] = bitcast <2 x i32> [[TMP0]] to x86_mmx
+; CHECK-NEXT:    [[_MSCMP:%.*]] = icmp ne i64 [[TMP3]], 0
+; CHECK-NEXT:    br i1 [[_MSCMP]], label [[TMP5:%.*]], label [[TMP6:%.*]], !prof [[PROF0]]
+; CHECK:       5:
+; CHECK-NEXT:    call void @__msan_warning_noreturn() #[[ATTR6]]
+; CHECK-NEXT:    unreachable
+; CHECK:       6:
+; CHECK-NEXT:    [[TMP2:%.*]] = tail call <2 x double> @llvm.x86.sse.cvtpi2pd(x86_mmx [[TMP1]]) #[[ATTR5]]
+; CHECK-NEXT:    store <2 x i64> zeroinitializer, ptr @__msan_retval_tls, align 8
+; CHECK-NEXT:    ret <2 x double> [[TMP2]]
+;
+entry:
+  %0 = bitcast <1 x i64> %a to <2 x i32>
+  %1 = bitcast <2 x i32> %0 to x86_mmx
+  %2 = tail call <2 x double> @llvm.x86.sse.cvtpi2pd(x86_mmx %1) nounwind readnone
+  ret <2 x double> %2
+}
+
+declare x86_mmx @llvm.x86.sse.cvttpd2pi(<2 x double>) nounwind readnone
+
+define i64 @test18(<2 x double> %a) #0 {
+; CHECK-LABEL: define i64 @test18(
+; CHECK-SAME: <2 x double> [[A:%.*]]) #[[ATTR1]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP4:%.*]] = load <2 x i64>, ptr @__msan_param_tls, align 8
+; CHECK-NEXT:    call void @llvm.donothing()
+; CHECK-NEXT:    [[TMP5:%.*]] = bitcast <2 x i64> [[TMP4]] to i128
+; CHECK-NEXT:    [[_MSCMP:%.*]] = icmp ne i128 [[TMP5]], 0
+; CHECK-NEXT:    br i1 [[_MSCMP]], label [[TMP6:%.*]], label [[TMP7:%.*]], !prof [[PROF0]]
+; CHECK:       2:
+; CHECK-NEXT:    call void @__msan_warning_noreturn() #[[ATTR6]]
+; CHECK-NEXT:    unreachable
+; CHECK:       3:
+; CHECK-NEXT:    [[TMP0:%.*]] = tail call x86_mmx @llvm.x86.sse.cvttpd2pi(<2 x double> [[A]]) #[[ATTR5]]
+; CHECK-NEXT:    [[TMP1:%.*]] = bitcast x86_mmx [[TMP0]] to <2 x i32>
+; CHECK-NEXT:    [[TMP2:%.*]] = bitcast <2 x i32> [[TMP1]] to <1 x i64>
+; CHECK-NEXT:    [[TMP3:%.*]] = extractelement <1 x i64> [[TMP2]], i32 0
+; CHECK-NEXT:    store i64 0, ptr @__msan_retval_tls, align 8
+; CHECK-NEXT:    ret i64 [[TMP3]]
+;
+entry:
+  %0 = tail call x86_mmx @llvm.x86.sse.cvttpd2pi(<2 x double> %a) nounwind readnone
+  %1 = bitcast x86_mmx %0 to <2 x i32>
+  %2 = bitcast <2 x i32> %1 to <1 x i64>
+  %3 = extractelement <1 x i64> %2, i32 0
+  ret i64 %3
+}
+
+declare x86_mmx @llvm.x86.sse.cvtpd2pi(<2 x double>) nounwind readnone
+
+define i64 @test17(<2 x double> %a) #0 {
+; CHECK-LABEL: define i64 @test17(
+; CHECK-SAME: <2 x double> [[A:%.*]]) #[[ATTR1]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP4:%.*]] = load <2 x i64>, ptr @__msan_param_tls, align 8
+; CHECK-NEXT:    call void @llvm.donothing()
+; CHECK-NEXT:    [[TMP5:%.*]] = bitcast <2 x i64> [[TMP4]] to i128
+; CHECK-NEXT:    [[_MSCMP:%.*]] = icmp ne i128 [[TMP5]], 0
+; CHECK-NEXT:    br i1 [[_MSCMP]], label [[TMP6:%.*]], label [[TMP7:%.*]], !prof [[PROF0]]
+; CHECK:       2:
+; CHECK-NEXT:    call void @__msan_warning_noreturn() #[[ATTR6]]
+; CHECK-NEXT:    unreachable
+; CHECK:       3:
+; CHECK-NEXT:    [[TMP0:%.*]] = tail call x86_mmx @llvm.x86.sse.cvtpd2pi(<2 x double> [[A]]) #[[ATTR5]]
+; CHECK-NEXT:    [[TMP1:%.*]] = bitcast x86_mmx [[TMP0]] to <2 x i32>
+; CHECK-NEXT:    [[TMP2:%.*]] = bitcast <2 x i32> [[TMP1]] to <1 x i64>
+; CHECK-NEXT:    [[TMP3:%.*]] = extractelement <1 x i64> [[TMP2]], i32 0
+; CHECK-NEXT:    store i64 0, ptr @__msan_retval_tls, align 8
+; CHECK-NEXT:    ret i64 [[TMP3]]
+;
+entry:
+  %0 = tail call x86_mmx @llvm.x86.sse.cvtpd2pi(<2 x double> %a) nounwind readnone
+  %1 = bitcast x86_mmx %0 to <2 x i32>
+  %2 = bitcast <2 x i32> %1 to <1 x i64>
+  %3 = extractelement <1 x i64> %2, i32 0
+  ret i64 %3
+}
+
+declare x86_mmx @llvm.x86.mmx.palignr.b(x86_mmx, x86_mmx, i8) nounwind readnone
+
+define i64 @test16(<1 x i64> %a, <1 x i64> %b) #0 {
+; CHECK-LABEL: define i64 @test16(
+; CHECK-SAME: <1 x i64> [[A:%.*]], <1 x i64> [[B:%.*]]) #[[ATTR1]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP6:%.*]] = load <1 x i64>, ptr @__msan_param_tls, align 8
+; CHECK-NEXT:    [[TMP7:%.*]] = load <1 x i64>, ptr inttoptr (i64 add (i64 ptrtoint (ptr @__msan_param_tls to i64), i64 8) to ptr), align 8
+; CHECK-NEXT:    call void @llvm.donothing()
+; CHECK-NEXT:    [[_MSPROP:%.*]] = extractelement <1 x i64> [[TMP6]], i32 0
+; CHECK-NEXT:    [[TMP0:%.*]] = extractelement <1 x i64> [[A]], i32 0
+; CHECK-NEXT:    [[MMX_VAR:%.*]] = bitcast i64 [[TMP0]] to x86_mmx
+; CHECK-NEXT:    [[_MSPROP1:%.*]] = extractelement <1 x i64> [[TMP7]], i32 0
+; CHECK-NEXT:    [[TMP1:%.*]] = extractelement <1 x i64> [[B]], i32 0
+; CHECK-NEXT:    [[MMX_VAR1:%.*]] = bitcast i64 [[TMP1]] to x86_mmx
+; CHECK-NEXT:    [[_MSCMP:%.*]] = icmp ne i64 [[_MSPROP]], 0
+; CHECK-NEXT:    [[_MSCMP2:%.*]] = icmp ne i64 [[_MSPROP1]], 0
+; CHECK-NEXT:    [[_MSOR:%.*]] = or i1 [[_MSCMP]], [[_MSCMP2]]
+; CHECK-NEXT:    br i1 [[_MSOR]], label [[TMP4:%.*]], label [[TMP5:%.*]], !prof [[PROF0]]
+; CHECK:       4:
+; CHECK-NEXT:    call void @__msan_warning_noreturn() #[[ATTR6]]
+; CHECK-NEXT:    unreachable
+; CHECK:       5:
+; CHECK-NEXT:    [[TMP2:%.*]] = tail call x86_mmx @llvm.x86.mmx.palignr.b(x86_mmx [[MMX_VAR]], x86_mmx [[MMX_VAR1]], i8 16)
+; CHECK-NEXT:    [[TMP3:%.*]] = bitcast x86_mmx [[TMP2]] to i64
+; CHECK-NEXT:    store i64 0, ptr @__msan_retval_tls, align 8
+; CHECK-NEXT:    ret i64 [[TMP3]]
+;
+entry:
+  %0 = extractelement <1 x i64> %a, i32 0
+  %mmx_var = bitcast i64 %0 to x86_mmx
+  %1 = extractelement <1 x i64> %b, i32 0
+  %mmx_var1 = bitcast i64 %1 to x86_mmx
+  %2 = tail call x86_mmx @llvm.x86.mmx.palignr.b(x86_mmx %mmx_var, x86_mmx %mmx_var1, i8 16)
+  %3 = bitcast x86_mmx %2 to i64
+  ret i64 %3
+}
+
+declare x86_mmx @llvm.x86.ssse3.pabs.d(x86_mmx) nounwind readnone
+
+define i64 @test15(<1 x i64> %a) #0 {
+; CHECK-LABEL: define i64 @test15(
+; CHECK-SAME: <1 x i64> [[A:%.*]]) #[[ATTR1]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP7:%.*]] = load <1 x i64>, ptr @__msan_param_tls, align 8
+; CHECK-NEXT:    call void @llvm.donothing()
+; CHECK-NEXT:    [[TMP8:%.*]] = bitcast <1 x i64> [[TMP7]] to <2 x i32>
+; CHECK-NEXT:    [[TMP0:%.*]] = bitcast <1 x i64> [[A]] to <2 x i32>
+; CHECK-NEXT:    [[TMP11:%.*]] = bitcast <2 x i32> [[TMP8]] to i64
+; CHECK-NEXT:    [[TMP1:%.*]] = bitcast <2 x i32> [[TMP0]] to x86_mmx
+; CHECK-NEXT:    [[TMP2:%.*]] = tail call x86_mmx @llvm.x86.ssse3.pabs.d(x86_mmx [[TMP1]]) #[[ATTR5]]
+; CHECK-NEXT:    [[TMP6:%.*]] = bitcast i64 [[TMP11]] to <2 x i32>
+; CHECK-NEXT:    [[TMP3:%.*]] = bitcast x86_mmx [[TMP2]] to <2 x i32>
+; CHECK-NEXT:    [[TMP4:%.*]] = bitcast <2 x i32> [[TMP6]] to <1 x i64>
+; CHECK-NEXT:    [[TMP9:%.*]] = bitcast <2 x i32> [[TMP3]] to <1 x i64>
+; CHECK-NEXT:    [[TMP5:%.*]] = extractelement <1 x i64> [[TMP4]], i32 0
+; CHECK-NEXT:    [[TMP10:%.*]] = extractelement <1 x i64> [[TMP9]], i32 0
+; CHECK-NEXT:    store i64 [[TMP5]], ptr @__msan_retval_tls, align 8
+; CHECK-NEXT:    ret i64 [[TMP10]]
+;
+entry:
+  %0 = bitcast <1 x i64> %a to <2 x i32>
+  %1 = bitcast <2 x i32> %0 to x86_mmx
+  %2 = tail call x86_mmx @llvm.x86.ssse3.pabs.d(x86_mmx %1) nounwind readnone
+  %3 = bitcast x86_mmx %2 to <2 x i32>
+  %4 = bitcast <2 x i32> %3 to <1 x i64>
+  %5 = extractelement <1 x i64> %4, i32 0
+  ret i64 %5
+}
+
+declare x86_mmx @llvm.x86.ssse3.pabs.w(x86_mmx) nounwind readnone
+
+define i64 @test14(<1 x i64> %a) #0 {
+; CHECK-LABEL: define i64 @test14(
+; CHECK-SAME: <1 x i64> [[A:%.*]]) #[[ATTR1]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP7:%.*]] = load <1 x i64>, ptr @__msan_param_tls, align 8
+; CHECK-NEXT:    call void @llvm.donothing()
+; CHECK-NEXT:    [[TMP8:%.*]] = bitcast <1 x i64> [[TMP7]] to <4 x i16>
+; CHECK-NEXT:    [[TMP0:%.*]] = bitcast <1 x i64> [[A]] to <4 x i16>
+; CHECK-NEXT:    [[TMP11:%.*]] = bitcast <4 x i16> [[TMP8]] to i64
+; CHECK-NEXT:    [[TMP1:%.*]] = bitcast <4 x i16> [[TMP0]] to x86_mmx
+; CHECK-NEXT:    [[TMP2:%.*]] = tail call x86_mmx @llvm.x86.ssse3.pabs.w(x86_mmx [[TMP1]]) #[[ATTR5]]
+; CHECK-NEXT:    [[TMP6:%.*]] = bitcast i64 [[TMP11]] to <4 x i16>
+; CHECK-NEXT:    [[TMP3:%.*]] = bitcast x86_mmx [[TMP2]] to <4 x i16>
+; CHECK-NEXT:    [[TMP4:%.*]] = bitcast <4 x i16> [[TMP6]] to <1 x i64>
+; CHECK-NEXT:    [[TMP9:%.*]] = bitcast <4 x i16> [[TMP3]] to <1 x i64>
+; CHECK-NEXT:    [[TMP5:%.*]] = extractelement <1 x i64> [[TMP4]], i32 0
+; CHECK-NEXT:    [[TMP10:%.*]] = extractelement <1 x i64> [[TMP9]], i32 0
+; CHECK-NEXT:    store i64 [[TMP5]], ptr @__msan_retval_tls, align 8
+; CHECK-NEXT:    ret i64 [[TMP10]]
+;
+entry:
+  %0 = bitcast <1 x i64> %a to <4 x i16>
+  %1 = bitcast <4 x i16> %0 to x86_mmx
+  %2 = tail call x86_mmx @llvm.x86.ssse3.pabs.w(x86_mmx %1) nounwind readnone
+  %3 = bitcast x86_mmx %2 to <4 x i16>
+  %4 = bitcast <4 x i16> %3 to <1 x i64>
+  %5 = extractelement <1 x i64> %4, i32 0
+  ret i64 %5
+}
+
+declare x86_mmx @llvm.x86.ssse3.pabs.b(x86_mmx) nounwind readnone
+
+define i64 @test13(<1 x i64> %a) #0 {
+; CHECK-LABEL: define i64 @test13(
+; CHECK-SAME: <1 x i64> [[A:%.*]]) #[[ATTR1]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP7:%.*]] = load <1 x i64>, ptr @__msan_param_tls, align 8
+; CHECK-NEXT:    call void @llvm.donothing()
+; CHECK-NEXT:    [[TMP8:%.*]] = bitcast <1 x i64> [[TMP7]] to <8 x i8>
+; CHECK-NEXT:    [[TMP0:%.*]] = bitcast <1 x i64> [[A]] to <8 x i8>
+; CHECK-NEXT:    [[TMP11:%.*]] = bitcast <8 x i8> [[TMP8]] to i64
+; CHECK-NEXT:    [[TMP1:%.*]] = bitcast <8 x i8> [[TMP0]] to x86_mmx
+; CHECK-NEXT:    [[TMP2:%.*]] = tail call x86_mmx @llvm.x86.ssse3.pabs.b(x86_mmx [[TMP1]]) #[[ATTR5]]
+; CHECK-NEXT:    [[TMP6:%.*]] = bitcast i64 [[TMP11]] to <8 x i8>
+; CHECK-NEXT:    [[TMP3:%.*]] = bitcast x86_mmx [[TMP2]] to <8 x i8>
+; CHECK-NEXT:    [[TMP4:%.*]] = bitcast <8 x i8> [[TMP6]] to <1 x i64>
+; CHECK-NEXT:    [[TMP9:%.*]] = bitcast <8 x i8> [[TMP3]] to <1 x i64>
+; CHECK-NEXT:    [[TMP5:%.*]] = extractelement <1 x i64> [[TMP4]], i32 0
+; CHECK-NEXT:    [[TMP10:%.*]] = extractelement <1 x i64> [[TMP9]], i32 0
+; CHECK-NEXT:    store i64 [[TMP5]], ptr @__msan_retval_tls, align 8
+; CHECK-NEXT:    ret i64 [[TMP10]]
+;
+entry:
+  %0 = bitcast <1 x i64> %a to <8 x i8>
+  %1 = bitcast <8 x i8> %0 to x86_mmx
+  %2 = tail call x86_mmx @llvm.x86.ssse3.pabs.b(x86_mmx %1) nounwind readnone
+  %3 = bitcast x86_mmx %2 to <8 x i8>
+  %4 = bitcast <8 x i8> %3 to <1 x i64>
+  %5 = extractelement <1 x i64> %4, i32 0
+  ret i64 %5
+}
+
+declare x86_mmx @llvm.x86.ssse3.psign.d(x86_mmx, x86_mmx) nounwind readnone
+
+define i64 @test12(<1 x i64> %a, <1 x i64> %b) #0 {
+; CHECK-LABEL: define i64 @test12(
+; CHECK-SAME: <1 x i64> [[A:%.*]], <1 x i64> [[B:%.*]]) #[[ATTR1]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP9:%.*]] = load <1 x i64>, ptr inttoptr (i64 add (i64 ptrtoint (ptr @__msan_param_tls to i64), i64 8) to ptr), align 8
+; CHECK-NEXT:    [[TMP10:%.*]] = load <1 x i64>, ptr @__msan_param_tls, align 8
+; CHECK-NEXT:    call void @llvm.donothing()
+; CHECK-NEXT:    [[TMP12:%.*]] = bitcast <1 x i64> [[TMP9]] to <2 x i32>
+; CHECK-NEXT:    [[TMP0:%.*]] = bitcast <1 x i64> [[B]] to <2 x i32>
+; CHECK-NEXT:    [[TMP13:%.*]] = bitcast <1 x i64> [[TMP10]] to <2 x i32>
+; CHECK-NEXT:    [[TMP1:%.*]] = bitcast <1 x i64> [[A]] to <2 x i32>
+; CHECK-NEXT:    [[TMP16:%.*]] = bitcast <2 x i32> [[TMP13]] to i64
+; CHECK-NEXT:    [[TMP2:%.*]] = bitcast <2 x i32> [[TMP1]] to x86_mmx
+; CHECK-NEXT:    [[TMP8:%.*]] = bitcast <2 x i32> [[TMP12]] to i64
+; CHECK-NEXT:    [[TMP3:%.*]] = bitcast <2 x i32> [[TMP0]] to x86_mmx
+; CHECK-NEXT:    [[_MSPROP:%.*]] = or i64 [[TMP16]], [[TMP8]]
+; CHECK-NEXT:    [[TMP4:%.*]] = tail call x86_mmx @llvm.x86.ssse3.psign.d(x86_mmx [[TMP2]], x86_mmx [[TMP3]]) #[[ATTR5]]
+; CHECK-NEXT:    [[TMP11:%.*]] = bitcast i64 [[_MSPROP]] to <2 x i32>
+; CHECK-NEXT:    [[TMP5:%.*]] = bitcast x86_mmx [[TMP4]] to <2 x i32>
+; CHECK-NEXT:    [[TMP6:%.*]] = bitcast <2 x i32> [[TMP11]] to <1 x i64>
+; CHECK-NEXT:    [[TMP14:%.*]] = bitcast <2 x i32> [[TMP5]] to <1 x i64>
+; CHECK-NEXT:    [[TMP7:%.*]] = extractelement <1 x i64> [[TMP6]], i32 0
+; CHECK-NEXT:    [[TMP15:%.*]] = extractelement <1 x i64> [[TMP14]], i32 0
+; CHECK-NEXT:    store i64 [[TMP7]], ptr @__msan_retval_tls, align 8
+; CHECK-NEXT:    ret i64 [[TMP15]]
+;
+entry:
+  %0 = bitcast <1 x i64> %b to <2 x i32>
+  %1 = bitcast <1 x i64> %a to <2 x i32>
+  %2 = bitcast <2 x i32> %1 to x86_mmx
+  %3 = bitcast <2 x i32> %0 to x86_mmx
+  %4 = tail call x86_mmx @llvm.x86.ssse3.psign.d(x86_mmx %2, x86_mmx %3) nounwind readnone
+  %5 = bitcast x86_mmx %4 to <2 x i32>
+  %6 = bitcast <2 x i32> %5 to <1 x i64>
+  %7 = extractelement <1 x i64> %6, i32 0
+  ret i64 %7
+}
+
+declare x86_mmx @llvm.x86.ssse3.psign.w(x86_mmx, x86_mmx) nounwind readnone
+
+define i64 @test11(<1 x i64> %a, <1 x i64> %b) #0 {
+; CHECK-LABEL: define i64 @test11(
+; CHECK-SAME: <1 x i64> [[A:%.*]], <1 x i64> [[B:%.*]]) #[[ATTR1]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP9:%.*]] = load <1 x i64>, ptr inttoptr (i64 add (i64 ptrtoint (ptr @__msan_param_tls to i64), i64 8) to ptr), align 8
+; CHECK-NEXT:    [[TMP10:%.*]] = load <1 x i64>, ptr @__msan_param_tls, align 8
+; CHECK-NEXT:    call void @llvm.donothing()
+; CHECK-NEXT:    [[TMP12:%.*]] = bitcast <1 x i64> [[TMP9]] to <4 x i16>
+; CHECK-NEXT:    [[TMP0:%.*]] = bitcast <1 x i64> [[B]] to <4 x i16>
+; CHECK-NEXT:    [[TMP13:%.*]] = bitcast <1 x i64> [[TMP10]] to <4 x i16>
+; CHECK-NEXT:    [[TMP1:%.*]] = bitcast <1 x i64> [[A]] to <4 x i16>
+; CHECK-NEXT:    [[TMP16:%.*]] = bitcast <4 x i16> [[TMP13]] to i64
+; CHECK-NEXT:    [[TMP2:%.*]] = bitcast <4 x i16> [[TMP1]] to x86_mmx
+; CHECK-NEXT:    [[TMP8:%.*]] = bitcast <4 x i16> [[TMP12]] to i64
+; CHECK-NEXT:    [[TMP3:%.*]] = bitcast <4 x i16> [[TMP0]] to x86_mmx
+; CHECK-NEXT:    [[_MSPROP:%.*]] = or i64 [[TMP16]], [[TMP8]]
+; CHECK-NEXT:    [[TMP4:%.*]] = tail call x86_mmx @llvm.x86.ssse3.psign.w(x86_mmx [[TMP2]], x86_mmx [[TMP3]]) #[[ATTR5]]
+; CHECK-NEXT:    [[TMP11:%.*]] = bitcast i64 [[_MSPROP]] to <4 x i16>
+; CHECK-NEXT:    [[TMP5:%.*]] = bitcast x86_mmx [[TMP4]] to <4 x i16>
+; CHECK-NEXT:    [[TMP6:%.*]] = bitcast <4 x i16> [[TMP11]] to <1 x i64>
+; CHECK-NEXT:    [[TMP14:%.*]] = bitcast <4 x i16> [[TMP5]] to <1 x i64>
+; CHECK-NEXT:    [[TMP7:%.*]] = extractelement <1 x i64> [[TMP6]], i32 0
+; CHECK-NEXT:    [[TMP15:%.*]] = extractelement <1 x i64> [[TMP14]], i32 0
+; CHECK-NEXT:    store i64 [[TMP7]], ptr @__msan_retval_tls, align 8
+; CHECK-NEXT:    ret i64 [[TMP15]]
+;
+entry:
+  %0 = bitcast <1 x i64> %b to <4 x i16>
+  %1 = bitcast <1 x i64> %a to <4 x i16>
+  %2 = bitcast <4 x i16> %1 to x86_mmx
+  %3 = bitcast <4 x i16> %0 to x86_mmx
+  %4 = tail call x86_mmx @llvm.x86.ssse3.psign.w(x86_mmx %2, x86_mmx %3) nounwind readnone
+  %5 = bitcast x86_mmx %4 to <4 x i16>
+  %6 = bitcast <4 x i16> %5 to <1 x i64>
+  %7 = extractelement <1 x i64> %6, i32 0
+  ret i64 %7
+}
+
+declare x86_mmx @llvm.x86.ssse3.psign.b(x86_mmx, x86_mmx) nounwind readnone
+
+define i64 @test10(<1 x i64> %a, <1 x i64> %b) #0 {
+; CHECK-LABEL: define i64 @test10(
+; CHECK-SAME: <1 x i64> [[A:%.*]], <1 x i64> [[B:%.*]]) #[[ATTR1]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP9:%.*]] = load <1 x i64>, ptr inttoptr (i64 add (i64 ptrtoint (ptr @__msan_param_tls to i64), i64 8) to ptr), align 8
+; CHECK-NEXT:    [[TMP10:%.*]] = load <1 x i64>, ptr @__msan_param_tls, align 8
+; CHECK-NEXT:    call void @llvm.donothing()
+; CHECK-NEXT:    [[TMP12:%.*]] = bitcast <1 x i64> [[TMP9]] to <8 x i8>
+; CHECK-NEXT:    [[TMP0:%.*]] = bitcast <1 x i64> [[B]] to <8 x i8>
+; CHECK-NEXT:    [[TMP13:%.*]] = bitcast <1 x i64> [[TMP10]] to <8 x i8>
+; CHECK-NEXT:    [[TMP1:%.*]] = bitcast <1 x i64> [[A]] to <8 x i8>
+; CHECK-NEXT:    [[TMP16:%.*]] = bitcast <8 x i8> [[TMP13]] to i64
+; CHECK-NEXT:    [[TMP2:%.*]] = bitcast <8 x i8> [[TMP1]] to x86_mmx
+; CHECK-NEXT:    [[TMP8:%.*]] = bitcast <8 x i8> [[TMP12]] to i64
+; CHECK-NEXT:    [[TMP3:%.*]] = bitcast <8 x i8> [[TMP0]] to x86_mmx
+; CHECK-NEXT:    [[_MSPROP:%.*]] = or i64 [[TMP16]], [[TMP8]]
+; CHECK-NEXT:    [[TMP4:%.*]] = tail call x86_mmx @llvm.x86.ssse3.psign.b(x86_mmx [[TMP2]], x86_mmx [[TMP3]]) #[[ATTR5]]
+; CHECK-NEXT:    [[TMP11:%.*]] = bitcast i64 [[_MSPROP]] to <8 x i8>
+; CHECK-NEXT:    [[TMP5:%.*]] = bitcast x86_mmx [[TMP4]] to <8 x i8>
+; CHECK-NEXT:    [[TMP6:%.*]] = bitcast <8 x i8> [[TMP11]] to <1 x i64>
+; CHECK-NEXT:    [[TMP14:%.*]] = bitcast <8 x i8> [[TMP5]] to <1 x i64>
+; CHECK-NEXT:    [[TMP7:%.*]] = extractelement <1 x i64> [[TMP6]], i32 0
+; CHECK-NEXT:    [[TMP15:%.*]] = extractelement <1 x i64> [[TMP14]], i32 0
+; CHECK-NEXT:    store i64 [[TMP7]], ptr @__msan_retval_tls, align 8
+; CHECK-NEXT:    ret i64 [[TMP15]]
+;
+entry:
+  %0 = bitcast <1 x i64> %b to <8 x i8>
+  %1 = bitcast <1 x i64> %a to <8 x i8>
+  %2 = bitcast <8 x i8> %1 to x86_mmx
+  %3 = bitcast <8 x i8> %0 to x86_mmx
+  %4 = tail call x86_mmx @llvm.x86.ssse3.psign.b(x86_mmx %2, x86_mmx %3) nounwind readnone
+  %5 = bitcast x86_mmx %4 to <8 x i8>
+  %6 = bitcast <8 x i8> %5 to <1 x i64>
+  %7 = extractelement <1 x i64> %6, i32 0
+  ret i64 %7
+}
+
+declare x86_mmx @llvm.x86.ssse3.pshuf.b(x86_mmx, x86_mmx) nounwind readnone
+
+define i64 @test9(<1 x i64> %a, <1 x i64> %b) #0 {
+; CHECK-LABEL: define i64 @test9(
+; CHECK-SAME: <1 x i64> [[A:%.*]], <1 x i64> [[B:%.*]]) #[[ATTR1]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP9:%.*]] = load <1 x i64>, ptr inttoptr (i64 add (i64 ptrtoint (ptr @__msan_param_tls to i64), i64 8) to ptr), align 8
+; CHECK-NEXT:    [[TMP10:%.*]] = load <1 x i64>, ptr @__msan_param_tls, align 8
+; CHECK-NEXT:    call void @llvm.donothing()
+; CHECK-NEXT:    [[TMP12:%.*]] = bitcast <1 x i64> [[TMP9]] to <8 x i8>
+; CHECK-NEXT:    [[TMP0:%.*]] = bitcast <1 x i64> [[B]] to <8 x i8>
+; CHECK-NEXT:    [[TMP13:%.*]] = bitcast <1 x i64> [[TMP10]] to <8 x i8>
+; CHECK-NEXT:    [[TMP1:%.*]] = bitcast <1 x i64> [[A]] to <8 x i8>
+; CHECK-NEXT:    [[TMP16:%.*]] = bitcast <8 x i8> [[TMP13]] to i64
+; CHECK-NEXT:    [[TMP2:%.*]] = bitcast <8 x i8> [[TMP1]] to x86_mmx
+; CHECK-NEXT:    [[TMP8:%.*]] = bitcast <8 x i8> [[TMP12]] to i64
+; CHECK-NEXT:    [[TMP3:%.*]] = bitcast <8 x i8> [[TMP0]] to x86_mmx
+; CHECK-NEXT:    [[_MSPROP:%.*]] = or i64 [[TMP16]], [[TMP8]]
+; CHECK-NEXT:    [[TMP4:%.*]] = tail call x86_mmx @llvm.x86.ssse3.pshuf.b(x86_mmx [[TMP2]], x86_mmx [[TMP3]]) #[[ATTR5]]
+; CHECK-NEXT:    [[TMP11:%.*]] = bitcast i64 [[_MSPROP]] to <8 x i8>
+; CHECK-NEXT:    [[TMP5:%.*]] = bitcast x86_mmx [[TMP4]] to <8 x i8>
+; CHECK-NEXT:    [[TMP6:%.*]] = bitcast <8 x i8> [[TMP11]] to <1 x i64>
+; CHECK-NEXT:    [[TMP14:%.*]] = bitcast <8 x i8> [[TMP5]] to <1 x i64>
+; CHECK-NEXT:    [[TMP7:%.*]] = extractelement <1 x i64> [[TMP6]], i32 0
+; CHECK-NEXT:    [[TMP15:%.*]] = extractelement <1 x i64> [[TMP14]], i32 0
+; CHECK-NEXT:    store i64 [[TMP7]], ptr @__msan_retval_tls, align 8
+; CHECK-NEXT:    ret i64 [[TMP15]]
+;
+entry:
+  %0 = bitcast <1 x i64> %b to <8 x i8>
+  %1 = bitcast <1 x i64> %a to <8 x i8>
+  %2 = bitcast <8 x i8> %1 to x86_mmx
+  %3 = bitcast <8 x i8> %0 to x86_mmx
+  %4 = tail call x86_mmx @llvm.x86.ssse3.pshuf.b(x86_mmx %2, x86_mmx %3) nounwind readnone
+  %5 = bitcast x86_mmx %4 to <8 x i8>
+  %6 = bitcast <8 x i8> %5 to <1 x i64>
+  %7 = extractelement <1 x i64> %6, i32 0
+  ret i64 %7
+}
+
+declare x86_mmx @llvm.x86.ssse3.pmul.hr.sw(x86_mmx, x86_mmx) nounwind readnone
+
+define i64 @test8(<1 x i64> %a, <1 x i64> %b) #0 {
+; CHECK-LABEL: define i64 @test8(
+; CHECK-SAME: <1 x i64> [[A:%.*]], <1 x i64> [[B:%.*]]) #[[ATTR1]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP9:%.*]] = load <1 x i64>, ptr inttoptr (i64 add (i64 ptrtoint (ptr @__msan_param_tls to i64), i64 8) to ptr), align 8
+; CHECK-NEXT:    [[TMP10:%.*]] = load <1 x i64>, ptr @__msan_param_tls, align 8
+; CHECK-NEXT:    call void @llvm.donothing()
+; CHECK-NEXT:    [[TMP12:%.*]] = bitcast <1 x i64> [[TMP9]] to <4 x i16>
+; CHECK-NEXT:    [[TMP0:%.*]] = bitcast <1 x i64> [[B]] to <4 x i16>
+; CHECK-NEXT:    [[TMP13:%.*]] = bitcast <1 x i64> [[TMP10]] to <4 x i16>
+; CHECK-NEXT:    [[TMP1:%.*]] = bitcast <1 x i64> [[A]] to <4 x i16>
+; CHECK-NEXT:    [[TMP16:%.*]] = bitcast <4 x i16> [[TMP13]] to i64
+; CHECK-NEXT:    [[TMP2:%.*]] = bitcast <4 x i16> [[TMP1]] to x86_mmx
+; CHECK-NEXT:    [[TMP8:%.*]] = bitcast <4 x i16> [[TMP12]] to i64
+; CHECK-NEXT:    [[TMP3:%.*]] = bitcast <4 x i16> [[TMP0]] to x86_mmx
+; CHECK-NEXT:    [[_MSPROP:%.*]] = or i64 [[TMP16]], [[TMP8]]
+; CHECK-NEXT:    [[TMP4:%.*]] = tail call x86_mmx @llvm.x86.ssse3.pmul.hr.sw(x86_mmx [[TMP2]], x86_mmx [[TMP3]]) #[[ATTR5]]
+; CHECK-NEXT:    [[TMP11:%.*]] = bitcast i64 [[_MSPROP]] to <4 x i16>
+; CHECK-NEXT:    [[TMP5:%.*]] = bitcast x86_mmx [[TMP4]] to <4 x i16>
+; CHECK-NEXT:    [[TMP6:%.*]] = bitcast <4 x i16> [[TMP11]] to <1 x i64>
+; CHECK-NEXT:    [[TMP14:%.*]] = bitcast <4 x i16> [[TMP5]] to <1 x i64>
+; CHECK-NEXT:    [[TMP7:%.*]] = extractelement <1 x i64> [[TMP6]], i32 0
+; CHECK-NEXT:    [[TMP15:%.*]] = extractelement <1 x i64> [[TMP14]], i32 0
+; CHECK-NEXT:    store i64 [[TMP7]], ptr @__msan_retval_tls, align 8
+; CHECK-NEXT:    ret i64 [[TMP15]]
+;
+entry:
+  %0 = bitcast <1 x i64> %b to <4 x i16>
+  %1 = bitcast <1 x i64> %a to <4 x i16>
+  %2 = bitcast <4 x i16> %1 to x86_mmx
+  %3 = bitcast <4 x i16> %0 to x86_mmx
+  %4 = tail call x86_mmx @llvm.x86.ssse3.pmul.hr.sw(x86_mmx %2, x86_mmx %3) nounwind readnone
+  %5 = bitcast x86_mmx %4 to <4 x i16>
+  %6 = bitcast <4 x i16> %5 to <1 x i64>
+  %7 = extractelement <1 x i64> %6, i32 0
+  ret i64 %7
+}
+
+declare x86_mmx @llvm.x86.ssse3.pmadd.ub.sw(x86_mmx, x86_mmx) nounwind readnone
+
+define i64 @test7(<1 x i64> %a, <1 x i64> %b) #0 {
+; CHECK-LABEL: define i64 @test7(
+; CHECK-SAME: <1 x i64> [[A:%.*]], <1 x i64> [[B:%.*]]) #[[ATTR1]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP9:%.*]] = load <1 x i64>, ptr inttoptr (i64 add (i64 ptrtoint (ptr @__msan_param_tls to i64), i64 8) to ptr), align 8
+; CHECK-NEXT:    [[TMP15:%.*]] = load <1 x i64>, ptr @__msan_param_tls, align 8
+; CHECK-NEXT:    call void @llvm.donothing()
+; CHECK-NEXT:    [[TMP17:%.*]] = bitcast <1 x i64> [[TMP9]] to <8 x i8>
+; CHECK-NEXT:    [[TMP0:%.*]] = bitcast <1 x i64> [[B]] to <8 x i8>
+; CHECK-NEXT:    [[TMP18:%.*]] = bitcast <1 x i64> [[TMP15]] to <8 x i8>
+; CHECK-NEXT:    [[TMP1:%.*]] = bitcast <1 x i64> [[A]] to <8 x i8>
+; CHECK-NEXT:    [[TMP21:%.*]] = bitcast <8 x i8> [[TMP18]] to i64
+; CHECK-NEXT:    [[TMP2:%.*]] = bitcast <8 x i8> [[TMP1]] to x86_mmx
+; CHECK-NEXT:    [[TMP8:%.*]] = bitcast <8 x i8> [[TMP17]] to i64
+; CHECK-NEXT:    [[TMP3:%.*]] = bitcast <8 x i8> [[TMP0]] to x86_mmx
+; CHECK-NEXT:    [[TMP10:%.*]] = or i64 [[TMP21]], [[TMP8]]
+; CHECK-NEXT:    [[TMP11:%.*]] = bitcast i64 [[TMP10]] to <4 x i16>
+; CHECK-NEXT:    [[TMP12:%.*]] = icmp ne <4 x i16> [[TMP11]], zeroinitializer
+; CHECK-NEXT:    [[TMP13:%.*]] = sext <4 x i1> [[TMP12]] to <4 x i16>
+; CHECK-NEXT:    [[TMP14:%.*]] = bitcast <4 x i16> [[TMP13]] to i64
+; CHECK-NEXT:    [[TMP4:%.*]] = tail call x86_mmx @llvm.x86.ssse3.pmadd.ub.sw(x86_mmx [[TMP2]], x86_mmx [[TMP3]]) #[[ATTR5]]
+; CHECK-NEXT:    [[TMP16:%.*]] = bitcast i64 [[TMP14]] to <8 x i8>
+; CHECK-NEXT:    [[TMP5:%.*]] = bitcast x86_mmx [[TMP4]] to <8 x i8>
+; CHECK-NEXT:    [[TMP6:%.*]] = bitcast <8 x i8> [[TMP16]] to <1 x i64>
+; CHECK-NEXT:    [[TMP19:%.*]] = bitcast <8 x i8> [[TMP5]] to <1 x i64>
+; CHECK-NEXT:    [[TMP7:%.*]] = extractelement <1 x i64> [[TMP6]], i32 0
+; CHECK-NEXT:    [[TMP20:%.*]] = extractelement <1 x i64> [[TMP19]], i32 0
+; CHECK-NEXT:    store i64 [[TMP7]], ptr @__msan_retval_tls, align 8
+; CHECK-NEXT:    ret i64 [[TMP20]]
+;
+entry:
+  %0 = bitcast <1 x i64> %b to <8 x i8>
+  %1 = bitcast <1 x i64> %a to <8 x i8>
+  %2 = bitcast <8 x i8> %1 to x86_mmx
+  %3 = bitcast <8 x i8> %0 to x86_mmx
+  %4 = tail call x86_mmx @llvm.x86.ssse3.pmadd.ub.sw(x86_mmx %2, x86_mmx %3) nounwind readnone
+  %5 = bitcast x86_mmx %4 to <8 x i8>
+  %6 = bitcast <8 x i8> %5 to <1 x i64>
+  %7 = extractelement <1 x i64> %6, i32 0
+  ret i64 %7
+}
+
+declare x86_mmx @llvm.x86.ssse3.phsub.sw(x86_mmx, x86_mmx) nounwind readnone
+
+define i64 @test6(<1 x i64> %a, <1 x i64> %b) #0 {
+; CHECK-LABEL: define i64 @test6(
+; CHECK-SAME: <1 x i64> [[A:%.*]], <1 x i64> [[B:%.*]]) #[[ATTR1]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP9:%.*]] = load <1 x i64>, ptr inttoptr (i64 add (i64 ptrtoint (ptr @__msan_param_tls to i64), i64 8) to ptr), align 8
+; CHECK-NEXT:    [[TMP10:%.*]] = load <1 x i64>, ptr @__msan_param_tls, align 8
+; CHECK-NEXT:    call void @llvm.donothing()
+; CHECK-NEXT:    [[TMP12:%.*]] = bitcast <1 x i64> [[TMP9]] to <4 x i16>
+; CHECK-NEXT:    [[TMP0:%.*]] = bitcast <1 x i64> [[B]] to <4 x i16>
+; CHECK-NEXT:    [[TMP13:%.*]] = bitcast <1 x i64> [[TMP10]] to <4 x i16>
+; CHECK-NEXT:    [[TMP1:%.*]] = bitcast <1 x i64> [[A]] to <4 x i16>
+; CHECK-NEXT:    [[TMP16:%.*]] = bitcast <4 x i16> [[TMP13]] to i64
+; CHECK-NEXT:    [[TMP2:%.*]] = bitcast <4 x i16> [[TMP1]] to x86_mmx
+; CHECK-NEXT:    [[TMP8:%.*]] = bitcast <4 x i16> [[TMP12]] to i64
+; CHECK-NEXT:    [[TMP3:%.*]] = bitcast <4 x i16> [[TMP0]] to x86_mmx
+; CHECK-NEXT:    [[_MSPROP:%.*]] = or i64 [[TMP16]], [[TMP8]]
+; CHECK-NEXT:    [[TMP4:%.*]] = tail call x86_mmx @llvm.x86.ssse3.phsub.sw(x86_mmx [[TMP2]], x86_mmx [[TMP3]]) #[[ATTR5]]
+; CHECK-NEXT:    [[TMP11:%.*]] = bitcast i64 [[_MSPROP]] to <4 x i16>
+; CHECK-NEXT:    [[TMP5:%.*]] = bitcast x86_mmx [[TMP4]] to <4 x i16>
+; CHECK-NEXT:    [[TMP6:%.*]] = bitcast <4 x i16> [[TMP11]] to <1 x i64>
+; CHECK-NEXT:    [[TMP14:%.*]] = bitcast <4 x i16> [[TMP5]] to <1 x i64>
+; CHECK-NEXT:    [[TMP7:%.*]] = extractelement <1 x i64> [[TMP6]], i32 0
+; CHECK-NEXT:    [[TMP15:%.*]] = extractelement <1 x i64> [[TMP14]], i32 0
+; CHECK-NEXT:    store i64 [[TMP7]], ptr @__msan_retval_tls, align 8
+; CHECK-NEXT:    ret i64 [[TMP15]]
+;
+entry:
+  %0 = bitcast <1 x i64> %b to <4 x i16>
+  %1 = bitcast <1 x i64> %a to <4 x i16>
+  %2 = bitcast <4 x i16> %1 to x86_mmx
+  %3 = bitcast <4 x i16> %0 to x86_mmx
+  %4 = tail call x86_mmx @llvm.x86.ssse3.phsub.sw(x86_mmx %2, x86_mmx %3) nounwind readnone
+  %5 = bitcast x86_mmx %4 to <4 x i16>
+  %6 = bitcast <4 x i16> %5 to <1 x i64>
+  %7 = extractelement <1 x i64> %6, i32 0
+  ret i64 %7
+}
+
+declare x86_mmx @llvm.x86.ssse3.phsub.d(x86_mmx, x86_mmx) nounwind readnone
+
+define i64 @test5(<1 x i64> %a, <1 x i64> %b) #0 {
+; CHECK-LABEL: define i64 @test5(
+; CHECK-SAME: <1 x i64> [[A:%.*]], <1 x i64> [[B:%.*]]) #[[ATTR1]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP9:%.*]] = load <1 x i64>, ptr inttoptr (i64 add (i64 ptrtoint (ptr @__msan_param_tls to i64), i64 8) to ptr), align 8
+; CHECK-NEXT:    [[TMP10:%.*]] = load <1 x i64>, ptr @__msan_param_tls, align 8
+; CHECK-NEXT:    call void @llvm.donothing()
+; CHECK-NEXT:    [[TMP12:%.*]] = bitcast <1 x i64> [[TMP9]] to <2 x i32>
+; CHECK-NEXT:    [[TMP0:%.*]] = bitcast <1 x i64> [[B]] to <2 x i32>
+; CHECK-NEXT:    [[TMP13:%.*]] = bitcast <1 x i64> [[TMP10]] to <2 x i32>
+; CHECK-NEXT:    [[TMP1:%.*]] = bitcast <1 x i64> [[A]] to <2 x i32>
+; CHECK-NEXT:    [[TMP16:%.*]] = bitcast <2 x i32> [[TMP13]] to i64
+; CHECK-NEXT:    [[TMP2:%.*]] = bitcast <2 x i32> [[TMP1]] to x86_mmx
+; CHECK-NEXT:    [[TMP8:%.*]] = bitcast <2 x i32> [[TMP12]] to i64
+; CHECK-NEXT:    [[TMP3:%.*]] = bitcast <2 x i32> [[TMP0]] to x86_mmx
+; CHECK-NEXT:    [[_MSPROP:%.*]] = or i64 [[TMP16]], [[TMP8]]
+; CHECK-NEXT:    [[TMP4:%.*]] = tail call x86_mmx @llvm.x86.ssse3.phsub.d(x86_mmx [[TMP2]], x86_mmx [[TMP3]]) #[[ATTR5]]
+; CHECK-NEXT:    [[TMP11:%.*]] = bitcast i64 [[_MSPROP]] to <2 x i32>
+; CHECK-NEXT:    [[TMP5:%.*]] = bitcast x86_mmx [[TMP4]] to <2 x i32>
+; CHECK-NEXT:    [[TMP6:%.*]] = bitcast <2 x i32> [[TMP11]] to <1 x i64>
+; CHECK-NEXT:    [[TMP14:%.*]] = bitcast <2 x i32> [[TMP5]] to <1 x i64>
+; CHECK-NEXT:    [[TMP7:%.*]] = extractelement <1 x i64> [[TMP6]], i32 0
+; CHECK-NEXT:    [[TMP15:%.*]] = extractelement <1 x i64> [[TMP14]], i32 0
+; CHECK-NEXT:    store i64 [[TMP7]], ptr @__msan_retval_tls, align 8
+; CHECK-NEXT:    ret i64 [[TMP15]]
+;
+entry:
+  %0 = bitcast <1 x i64> %b to <2 x i32>
+  %1 = bitcast <1 x i64> %a to <2 x i32>
+  %2 = bitcast <2 x i32> %1 to x86_mmx
+  %3 = bitcast <2 x i32> %0 to x86_mmx
+  %4 = tail call x86_mmx @llvm.x86.ssse3.phsub.d(x86_mmx %2, x86_mmx %3) nounwind readnone
+  %5 = bitcast x86_mmx %4 to <2 x i32>
+  %6 = bitcast <2 x i32> %5 to <1 x i64>
+  %7 = extractelement <1 x i64> %6, i32 0
+  ret i64 %7
+}
+
+declare x86_mmx @llvm.x86.ssse3.phsub.w(x86_mmx, x86_mmx) nounwind readnone
+
+define i64 @test4(<1 x i64> %a, <1 x i64> %b) #0 {
+; CHECK-LABEL: define i64 @test4(
+; CHECK-SAME: <1 x i64> [[A:%.*]], <1 x i64> [[B:%.*]]) #[[ATTR1]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP9:%.*]] = load <1 x i64>, ptr inttoptr (i64 add (i64 ptrtoint (ptr @__msan_param_tls to i64), i64 8) to ptr), align 8
+; CHECK-NEXT:    [[TMP10:%.*]] = load <1 x i64>, ptr @__msan_param_tls, align 8
+; CHECK-NEXT:    call void @llvm.donothing()
+; CHECK-NEXT:    [[TMP12:%.*]] = bitcast <1 x i64> [[TMP9]] to <4 x i16>
+; CHECK-NEXT:    [[TMP0:%.*]] = bitcast <1 x i64> [[B]] to <4 x i16>
+; CHECK-NEXT:    [[TMP13:%.*]] = bitcast <1 x i64> [[TMP10]] to <4 x i16>
+; CHECK-NEXT:    [[TMP1:%.*]] = bitcast <1 x i64> [[A]] to <4 x i16>
+; CHECK-NEXT:    [[TMP16:%.*]] = bitcast <4 x i16> [[TMP13]] to i64
+; CHECK-NEXT:    [[TMP2:%.*]] = bitcast <4 x i16> [[TMP1]] to x86_mmx
+; CHECK-NEXT:    [[TMP8:%.*]] = bitcast <4 x i16> [[TMP12]] to i64
+; CHECK-NEXT:    [[TMP3:%.*]] = bitcast <4 x i16> [[TMP0]] to x86_mmx
+; CHECK-NEXT:    [[_MSPROP:%.*]] = or i64 [[TMP16]], [[TMP8]]
+; CHECK-NEXT:    [[TMP4:%.*]] = tail call x86_mmx @llvm.x86.ssse3.phsub.w(x86_mmx [[TMP2]], x86_mmx [[TMP3]]) #[[ATTR5]]
+; CHECK-NEXT:    [[TMP11:%.*]] = bitcast i64 [[_MSPROP]] to <4 x i16>
+; CHECK-NEXT:    [[TMP5:%.*]] = bitcast x86_mmx [[TMP4]] to <4 x i16>
+; CHECK-NEXT:    [[TMP6:%.*]] = bitcast <4 x i16> [[TMP11]] to <1 x i64>
+; CHECK-NEXT:    [[TMP14:%.*]] = bitcast <4 x i16> [[TMP5]] to <1 x i64>
+; CHECK-NEXT:    [[TMP7:%.*]] = extractelement <1 x i64> [[TMP6]], i32 0
+; CHECK-NEXT:    [[TMP15:%.*]] = extractelement <1 x i64> [[TMP14]], i32 0
+; CHECK-NEXT:    store i64 [[TMP7]], ptr @__msan_retval_tls, align 8
+; CHECK-NEXT:    ret i64 [[TMP15]]
+;
+entry:
+  %0 = bitcast <1 x i64> %b to <4 x i16>
+  %1 = bitcast <1 x i64> %a to <4 x i16>
+  %2 = bitcast <4 x i16> %1 to x86_mmx
+  %3 = bitcast <4 x i16> %0 to x86_mmx
+  %4 = tail call x86_mmx @llvm.x86.ssse3.phsub.w(x86_mmx %2, x86_mmx %3) nounwind readnone
+  %5 = bitcast x86_mmx %4 to <4 x i16>
+  %6 = bitcast <4 x i16> %5 to <1 x i64>
+  %7 = extractelement <1 x i64> %6, i32 0
+  ret i64 %7
+}
+
+declare x86_mmx @llvm.x86.ssse3.phadd.sw(x86_mmx, x86_mmx) nounwind readnone
+
+define i64 @test3(<1 x i64> %a, <1 x i64> %b) #0 {
+; CHECK-LABEL: define i64 @test3(
+; CHECK-SAME: <1 x i64> [[A:%.*]], <1 x i64> [[B:%.*]]) #[[ATTR1]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP9:%.*]] = load <1 x i64>, ptr inttoptr (i64 add (i64 ptrtoint (ptr @__msan_param_tls to i64), i64 8) to ptr), align 8
+; CHECK-NEXT:    [[TMP10:%.*]] = load <1 x i64>, ptr @__msan_param_tls, align 8
+; CHECK-NEXT:    call void @llvm.donothing()
+; CHECK-NEXT:    [[TMP12:%.*]] = bitcast <1 x i64> [[TMP9]] to <4 x i16>
+; CHECK-NEXT:    [[TMP0:%.*]] = bitcast <1 x i64> [[B]] to <4 x i16>
+; CHECK-NEXT:    [[TMP13:%.*]] = bitcast <1 x i64> [[TMP10]] to <4 x i16>
+; CHECK-NEXT:    [[TMP1:%.*]] = bitcast <1 x i64> [[A]] to <4 x i16>
+; CHECK-NEXT:    [[TMP16:%.*]] = bitcast <4 x i16> [[TMP13]] to i64
+; CHECK-NEXT:    [[TMP2:%.*]] = bitcast <4 x i16> [[TMP1]] to x86_mmx
+; CHECK-NEXT:    [[TMP8:%.*]] = bitcast <4 x i16> [[TMP12]] to i64
+; CHECK-NEXT:    [[TMP3:%.*]] = bitcast <4 x i16> [[TMP0]] to x86_mmx
+; CHECK-NEXT:    [[_MSPROP:%.*]] = or i64 [[TMP16]], [[TMP8]]
+; CHECK-NEXT:    [[TMP4:%.*]] = tail call x86_mmx @llvm.x86.ssse3.phadd.sw(x86_mmx [[TMP2]], x86_mmx [[TMP3]]) #[[ATTR5]]
+; CHECK-NEXT:    [[TMP11:%.*]] = bitcast i64 [[_MSPROP]] to <4 x i16>
+; CHECK-NEXT:    [[TMP5:%.*]] = bitcast x86_mmx [[TMP4]] to <4 x i16>
+; CHECK-NEXT:    [[TMP6:%.*]] = bitcast <4 x i16> [[TMP11]] to <1 x i64>
+; CHECK-NEXT:    [[TMP14:%.*]] = bitcast <4 x i16> [[TMP5]] to <1 x i64>
+; CHECK-NEXT:    [[TMP7:%.*]] = extractelement <1 x i64> [[TMP6]], i32 0
+; CHECK-NEXT:    [[TMP15:%.*]] = extractelement <1 x i64> [[TMP14]], i32 0
+; CHECK-NEXT:    store i64 [[TMP7]], ptr @__msan_retval_tls, align 8
+; CHECK-NEXT:    ret i64 [[TMP15]]
+;
+entry:
+  %0 = bitcast <1 x i64> %b to <4 x i16>
+  %1 = bitcast <1 x i64> %a to <4 x i16>
+  %2 = bitcast <4 x i16> %1 to x86_mmx
+  %3 = bitcast <4 x i16> %0 to x86_mmx
+  %4 = tail call x86_mmx @llvm.x86.ssse3.phadd.sw(x86_mmx %2, x86_mmx %3) nounwind readnone
+  %5 = bitcast x86_mmx %4 to <4 x i16>
+  %6 = bitcast <4 x i16> %5 to <1 x i64>
+  %7 = extractelement <1 x i64> %6, i32 0
+  ret i64 %7
+}
+
+declare x86_mmx @llvm.x86.ssse3.phadd.d(x86_mmx, x86_mmx) nounwind readnone
+
+define i64 @test2(<1 x i64> %a, <1 x i64> %b) #0 {
+; CHECK-LABEL: define i64 @test2(
+; CHECK-SAME: <1 x i64> [[A:%.*]], <1 x i64> [[B:%.*]]) #[[ATTR1]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP9:%.*]] = load <1 x i64>, ptr inttoptr (i64 add (i64 ptrtoint (ptr @__msan_param_tls to i64), i64 8) to ptr), align 8
+; CHECK-NEXT:    [[TMP10:%.*]] = load <1 x i64>, ptr @__msan_param_tls, align 8
+; CHECK-NEXT:    call void @llvm.donothing()
+; CHECK-NEXT:    [[TMP12:%.*]] = bitcast <1 x i64> [[TMP9]] to <2 x i32>
+; CHECK-NEXT:    [[TMP0:%.*]] = bitcast <1 x i64> [[B]] to <2 x i32>
+; CHECK-NEXT:    [[TMP13:%.*]] = bitcast <1 x i64> [[TMP10]] to <2 x i32>
+; CHECK-NEXT:    [[TMP1:%.*]] = bitcast <1 x i64> [[A]] to <2 x i32>
+; CHECK-NEXT:    [[TMP16:%.*]] = bitcast <2 x i32> [[TMP13]] to i64
+; CHECK-NEXT:    [[TMP2:%.*]] = bitcast <2 x i32> [[TMP1]] to x86_mmx
+; CHECK-NEXT:    [[TMP8:%.*]] = bitcast <2 x i32> [[TMP12]] to i64
+; CHECK-NEXT:    [[TMP3:%.*]] = bitcast <2 x i32> [[TMP0]] to x86_mmx
+; CHECK-NEXT:    [[_MSPROP:%.*]] = or i64 [[TMP16]], [[TMP8]]
+; CHECK-NEXT:    [[TMP4:%.*]] = tail call x86_mmx @llvm.x86.ssse3.phadd.d(x86_mmx [[TMP2]], x86_mmx [[TMP3]]) #[[ATTR5]]
+; CHECK-NEXT:    [[TMP11:%.*]] = bitcast i64 [[_MSPROP]] to <2 x i32>
+; CHECK-NEXT:    [[TMP5:%.*]] = bitcast x86_mmx [[TMP4]] to <2 x i32>
+; CHECK-NEXT:    [[TMP6:%.*]] = bitcast <2 x i32> [[TMP11]] to <1 x i64>
+; CHECK-NEXT:    [[TMP14:%.*]] = bitcast <2 x i32> [[TMP5]] to <1 x i64>
+; CHECK-NEXT:    [[TMP7:%.*]] = extractelement <1 x i64> [[TMP6]], i32 0
+; CHECK-NEXT:    [[TMP15:%.*]] = extractelement <1 x i64> [[TMP14]], i32 0
+; CHECK-NEXT:    store i64 [[TMP7]], ptr @__msan_retval_tls, align 8
+; CHECK-NEXT:    ret i64 [[TMP15]]
+;
+entry:
+  %0 = bitcast <1 x i64> %b to <2 x i32>
+  %1 = bitcast <1 x i64> %a to <2 x i32>
+  %2 = bitcast <2 x i32> %1 to x86_mmx
+  %3 = bitcast <2 x i32> %0 to x86_mmx
+  %4 = tail call x86_mmx @llvm.x86.ssse3.phadd.d(x86_mmx %2, x86_mmx %3) nounwind readnone
+  %5 = bitcast x86_mmx %4 to <2 x i32>
+  %6 = bitcast <2 x i32> %5 to <1 x i64>
+  %7 = extractelement <1 x i64> %6, i32 0
+  ret i64 %7
+}
+
+define <4 x float> @test89(<4 x float> %a, x86_mmx %b) nounwind #0 {
+; ALL-LABEL: test89:
+; ALL:       # %bb.0:
+; ALL-NEXT:    cvtpi2ps %mm0, %xmm0
+; ALL-NEXT:    ret{{[l|q]}}
+; CHECK-LABEL: define <4 x float> @test89(
+; CHECK-SAME: <4 x float> [[A:%.*]], x86_mmx [[B:%.*]]) #[[ATTR4:[0-9]+]] {
+; CHECK-NEXT:    [[TMP1:%.*]] = load <4 x i32>, ptr @__msan_param_tls, align 8
+; CHECK-NEXT:    [[TMP2:%.*]] = load i64, ptr inttoptr (i64 add (i64 ptrtoint (ptr @__msan_param_tls to i64), i64 16) to ptr), align 8
+; CHECK-NEXT:    call void @llvm.donothing()
+; CHECK-NEXT:    [[TMP3:%.*]] = bitcast <4 x i32> [[TMP1]] to i128
+; CHECK-NEXT:    [[_MSCMP:%.*]] = icmp ne i128 [[TMP3]], 0
+; CHECK-NEXT:    [[_MSCMP1:%.*]] = icmp ne i64 [[TMP2]], 0
+; CHECK-NEXT:    [[_MSOR:%.*]] = or i1 [[_MSCMP]], [[_MSCMP1]]
+; CHECK-NEXT:    br i1 [[_MSOR]], label [[TMP4:%.*]], label [[TMP5:%.*]], !prof [[PROF0]]
+; CHECK:       4:
+; CHECK-NEXT:    call void @__msan_warning_noreturn() #[[ATTR6]]
+; CHECK-NEXT:    unreachable
+; CHECK:       5:
+; CHECK-NEXT:    [[C:%.*]] = tail call <4 x float> @llvm.x86.sse.cvtpi2ps(<4 x float> [[A]], x86_mmx [[B]])
+; CHECK-NEXT:    store <4 x i32> zeroinitializer, ptr @__msan_retval_tls, align 8
+; CHECK-NEXT:    ret <4 x float> [[C]]
+;
+  %c = tail call <4 x float> @llvm.x86.sse.cvtpi2ps(<4 x float> %a, x86_mmx %b)
+  ret <4 x float> %c
+}
+
+declare <4 x float> @llvm.x86.sse.cvtpi2ps(<4 x float>, x86_mmx) nounwind readnone
+
+define void @test90() #0 {
+; ALL-LABEL: test90:
+; ALL:       # %bb.0:
+; ALL-NEXT:    emms
+; ALL-NEXT:    ret{{[l|q]}}
+; CHECK-LABEL: define void @test90(
+; CHECK-SAME: ) #[[ATTR1]] {
+; CHECK-NEXT:    call void @llvm.donothing()
+; CHECK-NEXT:    call void @llvm.x86.mmx.emms()
+; CHECK-NEXT:    ret void
+;
+  call void @llvm.x86.mmx.emms()
+  ret void
+}
+
+declare void @llvm.x86.mmx.emms()
+
+define <1 x i64> @test_mm_insert_pi16(<1 x i64> %a.coerce, i32 %d) nounwind #0 {
+; CHECK-LABEL: define <1 x i64> @test_mm_insert_pi16(
+; CHECK-SAME: <1 x i64> [[A_COERCE:%.*]], i32 [[D:%.*]]) #[[ATTR4]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP3:%.*]] = load <1 x i64>, ptr @__msan_param_tls, align 8
+; CHECK-NEXT:    [[TMP6:%.*]] = load i32, ptr inttoptr (i64 add (i64 ptrtoint (ptr @__msan_param_tls to i64), i64 8) to ptr), align 8
+; CHECK-NEXT:    call void @llvm.donothing()
+; CHECK-NEXT:    [[TMP7:%.*]] = bitcast <1 x i64> [[TMP3]] to i64
+; CHECK-NEXT:    [[TMP0:%.*]] = bitcast <1 x i64> [[A_COERCE]] to x86_mmx
+; CHECK-NEXT:    [[_MSCMP:%.*]] = icmp ne i64 [[TMP7]], 0
+; CHECK-NEXT:    [[_MSCMP1:%.*]] = icmp ne i32 [[TMP6]], 0
+; CHECK-NEXT:    [[_MSOR:%.*]] = or i1 [[_MSCMP]], [[_MSCMP1]]
+; CHECK-NEXT:    br i1 [[_MSOR]], label [[TMP4:%.*]], label [[TMP5:%.*]], !prof [[PROF0]]
+; CHECK:       4:
+; CHECK-NEXT:    call void @__msan_warning_noreturn() #[[ATTR6]]
+; CHECK-NEXT:    unreachable
+; CHECK:       5:
+; CHECK-NEXT:    [[TMP1:%.*]] = tail call x86_mmx @llvm.x86.mmx.pinsr.w(x86_mmx [[TMP0]], i32 [[D]], i32 2)
+; CHECK-NEXT:    [[TMP2:%.*]] = bitcast x86_mmx [[TMP1]] to <1 x i64>
+; CHECK-NEXT:    store <1 x i64> zeroinitializer, ptr @__msan_retval_tls, align 8
+; CHECK-NEXT:    ret <1 x i64> [[TMP2]]
+;
+entry:
+  %0 = bitcast <1 x i64> %a.coerce to x86_mmx
+  %1 = tail call x86_mmx @llvm.x86.mmx.pinsr.w(x86_mmx %0, i32 %d, i32 2)
+  %2 = bitcast x86_mmx %1 to <1 x i64>
+  ret <1 x i64> %2
+}
+
+declare x86_mmx @llvm.x86.mmx.pinsr.w(x86_mmx, i32, i32 immarg)
+
+define i32 @test_mm_extract_pi16(<1 x i64> %a.coerce) nounwind #0 {
+; CHECK-LABEL: define i32 @test_mm_extract_pi16(
+; CHECK-SAME: <1 x i64> [[A_COERCE:%.*]]) #[[ATTR4]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP2:%.*]] = load <1 x i64>, ptr @__msan_param_tls, align 8
+; CHECK-NEXT:    call void @llvm.donothing()
+; CHECK-NEXT:    [[TMP5:%.*]] = bitcast <1 x i64> [[TMP2]] to i64
+; CHECK-NEXT:    [[TMP0:%.*]] = bitcast <1 x i64> [[A_COERCE]] to x86_mmx
+; CHECK-NEXT:    [[_MSCMP:%.*]] = icmp ne i64 [[TMP5]], 0
+; CHECK-NEXT:    br i1 [[_MSCMP]], label [[TMP3:%.*]], label [[TMP4:%.*]], !prof [[PROF0]]
+; CHECK:       3:
+; CHECK-NEXT:    call void @__msan_warning_noreturn() #[[ATTR6]]
+; CHECK-NEXT:    unreachable
+; CHECK:       4:
+; CHECK-NEXT:    [[TMP1:%.*]] = tail call i32 @llvm.x86.mmx.pextr.w(x86_mmx [[TMP0]], i32 2)
+; CHECK-NEXT:    store i32 0, ptr @__msan_retval_tls, align 8
+; CHECK-NEXT:    ret i32 [[TMP1]]
+;
+entry:
+  %0 = bitcast <1 x i64> %a.coerce to x86_mmx
+  %1 = tail call i32 @llvm.x86.mmx.pextr.w(x86_mmx %0, i32 2)
+  ret i32 %1
+}
+
+declare i32 @llvm.x86.mmx.pextr.w(x86_mmx, i32 immarg)
+
+attributes #0 = { sanitize_memory }
+;.
+; CHECK: [[PROF0]] = !{!"branch_weights", i32 1, i32 1048575}
+;.
diff --git a/llvm/test/Instrumentation/MemorySanitizer/vscale.ll b/llvm/test/Instrumentation/MemorySanitizer/vscale.ll
new file mode 100644
index 000000000000..e1a4a9b7aa68
--- /dev/null
+++ b/llvm/test/Instrumentation/MemorySanitizer/vscale.ll
@@ -0,0 +1,188 @@
+; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --version 4
+; RUN: opt < %s -S -msan-check-access-address=0 -passes="msan" 2>&1 | FileCheck %s
+
+target datalayout = "e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f32:32:32-f64:64:64-v64:64:64-v128:128:128-a0:0:64-s0:64:64-f80:128:128-n8:16:32:64-S128"
+target triple = "x86_64-unknown-linux-gnu"
+
+define void @test_load_store_i32(ptr %a, ptr %b) sanitize_memory {
+; CHECK-LABEL: define void @test_load_store_i32(
+; CHECK-SAME: ptr [[A:%.*]], ptr [[B:%.*]]) #[[ATTR0:[0-9]+]] {
+; CHECK-NEXT:    call void @llvm.donothing()
+; CHECK-NEXT:    [[TMP1:%.*]] = load <vscale x 4 x i32>, ptr [[A]], align 16
+; CHECK-NEXT:    [[TMP2:%.*]] = ptrtoint ptr [[A]] to i64
+; CHECK-NEXT:    [[TMP3:%.*]] = xor i64 [[TMP2]], 87960930222080
+; CHECK-NEXT:    [[TMP4:%.*]] = inttoptr i64 [[TMP3]] to ptr
+; CHECK-NEXT:    [[_MSLD:%.*]] = load <vscale x 4 x i32>, ptr [[TMP4]], align 16
+; CHECK-NEXT:    [[TMP5:%.*]] = ptrtoint ptr [[B]] to i64
+; CHECK-NEXT:    [[TMP6:%.*]] = xor i64 [[TMP5]], 87960930222080
+; CHECK-NEXT:    [[TMP7:%.*]] = inttoptr i64 [[TMP6]] to ptr
+; CHECK-NEXT:    store <vscale x 4 x i32> [[_MSLD]], ptr [[TMP7]], align 16
+; CHECK-NEXT:    store <vscale x 4 x i32> [[TMP1]], ptr [[B]], align 16
+; CHECK-NEXT:    ret void
+;
+  %1 = load <vscale x 4 x i32>, ptr %a
+  store <vscale x 4 x i32> %1, ptr %b
+  ret void
+}
+
+define void @test_load_store_add_int(ptr %a, ptr %b) sanitize_memory {
+; CHECK-LABEL: define void @test_load_store_add_int(
+; CHECK-SAME: ptr [[A:%.*]], ptr [[B:%.*]]) #[[ATTR0]] {
+; CHECK-NEXT:    call void @llvm.donothing()
+; CHECK-NEXT:    [[TMP1:%.*]] = load <vscale x 8 x i64>, ptr [[A]], align 64
+; CHECK-NEXT:    [[TMP2:%.*]] = ptrtoint ptr [[A]] to i64
+; CHECK-NEXT:    [[TMP3:%.*]] = xor i64 [[TMP2]], 87960930222080
+; CHECK-NEXT:    [[TMP4:%.*]] = inttoptr i64 [[TMP3]] to ptr
+; CHECK-NEXT:    [[_MSLD:%.*]] = load <vscale x 8 x i64>, ptr [[TMP4]], align 64
+; CHECK-NEXT:    [[TMP5:%.*]] = load <vscale x 8 x i64>, ptr [[B]], align 64
+; CHECK-NEXT:    [[TMP6:%.*]] = ptrtoint ptr [[B]] to i64
+; CHECK-NEXT:    [[TMP7:%.*]] = xor i64 [[TMP6]], 87960930222080
+; CHECK-NEXT:    [[TMP8:%.*]] = inttoptr i64 [[TMP7]] to ptr
+; CHECK-NEXT:    [[_MSLD1:%.*]] = load <vscale x 8 x i64>, ptr [[TMP8]], align 64
+; CHECK-NEXT:    [[_MSPROP:%.*]] = or <vscale x 8 x i64> [[_MSLD]], [[_MSLD1]]
+; CHECK-NEXT:    [[TMP9:%.*]] = add <vscale x 8 x i64> [[TMP1]], [[TMP5]]
+; CHECK-NEXT:    [[TMP10:%.*]] = ptrtoint ptr [[B]] to i64
+; CHECK-NEXT:    [[TMP11:%.*]] = xor i64 [[TMP10]], 87960930222080
+; CHECK-NEXT:    [[TMP12:%.*]] = inttoptr i64 [[TMP11]] to ptr
+; CHECK-NEXT:    store <vscale x 8 x i64> [[_MSLD1]], ptr [[TMP12]], align 64
+; CHECK-NEXT:    store <vscale x 8 x i64> [[TMP5]], ptr [[B]], align 64
+; CHECK-NEXT:    ret void
+;
+  %1 = load <vscale x 8 x i64>, ptr %a
+  %2 = load <vscale x 8 x i64>, ptr %b
+  %3 = add <vscale x 8 x i64> %1, %2
+  store <vscale x 8 x i64> %2, ptr %b
+  ret void
+}
+
+define void @test_load_store_float(ptr %a, ptr %b) sanitize_memory {
+; CHECK-LABEL: define void @test_load_store_float(
+; CHECK-SAME: ptr [[A:%.*]], ptr [[B:%.*]]) #[[ATTR0]] {
+; CHECK-NEXT:    call void @llvm.donothing()
+; CHECK-NEXT:    [[TMP1:%.*]] = load <vscale x 4 x float>, ptr [[A]], align 16
+; CHECK-NEXT:    [[TMP2:%.*]] = ptrtoint ptr [[A]] to i64
+; CHECK-NEXT:    [[TMP3:%.*]] = xor i64 [[TMP2]], 87960930222080
+; CHECK-NEXT:    [[TMP4:%.*]] = inttoptr i64 [[TMP3]] to ptr
+; CHECK-NEXT:    [[_MSLD:%.*]] = load <vscale x 4 x i32>, ptr [[TMP4]], align 16
+; CHECK-NEXT:    [[TMP5:%.*]] = ptrtoint ptr [[B]] to i64
+; CHECK-NEXT:    [[TMP6:%.*]] = xor i64 [[TMP5]], 87960930222080
+; CHECK-NEXT:    [[TMP7:%.*]] = inttoptr i64 [[TMP6]] to ptr
+; CHECK-NEXT:    store <vscale x 4 x i32> [[_MSLD]], ptr [[TMP7]], align 16
+; CHECK-NEXT:    store <vscale x 4 x float> [[TMP1]], ptr [[B]], align 16
+; CHECK-NEXT:    ret void
+;
+  %1 = load <vscale x 4 x float>, ptr %a
+  store <vscale x 4 x float> %1, ptr %b
+  ret void
+}
+
+define void @test_load_store_add_float(ptr %a, ptr %b) sanitize_memory {
+; CHECK-LABEL: define void @test_load_store_add_float(
+; CHECK-SAME: ptr [[A:%.*]], ptr [[B:%.*]]) #[[ATTR0]] {
+; CHECK-NEXT:    call void @llvm.donothing()
+; CHECK-NEXT:    [[TMP1:%.*]] = load <vscale x 2 x float>, ptr [[A]], align 8
+; CHECK-NEXT:    [[TMP2:%.*]] = ptrtoint ptr [[A]] to i64
+; CHECK-NEXT:    [[TMP3:%.*]] = xor i64 [[TMP2]], 87960930222080
+; CHECK-NEXT:    [[TMP4:%.*]] = inttoptr i64 [[TMP3]] to ptr
+; CHECK-NEXT:    [[_MSLD:%.*]] = load <vscale x 2 x i32>, ptr [[TMP4]], align 8
+; CHECK-NEXT:    [[TMP5:%.*]] = load <vscale x 2 x float>, ptr [[B]], align 8
+; CHECK-NEXT:    [[TMP6:%.*]] = ptrtoint ptr [[B]] to i64
+; CHECK-NEXT:    [[TMP7:%.*]] = xor i64 [[TMP6]], 87960930222080
+; CHECK-NEXT:    [[TMP8:%.*]] = inttoptr i64 [[TMP7]] to ptr
+; CHECK-NEXT:    [[_MSLD1:%.*]] = load <vscale x 2 x i32>, ptr [[TMP8]], align 8
+; CHECK-NEXT:    [[_MSPROP:%.*]] = or <vscale x 2 x i32> [[_MSLD]], [[_MSLD1]]
+; CHECK-NEXT:    [[TMP9:%.*]] = fadd <vscale x 2 x float> [[TMP1]], [[TMP5]]
+; CHECK-NEXT:    [[TMP10:%.*]] = ptrtoint ptr [[B]] to i64
+; CHECK-NEXT:    [[TMP11:%.*]] = xor i64 [[TMP10]], 87960930222080
+; CHECK-NEXT:    [[TMP12:%.*]] = inttoptr i64 [[TMP11]] to ptr
+; CHECK-NEXT:    store <vscale x 2 x i32> [[_MSLD1]], ptr [[TMP12]], align 8
+; CHECK-NEXT:    store <vscale x 2 x float> [[TMP5]], ptr [[B]], align 8
+; CHECK-NEXT:    ret void
+;
+  %1 = load <vscale x 2 x float>, ptr %a
+  %2 = load <vscale x 2 x float>, ptr %b
+  %3 = fadd <vscale x 2 x float> %1, %2
+  store <vscale x 2 x float> %2, ptr %b
+  ret void
+}
+
+define <vscale x 2 x float> @fn_ret(ptr %a) sanitize_memory {
+; CHECK-LABEL: define <vscale x 2 x float> @fn_ret(
+; CHECK-SAME: ptr [[A:%.*]]) #[[ATTR0]] {
+; CHECK-NEXT:    call void @llvm.donothing()
+; CHECK-NEXT:    [[TMP1:%.*]] = load <vscale x 2 x float>, ptr [[A]], align 8
+; CHECK-NEXT:    [[TMP2:%.*]] = ptrtoint ptr [[A]] to i64
+; CHECK-NEXT:    [[TMP3:%.*]] = xor i64 [[TMP2]], 87960930222080
+; CHECK-NEXT:    [[TMP4:%.*]] = inttoptr i64 [[TMP3]] to ptr
+; CHECK-NEXT:    [[_MSLD:%.*]] = load <vscale x 2 x i32>, ptr [[TMP4]], align 8
+; CHECK-NEXT:    store <vscale x 2 x i32> [[_MSLD]], ptr @__msan_retval_tls, align 8
+; CHECK-NEXT:    ret <vscale x 2 x float> [[TMP1]]
+;
+  %1 = load <vscale x 2 x float>, ptr %a
+  ret <vscale x 2 x float> %1
+}
+
+define void @test_ret(ptr %a, ptr %b) sanitize_memory {
+; CHECK-LABEL: define void @test_ret(
+; CHECK-SAME: ptr [[A:%.*]], ptr [[B:%.*]]) #[[ATTR0]] {
+; CHECK-NEXT:    [[TMP1:%.*]] = load i64, ptr @__msan_param_tls, align 8
+; CHECK-NEXT:    call void @llvm.donothing()
+; CHECK-NEXT:    store i64 [[TMP1]], ptr @__msan_param_tls, align 8
+; CHECK-NEXT:    store <vscale x 2 x i32> zeroinitializer, ptr @__msan_retval_tls, align 8
+; CHECK-NEXT:    [[TMP5:%.*]] = call <vscale x 2 x float> @fn_ret(ptr [[A]])
+; CHECK-NEXT:    [[_MSRET:%.*]] = load <vscale x 2 x i32>, ptr @__msan_retval_tls, align 8
+; CHECK-NEXT:    [[TMP2:%.*]] = ptrtoint ptr [[B]] to i64
+; CHECK-NEXT:    [[TMP3:%.*]] = xor i64 [[TMP2]], 87960930222080
+; CHECK-NEXT:    [[TMP4:%.*]] = inttoptr i64 [[TMP3]] to ptr
+; CHECK-NEXT:    store <vscale x 2 x i32> [[_MSRET]], ptr [[TMP4]], align 8
+; CHECK-NEXT:    store <vscale x 2 x float> [[TMP5]], ptr [[B]], align 8
+; CHECK-NEXT:    ret void
+;
+  %1 = call <vscale x 2 x float> @fn_ret(ptr %a)
+  store <vscale x 2 x float> %1, ptr %b
+  ret void
+}
+
+define void @fn_param(<vscale x 2 x float> %a, ptr %b) sanitize_memory {
+; CHECK-LABEL: define void @fn_param(
+; CHECK-SAME: <vscale x 2 x float> [[A:%.*]], ptr [[B:%.*]]) #[[ATTR0]] {
+; CHECK-NEXT:    call void @llvm.donothing()
+; CHECK-NEXT:    [[TMP1:%.*]] = ptrtoint ptr [[B]] to i64
+; CHECK-NEXT:    [[TMP2:%.*]] = xor i64 [[TMP1]], 87960930222080
+; CHECK-NEXT:    [[TMP3:%.*]] = inttoptr i64 [[TMP2]] to ptr
+; CHECK-NEXT:    store <vscale x 2 x i32> zeroinitializer, ptr [[TMP3]], align 8
+; CHECK-NEXT:    store <vscale x 2 x float> [[A]], ptr [[B]], align 8
+; CHECK-NEXT:    ret void
+;
+  store <vscale x 2 x float> %a, ptr %b
+  ret void
+}
+
+define void @test_param(ptr %a, ptr %b) sanitize_memory {
+; CHECK-LABEL: define void @test_param(
+; CHECK-SAME: ptr [[A:%.*]], ptr [[B:%.*]]) #[[ATTR0]] {
+; CHECK-NEXT:    [[TMP1:%.*]] = load i64, ptr inttoptr (i64 add (i64 ptrtoint (ptr @__msan_param_tls to i64), i64 8) to ptr), align 8
+; CHECK-NEXT:    call void @llvm.donothing()
+; CHECK-NEXT:    [[TMP2:%.*]] = load <vscale x 2 x float>, ptr [[A]], align 8
+; CHECK-NEXT:    [[TMP3:%.*]] = ptrtoint ptr [[A]] to i64
+; CHECK-NEXT:    [[TMP4:%.*]] = xor i64 [[TMP3]], 87960930222080
+; CHECK-NEXT:    [[TMP5:%.*]] = inttoptr i64 [[TMP4]] to ptr
+; CHECK-NEXT:    [[_MSLD:%.*]] = load <vscale x 2 x i32>, ptr [[TMP5]], align 8
+; CHECK-NEXT:    store i64 [[TMP1]], ptr @__msan_param_tls, align 8
+; CHECK-NEXT:    [[TMP6:%.*]] = call i32 @llvm.vector.reduce.or.nxv2i32(<vscale x 2 x i32> [[_MSLD]])
+; CHECK-NEXT:    [[_MSCMP:%.*]] = icmp ne i32 [[TMP6]], 0
+; CHECK-NEXT:    br i1 [[_MSCMP]], label [[TMP7:%.*]], label [[TMP8:%.*]], !prof [[PROF0:![0-9]+]]
+; CHECK:       7:
+; CHECK-NEXT:    call void @__msan_warning_noreturn() #[[ATTR4:[0-9]+]]
+; CHECK-NEXT:    unreachable
+; CHECK:       8:
+; CHECK-NEXT:    call void @fn_param(<vscale x 2 x float> [[TMP2]], ptr [[B]])
+; CHECK-NEXT:    ret void
+;
+  %1 = load <vscale x 2 x float>, ptr %a
+  call void @fn_param(<vscale x 2 x float> %1, ptr %b)
+  ret void
+}
+;.
+; CHECK: [[PROF0]] = !{!"branch_weights", i32 1, i32 1048575}
+;.
diff --git a/llvm/test/LTO/X86/codemodel-2.ll b/llvm/test/LTO/X86/codemodel-2.ll
index 5cd9731606f2..fc1074bcf223 100644
--- a/llvm/test/LTO/X86/codemodel-2.ll
+++ b/llvm/test/LTO/X86/codemodel-2.ll
@@ -1,5 +1,5 @@
 ; RUN: llvm-as %s -o %t.o
-; RUN: llvm-lto2 run -r %t.o,_start,px %t.o -o %t.s
+; RUN: llvm-lto2 run -r %t.o,_start,px -r %t.o,_GLOBAL_OFFSET_TABLE_, %t.o -o %t.s
 ; RUN: llvm-objdump --no-print-imm-hex -d %t.s.0 | FileCheck %s --check-prefix=CHECK-LARGE
 
 target triple = "x86_64-unknown-linux-gnu"
diff --git a/llvm/test/LTO/X86/codemodel-3.ll b/llvm/test/LTO/X86/codemodel-3.ll
index 947221e9f36d..13702dfbca2d 100644
--- a/llvm/test/LTO/X86/codemodel-3.ll
+++ b/llvm/test/LTO/X86/codemodel-3.ll
@@ -1,6 +1,7 @@
 ; RUN: llvm-as %s -o %t0.o
 ; RUN: llvm-as < %p/Inputs/codemodel-3.ll > %t1.o
-; RUN: not llvm-lto2 run -r %t0.o,_start,px -r %t1.o,bar,px %t0.o %t1.o -o %t2.s 2>&1 | FileCheck %s 
+; RUN: not llvm-lto2 run -r %t0.o,_start,px -r %t1.o,bar,px -r %t0.o,_GLOBAL_OFFSET_TABLE_, \
+; RUN:   -r %t1.o,_GLOBAL_OFFSET_TABLE_, %t0.o %t1.o -o %t2.s 2>&1 | FileCheck %s 
 
 target triple = "x86_64-unknown-linux-gnu"
 target datalayout = "e-m:e-p270:32:32-p271:32:32-p272:64:64-i64:64-f80:128-n8:16:32:64-S128"
diff --git a/llvm/test/LTO/X86/largedatathreshold-1.ll b/llvm/test/LTO/X86/largedatathreshold-1.ll
index e3be5c11baaa..dfd8319511b6 100644
--- a/llvm/test/LTO/X86/largedatathreshold-1.ll
+++ b/llvm/test/LTO/X86/largedatathreshold-1.ll
@@ -1,5 +1,5 @@
 ; RUN: llvm-as %s -o %t.o
-; RUN: llvm-lto2 run -r %t.o,_start,px %t.o -o %t.s
+; RUN: llvm-lto2 run -r %t.o,_start,px -r %t.o,_GLOBAL_OFFSET_TABLE_, %t.o -o %t.s
 ; RUN: llvm-objdump -d %t.s.0 | FileCheck %s
 
 target triple = "x86_64-unknown-linux-gnu"
diff --git a/llvm/test/LTO/X86/largedatathreshold-2.ll b/llvm/test/LTO/X86/largedatathreshold-2.ll
index 103c066b744d..59438bbdb502 100644
--- a/llvm/test/LTO/X86/largedatathreshold-2.ll
+++ b/llvm/test/LTO/X86/largedatathreshold-2.ll
@@ -1,5 +1,5 @@
 ; RUN: llvm-as %s -o %t.o
-; RUN: llvm-lto2 run -r %t.o,_start,px %t.o -o %t.s
+; RUN: llvm-lto2 run -r %t.o,_start,px -r %t.o,_GLOBAL_OFFSET_TABLE_, %t.o -o %t.s
 ; RUN: llvm-objdump -d %t.s.0 | FileCheck %s
 
 target triple = "x86_64-unknown-linux-gnu"
diff --git a/llvm/test/LTO/X86/largedatathreshold-3.ll b/llvm/test/LTO/X86/largedatathreshold-3.ll
index 3c0653db334d..fea7987ff155 100644
--- a/llvm/test/LTO/X86/largedatathreshold-3.ll
+++ b/llvm/test/LTO/X86/largedatathreshold-3.ll
@@ -1,6 +1,7 @@
 ; RUN: llvm-as %s -o %t0.o
 ; RUN: llvm-as < %p/Inputs/largedatathreshold.ll > %t1.o
-; RUN: not llvm-lto2 run -r %t0.o,_start,px -r %t1.o,bar,px %t0.o %t1.o -o %t2.s 2>&1 | FileCheck %s
+; RUN: not llvm-lto2 run -r %t0.o,_start,px -r %t1.o,bar,px -r %t0.o,_GLOBAL_OFFSET_TABLE_, \
+; RUN:   -r %t1.o,_GLOBAL_OFFSET_TABLE_, %t0.o %t1.o -o %t2.s 2>&1 | FileCheck %s
 
 ; CHECK: 'Large Data Threshold': IDs have conflicting values
 
diff --git a/llvm/test/MC/AArch64/SME2p1/fadd-diagnostics.s b/llvm/test/MC/AArch64/SME2p1/fadd-diagnostics.s
index c13a1be05b1c..a18989880a34 100644
--- a/llvm/test/MC/AArch64/SME2p1/fadd-diagnostics.s
+++ b/llvm/test/MC/AArch64/SME2p1/fadd-diagnostics.s
@@ -1,4 +1,4 @@
-// RUN: not llvm-mc -triple=aarch64 -show-encoding -mattr=+sme2p1,+sme-f16f16 2>&1 < %s | FileCheck %s
+// RUN: not llvm-mc -triple=aarch64 -show-encoding -mattr=+sme-f16f16 2>&1 < %s | FileCheck %s
 
 // --------------------------------------------------------------------------//
 // Out of range index offset
diff --git a/llvm/test/MC/AArch64/SME2p1/fadd.s b/llvm/test/MC/AArch64/SME2p1/fadd.s
index a8e64a63dbdb..bdb769093c83 100644
--- a/llvm/test/MC/AArch64/SME2p1/fadd.s
+++ b/llvm/test/MC/AArch64/SME2p1/fadd.s
@@ -1,300 +1,302 @@
-// RUN: llvm-mc -triple=aarch64 -show-encoding -mattr=+sme2p1,+sme-f16f16 < %s \
+// RUN: llvm-mc -triple=aarch64 -show-encoding -mattr=+sme-f16f16 < %s \
+// RUN:        | FileCheck %s --check-prefixes=CHECK-ENCODING,CHECK-INST
+// RUN: llvm-mc -triple=aarch64 -show-encoding -mattr=+sme-f8f16 < %s \
 // RUN:        | FileCheck %s --check-prefixes=CHECK-ENCODING,CHECK-INST
 // RUN: not llvm-mc -triple=aarch64 -show-encoding < %s 2>&1 \
 // RUN:        | FileCheck %s --check-prefix=CHECK-ERROR
-// RUN: llvm-mc -triple=aarch64 -filetype=obj -mattr=+sme2p1,+sme-f16f16 < %s \
-// RUN:        | llvm-objdump -d --mattr=+sme2p1,+sme-f16f16 - | FileCheck %s --check-prefix=CHECK-INST
-// RUN: llvm-mc -triple=aarch64 -filetype=obj -mattr=+sme2p1,+sme-f16f16 < %s \
+// RUN: llvm-mc -triple=aarch64 -filetype=obj -mattr=+sme-f16f16 < %s \
+// RUN:        | llvm-objdump -d --mattr=+sme-f16f16 - | FileCheck %s --check-prefix=CHECK-INST
+// RUN: llvm-mc -triple=aarch64 -filetype=obj -mattr=+sme-f16f16 < %s \
 // RUN:        | llvm-objdump -d --mattr=-sme2p1 - | FileCheck %s --check-prefix=CHECK-UNKNOWN
-// RUN: llvm-mc -triple=aarch64 -show-encoding -mattr=+sme2p1,+sme-f16f16 < %s \
+// RUN: llvm-mc -triple=aarch64 -show-encoding -mattr=+sme-f16f16 < %s \
 // RUN:        | sed '/.text/d' | sed 's/.*encoding: //g' \
-// RUN:        | llvm-mc -triple=aarch64 -mattr=+sme2p1,+sme-f16f16 -disassemble -show-encoding \
+// RUN:        | llvm-mc -triple=aarch64 -mattr=+sme-f16f16 -disassemble -show-encoding \
 // RUN:        | FileCheck %s --check-prefixes=CHECK-ENCODING,CHECK-INST
 
 fadd    za.h[w8, 0, vgx2], {z0.h, z1.h}  // 11000001-10100100-00011100-00000000
 // CHECK-INST: fadd    za.h[w8, 0, vgx2], { z0.h, z1.h }
 // CHECK-ENCODING: [0x00,0x1c,0xa4,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1
+// CHECK-ERROR: instruction requires: sme-f16f16 or sme-f8f16
 // CHECK-UNKNOWN: c1a41c00 <unknown>
 
 fadd    za.h[w8, 0], {z0.h - z1.h}  // 11000001-10100100-00011100-00000000
 // CHECK-INST: fadd    za.h[w8, 0, vgx2], { z0.h, z1.h }
 // CHECK-ENCODING: [0x00,0x1c,0xa4,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1
+// CHECK-ERROR: instruction requires: sme-f16f16 or sme-f8f16
 // CHECK-UNKNOWN: c1a41c00 <unknown>
 
 fadd    za.h[w10, 5, vgx2], {z10.h, z11.h}  // 11000001-10100100-01011101-01000101
 // CHECK-INST: fadd    za.h[w10, 5, vgx2], { z10.h, z11.h }
 // CHECK-ENCODING: [0x45,0x5d,0xa4,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1
+// CHECK-ERROR: instruction requires: sme-f16f16 or sme-f8f16
 // CHECK-UNKNOWN: c1a45d45 <unknown>
 
 fadd    za.h[w10, 5], {z10.h - z11.h}  // 11000001-10100100-01011101-01000101
 // CHECK-INST: fadd    za.h[w10, 5, vgx2], { z10.h, z11.h }
 // CHECK-ENCODING: [0x45,0x5d,0xa4,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1
+// CHECK-ERROR: instruction requires: sme-f16f16 or sme-f8f16
 // CHECK-UNKNOWN: c1a45d45 <unknown>
 
 fadd    za.h[w11, 7, vgx2], {z12.h, z13.h}  // 11000001-10100100-01111101-10000111
 // CHECK-INST: fadd    za.h[w11, 7, vgx2], { z12.h, z13.h }
 // CHECK-ENCODING: [0x87,0x7d,0xa4,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1
+// CHECK-ERROR: instruction requires: sme-f16f16 or sme-f8f16
 // CHECK-UNKNOWN: c1a47d87 <unknown>
 
 fadd    za.h[w11, 7], {z12.h - z13.h}  // 11000001-10100100-01111101-10000111
 // CHECK-INST: fadd    za.h[w11, 7, vgx2], { z12.h, z13.h }
 // CHECK-ENCODING: [0x87,0x7d,0xa4,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1
+// CHECK-ERROR: instruction requires: sme-f16f16 or sme-f8f16
 // CHECK-UNKNOWN: c1a47d87 <unknown>
 
 fadd    za.h[w11, 7, vgx2], {z30.h, z31.h}  // 11000001-10100100-01111111-11000111
 // CHECK-INST: fadd    za.h[w11, 7, vgx2], { z30.h, z31.h }
 // CHECK-ENCODING: [0xc7,0x7f,0xa4,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1
+// CHECK-ERROR: instruction requires: sme-f16f16 or sme-f8f16
 // CHECK-UNKNOWN: c1a47fc7 <unknown>
 
 fadd    za.h[w11, 7], {z30.h - z31.h}  // 11000001-10100100-01111111-11000111
 // CHECK-INST: fadd    za.h[w11, 7, vgx2], { z30.h, z31.h }
 // CHECK-ENCODING: [0xc7,0x7f,0xa4,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1
+// CHECK-ERROR: instruction requires: sme-f16f16 or sme-f8f16
 // CHECK-UNKNOWN: c1a47fc7 <unknown>
 
 fadd    za.h[w8, 5, vgx2], {z16.h, z17.h}  // 11000001-10100100-00011110-00000101
 // CHECK-INST: fadd    za.h[w8, 5, vgx2], { z16.h, z17.h }
 // CHECK-ENCODING: [0x05,0x1e,0xa4,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1
+// CHECK-ERROR: instruction requires: sme-f16f16 or sme-f8f16
 // CHECK-UNKNOWN: c1a41e05 <unknown>
 
 fadd    za.h[w8, 5], {z16.h - z17.h}  // 11000001-10100100-00011110-00000101
 // CHECK-INST: fadd    za.h[w8, 5, vgx2], { z16.h, z17.h }
 // CHECK-ENCODING: [0x05,0x1e,0xa4,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1
+// CHECK-ERROR: instruction requires: sme-f16f16 or sme-f8f16
 // CHECK-UNKNOWN: c1a41e05 <unknown>
 
 fadd    za.h[w8, 1, vgx2], {z0.h, z1.h}  // 11000001-10100100-00011100-00000001
 // CHECK-INST: fadd    za.h[w8, 1, vgx2], { z0.h, z1.h }
 // CHECK-ENCODING: [0x01,0x1c,0xa4,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1
+// CHECK-ERROR: instruction requires: sme-f16f16 or sme-f8f16
 // CHECK-UNKNOWN: c1a41c01 <unknown>
 
 fadd    za.h[w8, 1], {z0.h - z1.h}  // 11000001-10100100-00011100-00000001
 // CHECK-INST: fadd    za.h[w8, 1, vgx2], { z0.h, z1.h }
 // CHECK-ENCODING: [0x01,0x1c,0xa4,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1
+// CHECK-ERROR: instruction requires: sme-f16f16 or sme-f8f16
 // CHECK-UNKNOWN: c1a41c01 <unknown>
 
 fadd    za.h[w10, 0, vgx2], {z18.h, z19.h}  // 11000001-10100100-01011110, 01000000
 // CHECK-INST: fadd    za.h[w10, 0, vgx2], { z18.h, z19.h }
 // CHECK-ENCODING: [0x40,0x5e,0xa4,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1
+// CHECK-ERROR: instruction requires: sme-f16f16 or sme-f8f16
 // CHECK-UNKNOWN: c1a45e40 <unknown>
 
 fadd    za.h[w10, 0], {z18.h - z19.h}  // 11000001-10100100-01011110-01000000
 // CHECK-INST: fadd    za.h[w10, 0, vgx2], { z18.h, z19.h }
 // CHECK-ENCODING: [0x40,0x5e,0xa4,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1
+// CHECK-ERROR: instruction requires: sme-f16f16 or sme-f8f16
 // CHECK-UNKNOWN: c1a45e40 <unknown>
 
 fadd    za.h[w8, 0, vgx2], {z12.h, z13.h}  // 11000001-10100100-00011101-10000000
 // CHECK-INST: fadd    za.h[w8, 0, vgx2], { z12.h, z13.h }
 // CHECK-ENCODING: [0x80,0x1d,0xa4,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1
+// CHECK-ERROR: instruction requires: sme-f16f16 or sme-f8f16
 // CHECK-UNKNOWN: c1a41d80 <unknown>
 
 fadd    za.h[w8, 0], {z12.h - z13.h}  // 11000001-10100100-00011101-10000000
 // CHECK-INST: fadd    za.h[w8, 0, vgx2], { z12.h, z13.h }
 // CHECK-ENCODING: [0x80,0x1d,0xa4,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1
+// CHECK-ERROR: instruction requires: sme-f16f16 or sme-f8f16
 // CHECK-UNKNOWN: c1a41d80 <unknown>
 
 fadd    za.h[w10, 1, vgx2], {z0.h, z1.h}  // 11000001-10100100-01011100-00000001
 // CHECK-INST: fadd    za.h[w10, 1, vgx2], { z0.h, z1.h }
 // CHECK-ENCODING: [0x01,0x5c,0xa4,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1
+// CHECK-ERROR: instruction requires: sme-f16f16 or sme-f8f16
 // CHECK-UNKNOWN: c1a45c01 <unknown>
 
 fadd    za.h[w10, 1], {z0.h - z1.h}  // 11000001-10100100-01011100-00000001
 // CHECK-INST: fadd    za.h[w10, 1, vgx2], { z0.h, z1.h }
 // CHECK-ENCODING: [0x01,0x5c,0xa4,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1
+// CHECK-ERROR: instruction requires: sme-f16f16 or sme-f8f16
 // CHECK-UNKNOWN: c1a45c01 <unknown>
 
 fadd    za.h[w8, 5, vgx2], {z22.h, z23.h}  // 11000001-10100100-00011110, 11000101
 // CHECK-INST: fadd    za.h[w8, 5, vgx2], { z22.h, z23.h }
 // CHECK-ENCODING: [0xc5,0x1e,0xa4,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1
+// CHECK-ERROR: instruction requires: sme-f16f16 or sme-f8f16
 // CHECK-UNKNOWN: c1a41ec5 <unknown>
 
 fadd    za.h[w8, 5], {z22.h - z23.h}  // 11000001-10100100-00011110-11000101
 // CHECK-INST: fadd    za.h[w8, 5, vgx2], { z22.h, z23.h }
 // CHECK-ENCODING: [0xc5,0x1e,0xa4,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1
+// CHECK-ERROR: instruction requires: sme-f16f16 or sme-f8f16
 // CHECK-UNKNOWN: c1a41ec5 <unknown>
 
 fadd    za.h[w11, 2, vgx2], {z8.h, z9.h}  // 11000001-10100100-01111101-00000010
 // CHECK-INST: fadd    za.h[w11, 2, vgx2], { z8.h, z9.h }
 // CHECK-ENCODING: [0x02,0x7d,0xa4,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1
+// CHECK-ERROR: instruction requires: sme-f16f16 or sme-f8f16
 // CHECK-UNKNOWN: c1a47d02 <unknown>
 
 fadd    za.h[w11, 2], {z8.h - z9.h}  // 11000001-10100100-01111101-00000010
 // CHECK-INST: fadd    za.h[w11, 2, vgx2], { z8.h, z9.h }
 // CHECK-ENCODING: [0x02,0x7d,0xa4,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1
+// CHECK-ERROR: instruction requires: sme-f16f16 or sme-f8f16
 // CHECK-UNKNOWN: c1a47d02 <unknown>
 
 fadd    za.h[w9, 7, vgx2], {z12.h, z13.h}  // 11000001-10100100-00111101-10000111
 // CHECK-INST: fadd    za.h[w9, 7, vgx2], { z12.h, z13.h }
 // CHECK-ENCODING: [0x87,0x3d,0xa4,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1
+// CHECK-ERROR: instruction requires: sme-f16f16 or sme-f8f16
 // CHECK-UNKNOWN: c1a43d87 <unknown>
 
 fadd    za.h[w9, 7], {z12.h - z13.h}  // 11000001-10100100-00111101-10000111
 // CHECK-INST: fadd    za.h[w9, 7, vgx2], { z12.h, z13.h }
 // CHECK-ENCODING: [0x87,0x3d,0xa4,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1
+// CHECK-ERROR: instruction requires: sme-f16f16 or sme-f8f16
 // CHECK-UNKNOWN: c1a43d87 <unknown>
 
 fadd    za.h[w8, 0, vgx4], {z0.h - z3.h}  // 11000001-10100101-00011100-00000000
 // CHECK-INST: fadd    za.h[w8, 0, vgx4], { z0.h - z3.h }
 // CHECK-ENCODING: [0x00,0x1c,0xa5,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1
+// CHECK-ERROR: instruction requires: sme-f16f16 or sme-f8f16
 // CHECK-UNKNOWN: c1a51c00 <unknown>
 
 fadd    za.h[w8, 0], {z0.h - z3.h}  // 11000001-10100101-00011100-00000000
 // CHECK-INST: fadd    za.h[w8, 0, vgx4], { z0.h - z3.h }
 // CHECK-ENCODING: [0x00,0x1c,0xa5,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1
+// CHECK-ERROR: instruction requires: sme-f16f16 or sme-f8f16
 // CHECK-UNKNOWN: c1a51c00 <unknown>
 
 fadd    za.h[w10, 5, vgx4], {z8.h - z11.h}  // 11000001-10100101-01011101-00000101
 // CHECK-INST: fadd    za.h[w10, 5, vgx4], { z8.h - z11.h }
 // CHECK-ENCODING: [0x05,0x5d,0xa5,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1
+// CHECK-ERROR: instruction requires: sme-f16f16 or sme-f8f16
 // CHECK-UNKNOWN: c1a55d05 <unknown>
 
 fadd    za.h[w10, 5], {z8.h - z11.h}  // 11000001-10100101-01011101-00000101
 // CHECK-INST: fadd    za.h[w10, 5, vgx4], { z8.h - z11.h }
 // CHECK-ENCODING: [0x05,0x5d,0xa5,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1
+// CHECK-ERROR: instruction requires: sme-f16f16 or sme-f8f16
 // CHECK-UNKNOWN: c1a55d05 <unknown>
 
 fadd    za.h[w11, 7, vgx4], {z12.h - z15.h}  // 11000001-10100101-01111101-10000111
 // CHECK-INST: fadd    za.h[w11, 7, vgx4], { z12.h - z15.h }
 // CHECK-ENCODING: [0x87,0x7d,0xa5,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1
+// CHECK-ERROR: instruction requires: sme-f16f16 or sme-f8f16
 // CHECK-UNKNOWN: c1a57d87 <unknown>
 
 fadd    za.h[w11, 7], {z12.h - z15.h}  // 11000001-10100101-01111101-10000111
 // CHECK-INST: fadd    za.h[w11, 7, vgx4], { z12.h - z15.h }
 // CHECK-ENCODING: [0x87,0x7d,0xa5,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1
+// CHECK-ERROR: instruction requires: sme-f16f16 or sme-f8f16
 // CHECK-UNKNOWN: c1a57d87 <unknown>
 
 fadd    za.h[w11, 7, vgx4], {z28.h - z31.h}  // 11000001-10100101-01111111-10000111
 // CHECK-INST: fadd    za.h[w11, 7, vgx4], { z28.h - z31.h }
 // CHECK-ENCODING: [0x87,0x7f,0xa5,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1
+// CHECK-ERROR: instruction requires: sme-f16f16 or sme-f8f16
 // CHECK-UNKNOWN: c1a57f87 <unknown>
 
 fadd    za.h[w11, 7], {z28.h - z31.h}  // 11000001-10100101-01111111-10000111
 // CHECK-INST: fadd    za.h[w11, 7, vgx4], { z28.h - z31.h }
 // CHECK-ENCODING: [0x87,0x7f,0xa5,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1
+// CHECK-ERROR: instruction requires: sme-f16f16 or sme-f8f16
 // CHECK-UNKNOWN: c1a57f87 <unknown>
 
 fadd    za.h[w8, 5, vgx4], {z16.h - z19.h}  // 11000001-10100101-00011110-00000101
 // CHECK-INST: fadd    za.h[w8, 5, vgx4], { z16.h - z19.h }
 // CHECK-ENCODING: [0x05,0x1e,0xa5,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1
+// CHECK-ERROR: instruction requires: sme-f16f16 or sme-f8f16
 // CHECK-UNKNOWN: c1a51e05 <unknown>
 
 fadd    za.h[w8, 5], {z16.h - z19.h}  // 11000001-10100101-00011110-00000101
 // CHECK-INST: fadd    za.h[w8, 5, vgx4], { z16.h - z19.h }
 // CHECK-ENCODING: [0x05,0x1e,0xa5,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1
+// CHECK-ERROR: instruction requires: sme-f16f16 or sme-f8f16
 // CHECK-UNKNOWN: c1a51e05 <unknown>
 
 fadd    za.h[w8, 1, vgx4], {z0.h - z3.h}  // 11000001-10100101-00011100-00000001
 // CHECK-INST: fadd    za.h[w8, 1, vgx4], { z0.h - z3.h }
 // CHECK-ENCODING: [0x01,0x1c,0xa5,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1
+// CHECK-ERROR: instruction requires: sme-f16f16 or sme-f8f16
 // CHECK-UNKNOWN: c1a51c01 <unknown>
 
 fadd    za.h[w8, 1], {z0.h - z3.h}  // 11000001-10100101-00011100-00000001
 // CHECK-INST: fadd    za.h[w8, 1, vgx4], { z0.h - z3.h }
 // CHECK-ENCODING: [0x01,0x1c,0xa5,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1
+// CHECK-ERROR: instruction requires: sme-f16f16 or sme-f8f16
 // CHECK-UNKNOWN: c1a51c01 <unknown>
 
 fadd    za.h[w10, 0, vgx4], {z16.h - z19.h}  // 11000001-10100101-01011110-00000000
 // CHECK-INST: fadd    za.h[w10, 0, vgx4], { z16.h - z19.h }
 // CHECK-ENCODING: [0x00,0x5e,0xa5,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1
+// CHECK-ERROR: instruction requires: sme-f16f16 or sme-f8f16
 // CHECK-UNKNOWN: c1a55e00 <unknown>
 
 fadd    za.h[w10, 0], {z16.h - z19.h}  // 11000001-10100101-01011110-00000000
 // CHECK-INST: fadd    za.h[w10, 0, vgx4], { z16.h - z19.h }
 // CHECK-ENCODING: [0x00,0x5e,0xa5,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1
+// CHECK-ERROR: instruction requires: sme-f16f16 or sme-f8f16
 // CHECK-UNKNOWN: c1a55e00 <unknown>
 
 fadd    za.h[w8, 0, vgx4], {z12.h - z15.h}  // 11000001-10100101-00011101-10000000
 // CHECK-INST: fadd    za.h[w8, 0, vgx4], { z12.h - z15.h }
 // CHECK-ENCODING: [0x80,0x1d,0xa5,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1
+// CHECK-ERROR: instruction requires: sme-f16f16 or sme-f8f16
 // CHECK-UNKNOWN: c1a51d80 <unknown>
 
 fadd    za.h[w8, 0], {z12.h - z15.h}  // 11000001-10100101-00011101-10000000
 // CHECK-INST: fadd    za.h[w8, 0, vgx4], { z12.h - z15.h }
 // CHECK-ENCODING: [0x80,0x1d,0xa5,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1
+// CHECK-ERROR: instruction requires: sme-f16f16 or sme-f8f16
 // CHECK-UNKNOWN: c1a51d80 <unknown>
 
 fadd    za.h[w10, 1, vgx4], {z0.h - z3.h}  // 11000001-10100101-01011100-00000001
 // CHECK-INST: fadd    za.h[w10, 1, vgx4], { z0.h - z3.h }
 // CHECK-ENCODING: [0x01,0x5c,0xa5,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1
+// CHECK-ERROR: instruction requires: sme-f16f16 or sme-f8f16
 // CHECK-UNKNOWN: c1a55c01 <unknown>
 
 fadd    za.h[w10, 1], {z0.h - z3.h}  // 11000001-10100101-01011100-00000001
 // CHECK-INST: fadd    za.h[w10, 1, vgx4], { z0.h - z3.h }
 // CHECK-ENCODING: [0x01,0x5c,0xa5,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1
+// CHECK-ERROR: instruction requires: sme-f16f16 or sme-f8f16
 // CHECK-UNKNOWN: c1a55c01 <unknown>
 
 fadd    za.h[w8, 5, vgx4], {z20.h - z23.h}  // 11000001-10100101-00011110-10000101
 // CHECK-INST: fadd    za.h[w8, 5, vgx4], { z20.h - z23.h }
 // CHECK-ENCODING: [0x85,0x1e,0xa5,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1
+// CHECK-ERROR: instruction requires: sme-f16f16 or sme-f8f16
 // CHECK-UNKNOWN: c1a51e85 <unknown>
 
 fadd    za.h[w8, 5], {z20.h - z23.h}  // 11000001-10100101-00011110-10000101
 // CHECK-INST: fadd    za.h[w8, 5, vgx4], { z20.h - z23.h }
 // CHECK-ENCODING: [0x85,0x1e,0xa5,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1
+// CHECK-ERROR: instruction requires: sme-f16f16 or sme-f8f16
 // CHECK-UNKNOWN: c1a51e85 <unknown>
 
 fadd    za.h[w11, 2, vgx4], {z8.h - z11.h}  // 11000001-10100101-01111101-00000010
 // CHECK-INST: fadd    za.h[w11, 2, vgx4], { z8.h - z11.h }
 // CHECK-ENCODING: [0x02,0x7d,0xa5,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1
+// CHECK-ERROR: instruction requires: sme-f16f16 or sme-f8f16
 // CHECK-UNKNOWN: c1a57d02 <unknown>
 
 fadd    za.h[w11, 2], {z8.h - z11.h}  // 11000001-10100101-01111101-00000010
 // CHECK-INST: fadd    za.h[w11, 2, vgx4], { z8.h - z11.h }
 // CHECK-ENCODING: [0x02,0x7d,0xa5,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1
+// CHECK-ERROR: instruction requires: sme-f16f16 or sme-f8f16
 // CHECK-UNKNOWN: c1a57d02 <unknown>
 
 fadd    za.h[w9, 7, vgx4], {z12.h - z15.h}  // 11000001-10100101-00111101-10000111
 // CHECK-INST: fadd    za.h[w9, 7, vgx4], { z12.h - z15.h }
 // CHECK-ENCODING: [0x87,0x3d,0xa5,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1
+// CHECK-ERROR: instruction requires: sme-f16f16 or sme-f8f16
 // CHECK-UNKNOWN: c1a53d87 <unknown>
 
 fadd    za.h[w9, 7], {z12.h - z15.h}  // 11000001-10100101-00111101-10000111
 // CHECK-INST: fadd    za.h[w9, 7, vgx4], { z12.h - z15.h }
 // CHECK-ENCODING: [0x87,0x3d,0xa5,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1
+// CHECK-ERROR: instruction requires: sme-f16f16 or sme-f8f16
 // CHECK-UNKNOWN: c1a53d87 <unknown>
diff --git a/llvm/test/MC/AArch64/SME2p1/fcvt.s b/llvm/test/MC/AArch64/SME2p1/fcvt.s
index b5707bad0a24..2731055dedec 100644
--- a/llvm/test/MC/AArch64/SME2p1/fcvt.s
+++ b/llvm/test/MC/AArch64/SME2p1/fcvt.s
@@ -1,36 +1,36 @@
-// RUN: llvm-mc -triple=aarch64 -show-encoding -mattr=+sme2p1,+sme-f16f16 < %s \
+// RUN: llvm-mc -triple=aarch64 -show-encoding -mattr=+sme-f16f16 < %s \
 // RUN:        | FileCheck %s --check-prefixes=CHECK-ENCODING,CHECK-INST
 // RUN: not llvm-mc -triple=aarch64 -show-encoding < %s 2>&1 \
 // RUN:        | FileCheck %s --check-prefix=CHECK-ERROR
-// RUN: llvm-mc -triple=aarch64 -filetype=obj -mattr=+sme2p1,+sme-f16f16 < %s \
-// RUN:        | llvm-objdump -d --mattr=+sme2p1,+sme-f16f16 - | FileCheck %s --check-prefix=CHECK-INST
-// RUN: llvm-mc -triple=aarch64 -filetype=obj -mattr=+sme2p1,+sme-f16f16 < %s \
+// RUN: llvm-mc -triple=aarch64 -filetype=obj -mattr=+sme-f16f16 < %s \
+// RUN:        | llvm-objdump -d --mattr=+sme-f16f16 - | FileCheck %s --check-prefix=CHECK-INST
+// RUN: llvm-mc -triple=aarch64 -filetype=obj -mattr=+sme-f16f16 < %s \
 // RUN:        | llvm-objdump -d --mattr=-sme2p1 - | FileCheck %s --check-prefix=CHECK-UNKNOWN
-// RUN: llvm-mc -triple=aarch64 -show-encoding -mattr=+sme2p1,+sme-f16f16 < %s \
+// RUN: llvm-mc -triple=aarch64 -show-encoding -mattr=+sme-f16f16 < %s \
 // RUN:        | sed '/.text/d' | sed 's/.*encoding: //g' \
-// RUN:        | llvm-mc -triple=aarch64 -mattr=+sme2p1,+sme-f16f16 -disassemble -show-encoding \
+// RUN:        | llvm-mc -triple=aarch64 -mattr=+sme-f16f16 -disassemble -show-encoding \
 // RUN:        | FileCheck %s --check-prefixes=CHECK-ENCODING,CHECK-INST
 
 fcvt    {z0.s, z1.s}, z0.h  // 11000001-10100000-11100000-00000000
 // CHECK-INST: fcvt    { z0.s, z1.s }, z0.h
 // CHECK-ENCODING: [0x00,0xe0,0xa0,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1a0e000 <unknown>
 
 fcvt    {z20.s, z21.s}, z10.h  // 11000001-10100000-11100001-01010100
 // CHECK-INST: fcvt    { z20.s, z21.s }, z10.h
 // CHECK-ENCODING: [0x54,0xe1,0xa0,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1a0e154 <unknown>
 
 fcvt    {z22.s, z23.s}, z13.h  // 11000001-10100000-11100001-10110110
 // CHECK-INST: fcvt    { z22.s, z23.s }, z13.h
 // CHECK-ENCODING: [0xb6,0xe1,0xa0,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1a0e1b6 <unknown>
 
 fcvt    {z30.s, z31.s}, z31.h  // 11000001-10100000-11100011-11111110
 // CHECK-INST: fcvt    { z30.s, z31.s }, z31.h
 // CHECK-ENCODING: [0xfe,0xe3,0xa0,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1a0e3fe <unknown>
diff --git a/llvm/test/MC/AArch64/SME2p1/fcvtl-diagnostics.s b/llvm/test/MC/AArch64/SME2p1/fcvtl-diagnostics.s
index a723d2fc6f3a..ad3eaba7bdc2 100644
--- a/llvm/test/MC/AArch64/SME2p1/fcvtl-diagnostics.s
+++ b/llvm/test/MC/AArch64/SME2p1/fcvtl-diagnostics.s
@@ -1,4 +1,4 @@
-// RUN: not llvm-mc -triple=aarch64 -show-encoding -mattr=+sme2p1,+sme-f16f16 2>&1 < %s | FileCheck %s
+// RUN: not llvm-mc -triple=aarch64 -show-encoding -mattr=+sme-f16f16 2>&1 < %s | FileCheck %s
 
 // --------------------------------------------------------------------------//
 // Invalid vector list
diff --git a/llvm/test/MC/AArch64/SME2p1/fcvtl.s b/llvm/test/MC/AArch64/SME2p1/fcvtl.s
index 31cf90d03796..6284915e4983 100644
--- a/llvm/test/MC/AArch64/SME2p1/fcvtl.s
+++ b/llvm/test/MC/AArch64/SME2p1/fcvtl.s
@@ -1,36 +1,36 @@
-// RUN: llvm-mc -triple=aarch64 -show-encoding -mattr=+sme2p1,+sme-f16f16 < %s \
+// RUN: llvm-mc -triple=aarch64 -show-encoding -mattr=+sme-f16f16 < %s \
 // RUN:        | FileCheck %s --check-prefixes=CHECK-ENCODING,CHECK-INST
 // RUN: not llvm-mc -triple=aarch64 -show-encoding < %s 2>&1 \
 // RUN:        | FileCheck %s --check-prefix=CHECK-ERROR
-// RUN: llvm-mc -triple=aarch64 -filetype=obj -mattr=+sme2p1,+sme-f16f16 < %s \
-// RUN:        | llvm-objdump -d --mattr=+sme2p1,+sme-f16f16 - | FileCheck %s --check-prefix=CHECK-INST
-// RUN: llvm-mc -triple=aarch64 -filetype=obj -mattr=+sme2p1,+sme-f16f16 < %s \
+// RUN: llvm-mc -triple=aarch64 -filetype=obj -mattr=+sme-f16f16 < %s \
+// RUN:        | llvm-objdump -d --mattr=+sme-f16f16 - | FileCheck %s --check-prefix=CHECK-INST
+// RUN: llvm-mc -triple=aarch64 -filetype=obj -mattr=+sme-f16f16 < %s \
 // RUN:        | llvm-objdump -d --mattr=-sme2p1 - | FileCheck %s --check-prefix=CHECK-UNKNOWN
-// RUN: llvm-mc -triple=aarch64 -show-encoding -mattr=+sme2p1,+sme-f16f16 < %s \
+// RUN: llvm-mc -triple=aarch64 -show-encoding -mattr=+sme-f16f16 < %s \
 // RUN:        | sed '/.text/d' | sed 's/.*encoding: //g' \
-// RUN:        | llvm-mc -triple=aarch64 -mattr=+sme2p1,+sme-f16f16 -disassemble -show-encoding \
+// RUN:        | llvm-mc -triple=aarch64 -mattr=+sme-f16f16 -disassemble -show-encoding \
 // RUN:        | FileCheck %s --check-prefixes=CHECK-ENCODING,CHECK-INST
 
 fcvtl   {z0.s, z1.s}, z0.h  // 11000001-10100000-11100000-00000001
 // CHECK-INST: fcvtl   { z0.s, z1.s }, z0.h
 // CHECK-ENCODING: [0x01,0xe0,0xa0,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1a0e001 <unknown>
 
 fcvtl   {z20.s, z21.s}, z10.h  // 11000001-10100000-11100001-01010101
 // CHECK-INST: fcvtl   { z20.s, z21.s }, z10.h
 // CHECK-ENCODING: [0x55,0xe1,0xa0,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1a0e155 <unknown>
 
 fcvtl   {z22.s, z23.s}, z13.h  // 11000001-10100000-11100001-10110111
 // CHECK-INST: fcvtl   { z22.s, z23.s }, z13.h
 // CHECK-ENCODING: [0xb7,0xe1,0xa0,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1a0e1b7 <unknown>
 
 fcvtl   {z30.s, z31.s}, z31.h  // 11000001-10100000-11100011-11111111
 // CHECK-INST: fcvtl   { z30.s, z31.s }, z31.h
 // CHECK-ENCODING: [0xff,0xe3,0xa0,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1a0e3ff <unknown>
diff --git a/llvm/test/MC/AArch64/SME2p1/fmla-diagnostics.s b/llvm/test/MC/AArch64/SME2p1/fmla-diagnostics.s
index d32f795728a2..2f0dccb57c90 100644
--- a/llvm/test/MC/AArch64/SME2p1/fmla-diagnostics.s
+++ b/llvm/test/MC/AArch64/SME2p1/fmla-diagnostics.s
@@ -1,4 +1,4 @@
-// RUN: not llvm-mc -triple=aarch64 -show-encoding -mattr=+sme2p1,+sme-f16f16 2>&1 < %s | FileCheck %s
+// RUN: not llvm-mc -triple=aarch64 -show-encoding -mattr=+sme-f16f16 2>&1 < %s | FileCheck %s
 
 // --------------------------------------------------------------------------//
 // Invalid vector list
diff --git a/llvm/test/MC/AArch64/SME2p1/fmla.s b/llvm/test/MC/AArch64/SME2p1/fmla.s
index 10529d81eed6..df9ac8076e56 100644
--- a/llvm/test/MC/AArch64/SME2p1/fmla.s
+++ b/llvm/test/MC/AArch64/SME2p1/fmla.s
@@ -1,877 +1,877 @@
-// RUN: llvm-mc -triple=aarch64 -show-encoding -mattr=+sme2p1,+sme-f16f16 < %s \
+// RUN: llvm-mc -triple=aarch64 -show-encoding -mattr=+sme-f16f16 < %s \
 // RUN:        | FileCheck %s --check-prefixes=CHECK-ENCODING,CHECK-INST
 // RUN: not llvm-mc -triple=aarch64 -show-encoding < %s 2>&1 \
 // RUN:        | FileCheck %s --check-prefix=CHECK-ERROR
-// RUN: llvm-mc -triple=aarch64 -filetype=obj -mattr=+sme2p1,+sme-f16f16 < %s \
-// RUN:        | llvm-objdump -d --mattr=+sme2p1,+sme-f16f16 - | FileCheck %s --check-prefix=CHECK-INST
-// RUN: llvm-mc -triple=aarch64 -filetype=obj -mattr=+sme2p1,+sme-f16f16 < %s \
+// RUN: llvm-mc -triple=aarch64 -filetype=obj -mattr=+sme-f16f16 < %s \
+// RUN:        | llvm-objdump -d --mattr=+sme-f16f16 - | FileCheck %s --check-prefix=CHECK-INST
+// RUN: llvm-mc -triple=aarch64 -filetype=obj -mattr=+sme-f16f16 < %s \
 // RUN:        | llvm-objdump -d --mattr=-sme2p1 - | FileCheck %s --check-prefix=CHECK-UNKNOWN
-// RUN: llvm-mc -triple=aarch64 -show-encoding -mattr=+sme2p1,+sme-f16f16 < %s \
+// RUN: llvm-mc -triple=aarch64 -show-encoding -mattr=+sme-f16f16 < %s \
 // RUN:        | sed '/.text/d' | sed 's/.*encoding: //g' \
-// RUN:        | llvm-mc -triple=aarch64 -mattr=+sme2p1,+sme-f16f16 -disassemble -show-encoding \
+// RUN:        | llvm-mc -triple=aarch64 -mattr=+sme-f16f16 -disassemble -show-encoding \
 // RUN:        | FileCheck %s --check-prefixes=CHECK-ENCODING,CHECK-INST
 
 fmla    za.h[w8, 0, vgx2], {z0.h, z1.h}, z0.h  // 11000001-00100000-00011100-00000000
 // CHECK-INST: fmla    za.h[w8, 0, vgx2], { z0.h, z1.h }, z0.h
 // CHECK-ENCODING: [0x00,0x1c,0x20,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1201c00 <unknown>
 
 fmla    za.h[w8, 0], {z0.h - z1.h}, z0.h  // 11000001-00100000-00011100-00000000
 // CHECK-INST: fmla    za.h[w8, 0, vgx2], { z0.h, z1.h }, z0.h
 // CHECK-ENCODING: [0x00,0x1c,0x20,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1201c00 <unknown>
 
 fmla    za.h[w10, 5, vgx2], {z10.h, z11.h}, z5.h  // 11000001-00100101-01011101-01000101
 // CHECK-INST: fmla    za.h[w10, 5, vgx2], { z10.h, z11.h }, z5.h
 // CHECK-ENCODING: [0x45,0x5d,0x25,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1255d45 <unknown>
 
 fmla    za.h[w10, 5], {z10.h - z11.h}, z5.h  // 11000001-00100101-01011101-01000101
 // CHECK-INST: fmla    za.h[w10, 5, vgx2], { z10.h, z11.h }, z5.h
 // CHECK-ENCODING: [0x45,0x5d,0x25,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1255d45 <unknown>
 
 fmla    za.h[w11, 7, vgx2], {z13.h, z14.h}, z8.h  // 11000001-00101000-01111101-10100111
 // CHECK-INST: fmla    za.h[w11, 7, vgx2], { z13.h, z14.h }, z8.h
 // CHECK-ENCODING: [0xa7,0x7d,0x28,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1287da7 <unknown>
 
 fmla    za.h[w11, 7], {z13.h - z14.h}, z8.h  // 11000001-00101000-01111101-10100111
 // CHECK-INST: fmla    za.h[w11, 7, vgx2], { z13.h, z14.h }, z8.h
 // CHECK-ENCODING: [0xa7,0x7d,0x28,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1287da7 <unknown>
 
 fmla    za.h[w11, 7, vgx2], {z31.h, z0.h}, z15.h  // 11000001-00101111-01111111-11100111
 // CHECK-INST: fmla    za.h[w11, 7, vgx2], { z31.h, z0.h }, z15.h
 // CHECK-ENCODING: [0xe7,0x7f,0x2f,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c12f7fe7 <unknown>
 
 fmla    za.h[w11, 7], {z31.h - z0.h}, z15.h  // 11000001-00101111-01111111-11100111
 // CHECK-INST: fmla    za.h[w11, 7, vgx2], { z31.h, z0.h }, z15.h
 // CHECK-ENCODING: [0xe7,0x7f,0x2f,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c12f7fe7 <unknown>
 
 fmla    za.h[w8, 5, vgx2], {z17.h, z18.h}, z0.h  // 11000001-00100000-00011110-00100101
 // CHECK-INST: fmla    za.h[w8, 5, vgx2], { z17.h, z18.h }, z0.h
 // CHECK-ENCODING: [0x25,0x1e,0x20,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1201e25 <unknown>
 
 fmla    za.h[w8, 5], {z17.h - z18.h}, z0.h  // 11000001-00100000-00011110-00100101
 // CHECK-INST: fmla    za.h[w8, 5, vgx2], { z17.h, z18.h }, z0.h
 // CHECK-ENCODING: [0x25,0x1e,0x20,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1201e25 <unknown>
 
 fmla    za.h[w8, 1, vgx2], {z1.h, z2.h}, z14.h  // 11000001-00101110-00011100-00100001
 // CHECK-INST: fmla    za.h[w8, 1, vgx2], { z1.h, z2.h }, z14.h
 // CHECK-ENCODING: [0x21,0x1c,0x2e,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c12e1c21 <unknown>
 
 fmla    za.h[w8, 1], {z1.h - z2.h}, z14.h  // 11000001-00101110-00011100-00100001
 // CHECK-INST: fmla    za.h[w8, 1, vgx2], { z1.h, z2.h }, z14.h
 // CHECK-ENCODING: [0x21,0x1c,0x2e,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c12e1c21 <unknown>
 
 fmla    za.h[w10, 0, vgx2], {z19.h, z20.h}, z4.h  // 11000001-00100100-01011110-01100000
 // CHECK-INST: fmla    za.h[w10, 0, vgx2], { z19.h, z20.h }, z4.h
 // CHECK-ENCODING: [0x60,0x5e,0x24,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1245e60 <unknown>
 
 fmla    za.h[w10, 0], {z19.h - z20.h}, z4.h  // 11000001-00100100-01011110-01100000
 // CHECK-INST: fmla    za.h[w10, 0, vgx2], { z19.h, z20.h }, z4.h
 // CHECK-ENCODING: [0x60,0x5e,0x24,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1245e60 <unknown>
 
 fmla    za.h[w8, 0, vgx2], {z12.h, z13.h}, z2.h  // 11000001-00100010-00011101-10000000
 // CHECK-INST: fmla    za.h[w8, 0, vgx2], { z12.h, z13.h }, z2.h
 // CHECK-ENCODING: [0x80,0x1d,0x22,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1221d80 <unknown>
 
 fmla    za.h[w8, 0], {z12.h - z13.h}, z2.h  // 11000001-00100010-00011101-10000000
 // CHECK-INST: fmla    za.h[w8, 0, vgx2], { z12.h, z13.h }, z2.h
 // CHECK-ENCODING: [0x80,0x1d,0x22,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1221d80 <unknown>
 
 fmla    za.h[w10, 1, vgx2], {z1.h, z2.h}, z10.h  // 11000001-00101010-01011100-00100001
 // CHECK-INST: fmla    za.h[w10, 1, vgx2], { z1.h, z2.h }, z10.h
 // CHECK-ENCODING: [0x21,0x5c,0x2a,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c12a5c21 <unknown>
 
 fmla    za.h[w10, 1], {z1.h - z2.h}, z10.h  // 11000001-00101010-01011100-00100001
 // CHECK-INST: fmla    za.h[w10, 1, vgx2], { z1.h, z2.h }, z10.h
 // CHECK-ENCODING: [0x21,0x5c,0x2a,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c12a5c21 <unknown>
 
 fmla    za.h[w8, 5, vgx2], {z22.h, z23.h}, z14.h  // 11000001-00101110-00011110-11000101
 // CHECK-INST: fmla    za.h[w8, 5, vgx2], { z22.h, z23.h }, z14.h
 // CHECK-ENCODING: [0xc5,0x1e,0x2e,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c12e1ec5 <unknown>
 
 fmla    za.h[w8, 5], {z22.h - z23.h}, z14.h  // 11000001-00101110-00011110-11000101
 // CHECK-INST: fmla    za.h[w8, 5, vgx2], { z22.h, z23.h }, z14.h
 // CHECK-ENCODING: [0xc5,0x1e,0x2e,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c12e1ec5 <unknown>
 
 fmla    za.h[w11, 2, vgx2], {z9.h, z10.h}, z1.h  // 11000001-00100001-01111101-00100010
 // CHECK-INST: fmla    za.h[w11, 2, vgx2], { z9.h, z10.h }, z1.h
 // CHECK-ENCODING: [0x22,0x7d,0x21,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1217d22 <unknown>
 
 fmla    za.h[w11, 2], {z9.h - z10.h}, z1.h  // 11000001-00100001-01111101-00100010
 // CHECK-INST: fmla    za.h[w11, 2, vgx2], { z9.h, z10.h }, z1.h
 // CHECK-ENCODING: [0x22,0x7d,0x21,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1217d22 <unknown>
 
 fmla    za.h[w9, 7, vgx2], {z12.h, z13.h}, z11.h  // 11000001-00101011-00111101-10000111
 // CHECK-INST: fmla    za.h[w9, 7, vgx2], { z12.h, z13.h }, z11.h
 // CHECK-ENCODING: [0x87,0x3d,0x2b,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c12b3d87 <unknown>
 
 fmla    za.h[w9, 7], {z12.h - z13.h}, z11.h  // 11000001-00101011-00111101-10000111
 // CHECK-INST: fmla    za.h[w9, 7, vgx2], { z12.h, z13.h }, z11.h
 // CHECK-ENCODING: [0x87,0x3d,0x2b,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c12b3d87 <unknown>
 
 fmla    za.h[w8, 0, vgx2], {z0.h, z1.h}, z0.h[0]  // 11000001-00010000-00010000-00000000
 // CHECK-INST: fmla    za.h[w8, 0, vgx2], { z0.h, z1.h }, z0.h[0]
 // CHECK-ENCODING: [0x00,0x10,0x10,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1101000 <unknown>
 
 fmla    za.h[w8, 0], {z0.h - z1.h}, z0.h[0]  // 11000001-00010000-00010000-00000000
 // CHECK-INST: fmla    za.h[w8, 0, vgx2], { z0.h, z1.h }, z0.h[0]
 // CHECK-ENCODING: [0x00,0x10,0x10,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1101000 <unknown>
 
 fmla    za.h[w10, 5, vgx2], {z10.h, z11.h}, z5.h[2]  // 11000001-00010101-01010101-01000101
 // CHECK-INST: fmla    za.h[w10, 5, vgx2], { z10.h, z11.h }, z5.h[2]
 // CHECK-ENCODING: [0x45,0x55,0x15,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1155545 <unknown>
 
 fmla    za.h[w10, 5], {z10.h - z11.h}, z5.h[2]  // 11000001-00010101-01010101-01000101
 // CHECK-INST: fmla    za.h[w10, 5, vgx2], { z10.h, z11.h }, z5.h[2]
 // CHECK-ENCODING: [0x45,0x55,0x15,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1155545 <unknown>
 
 fmla    za.h[w11, 7, vgx2], {z12.h, z13.h}, z8.h[6]  // 11000001-00011000-01111101-10000111
 // CHECK-INST: fmla    za.h[w11, 7, vgx2], { z12.h, z13.h }, z8.h[6]
 // CHECK-ENCODING: [0x87,0x7d,0x18,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1187d87 <unknown>
 
 fmla    za.h[w11, 7], {z12.h - z13.h}, z8.h[6]  // 11000001-00011000-01111101-10000111
 // CHECK-INST: fmla    za.h[w11, 7, vgx2], { z12.h, z13.h }, z8.h[6]
 // CHECK-ENCODING: [0x87,0x7d,0x18,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1187d87 <unknown>
 
 fmla    za.h[w11, 7, vgx2], {z30.h, z31.h}, z15.h[7]  // 11000001-00011111-01111111-11001111
 // CHECK-INST: fmla    za.h[w11, 7, vgx2], { z30.h, z31.h }, z15.h[7]
 // CHECK-ENCODING: [0xcf,0x7f,0x1f,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c11f7fcf <unknown>
 
 fmla    za.h[w11, 7], {z30.h - z31.h}, z15.h[7]  // 11000001-00011111-01111111-11001111
 // CHECK-INST: fmla    za.h[w11, 7, vgx2], { z30.h, z31.h }, z15.h[7]
 // CHECK-ENCODING: [0xcf,0x7f,0x1f,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c11f7fcf <unknown>
 
 fmla    za.h[w8, 5, vgx2], {z16.h, z17.h}, z0.h[6]  // 11000001-00010000-00011110-00000101
 // CHECK-INST: fmla    za.h[w8, 5, vgx2], { z16.h, z17.h }, z0.h[6]
 // CHECK-ENCODING: [0x05,0x1e,0x10,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1101e05 <unknown>
 
 fmla    za.h[w8, 5], {z16.h - z17.h}, z0.h[6]  // 11000001-00010000-00011110-00000101
 // CHECK-INST: fmla    za.h[w8, 5, vgx2], { z16.h, z17.h }, z0.h[6]
 // CHECK-ENCODING: [0x05,0x1e,0x10,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1101e05 <unknown>
 
 fmla    za.h[w8, 1, vgx2], {z0.h, z1.h}, z14.h[2]  // 11000001-00011110-00010100-00000001
 // CHECK-INST: fmla    za.h[w8, 1, vgx2], { z0.h, z1.h }, z14.h[2]
 // CHECK-ENCODING: [0x01,0x14,0x1e,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c11e1401 <unknown>
 
 fmla    za.h[w8, 1], {z0.h - z1.h}, z14.h[2]  // 11000001-00011110-00010100-00000001
 // CHECK-INST: fmla    za.h[w8, 1, vgx2], { z0.h, z1.h }, z14.h[2]
 // CHECK-ENCODING: [0x01,0x14,0x1e,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c11e1401 <unknown>
 
 fmla    za.h[w10, 0, vgx2], {z18.h, z19.h}, z4.h[3]  // 11000001-00010100-01010110-01001000
 // CHECK-INST: fmla    za.h[w10, 0, vgx2], { z18.h, z19.h }, z4.h[3]
 // CHECK-ENCODING: [0x48,0x56,0x14,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1145648 <unknown>
 
 fmla    za.h[w10, 0], {z18.h - z19.h}, z4.h[3]  // 11000001-00010100-01010110-01001000
 // CHECK-INST: fmla    za.h[w10, 0, vgx2], { z18.h, z19.h }, z4.h[3]
 // CHECK-ENCODING: [0x48,0x56,0x14,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1145648 <unknown>
 
 fmla    za.h[w8, 0, vgx2], {z12.h, z13.h}, z2.h[4]  // 11000001-00010010-00011001-10000000
 // CHECK-INST: fmla    za.h[w8, 0, vgx2], { z12.h, z13.h }, z2.h[4]
 // CHECK-ENCODING: [0x80,0x19,0x12,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1121980 <unknown>
 
 fmla    za.h[w8, 0], {z12.h - z13.h}, z2.h[4]  // 11000001-00010010-00011001-10000000
 // CHECK-INST: fmla    za.h[w8, 0, vgx2], { z12.h, z13.h }, z2.h[4]
 // CHECK-ENCODING: [0x80,0x19,0x12,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1121980 <unknown>
 
 fmla    za.h[w10, 1, vgx2], {z0.h, z1.h}, z10.h[4]  // 11000001-00011010-01011000-00000001
 // CHECK-INST: fmla    za.h[w10, 1, vgx2], { z0.h, z1.h }, z10.h[4]
 // CHECK-ENCODING: [0x01,0x58,0x1a,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c11a5801 <unknown>
 
 fmla    za.h[w10, 1], {z0.h - z1.h}, z10.h[4]  // 11000001-00011010-01011000-00000001
 // CHECK-INST: fmla    za.h[w10, 1, vgx2], { z0.h, z1.h }, z10.h[4]
 // CHECK-ENCODING: [0x01,0x58,0x1a,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c11a5801 <unknown>
 
 fmla    za.h[w8, 5, vgx2], {z22.h, z23.h}, z14.h[5]  // 11000001-00011110-00011010-11001101
 // CHECK-INST: fmla    za.h[w8, 5, vgx2], { z22.h, z23.h }, z14.h[5]
 // CHECK-ENCODING: [0xcd,0x1a,0x1e,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c11e1acd <unknown>
 
 fmla    za.h[w8, 5], {z22.h - z23.h}, z14.h[5]  // 11000001-00011110-00011010-11001101
 // CHECK-INST: fmla    za.h[w8, 5, vgx2], { z22.h, z23.h }, z14.h[5]
 // CHECK-ENCODING: [0xcd,0x1a,0x1e,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c11e1acd <unknown>
 
 fmla    za.h[w11, 2, vgx2], {z8.h, z9.h}, z1.h[2]  // 11000001-00010001-01110101-00000010
 // CHECK-INST: fmla    za.h[w11, 2, vgx2], { z8.h, z9.h }, z1.h[2]
 // CHECK-ENCODING: [0x02,0x75,0x11,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1117502 <unknown>
 
 fmla    za.h[w11, 2], {z8.h - z9.h}, z1.h[2]  // 11000001-00010001-01110101-00000010
 // CHECK-INST: fmla    za.h[w11, 2, vgx2], { z8.h, z9.h }, z1.h[2]
 // CHECK-ENCODING: [0x02,0x75,0x11,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1117502 <unknown>
 
 fmla    za.h[w9, 7, vgx2], {z12.h, z13.h}, z11.h[4]  // 11000001-00011011-00111001-10000111
 // CHECK-INST: fmla    za.h[w9, 7, vgx2], { z12.h, z13.h }, z11.h[4]
 // CHECK-ENCODING: [0x87,0x39,0x1b,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c11b3987 <unknown>
 
 fmla    za.h[w9, 7], {z12.h - z13.h}, z11.h[4]  // 11000001-00011011-00111001-10000111
 // CHECK-INST: fmla    za.h[w9, 7, vgx2], { z12.h, z13.h }, z11.h[4]
 // CHECK-ENCODING: [0x87,0x39,0x1b,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c11b3987 <unknown>
 
 fmla    za.h[w8, 0, vgx2], {z0.h, z1.h}, {z0.h, z1.h}  // 11000001-10100000-00010000-00001000
 // CHECK-INST: fmla    za.h[w8, 0, vgx2], { z0.h, z1.h }, { z0.h, z1.h }
 // CHECK-ENCODING: [0x08,0x10,0xa0,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1a01008 <unknown>
 
 fmla    za.h[w8, 0], {z0.h - z1.h}, {z0.h - z1.h}  // 11000001-10100000-00010000-00001000
 // CHECK-INST: fmla    za.h[w8, 0, vgx2], { z0.h, z1.h }, { z0.h, z1.h }
 // CHECK-ENCODING: [0x08,0x10,0xa0,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1a01008 <unknown>
 
 fmla    za.h[w10, 5, vgx2], {z10.h, z11.h}, {z20.h, z21.h}  // 11000001-10110100-01010001-01001101
 // CHECK-INST: fmla    za.h[w10, 5, vgx2], { z10.h, z11.h }, { z20.h, z21.h }
 // CHECK-ENCODING: [0x4d,0x51,0xb4,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1b4514d <unknown>
 
 fmla    za.h[w10, 5], {z10.h - z11.h}, {z20.h - z21.h}  // 11000001-10110100-01010001-01001101
 // CHECK-INST: fmla    za.h[w10, 5, vgx2], { z10.h, z11.h }, { z20.h, z21.h }
 // CHECK-ENCODING: [0x4d,0x51,0xb4,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1b4514d <unknown>
 
 fmla    za.h[w11, 7, vgx2], {z12.h, z13.h}, {z8.h, z9.h}  // 11000001-10101000-01110001-10001111
 // CHECK-INST: fmla    za.h[w11, 7, vgx2], { z12.h, z13.h }, { z8.h, z9.h }
 // CHECK-ENCODING: [0x8f,0x71,0xa8,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1a8718f <unknown>
 
 fmla    za.h[w11, 7], {z12.h - z13.h}, {z8.h - z9.h}  // 11000001-10101000-01110001-10001111
 // CHECK-INST: fmla    za.h[w11, 7, vgx2], { z12.h, z13.h }, { z8.h, z9.h }
 // CHECK-ENCODING: [0x8f,0x71,0xa8,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1a8718f <unknown>
 
 fmla    za.h[w11, 7, vgx2], {z30.h, z31.h}, {z30.h, z31.h}  // 11000001-10111110-01110011-11001111
 // CHECK-INST: fmla    za.h[w11, 7, vgx2], { z30.h, z31.h }, { z30.h, z31.h }
 // CHECK-ENCODING: [0xcf,0x73,0xbe,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1be73cf <unknown>
 
 fmla    za.h[w11, 7], {z30.h - z31.h}, {z30.h - z31.h}  // 11000001-10111110-01110011-11001111
 // CHECK-INST: fmla    za.h[w11, 7, vgx2], { z30.h, z31.h }, { z30.h, z31.h }
 // CHECK-ENCODING: [0xcf,0x73,0xbe,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1be73cf <unknown>
 
 fmla    za.h[w8, 5, vgx2], {z16.h, z17.h}, {z16.h, z17.h}  // 11000001-10110000-00010010-00001101
 // CHECK-INST: fmla    za.h[w8, 5, vgx2], { z16.h, z17.h }, { z16.h, z17.h }
 // CHECK-ENCODING: [0x0d,0x12,0xb0,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1b0120d <unknown>
 
 fmla    za.h[w8, 5], {z16.h - z17.h}, {z16.h - z17.h}  // 11000001-10110000-00010010-00001101
 // CHECK-INST: fmla    za.h[w8, 5, vgx2], { z16.h, z17.h }, { z16.h, z17.h }
 // CHECK-ENCODING: [0x0d,0x12,0xb0,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1b0120d <unknown>
 
 fmla    za.h[w8, 1, vgx2], {z0.h, z1.h}, {z30.h, z31.h}  // 11000001-10111110-00010000-00001001
 // CHECK-INST: fmla    za.h[w8, 1, vgx2], { z0.h, z1.h }, { z30.h, z31.h }
 // CHECK-ENCODING: [0x09,0x10,0xbe,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1be1009 <unknown>
 
 fmla    za.h[w8, 1], {z0.h - z1.h}, {z30.h - z31.h}  // 11000001-10111110-00010000-00001001
 // CHECK-INST: fmla    za.h[w8, 1, vgx2], { z0.h, z1.h }, { z30.h, z31.h }
 // CHECK-ENCODING: [0x09,0x10,0xbe,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1be1009 <unknown>
 
 fmla    za.h[w10, 0, vgx2], {z18.h, z19.h}, {z20.h, z21.h}  // 11000001-10110100-01010010-01001000
 // CHECK-INST: fmla    za.h[w10, 0, vgx2], { z18.h, z19.h }, { z20.h, z21.h }
 // CHECK-ENCODING: [0x48,0x52,0xb4,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1b45248 <unknown>
 
 fmla    za.h[w10, 0], {z18.h - z19.h}, {z20.h - z21.h}  // 11000001-10110100-01010010-01001000
 // CHECK-INST: fmla    za.h[w10, 0, vgx2], { z18.h, z19.h }, { z20.h, z21.h }
 // CHECK-ENCODING: [0x48,0x52,0xb4,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1b45248 <unknown>
 
 fmla    za.h[w8, 0, vgx2], {z12.h, z13.h}, {z2.h, z3.h}  // 11000001-10100010-00010001-10001000
 // CHECK-INST: fmla    za.h[w8, 0, vgx2], { z12.h, z13.h }, { z2.h, z3.h }
 // CHECK-ENCODING: [0x88,0x11,0xa2,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1a21188 <unknown>
 
 fmla    za.h[w8, 0], {z12.h - z13.h}, {z2.h - z3.h}  // 11000001-10100010-00010001-10001000
 // CHECK-INST: fmla    za.h[w8, 0, vgx2], { z12.h, z13.h }, { z2.h, z3.h }
 // CHECK-ENCODING: [0x88,0x11,0xa2,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1a21188 <unknown>
 
 fmla    za.h[w10, 1, vgx2], {z0.h, z1.h}, {z26.h, z27.h}  // 11000001-10111010-01010000-00001001
 // CHECK-INST: fmla    za.h[w10, 1, vgx2], { z0.h, z1.h }, { z26.h, z27.h }
 // CHECK-ENCODING: [0x09,0x50,0xba,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1ba5009 <unknown>
 
 fmla    za.h[w10, 1], {z0.h - z1.h}, {z26.h - z27.h}  // 11000001-10111010-01010000-00001001
 // CHECK-INST: fmla    za.h[w10, 1, vgx2], { z0.h, z1.h }, { z26.h, z27.h }
 // CHECK-ENCODING: [0x09,0x50,0xba,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1ba5009 <unknown>
 
 fmla    za.h[w8, 5, vgx2], {z22.h, z23.h}, {z30.h, z31.h}  // 11000001-10111110-00010010-11001101
 // CHECK-INST: fmla    za.h[w8, 5, vgx2], { z22.h, z23.h }, { z30.h, z31.h }
 // CHECK-ENCODING: [0xcd,0x12,0xbe,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1be12cd <unknown>
 
 fmla    za.h[w8, 5], {z22.h - z23.h}, {z30.h - z31.h}  // 11000001-10111110-00010010-11001101
 // CHECK-INST: fmla    za.h[w8, 5, vgx2], { z22.h, z23.h }, { z30.h, z31.h }
 // CHECK-ENCODING: [0xcd,0x12,0xbe,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1be12cd <unknown>
 
 fmla    za.h[w11, 2, vgx2], {z8.h, z9.h}, {z0.h, z1.h}  // 11000001-10100000-01110001-00001010
 // CHECK-INST: fmla    za.h[w11, 2, vgx2], { z8.h, z9.h }, { z0.h, z1.h }
 // CHECK-ENCODING: [0x0a,0x71,0xa0,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1a0710a <unknown>
 
 fmla    za.h[w11, 2], {z8.h - z9.h}, {z0.h - z1.h}  // 11000001-10100000-01110001-00001010
 // CHECK-INST: fmla    za.h[w11, 2, vgx2], { z8.h, z9.h }, { z0.h, z1.h }
 // CHECK-ENCODING: [0x0a,0x71,0xa0,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1a0710a <unknown>
 
 fmla    za.h[w9, 7, vgx2], {z12.h, z13.h}, {z10.h, z11.h}  // 11000001-10101010-00110001-10001111
 // CHECK-INST: fmla    za.h[w9, 7, vgx2], { z12.h, z13.h }, { z10.h, z11.h }
 // CHECK-ENCODING: [0x8f,0x31,0xaa,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1aa318f <unknown>
 
 fmla    za.h[w9, 7], {z12.h - z13.h}, {z10.h - z11.h}  // 11000001-10101010-00110001-10001111
 // CHECK-INST: fmla    za.h[w9, 7, vgx2], { z12.h, z13.h }, { z10.h, z11.h }
 // CHECK-ENCODING: [0x8f,0x31,0xaa,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1aa318f <unknown>
 
 
 fmla    za.h[w8, 0, vgx4], {z0.h - z3.h}, z0.h  // 11000001-00110000-00011100-00000000
 // CHECK-INST: fmla    za.h[w8, 0, vgx4], { z0.h - z3.h }, z0.h
 // CHECK-ENCODING: [0x00,0x1c,0x30,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1301c00 <unknown>
 
 fmla    za.h[w8, 0], {z0.h - z3.h}, z0.h  // 11000001-00110000-00011100-00000000
 // CHECK-INST: fmla    za.h[w8, 0, vgx4], { z0.h - z3.h }, z0.h
 // CHECK-ENCODING: [0x00,0x1c,0x30,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1301c00 <unknown>
 
 fmla    za.h[w10, 5, vgx4], {z10.h - z13.h}, z5.h  // 11000001-00110101-01011101-01000101
 // CHECK-INST: fmla    za.h[w10, 5, vgx4], { z10.h - z13.h }, z5.h
 // CHECK-ENCODING: [0x45,0x5d,0x35,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1355d45 <unknown>
 
 fmla    za.h[w10, 5], {z10.h - z13.h}, z5.h  // 11000001-00110101-01011101-01000101
 // CHECK-INST: fmla    za.h[w10, 5, vgx4], { z10.h - z13.h }, z5.h
 // CHECK-ENCODING: [0x45,0x5d,0x35,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1355d45 <unknown>
 
 fmla    za.h[w11, 7, vgx4], {z13.h - z16.h}, z8.h  // 11000001-00111000-01111101-10100111
 // CHECK-INST: fmla    za.h[w11, 7, vgx4], { z13.h - z16.h }, z8.h
 // CHECK-ENCODING: [0xa7,0x7d,0x38,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1387da7 <unknown>
 
 fmla    za.h[w11, 7], {z13.h - z16.h}, z8.h  // 11000001-00111000-01111101-10100111
 // CHECK-INST: fmla    za.h[w11, 7, vgx4], { z13.h - z16.h }, z8.h
 // CHECK-ENCODING: [0xa7,0x7d,0x38,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1387da7 <unknown>
 
 fmla    za.h[w11, 7, vgx4], {z31.h, z0.h, z1.h, z2.h}, z15.h  // 11000001-00111111-01111111-11100111
 // CHECK-INST: fmla    za.h[w11, 7, vgx4], { z31.h, z0.h, z1.h, z2.h }, z15.h
 // CHECK-ENCODING: [0xe7,0x7f,0x3f,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c13f7fe7 <unknown>
 
 fmla    za.h[w11, 7], {z31.h, z0.h, z1.h, z2.h}, z15.h  // 11000001-00111111-01111111-11100111
 // CHECK-INST: fmla    za.h[w11, 7, vgx4], { z31.h, z0.h, z1.h, z2.h }, z15.h
 // CHECK-ENCODING: [0xe7,0x7f,0x3f,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c13f7fe7 <unknown>
 
 fmla    za.h[w8, 5, vgx4], {z17.h - z20.h}, z0.h  // 11000001-00110000-00011110-00100101
 // CHECK-INST: fmla    za.h[w8, 5, vgx4], { z17.h - z20.h }, z0.h
 // CHECK-ENCODING: [0x25,0x1e,0x30,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1301e25 <unknown>
 
 fmla    za.h[w8, 5], {z17.h - z20.h}, z0.h  // 11000001-00110000-00011110-00100101
 // CHECK-INST: fmla    za.h[w8, 5, vgx4], { z17.h - z20.h }, z0.h
 // CHECK-ENCODING: [0x25,0x1e,0x30,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1301e25 <unknown>
 
 fmla    za.h[w8, 1, vgx4], {z1.h - z4.h}, z14.h  // 11000001-00111110-00011100-00100001
 // CHECK-INST: fmla    za.h[w8, 1, vgx4], { z1.h - z4.h }, z14.h
 // CHECK-ENCODING: [0x21,0x1c,0x3e,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c13e1c21 <unknown>
 
 fmla    za.h[w8, 1], {z1.h - z4.h}, z14.h  // 11000001-00111110-00011100-00100001
 // CHECK-INST: fmla    za.h[w8, 1, vgx4], { z1.h - z4.h }, z14.h
 // CHECK-ENCODING: [0x21,0x1c,0x3e,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c13e1c21 <unknown>
 
 fmla    za.h[w10, 0, vgx4], {z19.h - z22.h}, z4.h  // 11000001-00110100-01011110-01100000
 // CHECK-INST: fmla    za.h[w10, 0, vgx4], { z19.h - z22.h }, z4.h
 // CHECK-ENCODING: [0x60,0x5e,0x34,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1345e60 <unknown>
 
 fmla    za.h[w10, 0], {z19.h - z22.h}, z4.h  // 11000001-00110100-01011110-01100000
 // CHECK-INST: fmla    za.h[w10, 0, vgx4], { z19.h - z22.h }, z4.h
 // CHECK-ENCODING: [0x60,0x5e,0x34,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1345e60 <unknown>
 
 fmla    za.h[w8, 0, vgx4], {z12.h - z15.h}, z2.h  // 11000001-00110010-00011101-10000000
 // CHECK-INST: fmla    za.h[w8, 0, vgx4], { z12.h - z15.h }, z2.h
 // CHECK-ENCODING: [0x80,0x1d,0x32,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1321d80 <unknown>
 
 fmla    za.h[w8, 0], {z12.h - z15.h}, z2.h  // 11000001-00110010-00011101-10000000
 // CHECK-INST: fmla    za.h[w8, 0, vgx4], { z12.h - z15.h }, z2.h
 // CHECK-ENCODING: [0x80,0x1d,0x32,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1321d80 <unknown>
 
 fmla    za.h[w10, 1, vgx4], {z1.h - z4.h}, z10.h  // 11000001-00111010-01011100-00100001
 // CHECK-INST: fmla    za.h[w10, 1, vgx4], { z1.h - z4.h }, z10.h
 // CHECK-ENCODING: [0x21,0x5c,0x3a,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c13a5c21 <unknown>
 
 fmla    za.h[w10, 1], {z1.h - z4.h}, z10.h  // 11000001-00111010-01011100-00100001
 // CHECK-INST: fmla    za.h[w10, 1, vgx4], { z1.h - z4.h }, z10.h
 // CHECK-ENCODING: [0x21,0x5c,0x3a,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c13a5c21 <unknown>
 
 fmla    za.h[w8, 5, vgx4], {z22.h - z25.h}, z14.h  // 11000001-00111110-00011110-11000101
 // CHECK-INST: fmla    za.h[w8, 5, vgx4], { z22.h - z25.h }, z14.h
 // CHECK-ENCODING: [0xc5,0x1e,0x3e,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c13e1ec5 <unknown>
 
 fmla    za.h[w8, 5], {z22.h - z25.h}, z14.h  // 11000001-00111110-00011110-11000101
 // CHECK-INST: fmla    za.h[w8, 5, vgx4], { z22.h - z25.h }, z14.h
 // CHECK-ENCODING: [0xc5,0x1e,0x3e,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c13e1ec5 <unknown>
 
 fmla    za.h[w11, 2, vgx4], {z9.h - z12.h}, z1.h  // 11000001-00110001-01111101-00100010
 // CHECK-INST: fmla    za.h[w11, 2, vgx4], { z9.h - z12.h }, z1.h
 // CHECK-ENCODING: [0x22,0x7d,0x31,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1317d22 <unknown>
 
 fmla    za.h[w11, 2], {z9.h - z12.h}, z1.h  // 11000001-00110001-01111101-00100010
 // CHECK-INST: fmla    za.h[w11, 2, vgx4], { z9.h - z12.h }, z1.h
 // CHECK-ENCODING: [0x22,0x7d,0x31,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1317d22 <unknown>
 
 fmla    za.h[w9, 7, vgx4], {z12.h - z15.h}, z11.h  // 11000001-00111011-00111101-10000111
 // CHECK-INST: fmla    za.h[w9, 7, vgx4], { z12.h - z15.h }, z11.h
 // CHECK-ENCODING: [0x87,0x3d,0x3b,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c13b3d87 <unknown>
 
 fmla    za.h[w9, 7], {z12.h - z15.h}, z11.h  // 11000001-00111011-00111101-10000111
 // CHECK-INST: fmla    za.h[w9, 7, vgx4], { z12.h - z15.h }, z11.h
 // CHECK-ENCODING: [0x87,0x3d,0x3b,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c13b3d87 <unknown>
 
 fmla    za.h[w8, 0, vgx4], {z0.h - z3.h}, z0.h[0]  // 11000001-00010000-10010000-00000000
 // CHECK-INST: fmla    za.h[w8, 0, vgx4], { z0.h - z3.h }, z0.h[0]
 // CHECK-ENCODING: [0x00,0x90,0x10,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1109000 <unknown>
 
 fmla    za.h[w8, 0], {z0.h - z3.h}, z0.h[0]  // 11000001-00010000-10010000-00000000
 // CHECK-INST: fmla    za.h[w8, 0, vgx4], { z0.h - z3.h }, z0.h[0]
 // CHECK-ENCODING: [0x00,0x90,0x10,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1109000 <unknown>
 
 fmla    za.h[w10, 5, vgx4], {z8.h - z11.h}, z5.h[2]  // 11000001-00010101-11010101-00000101
 // CHECK-INST: fmla    za.h[w10, 5, vgx4], { z8.h - z11.h }, z5.h[2]
 // CHECK-ENCODING: [0x05,0xd5,0x15,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c115d505 <unknown>
 
 fmla    za.h[w10, 5], {z8.h - z11.h}, z5.h[2]  // 11000001-00010101-11010101-00000101
 // CHECK-INST: fmla    za.h[w10, 5, vgx4], { z8.h - z11.h }, z5.h[2]
 // CHECK-ENCODING: [0x05,0xd5,0x15,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c115d505 <unknown>
 
 fmla    za.h[w11, 7, vgx4], {z12.h - z15.h}, z8.h[6]  // 11000001-00011000-11111101-10000111
 // CHECK-INST: fmla    za.h[w11, 7, vgx4], { z12.h - z15.h }, z8.h[6]
 // CHECK-ENCODING: [0x87,0xfd,0x18,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c118fd87 <unknown>
 
 fmla    za.h[w11, 7], {z12.h - z15.h}, z8.h[6]  // 11000001-00011000-11111101-10000111
 // CHECK-INST: fmla    za.h[w11, 7, vgx4], { z12.h - z15.h }, z8.h[6]
 // CHECK-ENCODING: [0x87,0xfd,0x18,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c118fd87 <unknown>
 
 fmla    za.h[w11, 7, vgx4], {z28.h - z31.h}, z15.h[7]  // 11000001-00011111-11111111-10001111
 // CHECK-INST: fmla    za.h[w11, 7, vgx4], { z28.h - z31.h }, z15.h[7]
 // CHECK-ENCODING: [0x8f,0xff,0x1f,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c11fff8f <unknown>
 
 fmla    za.h[w11, 7], {z28.h - z31.h}, z15.h[7]  // 11000001-00011111-11111111-10001111
 // CHECK-INST: fmla    za.h[w11, 7, vgx4], { z28.h - z31.h }, z15.h[7]
 // CHECK-ENCODING: [0x8f,0xff,0x1f,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c11fff8f <unknown>
 
 fmla    za.h[w8, 5, vgx4], {z16.h - z19.h}, z0.h[6]  // 11000001-00010000-10011110-00000101
 // CHECK-INST: fmla    za.h[w8, 5, vgx4], { z16.h - z19.h }, z0.h[6]
 // CHECK-ENCODING: [0x05,0x9e,0x10,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1109e05 <unknown>
 
 fmla    za.h[w8, 5], {z16.h - z19.h}, z0.h[6]  // 11000001-00010000-10011110-00000101
 // CHECK-INST: fmla    za.h[w8, 5, vgx4], { z16.h - z19.h }, z0.h[6]
 // CHECK-ENCODING: [0x05,0x9e,0x10,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1109e05 <unknown>
 
 fmla    za.h[w8, 1, vgx4], {z0.h - z3.h}, z14.h[2]  // 11000001-00011110-10010100-00000001
 // CHECK-INST: fmla    za.h[w8, 1, vgx4], { z0.h - z3.h }, z14.h[2]
 // CHECK-ENCODING: [0x01,0x94,0x1e,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c11e9401 <unknown>
 
 fmla    za.h[w8, 1], {z0.h - z3.h}, z14.h[2]  // 11000001-00011110-10010100-00000001
 // CHECK-INST: fmla    za.h[w8, 1, vgx4], { z0.h - z3.h }, z14.h[2]
 // CHECK-ENCODING: [0x01,0x94,0x1e,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c11e9401 <unknown>
 
 fmla    za.h[w10, 0, vgx4], {z16.h - z19.h}, z4.h[3]  // 11000001-00010100-11010110-00001000
 // CHECK-INST: fmla    za.h[w10, 0, vgx4], { z16.h - z19.h }, z4.h[3]
 // CHECK-ENCODING: [0x08,0xd6,0x14,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c114d608 <unknown>
 
 fmla    za.h[w10, 0], {z16.h - z19.h}, z4.h[3]  // 11000001-00010100-11010110-00001000
 // CHECK-INST: fmla    za.h[w10, 0, vgx4], { z16.h - z19.h }, z4.h[3]
 // CHECK-ENCODING: [0x08,0xd6,0x14,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c114d608 <unknown>
 
 fmla    za.h[w8, 0, vgx4], {z12.h - z15.h}, z2.h[4]  // 11000001-00010010-10011001-10000000
 // CHECK-INST: fmla    za.h[w8, 0, vgx4], { z12.h - z15.h }, z2.h[4]
 // CHECK-ENCODING: [0x80,0x99,0x12,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1129980 <unknown>
 
 fmla    za.h[w8, 0], {z12.h - z15.h}, z2.h[4]  // 11000001-00010010-10011001-10000000
 // CHECK-INST: fmla    za.h[w8, 0, vgx4], { z12.h - z15.h }, z2.h[4]
 // CHECK-ENCODING: [0x80,0x99,0x12,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1129980 <unknown>
 
 fmla    za.h[w10, 1, vgx4], {z0.h - z3.h}, z10.h[4]  // 11000001-00011010-11011000-00000001
 // CHECK-INST: fmla    za.h[w10, 1, vgx4], { z0.h - z3.h }, z10.h[4]
 // CHECK-ENCODING: [0x01,0xd8,0x1a,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c11ad801 <unknown>
 
 fmla    za.h[w10, 1], {z0.h - z3.h}, z10.h[4]  // 11000001-00011010-11011000-00000001
 // CHECK-INST: fmla    za.h[w10, 1, vgx4], { z0.h - z3.h }, z10.h[4]
 // CHECK-ENCODING: [0x01,0xd8,0x1a,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c11ad801 <unknown>
 
 fmla    za.h[w8, 5, vgx4], {z20.h - z23.h}, z14.h[5]  // 11000001-00011110-10011010-10001101
 // CHECK-INST: fmla    za.h[w8, 5, vgx4], { z20.h - z23.h }, z14.h[5]
 // CHECK-ENCODING: [0x8d,0x9a,0x1e,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c11e9a8d <unknown>
 
 fmla    za.h[w8, 5], {z20.h - z23.h}, z14.h[5]  // 11000001-00011110-10011010-10001101
 // CHECK-INST: fmla    za.h[w8, 5, vgx4], { z20.h - z23.h }, z14.h[5]
 // CHECK-ENCODING: [0x8d,0x9a,0x1e,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c11e9a8d <unknown>
 
 fmla    za.h[w11, 2, vgx4], {z8.h - z11.h}, z1.h[2]  // 11000001-00010001-11110101-00000010
 // CHECK-INST: fmla    za.h[w11, 2, vgx4], { z8.h - z11.h }, z1.h[2]
 // CHECK-ENCODING: [0x02,0xf5,0x11,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c111f502 <unknown>
 
 fmla    za.h[w11, 2], {z8.h - z11.h}, z1.h[2]  // 11000001-00010001-11110101-00000010
 // CHECK-INST: fmla    za.h[w11, 2, vgx4], { z8.h - z11.h }, z1.h[2]
 // CHECK-ENCODING: [0x02,0xf5,0x11,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c111f502 <unknown>
 
 fmla    za.h[w9, 7, vgx4], {z12.h - z15.h}, z11.h[4]  // 11000001-00011011-10111001-10000111
 // CHECK-INST: fmla    za.h[w9, 7, vgx4], { z12.h - z15.h }, z11.h[4]
 // CHECK-ENCODING: [0x87,0xb9,0x1b,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c11bb987 <unknown>
 
 fmla    za.h[w9, 7], {z12.h - z15.h}, z11.h[4]  // 11000001-00011011-10111001-10000111
 // CHECK-INST: fmla    za.h[w9, 7, vgx4], { z12.h - z15.h }, z11.h[4]
 // CHECK-ENCODING: [0x87,0xb9,0x1b,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c11bb987 <unknown>
 
 fmla    za.h[w8, 0, vgx4], {z0.h - z3.h}, {z0.h - z3.h}  // 11000001-10100001-00010000-00001000
 // CHECK-INST: fmla    za.h[w8, 0, vgx4], { z0.h - z3.h }, { z0.h - z3.h }
 // CHECK-ENCODING: [0x08,0x10,0xa1,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1a11008 <unknown>
 
 fmla    za.h[w8, 0], {z0.h - z3.h}, {z0.h - z3.h}  // 11000001-10100001-00010000-00001000
 // CHECK-INST: fmla    za.h[w8, 0, vgx4], { z0.h - z3.h }, { z0.h - z3.h }
 // CHECK-ENCODING: [0x08,0x10,0xa1,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1a11008 <unknown>
 
 fmla    za.h[w10, 5, vgx4], {z8.h - z11.h}, {z20.h - z23.h}  // 11000001-10110101-01010001-00001101
 // CHECK-INST: fmla    za.h[w10, 5, vgx4], { z8.h - z11.h }, { z20.h - z23.h }
 // CHECK-ENCODING: [0x0d,0x51,0xb5,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1b5510d <unknown>
 
 fmla    za.h[w10, 5], {z8.h - z11.h}, {z20.h - z23.h}  // 11000001-10110101-01010001-00001101
 // CHECK-INST: fmla    za.h[w10, 5, vgx4], { z8.h - z11.h }, { z20.h - z23.h }
 // CHECK-ENCODING: [0x0d,0x51,0xb5,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1b5510d <unknown>
 
 fmla    za.h[w11, 7, vgx4], {z12.h - z15.h}, {z8.h - z11.h}  // 11000001-10101001-01110001-10001111
 // CHECK-INST: fmla    za.h[w11, 7, vgx4], { z12.h - z15.h }, { z8.h - z11.h }
 // CHECK-ENCODING: [0x8f,0x71,0xa9,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1a9718f <unknown>
 
 fmla    za.h[w11, 7], {z12.h - z15.h}, {z8.h - z11.h}  // 11000001-10101001-01110001-10001111
 // CHECK-INST: fmla    za.h[w11, 7, vgx4], { z12.h - z15.h }, { z8.h - z11.h }
 // CHECK-ENCODING: [0x8f,0x71,0xa9,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1a9718f <unknown>
 
 fmla    za.h[w11, 7, vgx4], {z28.h - z31.h}, {z28.h - z31.h}  // 11000001-10111101-01110011-10001111
 // CHECK-INST: fmla    za.h[w11, 7, vgx4], { z28.h - z31.h }, { z28.h - z31.h }
 // CHECK-ENCODING: [0x8f,0x73,0xbd,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1bd738f <unknown>
 
 fmla    za.h[w11, 7], {z28.h - z31.h}, {z28.h - z31.h}  // 11000001-10111101-01110011-10001111
 // CHECK-INST: fmla    za.h[w11, 7, vgx4], { z28.h - z31.h }, { z28.h - z31.h }
 // CHECK-ENCODING: [0x8f,0x73,0xbd,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1bd738f <unknown>
 
 fmla    za.h[w8, 5, vgx4], {z16.h - z19.h}, {z16.h - z19.h}  // 11000001-10110001-00010010-00001101
 // CHECK-INST: fmla    za.h[w8, 5, vgx4], { z16.h - z19.h }, { z16.h - z19.h }
 // CHECK-ENCODING: [0x0d,0x12,0xb1,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1b1120d <unknown>
 
 fmla    za.h[w8, 5], {z16.h - z19.h}, {z16.h - z19.h}  // 11000001-10110001-00010010-00001101
 // CHECK-INST: fmla    za.h[w8, 5, vgx4], { z16.h - z19.h }, { z16.h - z19.h }
 // CHECK-ENCODING: [0x0d,0x12,0xb1,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1b1120d <unknown>
 
 fmla    za.h[w8, 1, vgx4], {z0.h - z3.h}, {z28.h - z31.h}  // 11000001-10111101-00010000-00001001
 // CHECK-INST: fmla    za.h[w8, 1, vgx4], { z0.h - z3.h }, { z28.h - z31.h }
 // CHECK-ENCODING: [0x09,0x10,0xbd,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1bd1009 <unknown>
 
 fmla    za.h[w8, 1], {z0.h - z3.h}, {z28.h - z31.h}  // 11000001-10111101-00010000-00001001
 // CHECK-INST: fmla    za.h[w8, 1, vgx4], { z0.h - z3.h }, { z28.h - z31.h }
 // CHECK-ENCODING: [0x09,0x10,0xbd,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1bd1009 <unknown>
 
 fmla    za.h[w10, 0, vgx4], {z16.h - z19.h}, {z20.h - z23.h}  // 11000001-10110101-01010010-00001000
 // CHECK-INST: fmla    za.h[w10, 0, vgx4], { z16.h - z19.h }, { z20.h - z23.h }
 // CHECK-ENCODING: [0x08,0x52,0xb5,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1b55208 <unknown>
 
 fmla    za.h[w10, 0], {z16.h - z19.h}, {z20.h - z23.h}  // 11000001-10110101-01010010-00001000
 // CHECK-INST: fmla    za.h[w10, 0, vgx4], { z16.h - z19.h }, { z20.h - z23.h }
 // CHECK-ENCODING: [0x08,0x52,0xb5,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1b55208 <unknown>
 
 fmla    za.h[w8, 0, vgx4], {z12.h - z15.h}, {z0.h - z3.h}  // 11000001-10100001-00010001-10001000
 // CHECK-INST: fmla    za.h[w8, 0, vgx4], { z12.h - z15.h }, { z0.h - z3.h }
 // CHECK-ENCODING: [0x88,0x11,0xa1,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1a11188 <unknown>
 
 fmla    za.h[w8, 0], {z12.h - z15.h}, {z0.h - z3.h}  // 11000001-10100001-00010001-10001000
 // CHECK-INST: fmla    za.h[w8, 0, vgx4], { z12.h - z15.h }, { z0.h - z3.h }
 // CHECK-ENCODING: [0x88,0x11,0xa1,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1a11188 <unknown>
 
 fmla    za.h[w10, 1, vgx4], {z0.h - z3.h}, {z24.h - z27.h}  // 11000001-10111001-01010000-00001001
 // CHECK-INST: fmla    za.h[w10, 1, vgx4], { z0.h - z3.h }, { z24.h - z27.h }
 // CHECK-ENCODING: [0x09,0x50,0xb9,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1b95009 <unknown>
 
 fmla    za.h[w10, 1], {z0.h - z3.h}, {z24.h - z27.h}  // 11000001-10111001-01010000-00001001
 // CHECK-INST: fmla    za.h[w10, 1, vgx4], { z0.h - z3.h }, { z24.h - z27.h }
 // CHECK-ENCODING: [0x09,0x50,0xb9,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1b95009 <unknown>
 
 fmla    za.h[w8, 5, vgx4], {z20.h - z23.h}, {z28.h - z31.h}  // 11000001-10111101-00010010-10001101
 // CHECK-INST: fmla    za.h[w8, 5, vgx4], { z20.h - z23.h }, { z28.h - z31.h }
 // CHECK-ENCODING: [0x8d,0x12,0xbd,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1bd128d <unknown>
 
 fmla    za.h[w8, 5], {z20.h - z23.h}, {z28.h - z31.h}  // 11000001-10111101-00010010-10001101
 // CHECK-INST: fmla    za.h[w8, 5, vgx4], { z20.h - z23.h }, { z28.h - z31.h }
 // CHECK-ENCODING: [0x8d,0x12,0xbd,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1bd128d <unknown>
 
 fmla    za.h[w11, 2, vgx4], {z8.h - z11.h}, {z0.h - z3.h}  // 11000001-10100001-01110001-00001010
 // CHECK-INST: fmla    za.h[w11, 2, vgx4], { z8.h - z11.h }, { z0.h - z3.h }
 // CHECK-ENCODING: [0x0a,0x71,0xa1,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1a1710a <unknown>
 
 fmla    za.h[w11, 2], {z8.h - z11.h}, {z0.h - z3.h}  // 11000001-10100001-01110001-00001010
 // CHECK-INST: fmla    za.h[w11, 2, vgx4], { z8.h - z11.h }, { z0.h - z3.h }
 // CHECK-ENCODING: [0x0a,0x71,0xa1,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1a1710a <unknown>
 
 fmla    za.h[w9, 7, vgx4], {z12.h - z15.h}, {z8.h - z11.h}  // 11000001-10101001-00110001-10001111
 // CHECK-INST: fmla    za.h[w9, 7, vgx4], { z12.h - z15.h }, { z8.h - z11.h }
 // CHECK-ENCODING: [0x8f,0x31,0xa9,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1a9318f <unknown>
 
 fmla    za.h[w9, 7], {z12.h - z15.h}, {z8.h - z11.h}  // 11000001-10101001-00110001-10001111
 // CHECK-INST: fmla    za.h[w9, 7, vgx4], { z12.h - z15.h }, { z8.h - z11.h }
 // CHECK-ENCODING: [0x8f,0x31,0xa9,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1a9318f <unknown>
diff --git a/llvm/test/MC/AArch64/SME2p1/fmls-diagnostics.s b/llvm/test/MC/AArch64/SME2p1/fmls-diagnostics.s
index 2174e4202ba0..3ff09321e343 100644
--- a/llvm/test/MC/AArch64/SME2p1/fmls-diagnostics.s
+++ b/llvm/test/MC/AArch64/SME2p1/fmls-diagnostics.s
@@ -1,4 +1,4 @@
-// RUN: not llvm-mc -triple=aarch64 -show-encoding -mattr=+sme2p1,+sme-f16f16 2>&1 < %s | FileCheck %s
+// RUN: not llvm-mc -triple=aarch64 -show-encoding -mattr=+sme-f16f16 2>&1 < %s | FileCheck %s
 
 // --------------------------------------------------------------------------//
 // Invalid vector list
diff --git a/llvm/test/MC/AArch64/SME2p1/fmls.s b/llvm/test/MC/AArch64/SME2p1/fmls.s
index 9bbb21869e37..67b1430240e8 100644
--- a/llvm/test/MC/AArch64/SME2p1/fmls.s
+++ b/llvm/test/MC/AArch64/SME2p1/fmls.s
@@ -1,878 +1,878 @@
-// RUN: llvm-mc -triple=aarch64 -show-encoding -mattr=+sme2p1,+sme-f16f16 < %s \
+// RUN: llvm-mc -triple=aarch64 -show-encoding -mattr=+sme-f16f16 < %s \
 // RUN:        | FileCheck %s --check-prefixes=CHECK-ENCODING,CHECK-INST
 // RUN: not llvm-mc -triple=aarch64 -show-encoding < %s 2>&1 \
 // RUN:        | FileCheck %s --check-prefix=CHECK-ERROR
-// RUN: llvm-mc -triple=aarch64 -filetype=obj -mattr=+sme2p1,+sme-f16f16 < %s \
-// RUN:        | llvm-objdump -d --mattr=+sme2p1,+sme-f16f16 - | FileCheck %s --check-prefix=CHECK-INST
-// RUN: llvm-mc -triple=aarch64 -filetype=obj -mattr=+sme2p1,+sme-f16f16 < %s \
+// RUN: llvm-mc -triple=aarch64 -filetype=obj -mattr=+sme-f16f16 < %s \
+// RUN:        | llvm-objdump -d --mattr=+sme-f16f16 - | FileCheck %s --check-prefix=CHECK-INST
+// RUN: llvm-mc -triple=aarch64 -filetype=obj -mattr=+sme-f16f16 < %s \
 // RUN:        | llvm-objdump -d  --mattr=-sme2p1 - | FileCheck %s --check-prefix=CHECK-UNKNOWN
-// RUN: llvm-mc -triple=aarch64 -show-encoding -mattr=+sme2p1,+sme-f16f16 < %s \
+// RUN: llvm-mc -triple=aarch64 -show-encoding -mattr=+sme-f16f16 < %s \
 // RUN:        | sed '/.text/d' | sed 's/.*encoding: //g' \
-// RUN:        | llvm-mc -triple=aarch64 -mattr=+sme2p1,+sme-f16f16 -disassemble -show-encoding \
+// RUN:        | llvm-mc -triple=aarch64 -mattr=+sme-f16f16 -disassemble -show-encoding \
 // RUN:        | FileCheck %s --check-prefixes=CHECK-ENCODING,CHECK-INST
 
 fmls    za.h[w8, 0, vgx2], {z0.h, z1.h}, z0.h  // 11000001-00100000-00011100-00001000
 // CHECK-INST: fmls    za.h[w8, 0, vgx2], { z0.h, z1.h }, z0.h
 // CHECK-ENCODING: [0x08,0x1c,0x20,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1201c08 <unknown>
 
 fmls    za.h[w8, 0], {z0.h - z1.h}, z0.h  // 11000001-00100000-00011100-00001000
 // CHECK-INST: fmls    za.h[w8, 0, vgx2], { z0.h, z1.h }, z0.h
 // CHECK-ENCODING: [0x08,0x1c,0x20,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1201c08 <unknown>
 
 fmls    za.h[w10, 5, vgx2], {z10.h, z11.h}, z5.h  // 11000001-00100101-01011101-01001101
 // CHECK-INST: fmls    za.h[w10, 5, vgx2], { z10.h, z11.h }, z5.h
 // CHECK-ENCODING: [0x4d,0x5d,0x25,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1255d4d <unknown>
 
 fmls    za.h[w10, 5], {z10.h - z11.h}, z5.h  // 11000001-00100101-01011101-01001101
 // CHECK-INST: fmls    za.h[w10, 5, vgx2], { z10.h, z11.h }, z5.h
 // CHECK-ENCODING: [0x4d,0x5d,0x25,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1255d4d <unknown>
 
 fmls    za.h[w11, 7, vgx2], {z13.h, z14.h}, z8.h  // 11000001-00101000-01111101-10101111
 // CHECK-INST: fmls    za.h[w11, 7, vgx2], { z13.h, z14.h }, z8.h
 // CHECK-ENCODING: [0xaf,0x7d,0x28,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1287daf <unknown>
 
 fmls    za.h[w11, 7], {z13.h - z14.h}, z8.h  // 11000001-00101000-01111101-10101111
 // CHECK-INST: fmls    za.h[w11, 7, vgx2], { z13.h, z14.h }, z8.h
 // CHECK-ENCODING: [0xaf,0x7d,0x28,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1287daf <unknown>
 
 fmls    za.h[w11, 7, vgx2], {z31.h, z0.h}, z15.h  // 11000001-00101111-01111111-11101111
 // CHECK-INST: fmls    za.h[w11, 7, vgx2], { z31.h, z0.h }, z15.h
 // CHECK-ENCODING: [0xef,0x7f,0x2f,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c12f7fef <unknown>
 
 fmls    za.h[w11, 7], {z31.h - z0.h}, z15.h  // 11000001-00101111-01111111-11101111
 // CHECK-INST: fmls    za.h[w11, 7, vgx2], { z31.h, z0.h }, z15.h
 // CHECK-ENCODING: [0xef,0x7f,0x2f,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c12f7fef <unknown>
 
 fmls    za.h[w8, 5, vgx2], {z17.h, z18.h}, z0.h  // 11000001-00100000-00011110-00101101
 // CHECK-INST: fmls    za.h[w8, 5, vgx2], { z17.h, z18.h }, z0.h
 // CHECK-ENCODING: [0x2d,0x1e,0x20,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1201e2d <unknown>
 
 fmls    za.h[w8, 5], {z17.h - z18.h}, z0.h  // 11000001-00100000-00011110-00101101
 // CHECK-INST: fmls    za.h[w8, 5, vgx2], { z17.h, z18.h }, z0.h
 // CHECK-ENCODING: [0x2d,0x1e,0x20,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1201e2d <unknown>
 
 fmls    za.h[w8, 1, vgx2], {z1.h, z2.h}, z14.h  // 11000001-00101110-00011100-00101001
 // CHECK-INST: fmls    za.h[w8, 1, vgx2], { z1.h, z2.h }, z14.h
 // CHECK-ENCODING: [0x29,0x1c,0x2e,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c12e1c29 <unknown>
 
 fmls    za.h[w8, 1], {z1.h - z2.h}, z14.h  // 11000001-00101110-00011100-00101001
 // CHECK-INST: fmls    za.h[w8, 1, vgx2], { z1.h, z2.h }, z14.h
 // CHECK-ENCODING: [0x29,0x1c,0x2e,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c12e1c29 <unknown>
 
 fmls    za.h[w10, 0, vgx2], {z19.h, z20.h}, z4.h  // 11000001-00100100-01011110-01101000
 // CHECK-INST: fmls    za.h[w10, 0, vgx2], { z19.h, z20.h }, z4.h
 // CHECK-ENCODING: [0x68,0x5e,0x24,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1245e68 <unknown>
 
 fmls    za.h[w10, 0], {z19.h - z20.h}, z4.h  // 11000001-00100100-01011110-01101000
 // CHECK-INST: fmls    za.h[w10, 0, vgx2], { z19.h, z20.h }, z4.h
 // CHECK-ENCODING: [0x68,0x5e,0x24,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1245e68 <unknown>
 
 fmls    za.h[w8, 0, vgx2], {z12.h, z13.h}, z2.h  // 11000001-00100010-00011101-10001000
 // CHECK-INST: fmls    za.h[w8, 0, vgx2], { z12.h, z13.h }, z2.h
 // CHECK-ENCODING: [0x88,0x1d,0x22,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1221d88 <unknown>
 
 fmls    za.h[w8, 0], {z12.h - z13.h}, z2.h  // 11000001-00100010-00011101-10001000
 // CHECK-INST: fmls    za.h[w8, 0, vgx2], { z12.h, z13.h }, z2.h
 // CHECK-ENCODING: [0x88,0x1d,0x22,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1221d88 <unknown>
 
 fmls    za.h[w10, 1, vgx2], {z1.h, z2.h}, z10.h  // 11000001-00101010-01011100-00101001
 // CHECK-INST: fmls    za.h[w10, 1, vgx2], { z1.h, z2.h }, z10.h
 // CHECK-ENCODING: [0x29,0x5c,0x2a,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c12a5c29 <unknown>
 
 fmls    za.h[w10, 1], {z1.h - z2.h}, z10.h  // 11000001-00101010-01011100-00101001
 // CHECK-INST: fmls    za.h[w10, 1, vgx2], { z1.h, z2.h }, z10.h
 // CHECK-ENCODING: [0x29,0x5c,0x2a,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c12a5c29 <unknown>
 
 fmls    za.h[w8, 5, vgx2], {z22.h, z23.h}, z14.h  // 11000001-00101110-00011110-11001101
 // CHECK-INST: fmls    za.h[w8, 5, vgx2], { z22.h, z23.h }, z14.h
 // CHECK-ENCODING: [0xcd,0x1e,0x2e,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c12e1ecd <unknown>
 
 fmls    za.h[w8, 5], {z22.h - z23.h}, z14.h  // 11000001-00101110-00011110-11001101
 // CHECK-INST: fmls    za.h[w8, 5, vgx2], { z22.h, z23.h }, z14.h
 // CHECK-ENCODING: [0xcd,0x1e,0x2e,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c12e1ecd <unknown>
 
 fmls    za.h[w11, 2, vgx2], {z9.h, z10.h}, z1.h  // 11000001-00100001-01111101-00101010
 // CHECK-INST: fmls    za.h[w11, 2, vgx2], { z9.h, z10.h }, z1.h
 // CHECK-ENCODING: [0x2a,0x7d,0x21,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1217d2a <unknown>
 
 fmls    za.h[w11, 2], {z9.h - z10.h}, z1.h  // 11000001-00100001-01111101-00101010
 // CHECK-INST: fmls    za.h[w11, 2, vgx2], { z9.h, z10.h }, z1.h
 // CHECK-ENCODING: [0x2a,0x7d,0x21,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1217d2a <unknown>
 
 fmls    za.h[w9, 7, vgx2], {z12.h, z13.h}, z11.h  // 11000001-00101011-00111101-10001111
 // CHECK-INST: fmls    za.h[w9, 7, vgx2], { z12.h, z13.h }, z11.h
 // CHECK-ENCODING: [0x8f,0x3d,0x2b,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c12b3d8f <unknown>
 
 fmls    za.h[w9, 7], {z12.h - z13.h}, z11.h  // 11000001-00101011-00111101-10001111
 // CHECK-INST: fmls    za.h[w9, 7, vgx2], { z12.h, z13.h }, z11.h
 // CHECK-ENCODING: [0x8f,0x3d,0x2b,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c12b3d8f <unknown>
 
 
 fmls    za.h[w8, 0, vgx2], {z0.h, z1.h}, z0.h[0]  // 11000001-00010000-00010000-00010000
 // CHECK-INST: fmls    za.h[w8, 0, vgx2], { z0.h, z1.h }, z0.h[0]
 // CHECK-ENCODING: [0x10,0x10,0x10,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1101010 <unknown>
 
 fmls    za.h[w8, 0], {z0.h - z1.h}, z0.h[0]  // 11000001-00010000-00010000-00010000
 // CHECK-INST: fmls    za.h[w8, 0, vgx2], { z0.h, z1.h }, z0.h[0]
 // CHECK-ENCODING: [0x10,0x10,0x10,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1101010 <unknown>
 
 fmls    za.h[w10, 5, vgx2], {z10.h, z11.h}, z5.h[2]  // 11000001-00010101-01010101-01010101
 // CHECK-INST: fmls    za.h[w10, 5, vgx2], { z10.h, z11.h }, z5.h[2]
 // CHECK-ENCODING: [0x55,0x55,0x15,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1155555 <unknown>
 
 fmls    za.h[w10, 5], {z10.h - z11.h}, z5.h[2]  // 11000001-00010101-01010101-01010101
 // CHECK-INST: fmls    za.h[w10, 5, vgx2], { z10.h, z11.h }, z5.h[2]
 // CHECK-ENCODING: [0x55,0x55,0x15,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1155555 <unknown>
 
 fmls    za.h[w11, 7, vgx2], {z12.h, z13.h}, z8.h[6]  // 11000001-00011000-01111101-10010111
 // CHECK-INST: fmls    za.h[w11, 7, vgx2], { z12.h, z13.h }, z8.h[6]
 // CHECK-ENCODING: [0x97,0x7d,0x18,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1187d97 <unknown>
 
 fmls    za.h[w11, 7], {z12.h - z13.h}, z8.h[6]  // 11000001-00011000-01111101-10010111
 // CHECK-INST: fmls    za.h[w11, 7, vgx2], { z12.h, z13.h }, z8.h[6]
 // CHECK-ENCODING: [0x97,0x7d,0x18,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1187d97 <unknown>
 
 fmls    za.h[w11, 7, vgx2], {z30.h, z31.h}, z15.h[7]  // 11000001-00011111-01111111-11011111
 // CHECK-INST: fmls    za.h[w11, 7, vgx2], { z30.h, z31.h }, z15.h[7]
 // CHECK-ENCODING: [0xdf,0x7f,0x1f,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c11f7fdf <unknown>
 
 fmls    za.h[w11, 7], {z30.h - z31.h}, z15.h[7]  // 11000001-00011111-01111111-11011111
 // CHECK-INST: fmls    za.h[w11, 7, vgx2], { z30.h, z31.h }, z15.h[7]
 // CHECK-ENCODING: [0xdf,0x7f,0x1f,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c11f7fdf <unknown>
 
 fmls    za.h[w8, 5, vgx2], {z16.h, z17.h}, z0.h[6]  // 11000001-00010000-00011110-00010101
 // CHECK-INST: fmls    za.h[w8, 5, vgx2], { z16.h, z17.h }, z0.h[6]
 // CHECK-ENCODING: [0x15,0x1e,0x10,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1101e15 <unknown>
 
 fmls    za.h[w8, 5], {z16.h - z17.h}, z0.h[6]  // 11000001-00010000-00011110-00010101
 // CHECK-INST: fmls    za.h[w8, 5, vgx2], { z16.h, z17.h }, z0.h[6]
 // CHECK-ENCODING: [0x15,0x1e,0x10,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1101e15 <unknown>
 
 fmls    za.h[w8, 1, vgx2], {z0.h, z1.h}, z14.h[2]  // 11000001-00011110-00010100-00010001
 // CHECK-INST: fmls    za.h[w8, 1, vgx2], { z0.h, z1.h }, z14.h[2]
 // CHECK-ENCODING: [0x11,0x14,0x1e,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c11e1411 <unknown>
 
 fmls    za.h[w8, 1], {z0.h - z1.h}, z14.h[2]  // 11000001-00011110-00010100-00010001
 // CHECK-INST: fmls    za.h[w8, 1, vgx2], { z0.h, z1.h }, z14.h[2]
 // CHECK-ENCODING: [0x11,0x14,0x1e,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c11e1411 <unknown>
 
 fmls    za.h[w10, 0, vgx2], {z18.h, z19.h}, z4.h[3]  // 11000001-00010100-01010110-01011000
 // CHECK-INST: fmls    za.h[w10, 0, vgx2], { z18.h, z19.h }, z4.h[3]
 // CHECK-ENCODING: [0x58,0x56,0x14,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1145658 <unknown>
 
 fmls    za.h[w10, 0], {z18.h - z19.h}, z4.h[3]  // 11000001-00010100-01010110-01011000
 // CHECK-INST: fmls    za.h[w10, 0, vgx2], { z18.h, z19.h }, z4.h[3]
 // CHECK-ENCODING: [0x58,0x56,0x14,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1145658 <unknown>
 
 fmls    za.h[w8, 0, vgx2], {z12.h, z13.h}, z2.h[4]  // 11000001-00010010-00011001-10010000
 // CHECK-INST: fmls    za.h[w8, 0, vgx2], { z12.h, z13.h }, z2.h[4]
 // CHECK-ENCODING: [0x90,0x19,0x12,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1121990 <unknown>
 
 fmls    za.h[w8, 0], {z12.h - z13.h}, z2.h[4]  // 11000001-00010010-00011001-10010000
 // CHECK-INST: fmls    za.h[w8, 0, vgx2], { z12.h, z13.h }, z2.h[4]
 // CHECK-ENCODING: [0x90,0x19,0x12,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1121990 <unknown>
 
 fmls    za.h[w10, 1, vgx2], {z0.h, z1.h}, z10.h[4]  // 11000001-00011010-01011000-00010001
 // CHECK-INST: fmls    za.h[w10, 1, vgx2], { z0.h, z1.h }, z10.h[4]
 // CHECK-ENCODING: [0x11,0x58,0x1a,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c11a5811 <unknown>
 
 fmls    za.h[w10, 1], {z0.h - z1.h}, z10.h[4]  // 11000001-00011010-01011000-00010001
 // CHECK-INST: fmls    za.h[w10, 1, vgx2], { z0.h, z1.h }, z10.h[4]
 // CHECK-ENCODING: [0x11,0x58,0x1a,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c11a5811 <unknown>
 
 fmls    za.h[w8, 5, vgx2], {z22.h, z23.h}, z14.h[5]  // 11000001-00011110-00011010-11011101
 // CHECK-INST: fmls    za.h[w8, 5, vgx2], { z22.h, z23.h }, z14.h[5]
 // CHECK-ENCODING: [0xdd,0x1a,0x1e,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c11e1add <unknown>
 
 fmls    za.h[w8, 5], {z22.h - z23.h}, z14.h[5]  // 11000001-00011110-00011010-11011101
 // CHECK-INST: fmls    za.h[w8, 5, vgx2], { z22.h, z23.h }, z14.h[5]
 // CHECK-ENCODING: [0xdd,0x1a,0x1e,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c11e1add <unknown>
 
 fmls    za.h[w11, 2, vgx2], {z8.h, z9.h}, z1.h[2]  // 11000001-00010001-01110101-00010010
 // CHECK-INST: fmls    za.h[w11, 2, vgx2], { z8.h, z9.h }, z1.h[2]
 // CHECK-ENCODING: [0x12,0x75,0x11,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1117512 <unknown>
 
 fmls    za.h[w11, 2], {z8.h - z9.h}, z1.h[2]  // 11000001-00010001-01110101-00010010
 // CHECK-INST: fmls    za.h[w11, 2, vgx2], { z8.h, z9.h }, z1.h[2]
 // CHECK-ENCODING: [0x12,0x75,0x11,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1117512 <unknown>
 
 fmls    za.h[w9, 7, vgx2], {z12.h, z13.h}, z11.h[4]  // 11000001-00011011-00111001-10010111
 // CHECK-INST: fmls    za.h[w9, 7, vgx2], { z12.h, z13.h }, z11.h[4]
 // CHECK-ENCODING: [0x97,0x39,0x1b,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c11b3997 <unknown>
 
 fmls    za.h[w9, 7], {z12.h - z13.h}, z11.h[4]  // 11000001-00011011-00111001-10010111
 // CHECK-INST: fmls    za.h[w9, 7, vgx2], { z12.h, z13.h }, z11.h[4]
 // CHECK-ENCODING: [0x97,0x39,0x1b,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c11b3997 <unknown>
 
 
 fmls    za.h[w8, 0, vgx2], {z0.h, z1.h}, {z0.h, z1.h}  // 11000001-10100000-00010000-00011000
 // CHECK-INST: fmls    za.h[w8, 0, vgx2], { z0.h, z1.h }, { z0.h, z1.h }
 // CHECK-ENCODING: [0x18,0x10,0xa0,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1a01018 <unknown>
 
 fmls    za.h[w8, 0], {z0.h - z1.h}, {z0.h - z1.h}  // 11000001-10100000-00010000-00011000
 // CHECK-INST: fmls    za.h[w8, 0, vgx2], { z0.h, z1.h }, { z0.h, z1.h }
 // CHECK-ENCODING: [0x18,0x10,0xa0,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1a01018 <unknown>
 
 fmls    za.h[w10, 5, vgx2], {z10.h, z11.h}, {z20.h, z21.h}  // 11000001-10110100-01010001-01011101
 // CHECK-INST: fmls    za.h[w10, 5, vgx2], { z10.h, z11.h }, { z20.h, z21.h }
 // CHECK-ENCODING: [0x5d,0x51,0xb4,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1b4515d <unknown>
 
 fmls    za.h[w10, 5], {z10.h - z11.h}, {z20.h - z21.h}  // 11000001-10110100-01010001-01011101
 // CHECK-INST: fmls    za.h[w10, 5, vgx2], { z10.h, z11.h }, { z20.h, z21.h }
 // CHECK-ENCODING: [0x5d,0x51,0xb4,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1b4515d <unknown>
 
 fmls    za.h[w11, 7, vgx2], {z12.h, z13.h}, {z8.h, z9.h}  // 11000001-10101000-01110001-10011111
 // CHECK-INST: fmls    za.h[w11, 7, vgx2], { z12.h, z13.h }, { z8.h, z9.h }
 // CHECK-ENCODING: [0x9f,0x71,0xa8,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1a8719f <unknown>
 
 fmls    za.h[w11, 7], {z12.h - z13.h}, {z8.h - z9.h}  // 11000001-10101000-01110001-10011111
 // CHECK-INST: fmls    za.h[w11, 7, vgx2], { z12.h, z13.h }, { z8.h, z9.h }
 // CHECK-ENCODING: [0x9f,0x71,0xa8,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1a8719f <unknown>
 
 fmls    za.h[w11, 7, vgx2], {z30.h, z31.h}, {z30.h, z31.h}  // 11000001-10111110-01110011-11011111
 // CHECK-INST: fmls    za.h[w11, 7, vgx2], { z30.h, z31.h }, { z30.h, z31.h }
 // CHECK-ENCODING: [0xdf,0x73,0xbe,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1be73df <unknown>
 
 fmls    za.h[w11, 7], {z30.h - z31.h}, {z30.h - z31.h}  // 11000001-10111110-01110011-11011111
 // CHECK-INST: fmls    za.h[w11, 7, vgx2], { z30.h, z31.h }, { z30.h, z31.h }
 // CHECK-ENCODING: [0xdf,0x73,0xbe,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1be73df <unknown>
 
 fmls    za.h[w8, 5, vgx2], {z16.h, z17.h}, {z16.h, z17.h}  // 11000001-10110000-00010010-00011101
 // CHECK-INST: fmls    za.h[w8, 5, vgx2], { z16.h, z17.h }, { z16.h, z17.h }
 // CHECK-ENCODING: [0x1d,0x12,0xb0,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1b0121d <unknown>
 
 fmls    za.h[w8, 5], {z16.h - z17.h}, {z16.h - z17.h}  // 11000001-10110000-00010010-00011101
 // CHECK-INST: fmls    za.h[w8, 5, vgx2], { z16.h, z17.h }, { z16.h, z17.h }
 // CHECK-ENCODING: [0x1d,0x12,0xb0,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1b0121d <unknown>
 
 fmls    za.h[w8, 1, vgx2], {z0.h, z1.h}, {z30.h, z31.h}  // 11000001-10111110-00010000-00011001
 // CHECK-INST: fmls    za.h[w8, 1, vgx2], { z0.h, z1.h }, { z30.h, z31.h }
 // CHECK-ENCODING: [0x19,0x10,0xbe,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1be1019 <unknown>
 
 fmls    za.h[w8, 1], {z0.h - z1.h}, {z30.h - z31.h}  // 11000001-10111110-00010000-00011001
 // CHECK-INST: fmls    za.h[w8, 1, vgx2], { z0.h, z1.h }, { z30.h, z31.h }
 // CHECK-ENCODING: [0x19,0x10,0xbe,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1be1019 <unknown>
 
 fmls    za.h[w10, 0, vgx2], {z18.h, z19.h}, {z20.h, z21.h}  // 11000001-10110100-01010010-01011000
 // CHECK-INST: fmls    za.h[w10, 0, vgx2], { z18.h, z19.h }, { z20.h, z21.h }
 // CHECK-ENCODING: [0x58,0x52,0xb4,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1b45258 <unknown>
 
 fmls    za.h[w10, 0], {z18.h - z19.h}, {z20.h - z21.h}  // 11000001-10110100-01010010-01011000
 // CHECK-INST: fmls    za.h[w10, 0, vgx2], { z18.h, z19.h }, { z20.h, z21.h }
 // CHECK-ENCODING: [0x58,0x52,0xb4,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1b45258 <unknown>
 
 fmls    za.h[w8, 0, vgx2], {z12.h, z13.h}, {z2.h, z3.h}  // 11000001-10100010-00010001-10011000
 // CHECK-INST: fmls    za.h[w8, 0, vgx2], { z12.h, z13.h }, { z2.h, z3.h }
 // CHECK-ENCODING: [0x98,0x11,0xa2,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1a21198 <unknown>
 
 fmls    za.h[w8, 0], {z12.h - z13.h}, {z2.h - z3.h}  // 11000001-10100010-00010001-10011000
 // CHECK-INST: fmls    za.h[w8, 0, vgx2], { z12.h, z13.h }, { z2.h, z3.h }
 // CHECK-ENCODING: [0x98,0x11,0xa2,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1a21198 <unknown>
 
 fmls    za.h[w10, 1, vgx2], {z0.h, z1.h}, {z26.h, z27.h}  // 11000001-10111010-01010000-00011001
 // CHECK-INST: fmls    za.h[w10, 1, vgx2], { z0.h, z1.h }, { z26.h, z27.h }
 // CHECK-ENCODING: [0x19,0x50,0xba,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1ba5019 <unknown>
 
 fmls    za.h[w10, 1], {z0.h - z1.h}, {z26.h - z27.h}  // 11000001-10111010-01010000-00011001
 // CHECK-INST: fmls    za.h[w10, 1, vgx2], { z0.h, z1.h }, { z26.h, z27.h }
 // CHECK-ENCODING: [0x19,0x50,0xba,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1ba5019 <unknown>
 
 fmls    za.h[w8, 5, vgx2], {z22.h, z23.h}, {z30.h, z31.h}  // 11000001-10111110-00010010-11011101
 // CHECK-INST: fmls    za.h[w8, 5, vgx2], { z22.h, z23.h }, { z30.h, z31.h }
 // CHECK-ENCODING: [0xdd,0x12,0xbe,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1be12dd <unknown>
 
 fmls    za.h[w8, 5], {z22.h - z23.h}, {z30.h - z31.h}  // 11000001-10111110-00010010-11011101
 // CHECK-INST: fmls    za.h[w8, 5, vgx2], { z22.h, z23.h }, { z30.h, z31.h }
 // CHECK-ENCODING: [0xdd,0x12,0xbe,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1be12dd <unknown>
 
 fmls    za.h[w11, 2, vgx2], {z8.h, z9.h}, {z0.h, z1.h}  // 11000001-10100000-01110001-00011010
 // CHECK-INST: fmls    za.h[w11, 2, vgx2], { z8.h, z9.h }, { z0.h, z1.h }
 // CHECK-ENCODING: [0x1a,0x71,0xa0,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1a0711a <unknown>
 
 fmls    za.h[w11, 2], {z8.h - z9.h}, {z0.h - z1.h}  // 11000001-10100000-01110001-00011010
 // CHECK-INST: fmls    za.h[w11, 2, vgx2], { z8.h, z9.h }, { z0.h, z1.h }
 // CHECK-ENCODING: [0x1a,0x71,0xa0,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1a0711a <unknown>
 
 fmls    za.h[w9, 7, vgx2], {z12.h, z13.h}, {z10.h, z11.h}  // 11000001-10101010-00110001-10011111
 // CHECK-INST: fmls    za.h[w9, 7, vgx2], { z12.h, z13.h }, { z10.h, z11.h }
 // CHECK-ENCODING: [0x9f,0x31,0xaa,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1aa319f <unknown>
 
 fmls    za.h[w9, 7], {z12.h - z13.h}, {z10.h - z11.h}  // 11000001-10101010-00110001-10011111
 // CHECK-INST: fmls    za.h[w9, 7, vgx2], { z12.h, z13.h }, { z10.h, z11.h }
 // CHECK-ENCODING: [0x9f,0x31,0xaa,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1aa319f <unknown>
 
 fmls    za.h[w8, 0, vgx4], {z0.h - z3.h}, z0.h  // 11000001-00110000-00011100-00001000
 // CHECK-INST: fmls    za.h[w8, 0, vgx4], { z0.h - z3.h }, z0.h
 // CHECK-ENCODING: [0x08,0x1c,0x30,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1301c08 <unknown>
 
 fmls    za.h[w8, 0], {z0.h - z3.h}, z0.h  // 11000001-00110000-00011100-00001000
 // CHECK-INST: fmls    za.h[w8, 0, vgx4], { z0.h - z3.h }, z0.h
 // CHECK-ENCODING: [0x08,0x1c,0x30,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1301c08 <unknown>
 
 fmls    za.h[w10, 5, vgx4], {z10.h - z13.h}, z5.h  // 11000001-00110101-01011101-01001101
 // CHECK-INST: fmls    za.h[w10, 5, vgx4], { z10.h - z13.h }, z5.h
 // CHECK-ENCODING: [0x4d,0x5d,0x35,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1355d4d <unknown>
 
 fmls    za.h[w10, 5], {z10.h - z13.h}, z5.h  // 11000001-00110101-01011101-01001101
 // CHECK-INST: fmls    za.h[w10, 5, vgx4], { z10.h - z13.h }, z5.h
 // CHECK-ENCODING: [0x4d,0x5d,0x35,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1355d4d <unknown>
 
 fmls    za.h[w11, 7, vgx4], {z13.h - z16.h}, z8.h  // 11000001-00111000-01111101-10101111
 // CHECK-INST: fmls    za.h[w11, 7, vgx4], { z13.h - z16.h }, z8.h
 // CHECK-ENCODING: [0xaf,0x7d,0x38,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1387daf <unknown>
 
 fmls    za.h[w11, 7], {z13.h - z16.h}, z8.h  // 11000001-00111000-01111101-10101111
 // CHECK-INST: fmls    za.h[w11, 7, vgx4], { z13.h - z16.h }, z8.h
 // CHECK-ENCODING: [0xaf,0x7d,0x38,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1387daf <unknown>
 
 fmls    za.h[w11, 7, vgx4], {z31.h, z0.h, z1.h, z2.h}, z15.h  // 11000001-00111111-01111111-11101111
 // CHECK-INST: fmls    za.h[w11, 7, vgx4], { z31.h, z0.h, z1.h, z2.h }, z15.h
 // CHECK-ENCODING: [0xef,0x7f,0x3f,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c13f7fef <unknown>
 
 fmls    za.h[w11, 7], {z31.h, z0.h, z1.h, z2.h}, z15.h  // 11000001-00111111-01111111-11101111
 // CHECK-INST: fmls    za.h[w11, 7, vgx4], { z31.h, z0.h, z1.h, z2.h }, z15.h
 // CHECK-ENCODING: [0xef,0x7f,0x3f,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c13f7fef <unknown>
 
 fmls    za.h[w8, 5, vgx4], {z17.h - z20.h}, z0.h  // 11000001-00110000-00011110-00101101
 // CHECK-INST: fmls    za.h[w8, 5, vgx4], { z17.h - z20.h }, z0.h
 // CHECK-ENCODING: [0x2d,0x1e,0x30,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1301e2d <unknown>
 
 fmls    za.h[w8, 5], {z17.h - z20.h}, z0.h  // 11000001-00110000-00011110-00101101
 // CHECK-INST: fmls    za.h[w8, 5, vgx4], { z17.h - z20.h }, z0.h
 // CHECK-ENCODING: [0x2d,0x1e,0x30,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1301e2d <unknown>
 
 fmls    za.h[w8, 1, vgx4], {z1.h - z4.h}, z14.h  // 11000001-00111110-00011100-00101001
 // CHECK-INST: fmls    za.h[w8, 1, vgx4], { z1.h - z4.h }, z14.h
 // CHECK-ENCODING: [0x29,0x1c,0x3e,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c13e1c29 <unknown>
 
 fmls    za.h[w8, 1], {z1.h - z4.h}, z14.h  // 11000001-00111110-00011100-00101001
 // CHECK-INST: fmls    za.h[w8, 1, vgx4], { z1.h - z4.h }, z14.h
 // CHECK-ENCODING: [0x29,0x1c,0x3e,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c13e1c29 <unknown>
 
 fmls    za.h[w10, 0, vgx4], {z19.h - z22.h}, z4.h  // 11000001-00110100-01011110-01101000
 // CHECK-INST: fmls    za.h[w10, 0, vgx4], { z19.h - z22.h }, z4.h
 // CHECK-ENCODING: [0x68,0x5e,0x34,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1345e68 <unknown>
 
 fmls    za.h[w10, 0], {z19.h - z22.h}, z4.h  // 11000001-00110100-01011110-01101000
 // CHECK-INST: fmls    za.h[w10, 0, vgx4], { z19.h - z22.h }, z4.h
 // CHECK-ENCODING: [0x68,0x5e,0x34,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1345e68 <unknown>
 
 fmls    za.h[w8, 0, vgx4], {z12.h - z15.h}, z2.h  // 11000001-00110010-00011101-10001000
 // CHECK-INST: fmls    za.h[w8, 0, vgx4], { z12.h - z15.h }, z2.h
 // CHECK-ENCODING: [0x88,0x1d,0x32,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1321d88 <unknown>
 
 fmls    za.h[w8, 0], {z12.h - z15.h}, z2.h  // 11000001-00110010-00011101-10001000
 // CHECK-INST: fmls    za.h[w8, 0, vgx4], { z12.h - z15.h }, z2.h
 // CHECK-ENCODING: [0x88,0x1d,0x32,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1321d88 <unknown>
 
 fmls    za.h[w10, 1, vgx4], {z1.h - z4.h}, z10.h  // 11000001-00111010-01011100-00101001
 // CHECK-INST: fmls    za.h[w10, 1, vgx4], { z1.h - z4.h }, z10.h
 // CHECK-ENCODING: [0x29,0x5c,0x3a,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c13a5c29 <unknown>
 
 fmls    za.h[w10, 1], {z1.h - z4.h}, z10.h  // 11000001-00111010-01011100-00101001
 // CHECK-INST: fmls    za.h[w10, 1, vgx4], { z1.h - z4.h }, z10.h
 // CHECK-ENCODING: [0x29,0x5c,0x3a,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c13a5c29 <unknown>
 
 fmls    za.h[w8, 5, vgx4], {z22.h - z25.h}, z14.h  // 11000001-00111110-00011110-11001101
 // CHECK-INST: fmls    za.h[w8, 5, vgx4], { z22.h - z25.h }, z14.h
 // CHECK-ENCODING: [0xcd,0x1e,0x3e,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c13e1ecd <unknown>
 
 fmls    za.h[w8, 5], {z22.h - z25.h}, z14.h  // 11000001-00111110-00011110-11001101
 // CHECK-INST: fmls    za.h[w8, 5, vgx4], { z22.h - z25.h }, z14.h
 // CHECK-ENCODING: [0xcd,0x1e,0x3e,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c13e1ecd <unknown>
 
 fmls    za.h[w11, 2, vgx4], {z9.h - z12.h}, z1.h  // 11000001-00110001-01111101-00101010
 // CHECK-INST: fmls    za.h[w11, 2, vgx4], { z9.h - z12.h }, z1.h
 // CHECK-ENCODING: [0x2a,0x7d,0x31,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1317d2a <unknown>
 
 fmls    za.h[w11, 2], {z9.h - z12.h}, z1.h  // 11000001-00110001-01111101-00101010
 // CHECK-INST: fmls    za.h[w11, 2, vgx4], { z9.h - z12.h }, z1.h
 // CHECK-ENCODING: [0x2a,0x7d,0x31,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1317d2a <unknown>
 
 fmls    za.h[w9, 7, vgx4], {z12.h - z15.h}, z11.h  // 11000001-00111011-00111101-10001111
 // CHECK-INST: fmls    za.h[w9, 7, vgx4], { z12.h - z15.h }, z11.h
 // CHECK-ENCODING: [0x8f,0x3d,0x3b,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c13b3d8f <unknown>
 
 fmls    za.h[w9, 7], {z12.h - z15.h}, z11.h  // 11000001-00111011-00111101-10001111
 // CHECK-INST: fmls    za.h[w9, 7, vgx4], { z12.h - z15.h }, z11.h
 // CHECK-ENCODING: [0x8f,0x3d,0x3b,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c13b3d8f <unknown>
 
 fmls    za.h[w8, 0, vgx4], {z0.h - z3.h}, z0.h[0]  // 11000001-00010000-10010000-00010000
 // CHECK-INST: fmls    za.h[w8, 0, vgx4], { z0.h - z3.h }, z0.h[0]
 // CHECK-ENCODING: [0x10,0x90,0x10,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1109010 <unknown>
 
 fmls    za.h[w8, 0], {z0.h - z3.h}, z0.h[0]  // 11000001-00010000-10010000-00010000
 // CHECK-INST: fmls    za.h[w8, 0, vgx4], { z0.h - z3.h }, z0.h[0]
 // CHECK-ENCODING: [0x10,0x90,0x10,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1109010 <unknown>
 
 fmls    za.h[w10, 5, vgx4], {z8.h - z11.h}, z5.h[2]  // 11000001-00010101-11010101-00010101
 // CHECK-INST: fmls    za.h[w10, 5, vgx4], { z8.h - z11.h }, z5.h[2]
 // CHECK-ENCODING: [0x15,0xd5,0x15,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c115d515 <unknown>
 
 fmls    za.h[w10, 5], {z8.h - z11.h}, z5.h[2]  // 11000001-00010101-11010101-00010101
 // CHECK-INST: fmls    za.h[w10, 5, vgx4], { z8.h - z11.h }, z5.h[2]
 // CHECK-ENCODING: [0x15,0xd5,0x15,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c115d515 <unknown>
 
 fmls    za.h[w11, 7, vgx4], {z12.h - z15.h}, z8.h[6]  // 11000001-00011000-11111101-10010111
 // CHECK-INST: fmls    za.h[w11, 7, vgx4], { z12.h - z15.h }, z8.h[6]
 // CHECK-ENCODING: [0x97,0xfd,0x18,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c118fd97 <unknown>
 
 fmls    za.h[w11, 7], {z12.h - z15.h}, z8.h[6]  // 11000001-00011000-11111101-10010111
 // CHECK-INST: fmls    za.h[w11, 7, vgx4], { z12.h - z15.h }, z8.h[6]
 // CHECK-ENCODING: [0x97,0xfd,0x18,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c118fd97 <unknown>
 
 fmls    za.h[w11, 7, vgx4], {z28.h - z31.h}, z15.h[7]  // 11000001-00011111-11111111-10011111
 // CHECK-INST: fmls    za.h[w11, 7, vgx4], { z28.h - z31.h }, z15.h[7]
 // CHECK-ENCODING: [0x9f,0xff,0x1f,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c11fff9f <unknown>
 
 fmls    za.h[w11, 7], {z28.h - z31.h}, z15.h[7]  // 11000001-00011111-11111111-10011111
 // CHECK-INST: fmls    za.h[w11, 7, vgx4], { z28.h - z31.h }, z15.h[7]
 // CHECK-ENCODING: [0x9f,0xff,0x1f,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c11fff9f <unknown>
 
 fmls    za.h[w8, 5, vgx4], {z16.h - z19.h}, z0.h[6]  // 11000001-00010000-10011110-00010101
 // CHECK-INST: fmls    za.h[w8, 5, vgx4], { z16.h - z19.h }, z0.h[6]
 // CHECK-ENCODING: [0x15,0x9e,0x10,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1109e15 <unknown>
 
 fmls    za.h[w8, 5], {z16.h - z19.h}, z0.h[6]  // 11000001-00010000-10011110-00010101
 // CHECK-INST: fmls    za.h[w8, 5, vgx4], { z16.h - z19.h }, z0.h[6]
 // CHECK-ENCODING: [0x15,0x9e,0x10,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1109e15 <unknown>
 
 fmls    za.h[w8, 1, vgx4], {z0.h - z3.h}, z14.h[2]  // 11000001-00011110-10010100-00010001
 // CHECK-INST: fmls    za.h[w8, 1, vgx4], { z0.h - z3.h }, z14.h[2]
 // CHECK-ENCODING: [0x11,0x94,0x1e,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c11e9411 <unknown>
 
 fmls    za.h[w8, 1], {z0.h - z3.h}, z14.h[2]  // 11000001-00011110-10010100-00010001
 // CHECK-INST: fmls    za.h[w8, 1, vgx4], { z0.h - z3.h }, z14.h[2]
 // CHECK-ENCODING: [0x11,0x94,0x1e,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c11e9411 <unknown>
 
 fmls    za.h[w10, 0, vgx4], {z16.h - z19.h}, z4.h[3]  // 11000001-00010100-11010110-00011000
 // CHECK-INST: fmls    za.h[w10, 0, vgx4], { z16.h - z19.h }, z4.h[3]
 // CHECK-ENCODING: [0x18,0xd6,0x14,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c114d618 <unknown>
 
 fmls    za.h[w10, 0], {z16.h - z19.h}, z4.h[3]  // 11000001-00010100-11010110-00011000
 // CHECK-INST: fmls    za.h[w10, 0, vgx4], { z16.h - z19.h }, z4.h[3]
 // CHECK-ENCODING: [0x18,0xd6,0x14,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c114d618 <unknown>
 
 fmls    za.h[w8, 0, vgx4], {z12.h - z15.h}, z2.h[4]  // 11000001-00010010-10011001-10010000
 // CHECK-INST: fmls    za.h[w8, 0, vgx4], { z12.h - z15.h }, z2.h[4]
 // CHECK-ENCODING: [0x90,0x99,0x12,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1129990 <unknown>
 
 fmls    za.h[w8, 0], {z12.h - z15.h}, z2.h[4]  // 11000001-00010010-10011001-10010000
 // CHECK-INST: fmls    za.h[w8, 0, vgx4], { z12.h - z15.h }, z2.h[4]
 // CHECK-ENCODING: [0x90,0x99,0x12,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1129990 <unknown>
 
 fmls    za.h[w10, 1, vgx4], {z0.h - z3.h}, z10.h[4]  // 11000001-00011010-11011000-00010001
 // CHECK-INST: fmls    za.h[w10, 1, vgx4], { z0.h - z3.h }, z10.h[4]
 // CHECK-ENCODING: [0x11,0xd8,0x1a,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c11ad811 <unknown>
 
 fmls    za.h[w10, 1], {z0.h - z3.h}, z10.h[4]  // 11000001-00011010-11011000-00010001
 // CHECK-INST: fmls    za.h[w10, 1, vgx4], { z0.h - z3.h }, z10.h[4]
 // CHECK-ENCODING: [0x11,0xd8,0x1a,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c11ad811 <unknown>
 
 fmls    za.h[w8, 5, vgx4], {z20.h - z23.h}, z14.h[5]  // 11000001-00011110-10011010-10011101
 // CHECK-INST: fmls    za.h[w8, 5, vgx4], { z20.h - z23.h }, z14.h[5]
 // CHECK-ENCODING: [0x9d,0x9a,0x1e,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c11e9a9d <unknown>
 
 fmls    za.h[w8, 5], {z20.h - z23.h}, z14.h[5]  // 11000001-00011110-10011010-10011101
 // CHECK-INST: fmls    za.h[w8, 5, vgx4], { z20.h - z23.h }, z14.h[5]
 // CHECK-ENCODING: [0x9d,0x9a,0x1e,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c11e9a9d <unknown>
 
 fmls    za.h[w11, 2, vgx4], {z8.h - z11.h}, z1.h[2]  // 11000001-00010001-11110101-00010010
 // CHECK-INST: fmls    za.h[w11, 2, vgx4], { z8.h - z11.h }, z1.h[2]
 // CHECK-ENCODING: [0x12,0xf5,0x11,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c111f512 <unknown>
 
 fmls    za.h[w11, 2], {z8.h - z11.h}, z1.h[2]  // 11000001-00010001-11110101-00010010
 // CHECK-INST: fmls    za.h[w11, 2, vgx4], { z8.h - z11.h }, z1.h[2]
 // CHECK-ENCODING: [0x12,0xf5,0x11,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c111f512 <unknown>
 
 fmls    za.h[w9, 7, vgx4], {z12.h - z15.h}, z11.h[4]  // 11000001-00011011-10111001-10010111
 // CHECK-INST: fmls    za.h[w9, 7, vgx4], { z12.h - z15.h }, z11.h[4]
 // CHECK-ENCODING: [0x97,0xb9,0x1b,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c11bb997 <unknown>
 
 fmls    za.h[w9, 7], {z12.h - z15.h}, z11.h[4]  // 11000001-00011011-10111001-10010111
 // CHECK-INST: fmls    za.h[w9, 7, vgx4], { z12.h - z15.h }, z11.h[4]
 // CHECK-ENCODING: [0x97,0xb9,0x1b,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c11bb997 <unknown>
 
 fmls    za.h[w8, 0, vgx4], {z0.h - z3.h}, {z0.h - z3.h}  // 11000001-10100001-00010000-00011000
 // CHECK-INST: fmls    za.h[w8, 0, vgx4], { z0.h - z3.h }, { z0.h - z3.h }
 // CHECK-ENCODING: [0x18,0x10,0xa1,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1a11018 <unknown>
 
 fmls    za.h[w8, 0], {z0.h - z3.h}, {z0.h - z3.h}  // 11000001-10100001-00010000-00011000
 // CHECK-INST: fmls    za.h[w8, 0, vgx4], { z0.h - z3.h }, { z0.h - z3.h }
 // CHECK-ENCODING: [0x18,0x10,0xa1,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1a11018 <unknown>
 
 fmls    za.h[w10, 5, vgx4], {z8.h - z11.h}, {z20.h - z23.h}  // 11000001-10110101-01010001-00011101
 // CHECK-INST: fmls    za.h[w10, 5, vgx4], { z8.h - z11.h }, { z20.h - z23.h }
 // CHECK-ENCODING: [0x1d,0x51,0xb5,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1b5511d <unknown>
 
 fmls    za.h[w10, 5], {z8.h - z11.h}, {z20.h - z23.h}  // 11000001-10110101-01010001-00011101
 // CHECK-INST: fmls    za.h[w10, 5, vgx4], { z8.h - z11.h }, { z20.h - z23.h }
 // CHECK-ENCODING: [0x1d,0x51,0xb5,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1b5511d <unknown>
 
 fmls    za.h[w11, 7, vgx4], {z12.h - z15.h}, {z8.h - z11.h}  // 11000001-10101001-01110001-10011111
 // CHECK-INST: fmls    za.h[w11, 7, vgx4], { z12.h - z15.h }, { z8.h - z11.h }
 // CHECK-ENCODING: [0x9f,0x71,0xa9,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1a9719f <unknown>
 
 fmls    za.h[w11, 7], {z12.h - z15.h}, {z8.h - z11.h}  // 11000001-10101001-01110001-10011111
 // CHECK-INST: fmls    za.h[w11, 7, vgx4], { z12.h - z15.h }, { z8.h - z11.h }
 // CHECK-ENCODING: [0x9f,0x71,0xa9,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1a9719f <unknown>
 
 fmls    za.h[w11, 7, vgx4], {z28.h - z31.h}, {z28.h - z31.h}  // 11000001-10111101-01110011-10011111
 // CHECK-INST: fmls    za.h[w11, 7, vgx4], { z28.h - z31.h }, { z28.h - z31.h }
 // CHECK-ENCODING: [0x9f,0x73,0xbd,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1bd739f <unknown>
 
 fmls    za.h[w11, 7], {z28.h - z31.h}, {z28.h - z31.h}  // 11000001-10111101-01110011-10011111
 // CHECK-INST: fmls    za.h[w11, 7, vgx4], { z28.h - z31.h }, { z28.h - z31.h }
 // CHECK-ENCODING: [0x9f,0x73,0xbd,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1bd739f <unknown>
 
 fmls    za.h[w8, 5, vgx4], {z16.h - z19.h}, {z16.h - z19.h}  // 11000001-10110001-00010010-00011101
 // CHECK-INST: fmls    za.h[w8, 5, vgx4], { z16.h - z19.h }, { z16.h - z19.h }
 // CHECK-ENCODING: [0x1d,0x12,0xb1,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1b1121d <unknown>
 
 fmls    za.h[w8, 5], {z16.h - z19.h}, {z16.h - z19.h}  // 11000001-10110001-00010010-00011101
 // CHECK-INST: fmls    za.h[w8, 5, vgx4], { z16.h - z19.h }, { z16.h - z19.h }
 // CHECK-ENCODING: [0x1d,0x12,0xb1,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1b1121d <unknown>
 
 fmls    za.h[w8, 1, vgx4], {z0.h - z3.h}, {z28.h - z31.h}  // 11000001-10111101-00010000-00011001
 // CHECK-INST: fmls    za.h[w8, 1, vgx4], { z0.h - z3.h }, { z28.h - z31.h }
 // CHECK-ENCODING: [0x19,0x10,0xbd,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1bd1019 <unknown>
 
 fmls    za.h[w8, 1], {z0.h - z3.h}, {z28.h - z31.h}  // 11000001-10111101-00010000-00011001
 // CHECK-INST: fmls    za.h[w8, 1, vgx4], { z0.h - z3.h }, { z28.h - z31.h }
 // CHECK-ENCODING: [0x19,0x10,0xbd,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1bd1019 <unknown>
 
 fmls    za.h[w10, 0, vgx4], {z16.h - z19.h}, {z20.h - z23.h}  // 11000001-10110101-01010010-00011000
 // CHECK-INST: fmls    za.h[w10, 0, vgx4], { z16.h - z19.h }, { z20.h - z23.h }
 // CHECK-ENCODING: [0x18,0x52,0xb5,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1b55218 <unknown>
 
 fmls    za.h[w10, 0], {z16.h - z19.h}, {z20.h - z23.h}  // 11000001-10110101-01010010-00011000
 // CHECK-INST: fmls    za.h[w10, 0, vgx4], { z16.h - z19.h }, { z20.h - z23.h }
 // CHECK-ENCODING: [0x18,0x52,0xb5,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1b55218 <unknown>
 
 fmls    za.h[w8, 0, vgx4], {z12.h - z15.h}, {z0.h - z3.h}  // 11000001-10100001-00010001-10011000
 // CHECK-INST: fmls    za.h[w8, 0, vgx4], { z12.h - z15.h }, { z0.h - z3.h }
 // CHECK-ENCODING: [0x98,0x11,0xa1,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1a11198 <unknown>
 
 fmls    za.h[w8, 0], {z12.h - z15.h}, {z0.h - z3.h}  // 11000001-10100001-00010001-10011000
 // CHECK-INST: fmls    za.h[w8, 0, vgx4], { z12.h - z15.h }, { z0.h - z3.h }
 // CHECK-ENCODING: [0x98,0x11,0xa1,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1a11198 <unknown>
 
 fmls    za.h[w10, 1, vgx4], {z0.h - z3.h}, {z24.h - z27.h}  // 11000001-10111001-01010000-00011001
 // CHECK-INST: fmls    za.h[w10, 1, vgx4], { z0.h - z3.h }, { z24.h - z27.h }
 // CHECK-ENCODING: [0x19,0x50,0xb9,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1b95019 <unknown>
 
 fmls    za.h[w10, 1], {z0.h - z3.h}, {z24.h - z27.h}  // 11000001-10111001-01010000-00011001
 // CHECK-INST: fmls    za.h[w10, 1, vgx4], { z0.h - z3.h }, { z24.h - z27.h }
 // CHECK-ENCODING: [0x19,0x50,0xb9,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1b95019 <unknown>
 
 fmls    za.h[w8, 5, vgx4], {z20.h - z23.h}, {z28.h - z31.h}  // 11000001-10111101-00010010-10011101
 // CHECK-INST: fmls    za.h[w8, 5, vgx4], { z20.h - z23.h }, { z28.h - z31.h }
 // CHECK-ENCODING: [0x9d,0x12,0xbd,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1bd129d <unknown>
 
 fmls    za.h[w8, 5], {z20.h - z23.h}, {z28.h - z31.h}  // 11000001-10111101-00010010-10011101
 // CHECK-INST: fmls    za.h[w8, 5, vgx4], { z20.h - z23.h }, { z28.h - z31.h }
 // CHECK-ENCODING: [0x9d,0x12,0xbd,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1bd129d <unknown>
 
 fmls    za.h[w11, 2, vgx4], {z8.h - z11.h}, {z0.h - z3.h}  // 11000001-10100001-01110001-00011010
 // CHECK-INST: fmls    za.h[w11, 2, vgx4], { z8.h - z11.h }, { z0.h - z3.h }
 // CHECK-ENCODING: [0x1a,0x71,0xa1,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1a1711a <unknown>
 
 fmls    za.h[w11, 2], {z8.h - z11.h}, {z0.h - z3.h}  // 11000001-10100001-01110001-00011010
 // CHECK-INST: fmls    za.h[w11, 2, vgx4], { z8.h - z11.h }, { z0.h - z3.h }
 // CHECK-ENCODING: [0x1a,0x71,0xa1,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1a1711a <unknown>
 
 fmls    za.h[w9, 7, vgx4], {z12.h - z15.h}, {z8.h - z11.h}  // 11000001-10101001-00110001-10011111
 // CHECK-INST: fmls    za.h[w9, 7, vgx4], { z12.h - z15.h }, { z8.h - z11.h }
 // CHECK-ENCODING: [0x9f,0x31,0xa9,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1a9319f <unknown>
 
 fmls    za.h[w9, 7], {z12.h - z15.h}, {z8.h - z11.h}  // 11000001-10101001-00110001-10011111
 // CHECK-INST: fmls    za.h[w9, 7, vgx4], { z12.h - z15.h }, { z8.h - z11.h }
 // CHECK-ENCODING: [0x9f,0x31,0xa9,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: c1a9319f <unknown>
diff --git a/llvm/test/MC/AArch64/SME2p1/fmopa-diagnostics.s b/llvm/test/MC/AArch64/SME2p1/fmopa-diagnostics.s
index def19a316c2a..1c561959c25e 100644
--- a/llvm/test/MC/AArch64/SME2p1/fmopa-diagnostics.s
+++ b/llvm/test/MC/AArch64/SME2p1/fmopa-diagnostics.s
@@ -1,4 +1,4 @@
-// RUN: not llvm-mc -triple=aarch64 -show-encoding -mattr=+sme2p1,+sme-f16f16 2>&1 < %s | FileCheck %s
+// RUN: not llvm-mc -triple=aarch64 -show-encoding -mattr=+sme-f16f16 2>&1 < %s | FileCheck %s
 
 // --------------------------------------------------------------------------//
 // Invalid predicate register
diff --git a/llvm/test/MC/AArch64/SME2p1/fmopa.s b/llvm/test/MC/AArch64/SME2p1/fmopa.s
index e53d21244fde..0a586d3acc42 100644
--- a/llvm/test/MC/AArch64/SME2p1/fmopa.s
+++ b/llvm/test/MC/AArch64/SME2p1/fmopa.s
@@ -1,85 +1,85 @@
-// RUN: llvm-mc -triple=aarch64 -show-encoding -mattr=+sme2p1,+sme-f16f16 < %s \
+// RUN: llvm-mc -triple=aarch64 -show-encoding -mattr=+sme-f16f16 < %s \
 // RUN:        | FileCheck %s --check-prefixes=CHECK-ENCODING,CHECK-INST
 // RUN: not llvm-mc -triple=aarch64 -show-encoding < %s 2>&1 \
 // RUN:        | FileCheck %s --check-prefix=CHECK-ERROR
-// RUN: llvm-mc -triple=aarch64 -filetype=obj -mattr=+sme2p1,+sme-f16f16 < %s \
-// RUN:        | llvm-objdump -d --mattr=+sme2p1,+sme-f16f16 - | FileCheck %s --check-prefix=CHECK-INST
-// RUN: llvm-mc -triple=aarch64 -filetype=obj -mattr=+sme2p1,+sme-f16f16 < %s \
+// RUN: llvm-mc -triple=aarch64 -filetype=obj -mattr=+sme-f16f16 < %s \
+// RUN:        | llvm-objdump -d --mattr=+sme-f16f16 - | FileCheck %s --check-prefix=CHECK-INST
+// RUN: llvm-mc -triple=aarch64 -filetype=obj -mattr=+sme-f16f16 < %s \
 // RUN:        | llvm-objdump -d  --mattr=-sme2p1 - | FileCheck %s --check-prefix=CHECK-UNKNOWN
-// RUN: llvm-mc -triple=aarch64 -show-encoding -mattr=+sme2p1,+sme-f16f16 < %s \
+// RUN: llvm-mc -triple=aarch64 -show-encoding -mattr=+sme-f16f16 < %s \
 // RUN:        | sed '/.text/d' | sed 's/.*encoding: //g' \
-// RUN:        | llvm-mc -triple=aarch64 -mattr=+sme2p1,+sme-f16f16 -disassemble -show-encoding \
+// RUN:        | llvm-mc -triple=aarch64 -mattr=+sme-f16f16 -disassemble -show-encoding \
 // RUN:        | FileCheck %s --check-prefixes=CHECK-ENCODING,CHECK-INST
 
 fmopa   za0.h, p0/m, p0/m, z0.h, z0.h  // 10000001-10000000-00000000-00001000
 // CHECK-INST: fmopa   za0.h, p0/m, p0/m, z0.h, z0.h
 // CHECK-ENCODING: [0x08,0x00,0x80,0x81]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: 81800008 <unknown>
 
 fmopa   za1.h, p5/m, p2/m, z10.h, z21.h  // 10000001-10010101-01010101-01001001
 // CHECK-INST: fmopa   za1.h, p5/m, p2/m, z10.h, z21.h
 // CHECK-ENCODING: [0x49,0x55,0x95,0x81]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: 81955549 <unknown>
 
 fmopa   za1.h, p3/m, p7/m, z13.h, z8.h  // 10000001-10001000-11101101-10101001
 // CHECK-INST: fmopa   za1.h, p3/m, p7/m, z13.h, z8.h
 // CHECK-ENCODING: [0xa9,0xed,0x88,0x81]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: 8188eda9 <unknown>
 
 fmopa   za1.h, p7/m, p7/m, z31.h, z31.h  // 10000001-10011111-11111111-11101001
 // CHECK-INST: fmopa   za1.h, p7/m, p7/m, z31.h, z31.h
 // CHECK-ENCODING: [0xe9,0xff,0x9f,0x81]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: 819fffe9 <unknown>
 
 fmopa   za1.h, p3/m, p0/m, z17.h, z16.h  // 10000001-10010000-00001110-00101001
 // CHECK-INST: fmopa   za1.h, p3/m, p0/m, z17.h, z16.h
 // CHECK-ENCODING: [0x29,0x0e,0x90,0x81]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: 81900e29 <unknown>
 
 fmopa   za1.h, p1/m, p4/m, z1.h, z30.h  // 10000001-10011110-10000100-00101001
 // CHECK-INST: fmopa   za1.h, p1/m, p4/m, z1.h, z30.h
 // CHECK-ENCODING: [0x29,0x84,0x9e,0x81]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: 819e8429 <unknown>
 
 fmopa   za0.h, p5/m, p2/m, z19.h, z20.h  // 10000001-10010100-01010110-01101000
 // CHECK-INST: fmopa   za0.h, p5/m, p2/m, z19.h, z20.h
 // CHECK-ENCODING: [0x68,0x56,0x94,0x81]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: 81945668 <unknown>
 
 fmopa   za0.h, p6/m, p0/m, z12.h, z2.h  // 10000001-10000010-00011001-10001000
 // CHECK-INST: fmopa   za0.h, p6/m, p0/m, z12.h, z2.h
 // CHECK-ENCODING: [0x88,0x19,0x82,0x81]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: 81821988 <unknown>
 
 fmopa   za1.h, p2/m, p6/m, z1.h, z26.h  // 10000001-10011010-11001000-00101001
 // CHECK-INST: fmopa   za1.h, p2/m, p6/m, z1.h, z26.h
 // CHECK-ENCODING: [0x29,0xc8,0x9a,0x81]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: 819ac829 <unknown>
 
 fmopa   za1.h, p2/m, p0/m, z22.h, z30.h  // 10000001-10011110-00001010-11001001
 // CHECK-INST: fmopa   za1.h, p2/m, p0/m, z22.h, z30.h
 // CHECK-ENCODING: [0xc9,0x0a,0x9e,0x81]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: 819e0ac9 <unknown>
 
 fmopa   za0.h, p5/m, p7/m, z9.h, z1.h  // 10000001-10000001-11110101-00101000
 // CHECK-INST: fmopa   za0.h, p5/m, p7/m, z9.h, z1.h
 // CHECK-ENCODING: [0x28,0xf5,0x81,0x81]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: 8181f528 <unknown>
 
 fmopa   za1.h, p2/m, p5/m, z12.h, z11.h  // 10000001-10001011-10101001-10001001
 // CHECK-INST: fmopa   za1.h, p2/m, p5/m, z12.h, z11.h
 // CHECK-ENCODING: [0x89,0xa9,0x8b,0x81]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: 818ba989 <unknown>
 
diff --git a/llvm/test/MC/AArch64/SME2p1/fmops-diagnostics.s b/llvm/test/MC/AArch64/SME2p1/fmops-diagnostics.s
index 75eea8113262..0ec227ae0e68 100644
--- a/llvm/test/MC/AArch64/SME2p1/fmops-diagnostics.s
+++ b/llvm/test/MC/AArch64/SME2p1/fmops-diagnostics.s
@@ -1,4 +1,4 @@
-// RUN: not llvm-mc -triple=aarch64 -show-encoding -mattr=+sme2p1,+sme-f16f16 2>&1 < %s | FileCheck %s
+// RUN: not llvm-mc -triple=aarch64 -show-encoding -mattr=+sme-f16f16 2>&1 < %s | FileCheck %s
 
 // --------------------------------------------------------------------------//
 // Invalid predicate register
diff --git a/llvm/test/MC/AArch64/SME2p1/fmops.s b/llvm/test/MC/AArch64/SME2p1/fmops.s
index 325d4c125b60..597665d59150 100644
--- a/llvm/test/MC/AArch64/SME2p1/fmops.s
+++ b/llvm/test/MC/AArch64/SME2p1/fmops.s
@@ -1,84 +1,84 @@
-// RUN: llvm-mc -triple=aarch64 -show-encoding -mattr=+sme2p1,+sme-f16f16 < %s \
+// RUN: llvm-mc -triple=aarch64 -show-encoding -mattr=+sme-f16f16 < %s \
 // RUN:        | FileCheck %s --check-prefixes=CHECK-ENCODING,CHECK-INST
 // RUN: not llvm-mc -triple=aarch64 -show-encoding < %s 2>&1 \
 // RUN:        | FileCheck %s --check-prefix=CHECK-ERROR
-// RUN: llvm-mc -triple=aarch64 -filetype=obj -mattr=+sme2p1,+sme-f16f16 < %s \
-// RUN:        | llvm-objdump -d --mattr=+sme2p1,+sme-f16f16 - | FileCheck %s --check-prefix=CHECK-INST
-// RUN: llvm-mc -triple=aarch64 -filetype=obj -mattr=+sme2p1,+sme-f16f16 < %s \
+// RUN: llvm-mc -triple=aarch64 -filetype=obj -mattr=+sme-f16f16 < %s \
+// RUN:        | llvm-objdump -d --mattr=+sme-f16f16 - | FileCheck %s --check-prefix=CHECK-INST
+// RUN: llvm-mc -triple=aarch64 -filetype=obj -mattr=+sme-f16f16 < %s \
 // RUN:        | llvm-objdump -d --mattr=-sme2p1 - | FileCheck %s --check-prefix=CHECK-UNKNOWN
-// RUN: llvm-mc -triple=aarch64 -show-encoding -mattr=+sme2p1,+sme-f16f16 < %s \
+// RUN: llvm-mc -triple=aarch64 -show-encoding -mattr=+sme-f16f16 < %s \
 // RUN:        | sed '/.text/d' | sed 's/.*encoding: //g' \
-// RUN:        | llvm-mc -triple=aarch64 -mattr=+sme2p1,+sme-f16f16 -disassemble -show-encoding \
+// RUN:        | llvm-mc -triple=aarch64 -mattr=+sme-f16f16 -disassemble -show-encoding \
 // RUN:        | FileCheck %s --check-prefixes=CHECK-ENCODING,CHECK-INST
 
 fmops   za0.h, p0/m, p0/m, z0.h, z0.h  // 10000001-10000000-00000000-00011000
 // CHECK-INST: fmops   za0.h, p0/m, p0/m, z0.h, z0.h
 // CHECK-ENCODING: [0x18,0x00,0x80,0x81]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: 81800018 <unknown>
 
 fmops   za1.h, p5/m, p2/m, z10.h, z21.h  // 10000001-10010101-01010101-01011001
 // CHECK-INST: fmops   za1.h, p5/m, p2/m, z10.h, z21.h
 // CHECK-ENCODING: [0x59,0x55,0x95,0x81]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: 81955559 <unknown>
 
 fmops   za1.h, p3/m, p7/m, z13.h, z8.h  // 10000001-10001000-11101101-10111001
 // CHECK-INST: fmops   za1.h, p3/m, p7/m, z13.h, z8.h
 // CHECK-ENCODING: [0xb9,0xed,0x88,0x81]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: 8188edb9 <unknown>
 
 fmops   za1.h, p7/m, p7/m, z31.h, z31.h  // 10000001-10011111-11111111-11111001
 // CHECK-INST: fmops   za1.h, p7/m, p7/m, z31.h, z31.h
 // CHECK-ENCODING: [0xf9,0xff,0x9f,0x81]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: 819ffff9 <unknown>
 
 fmops   za1.h, p3/m, p0/m, z17.h, z16.h  // 10000001-10010000-00001110-00111001
 // CHECK-INST: fmops   za1.h, p3/m, p0/m, z17.h, z16.h
 // CHECK-ENCODING: [0x39,0x0e,0x90,0x81]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: 81900e39 <unknown>
 
 fmops   za1.h, p1/m, p4/m, z1.h, z30.h  // 10000001-10011110-10000100-00111001
 // CHECK-INST: fmops   za1.h, p1/m, p4/m, z1.h, z30.h
 // CHECK-ENCODING: [0x39,0x84,0x9e,0x81]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: 819e8439 <unknown>
 
 fmops   za0.h, p5/m, p2/m, z19.h, z20.h  // 10000001-10010100-01010110-01111000
 // CHECK-INST: fmops   za0.h, p5/m, p2/m, z19.h, z20.h
 // CHECK-ENCODING: [0x78,0x56,0x94,0x81]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: 81945678 <unknown>
 
 fmops   za0.h, p6/m, p0/m, z12.h, z2.h  // 10000001-10000010-00011001-10011000
 // CHECK-INST: fmops   za0.h, p6/m, p0/m, z12.h, z2.h
 // CHECK-ENCODING: [0x98,0x19,0x82,0x81]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: 81821998 <unknown>
 
 fmops   za1.h, p2/m, p6/m, z1.h, z26.h  // 10000001-10011010-11001000-00111001
 // CHECK-INST: fmops   za1.h, p2/m, p6/m, z1.h, z26.h
 // CHECK-ENCODING: [0x39,0xc8,0x9a,0x81]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: 819ac839 <unknown>
 
 fmops   za1.h, p2/m, p0/m, z22.h, z30.h  // 10000001-10011110-00001010-11011001
 // CHECK-INST: fmops   za1.h, p2/m, p0/m, z22.h, z30.h
 // CHECK-ENCODING: [0xd9,0x0a,0x9e,0x81]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: 819e0ad9 <unknown>
 
 fmops   za0.h, p5/m, p7/m, z9.h, z1.h  // 10000001-10000001-11110101-00111000
 // CHECK-INST: fmops   za0.h, p5/m, p7/m, z9.h, z1.h
 // CHECK-ENCODING: [0x38,0xf5,0x81,0x81]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: 8181f538 <unknown>
 
 fmops   za1.h, p2/m, p5/m, z12.h, z11.h  // 10000001-10001011-10101001-10011001
 // CHECK-INST: fmops   za1.h, p2/m, p5/m, z12.h, z11.h
 // CHECK-ENCODING: [0x99,0xa9,0x8b,0x81]
-// CHECK-ERROR: instruction requires: sme2p1 sme-f16f16
+// CHECK-ERROR: instruction requires: sme-f16f16
 // CHECK-UNKNOWN: 818ba999 <unknown>
diff --git a/llvm/test/MC/AArch64/SME2p1/fsub-diagnostics.s b/llvm/test/MC/AArch64/SME2p1/fsub-diagnostics.s
index 716427a2f725..60cef4260be9 100644
--- a/llvm/test/MC/AArch64/SME2p1/fsub-diagnostics.s
+++ b/llvm/test/MC/AArch64/SME2p1/fsub-diagnostics.s
@@ -1,5 +1,4 @@
-// RUN: not llvm-mc -triple=aarch64 -show-encoding -mattr=+sme2p1,+sme-f16f16 2>&1 < %s | FileCheck %s
-
+// RUN: not llvm-mc -triple=aarch64 -show-encoding -mattr=+sme-f16f16 2>&1 < %s | FileCheck %s
 // --------------------------------------------------------------------------//
 // Out of range index offset
 
diff --git a/llvm/test/MC/AArch64/SME2p1/fsub.s b/llvm/test/MC/AArch64/SME2p1/fsub.s
index b3735d554765..66410008eb11 100644
--- a/llvm/test/MC/AArch64/SME2p1/fsub.s
+++ b/llvm/test/MC/AArch64/SME2p1/fsub.s
@@ -1,296 +1,298 @@
-// RUN: llvm-mc -triple=aarch64 -show-encoding -mattr=+sme2p1,+sme-f16f16 < %s \
+// RUN: llvm-mc -triple=aarch64 -show-encoding -mattr=+sme-f16f16 < %s \
+// RUN:        | FileCheck %s --check-prefixes=CHECK-ENCODING,CHECK-INST
+// RUN: llvm-mc -triple=aarch64 -show-encoding -mattr=+sme-f8f16 < %s \
 // RUN:        | FileCheck %s --check-prefixes=CHECK-ENCODING,CHECK-INST
 // RUN: not llvm-mc -triple=aarch64 -show-encoding < %s 2>&1 \
 // RUN:        | FileCheck %s --check-prefix=CHECK-ERROR
-// RUN: llvm-mc -triple=aarch64 -filetype=obj -mattr=+sme2p1,+sme-f16f16 < %s \
-// RUN:        | llvm-objdump -d --mattr=+sme2p1,+sme-f16f16 - | FileCheck %s --check-prefix=CHECK-INST
-// RUN: llvm-mc -triple=aarch64 -filetype=obj -mattr=+sme2p1,+sme-f16f16 < %s \
+// RUN: llvm-mc -triple=aarch64 -filetype=obj -mattr=+sme-f16f16 < %s \
+// RUN:        | llvm-objdump -d --mattr=+sme-f16f16 - | FileCheck %s --check-prefix=CHECK-INST
+// RUN: llvm-mc -triple=aarch64 -filetype=obj -mattr=+sme-f16f16 < %s \
 // RUN:        | llvm-objdump -d --mattr=-sme2p1 - | FileCheck %s --check-prefix=CHECK-UNKNOWN
-// RUN: llvm-mc -triple=aarch64 -show-encoding -mattr=+sme2p1,+sme-f16f16 < %s \
+// RUN: llvm-mc -triple=aarch64 -show-encoding -mattr=+sme-f16f16 < %s \
 // RUN:        | sed '/.text/d' | sed 's/.*encoding: //g' \
-// RUN:        | llvm-mc -triple=aarch64 -mattr=+sme2p1,+sme-f16f16 -disassemble -show-encoding \
+// RUN:        | llvm-mc -triple=aarch64 -mattr=+sme-f16f16 -disassemble -show-encoding \
 // RUN:        | FileCheck %s --check-prefixes=CHECK-ENCODING,CHECK-INST
 
 
 fsub    za.h[w8, 0], {z0.h - z1.h}  // 11000001-10100100-00011100-00001000
 // CHECK-INST: fsub    za.h[w8, 0, vgx2], { z0.h, z1.h }
 // CHECK-ENCODING: [0x08,0x1c,0xa4,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1
+// CHECK-ERROR: instruction requires: sme-f16f16 or sme-f8f16
 // CHECK-UNKNOWN: c1a41c08 <unknown>
 
 fsub    za.h[w10, 5, vgx2], {z10.h, z11.h}  // 11000001-10100100-01011101-01001101
 // CHECK-INST: fsub    za.h[w10, 5, vgx2], { z10.h, z11.h }
 // CHECK-ENCODING: [0x4d,0x5d,0xa4,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1
+// CHECK-ERROR: instruction requires: sme-f16f16 or sme-f8f16
 // CHECK-UNKNOWN: c1a45d4d <unknown>
 
 fsub    za.h[w10, 5], {z10.h - z11.h}  // 11000001-10100100-01011101-01001101
 // CHECK-INST: fsub    za.h[w10, 5, vgx2], { z10.h, z11.h }
 // CHECK-ENCODING: [0x4d,0x5d,0xa4,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1
+// CHECK-ERROR: instruction requires: sme-f16f16 or sme-f8f16
 // CHECK-UNKNOWN: c1a45d4d <unknown>
 
 fsub    za.h[w11, 7, vgx2], {z12.h, z13.h}  // 11000001-10100100-01111101-10001111
 // CHECK-INST: fsub    za.h[w11, 7, vgx2], { z12.h, z13.h }
 // CHECK-ENCODING: [0x8f,0x7d,0xa4,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1
+// CHECK-ERROR: instruction requires: sme-f16f16 or sme-f8f16
 // CHECK-UNKNOWN: c1a47d8f <unknown>
 
 fsub    za.h[w11, 7], {z12.h - z13.h}  // 11000001-10100100-01111101-10001111
 // CHECK-INST: fsub    za.h[w11, 7, vgx2], { z12.h, z13.h }
 // CHECK-ENCODING: [0x8f,0x7d,0xa4,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1
+// CHECK-ERROR: instruction requires: sme-f16f16 or sme-f8f16
 // CHECK-UNKNOWN: c1a47d8f <unknown>
 
 fsub    za.h[w11, 7, vgx2], {z30.h, z31.h}  // 11000001-10100100-01111111-11001111
 // CHECK-INST: fsub    za.h[w11, 7, vgx2], { z30.h, z31.h }
 // CHECK-ENCODING: [0xcf,0x7f,0xa4,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1
+// CHECK-ERROR: instruction requires: sme-f16f16 or sme-f8f16
 // CHECK-UNKNOWN: c1a47fcf <unknown>
 
 fsub    za.h[w11, 7], {z30.h - z31.h}  // 11000001-10100100-01111111-11001111
 // CHECK-INST: fsub    za.h[w11, 7, vgx2], { z30.h, z31.h }
 // CHECK-ENCODING: [0xcf,0x7f,0xa4,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1
+// CHECK-ERROR: instruction requires: sme-f16f16 or sme-f8f16
 // CHECK-UNKNOWN: c1a47fcf <unknown>
 
 fsub    za.h[w8, 5, vgx2], {z16.h, z17.h}  // 11000001-10100100-00011110-00001101
 // CHECK-INST: fsub    za.h[w8, 5, vgx2], { z16.h, z17.h }
 // CHECK-ENCODING: [0x0d,0x1e,0xa4,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1
+// CHECK-ERROR: instruction requires: sme-f16f16 or sme-f8f16
 // CHECK-UNKNOWN: c1a41e0d <unknown>
 
 fsub    za.h[w8, 5], {z16.h - z17.h}  // 11000001-10100100-00011110-00001101
 // CHECK-INST: fsub    za.h[w8, 5, vgx2], { z16.h, z17.h }
 // CHECK-ENCODING: [0x0d,0x1e,0xa4,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1
+// CHECK-ERROR: instruction requires: sme-f16f16 or sme-f8f16
 // CHECK-UNKNOWN: c1a41e0d <unknown>
 
 fsub    za.h[w8, 1, vgx2], {z0.h, z1.h}  // 11000001-10100100-00011100-00001001
 // CHECK-INST: fsub    za.h[w8, 1, vgx2], { z0.h, z1.h }
 // CHECK-ENCODING: [0x09,0x1c,0xa4,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1
+// CHECK-ERROR: instruction requires: sme-f16f16 or sme-f8f16
 // CHECK-UNKNOWN: c1a41c09 <unknown>
 
 fsub    za.h[w8, 1], {z0.h - z1.h}  // 11000001-10100100-00011100-00001001
 // CHECK-INST: fsub    za.h[w8, 1, vgx2], { z0.h, z1.h }
 // CHECK-ENCODING: [0x09,0x1c,0xa4,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1
+// CHECK-ERROR: instruction requires: sme-f16f16 or sme-f8f16
 // CHECK-UNKNOWN: c1a41c09 <unknown>
 
 fsub    za.h[w10, 0, vgx2], {z18.h, z19.h}  // 11000001-10100100-01011110, 01001000
 // CHECK-INST: fsub    za.h[w10, 0, vgx2], { z18.h, z19.h }
 // CHECK-ENCODING: [0x48,0x5e,0xa4,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1
+// CHECK-ERROR: instruction requires: sme-f16f16 or sme-f8f16
 // CHECK-UNKNOWN: c1a45e48 <unknown>
 
 fsub    za.h[w10, 0], {z18.h - z19.h}  // 11000001-10100100-01011110-01001000
 // CHECK-INST: fsub    za.h[w10, 0, vgx2], { z18.h, z19.h }
 // CHECK-ENCODING: [0x48,0x5e,0xa4,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1
+// CHECK-ERROR: instruction requires: sme-f16f16 or sme-f8f16
 // CHECK-UNKNOWN: c1a45e48 <unknown>
 
 fsub    za.h[w8, 0, vgx2], {z12.h, z13.h}  // 11000001-10100100-00011101-10001000
 // CHECK-INST: fsub    za.h[w8, 0, vgx2], { z12.h, z13.h }
 // CHECK-ENCODING: [0x88,0x1d,0xa4,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1
+// CHECK-ERROR: instruction requires: sme-f16f16 or sme-f8f16
 // CHECK-UNKNOWN: c1a41d88 <unknown>
 
 fsub    za.h[w8, 0], {z12.h - z13.h}  // 11000001-10100100-00011101-10001000
 // CHECK-INST: fsub    za.h[w8, 0, vgx2], { z12.h, z13.h }
 // CHECK-ENCODING: [0x88,0x1d,0xa4,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1
+// CHECK-ERROR: instruction requires: sme-f16f16 or sme-f8f16
 // CHECK-UNKNOWN: c1a41d88 <unknown>
 
 fsub    za.h[w10, 1, vgx2], {z0.h, z1.h}  // 11000001-10100100-01011100-00001001
 // CHECK-INST: fsub    za.h[w10, 1, vgx2], { z0.h, z1.h }
 // CHECK-ENCODING: [0x09,0x5c,0xa4,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1
+// CHECK-ERROR: instruction requires: sme-f16f16 or sme-f8f16
 // CHECK-UNKNOWN: c1a45c09 <unknown>
 
 fsub    za.h[w10, 1], {z0.h - z1.h}  // 11000001-10100100-01011100-00001001
 // CHECK-INST: fsub    za.h[w10, 1, vgx2], { z0.h, z1.h }
 // CHECK-ENCODING: [0x09,0x5c,0xa4,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1
+// CHECK-ERROR: instruction requires: sme-f16f16 or sme-f8f16
 // CHECK-UNKNOWN: c1a45c09 <unknown>
 
 fsub    za.h[w8, 5, vgx2], {z22.h, z23.h}  // 11000001-10100100-00011110, 11001101
 // CHECK-INST: fsub    za.h[w8, 5, vgx2], { z22.h, z23.h }
 // CHECK-ENCODING: [0xcd,0x1e,0xa4,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1
+// CHECK-ERROR: instruction requires: sme-f16f16 or sme-f8f16
 // CHECK-UNKNOWN: c1a41ecd <unknown>
 
 fsub    za.h[w8, 5], {z22.h - z23.h}  // 11000001-10100100-00011110-11001101
 // CHECK-INST: fsub    za.h[w8, 5, vgx2], { z22.h, z23.h }
 // CHECK-ENCODING: [0xcd,0x1e,0xa4,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1
+// CHECK-ERROR: instruction requires: sme-f16f16 or sme-f8f16
 // CHECK-UNKNOWN: c1a41ecd <unknown>
 
 fsub    za.h[w11, 2, vgx2], {z8.h, z9.h}  // 11000001-10100100-01111101-00001010
 // CHECK-INST: fsub    za.h[w11, 2, vgx2], { z8.h, z9.h }
 // CHECK-ENCODING: [0x0a,0x7d,0xa4,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1
+// CHECK-ERROR: instruction requires: sme-f16f16 or sme-f8f16
 // CHECK-UNKNOWN: c1a47d0a <unknown>
 
 fsub    za.h[w11, 2], {z8.h - z9.h}  // 11000001-10100100-01111101-00001010
 // CHECK-INST: fsub    za.h[w11, 2, vgx2], { z8.h, z9.h }
 // CHECK-ENCODING: [0x0a,0x7d,0xa4,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1
+// CHECK-ERROR: instruction requires: sme-f16f16 or sme-f8f16
 // CHECK-UNKNOWN: c1a47d0a <unknown>
 
 fsub    za.h[w9, 7, vgx2], {z12.h, z13.h}  // 11000001-10100100-00111101-10001111
 // CHECK-INST: fsub    za.h[w9, 7, vgx2], { z12.h, z13.h }
 // CHECK-ENCODING: [0x8f,0x3d,0xa4,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1
+// CHECK-ERROR: instruction requires: sme-f16f16 or sme-f8f16
 // CHECK-UNKNOWN: c1a43d8f <unknown>
 
 fsub    za.h[w9, 7], {z12.h - z13.h}  // 11000001-10100100-00111101-10001111
 // CHECK-INST: fsub    za.h[w9, 7, vgx2], { z12.h, z13.h }
 // CHECK-ENCODING: [0x8f,0x3d,0xa4,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1
+// CHECK-ERROR: instruction requires: sme-f16f16 or sme-f8f16
 // CHECK-UNKNOWN: c1a43d8f <unknown>
 
 
 fsub    za.h[w8, 0, vgx4], {z0.h - z3.h}  // 11000001-10100101-00011100-00001000
 // CHECK-INST: fsub    za.h[w8, 0, vgx4], { z0.h - z3.h }
 // CHECK-ENCODING: [0x08,0x1c,0xa5,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1
+// CHECK-ERROR: instruction requires: sme-f16f16 or sme-f8f16
 // CHECK-UNKNOWN: c1a51c08 <unknown>
 
 fsub    za.h[w8, 0], {z0.h - z3.h}  // 11000001-10100101-00011100-00001000
 // CHECK-INST: fsub    za.h[w8, 0, vgx4], { z0.h - z3.h }
 // CHECK-ENCODING: [0x08,0x1c,0xa5,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1
+// CHECK-ERROR: instruction requires: sme-f16f16 or sme-f8f16
 // CHECK-UNKNOWN: c1a51c08 <unknown>
 
 fsub    za.h[w10, 5, vgx4], {z8.h - z11.h}  // 11000001-10100101-01011101-00001101
 // CHECK-INST: fsub    za.h[w10, 5, vgx4], { z8.h - z11.h }
 // CHECK-ENCODING: [0x0d,0x5d,0xa5,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1
+// CHECK-ERROR: instruction requires: sme-f16f16 or sme-f8f16
 // CHECK-UNKNOWN: c1a55d0d <unknown>
 
 fsub    za.h[w10, 5], {z8.h - z11.h}  // 11000001-10100101-01011101-00001101
 // CHECK-INST: fsub    za.h[w10, 5, vgx4], { z8.h - z11.h }
 // CHECK-ENCODING: [0x0d,0x5d,0xa5,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1
+// CHECK-ERROR: instruction requires: sme-f16f16 or sme-f8f16
 // CHECK-UNKNOWN: c1a55d0d <unknown>
 
 fsub    za.h[w11, 7, vgx4], {z12.h - z15.h}  // 11000001-10100101-01111101-10001111
 // CHECK-INST: fsub    za.h[w11, 7, vgx4], { z12.h - z15.h }
 // CHECK-ENCODING: [0x8f,0x7d,0xa5,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1
+// CHECK-ERROR: instruction requires: sme-f16f16 or sme-f8f16
 // CHECK-UNKNOWN: c1a57d8f <unknown>
 
 fsub    za.h[w11, 7], {z12.h - z15.h}  // 11000001-10100101-01111101-10001111
 // CHECK-INST: fsub    za.h[w11, 7, vgx4], { z12.h - z15.h }
 // CHECK-ENCODING: [0x8f,0x7d,0xa5,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1
+// CHECK-ERROR: instruction requires: sme-f16f16 or sme-f8f16
 // CHECK-UNKNOWN: c1a57d8f <unknown>
 
 fsub    za.h[w11, 7, vgx4], {z28.h - z31.h}  // 11000001-10100101-01111111-10001111
 // CHECK-INST: fsub    za.h[w11, 7, vgx4], { z28.h - z31.h }
 // CHECK-ENCODING: [0x8f,0x7f,0xa5,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1
+// CHECK-ERROR: instruction requires: sme-f16f16 or sme-f8f16
 // CHECK-UNKNOWN: c1a57f8f <unknown>
 
 fsub    za.h[w11, 7], {z28.h - z31.h}  // 11000001-10100101-01111111-10001111
 // CHECK-INST: fsub    za.h[w11, 7, vgx4], { z28.h - z31.h }
 // CHECK-ENCODING: [0x8f,0x7f,0xa5,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1
+// CHECK-ERROR: instruction requires: sme-f16f16 or sme-f8f16
 // CHECK-UNKNOWN: c1a57f8f <unknown>
 
 fsub    za.h[w8, 5, vgx4], {z16.h - z19.h}  // 11000001-10100101-00011110-00001101
 // CHECK-INST: fsub    za.h[w8, 5, vgx4], { z16.h - z19.h }
 // CHECK-ENCODING: [0x0d,0x1e,0xa5,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1
+// CHECK-ERROR: instruction requires: sme-f16f16 or sme-f8f16
 // CHECK-UNKNOWN: c1a51e0d <unknown>
 
 fsub    za.h[w8, 5], {z16.h - z19.h}  // 11000001-10100101-00011110-00001101
 // CHECK-INST: fsub    za.h[w8, 5, vgx4], { z16.h - z19.h }
 // CHECK-ENCODING: [0x0d,0x1e,0xa5,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1
+// CHECK-ERROR: instruction requires: sme-f16f16 or sme-f8f16
 // CHECK-UNKNOWN: c1a51e0d <unknown>
 
 fsub    za.h[w8, 1, vgx4], {z0.h - z3.h}  // 11000001-10100101-00011100-00001001
 // CHECK-INST: fsub    za.h[w8, 1, vgx4], { z0.h - z3.h }
 // CHECK-ENCODING: [0x09,0x1c,0xa5,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1
+// CHECK-ERROR: instruction requires: sme-f16f16 or sme-f8f16
 // CHECK-UNKNOWN: c1a51c09 <unknown>
 
 fsub    za.h[w8, 1], {z0.h - z3.h}  // 11000001-10100101-00011100-00001001
 // CHECK-INST: fsub    za.h[w8, 1, vgx4], { z0.h - z3.h }
 // CHECK-ENCODING: [0x09,0x1c,0xa5,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1
+// CHECK-ERROR: instruction requires: sme-f16f16 or sme-f8f16
 // CHECK-UNKNOWN: c1a51c09 <unknown>
 
 fsub    za.h[w10, 0, vgx4], {z16.h - z19.h}  // 11000001-10100101-01011110-00001000
 // CHECK-INST: fsub    za.h[w10, 0, vgx4], { z16.h - z19.h }
 // CHECK-ENCODING: [0x08,0x5e,0xa5,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1
+// CHECK-ERROR: instruction requires: sme-f16f16 or sme-f8f16
 // CHECK-UNKNOWN: c1a55e08 <unknown>
 
 fsub    za.h[w10, 0], {z16.h - z19.h}  // 11000001-10100101-01011110-00001000
 // CHECK-INST: fsub    za.h[w10, 0, vgx4], { z16.h - z19.h }
 // CHECK-ENCODING: [0x08,0x5e,0xa5,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1
+// CHECK-ERROR: instruction requires: sme-f16f16 or sme-f8f16
 // CHECK-UNKNOWN: c1a55e08 <unknown>
 
 fsub    za.h[w8, 0, vgx4], {z12.h - z15.h}  // 11000001-10100101-00011101-10001000
 // CHECK-INST: fsub    za.h[w8, 0, vgx4], { z12.h - z15.h }
 // CHECK-ENCODING: [0x88,0x1d,0xa5,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1
+// CHECK-ERROR: instruction requires: sme-f16f16 or sme-f8f16
 // CHECK-UNKNOWN: c1a51d88 <unknown>
 
 fsub    za.h[w8, 0], {z12.h - z15.h}  // 11000001-10100101-00011101-10001000
 // CHECK-INST: fsub    za.h[w8, 0, vgx4], { z12.h - z15.h }
 // CHECK-ENCODING: [0x88,0x1d,0xa5,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1
+// CHECK-ERROR: instruction requires: sme-f16f16 or sme-f8f16
 // CHECK-UNKNOWN: c1a51d88 <unknown>
 
 fsub    za.h[w10, 1, vgx4], {z0.h - z3.h}  // 11000001-10100101-01011100-00001001
 // CHECK-INST: fsub    za.h[w10, 1, vgx4], { z0.h - z3.h }
 // CHECK-ENCODING: [0x09,0x5c,0xa5,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1
+// CHECK-ERROR: instruction requires: sme-f16f16 or sme-f8f16
 // CHECK-UNKNOWN: c1a55c09 <unknown>
 
 fsub    za.h[w10, 1], {z0.h - z3.h}  // 11000001-10100101-01011100-00001001
 // CHECK-INST: fsub    za.h[w10, 1, vgx4], { z0.h - z3.h }
 // CHECK-ENCODING: [0x09,0x5c,0xa5,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1
+// CHECK-ERROR: instruction requires: sme-f16f16 or sme-f8f16
 // CHECK-UNKNOWN: c1a55c09 <unknown>
 
 fsub    za.h[w8, 5, vgx4], {z20.h - z23.h}  // 11000001-10100101-00011110-10001101
 // CHECK-INST: fsub    za.h[w8, 5, vgx4], { z20.h - z23.h }
 // CHECK-ENCODING: [0x8d,0x1e,0xa5,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1
+// CHECK-ERROR: instruction requires: sme-f16f16 or sme-f8f16
 // CHECK-UNKNOWN: c1a51e8d <unknown>
 
 fsub    za.h[w8, 5], {z20.h - z23.h}  // 11000001-10100101-00011110-10001101
 // CHECK-INST: fsub    za.h[w8, 5, vgx4], { z20.h - z23.h }
 // CHECK-ENCODING: [0x8d,0x1e,0xa5,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1
+// CHECK-ERROR: instruction requires: sme-f16f16 or sme-f8f16
 // CHECK-UNKNOWN: c1a51e8d <unknown>
 
 fsub    za.h[w11, 2, vgx4], {z8.h - z11.h}  // 11000001-10100101-01111101-00001010
 // CHECK-INST: fsub    za.h[w11, 2, vgx4], { z8.h - z11.h }
 // CHECK-ENCODING: [0x0a,0x7d,0xa5,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1
+// CHECK-ERROR: instruction requires: sme-f16f16 or sme-f8f16
 // CHECK-UNKNOWN: c1a57d0a <unknown>
 
 fsub    za.h[w11, 2], {z8.h - z11.h}  // 11000001-10100101-01111101-00001010
 // CHECK-INST: fsub    za.h[w11, 2, vgx4], { z8.h - z11.h }
 // CHECK-ENCODING: [0x0a,0x7d,0xa5,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1
+// CHECK-ERROR: instruction requires: sme-f16f16 or sme-f8f16
 // CHECK-UNKNOWN: c1a57d0a <unknown>
 
 fsub    za.h[w9, 7, vgx4], {z12.h - z15.h}  // 11000001-10100101-00111101-10001111
 // CHECK-INST: fsub    za.h[w9, 7, vgx4], { z12.h - z15.h }
 // CHECK-ENCODING: [0x8f,0x3d,0xa5,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1
+// CHECK-ERROR: instruction requires: sme-f16f16 or sme-f8f16
 // CHECK-UNKNOWN: c1a53d8f <unknown>
 
 fsub    za.h[w9, 7], {z12.h - z15.h}  // 11000001-10100101-00111101-10001111
 // CHECK-INST: fsub    za.h[w9, 7, vgx4], { z12.h - z15.h }
 // CHECK-ENCODING: [0x8f,0x3d,0xa5,0xc1]
-// CHECK-ERROR: instruction requires: sme2p1
+// CHECK-ERROR: instruction requires: sme-f16f16 or sme-f8f16
 // CHECK-UNKNOWN: c1a53d8f <unknown>
diff --git a/llvm/test/MC/AArch64/ilp32-diagnostics.s b/llvm/test/MC/AArch64/ilp32-diagnostics.s
index 65c9e4ea5a1c..4ca15f160418 100644
--- a/llvm/test/MC/AArch64/ilp32-diagnostics.s
+++ b/llvm/test/MC/AArch64/ilp32-diagnostics.s
@@ -10,6 +10,14 @@
 // CHECK-ERROR: error: ILP32 8 byte absolute data relocation not supported (LP64 eqv: ABS64)
 // CHECK-ERROR: ^
 
+        .xword sym@AUTH(da,42)
+// CHECK-ERROR: error: ILP32 8 byte absolute data relocation not supported (LP64 eqv: AUTH_ABS64)
+// CHECK-ERROR: ^
+
+        .xword sym@AUTH(da,42,addr)
+// CHECK-ERROR: error: ILP32 8 byte absolute data relocation not supported (LP64 eqv: AUTH_ABS64)
+// CHECK-ERROR: ^
+
         movz x7, #:abs_g3:some_label
 // CHECK-ERROR: error: ILP32 absolute MOV relocation not supported (LP64 eqv: MOVW_UABS_G3)
 // CHECK-ERROR:        movz x7, #:abs_g3:some_label
diff --git a/llvm/test/MC/AMDGPU/ds-err.s b/llvm/test/MC/AMDGPU/ds-err.s
index 2d25fdf5e302..c31f4c759395 100644
--- a/llvm/test/MC/AMDGPU/ds-err.s
+++ b/llvm/test/MC/AMDGPU/ds-err.s
@@ -18,19 +18,19 @@ ds_write2_b32 v2, v4, v6 offset0:4 offset0:8
 ds_write2_b32 v2, v4, v6 offset1:4 offset1:8
 
 // offset0 too big
-// CHECK: :[[@LINE+1]]:{{[0-9]+}}: error: invalid operand for instruction
+// CHECK: :[[@LINE+1]]:{{[0-9]+}}: error: invalid offset0 value.
 ds_write2_b32 v2, v4, v6 offset0:1000000000
 
 // offset0 too big
-// CHECK: :[[@LINE+1]]:{{[0-9]+}}: error: invalid operand for instruction
+// CHECK: :[[@LINE+1]]:{{[0-9]+}}: error: invalid offset0 value.
 ds_write2_b32 v2, v4, v6 offset0:0x100
 
 // offset1 too big
-// CHECK: :[[@LINE+1]]:{{[0-9]+}}: error: invalid operand for instruction
+// CHECK: :[[@LINE+1]]:{{[0-9]+}}: error: invalid offset1 value.
 ds_write2_b32 v2, v4, v6 offset1:1000000000
 
 // offset1 too big
-// CHECK: :[[@LINE+1]]:{{[0-9]+}}: error: invalid operand for instruction
+// CHECK: :[[@LINE+1]]:{{[0-9]+}}: error: invalid offset1 value.
 ds_write2_b32 v2, v4, v6 offset1:0x100
 
 //===----------------------------------------------------------------------===//
diff --git a/llvm/test/MC/AMDGPU/gfx11_asm_err.s b/llvm/test/MC/AMDGPU/gfx11_asm_err.s
index 3ec31626be5b..7f99afe01925 100644
--- a/llvm/test/MC/AMDGPU/gfx11_asm_err.s
+++ b/llvm/test/MC/AMDGPU/gfx11_asm_err.s
@@ -22,13 +22,13 @@ s_delay_alu instid0(VALU_DEP_1) | SALU_CYCLE_1)
 // GFX11: :[[@LINE-1]]:{{[0-9]+}}: error: expected a left parenthesis
 
 lds_direct_load v15 wait_vdst:16
-// GFX11: :[[@LINE-1]]:{{[0-9]+}}: error: invalid operand for instruction
+// GFX11: :[[@LINE-1]]:{{[0-9]+}}: error: invalid wait_vdst value.
 
 lds_direct_load v15 wait_vdst
 // GFX11: :[[@LINE-1]]:{{[0-9]+}}: error: invalid operand for instruction
 
 v_interp_p10_f32 v0, v1, v2, v3 wait_exp:8
-// GFX11: :[[@LINE-1]]:{{[0-9]+}}: error: invalid operand for instruction
+// GFX11: :[[@LINE-1]]:{{[0-9]+}}: error: invalid wait_exp value.
 
 v_interp_p2_f32 v0, -v1, v2, v3 wait_exp
 // GFX11: :[[@LINE-1]]:{{[0-9]+}}: error: invalid operand for instruction
diff --git a/llvm/test/MC/AMDGPU/gfx12_asm_vop3.s b/llvm/test/MC/AMDGPU/gfx12_asm_vop3.s
index be9edc3e019e..b0854881d428 100644
--- a/llvm/test/MC/AMDGPU/gfx12_asm_vop3.s
+++ b/llvm/test/MC/AMDGPU/gfx12_asm_vop3.s
@@ -1126,6 +1126,18 @@ v_cvt_sr_fp8_f32 v10, s2, v5
 v_cvt_sr_fp8_f32 v5, -|v255|, v4
 // GFX12: encoding: [0x05,0x01,0x6b,0xd7,0xff,0x09,0x02,0x20]
 
+v_cvt_sr_fp8_f32 v1, v2, v3 byte_sel:0
+// GFX12: v_cvt_sr_fp8_f32 v1, v2, v3             ; encoding: [0x01,0x00,0x6b,0xd7,0x02,0x07,0x02,0x00]
+
+v_cvt_sr_fp8_f32 v1, v2, v3 byte_sel:1
+// GFX12: v_cvt_sr_fp8_f32 v1, v2, v3 byte_sel:1  ; encoding: [0x01,0x20,0x6b,0xd7,0x02,0x07,0x02,0x00]
+
+v_cvt_sr_fp8_f32 v1, v2, v3 byte_sel:2
+// GFX12: v_cvt_sr_fp8_f32 v1, v2, v3 byte_sel:2  ; encoding: [0x01,0x40,0x6b,0xd7,0x02,0x07,0x02,0x00]
+
+v_cvt_sr_fp8_f32 v1, v2, v3 byte_sel:3
+// GFX12: v_cvt_sr_fp8_f32 v1, v2, v3 byte_sel:3  ; encoding: [0x01,0x60,0x6b,0xd7,0x02,0x07,0x02,0x00]
+
 v_cvt_sr_bf8_f32 v1, v2, v3
 // GFX12: encoding: [0x01,0x00,0x6c,0xd7,0x02,0x07,0x02,0x00]
 
@@ -1135,6 +1147,18 @@ v_cvt_sr_bf8_f32 v10, s2, v5
 v_cvt_sr_bf8_f32 v5, -|v255|, v4
 // GFX12: encoding: [0x05,0x01,0x6c,0xd7,0xff,0x09,0x02,0x20]
 
+v_cvt_sr_bf8_f32 v1, v2, v3 byte_sel:0
+// GFX12: v_cvt_sr_bf8_f32 v1, v2, v3             ; encoding: [0x01,0x00,0x6c,0xd7,0x02,0x07,0x02,0x00]
+
+v_cvt_sr_bf8_f32 v1, v2, v3 byte_sel:1
+// GFX12: v_cvt_sr_bf8_f32 v1, v2, v3 byte_sel:1  ; encoding: [0x01,0x20,0x6c,0xd7,0x02,0x07,0x02,0x00]
+
+v_cvt_sr_bf8_f32 v1, v2, v3 byte_sel:2
+// GFX12: v_cvt_sr_bf8_f32 v1, v2, v3 byte_sel:2  ; encoding: [0x01,0x40,0x6c,0xd7,0x02,0x07,0x02,0x00]
+
+v_cvt_sr_bf8_f32 v1, v2, v3 byte_sel:3
+// GFX12: v_cvt_sr_bf8_f32 v1, v2, v3 byte_sel:3  ; encoding: [0x01,0x60,0x6c,0xd7,0x02,0x07,0x02,0x00]
+
 v_cvt_pk_i16_f32 v5, v1, v2
 // GFX12: encoding: [0x05,0x00,0x06,0xd7,0x01,0x05,0x02,0x00]
 
diff --git a/llvm/test/MC/AMDGPU/gfx12_asm_vop3_dpp16.s b/llvm/test/MC/AMDGPU/gfx12_asm_vop3_dpp16.s
index d0e309adce41..16cd8d5aa5e9 100644
--- a/llvm/test/MC/AMDGPU/gfx12_asm_vop3_dpp16.s
+++ b/llvm/test/MC/AMDGPU/gfx12_asm_vop3_dpp16.s
@@ -1192,6 +1192,18 @@ v_cvt_sr_bf8_f32_e64_dpp v1, -v2, v3 quad_perm:[3,2,1,0] row_mask:0xe bank_mask:
 v_cvt_sr_bf8_f32_e64_dpp v1, -v2, v3 quad_perm:[3,2,1,0] row_mask:0xe bank_mask:0xd fi:1
 // GFX12: encoding: [0x01,0x00,0x6c,0xd7,0xfa,0x06,0x02,0x20,0x02,0x1b,0x04,0xed]
 
+v_cvt_sr_bf8_f32 v1, v2, v3 byte_sel:0 quad_perm:[3,2,1,0]
+// GFX12: v_cvt_sr_bf8_f32_e64_dpp v1, v2, v3 quad_perm:[3,2,1,0] row_mask:0xf bank_mask:0xf ; encoding: [0x01,0x00,0x6c,0xd7,0xfa,0x06,0x02,0x00,0x02,0x1b,0x00,0xff]
+
+v_cvt_sr_bf8_f32 v1, v2, v3 byte_sel:1 quad_perm:[3,2,1,0]
+// GFX12: v_cvt_sr_bf8_f32_e64_dpp v1, v2, v3 byte_sel:1 quad_perm:[3,2,1,0] row_mask:0xf bank_mask:0xf ; encoding: [0x01,0x20,0x6c,0xd7,0xfa,0x06,0x02,0x00,0x02,0x1b,0x00,0xff]
+
+v_cvt_sr_bf8_f32 v1, v2, v3 byte_sel:2 quad_perm:[3,2,1,0]
+// GFX12: v_cvt_sr_bf8_f32_e64_dpp v1, v2, v3 byte_sel:2 quad_perm:[3,2,1,0] row_mask:0xf bank_mask:0xf ; encoding: [0x01,0x40,0x6c,0xd7,0xfa,0x06,0x02,0x00,0x02,0x1b,0x00,0xff]
+
+v_cvt_sr_bf8_f32 v1, v2, v3 byte_sel:3 quad_perm:[3,2,1,0]
+// GFX12: v_cvt_sr_bf8_f32_e64_dpp v1, v2, v3 byte_sel:3 quad_perm:[3,2,1,0] row_mask:0xf bank_mask:0xf ; encoding: [0x01,0x60,0x6c,0xd7,0xfa,0x06,0x02,0x00,0x02,0x1b,0x00,0xff]
+
 v_cvt_sr_fp8_f32_e64_dpp v1, -v2, v3 quad_perm:[3,2,1,0] row_mask:0xe bank_mask:0xd
 // GFX12: encoding: [0x01,0x00,0x6b,0xd7,0xfa,0x06,0x02,0x20,0x02,0x1b,0x00,0xed]
 
@@ -1219,6 +1231,18 @@ v_cvt_sr_fp8_f32_e64_dpp v1, -v2, v3 quad_perm:[3,2,1,0] row_mask:0xe bank_mask:
 v_cvt_sr_fp8_f32_e64_dpp v1, -v2, v3 quad_perm:[3,2,1,0] row_mask:0xe bank_mask:0xd fi:1
 // GFX12: encoding: [0x01,0x00,0x6b,0xd7,0xfa,0x06,0x02,0x20,0x02,0x1b,0x04,0xed]
 
+v_cvt_sr_fp8_f32 v1, v2, v3 byte_sel:0 quad_perm:[3,2,1,0]
+// GFX12: v_cvt_sr_fp8_f32_e64_dpp v1, v2, v3 quad_perm:[3,2,1,0] row_mask:0xf bank_mask:0xf ; encoding: [0x01,0x00,0x6b,0xd7,0xfa,0x06,0x02,0x00,0x02,0x1b,0x00,0xff]
+
+v_cvt_sr_fp8_f32 v1, v2, v3 byte_sel:1 quad_perm:[3,2,1,0]
+// GFX12: v_cvt_sr_fp8_f32_e64_dpp v1, v2, v3 byte_sel:1 quad_perm:[3,2,1,0] row_mask:0xf bank_mask:0xf ; encoding: [0x01,0x20,0x6b,0xd7,0xfa,0x06,0x02,0x00,0x02,0x1b,0x00,0xff]
+
+v_cvt_sr_fp8_f32 v1, v2, v3 byte_sel:2 quad_perm:[3,2,1,0]
+// GFX12: v_cvt_sr_fp8_f32_e64_dpp v1, v2, v3 byte_sel:2 quad_perm:[3,2,1,0] row_mask:0xf bank_mask:0xf ; encoding: [0x01,0x40,0x6b,0xd7,0xfa,0x06,0x02,0x00,0x02,0x1b,0x00,0xff]
+
+v_cvt_sr_fp8_f32 v1, v2, v3 byte_sel:3 quad_perm:[3,2,1,0]
+// GFX12: v_cvt_sr_fp8_f32_e64_dpp v1, v2, v3 byte_sel:3 quad_perm:[3,2,1,0] row_mask:0xf bank_mask:0xf ; encoding: [0x01,0x60,0x6b,0xd7,0xfa,0x06,0x02,0x00,0x02,0x1b,0x00,0xff]
+
 v_cvt_pk_i16_f32_e64_dpp v5, v1, v2 quad_perm:[3,2,1,0]
 // GFX12: [0x05,0x00,0x06,0xd7,0xfa,0x04,0x02,0x00,0x01,0x1b,0x00,0xff]
 
diff --git a/llvm/test/MC/AMDGPU/gfx12_asm_vop3_dpp8.s b/llvm/test/MC/AMDGPU/gfx12_asm_vop3_dpp8.s
index 25b13ac62e4a..d6ef14cff5fa 100644
--- a/llvm/test/MC/AMDGPU/gfx12_asm_vop3_dpp8.s
+++ b/llvm/test/MC/AMDGPU/gfx12_asm_vop3_dpp8.s
@@ -698,6 +698,18 @@ v_cvt_sr_fp8_f32_e64_dpp v5, -v1, v2 dpp8:[7,6,5,4,3,2,1,0]
 v_cvt_sr_fp8_f32_e64_dpp v255, -|v255|, v255 dpp8:[0,0,0,0,0,0,0,0]
 // GFX12: encoding: [0xff,0x01,0x6b,0xd7,0xe9,0xfe,0x03,0x20,0xff,0x00,0x00,0x00]
 
+v_cvt_sr_fp8_f32 v1, v2, v3 byte_sel:0 dpp8:[7,6,5,4,3,2,1,0]
+// GFX12: v_cvt_sr_fp8_f32_e64_dpp v1, v2, v3 dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0x01,0x00,0x6b,0xd7,0xe9,0x06,0x02,0x00,0x02,0x77,0x39,0x05]
+
+v_cvt_sr_fp8_f32 v1, v2, v3 byte_sel:1 dpp8:[7,6,5,4,3,2,1,0]
+// GFX12: v_cvt_sr_fp8_f32_e64_dpp v1, v2, v3 byte_sel:1 dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0x01,0x20,0x6b,0xd7,0xe9,0x06,0x02,0x00,0x02,0x77,0x39,0x05]
+
+v_cvt_sr_fp8_f32 v1, v2, v3 byte_sel:2 dpp8:[7,6,5,4,3,2,1,0]
+// GFX12: v_cvt_sr_fp8_f32_e64_dpp v1, v2, v3 byte_sel:2 dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0x01,0x40,0x6b,0xd7,0xe9,0x06,0x02,0x00,0x02,0x77,0x39,0x05]
+
+v_cvt_sr_fp8_f32 v1, v2, v3 byte_sel:3 dpp8:[7,6,5,4,3,2,1,0]
+// GFX12: v_cvt_sr_fp8_f32_e64_dpp v1, v2, v3 byte_sel:3 dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0x01,0x60,0x6b,0xd7,0xe9,0x06,0x02,0x00,0x02,0x77,0x39,0x05]
+
 v_cvt_sr_bf8_f32_e64_dpp v5, v1, v2 dpp8:[7,6,5,4,3,2,1,0]
 // GFX12: encoding: [0x05,0x00,0x6c,0xd7,0xe9,0x04,0x02,0x00,0x01,0x77,0x39,0x05]
 
@@ -710,6 +722,18 @@ v_cvt_sr_bf8_f32_e64_dpp v5, -v1, v2 dpp8:[7,6,5,4,3,2,1,0]
 v_cvt_sr_bf8_f32_e64_dpp v255, -|v255|, v255 dpp8:[0,0,0,0,0,0,0,0]
 // GFX12: encoding: [0xff,0x01,0x6c,0xd7,0xe9,0xfe,0x03,0x20,0xff,0x00,0x00,0x00]
 
+v_cvt_sr_bf8_f32 v1, v2, v3 byte_sel:0 dpp8:[7,6,5,4,3,2,1,0]
+// GFX12: v_cvt_sr_bf8_f32_e64_dpp v1, v2, v3 dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0x01,0x00,0x6c,0xd7,0xe9,0x06,0x02,0x00,0x02,0x77,0x39,0x05]
+
+v_cvt_sr_bf8_f32 v1, v2, v3 byte_sel:1 dpp8:[7,6,5,4,3,2,1,0]
+// GFX12: v_cvt_sr_bf8_f32_e64_dpp v1, v2, v3 byte_sel:1 dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0x01,0x20,0x6c,0xd7,0xe9,0x06,0x02,0x00,0x02,0x77,0x39,0x05]
+
+v_cvt_sr_bf8_f32 v1, v2, v3 byte_sel:2 dpp8:[7,6,5,4,3,2,1,0]
+// GFX12: v_cvt_sr_bf8_f32_e64_dpp v1, v2, v3 byte_sel:2 dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0x01,0x40,0x6c,0xd7,0xe9,0x06,0x02,0x00,0x02,0x77,0x39,0x05]
+
+v_cvt_sr_bf8_f32 v1, v2, v3 byte_sel:3 dpp8:[7,6,5,4,3,2,1,0]
+// GFX12: v_cvt_sr_bf8_f32_e64_dpp v1, v2, v3 byte_sel:3 dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0x01,0x60,0x6c,0xd7,0xe9,0x06,0x02,0x00,0x02,0x77,0x39,0x05]
+
 v_cvt_pk_i16_f32_e64_dpp v5, v1, v2 dpp8:[7,6,5,4,3,2,1,0]
 // GFX12: [0x05,0x00,0x06,0xd7,0xe9,0x04,0x02,0x00,0x01,0x77,0x39,0x05]
 
diff --git a/llvm/test/MC/AMDGPU/gfx12_asm_vop3_err.s b/llvm/test/MC/AMDGPU/gfx12_asm_vop3_err.s
index 55c5fcabea73..a9dd290ea67d 100644
--- a/llvm/test/MC/AMDGPU/gfx12_asm_vop3_err.s
+++ b/llvm/test/MC/AMDGPU/gfx12_asm_vop3_err.s
@@ -101,3 +101,8 @@ v_permlane16_var_b32 v5, v1, v2 op_sel:[0, 0, 1]
 // GFX12: error: invalid op_sel operand
 // GFX12-NEXT:{{^}}v_permlane16_var_b32 v5, v1, v2 op_sel:[0, 0, 1]
 // GFX12-NEXT:{{^}}                                ^
+
+v_cvt_sr_bf8_f32 v1, v2, v3 byte_sel:4
+// GFX12: :[[@LINE-1]]:{{[0-9]+}}: error: invalid byte_sel value.
+// GFX12-NEXT:{{^}}v_cvt_sr_bf8_f32 v1, v2, v3 byte_sel:4
+// GFX12-NEXT:{{^}}                            ^
diff --git a/llvm/test/MC/AMDGPU/pal-msgpack.s b/llvm/test/MC/AMDGPU/pal-msgpack.s
index 886cc8b0538b..03c6c547af8a 100644
--- a/llvm/test/MC/AMDGPU/pal-msgpack.s
+++ b/llvm/test/MC/AMDGPU/pal-msgpack.s
@@ -14,10 +14,10 @@ amdpal.pipelines:
       - 0x123456789abcdef0
       - 0xfedcba9876543210
     .registers:      
-      0x2c0a (SPI_SHADER_PGM_RSRC1_PS): 0
-      0x2c0b (SPI_SHADER_PGM_RSRC2_PS): 0x42000000
-      0xa1b3 (SPI_PS_INPUT_ENA): 0x1
-      0xa1b4 (SPI_PS_INPUT_ADDR): 0x1
+      '0x2c0a (SPI_SHADER_PGM_RSRC1_PS)': 0
+      '0x2c0b (SPI_SHADER_PGM_RSRC2_PS)': 0x42000000
+      '0xa1b3 (SPI_PS_INPUT_ENA)': 0x1
+      '0xa1b4 (SPI_PS_INPUT_ADDR)': 0x1
 ...
 	.end_amdgpu_pal_metadata
 
@@ -34,10 +34,10 @@ amdpal.pipelines:
 // ASM:       - 0x123456789abcdef0
 // ASM:       - 0xfedcba9876543210
 // ASM:     .registers:      
-// ASM:       0x2c0a (SPI_SHADER_PGM_RSRC1_PS): 0
-// ASM:       0x2c0b (SPI_SHADER_PGM_RSRC2_PS): 0x42000000
-// ASM:       0xa1b3 (SPI_PS_INPUT_ENA): 0x1
-// ASM:       0xa1b4 (SPI_PS_INPUT_ADDR): 0x1
+// ASM:       '0x2c0a (SPI_SHADER_PGM_RSRC1_PS)': 0
+// ASM:       '0x2c0b (SPI_SHADER_PGM_RSRC2_PS)': 0x42000000
+// ASM:       '0xa1b3 (SPI_PS_INPUT_ENA)': 0x1
+// ASM:       '0xa1b4 (SPI_PS_INPUT_ADDR)': 0x1
 // ASM: ...
 // ASM: 	.end_amdgpu_pal_metadata
 
diff --git a/llvm/test/MC/Disassembler/AMDGPU/gfx12_dasm_vop3.txt b/llvm/test/MC/Disassembler/AMDGPU/gfx12_dasm_vop3.txt
index 6acaa8152720..2c911777ef97 100644
--- a/llvm/test/MC/Disassembler/AMDGPU/gfx12_dasm_vop3.txt
+++ b/llvm/test/MC/Disassembler/AMDGPU/gfx12_dasm_vop3.txt
@@ -1020,6 +1020,15 @@
 # GFX12: v_cvt_sr_fp8_f32 v5, -|v255|, v4 ; encoding: [0x05,0x01,0x6b,0xd7,0xff,0x09,0x02,0x20]
 0x05,0x01,0x6b,0xd7,0xff,0x09,0x02,0x20
 
+# GFX12: v_cvt_sr_fp8_f32 v1, v2, v3 byte_sel:1  ; encoding: [0x01,0x20,0x6b,0xd7,0x02,0x07,0x02,0x00]
+0x01,0x20,0x6b,0xd7,0x02,0x07,0x02,0x00
+
+# GFX12: v_cvt_sr_fp8_f32 v1, v2, v3 byte_sel:2  ; encoding: [0x01,0x40,0x6b,0xd7,0x02,0x07,0x02,0x00]
+0x01,0x40,0x6b,0xd7,0x02,0x07,0x02,0x00
+
+# GFX12: v_cvt_sr_fp8_f32 v1, v2, v3 byte_sel:3  ; encoding: [0x01,0x60,0x6b,0xd7,0x02,0x07,0x02,0x00]
+0x01,0x60,0x6b,0xd7,0x02,0x07,0x02,0x00
+
 # GFX12: v_cvt_sr_bf8_f32 v1, v2, v3 ; encoding: [0x01,0x00,0x6c,0xd7,0x02,0x07,0x02,0x00]
 0x01,0x00,0x6c,0xd7,0x02,0x07,0x02,0x00
 
@@ -1029,6 +1038,15 @@
 # GFX12: v_cvt_sr_bf8_f32 v5, -|v255|, v4 ; encoding: [0x05,0x01,0x6c,0xd7,0xff,0x09,0x02,0x20]
 0x05,0x01,0x6c,0xd7,0xff,0x09,0x02,0x20
 
+# GFX12: v_cvt_sr_bf8_f32 v1, v2, v3 byte_sel:1  ; encoding: [0x01,0x20,0x6c,0xd7,0x02,0x07,0x02,0x00]
+0x01,0x20,0x6c,0xd7,0x02,0x07,0x02,0x00
+
+# GFX12: v_cvt_sr_bf8_f32 v1, v2, v3 byte_sel:2  ; encoding: [0x01,0x40,0x6c,0xd7,0x02,0x07,0x02,0x00]
+0x01,0x40,0x6c,0xd7,0x02,0x07,0x02,0x00
+
+# GFX12: v_cvt_sr_bf8_f32 v1, v2, v3 byte_sel:3  ; encoding: [0x01,0x60,0x6c,0xd7,0x02,0x07,0x02,0x00]
+0x01,0x60,0x6c,0xd7,0x02,0x07,0x02,0x00
+
 # GFX12: v_cvt_pk_i16_f32 v5, v1, v2             ; encoding: [0x05,0x00,0x06,0xd7,0x01,0x05,0x02,0x00]
 0x05,0x00,0x06,0xd7,0x01,0x05,0x02,0x00
 
diff --git a/llvm/test/MC/Disassembler/AMDGPU/gfx12_dasm_vop3_dpp16.txt b/llvm/test/MC/Disassembler/AMDGPU/gfx12_dasm_vop3_dpp16.txt
index 0771e6449b62..f9b6c1b73ddc 100644
--- a/llvm/test/MC/Disassembler/AMDGPU/gfx12_dasm_vop3_dpp16.txt
+++ b/llvm/test/MC/Disassembler/AMDGPU/gfx12_dasm_vop3_dpp16.txt
@@ -945,6 +945,15 @@
 # GFX12: v_cvt_sr_bf8_f32_e64_dpp v1, -v2, v3 quad_perm:[3,2,1,0] row_mask:0xe bank_mask:0xd fi:1 ; encoding: [0x01,0x00,0x6c,0xd7,0xfa,0x06,0x02,0x20,0x02,0x1b,0x04,0xed]
 0x01,0x00,0x6c,0xd7,0xfa,0x06,0x02,0x20,0x02,0x1b,0x04,0xed
 
+# GFX12: v_cvt_sr_bf8_f32_e64_dpp v1, v2, v3 byte_sel:1 quad_perm:[3,2,1,0] row_mask:0xf bank_mask:0xf ; encoding: [0x01,0x20,0x6c,0xd7,0xfa,0x06,0x02,0x00,0x02,0x1b,0x00,0xff]
+0x01,0x20,0x6c,0xd7,0xfa,0x06,0x02,0x00,0x02,0x1b,0x00,0xff
+
+# GFX12: v_cvt_sr_bf8_f32_e64_dpp v1, v2, v3 byte_sel:2 quad_perm:[3,2,1,0] row_mask:0xf bank_mask:0xf ; encoding: [0x01,0x40,0x6c,0xd7,0xfa,0x06,0x02,0x00,0x02,0x1b,0x00,0xff]
+0x01,0x40,0x6c,0xd7,0xfa,0x06,0x02,0x00,0x02,0x1b,0x00,0xff
+
+# GFX12: v_cvt_sr_bf8_f32_e64_dpp v1, v2, v3 byte_sel:3 quad_perm:[3,2,1,0] row_mask:0xf bank_mask:0xf ; encoding: [0x01,0x60,0x6c,0xd7,0xfa,0x06,0x02,0x00,0x02,0x1b,0x00,0xff]
+0x01,0x60,0x6c,0xd7,0xfa,0x06,0x02,0x00,0x02,0x1b,0x00,0xff
+
 # GFX12: v_cvt_sr_fp8_f32_e64_dpp v1, -v2, v3 quad_perm:[3,2,1,0] row_mask:0xe bank_mask:0xd ; encoding: [0x01,0x00,0x6b,0xd7,0xfa,0x06,0x02,0x20,0x02,0x1b,0x00,0xed]
 0x01,0x00,0x6b,0xd7,0xfa,0x06,0x02,0x20,0x02,0x1b,0x00,0xed
 
@@ -972,6 +981,15 @@
 # GFX12: v_cvt_sr_fp8_f32_e64_dpp v1, -v2, v3 quad_perm:[3,2,1,0] row_mask:0xe bank_mask:0xd fi:1 ; encoding: [0x01,0x00,0x6b,0xd7,0xfa,0x06,0x02,0x20,0x02,0x1b,0x04,0xed]
 0x01,0x00,0x6b,0xd7,0xfa,0x06,0x02,0x20,0x02,0x1b,0x04,0xed
 
+# GFX12: v_cvt_sr_fp8_f32_e64_dpp v1, v2, v3 byte_sel:1 quad_perm:[3,2,1,0] row_mask:0xf bank_mask:0xf ; encoding: [0x01,0x20,0x6b,0xd7,0xfa,0x06,0x02,0x00,0x02,0x1b,0x00,0xff]
+0x01,0x20,0x6b,0xd7,0xfa,0x06,0x02,0x00,0x02,0x1b,0x00,0xff
+
+# GFX12: v_cvt_sr_fp8_f32_e64_dpp v1, v2, v3 byte_sel:2 quad_perm:[3,2,1,0] row_mask:0xf bank_mask:0xf ; encoding: [0x01,0x40,0x6b,0xd7,0xfa,0x06,0x02,0x00,0x02,0x1b,0x00,0xff]
+0x01,0x40,0x6b,0xd7,0xfa,0x06,0x02,0x00,0x02,0x1b,0x00,0xff
+
+# GFX12: v_cvt_sr_fp8_f32_e64_dpp v1, v2, v3 byte_sel:3 quad_perm:[3,2,1,0] row_mask:0xf bank_mask:0xf ; encoding: [0x01,0x60,0x6b,0xd7,0xfa,0x06,0x02,0x00,0x02,0x1b,0x00,0xff]
+0x01,0x60,0x6b,0xd7,0xfa,0x06,0x02,0x00,0x02,0x1b,0x00,0xff
+
 # GFX12: v_cvt_pk_i16_f32_e64_dpp v5, v1, v2 quad_perm:[3,2,1,0] row_mask:0xf bank_mask:0xf ; encoding: [0x05,0x00,0x06,0xd7,0xfa,0x04,0x02,0x00,0x01,0x1b,0x00,0xff]
 0x05,0x00,0x06,0xd7,0xfa,0x04,0x02,0x00,0x01,0x1b,0x00,0xff
 
diff --git a/llvm/test/MC/Disassembler/AMDGPU/gfx12_dasm_vop3_dpp8.txt b/llvm/test/MC/Disassembler/AMDGPU/gfx12_dasm_vop3_dpp8.txt
index a836adafb31e..eedc6d491087 100644
--- a/llvm/test/MC/Disassembler/AMDGPU/gfx12_dasm_vop3_dpp8.txt
+++ b/llvm/test/MC/Disassembler/AMDGPU/gfx12_dasm_vop3_dpp8.txt
@@ -570,6 +570,15 @@
 # GFX12: v_cvt_sr_fp8_f32_e64_dpp v255, -|v255|, v255 dpp8:[0,0,0,0,0,0,0,0] ; encoding: [0xff,0x01,0x6b,0xd7,0xe9,0xfe,0x03,0x20,0xff,0x00,0x00,0x00]
 0xff,0x01,0x6b,0xd7,0xe9,0xfe,0x03,0x20,0xff,0x00,0x00,0x00
 
+# GFX12: v_cvt_sr_fp8_f32_e64_dpp v1, v2, v3 byte_sel:1 dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0x01,0x20,0x6b,0xd7,0xe9,0x06,0x02,0x00,0x02,0x77,0x39,0x05]
+0x01,0x20,0x6b,0xd7,0xe9,0x06,0x02,0x00,0x02,0x77,0x39,0x05
+
+# GFX12: v_cvt_sr_fp8_f32_e64_dpp v1, v2, v3 byte_sel:2 dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0x01,0x40,0x6b,0xd7,0xe9,0x06,0x02,0x00,0x02,0x77,0x39,0x05]
+0x01,0x40,0x6b,0xd7,0xe9,0x06,0x02,0x00,0x02,0x77,0x39,0x05
+
+# GFX12: v_cvt_sr_fp8_f32_e64_dpp v1, v2, v3 byte_sel:3 dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0x01,0x60,0x6b,0xd7,0xe9,0x06,0x02,0x00,0x02,0x77,0x39,0x05]
+0x01,0x60,0x6b,0xd7,0xe9,0x06,0x02,0x00,0x02,0x77,0x39,0x05
+
 # GFX12: v_cvt_sr_bf8_f32_e64_dpp v5, v1, v2 dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0x05,0x00,0x6c,0xd7,0xe9,0x04,0x02,0x00,0x01,0x77,0x39,0x05]
 0x05,0x00,0x6c,0xd7,0xe9,0x04,0x02,0x00,0x01,0x77,0x39,0x05
 
@@ -582,6 +591,15 @@
 # GFX12: v_cvt_sr_bf8_f32_e64_dpp v255, -|v255|, v255 dpp8:[0,0,0,0,0,0,0,0] ; encoding: [0xff,0x01,0x6c,0xd7,0xe9,0xfe,0x03,0x20,0xff,0x00,0x00,0x00]
 0xff,0x01,0x6c,0xd7,0xe9,0xfe,0x03,0x20,0xff,0x00,0x00,0x00
 
+# GFX12: v_cvt_sr_bf8_f32_e64_dpp v1, v2, v3 byte_sel:1 dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0x01,0x20,0x6c,0xd7,0xe9,0x06,0x02,0x00,0x02,0x77,0x39,0x05]
+0x01,0x20,0x6c,0xd7,0xe9,0x06,0x02,0x00,0x02,0x77,0x39,0x05
+
+# GFX12: v_cvt_sr_bf8_f32_e64_dpp v1, v2, v3 byte_sel:2 dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0x01,0x40,0x6c,0xd7,0xe9,0x06,0x02,0x00,0x02,0x77,0x39,0x05]
+0x01,0x40,0x6c,0xd7,0xe9,0x06,0x02,0x00,0x02,0x77,0x39,0x05
+
+# GFX12: v_cvt_sr_bf8_f32_e64_dpp v1, v2, v3 byte_sel:3 dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0x01,0x60,0x6c,0xd7,0xe9,0x06,0x02,0x00,0x02,0x77,0x39,0x05]
+0x01,0x60,0x6c,0xd7,0xe9,0x06,0x02,0x00,0x02,0x77,0x39,0x05
+
 # GFX12: v_cvt_pk_i16_f32_e64_dpp v5, v1, v2 dpp8:[7,6,5,4,3,2,1,0] ; encoding: [0x05,0x00,0x06,0xd7,0xe9,0x04,0x02,0x00,0x01,0x77,0x39,0x05]
 0x05,0x00,0x06,0xd7,0xe9,0x04,0x02,0x00,0x01,0x77,0x39,0x05
 
diff --git a/llvm/test/MC/Disassembler/M68k/data.txt b/llvm/test/MC/Disassembler/M68k/data.txt
index 8e2fb3f13560..3951ea677f11 100644
--- a/llvm/test/MC/Disassembler/M68k/data.txt
+++ b/llvm/test/MC/Disassembler/M68k/data.txt
@@ -36,6 +36,12 @@
 # CHECK: move.l (64,%sp,%a0), %d0
 0x20 0x37 0x88 0x40
 
+# CHECK: move.b #234, %d2
+0x14 0x3c 0x00 0xea
+
+# CHECK: moveq #100, %d2
+0x74 0x64
+
 # CHECK: move.l $f0000000, %a5
 0x2a 0x79 0xf0 0x00 0x00 0x00
 
diff --git a/llvm/test/MC/ELF/RISCV/gen-dwarf.s b/llvm/test/MC/ELF/RISCV/gen-dwarf.s
index 342ed1cc0e7e..34d02f5da44f 100644
--- a/llvm/test/MC/ELF/RISCV/gen-dwarf.s
+++ b/llvm/test/MC/ELF/RISCV/gen-dwarf.s
@@ -40,28 +40,28 @@
 # CHECK-NEXT: 0x00000020: [DW_RLE_end_of_list ]
 
 # RELOC:      Section ([[#]]) .rela.eh_frame {
-# RELOC-NEXT:   0x1C R_RISCV_32_PCREL <null> 0x0
-# RELOC-NEXT:   0x20 R_RISCV_ADD32 <null> 0x0
-# RELOC-NEXT:   0x20 R_RISCV_SUB32 <null> 0x0
-# RELOC-NEXT:   0x25 R_RISCV_SET6 <null> 0x0
-# RELOC-NEXT:   0x25 R_RISCV_SUB6 <null> 0x0
-# RELOC-NEXT:   0x34 R_RISCV_32_PCREL <null> 0x0
+# RELOC-NEXT:   0x1C R_RISCV_32_PCREL .L0  0x0
+# RELOC-NEXT:   0x20 R_RISCV_ADD32 .L0  0x0
+# RELOC-NEXT:   0x20 R_RISCV_SUB32 .L0  0x0
+# RELOC-NEXT:   0x25 R_RISCV_SET6 .L0  0x0
+# RELOC-NEXT:   0x25 R_RISCV_SUB6 .L0  0x0
+# RELOC-NEXT:   0x34 R_RISCV_32_PCREL .L0  0x0
 # RELOC-NEXT: }
 
 # RELOC:      Section ([[#]]) .rela.debug_rnglists {
 # RELOC-NEXT:   0xD R_RISCV_64 .text.foo 0x0
-# RELOC-NEXT:   0x15 R_RISCV_SET_ULEB128 <null> 0x0
+# RELOC-NEXT:   0x15 R_RISCV_SET_ULEB128 .L0  0x0
 # RELOC-NEXT:   0x15 R_RISCV_SUB_ULEB128 .text.foo 0x0
 # RELOC-NEXT:   0x17 R_RISCV_64 .text.bar 0x0
 # RELOC-NEXT: }
 
 # RELOC:      Section ([[#]]) .rela.debug_line {
-# RELOC:        R_RISCV_ADD16 <null> 0x0
-# RELOC-NEXT:   R_RISCV_SUB16 <null> 0x0
-# RELOC-NEXT:   R_RISCV_ADD16 <null> 0x0
-# RELOC-NEXT:   R_RISCV_SUB16 <null> 0x0
-# RELOC-NEXT:   R_RISCV_ADD16 <null> 0x0
-# RELOC-NEXT:   R_RISCV_SUB16 <null> 0x0
+# RELOC:        R_RISCV_ADD16 .L0  0x0
+# RELOC-NEXT:   R_RISCV_SUB16 .L0  0x0
+# RELOC-NEXT:   R_RISCV_ADD16 .L0  0x0
+# RELOC-NEXT:   R_RISCV_SUB16 .L0  0x0
+# RELOC-NEXT:   R_RISCV_ADD16 .L0  0x0
+# RELOC-NEXT:   R_RISCV_SUB16 .L0  0x0
 # RELOC:      }
 
 # RELOC:      Hex dump of section '.eh_frame':
diff --git a/llvm/test/MC/M68k/Data/Classes/MxMove_RI.s b/llvm/test/MC/M68k/Data/Classes/MxMove_RI.s
index 091367a68256..2081924d7b17 100644
--- a/llvm/test/MC/M68k/Data/Classes/MxMove_RI.s
+++ b/llvm/test/MC/M68k/Data/Classes/MxMove_RI.s
@@ -9,3 +9,6 @@ move.l	#42, %a1
 ; CHECK:      move.l  #-1, %a1
 ; CHECK-SAME: encoding: [0x22,0x7c,0xff,0xff,0xff,0xff]
 move.l	#-1, %a1
+; CHECK:      moveq  #-17, %d3
+; CHECK-SAME: encoding: [0x76,0xef]
+moveq	#-17, %d3
diff --git a/llvm/test/MC/RISCV/XTHeadVdot-valid.s b/llvm/test/MC/RISCV/XTHeadVdot-valid.s
index 2e00bd1cac3e..ab411dfac730 100644
--- a/llvm/test/MC/RISCV/XTHeadVdot-valid.s
+++ b/llvm/test/MC/RISCV/XTHeadVdot-valid.s
@@ -12,82 +12,82 @@ th.vmaqau.vv v8, v20, v4, v0.t
 # CHECK-INST: th.vmaqau.vv v8, v20, v4, v0.t
 # CHECK-ENCODING: [0x0b,0x64,0x4a,0x88]
 # CHECK-ERROR: instruction requires the following: 'xtheadvdot' (T-Head Vector Extensions for Dot){{$}}
-# CHECK-UNKNOWN: 0b 64 4a 88 <unknown>
+# CHECK-UNKNOWN: 884a640b <unknown>
 
 th.vmaqau.vv v8, v20, v4
 # CHECK-INST: th.vmaqau.vv v8, v20, v4
 # CHECK-ENCODING: [0x0b,0x64,0x4a,0x8a]
 # CHECK-ERROR: instruction requires the following: 'xtheadvdot' (T-Head Vector Extensions for Dot){{$}}
-# CHECK-UNKNOWN: 0b 64 4a 8a <unknown>
+# CHECK-UNKNOWN: 8a4a640b <unknown>
 
 th.vmaqau.vx v8, a0, v4, v0.t
 # CHECK-INST: th.vmaqau.vx v8, a0, v4, v0.t
 # CHECK-ENCODING: [0x0b,0x64,0x45,0x8c]
 # CHECK-ERROR: instruction requires the following: 'xtheadvdot' (T-Head Vector Extensions for Dot){{$}}
-# CHECK-UNKNOWN: 0b 64 45 8c <unknown>
+# CHECK-UNKNOWN: 8c45640b <unknown>
 
 th.vmaqau.vx v8, a0, v4
 # CHECK-INST: th.vmaqau.vx v8, a0, v4
 # CHECK-ENCODING: [0x0b,0x64,0x45,0x8e]
 # CHECK-ERROR: instruction requires the following: 'xtheadvdot' (T-Head Vector Extensions for Dot){{$}}
-# CHECK-UNKNOWN: 0b 64 45 8e <unknown>
+# CHECK-UNKNOWN: 8e45640b <unknown>
 
 th.vmaqa.vv v8, v20, v4, v0.t
 # CHECK-INST: th.vmaqa.vv v8, v20, v4, v0.t
 # CHECK-ENCODING: [0x0b,0x64,0x4a,0x80]
 # CHECK-ERROR: instruction requires the following: 'xtheadvdot' (T-Head Vector Extensions for Dot){{$}}
-# CHECK-UNKNOWN: 0b 64 4a 80 <unknown>
+# CHECK-UNKNOWN: 804a640b <unknown>
 
 th.vmaqa.vv v8, v20, v4
 # CHECK-INST: th.vmaqa.vv v8, v20, v4
 # CHECK-ENCODING: [0x0b,0x64,0x4a,0x82]
 # CHECK-ERROR: instruction requires the following: 'xtheadvdot' (T-Head Vector Extensions for Dot){{$}}
-# CHECK-UNKNOWN: 0b 64 4a 82 <unknown>
+# CHECK-UNKNOWN: 824a640b <unknown>
 
 th.vmaqa.vx v8, a0, v4, v0.t
 # CHECK-INST: th.vmaqa.vx v8, a0, v4, v0.t
 # CHECK-ENCODING: [0x0b,0x64,0x45,0x84]
 # CHECK-ERROR: instruction requires the following: 'xtheadvdot' (T-Head Vector Extensions for Dot){{$}}
-# CHECK-UNKNOWN: 0b 64 45 84 <unknown>
+# CHECK-UNKNOWN: 8445640b <unknown>
 
 th.vmaqa.vx v8, a0, v4
 # CHECK-INST: th.vmaqa.vx v8, a0, v4
 # CHECK-ENCODING: [0x0b,0x64,0x45,0x86]
 # CHECK-ERROR: instruction requires the following: 'xtheadvdot' (T-Head Vector Extensions for Dot){{$}}
-# CHECK-UNKNOWN: 0b 64 45 86 <unknown>
+# CHECK-UNKNOWN: 8645640b <unknown>
 
 th.vmaqasu.vv v8, v20, v4, v0.t
 # CHECK-INST: th.vmaqasu.vv v8, v20, v4, v0.t
 # CHECK-ENCODING: [0x0b,0x64,0x4a,0x90]
 # CHECK-ERROR: instruction requires the following: 'xtheadvdot' (T-Head Vector Extensions for Dot){{$}}
-# CHECK-UNKNOWN: 0b 64 4a 90 <unknown>
+# CHECK-UNKNOWN: 904a640b <unknown>
 
 th.vmaqasu.vv v8, v20, v4
 # CHECK-INST: th.vmaqasu.vv v8, v20, v4
 # CHECK-ENCODING: [0x0b,0x64,0x4a,0x92]
 # CHECK-ERROR: instruction requires the following: 'xtheadvdot' (T-Head Vector Extensions for Dot){{$}}
-# CHECK-UNKNOWN: 0b 64 4a 92 <unknown>
+# CHECK-UNKNOWN: 924a640b <unknown>
 
 th.vmaqasu.vx v8, a0, v4, v0.t
 # CHECK-INST: th.vmaqasu.vx v8, a0, v4, v0.t
 # CHECK-ENCODING: [0x0b,0x64,0x45,0x94]
 # CHECK-ERROR: instruction requires the following: 'xtheadvdot' (T-Head Vector Extensions for Dot){{$}}
-# CHECK-UNKNOWN: 0b 64 45 94 <unknown>
+# CHECK-UNKNOWN: 9445640b <unknown>
 
 th.vmaqasu.vx v8, a0, v4
 # CHECK-INST: th.vmaqasu.vx v8, a0, v4
 # CHECK-ENCODING: [0x0b,0x64,0x45,0x96]
 # CHECK-ERROR: instruction requires the following: 'xtheadvdot' (T-Head Vector Extensions for Dot){{$}}
-# CHECK-UNKNOWN: 0b 64 45 96 <unknown>
+# CHECK-UNKNOWN: 9645640b <unknown>
 
 th.vmaqaus.vx v8, a0, v4, v0.t
 # CHECK-INST: th.vmaqaus.vx v8, a0, v4, v0.t
 # CHECK-ENCODING: [0x0b,0x64,0x45,0x9c]
 # CHECK-ERROR: instruction requires the following: 'xtheadvdot' (T-Head Vector Extensions for Dot){{$}}
-# CHECK-UNKNOWN: 0b 64 45 9c <unknown>
+# CHECK-UNKNOWN: 9c45640b <unknown>
 
 th.vmaqaus.vx v8, a0, v4
 # CHECK-INST: th.vmaqaus.vx v8, a0, v4
 # CHECK-ENCODING: [0x0b,0x64,0x45,0x9e]
 # CHECK-ERROR: instruction requires the following: 'xtheadvdot' (T-Head Vector Extensions for Dot){{$}}
-# CHECK-UNKNOWN: 0b 64 45 9e <unknown>
+# CHECK-UNKNOWN: 9e45640b <unknown>
diff --git a/llvm/test/MC/RISCV/align.s b/llvm/test/MC/RISCV/align.s
index 2eb7186d0de9..32cc071b613c 100644
--- a/llvm/test/MC/RISCV/align.s
+++ b/llvm/test/MC/RISCV/align.s
@@ -98,11 +98,11 @@ test:
 # The behavior is the same as GNU assembler.
 	.p2align 4, 1
 # RELAX-RELOC-NOT: R_RISCV_ALIGN - 0xC
-# RELAX-INST:  01 01
-# RELAX-INST:  01 01
+# RELAX-INST:  0101
+# RELAX-INST:  0101
 # C-OR-ZCA-EXT-RELAX-RELOC-NOT: R_RISCV_ALIGN - 0xE
-# C-OR-ZCA-EXT-RELAX-INST:  01 01
-# C-EXT-INST:  01 01
+# C-OR-ZCA-EXT-RELAX-INST:  0101
+# C-EXT-INST:  0101
 	ret
 # NORELAX-RELOC-NOT: R_RISCV
 # C-OR-ZCA-EXT-NORELAX-RELOC-NOT: R_RISCV
diff --git a/llvm/test/MC/RISCV/attribute-arch.s b/llvm/test/MC/RISCV/attribute-arch.s
index 8835ff22446c..aea27146e370 100644
--- a/llvm/test/MC/RISCV/attribute-arch.s
+++ b/llvm/test/MC/RISCV/attribute-arch.s
@@ -366,7 +366,7 @@
 .attribute arch, "rv32i_zvfbfwma1p0"
 # CHECK: .attribute     5, "rv32i2p1_f2p2_zicsr2p0_zfbfmin1p0_zve32f1p0_zve32x1p0_zvfbfmin1p0_zvfbfwma1p0_zvl32b1p0"
 
-.attribute arch, "rv32izacas1p0"
+.attribute arch, "rv32ia_zacas1p0"
 # CHECK: attribute      5, "rv32i2p1_a2p1_zacas1p0"
 
 .attribute arch, "rv32izalasr0p1"
diff --git a/llvm/test/MC/RISCV/cfi-advance.s b/llvm/test/MC/RISCV/cfi-advance.s
index c4af390be757..b99af38f553a 100644
--- a/llvm/test/MC/RISCV/cfi-advance.s
+++ b/llvm/test/MC/RISCV/cfi-advance.s
@@ -1,13 +1,27 @@
 # RUN: llvm-mc -filetype=obj -triple riscv32 %s -o %t.o  
-# RUN: llvm-readobj -r %t.o | FileCheck -check-prefix=CHECK %s
+# RUN: llvm-readelf -sr %t.o | FileCheck %s
 # RUN: llvm-dwarfdump --debug-frame %t.o 2>&1 \
 # RUN:     | FileCheck -check-prefix=CHECK-DWARFDUMP %s
 
-# CHECK:      .rela.eh_frame {
-# CHECK-NEXT:   0x1C R_RISCV_32_PCREL <null> 0x0
-# CHECK-NEXT:   0x35 R_RISCV_SET6 <null> 0x0
-# CHECK-NEXT:   0x35 R_RISCV_SUB6 <null> 0x0
-# CHECK-NEXT: }
+
+# CHECK:      Relocation section '.rela.text1' at offset {{.*}} contains 1 entries:
+# CHECK-NEXT:  Offset     Info    Type                Sym. Value  Symbol's Name + Addend
+# CHECK-NEXT: 00000000  00000313 R_RISCV_CALL_PLT       00000004   .L0 + 0
+# CHECK-EMPTY:
+# CHECK-NEXT: Relocation section '.rela.eh_frame' at offset {{.*}} contains 3 entries:
+# CHECK:       Offset     Info    Type                Sym. Value  Symbol's Name + Addend
+# CHECK-NEXT: 0000001c  00000139 R_RISCV_32_PCREL       00000000   .L0 + 0
+# CHECK-NEXT: 00000035  00000b35 R_RISCV_SET6           00010178   .L0 + 0
+# CHECK-NEXT: 00000035  00000934 R_RISCV_SUB6           0001016e   .L0 + 0
+# CHECK-EMPTY:
+# CHECK:      Symbol table '.symtab' contains 15 entries:
+# CHECK-NEXT:    Num:    Value  Size Type    Bind   Vis       Ndx Name
+# CHECK-NEXT:      0: 00000000     0 NOTYPE  LOCAL  DEFAULT   UND
+# CHECK-NEXT:      1: 00000000     0 NOTYPE  LOCAL  DEFAULT     2 .L0 {{$}}
+# CHECK:           3: 00000004     0 NOTYPE  LOCAL  DEFAULT     2 .L0{{$}}
+# CHECK:           9: 0001016e     0 NOTYPE  LOCAL  DEFAULT     2 .L0 {{$}}
+# CHECK:          11: 00010178     0 NOTYPE  LOCAL  DEFAULT     2 .L0 {{$}}
+
 # CHECK-DWARFDUMP: DW_CFA_advance_loc1: 104
 # CHECK-DWARFDUMP-NEXT: DW_CFA_def_cfa_offset: +8
 # CHECK-DWARFDUMP-NEXT: DW_CFA_advance_loc2: 259
@@ -23,6 +37,9 @@
 test:
         .cfi_startproc
         nop
+## This looks similar to fake label names ".L0 ". Even if this is ".L0 ",
+## the assembler will not conflate it with fake labels.
+.L0:
         .zero 100, 0x90
         .cfi_def_cfa_offset 8
         nop
@@ -36,3 +53,6 @@ test:
         .cfi_def_cfa_offset 8
         nop
         .cfi_endproc
+
+.section .text1,"ax"
+call .L0
diff --git a/llvm/test/MC/RISCV/compress-cjal.s b/llvm/test/MC/RISCV/compress-cjal.s
index 31b9c30c2b01..d55586b005c7 100644
--- a/llvm/test/MC/RISCV/compress-cjal.s
+++ b/llvm/test/MC/RISCV/compress-cjal.s
@@ -11,7 +11,7 @@
 
 # c.jal is an rv32 only instruction.
 jal ra, 2046
-# CHECK-BYTES: fd 2f
+# CHECK-BYTES: 2ffd
 # CHECK-ALIASOBJ: jal 0x7fe
 # CHECK-ALIAS: jal 2046
 # CHECK-INST: c.jal 2046
diff --git a/llvm/test/MC/RISCV/compress-rv32d.s b/llvm/test/MC/RISCV/compress-rv32d.s
index bebc78ef8690..c41a08892862 100644
--- a/llvm/test/MC/RISCV/compress-rv32d.s
+++ b/llvm/test/MC/RISCV/compress-rv32d.s
@@ -43,22 +43,22 @@
 # Tests double precision floating point instructions available in rv32 and in rv64.
 
 fld ft0, 64(sp)
-# CHECK-BYTES: 06 20
+# CHECK-BYTES: 2006
 # CHECK-ALIAS: fld ft0, 64(sp)
 # CHECK-INST: c.fldsp ft0, 64(sp)
 # CHECK: # encoding:  [0x06,0x20]
 fsd ft0, 64(sp)
-# CHECK-BYTES: 82 a0
+# CHECK-BYTES: a082
 # CHECK-ALIAS: fsd ft0, 64(sp)
 # CHECK-INST: c.fsdsp ft0, 64(sp)
 # CHECK: # encoding:  [0x82,0xa0]
 fld fs0, 248(s0)
-# CHECK-BYTES: 60 3c
+# CHECK-BYTES: 3c60
 # CHECK-ALIAS: fld fs0, 248(s0)
 # CHECK-INST: c.fld fs0, 248(s0)
 # CHECK: # encoding:  [0x60,0x3c]
 fsd fs0, 248(s0)
-# CHECK-BYTES: 60 bc
+# CHECK-BYTES: bc60
 # CHECK-ALIAS: fsd fs0, 248(s0)
 # CHECK-INST: c.fsd fs0, 248(s0)
 # CHECK: # encoding:  [0x60,0xbc]
diff --git a/llvm/test/MC/RISCV/compress-rv32f.s b/llvm/test/MC/RISCV/compress-rv32f.s
index 3f0c69fb9893..afe15c598bb6 100644
--- a/llvm/test/MC/RISCV/compress-rv32f.s
+++ b/llvm/test/MC/RISCV/compress-rv32f.s
@@ -21,22 +21,22 @@
 
 # Instructions that are 32 bit only.
 flw ft0, 124(sp)
-# CHECK-BYTES: 76 70
+# CHECK-BYTES: 7076
 # CHECK-ALIAS: flw     ft0, 124(sp)
 # CHECK-INST: c.flwsp ft0, 124(sp)
 # CHECK:  # encoding: [0x76,0x70]
 fsw ft0, 124(sp)
-# CHECK-BYTES: 82 fe
+# CHECK-BYTES: fe82
 # CHECK-ALIAS: fsw ft0, 124(sp)
 # CHECK-INST: c.fswsp ft0, 124(sp)
 # CHECK:  # encoding: [0x82,0xfe]
 flw fs0, 124(s0)
-# CHECK-BYTES: 60 7c
+# CHECK-BYTES: 7c60
 # CHECK-ALIAS: flw fs0, 124(s0)
 # CHECK-INST: c.flw fs0, 124(s0)
 # CHECK:  # encoding:  [0x60,0x7c]
 fsw fs0, 124(s0)
-# CHECK-BYTES: 60 fc
+# CHECK-BYTES: fc60
 # CHECK-ALIAS: fsw fs0, 124(s0)
 # CHECK-INST: c.fsw fs0, 124(s0)
 # CHECK:  # encoding:  [0x60,0xfc]
diff --git a/llvm/test/MC/RISCV/compress-rv32i.s b/llvm/test/MC/RISCV/compress-rv32i.s
index b4fd72a0f81c..a75bea32ac0c 100644
--- a/llvm/test/MC/RISCV/compress-rv32i.s
+++ b/llvm/test/MC/RISCV/compress-rv32i.s
@@ -20,121 +20,121 @@
 # RUN:   | llvm-objdump  --triple=riscv64 --mattr=+c --no-print-imm-hex -d -M no-aliases - \
 # RUN:   | FileCheck -check-prefixes=CHECK-BYTES,CHECK-INST,CHECK-INSTOBJ64 %s
 
-# CHECK-BYTES: 2e 85
+# CHECK-BYTES: 852e
 # CHECK-ALIAS: mv a0, a1
 # CHECK-INST: c.mv a0, a1
 # CHECK: # encoding:  [0x2e,0x85]
 addi a0, a1, 0
 
-# CHECK-BYTES: e0 1f
+# CHECK-BYTES: 1fe0
 # CHECK-ALIAS: addi s0, sp, 1020
 # CHECK-INST: c.addi4spn s0, sp, 1020
 # CHECK: # encoding:  [0xe0,0x1f]
 addi s0, sp, 1020
 
-# CHECK-BYTES: e0 5f
+# CHECK-BYTES: 5fe0
 # CHECK-ALIAS: lw s0, 124(a5)
 # CHECK-INST: c.lw s0, 124(a5)
 # CHECK: # encoding: [0xe0,0x5f]
 lw s0, 124(a5)
 
-# CHECK-BYTES: e0 df
+# CHECK-BYTES: dfe0
 # CHECK-ALIAS: sw s0, 124(a5)
 # CHECK-INST: c.sw s0, 124(a5)
 # CHECK: # encoding: [0xe0,0xdf]
 sw s0, 124(a5)
 
-# CHECK-BYTES: 01 00
+# CHECK-BYTES: 0001
 # CHECK-ALIAS: nop
 # CHECK-INST: c.nop
 # CHECK: # encoding: [0x01,0x00]
 nop
 
-# CHECK-BYTES: 81 10
+# CHECK-BYTES: 1081
 # CHECK-ALIAS: addi ra, ra, -32
 # CHECK-INST: c.addi ra, -32
 # CHECK: # encoding:  [0x81,0x10]
 addi ra, ra, -32
 
-# CHECK-BYTES: 85 50
+# CHECK-BYTES: 5085
 # CHECK-ALIAS: li ra, -31
 # CHECK-INST: c.li ra, -31
 # CHECK: # encoding: [0x85,0x50]
 li ra, -31
 
-# CHECK-BYTES: 39 71
+# CHECK-BYTES: 7139
 # CHECK-ALIAS: addi sp, sp, -64
 # CHECK-INST: c.addi16sp sp, -64
 # CHECK:  # encoding: [0x39,0x71]
 addi sp, sp, -64
 
-# CHECK-BYTES: fd 61
+# CHECK-BYTES: 61fd
 # CHECK-ALIAS: lui gp, 31
 # CHECK-INST: c.lui gp, 31
 # CHECK: # encoding:  [0xfd,0x61]
 lui gp, 31
 
-# CHECK-BYTES: 7d 80
+# CHECK-BYTES: 807d
 # CHECK-ALIAS: srli s0, s0, 31
 # CHECK-INST: c.srli s0, 31
 # CHECK: # encoding:  [0x7d,0x80]
 srli s0, s0, 31
 
-# CHECK-BYTES: 7d 84
+# CHECK-BYTES: 847d
 # CHECK-ALIAS: srai s0, s0, 31
 # CHECK-INST: c.srai s0, 31
 # CHECK: # encoding: [0x7d,0x84]
 srai s0, s0, 31
 
-# CHECK-BYTES: 7d 88
+# CHECK-BYTES: 887d
 # CHECK-ALIAS: andi s0, s0, 31
 # CHECK-INST: c.andi s0, 31
 # CHECK: # encoding: [0x7d,0x88]
 andi s0, s0, 31
 
-# CHECK-BYTES: 1d 8c
+# CHECK-BYTES: 8c1d
 # CHECK-ALIAS: sub s0, s0, a5
 # CHECK-INST: c.sub s0, a5
 # CHECK: # encoding: [0x1d,0x8c]
 sub s0, s0, a5
 
-# CHECK-BYTES: 3d 8c
+# CHECK-BYTES: 8c3d
 # CHECK-ALIAS: xor s0, s0, a5
 # CHECK-INST: c.xor s0, a5
 # CHECK: # encoding: [0x3d,0x8c]
 xor s0, s0, a5
 
-# CHECK-BYTES: 3d 8c
+# CHECK-BYTES: 8c3d
 # CHECK-ALIAS: xor s0, s0, a5
 # CHECK-INST: c.xor s0, a5
 # CHECK: # encoding: [0x3d,0x8c]
 xor s0, a5, s0
 
-# CHECK-BYTES: 5d 8c
+# CHECK-BYTES: 8c5d
 # CHECK-ALIAS: or s0, s0, a5
 # CHECK-INST: c.or s0, a5
 # CHECK: # encoding:  [0x5d,0x8c]
 or s0, s0, a5
 
-# CHECK-BYTES: 45 8c
+# CHECK-BYTES: 8c45
 # CHECK-ALIAS: or s0, s0, s1
 # CHECK-INST: c.or s0, s1
 # CHECK:  # encoding: [0x45,0x8c]
 or  s0, s1, s0
 
-# CHECK-BYTES: 7d 8c
+# CHECK-BYTES: 8c7d
 # CHECK-ALIAS: and s0, s0, a5
 # CHECK-INST: c.and s0, a5
 # CHECK: # encoding: [0x7d,0x8c]
 and s0, s0, a5
 
-# CHECK-BYTES: 7d 8c
+# CHECK-BYTES: 8c7d
 # CHECK-ALIAS: and s0, s0, a5
 # CHECK-INST: c.and s0, a5
 # CHECK: # encoding: [0x7d,0x8c]
 and s0, a5, s0
 
-# CHECK-BYTES: 01 b0
+# CHECK-BYTES: b001
 # CHECK-ALIASASM: j -2048
 # CHECK-ALIASOBJ32: j 0xfffff826
 # CHECK-ALIASOBJ64: j 0xfffffffffffff826
@@ -144,7 +144,7 @@ and s0, a5, s0
 # CHECK:  # encoding: [0x01,0xb0]
 jal zero, -2048
 
-# CHECK-BYTES: 01 d0
+# CHECK-BYTES: d001
 # CHECK-ALIASASM: beqz s0, -256
 # CHECK-ALIASOBJ32: beqz s0, 0xffffff28
 # CHECK-ALIASOBJ64: beqz s0, 0xffffffffffffff28
@@ -154,7 +154,7 @@ jal zero, -2048
 # CHECK: # encoding: [0x01,0xd0]
 beq s0, zero, -256
 
-# CHECK-BYTES: 01 d0
+# CHECK-BYTES: d001
 # CHECK-ALIASASM: beqz s0, -256
 # CHECK-ALIASOBJ32: beqz s0, 0xffffff2a
 # CHECK-ALIASOBJ64: beqz s0, 0xffffffffffffff2a
@@ -164,7 +164,7 @@ beq s0, zero, -256
 # CHECK: # encoding: [0x01,0xd0]
 beq zero, s0, -256
 
-# CHECK-BYTES: 7d ec
+# CHECK-BYTES: ec7d
 # CHECK-ALIASASM: bnez s0, 254
 # CHECK-ALIASOBJ32: bnez s0, 0x12a
 # CHECK-ALIASOBJ64: bnez s0, 0x12a
@@ -174,7 +174,7 @@ beq zero, s0, -256
 # CHECK: # encoding: [0x7d,0xec]
 bne s0, zero, 254
 
-# CHECK-BYTES: 7d ec
+# CHECK-BYTES: ec7d
 # CHECK-ALIASASM: bnez s0, 254
 # CHECK-ALIASOBJ32: bnez s0, 0x12c
 # CHECK-ALIASOBJ64: bnez s0, 0x12c
@@ -184,67 +184,67 @@ bne s0, zero, 254
 # CHECK: # encoding: [0x7d,0xec]
 bne zero, s0, 254
 
-# CHECK-BYTES: 7e 04
+# CHECK-BYTES: 047e
 # CHECK-ALIAS: slli s0, s0, 31
 # CHECK-INST: c.slli s0, 31
 # CHECK: # encoding:  [0x7e,0x04]
 slli s0, s0, 31
 
-# CHECK-BYTES: fe 50
+# CHECK-BYTES: 50fe
 # CHECK-ALIAS: lw ra, 252(sp)
 # CHECK-INST: c.lwsp  ra, 252(sp)
 # CHECK: # encoding:  [0xfe,0x50]
 lw ra, 252(sp)
 
-# CHECK-BYTES: 82 80
+# CHECK-BYTES: 8082
 # CHECK-ALIAS: ret
 # CHECK-INST: c.jr ra
 # CHECK: # encoding:  [0x82,0x80]
 jalr zero, 0(ra)
 
-# CHECK-BYTES: 92 80
+# CHECK-BYTES: 8092
 # CHECK-ALIAS: mv ra, tp
 # CHECK-INST: c.mv ra, tp
 # CHECK:  # encoding: [0x92,0x80]
 add ra, zero, tp
 
-# CHECK-BYTES: 92 80
+# CHECK-BYTES: 8092
 # CHECK-ALIAS: mv ra, tp
 # CHECK-INST: c.mv ra, tp
 # CHECK:  # encoding: [0x92,0x80]
 add ra, tp, zero
 
-# CHECK-BYTES: 02 90
+# CHECK-BYTES: 9002
 # CHECK-ALIAS: ebreak
 # CHECK-INST: c.ebreak
 # CHECK: # encoding: [0x02,0x90]
 ebreak
 
-# CHECK-BYTES: 02 94
+# CHECK-BYTES: 9402
 # CHECK-ALIAS: jalr s0
 # CHECK-INST: c.jalr s0
 # CHECK: # encoding: [0x02,0x94]
 jalr ra, 0(s0)
 
-# CHECK-BYTES: 3e 94
+# CHECK-BYTES: 943e
 # CHECK-ALIAS: add s0, s0, a5
 # CHECK-INST: c.add s0, a5
 # CHECK: # encoding:  [0x3e,0x94]
 add s0, a5, s0
 
-# CHECK-BYTES: 3e 94
+# CHECK-BYTES: 943e
 # CHECK-ALIAS: add s0, s0, a5
 # CHECK-INST: c.add s0, a5
 # CHECK: # encoding:  [0x3e,0x94]
 add s0, s0, a5
 
-# CHECK-BYTES: 82 df
+# CHECK-BYTES: df82
 # CHECK-ALIAS: sw zero, 252(sp)
 # CHECK-INST: c.swsp zero, 252(sp)
 # CHECK: # encoding: [0x82,0xdf]
 sw zero, 252(sp)
 
-# CHECK-BYTES: 00 00
+# CHECK-BYTES: 0000
 # CHECK-ALIAS: unimp
 # CHECK-INST: c.unimp
 # CHECK: # encoding: [0x00,0x00]
diff --git a/llvm/test/MC/RISCV/compress-rv64i.s b/llvm/test/MC/RISCV/compress-rv64i.s
index 55d24f0d41c0..ab5b24307cd1 100644
--- a/llvm/test/MC/RISCV/compress-rv64i.s
+++ b/llvm/test/MC/RISCV/compress-rv64i.s
@@ -11,49 +11,49 @@
 
 # Tests compressed instructions available in rv64 and not in rv32.
 
-# CHECK-BYTES: e0 7f
+# CHECK-BYTES: 7fe0
 # CHECK-ALIAS: ld s0, 248(a5)
 # CHECK-INST: c.ld s0, 248(a5)
 # CHECK: # encoding: [0xe0,0x7f]
 ld s0, 248(a5)
 
-# CHECK-BYTES: a0 e3
+# CHECK-BYTES: e3a0
 # CHECK-ALIAS: sd s0, 64(a5)
 # CHECK-INST: c.sd s0, 64(a5)
 # CHECK: # encoding: [0xa0,0xe3]
 sd s0, 64(a5)
 
-# CHECK-BYTES: 7d 22
+# CHECK-BYTES: 227d
 # CHEACK-ALIAS: addiw tp, tp, 31
 # CHECK-INST: c.addiw  tp, 31
 # CHECK: # encoding: [0x7d,0x22]
 addiw tp, tp, 31
 
-# CHECK-BYTES: 1d 9c
+# CHECK-BYTES: 9c1d
 # CHEACK-ALIAS: subw s0, s0, a5
 # CHECK-INST: c.subw s0, a5
 # CHECK: # encoding:  [0x1d,0x9c]
 subw s0, s0, a5
 
-# CHECK-BYTES: 3d 9c
+# CHECK-BYTES: 9c3d
 # CHECK-ALIAS: addw s0, s0, a5
 # CHECK-INST: c.addw s0, a5
 # CHECK: # encoding: [0x3d,0x9c]
 addw s0, s0, a5
 
-# CHECK-BYTES: 3d 9c
+# CHECK-BYTES: 9c3d
 # CHECK-ALIAS: addw s0, s0, a5
 # CHECK-INST: c.addw s0, a5
 # CHECK: # encoding: [0x3d,0x9c]
 addw s0, a5, s0
 
-# CHECK-BYTES: ee 70
+# CHECK-BYTES: 70ee
 # CHECK-ALIAS: ld ra, 248(sp)
 # CHECK-INST: c.ldsp ra, 248(sp)
 # CHECK: # encoding:  [0xee,0x70]
 ld ra, 248(sp)
 
-# CHECK-BYTES: a2 e0
+# CHECK-BYTES: e0a2
 # CHECK-ALIAS: sd s0, 64(sp)
 # CHECK-INST: c.sdsp s0, 64(sp)
 # CHECK: # encoding: [0xa2,0xe0]
diff --git a/llvm/test/MC/RISCV/fde-reloc.s b/llvm/test/MC/RISCV/fde-reloc.s
index 1db8929e0747..81ec426c8b61 100644
--- a/llvm/test/MC/RISCV/fde-reloc.s
+++ b/llvm/test/MC/RISCV/fde-reloc.s
@@ -12,7 +12,7 @@ func:
 	.cfi_endproc
 
 # CHECK:   Section (4) .rela.eh_frame {
-# CHECK-NEXT:   0x1C R_RISCV_32_PCREL <null> 0x0
+# CHECK-NEXT:   0x1C R_RISCV_32_PCREL .L0  0x0
 # CHECK-NEXT: }
 # CHECK:      Hex dump of section '.eh_frame':
 # CHECK-NEXT: 0x00000000 10000000 00000000 017a5200 017c0101
diff --git a/llvm/test/MC/RISCV/large-instructions.s b/llvm/test/MC/RISCV/large-instructions.s
new file mode 100644
index 000000000000..b50dbde17d38
--- /dev/null
+++ b/llvm/test/MC/RISCV/large-instructions.s
@@ -0,0 +1,29 @@
+# RUN: llvm-mc -filetype=obj -triple riscv32 < %s \
+# RUN:     | llvm-objdump -d - | FileCheck %s
+
+# CHECK: 011f 4523 8967 <unknown>
+.byte 0x1f, 0x01, 0x23, 0x45, 0x67, 0x89
+
+# CHECK: 4523013f cdab8967 <unknown>
+.byte 0x3f, 0x01, 0x23, 0x45, 0x67, 0x89, 0xab, 0xcd
+
+# CHECK: 007f 4523 8967 cdab feef <unknown>
+.byte 0x7f, 0x00, 0x23, 0x45, 0x67, 0x89, 0xab, 0xcd, 0xef, 0xfe
+
+# CHECK: 4523107f cdab8967 badcfeef <unknown>
+.byte 0x7f, 0x10, 0x23, 0x45, 0x67, 0x89, 0xab, 0xcd, 0xef, 0xfe, 0xdc, 0xba
+
+# CHECK: 207f 4523 8967 cdab feef badc 7698 <unknown>
+.byte 0x7f, 0x20, 0x23, 0x45, 0x67, 0x89, 0xab, 0xcd, 0xef, 0xfe, 0xdc, 0xba, 0x98, 0x76
+
+# CHECK: 4523307f cdab8967 badcfeef 32547698 <unknown>
+.byte 0x7f, 0x30, 0x23, 0x45, 0x67, 0x89, 0xab, 0xcd, 0xef, 0xfe, 0xdc, 0xba, 0x98, 0x76, 0x54, 0x32
+
+# CHECK: 407f 4523 8967 cdab feef badc 7698 3254 1210 <unknown>
+.byte 0x7f, 0x40, 0x23, 0x45, 0x67, 0x89, 0xab, 0xcd, 0xef, 0xfe, 0xdc, 0xba, 0x98, 0x76, 0x54, 0x32, 0x10, 0x12
+
+# CHECK: 4523507f cdab8967 badcfeef 32547698 56341210 <unknown>
+.byte 0x7f, 0x50, 0x23, 0x45, 0x67, 0x89, 0xab, 0xcd, 0xef, 0xfe, 0xdc, 0xba, 0x98, 0x76, 0x54, 0x32, 0x10, 0x12, 0x34, 0x56
+
+# CHECK: 607f 4523 8967 cdab feef badc 7698 3254 1210 5634 9a78 <unknown>
+.byte 0x7f, 0x60, 0x23, 0x45, 0x67, 0x89, 0xab, 0xcd, 0xef, 0xfe, 0xdc, 0xba, 0x98, 0x76, 0x54, 0x32, 0x10, 0x12, 0x34, 0x56, 0x78, 0x9a
diff --git a/llvm/test/MC/RISCV/nop-slide.s b/llvm/test/MC/RISCV/nop-slide.s
index f280d6e521e3..4dc888b3ba77 100644
--- a/llvm/test/MC/RISCV/nop-slide.s
+++ b/llvm/test/MC/RISCV/nop-slide.s
@@ -10,18 +10,18 @@
 auipc a0, 0
 
 # CHECK-RVC-NORELAX: 0000000000000000 <.text>:
-# CHECK-RVC-NORELAX-NEXT: 0: 00 00        	unimp
-# CHECK-RVC-NORELAX-NEXT: 2: 01 00        	nop
-# CHECK-RVC-NORELAX-NEXT: 4: 17 05 00 00  	auipc	a0, 0x0
+# CHECK-RVC-NORELAX-NEXT: 0: 0000      	unimp
+# CHECK-RVC-NORELAX-NEXT: 2: 0001      	nop
+# CHECK-RVC-NORELAX-NEXT: 4: 00000517  	auipc	a0, 0x0
 
 # CHECK-RVC-RELAX: 0000000000000000 <.text>:
-# CHECK-RVC-RELAX-NEXT:   0: 01 00        	nop
-# CHECK-RVC-RELAX-NEXT:   2: 00 01        	addi	s0, sp, 0x80
-# CHECK-RVC-RELAX-NEXT:   4: 00 17        	addi	s0, sp, 0x3a0
-# CHECK-RVC-RELAX-NEXT:   6: 05 00        	c.nop	0x1
-# CHECK-RVC-RELAX-NEXT:   8: 00           	<unknown>
+# CHECK-RVC-RELAX-NEXT:   0: 0001      	nop
+# CHECK-RVC-RELAX-NEXT:   2: 0100      	addi	s0, sp, 0x80
+# CHECK-RVC-RELAX-NEXT:   4: 1700      	addi	s0, sp, 0x3a0
+# CHECK-RVC-RELAX-NEXT:   6: 0005      	c.nop	0x1
+# CHECK-RVC-RELAX-NEXT:   8: 00        	<unknown>
 
 # CHECK: 0000000000000000 <.text>:
-# CHECK-NEXT: 0: 00 00        	<unknown>
-# CHECK-NEXT: 2: 00 00        	<unknown>
-# CHECK-NEXT: 4: 17 05 00 00  	auipc	a0, 0x0
+# CHECK-NEXT: 0: 0000      	<unknown>
+# CHECK-NEXT: 2: 0000      	<unknown>
+# CHECK-NEXT: 4: 00000517  	auipc	a0, 0x0
diff --git a/llvm/test/MC/RISCV/option-pushpop.s b/llvm/test/MC/RISCV/option-pushpop.s
index c830d16e590b..9c61b5dab5f3 100644
--- a/llvm/test/MC/RISCV/option-pushpop.s
+++ b/llvm/test/MC/RISCV/option-pushpop.s
@@ -25,7 +25,7 @@
 call foo
 
 # CHECK-INST: addi s0, sp, 1020
-# CHECK-BYTES: 13 04 c1 3f
+# CHECK-BYTES: 3fc10413
 # CHECK-ALIAS: addi s0, sp, 1020
 addi s0, sp, 1020
 
@@ -45,14 +45,14 @@ call bar
 .option rvc
 # CHECK-INST: .option rvc
 # CHECK-INST: c.addi4spn s0, sp, 1020
-# CHECK-BYTES: e0 1f
+# CHECK-BYTES: 1fe0
 # CHECK-ALIAS: addi s0, sp, 1020
 addi s0, sp, 1020
 
 .option pop     # Pop relax=true, rvc=false
 # CHECK-INST: .option pop
 # CHECK-INST: addi s0, sp, 1020
-# CHECK-BYTES: 13 04 c1 3f
+# CHECK-BYTES: 3fc10413
 # CHECK-ALIAS: addi s0, sp, 1020
 addi s0, sp, 1020
 
@@ -69,7 +69,7 @@ call bar
 call baz
 
 # CHECK-INST: addi s0, sp, 1020
-# CHECK-BYTES: 13 04 c1 3f
+# CHECK-BYTES: 3fc10413
 # CHECK-ALIAS: addi s0, sp, 1020
 addi s0, sp, 1020
 
diff --git a/llvm/test/MC/RISCV/option-rvc.s b/llvm/test/MC/RISCV/option-rvc.s
index 00c8ea167bcd..894fbab562d7 100644
--- a/llvm/test/MC/RISCV/option-rvc.s
+++ b/llvm/test/MC/RISCV/option-rvc.s
@@ -20,13 +20,13 @@
 # RUN:   | llvm-objdump  --triple=riscv64 --mattr=+c --no-print-imm-hex -d -M no-aliases - \
 # RUN:   | FileCheck -check-prefixes=CHECK-BYTES,CHECK-INST %s
 
-# CHECK-BYTES: 13 85 05 00
+# CHECK-BYTES: 00058513
 # CHECK-ALIAS: mv a0, a1
 # CHECK-INST: addi a0, a1, 0
 # CHECK: # encoding:  [0x13,0x85,0x05,0x00]
 addi a0, a1, 0
 
-# CHECK-BYTES: 13 04 c1 3f
+# CHECK-BYTES: 3fc10413
 # CHECK-ALIAS: addi s0, sp, 1020
 # CHECK-INST: addi s0, sp, 1020
 # CHECK: # encoding:  [0x13,0x04,0xc1,0x3f]
@@ -35,13 +35,13 @@ addi s0, sp, 1020
 
 # CHECK: .option rvc
 .option rvc
-# CHECK-BYTES: 2e 85
+# CHECK-BYTES: 852e
 # CHECK-ALIAS: mv a0, a1
 # CHECK-INST: c.mv a0, a1
 # CHECK: # encoding:  [0x2e,0x85]
 addi a0, a1, 0
 
-# CHECK-BYTES: e0 1f
+# CHECK-BYTES: 1fe0
 # CHECK-ALIAS: addi s0, sp, 1020
 # CHECK-INST: c.addi4spn s0, sp, 1020
 # CHECK: # encoding:  [0xe0,0x1f]
@@ -49,13 +49,13 @@ addi s0, sp, 1020
 
 # CHECK: .option norvc
 .option norvc
-# CHECK-BYTES: 13 85 05 00
+# CHECK-BYTES: 00058513
 # CHECK-ALIAS: mv a0, a1
 # CHECK-INST: addi a0, a1, 0
 # CHECK: # encoding:  [0x13,0x85,0x05,0x00]
 addi a0, a1, 0
 
-# CHECK-BYTES: 13 04 c1 3f
+# CHECK-BYTES: 3fc10413
 # CHECK-ALIAS: addi s0, sp, 1020
 # CHECK-INST: addi s0, sp, 1020
 # CHECK: # encoding:  [0x13,0x04,0xc1,0x3f]
@@ -63,13 +63,13 @@ addi s0, sp, 1020
 
 # CHECK: .option rvc
 .option rvc
-# CHECK-BYTES: 2e 85
+# CHECK-BYTES: 852e
 # CHECK-ALIAS: mv a0, a1
 # CHECK-INST: c.mv a0, a1
 # CHECK: # encoding:  [0x2e,0x85]
 addi a0, a1, 0
 
-# CHECK-BYTES: e0 1f
+# CHECK-BYTES: 1fe0
 # CHECK-ALIAS: addi s0, sp, 1020
 # CHECK-INST: c.addi4spn s0, sp, 1020
 # CHECK: # encoding:  [0xe0,0x1f]
@@ -77,13 +77,13 @@ addi s0, sp, 1020
 
 # CHECK: .option norvc
 .option norvc
-# CHECK-BYTES: 13 85 05 00
+# CHECK-BYTES: 00058513
 # CHECK-ALIAS: mv a0, a1
 # CHECK-INST: addi a0, a1, 0
 # CHECK: # encoding:  [0x13,0x85,0x05,0x00]
 addi a0, a1, 0
 
-# CHECK-BYTES: 13 04 c1 3f
+# CHECK-BYTES: 3fc10413
 # CHECK-ALIAS: addi s0, sp, 1020
 # CHECK-INST: addi s0, sp, 1020
 # CHECK: # encoding:  [0x13,0x04,0xc1,0x3f]
diff --git a/llvm/test/MC/RISCV/rv32dc-valid.s b/llvm/test/MC/RISCV/rv32dc-valid.s
index 4e30fadac34b..201aee545d4a 100644
--- a/llvm/test/MC/RISCV/rv32dc-valid.s
+++ b/llvm/test/MC/RISCV/rv32dc-valid.s
@@ -12,9 +12,6 @@
 # RUN: not llvm-mc -triple riscv32 -mattr=+c \
 # RUN:     -riscv-no-aliases -show-encoding < %s 2>&1 \
 # RUN:     | FileCheck -check-prefixes=CHECK-NO-EXT-D %s
-# RUN: not llvm-mc -triple riscv32 -mattr=+zcd \
-# RUN:     -riscv-no-aliases -show-encoding < %s 2>&1 \
-# RUN:     | FileCheck -check-prefixes=CHECK-NO-EXT-D %s
 # RUN: not llvm-mc -triple riscv32 -riscv-no-aliases -show-encoding < %s 2>&1 \
 # RUN:     | FileCheck -check-prefixes=CHECK-NO-EXT-DC %s
 
diff --git a/llvm/test/MC/RISCV/rv32e-invalid.s b/llvm/test/MC/RISCV/rv32e-invalid.s
index 9c19d3f40bcf..95dc156f250a 100644
--- a/llvm/test/MC/RISCV/rv32e-invalid.s
+++ b/llvm/test/MC/RISCV/rv32e-invalid.s
@@ -11,100 +11,100 @@
 # are rejected for RV32E/RV64E, when both assembling and disassembling.
 
 
-# CHECK-DIS: 37 18 00 00 <unknown>
+# CHECK-DIS: 00001837 <unknown>
 # CHECK: :[[@LINE+1]]:5: error: invalid operand for instruction
 lui x16, 1
-# CHECK-DIS: b7 28 00 00 <unknown>
+# CHECK-DIS: 000028b7 <unknown>
 # CHECK: :[[@LINE+1]]:5: error: invalid operand for instruction
 lui x17, 2
-# CHECK-DIS: 37 39 00 00 <unknown>
+# CHECK-DIS: 00003937 <unknown>
 # CHECK: :[[@LINE+1]]:5: error: invalid operand for instruction
 lui x18, 3
-# CHECK-DIS: b7 49 00 00 <unknown>
+# CHECK-DIS: 000049b7 <unknown>
 # CHECK: :[[@LINE+1]]:5: error: invalid operand for instruction
 lui x19, 4
-# CHECK-DIS: 37 5a 00 00 <unknown>
+# CHECK-DIS: 00005a37 <unknown>
 # CHECK: :[[@LINE+1]]:5: error: invalid operand for instruction
 lui x20, 5
-# CHECK-DIS: b7 6a 00 00 <unknown>
+# CHECK-DIS: 00006ab7 <unknown>
 # CHECK: :[[@LINE+1]]:5: error: invalid operand for instruction
 lui x21, 6
-# CHECK-DIS: 37 7b 00 00 <unknown>
+# CHECK-DIS: 00007b37 <unknown>
 # CHECK: :[[@LINE+1]]:5: error: invalid operand for instruction
 lui x22, 7
-# CHECK-DIS: b7 8b 00 00 <unknown>
+# CHECK-DIS: 00008bb7 <unknown>
 # CHECK: :[[@LINE+1]]:5: error: invalid operand for instruction
 lui x23, 8
-# CHECK-DIS: 37 9c 00 00 <unknown>
+# CHECK-DIS: 00009c37 <unknown>
 # CHECK: :[[@LINE+1]]:5: error: invalid operand for instruction
 lui x24, 9
-# CHECK-DIS: b7 ac 00 00 <unknown>
+# CHECK-DIS: 0000acb7 <unknown>
 # CHECK: :[[@LINE+1]]:5: error: invalid operand for instruction
 lui x25, 10
-# CHECK-DIS: 37 bd 00 00 <unknown>
+# CHECK-DIS: 0000bd37 <unknown>
 # CHECK: :[[@LINE+1]]:5: error: invalid operand for instruction
 lui x26, 11
-# CHECK-DIS: b7 cd 00 00 <unknown>
+# CHECK-DIS: 0000cdb7 <unknown>
 # CHECK: :[[@LINE+1]]:5: error: invalid operand for instruction
 lui x27, 12
-# CHECK-DIS: 37 de 00 00 <unknown>
+# CHECK-DIS: 0000de37 <unknown>
 # CHECK: :[[@LINE+1]]:5: error: invalid operand for instruction
 lui x28, 13
-# CHECK-DIS: b7 ee 00 00 <unknown>
+# CHECK-DIS: 0000eeb7 <unknown>
 # CHECK: :[[@LINE+1]]:5: error: invalid operand for instruction
 lui x29, 14
-# CHECK-DIS: 37 ff 00 00 <unknown>
+# CHECK-DIS: 0000ff37 <unknown>
 # CHECK: :[[@LINE+1]]:5: error: invalid operand for instruction
 lui x30, 15
-# CHECK-DIS: b7 0f 01 00 <unknown>
+# CHECK-DIS: 00010fb7 <unknown>
 # CHECK: :[[@LINE+1]]:5: error: invalid operand for instruction
 lui x31, 16
 
-# CHECK-DIS: 17 18 01 00 <unknown>
+# CHECK-DIS: 00011817 <unknown>
 # CHECK: :[[@LINE+1]]:7: error: invalid operand for instruction
 auipc a6, 17
-# CHECK-DIS: 97 28 01 00 <unknown>
+# CHECK-DIS: 00012897 <unknown>
 # CHECK: :[[@LINE+1]]:7: error: invalid operand for instruction
 auipc a7, 18
-# CHECK-DIS: 17 39 01 00 <unknown>
+# CHECK-DIS: 00013917 <unknown>
 # CHECK: :[[@LINE+1]]:7: error: invalid operand for instruction
 auipc s2, 19
-# CHECK-DIS: 97 49 01 00 <unknown>
+# CHECK-DIS: 00014997 <unknown>
 # CHECK: :[[@LINE+1]]:7: error: invalid operand for instruction
 auipc s3, 20
-# CHECK-DIS: 17 5a 01 00 <unknown>
+# CHECK-DIS: 00015a17 <unknown>
 # CHECK: :[[@LINE+1]]:7: error: invalid operand for instruction
 auipc s4, 21
-# CHECK-DIS: 97 6a 01 00 <unknown>
+# CHECK-DIS: 00016a97 <unknown>
 # CHECK: :[[@LINE+1]]:7: error: invalid operand for instruction
 auipc s5, 22
-# CHECK-DIS: 17 7b 01 00 <unknown>
+# CHECK-DIS: 00017b17 <unknown>
 # CHECK: :[[@LINE+1]]:7: error: invalid operand for instruction
 auipc s6, 23
-# CHECK-DIS: 97 8b 01 00 <unknown>
+# CHECK-DIS: 00018b97 <unknown>
 # CHECK: :[[@LINE+1]]:7: error: invalid operand for instruction
 auipc s7, 24
-# CHECK-DIS: 17 9c 01 00 <unknown>
+# CHECK-DIS: 00019c17 <unknown>
 # CHECK: :[[@LINE+1]]:7: error: invalid operand for instruction
 auipc s8, 25
-# CHECK-DIS: 97 ac 01 00 <unknown>
+# CHECK-DIS: 0001ac97 <unknown>
 # CHECK: :[[@LINE+1]]:7: error: invalid operand for instruction
 auipc s9, 26
-# CHECK-DIS: 17 bd 01 00 <unknown>
+# CHECK-DIS: 0001bd17 <unknown>
 # CHECK: :[[@LINE+1]]:7: error: invalid operand for instruction
 auipc s10, 27
-# CHECK-DIS: 97 cd 01 00 <unknown>
+# CHECK-DIS: 0001cd97 <unknown>
 # CHECK: :[[@LINE+1]]:7: error: invalid operand for instruction
 auipc s11, 28
-# CHECK-DIS: 17 de 01 00 <unknown>
+# CHECK-DIS: 0001de17 <unknown>
 # CHECK: :[[@LINE+1]]:7: error: invalid operand for instruction
 auipc t3, 29
-# CHECK-DIS: 97 ee 01 00 <unknown>
+# CHECK-DIS: 0001ee97 <unknown>
 # CHECK: :[[@LINE+1]]:7: error: invalid operand for instruction
 auipc t4, 30
-# CHECK-DIS: 17 ff 01 00 <unknown>
+# CHECK-DIS: 0001ff17 <unknown>
 # CHECK: :[[@LINE+1]]:7: error: invalid operand for instruction
 auipc t5, 31
-# CHECK-DIS: 97 0f 02 00 <unknown>
+# CHECK-DIS: 00020f97 <unknown>
 # CHECK: :[[@LINE+1]]:7: error: invalid operand for instruction
 auipc t6, 32
diff --git a/llvm/test/MC/RISCV/rv32fc-valid.s b/llvm/test/MC/RISCV/rv32fc-valid.s
index f8be03367794..936032594457 100644
--- a/llvm/test/MC/RISCV/rv32fc-valid.s
+++ b/llvm/test/MC/RISCV/rv32fc-valid.s
@@ -12,9 +12,6 @@
 # RUN: not llvm-mc -triple riscv32 -mattr=+c \
 # RUN:     -riscv-no-aliases -show-encoding < %s 2>&1 \
 # RUN:     | FileCheck -check-prefixes=CHECK-NO-EXT-F %s
-# RUN: not llvm-mc -triple riscv32 -mattr=+zcf \
-# RUN:     -riscv-no-aliases -show-encoding < %s 2>&1 \
-# RUN:     | FileCheck -check-prefixes=CHECK-NO-EXT-F %s
 # RUN: not llvm-mc -triple riscv32 \
 # RUN:     -riscv-no-aliases -show-encoding < %s 2>&1 \
 # RUN:     | FileCheck -check-prefixes=CHECK-NO-EXT-FC %s
diff --git a/llvm/test/MC/RISCV/rv32zacas-invalid.s b/llvm/test/MC/RISCV/rv32zacas-invalid.s
index 11d20dacd8a7..66f939d139a1 100644
--- a/llvm/test/MC/RISCV/rv32zacas-invalid.s
+++ b/llvm/test/MC/RISCV/rv32zacas-invalid.s
@@ -1,4 +1,4 @@
-# RUN: not llvm-mc -triple riscv32 -mattr=+zacas < %s 2>&1 | FileCheck %s
+# RUN: not llvm-mc -triple riscv32 -mattr=+a,zacas < %s 2>&1 | FileCheck %s
 
 # Non-zero offsets not supported for the third operand (rs1).
 amocas.w a1, a3, 1(a5) # CHECK: :[[@LINE]]:18: error: optional integer offset must be 0
diff --git a/llvm/test/MC/RISCV/rv32zacas-valid.s b/llvm/test/MC/RISCV/rv32zacas-valid.s
index 05a9cdd5cc21..0e76f0239948 100644
--- a/llvm/test/MC/RISCV/rv32zacas-valid.s
+++ b/llvm/test/MC/RISCV/rv32zacas-valid.s
@@ -1,12 +1,12 @@
-# RUN: llvm-mc %s -triple=riscv32 -mattr=+zacas -riscv-no-aliases -show-encoding \
+# RUN: llvm-mc %s -triple=riscv32 -mattr=+a,+zacas -riscv-no-aliases -show-encoding \
 # RUN:     | FileCheck -check-prefixes=CHECK-ASM,CHECK-ASM-AND-OBJ %s
-# RUN: llvm-mc %s -triple=riscv64 -mattr=+zacas -riscv-no-aliases -show-encoding \
+# RUN: llvm-mc %s -triple=riscv64 -mattr=+a,+zacas -riscv-no-aliases -show-encoding \
 # RUN:     | FileCheck -check-prefixes=CHECK-ASM,CHECK-ASM-AND-OBJ %s
-# RUN: llvm-mc -filetype=obj -triple=riscv32 -mattr=+zacas < %s \
-# RUN:     | llvm-objdump --mattr=+zacas -M no-aliases -d -r - \
+# RUN: llvm-mc -filetype=obj -triple=riscv32 -mattr=+a,+zacas < %s \
+# RUN:     | llvm-objdump --mattr=+a,+zacas -M no-aliases -d -r - \
 # RUN:     | FileCheck --check-prefix=CHECK-ASM-AND-OBJ %s
-# RUN: llvm-mc -filetype=obj -triple=riscv64 -mattr=+zacas < %s \
-# RUN:     | llvm-objdump --mattr=+zacas -M no-aliases -d -r - \
+# RUN: llvm-mc -filetype=obj -triple=riscv64 -mattr=+a,+zacas < %s \
+# RUN:     | llvm-objdump --mattr=+a,+zacas -M no-aliases -d -r - \
 # RUN:     | FileCheck --check-prefix=CHECK-ASM-AND-OBJ %s
 # RUN: not llvm-mc -triple=riscv32 -mattr=+a -show-encoding %s 2>&1 \
 # RUN:        | FileCheck %s --check-prefix=CHECK-ERROR
diff --git a/llvm/test/MC/RISCV/rv64dc-valid.s b/llvm/test/MC/RISCV/rv64dc-valid.s
index 5347714761dc..83225b2c6856 100644
--- a/llvm/test/MC/RISCV/rv64dc-valid.s
+++ b/llvm/test/MC/RISCV/rv64dc-valid.s
@@ -12,9 +12,6 @@
 # RUN: not llvm-mc -triple riscv64 -mattr=+c \
 # RUN:     -riscv-no-aliases -show-encoding < %s 2>&1 \
 # RUN:     | FileCheck -check-prefixes=CHECK-NO-EXT-D %s
-# RUN: not llvm-mc -triple riscv64 -mattr=+zcd \
-# RUN:     -riscv-no-aliases -show-encoding < %s 2>&1 \
-# RUN:     | FileCheck -check-prefixes=CHECK-NO-EXT-D %s
 # RUN: not llvm-mc -triple riscv64 -riscv-no-aliases -show-encoding < %s 2>&1 \
 # RUN:     | FileCheck -check-prefixes=CHECK-NO-EXT-DC %s
 
diff --git a/llvm/test/MC/RISCV/rv64zacas-valid.s b/llvm/test/MC/RISCV/rv64zacas-valid.s
index 694f43b9b440..c6bf1252fa40 100644
--- a/llvm/test/MC/RISCV/rv64zacas-valid.s
+++ b/llvm/test/MC/RISCV/rv64zacas-valid.s
@@ -1,7 +1,7 @@
-# RUN: llvm-mc %s -triple=riscv64 -mattr=+zacas -riscv-no-aliases -show-encoding \
+# RUN: llvm-mc %s -triple=riscv64 -mattr=+a,zacas -riscv-no-aliases -show-encoding \
 # RUN:     | FileCheck -check-prefixes=CHECK-ASM,CHECK-ASM-AND-OBJ %s
-# RUN: llvm-mc -filetype=obj -triple=riscv64 -mattr=+zacas < %s \
-# RUN:     | llvm-objdump --mattr=+zacas -M no-aliases -d -r - \
+# RUN: llvm-mc -filetype=obj -triple=riscv64 -mattr=+a,zacas < %s \
+# RUN:     | llvm-objdump --mattr=+a,zacas -M no-aliases -d -r - \
 # RUN:     | FileCheck --check-prefix=CHECK-ASM-AND-OBJ %s
 # RUN: not llvm-mc -triple=riscv64 -mattr=+a -show-encoding %s 2>&1 \
 # RUN:        | FileCheck %s --check-prefix=CHECK-ERROR
diff --git a/llvm/test/MC/RISCV/rvv/add.s b/llvm/test/MC/RISCV/rvv/add.s
index 89cef5dc0a4c..ebfe50f2d958 100644
--- a/llvm/test/MC/RISCV/rvv/add.s
+++ b/llvm/test/MC/RISCV/rvv/add.s
@@ -12,352 +12,352 @@ vadd.vv v8, v4, v20, v0.t
 # CHECK-INST: vadd.vv v8, v4, v20, v0.t
 # CHECK-ENCODING: [0x57,0x04,0x4a,0x00]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 04 4a 00 <unknown>
+# CHECK-UNKNOWN: 004a0457 <unknown>
 
 vadd.vv v8, v4, v20
 # CHECK-INST: vadd.vv v8, v4, v20
 # CHECK-ENCODING: [0x57,0x04,0x4a,0x02]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 04 4a 02 <unknown>
+# CHECK-UNKNOWN: 024a0457 <unknown>
 
 vadd.vx v8, v4, a0, v0.t
 # CHECK-INST: vadd.vx v8, v4, a0, v0.t
 # CHECK-ENCODING: [0x57,0x44,0x45,0x00]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 44 45 00 <unknown>
+# CHECK-UNKNOWN: 00454457 <unknown>
 
 vadd.vx v8, v4, a0
 # CHECK-INST: vadd.vx v8, v4, a0
 # CHECK-ENCODING: [0x57,0x44,0x45,0x02]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 44 45 02 <unknown>
+# CHECK-UNKNOWN: 02454457 <unknown>
 
 vadd.vi v8, v4, 15, v0.t
 # CHECK-INST: vadd.vi v8, v4, 15, v0.t
 # CHECK-ENCODING: [0x57,0xb4,0x47,0x00]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 b4 47 00 <unknown>
+# CHECK-UNKNOWN: 0047b457 <unknown>
 
 vadd.vi v8, v4, 15
 # CHECK-INST: vadd.vi v8, v4, 15
 # CHECK-ENCODING: [0x57,0xb4,0x47,0x02]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 b4 47 02 <unknown>
+# CHECK-UNKNOWN: 0247b457 <unknown>
 
 vwaddu.vv v8, v4, v20, v0.t
 # CHECK-INST: vwaddu.vv v8, v4, v20, v0.t
 # CHECK-ENCODING: [0x57,0x24,0x4a,0xc0]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 24 4a c0 <unknown>
+# CHECK-UNKNOWN: c04a2457 <unknown>
 
 vwaddu.vv v8, v4, v20
 # CHECK-INST: vwaddu.vv v8, v4, v20
 # CHECK-ENCODING: [0x57,0x24,0x4a,0xc2]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 24 4a c2 <unknown>
+# CHECK-UNKNOWN: c24a2457 <unknown>
 
 vwaddu.vx v8, v4, a0, v0.t
 # CHECK-INST: vwaddu.vx v8, v4, a0, v0.t
 # CHECK-ENCODING: [0x57,0x64,0x45,0xc0]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 64 45 c0 <unknown>
+# CHECK-UNKNOWN: c0456457 <unknown>
 
 vwaddu.vx v8, v4, a0
 # CHECK-INST: vwaddu.vx v8, v4, a0
 # CHECK-ENCODING: [0x57,0x64,0x45,0xc2]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 64 45 c2 <unknown>
+# CHECK-UNKNOWN: c2456457 <unknown>
 
 vwadd.vv v8, v4, v20, v0.t
 # CHECK-INST: vwadd.vv v8, v4, v20, v0.t
 # CHECK-ENCODING: [0x57,0x24,0x4a,0xc4]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 24 4a c4 <unknown>
+# CHECK-UNKNOWN: c44a2457 <unknown>
 
 vwadd.vv v8, v4, v20
 # CHECK-INST: vwadd.vv v8, v4, v20
 # CHECK-ENCODING: [0x57,0x24,0x4a,0xc6]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 24 4a c6 <unknown>
+# CHECK-UNKNOWN: c64a2457 <unknown>
 
 vwadd.vx v8, v4, a0, v0.t
 # CHECK-INST: vwadd.vx v8, v4, a0, v0.t
 # CHECK-ENCODING: [0x57,0x64,0x45,0xc4]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 64 45 c4 <unknown>
+# CHECK-UNKNOWN: c4456457 <unknown>
 
 vwadd.vx v8, v4, a0
 # CHECK-INST: vwadd.vx v8, v4, a0
 # CHECK-ENCODING: [0x57,0x64,0x45,0xc6]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 64 45 c6 <unknown>
+# CHECK-UNKNOWN: c6456457 <unknown>
 
 vwaddu.wv v8, v4, v20, v0.t
 # CHECK-INST: vwaddu.wv v8, v4, v20, v0.t
 # CHECK-ENCODING: [0x57,0x24,0x4a,0xd0]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 24 4a d0 <unknown>
+# CHECK-UNKNOWN: d04a2457 <unknown>
 
 vwaddu.wv v8, v4, v20
 # CHECK-INST: vwaddu.wv v8, v4, v20
 # CHECK-ENCODING: [0x57,0x24,0x4a,0xd2]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 24 4a d2 <unknown>
+# CHECK-UNKNOWN: d24a2457 <unknown>
 
 vwaddu.wx v8, v4, a0, v0.t
 # CHECK-INST: vwaddu.wx v8, v4, a0, v0.t
 # CHECK-ENCODING: [0x57,0x64,0x45,0xd0]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 64 45 d0 <unknown>
+# CHECK-UNKNOWN: d0456457 <unknown>
 
 vwaddu.wx v8, v4, a0
 # CHECK-INST: vwaddu.wx v8, v4, a0
 # CHECK-ENCODING: [0x57,0x64,0x45,0xd2]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 64 45 d2 <unknown>
+# CHECK-UNKNOWN: d2456457 <unknown>
 
 vwadd.wv v8, v4, v20, v0.t
 # CHECK-INST: vwadd.wv v8, v4, v20, v0.t
 # CHECK-ENCODING: [0x57,0x24,0x4a,0xd4]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 24 4a d4 <unknown>
+# CHECK-UNKNOWN: d44a2457 <unknown>
 
 vwadd.wv v8, v4, v20
 # CHECK-INST: vwadd.wv v8, v4, v20
 # CHECK-ENCODING: [0x57,0x24,0x4a,0xd6]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 24 4a d6 <unknown>
+# CHECK-UNKNOWN: d64a2457 <unknown>
 
 vwadd.wx v8, v4, a0, v0.t
 # CHECK-INST: vwadd.wx v8, v4, a0, v0.t
 # CHECK-ENCODING: [0x57,0x64,0x45,0xd4]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 64 45 d4 <unknown>
+# CHECK-UNKNOWN: d4456457 <unknown>
 
 vwadd.wx v8, v4, a0
 # CHECK-INST: vwadd.wx v8, v4, a0
 # CHECK-ENCODING: [0x57,0x64,0x45,0xd6]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 64 45 d6 <unknown>
+# CHECK-UNKNOWN: d6456457 <unknown>
 
 vadc.vvm v8, v4, v20, v0
 # CHECK-INST: vadc.vvm v8, v4, v20, v0
 # CHECK-ENCODING: [0x57,0x04,0x4a,0x40]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 04 4a 40 <unknown>
+# CHECK-UNKNOWN: 404a0457 <unknown>
 
 vadc.vvm v4, v4, v20, v0
 # CHECK-INST: vadc.vvm v4, v4, v20, v0
 # CHECK-ENCODING: [0x57,0x02,0x4a,0x40]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 02 4a 40 <unknown>
+# CHECK-UNKNOWN: 404a0257 <unknown>
 
 vadc.vvm v8, v4, v8, v0
 # CHECK-INST: vadc.vvm v8, v4, v8, v0
 # CHECK-ENCODING: [0x57,0x04,0x44,0x40]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 04 44 40 <unknown>
+# CHECK-UNKNOWN: 40440457 <unknown>
 
 vadc.vxm v8, v4, a0, v0
 # CHECK-INST: vadc.vxm v8, v4, a0, v0
 # CHECK-ENCODING: [0x57,0x44,0x45,0x40]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 44 45 40 <unknown>
+# CHECK-UNKNOWN: 40454457 <unknown>
 
 vadc.vim v8, v4, 15, v0
 # CHECK-INST: vadc.vim v8, v4, 15, v0
 # CHECK-ENCODING: [0x57,0xb4,0x47,0x40]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 b4 47 40 <unknown>
+# CHECK-UNKNOWN: 4047b457 <unknown>
 
 vmadc.vvm v8, v4, v20, v0
 # CHECK-INST: vmadc.vvm v8, v4, v20, v0
 # CHECK-ENCODING: [0x57,0x04,0x4a,0x44]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 04 4a 44 <unknown>
+# CHECK-UNKNOWN: 444a0457 <unknown>
 
 vmadc.vvm v4, v4, v20, v0
 # CHECK-INST: vmadc.vvm v4, v4, v20, v0
 # CHECK-ENCODING: [0x57,0x02,0x4a,0x44]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 02 4a 44 <unknown>
+# CHECK-UNKNOWN: 444a0257 <unknown>
 
 vmadc.vvm v8, v4, v8, v0
 # CHECK-INST: vmadc.vvm v8, v4, v8, v0
 # CHECK-ENCODING: [0x57,0x04,0x44,0x44]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 04 44 44 <unknown>
+# CHECK-UNKNOWN: 44440457 <unknown>
 
 vmadc.vxm v8, v4, a0, v0
 # CHECK-INST: vmadc.vxm v8, v4, a0, v0
 # CHECK-ENCODING: [0x57,0x44,0x45,0x44]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 44 45 44 <unknown>
+# CHECK-UNKNOWN: 44454457 <unknown>
 
 vmadc.vim v8, v4, 15, v0
 # CHECK-INST: vmadc.vim v8, v4, 15, v0
 # CHECK-ENCODING: [0x57,0xb4,0x47,0x44]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 b4 47 44 <unknown>
+# CHECK-UNKNOWN: 4447b457 <unknown>
 
 vmadc.vv v8, v4, v20
 # CHECK-INST: vmadc.vv v8, v4, v20
 # CHECK-ENCODING: [0x57,0x04,0x4a,0x46]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 04 4a 46 <unknown>
+# CHECK-UNKNOWN: 464a0457 <unknown>
 
 vmadc.vx v8, v4, a0
 # CHECK-INST: vmadc.vx v8, v4, a0
 # CHECK-ENCODING: [0x57,0x44,0x45,0x46]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 44 45 46 <unknown>
+# CHECK-UNKNOWN: 46454457 <unknown>
 
 vmadc.vi v8, v4, 15
 # CHECK-INST: vmadc.vi v8, v4, 15
 # CHECK-ENCODING: [0x57,0xb4,0x47,0x46]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 b4 47 46 <unknown>
+# CHECK-UNKNOWN: 4647b457 <unknown>
 
 vsaddu.vv v8, v4, v20, v0.t
 # CHECK-INST: vsaddu.vv v8, v4, v20, v0.t
 # CHECK-ENCODING: [0x57,0x04,0x4a,0x80]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 04 4a 80 <unknown>
+# CHECK-UNKNOWN: 804a0457 <unknown>
 
 vsaddu.vv v8, v4, v20
 # CHECK-INST: vsaddu.vv v8, v4, v20
 # CHECK-ENCODING: [0x57,0x04,0x4a,0x82]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 04 4a 82 <unknown>
+# CHECK-UNKNOWN: 824a0457 <unknown>
 
 vsaddu.vx v8, v4, a0, v0.t
 # CHECK-INST: vsaddu.vx v8, v4, a0, v0.t
 # CHECK-ENCODING: [0x57,0x44,0x45,0x80]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 44 45 80 <unknown>
+# CHECK-UNKNOWN: 80454457 <unknown>
 
 vsaddu.vx v8, v4, a0
 # CHECK-INST: vsaddu.vx v8, v4, a0
 # CHECK-ENCODING: [0x57,0x44,0x45,0x82]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 44 45 82 <unknown>
+# CHECK-UNKNOWN: 82454457 <unknown>
 
 vsaddu.vi v8, v4, 15, v0.t
 # CHECK-INST: vsaddu.vi v8, v4, 15, v0.t
 # CHECK-ENCODING: [0x57,0xb4,0x47,0x80]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 b4 47 80 <unknown>
+# CHECK-UNKNOWN: 8047b457 <unknown>
 
 vsaddu.vi v8, v4, 15
 # CHECK-INST: vsaddu.vi v8, v4, 15
 # CHECK-ENCODING: [0x57,0xb4,0x47,0x82]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 b4 47 82 <unknown>
+# CHECK-UNKNOWN: 8247b457 <unknown>
 
 vsadd.vv v8, v4, v20, v0.t
 # CHECK-INST: vsadd.vv v8, v4, v20, v0.t
 # CHECK-ENCODING: [0x57,0x04,0x4a,0x84]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 04 4a 84 <unknown>
+# CHECK-UNKNOWN: 844a0457 <unknown>
 
 vsadd.vv v8, v4, v20
 # CHECK-INST: vsadd.vv v8, v4, v20
 # CHECK-ENCODING: [0x57,0x04,0x4a,0x86]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 04 4a 86 <unknown>
+# CHECK-UNKNOWN: 864a0457 <unknown>
 
 vsadd.vx v8, v4, a0, v0.t
 # CHECK-INST: vsadd.vx v8, v4, a0, v0.t
 # CHECK-ENCODING: [0x57,0x44,0x45,0x84]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 44 45 84 <unknown>
+# CHECK-UNKNOWN: 84454457 <unknown>
 
 vsadd.vx v8, v4, a0
 # CHECK-INST: vsadd.vx v8, v4, a0
 # CHECK-ENCODING: [0x57,0x44,0x45,0x86]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 44 45 86 <unknown>
+# CHECK-UNKNOWN: 86454457 <unknown>
 
 vsadd.vi v8, v4, 15, v0.t
 # CHECK-INST: vsadd.vi v8, v4, 15, v0.t
 # CHECK-ENCODING: [0x57,0xb4,0x47,0x84]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 b4 47 84 <unknown>
+# CHECK-UNKNOWN: 8447b457 <unknown>
 
 vsadd.vi v8, v4, 15
 # CHECK-INST: vsadd.vi v8, v4, 15
 # CHECK-ENCODING: [0x57,0xb4,0x47,0x86]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 b4 47 86 <unknown>
+# CHECK-UNKNOWN: 8647b457 <unknown>
 
 vaadd.vv v8, v4, v20, v0.t
 # CHECK-INST: vaadd.vv v8, v4, v20, v0.t
 # CHECK-ENCODING: [0x57,0x24,0x4a,0x24]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 24 4a 24 <unknown>
+# CHECK-UNKNOWN: 244a2457 <unknown>
 
 vaadd.vv v8, v4, v20
 # CHECK-INST: vaadd.vv v8, v4, v20
 # CHECK-ENCODING: [0x57,0x24,0x4a,0x26]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 24 4a 26 <unknown>
+# CHECK-UNKNOWN: 264a2457 <unknown>
 
 vaadd.vx v8, v4, a0, v0.t
 # CHECK-INST: vaadd.vx v8, v4, a0, v0.t
 # CHECK-ENCODING: [0x57,0x64,0x45,0x24]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 64 45 24 <unknown>
+# CHECK-UNKNOWN: 24456457 <unknown>
 
 vaadd.vx v8, v4, a0
 # CHECK-INST: vaadd.vx v8, v4, a0
 # CHECK-ENCODING: [0x57,0x64,0x45,0x26]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 64 45 26 <unknown>
+# CHECK-UNKNOWN: 26456457 <unknown>
 
 vaaddu.vv v8, v4, v20, v0.t
 # CHECK-INST: vaaddu.vv v8, v4, v20, v0.t
 # CHECK-ENCODING: [0x57,0x24,0x4a,0x20]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 24 4a 20 <unknown>
+# CHECK-UNKNOWN: 204a2457 <unknown>
 
 vaaddu.vv v8, v4, v20
 # CHECK-INST: vaaddu.vv v8, v4, v20
 # CHECK-ENCODING: [0x57,0x24,0x4a,0x22]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 24 4a 22 <unknown>
+# CHECK-UNKNOWN: 224a2457 <unknown>
 
 vaaddu.vx v8, v4, a0, v0.t
 # CHECK-INST: vaaddu.vx v8, v4, a0, v0.t
 # CHECK-ENCODING: [0x57,0x64,0x45,0x20]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 64 45 20 <unknown>
+# CHECK-UNKNOWN: 20456457 <unknown>
 
 vaaddu.vx v8, v4, a0
 # CHECK-INST: vaaddu.vx v8, v4, a0
 # CHECK-ENCODING: [0x57,0x64,0x45,0x22]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 64 45 22 <unknown>
+# CHECK-UNKNOWN: 22456457 <unknown>
 
 vwcvt.x.x.v v8, v4, v0.t
 # CHECK-INST: vwcvt.x.x.v v8, v4, v0.t
 # CHECK-ENCODING: [0x57,0x64,0x40,0xc4]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 64 40 c4 <unknown>
+# CHECK-UNKNOWN: c4406457 <unknown>
 
 vwcvt.x.x.v v8, v4
 # CHECK-INST: vwcvt.x.x.v v8, v4
 # CHECK-ENCODING: [0x57,0x64,0x40,0xc6]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 64 40 c6 <unknown>
+# CHECK-UNKNOWN: c6406457 <unknown>
 
 vwcvtu.x.x.v v8, v4, v0.t
 # CHECK-INST: vwcvtu.x.x.v v8, v4, v0.t
 # CHECK-ENCODING: [0x57,0x64,0x40,0xc0]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 64 40 c0 <unknown>
+# CHECK-UNKNOWN: c0406457 <unknown>
 
 vwcvtu.x.x.v v8, v4
 # CHECK-INST: vwcvtu.x.x.v v8, v4
 # CHECK-ENCODING: [0x57,0x64,0x40,0xc2]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 64 40 c2 <unknown>
+# CHECK-UNKNOWN: c2406457 <unknown>
diff --git a/llvm/test/MC/RISCV/rvv/and.s b/llvm/test/MC/RISCV/rvv/and.s
index 894263fe0152..b1182c175a5d 100644
--- a/llvm/test/MC/RISCV/rvv/and.s
+++ b/llvm/test/MC/RISCV/rvv/and.s
@@ -12,34 +12,34 @@ vand.vv v8, v4, v20, v0.t
 # CHECK-INST: vand.vv v8, v4, v20, v0.t
 # CHECK-ENCODING: [0x57,0x04,0x4a,0x24]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 04 4a 24 <unknown>
+# CHECK-UNKNOWN: 244a0457 <unknown>
 
 vand.vv v8, v4, v20
 # CHECK-INST: vand.vv v8, v4, v20
 # CHECK-ENCODING: [0x57,0x04,0x4a,0x26]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 04 4a 26 <unknown>
+# CHECK-UNKNOWN: 264a0457 <unknown>
 
 vand.vx v8, v4, a0, v0.t
 # CHECK-INST: vand.vx v8, v4, a0, v0.t
 # CHECK-ENCODING: [0x57,0x44,0x45,0x24]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 44 45 24 <unknown>
+# CHECK-UNKNOWN: 24454457 <unknown>
 
 vand.vx v8, v4, a0
 # CHECK-INST: vand.vx v8, v4, a0
 # CHECK-ENCODING: [0x57,0x44,0x45,0x26]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 44 45 26 <unknown>
+# CHECK-UNKNOWN: 26454457 <unknown>
 
 vand.vi v8, v4, 15, v0.t
 # CHECK-INST: vand.vi v8, v4, 15, v0.t
 # CHECK-ENCODING: [0x57,0xb4,0x47,0x24]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 b4 47 24 <unknown>
+# CHECK-UNKNOWN: 2447b457 <unknown>
 
 vand.vi v8, v4, 15
 # CHECK-INST: vand.vi v8, v4, 15
 # CHECK-ENCODING: [0x57,0xb4,0x47,0x26]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 b4 47 26 <unknown>
+# CHECK-UNKNOWN: 2647b457 <unknown>
diff --git a/llvm/test/MC/RISCV/rvv/clip.s b/llvm/test/MC/RISCV/rvv/clip.s
index f4fb2c576b30..70c23d8ec0b1 100644
--- a/llvm/test/MC/RISCV/rvv/clip.s
+++ b/llvm/test/MC/RISCV/rvv/clip.s
@@ -12,70 +12,70 @@ vnclipu.wv v8, v4, v20, v0.t
 # CHECK-INST: vnclipu.wv v8, v4, v20, v0.t
 # CHECK-ENCODING: [0x57,0x04,0x4a,0xb8]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 04 4a b8 <unknown>
+# CHECK-UNKNOWN: b84a0457 <unknown>
 
 vnclipu.wv v8, v4, v20
 # CHECK-INST: vnclipu.wv v8, v4, v20
 # CHECK-ENCODING: [0x57,0x04,0x4a,0xba]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 04 4a ba <unknown>
+# CHECK-UNKNOWN: ba4a0457 <unknown>
 
 vnclipu.wx v8, v4, a0, v0.t
 # CHECK-INST: vnclipu.wx v8, v4, a0, v0.t
 # CHECK-ENCODING: [0x57,0x44,0x45,0xb8]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 44 45 b8 <unknown>
+# CHECK-UNKNOWN: b8454457 <unknown>
 
 vnclipu.wx v8, v4, a0
 # CHECK-INST: vnclipu.wx v8, v4, a0
 # CHECK-ENCODING: [0x57,0x44,0x45,0xba]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 44 45 ba <unknown>
+# CHECK-UNKNOWN: ba454457 <unknown>
 
 vnclipu.wi v8, v4, 31, v0.t
 # CHECK-INST: vnclipu.wi v8, v4, 31, v0.t
 # CHECK-ENCODING: [0x57,0xb4,0x4f,0xb8]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 b4 4f b8 <unknown>
+# CHECK-UNKNOWN: b84fb457 <unknown>
 
 vnclipu.wi v8, v4, 31
 # CHECK-INST: vnclipu.wi v8, v4, 31
 # CHECK-ENCODING: [0x57,0xb4,0x4f,0xba]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 b4 4f ba <unknown>
+# CHECK-UNKNOWN: ba4fb457 <unknown>
 
 vnclip.wv v8, v4, v20, v0.t
 # CHECK-INST: vnclip.wv v8, v4, v20, v0.t
 # CHECK-ENCODING: [0x57,0x04,0x4a,0xbc]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 04 4a bc <unknown>
+# CHECK-UNKNOWN: bc4a0457 <unknown>
 
 vnclip.wv v8, v4, v20
 # CHECK-INST: vnclip.wv v8, v4, v20
 # CHECK-ENCODING: [0x57,0x04,0x4a,0xbe]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 04 4a be <unknown>
+# CHECK-UNKNOWN: be4a0457 <unknown>
 
 vnclip.wx v8, v4, a0, v0.t
 # CHECK-INST: vnclip.wx v8, v4, a0, v0.t
 # CHECK-ENCODING: [0x57,0x44,0x45,0xbc]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 44 45 bc <unknown>
+# CHECK-UNKNOWN: bc454457 <unknown>
 
 vnclip.wx v8, v4, a0
 # CHECK-INST: vnclip.wx v8, v4, a0
 # CHECK-ENCODING: [0x57,0x44,0x45,0xbe]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 44 45 be <unknown>
+# CHECK-UNKNOWN: be454457 <unknown>
 
 vnclip.wi v8, v4, 31, v0.t
 # CHECK-INST: vnclip.wi v8, v4, 31, v0.t
 # CHECK-ENCODING: [0x57,0xb4,0x4f,0xbc]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 b4 4f bc <unknown>
+# CHECK-UNKNOWN: bc4fb457 <unknown>
 
 vnclip.wi v8, v4, 31
 # CHECK-INST: vnclip.wi v8, v4, 31
 # CHECK-ENCODING: [0x57,0xb4,0x4f,0xbe]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 b4 4f be <unknown>
+# CHECK-UNKNOWN: be4fb457 <unknown>
diff --git a/llvm/test/MC/RISCV/rvv/compare.s b/llvm/test/MC/RISCV/rvv/compare.s
index fe7c1144a3c0..b1b9518a1d5a 100644
--- a/llvm/test/MC/RISCV/rvv/compare.s
+++ b/llvm/test/MC/RISCV/rvv/compare.s
@@ -12,367 +12,367 @@ vmslt.vv v0, v4, v20, v0.t
 # CHECK-INST: vmslt.vv v0, v4, v20, v0.t
 # CHECK-ENCODING: [0x57,0x00,0x4a,0x6c]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 00 4a 6c <unknown>
+# CHECK-UNKNOWN: 6c4a0057 <unknown>
 
 vmseq.vv v8, v4, v20, v0.t
 # CHECK-INST: vmseq.vv v8, v4, v20, v0.t
 # CHECK-ENCODING: [0x57,0x04,0x4a,0x60]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 04 4a 60 <unknown>
+# CHECK-UNKNOWN: 604a0457 <unknown>
 
 vmseq.vv v8, v4, v20
 # CHECK-INST: vmseq.vv v8, v4, v20
 # CHECK-ENCODING: [0x57,0x04,0x4a,0x62]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 04 4a 62 <unknown>
+# CHECK-UNKNOWN: 624a0457 <unknown>
 
 vmseq.vx v8, v4, a0, v0.t
 # CHECK-INST: vmseq.vx v8, v4, a0, v0.t
 # CHECK-ENCODING: [0x57,0x44,0x45,0x60]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 44 45 60 <unknown>
+# CHECK-UNKNOWN: 60454457 <unknown>
 
 vmseq.vx v8, v4, a0
 # CHECK-INST: vmseq.vx v8, v4, a0
 # CHECK-ENCODING: [0x57,0x44,0x45,0x62]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 44 45 62 <unknown>
+# CHECK-UNKNOWN: 62454457 <unknown>
 
 vmseq.vi v8, v4, 15, v0.t
 # CHECK-INST: vmseq.vi v8, v4, 0xf, v0.t
 # CHECK-ENCODING: [0x57,0xb4,0x47,0x60]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 b4 47 60 <unknown>
+# CHECK-UNKNOWN: 6047b457 <unknown>
 
 vmseq.vi v8, v4, 15
 # CHECK-INST: vmseq.vi v8, v4, 0xf
 # CHECK-ENCODING: [0x57,0xb4,0x47,0x62]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 b4 47 62 <unknown>
+# CHECK-UNKNOWN: 6247b457 <unknown>
 
 vmsne.vv v8, v4, v20, v0.t
 # CHECK-INST: vmsne.vv v8, v4, v20, v0.t
 # CHECK-ENCODING: [0x57,0x04,0x4a,0x64]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 04 4a 64 <unknown>
+# CHECK-UNKNOWN: 644a0457 <unknown>
 
 vmsne.vv v8, v4, v20
 # CHECK-INST: vmsne.vv v8, v4, v20
 # CHECK-ENCODING: [0x57,0x04,0x4a,0x66]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 04 4a 66 <unknown>
+# CHECK-UNKNOWN: 664a0457 <unknown>
 
 vmsne.vx v8, v4, a0, v0.t
 # CHECK-INST: vmsne.vx v8, v4, a0, v0.t
 # CHECK-ENCODING: [0x57,0x44,0x45,0x64]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 44 45 64 <unknown>
+# CHECK-UNKNOWN: 64454457 <unknown>
 
 vmsne.vx v8, v4, a0
 # CHECK-INST: vmsne.vx v8, v4, a0
 # CHECK-ENCODING: [0x57,0x44,0x45,0x66]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 44 45 66 <unknown>
+# CHECK-UNKNOWN: 66454457 <unknown>
 
 vmsne.vi v8, v4, 15, v0.t
 # CHECK-INST: vmsne.vi v8, v4, 0xf, v0.t
 # CHECK-ENCODING: [0x57,0xb4,0x47,0x64]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 b4 47 64 <unknown>
+# CHECK-UNKNOWN: 6447b457 <unknown>
 
 vmsne.vi v8, v4, 15
 # CHECK-INST: vmsne.vi v8, v4, 0xf
 # CHECK-ENCODING: [0x57,0xb4,0x47,0x66]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 b4 47 66 <unknown>
+# CHECK-UNKNOWN: 6647b457 <unknown>
 
 vmsltu.vv v8, v4, v20, v0.t
 # CHECK-INST: vmsltu.vv v8, v4, v20, v0.t
 # CHECK-ENCODING: [0x57,0x04,0x4a,0x68]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 04 4a 68 <unknown>
+# CHECK-UNKNOWN: 684a0457 <unknown>
 
 vmsltu.vv v8, v4, v20
 # CHECK-INST: vmsltu.vv v8, v4, v20
 # CHECK-ENCODING: [0x57,0x04,0x4a,0x6a]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 04 4a 6a <unknown>
+# CHECK-UNKNOWN: 6a4a0457 <unknown>
 
 vmsltu.vx v8, v4, a0, v0.t
 # CHECK-INST: vmsltu.vx v8, v4, a0, v0.t
 # CHECK-ENCODING: [0x57,0x44,0x45,0x68]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 44 45 68 <unknown>
+# CHECK-UNKNOWN: 68454457 <unknown>
 
 vmsltu.vx v8, v4, a0
 # CHECK-INST: vmsltu.vx v8, v4, a0
 # CHECK-ENCODING: [0x57,0x44,0x45,0x6a]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 44 45 6a <unknown>
+# CHECK-UNKNOWN: 6a454457 <unknown>
 
 vmslt.vv v8, v4, v20, v0.t
 # CHECK-INST: vmslt.vv v8, v4, v20, v0.t
 # CHECK-ENCODING: [0x57,0x04,0x4a,0x6c]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 04 4a 6c <unknown>
+# CHECK-UNKNOWN: 6c4a0457 <unknown>
 
 vmslt.vv v8, v4, v20
 # CHECK-INST: vmslt.vv v8, v4, v20
 # CHECK-ENCODING: [0x57,0x04,0x4a,0x6e]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 04 4a 6e <unknown>
+# CHECK-UNKNOWN: 6e4a0457 <unknown>
 
 vmslt.vx v8, v4, a0, v0.t
 # CHECK-INST: vmslt.vx v8, v4, a0, v0.t
 # CHECK-ENCODING: [0x57,0x44,0x45,0x6c]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 44 45 6c <unknown>
+# CHECK-UNKNOWN: 6c454457 <unknown>
 
 vmslt.vx v8, v4, a0
 # CHECK-INST: vmslt.vx v8, v4, a0
 # CHECK-ENCODING: [0x57,0x44,0x45,0x6e]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 44 45 6e <unknown>
+# CHECK-UNKNOWN: 6e454457 <unknown>
 
 vmsleu.vv v8, v4, v20, v0.t
 # CHECK-INST: vmsleu.vv v8, v4, v20, v0.t
 # CHECK-ENCODING: [0x57,0x04,0x4a,0x70]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 04 4a 70 <unknown>
+# CHECK-UNKNOWN: 704a0457 <unknown>
 
 vmsleu.vv v8, v4, v20
 # CHECK-INST: vmsleu.vv v8, v4, v20
 # CHECK-ENCODING: [0x57,0x04,0x4a,0x72]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 04 4a 72 <unknown>
+# CHECK-UNKNOWN: 724a0457 <unknown>
 
 vmsleu.vx v8, v4, a0, v0.t
 # CHECK-INST: vmsleu.vx v8, v4, a0, v0.t
 # CHECK-ENCODING: [0x57,0x44,0x45,0x70]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 44 45 70 <unknown>
+# CHECK-UNKNOWN: 70454457 <unknown>
 
 vmsleu.vx v8, v4, a0
 # CHECK-INST: vmsleu.vx v8, v4, a0
 # CHECK-ENCODING: [0x57,0x44,0x45,0x72]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 44 45 72 <unknown>
+# CHECK-UNKNOWN: 72454457 <unknown>
 
 vmsleu.vi v8, v4, 15, v0.t
 # CHECK-INST: vmsleu.vi v8, v4, 0xf, v0.t
 # CHECK-ENCODING: [0x57,0xb4,0x47,0x70]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 b4 47 70 <unknown>
+# CHECK-UNKNOWN: 7047b457 <unknown>
 
 vmsleu.vi v8, v4, 15
 # CHECK-INST: vmsleu.vi v8, v4, 0xf
 # CHECK-ENCODING: [0x57,0xb4,0x47,0x72]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 b4 47 72 <unknown>
+# CHECK-UNKNOWN: 7247b457 <unknown>
 
 vmsle.vv v8, v4, v20, v0.t
 # CHECK-INST: vmsle.vv v8, v4, v20, v0.t
 # CHECK-ENCODING: [0x57,0x04,0x4a,0x74]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 04 4a 74 <unknown>
+# CHECK-UNKNOWN: 744a0457 <unknown>
 
 vmsle.vv v8, v4, v20
 # CHECK-INST: vmsle.vv v8, v4, v20
 # CHECK-ENCODING: [0x57,0x04,0x4a,0x76]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 04 4a 76 <unknown>
+# CHECK-UNKNOWN: 764a0457 <unknown>
 
 vmsle.vx v8, v4, a0, v0.t
 # CHECK-INST: vmsle.vx v8, v4, a0, v0.t
 # CHECK-ENCODING: [0x57,0x44,0x45,0x74]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 44 45 74 <unknown>
+# CHECK-UNKNOWN: 74454457 <unknown>
 
 vmsle.vx v8, v4, a0
 # CHECK-INST: vmsle.vx v8, v4, a0
 # CHECK-ENCODING: [0x57,0x44,0x45,0x76]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 44 45 76 <unknown>
+# CHECK-UNKNOWN: 76454457 <unknown>
 
 vmsle.vi v8, v4, 15, v0.t
 # CHECK-INST: vmsle.vi v8, v4, 0xf, v0.t
 # CHECK-ENCODING: [0x57,0xb4,0x47,0x74]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 b4 47 74 <unknown>
+# CHECK-UNKNOWN: 7447b457 <unknown>
 
 vmsle.vi v8, v4, 15
 # CHECK-INST: vmsle.vi v8, v4, 0xf
 # CHECK-ENCODING: [0x57,0xb4,0x47,0x76]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 b4 47 76 <unknown>
+# CHECK-UNKNOWN: 7647b457 <unknown>
 
 vmsgtu.vx v8, v4, a0, v0.t
 # CHECK-INST: vmsgtu.vx v8, v4, a0, v0.t
 # CHECK-ENCODING: [0x57,0x44,0x45,0x78]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 44 45 78 <unknown>
+# CHECK-UNKNOWN: 78454457 <unknown>
 
 vmsgtu.vx v8, v4, a0
 # CHECK-INST: vmsgtu.vx v8, v4, a0
 # CHECK-ENCODING: [0x57,0x44,0x45,0x7a]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 44 45 7a <unknown>
+# CHECK-UNKNOWN: 7a454457 <unknown>
 
 vmsgtu.vi v8, v4, 15, v0.t
 # CHECK-INST: vmsgtu.vi v8, v4, 0xf, v0.t
 # CHECK-ENCODING: [0x57,0xb4,0x47,0x78]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 b4 47 78 <unknown>
+# CHECK-UNKNOWN: 7847b457 <unknown>
 
 vmsgtu.vi v8, v4, 15
 # CHECK-INST: vmsgtu.vi v8, v4, 0xf
 # CHECK-ENCODING: [0x57,0xb4,0x47,0x7a]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 b4 47 7a <unknown>
+# CHECK-UNKNOWN: 7a47b457 <unknown>
 
 vmsgt.vx v8, v4, a0, v0.t
 # CHECK-INST: vmsgt.vx v8, v4, a0, v0.t
 # CHECK-ENCODING: [0x57,0x44,0x45,0x7c]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 44 45 7c <unknown>
+# CHECK-UNKNOWN: 7c454457 <unknown>
 
 vmsgt.vx v8, v4, a0
 # CHECK-INST: vmsgt.vx v8, v4, a0
 # CHECK-ENCODING: [0x57,0x44,0x45,0x7e]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 44 45 7e <unknown>
+# CHECK-UNKNOWN: 7e454457 <unknown>
 
 vmsgt.vi v8, v4, 15, v0.t
 # CHECK-INST: vmsgt.vi v8, v4, 0xf, v0.t
 # CHECK-ENCODING: [0x57,0xb4,0x47,0x7c]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 b4 47 7c <unknown>
+# CHECK-UNKNOWN: 7c47b457 <unknown>
 
 vmsgt.vi v8, v4, 15
 # CHECK-INST: vmsgt.vi v8, v4, 0xf
 # CHECK-ENCODING: [0x57,0xb4,0x47,0x7e]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 b4 47 7e <unknown>
+# CHECK-UNKNOWN: 7e47b457 <unknown>
 
 vmsgtu.vv v8, v20, v4, v0.t
 # CHECK-INST: vmsltu.vv v8, v4, v20, v0.t
 # CHECK-ENCODING: [0x57,0x04,0x4a,0x68]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 04 4a 68 <unknown>
+# CHECK-UNKNOWN: 684a0457 <unknown>
 
 vmsgtu.vv v8, v20, v4
 # CHECK-INST: vmsltu.vv v8, v4, v20
 # CHECK-ENCODING: [0x57,0x04,0x4a,0x6a]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 04 4a 6a <unknown>
+# CHECK-UNKNOWN: 6a4a0457 <unknown>
 
 vmsgt.vv v8, v20, v4, v0.t
 # CHECK-INST: vmslt.vv v8, v4, v20, v0.t
 # CHECK-ENCODING: [0x57,0x04,0x4a,0x6c]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 04 4a 6c <unknown>
+# CHECK-UNKNOWN: 6c4a0457 <unknown>
 
 vmsgt.vv v8, v20, v4
 # CHECK-INST: vmslt.vv v8, v4, v20
 # CHECK-ENCODING: [0x57,0x04,0x4a,0x6e]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 04 4a 6e <unknown>
+# CHECK-UNKNOWN: 6e4a0457 <unknown>
 
 vmsgeu.vv v8, v20, v4, v0.t
 # CHECK-INST: vmsleu.vv v8, v4, v20, v0.t
 # CHECK-ENCODING: [0x57,0x04,0x4a,0x70]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 04 4a 70 <unknown>
+# CHECK-UNKNOWN: 704a0457 <unknown>
 
 vmsgeu.vv v8, v20, v4
 # CHECK-INST: vmsleu.vv v8, v4, v20
 # CHECK-ENCODING: [0x57,0x04,0x4a,0x72]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 04 4a 72 <unknown>
+# CHECK-UNKNOWN: 724a0457 <unknown>
 
 vmsge.vv v8, v20, v4, v0.t
 # CHECK-INST: vmsle.vv v8, v4, v20, v0.t
 # CHECK-ENCODING: [0x57,0x04,0x4a,0x74]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 04 4a 74 <unknown>
+# CHECK-UNKNOWN: 744a0457 <unknown>
 
 vmsge.vv v8, v20, v4
 # CHECK-INST: vmsle.vv v8, v4, v20
 # CHECK-ENCODING: [0x57,0x04,0x4a,0x76]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 04 4a 76 <unknown>
+# CHECK-UNKNOWN: 764a0457 <unknown>
 
 vmsltu.vi v8, v4, 16, v0.t
 # CHECK-INST: vmsleu.vi v8, v4, 0xf, v0.t
 # CHECK-ENCODING: [0x57,0xb4,0x47,0x70]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 b4 47 70 <unknown>
+# CHECK-UNKNOWN: 7047b457 <unknown>
 
 vmsltu.vi v8, v4, 16
 # CHECK-INST: vmsleu.vi v8, v4, 0xf
 # CHECK-ENCODING: [0x57,0xb4,0x47,0x72]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 b4 47 72 <unknown>
+# CHECK-UNKNOWN: 7247b457 <unknown>
 
 vmsltu.vi v8, v4, 0, v0.t
 # CHECK-INST: vmsne.vv v8, v4, v4, v0.t
 # CHECK-ENCODING: [0x57,0x04,0x42,0x64]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 04 42 64 <unknown>
+# CHECK-UNKNOWN: 64420457 <unknown>
 
 vmsltu.vi v8, v4, 0
 # CHECK-INST: vmsne.vv v8, v4, v4
 # CHECK-ENCODING: [0x57,0x04,0x42,0x66]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 04 42 66 <unknown>
+# CHECK-UNKNOWN: 66420457 <unknown>
 
 vmslt.vi v8, v4, 16, v0.t
 # CHECK-INST: vmsle.vi v8, v4, 0xf, v0.t
 # CHECK-ENCODING: [0x57,0xb4,0x47,0x74]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 b4 47 74 <unknown>
+# CHECK-UNKNOWN: 7447b457 <unknown>
 
 vmslt.vi v8, v4, 16
 # CHECK-INST: vmsle.vi v8, v4, 0xf
 # CHECK-ENCODING: [0x57,0xb4,0x47,0x76]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 b4 47 76 <unknown>
+# CHECK-UNKNOWN: 7647b457 <unknown>
 
 vmsgeu.vi v8, v4, 16, v0.t
 # CHECK-INST: vmsgtu.vi v8, v4, 0xf, v0.t
 # CHECK-ENCODING: [0x57,0xb4,0x47,0x78]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 b4 47 78 <unknown>
+# CHECK-UNKNOWN: 7847b457 <unknown>
 
 vmsgeu.vi v8, v4, 16
 # CHECK-INST: vmsgtu.vi v8, v4, 0xf
 # CHECK-ENCODING: [0x57,0xb4,0x47,0x7a]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 b4 47 7a <unknown>
+# CHECK-UNKNOWN: 7a47b457 <unknown>
 
 vmsgeu.vi v8, v4, 0, v0.t
 # CHECK-INST: vmseq.vv v8, v4, v4, v0.t
 # CHECK-ENCODING: [0x57,0x04,0x42,0x60]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 04 42 60 <unknown>
+# CHECK-UNKNOWN: 60420457 <unknown>
 
 vmsgeu.vi v8, v4, 0
 # CHECK-INST: vmseq.vv v8, v4, v4
 # CHECK-ENCODING: [0x57,0x04,0x42,0x62]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 04 42 62 <unknown>
+# CHECK-UNKNOWN: 62420457 <unknown>
 
 vmsge.vi v8, v4, 16, v0.t
 # CHECK-INST: vmsgt.vi v8, v4, 0xf, v0.t
 # CHECK-ENCODING: [0x57,0xb4,0x47,0x7c]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 b4 47 7c <unknown>
+# CHECK-UNKNOWN: 7c47b457 <unknown>
 
 vmsge.vi v8, v4, 16
 # CHECK-INST: vmsgt.vi v8, v4, 0xf
 # CHECK-ENCODING: [0x57,0xb4,0x47,0x7e]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 b4 47 7e <unknown>
+# CHECK-UNKNOWN: 7e47b457 <unknown>
 
 vmsgeu.vx v8, v4, a0
 # CHECK-INST: vmsltu.vx v8, v4, a0
@@ -380,8 +380,8 @@ vmsgeu.vx v8, v4, a0
 # CHECK-ENCODING: [0x57,0x44,0x45,0x6a]
 # CHECK-ENCODING: [0x57,0x24,0x84,0x76]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 44 45 6a <unknown>
-# CHECK-UNKNOWN: 57 24 84 76 <unknown>
+# CHECK-UNKNOWN: 6a454457 <unknown>
+# CHECK-UNKNOWN: 76842457 <unknown>
 
 vmsge.vx v0, v4, a0
 # CHECK-INST: vmslt.vx v0, v4, a0
@@ -389,8 +389,8 @@ vmsge.vx v0, v4, a0
 # CHECK-ENCODING: [0x57,0x40,0x45,0x6e]
 # CHECK-ENCODING: [0x57,0x20,0x00,0x76]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 40 45 6e <unknown>
-# CHECK-UNKNOWN: 57 20 00 76 <unknown>
+# CHECK-UNKNOWN: 6e454057 <unknown>
+# CHECK-UNKNOWN: 76002057 <unknown>
 
 vmsge.vx v8, v4, a0
 # CHECK-INST: vmslt.vx v8, v4, a0
@@ -398,8 +398,8 @@ vmsge.vx v8, v4, a0
 # CHECK-ENCODING: [0x57,0x44,0x45,0x6e]
 # CHECK-ENCODING: [0x57,0x24,0x84,0x76]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 44 45 6e <unknown>
-# CHECK-UNKNOWN: 57 24 84 76 <unknown>
+# CHECK-UNKNOWN: 6e454457 <unknown>
+# CHECK-UNKNOWN: 76842457 <unknown>
 
 vmsgeu.vx v8, v4, a0, v0.t
 # CHECK-INST: vmsltu.vx v8, v4, a0, v0.t
@@ -407,8 +407,8 @@ vmsgeu.vx v8, v4, a0, v0.t
 # CHECK-ENCODING: [0x57,0x44,0x45,0x68]
 # CHECK-ENCODING: [0x57,0x24,0x80,0x6e]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 44 45 68 <unknown>
-# CHECK-UNKNOWN: 57 24 80 6e <unknown>
+# CHECK-UNKNOWN: 68454457 <unknown>
+# CHECK-UNKNOWN: 6e802457 <unknown>
 
 vmsge.vx v8, v4, a0, v0.t
 # CHECK-INST: vmslt.vx v8, v4, a0, v0.t
@@ -416,8 +416,8 @@ vmsge.vx v8, v4, a0, v0.t
 # CHECK-ENCODING: [0x57,0x44,0x45,0x6c]
 # CHECK-ENCODING: [0x57,0x24,0x80,0x6e]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 44 45 6c <unknown>
-# CHECK-UNKNOWN: 57 24 80 6e <unknown>
+# CHECK-UNKNOWN: 6c454457 <unknown>
+# CHECK-UNKNOWN: 6e802457 <unknown>
 
 vmsgeu.vx v0, v4, a0, v0.t, v2
 # CHECK-INST: vmsltu.vx v2, v4, a0
@@ -425,8 +425,8 @@ vmsgeu.vx v0, v4, a0, v0.t, v2
 # CHECK-ENCODING: [0x57,0x41,0x45,0x6a]
 # CHECK-ENCODING: [0x57,0x20,0x01,0x62]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 41 45 6a <unknown>
-# CHECK-UNKNOWN: 57 20 01 62 <unknown>
+# CHECK-UNKNOWN: 6a454157 <unknown>
+# CHECK-UNKNOWN: 62012057 <unknown>
 
 vmsge.vx v0, v4, a0, v0.t, v2
 # CHECK-INST: vmslt.vx v2, v4, a0
@@ -434,8 +434,8 @@ vmsge.vx v0, v4, a0, v0.t, v2
 # CHECK-ENCODING: [0x57,0x41,0x45,0x6e]
 # CHECK-ENCODING: [0x57,0x20,0x01,0x62]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 41 45 6e <unknown>
-# CHECK-UNKNOWN: 57 20 01 62 <unknown>
+# CHECK-UNKNOWN: 6e454157 <unknown>
+# CHECK-UNKNOWN: 62012057 <unknown>
 
 vmsgeu.vx v9, v4, a0, v0.t, v2
 # CHECK-INST: vmsltu.vx v2, v4, a0
@@ -447,10 +447,10 @@ vmsgeu.vx v9, v4, a0, v0.t, v2
 # CHECK-ENCODING: [0xd7,0x24,0x90,0x62]
 # CHECK-ENCODING: [0xd7,0xa4,0x24,0x6a]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 41 45 6a <unknown>
-# CHECK-UNKNOWN: 57 21 01 62 <unknown>
-# CHECK-UNKNOWN: d7 24 90 62 <unknown>
-# CHECK-UNKNOWN: d7 a4 24 6a <unknown>
+# CHECK-UNKNOWN: 6a454157 <unknown>
+# CHECK-UNKNOWN: 62012157 <unknown>
+# CHECK-UNKNOWN: 629024d7 <unknown>
+# CHECK-UNKNOWN: 6a24a4d7 <unknown>
 
 vmsge.vx v8, v4, a0, v0.t, v2
 # CHECK-INST: vmslt.vx v2, v4, a0
@@ -462,7 +462,7 @@ vmsge.vx v8, v4, a0, v0.t, v2
 # CHECK-ENCODING: [0x57,0x24,0x80,0x62]
 # CHECK-ENCODING: [0x57,0x24,0x24,0x6a]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 41 45 6e <unknown>
-# CHECK-UNKNOWN: 57 21 01 62 <unknown>
-# CHECK-UNKNOWN: 57 24 80 62 <unknown>
-# CHECK-UNKNOWN: 57 24 24 6a <unknown>
+# CHECK-UNKNOWN: 6e454157 <unknown>
+# CHECK-UNKNOWN: 62012157 <unknown>
+# CHECK-UNKNOWN: 62802457 <unknown>
+# CHECK-UNKNOWN: 6a242457 <unknown>
diff --git a/llvm/test/MC/RISCV/rvv/convert.s b/llvm/test/MC/RISCV/rvv/convert.s
index 28c0a0fa837e..269c86cef473 100644
--- a/llvm/test/MC/RISCV/rvv/convert.s
+++ b/llvm/test/MC/RISCV/rvv/convert.s
@@ -15,256 +15,256 @@ vfcvt.xu.f.v v8, v4, v0.t
 # CHECK-INST: vfcvt.xu.f.v v8, v4, v0.t
 # CHECK-ENCODING: [0x57,0x14,0x40,0x48]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 14 40 48 <unknown>
+# CHECK-UNKNOWN: 48401457 <unknown>
 
 vfcvt.xu.f.v v8, v4
 # CHECK-INST: vfcvt.xu.f.v v8, v4
 # CHECK-ENCODING: [0x57,0x14,0x40,0x4a]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 14 40 4a <unknown>
+# CHECK-UNKNOWN: 4a401457 <unknown>
 
 vfcvt.x.f.v v8, v4, v0.t
 # CHECK-INST: vfcvt.x.f.v v8, v4, v0.t
 # CHECK-ENCODING: [0x57,0x94,0x40,0x48]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 94 40 48 <unknown>
+# CHECK-UNKNOWN: 48409457 <unknown>
 
 vfcvt.x.f.v v8, v4
 # CHECK-INST: vfcvt.x.f.v v8, v4
 # CHECK-ENCODING: [0x57,0x94,0x40,0x4a]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 94 40 4a <unknown>
+# CHECK-UNKNOWN: 4a409457 <unknown>
 
 vfcvt.f.xu.v v8, v4, v0.t
 # CHECK-INST: vfcvt.f.xu.v v8, v4, v0.t
 # CHECK-ENCODING: [0x57,0x14,0x41,0x48]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 14 41 48 <unknown>
+# CHECK-UNKNOWN: 48411457 <unknown>
 
 vfcvt.f.xu.v v8, v4
 # CHECK-INST: vfcvt.f.xu.v v8, v4
 # CHECK-ENCODING: [0x57,0x14,0x41,0x4a]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 14 41 4a <unknown>
+# CHECK-UNKNOWN: 4a411457 <unknown>
 
 vfcvt.f.x.v v8, v4, v0.t
 # CHECK-INST: vfcvt.f.x.v v8, v4, v0.t
 # CHECK-ENCODING: [0x57,0x94,0x41,0x48]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 94 41 48 <unknown>
+# CHECK-UNKNOWN: 48419457 <unknown>
 
 vfcvt.f.x.v v8, v4
 # CHECK-INST: vfcvt.f.x.v v8, v4
 # CHECK-ENCODING: [0x57,0x94,0x41,0x4a]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 94 41 4a <unknown>
+# CHECK-UNKNOWN: 4a419457 <unknown>
 
 vfcvt.rtz.xu.f.v v8, v4, v0.t
 # CHECK-INST: vfcvt.rtz.xu.f.v v8, v4, v0.t
 # CHECK-ENCODING: [0x57,0x14,0x43,0x48]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 14 43 48 <unknown>
+# CHECK-UNKNOWN: 48431457 <unknown>
 
 vfcvt.rtz.xu.f.v v8, v4
 # CHECK-INST: vfcvt.rtz.xu.f.v v8, v4
 # CHECK-ENCODING: [0x57,0x14,0x43,0x4a]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 14 43 4a <unknown>
+# CHECK-UNKNOWN: 4a431457 <unknown>
 
 vfcvt.rtz.x.f.v v8, v4, v0.t
 # CHECK-INST: vfcvt.rtz.x.f.v v8, v4, v0.t
 # CHECK-ENCODING: [0x57,0x94,0x43,0x48]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 94 43 48 <unknown>
+# CHECK-UNKNOWN: 48439457 <unknown>
 
 vfcvt.rtz.x.f.v v8, v4
 # CHECK-INST: vfcvt.rtz.x.f.v v8, v4
 # CHECK-ENCODING: [0x57,0x94,0x43,0x4a]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 94 43 4a <unknown>
+# CHECK-UNKNOWN: 4a439457 <unknown>
 
 vfwcvt.xu.f.v v8, v4, v0.t
 # CHECK-INST: vfwcvt.xu.f.v v8, v4, v0.t
 # CHECK-ENCODING: [0x57,0x14,0x44,0x48]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 14 44 48 <unknown>
+# CHECK-UNKNOWN: 48441457 <unknown>
 
 vfwcvt.xu.f.v v8, v4
 # CHECK-INST: vfwcvt.xu.f.v v8, v4
 # CHECK-ENCODING: [0x57,0x14,0x44,0x4a]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 14 44 4a <unknown>
+# CHECK-UNKNOWN: 4a441457 <unknown>
 
 vfwcvt.x.f.v v8, v4, v0.t
 # CHECK-INST: vfwcvt.x.f.v v8, v4, v0.t
 # CHECK-ENCODING: [0x57,0x94,0x44,0x48]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 94 44 48 <unknown>
+# CHECK-UNKNOWN: 48449457 <unknown>
 
 vfwcvt.x.f.v v8, v4
 # CHECK-INST: vfwcvt.x.f.v v8, v4
 # CHECK-ENCODING: [0x57,0x94,0x44,0x4a]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 94 44 4a <unknown>
+# CHECK-UNKNOWN: 4a449457 <unknown>
 
 vfwcvt.f.xu.v v8, v4, v0.t
 # CHECK-INST: vfwcvt.f.xu.v v8, v4, v0.t
 # CHECK-ENCODING: [0x57,0x14,0x45,0x48]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 14 45 48 <unknown>
+# CHECK-UNKNOWN: 48451457 <unknown>
 
 vfwcvt.f.xu.v v8, v4
 # CHECK-INST: vfwcvt.f.xu.v v8, v4
 # CHECK-ENCODING: [0x57,0x14,0x45,0x4a]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 14 45 4a <unknown>
+# CHECK-UNKNOWN: 4a451457 <unknown>
 
 vfwcvt.f.x.v v8, v4, v0.t
 # CHECK-INST: vfwcvt.f.x.v v8, v4, v0.t
 # CHECK-ENCODING: [0x57,0x94,0x45,0x48]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 94 45 48 <unknown>
+# CHECK-UNKNOWN: 48459457 <unknown>
 
 vfwcvt.f.x.v v8, v4
 # CHECK-INST: vfwcvt.f.x.v v8, v4
 # CHECK-ENCODING: [0x57,0x94,0x45,0x4a]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 94 45 4a <unknown>
+# CHECK-UNKNOWN: 4a459457 <unknown>
 
 vfwcvt.f.f.v v8, v4, v0.t
 # CHECK-INST: vfwcvt.f.f.v v8, v4, v0.t
 # CHECK-ENCODING: [0x57,0x14,0x46,0x48]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 14 46 48 <unknown>
+# CHECK-UNKNOWN: 48461457 <unknown>
 
 vfwcvt.f.f.v v8, v4
 # CHECK-INST: vfwcvt.f.f.v v8, v4
 # CHECK-ENCODING: [0x57,0x14,0x46,0x4a]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 14 46 4a <unknown>
+# CHECK-UNKNOWN: 4a461457 <unknown>
 
 vfwcvt.rtz.xu.f.v v8, v4, v0.t
 # CHECK-INST: vfwcvt.rtz.xu.f.v v8, v4, v0.t
 # CHECK-ENCODING: [0x57,0x14,0x47,0x48]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 14 47 48 <unknown>
+# CHECK-UNKNOWN: 48471457 <unknown>
 
 vfwcvt.rtz.xu.f.v v8, v4
 # CHECK-INST: vfwcvt.rtz.xu.f.v v8, v4
 # CHECK-ENCODING: [0x57,0x14,0x47,0x4a]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 14 47 4a <unknown>
+# CHECK-UNKNOWN: 4a471457 <unknown>
 
 vfwcvt.rtz.x.f.v v8, v4, v0.t
 # CHECK-INST: vfwcvt.rtz.x.f.v v8, v4, v0.t
 # CHECK-ENCODING: [0x57,0x94,0x47,0x48]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 94 47 48 <unknown>
+# CHECK-UNKNOWN: 48479457 <unknown>
 
 vfwcvt.rtz.x.f.v v8, v4
 # CHECK-INST: vfwcvt.rtz.x.f.v v8, v4
 # CHECK-ENCODING: [0x57,0x94,0x47,0x4a]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 94 47 4a <unknown>
+# CHECK-UNKNOWN: 4a479457 <unknown>
 
 vfncvt.xu.f.w v8, v4, v0.t
 # CHECK-INST: vfncvt.xu.f.w v8, v4, v0.t
 # CHECK-ENCODING: [0x57,0x14,0x48,0x48]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 14 48 48 <unknown>
+# CHECK-UNKNOWN: 48481457 <unknown>
 
 vfncvt.xu.f.w v4, v4, v0.t
 # CHECK-INST: vfncvt.xu.f.w v4, v4, v0.t
 # CHECK-ENCODING: [0x57,0x12,0x48,0x48]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 12 48 48 <unknown>
+# CHECK-UNKNOWN: 48481257 <unknown>
 
 vfncvt.xu.f.w v8, v4
 # CHECK-INST: vfncvt.xu.f.w v8, v4
 # CHECK-ENCODING: [0x57,0x14,0x48,0x4a]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 14 48 4a <unknown>
+# CHECK-UNKNOWN: 4a481457 <unknown>
 
 vfncvt.x.f.w v8, v4, v0.t
 # CHECK-INST: vfncvt.x.f.w v8, v4, v0.t
 # CHECK-ENCODING: [0x57,0x94,0x48,0x48]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 94 48 48 <unknown>
+# CHECK-UNKNOWN: 48489457 <unknown>
 
 vfncvt.x.f.w v8, v4
 # CHECK-INST: vfncvt.x.f.w v8, v4
 # CHECK-ENCODING: [0x57,0x94,0x48,0x4a]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 94 48 4a <unknown>
+# CHECK-UNKNOWN: 4a489457 <unknown>
 
 vfncvt.f.xu.w v8, v4, v0.t
 # CHECK-INST: vfncvt.f.xu.w v8, v4, v0.t
 # CHECK-ENCODING: [0x57,0x14,0x49,0x48]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 14 49 48 <unknown>
+# CHECK-UNKNOWN: 48491457 <unknown>
 
 vfncvt.f.xu.w v8, v4
 # CHECK-INST: vfncvt.f.xu.w v8, v4
 # CHECK-ENCODING: [0x57,0x14,0x49,0x4a]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 14 49 4a <unknown>
+# CHECK-UNKNOWN: 4a491457 <unknown>
 
 vfncvt.f.x.w v8, v4, v0.t
 # CHECK-INST: vfncvt.f.x.w v8, v4, v0.t
 # CHECK-ENCODING: [0x57,0x94,0x49,0x48]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 94 49 48 <unknown>
+# CHECK-UNKNOWN: 48499457 <unknown>
 
 vfncvt.f.x.w v8, v4
 # CHECK-INST: vfncvt.f.x.w v8, v4
 # CHECK-ENCODING: [0x57,0x94,0x49,0x4a]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 94 49 4a <unknown>
+# CHECK-UNKNOWN: 4a499457 <unknown>
 
 vfncvt.f.f.w v8, v4, v0.t
 # CHECK-INST: vfncvt.f.f.w v8, v4, v0.t
 # CHECK-ENCODING: [0x57,0x14,0x4a,0x48]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 14 4a 48 <unknown>
+# CHECK-UNKNOWN: 484a1457 <unknown>
 
 vfncvt.f.f.w v8, v4
 # CHECK-INST: vfncvt.f.f.w v8, v4
 # CHECK-ENCODING: [0x57,0x14,0x4a,0x4a]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 14 4a 4a <unknown>
+# CHECK-UNKNOWN: 4a4a1457 <unknown>
 
 vfncvt.rod.f.f.w v8, v4, v0.t
 # CHECK-INST: vfncvt.rod.f.f.w v8, v4, v0.t
 # CHECK-ENCODING: [0x57,0x94,0x4a,0x48]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 94 4a 48 <unknown>
+# CHECK-UNKNOWN: 484a9457 <unknown>
 
 vfncvt.rod.f.f.w v8, v4
 # CHECK-INST: vfncvt.rod.f.f.w v8, v4
 # CHECK-ENCODING: [0x57,0x94,0x4a,0x4a]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 94 4a 4a <unknown>
+# CHECK-UNKNOWN: 4a4a9457 <unknown>
 
 vfncvt.rtz.xu.f.w v8, v4, v0.t
 # CHECK-INST: vfncvt.rtz.xu.f.w v8, v4, v0.t
 # CHECK-ENCODING: [0x57,0x14,0x4b,0x48]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 14 4b 48 <unknown>
+# CHECK-UNKNOWN: 484b1457 <unknown>
 
 vfncvt.rtz.xu.f.w v8, v4
 # CHECK-INST: vfncvt.rtz.xu.f.w v8, v4
 # CHECK-ENCODING: [0x57,0x14,0x4b,0x4a]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 14 4b 4a <unknown>
+# CHECK-UNKNOWN: 4a4b1457 <unknown>
 
 vfncvt.rtz.x.f.w v8, v4, v0.t
 # CHECK-INST: vfncvt.rtz.x.f.w v8, v4, v0.t
 # CHECK-ENCODING: [0x57,0x94,0x4b,0x48]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 94 4b 48 <unknown>
+# CHECK-UNKNOWN: 484b9457 <unknown>
 
 vfncvt.rtz.x.f.w v8, v4
 # CHECK-INST: vfncvt.rtz.x.f.w v8, v4
 # CHECK-ENCODING: [0x57,0x94,0x4b,0x4a]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 94 4b 4a <unknown>
+# CHECK-UNKNOWN: 4a4b9457 <unknown>
diff --git a/llvm/test/MC/RISCV/rvv/div.s b/llvm/test/MC/RISCV/rvv/div.s
index 229124c671c6..aca04375f61e 100644
--- a/llvm/test/MC/RISCV/rvv/div.s
+++ b/llvm/test/MC/RISCV/rvv/div.s
@@ -12,94 +12,94 @@ vdivu.vv v8, v4, v20, v0.t
 # CHECK-INST: vdivu.vv v8, v4, v20, v0.t
 # CHECK-ENCODING: [0x57,0x24,0x4a,0x80]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 24 4a 80 <unknown>
+# CHECK-UNKNOWN: 804a2457 <unknown>
 
 vdivu.vv v8, v4, v20
 # CHECK-INST: vdivu.vv v8, v4, v20
 # CHECK-ENCODING: [0x57,0x24,0x4a,0x82]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 24 4a 82 <unknown>
+# CHECK-UNKNOWN: 824a2457 <unknown>
 
 vdivu.vx v8, v4, a0, v0.t
 # CHECK-INST: vdivu.vx v8, v4, a0, v0.t
 # CHECK-ENCODING: [0x57,0x64,0x45,0x80]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 64 45 80 <unknown>
+# CHECK-UNKNOWN: 80456457 <unknown>
 
 vdivu.vx v8, v4, a0
 # CHECK-INST: vdivu.vx v8, v4, a0
 # CHECK-ENCODING: [0x57,0x64,0x45,0x82]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 64 45 82 <unknown>
+# CHECK-UNKNOWN: 82456457 <unknown>
 
 vdiv.vv v8, v4, v20, v0.t
 # CHECK-INST: vdiv.vv v8, v4, v20, v0.t
 # CHECK-ENCODING: [0x57,0x24,0x4a,0x84]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 24 4a 84 <unknown>
+# CHECK-UNKNOWN: 844a2457 <unknown>
 
 vdiv.vv v8, v4, v20
 # CHECK-INST: vdiv.vv v8, v4, v20
 # CHECK-ENCODING: [0x57,0x24,0x4a,0x86]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 24 4a 86 <unknown>
+# CHECK-UNKNOWN: 864a2457 <unknown>
 
 vdiv.vx v8, v4, a0, v0.t
 # CHECK-INST: vdiv.vx v8, v4, a0, v0.t
 # CHECK-ENCODING: [0x57,0x64,0x45,0x84]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 64 45 84 <unknown>
+# CHECK-UNKNOWN: 84456457 <unknown>
 
 vdiv.vx v8, v4, a0
 # CHECK-INST: vdiv.vx v8, v4, a0
 # CHECK-ENCODING: [0x57,0x64,0x45,0x86]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 64 45 86 <unknown>
+# CHECK-UNKNOWN: 86456457 <unknown>
 
 vremu.vv v8, v4, v20, v0.t
 # CHECK-INST: vremu.vv v8, v4, v20, v0.t
 # CHECK-ENCODING: [0x57,0x24,0x4a,0x88]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 24 4a 88 <unknown>
+# CHECK-UNKNOWN: 884a2457 <unknown>
 
 vremu.vv v8, v4, v20
 # CHECK-INST: vremu.vv v8, v4, v20
 # CHECK-ENCODING: [0x57,0x24,0x4a,0x8a]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 24 4a 8a <unknown>
+# CHECK-UNKNOWN: 8a4a2457 <unknown>
 
 vremu.vx v8, v4, a0, v0.t
 # CHECK-INST: vremu.vx v8, v4, a0, v0.t
 # CHECK-ENCODING: [0x57,0x64,0x45,0x88]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 64 45 88 <unknown>
+# CHECK-UNKNOWN: 88456457 <unknown>
 
 vremu.vx v8, v4, a0
 # CHECK-INST: vremu.vx v8, v4, a0
 # CHECK-ENCODING: [0x57,0x64,0x45,0x8a]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 64 45 8a <unknown>
+# CHECK-UNKNOWN: 8a456457 <unknown>
 
 vrem.vv v8, v4, v20, v0.t
 # CHECK-INST: vrem.vv v8, v4, v20, v0.t
 # CHECK-ENCODING: [0x57,0x24,0x4a,0x8c]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 24 4a 8c <unknown>
+# CHECK-UNKNOWN: 8c4a2457 <unknown>
 
 vrem.vv v8, v4, v20
 # CHECK-INST: vrem.vv v8, v4, v20
 # CHECK-ENCODING: [0x57,0x24,0x4a,0x8e]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 24 4a 8e <unknown>
+# CHECK-UNKNOWN: 8e4a2457 <unknown>
 
 vrem.vx v8, v4, a0, v0.t
 # CHECK-INST: vrem.vx v8, v4, a0, v0.t
 # CHECK-ENCODING: [0x57,0x64,0x45,0x8c]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 64 45 8c <unknown>
+# CHECK-UNKNOWN: 8c456457 <unknown>
 
 vrem.vx v8, v4, a0
 # CHECK-INST: vrem.vx v8, v4, a0
 # CHECK-ENCODING: [0x57,0x64,0x45,0x8e]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 64 45 8e <unknown>
+# CHECK-UNKNOWN: 8e456457 <unknown>
diff --git a/llvm/test/MC/RISCV/rvv/ext.s b/llvm/test/MC/RISCV/rvv/ext.s
index 80cadc096104..3bf1351d77a7 100644
--- a/llvm/test/MC/RISCV/rvv/ext.s
+++ b/llvm/test/MC/RISCV/rvv/ext.s
@@ -12,70 +12,70 @@ vzext.vf2 v8, v4, v0.t
 # CHECK-INST: vzext.vf2 v8, v4, v0.t
 # CHECK-ENCODING: [0x57,0x24,0x43,0x48]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 24 43 48 <unknown>
+# CHECK-UNKNOWN: 48432457 <unknown>
 
 vzext.vf2 v8, v4
 # CHECK-INST: vzext.vf2 v8, v4
 # CHECK-ENCODING: [0x57,0x24,0x43,0x4a]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 24 43 4a <unknown>
+# CHECK-UNKNOWN: 4a432457 <unknown>
 
 vsext.vf2 v8, v4, v0.t
 # CHECK-INST: vsext.vf2 v8, v4, v0.t
 # CHECK-ENCODING: [0x57,0xa4,0x43,0x48]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 a4 43 48 <unknown>
+# CHECK-UNKNOWN: 4843a457 <unknown>
 
 vsext.vf2 v8, v4
 # CHECK-INST: vsext.vf2 v8, v4
 # CHECK-ENCODING: [0x57,0xa4,0x43,0x4a]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 a4 43 4a <unknown>
+# CHECK-UNKNOWN: 4a43a457 <unknown>
 
 vzext.vf4 v8, v4, v0.t
 # CHECK-INST: vzext.vf4 v8, v4, v0.t
 # CHECK-ENCODING: [0x57,0x24,0x42,0x48]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 24 42 48 <unknown>
+# CHECK-UNKNOWN: 48422457 <unknown>
 
 vzext.vf4 v8, v4
 # CHECK-INST: vzext.vf4 v8, v4
 # CHECK-ENCODING: [0x57,0x24,0x42,0x4a]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 24 42 4a <unknown>
+# CHECK-UNKNOWN: 4a422457 <unknown>
 
 vsext.vf4 v8, v4, v0.t
 # CHECK-INST: vsext.vf4 v8, v4, v0.t
 # CHECK-ENCODING: [0x57,0xa4,0x42,0x48]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 a4 42 48 <unknown>
+# CHECK-UNKNOWN: 4842a457 <unknown>
 
 vsext.vf4 v8, v4
 # CHECK-INST: vsext.vf4 v8, v4
 # CHECK-ENCODING: [0x57,0xa4,0x42,0x4a]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 a4 42 4a <unknown>
+# CHECK-UNKNOWN: 4a42a457 <unknown>
 
 vzext.vf8 v8, v4, v0.t
 # CHECK-INST: vzext.vf8 v8, v4, v0.t
 # CHECK-ENCODING: [0x57,0x24,0x41,0x48]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 24 41 48 <unknown>
+# CHECK-UNKNOWN: 48412457 <unknown>
 
 vzext.vf8 v8, v4
 # CHECK-INST: vzext.vf8 v8, v4
 # CHECK-ENCODING: [0x57,0x24,0x41,0x4a]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 24 41 4a <unknown>
+# CHECK-UNKNOWN: 4a412457 <unknown>
 
 vsext.vf8 v8, v4, v0.t
 # CHECK-INST: vsext.vf8 v8, v4, v0.t
 # CHECK-ENCODING: [0x57,0xa4,0x41,0x48]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 a4 41 48 <unknown>
+# CHECK-UNKNOWN: 4841a457 <unknown>
 
 vsext.vf8 v8, v4
 # CHECK-INST: vsext.vf8 v8, v4
 # CHECK-ENCODING: [0x57,0xa4,0x41,0x4a]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 a4 41 4a <unknown>
+# CHECK-UNKNOWN: 4a41a457 <unknown>
diff --git a/llvm/test/MC/RISCV/rvv/fadd.s b/llvm/test/MC/RISCV/rvv/fadd.s
index 60ffaf62ca6b..890b2c0ad68b 100644
--- a/llvm/test/MC/RISCV/rvv/fadd.s
+++ b/llvm/test/MC/RISCV/rvv/fadd.s
@@ -15,70 +15,70 @@ vfadd.vv v8, v4, v20, v0.t
 # CHECK-INST: vfadd.vv v8, v4, v20, v0.t
 # CHECK-ENCODING: [0x57,0x14,0x4a,0x00]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 14 4a 00 <unknown>
+# CHECK-UNKNOWN: 004a1457 <unknown>
 
 vfadd.vv v8, v4, v20
 # CHECK-INST: vfadd.vv v8, v4, v20
 # CHECK-ENCODING: [0x57,0x14,0x4a,0x02]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 14 4a 02 <unknown>
+# CHECK-UNKNOWN: 024a1457 <unknown>
 
 vfadd.vf v8, v4, fa0, v0.t
 # CHECK-INST: vfadd.vf v8, v4, fa0, v0.t
 # CHECK-ENCODING: [0x57,0x54,0x45,0x00]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 54 45 00 <unknown>
+# CHECK-UNKNOWN: 00455457 <unknown>
 
 vfadd.vf v8, v4, fa0
 # CHECK-INST: vfadd.vf v8, v4, fa0
 # CHECK-ENCODING: [0x57,0x54,0x45,0x02]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 54 45 02 <unknown>
+# CHECK-UNKNOWN: 02455457 <unknown>
 
 vfwadd.vv v8, v4, v20, v0.t
 # CHECK-INST: vfwadd.vv v8, v4, v20, v0.t
 # CHECK-ENCODING: [0x57,0x14,0x4a,0xc0]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 14 4a c0 <unknown>
+# CHECK-UNKNOWN: c04a1457 <unknown>
 
 vfwadd.vv v8, v4, v20
 # CHECK-INST: vfwadd.vv v8, v4, v20
 # CHECK-ENCODING: [0x57,0x14,0x4a,0xc2]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 14 4a c2 <unknown>
+# CHECK-UNKNOWN: c24a1457 <unknown>
 
 vfwadd.vf v8, v4, fa0, v0.t
 # CHECK-INST: vfwadd.vf v8, v4, fa0, v0.t
 # CHECK-ENCODING: [0x57,0x54,0x45,0xc0]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 54 45 c0 <unknown>
+# CHECK-UNKNOWN: c0455457 <unknown>
 
 vfwadd.vf v8, v4, fa0
 # CHECK-INST: vfwadd.vf v8, v4, fa0
 # CHECK-ENCODING: [0x57,0x54,0x45,0xc2]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 54 45 c2 <unknown>
+# CHECK-UNKNOWN: c2455457 <unknown>
 
 vfwadd.wv v8, v4, v20, v0.t
 # CHECK-INST: vfwadd.wv v8, v4, v20, v0.t
 # CHECK-ENCODING: [0x57,0x14,0x4a,0xd0]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 14 4a d0 <unknown>
+# CHECK-UNKNOWN: d04a1457 <unknown>
 
 vfwadd.wv v8, v4, v20
 # CHECK-INST: vfwadd.wv v8, v4, v20
 # CHECK-ENCODING: [0x57,0x14,0x4a,0xd2]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 14 4a d2 <unknown>
+# CHECK-UNKNOWN: d24a1457 <unknown>
 
 vfwadd.wf v8, v4, fa0, v0.t
 # CHECK-INST: vfwadd.wf v8, v4, fa0, v0.t
 # CHECK-ENCODING: [0x57,0x54,0x45,0xd0]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 54 45 d0 <unknown>
+# CHECK-UNKNOWN: d0455457 <unknown>
 
 vfwadd.wf v8, v4, fa0
 # CHECK-INST: vfwadd.wf v8, v4, fa0
 # CHECK-ENCODING: [0x57,0x54,0x45,0xd2]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 54 45 d2 <unknown>
+# CHECK-UNKNOWN: d2455457 <unknown>
diff --git a/llvm/test/MC/RISCV/rvv/fcompare.s b/llvm/test/MC/RISCV/rvv/fcompare.s
index 11dd7e05467b..3903bbdab650 100644
--- a/llvm/test/MC/RISCV/rvv/fcompare.s
+++ b/llvm/test/MC/RISCV/rvv/fcompare.s
@@ -15,148 +15,148 @@ vmfeq.vv v8, v4, v20, v0.t
 # CHECK-INST: vmfeq.vv v8, v4, v20, v0.t
 # CHECK-ENCODING: [0x57,0x14,0x4a,0x60]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 14 4a 60 <unknown>
+# CHECK-UNKNOWN: 604a1457 <unknown>
 
 vmfeq.vv v8, v4, v20
 # CHECK-INST: vmfeq.vv v8, v4, v20
 # CHECK-ENCODING: [0x57,0x14,0x4a,0x62]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 14 4a 62 <unknown>
+# CHECK-UNKNOWN: 624a1457 <unknown>
 
 vmfeq.vf v8, v4, fa0, v0.t
 # CHECK-INST: vmfeq.vf v8, v4, fa0, v0.t
 # CHECK-ENCODING: [0x57,0x54,0x45,0x60]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 54 45 60 <unknown>
+# CHECK-UNKNOWN: 60455457 <unknown>
 
 vmfeq.vf v8, v4, fa0
 # CHECK-INST: vmfeq.vf v8, v4, fa0
 # CHECK-ENCODING: [0x57,0x54,0x45,0x62]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 54 45 62 <unknown>
+# CHECK-UNKNOWN: 62455457 <unknown>
 
 vmfne.vv v8, v4, v20, v0.t
 # CHECK-INST: vmfne.vv v8, v4, v20, v0.t
 # CHECK-ENCODING: [0x57,0x14,0x4a,0x70]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 14 4a 70 <unknown>
+# CHECK-UNKNOWN: 704a1457 <unknown>
 
 vmfne.vv v8, v4, v20
 # CHECK-INST: vmfne.vv v8, v4, v20
 # CHECK-ENCODING: [0x57,0x14,0x4a,0x72]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 14 4a 72 <unknown>
+# CHECK-UNKNOWN: 724a1457 <unknown>
 
 vmfne.vf v8, v4, fa0, v0.t
 # CHECK-INST: vmfne.vf v8, v4, fa0, v0.t
 # CHECK-ENCODING: [0x57,0x54,0x45,0x70]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 54 45 70 <unknown>
+# CHECK-UNKNOWN: 70455457 <unknown>
 
 vmfne.vf v8, v4, fa0
 # CHECK-INST: vmfne.vf v8, v4, fa0
 # CHECK-ENCODING: [0x57,0x54,0x45,0x72]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 54 45 72 <unknown>
+# CHECK-UNKNOWN: 72455457 <unknown>
 
 vmflt.vv v8, v4, v20, v0.t
 # CHECK-INST: vmflt.vv v8, v4, v20, v0.t
 # CHECK-ENCODING: [0x57,0x14,0x4a,0x6c]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 14 4a 6c <unknown>
+# CHECK-UNKNOWN: 6c4a1457 <unknown>
 
 vmflt.vv v8, v4, v20
 # CHECK-INST: vmflt.vv v8, v4, v20
 # CHECK-ENCODING: [0x57,0x14,0x4a,0x6e]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 14 4a 6e <unknown>
+# CHECK-UNKNOWN: 6e4a1457 <unknown>
 
 vmflt.vf v8, v4, fa0, v0.t
 # CHECK-INST: vmflt.vf v8, v4, fa0, v0.t
 # CHECK-ENCODING: [0x57,0x54,0x45,0x6c]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 54 45 6c <unknown>
+# CHECK-UNKNOWN: 6c455457 <unknown>
 
 vmflt.vf v8, v4, fa0
 # CHECK-INST: vmflt.vf v8, v4, fa0
 # CHECK-ENCODING: [0x57,0x54,0x45,0x6e]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 54 45 6e <unknown>
+# CHECK-UNKNOWN: 6e455457 <unknown>
 
 vmfle.vv v8, v4, v20, v0.t
 # CHECK-INST: vmfle.vv v8, v4, v20, v0.t
 # CHECK-ENCODING: [0x57,0x14,0x4a,0x64]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 14 4a 64 <unknown>
+# CHECK-UNKNOWN: 644a1457 <unknown>
 
 vmfle.vv v8, v4, v20
 # CHECK-INST: vmfle.vv v8, v4, v20
 # CHECK-ENCODING: [0x57,0x14,0x4a,0x66]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 14 4a 66 <unknown>
+# CHECK-UNKNOWN: 664a1457 <unknown>
 
 vmfle.vf v8, v4, fa0, v0.t
 # CHECK-INST: vmfle.vf v8, v4, fa0, v0.t
 # CHECK-ENCODING: [0x57,0x54,0x45,0x64]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 54 45 64 <unknown>
+# CHECK-UNKNOWN: 64455457 <unknown>
 
 vmfle.vf v8, v4, fa0
 # CHECK-INST: vmfle.vf v8, v4, fa0
 # CHECK-ENCODING: [0x57,0x54,0x45,0x66]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 54 45 66 <unknown>
+# CHECK-UNKNOWN: 66455457 <unknown>
 
 vmfgt.vf v8, v4, fa0, v0.t
 # CHECK-INST: vmfgt.vf v8, v4, fa0, v0.t
 # CHECK-ENCODING: [0x57,0x54,0x45,0x74]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 54 45 74 <unknown>
+# CHECK-UNKNOWN: 74455457 <unknown>
 
 vmfgt.vf v8, v4, fa0
 # CHECK-INST: vmfgt.vf v8, v4, fa0
 # CHECK-ENCODING: [0x57,0x54,0x45,0x76]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 54 45 76 <unknown>
+# CHECK-UNKNOWN: 76455457 <unknown>
 
 vmfge.vf v8, v4, fa0, v0.t
 # CHECK-INST: vmfge.vf v8, v4, fa0, v0.t
 # CHECK-ENCODING: [0x57,0x54,0x45,0x7c]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 54 45 7c <unknown>
+# CHECK-UNKNOWN: 7c455457 <unknown>
 
 vmfge.vf v8, v4, fa0
 # CHECK-INST: vmfge.vf v8, v4, fa0
 # CHECK-ENCODING: [0x57,0x54,0x45,0x7e]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 54 45 7e <unknown>
+# CHECK-UNKNOWN: 7e455457 <unknown>
 
 vmfgt.vv v8, v20, v4, v0.t
 # CHECK-INST: vmflt.vv v8, v4, v20, v0.t
 # CHECK-ENCODING: [0x57,0x14,0x4a,0x6c]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 14 4a 6c <unknown>
+# CHECK-UNKNOWN: 6c4a1457 <unknown>
 
 vmfgt.vv v8, v20, v4
 # CHECK-INST: vmflt.vv v8, v4, v20
 # CHECK-ENCODING: [0x57,0x14,0x4a,0x6e]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 14 4a 6e <unknown>
+# CHECK-UNKNOWN: 6e4a1457 <unknown>
 
 vmfge.vv v8, v20, v4, v0.t
 # CHECK-INST: vmfle.vv v8, v4, v20, v0.t
 # CHECK-ENCODING: [0x57,0x14,0x4a,0x64]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 14 4a 64 <unknown>
+# CHECK-UNKNOWN: 644a1457 <unknown>
 
 vmfge.vv v8, v20, v4
 # CHECK-INST: vmfle.vv v8, v4, v20
 # CHECK-ENCODING: [0x57,0x14,0x4a,0x66]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 14 4a 66 <unknown>
+# CHECK-UNKNOWN: 664a1457 <unknown>
 
 vmfeq.vv v0, v4, v20, v0.t
 # CHECK-INST: vmfeq.vv v0, v4, v20, v0.t
 # CHECK-ENCODING: [0x57,0x10,0x4a,0x60]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 10 4a 60 <unknown>
+# CHECK-UNKNOWN: 604a1057 <unknown>
diff --git a/llvm/test/MC/RISCV/rvv/fdiv.s b/llvm/test/MC/RISCV/rvv/fdiv.s
index 7eb048f3cd58..aa3aae5841a2 100644
--- a/llvm/test/MC/RISCV/rvv/fdiv.s
+++ b/llvm/test/MC/RISCV/rvv/fdiv.s
@@ -15,34 +15,34 @@ vfdiv.vv v8, v4, v20, v0.t
 # CHECK-INST: vfdiv.vv v8, v4, v20, v0.t
 # CHECK-ENCODING: [0x57,0x14,0x4a,0x80]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 14 4a 80 <unknown>
+# CHECK-UNKNOWN: 804a1457 <unknown>
 
 vfdiv.vv v8, v4, v20
 # CHECK-INST: vfdiv.vv v8, v4, v20
 # CHECK-ENCODING: [0x57,0x14,0x4a,0x82]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 14 4a 82 <unknown>
+# CHECK-UNKNOWN: 824a1457 <unknown>
 
 vfdiv.vf v8, v4, fa0, v0.t
 # CHECK-INST: vfdiv.vf v8, v4, fa0, v0.t
 # CHECK-ENCODING: [0x57,0x54,0x45,0x80]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 54 45 80 <unknown>
+# CHECK-UNKNOWN: 80455457 <unknown>
 
 vfdiv.vf v8, v4, fa0
 # CHECK-INST: vfdiv.vf v8, v4, fa0
 # CHECK-ENCODING: [0x57,0x54,0x45,0x82]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 54 45 82 <unknown>
+# CHECK-UNKNOWN: 82455457 <unknown>
 
 vfrdiv.vf v8, v4, fa0, v0.t
 # CHECK-INST: vfrdiv.vf v8, v4, fa0, v0.t
 # CHECK-ENCODING: [0x57,0x54,0x45,0x84]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 54 45 84 <unknown>
+# CHECK-UNKNOWN: 84455457 <unknown>
 
 vfrdiv.vf v8, v4, fa0
 # CHECK-INST: vfrdiv.vf v8, v4, fa0
 # CHECK-ENCODING: [0x57,0x54,0x45,0x86]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 54 45 86 <unknown>
+# CHECK-UNKNOWN: 86455457 <unknown>
diff --git a/llvm/test/MC/RISCV/rvv/fmacc.s b/llvm/test/MC/RISCV/rvv/fmacc.s
index 129455901b93..8ca43da80961 100644
--- a/llvm/test/MC/RISCV/rvv/fmacc.s
+++ b/llvm/test/MC/RISCV/rvv/fmacc.s
@@ -15,286 +15,286 @@ vfmacc.vv v8, v20, v4, v0.t
 # CHECK-INST: vfmacc.vv v8, v20, v4, v0.t
 # CHECK-ENCODING: [0x57,0x14,0x4a,0xb0]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 14 4a b0 <unknown>
+# CHECK-UNKNOWN: b04a1457 <unknown>
 
 vfmacc.vv v8, v20, v4
 # CHECK-INST: vfmacc.vv v8, v20, v4
 # CHECK-ENCODING: [0x57,0x14,0x4a,0xb2]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 14 4a b2 <unknown>
+# CHECK-UNKNOWN: b24a1457 <unknown>
 
 vfmacc.vf v8, fa0, v4, v0.t
 # CHECK-INST: vfmacc.vf v8, fa0, v4, v0.t
 # CHECK-ENCODING: [0x57,0x54,0x45,0xb0]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 54 45 b0 <unknown>
+# CHECK-UNKNOWN: b0455457 <unknown>
 
 vfmacc.vf v8, fa0, v4
 # CHECK-INST: vfmacc.vf v8, fa0, v4
 # CHECK-ENCODING: [0x57,0x54,0x45,0xb2]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 54 45 b2 <unknown>
+# CHECK-UNKNOWN: b2455457 <unknown>
 
 vfnmacc.vv v8, v20, v4, v0.t
 # CHECK-INST: vfnmacc.vv v8, v20, v4, v0.t
 # CHECK-ENCODING: [0x57,0x14,0x4a,0xb4]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 14 4a b4 <unknown>
+# CHECK-UNKNOWN: b44a1457 <unknown>
 
 vfnmacc.vv v8, v20, v4
 # CHECK-INST: vfnmacc.vv v8, v20, v4
 # CHECK-ENCODING: [0x57,0x14,0x4a,0xb6]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 14 4a b6 <unknown>
+# CHECK-UNKNOWN: b64a1457 <unknown>
 
 vfnmacc.vf v8, fa0, v4, v0.t
 # CHECK-INST: vfnmacc.vf v8, fa0, v4, v0.t
 # CHECK-ENCODING: [0x57,0x54,0x45,0xb4]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 54 45 b4 <unknown>
+# CHECK-UNKNOWN: b4455457 <unknown>
 
 vfnmacc.vf v8, fa0, v4
 # CHECK-INST: vfnmacc.vf v8, fa0, v4
 # CHECK-ENCODING: [0x57,0x54,0x45,0xb6]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 54 45 b6 <unknown>
+# CHECK-UNKNOWN: b6455457 <unknown>
 
 vfmsac.vv v8, v20, v4, v0.t
 # CHECK-INST: vfmsac.vv v8, v20, v4, v0.t
 # CHECK-ENCODING: [0x57,0x14,0x4a,0xb8]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 14 4a b8 <unknown>
+# CHECK-UNKNOWN: b84a1457 <unknown>
 
 vfmsac.vv v8, v20, v4
 # CHECK-INST: vfmsac.vv v8, v20, v4
 # CHECK-ENCODING: [0x57,0x14,0x4a,0xba]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 14 4a ba <unknown>
+# CHECK-UNKNOWN: ba4a1457 <unknown>
 
 vfmsac.vf v8, fa0, v4, v0.t
 # CHECK-INST: vfmsac.vf v8, fa0, v4, v0.t
 # CHECK-ENCODING: [0x57,0x54,0x45,0xb8]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 54 45 b8 <unknown>
+# CHECK-UNKNOWN: b8455457 <unknown>
 
 vfmsac.vf v8, fa0, v4
 # CHECK-INST: vfmsac.vf v8, fa0, v4
 # CHECK-ENCODING: [0x57,0x54,0x45,0xba]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 54 45 ba <unknown>
+# CHECK-UNKNOWN: ba455457 <unknown>
 
 vfnmsac.vv v8, v20, v4, v0.t
 # CHECK-INST: vfnmsac.vv v8, v20, v4, v0.t
 # CHECK-ENCODING: [0x57,0x14,0x4a,0xbc]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 14 4a bc <unknown>
+# CHECK-UNKNOWN: bc4a1457 <unknown>
 
 vfnmsac.vv v8, v20, v4
 # CHECK-INST: vfnmsac.vv v8, v20, v4
 # CHECK-ENCODING: [0x57,0x14,0x4a,0xbe]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 14 4a be <unknown>
+# CHECK-UNKNOWN: be4a1457 <unknown>
 
 vfnmsac.vf v8, fa0, v4, v0.t
 # CHECK-INST: vfnmsac.vf v8, fa0, v4, v0.t
 # CHECK-ENCODING: [0x57,0x54,0x45,0xbc]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 54 45 bc <unknown>
+# CHECK-UNKNOWN: bc455457 <unknown>
 
 vfnmsac.vf v8, fa0, v4
 # CHECK-INST: vfnmsac.vf v8, fa0, v4
 # CHECK-ENCODING: [0x57,0x54,0x45,0xbe]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 54 45 be <unknown>
+# CHECK-UNKNOWN: be455457 <unknown>
 
 vfmadd.vv v8, v20, v4, v0.t
 # CHECK-INST: vfmadd.vv v8, v20, v4, v0.t
 # CHECK-ENCODING: [0x57,0x14,0x4a,0xa0]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 14 4a a0 <unknown>
+# CHECK-UNKNOWN: a04a1457 <unknown>
 
 vfmadd.vv v8, v20, v4
 # CHECK-INST: vfmadd.vv v8, v20, v4
 # CHECK-ENCODING: [0x57,0x14,0x4a,0xa2]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 14 4a a2 <unknown>
+# CHECK-UNKNOWN: a24a1457 <unknown>
 
 vfmadd.vf v8, fa0, v4, v0.t
 # CHECK-INST: vfmadd.vf v8, fa0, v4, v0.t
 # CHECK-ENCODING: [0x57,0x54,0x45,0xa0]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 54 45 a0 <unknown>
+# CHECK-UNKNOWN: a0455457 <unknown>
 
 vfmadd.vf v8, fa0, v4
 # CHECK-INST: vfmadd.vf v8, fa0, v4
 # CHECK-ENCODING: [0x57,0x54,0x45,0xa2]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 54 45 a2 <unknown>
+# CHECK-UNKNOWN: a2455457 <unknown>
 
 vfnmadd.vv v8, v20, v4, v0.t
 # CHECK-INST: vfnmadd.vv v8, v20, v4, v0.t
 # CHECK-ENCODING: [0x57,0x14,0x4a,0xa4]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 14 4a a4 <unknown>
+# CHECK-UNKNOWN: a44a1457 <unknown>
 
 vfnmadd.vv v8, v20, v4
 # CHECK-INST: vfnmadd.vv v8, v20, v4
 # CHECK-ENCODING: [0x57,0x14,0x4a,0xa6]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 14 4a a6 <unknown>
+# CHECK-UNKNOWN: a64a1457 <unknown>
 
 vfnmadd.vf v8, fa0, v4, v0.t
 # CHECK-INST: vfnmadd.vf v8, fa0, v4, v0.t
 # CHECK-ENCODING: [0x57,0x54,0x45,0xa4]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 54 45 a4 <unknown>
+# CHECK-UNKNOWN: a4455457 <unknown>
 
 vfnmadd.vf v8, fa0, v4
 # CHECK-INST: vfnmadd.vf v8, fa0, v4
 # CHECK-ENCODING: [0x57,0x54,0x45,0xa6]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 54 45 a6 <unknown>
+# CHECK-UNKNOWN: a6455457 <unknown>
 
 vfmsub.vv v8, v20, v4, v0.t
 # CHECK-INST: vfmsub.vv v8, v20, v4, v0.t
 # CHECK-ENCODING: [0x57,0x14,0x4a,0xa8]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 14 4a a8 <unknown>
+# CHECK-UNKNOWN: a84a1457 <unknown>
 
 vfmsub.vv v8, v20, v4
 # CHECK-INST: vfmsub.vv v8, v20, v4
 # CHECK-ENCODING: [0x57,0x14,0x4a,0xaa]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 14 4a aa <unknown>
+# CHECK-UNKNOWN: aa4a1457 <unknown>
 
 vfmsub.vf v8, fa0, v4, v0.t
 # CHECK-INST: vfmsub.vf v8, fa0, v4, v0.t
 # CHECK-ENCODING: [0x57,0x54,0x45,0xa8]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 54 45 a8 <unknown>
+# CHECK-UNKNOWN: a8455457 <unknown>
 
 vfmsub.vf v8, fa0, v4
 # CHECK-INST: vfmsub.vf v8, fa0, v4
 # CHECK-ENCODING: [0x57,0x54,0x45,0xaa]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 54 45 aa <unknown>
+# CHECK-UNKNOWN: aa455457 <unknown>
 
 vfnmsub.vv v8, v20, v4, v0.t
 # CHECK-INST: vfnmsub.vv v8, v20, v4, v0.t
 # CHECK-ENCODING: [0x57,0x14,0x4a,0xac]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 14 4a ac <unknown>
+# CHECK-UNKNOWN: ac4a1457 <unknown>
 
 vfnmsub.vv v8, v20, v4
 # CHECK-INST: vfnmsub.vv v8, v20, v4
 # CHECK-ENCODING: [0x57,0x14,0x4a,0xae]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 14 4a ae <unknown>
+# CHECK-UNKNOWN: ae4a1457 <unknown>
 
 vfnmsub.vf v8, fa0, v4, v0.t
 # CHECK-INST: vfnmsub.vf v8, fa0, v4, v0.t
 # CHECK-ENCODING: [0x57,0x54,0x45,0xac]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 54 45 ac <unknown>
+# CHECK-UNKNOWN: ac455457 <unknown>
 
 vfnmsub.vf v8, fa0, v4
 # CHECK-INST: vfnmsub.vf v8, fa0, v4
 # CHECK-ENCODING: [0x57,0x54,0x45,0xae]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 54 45 ae <unknown>
+# CHECK-UNKNOWN: ae455457 <unknown>
 
 vfwmacc.vv v8, v20, v4, v0.t
 # CHECK-INST: vfwmacc.vv v8, v20, v4, v0.t
 # CHECK-ENCODING: [0x57,0x14,0x4a,0xf0]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 14 4a f0 <unknown>
+# CHECK-UNKNOWN: f04a1457 <unknown>
 
 vfwmacc.vv v8, v20, v4
 # CHECK-INST: vfwmacc.vv v8, v20, v4
 # CHECK-ENCODING: [0x57,0x14,0x4a,0xf2]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 14 4a f2 <unknown>
+# CHECK-UNKNOWN: f24a1457 <unknown>
 
 vfwmacc.vf v8, fa0, v4, v0.t
 # CHECK-INST: vfwmacc.vf v8, fa0, v4, v0.t
 # CHECK-ENCODING: [0x57,0x54,0x45,0xf0]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 54 45 f0 <unknown>
+# CHECK-UNKNOWN: f0455457 <unknown>
 
 vfwmacc.vf v8, fa0, v4
 # CHECK-INST: vfwmacc.vf v8, fa0, v4
 # CHECK-ENCODING: [0x57,0x54,0x45,0xf2]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 54 45 f2 <unknown>
+# CHECK-UNKNOWN: f2455457 <unknown>
 
 vfwnmacc.vv v8, v20, v4, v0.t
 # CHECK-INST: vfwnmacc.vv v8, v20, v4, v0.t
 # CHECK-ENCODING: [0x57,0x14,0x4a,0xf4]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 14 4a f4 <unknown>
+# CHECK-UNKNOWN: f44a1457 <unknown>
 
 vfwnmacc.vv v8, v20, v4
 # CHECK-INST: vfwnmacc.vv v8, v20, v4
 # CHECK-ENCODING: [0x57,0x14,0x4a,0xf6]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 14 4a f6 <unknown>
+# CHECK-UNKNOWN: f64a1457 <unknown>
 
 vfwnmacc.vf v8, fa0, v4, v0.t
 # CHECK-INST: vfwnmacc.vf v8, fa0, v4, v0.t
 # CHECK-ENCODING: [0x57,0x54,0x45,0xf4]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 54 45 f4 <unknown>
+# CHECK-UNKNOWN: f4455457 <unknown>
 
 vfwnmacc.vf v8, fa0, v4
 # CHECK-INST: vfwnmacc.vf v8, fa0, v4
 # CHECK-ENCODING: [0x57,0x54,0x45,0xf6]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 54 45 f6 <unknown>
+# CHECK-UNKNOWN: f6455457 <unknown>
 
 vfwmsac.vv v8, v20, v4, v0.t
 # CHECK-INST: vfwmsac.vv v8, v20, v4, v0.t
 # CHECK-ENCODING: [0x57,0x14,0x4a,0xf8]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 14 4a f8 <unknown>
+# CHECK-UNKNOWN: f84a1457 <unknown>
 
 vfwmsac.vv v8, v20, v4
 # CHECK-INST: vfwmsac.vv v8, v20, v4
 # CHECK-ENCODING: [0x57,0x14,0x4a,0xfa]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 14 4a fa <unknown>
+# CHECK-UNKNOWN: fa4a1457 <unknown>
 
 vfwmsac.vf v8, fa0, v4, v0.t
 # CHECK-INST: vfwmsac.vf v8, fa0, v4, v0.t
 # CHECK-ENCODING: [0x57,0x54,0x45,0xf8]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 54 45 f8 <unknown>
+# CHECK-UNKNOWN: f8455457 <unknown>
 
 vfwmsac.vf v8, fa0, v4
 # CHECK-INST: vfwmsac.vf v8, fa0, v4
 # CHECK-ENCODING: [0x57,0x54,0x45,0xfa]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 54 45 fa <unknown>
+# CHECK-UNKNOWN: fa455457 <unknown>
 
 vfwnmsac.vv v8, v20, v4, v0.t
 # CHECK-INST: vfwnmsac.vv v8, v20, v4, v0.t
 # CHECK-ENCODING: [0x57,0x14,0x4a,0xfc]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 14 4a fc <unknown>
+# CHECK-UNKNOWN: fc4a1457 <unknown>
 
 vfwnmsac.vv v8, v20, v4
 # CHECK-INST: vfwnmsac.vv v8, v20, v4
 # CHECK-ENCODING: [0x57,0x14,0x4a,0xfe]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 14 4a fe <unknown>
+# CHECK-UNKNOWN: fe4a1457 <unknown>
 
 vfwnmsac.vf v8, fa0, v4, v0.t
 # CHECK-INST: vfwnmsac.vf v8, fa0, v4, v0.t
 # CHECK-ENCODING: [0x57,0x54,0x45,0xfc]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 54 45 fc <unknown>
+# CHECK-UNKNOWN: fc455457 <unknown>
 
 vfwnmsac.vf v8, fa0, v4
 # CHECK-INST: vfwnmsac.vf v8, fa0, v4
 # CHECK-ENCODING: [0x57,0x54,0x45,0xfe]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 54 45 fe <unknown>
+# CHECK-UNKNOWN: fe455457 <unknown>
diff --git a/llvm/test/MC/RISCV/rvv/fminmax.s b/llvm/test/MC/RISCV/rvv/fminmax.s
index c8aab38e1fc8..f7e85ed31c33 100644
--- a/llvm/test/MC/RISCV/rvv/fminmax.s
+++ b/llvm/test/MC/RISCV/rvv/fminmax.s
@@ -15,46 +15,46 @@ vfmin.vv v8, v4, v20, v0.t
 # CHECK-INST: vfmin.vv v8, v4, v20, v0.t
 # CHECK-ENCODING: [0x57,0x14,0x4a,0x10]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 14 4a 10 <unknown>
+# CHECK-UNKNOWN: 104a1457 <unknown>
 
 vfmin.vv v8, v4, v20
 # CHECK-INST: vfmin.vv v8, v4, v20
 # CHECK-ENCODING: [0x57,0x14,0x4a,0x12]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 14 4a 12 <unknown>
+# CHECK-UNKNOWN: 124a1457 <unknown>
 
 vfmin.vf v8, v4, fa0, v0.t
 # CHECK-INST: vfmin.vf v8, v4, fa0, v0.t
 # CHECK-ENCODING: [0x57,0x54,0x45,0x10]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 54 45 10 <unknown>
+# CHECK-UNKNOWN: 10455457 <unknown>
 
 vfmin.vf v8, v4, fa0
 # CHECK-INST: vfmin.vf v8, v4, fa0
 # CHECK-ENCODING: [0x57,0x54,0x45,0x12]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 54 45 12 <unknown>
+# CHECK-UNKNOWN: 12455457 <unknown>
 
 vfmax.vv v8, v4, v20, v0.t
 # CHECK-INST: vfmax.vv v8, v4, v20, v0.t
 # CHECK-ENCODING: [0x57,0x14,0x4a,0x18]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 14 4a 18 <unknown>
+# CHECK-UNKNOWN: 184a1457 <unknown>
 
 vfmax.vv v8, v4, v20
 # CHECK-INST: vfmax.vv v8, v4, v20
 # CHECK-ENCODING: [0x57,0x14,0x4a,0x1a]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 14 4a 1a <unknown>
+# CHECK-UNKNOWN: 1a4a1457 <unknown>
 
 vfmax.vf v8, v4, fa0, v0.t
 # CHECK-INST: vfmax.vf v8, v4, fa0, v0.t
 # CHECK-ENCODING: [0x57,0x54,0x45,0x18]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 54 45 18 <unknown>
+# CHECK-UNKNOWN: 18455457 <unknown>
 
 vfmax.vf v8, v4, fa0
 # CHECK-INST: vfmax.vf v8, v4, fa0
 # CHECK-ENCODING: [0x57,0x54,0x45,0x1a]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 54 45 1a <unknown>
+# CHECK-UNKNOWN: 1a455457 <unknown>
diff --git a/llvm/test/MC/RISCV/rvv/fmul.s b/llvm/test/MC/RISCV/rvv/fmul.s
index 42c37932043c..9cd6e5287750 100644
--- a/llvm/test/MC/RISCV/rvv/fmul.s
+++ b/llvm/test/MC/RISCV/rvv/fmul.s
@@ -15,46 +15,46 @@ vfmul.vv v8, v4, v20, v0.t
 # CHECK-INST: vfmul.vv v8, v4, v20, v0.t
 # CHECK-ENCODING: [0x57,0x14,0x4a,0x90]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 14 4a 90 <unknown>
+# CHECK-UNKNOWN: 904a1457 <unknown>
 
 vfmul.vv v8, v4, v20
 # CHECK-INST: vfmul.vv v8, v4, v20
 # CHECK-ENCODING: [0x57,0x14,0x4a,0x92]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 14 4a 92 <unknown>
+# CHECK-UNKNOWN: 924a1457 <unknown>
 
 vfmul.vf v8, v4, fa0, v0.t
 # CHECK-INST: vfmul.vf v8, v4, fa0, v0.t
 # CHECK-ENCODING: [0x57,0x54,0x45,0x90]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 54 45 90 <unknown>
+# CHECK-UNKNOWN: 90455457 <unknown>
 
 vfmul.vf v8, v4, fa0
 # CHECK-INST: vfmul.vf v8, v4, fa0
 # CHECK-ENCODING: [0x57,0x54,0x45,0x92]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 54 45 92 <unknown>
+# CHECK-UNKNOWN: 92455457 <unknown>
 
 vfwmul.vv v8, v4, v20, v0.t
 # CHECK-INST: vfwmul.vv v8, v4, v20, v0.t
 # CHECK-ENCODING: [0x57,0x14,0x4a,0xe0]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 14 4a e0 <unknown>
+# CHECK-UNKNOWN: e04a1457 <unknown>
 
 vfwmul.vv v8, v4, v20
 # CHECK-INST: vfwmul.vv v8, v4, v20
 # CHECK-ENCODING: [0x57,0x14,0x4a,0xe2]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 14 4a e2 <unknown>
+# CHECK-UNKNOWN: e24a1457 <unknown>
 
 vfwmul.vf v8, v4, fa0, v0.t
 # CHECK-INST: vfwmul.vf v8, v4, fa0, v0.t
 # CHECK-ENCODING: [0x57,0x54,0x45,0xe0]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 54 45 e0 <unknown>
+# CHECK-UNKNOWN: e0455457 <unknown>
 
 vfwmul.vf v8, v4, fa0
 # CHECK-INST: vfwmul.vf v8, v4, fa0
 # CHECK-ENCODING: [0x57,0x54,0x45,0xe2]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 54 45 e2 <unknown>
+# CHECK-UNKNOWN: e2455457 <unknown>
diff --git a/llvm/test/MC/RISCV/rvv/fmv.s b/llvm/test/MC/RISCV/rvv/fmv.s
index a5b814f130fb..2534b5171b3e 100644
--- a/llvm/test/MC/RISCV/rvv/fmv.s
+++ b/llvm/test/MC/RISCV/rvv/fmv.s
@@ -15,16 +15,16 @@ vfmv.v.f v8, fa0
 # CHECK-INST: vfmv.v.f v8, fa0
 # CHECK-ENCODING: [0x57,0x54,0x05,0x5e]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 54 05 5e <unknown>
+# CHECK-UNKNOWN: 5e055457 <unknown>
 
 vfmv.f.s fa0, v4
 # CHECK-INST: vfmv.f.s fa0, v4
 # CHECK-ENCODING: [0x57,0x15,0x40,0x42]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 15 40 42 <unknown>
+# CHECK-UNKNOWN: 42401557 <unknown>
 
 vfmv.s.f v8, fa0
 # CHECK-INST: vfmv.s.f v8, fa0
 # CHECK-ENCODING: [0x57,0x54,0x05,0x42]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 54 05 42 <unknown>
+# CHECK-UNKNOWN: 42055457 <unknown>
diff --git a/llvm/test/MC/RISCV/rvv/fothers.s b/llvm/test/MC/RISCV/rvv/fothers.s
index 451f6ca39ccc..997115f96bd9 100644
--- a/llvm/test/MC/RISCV/rvv/fothers.s
+++ b/llvm/test/MC/RISCV/rvv/fothers.s
@@ -13,76 +13,76 @@ vfsqrt.v v8, v4, v0.t
 # CHECK-INST: vfsqrt.v v8, v4, v0.t
 # CHECK-ENCODING: [0x57,0x14,0x40,0x4c]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 14 40 4c <unknown>
+# CHECK-UNKNOWN: 4c401457 <unknown>
 
 vfsqrt.v v8, v4
 # CHECK-INST: vfsqrt.v v8, v4
 # CHECK-ENCODING: [0x57,0x14,0x40,0x4e]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 14 40 4e <unknown>
+# CHECK-UNKNOWN: 4e401457 <unknown>
 
 vfrsqrt7.v v8, v4, v0.t
 # CHECK-INST: vfrsqrt7.v v8, v4, v0.t
 # CHECK-ENCODING: [0x57,0x14,0x42,0x4c]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 14 42 4c <unknown>
+# CHECK-UNKNOWN: 4c421457 <unknown>
 
 vfrsqrt7.v v8, v4
 # CHECK-INST: vfrsqrt7.v v8, v4
 # CHECK-ENCODING: [0x57,0x14,0x42,0x4e]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 14 42 4e <unknown>
+# CHECK-UNKNOWN: 4e421457 <unknown>
 
 vfrec7.v v8, v4, v0.t
 # CHECK-INST: vfrec7.v v8, v4, v0.t
 # CHECK-ENCODING: [0x57,0x94,0x42,0x4c]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 94 42 4c <unknown>
+# CHECK-UNKNOWN: 4c429457 <unknown>
 
 vfrec7.v v8, v4
 # CHECK-INST: vfrec7.v v8, v4
 # CHECK-ENCODING: [0x57,0x94,0x42,0x4e]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 94 42 4e <unknown>
+# CHECK-UNKNOWN: 4e429457 <unknown>
 
 vfclass.v v8, v4, v0.t
 # CHECK-INST: vfclass.v v8, v4, v0.t
 # CHECK-ENCODING: [0x57,0x14,0x48,0x4c]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 14 48 4c <unknown>
+# CHECK-UNKNOWN: 4c481457 <unknown>
 
 vfclass.v v8, v4
 # CHECK-INST: vfclass.v v8, v4
 # CHECK-ENCODING: [0x57,0x14,0x48,0x4e]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 14 48 4e <unknown>
+# CHECK-UNKNOWN: 4e481457 <unknown>
 
 vfmerge.vfm v8, v4, fa0, v0
 # CHECK-INST: vfmerge.vfm v8, v4, fa0, v0
 # CHECK-ENCODING: [0x57,0x54,0x45,0x5c]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 54 45 5c <unknown>
+# CHECK-UNKNOWN: 5c455457 <unknown>
 
 vfslide1up.vf v8, v4, fa0, v0.t
 # CHECK-INST: vfslide1up.vf v8, v4, fa0, v0.t
 # CHECK-ENCODING: [0x57,0x54,0x45,0x38]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 54 45 38 <unknown>
+# CHECK-UNKNOWN: 38455457 <unknown>
 
 vfslide1up.vf v8, v4, fa0
 # CHECK-INST: vfslide1up.vf v8, v4, fa0
 # CHECK-ENCODING: [0x57,0x54,0x45,0x3a]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 54 45 3a <unknown>
+# CHECK-UNKNOWN: 3a455457 <unknown>
 
 vfslide1down.vf v8, v4, fa0, v0.t
 # CHECK-INST: vfslide1down.vf v8, v4, fa0, v0.t
 # CHECK-ENCODING: [0x57,0x54,0x45,0x3c]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 54 45 3c <unknown>
+# CHECK-UNKNOWN: 3c455457 <unknown>
 
 vfslide1down.vf v8, v4, fa0
 # CHECK-INST: vfslide1down.vf v8, v4, fa0
 # CHECK-ENCODING: [0x57,0x54,0x45,0x3e]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 54 45 3e <unknown>
+# CHECK-UNKNOWN: 3e455457 <unknown>
diff --git a/llvm/test/MC/RISCV/rvv/freduction.s b/llvm/test/MC/RISCV/rvv/freduction.s
index fca64372b15f..12326942e6e8 100644
--- a/llvm/test/MC/RISCV/rvv/freduction.s
+++ b/llvm/test/MC/RISCV/rvv/freduction.s
@@ -15,76 +15,76 @@ vfredosum.vs v8, v4, v20, v0.t
 # CHECK-INST: vfredosum.vs v8, v4, v20, v0.t
 # CHECK-ENCODING: [0x57,0x14,0x4a,0x0c]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 14 4a 0c <unknown>
+# CHECK-UNKNOWN: 0c4a1457 <unknown>
 
 vfredosum.vs v8, v4, v20
 # CHECK-INST: vfredosum.vs v8, v4, v20
 # CHECK-ENCODING: [0x57,0x14,0x4a,0x0e]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 14 4a 0e <unknown>
+# CHECK-UNKNOWN: 0e4a1457 <unknown>
 
 vfredusum.vs v8, v4, v20, v0.t
 # CHECK-INST: vfredusum.vs v8, v4, v20, v0.t
 # CHECK-ENCODING: [0x57,0x14,0x4a,0x04]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 14 4a 04 <unknown>
+# CHECK-UNKNOWN: 044a1457 <unknown>
 
 vfredusum.vs v8, v4, v20
 # CHECK-INST: vfredusum.vs v8, v4, v20
 # CHECK-ENCODING: [0x57,0x14,0x4a,0x06]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 14 4a 06 <unknown>
+# CHECK-UNKNOWN: 064a1457 <unknown>
 
 vfredmax.vs v8, v4, v20, v0.t
 # CHECK-INST: vfredmax.vs v8, v4, v20, v0.t
 # CHECK-ENCODING: [0x57,0x14,0x4a,0x1c]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 14 4a 1c <unknown>
+# CHECK-UNKNOWN: 1c4a1457 <unknown>
 
 vfredmax.vs v8, v4, v20
 # CHECK-INST: vfredmax.vs v8, v4, v20
 # CHECK-ENCODING: [0x57,0x14,0x4a,0x1e]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 14 4a 1e <unknown>
+# CHECK-UNKNOWN: 1e4a1457 <unknown>
 
 vfredmin.vs v8, v4, v20, v0.t
 # CHECK-INST: vfredmin.vs v8, v4, v20, v0.t
 # CHECK-ENCODING: [0x57,0x14,0x4a,0x14]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 14 4a 14 <unknown>
+# CHECK-UNKNOWN: 144a1457 <unknown>
 
 vfredmin.vs v8, v4, v20
 # CHECK-INST: vfredmin.vs v8, v4, v20
 # CHECK-ENCODING: [0x57,0x14,0x4a,0x16]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 14 4a 16 <unknown>
+# CHECK-UNKNOWN: 164a1457 <unknown>
 
 vfwredosum.vs v8, v4, v20, v0.t
 # CHECK-INST: vfwredosum.vs v8, v4, v20, v0.t
 # CHECK-ENCODING: [0x57,0x14,0x4a,0xcc]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 14 4a cc <unknown>
+# CHECK-UNKNOWN: cc4a1457 <unknown>
 
 vfwredosum.vs v8, v4, v20
 # CHECK-INST: vfwredosum.vs v8, v4, v20
 # CHECK-ENCODING: [0x57,0x14,0x4a,0xce]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 14 4a ce <unknown>
+# CHECK-UNKNOWN: ce4a1457 <unknown>
 
 vfwredusum.vs v8, v4, v20, v0.t
 # CHECK-INST: vfwredusum.vs v8, v4, v20, v0.t
 # CHECK-ENCODING: [0x57,0x14,0x4a,0xc4]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 14 4a c4 <unknown>
+# CHECK-UNKNOWN: c44a1457 <unknown>
 
 vfwredusum.vs v8, v4, v20
 # CHECK-INST: vfwredusum.vs v8, v4, v20
 # CHECK-ENCODING: [0x57,0x14,0x4a,0xc6]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 14 4a c6 <unknown>
+# CHECK-UNKNOWN: c64a1457 <unknown>
 
 vfredosum.vs v0, v4, v20, v0.t
 # CHECK-INST: vfredosum.vs v0, v4, v20, v0.t
 # CHECK-ENCODING: [0x57,0x10,0x4a,0x0c]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 10 4a 0c <unknown>
+# CHECK-UNKNOWN: 0c4a1057 <unknown>
diff --git a/llvm/test/MC/RISCV/rvv/fsub.s b/llvm/test/MC/RISCV/rvv/fsub.s
index a8f2bc6260b9..62ff2e744c78 100644
--- a/llvm/test/MC/RISCV/rvv/fsub.s
+++ b/llvm/test/MC/RISCV/rvv/fsub.s
@@ -15,82 +15,82 @@ vfsub.vv v8, v4, v20, v0.t
 # CHECK-INST: vfsub.vv v8, v4, v20, v0.t
 # CHECK-ENCODING: [0x57,0x14,0x4a,0x08]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 14 4a 08 <unknown>
+# CHECK-UNKNOWN: 084a1457 <unknown>
 
 vfsub.vv v8, v4, v20
 # CHECK-INST: vfsub.vv v8, v4, v20
 # CHECK-ENCODING: [0x57,0x14,0x4a,0x0a]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 14 4a 0a <unknown>
+# CHECK-UNKNOWN: 0a4a1457 <unknown>
 
 vfsub.vf v8, v4, fa0, v0.t
 # CHECK-INST: vfsub.vf v8, v4, fa0, v0.t
 # CHECK-ENCODING: [0x57,0x54,0x45,0x08]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 54 45 08 <unknown>
+# CHECK-UNKNOWN: 08455457 <unknown>
 
 vfsub.vf v8, v4, fa0
 # CHECK-INST: vfsub.vf v8, v4, fa0
 # CHECK-ENCODING: [0x57,0x54,0x45,0x0a]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 54 45 0a <unknown>
+# CHECK-UNKNOWN: 0a455457 <unknown>
 
 vfrsub.vf v8, v4, fa0, v0.t
 # CHECK-INST: vfrsub.vf v8, v4, fa0, v0.t
 # CHECK-ENCODING: [0x57,0x54,0x45,0x9c]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 54 45 9c <unknown>
+# CHECK-UNKNOWN: 9c455457 <unknown>
 
 vfrsub.vf v8, v4, fa0
 # CHECK-INST: vfrsub.vf v8, v4, fa0
 # CHECK-ENCODING: [0x57,0x54,0x45,0x9e]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 54 45 9e <unknown>
+# CHECK-UNKNOWN: 9e455457 <unknown>
 
 vfwsub.vv v8, v4, v20, v0.t
 # CHECK-INST: vfwsub.vv v8, v4, v20, v0.t
 # CHECK-ENCODING: [0x57,0x14,0x4a,0xc8]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 14 4a c8 <unknown>
+# CHECK-UNKNOWN: c84a1457 <unknown>
 
 vfwsub.vv v8, v4, v20
 # CHECK-INST: vfwsub.vv v8, v4, v20
 # CHECK-ENCODING: [0x57,0x14,0x4a,0xca]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 14 4a ca <unknown>
+# CHECK-UNKNOWN: ca4a1457 <unknown>
 
 vfwsub.vf v8, v4, fa0, v0.t
 # CHECK-INST: vfwsub.vf v8, v4, fa0, v0.t
 # CHECK-ENCODING: [0x57,0x54,0x45,0xc8]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 54 45 c8 <unknown>
+# CHECK-UNKNOWN: c8455457 <unknown>
 
 vfwsub.vf v8, v4, fa0
 # CHECK-INST: vfwsub.vf v8, v4, fa0
 # CHECK-ENCODING: [0x57,0x54,0x45,0xca]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 54 45 ca <unknown>
+# CHECK-UNKNOWN: ca455457 <unknown>
 
 vfwsub.wv v8, v4, v20, v0.t
 # CHECK-INST: vfwsub.wv v8, v4, v20, v0.t
 # CHECK-ENCODING: [0x57,0x14,0x4a,0xd8]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 14 4a d8 <unknown>
+# CHECK-UNKNOWN: d84a1457 <unknown>
 
 vfwsub.wv v8, v4, v20
 # CHECK-INST: vfwsub.wv v8, v4, v20
 # CHECK-ENCODING: [0x57,0x14,0x4a,0xda]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 14 4a da <unknown>
+# CHECK-UNKNOWN: da4a1457 <unknown>
 
 vfwsub.wf v8, v4, fa0, v0.t
 # CHECK-INST: vfwsub.wf v8, v4, fa0, v0.t
 # CHECK-ENCODING: [0x57,0x54,0x45,0xd8]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 54 45 d8 <unknown>
+# CHECK-UNKNOWN: d8455457 <unknown>
 
 vfwsub.wf v8, v4, fa0
 # CHECK-INST: vfwsub.wf v8, v4, fa0
 # CHECK-ENCODING: [0x57,0x54,0x45,0xda]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 54 45 da <unknown>
+# CHECK-UNKNOWN: da455457 <unknown>
diff --git a/llvm/test/MC/RISCV/rvv/load.s b/llvm/test/MC/RISCV/rvv/load.s
index 23357df88d3f..3c251a3a8d75 100644
--- a/llvm/test/MC/RISCV/rvv/load.s
+++ b/llvm/test/MC/RISCV/rvv/load.s
@@ -12,382 +12,382 @@ vlm.v v0, (a0)
 # CHECK-INST: vlm.v v0, (a0)
 # CHECK-ENCODING: [0x07,0x00,0xb5,0x02]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 00 b5 02 <unknown>
+# CHECK-UNKNOWN: 02b50007 <unknown>
 
 vlm.v v8, (a0)
 # CHECK-INST: vlm.v v8, (a0)
 # CHECK-ENCODING: [0x07,0x04,0xb5,0x02]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 04 b5 02 <unknown>
+# CHECK-UNKNOWN: 02b50407 <unknown>
 
 vle8.v v8, (a0), v0.t
 # CHECK-INST: vle8.v v8, (a0), v0.t
 # CHECK-ENCODING: [0x07,0x04,0x05,0x00]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 04 05 00 <unknown>
+# CHECK-UNKNOWN: 00050407 <unknown>
 
 vle8.v v8, (a0)
 # CHECK-INST: vle8.v v8, (a0)
 # CHECK-ENCODING: [0x07,0x04,0x05,0x02]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 04 05 02 <unknown>
+# CHECK-UNKNOWN: 02050407 <unknown>
 
 vle16.v v8, (a0), v0.t
 # CHECK-INST: vle16.v v8, (a0), v0.t
 # CHECK-ENCODING: [0x07,0x54,0x05,0x00]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 54 05 00 <unknown>
+# CHECK-UNKNOWN: 00055407 <unknown>
 
 vle16.v v8, (a0)
 # CHECK-INST: vle16.v v8, (a0)
 # CHECK-ENCODING: [0x07,0x54,0x05,0x02]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 54 05 02 <unknown>
+# CHECK-UNKNOWN: 02055407 <unknown>
 
 vle32.v v8, (a0), v0.t
 # CHECK-INST: vle32.v v8, (a0), v0.t
 # CHECK-ENCODING: [0x07,0x64,0x05,0x00]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 64 05 00 <unknown>
+# CHECK-UNKNOWN: 00056407 <unknown>
 
 vle32.v v8, (a0)
 # CHECK-INST: vle32.v v8, (a0)
 # CHECK-ENCODING: [0x07,0x64,0x05,0x02]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 64 05 02 <unknown>
+# CHECK-UNKNOWN: 02056407 <unknown>
 
 vle64.v v8, (a0), v0.t
 # CHECK-INST: vle64.v v8, (a0), v0.t
 # CHECK-ENCODING: [0x07,0x74,0x05,0x00]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 74 05 00 <unknown>
+# CHECK-UNKNOWN: 00057407 <unknown>
 
 vle64.v v8, (a0)
 # CHECK-INST: vle64.v v8, (a0)
 # CHECK-ENCODING: [0x07,0x74,0x05,0x02]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 74 05 02 <unknown>
+# CHECK-UNKNOWN: 02057407 <unknown>
 
 vle8ff.v v8, (a0), v0.t
 # CHECK-INST: vle8ff.v v8, (a0), v0.t
 # CHECK-ENCODING: [0x07,0x04,0x05,0x01]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 04 05 01 <unknown>
+# CHECK-UNKNOWN: 01050407 <unknown>
 
 vle8ff.v v8, (a0)
 # CHECK-INST: vle8ff.v v8, (a0)
 # CHECK-ENCODING: [0x07,0x04,0x05,0x03]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 04 05 03 <unknown>
+# CHECK-UNKNOWN: 03050407 <unknown>
 
 vle16ff.v v8, (a0), v0.t
 # CHECK-INST: vle16ff.v v8, (a0), v0.t
 # CHECK-ENCODING: [0x07,0x54,0x05,0x01]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 54 05 01 <unknown>
+# CHECK-UNKNOWN: 01055407 <unknown>
 
 vle16ff.v v8, (a0)
 # CHECK-INST: vle16ff.v v8, (a0)
 # CHECK-ENCODING: [0x07,0x54,0x05,0x03]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 54 05 03 <unknown>
+# CHECK-UNKNOWN: 03055407 <unknown>
 
 vle32ff.v v8, (a0), v0.t
 # CHECK-INST: vle32ff.v v8, (a0), v0.t
 # CHECK-ENCODING: [0x07,0x64,0x05,0x01]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 64 05 01 <unknown>
+# CHECK-UNKNOWN: 01056407 <unknown>
 
 vle32ff.v v8, (a0)
 # CHECK-INST: vle32ff.v v8, (a0)
 # CHECK-ENCODING: [0x07,0x64,0x05,0x03]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 64 05 03 <unknown>
+# CHECK-UNKNOWN: 03056407 <unknown>
 
 vle64ff.v v8, (a0), v0.t
 # CHECK-INST: vle64ff.v v8, (a0), v0.t
 # CHECK-ENCODING: [0x07,0x74,0x05,0x01]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 74 05 01 <unknown>
+# CHECK-UNKNOWN: 01057407 <unknown>
 
 vle64ff.v v8, (a0)
 # CHECK-INST: vle64ff.v v8, (a0)
 # CHECK-ENCODING: [0x07,0x74,0x05,0x03]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 74 05 03 <unknown>
+# CHECK-UNKNOWN: 03057407 <unknown>
 
 vlse8.v v8, (a0), a1, v0.t
 # CHECK-INST: vlse8.v v8, (a0), a1, v0.t
 # CHECK-ENCODING: [0x07,0x04,0xb5,0x08]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 04 b5 08 <unknown>
+# CHECK-UNKNOWN: 08b50407 <unknown>
 
 vlse8.v v8, (a0), a1
 # CHECK-INST: vlse8.v v8, (a0), a1
 # CHECK-ENCODING: [0x07,0x04,0xb5,0x0a]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 04 b5 0a <unknown>
+# CHECK-UNKNOWN: 0ab50407 <unknown>
 
 vlse16.v v8, (a0), a1, v0.t
 # CHECK-INST: vlse16.v v8, (a0), a1, v0.t
 # CHECK-ENCODING: [0x07,0x54,0xb5,0x08]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 54 b5 08 <unknown>
+# CHECK-UNKNOWN: 08b55407 <unknown>
 
 vlse16.v v8, (a0), a1
 # CHECK-INST: vlse16.v v8, (a0), a1
 # CHECK-ENCODING: [0x07,0x54,0xb5,0x0a]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 54 b5 0a <unknown>
+# CHECK-UNKNOWN: 0ab55407 <unknown>
 
 vlse32.v v8, (a0), a1, v0.t
 # CHECK-INST: vlse32.v v8, (a0), a1, v0.t
 # CHECK-ENCODING: [0x07,0x64,0xb5,0x08]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 64 b5 08 <unknown>
+# CHECK-UNKNOWN: 08b56407 <unknown>
 
 vlse32.v v8, (a0), a1
 # CHECK-INST: vlse32.v v8, (a0), a1
 # CHECK-ENCODING: [0x07,0x64,0xb5,0x0a]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 64 b5 0a <unknown>
+# CHECK-UNKNOWN: 0ab56407 <unknown>
 
 vlse64.v v8, (a0), a1, v0.t
 # CHECK-INST: vlse64.v v8, (a0), a1, v0.t
 # CHECK-ENCODING: [0x07,0x74,0xb5,0x08]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 74 b5 08 <unknown>
+# CHECK-UNKNOWN: 08b57407 <unknown>
 
 vlse64.v v8, (a0), a1
 # CHECK-INST: vlse64.v v8, (a0), a1
 # CHECK-ENCODING: [0x07,0x74,0xb5,0x0a]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 74 b5 0a <unknown>
+# CHECK-UNKNOWN: 0ab57407 <unknown>
 
 vluxei8.v v8, (a0), v4, v0.t
 # CHECK-INST: vluxei8.v v8, (a0), v4, v0.t
 # CHECK-ENCODING: [0x07,0x04,0x45,0x04]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 04 45 04 <unknown>
+# CHECK-UNKNOWN: 04450407 <unknown>
 
 vluxei8.v v8, (a0), v4
 # CHECK-INST: vluxei8.v v8, (a0), v4
 # CHECK-ENCODING: [0x07,0x04,0x45,0x06]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 04 45 06 <unknown>
+# CHECK-UNKNOWN: 06450407 <unknown>
 
 vluxei16.v v8, (a0), v4, v0.t
 # CHECK-INST: vluxei16.v v8, (a0), v4, v0.t
 # CHECK-ENCODING: [0x07,0x54,0x45,0x04]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 54 45 04 <unknown>
+# CHECK-UNKNOWN: 04455407 <unknown>
 
 vluxei16.v v8, (a0), v4
 # CHECK-INST: vluxei16.v v8, (a0), v4
 # CHECK-ENCODING: [0x07,0x54,0x45,0x06]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 54 45 06 <unknown>
+# CHECK-UNKNOWN: 06455407 <unknown>
 
 vluxei32.v v8, (a0), v4, v0.t
 # CHECK-INST: vluxei32.v v8, (a0), v4, v0.t
 # CHECK-ENCODING: [0x07,0x64,0x45,0x04]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 64 45 04 <unknown>
+# CHECK-UNKNOWN: 04456407 <unknown>
 
 vluxei32.v v8, (a0), v4
 # CHECK-INST: vluxei32.v v8, (a0), v4
 # CHECK-ENCODING: [0x07,0x64,0x45,0x06]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 64 45 06 <unknown>
+# CHECK-UNKNOWN: 06456407 <unknown>
 
 vluxei64.v v8, (a0), v4, v0.t
 # CHECK-INST: vluxei64.v v8, (a0), v4, v0.t
 # CHECK-ENCODING: [0x07,0x74,0x45,0x04]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 74 45 04 <unknown>
+# CHECK-UNKNOWN: 04457407 <unknown>
 
 vluxei64.v v8, (a0), v4
 # CHECK-INST: vluxei64.v v8, (a0), v4
 # CHECK-ENCODING: [0x07,0x74,0x45,0x06]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 74 45 06 <unknown>
+# CHECK-UNKNOWN: 06457407 <unknown>
 
 vloxei8.v v8, (a0), v4, v0.t
 # CHECK-INST: vloxei8.v v8, (a0), v4, v0.t
 # CHECK-ENCODING: [0x07,0x04,0x45,0x0c]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 04 45 0c <unknown>
+# CHECK-UNKNOWN: 0c450407 <unknown>
 
 vloxei8.v v8, (a0), v4
 # CHECK-INST: vloxei8.v v8, (a0), v4
 # CHECK-ENCODING: [0x07,0x04,0x45,0x0e]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 04 45 0e <unknown>
+# CHECK-UNKNOWN: 0e450407 <unknown>
 
 vloxei16.v v8, (a0), v4, v0.t
 # CHECK-INST: vloxei16.v v8, (a0), v4, v0.t
 # CHECK-ENCODING: [0x07,0x54,0x45,0x0c]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 54 45 0c <unknown>
+# CHECK-UNKNOWN: 0c455407 <unknown>
 
 vloxei16.v v8, (a0), v4
 # CHECK-INST: vloxei16.v v8, (a0), v4
 # CHECK-ENCODING: [0x07,0x54,0x45,0x0e]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 54 45 0e <unknown>
+# CHECK-UNKNOWN: 0e455407 <unknown>
 
 vloxei32.v v8, (a0), v4, v0.t
 # CHECK-INST: vloxei32.v v8, (a0), v4, v0.t
 # CHECK-ENCODING: [0x07,0x64,0x45,0x0c]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 64 45 0c <unknown>
+# CHECK-UNKNOWN: 0c456407 <unknown>
 
 vloxei32.v v8, (a0), v4
 # CHECK-INST: vloxei32.v v8, (a0), v4
 # CHECK-ENCODING: [0x07,0x64,0x45,0x0e]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 64 45 0e <unknown>
+# CHECK-UNKNOWN: 0e456407 <unknown>
 
 vloxei64.v v8, (a0), v4, v0.t
 # CHECK-INST: vloxei64.v v8, (a0), v4, v0.t
 # CHECK-ENCODING: [0x07,0x74,0x45,0x0c]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 74 45 0c <unknown>
+# CHECK-UNKNOWN: 0c457407 <unknown>
 
 vloxei64.v v8, (a0), v4
 # CHECK-INST: vloxei64.v v8, (a0), v4
 # CHECK-ENCODING: [0x07,0x74,0x45,0x0e]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 74 45 0e <unknown>
+# CHECK-UNKNOWN: 0e457407 <unknown>
 
 vl1re8.v v8, (a0)
 # CHECK-INST: vl1re8.v v8, (a0)
 # CHECK-ENCODING: [0x07,0x04,0x85,0x02]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 04 85 02 <unknown>
+# CHECK-UNKNOWN: 02850407 <unknown>
 
 vl1re16.v v8, (a0)
 # CHECK-INST: vl1re16.v v8, (a0)
 # CHECK-ENCODING: [0x07,0x54,0x85,0x02]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 54 85 02 <unknown>
+# CHECK-UNKNOWN: 02855407 <unknown>
 
 vl1re32.v v8, (a0)
 # CHECK-INST: vl1re32.v v8, (a0)
 # CHECK-ENCODING: [0x07,0x64,0x85,0x02]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 64 85 02 <unknown>
+# CHECK-UNKNOWN: 02856407 <unknown>
 
 vl1re64.v v8, (a0)
 # CHECK-INST: vl1re64.v v8, (a0)
 # CHECK-ENCODING: [0x07,0x74,0x85,0x02]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 74 85 02 <unknown>
+# CHECK-UNKNOWN: 02857407 <unknown>
 
 vl2re8.v v8, (a0)
 # CHECK-INST: vl2re8.v v8, (a0)
 # CHECK-ENCODING: [0x07,0x04,0x85,0x22]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 04 85 22 <unknown>
+# CHECK-UNKNOWN: 22850407 <unknown>
 
 vl2re16.v v8, (a0)
 # CHECK-INST: vl2re16.v v8, (a0)
 # CHECK-ENCODING: [0x07,0x54,0x85,0x22]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 54 85 22 <unknown>
+# CHECK-UNKNOWN: 22855407 <unknown>
 
 vl2re32.v v8, (a0)
 # CHECK-INST: vl2re32.v v8, (a0)
 # CHECK-ENCODING: [0x07,0x64,0x85,0x22]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 64 85 22 <unknown>
+# CHECK-UNKNOWN: 22856407 <unknown>
 
 vl2re64.v v8, (a0)
 # CHECK-INST: vl2re64.v v8, (a0)
 # CHECK-ENCODING: [0x07,0x74,0x85,0x22]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 74 85 22 <unknown>
+# CHECK-UNKNOWN: 22857407 <unknown>
 
 vl4re8.v v8, (a0)
 # CHECK-INST: vl4re8.v v8, (a0)
 # CHECK-ENCODING: [0x07,0x04,0x85,0x62]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 04 85 62 <unknown>
+# CHECK-UNKNOWN: 62850407 <unknown>
 
 vl4re16.v v8, (a0)
 # CHECK-INST: vl4re16.v v8, (a0)
 # CHECK-ENCODING: [0x07,0x54,0x85,0x62]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 54 85 62 <unknown>
+# CHECK-UNKNOWN: 62855407 <unknown>
 
 vl4re32.v v8, (a0)
 # CHECK-INST: vl4re32.v v8, (a0)
 # CHECK-ENCODING: [0x07,0x64,0x85,0x62]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 64 85 62 <unknown>
+# CHECK-UNKNOWN: 62856407 <unknown>
 
 vl4re64.v v8, (a0)
 # CHECK-INST: vl4re64.v v8, (a0)
 # CHECK-ENCODING: [0x07,0x74,0x85,0x62]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 74 85 62 <unknown>
+# CHECK-UNKNOWN: 62857407 <unknown>
 
 vl8re8.v v8, (a0)
 # CHECK-INST: vl8re8.v v8, (a0)
 # CHECK-ENCODING: [0x07,0x04,0x85,0xe2]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 04 85 e2 <unknown>
+# CHECK-UNKNOWN: e2850407 <unknown>
 
 vl8re16.v v8, (a0)
 # CHECK-INST: vl8re16.v v8, (a0)
 # CHECK-ENCODING: [0x07,0x54,0x85,0xe2]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 54 85 e2 <unknown>
+# CHECK-UNKNOWN: e2855407 <unknown>
 
 vl8re32.v v8, (a0)
 # CHECK-INST: vl8re32.v v8, (a0)
 # CHECK-ENCODING: [0x07,0x64,0x85,0xe2]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 64 85 e2 <unknown>
+# CHECK-UNKNOWN: e2856407 <unknown>
 
 vl8re64.v v8, (a0)
 # CHECK-INST: vl8re64.v v8, (a0)
 # CHECK-ENCODING: [0x07,0x74,0x85,0xe2]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 74 85 e2 <unknown>
+# CHECK-UNKNOWN: e2857407 <unknown>
 
 vlm.v v0, 0(a0)
 # CHECK-INST: vlm.v v0, (a0)
 # CHECK-ENCODING: [0x07,0x00,0xb5,0x02]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 00 b5 02 <unknown>
+# CHECK-UNKNOWN: 02b50007 <unknown>
 
 vle8.v v8, 0(a0)
 # CHECK-INST: vle8.v v8, (a0)
 # CHECK-ENCODING: [0x07,0x04,0x05,0x02]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 04 05 02 <unknown>
+# CHECK-UNKNOWN: 02050407 <unknown>
 
 vle8ff.v v8, 0(a0), v0.t
 # CHECK-INST: vle8ff.v v8, (a0), v0.t
 # CHECK-ENCODING: [0x07,0x04,0x05,0x01]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 04 05 01 <unknown>
+# CHECK-UNKNOWN: 01050407 <unknown>
 
 vlse16.v v8, 0(a0), a1, v0.t
 # CHECK-INST: vlse16.v v8, (a0), a1, v0.t
 # CHECK-ENCODING: [0x07,0x54,0xb5,0x08]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 54 b5 08 <unknown>
+# CHECK-UNKNOWN: 08b55407 <unknown>
 
 vluxei32.v v8, 0(a0), v4
 # CHECK-INST: vluxei32.v v8, (a0), v4
 # CHECK-ENCODING: [0x07,0x64,0x45,0x06]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 64 45 06 <unknown>
+# CHECK-UNKNOWN: 06456407 <unknown>
 
 vloxei64.v v8, 0(a0), v4
 # CHECK-INST: vloxei64.v v8, (a0), v4
 # CHECK-ENCODING: [0x07,0x74,0x45,0x0e]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 74 45 0e <unknown>
+# CHECK-UNKNOWN: 0e457407 <unknown>
diff --git a/llvm/test/MC/RISCV/rvv/macc.s b/llvm/test/MC/RISCV/rvv/macc.s
index 0662620b455a..ffdc2d646929 100644
--- a/llvm/test/MC/RISCV/rvv/macc.s
+++ b/llvm/test/MC/RISCV/rvv/macc.s
@@ -12,178 +12,178 @@ vmacc.vv v8, v20, v4, v0.t
 # CHECK-INST: vmacc.vv v8, v20, v4, v0.t
 # CHECK-ENCODING: [0x57,0x24,0x4a,0xb4]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 24 4a b4 <unknown>
+# CHECK-UNKNOWN: b44a2457 <unknown>
 
 vmacc.vv v8, v20, v4
 # CHECK-INST: vmacc.vv v8, v20, v4
 # CHECK-ENCODING: [0x57,0x24,0x4a,0xb6]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 24 4a b6 <unknown>
+# CHECK-UNKNOWN: b64a2457 <unknown>
 
 vmacc.vx v8, a0, v4, v0.t
 # CHECK-INST: vmacc.vx v8, a0, v4, v0.t
 # CHECK-ENCODING: [0x57,0x64,0x45,0xb4]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 64 45 b4 <unknown>
+# CHECK-UNKNOWN: b4456457 <unknown>
 
 vmacc.vx v8, a0, v4
 # CHECK-INST: vmacc.vx v8, a0, v4
 # CHECK-ENCODING: [0x57,0x64,0x45,0xb6]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 64 45 b6 <unknown>
+# CHECK-UNKNOWN: b6456457 <unknown>
 
 vnmsac.vv v8, v20, v4, v0.t
 # CHECK-INST: vnmsac.vv v8, v20, v4, v0.t
 # CHECK-ENCODING: [0x57,0x24,0x4a,0xbc]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 24 4a bc <unknown>
+# CHECK-UNKNOWN: bc4a2457 <unknown>
 
 vnmsac.vv v8, v20, v4
 # CHECK-INST: vnmsac.vv v8, v20, v4
 # CHECK-ENCODING: [0x57,0x24,0x4a,0xbe]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 24 4a be <unknown>
+# CHECK-UNKNOWN: be4a2457 <unknown>
 
 vnmsac.vx v8, a0, v4, v0.t
 # CHECK-INST: vnmsac.vx v8, a0, v4, v0.t
 # CHECK-ENCODING: [0x57,0x64,0x45,0xbc]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 64 45 bc <unknown>
+# CHECK-UNKNOWN: bc456457 <unknown>
 
 vnmsac.vx v8, a0, v4
 # CHECK-INST: vnmsac.vx v8, a0, v4
 # CHECK-ENCODING: [0x57,0x64,0x45,0xbe]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 64 45 be <unknown>
+# CHECK-UNKNOWN: be456457 <unknown>
 
 vmadd.vv v8, v20, v4, v0.t
 # CHECK-INST: vmadd.vv v8, v20, v4, v0.t
 # CHECK-ENCODING: [0x57,0x24,0x4a,0xa4]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 24 4a a4 <unknown>
+# CHECK-UNKNOWN: a44a2457 <unknown>
 
 vmadd.vv v8, v20, v4
 # CHECK-INST: vmadd.vv v8, v20, v4
 # CHECK-ENCODING: [0x57,0x24,0x4a,0xa6]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 24 4a a6 <unknown>
+# CHECK-UNKNOWN: a64a2457 <unknown>
 
 vmadd.vx v8, a0, v4, v0.t
 # CHECK-INST: vmadd.vx v8, a0, v4, v0.t
 # CHECK-ENCODING: [0x57,0x64,0x45,0xa4]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 64 45 a4 <unknown>
+# CHECK-UNKNOWN: a4456457 <unknown>
 
 vmadd.vx v8, a0, v4
 # CHECK-INST: vmadd.vx v8, a0, v4
 # CHECK-ENCODING: [0x57,0x64,0x45,0xa6]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 64 45 a6 <unknown>
+# CHECK-UNKNOWN: a6456457 <unknown>
 
 vnmsub.vv v8, v20, v4, v0.t
 # CHECK-INST: vnmsub.vv v8, v20, v4, v0.t
 # CHECK-ENCODING: [0x57,0x24,0x4a,0xac]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 24 4a ac <unknown>
+# CHECK-UNKNOWN: ac4a2457 <unknown>
 
 vnmsub.vv v8, v20, v4
 # CHECK-INST: vnmsub.vv v8, v20, v4
 # CHECK-ENCODING: [0x57,0x24,0x4a,0xae]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 24 4a ae <unknown>
+# CHECK-UNKNOWN: ae4a2457 <unknown>
 
 vnmsub.vx v8, a0, v4, v0.t
 # CHECK-INST: vnmsub.vx v8, a0, v4, v0.t
 # CHECK-ENCODING: [0x57,0x64,0x45,0xac]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 64 45 ac <unknown>
+# CHECK-UNKNOWN: ac456457 <unknown>
 
 vnmsub.vx v8, a0, v4
 # CHECK-INST: vnmsub.vx v8, a0, v4
 # CHECK-ENCODING: [0x57,0x64,0x45,0xae]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 64 45 ae <unknown>
+# CHECK-UNKNOWN: ae456457 <unknown>
 
 vwmaccu.vv v8, v20, v4, v0.t
 # CHECK-INST: vwmaccu.vv v8, v20, v4, v0.t
 # CHECK-ENCODING: [0x57,0x24,0x4a,0xf0]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 24 4a f0 <unknown>
+# CHECK-UNKNOWN: f04a2457 <unknown>
 
 vwmaccu.vv v8, v20, v4
 # CHECK-INST: vwmaccu.vv v8, v20, v4
 # CHECK-ENCODING: [0x57,0x24,0x4a,0xf2]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 24 4a f2 <unknown>
+# CHECK-UNKNOWN: f24a2457 <unknown>
 
 vwmaccu.vx v8, a0, v4, v0.t
 # CHECK-INST: vwmaccu.vx v8, a0, v4, v0.t
 # CHECK-ENCODING: [0x57,0x64,0x45,0xf0]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 64 45 f0 <unknown>
+# CHECK-UNKNOWN: f0456457 <unknown>
 
 vwmaccu.vx v8, a0, v4
 # CHECK-INST: vwmaccu.vx v8, a0, v4
 # CHECK-ENCODING: [0x57,0x64,0x45,0xf2]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 64 45 f2 <unknown>
+# CHECK-UNKNOWN: f2456457 <unknown>
 
 vwmacc.vv v8, v20, v4, v0.t
 # CHECK-INST: vwmacc.vv v8, v20, v4, v0.t
 # CHECK-ENCODING: [0x57,0x24,0x4a,0xf4]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 24 4a f4 <unknown>
+# CHECK-UNKNOWN: f44a2457 <unknown>
 
 vwmacc.vv v8, v20, v4
 # CHECK-INST: vwmacc.vv v8, v20, v4
 # CHECK-ENCODING: [0x57,0x24,0x4a,0xf6]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 24 4a f6 <unknown>
+# CHECK-UNKNOWN: f64a2457 <unknown>
 
 vwmacc.vx v8, a0, v4, v0.t
 # CHECK-INST: vwmacc.vx v8, a0, v4, v0.t
 # CHECK-ENCODING: [0x57,0x64,0x45,0xf4]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 64 45 f4 <unknown>
+# CHECK-UNKNOWN: f4456457 <unknown>
 
 vwmacc.vx v8, a0, v4
 # CHECK-INST: vwmacc.vx v8, a0, v4
 # CHECK-ENCODING: [0x57,0x64,0x45,0xf6]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 64 45 f6 <unknown>
+# CHECK-UNKNOWN: f6456457 <unknown>
 
 vwmaccsu.vv v8, v20, v4, v0.t
 # CHECK-INST: vwmaccsu.vv v8, v20, v4, v0.t
 # CHECK-ENCODING: [0x57,0x24,0x4a,0xfc]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 24 4a fc <unknown>
+# CHECK-UNKNOWN: fc4a2457 <unknown>
 
 vwmaccsu.vv v8, v20, v4
 # CHECK-INST: vwmaccsu.vv v8, v20, v4
 # CHECK-ENCODING: [0x57,0x24,0x4a,0xfe]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 24 4a fe <unknown>
+# CHECK-UNKNOWN: fe4a2457 <unknown>
 
 vwmaccsu.vx v8, a0, v4, v0.t
 # CHECK-INST: vwmaccsu.vx v8, a0, v4, v0.t
 # CHECK-ENCODING: [0x57,0x64,0x45,0xfc]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 64 45 fc <unknown>
+# CHECK-UNKNOWN: fc456457 <unknown>
 
 vwmaccsu.vx v8, a0, v4
 # CHECK-INST: vwmaccsu.vx v8, a0, v4
 # CHECK-ENCODING: [0x57,0x64,0x45,0xfe]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 64 45 fe <unknown>
+# CHECK-UNKNOWN: fe456457 <unknown>
 
 vwmaccus.vx v8, a0, v4, v0.t
 # CHECK-INST: vwmaccus.vx v8, a0, v4, v0.t
 # CHECK-ENCODING: [0x57,0x64,0x45,0xf8]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 64 45 f8 <unknown>
+# CHECK-UNKNOWN: f8456457 <unknown>
 
 vwmaccus.vx v8, a0, v4
 # CHECK-INST: vwmaccus.vx v8, a0, v4
 # CHECK-ENCODING: [0x57,0x64,0x45,0xfa]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 64 45 fa <unknown>
+# CHECK-UNKNOWN: fa456457 <unknown>
diff --git a/llvm/test/MC/RISCV/rvv/mask.s b/llvm/test/MC/RISCV/rvv/mask.s
index c0691b97f1d9..c0dd44b6fc67 100644
--- a/llvm/test/MC/RISCV/rvv/mask.s
+++ b/llvm/test/MC/RISCV/rvv/mask.s
@@ -12,154 +12,154 @@ vmand.mm v8, v4, v20
 # CHECK-INST: vmand.mm v8, v4, v20
 # CHECK-ENCODING: [0x57,0x24,0x4a,0x66]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 24 4a 66 <unknown>
+# CHECK-UNKNOWN: 664a2457 <unknown>
 
 vmnand.mm v8, v4, v20
 # CHECK-INST: vmnand.mm v8, v4, v20
 # CHECK-ENCODING: [0x57,0x24,0x4a,0x76]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 24 4a 76 <unknown>
+# CHECK-UNKNOWN: 764a2457 <unknown>
 
 vmandn.mm v8, v4, v20
 # CHECK-INST: vmandn.mm v8, v4, v20
 # CHECK-ENCODING: [0x57,0x24,0x4a,0x62]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 24 4a 62 <unknown>
+# CHECK-UNKNOWN: 624a2457 <unknown>
 
 vmxor.mm v8, v4, v20
 # CHECK-INST: vmxor.mm v8, v4, v20
 # CHECK-ENCODING: [0x57,0x24,0x4a,0x6e]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 24 4a 6e <unknown>
+# CHECK-UNKNOWN: 6e4a2457 <unknown>
 
 vmor.mm v8, v4, v20
 # CHECK-INST: vmor.mm v8, v4, v20
 # CHECK-ENCODING: [0x57,0x24,0x4a,0x6a]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 24 4a 6a <unknown>
+# CHECK-UNKNOWN: 6a4a2457 <unknown>
 
 vmnor.mm v8, v4, v20
 # CHECK-INST: vmnor.mm v8, v4, v20
 # CHECK-ENCODING: [0x57,0x24,0x4a,0x7a]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 24 4a 7a <unknown>
+# CHECK-UNKNOWN: 7a4a2457 <unknown>
 
 vmorn.mm v8, v4, v20
 # CHECK-INST: vmorn.mm v8, v4, v20
 # CHECK-ENCODING: [0x57,0x24,0x4a,0x72]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 24 4a 72 <unknown>
+# CHECK-UNKNOWN: 724a2457 <unknown>
 
 vmxnor.mm v8, v4, v20
 # CHECK-INST: vmxnor.mm v8, v4, v20
 # CHECK-ENCODING: [0x57,0x24,0x4a,0x7e]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 24 4a 7e <unknown>
+# CHECK-UNKNOWN: 7e4a2457 <unknown>
 
 vcpop.m a2, v4, v0.t
 # CHECK-INST: vcpop.m a2, v4, v0.t
 # CHECK-ENCODING: [0x57,0x26,0x48,0x40]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 26 48 40 <unknown>
+# CHECK-UNKNOWN: 40482657 <unknown>
 
 vcpop.m a2, v4
 # CHECK-INST: vcpop.m a2, v4
 # CHECK-ENCODING: [0x57,0x26,0x48,0x42]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 26 48 42 <unknown>
+# CHECK-UNKNOWN: 42482657 <unknown>
 
 vfirst.m a2, v4, v0.t
 # CHECK-INST: vfirst.m a2, v4, v0.t
 # CHECK-ENCODING: [0x57,0xa6,0x48,0x40]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 a6 48 40 <unknown>
+# CHECK-UNKNOWN: 4048a657 <unknown>
 
 vfirst.m a2, v4
 # CHECK-INST: vfirst.m a2, v4
 # CHECK-ENCODING: [0x57,0xa6,0x48,0x42]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 a6 48 42 <unknown>
+# CHECK-UNKNOWN: 4248a657 <unknown>
 
 vmsbf.m v8, v4, v0.t
 # CHECK-INST: vmsbf.m v8, v4, v0.t
 # CHECK-ENCODING: [0x57,0xa4,0x40,0x50]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 a4 40 50 <unknown>
+# CHECK-UNKNOWN: 5040a457 <unknown>
 
 vmsbf.m v8, v4
 # CHECK-INST: vmsbf.m v8, v4
 # CHECK-ENCODING: [0x57,0xa4,0x40,0x52]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 a4 40 52 <unknown>
+# CHECK-UNKNOWN: 5240a457 <unknown>
 
 vmsif.m v8, v4, v0.t
 # CHECK-INST: vmsif.m v8, v4, v0.t
 # CHECK-ENCODING: [0x57,0xa4,0x41,0x50]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 a4 41 50 <unknown>
+# CHECK-UNKNOWN: 5041a457 <unknown>
 
 vmsif.m v8, v4
 # CHECK-INST: vmsif.m v8, v4
 # CHECK-ENCODING: [0x57,0xa4,0x41,0x52]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 a4 41 52 <unknown>
+# CHECK-UNKNOWN: 5241a457 <unknown>
 
 vmsof.m v8, v4, v0.t
 # CHECK-INST: vmsof.m v8, v4, v0.t
 # CHECK-ENCODING: [0x57,0x24,0x41,0x50]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 24 41 50 <unknown>
+# CHECK-UNKNOWN: 50412457 <unknown>
 
 vmsof.m v8, v4
 # CHECK-INST: vmsof.m v8, v4
 # CHECK-ENCODING: [0x57,0x24,0x41,0x52]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 24 41 52 <unknown>
+# CHECK-UNKNOWN: 52412457 <unknown>
 
 viota.m v8, v4, v0.t
 # CHECK-INST: viota.m v8, v4, v0.t
 # CHECK-ENCODING: [0x57,0x24,0x48,0x50]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 24 48 50 <unknown>
+# CHECK-UNKNOWN: 50482457 <unknown>
 
 viota.m v8, v4
 # CHECK-INST: viota.m v8, v4
 # CHECK-ENCODING: [0x57,0x24,0x48,0x52]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 24 48 52 <unknown>
+# CHECK-UNKNOWN: 52482457 <unknown>
 
 vid.v v8, v0.t
 # CHECK-INST: vid.v v8, v0.t
 # CHECK-ENCODING: [0x57,0xa4,0x08,0x50]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 a4 08 50 <unknown>
+# CHECK-UNKNOWN: 5008a457 <unknown>
 
 vid.v v8
 # CHECK-INST: vid.v v8
 # CHECK-ENCODING: [0x57,0xa4,0x08,0x52]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 a4 08 52 <unknown>
+# CHECK-UNKNOWN: 5208a457 <unknown>
 
 vmmv.m v8, v4
 # CHECK-INST: vmmv.m v8, v4
 # CHECK-ENCODING: [0x57,0x24,0x42,0x66]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 24 42 66 <unknown>
+# CHECK-UNKNOWN: 66422457 <unknown>
 
 vmclr.m v8
 # CHECK-INST: vmclr.m v8
 # CHECK-ENCODING: [0x57,0x24,0x84,0x6e]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 24 84 6e <unknown>
+# CHECK-UNKNOWN: 6e842457 <unknown>
 
 vmset.m v8
 # CHECK-INST: vmset.m v8
 # CHECK-ENCODING: [0x57,0x24,0x84,0x7e]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 24 84 7e <unknown>
+# CHECK-UNKNOWN: 7e842457 <unknown>
 
 vmnot.m v8, v4
 # CHECK-INST: vmnot.m v8, v4
 # CHECK-ENCODING: [0x57,0x24,0x42,0x76]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 24 42 76 <unknown>
+# CHECK-UNKNOWN: 76422457 <unknown>
diff --git a/llvm/test/MC/RISCV/rvv/minmax.s b/llvm/test/MC/RISCV/rvv/minmax.s
index 4eaf897b84c4..70fe040f07f9 100644
--- a/llvm/test/MC/RISCV/rvv/minmax.s
+++ b/llvm/test/MC/RISCV/rvv/minmax.s
@@ -12,94 +12,94 @@ vminu.vv v8, v4, v20, v0.t
 # CHECK-INST: vminu.vv v8, v4, v20, v0.t
 # CHECK-ENCODING: [0x57,0x04,0x4a,0x10]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 04 4a 10 <unknown>
+# CHECK-UNKNOWN: 104a0457 <unknown>
 
 vminu.vv v8, v4, v20
 # CHECK-INST: vminu.vv v8, v4, v20
 # CHECK-ENCODING: [0x57,0x04,0x4a,0x12]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 04 4a 12 <unknown>
+# CHECK-UNKNOWN: 124a0457 <unknown>
 
 vminu.vx v8, v4, a0, v0.t
 # CHECK-INST: vminu.vx v8, v4, a0, v0.t
 # CHECK-ENCODING: [0x57,0x44,0x45,0x10]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 44 45 10 <unknown>
+# CHECK-UNKNOWN: 10454457 <unknown>
 
 vminu.vx v8, v4, a0
 # CHECK-INST: vminu.vx v8, v4, a0
 # CHECK-ENCODING: [0x57,0x44,0x45,0x12]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 44 45 12 <unknown>
+# CHECK-UNKNOWN: 12454457 <unknown>
 
 vmin.vv v8, v4, v20, v0.t
 # CHECK-INST: vmin.vv v8, v4, v20, v0.t
 # CHECK-ENCODING: [0x57,0x04,0x4a,0x14]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 04 4a 14 <unknown>
+# CHECK-UNKNOWN: 144a0457 <unknown>
 
 vmin.vv v8, v4, v20
 # CHECK-INST: vmin.vv v8, v4, v20
 # CHECK-ENCODING: [0x57,0x04,0x4a,0x16]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 04 4a 16 <unknown>
+# CHECK-UNKNOWN: 164a0457 <unknown>
 
 vmin.vx v8, v4, a0, v0.t
 # CHECK-INST: vmin.vx v8, v4, a0, v0.t
 # CHECK-ENCODING: [0x57,0x44,0x45,0x14]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 44 45 14 <unknown>
+# CHECK-UNKNOWN: 14454457 <unknown>
 
 vmin.vx v8, v4, a0
 # CHECK-INST: vmin.vx v8, v4, a0
 # CHECK-ENCODING: [0x57,0x44,0x45,0x16]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 44 45 16 <unknown>
+# CHECK-UNKNOWN: 16454457 <unknown>
 
 vmaxu.vv v8, v4, v20, v0.t
 # CHECK-INST: vmaxu.vv v8, v4, v20, v0.t
 # CHECK-ENCODING: [0x57,0x04,0x4a,0x18]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 04 4a 18 <unknown>
+# CHECK-UNKNOWN: 184a0457 <unknown>
 
 vmaxu.vv v8, v4, v20
 # CHECK-INST: vmaxu.vv v8, v4, v20
 # CHECK-ENCODING: [0x57,0x04,0x4a,0x1a]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 04 4a 1a <unknown>
+# CHECK-UNKNOWN: 1a4a0457 <unknown>
 
 vmaxu.vx v8, v4, a0, v0.t
 # CHECK-INST: vmaxu.vx v8, v4, a0, v0.t
 # CHECK-ENCODING: [0x57,0x44,0x45,0x18]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 44 45 18 <unknown>
+# CHECK-UNKNOWN: 18454457 <unknown>
 
 vmaxu.vx v8, v4, a0
 # CHECK-INST: vmaxu.vx v8, v4, a0
 # CHECK-ENCODING: [0x57,0x44,0x45,0x1a]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 44 45 1a <unknown>
+# CHECK-UNKNOWN: 1a454457 <unknown>
 
 vmax.vv v8, v4, v20, v0.t
 # CHECK-INST: vmax.vv v8, v4, v20, v0.t
 # CHECK-ENCODING: [0x57,0x04,0x4a,0x1c]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 04 4a 1c <unknown>
+# CHECK-UNKNOWN: 1c4a0457 <unknown>
 
 vmax.vv v8, v4, v20
 # CHECK-INST: vmax.vv v8, v4, v20
 # CHECK-ENCODING: [0x57,0x04,0x4a,0x1e]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 04 4a 1e <unknown>
+# CHECK-UNKNOWN: 1e4a0457 <unknown>
 
 vmax.vx v8, v4, a0, v0.t
 # CHECK-INST: vmax.vx v8, v4, a0, v0.t
 # CHECK-ENCODING: [0x57,0x44,0x45,0x1c]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 44 45 1c <unknown>
+# CHECK-UNKNOWN: 1c454457 <unknown>
 
 vmax.vx v8, v4, a0
 # CHECK-INST: vmax.vx v8, v4, a0
 # CHECK-ENCODING: [0x57,0x44,0x45,0x1e]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 44 45 1e <unknown>
+# CHECK-UNKNOWN: 1e454457 <unknown>
diff --git a/llvm/test/MC/RISCV/rvv/mul.s b/llvm/test/MC/RISCV/rvv/mul.s
index 9f7d6182d80f..2782ea683f9c 100644
--- a/llvm/test/MC/RISCV/rvv/mul.s
+++ b/llvm/test/MC/RISCV/rvv/mul.s
@@ -12,190 +12,190 @@ vmul.vv v8, v4, v20, v0.t
 # CHECK-INST: vmul.vv v8, v4, v20, v0.t
 # CHECK-ENCODING: [0x57,0x24,0x4a,0x94]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 24 4a 94 <unknown>
+# CHECK-UNKNOWN: 944a2457 <unknown>
 
 vmul.vv v8, v4, v20
 # CHECK-INST: vmul.vv v8, v4, v20
 # CHECK-ENCODING: [0x57,0x24,0x4a,0x96]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 24 4a 96 <unknown>
+# CHECK-UNKNOWN: 964a2457 <unknown>
 
 vmul.vx v8, v4, a0, v0.t
 # CHECK-INST: vmul.vx v8, v4, a0, v0.t
 # CHECK-ENCODING: [0x57,0x64,0x45,0x94]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 64 45 94 <unknown>
+# CHECK-UNKNOWN: 94456457 <unknown>
 
 vmul.vx v8, v4, a0
 # CHECK-INST: vmul.vx v8, v4, a0
 # CHECK-ENCODING: [0x57,0x64,0x45,0x96]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 64 45 96 <unknown>
+# CHECK-UNKNOWN: 96456457 <unknown>
 
 vmulh.vv v8, v4, v20, v0.t
 # CHECK-INST: vmulh.vv v8, v4, v20, v0.t
 # CHECK-ENCODING: [0x57,0x24,0x4a,0x9c]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 24 4a 9c <unknown>
+# CHECK-UNKNOWN: 9c4a2457 <unknown>
 
 vmulh.vv v8, v4, v20
 # CHECK-INST: vmulh.vv v8, v4, v20
 # CHECK-ENCODING: [0x57,0x24,0x4a,0x9e]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 24 4a 9e <unknown>
+# CHECK-UNKNOWN: 9e4a2457 <unknown>
 
 vmulh.vx v8, v4, a0, v0.t
 # CHECK-INST: vmulh.vx v8, v4, a0, v0.t
 # CHECK-ENCODING: [0x57,0x64,0x45,0x9c]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 64 45 9c <unknown>
+# CHECK-UNKNOWN: 9c456457 <unknown>
 
 vmulh.vx v8, v4, a0
 # CHECK-INST: vmulh.vx v8, v4, a0
 # CHECK-ENCODING: [0x57,0x64,0x45,0x9e]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 64 45 9e <unknown>
+# CHECK-UNKNOWN: 9e456457 <unknown>
 
 vmulhu.vv v8, v4, v20, v0.t
 # CHECK-INST: vmulhu.vv v8, v4, v20, v0.t
 # CHECK-ENCODING: [0x57,0x24,0x4a,0x90]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 24 4a 90 <unknown>
+# CHECK-UNKNOWN: 904a2457 <unknown>
 
 vmulhu.vv v8, v4, v20
 # CHECK-INST: vmulhu.vv v8, v4, v20
 # CHECK-ENCODING: [0x57,0x24,0x4a,0x92]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 24 4a 92 <unknown>
+# CHECK-UNKNOWN: 924a2457 <unknown>
 
 vmulhu.vx v8, v4, a0, v0.t
 # CHECK-INST: vmulhu.vx v8, v4, a0, v0.t
 # CHECK-ENCODING: [0x57,0x64,0x45,0x90]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 64 45 90 <unknown>
+# CHECK-UNKNOWN: 90456457 <unknown>
 
 vmulhu.vx v8, v4, a0
 # CHECK-INST: vmulhu.vx v8, v4, a0
 # CHECK-ENCODING: [0x57,0x64,0x45,0x92]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 64 45 92 <unknown>
+# CHECK-UNKNOWN: 92456457 <unknown>
 
 vmulhsu.vv v8, v4, v20, v0.t
 # CHECK-INST: vmulhsu.vv v8, v4, v20, v0.t
 # CHECK-ENCODING: [0x57,0x24,0x4a,0x98]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 24 4a 98 <unknown>
+# CHECK-UNKNOWN: 984a2457 <unknown>
 
 vmulhsu.vv v8, v4, v20
 # CHECK-INST: vmulhsu.vv v8, v4, v20
 # CHECK-ENCODING: [0x57,0x24,0x4a,0x9a]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 24 4a 9a <unknown>
+# CHECK-UNKNOWN: 9a4a2457 <unknown>
 
 vmulhsu.vx v8, v4, a0, v0.t
 # CHECK-INST: vmulhsu.vx v8, v4, a0, v0.t
 # CHECK-ENCODING: [0x57,0x64,0x45,0x98]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 64 45 98 <unknown>
+# CHECK-UNKNOWN: 98456457 <unknown>
 
 vmulhsu.vx v8, v4, a0
 # CHECK-INST: vmulhsu.vx v8, v4, a0
 # CHECK-ENCODING: [0x57,0x64,0x45,0x9a]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 64 45 9a <unknown>
+# CHECK-UNKNOWN: 9a456457 <unknown>
 
 vwmul.vv v8, v4, v20, v0.t
 # CHECK-INST: vwmul.vv v8, v4, v20, v0.t
 # CHECK-ENCODING: [0x57,0x24,0x4a,0xec]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 24 4a ec <unknown>
+# CHECK-UNKNOWN: ec4a2457 <unknown>
 
 vwmul.vv v8, v4, v20
 # CHECK-INST: vwmul.vv v8, v4, v20
 # CHECK-ENCODING: [0x57,0x24,0x4a,0xee]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 24 4a ee <unknown>
+# CHECK-UNKNOWN: ee4a2457 <unknown>
 
 vwmul.vx v8, v4, a0, v0.t
 # CHECK-INST: vwmul.vx v8, v4, a0, v0.t
 # CHECK-ENCODING: [0x57,0x64,0x45,0xec]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 64 45 ec <unknown>
+# CHECK-UNKNOWN: ec456457 <unknown>
 
 vwmul.vx v8, v4, a0
 # CHECK-INST: vwmul.vx v8, v4, a0
 # CHECK-ENCODING: [0x57,0x64,0x45,0xee]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 64 45 ee <unknown>
+# CHECK-UNKNOWN: ee456457 <unknown>
 
 vwmulu.vv v8, v4, v20, v0.t
 # CHECK-INST: vwmulu.vv v8, v4, v20, v0.t
 # CHECK-ENCODING: [0x57,0x24,0x4a,0xe0]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 24 4a e0 <unknown>
+# CHECK-UNKNOWN: e04a2457 <unknown>
 
 vwmulu.vv v8, v4, v20
 # CHECK-INST: vwmulu.vv v8, v4, v20
 # CHECK-ENCODING: [0x57,0x24,0x4a,0xe2]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 24 4a e2 <unknown>
+# CHECK-UNKNOWN: e24a2457 <unknown>
 
 vwmulu.vx v8, v4, a0, v0.t
 # CHECK-INST: vwmulu.vx v8, v4, a0, v0.t
 # CHECK-ENCODING: [0x57,0x64,0x45,0xe0]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 64 45 e0 <unknown>
+# CHECK-UNKNOWN: e0456457 <unknown>
 
 vwmulu.vx v8, v4, a0
 # CHECK-INST: vwmulu.vx v8, v4, a0
 # CHECK-ENCODING: [0x57,0x64,0x45,0xe2]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 64 45 e2 <unknown>
+# CHECK-UNKNOWN: e2456457 <unknown>
 
 vwmulsu.vv v8, v4, v20, v0.t
 # CHECK-INST: vwmulsu.vv v8, v4, v20, v0.t
 # CHECK-ENCODING: [0x57,0x24,0x4a,0xe8]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 24 4a e8 <unknown>
+# CHECK-UNKNOWN: e84a2457 <unknown>
 
 vwmulsu.vv v8, v4, v20
 # CHECK-INST: vwmulsu.vv v8, v4, v20
 # CHECK-ENCODING: [0x57,0x24,0x4a,0xea]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 24 4a ea <unknown>
+# CHECK-UNKNOWN: ea4a2457 <unknown>
 
 vwmulsu.vx v8, v4, a0, v0.t
 # CHECK-INST: vwmulsu.vx v8, v4, a0, v0.t
 # CHECK-ENCODING: [0x57,0x64,0x45,0xe8]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 64 45 e8 <unknown>
+# CHECK-UNKNOWN: e8456457 <unknown>
 
 vwmulsu.vx v8, v4, a0
 # CHECK-INST: vwmulsu.vx v8, v4, a0
 # CHECK-ENCODING: [0x57,0x64,0x45,0xea]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 64 45 ea <unknown>
+# CHECK-UNKNOWN: ea456457 <unknown>
 
 vsmul.vv v8, v4, v20, v0.t
 # CHECK-INST: vsmul.vv v8, v4, v20, v0.t
 # CHECK-ENCODING: [0x57,0x04,0x4a,0x9c]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 04 4a 9c <unknown>
+# CHECK-UNKNOWN: 9c4a0457 <unknown>
 
 vsmul.vv v8, v4, v20
 # CHECK-INST: vsmul.vv v8, v4, v20
 # CHECK-ENCODING: [0x57,0x04,0x4a,0x9e]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 04 4a 9e <unknown>
+# CHECK-UNKNOWN: 9e4a0457 <unknown>
 
 vsmul.vx v8, v4, a0, v0.t
 # CHECK-INST: vsmul.vx v8, v4, a0, v0.t
 # CHECK-ENCODING: [0x57,0x44,0x45,0x9c]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 44 45 9c <unknown>
+# CHECK-UNKNOWN: 9c454457 <unknown>
 
 vsmul.vx v8, v4, a0
 # CHECK-INST: vsmul.vx v8, v4, a0
 # CHECK-ENCODING: [0x57,0x44,0x45,0x9e]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 44 45 9e <unknown>
+# CHECK-UNKNOWN: 9e454457 <unknown>
diff --git a/llvm/test/MC/RISCV/rvv/mv.s b/llvm/test/MC/RISCV/rvv/mv.s
index cd02aa94d18e..f96e14932ead 100644
--- a/llvm/test/MC/RISCV/rvv/mv.s
+++ b/llvm/test/MC/RISCV/rvv/mv.s
@@ -12,52 +12,52 @@ vmv.v.v v8, v20
 # CHECK-INST: vmv.v.v v8, v20
 # CHECK-ENCODING: [0x57,0x04,0x0a,0x5e]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 04 0a 5e <unknown>
+# CHECK-UNKNOWN: 5e0a0457 <unknown>
 
 vmv.v.x v8, a0
 # CHECK-INST: vmv.v.x v8, a0
 # CHECK-ENCODING: [0x57,0x44,0x05,0x5e]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 44 05 5e <unknown>
+# CHECK-UNKNOWN: 5e054457 <unknown>
 
 vmv.v.i v8, 15
 # CHECK-INST: vmv.v.i v8, 15
 # CHECK-ENCODING: [0x57,0xb4,0x07,0x5e]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 b4 07 5e <unknown>
+# CHECK-UNKNOWN: 5e07b457 <unknown>
 
 vmv.x.s a2, v4
 # CHECK-INST: vmv.x.s a2, v4
 # CHECK-ENCODING: [0x57,0x26,0x40,0x42]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 26 40 42 <unknown>
+# CHECK-UNKNOWN: 42402657 <unknown>
 
 vmv.s.x v8, a0
 # CHECK-INST: vmv.s.x v8, a0
 # CHECK-ENCODING: [0x57,0x64,0x05,0x42]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 64 05 42 <unknown>
+# CHECK-UNKNOWN: 42056457 <unknown>
 
 vmv1r.v v8, v4
 # CHECK-INST: vmv1r.v v8, v4
 # CHECK-ENCODING: [0x57,0x34,0x40,0x9e]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 34 40 9e <unknown>
+# CHECK-UNKNOWN: 9e403457 <unknown>
 
 vmv2r.v v8, v4
 # CHECK-INST: vmv2r.v v8, v4
 # CHECK-ENCODING: [0x57,0xb4,0x40,0x9e]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 b4 40 9e <unknown>
+# CHECK-UNKNOWN: 9e40b457 <unknown>
 
 vmv4r.v v8, v4
 # CHECK-INST: vmv4r.v v8, v4
 # CHECK-ENCODING: [0x57,0xb4,0x41,0x9e]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 b4 41 9e <unknown>
+# CHECK-UNKNOWN: 9e41b457 <unknown>
 
 vmv8r.v v8, v24
 # CHECK-INST: vmv8r.v v8, v24
 # CHECK-ENCODING: [0x57,0xb4,0x83,0x9f]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 b4 83 9f <unknown>
+# CHECK-UNKNOWN: 9f83b457 <unknown>
diff --git a/llvm/test/MC/RISCV/rvv/or.s b/llvm/test/MC/RISCV/rvv/or.s
index ef281fe80bd4..306d7ae81442 100644
--- a/llvm/test/MC/RISCV/rvv/or.s
+++ b/llvm/test/MC/RISCV/rvv/or.s
@@ -12,34 +12,34 @@ vor.vv v8, v4, v20, v0.t
 # CHECK-INST: vor.vv v8, v4, v20, v0.t
 # CHECK-ENCODING: [0x57,0x04,0x4a,0x28]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 04 4a 28 <unknown>
+# CHECK-UNKNOWN: 284a0457 <unknown>
 
 vor.vv v8, v4, v20
 # CHECK-INST: vor.vv v8, v4, v20
 # CHECK-ENCODING: [0x57,0x04,0x4a,0x2a]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 04 4a 2a <unknown>
+# CHECK-UNKNOWN: 2a4a0457 <unknown>
 
 vor.vx v8, v4, a0, v0.t
 # CHECK-INST: vor.vx v8, v4, a0, v0.t
 # CHECK-ENCODING: [0x57,0x44,0x45,0x28]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 44 45 28 <unknown>
+# CHECK-UNKNOWN: 28454457 <unknown>
 
 vor.vx v8, v4, a0
 # CHECK-INST: vor.vx v8, v4, a0
 # CHECK-ENCODING: [0x57,0x44,0x45,0x2a]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 44 45 2a <unknown>
+# CHECK-UNKNOWN: 2a454457 <unknown>
 
 vor.vi v8, v4, 15, v0.t
 # CHECK-INST: vor.vi v8, v4, 15, v0.t
 # CHECK-ENCODING: [0x57,0xb4,0x47,0x28]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 b4 47 28 <unknown>
+# CHECK-UNKNOWN: 2847b457 <unknown>
 
 vor.vi v8, v4, 15
 # CHECK-INST: vor.vi v8, v4, 15
 # CHECK-ENCODING: [0x57,0xb4,0x47,0x2a]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 b4 47 2a <unknown>
+# CHECK-UNKNOWN: 2a47b457 <unknown>
diff --git a/llvm/test/MC/RISCV/rvv/others.s b/llvm/test/MC/RISCV/rvv/others.s
index d1845e0bb238..cc16a8774b82 100644
--- a/llvm/test/MC/RISCV/rvv/others.s
+++ b/llvm/test/MC/RISCV/rvv/others.s
@@ -12,142 +12,142 @@ vmerge.vvm v8, v4, v20, v0
 # CHECK-INST: vmerge.vvm v8, v4, v20, v0
 # CHECK-ENCODING: [0x57,0x04,0x4a,0x5c]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 04 4a 5c <unknown>
+# CHECK-UNKNOWN: 5c4a0457 <unknown>
 
 vmerge.vxm v8, v4, a0, v0
 # CHECK-INST: vmerge.vxm v8, v4, a0, v0
 # CHECK-ENCODING: [0x57,0x44,0x45,0x5c]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 44 45 5c <unknown>
+# CHECK-UNKNOWN: 5c454457 <unknown>
 
 vmerge.vim v8, v4, 15, v0
 # CHECK-INST: vmerge.vim v8, v4, 15, v0
 # CHECK-ENCODING: [0x57,0xb4,0x47,0x5c]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 b4 47 5c <unknown>
+# CHECK-UNKNOWN: 5c47b457 <unknown>
 
 vslideup.vx v8, v4, a0, v0.t
 # CHECK-INST: vslideup.vx v8, v4, a0, v0.t
 # CHECK-ENCODING: [0x57,0x44,0x45,0x38]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 44 45 38 <unknown>
+# CHECK-UNKNOWN: 38454457 <unknown>
 
 vslideup.vx v8, v4, a0
 # CHECK-INST: vslideup.vx v8, v4, a0
 # CHECK-ENCODING: [0x57,0x44,0x45,0x3a]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 44 45 3a <unknown>
+# CHECK-UNKNOWN: 3a454457 <unknown>
 
 vslideup.vi v8, v4, 31, v0.t
 # CHECK-INST: vslideup.vi v8, v4, 31, v0.t
 # CHECK-ENCODING: [0x57,0xb4,0x4f,0x38]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 b4 4f 38 <unknown>
+# CHECK-UNKNOWN: 384fb457 <unknown>
 
 vslideup.vi v8, v4, 31
 # CHECK-INST: vslideup.vi v8, v4, 31
 # CHECK-ENCODING: [0x57,0xb4,0x4f,0x3a]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 b4 4f 3a <unknown>
+# CHECK-UNKNOWN: 3a4fb457 <unknown>
 
 vslidedown.vx v8, v4, a0, v0.t
 # CHECK-INST: vslidedown.vx v8, v4, a0, v0.t
 # CHECK-ENCODING: [0x57,0x44,0x45,0x3c]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 44 45 3c <unknown>
+# CHECK-UNKNOWN: 3c454457 <unknown>
 
 vslidedown.vx v8, v4, a0
 # CHECK-INST: vslidedown.vx v8, v4, a0
 # CHECK-ENCODING: [0x57,0x44,0x45,0x3e]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 44 45 3e <unknown>
+# CHECK-UNKNOWN: 3e454457 <unknown>
 
 vslidedown.vi v8, v4, 31, v0.t
 # CHECK-INST: vslidedown.vi v8, v4, 31, v0.t
 # CHECK-ENCODING: [0x57,0xb4,0x4f,0x3c]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 b4 4f 3c <unknown>
+# CHECK-UNKNOWN: 3c4fb457 <unknown>
 
 vslidedown.vi v8, v4, 31
 # CHECK-INST: vslidedown.vi v8, v4, 31
 # CHECK-ENCODING: [0x57,0xb4,0x4f,0x3e]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 b4 4f 3e <unknown>
+# CHECK-UNKNOWN: 3e4fb457 <unknown>
 
 vslide1up.vx v8, v4, a0, v0.t
 # CHECK-INST: vslide1up.vx v8, v4, a0, v0.t
 # CHECK-ENCODING: [0x57,0x64,0x45,0x38]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 64 45 38 <unknown>
+# CHECK-UNKNOWN: 38456457 <unknown>
 
 vslide1up.vx v8, v4, a0
 # CHECK-INST: vslide1up.vx v8, v4, a0
 # CHECK-ENCODING: [0x57,0x64,0x45,0x3a]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 64 45 3a <unknown>
+# CHECK-UNKNOWN: 3a456457 <unknown>
 
 vslide1down.vx v8, v4, a0, v0.t
 # CHECK-INST: vslide1down.vx v8, v4, a0, v0.t
 # CHECK-ENCODING: [0x57,0x64,0x45,0x3c]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 64 45 3c <unknown>
+# CHECK-UNKNOWN: 3c456457 <unknown>
 
 vslide1down.vx v8, v4, a0
 # CHECK-INST: vslide1down.vx v8, v4, a0
 # CHECK-ENCODING: [0x57,0x64,0x45,0x3e]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 64 45 3e <unknown>
+# CHECK-UNKNOWN: 3e456457 <unknown>
 
 vrgather.vv v8, v4, v20, v0.t
 # CHECK-INST: vrgather.vv v8, v4, v20, v0.t
 # CHECK-ENCODING: [0x57,0x04,0x4a,0x30]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 04 4a 30 <unknown>
+# CHECK-UNKNOWN: 304a0457 <unknown>
 
 vrgather.vv v8, v4, v20
 # CHECK-INST: vrgather.vv v8, v4, v20
 # CHECK-ENCODING: [0x57,0x04,0x4a,0x32]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 04 4a 32 <unknown>
+# CHECK-UNKNOWN: 324a0457 <unknown>
 
 vrgather.vx v8, v4, a0, v0.t
 # CHECK-INST: vrgather.vx v8, v4, a0, v0.t
 # CHECK-ENCODING: [0x57,0x44,0x45,0x30]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 44 45 30 <unknown>
+# CHECK-UNKNOWN: 30454457 <unknown>
 
 vrgather.vx v8, v4, a0
 # CHECK-INST: vrgather.vx v8, v4, a0
 # CHECK-ENCODING: [0x57,0x44,0x45,0x32]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 44 45 32 <unknown>
+# CHECK-UNKNOWN: 32454457 <unknown>
 
 vrgather.vi v8, v4, 31, v0.t
 # CHECK-INST: vrgather.vi v8, v4, 31, v0.t
 # CHECK-ENCODING: [0x57,0xb4,0x4f,0x30]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 b4 4f 30 <unknown>
+# CHECK-UNKNOWN: 304fb457 <unknown>
 
 vrgather.vi v8, v4, 31
 # CHECK-INST: vrgather.vi v8, v4, 31
 # CHECK-ENCODING: [0x57,0xb4,0x4f,0x32]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 b4 4f 32 <unknown>
+# CHECK-UNKNOWN: 324fb457 <unknown>
 
 vrgatherei16.vv v8, v4, v20, v0.t
 # CHECK-INST: vrgatherei16.vv v8, v4, v20, v0.t
 # CHECK-ENCODING: [0x57,0x04,0x4a,0x38]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 04 4a 38 <unknown>
+# CHECK-UNKNOWN: 384a0457 <unknown>
 
 vrgatherei16.vv v8, v4, v20
 # CHECK-INST: vrgatherei16.vv v8, v4, v20
 # CHECK-ENCODING: [0x57,0x04,0x4a,0x3a]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 04 4a 3a <unknown>
+# CHECK-UNKNOWN: 3a4a0457 <unknown>
 
 vcompress.vm v8, v4, v20
 # CHECK-INST: vcompress.vm v8, v4, v20
 # CHECK-ENCODING: [0x57,0x24,0x4a,0x5e]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 24 4a 5e <unknown>
+# CHECK-UNKNOWN: 5e4a2457 <unknown>
diff --git a/llvm/test/MC/RISCV/rvv/reduction.s b/llvm/test/MC/RISCV/rvv/reduction.s
index 2172589b7c38..006f54d9b545 100644
--- a/llvm/test/MC/RISCV/rvv/reduction.s
+++ b/llvm/test/MC/RISCV/rvv/reduction.s
@@ -12,124 +12,124 @@ vredsum.vs v8, v4, v20, v0.t
 # CHECK-INST: vredsum.vs v8, v4, v20, v0.t
 # CHECK-ENCODING: [0x57,0x24,0x4a,0x00]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 24 4a 00 <unknown>
+# CHECK-UNKNOWN: 004a2457 <unknown>
 
 vredsum.vs v8, v4, v20
 # CHECK-INST: vredsum.vs v8, v4, v20
 # CHECK-ENCODING: [0x57,0x24,0x4a,0x02]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 24 4a 02 <unknown>
+# CHECK-UNKNOWN: 024a2457 <unknown>
 
 vredmaxu.vs v8, v4, v20, v0.t
 # CHECK-INST: vredmaxu.vs v8, v4, v20, v0.t
 # CHECK-ENCODING: [0x57,0x24,0x4a,0x18]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 24 4a 18 <unknown>
+# CHECK-UNKNOWN: 184a2457 <unknown>
 
 vredmaxu.vs v8, v4, v20
 # CHECK-INST: vredmaxu.vs v8, v4, v20
 # CHECK-ENCODING: [0x57,0x24,0x4a,0x1a]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 24 4a 1a <unknown>
+# CHECK-UNKNOWN: 1a4a2457 <unknown>
 
 vredmax.vs v8, v4, v20, v0.t
 # CHECK-INST: vredmax.vs v8, v4, v20, v0.t
 # CHECK-ENCODING: [0x57,0x24,0x4a,0x1c]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 24 4a 1c <unknown>
+# CHECK-UNKNOWN: 1c4a2457 <unknown>
 
 vredmax.vs v8, v4, v20
 # CHECK-INST: vredmax.vs v8, v4, v20
 # CHECK-ENCODING: [0x57,0x24,0x4a,0x1e]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 24 4a 1e <unknown>
+# CHECK-UNKNOWN: 1e4a2457 <unknown>
 
 vredminu.vs v8, v4, v20, v0.t
 # CHECK-INST: vredminu.vs v8, v4, v20, v0.t
 # CHECK-ENCODING: [0x57,0x24,0x4a,0x10]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 24 4a 10 <unknown>
+# CHECK-UNKNOWN: 104a2457 <unknown>
 
 vredminu.vs v8, v4, v20
 # CHECK-INST: vredminu.vs v8, v4, v20
 # CHECK-ENCODING: [0x57,0x24,0x4a,0x12]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 24 4a 12 <unknown>
+# CHECK-UNKNOWN: 124a2457 <unknown>
 
 vredmin.vs v8, v4, v20, v0.t
 # CHECK-INST: vredmin.vs v8, v4, v20, v0.t
 # CHECK-ENCODING: [0x57,0x24,0x4a,0x14]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 24 4a 14 <unknown>
+# CHECK-UNKNOWN: 144a2457 <unknown>
 
 vredmin.vs v8, v4, v20
 # CHECK-INST: vredmin.vs v8, v4, v20
 # CHECK-ENCODING: [0x57,0x24,0x4a,0x16]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 24 4a 16 <unknown>
+# CHECK-UNKNOWN: 164a2457 <unknown>
 
 vredand.vs v8, v4, v20, v0.t
 # CHECK-INST: vredand.vs v8, v4, v20, v0.t
 # CHECK-ENCODING: [0x57,0x24,0x4a,0x04]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 24 4a 04 <unknown>
+# CHECK-UNKNOWN: 044a2457 <unknown>
 
 vredand.vs v8, v4, v20
 # CHECK-INST: vredand.vs v8, v4, v20
 # CHECK-ENCODING: [0x57,0x24,0x4a,0x06]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 24 4a 06 <unknown>
+# CHECK-UNKNOWN: 064a2457 <unknown>
 
 vredor.vs v8, v4, v20, v0.t
 # CHECK-INST: vredor.vs v8, v4, v20, v0.t
 # CHECK-ENCODING: [0x57,0x24,0x4a,0x08]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 24 4a 08 <unknown>
+# CHECK-UNKNOWN: 084a2457 <unknown>
 
 vredor.vs v8, v4, v20
 # CHECK-INST: vredor.vs v8, v4, v20
 # CHECK-ENCODING: [0x57,0x24,0x4a,0x0a]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 24 4a 0a <unknown>
+# CHECK-UNKNOWN: 0a4a2457 <unknown>
 
 vredxor.vs v8, v4, v20, v0.t
 # CHECK-INST: vredxor.vs v8, v4, v20, v0.t
 # CHECK-ENCODING: [0x57,0x24,0x4a,0x0c]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 24 4a 0c <unknown>
+# CHECK-UNKNOWN: 0c4a2457 <unknown>
 
 vredxor.vs v8, v4, v20
 # CHECK-INST: vredxor.vs v8, v4, v20
 # CHECK-ENCODING: [0x57,0x24,0x4a,0x0e]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 24 4a 0e <unknown>
+# CHECK-UNKNOWN: 0e4a2457 <unknown>
 
 vwredsumu.vs v8, v4, v20, v0.t
 # CHECK-INST: vwredsumu.vs v8, v4, v20, v0.t
 # CHECK-ENCODING: [0x57,0x04,0x4a,0xc0]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 04 4a c0 <unknown>
+# CHECK-UNKNOWN: c04a0457 <unknown>
 
 vwredsumu.vs v8, v4, v20
 # CHECK-INST: vwredsumu.vs v8, v4, v20
 # CHECK-ENCODING: [0x57,0x04,0x4a,0xc2]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 04 4a c2 <unknown>
+# CHECK-UNKNOWN: c24a0457 <unknown>
 
 vwredsum.vs v8, v4, v20, v0.t
 # CHECK-INST: vwredsum.vs v8, v4, v20, v0.t
 # CHECK-ENCODING: [0x57,0x04,0x4a,0xc4]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 04 4a c4 <unknown>
+# CHECK-UNKNOWN: c44a0457 <unknown>
 
 vwredsum.vs v8, v4, v20
 # CHECK-INST: vwredsum.vs v8, v4, v20
 # CHECK-ENCODING: [0x57,0x04,0x4a,0xc6]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 04 4a c6 <unknown>
+# CHECK-UNKNOWN: c64a0457 <unknown>
 
 vredsum.vs v0, v4, v20, v0.t
 # CHECK-INST: vredsum.vs v0, v4, v20, v0.t
 # CHECK-ENCODING: [0x57,0x20,0x4a,0x00]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 20 4a 00 <unknown>
+# CHECK-UNKNOWN: 004a2057 <unknown>
diff --git a/llvm/test/MC/RISCV/rvv/shift.s b/llvm/test/MC/RISCV/rvv/shift.s
index 8a2e82f3c8fe..017e12499dc2 100644
--- a/llvm/test/MC/RISCV/rvv/shift.s
+++ b/llvm/test/MC/RISCV/rvv/shift.s
@@ -12,256 +12,256 @@ vsll.vv v8, v4, v20, v0.t
 # CHECK-INST: vsll.vv v8, v4, v20, v0.t
 # CHECK-ENCODING: [0x57,0x04,0x4a,0x94]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 04 4a 94 <unknown>
+# CHECK-UNKNOWN: 944a0457 <unknown>
 
 vsll.vv v8, v4, v20
 # CHECK-INST: vsll.vv v8, v4, v20
 # CHECK-ENCODING: [0x57,0x04,0x4a,0x96]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 04 4a 96 <unknown>
+# CHECK-UNKNOWN: 964a0457 <unknown>
 
 vsll.vx v8, v4, a0, v0.t
 # CHECK-INST: vsll.vx v8, v4, a0, v0.t
 # CHECK-ENCODING: [0x57,0x44,0x45,0x94]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 44 45 94 <unknown>
+# CHECK-UNKNOWN: 94454457 <unknown>
 
 vsll.vx v8, v4, a0
 # CHECK-INST: vsll.vx v8, v4, a0
 # CHECK-ENCODING: [0x57,0x44,0x45,0x96]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 44 45 96 <unknown>
+# CHECK-UNKNOWN: 96454457 <unknown>
 
 vsll.vi v8, v4, 31, v0.t
 # CHECK-INST: vsll.vi v8, v4, 31, v0.t
 # CHECK-ENCODING: [0x57,0xb4,0x4f,0x94]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 b4 4f 94 <unknown>
+# CHECK-UNKNOWN: 944fb457 <unknown>
 
 vsll.vi v8, v4, 31
 # CHECK-INST: vsll.vi v8, v4, 31
 # CHECK-ENCODING: [0x57,0xb4,0x4f,0x96]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 b4 4f 96 <unknown>
+# CHECK-UNKNOWN: 964fb457 <unknown>
 
 vsrl.vv v8, v4, v20, v0.t
 # CHECK-INST: vsrl.vv v8, v4, v20, v0.t
 # CHECK-ENCODING: [0x57,0x04,0x4a,0xa0]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 04 4a a0 <unknown>
+# CHECK-UNKNOWN: a04a0457 <unknown>
 
 vsrl.vv v8, v4, v20
 # CHECK-INST: vsrl.vv v8, v4, v20
 # CHECK-ENCODING: [0x57,0x04,0x4a,0xa2]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 04 4a a2 <unknown>
+# CHECK-UNKNOWN: a24a0457 <unknown>
 
 vsrl.vx v8, v4, a0, v0.t
 # CHECK-INST: vsrl.vx v8, v4, a0, v0.t
 # CHECK-ENCODING: [0x57,0x44,0x45,0xa0]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 44 45 a0 <unknown>
+# CHECK-UNKNOWN: a0454457 <unknown>
 
 vsrl.vx v8, v4, a0
 # CHECK-INST: vsrl.vx v8, v4, a0
 # CHECK-ENCODING: [0x57,0x44,0x45,0xa2]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 44 45 a2 <unknown>
+# CHECK-UNKNOWN: a2454457 <unknown>
 
 vsrl.vi v8, v4, 31, v0.t
 # CHECK-INST: vsrl.vi v8, v4, 31, v0.t
 # CHECK-ENCODING: [0x57,0xb4,0x4f,0xa0]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 b4 4f a0 <unknown>
+# CHECK-UNKNOWN: a04fb457 <unknown>
 
 vsrl.vi v8, v4, 31
 # CHECK-INST: vsrl.vi v8, v4, 31
 # CHECK-ENCODING: [0x57,0xb4,0x4f,0xa2]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 b4 4f a2 <unknown>
+# CHECK-UNKNOWN: a24fb457 <unknown>
 
 vsra.vv v8, v4, v20, v0.t
 # CHECK-INST: vsra.vv v8, v4, v20, v0.t
 # CHECK-ENCODING: [0x57,0x04,0x4a,0xa4]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 04 4a a4 <unknown>
+# CHECK-UNKNOWN: a44a0457 <unknown>
 
 vsra.vv v8, v4, v20
 # CHECK-INST: vsra.vv v8, v4, v20
 # CHECK-ENCODING: [0x57,0x04,0x4a,0xa6]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 04 4a a6 <unknown>
+# CHECK-UNKNOWN: a64a0457 <unknown>
 
 vsra.vx v8, v4, a0, v0.t
 # CHECK-INST: vsra.vx v8, v4, a0, v0.t
 # CHECK-ENCODING: [0x57,0x44,0x45,0xa4]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 44 45 a4 <unknown>
+# CHECK-UNKNOWN: a4454457 <unknown>
 
 vsra.vx v8, v4, a0
 # CHECK-INST: vsra.vx v8, v4, a0
 # CHECK-ENCODING: [0x57,0x44,0x45,0xa6]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 44 45 a6 <unknown>
+# CHECK-UNKNOWN: a6454457 <unknown>
 
 vsra.vi v8, v4, 31, v0.t
 # CHECK-INST: vsra.vi v8, v4, 31, v0.t
 # CHECK-ENCODING: [0x57,0xb4,0x4f,0xa4]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 b4 4f a4 <unknown>
+# CHECK-UNKNOWN: a44fb457 <unknown>
 
 vsra.vi v8, v4, 31
 # CHECK-INST: vsra.vi v8, v4, 31
 # CHECK-ENCODING: [0x57,0xb4,0x4f,0xa6]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 b4 4f a6 <unknown>
+# CHECK-UNKNOWN: a64fb457 <unknown>
 
 vnsrl.wv v8, v4, v20, v0.t
 # CHECK-INST: vnsrl.wv v8, v4, v20, v0.t
 # CHECK-ENCODING: [0x57,0x04,0x4a,0xb0]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 04 4a b0 <unknown>
+# CHECK-UNKNOWN: b04a0457 <unknown>
 
 vnsrl.wv v4, v4, v20, v0.t
 # CHECK-INST: vnsrl.wv v4, v4, v20, v0.t
 # CHECK-ENCODING: [0x57,0x02,0x4a,0xb0]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 02 4a b0 <unknown>
+# CHECK-UNKNOWN: b04a0257 <unknown>
 
 vnsrl.wv v8, v4, v20
 # CHECK-INST: vnsrl.wv v8, v4, v20
 # CHECK-ENCODING: [0x57,0x04,0x4a,0xb2]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 04 4a b2 <unknown>
+# CHECK-UNKNOWN: b24a0457 <unknown>
 
 vnsrl.wx v8, v4, a0, v0.t
 # CHECK-INST: vnsrl.wx v8, v4, a0, v0.t
 # CHECK-ENCODING: [0x57,0x44,0x45,0xb0]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 44 45 b0 <unknown>
+# CHECK-UNKNOWN: b0454457 <unknown>
 
 vnsrl.wx v8, v4, a0
 # CHECK-INST: vnsrl.wx v8, v4, a0
 # CHECK-ENCODING: [0x57,0x44,0x45,0xb2]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 44 45 b2 <unknown>
+# CHECK-UNKNOWN: b2454457 <unknown>
 
 vnsrl.wi v8, v4, 31, v0.t
 # CHECK-INST: vnsrl.wi v8, v4, 31, v0.t
 # CHECK-ENCODING: [0x57,0xb4,0x4f,0xb0]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 b4 4f b0 <unknown>
+# CHECK-UNKNOWN: b04fb457 <unknown>
 
 vnsrl.wi v8, v4, 31
 # CHECK-INST: vnsrl.wi v8, v4, 31
 # CHECK-ENCODING: [0x57,0xb4,0x4f,0xb2]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 b4 4f b2 <unknown>
+# CHECK-UNKNOWN: b24fb457 <unknown>
 
 vnsra.wv v8, v4, v20, v0.t
 # CHECK-INST: vnsra.wv v8, v4, v20, v0.t
 # CHECK-ENCODING: [0x57,0x04,0x4a,0xb4]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 04 4a b4 <unknown>
+# CHECK-UNKNOWN: b44a0457 <unknown>
 
 vnsra.wv v8, v4, v20
 # CHECK-INST: vnsra.wv v8, v4, v20
 # CHECK-ENCODING: [0x57,0x04,0x4a,0xb6]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 04 4a b6 <unknown>
+# CHECK-UNKNOWN: b64a0457 <unknown>
 
 vnsra.wx v8, v4, a0, v0.t
 # CHECK-INST: vnsra.wx v8, v4, a0, v0.t
 # CHECK-ENCODING: [0x57,0x44,0x45,0xb4]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 44 45 b4 <unknown>
+# CHECK-UNKNOWN: b4454457 <unknown>
 
 vnsra.wx v8, v4, a0
 # CHECK-INST: vnsra.wx v8, v4, a0
 # CHECK-ENCODING: [0x57,0x44,0x45,0xb6]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 44 45 b6 <unknown>
+# CHECK-UNKNOWN: b6454457 <unknown>
 
 vnsra.wi v8, v4, 31, v0.t
 # CHECK-INST: vnsra.wi v8, v4, 31, v0.t
 # CHECK-ENCODING: [0x57,0xb4,0x4f,0xb4]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 b4 4f b4 <unknown>
+# CHECK-UNKNOWN: b44fb457 <unknown>
 
 vnsra.wi v8, v4, 31
 # CHECK-INST: vnsra.wi v8, v4, 31
 # CHECK-ENCODING: [0x57,0xb4,0x4f,0xb6]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 b4 4f b6 <unknown>
+# CHECK-UNKNOWN: b64fb457 <unknown>
 
 vssrl.vv v8, v4, v20, v0.t
 # CHECK-INST: vssrl.vv v8, v4, v20, v0.t
 # CHECK-ENCODING: [0x57,0x04,0x4a,0xa8]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 04 4a a8 <unknown>
+# CHECK-UNKNOWN: a84a0457 <unknown>
 
 vssrl.vv v8, v4, v20
 # CHECK-INST: vssrl.vv v8, v4, v20
 # CHECK-ENCODING: [0x57,0x04,0x4a,0xaa]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 04 4a aa <unknown>
+# CHECK-UNKNOWN: aa4a0457 <unknown>
 
 vssrl.vx v8, v4, a0, v0.t
 # CHECK-INST: vssrl.vx v8, v4, a0, v0.t
 # CHECK-ENCODING: [0x57,0x44,0x45,0xa8]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 44 45 a8 <unknown>
+# CHECK-UNKNOWN: a8454457 <unknown>
 
 vssrl.vx v8, v4, a0
 # CHECK-INST: vssrl.vx v8, v4, a0
 # CHECK-ENCODING: [0x57,0x44,0x45,0xaa]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 44 45 aa <unknown>
+# CHECK-UNKNOWN: aa454457 <unknown>
 
 vssrl.vi v8, v4, 31, v0.t
 # CHECK-INST: vssrl.vi v8, v4, 31, v0.t
 # CHECK-ENCODING: [0x57,0xb4,0x4f,0xa8]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 b4 4f a8 <unknown>
+# CHECK-UNKNOWN: a84fb457 <unknown>
 
 vssrl.vi v8, v4, 31
 # CHECK-INST: vssrl.vi v8, v4, 31
 # CHECK-ENCODING: [0x57,0xb4,0x4f,0xaa]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 b4 4f aa <unknown>
+# CHECK-UNKNOWN: aa4fb457 <unknown>
 
 vssra.vv v8, v4, v20, v0.t
 # CHECK-INST: vssra.vv v8, v4, v20, v0.t
 # CHECK-ENCODING: [0x57,0x04,0x4a,0xac]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 04 4a ac <unknown>
+# CHECK-UNKNOWN: ac4a0457 <unknown>
 
 vssra.vv v8, v4, v20
 # CHECK-INST: vssra.vv v8, v4, v20
 # CHECK-ENCODING: [0x57,0x04,0x4a,0xae]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 04 4a ae <unknown>
+# CHECK-UNKNOWN: ae4a0457 <unknown>
 
 vssra.vx v8, v4, a0, v0.t
 # CHECK-INST: vssra.vx v8, v4, a0, v0.t
 # CHECK-ENCODING: [0x57,0x44,0x45,0xac]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 44 45 ac <unknown>
+# CHECK-UNKNOWN: ac454457 <unknown>
 
 vssra.vx v8, v4, a0
 # CHECK-INST: vssra.vx v8, v4, a0
 # CHECK-ENCODING: [0x57,0x44,0x45,0xae]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 44 45 ae <unknown>
+# CHECK-UNKNOWN: ae454457 <unknown>
 
 vssra.vi v8, v4, 31, v0.t
 # CHECK-INST: vssra.vi v8, v4, 31, v0.t
 # CHECK-ENCODING: [0x57,0xb4,0x4f,0xac]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 b4 4f ac <unknown>
+# CHECK-UNKNOWN: ac4fb457 <unknown>
 
 vssra.vi v8, v4, 31
 # CHECK-INST: vssra.vi v8, v4, 31
 # CHECK-ENCODING: [0x57,0xb4,0x4f,0xae]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 b4 4f ae <unknown>
+# CHECK-UNKNOWN: ae4fb457 <unknown>
diff --git a/llvm/test/MC/RISCV/rvv/sign-injection.s b/llvm/test/MC/RISCV/rvv/sign-injection.s
index 96d37c42d208..23e9be868a42 100644
--- a/llvm/test/MC/RISCV/rvv/sign-injection.s
+++ b/llvm/test/MC/RISCV/rvv/sign-injection.s
@@ -15,70 +15,70 @@ vfsgnj.vv v8, v4, v20, v0.t
 # CHECK-INST: vfsgnj.vv v8, v4, v20, v0.t
 # CHECK-ENCODING: [0x57,0x14,0x4a,0x20]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 14 4a 20 <unknown>
+# CHECK-UNKNOWN: 204a1457 <unknown>
 
 vfsgnj.vv v8, v4, v20
 # CHECK-INST: vfsgnj.vv v8, v4, v20
 # CHECK-ENCODING: [0x57,0x14,0x4a,0x22]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 14 4a 22 <unknown>
+# CHECK-UNKNOWN: 224a1457 <unknown>
 
 vfsgnj.vf v8, v4, fa0, v0.t
 # CHECK-INST: vfsgnj.vf v8, v4, fa0, v0.t
 # CHECK-ENCODING: [0x57,0x54,0x45,0x20]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 54 45 20 <unknown>
+# CHECK-UNKNOWN: 20455457 <unknown>
 
 vfsgnj.vf v8, v4, fa0
 # CHECK-INST: vfsgnj.vf v8, v4, fa0
 # CHECK-ENCODING: [0x57,0x54,0x45,0x22]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 54 45 22 <unknown>
+# CHECK-UNKNOWN: 22455457 <unknown>
 
 vfsgnjn.vv v8, v4, v20, v0.t
 # CHECK-INST: vfsgnjn.vv v8, v4, v20, v0.t
 # CHECK-ENCODING: [0x57,0x14,0x4a,0x24]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 14 4a 24 <unknown>
+# CHECK-UNKNOWN: 244a1457 <unknown>
 
 vfsgnjn.vv v8, v4, v20
 # CHECK-INST: vfsgnjn.vv v8, v4, v20
 # CHECK-ENCODING: [0x57,0x14,0x4a,0x26]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 14 4a 26 <unknown>
+# CHECK-UNKNOWN: 264a1457 <unknown>
 
 vfsgnjn.vf v8, v4, fa0, v0.t
 # CHECK-INST: vfsgnjn.vf v8, v4, fa0, v0.t
 # CHECK-ENCODING: [0x57,0x54,0x45,0x24]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 54 45 24 <unknown>
+# CHECK-UNKNOWN: 24455457 <unknown>
 
 vfsgnjn.vf v8, v4, fa0
 # CHECK-INST: vfsgnjn.vf v8, v4, fa0
 # CHECK-ENCODING: [0x57,0x54,0x45,0x26]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 54 45 26 <unknown>
+# CHECK-UNKNOWN: 26455457 <unknown>
 
 vfsgnjx.vv v8, v4, v20, v0.t
 # CHECK-INST: vfsgnjx.vv v8, v4, v20, v0.t
 # CHECK-ENCODING: [0x57,0x14,0x4a,0x28]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 14 4a 28 <unknown>
+# CHECK-UNKNOWN: 284a1457 <unknown>
 
 vfsgnjx.vv v8, v4, v20
 # CHECK-INST: vfsgnjx.vv v8, v4, v20
 # CHECK-ENCODING: [0x57,0x14,0x4a,0x2a]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 14 4a 2a <unknown>
+# CHECK-UNKNOWN: 2a4a1457 <unknown>
 
 vfsgnjx.vf v8, v4, fa0, v0.t
 # CHECK-INST: vfsgnjx.vf v8, v4, fa0, v0.t
 # CHECK-ENCODING: [0x57,0x54,0x45,0x28]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 54 45 28 <unknown>
+# CHECK-UNKNOWN: 28455457 <unknown>
 
 vfsgnjx.vf v8, v4, fa0
 # CHECK-INST: vfsgnjx.vf v8, v4, fa0
 # CHECK-ENCODING: [0x57,0x54,0x45,0x2a]
 # CHECK-ERROR: instruction requires the following: 'V'{{.*}}'Zve32f' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 54 45 2a <unknown>
+# CHECK-UNKNOWN: 2a455457 <unknown>
diff --git a/llvm/test/MC/RISCV/rvv/snippet.s b/llvm/test/MC/RISCV/rvv/snippet.s
index c032e468dede..c3e57e193f84 100644
--- a/llvm/test/MC/RISCV/rvv/snippet.s
+++ b/llvm/test/MC/RISCV/rvv/snippet.s
@@ -6,27 +6,27 @@
 
 loop:
     vsetvli a3, a0, e16,m4,ta,ma  # vtype = 16-bit integer vectors
-# CHECK-INST: d7 76 a5 0c    vsetvli a3, a0, e16, m4, ta, ma
+# CHECK-INST: 0ca576d7 vsetvli a3, a0, e16, m4, ta, ma
     vle16.v v4, (a1)              # Get 16b vector
-# CHECK-INST: 07 d2 05 02    vle16.v   v4, (a1)
+# CHECK-INST: 0205d207 vle16.v v4, (a1)
     slli t1, a3, 1                # Multiply length by two bytes/element
-# CHECK-INST: 13 93 16 00    slli    t1, a3, 0x1
+# CHECK-INST: 00169313 slli t1, a3, 0x1
     add a1, a1, t1                # Bump pointer
-# CHECK-INST: b3 85 65 00    add     a1, a1, t1
+# CHECK-INST: 006585b3 add a1, a1, t1
     vwmul.vx v8, v4, x10          # 32b in <v8--v15>
-# CHECK-INST: 57 64 45 ee    vwmul.vx        v8, v4, a0
+# CHECK-INST: ee456457 vwmul.vx v8, v4, a0
 
     vsetvli x0, a0, e32,m8,ta,ma  # Operate on 32b values
-# CHECK-INST: 57 70 35 0d    vsetvli zero, a0, e32, m8, ta, ma
+# CHECK-INST: 0d357057 vsetvli zero, a0, e32, m8, ta, ma
     vsrl.vi v8, v8, 3
-# CHECK-INST: 57 b4 81 a2    vsrl.vi v8, v8, 0x3
+# CHECK-INST: a281b457 vsrl.vi v8, v8, 0x3
     vse32.v v8, (a2)              # Store vector of 32b
-# CHECK-INST: 27 64 06 02    vse32.v   v8, (a2)
+# CHECK-INST: 02066427 vse32.v v8, (a2)
     slli t1, a3, 2                # Multiply length by four bytes/element
-# CHECK-INST: 13 93 26 00    slli    t1, a3, 0x2
+# CHECK-INST: 00269313 slli t1, a3, 0x2
     add a2, a2, t1                # Bump pointer
-# CHECK-INST: 33 06 66 00    add     a2, a2, t1
+# CHECK-INST: 00660633 add a2, a2, t1
     sub a0, a0, a3                # Decrement count
-# CHECK-INST: 33 05 d5 40    sub     a0, a0, a3
+# CHECK-INST: 40d50533 sub a0, a0, a3
     bnez a0, loop                 # Any more?
-# CHECK-INST: e3 1a 05 fc    bnez    a0, 0x0
+# CHECK-INST: fc051ae3 bnez a0, 0x0
diff --git a/llvm/test/MC/RISCV/rvv/store.s b/llvm/test/MC/RISCV/rvv/store.s
index a38f19f266fa..c6a34705fa4a 100644
--- a/llvm/test/MC/RISCV/rvv/store.s
+++ b/llvm/test/MC/RISCV/rvv/store.s
@@ -12,250 +12,250 @@ vsm.v v24, (a0)
 # CHECK-INST: vsm.v v24, (a0)
 # CHECK-ENCODING: [0x27,0x0c,0xb5,0x02]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 0c b5 02 <unknown>
+# CHECK-UNKNOWN: 02b50c27 <unknown>
 
 vse8.v v24, (a0), v0.t
 # CHECK-INST: vse8.v v24, (a0), v0.t
 # CHECK-ENCODING: [0x27,0x0c,0x05,0x00]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 0c 05 00 <unknown>
+# CHECK-UNKNOWN: 00050c27 <unknown>
 
 vse8.v v24, (a0)
 # CHECK-INST: vse8.v v24, (a0)
 # CHECK-ENCODING: [0x27,0x0c,0x05,0x02]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 0c 05 02 <unknown>
+# CHECK-UNKNOWN: 02050c27 <unknown>
 
 vse16.v v24, (a0), v0.t
 # CHECK-INST: vse16.v v24, (a0), v0.t
 # CHECK-ENCODING: [0x27,0x5c,0x05,0x00]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 5c 05 00 <unknown>
+# CHECK-UNKNOWN: 00055c27 <unknown>
 
 vse16.v v24, (a0)
 # CHECK-INST: vse16.v v24, (a0)
 # CHECK-ENCODING: [0x27,0x5c,0x05,0x02]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 5c 05 02 <unknown>
+# CHECK-UNKNOWN: 02055c27 <unknown>
 
 vse32.v v24, (a0), v0.t
 # CHECK-INST: vse32.v v24, (a0), v0.t
 # CHECK-ENCODING: [0x27,0x6c,0x05,0x00]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 6c 05 00 <unknown>
+# CHECK-UNKNOWN: 00056c27 <unknown>
 
 vse32.v v24, (a0)
 # CHECK-INST: vse32.v v24, (a0)
 # CHECK-ENCODING: [0x27,0x6c,0x05,0x02]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 6c 05 02 <unknown>
+# CHECK-UNKNOWN: 02056c27 <unknown>
 
 vse64.v v24, (a0), v0.t
 # CHECK-INST: vse64.v v24, (a0), v0.t
 # CHECK-ENCODING: [0x27,0x7c,0x05,0x00]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 7c 05 00 <unknown>
+# CHECK-UNKNOWN: 00057c27 <unknown>
 
 vse64.v v24, (a0)
 # CHECK-INST: vse64.v v24, (a0)
 # CHECK-ENCODING: [0x27,0x7c,0x05,0x02]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 7c 05 02 <unknown>
+# CHECK-UNKNOWN: 02057c27 <unknown>
 
 vsse8.v v24, (a0), a1, v0.t
 # CHECK-INST: vsse8.v v24, (a0), a1, v0.t
 # CHECK-ENCODING: [0x27,0x0c,0xb5,0x08]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 0c b5 08 <unknown>
+# CHECK-UNKNOWN: 08b50c27 <unknown>
 
 vsse8.v v24, (a0), a1
 # CHECK-INST: vsse8.v v24, (a0), a1
 # CHECK-ENCODING: [0x27,0x0c,0xb5,0x0a]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 0c b5 0a <unknown>
+# CHECK-UNKNOWN: 0ab50c27 <unknown>
 
 vsse16.v v24, (a0), a1, v0.t
 # CHECK-INST: vsse16.v v24, (a0), a1, v0.t
 # CHECK-ENCODING: [0x27,0x5c,0xb5,0x08]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 5c b5 08 <unknown>
+# CHECK-UNKNOWN: 08b55c27 <unknown>
 
 vsse16.v v24, (a0), a1
 # CHECK-INST: vsse16.v v24, (a0), a1
 # CHECK-ENCODING: [0x27,0x5c,0xb5,0x0a]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 5c b5 0a <unknown>
+# CHECK-UNKNOWN: 0ab55c27 <unknown>
 
 vsse32.v v24, (a0), a1, v0.t
 # CHECK-INST: vsse32.v v24, (a0), a1, v0.t
 # CHECK-ENCODING: [0x27,0x6c,0xb5,0x08]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 6c b5 08 <unknown>
+# CHECK-UNKNOWN: 08b56c27 <unknown>
 
 vsse32.v v24, (a0), a1
 # CHECK-INST: vsse32.v v24, (a0), a1
 # CHECK-ENCODING: [0x27,0x6c,0xb5,0x0a]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 6c b5 0a <unknown>
+# CHECK-UNKNOWN: 0ab56c27 <unknown>
 
 vsse64.v v24, (a0), a1, v0.t
 # CHECK-INST: vsse64.v v24, (a0), a1, v0.t
 # CHECK-ENCODING: [0x27,0x7c,0xb5,0x08]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 7c b5 08 <unknown>
+# CHECK-UNKNOWN: 08b57c27 <unknown>
 
 vsse64.v v24, (a0), a1
 # CHECK-INST: vsse64.v v24, (a0), a1
 # CHECK-ENCODING: [0x27,0x7c,0xb5,0x0a]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 7c b5 0a <unknown>
+# CHECK-UNKNOWN: 0ab57c27 <unknown>
 
 vsuxei8.v v24, (a0), v4, v0.t
 # CHECK-INST: vsuxei8.v v24, (a0), v4, v0.t
 # CHECK-ENCODING: [0x27,0x0c,0x45,0x04]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 0c 45 04 <unknown>
+# CHECK-UNKNOWN: 04450c27 <unknown>
 
 vsuxei8.v v24, (a0), v4
 # CHECK-INST: vsuxei8.v v24, (a0), v4
 # CHECK-ENCODING: [0x27,0x0c,0x45,0x06]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 0c 45 06 <unknown>
+# CHECK-UNKNOWN: 06450c27 <unknown>
 
 vsuxei16.v v24, (a0), v4, v0.t
 # CHECK-INST: vsuxei16.v v24, (a0), v4, v0.t
 # CHECK-ENCODING: [0x27,0x5c,0x45,0x04]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 5c 45 04 <unknown>
+# CHECK-UNKNOWN: 04455c27 <unknown>
 
 vsuxei16.v v24, (a0), v4
 # CHECK-INST: vsuxei16.v v24, (a0), v4
 # CHECK-ENCODING: [0x27,0x5c,0x45,0x06]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 5c 45 06 <unknown>
+# CHECK-UNKNOWN: 06455c27 <unknown>
 
 vsuxei32.v v24, (a0), v4, v0.t
 # CHECK-INST: vsuxei32.v v24, (a0), v4, v0.t
 # CHECK-ENCODING: [0x27,0x6c,0x45,0x04]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 6c 45 04 <unknown>
+# CHECK-UNKNOWN: 04456c27 <unknown>
 
 vsuxei32.v v24, (a0), v4
 # CHECK-INST: vsuxei32.v v24, (a0), v4
 # CHECK-ENCODING: [0x27,0x6c,0x45,0x06]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 6c 45 06 <unknown>
+# CHECK-UNKNOWN: 06456c27 <unknown>
 
 vsuxei64.v v24, (a0), v4, v0.t
 # CHECK-INST: vsuxei64.v v24, (a0), v4, v0.t
 # CHECK-ENCODING: [0x27,0x7c,0x45,0x04]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 7c 45 04 <unknown>
+# CHECK-UNKNOWN: 04457c27 <unknown>
 
 vsuxei64.v v24, (a0), v4
 # CHECK-INST: vsuxei64.v v24, (a0), v4
 # CHECK-ENCODING: [0x27,0x7c,0x45,0x06]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 7c 45 06 <unknown>
+# CHECK-UNKNOWN: 06457c27 <unknown>
 
 vsoxei8.v v24, (a0), v4, v0.t
 # CHECK-INST: vsoxei8.v v24, (a0), v4, v0.t
 # CHECK-ENCODING: [0x27,0x0c,0x45,0x0c]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 0c 45 0c <unknown>
+# CHECK-UNKNOWN: 0c450c27 <unknown>
 
 vsoxei8.v v24, (a0), v4
 # CHECK-INST: vsoxei8.v v24, (a0), v4
 # CHECK-ENCODING: [0x27,0x0c,0x45,0x0e]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 0c 45 0e <unknown>
+# CHECK-UNKNOWN: 0e450c27 <unknown>
 
 vsoxei16.v v24, (a0), v4, v0.t
 # CHECK-INST: vsoxei16.v v24, (a0), v4, v0.t
 # CHECK-ENCODING: [0x27,0x5c,0x45,0x0c]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 5c 45 0c <unknown>
+# CHECK-UNKNOWN: 0c455c27 <unknown>
 
 vsoxei16.v v24, (a0), v4
 # CHECK-INST: vsoxei16.v v24, (a0), v4
 # CHECK-ENCODING: [0x27,0x5c,0x45,0x0e]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 5c 45 0e <unknown>
+# CHECK-UNKNOWN: 0e455c27 <unknown>
 
 vsoxei32.v v24, (a0), v4, v0.t
 # CHECK-INST: vsoxei32.v v24, (a0), v4, v0.t
 # CHECK-ENCODING: [0x27,0x6c,0x45,0x0c]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 6c 45 0c <unknown>
+# CHECK-UNKNOWN: 0c456c27 <unknown>
 
 vsoxei32.v v24, (a0), v4
 # CHECK-INST: vsoxei32.v v24, (a0), v4
 # CHECK-ENCODING: [0x27,0x6c,0x45,0x0e]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 6c 45 0e <unknown>
+# CHECK-UNKNOWN: 0e456c27 <unknown>
 
 vsoxei64.v v24, (a0), v4, v0.t
 # CHECK-INST: vsoxei64.v v24, (a0), v4, v0.t
 # CHECK-ENCODING: [0x27,0x7c,0x45,0x0c]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 7c 45 0c <unknown>
+# CHECK-UNKNOWN: 0c457c27 <unknown>
 
 vsoxei64.v v24, (a0), v4
 # CHECK-INST: vsoxei64.v v24, (a0), v4
 # CHECK-ENCODING: [0x27,0x7c,0x45,0x0e]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 7c 45 0e <unknown>
+# CHECK-UNKNOWN: 0e457c27 <unknown>
 
 vs1r.v v24, (a0)
 # CHECK-INST: vs1r.v v24, (a0)
 # CHECK-ENCODING: [0x27,0x0c,0x85,0x02]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 0c 85 02 <unknown>
+# CHECK-UNKNOWN: 02850c27 <unknown>
 
 vs2r.v v24, (a0)
 # CHECK-INST: vs2r.v v24, (a0)
 # CHECK-ENCODING: [0x27,0x0c,0x85,0x22]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 0c 85 22 <unknown>
+# CHECK-UNKNOWN: 22850c27 <unknown>
 
 vs4r.v v24, (a0)
 # CHECK-INST: vs4r.v v24, (a0)
 # CHECK-ENCODING: [0x27,0x0c,0x85,0x62]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 0c 85 62 <unknown>
+# CHECK-UNKNOWN: 62850c27 <unknown>
 
 vs8r.v v24, (a0)
 # CHECK-INST: vs8r.v v24, (a0)
 # CHECK-ENCODING: [0x27,0x0c,0x85,0xe2]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 0c 85 e2 <unknown>
+# CHECK-UNKNOWN: e2850c27 <unknown>
 
 vsm.v v24, 0(a0)
 # CHECK-INST: vsm.v v24, (a0)
 # CHECK-ENCODING: [0x27,0x0c,0xb5,0x02]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 0c b5 02 <unknown>
+# CHECK-UNKNOWN: 02b50c27 <unknown>
 
 vse8.v v24, 0(a0), v0.t
 # CHECK-INST: vse8.v v24, (a0), v0.t
 # CHECK-ENCODING: [0x27,0x0c,0x05,0x00]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 0c 05 00 <unknown>
+# CHECK-UNKNOWN: 00050c27 <unknown>
 
 vsse16.v v24, 0(a0), a1, v0.t
 # CHECK-INST: vsse16.v v24, (a0), a1, v0.t
 # CHECK-ENCODING: [0x27,0x5c,0xb5,0x08]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 5c b5 08 <unknown>
+# CHECK-UNKNOWN: 08b55c27 <unknown>
 
 vsuxei8.v v24, 0(a0), v4, v0.t
 # CHECK-INST: vsuxei8.v v24, (a0), v4, v0.t
 # CHECK-ENCODING: [0x27,0x0c,0x45,0x04]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 0c 45 04 <unknown>
+# CHECK-UNKNOWN: 04450c27 <unknown>
 
 vsoxei32.v v24, 0(a0), v4, v0.t
 # CHECK-INST: vsoxei32.v v24, (a0), v4, v0.t
 # CHECK-ENCODING: [0x27,0x6c,0x45,0x0c]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 6c 45 0c <unknown>
+# CHECK-UNKNOWN: 0c456c27 <unknown>
diff --git a/llvm/test/MC/RISCV/rvv/sub.s b/llvm/test/MC/RISCV/rvv/sub.s
index 3cc75ae730e7..6a637d9207ee 100644
--- a/llvm/test/MC/RISCV/rvv/sub.s
+++ b/llvm/test/MC/RISCV/rvv/sub.s
@@ -12,298 +12,298 @@ vsub.vv v8, v4, v20, v0.t
 # CHECK-INST: vsub.vv v8, v4, v20, v0.t
 # CHECK-ENCODING: [0x57,0x04,0x4a,0x08]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 04 4a 08 <unknown>
+# CHECK-UNKNOWN: 084a0457 <unknown>
 
 vsub.vv v8, v4, v20
 # CHECK-INST: vsub.vv v8, v4, v20
 # CHECK-ENCODING: [0x57,0x04,0x4a,0x0a]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 04 4a 0a <unknown>
+# CHECK-UNKNOWN: 0a4a0457 <unknown>
 
 vsub.vx v8, v4, a0, v0.t
 # CHECK-INST: vsub.vx v8, v4, a0, v0.t
 # CHECK-ENCODING: [0x57,0x44,0x45,0x08]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 44 45 08 <unknown>
+# CHECK-UNKNOWN: 08454457 <unknown>
 
 vsub.vx v8, v4, a0
 # CHECK-INST: vsub.vx v8, v4, a0
 # CHECK-ENCODING: [0x57,0x44,0x45,0x0a]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 44 45 0a <unknown>
+# CHECK-UNKNOWN: 0a454457 <unknown>
 
 vrsub.vx v8, v4, a0, v0.t
 # CHECK-INST: vrsub.vx v8, v4, a0, v0.t
 # CHECK-ENCODING: [0x57,0x44,0x45,0x0c]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 44 45 0c <unknown>
+# CHECK-UNKNOWN: 0c454457 <unknown>
 
 vrsub.vx v8, v4, a0
 # CHECK-INST: vrsub.vx v8, v4, a0
 # CHECK-ENCODING: [0x57,0x44,0x45,0x0e]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 44 45 0e <unknown>
+# CHECK-UNKNOWN: 0e454457 <unknown>
 
 vrsub.vi v8, v4, 15, v0.t
 # CHECK-INST: vrsub.vi v8, v4, 15, v0.t
 # CHECK-ENCODING: [0x57,0xb4,0x47,0x0c]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 b4 47 0c <unknown>
+# CHECK-UNKNOWN: 0c47b457 <unknown>
 
 vrsub.vi v8, v4, 15
 # CHECK-INST: vrsub.vi v8, v4, 15
 # CHECK-ENCODING: [0x57,0xb4,0x47,0x0e]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 b4 47 0e <unknown>
+# CHECK-UNKNOWN: 0e47b457 <unknown>
 
 vwsubu.vv v8, v4, v20, v0.t
 # CHECK-INST: vwsubu.vv v8, v4, v20, v0.t
 # CHECK-ENCODING: [0x57,0x24,0x4a,0xc8]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 24 4a c8 <unknown>
+# CHECK-UNKNOWN: c84a2457 <unknown>
 
 vwsubu.vv v8, v4, v20
 # CHECK-INST: vwsubu.vv v8, v4, v20
 # CHECK-ENCODING: [0x57,0x24,0x4a,0xca]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 24 4a ca <unknown>
+# CHECK-UNKNOWN: ca4a2457 <unknown>
 
 vwsubu.vx v8, v4, a0, v0.t
 # CHECK-INST: vwsubu.vx v8, v4, a0, v0.t
 # CHECK-ENCODING: [0x57,0x64,0x45,0xc8]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 64 45 c8 <unknown>
+# CHECK-UNKNOWN: c8456457 <unknown>
 
 vwsubu.vx v8, v4, a0
 # CHECK-INST: vwsubu.vx v8, v4, a0
 # CHECK-ENCODING: [0x57,0x64,0x45,0xca]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 64 45 ca <unknown>
+# CHECK-UNKNOWN: ca456457 <unknown>
 
 vwsub.vv v8, v4, v20, v0.t
 # CHECK-INST: vwsub.vv v8, v4, v20, v0.t
 # CHECK-ENCODING: [0x57,0x24,0x4a,0xcc]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 24 4a cc <unknown>
+# CHECK-UNKNOWN: cc4a2457 <unknown>
 
 vwsub.vv v8, v4, v20
 # CHECK-INST: vwsub.vv v8, v4, v20
 # CHECK-ENCODING: [0x57,0x24,0x4a,0xce]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 24 4a ce <unknown>
+# CHECK-UNKNOWN: ce4a2457 <unknown>
 
 vwsub.vx v8, v4, a0, v0.t
 # CHECK-INST: vwsub.vx v8, v4, a0, v0.t
 # CHECK-ENCODING: [0x57,0x64,0x45,0xcc]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 64 45 cc <unknown>
+# CHECK-UNKNOWN: cc456457 <unknown>
 
 vwsub.vx v8, v4, a0
 # CHECK-INST: vwsub.vx v8, v4, a0
 # CHECK-ENCODING: [0x57,0x64,0x45,0xce]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 64 45 ce <unknown>
+# CHECK-UNKNOWN: ce456457 <unknown>
 
 vwsubu.wv v8, v4, v20, v0.t
 # CHECK-INST: vwsubu.wv v8, v4, v20, v0.t
 # CHECK-ENCODING: [0x57,0x24,0x4a,0xd8]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 24 4a d8 <unknown>
+# CHECK-UNKNOWN: d84a2457 <unknown>
 
 vwsubu.wv v8, v4, v20
 # CHECK-INST: vwsubu.wv v8, v4, v20
 # CHECK-ENCODING: [0x57,0x24,0x4a,0xda]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 24 4a da <unknown>
+# CHECK-UNKNOWN: da4a2457 <unknown>
 
 vwsubu.wx v8, v4, a0, v0.t
 # CHECK-INST: vwsubu.wx v8, v4, a0, v0.t
 # CHECK-ENCODING: [0x57,0x64,0x45,0xd8]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 64 45 d8 <unknown>
+# CHECK-UNKNOWN: d8456457 <unknown>
 
 vwsubu.wx v8, v4, a0
 # CHECK-INST: vwsubu.wx v8, v4, a0
 # CHECK-ENCODING: [0x57,0x64,0x45,0xda]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 64 45 da <unknown>
+# CHECK-UNKNOWN: da456457 <unknown>
 
 vwsub.wv v8, v4, v20, v0.t
 # CHECK-INST: vwsub.wv v8, v4, v20, v0.t
 # CHECK-ENCODING: [0x57,0x24,0x4a,0xdc]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 24 4a dc <unknown>
+# CHECK-UNKNOWN: dc4a2457 <unknown>
 
 vwsub.wv v8, v4, v20
 # CHECK-INST: vwsub.wv v8, v4, v20
 # CHECK-ENCODING: [0x57,0x24,0x4a,0xde]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 24 4a de <unknown>
+# CHECK-UNKNOWN: de4a2457 <unknown>
 
 vwsub.wx v8, v4, a0, v0.t
 # CHECK-INST: vwsub.wx v8, v4, a0, v0.t
 # CHECK-ENCODING: [0x57,0x64,0x45,0xdc]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 64 45 dc <unknown>
+# CHECK-UNKNOWN: dc456457 <unknown>
 
 vwsub.wx v8, v4, a0
 # CHECK-INST: vwsub.wx v8, v4, a0
 # CHECK-ENCODING: [0x57,0x64,0x45,0xde]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 64 45 de <unknown>
+# CHECK-UNKNOWN: de456457 <unknown>
 
 vsbc.vvm v8, v4, v20, v0
 # CHECK-INST: vsbc.vvm v8, v4, v20, v0
 # CHECK-ENCODING: [0x57,0x04,0x4a,0x48]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 04 4a 48 <unknown>
+# CHECK-UNKNOWN: 484a0457 <unknown>
 
 vsbc.vvm v4, v4, v20, v0
 # CHECK-INST: vsbc.vvm v4, v4, v20, v0
 # CHECK-ENCODING: [0x57,0x02,0x4a,0x48]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 02 4a 48 <unknown>
+# CHECK-UNKNOWN: 484a0257 <unknown>
 
 vsbc.vvm v8, v4, v8, v0
 # CHECK-INST: vsbc.vvm v8, v4, v8, v0
 # CHECK-ENCODING: [0x57,0x04,0x44,0x48]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 04 44 48 <unknown>
+# CHECK-UNKNOWN: 48440457 <unknown>
 
 vsbc.vxm v8, v4, a0, v0
 # CHECK-INST: vsbc.vxm v8, v4, a0, v0
 # CHECK-ENCODING: [0x57,0x44,0x45,0x48]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 44 45 48 <unknown>
+# CHECK-UNKNOWN: 48454457 <unknown>
 
 vmsbc.vvm v8, v4, v20, v0
 # CHECK-INST: vmsbc.vvm v8, v4, v20, v0
 # CHECK-ENCODING: [0x57,0x04,0x4a,0x4c]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 04 4a 4c <unknown>
+# CHECK-UNKNOWN: 4c4a0457 <unknown>
 
 vmsbc.vvm v4, v4, v20, v0
 # CHECK-INST: vmsbc.vvm v4, v4, v20, v0
 # CHECK-ENCODING: [0x57,0x02,0x4a,0x4c]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 02 4a 4c <unknown>
+# CHECK-UNKNOWN: 4c4a0257 <unknown>
 
 vmsbc.vvm v8, v4, v8, v0
 # CHECK-INST: vmsbc.vvm v8, v4, v8, v0
 # CHECK-ENCODING: [0x57,0x04,0x44,0x4c]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 04 44 4c <unknown>
+# CHECK-UNKNOWN: 4c440457 <unknown>
 
 vmsbc.vxm v8, v4, a0, v0
 # CHECK-INST: vmsbc.vxm v8, v4, a0, v0
 # CHECK-ENCODING: [0x57,0x44,0x45,0x4c]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 44 45 4c <unknown>
+# CHECK-UNKNOWN: 4c454457 <unknown>
 
 vmsbc.vv v8, v4, v20
 # CHECK-INST: vmsbc.vv v8, v4, v20
 # CHECK-ENCODING: [0x57,0x04,0x4a,0x4e]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 04 4a 4e <unknown>
+# CHECK-UNKNOWN: 4e4a0457 <unknown>
 
 vmsbc.vx v8, v4, a0
 # CHECK-INST: vmsbc.vx v8, v4, a0
 # CHECK-ENCODING: [0x57,0x44,0x45,0x4e]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 44 45 4e <unknown>
+# CHECK-UNKNOWN: 4e454457 <unknown>
 
 vssubu.vv v8, v4, v20, v0.t
 # CHECK-INST: vssubu.vv v8, v4, v20, v0.t
 # CHECK-ENCODING: [0x57,0x04,0x4a,0x88]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 04 4a 88 <unknown>
+# CHECK-UNKNOWN: 884a0457 <unknown>
 
 vssubu.vv v8, v4, v20
 # CHECK-INST: vssubu.vv v8, v4, v20
 # CHECK-ENCODING: [0x57,0x04,0x4a,0x8a]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 04 4a 8a <unknown>
+# CHECK-UNKNOWN: 8a4a0457 <unknown>
 
 vssubu.vx v8, v4, a0, v0.t
 # CHECK-INST: vssubu.vx v8, v4, a0, v0.t
 # CHECK-ENCODING: [0x57,0x44,0x45,0x88]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 44 45 88 <unknown>
+# CHECK-UNKNOWN: 88454457 <unknown>
 
 vssubu.vx v8, v4, a0
 # CHECK-INST: vssubu.vx v8, v4, a0
 # CHECK-ENCODING: [0x57,0x44,0x45,0x8a]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 44 45 8a <unknown>
+# CHECK-UNKNOWN: 8a454457 <unknown>
 
 vssub.vv v8, v4, v20, v0.t
 # CHECK-INST: vssub.vv v8, v4, v20, v0.t
 # CHECK-ENCODING: [0x57,0x04,0x4a,0x8c]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 04 4a 8c <unknown>
+# CHECK-UNKNOWN: 8c4a0457 <unknown>
 
 vssub.vv v8, v4, v20
 # CHECK-INST: vssub.vv v8, v4, v20
 # CHECK-ENCODING: [0x57,0x04,0x4a,0x8e]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 04 4a 8e <unknown>
+# CHECK-UNKNOWN: 8e4a0457 <unknown>
 
 vssub.vx v8, v4, a0, v0.t
 # CHECK-INST: vssub.vx v8, v4, a0, v0.t
 # CHECK-ENCODING: [0x57,0x44,0x45,0x8c]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 44 45 8c <unknown>
+# CHECK-UNKNOWN: 8c454457 <unknown>
 
 vssub.vx v8, v4, a0
 # CHECK-INST: vssub.vx v8, v4, a0
 # CHECK-ENCODING: [0x57,0x44,0x45,0x8e]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 44 45 8e <unknown>
+# CHECK-UNKNOWN: 8e454457 <unknown>
 
 vasub.vv v8, v4, v20, v0.t
 # CHECK-INST: vasub.vv v8, v4, v20, v0.t
 # CHECK-ENCODING: [0x57,0x24,0x4a,0x2c]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 24 4a 2c <unknown>
+# CHECK-UNKNOWN: 2c4a2457 <unknown>
 
 vasub.vv v8, v4, v20
 # CHECK-INST: vasub.vv v8, v4, v20
 # CHECK-ENCODING: [0x57,0x24,0x4a,0x2e]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 24 4a 2e <unknown>
+# CHECK-UNKNOWN: 2e4a2457 <unknown>
 
 vasub.vx v8, v4, a0, v0.t
 # CHECK-INST: vasub.vx v8, v4, a0, v0.t
 # CHECK-ENCODING: [0x57,0x64,0x45,0x2c]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 64 45 2c <unknown>
+# CHECK-UNKNOWN: 2c456457 <unknown>
 
 vasub.vx v8, v4, a0
 # CHECK-INST: vasub.vx v8, v4, a0
 # CHECK-ENCODING: [0x57,0x64,0x45,0x2e]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 64 45 2e <unknown>
+# CHECK-UNKNOWN: 2e456457 <unknown>
 
 vasubu.vv v8, v4, v20, v0.t
 # CHECK-INST: vasubu.vv v8, v4, v20, v0.t
 # CHECK-ENCODING: [0x57,0x24,0x4a,0x28]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 24 4a 28 <unknown>
+# CHECK-UNKNOWN: 284a2457 <unknown>
 
 vasubu.vv v8, v4, v20
 # CHECK-INST: vasubu.vv v8, v4, v20
 # CHECK-ENCODING: [0x57,0x24,0x4a,0x2a]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 24 4a 2a <unknown>
+# CHECK-UNKNOWN: 2a4a2457 <unknown>
 
 vasubu.vx v8, v4, a0, v0.t
 # CHECK-INST: vasubu.vx v8, v4, a0, v0.t
 # CHECK-ENCODING: [0x57,0x64,0x45,0x28]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 64 45 28 <unknown>
+# CHECK-UNKNOWN: 28456457 <unknown>
 
 vasubu.vx v8, v4, a0
 # CHECK-INST: vasubu.vx v8, v4, a0
 # CHECK-ENCODING: [0x57,0x64,0x45,0x2a]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 64 45 2a <unknown>
+# CHECK-UNKNOWN: 2a456457 <unknown>
diff --git a/llvm/test/MC/RISCV/rvv/vsetvl.s b/llvm/test/MC/RISCV/rvv/vsetvl.s
index 69a48d24f190..c9197d8917a4 100644
--- a/llvm/test/MC/RISCV/rvv/vsetvl.s
+++ b/llvm/test/MC/RISCV/rvv/vsetvl.s
@@ -13,149 +13,149 @@ vsetvli a2, a0, 0x224
 # CHECK-INST: vsetvli a2, a0, 548
 # CHECK-ENCODING: [0x57,0x76,0x45,0x22]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 76 45 22 <unknown>
+# CHECK-UNKNOWN: 22457657 <unknown>
 
 vsetvli a2, a0, 0xd0
 # CHECK-INST: vsetvli a2, a0, e32, m1, ta, ma
 # CHECK-ENCODING: [0x57,0x76,0x05,0x0d]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 76 05 0d <unknown>
+# CHECK-UNKNOWN: 0d057657 <unknown>
 
 vsetvli a2, a0, 0xd1
 # CHECK-INST: vsetvli a2, a0, e32, m2, ta, ma
 # CHECK-ENCODING: [0x57,0x76,0x15,0x0d]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 76 15 0d <unknown>
+# CHECK-UNKNOWN: 0d157657 <unknown>
 
 vsetvli a2, a0, 0x50
 # CHECK-INST: vsetvli a2, a0, e32, m1, ta, mu
 # CHECK-ENCODING: [0x57,0x76,0x05,0x05]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 76 05 05 <unknown>
+# CHECK-UNKNOWN: 05057657 <unknown>
 
 vsetvli a2, a0, 0x90
 # CHECK-INST: vsetvli a2, a0, e32, m1, tu, ma
 # CHECK-ENCODING: [0x57,0x76,0x05,0x09]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 76 05 09 <unknown>
+# CHECK-UNKNOWN: 09057657 <unknown>
 
 vsetvli a2, a0, 144
 # CHECK-INST: vsetvli a2, a0, e32, m1, tu, ma
 # CHECK-ENCODING: [0x57,0x76,0x05,0x09]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 76 05 09 <unknown>
+# CHECK-UNKNOWN: 09057657 <unknown>
 
 vsetvli a2, a0, e32, m1, ta, ma
 # CHECK-INST: vsetvli a2, a0, e32,  m1,  ta,  ma
 # CHECK-ENCODING: [0x57,0x76,0x05,0x0d]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 76 05 0d <unknown>
+# CHECK-UNKNOWN: 0d057657 <unknown>
 
 vsetvli a2, a0, e32, m2, ta, ma
 # CHECK-INST: vsetvli a2, a0, e32, m2, ta, ma
 # CHECK-ENCODING: [0x57,0x76,0x15,0x0d]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 76 15 0d <unknown>
+# CHECK-UNKNOWN: 0d157657 <unknown>
 
 vsetvli a2, a0, e32, m4, ta, ma
 # CHECK-INST: vsetvli a2, a0, e32, m4, ta, ma
 # CHECK-ENCODING: [0x57,0x76,0x25,0x0d]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 76 25 0d <unknown>
+# CHECK-UNKNOWN: 0d257657 <unknown>
 
 vsetvli a2, a0, e32, m8, ta, ma
 # CHECK-INST: vsetvli a2, a0, e32, m8, ta, ma
 # CHECK-ENCODING: [0x57,0x76,0x35,0x0d]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 76 35 0d <unknown>
+# CHECK-UNKNOWN: 0d357657 <unknown>
 
 vsetvli a2, a0, e32, mf2, ta, ma
 # CHECK-INST: vsetvli a2, a0, e32, mf2, ta, ma
 # CHECK-ENCODING: [0x57,0x76,0x75,0x0d]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 76 75 0d <unknown>
+# CHECK-UNKNOWN: 0d757657 <unknown>
 
 vsetvli a2, a0, e32, mf4, ta, ma
 # CHECK-INST: vsetvli a2, a0, e32, mf4, ta, ma
 # CHECK-ENCODING: [0x57,0x76,0x65,0x0d]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 76 65 0d <unknown>
+# CHECK-UNKNOWN: 0d657657 <unknown>
 
 vsetvli a2, a0, e32, mf8, ta, ma
 # CHECK-INST: vsetvli a2, a0, e32, mf8, ta, ma
 # CHECK-ENCODING: [0x57,0x76,0x55,0x0d]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 76 55 0d <unknown>
+# CHECK-UNKNOWN: 0d557657 <unknown>
 
 vsetvli a2, a0, e32, m1, ta, ma
 # CHECK-INST: vsetvli a2, a0, e32, m1, ta, ma
 # CHECK-ENCODING: [0x57,0x76,0x05,0x0d]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 76 05 0d <unknown>
+# CHECK-UNKNOWN: 0d057657 <unknown>
 
 vsetvli a2, a0, e32, m1, tu, ma
 # CHECK-INST: vsetvli a2, a0, e32, m1, tu, ma
 # CHECK-ENCODING: [0x57,0x76,0x05,0x09]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 76 05 09 <unknown>
+# CHECK-UNKNOWN: 09057657 <unknown>
 
 vsetvli a2, a0, e32, m1, ta, mu
 # CHECK-INST: vsetvli a2, a0, e32, m1, ta, mu
 # CHECK-ENCODING: [0x57,0x76,0x05,0x05]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 76 05 05 <unknown>
+# CHECK-UNKNOWN: 05057657 <unknown>
 
 vsetvli a2, a0, e32, m1, tu, mu
 # CHECK-INST: vsetvli a2, a0, e32, m1
 # CHECK-ENCODING: [0x57,0x76,0x05,0x01]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 76 05 01 <unknown>
+# CHECK-UNKNOWN: 01057657 <unknown>
 
 vsetvl a2, a0, a1
 # CHECK-INST: vsetvl a2, a0, a1
 # CHECK-ENCODING: [0x57,0x76,0xb5,0x80]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 76 b5 80 <unknown>
+# CHECK-UNKNOWN: 80b57657 <unknown>
 
 # reserved filed: vlmul[2:0]=4, vsew[2:0]=0b1xx, non-zero bits 8/9/10.
 vsetivli a2, 0, 0x224
 # CHECK-INST: vsetivli a2, 0, 548
 # CHECK-ENCODING: [0x57,0x76,0x40,0xe2]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 76 40 e2 <unknown>
+# CHECK-UNKNOWN: e2407657 <unknown>
 
 vsetivli a2, 0, 0xd0
 # CHECK-INST: vsetivli a2, 0, e32, m1, ta, ma
 # CHECK-ENCODING: [0x57,0x76,0x00,0xcd]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 76 00 cd <unknown>
+# CHECK-UNKNOWN: cd007657 <unknown>
 
 vsetivli a2, 15, 0xd0
 # CHECK-INST: vsetivli a2, 15, e32, m1, ta, ma
 # CHECK-ENCODING: [0x57,0xf6,0x07,0xcd]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 f6 07 cd <unknown>
+# CHECK-UNKNOWN: cd07f657 <unknown>
 
 vsetivli a2, 15, 208
 # CHECK-INST: vsetivli a2, 15, e32, m1, ta, ma
 # CHECK-ENCODING: [0x57,0xf6,0x07,0xcd]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 f6 07 cd <unknown>
+# CHECK-UNKNOWN: cd07f657 <unknown>
 
 vsetivli a2, 0, e32, m1, ta, ma
 # CHECK-INST: vsetivli a2, 0, e32, m1, ta, ma
 # CHECK-ENCODING: [0x57,0x76,0x00,0xcd]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 76 00 cd <unknown>
+# CHECK-UNKNOWN: cd007657 <unknown>
 
 vsetivli a2, 15, e32, m1, ta, ma
 # CHECK-INST: vsetivli a2, 15, e32, m1, ta, ma
 # CHECK-ENCODING: [0x57,0xf6,0x07,0xcd]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 f6 07 cd <unknown>
+# CHECK-UNKNOWN: cd07f657 <unknown>
 
 vsetivli a2, 31, e32, m1, ta, ma
 # CHECK-INST: vsetivli a2, 31, e32, m1, ta, ma
 # CHECK-ENCODING: [0x57,0xf6,0x0f,0xcd]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 f6 0f cd <unknown>
+# CHECK-UNKNOWN: cd0ff657 <unknown>
diff --git a/llvm/test/MC/RISCV/rvv/xor.s b/llvm/test/MC/RISCV/rvv/xor.s
index 5ea0f694e0d6..572388ed2267 100644
--- a/llvm/test/MC/RISCV/rvv/xor.s
+++ b/llvm/test/MC/RISCV/rvv/xor.s
@@ -12,46 +12,46 @@ vxor.vv v8, v4, v20, v0.t
 # CHECK-INST: vxor.vv v8, v4, v20, v0.t
 # CHECK-ENCODING: [0x57,0x04,0x4a,0x2c]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 04 4a 2c <unknown>
+# CHECK-UNKNOWN: 2c4a0457 <unknown>
 
 vxor.vv v8, v4, v20
 # CHECK-INST: vxor.vv v8, v4, v20
 # CHECK-ENCODING: [0x57,0x04,0x4a,0x2e]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 04 4a 2e <unknown>
+# CHECK-UNKNOWN: 2e4a0457 <unknown>
 
 vxor.vx v8, v4, a0, v0.t
 # CHECK-INST: vxor.vx v8, v4, a0, v0.t
 # CHECK-ENCODING: [0x57,0x44,0x45,0x2c]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 44 45 2c <unknown>
+# CHECK-UNKNOWN: 2c454457 <unknown>
 
 vxor.vx v8, v4, a0
 # CHECK-INST: vxor.vx v8, v4, a0
 # CHECK-ENCODING: [0x57,0x44,0x45,0x2e]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 44 45 2e <unknown>
+# CHECK-UNKNOWN: 2e454457 <unknown>
 
 vxor.vi v8, v4, 15, v0.t
 # CHECK-INST: vxor.vi v8, v4, 15, v0.t
 # CHECK-ENCODING: [0x57,0xb4,0x47,0x2c]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 b4 47 2c <unknown>
+# CHECK-UNKNOWN: 2c47b457 <unknown>
 
 vxor.vi v8, v4, 15
 # CHECK-INST: vxor.vi v8, v4, 15
 # CHECK-ENCODING: [0x57,0xb4,0x47,0x2e]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 b4 47 2e <unknown>
+# CHECK-UNKNOWN: 2e47b457 <unknown>
 
 vnot.v v8, v4, v0.t
 # CHECK-INST: vnot.v v8, v4, v0.t
 # CHECK-ENCODING: [0x57,0xb4,0x4f,0x2c]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 b4 4f 2c <unknown>
+# CHECK-UNKNOWN: 2c4fb457 <unknown>
 
 vnot.v v8, v4
 # CHECK-INST: vnot.v v8, v4
 # CHECK-ENCODING: [0x57,0xb4,0x4f,0x2e]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 57 b4 4f 2e <unknown>
+# CHECK-UNKNOWN: 2e4fb457 <unknown>
diff --git a/llvm/test/MC/RISCV/rvv/xsfvcp.s b/llvm/test/MC/RISCV/rvv/xsfvcp.s
index a137311f575a..4298bf7a7b7a 100644
--- a/llvm/test/MC/RISCV/rvv/xsfvcp.s
+++ b/llvm/test/MC/RISCV/rvv/xsfvcp.s
@@ -21,166 +21,166 @@ sf.vc.x 0x3, 0xf, 0x1f, a1
 # CHECK-INST: sf.vc.x 3, 15, 31, a1
 # CHECK-ENCODING: [0xdb,0xcf,0xf5,0x0e]
 # CHECK-ERROR: instruction requires the following: 'XSfvcp' (SiFive Custom Vector Coprocessor Interface Instructions){{$}}
-# CHECK-UNKNOWN: db cf f5 0e <unknown>
+# CHECK-UNKNOWN: 0ef5cfdb <unknown>
 
 sf.vc.i 0x3, 0xf, 0x1f, 15
 # CHECK-INST: sf.vc.i 3, 15, 31, 15
 # CHECK-ENCODING: [0xdb,0xbf,0xf7,0x0e]
 # CHECK-ERROR: instruction requires the following: 'XSfvcp' (SiFive Custom Vector Coprocessor Interface Instructions){{$}}
-# CHECK-UNKNOWN: db bf f7 0e <unknown>
+# CHECK-UNKNOWN: 0ef7bfdb <unknown>
 
 sf.vc.vv 0x3, 0x1f, v2, v1
 # CHECK-INST: sf.vc.vv 3, 31, v2, v1
 # CHECK-ENCODING: [0xdb,0x8f,0x20,0x2e]
 # CHECK-ERROR: instruction requires the following: 'XSfvcp' (SiFive Custom Vector Coprocessor Interface Instructions){{$}}
-# CHECK-UNKNOWN: db 8f 20 2e <unknown>
+# CHECK-UNKNOWN: 2e208fdb <unknown>
 
 sf.vc.xv 0x3, 0x1f, v2, a1
 # CHECK-INST: sf.vc.xv 3, 31, v2, a1
 # CHECK-ENCODING: [0xdb,0xcf,0x25,0x2e]
 # CHECK-ERROR: instruction requires the following: 'XSfvcp' (SiFive Custom Vector Coprocessor Interface Instructions){{$}}
-# CHECK-UNKNOWN: db cf 25 2e <unknown>
+# CHECK-UNKNOWN: 2e25cfdb <unknown>
 
 sf.vc.iv 0x3, 0x1f, v2, 15
 # CHECK-INST: sf.vc.iv 3, 31, v2, 15
 # CHECK-ENCODING: [0xdb,0xbf,0x27,0x2e]
 # CHECK-ERROR: instruction requires the following: 'XSfvcp' (SiFive Custom Vector Coprocessor Interface Instructions){{$}}
-# CHECK-UNKNOWN: db bf 27 2e <unknown>
+# CHECK-UNKNOWN: 2e27bfdb <unknown>
 
 sf.vc.fv 0x1, 0x1f, v2, fa1
 # CHECK-INST: sf.vc.fv 1, 31, v2, fa1
 # CHECK-ENCODING: [0xdb,0xdf,0x25,0x2e]
 # CHECK-ERROR: instruction requires the following: 'XSfvcp' (SiFive Custom Vector Coprocessor Interface Instructions){{$}}
-# CHECK-UNKNOWN: db df 25 2e <unknown>
+# CHECK-UNKNOWN: 2e25dfdb <unknown>
 
 sf.vc.vvv 0x3, v0, v2, v1
 # CHECK-INST: sf.vc.vvv 3, v0, v2, v1
 # CHECK-ENCODING: [0x5b,0x80,0x20,0xae]
 # CHECK-ERROR: instruction requires the following: 'XSfvcp' (SiFive Custom Vector Coprocessor Interface Instructions){{$}}
-# CHECK-UNKNOWN: 5b 80 20 ae <unknown>
+# CHECK-UNKNOWN: ae20805b <unknown>
 
 sf.vc.xvv 0x3, v0, v2, a1
 # CHECK-INST: sf.vc.xvv 3, v0, v2, a1
 # CHECK-ENCODING: [0x5b,0xc0,0x25,0xae]
 # CHECK-ERROR: instruction requires the following: 'XSfvcp' (SiFive Custom Vector Coprocessor Interface Instructions){{$}}
-# CHECK-UNKNOWN: 5b c0 25 ae <unknown>
+# CHECK-UNKNOWN: ae25c05b <unknown>
 
 sf.vc.ivv 0x3, v0, v2, 15
 # CHECK-INST: sf.vc.ivv 3, v0, v2, 15
 # CHECK-ENCODING: [0x5b,0xb0,0x27,0xae]
 # CHECK-ERROR: instruction requires the following: 'XSfvcp' (SiFive Custom Vector Coprocessor Interface Instructions){{$}}
-# CHECK-UNKNOWN: 5b b0 27 ae <unknown>
+# CHECK-UNKNOWN: ae27b05b <unknown>
 
 sf.vc.fvv 0x1, v0, v2, fa1
 # CHECK-INST: sf.vc.fvv 1, v0, v2, fa1
 # CHECK-ENCODING: [0x5b,0xd0,0x25,0xae]
 # CHECK-ERROR: instruction requires the following: 'XSfvcp' (SiFive Custom Vector Coprocessor Interface Instructions){{$}}
-# CHECK-UNKNOWN: 5b d0 25 ae <unknown>
+# CHECK-UNKNOWN: ae25d05b <unknown>
 
 sf.vc.vvw 0x3, v0, v2, v1
 # CHECK-INST: sf.vc.vvw 3, v0, v2, v1
 # CHECK-ENCODING: [0x5b,0x80,0x20,0xfe]
 # CHECK-ERROR: instruction requires the following: 'XSfvcp' (SiFive Custom Vector Coprocessor Interface Instructions){{$}}
-# CHECK-UNKNOWN: 5b 80 20 fe <unknown>
+# CHECK-UNKNOWN: fe20805b <unknown>
 
 sf.vc.xvw 0x3, v0, v2, a1
 # CHECK-INST: sf.vc.xvw 3, v0, v2, a1
 # CHECK-ENCODING: [0x5b,0xc0,0x25,0xfe]
 # CHECK-ERROR: instruction requires the following: 'XSfvcp' (SiFive Custom Vector Coprocessor Interface Instructions){{$}}
-# CHECK-UNKNOWN: 5b c0 25 fe <unknown>
+# CHECK-UNKNOWN: fe25c05b <unknown>
 
 sf.vc.ivw 0x3, v0, v2, 15
 # CHECK-INST: sf.vc.ivw 3, v0, v2, 15
 # CHECK-ENCODING: [0x5b,0xb0,0x27,0xfe]
 # CHECK-ERROR: instruction requires the following: 'XSfvcp' (SiFive Custom Vector Coprocessor Interface Instructions){{$}}
-# CHECK-UNKNOWN: 5b b0 27 fe <unknown>
+# CHECK-UNKNOWN: fe27b05b <unknown>
 
 sf.vc.fvw 0x1, v0, v2, fa1
 # CHECK-INST: sf.vc.fvw 1, v0, v2, fa1
 # CHECK-ENCODING: [0x5b,0xd0,0x25,0xfe]
 # CHECK-ERROR: instruction requires the following: 'XSfvcp' (SiFive Custom Vector Coprocessor Interface Instructions){{$}}
-# CHECK-UNKNOWN: 5b d0 25 fe <unknown>
+# CHECK-UNKNOWN: fe25d05b <unknown>
 
 sf.vc.v.x 0x3, 0xf, v0, a1
 # CHECK-INST: sf.vc.v.x 3, 15, v0, a1
 # CHECK-ENCODING: [0x5b,0xc0,0xf5,0x0c]
 # CHECK-ERROR: instruction requires the following: 'XSfvcp' (SiFive Custom Vector Coprocessor Interface Instructions){{$}}
-# CHECK-UNKNOWN: 5b c0 f5 0c <unknown>
+# CHECK-UNKNOWN: 0cf5c05b <unknown>
 
 sf.vc.v.i 0x3, 0xf, v0, 15
 # CHECK-INST: sf.vc.v.i 3, 15, v0, 15
 # CHECK-ENCODING: [0x5b,0xb0,0xf7,0x0c]
 # CHECK-ERROR: instruction requires the following: 'XSfvcp' (SiFive Custom Vector Coprocessor Interface Instructions){{$}}
-# CHECK-UNKNOWN: 5b b0 f7 0c <unknown>
+# CHECK-UNKNOWN: 0cf7b05b <unknown>
 
 sf.vc.v.vv 0x3, v0, v2, v1
 # CHECK-INST: sf.vc.v.vv 3, v0, v2, v1
 # CHECK-ENCODING: [0x5b,0x80,0x20,0x2c]
 # CHECK-ERROR: instruction requires the following: 'XSfvcp' (SiFive Custom Vector Coprocessor Interface Instructions){{$}}
-# CHECK-UNKNOWN: 5b 80 20 2c <unknown>
+# CHECK-UNKNOWN: 2c20805b <unknown>
 
 sf.vc.v.xv 0x3, v0, v2, a1
 # CHECK-INST: sf.vc.v.xv 3, v0, v2, a1
 # CHECK-ENCODING: [0x5b,0xc0,0x25,0x2c]
 # CHECK-ERROR: instruction requires the following: 'XSfvcp' (SiFive Custom Vector Coprocessor Interface Instructions){{$}}
-# CHECK-UNKNOWN: 5b c0 25 2c <unknown>
+# CHECK-UNKNOWN: 2c25c05b <unknown>
 
 sf.vc.v.iv 0x3, v0, v2, 15
 # CHECK-INST: sf.vc.v.iv 3, v0, v2, 15
 # CHECK-ENCODING: [0x5b,0xb0,0x27,0x2c]
 # CHECK-ERROR: instruction requires the following: 'XSfvcp' (SiFive Custom Vector Coprocessor Interface Instructions){{$}}
-# CHECK-UNKNOWN: 5b b0 27 2c <unknown>
+# CHECK-UNKNOWN: 2c27b05b <unknown>
 
 sf.vc.v.fv 0x1, v0, v2, fa1
 # CHECK-INST: sf.vc.v.fv 1, v0, v2, fa1
 # CHECK-ENCODING: [0x5b,0xd0,0x25,0x2c]
 # CHECK-ERROR: instruction requires the following: 'XSfvcp' (SiFive Custom Vector Coprocessor Interface Instructions){{$}}
-# CHECK-UNKNOWN: 5b d0 25 2c <unknown>
+# CHECK-UNKNOWN: 2c25d05b <unknown>
 
 sf.vc.v.vvv 0x3, v0, v2, v1
 # CHECK-INST: sf.vc.v.vvv 3, v0, v2, v1
 # CHECK-ENCODING: [0x5b,0x80,0x20,0xac]
 # CHECK-ERROR: instruction requires the following: 'XSfvcp' (SiFive Custom Vector Coprocessor Interface Instructions){{$}}
-# CHECK-UNKNOWN: 5b 80 20 ac <unknown>
+# CHECK-UNKNOWN: ac20805b <unknown>
 
 sf.vc.v.xvv 0x3, v0, v2, a1
 # CHECK-INST: sf.vc.v.xvv 3, v0, v2, a1
 # CHECK-ENCODING: [0x5b,0xc0,0x25,0xac]
 # CHECK-ERROR: instruction requires the following: 'XSfvcp' (SiFive Custom Vector Coprocessor Interface Instructions){{$}}
-# CHECK-UNKNOWN: 5b c0 25 ac <unknown>
+# CHECK-UNKNOWN: ac25c05b <unknown>
 
 sf.vc.v.ivv 0x3, v0, v2, 15
 # CHECK-INST: sf.vc.v.ivv 3, v0, v2, 15
 # CHECK-ENCODING: [0x5b,0xb0,0x27,0xac]
 # CHECK-ERROR: instruction requires the following: 'XSfvcp' (SiFive Custom Vector Coprocessor Interface Instructions){{$}}
-# CHECK-UNKNOWN: 5b b0 27 ac <unknown>
+# CHECK-UNKNOWN: ac27b05b <unknown>
 
 sf.vc.v.fvv 0x1, v0, v2, fa1
 # CHECK-INST: sf.vc.v.fvv 1, v0, v2, fa1
 # CHECK-ENCODING: [0x5b,0xd0,0x25,0xac]
 # CHECK-ERROR: instruction requires the following: 'XSfvcp' (SiFive Custom Vector Coprocessor Interface Instructions){{$}}
-# CHECK-UNKNOWN: 5b d0 25 ac <unknown>
+# CHECK-UNKNOWN: ac25d05b <unknown>
 
 sf.vc.v.vvw 0x3, v0, v2, v1
 # CHECK-INST: sf.vc.v.vvw 3, v0, v2, v1
 # CHECK-ENCODING: [0x5b,0x80,0x20,0xfc]
 # CHECK-ERROR: instruction requires the following: 'XSfvcp' (SiFive Custom Vector Coprocessor Interface Instructions){{$}}
-# CHECK-UNKNOWN: 5b 80 20 fc <unknown>
+# CHECK-UNKNOWN: fc20805b <unknown>
 
 sf.vc.v.xvw 0x3, v0, v2, a1
 # CHECK-INST: sf.vc.v.xvw 3, v0, v2, a1
 # CHECK-ENCODING: [0x5b,0xc0,0x25,0xfc]
 # CHECK-ERROR: instruction requires the following: 'XSfvcp' (SiFive Custom Vector Coprocessor Interface Instructions){{$}}
-# CHECK-UNKNOWN: 5b c0 25 fc <unknown>
+# CHECK-UNKNOWN: fc25c05b <unknown>
 
 sf.vc.v.ivw 0x3, v0, v2, 15
 # CHECK-INST: sf.vc.v.ivw 3, v0, v2, 15
 # CHECK-ENCODING: [0x5b,0xb0,0x27,0xfc]
 # CHECK-ERROR: instruction requires the following: 'XSfvcp' (SiFive Custom Vector Coprocessor Interface Instructions){{$}}
-# CHECK-UNKNOWN: 5b b0 27 fc <unknown>
+# CHECK-UNKNOWN: fc27b05b <unknown>
 
 sf.vc.v.fvw 0x1, v0, v2, fa1
 # CHECK-INST: sf.vc.v.fvw 1, v0, v2, fa1
 # CHECK-ENCODING: [0x5b,0xd0,0x25,0xfc]
 # CHECK-ERROR: instruction requires the following: 'XSfvcp' (SiFive Custom Vector Coprocessor Interface Instructions){{$}}
-# CHECK-UNKNOWN: 5b d0 25 fc <unknown>
+# CHECK-UNKNOWN: fc25d05b <unknown>
diff --git a/llvm/test/MC/RISCV/rvv/xsfvfnrclip.s b/llvm/test/MC/RISCV/rvv/xsfvfnrclip.s
index d8b184659ac4..7508d44bc916 100644
--- a/llvm/test/MC/RISCV/rvv/xsfvfnrclip.s
+++ b/llvm/test/MC/RISCV/rvv/xsfvfnrclip.s
@@ -12,22 +12,22 @@ sf.vfnrclip.xu.f.qf v4, v8, fa2
 # CHECK-INST: sf.vfnrclip.xu.f.qf v4, v8, fa2
 # CHECK-ENCODING: [0x5b,0x52,0x86,0x8a]
 # CHECK-ERROR: instruction requires the following: 'XSfvfnrclipxfqf' (SiFive FP32-to-int8 Ranged Clip Instructions)
-# CHECK-UNKNOWN: 5b 52 86 8a <unknown>
+# CHECK-UNKNOWN: 8a86525b <unknown>
 
 sf.vfnrclip.xu.f.qf v4, v8, fa2, v0.t
 # CHECK-INST: sf.vfnrclip.xu.f.qf v4, v8, fa2
 # CHECK-ENCODING: [0x5b,0x52,0x86,0x88]
 # CHECK-ERROR: instruction requires the following: 'XSfvfnrclipxfqf' (SiFive FP32-to-int8 Ranged Clip Instructions)
-# CHECK-UNKNOWN: 5b 52 86 88 <unknown>
+# CHECK-UNKNOWN: 8886525b <unknown>
 
 sf.vfnrclip.x.f.qf v4, v8, fa2
 # CHECK-INST: sf.vfnrclip.x.f.qf v4, v8, fa2
 # CHECK-ENCODING: [0x5b,0x52,0x86,0x8e]
 # CHECK-ERROR: instruction requires the following: 'XSfvfnrclipxfqf' (SiFive FP32-to-int8 Ranged Clip Instructions)
-# CHECK-UNKNOWN: 5b 52 86 8e <unknown>
+# CHECK-UNKNOWN: 8e86525b <unknown>
 
 sf.vfnrclip.x.f.qf v4, v8, fa2, v0.t
 # CHECK-INST: sf.vfnrclip.x.f.qf v4, v8, fa2
 # CHECK-ENCODING: [0x5b,0x52,0x86,0x8c]
 # CHECK-ERROR: instruction requires the following: 'XSfvfnrclipxfqf' (SiFive FP32-to-int8 Ranged Clip Instructions)
-# CHECK-UNKNOWN: 5b 52 86 8c <unknown>
+# CHECK-UNKNOWN: 8c86525b <unknown>
diff --git a/llvm/test/MC/RISCV/rvv/xsfvfwmacc.s b/llvm/test/MC/RISCV/rvv/xsfvfwmacc.s
index ba054fff2bd8..a9843c350fc8 100644
--- a/llvm/test/MC/RISCV/rvv/xsfvfwmacc.s
+++ b/llvm/test/MC/RISCV/rvv/xsfvfwmacc.s
@@ -12,4 +12,4 @@ sf.vfwmacc.4x4x4 v8, v4, v20
 # CHECK-INST: sf.vfwmacc.4x4x4 v8, v4, v20
 # CHECK-ENCODING: [0x5b,0x14,0x42,0xf3]
 # CHECK-ERROR: instruction requires the following: 'XSfvfwmaccqqq' (SiFive Matrix Multiply Accumulate Instruction and 4-by-4))
-# CHECK-UNKNOWN: 5b 14 42 f3 <unknown>
+# CHECK-UNKNOWN: f342145b <unknown>
diff --git a/llvm/test/MC/RISCV/rvv/xsfvqmacc.s b/llvm/test/MC/RISCV/rvv/xsfvqmacc.s
index ba19f2184486..81703c847d74 100644
--- a/llvm/test/MC/RISCV/rvv/xsfvqmacc.s
+++ b/llvm/test/MC/RISCV/rvv/xsfvqmacc.s
@@ -12,46 +12,46 @@ sf.vqmaccu.2x8x2 v8, v4, v20
 # CHECK-INST: sf.vqmaccu.2x8x2 v8, v4, v20
 # CHECK-ENCODING: [0x5b,0x24,0x42,0xb3]
 # CHECK-ERROR: instruction requires the following: 'XSfvqmaccdod' (SiFive Int8 Matrix Multiplication Instructions (2-by-8 and 8-by-2))
-# CHECK-UNKNOWN: 5b 24 42 b3 <unknown>
+# CHECK-UNKNOWN: b342245b <unknown>
 
 sf.vqmacc.2x8x2 v8, v4, v20
 # CHECK-INST: sf.vqmacc.2x8x2 v8, v4, v20
 # CHECK-ENCODING: [0x5b,0x24,0x42,0xb7]
 # CHECK-ERROR: instruction requires the following: 'XSfvqmaccdod' (SiFive Int8 Matrix Multiplication Instructions (2-by-8 and 8-by-2))
-# CHECK-UNKNOWN: 5b 24 42 b7 <unknown>
+# CHECK-UNKNOWN: b742245b <unknown>
 
 sf.vqmaccus.2x8x2 v8, v4, v20
 # CHECK-INST: sf.vqmaccus.2x8x2 v8, v4, v20
 # CHECK-ENCODING: [0x5b,0x24,0x42,0xbb]
 # CHECK-ERROR: instruction requires the following: 'XSfvqmaccdod' (SiFive Int8 Matrix Multiplication Instructions (2-by-8 and 8-by-2))
-# CHECK-UNKNOWN: 5b 24 42 bb <unknown>
+# CHECK-UNKNOWN: bb42245b <unknown>
 
 sf.vqmaccsu.2x8x2 v8, v4, v20
 # CHECK-INST: sf.vqmaccsu.2x8x2 v8, v4, v20
 # CHECK-ENCODING: [0x5b,0x24,0x42,0xbf]
 # CHECK-ERROR: instruction requires the following: 'XSfvqmaccdod' (SiFive Int8 Matrix Multiplication Instructions (2-by-8 and 8-by-2))
-# CHECK-UNKNOWN: 5b 24 42 bf <unknown>
+# CHECK-UNKNOWN: bf42245b <unknown>
 
 sf.vqmaccu.4x8x4 v8, v4, v20
 # CHECK-INST: sf.vqmaccu.4x8x4 v8, v4, v20
 # CHECK-ENCODING: [0x5b,0x24,0x42,0xf3]
 # CHECK-ERROR: instruction requires the following: 'XSfvqmaccqoq' (SiFive Int8 Matrix Multiplication Instructions (4-by-8 and 8-by-4))
-# CHECK-UNKNOWN: 5b 24 42 f3 <unknown>
+# CHECK-UNKNOWN: f342245b <unknown>
 
 sf.vqmacc.4x8x4 v8, v4, v20
 # CHECK-INST: sf.vqmacc.4x8x4 v8, v4, v20
 # CHECK-ENCODING: [0x5b,0x24,0x42,0xf7]
 # CHECK-ERROR: instruction requires the following: 'XSfvqmaccqoq' (SiFive Int8 Matrix Multiplication Instructions (4-by-8 and 8-by-4))
-# CHECK-UNKNOWN: 5b 24 42 f7 <unknown>
+# CHECK-UNKNOWN: f742245b <unknown>
 
 sf.vqmaccus.4x8x4 v8, v4, v20
 # CHECK-INST: sf.vqmaccus.4x8x4 v8, v4, v20
 # CHECK-ENCODING: [0x5b,0x24,0x42,0xfb]
 # CHECK-ERROR: instruction requires the following: 'XSfvqmaccqoq' (SiFive Int8 Matrix Multiplication Instructions (4-by-8 and 8-by-4))
-# CHECK-UNKNOWN: 5b 24 42 fb <unknown>
+# CHECK-UNKNOWN: fb42245b <unknown>
 
 sf.vqmaccsu.4x8x4 v8, v4, v20
 # CHECK-INST: sf.vqmaccsu.4x8x4 v8, v4, v20
 # CHECK-ENCODING: [0x5b,0x24,0x42,0xff]
 # CHECK-ERROR: instruction requires the following: 'XSfvqmaccqoq' (SiFive Int8 Matrix Multiplication Instructions (4-by-8 and 8-by-4))
-# CHECK-UNKNOWN: 5b 24 42 ff <unknown>
+# CHECK-UNKNOWN: ff42245b <unknown>
diff --git a/llvm/test/MC/RISCV/rvv/zvbb.s b/llvm/test/MC/RISCV/rvv/zvbb.s
index 04e5ad6e03f7..d9d1f6f42d32 100644
--- a/llvm/test/MC/RISCV/rvv/zvbb.s
+++ b/llvm/test/MC/RISCV/rvv/zvbb.s
@@ -12,40 +12,40 @@ vbrev.v v10, v9, v0.t
 # CHECK-INST: vbrev.v v10, v9, v0.t
 # CHECK-ENCODING: [0x57,0x25,0x95,0x48]
 # CHECK-ERROR: instruction requires the following: 'Zvbb' (Vector basic bit-manipulation instructions){{$}}
-# CHECK-UNKNOWN: 57 25 95 48   <unknown>
+# CHECK-UNKNOWN: 48952557 <unknown>
 
 vclz.v v10, v9, v0.t
 # CHECK-INST: vclz.v v10, v9, v0.t
 # CHECK-ENCODING: [0x57,0x25,0x96,0x48]
 # CHECK-ERROR: instruction requires the following: 'Zvbb' (Vector basic bit-manipulation instructions){{$}}
-# CHECK-UNKNOWN: 57 25 96 48   <unknown>
+# CHECK-UNKNOWN: 48962557 <unknown>
 
 vcpop.v v10, v9, v0.t
 # CHECK-INST: vcpop.v v10, v9, v0.t
 # CHECK-ENCODING: [0x57,0x25,0x97,0x48]
 # CHECK-ERROR: instruction requires the following: 'Zvbb' (Vector basic bit-manipulation instructions){{$}}
-# CHECK-UNKNOWN: 57 25 97 48   <unknown>
+# CHECK-UNKNOWN: 48972557 <unknown>
 
 vctz.v v10, v9, v0.t
 # CHECK-INST: vctz.v v10, v9, v0.t
 # CHECK-ENCODING: [0x57,0xa5,0x96,0x48]
 # CHECK-ERROR: instruction requires the following: 'Zvbb' (Vector basic bit-manipulation instructions){{$}}
-# CHECK-UNKNOWN: 57 a5 96 48   <unknown>
+# CHECK-UNKNOWN: 4896a557 <unknown>
 
 vwsll.vv v10, v9, v8, v0.t
 # CHECK-INST: vwsll.vv v10, v9, v8, v0.t
 # CHECK-ENCODING: [0x57,0x05,0x94,0xd4]
 # CHECK-ERROR: instruction requires the following: 'Zvbb' (Vector basic bit-manipulation instructions){{$}}
-# CHECK-UNKNOWN: 57 05 94 d4   <unknown>
+# CHECK-UNKNOWN: d4940557 <unknown>
 
 vwsll.vx v10, v9, a0, v0.t
 # CHECK-INST: vwsll.vx v10, v9, a0, v0.t
 # CHECK-ENCODING: [0x57,0x45,0x95,0xd4]
 # CHECK-ERROR: instruction requires the following: 'Zvbb' (Vector basic bit-manipulation instructions){{$}}
-# CHECK-UNKNOWN: 57 45 95 d4   <unknown>
+# CHECK-UNKNOWN: d4954557 <unknown>
 
 vwsll.vi v10, v9, 29, v0.t
 # CHECK-INST: vwsll.vi v10, v9, 29, v0.t
 # CHECK-ENCODING: [0x57,0xb5,0x9e,0xd4]
 # CHECK-ERROR: instruction requires the following: 'Zvbb' (Vector basic bit-manipulation instructions){{$}}
-# CHECK-UNKNOWN: 57 b5 9e d4   <unknown>
+# CHECK-UNKNOWN: d49eb557 <unknown>
diff --git a/llvm/test/MC/RISCV/rvv/zvbc.s b/llvm/test/MC/RISCV/rvv/zvbc.s
index b32349a2db13..0eb02d153b79 100644
--- a/llvm/test/MC/RISCV/rvv/zvbc.s
+++ b/llvm/test/MC/RISCV/rvv/zvbc.s
@@ -12,22 +12,22 @@ vclmul.vv v10, v9, v8
 # CHECK-INST: vclmul.vv v10, v9, v8
 # CHECK-ENCODING: [0x57,0x25,0x94,0x32]
 # CHECK-ERROR: instruction requires the following: 'Zvbc' (Vector Carryless Multiplication){{$}}
-# CHECK-UNKNOWN: 57 25 94 32   <unknown>
+# CHECK-UNKNOWN: 32942557 <unknown>
 
 vclmul.vx v10, v9, a0
 # CHECK-INST: vclmul.vx v10, v9, a0
 # CHECK-ENCODING: [0x57,0x65,0x95,0x32]
 # CHECK-ERROR: instruction requires the following: 'Zvbc' (Vector Carryless Multiplication){{$}}
-# CHECK-UNKNOWN: 57 65 95 32   <unknown>
+# CHECK-UNKNOWN: 32956557 <unknown>
 
 vclmulh.vv v10, v9, v8
 # CHECK-INST: vclmulh.vv v10, v9, v8
 # CHECK-ENCODING: [0x57,0x25,0x94,0x36]
 # CHECK-ERROR: instruction requires the following: 'Zvbc' (Vector Carryless Multiplication){{$}}
-# CHECK-UNKNOWN: 57 25 94 36   <unknown>
+# CHECK-UNKNOWN: 36942557 <unknown>
 
 vclmulh.vx v10, v9, a0
 # CHECK-INST: vclmulh.vx v10, v9, a0
 # CHECK-ENCODING: [0x57,0x65,0x95,0x36]
 # CHECK-ERROR: instruction requires the following: 'Zvbc' (Vector Carryless Multiplication){{$}}
-# CHECK-UNKNOWN: 57 65 95 36   <unknown>
+# CHECK-UNKNOWN: 36956557 <unknown>
diff --git a/llvm/test/MC/RISCV/rvv/zvfbfmin.s b/llvm/test/MC/RISCV/rvv/zvfbfmin.s
index 1cbe027ef26c..7965c2482b00 100644
--- a/llvm/test/MC/RISCV/rvv/zvfbfmin.s
+++ b/llvm/test/MC/RISCV/rvv/zvfbfmin.s
@@ -20,23 +20,23 @@
 # CHECK-INST: vfncvtbf16.f.f.w v8, v4, v0.t
 # CHECK-ENCODING: [0x57,0x94,0x4e,0x48]
 # CHECK-ERROR: instruction requires the following: 'Zvfbfmin' (Vector BF16 Converts){{$}}
-# CHECK-UNKNOWN: 57 94 4e 48 <unknown>
+# CHECK-UNKNOWN: 484e9457 <unknown>
 vfncvtbf16.f.f.w v8, v4, v0.t
 
 # CHECK-INST: vfncvtbf16.f.f.w v8, v4
 # CHECK-ENCODING: [0x57,0x94,0x4e,0x4a]
 # CHECK-ERROR: instruction requires the following: 'Zvfbfmin' (Vector BF16 Converts){{$}}
-# CHECK-UNKNOWN: 57 94 4e 4a <unknown>
+# CHECK-UNKNOWN: 4a4e9457 <unknown>
 vfncvtbf16.f.f.w v8, v4
 
 # CHECK-INST: vfwcvtbf16.f.f.v v8, v4, v0.t
 # CHECK-ENCODING: [0x57,0x94,0x46,0x48]
 # CHECK-ERROR: instruction requires the following: 'Zvfbfmin' (Vector BF16 Converts){{$}}
-# CHECK-UNKNOWN: 57 94 46 48 <unknown>
+# CHECK-UNKNOWN: 48469457 <unknown>
 vfwcvtbf16.f.f.v v8, v4, v0.t
 
 # CHECK-INST: vfwcvtbf16.f.f.v v8, v4
 # CHECK-ENCODING: [0x57,0x94,0x46,0x4a]
 # CHECK-ERROR: instruction requires the following: 'Zvfbfmin' (Vector BF16 Converts){{$}}
-# CHECK-UNKNOWN: 57 94 46 4a <unknown>
+# CHECK-UNKNOWN: 4a469457 <unknown>
 vfwcvtbf16.f.f.v v8, v4
diff --git a/llvm/test/MC/RISCV/rvv/zvfbfwma.s b/llvm/test/MC/RISCV/rvv/zvfbfwma.s
index 5a30d9f19ab6..330dee58d836 100644
--- a/llvm/test/MC/RISCV/rvv/zvfbfwma.s
+++ b/llvm/test/MC/RISCV/rvv/zvfbfwma.s
@@ -20,25 +20,25 @@
 # CHECK-INST: vfwmaccbf16.vv v8, v20, v4, v0.t
 # CHECK-ENCODING: [0x57,0x14,0x4a,0xec]
 # CHECK-ERROR: instruction requires the following: 'Zvfbfwma' (Vector BF16 widening mul-add){{$}}
-# CHECK-UNKNOWN: 57 14 4a ec <unknown>
+# CHECK-UNKNOWN: ec4a1457 <unknown>
 vfwmaccbf16.vv v8, v20, v4, v0.t
 
 # CHECK-INST: vfwmaccbf16.vv v8, v20, v4
 # CHECK-ENCODING: [0x57,0x14,0x4a,0xee]
 # CHECK-ERROR: instruction requires the following: 'Zvfbfwma' (Vector BF16 widening mul-add){{$}}
-# CHECK-UNKNOWN: 57 14 4a ee <unknown>
+# CHECK-UNKNOWN: ee4a1457 <unknown>
 vfwmaccbf16.vv v8, v20, v4
 
 # CHECK-INST: vfwmaccbf16.vf v8, fa0, v4, v0.t
 # CHECK-ENCODING: [0x57,0x54,0x45,0xec]
 # CHECK-ERROR: instruction requires the following: 'Zvfbfwma' (Vector BF16 widening mul-add){{$}}
-# CHECK-UNKNOWN: 57 54 45 ec <unknown>
+# CHECK-UNKNOWN: ec455457 <unknown>
 vfwmaccbf16.vf v8, fa0, v4, v0.t
 
 # CHECK-INST: vfwmaccbf16.vf v8, fa0, v4
 # CHECK-ENCODING: [0x57,0x54,0x45,0xee]
 # CHECK-ERROR: instruction requires the following: 'Zvfbfwma' (Vector BF16 widening mul-add){{$}}
-# CHECK-UNKNOWN: 57 54 45 ee <unknown>
+# CHECK-UNKNOWN: ee455457 <unknown>
 vfwmaccbf16.vf v8, fa0, v4
 
 # Check scalar half FP load/store/move included in this extension.
@@ -46,23 +46,23 @@ vfwmaccbf16.vf v8, fa0, v4
 # CHECK-INST: flh ft0, 12(a0)
 # CHECK-ENCODING: [0x07,0x10,0xc5,0x00]
 # CHECK-ERROR: instruction requires the following: 'Zfh' (Half-Precision Floating-Point) or 'Zfhmin' (Half-Precision Floating-Point Minimal) or 'Zfbfmin' (Scalar BF16 Converts){{$}}
-# CHECK-UNKNOWN: 07 10 c5 00 <unknown>
+# CHECK-UNKNOWN: 00c51007 <unknown>
 flh f0, 12(a0)
 
 # CHECK-INST: fsh ft6, 2047(s4)
 # CHECK-ENCODING: [0xa7,0x1f,0x6a,0x7e]
 # CHECK-ERROR: instruction requires the following: 'Zfh' (Half-Precision Floating-Point) or 'Zfhmin' (Half-Precision Floating-Point Minimal) or 'Zfbfmin' (Scalar BF16 Converts){{$}}
-# CHECK-UNKNOWN: a7 1f 6a 7e <unknown>
+# CHECK-UNKNOWN: 7e6a1fa7 <unknown>
 fsh f6, 2047(s4)
 
 # CHECK-INST: fmv.x.h a2, fs7
 # CHECK-ENCODING: [0x53,0x86,0x0b,0xe4]
 # CHECK-ERROR: instruction requires the following: 'Zfh' (Half-Precision Floating-Point) or 'Zfhmin' (Half-Precision Floating-Point Minimal) or 'Zfbfmin' (Scalar BF16 Converts){{$}}
-# CHECK-UNKNOWN: 53 86 0b e4 <unknown>
+# CHECK-UNKNOWN: e40b8653 <unknown>
 fmv.x.h a2, fs7
 
 # CHECK-INST: fmv.h.x ft1, a6
 # CHECK-ENCODING: [0xd3,0x00,0x08,0xf4]
 # CHECK-ERROR: instruction requires the following: 'Zfh' (Half-Precision Floating-Point) or 'Zfhmin' (Half-Precision Floating-Point Minimal) or 'Zfbfmin' (Scalar BF16 Converts){{$}}
-# CHECK-UNKNOWN: d3 00 08 f4 <unknown>
+# CHECK-UNKNOWN: f40800d3 <unknown>
 fmv.h.x ft1, a6
diff --git a/llvm/test/MC/RISCV/rvv/zvkb.s b/llvm/test/MC/RISCV/rvv/zvkb.s
index ae2dec18d33c..1833ba860c90 100644
--- a/llvm/test/MC/RISCV/rvv/zvkb.s
+++ b/llvm/test/MC/RISCV/rvv/zvkb.s
@@ -12,52 +12,52 @@ vandn.vv v10, v9, v8, v0.t
 # CHECK-INST: vandn.vv v10, v9, v8, v0.t
 # CHECK-ENCODING: [0x57,0x05,0x94,0x04]
 # CHECK-ERROR: instruction requires the following: 'Zvkb' (Vector Bit-manipulation used in Cryptography){{$}}
-# CHECK-UNKNOWN: 57 05 94 04   <unknown>
+# CHECK-UNKNOWN: 04940557 <unknown>
 
 vandn.vx v10, v9, a0, v0.t
 # CHECK-INST: vandn.vx v10, v9, a0, v0.t
 # CHECK-ENCODING: [0x57,0x45,0x95,0x04]
 # CHECK-ERROR: instruction requires the following: 'Zvkb' (Vector Bit-manipulation used in Cryptography){{$}}
-# CHECK-UNKNOWN: 57 45 95 04   <unknown>
+# CHECK-UNKNOWN: 04954557 <unknown>
 
 vbrev8.v v10, v9, v0.t
 # CHECK-INST: vbrev8.v v10, v9, v0.t
 # CHECK-ENCODING: [0x57,0x25,0x94,0x48]
 # CHECK-ERROR: instruction requires the following: 'Zvkb' (Vector Bit-manipulation used in Cryptography){{$}}
-# CHECK-UNKNOWN: 57 25 94 48   <unknown>
+# CHECK-UNKNOWN: 48942557 <unknown>
 
 vrev8.v v10, v9, v0.t
 # CHECK-INST: vrev8.v v10, v9, v0.t
 # CHECK-ENCODING: [0x57,0xa5,0x94,0x48]
 # CHECK-ERROR: instruction requires the following: 'Zvkb' (Vector Bit-manipulation used in Cryptography){{$}}
-# CHECK-UNKNOWN: 57 a5 94 48   <unknown>
+# CHECK-UNKNOWN: 4894a557 <unknown>
 
 vrol.vv v10, v9, v8, v0.t
 # CHECK-INST: vrol.vv v10, v9, v8, v0.t
 # CHECK-ENCODING: [0x57,0x05,0x94,0x54]
 # CHECK-ERROR: instruction requires the following: 'Zvkb' (Vector Bit-manipulation used in Cryptography){{$}}
-# CHECK-UNKNOWN: 57 05 94 54   <unknown>
+# CHECK-UNKNOWN: 54940557 <unknown>
 
 vrol.vx v10, v9, a0, v0.t
 # CHECK-INST: vrol.vx v10, v9, a0, v0.t
 # CHECK-ENCODING: [0x57,0x45,0x95,0x54]
 # CHECK-ERROR: instruction requires the following: 'Zvkb' (Vector Bit-manipulation used in Cryptography){{$}}
-# CHECK-UNKNOWN: 57 45 95 54   <unknown>
+# CHECK-UNKNOWN: 54954557 <unknown>
 
 vror.vv v10, v9, v8, v0.t
 # CHECK-INST: vror.vv v10, v9, v8, v0.t
 # CHECK-ENCODING: [0x57,0x05,0x94,0x50]
 # CHECK-ERROR: instruction requires the following: 'Zvkb' (Vector Bit-manipulation used in Cryptography){{$}}
-# CHECK-UNKNOWN: 57 05 94 50   <unknown>
+# CHECK-UNKNOWN: 50940557 <unknown>
 
 vror.vx v10, v9, a0, v0.t
 # CHECK-INST: vror.vx v10, v9, a0, v0.t
 # CHECK-ENCODING: [0x57,0x45,0x95,0x50]
 # CHECK-ERROR: instruction requires the following: 'Zvkb' (Vector Bit-manipulation used in Cryptography){{$}}
-# CHECK-UNKNOWN: 57 45 95 50   <unknown>
+# CHECK-UNKNOWN: 50954557 <unknown>
 
 vror.vi v10, v9, 33, v0.t
 # CHECK-INST: vror.vi v10, v9, 33, v0.t
 # CHECK-ENCODING: [0x57,0xb5,0x90,0x54]
 # CHECK-ERROR: instruction requires the following: 'Zvkb' (Vector Bit-manipulation used in Cryptography){{$}}
-# CHECK-UNKNOWN: 57 b5 90 54   <unknown>
+# CHECK-UNKNOWN: 5490b557 <unknown>
diff --git a/llvm/test/MC/RISCV/rvv/zvkg.s b/llvm/test/MC/RISCV/rvv/zvkg.s
index f2016bc116b6..48b84659e0ae 100644
--- a/llvm/test/MC/RISCV/rvv/zvkg.s
+++ b/llvm/test/MC/RISCV/rvv/zvkg.s
@@ -12,10 +12,10 @@ vghsh.vv v10, v9, v8
 # CHECK-INST: vghsh.vv v10, v9, v8
 # CHECK-ENCODING: [0x77,0x25,0x94,0xb2]
 # CHECK-ERROR: instruction requires the following: 'Zvkg' (Vector GCM instructions for Cryptography){{$}}
-# CHECK-UNKNOWN: 77 25 94 b2   <unknown>
+# CHECK-UNKNOWN: b2942577 <unknown>
 
 vgmul.vv v10, v9
 # CHECK-INST: vgmul.vv v10, v9
 # CHECK-ENCODING: [0x77,0xa5,0x98,0xa2]
 # CHECK-ERROR: instruction requires the following: 'Zvkg' (Vector GCM instructions for Cryptography){{$}}
-# CHECK-UNKNOWN: 77 a5 98 a2   <unknown>
+# CHECK-UNKNOWN: a298a577 <unknown>
diff --git a/llvm/test/MC/RISCV/rvv/zvkned.s b/llvm/test/MC/RISCV/rvv/zvkned.s
index e51a9cc562f1..bee3d74ee88d 100644
--- a/llvm/test/MC/RISCV/rvv/zvkned.s
+++ b/llvm/test/MC/RISCV/rvv/zvkned.s
@@ -12,76 +12,76 @@ vaesdf.vv v10, v9
 # CHECK-INST: vaesdf.vv v10, v9
 # CHECK-ENCODING: [0x77,0xa5,0x90,0xa2]
 # CHECK-ERROR: instruction requires the following: 'Zvkned' (Vector AES Encryption & Decryption (Single Round)){{$}}
-# CHECK-UNKNOWN: 77 a5 90 a2   <unknown>
+# CHECK-UNKNOWN: a290a577 <unknown>
 
 vaesdf.vs v10, v9
 # CHECK-INST: vaesdf.vs v10, v9
 # CHECK-ENCODING: [0x77,0xa5,0x90,0xa6]
 # CHECK-ERROR: instruction requires the following: 'Zvkned' (Vector AES Encryption & Decryption (Single Round)){{$}}
-# CHECK-UNKNOWN: 77 a5 90 a6   <unknown>
+# CHECK-UNKNOWN: a690a577 <unknown>
 
 vaesef.vv v10, v9
 # CHECK-INST: vaesef.vv v10, v9
 # CHECK-ENCODING: [0x77,0xa5,0x91,0xa2]
 # CHECK-ERROR: instruction requires the following: 'Zvkned' (Vector AES Encryption & Decryption (Single Round)){{$}}
-# CHECK-UNKNOWN: 77 a5 91 a2   <unknown>
+# CHECK-UNKNOWN: a291a577 <unknown>
                        
 vaesef.vs v10, v9
 # CHECK-INST: vaesef.vs v10, v9
 # CHECK-ENCODING: [0x77,0xa5,0x91,0xa6]
 # CHECK-ERROR: instruction requires the following: 'Zvkned' (Vector AES Encryption & Decryption (Single Round)){{$}}
-# CHECK-UNKNOWN: 77 a5 91 a6   <unknown>
+# CHECK-UNKNOWN: a691a577 <unknown>
 
 vaesdm.vv v10, v9
 # CHECK-INST: vaesdm.vv v10, v9
 # CHECK-ENCODING: [0x77,0x25,0x90,0xa2]
 # CHECK-ERROR: instruction requires the following: 'Zvkned' (Vector AES Encryption & Decryption (Single Round)){{$}}
-# CHECK-UNKNOWN: 77 25 90 a2   <unknown>
+# CHECK-UNKNOWN: a2902577 <unknown>
                        
 vaesdm.vs v10, v9
 # CHECK-INST: vaesdm.vs v10, v9
 # CHECK-ENCODING: [0x77,0x25,0x90,0xa6]
 # CHECK-ERROR: instruction requires the following: 'Zvkned' (Vector AES Encryption & Decryption (Single Round)){{$}}
-# CHECK-UNKNOWN: 77 25 90 a6   <unknown>
+# CHECK-UNKNOWN: a6902577 <unknown>
 
 vaesem.vv v10, v9
 # CHECK-INST: vaesem.vv v10, v9
 # CHECK-ENCODING: [0x77,0x25,0x91,0xa2]
 # CHECK-ERROR: instruction requires the following: 'Zvkned' (Vector AES Encryption & Decryption (Single Round)){{$}}
-# CHECK-UNKNOWN: 77 25 91 a2   <unknown>
+# CHECK-UNKNOWN: a2912577 <unknown>
                        
 vaesem.vs v10, v9
 # CHECK-INST: vaesem.vs v10, v9
 # CHECK-ENCODING: [0x77,0x25,0x91,0xa6]
 # CHECK-ERROR: instruction requires the following: 'Zvkned' (Vector AES Encryption & Decryption (Single Round)){{$}}
-# CHECK-UNKNOWN: 77 25 91 a6   <unknown>
+# CHECK-UNKNOWN: a6912577 <unknown>
 
 vaeskf1.vi v10, v9, 1
 # CHECK-INST: vaeskf1.vi v10, v9, 1
 # CHECK-ENCODING: [0x77,0xa5,0x90,0x8a]
 # CHECK-ERROR: instruction requires the following: 'Zvkned' (Vector AES Encryption & Decryption (Single Round)){{$}}
-# CHECK-UNKNOWN: 77 a5 90 8a   <unknown>
+# CHECK-UNKNOWN: 8a90a577 <unknown>
 
 vaeskf1.vi v10, v9, 31
 # CHECK-INST: vaeskf1.vi v10, v9, 31
 # CHECK-ENCODING: [0x77,0xa5,0x9f,0x8a]
 # CHECK-ERROR: instruction requires the following: 'Zvkned' (Vector AES Encryption & Decryption (Single Round)){{$}}
-# CHECK-UNKNOWN: 77 a5 9f 8a   <unknown>
+# CHECK-UNKNOWN: 8a9fa577 <unknown>
 
 vaeskf2.vi v10, v9, 2
 # CHECK-INST: vaeskf2.vi v10, v9, 2
 # CHECK-ENCODING: [0x77,0x25,0x91,0xaa]
 # CHECK-ERROR: instruction requires the following: 'Zvkned' (Vector AES Encryption & Decryption (Single Round)){{$}}
-# CHECK-UNKNOWN: 77 25 91 aa   <unknown>
+# CHECK-UNKNOWN: aa912577 <unknown>
 
 vaeskf2.vi v10, v9, 31
 # CHECK-INST: vaeskf2.vi v10, v9, 31
 # CHECK-ENCODING: [0x77,0xa5,0x9f,0xaa]
 # CHECK-ERROR: instruction requires the following: 'Zvkned' (Vector AES Encryption & Decryption (Single Round)){{$}}
-# CHECK-UNKNOWN: 77 a5 9f aa   <unknown>
+# CHECK-UNKNOWN: aa9fa577 <unknown>
 
 vaesz.vs v10, v9
 # CHECK-INST: vaesz.vs v10, v9
 # CHECK-ENCODING: [0x77,0xa5,0x93,0xa6]
 # CHECK-ERROR: instruction requires the following: 'Zvkned' (Vector AES Encryption & Decryption (Single Round)){{$}}
-# CHECK-UNKNOWN: 77 a5 93 a6   <unknown>
+# CHECK-UNKNOWN: a693a577 <unknown>
diff --git a/llvm/test/MC/RISCV/rvv/zvknh.s b/llvm/test/MC/RISCV/rvv/zvknh.s
index aa8033a5f217..b16b9081f7e6 100644
--- a/llvm/test/MC/RISCV/rvv/zvknh.s
+++ b/llvm/test/MC/RISCV/rvv/zvknh.s
@@ -18,17 +18,17 @@
 vsha2ms.vv v10, v9, v8
 # CHECK-INST: vsha2ms.vv v10, v9, v8
 # CHECK-ENCODING: [0x77,0x25,0x94,0xb6]
-# CHECK-UNKNOWN: 77 25 94 b6   <unknown>
+# CHECK-UNKNOWN: b6942577 <unknown>
 # CHECK-ERROR: instruction requires the following: 'Zvknha' or 'Zvknhb' (Vector SHA-2){{$}}
 
 vsha2ch.vv v10, v9, v8
 # CHECK-INST: vsha2ch.vv v10, v9, v8
 # CHECK-ENCODING: [0x77,0x25,0x94,0xba]
-# CHECK-UNKNOWN: 77 25 94 ba   <unknown>
+# CHECK-UNKNOWN: ba942577 <unknown>
 # CHECK-ERROR: instruction requires the following: 'Zvknha' or 'Zvknhb' (Vector SHA-2){{$}}
 
 vsha2cl.vv v10, v9, v8
 # CHECK-INST: vsha2cl.vv v10, v9, v8
 # CHECK-ENCODING: [0x77,0x25,0x94,0xbe]
-# CHECK-UNKNOWN: 77 25 94 be   <unknown>
+# CHECK-UNKNOWN: be942577 <unknown>
 # CHECK-ERROR: instruction requires the following: 'Zvknha' or 'Zvknhb' (Vector SHA-2){{$}}
diff --git a/llvm/test/MC/RISCV/rvv/zvksed.s b/llvm/test/MC/RISCV/rvv/zvksed.s
index 87c9713f8c65..f7a0949272ff 100644
--- a/llvm/test/MC/RISCV/rvv/zvksed.s
+++ b/llvm/test/MC/RISCV/rvv/zvksed.s
@@ -12,22 +12,22 @@ vsm4k.vi v10, v9, 7
 # CHECK-INST: vsm4k.vi v10, v9, 7
 # CHECK-ENCODING: [0x77,0xa5,0x93,0x86]
 # CHECK-ERROR: instruction requires the following: 'Zvksed' (SM4 Block Cipher Instructions){{$}}
-# CHECK-UNKNOWN: 77 a5 93 86   <unknown>
+# CHECK-UNKNOWN: 8693a577 <unknown>
 
 vsm4k.vi v10, v9, 31
 # CHECK-INST: vsm4k.vi v10, v9, 31
 # CHECK-ENCODING: [0x77,0xa5,0x9f,0x86]
 # CHECK-ERROR: instruction requires the following: 'Zvksed' (SM4 Block Cipher Instructions){{$}}
-# CHECK-UNKNOWN: 77 a5 9f 86   <unknown>
+# CHECK-UNKNOWN: 869fa577 <unknown>
 
 vsm4r.vv v10, v9
 # CHECK-INST: vsm4r.vv v10, v9
 # CHECK-ENCODING: [0x77,0x25,0x98,0xa2]
 # CHECK-ERROR: instruction requires the following: 'Zvksed' (SM4 Block Cipher Instructions){{$}}
-# CHECK-UNKNOWN: 77 25 98 a2   <unknown>
+# CHECK-UNKNOWN: a2982577 <unknown>
 
 vsm4r.vs v10, v9
 # CHECK-INST: vsm4r.vs v10, v9
 # CHECK-ENCODING: [0x77,0x25,0x98,0xa6]
 # CHECK-ERROR: instruction requires the following: 'Zvksed' (SM4 Block Cipher Instructions){{$}}
-# CHECK-UNKNOWN: 77 25 98 a6   <unknown>
+# CHECK-UNKNOWN: a6982577 <unknown>
diff --git a/llvm/test/MC/RISCV/rvv/zvksh.s b/llvm/test/MC/RISCV/rvv/zvksh.s
index 06251ff6efe5..ef1c654b4605 100644
--- a/llvm/test/MC/RISCV/rvv/zvksh.s
+++ b/llvm/test/MC/RISCV/rvv/zvksh.s
@@ -12,17 +12,17 @@ vsm3c.vi v10, v9, 7
 # CHECK-INST: vsm3c.vi v10, v9, 7
 # CHECK-ENCODING: [0x77,0xa5,0x93,0xae]
 # CHECK-ERROR: instruction requires the following: 'Zvksh' (SM3 Hash Function Instructions){{$}}
-# CHECK-UNKNOWN: 77 a5 93 ae   <unknown>
+# CHECK-UNKNOWN: ae93a577 <unknown>
 
 vsm3me.vv v10, v9, v8
 # CHECK-INST: vsm3me.vv v10, v9, v8
 # CHECK-ENCODING: [0x77,0x25,0x94,0x82]
 # CHECK-ERROR: instruction requires the following: 'Zvksh' (SM3 Hash Function Instructions){{$}}
-# CHECK-UNKNOWN: 77 25 94 82   <unknown>
+# CHECK-UNKNOWN: 82942577 <unknown>
 
 # vs1 is allowed to overlap, but not vs2.
 vsm3me.vv v10, v9, v10
 # CHECK-INST: vsm3me.vv v10, v9, v10
 # CHECK-ENCODING: [0x77,0x25,0x95,0x82]
 # CHECK-ERROR: instruction requires the following: 'Zvksh' (SM3 Hash Function Instructions){{$}}
-# CHECK-UNKNOWN: 77 25 95 82   <unknown>
+# CHECK-UNKNOWN: 82952577 <unknown>
diff --git a/llvm/test/MC/RISCV/rvv/zvlsseg.s b/llvm/test/MC/RISCV/rvv/zvlsseg.s
index 9a83ea9f8721..65089e2261be 100644
--- a/llvm/test/MC/RISCV/rvv/zvlsseg.s
+++ b/llvm/test/MC/RISCV/rvv/zvlsseg.s
@@ -13,3076 +13,3076 @@ vlseg2e8.v v8, (a0), v0.t
 # CHECK-INST: vlseg2e8.v v8, (a0), v0.t
 # CHECK-ENCODING: [0x07,0x04,0x05,0x20]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 04 05 20 <unknown>
+# CHECK-UNKNOWN: 20050407 <unknown>
 
 vlseg2e8.v v8, (a0)
 # CHECK-INST: vlseg2e8.v v8, (a0)
 # CHECK-ENCODING: [0x07,0x04,0x05,0x22]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 04 05 22 <unknown>
+# CHECK-UNKNOWN: 22050407 <unknown>
 
 vlseg2e16.v v8, (a0), v0.t
 # CHECK-INST: vlseg2e16.v v8, (a0), v0.t
 # CHECK-ENCODING: [0x07,0x54,0x05,0x20]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 54 05 20 <unknown>
+# CHECK-UNKNOWN: 20055407 <unknown>
 
 vlseg2e16.v v8, (a0)
 # CHECK-INST: vlseg2e16.v v8, (a0)
 # CHECK-ENCODING: [0x07,0x54,0x05,0x22]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 54 05 22 <unknown>
+# CHECK-UNKNOWN: 22055407 <unknown>
 
 vlseg2e32.v v8, (a0), v0.t
 # CHECK-INST: vlseg2e32.v v8, (a0), v0.t
 # CHECK-ENCODING: [0x07,0x64,0x05,0x20]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 64 05 20 <unknown>
+# CHECK-UNKNOWN: 20056407 <unknown>
 
 vlseg2e32.v v8, (a0)
 # CHECK-INST: vlseg2e32.v v8, (a0)
 # CHECK-ENCODING: [0x07,0x64,0x05,0x22]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 64 05 22 <unknown>
+# CHECK-UNKNOWN: 22056407 <unknown>
 
 vlseg2e64.v v8, (a0), v0.t
 # CHECK-INST: vlseg2e64.v v8, (a0), v0.t
 # CHECK-ENCODING: [0x07,0x74,0x05,0x20]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 74 05 20 <unknown>
+# CHECK-UNKNOWN: 20057407 <unknown>
 
 vlseg2e64.v v8, (a0)
 # CHECK-INST: vlseg2e64.v v8, (a0)
 # CHECK-ENCODING: [0x07,0x74,0x05,0x22]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 74 05 22 <unknown>
+# CHECK-UNKNOWN: 22057407 <unknown>
 
 vlseg2e8ff.v v8, (a0), v0.t
 # CHECK-INST: vlseg2e8ff.v v8, (a0), v0.t
 # CHECK-ENCODING: [0x07,0x04,0x05,0x21]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 04 05 21 <unknown>
+# CHECK-UNKNOWN: 21050407 <unknown>
 
 vlseg2e8ff.v v8, (a0)
 # CHECK-INST: vlseg2e8ff.v v8, (a0)
 # CHECK-ENCODING: [0x07,0x04,0x05,0x23]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 04 05 23 <unknown>
+# CHECK-UNKNOWN: 23050407 <unknown>
 
 vlseg2e16ff.v v8, (a0), v0.t
 # CHECK-INST: vlseg2e16ff.v v8, (a0), v0.t
 # CHECK-ENCODING: [0x07,0x54,0x05,0x21]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 54 05 21 <unknown>
+# CHECK-UNKNOWN: 21055407 <unknown>
 
 vlseg2e16ff.v v8, (a0)
 # CHECK-INST: vlseg2e16ff.v v8, (a0)
 # CHECK-ENCODING: [0x07,0x54,0x05,0x23]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 54 05 23 <unknown>
+# CHECK-UNKNOWN: 23055407 <unknown>
 
 vlseg2e32ff.v v8, (a0), v0.t
 # CHECK-INST: vlseg2e32ff.v v8, (a0), v0.t
 # CHECK-ENCODING: [0x07,0x64,0x05,0x21]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 64 05 21 <unknown>
+# CHECK-UNKNOWN: 21056407 <unknown>
 
 vlseg2e32ff.v v8, (a0)
 # CHECK-INST: vlseg2e32ff.v v8, (a0)
 # CHECK-ENCODING: [0x07,0x64,0x05,0x23]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 64 05 23 <unknown>
+# CHECK-UNKNOWN: 23056407 <unknown>
 
 vlseg2e64ff.v v8, (a0), v0.t
 # CHECK-INST: vlseg2e64ff.v v8, (a0), v0.t
 # CHECK-ENCODING: [0x07,0x74,0x05,0x21]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 74 05 21 <unknown>
+# CHECK-UNKNOWN: 21057407 <unknown>
 
 vlseg2e64ff.v v8, (a0)
 # CHECK-INST: vlseg2e64ff.v v8, (a0)
 # CHECK-ENCODING: [0x07,0x74,0x05,0x23]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 74 05 23 <unknown>
+# CHECK-UNKNOWN: 23057407 <unknown>
 
 vlsseg2e8.v v8, (a0), a1, v0.t
 # CHECK-INST: vlsseg2e8.v v8, (a0), a1, v0.t
 # CHECK-ENCODING: [0x07,0x04,0xb5,0x28]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 04 b5 28 <unknown>
+# CHECK-UNKNOWN: 28b50407 <unknown>
 
 vlsseg2e8.v v8, (a0), a1
 # CHECK-INST: vlsseg2e8.v v8, (a0), a1
 # CHECK-ENCODING: [0x07,0x04,0xb5,0x2a]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 04 b5 2a <unknown>
+# CHECK-UNKNOWN: 2ab50407 <unknown>
 
 vlsseg2e16.v v8, (a0), a1, v0.t
 # CHECK-INST: vlsseg2e16.v v8, (a0), a1, v0.t
 # CHECK-ENCODING: [0x07,0x54,0xb5,0x28]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 54 b5 28 <unknown>
+# CHECK-UNKNOWN: 28b55407 <unknown>
 
 vlsseg2e16.v v8, (a0), a1
 # CHECK-INST: vlsseg2e16.v v8, (a0), a1
 # CHECK-ENCODING: [0x07,0x54,0xb5,0x2a]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 54 b5 2a <unknown>
+# CHECK-UNKNOWN: 2ab55407 <unknown>
 
 vlsseg2e32.v v8, (a0), a1, v0.t
 # CHECK-INST: vlsseg2e32.v v8, (a0), a1, v0.t
 # CHECK-ENCODING: [0x07,0x64,0xb5,0x28]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 64 b5 28 <unknown>
+# CHECK-UNKNOWN: 28b56407 <unknown>
 
 vlsseg2e32.v v8, (a0), a1
 # CHECK-INST: vlsseg2e32.v v8, (a0), a1
 # CHECK-ENCODING: [0x07,0x64,0xb5,0x2a]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 64 b5 2a <unknown>
+# CHECK-UNKNOWN: 2ab56407 <unknown>
 
 vlsseg2e64.v v8, (a0), a1, v0.t
 # CHECK-INST: vlsseg2e64.v v8, (a0), a1, v0.t
 # CHECK-ENCODING: [0x07,0x74,0xb5,0x28]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 74 b5 28 <unknown>
+# CHECK-UNKNOWN: 28b57407 <unknown>
 
 vlsseg2e64.v v8, (a0), a1
 # CHECK-INST: vlsseg2e64.v v8, (a0), a1
 # CHECK-ENCODING: [0x07,0x74,0xb5,0x2a]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 74 b5 2a <unknown>
+# CHECK-UNKNOWN: 2ab57407 <unknown>
 
 vluxseg2ei8.v v8, (a0), v4, v0.t
 # CHECK-INST: vluxseg2ei8.v v8, (a0), v4, v0.t
 # CHECK-ENCODING: [0x07,0x04,0x45,0x24]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 04 45 24 <unknown>
+# CHECK-UNKNOWN: 24450407 <unknown>
 
 vluxseg2ei8.v v8, (a0), v4
 # CHECK-INST: vluxseg2ei8.v v8, (a0), v4
 # CHECK-ENCODING: [0x07,0x04,0x45,0x26]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 04 45 26 <unknown>
+# CHECK-UNKNOWN: 26450407 <unknown>
 
 vluxseg2ei16.v v8, (a0), v4, v0.t
 # CHECK-INST: vluxseg2ei16.v v8, (a0), v4, v0.t
 # CHECK-ENCODING: [0x07,0x54,0x45,0x24]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 54 45 24 <unknown>
+# CHECK-UNKNOWN: 24455407 <unknown>
 
 vluxseg2ei16.v v8, (a0), v4
 # CHECK-INST: vluxseg2ei16.v v8, (a0), v4
 # CHECK-ENCODING: [0x07,0x54,0x45,0x26]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 54 45 26 <unknown>
+# CHECK-UNKNOWN: 26455407 <unknown>
 
 vluxseg2ei32.v v8, (a0), v4, v0.t
 # CHECK-INST: vluxseg2ei32.v v8, (a0), v4, v0.t
 # CHECK-ENCODING: [0x07,0x64,0x45,0x24]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 64 45 24 <unknown>
+# CHECK-UNKNOWN: 24456407 <unknown>
 
 vluxseg2ei32.v v8, (a0), v4
 # CHECK-INST: vluxseg2ei32.v v8, (a0), v4
 # CHECK-ENCODING: [0x07,0x64,0x45,0x26]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 64 45 26 <unknown>
+# CHECK-UNKNOWN: 26456407 <unknown>
 
 vluxseg2ei64.v v8, (a0), v4, v0.t
 # CHECK-INST: vluxseg2ei64.v v8, (a0), v4, v0.t
 # CHECK-ENCODING: [0x07,0x74,0x45,0x24]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 74 45 24 <unknown>
+# CHECK-UNKNOWN: 24457407 <unknown>
 
 vluxseg2ei64.v v8, (a0), v4
 # CHECK-INST: vluxseg2ei64.v v8, (a0), v4
 # CHECK-ENCODING: [0x07,0x74,0x45,0x26]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 74 45 26 <unknown>
+# CHECK-UNKNOWN: 26457407 <unknown>
 
 vloxseg2ei8.v v8, (a0), v4, v0.t
 # CHECK-INST: vloxseg2ei8.v v8, (a0), v4, v0.t
 # CHECK-ENCODING: [0x07,0x04,0x45,0x2c]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 04 45 2c <unknown>
+# CHECK-UNKNOWN: 2c450407 <unknown>
 
 vloxseg2ei8.v v8, (a0), v4
 # CHECK-INST: vloxseg2ei8.v v8, (a0), v4
 # CHECK-ENCODING: [0x07,0x04,0x45,0x2e]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 04 45 2e <unknown>
+# CHECK-UNKNOWN: 2e450407 <unknown>
 
 vloxseg2ei16.v v8, (a0), v4, v0.t
 # CHECK-INST: vloxseg2ei16.v v8, (a0), v4, v0.t
 # CHECK-ENCODING: [0x07,0x54,0x45,0x2c]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 54 45 2c <unknown>
+# CHECK-UNKNOWN: 2c455407 <unknown>
 
 vloxseg2ei16.v v8, (a0), v4
 # CHECK-INST: vloxseg2ei16.v v8, (a0), v4
 # CHECK-ENCODING: [0x07,0x54,0x45,0x2e]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 54 45 2e <unknown>
+# CHECK-UNKNOWN: 2e455407 <unknown>
 
 vloxseg2ei32.v v8, (a0), v4, v0.t
 # CHECK-INST: vloxseg2ei32.v v8, (a0), v4, v0.t
 # CHECK-ENCODING: [0x07,0x64,0x45,0x2c]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 64 45 2c <unknown>
+# CHECK-UNKNOWN: 2c456407 <unknown>
 
 vloxseg2ei32.v v8, (a0), v4
 # CHECK-INST: vloxseg2ei32.v v8, (a0), v4
 # CHECK-ENCODING: [0x07,0x64,0x45,0x2e]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 64 45 2e <unknown>
+# CHECK-UNKNOWN: 2e456407 <unknown>
 
 vloxseg2ei64.v v8, (a0), v4, v0.t
 # CHECK-INST: vloxseg2ei64.v v8, (a0), v4, v0.t
 # CHECK-ENCODING: [0x07,0x74,0x45,0x2c]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 74 45 2c <unknown>
+# CHECK-UNKNOWN: 2c457407 <unknown>
 
 vloxseg2ei64.v v8, (a0), v4
 # CHECK-INST: vloxseg2ei64.v v8, (a0), v4
 # CHECK-ENCODING: [0x07,0x74,0x45,0x2e]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 74 45 2e <unknown>
+# CHECK-UNKNOWN: 2e457407 <unknown>
 
 vlseg3e8.v v8, (a0), v0.t
 # CHECK-INST: vlseg3e8.v v8, (a0), v0.t
 # CHECK-ENCODING: [0x07,0x04,0x05,0x40]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 04 05 40 <unknown>
+# CHECK-UNKNOWN: 40050407 <unknown>
 
 vlseg3e8.v v8, (a0)
 # CHECK-INST: vlseg3e8.v v8, (a0)
 # CHECK-ENCODING: [0x07,0x04,0x05,0x42]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 04 05 42 <unknown>
+# CHECK-UNKNOWN: 42050407 <unknown>
 
 vlseg3e16.v v8, (a0), v0.t
 # CHECK-INST: vlseg3e16.v v8, (a0), v0.t
 # CHECK-ENCODING: [0x07,0x54,0x05,0x40]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 54 05 40 <unknown>
+# CHECK-UNKNOWN: 40055407 <unknown>
 
 vlseg3e16.v v8, (a0)
 # CHECK-INST: vlseg3e16.v v8, (a0)
 # CHECK-ENCODING: [0x07,0x54,0x05,0x42]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 54 05 42 <unknown>
+# CHECK-UNKNOWN: 42055407 <unknown>
 
 vlseg3e32.v v8, (a0), v0.t
 # CHECK-INST: vlseg3e32.v v8, (a0), v0.t
 # CHECK-ENCODING: [0x07,0x64,0x05,0x40]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 64 05 40 <unknown>
+# CHECK-UNKNOWN: 40056407 <unknown>
 
 vlseg3e32.v v8, (a0)
 # CHECK-INST: vlseg3e32.v v8, (a0)
 # CHECK-ENCODING: [0x07,0x64,0x05,0x42]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 64 05 42 <unknown>
+# CHECK-UNKNOWN: 42056407 <unknown>
 
 vlseg3e64.v v8, (a0), v0.t
 # CHECK-INST: vlseg3e64.v v8, (a0), v0.t
 # CHECK-ENCODING: [0x07,0x74,0x05,0x40]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 74 05 40 <unknown>
+# CHECK-UNKNOWN: 40057407 <unknown>
 
 vlseg3e64.v v8, (a0)
 # CHECK-INST: vlseg3e64.v v8, (a0)
 # CHECK-ENCODING: [0x07,0x74,0x05,0x42]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 74 05 42 <unknown>
+# CHECK-UNKNOWN: 42057407 <unknown>
 
 vlseg3e8ff.v v8, (a0), v0.t
 # CHECK-INST: vlseg3e8ff.v v8, (a0), v0.t
 # CHECK-ENCODING: [0x07,0x04,0x05,0x41]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 04 05 41 <unknown>
+# CHECK-UNKNOWN: 41050407 <unknown>
 
 vlseg3e8ff.v v8, (a0)
 # CHECK-INST: vlseg3e8ff.v v8, (a0)
 # CHECK-ENCODING: [0x07,0x04,0x05,0x43]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 04 05 43 <unknown>
+# CHECK-UNKNOWN: 43050407 <unknown>
 
 vlseg3e16ff.v v8, (a0), v0.t
 # CHECK-INST: vlseg3e16ff.v v8, (a0), v0.t
 # CHECK-ENCODING: [0x07,0x54,0x05,0x41]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 54 05 41 <unknown>
+# CHECK-UNKNOWN: 41055407 <unknown>
 
 vlseg3e16ff.v v8, (a0)
 # CHECK-INST: vlseg3e16ff.v v8, (a0)
 # CHECK-ENCODING: [0x07,0x54,0x05,0x43]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 54 05 43 <unknown>
+# CHECK-UNKNOWN: 43055407 <unknown>
 
 vlseg3e32ff.v v8, (a0), v0.t
 # CHECK-INST: vlseg3e32ff.v v8, (a0), v0.t
 # CHECK-ENCODING: [0x07,0x64,0x05,0x41]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 64 05 41 <unknown>
+# CHECK-UNKNOWN: 41056407 <unknown>
 
 vlseg3e32ff.v v8, (a0)
 # CHECK-INST: vlseg3e32ff.v v8, (a0)
 # CHECK-ENCODING: [0x07,0x64,0x05,0x43]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 64 05 43 <unknown>
+# CHECK-UNKNOWN: 43056407 <unknown>
 
 vlseg3e64ff.v v8, (a0), v0.t
 # CHECK-INST: vlseg3e64ff.v v8, (a0), v0.t
 # CHECK-ENCODING: [0x07,0x74,0x05,0x41]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 74 05 41 <unknown>
+# CHECK-UNKNOWN: 41057407 <unknown>
 
 vlseg3e64ff.v v8, (a0)
 # CHECK-INST: vlseg3e64ff.v v8, (a0)
 # CHECK-ENCODING: [0x07,0x74,0x05,0x43]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 74 05 43 <unknown>
+# CHECK-UNKNOWN: 43057407 <unknown>
 
 vlsseg3e8.v v8, (a0), a1, v0.t
 # CHECK-INST: vlsseg3e8.v v8, (a0), a1, v0.t
 # CHECK-ENCODING: [0x07,0x04,0xb5,0x48]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 04 b5 48 <unknown>
+# CHECK-UNKNOWN: 48b50407 <unknown>
 
 vlsseg3e8.v v8, (a0), a1
 # CHECK-INST: vlsseg3e8.v v8, (a0), a1
 # CHECK-ENCODING: [0x07,0x04,0xb5,0x4a]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 04 b5 4a <unknown>
+# CHECK-UNKNOWN: 4ab50407 <unknown>
 
 vlsseg3e16.v v8, (a0), a1, v0.t
 # CHECK-INST: vlsseg3e16.v v8, (a0), a1, v0.t
 # CHECK-ENCODING: [0x07,0x54,0xb5,0x48]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 54 b5 48 <unknown>
+# CHECK-UNKNOWN: 48b55407 <unknown>
 
 vlsseg3e16.v v8, (a0), a1
 # CHECK-INST: vlsseg3e16.v v8, (a0), a1
 # CHECK-ENCODING: [0x07,0x54,0xb5,0x4a]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 54 b5 4a <unknown>
+# CHECK-UNKNOWN: 4ab55407 <unknown>
 
 vlsseg3e32.v v8, (a0), a1, v0.t
 # CHECK-INST: vlsseg3e32.v v8, (a0), a1, v0.t
 # CHECK-ENCODING: [0x07,0x64,0xb5,0x48]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 64 b5 48 <unknown>
+# CHECK-UNKNOWN: 48b56407 <unknown>
 
 vlsseg3e32.v v8, (a0), a1
 # CHECK-INST: vlsseg3e32.v v8, (a0), a1
 # CHECK-ENCODING: [0x07,0x64,0xb5,0x4a]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 64 b5 4a <unknown>
+# CHECK-UNKNOWN: 4ab56407 <unknown>
 
 vlsseg3e64.v v8, (a0), a1, v0.t
 # CHECK-INST: vlsseg3e64.v v8, (a0), a1, v0.t
 # CHECK-ENCODING: [0x07,0x74,0xb5,0x48]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 74 b5 48 <unknown>
+# CHECK-UNKNOWN: 48b57407 <unknown>
 
 vlsseg3e64.v v8, (a0), a1
 # CHECK-INST: vlsseg3e64.v v8, (a0), a1
 # CHECK-ENCODING: [0x07,0x74,0xb5,0x4a]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 74 b5 4a <unknown>
+# CHECK-UNKNOWN: 4ab57407 <unknown>
 
 vluxseg3ei8.v v8, (a0), v4, v0.t
 # CHECK-INST: vluxseg3ei8.v v8, (a0), v4, v0.t
 # CHECK-ENCODING: [0x07,0x04,0x45,0x44]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 04 45 44 <unknown>
+# CHECK-UNKNOWN: 44450407 <unknown>
 
 vluxseg3ei8.v v8, (a0), v4
 # CHECK-INST: vluxseg3ei8.v v8, (a0), v4
 # CHECK-ENCODING: [0x07,0x04,0x45,0x46]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 04 45 46 <unknown>
+# CHECK-UNKNOWN: 46450407 <unknown>
 
 vluxseg3ei16.v v8, (a0), v4, v0.t
 # CHECK-INST: vluxseg3ei16.v v8, (a0), v4, v0.t
 # CHECK-ENCODING: [0x07,0x54,0x45,0x44]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 54 45 44 <unknown>
+# CHECK-UNKNOWN: 44455407 <unknown>
 
 vluxseg3ei16.v v8, (a0), v4
 # CHECK-INST: vluxseg3ei16.v v8, (a0), v4
 # CHECK-ENCODING: [0x07,0x54,0x45,0x46]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 54 45 46 <unknown>
+# CHECK-UNKNOWN: 46455407 <unknown>
 
 vluxseg3ei32.v v8, (a0), v4, v0.t
 # CHECK-INST: vluxseg3ei32.v v8, (a0), v4, v0.t
 # CHECK-ENCODING: [0x07,0x64,0x45,0x44]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 64 45 44 <unknown>
+# CHECK-UNKNOWN: 44456407 <unknown>
 
 vluxseg3ei32.v v8, (a0), v4
 # CHECK-INST: vluxseg3ei32.v v8, (a0), v4
 # CHECK-ENCODING: [0x07,0x64,0x45,0x46]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 64 45 46 <unknown>
+# CHECK-UNKNOWN: 46456407 <unknown>
 
 vluxseg3ei64.v v8, (a0), v4, v0.t
 # CHECK-INST: vluxseg3ei64.v v8, (a0), v4, v0.t
 # CHECK-ENCODING: [0x07,0x74,0x45,0x44]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 74 45 44 <unknown>
+# CHECK-UNKNOWN: 44457407 <unknown>
 
 vluxseg3ei64.v v8, (a0), v4
 # CHECK-INST: vluxseg3ei64.v v8, (a0), v4
 # CHECK-ENCODING: [0x07,0x74,0x45,0x46]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 74 45 46 <unknown>
+# CHECK-UNKNOWN: 46457407 <unknown>
 
 vloxseg3ei8.v v8, (a0), v4, v0.t
 # CHECK-INST: vloxseg3ei8.v v8, (a0), v4, v0.t
 # CHECK-ENCODING: [0x07,0x04,0x45,0x4c]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 04 45 4c <unknown>
+# CHECK-UNKNOWN: 4c450407 <unknown>
 
 vloxseg3ei8.v v8, (a0), v4
 # CHECK-INST: vloxseg3ei8.v v8, (a0), v4
 # CHECK-ENCODING: [0x07,0x04,0x45,0x4e]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 04 45 4e <unknown>
+# CHECK-UNKNOWN: 4e450407 <unknown>
 
 vloxseg3ei16.v v8, (a0), v4, v0.t
 # CHECK-INST: vloxseg3ei16.v v8, (a0), v4, v0.t
 # CHECK-ENCODING: [0x07,0x54,0x45,0x4c]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 54 45 4c <unknown>
+# CHECK-UNKNOWN: 4c455407 <unknown>
 
 vloxseg3ei16.v v8, (a0), v4
 # CHECK-INST: vloxseg3ei16.v v8, (a0), v4
 # CHECK-ENCODING: [0x07,0x54,0x45,0x4e]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 54 45 4e <unknown>
+# CHECK-UNKNOWN: 4e455407 <unknown>
 
 vloxseg3ei32.v v8, (a0), v4, v0.t
 # CHECK-INST: vloxseg3ei32.v v8, (a0), v4, v0.t
 # CHECK-ENCODING: [0x07,0x64,0x45,0x4c]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 64 45 4c <unknown>
+# CHECK-UNKNOWN: 4c456407 <unknown>
 
 vloxseg3ei32.v v8, (a0), v4
 # CHECK-INST: vloxseg3ei32.v v8, (a0), v4
 # CHECK-ENCODING: [0x07,0x64,0x45,0x4e]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 64 45 4e <unknown>
+# CHECK-UNKNOWN: 4e456407 <unknown>
 
 vloxseg3ei64.v v8, (a0), v4, v0.t
 # CHECK-INST: vloxseg3ei64.v v8, (a0), v4, v0.t
 # CHECK-ENCODING: [0x07,0x74,0x45,0x4c]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 74 45 4c <unknown>
+# CHECK-UNKNOWN: 4c457407 <unknown>
 
 vloxseg3ei64.v v8, (a0), v4
 # CHECK-INST: vloxseg3ei64.v v8, (a0), v4
 # CHECK-ENCODING: [0x07,0x74,0x45,0x4e]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 74 45 4e <unknown>
+# CHECK-UNKNOWN: 4e457407 <unknown>
 
 vlseg4e8.v v8, (a0), v0.t
 # CHECK-INST: vlseg4e8.v v8, (a0), v0.t
 # CHECK-ENCODING: [0x07,0x04,0x05,0x60]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 04 05 60 <unknown>
+# CHECK-UNKNOWN: 60050407 <unknown>
 
 vlseg4e8.v v8, (a0)
 # CHECK-INST: vlseg4e8.v v8, (a0)
 # CHECK-ENCODING: [0x07,0x04,0x05,0x62]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 04 05 62 <unknown>
+# CHECK-UNKNOWN: 62050407 <unknown>
 
 vlseg4e16.v v8, (a0), v0.t
 # CHECK-INST: vlseg4e16.v v8, (a0), v0.t
 # CHECK-ENCODING: [0x07,0x54,0x05,0x60]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 54 05 60 <unknown>
+# CHECK-UNKNOWN: 60055407 <unknown>
 
 vlseg4e16.v v8, (a0)
 # CHECK-INST: vlseg4e16.v v8, (a0)
 # CHECK-ENCODING: [0x07,0x54,0x05,0x62]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 54 05 62 <unknown>
+# CHECK-UNKNOWN: 62055407 <unknown>
 
 vlseg4e32.v v8, (a0), v0.t
 # CHECK-INST: vlseg4e32.v v8, (a0), v0.t
 # CHECK-ENCODING: [0x07,0x64,0x05,0x60]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 64 05 60 <unknown>
+# CHECK-UNKNOWN: 60056407 <unknown>
 
 vlseg4e32.v v8, (a0)
 # CHECK-INST: vlseg4e32.v v8, (a0)
 # CHECK-ENCODING: [0x07,0x64,0x05,0x62]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 64 05 62 <unknown>
+# CHECK-UNKNOWN: 62056407 <unknown>
 
 vlseg4e64.v v8, (a0), v0.t
 # CHECK-INST: vlseg4e64.v v8, (a0), v0.t
 # CHECK-ENCODING: [0x07,0x74,0x05,0x60]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 74 05 60 <unknown>
+# CHECK-UNKNOWN: 60057407 <unknown>
 
 vlseg4e64.v v8, (a0)
 # CHECK-INST: vlseg4e64.v v8, (a0)
 # CHECK-ENCODING: [0x07,0x74,0x05,0x62]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 74 05 62 <unknown>
+# CHECK-UNKNOWN: 62057407 <unknown>
 
 vlseg4e8ff.v v8, (a0), v0.t
 # CHECK-INST: vlseg4e8ff.v v8, (a0), v0.t
 # CHECK-ENCODING: [0x07,0x04,0x05,0x61]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 04 05 61 <unknown>
+# CHECK-UNKNOWN: 61050407 <unknown>
 
 vlseg4e8ff.v v8, (a0)
 # CHECK-INST: vlseg4e8ff.v v8, (a0)
 # CHECK-ENCODING: [0x07,0x04,0x05,0x63]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 04 05 63 <unknown>
+# CHECK-UNKNOWN: 63050407 <unknown>
 
 vlseg4e16ff.v v8, (a0), v0.t
 # CHECK-INST: vlseg4e16ff.v v8, (a0), v0.t
 # CHECK-ENCODING: [0x07,0x54,0x05,0x61]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 54 05 61 <unknown>
+# CHECK-UNKNOWN: 61055407 <unknown>
 
 vlseg4e16ff.v v8, (a0)
 # CHECK-INST: vlseg4e16ff.v v8, (a0)
 # CHECK-ENCODING: [0x07,0x54,0x05,0x63]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 54 05 63 <unknown>
+# CHECK-UNKNOWN: 63055407 <unknown>
 
 vlseg4e32ff.v v8, (a0), v0.t
 # CHECK-INST: vlseg4e32ff.v v8, (a0), v0.t
 # CHECK-ENCODING: [0x07,0x64,0x05,0x61]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 64 05 61 <unknown>
+# CHECK-UNKNOWN: 61056407 <unknown>
 
 vlseg4e32ff.v v8, (a0)
 # CHECK-INST: vlseg4e32ff.v v8, (a0)
 # CHECK-ENCODING: [0x07,0x64,0x05,0x63]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 64 05 63 <unknown>
+# CHECK-UNKNOWN: 63056407 <unknown>
 
 vlseg4e64ff.v v8, (a0), v0.t
 # CHECK-INST: vlseg4e64ff.v v8, (a0), v0.t
 # CHECK-ENCODING: [0x07,0x74,0x05,0x61]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 74 05 61 <unknown>
+# CHECK-UNKNOWN: 61057407 <unknown>
 
 vlseg4e64ff.v v8, (a0)
 # CHECK-INST: vlseg4e64ff.v v8, (a0)
 # CHECK-ENCODING: [0x07,0x74,0x05,0x63]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 74 05 63 <unknown>
+# CHECK-UNKNOWN: 63057407 <unknown>
 
 vlsseg4e8.v v8, (a0), a1, v0.t
 # CHECK-INST: vlsseg4e8.v v8, (a0), a1, v0.t
 # CHECK-ENCODING: [0x07,0x04,0xb5,0x68]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 04 b5 68 <unknown>
+# CHECK-UNKNOWN: 68b50407 <unknown>
 
 vlsseg4e8.v v8, (a0), a1
 # CHECK-INST: vlsseg4e8.v v8, (a0), a1
 # CHECK-ENCODING: [0x07,0x04,0xb5,0x6a]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 04 b5 6a <unknown>
+# CHECK-UNKNOWN: 6ab50407 <unknown>
 
 vlsseg4e16.v v8, (a0), a1, v0.t
 # CHECK-INST: vlsseg4e16.v v8, (a0), a1, v0.t
 # CHECK-ENCODING: [0x07,0x54,0xb5,0x68]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 54 b5 68 <unknown>
+# CHECK-UNKNOWN: 68b55407 <unknown>
 
 vlsseg4e16.v v8, (a0), a1
 # CHECK-INST: vlsseg4e16.v v8, (a0), a1
 # CHECK-ENCODING: [0x07,0x54,0xb5,0x6a]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 54 b5 6a <unknown>
+# CHECK-UNKNOWN: 6ab55407 <unknown>
 
 vlsseg4e32.v v8, (a0), a1, v0.t
 # CHECK-INST: vlsseg4e32.v v8, (a0), a1, v0.t
 # CHECK-ENCODING: [0x07,0x64,0xb5,0x68]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 64 b5 68 <unknown>
+# CHECK-UNKNOWN: 68b56407 <unknown>
 
 vlsseg4e32.v v8, (a0), a1
 # CHECK-INST: vlsseg4e32.v v8, (a0), a1
 # CHECK-ENCODING: [0x07,0x64,0xb5,0x6a]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 64 b5 6a <unknown>
+# CHECK-UNKNOWN: 6ab56407 <unknown>
 
 vlsseg4e64.v v8, (a0), a1, v0.t
 # CHECK-INST: vlsseg4e64.v v8, (a0), a1, v0.t
 # CHECK-ENCODING: [0x07,0x74,0xb5,0x68]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 74 b5 68 <unknown>
+# CHECK-UNKNOWN: 68b57407 <unknown>
 
 vlsseg4e64.v v8, (a0), a1
 # CHECK-INST: vlsseg4e64.v v8, (a0), a1
 # CHECK-ENCODING: [0x07,0x74,0xb5,0x6a]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 74 b5 6a <unknown>
+# CHECK-UNKNOWN: 6ab57407 <unknown>
 
 vluxseg4ei8.v v8, (a0), v4, v0.t
 # CHECK-INST: vluxseg4ei8.v v8, (a0), v4, v0.t
 # CHECK-ENCODING: [0x07,0x04,0x45,0x64]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 04 45 64 <unknown>
+# CHECK-UNKNOWN: 64450407 <unknown>
 
 vluxseg4ei8.v v8, (a0), v4
 # CHECK-INST: vluxseg4ei8.v v8, (a0), v4
 # CHECK-ENCODING: [0x07,0x04,0x45,0x66]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 04 45 66 <unknown>
+# CHECK-UNKNOWN: 66450407 <unknown>
 
 vluxseg4ei16.v v8, (a0), v4, v0.t
 # CHECK-INST: vluxseg4ei16.v v8, (a0), v4, v0.t
 # CHECK-ENCODING: [0x07,0x54,0x45,0x64]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 54 45 64 <unknown>
+# CHECK-UNKNOWN: 64455407 <unknown>
 
 vluxseg4ei16.v v8, (a0), v4
 # CHECK-INST: vluxseg4ei16.v v8, (a0), v4
 # CHECK-ENCODING: [0x07,0x54,0x45,0x66]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 54 45 66 <unknown>
+# CHECK-UNKNOWN: 66455407 <unknown>
 
 vluxseg4ei32.v v8, (a0), v4, v0.t
 # CHECK-INST: vluxseg4ei32.v v8, (a0), v4, v0.t
 # CHECK-ENCODING: [0x07,0x64,0x45,0x64]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 64 45 64 <unknown>
+# CHECK-UNKNOWN: 64456407 <unknown>
 
 vluxseg4ei32.v v8, (a0), v4
 # CHECK-INST: vluxseg4ei32.v v8, (a0), v4
 # CHECK-ENCODING: [0x07,0x64,0x45,0x66]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 64 45 66 <unknown>
+# CHECK-UNKNOWN: 66456407 <unknown>
 
 vluxseg4ei64.v v8, (a0), v4, v0.t
 # CHECK-INST: vluxseg4ei64.v v8, (a0), v4, v0.t
 # CHECK-ENCODING: [0x07,0x74,0x45,0x64]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 74 45 64 <unknown>
+# CHECK-UNKNOWN: 64457407 <unknown>
 
 vluxseg4ei64.v v8, (a0), v4
 # CHECK-INST: vluxseg4ei64.v v8, (a0), v4
 # CHECK-ENCODING: [0x07,0x74,0x45,0x66]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 74 45 66 <unknown>
+# CHECK-UNKNOWN: 66457407 <unknown>
 
 vloxseg4ei8.v v8, (a0), v4, v0.t
 # CHECK-INST: vloxseg4ei8.v v8, (a0), v4, v0.t
 # CHECK-ENCODING: [0x07,0x04,0x45,0x6c]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 04 45 6c <unknown>
+# CHECK-UNKNOWN: 6c450407 <unknown>
 
 vloxseg4ei8.v v8, (a0), v4
 # CHECK-INST: vloxseg4ei8.v v8, (a0), v4
 # CHECK-ENCODING: [0x07,0x04,0x45,0x6e]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 04 45 6e <unknown>
+# CHECK-UNKNOWN: 6e450407 <unknown>
 
 vloxseg4ei16.v v8, (a0), v4, v0.t
 # CHECK-INST: vloxseg4ei16.v v8, (a0), v4, v0.t
 # CHECK-ENCODING: [0x07,0x54,0x45,0x6c]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 54 45 6c <unknown>
+# CHECK-UNKNOWN: 6c455407 <unknown>
 
 vloxseg4ei16.v v8, (a0), v4
 # CHECK-INST: vloxseg4ei16.v v8, (a0), v4
 # CHECK-ENCODING: [0x07,0x54,0x45,0x6e]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 54 45 6e <unknown>
+# CHECK-UNKNOWN: 6e455407 <unknown>
 
 vloxseg4ei32.v v8, (a0), v4, v0.t
 # CHECK-INST: vloxseg4ei32.v v8, (a0), v4, v0.t
 # CHECK-ENCODING: [0x07,0x64,0x45,0x6c]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 64 45 6c <unknown>
+# CHECK-UNKNOWN: 6c456407 <unknown>
 
 vloxseg4ei32.v v8, (a0), v4
 # CHECK-INST: vloxseg4ei32.v v8, (a0), v4
 # CHECK-ENCODING: [0x07,0x64,0x45,0x6e]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 64 45 6e <unknown>
+# CHECK-UNKNOWN: 6e456407 <unknown>
 
 vloxseg4ei64.v v8, (a0), v4, v0.t
 # CHECK-INST: vloxseg4ei64.v v8, (a0), v4, v0.t
 # CHECK-ENCODING: [0x07,0x74,0x45,0x6c]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 74 45 6c <unknown>
+# CHECK-UNKNOWN: 6c457407 <unknown>
 
 vloxseg4ei64.v v8, (a0), v4
 # CHECK-INST: vloxseg4ei64.v v8, (a0), v4
 # CHECK-ENCODING: [0x07,0x74,0x45,0x6e]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 74 45 6e <unknown>
+# CHECK-UNKNOWN: 6e457407 <unknown>
 
 vlseg5e8.v v8, (a0), v0.t
 # CHECK-INST: vlseg5e8.v v8, (a0), v0.t
 # CHECK-ENCODING: [0x07,0x04,0x05,0x80]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 04 05 80 <unknown>
+# CHECK-UNKNOWN: 80050407 <unknown>
 
 vlseg5e8.v v8, (a0)
 # CHECK-INST: vlseg5e8.v v8, (a0)
 # CHECK-ENCODING: [0x07,0x04,0x05,0x82]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 04 05 82 <unknown>
+# CHECK-UNKNOWN: 82050407 <unknown>
 
 vlseg5e16.v v8, (a0), v0.t
 # CHECK-INST: vlseg5e16.v v8, (a0), v0.t
 # CHECK-ENCODING: [0x07,0x54,0x05,0x80]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 54 05 80 <unknown>
+# CHECK-UNKNOWN: 80055407 <unknown>
 
 vlseg5e16.v v8, (a0)
 # CHECK-INST: vlseg5e16.v v8, (a0)
 # CHECK-ENCODING: [0x07,0x54,0x05,0x82]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 54 05 82 <unknown>
+# CHECK-UNKNOWN: 82055407 <unknown>
 
 vlseg5e32.v v8, (a0), v0.t
 # CHECK-INST: vlseg5e32.v v8, (a0), v0.t
 # CHECK-ENCODING: [0x07,0x64,0x05,0x80]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 64 05 80 <unknown>
+# CHECK-UNKNOWN: 80056407 <unknown>
 
 vlseg5e32.v v8, (a0)
 # CHECK-INST: vlseg5e32.v v8, (a0)
 # CHECK-ENCODING: [0x07,0x64,0x05,0x82]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 64 05 82 <unknown>
+# CHECK-UNKNOWN: 82056407 <unknown>
 
 vlseg5e64.v v8, (a0), v0.t
 # CHECK-INST: vlseg5e64.v v8, (a0), v0.t
 # CHECK-ENCODING: [0x07,0x74,0x05,0x80]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 74 05 80 <unknown>
+# CHECK-UNKNOWN: 80057407 <unknown>
 
 vlseg5e64.v v8, (a0)
 # CHECK-INST: vlseg5e64.v v8, (a0)
 # CHECK-ENCODING: [0x07,0x74,0x05,0x82]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 74 05 82 <unknown>
+# CHECK-UNKNOWN: 82057407 <unknown>
 
 vlseg5e8ff.v v8, (a0), v0.t
 # CHECK-INST: vlseg5e8ff.v v8, (a0), v0.t
 # CHECK-ENCODING: [0x07,0x04,0x05,0x81]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 04 05 81 <unknown>
+# CHECK-UNKNOWN: 81050407 <unknown>
 
 vlseg5e8ff.v v8, (a0)
 # CHECK-INST: vlseg5e8ff.v v8, (a0)
 # CHECK-ENCODING: [0x07,0x04,0x05,0x83]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 04 05 83 <unknown>
+# CHECK-UNKNOWN: 83050407 <unknown>
 
 vlseg5e16ff.v v8, (a0), v0.t
 # CHECK-INST: vlseg5e16ff.v v8, (a0), v0.t
 # CHECK-ENCODING: [0x07,0x54,0x05,0x81]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 54 05 81 <unknown>
+# CHECK-UNKNOWN: 81055407 <unknown>
 
 vlseg5e16ff.v v8, (a0)
 # CHECK-INST: vlseg5e16ff.v v8, (a0)
 # CHECK-ENCODING: [0x07,0x54,0x05,0x83]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 54 05 83 <unknown>
+# CHECK-UNKNOWN: 83055407 <unknown>
 
 vlseg5e32ff.v v8, (a0), v0.t
 # CHECK-INST: vlseg5e32ff.v v8, (a0), v0.t
 # CHECK-ENCODING: [0x07,0x64,0x05,0x81]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 64 05 81 <unknown>
+# CHECK-UNKNOWN: 81056407 <unknown>
 
 vlseg5e32ff.v v8, (a0)
 # CHECK-INST: vlseg5e32ff.v v8, (a0)
 # CHECK-ENCODING: [0x07,0x64,0x05,0x83]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 64 05 83 <unknown>
+# CHECK-UNKNOWN: 83056407 <unknown>
 
 vlseg5e64ff.v v8, (a0), v0.t
 # CHECK-INST: vlseg5e64ff.v v8, (a0), v0.t
 # CHECK-ENCODING: [0x07,0x74,0x05,0x81]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 74 05 81 <unknown>
+# CHECK-UNKNOWN: 81057407 <unknown>
 
 vlseg5e64ff.v v8, (a0)
 # CHECK-INST: vlseg5e64ff.v v8, (a0)
 # CHECK-ENCODING: [0x07,0x74,0x05,0x83]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 74 05 83 <unknown>
+# CHECK-UNKNOWN: 83057407 <unknown>
 
 vlsseg5e8.v v8, (a0), a1, v0.t
 # CHECK-INST: vlsseg5e8.v v8, (a0), a1, v0.t
 # CHECK-ENCODING: [0x07,0x04,0xb5,0x88]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 04 b5 88 <unknown>
+# CHECK-UNKNOWN: 88b50407 <unknown>
 
 vlsseg5e8.v v8, (a0), a1
 # CHECK-INST: vlsseg5e8.v v8, (a0), a1
 # CHECK-ENCODING: [0x07,0x04,0xb5,0x8a]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 04 b5 8a <unknown>
+# CHECK-UNKNOWN: 8ab50407 <unknown>
 
 vlsseg5e16.v v8, (a0), a1, v0.t
 # CHECK-INST: vlsseg5e16.v v8, (a0), a1, v0.t
 # CHECK-ENCODING: [0x07,0x54,0xb5,0x88]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 54 b5 88 <unknown>
+# CHECK-UNKNOWN: 88b55407 <unknown>
 
 vlsseg5e16.v v8, (a0), a1
 # CHECK-INST: vlsseg5e16.v v8, (a0), a1
 # CHECK-ENCODING: [0x07,0x54,0xb5,0x8a]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 54 b5 8a <unknown>
+# CHECK-UNKNOWN: 8ab55407 <unknown>
 
 vlsseg5e32.v v8, (a0), a1, v0.t
 # CHECK-INST: vlsseg5e32.v v8, (a0), a1, v0.t
 # CHECK-ENCODING: [0x07,0x64,0xb5,0x88]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 64 b5 88 <unknown>
+# CHECK-UNKNOWN: 88b56407 <unknown>
 
 vlsseg5e32.v v8, (a0), a1
 # CHECK-INST: vlsseg5e32.v v8, (a0), a1
 # CHECK-ENCODING: [0x07,0x64,0xb5,0x8a]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 64 b5 8a <unknown>
+# CHECK-UNKNOWN: 8ab56407 <unknown>
 
 vlsseg5e64.v v8, (a0), a1, v0.t
 # CHECK-INST: vlsseg5e64.v v8, (a0), a1, v0.t
 # CHECK-ENCODING: [0x07,0x74,0xb5,0x88]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 74 b5 88 <unknown>
+# CHECK-UNKNOWN: 88b57407 <unknown>
 
 vlsseg5e64.v v8, (a0), a1
 # CHECK-INST: vlsseg5e64.v v8, (a0), a1
 # CHECK-ENCODING: [0x07,0x74,0xb5,0x8a]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 74 b5 8a <unknown>
+# CHECK-UNKNOWN: 8ab57407 <unknown>
 
 vluxseg5ei8.v v8, (a0), v4, v0.t
 # CHECK-INST: vluxseg5ei8.v v8, (a0), v4, v0.t
 # CHECK-ENCODING: [0x07,0x04,0x45,0x84]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 04 45 84 <unknown>
+# CHECK-UNKNOWN: 84450407 <unknown>
 
 vluxseg5ei8.v v8, (a0), v4
 # CHECK-INST: vluxseg5ei8.v v8, (a0), v4
 # CHECK-ENCODING: [0x07,0x04,0x45,0x86]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 04 45 86 <unknown>
+# CHECK-UNKNOWN: 86450407 <unknown>
 
 vluxseg5ei16.v v8, (a0), v4, v0.t
 # CHECK-INST: vluxseg5ei16.v v8, (a0), v4, v0.t
 # CHECK-ENCODING: [0x07,0x54,0x45,0x84]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 54 45 84 <unknown>
+# CHECK-UNKNOWN: 84455407 <unknown>
 
 vluxseg5ei16.v v8, (a0), v4
 # CHECK-INST: vluxseg5ei16.v v8, (a0), v4
 # CHECK-ENCODING: [0x07,0x54,0x45,0x86]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 54 45 86 <unknown>
+# CHECK-UNKNOWN: 86455407 <unknown>
 
 vluxseg5ei32.v v8, (a0), v4, v0.t
 # CHECK-INST: vluxseg5ei32.v v8, (a0), v4, v0.t
 # CHECK-ENCODING: [0x07,0x64,0x45,0x84]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 64 45 84 <unknown>
+# CHECK-UNKNOWN: 84456407 <unknown>
 
 vluxseg5ei32.v v8, (a0), v4
 # CHECK-INST: vluxseg5ei32.v v8, (a0), v4
 # CHECK-ENCODING: [0x07,0x64,0x45,0x86]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 64 45 86 <unknown>
+# CHECK-UNKNOWN: 86456407 <unknown>
 
 vluxseg5ei64.v v8, (a0), v4, v0.t
 # CHECK-INST: vluxseg5ei64.v v8, (a0), v4, v0.t
 # CHECK-ENCODING: [0x07,0x74,0x45,0x84]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 74 45 84 <unknown>
+# CHECK-UNKNOWN: 84457407 <unknown>
 
 vluxseg5ei64.v v8, (a0), v4
 # CHECK-INST: vluxseg5ei64.v v8, (a0), v4
 # CHECK-ENCODING: [0x07,0x74,0x45,0x86]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 74 45 86 <unknown>
+# CHECK-UNKNOWN: 86457407 <unknown>
 
 vloxseg5ei8.v v8, (a0), v4, v0.t
 # CHECK-INST: vloxseg5ei8.v v8, (a0), v4, v0.t
 # CHECK-ENCODING: [0x07,0x04,0x45,0x8c]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 04 45 8c <unknown>
+# CHECK-UNKNOWN: 8c450407 <unknown>
 
 vloxseg5ei8.v v8, (a0), v4
 # CHECK-INST: vloxseg5ei8.v v8, (a0), v4
 # CHECK-ENCODING: [0x07,0x04,0x45,0x8e]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 04 45 8e <unknown>
+# CHECK-UNKNOWN: 8e450407 <unknown>
 
 vloxseg5ei16.v v8, (a0), v4, v0.t
 # CHECK-INST: vloxseg5ei16.v v8, (a0), v4, v0.t
 # CHECK-ENCODING: [0x07,0x54,0x45,0x8c]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 54 45 8c <unknown>
+# CHECK-UNKNOWN: 8c455407 <unknown>
 
 vloxseg5ei16.v v8, (a0), v4
 # CHECK-INST: vloxseg5ei16.v v8, (a0), v4
 # CHECK-ENCODING: [0x07,0x54,0x45,0x8e]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 54 45 8e <unknown>
+# CHECK-UNKNOWN: 8e455407 <unknown>
 
 vloxseg5ei32.v v8, (a0), v4, v0.t
 # CHECK-INST: vloxseg5ei32.v v8, (a0), v4, v0.t
 # CHECK-ENCODING: [0x07,0x64,0x45,0x8c]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 64 45 8c <unknown>
+# CHECK-UNKNOWN: 8c456407 <unknown>
 
 vloxseg5ei32.v v8, (a0), v4
 # CHECK-INST: vloxseg5ei32.v v8, (a0), v4
 # CHECK-ENCODING: [0x07,0x64,0x45,0x8e]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 64 45 8e <unknown>
+# CHECK-UNKNOWN: 8e456407 <unknown>
 
 vloxseg5ei64.v v8, (a0), v4, v0.t
 # CHECK-INST: vloxseg5ei64.v v8, (a0), v4, v0.t
 # CHECK-ENCODING: [0x07,0x74,0x45,0x8c]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 74 45 8c <unknown>
+# CHECK-UNKNOWN: 8c457407 <unknown>
 
 vloxseg5ei64.v v8, (a0), v4
 # CHECK-INST: vloxseg5ei64.v v8, (a0), v4
 # CHECK-ENCODING: [0x07,0x74,0x45,0x8e]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 74 45 8e <unknown>
+# CHECK-UNKNOWN: 8e457407 <unknown>
 
 vlseg6e8.v v8, (a0), v0.t
 # CHECK-INST: vlseg6e8.v v8, (a0), v0.t
 # CHECK-ENCODING: [0x07,0x04,0x05,0xa0]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 04 05 a0 <unknown>
+# CHECK-UNKNOWN: a0050407 <unknown>
 
 vlseg6e8.v v8, (a0)
 # CHECK-INST: vlseg6e8.v v8, (a0)
 # CHECK-ENCODING: [0x07,0x04,0x05,0xa2]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 04 05 a2 <unknown>
+# CHECK-UNKNOWN: a2050407 <unknown>
 
 vlseg6e16.v v8, (a0), v0.t
 # CHECK-INST: vlseg6e16.v v8, (a0), v0.t
 # CHECK-ENCODING: [0x07,0x54,0x05,0xa0]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 54 05 a0 <unknown>
+# CHECK-UNKNOWN: a0055407 <unknown>
 
 vlseg6e16.v v8, (a0)
 # CHECK-INST: vlseg6e16.v v8, (a0)
 # CHECK-ENCODING: [0x07,0x54,0x05,0xa2]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 54 05 a2 <unknown>
+# CHECK-UNKNOWN: a2055407 <unknown>
 
 vlseg6e32.v v8, (a0), v0.t
 # CHECK-INST: vlseg6e32.v v8, (a0), v0.t
 # CHECK-ENCODING: [0x07,0x64,0x05,0xa0]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 64 05 a0 <unknown>
+# CHECK-UNKNOWN: a0056407 <unknown>
 
 vlseg6e32.v v8, (a0)
 # CHECK-INST: vlseg6e32.v v8, (a0)
 # CHECK-ENCODING: [0x07,0x64,0x05,0xa2]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 64 05 a2 <unknown>
+# CHECK-UNKNOWN: a2056407 <unknown>
 
 vlseg6e64.v v8, (a0), v0.t
 # CHECK-INST: vlseg6e64.v v8, (a0), v0.t
 # CHECK-ENCODING: [0x07,0x74,0x05,0xa0]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 74 05 a0 <unknown>
+# CHECK-UNKNOWN: a0057407 <unknown>
 
 vlseg6e64.v v8, (a0)
 # CHECK-INST: vlseg6e64.v v8, (a0)
 # CHECK-ENCODING: [0x07,0x74,0x05,0xa2]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 74 05 a2 <unknown>
+# CHECK-UNKNOWN: a2057407 <unknown>
 
 vlseg6e8ff.v v8, (a0), v0.t
 # CHECK-INST: vlseg6e8ff.v v8, (a0), v0.t
 # CHECK-ENCODING: [0x07,0x04,0x05,0xa1]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 04 05 a1 <unknown>
+# CHECK-UNKNOWN: a1050407 <unknown>
 
 vlseg6e8ff.v v8, (a0)
 # CHECK-INST: vlseg6e8ff.v v8, (a0)
 # CHECK-ENCODING: [0x07,0x04,0x05,0xa3]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 04 05 a3 <unknown>
+# CHECK-UNKNOWN: a3050407 <unknown>
 
 vlseg6e16ff.v v8, (a0), v0.t
 # CHECK-INST: vlseg6e16ff.v v8, (a0), v0.t
 # CHECK-ENCODING: [0x07,0x54,0x05,0xa1]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 54 05 a1 <unknown>
+# CHECK-UNKNOWN: a1055407 <unknown>
 
 vlseg6e16ff.v v8, (a0)
 # CHECK-INST: vlseg6e16ff.v v8, (a0)
 # CHECK-ENCODING: [0x07,0x54,0x05,0xa3]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 54 05 a3 <unknown>
+# CHECK-UNKNOWN: a3055407 <unknown>
 
 vlseg6e32ff.v v8, (a0), v0.t
 # CHECK-INST: vlseg6e32ff.v v8, (a0), v0.t
 # CHECK-ENCODING: [0x07,0x64,0x05,0xa1]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 64 05 a1 <unknown>
+# CHECK-UNKNOWN: a1056407 <unknown>
 
 vlseg6e32ff.v v8, (a0)
 # CHECK-INST: vlseg6e32ff.v v8, (a0)
 # CHECK-ENCODING: [0x07,0x64,0x05,0xa3]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 64 05 a3 <unknown>
+# CHECK-UNKNOWN: a3056407 <unknown>
 
 vlseg6e64ff.v v8, (a0), v0.t
 # CHECK-INST: vlseg6e64ff.v v8, (a0), v0.t
 # CHECK-ENCODING: [0x07,0x74,0x05,0xa1]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 74 05 a1 <unknown>
+# CHECK-UNKNOWN: a1057407 <unknown>
 
 vlseg6e64ff.v v8, (a0)
 # CHECK-INST: vlseg6e64ff.v v8, (a0)
 # CHECK-ENCODING: [0x07,0x74,0x05,0xa3]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 74 05 a3 <unknown>
+# CHECK-UNKNOWN: a3057407 <unknown>
 
 vlsseg6e8.v v8, (a0), a1, v0.t
 # CHECK-INST: vlsseg6e8.v v8, (a0), a1, v0.t
 # CHECK-ENCODING: [0x07,0x04,0xb5,0xa8]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 04 b5 a8 <unknown>
+# CHECK-UNKNOWN: a8b50407 <unknown>
 
 vlsseg6e8.v v8, (a0), a1
 # CHECK-INST: vlsseg6e8.v v8, (a0), a1
 # CHECK-ENCODING: [0x07,0x04,0xb5,0xaa]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 04 b5 aa <unknown>
+# CHECK-UNKNOWN: aab50407 <unknown>
 
 vlsseg6e16.v v8, (a0), a1, v0.t
 # CHECK-INST: vlsseg6e16.v v8, (a0), a1, v0.t
 # CHECK-ENCODING: [0x07,0x54,0xb5,0xa8]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 54 b5 a8 <unknown>
+# CHECK-UNKNOWN: a8b55407 <unknown>
 
 vlsseg6e16.v v8, (a0), a1
 # CHECK-INST: vlsseg6e16.v v8, (a0), a1
 # CHECK-ENCODING: [0x07,0x54,0xb5,0xaa]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 54 b5 aa <unknown>
+# CHECK-UNKNOWN: aab55407 <unknown>
 
 vlsseg6e32.v v8, (a0), a1, v0.t
 # CHECK-INST: vlsseg6e32.v v8, (a0), a1, v0.t
 # CHECK-ENCODING: [0x07,0x64,0xb5,0xa8]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 64 b5 a8 <unknown>
+# CHECK-UNKNOWN: a8b56407 <unknown>
 
 vlsseg6e32.v v8, (a0), a1
 # CHECK-INST: vlsseg6e32.v v8, (a0), a1
 # CHECK-ENCODING: [0x07,0x64,0xb5,0xaa]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 64 b5 aa <unknown>
+# CHECK-UNKNOWN: aab56407 <unknown>
 
 vlsseg6e64.v v8, (a0), a1, v0.t
 # CHECK-INST: vlsseg6e64.v v8, (a0), a1, v0.t
 # CHECK-ENCODING: [0x07,0x74,0xb5,0xa8]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 74 b5 a8 <unknown>
+# CHECK-UNKNOWN: a8b57407 <unknown>
 
 vlsseg6e64.v v8, (a0), a1
 # CHECK-INST: vlsseg6e64.v v8, (a0), a1
 # CHECK-ENCODING: [0x07,0x74,0xb5,0xaa]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 74 b5 aa <unknown>
+# CHECK-UNKNOWN: aab57407 <unknown>
 
 vluxseg6ei8.v v8, (a0), v4, v0.t
 # CHECK-INST: vluxseg6ei8.v v8, (a0), v4, v0.t
 # CHECK-ENCODING: [0x07,0x04,0x45,0xa4]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 04 45 a4 <unknown>
+# CHECK-UNKNOWN: a4450407 <unknown>
 
 vluxseg6ei8.v v8, (a0), v4
 # CHECK-INST: vluxseg6ei8.v v8, (a0), v4
 # CHECK-ENCODING: [0x07,0x04,0x45,0xa6]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 04 45 a6 <unknown>
+# CHECK-UNKNOWN: a6450407 <unknown>
 
 vluxseg6ei16.v v8, (a0), v4, v0.t
 # CHECK-INST: vluxseg6ei16.v v8, (a0), v4, v0.t
 # CHECK-ENCODING: [0x07,0x54,0x45,0xa4]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 54 45 a4 <unknown>
+# CHECK-UNKNOWN: a4455407 <unknown>
 
 vluxseg6ei16.v v8, (a0), v4
 # CHECK-INST: vluxseg6ei16.v v8, (a0), v4
 # CHECK-ENCODING: [0x07,0x54,0x45,0xa6]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 54 45 a6 <unknown>
+# CHECK-UNKNOWN: a6455407 <unknown>
 
 vluxseg6ei32.v v8, (a0), v4, v0.t
 # CHECK-INST: vluxseg6ei32.v v8, (a0), v4, v0.t
 # CHECK-ENCODING: [0x07,0x64,0x45,0xa4]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 64 45 a4 <unknown>
+# CHECK-UNKNOWN: a4456407 <unknown>
 
 vluxseg6ei32.v v8, (a0), v4
 # CHECK-INST: vluxseg6ei32.v v8, (a0), v4
 # CHECK-ENCODING: [0x07,0x64,0x45,0xa6]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 64 45 a6 <unknown>
+# CHECK-UNKNOWN: a6456407 <unknown>
 
 vluxseg6ei64.v v8, (a0), v4, v0.t
 # CHECK-INST: vluxseg6ei64.v v8, (a0), v4, v0.t
 # CHECK-ENCODING: [0x07,0x74,0x45,0xa4]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 74 45 a4 <unknown>
+# CHECK-UNKNOWN: a4457407 <unknown>
 
 vluxseg6ei64.v v8, (a0), v4
 # CHECK-INST: vluxseg6ei64.v v8, (a0), v4
 # CHECK-ENCODING: [0x07,0x74,0x45,0xa6]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 74 45 a6 <unknown>
+# CHECK-UNKNOWN: a6457407 <unknown>
 
 vloxseg6ei8.v v8, (a0), v4, v0.t
 # CHECK-INST: vloxseg6ei8.v v8, (a0), v4, v0.t
 # CHECK-ENCODING: [0x07,0x04,0x45,0xac]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 04 45 ac <unknown>
+# CHECK-UNKNOWN: ac450407 <unknown>
 
 vloxseg6ei8.v v8, (a0), v4
 # CHECK-INST: vloxseg6ei8.v v8, (a0), v4
 # CHECK-ENCODING: [0x07,0x04,0x45,0xae]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 04 45 ae <unknown>
+# CHECK-UNKNOWN: ae450407 <unknown>
 
 vloxseg6ei16.v v8, (a0), v4, v0.t
 # CHECK-INST: vloxseg6ei16.v v8, (a0), v4, v0.t
 # CHECK-ENCODING: [0x07,0x54,0x45,0xac]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 54 45 ac <unknown>
+# CHECK-UNKNOWN: ac455407 <unknown>
 
 vloxseg6ei16.v v8, (a0), v4
 # CHECK-INST: vloxseg6ei16.v v8, (a0), v4
 # CHECK-ENCODING: [0x07,0x54,0x45,0xae]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 54 45 ae <unknown>
+# CHECK-UNKNOWN: ae455407 <unknown>
 
 vloxseg6ei32.v v8, (a0), v4, v0.t
 # CHECK-INST: vloxseg6ei32.v v8, (a0), v4, v0.t
 # CHECK-ENCODING: [0x07,0x64,0x45,0xac]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 64 45 ac <unknown>
+# CHECK-UNKNOWN: ac456407 <unknown>
 
 vloxseg6ei32.v v8, (a0), v4
 # CHECK-INST: vloxseg6ei32.v v8, (a0), v4
 # CHECK-ENCODING: [0x07,0x64,0x45,0xae]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 64 45 ae <unknown>
+# CHECK-UNKNOWN: ae456407 <unknown>
 
 vloxseg6ei64.v v8, (a0), v4, v0.t
 # CHECK-INST: vloxseg6ei64.v v8, (a0), v4, v0.t
 # CHECK-ENCODING: [0x07,0x74,0x45,0xac]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 74 45 ac <unknown>
+# CHECK-UNKNOWN: ac457407 <unknown>
 
 vloxseg6ei64.v v8, (a0), v4
 # CHECK-INST: vloxseg6ei64.v v8, (a0), v4
 # CHECK-ENCODING: [0x07,0x74,0x45,0xae]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 74 45 ae <unknown>
+# CHECK-UNKNOWN: ae457407 <unknown>
 
 vlseg7e8.v v8, (a0), v0.t
 # CHECK-INST: vlseg7e8.v v8, (a0), v0.t
 # CHECK-ENCODING: [0x07,0x04,0x05,0xc0]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 04 05 c0 <unknown>
+# CHECK-UNKNOWN: c0050407 <unknown>
 
 vlseg7e8.v v8, (a0)
 # CHECK-INST: vlseg7e8.v v8, (a0)
 # CHECK-ENCODING: [0x07,0x04,0x05,0xc2]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 04 05 c2 <unknown>
+# CHECK-UNKNOWN: c2050407 <unknown>
 
 vlseg7e16.v v8, (a0), v0.t
 # CHECK-INST: vlseg7e16.v v8, (a0), v0.t
 # CHECK-ENCODING: [0x07,0x54,0x05,0xc0]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 54 05 c0 <unknown>
+# CHECK-UNKNOWN: c0055407 <unknown>
 
 vlseg7e16.v v8, (a0)
 # CHECK-INST: vlseg7e16.v v8, (a0)
 # CHECK-ENCODING: [0x07,0x54,0x05,0xc2]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 54 05 c2 <unknown>
+# CHECK-UNKNOWN: c2055407 <unknown>
 
 vlseg7e32.v v8, (a0), v0.t
 # CHECK-INST: vlseg7e32.v v8, (a0), v0.t
 # CHECK-ENCODING: [0x07,0x64,0x05,0xc0]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 64 05 c0 <unknown>
+# CHECK-UNKNOWN: c0056407 <unknown>
 
 vlseg7e32.v v8, (a0)
 # CHECK-INST: vlseg7e32.v v8, (a0)
 # CHECK-ENCODING: [0x07,0x64,0x05,0xc2]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 64 05 c2 <unknown>
+# CHECK-UNKNOWN: c2056407 <unknown>
 
 vlseg7e64.v v8, (a0), v0.t
 # CHECK-INST: vlseg7e64.v v8, (a0), v0.t
 # CHECK-ENCODING: [0x07,0x74,0x05,0xc0]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 74 05 c0 <unknown>
+# CHECK-UNKNOWN: c0057407 <unknown>
 
 vlseg7e64.v v8, (a0)
 # CHECK-INST: vlseg7e64.v v8, (a0)
 # CHECK-ENCODING: [0x07,0x74,0x05,0xc2]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 74 05 c2 <unknown>
+# CHECK-UNKNOWN: c2057407 <unknown>
 
 vlseg7e8ff.v v8, (a0), v0.t
 # CHECK-INST: vlseg7e8ff.v v8, (a0), v0.t
 # CHECK-ENCODING: [0x07,0x04,0x05,0xc1]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 04 05 c1 <unknown>
+# CHECK-UNKNOWN: c1050407 <unknown>
 
 vlseg7e8ff.v v8, (a0)
 # CHECK-INST: vlseg7e8ff.v v8, (a0)
 # CHECK-ENCODING: [0x07,0x04,0x05,0xc3]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 04 05 c3 <unknown>
+# CHECK-UNKNOWN: c3050407 <unknown>
 
 vlseg7e16ff.v v8, (a0), v0.t
 # CHECK-INST: vlseg7e16ff.v v8, (a0), v0.t
 # CHECK-ENCODING: [0x07,0x54,0x05,0xc1]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 54 05 c1 <unknown>
+# CHECK-UNKNOWN: c1055407 <unknown>
 
 vlseg7e16ff.v v8, (a0)
 # CHECK-INST: vlseg7e16ff.v v8, (a0)
 # CHECK-ENCODING: [0x07,0x54,0x05,0xc3]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 54 05 c3 <unknown>
+# CHECK-UNKNOWN: c3055407 <unknown>
 
 vlseg7e32ff.v v8, (a0), v0.t
 # CHECK-INST: vlseg7e32ff.v v8, (a0), v0.t
 # CHECK-ENCODING: [0x07,0x64,0x05,0xc1]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 64 05 c1 <unknown>
+# CHECK-UNKNOWN: c1056407 <unknown>
 
 vlseg7e32ff.v v8, (a0)
 # CHECK-INST: vlseg7e32ff.v v8, (a0)
 # CHECK-ENCODING: [0x07,0x64,0x05,0xc3]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 64 05 c3 <unknown>
+# CHECK-UNKNOWN: c3056407 <unknown>
 
 vlseg7e64ff.v v8, (a0), v0.t
 # CHECK-INST: vlseg7e64ff.v v8, (a0), v0.t
 # CHECK-ENCODING: [0x07,0x74,0x05,0xc1]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 74 05 c1 <unknown>
+# CHECK-UNKNOWN: c1057407 <unknown>
 
 vlseg7e64ff.v v8, (a0)
 # CHECK-INST: vlseg7e64ff.v v8, (a0)
 # CHECK-ENCODING: [0x07,0x74,0x05,0xc3]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 74 05 c3 <unknown>
+# CHECK-UNKNOWN: c3057407 <unknown>
 
 vlsseg7e8.v v8, (a0), a1, v0.t
 # CHECK-INST: vlsseg7e8.v v8, (a0), a1, v0.t
 # CHECK-ENCODING: [0x07,0x04,0xb5,0xc8]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 04 b5 c8 <unknown>
+# CHECK-UNKNOWN: c8b50407 <unknown>
 
 vlsseg7e8.v v8, (a0), a1
 # CHECK-INST: vlsseg7e8.v v8, (a0), a1
 # CHECK-ENCODING: [0x07,0x04,0xb5,0xca]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 04 b5 ca <unknown>
+# CHECK-UNKNOWN: cab50407 <unknown>
 
 vlsseg7e16.v v8, (a0), a1, v0.t
 # CHECK-INST: vlsseg7e16.v v8, (a0), a1, v0.t
 # CHECK-ENCODING: [0x07,0x54,0xb5,0xc8]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 54 b5 c8 <unknown>
+# CHECK-UNKNOWN: c8b55407 <unknown>
 
 vlsseg7e16.v v8, (a0), a1
 # CHECK-INST: vlsseg7e16.v v8, (a0), a1
 # CHECK-ENCODING: [0x07,0x54,0xb5,0xca]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 54 b5 ca <unknown>
+# CHECK-UNKNOWN: cab55407 <unknown>
 
 vlsseg7e32.v v8, (a0), a1, v0.t
 # CHECK-INST: vlsseg7e32.v v8, (a0), a1, v0.t
 # CHECK-ENCODING: [0x07,0x64,0xb5,0xc8]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 64 b5 c8 <unknown>
+# CHECK-UNKNOWN: c8b56407 <unknown>
 
 vlsseg7e32.v v8, (a0), a1
 # CHECK-INST: vlsseg7e32.v v8, (a0), a1
 # CHECK-ENCODING: [0x07,0x64,0xb5,0xca]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 64 b5 ca <unknown>
+# CHECK-UNKNOWN: cab56407 <unknown>
 
 vlsseg7e64.v v8, (a0), a1, v0.t
 # CHECK-INST: vlsseg7e64.v v8, (a0), a1, v0.t
 # CHECK-ENCODING: [0x07,0x74,0xb5,0xc8]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 74 b5 c8 <unknown>
+# CHECK-UNKNOWN: c8b57407 <unknown>
 
 vlsseg7e64.v v8, (a0), a1
 # CHECK-INST: vlsseg7e64.v v8, (a0), a1
 # CHECK-ENCODING: [0x07,0x74,0xb5,0xca]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 74 b5 ca <unknown>
+# CHECK-UNKNOWN: cab57407 <unknown>
 
 vluxseg7ei8.v v8, (a0), v4, v0.t
 # CHECK-INST: vluxseg7ei8.v v8, (a0), v4, v0.t
 # CHECK-ENCODING: [0x07,0x04,0x45,0xc4]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 04 45 c4 <unknown>
+# CHECK-UNKNOWN: c4450407 <unknown>
 
 vluxseg7ei8.v v8, (a0), v4
 # CHECK-INST: vluxseg7ei8.v v8, (a0), v4
 # CHECK-ENCODING: [0x07,0x04,0x45,0xc6]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 04 45 c6 <unknown>
+# CHECK-UNKNOWN: c6450407 <unknown>
 
 vluxseg7ei16.v v8, (a0), v4, v0.t
 # CHECK-INST: vluxseg7ei16.v v8, (a0), v4, v0.t
 # CHECK-ENCODING: [0x07,0x54,0x45,0xc4]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 54 45 c4 <unknown>
+# CHECK-UNKNOWN: c4455407 <unknown>
 
 vluxseg7ei16.v v8, (a0), v4
 # CHECK-INST: vluxseg7ei16.v v8, (a0), v4
 # CHECK-ENCODING: [0x07,0x54,0x45,0xc6]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 54 45 c6 <unknown>
+# CHECK-UNKNOWN: c6455407 <unknown>
 
 vluxseg7ei32.v v8, (a0), v4, v0.t
 # CHECK-INST: vluxseg7ei32.v v8, (a0), v4, v0.t
 # CHECK-ENCODING: [0x07,0x64,0x45,0xc4]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 64 45 c4 <unknown>
+# CHECK-UNKNOWN: c4456407 <unknown>
 
 vluxseg7ei32.v v8, (a0), v4
 # CHECK-INST: vluxseg7ei32.v v8, (a0), v4
 # CHECK-ENCODING: [0x07,0x64,0x45,0xc6]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 64 45 c6 <unknown>
+# CHECK-UNKNOWN: c6456407 <unknown>
 
 vluxseg7ei64.v v8, (a0), v4, v0.t
 # CHECK-INST: vluxseg7ei64.v v8, (a0), v4, v0.t
 # CHECK-ENCODING: [0x07,0x74,0x45,0xc4]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 74 45 c4 <unknown>
+# CHECK-UNKNOWN: c4457407 <unknown>
 
 vluxseg7ei64.v v8, (a0), v4
 # CHECK-INST: vluxseg7ei64.v v8, (a0), v4
 # CHECK-ENCODING: [0x07,0x74,0x45,0xc6]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 74 45 c6 <unknown>
+# CHECK-UNKNOWN: c6457407 <unknown>
 
 vloxseg7ei8.v v8, (a0), v4, v0.t
 # CHECK-INST: vloxseg7ei8.v v8, (a0), v4, v0.t
 # CHECK-ENCODING: [0x07,0x04,0x45,0xcc]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 04 45 cc <unknown>
+# CHECK-UNKNOWN: cc450407 <unknown>
 
 vloxseg7ei8.v v8, (a0), v4
 # CHECK-INST: vloxseg7ei8.v v8, (a0), v4
 # CHECK-ENCODING: [0x07,0x04,0x45,0xce]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 04 45 ce <unknown>
+# CHECK-UNKNOWN: ce450407 <unknown>
 
 vloxseg7ei16.v v8, (a0), v4, v0.t
 # CHECK-INST: vloxseg7ei16.v v8, (a0), v4, v0.t
 # CHECK-ENCODING: [0x07,0x54,0x45,0xcc]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 54 45 cc <unknown>
+# CHECK-UNKNOWN: cc455407 <unknown>
 
 vloxseg7ei16.v v8, (a0), v4
 # CHECK-INST: vloxseg7ei16.v v8, (a0), v4
 # CHECK-ENCODING: [0x07,0x54,0x45,0xce]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 54 45 ce <unknown>
+# CHECK-UNKNOWN: ce455407 <unknown>
 
 vloxseg7ei32.v v8, (a0), v4, v0.t
 # CHECK-INST: vloxseg7ei32.v v8, (a0), v4, v0.t
 # CHECK-ENCODING: [0x07,0x64,0x45,0xcc]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 64 45 cc <unknown>
+# CHECK-UNKNOWN: cc456407 <unknown>
 
 vloxseg7ei32.v v8, (a0), v4
 # CHECK-INST: vloxseg7ei32.v v8, (a0), v4
 # CHECK-ENCODING: [0x07,0x64,0x45,0xce]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 64 45 ce <unknown>
+# CHECK-UNKNOWN: ce456407 <unknown>
 
 vloxseg7ei64.v v8, (a0), v4, v0.t
 # CHECK-INST: vloxseg7ei64.v v8, (a0), v4, v0.t
 # CHECK-ENCODING: [0x07,0x74,0x45,0xcc]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 74 45 cc <unknown>
+# CHECK-UNKNOWN: cc457407 <unknown>
 
 vloxseg7ei64.v v8, (a0), v4
 # CHECK-INST: vloxseg7ei64.v v8, (a0), v4
 # CHECK-ENCODING: [0x07,0x74,0x45,0xce]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 74 45 ce <unknown>
+# CHECK-UNKNOWN: ce457407 <unknown>
 
 vlseg8e8.v v8, (a0), v0.t
 # CHECK-INST: vlseg8e8.v v8, (a0), v0.t
 # CHECK-ENCODING: [0x07,0x04,0x05,0xe0]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 04 05 e0 <unknown>
+# CHECK-UNKNOWN: e0050407 <unknown>
 
 vlseg8e8.v v8, (a0)
 # CHECK-INST: vlseg8e8.v v8, (a0)
 # CHECK-ENCODING: [0x07,0x04,0x05,0xe2]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 04 05 e2 <unknown>
+# CHECK-UNKNOWN: e2050407 <unknown>
 
 vlseg8e16.v v8, (a0), v0.t
 # CHECK-INST: vlseg8e16.v v8, (a0), v0.t
 # CHECK-ENCODING: [0x07,0x54,0x05,0xe0]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 54 05 e0 <unknown>
+# CHECK-UNKNOWN: e0055407 <unknown>
 
 vlseg8e16.v v8, (a0)
 # CHECK-INST: vlseg8e16.v v8, (a0)
 # CHECK-ENCODING: [0x07,0x54,0x05,0xe2]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 54 05 e2 <unknown>
+# CHECK-UNKNOWN: e2055407 <unknown>
 
 vlseg8e32.v v8, (a0), v0.t
 # CHECK-INST: vlseg8e32.v v8, (a0), v0.t
 # CHECK-ENCODING: [0x07,0x64,0x05,0xe0]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 64 05 e0 <unknown>
+# CHECK-UNKNOWN: e0056407 <unknown>
 
 vlseg8e32.v v8, (a0)
 # CHECK-INST: vlseg8e32.v v8, (a0)
 # CHECK-ENCODING: [0x07,0x64,0x05,0xe2]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 64 05 e2 <unknown>
+# CHECK-UNKNOWN: e2056407 <unknown>
 
 vlseg8e64.v v8, (a0), v0.t
 # CHECK-INST: vlseg8e64.v v8, (a0), v0.t
 # CHECK-ENCODING: [0x07,0x74,0x05,0xe0]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 74 05 e0 <unknown>
+# CHECK-UNKNOWN: e0057407 <unknown>
 
 vlseg8e64.v v8, (a0)
 # CHECK-INST: vlseg8e64.v v8, (a0)
 # CHECK-ENCODING: [0x07,0x74,0x05,0xe2]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 74 05 e2 <unknown>
+# CHECK-UNKNOWN: e2057407 <unknown>
 
 vlseg8e8ff.v v8, (a0), v0.t
 # CHECK-INST: vlseg8e8ff.v v8, (a0), v0.t
 # CHECK-ENCODING: [0x07,0x04,0x05,0xe1]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 04 05 e1 <unknown>
+# CHECK-UNKNOWN: e1050407 <unknown>
 
 vlseg8e8ff.v v8, (a0)
 # CHECK-INST: vlseg8e8ff.v v8, (a0)
 # CHECK-ENCODING: [0x07,0x04,0x05,0xe3]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 04 05 e3 <unknown>
+# CHECK-UNKNOWN: e3050407 <unknown>
 
 vlseg8e16ff.v v8, (a0), v0.t
 # CHECK-INST: vlseg8e16ff.v v8, (a0), v0.t
 # CHECK-ENCODING: [0x07,0x54,0x05,0xe1]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 54 05 e1 <unknown>
+# CHECK-UNKNOWN: e1055407 <unknown>
 
 vlseg8e16ff.v v8, (a0)
 # CHECK-INST: vlseg8e16ff.v v8, (a0)
 # CHECK-ENCODING: [0x07,0x54,0x05,0xe3]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 54 05 e3 <unknown>
+# CHECK-UNKNOWN: e3055407 <unknown>
 
 vlseg8e32ff.v v8, (a0), v0.t
 # CHECK-INST: vlseg8e32ff.v v8, (a0), v0.t
 # CHECK-ENCODING: [0x07,0x64,0x05,0xe1]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 64 05 e1 <unknown>
+# CHECK-UNKNOWN: e1056407 <unknown>
 
 vlseg8e32ff.v v8, (a0)
 # CHECK-INST: vlseg8e32ff.v v8, (a0)
 # CHECK-ENCODING: [0x07,0x64,0x05,0xe3]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 64 05 e3 <unknown>
+# CHECK-UNKNOWN: e3056407 <unknown>
 
 vlseg8e64ff.v v8, (a0), v0.t
 # CHECK-INST: vlseg8e64ff.v v8, (a0), v0.t
 # CHECK-ENCODING: [0x07,0x74,0x05,0xe1]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 74 05 e1 <unknown>
+# CHECK-UNKNOWN: e1057407 <unknown>
 
 vlseg8e64ff.v v8, (a0)
 # CHECK-INST: vlseg8e64ff.v v8, (a0)
 # CHECK-ENCODING: [0x07,0x74,0x05,0xe3]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 74 05 e3 <unknown>
+# CHECK-UNKNOWN: e3057407 <unknown>
 
 vlsseg8e8.v v8, (a0), a1, v0.t
 # CHECK-INST: vlsseg8e8.v v8, (a0), a1, v0.t
 # CHECK-ENCODING: [0x07,0x04,0xb5,0xe8]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 04 b5 e8 <unknown>
+# CHECK-UNKNOWN: e8b50407 <unknown>
 
 vlsseg8e8.v v8, (a0), a1
 # CHECK-INST: vlsseg8e8.v v8, (a0), a1
 # CHECK-ENCODING: [0x07,0x04,0xb5,0xea]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 04 b5 ea <unknown>
+# CHECK-UNKNOWN: eab50407 <unknown>
 
 vlsseg8e16.v v8, (a0), a1, v0.t
 # CHECK-INST: vlsseg8e16.v v8, (a0), a1, v0.t
 # CHECK-ENCODING: [0x07,0x54,0xb5,0xe8]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 54 b5 e8 <unknown>
+# CHECK-UNKNOWN: e8b55407 <unknown>
 
 vlsseg8e16.v v8, (a0), a1
 # CHECK-INST: vlsseg8e16.v v8, (a0), a1
 # CHECK-ENCODING: [0x07,0x54,0xb5,0xea]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 54 b5 ea <unknown>
+# CHECK-UNKNOWN: eab55407 <unknown>
 
 vlsseg8e32.v v8, (a0), a1, v0.t
 # CHECK-INST: vlsseg8e32.v v8, (a0), a1, v0.t
 # CHECK-ENCODING: [0x07,0x64,0xb5,0xe8]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 64 b5 e8 <unknown>
+# CHECK-UNKNOWN: e8b56407 <unknown>
 
 vlsseg8e32.v v8, (a0), a1
 # CHECK-INST: vlsseg8e32.v v8, (a0), a1
 # CHECK-ENCODING: [0x07,0x64,0xb5,0xea]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 64 b5 ea <unknown>
+# CHECK-UNKNOWN: eab56407 <unknown>
 
 vlsseg8e64.v v8, (a0), a1, v0.t
 # CHECK-INST: vlsseg8e64.v v8, (a0), a1, v0.t
 # CHECK-ENCODING: [0x07,0x74,0xb5,0xe8]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 74 b5 e8 <unknown>
+# CHECK-UNKNOWN: e8b57407 <unknown>
 
 vlsseg8e64.v v8, (a0), a1
 # CHECK-INST: vlsseg8e64.v v8, (a0), a1
 # CHECK-ENCODING: [0x07,0x74,0xb5,0xea]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 74 b5 ea <unknown>
+# CHECK-UNKNOWN: eab57407 <unknown>
 
 vluxseg8ei8.v v8, (a0), v4, v0.t
 # CHECK-INST: vluxseg8ei8.v v8, (a0), v4, v0.t
 # CHECK-ENCODING: [0x07,0x04,0x45,0xe4]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 04 45 e4 <unknown>
+# CHECK-UNKNOWN: e4450407 <unknown>
 
 vluxseg8ei8.v v8, (a0), v4
 # CHECK-INST: vluxseg8ei8.v v8, (a0), v4
 # CHECK-ENCODING: [0x07,0x04,0x45,0xe6]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 04 45 e6 <unknown>
+# CHECK-UNKNOWN: e6450407 <unknown>
 
 vluxseg8ei16.v v8, (a0), v4, v0.t
 # CHECK-INST: vluxseg8ei16.v v8, (a0), v4, v0.t
 # CHECK-ENCODING: [0x07,0x54,0x45,0xe4]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 54 45 e4 <unknown>
+# CHECK-UNKNOWN: e4455407 <unknown>
 
 vluxseg8ei16.v v8, (a0), v4
 # CHECK-INST: vluxseg8ei16.v v8, (a0), v4
 # CHECK-ENCODING: [0x07,0x54,0x45,0xe6]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 54 45 e6 <unknown>
+# CHECK-UNKNOWN: e6455407 <unknown>
 
 vluxseg8ei32.v v8, (a0), v4, v0.t
 # CHECK-INST: vluxseg8ei32.v v8, (a0), v4, v0.t
 # CHECK-ENCODING: [0x07,0x64,0x45,0xe4]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 64 45 e4 <unknown>
+# CHECK-UNKNOWN: e4456407 <unknown>
 
 vluxseg8ei32.v v8, (a0), v4
 # CHECK-INST: vluxseg8ei32.v v8, (a0), v4
 # CHECK-ENCODING: [0x07,0x64,0x45,0xe6]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 64 45 e6 <unknown>
+# CHECK-UNKNOWN: e6456407 <unknown>
 
 vluxseg8ei64.v v8, (a0), v4, v0.t
 # CHECK-INST: vluxseg8ei64.v v8, (a0), v4, v0.t
 # CHECK-ENCODING: [0x07,0x74,0x45,0xe4]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 74 45 e4 <unknown>
+# CHECK-UNKNOWN: e4457407 <unknown>
 
 vluxseg8ei64.v v8, (a0), v4
 # CHECK-INST: vluxseg8ei64.v v8, (a0), v4
 # CHECK-ENCODING: [0x07,0x74,0x45,0xe6]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 74 45 e6 <unknown>
+# CHECK-UNKNOWN: e6457407 <unknown>
 
 vloxseg8ei8.v v8, (a0), v4, v0.t
 # CHECK-INST: vloxseg8ei8.v v8, (a0), v4, v0.t
 # CHECK-ENCODING: [0x07,0x04,0x45,0xec]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 04 45 ec <unknown>
+# CHECK-UNKNOWN: ec450407 <unknown>
 
 vloxseg8ei8.v v8, (a0), v4
 # CHECK-INST: vloxseg8ei8.v v8, (a0), v4
 # CHECK-ENCODING: [0x07,0x04,0x45,0xee]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 04 45 ee <unknown>
+# CHECK-UNKNOWN: ee450407 <unknown>
 
 vloxseg8ei16.v v8, (a0), v4, v0.t
 # CHECK-INST: vloxseg8ei16.v v8, (a0), v4, v0.t
 # CHECK-ENCODING: [0x07,0x54,0x45,0xec]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 54 45 ec <unknown>
+# CHECK-UNKNOWN: ec455407 <unknown>
 
 vloxseg8ei16.v v8, (a0), v4
 # CHECK-INST: vloxseg8ei16.v v8, (a0), v4
 # CHECK-ENCODING: [0x07,0x54,0x45,0xee]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 54 45 ee <unknown>
+# CHECK-UNKNOWN: ee455407 <unknown>
 
 vloxseg8ei32.v v8, (a0), v4, v0.t
 # CHECK-INST: vloxseg8ei32.v v8, (a0), v4, v0.t
 # CHECK-ENCODING: [0x07,0x64,0x45,0xec]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 64 45 ec <unknown>
+# CHECK-UNKNOWN: ec456407 <unknown>
 
 vloxseg8ei32.v v8, (a0), v4
 # CHECK-INST: vloxseg8ei32.v v8, (a0), v4
 # CHECK-ENCODING: [0x07,0x64,0x45,0xee]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 64 45 ee <unknown>
+# CHECK-UNKNOWN: ee456407 <unknown>
 
 vloxseg8ei64.v v8, (a0), v4, v0.t
 # CHECK-INST: vloxseg8ei64.v v8, (a0), v4, v0.t
 # CHECK-ENCODING: [0x07,0x74,0x45,0xec]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 74 45 ec <unknown>
+# CHECK-UNKNOWN: ec457407 <unknown>
 
 vloxseg8ei64.v v8, (a0), v4
 # CHECK-INST: vloxseg8ei64.v v8, (a0), v4
 # CHECK-ENCODING: [0x07,0x74,0x45,0xee]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 74 45 ee <unknown>
+# CHECK-UNKNOWN: ee457407 <unknown>
 
 vsseg2e8.v v24, (a0), v0.t
 # CHECK-INST: vsseg2e8.v v24, (a0), v0.t
 # CHECK-ENCODING: [0x27,0x0c,0x05,0x20]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 0c 05 20 <unknown>
+# CHECK-UNKNOWN: 20050c27 <unknown>
 
 vsseg2e8.v v24, (a0)
 # CHECK-INST: vsseg2e8.v v24, (a0)
 # CHECK-ENCODING: [0x27,0x0c,0x05,0x22]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 0c 05 22 <unknown>
+# CHECK-UNKNOWN: 22050c27 <unknown>
 
 vsseg2e16.v v24, (a0), v0.t
 # CHECK-INST: vsseg2e16.v v24, (a0), v0.t
 # CHECK-ENCODING: [0x27,0x5c,0x05,0x20]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 5c 05 20 <unknown>
+# CHECK-UNKNOWN: 20055c27 <unknown>
 
 vsseg2e16.v v24, (a0)
 # CHECK-INST: vsseg2e16.v v24, (a0)
 # CHECK-ENCODING: [0x27,0x5c,0x05,0x22]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 5c 05 22 <unknown>
+# CHECK-UNKNOWN: 22055c27 <unknown>
 
 vsseg2e32.v v24, (a0), v0.t
 # CHECK-INST: vsseg2e32.v v24, (a0), v0.t
 # CHECK-ENCODING: [0x27,0x6c,0x05,0x20]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 6c 05 20 <unknown>
+# CHECK-UNKNOWN: 20056c27 <unknown>
 
 vsseg2e32.v v24, (a0)
 # CHECK-INST: vsseg2e32.v v24, (a0)
 # CHECK-ENCODING: [0x27,0x6c,0x05,0x22]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 6c 05 22 <unknown>
+# CHECK-UNKNOWN: 22056c27 <unknown>
 
 vsseg2e64.v v24, (a0), v0.t
 # CHECK-INST: vsseg2e64.v v24, (a0), v0.t
 # CHECK-ENCODING: [0x27,0x7c,0x05,0x20]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 7c 05 20 <unknown>
+# CHECK-UNKNOWN: 20057c27 <unknown>
 
 vsseg2e64.v v24, (a0)
 # CHECK-INST: vsseg2e64.v v24, (a0)
 # CHECK-ENCODING: [0x27,0x7c,0x05,0x22]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 7c 05 22 <unknown>
+# CHECK-UNKNOWN: 22057c27 <unknown>
 
 vssseg2e8.v v24, (a0), a1, v0.t
 # CHECK-INST: vssseg2e8.v v24, (a0), a1, v0.t
 # CHECK-ENCODING: [0x27,0x0c,0xb5,0x28]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 0c b5 28 <unknown>
+# CHECK-UNKNOWN: 28b50c27 <unknown>
 
 vssseg2e8.v v24, (a0), a1
 # CHECK-INST: vssseg2e8.v v24, (a0), a1
 # CHECK-ENCODING: [0x27,0x0c,0xb5,0x2a]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 0c b5 2a <unknown>
+# CHECK-UNKNOWN: 2ab50c27 <unknown>
 
 vssseg2e16.v v24, (a0), a1, v0.t
 # CHECK-INST: vssseg2e16.v v24, (a0), a1, v0.t
 # CHECK-ENCODING: [0x27,0x5c,0xb5,0x28]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 5c b5 28 <unknown>
+# CHECK-UNKNOWN: 28b55c27 <unknown>
 
 vssseg2e16.v v24, (a0), a1
 # CHECK-INST: vssseg2e16.v v24, (a0), a1
 # CHECK-ENCODING: [0x27,0x5c,0xb5,0x2a]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 5c b5 2a <unknown>
+# CHECK-UNKNOWN: 2ab55c27 <unknown>
 
 vssseg2e32.v v24, (a0), a1, v0.t
 # CHECK-INST: vssseg2e32.v v24, (a0), a1, v0.t
 # CHECK-ENCODING: [0x27,0x6c,0xb5,0x28]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 6c b5 28 <unknown>
+# CHECK-UNKNOWN: 28b56c27 <unknown>
 
 vssseg2e32.v v24, (a0), a1
 # CHECK-INST: vssseg2e32.v v24, (a0), a1
 # CHECK-ENCODING: [0x27,0x6c,0xb5,0x2a]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 6c b5 2a <unknown>
+# CHECK-UNKNOWN: 2ab56c27 <unknown>
 
 vssseg2e64.v v24, (a0), a1, v0.t
 # CHECK-INST: vssseg2e64.v v24, (a0), a1, v0.t
 # CHECK-ENCODING: [0x27,0x7c,0xb5,0x28]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 7c b5 28 <unknown>
+# CHECK-UNKNOWN: 28b57c27 <unknown>
 
 vssseg2e64.v v24, (a0), a1
 # CHECK-INST: vssseg2e64.v v24, (a0), a1
 # CHECK-ENCODING: [0x27,0x7c,0xb5,0x2a]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 7c b5 2a <unknown>
+# CHECK-UNKNOWN: 2ab57c27 <unknown>
 
 vsuxseg2ei8.v v24, (a0), v4, v0.t
 # CHECK-INST: vsuxseg2ei8.v v24, (a0), v4, v0.t
 # CHECK-ENCODING: [0x27,0x0c,0x45,0x24]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 0c 45 24 <unknown>
+# CHECK-UNKNOWN: 24450c27 <unknown>
 
 vsuxseg2ei8.v v24, (a0), v4
 # CHECK-INST: vsuxseg2ei8.v v24, (a0), v4
 # CHECK-ENCODING: [0x27,0x0c,0x45,0x26]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 0c 45 26 <unknown>
+# CHECK-UNKNOWN: 26450c27 <unknown>
 
 vsuxseg2ei16.v v24, (a0), v4, v0.t
 # CHECK-INST: vsuxseg2ei16.v v24, (a0), v4, v0.t
 # CHECK-ENCODING: [0x27,0x5c,0x45,0x24]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 5c 45 24 <unknown>
+# CHECK-UNKNOWN: 24455c27 <unknown>
 
 vsuxseg2ei16.v v24, (a0), v4
 # CHECK-INST: vsuxseg2ei16.v v24, (a0), v4
 # CHECK-ENCODING: [0x27,0x5c,0x45,0x26]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 5c 45 26 <unknown>
+# CHECK-UNKNOWN: 26455c27 <unknown>
 
 vsuxseg2ei32.v v24, (a0), v4, v0.t
 # CHECK-INST: vsuxseg2ei32.v v24, (a0), v4, v0.t
 # CHECK-ENCODING: [0x27,0x6c,0x45,0x24]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 6c 45 24 <unknown>
+# CHECK-UNKNOWN: 24456c27 <unknown>
 
 vsuxseg2ei32.v v24, (a0), v4
 # CHECK-INST: vsuxseg2ei32.v v24, (a0), v4
 # CHECK-ENCODING: [0x27,0x6c,0x45,0x26]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 6c 45 26 <unknown>
+# CHECK-UNKNOWN: 26456c27 <unknown>
 
 vsuxseg2ei64.v v24, (a0), v4, v0.t
 # CHECK-INST: vsuxseg2ei64.v v24, (a0), v4, v0.t
 # CHECK-ENCODING: [0x27,0x7c,0x45,0x24]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 7c 45 24 <unknown>
+# CHECK-UNKNOWN: 24457c27 <unknown>
 
 vsuxseg2ei64.v v24, (a0), v4
 # CHECK-INST: vsuxseg2ei64.v v24, (a0), v4
 # CHECK-ENCODING: [0x27,0x7c,0x45,0x26]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 7c 45 26 <unknown>
+# CHECK-UNKNOWN: 26457c27 <unknown>
 
 vsoxseg2ei8.v v24, (a0), v4, v0.t
 # CHECK-INST: vsoxseg2ei8.v v24, (a0), v4, v0.t
 # CHECK-ENCODING: [0x27,0x0c,0x45,0x2c]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 0c 45 2c <unknown>
+# CHECK-UNKNOWN: 2c450c27 <unknown>
 
 vsoxseg2ei8.v v24, (a0), v4
 # CHECK-INST: vsoxseg2ei8.v v24, (a0), v4
 # CHECK-ENCODING: [0x27,0x0c,0x45,0x2e]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 0c 45 2e <unknown>
+# CHECK-UNKNOWN: 2e450c27 <unknown>
 
 vsoxseg2ei16.v v24, (a0), v4, v0.t
 # CHECK-INST: vsoxseg2ei16.v v24, (a0), v4, v0.t
 # CHECK-ENCODING: [0x27,0x5c,0x45,0x2c]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 5c 45 2c <unknown>
+# CHECK-UNKNOWN: 2c455c27 <unknown>
 
 vsoxseg2ei16.v v24, (a0), v4
 # CHECK-INST: vsoxseg2ei16.v v24, (a0), v4
 # CHECK-ENCODING: [0x27,0x5c,0x45,0x2e]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 5c 45 2e <unknown>
+# CHECK-UNKNOWN: 2e455c27 <unknown>
 
 vsoxseg2ei32.v v24, (a0), v4, v0.t
 # CHECK-INST: vsoxseg2ei32.v v24, (a0), v4, v0.t
 # CHECK-ENCODING: [0x27,0x6c,0x45,0x2c]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 6c 45 2c <unknown>
+# CHECK-UNKNOWN: 2c456c27 <unknown>
 
 vsoxseg2ei32.v v24, (a0), v4
 # CHECK-INST: vsoxseg2ei32.v v24, (a0), v4
 # CHECK-ENCODING: [0x27,0x6c,0x45,0x2e]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 6c 45 2e <unknown>
+# CHECK-UNKNOWN: 2e456c27 <unknown>
 
 vsoxseg2ei64.v v24, (a0), v4, v0.t
 # CHECK-INST: vsoxseg2ei64.v v24, (a0), v4, v0.t
 # CHECK-ENCODING: [0x27,0x7c,0x45,0x2c]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 7c 45 2c <unknown>
+# CHECK-UNKNOWN: 2c457c27 <unknown>
 
 vsoxseg2ei64.v v24, (a0), v4
 # CHECK-INST: vsoxseg2ei64.v v24, (a0), v4
 # CHECK-ENCODING: [0x27,0x7c,0x45,0x2e]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 7c 45 2e <unknown>
+# CHECK-UNKNOWN: 2e457c27 <unknown>
 
 vsseg3e8.v v24, (a0), v0.t
 # CHECK-INST: vsseg3e8.v v24, (a0), v0.t
 # CHECK-ENCODING: [0x27,0x0c,0x05,0x40]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 0c 05 40 <unknown>
+# CHECK-UNKNOWN: 40050c27 <unknown>
 
 vsseg3e8.v v24, (a0)
 # CHECK-INST: vsseg3e8.v v24, (a0)
 # CHECK-ENCODING: [0x27,0x0c,0x05,0x42]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 0c 05 42 <unknown>
+# CHECK-UNKNOWN: 42050c27 <unknown>
 
 vsseg3e16.v v24, (a0), v0.t
 # CHECK-INST: vsseg3e16.v v24, (a0), v0.t
 # CHECK-ENCODING: [0x27,0x5c,0x05,0x40]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 5c 05 40 <unknown>
+# CHECK-UNKNOWN: 40055c27 <unknown>
 
 vsseg3e16.v v24, (a0)
 # CHECK-INST: vsseg3e16.v v24, (a0)
 # CHECK-ENCODING: [0x27,0x5c,0x05,0x42]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 5c 05 42 <unknown>
+# CHECK-UNKNOWN: 42055c27 <unknown>
 
 vsseg3e32.v v24, (a0), v0.t
 # CHECK-INST: vsseg3e32.v v24, (a0), v0.t
 # CHECK-ENCODING: [0x27,0x6c,0x05,0x40]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 6c 05 40 <unknown>
+# CHECK-UNKNOWN: 40056c27 <unknown>
 
 vsseg3e32.v v24, (a0)
 # CHECK-INST: vsseg3e32.v v24, (a0)
 # CHECK-ENCODING: [0x27,0x6c,0x05,0x42]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 6c 05 42 <unknown>
+# CHECK-UNKNOWN: 42056c27 <unknown>
 
 vsseg3e64.v v24, (a0), v0.t
 # CHECK-INST: vsseg3e64.v v24, (a0), v0.t
 # CHECK-ENCODING: [0x27,0x7c,0x05,0x40]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 7c 05 40 <unknown>
+# CHECK-UNKNOWN: 40057c27 <unknown>
 
 vsseg3e64.v v24, (a0)
 # CHECK-INST: vsseg3e64.v v24, (a0)
 # CHECK-ENCODING: [0x27,0x7c,0x05,0x42]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 7c 05 42 <unknown>
+# CHECK-UNKNOWN: 42057c27 <unknown>
 
 vssseg3e8.v v24, (a0), a1, v0.t
 # CHECK-INST: vssseg3e8.v v24, (a0), a1, v0.t
 # CHECK-ENCODING: [0x27,0x0c,0xb5,0x48]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 0c b5 48 <unknown>
+# CHECK-UNKNOWN: 48b50c27 <unknown>
 
 vssseg3e8.v v24, (a0), a1
 # CHECK-INST: vssseg3e8.v v24, (a0), a1
 # CHECK-ENCODING: [0x27,0x0c,0xb5,0x4a]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 0c b5 4a <unknown>
+# CHECK-UNKNOWN: 4ab50c27 <unknown>
 
 vssseg3e16.v v24, (a0), a1, v0.t
 # CHECK-INST: vssseg3e16.v v24, (a0), a1, v0.t
 # CHECK-ENCODING: [0x27,0x5c,0xb5,0x48]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 5c b5 48 <unknown>
+# CHECK-UNKNOWN: 48b55c27 <unknown>
 
 vssseg3e16.v v24, (a0), a1
 # CHECK-INST: vssseg3e16.v v24, (a0), a1
 # CHECK-ENCODING: [0x27,0x5c,0xb5,0x4a]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 5c b5 4a <unknown>
+# CHECK-UNKNOWN: 4ab55c27 <unknown>
 
 vssseg3e32.v v24, (a0), a1, v0.t
 # CHECK-INST: vssseg3e32.v v24, (a0), a1, v0.t
 # CHECK-ENCODING: [0x27,0x6c,0xb5,0x48]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 6c b5 48 <unknown>
+# CHECK-UNKNOWN: 48b56c27 <unknown>
 
 vssseg3e32.v v24, (a0), a1
 # CHECK-INST: vssseg3e32.v v24, (a0), a1
 # CHECK-ENCODING: [0x27,0x6c,0xb5,0x4a]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 6c b5 4a <unknown>
+# CHECK-UNKNOWN: 4ab56c27 <unknown>
 
 vssseg3e64.v v24, (a0), a1, v0.t
 # CHECK-INST: vssseg3e64.v v24, (a0), a1, v0.t
 # CHECK-ENCODING: [0x27,0x7c,0xb5,0x48]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 7c b5 48 <unknown>
+# CHECK-UNKNOWN: 48b57c27 <unknown>
 
 vssseg3e64.v v24, (a0), a1
 # CHECK-INST: vssseg3e64.v v24, (a0), a1
 # CHECK-ENCODING: [0x27,0x7c,0xb5,0x4a]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 7c b5 4a <unknown>
+# CHECK-UNKNOWN: 4ab57c27 <unknown>
 
 vsuxseg3ei8.v v24, (a0), v4, v0.t
 # CHECK-INST: vsuxseg3ei8.v v24, (a0), v4, v0.t
 # CHECK-ENCODING: [0x27,0x0c,0x45,0x44]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 0c 45 44 <unknown>
+# CHECK-UNKNOWN: 44450c27 <unknown>
 
 vsuxseg3ei8.v v24, (a0), v4
 # CHECK-INST: vsuxseg3ei8.v v24, (a0), v4
 # CHECK-ENCODING: [0x27,0x0c,0x45,0x46]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 0c 45 46 <unknown>
+# CHECK-UNKNOWN: 46450c27 <unknown>
 
 vsuxseg3ei16.v v24, (a0), v4, v0.t
 # CHECK-INST: vsuxseg3ei16.v v24, (a0), v4, v0.t
 # CHECK-ENCODING: [0x27,0x5c,0x45,0x44]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 5c 45 44 <unknown>
+# CHECK-UNKNOWN: 44455c27 <unknown>
 
 vsuxseg3ei16.v v24, (a0), v4
 # CHECK-INST: vsuxseg3ei16.v v24, (a0), v4
 # CHECK-ENCODING: [0x27,0x5c,0x45,0x46]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 5c 45 46 <unknown>
+# CHECK-UNKNOWN: 46455c27 <unknown>
 
 vsuxseg3ei32.v v24, (a0), v4, v0.t
 # CHECK-INST: vsuxseg3ei32.v v24, (a0), v4, v0.t
 # CHECK-ENCODING: [0x27,0x6c,0x45,0x44]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 6c 45 44 <unknown>
+# CHECK-UNKNOWN: 44456c27 <unknown>
 
 vsuxseg3ei32.v v24, (a0), v4
 # CHECK-INST: vsuxseg3ei32.v v24, (a0), v4
 # CHECK-ENCODING: [0x27,0x6c,0x45,0x46]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 6c 45 46 <unknown>
+# CHECK-UNKNOWN: 46456c27 <unknown>
 
 vsuxseg3ei64.v v24, (a0), v4, v0.t
 # CHECK-INST: vsuxseg3ei64.v v24, (a0), v4, v0.t
 # CHECK-ENCODING: [0x27,0x7c,0x45,0x44]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 7c 45 44 <unknown>
+# CHECK-UNKNOWN: 44457c27 <unknown>
 
 vsuxseg3ei64.v v24, (a0), v4
 # CHECK-INST: vsuxseg3ei64.v v24, (a0), v4
 # CHECK-ENCODING: [0x27,0x7c,0x45,0x46]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 7c 45 46 <unknown>
+# CHECK-UNKNOWN: 46457c27 <unknown>
 
 vsoxseg3ei8.v v24, (a0), v4, v0.t
 # CHECK-INST: vsoxseg3ei8.v v24, (a0), v4, v0.t
 # CHECK-ENCODING: [0x27,0x0c,0x45,0x4c]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 0c 45 4c <unknown>
+# CHECK-UNKNOWN: 4c450c27 <unknown>
 
 vsoxseg3ei8.v v24, (a0), v4
 # CHECK-INST: vsoxseg3ei8.v v24, (a0), v4
 # CHECK-ENCODING: [0x27,0x0c,0x45,0x4e]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 0c 45 4e <unknown>
+# CHECK-UNKNOWN: 4e450c27 <unknown>
 
 vsoxseg3ei16.v v24, (a0), v4, v0.t
 # CHECK-INST: vsoxseg3ei16.v v24, (a0), v4, v0.t
 # CHECK-ENCODING: [0x27,0x5c,0x45,0x4c]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 5c 45 4c <unknown>
+# CHECK-UNKNOWN: 4c455c27 <unknown>
 
 vsoxseg3ei16.v v24, (a0), v4
 # CHECK-INST: vsoxseg3ei16.v v24, (a0), v4
 # CHECK-ENCODING: [0x27,0x5c,0x45,0x4e]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 5c 45 4e <unknown>
+# CHECK-UNKNOWN: 4e455c27 <unknown>
 
 vsoxseg3ei32.v v24, (a0), v4, v0.t
 # CHECK-INST: vsoxseg3ei32.v v24, (a0), v4, v0.t
 # CHECK-ENCODING: [0x27,0x6c,0x45,0x4c]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 6c 45 4c <unknown>
+# CHECK-UNKNOWN: 4c456c27 <unknown>
 
 vsoxseg3ei32.v v24, (a0), v4
 # CHECK-INST: vsoxseg3ei32.v v24, (a0), v4
 # CHECK-ENCODING: [0x27,0x6c,0x45,0x4e]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 6c 45 4e <unknown>
+# CHECK-UNKNOWN: 4e456c27 <unknown>
 
 vsoxseg3ei64.v v24, (a0), v4, v0.t
 # CHECK-INST: vsoxseg3ei64.v v24, (a0), v4, v0.t
 # CHECK-ENCODING: [0x27,0x7c,0x45,0x4c]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 7c 45 4c <unknown>
+# CHECK-UNKNOWN: 4c457c27 <unknown>
 
 vsoxseg3ei64.v v24, (a0), v4
 # CHECK-INST: vsoxseg3ei64.v v24, (a0), v4
 # CHECK-ENCODING: [0x27,0x7c,0x45,0x4e]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 7c 45 4e <unknown>
+# CHECK-UNKNOWN: 4e457c27 <unknown>
 
 vsseg4e8.v v24, (a0), v0.t
 # CHECK-INST: vsseg4e8.v v24, (a0), v0.t
 # CHECK-ENCODING: [0x27,0x0c,0x05,0x60]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 0c 05 60 <unknown>
+# CHECK-UNKNOWN: 60050c27 <unknown>
 
 vsseg4e8.v v24, (a0)
 # CHECK-INST: vsseg4e8.v v24, (a0)
 # CHECK-ENCODING: [0x27,0x0c,0x05,0x62]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 0c 05 62 <unknown>
+# CHECK-UNKNOWN: 62050c27 <unknown>
 
 vsseg4e16.v v24, (a0), v0.t
 # CHECK-INST: vsseg4e16.v v24, (a0), v0.t
 # CHECK-ENCODING: [0x27,0x5c,0x05,0x60]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 5c 05 60 <unknown>
+# CHECK-UNKNOWN: 60055c27 <unknown>
 
 vsseg4e16.v v24, (a0)
 # CHECK-INST: vsseg4e16.v v24, (a0)
 # CHECK-ENCODING: [0x27,0x5c,0x05,0x62]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 5c 05 62 <unknown>
+# CHECK-UNKNOWN: 62055c27 <unknown>
 
 vsseg4e32.v v24, (a0), v0.t
 # CHECK-INST: vsseg4e32.v v24, (a0), v0.t
 # CHECK-ENCODING: [0x27,0x6c,0x05,0x60]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 6c 05 60 <unknown>
+# CHECK-UNKNOWN: 60056c27 <unknown>
 
 vsseg4e32.v v24, (a0)
 # CHECK-INST: vsseg4e32.v v24, (a0)
 # CHECK-ENCODING: [0x27,0x6c,0x05,0x62]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 6c 05 62 <unknown>
+# CHECK-UNKNOWN: 62056c27 <unknown>
 
 vsseg4e64.v v24, (a0), v0.t
 # CHECK-INST: vsseg4e64.v v24, (a0), v0.t
 # CHECK-ENCODING: [0x27,0x7c,0x05,0x60]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 7c 05 60 <unknown>
+# CHECK-UNKNOWN: 60057c27 <unknown>
 
 vsseg4e64.v v24, (a0)
 # CHECK-INST: vsseg4e64.v v24, (a0)
 # CHECK-ENCODING: [0x27,0x7c,0x05,0x62]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 7c 05 62 <unknown>
+# CHECK-UNKNOWN: 62057c27 <unknown>
 
 vssseg4e8.v v24, (a0), a1, v0.t
 # CHECK-INST: vssseg4e8.v v24, (a0), a1, v0.t
 # CHECK-ENCODING: [0x27,0x0c,0xb5,0x68]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 0c b5 68 <unknown>
+# CHECK-UNKNOWN: 68b50c27 <unknown>
 
 vssseg4e8.v v24, (a0), a1
 # CHECK-INST: vssseg4e8.v v24, (a0), a1
 # CHECK-ENCODING: [0x27,0x0c,0xb5,0x6a]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 0c b5 6a <unknown>
+# CHECK-UNKNOWN: 6ab50c27 <unknown>
 
 vssseg4e16.v v24, (a0), a1, v0.t
 # CHECK-INST: vssseg4e16.v v24, (a0), a1, v0.t
 # CHECK-ENCODING: [0x27,0x5c,0xb5,0x68]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 5c b5 68 <unknown>
+# CHECK-UNKNOWN: 68b55c27 <unknown>
 
 vssseg4e16.v v24, (a0), a1
 # CHECK-INST: vssseg4e16.v v24, (a0), a1
 # CHECK-ENCODING: [0x27,0x5c,0xb5,0x6a]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 5c b5 6a <unknown>
+# CHECK-UNKNOWN: 6ab55c27 <unknown>
 
 vssseg4e32.v v24, (a0), a1, v0.t
 # CHECK-INST: vssseg4e32.v v24, (a0), a1, v0.t
 # CHECK-ENCODING: [0x27,0x6c,0xb5,0x68]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 6c b5 68 <unknown>
+# CHECK-UNKNOWN: 68b56c27 <unknown>
 
 vssseg4e32.v v24, (a0), a1
 # CHECK-INST: vssseg4e32.v v24, (a0), a1
 # CHECK-ENCODING: [0x27,0x6c,0xb5,0x6a]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 6c b5 6a <unknown>
+# CHECK-UNKNOWN: 6ab56c27 <unknown>
 
 vssseg4e64.v v24, (a0), a1, v0.t
 # CHECK-INST: vssseg4e64.v v24, (a0), a1, v0.t
 # CHECK-ENCODING: [0x27,0x7c,0xb5,0x68]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 7c b5 68 <unknown>
+# CHECK-UNKNOWN: 68b57c27 <unknown>
 
 vssseg4e64.v v24, (a0), a1
 # CHECK-INST: vssseg4e64.v v24, (a0), a1
 # CHECK-ENCODING: [0x27,0x7c,0xb5,0x6a]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 7c b5 6a <unknown>
+# CHECK-UNKNOWN: 6ab57c27 <unknown>
 
 vsuxseg4ei8.v v24, (a0), v4, v0.t
 # CHECK-INST: vsuxseg4ei8.v v24, (a0), v4, v0.t
 # CHECK-ENCODING: [0x27,0x0c,0x45,0x64]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 0c 45 64 <unknown>
+# CHECK-UNKNOWN: 64450c27 <unknown>
 
 vsuxseg4ei8.v v24, (a0), v4
 # CHECK-INST: vsuxseg4ei8.v v24, (a0), v4
 # CHECK-ENCODING: [0x27,0x0c,0x45,0x66]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 0c 45 66 <unknown>
+# CHECK-UNKNOWN: 66450c27 <unknown>
 
 vsuxseg4ei16.v v24, (a0), v4, v0.t
 # CHECK-INST: vsuxseg4ei16.v v24, (a0), v4, v0.t
 # CHECK-ENCODING: [0x27,0x5c,0x45,0x64]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 5c 45 64 <unknown>
+# CHECK-UNKNOWN: 64455c27 <unknown>
 
 vsuxseg4ei16.v v24, (a0), v4
 # CHECK-INST: vsuxseg4ei16.v v24, (a0), v4
 # CHECK-ENCODING: [0x27,0x5c,0x45,0x66]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 5c 45 66 <unknown>
+# CHECK-UNKNOWN: 66455c27 <unknown>
 
 vsuxseg4ei32.v v24, (a0), v4, v0.t
 # CHECK-INST: vsuxseg4ei32.v v24, (a0), v4, v0.t
 # CHECK-ENCODING: [0x27,0x6c,0x45,0x64]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 6c 45 64 <unknown>
+# CHECK-UNKNOWN: 64456c27 <unknown>
 
 vsuxseg4ei32.v v24, (a0), v4
 # CHECK-INST: vsuxseg4ei32.v v24, (a0), v4
 # CHECK-ENCODING: [0x27,0x6c,0x45,0x66]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 6c 45 66 <unknown>
+# CHECK-UNKNOWN: 66456c27 <unknown>
 
 vsuxseg4ei64.v v24, (a0), v4, v0.t
 # CHECK-INST: vsuxseg4ei64.v v24, (a0), v4, v0.t
 # CHECK-ENCODING: [0x27,0x7c,0x45,0x64]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 7c 45 64 <unknown>
+# CHECK-UNKNOWN: 64457c27 <unknown>
 
 vsuxseg4ei64.v v24, (a0), v4
 # CHECK-INST: vsuxseg4ei64.v v24, (a0), v4
 # CHECK-ENCODING: [0x27,0x7c,0x45,0x66]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 7c 45 66 <unknown>
+# CHECK-UNKNOWN: 66457c27 <unknown>
 
 vsoxseg4ei8.v v24, (a0), v4, v0.t
 # CHECK-INST: vsoxseg4ei8.v v24, (a0), v4, v0.t
 # CHECK-ENCODING: [0x27,0x0c,0x45,0x6c]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 0c 45 6c <unknown>
+# CHECK-UNKNOWN: 6c450c27 <unknown>
 
 vsoxseg4ei8.v v24, (a0), v4
 # CHECK-INST: vsoxseg4ei8.v v24, (a0), v4
 # CHECK-ENCODING: [0x27,0x0c,0x45,0x6e]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 0c 45 6e <unknown>
+# CHECK-UNKNOWN: 6e450c27 <unknown>
 
 vsoxseg4ei16.v v24, (a0), v4, v0.t
 # CHECK-INST: vsoxseg4ei16.v v24, (a0), v4, v0.t
 # CHECK-ENCODING: [0x27,0x5c,0x45,0x6c]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 5c 45 6c <unknown>
+# CHECK-UNKNOWN: 6c455c27 <unknown>
 
 vsoxseg4ei16.v v24, (a0), v4
 # CHECK-INST: vsoxseg4ei16.v v24, (a0), v4
 # CHECK-ENCODING: [0x27,0x5c,0x45,0x6e]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 5c 45 6e <unknown>
+# CHECK-UNKNOWN: 6e455c27 <unknown>
 
 vsoxseg4ei32.v v24, (a0), v4, v0.t
 # CHECK-INST: vsoxseg4ei32.v v24, (a0), v4, v0.t
 # CHECK-ENCODING: [0x27,0x6c,0x45,0x6c]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 6c 45 6c <unknown>
+# CHECK-UNKNOWN: 6c456c27 <unknown>
 
 vsoxseg4ei32.v v24, (a0), v4
 # CHECK-INST: vsoxseg4ei32.v v24, (a0), v4
 # CHECK-ENCODING: [0x27,0x6c,0x45,0x6e]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 6c 45 6e <unknown>
+# CHECK-UNKNOWN: 6e456c27 <unknown>
 
 vsoxseg4ei64.v v24, (a0), v4, v0.t
 # CHECK-INST: vsoxseg4ei64.v v24, (a0), v4, v0.t
 # CHECK-ENCODING: [0x27,0x7c,0x45,0x6c]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 7c 45 6c <unknown>
+# CHECK-UNKNOWN: 6c457c27 <unknown>
 
 vsoxseg4ei64.v v24, (a0), v4
 # CHECK-INST: vsoxseg4ei64.v v24, (a0), v4
 # CHECK-ENCODING: [0x27,0x7c,0x45,0x6e]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 7c 45 6e <unknown>
+# CHECK-UNKNOWN: 6e457c27 <unknown>
 
 vsseg5e8.v v24, (a0), v0.t
 # CHECK-INST: vsseg5e8.v v24, (a0), v0.t
 # CHECK-ENCODING: [0x27,0x0c,0x05,0x80]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 0c 05 80 <unknown>
+# CHECK-UNKNOWN: 80050c27 <unknown>
 
 vsseg5e8.v v24, (a0)
 # CHECK-INST: vsseg5e8.v v24, (a0)
 # CHECK-ENCODING: [0x27,0x0c,0x05,0x82]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 0c 05 82 <unknown>
+# CHECK-UNKNOWN: 82050c27 <unknown>
 
 vsseg5e16.v v24, (a0), v0.t
 # CHECK-INST: vsseg5e16.v v24, (a0), v0.t
 # CHECK-ENCODING: [0x27,0x5c,0x05,0x80]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 5c 05 80 <unknown>
+# CHECK-UNKNOWN: 80055c27 <unknown>
 
 vsseg5e16.v v24, (a0)
 # CHECK-INST: vsseg5e16.v v24, (a0)
 # CHECK-ENCODING: [0x27,0x5c,0x05,0x82]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 5c 05 82 <unknown>
+# CHECK-UNKNOWN: 82055c27 <unknown>
 
 vsseg5e32.v v24, (a0), v0.t
 # CHECK-INST: vsseg5e32.v v24, (a0), v0.t
 # CHECK-ENCODING: [0x27,0x6c,0x05,0x80]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 6c 05 80 <unknown>
+# CHECK-UNKNOWN: 80056c27 <unknown>
 
 vsseg5e32.v v24, (a0)
 # CHECK-INST: vsseg5e32.v v24, (a0)
 # CHECK-ENCODING: [0x27,0x6c,0x05,0x82]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 6c 05 82 <unknown>
+# CHECK-UNKNOWN: 82056c27 <unknown>
 
 vsseg5e64.v v24, (a0), v0.t
 # CHECK-INST: vsseg5e64.v v24, (a0), v0.t
 # CHECK-ENCODING: [0x27,0x7c,0x05,0x80]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 7c 05 80 <unknown>
+# CHECK-UNKNOWN: 80057c27 <unknown>
 
 vsseg5e64.v v24, (a0)
 # CHECK-INST: vsseg5e64.v v24, (a0)
 # CHECK-ENCODING: [0x27,0x7c,0x05,0x82]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 7c 05 82 <unknown>
+# CHECK-UNKNOWN: 82057c27 <unknown>
 
 vssseg5e8.v v24, (a0), a1, v0.t
 # CHECK-INST: vssseg5e8.v v24, (a0), a1, v0.t
 # CHECK-ENCODING: [0x27,0x0c,0xb5,0x88]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 0c b5 88 <unknown>
+# CHECK-UNKNOWN: 88b50c27 <unknown>
 
 vssseg5e8.v v24, (a0), a1
 # CHECK-INST: vssseg5e8.v v24, (a0), a1
 # CHECK-ENCODING: [0x27,0x0c,0xb5,0x8a]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 0c b5 8a <unknown>
+# CHECK-UNKNOWN: 8ab50c27 <unknown>
 
 vssseg5e16.v v24, (a0), a1, v0.t
 # CHECK-INST: vssseg5e16.v v24, (a0), a1, v0.t
 # CHECK-ENCODING: [0x27,0x5c,0xb5,0x88]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 5c b5 88 <unknown>
+# CHECK-UNKNOWN: 88b55c27 <unknown>
 
 vssseg5e16.v v24, (a0), a1
 # CHECK-INST: vssseg5e16.v v24, (a0), a1
 # CHECK-ENCODING: [0x27,0x5c,0xb5,0x8a]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 5c b5 8a <unknown>
+# CHECK-UNKNOWN: 8ab55c27 <unknown>
 
 vssseg5e32.v v24, (a0), a1, v0.t
 # CHECK-INST: vssseg5e32.v v24, (a0), a1, v0.t
 # CHECK-ENCODING: [0x27,0x6c,0xb5,0x88]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 6c b5 88 <unknown>
+# CHECK-UNKNOWN: 88b56c27 <unknown>
 
 vssseg5e32.v v24, (a0), a1
 # CHECK-INST: vssseg5e32.v v24, (a0), a1
 # CHECK-ENCODING: [0x27,0x6c,0xb5,0x8a]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 6c b5 8a <unknown>
+# CHECK-UNKNOWN: 8ab56c27 <unknown>
 
 vssseg5e64.v v24, (a0), a1, v0.t
 # CHECK-INST: vssseg5e64.v v24, (a0), a1, v0.t
 # CHECK-ENCODING: [0x27,0x7c,0xb5,0x88]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 7c b5 88 <unknown>
+# CHECK-UNKNOWN: 88b57c27 <unknown>
 
 vssseg5e64.v v24, (a0), a1
 # CHECK-INST: vssseg5e64.v v24, (a0), a1
 # CHECK-ENCODING: [0x27,0x7c,0xb5,0x8a]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 7c b5 8a <unknown>
+# CHECK-UNKNOWN: 8ab57c27 <unknown>
 
 vsuxseg5ei8.v v24, (a0), v4, v0.t
 # CHECK-INST: vsuxseg5ei8.v v24, (a0), v4, v0.t
 # CHECK-ENCODING: [0x27,0x0c,0x45,0x84]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 0c 45 84 <unknown>
+# CHECK-UNKNOWN: 84450c27 <unknown>
 
 vsuxseg5ei8.v v24, (a0), v4
 # CHECK-INST: vsuxseg5ei8.v v24, (a0), v4
 # CHECK-ENCODING: [0x27,0x0c,0x45,0x86]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 0c 45 86 <unknown>
+# CHECK-UNKNOWN: 86450c27 <unknown>
 
 vsuxseg5ei16.v v24, (a0), v4, v0.t
 # CHECK-INST: vsuxseg5ei16.v v24, (a0), v4, v0.t
 # CHECK-ENCODING: [0x27,0x5c,0x45,0x84]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 5c 45 84 <unknown>
+# CHECK-UNKNOWN: 84455c27 <unknown>
 
 vsuxseg5ei16.v v24, (a0), v4
 # CHECK-INST: vsuxseg5ei16.v v24, (a0), v4
 # CHECK-ENCODING: [0x27,0x5c,0x45,0x86]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 5c 45 86 <unknown>
+# CHECK-UNKNOWN: 86455c27 <unknown>
 
 vsuxseg5ei32.v v24, (a0), v4, v0.t
 # CHECK-INST: vsuxseg5ei32.v v24, (a0), v4, v0.t
 # CHECK-ENCODING: [0x27,0x6c,0x45,0x84]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 6c 45 84 <unknown>
+# CHECK-UNKNOWN: 84456c27 <unknown>
 
 vsuxseg5ei32.v v24, (a0), v4
 # CHECK-INST: vsuxseg5ei32.v v24, (a0), v4
 # CHECK-ENCODING: [0x27,0x6c,0x45,0x86]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 6c 45 86 <unknown>
+# CHECK-UNKNOWN: 86456c27 <unknown>
 
 vsuxseg5ei64.v v24, (a0), v4, v0.t
 # CHECK-INST: vsuxseg5ei64.v v24, (a0), v4, v0.t
 # CHECK-ENCODING: [0x27,0x7c,0x45,0x84]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 7c 45 84 <unknown>
+# CHECK-UNKNOWN: 84457c27 <unknown>
 
 vsuxseg5ei64.v v24, (a0), v4
 # CHECK-INST: vsuxseg5ei64.v v24, (a0), v4
 # CHECK-ENCODING: [0x27,0x7c,0x45,0x86]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 7c 45 86 <unknown>
+# CHECK-UNKNOWN: 86457c27 <unknown>
 
 vsoxseg5ei8.v v24, (a0), v4, v0.t
 # CHECK-INST: vsoxseg5ei8.v v24, (a0), v4, v0.t
 # CHECK-ENCODING: [0x27,0x0c,0x45,0x8c]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 0c 45 8c <unknown>
+# CHECK-UNKNOWN: 8c450c27 <unknown>
 
 vsoxseg5ei8.v v24, (a0), v4
 # CHECK-INST: vsoxseg5ei8.v v24, (a0), v4
 # CHECK-ENCODING: [0x27,0x0c,0x45,0x8e]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 0c 45 8e <unknown>
+# CHECK-UNKNOWN: 8e450c27 <unknown>
 
 vsoxseg5ei16.v v24, (a0), v4, v0.t
 # CHECK-INST: vsoxseg5ei16.v v24, (a0), v4, v0.t
 # CHECK-ENCODING: [0x27,0x5c,0x45,0x8c]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 5c 45 8c <unknown>
+# CHECK-UNKNOWN: 8c455c27 <unknown>
 
 vsoxseg5ei16.v v24, (a0), v4
 # CHECK-INST: vsoxseg5ei16.v v24, (a0), v4
 # CHECK-ENCODING: [0x27,0x5c,0x45,0x8e]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 5c 45 8e <unknown>
+# CHECK-UNKNOWN: 8e455c27 <unknown>
 
 vsoxseg5ei32.v v24, (a0), v4, v0.t
 # CHECK-INST: vsoxseg5ei32.v v24, (a0), v4, v0.t
 # CHECK-ENCODING: [0x27,0x6c,0x45,0x8c]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 6c 45 8c <unknown>
+# CHECK-UNKNOWN: 8c456c27 <unknown>
 
 vsoxseg5ei32.v v24, (a0), v4
 # CHECK-INST: vsoxseg5ei32.v v24, (a0), v4
 # CHECK-ENCODING: [0x27,0x6c,0x45,0x8e]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 6c 45 8e <unknown>
+# CHECK-UNKNOWN: 8e456c27 <unknown>
 
 vsoxseg5ei64.v v24, (a0), v4, v0.t
 # CHECK-INST: vsoxseg5ei64.v v24, (a0), v4, v0.t
 # CHECK-ENCODING: [0x27,0x7c,0x45,0x8c]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 7c 45 8c <unknown>
+# CHECK-UNKNOWN: 8c457c27 <unknown>
 
 vsoxseg5ei64.v v24, (a0), v4
 # CHECK-INST: vsoxseg5ei64.v v24, (a0), v4
 # CHECK-ENCODING: [0x27,0x7c,0x45,0x8e]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 7c 45 8e <unknown>
+# CHECK-UNKNOWN: 8e457c27 <unknown>
 
 vsseg6e8.v v24, (a0), v0.t
 # CHECK-INST: vsseg6e8.v v24, (a0), v0.t
 # CHECK-ENCODING: [0x27,0x0c,0x05,0xa0]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 0c 05 a0 <unknown>
+# CHECK-UNKNOWN: a0050c27 <unknown>
 
 vsseg6e8.v v24, (a0)
 # CHECK-INST: vsseg6e8.v v24, (a0)
 # CHECK-ENCODING: [0x27,0x0c,0x05,0xa2]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 0c 05 a2 <unknown>
+# CHECK-UNKNOWN: a2050c27 <unknown>
 
 vsseg6e16.v v24, (a0), v0.t
 # CHECK-INST: vsseg6e16.v v24, (a0), v0.t
 # CHECK-ENCODING: [0x27,0x5c,0x05,0xa0]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 5c 05 a0 <unknown>
+# CHECK-UNKNOWN: a0055c27 <unknown>
 
 vsseg6e16.v v24, (a0)
 # CHECK-INST: vsseg6e16.v v24, (a0)
 # CHECK-ENCODING: [0x27,0x5c,0x05,0xa2]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 5c 05 a2 <unknown>
+# CHECK-UNKNOWN: a2055c27 <unknown>
 
 vsseg6e32.v v24, (a0), v0.t
 # CHECK-INST: vsseg6e32.v v24, (a0), v0.t
 # CHECK-ENCODING: [0x27,0x6c,0x05,0xa0]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 6c 05 a0 <unknown>
+# CHECK-UNKNOWN: a0056c27 <unknown>
 
 vsseg6e32.v v24, (a0)
 # CHECK-INST: vsseg6e32.v v24, (a0)
 # CHECK-ENCODING: [0x27,0x6c,0x05,0xa2]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 6c 05 a2 <unknown>
+# CHECK-UNKNOWN: a2056c27 <unknown>
 
 vsseg6e64.v v24, (a0), v0.t
 # CHECK-INST: vsseg6e64.v v24, (a0), v0.t
 # CHECK-ENCODING: [0x27,0x7c,0x05,0xa0]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 7c 05 a0 <unknown>
+# CHECK-UNKNOWN: a0057c27 <unknown>
 
 vsseg6e64.v v24, (a0)
 # CHECK-INST: vsseg6e64.v v24, (a0)
 # CHECK-ENCODING: [0x27,0x7c,0x05,0xa2]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 7c 05 a2 <unknown>
+# CHECK-UNKNOWN: a2057c27 <unknown>
 
 vssseg6e8.v v24, (a0), a1, v0.t
 # CHECK-INST: vssseg6e8.v v24, (a0), a1, v0.t
 # CHECK-ENCODING: [0x27,0x0c,0xb5,0xa8]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 0c b5 a8 <unknown>
+# CHECK-UNKNOWN: a8b50c27 <unknown>
 
 vssseg6e8.v v24, (a0), a1
 # CHECK-INST: vssseg6e8.v v24, (a0), a1
 # CHECK-ENCODING: [0x27,0x0c,0xb5,0xaa]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 0c b5 aa <unknown>
+# CHECK-UNKNOWN: aab50c27 <unknown>
 
 vssseg6e16.v v24, (a0), a1, v0.t
 # CHECK-INST: vssseg6e16.v v24, (a0), a1, v0.t
 # CHECK-ENCODING: [0x27,0x5c,0xb5,0xa8]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 5c b5 a8 <unknown>
+# CHECK-UNKNOWN: a8b55c27 <unknown>
 
 vssseg6e16.v v24, (a0), a1
 # CHECK-INST: vssseg6e16.v v24, (a0), a1
 # CHECK-ENCODING: [0x27,0x5c,0xb5,0xaa]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 5c b5 aa <unknown>
+# CHECK-UNKNOWN: aab55c27 <unknown>
 
 vssseg6e32.v v24, (a0), a1, v0.t
 # CHECK-INST: vssseg6e32.v v24, (a0), a1, v0.t
 # CHECK-ENCODING: [0x27,0x6c,0xb5,0xa8]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 6c b5 a8 <unknown>
+# CHECK-UNKNOWN: a8b56c27 <unknown>
 
 vssseg6e32.v v24, (a0), a1
 # CHECK-INST: vssseg6e32.v v24, (a0), a1
 # CHECK-ENCODING: [0x27,0x6c,0xb5,0xaa]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 6c b5 aa <unknown>
+# CHECK-UNKNOWN: aab56c27 <unknown>
 
 vssseg6e64.v v24, (a0), a1, v0.t
 # CHECK-INST: vssseg6e64.v v24, (a0), a1, v0.t
 # CHECK-ENCODING: [0x27,0x7c,0xb5,0xa8]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 7c b5 a8 <unknown>
+# CHECK-UNKNOWN: a8b57c27 <unknown>
 
 vssseg6e64.v v24, (a0), a1
 # CHECK-INST: vssseg6e64.v v24, (a0), a1
 # CHECK-ENCODING: [0x27,0x7c,0xb5,0xaa]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 7c b5 aa <unknown>
+# CHECK-UNKNOWN: aab57c27 <unknown>
 
 vsuxseg6ei8.v v24, (a0), v4, v0.t
 # CHECK-INST: vsuxseg6ei8.v v24, (a0), v4, v0.t
 # CHECK-ENCODING: [0x27,0x0c,0x45,0xa4]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 0c 45 a4 <unknown>
+# CHECK-UNKNOWN: a4450c27 <unknown>
 
 vsuxseg6ei8.v v24, (a0), v4
 # CHECK-INST: vsuxseg6ei8.v v24, (a0), v4
 # CHECK-ENCODING: [0x27,0x0c,0x45,0xa6]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 0c 45 a6 <unknown>
+# CHECK-UNKNOWN: a6450c27 <unknown>
 
 vsuxseg6ei16.v v24, (a0), v4, v0.t
 # CHECK-INST: vsuxseg6ei16.v v24, (a0), v4, v0.t
 # CHECK-ENCODING: [0x27,0x5c,0x45,0xa4]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 5c 45 a4 <unknown>
+# CHECK-UNKNOWN: a4455c27 <unknown>
 
 vsuxseg6ei16.v v24, (a0), v4
 # CHECK-INST: vsuxseg6ei16.v v24, (a0), v4
 # CHECK-ENCODING: [0x27,0x5c,0x45,0xa6]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 5c 45 a6 <unknown>
+# CHECK-UNKNOWN: a6455c27 <unknown>
 
 vsuxseg6ei32.v v24, (a0), v4, v0.t
 # CHECK-INST: vsuxseg6ei32.v v24, (a0), v4, v0.t
 # CHECK-ENCODING: [0x27,0x6c,0x45,0xa4]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 6c 45 a4 <unknown>
+# CHECK-UNKNOWN: a4456c27 <unknown>
 
 vsuxseg6ei32.v v24, (a0), v4
 # CHECK-INST: vsuxseg6ei32.v v24, (a0), v4
 # CHECK-ENCODING: [0x27,0x6c,0x45,0xa6]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 6c 45 a6 <unknown>
+# CHECK-UNKNOWN: a6456c27 <unknown>
 
 vsuxseg6ei64.v v24, (a0), v4, v0.t
 # CHECK-INST: vsuxseg6ei64.v v24, (a0), v4, v0.t
 # CHECK-ENCODING: [0x27,0x7c,0x45,0xa4]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 7c 45 a4 <unknown>
+# CHECK-UNKNOWN: a4457c27 <unknown>
 
 vsuxseg6ei64.v v24, (a0), v4
 # CHECK-INST: vsuxseg6ei64.v v24, (a0), v4
 # CHECK-ENCODING: [0x27,0x7c,0x45,0xa6]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 7c 45 a6 <unknown>
+# CHECK-UNKNOWN: a6457c27 <unknown>
 
 vsoxseg6ei8.v v24, (a0), v4, v0.t
 # CHECK-INST: vsoxseg6ei8.v v24, (a0), v4, v0.t
 # CHECK-ENCODING: [0x27,0x0c,0x45,0xac]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 0c 45 ac <unknown>
+# CHECK-UNKNOWN: ac450c27 <unknown>
 
 vsoxseg6ei8.v v24, (a0), v4
 # CHECK-INST: vsoxseg6ei8.v v24, (a0), v4
 # CHECK-ENCODING: [0x27,0x0c,0x45,0xae]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 0c 45 ae <unknown>
+# CHECK-UNKNOWN: ae450c27 <unknown>
 
 vsoxseg6ei16.v v24, (a0), v4, v0.t
 # CHECK-INST: vsoxseg6ei16.v v24, (a0), v4, v0.t
 # CHECK-ENCODING: [0x27,0x5c,0x45,0xac]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 5c 45 ac <unknown>
+# CHECK-UNKNOWN: ac455c27 <unknown>
 
 vsoxseg6ei16.v v24, (a0), v4
 # CHECK-INST: vsoxseg6ei16.v v24, (a0), v4
 # CHECK-ENCODING: [0x27,0x5c,0x45,0xae]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 5c 45 ae <unknown>
+# CHECK-UNKNOWN: ae455c27 <unknown>
 
 vsoxseg6ei32.v v24, (a0), v4, v0.t
 # CHECK-INST: vsoxseg6ei32.v v24, (a0), v4, v0.t
 # CHECK-ENCODING: [0x27,0x6c,0x45,0xac]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 6c 45 ac <unknown>
+# CHECK-UNKNOWN: ac456c27 <unknown>
 
 vsoxseg6ei32.v v24, (a0), v4
 # CHECK-INST: vsoxseg6ei32.v v24, (a0), v4
 # CHECK-ENCODING: [0x27,0x6c,0x45,0xae]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 6c 45 ae <unknown>
+# CHECK-UNKNOWN: ae456c27 <unknown>
 
 vsoxseg6ei64.v v24, (a0), v4, v0.t
 # CHECK-INST: vsoxseg6ei64.v v24, (a0), v4, v0.t
 # CHECK-ENCODING: [0x27,0x7c,0x45,0xac]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 7c 45 ac <unknown>
+# CHECK-UNKNOWN: ac457c27 <unknown>
 
 vsoxseg6ei64.v v24, (a0), v4
 # CHECK-INST: vsoxseg6ei64.v v24, (a0), v4
 # CHECK-ENCODING: [0x27,0x7c,0x45,0xae]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 7c 45 ae <unknown>
+# CHECK-UNKNOWN: ae457c27 <unknown>
 
 vsseg7e8.v v24, (a0), v0.t
 # CHECK-INST: vsseg7e8.v v24, (a0), v0.t
 # CHECK-ENCODING: [0x27,0x0c,0x05,0xc0]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 0c 05 c0 <unknown>
+# CHECK-UNKNOWN: c0050c27 <unknown>
 
 vsseg7e8.v v24, (a0)
 # CHECK-INST: vsseg7e8.v v24, (a0)
 # CHECK-ENCODING: [0x27,0x0c,0x05,0xc2]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 0c 05 c2 <unknown>
+# CHECK-UNKNOWN: c2050c27 <unknown>
 
 vsseg7e16.v v24, (a0), v0.t
 # CHECK-INST: vsseg7e16.v v24, (a0), v0.t
 # CHECK-ENCODING: [0x27,0x5c,0x05,0xc0]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 5c 05 c0 <unknown>
+# CHECK-UNKNOWN: c0055c27 <unknown>
 
 vsseg7e16.v v24, (a0)
 # CHECK-INST: vsseg7e16.v v24, (a0)
 # CHECK-ENCODING: [0x27,0x5c,0x05,0xc2]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 5c 05 c2 <unknown>
+# CHECK-UNKNOWN: c2055c27 <unknown>
 
 vsseg7e32.v v24, (a0), v0.t
 # CHECK-INST: vsseg7e32.v v24, (a0), v0.t
 # CHECK-ENCODING: [0x27,0x6c,0x05,0xc0]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 6c 05 c0 <unknown>
+# CHECK-UNKNOWN: c0056c27 <unknown>
 
 vsseg7e32.v v24, (a0)
 # CHECK-INST: vsseg7e32.v v24, (a0)
 # CHECK-ENCODING: [0x27,0x6c,0x05,0xc2]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 6c 05 c2 <unknown>
+# CHECK-UNKNOWN: c2056c27 <unknown>
 
 vsseg7e64.v v24, (a0), v0.t
 # CHECK-INST: vsseg7e64.v v24, (a0), v0.t
 # CHECK-ENCODING: [0x27,0x7c,0x05,0xc0]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 7c 05 c0 <unknown>
+# CHECK-UNKNOWN: c0057c27 <unknown>
 
 vsseg7e64.v v24, (a0)
 # CHECK-INST: vsseg7e64.v v24, (a0)
 # CHECK-ENCODING: [0x27,0x7c,0x05,0xc2]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 7c 05 c2 <unknown>
+# CHECK-UNKNOWN: c2057c27 <unknown>
 
 vssseg7e8.v v24, (a0), a1, v0.t
 # CHECK-INST: vssseg7e8.v v24, (a0), a1, v0.t
 # CHECK-ENCODING: [0x27,0x0c,0xb5,0xc8]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 0c b5 c8 <unknown>
+# CHECK-UNKNOWN: c8b50c27 <unknown>
 
 vssseg7e8.v v24, (a0), a1
 # CHECK-INST: vssseg7e8.v v24, (a0), a1
 # CHECK-ENCODING: [0x27,0x0c,0xb5,0xca]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 0c b5 ca <unknown>
+# CHECK-UNKNOWN: cab50c27 <unknown>
 
 vssseg7e16.v v24, (a0), a1, v0.t
 # CHECK-INST: vssseg7e16.v v24, (a0), a1, v0.t
 # CHECK-ENCODING: [0x27,0x5c,0xb5,0xc8]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 5c b5 c8 <unknown>
+# CHECK-UNKNOWN: c8b55c27 <unknown>
 
 vssseg7e16.v v24, (a0), a1
 # CHECK-INST: vssseg7e16.v v24, (a0), a1
 # CHECK-ENCODING: [0x27,0x5c,0xb5,0xca]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 5c b5 ca <unknown>
+# CHECK-UNKNOWN: cab55c27 <unknown>
 
 vssseg7e32.v v24, (a0), a1, v0.t
 # CHECK-INST: vssseg7e32.v v24, (a0), a1, v0.t
 # CHECK-ENCODING: [0x27,0x6c,0xb5,0xc8]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 6c b5 c8 <unknown>
+# CHECK-UNKNOWN: c8b56c27 <unknown>
 
 vssseg7e32.v v24, (a0), a1
 # CHECK-INST: vssseg7e32.v v24, (a0), a1
 # CHECK-ENCODING: [0x27,0x6c,0xb5,0xca]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 6c b5 ca <unknown>
+# CHECK-UNKNOWN: cab56c27 <unknown>
 
 vssseg7e64.v v24, (a0), a1, v0.t
 # CHECK-INST: vssseg7e64.v v24, (a0), a1, v0.t
 # CHECK-ENCODING: [0x27,0x7c,0xb5,0xc8]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 7c b5 c8 <unknown>
+# CHECK-UNKNOWN: c8b57c27 <unknown>
 
 vssseg7e64.v v24, (a0), a1
 # CHECK-INST: vssseg7e64.v v24, (a0), a1
 # CHECK-ENCODING: [0x27,0x7c,0xb5,0xca]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 7c b5 ca <unknown>
+# CHECK-UNKNOWN: cab57c27 <unknown>
 
 vsuxseg7ei8.v v24, (a0), v4, v0.t
 # CHECK-INST: vsuxseg7ei8.v v24, (a0), v4, v0.t
 # CHECK-ENCODING: [0x27,0x0c,0x45,0xc4]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 0c 45 c4 <unknown>
+# CHECK-UNKNOWN: c4450c27 <unknown>
 
 vsuxseg7ei8.v v24, (a0), v4
 # CHECK-INST: vsuxseg7ei8.v v24, (a0), v4
 # CHECK-ENCODING: [0x27,0x0c,0x45,0xc6]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 0c 45 c6 <unknown>
+# CHECK-UNKNOWN: c6450c27 <unknown>
 
 vsuxseg7ei16.v v24, (a0), v4, v0.t
 # CHECK-INST: vsuxseg7ei16.v v24, (a0), v4, v0.t
 # CHECK-ENCODING: [0x27,0x5c,0x45,0xc4]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 5c 45 c4 <unknown>
+# CHECK-UNKNOWN: c4455c27 <unknown>
 
 vsuxseg7ei16.v v24, (a0), v4
 # CHECK-INST: vsuxseg7ei16.v v24, (a0), v4
 # CHECK-ENCODING: [0x27,0x5c,0x45,0xc6]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 5c 45 c6 <unknown>
+# CHECK-UNKNOWN: c6455c27 <unknown>
 
 vsuxseg7ei32.v v24, (a0), v4, v0.t
 # CHECK-INST: vsuxseg7ei32.v v24, (a0), v4, v0.t
 # CHECK-ENCODING: [0x27,0x6c,0x45,0xc4]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 6c 45 c4 <unknown>
+# CHECK-UNKNOWN: c4456c27 <unknown>
 
 vsuxseg7ei32.v v24, (a0), v4
 # CHECK-INST: vsuxseg7ei32.v v24, (a0), v4
 # CHECK-ENCODING: [0x27,0x6c,0x45,0xc6]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 6c 45 c6 <unknown>
+# CHECK-UNKNOWN: c6456c27 <unknown>
 
 vsuxseg7ei64.v v24, (a0), v4, v0.t
 # CHECK-INST: vsuxseg7ei64.v v24, (a0), v4, v0.t
 # CHECK-ENCODING: [0x27,0x7c,0x45,0xc4]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 7c 45 c4 <unknown>
+# CHECK-UNKNOWN: c4457c27 <unknown>
 
 vsuxseg7ei64.v v24, (a0), v4
 # CHECK-INST: vsuxseg7ei64.v v24, (a0), v4
 # CHECK-ENCODING: [0x27,0x7c,0x45,0xc6]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 7c 45 c6 <unknown>
+# CHECK-UNKNOWN: c6457c27 <unknown>
 
 vsoxseg7ei8.v v24, (a0), v4, v0.t
 # CHECK-INST: vsoxseg7ei8.v v24, (a0), v4, v0.t
 # CHECK-ENCODING: [0x27,0x0c,0x45,0xcc]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 0c 45 cc <unknown>
+# CHECK-UNKNOWN: cc450c27 <unknown>
 
 vsoxseg7ei8.v v24, (a0), v4
 # CHECK-INST: vsoxseg7ei8.v v24, (a0), v4
 # CHECK-ENCODING: [0x27,0x0c,0x45,0xce]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 0c 45 ce <unknown>
+# CHECK-UNKNOWN: ce450c27 <unknown>
 
 vsoxseg7ei16.v v24, (a0), v4, v0.t
 # CHECK-INST: vsoxseg7ei16.v v24, (a0), v4, v0.t
 # CHECK-ENCODING: [0x27,0x5c,0x45,0xcc]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 5c 45 cc <unknown>
+# CHECK-UNKNOWN: cc455c27 <unknown>
 
 vsoxseg7ei16.v v24, (a0), v4
 # CHECK-INST: vsoxseg7ei16.v v24, (a0), v4
 # CHECK-ENCODING: [0x27,0x5c,0x45,0xce]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 5c 45 ce <unknown>
+# CHECK-UNKNOWN: ce455c27 <unknown>
 
 vsoxseg7ei32.v v24, (a0), v4, v0.t
 # CHECK-INST: vsoxseg7ei32.v v24, (a0), v4, v0.t
 # CHECK-ENCODING: [0x27,0x6c,0x45,0xcc]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 6c 45 cc <unknown>
+# CHECK-UNKNOWN: cc456c27 <unknown>
 
 vsoxseg7ei32.v v24, (a0), v4
 # CHECK-INST: vsoxseg7ei32.v v24, (a0), v4
 # CHECK-ENCODING: [0x27,0x6c,0x45,0xce]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 6c 45 ce <unknown>
+# CHECK-UNKNOWN: ce456c27 <unknown>
 
 vsoxseg7ei64.v v24, (a0), v4, v0.t
 # CHECK-INST: vsoxseg7ei64.v v24, (a0), v4, v0.t
 # CHECK-ENCODING: [0x27,0x7c,0x45,0xcc]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 7c 45 cc <unknown>
+# CHECK-UNKNOWN: cc457c27 <unknown>
 
 vsoxseg7ei64.v v24, (a0), v4
 # CHECK-INST: vsoxseg7ei64.v v24, (a0), v4
 # CHECK-ENCODING: [0x27,0x7c,0x45,0xce]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 7c 45 ce <unknown>
+# CHECK-UNKNOWN: ce457c27 <unknown>
 
 vsseg8e8.v v24, (a0), v0.t
 # CHECK-INST: vsseg8e8.v v24, (a0), v0.t
 # CHECK-ENCODING: [0x27,0x0c,0x05,0xe0]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 0c 05 e0 <unknown>
+# CHECK-UNKNOWN: e0050c27 <unknown>
 
 vsseg8e8.v v24, (a0)
 # CHECK-INST: vsseg8e8.v v24, (a0)
 # CHECK-ENCODING: [0x27,0x0c,0x05,0xe2]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 0c 05 e2 <unknown>
+# CHECK-UNKNOWN: e2050c27 <unknown>
 
 vsseg8e16.v v24, (a0), v0.t
 # CHECK-INST: vsseg8e16.v v24, (a0), v0.t
 # CHECK-ENCODING: [0x27,0x5c,0x05,0xe0]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 5c 05 e0 <unknown>
+# CHECK-UNKNOWN: e0055c27 <unknown>
 
 vsseg8e16.v v24, (a0)
 # CHECK-INST: vsseg8e16.v v24, (a0)
 # CHECK-ENCODING: [0x27,0x5c,0x05,0xe2]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 5c 05 e2 <unknown>
+# CHECK-UNKNOWN: e2055c27 <unknown>
 
 vsseg8e32.v v24, (a0), v0.t
 # CHECK-INST: vsseg8e32.v v24, (a0), v0.t
 # CHECK-ENCODING: [0x27,0x6c,0x05,0xe0]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 6c 05 e0 <unknown>
+# CHECK-UNKNOWN: e0056c27 <unknown>
 
 vsseg8e32.v v24, (a0)
 # CHECK-INST: vsseg8e32.v v24, (a0)
 # CHECK-ENCODING: [0x27,0x6c,0x05,0xe2]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 6c 05 e2 <unknown>
+# CHECK-UNKNOWN: e2056c27 <unknown>
 
 vsseg8e64.v v24, (a0), v0.t
 # CHECK-INST: vsseg8e64.v v24, (a0), v0.t
 # CHECK-ENCODING: [0x27,0x7c,0x05,0xe0]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 7c 05 e0 <unknown>
+# CHECK-UNKNOWN: e0057c27 <unknown>
 
 vsseg8e64.v v24, (a0)
 # CHECK-INST: vsseg8e64.v v24, (a0)
 # CHECK-ENCODING: [0x27,0x7c,0x05,0xe2]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 7c 05 e2 <unknown>
+# CHECK-UNKNOWN: e2057c27 <unknown>
 
 vssseg8e8.v v24, (a0), a1, v0.t
 # CHECK-INST: vssseg8e8.v v24, (a0), a1, v0.t
 # CHECK-ENCODING: [0x27,0x0c,0xb5,0xe8]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 0c b5 e8 <unknown>
+# CHECK-UNKNOWN: e8b50c27 <unknown>
 
 vssseg8e8.v v24, (a0), a1
 # CHECK-INST: vssseg8e8.v v24, (a0), a1
 # CHECK-ENCODING: [0x27,0x0c,0xb5,0xea]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 0c b5 ea <unknown>
+# CHECK-UNKNOWN: eab50c27 <unknown>
 
 vssseg8e16.v v24, (a0), a1, v0.t
 # CHECK-INST: vssseg8e16.v v24, (a0), a1, v0.t
 # CHECK-ENCODING: [0x27,0x5c,0xb5,0xe8]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 5c b5 e8 <unknown>
+# CHECK-UNKNOWN: e8b55c27 <unknown>
 
 vssseg8e16.v v24, (a0), a1
 # CHECK-INST: vssseg8e16.v v24, (a0), a1
 # CHECK-ENCODING: [0x27,0x5c,0xb5,0xea]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 5c b5 ea <unknown>
+# CHECK-UNKNOWN: eab55c27 <unknown>
 
 vssseg8e32.v v24, (a0), a1, v0.t
 # CHECK-INST: vssseg8e32.v v24, (a0), a1, v0.t
 # CHECK-ENCODING: [0x27,0x6c,0xb5,0xe8]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 6c b5 e8 <unknown>
+# CHECK-UNKNOWN: e8b56c27 <unknown>
 
 vssseg8e32.v v24, (a0), a1
 # CHECK-INST: vssseg8e32.v v24, (a0), a1
 # CHECK-ENCODING: [0x27,0x6c,0xb5,0xea]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 6c b5 ea <unknown>
+# CHECK-UNKNOWN: eab56c27 <unknown>
 
 vssseg8e64.v v24, (a0), a1, v0.t
 # CHECK-INST: vssseg8e64.v v24, (a0), a1, v0.t
 # CHECK-ENCODING: [0x27,0x7c,0xb5,0xe8]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 7c b5 e8 <unknown>
+# CHECK-UNKNOWN: e8b57c27 <unknown>
 
 vssseg8e64.v v24, (a0), a1
 # CHECK-INST: vssseg8e64.v v24, (a0), a1
 # CHECK-ENCODING: [0x27,0x7c,0xb5,0xea]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 7c b5 ea <unknown>
+# CHECK-UNKNOWN: eab57c27 <unknown>
 
 vsuxseg8ei8.v v24, (a0), v4, v0.t
 # CHECK-INST: vsuxseg8ei8.v v24, (a0), v4, v0.t
 # CHECK-ENCODING: [0x27,0x0c,0x45,0xe4]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 0c 45 e4 <unknown>
+# CHECK-UNKNOWN: e4450c27 <unknown>
 
 vsuxseg8ei8.v v24, (a0), v4
 # CHECK-INST: vsuxseg8ei8.v v24, (a0), v4
 # CHECK-ENCODING: [0x27,0x0c,0x45,0xe6]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 0c 45 e6 <unknown>
+# CHECK-UNKNOWN: e6450c27 <unknown>
 
 vsuxseg8ei16.v v24, (a0), v4, v0.t
 # CHECK-INST: vsuxseg8ei16.v v24, (a0), v4, v0.t
 # CHECK-ENCODING: [0x27,0x5c,0x45,0xe4]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 5c 45 e4 <unknown>
+# CHECK-UNKNOWN: e4455c27 <unknown>
 
 vsuxseg8ei16.v v24, (a0), v4
 # CHECK-INST: vsuxseg8ei16.v v24, (a0), v4
 # CHECK-ENCODING: [0x27,0x5c,0x45,0xe6]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 5c 45 e6 <unknown>
+# CHECK-UNKNOWN: e6455c27 <unknown>
 
 vsuxseg8ei32.v v24, (a0), v4, v0.t
 # CHECK-INST: vsuxseg8ei32.v v24, (a0), v4, v0.t
 # CHECK-ENCODING: [0x27,0x6c,0x45,0xe4]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 6c 45 e4 <unknown>
+# CHECK-UNKNOWN: e4456c27 <unknown>
 
 vsuxseg8ei32.v v24, (a0), v4
 # CHECK-INST: vsuxseg8ei32.v v24, (a0), v4
 # CHECK-ENCODING: [0x27,0x6c,0x45,0xe6]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 6c 45 e6 <unknown>
+# CHECK-UNKNOWN: e6456c27 <unknown>
 
 vsuxseg8ei64.v v24, (a0), v4, v0.t
 # CHECK-INST: vsuxseg8ei64.v v24, (a0), v4, v0.t
 # CHECK-ENCODING: [0x27,0x7c,0x45,0xe4]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 7c 45 e4 <unknown>
+# CHECK-UNKNOWN: e4457c27 <unknown>
 
 vsuxseg8ei64.v v24, (a0), v4
 # CHECK-INST: vsuxseg8ei64.v v24, (a0), v4
 # CHECK-ENCODING: [0x27,0x7c,0x45,0xe6]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 7c 45 e6 <unknown>
+# CHECK-UNKNOWN: e6457c27 <unknown>
 
 vsoxseg8ei8.v v24, (a0), v4, v0.t
 # CHECK-INST: vsoxseg8ei8.v v24, (a0), v4, v0.t
 # CHECK-ENCODING: [0x27,0x0c,0x45,0xec]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 0c 45 ec <unknown>
+# CHECK-UNKNOWN: ec450c27 <unknown>
 
 vsoxseg8ei8.v v24, (a0), v4
 # CHECK-INST: vsoxseg8ei8.v v24, (a0), v4
 # CHECK-ENCODING: [0x27,0x0c,0x45,0xee]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 0c 45 ee <unknown>
+# CHECK-UNKNOWN: ee450c27 <unknown>
 
 vsoxseg8ei16.v v24, (a0), v4, v0.t
 # CHECK-INST: vsoxseg8ei16.v v24, (a0), v4, v0.t
 # CHECK-ENCODING: [0x27,0x5c,0x45,0xec]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 5c 45 ec <unknown>
+# CHECK-UNKNOWN: ec455c27 <unknown>
 
 vsoxseg8ei16.v v24, (a0), v4
 # CHECK-INST: vsoxseg8ei16.v v24, (a0), v4
 # CHECK-ENCODING: [0x27,0x5c,0x45,0xee]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 5c 45 ee <unknown>
+# CHECK-UNKNOWN: ee455c27 <unknown>
 
 vsoxseg8ei32.v v24, (a0), v4, v0.t
 # CHECK-INST: vsoxseg8ei32.v v24, (a0), v4, v0.t
 # CHECK-ENCODING: [0x27,0x6c,0x45,0xec]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 6c 45 ec <unknown>
+# CHECK-UNKNOWN: ec456c27 <unknown>
 
 vsoxseg8ei32.v v24, (a0), v4
 # CHECK-INST: vsoxseg8ei32.v v24, (a0), v4
 # CHECK-ENCODING: [0x27,0x6c,0x45,0xee]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 6c 45 ee <unknown>
+# CHECK-UNKNOWN: ee456c27 <unknown>
 
 vsoxseg8ei64.v v24, (a0), v4, v0.t
 # CHECK-INST: vsoxseg8ei64.v v24, (a0), v4, v0.t
 # CHECK-ENCODING: [0x27,0x7c,0x45,0xec]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 7c 45 ec <unknown>
+# CHECK-UNKNOWN: ec457c27 <unknown>
 
 vsoxseg8ei64.v v24, (a0), v4
 # CHECK-INST: vsoxseg8ei64.v v24, (a0), v4
 # CHECK-ENCODING: [0x27,0x7c,0x45,0xee]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 7c 45 ee <unknown>
+# CHECK-UNKNOWN: ee457c27 <unknown>
 
 vlseg2e8.v v8, 0(a0), v0.t
 # CHECK-INST: vlseg2e8.v v8, (a0), v0.t
 # CHECK-ENCODING: [0x07,0x04,0x05,0x20]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 04 05 20 <unknown>
+# CHECK-UNKNOWN: 20050407 <unknown>
 
 vlseg2e16ff.v v8, 0(a0)
 # CHECK-INST: vlseg2e16ff.v v8, (a0)
 # CHECK-ENCODING: [0x07,0x54,0x05,0x23]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 54 05 23 <unknown>
+# CHECK-UNKNOWN: 23055407 <unknown>
 
 vlsseg2e8.v v8, 0(a0), a1
 # CHECK-INST: vlsseg2e8.v v8, (a0), a1
 # CHECK-ENCODING: [0x07,0x04,0xb5,0x2a]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 04 b5 2a <unknown>
+# CHECK-UNKNOWN: 2ab50407 <unknown>
 
 vluxseg3ei16.v v8, 0(a0), v4
 # CHECK-INST: vluxseg3ei16.v v8, (a0), v4
 # CHECK-ENCODING: [0x07,0x54,0x45,0x46]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 54 45 46 <unknown>
+# CHECK-UNKNOWN: 46455407 <unknown>
 
 vloxseg4ei64.v v8, 0(a0), v4, v0.t
 # CHECK-INST: vloxseg4ei64.v v8, (a0), v4, v0.t
 # CHECK-ENCODING: [0x07,0x74,0x45,0x6c]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors) or 'Zve64x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 07 74 45 6c <unknown>
+# CHECK-UNKNOWN: 6c457407 <unknown>
 
 vsseg5e32.v v24, 0(a0), v0.t
 # CHECK-INST: vsseg5e32.v v24, (a0), v0.t
 # CHECK-ENCODING: [0x27,0x6c,0x05,0x80]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 6c 05 80 <unknown>
+# CHECK-UNKNOWN: 80056c27 <unknown>
 
 vssseg2e8.v v24, 0(a0), a1, v0.t
 # CHECK-INST: vssseg2e8.v v24, (a0), a1, v0.t
 # CHECK-ENCODING: [0x27,0x0c,0xb5,0x28]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 0c b5 28 <unknown>
+# CHECK-UNKNOWN: 28b50c27 <unknown>
 
 vsoxseg7ei16.v v24, 0(a0), v4
 # CHECK-INST: vsoxseg7ei16.v v24, (a0), v4
 # CHECK-ENCODING: [0x27,0x5c,0x45,0xce]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 5c 45 ce <unknown>
+# CHECK-UNKNOWN: ce455c27 <unknown>
 
 vsuxseg6ei32.v v24, 0(a0), v4, v0.t
 # CHECK-INST: vsuxseg6ei32.v v24, (a0), v4, v0.t
 # CHECK-ENCODING: [0x27,0x6c,0x45,0xa4]
 # CHECK-ERROR: instruction requires the following: 'V' (Vector Extension for Application Processors), 'Zve32x' (Vector Extensions for Embedded Processors){{$}}
-# CHECK-UNKNOWN: 27 6c 45 a4 <unknown>
+# CHECK-UNKNOWN: a4456c27 <unknown>
diff --git a/llvm/test/MC/RISCV/rvzabha-invalid.s b/llvm/test/MC/RISCV/rvzabha-invalid.s
index 091f7e59ae18..62cb1d8c8913 100644
--- a/llvm/test/MC/RISCV/rvzabha-invalid.s
+++ b/llvm/test/MC/RISCV/rvzabha-invalid.s
@@ -1,5 +1,5 @@
-# RUN: not llvm-mc -triple riscv32 -mattr=+experimental-zabha < %s 2>&1 | FileCheck %s
-# RUN: not llvm-mc -triple riscv64 -mattr=+experimental-zabha < %s 2>&1 | FileCheck %s
+# RUN: not llvm-mc -triple riscv32 -mattr=+a,+experimental-zabha < %s 2>&1 | FileCheck %s
+# RUN: not llvm-mc -triple riscv64 -mattr=+a,+experimental-zabha < %s 2>&1 | FileCheck %s
 
 # Final operand must have parentheses
 amoswap.b a1, a2, a3 # CHECK: :[[@LINE]]:19: error: expected '(' or optional integer offset
diff --git a/llvm/test/MC/RISCV/rvzabha-valid.s b/llvm/test/MC/RISCV/rvzabha-valid.s
index c1f7f44a04dd..333e58d1268e 100644
--- a/llvm/test/MC/RISCV/rvzabha-valid.s
+++ b/llvm/test/MC/RISCV/rvzabha-valid.s
@@ -1,12 +1,12 @@
-# RUN: llvm-mc %s -triple=riscv32 -mattr=+experimental-zabha -riscv-no-aliases -show-encoding \
+# RUN: llvm-mc %s -triple=riscv32 -mattr=+a,+experimental-zabha -riscv-no-aliases -show-encoding \
 # RUN:     | FileCheck -check-prefixes=CHECK-ASM,CHECK-ASM-AND-OBJ %s
-# RUN: llvm-mc %s -triple=riscv64 -mattr=+experimental-zabha -riscv-no-aliases -show-encoding \
+# RUN: llvm-mc %s -triple=riscv64 -mattr=+a,+experimental-zabha -riscv-no-aliases -show-encoding \
 # RUN:     | FileCheck -check-prefixes=CHECK-ASM,CHECK-ASM-AND-OBJ %s
-# RUN: llvm-mc -filetype=obj -triple=riscv32 -mattr=+experimental-zabha < %s \
-# RUN:     | llvm-objdump --mattr=+experimental-zabha -M no-aliases -d -r - \
+# RUN: llvm-mc -filetype=obj -triple=riscv32 -mattr=+a,+experimental-zabha < %s \
+# RUN:     | llvm-objdump --mattr=+a,+experimental-zabha -M no-aliases -d -r - \
 # RUN:     | FileCheck --check-prefix=CHECK-ASM-AND-OBJ %s
-# RUN: llvm-mc -filetype=obj -triple=riscv64 -mattr=+experimental-zabha < %s \
-# RUN:     | llvm-objdump --mattr=+experimental-zabha -M no-aliases -d -r - \
+# RUN: llvm-mc -filetype=obj -triple=riscv64 -mattr=+a,+experimental-zabha < %s \
+# RUN:     | llvm-objdump --mattr=+a,+experimental-zabha -M no-aliases -d -r - \
 # RUN:     | FileCheck --check-prefix=CHECK-ASM-AND-OBJ %s
 
 # CHECK-ASM-AND-OBJ: amoswap.b a4, ra, (s0)
diff --git a/llvm/test/MC/RISCV/rvzabha-zacas-valid.s b/llvm/test/MC/RISCV/rvzabha-zacas-valid.s
index f1f705e625b8..994112c6b6e5 100644
--- a/llvm/test/MC/RISCV/rvzabha-zacas-valid.s
+++ b/llvm/test/MC/RISCV/rvzabha-zacas-valid.s
@@ -1,16 +1,16 @@
-# RUN: llvm-mc %s -triple=riscv32 -mattr=+experimental-zabha,+zacas -riscv-no-aliases -show-encoding \
+# RUN: llvm-mc %s -triple=riscv32 -mattr=+a,+experimental-zabha,+zacas -riscv-no-aliases -show-encoding \
 # RUN:     | FileCheck -check-prefixes=CHECK-ASM,CHECK-ASM-AND-OBJ %s
-# RUN: llvm-mc %s -triple=riscv64 -mattr=+experimental-zabha,+zacas -riscv-no-aliases -show-encoding \
+# RUN: llvm-mc %s -triple=riscv64 -mattr=+a,+experimental-zabha,+zacas -riscv-no-aliases -show-encoding \
 # RUN:     | FileCheck -check-prefixes=CHECK-ASM,CHECK-ASM-AND-OBJ %s
-# RUN: llvm-mc -filetype=obj -triple=riscv32 -mattr=+experimental-zabha,+zacas < %s \
-# RUN:     | llvm-objdump --mattr=+experimental-zabha,+zacas -M no-aliases -d -r - \
+# RUN: llvm-mc -filetype=obj -triple=riscv32 -mattr=+a,+experimental-zabha,+zacas < %s \
+# RUN:     | llvm-objdump --mattr=+a,+experimental-zabha,+zacas -M no-aliases -d -r - \
 # RUN:     | FileCheck --check-prefix=CHECK-ASM-AND-OBJ %s
-# RUN: llvm-mc -filetype=obj -triple=riscv64 -mattr=+experimental-zabha,+zacas < %s \
-# RUN:     | llvm-objdump --mattr=+experimental-zabha,+zacas -M no-aliases -d -r - \
+# RUN: llvm-mc -filetype=obj -triple=riscv64 -mattr=+a,+experimental-zabha,+zacas < %s \
+# RUN:     | llvm-objdump --mattr=+a,+experimental-zabha,+zacas -M no-aliases -d -r - \
 # RUN:     | FileCheck --check-prefix=CHECK-ASM-AND-OBJ %s
-# RUN: not llvm-mc -triple=riscv32 -mattr=+experimental-zabha -show-encoding %s 2>&1 \
+# RUN: not llvm-mc -triple=riscv32 -mattr=+a,+experimental-zabha -show-encoding %s 2>&1 \
 # RUN:        | FileCheck %s --check-prefix=CHECK-ERROR
-# RUN: not llvm-mc -triple=riscv64 -mattr=+experimental-zabha -show-encoding %s 2>&1 \
+# RUN: not llvm-mc -triple=riscv64 -mattr=+a,+experimental-zabha -show-encoding %s 2>&1 \
 # RUN:        | FileCheck %s --check-prefix=CHECK-ERROR
 
 # CHECK-ASM-AND-OBJ: amocas.b a1, a3, (a5)
diff --git a/llvm/test/MC/RISCV/scoped-relaxation.s b/llvm/test/MC/RISCV/scoped-relaxation.s
index 0b797ee5aca5..56394fd80532 100644
--- a/llvm/test/MC/RISCV/scoped-relaxation.s
+++ b/llvm/test/MC/RISCV/scoped-relaxation.s
@@ -9,7 +9,7 @@
 .dword function - .
 
 # CHECK: 0x0 R_RISCV_ADD64 function 0x0
-# CHECK-NEXT: 0x0 R_RISCV_SUB64 <null> 0x0
+# CHECK-NEXT: 0x0 R_RISCV_SUB64 .L0  0x0
 
 # Relaxed reference, this will resolve to a pair of `RISCV_ADD64` and
 # `RISCV_SUB64` relocation.
@@ -19,7 +19,7 @@
 .option pop
 
 # CHECK: 0x8 R_RISCV_ADD64 function 0x0
-# CHECK-NEXT: 0x8 R_RISCV_SUB64 <null> 0x0
+# CHECK-NEXT: 0x8 R_RISCV_SUB64 .L0  0x0
 
 # Unrelaxed reference, this will resolve to a pair of `RISCV_ADD64` and
 # `RISCV_SUB64` relocation due to relaxation being sticky to the file.
@@ -29,6 +29,6 @@
 .option pop
 
 # CHECK: 0x10 R_RISCV_ADD64 function 0x0
-# CHECK-NEXT: 0x10 R_RISCV_SUB64 <null> 0x0
+# CHECK-NEXT: 0x10 R_RISCV_SUB64 .L0  0x0
 
 # CHECK: }
diff --git a/llvm/test/MC/WebAssembly/global-ctor-dtor.ll b/llvm/test/MC/WebAssembly/global-ctor-dtor.ll
index bc1be7931349..f1ec71da1ebb 100644
--- a/llvm/test/MC/WebAssembly/global-ctor-dtor.ll
+++ b/llvm/test/MC/WebAssembly/global-ctor-dtor.ll
@@ -80,29 +80,29 @@ declare void @func3()
 ; CHECK-NEXT:         Offset:          0x1D
 ; CHECK-NEXT:       - Type:            R_WASM_FUNCTION_INDEX_LEB
 ; CHECK-NEXT:         Index:           6
-; CHECK-NEXT:         Offset:          0x2C
+; CHECK-NEXT:         Offset:          0x2B
 ; CHECK-NEXT:       - Type:            R_WASM_TABLE_INDEX_SLEB
 ; CHECK-NEXT:         Index:           5
-; CHECK-NEXT:         Offset:          0x37
+; CHECK-NEXT:         Offset:          0x36
 ; CHECK-NEXT:       - Type:            R_WASM_MEMORY_ADDR_SLEB
 ; CHECK-NEXT:         Index:           3
-; CHECK-NEXT:         Offset:          0x3F
+; CHECK-NEXT:         Offset:          0x3E
 ; CHECK-NEXT:       - Type:            R_WASM_FUNCTION_INDEX_LEB
 ; CHECK-NEXT:         Index:           4
-; CHECK-NEXT:         Offset:          0x45
+; CHECK-NEXT:         Offset:          0x44
 ; CHECK-NEXT:     Functions:
 ; CHECK-NEXT:       - Index:           5
 ; CHECK-NEXT:         Locals:
 ; CHECK-NEXT:         Body:            1080808080000B
 ; CHECK-NEXT:       - Index:           6
 ; CHECK-NEXT:         Locals:
-; CHECK-NEXT:         Body:            02404181808080004100418080808000108180808000450D0000000B0B
+; CHECK-NEXT:         Body:            02404181808080004100418080808000108180808000450D00000B0B
 ; CHECK-NEXT:       - Index:           7
 ; CHECK-NEXT:         Locals:
 ; CHECK-NEXT:         Body:            1082808080000B
 ; CHECK-NEXT:       - Index:           8
 ; CHECK-NEXT:         Locals:
-; CHECK-NEXT:         Body:            02404182808080004100418080808000108180808000450D0000000B0B
+; CHECK-NEXT:         Body:            02404182808080004100418080808000108180808000450D00000B0B
 ; CHECK-NEXT:   - Type:            DATA
 ; CHECK-NEXT:     Segments:
 ; CHECK-NEXT:       - SectionOffset:   6
diff --git a/llvm/test/MC/Xtensa/Core/invalid.s b/llvm/test/MC/Xtensa/Core/invalid.s
index d3d8fba8169a..c7473e90c10b 100644
--- a/llvm/test/MC/Xtensa/Core/invalid.s
+++ b/llvm/test/MC/Xtensa/Core/invalid.s
@@ -4,10 +4,6 @@ LBL0:
 
 # Out of range immediates
 
-# imm12m
-movi a1, 3000
-# CHECK: :[[#@LINE-1]]:10: error: expected immediate in range [-2048, 2047]
-
 # imm8
 addi a1, a2, 300
 # CHECK: :[[#@LINE-1]]:14: error: expected immediate in range [-128, 127]
diff --git a/llvm/test/MC/Xtensa/directive-literal.s b/llvm/test/MC/Xtensa/directive-literal.s
new file mode 100644
index 000000000000..269cf20ed45e
--- /dev/null
+++ b/llvm/test/MC/Xtensa/directive-literal.s
@@ -0,0 +1,42 @@
+# RUN: llvm-mc -triple=xtensa -filetype obj -o - %s \
+# RUN:   | llvm-readobj -S --sd - \
+# RUN:   | FileCheck -check-prefix=CHECK-LITERAL %s
+
+# RUN: llvm-mc %s -triple=xtensa  -show-encoding \
+# RUN:   | FileCheck -check-prefix=CHECK-INST %s
+
+	.text
+	.literal_position
+	.literal .LCPI0_0, 305419896
+	.literal .LCPI1_0, ext_var
+	.global	test_literal
+	.p2align	2
+	.type	test_literal,@function
+test_literal:
+	l32r	a2, .LCPI0_0
+	l32r	a3, .LCPI1_0
+	movi    a4, 30000
+	movi    a5, 1000
+	ret
+
+# CHECK-LITERAL: Section {
+# CHECK-LITERAL:   Name: .literal
+# CHECK-LITERAL:   SectionData (
+# CHECK-LITERAL:     0000: 78563412 00000000 30750000
+# CHECK-LITERAL:   )
+# CHECK-LITERAL: }
+
+# CHECK-INST: .literal_position
+# CHECK-INST: .literal .LCPI0_0, 305419896
+# CHECK-INST: .literal .LCPI1_0, ext_var
+# CHECK-INST: .global test_literal
+# CHECK-INST: .p2align 2
+# CHECK-INST: .type test_literal,@function
+# CHECK-INST: test_literal:
+# CHECK-INST: l32r a2, .LCPI0_0
+# CHECK-INST: l32r a3, .LCPI1_0
+# CHECK-INST: .literal .Ltmp0, 30000
+# CHECK-INST: l32r a4, .Ltmp0
+# CHECK-INST: movi a5, 1000
+# CHECK-INST: ret
+
diff --git a/llvm/test/MC/Xtensa/invalid-literal.s b/llvm/test/MC/Xtensa/invalid-literal.s
new file mode 100644
index 000000000000..ebb37441059c
--- /dev/null
+++ b/llvm/test/MC/Xtensa/invalid-literal.s
@@ -0,0 +1,10 @@
+# RUN: not llvm-mc %s -triple=xtensa -filetype=asm 2>&1 | FileCheck %s
+
+.text
+.literal_position
+.literal .LCPI0_0  a
+# CHECK: [[@LINE-1]]:20: error: expected comma
+.literal 123, a
+# CHECK: [[@LINE-1]]:10: error: literal label must be a symbol
+.literal .LCPI1_0,
+# CHECK: [[@LINE-1]]:19: error: expected value
diff --git a/llvm/test/TableGen/ContextlessPredicates.td b/llvm/test/TableGen/ContextlessPredicates.td
index 5e4e69069c3e..eead9655111e 100644
--- a/llvm/test/TableGen/ContextlessPredicates.td
+++ b/llvm/test/TableGen/ContextlessPredicates.td
@@ -22,26 +22,26 @@ def : Pat<(test_atomic_op_frag GPR32:$ptr, GPR32:$val) ,
 
 // CHECK_NOPT-LABEL: const uint8_t *MyTargetInstructionSelector::getMatchTable() const {
 // CHECK_NOPT-NEXT:    constexpr static uint8_t MatchTable0[] = {
-// CHECK_NOPT-NEXT:      GIM_Try, /*On fail goto*//*Label 0*/ GIMT_Encode4(58), // Rule ID 0 //
+// CHECK_NOPT-NEXT:      GIM_Try, /*On fail goto*//*Label 0*/ GIMT_Encode4(52), // Rule ID 0 //
 // CHECK_NOPT-NEXT:        GIM_CheckNumOperands, /*MI*/0, /*Expected*/3,
 // CHECK_NOPT-NEXT:        GIM_CheckOpcode, /*MI*/0, GIMT_Encode2(TargetOpcode::G_ATOMICRMW_XCHG),
 // CHECK_NOPT-NEXT:        GIM_CheckMemorySizeEqualTo, /*MI*/0, /*MMO*/0, /*Size*/GIMT_Encode4(4),
 // CHECK_NOPT-NEXT:        // MIs[0] DstI[dst]
-// CHECK_NOPT-NEXT:        GIM_CheckType, /*MI*/0, /*Op*/0, /*Type*/GILLT_s32,
-// CHECK_NOPT-NEXT:        GIM_CheckRegBankForClass, /*MI*/0, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
+// CHECK_NOPT-NEXT:        GIM_RootCheckType, /*Op*/0, /*Type*/GILLT_s32,
+// CHECK_NOPT-NEXT:        GIM_RootCheckRegBankForClass, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
 // CHECK_NOPT-NEXT:        // MIs[0] ptr
 // CHECK_NOPT-NEXT:        GIM_CheckPointerToAny, /*MI*/0, /*Op*/1, /*SizeInBits*/32,
-// CHECK_NOPT-NEXT:        GIM_CheckRegBankForClass, /*MI*/0, /*Op*/1, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
+// CHECK_NOPT-NEXT:        GIM_RootCheckRegBankForClass, /*Op*/1, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
 // CHECK_NOPT-NEXT:        // MIs[0] val
-// CHECK_NOPT-NEXT:        GIM_CheckType, /*MI*/0, /*Op*/2, /*Type*/GILLT_s32,
-// CHECK_NOPT-NEXT:        GIM_CheckRegBankForClass, /*MI*/0, /*Op*/2, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
+// CHECK_NOPT-NEXT:        GIM_RootCheckType, /*Op*/2, /*Type*/GILLT_s32,
+// CHECK_NOPT-NEXT:        GIM_RootCheckRegBankForClass, /*Op*/2, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
 // CHECK_NOPT-NEXT:        GIM_CheckCxxInsnPredicate, /*MI*/0, /*FnId*/GIMT_Encode2(GICXXPred_MI_Predicate_test_atomic_op_frag),
 // CHECK_NOPT-NEXT:        // (atomic_swap:{ *:[i32] } GPR32:{ *:[i32] }:$ptr, GPR32:{ *:[i32] }:$val)<<P:Predicate_test_atomic_op_frag>>  =>  (INSN:{ *:[i32] } GPR32:{ *:[i32] }:$ptr, GPR32:{ *:[i32] }:$val)
 // CHECK_NOPT-NEXT:        GIR_MutateOpcode, /*InsnID*/0, /*RecycleInsnID*/0, /*Opcode*/GIMT_Encode2(MyTarget::INSN),
-// CHECK_NOPT-NEXT:        GIR_ConstrainSelectedInstOperands, /*InsnID*/0,
+// CHECK_NOPT-NEXT:        GIR_RootConstrainSelectedInstOperands,
 // CHECK_NOPT-NEXT:        // GIR_Coverage, 0,
 // CHECK_NOPT-NEXT:        GIR_Done,
-// CHECK_NOPT-NEXT:      // Label 0: @58
+// CHECK_NOPT-NEXT:      // Label 0: @52
 // CHECK_NOPT-NEXT:      GIM_Reject,
 // CHECK_NOPT-NEXT:      };
 // CHECK_NOPT-NEXT:    return MatchTable0;
@@ -49,23 +49,23 @@ def : Pat<(test_atomic_op_frag GPR32:$ptr, GPR32:$val) ,
 
 // CHECK_OPT-LABEL: const uint8_t *MyTargetInstructionSelector::getMatchTable() const {
 // CHECK_OPT-NEXT:   constexpr static uint8_t MatchTable0[] = {
-// CHECK_OPT-NEXT:     GIM_Try, /*On fail goto*//*Label 0*/ GIMT_Encode4(55), // Rule ID 0 //
+// CHECK_OPT-NEXT:     GIM_Try, /*On fail goto*//*Label 0*/ GIMT_Encode4(49), // Rule ID 0 //
 // CHECK_OPT-NEXT:       GIM_CheckOpcode, /*MI*/0, GIMT_Encode2(TargetOpcode::G_ATOMICRMW_XCHG),
-// CHECK_OPT-NEXT:       GIM_CheckType, /*MI*/0, /*Op*/0, /*Type*/GILLT_s32,
-// CHECK_OPT-NEXT:       GIM_CheckType, /*MI*/0, /*Op*/2, /*Type*/GILLT_s32,
+// CHECK_OPT-NEXT:       GIM_RootCheckType, /*Op*/0, /*Type*/GILLT_s32,
+// CHECK_OPT-NEXT:       GIM_RootCheckType, /*Op*/2, /*Type*/GILLT_s32,
 // CHECK_OPT-NEXT:       GIM_CheckMemorySizeEqualTo, /*MI*/0, /*MMO*/0, /*Size*/GIMT_Encode4(4),
-// CHECK_OPT-NEXT:       GIM_CheckRegBankForClass, /*MI*/0, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
+// CHECK_OPT-NEXT:       GIM_RootCheckRegBankForClass, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
 // CHECK_OPT-NEXT:       // MIs[0] ptr
 // CHECK_OPT-NEXT:       GIM_CheckPointerToAny, /*MI*/0, /*Op*/1, /*SizeInBits*/32,
-// CHECK_OPT-NEXT:       GIM_CheckRegBankForClass, /*MI*/0, /*Op*/1, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
-// CHECK_OPT-NEXT:       GIM_CheckRegBankForClass, /*MI*/0, /*Op*/2, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
+// CHECK_OPT-NEXT:       GIM_RootCheckRegBankForClass, /*Op*/1, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
+// CHECK_OPT-NEXT:       GIM_RootCheckRegBankForClass, /*Op*/2, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
 // CHECK_OPT-NEXT:       GIM_CheckCxxInsnPredicate, /*MI*/0, /*FnId*/GIMT_Encode2(GICXXPred_MI_Predicate_test_atomic_op_frag),
 // CHECK_OPT-NEXT:       // (atomic_swap:{ *:[i32] } GPR32:{ *:[i32] }:$ptr, GPR32:{ *:[i32] }:$val)<<P:Predicate_test_atomic_op_frag>>  =>  (INSN:{ *:[i32] } GPR32:{ *:[i32] }:$ptr, GPR32:{ *:[i32] }:$val)
 // CHECK_OPT-NEXT:       GIR_MutateOpcode, /*InsnID*/0, /*RecycleInsnID*/0, /*Opcode*/GIMT_Encode2(MyTarget::INSN),
-// CHECK_OPT-NEXT:       GIR_ConstrainSelectedInstOperands, /*InsnID*/0,
+// CHECK_OPT-NEXT:       GIR_RootConstrainSelectedInstOperands,
 // CHECK_OPT-NEXT:       // GIR_Coverage, 0,
 // CHECK_OPT-NEXT:       GIR_Done,
-// CHECK_OPT-NEXT:     // Label 0: @55
+// CHECK_OPT-NEXT:     // Label 0: @49
 // CHECK_OPT-NEXT:     GIM_Reject,
 // CHECK_OPT-NEXT:     };
 // CHECK_OPT-NEXT:   return MatchTable0;
diff --git a/llvm/test/TableGen/DefaultOpsGlobalISel.td b/llvm/test/TableGen/DefaultOpsGlobalISel.td
index 0c5aa0b912f5..8f4176a2aa73 100644
--- a/llvm/test/TableGen/DefaultOpsGlobalISel.td
+++ b/llvm/test/TableGen/DefaultOpsGlobalISel.td
@@ -33,101 +33,97 @@ def clamp : OperandWithDefaultOps <i1, (ops (i1 0))>;
 
 // CHECK:      const uint8_t *MyTargetInstructionSelector::getMatchTable() const {
 // CHECK-NEXT:   constexpr static uint8_t MatchTable0[] = {
-// CHECK-NEXT:     GIM_Try, /*On fail goto*//*Label 0*/ GIMT_Encode4(79), // Rule ID 3 //
+// CHECK-NEXT:     GIM_Try, /*On fail goto*//*Label 0*/ GIMT_Encode4(69), // Rule ID 3 //
 // CHECK-NEXT:       GIM_CheckNumOperands, /*MI*/0, /*Expected*/3,
 // CHECK-NEXT:       GIM_CheckOpcode, /*MI*/0, GIMT_Encode2(TargetOpcode::G_FMAXNUM),
 // CHECK-NEXT:       // MIs[0] DstI[dst]
-// CHECK-NEXT:       GIM_CheckType, /*MI*/0, /*Op*/0, /*Type*/GILLT_s32,
-// CHECK-NEXT:       GIM_CheckRegBankForClass, /*MI*/0, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::FPR32RegClassID),
+// CHECK-NEXT:       GIM_RootCheckType, /*Op*/0, /*Type*/GILLT_s32,
+// CHECK-NEXT:       GIM_RootCheckRegBankForClass, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::FPR32RegClassID),
 // CHECK-NEXT:       // MIs[0] SelectSrcMods:src0:mods0
-// CHECK-NEXT:       GIM_CheckType, /*MI*/0, /*Op*/1, /*Type*/GILLT_s32,
+// CHECK-NEXT:       GIM_RootCheckType, /*Op*/1, /*Type*/GILLT_s32,
 // CHECK-NEXT:       GIM_CheckComplexPattern, /*MI*/0, /*Op*/1, /*Renderer*/GIMT_Encode2(0), GIMT_Encode2(GICP_gi_SelectSrcMods),
 // CHECK-NEXT:       // MIs[0] SelectSrcMods:src1:mods1
-// CHECK-NEXT:       GIM_CheckType, /*MI*/0, /*Op*/2, /*Type*/GILLT_s32,
+// CHECK-NEXT:       GIM_RootCheckType, /*Op*/2, /*Type*/GILLT_s32,
 // CHECK-NEXT:       GIM_CheckComplexPattern, /*MI*/0, /*Op*/2, /*Renderer*/GIMT_Encode2(1), GIMT_Encode2(GICP_gi_SelectSrcMods),
 // CHECK-NEXT:       // (fmaxnum:{ *:[f32] } (SelectSrcMods:{ *:[f32] } f32:{ *:[f32] }:$src0, src_mods:{ *:[i32] }:$mods0), (SelectSrcMods:{ *:[f32] } f32:{ *:[f32] }:$src1, src_mods:{ *:[i32] }:$mods1))  =>  (FMAX:{ *:[f32] } src_mods:{ *:[i32] }:$mods0, f32:{ *:[f32] }:$src0, src_mods:{ *:[i32] }:$mods1, f32:{ *:[f32] }:$src1)
-// CHECK-NEXT:       GIR_BuildMI, /*InsnID*/0, /*Opcode*/GIMT_Encode2(MyTarget::FMAX),
-// CHECK-NEXT:       GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/0, // DstI[dst]
+// CHECK-NEXT:       GIR_BuildRootMI, /*Opcode*/GIMT_Encode2(MyTarget::FMAX),
+// CHECK-NEXT:       GIR_RootToRootCopy, /*OpIdx*/0, // DstI[dst]
 // CHECK-NEXT:       GIR_ComplexSubOperandRenderer, /*InsnID*/0, /*RendererID*/GIMT_Encode2(0), /*SubOperand*/1, // mods0
 // CHECK-NEXT:       GIR_ComplexSubOperandRenderer, /*InsnID*/0, /*RendererID*/GIMT_Encode2(0), /*SubOperand*/0, // src0
 // CHECK-NEXT:       GIR_ComplexSubOperandRenderer, /*InsnID*/0, /*RendererID*/GIMT_Encode2(1), /*SubOperand*/1, // mods1
 // CHECK-NEXT:       GIR_ComplexSubOperandRenderer, /*InsnID*/0, /*RendererID*/GIMT_Encode2(1), /*SubOperand*/0, // src1
 // CHECK-NEXT:       GIR_AddImm8, /*InsnID*/0, /*Imm*/0,
-// CHECK-NEXT:       GIR_ConstrainSelectedInstOperands, /*InsnID*/0,
-// CHECK-NEXT:       GIR_EraseFromParent, /*InsnID*/0,
+// CHECK-NEXT:       GIR_RootConstrainSelectedInstOperands,
 // CHECK-NEXT:       // GIR_Coverage, 3,
-// CHECK-NEXT:       GIR_Done,
-// CHECK-NEXT:     // Label 0: @79
-// CHECK-NEXT:     GIM_Try, /*On fail goto*//*Label 1*/ GIMT_Encode4(139), // Rule ID 2 //
+// CHECK-NEXT:       GIR_EraseRootFromParent_Done,
+// CHECK-NEXT:     // Label 0: @69
+// CHECK-NEXT:     GIM_Try, /*On fail goto*//*Label 1*/ GIMT_Encode4(120), // Rule ID 2 //
 // CHECK-NEXT:       GIM_CheckNumOperands, /*MI*/0, /*Expected*/2,
 // CHECK-NEXT:       GIM_CheckOpcode, /*MI*/0, GIMT_Encode2(TargetOpcode::G_FFLOOR),
 // CHECK-NEXT:       // MIs[0] DstI[dst]
-// CHECK-NEXT:       GIM_CheckType, /*MI*/0, /*Op*/0, /*Type*/GILLT_s32,
-// CHECK-NEXT:       GIM_CheckRegBankForClass, /*MI*/0, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::FPR32RegClassID),
+// CHECK-NEXT:       GIM_RootCheckType, /*Op*/0, /*Type*/GILLT_s32,
+// CHECK-NEXT:       GIM_RootCheckRegBankForClass, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::FPR32RegClassID),
 // CHECK-NEXT:       // MIs[0] SelectClampOMod:src0:omod:clamp
-// CHECK-NEXT:       GIM_CheckType, /*MI*/0, /*Op*/1, /*Type*/GILLT_s32,
+// CHECK-NEXT:       GIM_RootCheckType, /*Op*/1, /*Type*/GILLT_s32,
 // CHECK-NEXT:       GIM_CheckComplexPattern, /*MI*/0, /*Op*/1, /*Renderer*/GIMT_Encode2(0), GIMT_Encode2(GICP_gi_SelectClampOMod),
 // CHECK-NEXT:       // (ffloor:{ *:[f32] } (SelectClampOMod:{ *:[f32] } f32:{ *:[f32] }:$src0, omod:{ *:[i32] }:$omod, i1:{ *:[i1] }:$clamp))  =>  (FLOMP:{ *:[f32] } f32:{ *:[f32] }:$src0, i1:{ *:[i1] }:$clamp, omod:{ *:[i32] }:$omod)
-// CHECK-NEXT:       GIR_BuildMI, /*InsnID*/0, /*Opcode*/GIMT_Encode2(MyTarget::FLOMP),
-// CHECK-NEXT:       GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/0, // DstI[dst]
+// CHECK-NEXT:       GIR_BuildRootMI, /*Opcode*/GIMT_Encode2(MyTarget::FLOMP),
+// CHECK-NEXT:       GIR_RootToRootCopy, /*OpIdx*/0, // DstI[dst]
 // CHECK-NEXT:       GIR_ComplexSubOperandRenderer, /*InsnID*/0, /*RendererID*/GIMT_Encode2(0), /*SubOperand*/0, // src0
 // CHECK-NEXT:       GIR_ComplexSubOperandRenderer, /*InsnID*/0, /*RendererID*/GIMT_Encode2(0), /*SubOperand*/2, // clamp
 // CHECK-NEXT:       GIR_ComplexSubOperandRenderer, /*InsnID*/0, /*RendererID*/GIMT_Encode2(0), /*SubOperand*/1, // omod
-// CHECK-NEXT:       GIR_ConstrainSelectedInstOperands, /*InsnID*/0,
-// CHECK-NEXT:       GIR_EraseFromParent, /*InsnID*/0,
+// CHECK-NEXT:       GIR_RootConstrainSelectedInstOperands,
 // CHECK-NEXT:       // GIR_Coverage, 2,
-// CHECK-NEXT:       GIR_Done,
-// CHECK-NEXT:     // Label 1: @139
-// CHECK-NEXT:     GIM_Try, /*On fail goto*//*Label 2*/ GIMT_Encode4(207), // Rule ID 8 //
+// CHECK-NEXT:       GIR_EraseRootFromParent_Done,
+// CHECK-NEXT:     // Label 1: @120
+// CHECK-NEXT:     GIM_Try, /*On fail goto*//*Label 2*/ GIMT_Encode4(179), // Rule ID 8 //
 // CHECK-NEXT:       GIM_CheckNumOperands, /*MI*/0, /*Expected*/2,
 // CHECK-NEXT:       GIM_CheckOpcode, /*MI*/0, GIMT_Encode2(TargetOpcode::G_FCANONICALIZE),
 // CHECK-NEXT:       // MIs[0] DstI[dst]
-// CHECK-NEXT:       GIM_CheckType, /*MI*/0, /*Op*/0, /*Type*/GILLT_s32,
-// CHECK-NEXT:       GIM_CheckRegBankForClass, /*MI*/0, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::FPR32RegClassID),
+// CHECK-NEXT:       GIM_RootCheckType, /*Op*/0, /*Type*/GILLT_s32,
+// CHECK-NEXT:       GIM_RootCheckRegBankForClass, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::FPR32RegClassID),
 // CHECK-NEXT:       // MIs[0] SelectSrcMods:src:mods
-// CHECK-NEXT:       GIM_CheckType, /*MI*/0, /*Op*/1, /*Type*/GILLT_s32,
+// CHECK-NEXT:       GIM_RootCheckType, /*Op*/1, /*Type*/GILLT_s32,
 // CHECK-NEXT:       GIM_CheckComplexPattern, /*MI*/0, /*Op*/1, /*Renderer*/GIMT_Encode2(0), GIMT_Encode2(GICP_gi_SelectSrcMods),
 // CHECK-NEXT:       // (fcanonicalize:{ *:[f32] } (SelectSrcMods:{ *:[f32] } f32:{ *:[f32] }:$src, i32:{ *:[i32] }:$mods))  =>  (FMAX:{ *:[f32] } ?:{ *:[i32] }:$mods, ?:{ *:[f32] }:$src, ?:{ *:[i32] }:$mods, ?:{ *:[f32] }:$src, 0:{ *:[i1] })
-// CHECK-NEXT:       GIR_BuildMI, /*InsnID*/0, /*Opcode*/GIMT_Encode2(MyTarget::FMAX),
-// CHECK-NEXT:       GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/0, // DstI[dst]
+// CHECK-NEXT:       GIR_BuildRootMI, /*Opcode*/GIMT_Encode2(MyTarget::FMAX),
+// CHECK-NEXT:       GIR_RootToRootCopy, /*OpIdx*/0, // DstI[dst]
 // CHECK-NEXT:       GIR_ComplexSubOperandRenderer, /*InsnID*/0, /*RendererID*/GIMT_Encode2(0), /*SubOperand*/1, // mods
 // CHECK-NEXT:       GIR_ComplexSubOperandRenderer, /*InsnID*/0, /*RendererID*/GIMT_Encode2(0), /*SubOperand*/0, // src
 // CHECK-NEXT:       GIR_ComplexSubOperandRenderer, /*InsnID*/0, /*RendererID*/GIMT_Encode2(0), /*SubOperand*/1, // mods
 // CHECK-NEXT:       GIR_ComplexSubOperandRenderer, /*InsnID*/0, /*RendererID*/GIMT_Encode2(0), /*SubOperand*/0, // src
 // CHECK-NEXT:       GIR_AddImm8, /*InsnID*/0, /*Imm*/0,
-// CHECK-NEXT:       GIR_ConstrainSelectedInstOperands, /*InsnID*/0,
-// CHECK-NEXT:       GIR_EraseFromParent, /*InsnID*/0,
+// CHECK-NEXT:       GIR_RootConstrainSelectedInstOperands,
 // CHECK-NEXT:       // GIR_Coverage, 8,
-// CHECK-NEXT:       GIR_Done,
-// CHECK-NEXT:     // Label 2: @207
-// CHECK-NEXT:     GIM_Try, /*On fail goto*//*Label 3*/ GIMT_Encode4(265), // Rule ID 5 //
+// CHECK-NEXT:       GIR_EraseRootFromParent_Done,
+// CHECK-NEXT:     // Label 2: @179
+// CHECK-NEXT:     GIM_Try, /*On fail goto*//*Label 3*/ GIMT_Encode4(228), // Rule ID 5 //
 // CHECK-NEXT:       GIM_CheckNumOperands, /*MI*/0, /*Expected*/2,
 // CHECK-NEXT:       GIM_CheckOpcode, /*MI*/0, GIMT_Encode2(TargetOpcode::G_FCOS),
 // CHECK-NEXT:       // MIs[0] DstI[dst]
-// CHECK-NEXT:       GIM_CheckType, /*MI*/0, /*Op*/0, /*Type*/GILLT_s32,
-// CHECK-NEXT:       GIM_CheckRegBankForClass, /*MI*/0, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::FPR32RegClassID),
+// CHECK-NEXT:       GIM_RootCheckType, /*Op*/0, /*Type*/GILLT_s32,
+// CHECK-NEXT:       GIM_RootCheckRegBankForClass, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::FPR32RegClassID),
 // CHECK-NEXT:       // MIs[0] SelectOMod:src0:omod
-// CHECK-NEXT:       GIM_CheckType, /*MI*/0, /*Op*/1, /*Type*/GILLT_s32,
+// CHECK-NEXT:       GIM_RootCheckType, /*Op*/1, /*Type*/GILLT_s32,
 // CHECK-NEXT:       GIM_CheckComplexPattern, /*MI*/0, /*Op*/1, /*Renderer*/GIMT_Encode2(0), GIMT_Encode2(GICP_gi_SelectOMod),
 // CHECK-NEXT:       // (fcos:{ *:[f32] } (SelectOMod:{ *:[f32] } f32:{ *:[f32] }:$src0, i32:{ *:[i32] }:$omod))  =>  (FLAMP:{ *:[f32] } FPR32:{ *:[f32] }:$src0, omod:{ *:[i32] }:$omod)
-// CHECK-NEXT:       GIR_BuildMI, /*InsnID*/0, /*Opcode*/GIMT_Encode2(MyTarget::FLAMP),
-// CHECK-NEXT:       GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/0, // DstI[dst]
+// CHECK-NEXT:       GIR_BuildRootMI, /*Opcode*/GIMT_Encode2(MyTarget::FLAMP),
+// CHECK-NEXT:       GIR_RootToRootCopy, /*OpIdx*/0, // DstI[dst]
 // CHECK-NEXT:       GIR_ComplexSubOperandRenderer, /*InsnID*/0, /*RendererID*/GIMT_Encode2(0), /*SubOperand*/0, // src0
 // CHECK-NEXT:       GIR_ComplexSubOperandRenderer, /*InsnID*/0, /*RendererID*/GIMT_Encode2(0), /*SubOperand*/1, // omod
 // CHECK-NEXT:       GIR_AddImm8, /*InsnID*/0, /*Imm*/0,
-// CHECK-NEXT:       GIR_ConstrainSelectedInstOperands, /*InsnID*/0,
-// CHECK-NEXT:       GIR_EraseFromParent, /*InsnID*/0,
+// CHECK-NEXT:       GIR_RootConstrainSelectedInstOperands,
 // CHECK-NEXT:       // GIR_Coverage, 5,
-// CHECK-NEXT:       GIR_Done,
-// CHECK-NEXT:     // Label 3: @265
-// CHECK-NEXT:     GIM_Try, /*On fail goto*//*Label 4*/ GIMT_Encode4(345), // Rule ID 7 //
+// CHECK-NEXT:       GIR_EraseRootFromParent_Done,
+// CHECK-NEXT:     // Label 3: @228
+// CHECK-NEXT:     GIM_Try, /*On fail goto*//*Label 4*/ GIMT_Encode4(299), // Rule ID 7 //
 // CHECK-NEXT:       GIM_CheckNumOperands, /*MI*/0, /*Expected*/2,
 // CHECK-NEXT:       GIM_CheckOpcode, /*MI*/0, GIMT_Encode2(TargetOpcode::G_FEXP2),
 // CHECK-NEXT:       // MIs[0] DstI[dst]
-// CHECK-NEXT:       GIM_CheckType, /*MI*/0, /*Op*/0, /*Type*/GILLT_s32,
-// CHECK-NEXT:       GIM_CheckRegBankForClass, /*MI*/0, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::FPR32RegClassID),
+// CHECK-NEXT:       GIM_RootCheckType, /*Op*/0, /*Type*/GILLT_s32,
+// CHECK-NEXT:       GIM_RootCheckRegBankForClass, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::FPR32RegClassID),
 // CHECK-NEXT:       // MIs[0] SelectClamp:src0:clamp
-// CHECK-NEXT:       GIM_CheckType, /*MI*/0, /*Op*/1, /*Type*/GILLT_s32,
+// CHECK-NEXT:       GIM_RootCheckType, /*Op*/1, /*Type*/GILLT_s32,
 // CHECK-NEXT:       GIM_CheckComplexPattern, /*MI*/0, /*Op*/1, /*Renderer*/GIMT_Encode2(0), GIMT_Encode2(GICP_gi_SelectClamp),
 // CHECK-NEXT:       // (fexp2:{ *:[f32] } (SelectClamp:{ *:[f32] } f32:{ *:[f32] }:$src0, i1:{ *:[i1] }:$clamp))  =>  (FEEPLE:{ *:[f32] } FPR32:{ *:[f32] }:$src0, (FFOO:{ *:[f32] } FPR32:{ *:[f32] }:$src0), clamp:{ *:[i1] }:$clamp)
 // CHECK-NEXT:       GIR_MakeTempReg, /*TempRegID*/0, /*TypeID*/GILLT_s32,
@@ -136,93 +132,88 @@ def clamp : OperandWithDefaultOps <i1, (ops (i1 0))>;
 // CHECK-NEXT:       GIR_ComplexSubOperandRenderer, /*InsnID*/1, /*RendererID*/GIMT_Encode2(0), /*SubOperand*/0, // src0
 // CHECK-NEXT:       GIR_AddImm8, /*InsnID*/1, /*Imm*/0,
 // CHECK-NEXT:       GIR_ConstrainSelectedInstOperands, /*InsnID*/1,
-// CHECK-NEXT:       GIR_BuildMI, /*InsnID*/0, /*Opcode*/GIMT_Encode2(MyTarget::FEEPLE),
-// CHECK-NEXT:       GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/0, // DstI[dst]
+// CHECK-NEXT:       GIR_BuildRootMI, /*Opcode*/GIMT_Encode2(MyTarget::FEEPLE),
+// CHECK-NEXT:       GIR_RootToRootCopy, /*OpIdx*/0, // DstI[dst]
 // CHECK-NEXT:       GIR_ComplexSubOperandRenderer, /*InsnID*/0, /*RendererID*/GIMT_Encode2(0), /*SubOperand*/0, // src0
 // CHECK-NEXT:       GIR_AddSimpleTempRegister, /*InsnID*/0, /*TempRegID*/0,
 // CHECK-NEXT:       GIR_ComplexSubOperandRenderer, /*InsnID*/0, /*RendererID*/GIMT_Encode2(0), /*SubOperand*/1, // clamp
-// CHECK-NEXT:       GIR_ConstrainSelectedInstOperands, /*InsnID*/0,
-// CHECK-NEXT:       GIR_EraseFromParent, /*InsnID*/0,
+// CHECK-NEXT:       GIR_RootConstrainSelectedInstOperands,
 // CHECK-NEXT:       // GIR_Coverage, 7,
-// CHECK-NEXT:       GIR_Done,
-// CHECK-NEXT:     // Label 4: @345
-// CHECK-NEXT:     GIM_Try, /*On fail goto*//*Label 5*/ GIMT_Encode4(400), // Rule ID 0 //
+// CHECK-NEXT:       GIR_EraseRootFromParent_Done,
+// CHECK-NEXT:     // Label 4: @299
+// CHECK-NEXT:     GIM_Try, /*On fail goto*//*Label 5*/ GIMT_Encode4(345), // Rule ID 0 //
 // CHECK-NEXT:       GIM_CheckNumOperands, /*MI*/0, /*Expected*/2,
 // CHECK-NEXT:       GIM_CheckOpcode, /*MI*/0, GIMT_Encode2(TargetOpcode::G_FSIN),
 // CHECK-NEXT:       // MIs[0] DstI[dst]
-// CHECK-NEXT:       GIM_CheckType, /*MI*/0, /*Op*/0, /*Type*/GILLT_s32,
-// CHECK-NEXT:       GIM_CheckRegBankForClass, /*MI*/0, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::FPR32RegClassID),
+// CHECK-NEXT:       GIM_RootCheckType, /*Op*/0, /*Type*/GILLT_s32,
+// CHECK-NEXT:       GIM_RootCheckRegBankForClass, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::FPR32RegClassID),
 // CHECK-NEXT:       // MIs[0] SelectClamp:src0:clamp
-// CHECK-NEXT:       GIM_CheckType, /*MI*/0, /*Op*/1, /*Type*/GILLT_s32,
+// CHECK-NEXT:       GIM_RootCheckType, /*Op*/1, /*Type*/GILLT_s32,
 // CHECK-NEXT:       GIM_CheckComplexPattern, /*MI*/0, /*Op*/1, /*Renderer*/GIMT_Encode2(0), GIMT_Encode2(GICP_gi_SelectClamp),
 // CHECK-NEXT:       // (fsin:{ *:[f32] } (SelectClamp:{ *:[f32] } f32:{ *:[f32] }:$src0, i1:{ *:[i1] }:$clamp))  =>  (FFOO:{ *:[f32] } f32:{ *:[f32] }:$src0, i1:{ *:[i1] }:$clamp)
-// CHECK-NEXT:       GIR_BuildMI, /*InsnID*/0, /*Opcode*/GIMT_Encode2(MyTarget::FFOO),
-// CHECK-NEXT:       GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/0, // DstI[dst]
+// CHECK-NEXT:       GIR_BuildRootMI, /*Opcode*/GIMT_Encode2(MyTarget::FFOO),
+// CHECK-NEXT:       GIR_RootToRootCopy, /*OpIdx*/0, // DstI[dst]
 // CHECK-NEXT:       GIR_ComplexSubOperandRenderer, /*InsnID*/0, /*RendererID*/GIMT_Encode2(0), /*SubOperand*/0, // src0
 // CHECK-NEXT:       GIR_ComplexSubOperandRenderer, /*InsnID*/0, /*RendererID*/GIMT_Encode2(0), /*SubOperand*/1, // clamp
-// CHECK-NEXT:       GIR_ConstrainSelectedInstOperands, /*InsnID*/0,
-// CHECK-NEXT:       GIR_EraseFromParent, /*InsnID*/0,
+// CHECK-NEXT:       GIR_RootConstrainSelectedInstOperands,
 // CHECK-NEXT:       // GIR_Coverage, 0,
-// CHECK-NEXT:       GIR_Done,
-// CHECK-NEXT:     // Label 5: @400
-// CHECK-NEXT:     GIM_Try, /*On fail goto*//*Label 6*/ GIMT_Encode4(458), // Rule ID 6 //
+// CHECK-NEXT:       GIR_EraseRootFromParent_Done,
+// CHECK-NEXT:     // Label 5: @345
+// CHECK-NEXT:     GIM_Try, /*On fail goto*//*Label 6*/ GIMT_Encode4(394), // Rule ID 6 //
 // CHECK-NEXT:       GIM_CheckNumOperands, /*MI*/0, /*Expected*/2,
 // CHECK-NEXT:       GIM_CheckOpcode, /*MI*/0, GIMT_Encode2(TargetOpcode::G_FSQRT),
 // CHECK-NEXT:       // MIs[0] DstI[dst]
-// CHECK-NEXT:       GIM_CheckType, /*MI*/0, /*Op*/0, /*Type*/GILLT_s32,
-// CHECK-NEXT:       GIM_CheckRegBankForClass, /*MI*/0, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::FPR32RegClassID),
+// CHECK-NEXT:       GIM_RootCheckType, /*Op*/0, /*Type*/GILLT_s32,
+// CHECK-NEXT:       GIM_RootCheckRegBankForClass, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::FPR32RegClassID),
 // CHECK-NEXT:       // MIs[0] SelectClamp:src0:clamp
-// CHECK-NEXT:       GIM_CheckType, /*MI*/0, /*Op*/1, /*Type*/GILLT_s32,
+// CHECK-NEXT:       GIM_RootCheckType, /*Op*/1, /*Type*/GILLT_s32,
 // CHECK-NEXT:       GIM_CheckComplexPattern, /*MI*/0, /*Op*/1, /*Renderer*/GIMT_Encode2(0), GIMT_Encode2(GICP_gi_SelectClamp),
 // CHECK-NEXT:       // (fsqrt:{ *:[f32] } (SelectClamp:{ *:[f32] } f32:{ *:[f32] }:$src0, i1:{ *:[i1] }:$clamp))  =>  (FLAMP:{ *:[f32] } FPR32:{ *:[f32] }:$src0, 93:{ *:[i32] }, clamp:{ *:[i1] }:$clamp)
-// CHECK-NEXT:       GIR_BuildMI, /*InsnID*/0, /*Opcode*/GIMT_Encode2(MyTarget::FLAMP),
-// CHECK-NEXT:       GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/0, // DstI[dst]
+// CHECK-NEXT:       GIR_BuildRootMI, /*Opcode*/GIMT_Encode2(MyTarget::FLAMP),
+// CHECK-NEXT:       GIR_RootToRootCopy, /*OpIdx*/0, // DstI[dst]
 // CHECK-NEXT:       GIR_ComplexSubOperandRenderer, /*InsnID*/0, /*RendererID*/GIMT_Encode2(0), /*SubOperand*/0, // src0
 // CHECK-NEXT:       GIR_AddImm8, /*InsnID*/0, /*Imm*/93,
 // CHECK-NEXT:       GIR_ComplexSubOperandRenderer, /*InsnID*/0, /*RendererID*/GIMT_Encode2(0), /*SubOperand*/1, // clamp
-// CHECK-NEXT:       GIR_ConstrainSelectedInstOperands, /*InsnID*/0,
-// CHECK-NEXT:       GIR_EraseFromParent, /*InsnID*/0,
+// CHECK-NEXT:       GIR_RootConstrainSelectedInstOperands,
 // CHECK-NEXT:       // GIR_Coverage, 6,
-// CHECK-NEXT:       GIR_Done,
-// CHECK-NEXT:     // Label 6: @458
-// CHECK-NEXT:     GIM_Try, /*On fail goto*//*Label 7*/ GIMT_Encode4(503), // Rule ID 1 //
+// CHECK-NEXT:       GIR_EraseRootFromParent_Done,
+// CHECK-NEXT:     // Label 6: @394
+// CHECK-NEXT:     GIM_Try, /*On fail goto*//*Label 7*/ GIMT_Encode4(428), // Rule ID 1 //
 // CHECK-NEXT:       GIM_CheckNumOperands, /*MI*/0, /*Expected*/2,
 // CHECK-NEXT:       GIM_CheckOpcode, /*MI*/0, GIMT_Encode2(TargetOpcode::G_INTRINSIC_ROUND),
 // CHECK-NEXT:       // MIs[0] DstI[dst]
-// CHECK-NEXT:       GIM_CheckType, /*MI*/0, /*Op*/0, /*Type*/GILLT_s32,
-// CHECK-NEXT:       GIM_CheckRegBankForClass, /*MI*/0, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::FPR32RegClassID),
+// CHECK-NEXT:       GIM_RootCheckType, /*Op*/0, /*Type*/GILLT_s32,
+// CHECK-NEXT:       GIM_RootCheckRegBankForClass, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::FPR32RegClassID),
 // CHECK-NEXT:       // MIs[0] src0
-// CHECK-NEXT:       GIM_CheckType, /*MI*/0, /*Op*/1, /*Type*/GILLT_s32,
+// CHECK-NEXT:       GIM_RootCheckType, /*Op*/1, /*Type*/GILLT_s32,
 // CHECK-NEXT:       // (fround:{ *:[f32] } f32:{ *:[f32] }:$src0)  =>  (FBAR:{ *:[f32] } f32:{ *:[f32] }:$src0)
-// CHECK-NEXT:       GIR_BuildMI, /*InsnID*/0, /*Opcode*/GIMT_Encode2(MyTarget::FBAR),
-// CHECK-NEXT:       GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/0, // DstI[dst]
-// CHECK-NEXT:       GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/1, // src0
+// CHECK-NEXT:       GIR_BuildRootMI, /*Opcode*/GIMT_Encode2(MyTarget::FBAR),
+// CHECK-NEXT:       GIR_RootToRootCopy, /*OpIdx*/0, // DstI[dst]
+// CHECK-NEXT:       GIR_RootToRootCopy, /*OpIdx*/1, // src0
 // CHECK-NEXT:       GIR_AddImm8, /*InsnID*/0, /*Imm*/0,
-// CHECK-NEXT:       GIR_ConstrainSelectedInstOperands, /*InsnID*/0,
-// CHECK-NEXT:       GIR_EraseFromParent, /*InsnID*/0,
+// CHECK-NEXT:       GIR_RootConstrainSelectedInstOperands,
 // CHECK-NEXT:       // GIR_Coverage, 1,
-// CHECK-NEXT:       GIR_Done,
-// CHECK-NEXT:     // Label 7: @503
-// CHECK-NEXT:     GIM_Try, /*On fail goto*//*Label 8*/ GIMT_Encode4(548), // Rule ID 4 //
+// CHECK-NEXT:       GIR_EraseRootFromParent_Done,
+// CHECK-NEXT:     // Label 7: @428
+// CHECK-NEXT:     GIM_Try, /*On fail goto*//*Label 8*/ GIMT_Encode4(462), // Rule ID 4 //
 // CHECK-NEXT:       GIM_CheckNumOperands, /*MI*/0, /*Expected*/2,
 // CHECK-NEXT:       GIM_CheckOpcode, /*MI*/0, GIMT_Encode2(TargetOpcode::G_INTRINSIC_TRUNC),
 // CHECK-NEXT:       // MIs[0] DstI[dst]
-// CHECK-NEXT:       GIM_CheckType, /*MI*/0, /*Op*/0, /*Type*/GILLT_s32,
-// CHECK-NEXT:       GIM_CheckRegBankForClass, /*MI*/0, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::FPR32RegClassID),
+// CHECK-NEXT:       GIM_RootCheckType, /*Op*/0, /*Type*/GILLT_s32,
+// CHECK-NEXT:       GIM_RootCheckRegBankForClass, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::FPR32RegClassID),
 // CHECK-NEXT:       // MIs[0] src0
-// CHECK-NEXT:       GIM_CheckType, /*MI*/0, /*Op*/1, /*Type*/GILLT_s32,
+// CHECK-NEXT:       GIM_RootCheckType, /*Op*/1, /*Type*/GILLT_s32,
 // CHECK-NEXT:       // (ftrunc:{ *:[f32] } f32:{ *:[f32] }:$src0)  =>  (FFOO:{ *:[f32] } FPR32:{ *:[f32] }:$src0)
-// CHECK-NEXT:       GIR_BuildMI, /*InsnID*/0, /*Opcode*/GIMT_Encode2(MyTarget::FFOO),
-// CHECK-NEXT:       GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/0, // DstI[dst]
-// CHECK-NEXT:       GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/1, // src0
+// CHECK-NEXT:       GIR_BuildRootMI, /*Opcode*/GIMT_Encode2(MyTarget::FFOO),
+// CHECK-NEXT:       GIR_RootToRootCopy, /*OpIdx*/0, // DstI[dst]
+// CHECK-NEXT:       GIR_RootToRootCopy, /*OpIdx*/1, // src0
 // CHECK-NEXT:       GIR_AddImm8, /*InsnID*/0, /*Imm*/0,
-// CHECK-NEXT:       GIR_ConstrainSelectedInstOperands, /*InsnID*/0,
-// CHECK-NEXT:       GIR_EraseFromParent, /*InsnID*/0,
+// CHECK-NEXT:       GIR_RootConstrainSelectedInstOperands,
 // CHECK-NEXT:       // GIR_Coverage, 4,
-// CHECK-NEXT:       GIR_Done,
-// CHECK-NEXT:     // Label 8: @548
+// CHECK-NEXT:       GIR_EraseRootFromParent_Done,
+// CHECK-NEXT:     // Label 8: @462
 // CHECK-NEXT:     GIM_Reject,
-// CHECK-NEXT:     }; // Size: 549 bytes
+// CHECK-NEXT:     }; // Size: 463 bytes
 // CHECK-NEXT:   return MatchTable0;
 // CHECK-NEXT: }
 
diff --git a/llvm/test/TableGen/GlobalISelCombinerEmitter/builtins/match-table-eraseroot.td b/llvm/test/TableGen/GlobalISelCombinerEmitter/builtins/match-table-eraseroot.td
index 25fdd887b20b..c227737080e1 100644
--- a/llvm/test/TableGen/GlobalISelCombinerEmitter/builtins/match-table-eraseroot.td
+++ b/llvm/test/TableGen/GlobalISelCombinerEmitter/builtins/match-table-eraseroot.td
@@ -19,7 +19,7 @@ def MyCombiner: GICombiner<"GenMyCombiner", [
 
 // CHECK:      const uint8_t *GenMyCombiner::getMatchTable() const {
 // CHECK-NEXT:   constexpr static uint8_t MatchTable0[] = {
-// CHECK-NEXT:     GIM_Try, /*On fail goto*//*Label 0*/ GIMT_Encode4(15), // Rule ID 0 //
+// CHECK-NEXT:     GIM_Try, /*On fail goto*//*Label 0*/ GIMT_Encode4(13), // Rule ID 0 //
 // CHECK-NEXT:       GIM_CheckSimplePredicate, GIMT_Encode2(GICXXPred_Simple_IsRule0Enabled),
 // CHECK-NEXT:       GIM_CheckOpcode, /*MI*/0, GIMT_Encode2(TargetOpcode::G_STORE),
 // CHECK-NEXT:       // MIs[0] a
@@ -27,9 +27,8 @@ def MyCombiner: GICombiner<"GenMyCombiner", [
 // CHECK-NEXT:       // MIs[0] b
 // CHECK-NEXT:       // No operand predicates
 // CHECK-NEXT:       // Combiner Rule #0: Test0
-// CHECK-NEXT:       GIR_EraseFromParent, /*InsnID*/0,
-// CHECK-NEXT:       GIR_Done,
-// CHECK-NEXT:     // Label 0: @15
+// CHECK-NEXT:       GIR_EraseRootFromParent_Done,
+// CHECK-NEXT:     // Label 0: @13
 // CHECK-NEXT:     GIM_Reject,
 // CHECK-NEXT:     };
 // CHECK-NEXT:   return MatchTable0;
diff --git a/llvm/test/TableGen/GlobalISelCombinerEmitter/builtins/match-table-replacerreg.td b/llvm/test/TableGen/GlobalISelCombinerEmitter/builtins/match-table-replacerreg.td
index ebb95ccb2104..9c9b39027f8f 100644
--- a/llvm/test/TableGen/GlobalISelCombinerEmitter/builtins/match-table-replacerreg.td
+++ b/llvm/test/TableGen/GlobalISelCombinerEmitter/builtins/match-table-replacerreg.td
@@ -47,16 +47,15 @@ def MyCombiner: GICombiner<"GenMyCombiner", [
 // CHECK-NEXT:       // No operand predicates
 // CHECK-NEXT:       // MIs[1] y
 // CHECK-NEXT:       // No operand predicates
-// CHECK-NEXT:       GIM_CheckIsSafeToFold, /*InsnID*/1,
+// CHECK-NEXT:       GIM_CheckIsSafeToFold, /*NumInsns*/1,
 // CHECK-NEXT:       GIR_MakeTempReg, /*TempRegID*/0, /*TypeID*/GILLT_s32,
 // CHECK-NEXT:       // Combiner Rule #1: ReplaceTemp
-// CHECK-NEXT:       GIR_BuildMI, /*InsnID*/0, /*Opcode*/GIMT_Encode2(TargetOpcode::G_UNMERGE_VALUES),
-// CHECK-NEXT:       GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/0, // a
+// CHECK-NEXT:       GIR_BuildRootMI, /*Opcode*/GIMT_Encode2(TargetOpcode::G_UNMERGE_VALUES),
+// CHECK-NEXT:       GIR_RootToRootCopy, /*OpIdx*/0, // a
 // CHECK-NEXT:       GIR_AddTempRegister, /*InsnID*/0, /*TempRegID*/0, /*TempRegFlags*/GIMT_Encode2(RegState::Define),
 // CHECK-NEXT:       GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/1, /*OpIdx*/2, // y
 // CHECK-NEXT:       GIR_ReplaceRegWithTempReg, /*OldInsnID*/0, /*OldOpIdx*/1, /*TempRegID*/0,
-// CHECK-NEXT:       GIR_EraseFromParent, /*InsnID*/0,
-// CHECK-NEXT:       GIR_Done,
+// CHECK-NEXT:       GIR_EraseRootFromParent_Done,
 // CHECK-NEXT:     // Label 3: @[[L529]]
 // CHECK-NEXT:     GIM_Reject,
 // CHECK-NEXT:     // Label 1: @[[L530]]
@@ -70,11 +69,10 @@ def MyCombiner: GICombiner<"GenMyCombiner", [
 // CHECK-NEXT:       // MIs[1] src
 // CHECK-NEXT:       // No operand predicates
 // CHECK-NEXT:       GIM_CheckCanReplaceReg, /*OldInsnID*/0, /*OldOpIdx*/0, /*NewInsnId*/1, /*NewOpIdx*/1,
-// CHECK-NEXT:       GIM_CheckIsSafeToFold, /*InsnID*/1,
+// CHECK-NEXT:       GIM_CheckIsSafeToFold, /*NumInsns*/1,
 // CHECK-NEXT:       // Combiner Rule #0: ReplaceMatched
 // CHECK-NEXT:       GIR_ReplaceReg, /*OldInsnID*/0, /*OldOpIdx*/0, /*NewInsnId*/1, /*NewOpIdx*/1,
-// CHECK-NEXT:       GIR_EraseFromParent, /*InsnID*/0,
-// CHECK-NEXT:       GIR_Done,
+// CHECK-NEXT:       GIR_EraseRootFromParent_Done,
 // CHECK-NEXT:     // Label 4: @[[L561]]
 // CHECK-NEXT:     GIM_Reject,
 // CHECK-NEXT:     // Label 2: @[[L562]]
diff --git a/llvm/test/TableGen/GlobalISelCombinerEmitter/match-table-imms.td b/llvm/test/TableGen/GlobalISelCombinerEmitter/match-table-imms.td
index 6004a17d351b..d9a8854cd018 100644
--- a/llvm/test/TableGen/GlobalISelCombinerEmitter/match-table-imms.td
+++ b/llvm/test/TableGen/GlobalISelCombinerEmitter/match-table-imms.td
@@ -41,31 +41,29 @@ def MyCombiner: GICombiner<"GenMyCombiner", [
 // CHECK-NEXT:     // Label 0: @[[L462]]
 // CHECK-NEXT:     GIM_Try, /*On fail goto*//*Label 4*/ GIMT_Encode4([[L492:[0-9]+]]), // Rule ID 0 //
 // CHECK-NEXT:       GIM_CheckSimplePredicate, GIMT_Encode2(GICXXPred_Simple_IsRule0Enabled),
-// CHECK-NEXT:       GIM_CheckType, /*MI*/0, /*Op*/1, /*Type*/GILLT_s32,
+// CHECK-NEXT:       GIM_RootCheckType, /*Op*/1, /*Type*/GILLT_s32,
 // CHECK-NEXT:       // MIs[0] a
 // CHECK-NEXT:       // No operand predicates
 // CHECK-NEXT:       GIM_CheckConstantInt8, /*MI*/0, /*Op*/1, 0,
 // CHECK-NEXT:       // Combiner Rule #0: InstTest0
-// CHECK-NEXT:       GIR_BuildMI, /*InsnID*/0, /*Opcode*/GIMT_Encode2(TargetOpcode::COPY),
-// CHECK-NEXT:       GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/0, // a
+// CHECK-NEXT:       GIR_BuildRootMI, /*Opcode*/GIMT_Encode2(TargetOpcode::COPY),
+// CHECK-NEXT:       GIR_RootToRootCopy, /*OpIdx*/0, // a
 // CHECK-NEXT:       GIR_AddImm8, /*InsnID*/0, /*Imm*/0,
-// CHECK-NEXT:       GIR_EraseFromParent, /*InsnID*/0,
-// CHECK-NEXT:       GIR_Done,
+// CHECK-NEXT:       GIR_EraseRootFromParent_Done,
 // CHECK-NEXT:     // Label 4: @[[L492]]
 // CHECK-NEXT:     GIM_Reject,
 // CHECK-NEXT:     // Label 1: @[[L493]]
 // CHECK-NEXT:     GIM_Try, /*On fail goto*//*Label 5*/ GIMT_Encode4([[L538:[0-9]+]]), // Rule ID 2 //
 // CHECK-NEXT:       GIM_CheckSimplePredicate, GIMT_Encode2(GICXXPred_Simple_IsRule2Enabled),
-// CHECK-NEXT:       GIM_CheckType, /*MI*/0, /*Op*/1, /*Type*/GILLT_s32,
+// CHECK-NEXT:       GIM_RootCheckType, /*Op*/1, /*Type*/GILLT_s32,
 // CHECK-NEXT:       // MIs[0] a
 // CHECK-NEXT:       // No operand predicates
 // CHECK-NEXT:       GIM_CheckLiteralInt, /*MI*/0, /*Op*/1, GIMT_Encode8(0),
 // CHECK-NEXT:       // Combiner Rule #2: CImmInstTest1
-// CHECK-NEXT:       GIR_BuildMI, /*InsnID*/0, /*Opcode*/GIMT_Encode2(TargetOpcode::G_CONSTANT),
-// CHECK-NEXT:       GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/0, // a
+// CHECK-NEXT:       GIR_BuildRootMI, /*Opcode*/GIMT_Encode2(TargetOpcode::G_CONSTANT),
+// CHECK-NEXT:       GIR_RootToRootCopy, /*OpIdx*/0, // a
 // CHECK-NEXT:       GIR_AddCImm, /*InsnID*/0, /*Type*/GILLT_s32, /*Imm*/GIMT_Encode8(42),
-// CHECK-NEXT:       GIR_EraseFromParent, /*InsnID*/0,
-// CHECK-NEXT:       GIR_Done,
+// CHECK-NEXT:       GIR_EraseRootFromParent_Done,
 // CHECK-NEXT:     // Label 5: @[[L538]]
 // CHECK-NEXT:     GIM_Reject,
 // CHECK-NEXT:     // Label 2: @{{[0-9]+}}
@@ -78,11 +76,10 @@ def MyCombiner: GICombiner<"GenMyCombiner", [
 // CHECK-NEXT:       GIR_MakeTempReg, /*TempRegID*/0, /*TypeID*/GILLT_s32,
 // CHECK-NEXT:       GIR_BuildConstant, /*TempRegID*/0, /*Val*/GIMT_Encode8(0),
 // CHECK-NEXT:       // Combiner Rule #1: InstTest1
-// CHECK-NEXT:       GIR_BuildMI, /*InsnID*/0, /*Opcode*/GIMT_Encode2(TargetOpcode::COPY),
-// CHECK-NEXT:       GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/0, // a
+// CHECK-NEXT:       GIR_BuildRootMI, /*Opcode*/GIMT_Encode2(TargetOpcode::COPY),
+// CHECK-NEXT:       GIR_RootToRootCopy, /*OpIdx*/0, // a
 // CHECK-NEXT:       GIR_AddSimpleTempRegister, /*InsnID*/0, /*TempRegID*/0,
-// CHECK-NEXT:       GIR_EraseFromParent, /*InsnID*/0,
-// CHECK-NEXT:       GIR_Done,
+// CHECK-NEXT:       GIR_EraseRootFromParent_Done,
 // CHECK-NEXT:     // Label 6: @[[L578]]
 // CHECK-NEXT:     GIM_Reject,
 // CHECK-NEXT:     // Label 3: @[[L579]]
diff --git a/llvm/test/TableGen/GlobalISelCombinerEmitter/match-table-intrinsics.td b/llvm/test/TableGen/GlobalISelCombinerEmitter/match-table-intrinsics.td
index b2dd8b6684b1..365d0c9fbff4 100644
--- a/llvm/test/TableGen/GlobalISelCombinerEmitter/match-table-intrinsics.td
+++ b/llvm/test/TableGen/GlobalISelCombinerEmitter/match-table-intrinsics.td
@@ -43,15 +43,14 @@ def MyCombiner: GICombiner<"GenMyCombiner", [
 // CHECK-NEXT:       GIM_CheckConstantInt8, /*MI*/0, /*Op*/2, 0,
 // CHECK-NEXT:       GIR_MakeTempReg, /*TempRegID*/0, /*TypeID*/GILLT_s32,
 // CHECK-NEXT:       // Combiner Rule #0: IntrinTest0
-// CHECK-NEXT:       GIR_BuildMI, /*InsnID*/0, /*Opcode*/GIMT_Encode2(TargetOpcode::G_INTRINSIC),
+// CHECK-NEXT:       GIR_BuildRootMI, /*Opcode*/GIMT_Encode2(TargetOpcode::G_INTRINSIC),
 // CHECK-NEXT:       GIR_AddTempRegister, /*InsnID*/0, /*TempRegID*/0, /*TempRegFlags*/GIMT_Encode2(RegState::Define),
 // CHECK-NEXT:       GIR_AddIntrinsicID, /*MI*/0, GIMT_Encode2(Intrinsic::0in_1out),
 // CHECK-NEXT:       GIR_BuildMI, /*InsnID*/1, /*Opcode*/GIMT_Encode2(TargetOpcode::G_INTRINSIC),
 // CHECK-NEXT:       GIR_Copy, /*NewInsnID*/1, /*OldInsnID*/0, /*OpIdx*/0, // a
 // CHECK-NEXT:       GIR_AddIntrinsicID, /*MI*/1, GIMT_Encode2(Intrinsic::1in_1out),
 // CHECK-NEXT:       GIR_AddSimpleTempRegister, /*InsnID*/1, /*TempRegID*/0,
-// CHECK-NEXT:       GIR_EraseFromParent, /*InsnID*/0,
-// CHECK-NEXT:       GIR_Done,
+// CHECK-NEXT:       GIR_EraseRootFromParent_Done,
 // CHECK-NEXT:     // Label 3: @[[L72]]
 // CHECK-NEXT:     GIM_Reject,
 // CHECK-NEXT:     // Label 1: @[[L73]]
@@ -65,21 +64,20 @@ def MyCombiner: GICombiner<"GenMyCombiner", [
 // CHECK-NEXT:       // No operand predicates
 // CHECK-NEXT:       GIR_MakeTempReg, /*TempRegID*/0, /*TypeID*/GILLT_s32,
 // CHECK-NEXT:       // Combiner Rule #1: SpecialIntrins
-// CHECK-NEXT:       GIR_BuildMI, /*InsnID*/0, /*Opcode*/GIMT_Encode2(TargetOpcode::G_INTRINSIC_CONVERGENT),
+// CHECK-NEXT:       GIR_BuildRootMI, /*Opcode*/GIMT_Encode2(TargetOpcode::G_INTRINSIC_CONVERGENT),
 // CHECK-NEXT:       GIR_AddTempRegister, /*InsnID*/0, /*TempRegID*/0, /*TempRegFlags*/GIMT_Encode2(RegState::Define),
 // CHECK-NEXT:       GIR_AddIntrinsicID, /*MI*/0, GIMT_Encode2(Intrinsic::convergent_1in_1out),
-// CHECK-NEXT:       GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/2, // b
+// CHECK-NEXT:       GIR_RootToRootCopy, /*OpIdx*/2, // b
 // CHECK-NEXT:       GIR_BuildMI, /*InsnID*/1, /*Opcode*/GIMT_Encode2(TargetOpcode::G_INTRINSIC_CONVERGENT_W_SIDE_EFFECTS),
 // CHECK-NEXT:       GIR_Copy, /*NewInsnID*/1, /*OldInsnID*/0, /*OpIdx*/0, // a
 // CHECK-NEXT:       GIR_AddIntrinsicID, /*MI*/1, GIMT_Encode2(Intrinsic::convergent_sideeffects_1in_1out),
 // CHECK-NEXT:       GIR_AddSimpleTempRegister, /*InsnID*/1, /*TempRegID*/0,
 // CHECK-NEXT:       GIR_MergeMemOperands, /*InsnID*/1, /*NumInsns*/1, /*MergeInsnID's*/0,
-// CHECK-NEXT:       GIR_EraseFromParent, /*InsnID*/0,
-// CHECK-NEXT:       GIR_Done,
+// CHECK-NEXT:       GIR_EraseRootFromParent_Done,
 // CHECK-NEXT:     // Label 4: @[[L131]]
 // CHECK-NEXT:     GIM_Reject,
 // CHECK-NEXT:     // Label 2: @[[L132]]
 // CHECK-NEXT:     GIM_Reject,
-// CHECK-NEXT:     }; // Size: 133 bytes
+// CHECK-NEXT:     }; // Size: 125 bytes
 // CHECK-NEXT:   return MatchTable0;
 // CHECK-NEXT: }
diff --git a/llvm/test/TableGen/GlobalISelCombinerEmitter/match-table-miflags.td b/llvm/test/TableGen/GlobalISelCombinerEmitter/match-table-miflags.td
index 22e4d2d5d9d1..24864e8aef45 100644
--- a/llvm/test/TableGen/GlobalISelCombinerEmitter/match-table-miflags.td
+++ b/llvm/test/TableGen/GlobalISelCombinerEmitter/match-table-miflags.td
@@ -17,7 +17,7 @@ def MyCombiner: GICombiner<"GenMyCombiner", [MIFlagsTest]>;
 
 // CHECK:      const uint8_t *GenMyCombiner::getMatchTable() const {
 // CHECK-NEXT:   constexpr static uint8_t MatchTable0[] = {
-// CHECK-NEXT:     GIM_Try, /*On fail goto*//*Label 0*/ GIMT_Encode4(68), // Rule ID 0 //
+// CHECK-NEXT:     GIM_Try, /*On fail goto*//*Label 0*/ GIMT_Encode4(63), // Rule ID 0 //
 // CHECK-NEXT:       GIM_CheckSimplePredicate, GIMT_Encode2(GICXXPred_Simple_IsRule0Enabled),
 // CHECK-NEXT:       GIM_CheckOpcode, /*MI*/0, GIMT_Encode2(TargetOpcode::G_SEXT),
 // CHECK-NEXT:       // MIs[0] dst
@@ -29,18 +29,17 @@ def MyCombiner: GICombiner<"GenMyCombiner", [MIFlagsTest]>;
 // CHECK-NEXT:       GIM_MIFlagsNot, /*MI*/1, GIMT_Encode4(MachineInstr::FmArcp | MachineInstr::FmNoNans),
 // CHECK-NEXT:       // MIs[1] src
 // CHECK-NEXT:       // No operand predicates
-// CHECK-NEXT:       GIM_CheckIsSafeToFold, /*InsnID*/1,
+// CHECK-NEXT:       GIM_CheckIsSafeToFold, /*NumInsns*/1,
 // CHECK-NEXT:       // Combiner Rule #0: MIFlagsTest
-// CHECK-NEXT:       GIR_BuildMI, /*InsnID*/0, /*Opcode*/GIMT_Encode2(TargetOpcode::G_MUL),
-// CHECK-NEXT:       GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/0, // dst
+// CHECK-NEXT:       GIR_BuildRootMI, /*Opcode*/GIMT_Encode2(TargetOpcode::G_MUL),
+// CHECK-NEXT:       GIR_RootToRootCopy, /*OpIdx*/0, // dst
 // CHECK-NEXT:       GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/1, /*OpIdx*/1, // src
 // CHECK-NEXT:       GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/1, /*OpIdx*/1, // src
 // CHECK-NEXT:       GIR_CopyMIFlags, /*InsnID*/0, /*OldInsnID*/1,
 // CHECK-NEXT:       GIR_SetMIFlags, /*InsnID*/0, GIMT_Encode4(MachineInstr::FmReassoc),
 // CHECK-NEXT:       GIR_UnsetMIFlags, /*InsnID*/0, GIMT_Encode4(MachineInstr::FmNsz | MachineInstr::FmArcp),
-// CHECK-NEXT:       GIR_EraseFromParent, /*InsnID*/0,
-// CHECK-NEXT:       GIR_Done,
-// CHECK-NEXT:     // Label 0: @68
+// CHECK-NEXT:       GIR_EraseRootFromParent_Done,
+// CHECK-NEXT:     // Label 0: @63
 // CHECK-NEXT:     GIM_Reject,
 // CHECK-NEXT:     };
 // CHECK-NEXT:   return MatchTable0;
diff --git a/llvm/test/TableGen/GlobalISelCombinerEmitter/match-table-operand-types.td b/llvm/test/TableGen/GlobalISelCombinerEmitter/match-table-operand-types.td
index 0fb63bce1d6a..a23b54afb512 100644
--- a/llvm/test/TableGen/GlobalISelCombinerEmitter/match-table-operand-types.td
+++ b/llvm/test/TableGen/GlobalISelCombinerEmitter/match-table-operand-types.td
@@ -21,32 +21,31 @@ def MyCombiner: GICombiner<"GenMyCombiner", [
 
 // CHECK:      const uint8_t *GenMyCombiner::getMatchTable() const {
 // CHECK-NEXT:   constexpr static uint8_t MatchTable0[] = {
-// CHECK-NEXT:     GIM_Try, /*On fail goto*//*Label 0*/ GIMT_Encode4(81), // Rule ID 0 //
+// CHECK-NEXT:     GIM_Try, /*On fail goto*//*Label 0*/ GIMT_Encode4(73), // Rule ID 0 //
 // CHECK-NEXT:       GIM_CheckSimplePredicate, GIMT_Encode2(GICXXPred_Simple_IsRule0Enabled),
 // CHECK-NEXT:       GIM_CheckOpcode, /*MI*/0, GIMT_Encode2(TargetOpcode::G_MUL),
-// CHECK-NEXT:       GIM_CheckType, /*MI*/0, /*Op*/0, /*Type*/GILLT_s8,
-// CHECK-NEXT:       GIM_CheckType, /*MI*/0, /*Op*/1, /*Type*/GILLT_s32,
-// CHECK-NEXT:       GIM_CheckType, /*MI*/0, /*Op*/2, /*Type*/GILLT_s32,
+// CHECK-NEXT:       GIM_RootCheckType, /*Op*/0, /*Type*/GILLT_s8,
+// CHECK-NEXT:       GIM_RootCheckType, /*Op*/1, /*Type*/GILLT_s32,
+// CHECK-NEXT:       GIM_RootCheckType, /*Op*/2, /*Type*/GILLT_s32,
 // CHECK-NEXT:       GIM_RecordInsnIgnoreCopies, /*DefineMI*/1, /*MI*/0, /*OpIdx*/2, // MIs[1]
 // CHECK-NEXT:       GIM_CheckOpcode, /*MI*/1, GIMT_Encode2(TargetOpcode::G_MUL),
 // CHECK-NEXT:       GIM_CheckType, /*MI*/1, /*Op*/2, /*Type*/GILLT_s32,
 // CHECK-NEXT:       // MIs[1] b
 // CHECK-NEXT:       GIM_CheckIsSameOperandIgnoreCopies, /*MI*/1, /*OpIdx*/1, /*OtherMI*/0, /*OtherOpIdx*/1,
-// CHECK-NEXT:       GIM_CheckIsSafeToFold, /*InsnID*/1,
+// CHECK-NEXT:       GIM_CheckIsSafeToFold, /*NumInsns*/1,
 // CHECK-NEXT:       GIR_MakeTempReg, /*TempRegID*/0, /*TypeID*/GILLT_s64,
 // CHECK-NEXT:       // Combiner Rule #0: InstTest0
-// CHECK-NEXT:       GIR_BuildMI, /*InsnID*/0, /*Opcode*/GIMT_Encode2(TargetOpcode::G_ADD),
+// CHECK-NEXT:       GIR_BuildRootMI, /*Opcode*/GIMT_Encode2(TargetOpcode::G_ADD),
 // CHECK-NEXT:       GIR_AddTempRegister, /*InsnID*/0, /*TempRegID*/0, /*TempRegFlags*/GIMT_Encode2(RegState::Define),
-// CHECK-NEXT:       GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/1, // b
+// CHECK-NEXT:       GIR_RootToRootCopy, /*OpIdx*/1, // b
 // CHECK-NEXT:       GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/1, /*OpIdx*/2, // c
 // CHECK-NEXT:       GIR_BuildMI, /*InsnID*/1, /*Opcode*/GIMT_Encode2(TargetOpcode::G_ADD),
 // CHECK-NEXT:       GIR_Copy, /*NewInsnID*/1, /*OldInsnID*/0, /*OpIdx*/0, // a
 // CHECK-NEXT:       GIR_Copy, /*NewInsnID*/1, /*OldInsnID*/0, /*OpIdx*/1, // b
 // CHECK-NEXT:       GIR_AddSimpleTempRegister, /*InsnID*/1, /*TempRegID*/0,
-// CHECK-NEXT:       GIR_EraseFromParent, /*InsnID*/0,
-// CHECK-NEXT:       GIR_Done,
-// CHECK-NEXT:     // Label 0: @81
+// CHECK-NEXT:       GIR_EraseRootFromParent_Done,
+// CHECK-NEXT:     // Label 0: @73
 // CHECK-NEXT:     GIM_Reject,
-// CHECK-NEXT:     }; // Size: 82 bytes
+// CHECK-NEXT:     }; // Size: 74 bytes
 // CHECK-NEXT:   return MatchTable0;
 // CHECK-NEXT: }
diff --git a/llvm/test/TableGen/GlobalISelCombinerEmitter/match-table-patfrag-root.td b/llvm/test/TableGen/GlobalISelCombinerEmitter/match-table-patfrag-root.td
index 016ab05ca01e..5a8b51dfc832 100644
--- a/llvm/test/TableGen/GlobalISelCombinerEmitter/match-table-patfrag-root.td
+++ b/llvm/test/TableGen/GlobalISelCombinerEmitter/match-table-patfrag-root.td
@@ -42,11 +42,10 @@ def MyCombiner: GICombiner<"GenMyCombiner", [
 // CHECK-NEXT:       GIR_MakeTempReg, /*TempRegID*/0, /*TypeID*/GILLT_s32,
 // CHECK-NEXT:       GIR_BuildConstant, /*TempRegID*/0, /*Val*/GIMT_Encode8(0),
 // CHECK-NEXT:       // Combiner Rule #0: Test0 @ [__Test0_match_0[1]]
-// CHECK-NEXT:       GIR_BuildMI, /*InsnID*/0, /*Opcode*/GIMT_Encode2(TargetOpcode::COPY),
-// CHECK-NEXT:       GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/0, // root
+// CHECK-NEXT:       GIR_BuildRootMI, /*Opcode*/GIMT_Encode2(TargetOpcode::COPY),
+// CHECK-NEXT:       GIR_RootToRootCopy, /*OpIdx*/0, // root
 // CHECK-NEXT:       GIR_AddSimpleTempRegister, /*InsnID*/0, /*TempRegID*/0,
-// CHECK-NEXT:       GIR_EraseFromParent, /*InsnID*/0,
-// CHECK-NEXT:       GIR_Done,
+// CHECK-NEXT:       GIR_EraseRootFromParent_Done,
 // CHECK-NEXT:     // Label 4: @[[L297]]
 // CHECK-NEXT:     GIM_Reject,
 // CHECK-NEXT:     // Label 1: @[[L298]]
@@ -59,15 +58,14 @@ def MyCombiner: GICombiner<"GenMyCombiner", [
 // CHECK-NEXT:       GIM_CheckOpcode, /*MI*/1, GIMT_Encode2(TargetOpcode::G_TRUNC),
 // CHECK-NEXT:       // MIs[1] __Test0_match_0.x
 // CHECK-NEXT:       // No operand predicates
-// CHECK-NEXT:       GIM_CheckIsSafeToFold, /*InsnID*/1,
+// CHECK-NEXT:       GIM_CheckIsSafeToFold, /*NumInsns*/1,
 // CHECK-NEXT:       GIR_MakeTempReg, /*TempRegID*/0, /*TypeID*/GILLT_s32,
 // CHECK-NEXT:       GIR_BuildConstant, /*TempRegID*/0, /*Val*/GIMT_Encode8(0),
 // CHECK-NEXT:       // Combiner Rule #0: Test0 @ [__Test0_match_0[0]]
-// CHECK-NEXT:       GIR_BuildMI, /*InsnID*/0, /*Opcode*/GIMT_Encode2(TargetOpcode::COPY),
-// CHECK-NEXT:       GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/0, // root
+// CHECK-NEXT:       GIR_BuildRootMI, /*Opcode*/GIMT_Encode2(TargetOpcode::COPY),
+// CHECK-NEXT:       GIR_RootToRootCopy, /*OpIdx*/0, // root
 // CHECK-NEXT:       GIR_AddSimpleTempRegister, /*InsnID*/0, /*TempRegID*/0,
-// CHECK-NEXT:       GIR_EraseFromParent, /*InsnID*/0,
-// CHECK-NEXT:       GIR_Done,
+// CHECK-NEXT:       GIR_EraseRootFromParent_Done,
 // CHECK-NEXT:     // Label 5: @[[L343]]
 // CHECK-NEXT:     GIM_Reject,
 // CHECK-NEXT:     // Label 2: @[[L344]]
@@ -80,11 +78,10 @@ def MyCombiner: GICombiner<"GenMyCombiner", [
 // CHECK-NEXT:       GIR_MakeTempReg, /*TempRegID*/0, /*TypeID*/GILLT_s32,
 // CHECK-NEXT:       GIR_BuildConstant, /*TempRegID*/0, /*Val*/GIMT_Encode8(0),
 // CHECK-NEXT:       // Combiner Rule #0: Test0 @ [__Test0_match_0[2]]
-// CHECK-NEXT:       GIR_BuildMI, /*InsnID*/0, /*Opcode*/GIMT_Encode2(TargetOpcode::COPY),
-// CHECK-NEXT:       GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/0, // root
+// CHECK-NEXT:       GIR_BuildRootMI, /*Opcode*/GIMT_Encode2(TargetOpcode::COPY),
+// CHECK-NEXT:       GIR_RootToRootCopy, /*OpIdx*/0, // root
 // CHECK-NEXT:       GIR_AddSimpleTempRegister, /*InsnID*/0, /*TempRegID*/0,
-// CHECK-NEXT:       GIR_EraseFromParent, /*InsnID*/0,
-// CHECK-NEXT:       GIR_Done,
+// CHECK-NEXT:       GIR_EraseRootFromParent_Done,
 // CHECK-NEXT:     // Label 6: @[[L379]]
 // CHECK-NEXT:     GIM_Reject,
 // CHECK-NEXT:     // Label 3: @[[L380]]
diff --git a/llvm/test/TableGen/GlobalISelCombinerEmitter/match-table-permutations.td b/llvm/test/TableGen/GlobalISelCombinerEmitter/match-table-permutations.td
index c38c4be9d545..fda57d5b64e0 100644
--- a/llvm/test/TableGen/GlobalISelCombinerEmitter/match-table-permutations.td
+++ b/llvm/test/TableGen/GlobalISelCombinerEmitter/match-table-permutations.td
@@ -159,9 +159,9 @@ def MyCombiner: GICombiner<"GenMyCombiner", [
 
 // CHECK:      const uint8_t *GenMyCombiner::getMatchTable() const {
 // CHECK-NEXT:   constexpr static uint8_t MatchTable0[] = {
-// CHECK-NEXT:     GIM_Try, /*On fail goto*//*Label 0*/ GIMT_Encode4(850),
+// CHECK-NEXT:     GIM_Try, /*On fail goto*//*Label 0*/ GIMT_Encode4(738),
 // CHECK-NEXT:       GIM_CheckOpcode, /*MI*/0, GIMT_Encode2(TargetOpcode::G_AND),
-// CHECK-NEXT:       GIM_Try, /*On fail goto*//*Label 1*/ GIMT_Encode4(99), // Rule ID 7 //
+// CHECK-NEXT:       GIM_Try, /*On fail goto*//*Label 1*/ GIMT_Encode4(88), // Rule ID 7 //
 // CHECK-NEXT:         GIM_CheckSimplePredicate, GIMT_Encode2(GICXXPred_Simple_IsRule0Enabled),
 // CHECK-NEXT:         // MIs[0] dst
 // CHECK-NEXT:         // No operand predicates
@@ -186,21 +186,17 @@ def MyCombiner: GICombiner<"GenMyCombiner", [
 // CHECK-NEXT:         GIM_CheckCxxInsnPredicate, /*MI*/0, /*FnId*/GIMT_Encode2(GICXXPred_MI_Predicate_GICombiner21),
 // CHECK-NEXT:         GIM_CheckCxxInsnPredicate, /*MI*/0, /*FnId*/GIMT_Encode2(GICXXPred_MI_Predicate_GICombiner22),
 // CHECK-NEXT:         GIM_CheckCxxInsnPredicate, /*MI*/0, /*FnId*/GIMT_Encode2(GICXXPred_MI_Predicate_GICombiner23),
-// CHECK-NEXT:         GIM_CheckIsSafeToFold, /*InsnID*/1,
-// CHECK-NEXT:         GIM_CheckIsSafeToFold, /*InsnID*/2,
-// CHECK-NEXT:         GIM_CheckIsSafeToFold, /*InsnID*/3,
-// CHECK-NEXT:         GIM_CheckIsSafeToFold, /*InsnID*/4,
+// CHECK-NEXT:         GIM_CheckIsSafeToFold, /*NumInsns*/4,
 // CHECK-NEXT:         GIR_MakeTempReg, /*TempRegID*/0, /*TypeID*/GILLT_s32,
 // CHECK-NEXT:         GIR_BuildConstant, /*TempRegID*/0, /*Val*/GIMT_Encode8(0),
 // CHECK-NEXT:         // Combiner Rule #0: Test0 @ [a[1], b[1], c[1]]
-// CHECK-NEXT:         GIR_BuildMI, /*InsnID*/0, /*Opcode*/GIMT_Encode2(TargetOpcode::COPY),
-// CHECK-NEXT:         GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/0, // dst
+// CHECK-NEXT:         GIR_BuildRootMI, /*Opcode*/GIMT_Encode2(TargetOpcode::COPY),
+// CHECK-NEXT:         GIR_RootToRootCopy, /*OpIdx*/0, // dst
 // CHECK-NEXT:         GIR_AddSimpleTempRegister, /*InsnID*/0, /*TempRegID*/0,
 // CHECK-NEXT:         GIR_CustomAction, GIMT_Encode2(GICXXCustomAction_CombineApplyGICombiner0),
-// CHECK-NEXT:         GIR_EraseFromParent, /*InsnID*/0,
-// CHECK-NEXT:         GIR_Done,
-// CHECK-NEXT:       // Label 1: @99
-// CHECK-NEXT:       GIM_Try, /*On fail goto*//*Label 2*/ GIMT_Encode4(199), // Rule ID 6 //
+// CHECK-NEXT:         GIR_EraseRootFromParent_Done,
+// CHECK-NEXT:       // Label 1: @88
+// CHECK-NEXT:       GIM_Try, /*On fail goto*//*Label 2*/ GIMT_Encode4(175), // Rule ID 6 //
 // CHECK-NEXT:         GIM_CheckSimplePredicate, GIMT_Encode2(GICXXPred_Simple_IsRule0Enabled),
 // CHECK-NEXT:         // MIs[0] dst
 // CHECK-NEXT:         // No operand predicates
@@ -228,22 +224,17 @@ def MyCombiner: GICombiner<"GenMyCombiner", [
 // CHECK-NEXT:         GIM_CheckCxxInsnPredicate, /*MI*/0, /*FnId*/GIMT_Encode2(GICXXPred_MI_Predicate_GICombiner18),
 // CHECK-NEXT:         GIM_CheckCxxInsnPredicate, /*MI*/0, /*FnId*/GIMT_Encode2(GICXXPred_MI_Predicate_GICombiner19),
 // CHECK-NEXT:         GIM_CheckCxxInsnPredicate, /*MI*/0, /*FnId*/GIMT_Encode2(GICXXPred_MI_Predicate_GICombiner20),
-// CHECK-NEXT:         GIM_CheckIsSafeToFold, /*InsnID*/1,
-// CHECK-NEXT:         GIM_CheckIsSafeToFold, /*InsnID*/2,
-// CHECK-NEXT:         GIM_CheckIsSafeToFold, /*InsnID*/3,
-// CHECK-NEXT:         GIM_CheckIsSafeToFold, /*InsnID*/4,
-// CHECK-NEXT:         GIM_CheckIsSafeToFold, /*InsnID*/5,
+// CHECK-NEXT:         GIM_CheckIsSafeToFold, /*NumInsns*/5,
 // CHECK-NEXT:         GIR_MakeTempReg, /*TempRegID*/0, /*TypeID*/GILLT_s32,
 // CHECK-NEXT:         GIR_BuildConstant, /*TempRegID*/0, /*Val*/GIMT_Encode8(0),
 // CHECK-NEXT:         // Combiner Rule #0: Test0 @ [a[1], b[1], c[0]]
-// CHECK-NEXT:         GIR_BuildMI, /*InsnID*/0, /*Opcode*/GIMT_Encode2(TargetOpcode::COPY),
-// CHECK-NEXT:         GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/0, // dst
+// CHECK-NEXT:         GIR_BuildRootMI, /*Opcode*/GIMT_Encode2(TargetOpcode::COPY),
+// CHECK-NEXT:         GIR_RootToRootCopy, /*OpIdx*/0, // dst
 // CHECK-NEXT:         GIR_AddSimpleTempRegister, /*InsnID*/0, /*TempRegID*/0,
 // CHECK-NEXT:         GIR_CustomAction, GIMT_Encode2(GICXXCustomAction_CombineApplyGICombiner0),
-// CHECK-NEXT:         GIR_EraseFromParent, /*InsnID*/0,
-// CHECK-NEXT:         GIR_Done,
-// CHECK-NEXT:       // Label 2: @199
-// CHECK-NEXT:       GIM_Try, /*On fail goto*//*Label 3*/ GIMT_Encode4(299), // Rule ID 5 //
+// CHECK-NEXT:         GIR_EraseRootFromParent_Done,
+// CHECK-NEXT:       // Label 2: @175
+// CHECK-NEXT:       GIM_Try, /*On fail goto*//*Label 3*/ GIMT_Encode4(262), // Rule ID 5 //
 // CHECK-NEXT:         GIM_CheckSimplePredicate, GIMT_Encode2(GICXXPred_Simple_IsRule0Enabled),
 // CHECK-NEXT:         // MIs[0] dst
 // CHECK-NEXT:         // No operand predicates
@@ -271,22 +262,17 @@ def MyCombiner: GICombiner<"GenMyCombiner", [
 // CHECK-NEXT:         GIM_CheckCxxInsnPredicate, /*MI*/0, /*FnId*/GIMT_Encode2(GICXXPred_MI_Predicate_GICombiner15),
 // CHECK-NEXT:         GIM_CheckCxxInsnPredicate, /*MI*/0, /*FnId*/GIMT_Encode2(GICXXPred_MI_Predicate_GICombiner16),
 // CHECK-NEXT:         GIM_CheckCxxInsnPredicate, /*MI*/0, /*FnId*/GIMT_Encode2(GICXXPred_MI_Predicate_GICombiner17),
-// CHECK-NEXT:         GIM_CheckIsSafeToFold, /*InsnID*/1,
-// CHECK-NEXT:         GIM_CheckIsSafeToFold, /*InsnID*/2,
-// CHECK-NEXT:         GIM_CheckIsSafeToFold, /*InsnID*/3,
-// CHECK-NEXT:         GIM_CheckIsSafeToFold, /*InsnID*/4,
-// CHECK-NEXT:         GIM_CheckIsSafeToFold, /*InsnID*/5,
+// CHECK-NEXT:         GIM_CheckIsSafeToFold, /*NumInsns*/5,
 // CHECK-NEXT:         GIR_MakeTempReg, /*TempRegID*/0, /*TypeID*/GILLT_s32,
 // CHECK-NEXT:         GIR_BuildConstant, /*TempRegID*/0, /*Val*/GIMT_Encode8(0),
 // CHECK-NEXT:         // Combiner Rule #0: Test0 @ [a[1], b[0], c[1]]
-// CHECK-NEXT:         GIR_BuildMI, /*InsnID*/0, /*Opcode*/GIMT_Encode2(TargetOpcode::COPY),
-// CHECK-NEXT:         GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/0, // dst
+// CHECK-NEXT:         GIR_BuildRootMI, /*Opcode*/GIMT_Encode2(TargetOpcode::COPY),
+// CHECK-NEXT:         GIR_RootToRootCopy, /*OpIdx*/0, // dst
 // CHECK-NEXT:         GIR_AddSimpleTempRegister, /*InsnID*/0, /*TempRegID*/0,
 // CHECK-NEXT:         GIR_CustomAction, GIMT_Encode2(GICXXCustomAction_CombineApplyGICombiner0),
-// CHECK-NEXT:         GIR_EraseFromParent, /*InsnID*/0,
-// CHECK-NEXT:         GIR_Done,
-// CHECK-NEXT:       // Label 3: @299
-// CHECK-NEXT:       GIM_Try, /*On fail goto*//*Label 4*/ GIMT_Encode4(409), // Rule ID 4 //
+// CHECK-NEXT:         GIR_EraseRootFromParent_Done,
+// CHECK-NEXT:       // Label 3: @262
+// CHECK-NEXT:       GIM_Try, /*On fail goto*//*Label 4*/ GIMT_Encode4(357), // Rule ID 4 //
 // CHECK-NEXT:         GIM_CheckSimplePredicate, GIMT_Encode2(GICXXPred_Simple_IsRule0Enabled),
 // CHECK-NEXT:         // MIs[0] dst
 // CHECK-NEXT:         // No operand predicates
@@ -317,23 +303,17 @@ def MyCombiner: GICombiner<"GenMyCombiner", [
 // CHECK-NEXT:         GIM_CheckCxxInsnPredicate, /*MI*/0, /*FnId*/GIMT_Encode2(GICXXPred_MI_Predicate_GICombiner12),
 // CHECK-NEXT:         GIM_CheckCxxInsnPredicate, /*MI*/0, /*FnId*/GIMT_Encode2(GICXXPred_MI_Predicate_GICombiner13),
 // CHECK-NEXT:         GIM_CheckCxxInsnPredicate, /*MI*/0, /*FnId*/GIMT_Encode2(GICXXPred_MI_Predicate_GICombiner14),
-// CHECK-NEXT:         GIM_CheckIsSafeToFold, /*InsnID*/1,
-// CHECK-NEXT:         GIM_CheckIsSafeToFold, /*InsnID*/2,
-// CHECK-NEXT:         GIM_CheckIsSafeToFold, /*InsnID*/3,
-// CHECK-NEXT:         GIM_CheckIsSafeToFold, /*InsnID*/4,
-// CHECK-NEXT:         GIM_CheckIsSafeToFold, /*InsnID*/5,
-// CHECK-NEXT:         GIM_CheckIsSafeToFold, /*InsnID*/6,
+// CHECK-NEXT:         GIM_CheckIsSafeToFold, /*NumInsns*/6,
 // CHECK-NEXT:         GIR_MakeTempReg, /*TempRegID*/0, /*TypeID*/GILLT_s32,
 // CHECK-NEXT:         GIR_BuildConstant, /*TempRegID*/0, /*Val*/GIMT_Encode8(0),
 // CHECK-NEXT:         // Combiner Rule #0: Test0 @ [a[1], b[0], c[0]]
-// CHECK-NEXT:         GIR_BuildMI, /*InsnID*/0, /*Opcode*/GIMT_Encode2(TargetOpcode::COPY),
-// CHECK-NEXT:         GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/0, // dst
+// CHECK-NEXT:         GIR_BuildRootMI, /*Opcode*/GIMT_Encode2(TargetOpcode::COPY),
+// CHECK-NEXT:         GIR_RootToRootCopy, /*OpIdx*/0, // dst
 // CHECK-NEXT:         GIR_AddSimpleTempRegister, /*InsnID*/0, /*TempRegID*/0,
 // CHECK-NEXT:         GIR_CustomAction, GIMT_Encode2(GICXXCustomAction_CombineApplyGICombiner0),
-// CHECK-NEXT:         GIR_EraseFromParent, /*InsnID*/0,
-// CHECK-NEXT:         GIR_Done,
-// CHECK-NEXT:       // Label 4: @409
-// CHECK-NEXT:       GIM_Try, /*On fail goto*//*Label 5*/ GIMT_Encode4(509), // Rule ID 3 //
+// CHECK-NEXT:         GIR_EraseRootFromParent_Done,
+// CHECK-NEXT:       // Label 4: @357
+// CHECK-NEXT:       GIM_Try, /*On fail goto*//*Label 5*/ GIMT_Encode4(444), // Rule ID 3 //
 // CHECK-NEXT:         GIM_CheckSimplePredicate, GIMT_Encode2(GICXXPred_Simple_IsRule0Enabled),
 // CHECK-NEXT:         // MIs[0] dst
 // CHECK-NEXT:         // No operand predicates
@@ -361,22 +341,17 @@ def MyCombiner: GICombiner<"GenMyCombiner", [
 // CHECK-NEXT:         GIM_CheckCxxInsnPredicate, /*MI*/0, /*FnId*/GIMT_Encode2(GICXXPred_MI_Predicate_GICombiner9),
 // CHECK-NEXT:         GIM_CheckCxxInsnPredicate, /*MI*/0, /*FnId*/GIMT_Encode2(GICXXPred_MI_Predicate_GICombiner10),
 // CHECK-NEXT:         GIM_CheckCxxInsnPredicate, /*MI*/0, /*FnId*/GIMT_Encode2(GICXXPred_MI_Predicate_GICombiner11),
-// CHECK-NEXT:         GIM_CheckIsSafeToFold, /*InsnID*/1,
-// CHECK-NEXT:         GIM_CheckIsSafeToFold, /*InsnID*/2,
-// CHECK-NEXT:         GIM_CheckIsSafeToFold, /*InsnID*/3,
-// CHECK-NEXT:         GIM_CheckIsSafeToFold, /*InsnID*/4,
-// CHECK-NEXT:         GIM_CheckIsSafeToFold, /*InsnID*/5,
+// CHECK-NEXT:         GIM_CheckIsSafeToFold, /*NumInsns*/5,
 // CHECK-NEXT:         GIR_MakeTempReg, /*TempRegID*/0, /*TypeID*/GILLT_s32,
 // CHECK-NEXT:         GIR_BuildConstant, /*TempRegID*/0, /*Val*/GIMT_Encode8(0),
 // CHECK-NEXT:         // Combiner Rule #0: Test0 @ [a[0], b[1], c[1]]
-// CHECK-NEXT:         GIR_BuildMI, /*InsnID*/0, /*Opcode*/GIMT_Encode2(TargetOpcode::COPY),
-// CHECK-NEXT:         GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/0, // dst
+// CHECK-NEXT:         GIR_BuildRootMI, /*Opcode*/GIMT_Encode2(TargetOpcode::COPY),
+// CHECK-NEXT:         GIR_RootToRootCopy, /*OpIdx*/0, // dst
 // CHECK-NEXT:         GIR_AddSimpleTempRegister, /*InsnID*/0, /*TempRegID*/0,
 // CHECK-NEXT:         GIR_CustomAction, GIMT_Encode2(GICXXCustomAction_CombineApplyGICombiner0),
-// CHECK-NEXT:         GIR_EraseFromParent, /*InsnID*/0,
-// CHECK-NEXT:         GIR_Done,
-// CHECK-NEXT:       // Label 5: @509
-// CHECK-NEXT:       GIM_Try, /*On fail goto*//*Label 6*/ GIMT_Encode4(619), // Rule ID 2 //
+// CHECK-NEXT:         GIR_EraseRootFromParent_Done,
+// CHECK-NEXT:       // Label 5: @444
+// CHECK-NEXT:       GIM_Try, /*On fail goto*//*Label 6*/ GIMT_Encode4(539), // Rule ID 2 //
 // CHECK-NEXT:         GIM_CheckSimplePredicate, GIMT_Encode2(GICXXPred_Simple_IsRule0Enabled),
 // CHECK-NEXT:         // MIs[0] dst
 // CHECK-NEXT:         // No operand predicates
@@ -407,23 +382,17 @@ def MyCombiner: GICombiner<"GenMyCombiner", [
 // CHECK-NEXT:         GIM_CheckCxxInsnPredicate, /*MI*/0, /*FnId*/GIMT_Encode2(GICXXPred_MI_Predicate_GICombiner6),
 // CHECK-NEXT:         GIM_CheckCxxInsnPredicate, /*MI*/0, /*FnId*/GIMT_Encode2(GICXXPred_MI_Predicate_GICombiner7),
 // CHECK-NEXT:         GIM_CheckCxxInsnPredicate, /*MI*/0, /*FnId*/GIMT_Encode2(GICXXPred_MI_Predicate_GICombiner8),
-// CHECK-NEXT:         GIM_CheckIsSafeToFold, /*InsnID*/1,
-// CHECK-NEXT:         GIM_CheckIsSafeToFold, /*InsnID*/2,
-// CHECK-NEXT:         GIM_CheckIsSafeToFold, /*InsnID*/3,
-// CHECK-NEXT:         GIM_CheckIsSafeToFold, /*InsnID*/4,
-// CHECK-NEXT:         GIM_CheckIsSafeToFold, /*InsnID*/5,
-// CHECK-NEXT:         GIM_CheckIsSafeToFold, /*InsnID*/6,
+// CHECK-NEXT:         GIM_CheckIsSafeToFold, /*NumInsns*/6,
 // CHECK-NEXT:         GIR_MakeTempReg, /*TempRegID*/0, /*TypeID*/GILLT_s32,
 // CHECK-NEXT:         GIR_BuildConstant, /*TempRegID*/0, /*Val*/GIMT_Encode8(0),
 // CHECK-NEXT:         // Combiner Rule #0: Test0 @ [a[0], b[1], c[0]]
-// CHECK-NEXT:         GIR_BuildMI, /*InsnID*/0, /*Opcode*/GIMT_Encode2(TargetOpcode::COPY),
-// CHECK-NEXT:         GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/0, // dst
+// CHECK-NEXT:         GIR_BuildRootMI, /*Opcode*/GIMT_Encode2(TargetOpcode::COPY),
+// CHECK-NEXT:         GIR_RootToRootCopy, /*OpIdx*/0, // dst
 // CHECK-NEXT:         GIR_AddSimpleTempRegister, /*InsnID*/0, /*TempRegID*/0,
 // CHECK-NEXT:         GIR_CustomAction, GIMT_Encode2(GICXXCustomAction_CombineApplyGICombiner0),
-// CHECK-NEXT:         GIR_EraseFromParent, /*InsnID*/0,
-// CHECK-NEXT:         GIR_Done,
-// CHECK-NEXT:       // Label 6: @619
-// CHECK-NEXT:       GIM_Try, /*On fail goto*//*Label 7*/ GIMT_Encode4(729), // Rule ID 1 //
+// CHECK-NEXT:         GIR_EraseRootFromParent_Done,
+// CHECK-NEXT:       // Label 6: @539
+// CHECK-NEXT:       GIM_Try, /*On fail goto*//*Label 7*/ GIMT_Encode4(634), // Rule ID 1 //
 // CHECK-NEXT:         GIM_CheckSimplePredicate, GIMT_Encode2(GICXXPred_Simple_IsRule0Enabled),
 // CHECK-NEXT:         // MIs[0] dst
 // CHECK-NEXT:         // No operand predicates
@@ -454,23 +423,17 @@ def MyCombiner: GICombiner<"GenMyCombiner", [
 // CHECK-NEXT:         GIM_CheckCxxInsnPredicate, /*MI*/0, /*FnId*/GIMT_Encode2(GICXXPred_MI_Predicate_GICombiner3),
 // CHECK-NEXT:         GIM_CheckCxxInsnPredicate, /*MI*/0, /*FnId*/GIMT_Encode2(GICXXPred_MI_Predicate_GICombiner4),
 // CHECK-NEXT:         GIM_CheckCxxInsnPredicate, /*MI*/0, /*FnId*/GIMT_Encode2(GICXXPred_MI_Predicate_GICombiner5),
-// CHECK-NEXT:         GIM_CheckIsSafeToFold, /*InsnID*/1,
-// CHECK-NEXT:         GIM_CheckIsSafeToFold, /*InsnID*/2,
-// CHECK-NEXT:         GIM_CheckIsSafeToFold, /*InsnID*/3,
-// CHECK-NEXT:         GIM_CheckIsSafeToFold, /*InsnID*/4,
-// CHECK-NEXT:         GIM_CheckIsSafeToFold, /*InsnID*/5,
-// CHECK-NEXT:         GIM_CheckIsSafeToFold, /*InsnID*/6,
+// CHECK-NEXT:         GIM_CheckIsSafeToFold, /*NumInsns*/6,
 // CHECK-NEXT:         GIR_MakeTempReg, /*TempRegID*/0, /*TypeID*/GILLT_s32,
 // CHECK-NEXT:         GIR_BuildConstant, /*TempRegID*/0, /*Val*/GIMT_Encode8(0),
 // CHECK-NEXT:         // Combiner Rule #0: Test0 @ [a[0], b[0], c[1]]
-// CHECK-NEXT:         GIR_BuildMI, /*InsnID*/0, /*Opcode*/GIMT_Encode2(TargetOpcode::COPY),
-// CHECK-NEXT:         GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/0, // dst
+// CHECK-NEXT:         GIR_BuildRootMI, /*Opcode*/GIMT_Encode2(TargetOpcode::COPY),
+// CHECK-NEXT:         GIR_RootToRootCopy, /*OpIdx*/0, // dst
 // CHECK-NEXT:         GIR_AddSimpleTempRegister, /*InsnID*/0, /*TempRegID*/0,
 // CHECK-NEXT:         GIR_CustomAction, GIMT_Encode2(GICXXCustomAction_CombineApplyGICombiner0),
-// CHECK-NEXT:         GIR_EraseFromParent, /*InsnID*/0,
-// CHECK-NEXT:         GIR_Done,
-// CHECK-NEXT:       // Label 7: @729
-// CHECK-NEXT:       GIM_Try, /*On fail goto*//*Label 8*/ GIMT_Encode4(849), // Rule ID 0 //
+// CHECK-NEXT:         GIR_EraseRootFromParent_Done,
+// CHECK-NEXT:       // Label 7: @634
+// CHECK-NEXT:       GIM_Try, /*On fail goto*//*Label 8*/ GIMT_Encode4(737), // Rule ID 0 //
 // CHECK-NEXT:         GIM_CheckSimplePredicate, GIMT_Encode2(GICXXPred_Simple_IsRule0Enabled),
 // CHECK-NEXT:         // MIs[0] dst
 // CHECK-NEXT:         // No operand predicates
@@ -504,26 +467,19 @@ def MyCombiner: GICombiner<"GenMyCombiner", [
 // CHECK-NEXT:         GIM_CheckCxxInsnPredicate, /*MI*/0, /*FnId*/GIMT_Encode2(GICXXPred_MI_Predicate_GICombiner0),
 // CHECK-NEXT:         GIM_CheckCxxInsnPredicate, /*MI*/0, /*FnId*/GIMT_Encode2(GICXXPred_MI_Predicate_GICombiner1),
 // CHECK-NEXT:         GIM_CheckCxxInsnPredicate, /*MI*/0, /*FnId*/GIMT_Encode2(GICXXPred_MI_Predicate_GICombiner2),
-// CHECK-NEXT:         GIM_CheckIsSafeToFold, /*InsnID*/1,
-// CHECK-NEXT:         GIM_CheckIsSafeToFold, /*InsnID*/2,
-// CHECK-NEXT:         GIM_CheckIsSafeToFold, /*InsnID*/3,
-// CHECK-NEXT:         GIM_CheckIsSafeToFold, /*InsnID*/4,
-// CHECK-NEXT:         GIM_CheckIsSafeToFold, /*InsnID*/5,
-// CHECK-NEXT:         GIM_CheckIsSafeToFold, /*InsnID*/6,
-// CHECK-NEXT:         GIM_CheckIsSafeToFold, /*InsnID*/7,
+// CHECK-NEXT:         GIM_CheckIsSafeToFold, /*NumInsns*/7,
 // CHECK-NEXT:         GIR_MakeTempReg, /*TempRegID*/0, /*TypeID*/GILLT_s32,
 // CHECK-NEXT:         GIR_BuildConstant, /*TempRegID*/0, /*Val*/GIMT_Encode8(0),
 // CHECK-NEXT:         // Combiner Rule #0: Test0 @ [a[0], b[0], c[0]]
-// CHECK-NEXT:         GIR_BuildMI, /*InsnID*/0, /*Opcode*/GIMT_Encode2(TargetOpcode::COPY),
-// CHECK-NEXT:         GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/0, // dst
+// CHECK-NEXT:         GIR_BuildRootMI, /*Opcode*/GIMT_Encode2(TargetOpcode::COPY),
+// CHECK-NEXT:         GIR_RootToRootCopy, /*OpIdx*/0, // dst
 // CHECK-NEXT:         GIR_AddSimpleTempRegister, /*InsnID*/0, /*TempRegID*/0,
 // CHECK-NEXT:         GIR_CustomAction, GIMT_Encode2(GICXXCustomAction_CombineApplyGICombiner0),
-// CHECK-NEXT:         GIR_EraseFromParent, /*InsnID*/0,
-// CHECK-NEXT:         GIR_Done,
-// CHECK-NEXT:       // Label 8: @849
+// CHECK-NEXT:         GIR_EraseRootFromParent_Done,
+// CHECK-NEXT:       // Label 8: @737
 // CHECK-NEXT:       GIM_Reject,
-// CHECK-NEXT:     // Label 0: @850
+// CHECK-NEXT:     // Label 0: @738
 // CHECK-NEXT:     GIM_Reject,
-// CHECK-NEXT:     }; // Size: 851 bytes
+// CHECK-NEXT:     }; // Size: 739 bytes
 // CHECK-NEXT:   return MatchTable0;
 // CHECK-NEXT: }
diff --git a/llvm/test/TableGen/GlobalISelCombinerEmitter/match-table-temp-defs.td b/llvm/test/TableGen/GlobalISelCombinerEmitter/match-table-temp-defs.td
index 4e473355e14c..9a7716e54b27 100644
--- a/llvm/test/TableGen/GlobalISelCombinerEmitter/match-table-temp-defs.td
+++ b/llvm/test/TableGen/GlobalISelCombinerEmitter/match-table-temp-defs.td
@@ -36,21 +36,21 @@ def MyCombiner: GICombiner<"GenMyCombiner", [
 ]>;
 
 // CHECK:       // Combiner Rule #0: Test0
-// CHECK-NEXT:  GIR_BuildMI, /*InsnID*/0, /*Opcode*/GIMT_Encode2(TargetOpcode::G_UDIVREM),
+// CHECK-NEXT:  GIR_BuildRootMI, /*Opcode*/GIMT_Encode2(TargetOpcode::G_UDIVREM),
 // CHECK-NEXT:  GIR_AddTempRegister, /*InsnID*/0, /*TempRegID*/0, /*TempRegFlags*/GIMT_Encode2(RegState::Define),
-// CHECK-NEXT:  GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/0, // dst
-// CHECK-NEXT:  GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/1, // lhs
-// CHECK-NEXT:  GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/2, // rhs
+// CHECK-NEXT:  GIR_RootToRootCopy, /*OpIdx*/0, // dst
+// CHECK-NEXT:  GIR_RootToRootCopy, /*OpIdx*/1, // lhs
+// CHECK-NEXT:  GIR_RootToRootCopy, /*OpIdx*/2, // rhs
 
 // CHECK:       // Combiner Rule #1: Test1
-// CHECK-NEXT:  GIR_BuildMI, /*InsnID*/0, /*Opcode*/GIMT_Encode2(TargetOpcode::G_UDIVREM),
-// CHECK-NEXT:  GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/0, // dst
+// CHECK-NEXT:  GIR_BuildRootMI, /*Opcode*/GIMT_Encode2(TargetOpcode::G_UDIVREM),
+// CHECK-NEXT:  GIR_RootToRootCopy, /*OpIdx*/0, // dst
 // CHECK-NEXT:  GIR_AddTempRegister, /*InsnID*/0, /*TempRegID*/0, /*TempRegFlags*/GIMT_Encode2(RegState::Define),
-// CHECK-NEXT:  GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/1, // lhs
-// CHECK-NEXT:  GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/2, // rhs
+// CHECK-NEXT:  GIR_RootToRootCopy, /*OpIdx*/1, // lhs
+// CHECK-NEXT:  GIR_RootToRootCopy, /*OpIdx*/2, // rhs
 
 // CHECK:       // Combiner Rule #2: Test2
-// CHECK-NEXT:  GIR_BuildMI, /*InsnID*/0, /*Opcode*/GIMT_Encode2(TargetOpcode::G_ADD),
+// CHECK-NEXT:  GIR_BuildRootMI, /*Opcode*/GIMT_Encode2(TargetOpcode::G_ADD),
 // CHECK-NEXT:  GIR_AddTempRegister, /*InsnID*/0, /*TempRegID*/0, /*TempRegFlags*/GIMT_Encode2(RegState::Define),
 // CHECK-NEXT:  GIR_AddSimpleTempRegister, /*InsnID*/0, /*TempRegID*/1,
-// CHECK-NEXT:  GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/1, // lhs
+// CHECK-NEXT:  GIR_RootToRootCopy, /*OpIdx*/1, // lhs
diff --git a/llvm/test/TableGen/GlobalISelCombinerEmitter/match-table-typeof.td b/llvm/test/TableGen/GlobalISelCombinerEmitter/match-table-typeof.td
index ca653674d9c2..7fe63b1298ae 100644
--- a/llvm/test/TableGen/GlobalISelCombinerEmitter/match-table-typeof.td
+++ b/llvm/test/TableGen/GlobalISelCombinerEmitter/match-table-typeof.td
@@ -16,7 +16,7 @@ def Test0 : GICombineRule<
 
 // CHECK:      const uint8_t *GenMyCombiner::getMatchTable() const {
 // CHECK-NEXT:   constexpr static uint8_t MatchTable0[] = {
-// CHECK-NEXT:     GIM_Try, /*On fail goto*//*Label 0*/ GIMT_Encode4(77), // Rule ID 0 //
+// CHECK-NEXT:     GIM_Try, /*On fail goto*//*Label 0*/ GIMT_Encode4(74), // Rule ID 0 //
 // CHECK-NEXT:       GIM_CheckSimplePredicate, GIMT_Encode2(GICXXPred_Simple_IsRule0Enabled),
 // CHECK-NEXT:       GIM_CheckOpcode, /*MI*/0, GIMT_Encode2(TargetOpcode::G_MUL),
 // CHECK-NEXT:       // MIs[0] dst
@@ -29,18 +29,17 @@ def Test0 : GICombineRule<
 // CHECK-NEXT:       GIR_BuildConstant, /*TempRegID*/1, /*Val*/GIMT_Encode8(0),
 // CHECK-NEXT:       GIR_MakeTempReg, /*TempRegID*/0, /*TypeID*/uint8_t(-1),
 // CHECK-NEXT:       // Combiner Rule #0: Test0
-// CHECK-NEXT:       GIR_BuildMI, /*InsnID*/0, /*Opcode*/GIMT_Encode2(TargetOpcode::G_CONSTANT),
+// CHECK-NEXT:       GIR_BuildRootMI, /*Opcode*/GIMT_Encode2(TargetOpcode::G_CONSTANT),
 // CHECK-NEXT:       GIR_AddTempRegister, /*InsnID*/0, /*TempRegID*/0, /*TempRegFlags*/GIMT_Encode2(RegState::Define),
 // CHECK-NEXT:       GIR_AddCImm, /*InsnID*/0, /*Type*/uint8_t(-2), /*Imm*/GIMT_Encode8(42),
 // CHECK-NEXT:       GIR_BuildMI, /*InsnID*/1, /*Opcode*/GIMT_Encode2(TargetOpcode::G_SUB),
 // CHECK-NEXT:       GIR_Copy, /*NewInsnID*/1, /*OldInsnID*/0, /*OpIdx*/0, // dst
 // CHECK-NEXT:       GIR_AddSimpleTempRegister, /*InsnID*/1, /*TempRegID*/1,
 // CHECK-NEXT:       GIR_AddSimpleTempRegister, /*InsnID*/1, /*TempRegID*/0,
-// CHECK-NEXT:       GIR_EraseFromParent, /*InsnID*/0,
-// CHECK-NEXT:       GIR_Done,
-// CHECK-NEXT:     // Label 0: @77
+// CHECK-NEXT:       GIR_EraseRootFromParent_Done,
+// CHECK-NEXT:     // Label 0: @74
 // CHECK-NEXT:     GIM_Reject,
-// CHECK-NEXT:     }; // Size: 78 bytes
+// CHECK-NEXT:     }; // Size: 75 bytes
 // CHECK-NEXT:   return MatchTable0;
 // CHECK-NEXT: }
 
diff --git a/llvm/test/TableGen/GlobalISelCombinerEmitter/match-table.td b/llvm/test/TableGen/GlobalISelCombinerEmitter/match-table.td
index 02085c1fd266..1052e31b2d05 100644
--- a/llvm/test/TableGen/GlobalISelCombinerEmitter/match-table.td
+++ b/llvm/test/TableGen/GlobalISelCombinerEmitter/match-table.td
@@ -154,7 +154,7 @@ def MyCombiner: GICombiner<"GenMyCombiner", [
 // CHECK-NEXT:       // MIs[1] c
 // CHECK-NEXT:       // No operand predicates
 // CHECK-NEXT:       GIM_CheckCxxInsnPredicate, /*MI*/0, /*FnId*/GIMT_Encode2(GICXXPred_MI_Predicate_GICombiner0),
-// CHECK-NEXT:       GIM_CheckIsSafeToFold, /*InsnID*/1,
+// CHECK-NEXT:       GIM_CheckIsSafeToFold, /*NumInsns*/1,
 // CHECK-NEXT:       // Combiner Rule #3: InstTest1
 // CHECK-NEXT:       GIR_CustomAction, GIMT_Encode2(GICXXCustomAction_CombineApplyGICombiner0),
 // CHECK-NEXT:       GIR_Done,
@@ -173,7 +173,7 @@ def MyCombiner: GICombiner<"GenMyCombiner", [
 // CHECK-NEXT:     // Label 1: @[[L504]]
 // CHECK-NEXT:     GIM_Try, /*On fail goto*//*Label 9*/ GIMT_Encode4([[L556:[0-9]+]]), // Rule ID 6 //
 // CHECK-NEXT:       GIM_CheckSimplePredicate, GIMT_Encode2(GICXXPred_Simple_IsRule5Enabled),
-// CHECK-NEXT:       GIM_CheckType, /*MI*/0, /*Op*/2, /*Type*/GILLT_s32,
+// CHECK-NEXT:       GIM_RootCheckType, /*Op*/2, /*Type*/GILLT_s32,
 // CHECK-NEXT:       // MIs[0] dst
 // CHECK-NEXT:       // No operand predicates
 // CHECK-NEXT:       // MIs[0] x
@@ -182,13 +182,12 @@ def MyCombiner: GICombiner<"GenMyCombiner", [
 // CHECK-NEXT:       // MIs[1] z
 // CHECK-NEXT:       GIM_CheckLiteralInt, /*MI*/1, /*Op*/1, GIMT_Encode8(-42),
 // CHECK-NEXT:       GIM_CheckConstantInt8, /*MI*/0, /*Op*/2, 43,
-// CHECK-NEXT:       GIM_CheckIsSafeToFold, /*InsnID*/1,
+// CHECK-NEXT:       GIM_CheckIsSafeToFold, /*NumInsns*/1,
 // CHECK-NEXT:       // Combiner Rule #5: InOutInstTest1
-// CHECK-NEXT:       GIR_BuildMI, /*InsnID*/0, /*Opcode*/GIMT_Encode2(TargetOpcode::G_TRUNC),
-// CHECK-NEXT:       GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/0, // dst
+// CHECK-NEXT:       GIR_BuildRootMI, /*Opcode*/GIMT_Encode2(TargetOpcode::G_TRUNC),
+// CHECK-NEXT:       GIR_RootToRootCopy, /*OpIdx*/0, // dst
 // CHECK-NEXT:       GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/1, /*OpIdx*/1, // z
-// CHECK-NEXT:       GIR_EraseFromParent, /*InsnID*/0,
-// CHECK-NEXT:       GIR_Done,
+// CHECK-NEXT:       GIR_EraseRootFromParent_Done,
 // CHECK-NEXT:     // Label 9: @[[L556]]
 // CHECK-NEXT:     GIM_Reject,
 // CHECK-NEXT:     // Label 2: @[[L557]]
@@ -201,15 +200,14 @@ def MyCombiner: GICombiner<"GenMyCombiner", [
 // CHECK-NEXT:       // No operand predicates
 // CHECK-NEXT:       // MIs[0] ptr
 // CHECK-NEXT:       // No operand predicates
-// CHECK-NEXT:       GIM_CheckIsSafeToFold, /*InsnID*/1,
+// CHECK-NEXT:       GIM_CheckIsSafeToFold, /*NumInsns*/1,
 // CHECK-NEXT:       // Combiner Rule #4: InOutInstTest0
-// CHECK-NEXT:       GIR_BuildMI, /*InsnID*/0, /*Opcode*/GIMT_Encode2(TargetOpcode::G_STORE),
+// CHECK-NEXT:       GIR_BuildRootMI, /*Opcode*/GIMT_Encode2(TargetOpcode::G_STORE),
 // CHECK-NEXT:       GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/1, /*OpIdx*/1, // ext
-// CHECK-NEXT:       GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/1, // ptr
+// CHECK-NEXT:       GIR_RootToRootCopy, /*OpIdx*/1, // ptr
 // CHECK-NEXT:       GIR_MergeMemOperands, /*InsnID*/0, /*NumInsns*/2, /*MergeInsnID's*/0, 1,
 // CHECK-NEXT:       GIR_CustomAction, GIMT_Encode2(GICXXCustomAction_CombineApplyGICombiner2),
-// CHECK-NEXT:       GIR_EraseFromParent, /*InsnID*/0,
-// CHECK-NEXT:       GIR_Done,
+// CHECK-NEXT:       GIR_EraseRootFromParent_Done,
 // CHECK-NEXT:     // Label 10: @[[L598]]
 // CHECK-NEXT:     GIM_Reject,
 // CHECK-NEXT:     // Label 3: @[[L599]]
@@ -245,11 +243,10 @@ def MyCombiner: GICombiner<"GenMyCombiner", [
 // CHECK-NEXT:       GIR_MakeTempReg, /*TempRegID*/0, /*TypeID*/GILLT_s32,
 // CHECK-NEXT:       GIR_BuildConstant, /*TempRegID*/0, /*Val*/GIMT_Encode8(0),
 // CHECK-NEXT:       // Combiner Rule #6: PatFragTest0 @ [__PatFragTest0_match_1[0]]
-// CHECK-NEXT:       GIR_BuildMI, /*InsnID*/0, /*Opcode*/GIMT_Encode2(TargetOpcode::COPY),
-// CHECK-NEXT:       GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/0, // dst
+// CHECK-NEXT:       GIR_BuildRootMI, /*Opcode*/GIMT_Encode2(TargetOpcode::COPY),
+// CHECK-NEXT:       GIR_RootToRootCopy, /*OpIdx*/0, // dst
 // CHECK-NEXT:       GIR_AddSimpleTempRegister, /*InsnID*/0, /*TempRegID*/0,
-// CHECK-NEXT:       GIR_EraseFromParent, /*InsnID*/0,
-// CHECK-NEXT:       GIR_Done,
+// CHECK-NEXT:       GIR_EraseRootFromParent_Done,
 // CHECK-NEXT:     // Label 14: @[[L676]]
 // CHECK-NEXT:     GIM_Reject,
 // CHECK-NEXT:     // Label 6: @[[L677]]
diff --git a/llvm/test/TableGen/GlobalISelEmitter-atomic_store.td b/llvm/test/TableGen/GlobalISelEmitter-atomic_store.td
index 081c9e863459..da2dfe800428 100644
--- a/llvm/test/TableGen/GlobalISelEmitter-atomic_store.td
+++ b/llvm/test/TableGen/GlobalISelEmitter-atomic_store.td
@@ -9,7 +9,7 @@ def ST_ATOM_B32 : I<(outs), (ins GPR32Op:$val, GPR32Op:$ptr), []>;
 // GISEL-NEXT: GIM_CheckMemorySizeEqualTo, /*MI*/0, /*MMO*/0, /*Size*/GIMT_Encode4(1),
 // GISEL-NEXT: GIM_CheckAtomicOrderingOrStrongerThan, /*MI*/0, /*Order*/(uint8_t)AtomicOrdering::Unordered,
 // GISEL-NEXT: // MIs[0] val
-// GISEL-NEXT: GIM_CheckType, /*MI*/0, /*Op*/0, /*Type*/GILLT_s32,
+// GISEL-NEXT: GIM_RootCheckType, /*Op*/0, /*Type*/GILLT_s32,
 // GISEL-NEXT: // MIs[0] ptr
 // GISEL-NEXT: GIM_CheckPointerToAny, /*MI*/0, /*Op*/1, /*SizeInBits*/0,
 // GISEL-NEXT: // (atomic_store i32:{ *:[i32] }:$val, iPTR:{ *:[iPTR] }:$ptr)<<P:Predicate_atomic_store_8>>  =>  (ST_ATOM_B32 GPR32Op:{ *:[i32] }:$val, GPR32Op:{ *:[i32] }:$ptr)
diff --git a/llvm/test/TableGen/GlobalISelEmitter-frameindex.td b/llvm/test/TableGen/GlobalISelEmitter-frameindex.td
new file mode 100644
index 000000000000..232691465bb3
--- /dev/null
+++ b/llvm/test/TableGen/GlobalISelEmitter-frameindex.td
@@ -0,0 +1,29 @@
+// RUN: llvm-tblgen -gen-global-isel -optimize-match-table=false -I %p/../../include -I %p/Common %s -o - < %s | FileCheck %s
+
+include "llvm/Target/Target.td"
+include "GlobalISelEmitterCommon.td"
+
+def ADD : I<(outs GPR32:$dst), (ins GPR32:$src1, GPR32:$src2), []>;
+
+//===- Test a simple pattern with frame index operands. ----------------------===//
+//
+// CHECK:       GIM_Try, /*On fail goto*//*Label [[LABEL_NUM:[0-9]+]]*/ GIMT_Encode4([[LABEL:[0-9]+]]),
+// CHECK-NEXT:    GIM_CheckNumOperands, /*MI*/0, /*Expected*/2,
+// CHECK-NEXT:    GIM_CheckOpcode, /*MI*/0, GIMT_Encode2(TargetOpcode::G_FRAME_INDEX),
+// CHECK-NEXT:    // MIs[0] DstI[dst]
+// CHECK-NEXT:    GIM_RootCheckType, /*Op*/0, /*Type*/GILLT_p0s32,
+// CHECK-NEXT:    GIM_RootCheckRegBankForClass, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
+// CHECK-NEXT:    // MIs[0] fi
+// CHECK-NEXT:    // No operand predicates
+// CHECK-NEXT:    // (frameindex:{ *:[i32] }):$fi  =>  (ADD:{ *:[i32] } (tframeindex:{ *:[i32] }):$fi, 0:{ *:[i32] })
+// CHECK-NEXT:    GIR_BuildRootMI, /*Opcode*/GIMT_Encode2(MyTarget::ADD),
+// CHECK-NEXT:    GIR_RootToRootCopy, /*OpIdx*/0, // DstI[dst]
+// CHECK-NEXT:    GIR_RootToRootCopy, /*OpIdx*/1, // fi
+// CHECK-NEXT:    GIR_AddImm8, /*InsnID*/0, /*Imm*/0,
+// CHECK-NEXT:    GIR_RootConstrainSelectedInstOperands,
+// CHECK-NEXT:    // GIR_Coverage, 0,
+// CHECK-NEXT:    GIR_EraseRootFromParent_Done,
+// CHECK-NEXT:  // Label [[LABEL_NUM]]: @[[LABEL]]
+// CHECK-NEXT:  GIM_Reject,
+
+def : Pat<(p0 frameindex:$fi), (ADD tframeindex:$fi, 0)>;
diff --git a/llvm/test/TableGen/GlobalISelEmitter-immAllZeroOne.td b/llvm/test/TableGen/GlobalISelEmitter-immAllZeroOne.td
index eae29308aa1b..0125aa5c30fa 100644
--- a/llvm/test/TableGen/GlobalISelEmitter-immAllZeroOne.td
+++ b/llvm/test/TableGen/GlobalISelEmitter-immAllZeroOne.td
@@ -11,12 +11,12 @@ include "GlobalISelEmitterCommon.td"
 
 
 // GISEL-OPT: GIM_Try,
-// GISEL-OPT: GIM_CheckType, /*MI*/0, /*Op*/2, /*Type*/GILLT_v4s16,
+// GISEL-OPT: GIM_RootCheckType, /*Op*/2, /*Type*/GILLT_v4s16,
 // GISEL-OPT: GIM_CheckOpcodeIsEither, /*MI*/1, GIMT_Encode2(TargetOpcode::G_BUILD_VECTOR), GIMT_Encode2(TargetOpcode::G_BUILD_VECTOR_TRUNC),
 // GISEL-OPT: GIM_CheckIsBuildVectorAllZeros, /*MI*/1,
 
 // GISEL-OPT: GIM_Try,
-// GISEL-OPT: GIM_CheckType, /*MI*/0, /*Op*/2, /*Type*/GILLT_v4s16,
+// GISEL-OPT: GIM_RootCheckType, /*Op*/2, /*Type*/GILLT_v4s16,
 // GISEL-OPT: GIM_CheckOpcodeIsEither, /*MI*/1, GIMT_Encode2(TargetOpcode::G_BUILD_VECTOR), GIMT_Encode2(TargetOpcode::G_BUILD_VECTOR_TRUNC),
 // GISEL-OPT: GIM_CheckIsBuildVectorAllOnes, /*MI*/1,
 
@@ -24,13 +24,13 @@ include "GlobalISelEmitterCommon.td"
 
 // GISEL-NOOPT: GIM_CheckOpcode, /*MI*/0, GIMT_Encode2(TargetOpcode::G_LSHR),
 // GISEL-NOOPT: // MIs[0] Operand 2
-// GISEL-NOOPT-NEXT: GIM_CheckType, /*MI*/0, /*Op*/2, /*Type*/GILLT_v4s16,
+// GISEL-NOOPT-NEXT: GIM_RootCheckType, /*Op*/2, /*Type*/GILLT_v4s16,
 // GISEL-NOOPT-NEXT: GIM_RecordInsn, /*DefineMI*/1, /*MI*/0, /*OpIdx*/2, // MIs[1]
 // GISEL-NOOPT-NEXT: GIM_CheckOpcodeIsEither, /*MI*/1, GIMT_Encode2(TargetOpcode::G_BUILD_VECTOR), GIMT_Encode2(TargetOpcode::G_BUILD_VECTOR_TRUNC),
 // GISEL-NOOPT-NEXT: GIM_CheckIsBuildVectorAllOnes, /*MI*/1,
 // GISEL-NOOPT-NEXT: // MIs[1] Operand 0
 // GISEL-NOOPT-NEXT: GIM_CheckType, /*MI*/1, /*Op*/0, /*Type*/GILLT_v4s16,
-// GISEL-NOOPT-NEXT: GIM_CheckIsSafeToFold, /*InsnID*/1,
+// GISEL-NOOPT-NEXT: GIM_CheckIsSafeToFold, /*NumInsns*/1,
 // GISEL-NOOPT-NEXT: // (srl:{ *:[v4i32] } v4i32:{ *:[v4i32] }:$src0, immAllOnesV:{ *:[v4i16] })  =>  (VFOOONES:{ *:[v4i32] } v4i32:{ *:[v4i32] }:$src0)
 def VFOOONES : I<(outs VecReg128:$dst), (ins VecReg128:$src0),
   [(set v4i32:$dst, (srl v4i32:$src0, (v4i16 immAllOnesV)))]
@@ -39,13 +39,13 @@ def VFOOONES : I<(outs VecReg128:$dst), (ins VecReg128:$src0),
 
 // GISEL-NOOPT: GIM_CheckOpcode, /*MI*/0, GIMT_Encode2(TargetOpcode::G_SHL),
 // GISEL-NOOPT: // MIs[0] Operand 2
-// GISEL-NOOPT-NEXT: GIM_CheckType, /*MI*/0, /*Op*/2, /*Type*/GILLT_v4s16,
+// GISEL-NOOPT-NEXT: GIM_RootCheckType, /*Op*/2, /*Type*/GILLT_v4s16,
 // GISEL-NOOPT-NEXT: GIM_RecordInsn, /*DefineMI*/1, /*MI*/0, /*OpIdx*/2, // MIs[1]
 // GISEL-NOOPT-NEXT: GIM_CheckOpcodeIsEither, /*MI*/1, GIMT_Encode2(TargetOpcode::G_BUILD_VECTOR), GIMT_Encode2(TargetOpcode::G_BUILD_VECTOR_TRUNC),
 // GISEL-NOOPT-NEXT: GIM_CheckIsBuildVectorAllZeros, /*MI*/1,
 // GISEL-NOOPT-NEXT: // MIs[1] Operand 0
 // GISEL-NOOPT-NEXT: GIM_CheckType, /*MI*/1, /*Op*/0, /*Type*/GILLT_v4s16,
-// GISEL-NOOPT-NEXT: GIM_CheckIsSafeToFold, /*InsnID*/1,
+// GISEL-NOOPT-NEXT: GIM_CheckIsSafeToFold, /*NumInsns*/1,
 // GISEL-NOOPT-NEXT: // (shl:{ *:[v4i32] } v4i32:{ *:[v4i32] }:$src0, immAllZerosV:{ *:[v4i16] })  =>  (VFOOZERO:{ *:[v4i32] } v4i32:{ *:[v4i32] }:$src0)
 def VFOOZERO : I<(outs VecReg128:$dst), (ins VecReg128:$src0),
   [(set v4i32:$dst, (shl v4i32:$src0, (v4i16 immAllZerosV)))]
diff --git a/llvm/test/TableGen/GlobalISelEmitter-immarg-literal-pattern.td b/llvm/test/TableGen/GlobalISelEmitter-immarg-literal-pattern.td
index bfbeee466b1c..6b4012eb736c 100644
--- a/llvm/test/TableGen/GlobalISelEmitter-immarg-literal-pattern.td
+++ b/llvm/test/TableGen/GlobalISelEmitter-immarg-literal-pattern.td
@@ -43,7 +43,7 @@ def : Pat<
 // Check a non-intrinsic instruction with an immediate parameter.
 
 // GISEL: GIM_CheckOpcode, /*MI*/0, GIMT_Encode2(MyTarget::G_TGT_CAT),
-// GISEL: GIM_CheckType, /*MI*/0, /*Op*/1, /*Type*/GILLT_s32,
+// GISEL: GIM_RootCheckType, /*Op*/1, /*Type*/GILLT_s32,
 // GISEL-NEXT: // MIs[0] Operand 2
 // GISEL-NEXT: GIM_CheckLiteralInt, /*MI*/0, /*Op*/2, GIMT_Encode8(0),
 def : Pat<
@@ -52,7 +52,7 @@ def : Pat<
 >;
 
 // GISEL: GIM_CheckOpcode, /*MI*/0, GIMT_Encode2(MyTarget::G_TGT_CAT),
-// GISEL: GIM_CheckType, /*MI*/0, /*Op*/1, /*Type*/GILLT_s32,
+// GISEL: GIM_RootCheckType, /*Op*/1, /*Type*/GILLT_s32,
 // GISEL-NEXT: // MIs[0] Operand 2
 // GISEL-NEXT: GIM_CheckLiteralInt, /*MI*/0, /*Op*/2, GIMT_Encode8(93),
 def : Pat<
diff --git a/llvm/test/TableGen/GlobalISelEmitter-input-discard.td b/llvm/test/TableGen/GlobalISelEmitter-input-discard.td
index 8d3c6cb180ae..202ff4a5758d 100644
--- a/llvm/test/TableGen/GlobalISelEmitter-input-discard.td
+++ b/llvm/test/TableGen/GlobalISelEmitter-input-discard.td
@@ -11,21 +11,22 @@ def FOO : I<(outs GPR32:$dst), (ins GPR32Op:$src0, GPR32Op:$src1), []>;
 // GISEL: GIM_CheckOpcode, /*MI*/0, GIMT_Encode2(TargetOpcode::G_INTRINSIC_W_SIDE_EFFECTS),
 // GISEL-NEXT: GIM_CheckNumOperands, /*MI*/0, /*Expected*/4,
 // GISEL-NEXT: GIM_CheckIntrinsicID, /*MI*/0, /*Op*/1, GIMT_Encode2(Intrinsic::tgt_foo),
-// GISEL-NEXT: GIM_CheckType, /*MI*/0, /*Op*/0, /*Type*/GILLT_s32,
-// GISEL-NEXT: GIM_CheckType, /*MI*/0, /*Op*/2, /*Type*/GILLT_s32,
-// GISEL-NEXT: GIM_CheckType, /*MI*/0, /*Op*/3, /*Type*/GILLT_s32,
-// GISEL-NEXT: GIM_CheckRegBankForClass, /*MI*/0, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
+// GISEL-NEXT: GIM_RootCheckType, /*Op*/0, /*Type*/GILLT_s32,
+// GISEL-NEXT: GIM_RootCheckType, /*Op*/2, /*Type*/GILLT_s32,
+// GISEL-NEXT: GIM_RootCheckType, /*Op*/3, /*Type*/GILLT_s32,
+// GISEL-NEXT: GIM_RootCheckRegBankForClass, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
 // GISEL-NEXT: // (intrinsic_w_chain:{ *:[i32] } {{[0-9]+}}:{ *:[iPTR] }, srcvalue:{ *:[i32] }, i32:{ *:[i32] }:$src1)  =>  (FOO:{ *:[i32] } (IMPLICIT_DEF:{ *:[i32] }), GPR32:{ *:[i32] }:$src1)
 // GISEL-NEXT: GIR_MakeTempReg, /*TempRegID*/0, /*TypeID*/GILLT_s32,
 // GISEL-NEXT: GIR_BuildMI, /*InsnID*/1, /*Opcode*/GIMT_Encode2(TargetOpcode::IMPLICIT_DEF),
 // GISEL-NEXT: GIR_AddTempRegister, /*InsnID*/1, /*TempRegID*/0, /*TempRegFlags*/GIMT_Encode2(RegState::Define),
 // GISEL-NEXT: GIR_ConstrainSelectedInstOperands, /*InsnID*/1,
-// GISEL-NEXT: GIR_BuildMI, /*InsnID*/0, /*Opcode*/GIMT_Encode2(MyTarget::FOO),
-// GISEL-NEXT: GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/0, // DstI[dst]
+// GISEL-NEXT: GIR_BuildRootMI, /*Opcode*/GIMT_Encode2(MyTarget::FOO),
+// GISEL-NEXT: GIR_RootToRootCopy, /*OpIdx*/0, // DstI[dst]
 // GISEL-NEXT: GIR_AddSimpleTempRegister, /*InsnID*/0, /*TempRegID*/0,
-// GISEL-NEXT: GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/3, // src1
-// GISEL-NEXT: GIR_ConstrainSelectedInstOperands, /*InsnID*/0,
-// GISEL-NEXT: GIR_EraseFromParent, /*InsnID*/0,
+// GISEL-NEXT: GIR_RootToRootCopy, /*OpIdx*/3, // src1
+// GISEL-NEXT: GIR_RootConstrainSelectedInstOperands,
+// GISEL-NEXT: // GIR_Coverage
+// GISEL-NEXT: GIR_EraseRootFromParent_Done,
 def : Pat <
   (int_tgt_foo (i32 srcvalue), i32:$src1),
   (FOO (IMPLICIT_DEF), GPR32:$src1)
diff --git a/llvm/test/TableGen/GlobalISelEmitter-multiple-output-discard.td b/llvm/test/TableGen/GlobalISelEmitter-multiple-output-discard.td
index 70991ea3b69c..2d968bebbc65 100644
--- a/llvm/test/TableGen/GlobalISelEmitter-multiple-output-discard.td
+++ b/llvm/test/TableGen/GlobalISelEmitter-multiple-output-discard.td
@@ -23,20 +23,21 @@ def : Pat<(two_out GPR32:$val), (THREE_OUTS GPR32:$val)>;
 
 // CHECK:      GIM_CheckOpcode, /*MI*/0, GIMT_Encode2(MyTarget::G_TWO_OUT),
 // CHECK-NEXT: // MIs[0] DstI[out1]
-// CHECK-NEXT: GIM_CheckType, /*MI*/0, /*Op*/0, /*Type*/GILLT_s32,
-// CHECK-NEXT: GIM_CheckRegBankForClass, /*MI*/0, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
+// CHECK-NEXT: GIM_RootCheckType, /*Op*/0, /*Type*/GILLT_s32,
+// CHECK-NEXT: GIM_RootCheckRegBankForClass, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
 // CHECK-NEXT: // MIs[0] DstI[out2]
-// CHECK-NEXT: GIM_CheckType, /*MI*/0, /*Op*/1, /*Type*/GILLT_s32,
-// CHECK-NEXT: GIM_CheckRegBankForClass, /*MI*/0, /*Op*/1, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
+// CHECK-NEXT: GIM_RootCheckType, /*Op*/1, /*Type*/GILLT_s32,
+// CHECK-NEXT: GIM_RootCheckRegBankForClass, /*Op*/1, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
 // CHECK-NEXT: // MIs[0] val
-// CHECK-NEXT: GIM_CheckType, /*MI*/0, /*Op*/2, /*Type*/GILLT_s32,
-// CHECK-NEXT: GIM_CheckRegBankForClass, /*MI*/0, /*Op*/2, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
+// CHECK-NEXT: GIM_RootCheckType, /*Op*/2, /*Type*/GILLT_s32,
+// CHECK-NEXT: GIM_RootCheckRegBankForClass, /*Op*/2, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
 // CHECK-NEXT: // (two_out:{ *:[i32] }:{ *:[i32] } GPR32:{ *:[i32] }:$val)  =>  (THREE_OUTS:{ *:[i32] }:{ *:[i32] }:{ *:[i32] } GPR32:{ *:[i32] }:$val)
 // CHECK-NEXT: GIR_MakeTempReg, /*TempRegID*/0, /*TypeID*/GILLT_s32,
-// CHECK-NEXT: GIR_BuildMI, /*InsnID*/0, /*Opcode*/GIMT_Encode2(MyTarget::THREE_OUTS),
-// CHECK-NEXT: GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/0, // DstI[out1]
-// CHECK-NEXT: GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/1, // DstI[out2]
+// CHECK-NEXT: GIR_BuildRootMI, /*Opcode*/GIMT_Encode2(MyTarget::THREE_OUTS),
+// CHECK-NEXT: GIR_RootToRootCopy, /*OpIdx*/0, // DstI[out1]
+// CHECK-NEXT: GIR_RootToRootCopy, /*OpIdx*/1, // DstI[out2]
 // CHECK-NEXT: GIR_AddTempRegister, /*InsnID*/0, /*TempRegID*/0, /*TempRegFlags*/GIMT_Encode2(RegState::Define|RegState::Dead),
-// CHECK-NEXT: GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/2, // val
-// CHECK-NEXT: GIR_ConstrainSelectedInstOperands, /*InsnID*/0,
-// CHECK-NEXT: GIR_EraseFromParent, /*InsnID*/0,
+// CHECK-NEXT: GIR_RootToRootCopy, /*OpIdx*/2, // val
+// CHECK-NEXT: GIR_RootConstrainSelectedInstOperands,
+// CHECK-NEXT: // GIR_Coverage
+// CHECK-NEXT: GIR_EraseRootFromParent_Done,
diff --git a/llvm/test/TableGen/GlobalISelEmitter-multiple-output.td b/llvm/test/TableGen/GlobalISelEmitter-multiple-output.td
index 94c9f60eabd3..dea3b5496071 100644
--- a/llvm/test/TableGen/GlobalISelEmitter-multiple-output.td
+++ b/llvm/test/TableGen/GlobalISelEmitter-multiple-output.td
@@ -32,20 +32,20 @@ def : Pat<(loadpost (p0 GPR32:$addr), (i32 GPR32:$off)),
 
 // CHECK:      GIM_CheckOpcode, /*MI*/0, GIMT_Encode2(MyTarget::G_POST_LOAD),
 // CHECK-NEXT: // MIs[0] DstI[val]
-// CHECK-NEXT: GIM_CheckType, /*MI*/0, /*Op*/0, /*Type*/GILLT_s32,
-// CHECK-NEXT: GIM_CheckRegBankForClass, /*MI*/0, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
+// CHECK-NEXT: GIM_RootCheckType, /*Op*/0, /*Type*/GILLT_s32,
+// CHECK-NEXT: GIM_RootCheckRegBankForClass, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
 // CHECK-NEXT: // MIs[0] DstI[ptr_out]
-// CHECK-NEXT: GIM_CheckType, /*MI*/0, /*Op*/1, /*Type*/GILLT_p0s32,
-// CHECK-NEXT: GIM_CheckRegBankForClass, /*MI*/0, /*Op*/1, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
+// CHECK-NEXT: GIM_RootCheckType, /*Op*/1, /*Type*/GILLT_p0s32,
+// CHECK-NEXT: GIM_RootCheckRegBankForClass, /*Op*/1, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
 // CHECK-NEXT: // MIs[0] addr
-// CHECK-NEXT: GIM_CheckType, /*MI*/0, /*Op*/2, /*Type*/GILLT_p0s32,
-// CHECK-NEXT: GIM_CheckRegBankForClass, /*MI*/0, /*Op*/2, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
+// CHECK-NEXT: GIM_RootCheckType, /*Op*/2, /*Type*/GILLT_p0s32,
+// CHECK-NEXT: GIM_RootCheckRegBankForClass, /*Op*/2, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
 // CHECK-NEXT: // MIs[0] off
-// CHECK-NEXT: GIM_CheckType, /*MI*/0, /*Op*/3, /*Type*/GILLT_s32,
-// CHECK-NEXT: GIM_CheckRegBankForClass, /*MI*/0, /*Op*/3, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
+// CHECK-NEXT: GIM_RootCheckType, /*Op*/3, /*Type*/GILLT_s32,
+// CHECK-NEXT: GIM_RootCheckRegBankForClass, /*Op*/3, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
 // CHECK-NEXT: // (loadpost:{ *:[i32] }:{ *:[i32] } GPR32:{ *:[i32] }:$addr, GPR32:{ *:[i32] }:$off)  =>  (LDPost:{ *:[i32] }:{ *:[i32] } GPR32:{ *:[i32] }:$addr, GPR32:{ *:[i32] }:$off)
 // CHECK-NEXT: GIR_MutateOpcode, /*InsnID*/0, /*RecycleInsnID*/0, /*Opcode*/GIMT_Encode2(MyTarget::LDPost),
-// CHECK-NEXT: GIR_ConstrainSelectedInstOperands, /*InsnID*/0,
+// CHECK-NEXT: GIR_RootConstrainSelectedInstOperands,
 
 //-----------------------------------------------------------------------------
 // Test where a whole new MIR instruction is created during ISel
@@ -67,25 +67,26 @@ def : Pat<(two_in GPR32:$i1, GPR32:$i2), (TWO_INS GPR32:$i2, GPR32:$i1)>;
 
 // CHECK:      GIM_CheckOpcode, /*MI*/0, GIMT_Encode2(MyTarget::G_TWO_IN),
 // CHECK-NEXT: // MIs[0] DstI[out1]
-// CHECK-NEXT: GIM_CheckType, /*MI*/0, /*Op*/0, /*Type*/GILLT_s32,
-// CHECK-NEXT: GIM_CheckRegBankForClass, /*MI*/0, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
+// CHECK-NEXT: GIM_RootCheckType, /*Op*/0, /*Type*/GILLT_s32,
+// CHECK-NEXT: GIM_RootCheckRegBankForClass, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
 // CHECK-NEXT: // MIs[0] DstI[out2]
-// CHECK-NEXT: GIM_CheckType, /*MI*/0, /*Op*/1, /*Type*/GILLT_s32,
-// CHECK-NEXT: GIM_CheckRegBankForClass, /*MI*/0, /*Op*/1, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
+// CHECK-NEXT: GIM_RootCheckType, /*Op*/1, /*Type*/GILLT_s32,
+// CHECK-NEXT: GIM_RootCheckRegBankForClass, /*Op*/1, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
 // CHECK-NEXT: // MIs[0] i1
-// CHECK-NEXT: GIM_CheckType, /*MI*/0, /*Op*/2, /*Type*/GILLT_s32,
-// CHECK-NEXT: GIM_CheckRegBankForClass, /*MI*/0, /*Op*/2, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
+// CHECK-NEXT: GIM_RootCheckType, /*Op*/2, /*Type*/GILLT_s32,
+// CHECK-NEXT: GIM_RootCheckRegBankForClass, /*Op*/2, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
 // CHECK-NEXT: // MIs[0] i2
-// CHECK-NEXT: GIM_CheckType, /*MI*/0, /*Op*/3, /*Type*/GILLT_s32,
-// CHECK-NEXT: GIM_CheckRegBankForClass, /*MI*/0, /*Op*/3, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
+// CHECK-NEXT: GIM_RootCheckType, /*Op*/3, /*Type*/GILLT_s32,
+// CHECK-NEXT: GIM_RootCheckRegBankForClass, /*Op*/3, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
 // CHECK-NEXT: // (two_in:{ *:[i32] }:{ *:[i32] } GPR32:{ *:[i32] }:$i1, GPR32:{ *:[i32] }:$i2)  =>  (TWO_INS:{ *:[i32] }:{ *:[i32] } GPR32:{ *:[i32] }:$i2, GPR32:{ *:[i32] }:$i1)
-// CHECK-NEXT: GIR_BuildMI, /*InsnID*/0, /*Opcode*/GIMT_Encode2(MyTarget::TWO_INS),
-// CHECK-NEXT: GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/0, // DstI[out1]
-// CHECK-NEXT: GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/1, // DstI[out2]
-// CHECK-NEXT: GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/3, // i2
-// CHECK-NEXT: GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/2, // i1
-// CHECK-NEXT: GIR_ConstrainSelectedInstOperands, /*InsnID*/0,
-// CHECK-NEXT: GIR_EraseFromParent, /*InsnID*/0,
+// CHECK-NEXT: GIR_BuildRootMI, /*Opcode*/GIMT_Encode2(MyTarget::TWO_INS),
+// CHECK-NEXT: GIR_RootToRootCopy, /*OpIdx*/0, // DstI[out1]
+// CHECK-NEXT: GIR_RootToRootCopy, /*OpIdx*/1, // DstI[out2]
+// CHECK-NEXT: GIR_RootToRootCopy, /*OpIdx*/3, // i2
+// CHECK-NEXT: GIR_RootToRootCopy, /*OpIdx*/2, // i1
+// CHECK-NEXT: GIR_RootConstrainSelectedInstOperands,
+// CHECK-NEXT: // GIR_Coverage
+// CHECK-NEXT: GIR_EraseRootFromParent_Done,
 
 //-----------------------------------------------------------------------------
 // Test where implicit defs are added using Defs.
@@ -99,10 +100,10 @@ def : Pat<(i32 (add i32:$src, i32:$src)),
 
 // CHECK: GIM_CheckOpcode, /*MI*/0, GIMT_Encode2(TargetOpcode::G_ADD),
 // CHECK-NEXT: // MIs[0] DstI[dst]
-// CHECK-NEXT: GIM_CheckType, /*MI*/0, /*Op*/0, /*Type*/GILLT_s32,
-// CHECK-NEXT: GIM_CheckRegBankForClass, /*MI*/0, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
+// CHECK-NEXT: GIM_RootCheckType, /*Op*/0, /*Type*/GILLT_s32,
+// CHECK-NEXT: GIM_RootCheckRegBankForClass, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
 // CHECK-NEXT: // MIs[0] src
-// CHECK-NEXT: GIM_CheckType, /*MI*/0, /*Op*/1, /*Type*/GILLT_s32,
+// CHECK-NEXT: GIM_RootCheckType, /*Op*/1, /*Type*/GILLT_s32,
 // CHECK-NEXT: // MIs[0] src
 // CHECK-NEXT: GIM_CheckIsSameOperand, /*MI*/0, /*OpIdx*/2, /*OtherMI*/0, /*OtherOpIdx*/1,
 // CHECK-NEXT: // (add:{ *:[i32] } i32:{ *:[i32] }:$src, i32:{ *:[i32] }:$src)  =>  (OtherInstr:{ *:[i32] } (ImplicitDefInstr:{ *:[i32] }:{ *:[i32] } GPR32:{ *:[i32] }:$src))
@@ -112,21 +113,22 @@ def : Pat<(i32 (add i32:$src, i32:$src)),
 // CHECK-NEXT: GIR_Copy, /*NewInsnID*/1, /*OldInsnID*/0, /*OpIdx*/1, // src
 // CHECK-NEXT: GIR_SetImplicitDefDead, /*InsnID*/1, /*OpIdx for MyTarget::R0*/0,
 // CHECK-NEXT: GIR_ConstrainSelectedInstOperands, /*InsnID*/1,
-// CHECK-NEXT: GIR_BuildMI, /*InsnID*/0, /*Opcode*/GIMT_Encode2(MyTarget::OtherInstr),
-// CHECK-NEXT: GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/0, // DstI[dst]
+// CHECK-NEXT: GIR_BuildRootMI, /*Opcode*/GIMT_Encode2(MyTarget::OtherInstr),
+// CHECK-NEXT: GIR_RootToRootCopy, /*OpIdx*/0, // DstI[dst]
 // CHECK-NEXT: GIR_AddSimpleTempRegister, /*InsnID*/0, /*TempRegID*/0,
-// CHECK-NEXT: GIR_ConstrainSelectedInstOperands, /*InsnID*/0,
-// CHECK-NEXT: GIR_EraseFromParent, /*InsnID*/0,
+// CHECK-NEXT: GIR_RootConstrainSelectedInstOperands,
+// CHECK-NEXT: // GIR_Coverage
+// CHECK-NEXT: GIR_EraseRootFromParent_Done,
 
 //-----------------------------------------------------------------------------
 // Test when the inner instruction in the output pattern has two outs
 
 // CHECK:      GIM_CheckOpcode, /*MI*/0, GIMT_Encode2(TargetOpcode::G_ADD),
 // CHECK-NEXT: // MIs[0] DstI[dst]
-// CHECK-NEXT: GIM_CheckType, /*MI*/0, /*Op*/0, /*Type*/GILLT_s32,
-// CHECK-NEXT: GIM_CheckRegBankForClass, /*MI*/0, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
+// CHECK-NEXT: GIM_RootCheckType, /*Op*/0, /*Type*/GILLT_s32,
+// CHECK-NEXT: GIM_RootCheckRegBankForClass, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
 // CHECK-NEXT: // MIs[0] src
-// CHECK-NEXT: GIM_CheckType, /*MI*/0, /*Op*/1, /*Type*/GILLT_s32,
+// CHECK-NEXT: GIM_RootCheckType, /*Op*/1, /*Type*/GILLT_s32,
 // CHECK-NEXT: // MIs[0] src
 // CHECK-NEXT: GIM_CheckIsSameOperand, /*MI*/0, /*OpIdx*/2, /*OtherMI*/0, /*OtherOpIdx*/1,
 // CHECK-NEXT: // (add:{ *:[i32] } i32:{ *:[i32] }:$src, i32:{ *:[i32] }:$src)  =>  (OtherInstr:{ *:[i32] } (TwoOutsInstr:{ *:[i32] }:{ *:[i32] } GPR32:{ *:[i32] }:$src))
@@ -137,11 +139,12 @@ def : Pat<(i32 (add i32:$src, i32:$src)),
 // CHECK-NEXT: GIR_AddTempRegister, /*InsnID*/1, /*TempRegID*/1, /*TempRegFlags*/GIMT_Encode2(RegState::Define|RegState::Dead),
 // CHECK-NEXT: GIR_Copy, /*NewInsnID*/1, /*OldInsnID*/0, /*OpIdx*/1, // src
 // CHECK-NEXT: GIR_ConstrainSelectedInstOperands, /*InsnID*/1,
-// CHECK-NEXT: GIR_BuildMI, /*InsnID*/0, /*Opcode*/GIMT_Encode2(MyTarget::OtherInstr),
-// CHECK-NEXT: GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/0, // DstI[dst]
+// CHECK-NEXT: GIR_BuildRootMI, /*Opcode*/GIMT_Encode2(MyTarget::OtherInstr),
+// CHECK-NEXT: GIR_RootToRootCopy, /*OpIdx*/0, // DstI[dst]
 // CHECK-NEXT: GIR_AddSimpleTempRegister, /*InsnID*/0, /*TempRegID*/0,
-// CHECK-NEXT: GIR_ConstrainSelectedInstOperands, /*InsnID*/0,
-// CHECK-NEXT: GIR_EraseFromParent, /*InsnID*/0,
+// CHECK-NEXT: GIR_RootConstrainSelectedInstOperands,
+// CHECK-NEXT: // GIR_Coverage
+// CHECK-NEXT: GIR_EraseRootFromParent_Done,
 
 def TwoOutsInstr : I<(outs GPR32:$out1, GPR32:$out2), (ins GPR32:$src), []>;
 
diff --git a/llvm/test/TableGen/GlobalISelEmitter-nested-subregs.td b/llvm/test/TableGen/GlobalISelEmitter-nested-subregs.td
index 234b19a146c1..25a39a40da61 100644
--- a/llvm/test/TableGen/GlobalISelEmitter-nested-subregs.td
+++ b/llvm/test/TableGen/GlobalISelEmitter-nested-subregs.td
@@ -33,10 +33,10 @@ def A0  : RegisterClass<"MyTarget", [i32], 32, (add a0)>;
 // CHECK: GIM_CheckNumOperands, /*MI*/0, /*Expected*/2,
 // CHECK-NEXT: GIM_CheckOpcode, /*MI*/0, GIMT_Encode2(TargetOpcode::G_ANYEXT),
 // CHECK-NEXT: // MIs[0] DstI[dst]
-// CHECK-NEXT: GIM_CheckType, /*MI*/0, /*Op*/0, /*Type*/GILLT_s16,
-// CHECK-NEXT: GIM_CheckRegBankForClass, /*MI*/0, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::A0RegClassID),
+// CHECK-NEXT: GIM_RootCheckType, /*Op*/0, /*Type*/GILLT_s16,
+// CHECK-NEXT: GIM_RootCheckRegBankForClass, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::A0RegClassID),
 // CHECK-NEXT: // MIs[0] src
-// CHECK-NEXT: GIM_CheckType, /*MI*/0, /*Op*/1, /*Type*/GILLT_s8,
+// CHECK-NEXT: GIM_RootCheckType, /*Op*/1, /*Type*/GILLT_s8,
 // CHECK-NEXT: // (anyext:{ *:[i16] } i8:{ *:[i8] }:$src)  =>  (EXTRACT_SUBREG:{ *:[i16] } (INSERT_SUBREG:{ *:[i32] } (IMPLICIT_DEF:{ *:[i32] }), A0b:{ *:[i8] }:$src, lo8:{ *:[i32] }), lo16:{ *:[i32] })
 // CHECK-NEXT: GIR_MakeTempReg, /*TempRegID*/0, /*TypeID*/GILLT_s32,
 // CHECK-NEXT: GIR_MakeTempReg, /*TempRegID*/1, /*TypeID*/GILLT_s32,
@@ -51,12 +51,13 @@ def A0  : RegisterClass<"MyTarget", [i32], 32, (add a0)>;
 // CHECK-NEXT: GIR_ConstrainOperandRC, /*InsnID*/1, /*Op*/0, GIMT_Encode2(MyTarget::A0RegClassID),
 // CHECK-NEXT: GIR_ConstrainOperandRC, /*InsnID*/1, /*Op*/1, GIMT_Encode2(MyTarget::A0RegClassID),
 // CHECK-NEXT: GIR_ConstrainOperandRC, /*InsnID*/1, /*Op*/2, GIMT_Encode2(MyTarget::A0bRegClassID),
-// CHECK-NEXT: GIR_BuildMI, /*InsnID*/0, /*Opcode*/GIMT_Encode2(TargetOpcode::COPY),
-// CHECK-NEXT: GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/0, // DstI[dst]
+// CHECK-NEXT: GIR_BuildRootMI, /*Opcode*/GIMT_Encode2(TargetOpcode::COPY),
+// CHECK-NEXT: GIR_RootToRootCopy, /*OpIdx*/0, // DstI[dst]
 // CHECK-NEXT: GIR_AddTempSubRegister, /*InsnID*/0, /*TempRegID*/0, /*TempRegFlags*/GIMT_Encode2(0), GIMT_Encode2(MyTarget::lo16),
 // CHECK-NEXT: GIR_ConstrainOperandRC, /*InsnID*/0, /*Op*/0, GIMT_Encode2(MyTarget::A0wRegClassID),
 // CHECK-NEXT: GIR_ConstrainOperandRC, /*InsnID*/0, /*Op*/1, GIMT_Encode2(MyTarget::A0RegClassID),
-// CHECK-NEXT: GIR_EraseFromParent, /*InsnID*/0,
+// CHECK-NEXT: // GIR_Coverage
+// CHECK-NEXT: GIR_EraseRootFromParent_Done,
 def : Pat<(i16 (anyext i8:$src)),
           (i16 (EXTRACT_SUBREG
                  (i32 (INSERT_SUBREG
diff --git a/llvm/test/TableGen/GlobalISelEmitter-notype-output-pattern.td b/llvm/test/TableGen/GlobalISelEmitter-notype-output-pattern.td
index 86a6e1651fca..622d7fa1f795 100644
--- a/llvm/test/TableGen/GlobalISelEmitter-notype-output-pattern.td
+++ b/llvm/test/TableGen/GlobalISelEmitter-notype-output-pattern.td
@@ -6,11 +6,11 @@ include "GlobalISelEmitterCommon.td"
 // CHECK:      constexpr static uint8_t MatchTable0[] = {
 // CHECK-NEXT:   GIM_Try,
 // CHECK-NEXT:     GIM_CheckOpcode{{.*}}GIMT_Encode2(TargetOpcode::G_ANYEXT),
-// CHECK-NEXT:     GIM_CheckType{{.*}}/*Type*/GILLT_s32,
-// CHECK-NEXT:     GIM_CheckType{{.*}}/*Type*/GILLT_s8,
-// CHECK-NEXT:     GIM_CheckRegBankForClass{{.*}}/*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
+// CHECK-NEXT:     GIM_RootCheckType{{.*}}/*Type*/GILLT_s32,
+// CHECK-NEXT:     GIM_RootCheckType{{.*}}/*Type*/GILLT_s8,
+// CHECK-NEXT:     GIM_RootCheckRegBankForClass{{.*}}/*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
 // CHECK-NEXT:     // (anyext:{{.*}}=>{{.*}}(SELECT_I4:
-// CHECK:          GIR_Done,
+// CHECK:          GIR_EraseRootFromParent_Done,
 // CHECK-NEXT:   // Label 0:
 // CHECK-NEXT:   GIM_Reject,
 // CHECK-NEXT:   };
diff --git a/llvm/test/TableGen/GlobalISelEmitter-output-discard.td b/llvm/test/TableGen/GlobalISelEmitter-output-discard.td
index 30f9c5f47555..7a0242d9a992 100644
--- a/llvm/test/TableGen/GlobalISelEmitter-output-discard.td
+++ b/llvm/test/TableGen/GlobalISelEmitter-output-discard.td
@@ -8,19 +8,20 @@ def ADD_CO : I<(outs GPR32:$dst, GPR8:$flag),
                (ins GPR32Op:$src0, GPR32Op:$src1), []>;
 
 // GISEL: GIM_CheckOpcode, /*MI*/0, GIMT_Encode2(TargetOpcode::G_ADD),
-// GISEL-NEXT: GIM_CheckType, /*MI*/0, /*Op*/0, /*Type*/GILLT_s32,
-// GISEL-NEXT: GIM_CheckType, /*MI*/0, /*Op*/1, /*Type*/GILLT_s32,
-// GISEL-NEXT: GIM_CheckType, /*MI*/0, /*Op*/2, /*Type*/GILLT_s32,
-// GISEL-NEXT: GIM_CheckRegBankForClass, /*MI*/0, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
+// GISEL-NEXT: GIM_RootCheckType, /*Op*/0, /*Type*/GILLT_s32,
+// GISEL-NEXT: GIM_RootCheckType, /*Op*/1, /*Type*/GILLT_s32,
+// GISEL-NEXT: GIM_RootCheckType, /*Op*/2, /*Type*/GILLT_s32,
+// GISEL-NEXT: GIM_RootCheckRegBankForClass, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
 // GISEL-NEXT: // (add:{ *:[i32] } i32:{ *:[i32] }:$src0, i32:{ *:[i32] }:$src1)  =>  (ADD_CO:{ *:[i32] }:{ *:[i8] } GPR32:{ *:[i32] }:$src0, GPR32:{ *:[i32] }:$src1)
 // GISEL-NEXT: GIR_MakeTempReg, /*TempRegID*/0, /*TypeID*/GILLT_s8,
-// GISEL-NEXT: GIR_BuildMI, /*InsnID*/0, /*Opcode*/GIMT_Encode2(MyTarget::ADD_CO),
-// GISEL-NEXT: GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/0, // DstI[dst]
+// GISEL-NEXT: GIR_BuildRootMI, /*Opcode*/GIMT_Encode2(MyTarget::ADD_CO),
+// GISEL-NEXT: GIR_RootToRootCopy, /*OpIdx*/0, // DstI[dst]
 // GISEL-NEXT: GIR_AddTempRegister, /*InsnID*/0, /*TempRegID*/0, /*TempRegFlags*/GIMT_Encode2(RegState::Define|RegState::Dead),
-// GISEL-NEXT: GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/1, // src0
-// GISEL-NEXT: GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/2, // src1
-// GISEL-NEXT: GIR_ConstrainSelectedInstOperands, /*InsnID*/0,
-// GISEL-NEXT: GIR_EraseFromParent, /*InsnID*/0,
+// GISEL-NEXT: GIR_RootToRootCopy, /*OpIdx*/1, // src0
+// GISEL-NEXT: GIR_RootToRootCopy, /*OpIdx*/2, // src1
+// GISEL-NEXT: GIR_RootConstrainSelectedInstOperands,
+// GISEL-NEXT: // GIR_Coverage
+// GISEL-NEXT: GIR_EraseRootFromParent_Done,
 def : Pat <
   (add i32:$src0, i32:$src1),
   (ADD_CO GPR32:$src0, GPR32:$src1)
diff --git a/llvm/test/TableGen/GlobalISelEmitter-zero-reg.td b/llvm/test/TableGen/GlobalISelEmitter-zero-reg.td
index dd47b905e57d..ddf02240ee1f 100644
--- a/llvm/test/TableGen/GlobalISelEmitter-zero-reg.td
+++ b/llvm/test/TableGen/GlobalISelEmitter-zero-reg.td
@@ -25,18 +25,19 @@ def INST : PredI<(outs GPR32:$dst), (ins GPR32:$src), []>;
 // CHECK-NEXT: GIM_CheckMemorySizeEqualToLLT, /*MI*/0, /*MMO*/0, /*OpIdx*/0,
 // CHECK-NEXT: GIM_CheckAtomicOrdering, /*MI*/0, /*Order*/(uint8_t)AtomicOrdering::NotAtomic,
 // CHECK-NEXT: // MIs[0] DstI[dst]
-// CHECK-NEXT: GIM_CheckType, /*MI*/0, /*Op*/0, /*Type*/GILLT_s32,
-// CHECK-NEXT: GIM_CheckRegBankForClass, /*MI*/0, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
+// CHECK-NEXT: GIM_RootCheckType, /*Op*/0, /*Type*/GILLT_s32,
+// CHECK-NEXT: GIM_RootCheckRegBankForClass, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
 // CHECK-NEXT: // MIs[0] src
 // CHECK-NEXT: GIM_CheckPointerToAny, /*MI*/0, /*Op*/1, /*SizeInBits*/32,
-// CHECK-NEXT: GIM_CheckRegBankForClass, /*MI*/0, /*Op*/1, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
+// CHECK-NEXT: GIM_RootCheckRegBankForClass, /*Op*/1, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
 // CHECK-NEXT: // (ld:{ *:[i32] } GPR32:{ *:[i32] }:$src)<<P:Predicate_unindexedload>><<P:Predicate_load>>  =>  (INST:{ *:[i32] } GPR32:{ *:[i32] }:$src)
-// CHECK-NEXT: GIR_BuildMI, /*InsnID*/0, /*Opcode*/GIMT_Encode2(MyTarget::INST),
-// CHECK-NEXT: GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/0, // DstI[dst]
-// CHECK-NEXT: GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/1, // src
+// CHECK-NEXT: GIR_BuildRootMI, /*Opcode*/GIMT_Encode2(MyTarget::INST),
+// CHECK-NEXT: GIR_RootToRootCopy, /*OpIdx*/0, // DstI[dst]
+// CHECK-NEXT: GIR_RootToRootCopy, /*OpIdx*/1, // src
 // CHECK-NEXT: GIR_AddRegister, /*InsnID*/0, GIMT_Encode2(MyTarget::NoRegister), /*AddRegisterRegFlags*/GIMT_Encode2(0),
 // CHECK-NEXT: GIR_MergeMemOperands, /*InsnID*/0, /*NumInsns*/1, /*MergeInsnID's*/0,
-// CHECK-NEXT: GIR_ConstrainSelectedInstOperands, /*InsnID*/0,
-// CHECK-NEXT: GIR_EraseFromParent, /*InsnID*/0,
+// CHECK-NEXT: GIR_RootConstrainSelectedInstOperands,
+// CHECK-NEXT: // GIR_Coverage
+// CHECK-NEXT: GIR_EraseRootFromParent_Done,
 def : Pat<(i32 (load GPR32:$src)),
           (INST GPR32:$src)>;
diff --git a/llvm/test/TableGen/GlobalISelEmitter.td b/llvm/test/TableGen/GlobalISelEmitter.td
index 82ecc4495e80..23b3d6f59b38 100644
--- a/llvm/test/TableGen/GlobalISelEmitter.td
+++ b/llvm/test/TableGen/GlobalISelEmitter.td
@@ -110,7 +110,7 @@ def HasC : Predicate<"Subtarget->hasC()"> { let RecomputePerFunction = 1; }
 
 // CHECK-LABEL: PredicateBitset MyTargetInstructionSelector::
 // CHECK-NEXT:  computeAvailableModuleFeatures(const MyTargetSubtarget *Subtarget) const {
-// CHECK-NEXT:    PredicateBitset Features;
+// CHECK-NEXT:    PredicateBitset Features{};
 // CHECK-NEXT:    if (Subtarget->hasA())
 // CHECK-NEXT:      Features.set(Feature_HasABit);
 // CHECK-NEXT:    if (Subtarget->hasB())
@@ -120,7 +120,7 @@ def HasC : Predicate<"Subtarget->hasC()"> { let RecomputePerFunction = 1; }
 
 // CHECK-LABEL: PredicateBitset MyTargetInstructionSelector::
 // CHECK-NEXT:  computeAvailableFunctionFeatures(const MyTargetSubtarget *Subtarget, const MachineFunction *MF) const {
-// CHECK-NEXT:    PredicateBitset Features;
+// CHECK-NEXT:    PredicateBitset Features{};
 // CHECK-NEXT:    if (Subtarget->hasC())
 // CHECK-NEXT:      Features.set(Feature_HasCBit);
 // CHECK-NEXT:    return Features;
@@ -237,29 +237,29 @@ def HasC : Predicate<"Subtarget->hasC()"> { let RecomputePerFunction = 1; }
 // R19O:       // Label [[CASE_ADD_NUM]]: @[[CASE_ADD]]
 // R19O:       // Label [[CASE_SELECT_NUM]]: @[[CASE_SELECT]]
 // R19O-NEXT:  GIM_Try, /*On fail goto*//*Label [[GROUP_NUM:[0-9]+]]*/ GIMT_Encode4([[GROUP:[0-9]+]]),
-// R19O-NEXT:    GIM_CheckType, /*MI*/0, /*Op*/0, /*Type*/GILLT_s32,
-// R19O-NEXT:    GIM_CheckType, /*MI*/0, /*Op*/1, /*Type*/GILLT_s32,
-// R19O-NEXT:    GIM_CheckType, /*MI*/0, /*Op*/2, /*Type*/GILLT_s32,
-// R19O-NEXT:    GIM_CheckType, /*MI*/0, /*Op*/3, /*Type*/GILLT_s32,
+// R19O-NEXT:    GIM_RootCheckType, /*Op*/0, /*Type*/GILLT_s32,
+// R19O-NEXT:    GIM_RootCheckType, /*Op*/1, /*Type*/GILLT_s32,
+// R19O-NEXT:    GIM_RootCheckType, /*Op*/2, /*Type*/GILLT_s32,
+// R19O-NEXT:    GIM_RootCheckType, /*Op*/3, /*Type*/GILLT_s32,
 //
 // R19C-NEXT:  GIM_Try, /*On fail goto*//*Label [[LABEL_NUM:[0-9]+]]*/ GIMT_Encode4([[LABEL:[0-9]+]]),
 //
-// R19O-NEXT:    GIM_CheckRegBankForClass, /*MI*/0, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
-// R19O-NEXT:    GIM_CheckRegBankForClass, /*MI*/0, /*Op*/1, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
+// R19O-NEXT:    GIM_RootCheckRegBankForClass, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
+// R19O-NEXT:    GIM_RootCheckRegBankForClass, /*Op*/1, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
 // R19N-NEXT:    GIM_CheckNumOperands, /*MI*/0, /*Expected*/4,
 // R19N-NEXT:    GIM_CheckOpcode, /*MI*/0, GIMT_Encode2(TargetOpcode::G_SELECT),
 // R19N-NEXT:    // MIs[0] DstI[dst]
-// R19N-NEXT:    GIM_CheckType, /*MI*/0, /*Op*/0, /*Type*/GILLT_s32,
-// R19N-NEXT:    GIM_CheckRegBankForClass, /*MI*/0, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
+// R19N-NEXT:    GIM_RootCheckType, /*Op*/0, /*Type*/GILLT_s32,
+// R19N-NEXT:    GIM_RootCheckRegBankForClass, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
 // R19N-NEXT:    // MIs[0] src1
-// R19N-NEXT:    GIM_CheckType, /*MI*/0, /*Op*/1, /*Type*/GILLT_s32,
-// R19N-NEXT:    GIM_CheckRegBankForClass, /*MI*/0, /*Op*/1, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
+// R19N-NEXT:    GIM_RootCheckType, /*Op*/1, /*Type*/GILLT_s32,
+// R19N-NEXT:    GIM_RootCheckRegBankForClass, /*Op*/1, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
 // R19N-NEXT:    // MIs[0] complex_rr:src2a:src2b
-// R19N-NEXT:    GIM_CheckType, /*MI*/0, /*Op*/2, /*Type*/GILLT_s32,
+// R19N-NEXT:    GIM_RootCheckType, /*Op*/2, /*Type*/GILLT_s32,
 //
 // R19N-NEXT:    GIM_CheckComplexPattern, /*MI*/0, /*Op*/2, /*Renderer*/GIMT_Encode2(0), GIMT_Encode2(GICP_gi_complex_rr),
 // R19N-NEXT:    // MIs[0] Operand 3
-// R19N-NEXT:    GIM_CheckType, /*MI*/0, /*Op*/3, /*Type*/GILLT_s32,
+// R19N-NEXT:    GIM_RootCheckType, /*Op*/3, /*Type*/GILLT_s32,
 // R19C-NEXT:    GIM_RecordInsn, /*DefineMI*/1, /*MI*/0, /*OpIdx*/3, // MIs[1]
 // R19N-NEXT:    GIM_CheckNumOperands, /*MI*/1, /*Expected*/4,
 // R19C-NEXT:    GIM_CheckOpcode, /*MI*/1, GIMT_Encode2(TargetOpcode::G_SELECT),
@@ -277,7 +277,7 @@ def HasC : Predicate<"Subtarget->hasC()"> { let RecomputePerFunction = 1; }
 // R19N-NEXT:    GIM_CheckType, /*MI*/1, /*Op*/3, /*Type*/GILLT_s32,
 // R19N-NEXT:    GIM_CheckComplexPattern, /*MI*/1, /*Op*/3, /*Renderer*/GIMT_Encode2(2), GIMT_Encode2(GICP_gi_complex),
 // R19O-NEXT:    GIM_CheckRegBankForClass, /*MI*/1, /*Op*/1, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
-// R19C-NEXT:    GIM_CheckIsSafeToFold, /*InsnID*/1,
+// R19C-NEXT:    GIM_CheckIsSafeToFold, /*NumInsns*/1,
 // R19O-NEXT:    GIM_CheckComplexPattern, /*MI*/0, /*Op*/2, /*Renderer*/GIMT_Encode2(0), GIMT_Encode2(GICP_gi_complex_rr),
 // R19O-NEXT:    GIM_CheckComplexPattern, /*MI*/1, /*Op*/2, /*Renderer*/GIMT_Encode2(1), GIMT_Encode2(GICP_gi_complex),
 // R19O-NEXT:    GIM_CheckComplexPattern, /*MI*/1, /*Op*/3, /*Renderer*/GIMT_Encode2(2), GIMT_Encode2(GICP_gi_complex),
@@ -290,16 +290,15 @@ def HasC : Predicate<"Subtarget->hasC()"> { let RecomputePerFunction = 1; }
 // R19C-NEXT:    GIR_ComplexSubOperandRenderer, /*InsnID*/1, /*RendererID*/GIMT_Encode2(2), /*SubOperand*/0, // src5a
 // R19C-NEXT:    GIR_ComplexSubOperandRenderer, /*InsnID*/1, /*RendererID*/GIMT_Encode2(2), /*SubOperand*/1, // src5b
 // R19C-NEXT:    GIR_ConstrainSelectedInstOperands, /*InsnID*/1,
-// R19C-NEXT:    GIR_BuildMI, /*InsnID*/0, /*Opcode*/GIMT_Encode2(MyTarget::INSN3),
-// R19C-NEXT:    GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/0, //  DstI[dst]
-// R19C-NEXT:    GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/1, // src1
+// R19C-NEXT:    GIR_BuildRootMI, /*Opcode*/GIMT_Encode2(MyTarget::INSN3),
+// R19C-NEXT:    GIR_RootToRootCopy, /*OpIdx*/0, //  DstI[dst]
+// R19C-NEXT:    GIR_RootToRootCopy, /*OpIdx*/1, // src1
 // R19C-NEXT:    GIR_ComplexSubOperandRenderer, /*InsnID*/0, /*RendererID*/GIMT_Encode2(0), /*SubOperand*/1, // src2b
 // R19C-NEXT:    GIR_ComplexSubOperandRenderer, /*InsnID*/0, /*RendererID*/GIMT_Encode2(0), /*SubOperand*/0, // src2a
 // R19C-NEXT:    GIR_AddSimpleTempRegister, /*InsnID*/0, /*TempRegID*/0,
-// R19C-NEXT:    GIR_ConstrainSelectedInstOperands, /*InsnID*/0,
-// R19C-NEXT:    GIR_EraseFromParent, /*InsnID*/0,
+// R19C-NEXT:    GIR_RootConstrainSelectedInstOperands,
 // R19C-NEXT:    // GIR_Coverage, 20,
-// R19C-NEXT:    GIR_Done,
+// R19C-NEXT:    GIR_EraseRootFromParent_Done,
 // R19C-NEXT:  // Label [[LABEL_NUM]]: @[[LABEL]]
 //
 // R19O:       // Label [[GROUP_NUM]]: @[[GROUP]]
@@ -326,49 +325,48 @@ def : Pat<(select GPR32:$src1, (complex_rr GPR32:$src2a, GPR32:$src2b),
 // R21O:       // Label [[CASE_ADD_NUM]]: @[[CASE_ADD]]
 // R21O:       // Label [[CASE_SELECT_NUM]]: @[[CASE_SELECT]]
 // R21O-NEXT:  GIM_Try, /*On fail goto*//*Label [[GROUP_NUM:[0-9]+]]*/ GIMT_Encode4([[GROUP:[0-9]+]]),
-// R21O-NEXT:    GIM_CheckType, /*MI*/0, /*Op*/0, /*Type*/GILLT_s32,
-// R21O-NEXT:    GIM_CheckType, /*MI*/0, /*Op*/1, /*Type*/GILLT_s32,
-// R21O-NEXT:    GIM_CheckType, /*MI*/0, /*Op*/2, /*Type*/GILLT_s32,
-// R21O-NEXT:    GIM_CheckType, /*MI*/0, /*Op*/3, /*Type*/GILLT_s32,
+// R21O-NEXT:    GIM_RootCheckType, /*Op*/0, /*Type*/GILLT_s32,
+// R21O-NEXT:    GIM_RootCheckType, /*Op*/1, /*Type*/GILLT_s32,
+// R21O-NEXT:    GIM_RootCheckType, /*Op*/2, /*Type*/GILLT_s32,
+// R21O-NEXT:    GIM_RootCheckType, /*Op*/3, /*Type*/GILLT_s32,
 //
 // R21C-NEXT:  GIM_Try, /*On fail goto*//*Label [[PREV_NUM:[0-9]+]]*/ GIMT_Encode4([[PREV:[0-9]+]]), // Rule ID 20 //
-// R21C-NOT:     GIR_Done,
+// R21C-NOT:     GIR_EraseRootFromParent_Done,
 // R21C:         // GIR_Coverage, 20,
-// R21C-NEXT:    GIR_Done,
+// R21C-NEXT:    GIR_EraseRootFromParent_Done,
 // R21C-NEXT:  // Label [[PREV_NUM]]: @[[PREV]]
 // R21C-NEXT:  GIM_Try, /*On fail goto*//*Label [[LABEL_NUM:[0-9]+]]*/ GIMT_Encode4([[LABEL:[0-9]+]]), // Rule ID 22 //
 //
-// R21O-NEXT:    GIM_CheckRegBankForClass, /*MI*/0, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
-// R21O-NEXT:    GIM_CheckRegBankForClass, /*MI*/0, /*Op*/1, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
+// R21O-NEXT:    GIM_RootCheckRegBankForClass, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
+// R21O-NEXT:    GIM_RootCheckRegBankForClass, /*Op*/1, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
 // R21N-NEXT:    GIM_CheckNumOperands, /*MI*/0, /*Expected*/4,
 // R21N-NEXT:    GIM_CheckOpcode, /*MI*/0, GIMT_Encode2(TargetOpcode::G_SELECT),
 // R21N-NEXT:    // MIs[0] DstI[dst]
-// R21N-NEXT:    GIM_CheckType, /*MI*/0, /*Op*/0, /*Type*/GILLT_s32,
-// R21N-NEXT:    GIM_CheckRegBankForClass, /*MI*/0, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
+// R21N-NEXT:    GIM_RootCheckType, /*Op*/0, /*Type*/GILLT_s32,
+// R21N-NEXT:    GIM_RootCheckRegBankForClass, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
 // R21N-NEXT:    // MIs[0] src1
-// R21N-NEXT:    GIM_CheckType, /*MI*/0, /*Op*/1, /*Type*/GILLT_s32,
-// R21N-NEXT:    GIM_CheckRegBankForClass, /*MI*/0, /*Op*/1, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
+// R21N-NEXT:    GIM_RootCheckType, /*Op*/1, /*Type*/GILLT_s32,
+// R21N-NEXT:    GIM_RootCheckRegBankForClass, /*Op*/1, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
 // R21N-NEXT:    // MIs[0] src2
-// R21N-NEXT:    GIM_CheckType, /*MI*/0, /*Op*/2, /*Type*/GILLT_s32,
+// R21N-NEXT:    GIM_RootCheckType, /*Op*/2, /*Type*/GILLT_s32,
 //
 // R21O-NEXT:    GIM_CheckCxxInsnPredicate, /*MI*/0, /*FnId*/GIMT_Encode2(GICXXPred_MI_Predicate_frag),
 // R21C-NEXT:    GIM_CheckComplexPattern, /*MI*/0, /*Op*/2, /*Renderer*/GIMT_Encode2(0), GIMT_Encode2(GICP_gi_complex),
 // R21N-NEXT:    // MIs[0] src3
-// R21N-NEXT:    GIM_CheckType, /*MI*/0, /*Op*/3, /*Type*/GILLT_s32,
+// R21N-NEXT:    GIM_RootCheckType, /*Op*/3, /*Type*/GILLT_s32,
 // R21C-NEXT:    GIM_CheckComplexPattern, /*MI*/0, /*Op*/3, /*Renderer*/GIMT_Encode2(1), GIMT_Encode2(GICP_gi_complex),
 // R21N-NEXT:    GIM_CheckCxxInsnPredicate, /*MI*/0, /*FnId*/GIMT_Encode2(GICXXPred_MI_Predicate_frag),
 // R21C-NEXT:    // (select:{ *:[i32] } GPR32:{ *:[i32] }:$src1, complex:{ *:[i32] }:$src2, complex:{ *:[i32] }:$src3)<<P:Predicate_frag>> => (INSN2:{ *:[i32] } GPR32:{ *:[i32] }:$src1, complex:{ *:[i32] }:$src3, complex:{ *:[i32] }:$src2)
 
-// R21C-NEXT:    GIR_BuildMI, /*InsnID*/0, /*Opcode*/GIMT_Encode2(MyTarget::INSN2),
-// R21C-NEXT:    GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/0, //  DstI[dst]
-// R21C-NEXT:    GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/1, // src1
+// R21C-NEXT:    GIR_BuildRootMI, /*Opcode*/GIMT_Encode2(MyTarget::INSN2),
+// R21C-NEXT:    GIR_RootToRootCopy, /*OpIdx*/0, //  DstI[dst]
+// R21C-NEXT:    GIR_RootToRootCopy, /*OpIdx*/1, // src1
 // R21C-NEXT:    GIR_ComplexRenderer, /*InsnID*/0, /*RendererID*/GIMT_Encode2(1),
 // R21C-NEXT:    GIR_ComplexRenderer, /*InsnID*/0, /*RendererID*/GIMT_Encode2(0),
 // R21C-NEXT:    GIR_MergeMemOperands, /*InsnID*/0, /*NumInsns*/1, /*MergeInsnID's*/0
-// R21C-NEXT:    GIR_ConstrainSelectedInstOperands, /*InsnID*/0,
-// R21C-NEXT:    GIR_EraseFromParent, /*InsnID*/0,
+// R21C-NEXT:    GIR_RootConstrainSelectedInstOperands,
 // R21C-NEXT:    // GIR_Coverage, 22,
-// R21C-NEXT:    GIR_Done,
+// R21C-NEXT:    GIR_EraseRootFromParent_Done,
 // R21C-NEXT:  // Label [[LABEL_NUM]]: @[[LABEL]]
 //
 // R21O-NEXT:  GIM_Reject,
@@ -386,10 +384,10 @@ def : Pat<(select GPR32:$src1, (complex_rr GPR32:$src2a, GPR32:$src2b),
 // R20O:       // Label [[CASE_ADD_NUM]]: @[[CASE_ADD]]
 // R20O:       // Label [[CASE_SUB_NUM]]: @[[CASE_SUB]]
 // R20O-NEXT:  GIM_Try, /*On fail goto*//*Label [[GROUP_NUM:[0-9]+]]*/ GIMT_Encode4([[GROUP:[0-9]+]]),
-// R20O-NEXT:    GIM_CheckType, /*MI*/0, /*Op*/0, /*Type*/GILLT_s32,
-// R20O-NEXT:    GIM_CheckType, /*MI*/0, /*Op*/1, /*Type*/GILLT_s32,
-// R20O-NEXT:    GIM_CheckType, /*MI*/0, /*Op*/2, /*Type*/GILLT_s32,
-// R20O-NEXT:    GIM_CheckRegBankForClass, /*MI*/0, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
+// R20O-NEXT:    GIM_RootCheckType, /*Op*/0, /*Type*/GILLT_s32,
+// R20O-NEXT:    GIM_RootCheckType, /*Op*/1, /*Type*/GILLT_s32,
+// R20O-NEXT:    GIM_RootCheckType, /*Op*/2, /*Type*/GILLT_s32,
+// R20O-NEXT:    GIM_RootCheckRegBankForClass, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
 //
 // R20N:       GIM_Try, /*On fail goto*//*Label [[PREV_NUM:[0-9]+]]*/ GIMT_Encode4([[PREV:[0-9]+]]), // Rule ID 22 //
 // R20N:       // Label [[PREV_NUM]]: @[[PREV]]
@@ -399,25 +397,24 @@ def : Pat<(select GPR32:$src1, (complex_rr GPR32:$src2a, GPR32:$src2b),
 // R20N-NEXT:    GIM_CheckNumOperands, /*MI*/0, /*Expected*/3,
 // R20N-NEXT:    GIM_CheckOpcode, /*MI*/0, GIMT_Encode2(TargetOpcode::G_SUB),
 // R20N-NEXT:    // MIs[0] DstI[dst]
-// R20N-NEXT:    GIM_CheckType, /*MI*/0, /*Op*/0, /*Type*/GILLT_s32,
-// R20N-NEXT:    GIM_CheckRegBankForClass, /*MI*/0, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
+// R20N-NEXT:    GIM_RootCheckType, /*Op*/0, /*Type*/GILLT_s32,
+// R20N-NEXT:    GIM_RootCheckRegBankForClass, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
 // R20N-NEXT:    // MIs[0] src1
-// R20N-NEXT:    GIM_CheckType, /*MI*/0, /*Op*/1, /*Type*/GILLT_s32,
+// R20N-NEXT:    GIM_RootCheckType, /*Op*/1, /*Type*/GILLT_s32,
 //
-// R20N-NEXT:    GIM_CheckRegBankForClass, /*MI*/0, /*Op*/1, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
+// R20N-NEXT:    GIM_RootCheckRegBankForClass, /*Op*/1, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
 // R20N-NEXT:    // MIs[0] src2
-// R20N-NEXT:    GIM_CheckType, /*MI*/0, /*Op*/2, /*Type*/GILLT_s32,
-// R20O-NEXT:    GIM_CheckRegBankForClass, /*MI*/0, /*Op*/1, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
+// R20N-NEXT:    GIM_RootCheckType, /*Op*/2, /*Type*/GILLT_s32,
+// R20O-NEXT:    GIM_RootCheckRegBankForClass, /*Op*/1, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
 // R20C-NEXT:    GIM_CheckComplexPattern, /*MI*/0, /*Op*/2, /*Renderer*/GIMT_Encode2(0), GIMT_Encode2(GICP_gi_complex),
 // R20C-NEXT:    // (sub:{ *:[i32] } GPR32:{ *:[i32] }:$src1, complex:{ *:[i32] }:$src2) => (INSN1:{ *:[i32] } GPR32:{ *:[i32] }:$src1, complex:{ *:[i32] }:$src2)
-// R20C-NEXT:    GIR_BuildMI, /*InsnID*/0, /*Opcode*/GIMT_Encode2(MyTarget::INSN1),
-// R20C-NEXT:    GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/0, //  DstI[dst]
-// R20C-NEXT:    GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/1, // src1
+// R20C-NEXT:    GIR_BuildRootMI, /*Opcode*/GIMT_Encode2(MyTarget::INSN1),
+// R20C-NEXT:    GIR_RootToRootCopy, /*OpIdx*/0, //  DstI[dst]
+// R20C-NEXT:    GIR_RootToRootCopy, /*OpIdx*/1, // src1
 // R20C-NEXT:    GIR_ComplexRenderer, /*InsnID*/0, /*RendererID*/GIMT_Encode2(0),
-// R20C-NEXT:    GIR_ConstrainSelectedInstOperands, /*InsnID*/0,
-// R20C-NEXT:    GIR_EraseFromParent, /*InsnID*/0,
+// R20C-NEXT:    GIR_RootConstrainSelectedInstOperands,
 // R20C-NEXT:    // GIR_Coverage, 21,
-// R20C-NEXT:    GIR_Done,
+// R20C-NEXT:    GIR_EraseRootFromParent_Done,
 // R20C-NEXT:  // Label [[LABEL_NUM]]: @[[LABEL]]
 //
 // R20O:       // Label [[GROUP_NUM]]: @[[GROUP]]
@@ -451,10 +448,10 @@ def : Pat<(frag GPR32:$src1, complex:$src2, complex:$src3),
 // R00O:       // Label [[CASE_ADD_NUM]]: @[[CASE_ADD]]
 // R00O:       // Label [[CASE_SUB_NUM]]: @[[CASE_SUB]]
 // R00O-NEXT:  GIM_Try, /*On fail goto*//*Label [[GROUP_NUM:[0-9]+]]*/ GIMT_Encode4([[GROUP:[0-9]+]]),
-// R00O-NEXT:    GIM_CheckType, /*MI*/0, /*Op*/0, /*Type*/GILLT_s32,
-// R00O-NEXT:    GIM_CheckType, /*MI*/0, /*Op*/1, /*Type*/GILLT_s32,
-// R00O-NEXT:    GIM_CheckType, /*MI*/0, /*Op*/2, /*Type*/GILLT_s32,
-// R00O-NEXT:    GIM_CheckRegBankForClass, /*MI*/0, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
+// R00O-NEXT:    GIM_RootCheckType, /*Op*/0, /*Type*/GILLT_s32,
+// R00O-NEXT:    GIM_RootCheckType, /*Op*/1, /*Type*/GILLT_s32,
+// R00O-NEXT:    GIM_RootCheckType, /*Op*/2, /*Type*/GILLT_s32,
+// R00O-NEXT:    GIM_RootCheckRegBankForClass, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
 //
 // R00C:       GIM_Try, /*On fail goto*//*Label [[PREV_NUM:[0-9]+]]*/ GIMT_Encode4([[PREV:[0-9]+]]), // Rule ID 21 //
 // R00C:       // Label [[PREV_NUM]]: @[[PREV]]
@@ -464,10 +461,10 @@ def : Pat<(frag GPR32:$src1, complex:$src2, complex:$src3),
 // R00N-NEXT:    GIM_CheckNumOperands, /*MI*/0, /*Expected*/3,
 // R00N-NEXT:    GIM_CheckOpcode, /*MI*/0, GIMT_Encode2(TargetOpcode::G_SUB),
 // R00N-NEXT:    // MIs[0] DstI[dst]
-// R00N-NEXT:    GIM_CheckType, /*MI*/0, /*Op*/0, /*Type*/GILLT_s32,
-// R00N-NEXT:    GIM_CheckRegBankForClass, /*MI*/0, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
+// R00N-NEXT:    GIM_RootCheckType, /*Op*/0, /*Type*/GILLT_s32,
+// R00N-NEXT:    GIM_RootCheckRegBankForClass, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
 // R00N-NEXT:    // MIs[0] Operand 1
-// R00N-NEXT:    GIM_CheckType, /*MI*/0, /*Op*/1, /*Type*/GILLT_s32,
+// R00N-NEXT:    GIM_RootCheckType, /*Op*/1, /*Type*/GILLT_s32,
 // R00C-NEXT:    GIM_RecordInsn, /*DefineMI*/1, /*MI*/0, /*OpIdx*/1, // MIs[1]
 // R00N-NEXT:    GIM_CheckNumOperands, /*MI*/1, /*Expected*/3,
 // R00C-NEXT:    GIM_CheckOpcode, /*MI*/1, GIMT_Encode2(TargetOpcode::G_SUB),
@@ -481,7 +478,7 @@ def : Pat<(frag GPR32:$src1, complex:$src2, complex:$src3),
 // R00N-NEXT:    GIM_CheckType, /*MI*/1, /*Op*/2, /*Type*/GILLT_s32,
 // R00N-NEXT:    GIM_CheckRegBankForClass, /*MI*/1, /*Op*/2, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
 // R00N-NEXT:    // MIs[0] Operand 2
-// R00N-NEXT:    GIM_CheckType, /*MI*/0, /*Op*/2, /*Type*/GILLT_s32,
+// R00N-NEXT:    GIM_RootCheckType, /*Op*/2, /*Type*/GILLT_s32,
 // R00O-NEXT:    GIM_CheckRegBankForClass, /*MI*/1, /*Op*/1, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
 // R00O-NEXT:    GIM_CheckRegBankForClass, /*MI*/1, /*Op*/2, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
 // R00C-NEXT:    GIM_RecordInsn, /*DefineMI*/2, /*MI*/0, /*OpIdx*/2, // MIs[2]
@@ -498,19 +495,17 @@ def : Pat<(frag GPR32:$src1, complex:$src2, complex:$src3),
 // R00N-NEXT:    GIM_CheckRegBankForClass, /*MI*/2, /*Op*/2, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
 // R00O-NEXT:    GIM_CheckRegBankForClass, /*MI*/2, /*Op*/1, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
 // R00O-NEXT:    GIM_CheckRegBankForClass, /*MI*/2, /*Op*/2, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
-// R00C-NEXT:    GIM_CheckIsSafeToFold, /*InsnID*/1,
-// R00C-NEXT:    GIM_CheckIsSafeToFold, /*InsnID*/2,
+// R00C-NEXT:    GIM_CheckIsSafeToFold, /*NumInsns*/2,
 // R00C-NEXT:    // (sub:{ *:[i32] } (sub:{ *:[i32] } GPR32:{ *:[i32] }:$src1, GPR32:{ *:[i32] }:$src2), (sub:{ *:[i32] } GPR32:{ *:[i32] }:$src3, GPR32:{ *:[i32] }:$src4)) => (INSNBOB:{ *:[i32] } GPR32:{ *:[i32] }:$src1, GPR32:{ *:[i32] }:$src2, GPR32:{ *:[i32] }:$src3, GPR32:{ *:[i32] }:$src4)
-// R00C-NEXT:    GIR_BuildMI, /*InsnID*/0, /*Opcode*/GIMT_Encode2(MyTarget::INSNBOB),
-// R00C-NEXT:    GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/0, //  DstI[dst]
+// R00C-NEXT:    GIR_BuildRootMI, /*Opcode*/GIMT_Encode2(MyTarget::INSNBOB),
+// R00C-NEXT:    GIR_RootToRootCopy, /*OpIdx*/0, //  DstI[dst]
 // R00C-NEXT:    GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/1, /*OpIdx*/1, // src1
 // R00C-NEXT:    GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/1, /*OpIdx*/2, // src2
 // R00C-NEXT:    GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/2, /*OpIdx*/1, // src3
 // R00C-NEXT:    GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/2, /*OpIdx*/2, // src4
-// R00C-NEXT:    GIR_ConstrainSelectedInstOperands, /*InsnID*/0,
-// R00C-NEXT:    GIR_EraseFromParent, /*InsnID*/0,
+// R00C-NEXT:    GIR_RootConstrainSelectedInstOperands,
 // R00C-NEXT:    // GIR_Coverage, 0,
-// R00C-NEXT:    GIR_Done,
+// R00C-NEXT:    GIR_EraseRootFromParent_Done,
 // R00C-NEXT:  // Label [[LABEL_NUM]]: @[[LABEL]]
 //
 // R00O-NEXT:  GIM_Reject,
@@ -518,7 +513,7 @@ def : Pat<(frag GPR32:$src1, complex:$src2, complex:$src3),
 // R00O-NEXT:  GIM_Reject,
 // R00O:       // Label [[DEFAULT_NUM]]: @[[DEFAULT]]
 // R00O-NEXT:  GIM_Reject,
-// R00O-NEXT:  }; // Size: 2027 bytes
+// R00O-NEXT:  }; // Size: 1804 bytes
 
 def INSNBOB : I<(outs GPR32:$dst), (ins GPR32:$src1, GPR32:$src2, GPR32:$src3, GPR32:$src4),
                  [(set GPR32:$dst,
@@ -540,28 +535,27 @@ def INSNBOB : I<(outs GPR32:$dst), (ins GPR32:$src1, GPR32:$src2, GPR32:$src3, G
 // R01C-NEXT:    GIM_CheckNumOperands, /*MI*/0, /*Expected*/3,
 //
 // R01O-NEXT:    GIM_CheckIntrinsicID, /*MI*/0, /*Op*/1, GIMT_Encode2(Intrinsic::mytarget_nop),
-// R01O-NEXT:    GIM_CheckType, /*MI*/0, /*Op*/0, /*Type*/GILLT_s32,
-// R01O-NEXT:    GIM_CheckType, /*MI*/0, /*Op*/2, /*Type*/GILLT_s32,
-// R01O-NEXT:    GIM_CheckRegBankForClass, /*MI*/0, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
+// R01O-NEXT:    GIM_RootCheckType, /*Op*/0, /*Type*/GILLT_s32,
+// R01O-NEXT:    GIM_RootCheckType, /*Op*/2, /*Type*/GILLT_s32,
+// R01O-NEXT:    GIM_RootCheckRegBankForClass, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
 //
 // R01N-NEXT:    GIM_CheckOpcode, /*MI*/0, GIMT_Encode2(TargetOpcode::G_INTRINSIC),
 // R01N-NEXT:    // MIs[0] DstI[dst]
-// R01N-NEXT:    GIM_CheckType, /*MI*/0, /*Op*/0, /*Type*/GILLT_s32,
-// R01N-NEXT:    GIM_CheckRegBankForClass, /*MI*/0, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
+// R01N-NEXT:    GIM_RootCheckType, /*Op*/0, /*Type*/GILLT_s32,
+// R01N-NEXT:    GIM_RootCheckRegBankForClass, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
 // R01N-NEXT:    // MIs[0] Operand 1
 // R01N-NEXT:    GIM_CheckIntrinsicID, /*MI*/0, /*Op*/1, GIMT_Encode2(Intrinsic::mytarget_nop),
 // R01N-NEXT:    // MIs[0] src1
-// R01N-NEXT:    GIM_CheckType, /*MI*/0, /*Op*/2, /*Type*/GILLT_s32,
+// R01N-NEXT:    GIM_RootCheckType, /*Op*/2, /*Type*/GILLT_s32,
 //
-// R01C-NEXT:    GIM_CheckRegBankForClass, /*MI*/0, /*Op*/2, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
+// R01C-NEXT:    GIM_RootCheckRegBankForClass, /*Op*/2, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
 // R01C-NEXT:    // (intrinsic_wo_chain:{ *:[i32] } [[ID:[0-9]+]]:{ *:[iPTR] }, GPR32:{ *:[i32] }:$src1) => (MOV:{ *:[i32] } GPR32:{ *:[i32] }:$src1)
-// R01C-NEXT:    GIR_BuildMI, /*InsnID*/0, /*Opcode*/GIMT_Encode2(MyTarget::MOV),
-// R01C-NEXT:    GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/0, //  DstI[dst]
-// R01C-NEXT:    GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/2, // src1
-// R01C-NEXT:    GIR_ConstrainSelectedInstOperands, /*InsnID*/0,
-// R01C-NEXT:    GIR_EraseFromParent, /*InsnID*/0,
+// R01C-NEXT:    GIR_BuildRootMI, /*Opcode*/GIMT_Encode2(MyTarget::MOV),
+// R01C-NEXT:    GIR_RootToRootCopy, /*OpIdx*/0, //  DstI[dst]
+// R01C-NEXT:    GIR_RootToRootCopy, /*OpIdx*/2, // src1
+// R01C-NEXT:    GIR_RootConstrainSelectedInstOperands,
 // R01C-NEXT:    // GIR_Coverage, 1,
-// R01C-NEXT:    GIR_Done,
+// R01C-NEXT:    GIR_EraseRootFromParent_Done,
 // R01C-NEXT:  // Label [[LABEL_NUM]]: @[[LABEL]]
 //
 // R01O-NEXT:  GIM_Reject,
@@ -579,11 +573,11 @@ def MOV : I<(outs GPR32:$dst), (ins GPR32:$src1),
 // R02O:       // Label [[CASE_ADD_NUM]]: @[[CASE_ADD]]
 // R02O:       // Label [[CASE_XOR_NUM]]: @[[CASE_XOR]]
 // R02O-NEXT:  GIM_Try, /*On fail goto*//*Label [[GROUP_NUM:[0-9]+]]*/ GIMT_Encode4([[GROUP:[0-9]+]]),
-// R02O-NEXT:    GIM_CheckType, /*MI*/0, /*Op*/0, /*Type*/GILLT_s32,
-// R02O-NEXT:    GIM_CheckType, /*MI*/0, /*Op*/1, /*Type*/GILLT_s32,
-// R02O-NEXT:    GIM_CheckType, /*MI*/0, /*Op*/2, /*Type*/GILLT_s32,
-// R02O-NEXT:    GIM_CheckRegBankForClass, /*MI*/0, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
-// R02O-NEXT:    GIM_CheckRegBankForClass, /*MI*/0, /*Op*/1, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
+// R02O-NEXT:    GIM_RootCheckType, /*Op*/0, /*Type*/GILLT_s32,
+// R02O-NEXT:    GIM_RootCheckType, /*Op*/1, /*Type*/GILLT_s32,
+// R02O-NEXT:    GIM_RootCheckType, /*Op*/2, /*Type*/GILLT_s32,
+// R02O-NEXT:    GIM_RootCheckRegBankForClass, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
+// R02O-NEXT:    GIM_RootCheckRegBankForClass, /*Op*/1, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
 //
 // R02N:       GIM_Try, /*On fail goto*//*Label [[PREV_NUM:[0-9]+]]*/ GIMT_Encode4([[PREV:[0-9]+]]), // Rule ID 1 //
 // R02N:       // Label [[PREV_NUM]]: @[[PREV]]
@@ -593,24 +587,23 @@ def MOV : I<(outs GPR32:$dst), (ins GPR32:$src1),
 // R02N-NEXT:    GIM_CheckNumOperands, /*MI*/0, /*Expected*/3,
 // R02N-NEXT:    GIM_CheckOpcode, /*MI*/0, GIMT_Encode2(TargetOpcode::G_XOR),
 // R02N-NEXT:    // MIs[0] DstI[dst]
-// R02N-NEXT:    GIM_CheckType, /*MI*/0, /*Op*/0, /*Type*/GILLT_s32,
-// R02N-NEXT:    GIM_CheckRegBankForClass, /*MI*/0, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
+// R02N-NEXT:    GIM_RootCheckType, /*Op*/0, /*Type*/GILLT_s32,
+// R02N-NEXT:    GIM_RootCheckRegBankForClass, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
 // R02N-NEXT:    // MIs[0] src1
-// R02N-NEXT:    GIM_CheckType, /*MI*/0, /*Op*/1, /*Type*/GILLT_s32,
-// R02N-NEXT:    GIM_CheckRegBankForClass, /*MI*/0, /*Op*/1, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
+// R02N-NEXT:    GIM_RootCheckType, /*Op*/1, /*Type*/GILLT_s32,
+// R02N-NEXT:    GIM_RootCheckRegBankForClass, /*Op*/1, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
 // R02N-NEXT:    // MIs[0] Operand 2
-// R02N-NEXT:    GIM_CheckType, /*MI*/0, /*Op*/2, /*Type*/GILLT_s32,
+// R02N-NEXT:    GIM_RootCheckType, /*Op*/2, /*Type*/GILLT_s32,
 //
 // R02C-NEXT:    GIM_CheckConstantInt8, /*MI*/0, /*Op*/2, uint8_t(-2)
 // R02C-NEXT:    // (xor:{ *:[i32] } GPR32:{ *:[i32] }:$src1, -2:{ *:[i32] }) => (XORI:{ *:[i32] } GPR32:{ *:[i32] }:$src1)
-// R02C-NEXT:    GIR_BuildMI, /*InsnID*/0, /*Opcode*/GIMT_Encode2(MyTarget::XORI),
-// R02C-NEXT:    GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/0, // DstI[dst]
+// R02C-NEXT:    GIR_BuildRootMI, /*Opcode*/GIMT_Encode2(MyTarget::XORI),
+// R02C-NEXT:    GIR_RootToRootCopy, /*OpIdx*/0, // DstI[dst]
 // R02C-NEXT:    GIR_AddImm8, /*InsnID*/0, /*Imm*/uint8_t(-1),
-// R02C-NEXT:    GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/1, // src1
-// R02C-NEXT:    GIR_ConstrainSelectedInstOperands, /*InsnID*/0,
-// R02C-NEXT:    GIR_EraseFromParent, /*InsnID*/0,
+// R02C-NEXT:    GIR_RootToRootCopy, /*OpIdx*/1, // src1
+// R02C-NEXT:    GIR_RootConstrainSelectedInstOperands,
 // R02C-NEXT:    // GIR_Coverage, 2,
-// R02C-NEXT:    GIR_Done,
+// R02C-NEXT:    GIR_EraseRootFromParent_Done,
 // R02C-NEXT:  // Label [[LABEL_NUM]]: @[[LABEL]]
 //
 // R02O:       // Label [[DEFAULT_NUM]]: @[[DEFAULT]]
@@ -626,23 +619,22 @@ def XORI : I<(outs GPR32:$dst), (ins m1:$src2, GPR32:$src1),
 // NOOPT-NEXT:    GIM_CheckNumOperands, /*MI*/0, /*Expected*/3,
 // NOOPT-NEXT:    GIM_CheckOpcode, /*MI*/0, GIMT_Encode2(TargetOpcode::G_XOR),
 // NOOPT-NEXT:    // MIs[0] DstI[dst]
-// NOOPT-NEXT:    GIM_CheckType, /*MI*/0, /*Op*/0, /*Type*/GILLT_s32,
-// NOOPT-NEXT:    GIM_CheckRegBankForClass, /*MI*/0, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
+// NOOPT-NEXT:    GIM_RootCheckType, /*Op*/0, /*Type*/GILLT_s32,
+// NOOPT-NEXT:    GIM_RootCheckRegBankForClass, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
 // NOOPT-NEXT:    // MIs[0] src1
-// NOOPT-NEXT:    GIM_CheckType, /*MI*/0, /*Op*/1, /*Type*/GILLT_s32,
-// NOOPT-NEXT:    GIM_CheckRegBankForClass, /*MI*/0, /*Op*/1, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
+// NOOPT-NEXT:    GIM_RootCheckType, /*Op*/1, /*Type*/GILLT_s32,
+// NOOPT-NEXT:    GIM_RootCheckRegBankForClass, /*Op*/1, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
 // NOOPT-NEXT:    // MIs[0] Operand 2
-// NOOPT-NEXT:    GIM_CheckType, /*MI*/0, /*Op*/2, /*Type*/GILLT_s32,
+// NOOPT-NEXT:    GIM_RootCheckType, /*Op*/2, /*Type*/GILLT_s32,
 // NOOPT-NEXT:    GIM_CheckConstantInt8, /*MI*/0, /*Op*/2, uint8_t(-3)
 // NOOPT-NEXT:    // (xor:{ *:[i32] } GPR32:{ *:[i32] }:$src1, -3:{ *:[i32] }) => (XOR:{ *:[i32] } GPR32:{ *:[i32] }:$src1)
-// NOOPT-NEXT:    GIR_BuildMI, /*InsnID*/0, /*Opcode*/GIMT_Encode2(MyTarget::XOR),
-// NOOPT-NEXT:    GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/0, //  DstI[dst]
+// NOOPT-NEXT:    GIR_BuildRootMI, /*Opcode*/GIMT_Encode2(MyTarget::XOR),
+// NOOPT-NEXT:    GIR_RootToRootCopy, /*OpIdx*/0, //  DstI[dst]
 // NOOPT-NEXT:    GIR_AddRegister, /*InsnID*/0, GIMT_Encode2(MyTarget::R0),
-// NOOPT-NEXT:    GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/1, // src1
-// NOOPT-NEXT:    GIR_ConstrainSelectedInstOperands, /*InsnID*/0,
-// NOOPT-NEXT:    GIR_EraseFromParent, /*InsnID*/0,
+// NOOPT-NEXT:    GIR_RootToRootCopy, /*OpIdx*/1, // src1
+// NOOPT-NEXT:    GIR_RootConstrainSelectedInstOperands,
 // NOOPT-NEXT:    // GIR_Coverage, 3,
-// NOOPT-NEXT:    GIR_Done,
+// NOOPT-NEXT:    GIR_EraseRootFromParent_Done,
 // NOOPT-NEXT:  // Label [[LABEL_NUM]]: @[[LABEL]]
 
 // The -3 is just to distinguish it from the 'not' case below and the other default op case above.
@@ -655,24 +647,23 @@ def XOR : I<(outs GPR32:$dst), (ins Z:$src2, GPR32:$src1),
 // NOOPT-NEXT:    GIM_CheckNumOperands, /*MI*/0, /*Expected*/3,
 // NOOPT-NEXT:    GIM_CheckOpcode, /*MI*/0, GIMT_Encode2(TargetOpcode::G_XOR),
 // NOOPT-NEXT:    // MIs[0] DstI[dst]
-// NOOPT-NEXT:    GIM_CheckType, /*MI*/0, /*Op*/0, /*Type*/GILLT_s32,
-// NOOPT-NEXT:    GIM_CheckRegBankForClass, /*MI*/0, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
+// NOOPT-NEXT:    GIM_RootCheckType, /*Op*/0, /*Type*/GILLT_s32,
+// NOOPT-NEXT:    GIM_RootCheckRegBankForClass, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
 // NOOPT-NEXT:    // MIs[0] src1
-// NOOPT-NEXT:    GIM_CheckType, /*MI*/0, /*Op*/1, /*Type*/GILLT_s32,
-// NOOPT-NEXT:    GIM_CheckRegBankForClass, /*MI*/0, /*Op*/1, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
+// NOOPT-NEXT:    GIM_RootCheckType, /*Op*/1, /*Type*/GILLT_s32,
+// NOOPT-NEXT:    GIM_RootCheckRegBankForClass, /*Op*/1, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
 // NOOPT-NEXT:    // MIs[0] Operand 2
-// NOOPT-NEXT:    GIM_CheckType, /*MI*/0, /*Op*/2, /*Type*/GILLT_s32,
+// NOOPT-NEXT:    GIM_RootCheckType, /*Op*/2, /*Type*/GILLT_s32,
 // NOOPT-NEXT:    GIM_CheckConstantInt8, /*MI*/0, /*Op*/2, uint8_t(-4)
 // NOOPT-NEXT:    // (xor:{ *:[i32] } GPR32:{ *:[i32] }:$src1, -4:{ *:[i32] }) => (XORlike:{ *:[i32] } GPR32:{ *:[i32] }:$src1)
-// NOOPT-NEXT:    GIR_BuildMI, /*InsnID*/0, /*Opcode*/GIMT_Encode2(MyTarget::XORlike),
-// NOOPT-NEXT:    GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/0, // DstI[dst]
+// NOOPT-NEXT:    GIR_BuildRootMI, /*Opcode*/GIMT_Encode2(MyTarget::XORlike),
+// NOOPT-NEXT:    GIR_RootToRootCopy, /*OpIdx*/0, // DstI[dst]
 // NOOPT-NEXT:    GIR_AddImm8, /*InsnID*/0, /*Imm*/uint8_t(-1),
 // NOOPT-NEXT:    GIR_AddRegister, /*InsnID*/0, GIMT_Encode2(MyTarget::R0),
-// NOOPT-NEXT:    GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/1, // src1
-// NOOPT-NEXT:    GIR_ConstrainSelectedInstOperands, /*InsnID*/0,
-// NOOPT-NEXT:    GIR_EraseFromParent, /*InsnID*/0,
+// NOOPT-NEXT:    GIR_RootToRootCopy, /*OpIdx*/1, // src1
+// NOOPT-NEXT:    GIR_RootConstrainSelectedInstOperands,
 // NOOPT-NEXT:    // GIR_Coverage, 4,
-// NOOPT-NEXT:    GIR_Done,
+// NOOPT-NEXT:    GIR_EraseRootFromParent_Done,
 // NOOPT-NEXT:  // Label [[LABEL_NUM]]: @[[LABEL]]
 
 // The -4 is just to distinguish it from the other 'not' cases.
@@ -685,25 +676,24 @@ def XORlike : I<(outs GPR32:$dst), (ins m1Z:$src2, GPR32:$src1),
 // NOOPT-NEXT:    GIM_CheckNumOperands, /*MI*/0, /*Expected*/3,
 // NOOPT-NEXT:    GIM_CheckOpcode, /*MI*/0, GIMT_Encode2(TargetOpcode::G_XOR),
 // NOOPT-NEXT:    // MIs[0] DstI[dst]
-// NOOPT-NEXT:    GIM_CheckType, /*MI*/0, /*Op*/0, /*Type*/GILLT_s32,
-// NOOPT-NEXT:    GIM_CheckRegBankForClass, /*MI*/0, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
+// NOOPT-NEXT:    GIM_RootCheckType, /*Op*/0, /*Type*/GILLT_s32,
+// NOOPT-NEXT:    GIM_RootCheckRegBankForClass, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
 // NOOPT-NEXT:    // MIs[0] src1
-// NOOPT-NEXT:    GIM_CheckType, /*MI*/0, /*Op*/1, /*Type*/GILLT_s32,
-// NOOPT-NEXT:    GIM_CheckRegBankForClass, /*MI*/0, /*Op*/1, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
+// NOOPT-NEXT:    GIM_RootCheckType, /*Op*/1, /*Type*/GILLT_s32,
+// NOOPT-NEXT:    GIM_RootCheckRegBankForClass, /*Op*/1, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
 // NOOPT-NEXT:    // MIs[0] Operand 2
-// NOOPT-NEXT:    GIM_CheckType, /*MI*/0, /*Op*/2, /*Type*/GILLT_s32,
+// NOOPT-NEXT:    GIM_RootCheckType, /*Op*/2, /*Type*/GILLT_s32,
 // NOOPT-NEXT:    GIM_CheckConstantInt8, /*MI*/0, /*Op*/2, uint8_t(-5),
 // NOOPT-NEXT:    // (xor:{ *:[i32] } GPR32:{ *:[i32] }:$src1, -5:{ *:[i32] }) => (XORManyDefaults:{ *:[i32] } GPR32:{ *:[i32] }:$src1)
-// NOOPT-NEXT:    GIR_BuildMI, /*InsnID*/0, /*Opcode*/GIMT_Encode2(MyTarget::XORManyDefaults),
-// NOOPT-NEXT:    GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/0, // DstI[dst]
+// NOOPT-NEXT:    GIR_BuildRootMI, /*Opcode*/GIMT_Encode2(MyTarget::XORManyDefaults),
+// NOOPT-NEXT:    GIR_RootToRootCopy, /*OpIdx*/0, // DstI[dst]
 // NOOPT-NEXT:    GIR_AddImm8, /*InsnID*/0, /*Imm*/uint8_t(-1),
 // NOOPT-NEXT:    GIR_AddRegister, /*InsnID*/0, GIMT_Encode2(MyTarget::R0),
 // NOOPT-NEXT:    GIR_AddRegister, /*InsnID*/0, GIMT_Encode2(MyTarget::R0),
-// NOOPT-NEXT:    GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/1, // src1
-// NOOPT-NEXT:    GIR_ConstrainSelectedInstOperands, /*InsnID*/0,
-// NOOPT-NEXT:    GIR_EraseFromParent, /*InsnID*/0,
+// NOOPT-NEXT:    GIR_RootToRootCopy, /*OpIdx*/1, // src1
+// NOOPT-NEXT:    GIR_RootConstrainSelectedInstOperands,
 // NOOPT-NEXT:    // GIR_Coverage, 5,
-// NOOPT-NEXT:    GIR_Done,
+// NOOPT-NEXT:    GIR_EraseRootFromParent_Done,
 // NOOPT-NEXT:  // Label [[LABEL_NUM]]: @[[LABEL]]
 
 // The -5 is just to distinguish it from the other cases.
@@ -716,23 +706,22 @@ def XORManyDefaults : I<(outs GPR32:$dst), (ins m1Z:$src3, Z:$src2, GPR32:$src1)
 // NOOPT-NEXT:    GIM_CheckNumOperands, /*MI*/0, /*Expected*/3,
 // NOOPT-NEXT:    GIM_CheckOpcode, /*MI*/0, GIMT_Encode2(TargetOpcode::G_XOR),
 // NOOPT-NEXT:    // MIs[0] DstI[dst]
-// NOOPT-NEXT:    GIM_CheckType, /*MI*/0, /*Op*/0, /*Type*/GILLT_s32,
-// NOOPT-NEXT:    GIM_CheckRegBankForClass, /*MI*/0, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
+// NOOPT-NEXT:    GIM_RootCheckType, /*Op*/0, /*Type*/GILLT_s32,
+// NOOPT-NEXT:    GIM_RootCheckRegBankForClass, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
 // NOOPT-NEXT:    // MIs[0] src1
-// NOOPT-NEXT:    GIM_CheckType, /*MI*/0, /*Op*/1, /*Type*/GILLT_s32,
-// NOOPT-NEXT:    GIM_CheckRegBankForClass, /*MI*/0, /*Op*/1, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
+// NOOPT-NEXT:    GIM_RootCheckType, /*Op*/1, /*Type*/GILLT_s32,
+// NOOPT-NEXT:    GIM_RootCheckRegBankForClass, /*Op*/1, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
 // NOOPT-NEXT:    // MIs[0] Operand 2
-// NOOPT-NEXT:    GIM_CheckType, /*MI*/0, /*Op*/2, /*Type*/GILLT_s32,
+// NOOPT-NEXT:    GIM_RootCheckType, /*Op*/2, /*Type*/GILLT_s32,
 // NOOPT-NEXT:    GIM_CheckConstantInt8, /*MI*/0, /*Op*/2, uint8_t(-6)
 // NOOPT-NEXT:    // (xor:{ *:[i32] } GPR32:{ *:[i32] }:$src1, -6:{ *:[i32] }) => (XORIb:{ *:[i32] } GPR32:{ *:[i32] }:$src1)
-// NOOPT-NEXT:    GIR_BuildMI, /*InsnID*/0, /*Opcode*/GIMT_Encode2(MyTarget::XORIb),
-// NOOPT-NEXT:    GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/0, // DstI[dst]
+// NOOPT-NEXT:    GIR_BuildRootMI, /*Opcode*/GIMT_Encode2(MyTarget::XORIb),
+// NOOPT-NEXT:    GIR_RootToRootCopy, /*OpIdx*/0, // DstI[dst]
 // NOOPT-NEXT:    GIR_AddImm8, /*InsnID*/0, /*Imm*/13,
-// NOOPT-NEXT:    GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/1, // src1
-// NOOPT-NEXT:    GIR_ConstrainSelectedInstOperands, /*InsnID*/0,
-// NOOPT-NEXT:    GIR_EraseFromParent, /*InsnID*/0,
+// NOOPT-NEXT:    GIR_RootToRootCopy, /*OpIdx*/1, // src1
+// NOOPT-NEXT:    GIR_RootConstrainSelectedInstOperands,
 // NOOPT-NEXT:    // GIR_Coverage, 6,
-// NOOPT-NEXT:    GIR_Done,
+// NOOPT-NEXT:    GIR_EraseRootFromParent_Done,
 // NOOPT-NEXT:  // Label [[LABEL_NUM]]: @[[LABEL]]
 
 // The -6 is just to distinguish it from the other cases.
@@ -748,23 +737,22 @@ def XORIb : I<(outs GPR32:$dst), (ins mb:$src2, GPR32:$src1),
 // NOOPT-NEXT:    GIM_CheckNumOperands, /*MI*/0, /*Expected*/3,
 // NOOPT-NEXT:    GIM_CheckOpcode, /*MI*/0, GIMT_Encode2(TargetOpcode::G_XOR),
 // NOOPT-NEXT:    // MIs[0] DstI[dst]
-// NOOPT-NEXT:    GIM_CheckType, /*MI*/0, /*Op*/0, /*Type*/GILLT_s32,
-// NOOPT-NEXT:    GIM_CheckRegBankForClass, /*MI*/0, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
+// NOOPT-NEXT:    GIM_RootCheckType, /*Op*/0, /*Type*/GILLT_s32,
+// NOOPT-NEXT:    GIM_RootCheckRegBankForClass, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
 // NOOPT-NEXT:    // MIs[0] Wm
-// NOOPT-NEXT:    GIM_CheckType, /*MI*/0, /*Op*/1, /*Type*/GILLT_s32,
-// NOOPT-NEXT:    GIM_CheckRegBankForClass, /*MI*/0, /*Op*/1, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
+// NOOPT-NEXT:    GIM_RootCheckType, /*Op*/1, /*Type*/GILLT_s32,
+// NOOPT-NEXT:    GIM_RootCheckRegBankForClass, /*Op*/1, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
 // NOOPT-NEXT:    // MIs[0] Operand 2
-// NOOPT-NEXT:    GIM_CheckType, /*MI*/0, /*Op*/2, /*Type*/GILLT_s32,
+// NOOPT-NEXT:    GIM_RootCheckType, /*Op*/2, /*Type*/GILLT_s32,
 // NOOPT-NEXT:    GIM_CheckConstantInt8, /*MI*/0, /*Op*/2, uint8_t(-1),
 // NOOPT-NEXT:    // (xor:{ *:[i32] } GPR32:{ *:[i32] }:$Wm, -1:{ *:[i32] }) => (ORN:{ *:[i32] } R0:{ *:[i32] }, GPR32:{ *:[i32] }:$Wm)
-// NOOPT-NEXT:    GIR_BuildMI, /*InsnID*/0, /*Opcode*/GIMT_Encode2(MyTarget::ORN),
-// NOOPT-NEXT:    GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/0, //  DstI[dst]
+// NOOPT-NEXT:    GIR_BuildRootMI, /*Opcode*/GIMT_Encode2(MyTarget::ORN),
+// NOOPT-NEXT:    GIR_RootToRootCopy, /*OpIdx*/0, //  DstI[dst]
 // NOOPT-NEXT:    GIR_AddRegister, /*InsnID*/0, GIMT_Encode2(MyTarget::R0),
-// NOOPT-NEXT:    GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/1, // Wm
-// NOOPT-NEXT:    GIR_ConstrainSelectedInstOperands, /*InsnID*/0,
-// NOOPT-NEXT:    GIR_EraseFromParent, /*InsnID*/0,
+// NOOPT-NEXT:    GIR_RootToRootCopy, /*OpIdx*/1, // Wm
+// NOOPT-NEXT:    GIR_RootConstrainSelectedInstOperands,
 // NOOPT-NEXT:    // GIR_Coverage, 23,
-// NOOPT-NEXT:    GIR_Done,
+// NOOPT-NEXT:    GIR_EraseRootFromParent_Done,
 // NOOPT-NEXT:  // Label [[LABEL_NUM]]: @[[LABEL]]
 
 def ORN : I<(outs GPR32:$dst), (ins GPR32:$src1, GPR32:$src2), []>;
@@ -777,10 +765,10 @@ def : Pat<(not GPR32:$Wm), (ORN R0, GPR32:$Wm)>;
 // NOOPT-NEXT:    GIM_CheckNumOperands, /*MI*/0, /*Expected*/3,
 // NOOPT-NEXT:    GIM_CheckOpcode, /*MI*/0, GIMT_Encode2(TargetOpcode::G_MUL),
 // NOOPT-NEXT:    // MIs[0] DstI[dst]
-// NOOPT-NEXT:    GIM_CheckType, /*MI*/0, /*Op*/0, /*Type*/GILLT_s32,
-// NOOPT-NEXT:    GIM_CheckRegBankForClass, /*MI*/0, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
+// NOOPT-NEXT:    GIM_RootCheckType, /*Op*/0, /*Type*/GILLT_s32,
+// NOOPT-NEXT:    GIM_RootCheckRegBankForClass, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
 // NOOPT-NEXT:    // MIs[0] Operand 1
-// NOOPT-NEXT:    GIM_CheckType, /*MI*/0, /*Op*/1, /*Type*/GILLT_s32,
+// NOOPT-NEXT:    GIM_RootCheckType, /*Op*/1, /*Type*/GILLT_s32,
 // NOOPT-NEXT:    GIM_RecordInsn, /*DefineMI*/1, /*MI*/0, /*OpIdx*/1, // MIs[1]
 // NOOPT-NEXT:    GIM_CheckNumOperands, /*MI*/1, /*Expected*/3,
 // NOOPT-NEXT:    GIM_CheckOpcode, /*MI*/1, GIMT_Encode2(TargetOpcode::G_ADD),
@@ -793,19 +781,18 @@ def : Pat<(not GPR32:$Wm), (ORN R0, GPR32:$Wm)>;
 // NOOPT-NEXT:    GIM_CheckType, /*MI*/1, /*Op*/2, /*Type*/GILLT_s32,
 // NOOPT-NEXT:    GIM_CheckRegBankForClass, /*MI*/1, /*Op*/2, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
 // NOOPT-NEXT:    // MIs[0] src3
-// NOOPT-NEXT:    GIM_CheckType, /*MI*/0, /*Op*/2, /*Type*/GILLT_s32,
-// NOOPT-NEXT:    GIM_CheckRegBankForClass, /*MI*/0, /*Op*/2, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
-// NOOPT-NEXT:    GIM_CheckIsSafeToFold, /*InsnID*/1,
+// NOOPT-NEXT:    GIM_RootCheckType, /*Op*/2, /*Type*/GILLT_s32,
+// NOOPT-NEXT:    GIM_RootCheckRegBankForClass, /*Op*/2, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
+// NOOPT-NEXT:    GIM_CheckIsSafeToFold, /*NumInsns*/1,
 // NOOPT-NEXT:    // (mul:{ *:[i32] } (add:{ *:[i32] } GPR32:{ *:[i32] }:$src1, GPR32:{ *:[i32] }:$src2), GPR32:{ *:[i32] }:$src3)  =>  (MULADD:{ *:[i32] } GPR32:{ *:[i32] }:$src1, GPR32:{ *:[i32] }:$src2, GPR32:{ *:[i32] }:$src3)
-// NOOPT-NEXT:    GIR_BuildMI, /*InsnID*/0, /*Opcode*/GIMT_Encode2(MyTarget::MULADD),
-// NOOPT-NEXT:    GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/0, //  DstI[dst]
+// NOOPT-NEXT:    GIR_BuildRootMI, /*Opcode*/GIMT_Encode2(MyTarget::MULADD),
+// NOOPT-NEXT:    GIR_RootToRootCopy, /*OpIdx*/0, //  DstI[dst]
 // NOOPT-NEXT:    GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/1, /*OpIdx*/1, // src1
 // NOOPT-NEXT:    GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/1, /*OpIdx*/2, // src2
-// NOOPT-NEXT:    GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/2, // src3
-// NOOPT-NEXT:    GIR_ConstrainSelectedInstOperands, /*InsnID*/0,
-// NOOPT-NEXT:    GIR_EraseFromParent, /*InsnID*/0,
+// NOOPT-NEXT:    GIR_RootToRootCopy, /*OpIdx*/2, // src3
+// NOOPT-NEXT:    GIR_RootConstrainSelectedInstOperands,
 // NOOPT-NEXT:    // GIR_Coverage, 7,
-// NOOPT-NEXT:    GIR_Done,
+// NOOPT-NEXT:    GIR_EraseRootFromParent_Done,
 // NOOPT-NEXT:  // Label [[LABEL_NUM]]: @[[LABEL]]
 
 // We also get a second rule by commutativity.
@@ -815,13 +802,13 @@ def : Pat<(not GPR32:$Wm), (ORN R0, GPR32:$Wm)>;
 // NOOPT-NEXT:    GIM_CheckNumOperands, /*MI*/0, /*Expected*/3,
 // NOOPT-NEXT:    GIM_CheckOpcode, /*MI*/0, GIMT_Encode2(TargetOpcode::G_MUL),
 // NOOPT-NEXT:    // MIs[0] DstI[dst]
-// NOOPT-NEXT:    GIM_CheckType, /*MI*/0, /*Op*/0, /*Type*/GILLT_s32,
-// NOOPT-NEXT:    GIM_CheckRegBankForClass, /*MI*/0, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
+// NOOPT-NEXT:    GIM_RootCheckType, /*Op*/0, /*Type*/GILLT_s32,
+// NOOPT-NEXT:    GIM_RootCheckRegBankForClass, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
 // NOOPT-NEXT:    // MIs[0] src3
-// NOOPT-NEXT:    GIM_CheckType, /*MI*/0, /*Op*/1, /*Type*/GILLT_s32,
-// NOOPT-NEXT:    GIM_CheckRegBankForClass, /*MI*/0, /*Op*/1, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
+// NOOPT-NEXT:    GIM_RootCheckType, /*Op*/1, /*Type*/GILLT_s32,
+// NOOPT-NEXT:    GIM_RootCheckRegBankForClass, /*Op*/1, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
 // NOOPT-NEXT:    // MIs[0] Operand 2
-// NOOPT-NEXT:    GIM_CheckType, /*MI*/0, /*Op*/2, /*Type*/GILLT_s32,
+// NOOPT-NEXT:    GIM_RootCheckType, /*Op*/2, /*Type*/GILLT_s32,
 // NOOPT-NEXT:    GIM_RecordInsn, /*DefineMI*/1, /*MI*/0, /*OpIdx*/2,
 // NOOPT-NEXT:    GIM_CheckNumOperands, /*MI*/1, /*Expected*/3,
 // NOOPT-NEXT:    GIM_CheckOpcode, /*MI*/1, GIMT_Encode2(TargetOpcode::G_ADD),
@@ -833,17 +820,16 @@ def : Pat<(not GPR32:$Wm), (ORN R0, GPR32:$Wm)>;
 // NOOPT-NEXT:    // MIs[1] src2
 // NOOPT-NEXT:    GIM_CheckType, /*MI*/1, /*Op*/2, /*Type*/GILLT_s32,
 // NOOPT-NEXT:    GIM_CheckRegBankForClass, /*MI*/1, /*Op*/2, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
-// NOOPT-NEXT:    GIM_CheckIsSafeToFold, /*InsnID*/1,
+// NOOPT-NEXT:    GIM_CheckIsSafeToFold, /*NumInsns*/1,
 // NOOPT-NEXT:    // (mul:{ *:[i32] } GPR32:{ *:[i32] }:$src3, (add:{ *:[i32] } GPR32:{ *:[i32] }:$src1, GPR32:{ *:[i32] }:$src2))  =>  (MULADD:{ *:[i32] } GPR32:{ *:[i32] }:$src1, GPR32:{ *:[i32] }:$src2, GPR32:{ *:[i32] }:$src3)
-// NOOPT-NEXT:    GIR_BuildMI, /*InsnID*/0, /*Opcode*/GIMT_Encode2(MyTarget::MULADD),
-// NOOPT-NEXT:    GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/0, //  DstI[dst]
+// NOOPT-NEXT:    GIR_BuildRootMI, /*Opcode*/GIMT_Encode2(MyTarget::MULADD),
+// NOOPT-NEXT:    GIR_RootToRootCopy, /*OpIdx*/0, //  DstI[dst]
 // NOOPT-NEXT:    GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/1, /*OpIdx*/1, // src1
 // NOOPT-NEXT:    GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/1, /*OpIdx*/2, // src2
-// NOOPT-NEXT:    GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/1, // src3
-// NOOPT-NEXT:    GIR_ConstrainSelectedInstOperands, /*InsnID*/0,
-// NOOPT-NEXT:    GIR_EraseFromParent, /*InsnID*/0,
+// NOOPT-NEXT:    GIR_RootToRootCopy, /*OpIdx*/1, // src3
+// NOOPT-NEXT:    GIR_RootConstrainSelectedInstOperands,
 // NOOPT-NEXT:    // GIR_Coverage, 28,
-// NOOPT-NEXT:    GIR_Done,
+// NOOPT-NEXT:    GIR_EraseRootFromParent_Done,
 // NOOPT-NEXT:  // Label [[LABEL_NUM]]: @[[LABEL]]
 
 def MULADD : I<(outs GPR32:$dst), (ins GPR32:$src1, GPR32:$src2, GPR32:$src3),
@@ -857,17 +843,16 @@ def MULADD : I<(outs GPR32:$dst), (ins GPR32:$src1, GPR32:$src2, GPR32:$src3),
 // NOOPT-NEXT:    GIM_CheckNumOperands, /*MI*/0, /*Expected*/2,
 // NOOPT-NEXT:    GIM_CheckOpcode, /*MI*/0, GIMT_Encode2(TargetOpcode::G_CONSTANT),
 // NOOPT-NEXT:    // MIs[0] DstI[dst]
-// NOOPT-NEXT:    GIM_CheckType, /*MI*/0, /*Op*/0, /*Type*/GILLT_s32,
-// NOOPT-NEXT:    GIM_CheckRegBankForClass, /*MI*/0, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
+// NOOPT-NEXT:    GIM_RootCheckType, /*Op*/0, /*Type*/GILLT_s32,
+// NOOPT-NEXT:    GIM_RootCheckRegBankForClass, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
 // NOOPT-NEXT:    // MIs[0] Operand 1
 // NOOPT-NEXT:    GIM_CheckLiteralInt, /*MI*/0, /*Op*/1, GIMT_Encode8(1),
 // NOOPT-NEXT:    // 1:{ *:[i32] }  =>  (MOV1:{ *:[i32] })
-// NOOPT-NEXT:    GIR_BuildMI, /*InsnID*/0, /*Opcode*/GIMT_Encode2(MyTarget::MOV1),
-// NOOPT-NEXT:    GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/0, //  DstI[dst]
-// NOOPT-NEXT:    GIR_ConstrainSelectedInstOperands, /*InsnID*/0,
-// NOOPT-NEXT:    GIR_EraseFromParent, /*InsnID*/0,
+// NOOPT-NEXT:    GIR_BuildRootMI, /*Opcode*/GIMT_Encode2(MyTarget::MOV1),
+// NOOPT-NEXT:    GIR_RootToRootCopy, /*OpIdx*/0, //  DstI[dst]
+// NOOPT-NEXT:    GIR_RootConstrainSelectedInstOperands,
 // NOOPT-NEXT:    // GIR_Coverage, 8,
-// NOOPT-NEXT:    GIR_Done,
+// NOOPT-NEXT:    GIR_EraseRootFromParent_Done,
 // NOOPT-NEXT:  // Label [[LABEL_NUM]]: @[[LABEL]]
 
 def MOV1 : I<(outs GPR32:$dst), (ins), [(set GPR32:$dst, 1)]>;
@@ -879,18 +864,17 @@ def MOV1 : I<(outs GPR32:$dst), (ins), [(set GPR32:$dst, 1)]>;
 // NOOPT-NEXT:    GIM_CheckOpcode, /*MI*/0, GIMT_Encode2(TargetOpcode::G_CONSTANT),
 // NOOPT-NEXT:    GIM_CheckI64ImmPredicate, /*MI*/0, /*Predicate*/GIMT_Encode2(GICXXPred_I64_Predicate_simm8),
 // NOOPT-NEXT:    // MIs[0] DstI[dst]
-// NOOPT-NEXT:    GIM_CheckType, /*MI*/0, /*Op*/0, /*Type*/GILLT_s32,
-// NOOPT-NEXT:    GIM_CheckRegBankForClass, /*MI*/0, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
+// NOOPT-NEXT:    GIM_RootCheckType, /*Op*/0, /*Type*/GILLT_s32,
+// NOOPT-NEXT:    GIM_RootCheckRegBankForClass, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
 // NOOPT-NEXT:    // MIs[0] Operand 1
 // NOOPT-NEXT:    // No operand predicates
 // NOOPT-NEXT:    // (imm:{ *:[i32] })<<P:Predicate_simm8>>:$imm => (MOVimm8:{ *:[i32] } (imm:{ *:[i32] }):$imm)
-// NOOPT-NEXT:    GIR_BuildMI, /*InsnID*/0, /*Opcode*/GIMT_Encode2(MyTarget::MOVimm8),
-// NOOPT-NEXT:    GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/0, //  DstI[dst]
+// NOOPT-NEXT:    GIR_BuildRootMI, /*Opcode*/GIMT_Encode2(MyTarget::MOVimm8),
+// NOOPT-NEXT:    GIR_RootToRootCopy, /*OpIdx*/0, //  DstI[dst]
 // NOOPT-NEXT:    GIR_CopyConstantAsSImm, /*NewInsnID*/0, /*OldInsnID*/0, // imm
-// NOOPT-NEXT:    GIR_ConstrainSelectedInstOperands, /*InsnID*/0,
-// NOOPT-NEXT:    GIR_EraseFromParent, /*InsnID*/0,
+// NOOPT-NEXT:    GIR_RootConstrainSelectedInstOperands,
 // NOOPT-NEXT:    // GIR_Coverage, 9,
-// NOOPT-NEXT:    GIR_Done,
+// NOOPT-NEXT:    GIR_EraseRootFromParent_Done,
 // NOOPT-NEXT:  // Label [[LABEL_NUM]]: @[[LABEL]]
 
 def simm8 : ImmLeaf<i32, [{ return isInt<8>(Imm); }]>;
@@ -903,18 +887,17 @@ def MOVimm8 : I<(outs GPR32:$dst), (ins i32imm:$imm), [(set GPR32:$dst, simm8:$i
 // NOOPT-NEXT:    GIM_CheckOpcode, /*MI*/0, GIMT_Encode2(TargetOpcode::G_CONSTANT),
 // NOOPT-NEXT:    GIM_CheckAPIntImmPredicate, /*MI*/0, /*Predicate*/GIMT_Encode2(GICXXPred_APInt_Predicate_simm9),
 // NOOPT-NEXT:    // MIs[0] DstI[dst]
-// NOOPT-NEXT:    GIM_CheckType, /*MI*/0, /*Op*/0, /*Type*/GILLT_s32,
-// NOOPT-NEXT:    GIM_CheckRegBankForClass, /*MI*/0, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
+// NOOPT-NEXT:    GIM_RootCheckType, /*Op*/0, /*Type*/GILLT_s32,
+// NOOPT-NEXT:    GIM_RootCheckRegBankForClass, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
 // NOOPT-NEXT:    // MIs[0] Operand 1
 // NOOPT-NEXT:    // No operand predicates
 // NOOPT-NEXT:    // (imm:{ *:[i32] })<<P:Predicate_simm9>>:$imm =>  (MOVimm9:{ *:[i32] } (imm:{ *:[i32] }):$imm)
-// NOOPT-NEXT:    GIR_BuildMI, /*InsnID*/0, /*Opcode*/GIMT_Encode2(MyTarget::MOVimm9),
-// NOOPT-NEXT:    GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/0, //  DstI[dst]
+// NOOPT-NEXT:    GIR_BuildRootMI, /*Opcode*/GIMT_Encode2(MyTarget::MOVimm9),
+// NOOPT-NEXT:    GIR_RootToRootCopy, /*OpIdx*/0, //  DstI[dst]
 // NOOPT-NEXT:    GIR_CopyConstantAsSImm, /*NewInsnID*/0, /*OldInsnID*/0, // imm
-// NOOPT-NEXT:    GIR_ConstrainSelectedInstOperands, /*InsnID*/0,
-// NOOPT-NEXT:    GIR_EraseFromParent, /*InsnID*/0,
+// NOOPT-NEXT:    GIR_RootConstrainSelectedInstOperands,
 // NOOPT-NEXT:    // GIR_Coverage, 10,
-// NOOPT-NEXT:    GIR_Done,
+// NOOPT-NEXT:    GIR_EraseRootFromParent_Done,
 // NOOPT-NEXT:  // Label [[LABEL_NUM]]: @[[LABEL]]
 
 def simm9 : IntImmLeaf<i32, [{ return isInt<9>(Imm->getSExtValue()); }]>;
@@ -927,18 +910,17 @@ def MOVimm9 : I<(outs GPR32:$dst), (ins i32imm:$imm), [(set GPR32:$dst, simm9:$i
 // NOOPT-NEXT:    GIM_CheckOpcode, /*MI*/0, GIMT_Encode2(TargetOpcode::G_CONSTANT),
 // NOOPT-NEXT:    GIM_CheckI64ImmPredicate, /*MI*/0, /*Predicate*/GIMT_Encode2(GICXXPred_I64_Predicate_cimm8),
 // NOOPT-NEXT:    // MIs[0] DstI[dst]
-// NOOPT-NEXT:    GIM_CheckType, /*MI*/0, /*Op*/0, /*Type*/GILLT_s32,
-// NOOPT-NEXT:    GIM_CheckRegBankForClass, /*MI*/0, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
+// NOOPT-NEXT:    GIM_RootCheckType, /*Op*/0, /*Type*/GILLT_s32,
+// NOOPT-NEXT:    GIM_RootCheckRegBankForClass, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
 // NOOPT-NEXT:    // MIs[0] Operand 1
 // NOOPT-NEXT:    // No operand predicates
 // NOOPT-NEXT:    // (imm:{ *:[i32] })<<P:Predicate_cimm8>><<X:cimm8_xform>>:$imm  =>  (MOVcimm8:{ *:[i32] } (cimm8_xform:{ *:[i32] } (imm:{ *:[i32] }):$imm))
-// NOOPT-NEXT:    GIR_BuildMI, /*InsnID*/0, /*Opcode*/GIMT_Encode2(MyTarget::MOVcimm8),
-// NOOPT-NEXT:    GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/0, //  DstI[dst]
+// NOOPT-NEXT:    GIR_BuildRootMI, /*Opcode*/GIMT_Encode2(MyTarget::MOVcimm8),
+// NOOPT-NEXT:    GIR_RootToRootCopy, /*OpIdx*/0, //  DstI[dst]
 // NOOPT-NEXT:    GIR_CustomRenderer, /*InsnID*/0, /*OldInsnID*/0, /*Renderer*/GIMT_Encode2(GICR_renderImm), // imm
-// NOOPT-NEXT:    GIR_ConstrainSelectedInstOperands, /*InsnID*/0,
-// NOOPT-NEXT:    GIR_EraseFromParent, /*InsnID*/0,
+// NOOPT-NEXT:    GIR_RootConstrainSelectedInstOperands,
 // NOOPT-NEXT:    // GIR_Coverage, 11,
-// NOOPT-NEXT:    GIR_Done,
+// NOOPT-NEXT:    GIR_EraseRootFromParent_Done,
 // NOOPT-NEXT:  // Label [[LABEL_NUM]]: @[[LABEL]]
 
 def MOVcimm8 : I<(outs GPR32:$dst), (ins i32imm:$imm), [(set GPR32:$dst, cimm8:$imm)]>;
@@ -950,18 +932,17 @@ def MOVcimm8 : I<(outs GPR32:$dst), (ins i32imm:$imm), [(set GPR32:$dst, cimm8:$
 // NOOPT-NEXT:    GIM_CheckOpcode, /*MI*/0, GIMT_Encode2(TargetOpcode::G_FCONSTANT),
 // NOOPT-NEXT:    GIM_CheckAPFloatImmPredicate, /*MI*/0, /*Predicate*/GIMT_Encode2(GICXXPred_APFloat_Predicate_fpimmz),
 // NOOPT-NEXT:    // MIs[0] DstI[dst]
-// NOOPT-NEXT:    GIM_CheckType, /*MI*/0, /*Op*/0, /*Type*/GILLT_s32,
-// NOOPT-NEXT:    GIM_CheckRegBankForClass, /*MI*/0, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::FPR32RegClassID),
+// NOOPT-NEXT:    GIM_RootCheckType, /*Op*/0, /*Type*/GILLT_s32,
+// NOOPT-NEXT:    GIM_RootCheckRegBankForClass, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::FPR32RegClassID),
 // NOOPT-NEXT:    // MIs[0] Operand 1
 // NOOPT-NEXT:    // No operand predicates
 // NOOPT-NEXT:    // (fpimm:{ *:[f32] })<<P:Predicate_fpimmz>>:$imm =>  (MOVfpimmz:{ *:[f32] } (fpimm:{ *:[f32] }):$imm)
-// NOOPT-NEXT:    GIR_BuildMI, /*InsnID*/0, /*Opcode*/GIMT_Encode2(MyTarget::MOVfpimmz),
-// NOOPT-NEXT:    GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/0, //  DstI[dst]
+// NOOPT-NEXT:    GIR_BuildRootMI, /*Opcode*/GIMT_Encode2(MyTarget::MOVfpimmz),
+// NOOPT-NEXT:    GIR_RootToRootCopy, /*OpIdx*/0, //  DstI[dst]
 // NOOPT-NEXT:    GIR_CopyFConstantAsFPImm, /*NewInsnID*/0, /*OldInsnID*/0, // imm
-// NOOPT-NEXT:    GIR_ConstrainSelectedInstOperands, /*InsnID*/0,
-// NOOPT-NEXT:    GIR_EraseFromParent, /*InsnID*/0,
+// NOOPT-NEXT:    GIR_RootConstrainSelectedInstOperands,
 // NOOPT-NEXT:    // GIR_Coverage, 18,
-// NOOPT-NEXT:    GIR_Done,
+// NOOPT-NEXT:    GIR_EraseRootFromParent_Done,
 // NOOPT-NEXT:  // Label [[LABEL_NUM]]: @[[LABEL]]
 
 //===- Test a simple pattern with inferred pointer operands. ---------------===//
@@ -972,14 +953,14 @@ def MOVcimm8 : I<(outs GPR32:$dst), (ins i32imm:$imm), [(set GPR32:$dst, cimm8:$
 // NOOPT-NEXT:    GIM_CheckMemorySizeEqualToLLT, /*MI*/0, /*MMO*/0, /*OpIdx*/0,
 // NOOPT-NEXT:    GIM_CheckAtomicOrdering, /*MI*/0, /*Order*/(uint8_t)AtomicOrdering::NotAtomic,
 // NOOPT-NEXT:    // MIs[0] DstI[dst]
-// NOOPT-NEXT:    GIM_CheckType, /*MI*/0, /*Op*/0, /*Type*/GILLT_s32,
-// NOOPT-NEXT:    GIM_CheckRegBankForClass, /*MI*/0, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
+// NOOPT-NEXT:    GIM_RootCheckType, /*Op*/0, /*Type*/GILLT_s32,
+// NOOPT-NEXT:    GIM_RootCheckRegBankForClass, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
 // NOOPT-NEXT:    // MIs[0] src1
 // NOOPT-NEXT:    GIM_CheckPointerToAny, /*MI*/0, /*Op*/1, /*SizeInBits*/32,
-// NOOPT-NEXT:    GIM_CheckRegBankForClass, /*MI*/0, /*Op*/1, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
+// NOOPT-NEXT:    GIM_RootCheckRegBankForClass, /*Op*/1, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
 // NOOPT-NEXT:    // (ld:{ *:[i32] } GPR32:{ *:[i32] }:$src1)<<P:Predicate_unindexedload>><<P:Predicate_load>> => (LOAD:{ *:[i32] } GPR32:{ *:[i32] }:$src1)
 // NOOPT-NEXT:    GIR_MutateOpcode, /*InsnID*/0, /*RecycleInsnID*/0, /*Opcode*/GIMT_Encode2(MyTarget::LOAD),
-// NOOPT-NEXT:    GIR_ConstrainSelectedInstOperands, /*InsnID*/0,
+// NOOPT-NEXT:    GIR_RootConstrainSelectedInstOperands,
 // NOOPT-NEXT:    // GIR_Coverage, 12,
 // NOOPT-NEXT:    GIR_Done,
 // NOOPT-NEXT:  // Label [[LABEL_NUM]]: @[[LABEL]]
@@ -995,14 +976,14 @@ def LOAD : I<(outs GPR32:$dst), (ins GPR32:$src1),
 // NOOPT-NEXT:    GIM_CheckMemorySizeEqualToLLT, /*MI*/0, /*MMO*/0, /*OpIdx*/0,
 // NOOPT-NEXT:    GIM_CheckAtomicOrdering, /*MI*/0, /*Order*/(uint8_t)AtomicOrdering::NotAtomic,
 // NOOPT-NEXT:    // MIs[0] DstI[dst]
-// NOOPT-NEXT:    GIM_CheckType, /*MI*/0, /*Op*/0, /*Type*/GILLT_p0s32,
-// NOOPT-NEXT:    GIM_CheckRegBankForClass, /*MI*/0, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
+// NOOPT-NEXT:    GIM_RootCheckType, /*Op*/0, /*Type*/GILLT_p0s32,
+// NOOPT-NEXT:    GIM_RootCheckRegBankForClass, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
 // NOOPT-NEXT:    // MIs[0] src
 // NOOPT-NEXT:    GIM_CheckPointerToAny, /*MI*/0, /*Op*/1, /*SizeInBits*/32,
-// NOOPT-NEXT:    GIM_CheckRegBankForClass, /*MI*/0, /*Op*/1, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
+// NOOPT-NEXT:    GIM_RootCheckRegBankForClass, /*Op*/1, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
 // NOOPT-NEXT:    // (ld:{ *:[i32] } GPR32:{ *:[i32] }:$src)<<P:Predicate_unindexedload>><<P:Predicate_load>>  =>  (LOAD:{ *:[i32] } GPR32:{ *:[i32] }:$src)
 // NOOPT-NEXT:    GIR_MutateOpcode, /*InsnID*/0, /*RecycleInsnID*/0, /*Opcode*/GIMT_Encode2(MyTarget::LOAD),
-// NOOPT-NEXT:    GIR_ConstrainSelectedInstOperands, /*InsnID*/0,
+// NOOPT-NEXT:    GIR_RootConstrainSelectedInstOperands,
 // NOOPT-NEXT:    // GIR_Coverage, 24,
 // NOOPT-NEXT:    GIR_Done,
 // NOOPT-NEXT:  // Label [[LABEL_NUM]]: @[[LABEL]]
@@ -1018,14 +999,14 @@ def : Pat<(load GPR32:$src),
 // NOOPT-NEXT:    GIM_CheckMemorySizeEqualTo, /*MI*/0, /*MMO*/0, /*Size*/GIMT_Encode4(2),
 // NOOPT-NEXT:    GIM_CheckAtomicOrdering, /*MI*/0, /*Order*/(uint8_t)AtomicOrdering::NotAtomic,
 // NOOPT-NEXT:    // MIs[0] DstI[dst]
-// NOOPT-NEXT:    GIM_CheckType, /*MI*/0, /*Op*/0, /*Type*/GILLT_s32,
-// NOOPT-NEXT:    GIM_CheckRegBankForClass, /*MI*/0, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
+// NOOPT-NEXT:    GIM_RootCheckType, /*Op*/0, /*Type*/GILLT_s32,
+// NOOPT-NEXT:    GIM_RootCheckRegBankForClass, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
 // NOOPT-NEXT:    // MIs[0] src1
 // NOOPT-NEXT:    GIM_CheckPointerToAny, /*MI*/0, /*Op*/1, /*SizeInBits*/32,
-// NOOPT-NEXT:    GIM_CheckRegBankForClass, /*MI*/0, /*Op*/1, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
+// NOOPT-NEXT:    GIM_RootCheckRegBankForClass, /*Op*/1, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
 // NOOPT-NEXT:    // (ld:{ *:[i32] } GPR32:{ *:[i32] }:$src1)<<P:Predicate_unindexedload>><<P:Predicate_sextload>><<P:Predicate_sextloadi16>>  =>  (SEXTLOAD:{ *:[i32] } GPR32:{ *:[i32] }:$src1)
 // NOOPT-NEXT:    GIR_MutateOpcode, /*InsnID*/0, /*RecycleInsnID*/0, /*Opcode*/GIMT_Encode2(MyTarget::SEXTLOAD),
-// NOOPT-NEXT:    GIR_ConstrainSelectedInstOperands, /*InsnID*/0,
+// NOOPT-NEXT:    GIR_RootConstrainSelectedInstOperands,
 // NOOPT-NEXT:    // GIR_Coverage, 13,
 // NOOPT-NEXT:    GIR_Done,
 // NOOPT-NEXT:  // Label [[LABEL_NUM]]: @[[LABEL]]
@@ -1039,17 +1020,17 @@ def SEXTLOAD : I<(outs GPR32:$dst), (ins GPR32:$src1),
 // NOOPT-NEXT:    GIM_CheckNumOperands, /*MI*/0, /*Expected*/3,
 // NOOPT-NEXT:    GIM_CheckOpcode, /*MI*/0, GIMT_Encode2(TargetOpcode::G_ADD),
 // NOOPT-NEXT:    // MIs[0] DstI[dst]
-// NOOPT-NEXT:    GIM_CheckType, /*MI*/0, /*Op*/0, /*Type*/GILLT_s32,
-// NOOPT-NEXT:    GIM_CheckRegBankForClass, /*MI*/0, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
+// NOOPT-NEXT:    GIM_RootCheckType, /*Op*/0, /*Type*/GILLT_s32,
+// NOOPT-NEXT:    GIM_RootCheckRegBankForClass, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
 // NOOPT-NEXT:    // MIs[0] src1
-// NOOPT-NEXT:    GIM_CheckType, /*MI*/0, /*Op*/1, /*Type*/GILLT_s32,
-// NOOPT-NEXT:    GIM_CheckRegBankForClass, /*MI*/0, /*Op*/1, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID)
+// NOOPT-NEXT:    GIM_RootCheckType, /*Op*/1, /*Type*/GILLT_s32,
+// NOOPT-NEXT:    GIM_RootCheckRegBankForClass, /*Op*/1, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID)
 // NOOPT-NEXT:    // MIs[0] src2
-// NOOPT-NEXT:    GIM_CheckType, /*MI*/0, /*Op*/2, /*Type*/GILLT_s32,
-// NOOPT-NEXT:    GIM_CheckRegBankForClass, /*MI*/0, /*Op*/2, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
+// NOOPT-NEXT:    GIM_RootCheckType, /*Op*/2, /*Type*/GILLT_s32,
+// NOOPT-NEXT:    GIM_RootCheckRegBankForClass, /*Op*/2, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
 // NOOPT-NEXT:    // (add:{ *:[i32] } GPR32:{ *:[i32] }:$src1, GPR32:{ *:[i32] }:$src2) => (ADD:{ *:[i32] } GPR32:{ *:[i32] }:$src1, GPR32:{ *:[i32] }:$src2)
 // NOOPT-NEXT:    GIR_MutateOpcode, /*InsnID*/0, /*RecycleInsnID*/0, /*Opcode*/GIMT_Encode2(MyTarget::ADD),
-// NOOPT-NEXT:    GIR_ConstrainSelectedInstOperands, /*InsnID*/0,
+// NOOPT-NEXT:    GIR_RootConstrainSelectedInstOperands,
 // NOOPT-NEXT:    // GIR_Coverage, 14,
 // NOOPT-NEXT:    GIR_Done,
 // NOOPT-NEXT:  // Label [[LABEL_NUM]]: @[[LABEL]]
@@ -1063,21 +1044,20 @@ def ADD : I<(outs GPR32:$dst), (ins GPR32:$src1, GPR32:$src2),
 // NOOPT-NEXT:    GIM_CheckNumOperands, /*MI*/0, /*Expected*/3,
 // NOOPT-NEXT:    GIM_CheckOpcode, /*MI*/0, GIMT_Encode2(TargetOpcode::G_ADD),
 // NOOPT-NEXT:    // MIs[0] DstI[dst]
-// NOOPT-NEXT:    GIM_CheckType, /*MI*/0, /*Op*/0, /*Type*/GILLT_s32,
-// NOOPT-NEXT:    GIM_CheckRegBankForClass, /*MI*/0, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
+// NOOPT-NEXT:    GIM_RootCheckType, /*Op*/0, /*Type*/GILLT_s32,
+// NOOPT-NEXT:    GIM_RootCheckRegBankForClass, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
 // NOOPT-NEXT:    // MIs[0] src{{$}}
-// NOOPT-NEXT:    GIM_CheckType, /*MI*/0, /*Op*/1, /*Type*/GILLT_s32,
-// NOOPT-NEXT:    GIM_CheckRegBankForClass, /*MI*/0, /*Op*/1, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
+// NOOPT-NEXT:    GIM_RootCheckType, /*Op*/1, /*Type*/GILLT_s32,
+// NOOPT-NEXT:    GIM_RootCheckRegBankForClass, /*Op*/1, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
 // NOOPT-NEXT:    // MIs[0] src{{$}}
 // NOOPT-NEXT:    GIM_CheckIsSameOperand, /*MI*/0, /*OpIdx*/2, /*OtherMI*/0, /*OtherOpIdx*/1,
 // NOOPT-NEXT:    // (add:{ *:[i32] } GPR32:{ *:[i32] }:$src, GPR32:{ *:[i32] }:$src) => (DOUBLE:{ *:[i32] } GPR32:{ *:[i32] }:$src)
-// NOOPT-NEXT:    GIR_BuildMI, /*InsnID*/0, /*Opcode*/GIMT_Encode2(MyTarget::DOUBLE),
-// NOOPT-NEXT:    GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/0, //  DstI[dst]
-// NOOPT-NEXT:    GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/1, // src
-// NOOPT-NEXT:    GIR_ConstrainSelectedInstOperands, /*InsnID*/0,
-// NOOPT-NEXT:    GIR_EraseFromParent, /*InsnID*/0,
+// NOOPT-NEXT:    GIR_BuildRootMI, /*Opcode*/GIMT_Encode2(MyTarget::DOUBLE),
+// NOOPT-NEXT:    GIR_RootToRootCopy, /*OpIdx*/0, //  DstI[dst]
+// NOOPT-NEXT:    GIR_RootToRootCopy, /*OpIdx*/1, // src
+// NOOPT-NEXT:    GIR_RootConstrainSelectedInstOperands,
 // NOOPT-NEXT:    // GIR_Coverage, 15,
-// NOOPT-NEXT:    GIR_Done,
+// NOOPT-NEXT:    GIR_EraseRootFromParent_Done,
 // NOOPT-NEXT:  // Label [[LABEL_NUM]]: @[[LABEL]]
 
 def DOUBLE : I<(outs GPR32:$dst), (ins GPR32:$src), [(set GPR32:$dst, (add GPR32:$src, GPR32:$src))]>;
@@ -1094,15 +1074,15 @@ def DOUBLE : I<(outs GPR32:$dst), (ins GPR32:$src), [(set GPR32:$dst, (add GPR32
 // NOOPT-NEXT:    GIM_CheckNumOperands, /*MI*/0, /*Expected*/3,
 // NOOPT-NEXT:    GIM_CheckOpcode, /*MI*/0, GIMT_Encode2(TargetOpcode::G_ADD),
 // NOOPT-NEXT:    // MIs[0] DstI[samename]
-// NOOPT-NEXT:    GIM_CheckType, /*MI*/0, /*Op*/0, /*Type*/GILLT_s32,
-// NOOPT-NEXT:    GIM_CheckRegBankForClass, /*MI*/0, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
+// NOOPT-NEXT:    GIM_RootCheckType, /*Op*/0, /*Type*/GILLT_s32,
+// NOOPT-NEXT:    GIM_RootCheckRegBankForClass, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
 // NOOPT-NEXT:    // MIs[0] samename
-// NOOPT-NEXT:    GIM_CheckType, /*MI*/0, /*Op*/1, /*Type*/GILLT_s32,
+// NOOPT-NEXT:    GIM_RootCheckType, /*Op*/1, /*Type*/GILLT_s32,
 // NOOPT-NEXT:    // MIs[0] othername
-// NOOPT-NEXT:    GIM_CheckType, /*MI*/0, /*Op*/2, /*Type*/GILLT_s32,
+// NOOPT-NEXT:    GIM_RootCheckType, /*Op*/2, /*Type*/GILLT_s32,
 // NOOPT-NEXT:    // (add:{ *:[i32] } i32:{ *:[i32] }:$samename, i32:{ *:[i32] }:$othername)  =>  (InsnWithSpeciallyNamedDef:{ *:[i32] } i32:{ *:[i32] }:$samename, i32:{ *:[i32] }:$othername)
 // NOOPT-NEXT:    GIR_MutateOpcode, /*InsnID*/0, /*RecycleInsnID*/0, /*Opcode*/GIMT_Encode2(MyTarget::InsnWithSpeciallyNamedDef),
-// NOOPT-NEXT:    GIR_ConstrainSelectedInstOperands, /*InsnID*/0,
+// NOOPT-NEXT:    GIR_RootConstrainSelectedInstOperands,
 // NOOPT-NEXT:    // GIR_Coverage, 25,
 // NOOPT-NEXT:    GIR_Done,
 // NOOPT-NEXT:  // Label [[LABEL_NUM]]: @[[LABEL]]
@@ -1117,15 +1097,15 @@ def : Pat<(add i32:$samename, i32:$othername),
 // NOOPT-NEXT:    GIM_CheckNumOperands, /*MI*/0, /*Expected*/3,
 // NOOPT-NEXT:    GIM_CheckOpcode, /*MI*/0, GIMT_Encode2(TargetOpcode::G_ADD),
 // NOOPT-NEXT:    // MIs[0] DstI[dst]
-// NOOPT-NEXT:    GIM_CheckType, /*MI*/0, /*Op*/0, /*Type*/GILLT_s32,
-// NOOPT-NEXT:    GIM_CheckRegBankForClass, /*MI*/0, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
+// NOOPT-NEXT:    GIM_RootCheckType, /*Op*/0, /*Type*/GILLT_s32,
+// NOOPT-NEXT:    GIM_RootCheckRegBankForClass, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
 // NOOPT-NEXT:    // MIs[0] src1
-// NOOPT-NEXT:    GIM_CheckType, /*MI*/0, /*Op*/1, /*Type*/GILLT_s32,
+// NOOPT-NEXT:    GIM_RootCheckType, /*Op*/1, /*Type*/GILLT_s32,
 // NOOPT-NEXT:    // MIs[0] src2
-// NOOPT-NEXT:    GIM_CheckType, /*MI*/0, /*Op*/2, /*Type*/GILLT_s32,
+// NOOPT-NEXT:    GIM_RootCheckType, /*Op*/2, /*Type*/GILLT_s32,
 // NOOPT-NEXT:    // (add:{ *:[i32] } i32:{ *:[i32] }:$src1, i32:{ *:[i32] }:$src2) => (ADD:{ *:[i32] } i32:{ *:[i32] }:$src1, i32:{ *:[i32] }:$src2)
 // NOOPT-NEXT:    GIR_MutateOpcode, /*InsnID*/0, /*RecycleInsnID*/0, /*Opcode*/GIMT_Encode2(MyTarget::ADD),
-// NOOPT-NEXT:    GIR_ConstrainSelectedInstOperands, /*InsnID*/0,
+// NOOPT-NEXT:    GIR_RootConstrainSelectedInstOperands,
 // NOOPT-NEXT:    // GIR_Coverage, 26,
 // NOOPT-NEXT:    GIR_Done,
 // NOOPT-NEXT:  // Label [[LABEL_NUM]]: @[[LABEL]]
@@ -1140,23 +1120,22 @@ def : Pat<(add i32:$src1, i32:$src2),
 // NOOPT-NEXT:    GIM_CheckNumOperands, /*MI*/0, /*Expected*/3,
 // NOOPT-NEXT:    GIM_CheckOpcode, /*MI*/0, GIMT_Encode2(TargetOpcode::G_MUL),
 // NOOPT-NEXT:    // MIs[0] DstI[dst]
-// NOOPT-NEXT:    GIM_CheckType, /*MI*/0, /*Op*/0, /*Type*/GILLT_s32,
-// NOOPT-NEXT:    GIM_CheckRegBankForClass, /*MI*/0, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
+// NOOPT-NEXT:    GIM_RootCheckType, /*Op*/0, /*Type*/GILLT_s32,
+// NOOPT-NEXT:    GIM_RootCheckRegBankForClass, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
 // NOOPT-NEXT:    // MIs[0] src1
-// NOOPT-NEXT:    GIM_CheckType, /*MI*/0, /*Op*/1, /*Type*/GILLT_s32,
-// NOOPT-NEXT:    GIM_CheckRegBankForClass, /*MI*/0, /*Op*/1, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
+// NOOPT-NEXT:    GIM_RootCheckType, /*Op*/1, /*Type*/GILLT_s32,
+// NOOPT-NEXT:    GIM_RootCheckRegBankForClass, /*Op*/1, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
 // NOOPT-NEXT:    // MIs[0] src2
-// NOOPT-NEXT:    GIM_CheckType, /*MI*/0, /*Op*/2, /*Type*/GILLT_s32,
-// NOOPT-NEXT:    GIM_CheckRegBankForClass, /*MI*/0, /*Op*/2, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
+// NOOPT-NEXT:    GIM_RootCheckType, /*Op*/2, /*Type*/GILLT_s32,
+// NOOPT-NEXT:    GIM_RootCheckRegBankForClass, /*Op*/2, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
 // NOOPT-NEXT:    // (mul:{ *:[i32] } GPR32:{ *:[i32] }:$src1, GPR32:{ *:[i32] }:$src2) => (MUL:{ *:[i32] } GPR32:{ *:[i32] }:$src2, GPR32:{ *:[i32] }:$src1)
-// NOOPT-NEXT:    GIR_BuildMI, /*InsnID*/0, /*Opcode*/GIMT_Encode2(MyTarget::MUL),
-// NOOPT-NEXT:    GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/0, //  DstI[dst]
-// NOOPT-NEXT:    GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/2, // src2
-// NOOPT-NEXT:    GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/1, // src1
-// NOOPT-NEXT:    GIR_ConstrainSelectedInstOperands, /*InsnID*/0,
-// NOOPT-NEXT:    GIR_EraseFromParent, /*InsnID*/0,
+// NOOPT-NEXT:    GIR_BuildRootMI, /*Opcode*/GIMT_Encode2(MyTarget::MUL),
+// NOOPT-NEXT:    GIR_RootToRootCopy, /*OpIdx*/0, //  DstI[dst]
+// NOOPT-NEXT:    GIR_RootToRootCopy, /*OpIdx*/2, // src2
+// NOOPT-NEXT:    GIR_RootToRootCopy, /*OpIdx*/1, // src1
+// NOOPT-NEXT:    GIR_RootConstrainSelectedInstOperands,
 // NOOPT-NEXT:    // GIR_Coverage, 16,
-// NOOPT-NEXT:    GIR_Done,
+// NOOPT-NEXT:    GIR_EraseRootFromParent_Done,
 // NOOPT-NEXT:  // Label [[LABEL_NUM]]: @[[LABEL]]
 
 def MUL : I<(outs GPR32:$dst), (ins GPR32:$src2, GPR32:$src1),
@@ -1170,11 +1149,11 @@ def MUL : I<(outs GPR32:$dst), (ins GPR32:$src2, GPR32:$src1),
 // NOOPT-NEXT:    GIM_CheckNumOperands, /*MI*/0, /*Expected*/2,
 // NOOPT-NEXT:    GIM_CheckOpcode, /*MI*/0, GIMT_Encode2(TargetOpcode::G_BITCAST),
 // NOOPT-NEXT:    // MIs[0] DstI[dst]
-// NOOPT-NEXT:    GIM_CheckType, /*MI*/0, /*Op*/0, /*Type*/GILLT_s32,
-// NOOPT-NEXT:    GIM_CheckRegBankForClass, /*MI*/0, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
+// NOOPT-NEXT:    GIM_RootCheckType, /*Op*/0, /*Type*/GILLT_s32,
+// NOOPT-NEXT:    GIM_RootCheckRegBankForClass, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
 // NOOPT-NEXT:    // MIs[0] src1
-// NOOPT-NEXT:    GIM_CheckType, /*MI*/0, /*Op*/1, /*Type*/GILLT_s32,
-// NOOPT-NEXT:    GIM_CheckRegBankForClass, /*MI*/0, /*Op*/1, /*RC*/GIMT_Encode2(MyTarget::FPR32RegClassID),
+// NOOPT-NEXT:    GIM_RootCheckType, /*Op*/1, /*Type*/GILLT_s32,
+// NOOPT-NEXT:    GIM_RootCheckRegBankForClass, /*Op*/1, /*RC*/GIMT_Encode2(MyTarget::FPR32RegClassID),
 // NOOPT-NEXT:    // (bitconvert:{ *:[i32] } FPR32:{ *:[f32] }:$src1) => (COPY_TO_REGCLASS:{ *:[i32] } FPR32:{ *:[f32] }:$src1, GPR32:{ *:[i32] })
 // NOOPT-NEXT:    GIR_MutateOpcode, /*InsnID*/0, /*RecycleInsnID*/0, /*Opcode*/GIMT_Encode2(TargetOpcode::COPY),
 // NOOPT-NEXT:    GIR_ConstrainOperandRC, /*InsnID*/0, /*Op*/0, GIMT_Encode2(MyTarget::GPR32RegClassID),
@@ -1191,18 +1170,17 @@ def : Pat<(i32 (bitconvert FPR32:$src1)),
 // NOOPT-NEXT:    GIM_CheckNumOperands, /*MI*/0, /*Expected*/2,
 // NOOPT-NEXT:    GIM_CheckOpcode, /*MI*/0, GIMT_Encode2(TargetOpcode::G_CONSTANT),
 // NOOPT-NEXT:    // MIs[0] DstI[dst]
-// NOOPT-NEXT:    GIM_CheckType, /*MI*/0, /*Op*/0, /*Type*/GILLT_s32,
-// NOOPT-NEXT:    GIM_CheckRegBankForClass, /*MI*/0, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
+// NOOPT-NEXT:    GIM_RootCheckType, /*Op*/0, /*Type*/GILLT_s32,
+// NOOPT-NEXT:    GIM_RootCheckRegBankForClass, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
 // NOOPT-NEXT:    // MIs[0] Operand 1
 // NOOPT-NEXT:    // No operand predicates
 // NOOPT-NEXT:    // (imm:{ *:[i32] }):$imm =>  (MOVimm:{ *:[i32] } (imm:{ *:[i32] }):$imm)
-// NOOPT-NEXT:    GIR_BuildMI, /*InsnID*/0, /*Opcode*/GIMT_Encode2(MyTarget::MOVimm),
-// NOOPT-NEXT:    GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/0, //  DstI[dst]
+// NOOPT-NEXT:    GIR_BuildRootMI, /*Opcode*/GIMT_Encode2(MyTarget::MOVimm),
+// NOOPT-NEXT:    GIR_RootToRootCopy, /*OpIdx*/0, //  DstI[dst]
 // NOOPT-NEXT:    GIR_CopyConstantAsSImm, /*NewInsnID*/0, /*OldInsnID*/0, // imm
-// NOOPT-NEXT:    GIR_ConstrainSelectedInstOperands, /*InsnID*/0,
-// NOOPT-NEXT:    GIR_EraseFromParent, /*InsnID*/0,
+// NOOPT-NEXT:    GIR_RootConstrainSelectedInstOperands,
 // NOOPT-NEXT:    // GIR_Coverage, 17,
-// NOOPT-NEXT:    GIR_Done,
+// NOOPT-NEXT:    GIR_EraseRootFromParent_Done,
 // NOOPT-NEXT:  // Label [[LABEL_NUM]]: @[[LABEL]]
 
 def MOVimm : I<(outs GPR32:$dst), (ins i32imm:$imm), [(set GPR32:$dst, imm:$imm)]>;
@@ -1219,7 +1197,7 @@ def MOVfpimmz : I<(outs FPR32:$dst), (ins f32imm:$imm), [(set FPR32:$dst, fpimmz
 // NOOPT-NEXT:    GIM_CheckIsMBB, /*MI*/0, /*Op*/0,
 // NOOPT-NEXT:    // (br (bb:{ *:[Other] }):$target) => (BR (bb:{ *:[Other] }):$target)
 // NOOPT-NEXT:    GIR_MutateOpcode, /*InsnID*/0, /*RecycleInsnID*/0, /*Opcode*/GIMT_Encode2(MyTarget::BR),
-// NOOPT-NEXT:    GIR_ConstrainSelectedInstOperands, /*InsnID*/0,
+// NOOPT-NEXT:    GIR_RootConstrainSelectedInstOperands,
 // NOOPT-NEXT:    // GIR_Coverage, 19,
 // NOOPT-NEXT:    GIR_Done,
 // NOOPT-NEXT:  // Label [[LABEL_NUM]]: @[[LABEL]]
@@ -1228,5 +1206,5 @@ def BR : I<(outs), (ins unknown:$target),
             [(br bb:$target)]>;
 
 // NOOPT-NEXT:    GIM_Reject,
-// NOOPT-NEXT:  }; // Size: 1738 bytes
+// NOOPT-NEXT:  }; // Size: 1459 bytes
 // NOOPT-NEXT:  return MatchTable0;
diff --git a/llvm/test/TableGen/GlobalISelEmitterCustomPredicate.td b/llvm/test/TableGen/GlobalISelEmitterCustomPredicate.td
index ed43ff2f14fa..3ceadf32f064 100644
--- a/llvm/test/TableGen/GlobalISelEmitterCustomPredicate.td
+++ b/llvm/test/TableGen/GlobalISelEmitterCustomPredicate.td
@@ -73,18 +73,18 @@ def and_or_pat : PatFrag<
   let PredicateCodeUsesOperands = 1;
 }
 
-// CHECK:      GIM_Try, /*On fail goto*//*Label 0*/ GIMT_Encode4(110), // Rule ID 7 //
+// CHECK:      GIM_Try, /*On fail goto*//*Label 0*/ GIMT_Encode4(97), // Rule ID 7 //
 // CHECK-NEXT:   GIM_CheckNumOperands, /*MI*/0, /*Expected*/3,
 // CHECK-NEXT:   GIM_CheckOpcode, /*MI*/0, GIMT_Encode2(TargetOpcode::G_AND),
 // CHECK-NEXT:   // MIs[0] DstI[dst]
-// CHECK-NEXT:   GIM_CheckType, /*MI*/0, /*Op*/0, /*Type*/GILLT_s32,
-// CHECK-NEXT:   GIM_CheckRegBankForClass, /*MI*/0, /*Op*/0, /*RC*/GIMT_Encode2(Test::DRegsRegClassID),
+// CHECK-NEXT:   GIM_RootCheckType, /*Op*/0, /*Type*/GILLT_s32,
+// CHECK-NEXT:   GIM_RootCheckRegBankForClass, /*Op*/0, /*RC*/GIMT_Encode2(Test::DRegsRegClassID),
 // CHECK-NEXT:   // MIs[0] src2
-// CHECK-NEXT:   GIM_CheckType, /*MI*/0, /*Op*/1, /*Type*/GILLT_s32,
+// CHECK-NEXT:   GIM_RootCheckType, /*Op*/1, /*Type*/GILLT_s32,
 // CHECK-NEXT:   GIM_RecordNamedOperand, /*MI*/0, /*Op*/1, /*StoreIdx*/2, // Name : pred:3:z
-// CHECK-NEXT:   GIM_CheckRegBankForClass, /*MI*/0, /*Op*/1, /*RC*/GIMT_Encode2(Test::DRegsRegClassID),
+// CHECK-NEXT:   GIM_RootCheckRegBankForClass, /*Op*/1, /*RC*/GIMT_Encode2(Test::DRegsRegClassID),
 // CHECK-NEXT:   // MIs[0] Operand 2
-// CHECK-NEXT:   GIM_CheckType, /*MI*/0, /*Op*/2, /*Type*/GILLT_s32,
+// CHECK-NEXT:   GIM_RootCheckType, /*Op*/2, /*Type*/GILLT_s32,
 // CHECK-NEXT:   GIM_RecordInsn, /*DefineMI*/1, /*MI*/0, /*OpIdx*/2, // MIs[1]
 // CHECK-NEXT:   GIM_CheckNumOperands, /*MI*/1, /*Expected*/3,
 // CHECK-NEXT:   GIM_CheckOpcode, /*MI*/1, GIMT_Encode2(TargetOpcode::G_OR),
@@ -99,18 +99,18 @@ def and_or_pat : PatFrag<
 // CHECK-NEXT:   GIM_RecordNamedOperand, /*MI*/1, /*Op*/2, /*StoreIdx*/1, // Name : pred:3:y
 // CHECK-NEXT:   GIM_CheckRegBankForClass, /*MI*/1, /*Op*/2, /*RC*/GIMT_Encode2(Test::DRegsRegClassID),
 // CHECK-NEXT:   GIM_CheckCxxInsnPredicate, /*MI*/0, /*FnId*/GIMT_Encode2(GICXXPred_MI_Predicate_and_or_pat),
-// CHECK-NEXT:   GIM_CheckIsSafeToFold, /*InsnID*/1,
+// CHECK-NEXT:   GIM_CheckIsSafeToFold, /*NumInsns*/1,
 // CHECK-NEXT:   // (and:{ *:[i32] } DOP:{ *:[i32] }:$src2:$pred:3:z, (or:{ *:[i32] } DOP:{ *:[i32] }:$src0:$pred:3:x, DOP:{ *:[i32] }:$src1:$pred:3:y))<<P:3:Predicate_and_or_pat>>  =>  (AND_OR:{ *:[i32] } DOP:{ *:[i32] }:$src0, DOP:{ *:[i32] }:$src1, DOP:{ *:[i32] }:$src2)
-// CHECK-NEXT:   GIR_BuildMI, /*InsnID*/0, /*Opcode*/GIMT_Encode2(MyTarget::AND_OR),
+// CHECK-NEXT:   GIR_BuildRootMI, /*Opcode*/GIMT_Encode2(MyTarget::AND_OR),
 
-// CHECK:      GIM_Try, /*On fail goto*//*Label 1*/ GIMT_Encode4(220), // Rule ID 3 //
+// CHECK:      GIM_Try, /*On fail goto*//*Label 1*/ GIMT_Encode4(194), // Rule ID 3 //
 // CHECK-NEXT:   GIM_CheckNumOperands, /*MI*/0, /*Expected*/3,
 // CHECK-NEXT:   GIM_CheckOpcode, /*MI*/0, GIMT_Encode2(TargetOpcode::G_AND),
 // CHECK-NEXT:   // MIs[0] DstI[dst]
-// CHECK-NEXT:   GIM_CheckType, /*MI*/0, /*Op*/0, /*Type*/GILLT_s32,
-// CHECK-NEXT:   GIM_CheckRegBankForClass, /*MI*/0, /*Op*/0, /*RC*/GIMT_Encode2(Test::DRegsRegClassID),
+// CHECK-NEXT:   GIM_RootCheckType, /*Op*/0, /*Type*/GILLT_s32,
+// CHECK-NEXT:   GIM_RootCheckRegBankForClass, /*Op*/0, /*RC*/GIMT_Encode2(Test::DRegsRegClassID),
 // CHECK-NEXT:   // MIs[0] Operand 1
-// CHECK-NEXT:   GIM_CheckType, /*MI*/0, /*Op*/1, /*Type*/GILLT_s32,
+// CHECK-NEXT:   GIM_RootCheckType, /*Op*/1, /*Type*/GILLT_s32,
 // CHECK-NEXT:   GIM_RecordInsn, /*DefineMI*/1, /*MI*/0, /*OpIdx*/1, // MIs[1]
 // CHECK-NEXT:   GIM_CheckNumOperands, /*MI*/1, /*Expected*/3,
 // CHECK-NEXT:   GIM_CheckOpcode, /*MI*/1, GIMT_Encode2(TargetOpcode::G_OR),
@@ -125,13 +125,13 @@ def and_or_pat : PatFrag<
 // CHECK-NEXT:   GIM_RecordNamedOperand, /*MI*/1, /*Op*/2, /*StoreIdx*/1, // Name : pred:3:y
 // CHECK-NEXT:   GIM_CheckRegBankForClass, /*MI*/1, /*Op*/2, /*RC*/GIMT_Encode2(Test::DRegsRegClassID),
 // CHECK-NEXT:   // MIs[0] src2
-// CHECK-NEXT:   GIM_CheckType, /*MI*/0, /*Op*/2, /*Type*/GILLT_s32,
+// CHECK-NEXT:   GIM_RootCheckType, /*Op*/2, /*Type*/GILLT_s32,
 // CHECK-NEXT:   GIM_RecordNamedOperand, /*MI*/0, /*Op*/2, /*StoreIdx*/2, // Name : pred:3:z
-// CHECK-NEXT:   GIM_CheckRegBankForClass, /*MI*/0, /*Op*/2, /*RC*/GIMT_Encode2(Test::DRegsRegClassID),
+// CHECK-NEXT:   GIM_RootCheckRegBankForClass, /*Op*/2, /*RC*/GIMT_Encode2(Test::DRegsRegClassID),
 // CHECK-NEXT:   GIM_CheckCxxInsnPredicate, /*MI*/0, /*FnId*/GIMT_Encode2(GICXXPred_MI_Predicate_and_or_pat),
-// CHECK-NEXT:   GIM_CheckIsSafeToFold, /*InsnID*/1,
+// CHECK-NEXT:   GIM_CheckIsSafeToFold, /*NumInsns*/1,
 // CHECK-NEXT:   // (and:{ *:[i32] } (or:{ *:[i32] } DOP:{ *:[i32] }:$src0:$pred:3:x, DOP:{ *:[i32] }:$src1:$pred:3:y), DOP:{ *:[i32] }:$src2:$pred:3:z)<<P:3:Predicate_and_or_pat>>  =>  (AND_OR:{ *:[i32] } DOP:{ *:[i32] }:$src0, DOP:{ *:[i32] }:$src1, DOP:{ *:[i32] }:$src2)
-// CHECK-NEXT:   GIR_BuildMI, /*InsnID*/0, /*Opcode*/GIMT_Encode2(MyTarget::AND_OR),
+// CHECK-NEXT:   GIR_BuildRootMI, /*Opcode*/GIMT_Encode2(MyTarget::AND_OR),
 
 // Test commutative, standalone pattern.
 def : Pat<
@@ -148,14 +148,14 @@ def mul_pat : PatFrag<
   let PredicateCodeUsesOperands = 1;
 }
 
-// CHECK:      GIM_Try, /*On fail goto*//*Label 2*/ GIMT_Encode4(326), // Rule ID 4 //
+// CHECK:      GIM_Try, /*On fail goto*//*Label 2*/ GIMT_Encode4(287), // Rule ID 4 //
 // CHECK-NEXT:   GIM_CheckNumOperands, /*MI*/0, /*Expected*/3,
 // CHECK-NEXT:   GIM_CheckOpcode, /*MI*/0, GIMT_Encode2(TargetOpcode::G_MUL),
 // CHECK-NEXT:   // MIs[0] DstI[dst]
-// CHECK-NEXT:   GIM_CheckType, /*MI*/0, /*Op*/0, /*Type*/GILLT_s32,
-// CHECK-NEXT:   GIM_CheckRegBankForClass, /*MI*/0, /*Op*/0, /*RC*/GIMT_Encode2(Test::DRegsRegClassID),
+// CHECK-NEXT:   GIM_RootCheckType, /*Op*/0, /*Type*/GILLT_s32,
+// CHECK-NEXT:   GIM_RootCheckRegBankForClass, /*Op*/0, /*RC*/GIMT_Encode2(Test::DRegsRegClassID),
 // CHECK-NEXT:   // MIs[0] Operand 1
-// CHECK-NEXT:   GIM_CheckType, /*MI*/0, /*Op*/1, /*Type*/GILLT_s32,
+// CHECK-NEXT:   GIM_RootCheckType, /*Op*/1, /*Type*/GILLT_s32,
 // CHECK-NEXT:   GIM_RecordNamedOperand, /*MI*/0, /*Op*/1, /*StoreIdx*/0, // Name : pred:4:x
 // CHECK-NEXT:   GIM_RecordInsn, /*DefineMI*/1, /*MI*/0, /*OpIdx*/1, // MIs[1]
 // CHECK-NEXT:   GIM_CheckNumOperands, /*MI*/1, /*Expected*/3,
@@ -169,26 +169,26 @@ def mul_pat : PatFrag<
 // CHECK-NEXT:   GIM_CheckType, /*MI*/1, /*Op*/2, /*Type*/GILLT_s32,
 // CHECK-NEXT:   GIM_CheckRegBankForClass, /*MI*/1, /*Op*/2, /*RC*/GIMT_Encode2(Test::DRegsRegClassID),
 // CHECK-NEXT:   // MIs[0] src2
-// CHECK-NEXT:   GIM_CheckType, /*MI*/0, /*Op*/2, /*Type*/GILLT_s32,
+// CHECK-NEXT:   GIM_RootCheckType, /*Op*/2, /*Type*/GILLT_s32,
 // CHECK-NEXT:   GIM_RecordNamedOperand, /*MI*/0, /*Op*/2, /*StoreIdx*/1, // Name : pred:4:y
-// CHECK-NEXT:   GIM_CheckRegBankForClass, /*MI*/0, /*Op*/2, /*RC*/GIMT_Encode2(Test::DRegsRegClassID),
+// CHECK-NEXT:   GIM_RootCheckRegBankForClass, /*Op*/2, /*RC*/GIMT_Encode2(Test::DRegsRegClassID),
 // CHECK-NEXT:   GIM_CheckCxxInsnPredicate, /*MI*/0, /*FnId*/GIMT_Encode2(GICXXPred_MI_Predicate_mul_pat),
-// CHECK-NEXT:   GIM_CheckIsSafeToFold, /*InsnID*/1,
+// CHECK-NEXT:   GIM_CheckIsSafeToFold, /*NumInsns*/1,
 // CHECK-NEXT:   // (mul:{ *:[i32] } (or:{ *:[i32] } DOP:{ *:[i32] }:$src0, DOP:{ *:[i32] }:$src1):$pred:4:x, DOP:{ *:[i32] }:$src2:$pred:4:y)<<P:4:Predicate_mul_pat>>  =>  (MUL_OR:{ *:[i32] } DOP:{ *:[i32] }:$src0, DOP:{ *:[i32] }:$src1, DOP:{ *:[i32] }:$src2)
-// CHECK-NEXT:   GIR_BuildMI, /*InsnID*/0, /*Opcode*/GIMT_Encode2(MyTarget::MUL_OR),
+// CHECK-NEXT:   GIR_BuildRootMI, /*Opcode*/GIMT_Encode2(MyTarget::MUL_OR),
 
-// CHECK:      GIM_Try, /*On fail goto*//*Label 3*/ GIMT_Encode4(432), // Rule ID 8 //
+// CHECK:      GIM_Try, /*On fail goto*//*Label 3*/ GIMT_Encode4(380), // Rule ID 8 //
 // CHECK-NEXT:   GIM_CheckNumOperands, /*MI*/0, /*Expected*/3,
 // CHECK-NEXT:   GIM_CheckOpcode, /*MI*/0, GIMT_Encode2(TargetOpcode::G_MUL),
 // CHECK-NEXT:   // MIs[0] DstI[dst]
-// CHECK-NEXT:   GIM_CheckType, /*MI*/0, /*Op*/0, /*Type*/GILLT_s32,
-// CHECK-NEXT:   GIM_CheckRegBankForClass, /*MI*/0, /*Op*/0, /*RC*/GIMT_Encode2(Test::DRegsRegClassID),
+// CHECK-NEXT:   GIM_RootCheckType, /*Op*/0, /*Type*/GILLT_s32,
+// CHECK-NEXT:   GIM_RootCheckRegBankForClass, /*Op*/0, /*RC*/GIMT_Encode2(Test::DRegsRegClassID),
 // CHECK-NEXT:   // MIs[0] src2
-// CHECK-NEXT:   GIM_CheckType, /*MI*/0, /*Op*/1, /*Type*/GILLT_s32,
+// CHECK-NEXT:   GIM_RootCheckType, /*Op*/1, /*Type*/GILLT_s32,
 // CHECK-NEXT:   GIM_RecordNamedOperand, /*MI*/0, /*Op*/1, /*StoreIdx*/1, // Name : pred:4:y
-// CHECK-NEXT:   GIM_CheckRegBankForClass, /*MI*/0, /*Op*/1, /*RC*/GIMT_Encode2(Test::DRegsRegClassID),
+// CHECK-NEXT:   GIM_RootCheckRegBankForClass, /*Op*/1, /*RC*/GIMT_Encode2(Test::DRegsRegClassID),
 // CHECK-NEXT:   // MIs[0] Operand 2
-// CHECK-NEXT:   GIM_CheckType, /*MI*/0, /*Op*/2, /*Type*/GILLT_s32,
+// CHECK-NEXT:   GIM_RootCheckType, /*Op*/2, /*Type*/GILLT_s32,
 // CHECK-NEXT:   GIM_RecordNamedOperand, /*MI*/0, /*Op*/2, /*StoreIdx*/0, // Name : pred:4:x
 // CHECK-NEXT:   GIM_RecordInsn, /*DefineMI*/1, /*MI*/0, /*OpIdx*/2, // MIs[1]
 // CHECK-NEXT:   GIM_CheckNumOperands, /*MI*/1, /*Expected*/3,
@@ -202,9 +202,9 @@ def mul_pat : PatFrag<
 // CHECK-NEXT:   GIM_CheckType, /*MI*/1, /*Op*/2, /*Type*/GILLT_s32,
 // CHECK-NEXT:   GIM_CheckRegBankForClass, /*MI*/1, /*Op*/2, /*RC*/GIMT_Encode2(Test::DRegsRegClassID),
 // CHECK-NEXT:   GIM_CheckCxxInsnPredicate, /*MI*/0, /*FnId*/GIMT_Encode2(GICXXPred_MI_Predicate_mul_pat),
-// CHECK-NEXT:   GIM_CheckIsSafeToFold, /*InsnID*/1,
+// CHECK-NEXT:   GIM_CheckIsSafeToFold, /*NumInsns*/1,
 // CHECK-NEXT:   // (mul:{ *:[i32] } DOP:{ *:[i32] }:$src2:$pred:4:y, (or:{ *:[i32] } DOP:{ *:[i32] }:$src0, DOP:{ *:[i32] }:$src1):$pred:4:x)<<P:4:Predicate_mul_pat>>  =>  (MUL_OR:{ *:[i32] } DOP:{ *:[i32] }:$src0, DOP:{ *:[i32] }:$src1, DOP:{ *:[i32] }:$src2)
-// CHECK-NEXT:   GIR_BuildMI, /*InsnID*/0, /*Opcode*/GIMT_Encode2(MyTarget::MUL_OR),
+// CHECK-NEXT:   GIR_BuildRootMI, /*Opcode*/GIMT_Encode2(MyTarget::MUL_OR),
 
 // Test commutative patterns where named operands in the source pattern are not
 // directly bound to PatFrag's operands.
@@ -223,14 +223,14 @@ def sub3_pat : PatFrag<
   let PredicateCodeUsesOperands = 1;
 }
 
-// CHECK:      GIM_Try, /*On fail goto*//*Label 4*/ GIMT_Encode4(527), // Rule ID 0 //
+// CHECK:      GIM_Try, /*On fail goto*//*Label 4*/ GIMT_Encode4(463), // Rule ID 0 //
 // CHECK-NEXT:   GIM_CheckNumOperands, /*MI*/0, /*Expected*/3,
 // CHECK-NEXT:   GIM_CheckOpcode, /*MI*/0, GIMT_Encode2(TargetOpcode::G_SUB),
 // CHECK-NEXT:   // MIs[0] DstI[dst]
-// CHECK-NEXT:   GIM_CheckType, /*MI*/0, /*Op*/0, /*Type*/GILLT_s32,
-// CHECK-NEXT:   GIM_CheckRegBankForClass, /*MI*/0, /*Op*/0, /*RC*/GIMT_Encode2(Test::DRegsRegClassID),
+// CHECK-NEXT:   GIM_RootCheckType, /*Op*/0, /*Type*/GILLT_s32,
+// CHECK-NEXT:   GIM_RootCheckRegBankForClass, /*Op*/0, /*RC*/GIMT_Encode2(Test::DRegsRegClassID),
 // CHECK-NEXT:   // MIs[0] Operand 1
-// CHECK-NEXT:   GIM_CheckType, /*MI*/0, /*Op*/1, /*Type*/GILLT_s32,
+// CHECK-NEXT:   GIM_RootCheckType, /*Op*/1, /*Type*/GILLT_s32,
 // CHECK-NEXT:   GIM_RecordInsn, /*DefineMI*/1, /*MI*/0, /*OpIdx*/1, // MIs[1]
 // CHECK-NEXT:   GIM_CheckNumOperands, /*MI*/1, /*Expected*/3,
 // CHECK-NEXT:   GIM_CheckOpcode, /*MI*/1, GIMT_Encode2(TargetOpcode::G_SUB),
@@ -243,12 +243,12 @@ def sub3_pat : PatFrag<
 // CHECK-NEXT:   GIM_CheckType, /*MI*/1, /*Op*/2, /*Type*/GILLT_s32,
 // CHECK-NEXT:   GIM_RecordNamedOperand, /*MI*/1, /*Op*/2, /*StoreIdx*/1, // Name : pred:1:y
 // CHECK-NEXT:   // MIs[0] src2
-// CHECK-NEXT:   GIM_CheckType, /*MI*/0, /*Op*/2, /*Type*/GILLT_s32,
+// CHECK-NEXT:   GIM_RootCheckType, /*Op*/2, /*Type*/GILLT_s32,
 // CHECK-NEXT:   GIM_RecordNamedOperand, /*MI*/0, /*Op*/2, /*StoreIdx*/2, // Name : pred:1:z
 // CHECK-NEXT:   GIM_CheckCxxInsnPredicate, /*MI*/0, /*FnId*/GIMT_Encode2(GICXXPred_MI_Predicate_sub3_pat),
-// CHECK-NEXT:   GIM_CheckIsSafeToFold, /*InsnID*/1,
+// CHECK-NEXT:   GIM_CheckIsSafeToFold, /*NumInsns*/1,
 // CHECK-NEXT:   // (sub:{ *:[i32] } (sub:{ *:[i32] } i32:{ *:[i32] }:$src0:$pred:1:x, i32:{ *:[i32] }:$src1:$pred:1:y), i32:{ *:[i32] }:$src2:$pred:1:z)<<P:1:Predicate_sub3_pat>>  =>  (SUB3:{ *:[i32] } i32:{ *:[i32] }:$src0, i32:{ *:[i32] }:$src1, i32:{ *:[i32] }:$src2)
-// CHECK-NEXT:   GIR_BuildMI, /*InsnID*/0, /*Opcode*/GIMT_Encode2(MyTarget::SUB3)
+// CHECK-NEXT:   GIR_BuildRootMI, /*Opcode*/GIMT_Encode2(MyTarget::SUB3)
 
 // Test a non-commutative pattern.
 def SUB3 : I<(outs DRegs:$dst),
@@ -269,16 +269,16 @@ def patfrags_test_pat : PatFrags<
   let PredicateCodeUsesOperands = 1;
 }
 
-// CHECK: GIM_Try, /*On fail goto*//*Label 5*/ GIMT_Encode4(622), // Rule ID 1 //
+// CHECK: GIM_Try, /*On fail goto*//*Label 5*/ GIMT_Encode4(546), // Rule ID 1 //
 // CHECK: // (xor:{ *:[i32] } (add:{ *:[i32] } i32:{ *:[i32] }:$src0:$pred:2:x, i32:{ *:[i32] }:$src1:$pred:2:y), i32:{ *:[i32] }:$src2:$pred:2:z)<<P:2:Predicate_patfrags_test_pat>>  =>  (PATFRAGS:{ *:[i32] } i32:{ *:[i32] }:$src0, i32:{ *:[i32] }:$src1, i32:{ *:[i32] }:$src2)
 
-// CHECK: GIM_Try, /*On fail goto*//*Label 6*/ GIMT_Encode4(717), // Rule ID 2 //
+// CHECK: GIM_Try, /*On fail goto*//*Label 6*/ GIMT_Encode4(629), // Rule ID 2 //
 // CHECK: // (xor:{ *:[i32] } (sub:{ *:[i32] } i32:{ *:[i32] }:$src0:$pred:2:x, i32:{ *:[i32] }:$src1:$pred:2:y), i32:{ *:[i32] }:$src2:$pred:2:z)<<P:2:Predicate_patfrags_test_pat>>  =>  (PATFRAGS:{ *:[i32] } i32:{ *:[i32] }:$src0, i32:{ *:[i32] }:$src1, i32:{ *:[i32] }:$src2)
 
-// CHECK: GIM_Try, /*On fail goto*//*Label 7*/ GIMT_Encode4(812), // Rule ID 5 //
+// CHECK: GIM_Try, /*On fail goto*//*Label 7*/ GIMT_Encode4(712), // Rule ID 5 //
 // CHECK: // (xor:{ *:[i32] } i32:{ *:[i32] }:$src2:$pred:2:z, (add:{ *:[i32] } i32:{ *:[i32] }:$src0:$pred:2:x, i32:{ *:[i32] }:$src1:$pred:2:y))<<P:2:Predicate_patfrags_test_pat>>  =>  (PATFRAGS:{ *:[i32] } i32:{ *:[i32] }:$src0, i32:{ *:[i32] }:$src1, i32:{ *:[i32] }:$src2)
 
-// CHECK: GIM_Try, /*On fail goto*//*Label 8*/ GIMT_Encode4(907), // Rule ID 6 //
+// CHECK: GIM_Try, /*On fail goto*//*Label 8*/ GIMT_Encode4(795), // Rule ID 6 //
 // CHECK: // (xor:{ *:[i32] } i32:{ *:[i32] }:$src2:$pred:2:z, (sub:{ *:[i32] } i32:{ *:[i32] }:$src0:$pred:2:x, i32:{ *:[i32] }:$src1:$pred:2:y))<<P:2:Predicate_patfrags_test_pat>>  =>  (PATFRAGS:{ *:[i32] } i32:{ *:[i32] }:$src0, i32:{ *:[i32] }:$src1, i32:{ *:[i32] }:$src2)
 
 
diff --git a/llvm/test/TableGen/GlobalISelEmitterFlags.td b/llvm/test/TableGen/GlobalISelEmitterFlags.td
index 2025a2c1048e..fa8f2a79fbce 100644
--- a/llvm/test/TableGen/GlobalISelEmitterFlags.td
+++ b/llvm/test/TableGen/GlobalISelEmitterFlags.td
@@ -42,7 +42,7 @@ def : Pat<
 // CHECK:    GIM_CheckIsSameOperand, /*MI*/2, /*OpIdx*/2, /*OtherMI*/2, /*OtherOpIdx*/1
 // CHECK:    GIM_CheckIsSameOperandIgnoreCopies, /*MI*/0, /*OpIdx*/2, /*OtherMI*/2, /*OtherOpIdx*/1
 // CHECK:    // (srl:{ *:[i32] } (srl:{ *:[i32] } (add:{ *:[i32] } i32:{ *:[i32] }:$src0, i32:{ *:[i32] }:$src0), i32:{ *:[i32] }:$src1), i32:{ *:[i32] }:$src0)
-// CHECK:   GIR_Done
+// CHECK:   GIR_EraseRootFromParent_Done
 // CHECK: GIM_Try
 // CHECK:   GIM_RecordInsnIgnoreCopies, /*DefineMI*/1, /*MI*/0, /*OpIdx*/1, // MIs[1]
 // CHECK:   GIM_CheckOpcode, /*MI*/1, GIMT_Encode2(TargetOpcode::G_SHL)
@@ -51,10 +51,10 @@ def : Pat<
 // CHECK:   GIM_CheckIsSameOperandIgnoreCopies, /*MI*/2, /*OpIdx*/2, /*OtherMI*/2, /*OtherOpIdx*/1
 // CHECK:   GIM_CheckIsSameOperandIgnoreCopies, /*MI*/0, /*OpIdx*/2, /*OtherMI*/2, /*OtherOpIdx*/1
 // CHECK:   // (srl:{ *:[i32] } (shl:{ *:[i32] } (mul:{ *:[i32] } i32:{ *:[i32] }:$src0, i32:{ *:[i32] }:$src0), i32:{ *:[i32] }:$src1), i32:{ *:[i32] }:$src0)
-// CHECK:   GIR_Done
+// CHECK:   GIR_EraseRootFromParent_Done
 // CHECK: GIM_Try
 // CHECK:   GIM_RecordInsnIgnoreCopies, /*DefineMI*/1, /*MI*/0, /*OpIdx*/1, // MIs[1]
 // CHECK:   GIM_CheckOpcode, /*MI*/1, GIMT_Encode2(TargetOpcode::G_MUL)
 // CHECK:   GIM_CheckIsSameOperandIgnoreCopies, /*MI*/1, /*OpIdx*/2, /*OtherMI*/1, /*OtherOpIdx*/1
 // CHECK:   // (sub:{ *:[i32] } (mul:{ *:[i32] } i32:{ *:[i32] }:$src0, i32:{ *:[i32] }:$src0), i32:{ *:[i32] }:$src1)  =>  (InstThreeOperands:{ *:[i32] } GPR32:{ *:[i32] }:$src0, GPR32:{ *:[i32] }:$src1, GPR32:{ *:[i32] }:$src1)
-// CHECK:   GIR_Done
+// CHECK:   GIR_EraseRootFromParent_Done
diff --git a/llvm/test/TableGen/GlobalISelEmitterHwModes.td b/llvm/test/TableGen/GlobalISelEmitterHwModes.td
index 0c826fb50beb..7c8340a154e8 100644
--- a/llvm/test/TableGen/GlobalISelEmitterHwModes.td
+++ b/llvm/test/TableGen/GlobalISelEmitterHwModes.td
@@ -85,7 +85,7 @@ class I<dag OOps, dag IOps, list<dag> Pat>
 
 // CHECK-LABEL: PredicateBitset MyTargetInstructionSelector::
 // CHECK-NEXT:  computeAvailableModuleFeatures(const MyTargetSubtarget *Subtarget) const {
-// CHECK-NEXT:    PredicateBitset Features;
+// CHECK-NEXT:    PredicateBitset Features{};
 // CHECK-NEXT:    if (!((Subtarget->has64())))
 // CHECK-NEXT:      Features.set(Feature_HwMode1Bit);
 // CHECK-NEXT:    if ((Subtarget->has64()))
@@ -95,7 +95,7 @@ class I<dag OOps, dag IOps, list<dag> Pat>
 
 // CHECK-LABEL: PredicateBitset MyTargetInstructionSelector::
 // CHECK-NEXT:  computeAvailableFunctionFeatures(const MyTargetSubtarget *Subtarget, const MachineFunction *MF) const {
-// CHECK-NEXT:    PredicateBitset Features;
+// CHECK-NEXT:    PredicateBitset Features{};
 // CHECK-NEXT:    return Features;
 // CHECK-NEXT:  }
 
@@ -134,14 +134,14 @@ class I<dag OOps, dag IOps, list<dag> Pat>
 // CHECK-NEXT:    GIM_CheckMemorySizeEqualToLLT, /*MI*/0, /*MMO*/0, /*OpIdx*/0,
 // CHECK-NEXT:    GIM_CheckAtomicOrdering, /*MI*/0, /*Order*/(uint8_t)AtomicOrdering::NotAtomic,
 // CHECK-NEXT:    // MIs[0] DstI[dst]
-// CHECK-NEXT:    GIM_CheckType, /*MI*/0, /*Op*/0, /*Type*/GILLT_s64,
-// CHECK-NEXT:    GIM_CheckRegBankForClass, /*MI*/0, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPRRegClassID),
+// CHECK-NEXT:    GIM_RootCheckType, /*Op*/0, /*Type*/GILLT_s64,
+// CHECK-NEXT:    GIM_RootCheckRegBankForClass, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPRRegClassID),
 // CHECK-NEXT:    // MIs[0] src1
 // CHECK-NEXT:    GIM_CheckPointerToAny, /*MI*/0, /*Op*/1, /*SizeInBits*/64,
-// CHECK-NEXT:    GIM_CheckRegBankForClass, /*MI*/0, /*Op*/1, /*RC*/GIMT_Encode2(MyTarget::GPRRegClassID),
+// CHECK-NEXT:    GIM_RootCheckRegBankForClass, /*Op*/1, /*RC*/GIMT_Encode2(MyTarget::GPRRegClassID),
 // CHECK-NEXT:    // (ld:{ *:[i64] } GPR:{ *:[i64] }:$src1)<<P:Predicate_unindexedload>><<P:Predicate_load>>  =>  (LOAD:{ *:[i64] } GPR:{ *:[i64] }:$src1)
 // CHECK-NEXT:    GIR_MutateOpcode, /*InsnID*/0, /*RecycleInsnID*/0, /*Opcode*/GIMT_Encode2(MyTarget::LOAD),
-// CHECK-NEXT:    GIR_ConstrainSelectedInstOperands, /*InsnID*/0,
+// CHECK-NEXT:    GIR_RootConstrainSelectedInstOperands,
 // CHECK-NEXT:    // GIR_Coverage, 0,
 // CHECK-NEXT:    GIR_Done,
 // CHECK-NEXT:  // Label [[LABEL_NUM]]: @[[LABEL]]
@@ -152,14 +152,14 @@ class I<dag OOps, dag IOps, list<dag> Pat>
 // CHECK-NEXT:    GIM_CheckMemorySizeEqualToLLT, /*MI*/0, /*MMO*/0, /*OpIdx*/0,
 // CHECK-NEXT:    GIM_CheckAtomicOrdering, /*MI*/0, /*Order*/(uint8_t)AtomicOrdering::NotAtomic,
 // CHECK-NEXT:    // MIs[0] DstI[dst]
-// CHECK-NEXT:    GIM_CheckType, /*MI*/0, /*Op*/0, /*Type*/GILLT_s32,
-// CHECK-NEXT:    GIM_CheckRegBankForClass, /*MI*/0, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPRRegClassID),
+// CHECK-NEXT:    GIM_RootCheckType, /*Op*/0, /*Type*/GILLT_s32,
+// CHECK-NEXT:    GIM_RootCheckRegBankForClass, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPRRegClassID),
 // CHECK-NEXT:    // MIs[0] src1
 // CHECK-NEXT:    GIM_CheckPointerToAny, /*MI*/0, /*Op*/1, /*SizeInBits*/32,
-// CHECK-NEXT:    GIM_CheckRegBankForClass, /*MI*/0, /*Op*/1, /*RC*/GIMT_Encode2(MyTarget::GPRRegClassID),
+// CHECK-NEXT:    GIM_RootCheckRegBankForClass, /*Op*/1, /*RC*/GIMT_Encode2(MyTarget::GPRRegClassID),
 // CHECK-NEXT:    // (ld:{ *:[i32] } GPR:{ *:[i32] }:$src1)<<P:Predicate_unindexedload>><<P:Predicate_load>>  =>  (LOAD:{ *:[i32] } GPR:{ *:[i32] }:$src1)
 // CHECK-NEXT:    GIR_MutateOpcode, /*InsnID*/0, /*RecycleInsnID*/0, /*Opcode*/GIMT_Encode2(MyTarget::LOAD),
-// CHECK-NEXT:    GIR_ConstrainSelectedInstOperands, /*InsnID*/0,
+// CHECK-NEXT:    GIR_RootConstrainSelectedInstOperands,
 // CHECK-NEXT:    // GIR_Coverage, 1,
 // CHECK-NEXT:    GIR_Done,
 // CHECK-NEXT:  // Label [[LABEL_NUM]]: @[[LABEL]]
@@ -176,14 +176,14 @@ def LOAD : I<(outs GPR:$dst), (ins GPR:$src1),
 // CHECK-NEXT:    GIM_CheckMemorySizeEqualToLLT, /*MI*/0, /*MMO*/0, /*OpIdx*/0,
 // CHECK-NEXT:    GIM_CheckAtomicOrdering, /*MI*/0, /*Order*/(uint8_t)AtomicOrdering::NotAtomic,
 // CHECK-NEXT:    // MIs[0] DstI[dst]
-// CHECK-NEXT:    GIM_CheckType, /*MI*/0, /*Op*/0, /*Type*/GILLT_p0s64,
-// CHECK-NEXT:    GIM_CheckRegBankForClass, /*MI*/0, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPRRegClassID),
+// CHECK-NEXT:    GIM_RootCheckType, /*Op*/0, /*Type*/GILLT_p0s64,
+// CHECK-NEXT:    GIM_RootCheckRegBankForClass, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPRRegClassID),
 // CHECK-NEXT:    // MIs[0] src
 // CHECK-NEXT:    GIM_CheckPointerToAny, /*MI*/0, /*Op*/1, /*SizeInBits*/64,
-// CHECK-NEXT:    GIM_CheckRegBankForClass, /*MI*/0, /*Op*/1, /*RC*/GIMT_Encode2(MyTarget::GPRRegClassID),
+// CHECK-NEXT:    GIM_RootCheckRegBankForClass, /*Op*/1, /*RC*/GIMT_Encode2(MyTarget::GPRRegClassID),
 // CHECK-NEXT:    // (ld:{ *:[i64] } GPR:{ *:[i64] }:$src)<<P:Predicate_unindexedload>><<P:Predicate_load>>  =>  (LOAD:{ *:[i64] } GPR:{ *:[i64] }:$src)
 // CHECK-NEXT:    GIR_MutateOpcode, /*InsnID*/0, /*RecycleInsnID*/0, /*Opcode*/GIMT_Encode2(MyTarget::LOAD),
-// CHECK-NEXT:    GIR_ConstrainSelectedInstOperands, /*InsnID*/0,
+// CHECK-NEXT:    GIR_RootConstrainSelectedInstOperands,
 // CHECK-NEXT:    // GIR_Coverage, 2,
 // CHECK-NEXT:    GIR_Done,
 // CHECK-NEXT:  // Label [[LABEL_NUM]]: @[[LABEL]]
@@ -194,14 +194,14 @@ def LOAD : I<(outs GPR:$dst), (ins GPR:$src1),
 // CHECK-NEXT:    GIM_CheckMemorySizeEqualToLLT, /*MI*/0, /*MMO*/0, /*OpIdx*/0,
 // CHECK-NEXT:    GIM_CheckAtomicOrdering, /*MI*/0, /*Order*/(uint8_t)AtomicOrdering::NotAtomic,
 // CHECK-NEXT:    // MIs[0] DstI[dst]
-// CHECK-NEXT:    GIM_CheckType, /*MI*/0, /*Op*/0, /*Type*/GILLT_p0s32,
-// CHECK-NEXT:    GIM_CheckRegBankForClass, /*MI*/0, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPRRegClassID),
+// CHECK-NEXT:    GIM_RootCheckType, /*Op*/0, /*Type*/GILLT_p0s32,
+// CHECK-NEXT:    GIM_RootCheckRegBankForClass, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPRRegClassID),
 // CHECK-NEXT:    // MIs[0] src
 // CHECK-NEXT:    GIM_CheckPointerToAny, /*MI*/0, /*Op*/1, /*SizeInBits*/32,
-// CHECK-NEXT:    GIM_CheckRegBankForClass, /*MI*/0, /*Op*/1, /*RC*/GIMT_Encode2(MyTarget::GPRRegClassID),
+// CHECK-NEXT:    GIM_RootCheckRegBankForClass, /*Op*/1, /*RC*/GIMT_Encode2(MyTarget::GPRRegClassID),
 // CHECK-NEXT:    // (ld:{ *:[i32] } GPR:{ *:[i32] }:$src)<<P:Predicate_unindexedload>><<P:Predicate_load>>  =>  (LOAD:{ *:[i32] } GPR:{ *:[i32] }:$src)
 // CHECK-NEXT:    GIR_MutateOpcode, /*InsnID*/0, /*RecycleInsnID*/0, /*Opcode*/GIMT_Encode2(MyTarget::LOAD),
-// CHECK-NEXT:    GIR_ConstrainSelectedInstOperands, /*InsnID*/0,
+// CHECK-NEXT:    GIR_RootConstrainSelectedInstOperands,
 // CHECK-NEXT:    // GIR_Coverage, 3,
 // CHECK-NEXT:    GIR_Done,
 // CHECK-NEXT:  // Label [[LABEL_NUM]]: @[[LABEL]]
diff --git a/llvm/test/TableGen/GlobalISelEmitterMatchTableOptimizer.td b/llvm/test/TableGen/GlobalISelEmitterMatchTableOptimizer.td
index 5b534970d4d4..3db31bea8612 100644
--- a/llvm/test/TableGen/GlobalISelEmitterMatchTableOptimizer.td
+++ b/llvm/test/TableGen/GlobalISelEmitterMatchTableOptimizer.td
@@ -11,24 +11,24 @@ def LOAD32 : I<(outs GPR8:$dst), (ins GPR32:$src), []>;
 // CHECK-NEXT: GIM_Try, /*On fail goto*//*Label [[L1_ID:[0-9]+]]*/ GIMT_Encode4([[L1_AT:[0-9]+]]),
 // CHECK-NEXT: GIM_CheckMemorySizeEqualToLLT, /*MI*/0, /*MMO*/0, /*OpIdx*/0,
 // CHECK-NEXT:   GIM_CheckAtomicOrdering, /*MI*/0, /*Order*/(uint8_t)AtomicOrdering::NotAtomic,
-// CHECK-NEXT: GIM_CheckRegBankForClass, /*MI*/0, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR8RegClassID),
+// CHECK-NEXT: GIM_RootCheckRegBankForClass, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR8RegClassID),
 // CHECK-NEXT: GIM_Try, /*On fail goto*//*Label [[L2_ID:[0-9]+]]*/ GIMT_Encode4([[L2_AT:[0-9]+]]),
 // CHECK-NEXT: // MIs[0] src
 // CHECK-NEXT: GIM_CheckPointerToAny, /*MI*/0, /*Op*/1, /*SizeInBits*/8,
-// CHECK-NEXT: GIM_CheckRegBankForClass, /*MI*/0, /*Op*/1, /*RC*/GIMT_Encode2(MyTarget::GPR8RegClassID),
+// CHECK-NEXT: GIM_RootCheckRegBankForClass, /*Op*/1, /*RC*/GIMT_Encode2(MyTarget::GPR8RegClassID),
 // CHECK-NEXT: // (ld:{ *:[i8] } GPR8:{ *:[i8] }:$src)<<P:Predicate_unindexedload>><<P:Predicate_load>>  =>  (LOAD8:{ *:[i8] } GPR8:{ *:[i8] }:$src)
 // CHECK-NEXT: GIR_MutateOpcode, /*InsnID*/0, /*RecycleInsnID*/0, /*Opcode*/GIMT_Encode2(MyTarget::LOAD8),
-// CHECK-NEXT: GIR_ConstrainSelectedInstOperands, /*InsnID*/0,
+// CHECK-NEXT: GIR_RootConstrainSelectedInstOperands,
 // CHECK-NEXT: // GIR_Coverage, 0,
 // CHECK-NEXT: GIR_Done,
 // CHECK-NEXT: // Label [[L2_ID]]: @[[L2_AT]]
 // CHECK-NEXT: GIM_Try, /*On fail goto*//*Label [[L3_ID:[0-9]+]]*/ GIMT_Encode4([[L3_AT:[0-9]+]]),
 // CHECK-NEXT: // MIs[0] src
 // CHECK-NEXT: GIM_CheckPointerToAny, /*MI*/0, /*Op*/1, /*SizeInBits*/32,
-// CHECK-NEXT: GIM_CheckRegBankForClass, /*MI*/0, /*Op*/1, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
+// CHECK-NEXT: GIM_RootCheckRegBankForClass, /*Op*/1, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
 // CHECK-NEXT: // (ld:{ *:[i8] } GPR32:{ *:[i32] }:$src)<<P:Predicate_unindexedload>><<P:Predicate_load>>  =>  (LOAD32:{ *:[i8] } GPR32:{ *:[i32] }:$src)
 // CHECK-NEXT: GIR_MutateOpcode, /*InsnID*/0, /*RecycleInsnID*/0, /*Opcode*/GIMT_Encode2(MyTarget::LOAD32),
-// CHECK-NEXT: GIR_ConstrainSelectedInstOperands, /*InsnID*/0,
+// CHECK-NEXT: GIR_RootConstrainSelectedInstOperands,
 // CHECK-NEXT: // GIR_Coverage, 1,
 // CHECK-NEXT: GIR_Done,
 // CHECK-NEXT: // Label [[L3_ID]]: @[[L3_AT]]
@@ -49,7 +49,7 @@ def LOAD16Imm : I<(outs GPR16:$dst), (ins GPR16:$src), []>;
 // CHECK-NEXT: GIM_Try, /*On fail goto*//*Label [[L1_ID:[0-9]+]]*/ GIMT_Encode4([[L1_AT:[0-9]+]]),
 // CHECK-NEXT: GIM_CheckMemorySizeEqualToLLT, /*MI*/0, /*MMO*/0, /*OpIdx*/0,
 // CHECK-NEXT: GIM_CheckAtomicOrdering, /*MI*/0, /*Order*/(uint8_t)AtomicOrdering::NotAtomic,
-// CHECK-NEXT: GIM_CheckRegBankForClass, /*MI*/0, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR16RegClassID),
+// CHECK-NEXT: GIM_RootCheckRegBankForClass, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR16RegClassID),
 // CHECK-NEXT: GIM_CheckPointerToAny, /*MI*/0, /*Op*/1, /*SizeInBits*/16,
 // CHECK-NEXT: GIM_Try, /*On fail goto*//*Label [[L2_ID:[0-9]+]]*/ GIMT_Encode4([[L2_AT:[0-9]+]]),
 // CHECK-NEXT: GIM_RecordInsn, /*DefineMI*/1, /*MI*/0, /*OpIdx*/1, // MIs[1]
@@ -58,22 +58,21 @@ def LOAD16Imm : I<(outs GPR16:$dst), (ins GPR16:$src), []>;
 // CHECK-NEXT: GIM_CheckType, /*MI*/1, /*Op*/2, /*Type*/GILLT_s16,
 // CHECK-NEXT: GIM_CheckRegBankForClass, /*MI*/1, /*Op*/1, /*RC*/GIMT_Encode2(MyTarget::GPR16RegClassID),
 // CHECK-NEXT: GIM_CheckConstantInt8, /*MI*/1, /*Op*/2, 10,
-// CHECK-NEXT: GIM_CheckIsSafeToFold, /*InsnID*/1,
+// CHECK-NEXT: GIM_CheckIsSafeToFold, /*NumInsns*/1,
 // CHECK-NEXT: // (ld:{ *:[i16] } (add:{ *:[i16] } GPR16:{ *:[i16] }:$src, 10:{ *:[i16] }))<<P:Predicate_unindexedload>><<P:Predicate_load>>  =>  (LOAD16Imm:{ *:[i16] } GPR16:{ *:[i16] }:$src)
-// CHECK-NEXT: GIR_BuildMI, /*InsnID*/0, /*Opcode*/GIMT_Encode2(MyTarget::LOAD16Imm),
-// CHECK-NEXT: GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/0, // DstI[dst]
+// CHECK-NEXT: GIR_BuildRootMI, /*Opcode*/GIMT_Encode2(MyTarget::LOAD16Imm),
+// CHECK-NEXT: GIR_RootToRootCopy, /*OpIdx*/0, // DstI[dst]
 // CHECK-NEXT: GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/1, /*OpIdx*/1, // src
 // CHECK-NEXT: GIR_MergeMemOperands, /*InsnID*/0, /*NumInsns*/2, /*MergeInsnID's*/0, 1,
-// CHECK-NEXT: GIR_ConstrainSelectedInstOperands, /*InsnID*/0,
-// CHECK-NEXT: GIR_EraseFromParent, /*InsnID*/0,
+// CHECK-NEXT: GIR_RootConstrainSelectedInstOperands,
 // CHECK-NEXT: // GIR_Coverage, 3,
-// CHECK-NEXT: GIR_Done,
+// CHECK-NEXT: GIR_EraseRootFromParent_Done,
 // CHECK-NEXT: // Label [[L2_ID]]: @[[L2_AT]]
 // CHECK-NEXT: GIM_Try, /*On fail goto*//*Label [[L3_ID:[0-9]+]]*/ GIMT_Encode4([[L3_AT:[0-9]+]]),
-// CHECK-NEXT: GIM_CheckRegBankForClass, /*MI*/0, /*Op*/1, /*RC*/GIMT_Encode2(MyTarget::GPR16RegClassID),
+// CHECK-NEXT: GIM_RootCheckRegBankForClass, /*Op*/1, /*RC*/GIMT_Encode2(MyTarget::GPR16RegClassID),
 // CHECK-NEXT: // (ld:{ *:[i16] } GPR16:{ *:[i16] }:$src)<<P:Predicate_unindexedload>><<P:Predicate_load>>  =>  (LOAD16:{ *:[i16] } GPR16:{ *:[i16] }:$src)
 // CHECK-NEXT: GIR_MutateOpcode, /*InsnID*/0, /*RecycleInsnID*/0, /*Opcode*/GIMT_Encode2(MyTarget::LOAD16),
-// CHECK-NEXT: GIR_ConstrainSelectedInstOperands, /*InsnID*/0,
+// CHECK-NEXT: GIR_RootConstrainSelectedInstOperands,
 // CHECK-NEXT: // GIR_Coverage, 2,
 // CHECK-NEXT: GIR_Done,
 // CHECK-NEXT: // Label [[L3_ID]]: @[[L3_AT]]
diff --git a/llvm/test/TableGen/GlobalISelEmitterMatchTableOptimizerSameOperand-invalid.td b/llvm/test/TableGen/GlobalISelEmitterMatchTableOptimizerSameOperand-invalid.td
index 729a30e8cbf3..d93805b612a1 100644
--- a/llvm/test/TableGen/GlobalISelEmitterMatchTableOptimizerSameOperand-invalid.td
+++ b/llvm/test/TableGen/GlobalISelEmitterMatchTableOptimizerSameOperand-invalid.td
@@ -5,88 +5,83 @@ include "GlobalISelEmitterCommon.td"
 
 def InstTwoOperands : I<(outs GPR32:$dst), (ins GPR32:$src1, GPR32:$src2), []>;
 def InstThreeOperands : I<(outs GPR32:$dst), (ins GPR32:$cond, GPR32:$src,GPR32:$src2), []>;
-
-// CHECK:      GIM_Try, /*On fail goto*//*Label 0*/ GIMT_Encode4(255),
-// CHECK-NEXT:      GIM_CheckOpcode, /*MI*/0, GIMT_Encode2(TargetOpcode::G_SELECT),
-// CHECK-NEXT:      GIM_CheckType, /*MI*/0, /*Op*/0, /*Type*/GILLT_s32,
-// CHECK-NEXT:      GIM_CheckType, /*MI*/0, /*Op*/1, /*Type*/GILLT_s32,
-// CHECK-NEXT:      GIM_CheckType, /*MI*/0, /*Op*/2, /*Type*/GILLT_s32,
-// CHECK-NEXT:      GIM_Try, /*On fail goto*//*Label 1*/ GIMT_Encode4(217),
-// CHECK-NEXT:        GIM_CheckRegBankForClass, /*MI*/0, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
-// CHECK-NEXT:        GIM_CheckIsSameOperand, /*MI*/0, /*OpIdx*/3, /*OtherMI*/2, /*OtherOpIdx*/2,
-// CHECK-NEXT:        GIM_Try, /*On fail goto*//*Label 2*/ GIMT_Encode4(126), // Rule ID 1 //
-// CHECK-NEXT:          GIM_RecordInsn, /*DefineMI*/1, /*MI*/0, /*OpIdx*/1, // MIs[1]
-// CHECK-NEXT:          GIM_CheckOpcode, /*MI*/1, GIMT_Encode2(TargetOpcode::G_ICMP),
-// CHECK-NEXT:          GIM_CheckType, /*MI*/1, /*Op*/2, /*Type*/GILLT_s32,
-// CHECK-NEXT:          GIM_CheckType, /*MI*/1, /*Op*/3, /*Type*/GILLT_s32,
-// CHECK-NEXT:          // MIs[1] Operand 1
-// CHECK-NEXT:          GIM_CheckCmpPredicate, /*MI*/1, /*Op*/1, /*Predicate*/GIMT_Encode2(CmpInst::ICMP_EQ),
-// CHECK-NEXT:          GIM_CheckRegBankForClass, /*MI*/1, /*Op*/2, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
-// CHECK-NEXT:          GIM_CheckConstantInt8, /*MI*/1, /*Op*/3, 0,
-// CHECK-NEXT:          GIM_RecordInsn, /*DefineMI*/2, /*MI*/0, /*OpIdx*/2, // MIs[2]
-// CHECK-NEXT:          GIM_CheckOpcode, /*MI*/2, GIMT_Encode2(TargetOpcode::G_SUB),
-// CHECK-NEXT:          GIM_CheckType, /*MI*/2, /*Op*/1, /*Type*/GILLT_s32,
-// CHECK-NEXT:          GIM_CheckType, /*MI*/2, /*Op*/2, /*Type*/GILLT_s32,
-// CHECK-NEXT:          GIM_CheckRegBankForClass, /*MI*/2, /*Op*/1, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
-// CHECK-NEXT:          GIM_CheckRegBankForClass, /*MI*/2, /*Op*/2, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
-// CHECK-NEXT:          GIM_CheckIsSafeToFold, /*InsnID*/1,
-// CHECK-NEXT:          GIM_CheckIsSafeToFold, /*InsnID*/2,
-// CHECK-NEXT:          // (select:{ *:[i32] } (setcc:{ *:[i32] } GPR32:{ *:[i32] }:$cond, 0:{ *:[i32] }, SETEQ:{ *:[Other] }), (sub:{ *:[i32] } GPR32:{ *:[i32] }:$src1, GPR32:{ *:[i32] }:$src2), GPR32:{ *:[i32] }:$src2)  =>  (InstThreeOperands:{ *:[i32] } GPR32:{ *:[i32] }:$cond, GPR32:{ *:[i32] }:$src1, GPR32:{ *:[i32] }:$src2)
-// CHECK-NEXT:          GIR_BuildMI, /*InsnID*/0, /*Opcode*/GIMT_Encode2(MyTarget::InstThreeOperands),
-// CHECK-NEXT:          GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/0, // DstI[dst]
-// CHECK-NEXT:          GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/1, /*OpIdx*/2, // cond
-// CHECK-NEXT:          GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/2, /*OpIdx*/1, // src1
-// CHECK-NEXT:          GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/2, /*OpIdx*/2, // src2
-// CHECK-NEXT:          GIR_ConstrainSelectedInstOperands, /*InsnID*/0,
-// CHECK-NEXT:          GIR_EraseFromParent, /*InsnID*/0,
-// CHECK-NEXT:          // GIR_Coverage, 1,
-// CHECK-NEXT:          GIR_Done,
-// CHECK-NEXT:        // Label 2: @126
-// CHECK-NEXT:        GIM_Try, /*On fail goto*//*Label 3*/ GIMT_Encode4(216), // Rule ID 2 //
-// CHECK-NEXT:          GIM_RecordInsn, /*DefineMI*/1, /*MI*/0, /*OpIdx*/1, // MIs[1]
-// CHECK-NEXT:          GIM_CheckOpcode, /*MI*/1, GIMT_Encode2(TargetOpcode::G_ICMP),
-// CHECK-NEXT:          GIM_CheckType, /*MI*/1, /*Op*/2, /*Type*/GILLT_s32,
-// CHECK-NEXT:          GIM_CheckType, /*MI*/1, /*Op*/3, /*Type*/GILLT_s32,
-// CHECK-NEXT:          // MIs[1] Operand 1
-// CHECK-NEXT:          GIM_CheckCmpPredicate, /*MI*/1, /*Op*/1, /*Predicate*/GIMT_Encode2(CmpInst::ICMP_NE),
-// CHECK-NEXT:          GIM_CheckRegBankForClass, /*MI*/1, /*Op*/2, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
-// CHECK-NEXT:          GIM_CheckConstantInt8, /*MI*/1, /*Op*/3, 0,
-// CHECK-NEXT:          GIM_RecordInsn, /*DefineMI*/2, /*MI*/0, /*OpIdx*/2, // MIs[2]
-// CHECK-NEXT:          GIM_CheckOpcode, /*MI*/2, GIMT_Encode2(TargetOpcode::G_SUB),
-// CHECK-NEXT:          GIM_CheckType, /*MI*/2, /*Op*/1, /*Type*/GILLT_s32,
-// CHECK-NEXT:          GIM_CheckType, /*MI*/2, /*Op*/2, /*Type*/GILLT_s32,
-// CHECK-NEXT:          GIM_CheckRegBankForClass, /*MI*/2, /*Op*/1, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
-// CHECK-NEXT:          GIM_CheckRegBankForClass, /*MI*/2, /*Op*/2, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
-// CHECK-NEXT:          GIM_CheckIsSafeToFold, /*InsnID*/1,
-// CHECK-NEXT:          GIM_CheckIsSafeToFold, /*InsnID*/2,
-// CHECK-NEXT:          // (select:{ *:[i32] } (setcc:{ *:[i32] } GPR32:{ *:[i32] }:$cond, 0:{ *:[i32] }, SETNE:{ *:[Other] }), (sub:{ *:[i32] } GPR32:{ *:[i32] }:$src1, GPR32:{ *:[i32] }:$src2), GPR32:{ *:[i32] }:$src2)  =>  (InstThreeOperands:{ *:[i32] } GPR32:{ *:[i32] }:$cond, GPR32:{ *:[i32] }:$src1, GPR32:{ *:[i32] }:$src2)
-// CHECK-NEXT:          GIR_BuildMI, /*InsnID*/0, /*Opcode*/GIMT_Encode2(MyTarget::InstThreeOperands),
-// CHECK-NEXT:          GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/0, // DstI[dst]
-// CHECK-NEXT:          GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/1, /*OpIdx*/2, // cond
-// CHECK-NEXT:          GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/2, /*OpIdx*/1, // src1
-// CHECK-NEXT:          GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/2, /*OpIdx*/2, // src2
-// CHECK-NEXT:          GIR_ConstrainSelectedInstOperands, /*InsnID*/0,
-// CHECK-NEXT:          GIR_EraseFromParent, /*InsnID*/0,
-// CHECK-NEXT:          // GIR_Coverage, 2,
-// CHECK-NEXT:          GIR_Done,
-// CHECK-NEXT:        // Label 3: @216
-// CHECK-NEXT:        GIM_Reject,
-// CHECK-NEXT:      // Label 1: @217
-// CHECK-NEXT:      GIM_Try, /*On fail goto*//*Label 4*/ GIMT_Encode4(254), // Rule ID 0 //
-// CHECK-NEXT:        GIM_CheckType, /*MI*/0, /*Op*/3, /*Type*/GILLT_s32,
-// CHECK-NEXT:        GIM_CheckRegBankForClass, /*MI*/0, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
-// CHECK-NEXT:        GIM_CheckRegBankForClass, /*MI*/0, /*Op*/1, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
-// CHECK-NEXT:        GIM_CheckRegBankForClass, /*MI*/0, /*Op*/2, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
-// CHECK-NEXT:        GIM_CheckRegBankForClass, /*MI*/0, /*Op*/3, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
-// CHECK-NEXT:        // (select:{ *:[i32] } GPR32:{ *:[i32] }:$cond, GPR32:{ *:[i32] }:$src1, GPR32:{ *:[i32] }:$src2)  =>  (InstThreeOperands:{ *:[i32] } GPR32:{ *:[i32] }:$cond, GPR32:{ *:[i32] }:$src1, GPR32:{ *:[i32] }:$src2)
-// CHECK-NEXT:        GIR_MutateOpcode, /*InsnID*/0, /*RecycleInsnID*/0, /*Opcode*/GIMT_Encode2(MyTarget::InstThreeOperands),
-// CHECK-NEXT:        GIR_ConstrainSelectedInstOperands, /*InsnID*/0,
-// CHECK-NEXT:        // GIR_Coverage, 0,
-// CHECK-NEXT:        GIR_Done,
-// CHECK-NEXT:      // Label 4: @254
-// CHECK-NEXT:      GIM_Reject,
-// CHECK-NEXT:    // Label 0: @255
-// CHECK-NEXT:    GIM_Reject,
+// CHECK:      GIM_Try, /*On fail goto*//*Label 0*/ GIMT_Encode4(229),
+// CHECK-NEXT:   GIM_CheckOpcode, /*MI*/0, GIMT_Encode2(TargetOpcode::G_SELECT),
+// CHECK-NEXT:   GIM_RootCheckType, /*Op*/0, /*Type*/GILLT_s32,
+// CHECK-NEXT:   GIM_RootCheckType, /*Op*/1, /*Type*/GILLT_s32,
+// CHECK-NEXT:   GIM_RootCheckType, /*Op*/2, /*Type*/GILLT_s32,
+// CHECK-NEXT:   GIM_Try, /*On fail goto*//*Label 1*/ GIMT_Encode4(197),
+// CHECK-NEXT:     GIM_RootCheckRegBankForClass, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
+// CHECK-NEXT:     GIM_CheckIsSameOperand, /*MI*/0, /*OpIdx*/3, /*OtherMI*/2, /*OtherOpIdx*/2,
+// CHECK-NEXT:     GIM_Try, /*On fail goto*//*Label 2*/ GIMT_Encode4(114), // Rule ID 1 //
+// CHECK-NEXT:       GIM_RecordInsn, /*DefineMI*/1, /*MI*/0, /*OpIdx*/1, // MIs[1]
+// CHECK-NEXT:       GIM_CheckOpcode, /*MI*/1, GIMT_Encode2(TargetOpcode::G_ICMP),
+// CHECK-NEXT:       GIM_CheckType, /*MI*/1, /*Op*/2, /*Type*/GILLT_s32,
+// CHECK-NEXT:       GIM_CheckType, /*MI*/1, /*Op*/3, /*Type*/GILLT_s32,
+// CHECK-NEXT:       // MIs[1] Operand 1
+// CHECK-NEXT:       GIM_CheckCmpPredicate, /*MI*/1, /*Op*/1, /*Predicate*/GIMT_Encode2(CmpInst::ICMP_EQ),
+// CHECK-NEXT:       GIM_CheckRegBankForClass, /*MI*/1, /*Op*/2, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
+// CHECK-NEXT:       GIM_CheckConstantInt8, /*MI*/1, /*Op*/3, 0,
+// CHECK-NEXT:       GIM_RecordInsn, /*DefineMI*/2, /*MI*/0, /*OpIdx*/2, // MIs[2]
+// CHECK-NEXT:       GIM_CheckOpcode, /*MI*/2, GIMT_Encode2(TargetOpcode::G_SUB),
+// CHECK-NEXT:       GIM_CheckType, /*MI*/2, /*Op*/1, /*Type*/GILLT_s32,
+// CHECK-NEXT:       GIM_CheckType, /*MI*/2, /*Op*/2, /*Type*/GILLT_s32,
+// CHECK-NEXT:       GIM_CheckRegBankForClass, /*MI*/2, /*Op*/1, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
+// CHECK-NEXT:       GIM_CheckRegBankForClass, /*MI*/2, /*Op*/2, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
+// CHECK-NEXT:       GIM_CheckIsSafeToFold, /*NumInsns*/2,
+// CHECK-NEXT:       // (select:{ *:[i32] } (setcc:{ *:[i32] } GPR32:{ *:[i32] }:$cond, 0:{ *:[i32] }, SETEQ:{ *:[Other] }), (sub:{ *:[i32] } GPR32:{ *:[i32] }:$src1, GPR32:{ *:[i32] }:$src2), GPR32:{ *:[i32] }:$src2)  =>  (InstThreeOperands:{ *:[i32] } GPR32:{ *:[i32] }:$cond, GPR32:{ *:[i32] }:$src1, GPR32:{ *:[i32] }:$src2)
+// CHECK-NEXT:       GIR_BuildRootMI, /*Opcode*/GIMT_Encode2(MyTarget::InstThreeOperands),
+// CHECK-NEXT:       GIR_RootToRootCopy, /*OpIdx*/0, // DstI[dst]
+// CHECK-NEXT:       GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/1, /*OpIdx*/2, // cond
+// CHECK-NEXT:       GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/2, /*OpIdx*/1, // src1
+// CHECK-NEXT:       GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/2, /*OpIdx*/2, // src2
+// CHECK-NEXT:       GIR_RootConstrainSelectedInstOperands,
+// CHECK-NEXT:       // GIR_Coverage, 1,
+// CHECK-NEXT:       GIR_EraseRootFromParent_Done,
+// CHECK-NEXT:     // Label 2: @114
+// CHECK-NEXT:     GIM_Try, /*On fail goto*//*Label 3*/ GIMT_Encode4(196), // Rule ID 2 //
+// CHECK-NEXT:       GIM_RecordInsn, /*DefineMI*/1, /*MI*/0, /*OpIdx*/1, // MIs[1]
+// CHECK-NEXT:       GIM_CheckOpcode, /*MI*/1, GIMT_Encode2(TargetOpcode::G_ICMP),
+// CHECK-NEXT:       GIM_CheckType, /*MI*/1, /*Op*/2, /*Type*/GILLT_s32,
+// CHECK-NEXT:       GIM_CheckType, /*MI*/1, /*Op*/3, /*Type*/GILLT_s32,
+// CHECK-NEXT:       // MIs[1] Operand 1
+// CHECK-NEXT:       GIM_CheckCmpPredicate, /*MI*/1, /*Op*/1, /*Predicate*/GIMT_Encode2(CmpInst::ICMP_NE),
+// CHECK-NEXT:       GIM_CheckRegBankForClass, /*MI*/1, /*Op*/2, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
+// CHECK-NEXT:       GIM_CheckConstantInt8, /*MI*/1, /*Op*/3, 0,
+// CHECK-NEXT:       GIM_RecordInsn, /*DefineMI*/2, /*MI*/0, /*OpIdx*/2, // MIs[2]
+// CHECK-NEXT:       GIM_CheckOpcode, /*MI*/2, GIMT_Encode2(TargetOpcode::G_SUB),
+// CHECK-NEXT:       GIM_CheckType, /*MI*/2, /*Op*/1, /*Type*/GILLT_s32,
+// CHECK-NEXT:       GIM_CheckType, /*MI*/2, /*Op*/2, /*Type*/GILLT_s32,
+// CHECK-NEXT:       GIM_CheckRegBankForClass, /*MI*/2, /*Op*/1, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
+// CHECK-NEXT:       GIM_CheckRegBankForClass, /*MI*/2, /*Op*/2, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
+// CHECK-NEXT:       GIM_CheckIsSafeToFold, /*NumInsns*/2,
+// CHECK-NEXT:       // (select:{ *:[i32] } (setcc:{ *:[i32] } GPR32:{ *:[i32] }:$cond, 0:{ *:[i32] }, SETNE:{ *:[Other] }), (sub:{ *:[i32] } GPR32:{ *:[i32] }:$src1, GPR32:{ *:[i32] }:$src2), GPR32:{ *:[i32] }:$src2)  =>  (InstThreeOperands:{ *:[i32] } GPR32:{ *:[i32] }:$cond, GPR32:{ *:[i32] }:$src1, GPR32:{ *:[i32] }:$src2)
+// CHECK-NEXT:       GIR_BuildRootMI, /*Opcode*/GIMT_Encode2(MyTarget::InstThreeOperands),
+// CHECK-NEXT:       GIR_RootToRootCopy, /*OpIdx*/0, // DstI[dst]
+// CHECK-NEXT:       GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/1, /*OpIdx*/2, // cond
+// CHECK-NEXT:       GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/2, /*OpIdx*/1, // src1
+// CHECK-NEXT:       GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/2, /*OpIdx*/2, // src2
+// CHECK-NEXT:       GIR_RootConstrainSelectedInstOperands,
+// CHECK-NEXT:       // GIR_Coverage, 2,
+// CHECK-NEXT:       GIR_EraseRootFromParent_Done,
+// CHECK-NEXT:     // Label 3: @196
+// CHECK-NEXT:     GIM_Reject,
+// CHECK-NEXT:   // Label 1: @197
+// CHECK-NEXT:   GIM_Try, /*On fail goto*//*Label 4*/ GIMT_Encode4(228), // Rule ID 0 //
+// CHECK-NEXT:     GIM_RootCheckType, /*Op*/3, /*Type*/GILLT_s32,
+// CHECK-NEXT:     GIM_RootCheckRegBankForClass, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
+// CHECK-NEXT:     GIM_RootCheckRegBankForClass, /*Op*/1, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
+// CHECK-NEXT:     GIM_RootCheckRegBankForClass, /*Op*/2, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
+// CHECK-NEXT:     GIM_RootCheckRegBankForClass, /*Op*/3, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
+// CHECK-NEXT:     // (select:{ *:[i32] } GPR32:{ *:[i32] }:$cond, GPR32:{ *:[i32] }:$src1, GPR32:{ *:[i32] }:$src2)  =>  (InstThreeOperands:{ *:[i32] } GPR32:{ *:[i32] }:$cond, GPR32:{ *:[i32] }:$src1, GPR32:{ *:[i32] }:$src2)
+// CHECK-NEXT:     GIR_MutateOpcode, /*InsnID*/0, /*RecycleInsnID*/0, /*Opcode*/GIMT_Encode2(MyTarget::InstThreeOperands),
+// CHECK-NEXT:     GIR_RootConstrainSelectedInstOperands,
+// CHECK-NEXT:     // GIR_Coverage, 0,
+// CHECK-NEXT:     GIR_Done,
+// CHECK-NEXT:   // Label 4: @228
+// CHECK-NEXT:   GIM_Reject,
+// CHECK-NEXT: // Label 0: @229
+// CHECK-NEXT: GIM_Reject,
 def : Pat<(i32 (select GPR32:$cond, GPR32:$src1, GPR32:$src2)),
           (InstThreeOperands GPR32:$cond, GPR32:$src1, GPR32:$src2)>;
 
diff --git a/llvm/test/TableGen/GlobalISelEmitterMatchTableOptimizerSameOperand.td b/llvm/test/TableGen/GlobalISelEmitterMatchTableOptimizerSameOperand.td
index 00ece5f4ca79..1ac33990ab3b 100644
--- a/llvm/test/TableGen/GlobalISelEmitterMatchTableOptimizerSameOperand.td
+++ b/llvm/test/TableGen/GlobalISelEmitterMatchTableOptimizerSameOperand.td
@@ -9,7 +9,7 @@ def InstThreeOperands : I<(outs GPR32:$dst), (ins GPR32:$cond, GPR32:$src,GPR32:
 // Make sure the GIM_CheckIsSameOperand check is not hoisted into the common header group
 
 // CHECK:       GIM_Try, /*On fail goto*//*Label 1*/
-// CHECK-NEXT:  GIM_CheckRegBankForClass, /*MI*/0, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
+// CHECK-NEXT:  GIM_RootCheckRegBankForClass, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
 // CHECK-NOT:   GIM_CheckIsSameOperand
 // CHECK-NEXT:  GIM_Try, /*On fail goto*//*Label 2*/
 // CHECK:       GIM_CheckIsSameOperand, /*MI*/0, /*OpIdx*/3, /*OtherMI*/2, /*OtherOpIdx*/1,
diff --git a/llvm/test/TableGen/GlobalISelEmitterOverloadedPtr.td b/llvm/test/TableGen/GlobalISelEmitterOverloadedPtr.td
index 64723a0bbd4b..422edbba0e7a 100644
--- a/llvm/test/TableGen/GlobalISelEmitterOverloadedPtr.td
+++ b/llvm/test/TableGen/GlobalISelEmitterOverloadedPtr.td
@@ -12,14 +12,14 @@ let TargetPrefix = "mytarget" in {
 // GIM_CheckPointerToAny rather than a GIM_CheckType.
 //
 // CHECK: GIM_CheckIntrinsicID, /*MI*/0, /*Op*/1, GIMT_Encode2(Intrinsic::mytarget_anyptr),
-// CHECK-NEXT: GIM_CheckType, /*MI*/0, /*Op*/0, /*Type*/GILLT_s32,
-// CHECK-NEXT: GIM_CheckRegBankForClass, /*MI*/0, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
+// CHECK-NEXT: GIM_RootCheckType, /*Op*/0, /*Type*/GILLT_s32,
+// CHECK-NEXT: GIM_RootCheckRegBankForClass, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
 // CHECK-NEXT: // MIs[0] src
 // CHECK-NEXT: GIM_CheckPointerToAny, /*MI*/0, /*Op*/2, /*SizeInBits*/32,
-// CHECK-NEXT: GIM_CheckRegBankForClass, /*MI*/0, /*Op*/2, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
+// CHECK-NEXT: GIM_RootCheckRegBankForClass, /*Op*/2, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
 // CHECK-NEXT: GIM_CheckCxxInsnPredicate, /*MI*/0, /*FnId*/GIMT_Encode2(GICXXPred_MI_Predicate_frag_anyptr),
 // CHECK-NEXT: // (intrinsic_w_chain:{ *:[i32] } {{[0-9]+}}:{ *:[iPTR] }, GPR32:{ *:[i32] }:$src)<<P:Predicate_frag_anyptr>>  =>  (ANYLOAD:{ *:[i32] } GPR32:{ *:[i32] }:$src)
-// CHECK-NEXT: GIR_BuildMI, /*InsnID*/0, /*Opcode*/GIMT_Encode2(MyTarget::ANYLOAD),
+// CHECK-NEXT: GIR_BuildRootMI, /*Opcode*/GIMT_Encode2(MyTarget::ANYLOAD),
 let hasSideEffects = 1 in {
   def ANYLOAD : I<(outs GPR32:$dst), (ins GPR32:$src1),
             [(set GPR32:$dst, (load GPR32:$src1))]>;
diff --git a/llvm/test/TableGen/GlobalISelEmitterRegSequence.td b/llvm/test/TableGen/GlobalISelEmitterRegSequence.td
index 42ac68b9d91d..3829070b28ef 100644
--- a/llvm/test/TableGen/GlobalISelEmitterRegSequence.td
+++ b/llvm/test/TableGen/GlobalISelEmitterRegSequence.td
@@ -33,11 +33,11 @@ def SUBSOME_INSN : I<(outs SRegs:$dst), (ins SOP:$src), []>;
 // CHECK: GIM_CheckNumOperands, /*MI*/0, /*Expected*/2,
 // CHECK-NEXT: GIM_CheckOpcode, /*MI*/0, GIMT_Encode2(TargetOpcode::G_SEXT),
 // CHECK-NEXT: // MIs[0] DstI[dst]
-// CHECK-NEXT: GIM_CheckType, /*MI*/0, /*Op*/0, /*Type*/GILLT_s32,
-// CHECK-NEXT: GIM_CheckRegBankForClass, /*MI*/0, /*Op*/0, /*RC*/GIMT_Encode2(Test::DRegsRegClassID),
+// CHECK-NEXT: GIM_RootCheckType, /*Op*/0, /*Type*/GILLT_s32,
+// CHECK-NEXT: GIM_RootCheckRegBankForClass, /*Op*/0, /*RC*/GIMT_Encode2(Test::DRegsRegClassID),
 // CHECK-NEXT: // MIs[0] src
-// CHECK-NEXT: GIM_CheckType, /*MI*/0, /*Op*/1, /*Type*/GILLT_s16,
-// CHECK-NEXT: GIM_CheckRegBankForClass, /*MI*/0, /*Op*/1, /*RC*/GIMT_Encode2(Test::SRegsRegClassID),
+// CHECK-NEXT: GIM_RootCheckType, /*Op*/1, /*Type*/GILLT_s16,
+// CHECK-NEXT: GIM_RootCheckRegBankForClass, /*Op*/1, /*RC*/GIMT_Encode2(Test::SRegsRegClassID),
 // CHECK-NEXT: // (sext:{ *:[i32] } SOP:{ *:[i16] }:$src)  =>  (REG_SEQUENCE:{ *:[i32] } DRegs:{ *:[i32] }, (SUBSOME_INSN:{ *:[i16] } SOP:{ *:[i16] }:$src), sub0:{ *:[i32] }, (SUBSOME_INSN:{ *:[i16] } SOP:{ *:[i16] }:$src), sub1:{ *:[i32] })
 // CHECK-NEXT: GIR_MakeTempReg, /*TempRegID*/0, /*TypeID*/GILLT_s16,
 // CHECK-NEXT: GIR_MakeTempReg, /*TempRegID*/1, /*TypeID*/GILLT_s16,
@@ -49,8 +49,8 @@ def SUBSOME_INSN : I<(outs SRegs:$dst), (ins SOP:$src), []>;
 // CHECK-NEXT: GIR_AddTempRegister, /*InsnID*/1, /*TempRegID*/0, /*TempRegFlags*/GIMT_Encode2(RegState::Define),
 // CHECK-NEXT: GIR_Copy, /*NewInsnID*/1, /*OldInsnID*/0, /*OpIdx*/1, // src
 // CHECK-NEXT: GIR_ConstrainSelectedInstOperands, /*InsnID*/1,
-// CHECK-NEXT: GIR_BuildMI, /*InsnID*/0, /*Opcode*/GIMT_Encode2(TargetOpcode::REG_SEQUENCE),
-// CHECK-NEXT: GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/0, // DstI[dst]
+// CHECK-NEXT: GIR_BuildRootMI, /*Opcode*/GIMT_Encode2(TargetOpcode::REG_SEQUENCE),
+// CHECK-NEXT: GIR_RootToRootCopy, /*OpIdx*/0, // DstI[dst]
 // CHECK-NEXT: GIR_AddSimpleTempRegister, /*InsnID*/0, /*TempRegID*/0,
 // CHECK-NEXT: GIR_AddImm8, /*InsnID*/0, /*SubRegIndex*/1,
 // CHECK-NEXT: GIR_AddSimpleTempRegister, /*InsnID*/0, /*TempRegID*/1,
@@ -58,7 +58,8 @@ def SUBSOME_INSN : I<(outs SRegs:$dst), (ins SOP:$src), []>;
 // CHECK-NEXT: GIR_ConstrainOperandRC, /*InsnID*/0, /*Op*/0, GIMT_Encode2(Test::DRegsRegClassID),
 // CHECK-NEXT: GIR_ConstrainOperandRC, /*InsnID*/0, /*Op*/1, GIMT_Encode2(Test::SRegsRegClassID),
 // CHECK-NEXT: GIR_ConstrainOperandRC, /*InsnID*/0, /*Op*/3, GIMT_Encode2(Test::SRegsRegClassID),
-// CHECK-NEXT: GIR_EraseFromParent, /*InsnID*/0,
+// CHECK-NEXT: // GIR_Coverage, 0,
+// CHECK-NEXT: GIR_EraseRootFromParent_Done,
 def : Pat<(i32 (sext SOP:$src)),
           (REG_SEQUENCE DRegs, (SUBSOME_INSN SOP:$src), sub0,
                                (SUBSOME_INSN SOP:$src), sub1)>;
@@ -74,7 +75,7 @@ def : Pat<(i32 (sext SOP:$src)),
 // CHECK-NEXT: GIR_ConstrainOperandRC, /*InsnID*/1, /*Op*/0, GIMT_Encode2(Test::DRegsRegClassID),
 // CHECK-NEXT: GIR_ConstrainOperandRC, /*InsnID*/1, /*Op*/1, GIMT_Encode2(Test::SRegsRegClassID),
 // CHECK-NEXT: GIR_ConstrainOperandRC, /*InsnID*/1, /*Op*/3, GIMT_Encode2(Test::SRegsRegClassID),
-// CHECK-NEXT: GIR_BuildMI, /*InsnID*/0, /*Opcode*/GIMT_Encode2(MyTarget::SOME_INSN),
+// CHECK-NEXT: GIR_BuildRootMI, /*Opcode*/GIMT_Encode2(MyTarget::SOME_INSN),
 // Make sure operands are constrained when REG_SEQUENCE isn't the root instruction.
 def : Pat<(i32 (zext SOP:$src)),
           (SOME_INSN (REG_SEQUENCE DRegs, (SUBSOME_INSN SOP:$src), sub0,
diff --git a/llvm/test/TableGen/GlobalISelEmitterSkippedPatterns.td b/llvm/test/TableGen/GlobalISelEmitterSkippedPatterns.td
index 7c9df02ebd87..fc8abc6fbc54 100644
--- a/llvm/test/TableGen/GlobalISelEmitterSkippedPatterns.td
+++ b/llvm/test/TableGen/GlobalISelEmitterSkippedPatterns.td
@@ -1,4 +1,6 @@
 // RUN: llvm-tblgen -warn-on-skipped-patterns -gen-global-isel -I %p/../../include %s -I %p/Common -o /dev/null 2>&1 | FileCheck %s
+// RUN: llvm-tblgen -warn-on-skipped-patterns -gen-global-isel -I %p/../../include %s -I %p/Common -o /dev/null -DIGNORE 2>&1 | FileCheck --allow-empty --check-prefix=IGNORED %s
+
 include "llvm/Target/Target.td"
 include "GlobalISelEmitterCommon.td"
 
@@ -23,6 +25,10 @@ def INSN : I<(outs GPR32:$dst), (ins GPR32:$src1, complex:$src2), []>;
 
 //===- Bail out when we define a variable twice wrt complex suboperands. -===//
 
+#ifdef IGNORE
+let GISelShouldIgnore = 1 in
+#endif
+// IGNORED-NOT: warning: Skipped pattern: Error: {{.*}}
 // CHECK: warning: Skipped pattern: Error: Complex suboperand x referenced by different operands: complex_rr:x:y and complex_rr:x:z.
 def : Pat<(add (complex_rr GPR32:$x, GPR32:$y),
                (complex_rr GPR32:$x, GPR32:$z)),
diff --git a/llvm/test/TableGen/GlobalISelEmitterSubreg.td b/llvm/test/TableGen/GlobalISelEmitterSubreg.td
index 94e321e638c5..8df3238f6cc2 100644
--- a/llvm/test/TableGen/GlobalISelEmitterSubreg.td
+++ b/llvm/test/TableGen/GlobalISelEmitterSubreg.td
@@ -71,12 +71,13 @@ def : Pat<(sub (complex DOP:$src1, DOP:$src2), 77),
 // CHECK-NEXT: GIR_ComplexSubOperandSubRegRenderer, /*InsnID*/1, /*RendererID*/GIMT_Encode2(0), /*SubOperand*/0, /*SubRegIdx*/GIMT_Encode2(1), // src1
 // CHECK-NEXT: GIR_ConstrainOperandRC, /*InsnID*/1, /*Op*/0, GIMT_Encode2(Test::SRegsRegClassID),
 // CHECK-NEXT: GIR_ConstrainOperandRC, /*InsnID*/1, /*Op*/1, GIMT_Encode2(Test::DRegsRegClassID),
-// CHECK-NEXT: GIR_BuildMI, /*InsnID*/0, /*Opcode*/GIMT_Encode2(MyTarget::SOME_INSN2),
-// CHECK-NEXT: GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/0, // DstI[dst]
+// CHECK-NEXT: GIR_BuildRootMI, /*Opcode*/GIMT_Encode2(MyTarget::SOME_INSN2),
+// CHECK-NEXT: GIR_RootToRootCopy, /*OpIdx*/0, // DstI[dst]
 // CHECK-NEXT: GIR_AddSimpleTempRegister, /*InsnID*/0, /*TempRegID*/0,
 // CHECK-NEXT: GIR_AddSimpleTempRegister, /*InsnID*/0, /*TempRegID*/1,
-// CHECK-NEXT: GIR_ConstrainSelectedInstOperands, /*InsnID*/0,
-// CHECK-NEXT: GIR_EraseFromParent, /*InsnID*/0,
+// CHECK-NEXT: GIR_RootConstrainSelectedInstOperands,
+// CHECK-NEXT: // GIR_Coverage, 2,
+// CHECK-NEXT: GIR_EraseRootFromParent_Done,
 
 // Test that we import INSERT_SUBREG when its subregister source has a given
 // class.
@@ -86,15 +87,16 @@ def : Pat<(i32 (anyext i16:$src)), (INSERT_SUBREG (i32 (IMPLICIT_DEF)), SOP:$src
 // CHECK-NEXT:            GIR_BuildMI, /*InsnID*/1, /*Opcode*/GIMT_Encode2(TargetOpcode::IMPLICIT_DEF),
 // CHECK-NEXT:            GIR_AddTempRegister, /*InsnID*/1, /*TempRegID*/0, /*TempRegFlags*/GIMT_Encode2(RegState::Define),
 // CHECK-NEXT:            GIR_ConstrainSelectedInstOperands, /*InsnID*/1,
-// CHECK-NEXT:            GIR_BuildMI, /*InsnID*/0, /*Opcode*/GIMT_Encode2(TargetOpcode::INSERT_SUBREG),
-// CHECK-NEXT:            GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/0, // DstI[dst]
+// CHECK-NEXT:            GIR_BuildRootMI, /*Opcode*/GIMT_Encode2(TargetOpcode::INSERT_SUBREG),
+// CHECK-NEXT:            GIR_RootToRootCopy, /*OpIdx*/0, // DstI[dst]
 // CHECK-NEXT:            GIR_AddSimpleTempRegister, /*InsnID*/0, /*TempRegID*/0,
-// CHECK-NEXT:            GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/1, // src
+// CHECK-NEXT:            GIR_RootToRootCopy, /*OpIdx*/1, // src
 // CHECK-NEXT:            GIR_AddImm8, /*InsnID*/0, /*Imm*/1,
 // CHECK-NEXT:            GIR_ConstrainOperandRC, /*InsnID*/0, /*Op*/0, GIMT_Encode2(Test::DRegsRegClassID),
 // CHECK-NEXT:            GIR_ConstrainOperandRC, /*InsnID*/0, /*Op*/1, GIMT_Encode2(Test::DRegsRegClassID),
 // CHECK-NEXT:            GIR_ConstrainOperandRC, /*InsnID*/0, /*Op*/2, GIMT_Encode2(Test::SRegsRegClassID),
-// CHECK-NEXT:            GIR_EraseFromParent, /*InsnID*/0,
+// CHECK-NEXT:            // GIR_Coverage, 3,
+// CHECK-NEXT:            GIR_EraseRootFromParent_Done,
 
 
 // Test that we can import INSERT_SUBREG when it is a subinstruction of another
@@ -114,11 +116,12 @@ def : Pat<(i32 (anyext i16:$src)), (SOME_INSN (INSERT_SUBREG (i32 (IMPLICIT_DEF)
 // CHECK-NEXT:            GIR_ConstrainOperandRC, /*InsnID*/1, /*Op*/0, GIMT_Encode2(Test::DRegsRegClassID),
 // CHECK-NEXT:            GIR_ConstrainOperandRC, /*InsnID*/1, /*Op*/1, GIMT_Encode2(Test::DRegsRegClassID),
 // CHECK-NEXT:            GIR_ConstrainOperandRC, /*InsnID*/1, /*Op*/2, GIMT_Encode2(Test::SRegsRegClassID),
-// CHECK-NEXT:            GIR_BuildMI, /*InsnID*/0, /*Opcode*/GIMT_Encode2(MyTarget::SOME_INSN),
-// CHECK-NEXT:            GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/0, // DstI[dst]
+// CHECK-NEXT:            GIR_BuildRootMI, /*Opcode*/GIMT_Encode2(MyTarget::SOME_INSN),
+// CHECK-NEXT:            GIR_RootToRootCopy, /*OpIdx*/0, // DstI[dst]
 // CHECK-NEXT:            GIR_AddSimpleTempRegister, /*InsnID*/0, /*TempRegID*/0,
-// CHECK-NEXT:            GIR_ConstrainSelectedInstOperands, /*InsnID*/0,
-// CHECK-NEXT:            GIR_EraseFromParent, /*InsnID*/0,
+// CHECK-NEXT:            GIR_RootConstrainSelectedInstOperands,
+// CHECK-NEXT:            // GIR_Coverage, 4,
+// CHECK-NEXT:            GIR_EraseRootFromParent_Done,
 
 
 // Test that we correctly infer the super register class for INSERT_SUBREG when
@@ -126,7 +129,7 @@ def : Pat<(i32 (anyext i16:$src)), (SOME_INSN (INSERT_SUBREG (i32 (IMPLICIT_DEF)
 // not a D register.
 def : Pat<(i32 (anyext i16:$src)), (INSERT_SUBREG (i32 (COPY_TO_REGCLASS SOP:$src, ERegs)), SOP:$src, sub0)>;
 // CHECK-LABEL:  (anyext:{ *:[i32] } i16:{ *:[i16] }:$src)  =>  (INSERT_SUBREG:{ *:[i32] } (COPY_TO_REGCLASS:{ *:[i32] } SOP:{ *:[i16] }:$src, ERegs:{ *:[i32] }), SOP:{ *:[i16] }:$src, sub0:{ *:[i32] })
-// CHECK:                GIR_BuildMI, /*InsnID*/0, /*Opcode*/GIMT_Encode2(TargetOpcode::INSERT_SUBREG),
+// CHECK:                GIR_BuildRootMI, /*Opcode*/GIMT_Encode2(TargetOpcode::INSERT_SUBREG),
 // CHECK-DAG:            GIR_ConstrainOperandRC, /*InsnID*/0, /*Op*/0, GIMT_Encode2(Test::ERegsRegClassID),
 // CHECK-NEXT:           GIR_ConstrainOperandRC, /*InsnID*/0, /*Op*/1, GIMT_Encode2(Test::ERegsRegClassID),
 // CHECK-NEXT:           GIR_ConstrainOperandRC, /*InsnID*/0, /*Op*/2, GIMT_Encode2(Test::SRegsRegClassID),
@@ -144,15 +147,16 @@ def : Pat<(i32 (anyext i16:$src)), (INSERT_SUBREG (i32 (IMPLICIT_DEF)), (SUBSOME
 // CHECK-NEXT:          GIR_BuildMI, /*InsnID*/1, /*Opcode*/GIMT_Encode2(TargetOpcode::IMPLICIT_DEF),
 // CHECK-NEXT:          GIR_AddTempRegister, /*InsnID*/1, /*TempRegID*/0, /*TempRegFlags*/GIMT_Encode2(RegState::Define),
 // CHECK-NEXT:          GIR_ConstrainSelectedInstOperands, /*InsnID*/1,
-// CHECK-NEXT:          GIR_BuildMI, /*InsnID*/0, /*Opcode*/GIMT_Encode2(TargetOpcode::INSERT_SUBREG),
-// CHECK-NEXT:          GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/0, // DstI[dst]
+// CHECK-NEXT:          GIR_BuildRootMI, /*Opcode*/GIMT_Encode2(TargetOpcode::INSERT_SUBREG),
+// CHECK-NEXT:          GIR_RootToRootCopy, /*OpIdx*/0, // DstI[dst]
 // CHECK-NEXT:          GIR_AddSimpleTempRegister, /*InsnID*/0, /*TempRegID*/0,
 // CHECK-NEXT:          GIR_AddSimpleTempRegister, /*InsnID*/0, /*TempRegID*/1,
 // CHECK-NEXT:          GIR_AddImm8, /*InsnID*/0, /*Imm*/1,
 // CHECK-NEXT:          GIR_ConstrainOperandRC, /*InsnID*/0, /*Op*/0, GIMT_Encode2(Test::DRegsRegClassID),
 // CHECK-NEXT:          GIR_ConstrainOperandRC, /*InsnID*/0, /*Op*/1, GIMT_Encode2(Test::DRegsRegClassID),
 // CHECK-NEXT:          GIR_ConstrainOperandRC, /*InsnID*/0, /*Op*/2, GIMT_Encode2(Test::SRegsRegClassID),
-// CHECK-NEXT:          GIR_EraseFromParent, /*InsnID*/0,
+// CHECK-NEXT:          // GIR_Coverage, 6,
+// CHECK-NEXT:          GIR_EraseRootFromParent_Done,
 
 // Test an EXTRACT_SUBREG that is a sub instruction. The individual
 // operands should be constrained to specific register classes, and
@@ -166,7 +170,7 @@ def : Pat<(i16 (trunc (not DOP:$src))),
 // CHECK-NEXT: GIR_CopySubReg, /*NewInsnID*/1, /*OldInsnID*/1, /*OpIdx*/1, /*SubRegIdx*/GIMT_Encode2(1), // src
 // CHECK-NEXT: GIR_ConstrainOperandRC, /*InsnID*/1, /*Op*/0, GIMT_Encode2(Test::SRegsRegClassID),
 // CHECK-NEXT: GIR_ConstrainOperandRC, /*InsnID*/1, /*Op*/1, GIMT_Encode2(Test::DRegsRegClassID),
-// CHECK-NEXT: GIR_BuildMI, /*InsnID*/0, /*Opcode*/GIMT_Encode2(MyTarget::SUBSOME_INSN),
+// CHECK-NEXT: GIR_BuildRootMI, /*Opcode*/GIMT_Encode2(MyTarget::SUBSOME_INSN),
 
 // Test an extract from an output instruction result (nonleaf)
 def : Pat<(i16 (trunc (bitreverse DOP:$src))),
@@ -174,26 +178,27 @@ def : Pat<(i16 (trunc (bitreverse DOP:$src))),
 // CHECK-LABEL: GIM_CheckOpcode, /*MI*/1, GIMT_Encode2(TargetOpcode::G_BITREVERSE),
 // CHECK-NEXT:  GIM_CheckType, /*MI*/1, /*Op*/1, /*Type*/GILLT_s32,
 // CHECK-NEXT:  GIM_CheckRegBankForClass, /*MI*/1, /*Op*/1, /*RC*/GIMT_Encode2(Test::DRegsRegClassID),
-// CHECK-NEXT:  GIM_CheckIsSafeToFold, /*InsnID*/1,
+// CHECK-NEXT:  GIM_CheckIsSafeToFold, /*NumInsns*/1,
 // CHECK-NEXT:  // (trunc:{ *:[i16] } (bitreverse:{ *:[i32] } DOP:{ *:[i32] }:$src))  =>  (EXTRACT_SUBREG:{ *:[i16] } (SOME_INSN:{ *:[i32] } DOP:{ *:[i32] }:$src), sub0:{ *:[i32] })
 // CHECK-NEXT:  GIR_MakeTempReg, /*TempRegID*/0, /*TypeID*/GILLT_s32,
 // CHECK-NEXT:  GIR_BuildMI, /*InsnID*/1, /*Opcode*/GIMT_Encode2(MyTarget::SOME_INSN),
 // CHECK-NEXT:  GIR_AddTempRegister, /*InsnID*/1, /*TempRegID*/0, /*TempRegFlags*/GIMT_Encode2(RegState::Define),
 // CHECK-NEXT:  GIR_Copy, /*NewInsnID*/1, /*OldInsnID*/1, /*OpIdx*/1, // src
 // CHECK-NEXT:  GIR_ConstrainSelectedInstOperands, /*InsnID*/1,
-// CHECK-NEXT:  GIR_BuildMI, /*InsnID*/0, /*Opcode*/GIMT_Encode2(TargetOpcode::COPY),
-// CHECK-NEXT:  GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/0, // DstI[dst]
+// CHECK-NEXT:  GIR_BuildRootMI, /*Opcode*/GIMT_Encode2(TargetOpcode::COPY),
+// CHECK-NEXT:  GIR_RootToRootCopy, /*OpIdx*/0, // DstI[dst]
 // CHECK-NEXT:  GIR_AddTempSubRegister, /*InsnID*/0, /*TempRegID*/0, /*TempRegFlags*/GIMT_Encode2(0), GIMT_Encode2(sub0),
 // CHECK-NEXT:  GIR_ConstrainOperandRC, /*InsnID*/0, /*Op*/0, GIMT_Encode2(Test::SRegsRegClassID),
 // CHECK-NEXT:  GIR_ConstrainOperandRC, /*InsnID*/0, /*Op*/1, GIMT_Encode2(Test::DRegsRegClassID),
-// CHECK-NEXT:  GIR_EraseFromParent, /*InsnID*/0,
+// CHECK-NEXT:  // GIR_Coverage, 8,
+// CHECK-NEXT:  GIR_EraseRootFromParent_Done,
 
 // EXTRACT_SUBREG is subinstruction, but also doesn't have a leaf input
 
 // CHECK-LABEL: GIM_CheckOpcode, /*MI*/1, GIMT_Encode2(TargetOpcode::G_CTPOP),
 // CHECK-NEXT: GIM_CheckType, /*MI*/1, /*Op*/1, /*Type*/GILLT_s32,
 // CHECK-NEXT: GIM_CheckRegBankForClass, /*MI*/1, /*Op*/1, /*RC*/GIMT_Encode2(Test::DRegsRegClassID),
-// CHECK-NEXT: GIM_CheckIsSafeToFold, /*InsnID*/1,
+// CHECK-NEXT: GIM_CheckIsSafeToFold, /*NumInsns*/1,
 // CHECK-NEXT: // (trunc:{ *:[i16] } (ctpop:{ *:[i32] } DOP:{ *:[i32] }:$src))  =>  (SUBSOME_INSN2:{ *:[i16] } (EXTRACT_SUBREG:{ *:[i16] } (SOME_INSN:{ *:[i32] } DOP:{ *:[i32] }:$src), sub0:{ *:[i32] }))
 // CHECK-NEXT: GIR_MakeTempReg, /*TempRegID*/0, /*TypeID*/GILLT_s16,
 // CHECK-NEXT: GIR_MakeTempReg, /*TempRegID*/1, /*TypeID*/GILLT_s32,
@@ -206,11 +211,12 @@ def : Pat<(i16 (trunc (bitreverse DOP:$src))),
 // CHECK-NEXT: GIR_AddTempSubRegister, /*InsnID*/1, /*TempRegID*/1, /*TempRegFlags*/GIMT_Encode2(0), GIMT_Encode2(sub0),
 // CHECK-NEXT: GIR_ConstrainOperandRC, /*InsnID*/1, /*Op*/0, GIMT_Encode2(Test::SRegsRegClassID),
 // CHECK-NEXT: GIR_ConstrainOperandRC, /*InsnID*/1, /*Op*/1, GIMT_Encode2(Test::DRegsRegClassID),
-// CHECK-NEXT: GIR_BuildMI, /*InsnID*/0, /*Opcode*/GIMT_Encode2(MyTarget::SUBSOME_INSN2),
-// CHECK-NEXT: GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/0, // DstI[dst]
+// CHECK-NEXT: GIR_BuildRootMI, /*Opcode*/GIMT_Encode2(MyTarget::SUBSOME_INSN2),
+// CHECK-NEXT: GIR_RootToRootCopy, /*OpIdx*/0, // DstI[dst]
 // CHECK-NEXT: GIR_AddSimpleTempRegister, /*InsnID*/0, /*TempRegID*/0,
-// CHECK-NEXT: GIR_ConstrainSelectedInstOperands, /*InsnID*/0,
-// CHECK-NEXT: GIR_EraseFromParent, /*InsnID*/0,
+// CHECK-NEXT: GIR_RootConstrainSelectedInstOperands,
+// CHECK-NEXT: // GIR_Coverage, 9,
+// CHECK-NEXT: GIR_EraseRootFromParent_Done,
 def : Pat<(i16 (trunc (ctpop DOP:$src))),
            (SUBSOME_INSN2 (EXTRACT_SUBREG (SOME_INSN DOP:$src), sub0))>;
 
@@ -218,13 +224,13 @@ def : Pat<(i16 (trunc (ctpop DOP:$src))),
 def : Pat<(i16 (trunc DOP:$src)),
            (EXTRACT_SUBREG DOP:$src, sub0)>;
 // CHECK-LABEL: // (trunc:{ *:[i16] } DOP:{ *:[i32] }:$src)  =>  (EXTRACT_SUBREG:{ *:[i16] } DOP:{ *:[i32] }:$src, sub0:{ *:[i32] })
-// CHECK-NEXT: GIR_BuildMI, /*InsnID*/0, /*Opcode*/GIMT_Encode2(TargetOpcode::COPY),
-// CHECK-NEXT: GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/0, // DstI[dst]
+// CHECK-NEXT: GIR_BuildRootMI, /*Opcode*/GIMT_Encode2(TargetOpcode::COPY),
+// CHECK-NEXT: GIR_RootToRootCopy, /*OpIdx*/0, // DstI[dst]
 // CHECK-NEXT: GIR_CopySubReg, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/1, /*SubRegIdx*/GIMT_Encode2(1), // src
 // CHECK-NEXT: GIR_ConstrainOperandRC, /*InsnID*/0, /*Op*/0, GIMT_Encode2(Test::SRegsRegClassID),
 // CHECK-NEXT: GIR_ConstrainOperandRC, /*InsnID*/0, /*Op*/1, GIMT_Encode2(Test::DRegsRegClassID),
-// CHECK-NEXT: GIR_EraseFromParent, /*InsnID*/0,
-
+// CHECK-NEXT: // GIR_Coverage, 10,
+// CHECK-NEXT: GIR_EraseRootFromParent_Done,
 
 // Test that we can import SUBREG_TO_REG
 def : Pat<(i32 (zext SOP:$src)),
@@ -235,11 +241,12 @@ def : Pat<(i32 (zext SOP:$src)),
 // CHECK-NEXT:        GIR_AddTempRegister, /*InsnID*/1, /*TempRegID*/0, /*TempRegFlags*/GIMT_Encode2(RegState::Define),
 // CHECK-NEXT:        GIR_Copy, /*NewInsnID*/1, /*OldInsnID*/0, /*OpIdx*/1, // src
 // CHECK-NEXT:        GIR_ConstrainSelectedInstOperands, /*InsnID*/1,
-// CHECK-NEXT:        GIR_BuildMI, /*InsnID*/0, /*Opcode*/GIMT_Encode2(TargetOpcode::SUBREG_TO_REG),
-// CHECK-NEXT:        GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/0, // DstI[dst]
+// CHECK-NEXT:        GIR_BuildRootMI, /*Opcode*/GIMT_Encode2(TargetOpcode::SUBREG_TO_REG),
+// CHECK-NEXT:        GIR_RootToRootCopy, /*OpIdx*/0, // DstI[dst]
 // CHECK-NEXT:        GIR_AddImm8, /*InsnID*/0, /*Imm*/0,
 // CHECK-NEXT:        GIR_AddSimpleTempRegister, /*InsnID*/0, /*TempRegID*/0,
 // CHECK-NEXT:        GIR_AddImm8, /*InsnID*/0, /*Imm*/1,
 // CHECK-NEXT:        GIR_ConstrainOperandRC, /*InsnID*/0, /*Op*/0, GIMT_Encode2(Test::DRegsRegClassID),
 // CHECK-NEXT:        GIR_ConstrainOperandRC, /*InsnID*/0, /*Op*/2, GIMT_Encode2(Test::SRegsRegClassID),
-// CHECK-NEXT:        GIR_EraseFromParent, /*InsnID*/0,
+// CHECK-NEXT:        // GIR_Coverage, 11,
+// CHECK-NEXT:        GIR_EraseRootFromParent_Done,
diff --git a/llvm/test/TableGen/GlobalISelEmitterVariadic.td b/llvm/test/TableGen/GlobalISelEmitterVariadic.td
index ba6a93bd4837..992e1a4b907c 100644
--- a/llvm/test/TableGen/GlobalISelEmitterVariadic.td
+++ b/llvm/test/TableGen/GlobalISelEmitterVariadic.td
@@ -26,27 +26,27 @@ def : Pat<(build_vector GPR32:$src1, GPR32:$src2),
 // CHECK-NEXT:    GIM_CheckOpcode, /*MI*/0, GIMT_Encode2(TargetOpcode::G_BUILD_VECTOR),
 // CHECK-NEXT:    GIM_Try, /*On fail goto*//*Label 1*/ GIMT_Encode4([[NEXT_NUM_OPERANDS_LABEL_1:[0-9]+]]), // Rule ID 0 //
 // CHECK-NEXT:      GIM_CheckNumOperands, /*MI*/0, /*Expected*/2,
-// CHECK-NEXT:      GIM_CheckType, /*MI*/0, /*Op*/0, /*Type*/GILLT_s32,
-// CHECK-NEXT:      GIM_CheckType, /*MI*/0, /*Op*/1, /*Type*/GILLT_s32,
-// CHECK-NEXT:      GIM_CheckRegBankForClass, /*MI*/0, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
-// CHECK-NEXT:      GIM_CheckRegBankForClass, /*MI*/0, /*Op*/1, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
+// CHECK-NEXT:      GIM_RootCheckType, /*Op*/0, /*Type*/GILLT_s32,
+// CHECK-NEXT:      GIM_RootCheckType, /*Op*/1, /*Type*/GILLT_s32,
+// CHECK-NEXT:      GIM_RootCheckRegBankForClass, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
+// CHECK-NEXT:      GIM_RootCheckRegBankForClass, /*Op*/1, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
 // CHECK-NEXT:      // (build_vector:{ *:[i32] } GPR32:{ *:[i32] }:$src1)  =>  (ONE:{ *:[i32] } GPR32:{ *:[i32] }:$src1)
 // CHECK-NEXT:      GIR_MutateOpcode, /*InsnID*/0, /*RecycleInsnID*/0, /*Opcode*/GIMT_Encode2(MyTarget::ONE),
-// CHECK-NEXT:      GIR_ConstrainSelectedInstOperands, /*InsnID*/0,
+// CHECK-NEXT:      GIR_RootConstrainSelectedInstOperands,
 // CHECK-NEXT:      // GIR_Coverage, 0,
 // CHECK-NEXT:      GIR_Done,
 // CHECK-NEXT:    // Label 1: @[[NEXT_NUM_OPERANDS_LABEL_1]]
 // CHECK-NEXT:    GIM_Try, /*On fail goto*//*Label 2*/ GIMT_Encode4([[NEXT_NUM_OPERANDS_LABEL_2:[0-9]+]]), // Rule ID 1 //
 // CHECK-NEXT:      GIM_CheckNumOperands, /*MI*/0, /*Expected*/3,
-// CHECK-NEXT:      GIM_CheckType, /*MI*/0, /*Op*/0, /*Type*/GILLT_s32,
-// CHECK-NEXT:      GIM_CheckType, /*MI*/0, /*Op*/1, /*Type*/GILLT_s32,
-// CHECK-NEXT:      GIM_CheckType, /*MI*/0, /*Op*/2, /*Type*/GILLT_s32,
-// CHECK-NEXT:      GIM_CheckRegBankForClass, /*MI*/0, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
-// CHECK-NEXT:      GIM_CheckRegBankForClass, /*MI*/0, /*Op*/1, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
-// CHECK-NEXT:      GIM_CheckRegBankForClass, /*MI*/0, /*Op*/2, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
+// CHECK-NEXT:      GIM_RootCheckType, /*Op*/0, /*Type*/GILLT_s32,
+// CHECK-NEXT:      GIM_RootCheckType, /*Op*/1, /*Type*/GILLT_s32,
+// CHECK-NEXT:      GIM_RootCheckType, /*Op*/2, /*Type*/GILLT_s32,
+// CHECK-NEXT:      GIM_RootCheckRegBankForClass, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
+// CHECK-NEXT:      GIM_RootCheckRegBankForClass, /*Op*/1, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
+// CHECK-NEXT:      GIM_RootCheckRegBankForClass, /*Op*/2, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
 // CHECK-NEXT:      // (build_vector:{ *:[i32] } GPR32:{ *:[i32] }:$src1, GPR32:{ *:[i32] }:$src2)  =>  (TWO:{ *:[i32] } GPR32:{ *:[i32] }:$src1, GPR32:{ *:[i32] }:$src2)
 // CHECK-NEXT:      GIR_MutateOpcode, /*InsnID*/0, /*RecycleInsnID*/0, /*Opcode*/GIMT_Encode2(MyTarget::TWO),
-// CHECK-NEXT:      GIR_ConstrainSelectedInstOperands, /*InsnID*/0,
+// CHECK-NEXT:      GIR_RootConstrainSelectedInstOperands,
 // CHECK-NEXT:      // GIR_Coverage, 1,
 // CHECK-NEXT:      GIR_Done,
 // CHECK-NEXT:    // Label 2: @[[NEXT_NUM_OPERANDS_LABEL_2]]
diff --git a/llvm/test/TableGen/HasNoUse.td b/llvm/test/TableGen/HasNoUse.td
index 4fe10cd44e4f..030598d1cbee 100644
--- a/llvm/test/TableGen/HasNoUse.td
+++ b/llvm/test/TableGen/HasNoUse.td
@@ -17,19 +17,20 @@ def NO_RET_ATOMIC_ADD : I<(outs), (ins GPR32Op:$src0, GPR32Op:$src1), []>;
 // SDAG-NEXT: return true;
 
 // GISEL: GIM_CheckOpcode, /*MI*/0, GIMT_Encode2(TargetOpcode::G_ATOMICRMW_ADD),
-// GISEL-NEXT: GIM_CheckType, /*MI*/0, /*Op*/0, /*Type*/GILLT_s32,
-// GISEL-NEXT: GIM_CheckType, /*MI*/0, /*Op*/2, /*Type*/GILLT_s32,
+// GISEL-NEXT: GIM_RootCheckType, /*Op*/0, /*Type*/GILLT_s32,
+// GISEL-NEXT: GIM_RootCheckType, /*Op*/2, /*Type*/GILLT_s32,
 // GISEL-NEXT: GIM_CheckMemorySizeEqualTo, /*MI*/0, /*MMO*/0, /*Size*/GIMT_Encode4(4),
 // GISEL-NEXT: GIM_CheckHasNoUse, /*MI*/0,
 // GISEL-NEXT: // MIs[0] src0
 // GISEL-NEXT: GIM_CheckPointerToAny, /*MI*/0, /*Op*/1, /*SizeInBits*/0,
 // GISEL-NEXT: // (atomic_load_add:{ *:[i32] } iPTR:{ *:[iPTR] }:$src0, i32:{ *:[i32] }:$src1)<<P:Predicate_atomic_load_add_no_ret_32>>  =>  (NO_RET_ATOMIC_ADD GPR32:{ *:[i32] }:$src0, GPR32:{ *:[i32] }:$src1)
-// GISEL-NEXT: GIR_BuildMI, /*InsnID*/0, /*Opcode*/GIMT_Encode2(MyTarget::NO_RET_ATOMIC_ADD),
-// GISEL-NEXT: GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/1, // src0
-// GISEL-NEXT: GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/2, // src1
+// GISEL-NEXT: GIR_BuildRootMI, /*Opcode*/GIMT_Encode2(MyTarget::NO_RET_ATOMIC_ADD),
+// GISEL-NEXT: GIR_RootToRootCopy, /*OpIdx*/1, // src0
+// GISEL-NEXT: GIR_RootToRootCopy, /*OpIdx*/2, // src1
 // GISEL-NEXT: GIR_MergeMemOperands, /*InsnID*/0, /*NumInsns*/1, /*MergeInsnID's*/0,
-// GISEL-NEXT: GIR_ConstrainSelectedInstOperands, /*InsnID*/0,
-// GISEL-NEXT: GIR_EraseFromParent, /*InsnID*/0,
+// GISEL-NEXT: GIR_RootConstrainSelectedInstOperands,
+// GISEL-NEXT: // GIR_Coverage, 0,
+// GISEL-NEXT: GIR_EraseRootFromParent_Done,
 let HasNoUse = true in
 defm atomic_load_add_no_ret : binary_atomic_op<atomic_load_add>;
 
diff --git a/llvm/test/TableGen/address-space-patfrags.td b/llvm/test/TableGen/address-space-patfrags.td
index 46050a70720f..582b97d55a51 100644
--- a/llvm/test/TableGen/address-space-patfrags.td
+++ b/llvm/test/TableGen/address-space-patfrags.td
@@ -102,7 +102,7 @@ def truncstorei16_addrspace : PatFrag<(ops node:$val, node:$ptr),
 // GISEL-NEXT: GIM_CheckMemorySizeLessThanLLT, /*MI*/0, /*MMO*/0, /*OpIdx*/0,
 // GISEL-NEXT: GIM_CheckAtomicOrdering, /*MI*/0, /*Order*/(uint8_t)AtomicOrdering::NotAtomic,
 // GISEL-NEXT: // MIs[0] src0
-// GISEL-NEXT: GIM_CheckType, /*MI*/0, /*Op*/0, /*Type*/GILLT_s32,
+// GISEL-NEXT: GIM_RootCheckType, /*Op*/0, /*Type*/GILLT_s32,
 def : Pat <
   (truncstore GPR32:$src0, GPR32:$src1),
   (inst_c GPR32:$src0, GPR32:$src1)
diff --git a/llvm/test/TableGen/gisel-physreg-input.td b/llvm/test/TableGen/gisel-physreg-input.td
index b0af5b7dd3c1..f19872a331fc 100644
--- a/llvm/test/TableGen/gisel-physreg-input.td
+++ b/llvm/test/TableGen/gisel-physreg-input.td
@@ -28,23 +28,24 @@ class I<dag OOps, dag IOps, list<dag> Pat>
 // GISEL-NEXT: GIM_CheckNumOperands, /*MI*/0, /*Expected*/3,
 // GISEL-NEXT: GIM_CheckOpcode, /*MI*/0, GIMT_Encode2(TargetOpcode::G_ADD),
 // GISEL-NEXT: // MIs[0] DstI[dst]
-// GISEL-NEXT: GIM_CheckType, /*MI*/0, /*Op*/0, /*Type*/GILLT_s32,
-// GISEL-NEXT: GIM_CheckRegBankForClass, /*MI*/0, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
+// GISEL-NEXT: GIM_RootCheckType, /*Op*/0, /*Type*/GILLT_s32,
+// GISEL-NEXT: GIM_RootCheckRegBankForClass, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
 // GISEL-NEXT: // MIs[0] src0
-// GISEL-NEXT: GIM_CheckType, /*MI*/0, /*Op*/1, /*Type*/GILLT_s32,
-// GISEL-NEXT: GIM_CheckRegBankForClass, /*MI*/0, /*Op*/1, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
+// GISEL-NEXT: GIM_RootCheckType, /*Op*/1, /*Type*/GILLT_s32,
+// GISEL-NEXT: GIM_RootCheckRegBankForClass, /*Op*/1, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
 // GISEL-NEXT: // MIs[0] Operand 2
-// GISEL-NEXT: GIM_CheckType, /*MI*/0, /*Op*/2, /*Type*/GILLT_s32,
-// GISEL-NEXT: GIM_CheckRegBankForClass, /*MI*/0, /*Op*/2, /*RC*/GIMT_Encode2(MyTarget::Special32RegClassID),
+// GISEL-NEXT: GIM_RootCheckType, /*Op*/2, /*Type*/GILLT_s32,
+// GISEL-NEXT: GIM_RootCheckRegBankForClass, /*Op*/2, /*RC*/GIMT_Encode2(MyTarget::Special32RegClassID),
 // GISEL-NEXT: // (add:{ *:[i32] } GPR32:{ *:[i32] }:$src0, SPECIAL:{ *:[i32] })  =>  (ADD_PHYS:{ *:[i32] } GPR32:{ *:[i32] }:$src0)
 // GISEL-NEXT: GIR_BuildMI, /*InsnID*/1, /*Opcode*/GIMT_Encode2(TargetOpcode::COPY),
 // GISEL-NEXT: GIR_AddRegister, /*InsnID*/1, GIMT_Encode2(MyTarget::SPECIAL), /*AddRegisterRegFlags*/GIMT_Encode2(RegState::Define),
 // GISEL-NEXT: GIR_Copy, /*NewInsnID*/1, /*OldInsnID*/0, /*OpIdx*/2, // SPECIAL
-// GISEL-NEXT: GIR_BuildMI, /*InsnID*/0, /*Opcode*/GIMT_Encode2(MyTarget::ADD_PHYS),
-// GISEL-NEXT: GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/0, // DstI[dst]
-// GISEL-NEXT: GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/1, // src0
-// GISEL-NEXT: GIR_ConstrainSelectedInstOperands, /*InsnID*/0,
-// GISEL-NEXT: GIR_EraseFromParent, /*InsnID*/0,
+// GISEL-NEXT: GIR_BuildRootMI, /*Opcode*/GIMT_Encode2(MyTarget::ADD_PHYS),
+// GISEL-NEXT: GIR_RootToRootCopy, /*OpIdx*/0, // DstI[dst]
+// GISEL-NEXT: GIR_RootToRootCopy, /*OpIdx*/1, // src0
+// GISEL-NEXT: GIR_RootConstrainSelectedInstOperands,
+// GISEL-NEXT: // GIR_Coverage, 0,
+// GISEL-NEXT: GIR_EraseRootFromParent_Done,
 def ADD_PHYS : I<(outs GPR32:$dst), (ins GPR32:$src0),
     [(set GPR32:$dst, (add GPR32:$src0, SPECIAL))]> {
   let Uses = [SPECIAL];
@@ -56,23 +57,24 @@ def ADD_PHYS : I<(outs GPR32:$dst), (ins GPR32:$src0),
 // GISEL-NEXT: GIM_CheckNumOperands, /*MI*/0, /*Expected*/3,
 // GISEL-NEXT: GIM_CheckOpcode, /*MI*/0, GIMT_Encode2(TargetOpcode::G_MUL),
 // GISEL-NEXT: // MIs[0] DstI[dst]
-// GISEL-NEXT: GIM_CheckType, /*MI*/0, /*Op*/0, /*Type*/GILLT_s32,
-// GISEL-NEXT: GIM_CheckRegBankForClass, /*MI*/0, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
+// GISEL-NEXT: GIM_RootCheckType, /*Op*/0, /*Type*/GILLT_s32,
+// GISEL-NEXT: GIM_RootCheckRegBankForClass, /*Op*/0, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
 // GISEL-NEXT: // MIs[0] SPECIAL
-// GISEL-NEXT: GIM_CheckType, /*MI*/0, /*Op*/1, /*Type*/GILLT_s32,
-// GISEL-NEXT: GIM_CheckRegBankForClass, /*MI*/0, /*Op*/1, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
+// GISEL-NEXT: GIM_RootCheckType, /*Op*/1, /*Type*/GILLT_s32,
+// GISEL-NEXT: GIM_RootCheckRegBankForClass, /*Op*/1, /*RC*/GIMT_Encode2(MyTarget::GPR32RegClassID),
 // GISEL-NEXT: // MIs[0] Operand 2
-// GISEL-NEXT: GIM_CheckType, /*MI*/0, /*Op*/2, /*Type*/GILLT_s32,
-// GISEL-NEXT: GIM_CheckRegBankForClass, /*MI*/0, /*Op*/2, /*RC*/GIMT_Encode2(MyTarget::Special32RegClassID),
+// GISEL-NEXT: GIM_RootCheckType, /*Op*/2, /*Type*/GILLT_s32,
+// GISEL-NEXT: GIM_RootCheckRegBankForClass, /*Op*/2, /*RC*/GIMT_Encode2(MyTarget::Special32RegClassID),
 // GISEL-NEXT: // (mul:{ *:[i32] } GPR32:{ *:[i32] }:$SPECIAL, SPECIAL:{ *:[i32] })  =>  (MUL_PHYS:{ *:[i32] } GPR32:{ *:[i32] }:$SPECIAL)
 // GISEL-NEXT: GIR_BuildMI, /*InsnID*/1, /*Opcode*/GIMT_Encode2(TargetOpcode::COPY),
 // GISEL-NEXT: GIR_AddRegister, /*InsnID*/1, GIMT_Encode2(MyTarget::SPECIAL), /*AddRegisterRegFlags*/GIMT_Encode2(RegState::Define),
 // GISEL-NEXT: GIR_Copy, /*NewInsnID*/1, /*OldInsnID*/0, /*OpIdx*/2, // SPECIAL
-// GISEL-NEXT: GIR_BuildMI, /*InsnID*/0, /*Opcode*/GIMT_Encode2(MyTarget::MUL_PHYS),
-// GISEL-NEXT: GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/0, // DstI[dst]
-// GISEL-NEXT: GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/1, // SPECIAL
-// GISEL-NEXT: GIR_ConstrainSelectedInstOperands, /*InsnID*/0,
-// GISEL-NEXT: GIR_EraseFromParent, /*InsnID*/0,
+// GISEL-NEXT: GIR_BuildRootMI, /*Opcode*/GIMT_Encode2(MyTarget::MUL_PHYS),
+// GISEL-NEXT: GIR_RootToRootCopy, /*OpIdx*/0, // DstI[dst]
+// GISEL-NEXT: GIR_RootToRootCopy, /*OpIdx*/1, // SPECIAL
+// GISEL-NEXT: GIR_RootConstrainSelectedInstOperands,
+// GISEL-NEXT: // GIR_Coverage, 1,
+// GISEL-NEXT: GIR_EraseRootFromParent_Done,
 def MUL_PHYS : I<(outs GPR32:$dst), (ins GPR32:$SPECIAL),
     [(set GPR32:$dst, (mul GPR32:$SPECIAL, SPECIAL))]> {
   let Uses = [SPECIAL];
diff --git a/llvm/test/TableGen/immarg-predicated.td b/llvm/test/TableGen/immarg-predicated.td
index 320018010cc7..dcacb2f8f1de 100644
--- a/llvm/test/TableGen/immarg-predicated.td
+++ b/llvm/test/TableGen/immarg-predicated.td
@@ -14,8 +14,8 @@ def int_mytarget_sleep0 : Intrinsic<[], [llvm_i32_ty], [ImmArg<ArgIndex<0>>]>;
 // GISEL-NEXT: GIM_CheckIsImm, /*MI*/0, /*Op*/1,
 // GISEL-NEXT: GIM_CheckImmOperandPredicate, /*MI*/0, /*MO*/1, /*Predicate*/GIMT_Encode2(GICXXPred_I64_Predicate_tuimm9),
 // GISEL-NEXT: // (intrinsic_void {{[0-9]+}}:{ *:[iPTR] }, (timm:{ *:[i32] })<<P:Predicate_tuimm9>>:$src) => (SLEEP0 (timm:{ *:[i32] }):$src)
-// GISEL-NEXT: GIR_BuildMI, /*InsnID*/0, /*Opcode*/GIMT_Encode2(MyTarget::SLEEP0),
-// GISEL-NEXT: GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/1, // src
+// GISEL-NEXT: GIR_BuildRootMI, /*Opcode*/GIMT_Encode2(MyTarget::SLEEP0),
+// GISEL-NEXT: GIR_RootToRootCopy, /*OpIdx*/1, // src
 def tuimm9 : TImmLeaf<i32, [{ return isUInt<9>(Imm); }]>;
 def SLEEP0 : I<(outs), (ins i32imm:$src),
   [(int_mytarget_sleep0 tuimm9:$src)]
diff --git a/llvm/test/TableGen/immarg.td b/llvm/test/TableGen/immarg.td
index 80849d512bee..e5fd06ce6c08 100644
--- a/llvm/test/TableGen/immarg.td
+++ b/llvm/test/TableGen/immarg.td
@@ -14,8 +14,8 @@ def int_mytarget_sleep1 : Intrinsic<[], [llvm_i32_ty], [ImmArg<ArgIndex<0>>]>;
 // GISEL-NEXT: // MIs[0] src
 // GISEL-NEXT: GIM_CheckIsImm, /*MI*/0, /*Op*/1,
 // GISEL-NEXT: // (intrinsic_void {{[0-9]+}}:{ *:[iPTR] }, (timm:{ *:[i32] }):$src)  =>  (SLEEP0 (timm:{ *:[i32] }):$src)
-// GISEL-NEXT: GIR_BuildMI, /*InsnID*/0, /*Opcode*/GIMT_Encode2(MyTarget::SLEEP0),
-// GISEL-NEXT: GIR_Copy, /*NewInsnID*/0, /*OldInsnID*/0, /*OpIdx*/1, // src
+// GISEL-NEXT: GIR_BuildRootMI, /*Opcode*/GIMT_Encode2(MyTarget::SLEEP0),
+// GISEL-NEXT: GIR_RootToRootCopy, /*OpIdx*/1, // src
 def SLEEP0 : I<(outs), (ins i32imm:$src),
   [(int_mytarget_sleep0 timm:$src)]
 >;
diff --git a/llvm/test/TableGen/riscv-target-def.td b/llvm/test/TableGen/riscv-target-def.td
new file mode 100644
index 000000000000..7f3d9bdb278c
--- /dev/null
+++ b/llvm/test/TableGen/riscv-target-def.td
@@ -0,0 +1,162 @@
+// RUN: llvm-tblgen -gen-riscv-target-def -I %p/../../include %s | FileCheck %s
+
+include "llvm/Target/Target.td"
+
+class RISCVExtension<string name, int major, int minor, string desc,
+                     list<SubtargetFeature> implies = [],
+                     string fieldname = !subst("Feature", "Has", NAME),
+                     string value = "true">
+    : SubtargetFeature<name, fieldname, value, desc, implies> {
+  int MajorVersion = major;
+  int MinorVersion = minor;
+  bit Experimental = false;
+}
+
+class RISCVExperimentalExtension<string name, int major, int minor, string desc,
+                                 list<RISCVExtension> implies = [],
+                                 string fieldname = !subst("Feature", "Has", NAME),
+                                 string value = "true">
+    : RISCVExtension<"experimental-"#name, major, minor, desc, implies,
+                     fieldname, value> {
+  let Experimental = true;
+}
+
+def FeatureStdExtI
+    : RISCVExtension<"i", 2, 1,
+                     "'I' (Base Integer Instruction Set)">;
+
+def FeatureStdExtZicsr
+    : RISCVExtension<"zicsr", 2, 0,
+                     "'zicsr' (CSRs)">;
+
+def FeatureStdExtZifencei
+    : RISCVExtension<"zifencei", 2, 0,
+                     "'Zifencei' (fence.i)">;
+
+def FeatureStdExtF
+    : RISCVExtension<"f", 2, 2,
+                     "'F' (Single-Precision Floating-Point)",
+                     [FeatureStdExtZicsr]>;
+
+def FeatureStdExtZidummy
+    : RISCVExperimentalExtension<"zidummy", 0, 1,
+                                 "Dummy">;
+
+def Feature32Bit
+    : SubtargetFeature<"32bit", "IsRV32", "true", "Implements RV32">;
+def Feature64Bit
+    : SubtargetFeature<"64bit", "IsRV64", "true", "Implements RV64">;
+
+// Dummy feature that isn't an extension.
+def FeatureDummy
+    : SubtargetFeature<"dummy", "Dummy", "true", "Dummy">;
+
+class RISCVProfile<string name, list<SubtargetFeature> features>
+    : SubtargetFeature<name, "Is" # NAME, "true",
+                       "RISC-V " # name # " profile", features>;
+
+def RVI20U32 : RISCVProfile<"rvi20u32", [Feature32Bit, FeatureStdExtI]>;
+def RVI20U64 : RISCVProfile<"rvi20u64", [Feature64Bit, FeatureStdExtI]>;
+def ProfileDummy : RISCVProfile<"dummy", [Feature64Bit, FeatureStdExtI,
+                                          FeatureStdExtF, FeatureStdExtZidummy]>;
+
+class RISCVProcessorModel<string n,
+                          SchedMachineModel m,
+                          list<SubtargetFeature> f,
+                          list<SubtargetFeature> tunef = [],
+                          string default_march = "">
+    :  ProcessorModel<n, m, f, tunef> {
+  string DefaultMarch = default_march;
+}
+
+class RISCVTuneProcessorModel<string n,
+                              SchedMachineModel m,
+                              list<SubtargetFeature> tunef = [],
+                              list<SubtargetFeature> f = []>
+    : ProcessorModel<n, m, f,tunef>;
+
+def GENERIC_RV32 : RISCVProcessorModel<"generic-rv32",
+                                       NoSchedModel,
+                                       [Feature32Bit,
+                                        FeatureStdExtI]>;
+def GENERIC_RV64 : RISCVProcessorModel<"generic-rv64",
+                                       NoSchedModel,
+                                       [Feature64Bit,
+                                        FeatureStdExtI]>;
+def GENERIC : RISCVTuneProcessorModel<"generic", NoSchedModel>;
+
+
+def ROCKET_RV32 : RISCVProcessorModel<"rocket-rv32",
+                                      NoSchedModel,
+                                      [Feature32Bit,
+                                       FeatureStdExtI,
+                                       FeatureStdExtZifencei,
+                                       FeatureStdExtZicsr,
+                                       FeatureStdExtZidummy,
+                                       FeatureDummy]>;
+def ROCKET_RV64 : RISCVProcessorModel<"rocket-rv64",
+                                      NoSchedModel,
+                                      [Feature64Bit,
+                                       FeatureStdExtI,
+                                       FeatureStdExtZifencei,
+                                       FeatureStdExtZicsr,
+                                       FeatureStdExtZidummy,
+                                       FeatureDummy]>;
+def ROCKET : RISCVTuneProcessorModel<"rocket",
+                                     NoSchedModel>;
+
+// CHECK:      #ifdef GET_SUPPORTED_EXTENSIONS
+// CHECK-NEXT: #undef GET_SUPPORTED_EXTENSIONS
+
+// CHECK:      static const RISCVSupportedExtension SupportedExtensions[] = {
+// CHECK-NEXT:     {"f", {2, 2}},
+// CHECK-NEXT:     {"i", {2, 1}},
+// CHECK-NEXT:     {"zicsr", {2, 0}},
+// CHECK-NEXT:     {"zifencei", {2, 0}},
+// CHECK-NEXT: };
+
+// CHECK:      static const RISCVSupportedExtension SupportedExperimentalExtensions[] = {
+// CHECK-NEXT:     {"zidummy", {0, 1}},
+// CHECK-NEXT: };
+
+// CHECK:      #endif // GET_SUPPORTED_EXTENSIONS
+
+// CHECK:      #ifdef GET_IMPLIED_EXTENSIONS
+// CHECK-NEXT: #undef GET_IMPLIED_EXTENSIONS
+
+// CHECK:      static constexpr ImpliedExtsEntry ImpliedExts[] = {
+// CHECK-NEXT:       { {"f"}, "zicsr"},
+// CHECK-NEXT: };
+
+// CHECK:      #endif // GET_IMPLIED_EXTENSIONS
+
+// CHECK:      #ifdef GET_SUPPORTED_PROFILES
+// CHECK-NEXT: #undef GET_SUPPORTED_PROFILES
+
+// CHECK:      static constexpr RISCVProfile SupportedProfiles[] = {
+// CHECK-NEXT:     {"dummy","rv64i2p1_f2p2_zidummy0p1"},
+// CHECK-NEXT:     {"rvi20u32","rv32i2p1"},
+// CHECK-NEXT:     {"rvi20u64","rv64i2p1"},
+// CHECK-NEXT: };
+
+// CHECK:      #endif // GET_SUPPORTED_PROFILES
+
+// CHECK:      #ifndef PROC
+// CHECK-NEXT: #define PROC(ENUM, NAME, DEFAULT_MARCH, FAST_UNALIGNED_ACCESS)
+// CHECK-NEXT: #endif
+
+// CHECK:      PROC(GENERIC_RV32, {"generic-rv32"}, {"rv32i2p1"}, 0)
+// CHECK-NEXT: PROC(GENERIC_RV64, {"generic-rv64"}, {"rv64i2p1"}, 0)
+// CHECK-NEXT: PROC(ROCKET_RV32, {"rocket-rv32"}, {"rv32i2p1_zicsr2p0_zidummy0p1_zifencei2p0"}, 0)
+// CHECK-NEXT: PROC(ROCKET_RV64, {"rocket-rv64"}, {"rv64i2p1_zicsr2p0_zidummy0p1_zifencei2p0"}, 0)
+
+// CHECK: #undef PROC
+
+// CHECK:      #ifndef TUNE_PROC
+// CHECK-NEXT: #define TUNE_PROC(ENUM, NAME)
+// CHECK-NEXT: #endif
+
+// CHECK: TUNE_PROC(GENERIC, "generic")
+// CHECK-NEXT: TUNE_PROC(ROCKET, "rocket")
+
+// CHECK: #undef TUNE_PROC
diff --git a/llvm/test/TableGen/simplify-patfrag.td b/llvm/test/TableGen/simplify-patfrag.td
index 904c29696a6e..fbb6f97f2863 100644
--- a/llvm/test/TableGen/simplify-patfrag.td
+++ b/llvm/test/TableGen/simplify-patfrag.td
@@ -1,4 +1,5 @@
 // RUN: llvm-tblgen -gen-dag-isel -I %p/../../include %s 2>&1 | FileCheck %s
+// RUN: llvm-tblgen -gen-dag-isel -I %p/../../include -DIGNORE %s 2>&1 | FileCheck %s
 
 include "llvm/Target/Target.td"
 
@@ -29,6 +30,10 @@ def anyconvert : PatFrags<(ops node:$src),
                           [(bitconvert node:$src),
                            (specialconvert node:$src)]>;
 
+#ifdef IGNORE
+// Ensure ShouldIgnore does not disable records in dag isel emitter
+let GISelShouldIgnore = 1 in
+#endif
 // And a rule that matches that PatFrag and turns it into i2f
 def : Pat<(f32 (anyconvert (i32 GPR:$val))), (i2f GPR:$val)>;
 
diff --git a/llvm/test/Transforms/AtomicExpand/AMDGPU/expand-atomic-i16-system.ll b/llvm/test/Transforms/AtomicExpand/AMDGPU/expand-atomic-i16-system.ll
index 08d990fb5803..78468b933ff5 100644
--- a/llvm/test/Transforms/AtomicExpand/AMDGPU/expand-atomic-i16-system.ll
+++ b/llvm/test/Transforms/AtomicExpand/AMDGPU/expand-atomic-i16-system.ll
@@ -152,7 +152,7 @@ define i16 @test_atomicrmw_and_i16_global_system(ptr addrspace(1) %ptr, i16 %val
 ; CHECK-NEXT:    [[INV_MASK:%.*]] = xor i32 [[MASK]], -1
 ; CHECK-NEXT:    [[TMP3:%.*]] = zext i16 [[VALUE:%.*]] to i32
 ; CHECK-NEXT:    [[VALOPERAND_SHIFTED:%.*]] = shl i32 [[TMP3]], [[SHIFTAMT]]
-; CHECK-NEXT:    [[ANDOPERAND:%.*]] = or i32 [[INV_MASK]], [[VALOPERAND_SHIFTED]]
+; CHECK-NEXT:    [[ANDOPERAND:%.*]] = or i32 [[VALOPERAND_SHIFTED]], [[INV_MASK]]
 ; CHECK-NEXT:    [[TMP4:%.*]] = atomicrmw and ptr addrspace(1) [[ALIGNEDADDR]], i32 [[ANDOPERAND]] seq_cst, align 4
 ; CHECK-NEXT:    [[SHIFTED:%.*]] = lshr i32 [[TMP4]], [[SHIFTAMT]]
 ; CHECK-NEXT:    [[EXTRACTED:%.*]] = trunc i32 [[SHIFTED]] to i16
diff --git a/llvm/test/Transforms/AtomicExpand/AMDGPU/expand-atomic-i16.ll b/llvm/test/Transforms/AtomicExpand/AMDGPU/expand-atomic-i16.ll
index 94f1b733877e..324b6d2f6596 100644
--- a/llvm/test/Transforms/AtomicExpand/AMDGPU/expand-atomic-i16.ll
+++ b/llvm/test/Transforms/AtomicExpand/AMDGPU/expand-atomic-i16.ll
@@ -153,7 +153,7 @@ define i16 @test_atomicrmw_and_i16_global_agent(ptr addrspace(1) %ptr, i16 %valu
 ; CHECK-NEXT:    [[INV_MASK:%.*]] = xor i32 [[MASK]], -1
 ; CHECK-NEXT:    [[TMP3:%.*]] = zext i16 [[VALUE:%.*]] to i32
 ; CHECK-NEXT:    [[VALOPERAND_SHIFTED:%.*]] = shl i32 [[TMP3]], [[SHIFTAMT]]
-; CHECK-NEXT:    [[ANDOPERAND:%.*]] = or i32 [[INV_MASK]], [[VALOPERAND_SHIFTED]]
+; CHECK-NEXT:    [[ANDOPERAND:%.*]] = or i32 [[VALOPERAND_SHIFTED]], [[INV_MASK]]
 ; CHECK-NEXT:    [[TMP4:%.*]] = atomicrmw and ptr addrspace(1) [[ALIGNEDADDR]], i32 [[ANDOPERAND]] syncscope("agent") seq_cst, align 4
 ; CHECK-NEXT:    [[SHIFTED:%.*]] = lshr i32 [[TMP4]], [[SHIFTAMT]]
 ; CHECK-NEXT:    [[EXTRACTED:%.*]] = trunc i32 [[SHIFTED]] to i16
@@ -918,7 +918,7 @@ define half @test_atomicrmw_xchg_f16_global_agent(ptr addrspace(1) %ptr, half %v
 ; CHECK-NEXT:    [[LOADED:%.*]] = phi i32 [ [[TMP5]], [[TMP0:%.*]] ], [ [[NEWLOADED:%.*]], [[ATOMICRMW_START]] ]
 ; CHECK-NEXT:    [[TMP6:%.*]] = and i32 [[LOADED]], [[INV_MASK]]
 ; CHECK-NEXT:    [[TMP7:%.*]] = or i32 [[TMP6]], [[VALOPERAND_SHIFTED]]
-; CHECK-NEXT:    [[TMP8:%.*]] = cmpxchg ptr addrspace(1) [[ALIGNEDADDR]], i32 [[LOADED]], i32 [[TMP7]] seq_cst seq_cst, align 4
+; CHECK-NEXT:    [[TMP8:%.*]] = cmpxchg ptr addrspace(1) [[ALIGNEDADDR]], i32 [[LOADED]], i32 [[TMP7]] syncscope("agent") seq_cst seq_cst, align 4
 ; CHECK-NEXT:    [[SUCCESS:%.*]] = extractvalue { i32, i1 } [[TMP8]], 1
 ; CHECK-NEXT:    [[NEWLOADED]] = extractvalue { i32, i1 } [[TMP8]], 0
 ; CHECK-NEXT:    br i1 [[SUCCESS]], label [[ATOMICRMW_END:%.*]], label [[ATOMICRMW_START]]
@@ -942,7 +942,7 @@ define half @test_atomicrmw_xchg_f16_global_agent_align4(ptr addrspace(1) %ptr,
 ; CHECK-NEXT:    [[LOADED:%.*]] = phi i32 [ [[TMP3]], [[TMP0:%.*]] ], [ [[NEWLOADED:%.*]], [[ATOMICRMW_START]] ]
 ; CHECK-NEXT:    [[TMP4:%.*]] = and i32 [[LOADED]], -65536
 ; CHECK-NEXT:    [[TMP5:%.*]] = or i32 [[TMP4]], [[TMP2]]
-; CHECK-NEXT:    [[TMP6:%.*]] = cmpxchg ptr addrspace(1) [[PTR]], i32 [[LOADED]], i32 [[TMP5]] seq_cst seq_cst, align 4
+; CHECK-NEXT:    [[TMP6:%.*]] = cmpxchg ptr addrspace(1) [[PTR]], i32 [[LOADED]], i32 [[TMP5]] syncscope("agent") seq_cst seq_cst, align 4
 ; CHECK-NEXT:    [[SUCCESS:%.*]] = extractvalue { i32, i1 } [[TMP6]], 1
 ; CHECK-NEXT:    [[NEWLOADED]] = extractvalue { i32, i1 } [[TMP6]], 0
 ; CHECK-NEXT:    br i1 [[SUCCESS]], label [[ATOMICRMW_END:%.*]], label [[ATOMICRMW_START]]
@@ -973,7 +973,7 @@ define half @test_atomicrmw_xchg_f16_flat_agent(ptr %ptr, half %value) {
 ; CHECK-NEXT:    [[LOADED:%.*]] = phi i32 [ [[TMP5]], [[TMP0:%.*]] ], [ [[NEWLOADED:%.*]], [[ATOMICRMW_START]] ]
 ; CHECK-NEXT:    [[TMP6:%.*]] = and i32 [[LOADED]], [[INV_MASK]]
 ; CHECK-NEXT:    [[TMP7:%.*]] = or i32 [[TMP6]], [[VALOPERAND_SHIFTED]]
-; CHECK-NEXT:    [[TMP8:%.*]] = cmpxchg ptr [[ALIGNEDADDR]], i32 [[LOADED]], i32 [[TMP7]] seq_cst seq_cst, align 4
+; CHECK-NEXT:    [[TMP8:%.*]] = cmpxchg ptr [[ALIGNEDADDR]], i32 [[LOADED]], i32 [[TMP7]] syncscope("agent") seq_cst seq_cst, align 4
 ; CHECK-NEXT:    [[SUCCESS:%.*]] = extractvalue { i32, i1 } [[TMP8]], 1
 ; CHECK-NEXT:    [[NEWLOADED]] = extractvalue { i32, i1 } [[TMP8]], 0
 ; CHECK-NEXT:    br i1 [[SUCCESS]], label [[ATOMICRMW_END:%.*]], label [[ATOMICRMW_START]]
@@ -997,7 +997,7 @@ define half @test_atomicrmw_xchg_f16_flat_agent_align4(ptr %ptr, half %value) {
 ; CHECK-NEXT:    [[LOADED:%.*]] = phi i32 [ [[TMP3]], [[TMP0:%.*]] ], [ [[NEWLOADED:%.*]], [[ATOMICRMW_START]] ]
 ; CHECK-NEXT:    [[TMP4:%.*]] = and i32 [[LOADED]], -65536
 ; CHECK-NEXT:    [[TMP5:%.*]] = or i32 [[TMP4]], [[TMP2]]
-; CHECK-NEXT:    [[TMP6:%.*]] = cmpxchg ptr [[PTR]], i32 [[LOADED]], i32 [[TMP5]] seq_cst seq_cst, align 4
+; CHECK-NEXT:    [[TMP6:%.*]] = cmpxchg ptr [[PTR]], i32 [[LOADED]], i32 [[TMP5]] syncscope("agent") seq_cst seq_cst, align 4
 ; CHECK-NEXT:    [[SUCCESS:%.*]] = extractvalue { i32, i1 } [[TMP6]], 1
 ; CHECK-NEXT:    [[NEWLOADED]] = extractvalue { i32, i1 } [[TMP6]], 0
 ; CHECK-NEXT:    br i1 [[SUCCESS]], label [[ATOMICRMW_END:%.*]], label [[ATOMICRMW_START]]
@@ -1028,7 +1028,7 @@ define bfloat @test_atomicrmw_xchg_bf16_global_agent(ptr addrspace(1) %ptr, bflo
 ; CHECK-NEXT:    [[LOADED:%.*]] = phi i32 [ [[TMP5]], [[TMP0:%.*]] ], [ [[NEWLOADED:%.*]], [[ATOMICRMW_START]] ]
 ; CHECK-NEXT:    [[TMP6:%.*]] = and i32 [[LOADED]], [[INV_MASK]]
 ; CHECK-NEXT:    [[TMP7:%.*]] = or i32 [[TMP6]], [[VALOPERAND_SHIFTED]]
-; CHECK-NEXT:    [[TMP8:%.*]] = cmpxchg ptr addrspace(1) [[ALIGNEDADDR]], i32 [[LOADED]], i32 [[TMP7]] seq_cst seq_cst, align 4
+; CHECK-NEXT:    [[TMP8:%.*]] = cmpxchg ptr addrspace(1) [[ALIGNEDADDR]], i32 [[LOADED]], i32 [[TMP7]] syncscope("agent") seq_cst seq_cst, align 4
 ; CHECK-NEXT:    [[SUCCESS:%.*]] = extractvalue { i32, i1 } [[TMP8]], 1
 ; CHECK-NEXT:    [[NEWLOADED]] = extractvalue { i32, i1 } [[TMP8]], 0
 ; CHECK-NEXT:    br i1 [[SUCCESS]], label [[ATOMICRMW_END:%.*]], label [[ATOMICRMW_START]]
@@ -1052,7 +1052,7 @@ define bfloat @test_atomicrmw_xchg_bf16_global_agent_align4(ptr addrspace(1) %pt
 ; CHECK-NEXT:    [[LOADED:%.*]] = phi i32 [ [[TMP3]], [[TMP0:%.*]] ], [ [[NEWLOADED:%.*]], [[ATOMICRMW_START]] ]
 ; CHECK-NEXT:    [[TMP4:%.*]] = and i32 [[LOADED]], -65536
 ; CHECK-NEXT:    [[TMP5:%.*]] = or i32 [[TMP4]], [[TMP2]]
-; CHECK-NEXT:    [[TMP6:%.*]] = cmpxchg ptr addrspace(1) [[PTR]], i32 [[LOADED]], i32 [[TMP5]] seq_cst seq_cst, align 4
+; CHECK-NEXT:    [[TMP6:%.*]] = cmpxchg ptr addrspace(1) [[PTR]], i32 [[LOADED]], i32 [[TMP5]] syncscope("agent") seq_cst seq_cst, align 4
 ; CHECK-NEXT:    [[SUCCESS:%.*]] = extractvalue { i32, i1 } [[TMP6]], 1
 ; CHECK-NEXT:    [[NEWLOADED]] = extractvalue { i32, i1 } [[TMP6]], 0
 ; CHECK-NEXT:    br i1 [[SUCCESS]], label [[ATOMICRMW_END:%.*]], label [[ATOMICRMW_START]]
diff --git a/llvm/test/Transforms/AtomicExpand/AMDGPU/expand-atomic-i8-system.ll b/llvm/test/Transforms/AtomicExpand/AMDGPU/expand-atomic-i8-system.ll
index 80cf19ed8c63..088371f461ec 100644
--- a/llvm/test/Transforms/AtomicExpand/AMDGPU/expand-atomic-i8-system.ll
+++ b/llvm/test/Transforms/AtomicExpand/AMDGPU/expand-atomic-i8-system.ll
@@ -161,7 +161,7 @@ define i8 @test_atomicrmw_and_i8_global_system(ptr addrspace(1) %ptr, i8 %value)
 ; CHECK-NEXT:    [[INV_MASK:%.*]] = xor i32 [[MASK]], -1
 ; CHECK-NEXT:    [[TMP3:%.*]] = zext i8 [[VALUE:%.*]] to i32
 ; CHECK-NEXT:    [[VALOPERAND_SHIFTED:%.*]] = shl i32 [[TMP3]], [[SHIFTAMT]]
-; CHECK-NEXT:    [[ANDOPERAND:%.*]] = or i32 [[INV_MASK]], [[VALOPERAND_SHIFTED]]
+; CHECK-NEXT:    [[ANDOPERAND:%.*]] = or i32 [[VALOPERAND_SHIFTED]], [[INV_MASK]]
 ; CHECK-NEXT:    [[TMP4:%.*]] = atomicrmw and ptr addrspace(1) [[ALIGNEDADDR]], i32 [[ANDOPERAND]] seq_cst, align 4
 ; CHECK-NEXT:    [[SHIFTED:%.*]] = lshr i32 [[TMP4]], [[SHIFTAMT]]
 ; CHECK-NEXT:    [[EXTRACTED:%.*]] = trunc i32 [[SHIFTED]] to i8
diff --git a/llvm/test/Transforms/AtomicExpand/AMDGPU/expand-atomic-i8.ll b/llvm/test/Transforms/AtomicExpand/AMDGPU/expand-atomic-i8.ll
index 711580158e72..97651c8d23a1 100644
--- a/llvm/test/Transforms/AtomicExpand/AMDGPU/expand-atomic-i8.ll
+++ b/llvm/test/Transforms/AtomicExpand/AMDGPU/expand-atomic-i8.ll
@@ -264,7 +264,7 @@ define i8 @test_atomicrmw_and_i8_global_agent(ptr addrspace(1) %ptr, i8 %value)
 ; GCN-NEXT:    [[INV_MASK:%.*]] = xor i32 [[MASK]], -1
 ; GCN-NEXT:    [[TMP3:%.*]] = zext i8 [[VALUE:%.*]] to i32
 ; GCN-NEXT:    [[VALOPERAND_SHIFTED:%.*]] = shl i32 [[TMP3]], [[SHIFTAMT]]
-; GCN-NEXT:    [[ANDOPERAND:%.*]] = or i32 [[INV_MASK]], [[VALOPERAND_SHIFTED]]
+; GCN-NEXT:    [[ANDOPERAND:%.*]] = or i32 [[VALOPERAND_SHIFTED]], [[INV_MASK]]
 ; GCN-NEXT:    [[TMP4:%.*]] = atomicrmw and ptr addrspace(1) [[ALIGNEDADDR]], i32 [[ANDOPERAND]] syncscope("agent") seq_cst, align 4
 ; GCN-NEXT:    [[SHIFTED:%.*]] = lshr i32 [[TMP4]], [[SHIFTAMT]]
 ; GCN-NEXT:    [[EXTRACTED:%.*]] = trunc i32 [[SHIFTED]] to i8
@@ -279,7 +279,7 @@ define i8 @test_atomicrmw_and_i8_global_agent(ptr addrspace(1) %ptr, i8 %value)
 ; R600-NEXT:    [[INV_MASK:%.*]] = xor i32 [[MASK]], -1
 ; R600-NEXT:    [[TMP3:%.*]] = zext i8 [[VALUE:%.*]] to i32
 ; R600-NEXT:    [[VALOPERAND_SHIFTED:%.*]] = shl i32 [[TMP3]], [[TMP2]]
-; R600-NEXT:    [[ANDOPERAND:%.*]] = or i32 [[INV_MASK]], [[VALOPERAND_SHIFTED]]
+; R600-NEXT:    [[ANDOPERAND:%.*]] = or i32 [[VALOPERAND_SHIFTED]], [[INV_MASK]]
 ; R600-NEXT:    [[TMP4:%.*]] = atomicrmw and ptr addrspace(1) [[ALIGNEDADDR]], i32 [[ANDOPERAND]] syncscope("agent") seq_cst, align 4
 ; R600-NEXT:    [[SHIFTED:%.*]] = lshr i32 [[TMP4]], [[TMP2]]
 ; R600-NEXT:    [[EXTRACTED:%.*]] = trunc i32 [[SHIFTED]] to i8
diff --git a/llvm/test/Transforms/AtomicExpand/AMDGPU/expand-atomic-mmra.ll b/llvm/test/Transforms/AtomicExpand/AMDGPU/expand-atomic-mmra.ll
new file mode 100644
index 000000000000..d51e9291a611
--- /dev/null
+++ b/llvm/test/Transforms/AtomicExpand/AMDGPU/expand-atomic-mmra.ll
@@ -0,0 +1,204 @@
+; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --version 4
+
+; RUN: opt -S -mtriple=amdgcn-amd-amdhsa -mcpu=gfx90a -verify-each -atomic-expand %s | FileCheck -check-prefix=GFX90A %s
+; RUN: opt -S -mtriple=amdgcn-amd-amdhsa -mcpu=gfx1100 -verify-each -atomic-expand %s | FileCheck -check-prefix=GFX1100 %s
+
+; Contains a variety of tests with different types of atomic expansions to check that MMRAs are
+; preserved.
+
+define i16 @test_atomicrmw_xchg_i16_global_agent(ptr addrspace(1) %ptr, i16 %value) {
+; GFX90A-LABEL: define i16 @test_atomicrmw_xchg_i16_global_agent(
+; GFX90A-SAME: ptr addrspace(1) [[PTR:%.*]], i16 [[VALUE:%.*]]) #[[ATTR0:[0-9]+]] {
+; GFX90A-NEXT:    [[ALIGNEDADDR:%.*]] = call ptr addrspace(1) @llvm.ptrmask.p1.i64(ptr addrspace(1) [[PTR]], i64 -4)
+; GFX90A-NEXT:    [[TMP1:%.*]] = ptrtoint ptr addrspace(1) [[PTR]] to i64
+; GFX90A-NEXT:    [[PTRLSB:%.*]] = and i64 [[TMP1]], 3
+; GFX90A-NEXT:    [[TMP2:%.*]] = shl i64 [[PTRLSB]], 3
+; GFX90A-NEXT:    [[SHIFTAMT:%.*]] = trunc i64 [[TMP2]] to i32
+; GFX90A-NEXT:    [[MASK:%.*]] = shl i32 65535, [[SHIFTAMT]]
+; GFX90A-NEXT:    [[INV_MASK:%.*]] = xor i32 [[MASK]], -1
+; GFX90A-NEXT:    [[TMP3:%.*]] = zext i16 [[VALUE]] to i32
+; GFX90A-NEXT:    [[VALOPERAND_SHIFTED:%.*]] = shl i32 [[TMP3]], [[SHIFTAMT]]
+; GFX90A-NEXT:    [[TMP4:%.*]] = load i32, ptr addrspace(1) [[ALIGNEDADDR]], align 4, !mmra [[META0:![0-9]+]]
+; GFX90A-NEXT:    br label [[ATOMICRMW_START:%.*]]
+; GFX90A:       atomicrmw.start:
+; GFX90A-NEXT:    [[LOADED:%.*]] = phi i32 [ [[TMP4]], [[TMP0:%.*]] ], [ [[NEWLOADED:%.*]], [[ATOMICRMW_START]] ]
+; GFX90A-NEXT:    [[TMP5:%.*]] = and i32 [[LOADED]], [[INV_MASK]]
+; GFX90A-NEXT:    [[TMP6:%.*]] = or i32 [[TMP5]], [[VALOPERAND_SHIFTED]]
+; GFX90A-NEXT:    [[TMP7:%.*]] = cmpxchg ptr addrspace(1) [[ALIGNEDADDR]], i32 [[LOADED]], i32 [[TMP6]] syncscope("agent") seq_cst seq_cst, align 4, !mmra [[META0]]
+; GFX90A-NEXT:    [[SUCCESS:%.*]] = extractvalue { i32, i1 } [[TMP7]], 1
+; GFX90A-NEXT:    [[NEWLOADED]] = extractvalue { i32, i1 } [[TMP7]], 0
+; GFX90A-NEXT:    br i1 [[SUCCESS]], label [[ATOMICRMW_END:%.*]], label [[ATOMICRMW_START]]
+; GFX90A:       atomicrmw.end:
+; GFX90A-NEXT:    [[SHIFTED:%.*]] = lshr i32 [[NEWLOADED]], [[SHIFTAMT]]
+; GFX90A-NEXT:    [[EXTRACTED:%.*]] = trunc i32 [[SHIFTED]] to i16
+; GFX90A-NEXT:    ret i16 [[EXTRACTED]]
+;
+; GFX1100-LABEL: define i16 @test_atomicrmw_xchg_i16_global_agent(
+; GFX1100-SAME: ptr addrspace(1) [[PTR:%.*]], i16 [[VALUE:%.*]]) #[[ATTR0:[0-9]+]] {
+; GFX1100-NEXT:    [[ALIGNEDADDR:%.*]] = call ptr addrspace(1) @llvm.ptrmask.p1.i64(ptr addrspace(1) [[PTR]], i64 -4)
+; GFX1100-NEXT:    [[TMP1:%.*]] = ptrtoint ptr addrspace(1) [[PTR]] to i64
+; GFX1100-NEXT:    [[PTRLSB:%.*]] = and i64 [[TMP1]], 3
+; GFX1100-NEXT:    [[TMP2:%.*]] = shl i64 [[PTRLSB]], 3
+; GFX1100-NEXT:    [[SHIFTAMT:%.*]] = trunc i64 [[TMP2]] to i32
+; GFX1100-NEXT:    [[MASK:%.*]] = shl i32 65535, [[SHIFTAMT]]
+; GFX1100-NEXT:    [[INV_MASK:%.*]] = xor i32 [[MASK]], -1
+; GFX1100-NEXT:    [[TMP3:%.*]] = zext i16 [[VALUE]] to i32
+; GFX1100-NEXT:    [[VALOPERAND_SHIFTED:%.*]] = shl i32 [[TMP3]], [[SHIFTAMT]]
+; GFX1100-NEXT:    [[TMP4:%.*]] = load i32, ptr addrspace(1) [[ALIGNEDADDR]], align 4, !mmra [[META0:![0-9]+]]
+; GFX1100-NEXT:    br label [[ATOMICRMW_START:%.*]]
+; GFX1100:       atomicrmw.start:
+; GFX1100-NEXT:    [[LOADED:%.*]] = phi i32 [ [[TMP4]], [[TMP0:%.*]] ], [ [[NEWLOADED:%.*]], [[ATOMICRMW_START]] ]
+; GFX1100-NEXT:    [[TMP5:%.*]] = and i32 [[LOADED]], [[INV_MASK]]
+; GFX1100-NEXT:    [[TMP6:%.*]] = or i32 [[TMP5]], [[VALOPERAND_SHIFTED]]
+; GFX1100-NEXT:    [[TMP7:%.*]] = cmpxchg ptr addrspace(1) [[ALIGNEDADDR]], i32 [[LOADED]], i32 [[TMP6]] syncscope("agent") seq_cst seq_cst, align 4, !mmra [[META0]]
+; GFX1100-NEXT:    [[SUCCESS:%.*]] = extractvalue { i32, i1 } [[TMP7]], 1
+; GFX1100-NEXT:    [[NEWLOADED]] = extractvalue { i32, i1 } [[TMP7]], 0
+; GFX1100-NEXT:    br i1 [[SUCCESS]], label [[ATOMICRMW_END:%.*]], label [[ATOMICRMW_START]]
+; GFX1100:       atomicrmw.end:
+; GFX1100-NEXT:    [[SHIFTED:%.*]] = lshr i32 [[NEWLOADED]], [[SHIFTAMT]]
+; GFX1100-NEXT:    [[EXTRACTED:%.*]] = trunc i32 [[SHIFTED]] to i16
+; GFX1100-NEXT:    ret i16 [[EXTRACTED]]
+;
+  %res = atomicrmw xchg ptr addrspace(1) %ptr, i16 %value syncscope("agent") seq_cst, !mmra !2
+  ret i16 %res
+}
+
+define i16 @test_cmpxchg_i16_global_agent_align4(ptr addrspace(1) %out, i16 %in, i16 %old) {
+; GFX90A-LABEL: define i16 @test_cmpxchg_i16_global_agent_align4(
+; GFX90A-SAME: ptr addrspace(1) [[OUT:%.*]], i16 [[IN:%.*]], i16 [[OLD:%.*]]) #[[ATTR0]] {
+; GFX90A-NEXT:    [[GEP:%.*]] = getelementptr i16, ptr addrspace(1) [[OUT]], i64 4
+; GFX90A-NEXT:    [[TMP1:%.*]] = zext i16 [[IN]] to i32
+; GFX90A-NEXT:    [[TMP2:%.*]] = zext i16 [[OLD]] to i32
+; GFX90A-NEXT:    [[TMP3:%.*]] = load i32, ptr addrspace(1) [[GEP]], align 4, !mmra [[META0]]
+; GFX90A-NEXT:    [[TMP4:%.*]] = and i32 [[TMP3]], -65536
+; GFX90A-NEXT:    br label [[PARTWORD_CMPXCHG_LOOP:%.*]]
+; GFX90A:       partword.cmpxchg.loop:
+; GFX90A-NEXT:    [[TMP5:%.*]] = phi i32 [ [[TMP4]], [[TMP0:%.*]] ], [ [[TMP11:%.*]], [[PARTWORD_CMPXCHG_FAILURE:%.*]] ]
+; GFX90A-NEXT:    [[TMP6:%.*]] = or i32 [[TMP5]], [[TMP1]]
+; GFX90A-NEXT:    [[TMP7:%.*]] = or i32 [[TMP5]], [[TMP2]]
+; GFX90A-NEXT:    [[TMP8:%.*]] = cmpxchg ptr addrspace(1) [[GEP]], i32 [[TMP7]], i32 [[TMP6]] seq_cst seq_cst, align 4, !mmra [[META0]]
+; GFX90A-NEXT:    [[TMP9:%.*]] = extractvalue { i32, i1 } [[TMP8]], 0
+; GFX90A-NEXT:    [[TMP10:%.*]] = extractvalue { i32, i1 } [[TMP8]], 1
+; GFX90A-NEXT:    br i1 [[TMP10]], label [[PARTWORD_CMPXCHG_END:%.*]], label [[PARTWORD_CMPXCHG_FAILURE]]
+; GFX90A:       partword.cmpxchg.failure:
+; GFX90A-NEXT:    [[TMP11]] = and i32 [[TMP9]], -65536
+; GFX90A-NEXT:    [[TMP12:%.*]] = icmp ne i32 [[TMP5]], [[TMP11]]
+; GFX90A-NEXT:    br i1 [[TMP12]], label [[PARTWORD_CMPXCHG_LOOP]], label [[PARTWORD_CMPXCHG_END]]
+; GFX90A:       partword.cmpxchg.end:
+; GFX90A-NEXT:    [[EXTRACTED:%.*]] = trunc i32 [[TMP9]] to i16
+; GFX90A-NEXT:    [[TMP13:%.*]] = insertvalue { i16, i1 } poison, i16 [[EXTRACTED]], 0
+; GFX90A-NEXT:    [[TMP14:%.*]] = insertvalue { i16, i1 } [[TMP13]], i1 [[TMP10]], 1
+; GFX90A-NEXT:    [[EXTRACT:%.*]] = extractvalue { i16, i1 } [[TMP14]], 0
+; GFX90A-NEXT:    ret i16 [[EXTRACT]]
+;
+; GFX1100-LABEL: define i16 @test_cmpxchg_i16_global_agent_align4(
+; GFX1100-SAME: ptr addrspace(1) [[OUT:%.*]], i16 [[IN:%.*]], i16 [[OLD:%.*]]) #[[ATTR0]] {
+; GFX1100-NEXT:    [[GEP:%.*]] = getelementptr i16, ptr addrspace(1) [[OUT]], i64 4
+; GFX1100-NEXT:    [[TMP1:%.*]] = zext i16 [[IN]] to i32
+; GFX1100-NEXT:    [[TMP2:%.*]] = zext i16 [[OLD]] to i32
+; GFX1100-NEXT:    [[TMP3:%.*]] = load i32, ptr addrspace(1) [[GEP]], align 4, !mmra [[META0]]
+; GFX1100-NEXT:    [[TMP4:%.*]] = and i32 [[TMP3]], -65536
+; GFX1100-NEXT:    br label [[PARTWORD_CMPXCHG_LOOP:%.*]]
+; GFX1100:       partword.cmpxchg.loop:
+; GFX1100-NEXT:    [[TMP5:%.*]] = phi i32 [ [[TMP4]], [[TMP0:%.*]] ], [ [[TMP11:%.*]], [[PARTWORD_CMPXCHG_FAILURE:%.*]] ]
+; GFX1100-NEXT:    [[TMP6:%.*]] = or i32 [[TMP5]], [[TMP1]]
+; GFX1100-NEXT:    [[TMP7:%.*]] = or i32 [[TMP5]], [[TMP2]]
+; GFX1100-NEXT:    [[TMP8:%.*]] = cmpxchg ptr addrspace(1) [[GEP]], i32 [[TMP7]], i32 [[TMP6]] seq_cst seq_cst, align 4, !mmra [[META0]]
+; GFX1100-NEXT:    [[TMP9:%.*]] = extractvalue { i32, i1 } [[TMP8]], 0
+; GFX1100-NEXT:    [[TMP10:%.*]] = extractvalue { i32, i1 } [[TMP8]], 1
+; GFX1100-NEXT:    br i1 [[TMP10]], label [[PARTWORD_CMPXCHG_END:%.*]], label [[PARTWORD_CMPXCHG_FAILURE]]
+; GFX1100:       partword.cmpxchg.failure:
+; GFX1100-NEXT:    [[TMP11]] = and i32 [[TMP9]], -65536
+; GFX1100-NEXT:    [[TMP12:%.*]] = icmp ne i32 [[TMP5]], [[TMP11]]
+; GFX1100-NEXT:    br i1 [[TMP12]], label [[PARTWORD_CMPXCHG_LOOP]], label [[PARTWORD_CMPXCHG_END]]
+; GFX1100:       partword.cmpxchg.end:
+; GFX1100-NEXT:    [[EXTRACTED:%.*]] = trunc i32 [[TMP9]] to i16
+; GFX1100-NEXT:    [[TMP13:%.*]] = insertvalue { i16, i1 } poison, i16 [[EXTRACTED]], 0
+; GFX1100-NEXT:    [[TMP14:%.*]] = insertvalue { i16, i1 } [[TMP13]], i1 [[TMP10]], 1
+; GFX1100-NEXT:    [[EXTRACT:%.*]] = extractvalue { i16, i1 } [[TMP14]], 0
+; GFX1100-NEXT:    ret i16 [[EXTRACT]]
+;
+  %gep = getelementptr i16, ptr addrspace(1) %out, i64 4
+  %res = cmpxchg ptr addrspace(1) %gep, i16 %old, i16 %in seq_cst seq_cst, align 4, !mmra !2
+  %extract = extractvalue {i16, i1} %res, 0
+  ret i16 %extract
+}
+
+define void @syncscope_workgroup_nortn(ptr %addr, float %val) #0 {
+; GFX90A-LABEL: define void @syncscope_workgroup_nortn(
+; GFX90A-SAME: ptr [[ADDR:%.*]], float [[VAL:%.*]]) #[[ATTR1:[0-9]+]] {
+; GFX90A-NEXT:    br label [[ATOMICRMW_CHECK_SHARED:%.*]]
+; GFX90A:       atomicrmw.check.shared:
+; GFX90A-NEXT:    [[IS_SHARED:%.*]] = call i1 @llvm.amdgcn.is.shared(ptr [[ADDR]])
+; GFX90A-NEXT:    br i1 [[IS_SHARED]], label [[ATOMICRMW_SHARED:%.*]], label [[ATOMICRMW_CHECK_PRIVATE:%.*]]
+; GFX90A:       atomicrmw.shared:
+; GFX90A-NEXT:    [[TMP1:%.*]] = addrspacecast ptr [[ADDR]] to ptr addrspace(3)
+; GFX90A-NEXT:    [[TMP2:%.*]] = atomicrmw fadd ptr addrspace(3) [[TMP1]], float [[VAL]] syncscope("workgroup") seq_cst, align 4, !mmra [[META0]]
+; GFX90A-NEXT:    br label [[ATOMICRMW_PHI:%.*]]
+; GFX90A:       atomicrmw.check.private:
+; GFX90A-NEXT:    [[IS_PRIVATE:%.*]] = call i1 @llvm.amdgcn.is.private(ptr [[ADDR]])
+; GFX90A-NEXT:    br i1 [[IS_PRIVATE]], label [[ATOMICRMW_PRIVATE:%.*]], label [[ATOMICRMW_GLOBAL:%.*]]
+; GFX90A:       atomicrmw.private:
+; GFX90A-NEXT:    [[TMP3:%.*]] = addrspacecast ptr [[ADDR]] to ptr addrspace(5)
+; GFX90A-NEXT:    [[LOADED_PRIVATE:%.*]] = load float, ptr addrspace(5) [[TMP3]], align 4
+; GFX90A-NEXT:    [[VAL_NEW:%.*]] = fadd float [[LOADED_PRIVATE]], [[VAL]]
+; GFX90A-NEXT:    store float [[VAL_NEW]], ptr addrspace(5) [[TMP3]], align 4
+; GFX90A-NEXT:    br label [[ATOMICRMW_PHI]]
+; GFX90A:       atomicrmw.global:
+; GFX90A-NEXT:    [[TMP4:%.*]] = addrspacecast ptr [[ADDR]] to ptr addrspace(1)
+; GFX90A-NEXT:    [[TMP5:%.*]] = atomicrmw fadd ptr addrspace(1) [[TMP4]], float [[VAL]] syncscope("workgroup") seq_cst, align 4, !mmra [[META0]]
+; GFX90A-NEXT:    br label [[ATOMICRMW_PHI]]
+; GFX90A:       atomicrmw.phi:
+; GFX90A-NEXT:    [[LOADED_PHI:%.*]] = phi float [ [[TMP2]], [[ATOMICRMW_SHARED]] ], [ [[LOADED_PRIVATE]], [[ATOMICRMW_PRIVATE]] ], [ [[TMP5]], [[ATOMICRMW_GLOBAL]] ]
+; GFX90A-NEXT:    br label [[ATOMICRMW_END:%.*]]
+; GFX90A:       atomicrmw.end:
+; GFX90A-NEXT:    ret void
+;
+; GFX1100-LABEL: define void @syncscope_workgroup_nortn(
+; GFX1100-SAME: ptr [[ADDR:%.*]], float [[VAL:%.*]]) #[[ATTR1:[0-9]+]] {
+; GFX1100-NEXT:    [[RES:%.*]] = atomicrmw fadd ptr [[ADDR]], float [[VAL]] syncscope("workgroup") seq_cst, align 4, !mmra [[META0]]
+; GFX1100-NEXT:    ret void
+;
+  %res = atomicrmw fadd ptr %addr, float %val syncscope("workgroup") seq_cst, !mmra !2
+  ret void
+}
+
+define i32 @atomic_load_global_align1(ptr addrspace(1) %ptr) {
+; GFX90A-LABEL: define i32 @atomic_load_global_align1(
+; GFX90A-SAME: ptr addrspace(1) [[PTR:%.*]]) #[[ATTR0]] {
+; GFX90A-NEXT:    [[TMP1:%.*]] = addrspacecast ptr addrspace(1) [[PTR]] to ptr
+; GFX90A-NEXT:    [[TMP2:%.*]] = alloca i32, align 4, addrspace(5)
+; GFX90A-NEXT:    call void @llvm.lifetime.start.p5(i64 4, ptr addrspace(5) [[TMP2]])
+; GFX90A-NEXT:    call void @__atomic_load(i64 4, ptr [[TMP1]], ptr addrspace(5) [[TMP2]], i32 5)
+; GFX90A-NEXT:    [[TMP3:%.*]] = load i32, ptr addrspace(5) [[TMP2]], align 4
+; GFX90A-NEXT:    call void @llvm.lifetime.end.p5(i64 4, ptr addrspace(5) [[TMP2]])
+; GFX90A-NEXT:    ret i32 [[TMP3]]
+;
+; GFX1100-LABEL: define i32 @atomic_load_global_align1(
+; GFX1100-SAME: ptr addrspace(1) [[PTR:%.*]]) #[[ATTR0]] {
+; GFX1100-NEXT:    [[TMP1:%.*]] = addrspacecast ptr addrspace(1) [[PTR]] to ptr
+; GFX1100-NEXT:    [[TMP2:%.*]] = alloca i32, align 4, addrspace(5)
+; GFX1100-NEXT:    call void @llvm.lifetime.start.p5(i64 4, ptr addrspace(5) [[TMP2]])
+; GFX1100-NEXT:    call void @__atomic_load(i64 4, ptr [[TMP1]], ptr addrspace(5) [[TMP2]], i32 5)
+; GFX1100-NEXT:    [[TMP3:%.*]] = load i32, ptr addrspace(5) [[TMP2]], align 4
+; GFX1100-NEXT:    call void @llvm.lifetime.end.p5(i64 4, ptr addrspace(5) [[TMP2]])
+; GFX1100-NEXT:    ret i32 [[TMP3]]
+;
+  %val = load atomic i32, ptr addrspace(1) %ptr  seq_cst, align 1, !mmra !2
+  ret i32 %val
+}
+
+attributes #0 = { "amdgpu-unsafe-fp-atomics"="true" }
+
+!0 = !{!"foo", !"bar"}
+!1 = !{!"bux", !"baz"}
+!2 = !{!0, !1}
+;.
+; GFX90A: [[META0]] = !{[[META1:![0-9]+]], [[META2:![0-9]+]]}
+; GFX90A: [[META1]] = !{!"foo", !"bar"}
+; GFX90A: [[META2]] = !{!"bux", !"baz"}
+;.
+; GFX1100: [[META0]] = !{[[META1:![0-9]+]], [[META2:![0-9]+]]}
+; GFX1100: [[META1]] = !{!"foo", !"bar"}
+; GFX1100: [[META2]] = !{!"bux", !"baz"}
+;.
diff --git a/llvm/test/Transforms/AtomicExpand/AMDGPU/no-expand-atomic-load.ll b/llvm/test/Transforms/AtomicExpand/AMDGPU/no-expand-atomic-load.ll
new file mode 100644
index 000000000000..b1497aefe9b9
--- /dev/null
+++ b/llvm/test/Transforms/AtomicExpand/AMDGPU/no-expand-atomic-load.ll
@@ -0,0 +1,181 @@
+; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --version 4
+; RUN: opt -S -mtriple=amdgcn-amd-amdhsa -passes=atomic-expand %s | FileCheck %s
+
+; Make sure atomic loads are not bitcasted and lose metadata
+
+define float @load_atomic_f32_global_system(ptr addrspace(1) %ptr) {
+; CHECK-LABEL: define float @load_atomic_f32_global_system(
+; CHECK-SAME: ptr addrspace(1) [[PTR:%.*]]) {
+; CHECK-NEXT:    [[TMP2:%.*]] = load atomic float, ptr addrspace(1) [[PTR]] seq_cst, align 4, !some.unknown.md [[META0:![0-9]+]]
+; CHECK-NEXT:    ret float [[TMP2]]
+;
+  %ld = load atomic float, ptr addrspace(1) %ptr seq_cst, align 4, !some.unknown.md !0
+  ret float %ld
+}
+
+define float @load_atomic_f32_global_agent(ptr addrspace(1) %ptr) {
+; CHECK-LABEL: define float @load_atomic_f32_global_agent(
+; CHECK-SAME: ptr addrspace(1) [[PTR:%.*]]) {
+; CHECK-NEXT:    [[TMP2:%.*]] = load atomic float, ptr addrspace(1) [[PTR]] syncscope("agent") seq_cst, align 4, !some.unknown.md [[META0]]
+; CHECK-NEXT:    ret float [[TMP2]]
+;
+  %ld = load atomic float, ptr addrspace(1) %ptr syncscope("agent") seq_cst, align 4, !some.unknown.md !0
+  ret float %ld
+}
+
+define float @load_atomic_f32_local(ptr addrspace(3) %ptr) {
+; CHECK-LABEL: define float @load_atomic_f32_local(
+; CHECK-SAME: ptr addrspace(3) [[PTR:%.*]]) {
+; CHECK-NEXT:    [[TMP2:%.*]] = load atomic float, ptr addrspace(3) [[PTR]] seq_cst, align 4, !some.unknown.md [[META0]]
+; CHECK-NEXT:    ret float [[TMP2]]
+;
+  %ld = load atomic float, ptr addrspace(3) %ptr seq_cst, align 4, !some.unknown.md !0
+  ret float %ld
+}
+
+define float @load_atomic_f32_flat_system(ptr %ptr) {
+; CHECK-LABEL: define float @load_atomic_f32_flat_system(
+; CHECK-SAME: ptr [[PTR:%.*]]) {
+; CHECK-NEXT:    [[TMP2:%.*]] = load atomic float, ptr [[PTR]] seq_cst, align 4, !some.unknown.md [[META0]]
+; CHECK-NEXT:    ret float [[TMP2]]
+;
+  %ld = load atomic float, ptr %ptr seq_cst, align 4, !some.unknown.md !0
+  ret float %ld
+}
+
+define float @load_atomic_f32_flat_agent(ptr %ptr) {
+; CHECK-LABEL: define float @load_atomic_f32_flat_agent(
+; CHECK-SAME: ptr [[PTR:%.*]]) {
+; CHECK-NEXT:    [[TMP2:%.*]] = load atomic float, ptr [[PTR]] syncscope("agent") seq_cst, align 4, !some.unknown.md [[META0]]
+; CHECK-NEXT:    ret float [[TMP2]]
+;
+  %ld = load atomic float, ptr %ptr syncscope("agent") seq_cst, align 4, !some.unknown.md !0
+  ret float %ld
+}
+
+define half @load_atomic_f16_global_system(ptr addrspace(1) %ptr) {
+; CHECK-LABEL: define half @load_atomic_f16_global_system(
+; CHECK-SAME: ptr addrspace(1) [[PTR:%.*]]) {
+; CHECK-NEXT:    [[TMP2:%.*]] = load atomic half, ptr addrspace(1) [[PTR]] seq_cst, align 4, !some.unknown.md [[META0]]
+; CHECK-NEXT:    ret half [[TMP2]]
+;
+  %ld = load atomic half, ptr addrspace(1) %ptr seq_cst, align 4, !some.unknown.md !0
+  ret half %ld
+}
+
+define half @load_atomic_f16_global_agent(ptr addrspace(1) %ptr) {
+; CHECK-LABEL: define half @load_atomic_f16_global_agent(
+; CHECK-SAME: ptr addrspace(1) [[PTR:%.*]]) {
+; CHECK-NEXT:    [[TMP2:%.*]] = load atomic half, ptr addrspace(1) [[PTR]] syncscope("agent") seq_cst, align 4, !some.unknown.md [[META0]]
+; CHECK-NEXT:    ret half [[TMP2]]
+;
+  %ld = load atomic half, ptr addrspace(1) %ptr syncscope("agent") seq_cst, align 4, !some.unknown.md !0
+  ret half %ld
+}
+
+define half @load_atomic_f16_local(ptr addrspace(3) %ptr) {
+; CHECK-LABEL: define half @load_atomic_f16_local(
+; CHECK-SAME: ptr addrspace(3) [[PTR:%.*]]) {
+; CHECK-NEXT:    [[TMP2:%.*]] = load atomic half, ptr addrspace(3) [[PTR]] seq_cst, align 2, !some.unknown.md [[META0]]
+; CHECK-NEXT:    ret half [[TMP2]]
+;
+  %ld = load atomic half, ptr addrspace(3) %ptr seq_cst, align 2, !some.unknown.md !0
+  ret half %ld
+}
+
+define bfloat @load_atomic_bf16_global_system(ptr addrspace(1) %ptr) {
+; CHECK-LABEL: define bfloat @load_atomic_bf16_global_system(
+; CHECK-SAME: ptr addrspace(1) [[PTR:%.*]]) {
+; CHECK-NEXT:    [[TMP2:%.*]] = load atomic bfloat, ptr addrspace(1) [[PTR]] seq_cst, align 2, !some.unknown.md [[META0]]
+; CHECK-NEXT:    ret bfloat [[TMP2]]
+;
+  %ld = load atomic bfloat, ptr addrspace(1) %ptr seq_cst, align 2, !some.unknown.md !0
+  ret bfloat %ld
+}
+
+define bfloat @load_atomic_bf16_global_agent(ptr addrspace(1) %ptr) {
+; CHECK-LABEL: define bfloat @load_atomic_bf16_global_agent(
+; CHECK-SAME: ptr addrspace(1) [[PTR:%.*]]) {
+; CHECK-NEXT:    [[TMP2:%.*]] = load atomic bfloat, ptr addrspace(1) [[PTR]] syncscope("agent") seq_cst, align 2, !some.unknown.md [[META0]]
+; CHECK-NEXT:    ret bfloat [[TMP2]]
+;
+  %ld = load atomic bfloat, ptr addrspace(1) %ptr syncscope("agent") seq_cst, align 2, !some.unknown.md !0
+  ret bfloat %ld
+}
+
+define bfloat @load_atomic_bf16_local(ptr addrspace(3) %ptr) {
+; CHECK-LABEL: define bfloat @load_atomic_bf16_local(
+; CHECK-SAME: ptr addrspace(3) [[PTR:%.*]]) {
+; CHECK-NEXT:    [[TMP2:%.*]] = load atomic bfloat, ptr addrspace(3) [[PTR]] seq_cst, align 2, !some.unknown.md [[META0]]
+; CHECK-NEXT:    ret bfloat [[TMP2]]
+;
+  %ld = load atomic bfloat, ptr addrspace(3) %ptr seq_cst, align 2, !some.unknown.md !0
+  ret bfloat %ld
+}
+
+define bfloat @load_atomic_bf16_flat(ptr %ptr) {
+; CHECK-LABEL: define bfloat @load_atomic_bf16_flat(
+; CHECK-SAME: ptr [[PTR:%.*]]) {
+; CHECK-NEXT:    [[TMP2:%.*]] = load atomic bfloat, ptr [[PTR]] seq_cst, align 2, !some.unknown.md [[META0]]
+; CHECK-NEXT:    ret bfloat [[TMP2]]
+;
+  %ld = load atomic bfloat, ptr %ptr seq_cst, align 2, !some.unknown.md !0
+  ret bfloat %ld
+}
+
+define double @load_atomic_f64_global_system(ptr addrspace(1) %ptr) {
+; CHECK-LABEL: define double @load_atomic_f64_global_system(
+; CHECK-SAME: ptr addrspace(1) [[PTR:%.*]]) {
+; CHECK-NEXT:    [[TMP2:%.*]] = load atomic double, ptr addrspace(1) [[PTR]] seq_cst, align 8, !some.unknown.md [[META0]]
+; CHECK-NEXT:    ret double [[TMP2]]
+;
+  %ld = load atomic double, ptr addrspace(1) %ptr seq_cst, align 8, !some.unknown.md !0
+  ret double %ld
+}
+
+define double @load_atomic_f64_global_agent(ptr addrspace(1) %ptr) {
+; CHECK-LABEL: define double @load_atomic_f64_global_agent(
+; CHECK-SAME: ptr addrspace(1) [[PTR:%.*]]) {
+; CHECK-NEXT:    [[TMP2:%.*]] = load atomic double, ptr addrspace(1) [[PTR]] syncscope("agent") seq_cst, align 8, !some.unknown.md [[META0]]
+; CHECK-NEXT:    ret double [[TMP2]]
+;
+  %ld = load atomic double, ptr addrspace(1) %ptr syncscope("agent") seq_cst, align 8, !some.unknown.md !0
+  ret double %ld
+}
+
+define double @load_atomic_f64_local(ptr addrspace(3) %ptr) {
+; CHECK-LABEL: define double @load_atomic_f64_local(
+; CHECK-SAME: ptr addrspace(3) [[PTR:%.*]]) {
+; CHECK-NEXT:    [[TMP2:%.*]] = load atomic double, ptr addrspace(3) [[PTR]] seq_cst, align 8, !some.unknown.md [[META0]]
+; CHECK-NEXT:    ret double [[TMP2]]
+;
+  %ld = load atomic double, ptr addrspace(3) %ptr seq_cst, align 8, !some.unknown.md !0
+  ret double %ld
+}
+
+define double @load_atomic_f64_flat_system(ptr %ptr) {
+; CHECK-LABEL: define double @load_atomic_f64_flat_system(
+; CHECK-SAME: ptr [[PTR:%.*]]) {
+; CHECK-NEXT:    [[TMP2:%.*]] = load atomic double, ptr [[PTR]] seq_cst, align 8, !some.unknown.md [[META0]]
+; CHECK-NEXT:    ret double [[TMP2]]
+;
+  %ld = load atomic double, ptr %ptr seq_cst, align 8, !some.unknown.md !0
+  ret double %ld
+}
+
+define double @load_atomic_f64_flat_agent(ptr %ptr) {
+; CHECK-LABEL: define double @load_atomic_f64_flat_agent(
+; CHECK-SAME: ptr [[PTR:%.*]]) {
+; CHECK-NEXT:    [[TMP2:%.*]] = load atomic double, ptr [[PTR]] syncscope("agent") seq_cst, align 8, !some.unknown.md [[META0]]
+; CHECK-NEXT:    ret double [[TMP2]]
+;
+  %ld = load atomic double, ptr %ptr syncscope("agent") seq_cst, align 8, !some.unknown.md !0
+  ret double %ld
+}
+
+!0 = !{}
+
+
+;.
+; CHECK: [[META0]] = !{}
+;.
diff --git a/llvm/test/Transforms/AtomicExpand/AMDGPU/no-expand-atomic-store.ll b/llvm/test/Transforms/AtomicExpand/AMDGPU/no-expand-atomic-store.ll
new file mode 100644
index 000000000000..db0c3a20e62f
--- /dev/null
+++ b/llvm/test/Transforms/AtomicExpand/AMDGPU/no-expand-atomic-store.ll
@@ -0,0 +1,179 @@
+; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --version 4
+; RUN: opt -S -mtriple=amdgcn-amd-amdhsa -passes=atomic-expand %s | FileCheck %s
+
+define void @store_atomic_f32_global_system(float %val, ptr addrspace(1) %ptr) {
+; CHECK-LABEL: define void @store_atomic_f32_global_system(
+; CHECK-SAME: float [[VAL:%.*]], ptr addrspace(1) [[PTR:%.*]]) {
+; CHECK-NEXT:    [[TMP1:%.*]] = bitcast float [[VAL]] to i32
+; CHECK-NEXT:    store atomic i32 [[TMP1]], ptr addrspace(1) [[PTR]] seq_cst, align 4
+; CHECK-NEXT:    ret void
+;
+  store atomic float %val, ptr addrspace(1) %ptr seq_cst, align 4, !some.unknown.md !0
+  ret void
+}
+
+define void @store_atomic_f32_global_agent(float %val, ptr addrspace(1) %ptr) {
+; CHECK-LABEL: define void @store_atomic_f32_global_agent(
+; CHECK-SAME: float [[VAL:%.*]], ptr addrspace(1) [[PTR:%.*]]) {
+; CHECK-NEXT:    [[TMP1:%.*]] = bitcast float [[VAL]] to i32
+; CHECK-NEXT:    store atomic i32 [[TMP1]], ptr addrspace(1) [[PTR]] syncscope("agent") seq_cst, align 4
+; CHECK-NEXT:    ret void
+;
+  store atomic float %val, ptr addrspace(1) %ptr syncscope("agent") seq_cst, align 4, !some.unknown.md !0
+  ret void
+}
+
+define void @store_atomic_f32_local(float %val, ptr addrspace(3) %ptr) {
+; CHECK-LABEL: define void @store_atomic_f32_local(
+; CHECK-SAME: float [[VAL:%.*]], ptr addrspace(3) [[PTR:%.*]]) {
+; CHECK-NEXT:    [[TMP1:%.*]] = bitcast float [[VAL]] to i32
+; CHECK-NEXT:    store atomic i32 [[TMP1]], ptr addrspace(3) [[PTR]] seq_cst, align 4
+; CHECK-NEXT:    ret void
+;
+  store atomic float %val, ptr addrspace(3) %ptr seq_cst, align 4, !some.unknown.md !0
+  ret void
+}
+
+define void @store_atomic_f32_flat(float %val, ptr %ptr) {
+; CHECK-LABEL: define void @store_atomic_f32_flat(
+; CHECK-SAME: float [[VAL:%.*]], ptr [[PTR:%.*]]) {
+; CHECK-NEXT:    [[TMP1:%.*]] = bitcast float [[VAL]] to i32
+; CHECK-NEXT:    store atomic i32 [[TMP1]], ptr [[PTR]] seq_cst, align 4
+; CHECK-NEXT:    ret void
+;
+  store atomic float %val, ptr %ptr seq_cst, align 4, !some.unknown.md !0
+  ret void
+}
+
+define void @store_atomic_f16_global_system(half %val, ptr addrspace(1) %ptr) {
+; CHECK-LABEL: define void @store_atomic_f16_global_system(
+; CHECK-SAME: half [[VAL:%.*]], ptr addrspace(1) [[PTR:%.*]]) {
+; CHECK-NEXT:    [[TMP1:%.*]] = bitcast half [[VAL]] to i16
+; CHECK-NEXT:    store atomic i16 [[TMP1]], ptr addrspace(1) [[PTR]] seq_cst, align 4
+; CHECK-NEXT:    ret void
+;
+  store atomic half %val, ptr addrspace(1) %ptr seq_cst, align 4, !some.unknown.md !0
+  ret void
+}
+
+define void @store_atomic_f16_global_agent(half %val, ptr addrspace(1) %ptr) {
+; CHECK-LABEL: define void @store_atomic_f16_global_agent(
+; CHECK-SAME: half [[VAL:%.*]], ptr addrspace(1) [[PTR:%.*]]) {
+; CHECK-NEXT:    [[TMP1:%.*]] = bitcast half [[VAL]] to i16
+; CHECK-NEXT:    store atomic i16 [[TMP1]], ptr addrspace(1) [[PTR]] syncscope("agent") seq_cst, align 4
+; CHECK-NEXT:    ret void
+;
+  store atomic half %val, ptr addrspace(1) %ptr syncscope("agent") seq_cst, align 4, !some.unknown.md !0
+  ret void
+}
+
+define void @store_atomic_f16_local(half %val, ptr addrspace(3) %ptr) {
+; CHECK-LABEL: define void @store_atomic_f16_local(
+; CHECK-SAME: half [[VAL:%.*]], ptr addrspace(3) [[PTR:%.*]]) {
+; CHECK-NEXT:    [[TMP1:%.*]] = bitcast half [[VAL]] to i16
+; CHECK-NEXT:    store atomic i16 [[TMP1]], ptr addrspace(3) [[PTR]] seq_cst, align 4
+; CHECK-NEXT:    ret void
+;
+  store atomic half %val, ptr addrspace(3) %ptr seq_cst, align 4, !some.unknown.md !0
+  ret void
+}
+
+define void @store_atomic_f16_flat(half %val, ptr %ptr) {
+; CHECK-LABEL: define void @store_atomic_f16_flat(
+; CHECK-SAME: half [[VAL:%.*]], ptr [[PTR:%.*]]) {
+; CHECK-NEXT:    [[TMP1:%.*]] = bitcast half [[VAL]] to i16
+; CHECK-NEXT:    store atomic i16 [[TMP1]], ptr [[PTR]] seq_cst, align 4
+; CHECK-NEXT:    ret void
+;
+  store atomic half %val, ptr %ptr seq_cst, align 4, !some.unknown.md !0
+  ret void
+}
+
+define void @store_atomic_bf16_global_system(bfloat %val, ptr addrspace(1) %ptr) {
+; CHECK-LABEL: define void @store_atomic_bf16_global_system(
+; CHECK-SAME: bfloat [[VAL:%.*]], ptr addrspace(1) [[PTR:%.*]]) {
+; CHECK-NEXT:    [[TMP1:%.*]] = bitcast bfloat [[VAL]] to i16
+; CHECK-NEXT:    store atomic i16 [[TMP1]], ptr addrspace(1) [[PTR]] seq_cst, align 4
+; CHECK-NEXT:    ret void
+;
+  store atomic bfloat %val, ptr addrspace(1) %ptr seq_cst, align 4, !some.unknown.md !0
+  ret void
+}
+
+define void @store_atomic_bf16_global_agent(bfloat %val, ptr addrspace(1) %ptr) {
+; CHECK-LABEL: define void @store_atomic_bf16_global_agent(
+; CHECK-SAME: bfloat [[VAL:%.*]], ptr addrspace(1) [[PTR:%.*]]) {
+; CHECK-NEXT:    [[TMP1:%.*]] = bitcast bfloat [[VAL]] to i16
+; CHECK-NEXT:    store atomic i16 [[TMP1]], ptr addrspace(1) [[PTR]] syncscope("agent") seq_cst, align 4
+; CHECK-NEXT:    ret void
+;
+  store atomic bfloat %val, ptr addrspace(1) %ptr syncscope("agent") seq_cst, align 4, !some.unknown.md !0
+  ret void
+}
+
+define void @store_atomic_bf16_local(bfloat %val, ptr addrspace(3) %ptr) {
+; CHECK-LABEL: define void @store_atomic_bf16_local(
+; CHECK-SAME: bfloat [[VAL:%.*]], ptr addrspace(3) [[PTR:%.*]]) {
+; CHECK-NEXT:    [[TMP1:%.*]] = bitcast bfloat [[VAL]] to i16
+; CHECK-NEXT:    store atomic i16 [[TMP1]], ptr addrspace(3) [[PTR]] seq_cst, align 4
+; CHECK-NEXT:    ret void
+;
+  store atomic bfloat %val, ptr addrspace(3) %ptr seq_cst, align 4, !some.unknown.md !0
+  ret void
+}
+
+define void @store_atomic_bf16_flat(bfloat %val, ptr %ptr) {
+; CHECK-LABEL: define void @store_atomic_bf16_flat(
+; CHECK-SAME: bfloat [[VAL:%.*]], ptr [[PTR:%.*]]) {
+; CHECK-NEXT:    [[TMP1:%.*]] = bitcast bfloat [[VAL]] to i16
+; CHECK-NEXT:    store atomic i16 [[TMP1]], ptr [[PTR]] seq_cst, align 4
+; CHECK-NEXT:    ret void
+;
+  store atomic bfloat %val, ptr %ptr seq_cst, align 4, !some.unknown.md !0
+  ret void
+}
+define void @store_atomic_f64_global_system(double %val, ptr addrspace(1) %ptr) {
+; CHECK-LABEL: define void @store_atomic_f64_global_system(
+; CHECK-SAME: double [[VAL:%.*]], ptr addrspace(1) [[PTR:%.*]]) {
+; CHECK-NEXT:    [[TMP1:%.*]] = bitcast double [[VAL]] to i64
+; CHECK-NEXT:    store atomic i64 [[TMP1]], ptr addrspace(1) [[PTR]] seq_cst, align 8
+; CHECK-NEXT:    ret void
+;
+  store atomic double %val, ptr addrspace(1) %ptr seq_cst, align 8, !some.unknown.md !0
+  ret void
+}
+
+define void @store_atomic_f64_global_agent(double %val, ptr addrspace(1) %ptr) {
+; CHECK-LABEL: define void @store_atomic_f64_global_agent(
+; CHECK-SAME: double [[VAL:%.*]], ptr addrspace(1) [[PTR:%.*]]) {
+; CHECK-NEXT:    [[TMP1:%.*]] = bitcast double [[VAL]] to i64
+; CHECK-NEXT:    store atomic i64 [[TMP1]], ptr addrspace(1) [[PTR]] syncscope("agent") seq_cst, align 8
+; CHECK-NEXT:    ret void
+;
+  store atomic double %val, ptr addrspace(1) %ptr syncscope("agent") seq_cst, align 8, !some.unknown.md !0
+  ret void
+}
+
+define void @store_atomic_f64_local(double %val, ptr addrspace(3) %ptr) {
+; CHECK-LABEL: define void @store_atomic_f64_local(
+; CHECK-SAME: double [[VAL:%.*]], ptr addrspace(3) [[PTR:%.*]]) {
+; CHECK-NEXT:    [[TMP1:%.*]] = bitcast double [[VAL]] to i64
+; CHECK-NEXT:    store atomic i64 [[TMP1]], ptr addrspace(3) [[PTR]] seq_cst, align 8
+; CHECK-NEXT:    ret void
+;
+  store atomic double %val, ptr addrspace(3) %ptr seq_cst, align 8, !some.unknown.md !0
+  ret void
+}
+
+define void @store_atomic_f64_flat(double %val, ptr %ptr) {
+; CHECK-LABEL: define void @store_atomic_f64_flat(
+; CHECK-SAME: double [[VAL:%.*]], ptr [[PTR:%.*]]) {
+; CHECK-NEXT:    [[TMP1:%.*]] = bitcast double [[VAL]] to i64
+; CHECK-NEXT:    store atomic i64 [[TMP1]], ptr [[PTR]] seq_cst, align 8
+; CHECK-NEXT:    ret void
+;
+  store atomic double %val, ptr %ptr seq_cst, align 8, !some.unknown.md !0
+  ret void
+}
+
+!0 = !{}
diff --git a/llvm/test/Transforms/AtomicExpand/SPARC/partword.ll b/llvm/test/Transforms/AtomicExpand/SPARC/partword.ll
index e8be1bc18748..3a306a4d9861 100644
--- a/llvm/test/Transforms/AtomicExpand/SPARC/partword.ll
+++ b/llvm/test/Transforms/AtomicExpand/SPARC/partword.ll
@@ -220,7 +220,7 @@ define i16 @test_and_i16(ptr %arg, i16 %val) {
 ; CHECK-NEXT:    [[INV_MASK:%.*]] = xor i32 [[MASK]], -1
 ; CHECK-NEXT:    [[TMP3:%.*]] = zext i16 [[VAL:%.*]] to i32
 ; CHECK-NEXT:    [[VALOPERAND_SHIFTED:%.*]] = shl i32 [[TMP3]], [[SHIFTAMT]]
-; CHECK-NEXT:    [[ANDOPERAND:%.*]] = or i32 [[INV_MASK]], [[VALOPERAND_SHIFTED]]
+; CHECK-NEXT:    [[ANDOPERAND:%.*]] = or i32 [[VALOPERAND_SHIFTED]], [[INV_MASK]]
 ; CHECK-NEXT:    [[TMP4:%.*]] = load i32, ptr [[ALIGNEDADDR]], align 4
 ; CHECK-NEXT:    br label [[ATOMICRMW_START:%.*]]
 ; CHECK:       atomicrmw.start:
@@ -262,18 +262,18 @@ define i16 @test_min_i16(ptr %arg, i16 %val) {
 ; CHECK-NEXT:    [[TMP4:%.*]] = icmp sle i16 [[EXTRACTED]], [[VAL:%.*]]
 ; CHECK-NEXT:    [[NEW:%.*]] = select i1 [[TMP4]], i16 [[EXTRACTED]], i16 [[VAL]]
 ; CHECK-NEXT:    [[EXTENDED:%.*]] = zext i16 [[NEW]] to i32
-; CHECK-NEXT:    [[SHIFTED2:%.*]] = shl nuw i32 [[EXTENDED]], [[SHIFTAMT]]
+; CHECK-NEXT:    [[SHIFTED1:%.*]] = shl nuw i32 [[EXTENDED]], [[SHIFTAMT]]
 ; CHECK-NEXT:    [[UNMASKED:%.*]] = and i32 [[LOADED]], [[INV_MASK]]
-; CHECK-NEXT:    [[INSERTED:%.*]] = or i32 [[UNMASKED]], [[SHIFTED2]]
+; CHECK-NEXT:    [[INSERTED:%.*]] = or i32 [[UNMASKED]], [[SHIFTED1]]
 ; CHECK-NEXT:    [[TMP5:%.*]] = cmpxchg ptr [[ALIGNEDADDR]], i32 [[LOADED]], i32 [[INSERTED]] monotonic monotonic, align 4
 ; CHECK-NEXT:    [[SUCCESS:%.*]] = extractvalue { i32, i1 } [[TMP5]], 1
 ; CHECK-NEXT:    [[NEWLOADED]] = extractvalue { i32, i1 } [[TMP5]], 0
 ; CHECK-NEXT:    br i1 [[SUCCESS]], label [[ATOMICRMW_END:%.*]], label [[ATOMICRMW_START]]
 ; CHECK:       atomicrmw.end:
-; CHECK-NEXT:    [[SHIFTED3:%.*]] = lshr i32 [[NEWLOADED]], [[SHIFTAMT]]
-; CHECK-NEXT:    [[EXTRACTED4:%.*]] = trunc i32 [[SHIFTED3]] to i16
+; CHECK-NEXT:    [[SHIFTED2:%.*]] = lshr i32 [[NEWLOADED]], [[SHIFTAMT]]
+; CHECK-NEXT:    [[EXTRACTED3:%.*]] = trunc i32 [[SHIFTED2]] to i16
 ; CHECK-NEXT:    fence seq_cst
-; CHECK-NEXT:    ret i16 [[EXTRACTED4]]
+; CHECK-NEXT:    ret i16 [[EXTRACTED3]]
 ;
 entry:
   %ret = atomicrmw min ptr %arg, i16 %val seq_cst
@@ -301,17 +301,17 @@ define half @test_atomicrmw_fadd_f16(ptr %ptr, half %value) {
 ; CHECK-NEXT:    [[NEW:%.*]] = fadd half [[TMP5]], [[VALUE:%.*]]
 ; CHECK-NEXT:    [[TMP6:%.*]] = bitcast half [[NEW]] to i16
 ; CHECK-NEXT:    [[EXTENDED:%.*]] = zext i16 [[TMP6]] to i32
-; CHECK-NEXT:    [[SHIFTED2:%.*]] = shl nuw i32 [[EXTENDED]], [[SHIFTAMT]]
+; CHECK-NEXT:    [[SHIFTED1:%.*]] = shl nuw i32 [[EXTENDED]], [[SHIFTAMT]]
 ; CHECK-NEXT:    [[UNMASKED:%.*]] = and i32 [[LOADED]], [[INV_MASK]]
-; CHECK-NEXT:    [[INSERTED:%.*]] = or i32 [[UNMASKED]], [[SHIFTED2]]
+; CHECK-NEXT:    [[INSERTED:%.*]] = or i32 [[UNMASKED]], [[SHIFTED1]]
 ; CHECK-NEXT:    [[TMP7:%.*]] = cmpxchg ptr [[ALIGNEDADDR]], i32 [[LOADED]], i32 [[INSERTED]] monotonic monotonic, align 4
 ; CHECK-NEXT:    [[SUCCESS:%.*]] = extractvalue { i32, i1 } [[TMP7]], 1
 ; CHECK-NEXT:    [[NEWLOADED]] = extractvalue { i32, i1 } [[TMP7]], 0
 ; CHECK-NEXT:    br i1 [[SUCCESS]], label [[ATOMICRMW_END:%.*]], label [[ATOMICRMW_START]]
 ; CHECK:       atomicrmw.end:
-; CHECK-NEXT:    [[SHIFTED3:%.*]] = lshr i32 [[NEWLOADED]], [[SHIFTAMT]]
-; CHECK-NEXT:    [[EXTRACTED4:%.*]] = trunc i32 [[SHIFTED3]] to i16
-; CHECK-NEXT:    [[TMP8:%.*]] = bitcast i16 [[EXTRACTED4]] to half
+; CHECK-NEXT:    [[SHIFTED2:%.*]] = lshr i32 [[NEWLOADED]], [[SHIFTAMT]]
+; CHECK-NEXT:    [[EXTRACTED3:%.*]] = trunc i32 [[SHIFTED2]] to i16
+; CHECK-NEXT:    [[TMP8:%.*]] = bitcast i16 [[EXTRACTED3]] to half
 ; CHECK-NEXT:    fence seq_cst
 ; CHECK-NEXT:    ret half [[TMP8]]
 ;
diff --git a/llvm/test/Transforms/CallSiteSplitting/callsite-split-debug.ll b/llvm/test/Transforms/CallSiteSplitting/callsite-split-debug.ll
index 8f10dcb30d7b..68c906d616c9 100644
--- a/llvm/test/Transforms/CallSiteSplitting/callsite-split-debug.ll
+++ b/llvm/test/Transforms/CallSiteSplitting/callsite-split-debug.ll
@@ -1,4 +1,4 @@
-; RUN: opt -S -passes=callsite-splitting -o - < %s | FileCheck %s
+; RUN: opt -S -passes=callsite-splitting -o - < %s | FileCheck %s --check-prefixes=CHECK,CHECK-DEBUG
 ; RUN: opt -S -strip-debug -passes=callsite-splitting -o - < %s | FileCheck %s
 
 define internal i16 @bar(i16 %p1, i16 %p2) {
@@ -8,6 +8,9 @@ define internal i16 @bar(i16 %p1, i16 %p2) {
 
 define i16 @foo(i16 %in) {
 bb0:
+  %a = alloca i16, align 4, !DIAssignID !12
+  call void @llvm.dbg.assign(metadata i1 undef, metadata !11, metadata !DIExpression(), metadata !12, metadata ptr %a, metadata !DIExpression()), !dbg !8
+  store i16 7, ptr %a, align 4, !DIAssignID !13
   br label %bb1
 
 bb1:
@@ -20,13 +23,21 @@ bb2:
 CallsiteBB:
   %1 = phi i16 [ 0, %bb1 ], [ 1, %bb2 ]
   %c = phi i16 [ 2, %bb1 ], [ 3, %bb2 ]
+  %p = phi ptr [ %a, %bb1 ], [ %a, %bb2 ]
+  call void @llvm.dbg.value(metadata i16 %1, metadata !7, metadata !DIExpression()), !dbg !8
   call void @llvm.dbg.value(metadata i16 %c, metadata !7, metadata !DIExpression()), !dbg !8
+  call void @llvm.dbg.value(metadata !DIArgList(i16 %1, i16 %c), metadata !7, metadata !DIExpression()), !dbg !8
+  call void @llvm.dbg.value(metadata !DIArgList(i16 %c, i16 %c), metadata !7, metadata !DIExpression()), !dbg !8
+  call void @llvm.dbg.assign(metadata i16 %1, metadata !11, metadata !DIExpression(), metadata !13, metadata ptr %a, metadata !DIExpression()), !dbg !8
+  call void @llvm.dbg.assign(metadata i16 %c, metadata !11, metadata !DIExpression(), metadata !13, metadata ptr %a, metadata !DIExpression()), !dbg !8
+  call void @llvm.dbg.assign(metadata i16 %1, metadata !11, metadata !DIExpression(), metadata !13, metadata ptr %p, metadata !DIExpression()), !dbg !8
   %2 = call i16 @bar(i16 %1, i16 5)
   ret i16 %2
 }
 
 ; Function Attrs: nounwind readnone speculatable
 declare void @llvm.dbg.value(metadata, metadata, metadata) #0
+declare void @llvm.dbg.assign(metadata, metadata, metadata, metadata, metadata, metadata)
 
 attributes #0 = { nounwind readnone speculatable }
 
@@ -43,14 +54,37 @@ attributes #0 = { nounwind readnone speculatable }
 !6 = distinct !DISubprogram(name: "foo", scope: !1, file: !1, line: 4, unit: !0)
 !7 = !DILocalVariable(name: "c", scope: !6, line: 5, type: !5)
 !8 = !DILocation(line: 5, column: 7, scope: !6)
+!11 = !DILocalVariable(name: "a", scope: !6, line: 6, type: !5)
+!12 = distinct !DIAssignID()
+!13 = distinct !DIAssignID()
 
 ; The optimization should trigger even in the presence of the dbg.value in
 ; CallSiteBB.
 
 ; CHECK-LABEL: @foo
 ; CHECK-LABEL: bb1.split:
+; CHECK-DEBUG: call void @llvm.dbg.value(metadata i16 0, metadata ![[DBG_1:[0-9]+]], {{.*}}
+; CHECK-DEBUG: call void @llvm.dbg.value(metadata i16 2, metadata ![[DBG_1]], {{.*}}
+; CHECK-DEBUG: call void @llvm.dbg.value(metadata !DIArgList(i16 0, i16 2), {{.*}}
+; CHECK-DEBUG: call void @llvm.dbg.value(metadata !DIArgList(i16 2, i16 2), {{.*}}
+; CHECK-DEBUG: call void @llvm.dbg.assign(metadata i16 0, metadata ![[DBG_2:[0-9]+]], {{.*}}
+; CHECK-DEBUG: call void @llvm.dbg.assign(metadata i16 2, metadata ![[DBG_2]], {{.*}}
+; CHECK-DEBUG: call void @llvm.dbg.assign(metadata i16 0, metadata ![[DBG_2]], metadata !DIExpression(), metadata ![[ID_1:[0-9]+]], metadata ptr %a, {{.*}}
 ; CHECK: [[TMP1:%[0-9]+]] = call i16 @bar(i16 0, i16 5)
+
 ; CHECK-LABEL: bb2.split:
+; CHECK-DEBUG: call void @llvm.dbg.value(metadata i16 1, metadata ![[DBG_1]], {{.*}}
+; CHECK-DEBUG: call void @llvm.dbg.value(metadata i16 3, metadata ![[DBG_1]], {{.*}}
+; CHECK-DEBUG: call void @llvm.dbg.value(metadata !DIArgList(i16 1, i16 3), {{.*}}
+; CHECK-DEBUG: call void @llvm.dbg.value(metadata !DIArgList(i16 3, i16 3), {{.*}}
+; CHECK-DEBUG: call void @llvm.dbg.assign(metadata i16 1, metadata ![[DBG_2]], {{.*}}
+; CHECK-DEBUG: call void @llvm.dbg.assign(metadata i16 3, metadata ![[DBG_2]], {{.*}}
+; CHECK-DEBUG: call void @llvm.dbg.assign(metadata i16 1, metadata ![[DBG_2]], metadata !DIExpression(), metadata ![[ID_1:[0-9]+]], metadata ptr %a, {{.*}}
 ; CHECK: [[TMP2:%[0-9]+]] = call i16 @bar(i16 1, i16 5)
+
 ; CHECK-LABEL: CallsiteBB
 ; CHECK: %phi.call = phi i16 [ [[TMP2]], %bb2.split ], [ [[TMP1]], %bb1.split
+
+; CHECK-DEBUG-DAG: ![[DBG_1]] = !DILocalVariable(name: "c"{{.*}})
+; CHECK-DEBUG-DAG: ![[DBG_2]] = !DILocalVariable(name: "a"{{.*}})
+; CHECK-DEBUG-DAG: ![[ID_1]] = distinct !DIAssignID()
diff --git a/llvm/test/Transforms/CodeGenPrepare/ARM/branch-on-zero.ll b/llvm/test/Transforms/CodeGenPrepare/ARM/branch-on-zero.ll
index ff5cef7e781f..25dfb3c53a07 100644
--- a/llvm/test/Transforms/CodeGenPrepare/ARM/branch-on-zero.ll
+++ b/llvm/test/Transforms/CodeGenPrepare/ARM/branch-on-zero.ll
@@ -211,6 +211,29 @@ else:
   ret i32 %l
 }
 
+define i32 @sub10_else_drop_nuw(i32 %a) {
+; CHECK-LABEL: @sub10_else_drop_nuw(
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[L:%.*]] = sub i32 [[A:%.*]], 10
+; CHECK-NEXT:    [[TMP0:%.*]] = icmp eq i32 [[L]], 0
+; CHECK-NEXT:    br i1 [[TMP0]], label [[THEN:%.*]], label [[ELSE:%.*]]
+; CHECK:       then:
+; CHECK-NEXT:    ret i32 0
+; CHECK:       else:
+; CHECK-NEXT:    ret i32 [[L]]
+;
+entry:
+  %c = icmp eq i32 %a, 10
+  br i1 %c, label %then, label %else
+
+then:
+  ret i32 0
+
+else:
+  %l = sub nuw i32 %a, 10
+  ret i32 %l
+}
+
 define i32 @subm10_then(i32 %a) {
 ; CHECK-LABEL: @subm10_then(
 ; CHECK-NEXT:  entry:
diff --git a/llvm/test/Transforms/CodeGenPrepare/RISCV/convert-to-eqz.ll b/llvm/test/Transforms/CodeGenPrepare/RISCV/convert-to-eqz.ll
new file mode 100644
index 000000000000..a6909d149134
--- /dev/null
+++ b/llvm/test/Transforms/CodeGenPrepare/RISCV/convert-to-eqz.ll
@@ -0,0 +1,80 @@
+; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --version 4
+; RUN: opt -codegenprepare -S -mtriple=riscv64 < %s | FileCheck %s
+
+define i8 @hoist_add(i8 %x) {
+; CHECK-LABEL: define i8 @hoist_add(
+; CHECK-SAME: i8 [[X:%.*]]) {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[INC:%.*]] = add i8 [[X]], 1
+; CHECK-NEXT:    [[TMP0:%.*]] = icmp eq i8 [[INC]], 0
+; CHECK-NEXT:    br i1 [[TMP0]], label [[EXIT:%.*]], label [[IF_THEN:%.*]]
+; CHECK:       if.then:
+; CHECK-NEXT:    br label [[EXIT]]
+; CHECK:       exit:
+; CHECK-NEXT:    [[RETVAL:%.*]] = phi i8 [ [[INC]], [[IF_THEN]] ], [ -1, [[ENTRY:%.*]] ]
+; CHECK-NEXT:    ret i8 [[RETVAL]]
+;
+entry:
+  %cmp = icmp eq i8 %x, -1
+  br i1 %cmp, label %exit, label %if.then
+
+if.then:
+  %inc = add nuw nsw i8 %x, 1
+  br label %exit
+
+exit:
+  %retval = phi i8 [ %inc, %if.then ], [ -1, %entry ]
+  ret i8 %retval
+}
+
+define i8 @hoist_lshr(i8 %x) {
+; CHECK-LABEL: define i8 @hoist_lshr(
+; CHECK-SAME: i8 [[X:%.*]]) {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[INC:%.*]] = lshr i8 [[X]], 3
+; CHECK-NEXT:    [[TMP0:%.*]] = icmp eq i8 [[INC]], 0
+; CHECK-NEXT:    br i1 [[TMP0]], label [[EXIT:%.*]], label [[IF_THEN:%.*]]
+; CHECK:       if.then:
+; CHECK-NEXT:    br label [[EXIT]]
+; CHECK:       exit:
+; CHECK-NEXT:    [[RETVAL:%.*]] = phi i8 [ [[INC]], [[IF_THEN]] ], [ -1, [[ENTRY:%.*]] ]
+; CHECK-NEXT:    ret i8 [[RETVAL]]
+;
+entry:
+  %cmp = icmp ult i8 %x, 8
+  br i1 %cmp, label %exit, label %if.then
+
+if.then:
+  %inc = lshr exact i8 %x, 3
+  br label %exit
+
+exit:
+  %retval = phi i8 [ %inc, %if.then ], [ -1, %entry ]
+  ret i8 %retval
+}
+
+define i8 @nomove_add(i8 %x) {
+; CHECK-LABEL: define i8 @nomove_add(
+; CHECK-SAME: i8 [[X:%.*]]) {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[INC:%.*]] = add i8 [[X]], 1
+; CHECK-NEXT:    [[TMP0:%.*]] = icmp eq i8 [[INC]], 0
+; CHECK-NEXT:    br i1 [[TMP0]], label [[EXIT:%.*]], label [[IF_THEN:%.*]]
+; CHECK:       if.then:
+; CHECK-NEXT:    br label [[EXIT]]
+; CHECK:       exit:
+; CHECK-NEXT:    [[RETVAL:%.*]] = phi i8 [ [[INC]], [[IF_THEN]] ], [ -1, [[ENTRY:%.*]] ]
+; CHECK-NEXT:    ret i8 [[RETVAL]]
+;
+entry:
+  %inc = add nuw nsw i8 %x, 1
+  %cmp = icmp eq i8 %x, -1
+  br i1 %cmp, label %exit, label %if.then
+
+if.then:
+  br label %exit
+
+exit:
+  %retval = phi i8 [ %inc, %if.then ], [ -1, %entry ]
+  ret i8 %retval
+}
diff --git a/llvm/test/Transforms/CodeGenPrepare/RISCV/noop-copy-sink.ll b/llvm/test/Transforms/CodeGenPrepare/RISCV/noop-copy-sink.ll
new file mode 100644
index 000000000000..55cde6c1431f
--- /dev/null
+++ b/llvm/test/Transforms/CodeGenPrepare/RISCV/noop-copy-sink.ll
@@ -0,0 +1,30 @@
+; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
+; RUN: opt -S -passes='require<profile-summary>,function(codegenprepare)' -mtriple=riscv64 %s \
+; RUN:   | FileCheck --check-prefixes=CHECK %s
+
+define i16 @sink_trunc1(i64 %a) {
+; CHECK-LABEL: @sink_trunc1(
+; CHECK-NEXT:  fnend:
+; CHECK-NEXT:    [[TMP0:%.*]] = trunc i64 [[A:%.*]] to i16
+; CHECK-NEXT:    ret i16 [[TMP0]]
+;
+  %trunc = trunc i64 %a to i16
+  br label %fnend
+
+fnend:
+  ret i16 %trunc
+}
+
+; The flags on the original trunc should be preserved.
+define i16 @sink_trunc2(i64 %a) {
+; CHECK-LABEL: @sink_trunc2(
+; CHECK-NEXT:  fnend:
+; CHECK-NEXT:    [[TMP0:%.*]] = trunc nuw nsw i64 [[A:%.*]] to i16
+; CHECK-NEXT:    ret i16 [[TMP0]]
+;
+  %trunc = trunc nuw nsw i64 %a to i16
+  br label %fnend
+
+fnend:
+  ret i16 %trunc
+}
diff --git a/llvm/test/Transforms/Coroutines/coro-split-musttail10.ll b/llvm/test/Transforms/Coroutines/coro-split-musttail10.ll
index cdd58b2a084f..3e91b79c10f7 100644
--- a/llvm/test/Transforms/Coroutines/coro-split-musttail10.ll
+++ b/llvm/test/Transforms/Coroutines/coro-split-musttail10.ll
@@ -1,9 +1,12 @@
 ; Tests that we would convert coro.resume to a musttail call if the target is
-; Wasm64 with tail-call support.
-; RUN: opt < %s -passes='cgscc(coro-split),simplifycfg,early-cse' -S | FileCheck %s
-; RUN: opt < %s -passes='pgo-instr-gen,cgscc(coro-split),simplifycfg,early-cse' -S | FileCheck %s
+; Wasm64 or Wasm32 with tail-call support.
+; REQUIRES: webassembly-registered-target
 
-target triple = "wasm64-unknown-unknown"
+; RUN: opt -mtriple=wasm64-unknown-unknown < %s -passes='cgscc(coro-split),simplifycfg,early-cse' -S | FileCheck %s
+; RUN: opt -mtriple=wasm64-unknown-unknown < %s -passes='pgo-instr-gen,cgscc(coro-split),simplifycfg,early-cse' -S | FileCheck %s
+
+; RUN: opt -mtriple=wasm32-unknown-unknown < %s -passes='cgscc(coro-split),simplifycfg,early-cse' -S | FileCheck %s
+; RUN: opt -mtriple=wasm32-unknown-unknown < %s -passes='pgo-instr-gen,cgscc(coro-split),simplifycfg,early-cse' -S | FileCheck %s
 
 define void @f() #0 {
 entry:
diff --git a/llvm/test/Transforms/Coroutines/coro-split-musttail11.ll b/llvm/test/Transforms/Coroutines/coro-split-musttail11.ll
deleted file mode 100644
index da5d868280e9..000000000000
--- a/llvm/test/Transforms/Coroutines/coro-split-musttail11.ll
+++ /dev/null
@@ -1,55 +0,0 @@
-; Tests that we would convert coro.resume to a musttail call if the target is
-; Wasm32 with tail-call support.
-; RUN: opt < %s -passes='cgscc(coro-split),simplifycfg,early-cse' -S | FileCheck %s
-; RUN: opt < %s -passes='pgo-instr-gen,cgscc(coro-split),simplifycfg,early-cse' -S | FileCheck %s
-
-target triple = "wasm32-unknown-unknown"
-
-define void @f() #0 {
-entry:
-  %id = call token @llvm.coro.id(i32 0, ptr null, ptr null, ptr null)
-  %alloc = call ptr @malloc(i64 16) #3
-  %vFrame = call noalias nonnull ptr @llvm.coro.begin(token %id, ptr %alloc)
-
-  %save = call token @llvm.coro.save(ptr null)
-  %addr1 = call ptr @llvm.coro.subfn.addr(ptr null, i8 0)
-  call fastcc void %addr1(ptr null)
-
-  %suspend = call i8 @llvm.coro.suspend(token %save, i1 false)
-  switch i8 %suspend, label %exit [
-    i8 0, label %await.ready
-    i8 1, label %exit
-  ]
-await.ready:
-  %save2 = call token @llvm.coro.save(ptr null)
-  %addr2 = call ptr @llvm.coro.subfn.addr(ptr null, i8 0)
-  call fastcc void %addr2(ptr null)
-
-  %suspend2 = call i8 @llvm.coro.suspend(token %save2, i1 false)
-  switch i8 %suspend2, label %exit [
-    i8 0, label %exit
-    i8 1, label %exit
-  ]
-exit:
-  call i1 @llvm.coro.end(ptr null, i1 false, token none)
-  ret void
-}
-
-; CHECK: musttail call
-
-declare token @llvm.coro.id(i32, ptr readnone, ptr nocapture readonly, ptr) #1
-declare i1 @llvm.coro.alloc(token) #2
-declare i64 @llvm.coro.size.i64() #3
-declare ptr @llvm.coro.begin(token, ptr writeonly) #2
-declare token @llvm.coro.save(ptr) #2
-declare ptr @llvm.coro.frame() #3
-declare i8 @llvm.coro.suspend(token, i1) #2
-declare ptr @llvm.coro.free(token, ptr nocapture readonly) #1
-declare i1 @llvm.coro.end(ptr, i1, token) #2
-declare ptr @llvm.coro.subfn.addr(ptr nocapture readonly, i8) #1
-declare ptr @malloc(i64)
-
-attributes #0 = { presplitcoroutine "target-features"="+tail-call" }
-attributes #1 = { argmemonly nounwind readonly }
-attributes #2 = { nounwind }
-attributes #3 = { nounwind readnone }
diff --git a/llvm/test/Transforms/FunctionAttrs/noundef.ll b/llvm/test/Transforms/FunctionAttrs/noundef.ll
index 64deda546c06..b7c583880501 100644
--- a/llvm/test/Transforms/FunctionAttrs/noundef.ll
+++ b/llvm/test/Transforms/FunctionAttrs/noundef.ll
@@ -167,9 +167,8 @@ define i64 @test_trunc_with_constexpr() {
   ret i64 %conv
 }
 
-; FIXME: This is a miscompile.
 define align 4 ptr @maybe_not_aligned(ptr noundef %p) {
-; CHECK-LABEL: define noundef align 4 ptr @maybe_not_aligned(
+; CHECK-LABEL: define align 4 ptr @maybe_not_aligned(
 ; CHECK-SAME: ptr noundef readnone returned [[P:%.*]]) #[[ATTR0]] {
 ; CHECK-NEXT:    ret ptr [[P]]
 ;
@@ -184,9 +183,8 @@ define align 4 ptr @definitely_aligned(ptr noundef align 4 %p) {
   ret ptr %p
 }
 
-; FIXME: This is a miscompile.
 define nonnull ptr @maybe_not_nonnull(ptr noundef %p) {
-; CHECK-LABEL: define noundef nonnull ptr @maybe_not_nonnull(
+; CHECK-LABEL: define nonnull ptr @maybe_not_nonnull(
 ; CHECK-SAME: ptr noundef readnone returned [[P:%.*]]) #[[ATTR0]] {
 ; CHECK-NEXT:    ret ptr [[P]]
 ;
@@ -201,9 +199,8 @@ define nonnull ptr @definitely_nonnull(ptr noundef nonnull %p) {
   ret ptr %p
 }
 
-; FIXME: This is a miscompile.
 define range(i8 0, 10) i8 @maybe_not_in_range(i8 noundef %v) {
-; CHECK-LABEL: define noundef range(i8 0, 10) i8 @maybe_not_in_range(
+; CHECK-LABEL: define range(i8 0, 10) i8 @maybe_not_in_range(
 ; CHECK-SAME: i8 noundef returned [[V:%.*]]) #[[ATTR0]] {
 ; CHECK-NEXT:    ret i8 [[V]]
 ;
diff --git a/llvm/test/Transforms/FunctionSpecialization/discover-transitive-phis.ll b/llvm/test/Transforms/FunctionSpecialization/discover-transitive-phis.ll
index b4c24715037b..d0095231a30f 100644
--- a/llvm/test/Transforms/FunctionSpecialization/discover-transitive-phis.ll
+++ b/llvm/test/Transforms/FunctionSpecialization/discover-transitive-phis.ll
@@ -1,22 +1,22 @@
-; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
+; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --version 4
 ;
 ; RUN: opt -passes="ipsccp<func-spec>" -funcspec-min-function-size=20 -funcspec-for-literal-constant -S < %s | FileCheck %s --check-prefix=FUNCSPEC
 ; RUN: opt -passes="ipsccp<func-spec>" -funcspec-min-function-size=20 -funcspec-for-literal-constant -funcspec-max-discovery-iterations=16 -S < %s | FileCheck %s --check-prefix=NOFUNCSPEC
 
 define i64 @bar(i1 %c1, i1 %c2, i1 %c3, i1 %c4, i1 %c5, i1 %c6, i1 %c7, i1 %c8, i1 %c9, i1 %c10) {
-; FUNCSPEC-LABEL: define i64 @bar(
+; FUNCSPEC-LABEL: define range(i64 4, 13) i64 @bar(
 ; FUNCSPEC-SAME: i1 [[C1:%.*]], i1 [[C2:%.*]], i1 [[C3:%.*]], i1 [[C4:%.*]], i1 [[C5:%.*]], i1 [[C6:%.*]], i1 [[C7:%.*]], i1 [[C8:%.*]], i1 [[C9:%.*]], i1 [[C10:%.*]]) {
 ; FUNCSPEC-NEXT:  entry:
-; FUNCSPEC-NEXT:    [[F1:%.*]] = call i64 @foo.specialized.1(i64 3, i1 [[C1]], i1 [[C2]], i1 [[C3]], i1 [[C4]], i1 [[C5]], i1 [[C6]], i1 [[C7]], i1 [[C8]], i1 [[C9]], i1 [[C10]]), !range [[RNG0:![0-9]+]]
-; FUNCSPEC-NEXT:    [[F2:%.*]] = call i64 @foo.specialized.2(i64 4, i1 [[C1]], i1 [[C2]], i1 [[C3]], i1 [[C4]], i1 [[C5]], i1 [[C6]], i1 [[C7]], i1 [[C8]], i1 [[C9]], i1 [[C10]]), !range [[RNG1:![0-9]+]]
+; FUNCSPEC-NEXT:    [[F1:%.*]] = call i64 @foo.specialized.1(i64 3, i1 [[C1]], i1 [[C2]], i1 [[C3]], i1 [[C4]], i1 [[C5]], i1 [[C6]], i1 [[C7]], i1 [[C8]], i1 [[C9]], i1 [[C10]])
+; FUNCSPEC-NEXT:    [[F2:%.*]] = call i64 @foo.specialized.2(i64 4, i1 [[C1]], i1 [[C2]], i1 [[C3]], i1 [[C4]], i1 [[C5]], i1 [[C6]], i1 [[C7]], i1 [[C8]], i1 [[C9]], i1 [[C10]])
 ; FUNCSPEC-NEXT:    [[ADD:%.*]] = add nuw nsw i64 [[F1]], [[F2]]
 ; FUNCSPEC-NEXT:    ret i64 [[ADD]]
 ;
-; NOFUNCSPEC-LABEL: define i64 @bar(
+; NOFUNCSPEC-LABEL: define range(i64 4, 13) i64 @bar(
 ; NOFUNCSPEC-SAME: i1 [[C1:%.*]], i1 [[C2:%.*]], i1 [[C3:%.*]], i1 [[C4:%.*]], i1 [[C5:%.*]], i1 [[C6:%.*]], i1 [[C7:%.*]], i1 [[C8:%.*]], i1 [[C9:%.*]], i1 [[C10:%.*]]) {
 ; NOFUNCSPEC-NEXT:  entry:
-; NOFUNCSPEC-NEXT:    [[F1:%.*]] = call i64 @foo(i64 3, i1 [[C1]], i1 [[C2]], i1 [[C3]], i1 [[C4]], i1 [[C5]], i1 [[C6]], i1 [[C7]], i1 [[C8]], i1 [[C9]], i1 [[C10]]), !range [[RNG0:![0-9]+]]
-; NOFUNCSPEC-NEXT:    [[F2:%.*]] = call i64 @foo(i64 4, i1 [[C1]], i1 [[C2]], i1 [[C3]], i1 [[C4]], i1 [[C5]], i1 [[C6]], i1 [[C7]], i1 [[C8]], i1 [[C9]], i1 [[C10]]), !range [[RNG0]]
+; NOFUNCSPEC-NEXT:    [[F1:%.*]] = call i64 @foo(i64 3, i1 [[C1]], i1 [[C2]], i1 [[C3]], i1 [[C4]], i1 [[C5]], i1 [[C6]], i1 [[C7]], i1 [[C8]], i1 [[C9]], i1 [[C10]])
+; NOFUNCSPEC-NEXT:    [[F2:%.*]] = call i64 @foo(i64 4, i1 [[C1]], i1 [[C2]], i1 [[C3]], i1 [[C4]], i1 [[C5]], i1 [[C6]], i1 [[C7]], i1 [[C8]], i1 [[C9]], i1 [[C10]])
 ; NOFUNCSPEC-NEXT:    [[ADD:%.*]] = add nuw nsw i64 [[F1]], [[F2]]
 ; NOFUNCSPEC-NEXT:    ret i64 [[ADD]]
 ;
@@ -28,6 +28,50 @@ entry:
 }
 
 define internal i64 @foo(i64 %n, i1 %c1, i1 %c2, i1 %c3, i1 %c4, i1 %c5, i1 %c6, i1 %c7, i1 %c8, i1 %c9, i1 %c10) {
+; NOFUNCSPEC-LABEL: define internal range(i64 2, 7) i64 @foo(
+; NOFUNCSPEC-SAME: i64 [[N:%.*]], i1 [[C1:%.*]], i1 [[C2:%.*]], i1 [[C3:%.*]], i1 [[C4:%.*]], i1 [[C5:%.*]], i1 [[C6:%.*]], i1 [[C7:%.*]], i1 [[C8:%.*]], i1 [[C9:%.*]], i1 [[C10:%.*]]) {
+; NOFUNCSPEC-NEXT:  entry:
+; NOFUNCSPEC-NEXT:    br i1 [[C1]], label [[L1:%.*]], label [[L9:%.*]]
+; NOFUNCSPEC:       l1:
+; NOFUNCSPEC-NEXT:    [[PHI1:%.*]] = phi i64 [ [[N]], [[ENTRY:%.*]] ], [ [[PHI2:%.*]], [[L2:%.*]] ]
+; NOFUNCSPEC-NEXT:    [[ADD:%.*]] = add nuw nsw i64 [[PHI1]], 1
+; NOFUNCSPEC-NEXT:    br i1 [[C2]], label [[L1_5:%.*]], label [[EXIT:%.*]]
+; NOFUNCSPEC:       l1_5:
+; NOFUNCSPEC-NEXT:    br i1 [[C3]], label [[L1_75:%.*]], label [[L6:%.*]]
+; NOFUNCSPEC:       l1_75:
+; NOFUNCSPEC-NEXT:    br i1 [[C4]], label [[L2]], label [[L3:%.*]]
+; NOFUNCSPEC:       l2:
+; NOFUNCSPEC-NEXT:    [[PHI2]] = phi i64 [ [[PHI1]], [[L1_75]] ], [ [[PHI3:%.*]], [[L3]] ]
+; NOFUNCSPEC-NEXT:    br label [[L1]]
+; NOFUNCSPEC:       l3:
+; NOFUNCSPEC-NEXT:    [[PHI3]] = phi i64 [ [[PHI1]], [[L1_75]] ], [ [[PHI4:%.*]], [[L4:%.*]] ]
+; NOFUNCSPEC-NEXT:    br label [[L2]]
+; NOFUNCSPEC:       l4:
+; NOFUNCSPEC-NEXT:    [[PHI4]] = phi i64 [ [[PHI5:%.*]], [[L5:%.*]] ], [ [[PHI6:%.*]], [[L6]] ]
+; NOFUNCSPEC-NEXT:    br i1 [[C5]], label [[L3]], label [[L6]]
+; NOFUNCSPEC:       l5:
+; NOFUNCSPEC-NEXT:    [[PHI5]] = phi i64 [ [[PHI6]], [[L6_5:%.*]] ], [ [[PHI7:%.*]], [[L7:%.*]] ]
+; NOFUNCSPEC-NEXT:    br label [[L4]]
+; NOFUNCSPEC:       l6:
+; NOFUNCSPEC-NEXT:    [[PHI6]] = phi i64 [ [[PHI4]], [[L4]] ], [ [[PHI1]], [[L1_5]] ]
+; NOFUNCSPEC-NEXT:    br i1 [[C6]], label [[L4]], label [[L6_5]]
+; NOFUNCSPEC:       l6_5:
+; NOFUNCSPEC-NEXT:    br i1 [[C7]], label [[L5]], label [[L8:%.*]]
+; NOFUNCSPEC:       l7:
+; NOFUNCSPEC-NEXT:    [[PHI7]] = phi i64 [ [[PHI9:%.*]], [[L9]] ], [ [[PHI8:%.*]], [[L8]] ]
+; NOFUNCSPEC-NEXT:    br i1 [[C8]], label [[L5]], label [[L8]]
+; NOFUNCSPEC:       l8:
+; NOFUNCSPEC-NEXT:    [[PHI8]] = phi i64 [ [[PHI6]], [[L6_5]] ], [ [[PHI7]], [[L7]] ]
+; NOFUNCSPEC-NEXT:    br i1 [[C9]], label [[L7]], label [[L9]]
+; NOFUNCSPEC:       l9:
+; NOFUNCSPEC-NEXT:    [[PHI9]] = phi i64 [ [[N]], [[ENTRY]] ], [ [[PHI8]], [[L8]] ]
+; NOFUNCSPEC-NEXT:    [[SUB:%.*]] = sub nuw nsw i64 [[PHI9]], 1
+; NOFUNCSPEC-NEXT:    [[MUL:%.*]] = mul nuw nsw i64 [[SUB]], 2
+; NOFUNCSPEC-NEXT:    br i1 [[C10]], label [[L7]], label [[EXIT]]
+; NOFUNCSPEC:       exit:
+; NOFUNCSPEC-NEXT:    [[RES:%.*]] = phi i64 [ 2, [[L1]] ], [ [[MUL]], [[L9]] ]
+; NOFUNCSPEC-NEXT:    ret i64 [[RES]]
+;
 entry:
   br i1 %c1, label %l1, label %l9
 
diff --git a/llvm/test/Transforms/FunctionSpecialization/global-var-constants.ll b/llvm/test/Transforms/FunctionSpecialization/global-var-constants.ll
index b9481baae60b..a576d9aa32e1 100644
--- a/llvm/test/Transforms/FunctionSpecialization/global-var-constants.ll
+++ b/llvm/test/Transforms/FunctionSpecialization/global-var-constants.ll
@@ -49,10 +49,10 @@ entry:
 ; Check if specialisation on the address of a non-const global variable
 ; is not allowed, then it is not performed.
 
-; NO-GLOBALS-LABEL: define internal i32 @g()
+; NO-GLOBALS-LABEL: define internal range(i32 -2147483646, -2147483648) i32 @g()
 ; NO-GLOBALS: call i32 @f(ptr @G)
 
-; NO-GLOBALS-LABEL: define i32 @h0(ptr %p)
+; NO-GLOBALS-LABEL: define range(i32 -2147483646, -2147483648) i32 @h0(ptr %p)
 ; NO-GLOBALS:call i32 @g()
 
 ; NO-GLOBALS-LABEL: define i32 @h1()
@@ -64,10 +64,10 @@ entry:
 ; Check if specialisation on the address of a non-const global variable
 ; is allowed, then it is performed where possible.
 
-; GLOBALS-LABEL: define internal i32 @g()
+; GLOBALS-LABEL: define internal range(i32 -2147483646, -2147483648) i32 @g()
 ; GLOBALS: call i32 @f.specialized.2()
 
-; GLOBALS-LABEL: define i32 @h0(ptr %p)
+; GLOBALS-LABEL: define range(i32 -2147483646, -2147483648) i32 @h0(ptr %p)
 ; GLOBALS: call i32 @g()
 
 ; GLOBALS-LABEL: define i32 @h1()
diff --git a/llvm/test/Transforms/FunctionSpecialization/literal-const.ll b/llvm/test/Transforms/FunctionSpecialization/literal-const.ll
index f107ffe0ec7e..3eae3dc261fb 100644
--- a/llvm/test/Transforms/FunctionSpecialization/literal-const.ll
+++ b/llvm/test/Transforms/FunctionSpecialization/literal-const.ll
@@ -71,10 +71,10 @@ entry:
 ; CHECK-LIT-LABEL: define i32 @f1
 ; CHECK-LIT: call i32 @neg.specialized.[[#B:]]
 
-; CHECK-LIT-LABEL: define i32 @g0
+; CHECK-LIT-LABEL: define range(i32 -2147483647, -2147483648) i32 @g0
 ; CHECK-LIT: call i32 @add.specialized.[[#C:]]
 
-; CHECK-LIT-LABEL: define i32 @g1
+; CHECK-LIT-LABEL: define range(i32 -2147483647, -2147483648) i32 @g1
 ; CHECK-LIT: call i32 @add.specialized.[[#D:]]
 
 ; CHECK-LIT-LABEL: define float @h0
diff --git a/llvm/test/Transforms/GVN/condprop.ll b/llvm/test/Transforms/GVN/condprop.ll
index 6b1e4d106010..6402a2315772 100644
--- a/llvm/test/Transforms/GVN/condprop.ll
+++ b/llvm/test/Transforms/GVN/condprop.ll
@@ -214,11 +214,11 @@ define void @test4(i1 %b, i32 %x) {
 ; CHECK-NEXT:    br i1 [[B:%.*]], label [[SW:%.*]], label [[CASE3:%.*]]
 ; CHECK:       sw:
 ; CHECK-NEXT:    switch i32 [[X:%.*]], label [[DEFAULT:%.*]] [
-; CHECK-NEXT:    i32 0, label [[CASE0:%.*]]
-; CHECK-NEXT:    i32 1, label [[CASE1:%.*]]
-; CHECK-NEXT:    i32 2, label [[CASE0]]
-; CHECK-NEXT:    i32 3, label [[CASE3]]
-; CHECK-NEXT:    i32 4, label [[DEFAULT]]
+; CHECK-NEXT:      i32 0, label [[CASE0:%.*]]
+; CHECK-NEXT:      i32 1, label [[CASE1:%.*]]
+; CHECK-NEXT:      i32 2, label [[CASE0]]
+; CHECK-NEXT:      i32 3, label [[CASE3]]
+; CHECK-NEXT:      i32 4, label [[DEFAULT]]
 ; CHECK-NEXT:    ]
 ; CHECK:       default:
 ; CHECK-NEXT:    call void @bar(i32 [[X]])
@@ -521,15 +521,16 @@ define i32 @test13(ptr %ptr1, ptr %ptr2) {
 ; CHECK-NEXT:    [[GEP1:%.*]] = getelementptr i32, ptr [[PTR2:%.*]], i32 1
 ; CHECK-NEXT:    [[GEP2:%.*]] = getelementptr i32, ptr [[PTR2]], i32 2
 ; CHECK-NEXT:    [[CMP:%.*]] = icmp eq ptr [[PTR1:%.*]], [[PTR2]]
-; CHECK-NEXT:    [[VAL2_PRE:%.*]] = load i32, ptr [[GEP2]], align 4
 ; CHECK-NEXT:    br i1 [[CMP]], label [[IF:%.*]], label [[END:%.*]]
 ; CHECK:       if:
+; CHECK-NEXT:    [[VAL1:%.*]] = load i32, ptr [[GEP2]], align 4
 ; CHECK-NEXT:    br label [[END]]
 ; CHECK:       end:
-; CHECK-NEXT:    [[PHI1:%.*]] = phi ptr [ [[PTR2]], [[IF]] ], [ [[GEP1]], [[ENTRY:%.*]] ]
-; CHECK-NEXT:    [[PHI2:%.*]] = phi i32 [ [[VAL2_PRE]], [[IF]] ], [ 0, [[ENTRY]] ]
+; CHECK-NEXT:    [[PHI1:%.*]] = phi ptr [ [[PTR1]], [[IF]] ], [ [[GEP1]], [[ENTRY:%.*]] ]
+; CHECK-NEXT:    [[PHI2:%.*]] = phi i32 [ [[VAL1]], [[IF]] ], [ 0, [[ENTRY]] ]
 ; CHECK-NEXT:    store i32 0, ptr [[PHI1]], align 4
-; CHECK-NEXT:    [[RET:%.*]] = add i32 [[PHI2]], [[VAL2_PRE]]
+; CHECK-NEXT:    [[VAL2:%.*]] = load i32, ptr [[GEP2]], align 4
+; CHECK-NEXT:    [[RET:%.*]] = add i32 [[PHI2]], [[VAL2]]
 ; CHECK-NEXT:    ret i32 [[RET]]
 ;
 entry:
@@ -552,14 +553,14 @@ end:
   ret i32 %ret
 }
 
-define void @test14(ptr %ptr1, ptr noalias %ptr2) {
+define void @test14(ptr %ptr1, ptr noalias %ptr2, i1 %b1, i1 %b2) {
 ; CHECK-LABEL: @test14(
 ; CHECK-NEXT:  entry:
 ; CHECK-NEXT:    [[GEP1:%.*]] = getelementptr inbounds i32, ptr [[PTR1:%.*]], i32 1
 ; CHECK-NEXT:    [[GEP2:%.*]] = getelementptr inbounds i32, ptr [[PTR1]], i32 2
 ; CHECK-NEXT:    br label [[LOOP:%.*]]
 ; CHECK:       loop:
-; CHECK-NEXT:    br i1 undef, label [[LOOP_IF1_CRIT_EDGE:%.*]], label [[THEN:%.*]]
+; CHECK-NEXT:    br i1 [[B1:%.*]], label [[LOOP_IF1_CRIT_EDGE:%.*]], label [[THEN:%.*]]
 ; CHECK:       loop.if1_crit_edge:
 ; CHECK-NEXT:    [[VAL2_PRE:%.*]] = load i32, ptr [[GEP2]], align 4
 ; CHECK-NEXT:    br label [[IF1:%.*]]
@@ -574,10 +575,10 @@ define void @test14(ptr %ptr1, ptr noalias %ptr2) {
 ; CHECK:       if2:
 ; CHECK-NEXT:    br label [[LOOP_END]]
 ; CHECK:       loop.end:
-; CHECK-NEXT:    [[PHI3:%.*]] = phi ptr [ [[PTR2]], [[THEN]] ], [ [[PTR1]], [[IF2]] ]
+; CHECK-NEXT:    [[PHI3:%.*]] = phi ptr [ [[GEP2]], [[THEN]] ], [ [[PTR1]], [[IF2]] ]
 ; CHECK-NEXT:    [[VAL3]] = load i32, ptr [[GEP2]], align 4
 ; CHECK-NEXT:    store i32 [[VAL3]], ptr [[PHI3]], align 4
-; CHECK-NEXT:    br i1 undef, label [[LOOP]], label [[IF1]]
+; CHECK-NEXT:    br i1 [[B2:%.*]], label [[LOOP]], label [[IF1]]
 ;
 entry:
   %gep1 = getelementptr inbounds i32, ptr %ptr1, i32 1
@@ -586,7 +587,7 @@ entry:
 
 loop:
   %phi1 = phi ptr [ %gep3, %loop.end ], [ %gep1, %entry ]
-  br i1 undef, label %if1, label %then
+  br i1 %b1, label %if1, label %then
 
 
 if1:
@@ -607,5 +608,201 @@ loop.end:
   %val3 = load i32, ptr %gep2, align 4
   store i32 %val3, ptr %phi3, align 4
   %gep3 = getelementptr inbounds i32, ptr %ptr1, i32 1
-  br i1 undef, label %loop, label %if1
+  br i1 %b2, label %loop, label %if1
+}
+
+; Make sure that the call to use_ptr does not have %p1
+define void @single_phi1(ptr %p0, ptr %p1, i8 %s) {
+; CHECK-LABEL: @single_phi1(
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[P2:%.*]] = load ptr, ptr [[P0:%.*]], align 8
+; CHECK-NEXT:    [[CMP1:%.*]] = icmp eq ptr [[P2]], [[P1:%.*]]
+; CHECK-NEXT:    br i1 [[CMP1]], label [[BB4:%.*]], label [[BB1:%.*]]
+; CHECK:       bb1:
+; CHECK-NEXT:    switch i8 [[S:%.*]], label [[BB2:%.*]] [
+; CHECK-NEXT:      i8 0, label [[BB1]]
+; CHECK-NEXT:      i8 1, label [[BB3:%.*]]
+; CHECK-NEXT:    ]
+; CHECK:       bb2:
+; CHECK-NEXT:    unreachable
+; CHECK:       bb3:
+; CHECK-NEXT:    br label [[BB4]]
+; CHECK:       bb4:
+; CHECK-NEXT:    call void @use_bool(i1 [[CMP1]])
+; CHECK-NEXT:    call void @use_ptr(ptr [[P2]])
+; CHECK-NEXT:    ret void
+;
+entry:
+  %p2 = load ptr, ptr %p0, align 8
+  %cmp1 = icmp eq ptr %p2, %p1
+  br i1 %cmp1, label %bb4, label %bb1
+
+bb1:
+  switch i8 %s, label %bb2 [
+  i8 0, label %bb1
+  i8 1, label %bb3
+  ]
+
+bb2:
+  unreachable
+
+bb3:
+  br label %bb4
+
+bb4:
+  %phi1 = phi ptr [ %p2, %entry ], [ %p2, %bb3 ]
+  %cmp2 = icmp eq ptr %phi1, %p1
+  call void @use_bool(i1 %cmp2)
+  call void @use_ptr(ptr %phi1)
+  ret void
+}
+
+define void @single_phi2(ptr %p0, ptr %p1, i8 %s) {
+; CHECK-LABEL: @single_phi2(
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[P2:%.*]] = load ptr, ptr [[P0:%.*]], align 8
+; CHECK-NEXT:    [[CMP1:%.*]] = icmp eq ptr [[P2]], [[P1:%.*]]
+; CHECK-NEXT:    br i1 [[CMP1]], label [[BB4:%.*]], label [[BB1:%.*]]
+; CHECK:       bb1:
+; CHECK-NEXT:    switch i8 [[S:%.*]], label [[BB2:%.*]] [
+; CHECK-NEXT:      i8 0, label [[BB1]]
+; CHECK-NEXT:      i8 1, label [[BB3:%.*]]
+; CHECK-NEXT:    ]
+; CHECK:       bb2:
+; CHECK-NEXT:    br label [[BB4]]
+; CHECK:       bb3:
+; CHECK-NEXT:    br label [[BB4]]
+; CHECK:       bb4:
+; CHECK-NEXT:    call void @use_bool(i1 [[CMP1]])
+; CHECK-NEXT:    call void @use_ptr(ptr [[P2]])
+; CHECK-NEXT:    ret void
+;
+entry:
+  %p2 = load ptr, ptr %p0, align 8
+  %cmp1 = icmp eq ptr %p2, %p1
+  br i1 %cmp1, label %bb4, label %bb1
+
+bb1:
+  switch i8 %s, label %bb2 [
+  i8 0, label %bb1
+  i8 1, label %bb3
+  ]
+
+bb2:
+  br label %bb4
+
+bb3:
+  br label %bb4
+
+bb4:
+  %phi1 = phi ptr [ %p2, %entry ], [ %p2, %bb2 ], [ %p2, %bb3 ]
+  %cmp2 = icmp eq ptr %phi1, %p1
+  call void @use_bool(i1 %cmp2)
+  call void @use_ptr(ptr %phi1)
+  ret void
 }
+
+define void @multiple_phi1(ptr %p0, ptr %p1, i8 %s) {
+; CHECK-LABEL: @multiple_phi1(
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[P2:%.*]] = load ptr, ptr [[P0:%.*]], align 8
+; CHECK-NEXT:    [[CMP1:%.*]] = icmp eq ptr [[P2]], [[P1:%.*]]
+; CHECK-NEXT:    br i1 [[CMP1]], label [[BB4:%.*]], label [[BB1:%.*]]
+; CHECK:       bb1:
+; CHECK-NEXT:    switch i8 [[S:%.*]], label [[BB2:%.*]] [
+; CHECK-NEXT:      i8 0, label [[BB1]]
+; CHECK-NEXT:      i8 1, label [[BB3:%.*]]
+; CHECK-NEXT:    ]
+; CHECK:       bb2:
+; CHECK-NEXT:    unreachable
+; CHECK:       bb3:
+; CHECK-NEXT:    br label [[BB4]]
+; CHECK:       bb4:
+; CHECK-NEXT:    call void @use_bool(i1 [[CMP1]])
+; CHECK-NEXT:    br label [[BB5:%.*]]
+; CHECK:       bb5:
+; CHECK-NEXT:    call void @use_ptr(ptr [[P2]])
+; CHECK-NEXT:    br label [[BB5]]
+;
+entry:
+  %p2 = load ptr, ptr %p0, align 8
+  %cmp1 = icmp eq ptr %p2, %p1
+  br i1 %cmp1, label %bb4, label %bb1
+
+bb1:
+  switch i8 %s, label %bb2 [
+  i8 0, label %bb1
+  i8 1, label %bb3
+  ]
+
+bb2:
+  unreachable
+
+bb3:
+  br label %bb4
+
+bb4:
+  %phi1 = phi ptr [ %p2, %entry ], [ poison, %bb3 ]
+  %cmp2 = icmp eq ptr %phi1, %p1
+  call void @use_bool(i1 %cmp2)
+  br label %bb5
+
+bb5:
+  %phi2 = phi ptr [ poison, %bb5 ], [ %phi1, %bb4 ]
+  call void @use_ptr(ptr %phi2)
+  br label %bb5
+}
+
+define void @multiple_phi2(ptr %p0, ptr %p1, i8 %s) {
+; CHECK-LABEL: @multiple_phi2(
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[P2:%.*]] = load ptr, ptr [[P0:%.*]], align 8
+; CHECK-NEXT:    [[CMP1:%.*]] = icmp eq ptr [[P2]], [[P1:%.*]]
+; CHECK-NEXT:    br i1 [[CMP1]], label [[BB4:%.*]], label [[BB1:%.*]]
+; CHECK:       bb1:
+; CHECK-NEXT:    switch i8 [[S:%.*]], label [[BB2:%.*]] [
+; CHECK-NEXT:      i8 0, label [[BB1]]
+; CHECK-NEXT:      i8 1, label [[BB3:%.*]]
+; CHECK-NEXT:    ]
+; CHECK:       bb2:
+; CHECK-NEXT:    br label [[BB4]]
+; CHECK:       bb3:
+; CHECK-NEXT:    br label [[BB4]]
+; CHECK:       bb4:
+; CHECK-NEXT:    call void @use_bool(i1 [[CMP1]])
+; CHECK-NEXT:    br label [[BB5:%.*]]
+; CHECK:       bb5:
+; CHECK-NEXT:    call void @use_ptr(ptr [[P2]])
+; CHECK-NEXT:    br label [[BB5]]
+;
+entry:
+  %p2 = load ptr, ptr %p0, align 8
+  %cmp1 = icmp eq ptr %p2, %p1
+  br i1 %cmp1, label %bb4, label %bb1
+
+bb1:
+  switch i8 %s, label %bb2 [
+  i8 0, label %bb1
+  i8 1, label %bb3
+  ]
+
+bb2:
+  br label %bb4
+
+bb3:
+  br label %bb4
+
+bb4:
+  %phi1 = phi ptr [ %p2, %entry ], [ %p2, %bb2 ], [ poison, %bb3 ]
+  %cmp2 = icmp eq ptr %phi1, %p1
+  call void @use_bool(i1 %cmp2)
+  br label %bb5
+
+bb5:
+  %phi2 = phi ptr [ poison, %bb5 ], [ %phi1, %bb4 ]
+  call void @use_ptr(ptr %phi2)
+  br label %bb5
+}
+
+declare void @use_bool(i1)
+declare void @use_ptr(ptr)
diff --git a/llvm/test/Transforms/GlobalOpt/basictest.ll b/llvm/test/Transforms/GlobalOpt/basictest.ll
index 6d7fcdd96dfd..72d38a1e8845 100644
--- a/llvm/test/Transforms/GlobalOpt/basictest.ll
+++ b/llvm/test/Transforms/GlobalOpt/basictest.ll
@@ -1,9 +1,23 @@
+; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --version 4
 ; RUN: opt < %s -passes=globalopt -S | FileCheck %s
 
-; CHECK-NOT: global
 @X = internal global i32 4              ; <ptr> [#uses=1]
 
 define i32 @foo() {
-        %V = load i32, ptr @X               ; <i32> [#uses=1]
-        ret i32 %V
+; CHECK-LABEL: define i32 @foo() local_unnamed_addr {
+; CHECK-NEXT:    ret i32 4
+;
+  %V = load i32, ptr @X               ; <i32> [#uses=1]
+  ret i32 %V
+}
+
+@X_tls = internal thread_local global i32 13
+
+define i32 @bar() {
+; CHECK-LABEL: define i32 @bar() local_unnamed_addr {
+; CHECK-NEXT:    ret i32 13
+;
+  %p = call ptr @llvm.threadlocal.address(ptr @X_tls)
+  %v = load i32, ptr %p
+  ret i32 %v
 }
diff --git a/llvm/test/Transforms/GlobalOpt/constantfold-initializers.ll b/llvm/test/Transforms/GlobalOpt/constantfold-initializers.ll
index ca844f63937c..f82942e73d92 100644
--- a/llvm/test/Transforms/GlobalOpt/constantfold-initializers.ll
+++ b/llvm/test/Transforms/GlobalOpt/constantfold-initializers.ll
@@ -72,11 +72,12 @@ entry:
 }
 
 @threadlocalptr = global ptr null, align 4
-; CHECK: @threadlocalptr = global ptr null, align 4
+; CHECK: @threadlocalptr = local_unnamed_addr global ptr null, align 4
 @threadlocalvar = external thread_local global i32
 define internal void @test5() {
 entry:
-  store ptr @threadlocalvar, ptr @threadlocalptr, align 4
+  %p = call ptr @llvm.threadlocal.address(ptr @threadlocalvar)
+  store ptr %p, ptr @threadlocalptr, align 4
   ret void
 }
 
diff --git a/llvm/test/Transforms/GlobalOpt/stored-once-forward-value.ll b/llvm/test/Transforms/GlobalOpt/stored-once-forward-value.ll
index 7b845070bbd0..2b7ceb4169f3 100644
--- a/llvm/test/Transforms/GlobalOpt/stored-once-forward-value.ll
+++ b/llvm/test/Transforms/GlobalOpt/stored-once-forward-value.ll
@@ -39,12 +39,14 @@ define i32 @dom_arg(i32 %a) {
 
 define ptr @dom_thread_local_global() {
 ; CHECK-LABEL: @dom_thread_local_global(
-; CHECK-NEXT:    store ptr @tl, ptr @g3, align 8
+; CHECK-NEXT:    [[P:%.*]] = call ptr @llvm.threadlocal.address.p0(ptr @tl)
+; CHECK-NEXT:    store ptr [[P]], ptr @g3, align 8
 ; CHECK-NEXT:    call void @b()
 ; CHECK-NEXT:    [[R:%.*]] = load ptr, ptr @g3, align 8
 ; CHECK-NEXT:    ret ptr [[R]]
 ;
-  store ptr @tl, ptr @g3
+  %p = call ptr @llvm.threadlocal.address(ptr @tl)
+  store ptr %p, ptr @g3
   call void @b()
   %r = load ptr, ptr @g3
   ret ptr %r
diff --git a/llvm/test/Transforms/GlobalOpt/tls.ll b/llvm/test/Transforms/GlobalOpt/tls.ll
index 6ba003ff30b2..2cc2ea4e366e 100644
--- a/llvm/test/Transforms/GlobalOpt/tls.ll
+++ b/llvm/test/Transforms/GlobalOpt/tls.ll
@@ -15,14 +15,16 @@ declare void @start_thread(ptr)
 define i32 @f() {
 entry:
   ; Set @ip to point to x[1] for thread 1.
-  store ptr getelementptr inbounds ([100 x i32], ptr @x, i64 0, i64 1), ptr @ip, align 8
+  %p = call ptr @llvm.threadlocal.address(ptr @x)
+  %addr = getelementptr inbounds [100 x i32], ptr %p, i64 0, i64 1
+  store ptr %addr, ptr @ip, align 8
 
   ; Run g on a new thread.
   tail call void @start_thread(ptr @g) nounwind
   tail call void @wait() nounwind
 
   ; Reset x[1] for thread 1.
-  store i32 0, ptr getelementptr inbounds ([100 x i32], ptr @x, i64 0, i64 1), align 4
+  store i32 0, ptr %addr, align 4
 
   ; Read the value of @ip, which now points at x[1] for thread 2.
   %0 = load ptr, ptr @ip, align 8
@@ -39,10 +41,12 @@ entry:
 define internal void @g() nounwind uwtable {
 entry:
   ; Set @ip to point to x[1] for thread 2.
-  store ptr getelementptr inbounds ([100 x i32], ptr @x, i64 0, i64 1), ptr @ip, align 8
+  %p = call ptr @llvm.threadlocal.address(ptr @x)
+  %addr = getelementptr inbounds [100 x i32], ptr %p, i64 0, i64 1
+  store ptr %addr, ptr @ip, align 8
 
   ; Store 50 in x[1] for thread 2.
-  store i32 50, ptr getelementptr inbounds ([100 x i32], ptr @x, i64 0, i64 1), align 4
+  store i32 50, ptr %addr, align 4
 
   tail call void @signal() nounwind
   ret void
diff --git a/llvm/test/Transforms/IRCE/pr89959.ll b/llvm/test/Transforms/IRCE/pr89959.ll
new file mode 100644
index 000000000000..dc7c0dfbc57a
--- /dev/null
+++ b/llvm/test/Transforms/IRCE/pr89959.ll
@@ -0,0 +1,33 @@
+; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --version 4
+; RUN: opt -passes=irce -S < %s 2>&1 | FileCheck %s
+
+; Make sure we don't crash.
+define void @pr89959() {
+; CHECK-LABEL: define void @pr89959() {
+; CHECK-NEXT:  top:
+; CHECK-NEXT:    br label [[L3:%.*]]
+; CHECK:       L3:
+; CHECK-NEXT:    [[VALUE_PHI:%.*]] = phi ptr [ null, [[TOP:%.*]] ], [ [[TMP0:%.*]], [[L13:%.*]] ]
+; CHECK-NEXT:    [[TMP0]] = getelementptr i8, ptr [[VALUE_PHI]], i64 8
+; CHECK-NEXT:    [[DOTNOT:%.*]] = icmp ule ptr [[VALUE_PHI]], null
+; CHECK-NEXT:    br i1 [[DOTNOT]], label [[L13]], label [[L15:%.*]]
+; CHECK:       L13:
+; CHECK-NEXT:    br label [[L3]]
+; CHECK:       L15:
+; CHECK-NEXT:    ret void
+;
+top:
+  br label %L3
+
+L3:
+  %value_phi = phi ptr [ null, %top ], [ %0, %L13 ]
+  %0 = getelementptr i8, ptr %value_phi, i64 8
+  %.not = icmp ule ptr %value_phi, null
+  br i1 %.not, label %L13, label %L15
+
+L13:
+  br label %L3
+
+L15:
+  ret void
+}
diff --git a/llvm/test/Transforms/Inline/inline-deferred-instsimplify.ll b/llvm/test/Transforms/Inline/inline-deferred-instsimplify.ll
new file mode 100644
index 000000000000..4a9c576f0271
--- /dev/null
+++ b/llvm/test/Transforms/Inline/inline-deferred-instsimplify.ll
@@ -0,0 +1,76 @@
+; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --version 4
+; RUN: opt < %s -passes=inline -S | FileCheck %s
+; RUN: opt < %s -passes='cgscc(inline)' -S | FileCheck %s
+
+%struct.a = type { i32, i32, i32, i32, i32 }
+
+@g_var = global %struct.a { i32 1, i32 0, i32 0, i32 0, i32 0 }, align 8
+@other_g_var = global %struct.a zeroinitializer, align 4
+
+define void @callee(ptr noundef byval(%struct.a) align 8 %ptr) {
+; CHECK-LABEL: define void @callee(
+; CHECK-SAME: ptr noundef byval([[STRUCT_A:%.*]]) align 8 [[PTR:%.*]]) {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[VAL:%.*]] = load i32, ptr [[PTR]], align 8
+; CHECK-NEXT:    [[DOTNOT:%.*]] = icmp eq i32 [[VAL]], 0
+; CHECK-NEXT:    br i1 [[DOTNOT]], label [[CHECK_POINTERS_ARE_EQUAL:%.*]], label [[STORE_PTR_IN_GVAR:%.*]]
+; CHECK:       store_ptr_in_gvar:
+; CHECK-NEXT:    store ptr [[PTR]], ptr @other_g_var, align 8
+; CHECK-NEXT:    br label [[CHECK_POINTERS_ARE_EQUAL]]
+; CHECK:       check_pointers_are_equal:
+; CHECK-NEXT:    [[PHI:%.*]] = phi ptr [ [[PTR]], [[STORE_PTR_IN_GVAR]] ], [ @other_g_var, [[ENTRY:%.*]] ]
+; CHECK-NEXT:    [[DOTNOT1:%.*]] = icmp eq ptr [[PHI]], [[PTR]]
+; CHECK-NEXT:    br i1 [[DOTNOT1]], label [[RETURN:%.*]], label [[ABORT:%.*]]
+; CHECK:       abort:
+; CHECK-NEXT:    call void @abort()
+; CHECK-NEXT:    unreachable
+; CHECK:       return:
+; CHECK-NEXT:    ret void
+;
+entry:
+  %val = load i32, ptr %ptr, align 8
+  %.not = icmp eq i32 %val, 0
+  br i1 %.not, label %check_pointers_are_equal, label %store_ptr_in_gvar
+
+store_ptr_in_gvar:                                ; preds = %entry
+  store ptr %ptr, ptr @other_g_var, align 8
+  br label %check_pointers_are_equal
+
+check_pointers_are_equal:                         ; preds = %store_ptr_in_gvar, %entry
+  %phi = phi ptr [ %ptr, %store_ptr_in_gvar ], [ @other_g_var, %entry ]
+; FIXME: While inlining, the following is miscompiled to i1 false,
+; as %ptr in the phi-node is not taken into account.
+  %.not1 = icmp eq ptr %phi, %ptr
+  br i1 %.not1, label %return, label %abort
+
+abort:                                            ; preds = %check_pointers_are_equal
+  call void @abort()
+  unreachable
+
+return:                                           ; preds = %check_pointers_are_equal
+  ret void
+}
+
+define i32 @main() {
+; CHECK-LABEL: define i32 @main() {
+; CHECK-NEXT:    [[G_VAR:%.*]] = alloca [[STRUCT_A:%.*]], align 8
+; CHECK-NEXT:    call void @llvm.lifetime.start.p0(i64 20, ptr [[G_VAR]])
+; CHECK-NEXT:    call void @llvm.memcpy.p0.p0.i64(ptr align 1 [[G_VAR]], ptr align 1 @g_var, i64 20, i1 false)
+; CHECK-NEXT:    [[VAL_I:%.*]] = load i32, ptr [[G_VAR]], align 8
+; CHECK-NEXT:    [[DOTNOT_I:%.*]] = icmp eq i32 [[VAL_I]], 0
+; CHECK-NEXT:    br i1 [[DOTNOT_I]], label [[CHECK_POINTERS_ARE_EQUAL_I:%.*]], label [[STORE_PTR_IN_GVAR_I:%.*]]
+; CHECK:       store_ptr_in_gvar.i:
+; CHECK-NEXT:    store ptr [[G_VAR]], ptr @other_g_var, align 8
+; CHECK-NEXT:    br label [[CHECK_POINTERS_ARE_EQUAL_I]]
+; CHECK:       check_pointers_are_equal.i:
+; CHECK-NEXT:    [[PHI_I:%.*]] = phi ptr [ [[G_VAR]], [[STORE_PTR_IN_GVAR_I]] ], [ @other_g_var, [[TMP0:%.*]] ]
+; CHECK-NEXT:    call void @abort()
+; CHECK-NEXT:    unreachable
+; CHECK:       callee.exit:
+; CHECK-NEXT:    ret i32 0
+;
+  call void @callee(ptr noundef byval(%struct.a) align 8 @g_var)
+  ret i32 0
+}
+
+declare void @abort()
diff --git a/llvm/test/Transforms/InstCombine/add4.ll b/llvm/test/Transforms/InstCombine/add4.ll
index 7fc164c8b9a7..77f7fc7b35cd 100644
--- a/llvm/test/Transforms/InstCombine/add4.ll
+++ b/llvm/test/Transforms/InstCombine/add4.ll
@@ -1,6 +1,8 @@
 ; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
 ; RUN: opt < %s -passes=instcombine -S | FileCheck %s
 
+declare void @use(i32)
+
 define i64 @match_unsigned(i64 %x) {
 ; CHECK-LABEL: @match_unsigned(
 ; CHECK-NEXT:    [[UREM:%.*]] = urem i64 [[X:%.*]], 19136
@@ -127,3 +129,163 @@ define i32 @not_match_overflow(i32 %x) {
   %t4 = add i32 %t, %t3
   ret i32 %t4
 }
+
+; Tests from PR76128.
+define i32 @fold_add_udiv_urem(i32 noundef %val) {
+; CHECK-LABEL: @fold_add_udiv_urem(
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[DIV:%.*]] = udiv i32 [[VAL:%.*]], 10
+; CHECK-NEXT:    [[TMP0:%.*]] = mul nuw i32 [[DIV]], 6
+; CHECK-NEXT:    [[ADD:%.*]] = add i32 [[TMP0]], [[VAL]]
+; CHECK-NEXT:    ret i32 [[ADD]]
+;
+entry:
+  %div = udiv i32 %val, 10
+  %shl = shl i32 %div, 4
+  %rem = urem i32 %val, 10
+  %add = add i32 %shl, %rem
+  ret i32 %add
+}
+define i32 @fold_add_sdiv_srem(i32 noundef %val) {
+; CHECK-LABEL: @fold_add_sdiv_srem(
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[DIV:%.*]] = sdiv i32 [[VAL:%.*]], 10
+; CHECK-NEXT:    [[TMP0:%.*]] = mul nsw i32 [[DIV]], 6
+; CHECK-NEXT:    [[ADD:%.*]] = add i32 [[TMP0]], [[VAL]]
+; CHECK-NEXT:    ret i32 [[ADD]]
+;
+entry:
+  %div = sdiv i32 %val, 10
+  %shl = shl i32 %div, 4
+  %rem = srem i32 %val, 10
+  %add = add i32 %shl, %rem
+  ret i32 %add
+}
+define i32 @fold_add_udiv_urem_to_mul(i32 noundef %val) {
+; CHECK-LABEL: @fold_add_udiv_urem_to_mul(
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[ADD:%.*]] = mul i32 [[VAL:%.*]], 3
+; CHECK-NEXT:    ret i32 [[ADD]]
+;
+entry:
+  %div = udiv i32 %val, 7
+  %mul1 = mul i32 %div, 21
+  %rem = urem i32 %val, 7
+  %mul2 = mul i32 %rem, 3
+  %add = add i32 %mul1, %mul2
+  ret i32 %add
+}
+define i32 @fold_add_udiv_urem_to_mul_multiuse(i32 noundef %val) {
+; CHECK-LABEL: @fold_add_udiv_urem_to_mul_multiuse(
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[REM:%.*]] = urem i32 [[VAL:%.*]], 7
+; CHECK-NEXT:    call void @use(i32 [[REM]])
+; CHECK-NEXT:    [[ADD:%.*]] = mul i32 [[VAL]], 3
+; CHECK-NEXT:    ret i32 [[ADD]]
+;
+entry:
+  %div = udiv i32 %val, 7
+  %mul1 = mul i32 %div, 21
+  %rem = urem i32 %val, 7
+  call void @use(i32 %rem)
+  %mul2 = mul i32 %rem, 3
+  %add = add i32 %mul1, %mul2
+  ret i32 %add
+}
+define i32 @fold_add_udiv_urem_commuted(i32 noundef %val) {
+; CHECK-LABEL: @fold_add_udiv_urem_commuted(
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[DIV:%.*]] = udiv i32 [[VAL:%.*]], 10
+; CHECK-NEXT:    [[TMP0:%.*]] = mul nuw i32 [[DIV]], 6
+; CHECK-NEXT:    [[ADD:%.*]] = add i32 [[TMP0]], [[VAL]]
+; CHECK-NEXT:    ret i32 [[ADD]]
+;
+entry:
+  %div = udiv i32 %val, 10
+  %shl = shl i32 %div, 4
+  %rem = urem i32 %val, 10
+  %add = add i32 %rem, %shl
+  ret i32 %add
+}
+define i32 @fold_add_udiv_urem_or_disjoint(i32 noundef %val) {
+; CHECK-LABEL: @fold_add_udiv_urem_or_disjoint(
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[DIV:%.*]] = udiv i32 [[VAL:%.*]], 10
+; CHECK-NEXT:    [[TMP0:%.*]] = mul nuw i32 [[DIV]], 6
+; CHECK-NEXT:    [[ADD:%.*]] = add i32 [[TMP0]], [[VAL]]
+; CHECK-NEXT:    ret i32 [[ADD]]
+;
+entry:
+  %div = udiv i32 %val, 10
+  %shl = shl i32 %div, 4
+  %rem = urem i32 %val, 10
+  %add = or disjoint i32 %shl, %rem
+  ret i32 %add
+}
+; Negative tests
+define i32 @fold_add_udiv_urem_without_noundef(i32 %val) {
+; CHECK-LABEL: @fold_add_udiv_urem_without_noundef(
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[DIV:%.*]] = udiv i32 [[VAL:%.*]], 10
+; CHECK-NEXT:    [[SHL:%.*]] = shl i32 [[DIV]], 4
+; CHECK-NEXT:    [[REM:%.*]] = urem i32 [[VAL]], 10
+; CHECK-NEXT:    [[ADD:%.*]] = or disjoint i32 [[SHL]], [[REM]]
+; CHECK-NEXT:    ret i32 [[ADD]]
+;
+entry:
+  %div = udiv i32 %val, 10
+  %shl = shl i32 %div, 4
+  %rem = urem i32 %val, 10
+  %add = add i32 %shl, %rem
+  ret i32 %add
+}
+define i32 @fold_add_udiv_urem_multiuse_mul(i32 noundef %val) {
+; CHECK-LABEL: @fold_add_udiv_urem_multiuse_mul(
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[DIV:%.*]] = udiv i32 [[VAL:%.*]], 10
+; CHECK-NEXT:    [[SHL:%.*]] = shl i32 [[DIV]], 4
+; CHECK-NEXT:    call void @use(i32 [[SHL]])
+; CHECK-NEXT:    [[REM:%.*]] = urem i32 [[VAL]], 10
+; CHECK-NEXT:    [[ADD:%.*]] = or disjoint i32 [[SHL]], [[REM]]
+; CHECK-NEXT:    ret i32 [[ADD]]
+;
+entry:
+  %div = udiv i32 %val, 10
+  %shl = shl i32 %div, 4
+  call void @use(i32 %shl)
+  %rem = urem i32 %val, 10
+  %add = add i32 %shl, %rem
+  ret i32 %add
+}
+define i32 @fold_add_udiv_srem(i32 noundef %val) {
+; CHECK-LABEL: @fold_add_udiv_srem(
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[DIV:%.*]] = udiv i32 [[VAL:%.*]], 10
+; CHECK-NEXT:    [[SHL:%.*]] = shl i32 [[DIV]], 4
+; CHECK-NEXT:    [[REM:%.*]] = srem i32 [[VAL]], 10
+; CHECK-NEXT:    [[ADD:%.*]] = add i32 [[SHL]], [[REM]]
+; CHECK-NEXT:    ret i32 [[ADD]]
+;
+entry:
+  %div = udiv i32 %val, 10
+  %shl = shl i32 %div, 4
+  %rem = srem i32 %val, 10
+  %add = add i32 %shl, %rem
+  ret i32 %add
+}
+define i32 @fold_add_udiv_urem_non_constant(i32 noundef %val, i32 noundef %c) {
+; CHECK-LABEL: @fold_add_udiv_urem_non_constant(
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[DIV:%.*]] = udiv i32 [[VAL:%.*]], [[C:%.*]]
+; CHECK-NEXT:    [[SHL:%.*]] = shl i32 [[DIV]], 4
+; CHECK-NEXT:    [[REM:%.*]] = urem i32 [[VAL]], [[C]]
+; CHECK-NEXT:    [[ADD:%.*]] = add i32 [[SHL]], [[REM]]
+; CHECK-NEXT:    ret i32 [[ADD]]
+;
+entry:
+  %div = udiv i32 %val, %c
+  %shl = shl i32 %div, 4
+  %rem = urem i32 %val, %c
+  %add = add i32 %shl, %rem
+  ret i32 %add
+}
diff --git a/llvm/test/Transforms/InstCombine/array.ll b/llvm/test/Transforms/InstCombine/array.ll
index 236821d8ba4c..4f4ae17bebc5 100644
--- a/llvm/test/Transforms/InstCombine/array.ll
+++ b/llvm/test/Transforms/InstCombine/array.ll
@@ -108,3 +108,163 @@ entry:
   store i32 %b, ptr %gep
   ret void
 }
+
+define ptr @gep_inbounds_add_nsw_nonneg(ptr %ptr, i64 %a, i64 %b) {
+; CHECK-LABEL: define ptr @gep_inbounds_add_nsw_nonneg(
+; CHECK-SAME: ptr [[PTR:%.*]], i64 [[A:%.*]], i64 [[B:%.*]]) {
+; CHECK-NEXT:    [[A_NNEG:%.*]] = icmp sgt i64 [[A]], -1
+; CHECK-NEXT:    call void @llvm.assume(i1 [[A_NNEG]])
+; CHECK-NEXT:    [[B_NNEG:%.*]] = icmp sgt i64 [[B]], -1
+; CHECK-NEXT:    call void @llvm.assume(i1 [[B_NNEG]])
+; CHECK-NEXT:    [[TMP1:%.*]] = getelementptr inbounds i32, ptr [[PTR]], i64 [[A]]
+; CHECK-NEXT:    [[GEP:%.*]] = getelementptr inbounds i32, ptr [[TMP1]], i64 [[B]]
+; CHECK-NEXT:    ret ptr [[GEP]]
+;
+  %a.nneg = icmp sgt i64 %a, -1
+  call void @llvm.assume(i1 %a.nneg)
+  %b.nneg = icmp sgt i64 %b, -1
+  call void @llvm.assume(i1 %b.nneg)
+  %add = add nsw i64 %a, %b
+  %gep = getelementptr inbounds i32, ptr %ptr, i64 %add
+  ret ptr %gep
+}
+
+define ptr @gep_inbounds_add_nsw_not_nonneg1(ptr %ptr, i64 %a, i64 %b) {
+; CHECK-LABEL: define ptr @gep_inbounds_add_nsw_not_nonneg1(
+; CHECK-SAME: ptr [[PTR:%.*]], i64 [[A:%.*]], i64 [[B:%.*]]) {
+; CHECK-NEXT:    [[A_NNEG:%.*]] = icmp sgt i64 [[A]], -1
+; CHECK-NEXT:    call void @llvm.assume(i1 [[A_NNEG]])
+; CHECK-NEXT:    [[TMP1:%.*]] = getelementptr i32, ptr [[PTR]], i64 [[A]]
+; CHECK-NEXT:    [[GEP:%.*]] = getelementptr i32, ptr [[TMP1]], i64 [[B]]
+; CHECK-NEXT:    ret ptr [[GEP]]
+;
+  %a.nneg = icmp sgt i64 %a, -1
+  call void @llvm.assume(i1 %a.nneg)
+  %add = add nsw i64 %a, %b
+  %gep = getelementptr inbounds i32, ptr %ptr, i64 %add
+  ret ptr %gep
+}
+
+define ptr @gep_inbounds_add_nsw_not_nonneg2(ptr %ptr, i64 %a, i64 %b) {
+; CHECK-LABEL: define ptr @gep_inbounds_add_nsw_not_nonneg2(
+; CHECK-SAME: ptr [[PTR:%.*]], i64 [[A:%.*]], i64 [[B:%.*]]) {
+; CHECK-NEXT:    [[B_NNEG:%.*]] = icmp sgt i64 [[B]], -1
+; CHECK-NEXT:    call void @llvm.assume(i1 [[B_NNEG]])
+; CHECK-NEXT:    [[TMP1:%.*]] = getelementptr i32, ptr [[PTR]], i64 [[A]]
+; CHECK-NEXT:    [[GEP:%.*]] = getelementptr i32, ptr [[TMP1]], i64 [[B]]
+; CHECK-NEXT:    ret ptr [[GEP]]
+;
+  %b.nneg = icmp sgt i64 %b, -1
+  call void @llvm.assume(i1 %b.nneg)
+  %add = add nsw i64 %a, %b
+  %gep = getelementptr inbounds i32, ptr %ptr, i64 %add
+  ret ptr %gep
+}
+
+define ptr @gep_not_inbounds_add_nsw_nonneg(ptr %ptr, i64 %a, i64 %b) {
+; CHECK-LABEL: define ptr @gep_not_inbounds_add_nsw_nonneg(
+; CHECK-SAME: ptr [[PTR:%.*]], i64 [[A:%.*]], i64 [[B:%.*]]) {
+; CHECK-NEXT:    [[A_NNEG:%.*]] = icmp sgt i64 [[A]], -1
+; CHECK-NEXT:    call void @llvm.assume(i1 [[A_NNEG]])
+; CHECK-NEXT:    [[B_NNEG:%.*]] = icmp sgt i64 [[B]], -1
+; CHECK-NEXT:    call void @llvm.assume(i1 [[B_NNEG]])
+; CHECK-NEXT:    [[TMP1:%.*]] = getelementptr i32, ptr [[PTR]], i64 [[A]]
+; CHECK-NEXT:    [[GEP:%.*]] = getelementptr i32, ptr [[TMP1]], i64 [[B]]
+; CHECK-NEXT:    ret ptr [[GEP]]
+;
+  %a.nneg = icmp sgt i64 %a, -1
+  call void @llvm.assume(i1 %a.nneg)
+  %b.nneg = icmp sgt i64 %b, -1
+  call void @llvm.assume(i1 %b.nneg)
+  %add = add nsw i64 %a, %b
+  %gep = getelementptr i32, ptr %ptr, i64 %add
+  ret ptr %gep
+}
+
+define ptr @gep_inbounds_add_not_nsw_nonneg(ptr %ptr, i64 %a, i64 %b) {
+; CHECK-LABEL: define ptr @gep_inbounds_add_not_nsw_nonneg(
+; CHECK-SAME: ptr [[PTR:%.*]], i64 [[A:%.*]], i64 [[B:%.*]]) {
+; CHECK-NEXT:    [[A_NNEG:%.*]] = icmp sgt i64 [[A]], -1
+; CHECK-NEXT:    call void @llvm.assume(i1 [[A_NNEG]])
+; CHECK-NEXT:    [[B_NNEG:%.*]] = icmp sgt i64 [[B]], -1
+; CHECK-NEXT:    call void @llvm.assume(i1 [[B_NNEG]])
+; CHECK-NEXT:    [[TMP1:%.*]] = getelementptr i32, ptr [[PTR]], i64 [[A]]
+; CHECK-NEXT:    [[GEP:%.*]] = getelementptr i32, ptr [[TMP1]], i64 [[B]]
+; CHECK-NEXT:    ret ptr [[GEP]]
+;
+  %a.nneg = icmp sgt i64 %a, -1
+  call void @llvm.assume(i1 %a.nneg)
+  %b.nneg = icmp sgt i64 %b, -1
+  call void @llvm.assume(i1 %b.nneg)
+  %add = add i64 %a, %b
+  %gep = getelementptr inbounds i32, ptr %ptr, i64 %add
+  ret ptr %gep
+}
+
+define ptr @gep_inbounds_sext_add_nonneg(ptr %ptr, i32 %a) {
+; CHECK-LABEL: define ptr @gep_inbounds_sext_add_nonneg(
+; CHECK-SAME: ptr [[PTR:%.*]], i32 [[A:%.*]]) {
+; CHECK-NEXT:    [[A_NNEG:%.*]] = icmp sgt i32 [[A]], -1
+; CHECK-NEXT:    call void @llvm.assume(i1 [[A_NNEG]])
+; CHECK-NEXT:    [[TMP1:%.*]] = zext nneg i32 [[A]] to i64
+; CHECK-NEXT:    [[TMP2:%.*]] = getelementptr inbounds i32, ptr [[PTR]], i64 [[TMP1]]
+; CHECK-NEXT:    [[GEP:%.*]] = getelementptr inbounds i8, ptr [[TMP2]], i64 40
+; CHECK-NEXT:    ret ptr [[GEP]]
+;
+  %a.nneg = icmp sgt i32 %a, -1
+  call void @llvm.assume(i1 %a.nneg)
+  %add = add nsw i32 %a, 10
+  %idx = sext i32 %add to i64
+  %gep = getelementptr inbounds i32, ptr %ptr, i64 %idx
+  ret ptr %gep
+}
+
+define ptr @gep_inbounds_sext_add_not_nonneg_1(ptr %ptr, i32 %a) {
+; CHECK-LABEL: define ptr @gep_inbounds_sext_add_not_nonneg_1(
+; CHECK-SAME: ptr [[PTR:%.*]], i32 [[A:%.*]]) {
+; CHECK-NEXT:    [[A_NNEG:%.*]] = icmp sgt i32 [[A]], -1
+; CHECK-NEXT:    call void @llvm.assume(i1 [[A_NNEG]])
+; CHECK-NEXT:    [[TMP1:%.*]] = zext nneg i32 [[A]] to i64
+; CHECK-NEXT:    [[TMP2:%.*]] = getelementptr i32, ptr [[PTR]], i64 [[TMP1]]
+; CHECK-NEXT:    [[GEP:%.*]] = getelementptr i8, ptr [[TMP2]], i64 -40
+; CHECK-NEXT:    ret ptr [[GEP]]
+;
+  %a.nneg = icmp sgt i32 %a, -1
+  call void @llvm.assume(i1 %a.nneg)
+  %add = add nsw i32 %a, -10
+  %idx = sext i32 %add to i64
+  %gep = getelementptr inbounds i32, ptr %ptr, i64 %idx
+  ret ptr %gep
+}
+
+define ptr @gep_inbounds_sext_add_not_nonneg_2(ptr %ptr, i32 %a) {
+; CHECK-LABEL: define ptr @gep_inbounds_sext_add_not_nonneg_2(
+; CHECK-SAME: ptr [[PTR:%.*]], i32 [[A:%.*]]) {
+; CHECK-NEXT:    [[TMP1:%.*]] = sext i32 [[A]] to i64
+; CHECK-NEXT:    [[TMP2:%.*]] = getelementptr i32, ptr [[PTR]], i64 [[TMP1]]
+; CHECK-NEXT:    [[GEP:%.*]] = getelementptr i8, ptr [[TMP2]], i64 40
+; CHECK-NEXT:    ret ptr [[GEP]]
+;
+  %add = add nsw i32 %a, 10
+  %idx = sext i32 %add to i64
+  %gep = getelementptr inbounds i32, ptr %ptr, i64 %idx
+  ret ptr %gep
+}
+
+define ptr @gep_not_inbounds_sext_add_nonneg(ptr %ptr, i32 %a) {
+; CHECK-LABEL: define ptr @gep_not_inbounds_sext_add_nonneg(
+; CHECK-SAME: ptr [[PTR:%.*]], i32 [[A:%.*]]) {
+; CHECK-NEXT:    [[A_NNEG:%.*]] = icmp sgt i32 [[A]], -1
+; CHECK-NEXT:    call void @llvm.assume(i1 [[A_NNEG]])
+; CHECK-NEXT:    [[TMP1:%.*]] = zext nneg i32 [[A]] to i64
+; CHECK-NEXT:    [[TMP2:%.*]] = getelementptr i32, ptr [[PTR]], i64 [[TMP1]]
+; CHECK-NEXT:    [[GEP:%.*]] = getelementptr i8, ptr [[TMP2]], i64 40
+; CHECK-NEXT:    ret ptr [[GEP]]
+;
+  %a.nneg = icmp sgt i32 %a, -1
+  call void @llvm.assume(i1 %a.nneg)
+  %add = add nsw i32 %a, 10
+  %idx = sext i32 %add to i64
+  %gep = getelementptr i32, ptr %ptr, i64 %idx
+  ret ptr %gep
+}
diff --git a/llvm/test/Transforms/InstCombine/bit_ceil.ll b/llvm/test/Transforms/InstCombine/bit_ceil.ll
index 52e70c78ba54..16631afa4878 100644
--- a/llvm/test/Transforms/InstCombine/bit_ceil.ll
+++ b/llvm/test/Transforms/InstCombine/bit_ceil.ll
@@ -5,7 +5,7 @@
 define i32 @bit_ceil_32(i32 %x) {
 ; CHECK-LABEL: @bit_ceil_32(
 ; CHECK-NEXT:    [[DEC:%.*]] = add i32 [[X:%.*]], -1
-; CHECK-NEXT:    [[CTLZ:%.*]] = tail call i32 @llvm.ctlz.i32(i32 [[DEC]], i1 false), !range [[RNG0:![0-9]+]]
+; CHECK-NEXT:    [[CTLZ:%.*]] = tail call range(i32 0, 33) i32 @llvm.ctlz.i32(i32 [[DEC]], i1 false)
 ; CHECK-NEXT:    [[TMP1:%.*]] = sub nsw i32 0, [[CTLZ]]
 ; CHECK-NEXT:    [[TMP2:%.*]] = and i32 [[TMP1]], 31
 ; CHECK-NEXT:    [[SEL:%.*]] = shl nuw i32 1, [[TMP2]]
@@ -24,7 +24,7 @@ define i32 @bit_ceil_32(i32 %x) {
 define i64 @bit_ceil_64(i64 %x) {
 ; CHECK-LABEL: @bit_ceil_64(
 ; CHECK-NEXT:    [[DEC:%.*]] = add i64 [[X:%.*]], -1
-; CHECK-NEXT:    [[CTLZ:%.*]] = tail call i64 @llvm.ctlz.i64(i64 [[DEC]], i1 false), !range [[RNG1:![0-9]+]]
+; CHECK-NEXT:    [[CTLZ:%.*]] = tail call range(i64 0, 65) i64 @llvm.ctlz.i64(i64 [[DEC]], i1 false)
 ; CHECK-NEXT:    [[TMP1:%.*]] = sub nsw i64 0, [[CTLZ]]
 ; CHECK-NEXT:    [[TMP2:%.*]] = and i64 [[TMP1]], 63
 ; CHECK-NEXT:    [[SEL:%.*]] = shl nuw i64 1, [[TMP2]]
@@ -44,7 +44,7 @@ define i32 @bit_ceil_32_minus_1(i32 %x) {
 ; CHECK-LABEL: @bit_ceil_32_minus_1(
 ; CHECK-NEXT:  entry:
 ; CHECK-NEXT:    [[SUB:%.*]] = add i32 [[X:%.*]], -2
-; CHECK-NEXT:    [[CTLZ:%.*]] = tail call i32 @llvm.ctlz.i32(i32 [[SUB]], i1 false), !range [[RNG0]]
+; CHECK-NEXT:    [[CTLZ:%.*]] = tail call range(i32 0, 33) i32 @llvm.ctlz.i32(i32 [[SUB]], i1 false)
 ; CHECK-NEXT:    [[TMP0:%.*]] = sub nsw i32 0, [[CTLZ]]
 ; CHECK-NEXT:    [[TMP1:%.*]] = and i32 [[TMP0]], 31
 ; CHECK-NEXT:    [[SEL:%.*]] = shl nuw i32 1, [[TMP1]]
@@ -64,7 +64,7 @@ entry:
 ; std::bit_ceil<uint32_t>(x + 1)
 define i32 @bit_ceil_32_plus_1(i32 %x) {
 ; CHECK-LABEL: @bit_ceil_32_plus_1(
-; CHECK-NEXT:    [[CTLZ:%.*]] = tail call i32 @llvm.ctlz.i32(i32 [[X:%.*]], i1 false), !range [[RNG0]]
+; CHECK-NEXT:    [[CTLZ:%.*]] = tail call range(i32 0, 33) i32 @llvm.ctlz.i32(i32 [[X:%.*]], i1 false)
 ; CHECK-NEXT:    [[TMP1:%.*]] = sub nsw i32 0, [[CTLZ]]
 ; CHECK-NEXT:    [[TMP2:%.*]] = and i32 [[TMP1]], 31
 ; CHECK-NEXT:    [[SEL:%.*]] = shl nuw i32 1, [[TMP2]]
@@ -84,7 +84,7 @@ define i32 @bit_ceil_plus_2(i32 %x) {
 ; CHECK-LABEL: @bit_ceil_plus_2(
 ; CHECK-NEXT:  entry:
 ; CHECK-NEXT:    [[SUB:%.*]] = add i32 [[X:%.*]], 1
-; CHECK-NEXT:    [[CTLZ:%.*]] = tail call i32 @llvm.ctlz.i32(i32 [[SUB]], i1 false), !range [[RNG0]]
+; CHECK-NEXT:    [[CTLZ:%.*]] = tail call range(i32 0, 33) i32 @llvm.ctlz.i32(i32 [[SUB]], i1 false)
 ; CHECK-NEXT:    [[TMP0:%.*]] = sub nsw i32 0, [[CTLZ]]
 ; CHECK-NEXT:    [[TMP1:%.*]] = and i32 [[TMP0]], 31
 ; CHECK-NEXT:    [[SEL:%.*]] = shl nuw i32 1, [[TMP1]]
@@ -105,7 +105,7 @@ define i32 @bit_ceil_32_neg(i32 %x) {
 ; CHECK-LABEL: @bit_ceil_32_neg(
 ; CHECK-NEXT:  entry:
 ; CHECK-NEXT:    [[SUB:%.*]] = xor i32 [[X:%.*]], -1
-; CHECK-NEXT:    [[CTLZ:%.*]] = tail call i32 @llvm.ctlz.i32(i32 [[SUB]], i1 false), !range [[RNG0]]
+; CHECK-NEXT:    [[CTLZ:%.*]] = tail call range(i32 0, 33) i32 @llvm.ctlz.i32(i32 [[SUB]], i1 false)
 ; CHECK-NEXT:    [[TMP0:%.*]] = sub nsw i32 0, [[CTLZ]]
 ; CHECK-NEXT:    [[TMP1:%.*]] = and i32 [[TMP0]], 31
 ; CHECK-NEXT:    [[SEL:%.*]] = shl nuw i32 1, [[TMP1]]
@@ -127,7 +127,7 @@ define i32 @bit_ceil_not(i32 %x) {
 ; CHECK-LABEL: @bit_ceil_not(
 ; CHECK-NEXT:  entry:
 ; CHECK-NEXT:    [[SUB:%.*]] = sub i32 -2, [[X:%.*]]
-; CHECK-NEXT:    [[CTLZ:%.*]] = tail call i32 @llvm.ctlz.i32(i32 [[SUB]], i1 false), !range [[RNG0]]
+; CHECK-NEXT:    [[CTLZ:%.*]] = tail call range(i32 0, 33) i32 @llvm.ctlz.i32(i32 [[SUB]], i1 false)
 ; CHECK-NEXT:    [[TMP0:%.*]] = sub nsw i32 0, [[CTLZ]]
 ; CHECK-NEXT:    [[TMP1:%.*]] = and i32 [[TMP0]], 31
 ; CHECK-NEXT:    [[SEL:%.*]] = shl nuw i32 1, [[TMP1]]
@@ -147,7 +147,7 @@ entry:
 define i32 @bit_ceil_commuted_operands(i32 %x) {
 ; CHECK-LABEL: @bit_ceil_commuted_operands(
 ; CHECK-NEXT:    [[DEC:%.*]] = add i32 [[X:%.*]], -1
-; CHECK-NEXT:    [[CTLZ:%.*]] = tail call i32 @llvm.ctlz.i32(i32 [[DEC]], i1 false), !range [[RNG0]]
+; CHECK-NEXT:    [[CTLZ:%.*]] = tail call range(i32 0, 33) i32 @llvm.ctlz.i32(i32 [[DEC]], i1 false)
 ; CHECK-NEXT:    [[SUB:%.*]] = sub nuw nsw i32 32, [[CTLZ]]
 ; CHECK-NEXT:    [[SHL:%.*]] = shl nuw i32 1, [[SUB]]
 ; CHECK-NEXT:    ret i32 [[SHL]]
@@ -165,7 +165,7 @@ define i32 @bit_ceil_commuted_operands(i32 %x) {
 define i32 @bit_ceil_wrong_select_constant(i32 %x) {
 ; CHECK-LABEL: @bit_ceil_wrong_select_constant(
 ; CHECK-NEXT:    [[DEC:%.*]] = add i32 [[X:%.*]], -1
-; CHECK-NEXT:    [[CTLZ:%.*]] = tail call i32 @llvm.ctlz.i32(i32 [[DEC]], i1 false), !range [[RNG0]]
+; CHECK-NEXT:    [[CTLZ:%.*]] = tail call range(i32 0, 33) i32 @llvm.ctlz.i32(i32 [[DEC]], i1 false)
 ; CHECK-NEXT:    [[SUB:%.*]] = sub nuw nsw i32 32, [[CTLZ]]
 ; CHECK-NEXT:    [[SHL:%.*]] = shl nuw i32 1, [[SUB]]
 ; CHECK-NEXT:    [[UGT_INV:%.*]] = icmp ult i32 [[X]], 2
@@ -185,7 +185,7 @@ define i32 @bit_ceil_wrong_select_constant(i32 %x) {
 define i32 @bit_ceil_32_wrong_cond(i32 %x) {
 ; CHECK-LABEL: @bit_ceil_32_wrong_cond(
 ; CHECK-NEXT:    [[DEC:%.*]] = add i32 [[X:%.*]], -1
-; CHECK-NEXT:    [[CTLZ:%.*]] = tail call i32 @llvm.ctlz.i32(i32 [[DEC]], i1 false), !range [[RNG0]]
+; CHECK-NEXT:    [[CTLZ:%.*]] = tail call range(i32 0, 33) i32 @llvm.ctlz.i32(i32 [[DEC]], i1 false)
 ; CHECK-NEXT:    [[SUB:%.*]] = sub nuw nsw i32 32, [[CTLZ]]
 ; CHECK-NEXT:    [[SHL:%.*]] = shl nuw i32 1, [[SUB]]
 ; CHECK-NEXT:    [[UGT:%.*]] = icmp ugt i32 [[X]], 2
@@ -205,7 +205,7 @@ define i32 @bit_ceil_32_wrong_cond(i32 %x) {
 define i32 @bit_ceil_wrong_sub_constant(i32 %x) {
 ; CHECK-LABEL: @bit_ceil_wrong_sub_constant(
 ; CHECK-NEXT:    [[DEC:%.*]] = add i32 [[X:%.*]], -1
-; CHECK-NEXT:    [[CTLZ:%.*]] = tail call i32 @llvm.ctlz.i32(i32 [[DEC]], i1 false), !range [[RNG0]]
+; CHECK-NEXT:    [[CTLZ:%.*]] = tail call range(i32 0, 33) i32 @llvm.ctlz.i32(i32 [[DEC]], i1 false)
 ; CHECK-NEXT:    [[SUB:%.*]] = sub nuw nsw i32 33, [[CTLZ]]
 ; CHECK-NEXT:    [[SHL:%.*]] = shl nuw i32 1, [[SUB]]
 ; CHECK-NEXT:    [[UGT:%.*]] = icmp ugt i32 [[X]], 1
@@ -225,7 +225,7 @@ define i32 @bit_ceil_wrong_sub_constant(i32 %x) {
 define i32 @bit_ceil_32_shl_used_twice(i32 %x, ptr %p) {
 ; CHECK-LABEL: @bit_ceil_32_shl_used_twice(
 ; CHECK-NEXT:    [[DEC:%.*]] = add i32 [[X:%.*]], -1
-; CHECK-NEXT:    [[CTLZ:%.*]] = tail call i32 @llvm.ctlz.i32(i32 [[DEC]], i1 false), !range [[RNG0]]
+; CHECK-NEXT:    [[CTLZ:%.*]] = tail call range(i32 0, 33) i32 @llvm.ctlz.i32(i32 [[DEC]], i1 false)
 ; CHECK-NEXT:    [[SUB:%.*]] = sub nuw nsw i32 32, [[CTLZ]]
 ; CHECK-NEXT:    [[SHL:%.*]] = shl nuw i32 1, [[SUB]]
 ; CHECK-NEXT:    [[UGT:%.*]] = icmp ugt i32 [[X]], 1
@@ -247,7 +247,7 @@ define i32 @bit_ceil_32_shl_used_twice(i32 %x, ptr %p) {
 define i32 @bit_ceil_32_sub_used_twice(i32 %x, ptr %p) {
 ; CHECK-LABEL: @bit_ceil_32_sub_used_twice(
 ; CHECK-NEXT:    [[DEC:%.*]] = add i32 [[X:%.*]], -1
-; CHECK-NEXT:    [[CTLZ:%.*]] = tail call i32 @llvm.ctlz.i32(i32 [[DEC]], i1 false), !range [[RNG0]]
+; CHECK-NEXT:    [[CTLZ:%.*]] = tail call range(i32 0, 33) i32 @llvm.ctlz.i32(i32 [[DEC]], i1 false)
 ; CHECK-NEXT:    [[SUB:%.*]] = sub nuw nsw i32 32, [[CTLZ]]
 ; CHECK-NEXT:    [[SHL:%.*]] = shl nuw i32 1, [[SUB]]
 ; CHECK-NEXT:    [[UGT:%.*]] = icmp ugt i32 [[X]], 1
@@ -269,7 +269,7 @@ define i32 @bit_ceil_32_sub_used_twice(i32 %x, ptr %p) {
 define <4 x i32> @bit_ceil_v4i32(<4 x i32> %x) {
 ; CHECK-LABEL: @bit_ceil_v4i32(
 ; CHECK-NEXT:    [[DEC:%.*]] = add <4 x i32> [[X:%.*]], <i32 -1, i32 -1, i32 -1, i32 -1>
-; CHECK-NEXT:    [[CTLZ:%.*]] = tail call <4 x i32> @llvm.ctlz.v4i32(<4 x i32> [[DEC]], i1 false), !range [[RNG0]]
+; CHECK-NEXT:    [[CTLZ:%.*]] = tail call range(i32 0, 33) <4 x i32> @llvm.ctlz.v4i32(<4 x i32> [[DEC]], i1 false)
 ; CHECK-NEXT:    [[TMP1:%.*]] = sub nsw <4 x i32> zeroinitializer, [[CTLZ]]
 ; CHECK-NEXT:    [[TMP2:%.*]] = and <4 x i32> [[TMP1]], <i32 31, i32 31, i32 31, i32 31>
 ; CHECK-NEXT:    [[SEL:%.*]] = shl nuw <4 x i32> <i32 1, i32 1, i32 1, i32 1>, [[TMP2]]
diff --git a/llvm/test/Transforms/InstCombine/bit_floor.ll b/llvm/test/Transforms/InstCombine/bit_floor.ll
index 9daa8eee8969..bd8aabf4431c 100644
--- a/llvm/test/Transforms/InstCombine/bit_floor.ll
+++ b/llvm/test/Transforms/InstCombine/bit_floor.ll
@@ -5,7 +5,7 @@ define i32 @bit_floor_32(i32 %x) {
 ; CHECK-LABEL: @bit_floor_32(
 ; CHECK-NEXT:    [[EQ0:%.*]] = icmp eq i32 [[X:%.*]], 0
 ; CHECK-NEXT:    [[LSHR:%.*]] = lshr i32 [[X]], 1
-; CHECK-NEXT:    [[CTLZ:%.*]] = tail call i32 @llvm.ctlz.i32(i32 [[LSHR]], i1 false), !range [[RNG0:![0-9]+]]
+; CHECK-NEXT:    [[CTLZ:%.*]] = tail call range(i32 1, 33) i32 @llvm.ctlz.i32(i32 [[LSHR]], i1 false)
 ; CHECK-NEXT:    [[SUB:%.*]] = sub nuw nsw i32 32, [[CTLZ]]
 ; CHECK-NEXT:    [[SHL:%.*]] = shl nuw i32 1, [[SUB]]
 ; CHECK-NEXT:    [[SEL:%.*]] = select i1 [[EQ0]], i32 0, i32 [[SHL]]
@@ -24,7 +24,7 @@ define i64 @bit_floor_64(i64 %x) {
 ; CHECK-LABEL: @bit_floor_64(
 ; CHECK-NEXT:    [[EQ0:%.*]] = icmp eq i64 [[X:%.*]], 0
 ; CHECK-NEXT:    [[LSHR:%.*]] = lshr i64 [[X]], 1
-; CHECK-NEXT:    [[CTLZ:%.*]] = tail call i64 @llvm.ctlz.i64(i64 [[LSHR]], i1 false), !range [[RNG1:![0-9]+]]
+; CHECK-NEXT:    [[CTLZ:%.*]] = tail call range(i64 1, 65) i64 @llvm.ctlz.i64(i64 [[LSHR]], i1 false)
 ; CHECK-NEXT:    [[SUB:%.*]] = sub nuw nsw i64 64, [[CTLZ]]
 ; CHECK-NEXT:    [[SHL:%.*]] = shl nuw i64 1, [[SUB]]
 ; CHECK-NEXT:    [[SEL:%.*]] = select i1 [[EQ0]], i64 0, i64 [[SHL]]
@@ -44,7 +44,7 @@ define i32 @bit_floor_commuted_operands(i32 %x) {
 ; CHECK-LABEL: @bit_floor_commuted_operands(
 ; CHECK-NEXT:    [[NE0_NOT:%.*]] = icmp eq i32 [[X:%.*]], 0
 ; CHECK-NEXT:    [[LSHR:%.*]] = lshr i32 [[X]], 1
-; CHECK-NEXT:    [[CTLZ:%.*]] = tail call i32 @llvm.ctlz.i32(i32 [[LSHR]], i1 false), !range [[RNG0]]
+; CHECK-NEXT:    [[CTLZ:%.*]] = tail call range(i32 1, 33) i32 @llvm.ctlz.i32(i32 [[LSHR]], i1 false)
 ; CHECK-NEXT:    [[SUB:%.*]] = sub nuw nsw i32 32, [[CTLZ]]
 ; CHECK-NEXT:    [[SHL:%.*]] = shl nuw i32 1, [[SUB]]
 ; CHECK-NEXT:    [[SEL:%.*]] = select i1 [[NE0_NOT]], i32 0, i32 [[SHL]]
@@ -64,7 +64,7 @@ define i32 @bit_floor_lshr_used_twice(i32 %x, ptr %p) {
 ; CHECK-LABEL: @bit_floor_lshr_used_twice(
 ; CHECK-NEXT:    [[EQ0:%.*]] = icmp eq i32 [[X:%.*]], 0
 ; CHECK-NEXT:    [[LSHR:%.*]] = lshr i32 [[X]], 1
-; CHECK-NEXT:    [[CTLZ:%.*]] = tail call i32 @llvm.ctlz.i32(i32 [[LSHR]], i1 false), !range [[RNG0]]
+; CHECK-NEXT:    [[CTLZ:%.*]] = tail call range(i32 1, 33) i32 @llvm.ctlz.i32(i32 [[LSHR]], i1 false)
 ; CHECK-NEXT:    [[SUB:%.*]] = sub nuw nsw i32 32, [[CTLZ]]
 ; CHECK-NEXT:    [[SHL:%.*]] = shl nuw i32 1, [[SUB]]
 ; CHECK-NEXT:    [[SEL:%.*]] = select i1 [[EQ0]], i32 0, i32 [[SHL]]
@@ -86,7 +86,7 @@ define i32 @bit_floor_ctlz_used_twice(i32 %x, ptr %p) {
 ; CHECK-LABEL: @bit_floor_ctlz_used_twice(
 ; CHECK-NEXT:    [[EQ0:%.*]] = icmp eq i32 [[X:%.*]], 0
 ; CHECK-NEXT:    [[LSHR:%.*]] = lshr i32 [[X]], 1
-; CHECK-NEXT:    [[CTLZ:%.*]] = tail call i32 @llvm.ctlz.i32(i32 [[LSHR]], i1 false), !range [[RNG0]]
+; CHECK-NEXT:    [[CTLZ:%.*]] = tail call range(i32 1, 33) i32 @llvm.ctlz.i32(i32 [[LSHR]], i1 false)
 ; CHECK-NEXT:    [[SUB:%.*]] = sub nuw nsw i32 32, [[CTLZ]]
 ; CHECK-NEXT:    [[SHL:%.*]] = shl nuw i32 1, [[SUB]]
 ; CHECK-NEXT:    [[SEL:%.*]] = select i1 [[EQ0]], i32 0, i32 [[SHL]]
@@ -108,7 +108,7 @@ define i32 @bit_floor_sub_used_twice(i32 %x, ptr %p) {
 ; CHECK-LABEL: @bit_floor_sub_used_twice(
 ; CHECK-NEXT:    [[EQ0:%.*]] = icmp eq i32 [[X:%.*]], 0
 ; CHECK-NEXT:    [[LSHR:%.*]] = lshr i32 [[X]], 1
-; CHECK-NEXT:    [[CTLZ:%.*]] = tail call i32 @llvm.ctlz.i32(i32 [[LSHR]], i1 false), !range [[RNG0]]
+; CHECK-NEXT:    [[CTLZ:%.*]] = tail call range(i32 1, 33) i32 @llvm.ctlz.i32(i32 [[LSHR]], i1 false)
 ; CHECK-NEXT:    [[SUB:%.*]] = sub nuw nsw i32 32, [[CTLZ]]
 ; CHECK-NEXT:    [[SHL:%.*]] = shl nuw i32 1, [[SUB]]
 ; CHECK-NEXT:    [[SEL:%.*]] = select i1 [[EQ0]], i32 0, i32 [[SHL]]
@@ -130,7 +130,7 @@ define i32 @bit_floor_shl_used_twice(i32 %x, ptr %p) {
 ; CHECK-LABEL: @bit_floor_shl_used_twice(
 ; CHECK-NEXT:    [[EQ0:%.*]] = icmp eq i32 [[X:%.*]], 0
 ; CHECK-NEXT:    [[LSHR:%.*]] = lshr i32 [[X]], 1
-; CHECK-NEXT:    [[CTLZ:%.*]] = tail call i32 @llvm.ctlz.i32(i32 [[LSHR]], i1 false), !range [[RNG0]]
+; CHECK-NEXT:    [[CTLZ:%.*]] = tail call range(i32 1, 33) i32 @llvm.ctlz.i32(i32 [[LSHR]], i1 false)
 ; CHECK-NEXT:    [[SUB:%.*]] = sub nuw nsw i32 32, [[CTLZ]]
 ; CHECK-NEXT:    [[SHL:%.*]] = shl nuw i32 1, [[SUB]]
 ; CHECK-NEXT:    [[SEL:%.*]] = select i1 [[EQ0]], i32 0, i32 [[SHL]]
@@ -152,7 +152,7 @@ define <4 x i32> @bit_floor_v4i32(<4 x i32> %x) {
 ; CHECK-LABEL: @bit_floor_v4i32(
 ; CHECK-NEXT:    [[EQ0:%.*]] = icmp eq <4 x i32> [[X:%.*]], zeroinitializer
 ; CHECK-NEXT:    [[LSHR:%.*]] = lshr <4 x i32> [[X]], <i32 1, i32 1, i32 1, i32 1>
-; CHECK-NEXT:    [[CTLZ:%.*]] = tail call <4 x i32> @llvm.ctlz.v4i32(<4 x i32> [[LSHR]], i1 false), !range [[RNG0]]
+; CHECK-NEXT:    [[CTLZ:%.*]] = tail call range(i32 1, 33) <4 x i32> @llvm.ctlz.v4i32(<4 x i32> [[LSHR]], i1 false)
 ; CHECK-NEXT:    [[SUB:%.*]] = sub nuw nsw <4 x i32> <i32 32, i32 32, i32 32, i32 32>, [[CTLZ]]
 ; CHECK-NEXT:    [[SHL:%.*]] = shl nuw <4 x i32> <i32 1, i32 1, i32 1, i32 1>, [[SUB]]
 ; CHECK-NEXT:    [[SEL:%.*]] = select <4 x i1> [[EQ0]], <4 x i32> zeroinitializer, <4 x i32> [[SHL]]
diff --git a/llvm/test/Transforms/InstCombine/cmp-intrinsic.ll b/llvm/test/Transforms/InstCombine/cmp-intrinsic.ll
index 66cbb2636cbc..9a9f359fa80b 100644
--- a/llvm/test/Transforms/InstCombine/cmp-intrinsic.ll
+++ b/llvm/test/Transforms/InstCombine/cmp-intrinsic.ll
@@ -125,7 +125,7 @@ define <2 x i1> @ctlz_ne_other_v2i32(<2 x i32> %a) {
 
 define i1 @ctlz_eq_other_i32_multiuse(i32 %x, ptr %p) {
 ; CHECK-LABEL: @ctlz_eq_other_i32_multiuse(
-; CHECK-NEXT:    [[LZ:%.*]] = tail call i32 @llvm.ctlz.i32(i32 [[X:%.*]], i1 false), !range [[RNG0:![0-9]+]]
+; CHECK-NEXT:    [[LZ:%.*]] = tail call range(i32 0, 33) i32 @llvm.ctlz.i32(i32 [[X:%.*]], i1 false)
 ; CHECK-NEXT:    store i32 [[LZ]], ptr [[P:%.*]], align 4
 ; CHECK-NEXT:    [[CMP:%.*]] = icmp eq i32 [[LZ]], 24
 ; CHECK-NEXT:    ret i1 [[CMP]]
@@ -178,7 +178,7 @@ define i1 @ctlz_ugt_other_i32(i32 %x) {
 
 define i1 @ctlz_ugt_other_multiuse_i32(i32 %x, ptr %p) {
 ; CHECK-LABEL: @ctlz_ugt_other_multiuse_i32(
-; CHECK-NEXT:    [[LZ:%.*]] = tail call i32 @llvm.ctlz.i32(i32 [[X:%.*]], i1 false), !range [[RNG0]]
+; CHECK-NEXT:    [[LZ:%.*]] = tail call range(i32 0, 33) i32 @llvm.ctlz.i32(i32 [[X:%.*]], i1 false)
 ; CHECK-NEXT:    store i32 [[LZ]], ptr [[P:%.*]], align 4
 ; CHECK-NEXT:    [[CMP:%.*]] = icmp ult i32 [[X]], 32768
 ; CHECK-NEXT:    ret i1 [[CMP]]
@@ -221,7 +221,7 @@ define <2 x i1> @ctlz_ult_other_v2i32(<2 x i32> %x) {
 
 define <2 x i1> @ctlz_ult_other_multiuse_v2i32(<2 x i32> %x, ptr %p) {
 ; CHECK-LABEL: @ctlz_ult_other_multiuse_v2i32(
-; CHECK-NEXT:    [[LZ:%.*]] = tail call <2 x i32> @llvm.ctlz.v2i32(<2 x i32> [[X:%.*]], i1 false), !range [[RNG0]]
+; CHECK-NEXT:    [[LZ:%.*]] = tail call range(i32 0, 33) <2 x i32> @llvm.ctlz.v2i32(<2 x i32> [[X:%.*]], i1 false)
 ; CHECK-NEXT:    store <2 x i32> [[LZ]], ptr [[P:%.*]], align 8
 ; CHECK-NEXT:    [[CMP:%.*]] = icmp ugt <2 x i32> [[X]], <i32 65535, i32 65535>
 ; CHECK-NEXT:    ret <2 x i1> [[CMP]]
@@ -338,7 +338,7 @@ define <2 x i1> @cttz_ne_other_v2i32(<2 x i32> %a) {
 
 define i1 @cttz_eq_other_i33_multiuse(i33 %x, ptr %p) {
 ; CHECK-LABEL: @cttz_eq_other_i33_multiuse(
-; CHECK-NEXT:    [[TZ:%.*]] = tail call i33 @llvm.cttz.i33(i33 [[X:%.*]], i1 false), !range [[RNG1:![0-9]+]]
+; CHECK-NEXT:    [[TZ:%.*]] = tail call range(i33 0, 34) i33 @llvm.cttz.i33(i33 [[X:%.*]], i1 false)
 ; CHECK-NEXT:    store i33 [[TZ]], ptr [[P:%.*]], align 4
 ; CHECK-NEXT:    [[CMP:%.*]] = icmp eq i33 [[TZ]], 4
 ; CHECK-NEXT:    ret i1 [[CMP]]
@@ -384,7 +384,7 @@ define i1 @cttz_ugt_other_i33(i33 %x) {
 
 define i1 @cttz_ugt_other_multiuse_i33(i33 %x, ptr %p) {
 ; CHECK-LABEL: @cttz_ugt_other_multiuse_i33(
-; CHECK-NEXT:    [[TZ:%.*]] = tail call i33 @llvm.cttz.i33(i33 [[X:%.*]], i1 false), !range [[RNG1]]
+; CHECK-NEXT:    [[TZ:%.*]] = tail call range(i33 0, 34) i33 @llvm.cttz.i33(i33 [[X:%.*]], i1 false)
 ; CHECK-NEXT:    store i33 [[TZ]], ptr [[P:%.*]], align 4
 ; CHECK-NEXT:    [[CMP:%.*]] = icmp ugt i33 [[TZ]], 16
 ; CHECK-NEXT:    ret i1 [[CMP]]
@@ -428,7 +428,7 @@ define <2 x i1> @cttz_ult_other_v2i32(<2 x i32> %x) {
 
 define <2 x i1> @cttz_ult_other_multiuse_v2i32(<2 x i32> %x, ptr %p) {
 ; CHECK-LABEL: @cttz_ult_other_multiuse_v2i32(
-; CHECK-NEXT:    [[TZ:%.*]] = tail call <2 x i32> @llvm.cttz.v2i32(<2 x i32> [[X:%.*]], i1 false), !range [[RNG0]]
+; CHECK-NEXT:    [[TZ:%.*]] = tail call range(i32 0, 33) <2 x i32> @llvm.cttz.v2i32(<2 x i32> [[X:%.*]], i1 false)
 ; CHECK-NEXT:    store <2 x i32> [[TZ]], ptr [[P:%.*]], align 8
 ; CHECK-NEXT:    [[CMP:%.*]] = icmp ult <2 x i32> [[TZ]], <i32 16, i32 16>
 ; CHECK-NEXT:    ret <2 x i1> [[CMP]]
@@ -502,7 +502,7 @@ define <2 x i1> @ctpop_ne_bitwidth_v2i32(<2 x i32> %x) {
 
 define i1 @ctpop_ugt_bitwidth_minus_one_i8(i8 %x, ptr %p) {
 ; CHECK-LABEL: @ctpop_ugt_bitwidth_minus_one_i8(
-; CHECK-NEXT:    [[POP:%.*]] = tail call i8 @llvm.ctpop.i8(i8 [[X:%.*]]), !range [[RNG2:![0-9]+]]
+; CHECK-NEXT:    [[POP:%.*]] = tail call range(i8 0, 9) i8 @llvm.ctpop.i8(i8 [[X:%.*]])
 ; CHECK-NEXT:    store i8 [[POP]], ptr [[P:%.*]], align 1
 ; CHECK-NEXT:    [[CMP:%.*]] = icmp eq i8 [[X]], -1
 ; CHECK-NEXT:    ret i1 [[CMP]]
@@ -563,7 +563,7 @@ define i1 @trunc_cttz_ult_other_i33_i6(i33 %x) {
 
 define i1 @trunc_cttz_ult_other_i33_i5(i33 %x) {
 ; CHECK-LABEL: @trunc_cttz_ult_other_i33_i5(
-; CHECK-NEXT:    [[TZ:%.*]] = tail call i33 @llvm.cttz.i33(i33 [[X:%.*]], i1 true), !range [[RNG1]]
+; CHECK-NEXT:    [[TZ:%.*]] = tail call range(i33 0, 34) i33 @llvm.cttz.i33(i33 [[X:%.*]], i1 true)
 ; CHECK-NEXT:    [[TRUNC:%.*]] = trunc i33 [[TZ]] to i5
 ; CHECK-NEXT:    [[CMP:%.*]] = icmp ult i5 [[TRUNC]], 7
 ; CHECK-NEXT:    ret i1 [[CMP]]
@@ -590,7 +590,7 @@ define i1 @trunc_cttz_true_ult_other_i32_i5(i32 %x) {
 
 define i1 @trunc_cttz_false_ult_other_i32_i5(i32 %x) {
 ; CHECK-LABEL: @trunc_cttz_false_ult_other_i32_i5(
-; CHECK-NEXT:    [[TZ:%.*]] = tail call i32 @llvm.cttz.i32(i32 [[X:%.*]], i1 false), !range [[RNG0]]
+; CHECK-NEXT:    [[TZ:%.*]] = tail call range(i32 0, 33) i32 @llvm.cttz.i32(i32 [[X:%.*]], i1 false)
 ; CHECK-NEXT:    [[TRUNC:%.*]] = trunc i32 [[TZ]] to i5
 ; CHECK-NEXT:    [[CMP:%.*]] = icmp ult i5 [[TRUNC]], 7
 ; CHECK-NEXT:    ret i1 [[CMP]]
@@ -617,7 +617,7 @@ define i1 @trunc_cttz_false_ult_other_i32_i6(i32 %x) {
 
 define i1 @trunc_cttz_false_ult_other_i32_i6_extra_use(i32 %x) {
 ; CHECK-LABEL: @trunc_cttz_false_ult_other_i32_i6_extra_use(
-; CHECK-NEXT:    [[TZ:%.*]] = tail call i32 @llvm.cttz.i32(i32 [[X:%.*]], i1 false), !range [[RNG0]]
+; CHECK-NEXT:    [[TZ:%.*]] = tail call range(i32 0, 33) i32 @llvm.cttz.i32(i32 [[X:%.*]], i1 false)
 ; CHECK-NEXT:    [[TRUNC:%.*]] = trunc nuw i32 [[TZ]] to i6
 ; CHECK-NEXT:    call void @use6(i6 [[TRUNC]])
 ; CHECK-NEXT:    [[CMP:%.*]] = icmp ult i6 [[TRUNC]], 7
@@ -667,7 +667,7 @@ define i1 @trunc_ctlz_ugt_other_i33_i6(i33 %x) {
 
 define i1 @trunc_ctlz_ugt_other_i33_i5(i33 %x) {
 ; CHECK-LABEL: @trunc_ctlz_ugt_other_i33_i5(
-; CHECK-NEXT:    [[LZ:%.*]] = tail call i33 @llvm.ctlz.i33(i33 [[X:%.*]], i1 true), !range [[RNG1]]
+; CHECK-NEXT:    [[LZ:%.*]] = tail call range(i33 0, 34) i33 @llvm.ctlz.i33(i33 [[X:%.*]], i1 true)
 ; CHECK-NEXT:    [[TRUNC:%.*]] = trunc i33 [[LZ]] to i5
 ; CHECK-NEXT:    [[CMP:%.*]] = icmp ugt i5 [[TRUNC]], 4
 ; CHECK-NEXT:    ret i1 [[CMP]]
@@ -693,7 +693,7 @@ define i1 @trunc_ctlz_true_ugt_other_i32_i5(i32 %x) {
 
 define i1 @trunc_ctlz_false_ugt_other_i32_i5(i32 %x) {
 ; CHECK-LABEL: @trunc_ctlz_false_ugt_other_i32_i5(
-; CHECK-NEXT:    [[LZ:%.*]] = tail call i32 @llvm.ctlz.i32(i32 [[X:%.*]], i1 false), !range [[RNG0]]
+; CHECK-NEXT:    [[LZ:%.*]] = tail call range(i32 0, 33) i32 @llvm.ctlz.i32(i32 [[X:%.*]], i1 false)
 ; CHECK-NEXT:    [[TRUNC:%.*]] = trunc i32 [[LZ]] to i5
 ; CHECK-NEXT:    [[CMP:%.*]] = icmp ugt i5 [[TRUNC]], 4
 ; CHECK-NEXT:    ret i1 [[CMP]]
@@ -719,7 +719,7 @@ define i1 @trunc_ctlz_false_ugt_other_i32_i6(i32 %x) {
 
 define i1 @trunc_ctlz_false_ugt_other_i32_i6_extra_use(i32 %x) {
 ; CHECK-LABEL: @trunc_ctlz_false_ugt_other_i32_i6_extra_use(
-; CHECK-NEXT:    [[LZ:%.*]] = tail call i32 @llvm.ctlz.i32(i32 [[X:%.*]], i1 false), !range [[RNG0]]
+; CHECK-NEXT:    [[LZ:%.*]] = tail call range(i32 0, 33) i32 @llvm.ctlz.i32(i32 [[X:%.*]], i1 false)
 ; CHECK-NEXT:    [[TRUNC:%.*]] = trunc nuw i32 [[LZ]] to i6
 ; CHECK-NEXT:    call void @use6(i6 [[TRUNC]])
 ; CHECK-NEXT:    [[CMP:%.*]] = icmp ugt i6 [[TRUNC]], 4
@@ -758,7 +758,7 @@ define i1 @trunc_ctpop_eq_bitwidth_i8(i8 %x) {
 
 define i1 @trunc_negative_destbits_not_enough(i33 %x) {
 ; CHECK-LABEL: @trunc_negative_destbits_not_enough(
-; CHECK-NEXT:    [[TZ:%.*]] = tail call i33 @llvm.cttz.i33(i33 [[X:%.*]], i1 false), !range [[RNG1]]
+; CHECK-NEXT:    [[TZ:%.*]] = tail call range(i33 0, 34) i33 @llvm.cttz.i33(i33 [[X:%.*]], i1 false)
 ; CHECK-NEXT:    [[TRUNC:%.*]] = trunc i33 [[TZ]] to i4
 ; CHECK-NEXT:    [[CMP:%.*]] = icmp ult i4 [[TRUNC]], 7
 ; CHECK-NEXT:    ret i1 [[CMP]]
diff --git a/llvm/test/Transforms/InstCombine/ctlz-cttz-bitreverse.ll b/llvm/test/Transforms/InstCombine/ctlz-cttz-bitreverse.ll
index a5189f476505..ec822f4b8fb3 100644
--- a/llvm/test/Transforms/InstCombine/ctlz-cttz-bitreverse.ll
+++ b/llvm/test/Transforms/InstCombine/ctlz-cttz-bitreverse.ll
@@ -3,7 +3,7 @@
 
 define i32 @ctlz_true_bitreverse(i32 %x) {
 ; CHECK-LABEL: @ctlz_true_bitreverse(
-; CHECK-NEXT:    [[B:%.*]] = call i32 @llvm.cttz.i32(i32 [[X:%.*]], i1 true), !range [[RNG0:![0-9]+]]
+; CHECK-NEXT:    [[B:%.*]] = call range(i32 0, 33) i32 @llvm.cttz.i32(i32 [[X:%.*]], i1 true)
 ; CHECK-NEXT:    ret i32 [[B]]
 ;
   %a = tail call i32 @llvm.bitreverse.i32(i32 %x)
@@ -13,7 +13,7 @@ define i32 @ctlz_true_bitreverse(i32 %x) {
 
 define <2 x i64> @ctlz_true_bitreverse_vec(<2 x i64> %x) {
 ; CHECK-LABEL: @ctlz_true_bitreverse_vec(
-; CHECK-NEXT:    [[B:%.*]] = call <2 x i64> @llvm.cttz.v2i64(<2 x i64> [[X:%.*]], i1 true), !range [[RNG1:![0-9]+]]
+; CHECK-NEXT:    [[B:%.*]] = call range(i64 0, 65) <2 x i64> @llvm.cttz.v2i64(<2 x i64> [[X:%.*]], i1 true)
 ; CHECK-NEXT:    ret <2 x i64> [[B]]
 ;
   %a = tail call <2 x i64> @llvm.bitreverse.v2i64(<2 x i64> %x)
@@ -23,7 +23,7 @@ define <2 x i64> @ctlz_true_bitreverse_vec(<2 x i64> %x) {
 
 define i32 @ctlz_false_bitreverse(i32 %x) {
 ; CHECK-LABEL: @ctlz_false_bitreverse(
-; CHECK-NEXT:    [[B:%.*]] = call i32 @llvm.cttz.i32(i32 [[X:%.*]], i1 false), !range [[RNG0]]
+; CHECK-NEXT:    [[B:%.*]] = call range(i32 0, 33) i32 @llvm.cttz.i32(i32 [[X:%.*]], i1 false)
 ; CHECK-NEXT:    ret i32 [[B]]
 ;
   %a = tail call i32 @llvm.bitreverse.i32(i32 %x)
@@ -33,7 +33,7 @@ define i32 @ctlz_false_bitreverse(i32 %x) {
 
 define i32 @cttz_true_bitreverse(i32 %x) {
 ; CHECK-LABEL: @cttz_true_bitreverse(
-; CHECK-NEXT:    [[B:%.*]] = call i32 @llvm.ctlz.i32(i32 [[X:%.*]], i1 true), !range [[RNG0]]
+; CHECK-NEXT:    [[B:%.*]] = call range(i32 0, 33) i32 @llvm.ctlz.i32(i32 [[X:%.*]], i1 true)
 ; CHECK-NEXT:    ret i32 [[B]]
 ;
   %a = tail call i32 @llvm.bitreverse.i32(i32 %x)
@@ -43,7 +43,7 @@ define i32 @cttz_true_bitreverse(i32 %x) {
 
 define <2 x i64> @cttz_true_bitreverse_vec(<2 x i64> %x) {
 ; CHECK-LABEL: @cttz_true_bitreverse_vec(
-; CHECK-NEXT:    [[B:%.*]] = call <2 x i64> @llvm.ctlz.v2i64(<2 x i64> [[X:%.*]], i1 true), !range [[RNG1]]
+; CHECK-NEXT:    [[B:%.*]] = call range(i64 0, 65) <2 x i64> @llvm.ctlz.v2i64(<2 x i64> [[X:%.*]], i1 true)
 ; CHECK-NEXT:    ret <2 x i64> [[B]]
 ;
   %a = tail call <2 x i64> @llvm.bitreverse.v2i64(<2 x i64> %x)
@@ -53,7 +53,7 @@ define <2 x i64> @cttz_true_bitreverse_vec(<2 x i64> %x) {
 
 define i32 @cttz_false_bitreverse(i32 %x) {
 ; CHECK-LABEL: @cttz_false_bitreverse(
-; CHECK-NEXT:    [[B:%.*]] = call i32 @llvm.ctlz.i32(i32 [[X:%.*]], i1 false), !range [[RNG0]]
+; CHECK-NEXT:    [[B:%.*]] = call range(i32 0, 33) i32 @llvm.ctlz.i32(i32 [[X:%.*]], i1 false)
 ; CHECK-NEXT:    ret i32 [[B]]
 ;
   %a = tail call i32 @llvm.bitreverse.i32(i32 %x)
diff --git a/llvm/test/Transforms/InstCombine/ctlz-cttz-shifts.ll b/llvm/test/Transforms/InstCombine/ctlz-cttz-shifts.ll
index 86fef51872b1..15aa87f72c49 100644
--- a/llvm/test/Transforms/InstCombine/ctlz-cttz-shifts.ll
+++ b/llvm/test/Transforms/InstCombine/ctlz-cttz-shifts.ll
@@ -122,7 +122,7 @@ define <2 x i32> @vec2_shl_nsw_ctlz_true_neg(<2 x i32>) {
 ; CHECK-LABEL: define <2 x i32> @vec2_shl_nsw_ctlz_true_neg(
 ; CHECK-SAME: <2 x i32> [[TMP0:%.*]]) {
 ; CHECK-NEXT:    [[SHL:%.*]] = shl nsw <2 x i32> <i32 8387584, i32 4276440>, [[TMP0]]
-; CHECK-NEXT:    [[CTLZ:%.*]] = call <2 x i32> @llvm.ctlz.v2i32(<2 x i32> [[SHL]], i1 true), !range [[RNG0:![0-9]+]]
+; CHECK-NEXT:    [[CTLZ:%.*]] = call range(i32 1, 33) <2 x i32> @llvm.ctlz.v2i32(<2 x i32> [[SHL]], i1 true)
 ; CHECK-NEXT:    ret <2 x i32> [[CTLZ]]
 ;
   %shl = shl nsw <2 x i32> <i32 8387584, i32 4276440>, %0
@@ -134,7 +134,7 @@ define <2 x i32> @vec2_lshr_ctlz_false_neg(<2 x i32>) {
 ; CHECK-LABEL: define <2 x i32> @vec2_lshr_ctlz_false_neg(
 ; CHECK-SAME: <2 x i32> [[TMP0:%.*]]) {
 ; CHECK-NEXT:    [[DIV:%.*]] = lshr <2 x i32> <i32 8387584, i32 4276440>, [[TMP0]]
-; CHECK-NEXT:    [[CTLZ:%.*]] = call <2 x i32> @llvm.ctlz.v2i32(<2 x i32> [[DIV]], i1 false), !range [[RNG1:![0-9]+]]
+; CHECK-NEXT:    [[CTLZ:%.*]] = call range(i32 9, 33) <2 x i32> @llvm.ctlz.v2i32(<2 x i32> [[DIV]], i1 false)
 ; CHECK-NEXT:    ret <2 x i32> [[CTLZ]]
 ;
   %div = lshr <2 x i32> <i32 8387584, i32 4276440>, %0
@@ -146,7 +146,7 @@ define <2 x i32> @vec2_shl_ctlz_false_neg(<2 x i32>) {
 ; CHECK-LABEL: define <2 x i32> @vec2_shl_ctlz_false_neg(
 ; CHECK-SAME: <2 x i32> [[TMP0:%.*]]) {
 ; CHECK-NEXT:    [[SHL:%.*]] = shl <2 x i32> <i32 8387584, i32 4276440>, [[TMP0]]
-; CHECK-NEXT:    [[CTLZ:%.*]] = call <2 x i32> @llvm.ctlz.v2i32(<2 x i32> [[SHL]], i1 false), !range [[RNG2:![0-9]+]]
+; CHECK-NEXT:    [[CTLZ:%.*]] = call range(i32 0, 33) <2 x i32> @llvm.ctlz.v2i32(<2 x i32> [[SHL]], i1 false)
 ; CHECK-NEXT:    ret <2 x i32> [[CTLZ]]
 ;
   %shl = shl <2 x i32> <i32 8387584, i32 4276440>, %0
@@ -158,7 +158,7 @@ define <2 x i32> @vec2_lshr_cttz_false_neg(<2 x i32>) {
 ; CHECK-LABEL: define <2 x i32> @vec2_lshr_cttz_false_neg(
 ; CHECK-SAME: <2 x i32> [[TMP0:%.*]]) {
 ; CHECK-NEXT:    [[LSHR:%.*]] = lshr <2 x i32> <i32 8387584, i32 4276440>, [[TMP0]]
-; CHECK-NEXT:    [[CTTZ:%.*]] = call <2 x i32> @llvm.cttz.v2i32(<2 x i32> [[LSHR]], i1 false), !range [[RNG2]]
+; CHECK-NEXT:    [[CTTZ:%.*]] = call range(i32 0, 33) <2 x i32> @llvm.cttz.v2i32(<2 x i32> [[LSHR]], i1 false)
 ; CHECK-NEXT:    ret <2 x i32> [[CTTZ]]
 ;
   %lshr = lshr <2 x i32> <i32 8387584, i32 4276440>, %0
@@ -170,7 +170,7 @@ define <2 x i32> @vec2_shl_cttz_false_neg(<2 x i32>) {
 ; CHECK-LABEL: define <2 x i32> @vec2_shl_cttz_false_neg(
 ; CHECK-SAME: <2 x i32> [[TMP0:%.*]]) {
 ; CHECK-NEXT:    [[SHL:%.*]] = shl <2 x i32> <i32 8387584, i32 4276440>, [[TMP0]]
-; CHECK-NEXT:    [[CTTZ:%.*]] = call <2 x i32> @llvm.cttz.v2i32(<2 x i32> [[SHL]], i1 false), !range [[RNG3:![0-9]+]]
+; CHECK-NEXT:    [[CTTZ:%.*]] = call range(i32 3, 33) <2 x i32> @llvm.cttz.v2i32(<2 x i32> [[SHL]], i1 false)
 ; CHECK-NEXT:    ret <2 x i32> [[CTTZ]]
 ;
   %shl = shl <2 x i32> <i32 8387584, i32 4276440>, %0
@@ -182,7 +182,7 @@ define i32 @lshr_ctlz_faslse_neg(i32) {
 ; CHECK-LABEL: define i32 @lshr_ctlz_faslse_neg(
 ; CHECK-SAME: i32 [[TMP0:%.*]]) {
 ; CHECK-NEXT:    [[LSHR:%.*]] = lshr i32 8387584, [[TMP0]]
-; CHECK-NEXT:    [[CTLZ:%.*]] = call i32 @llvm.ctlz.i32(i32 [[LSHR]], i1 false), !range [[RNG1]]
+; CHECK-NEXT:    [[CTLZ:%.*]] = call range(i32 9, 33) i32 @llvm.ctlz.i32(i32 [[LSHR]], i1 false)
 ; CHECK-NEXT:    ret i32 [[CTLZ]]
 ;
   %lshr = lshr i32 8387584, %0
@@ -194,7 +194,7 @@ define i32 @shl_ctlz_false_neg(i32) {
 ; CHECK-LABEL: define i32 @shl_ctlz_false_neg(
 ; CHECK-SAME: i32 [[TMP0:%.*]]) {
 ; CHECK-NEXT:    [[SHL:%.*]] = shl i32 8387584, [[TMP0]]
-; CHECK-NEXT:    [[CTLZ:%.*]] = call i32 @llvm.ctlz.i32(i32 [[SHL]], i1 false), !range [[RNG2]]
+; CHECK-NEXT:    [[CTLZ:%.*]] = call range(i32 0, 33) i32 @llvm.ctlz.i32(i32 [[SHL]], i1 false)
 ; CHECK-NEXT:    ret i32 [[CTLZ]]
 ;
   %shl = shl i32 8387584, %0
@@ -206,7 +206,7 @@ define i32 @lshr_cttz_false_neg(i32) {
 ; CHECK-LABEL: define i32 @lshr_cttz_false_neg(
 ; CHECK-SAME: i32 [[TMP0:%.*]]) {
 ; CHECK-NEXT:    [[LSHR:%.*]] = lshr i32 8387584, [[TMP0]]
-; CHECK-NEXT:    [[CTTZ:%.*]] = call i32 @llvm.cttz.i32(i32 [[LSHR]], i1 false), !range [[RNG2]]
+; CHECK-NEXT:    [[CTTZ:%.*]] = call range(i32 0, 33) i32 @llvm.cttz.i32(i32 [[LSHR]], i1 false)
 ; CHECK-NEXT:    ret i32 [[CTTZ]]
 ;
   %lshr = lshr i32 8387584, %0
@@ -218,17 +218,10 @@ define i32 @shl_cttz_false_neg(i32) {
 ; CHECK-LABEL: define i32 @shl_cttz_false_neg(
 ; CHECK-SAME: i32 [[TMP0:%.*]]) {
 ; CHECK-NEXT:    [[SHL:%.*]] = shl i32 8387584, [[TMP0]]
-; CHECK-NEXT:    [[CTTZ:%.*]] = call i32 @llvm.cttz.i32(i32 [[SHL]], i1 false), !range [[RNG4:![0-9]+]]
+; CHECK-NEXT:    [[CTTZ:%.*]] = call range(i32 10, 33) i32 @llvm.cttz.i32(i32 [[SHL]], i1 false)
 ; CHECK-NEXT:    ret i32 [[CTTZ]]
 ;
   %shl = shl i32 8387584, %0
   %cttz = call i32 @llvm.cttz.i32(i32 %shl, i1 false)
   ret i32 %cttz
 }
-;.
-; CHECK: [[RNG0]] = !{i32 1, i32 33}
-; CHECK: [[RNG1]] = !{i32 9, i32 33}
-; CHECK: [[RNG2]] = !{i32 0, i32 33}
-; CHECK: [[RNG3]] = !{i32 3, i32 33}
-; CHECK: [[RNG4]] = !{i32 10, i32 33}
-;.
diff --git a/llvm/test/Transforms/InstCombine/ctpop-bswap-bitreverse.ll b/llvm/test/Transforms/InstCombine/ctpop-bswap-bitreverse.ll
index 2f523f90edda..7e4050873dc9 100644
--- a/llvm/test/Transforms/InstCombine/ctpop-bswap-bitreverse.ll
+++ b/llvm/test/Transforms/InstCombine/ctpop-bswap-bitreverse.ll
@@ -3,7 +3,7 @@
 
 define i32 @ctpop_bitreverse(i32 %x) {
 ; CHECK-LABEL: @ctpop_bitreverse(
-; CHECK-NEXT:    [[B:%.*]] = tail call i32 @llvm.ctpop.i32(i32 [[X:%.*]]), !range [[RNG0:![0-9]+]]
+; CHECK-NEXT:    [[B:%.*]] = tail call range(i32 0, 33) i32 @llvm.ctpop.i32(i32 [[X:%.*]])
 ; CHECK-NEXT:    ret i32 [[B]]
 ;
   %a = tail call i32 @llvm.bitreverse.i32(i32 %x)
@@ -13,7 +13,7 @@ define i32 @ctpop_bitreverse(i32 %x) {
 
 define <2 x i64> @ctpop_bitreverse_vec(<2 x i64> %x) {
 ; CHECK-LABEL: @ctpop_bitreverse_vec(
-; CHECK-NEXT:    [[B:%.*]] = tail call <2 x i64> @llvm.ctpop.v2i64(<2 x i64> [[X:%.*]]), !range [[RNG1:![0-9]+]]
+; CHECK-NEXT:    [[B:%.*]] = tail call range(i64 0, 65) <2 x i64> @llvm.ctpop.v2i64(<2 x i64> [[X:%.*]])
 ; CHECK-NEXT:    ret <2 x i64> [[B]]
 ;
   %a = tail call <2 x i64> @llvm.bitreverse.v2i64(<2 x i64> %x)
@@ -23,7 +23,7 @@ define <2 x i64> @ctpop_bitreverse_vec(<2 x i64> %x) {
 
 define i32 @ctpop_bswap(i32 %x) {
 ; CHECK-LABEL: @ctpop_bswap(
-; CHECK-NEXT:    [[B:%.*]] = tail call i32 @llvm.ctpop.i32(i32 [[X:%.*]]), !range [[RNG0]]
+; CHECK-NEXT:    [[B:%.*]] = tail call range(i32 0, 33) i32 @llvm.ctpop.i32(i32 [[X:%.*]])
 ; CHECK-NEXT:    ret i32 [[B]]
 ;
   %a = tail call i32 @llvm.bswap.i32(i32 %x)
@@ -33,7 +33,7 @@ define i32 @ctpop_bswap(i32 %x) {
 
 define <2 x i64> @ctpop_bswap_vec(<2 x i64> %x) {
 ; CHECK-LABEL: @ctpop_bswap_vec(
-; CHECK-NEXT:    [[B:%.*]] = tail call <2 x i64> @llvm.ctpop.v2i64(<2 x i64> [[X:%.*]]), !range [[RNG1]]
+; CHECK-NEXT:    [[B:%.*]] = tail call range(i64 0, 65) <2 x i64> @llvm.ctpop.v2i64(<2 x i64> [[X:%.*]])
 ; CHECK-NEXT:    ret <2 x i64> [[B]]
 ;
   %a = tail call <2 x i64> @llvm.bswap.v2i64(<2 x i64> %x)
diff --git a/llvm/test/Transforms/InstCombine/ctpop-cttz.ll b/llvm/test/Transforms/InstCombine/ctpop-cttz.ll
index 70868554bdc1..a505654fa96e 100644
--- a/llvm/test/Transforms/InstCombine/ctpop-cttz.ll
+++ b/llvm/test/Transforms/InstCombine/ctpop-cttz.ll
@@ -8,7 +8,7 @@ declare <2 x i32> @llvm.ctpop.v2i32(<2 x i32>)
 ; __builtin_popcount(i | -i) -> 32 - __builtin_cttz(i, false)
 define i32 @ctpop1(i32 %0) {
 ; CHECK-LABEL: @ctpop1(
-; CHECK-NEXT:    [[TMP2:%.*]] = call i32 @llvm.cttz.i32(i32 [[TMP0:%.*]], i1 false), !range [[RNG0:![0-9]+]]
+; CHECK-NEXT:    [[TMP2:%.*]] = call range(i32 0, 33) i32 @llvm.cttz.i32(i32 [[TMP0:%.*]], i1 false)
 ; CHECK-NEXT:    ret i32 [[TMP2]]
 ;
   %2 = sub i32 0, %0
@@ -20,7 +20,7 @@ define i32 @ctpop1(i32 %0) {
 
 define <2 x i32> @ctpop1v(<2 x i32> %0) {
 ; CHECK-LABEL: @ctpop1v(
-; CHECK-NEXT:    [[TMP2:%.*]] = call <2 x i32> @llvm.cttz.v2i32(<2 x i32> [[TMP0:%.*]], i1 false), !range [[RNG0]]
+; CHECK-NEXT:    [[TMP2:%.*]] = call range(i32 0, 33) <2 x i32> @llvm.cttz.v2i32(<2 x i32> [[TMP0:%.*]], i1 false)
 ; CHECK-NEXT:    [[TMP3:%.*]] = sub nuw nsw <2 x i32> <i32 32, i32 32>, [[TMP2]]
 ; CHECK-NEXT:    ret <2 x i32> [[TMP3]]
 ;
@@ -35,7 +35,7 @@ define i32 @ctpop1_multiuse(i32 %0) {
 ; CHECK-NEXT:    [[TMP2:%.*]] = sub i32 0, [[TMP0:%.*]]
 ; CHECK-NEXT:    [[TMP3:%.*]] = or i32 [[TMP2]], [[TMP0]]
 ; CHECK-NEXT:    [[TMP4:%.*]] = xor i32 [[TMP3]], -1
-; CHECK-NEXT:    [[TMP5:%.*]] = call i32 @llvm.ctpop.i32(i32 [[TMP4]]), !range [[RNG0]]
+; CHECK-NEXT:    [[TMP5:%.*]] = call range(i32 0, 33) i32 @llvm.ctpop.i32(i32 [[TMP4]])
 ; CHECK-NEXT:    [[TMP6:%.*]] = add i32 [[TMP5]], [[TMP3]]
 ; CHECK-NEXT:    ret i32 [[TMP6]]
 ;
@@ -51,7 +51,7 @@ define i32 @ctpop1_multiuse(i32 %0) {
 ; __builtin_popcount(~i & (i-1)) -> __builtin_cttz(i, false)
 define i32 @ctpop2(i32 %0) {
 ; CHECK-LABEL: @ctpop2(
-; CHECK-NEXT:    [[TMP2:%.*]] = call i32 @llvm.cttz.i32(i32 [[TMP0:%.*]], i1 false), !range [[RNG0]]
+; CHECK-NEXT:    [[TMP2:%.*]] = call range(i32 0, 33) i32 @llvm.cttz.i32(i32 [[TMP0:%.*]], i1 false)
 ; CHECK-NEXT:    ret i32 [[TMP2]]
 ;
   %2 = xor i32 %0, -1
@@ -63,7 +63,7 @@ define i32 @ctpop2(i32 %0) {
 
 define <2 x i32> @ctpop2v(<2 x i32> %0) {
 ; CHECK-LABEL: @ctpop2v(
-; CHECK-NEXT:    [[TMP2:%.*]] = call <2 x i32> @llvm.cttz.v2i32(<2 x i32> [[TMP0:%.*]], i1 false), !range [[RNG0]]
+; CHECK-NEXT:    [[TMP2:%.*]] = call range(i32 0, 33) <2 x i32> @llvm.cttz.v2i32(<2 x i32> [[TMP0:%.*]], i1 false)
 ; CHECK-NEXT:    ret <2 x i32> [[TMP2]]
 ;
   %2 = xor <2 x i32> %0, <i32 -1, i32 -1>
@@ -78,7 +78,7 @@ define i32 @ctpop2_multiuse(i32 %0) {
 ; CHECK-NEXT:    [[TMP2:%.*]] = xor i32 [[TMP0:%.*]], -1
 ; CHECK-NEXT:    [[TMP3:%.*]] = add i32 [[TMP0]], -1
 ; CHECK-NEXT:    [[TMP4:%.*]] = and i32 [[TMP3]], [[TMP2]]
-; CHECK-NEXT:    [[TMP5:%.*]] = call i32 @llvm.cttz.i32(i32 [[TMP0]], i1 false), !range [[RNG0]]
+; CHECK-NEXT:    [[TMP5:%.*]] = call range(i32 0, 33) i32 @llvm.cttz.i32(i32 [[TMP0]], i1 false)
 ; CHECK-NEXT:    [[TMP6:%.*]] = add i32 [[TMP5]], [[TMP4]]
 ; CHECK-NEXT:    ret i32 [[TMP6]]
 ;
@@ -94,7 +94,7 @@ define i32 @ctpop2_multiuse(i32 %0) {
 ; __builtin_popcount((i & -i) - 1) -> __builtin_cttz(i, false)
 define i32 @ctpop3(i32 %0) {
 ; CHECK-LABEL: @ctpop3(
-; CHECK-NEXT:    [[TMP2:%.*]] = call i32 @llvm.cttz.i32(i32 [[TMP0:%.*]], i1 false), !range [[RNG0]]
+; CHECK-NEXT:    [[TMP2:%.*]] = call range(i32 0, 33) i32 @llvm.cttz.i32(i32 [[TMP0:%.*]], i1 false)
 ; CHECK-NEXT:    ret i32 [[TMP2]]
 ;
   %2 = sub i32 0, %0
@@ -106,7 +106,7 @@ define i32 @ctpop3(i32 %0) {
 
 define <2 x i32> @ctpop3v(<2 x i32> %0) {
 ; CHECK-LABEL: @ctpop3v(
-; CHECK-NEXT:    [[TMP2:%.*]] = call <2 x i32> @llvm.cttz.v2i32(<2 x i32> [[TMP0:%.*]], i1 false), !range [[RNG0]]
+; CHECK-NEXT:    [[TMP2:%.*]] = call range(i32 0, 33) <2 x i32> @llvm.cttz.v2i32(<2 x i32> [[TMP0:%.*]], i1 false)
 ; CHECK-NEXT:    ret <2 x i32> [[TMP2]]
 ;
   %2 = sub <2 x i32> zeroinitializer, %0
@@ -118,7 +118,7 @@ define <2 x i32> @ctpop3v(<2 x i32> %0) {
 
 define <2 x i32> @ctpop3v_poison(<2 x i32> %0) {
 ; CHECK-LABEL: @ctpop3v_poison(
-; CHECK-NEXT:    [[TMP2:%.*]] = call <2 x i32> @llvm.cttz.v2i32(<2 x i32> [[TMP0:%.*]], i1 false), !range [[RNG0]]
+; CHECK-NEXT:    [[TMP2:%.*]] = call range(i32 0, 33) <2 x i32> @llvm.cttz.v2i32(<2 x i32> [[TMP0:%.*]], i1 false)
 ; CHECK-NEXT:    ret <2 x i32> [[TMP2]]
 ;
   %2 = sub <2 x i32> zeroinitializer, %0
diff --git a/llvm/test/Transforms/InstCombine/ctpop-pow2.ll b/llvm/test/Transforms/InstCombine/ctpop-pow2.ll
index f6757c2ff33c..7facdaf7590d 100644
--- a/llvm/test/Transforms/InstCombine/ctpop-pow2.ll
+++ b/llvm/test/Transforms/InstCombine/ctpop-pow2.ll
@@ -60,7 +60,7 @@ define i8 @ctpop_imin_plus1_lshr_nz(i8 %x) {
 ; CHECK-NEXT:    [[CMP:%.*]] = icmp ne i8 [[X:%.*]], 0
 ; CHECK-NEXT:    call void @llvm.assume(i1 [[CMP]])
 ; CHECK-NEXT:    [[V:%.*]] = lshr i8 -127, [[X]]
-; CHECK-NEXT:    [[CNT:%.*]] = call i8 @llvm.ctpop.i8(i8 [[V]]), !range [[RNG0:![0-9]+]]
+; CHECK-NEXT:    [[CNT:%.*]] = call range(i8 0, 9) i8 @llvm.ctpop.i8(i8 [[V]])
 ; CHECK-NEXT:    ret i8 [[CNT]]
 ;
   %cmp = icmp ne i8 %x, 0
@@ -104,7 +104,7 @@ define <2 x i32> @ctpop_lshr_intmin_intmin_plus1_vec_nz(<2 x i32> %x) {
 ; CHECK-LABEL: @ctpop_lshr_intmin_intmin_plus1_vec_nz(
 ; CHECK-NEXT:    [[X1:%.*]] = or <2 x i32> [[X:%.*]], <i32 1, i32 1>
 ; CHECK-NEXT:    [[SHR:%.*]] = lshr <2 x i32> <i32 -2147483648, i32 -2147483647>, [[X1]]
-; CHECK-NEXT:    [[CNT:%.*]] = call <2 x i32> @llvm.ctpop.v2i32(<2 x i32> [[SHR]]), !range [[RNG1:![0-9]+]]
+; CHECK-NEXT:    [[CNT:%.*]] = call range(i32 0, 17) <2 x i32> @llvm.ctpop.v2i32(<2 x i32> [[SHR]])
 ; CHECK-NEXT:    ret <2 x i32> [[CNT]]
 ;
   %x1 = or <2 x i32> %x, <i32 1 ,i32 1>
diff --git a/llvm/test/Transforms/InstCombine/ctpop.ll b/llvm/test/Transforms/InstCombine/ctpop.ll
index b3653e5071ba..83700e72de08 100644
--- a/llvm/test/Transforms/InstCombine/ctpop.ll
+++ b/llvm/test/Transforms/InstCombine/ctpop.ll
@@ -49,7 +49,7 @@ define i1 @test3(i32 %arg) {
 ; Negative test for when we know nothing
 define i1 @test4(i8 %arg) {
 ; CHECK-LABEL: @test4(
-; CHECK-NEXT:    [[CNT:%.*]] = call i8 @llvm.ctpop.i8(i8 [[ARG:%.*]]), !range [[RNG0:![0-9]+]]
+; CHECK-NEXT:    [[CNT:%.*]] = call range(i8 0, 9) i8 @llvm.ctpop.i8(i8 [[ARG:%.*]])
 ; CHECK-NEXT:    [[RES:%.*]] = icmp eq i8 [[CNT]], 2
 ; CHECK-NEXT:    ret i1 [[RES]]
 ;
@@ -118,7 +118,7 @@ define <2 x i32> @mask_one_bit_splat(<2 x i32> %x, ptr %p) {
 
 define i32 @_parity_of_not(i32 %x) {
 ; CHECK-LABEL: @_parity_of_not(
-; CHECK-NEXT:    [[TMP1:%.*]] = call i32 @llvm.ctpop.i32(i32 [[X:%.*]]), !range [[RNG1:![0-9]+]]
+; CHECK-NEXT:    [[TMP1:%.*]] = call range(i32 0, 33) i32 @llvm.ctpop.i32(i32 [[X:%.*]])
 ; CHECK-NEXT:    [[R:%.*]] = and i32 [[TMP1]], 1
 ; CHECK-NEXT:    ret i32 [[R]]
 ;
@@ -133,7 +133,7 @@ define i32 @_parity_of_not(i32 %x) {
 define i7 @_parity_of_not_odd_type(i7 %x) {
 ; CHECK-LABEL: @_parity_of_not_odd_type(
 ; CHECK-NEXT:    [[NEG:%.*]] = xor i7 [[X:%.*]], -1
-; CHECK-NEXT:    [[CNT:%.*]] = tail call i7 @llvm.ctpop.i7(i7 [[NEG]]), !range [[RNG2:![0-9]+]]
+; CHECK-NEXT:    [[CNT:%.*]] = tail call range(i7 0, 8) i7 @llvm.ctpop.i7(i7 [[NEG]])
 ; CHECK-NEXT:    [[R:%.*]] = and i7 [[CNT]], 1
 ; CHECK-NEXT:    ret i7 [[R]]
 ;
@@ -145,7 +145,7 @@ define i7 @_parity_of_not_odd_type(i7 %x) {
 
 define <2 x i32> @_parity_of_not_vec(<2 x i32> %x) {
 ; CHECK-LABEL: @_parity_of_not_vec(
-; CHECK-NEXT:    [[TMP1:%.*]] = call <2 x i32> @llvm.ctpop.v2i32(<2 x i32> [[X:%.*]]), !range [[RNG1]]
+; CHECK-NEXT:    [[TMP1:%.*]] = call range(i32 0, 33) <2 x i32> @llvm.ctpop.v2i32(<2 x i32> [[X:%.*]])
 ; CHECK-NEXT:    [[R:%.*]] = and <2 x i32> [[TMP1]], <i32 1, i32 1>
 ; CHECK-NEXT:    ret <2 x i32> [[R]]
 ;
@@ -157,7 +157,7 @@ define <2 x i32> @_parity_of_not_vec(<2 x i32> %x) {
 
 define <2 x i32> @_parity_of_not_poison(<2 x i32> %x) {
 ; CHECK-LABEL: @_parity_of_not_poison(
-; CHECK-NEXT:    [[TMP1:%.*]] = call <2 x i32> @llvm.ctpop.v2i32(<2 x i32> [[X:%.*]]), !range [[RNG1]]
+; CHECK-NEXT:    [[TMP1:%.*]] = call range(i32 0, 33) <2 x i32> @llvm.ctpop.v2i32(<2 x i32> [[X:%.*]])
 ; CHECK-NEXT:    [[R:%.*]] = and <2 x i32> [[TMP1]], <i32 1, i32 1>
 ; CHECK-NEXT:    ret <2 x i32> [[R]]
 ;
@@ -169,7 +169,7 @@ define <2 x i32> @_parity_of_not_poison(<2 x i32> %x) {
 
 define <2 x i32> @_parity_of_not_poison2(<2 x i32> %x) {
 ; CHECK-LABEL: @_parity_of_not_poison2(
-; CHECK-NEXT:    [[CNT:%.*]] = call <2 x i32> @llvm.ctpop.v2i32(<2 x i32> [[X:%.*]]), !range [[RNG1]]
+; CHECK-NEXT:    [[CNT:%.*]] = call range(i32 0, 33) <2 x i32> @llvm.ctpop.v2i32(<2 x i32> [[X:%.*]])
 ; CHECK-NEXT:    [[R:%.*]] = and <2 x i32> [[CNT]], <i32 1, i32 poison>
 ; CHECK-NEXT:    ret <2 x i32> [[R]]
 ;
@@ -200,7 +200,7 @@ define i32 @ctpop_add(i32 %a, i32 %b) {
 define i32 @ctpop_add_no_common_bits(i32 %a, i32 %b) {
 ; CHECK-LABEL: @ctpop_add_no_common_bits(
 ; CHECK-NEXT:    [[TMP1:%.*]] = call i32 @llvm.fshl.i32(i32 [[A:%.*]], i32 [[B:%.*]], i32 16)
-; CHECK-NEXT:    [[RES:%.*]] = call i32 @llvm.ctpop.i32(i32 [[TMP1]]), !range [[RNG1]]
+; CHECK-NEXT:    [[RES:%.*]] = call range(i32 0, 33) i32 @llvm.ctpop.i32(i32 [[TMP1]])
 ; CHECK-NEXT:    ret i32 [[RES]]
 ;
   %shl16 = shl i32 %a, 16
@@ -214,7 +214,7 @@ define i32 @ctpop_add_no_common_bits(i32 %a, i32 %b) {
 define <2 x i32> @ctpop_add_no_common_bits_vec(<2 x i32> %a, <2 x i32> %b) {
 ; CHECK-LABEL: @ctpop_add_no_common_bits_vec(
 ; CHECK-NEXT:    [[TMP1:%.*]] = call <2 x i32> @llvm.fshl.v2i32(<2 x i32> [[A:%.*]], <2 x i32> [[B:%.*]], <2 x i32> <i32 16, i32 16>)
-; CHECK-NEXT:    [[RES:%.*]] = call <2 x i32> @llvm.ctpop.v2i32(<2 x i32> [[TMP1]]), !range [[RNG1]]
+; CHECK-NEXT:    [[RES:%.*]] = call range(i32 0, 33) <2 x i32> @llvm.ctpop.v2i32(<2 x i32> [[TMP1]])
 ; CHECK-NEXT:    ret <2 x i32> [[RES]]
 ;
   %shl16 = shl <2 x i32> %a, <i32 16, i32 16>
@@ -228,9 +228,9 @@ define <2 x i32> @ctpop_add_no_common_bits_vec(<2 x i32> %a, <2 x i32> %b) {
 define <2 x i32> @ctpop_add_no_common_bits_vec_use(<2 x i32> %a, <2 x i32> %b, ptr %p) {
 ; CHECK-LABEL: @ctpop_add_no_common_bits_vec_use(
 ; CHECK-NEXT:    [[SHL16:%.*]] = shl <2 x i32> [[A:%.*]], <i32 16, i32 16>
-; CHECK-NEXT:    [[CTPOP1:%.*]] = tail call <2 x i32> @llvm.ctpop.v2i32(<2 x i32> [[SHL16]]), !range [[RNG3:![0-9]+]]
+; CHECK-NEXT:    [[CTPOP1:%.*]] = tail call range(i32 0, 17) <2 x i32> @llvm.ctpop.v2i32(<2 x i32> [[SHL16]])
 ; CHECK-NEXT:    [[LSHL16:%.*]] = lshr <2 x i32> [[B:%.*]], <i32 16, i32 16>
-; CHECK-NEXT:    [[CTPOP2:%.*]] = tail call <2 x i32> @llvm.ctpop.v2i32(<2 x i32> [[LSHL16]]), !range [[RNG3]]
+; CHECK-NEXT:    [[CTPOP2:%.*]] = tail call range(i32 0, 17) <2 x i32> @llvm.ctpop.v2i32(<2 x i32> [[LSHL16]])
 ; CHECK-NEXT:    store <2 x i32> [[CTPOP2]], ptr [[P:%.*]], align 8
 ; CHECK-NEXT:    [[RES:%.*]] = add nuw nsw <2 x i32> [[CTPOP1]], [[CTPOP2]]
 ; CHECK-NEXT:    ret <2 x i32> [[RES]]
@@ -247,10 +247,10 @@ define <2 x i32> @ctpop_add_no_common_bits_vec_use(<2 x i32> %a, <2 x i32> %b, p
 define <2 x i32> @ctpop_add_no_common_bits_vec_use2(<2 x i32> %a, <2 x i32> %b, ptr %p) {
 ; CHECK-LABEL: @ctpop_add_no_common_bits_vec_use2(
 ; CHECK-NEXT:    [[SHL16:%.*]] = shl <2 x i32> [[A:%.*]], <i32 16, i32 16>
-; CHECK-NEXT:    [[CTPOP1:%.*]] = tail call <2 x i32> @llvm.ctpop.v2i32(<2 x i32> [[SHL16]]), !range [[RNG3]]
+; CHECK-NEXT:    [[CTPOP1:%.*]] = tail call range(i32 0, 17) <2 x i32> @llvm.ctpop.v2i32(<2 x i32> [[SHL16]])
 ; CHECK-NEXT:    store <2 x i32> [[CTPOP1]], ptr [[P:%.*]], align 8
 ; CHECK-NEXT:    [[LSHL16:%.*]] = lshr <2 x i32> [[B:%.*]], <i32 16, i32 16>
-; CHECK-NEXT:    [[CTPOP2:%.*]] = tail call <2 x i32> @llvm.ctpop.v2i32(<2 x i32> [[LSHL16]]), !range [[RNG3]]
+; CHECK-NEXT:    [[CTPOP2:%.*]] = tail call range(i32 0, 17) <2 x i32> @llvm.ctpop.v2i32(<2 x i32> [[LSHL16]])
 ; CHECK-NEXT:    [[RES:%.*]] = add nuw nsw <2 x i32> [[CTPOP1]], [[CTPOP2]]
 ; CHECK-NEXT:    ret <2 x i32> [[RES]]
 ;
@@ -265,7 +265,7 @@ define <2 x i32> @ctpop_add_no_common_bits_vec_use2(<2 x i32> %a, <2 x i32> %b,
 
 define i8 @ctpop_rotate_left(i8 %a, i8 %amt)  {
 ; CHECK-LABEL: @ctpop_rotate_left(
-; CHECK-NEXT:    [[RES:%.*]] = tail call i8 @llvm.ctpop.i8(i8 [[A:%.*]]), !range [[RNG0]]
+; CHECK-NEXT:    [[RES:%.*]] = tail call range(i8 0, 9) i8 @llvm.ctpop.i8(i8 [[A:%.*]])
 ; CHECK-NEXT:    ret i8 [[RES]]
 ;
   %rotl = tail call i8 @llvm.fshl.i8(i8 %a, i8 %a, i8 %amt)
@@ -275,7 +275,7 @@ define i8 @ctpop_rotate_left(i8 %a, i8 %amt)  {
 
 define i8 @ctpop_rotate_right(i8 %a, i8 %amt)  {
 ; CHECK-LABEL: @ctpop_rotate_right(
-; CHECK-NEXT:    [[RES:%.*]] = tail call i8 @llvm.ctpop.i8(i8 [[A:%.*]]), !range [[RNG0]]
+; CHECK-NEXT:    [[RES:%.*]] = tail call range(i8 0, 9) i8 @llvm.ctpop.i8(i8 [[A:%.*]])
 ; CHECK-NEXT:    ret i8 [[RES]]
 ;
   %rotr = tail call i8 @llvm.fshr.i8(i8 %a, i8 %a, i8 %amt)
@@ -289,7 +289,7 @@ declare i8 @llvm.fshr.i8(i8, i8, i8)
 define i8 @sub_ctpop(i8 %a)  {
 ; CHECK-LABEL: @sub_ctpop(
 ; CHECK-NEXT:    [[TMP1:%.*]] = xor i8 [[A:%.*]], -1
-; CHECK-NEXT:    [[RES:%.*]] = call i8 @llvm.ctpop.i8(i8 [[TMP1]]), !range [[RNG0]]
+; CHECK-NEXT:    [[RES:%.*]] = call range(i8 0, 9) i8 @llvm.ctpop.i8(i8 [[TMP1]])
 ; CHECK-NEXT:    ret i8 [[RES]]
 ;
   %cnt = tail call i8 @llvm.ctpop.i8(i8 %a)
@@ -299,7 +299,7 @@ define i8 @sub_ctpop(i8 %a)  {
 
 define i8 @sub_ctpop_wrong_cst(i8 %a)  {
 ; CHECK-LABEL: @sub_ctpop_wrong_cst(
-; CHECK-NEXT:    [[CNT:%.*]] = tail call i8 @llvm.ctpop.i8(i8 [[A:%.*]]), !range [[RNG0]]
+; CHECK-NEXT:    [[CNT:%.*]] = tail call range(i8 0, 9) i8 @llvm.ctpop.i8(i8 [[A:%.*]])
 ; CHECK-NEXT:    [[RES:%.*]] = sub nsw i8 5, [[CNT]]
 ; CHECK-NEXT:    ret i8 [[RES]]
 ;
@@ -310,7 +310,7 @@ define i8 @sub_ctpop_wrong_cst(i8 %a)  {
 
 define i8 @sub_ctpop_unknown(i8 %a, i8 %b)  {
 ; CHECK-LABEL: @sub_ctpop_unknown(
-; CHECK-NEXT:    [[CNT:%.*]] = tail call i8 @llvm.ctpop.i8(i8 [[A:%.*]]), !range [[RNG0]]
+; CHECK-NEXT:    [[CNT:%.*]] = tail call range(i8 0, 9) i8 @llvm.ctpop.i8(i8 [[A:%.*]])
 ; CHECK-NEXT:    [[RES:%.*]] = sub i8 [[B:%.*]], [[CNT]]
 ; CHECK-NEXT:    ret i8 [[RES]]
 ;
@@ -322,7 +322,7 @@ define i8 @sub_ctpop_unknown(i8 %a, i8 %b)  {
 define <2 x i32> @sub_ctpop_vec(<2 x i32> %a) {
 ; CHECK-LABEL: @sub_ctpop_vec(
 ; CHECK-NEXT:    [[TMP1:%.*]] = xor <2 x i32> [[A:%.*]], <i32 -1, i32 -1>
-; CHECK-NEXT:    [[RES:%.*]] = call <2 x i32> @llvm.ctpop.v2i32(<2 x i32> [[TMP1]]), !range [[RNG1]]
+; CHECK-NEXT:    [[RES:%.*]] = call range(i32 0, 33) <2 x i32> @llvm.ctpop.v2i32(<2 x i32> [[TMP1]])
 ; CHECK-NEXT:    ret <2 x i32> [[RES]]
 ;
   %cnt = tail call <2 x i32> @llvm.ctpop.v2i32(<2 x i32> %a)
@@ -332,7 +332,7 @@ define <2 x i32> @sub_ctpop_vec(<2 x i32> %a) {
 
 define <2 x i32> @sub_ctpop_vec_extra_use(<2 x i32> %a, ptr %p) {
 ; CHECK-LABEL: @sub_ctpop_vec_extra_use(
-; CHECK-NEXT:    [[CNT:%.*]] = tail call <2 x i32> @llvm.ctpop.v2i32(<2 x i32> [[A:%.*]]), !range [[RNG1]]
+; CHECK-NEXT:    [[CNT:%.*]] = tail call range(i32 0, 33) <2 x i32> @llvm.ctpop.v2i32(<2 x i32> [[A:%.*]])
 ; CHECK-NEXT:    store <2 x i32> [[CNT]], ptr [[P:%.*]], align 8
 ; CHECK-NEXT:    [[RES:%.*]] = sub nuw nsw <2 x i32> <i32 32, i32 32>, [[CNT]]
 ; CHECK-NEXT:    ret <2 x i32> [[RES]]
@@ -345,7 +345,7 @@ define <2 x i32> @sub_ctpop_vec_extra_use(<2 x i32> %a, ptr %p) {
 
 define i32 @zext_ctpop(i16 %x) {
 ; CHECK-LABEL: @zext_ctpop(
-; CHECK-NEXT:    [[TMP1:%.*]] = call i16 @llvm.ctpop.i16(i16 [[X:%.*]]), !range [[RNG4:![0-9]+]]
+; CHECK-NEXT:    [[TMP1:%.*]] = call range(i16 0, 17) i16 @llvm.ctpop.i16(i16 [[X:%.*]])
 ; CHECK-NEXT:    [[P:%.*]] = zext nneg i16 [[TMP1]] to i32
 ; CHECK-NEXT:    ret i32 [[P]]
 ;
@@ -356,7 +356,7 @@ define i32 @zext_ctpop(i16 %x) {
 
 define <2 x i32> @zext_ctpop_vec(<2 x i7> %x) {
 ; CHECK-LABEL: @zext_ctpop_vec(
-; CHECK-NEXT:    [[TMP1:%.*]] = call <2 x i7> @llvm.ctpop.v2i7(<2 x i7> [[X:%.*]]), !range [[RNG2]]
+; CHECK-NEXT:    [[TMP1:%.*]] = call range(i7 0, 8) <2 x i7> @llvm.ctpop.v2i7(<2 x i7> [[X:%.*]])
 ; CHECK-NEXT:    [[P:%.*]] = zext nneg <2 x i7> [[TMP1]] to <2 x i32>
 ; CHECK-NEXT:    ret <2 x i32> [[P]]
 ;
@@ -369,7 +369,7 @@ define i32 @zext_ctpop_extra_use(i16 %x, ptr %q) {
 ; CHECK-LABEL: @zext_ctpop_extra_use(
 ; CHECK-NEXT:    [[Z:%.*]] = zext i16 [[X:%.*]] to i32
 ; CHECK-NEXT:    store i32 [[Z]], ptr [[Q:%.*]], align 4
-; CHECK-NEXT:    [[P:%.*]] = call i32 @llvm.ctpop.i32(i32 [[Z]]), !range [[RNG3]]
+; CHECK-NEXT:    [[P:%.*]] = call range(i32 0, 17) i32 @llvm.ctpop.i32(i32 [[Z]])
 ; CHECK-NEXT:    ret i32 [[P]]
 ;
   %z = zext i16 %x to i32
@@ -381,7 +381,7 @@ define i32 @zext_ctpop_extra_use(i16 %x, ptr %q) {
 define i32 @parity_xor(i32 %arg, i32 %arg1) {
 ; CHECK-LABEL: @parity_xor(
 ; CHECK-NEXT:    [[TMP1:%.*]] = xor i32 [[ARG1:%.*]], [[ARG:%.*]]
-; CHECK-NEXT:    [[TMP2:%.*]] = call i32 @llvm.ctpop.i32(i32 [[TMP1]]), !range [[RNG1]]
+; CHECK-NEXT:    [[TMP2:%.*]] = call range(i32 0, 33) i32 @llvm.ctpop.i32(i32 [[TMP1]])
 ; CHECK-NEXT:    [[I4:%.*]] = and i32 [[TMP2]], 1
 ; CHECK-NEXT:    ret i32 [[I4]]
 ;
@@ -395,7 +395,7 @@ define i32 @parity_xor(i32 %arg, i32 %arg1) {
 define i32 @parity_xor_trunc(i64 %arg, i64 %arg1) {
 ; CHECK-LABEL: @parity_xor_trunc(
 ; CHECK-NEXT:    [[TMP1:%.*]] = xor i64 [[ARG1:%.*]], [[ARG:%.*]]
-; CHECK-NEXT:    [[TMP2:%.*]] = call i64 @llvm.ctpop.i64(i64 [[TMP1]]), !range [[RNG5:![0-9]+]]
+; CHECK-NEXT:    [[TMP2:%.*]] = call range(i64 0, 65) i64 @llvm.ctpop.i64(i64 [[TMP1]])
 ; CHECK-NEXT:    [[I4:%.*]] = trunc nuw nsw i64 [[TMP2]] to i32
 ; CHECK-NEXT:    [[I5:%.*]] = and i32 [[I4]], 1
 ; CHECK-NEXT:    ret i32 [[I5]]
@@ -411,7 +411,7 @@ define i32 @parity_xor_trunc(i64 %arg, i64 %arg1) {
 define <2 x i32> @parity_xor_vec(<2 x i32> %arg, <2 x i32> %arg1) {
 ; CHECK-LABEL: @parity_xor_vec(
 ; CHECK-NEXT:    [[TMP1:%.*]] = xor <2 x i32> [[ARG1:%.*]], [[ARG:%.*]]
-; CHECK-NEXT:    [[TMP2:%.*]] = call <2 x i32> @llvm.ctpop.v2i32(<2 x i32> [[TMP1]]), !range [[RNG1]]
+; CHECK-NEXT:    [[TMP2:%.*]] = call range(i32 0, 33) <2 x i32> @llvm.ctpop.v2i32(<2 x i32> [[TMP1]])
 ; CHECK-NEXT:    [[I4:%.*]] = and <2 x i32> [[TMP2]], <i32 1, i32 1>
 ; CHECK-NEXT:    ret <2 x i32> [[I4]]
 ;
@@ -424,8 +424,8 @@ define <2 x i32> @parity_xor_vec(<2 x i32> %arg, <2 x i32> %arg1) {
 
 define i32 @parity_xor_wrong_cst(i32 %arg, i32 %arg1) {
 ; CHECK-LABEL: @parity_xor_wrong_cst(
-; CHECK-NEXT:    [[I:%.*]] = tail call i32 @llvm.ctpop.i32(i32 [[ARG:%.*]]), !range [[RNG1]]
-; CHECK-NEXT:    [[I2:%.*]] = tail call i32 @llvm.ctpop.i32(i32 [[ARG1:%.*]]), !range [[RNG1]]
+; CHECK-NEXT:    [[I:%.*]] = tail call range(i32 0, 33) i32 @llvm.ctpop.i32(i32 [[ARG:%.*]])
+; CHECK-NEXT:    [[I2:%.*]] = tail call range(i32 0, 33) i32 @llvm.ctpop.i32(i32 [[ARG1:%.*]])
 ; CHECK-NEXT:    [[I3:%.*]] = xor i32 [[I2]], [[I]]
 ; CHECK-NEXT:    [[I4:%.*]] = and i32 [[I3]], 3
 ; CHECK-NEXT:    ret i32 [[I4]]
@@ -439,11 +439,11 @@ define i32 @parity_xor_wrong_cst(i32 %arg, i32 %arg1) {
 
 define i32 @parity_xor_extra_use(i32 %arg, i32 %arg1) {
 ; CHECK-LABEL: @parity_xor_extra_use(
-; CHECK-NEXT:    [[I:%.*]] = tail call i32 @llvm.ctpop.i32(i32 [[ARG:%.*]]), !range [[RNG1]]
+; CHECK-NEXT:    [[I:%.*]] = tail call range(i32 0, 33) i32 @llvm.ctpop.i32(i32 [[ARG:%.*]])
 ; CHECK-NEXT:    [[I2:%.*]] = and i32 [[I]], 1
 ; CHECK-NEXT:    tail call void @use(i32 [[I2]])
 ; CHECK-NEXT:    [[TMP1:%.*]] = xor i32 [[ARG1:%.*]], [[ARG]]
-; CHECK-NEXT:    [[TMP2:%.*]] = call i32 @llvm.ctpop.i32(i32 [[TMP1]]), !range [[RNG1]]
+; CHECK-NEXT:    [[TMP2:%.*]] = call range(i32 0, 33) i32 @llvm.ctpop.i32(i32 [[TMP1]])
 ; CHECK-NEXT:    [[I5:%.*]] = and i32 [[TMP2]], 1
 ; CHECK-NEXT:    ret i32 [[I5]]
 ;
@@ -458,11 +458,11 @@ define i32 @parity_xor_extra_use(i32 %arg, i32 %arg1) {
 
 define i32 @parity_xor_extra_use2(i32 %arg, i32 %arg1) {
 ; CHECK-LABEL: @parity_xor_extra_use2(
-; CHECK-NEXT:    [[I:%.*]] = tail call i32 @llvm.ctpop.i32(i32 [[ARG1:%.*]]), !range [[RNG1]]
+; CHECK-NEXT:    [[I:%.*]] = tail call range(i32 0, 33) i32 @llvm.ctpop.i32(i32 [[ARG1:%.*]])
 ; CHECK-NEXT:    [[I2:%.*]] = and i32 [[I]], 1
 ; CHECK-NEXT:    tail call void @use(i32 [[I2]])
 ; CHECK-NEXT:    [[TMP1:%.*]] = xor i32 [[ARG1]], [[ARG:%.*]]
-; CHECK-NEXT:    [[TMP2:%.*]] = call i32 @llvm.ctpop.i32(i32 [[TMP1]]), !range [[RNG1]]
+; CHECK-NEXT:    [[TMP2:%.*]] = call range(i32 0, 33) i32 @llvm.ctpop.i32(i32 [[TMP1]])
 ; CHECK-NEXT:    [[I5:%.*]] = and i32 [[TMP2]], 1
 ; CHECK-NEXT:    ret i32 [[I5]]
 ;
@@ -477,7 +477,7 @@ define i32 @parity_xor_extra_use2(i32 %arg, i32 %arg1) {
 
 define i32 @select_ctpop_zero(i32 %x) {
 ; CHECK-LABEL: @select_ctpop_zero(
-; CHECK-NEXT:    [[CTPOP:%.*]] = call i32 @llvm.ctpop.i32(i32 [[X:%.*]]), !range [[RNG1]]
+; CHECK-NEXT:    [[CTPOP:%.*]] = call range(i32 0, 33) i32 @llvm.ctpop.i32(i32 [[X:%.*]])
 ; CHECK-NEXT:    ret i32 [[CTPOP]]
 ;
   %ctpop = call i32 @llvm.ctpop.i32(i32 %x)
diff --git a/llvm/test/Transforms/InstCombine/cttz-abs.ll b/llvm/test/Transforms/InstCombine/cttz-abs.ll
index 0141b2cd71ce..9d3b6168e710 100644
--- a/llvm/test/Transforms/InstCombine/cttz-abs.ll
+++ b/llvm/test/Transforms/InstCombine/cttz-abs.ll
@@ -3,7 +3,7 @@
 
 define i32 @cttz_abs(i32 %x) {
 ; CHECK-LABEL: @cttz_abs(
-; CHECK-NEXT:    [[R:%.*]] = tail call i32 @llvm.cttz.i32(i32 [[X:%.*]], i1 true), !range [[RNG0:![0-9]+]]
+; CHECK-NEXT:    [[R:%.*]] = tail call range(i32 0, 33) i32 @llvm.cttz.i32(i32 [[X:%.*]], i1 true)
 ; CHECK-NEXT:    ret i32 [[R]]
 ;
   %c = icmp slt i32 %x, 0
@@ -15,7 +15,7 @@ define i32 @cttz_abs(i32 %x) {
 
 define <2 x i64> @cttz_abs_vec(<2 x i64> %x) {
 ; CHECK-LABEL: @cttz_abs_vec(
-; CHECK-NEXT:    [[R:%.*]] = call <2 x i64> @llvm.cttz.v2i64(<2 x i64> [[X:%.*]], i1 false), !range [[RNG1:![0-9]+]]
+; CHECK-NEXT:    [[R:%.*]] = call range(i64 0, 65) <2 x i64> @llvm.cttz.v2i64(<2 x i64> [[X:%.*]], i1 false)
 ; CHECK-NEXT:    ret <2 x i64> [[R]]
 ;
   %c = icmp slt <2 x i64> %x, zeroinitializer
@@ -29,7 +29,7 @@ define i32 @cttz_abs2(i32 %x) {
 ; CHECK-LABEL: @cttz_abs2(
 ; CHECK-NEXT:    [[C:%.*]] = icmp sgt i32 [[X:%.*]], 0
 ; CHECK-NEXT:    call void @use_cond(i1 [[C]])
-; CHECK-NEXT:    [[R:%.*]] = tail call i32 @llvm.cttz.i32(i32 [[X]], i1 true), !range [[RNG0]]
+; CHECK-NEXT:    [[R:%.*]] = tail call range(i32 0, 33) i32 @llvm.cttz.i32(i32 [[X]], i1 true)
 ; CHECK-NEXT:    ret i32 [[R]]
 ;
   %c = icmp sgt i32 %x, 0
@@ -44,7 +44,7 @@ define i32 @cttz_abs3(i32 %x) {
 ; CHECK-LABEL: @cttz_abs3(
 ; CHECK-NEXT:    [[C:%.*]] = icmp sgt i32 [[X:%.*]], -1
 ; CHECK-NEXT:    call void @use_cond(i1 [[C]])
-; CHECK-NEXT:    [[R:%.*]] = tail call i32 @llvm.cttz.i32(i32 [[X]], i1 true), !range [[RNG0]]
+; CHECK-NEXT:    [[R:%.*]] = tail call range(i32 0, 33) i32 @llvm.cttz.i32(i32 [[X]], i1 true)
 ; CHECK-NEXT:    ret i32 [[R]]
 ;
   %c = icmp sgt i32 %x, -1
@@ -57,7 +57,7 @@ define i32 @cttz_abs3(i32 %x) {
 
 define i32 @cttz_abs4(i32 %x) {
 ; CHECK-LABEL: @cttz_abs4(
-; CHECK-NEXT:    [[R:%.*]] = tail call i32 @llvm.cttz.i32(i32 [[X:%.*]], i1 true), !range [[RNG0]]
+; CHECK-NEXT:    [[R:%.*]] = tail call range(i32 0, 33) i32 @llvm.cttz.i32(i32 [[X:%.*]], i1 true)
 ; CHECK-NEXT:    ret i32 [[R]]
 ;
   %c = icmp slt i32 %x, 1
@@ -69,7 +69,7 @@ define i32 @cttz_abs4(i32 %x) {
 
 define i32 @cttz_nabs(i32 %x) {
 ; CHECK-LABEL: @cttz_nabs(
-; CHECK-NEXT:    [[R:%.*]] = tail call i32 @llvm.cttz.i32(i32 [[X:%.*]], i1 false), !range [[RNG0]]
+; CHECK-NEXT:    [[R:%.*]] = tail call range(i32 0, 33) i32 @llvm.cttz.i32(i32 [[X:%.*]], i1 false)
 ; CHECK-NEXT:    ret i32 [[R]]
 ;
   %c = icmp slt i32 %x, 0
@@ -81,7 +81,7 @@ define i32 @cttz_nabs(i32 %x) {
 
 define <2 x i64> @cttz_nabs_vec(<2 x i64> %x) {
 ; CHECK-LABEL: @cttz_nabs_vec(
-; CHECK-NEXT:    [[R:%.*]] = call <2 x i64> @llvm.cttz.v2i64(<2 x i64> [[X:%.*]], i1 false), !range [[RNG1]]
+; CHECK-NEXT:    [[R:%.*]] = call range(i64 0, 65) <2 x i64> @llvm.cttz.v2i64(<2 x i64> [[X:%.*]], i1 false)
 ; CHECK-NEXT:    ret <2 x i64> [[R]]
 ;
   %c = icmp slt <2 x i64> %x, zeroinitializer
@@ -93,7 +93,7 @@ define <2 x i64> @cttz_nabs_vec(<2 x i64> %x) {
 
 define i64 @cttz_abs_64(i64 %x) {
 ; CHECK-LABEL: @cttz_abs_64(
-; CHECK-NEXT:    [[R:%.*]] = call i64 @llvm.cttz.i64(i64 [[X:%.*]], i1 false), !range [[RNG1]]
+; CHECK-NEXT:    [[R:%.*]] = call range(i64 0, 65) i64 @llvm.cttz.i64(i64 [[X:%.*]], i1 false)
 ; CHECK-NEXT:    ret i64 [[R]]
 ;
   %c = icmp slt i64 %x, 0
@@ -107,7 +107,7 @@ define i32 @cttz_abs_multiuse(i32 %x) {
 ; CHECK-LABEL: @cttz_abs_multiuse(
 ; CHECK-NEXT:    [[D:%.*]] = call i32 @llvm.abs.i32(i32 [[X:%.*]], i1 false)
 ; CHECK-NEXT:    call void @use_abs(i32 [[D]])
-; CHECK-NEXT:    [[R:%.*]] = tail call i32 @llvm.cttz.i32(i32 [[X]], i1 true), !range [[RNG0]]
+; CHECK-NEXT:    [[R:%.*]] = tail call range(i32 0, 33) i32 @llvm.cttz.i32(i32 [[X]], i1 true)
 ; CHECK-NEXT:    ret i32 [[R]]
 ;
   %c = icmp slt i32 %x, 1
@@ -123,7 +123,7 @@ define i32 @cttz_nabs_multiuse(i32 %x) {
 ; CHECK-NEXT:    [[TMP1:%.*]] = call i32 @llvm.abs.i32(i32 [[X:%.*]], i1 false)
 ; CHECK-NEXT:    [[D:%.*]] = sub i32 0, [[TMP1]]
 ; CHECK-NEXT:    call void @use_abs(i32 [[D]])
-; CHECK-NEXT:    [[R:%.*]] = tail call i32 @llvm.cttz.i32(i32 [[X]], i1 true), !range [[RNG0]]
+; CHECK-NEXT:    [[R:%.*]] = tail call range(i32 0, 33) i32 @llvm.cttz.i32(i32 [[X]], i1 true)
 ; CHECK-NEXT:    ret i32 [[R]]
 ;
   %c = icmp slt i32 %x, 1
@@ -141,7 +141,7 @@ define i32 @no_cttz_abs(i32 %x) {
 ; CHECK-NEXT:    [[C:%.*]] = icmp slt i32 [[X:%.*]], 2
 ; CHECK-NEXT:    [[S:%.*]] = sub i32 0, [[X]]
 ; CHECK-NEXT:    [[D:%.*]] = select i1 [[C]], i32 [[S]], i32 [[X]]
-; CHECK-NEXT:    [[R:%.*]] = tail call i32 @llvm.cttz.i32(i32 [[D]], i1 true), !range [[RNG0]]
+; CHECK-NEXT:    [[R:%.*]] = tail call range(i32 0, 33) i32 @llvm.cttz.i32(i32 [[D]], i1 true)
 ; CHECK-NEXT:    ret i32 [[R]]
 ;
   %c = icmp slt i32 %x, 2
@@ -156,7 +156,7 @@ define i32 @no_cttz_abs2(i32 %x) {
 ; CHECK-NEXT:    [[C:%.*]] = icmp slt i32 [[X:%.*]], 0
 ; CHECK-NEXT:    [[S:%.*]] = sub i32 1, [[X]]
 ; CHECK-NEXT:    [[D:%.*]] = select i1 [[C]], i32 [[S]], i32 [[X]]
-; CHECK-NEXT:    [[R:%.*]] = tail call i32 @llvm.cttz.i32(i32 [[D]], i1 true), !range [[RNG0]]
+; CHECK-NEXT:    [[R:%.*]] = tail call range(i32 0, 33) i32 @llvm.cttz.i32(i32 [[D]], i1 true)
 ; CHECK-NEXT:    ret i32 [[R]]
 ;
   %c = icmp slt i32 %x, 0
@@ -172,7 +172,7 @@ define i32 @no_cttz_abs3(i32 %x) {
 ; CHECK-NEXT:    call void @use_cond(i1 [[C]])
 ; CHECK-NEXT:    [[S:%.*]] = sub i32 0, [[X]]
 ; CHECK-NEXT:    [[D:%.*]] = select i1 [[C]], i32 [[X]], i32 [[S]]
-; CHECK-NEXT:    [[R:%.*]] = tail call i32 @llvm.cttz.i32(i32 [[D]], i1 true), !range [[RNG0]]
+; CHECK-NEXT:    [[R:%.*]] = tail call range(i32 0, 33) i32 @llvm.cttz.i32(i32 [[D]], i1 true)
 ; CHECK-NEXT:    ret i32 [[R]]
 ;
   %c = icmp sgt i32 %x, -2
@@ -188,7 +188,7 @@ define <2 x i64> @no_cttz_abs_vec(<2 x i64> %x) {
 ; CHECK-NEXT:    [[C:%.*]] = icmp slt <2 x i64> [[X:%.*]], <i64 2, i64 1>
 ; CHECK-NEXT:    [[S:%.*]] = sub <2 x i64> <i64 1, i64 0>, [[X]]
 ; CHECK-NEXT:    [[D:%.*]] = select <2 x i1> [[C]], <2 x i64> [[S]], <2 x i64> [[X]]
-; CHECK-NEXT:    [[R:%.*]] = call <2 x i64> @llvm.cttz.v2i64(<2 x i64> [[D]], i1 false), !range [[RNG1]]
+; CHECK-NEXT:    [[R:%.*]] = call range(i64 0, 65) <2 x i64> @llvm.cttz.v2i64(<2 x i64> [[D]], i1 false)
 ; CHECK-NEXT:    ret <2 x i64> [[R]]
 ;
   %c = icmp slt <2 x i64> %x, <i64 2, i64 1>
@@ -203,7 +203,7 @@ define <2 x i64> @no_cttz_nabs_vec(<2 x i64> %x) {
 ; CHECK-NEXT:    [[C:%.*]] = icmp slt <2 x i64> [[X:%.*]], <i64 2, i64 1>
 ; CHECK-NEXT:    [[S:%.*]] = sub <2 x i64> <i64 1, i64 0>, [[X]]
 ; CHECK-NEXT:    [[D:%.*]] = select <2 x i1> [[C]], <2 x i64> [[X]], <2 x i64> [[S]]
-; CHECK-NEXT:    [[R:%.*]] = call <2 x i64> @llvm.cttz.v2i64(<2 x i64> [[D]], i1 false), !range [[RNG1]]
+; CHECK-NEXT:    [[R:%.*]] = call range(i64 0, 65) <2 x i64> @llvm.cttz.v2i64(<2 x i64> [[D]], i1 false)
 ; CHECK-NEXT:    ret <2 x i64> [[R]]
 ;
   %c = icmp slt <2 x i64> %x, <i64 2, i64 1>
@@ -215,7 +215,7 @@ define <2 x i64> @no_cttz_nabs_vec(<2 x i64> %x) {
 
 define i32 @cttz_abs_intrin(i32 %x) {
 ; CHECK-LABEL: @cttz_abs_intrin(
-; CHECK-NEXT:    [[R:%.*]] = call i32 @llvm.cttz.i32(i32 [[X:%.*]], i1 false), !range [[RNG0]]
+; CHECK-NEXT:    [[R:%.*]] = call range(i32 0, 33) i32 @llvm.cttz.i32(i32 [[X:%.*]], i1 false)
 ; CHECK-NEXT:    ret i32 [[R]]
 ;
   %a = call i32 @llvm.abs.i32(i32 %x, i1 false)
@@ -225,7 +225,7 @@ define i32 @cttz_abs_intrin(i32 %x) {
 
 define i32 @cttz_nabs_intrin(i32 %x) {
 ; CHECK-LABEL: @cttz_nabs_intrin(
-; CHECK-NEXT:    [[R:%.*]] = call i32 @llvm.cttz.i32(i32 [[X:%.*]], i1 false), !range [[RNG0]]
+; CHECK-NEXT:    [[R:%.*]] = call range(i32 0, 33) i32 @llvm.cttz.i32(i32 [[X:%.*]], i1 false)
 ; CHECK-NEXT:    ret i32 [[R]]
 ;
   %a = call i32 @llvm.abs.i32(i32 %x, i1 false)
diff --git a/llvm/test/Transforms/InstCombine/cttz-negative.ll b/llvm/test/Transforms/InstCombine/cttz-negative.ll
index 139da840af9e..6f812e420935 100644
--- a/llvm/test/Transforms/InstCombine/cttz-negative.ll
+++ b/llvm/test/Transforms/InstCombine/cttz-negative.ll
@@ -3,7 +3,7 @@
 
 define i32 @cttz_neg_value(i32 %x) {
 ; CHECK-LABEL: @cttz_neg_value(
-; CHECK-NEXT:    [[B:%.*]] = call i32 @llvm.cttz.i32(i32 [[X:%.*]], i1 false), !range [[RNG0:![0-9]+]]
+; CHECK-NEXT:    [[B:%.*]] = call range(i32 0, 33) i32 @llvm.cttz.i32(i32 [[X:%.*]], i1 false)
 ; CHECK-NEXT:    ret i32 [[B]]
 ;
   %a = sub i32 0, %x
@@ -15,7 +15,7 @@ define i32 @cttz_neg_value_multiuse(i32 %x) {
 ; CHECK-LABEL: @cttz_neg_value_multiuse(
 ; CHECK-NEXT:    [[A:%.*]] = sub i32 0, [[X:%.*]]
 ; CHECK-NEXT:    call void @use(i32 [[A]])
-; CHECK-NEXT:    [[B:%.*]] = call i32 @llvm.cttz.i32(i32 [[X]], i1 false), !range [[RNG0]]
+; CHECK-NEXT:    [[B:%.*]] = call range(i32 0, 33) i32 @llvm.cttz.i32(i32 [[X]], i1 false)
 ; CHECK-NEXT:    ret i32 [[B]]
 ;
   %a = sub i32 0, %x
@@ -26,7 +26,7 @@ define i32 @cttz_neg_value_multiuse(i32 %x) {
 
 define i64 @cttz_neg_value_64(i64 %x) {
 ; CHECK-LABEL: @cttz_neg_value_64(
-; CHECK-NEXT:    [[B:%.*]] = tail call i64 @llvm.cttz.i64(i64 [[X:%.*]], i1 true), !range [[RNG1:![0-9]+]]
+; CHECK-NEXT:    [[B:%.*]] = tail call range(i64 0, 65) i64 @llvm.cttz.i64(i64 [[X:%.*]], i1 true)
 ; CHECK-NEXT:    ret i64 [[B]]
 ;
   %a = sub i64 0, %x
@@ -36,7 +36,7 @@ define i64 @cttz_neg_value_64(i64 %x) {
 
 define i64 @cttz_neg_value2_64(i64 %x) {
 ; CHECK-LABEL: @cttz_neg_value2_64(
-; CHECK-NEXT:    [[B:%.*]] = tail call i64 @llvm.cttz.i64(i64 [[X:%.*]], i1 false), !range [[RNG1]]
+; CHECK-NEXT:    [[B:%.*]] = tail call range(i64 0, 65) i64 @llvm.cttz.i64(i64 [[X:%.*]], i1 false)
 ; CHECK-NEXT:    ret i64 [[B]]
 ;
   %a = sub i64 0, %x
@@ -46,7 +46,7 @@ define i64 @cttz_neg_value2_64(i64 %x) {
 
 define <2 x i64> @cttz_neg_value_vec(<2 x i64> %x) {
 ; CHECK-LABEL: @cttz_neg_value_vec(
-; CHECK-NEXT:    [[B:%.*]] = call <2 x i64> @llvm.cttz.v2i64(<2 x i64> [[X:%.*]], i1 false), !range [[RNG1]]
+; CHECK-NEXT:    [[B:%.*]] = call range(i64 0, 65) <2 x i64> @llvm.cttz.v2i64(<2 x i64> [[X:%.*]], i1 false)
 ; CHECK-NEXT:    ret <2 x i64> [[B]]
 ;
   %a = sub  <2 x i64> zeroinitializer, %x
@@ -59,7 +59,7 @@ define <2 x i64> @cttz_neg_value_vec(<2 x i64> %x) {
 define i32 @cttz_nonneg_value(i32 %x) {
 ; CHECK-LABEL: @cttz_nonneg_value(
 ; CHECK-NEXT:    [[A:%.*]] = sub i32 1, [[X:%.*]]
-; CHECK-NEXT:    [[B:%.*]] = call i32 @llvm.cttz.i32(i32 [[A]], i1 false), !range [[RNG0]]
+; CHECK-NEXT:    [[B:%.*]] = call range(i32 0, 33) i32 @llvm.cttz.i32(i32 [[A]], i1 false)
 ; CHECK-NEXT:    ret i32 [[B]]
 ;
   %a = sub i32 1, %x
@@ -70,7 +70,7 @@ define i32 @cttz_nonneg_value(i32 %x) {
 define <2 x i64> @cttz_nonneg_value_vec(<2 x i64> %x) {
 ; CHECK-LABEL: @cttz_nonneg_value_vec(
 ; CHECK-NEXT:    [[A:%.*]] = sub <2 x i64> <i64 1, i64 0>, [[X:%.*]]
-; CHECK-NEXT:    [[B:%.*]] = call <2 x i64> @llvm.cttz.v2i64(<2 x i64> [[A]], i1 false), !range [[RNG1]]
+; CHECK-NEXT:    [[B:%.*]] = call range(i64 0, 65) <2 x i64> @llvm.cttz.v2i64(<2 x i64> [[A]], i1 false)
 ; CHECK-NEXT:    ret <2 x i64> [[B]]
 ;
   %a = sub  <2 x i64> <i64 1, i64 0>, %x
diff --git a/llvm/test/Transforms/InstCombine/cttz.ll b/llvm/test/Transforms/InstCombine/cttz.ll
index 6ea5e5e141b2..3595cff5f1ae 100644
--- a/llvm/test/Transforms/InstCombine/cttz.ll
+++ b/llvm/test/Transforms/InstCombine/cttz.ll
@@ -8,7 +8,7 @@ declare void @use(i32)
 
 define i32 @cttz_zext_zero_undef(i16 %x) {
 ; CHECK-LABEL: @cttz_zext_zero_undef(
-; CHECK-NEXT:    [[TMP1:%.*]] = call i16 @llvm.cttz.i16(i16 [[X:%.*]], i1 true), !range [[RNG0:![0-9]+]]
+; CHECK-NEXT:    [[TMP1:%.*]] = call range(i16 0, 17) i16 @llvm.cttz.i16(i16 [[X:%.*]], i1 true)
 ; CHECK-NEXT:    [[TZ:%.*]] = zext nneg i16 [[TMP1]] to i32
 ; CHECK-NEXT:    ret i32 [[TZ]]
 ;
@@ -20,7 +20,7 @@ define i32 @cttz_zext_zero_undef(i16 %x) {
 define i32 @cttz_zext_zero_def(i16 %x) {
 ; CHECK-LABEL: @cttz_zext_zero_def(
 ; CHECK-NEXT:    [[Z:%.*]] = zext i16 [[X:%.*]] to i32
-; CHECK-NEXT:    [[TZ:%.*]] = call i32 @llvm.cttz.i32(i32 [[Z]], i1 false), !range [[RNG1:![0-9]+]]
+; CHECK-NEXT:    [[TZ:%.*]] = call range(i32 0, 33) i32 @llvm.cttz.i32(i32 [[Z]], i1 false)
 ; CHECK-NEXT:    ret i32 [[TZ]]
 ;
   %z = zext i16 %x to i32
@@ -32,7 +32,7 @@ define i32 @cttz_zext_zero_undef_extra_use(i16 %x) {
 ; CHECK-LABEL: @cttz_zext_zero_undef_extra_use(
 ; CHECK-NEXT:    [[Z:%.*]] = zext i16 [[X:%.*]] to i32
 ; CHECK-NEXT:    call void @use(i32 [[Z]])
-; CHECK-NEXT:    [[TZ:%.*]] = call i32 @llvm.cttz.i32(i32 [[Z]], i1 true), !range [[RNG1]]
+; CHECK-NEXT:    [[TZ:%.*]] = call range(i32 0, 33) i32 @llvm.cttz.i32(i32 [[Z]], i1 true)
 ; CHECK-NEXT:    ret i32 [[TZ]]
 ;
   %z = zext i16 %x to i32
@@ -43,7 +43,7 @@ define i32 @cttz_zext_zero_undef_extra_use(i16 %x) {
 
 define <2 x i64> @cttz_zext_zero_undef_vec(<2 x i32> %x) {
 ; CHECK-LABEL: @cttz_zext_zero_undef_vec(
-; CHECK-NEXT:    [[TMP1:%.*]] = call <2 x i32> @llvm.cttz.v2i32(<2 x i32> [[X:%.*]], i1 true), !range [[RNG1]]
+; CHECK-NEXT:    [[TMP1:%.*]] = call range(i32 0, 33) <2 x i32> @llvm.cttz.v2i32(<2 x i32> [[X:%.*]], i1 true)
 ; CHECK-NEXT:    [[TZ:%.*]] = zext nneg <2 x i32> [[TMP1]] to <2 x i64>
 ; CHECK-NEXT:    ret <2 x i64> [[TZ]]
 ;
@@ -55,7 +55,7 @@ define <2 x i64> @cttz_zext_zero_undef_vec(<2 x i32> %x) {
 define <2 x i64> @cttz_zext_zero_def_vec(<2 x i32> %x) {
 ; CHECK-LABEL: @cttz_zext_zero_def_vec(
 ; CHECK-NEXT:    [[Z:%.*]] = zext <2 x i32> [[X:%.*]] to <2 x i64>
-; CHECK-NEXT:    [[TZ:%.*]] = tail call <2 x i64> @llvm.cttz.v2i64(<2 x i64> [[Z]], i1 false), !range [[RNG2:![0-9]+]]
+; CHECK-NEXT:    [[TZ:%.*]] = tail call range(i64 0, 65) <2 x i64> @llvm.cttz.v2i64(<2 x i64> [[Z]], i1 false)
 ; CHECK-NEXT:    ret <2 x i64> [[TZ]]
 ;
   %z = zext <2 x i32> %x to <2 x i64>
@@ -65,7 +65,7 @@ define <2 x i64> @cttz_zext_zero_def_vec(<2 x i32> %x) {
 
 define i32 @cttz_sext_zero_undef(i16 %x) {
 ; CHECK-LABEL: @cttz_sext_zero_undef(
-; CHECK-NEXT:    [[TMP1:%.*]] = call i16 @llvm.cttz.i16(i16 [[X:%.*]], i1 true), !range [[RNG0]]
+; CHECK-NEXT:    [[TMP1:%.*]] = call range(i16 0, 17) i16 @llvm.cttz.i16(i16 [[X:%.*]], i1 true)
 ; CHECK-NEXT:    [[TZ:%.*]] = zext nneg i16 [[TMP1]] to i32
 ; CHECK-NEXT:    ret i32 [[TZ]]
 ;
@@ -77,7 +77,7 @@ define i32 @cttz_sext_zero_undef(i16 %x) {
 define i32 @cttz_sext_zero_def(i16 %x) {
 ; CHECK-LABEL: @cttz_sext_zero_def(
 ; CHECK-NEXT:    [[TMP1:%.*]] = zext i16 [[X:%.*]] to i32
-; CHECK-NEXT:    [[TZ:%.*]] = call i32 @llvm.cttz.i32(i32 [[TMP1]], i1 false), !range [[RNG1]]
+; CHECK-NEXT:    [[TZ:%.*]] = call range(i32 0, 33) i32 @llvm.cttz.i32(i32 [[TMP1]], i1 false)
 ; CHECK-NEXT:    ret i32 [[TZ]]
 ;
   %s = sext i16 %x to i32
@@ -89,7 +89,7 @@ define i32 @cttz_sext_zero_undef_extra_use(i16 %x) {
 ; CHECK-LABEL: @cttz_sext_zero_undef_extra_use(
 ; CHECK-NEXT:    [[S:%.*]] = sext i16 [[X:%.*]] to i32
 ; CHECK-NEXT:    call void @use(i32 [[S]])
-; CHECK-NEXT:    [[TZ:%.*]] = call i32 @llvm.cttz.i32(i32 [[S]], i1 true), !range [[RNG1]]
+; CHECK-NEXT:    [[TZ:%.*]] = call range(i32 0, 33) i32 @llvm.cttz.i32(i32 [[S]], i1 true)
 ; CHECK-NEXT:    ret i32 [[TZ]]
 ;
   %s = sext i16 %x to i32
@@ -100,7 +100,7 @@ define i32 @cttz_sext_zero_undef_extra_use(i16 %x) {
 
 define <2 x i64> @cttz_sext_zero_undef_vec(<2 x i32> %x) {
 ; CHECK-LABEL: @cttz_sext_zero_undef_vec(
-; CHECK-NEXT:    [[TMP1:%.*]] = call <2 x i32> @llvm.cttz.v2i32(<2 x i32> [[X:%.*]], i1 true), !range [[RNG1]]
+; CHECK-NEXT:    [[TMP1:%.*]] = call range(i32 0, 33) <2 x i32> @llvm.cttz.v2i32(<2 x i32> [[X:%.*]], i1 true)
 ; CHECK-NEXT:    [[TZ:%.*]] = zext nneg <2 x i32> [[TMP1]] to <2 x i64>
 ; CHECK-NEXT:    ret <2 x i64> [[TZ]]
 ;
@@ -112,7 +112,7 @@ define <2 x i64> @cttz_sext_zero_undef_vec(<2 x i32> %x) {
 define <2 x i64> @cttz_sext_zero_def_vec(<2 x i32> %x) {
 ; CHECK-LABEL: @cttz_sext_zero_def_vec(
 ; CHECK-NEXT:    [[TMP1:%.*]] = zext <2 x i32> [[X:%.*]] to <2 x i64>
-; CHECK-NEXT:    [[TZ:%.*]] = call <2 x i64> @llvm.cttz.v2i64(<2 x i64> [[TMP1]], i1 false), !range [[RNG2]]
+; CHECK-NEXT:    [[TZ:%.*]] = call range(i64 0, 65) <2 x i64> @llvm.cttz.v2i64(<2 x i64> [[TMP1]], i1 false)
 ; CHECK-NEXT:    ret <2 x i64> [[TZ]]
 ;
   %s = sext <2 x i32> %x to <2 x i64>
@@ -122,7 +122,7 @@ define <2 x i64> @cttz_sext_zero_def_vec(<2 x i32> %x) {
 
 define i32 @cttz_of_lowest_set_bit(i32 %x) {
 ; CHECK-LABEL: @cttz_of_lowest_set_bit(
-; CHECK-NEXT:    [[TZ:%.*]] = call i32 @llvm.cttz.i32(i32 [[X:%.*]], i1 false), !range [[RNG1]]
+; CHECK-NEXT:    [[TZ:%.*]] = call range(i32 0, 33) i32 @llvm.cttz.i32(i32 [[X:%.*]], i1 false)
 ; CHECK-NEXT:    ret i32 [[TZ]]
 ;
   %sub = sub i32 0, %x
@@ -134,7 +134,7 @@ define i32 @cttz_of_lowest_set_bit(i32 %x) {
 define i32 @cttz_of_lowest_set_bit_commuted(i32 %xx) {
 ; CHECK-LABEL: @cttz_of_lowest_set_bit_commuted(
 ; CHECK-NEXT:    [[X:%.*]] = udiv i32 42, [[XX:%.*]]
-; CHECK-NEXT:    [[TZ:%.*]] = call i32 @llvm.cttz.i32(i32 [[X]], i1 false), !range [[RNG1]]
+; CHECK-NEXT:    [[TZ:%.*]] = call range(i32 0, 33) i32 @llvm.cttz.i32(i32 [[X]], i1 false)
 ; CHECK-NEXT:    ret i32 [[TZ]]
 ;
   %x = udiv i32 42, %xx ; thwart complexity-based canonicalization
@@ -146,7 +146,7 @@ define i32 @cttz_of_lowest_set_bit_commuted(i32 %xx) {
 
 define i32 @cttz_of_lowest_set_bit_poison_flag(i32 %x) {
 ; CHECK-LABEL: @cttz_of_lowest_set_bit_poison_flag(
-; CHECK-NEXT:    [[TZ:%.*]] = call i32 @llvm.cttz.i32(i32 [[X:%.*]], i1 true), !range [[RNG1]]
+; CHECK-NEXT:    [[TZ:%.*]] = call range(i32 0, 33) i32 @llvm.cttz.i32(i32 [[X:%.*]], i1 true)
 ; CHECK-NEXT:    ret i32 [[TZ]]
 ;
   %sub = sub i32 0, %x
@@ -157,7 +157,7 @@ define i32 @cttz_of_lowest_set_bit_poison_flag(i32 %x) {
 
 define <2 x i64> @cttz_of_lowest_set_bit_vec(<2 x i64> %x) {
 ; CHECK-LABEL: @cttz_of_lowest_set_bit_vec(
-; CHECK-NEXT:    [[TZ:%.*]] = call <2 x i64> @llvm.cttz.v2i64(<2 x i64> [[X:%.*]], i1 false), !range [[RNG2]]
+; CHECK-NEXT:    [[TZ:%.*]] = call range(i64 0, 65) <2 x i64> @llvm.cttz.v2i64(<2 x i64> [[X:%.*]], i1 false)
 ; CHECK-NEXT:    ret <2 x i64> [[TZ]]
 ;
   %sub = sub <2 x i64> zeroinitializer, %x
@@ -168,7 +168,7 @@ define <2 x i64> @cttz_of_lowest_set_bit_vec(<2 x i64> %x) {
 
 define <2 x i64> @cttz_of_lowest_set_bit_vec_undef(<2 x i64> %x) {
 ; CHECK-LABEL: @cttz_of_lowest_set_bit_vec_undef(
-; CHECK-NEXT:    [[TZ:%.*]] = call <2 x i64> @llvm.cttz.v2i64(<2 x i64> [[X:%.*]], i1 false), !range [[RNG2]]
+; CHECK-NEXT:    [[TZ:%.*]] = call range(i64 0, 65) <2 x i64> @llvm.cttz.v2i64(<2 x i64> [[X:%.*]], i1 false)
 ; CHECK-NEXT:    ret <2 x i64> [[TZ]]
 ;
   %sub = sub <2 x i64> zeroinitializer, %x
@@ -181,7 +181,7 @@ define i32 @cttz_of_lowest_set_bit_wrong_const(i32 %x) {
 ; CHECK-LABEL: @cttz_of_lowest_set_bit_wrong_const(
 ; CHECK-NEXT:    [[SUB:%.*]] = sub i32 1, [[X:%.*]]
 ; CHECK-NEXT:    [[AND:%.*]] = and i32 [[SUB]], [[X]]
-; CHECK-NEXT:    [[TZ:%.*]] = call i32 @llvm.cttz.i32(i32 [[AND]], i1 false), !range [[RNG3:![0-9]+]]
+; CHECK-NEXT:    [[TZ:%.*]] = call range(i32 1, 33) i32 @llvm.cttz.i32(i32 [[AND]], i1 false)
 ; CHECK-NEXT:    ret i32 [[TZ]]
 ;
   %sub = sub i32 1, %x
@@ -194,7 +194,7 @@ define i32 @cttz_of_lowest_set_bit_wrong_operand(i32 %x, i32 %y) {
 ; CHECK-LABEL: @cttz_of_lowest_set_bit_wrong_operand(
 ; CHECK-NEXT:    [[SUB:%.*]] = sub i32 0, [[Y:%.*]]
 ; CHECK-NEXT:    [[AND:%.*]] = and i32 [[SUB]], [[X:%.*]]
-; CHECK-NEXT:    [[TZ:%.*]] = call i32 @llvm.cttz.i32(i32 [[AND]], i1 false), !range [[RNG1]]
+; CHECK-NEXT:    [[TZ:%.*]] = call range(i32 0, 33) i32 @llvm.cttz.i32(i32 [[AND]], i1 false)
 ; CHECK-NEXT:    ret i32 [[TZ]]
 ;
   %sub = sub i32 0, %y
@@ -207,7 +207,7 @@ define i32 @cttz_of_lowest_set_bit_wrong_intrinsic(i32 %x) {
 ; CHECK-LABEL: @cttz_of_lowest_set_bit_wrong_intrinsic(
 ; CHECK-NEXT:    [[SUB:%.*]] = sub i32 0, [[X:%.*]]
 ; CHECK-NEXT:    [[AND:%.*]] = and i32 [[SUB]], [[X]]
-; CHECK-NEXT:    [[TZ:%.*]] = call i32 @llvm.ctlz.i32(i32 [[AND]], i1 false), !range [[RNG1]]
+; CHECK-NEXT:    [[TZ:%.*]] = call range(i32 0, 33) i32 @llvm.ctlz.i32(i32 [[AND]], i1 false)
 ; CHECK-NEXT:    ret i32 [[TZ]]
 ;
   %sub = sub i32 0, %x
diff --git a/llvm/test/Transforms/InstCombine/fcmp-select.ll b/llvm/test/Transforms/InstCombine/fcmp-select.ll
index f37c586845b1..028de1ff8a99 100644
--- a/llvm/test/Transforms/InstCombine/fcmp-select.ll
+++ b/llvm/test/Transforms/InstCombine/fcmp-select.ll
@@ -2,6 +2,7 @@
 ; RUN: opt < %s -passes=instcombine -S | FileCheck %s
 
 declare void @use(i1)
+declare void @usef64(double)
 
 ; X == 42.0 ? X : 42.0 --> 42.0
 
@@ -148,3 +149,122 @@ define i1 @fcmp_ogt_select(i1 %cond, float %a, float %b) {
   %res = fcmp ogt float %lhs, %rhs
   ret i1 %res
 }
+
+define i1 @test_fcmp_select_const_const(double %x) {
+; CHECK-LABEL: @test_fcmp_select_const_const(
+; CHECK-NEXT:    [[CMP1:%.*]] = fcmp uno double [[X:%.*]], 0.000000e+00
+; CHECK-NEXT:    ret i1 [[CMP1]]
+;
+  %cmp1 = fcmp ord double %x, 0.000000e+00
+  %sel = select i1 %cmp1, double 0xFFFFFFFFFFFFFFFF, double 0.000000e+00
+  %cmp2 = fcmp oeq double %sel, 0.000000e+00
+  ret i1 %cmp2
+}
+
+define i1 @test_fcmp_select_var_const(double %x, double %y) {
+; CHECK-LABEL: @test_fcmp_select_var_const(
+; CHECK-NEXT:    [[CMP1:%.*]] = fcmp ule double [[X:%.*]], 0x3E80000000000000
+; CHECK-NEXT:    [[TMP1:%.*]] = fcmp olt double [[Y:%.*]], 0x3E80000000000000
+; CHECK-NEXT:    [[CMP2:%.*]] = select i1 [[CMP1]], i1 true, i1 [[TMP1]]
+; CHECK-NEXT:    ret i1 [[CMP2]]
+;
+  %cmp1 = fcmp ogt double %x, 0x3E80000000000000
+  %sel = select i1 %cmp1, double %y, double 0.000000e+00
+  %cmp2 = fcmp olt double %sel, 0x3E80000000000000
+  ret i1 %cmp2
+}
+
+define i1 @test_fcmp_select_var_const_fmf(double %x, double %y) {
+; CHECK-LABEL: @test_fcmp_select_var_const_fmf(
+; CHECK-NEXT:    [[CMP1:%.*]] = fcmp ule double [[X:%.*]], 0x3E80000000000000
+; CHECK-NEXT:    [[TMP1:%.*]] = fcmp nnan olt double [[Y:%.*]], 0x3E80000000000000
+; CHECK-NEXT:    [[CMP2:%.*]] = select i1 [[CMP1]], i1 true, i1 [[TMP1]]
+; CHECK-NEXT:    ret i1 [[CMP2]]
+;
+  %cmp1 = fcmp ogt double %x, 0x3E80000000000000
+  %sel = select i1 %cmp1, double %y, double 0.000000e+00
+  %cmp2 = fcmp nnan olt double %sel, 0x3E80000000000000
+  ret i1 %cmp2
+}
+
+define <2 x i1> @test_fcmp_select_const_const_vec(<2 x double> %x) {
+; CHECK-LABEL: @test_fcmp_select_const_const_vec(
+; CHECK-NEXT:    [[CMP1:%.*]] = fcmp uno <2 x double> [[X:%.*]], zeroinitializer
+; CHECK-NEXT:    ret <2 x i1> [[CMP1]]
+;
+  %cmp1 = fcmp ord <2 x double> %x, zeroinitializer
+  %sel = select <2 x i1> %cmp1, <2 x double> <double 0xFFFFFFFFFFFFFFFF, double 0xFFFFFFFFFFFFFFFF>, <2 x double> zeroinitializer
+  %cmp2 = fcmp oeq <2 x double> %sel, zeroinitializer
+  ret <2 x i1> %cmp2
+}
+
+; Don't break clamp idioms
+
+define double @test_fcmp_select_clamp(double %x) {
+; CHECK-LABEL: @test_fcmp_select_clamp(
+; CHECK-NEXT:    [[CMP1:%.*]] = fcmp ogt double [[X:%.*]], 9.000000e-01
+; CHECK-NEXT:    [[SEL1:%.*]] = select i1 [[CMP1]], double 9.000000e-01, double [[X]]
+; CHECK-NEXT:    [[CMP2:%.*]] = fcmp olt double [[SEL1]], 5.000000e-01
+; CHECK-NEXT:    [[SEL2:%.*]] = select i1 [[CMP2]], double 5.000000e-01, double [[SEL1]]
+; CHECK-NEXT:    ret double [[SEL2]]
+;
+  %cmp1 = fcmp ogt double %x, 9.000000e-01
+  %sel1 = select i1 %cmp1, double 9.000000e-01, double %x
+  %cmp2 = fcmp olt double %sel1, 5.000000e-01
+  %sel2 = select i1 %cmp2, double 5.000000e-01, double %sel1
+  ret double %sel2
+}
+
+; Don't break fmin/fmax idioms
+
+define double @test_fcmp_select_maxnum(double %x) {
+; CHECK-LABEL: @test_fcmp_select_maxnum(
+; CHECK-NEXT:    [[SEL1:%.*]] = call nnan nsz double @llvm.maxnum.f64(double [[X:%.*]], double 1.000000e+00)
+; CHECK-NEXT:    [[SEL2:%.*]] = call nnan nsz double @llvm.minnum.f64(double [[SEL1]], double 2.550000e+02)
+; CHECK-NEXT:    ret double [[SEL2]]
+;
+  %cmp1 = fcmp ogt double %x, 1.0
+  %sel1 = select nnan nsz i1 %cmp1, double %x, double 1.0
+  %cmp2 = fcmp olt double %sel1, 255.0
+  %sel2 = select nnan nsz i1 %cmp2, double %sel1, double 255.0
+  ret double %sel2
+}
+
+define i1 @test_fcmp_select_const_const_multiuse(double %x) {
+; CHECK-LABEL: @test_fcmp_select_const_const_multiuse(
+; CHECK-NEXT:    [[CMP1:%.*]] = fcmp ord double [[X:%.*]], 0.000000e+00
+; CHECK-NEXT:    [[SEL:%.*]] = select i1 [[CMP1]], double 0xFFFFFFFFFFFFFFFF, double 0.000000e+00
+; CHECK-NEXT:    call void @usef64(double [[SEL]])
+; CHECK-NEXT:    [[CMP2:%.*]] = fcmp oeq double [[SEL]], 0.000000e+00
+; CHECK-NEXT:    ret i1 [[CMP2]]
+;
+  %cmp1 = fcmp ord double %x, 0.000000e+00
+  %sel = select i1 %cmp1, double 0xFFFFFFFFFFFFFFFF, double 0.000000e+00
+  call void @usef64(double %sel)
+  %cmp2 = fcmp oeq double %sel, 0.000000e+00
+  ret i1 %cmp2
+}
+
+define i1 @test_fcmp_select_const_const_unordered(double %x) {
+; CHECK-LABEL: @test_fcmp_select_const_const_unordered(
+; CHECK-NEXT:    [[CMP1:%.*]] = fcmp ord double [[X:%.*]], 0.000000e+00
+; CHECK-NEXT:    ret i1 [[CMP1]]
+;
+  %cmp1 = fcmp uno double %x, 0.000000e+00
+  %sel = select i1 %cmp1, double 0xFFFFFFFFFFFFFFFF, double 0.000000e+00
+  %cmp2 = fcmp oeq double %sel, 0.000000e+00
+  ret i1 %cmp2
+}
+
+define i1 @test_fcmp_select_var_const_unordered(double %x, double %y) {
+; CHECK-LABEL: @test_fcmp_select_var_const_unordered(
+; CHECK-NEXT:    [[CMP1:%.*]] = fcmp ult double [[X:%.*]], 0x3E80000000000000
+; CHECK-NEXT:    [[TMP1:%.*]] = fcmp ugt double [[Y:%.*]], 0x3E80000000000000
+; CHECK-NEXT:    [[CMP2:%.*]] = select i1 [[CMP1]], i1 [[TMP1]], i1 false
+; CHECK-NEXT:    ret i1 [[CMP2]]
+;
+  %cmp1 = fcmp ult double %x, 0x3E80000000000000
+  %sel = select i1 %cmp1, double %y, double 0.000000e+00
+  %cmp2 = fcmp ugt double %sel, 0x3E80000000000000
+  ret i1 %cmp2
+}
diff --git a/llvm/test/Transforms/InstCombine/ffs-1.ll b/llvm/test/Transforms/InstCombine/ffs-1.ll
index 7cf080765bb1..db01801b1225 100644
--- a/llvm/test/Transforms/InstCombine/ffs-1.ll
+++ b/llvm/test/Transforms/InstCombine/ffs-1.ll
@@ -148,7 +148,7 @@ define i32 @test_simplify12() {
 
 define i32 @test_simplify13(i32 %x) {
 ; ALL-LABEL: @test_simplify13(
-; ALL-NEXT:    [[CTTZ:%.*]] = call i32 @llvm.cttz.i32(i32 %x, i1 true), !range !0
+; ALL-NEXT:    [[CTTZ:%.*]] = call range(i32 0, 33) i32 @llvm.cttz.i32(i32 %x, i1 true)
 ; ALL-NEXT:    [[TMP1:%.*]] = add nuw nsw i32 [[CTTZ]], 1
 ; ALL-NEXT:    [[TMP2:%.*]] = icmp eq i32 %x, 0
 ; ALL-NEXT:    [[TMP3:%.*]] = select i1 [[TMP2]], i32 0, i32 [[TMP1]]
@@ -164,7 +164,7 @@ define i32 @test_simplify14(i32 %x) {
 ; GENERIC-NEXT:    ret i32 [[RET]]
 ;
 ; TARGET-LABEL: @test_simplify14(
-; TARGET-NEXT:    [[CTTZ:%.*]] = call i32 @llvm.cttz.i32(i32 %x, i1 true), !range !0
+; TARGET-NEXT:    [[CTTZ:%.*]] = call range(i32 0, 33) i32 @llvm.cttz.i32(i32 %x, i1 true)
 ; TARGET-NEXT:    [[TMP1:%.*]] = add nuw nsw i32 [[CTTZ]], 1
 ; TARGET-NEXT:    [[TMP2:%.*]] = icmp eq i32 %x, 0
 ; TARGET-NEXT:    [[TMP3:%.*]] = select i1 [[TMP2]], i32 0, i32 [[TMP1]]
@@ -180,7 +180,7 @@ define i32 @test_simplify15(i64 %x) {
 ; GENERIC-NEXT:    ret i32 [[RET]]
 ;
 ; TARGET-LABEL: @test_simplify15(
-; TARGET-NEXT:    [[CTTZ:%.*]] = call i64 @llvm.cttz.i64(i64 %x, i1 true), !range !1
+; TARGET-NEXT:    [[CTTZ:%.*]] = call range(i64 0, 65) i64 @llvm.cttz.i64(i64 %x, i1 true)
 ; TARGET-NEXT:    [[TMP1:%.*]] = trunc nuw nsw i64 [[CTTZ]] to i32
 ; TARGET-NEXT:    [[TMP2:%.*]] = add nuw nsw i32 [[TMP1]], 1
 ; TARGET-NEXT:    [[TMP3:%.*]] = icmp eq i64 %x, 0
diff --git a/llvm/test/Transforms/InstCombine/ffs-i16.ll b/llvm/test/Transforms/InstCombine/ffs-i16.ll
index f2b1f074f86d..ab5f1307f4d9 100644
--- a/llvm/test/Transforms/InstCombine/ffs-i16.ll
+++ b/llvm/test/Transforms/InstCombine/ffs-i16.ll
@@ -17,7 +17,7 @@ define void @fold_ffs(i16 %x) {
 ; AVR-LABEL: @fold_ffs(
 ; AVR-NEXT:    call addrspace(1) void @sink(i16 0)
 ; AVR-NEXT:    call addrspace(1) void @sink(i16 1)
-; AVR-NEXT:    [[CTTZ:%.*]] = call addrspace(1) i16 @llvm.cttz.i16(i16 [[X:%.*]], i1 true), !range [[RNG0:![0-9]+]]
+; AVR-NEXT:    [[CTTZ:%.*]] = call range(i16 0, 17) addrspace(1) i16 @llvm.cttz.i16(i16 [[X:%.*]], i1 true)
 ; AVR-NEXT:    [[TMP1:%.*]] = add nuw nsw i16 [[CTTZ]], 1
 ; AVR-NEXT:    [[DOTNOT:%.*]] = icmp eq i16 [[X]], 0
 ; AVR-NEXT:    [[NX:%.*]] = select i1 [[DOTNOT]], i16 0, i16 [[TMP1]]
@@ -27,7 +27,7 @@ define void @fold_ffs(i16 %x) {
 ; MSP430-LABEL: @fold_ffs(
 ; MSP430-NEXT:    call void @sink(i16 0)
 ; MSP430-NEXT:    call void @sink(i16 1)
-; MSP430-NEXT:    [[CTTZ:%.*]] = call i16 @llvm.cttz.i16(i16 [[X:%.*]], i1 true), !range [[RNG0:![0-9]+]]
+; MSP430-NEXT:    [[CTTZ:%.*]] = call range(i16 0, 17) i16 @llvm.cttz.i16(i16 [[X:%.*]], i1 true)
 ; MSP430-NEXT:    [[TMP1:%.*]] = add nuw nsw i16 [[CTTZ]], 1
 ; MSP430-NEXT:    [[DOTNOT:%.*]] = icmp eq i16 [[X]], 0
 ; MSP430-NEXT:    [[NX:%.*]] = select i1 [[DOTNOT]], i16 0, i16 [[TMP1]]
diff --git a/llvm/test/Transforms/InstCombine/fls-i16.ll b/llvm/test/Transforms/InstCombine/fls-i16.ll
index 54692fde5303..e48397f5116a 100644
--- a/llvm/test/Transforms/InstCombine/fls-i16.ll
+++ b/llvm/test/Transforms/InstCombine/fls-i16.ll
@@ -17,7 +17,7 @@ define void @fold_fls(i16 %x) {
 ; AVR-LABEL: @fold_fls(
 ; AVR-NEXT:    call addrspace(1) void @sink(i16 0)
 ; AVR-NEXT:    call addrspace(1) void @sink(i16 1)
-; AVR-NEXT:    [[CTLZ:%.*]] = call addrspace(1) i16 @llvm.ctlz.i16(i16 [[X:%.*]], i1 false), !range [[RNG0:![0-9]+]]
+; AVR-NEXT:    [[CTLZ:%.*]] = call range(i16 0, 17) addrspace(1) i16 @llvm.ctlz.i16(i16 [[X:%.*]], i1 false)
 ; AVR-NEXT:    [[NX:%.*]] = sub nuw nsw i16 16, [[CTLZ]]
 ; AVR-NEXT:    call addrspace(1) void @sink(i16 [[NX]])
 ; AVR-NEXT:    ret void
@@ -25,7 +25,7 @@ define void @fold_fls(i16 %x) {
 ; MSP430-LABEL: @fold_fls(
 ; MSP430-NEXT:    call void @sink(i16 0)
 ; MSP430-NEXT:    call void @sink(i16 1)
-; MSP430-NEXT:    [[CTLZ:%.*]] = call i16 @llvm.ctlz.i16(i16 [[X:%.*]], i1 false), !range [[RNG0:![0-9]+]]
+; MSP430-NEXT:    [[CTLZ:%.*]] = call range(i16 0, 17) i16 @llvm.ctlz.i16(i16 [[X:%.*]], i1 false)
 ; MSP430-NEXT:    [[NX:%.*]] = sub nuw nsw i16 16, [[CTLZ]]
 ; MSP430-NEXT:    call void @sink(i16 [[NX]])
 ; MSP430-NEXT:    ret void
diff --git a/llvm/test/Transforms/InstCombine/fls.ll b/llvm/test/Transforms/InstCombine/fls.ll
index 7710093e195a..68bc0a2fc8a1 100644
--- a/llvm/test/Transforms/InstCombine/fls.ll
+++ b/llvm/test/Transforms/InstCombine/fls.ll
@@ -31,7 +31,7 @@ define i32 @myflsll() {
 
 define i32 @flsnotconst(i64 %z) {
 ; CHECK-LABEL: @flsnotconst(
-; CHECK-NEXT:    [[CTLZ:%.*]] = call i64 @llvm.ctlz.i64(i64 [[Z:%.*]], i1 false), !range [[RNG0:![0-9]+]]
+; CHECK-NEXT:    [[CTLZ:%.*]] = call range(i64 0, 65) i64 @llvm.ctlz.i64(i64 [[Z:%.*]], i1 false)
 ; CHECK-NEXT:    [[TMP1:%.*]] = trunc nuw nsw i64 [[CTLZ]] to i32
 ; CHECK-NEXT:    [[GOO:%.*]] = sub nsw i32 64, [[TMP1]]
 ; CHECK-NEXT:    ret i32 [[GOO]]
diff --git a/llvm/test/Transforms/InstCombine/fneg.ll b/llvm/test/Transforms/InstCombine/fneg.ll
index 7c9289c44711..3c4088832fea 100644
--- a/llvm/test/Transforms/InstCombine/fneg.ll
+++ b/llvm/test/Transforms/InstCombine/fneg.ll
@@ -980,7 +980,7 @@ define float @fneg_ldexp_contract(float %x, i32 %n) {
 define float @fneg_ldexp_metadata(float %x, i32 %n) {
 ; CHECK-LABEL: @fneg_ldexp_metadata(
 ; CHECK-NEXT:    [[TMP1:%.*]] = fneg float [[X:%.*]]
-; CHECK-NEXT:    [[NEG:%.*]] = call float @llvm.ldexp.f32.i32(float [[TMP1]], i32 [[N:%.*]]), !arst !0
+; CHECK-NEXT:    [[NEG:%.*]] = call float @llvm.ldexp.f32.i32(float [[TMP1]], i32 [[N:%.*]]), !arst [[META0:![0-9]+]]
 ; CHECK-NEXT:    ret float [[NEG]]
 ;
   %ldexp = call float @llvm.ldexp.f32.i32(float %x, i32 %n), !arst !0
@@ -988,4 +988,125 @@ define float @fneg_ldexp_metadata(float %x, i32 %n) {
   ret float %neg
 }
 
+define float @test_fneg_select_constants(i1 %cond) {
+; CHECK-LABEL: @test_fneg_select_constants(
+; CHECK-NEXT:    [[NEG:%.*]] = select i1 [[COND:%.*]], float -0.000000e+00, float 0.000000e+00
+; CHECK-NEXT:    ret float [[NEG]]
+;
+  %sel1 = select i1 %cond, float 0.0, float -0.0
+  %neg = fneg float %sel1
+  ret float %neg
+}
+
+define <2 x float> @test_fneg_vec(<2 x i1> %cond) {
+; CHECK-LABEL: @test_fneg_vec(
+; CHECK-NEXT:    [[NEG:%.*]] = select <2 x i1> [[COND:%.*]], <2 x float> <float -0.000000e+00, float 0.000000e+00>, <2 x float> <float 0.000000e+00, float -0.000000e+00>
+; CHECK-NEXT:    ret <2 x float> [[NEG]]
+;
+  %sel1 = select <2 x i1> %cond, <2 x float> <float 0.0, float -0.0>, <2 x float> <float -0.0, float 0.0>
+  %neg = fneg <2 x float> %sel1
+  ret <2 x float> %neg
+}
+
+define float @test_fneg_select_var_constant(i1 %cond, float %x) {
+; CHECK-LABEL: @test_fneg_select_var_constant(
+; CHECK-NEXT:    [[X_NEG:%.*]] = fneg float [[X:%.*]]
+; CHECK-NEXT:    [[NEG:%.*]] = select i1 [[COND:%.*]], float [[X_NEG]], float 0.000000e+00
+; CHECK-NEXT:    ret float [[NEG]]
+;
+  %sel1 = select i1 %cond, float %x, float -0.0
+  %neg = fneg float %sel1
+  ret float %neg
+}
+
+; nsz can be preserved.
+
+define float @test_fneg_select_var_constant_fmf1(i1 %cond, float %x) {
+; CHECK-LABEL: @test_fneg_select_var_constant_fmf1(
+; CHECK-NEXT:    [[X_NEG:%.*]] = fneg float [[X:%.*]]
+; CHECK-NEXT:    [[NEG:%.*]] = select nnan ninf nsz i1 [[COND:%.*]], float [[X_NEG]], float -1.000000e+00
+; CHECK-NEXT:    ret float [[NEG]]
+;
+  %sel1 = select nnan ninf nsz i1 %cond, float %x, float 1.0
+  %neg = fneg float %sel1
+  ret float %neg
+}
+
+define float @test_fneg_select_var_constant_fmf2(i1 %cond, float %x) {
+; CHECK-LABEL: @test_fneg_select_var_constant_fmf2(
+; CHECK-NEXT:    [[X_NEG:%.*]] = fneg nnan ninf nsz float [[X:%.*]]
+; CHECK-NEXT:    [[NEG:%.*]] = select nnan ninf nsz i1 [[COND:%.*]], float [[X_NEG]], float -1.000000e+00
+; CHECK-NEXT:    ret float [[NEG]]
+;
+  %sel1 = select i1 %cond, float %x, float 1.0
+  %neg = fneg nnan ninf nsz float %sel1
+  ret float %neg
+}
+
+define float @test_fneg_select_constant_var(i1 %cond, float %x) {
+; CHECK-LABEL: @test_fneg_select_constant_var(
+; CHECK-NEXT:    [[X_NEG:%.*]] = fneg float [[X:%.*]]
+; CHECK-NEXT:    [[NEG:%.*]] = select i1 [[COND:%.*]], float -0.000000e+00, float [[X_NEG]]
+; CHECK-NEXT:    ret float [[NEG]]
+;
+  %sel1 = select i1 %cond, float 0.0, float %x
+  %neg = fneg float %sel1
+  ret float %neg
+}
+
+; Make sure nabs is generated.
+
+define float @test_fneg_select_abs(i1 %cond, float %x) {
+; CHECK-LABEL: @test_fneg_select_abs(
+; CHECK-NEXT:    [[ABSX:%.*]] = call float @llvm.fabs.f32(float [[X:%.*]])
+; CHECK-NEXT:    [[ABSX_NEG:%.*]] = fneg float [[ABSX]]
+; CHECK-NEXT:    [[NEG:%.*]] = select i1 [[COND:%.*]], float -0.000000e+00, float [[ABSX_NEG]]
+; CHECK-NEXT:    ret float [[NEG]]
+;
+  %absx = call float @llvm.fabs.f32(float %x)
+  %sel1 = select i1 %cond, float 0.0, float %absx
+  %neg = fneg float %sel1
+  ret float %neg
+}
+
+define float @test_fneg_fabs_select(i1 %cond, float %x) {
+; CHECK-LABEL: @test_fneg_fabs_select(
+; CHECK-NEXT:    [[TMP1:%.*]] = call float @llvm.fabs.f32(float [[X:%.*]])
+; CHECK-NEXT:    [[DOTNEG:%.*]] = fneg float [[TMP1]]
+; CHECK-NEXT:    [[NEG:%.*]] = select i1 [[COND:%.*]], float -0.000000e+00, float [[DOTNEG]]
+; CHECK-NEXT:    ret float [[NEG]]
+;
+  %sel1 = select i1 %cond, float 0.0, float %x
+  %abs = call float @llvm.fabs.f32(float %sel1)
+  %neg = fneg float %abs
+  ret float %neg
+}
+
+define float @test_fneg_select_constant_var_multiuse(i1 %cond, float %x) {
+; CHECK-LABEL: @test_fneg_select_constant_var_multiuse(
+; CHECK-NEXT:    [[SEL1:%.*]] = select i1 [[COND:%.*]], float 0.000000e+00, float [[X:%.*]]
+; CHECK-NEXT:    call void @use(float [[SEL1]])
+; CHECK-NEXT:    [[NEG:%.*]] = fneg float [[SEL1]]
+; CHECK-NEXT:    ret float [[NEG]]
+;
+  %sel1 = select i1 %cond, float 0.0, float %x
+  call void @use(float %sel1)
+  %neg = fneg float %sel1
+  ret float %neg
+}
+
+; Don't break fmax idioms.
+
+define float @test_fneg_select_maxnum(float %x) {
+; CHECK-LABEL: @test_fneg_select_maxnum(
+; CHECK-NEXT:    [[SEL1:%.*]] = call nnan nsz float @llvm.maxnum.f32(float [[X:%.*]], float 1.000000e+00)
+; CHECK-NEXT:    [[NEG:%.*]] = fneg float [[SEL1]]
+; CHECK-NEXT:    ret float [[NEG]]
+;
+  %cmp1 = fcmp ogt float %x, 1.0
+  %sel1 = select nnan nsz i1 %cmp1, float %x, float 1.0
+  %neg = fneg float %sel1
+  ret float %neg
+}
+
 !0 = !{}
diff --git a/llvm/test/Transforms/InstCombine/fold-ctpop-of-not.ll b/llvm/test/Transforms/InstCombine/fold-ctpop-of-not.ll
index bbce5c2d625e..4626d19bd289 100644
--- a/llvm/test/Transforms/InstCombine/fold-ctpop-of-not.ll
+++ b/llvm/test/Transforms/InstCombine/fold-ctpop-of-not.ll
@@ -8,7 +8,7 @@ declare <2 x i8> @llvm.ctpop.v2i8(<2 x i8>)
 
 define i8 @fold_sub_c_ctpop(i8 %x) {
 ; CHECK-LABEL: @fold_sub_c_ctpop(
-; CHECK-NEXT:    [[TMP1:%.*]] = call i8 @llvm.ctpop.i8(i8 [[X:%.*]]), !range [[RNG0:![0-9]+]]
+; CHECK-NEXT:    [[TMP1:%.*]] = call range(i8 0, 9) i8 @llvm.ctpop.i8(i8 [[X:%.*]])
 ; CHECK-NEXT:    [[R:%.*]] = add nuw nsw i8 [[TMP1]], 4
 ; CHECK-NEXT:    ret i8 [[R]]
 ;
@@ -21,7 +21,7 @@ define i8 @fold_sub_c_ctpop(i8 %x) {
 define i8 @fold_sub_var_ctpop_fail(i8 %x, i8 %y) {
 ; CHECK-LABEL: @fold_sub_var_ctpop_fail(
 ; CHECK-NEXT:    [[NX:%.*]] = xor i8 [[X:%.*]], -1
-; CHECK-NEXT:    [[CNT:%.*]] = call i8 @llvm.ctpop.i8(i8 [[NX]]), !range [[RNG0]]
+; CHECK-NEXT:    [[CNT:%.*]] = call range(i8 0, 9) i8 @llvm.ctpop.i8(i8 [[NX]])
 ; CHECK-NEXT:    [[R:%.*]] = sub i8 [[Y:%.*]], [[CNT]]
 ; CHECK-NEXT:    ret i8 [[R]]
 ;
@@ -33,7 +33,7 @@ define i8 @fold_sub_var_ctpop_fail(i8 %x, i8 %y) {
 
 define <2 x i8> @fold_sub_ctpop_c(<2 x i8> %x) {
 ; CHECK-LABEL: @fold_sub_ctpop_c(
-; CHECK-NEXT:    [[TMP1:%.*]] = call <2 x i8> @llvm.ctpop.v2i8(<2 x i8> [[X:%.*]]), !range [[RNG0]]
+; CHECK-NEXT:    [[TMP1:%.*]] = call range(i8 0, 9) <2 x i8> @llvm.ctpop.v2i8(<2 x i8> [[X:%.*]])
 ; CHECK-NEXT:    [[R:%.*]] = sub nuw nsw <2 x i8> <i8 -55, i8 -56>, [[TMP1]]
 ; CHECK-NEXT:    ret <2 x i8> [[R]]
 ;
@@ -45,7 +45,7 @@ define <2 x i8> @fold_sub_ctpop_c(<2 x i8> %x) {
 
 define i8 @fold_add_ctpop_c(i8 %x) {
 ; CHECK-LABEL: @fold_add_ctpop_c(
-; CHECK-NEXT:    [[TMP1:%.*]] = call i8 @llvm.ctpop.i8(i8 [[X:%.*]]), !range [[RNG0]]
+; CHECK-NEXT:    [[TMP1:%.*]] = call range(i8 0, 9) i8 @llvm.ctpop.i8(i8 [[X:%.*]])
 ; CHECK-NEXT:    [[R:%.*]] = sub nuw nsw i8 71, [[TMP1]]
 ; CHECK-NEXT:    ret i8 [[R]]
 ;
@@ -57,7 +57,7 @@ define i8 @fold_add_ctpop_c(i8 %x) {
 
 define i8 @fold_distjoint_or_ctpop_c(i8 %x) {
 ; CHECK-LABEL: @fold_distjoint_or_ctpop_c(
-; CHECK-NEXT:    [[TMP1:%.*]] = call i8 @llvm.ctpop.i8(i8 [[X:%.*]]), !range [[RNG0]]
+; CHECK-NEXT:    [[TMP1:%.*]] = call range(i8 0, 9) i8 @llvm.ctpop.i8(i8 [[X:%.*]])
 ; CHECK-NEXT:    [[R:%.*]] = sub nuw nsw i8 72, [[TMP1]]
 ; CHECK-NEXT:    ret i8 [[R]]
 ;
@@ -70,7 +70,7 @@ define i8 @fold_distjoint_or_ctpop_c(i8 %x) {
 define i8 @fold_or_ctpop_c_fail(i8 %x) {
 ; CHECK-LABEL: @fold_or_ctpop_c_fail(
 ; CHECK-NEXT:    [[NX:%.*]] = xor i8 [[X:%.*]], -1
-; CHECK-NEXT:    [[CNT:%.*]] = call i8 @llvm.ctpop.i8(i8 [[NX]]), !range [[RNG0]]
+; CHECK-NEXT:    [[CNT:%.*]] = call range(i8 0, 9) i8 @llvm.ctpop.i8(i8 [[NX]])
 ; CHECK-NEXT:    [[R:%.*]] = or i8 [[CNT]], 65
 ; CHECK-NEXT:    ret i8 [[R]]
 ;
@@ -83,7 +83,7 @@ define i8 @fold_or_ctpop_c_fail(i8 %x) {
 define i8 @fold_add_ctpop_var_fail(i8 %x, i8 %y) {
 ; CHECK-LABEL: @fold_add_ctpop_var_fail(
 ; CHECK-NEXT:    [[NX:%.*]] = xor i8 [[X:%.*]], -1
-; CHECK-NEXT:    [[CNT:%.*]] = call i8 @llvm.ctpop.i8(i8 [[NX]]), !range [[RNG0]]
+; CHECK-NEXT:    [[CNT:%.*]] = call range(i8 0, 9) i8 @llvm.ctpop.i8(i8 [[NX]])
 ; CHECK-NEXT:    [[R:%.*]] = add i8 [[CNT]], [[Y:%.*]]
 ; CHECK-NEXT:    ret i8 [[R]]
 ;
@@ -105,7 +105,7 @@ define i1 @fold_icmp_sgt_ctpop_c_i2_fail(i2 %x, i2 %C) {
 
 define i1 @fold_cmp_eq_ctpop_c(i8 %x) {
 ; CHECK-LABEL: @fold_cmp_eq_ctpop_c(
-; CHECK-NEXT:    [[TMP1:%.*]] = call i8 @llvm.ctpop.i8(i8 [[X:%.*]]), !range [[RNG0]]
+; CHECK-NEXT:    [[TMP1:%.*]] = call range(i8 0, 9) i8 @llvm.ctpop.i8(i8 [[X:%.*]])
 ; CHECK-NEXT:    [[R:%.*]] = icmp eq i8 [[TMP1]], 6
 ; CHECK-NEXT:    ret i1 [[R]]
 ;
@@ -118,7 +118,7 @@ define i1 @fold_cmp_eq_ctpop_c(i8 %x) {
 define i1 @fold_cmp_eq_ctpop_c_multiuse_fail(i8 %x) {
 ; CHECK-LABEL: @fold_cmp_eq_ctpop_c_multiuse_fail(
 ; CHECK-NEXT:    [[NX:%.*]] = xor i8 [[X:%.*]], -1
-; CHECK-NEXT:    [[CNT:%.*]] = call i8 @llvm.ctpop.i8(i8 [[NX]]), !range [[RNG0]]
+; CHECK-NEXT:    [[CNT:%.*]] = call range(i8 0, 9) i8 @llvm.ctpop.i8(i8 [[NX]])
 ; CHECK-NEXT:    [[R:%.*]] = icmp eq i8 [[CNT]], 2
 ; CHECK-NEXT:    call void @use.i8(i8 [[CNT]])
 ; CHECK-NEXT:    ret i1 [[R]]
@@ -132,7 +132,7 @@ define i1 @fold_cmp_eq_ctpop_c_multiuse_fail(i8 %x) {
 
 define <2 x i1> @fold_cmp_ne_ctpop_c(<2 x i8> %x) {
 ; CHECK-LABEL: @fold_cmp_ne_ctpop_c(
-; CHECK-NEXT:    [[TMP1:%.*]] = call <2 x i8> @llvm.ctpop.v2i8(<2 x i8> [[X:%.*]]), !range [[RNG0]]
+; CHECK-NEXT:    [[TMP1:%.*]] = call range(i8 0, 9) <2 x i8> @llvm.ctpop.v2i8(<2 x i8> [[X:%.*]])
 ; CHECK-NEXT:    [[R:%.*]] = icmp ne <2 x i8> [[TMP1]], <i8 -36, i8 5>
 ; CHECK-NEXT:    ret <2 x i1> [[R]]
 ;
@@ -145,7 +145,7 @@ define <2 x i1> @fold_cmp_ne_ctpop_c(<2 x i8> %x) {
 define <2 x i1> @fold_cmp_ne_ctpop_var_fail(<2 x i8> %x, <2 x i8> %y) {
 ; CHECK-LABEL: @fold_cmp_ne_ctpop_var_fail(
 ; CHECK-NEXT:    [[NX:%.*]] = xor <2 x i8> [[X:%.*]], <i8 -1, i8 -1>
-; CHECK-NEXT:    [[CNT:%.*]] = call <2 x i8> @llvm.ctpop.v2i8(<2 x i8> [[NX]]), !range [[RNG0]]
+; CHECK-NEXT:    [[CNT:%.*]] = call range(i8 0, 9) <2 x i8> @llvm.ctpop.v2i8(<2 x i8> [[NX]])
 ; CHECK-NEXT:    [[R:%.*]] = icmp ne <2 x i8> [[CNT]], [[Y:%.*]]
 ; CHECK-NEXT:    ret <2 x i1> [[R]]
 ;
@@ -159,7 +159,7 @@ define i1 @fold_cmp_ult_ctpop_c(i8 %x, i8 %y, i1 %cond) {
 ; CHECK-LABEL: @fold_cmp_ult_ctpop_c(
 ; CHECK-NEXT:    [[TMP1:%.*]] = sub i8 -16, [[Y:%.*]]
 ; CHECK-NEXT:    [[TMP2:%.*]] = select i1 [[COND:%.*]], i8 [[X:%.*]], i8 [[TMP1]]
-; CHECK-NEXT:    [[TMP3:%.*]] = call i8 @llvm.ctpop.i8(i8 [[TMP2]]), !range [[RNG0]]
+; CHECK-NEXT:    [[TMP3:%.*]] = call range(i8 0, 9) i8 @llvm.ctpop.i8(i8 [[TMP2]])
 ; CHECK-NEXT:    [[R:%.*]] = icmp ugt i8 [[TMP3]], 3
 ; CHECK-NEXT:    ret i1 [[R]]
 ;
@@ -175,7 +175,7 @@ define i1 @fold_cmp_sle_ctpop_c(i8 %x, i8 %y, i1 %cond) {
 ; CHECK-LABEL: @fold_cmp_sle_ctpop_c(
 ; CHECK-NEXT:    [[TMP1:%.*]] = sub i8 -16, [[Y:%.*]]
 ; CHECK-NEXT:    [[TMP2:%.*]] = select i1 [[COND:%.*]], i8 [[X:%.*]], i8 [[TMP1]]
-; CHECK-NEXT:    [[TMP3:%.*]] = call i8 @llvm.ctpop.i8(i8 [[TMP2]]), !range [[RNG0]]
+; CHECK-NEXT:    [[TMP3:%.*]] = call range(i8 0, 9) i8 @llvm.ctpop.i8(i8 [[TMP2]])
 ; CHECK-NEXT:    [[R:%.*]] = icmp ugt i8 [[TMP3]], 4
 ; CHECK-NEXT:    ret i1 [[R]]
 ;
@@ -190,7 +190,7 @@ define i1 @fold_cmp_sle_ctpop_c(i8 %x, i8 %y, i1 %cond) {
 define i1 @fold_cmp_ult_ctpop_c_no_not_inst_save_fail(i8 %x) {
 ; CHECK-LABEL: @fold_cmp_ult_ctpop_c_no_not_inst_save_fail(
 ; CHECK-NEXT:    [[NX:%.*]] = xor i8 [[X:%.*]], -2
-; CHECK-NEXT:    [[CNT:%.*]] = call i8 @llvm.ctpop.i8(i8 [[NX]]), !range [[RNG0]]
+; CHECK-NEXT:    [[CNT:%.*]] = call range(i8 0, 9) i8 @llvm.ctpop.i8(i8 [[NX]])
 ; CHECK-NEXT:    [[R:%.*]] = icmp ult i8 [[CNT]], 5
 ; CHECK-NEXT:    ret i1 [[R]]
 ;
@@ -202,7 +202,7 @@ define i1 @fold_cmp_ult_ctpop_c_no_not_inst_save_fail(i8 %x) {
 
 define <2 x i1> @fold_cmp_ugt_ctpop_c(<2 x i8> %x) {
 ; CHECK-LABEL: @fold_cmp_ugt_ctpop_c(
-; CHECK-NEXT:    [[TMP1:%.*]] = call <2 x i8> @llvm.ctpop.v2i8(<2 x i8> [[X:%.*]]), !range [[RNG0]]
+; CHECK-NEXT:    [[TMP1:%.*]] = call range(i8 0, 9) <2 x i8> @llvm.ctpop.v2i8(<2 x i8> [[X:%.*]])
 ; CHECK-NEXT:    [[R:%.*]] = icmp ult <2 x i8> [[TMP1]], <i8 0, i8 2>
 ; CHECK-NEXT:    ret <2 x i1> [[R]]
 ;
@@ -215,7 +215,7 @@ define <2 x i1> @fold_cmp_ugt_ctpop_c(<2 x i8> %x) {
 define <2 x i1> @fold_cmp_ugt_ctpop_c_out_of_range_fail(<2 x i8> %x) {
 ; CHECK-LABEL: @fold_cmp_ugt_ctpop_c_out_of_range_fail(
 ; CHECK-NEXT:    [[NX:%.*]] = xor <2 x i8> [[X:%.*]], <i8 -1, i8 -1>
-; CHECK-NEXT:    [[CNT:%.*]] = call <2 x i8> @llvm.ctpop.v2i8(<2 x i8> [[NX]]), !range [[RNG0]]
+; CHECK-NEXT:    [[CNT:%.*]] = call range(i8 0, 9) <2 x i8> @llvm.ctpop.v2i8(<2 x i8> [[NX]])
 ; CHECK-NEXT:    [[R:%.*]] = icmp ugt <2 x i8> [[CNT]], <i8 2, i8 10>
 ; CHECK-NEXT:    ret <2 x i1> [[R]]
 ;
diff --git a/llvm/test/Transforms/InstCombine/fold-log2-ceil-idiom.ll b/llvm/test/Transforms/InstCombine/fold-log2-ceil-idiom.ll
index a631aacd97ff..17e51e73201b 100644
--- a/llvm/test/Transforms/InstCombine/fold-log2-ceil-idiom.ll
+++ b/llvm/test/Transforms/InstCombine/fold-log2-ceil-idiom.ll
@@ -5,7 +5,7 @@ define i32 @log2_ceil_idiom(i32 %x) {
 ; CHECK-LABEL: define i32 @log2_ceil_idiom(
 ; CHECK-SAME: i32 [[X:%.*]]) {
 ; CHECK-NEXT:    [[TMP1:%.*]] = add i32 [[X]], -1
-; CHECK-NEXT:    [[TMP2:%.*]] = call i32 @llvm.ctlz.i32(i32 [[TMP1]], i1 false), !range [[RNG0:![0-9]+]]
+; CHECK-NEXT:    [[TMP2:%.*]] = call range(i32 0, 33) i32 @llvm.ctlz.i32(i32 [[TMP1]], i1 false)
 ; CHECK-NEXT:    [[RET:%.*]] = sub nuw nsw i32 32, [[TMP2]]
 ; CHECK-NEXT:    ret i32 [[RET]]
 ;
@@ -22,7 +22,7 @@ define i5 @log2_ceil_idiom_trunc(i32 %x) {
 ; CHECK-LABEL: define i5 @log2_ceil_idiom_trunc(
 ; CHECK-SAME: i32 [[X:%.*]]) {
 ; CHECK-NEXT:    [[TMP1:%.*]] = add i32 [[X]], -1
-; CHECK-NEXT:    [[TMP2:%.*]] = call i32 @llvm.ctlz.i32(i32 [[TMP1]], i1 false), !range [[RNG0]]
+; CHECK-NEXT:    [[TMP2:%.*]] = call range(i32 0, 33) i32 @llvm.ctlz.i32(i32 [[TMP1]], i1 false)
 ; CHECK-NEXT:    [[TMP3:%.*]] = sub nsw i32 0, [[TMP2]]
 ; CHECK-NEXT:    [[RET:%.*]] = trunc i32 [[TMP3]] to i5
 ; CHECK-NEXT:    ret i5 [[RET]]
@@ -41,7 +41,7 @@ define i64 @log2_ceil_idiom_zext(i32 %x) {
 ; CHECK-LABEL: define i64 @log2_ceil_idiom_zext(
 ; CHECK-SAME: i32 [[X:%.*]]) {
 ; CHECK-NEXT:    [[TMP1:%.*]] = add i32 [[X]], -1
-; CHECK-NEXT:    [[TMP2:%.*]] = call i32 @llvm.ctlz.i32(i32 [[TMP1]], i1 false), !range [[RNG0]]
+; CHECK-NEXT:    [[TMP2:%.*]] = call range(i32 0, 33) i32 @llvm.ctlz.i32(i32 [[TMP1]], i1 false)
 ; CHECK-NEXT:    [[TMP3:%.*]] = sub nuw nsw i32 32, [[TMP2]]
 ; CHECK-NEXT:    [[RET:%.*]] = zext nneg i32 [[TMP3]] to i64
 ; CHECK-NEXT:    ret i64 [[RET]]
@@ -60,7 +60,7 @@ define i32 @log2_ceil_idiom_power2_test2(i32 %x) {
 ; CHECK-LABEL: define i32 @log2_ceil_idiom_power2_test2(
 ; CHECK-SAME: i32 [[X:%.*]]) {
 ; CHECK-NEXT:    [[TMP1:%.*]] = add i32 [[X]], -1
-; CHECK-NEXT:    [[TMP2:%.*]] = call i32 @llvm.ctlz.i32(i32 [[TMP1]], i1 false), !range [[RNG0]]
+; CHECK-NEXT:    [[TMP2:%.*]] = call range(i32 0, 33) i32 @llvm.ctlz.i32(i32 [[TMP1]], i1 false)
 ; CHECK-NEXT:    [[RET:%.*]] = sub nuw nsw i32 32, [[TMP2]]
 ; CHECK-NEXT:    ret i32 [[RET]]
 ;
@@ -77,7 +77,7 @@ define i32 @log2_ceil_idiom_commuted(i32 %x) {
 ; CHECK-LABEL: define i32 @log2_ceil_idiom_commuted(
 ; CHECK-SAME: i32 [[X:%.*]]) {
 ; CHECK-NEXT:    [[TMP1:%.*]] = add i32 [[X]], -1
-; CHECK-NEXT:    [[TMP2:%.*]] = call i32 @llvm.ctlz.i32(i32 [[TMP1]], i1 false), !range [[RNG0]]
+; CHECK-NEXT:    [[TMP2:%.*]] = call range(i32 0, 33) i32 @llvm.ctlz.i32(i32 [[TMP1]], i1 false)
 ; CHECK-NEXT:    [[RET:%.*]] = sub nuw nsw i32 32, [[TMP2]]
 ; CHECK-NEXT:    ret i32 [[RET]]
 ;
@@ -93,10 +93,10 @@ define i32 @log2_ceil_idiom_commuted(i32 %x) {
 define i32 @log2_ceil_idiom_multiuse1(i32 %x) {
 ; CHECK-LABEL: define i32 @log2_ceil_idiom_multiuse1(
 ; CHECK-SAME: i32 [[X:%.*]]) {
-; CHECK-NEXT:    [[CTPOP:%.*]] = tail call i32 @llvm.ctpop.i32(i32 [[X]]), !range [[RNG0]]
+; CHECK-NEXT:    [[CTPOP:%.*]] = tail call range(i32 0, 33) i32 @llvm.ctpop.i32(i32 [[X]])
 ; CHECK-NEXT:    call void @use32(i32 [[CTPOP]])
 ; CHECK-NEXT:    [[TMP1:%.*]] = add i32 [[X]], -1
-; CHECK-NEXT:    [[TMP2:%.*]] = call i32 @llvm.ctlz.i32(i32 [[TMP1]], i1 false), !range [[RNG0]]
+; CHECK-NEXT:    [[TMP2:%.*]] = call range(i32 0, 33) i32 @llvm.ctlz.i32(i32 [[TMP1]], i1 false)
 ; CHECK-NEXT:    [[RET:%.*]] = sub nuw nsw i32 32, [[TMP2]]
 ; CHECK-NEXT:    ret i32 [[RET]]
 ;
@@ -115,9 +115,9 @@ define i32 @log2_ceil_idiom_multiuse1(i32 %x) {
 define i32 @log2_ceil_idiom_x_may_be_zero(i32 %x) {
 ; CHECK-LABEL: define i32 @log2_ceil_idiom_x_may_be_zero(
 ; CHECK-SAME: i32 [[X:%.*]]) {
-; CHECK-NEXT:    [[CTLZ:%.*]] = tail call i32 @llvm.ctlz.i32(i32 [[X]], i1 false), !range [[RNG0]]
+; CHECK-NEXT:    [[CTLZ:%.*]] = tail call range(i32 0, 33) i32 @llvm.ctlz.i32(i32 [[X]], i1 false)
 ; CHECK-NEXT:    [[XOR:%.*]] = xor i32 [[CTLZ]], 31
-; CHECK-NEXT:    [[CTPOP:%.*]] = tail call i32 @llvm.ctpop.i32(i32 [[X]]), !range [[RNG0]]
+; CHECK-NEXT:    [[CTPOP:%.*]] = tail call range(i32 0, 33) i32 @llvm.ctpop.i32(i32 [[X]])
 ; CHECK-NEXT:    [[CMP:%.*]] = icmp ugt i32 [[CTPOP]], 1
 ; CHECK-NEXT:    [[ZEXT:%.*]] = zext i1 [[CMP]] to i32
 ; CHECK-NEXT:    [[RET:%.*]] = add nuw nsw i32 [[XOR]], [[ZEXT]]
@@ -135,10 +135,10 @@ define i32 @log2_ceil_idiom_x_may_be_zero(i32 %x) {
 define i4 @log2_ceil_idiom_trunc_too_short(i32 %x) {
 ; CHECK-LABEL: define i4 @log2_ceil_idiom_trunc_too_short(
 ; CHECK-SAME: i32 [[X:%.*]]) {
-; CHECK-NEXT:    [[CTLZ:%.*]] = tail call i32 @llvm.ctlz.i32(i32 [[X]], i1 true), !range [[RNG0]]
+; CHECK-NEXT:    [[CTLZ:%.*]] = tail call range(i32 0, 33) i32 @llvm.ctlz.i32(i32 [[X]], i1 true)
 ; CHECK-NEXT:    [[TRUNC:%.*]] = trunc i32 [[CTLZ]] to i4
 ; CHECK-NEXT:    [[XOR:%.*]] = xor i4 [[TRUNC]], -1
-; CHECK-NEXT:    [[CTPOP:%.*]] = tail call i32 @llvm.ctpop.i32(i32 [[X]]), !range [[RNG0]]
+; CHECK-NEXT:    [[CTPOP:%.*]] = tail call range(i32 0, 33) i32 @llvm.ctpop.i32(i32 [[X]])
 ; CHECK-NEXT:    [[CMP:%.*]] = icmp ugt i32 [[CTPOP]], 1
 ; CHECK-NEXT:    [[ZEXT:%.*]] = zext i1 [[CMP]] to i4
 ; CHECK-NEXT:    [[RET:%.*]] = add i4 [[XOR]], [[ZEXT]]
@@ -157,9 +157,9 @@ define i4 @log2_ceil_idiom_trunc_too_short(i32 %x) {
 define i32 @log2_ceil_idiom_mismatched_operands(i32 %x, i32 %y) {
 ; CHECK-LABEL: define i32 @log2_ceil_idiom_mismatched_operands(
 ; CHECK-SAME: i32 [[X:%.*]], i32 [[Y:%.*]]) {
-; CHECK-NEXT:    [[CTLZ:%.*]] = tail call i32 @llvm.ctlz.i32(i32 [[X]], i1 true), !range [[RNG0]]
+; CHECK-NEXT:    [[CTLZ:%.*]] = tail call range(i32 0, 33) i32 @llvm.ctlz.i32(i32 [[X]], i1 true)
 ; CHECK-NEXT:    [[XOR:%.*]] = xor i32 [[CTLZ]], 31
-; CHECK-NEXT:    [[CTPOP:%.*]] = tail call i32 @llvm.ctpop.i32(i32 [[Y]]), !range [[RNG0]]
+; CHECK-NEXT:    [[CTPOP:%.*]] = tail call range(i32 0, 33) i32 @llvm.ctpop.i32(i32 [[Y]])
 ; CHECK-NEXT:    [[CMP:%.*]] = icmp ugt i32 [[CTPOP]], 1
 ; CHECK-NEXT:    [[ZEXT:%.*]] = zext i1 [[CMP]] to i32
 ; CHECK-NEXT:    [[RET:%.*]] = add nuw nsw i32 [[XOR]], [[ZEXT]]
@@ -177,9 +177,9 @@ define i32 @log2_ceil_idiom_mismatched_operands(i32 %x, i32 %y) {
 define i32 @log2_ceil_idiom_wrong_constant(i32 %x) {
 ; CHECK-LABEL: define i32 @log2_ceil_idiom_wrong_constant(
 ; CHECK-SAME: i32 [[X:%.*]]) {
-; CHECK-NEXT:    [[CTLZ:%.*]] = tail call i32 @llvm.ctlz.i32(i32 [[X]], i1 true), !range [[RNG0]]
+; CHECK-NEXT:    [[CTLZ:%.*]] = tail call range(i32 0, 33) i32 @llvm.ctlz.i32(i32 [[X]], i1 true)
 ; CHECK-NEXT:    [[XOR:%.*]] = xor i32 [[CTLZ]], 30
-; CHECK-NEXT:    [[CTPOP:%.*]] = tail call i32 @llvm.ctpop.i32(i32 [[X]]), !range [[RNG0]]
+; CHECK-NEXT:    [[CTPOP:%.*]] = tail call range(i32 0, 33) i32 @llvm.ctpop.i32(i32 [[X]])
 ; CHECK-NEXT:    [[CMP:%.*]] = icmp ugt i32 [[CTPOP]], 1
 ; CHECK-NEXT:    [[ZEXT:%.*]] = zext i1 [[CMP]] to i32
 ; CHECK-NEXT:    [[RET:%.*]] = add nuw nsw i32 [[XOR]], [[ZEXT]]
@@ -197,9 +197,9 @@ define i32 @log2_ceil_idiom_wrong_constant(i32 %x) {
 define i32 @log2_ceil_idiom_not_a_power2_test1(i32 %x) {
 ; CHECK-LABEL: define i32 @log2_ceil_idiom_not_a_power2_test1(
 ; CHECK-SAME: i32 [[X:%.*]]) {
-; CHECK-NEXT:    [[CTLZ:%.*]] = tail call i32 @llvm.ctlz.i32(i32 [[X]], i1 true), !range [[RNG0]]
+; CHECK-NEXT:    [[CTLZ:%.*]] = tail call range(i32 0, 33) i32 @llvm.ctlz.i32(i32 [[X]], i1 true)
 ; CHECK-NEXT:    [[XOR:%.*]] = xor i32 [[CTLZ]], 31
-; CHECK-NEXT:    [[CTPOP:%.*]] = tail call i32 @llvm.ctpop.i32(i32 [[X]]), !range [[RNG0]]
+; CHECK-NEXT:    [[CTPOP:%.*]] = tail call range(i32 0, 33) i32 @llvm.ctpop.i32(i32 [[X]])
 ; CHECK-NEXT:    [[CMP:%.*]] = icmp eq i32 [[CTPOP]], 1
 ; CHECK-NEXT:    [[ZEXT:%.*]] = zext i1 [[CMP]] to i32
 ; CHECK-NEXT:    [[RET:%.*]] = add nuw nsw i32 [[XOR]], [[ZEXT]]
@@ -217,9 +217,9 @@ define i32 @log2_ceil_idiom_not_a_power2_test1(i32 %x) {
 define i32 @log2_ceil_idiom_not_a_power2_test2(i32 %x) {
 ; CHECK-LABEL: define i32 @log2_ceil_idiom_not_a_power2_test2(
 ; CHECK-SAME: i32 [[X:%.*]]) {
-; CHECK-NEXT:    [[CTLZ:%.*]] = tail call i32 @llvm.ctlz.i32(i32 [[X]], i1 true), !range [[RNG0]]
+; CHECK-NEXT:    [[CTLZ:%.*]] = tail call range(i32 0, 33) i32 @llvm.ctlz.i32(i32 [[X]], i1 true)
 ; CHECK-NEXT:    [[XOR:%.*]] = xor i32 [[CTLZ]], 31
-; CHECK-NEXT:    [[CTPOP:%.*]] = tail call i32 @llvm.ctpop.i32(i32 [[X]]), !range [[RNG0]]
+; CHECK-NEXT:    [[CTPOP:%.*]] = tail call range(i32 0, 33) i32 @llvm.ctpop.i32(i32 [[X]])
 ; CHECK-NEXT:    [[CMP:%.*]] = icmp ugt i32 [[CTPOP]], 2
 ; CHECK-NEXT:    [[ZEXT:%.*]] = zext i1 [[CMP]] to i32
 ; CHECK-NEXT:    [[RET:%.*]] = add nuw nsw i32 [[XOR]], [[ZEXT]]
@@ -237,10 +237,10 @@ define i32 @log2_ceil_idiom_not_a_power2_test2(i32 %x) {
 define i32 @log2_ceil_idiom_multiuse2(i32 %x) {
 ; CHECK-LABEL: define i32 @log2_ceil_idiom_multiuse2(
 ; CHECK-SAME: i32 [[X:%.*]]) {
-; CHECK-NEXT:    [[CTLZ:%.*]] = tail call i32 @llvm.ctlz.i32(i32 [[X]], i1 true), !range [[RNG0]]
+; CHECK-NEXT:    [[CTLZ:%.*]] = tail call range(i32 0, 33) i32 @llvm.ctlz.i32(i32 [[X]], i1 true)
 ; CHECK-NEXT:    call void @use32(i32 [[CTLZ]])
 ; CHECK-NEXT:    [[XOR:%.*]] = xor i32 [[CTLZ]], 31
-; CHECK-NEXT:    [[CTPOP:%.*]] = tail call i32 @llvm.ctpop.i32(i32 [[X]]), !range [[RNG0]]
+; CHECK-NEXT:    [[CTPOP:%.*]] = tail call range(i32 0, 33) i32 @llvm.ctpop.i32(i32 [[X]])
 ; CHECK-NEXT:    [[CMP:%.*]] = icmp ugt i32 [[CTPOP]], 1
 ; CHECK-NEXT:    [[ZEXT:%.*]] = zext i1 [[CMP]] to i32
 ; CHECK-NEXT:    [[RET:%.*]] = add nuw nsw i32 [[XOR]], [[ZEXT]]
@@ -259,10 +259,10 @@ define i32 @log2_ceil_idiom_multiuse2(i32 %x) {
 define i32 @log2_ceil_idiom_multiuse3(i32 %x) {
 ; CHECK-LABEL: define i32 @log2_ceil_idiom_multiuse3(
 ; CHECK-SAME: i32 [[X:%.*]]) {
-; CHECK-NEXT:    [[CTLZ:%.*]] = tail call i32 @llvm.ctlz.i32(i32 [[X]], i1 true), !range [[RNG0]]
+; CHECK-NEXT:    [[CTLZ:%.*]] = tail call range(i32 0, 33) i32 @llvm.ctlz.i32(i32 [[X]], i1 true)
 ; CHECK-NEXT:    [[XOR:%.*]] = xor i32 [[CTLZ]], 31
 ; CHECK-NEXT:    call void @use32(i32 [[XOR]])
-; CHECK-NEXT:    [[CTPOP:%.*]] = tail call i32 @llvm.ctpop.i32(i32 [[X]]), !range [[RNG0]]
+; CHECK-NEXT:    [[CTPOP:%.*]] = tail call range(i32 0, 33) i32 @llvm.ctpop.i32(i32 [[X]])
 ; CHECK-NEXT:    [[CMP:%.*]] = icmp ugt i32 [[CTPOP]], 1
 ; CHECK-NEXT:    [[ZEXT:%.*]] = zext i1 [[CMP]] to i32
 ; CHECK-NEXT:    [[RET:%.*]] = add nuw nsw i32 [[XOR]], [[ZEXT]]
@@ -281,11 +281,11 @@ define i32 @log2_ceil_idiom_multiuse3(i32 %x) {
 define i5 @log2_ceil_idiom_trunc_multiuse4(i32 %x) {
 ; CHECK-LABEL: define i5 @log2_ceil_idiom_trunc_multiuse4(
 ; CHECK-SAME: i32 [[X:%.*]]) {
-; CHECK-NEXT:    [[CTLZ:%.*]] = tail call i32 @llvm.ctlz.i32(i32 [[X]], i1 true), !range [[RNG0]]
+; CHECK-NEXT:    [[CTLZ:%.*]] = tail call range(i32 0, 33) i32 @llvm.ctlz.i32(i32 [[X]], i1 true)
 ; CHECK-NEXT:    [[TRUNC:%.*]] = trunc nuw i32 [[CTLZ]] to i5
 ; CHECK-NEXT:    call void @use5(i5 [[TRUNC]])
 ; CHECK-NEXT:    [[XOR:%.*]] = xor i5 [[TRUNC]], -1
-; CHECK-NEXT:    [[CTPOP:%.*]] = tail call i32 @llvm.ctpop.i32(i32 [[X]]), !range [[RNG0]]
+; CHECK-NEXT:    [[CTPOP:%.*]] = tail call range(i32 0, 33) i32 @llvm.ctpop.i32(i32 [[X]])
 ; CHECK-NEXT:    [[CMP:%.*]] = icmp ugt i32 [[CTPOP]], 1
 ; CHECK-NEXT:    [[ZEXT:%.*]] = zext i1 [[CMP]] to i5
 ; CHECK-NEXT:    [[RET:%.*]] = add i5 [[XOR]], [[ZEXT]]
@@ -305,11 +305,11 @@ define i5 @log2_ceil_idiom_trunc_multiuse4(i32 %x) {
 define i64 @log2_ceil_idiom_zext_multiuse5(i32 %x) {
 ; CHECK-LABEL: define i64 @log2_ceil_idiom_zext_multiuse5(
 ; CHECK-SAME: i32 [[X:%.*]]) {
-; CHECK-NEXT:    [[CTLZ:%.*]] = tail call i32 @llvm.ctlz.i32(i32 [[X]], i1 true), !range [[RNG0]]
+; CHECK-NEXT:    [[CTLZ:%.*]] = tail call range(i32 0, 33) i32 @llvm.ctlz.i32(i32 [[X]], i1 true)
 ; CHECK-NEXT:    [[XOR:%.*]] = xor i32 [[CTLZ]], 31
 ; CHECK-NEXT:    [[EXT:%.*]] = zext nneg i32 [[XOR]] to i64
 ; CHECK-NEXT:    call void @use64(i64 [[EXT]])
-; CHECK-NEXT:    [[CTPOP:%.*]] = tail call i32 @llvm.ctpop.i32(i32 [[X]]), !range [[RNG0]]
+; CHECK-NEXT:    [[CTPOP:%.*]] = tail call range(i32 0, 33) i32 @llvm.ctpop.i32(i32 [[X]])
 ; CHECK-NEXT:    [[CMP:%.*]] = icmp ugt i32 [[CTPOP]], 1
 ; CHECK-NEXT:    [[ZEXT:%.*]] = zext i1 [[CMP]] to i64
 ; CHECK-NEXT:    [[RET:%.*]] = add nuw nsw i64 [[EXT]], [[ZEXT]]
@@ -332,6 +332,3 @@ declare void @use64(i64)
 
 declare i32 @llvm.ctlz.i32(i32, i1)
 declare i32 @llvm.ctpop.i32(i32)
-;.
-; CHECK: [[RNG0]] = !{i32 0, i32 33}
-;.
diff --git a/llvm/test/Transforms/InstCombine/freeze-integer-intrinsics.ll b/llvm/test/Transforms/InstCombine/freeze-integer-intrinsics.ll
index 105bd28fb052..c4a590e1a12b 100644
--- a/llvm/test/Transforms/InstCombine/freeze-integer-intrinsics.ll
+++ b/llvm/test/Transforms/InstCombine/freeze-integer-intrinsics.ll
@@ -3,7 +3,7 @@
 
 define i32 @ctlz_true_freeze(i32 %arg) {
 ; CHECK-LABEL: @ctlz_true_freeze(
-; CHECK-NEXT:    [[CALL:%.*]] = call i32 @llvm.ctlz.i32(i32 [[ARG:%.*]], i1 true), !range [[RNG0:![0-9]+]]
+; CHECK-NEXT:    [[CALL:%.*]] = call range(i32 0, 33) i32 @llvm.ctlz.i32(i32 [[ARG:%.*]], i1 true)
 ; CHECK-NEXT:    [[FREEZE:%.*]] = freeze i32 [[CALL]]
 ; CHECK-NEXT:    ret i32 [[FREEZE]]
 ;
@@ -15,7 +15,7 @@ define i32 @ctlz_true_freeze(i32 %arg) {
 define i32 @ctlz_false_freeze(i32 %arg) {
 ; CHECK-LABEL: @ctlz_false_freeze(
 ; CHECK-NEXT:    [[ARG_FR:%.*]] = freeze i32 [[ARG:%.*]]
-; CHECK-NEXT:    [[CALL:%.*]] = call i32 @llvm.ctlz.i32(i32 [[ARG_FR]], i1 false), !range [[RNG0]]
+; CHECK-NEXT:    [[CALL:%.*]] = call range(i32 0, 33) i32 @llvm.ctlz.i32(i32 [[ARG_FR]], i1 false)
 ; CHECK-NEXT:    ret i32 [[CALL]]
 ;
   %call = call i32 @llvm.ctlz.i32(i32 %arg, i1 false)
@@ -25,7 +25,7 @@ define i32 @ctlz_false_freeze(i32 %arg) {
 
 define i32 @ctlz_true_noundef_freeze(i32 %arg) {
 ; CHECK-LABEL: @ctlz_true_noundef_freeze(
-; CHECK-NEXT:    [[CALL:%.*]] = call noundef i32 @llvm.ctlz.i32(i32 [[ARG:%.*]], i1 true), !range [[RNG0]]
+; CHECK-NEXT:    [[CALL:%.*]] = call noundef range(i32 0, 33) i32 @llvm.ctlz.i32(i32 [[ARG:%.*]], i1 true)
 ; CHECK-NEXT:    ret i32 [[CALL]]
 ;
   %call = call noundef i32 @llvm.ctlz.i32(i32 %arg, i1 true)
@@ -35,7 +35,7 @@ define i32 @ctlz_true_noundef_freeze(i32 %arg) {
 
 define i32 @cttz_true_freeze(i32 %arg) {
 ; CHECK-LABEL: @cttz_true_freeze(
-; CHECK-NEXT:    [[CALL:%.*]] = call i32 @llvm.cttz.i32(i32 [[ARG:%.*]], i1 true), !range [[RNG0]]
+; CHECK-NEXT:    [[CALL:%.*]] = call range(i32 0, 33) i32 @llvm.cttz.i32(i32 [[ARG:%.*]], i1 true)
 ; CHECK-NEXT:    [[FREEZE:%.*]] = freeze i32 [[CALL]]
 ; CHECK-NEXT:    ret i32 [[FREEZE]]
 ;
@@ -46,7 +46,7 @@ define i32 @cttz_true_freeze(i32 %arg) {
 
 define i32 @cttz_true_noundef_freeze(i32 %arg) {
 ; CHECK-LABEL: @cttz_true_noundef_freeze(
-; CHECK-NEXT:    [[CALL:%.*]] = call noundef i32 @llvm.cttz.i32(i32 [[ARG:%.*]], i1 true), !range [[RNG0]]
+; CHECK-NEXT:    [[CALL:%.*]] = call noundef range(i32 0, 33) i32 @llvm.cttz.i32(i32 [[ARG:%.*]], i1 true)
 ; CHECK-NEXT:    ret i32 [[CALL]]
 ;
   %call = call noundef i32 @llvm.cttz.i32(i32 %arg, i1 true)
@@ -57,7 +57,7 @@ define i32 @cttz_true_noundef_freeze(i32 %arg) {
 define i32 @freeze_cttz_true(i32 %arg) {
 ; CHECK-LABEL: @freeze_cttz_true(
 ; CHECK-NEXT:    [[FREEZE:%.*]] = freeze i32 [[ARG:%.*]]
-; CHECK-NEXT:    [[CALL:%.*]] = call i32 @llvm.cttz.i32(i32 [[FREEZE]], i1 true), !range [[RNG0]]
+; CHECK-NEXT:    [[CALL:%.*]] = call range(i32 0, 33) i32 @llvm.cttz.i32(i32 [[FREEZE]], i1 true)
 ; CHECK-NEXT:    ret i32 [[CALL]]
 ;
   %freeze = freeze i32 %arg
@@ -68,7 +68,7 @@ define i32 @freeze_cttz_true(i32 %arg) {
 define i32 @cttz_false_freeze(i32 %arg) {
 ; CHECK-LABEL: @cttz_false_freeze(
 ; CHECK-NEXT:    [[ARG_FR:%.*]] = freeze i32 [[ARG:%.*]]
-; CHECK-NEXT:    [[CALL:%.*]] = call i32 @llvm.cttz.i32(i32 [[ARG_FR]], i1 false), !range [[RNG0]]
+; CHECK-NEXT:    [[CALL:%.*]] = call range(i32 0, 33) i32 @llvm.cttz.i32(i32 [[ARG_FR]], i1 false)
 ; CHECK-NEXT:    ret i32 [[CALL]]
 ;
   %call = call i32 @llvm.cttz.i32(i32 %arg, i1 false)
@@ -79,7 +79,7 @@ define i32 @cttz_false_freeze(i32 %arg) {
 define i32 @freeze_cttz_false(i32 %arg) {
 ; CHECK-LABEL: @freeze_cttz_false(
 ; CHECK-NEXT:    [[FREEZE:%.*]] = freeze i32 [[ARG:%.*]]
-; CHECK-NEXT:    [[CALL:%.*]] = call i32 @llvm.cttz.i32(i32 [[FREEZE]], i1 false), !range [[RNG0]]
+; CHECK-NEXT:    [[CALL:%.*]] = call range(i32 0, 33) i32 @llvm.cttz.i32(i32 [[FREEZE]], i1 false)
 ; CHECK-NEXT:    ret i32 [[CALL]]
 ;
   %freeze = freeze i32 %arg
diff --git a/llvm/test/Transforms/InstCombine/freeze.ll b/llvm/test/Transforms/InstCombine/freeze.ll
index adcf7d50f413..391d626a795c 100644
--- a/llvm/test/Transforms/InstCombine/freeze.ll
+++ b/llvm/test/Transforms/InstCombine/freeze.ll
@@ -1107,7 +1107,7 @@ define i32 @freeze_ctpop(i32 %x) {
 ; CHECK-LABEL: @freeze_ctpop(
 ; CHECK-NEXT:    [[Y:%.*]] = lshr i32 2047, [[X:%.*]]
 ; CHECK-NEXT:    [[Y_FR:%.*]] = freeze i32 [[Y]]
-; CHECK-NEXT:    [[CTPOP:%.*]] = call i32 @llvm.ctpop.i32(i32 [[Y_FR]]), !range [[RNG3:![0-9]+]]
+; CHECK-NEXT:    [[CTPOP:%.*]] = call range(i32 0, 33) i32 @llvm.ctpop.i32(i32 [[Y_FR]])
 ; CHECK-NEXT:    ret i32 [[CTPOP]]
 ;
   %y = lshr i32 2047, %x
@@ -1209,5 +1209,4 @@ define ptr @freeze_ptrmask_nonnull(ptr %p, i64 noundef %m) {
 ; CHECK: [[META0]] = !{}
 ; CHECK: [[META1]] = !{i64 4}
 ; CHECK: [[RNG2]] = !{i32 0, i32 100}
-; CHECK: [[RNG3]] = !{i32 0, i32 33}
 ;.
diff --git a/llvm/test/Transforms/InstCombine/gepofconstgepi8.ll b/llvm/test/Transforms/InstCombine/gepofconstgepi8.ll
new file mode 100644
index 000000000000..7b7c6fba699c
--- /dev/null
+++ b/llvm/test/Transforms/InstCombine/gepofconstgepi8.ll
@@ -0,0 +1,292 @@
+; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --version 4
+; RUN: opt < %s -S -passes=instcombine | FileCheck %s
+
+declare void @use64(i64)
+declare void @useptr(ptr)
+
+define ptr @test_zero(ptr %base, i64 %a) {
+; CHECK-LABEL: define ptr @test_zero(
+; CHECK-SAME: ptr [[BASE:%.*]], i64 [[A:%.*]]) {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[P2:%.*]] = getelementptr i32, ptr [[BASE]], i64 [[A]]
+; CHECK-NEXT:    ret ptr [[P2]]
+;
+entry:
+  %p1 = getelementptr i8, ptr %base, i64 -4
+  %index = add i64 %a, 1
+  %p2 = getelementptr i32, ptr %p1, i64 %index
+  ret ptr %p2
+}
+
+define ptr @test_nonzero(ptr %base, i64 %a) {
+; CHECK-LABEL: define ptr @test_nonzero(
+; CHECK-SAME: ptr [[BASE:%.*]], i64 [[A:%.*]]) {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP0:%.*]] = getelementptr i8, ptr [[BASE]], i64 4
+; CHECK-NEXT:    [[P2:%.*]] = getelementptr i32, ptr [[TMP0]], i64 [[A]]
+; CHECK-NEXT:    ret ptr [[P2]]
+;
+entry:
+  %p1 = getelementptr i8, ptr %base, i64 -4
+  %index = add i64 %a, 2
+  %p2 = getelementptr i32, ptr %p1, i64 %index
+  ret ptr %p2
+}
+
+define ptr @test_or_disjoint(ptr %base, i64 %a) {
+; CHECK-LABEL: define ptr @test_or_disjoint(
+; CHECK-SAME: ptr [[BASE:%.*]], i64 [[A:%.*]]) {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[P2:%.*]] = getelementptr i32, ptr [[BASE]], i64 [[A]]
+; CHECK-NEXT:    ret ptr [[P2]]
+;
+entry:
+  %p1 = getelementptr i8, ptr %base, i64 -4
+  %index = or disjoint i64 %a, 1
+  %p2 = getelementptr i32, ptr %p1, i64 %index
+  ret ptr %p2
+}
+
+define ptr @test_zero_multiuse_index(ptr %base, i64 %a) {
+; CHECK-LABEL: define ptr @test_zero_multiuse_index(
+; CHECK-SAME: ptr [[BASE:%.*]], i64 [[A:%.*]]) {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[INDEX:%.*]] = add i64 [[A]], 1
+; CHECK-NEXT:    call void @use64(i64 [[INDEX]])
+; CHECK-NEXT:    [[P2:%.*]] = getelementptr i32, ptr [[BASE]], i64 [[A]]
+; CHECK-NEXT:    ret ptr [[P2]]
+;
+entry:
+  %p1 = getelementptr i8, ptr %base, i64 -4
+  %index = add i64 %a, 1
+  call void @use64(i64 %index)
+  %p2 = getelementptr i32, ptr %p1, i64 %index
+  ret ptr %p2
+}
+
+define ptr @test_zero_multiuse_ptr(ptr %base, i64 %a) {
+; CHECK-LABEL: define ptr @test_zero_multiuse_ptr(
+; CHECK-SAME: ptr [[BASE:%.*]], i64 [[A:%.*]]) {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[P1:%.*]] = getelementptr i8, ptr [[BASE]], i64 -4
+; CHECK-NEXT:    call void @useptr(ptr [[P1]])
+; CHECK-NEXT:    [[P2:%.*]] = getelementptr i32, ptr [[BASE]], i64 [[A]]
+; CHECK-NEXT:    ret ptr [[P2]]
+;
+entry:
+  %p1 = getelementptr i8, ptr %base, i64 -4
+  call void @useptr(ptr %p1)
+  %index = add i64 %a, 1
+  %p2 = getelementptr i32, ptr %p1, i64 %index
+  ret ptr %p2
+}
+
+define ptr @test_zero_sext_add_nsw(ptr %base, i32 %a) {
+; CHECK-LABEL: define ptr @test_zero_sext_add_nsw(
+; CHECK-SAME: ptr [[BASE:%.*]], i32 [[A:%.*]]) {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[P1:%.*]] = getelementptr i8, ptr [[BASE]], i64 -4
+; CHECK-NEXT:    [[TMP0:%.*]] = sext i32 [[A]] to i64
+; CHECK-NEXT:    [[TMP1:%.*]] = getelementptr i32, ptr [[P1]], i64 [[TMP0]]
+; CHECK-NEXT:    [[P2:%.*]] = getelementptr i8, ptr [[TMP1]], i64 4
+; CHECK-NEXT:    ret ptr [[P2]]
+;
+entry:
+  %p1 = getelementptr i8, ptr %base, i64 -4
+  %index = add nsw i32 %a, 1
+  %p2 = getelementptr i32, ptr %p1, i32 %index
+  ret ptr %p2
+}
+
+define ptr @test_zero_trunc_add(ptr %base, i128 %a) {
+; CHECK-LABEL: define ptr @test_zero_trunc_add(
+; CHECK-SAME: ptr [[BASE:%.*]], i128 [[A:%.*]]) {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP0:%.*]] = trunc i128 [[A]] to i64
+; CHECK-NEXT:    [[P2:%.*]] = getelementptr i32, ptr [[BASE]], i64 [[TMP0]]
+; CHECK-NEXT:    ret ptr [[P2]]
+;
+entry:
+  %p1 = getelementptr i8, ptr %base, i64 -4
+  %index = add i128 %a, 1
+  %p2 = getelementptr i32, ptr %p1, i128 %index
+  ret ptr %p2
+}
+
+define ptr @test_non_i8(ptr %base, i64 %a) {
+; CHECK-LABEL: define ptr @test_non_i8(
+; CHECK-SAME: ptr [[BASE:%.*]], i64 [[A:%.*]]) {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[P1:%.*]] = getelementptr i8, ptr [[BASE]], i64 -4
+; CHECK-NEXT:    [[TMP0:%.*]] = getelementptr i32, ptr [[P1]], i64 [[A]]
+; CHECK-NEXT:    ret ptr [[TMP0]]
+;
+entry:
+  %p1 = getelementptr i16, ptr %base, i64 -4
+  %index = add i64 %a, 1
+  %p2 = getelementptr i32, ptr %p1, i64 %index
+  ret ptr %p2
+}
+
+define ptr @test_non_const(ptr %base, i64 %a, i64 %b) {
+; CHECK-LABEL: define ptr @test_non_const(
+; CHECK-SAME: ptr [[BASE:%.*]], i64 [[A:%.*]], i64 [[B:%.*]]) {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[P1:%.*]] = getelementptr i8, ptr [[BASE]], i64 [[B]]
+; CHECK-NEXT:    [[TMP0:%.*]] = getelementptr i32, ptr [[P1]], i64 [[A]]
+; CHECK-NEXT:    [[P2:%.*]] = getelementptr i8, ptr [[TMP0]], i64 4
+; CHECK-NEXT:    ret ptr [[P2]]
+;
+entry:
+  %p1 = getelementptr i8, ptr %base, i64 %b
+  %index = add i64 %a, 1
+  %p2 = getelementptr i32, ptr %p1, i64 %index
+  ret ptr %p2
+}
+
+define ptr @test_too_many_indices(ptr %base, i64 %a, i64 %b) {
+; CHECK-LABEL: define ptr @test_too_many_indices(
+; CHECK-SAME: ptr [[BASE:%.*]], i64 [[A:%.*]], i64 [[B:%.*]]) {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[P1:%.*]] = getelementptr i8, ptr [[BASE]], i64 [[B]]
+; CHECK-NEXT:    [[INDEX:%.*]] = add i64 [[A]], 1
+; CHECK-NEXT:    [[P2:%.*]] = getelementptr [8 x i32], ptr [[P1]], i64 1, i64 [[INDEX]]
+; CHECK-NEXT:    ret ptr [[P2]]
+;
+entry:
+  %p1 = getelementptr i8, ptr %base, i64 %b
+  %index = add i64 %a, 1
+  %p2 = getelementptr [8 x i32], ptr %p1, i64 1, i64 %index
+  ret ptr %p2
+}
+
+define ptr @test_wrong_op(ptr %base, i64 %a) {
+; CHECK-LABEL: define ptr @test_wrong_op(
+; CHECK-SAME: ptr [[BASE:%.*]], i64 [[A:%.*]]) {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[P1:%.*]] = getelementptr i8, ptr [[BASE]], i64 -4
+; CHECK-NEXT:    [[INDEX:%.*]] = xor i64 [[A]], 1
+; CHECK-NEXT:    [[P2:%.*]] = getelementptr i32, ptr [[P1]], i64 [[INDEX]]
+; CHECK-NEXT:    ret ptr [[P2]]
+;
+entry:
+  %p1 = getelementptr i8, ptr %base, i64 -4
+  %index = xor i64 %a, 1
+  %p2 = getelementptr i32, ptr %p1, i64 %index
+  ret ptr %p2
+}
+
+define ptr @test_sext_add_without_nsw(ptr %base, i32 %a) {
+; CHECK-LABEL: define ptr @test_sext_add_without_nsw(
+; CHECK-SAME: ptr [[BASE:%.*]], i32 [[A:%.*]]) {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[P1:%.*]] = getelementptr i8, ptr [[BASE]], i64 -4
+; CHECK-NEXT:    [[INDEX:%.*]] = add i32 [[A]], 1
+; CHECK-NEXT:    [[TMP0:%.*]] = sext i32 [[INDEX]] to i64
+; CHECK-NEXT:    [[P2:%.*]] = getelementptr i32, ptr [[P1]], i64 [[TMP0]]
+; CHECK-NEXT:    ret ptr [[P2]]
+;
+entry:
+  %p1 = getelementptr i8, ptr %base, i64 -4
+  %index = add i32 %a, 1
+  %p2 = getelementptr i32, ptr %p1, i32 %index
+  ret ptr %p2
+}
+
+define ptr @test_or_without_disjoint(ptr %base, i64 %a) {
+; CHECK-LABEL: define ptr @test_or_without_disjoint(
+; CHECK-SAME: ptr [[BASE:%.*]], i64 [[A:%.*]]) {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[P1:%.*]] = getelementptr i8, ptr [[BASE]], i64 -4
+; CHECK-NEXT:    [[INDEX:%.*]] = or i64 [[A]], 1
+; CHECK-NEXT:    [[P2:%.*]] = getelementptr i32, ptr [[P1]], i64 [[INDEX]]
+; CHECK-NEXT:    ret ptr [[P2]]
+;
+entry:
+  %p1 = getelementptr i8, ptr %base, i64 -4
+  %index = or i64 %a, 1
+  %p2 = getelementptr i32, ptr %p1, i64 %index
+  ret ptr %p2
+}
+
+define ptr @test_smul_overflow(ptr %base, i64 %a) {
+; CHECK-LABEL: define ptr @test_smul_overflow(
+; CHECK-SAME: ptr [[BASE:%.*]], i64 [[A:%.*]]) {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[P1:%.*]] = getelementptr i8, ptr [[BASE]], i64 -12
+; CHECK-NEXT:    [[TMP0:%.*]] = getelementptr i32, ptr [[P1]], i64 [[A]]
+; CHECK-NEXT:    ret ptr [[TMP0]]
+;
+entry:
+  %p1 = getelementptr i8, ptr %base, i64 -4
+  %index = add i64 %a, 9223372036854775806
+  %p2 = getelementptr i32, ptr %p1, i64 %index
+  ret ptr %p2
+}
+
+define ptr @test_sadd_overflow(ptr %base, i64 %a) {
+; CHECK-LABEL: define ptr @test_sadd_overflow(
+; CHECK-SAME: ptr [[BASE:%.*]], i64 [[A:%.*]]) {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[P1:%.*]] = getelementptr i8, ptr [[BASE]], i64 -9223372036854775808
+; CHECK-NEXT:    [[TMP0:%.*]] = getelementptr i32, ptr [[P1]], i64 [[A]]
+; CHECK-NEXT:    ret ptr [[TMP0]]
+;
+entry:
+  %p1 = getelementptr i8, ptr %base, i64 9223372036854775804
+  %index = add i64 %a, 1
+  %p2 = getelementptr i32, ptr %p1, i64 %index
+  ret ptr %p2
+}
+
+define ptr @test_nonzero_multiuse_index(ptr %base, i64 %a) {
+; CHECK-LABEL: define ptr @test_nonzero_multiuse_index(
+; CHECK-SAME: ptr [[BASE:%.*]], i64 [[A:%.*]]) {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[P1:%.*]] = getelementptr i8, ptr [[BASE]], i64 -4
+; CHECK-NEXT:    [[INDEX:%.*]] = add i64 [[A]], 2
+; CHECK-NEXT:    call void @use64(i64 [[INDEX]])
+; CHECK-NEXT:    [[P2:%.*]] = getelementptr i32, ptr [[P1]], i64 [[INDEX]]
+; CHECK-NEXT:    ret ptr [[P2]]
+;
+entry:
+  %p1 = getelementptr i8, ptr %base, i64 -4
+  %index = add i64 %a, 2
+  call void @use64(i64 %index)
+  %p2 = getelementptr i32, ptr %p1, i64 %index
+  ret ptr %p2
+}
+
+define ptr @test_nonzero_multiuse_ptr(ptr %base, i64 %a) {
+; CHECK-LABEL: define ptr @test_nonzero_multiuse_ptr(
+; CHECK-SAME: ptr [[BASE:%.*]], i64 [[A:%.*]]) {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[P1:%.*]] = getelementptr i8, ptr [[BASE]], i64 -4
+; CHECK-NEXT:    call void @useptr(ptr [[P1]])
+; CHECK-NEXT:    [[TMP0:%.*]] = getelementptr i32, ptr [[P1]], i64 [[A]]
+; CHECK-NEXT:    [[P2:%.*]] = getelementptr i8, ptr [[TMP0]], i64 8
+; CHECK-NEXT:    ret ptr [[P2]]
+;
+entry:
+  %p1 = getelementptr i8, ptr %base, i64 -4
+  call void @useptr(ptr %p1)
+  %index = add i64 %a, 2
+  %p2 = getelementptr i32, ptr %p1, i64 %index
+  ret ptr %p2
+}
+
+define ptr @test_scalable(ptr %base, i64 %a) {
+; CHECK-LABEL: define ptr @test_scalable(
+; CHECK-SAME: ptr [[BASE:%.*]], i64 [[A:%.*]]) {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[P1:%.*]] = getelementptr i8, ptr [[BASE]], i64 -4
+; CHECK-NEXT:    [[TMP0:%.*]] = getelementptr <vscale x 4 x i32>, ptr [[P1]], i64 [[A]]
+; CHECK-NEXT:    [[P2:%.*]] = getelementptr <vscale x 4 x i32>, ptr [[TMP0]], i64 1
+; CHECK-NEXT:    ret ptr [[P2]]
+;
+entry:
+  %p1 = getelementptr i8, ptr %base, i64 -4
+  %index = add i64 %a, 1
+  %p2 = getelementptr <vscale x 4 x i32>, ptr %p1, i64 %index
+  ret ptr %p2
+}
diff --git a/llvm/test/Transforms/InstCombine/icmp-ne-pow2.ll b/llvm/test/Transforms/InstCombine/icmp-ne-pow2.ll
index 70a2b33d17dd..e9ec6b415d46 100644
--- a/llvm/test/Transforms/InstCombine/icmp-ne-pow2.ll
+++ b/llvm/test/Transforms/InstCombine/icmp-ne-pow2.ll
@@ -284,7 +284,7 @@ False:
 
 define i32 @pow2_32_nonconst_assume(i32 %x, i32 %y) {
 ; CHECK-LABEL: @pow2_32_nonconst_assume(
-; CHECK-NEXT:    [[CTPOP:%.*]] = call i32 @llvm.ctpop.i32(i32 [[Y:%.*]]), !range [[RNG0:![0-9]+]]
+; CHECK-NEXT:    [[CTPOP:%.*]] = call range(i32 0, 33) i32 @llvm.ctpop.i32(i32 [[Y:%.*]])
 ; CHECK-NEXT:    [[YP2:%.*]] = icmp eq i32 [[CTPOP]], 1
 ; CHECK-NEXT:    call void @llvm.assume(i1 [[YP2]])
 ; CHECK-NEXT:    [[AND:%.*]] = and i32 [[X:%.*]], [[Y]]
@@ -306,7 +306,7 @@ define i32 @pow2_32_nonconst_assume(i32 %x, i32 %y) {
 
 define i32 @pow2_32_gtnonconst_assume(i32 %x, i32 %y) {
 ; CHECK-LABEL: @pow2_32_gtnonconst_assume(
-; CHECK-NEXT:    [[CTPOP:%.*]] = call i32 @llvm.ctpop.i32(i32 [[Y:%.*]]), !range [[RNG0]]
+; CHECK-NEXT:    [[CTPOP:%.*]] = call range(i32 0, 33) i32 @llvm.ctpop.i32(i32 [[Y:%.*]])
 ; CHECK-NEXT:    [[YP2:%.*]] = icmp eq i32 [[CTPOP]], 1
 ; CHECK-NEXT:    call void @llvm.assume(i1 [[YP2]])
 ; CHECK-NEXT:    [[YGT:%.*]] = icmp ugt i32 [[Y]], [[X:%.*]]
@@ -327,7 +327,7 @@ define i32 @pow2_32_gtnonconst_assume(i32 %x, i32 %y) {
 
 define i32 @not_pow2_32_nonconst_assume(i32 %x, i32 %y) {
 ; CHECK-LABEL: @not_pow2_32_nonconst_assume(
-; CHECK-NEXT:    [[CTPOP:%.*]] = call i32 @llvm.ctpop.i32(i32 [[Y:%.*]]), !range [[RNG0]]
+; CHECK-NEXT:    [[CTPOP:%.*]] = call range(i32 0, 33) i32 @llvm.ctpop.i32(i32 [[Y:%.*]])
 ; CHECK-NEXT:    [[YP2:%.*]] = icmp ne i32 [[CTPOP]], 1
 ; CHECK-NEXT:    call void @llvm.assume(i1 [[YP2]])
 ; CHECK-NEXT:    [[AND:%.*]] = and i32 [[X:%.*]], [[Y]]
@@ -349,7 +349,7 @@ define i32 @not_pow2_32_nonconst_assume(i32 %x, i32 %y) {
 
 define i32 @pow2_or_zero_32_nonconst_assume(i32 %x, i32 %y) {
 ; CHECK-LABEL: @pow2_or_zero_32_nonconst_assume(
-; CHECK-NEXT:    [[CTPOP:%.*]] = call i32 @llvm.ctpop.i32(i32 [[Y:%.*]]), !range [[RNG0]]
+; CHECK-NEXT:    [[CTPOP:%.*]] = call range(i32 0, 33) i32 @llvm.ctpop.i32(i32 [[Y:%.*]])
 ; CHECK-NEXT:    [[YP2:%.*]] = icmp ult i32 [[CTPOP]], 2
 ; CHECK-NEXT:    call void @llvm.assume(i1 [[YP2]])
 ; CHECK-NEXT:    [[AND:%.*]] = and i32 [[X:%.*]], [[Y]]
@@ -371,7 +371,7 @@ define i32 @pow2_or_zero_32_nonconst_assume(i32 %x, i32 %y) {
 
 define i32 @pow2_32_nonconst_assume_br(i32 %x, i32 %y) {
 ; CHECK-LABEL: @pow2_32_nonconst_assume_br(
-; CHECK-NEXT:    [[CTPOP:%.*]] = call i32 @llvm.ctpop.i32(i32 [[Y:%.*]]), !range [[RNG0]]
+; CHECK-NEXT:    [[CTPOP:%.*]] = call range(i32 0, 33) i32 @llvm.ctpop.i32(i32 [[Y:%.*]])
 ; CHECK-NEXT:    [[YP2:%.*]] = icmp eq i32 [[CTPOP]], 1
 ; CHECK-NEXT:    call void @llvm.assume(i1 [[YP2]])
 ; CHECK-NEXT:    [[AND:%.*]] = and i32 [[X:%.*]], [[Y]]
@@ -398,7 +398,7 @@ False:
 
 define i32 @not_pow2_32_nonconst_assume_br(i32 %x, i32 %y) {
 ; CHECK-LABEL: @not_pow2_32_nonconst_assume_br(
-; CHECK-NEXT:    [[CTPOP:%.*]] = call i32 @llvm.ctpop.i32(i32 [[Y:%.*]]), !range [[RNG0]]
+; CHECK-NEXT:    [[CTPOP:%.*]] = call range(i32 0, 33) i32 @llvm.ctpop.i32(i32 [[Y:%.*]])
 ; CHECK-NEXT:    [[YP2:%.*]] = icmp ne i32 [[CTPOP]], 1
 ; CHECK-NEXT:    call void @llvm.assume(i1 [[YP2]])
 ; CHECK-NEXT:    [[AND:%.*]] = and i32 [[X:%.*]], [[Y]]
@@ -425,7 +425,7 @@ False:
 
 define i32 @pow2_or_zero_32_nonconst_assume_br(i32 %x, i32 %y) {
 ; CHECK-LABEL: @pow2_or_zero_32_nonconst_assume_br(
-; CHECK-NEXT:    [[CTPOP:%.*]] = call i32 @llvm.ctpop.i32(i32 [[Y:%.*]]), !range [[RNG0]]
+; CHECK-NEXT:    [[CTPOP:%.*]] = call range(i32 0, 33) i32 @llvm.ctpop.i32(i32 [[Y:%.*]])
 ; CHECK-NEXT:    [[YP2:%.*]] = icmp ult i32 [[CTPOP]], 2
 ; CHECK-NEXT:    call void @llvm.assume(i1 [[YP2]])
 ; CHECK-NEXT:    [[AND:%.*]] = and i32 [[X:%.*]], [[Y]]
@@ -452,7 +452,7 @@ False:
 
 define i32 @pow2_32_nonconst_br1_br(i32 %x, i32 %y) {
 ; CHECK-LABEL: @pow2_32_nonconst_br1_br(
-; CHECK-NEXT:    [[CTPOP:%.*]] = call i32 @llvm.ctpop.i32(i32 [[Y:%.*]]), !range [[RNG0]]
+; CHECK-NEXT:    [[CTPOP:%.*]] = call range(i32 0, 33) i32 @llvm.ctpop.i32(i32 [[Y:%.*]])
 ; CHECK-NEXT:    [[YP2:%.*]] = icmp eq i32 [[CTPOP]], 1
 ; CHECK-NEXT:    br i1 [[YP2]], label [[CONT:%.*]], label [[FALSE:%.*]]
 ; CHECK:       Cont:
@@ -481,7 +481,7 @@ False:
 
 define i32 @not_pow2_32_nonconst_br1_br(i32 %x, i32 %y) {
 ; CHECK-LABEL: @not_pow2_32_nonconst_br1_br(
-; CHECK-NEXT:    [[CTPOP:%.*]] = call i32 @llvm.ctpop.i32(i32 [[Y:%.*]]), !range [[RNG0]]
+; CHECK-NEXT:    [[CTPOP:%.*]] = call range(i32 0, 33) i32 @llvm.ctpop.i32(i32 [[Y:%.*]])
 ; CHECK-NEXT:    [[YP2_NOT:%.*]] = icmp eq i32 [[CTPOP]], 1
 ; CHECK-NEXT:    br i1 [[YP2_NOT]], label [[FALSE:%.*]], label [[CONT:%.*]]
 ; CHECK:       Cont:
@@ -513,7 +513,7 @@ define i32 @maybe_pow2_32_noncont(i32 %x, i32 %y) {
 ; CHECK-NEXT:    [[YGT8:%.*]] = icmp ugt i32 [[Y:%.*]], 8
 ; CHECK-NEXT:    br i1 [[YGT8]], label [[CONT1:%.*]], label [[CONT2:%.*]]
 ; CHECK:       Cont1:
-; CHECK-NEXT:    [[CTPOP:%.*]] = call i32 @llvm.ctpop.i32(i32 [[Y]]), !range [[RNG0]]
+; CHECK-NEXT:    [[CTPOP:%.*]] = call range(i32 0, 33) i32 @llvm.ctpop.i32(i32 [[Y]])
 ; CHECK-NEXT:    [[YP2:%.*]] = icmp eq i32 [[CTPOP]], 1
 ; CHECK-NEXT:    call void @llvm.assume(i1 [[YP2]])
 ; CHECK-NEXT:    br i1 true, label [[CONT2]], label [[FALSE:%.*]]
diff --git a/llvm/test/Transforms/InstCombine/icmp-of-trunc-ext.ll b/llvm/test/Transforms/InstCombine/icmp-of-trunc-ext.ll
index 7f616bbb2a83..a61694919ab0 100644
--- a/llvm/test/Transforms/InstCombine/icmp-of-trunc-ext.ll
+++ b/llvm/test/Transforms/InstCombine/icmp-of-trunc-ext.ll
@@ -268,3 +268,328 @@ define i1 @icmp_trunc_x_zext_y_fail_multiuse(i32 %x, i8 %y) {
   %r = icmp ule i16 %x16, %y16
   ret i1 %r
 }
+
+define i1 @trunc_unsigned_nuw(i16 %x, i16 %y) {
+; CHECK-LABEL: @trunc_unsigned_nuw(
+; CHECK-NEXT:    [[XT:%.*]] = trunc nuw i16 [[X:%.*]] to i8
+; CHECK-NEXT:    [[YT:%.*]] = trunc nuw i16 [[Y:%.*]] to i8
+; CHECK-NEXT:    [[C:%.*]] = icmp ult i8 [[XT]], [[YT]]
+; CHECK-NEXT:    ret i1 [[C]]
+;
+  %xt = trunc nuw i16 %x to i8
+  %yt = trunc nuw i16 %y to i8
+  %c = icmp ult i8 %xt, %yt
+  ret i1 %c
+}
+
+define i1 @trunc_unsigned_nsw(i16 %x, i16 %y) {
+; CHECK-LABEL: @trunc_unsigned_nsw(
+; CHECK-NEXT:    [[XT:%.*]] = trunc nsw i16 [[X:%.*]] to i8
+; CHECK-NEXT:    [[YT:%.*]] = trunc nsw i16 [[Y:%.*]] to i8
+; CHECK-NEXT:    [[C:%.*]] = icmp ult i8 [[XT]], [[YT]]
+; CHECK-NEXT:    ret i1 [[C]]
+;
+  %xt = trunc nsw i16 %x to i8
+  %yt = trunc nsw i16 %y to i8
+  %c = icmp ult i8 %xt, %yt
+  ret i1 %c
+}
+
+define i1 @trunc_unsigned_both(i16 %x, i16 %y) {
+; CHECK-LABEL: @trunc_unsigned_both(
+; CHECK-NEXT:    [[XT:%.*]] = trunc nuw nsw i16 [[X:%.*]] to i8
+; CHECK-NEXT:    [[YT:%.*]] = trunc nuw nsw i16 [[Y:%.*]] to i8
+; CHECK-NEXT:    [[C:%.*]] = icmp ult i8 [[XT]], [[YT]]
+; CHECK-NEXT:    ret i1 [[C]]
+;
+  %xt = trunc nuw nsw i16 %x to i8
+  %yt = trunc nuw nsw i16 %y to i8
+  %c = icmp ult i8 %xt, %yt
+  ret i1 %c
+}
+
+define i1 @trunc_unsigned_either(i16 %x, i16 %y) {
+; CHECK-LABEL: @trunc_unsigned_either(
+; CHECK-NEXT:    [[XT:%.*]] = trunc nuw i16 [[X:%.*]] to i8
+; CHECK-NEXT:    [[YT:%.*]] = trunc nsw i16 [[Y:%.*]] to i8
+; CHECK-NEXT:    [[C:%.*]] = icmp ult i8 [[XT]], [[YT]]
+; CHECK-NEXT:    ret i1 [[C]]
+;
+  %xt = trunc nuw i16 %x to i8
+  %yt = trunc nsw i16 %y to i8
+  %c = icmp ult i8 %xt, %yt
+  ret i1 %c
+}
+
+define i1 @trunc_signed_nuw(i16 %x, i16 %y) {
+; CHECK-LABEL: @trunc_signed_nuw(
+; CHECK-NEXT:    [[XT:%.*]] = trunc nuw i16 [[X:%.*]] to i8
+; CHECK-NEXT:    [[YT:%.*]] = trunc nuw i16 [[Y:%.*]] to i8
+; CHECK-NEXT:    [[C:%.*]] = icmp slt i8 [[XT]], [[YT]]
+; CHECK-NEXT:    ret i1 [[C]]
+;
+  %xt = trunc nuw i16 %x to i8
+  %yt = trunc nuw i16 %y to i8
+  %c = icmp slt i8 %xt, %yt
+  ret i1 %c
+}
+
+define i1 @trunc_signed_nsw(i16 %x, i16 %y) {
+; CHECK-LABEL: @trunc_signed_nsw(
+; CHECK-NEXT:    [[XT:%.*]] = trunc nsw i16 [[X:%.*]] to i8
+; CHECK-NEXT:    [[YT:%.*]] = trunc nsw i16 [[Y:%.*]] to i8
+; CHECK-NEXT:    [[C:%.*]] = icmp slt i8 [[XT]], [[YT]]
+; CHECK-NEXT:    ret i1 [[C]]
+;
+  %xt = trunc nsw i16 %x to i8
+  %yt = trunc nsw i16 %y to i8
+  %c = icmp slt i8 %xt, %yt
+  ret i1 %c
+}
+
+define i1 @trunc_signed_both(i16 %x, i16 %y) {
+; CHECK-LABEL: @trunc_signed_both(
+; CHECK-NEXT:    [[XT:%.*]] = trunc nuw nsw i16 [[X:%.*]] to i8
+; CHECK-NEXT:    [[YT:%.*]] = trunc nuw nsw i16 [[Y:%.*]] to i8
+; CHECK-NEXT:    [[C:%.*]] = icmp slt i8 [[XT]], [[YT]]
+; CHECK-NEXT:    ret i1 [[C]]
+;
+  %xt = trunc nuw nsw i16 %x to i8
+  %yt = trunc nuw nsw i16 %y to i8
+  %c = icmp slt i8 %xt, %yt
+  ret i1 %c
+}
+
+define i1 @trunc_signed_either(i16 %x, i16 %y) {
+; CHECK-LABEL: @trunc_signed_either(
+; CHECK-NEXT:    [[XT:%.*]] = trunc nuw i16 [[X:%.*]] to i8
+; CHECK-NEXT:    [[YT:%.*]] = trunc nsw i16 [[Y:%.*]] to i8
+; CHECK-NEXT:    [[C:%.*]] = icmp slt i8 [[XT]], [[YT]]
+; CHECK-NEXT:    ret i1 [[C]]
+;
+  %xt = trunc nuw i16 %x to i8
+  %yt = trunc nsw i16 %y to i8
+  %c = icmp slt i8 %xt, %yt
+  ret i1 %c
+}
+
+define i1 @trunc_equality_nuw(i16 %x, i16 %y) {
+; CHECK-LABEL: @trunc_equality_nuw(
+; CHECK-NEXT:    [[XT:%.*]] = trunc nuw i16 [[X:%.*]] to i8
+; CHECK-NEXT:    [[YT:%.*]] = trunc nuw i16 [[Y:%.*]] to i8
+; CHECK-NEXT:    [[C:%.*]] = icmp eq i8 [[XT]], [[YT]]
+; CHECK-NEXT:    ret i1 [[C]]
+;
+  %xt = trunc nuw i16 %x to i8
+  %yt = trunc nuw i16 %y to i8
+  %c = icmp eq i8 %xt, %yt
+  ret i1 %c
+}
+
+define i1 @trunc_equality_nsw(i16 %x, i16 %y) {
+; CHECK-LABEL: @trunc_equality_nsw(
+; CHECK-NEXT:    [[XT:%.*]] = trunc nsw i16 [[X:%.*]] to i8
+; CHECK-NEXT:    [[YT:%.*]] = trunc nsw i16 [[Y:%.*]] to i8
+; CHECK-NEXT:    [[C:%.*]] = icmp eq i8 [[XT]], [[YT]]
+; CHECK-NEXT:    ret i1 [[C]]
+;
+  %xt = trunc nsw i16 %x to i8
+  %yt = trunc nsw i16 %y to i8
+  %c = icmp eq i8 %xt, %yt
+  ret i1 %c
+}
+
+define i1 @trunc_equality_both(i16 %x, i16 %y) {
+; CHECK-LABEL: @trunc_equality_both(
+; CHECK-NEXT:    [[XT:%.*]] = trunc nuw nsw i16 [[X:%.*]] to i8
+; CHECK-NEXT:    [[YT:%.*]] = trunc nuw nsw i16 [[Y:%.*]] to i8
+; CHECK-NEXT:    [[C:%.*]] = icmp eq i8 [[XT]], [[YT]]
+; CHECK-NEXT:    ret i1 [[C]]
+;
+  %xt = trunc nuw nsw i16 %x to i8
+  %yt = trunc nuw nsw i16 %y to i8
+  %c = icmp eq i8 %xt, %yt
+  ret i1 %c
+}
+
+define i1 @trunc_equality_either(i16 %x, i16 %y) {
+; CHECK-LABEL: @trunc_equality_either(
+; CHECK-NEXT:    [[XT:%.*]] = trunc nuw i16 [[X:%.*]] to i8
+; CHECK-NEXT:    [[YT:%.*]] = trunc nsw i16 [[Y:%.*]] to i8
+; CHECK-NEXT:    [[C:%.*]] = icmp eq i8 [[XT]], [[YT]]
+; CHECK-NEXT:    ret i1 [[C]]
+;
+  %xt = trunc nuw i16 %x to i8
+  %yt = trunc nsw i16 %y to i8
+  %c = icmp eq i8 %xt, %yt
+  ret i1 %c
+}
+
+define i1 @trunc_unsigned_nuw_zext(i32 %x, i8 %y) {
+; CHECK-LABEL: @trunc_unsigned_nuw_zext(
+; CHECK-NEXT:    [[XT:%.*]] = trunc nuw i32 [[X:%.*]] to i16
+; CHECK-NEXT:    [[YE:%.*]] = zext i8 [[Y:%.*]] to i16
+; CHECK-NEXT:    [[C:%.*]] = icmp ult i16 [[XT]], [[YE]]
+; CHECK-NEXT:    ret i1 [[C]]
+;
+  %xt = trunc nuw i32 %x to i16
+  %ye = zext i8 %y to i16
+  %c = icmp ult i16 %xt, %ye
+  ret i1 %c
+}
+
+define i1 @trunc_unsigned_nuw_sext(i32 %x, i8 %y) {
+; CHECK-LABEL: @trunc_unsigned_nuw_sext(
+; CHECK-NEXT:    [[XT:%.*]] = trunc nuw i32 [[X:%.*]] to i16
+; CHECK-NEXT:    [[YE:%.*]] = sext i8 [[Y:%.*]] to i16
+; CHECK-NEXT:    [[C:%.*]] = icmp ult i16 [[XT]], [[YE]]
+; CHECK-NEXT:    ret i1 [[C]]
+;
+  %xt = trunc nuw i32 %x to i16
+  %ye = sext i8 %y to i16
+  %c = icmp ult i16 %xt, %ye
+  ret i1 %c
+}
+
+define i1 @trunc_unsigned_nsw_zext(i32 %x, i8 %y) {
+; CHECK-LABEL: @trunc_unsigned_nsw_zext(
+; CHECK-NEXT:    [[XT:%.*]] = trunc nsw i32 [[X:%.*]] to i16
+; CHECK-NEXT:    [[YE:%.*]] = zext i8 [[Y:%.*]] to i16
+; CHECK-NEXT:    [[C:%.*]] = icmp ult i16 [[XT]], [[YE]]
+; CHECK-NEXT:    ret i1 [[C]]
+;
+  %xt = trunc nsw i32 %x to i16
+  %ye = zext i8 %y to i16
+  %c = icmp ult i16 %xt, %ye
+  ret i1 %c
+}
+
+define i1 @trunc_unsigned_nsw_sext(i32 %x, i8 %y) {
+; CHECK-LABEL: @trunc_unsigned_nsw_sext(
+; CHECK-NEXT:    [[XT:%.*]] = trunc nsw i32 [[X:%.*]] to i16
+; CHECK-NEXT:    [[YE:%.*]] = sext i8 [[Y:%.*]] to i16
+; CHECK-NEXT:    [[C:%.*]] = icmp ult i16 [[XT]], [[YE]]
+; CHECK-NEXT:    ret i1 [[C]]
+;
+  %xt = trunc nsw i32 %x to i16
+  %ye = sext i8 %y to i16
+  %c = icmp ult i16 %xt, %ye
+  ret i1 %c
+}
+
+define i1 @trunc_signed_nsw_sext(i32 %x, i8 %y) {
+; CHECK-LABEL: @trunc_signed_nsw_sext(
+; CHECK-NEXT:    [[XT:%.*]] = trunc nsw i32 [[X:%.*]] to i16
+; CHECK-NEXT:    [[YE:%.*]] = sext i8 [[Y:%.*]] to i16
+; CHECK-NEXT:    [[C:%.*]] = icmp slt i16 [[XT]], [[YE]]
+; CHECK-NEXT:    ret i1 [[C]]
+;
+  %xt = trunc nsw i32 %x to i16
+  %ye = sext i8 %y to i16
+  %c = icmp slt i16 %xt, %ye
+  ret i1 %c
+}
+
+define i1 @trunc_signed_nsw_zext(i32 %x, i8 %y) {
+; CHECK-LABEL: @trunc_signed_nsw_zext(
+; CHECK-NEXT:    [[XT:%.*]] = trunc nsw i32 [[X:%.*]] to i16
+; CHECK-NEXT:    [[YE:%.*]] = zext i8 [[Y:%.*]] to i16
+; CHECK-NEXT:    [[C:%.*]] = icmp slt i16 [[XT]], [[YE]]
+; CHECK-NEXT:    ret i1 [[C]]
+;
+  %xt = trunc nsw i32 %x to i16
+  %ye = zext i8 %y to i16
+  %c = icmp slt i16 %xt, %ye
+  ret i1 %c
+}
+
+define i1 @trunc_signed_nuw_sext(i32 %x, i8 %y) {
+; CHECK-LABEL: @trunc_signed_nuw_sext(
+; CHECK-NEXT:    [[XT:%.*]] = trunc nuw i32 [[X:%.*]] to i16
+; CHECK-NEXT:    [[YE:%.*]] = sext i8 [[Y:%.*]] to i16
+; CHECK-NEXT:    [[C:%.*]] = icmp slt i16 [[XT]], [[YE]]
+; CHECK-NEXT:    ret i1 [[C]]
+;
+  %xt = trunc nuw i32 %x to i16
+  %ye = sext i8 %y to i16
+  %c = icmp slt i16 %xt, %ye
+  ret i1 %c
+}
+
+define i1 @trunc_signed_nuw_zext(i32 %x, i8 %y) {
+; CHECK-LABEL: @trunc_signed_nuw_zext(
+; CHECK-NEXT:    [[XT:%.*]] = trunc nuw i32 [[X:%.*]] to i16
+; CHECK-NEXT:    [[YE:%.*]] = zext i8 [[Y:%.*]] to i16
+; CHECK-NEXT:    [[C:%.*]] = icmp slt i16 [[XT]], [[YE]]
+; CHECK-NEXT:    ret i1 [[C]]
+;
+  %xt = trunc nuw i32 %x to i16
+  %ye = zext i8 %y to i16
+  %c = icmp slt i16 %xt, %ye
+  ret i1 %c
+}
+
+define i1 @trunc_equality_nuw_zext(i32 %x, i8 %y) {
+; CHECK-LABEL: @trunc_equality_nuw_zext(
+; CHECK-NEXT:    [[XT:%.*]] = trunc nuw i32 [[X:%.*]] to i16
+; CHECK-NEXT:    [[YE:%.*]] = zext i8 [[Y:%.*]] to i16
+; CHECK-NEXT:    [[C:%.*]] = icmp ne i16 [[XT]], [[YE]]
+; CHECK-NEXT:    ret i1 [[C]]
+;
+  %xt = trunc nuw i32 %x to i16
+  %ye = zext i8 %y to i16
+  %c = icmp ne i16 %xt, %ye
+  ret i1 %c
+}
+
+define i1 @trunc_equality_nuw_sext(i32 %x, i8 %y) {
+; CHECK-LABEL: @trunc_equality_nuw_sext(
+; CHECK-NEXT:    [[XT:%.*]] = trunc nuw i32 [[X:%.*]] to i16
+; CHECK-NEXT:    [[YE:%.*]] = sext i8 [[Y:%.*]] to i16
+; CHECK-NEXT:    [[C:%.*]] = icmp ne i16 [[XT]], [[YE]]
+; CHECK-NEXT:    ret i1 [[C]]
+;
+  %xt = trunc nuw i32 %x to i16
+  %ye = sext i8 %y to i16
+  %c = icmp ne i16 %xt, %ye
+  ret i1 %c
+}
+
+define i1 @trunc_equality_nsw_zext(i32 %x, i8 %y) {
+; CHECK-LABEL: @trunc_equality_nsw_zext(
+; CHECK-NEXT:    [[XT:%.*]] = trunc nsw i32 [[X:%.*]] to i16
+; CHECK-NEXT:    [[YE:%.*]] = zext i8 [[Y:%.*]] to i16
+; CHECK-NEXT:    [[C:%.*]] = icmp ne i16 [[XT]], [[YE]]
+; CHECK-NEXT:    ret i1 [[C]]
+;
+  %xt = trunc nsw i32 %x to i16
+  %ye = zext i8 %y to i16
+  %c = icmp ne i16 %xt, %ye
+  ret i1 %c
+}
+
+define i1 @trunc_equality_nsw_sext(i32 %x, i8 %y) {
+; CHECK-LABEL: @trunc_equality_nsw_sext(
+; CHECK-NEXT:    [[XT:%.*]] = trunc nsw i32 [[X:%.*]] to i16
+; CHECK-NEXT:    [[YE:%.*]] = sext i8 [[Y:%.*]] to i16
+; CHECK-NEXT:    [[C:%.*]] = icmp ne i16 [[XT]], [[YE]]
+; CHECK-NEXT:    ret i1 [[C]]
+;
+  %xt = trunc nsw i32 %x to i16
+  %ye = sext i8 %y to i16
+  %c = icmp ne i16 %xt, %ye
+  ret i1 %c
+}
+
+define i1 @trunc_equality_both_sext(i32 %x, i8 %y) {
+; CHECK-LABEL: @trunc_equality_both_sext(
+; CHECK-NEXT:    [[XT:%.*]] = trunc nuw nsw i32 [[X:%.*]] to i16
+; CHECK-NEXT:    [[YE:%.*]] = sext i8 [[Y:%.*]] to i16
+; CHECK-NEXT:    [[C:%.*]] = icmp ne i16 [[XT]], [[YE]]
+; CHECK-NEXT:    ret i1 [[C]]
+;
+  %xt = trunc nuw nsw i32 %x to i16
+  %ye = sext i8 %y to i16
+  %c = icmp ne i16 %xt, %ye
+  ret i1 %c
+}
diff --git a/llvm/test/Transforms/InstCombine/intrinsic-select.ll b/llvm/test/Transforms/InstCombine/intrinsic-select.ll
index 1727d8f2758b..f110d7765830 100644
--- a/llvm/test/Transforms/InstCombine/intrinsic-select.ll
+++ b/llvm/test/Transforms/InstCombine/intrinsic-select.ll
@@ -26,7 +26,7 @@ define i32 @ctlz_sel_const_true_false(i1 %b) {
 
 define i32 @ctlz_sel_const_true(i1 %b, i32 %x) {
 ; CHECK-LABEL: @ctlz_sel_const_true(
-; CHECK-NEXT:    [[TMP1:%.*]] = call i32 @llvm.ctlz.i32(i32 [[X:%.*]], i1 false), !range [[RNG0:![0-9]+]]
+; CHECK-NEXT:    [[TMP1:%.*]] = call range(i32 0, 33) i32 @llvm.ctlz.i32(i32 [[X:%.*]], i1 false)
 ; CHECK-NEXT:    [[C:%.*]] = select i1 [[B:%.*]], i32 29, i32 [[TMP1]]
 ; CHECK-NEXT:    ret i32 [[C]]
 ;
@@ -37,7 +37,7 @@ define i32 @ctlz_sel_const_true(i1 %b, i32 %x) {
 
 define <3 x i17> @ctlz_sel_const_false(<3 x i1> %b, <3 x i17> %x) {
 ; CHECK-LABEL: @ctlz_sel_const_false(
-; CHECK-NEXT:    [[TMP1:%.*]] = call <3 x i17> @llvm.ctlz.v3i17(<3 x i17> [[X:%.*]], i1 true), !range [[RNG1:![0-9]+]]
+; CHECK-NEXT:    [[TMP1:%.*]] = call range(i17 0, 18) <3 x i17> @llvm.ctlz.v3i17(<3 x i17> [[X:%.*]], i1 true)
 ; CHECK-NEXT:    [[C:%.*]] = select <3 x i1> [[B:%.*]], <3 x i17> [[TMP1]], <3 x i17> <i17 14, i17 0, i17 poison>
 ; CHECK-NEXT:    ret <3 x i17> [[C]]
 ;
@@ -50,7 +50,7 @@ define i32 @ctlz_sel_const_true_false_extra_use(i1 %b) {
 ; CHECK-LABEL: @ctlz_sel_const_true_false_extra_use(
 ; CHECK-NEXT:    [[S:%.*]] = select i1 [[B:%.*]], i32 -1, i32 7
 ; CHECK-NEXT:    call void @use(i32 [[S]])
-; CHECK-NEXT:    [[C:%.*]] = call i32 @llvm.ctlz.i32(i32 [[S]], i1 true), !range [[RNG2:![0-9]+]]
+; CHECK-NEXT:    [[C:%.*]] = call range(i32 0, 30) i32 @llvm.ctlz.i32(i32 [[S]], i1 true)
 ; CHECK-NEXT:    ret i32 [[C]]
 ;
   %s = select i1 %b, i32 -1, i32 7
@@ -71,7 +71,7 @@ define i32 @cttz_sel_const_true_false(i1 %b) {
 
 define i32 @cttz_sel_const_true(i1 %b, i32 %x) {
 ; CHECK-LABEL: @cttz_sel_const_true(
-; CHECK-NEXT:    [[TMP1:%.*]] = call i32 @llvm.cttz.i32(i32 [[X:%.*]], i1 true), !range [[RNG0]]
+; CHECK-NEXT:    [[TMP1:%.*]] = call range(i32 0, 33) i32 @llvm.cttz.i32(i32 [[X:%.*]], i1 true)
 ; CHECK-NEXT:    [[C:%.*]] = select i1 [[B:%.*]], i32 0, i32 [[TMP1]]
 ; CHECK-NEXT:    ret i32 [[C]]
 ;
@@ -82,7 +82,7 @@ define i32 @cttz_sel_const_true(i1 %b, i32 %x) {
 
 define <3 x i5> @cttz_sel_const_false(<3 x i1> %b, <3 x i5> %x) {
 ; CHECK-LABEL: @cttz_sel_const_false(
-; CHECK-NEXT:    [[TMP1:%.*]] = call <3 x i5> @llvm.cttz.v3i5(<3 x i5> [[X:%.*]], i1 false), !range [[RNG3:![0-9]+]]
+; CHECK-NEXT:    [[TMP1:%.*]] = call range(i5 0, 6) <3 x i5> @llvm.cttz.v3i5(<3 x i5> [[X:%.*]], i1 false)
 ; CHECK-NEXT:    [[C:%.*]] = select <3 x i1> [[B:%.*]], <3 x i5> [[TMP1]], <3 x i5> <i5 0, i5 0, i5 5>
 ; CHECK-NEXT:    ret <3 x i5> [[C]]
 ;
@@ -95,7 +95,7 @@ define i32 @cttz_sel_const_true_false_extra_use(i1 %b) {
 ; CHECK-LABEL: @cttz_sel_const_true_false_extra_use(
 ; CHECK-NEXT:    [[S:%.*]] = select i1 [[B:%.*]], i32 5, i32 -8
 ; CHECK-NEXT:    call void @use(i32 [[S]])
-; CHECK-NEXT:    [[C:%.*]] = call i32 @llvm.cttz.i32(i32 [[S]], i1 true), !range [[RNG0]]
+; CHECK-NEXT:    [[C:%.*]] = call range(i32 0, 33) i32 @llvm.cttz.i32(i32 [[S]], i1 true)
 ; CHECK-NEXT:    ret i32 [[C]]
 ;
   %s = select i1 %b, i32 5, i32 -8
@@ -116,7 +116,7 @@ define i32 @ctpop_sel_const_true_false(i1 %b) {
 
 define i32 @ctpop_sel_const_true(i1 %b, i32 %x) {
 ; CHECK-LABEL: @ctpop_sel_const_true(
-; CHECK-NEXT:    [[TMP1:%.*]] = call i32 @llvm.ctpop.i32(i32 [[X:%.*]]), !range [[RNG0]]
+; CHECK-NEXT:    [[TMP1:%.*]] = call range(i32 0, 33) i32 @llvm.ctpop.i32(i32 [[X:%.*]])
 ; CHECK-NEXT:    [[C:%.*]] = select i1 [[B:%.*]], i32 2, i32 [[TMP1]]
 ; CHECK-NEXT:    ret i32 [[C]]
 ;
@@ -127,7 +127,7 @@ define i32 @ctpop_sel_const_true(i1 %b, i32 %x) {
 
 define <3 x i7> @ctpop_sel_const_false(<3 x i1> %b, <3 x i7> %x) {
 ; CHECK-LABEL: @ctpop_sel_const_false(
-; CHECK-NEXT:    [[TMP1:%.*]] = call <3 x i7> @llvm.ctpop.v3i7(<3 x i7> [[X:%.*]]), !range [[RNG4:![0-9]+]]
+; CHECK-NEXT:    [[TMP1:%.*]] = call range(i7 0, 8) <3 x i7> @llvm.ctpop.v3i7(<3 x i7> [[X:%.*]])
 ; CHECK-NEXT:    [[C:%.*]] = select <3 x i1> [[B:%.*]], <3 x i7> [[TMP1]], <3 x i7> <i7 3, i7 7, i7 0>
 ; CHECK-NEXT:    ret <3 x i7> [[C]]
 ;
@@ -140,7 +140,7 @@ define i32 @ctpop_sel_const_true_false_extra_use(i1 %b) {
 ; CHECK-LABEL: @ctpop_sel_const_true_false_extra_use(
 ; CHECK-NEXT:    [[S:%.*]] = select i1 [[B:%.*]], i32 5, i32 7
 ; CHECK-NEXT:    call void @use(i32 [[S]])
-; CHECK-NEXT:    [[C:%.*]] = call i32 @llvm.ctpop.i32(i32 [[S]]), !range [[RNG5:![0-9]+]]
+; CHECK-NEXT:    [[C:%.*]] = call range(i32 2, 4) i32 @llvm.ctpop.i32(i32 [[S]])
 ; CHECK-NEXT:    ret i32 [[C]]
 ;
   %s = select i1 %b, i32 5, i32 7
diff --git a/llvm/test/Transforms/InstCombine/intrinsics.ll b/llvm/test/Transforms/InstCombine/intrinsics.ll
index d90b0ebd400c..c8d70e17cd39 100644
--- a/llvm/test/Transforms/InstCombine/intrinsics.ll
+++ b/llvm/test/Transforms/InstCombine/intrinsics.ll
@@ -139,7 +139,7 @@ define <vscale x 1 x i1> @cttz_knownbits_scalable_vec(<vscale x 1 x i32> %arg) {
 define i32 @cttz_knownbits2(i32 %arg) {
 ; CHECK-LABEL: @cttz_knownbits2(
 ; CHECK-NEXT:    [[OR:%.*]] = or i32 [[ARG:%.*]], 4
-; CHECK-NEXT:    [[CNT:%.*]] = call i32 @llvm.cttz.i32(i32 [[OR]], i1 true) #[[ATTR2:[0-9]+]], !range [[RNG0:![0-9]+]]
+; CHECK-NEXT:    [[CNT:%.*]] = call range(i32 0, 3) i32 @llvm.cttz.i32(i32 [[OR]], i1 true) #[[ATTR2:[0-9]+]]
 ; CHECK-NEXT:    ret i32 [[CNT]]
 ;
   %or = or i32 %arg, 4
@@ -150,7 +150,7 @@ define i32 @cttz_knownbits2(i32 %arg) {
 define <2 x i32> @cttz_knownbits2_vec(<2 x i32> %arg) {
 ; CHECK-LABEL: @cttz_knownbits2_vec(
 ; CHECK-NEXT:    [[OR:%.*]] = or <2 x i32> [[ARG:%.*]], <i32 4, i32 4>
-; CHECK-NEXT:    [[CNT:%.*]] = call <2 x i32> @llvm.cttz.v2i32(<2 x i32> [[OR]], i1 true) #[[ATTR2]], !range [[RNG0]]
+; CHECK-NEXT:    [[CNT:%.*]] = call range(i32 0, 3) <2 x i32> @llvm.cttz.v2i32(<2 x i32> [[OR]], i1 true) #[[ATTR2]]
 ; CHECK-NEXT:    ret <2 x i32> [[CNT]]
 ;
   %or = or <2 x i32> %arg, <i32 4, i32 4>
@@ -256,7 +256,7 @@ define <2 x i1> @ctlz_knownbits_vec(<2 x i8> %arg) {
 define i8 @ctlz_knownbits2(i8 %arg) {
 ; CHECK-LABEL: @ctlz_knownbits2(
 ; CHECK-NEXT:    [[OR:%.*]] = or i8 [[ARG:%.*]], 32
-; CHECK-NEXT:    [[CNT:%.*]] = call i8 @llvm.ctlz.i8(i8 [[OR]], i1 true) #[[ATTR2]], !range [[RNG1:![0-9]+]]
+; CHECK-NEXT:    [[CNT:%.*]] = call range(i8 0, 3) i8 @llvm.ctlz.i8(i8 [[OR]], i1 true) #[[ATTR2]]
 ; CHECK-NEXT:    ret i8 [[CNT]]
 ;
   %or = or i8 %arg, 32
@@ -267,7 +267,7 @@ define i8 @ctlz_knownbits2(i8 %arg) {
 define <2 x i8> @ctlz_knownbits2_vec(<2 x i8> %arg) {
 ; CHECK-LABEL: @ctlz_knownbits2_vec(
 ; CHECK-NEXT:    [[OR:%.*]] = or <2 x i8> [[ARG:%.*]], <i8 32, i8 32>
-; CHECK-NEXT:    [[CNT:%.*]] = call <2 x i8> @llvm.ctlz.v2i8(<2 x i8> [[OR]], i1 true) #[[ATTR2]], !range [[RNG1]]
+; CHECK-NEXT:    [[CNT:%.*]] = call range(i8 0, 3) <2 x i8> @llvm.ctlz.v2i8(<2 x i8> [[OR]], i1 true) #[[ATTR2]]
 ; CHECK-NEXT:    ret <2 x i8> [[CNT]]
 ;
   %or = or <2 x i8> %arg, <i8 32, i8 32>
@@ -314,7 +314,7 @@ define <2 x i32> @ctlz_poison_vec(<2 x i32> %Value) {
 define i32 @ctlz_no_zero(i32 %a) {
 ; CHECK-LABEL: @ctlz_no_zero(
 ; CHECK-NEXT:    [[OR:%.*]] = or i32 [[A:%.*]], 8
-; CHECK-NEXT:    [[CTLZ:%.*]] = tail call i32 @llvm.ctlz.i32(i32 [[OR]], i1 true), !range [[RNG2:![0-9]+]]
+; CHECK-NEXT:    [[CTLZ:%.*]] = tail call range(i32 0, 29) i32 @llvm.ctlz.i32(i32 [[OR]], i1 true)
 ; CHECK-NEXT:    ret i32 [[CTLZ]]
 ;
   %or = or i32 %a, 8
@@ -325,7 +325,7 @@ define i32 @ctlz_no_zero(i32 %a) {
 define <2 x i32> @ctlz_no_zero_vec(<2 x i32> %a) {
 ; CHECK-LABEL: @ctlz_no_zero_vec(
 ; CHECK-NEXT:    [[OR:%.*]] = or <2 x i32> [[A:%.*]], <i32 8, i32 8>
-; CHECK-NEXT:    [[CTLZ:%.*]] = tail call <2 x i32> @llvm.ctlz.v2i32(<2 x i32> [[OR]], i1 true), !range [[RNG2]]
+; CHECK-NEXT:    [[CTLZ:%.*]] = tail call range(i32 0, 29) <2 x i32> @llvm.ctlz.v2i32(<2 x i32> [[OR]], i1 true)
 ; CHECK-NEXT:    ret <2 x i32> [[CTLZ]]
 ;
   %or = or <2 x i32> %a, <i32 8, i32 8>
@@ -352,7 +352,7 @@ define <2 x i32> @cttz_poison_vec(<2 x i32> %Value) {
 define i32 @cttz_no_zero(i32 %a) {
 ; CHECK-LABEL: @cttz_no_zero(
 ; CHECK-NEXT:    [[OR:%.*]] = or i32 [[A:%.*]], 8
-; CHECK-NEXT:    [[CTTZ:%.*]] = tail call i32 @llvm.cttz.i32(i32 [[OR]], i1 true), !range [[RNG3:![0-9]+]]
+; CHECK-NEXT:    [[CTTZ:%.*]] = tail call range(i32 0, 4) i32 @llvm.cttz.i32(i32 [[OR]], i1 true)
 ; CHECK-NEXT:    ret i32 [[CTTZ]]
 ;
   %or = or i32 %a, 8
@@ -363,7 +363,7 @@ define i32 @cttz_no_zero(i32 %a) {
 define <2 x i32> @cttz_no_zero_vec(<2 x i32> %a) {
 ; CHECK-LABEL: @cttz_no_zero_vec(
 ; CHECK-NEXT:    [[OR:%.*]] = or <2 x i32> [[A:%.*]], <i32 8, i32 8>
-; CHECK-NEXT:    [[CTTZ:%.*]] = tail call <2 x i32> @llvm.cttz.v2i32(<2 x i32> [[OR]], i1 true), !range [[RNG3]]
+; CHECK-NEXT:    [[CTTZ:%.*]] = tail call range(i32 0, 4) <2 x i32> @llvm.cttz.v2i32(<2 x i32> [[OR]], i1 true)
 ; CHECK-NEXT:    ret <2 x i32> [[CTTZ]]
 ;
   %or = or <2 x i32> %a, <i32 8, i32 8>
@@ -373,7 +373,7 @@ define <2 x i32> @cttz_no_zero_vec(<2 x i32> %a) {
 
 define i32 @ctlz_select(i32 %Value) nounwind {
 ; CHECK-LABEL: @ctlz_select(
-; CHECK-NEXT:    [[CTLZ:%.*]] = call i32 @llvm.ctlz.i32(i32 [[VALUE:%.*]], i1 false), !range [[RNG4:![0-9]+]]
+; CHECK-NEXT:    [[CTLZ:%.*]] = call range(i32 0, 33) i32 @llvm.ctlz.i32(i32 [[VALUE:%.*]], i1 false)
 ; CHECK-NEXT:    ret i32 [[CTLZ]]
 ;
   %tobool = icmp ne i32 %Value, 0
@@ -384,7 +384,7 @@ define i32 @ctlz_select(i32 %Value) nounwind {
 
 define <2 x i32> @ctlz_select_vec(<2 x i32> %Value) nounwind {
 ; CHECK-LABEL: @ctlz_select_vec(
-; CHECK-NEXT:    [[CTLZ:%.*]] = call <2 x i32> @llvm.ctlz.v2i32(<2 x i32> [[VALUE:%.*]], i1 false), !range [[RNG4]]
+; CHECK-NEXT:    [[CTLZ:%.*]] = call range(i32 0, 33) <2 x i32> @llvm.ctlz.v2i32(<2 x i32> [[VALUE:%.*]], i1 false)
 ; CHECK-NEXT:    ret <2 x i32> [[CTLZ]]
 ;
   %tobool = icmp ne <2 x i32> %Value, zeroinitializer
@@ -395,7 +395,7 @@ define <2 x i32> @ctlz_select_vec(<2 x i32> %Value) nounwind {
 
 define i32 @cttz_select(i32 %Value) nounwind {
 ; CHECK-LABEL: @cttz_select(
-; CHECK-NEXT:    [[CTTZ:%.*]] = call i32 @llvm.cttz.i32(i32 [[VALUE:%.*]], i1 false), !range [[RNG4]]
+; CHECK-NEXT:    [[CTTZ:%.*]] = call range(i32 0, 33) i32 @llvm.cttz.i32(i32 [[VALUE:%.*]], i1 false)
 ; CHECK-NEXT:    ret i32 [[CTTZ]]
 ;
   %tobool = icmp ne i32 %Value, 0
@@ -406,7 +406,7 @@ define i32 @cttz_select(i32 %Value) nounwind {
 
 define <2 x i32> @cttz_select_vec(<2 x i32> %Value) nounwind {
 ; CHECK-LABEL: @cttz_select_vec(
-; CHECK-NEXT:    [[CTTZ:%.*]] = call <2 x i32> @llvm.cttz.v2i32(<2 x i32> [[VALUE:%.*]], i1 false), !range [[RNG4]]
+; CHECK-NEXT:    [[CTTZ:%.*]] = call range(i32 0, 33) <2 x i32> @llvm.cttz.v2i32(<2 x i32> [[VALUE:%.*]], i1 false)
 ; CHECK-NEXT:    ret <2 x i32> [[CTTZ]]
 ;
   %tobool = icmp ne <2 x i32> %Value, zeroinitializer
diff --git a/llvm/test/Transforms/InstCombine/ispow2.ll b/llvm/test/Transforms/InstCombine/ispow2.ll
index cc50c5cd1e66..a143b1347cce 100644
--- a/llvm/test/Transforms/InstCombine/ispow2.ll
+++ b/llvm/test/Transforms/InstCombine/ispow2.ll
@@ -3,7 +3,7 @@
 
 define i1 @is_pow2or0_negate_op(i32 %x) {
 ; CHECK-LABEL: @is_pow2or0_negate_op(
-; CHECK-NEXT:    [[TMP1:%.*]] = call i32 @llvm.ctpop.i32(i32 [[X:%.*]]), !range [[RNG0:![0-9]+]]
+; CHECK-NEXT:    [[TMP1:%.*]] = call range(i32 0, 33) i32 @llvm.ctpop.i32(i32 [[X:%.*]])
 ; CHECK-NEXT:    [[CMP:%.*]] = icmp ult i32 [[TMP1]], 2
 ; CHECK-NEXT:    ret i1 [[CMP]]
 ;
@@ -15,7 +15,7 @@ define i1 @is_pow2or0_negate_op(i32 %x) {
 
 define <2 x i1> @is_pow2or0_negate_op_vec(<2 x i32> %x) {
 ; CHECK-LABEL: @is_pow2or0_negate_op_vec(
-; CHECK-NEXT:    [[TMP1:%.*]] = call <2 x i32> @llvm.ctpop.v2i32(<2 x i32> [[X:%.*]]), !range [[RNG0]]
+; CHECK-NEXT:    [[TMP1:%.*]] = call range(i32 0, 33) <2 x i32> @llvm.ctpop.v2i32(<2 x i32> [[X:%.*]])
 ; CHECK-NEXT:    [[CMP:%.*]] = icmp ult <2 x i32> [[TMP1]], <i32 2, i32 2>
 ; CHECK-NEXT:    ret <2 x i1> [[CMP]]
 ;
@@ -27,7 +27,7 @@ define <2 x i1> @is_pow2or0_negate_op_vec(<2 x i32> %x) {
 
 define i1 @is_pow2or0_decrement_op(i8 %x) {
 ; CHECK-LABEL: @is_pow2or0_decrement_op(
-; CHECK-NEXT:    [[TMP1:%.*]] = call i8 @llvm.ctpop.i8(i8 [[X:%.*]]), !range [[RNG1:![0-9]+]]
+; CHECK-NEXT:    [[TMP1:%.*]] = call range(i8 0, 9) i8 @llvm.ctpop.i8(i8 [[X:%.*]])
 ; CHECK-NEXT:    [[CMP:%.*]] = icmp ult i8 [[TMP1]], 2
 ; CHECK-NEXT:    ret i1 [[CMP]]
 ;
@@ -39,7 +39,7 @@ define i1 @is_pow2or0_decrement_op(i8 %x) {
 
 define <2 x i1> @is_pow2or0_decrement_op_vec(<2 x i8> %x) {
 ; CHECK-LABEL: @is_pow2or0_decrement_op_vec(
-; CHECK-NEXT:    [[TMP1:%.*]] = call <2 x i8> @llvm.ctpop.v2i8(<2 x i8> [[X:%.*]]), !range [[RNG1]]
+; CHECK-NEXT:    [[TMP1:%.*]] = call range(i8 0, 9) <2 x i8> @llvm.ctpop.v2i8(<2 x i8> [[X:%.*]])
 ; CHECK-NEXT:    [[CMP:%.*]] = icmp ult <2 x i8> [[TMP1]], <i8 2, i8 2>
 ; CHECK-NEXT:    ret <2 x i1> [[CMP]]
 ;
@@ -51,7 +51,7 @@ define <2 x i1> @is_pow2or0_decrement_op_vec(<2 x i8> %x) {
 
 define i1 @isnot_pow2or0_negate_op(i32 %x) {
 ; CHECK-LABEL: @isnot_pow2or0_negate_op(
-; CHECK-NEXT:    [[TMP1:%.*]] = call i32 @llvm.ctpop.i32(i32 [[X:%.*]]), !range [[RNG0]]
+; CHECK-NEXT:    [[TMP1:%.*]] = call range(i32 0, 33) i32 @llvm.ctpop.i32(i32 [[X:%.*]])
 ; CHECK-NEXT:    [[CMP:%.*]] = icmp ugt i32 [[TMP1]], 1
 ; CHECK-NEXT:    ret i1 [[CMP]]
 ;
@@ -63,7 +63,7 @@ define i1 @isnot_pow2or0_negate_op(i32 %x) {
 
 define <2 x i1> @isnot_pow2or0_negate_op_vec(<2 x i32> %x) {
 ; CHECK-LABEL: @isnot_pow2or0_negate_op_vec(
-; CHECK-NEXT:    [[TMP1:%.*]] = call <2 x i32> @llvm.ctpop.v2i32(<2 x i32> [[X:%.*]]), !range [[RNG0]]
+; CHECK-NEXT:    [[TMP1:%.*]] = call range(i32 0, 33) <2 x i32> @llvm.ctpop.v2i32(<2 x i32> [[X:%.*]])
 ; CHECK-NEXT:    [[CMP:%.*]] = icmp ugt <2 x i32> [[TMP1]], <i32 1, i32 1>
 ; CHECK-NEXT:    ret <2 x i1> [[CMP]]
 ;
@@ -75,7 +75,7 @@ define <2 x i1> @isnot_pow2or0_negate_op_vec(<2 x i32> %x) {
 
 define i1 @isnot_pow2or0_decrement_op(i8 %x) {
 ; CHECK-LABEL: @isnot_pow2or0_decrement_op(
-; CHECK-NEXT:    [[TMP1:%.*]] = call i8 @llvm.ctpop.i8(i8 [[X:%.*]]), !range [[RNG1]]
+; CHECK-NEXT:    [[TMP1:%.*]] = call range(i8 0, 9) i8 @llvm.ctpop.i8(i8 [[X:%.*]])
 ; CHECK-NEXT:    [[CMP:%.*]] = icmp ugt i8 [[TMP1]], 1
 ; CHECK-NEXT:    ret i1 [[CMP]]
 ;
@@ -87,7 +87,7 @@ define i1 @isnot_pow2or0_decrement_op(i8 %x) {
 
 define <2 x i1> @isnot_pow2or0_decrement_op_vec(<2 x i8> %x) {
 ; CHECK-LABEL: @isnot_pow2or0_decrement_op_vec(
-; CHECK-NEXT:    [[TMP1:%.*]] = call <2 x i8> @llvm.ctpop.v2i8(<2 x i8> [[X:%.*]]), !range [[RNG1]]
+; CHECK-NEXT:    [[TMP1:%.*]] = call range(i8 0, 9) <2 x i8> @llvm.ctpop.v2i8(<2 x i8> [[X:%.*]])
 ; CHECK-NEXT:    [[CMP:%.*]] = icmp ugt <2 x i8> [[TMP1]], <i8 1, i8 1>
 ; CHECK-NEXT:    ret <2 x i1> [[CMP]]
 ;
@@ -100,7 +100,7 @@ define <2 x i1> @isnot_pow2or0_decrement_op_vec(<2 x i8> %x) {
 define i1 @is_pow2or0_negate_op_commute1(i32 %p) {
 ; CHECK-LABEL: @is_pow2or0_negate_op_commute1(
 ; CHECK-NEXT:    [[X:%.*]] = srem i32 42, [[P:%.*]]
-; CHECK-NEXT:    [[TMP1:%.*]] = call i32 @llvm.ctpop.i32(i32 [[X]]), !range [[RNG2:![0-9]+]]
+; CHECK-NEXT:    [[TMP1:%.*]] = call range(i32 0, 7) i32 @llvm.ctpop.i32(i32 [[X]])
 ; CHECK-NEXT:    [[CMP:%.*]] = icmp ult i32 [[TMP1]], 2
 ; CHECK-NEXT:    ret i1 [[CMP]]
 ;
@@ -116,7 +116,7 @@ define i1 @is_pow2or0_negate_op_commute1(i32 %p) {
 define i1 @isnot_pow2or0_negate_op_commute2(i32 %p) {
 ; CHECK-LABEL: @isnot_pow2or0_negate_op_commute2(
 ; CHECK-NEXT:    [[X:%.*]] = urem i32 42, [[P:%.*]]
-; CHECK-NEXT:    [[TMP1:%.*]] = call i32 @llvm.ctpop.i32(i32 [[X]]), !range [[RNG2]]
+; CHECK-NEXT:    [[TMP1:%.*]] = call range(i32 0, 7) i32 @llvm.ctpop.i32(i32 [[X]])
 ; CHECK-NEXT:    [[CMP:%.*]] = icmp ugt i32 [[TMP1]], 1
 ; CHECK-NEXT:    ret i1 [[CMP]]
 ;
@@ -130,7 +130,7 @@ define i1 @isnot_pow2or0_negate_op_commute2(i32 %p) {
 define i1 @isnot_pow2or0_negate_op_commute3(i32 %p) {
 ; CHECK-LABEL: @isnot_pow2or0_negate_op_commute3(
 ; CHECK-NEXT:    [[X:%.*]] = urem i32 42, [[P:%.*]]
-; CHECK-NEXT:    [[TMP1:%.*]] = call i32 @llvm.ctpop.i32(i32 [[X]]), !range [[RNG2]]
+; CHECK-NEXT:    [[TMP1:%.*]] = call range(i32 0, 7) i32 @llvm.ctpop.i32(i32 [[X]])
 ; CHECK-NEXT:    [[CMP:%.*]] = icmp ugt i32 [[TMP1]], 1
 ; CHECK-NEXT:    ret i1 [[CMP]]
 ;
@@ -147,7 +147,7 @@ define i1 @is_pow2or0_negate_op_extra_use1(i32 %x) {
 ; CHECK-LABEL: @is_pow2or0_negate_op_extra_use1(
 ; CHECK-NEXT:    [[NEG:%.*]] = sub i32 0, [[X:%.*]]
 ; CHECK-NEXT:    call void @use(i32 [[NEG]])
-; CHECK-NEXT:    [[TMP1:%.*]] = call i32 @llvm.ctpop.i32(i32 [[X]]), !range [[RNG0]]
+; CHECK-NEXT:    [[TMP1:%.*]] = call range(i32 0, 33) i32 @llvm.ctpop.i32(i32 [[X]])
 ; CHECK-NEXT:    [[CMP:%.*]] = icmp ult i32 [[TMP1]], 2
 ; CHECK-NEXT:    ret i1 [[CMP]]
 ;
@@ -181,7 +181,7 @@ declare void @llvm.assume(i1)
 
 define i1 @is_pow2_ctpop(i32 %x) {
 ; CHECK-LABEL: @is_pow2_ctpop(
-; CHECK-NEXT:    [[T0:%.*]] = tail call i32 @llvm.ctpop.i32(i32 [[X:%.*]]), !range [[RNG0]]
+; CHECK-NEXT:    [[T0:%.*]] = tail call range(i32 0, 33) i32 @llvm.ctpop.i32(i32 [[X:%.*]])
 ; CHECK-NEXT:    [[R:%.*]] = icmp eq i32 [[T0]], 1
 ; CHECK-NEXT:    ret i1 [[R]]
 ;
@@ -197,7 +197,7 @@ define i1 @is_pow2_non_zero_ult_2(i32 %x) {
 ; CHECK-LABEL: @is_pow2_non_zero_ult_2(
 ; CHECK-NEXT:    [[NOTZERO:%.*]] = icmp ne i32 [[X:%.*]], 0
 ; CHECK-NEXT:    call void @llvm.assume(i1 [[NOTZERO]])
-; CHECK-NEXT:    [[T0:%.*]] = tail call i32 @llvm.ctpop.i32(i32 [[X]]), !range [[RNG0]]
+; CHECK-NEXT:    [[T0:%.*]] = tail call range(i32 0, 33) i32 @llvm.ctpop.i32(i32 [[X]])
 ; CHECK-NEXT:    [[CMP:%.*]] = icmp ult i32 [[T0]], 2
 ; CHECK-NEXT:    ret i1 [[CMP]]
 ;
@@ -212,7 +212,7 @@ define i1 @is_pow2_non_zero_eq_1(i32 %x) {
 ; CHECK-LABEL: @is_pow2_non_zero_eq_1(
 ; CHECK-NEXT:    [[NOTZERO:%.*]] = icmp ne i32 [[X:%.*]], 0
 ; CHECK-NEXT:    call void @llvm.assume(i1 [[NOTZERO]])
-; CHECK-NEXT:    [[T0:%.*]] = tail call i32 @llvm.ctpop.i32(i32 [[X]]), !range [[RNG0]]
+; CHECK-NEXT:    [[T0:%.*]] = tail call range(i32 0, 33) i32 @llvm.ctpop.i32(i32 [[X]])
 ; CHECK-NEXT:    [[CMP:%.*]] = icmp eq i32 [[T0]], 1
 ; CHECK-NEXT:    ret i1 [[CMP]]
 ;
@@ -227,7 +227,7 @@ define i1 @is_pow2_non_zero_ugt_1(i32 %x) {
 ; CHECK-LABEL: @is_pow2_non_zero_ugt_1(
 ; CHECK-NEXT:    [[NOTZERO:%.*]] = icmp ne i32 [[X:%.*]], 0
 ; CHECK-NEXT:    call void @llvm.assume(i1 [[NOTZERO]])
-; CHECK-NEXT:    [[T0:%.*]] = tail call i32 @llvm.ctpop.i32(i32 [[X]]), !range [[RNG0]]
+; CHECK-NEXT:    [[T0:%.*]] = tail call range(i32 0, 33) i32 @llvm.ctpop.i32(i32 [[X]])
 ; CHECK-NEXT:    [[CMP:%.*]] = icmp ugt i32 [[T0]], 1
 ; CHECK-NEXT:    ret i1 [[CMP]]
 ;
@@ -242,7 +242,7 @@ define i1 @is_pow2_non_zero_ne_1(i32 %x) {
 ; CHECK-LABEL: @is_pow2_non_zero_ne_1(
 ; CHECK-NEXT:    [[NOTZERO:%.*]] = icmp ne i32 [[X:%.*]], 0
 ; CHECK-NEXT:    call void @llvm.assume(i1 [[NOTZERO]])
-; CHECK-NEXT:    [[T0:%.*]] = tail call i32 @llvm.ctpop.i32(i32 [[X]]), !range [[RNG0]]
+; CHECK-NEXT:    [[T0:%.*]] = tail call range(i32 0, 33) i32 @llvm.ctpop.i32(i32 [[X]])
 ; CHECK-NEXT:    [[CMP:%.*]] = icmp ne i32 [[T0]], 1
 ; CHECK-NEXT:    ret i1 [[CMP]]
 ;
@@ -255,7 +255,7 @@ define i1 @is_pow2_non_zero_ne_1(i32 %x) {
 
 define i1 @is_pow2_ctpop_logical(i32 %x) {
 ; CHECK-LABEL: @is_pow2_ctpop_logical(
-; CHECK-NEXT:    [[T0:%.*]] = tail call i32 @llvm.ctpop.i32(i32 [[X:%.*]]), !range [[RNG0]]
+; CHECK-NEXT:    [[T0:%.*]] = tail call range(i32 0, 33) i32 @llvm.ctpop.i32(i32 [[X:%.*]])
 ; CHECK-NEXT:    [[R:%.*]] = icmp eq i32 [[T0]], 1
 ; CHECK-NEXT:    ret i1 [[R]]
 ;
@@ -271,7 +271,7 @@ declare void @use_i1(i1)
 
 define i1 @is_pow2_ctpop_extra_uses(i32 %x) {
 ; CHECK-LABEL: @is_pow2_ctpop_extra_uses(
-; CHECK-NEXT:    [[T0:%.*]] = tail call i32 @llvm.ctpop.i32(i32 [[X:%.*]]), !range [[RNG0]]
+; CHECK-NEXT:    [[T0:%.*]] = tail call range(i32 0, 33) i32 @llvm.ctpop.i32(i32 [[X:%.*]])
 ; CHECK-NEXT:    [[CMP:%.*]] = icmp ult i32 [[T0]], 2
 ; CHECK-NEXT:    call void @use_i1(i1 [[CMP]])
 ; CHECK-NEXT:    [[NOTZERO:%.*]] = icmp ne i32 [[X]], 0
@@ -290,7 +290,7 @@ define i1 @is_pow2_ctpop_extra_uses(i32 %x) {
 
 define i1 @is_pow2_ctpop_extra_uses_logical(i32 %x) {
 ; CHECK-LABEL: @is_pow2_ctpop_extra_uses_logical(
-; CHECK-NEXT:    [[T0:%.*]] = tail call i32 @llvm.ctpop.i32(i32 [[X:%.*]]), !range [[RNG0]]
+; CHECK-NEXT:    [[T0:%.*]] = tail call range(i32 0, 33) i32 @llvm.ctpop.i32(i32 [[X:%.*]])
 ; CHECK-NEXT:    [[CMP:%.*]] = icmp ult i32 [[T0]], 2
 ; CHECK-NEXT:    call void @use_i1(i1 [[CMP]])
 ; CHECK-NEXT:    [[NOTZERO:%.*]] = icmp ne i32 [[X]], 0
@@ -311,7 +311,7 @@ define i1 @is_pow2_ctpop_extra_uses_logical(i32 %x) {
 
 define <2 x i1> @is_pow2_ctpop_commute_vec(<2 x i8> %x) {
 ; CHECK-LABEL: @is_pow2_ctpop_commute_vec(
-; CHECK-NEXT:    [[T0:%.*]] = tail call <2 x i8> @llvm.ctpop.v2i8(<2 x i8> [[X:%.*]]), !range [[RNG1]]
+; CHECK-NEXT:    [[T0:%.*]] = tail call range(i8 0, 9) <2 x i8> @llvm.ctpop.v2i8(<2 x i8> [[X:%.*]])
 ; CHECK-NEXT:    [[R:%.*]] = icmp eq <2 x i8> [[T0]], <i8 1, i8 1>
 ; CHECK-NEXT:    ret <2 x i1> [[R]]
 ;
@@ -326,7 +326,7 @@ define <2 x i1> @is_pow2_ctpop_commute_vec(<2 x i8> %x) {
 
 define i1 @is_pow2_ctpop_wrong_cmp_op1(i32 %x) {
 ; CHECK-LABEL: @is_pow2_ctpop_wrong_cmp_op1(
-; CHECK-NEXT:    [[T0:%.*]] = tail call i32 @llvm.ctpop.i32(i32 [[X:%.*]]), !range [[RNG0]]
+; CHECK-NEXT:    [[T0:%.*]] = tail call range(i32 0, 33) i32 @llvm.ctpop.i32(i32 [[X:%.*]])
 ; CHECK-NEXT:    [[CMP:%.*]] = icmp ult i32 [[T0]], 3
 ; CHECK-NEXT:    [[NOTZERO:%.*]] = icmp ne i32 [[X]], 0
 ; CHECK-NEXT:    [[R:%.*]] = and i1 [[NOTZERO]], [[CMP]]
@@ -341,7 +341,7 @@ define i1 @is_pow2_ctpop_wrong_cmp_op1(i32 %x) {
 
 define i1 @is_pow2_ctpop_wrong_cmp_op1_logical(i32 %x) {
 ; CHECK-LABEL: @is_pow2_ctpop_wrong_cmp_op1_logical(
-; CHECK-NEXT:    [[T0:%.*]] = tail call i32 @llvm.ctpop.i32(i32 [[X:%.*]]), !range [[RNG0]]
+; CHECK-NEXT:    [[T0:%.*]] = tail call range(i32 0, 33) i32 @llvm.ctpop.i32(i32 [[X:%.*]])
 ; CHECK-NEXT:    [[CMP:%.*]] = icmp ult i32 [[T0]], 3
 ; CHECK-NEXT:    [[NOTZERO:%.*]] = icmp ne i32 [[X]], 0
 ; CHECK-NEXT:    [[R:%.*]] = select i1 [[NOTZERO]], i1 [[CMP]], i1 false
@@ -358,7 +358,7 @@ define i1 @is_pow2_ctpop_wrong_cmp_op1_logical(i32 %x) {
 
 define i1 @is_pow2_ctpop_wrong_cmp_op2(i32 %x) {
 ; CHECK-LABEL: @is_pow2_ctpop_wrong_cmp_op2(
-; CHECK-NEXT:    [[T0:%.*]] = tail call i32 @llvm.ctpop.i32(i32 [[X:%.*]]), !range [[RNG0]]
+; CHECK-NEXT:    [[T0:%.*]] = tail call range(i32 0, 33) i32 @llvm.ctpop.i32(i32 [[X:%.*]])
 ; CHECK-NEXT:    [[CMP:%.*]] = icmp ult i32 [[T0]], 2
 ; CHECK-NEXT:    [[NOTZERO:%.*]] = icmp ne i32 [[X]], 1
 ; CHECK-NEXT:    [[R:%.*]] = and i1 [[NOTZERO]], [[CMP]]
@@ -373,7 +373,7 @@ define i1 @is_pow2_ctpop_wrong_cmp_op2(i32 %x) {
 
 define i1 @is_pow2_ctpop_wrong_cmp_op2_logical(i32 %x) {
 ; CHECK-LABEL: @is_pow2_ctpop_wrong_cmp_op2_logical(
-; CHECK-NEXT:    [[T0:%.*]] = tail call i32 @llvm.ctpop.i32(i32 [[X:%.*]]), !range [[RNG0]]
+; CHECK-NEXT:    [[T0:%.*]] = tail call range(i32 0, 33) i32 @llvm.ctpop.i32(i32 [[X:%.*]])
 ; CHECK-NEXT:    [[CMP:%.*]] = icmp ult i32 [[T0]], 2
 ; CHECK-NEXT:    [[NOTZERO:%.*]] = icmp ne i32 [[X]], 1
 ; CHECK-NEXT:    [[R:%.*]] = select i1 [[NOTZERO]], i1 [[CMP]], i1 false
@@ -390,7 +390,7 @@ define i1 @is_pow2_ctpop_wrong_cmp_op2_logical(i32 %x) {
 
 define i1 @is_pow2_ctpop_wrong_pred1(i32 %x) {
 ; CHECK-LABEL: @is_pow2_ctpop_wrong_pred1(
-; CHECK-NEXT:    [[T0:%.*]] = tail call i32 @llvm.ctpop.i32(i32 [[X:%.*]]), !range [[RNG0]]
+; CHECK-NEXT:    [[T0:%.*]] = tail call range(i32 0, 33) i32 @llvm.ctpop.i32(i32 [[X:%.*]])
 ; CHECK-NEXT:    [[CMP:%.*]] = icmp ugt i32 [[T0]], 2
 ; CHECK-NEXT:    ret i1 [[CMP]]
 ;
@@ -403,7 +403,7 @@ define i1 @is_pow2_ctpop_wrong_pred1(i32 %x) {
 
 define i1 @is_pow2_ctpop_wrong_pred1_logical(i32 %x) {
 ; CHECK-LABEL: @is_pow2_ctpop_wrong_pred1_logical(
-; CHECK-NEXT:    [[T0:%.*]] = tail call i32 @llvm.ctpop.i32(i32 [[X:%.*]]), !range [[RNG0]]
+; CHECK-NEXT:    [[T0:%.*]] = tail call range(i32 0, 33) i32 @llvm.ctpop.i32(i32 [[X:%.*]])
 ; CHECK-NEXT:    [[CMP:%.*]] = icmp ugt i32 [[T0]], 2
 ; CHECK-NEXT:    ret i1 [[CMP]]
 ;
@@ -418,7 +418,7 @@ define i1 @is_pow2_ctpop_wrong_pred1_logical(i32 %x) {
 
 define i1 @is_pow2_ctpop_wrong_pred2(i32 %x) {
 ; CHECK-LABEL: @is_pow2_ctpop_wrong_pred2(
-; CHECK-NEXT:    [[T0:%.*]] = tail call i32 @llvm.ctpop.i32(i32 [[X:%.*]]), !range [[RNG0]]
+; CHECK-NEXT:    [[T0:%.*]] = tail call range(i32 0, 33) i32 @llvm.ctpop.i32(i32 [[X:%.*]])
 ; CHECK-NEXT:    [[CMP:%.*]] = icmp ult i32 [[T0]], 2
 ; CHECK-NEXT:    [[CMP2:%.*]] = icmp sgt i32 [[X]], 0
 ; CHECK-NEXT:    [[R:%.*]] = and i1 [[CMP2]], [[CMP]]
@@ -433,7 +433,7 @@ define i1 @is_pow2_ctpop_wrong_pred2(i32 %x) {
 
 define i1 @is_pow2_ctpop_wrong_pred2_logical(i32 %x) {
 ; CHECK-LABEL: @is_pow2_ctpop_wrong_pred2_logical(
-; CHECK-NEXT:    [[T0:%.*]] = tail call i32 @llvm.ctpop.i32(i32 [[X:%.*]]), !range [[RNG0]]
+; CHECK-NEXT:    [[T0:%.*]] = tail call range(i32 0, 33) i32 @llvm.ctpop.i32(i32 [[X:%.*]])
 ; CHECK-NEXT:    [[CMP:%.*]] = icmp ult i32 [[T0]], 2
 ; CHECK-NEXT:    [[CMP2:%.*]] = icmp sgt i32 [[X]], 0
 ; CHECK-NEXT:    [[R:%.*]] = select i1 [[CMP2]], i1 [[CMP]], i1 false
@@ -450,7 +450,7 @@ define i1 @is_pow2_ctpop_wrong_pred2_logical(i32 %x) {
 
 define i1 @isnot_pow2_ctpop(i32 %x) {
 ; CHECK-LABEL: @isnot_pow2_ctpop(
-; CHECK-NEXT:    [[T0:%.*]] = tail call i32 @llvm.ctpop.i32(i32 [[X:%.*]]), !range [[RNG0]]
+; CHECK-NEXT:    [[T0:%.*]] = tail call range(i32 0, 33) i32 @llvm.ctpop.i32(i32 [[X:%.*]])
 ; CHECK-NEXT:    [[R:%.*]] = icmp ne i32 [[T0]], 1
 ; CHECK-NEXT:    ret i1 [[R]]
 ;
@@ -463,7 +463,7 @@ define i1 @isnot_pow2_ctpop(i32 %x) {
 
 define i1 @isnot_pow2_ctpop_logical(i32 %x) {
 ; CHECK-LABEL: @isnot_pow2_ctpop_logical(
-; CHECK-NEXT:    [[T0:%.*]] = tail call i32 @llvm.ctpop.i32(i32 [[X:%.*]]), !range [[RNG0]]
+; CHECK-NEXT:    [[T0:%.*]] = tail call range(i32 0, 33) i32 @llvm.ctpop.i32(i32 [[X:%.*]])
 ; CHECK-NEXT:    [[R:%.*]] = icmp ne i32 [[T0]], 1
 ; CHECK-NEXT:    ret i1 [[R]]
 ;
@@ -478,7 +478,7 @@ define i1 @isnot_pow2_ctpop_logical(i32 %x) {
 
 define i1 @isnot_pow2_ctpop_extra_uses(i32 %x) {
 ; CHECK-LABEL: @isnot_pow2_ctpop_extra_uses(
-; CHECK-NEXT:    [[T0:%.*]] = tail call i32 @llvm.ctpop.i32(i32 [[X:%.*]]), !range [[RNG0]]
+; CHECK-NEXT:    [[T0:%.*]] = tail call range(i32 0, 33) i32 @llvm.ctpop.i32(i32 [[X:%.*]])
 ; CHECK-NEXT:    [[CMP:%.*]] = icmp ugt i32 [[T0]], 1
 ; CHECK-NEXT:    call void @use_i1(i1 [[CMP]])
 ; CHECK-NEXT:    [[ISZERO:%.*]] = icmp eq i32 [[X]], 0
@@ -497,7 +497,7 @@ define i1 @isnot_pow2_ctpop_extra_uses(i32 %x) {
 
 define i1 @isnot_pow2_ctpop_extra_uses_logical(i32 %x) {
 ; CHECK-LABEL: @isnot_pow2_ctpop_extra_uses_logical(
-; CHECK-NEXT:    [[T0:%.*]] = tail call i32 @llvm.ctpop.i32(i32 [[X:%.*]]), !range [[RNG0]]
+; CHECK-NEXT:    [[T0:%.*]] = tail call range(i32 0, 33) i32 @llvm.ctpop.i32(i32 [[X:%.*]])
 ; CHECK-NEXT:    [[CMP:%.*]] = icmp ugt i32 [[T0]], 1
 ; CHECK-NEXT:    call void @use_i1(i1 [[CMP]])
 ; CHECK-NEXT:    [[ISZERO:%.*]] = icmp eq i32 [[X]], 0
@@ -518,7 +518,7 @@ define i1 @isnot_pow2_ctpop_extra_uses_logical(i32 %x) {
 
 define <2 x i1> @isnot_pow2_ctpop_commute_vec(<2 x i8> %x) {
 ; CHECK-LABEL: @isnot_pow2_ctpop_commute_vec(
-; CHECK-NEXT:    [[T0:%.*]] = tail call <2 x i8> @llvm.ctpop.v2i8(<2 x i8> [[X:%.*]]), !range [[RNG1]]
+; CHECK-NEXT:    [[T0:%.*]] = tail call range(i8 0, 9) <2 x i8> @llvm.ctpop.v2i8(<2 x i8> [[X:%.*]])
 ; CHECK-NEXT:    [[R:%.*]] = icmp ne <2 x i8> [[T0]], <i8 1, i8 1>
 ; CHECK-NEXT:    ret <2 x i1> [[R]]
 ;
@@ -533,7 +533,7 @@ define <2 x i1> @isnot_pow2_ctpop_commute_vec(<2 x i8> %x) {
 
 define i1 @isnot_pow2_ctpop_wrong_cmp_op1(i32 %x) {
 ; CHECK-LABEL: @isnot_pow2_ctpop_wrong_cmp_op1(
-; CHECK-NEXT:    [[T0:%.*]] = tail call i32 @llvm.ctpop.i32(i32 [[X:%.*]]), !range [[RNG0]]
+; CHECK-NEXT:    [[T0:%.*]] = tail call range(i32 0, 33) i32 @llvm.ctpop.i32(i32 [[X:%.*]])
 ; CHECK-NEXT:    [[CMP:%.*]] = icmp ugt i32 [[T0]], 2
 ; CHECK-NEXT:    [[ISZERO:%.*]] = icmp eq i32 [[X]], 0
 ; CHECK-NEXT:    [[R:%.*]] = or i1 [[ISZERO]], [[CMP]]
@@ -548,7 +548,7 @@ define i1 @isnot_pow2_ctpop_wrong_cmp_op1(i32 %x) {
 
 define i1 @isnot_pow2_ctpop_wrong_cmp_op1_logical(i32 %x) {
 ; CHECK-LABEL: @isnot_pow2_ctpop_wrong_cmp_op1_logical(
-; CHECK-NEXT:    [[T0:%.*]] = tail call i32 @llvm.ctpop.i32(i32 [[X:%.*]]), !range [[RNG0]]
+; CHECK-NEXT:    [[T0:%.*]] = tail call range(i32 0, 33) i32 @llvm.ctpop.i32(i32 [[X:%.*]])
 ; CHECK-NEXT:    [[CMP:%.*]] = icmp ugt i32 [[T0]], 2
 ; CHECK-NEXT:    [[ISZERO:%.*]] = icmp eq i32 [[X]], 0
 ; CHECK-NEXT:    [[R:%.*]] = select i1 [[ISZERO]], i1 true, i1 [[CMP]]
@@ -565,7 +565,7 @@ define i1 @isnot_pow2_ctpop_wrong_cmp_op1_logical(i32 %x) {
 
 define i1 @isnot_pow2_ctpop_wrong_cmp_op2(i32 %x) {
 ; CHECK-LABEL: @isnot_pow2_ctpop_wrong_cmp_op2(
-; CHECK-NEXT:    [[T0:%.*]] = tail call i32 @llvm.ctpop.i32(i32 [[X:%.*]]), !range [[RNG0]]
+; CHECK-NEXT:    [[T0:%.*]] = tail call range(i32 0, 33) i32 @llvm.ctpop.i32(i32 [[X:%.*]])
 ; CHECK-NEXT:    [[CMP:%.*]] = icmp ugt i32 [[T0]], 1
 ; CHECK-NEXT:    [[ISZERO:%.*]] = icmp eq i32 [[X]], 1
 ; CHECK-NEXT:    [[R:%.*]] = or i1 [[ISZERO]], [[CMP]]
@@ -580,7 +580,7 @@ define i1 @isnot_pow2_ctpop_wrong_cmp_op2(i32 %x) {
 
 define i1 @isnot_pow2_ctpop_wrong_cmp_op2_logical(i32 %x) {
 ; CHECK-LABEL: @isnot_pow2_ctpop_wrong_cmp_op2_logical(
-; CHECK-NEXT:    [[T0:%.*]] = tail call i32 @llvm.ctpop.i32(i32 [[X:%.*]]), !range [[RNG0]]
+; CHECK-NEXT:    [[T0:%.*]] = tail call range(i32 0, 33) i32 @llvm.ctpop.i32(i32 [[X:%.*]])
 ; CHECK-NEXT:    [[CMP:%.*]] = icmp ugt i32 [[T0]], 1
 ; CHECK-NEXT:    [[ISZERO:%.*]] = icmp eq i32 [[X]], 1
 ; CHECK-NEXT:    [[R:%.*]] = select i1 [[ISZERO]], i1 true, i1 [[CMP]]
@@ -597,7 +597,7 @@ define i1 @isnot_pow2_ctpop_wrong_cmp_op2_logical(i32 %x) {
 
 define i1 @isnot_pow2_ctpop_wrong_pred2(i32 %x) {
 ; CHECK-LABEL: @isnot_pow2_ctpop_wrong_pred2(
-; CHECK-NEXT:    [[T0:%.*]] = tail call i32 @llvm.ctpop.i32(i32 [[X:%.*]]), !range [[RNG0]]
+; CHECK-NEXT:    [[T0:%.*]] = tail call range(i32 0, 33) i32 @llvm.ctpop.i32(i32 [[X:%.*]])
 ; CHECK-NEXT:    [[CMP:%.*]] = icmp ugt i32 [[T0]], 1
 ; CHECK-NEXT:    [[CMP2:%.*]] = icmp slt i32 [[X]], 0
 ; CHECK-NEXT:    [[R:%.*]] = or i1 [[CMP2]], [[CMP]]
@@ -612,7 +612,7 @@ define i1 @isnot_pow2_ctpop_wrong_pred2(i32 %x) {
 
 define i1 @isnot_pow2_ctpop_wrong_pred2_logical(i32 %x) {
 ; CHECK-LABEL: @isnot_pow2_ctpop_wrong_pred2_logical(
-; CHECK-NEXT:    [[T0:%.*]] = tail call i32 @llvm.ctpop.i32(i32 [[X:%.*]]), !range [[RNG0]]
+; CHECK-NEXT:    [[T0:%.*]] = tail call range(i32 0, 33) i32 @llvm.ctpop.i32(i32 [[X:%.*]])
 ; CHECK-NEXT:    [[CMP:%.*]] = icmp ugt i32 [[T0]], 1
 ; CHECK-NEXT:    [[CMP2:%.*]] = icmp slt i32 [[X]], 0
 ; CHECK-NEXT:    [[R:%.*]] = select i1 [[CMP2]], i1 true, i1 [[CMP]]
@@ -627,7 +627,7 @@ define i1 @isnot_pow2_ctpop_wrong_pred2_logical(i32 %x) {
 
 define i1 @is_pow2_negate_op(i32 %x) {
 ; CHECK-LABEL: @is_pow2_negate_op(
-; CHECK-NEXT:    [[TMP1:%.*]] = call i32 @llvm.ctpop.i32(i32 [[X:%.*]]), !range [[RNG0]]
+; CHECK-NEXT:    [[TMP1:%.*]] = call range(i32 0, 33) i32 @llvm.ctpop.i32(i32 [[X:%.*]])
 ; CHECK-NEXT:    [[R:%.*]] = icmp eq i32 [[TMP1]], 1
 ; CHECK-NEXT:    ret i1 [[R]]
 ;
@@ -641,7 +641,7 @@ define i1 @is_pow2_negate_op(i32 %x) {
 
 define i1 @is_pow2_negate_op_logical(i32 %x) {
 ; CHECK-LABEL: @is_pow2_negate_op_logical(
-; CHECK-NEXT:    [[TMP1:%.*]] = call i32 @llvm.ctpop.i32(i32 [[X:%.*]]), !range [[RNG0]]
+; CHECK-NEXT:    [[TMP1:%.*]] = call range(i32 0, 33) i32 @llvm.ctpop.i32(i32 [[X:%.*]])
 ; CHECK-NEXT:    [[R:%.*]] = icmp eq i32 [[TMP1]], 1
 ; CHECK-NEXT:    ret i1 [[R]]
 ;
@@ -655,7 +655,7 @@ define i1 @is_pow2_negate_op_logical(i32 %x) {
 
 define <2 x i1> @is_pow2_negate_op_vec(<2 x i32> %x) {
 ; CHECK-LABEL: @is_pow2_negate_op_vec(
-; CHECK-NEXT:    [[TMP1:%.*]] = call <2 x i32> @llvm.ctpop.v2i32(<2 x i32> [[X:%.*]]), !range [[RNG0]]
+; CHECK-NEXT:    [[TMP1:%.*]] = call range(i32 0, 33) <2 x i32> @llvm.ctpop.v2i32(<2 x i32> [[X:%.*]])
 ; CHECK-NEXT:    [[R:%.*]] = icmp eq <2 x i32> [[TMP1]], <i32 1, i32 1>
 ; CHECK-NEXT:    ret <2 x i1> [[R]]
 ;
@@ -669,7 +669,7 @@ define <2 x i1> @is_pow2_negate_op_vec(<2 x i32> %x) {
 
 define i1 @is_pow2_decrement_op(i8 %x) {
 ; CHECK-LABEL: @is_pow2_decrement_op(
-; CHECK-NEXT:    [[TMP1:%.*]] = call i8 @llvm.ctpop.i8(i8 [[X:%.*]]), !range [[RNG1]]
+; CHECK-NEXT:    [[TMP1:%.*]] = call range(i8 0, 9) i8 @llvm.ctpop.i8(i8 [[X:%.*]])
 ; CHECK-NEXT:    [[R:%.*]] = icmp eq i8 [[TMP1]], 1
 ; CHECK-NEXT:    ret i1 [[R]]
 ;
@@ -683,7 +683,7 @@ define i1 @is_pow2_decrement_op(i8 %x) {
 
 define i1 @is_pow2_decrement_op_logical(i8 %x) {
 ; CHECK-LABEL: @is_pow2_decrement_op_logical(
-; CHECK-NEXT:    [[TMP1:%.*]] = call i8 @llvm.ctpop.i8(i8 [[X:%.*]]), !range [[RNG1]]
+; CHECK-NEXT:    [[TMP1:%.*]] = call range(i8 0, 9) i8 @llvm.ctpop.i8(i8 [[X:%.*]])
 ; CHECK-NEXT:    [[R:%.*]] = icmp eq i8 [[TMP1]], 1
 ; CHECK-NEXT:    ret i1 [[R]]
 ;
@@ -697,7 +697,7 @@ define i1 @is_pow2_decrement_op_logical(i8 %x) {
 
 define <2 x i1> @is_pow2_decrement_op_vec(<2 x i8> %x) {
 ; CHECK-LABEL: @is_pow2_decrement_op_vec(
-; CHECK-NEXT:    [[TMP1:%.*]] = call <2 x i8> @llvm.ctpop.v2i8(<2 x i8> [[X:%.*]]), !range [[RNG1]]
+; CHECK-NEXT:    [[TMP1:%.*]] = call range(i8 0, 9) <2 x i8> @llvm.ctpop.v2i8(<2 x i8> [[X:%.*]])
 ; CHECK-NEXT:    [[R:%.*]] = icmp eq <2 x i8> [[TMP1]], <i8 1, i8 1>
 ; CHECK-NEXT:    ret <2 x i1> [[R]]
 ;
@@ -711,7 +711,7 @@ define <2 x i1> @is_pow2_decrement_op_vec(<2 x i8> %x) {
 
 define i1 @isnot_pow2_negate_op(i32 %x) {
 ; CHECK-LABEL: @isnot_pow2_negate_op(
-; CHECK-NEXT:    [[TMP1:%.*]] = call i32 @llvm.ctpop.i32(i32 [[X:%.*]]), !range [[RNG0]]
+; CHECK-NEXT:    [[TMP1:%.*]] = call range(i32 0, 33) i32 @llvm.ctpop.i32(i32 [[X:%.*]])
 ; CHECK-NEXT:    [[R:%.*]] = icmp ne i32 [[TMP1]], 1
 ; CHECK-NEXT:    ret i1 [[R]]
 ;
@@ -725,7 +725,7 @@ define i1 @isnot_pow2_negate_op(i32 %x) {
 
 define i1 @isnot_pow2_negate_op_logical(i32 %x) {
 ; CHECK-LABEL: @isnot_pow2_negate_op_logical(
-; CHECK-NEXT:    [[TMP1:%.*]] = call i32 @llvm.ctpop.i32(i32 [[X:%.*]]), !range [[RNG0]]
+; CHECK-NEXT:    [[TMP1:%.*]] = call range(i32 0, 33) i32 @llvm.ctpop.i32(i32 [[X:%.*]])
 ; CHECK-NEXT:    [[R:%.*]] = icmp ne i32 [[TMP1]], 1
 ; CHECK-NEXT:    ret i1 [[R]]
 ;
@@ -739,7 +739,7 @@ define i1 @isnot_pow2_negate_op_logical(i32 %x) {
 
 define <2 x i1> @isnot_pow2_negate_op_vec(<2 x i32> %x) {
 ; CHECK-LABEL: @isnot_pow2_negate_op_vec(
-; CHECK-NEXT:    [[TMP1:%.*]] = call <2 x i32> @llvm.ctpop.v2i32(<2 x i32> [[X:%.*]]), !range [[RNG0]]
+; CHECK-NEXT:    [[TMP1:%.*]] = call range(i32 0, 33) <2 x i32> @llvm.ctpop.v2i32(<2 x i32> [[X:%.*]])
 ; CHECK-NEXT:    [[R:%.*]] = icmp ne <2 x i32> [[TMP1]], <i32 1, i32 1>
 ; CHECK-NEXT:    ret <2 x i1> [[R]]
 ;
@@ -753,7 +753,7 @@ define <2 x i1> @isnot_pow2_negate_op_vec(<2 x i32> %x) {
 
 define i1 @isnot_pow2_decrement_op(i8 %x) {
 ; CHECK-LABEL: @isnot_pow2_decrement_op(
-; CHECK-NEXT:    [[TMP1:%.*]] = call i8 @llvm.ctpop.i8(i8 [[X:%.*]]), !range [[RNG1]]
+; CHECK-NEXT:    [[TMP1:%.*]] = call range(i8 0, 9) i8 @llvm.ctpop.i8(i8 [[X:%.*]])
 ; CHECK-NEXT:    [[R:%.*]] = icmp ne i8 [[TMP1]], 1
 ; CHECK-NEXT:    ret i1 [[R]]
 ;
@@ -767,7 +767,7 @@ define i1 @isnot_pow2_decrement_op(i8 %x) {
 
 define i1 @isnot_pow2_decrement_op_logical(i8 %x) {
 ; CHECK-LABEL: @isnot_pow2_decrement_op_logical(
-; CHECK-NEXT:    [[TMP1:%.*]] = call i8 @llvm.ctpop.i8(i8 [[X:%.*]]), !range [[RNG1]]
+; CHECK-NEXT:    [[TMP1:%.*]] = call range(i8 0, 9) i8 @llvm.ctpop.i8(i8 [[X:%.*]])
 ; CHECK-NEXT:    [[R:%.*]] = icmp ne i8 [[TMP1]], 1
 ; CHECK-NEXT:    ret i1 [[R]]
 ;
@@ -781,7 +781,7 @@ define i1 @isnot_pow2_decrement_op_logical(i8 %x) {
 
 define <2 x i1> @isnot_pow2_decrement_op_vec(<2 x i8> %x) {
 ; CHECK-LABEL: @isnot_pow2_decrement_op_vec(
-; CHECK-NEXT:    [[TMP1:%.*]] = call <2 x i8> @llvm.ctpop.v2i8(<2 x i8> [[X:%.*]]), !range [[RNG1]]
+; CHECK-NEXT:    [[TMP1:%.*]] = call range(i8 0, 9) <2 x i8> @llvm.ctpop.v2i8(<2 x i8> [[X:%.*]])
 ; CHECK-NEXT:    [[R:%.*]] = icmp ne <2 x i8> [[TMP1]], <i8 1, i8 1>
 ; CHECK-NEXT:    ret <2 x i1> [[R]]
 ;
@@ -797,7 +797,7 @@ define <2 x i1> @isnot_pow2_decrement_op_vec(<2 x i8> %x) {
 
 define i1 @is_pow2or0_ctpop(i32 %x) {
 ; CHECK-LABEL: @is_pow2or0_ctpop(
-; CHECK-NEXT:    [[T0:%.*]] = tail call i32 @llvm.ctpop.i32(i32 [[X:%.*]]), !range [[RNG0]]
+; CHECK-NEXT:    [[T0:%.*]] = tail call range(i32 0, 33) i32 @llvm.ctpop.i32(i32 [[X:%.*]])
 ; CHECK-NEXT:    [[R:%.*]] = icmp ult i32 [[T0]], 2
 ; CHECK-NEXT:    ret i1 [[R]]
 ;
@@ -810,7 +810,7 @@ define i1 @is_pow2or0_ctpop(i32 %x) {
 
 define i1 @is_pow2or0_ctpop_swap_cmp(i32 %x) {
 ; CHECK-LABEL: @is_pow2or0_ctpop_swap_cmp(
-; CHECK-NEXT:    [[T0:%.*]] = tail call i32 @llvm.ctpop.i32(i32 [[X:%.*]]), !range [[RNG0]]
+; CHECK-NEXT:    [[T0:%.*]] = tail call range(i32 0, 33) i32 @llvm.ctpop.i32(i32 [[X:%.*]])
 ; CHECK-NEXT:    [[R:%.*]] = icmp ult i32 [[T0]], 2
 ; CHECK-NEXT:    ret i1 [[R]]
 ;
@@ -823,7 +823,7 @@ define i1 @is_pow2or0_ctpop_swap_cmp(i32 %x) {
 
 define i1 @is_pow2or0_ctpop_logical(i32 %x) {
 ; CHECK-LABEL: @is_pow2or0_ctpop_logical(
-; CHECK-NEXT:    [[T0:%.*]] = tail call i32 @llvm.ctpop.i32(i32 [[X:%.*]]), !range [[RNG0]]
+; CHECK-NEXT:    [[T0:%.*]] = tail call range(i32 0, 33) i32 @llvm.ctpop.i32(i32 [[X:%.*]])
 ; CHECK-NEXT:    [[R:%.*]] = icmp ult i32 [[T0]], 2
 ; CHECK-NEXT:    ret i1 [[R]]
 ;
@@ -836,7 +836,7 @@ define i1 @is_pow2or0_ctpop_logical(i32 %x) {
 
 define <2 x i1> @is_pow2or0_ctpop_commute_vec(<2 x i8> %x) {
 ; CHECK-LABEL: @is_pow2or0_ctpop_commute_vec(
-; CHECK-NEXT:    [[T0:%.*]] = tail call <2 x i8> @llvm.ctpop.v2i8(<2 x i8> [[X:%.*]]), !range [[RNG1]]
+; CHECK-NEXT:    [[T0:%.*]] = tail call range(i8 0, 9) <2 x i8> @llvm.ctpop.v2i8(<2 x i8> [[X:%.*]])
 ; CHECK-NEXT:    [[R:%.*]] = icmp ult <2 x i8> [[T0]], <i8 2, i8 2>
 ; CHECK-NEXT:    ret <2 x i1> [[R]]
 ;
@@ -851,7 +851,7 @@ define <2 x i1> @is_pow2or0_ctpop_commute_vec(<2 x i8> %x) {
 
 define i1 @is_pow2or0_ctpop_extra_uses(i32 %x) {
 ; CHECK-LABEL: @is_pow2or0_ctpop_extra_uses(
-; CHECK-NEXT:    [[T0:%.*]] = tail call i32 @llvm.ctpop.i32(i32 [[X:%.*]]), !range [[RNG0]]
+; CHECK-NEXT:    [[T0:%.*]] = tail call range(i32 0, 33) i32 @llvm.ctpop.i32(i32 [[X:%.*]])
 ; CHECK-NEXT:    call void @use(i32 [[T0]])
 ; CHECK-NEXT:    [[CMP:%.*]] = icmp eq i32 [[T0]], 1
 ; CHECK-NEXT:    call void @use_i1(i1 [[CMP]])
@@ -872,7 +872,7 @@ define i1 @is_pow2or0_ctpop_extra_uses(i32 %x) {
 
 define i1 @is_pow2or0_ctpop_logical_extra_uses(i32 %x) {
 ; CHECK-LABEL: @is_pow2or0_ctpop_logical_extra_uses(
-; CHECK-NEXT:    [[T0:%.*]] = tail call i32 @llvm.ctpop.i32(i32 [[X:%.*]]), !range [[RNG0]]
+; CHECK-NEXT:    [[T0:%.*]] = tail call range(i32 0, 33) i32 @llvm.ctpop.i32(i32 [[X:%.*]])
 ; CHECK-NEXT:    call void @use(i32 [[T0]])
 ; CHECK-NEXT:    [[CMP:%.*]] = icmp eq i32 [[T0]], 1
 ; CHECK-NEXT:    call void @use_i1(i1 [[CMP]])
@@ -895,7 +895,7 @@ define i1 @is_pow2or0_ctpop_logical_extra_uses(i32 %x) {
 
 define i1 @is_pow2or0_ctpop_wrong_cmp_op1(i32 %x) {
 ; CHECK-LABEL: @is_pow2or0_ctpop_wrong_cmp_op1(
-; CHECK-NEXT:    [[T0:%.*]] = tail call i32 @llvm.ctpop.i32(i32 [[X:%.*]]), !range [[RNG0]]
+; CHECK-NEXT:    [[T0:%.*]] = tail call range(i32 0, 33) i32 @llvm.ctpop.i32(i32 [[X:%.*]])
 ; CHECK-NEXT:    [[CMP:%.*]] = icmp eq i32 [[T0]], 2
 ; CHECK-NEXT:    [[ISZERO:%.*]] = icmp eq i32 [[X]], 0
 ; CHECK-NEXT:    [[R:%.*]] = or i1 [[ISZERO]], [[CMP]]
@@ -910,7 +910,7 @@ define i1 @is_pow2or0_ctpop_wrong_cmp_op1(i32 %x) {
 
 define i1 @is_pow2or0_ctpop_wrong_cmp_op1_logical(i32 %x) {
 ; CHECK-LABEL: @is_pow2or0_ctpop_wrong_cmp_op1_logical(
-; CHECK-NEXT:    [[T0:%.*]] = tail call i32 @llvm.ctpop.i32(i32 [[X:%.*]]), !range [[RNG0]]
+; CHECK-NEXT:    [[T0:%.*]] = tail call range(i32 0, 33) i32 @llvm.ctpop.i32(i32 [[X:%.*]])
 ; CHECK-NEXT:    [[CMP:%.*]] = icmp eq i32 [[T0]], 3
 ; CHECK-NEXT:    [[ISZERO:%.*]] = icmp eq i32 [[X]], 0
 ; CHECK-NEXT:    [[R:%.*]] = select i1 [[ISZERO]], i1 true, i1 [[CMP]]
@@ -925,7 +925,7 @@ define i1 @is_pow2or0_ctpop_wrong_cmp_op1_logical(i32 %x) {
 
 define <2 x i1> @is_pow2or0_ctpop_commute_vec_wrong_cmp_op1(<2 x i8> %x) {
 ; CHECK-LABEL: @is_pow2or0_ctpop_commute_vec_wrong_cmp_op1(
-; CHECK-NEXT:    [[T0:%.*]] = tail call <2 x i8> @llvm.ctpop.v2i8(<2 x i8> [[X:%.*]]), !range [[RNG1]]
+; CHECK-NEXT:    [[T0:%.*]] = tail call range(i8 0, 9) <2 x i8> @llvm.ctpop.v2i8(<2 x i8> [[X:%.*]])
 ; CHECK-NEXT:    [[CMP:%.*]] = icmp eq <2 x i8> [[T0]], <i8 -1, i8 1>
 ; CHECK-NEXT:    [[ISZERO:%.*]] = icmp eq <2 x i8> [[X]], zeroinitializer
 ; CHECK-NEXT:    [[R:%.*]] = or <2 x i1> [[CMP]], [[ISZERO]]
@@ -942,7 +942,7 @@ define <2 x i1> @is_pow2or0_ctpop_commute_vec_wrong_cmp_op1(<2 x i8> %x) {
 
 define i1 @is_pow2or0_ctpop_wrong_pred1(i32 %x) {
 ; CHECK-LABEL: @is_pow2or0_ctpop_wrong_pred1(
-; CHECK-NEXT:    [[T0:%.*]] = tail call i32 @llvm.ctpop.i32(i32 [[X:%.*]]), !range [[RNG0]]
+; CHECK-NEXT:    [[T0:%.*]] = tail call range(i32 0, 33) i32 @llvm.ctpop.i32(i32 [[X:%.*]])
 ; CHECK-NEXT:    [[CMP:%.*]] = icmp ne i32 [[T0]], 1
 ; CHECK-NEXT:    ret i1 [[CMP]]
 ;
@@ -977,7 +977,7 @@ define i1 @is_pow2or0_ctpop_wrong_pred2_logical(i32 %x) {
 
 define <2 x i1> @is_pow2or0_ctpop_commute_vec_wrong_pred3(<2 x i8> %x) {
 ; CHECK-LABEL: @is_pow2or0_ctpop_commute_vec_wrong_pred3(
-; CHECK-NEXT:    [[T0:%.*]] = tail call <2 x i8> @llvm.ctpop.v2i8(<2 x i8> [[X:%.*]]), !range [[RNG1]]
+; CHECK-NEXT:    [[T0:%.*]] = tail call range(i8 0, 9) <2 x i8> @llvm.ctpop.v2i8(<2 x i8> [[X:%.*]])
 ; CHECK-NEXT:    [[CMP:%.*]] = icmp eq <2 x i8> [[T0]], <i8 1, i8 1>
 ; CHECK-NEXT:    [[ISZERO:%.*]] = icmp eq <2 x i8> [[X]], zeroinitializer
 ; CHECK-NEXT:    [[R:%.*]] = and <2 x i1> [[CMP]], [[ISZERO]]
@@ -994,7 +994,7 @@ define <2 x i1> @is_pow2or0_ctpop_commute_vec_wrong_pred3(<2 x i8> %x) {
 
 define i1 @isnot_pow2nor0_ctpop(i32 %x) {
 ; CHECK-LABEL: @isnot_pow2nor0_ctpop(
-; CHECK-NEXT:    [[T0:%.*]] = tail call i32 @llvm.ctpop.i32(i32 [[X:%.*]]), !range [[RNG0]]
+; CHECK-NEXT:    [[T0:%.*]] = tail call range(i32 0, 33) i32 @llvm.ctpop.i32(i32 [[X:%.*]])
 ; CHECK-NEXT:    [[R:%.*]] = icmp ugt i32 [[T0]], 1
 ; CHECK-NEXT:    ret i1 [[R]]
 ;
@@ -1007,7 +1007,7 @@ define i1 @isnot_pow2nor0_ctpop(i32 %x) {
 
 define i1 @isnot_pow2nor0_ctpop_swap_cmp(i32 %x) {
 ; CHECK-LABEL: @isnot_pow2nor0_ctpop_swap_cmp(
-; CHECK-NEXT:    [[T0:%.*]] = tail call i32 @llvm.ctpop.i32(i32 [[X:%.*]]), !range [[RNG0]]
+; CHECK-NEXT:    [[T0:%.*]] = tail call range(i32 0, 33) i32 @llvm.ctpop.i32(i32 [[X:%.*]])
 ; CHECK-NEXT:    [[R:%.*]] = icmp ugt i32 [[T0]], 1
 ; CHECK-NEXT:    ret i1 [[R]]
 ;
@@ -1020,7 +1020,7 @@ define i1 @isnot_pow2nor0_ctpop_swap_cmp(i32 %x) {
 
 define i1 @isnot_pow2nor0_ctpop_logical(i32 %x) {
 ; CHECK-LABEL: @isnot_pow2nor0_ctpop_logical(
-; CHECK-NEXT:    [[T0:%.*]] = tail call i32 @llvm.ctpop.i32(i32 [[X:%.*]]), !range [[RNG0]]
+; CHECK-NEXT:    [[T0:%.*]] = tail call range(i32 0, 33) i32 @llvm.ctpop.i32(i32 [[X:%.*]])
 ; CHECK-NEXT:    [[R:%.*]] = icmp ugt i32 [[T0]], 1
 ; CHECK-NEXT:    ret i1 [[R]]
 ;
@@ -1033,7 +1033,7 @@ define i1 @isnot_pow2nor0_ctpop_logical(i32 %x) {
 
 define <2 x i1> @isnot_pow2nor0_ctpop_commute_vec(<2 x i8> %x) {
 ; CHECK-LABEL: @isnot_pow2nor0_ctpop_commute_vec(
-; CHECK-NEXT:    [[T0:%.*]] = tail call <2 x i8> @llvm.ctpop.v2i8(<2 x i8> [[X:%.*]]), !range [[RNG1]]
+; CHECK-NEXT:    [[T0:%.*]] = tail call range(i8 0, 9) <2 x i8> @llvm.ctpop.v2i8(<2 x i8> [[X:%.*]])
 ; CHECK-NEXT:    [[R:%.*]] = icmp ugt <2 x i8> [[T0]], <i8 1, i8 1>
 ; CHECK-NEXT:    ret <2 x i1> [[R]]
 ;
@@ -1048,7 +1048,7 @@ define <2 x i1> @isnot_pow2nor0_ctpop_commute_vec(<2 x i8> %x) {
 
 define i1 @isnot_pow2nor0_ctpop_extra_uses(i32 %x) {
 ; CHECK-LABEL: @isnot_pow2nor0_ctpop_extra_uses(
-; CHECK-NEXT:    [[T0:%.*]] = tail call i32 @llvm.ctpop.i32(i32 [[X:%.*]]), !range [[RNG0]]
+; CHECK-NEXT:    [[T0:%.*]] = tail call range(i32 0, 33) i32 @llvm.ctpop.i32(i32 [[X:%.*]])
 ; CHECK-NEXT:    call void @use(i32 [[T0]])
 ; CHECK-NEXT:    [[CMP:%.*]] = icmp ne i32 [[T0]], 1
 ; CHECK-NEXT:    call void @use_i1(i1 [[CMP]])
@@ -1069,7 +1069,7 @@ define i1 @isnot_pow2nor0_ctpop_extra_uses(i32 %x) {
 
 define i1 @isnot_pow2nor0_ctpop_logical_extra_uses(i32 %x) {
 ; CHECK-LABEL: @isnot_pow2nor0_ctpop_logical_extra_uses(
-; CHECK-NEXT:    [[T0:%.*]] = tail call i32 @llvm.ctpop.i32(i32 [[X:%.*]]), !range [[RNG0]]
+; CHECK-NEXT:    [[T0:%.*]] = tail call range(i32 0, 33) i32 @llvm.ctpop.i32(i32 [[X:%.*]])
 ; CHECK-NEXT:    call void @use(i32 [[T0]])
 ; CHECK-NEXT:    [[CMP:%.*]] = icmp ne i32 [[T0]], 1
 ; CHECK-NEXT:    call void @use_i1(i1 [[CMP]])
@@ -1092,7 +1092,7 @@ define i1 @isnot_pow2nor0_ctpop_logical_extra_uses(i32 %x) {
 
 define i1 @isnot_pow2nor0_ctpop_wrong_cmp_op1(i32 %x) {
 ; CHECK-LABEL: @isnot_pow2nor0_ctpop_wrong_cmp_op1(
-; CHECK-NEXT:    [[T0:%.*]] = tail call i32 @llvm.ctpop.i32(i32 [[X:%.*]]), !range [[RNG0]]
+; CHECK-NEXT:    [[T0:%.*]] = tail call range(i32 0, 33) i32 @llvm.ctpop.i32(i32 [[X:%.*]])
 ; CHECK-NEXT:    [[CMP:%.*]] = icmp ne i32 [[T0]], 4
 ; CHECK-NEXT:    [[NOTZERO:%.*]] = icmp ne i32 [[X]], 0
 ; CHECK-NEXT:    [[R:%.*]] = and i1 [[NOTZERO]], [[CMP]]
@@ -1107,7 +1107,7 @@ define i1 @isnot_pow2nor0_ctpop_wrong_cmp_op1(i32 %x) {
 
 define i1 @isnot_pow2nor0_ctpop_wrong_cmp_op1_logical(i32 %x) {
 ; CHECK-LABEL: @isnot_pow2nor0_ctpop_wrong_cmp_op1_logical(
-; CHECK-NEXT:    [[T0:%.*]] = tail call i32 @llvm.ctpop.i32(i32 [[X:%.*]]), !range [[RNG0]]
+; CHECK-NEXT:    [[T0:%.*]] = tail call range(i32 0, 33) i32 @llvm.ctpop.i32(i32 [[X:%.*]])
 ; CHECK-NEXT:    [[CMP:%.*]] = icmp ne i32 [[T0]], 5
 ; CHECK-NEXT:    [[NOTZERO:%.*]] = icmp ne i32 [[X]], 0
 ; CHECK-NEXT:    [[R:%.*]] = select i1 [[NOTZERO]], i1 [[CMP]], i1 false
@@ -1122,7 +1122,7 @@ define i1 @isnot_pow2nor0_ctpop_wrong_cmp_op1_logical(i32 %x) {
 
 define <2 x i1> @isnot_pow2nor0_ctpop_commute_vec_wrong_cmp_op1(<2 x i8> %x) {
 ; CHECK-LABEL: @isnot_pow2nor0_ctpop_commute_vec_wrong_cmp_op1(
-; CHECK-NEXT:    [[T0:%.*]] = tail call <2 x i8> @llvm.ctpop.v2i8(<2 x i8> [[X:%.*]]), !range [[RNG1]]
+; CHECK-NEXT:    [[T0:%.*]] = tail call range(i8 0, 9) <2 x i8> @llvm.ctpop.v2i8(<2 x i8> [[X:%.*]])
 ; CHECK-NEXT:    [[CMP:%.*]] = icmp ne <2 x i8> [[T0]], <i8 0, i8 -1>
 ; CHECK-NEXT:    [[NOTZERO:%.*]] = icmp ne <2 x i8> [[X]], zeroinitializer
 ; CHECK-NEXT:    [[R:%.*]] = and <2 x i1> [[CMP]], [[NOTZERO]]
@@ -1139,7 +1139,7 @@ define <2 x i1> @isnot_pow2nor0_ctpop_commute_vec_wrong_cmp_op1(<2 x i8> %x) {
 
 define i1 @isnot_pow2nor0_ctpop_wrong_pred1(i32 %x) {
 ; CHECK-LABEL: @isnot_pow2nor0_ctpop_wrong_pred1(
-; CHECK-NEXT:    [[T0:%.*]] = tail call i32 @llvm.ctpop.i32(i32 [[X:%.*]]), !range [[RNG0]]
+; CHECK-NEXT:    [[T0:%.*]] = tail call range(i32 0, 33) i32 @llvm.ctpop.i32(i32 [[X:%.*]])
 ; CHECK-NEXT:    [[CMP:%.*]] = icmp eq i32 [[T0]], 1
 ; CHECK-NEXT:    ret i1 [[CMP]]
 ;
@@ -1174,7 +1174,7 @@ define i1 @isnot_pow2nor0_ctpop_wrong_pred2_logical(i32 %x) {
 
 define <2 x i1> @isnot_pow2nor0_wrong_pred3_ctpop_commute_vec(<2 x i8> %x) {
 ; CHECK-LABEL: @isnot_pow2nor0_wrong_pred3_ctpop_commute_vec(
-; CHECK-NEXT:    [[T0:%.*]] = tail call <2 x i8> @llvm.ctpop.v2i8(<2 x i8> [[X:%.*]]), !range [[RNG1]]
+; CHECK-NEXT:    [[T0:%.*]] = tail call range(i8 0, 9) <2 x i8> @llvm.ctpop.v2i8(<2 x i8> [[X:%.*]])
 ; CHECK-NEXT:    [[CMP:%.*]] = icmp ne <2 x i8> [[T0]], <i8 1, i8 1>
 ; CHECK-NEXT:    [[NOTZERO:%.*]] = icmp ne <2 x i8> [[X]], zeroinitializer
 ; CHECK-NEXT:    [[R:%.*]] = or <2 x i1> [[CMP]], [[NOTZERO]]
@@ -1217,7 +1217,7 @@ define i1 @blsmsk_is_p2_or_z(i32 %xx, i32 %yy) {
 
 define i1 @blsmsk_isnt_p2_or_z(i32 %x) {
 ; CHECK-LABEL: @blsmsk_isnt_p2_or_z(
-; CHECK-NEXT:    [[TMP1:%.*]] = call i32 @llvm.ctpop.i32(i32 [[X:%.*]]), !range [[RNG0]]
+; CHECK-NEXT:    [[TMP1:%.*]] = call range(i32 0, 33) i32 @llvm.ctpop.i32(i32 [[X:%.*]])
 ; CHECK-NEXT:    [[R:%.*]] = icmp ugt i32 [[TMP1]], 1
 ; CHECK-NEXT:    ret i1 [[R]]
 ;
@@ -1230,7 +1230,7 @@ define i1 @blsmsk_isnt_p2_or_z(i32 %x) {
 define i1 @blsmsk_is_p2_or_z_fail(i32 %xx, i32 %yy) {
 ; CHECK-LABEL: @blsmsk_is_p2_or_z_fail(
 ; CHECK-NEXT:    [[X:%.*]] = or i32 [[XX:%.*]], [[YY:%.*]]
-; CHECK-NEXT:    [[TMP1:%.*]] = call i32 @llvm.ctpop.i32(i32 [[X]]), !range [[RNG0]]
+; CHECK-NEXT:    [[TMP1:%.*]] = call range(i32 0, 33) i32 @llvm.ctpop.i32(i32 [[X]])
 ; CHECK-NEXT:    [[R:%.*]] = icmp ugt i32 [[TMP1]], 1
 ; CHECK-NEXT:    ret i1 [[R]]
 ;
@@ -1315,7 +1315,7 @@ define i1 @blsmsk_is_p2_or_z_fail_bad_cmp(i32 %x, i32 %z) {
 define i1 @blsmsk_is_p2_or_z_ule_xy(i8 %xx, i8 %yy) {
 ; CHECK-LABEL: @blsmsk_is_p2_or_z_ule_xy(
 ; CHECK-NEXT:    [[X:%.*]] = or i8 [[XX:%.*]], [[YY:%.*]]
-; CHECK-NEXT:    [[TMP1:%.*]] = call i8 @llvm.ctpop.i8(i8 [[X]]), !range [[RNG1]]
+; CHECK-NEXT:    [[TMP1:%.*]] = call range(i8 0, 9) i8 @llvm.ctpop.i8(i8 [[X]])
 ; CHECK-NEXT:    [[R:%.*]] = icmp ult i8 [[TMP1]], 2
 ; CHECK-NEXT:    ret i1 [[R]]
 ;
@@ -1346,7 +1346,7 @@ define i1 @blsmsk_is_p2_or_z_ule_yx_fail(i8 %xx, i8 %yy) {
 define i1 @blsmsk_is_p2_or_z_uge_yx(i8 %xx, i8 %yy) {
 ; CHECK-LABEL: @blsmsk_is_p2_or_z_uge_yx(
 ; CHECK-NEXT:    [[X:%.*]] = or i8 [[XX:%.*]], [[YY:%.*]]
-; CHECK-NEXT:    [[TMP1:%.*]] = call i8 @llvm.ctpop.i8(i8 [[X]]), !range [[RNG1]]
+; CHECK-NEXT:    [[TMP1:%.*]] = call range(i8 0, 9) i8 @llvm.ctpop.i8(i8 [[X]])
 ; CHECK-NEXT:    [[R:%.*]] = icmp ult i8 [[TMP1]], 2
 ; CHECK-NEXT:    ret i1 [[R]]
 ;
@@ -1376,7 +1376,7 @@ define i1 @blsmsk_is_p2_or_z_uge_xy_fail(i8 %xx, i8 %yy) {
 define i1 @blsmsk_isnt_p2_or_z_ugt_xy(i8 %xx, i8 %yy) {
 ; CHECK-LABEL: @blsmsk_isnt_p2_or_z_ugt_xy(
 ; CHECK-NEXT:    [[X:%.*]] = or i8 [[XX:%.*]], [[YY:%.*]]
-; CHECK-NEXT:    [[TMP1:%.*]] = call i8 @llvm.ctpop.i8(i8 [[X]]), !range [[RNG1]]
+; CHECK-NEXT:    [[TMP1:%.*]] = call range(i8 0, 9) i8 @llvm.ctpop.i8(i8 [[X]])
 ; CHECK-NEXT:    [[R:%.*]] = icmp ugt i8 [[TMP1]], 1
 ; CHECK-NEXT:    ret i1 [[R]]
 ;
@@ -1407,7 +1407,7 @@ define i1 @blsmsk_isnt_p2_or_z_ugt_yx_fail(i8 %xx, i8 %yy) {
 define i1 @blsmsk_isnt_p2_or_z_ult_yx(i8 %xx, i8 %yy) {
 ; CHECK-LABEL: @blsmsk_isnt_p2_or_z_ult_yx(
 ; CHECK-NEXT:    [[X:%.*]] = or i8 [[XX:%.*]], [[YY:%.*]]
-; CHECK-NEXT:    [[TMP1:%.*]] = call i8 @llvm.ctpop.i8(i8 [[X]]), !range [[RNG1]]
+; CHECK-NEXT:    [[TMP1:%.*]] = call range(i8 0, 9) i8 @llvm.ctpop.i8(i8 [[X]])
 ; CHECK-NEXT:    [[R:%.*]] = icmp ugt i8 [[TMP1]], 1
 ; CHECK-NEXT:    ret i1 [[R]]
 ;
@@ -1450,7 +1450,7 @@ define i1 @is_pow2_nz_known_bits(i32 %xin) {
 define i1 @is_pow2_nz_known_bits_fail_multiuse(i32 %xin) {
 ; CHECK-LABEL: @is_pow2_nz_known_bits_fail_multiuse(
 ; CHECK-NEXT:    [[X:%.*]] = or i32 [[XIN:%.*]], 64
-; CHECK-NEXT:    [[CNT:%.*]] = call i32 @llvm.ctpop.i32(i32 [[X]]), !range [[RNG3:![0-9]+]]
+; CHECK-NEXT:    [[CNT:%.*]] = call range(i32 1, 33) i32 @llvm.ctpop.i32(i32 [[X]])
 ; CHECK-NEXT:    call void @use.i32(i32 [[CNT]])
 ; CHECK-NEXT:    [[R:%.*]] = icmp eq i32 [[CNT]], 1
 ; CHECK-NEXT:    ret i1 [[R]]
@@ -1476,7 +1476,7 @@ define i1 @not_pow2_nz_known_bits(i32 %xin) {
 define i1 @not_pow2_nz_known_bits_fail_not_p2_test(i32 %xin) {
 ; CHECK-LABEL: @not_pow2_nz_known_bits_fail_not_p2_test(
 ; CHECK-NEXT:    [[X:%.*]] = or i32 [[XIN:%.*]], 1
-; CHECK-NEXT:    [[CNT:%.*]] = call i32 @llvm.ctpop.i32(i32 [[X]]), !range [[RNG3]]
+; CHECK-NEXT:    [[CNT:%.*]] = call range(i32 1, 33) i32 @llvm.ctpop.i32(i32 [[X]])
 ; CHECK-NEXT:    [[R:%.*]] = icmp ne i32 [[CNT]], 2
 ; CHECK-NEXT:    ret i1 [[R]]
 ;
@@ -1513,7 +1513,7 @@ define <2 x i1> @not_pow2_or_z_known_bits(<2 x i32> %xin) {
 define <2 x i1> @not_pow2_or_z_known_bits_fail_wrong_cmp(<2 x i32> %xin) {
 ; CHECK-LABEL: @not_pow2_or_z_known_bits_fail_wrong_cmp(
 ; CHECK-NEXT:    [[X:%.*]] = or <2 x i32> [[XIN:%.*]], <i32 64, i32 64>
-; CHECK-NEXT:    [[CNT:%.*]] = call <2 x i32> @llvm.ctpop.v2i32(<2 x i32> [[X]]), !range [[RNG3]]
+; CHECK-NEXT:    [[CNT:%.*]] = call range(i32 1, 33) <2 x i32> @llvm.ctpop.v2i32(<2 x i32> [[X]])
 ; CHECK-NEXT:    [[R:%.*]] = icmp ugt <2 x i32> [[CNT]], <i32 2, i32 2>
 ; CHECK-NEXT:    ret <2 x i1> [[R]]
 ;
diff --git a/llvm/test/Transforms/InstCombine/known-non-zero.ll b/llvm/test/Transforms/InstCombine/known-non-zero.ll
index f1c757cafefb..b77c04eb8147 100644
--- a/llvm/test/Transforms/InstCombine/known-non-zero.ll
+++ b/llvm/test/Transforms/InstCombine/known-non-zero.ll
@@ -13,7 +13,7 @@ define i32 @test0(i64 %x) {
 ; CHECK-NEXT:    [[C:%.*]] = icmp eq i64 [[X:%.*]], 0
 ; CHECK-NEXT:    br i1 [[C]], label [[EXIT:%.*]], label [[NON_ZERO:%.*]]
 ; CHECK:       non_zero:
-; CHECK-NEXT:    [[CTZ:%.*]] = call i64 @llvm.cttz.i64(i64 [[X]], i1 true), !range [[RNG0:![0-9]+]]
+; CHECK-NEXT:    [[CTZ:%.*]] = call range(i64 0, 65) i64 @llvm.cttz.i64(i64 [[X]], i1 true)
 ; CHECK-NEXT:    [[CTZ32:%.*]] = trunc nuw nsw i64 [[CTZ]] to i32
 ; CHECK-NEXT:    br label [[EXIT]]
 ; CHECK:       exit:
@@ -40,7 +40,7 @@ define i32 @test1(i64 %x) {
 ; CHECK-NEXT:    [[C:%.*]] = icmp eq i64 [[X:%.*]], 0
 ; CHECK-NEXT:    br i1 [[C]], label [[EXIT:%.*]], label [[NON_ZERO:%.*]]
 ; CHECK:       non_zero:
-; CHECK-NEXT:    [[CTZ:%.*]] = call i64 @llvm.ctlz.i64(i64 [[X]], i1 true), !range [[RNG0]]
+; CHECK-NEXT:    [[CTZ:%.*]] = call range(i64 0, 65) i64 @llvm.ctlz.i64(i64 [[X]], i1 true)
 ; CHECK-NEXT:    [[CTZ32:%.*]] = trunc nuw nsw i64 [[CTZ]] to i32
 ; CHECK-NEXT:    br label [[EXIT]]
 ; CHECK:       exit:
@@ -69,7 +69,7 @@ define <8 x i64> @test2(<8 x i64> %x) {
 ; CHECK-NEXT:    [[C:%.*]] = icmp eq i8 [[B]], 0
 ; CHECK-NEXT:    br i1 [[C]], label [[EXIT:%.*]], label [[NON_ZERO:%.*]]
 ; CHECK:       non_zero:
-; CHECK-NEXT:    [[CTZ:%.*]] = call <8 x i64> @llvm.cttz.v8i64(<8 x i64> [[X]], i1 false), !range [[RNG0]]
+; CHECK-NEXT:    [[CTZ:%.*]] = call range(i64 0, 65) <8 x i64> @llvm.cttz.v8i64(<8 x i64> [[X]], i1 false)
 ; CHECK-NEXT:    br label [[EXIT]]
 ; CHECK:       exit:
 ; CHECK-NEXT:    [[RES:%.*]] = phi <8 x i64> [ [[CTZ]], [[NON_ZERO]] ], [ zeroinitializer, [[START:%.*]] ]
@@ -140,7 +140,7 @@ define i64 @test_sgt_zero(i64 %x) {
 ; CHECK-NEXT:    [[C:%.*]] = icmp sgt i64 [[X:%.*]], 0
 ; CHECK-NEXT:    br i1 [[C]], label [[NON_ZERO:%.*]], label [[EXIT:%.*]]
 ; CHECK:       non_zero:
-; CHECK-NEXT:    [[CTZ:%.*]] = call i64 @llvm.ctlz.i64(i64 [[X]], i1 true), !range [[RNG1:![0-9]+]]
+; CHECK-NEXT:    [[CTZ:%.*]] = call range(i64 1, 65) i64 @llvm.ctlz.i64(i64 [[X]], i1 true)
 ; CHECK-NEXT:    ret i64 [[CTZ]]
 ; CHECK:       exit:
 ; CHECK-NEXT:    ret i64 -1
@@ -185,7 +185,7 @@ define i64 @test_slt_ten(i64 %x) {
 ; CHECK-NEXT:    [[C:%.*]] = icmp slt i64 [[X:%.*]], 10
 ; CHECK-NEXT:    br i1 [[C]], label [[MAYBE_ZERO:%.*]], label [[EXIT:%.*]]
 ; CHECK:       maybe_zero:
-; CHECK-NEXT:    [[CTZ:%.*]] = call i64 @llvm.ctlz.i64(i64 [[X]], i1 false), !range [[RNG0]]
+; CHECK-NEXT:    [[CTZ:%.*]] = call range(i64 0, 65) i64 @llvm.ctlz.i64(i64 [[X]], i1 false)
 ; CHECK-NEXT:    ret i64 [[CTZ]]
 ; CHECK:       exit:
 ; CHECK-NEXT:    ret i64 -1
@@ -208,7 +208,7 @@ define i64 @test_ugt_unknown(i64 %x, i64 %y) {
 ; CHECK-NEXT:    [[C:%.*]] = icmp ugt i64 [[X:%.*]], [[Y:%.*]]
 ; CHECK-NEXT:    br i1 [[C]], label [[NON_ZERO:%.*]], label [[EXIT:%.*]]
 ; CHECK:       non_zero:
-; CHECK-NEXT:    [[CTZ:%.*]] = call i64 @llvm.ctlz.i64(i64 [[X]], i1 true), !range [[RNG0]]
+; CHECK-NEXT:    [[CTZ:%.*]] = call range(i64 0, 65) i64 @llvm.ctlz.i64(i64 [[X]], i1 true)
 ; CHECK-NEXT:    ret i64 [[CTZ]]
 ; CHECK:       exit:
 ; CHECK-NEXT:    ret i64 -1
@@ -231,7 +231,7 @@ define i64 @test_sle_zero(i64 %x) {
 ; CHECK-NEXT:    [[C:%.*]] = icmp slt i64 [[X:%.*]], 1
 ; CHECK-NEXT:    br i1 [[C]], label [[EXIT:%.*]], label [[NON_ZERO:%.*]]
 ; CHECK:       non_zero:
-; CHECK-NEXT:    [[CTZ:%.*]] = call i64 @llvm.ctlz.i64(i64 [[X]], i1 true), !range [[RNG1]]
+; CHECK-NEXT:    [[CTZ:%.*]] = call range(i64 1, 65) i64 @llvm.ctlz.i64(i64 [[X]], i1 true)
 ; CHECK-NEXT:    ret i64 [[CTZ]]
 ; CHECK:       exit:
 ; CHECK-NEXT:    ret i64 -1
@@ -276,7 +276,7 @@ define i64 @test_sge_ten(i64 %x) {
 ; CHECK-NEXT:    [[C:%.*]] = icmp sgt i64 [[X:%.*]], 9
 ; CHECK-NEXT:    br i1 [[C]], label [[EXIT:%.*]], label [[MAYBE_ZERO:%.*]]
 ; CHECK:       maybe_zero:
-; CHECK-NEXT:    [[CTZ:%.*]] = call i64 @llvm.ctlz.i64(i64 [[X]], i1 false), !range [[RNG0]]
+; CHECK-NEXT:    [[CTZ:%.*]] = call range(i64 0, 65) i64 @llvm.ctlz.i64(i64 [[X]], i1 false)
 ; CHECK-NEXT:    ret i64 [[CTZ]]
 ; CHECK:       exit:
 ; CHECK-NEXT:    ret i64 -1
@@ -299,7 +299,7 @@ define i64 @test_ule_unknown(i64 %x, i64 %y) {
 ; CHECK-NEXT:    [[C_NOT:%.*]] = icmp ugt i64 [[X:%.*]], [[Y:%.*]]
 ; CHECK-NEXT:    br i1 [[C_NOT]], label [[NON_ZERO:%.*]], label [[EXIT:%.*]]
 ; CHECK:       non_zero:
-; CHECK-NEXT:    [[CTZ:%.*]] = call i64 @llvm.ctlz.i64(i64 [[X]], i1 true), !range [[RNG0]]
+; CHECK-NEXT:    [[CTZ:%.*]] = call range(i64 0, 65) i64 @llvm.ctlz.i64(i64 [[X]], i1 true)
 ; CHECK-NEXT:    ret i64 [[CTZ]]
 ; CHECK:       exit:
 ; CHECK-NEXT:    ret i64 -1
diff --git a/llvm/test/Transforms/InstCombine/known-phi-recurse.ll b/llvm/test/Transforms/InstCombine/known-phi-recurse.ll
index d33e08ffaf9b..c2007d16ae93 100644
--- a/llvm/test/Transforms/InstCombine/known-phi-recurse.ll
+++ b/llvm/test/Transforms/InstCombine/known-phi-recurse.ll
@@ -16,7 +16,7 @@ define i32 @single_entry_phi(i64 %x, i1 %c) {
 ; CHECK:       body:
 ; CHECK-NEXT:    br i1 [[C:%.*]], label [[END:%.*]], label [[BODY]]
 ; CHECK:       end:
-; CHECK-NEXT:    [[Y:%.*]] = call i64 @llvm.ctpop.i64(i64 [[X:%.*]]), !range [[RNG0:![0-9]+]]
+; CHECK-NEXT:    [[Y:%.*]] = call range(i64 0, 65) i64 @llvm.ctpop.i64(i64 [[X:%.*]])
 ; CHECK-NEXT:    [[TRUNC:%.*]] = trunc nuw nsw i64 [[Y]] to i32
 ; CHECK-NEXT:    ret i32 [[TRUNC]]
 ;
@@ -36,7 +36,7 @@ end:
 define i32 @two_entry_phi_with_constant(i64 %x, i1 %c) {
 ; CHECK-LABEL: @two_entry_phi_with_constant(
 ; CHECK-NEXT:  entry:
-; CHECK-NEXT:    [[Y:%.*]] = call i64 @llvm.ctpop.i64(i64 [[X:%.*]]), !range [[RNG0]]
+; CHECK-NEXT:    [[Y:%.*]] = call range(i64 0, 65) i64 @llvm.ctpop.i64(i64 [[X:%.*]])
 ; CHECK-NEXT:    [[TRUNC:%.*]] = trunc nuw nsw i64 [[Y]] to i32
 ; CHECK-NEXT:    br i1 [[C:%.*]], label [[END:%.*]], label [[BODY:%.*]]
 ; CHECK:       body:
@@ -61,11 +61,11 @@ end:
 define i32 @two_entry_phi_non_constant(i64 %x, i64 %x2, i1 %c) {
 ; CHECK-LABEL: @two_entry_phi_non_constant(
 ; CHECK-NEXT:  entry:
-; CHECK-NEXT:    [[Y:%.*]] = call i64 @llvm.ctpop.i64(i64 [[X:%.*]]), !range [[RNG0]]
+; CHECK-NEXT:    [[Y:%.*]] = call range(i64 0, 65) i64 @llvm.ctpop.i64(i64 [[X:%.*]])
 ; CHECK-NEXT:    [[TRUNC:%.*]] = trunc nuw nsw i64 [[Y]] to i32
 ; CHECK-NEXT:    br i1 [[C:%.*]], label [[END:%.*]], label [[BODY:%.*]]
 ; CHECK:       body:
-; CHECK-NEXT:    [[Y2:%.*]] = call i64 @llvm.ctpop.i64(i64 [[X2:%.*]]), !range [[RNG0]]
+; CHECK-NEXT:    [[Y2:%.*]] = call range(i64 0, 65) i64 @llvm.ctpop.i64(i64 [[X2:%.*]])
 ; CHECK-NEXT:    [[TRUNC2:%.*]] = trunc nuw nsw i64 [[Y2]] to i32
 ; CHECK-NEXT:    br label [[END]]
 ; CHECK:       end:
@@ -90,7 +90,7 @@ end:
 define i32 @neg_many_branches(i64 %x) {
 ; CHECK-LABEL: @neg_many_branches(
 ; CHECK-NEXT:  entry:
-; CHECK-NEXT:    [[Y:%.*]] = call i64 @llvm.ctpop.i64(i64 [[X:%.*]]), !range [[RNG0]]
+; CHECK-NEXT:    [[Y:%.*]] = call range(i64 0, 65) i64 @llvm.ctpop.i64(i64 [[X:%.*]])
 ; CHECK-NEXT:    [[TRUNC:%.*]] = trunc nuw nsw i64 [[Y]] to i32
 ; CHECK-NEXT:    switch i32 [[TRUNC]], label [[END:%.*]] [
 ; CHECK-NEXT:      i32 1, label [[ONE:%.*]]
diff --git a/llvm/test/Transforms/InstCombine/loadstore-alignment.ll b/llvm/test/Transforms/InstCombine/loadstore-alignment.ll
index 0fc82a1d5343..1027468d6715 100644
--- a/llvm/test/Transforms/InstCombine/loadstore-alignment.ll
+++ b/llvm/test/Transforms/InstCombine/loadstore-alignment.ll
@@ -9,24 +9,24 @@ target datalayout = "E-p:64:64:64-p1:64:64:64-p2:32:32:32-a0:0:8-f32:32:32-f64:6
 
 define <2 x i64> @static_hem() {
 ; CHECK-LABEL: @static_hem(
-; CHECK-NEXT:    [[TMP1:%.*]] = load <2 x i64>, ptr getelementptr (<2 x i64>, ptr @x, i64 7), align 1
-; CHECK-NEXT:    ret <2 x i64> [[TMP1]]
+; CHECK-NEXT:    [[L:%.*]] = load <2 x i64>, ptr getelementptr (<2 x i64>, ptr @x, i64 7), align 1
+; CHECK-NEXT:    ret <2 x i64> [[L]]
 ;
   %t = getelementptr <2 x i64>, ptr @x, i32 7
-  %tmp1 = load <2 x i64>, ptr %t, align 1
-  ret <2 x i64> %tmp1
+  %l = load <2 x i64>, ptr %t, align 1
+  ret <2 x i64> %l
 }
 
 define <2 x i64> @hem(i32 %i) {
 ; CHECK-LABEL: @hem(
 ; CHECK-NEXT:    [[TMP1:%.*]] = sext i32 [[I:%.*]] to i64
 ; CHECK-NEXT:    [[T:%.*]] = getelementptr <2 x i64>, ptr @x, i64 [[TMP1]]
-; CHECK-NEXT:    [[TMP1:%.*]] = load <2 x i64>, ptr [[T]], align 1
-; CHECK-NEXT:    ret <2 x i64> [[TMP1]]
+; CHECK-NEXT:    [[L:%.*]] = load <2 x i64>, ptr [[T]], align 1
+; CHECK-NEXT:    ret <2 x i64> [[L]]
 ;
   %t = getelementptr <2 x i64>, ptr @x, i32 %i
-  %tmp1 = load <2 x i64>, ptr %t, align 1
-  ret <2 x i64> %tmp1
+  %l = load <2 x i64>, ptr %t, align 1
+  ret <2 x i64> %l
 }
 
 define <2 x i64> @hem_2d(i32 %i, i32 %j) {
@@ -34,34 +34,34 @@ define <2 x i64> @hem_2d(i32 %i, i32 %j) {
 ; CHECK-NEXT:    [[TMP1:%.*]] = sext i32 [[I:%.*]] to i64
 ; CHECK-NEXT:    [[TMP2:%.*]] = sext i32 [[J:%.*]] to i64
 ; CHECK-NEXT:    [[T:%.*]] = getelementptr [13 x <2 x i64>], ptr @xx, i64 [[TMP1]], i64 [[TMP2]]
-; CHECK-NEXT:    [[TMP1:%.*]] = load <2 x i64>, ptr [[T]], align 1
-; CHECK-NEXT:    ret <2 x i64> [[TMP1]]
+; CHECK-NEXT:    [[L:%.*]] = load <2 x i64>, ptr [[T]], align 1
+; CHECK-NEXT:    ret <2 x i64> [[L]]
 ;
   %t = getelementptr [13 x <2 x i64>], ptr @xx, i32 %i, i32 %j
-  %tmp1 = load <2 x i64>, ptr %t, align 1
-  ret <2 x i64> %tmp1
+  %l = load <2 x i64>, ptr %t, align 1
+  ret <2 x i64> %l
 }
 
 define <2 x i64> @foo() {
 ; CHECK-LABEL: @foo(
-; CHECK-NEXT:    [[TMP1:%.*]] = load <2 x i64>, ptr @x, align 1
-; CHECK-NEXT:    ret <2 x i64> [[TMP1]]
+; CHECK-NEXT:    [[L:%.*]] = load <2 x i64>, ptr @x, align 1
+; CHECK-NEXT:    ret <2 x i64> [[L]]
 ;
-  %tmp1 = load <2 x i64>, ptr @x, align 1
-  ret <2 x i64> %tmp1
+  %l = load <2 x i64>, ptr @x, align 1
+  ret <2 x i64> %l
 }
 
 define <2 x i64> @bar() {
 ; CHECK-LABEL: @bar(
 ; CHECK-NEXT:    [[T:%.*]] = alloca <2 x i64>, align 16
 ; CHECK-NEXT:    call void @kip(ptr nonnull [[T]])
-; CHECK-NEXT:    [[TMP1:%.*]] = load <2 x i64>, ptr [[T]], align 1
-; CHECK-NEXT:    ret <2 x i64> [[TMP1]]
+; CHECK-NEXT:    [[L:%.*]] = load <2 x i64>, ptr [[T]], align 1
+; CHECK-NEXT:    ret <2 x i64> [[L]]
 ;
   %t = alloca <2 x i64>
   call void @kip(ptr %t)
-  %tmp1 = load <2 x i64>, ptr %t, align 1
-  ret <2 x i64> %tmp1
+  %l = load <2 x i64>, ptr %t, align 1
+  ret <2 x i64> %l
 }
 
 define void @static_hem_store(<2 x i64> %y) {
diff --git a/llvm/test/Transforms/InstCombine/memcpy-from-global.ll b/llvm/test/Transforms/InstCombine/memcpy-from-global.ll
index aeca0cd2924e..e9ff34735f1c 100644
--- a/llvm/test/Transforms/InstCombine/memcpy-from-global.ll
+++ b/llvm/test/Transforms/InstCombine/memcpy-from-global.ll
@@ -6,60 +6,60 @@ target datalayout = "E-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:32:64-f3
 define float @test1(i32 %hash, float %x, float %y, float %z, float %w) {
 ; CHECK-LABEL: @test1(
 ; CHECK-NEXT:  entry:
-; CHECK-NEXT:    [[TMP3:%.*]] = shl i32 [[HASH:%.*]], 2
-; CHECK-NEXT:    [[TMP5:%.*]] = and i32 [[TMP3]], 124
-; CHECK-NEXT:    [[TMP0:%.*]] = zext nneg i32 [[TMP5]] to i64
-; CHECK-NEXT:    [[TMP753:%.*]] = getelementptr [128 x float], ptr @C.0.1248, i64 0, i64 [[TMP0]]
-; CHECK-NEXT:    [[TMP9:%.*]] = load float, ptr [[TMP753]], align 4
-; CHECK-NEXT:    [[TMP11:%.*]] = fmul float [[TMP9]], [[X:%.*]]
-; CHECK-NEXT:    [[TMP13:%.*]] = fadd float [[TMP11]], 0.000000e+00
-; CHECK-NEXT:    [[TMP17_SUM52:%.*]] = or disjoint i32 [[TMP5]], 1
-; CHECK-NEXT:    [[TMP1:%.*]] = zext nneg i32 [[TMP17_SUM52]] to i64
-; CHECK-NEXT:    [[TMP1851:%.*]] = getelementptr [128 x float], ptr @C.0.1248, i64 0, i64 [[TMP1]]
-; CHECK-NEXT:    [[TMP19:%.*]] = load float, ptr [[TMP1851]], align 4
-; CHECK-NEXT:    [[TMP21:%.*]] = fmul float [[TMP19]], [[Y:%.*]]
-; CHECK-NEXT:    [[TMP23:%.*]] = fadd float [[TMP21]], [[TMP13]]
-; CHECK-NEXT:    [[TMP27_SUM50:%.*]] = or disjoint i32 [[TMP5]], 2
-; CHECK-NEXT:    [[TMP2:%.*]] = zext nneg i32 [[TMP27_SUM50]] to i64
-; CHECK-NEXT:    [[TMP2849:%.*]] = getelementptr [128 x float], ptr @C.0.1248, i64 0, i64 [[TMP2]]
-; CHECK-NEXT:    [[TMP29:%.*]] = load float, ptr [[TMP2849]], align 4
-; CHECK-NEXT:    [[TMP31:%.*]] = fmul float [[TMP29]], [[Z:%.*]]
-; CHECK-NEXT:    [[TMP33:%.*]] = fadd float [[TMP31]], [[TMP23]]
-; CHECK-NEXT:    [[TMP37_SUM48:%.*]] = or disjoint i32 [[TMP5]], 3
-; CHECK-NEXT:    [[TMP3:%.*]] = zext nneg i32 [[TMP37_SUM48]] to i64
-; CHECK-NEXT:    [[TMP3847:%.*]] = getelementptr [128 x float], ptr @C.0.1248, i64 0, i64 [[TMP3]]
-; CHECK-NEXT:    [[TMP39:%.*]] = load float, ptr [[TMP3847]], align 4
-; CHECK-NEXT:    [[TMP41:%.*]] = fmul float [[TMP39]], [[W:%.*]]
-; CHECK-NEXT:    [[TMP43:%.*]] = fadd float [[TMP41]], [[TMP33]]
-; CHECK-NEXT:    ret float [[TMP43]]
+; CHECK-NEXT:    [[T3:%.*]] = shl i32 [[HASH:%.*]], 2
+; CHECK-NEXT:    [[T5:%.*]] = and i32 [[T3]], 124
+; CHECK-NEXT:    [[TMP0:%.*]] = zext nneg i32 [[T5]] to i64
+; CHECK-NEXT:    [[T753:%.*]] = getelementptr [128 x float], ptr @C.0.1248, i64 0, i64 [[TMP0]]
+; CHECK-NEXT:    [[T9:%.*]] = load float, ptr [[T753]], align 4
+; CHECK-NEXT:    [[T11:%.*]] = fmul float [[T9]], [[X:%.*]]
+; CHECK-NEXT:    [[T13:%.*]] = fadd float [[T11]], 0.000000e+00
+; CHECK-NEXT:    [[T17_SUM52:%.*]] = or disjoint i32 [[T5]], 1
+; CHECK-NEXT:    [[TMP1:%.*]] = zext nneg i32 [[T17_SUM52]] to i64
+; CHECK-NEXT:    [[T1851:%.*]] = getelementptr [128 x float], ptr @C.0.1248, i64 0, i64 [[TMP1]]
+; CHECK-NEXT:    [[T19:%.*]] = load float, ptr [[T1851]], align 4
+; CHECK-NEXT:    [[T21:%.*]] = fmul float [[T19]], [[Y:%.*]]
+; CHECK-NEXT:    [[T23:%.*]] = fadd float [[T21]], [[T13]]
+; CHECK-NEXT:    [[T27_SUM50:%.*]] = or disjoint i32 [[T5]], 2
+; CHECK-NEXT:    [[TMP2:%.*]] = zext nneg i32 [[T27_SUM50]] to i64
+; CHECK-NEXT:    [[T2849:%.*]] = getelementptr [128 x float], ptr @C.0.1248, i64 0, i64 [[TMP2]]
+; CHECK-NEXT:    [[T29:%.*]] = load float, ptr [[T2849]], align 4
+; CHECK-NEXT:    [[T31:%.*]] = fmul float [[T29]], [[Z:%.*]]
+; CHECK-NEXT:    [[T33:%.*]] = fadd float [[T31]], [[T23]]
+; CHECK-NEXT:    [[T37_SUM48:%.*]] = or disjoint i32 [[T5]], 3
+; CHECK-NEXT:    [[TMP3:%.*]] = zext nneg i32 [[T37_SUM48]] to i64
+; CHECK-NEXT:    [[T3847:%.*]] = getelementptr [128 x float], ptr @C.0.1248, i64 0, i64 [[TMP3]]
+; CHECK-NEXT:    [[T39:%.*]] = load float, ptr [[T3847]], align 4
+; CHECK-NEXT:    [[T41:%.*]] = fmul float [[T39]], [[W:%.*]]
+; CHECK-NEXT:    [[T43:%.*]] = fadd float [[T41]], [[T33]]
+; CHECK-NEXT:    ret float [[T43]]
 ;
 entry:
-  %lookupTable = alloca [128 x float], align 16		; <ptr> [#uses=5]
+  %lookupTable = alloca [128 x float], align 16
   call void @llvm.memcpy.p0.p0.i64(ptr align 16 %lookupTable, ptr align 16 @C.0.1248, i64 512, i1 false)
 
 
-  %tmp3 = shl i32 %hash, 2		; <i32> [#uses=1]
-  %tmp5 = and i32 %tmp3, 124		; <i32> [#uses=4]
-  %tmp753 = getelementptr [128 x float], ptr %lookupTable, i32 0, i32 %tmp5		; <ptr> [#uses=1]
-  %tmp9 = load float, ptr %tmp753		; <float> [#uses=1]
-  %tmp11 = fmul float %tmp9, %x		; <float> [#uses=1]
-  %tmp13 = fadd float %tmp11, 0.000000e+00		; <float> [#uses=1]
-  %tmp17.sum52 = or i32 %tmp5, 1		; <i32> [#uses=1]
-  %tmp1851 = getelementptr [128 x float], ptr %lookupTable, i32 0, i32 %tmp17.sum52		; <ptr> [#uses=1]
-  %tmp19 = load float, ptr %tmp1851		; <float> [#uses=1]
-  %tmp21 = fmul float %tmp19, %y		; <float> [#uses=1]
-  %tmp23 = fadd float %tmp21, %tmp13		; <float> [#uses=1]
-  %tmp27.sum50 = or i32 %tmp5, 2		; <i32> [#uses=1]
-  %tmp2849 = getelementptr [128 x float], ptr %lookupTable, i32 0, i32 %tmp27.sum50		; <ptr> [#uses=1]
-  %tmp29 = load float, ptr %tmp2849		; <float> [#uses=1]
-  %tmp31 = fmul float %tmp29, %z		; <float> [#uses=1]
-  %tmp33 = fadd float %tmp31, %tmp23		; <float> [#uses=1]
-  %tmp37.sum48 = or i32 %tmp5, 3		; <i32> [#uses=1]
-  %tmp3847 = getelementptr [128 x float], ptr %lookupTable, i32 0, i32 %tmp37.sum48		; <ptr> [#uses=1]
-  %tmp39 = load float, ptr %tmp3847		; <float> [#uses=1]
-  %tmp41 = fmul float %tmp39, %w		; <float> [#uses=1]
-  %tmp43 = fadd float %tmp41, %tmp33		; <float> [#uses=1]
-  ret float %tmp43
+  %t3 = shl i32 %hash, 2
+  %t5 = and i32 %t3, 124
+  %t753 = getelementptr [128 x float], ptr %lookupTable, i32 0, i32 %t5
+  %t9 = load float, ptr %t753
+  %t11 = fmul float %t9, %x
+  %t13 = fadd float %t11, 0.000000e+00
+  %t17.sum52 = or i32 %t5, 1
+  %t1851 = getelementptr [128 x float], ptr %lookupTable, i32 0, i32 %t17.sum52
+  %t19 = load float, ptr %t1851
+  %t21 = fmul float %t19, %y
+  %t23 = fadd float %t21, %t13
+  %t27.sum50 = or i32 %t5, 2
+  %t2849 = getelementptr [128 x float], ptr %lookupTable, i32 0, i32 %t27.sum50
+  %t29 = load float, ptr %t2849
+  %t31 = fmul float %t29, %z
+  %t33 = fadd float %t31, %t23
+  %t37.sum48 = or i32 %t5, 3
+  %t3847 = getelementptr [128 x float], ptr %lookupTable, i32 0, i32 %t37.sum48
+  %t39 = load float, ptr %t3847
+  %t41 = fmul float %t39, %w
+  %t43 = fadd float %t41, %t33
+  ret float %t43
 }
 
 declare void @llvm.memcpy.p0.p0.i64(ptr nocapture, ptr nocapture, i64, i1) nounwind
diff --git a/llvm/test/Transforms/InstCombine/merging-multiple-stores-into-successor.ll b/llvm/test/Transforms/InstCombine/merging-multiple-stores-into-successor.ll
index 866381ff2887..9c5bf3cb5a41 100644
--- a/llvm/test/Transforms/InstCombine/merging-multiple-stores-into-successor.ll
+++ b/llvm/test/Transforms/InstCombine/merging-multiple-stores-into-successor.ll
@@ -105,11 +105,11 @@ define i32 @diff_types_diff_width_no_merge(i1 %cond, i32 %a, i64 %b) {
 ; CHECK-LABEL: @diff_types_diff_width_no_merge(
 ; CHECK-NEXT:  entry:
 ; CHECK-NEXT:    [[ALLOCA:%.*]] = alloca i64, align 8
-; CHECK-NEXT:    br i1 [[COND:%.*]], label [[A:%.*]], label [[B:%.*]]
-; CHECK:       A:
+; CHECK-NEXT:    br i1 [[COND:%.*]], label [[IF:%.*]], label [[ELSE:%.*]]
+; CHECK:       if:
 ; CHECK-NEXT:    store i32 [[A:%.*]], ptr [[ALLOCA]], align 4
 ; CHECK-NEXT:    br label [[SINK:%.*]]
-; CHECK:       B:
+; CHECK:       else:
 ; CHECK-NEXT:    store i64 [[B:%.*]], ptr [[ALLOCA]], align 4
 ; CHECK-NEXT:    br label [[SINK]]
 ; CHECK:       sink:
@@ -118,11 +118,11 @@ define i32 @diff_types_diff_width_no_merge(i1 %cond, i32 %a, i64 %b) {
 ;
 entry:
   %alloca = alloca i64
-  br i1 %cond, label %A, label %B
-A:
+  br i1 %cond, label %if, label %else
+if:
   store i32 %a, ptr %alloca
   br label %sink
-B:
+  else:
   store i64 %b, ptr %alloca
   br label %sink
 sink:
@@ -134,11 +134,11 @@ define <4 x i32> @vec_no_merge(i1 %cond, <2 x i32> %a, <4 x i32> %b) {
 ; CHECK-LABEL: @vec_no_merge(
 ; CHECK-NEXT:  entry:
 ; CHECK-NEXT:    [[ALLOCA:%.*]] = alloca i64, align 8
-; CHECK-NEXT:    br i1 [[COND:%.*]], label [[A:%.*]], label [[B:%.*]]
-; CHECK:       A:
+; CHECK-NEXT:    br i1 [[COND:%.*]], label [[IF:%.*]], label [[ELSE:%.*]]
+; CHECK:       if:
 ; CHECK-NEXT:    store <2 x i32> [[A:%.*]], ptr [[ALLOCA]], align 8
 ; CHECK-NEXT:    br label [[SINK:%.*]]
-; CHECK:       B:
+; CHECK:       else:
 ; CHECK-NEXT:    store <4 x i32> [[B:%.*]], ptr [[ALLOCA]], align 16
 ; CHECK-NEXT:    br label [[SINK]]
 ; CHECK:       sink:
@@ -147,11 +147,11 @@ define <4 x i32> @vec_no_merge(i1 %cond, <2 x i32> %a, <4 x i32> %b) {
 ;
 entry:
   %alloca = alloca i64
-  br i1 %cond, label %A, label %B
-A:
+  br i1 %cond, label %if, label %else
+if:
   store <2 x i32> %a, ptr %alloca
   br label %sink
-B:
+else:
   store <4 x i32> %b, ptr %alloca
   br label %sink
 sink:
@@ -195,11 +195,11 @@ define %struct.tup @multi_elem_struct_no_merge(i1 %cond, %struct.tup %a, half %b
 ; CHECK-LABEL: @multi_elem_struct_no_merge(
 ; CHECK-NEXT:  entry:
 ; CHECK-NEXT:    [[ALLOCA:%.*]] = alloca i64, align 8
-; CHECK-NEXT:    br i1 [[COND:%.*]], label [[A:%.*]], label [[B:%.*]]
-; CHECK:       A:
+; CHECK-NEXT:    br i1 [[COND:%.*]], label [[IF:%.*]], label [[ELSE:%.*]]
+; CHECK:       if:
 ; CHECK-NEXT:    store [[STRUCT_TUP:%.*]] [[A:%.*]], ptr [[ALLOCA]], align 4
 ; CHECK-NEXT:    br label [[SINK:%.*]]
-; CHECK:       B:
+; CHECK:       else:
 ; CHECK-NEXT:    store half [[B:%.*]], ptr [[ALLOCA]], align 2
 ; CHECK-NEXT:    br label [[SINK]]
 ; CHECK:       sink:
@@ -208,11 +208,11 @@ define %struct.tup @multi_elem_struct_no_merge(i1 %cond, %struct.tup %a, half %b
 ;
 entry:
   %alloca = alloca i64
-  br i1 %cond, label %A, label %B
-A:
+  br i1 %cond, label %if, label %else
+if:
   store %struct.tup %a, ptr %alloca
   br label %sink
-B:
+else:
   store half %b, ptr %alloca
   br label %sink
 sink:
diff --git a/llvm/test/Transforms/InstCombine/minmax-fold.ll b/llvm/test/Transforms/InstCombine/minmax-fold.ll
index 8b47dc7a2807..3e870c695cf1 100644
--- a/llvm/test/Transforms/InstCombine/minmax-fold.ll
+++ b/llvm/test/Transforms/InstCombine/minmax-fold.ll
@@ -1524,7 +1524,7 @@ define i32 @test_smin_umin4(i32 %x) {
 
 define i32 @test_umax_nonminmax(i32 %x) {
 ; CHECK-LABEL: @test_umax_nonminmax(
-; CHECK-NEXT:    [[Y:%.*]] = call i32 @llvm.ctpop.i32(i32 [[X:%.*]]), !range [[RNG0:![0-9]+]]
+; CHECK-NEXT:    [[Y:%.*]] = call range(i32 0, 33) i32 @llvm.ctpop.i32(i32 [[X:%.*]])
 ; CHECK-NEXT:    [[UMAX:%.*]] = call i32 @llvm.umax.i32(i32 [[Y]], i32 1)
 ; CHECK-NEXT:    ret i32 [[UMAX]]
 ;
diff --git a/llvm/test/Transforms/InstCombine/mul.ll b/llvm/test/Transforms/InstCombine/mul.ll
index 227ca4a6d5cf..4fb3c0b1ad49 100644
--- a/llvm/test/Transforms/InstCombine/mul.ll
+++ b/llvm/test/Transforms/InstCombine/mul.ll
@@ -2061,8 +2061,8 @@ define i32 @mul_sext_icmp_with_zero(i32 %x) {
 
 define i32 @test_mul_sext_bool(i1 %x, i32 %y) {
 ; CHECK-LABEL: @test_mul_sext_bool(
-; CHECK-NEXT:    [[Y_NEG:%.*]] = sub i32 0, [[Y:%.*]]
-; CHECK-NEXT:    [[MUL:%.*]] = select i1 [[X:%.*]], i32 [[Y_NEG]], i32 0
+; CHECK-NEXT:    [[TMP1:%.*]] = sub i32 0, [[Y:%.*]]
+; CHECK-NEXT:    [[MUL:%.*]] = select i1 [[X:%.*]], i32 [[TMP1]], i32 0
 ; CHECK-NEXT:    ret i32 [[MUL]]
 ;
   %sext = sext i1 %x to i32
@@ -2072,8 +2072,8 @@ define i32 @test_mul_sext_bool(i1 %x, i32 %y) {
 
 define i32 @test_mul_sext_bool_nuw(i1 %x, i32 %y) {
 ; CHECK-LABEL: @test_mul_sext_bool_nuw(
-; CHECK-NEXT:    [[Y_NEG:%.*]] = sub i32 0, [[Y:%.*]]
-; CHECK-NEXT:    [[MUL:%.*]] = select i1 [[X:%.*]], i32 [[Y_NEG]], i32 0
+; CHECK-NEXT:    [[TMP1:%.*]] = sub i32 0, [[Y:%.*]]
+; CHECK-NEXT:    [[MUL:%.*]] = select i1 [[X:%.*]], i32 [[TMP1]], i32 0
 ; CHECK-NEXT:    ret i32 [[MUL]]
 ;
   %sext = sext i1 %x to i32
@@ -2083,8 +2083,8 @@ define i32 @test_mul_sext_bool_nuw(i1 %x, i32 %y) {
 
 define i32 @test_mul_sext_bool_nsw(i1 %x, i32 %y) {
 ; CHECK-LABEL: @test_mul_sext_bool_nsw(
-; CHECK-NEXT:    [[Y_NEG:%.*]] = sub nsw i32 0, [[Y:%.*]]
-; CHECK-NEXT:    [[MUL:%.*]] = select i1 [[X:%.*]], i32 [[Y_NEG]], i32 0
+; CHECK-NEXT:    [[TMP1:%.*]] = sub nsw i32 0, [[Y:%.*]]
+; CHECK-NEXT:    [[MUL:%.*]] = select i1 [[X:%.*]], i32 [[TMP1]], i32 0
 ; CHECK-NEXT:    ret i32 [[MUL]]
 ;
   %sext = sext i1 %x to i32
@@ -2094,8 +2094,8 @@ define i32 @test_mul_sext_bool_nsw(i1 %x, i32 %y) {
 
 define i32 @test_mul_sext_bool_nuw_nsw(i1 %x, i32 %y) {
 ; CHECK-LABEL: @test_mul_sext_bool_nuw_nsw(
-; CHECK-NEXT:    [[Y_NEG:%.*]] = sub nsw i32 0, [[Y:%.*]]
-; CHECK-NEXT:    [[MUL:%.*]] = select i1 [[X:%.*]], i32 [[Y_NEG]], i32 0
+; CHECK-NEXT:    [[TMP1:%.*]] = sub nsw i32 0, [[Y:%.*]]
+; CHECK-NEXT:    [[MUL:%.*]] = select i1 [[X:%.*]], i32 [[TMP1]], i32 0
 ; CHECK-NEXT:    ret i32 [[MUL]]
 ;
   %sext = sext i1 %x to i32
@@ -2106,8 +2106,8 @@ define i32 @test_mul_sext_bool_nuw_nsw(i1 %x, i32 %y) {
 define i32 @test_mul_sext_bool_commuted(i1 %x, i32 %y) {
 ; CHECK-LABEL: @test_mul_sext_bool_commuted(
 ; CHECK-NEXT:    [[TMP1:%.*]] = xor i32 [[Y:%.*]], -2
-; CHECK-NEXT:    [[YY_NEG1:%.*]] = add i32 [[TMP1]], 1
-; CHECK-NEXT:    [[MUL:%.*]] = select i1 [[X:%.*]], i32 [[YY_NEG1]], i32 0
+; CHECK-NEXT:    [[YY_NEG:%.*]] = add i32 [[TMP1]], 1
+; CHECK-NEXT:    [[MUL:%.*]] = select i1 [[X:%.*]], i32 [[YY_NEG]], i32 0
 ; CHECK-NEXT:    ret i32 [[MUL]]
 ;
   %yy = xor i32 %y, 1
@@ -2139,3 +2139,63 @@ define i32 @test_mul_sext_multiuse(i1 %x, i32 %y) {
   %mul = mul i32 %sext, %y
   ret i32 %mul
 }
+
+define i8 @mul_nsw_nonneg(i8 %x, i8 %y) {
+; CHECK-LABEL: @mul_nsw_nonneg(
+; CHECK-NEXT:    [[X_NNEG:%.*]] = icmp sgt i8 [[X:%.*]], -1
+; CHECK-NEXT:    call void @llvm.assume(i1 [[X_NNEG]])
+; CHECK-NEXT:    [[Y_NNEG:%.*]] = icmp sgt i8 [[Y:%.*]], -1
+; CHECK-NEXT:    call void @llvm.assume(i1 [[Y_NNEG]])
+; CHECK-NEXT:    [[MUL:%.*]] = mul nuw nsw i8 [[X]], [[Y]]
+; CHECK-NEXT:    ret i8 [[MUL]]
+;
+  %x.nneg = icmp sge i8 %x, 0
+  call void @llvm.assume(i1 %x.nneg)
+  %y.nneg = icmp sge i8 %y, 0
+  call void @llvm.assume(i1 %y.nneg)
+  %mul = mul nsw i8 %x, %y
+  ret i8 %mul
+}
+
+define i8 @mul_nsw_not_nonneg1(i8 %x, i8 %y) {
+; CHECK-LABEL: @mul_nsw_not_nonneg1(
+; CHECK-NEXT:    [[Y_NNEG:%.*]] = icmp sgt i8 [[Y:%.*]], -1
+; CHECK-NEXT:    call void @llvm.assume(i1 [[Y_NNEG]])
+; CHECK-NEXT:    [[MUL:%.*]] = mul nsw i8 [[X:%.*]], [[Y]]
+; CHECK-NEXT:    ret i8 [[MUL]]
+;
+  %y.nneg = icmp sge i8 %y, 0
+  call void @llvm.assume(i1 %y.nneg)
+  %mul = mul nsw i8 %x, %y
+  ret i8 %mul
+}
+
+define i8 @mul_nsw_not_nonneg2(i8 %x, i8 %y) {
+; CHECK-LABEL: @mul_nsw_not_nonneg2(
+; CHECK-NEXT:    [[X_NNEG:%.*]] = icmp sgt i8 [[X:%.*]], -1
+; CHECK-NEXT:    call void @llvm.assume(i1 [[X_NNEG]])
+; CHECK-NEXT:    [[MUL:%.*]] = mul nsw i8 [[X]], [[Y:%.*]]
+; CHECK-NEXT:    ret i8 [[MUL]]
+;
+  %x.nneg = icmp sge i8 %x, 0
+  call void @llvm.assume(i1 %x.nneg)
+  %mul = mul nsw i8 %x, %y
+  ret i8 %mul
+}
+
+define i8 @mul_not_nsw_nonneg(i8 %x, i8 %y) {
+; CHECK-LABEL: @mul_not_nsw_nonneg(
+; CHECK-NEXT:    [[X_NNEG:%.*]] = icmp sgt i8 [[X:%.*]], -1
+; CHECK-NEXT:    call void @llvm.assume(i1 [[X_NNEG]])
+; CHECK-NEXT:    [[Y_NNEG:%.*]] = icmp sgt i8 [[Y:%.*]], -1
+; CHECK-NEXT:    call void @llvm.assume(i1 [[Y_NNEG]])
+; CHECK-NEXT:    [[MUL:%.*]] = mul i8 [[X]], [[Y]]
+; CHECK-NEXT:    ret i8 [[MUL]]
+;
+  %x.nneg = icmp sge i8 %x, 0
+  call void @llvm.assume(i1 %x.nneg)
+  %y.nneg = icmp sge i8 %y, 0
+  call void @llvm.assume(i1 %y.nneg)
+  %mul = mul i8 %x, %y
+  ret i8 %mul
+}
diff --git a/llvm/test/Transforms/InstCombine/reduction-add-sext-zext-i1.ll b/llvm/test/Transforms/InstCombine/reduction-add-sext-zext-i1.ll
index b94be990199b..ca586a71b42c 100644
--- a/llvm/test/Transforms/InstCombine/reduction-add-sext-zext-i1.ll
+++ b/llvm/test/Transforms/InstCombine/reduction-add-sext-zext-i1.ll
@@ -4,7 +4,7 @@
 define i1 @reduce_add_self(<8 x i1> %x) {
 ; CHECK-LABEL: @reduce_add_self(
 ; CHECK-NEXT:    [[TMP1:%.*]] = bitcast <8 x i1> [[X:%.*]] to i8
-; CHECK-NEXT:    [[TMP2:%.*]] = call i8 @llvm.ctpop.i8(i8 [[TMP1]]), !range [[RNG0:![0-9]+]]
+; CHECK-NEXT:    [[TMP2:%.*]] = call range(i8 0, 9) i8 @llvm.ctpop.i8(i8 [[TMP1]])
 ; CHECK-NEXT:    [[RES:%.*]] = trunc i8 [[TMP2]] to i1
 ; CHECK-NEXT:    ret i1 [[RES]]
 ;
@@ -15,7 +15,7 @@ define i1 @reduce_add_self(<8 x i1> %x) {
 define i32 @reduce_add_sext(<4 x i1> %x) {
 ; CHECK-LABEL: @reduce_add_sext(
 ; CHECK-NEXT:    [[TMP1:%.*]] = bitcast <4 x i1> [[X:%.*]] to i4
-; CHECK-NEXT:    [[TMP2:%.*]] = call i4 @llvm.ctpop.i4(i4 [[TMP1]]), !range [[RNG1:![0-9]+]]
+; CHECK-NEXT:    [[TMP2:%.*]] = call range(i4 0, 5) i4 @llvm.ctpop.i4(i4 [[TMP1]])
 ; CHECK-NEXT:    [[TMP3:%.*]] = zext nneg i4 [[TMP2]] to i32
 ; CHECK-NEXT:    [[RES:%.*]] = sub nsw i32 0, [[TMP3]]
 ; CHECK-NEXT:    ret i32 [[RES]]
@@ -28,7 +28,7 @@ define i32 @reduce_add_sext(<4 x i1> %x) {
 define i64 @reduce_add_zext(<8 x i1> %x) {
 ; CHECK-LABEL: @reduce_add_zext(
 ; CHECK-NEXT:    [[TMP1:%.*]] = bitcast <8 x i1> [[X:%.*]] to i8
-; CHECK-NEXT:    [[TMP2:%.*]] = call i8 @llvm.ctpop.i8(i8 [[TMP1]]), !range [[RNG0]]
+; CHECK-NEXT:    [[TMP2:%.*]] = call range(i8 0, 9) i8 @llvm.ctpop.i8(i8 [[TMP1]])
 ; CHECK-NEXT:    [[RES:%.*]] = zext nneg i8 [[TMP2]] to i64
 ; CHECK-NEXT:    ret i64 [[RES]]
 ;
@@ -40,7 +40,7 @@ define i64 @reduce_add_zext(<8 x i1> %x) {
 define i16 @reduce_add_sext_same(<16 x i1> %x) {
 ; CHECK-LABEL: @reduce_add_sext_same(
 ; CHECK-NEXT:    [[TMP1:%.*]] = bitcast <16 x i1> [[X:%.*]] to i16
-; CHECK-NEXT:    [[TMP2:%.*]] = call i16 @llvm.ctpop.i16(i16 [[TMP1]]), !range [[RNG2:![0-9]+]]
+; CHECK-NEXT:    [[TMP2:%.*]] = call range(i16 0, 17) i16 @llvm.ctpop.i16(i16 [[TMP1]])
 ; CHECK-NEXT:    [[RES:%.*]] = sub nsw i16 0, [[TMP2]]
 ; CHECK-NEXT:    ret i16 [[RES]]
 ;
@@ -52,7 +52,7 @@ define i16 @reduce_add_sext_same(<16 x i1> %x) {
 define i8 @reduce_add_zext_long(<128 x i1> %x) {
 ; CHECK-LABEL: @reduce_add_zext_long(
 ; CHECK-NEXT:    [[TMP1:%.*]] = bitcast <128 x i1> [[X:%.*]] to i128
-; CHECK-NEXT:    [[TMP2:%.*]] = call i128 @llvm.ctpop.i128(i128 [[TMP1]]), !range [[RNG3:![0-9]+]]
+; CHECK-NEXT:    [[TMP2:%.*]] = call range(i128 0, 129) i128 @llvm.ctpop.i128(i128 [[TMP1]])
 ; CHECK-NEXT:    [[TMP3:%.*]] = trunc nuw i128 [[TMP2]] to i8
 ; CHECK-NEXT:    [[RES:%.*]] = sub i8 0, [[TMP3]]
 ; CHECK-NEXT:    ret i8 [[RES]]
@@ -66,7 +66,7 @@ define i8 @reduce_add_zext_long(<128 x i1> %x) {
 define i8 @reduce_add_zext_long_external_use(<128 x i1> %x) {
 ; CHECK-LABEL: @reduce_add_zext_long_external_use(
 ; CHECK-NEXT:    [[TMP1:%.*]] = bitcast <128 x i1> [[X:%.*]] to i128
-; CHECK-NEXT:    [[TMP2:%.*]] = call i128 @llvm.ctpop.i128(i128 [[TMP1]]), !range [[RNG3]]
+; CHECK-NEXT:    [[TMP2:%.*]] = call range(i128 0, 129) i128 @llvm.ctpop.i128(i128 [[TMP1]])
 ; CHECK-NEXT:    [[TMP3:%.*]] = trunc nuw i128 [[TMP2]] to i8
 ; CHECK-NEXT:    [[RES:%.*]] = sub i8 0, [[TMP3]]
 ; CHECK-NEXT:    [[TMP4:%.*]] = extractelement <128 x i1> [[X]], i64 0
@@ -85,7 +85,7 @@ define i8 @reduce_add_zext_long_external_use(<128 x i1> %x) {
 define i64 @reduce_add_zext_external_use(<8 x i1> %x) {
 ; CHECK-LABEL: @reduce_add_zext_external_use(
 ; CHECK-NEXT:    [[TMP1:%.*]] = bitcast <8 x i1> [[X:%.*]] to i8
-; CHECK-NEXT:    [[TMP2:%.*]] = call i8 @llvm.ctpop.i8(i8 [[TMP1]]), !range [[RNG0]]
+; CHECK-NEXT:    [[TMP2:%.*]] = call range(i8 0, 9) i8 @llvm.ctpop.i8(i8 [[TMP1]])
 ; CHECK-NEXT:    [[RES:%.*]] = zext nneg i8 [[TMP2]] to i64
 ; CHECK-NEXT:    [[TMP3:%.*]] = extractelement <8 x i1> [[X]], i64 0
 ; CHECK-NEXT:    [[EXT:%.*]] = zext i1 [[TMP3]] to i64
diff --git a/llvm/test/Transforms/InstCombine/reduction-xor-sext-zext-i1.ll b/llvm/test/Transforms/InstCombine/reduction-xor-sext-zext-i1.ll
index 84ac9369b5ff..b5baf8ec71a4 100644
--- a/llvm/test/Transforms/InstCombine/reduction-xor-sext-zext-i1.ll
+++ b/llvm/test/Transforms/InstCombine/reduction-xor-sext-zext-i1.ll
@@ -4,7 +4,7 @@
 define i1 @reduce_xor_self(<8 x i1> %x) {
 ; CHECK-LABEL: @reduce_xor_self(
 ; CHECK-NEXT:    [[TMP1:%.*]] = bitcast <8 x i1> [[X:%.*]] to i8
-; CHECK-NEXT:    [[TMP2:%.*]] = call i8 @llvm.ctpop.i8(i8 [[TMP1]]), !range [[RNG0:![0-9]+]]
+; CHECK-NEXT:    [[TMP2:%.*]] = call range(i8 0, 9) i8 @llvm.ctpop.i8(i8 [[TMP1]])
 ; CHECK-NEXT:    [[RES:%.*]] = trunc i8 [[TMP2]] to i1
 ; CHECK-NEXT:    ret i1 [[RES]]
 ;
@@ -15,7 +15,7 @@ define i1 @reduce_xor_self(<8 x i1> %x) {
 define i32 @reduce_xor_sext(<4 x i1> %x) {
 ; CHECK-LABEL: @reduce_xor_sext(
 ; CHECK-NEXT:    [[TMP1:%.*]] = bitcast <4 x i1> [[X:%.*]] to i4
-; CHECK-NEXT:    [[TMP2:%.*]] = call i4 @llvm.ctpop.i4(i4 [[TMP1]]), !range [[RNG1:![0-9]+]]
+; CHECK-NEXT:    [[TMP2:%.*]] = call range(i4 0, 5) i4 @llvm.ctpop.i4(i4 [[TMP1]])
 ; CHECK-NEXT:    [[TMP3:%.*]] = trunc i4 [[TMP2]] to i1
 ; CHECK-NEXT:    [[RES:%.*]] = sext i1 [[TMP3]] to i32
 ; CHECK-NEXT:    ret i32 [[RES]]
@@ -28,7 +28,7 @@ define i32 @reduce_xor_sext(<4 x i1> %x) {
 define i64 @reduce_xor_zext(<8 x i1> %x) {
 ; CHECK-LABEL: @reduce_xor_zext(
 ; CHECK-NEXT:    [[TMP1:%.*]] = bitcast <8 x i1> [[X:%.*]] to i8
-; CHECK-NEXT:    [[TMP2:%.*]] = call i8 @llvm.ctpop.i8(i8 [[TMP1]]), !range [[RNG0]]
+; CHECK-NEXT:    [[TMP2:%.*]] = call range(i8 0, 9) i8 @llvm.ctpop.i8(i8 [[TMP1]])
 ; CHECK-NEXT:    [[TMP3:%.*]] = and i8 [[TMP2]], 1
 ; CHECK-NEXT:    [[RES:%.*]] = zext nneg i8 [[TMP3]] to i64
 ; CHECK-NEXT:    ret i64 [[RES]]
@@ -41,7 +41,7 @@ define i64 @reduce_xor_zext(<8 x i1> %x) {
 define i16 @reduce_xor_sext_same(<16 x i1> %x) {
 ; CHECK-LABEL: @reduce_xor_sext_same(
 ; CHECK-NEXT:    [[TMP1:%.*]] = bitcast <16 x i1> [[X:%.*]] to i16
-; CHECK-NEXT:    [[TMP2:%.*]] = call i16 @llvm.ctpop.i16(i16 [[TMP1]]), !range [[RNG2:![0-9]+]]
+; CHECK-NEXT:    [[TMP2:%.*]] = call range(i16 0, 17) i16 @llvm.ctpop.i16(i16 [[TMP1]])
 ; CHECK-NEXT:    [[TMP3:%.*]] = and i16 [[TMP2]], 1
 ; CHECK-NEXT:    [[SEXT:%.*]] = sub nsw i16 0, [[TMP3]]
 ; CHECK-NEXT:    ret i16 [[SEXT]]
@@ -54,7 +54,7 @@ define i16 @reduce_xor_sext_same(<16 x i1> %x) {
 define i8 @reduce_xor_zext_long(<128 x i1> %x) {
 ; CHECK-LABEL: @reduce_xor_zext_long(
 ; CHECK-NEXT:    [[TMP1:%.*]] = bitcast <128 x i1> [[X:%.*]] to i128
-; CHECK-NEXT:    [[TMP2:%.*]] = call i128 @llvm.ctpop.i128(i128 [[TMP1]]), !range [[RNG3:![0-9]+]]
+; CHECK-NEXT:    [[TMP2:%.*]] = call range(i128 0, 129) i128 @llvm.ctpop.i128(i128 [[TMP1]])
 ; CHECK-NEXT:    [[TMP3:%.*]] = trunc i128 [[TMP2]] to i1
 ; CHECK-NEXT:    [[RES:%.*]] = sext i1 [[TMP3]] to i8
 ; CHECK-NEXT:    ret i8 [[RES]]
@@ -68,7 +68,7 @@ define i8 @reduce_xor_zext_long(<128 x i1> %x) {
 define i8 @reduce_xor_zext_long_external_use(<128 x i1> %x) {
 ; CHECK-LABEL: @reduce_xor_zext_long_external_use(
 ; CHECK-NEXT:    [[TMP1:%.*]] = bitcast <128 x i1> [[X:%.*]] to i128
-; CHECK-NEXT:    [[TMP2:%.*]] = call i128 @llvm.ctpop.i128(i128 [[TMP1]]), !range [[RNG3]]
+; CHECK-NEXT:    [[TMP2:%.*]] = call range(i128 0, 129) i128 @llvm.ctpop.i128(i128 [[TMP1]])
 ; CHECK-NEXT:    [[TMP3:%.*]] = trunc i128 [[TMP2]] to i1
 ; CHECK-NEXT:    [[RES:%.*]] = sext i1 [[TMP3]] to i8
 ; CHECK-NEXT:    [[TMP5:%.*]] = extractelement <128 x i1> [[X]], i64 0
@@ -87,7 +87,7 @@ define i8 @reduce_xor_zext_long_external_use(<128 x i1> %x) {
 define i64 @reduce_xor_zext_external_use(<8 x i1> %x) {
 ; CHECK-LABEL: @reduce_xor_zext_external_use(
 ; CHECK-NEXT:    [[TMP1:%.*]] = bitcast <8 x i1> [[X:%.*]] to i8
-; CHECK-NEXT:    [[TMP2:%.*]] = call i8 @llvm.ctpop.i8(i8 [[TMP1]]), !range [[RNG0]]
+; CHECK-NEXT:    [[TMP2:%.*]] = call range(i8 0, 9) i8 @llvm.ctpop.i8(i8 [[TMP1]])
 ; CHECK-NEXT:    [[TMP3:%.*]] = and i8 [[TMP2]], 1
 ; CHECK-NEXT:    [[RES:%.*]] = zext nneg i8 [[TMP3]] to i64
 ; CHECK-NEXT:    [[TMP4:%.*]] = extractelement <8 x i1> [[X]], i64 0
diff --git a/llvm/test/Transforms/InstCombine/select-cmp-cttz-ctlz.ll b/llvm/test/Transforms/InstCombine/select-cmp-cttz-ctlz.ll
index 69896f855f5f..3d5f2c209edf 100644
--- a/llvm/test/Transforms/InstCombine/select-cmp-cttz-ctlz.ll
+++ b/llvm/test/Transforms/InstCombine/select-cmp-cttz-ctlz.ll
@@ -7,7 +7,7 @@
 
 define i16 @test1(i16 %x) {
 ; CHECK-LABEL: @test1(
-; CHECK-NEXT:    [[CT:%.*]] = tail call i16 @llvm.ctlz.i16(i16 [[X:%.*]], i1 false), !range [[RNG0:![0-9]+]]
+; CHECK-NEXT:    [[CT:%.*]] = tail call range(i16 0, 17) i16 @llvm.ctlz.i16(i16 [[X:%.*]], i1 false)
 ; CHECK-NEXT:    ret i16 [[CT]]
 ;
   %ct = tail call i16 @llvm.ctlz.i16(i16 %x, i1 true)
@@ -18,7 +18,7 @@ define i16 @test1(i16 %x) {
 
 define i32 @test2(i32 %x) {
 ; CHECK-LABEL: @test2(
-; CHECK-NEXT:    [[CT:%.*]] = tail call i32 @llvm.ctlz.i32(i32 [[X:%.*]], i1 false), !range [[RNG1:![0-9]+]]
+; CHECK-NEXT:    [[CT:%.*]] = tail call range(i32 0, 33) i32 @llvm.ctlz.i32(i32 [[X:%.*]], i1 false)
 ; CHECK-NEXT:    ret i32 [[CT]]
 ;
   %ct = tail call i32 @llvm.ctlz.i32(i32 %x, i1 true)
@@ -29,7 +29,7 @@ define i32 @test2(i32 %x) {
 
 define i64 @test3(i64 %x) {
 ; CHECK-LABEL: @test3(
-; CHECK-NEXT:    [[CT:%.*]] = tail call i64 @llvm.ctlz.i64(i64 [[X:%.*]], i1 false), !range [[RNG2:![0-9]+]]
+; CHECK-NEXT:    [[CT:%.*]] = tail call range(i64 0, 65) i64 @llvm.ctlz.i64(i64 [[X:%.*]], i1 false)
 ; CHECK-NEXT:    ret i64 [[CT]]
 ;
   %ct = tail call i64 @llvm.ctlz.i64(i64 %x, i1 true)
@@ -40,7 +40,7 @@ define i64 @test3(i64 %x) {
 
 define i16 @test4(i16 %x) {
 ; CHECK-LABEL: @test4(
-; CHECK-NEXT:    [[CT:%.*]] = tail call i16 @llvm.ctlz.i16(i16 [[X:%.*]], i1 false), !range [[RNG0]]
+; CHECK-NEXT:    [[CT:%.*]] = tail call range(i16 0, 17) i16 @llvm.ctlz.i16(i16 [[X:%.*]], i1 false)
 ; CHECK-NEXT:    ret i16 [[CT]]
 ;
   %ct = tail call i16 @llvm.ctlz.i16(i16 %x, i1 true)
@@ -51,7 +51,7 @@ define i16 @test4(i16 %x) {
 
 define i32 @test5(i32 %x) {
 ; CHECK-LABEL: @test5(
-; CHECK-NEXT:    [[CT:%.*]] = tail call i32 @llvm.ctlz.i32(i32 [[X:%.*]], i1 false), !range [[RNG1]]
+; CHECK-NEXT:    [[CT:%.*]] = tail call range(i32 0, 33) i32 @llvm.ctlz.i32(i32 [[X:%.*]], i1 false)
 ; CHECK-NEXT:    ret i32 [[CT]]
 ;
   %ct = tail call i32 @llvm.ctlz.i32(i32 %x, i1 true)
@@ -62,7 +62,7 @@ define i32 @test5(i32 %x) {
 
 define i64 @test6(i64 %x) {
 ; CHECK-LABEL: @test6(
-; CHECK-NEXT:    [[CT:%.*]] = tail call i64 @llvm.ctlz.i64(i64 [[X:%.*]], i1 false), !range [[RNG2]]
+; CHECK-NEXT:    [[CT:%.*]] = tail call range(i64 0, 65) i64 @llvm.ctlz.i64(i64 [[X:%.*]], i1 false)
 ; CHECK-NEXT:    ret i64 [[CT]]
 ;
   %ct = tail call i64 @llvm.ctlz.i64(i64 %x, i1 true)
@@ -73,7 +73,7 @@ define i64 @test6(i64 %x) {
 
 define i16 @test1b(i16 %x) {
 ; CHECK-LABEL: @test1b(
-; CHECK-NEXT:    [[CT:%.*]] = tail call i16 @llvm.cttz.i16(i16 [[X:%.*]], i1 false), !range [[RNG0]]
+; CHECK-NEXT:    [[CT:%.*]] = tail call range(i16 0, 17) i16 @llvm.cttz.i16(i16 [[X:%.*]], i1 false)
 ; CHECK-NEXT:    ret i16 [[CT]]
 ;
   %ct = tail call i16 @llvm.cttz.i16(i16 %x, i1 true)
@@ -84,7 +84,7 @@ define i16 @test1b(i16 %x) {
 
 define i32 @test2b(i32 %x) {
 ; CHECK-LABEL: @test2b(
-; CHECK-NEXT:    [[CT:%.*]] = tail call i32 @llvm.cttz.i32(i32 [[X:%.*]], i1 false), !range [[RNG1]]
+; CHECK-NEXT:    [[CT:%.*]] = tail call range(i32 0, 33) i32 @llvm.cttz.i32(i32 [[X:%.*]], i1 false)
 ; CHECK-NEXT:    ret i32 [[CT]]
 ;
   %ct = tail call i32 @llvm.cttz.i32(i32 %x, i1 true)
@@ -95,7 +95,7 @@ define i32 @test2b(i32 %x) {
 
 define i64 @test3b(i64 %x) {
 ; CHECK-LABEL: @test3b(
-; CHECK-NEXT:    [[CT:%.*]] = tail call i64 @llvm.cttz.i64(i64 [[X:%.*]], i1 false), !range [[RNG2]]
+; CHECK-NEXT:    [[CT:%.*]] = tail call range(i64 0, 65) i64 @llvm.cttz.i64(i64 [[X:%.*]], i1 false)
 ; CHECK-NEXT:    ret i64 [[CT]]
 ;
   %ct = tail call i64 @llvm.cttz.i64(i64 %x, i1 true)
@@ -106,7 +106,7 @@ define i64 @test3b(i64 %x) {
 
 define i16 @test4b(i16 %x) {
 ; CHECK-LABEL: @test4b(
-; CHECK-NEXT:    [[CT:%.*]] = tail call i16 @llvm.cttz.i16(i16 [[X:%.*]], i1 false), !range [[RNG0]]
+; CHECK-NEXT:    [[CT:%.*]] = tail call range(i16 0, 17) i16 @llvm.cttz.i16(i16 [[X:%.*]], i1 false)
 ; CHECK-NEXT:    ret i16 [[CT]]
 ;
   %ct = tail call i16 @llvm.cttz.i16(i16 %x, i1 true)
@@ -118,7 +118,7 @@ define i16 @test4b(i16 %x) {
 define i32 @test5b(i32 %x) {
 ; CHECK-LABEL: @test5b(
 ; CHECK-NEXT:  entry:
-; CHECK-NEXT:    [[CT:%.*]] = tail call i32 @llvm.cttz.i32(i32 [[X:%.*]], i1 false), !range [[RNG1]]
+; CHECK-NEXT:    [[CT:%.*]] = tail call range(i32 0, 33) i32 @llvm.cttz.i32(i32 [[X:%.*]], i1 false)
 ; CHECK-NEXT:    ret i32 [[CT]]
 ;
 entry:
@@ -130,7 +130,7 @@ entry:
 
 define i64 @test6b(i64 %x) {
 ; CHECK-LABEL: @test6b(
-; CHECK-NEXT:    [[CT:%.*]] = tail call i64 @llvm.cttz.i64(i64 [[X:%.*]], i1 false), !range [[RNG2]]
+; CHECK-NEXT:    [[CT:%.*]] = tail call range(i64 0, 65) i64 @llvm.cttz.i64(i64 [[X:%.*]], i1 false)
 ; CHECK-NEXT:    ret i64 [[CT]]
 ;
   %ct = tail call i64 @llvm.cttz.i64(i64 %x, i1 true)
@@ -141,7 +141,7 @@ define i64 @test6b(i64 %x) {
 
 define i32 @test1c(i16 %x) {
 ; CHECK-LABEL: @test1c(
-; CHECK-NEXT:    [[CT:%.*]] = tail call i16 @llvm.cttz.i16(i16 [[X:%.*]], i1 false), !range [[RNG0]]
+; CHECK-NEXT:    [[CT:%.*]] = tail call range(i16 0, 17) i16 @llvm.cttz.i16(i16 [[X:%.*]], i1 false)
 ; CHECK-NEXT:    [[CAST2:%.*]] = zext nneg i16 [[CT]] to i32
 ; CHECK-NEXT:    ret i32 [[CAST2]]
 ;
@@ -154,7 +154,7 @@ define i32 @test1c(i16 %x) {
 
 define i64 @test2c(i16 %x) {
 ; CHECK-LABEL: @test2c(
-; CHECK-NEXT:    [[CT:%.*]] = tail call i16 @llvm.cttz.i16(i16 [[X:%.*]], i1 false), !range [[RNG0]]
+; CHECK-NEXT:    [[CT:%.*]] = tail call range(i16 0, 17) i16 @llvm.cttz.i16(i16 [[X:%.*]], i1 false)
 ; CHECK-NEXT:    [[CONV:%.*]] = zext nneg i16 [[CT]] to i64
 ; CHECK-NEXT:    ret i64 [[CONV]]
 ;
@@ -167,7 +167,7 @@ define i64 @test2c(i16 %x) {
 
 define i64 @test3c(i32 %x) {
 ; CHECK-LABEL: @test3c(
-; CHECK-NEXT:    [[CT:%.*]] = tail call i32 @llvm.cttz.i32(i32 [[X:%.*]], i1 false), !range [[RNG1]]
+; CHECK-NEXT:    [[CT:%.*]] = tail call range(i32 0, 33) i32 @llvm.cttz.i32(i32 [[X:%.*]], i1 false)
 ; CHECK-NEXT:    [[CONV:%.*]] = zext nneg i32 [[CT]] to i64
 ; CHECK-NEXT:    ret i64 [[CONV]]
 ;
@@ -180,7 +180,7 @@ define i64 @test3c(i32 %x) {
 
 define i32 @test4c(i16 %x) {
 ; CHECK-LABEL: @test4c(
-; CHECK-NEXT:    [[CT:%.*]] = tail call i16 @llvm.ctlz.i16(i16 [[X:%.*]], i1 false), !range [[RNG0]]
+; CHECK-NEXT:    [[CT:%.*]] = tail call range(i16 0, 17) i16 @llvm.ctlz.i16(i16 [[X:%.*]], i1 false)
 ; CHECK-NEXT:    [[CAST:%.*]] = zext nneg i16 [[CT]] to i32
 ; CHECK-NEXT:    ret i32 [[CAST]]
 ;
@@ -193,7 +193,7 @@ define i32 @test4c(i16 %x) {
 
 define i64 @test5c(i16 %x) {
 ; CHECK-LABEL: @test5c(
-; CHECK-NEXT:    [[CT:%.*]] = tail call i16 @llvm.ctlz.i16(i16 [[X:%.*]], i1 false), !range [[RNG0]]
+; CHECK-NEXT:    [[CT:%.*]] = tail call range(i16 0, 17) i16 @llvm.ctlz.i16(i16 [[X:%.*]], i1 false)
 ; CHECK-NEXT:    [[CAST:%.*]] = zext nneg i16 [[CT]] to i64
 ; CHECK-NEXT:    ret i64 [[CAST]]
 ;
@@ -206,7 +206,7 @@ define i64 @test5c(i16 %x) {
 
 define i64 @test6c(i32 %x) {
 ; CHECK-LABEL: @test6c(
-; CHECK-NEXT:    [[CT:%.*]] = tail call i32 @llvm.ctlz.i32(i32 [[X:%.*]], i1 false), !range [[RNG1]]
+; CHECK-NEXT:    [[CT:%.*]] = tail call range(i32 0, 33) i32 @llvm.ctlz.i32(i32 [[X:%.*]], i1 false)
 ; CHECK-NEXT:    [[CAST:%.*]] = zext nneg i32 [[CT]] to i64
 ; CHECK-NEXT:    ret i64 [[CAST]]
 ;
@@ -219,7 +219,7 @@ define i64 @test6c(i32 %x) {
 
 define i16 @test1d(i64 %x) {
 ; CHECK-LABEL: @test1d(
-; CHECK-NEXT:    [[CT:%.*]] = tail call i64 @llvm.cttz.i64(i64 [[X:%.*]], i1 false), !range [[RNG2]]
+; CHECK-NEXT:    [[CT:%.*]] = tail call range(i64 0, 65) i64 @llvm.cttz.i64(i64 [[X:%.*]], i1 false)
 ; CHECK-NEXT:    [[CONV:%.*]] = trunc nuw nsw i64 [[CT]] to i16
 ; CHECK-NEXT:    ret i16 [[CONV]]
 ;
@@ -232,7 +232,7 @@ define i16 @test1d(i64 %x) {
 
 define i32 @test2d(i64 %x) {
 ; CHECK-LABEL: @test2d(
-; CHECK-NEXT:    [[CT:%.*]] = tail call i64 @llvm.cttz.i64(i64 [[X:%.*]], i1 false), !range [[RNG2]]
+; CHECK-NEXT:    [[CT:%.*]] = tail call range(i64 0, 65) i64 @llvm.cttz.i64(i64 [[X:%.*]], i1 false)
 ; CHECK-NEXT:    [[CAST:%.*]] = trunc nuw nsw i64 [[CT]] to i32
 ; CHECK-NEXT:    ret i32 [[CAST]]
 ;
@@ -245,7 +245,7 @@ define i32 @test2d(i64 %x) {
 
 define i16 @test3d(i32 %x) {
 ; CHECK-LABEL: @test3d(
-; CHECK-NEXT:    [[CT:%.*]] = tail call i32 @llvm.cttz.i32(i32 [[X:%.*]], i1 false), !range [[RNG1]]
+; CHECK-NEXT:    [[CT:%.*]] = tail call range(i32 0, 33) i32 @llvm.cttz.i32(i32 [[X:%.*]], i1 false)
 ; CHECK-NEXT:    [[CAST:%.*]] = trunc nuw nsw i32 [[CT]] to i16
 ; CHECK-NEXT:    ret i16 [[CAST]]
 ;
@@ -258,7 +258,7 @@ define i16 @test3d(i32 %x) {
 
 define i16 @test4d(i64 %x) {
 ; CHECK-LABEL: @test4d(
-; CHECK-NEXT:    [[CT:%.*]] = tail call i64 @llvm.ctlz.i64(i64 [[X:%.*]], i1 false), !range [[RNG2]]
+; CHECK-NEXT:    [[CT:%.*]] = tail call range(i64 0, 65) i64 @llvm.ctlz.i64(i64 [[X:%.*]], i1 false)
 ; CHECK-NEXT:    [[CAST:%.*]] = trunc nuw nsw i64 [[CT]] to i16
 ; CHECK-NEXT:    ret i16 [[CAST]]
 ;
@@ -271,7 +271,7 @@ define i16 @test4d(i64 %x) {
 
 define i32 @test5d(i64 %x) {
 ; CHECK-LABEL: @test5d(
-; CHECK-NEXT:    [[CT:%.*]] = tail call i64 @llvm.ctlz.i64(i64 [[X:%.*]], i1 false), !range [[RNG2]]
+; CHECK-NEXT:    [[CT:%.*]] = tail call range(i64 0, 65) i64 @llvm.ctlz.i64(i64 [[X:%.*]], i1 false)
 ; CHECK-NEXT:    [[CAST:%.*]] = trunc nuw nsw i64 [[CT]] to i32
 ; CHECK-NEXT:    ret i32 [[CAST]]
 ;
@@ -287,7 +287,7 @@ define i32 @test5d(i64 %x) {
 define i32 @not_op_ctlz(i64 %x) {
 ; CHECK-LABEL: @not_op_ctlz(
 ; CHECK-NEXT:    [[N:%.*]] = xor i64 [[X:%.*]], -1
-; CHECK-NEXT:    [[CT:%.*]] = tail call i64 @llvm.ctlz.i64(i64 [[N]], i1 false), !range [[RNG2]]
+; CHECK-NEXT:    [[CT:%.*]] = tail call range(i64 0, 65) i64 @llvm.ctlz.i64(i64 [[N]], i1 false)
 ; CHECK-NEXT:    [[CAST:%.*]] = trunc nuw nsw i64 [[CT]] to i32
 ; CHECK-NEXT:    ret i32 [[CAST]]
 ;
@@ -302,7 +302,7 @@ define i32 @not_op_ctlz(i64 %x) {
 define i32 @not_op_cttz(i64 %x) {
 ; CHECK-LABEL: @not_op_cttz(
 ; CHECK-NEXT:    [[N:%.*]] = xor i64 [[X:%.*]], -1
-; CHECK-NEXT:    [[CT:%.*]] = tail call i64 @llvm.cttz.i64(i64 [[N]], i1 false), !range [[RNG2]]
+; CHECK-NEXT:    [[CT:%.*]] = tail call range(i64 0, 65) i64 @llvm.cttz.i64(i64 [[N]], i1 false)
 ; CHECK-NEXT:    [[CAST:%.*]] = trunc nuw nsw i64 [[CT]] to i32
 ; CHECK-NEXT:    ret i32 [[CAST]]
 ;
@@ -319,7 +319,7 @@ define i32 @not_op_cttz(i64 %x) {
 define i32 @not_op_ctlz_wrong_xor_op1(i64 %x) {
 ; CHECK-LABEL: @not_op_ctlz_wrong_xor_op1(
 ; CHECK-NEXT:    [[N:%.*]] = xor i64 [[X:%.*]], -2
-; CHECK-NEXT:    [[CT:%.*]] = tail call i64 @llvm.ctlz.i64(i64 [[N]], i1 true), !range [[RNG2]]
+; CHECK-NEXT:    [[CT:%.*]] = tail call range(i64 0, 65) i64 @llvm.ctlz.i64(i64 [[N]], i1 true)
 ; CHECK-NEXT:    [[CAST:%.*]] = trunc nuw nsw i64 [[CT]] to i32
 ; CHECK-NEXT:    [[TOBOOL:%.*]] = icmp eq i64 [[X]], -1
 ; CHECK-NEXT:    [[R:%.*]] = select i1 [[TOBOOL]], i32 64, i32 [[CAST]]
@@ -338,7 +338,7 @@ define i32 @not_op_ctlz_wrong_xor_op1(i64 %x) {
 define i32 @not_op_ctlz_wrong_xor_op0(i64 %x, i64 %y) {
 ; CHECK-LABEL: @not_op_ctlz_wrong_xor_op0(
 ; CHECK-NEXT:    [[N:%.*]] = xor i64 [[Y:%.*]], -1
-; CHECK-NEXT:    [[CT:%.*]] = tail call i64 @llvm.ctlz.i64(i64 [[N]], i1 true), !range [[RNG2]]
+; CHECK-NEXT:    [[CT:%.*]] = tail call range(i64 0, 65) i64 @llvm.ctlz.i64(i64 [[N]], i1 true)
 ; CHECK-NEXT:    [[CAST:%.*]] = trunc nuw nsw i64 [[CT]] to i32
 ; CHECK-NEXT:    [[TOBOOL:%.*]] = icmp eq i64 [[X:%.*]], -1
 ; CHECK-NEXT:    [[R:%.*]] = select i1 [[TOBOOL]], i32 64, i32 [[CAST]]
@@ -357,7 +357,7 @@ define i32 @not_op_ctlz_wrong_xor_op0(i64 %x, i64 %y) {
 define i32 @not_op_cttz_wrong_cmp(i64 %x) {
 ; CHECK-LABEL: @not_op_cttz_wrong_cmp(
 ; CHECK-NEXT:    [[N:%.*]] = xor i64 [[X:%.*]], -1
-; CHECK-NEXT:    [[CT:%.*]] = tail call i64 @llvm.cttz.i64(i64 [[N]], i1 true), !range [[RNG2]]
+; CHECK-NEXT:    [[CT:%.*]] = tail call range(i64 0, 65) i64 @llvm.cttz.i64(i64 [[N]], i1 true)
 ; CHECK-NEXT:    [[CAST:%.*]] = trunc nuw nsw i64 [[CT]] to i32
 ; CHECK-NEXT:    [[TOBOOL:%.*]] = icmp eq i64 [[X]], 0
 ; CHECK-NEXT:    [[R:%.*]] = select i1 [[TOBOOL]], i32 64, i32 [[CAST]]
@@ -373,7 +373,7 @@ define i32 @not_op_cttz_wrong_cmp(i64 %x) {
 
 define i16 @test6d(i32 %x) {
 ; CHECK-LABEL: @test6d(
-; CHECK-NEXT:    [[CT:%.*]] = tail call i32 @llvm.ctlz.i32(i32 [[X:%.*]], i1 false), !range [[RNG1]]
+; CHECK-NEXT:    [[CT:%.*]] = tail call range(i32 0, 33) i32 @llvm.ctlz.i32(i32 [[X:%.*]], i1 false)
 ; CHECK-NEXT:    [[CAST:%.*]] = trunc nuw nsw i32 [[CT]] to i16
 ; CHECK-NEXT:    ret i16 [[CAST]]
 ;
@@ -386,7 +386,7 @@ define i16 @test6d(i32 %x) {
 
 define i64 @select_bug1(i32 %x) {
 ; CHECK-LABEL: @select_bug1(
-; CHECK-NEXT:    [[CT:%.*]] = tail call i32 @llvm.cttz.i32(i32 [[X:%.*]], i1 false), !range [[RNG1]]
+; CHECK-NEXT:    [[CT:%.*]] = tail call range(i32 0, 33) i32 @llvm.cttz.i32(i32 [[X:%.*]], i1 false)
 ; CHECK-NEXT:    [[CONV:%.*]] = zext nneg i32 [[CT]] to i64
 ; CHECK-NEXT:    ret i64 [[CONV]]
 ;
@@ -399,7 +399,7 @@ define i64 @select_bug1(i32 %x) {
 
 define i16 @select_bug2(i32 %x) {
 ; CHECK-LABEL: @select_bug2(
-; CHECK-NEXT:    [[CT:%.*]] = tail call i32 @llvm.cttz.i32(i32 [[X:%.*]], i1 false), !range [[RNG1]]
+; CHECK-NEXT:    [[CT:%.*]] = tail call range(i32 0, 33) i32 @llvm.cttz.i32(i32 [[X:%.*]], i1 false)
 ; CHECK-NEXT:    [[CONV:%.*]] = trunc nuw nsw i32 [[CT]] to i16
 ; CHECK-NEXT:    ret i16 [[CONV]]
 ;
@@ -412,7 +412,7 @@ define i16 @select_bug2(i32 %x) {
 
 define i128 @test7(i128 %x) {
 ; CHECK-LABEL: @test7(
-; CHECK-NEXT:    [[CT:%.*]] = tail call i128 @llvm.ctlz.i128(i128 [[X:%.*]], i1 false), !range [[RNG3:![0-9]+]]
+; CHECK-NEXT:    [[CT:%.*]] = tail call range(i128 0, 129) i128 @llvm.ctlz.i128(i128 [[X:%.*]], i1 false)
 ; CHECK-NEXT:    ret i128 [[CT]]
 ;
   %ct = tail call i128 @llvm.ctlz.i128(i128 %x, i1 true)
@@ -423,7 +423,7 @@ define i128 @test7(i128 %x) {
 
 define i128 @test8(i128 %x) {
 ; CHECK-LABEL: @test8(
-; CHECK-NEXT:    [[CT:%.*]] = tail call i128 @llvm.cttz.i128(i128 [[X:%.*]], i1 false), !range [[RNG3]]
+; CHECK-NEXT:    [[CT:%.*]] = tail call range(i128 0, 129) i128 @llvm.cttz.i128(i128 [[X:%.*]], i1 false)
 ; CHECK-NEXT:    ret i128 [[CT]]
 ;
   %ct = tail call i128 @llvm.cttz.i128(i128 %x, i1 true)
@@ -434,7 +434,7 @@ define i128 @test8(i128 %x) {
 
 define i32 @test_ctlz_not_bw(i32 %x) {
 ; CHECK-LABEL: @test_ctlz_not_bw(
-; CHECK-NEXT:    [[CT:%.*]] = tail call i32 @llvm.ctlz.i32(i32 [[X:%.*]], i1 true), !range [[RNG1]]
+; CHECK-NEXT:    [[CT:%.*]] = tail call range(i32 0, 33) i32 @llvm.ctlz.i32(i32 [[X:%.*]], i1 true)
 ; CHECK-NEXT:    [[CMP_NOT:%.*]] = icmp eq i32 [[X]], 0
 ; CHECK-NEXT:    [[RES:%.*]] = select i1 [[CMP_NOT]], i32 123, i32 [[CT]]
 ; CHECK-NEXT:    ret i32 [[RES]]
@@ -447,7 +447,7 @@ define i32 @test_ctlz_not_bw(i32 %x) {
 
 define i32 @test_ctlz_not_bw_multiuse(i32 %x) {
 ; CHECK-LABEL: @test_ctlz_not_bw_multiuse(
-; CHECK-NEXT:    [[CT:%.*]] = tail call i32 @llvm.ctlz.i32(i32 [[X:%.*]], i1 false), !range [[RNG1]]
+; CHECK-NEXT:    [[CT:%.*]] = tail call range(i32 0, 33) i32 @llvm.ctlz.i32(i32 [[X:%.*]], i1 false)
 ; CHECK-NEXT:    [[CMP_NOT:%.*]] = icmp eq i32 [[X]], 0
 ; CHECK-NEXT:    [[SEL:%.*]] = select i1 [[CMP_NOT]], i32 123, i32 [[CT]]
 ; CHECK-NEXT:    [[RES:%.*]] = or i32 [[SEL]], [[CT]]
@@ -462,7 +462,7 @@ define i32 @test_ctlz_not_bw_multiuse(i32 %x) {
 
 define i32 @test_cttz_not_bw(i32 %x) {
 ; CHECK-LABEL: @test_cttz_not_bw(
-; CHECK-NEXT:    [[CT:%.*]] = tail call i32 @llvm.cttz.i32(i32 [[X:%.*]], i1 true), !range [[RNG1]]
+; CHECK-NEXT:    [[CT:%.*]] = tail call range(i32 0, 33) i32 @llvm.cttz.i32(i32 [[X:%.*]], i1 true)
 ; CHECK-NEXT:    [[CMP_NOT:%.*]] = icmp eq i32 [[X]], 0
 ; CHECK-NEXT:    [[RES:%.*]] = select i1 [[CMP_NOT]], i32 123, i32 [[CT]]
 ; CHECK-NEXT:    ret i32 [[RES]]
@@ -475,7 +475,7 @@ define i32 @test_cttz_not_bw(i32 %x) {
 
 define i32 @test_cttz_not_bw_multiuse(i32 %x) {
 ; CHECK-LABEL: @test_cttz_not_bw_multiuse(
-; CHECK-NEXT:    [[CT:%.*]] = tail call i32 @llvm.cttz.i32(i32 [[X:%.*]], i1 false), !range [[RNG1]]
+; CHECK-NEXT:    [[CT:%.*]] = tail call range(i32 0, 33) i32 @llvm.cttz.i32(i32 [[X:%.*]], i1 false)
 ; CHECK-NEXT:    [[CMP_NOT:%.*]] = icmp eq i32 [[X]], 0
 ; CHECK-NEXT:    [[SEL:%.*]] = select i1 [[CMP_NOT]], i32 123, i32 [[CT]]
 ; CHECK-NEXT:    [[RES:%.*]] = or i32 [[SEL]], [[CT]]
@@ -490,7 +490,7 @@ define i32 @test_cttz_not_bw_multiuse(i32 %x) {
 
 define <2 x i32> @test_ctlz_bw_vec(<2 x i32> %x) {
 ; CHECK-LABEL: @test_ctlz_bw_vec(
-; CHECK-NEXT:    [[CT:%.*]] = tail call <2 x i32> @llvm.ctlz.v2i32(<2 x i32> [[X:%.*]], i1 false), !range [[RNG1]]
+; CHECK-NEXT:    [[CT:%.*]] = tail call range(i32 0, 33) <2 x i32> @llvm.ctlz.v2i32(<2 x i32> [[X:%.*]], i1 false)
 ; CHECK-NEXT:    ret <2 x i32> [[CT]]
 ;
   %ct = tail call <2 x i32> @llvm.ctlz.v2i32(<2 x i32> %x, i1 true)
@@ -501,7 +501,7 @@ define <2 x i32> @test_ctlz_bw_vec(<2 x i32> %x) {
 
 define <2 x i32> @test_ctlz_not_bw_vec(<2 x i32> %x) {
 ; CHECK-LABEL: @test_ctlz_not_bw_vec(
-; CHECK-NEXT:    [[CT:%.*]] = tail call <2 x i32> @llvm.ctlz.v2i32(<2 x i32> [[X:%.*]], i1 true), !range [[RNG1]]
+; CHECK-NEXT:    [[CT:%.*]] = tail call range(i32 0, 33) <2 x i32> @llvm.ctlz.v2i32(<2 x i32> [[X:%.*]], i1 true)
 ; CHECK-NEXT:    [[CMP_NOT:%.*]] = icmp eq <2 x i32> [[X]], zeroinitializer
 ; CHECK-NEXT:    [[RES:%.*]] = select <2 x i1> [[CMP_NOT]], <2 x i32> zeroinitializer, <2 x i32> [[CT]]
 ; CHECK-NEXT:    ret <2 x i32> [[RES]]
@@ -514,7 +514,7 @@ define <2 x i32> @test_ctlz_not_bw_vec(<2 x i32> %x) {
 
 define <2 x i32> @test_cttz_bw_vec(<2 x i32> %x) {
 ; CHECK-LABEL: @test_cttz_bw_vec(
-; CHECK-NEXT:    [[CT:%.*]] = tail call <2 x i32> @llvm.cttz.v2i32(<2 x i32> [[X:%.*]], i1 false), !range [[RNG1]]
+; CHECK-NEXT:    [[CT:%.*]] = tail call range(i32 0, 33) <2 x i32> @llvm.cttz.v2i32(<2 x i32> [[X:%.*]], i1 false)
 ; CHECK-NEXT:    ret <2 x i32> [[CT]]
 ;
   %ct = tail call <2 x i32> @llvm.cttz.v2i32(<2 x i32> %x, i1 true)
@@ -525,7 +525,7 @@ define <2 x i32> @test_cttz_bw_vec(<2 x i32> %x) {
 
 define <2 x i32> @test_cttz_not_bw_vec(<2 x i32> %x) {
 ; CHECK-LABEL: @test_cttz_not_bw_vec(
-; CHECK-NEXT:    [[CT:%.*]] = tail call <2 x i32> @llvm.cttz.v2i32(<2 x i32> [[X:%.*]], i1 true), !range [[RNG1]]
+; CHECK-NEXT:    [[CT:%.*]] = tail call range(i32 0, 33) <2 x i32> @llvm.cttz.v2i32(<2 x i32> [[X:%.*]], i1 true)
 ; CHECK-NEXT:    [[CMP_NOT:%.*]] = icmp eq <2 x i32> [[X]], zeroinitializer
 ; CHECK-NEXT:    [[RES:%.*]] = select <2 x i1> [[CMP_NOT]], <2 x i32> zeroinitializer, <2 x i32> [[CT]]
 ; CHECK-NEXT:    ret <2 x i32> [[RES]]
@@ -538,7 +538,7 @@ define <2 x i32> @test_cttz_not_bw_vec(<2 x i32> %x) {
 
 define i32 @test_multiuse_def(i32 %x, ptr %p) {
 ; CHECK-LABEL: @test_multiuse_def(
-; CHECK-NEXT:    [[CT:%.*]] = tail call i32 @llvm.ctlz.i32(i32 [[X:%.*]], i1 false), !range [[RNG1]]
+; CHECK-NEXT:    [[CT:%.*]] = tail call range(i32 0, 33) i32 @llvm.ctlz.i32(i32 [[X:%.*]], i1 false)
 ; CHECK-NEXT:    store i32 [[CT]], ptr [[P:%.*]], align 4
 ; CHECK-NEXT:    ret i32 [[CT]]
 ;
@@ -551,7 +551,7 @@ define i32 @test_multiuse_def(i32 %x, ptr %p) {
 
 define i32 @test_multiuse_undef(i32 %x, ptr %p) {
 ; CHECK-LABEL: @test_multiuse_undef(
-; CHECK-NEXT:    [[CT:%.*]] = tail call i32 @llvm.ctlz.i32(i32 [[X:%.*]], i1 false), !range [[RNG1]]
+; CHECK-NEXT:    [[CT:%.*]] = tail call range(i32 0, 33) i32 @llvm.ctlz.i32(i32 [[X:%.*]], i1 false)
 ; CHECK-NEXT:    store i32 [[CT]], ptr [[P:%.*]], align 4
 ; CHECK-NEXT:    ret i32 [[CT]]
 ;
@@ -564,7 +564,7 @@ define i32 @test_multiuse_undef(i32 %x, ptr %p) {
 
 define i64 @test_multiuse_zext_def(i32 %x, ptr %p) {
 ; CHECK-LABEL: @test_multiuse_zext_def(
-; CHECK-NEXT:    [[CT:%.*]] = tail call i32 @llvm.cttz.i32(i32 [[X:%.*]], i1 false), !range [[RNG1]]
+; CHECK-NEXT:    [[CT:%.*]] = tail call range(i32 0, 33) i32 @llvm.cttz.i32(i32 [[X:%.*]], i1 false)
 ; CHECK-NEXT:    [[CONV:%.*]] = zext nneg i32 [[CT]] to i64
 ; CHECK-NEXT:    store i64 [[CONV]], ptr [[P:%.*]], align 4
 ; CHECK-NEXT:    ret i64 [[CONV]]
@@ -579,7 +579,7 @@ define i64 @test_multiuse_zext_def(i32 %x, ptr %p) {
 
 define i64 @test_multiuse_zext_undef(i32 %x, ptr %p) {
 ; CHECK-LABEL: @test_multiuse_zext_undef(
-; CHECK-NEXT:    [[CT:%.*]] = tail call i32 @llvm.cttz.i32(i32 [[X:%.*]], i1 false), !range [[RNG1]]
+; CHECK-NEXT:    [[CT:%.*]] = tail call range(i32 0, 33) i32 @llvm.cttz.i32(i32 [[X:%.*]], i1 false)
 ; CHECK-NEXT:    [[CONV:%.*]] = zext nneg i32 [[CT]] to i64
 ; CHECK-NEXT:    store i64 [[CONV]], ptr [[P:%.*]], align 4
 ; CHECK-NEXT:    ret i64 [[CONV]]
@@ -594,7 +594,7 @@ define i64 @test_multiuse_zext_undef(i32 %x, ptr %p) {
 
 define i16 @test_multiuse_trunc_def(i64 %x, ptr %p) {
 ; CHECK-LABEL: @test_multiuse_trunc_def(
-; CHECK-NEXT:    [[CT:%.*]] = tail call i64 @llvm.cttz.i64(i64 [[X:%.*]], i1 false), !range [[RNG2]]
+; CHECK-NEXT:    [[CT:%.*]] = tail call range(i64 0, 65) i64 @llvm.cttz.i64(i64 [[X:%.*]], i1 false)
 ; CHECK-NEXT:    [[CONV:%.*]] = trunc nuw nsw i64 [[CT]] to i16
 ; CHECK-NEXT:    store i16 [[CONV]], ptr [[P:%.*]], align 2
 ; CHECK-NEXT:    ret i16 [[CONV]]
@@ -609,7 +609,7 @@ define i16 @test_multiuse_trunc_def(i64 %x, ptr %p) {
 
 define i16 @test_multiuse_trunc_undef(i64 %x, ptr %p) {
 ; CHECK-LABEL: @test_multiuse_trunc_undef(
-; CHECK-NEXT:    [[CT:%.*]] = tail call i64 @llvm.cttz.i64(i64 [[X:%.*]], i1 false), !range [[RNG2]]
+; CHECK-NEXT:    [[CT:%.*]] = tail call range(i64 0, 65) i64 @llvm.cttz.i64(i64 [[X:%.*]], i1 false)
 ; CHECK-NEXT:    [[CONV:%.*]] = trunc nuw nsw i64 [[CT]] to i16
 ; CHECK-NEXT:    store i16 [[CONV]], ptr [[P:%.*]], align 2
 ; CHECK-NEXT:    ret i16 [[CONV]]
diff --git a/llvm/test/Transforms/InstCombine/select-ctlz-to-cttz.ll b/llvm/test/Transforms/InstCombine/select-ctlz-to-cttz.ll
index fa8443d4c957..59d33ee3b39d 100644
--- a/llvm/test/Transforms/InstCombine/select-ctlz-to-cttz.ll
+++ b/llvm/test/Transforms/InstCombine/select-ctlz-to-cttz.ll
@@ -16,7 +16,7 @@ declare void @use2(i1)
 
 define i32 @select_clz_to_ctz(i32 %a) {
 ; CHECK-LABEL: @select_clz_to_ctz(
-; CHECK-NEXT:    [[SUB1:%.*]] = call i32 @llvm.cttz.i32(i32 [[A:%.*]], i1 true), !range [[RNG0:![0-9]+]]
+; CHECK-NEXT:    [[SUB1:%.*]] = call range(i32 0, 33) i32 @llvm.cttz.i32(i32 [[A:%.*]], i1 true)
 ; CHECK-NEXT:    ret i32 [[SUB1]]
 ;
   %sub = sub i32 0, %a
@@ -30,7 +30,7 @@ define i32 @select_clz_to_ctz(i32 %a) {
 
 define i32 @select_clz_to_ctz_preserve_flag(i32 %a) {
 ; CHECK-LABEL: @select_clz_to_ctz_preserve_flag(
-; CHECK-NEXT:    [[COND:%.*]] = call i32 @llvm.cttz.i32(i32 [[A:%.*]], i1 false), !range [[RNG0]]
+; CHECK-NEXT:    [[COND:%.*]] = call range(i32 0, 33) i32 @llvm.cttz.i32(i32 [[A:%.*]], i1 false)
 ; CHECK-NEXT:    ret i32 [[COND]]
 ;
   %sub = sub i32 0, %a
@@ -44,7 +44,7 @@ define i32 @select_clz_to_ctz_preserve_flag(i32 %a) {
 
 define i32 @select_clz_to_ctz_constant_for_zero(i32 %a) {
 ; CHECK-LABEL: @select_clz_to_ctz_constant_for_zero(
-; CHECK-NEXT:    [[COND:%.*]] = call i32 @llvm.cttz.i32(i32 [[A:%.*]], i1 false), !range [[RNG0]]
+; CHECK-NEXT:    [[COND:%.*]] = call range(i32 0, 33) i32 @llvm.cttz.i32(i32 [[A:%.*]], i1 false)
 ; CHECK-NEXT:    ret i32 [[COND]]
 ;
   %sub = sub i32 0, %a
@@ -58,7 +58,7 @@ define i32 @select_clz_to_ctz_constant_for_zero(i32 %a) {
 
 define <2 x i32> @select_clz_to_ctz_vec(<2 x i32> %a) {
 ; CHECK-LABEL: @select_clz_to_ctz_vec(
-; CHECK-NEXT:    [[COND:%.*]] = call <2 x i32> @llvm.cttz.v2i32(<2 x i32> [[A:%.*]], i1 true), !range [[RNG0]]
+; CHECK-NEXT:    [[COND:%.*]] = call range(i32 0, 33) <2 x i32> @llvm.cttz.v2i32(<2 x i32> [[A:%.*]], i1 true)
 ; CHECK-NEXT:    ret <2 x i32> [[COND]]
 ;
   %sub = sub <2 x i32> zeroinitializer, %a
@@ -72,7 +72,7 @@ define <2 x i32> @select_clz_to_ctz_vec(<2 x i32> %a) {
 
 define i32 @select_clz_to_ctz_extra_use(i32 %a) {
 ; CHECK-LABEL: @select_clz_to_ctz_extra_use(
-; CHECK-NEXT:    [[SUB1:%.*]] = call i32 @llvm.cttz.i32(i32 [[A:%.*]], i1 true), !range [[RNG0]]
+; CHECK-NEXT:    [[SUB1:%.*]] = call range(i32 0, 33) i32 @llvm.cttz.i32(i32 [[A:%.*]], i1 true)
 ; CHECK-NEXT:    call void @use(i32 [[SUB1]])
 ; CHECK-NEXT:    ret i32 [[SUB1]]
 ;
@@ -88,7 +88,7 @@ define i32 @select_clz_to_ctz_extra_use(i32 %a) {
 
 define i32 @select_clz_to_ctz_and_commuted(i32 %a) {
 ; CHECK-LABEL: @select_clz_to_ctz_and_commuted(
-; CHECK-NEXT:    [[SUB1:%.*]] = call i32 @llvm.cttz.i32(i32 [[A:%.*]], i1 true), !range [[RNG0]]
+; CHECK-NEXT:    [[SUB1:%.*]] = call range(i32 0, 33) i32 @llvm.cttz.i32(i32 [[A:%.*]], i1 true)
 ; CHECK-NEXT:    ret i32 [[SUB1]]
 ;
   %sub = sub i32 0, %a
@@ -104,7 +104,7 @@ define i32 @select_clz_to_ctz_icmp_ne(i32 %a) {
 ; CHECK-LABEL: @select_clz_to_ctz_icmp_ne(
 ; CHECK-NEXT:    [[TOBOOL:%.*]] = icmp ne i32 [[A:%.*]], 0
 ; CHECK-NEXT:    call void @use2(i1 [[TOBOOL]])
-; CHECK-NEXT:    [[SUB1:%.*]] = call i32 @llvm.cttz.i32(i32 [[A]], i1 true), !range [[RNG0]]
+; CHECK-NEXT:    [[SUB1:%.*]] = call range(i32 0, 33) i32 @llvm.cttz.i32(i32 [[A]], i1 true)
 ; CHECK-NEXT:    ret i32 [[SUB1]]
 ;
   %sub = sub i32 0, %a
@@ -119,7 +119,7 @@ define i32 @select_clz_to_ctz_icmp_ne(i32 %a) {
 
 define i64 @select_clz_to_ctz_i64(i64 %a) {
 ; CHECK-LABEL: @select_clz_to_ctz_i64(
-; CHECK-NEXT:    [[SUB1:%.*]] = call i64 @llvm.cttz.i64(i64 [[A:%.*]], i1 true), !range [[RNG1:![0-9]+]]
+; CHECK-NEXT:    [[SUB1:%.*]] = call range(i64 0, 65) i64 @llvm.cttz.i64(i64 [[A:%.*]], i1 true)
 ; CHECK-NEXT:    ret i64 [[SUB1]]
 ;
   %sub = sub i64 0, %a
@@ -137,7 +137,7 @@ define i32 @select_clz_to_ctz_wrong_sub(i32 %a) {
 ; CHECK-LABEL: @select_clz_to_ctz_wrong_sub(
 ; CHECK-NEXT:    [[SUB:%.*]] = sub i32 1, [[A:%.*]]
 ; CHECK-NEXT:    [[AND:%.*]] = and i32 [[SUB]], [[A]]
-; CHECK-NEXT:    [[LZ:%.*]] = tail call i32 @llvm.ctlz.i32(i32 [[AND]], i1 true), !range [[RNG0]]
+; CHECK-NEXT:    [[LZ:%.*]] = tail call range(i32 0, 33) i32 @llvm.ctlz.i32(i32 [[AND]], i1 true)
 ; CHECK-NEXT:    [[SUB1:%.*]] = xor i32 [[LZ]], 31
 ; CHECK-NEXT:    ret i32 [[SUB1]]
 ;
@@ -155,7 +155,7 @@ define i64 @select_clz_to_ctz_i64_wrong_xor(i64 %a) {
 ; CHECK-LABEL: @select_clz_to_ctz_i64_wrong_xor(
 ; CHECK-NEXT:    [[SUB:%.*]] = sub i64 0, [[A:%.*]]
 ; CHECK-NEXT:    [[AND:%.*]] = and i64 [[SUB]], [[A]]
-; CHECK-NEXT:    [[LZ:%.*]] = tail call i64 @llvm.ctlz.i64(i64 [[AND]], i1 true), !range [[RNG1]]
+; CHECK-NEXT:    [[LZ:%.*]] = tail call range(i64 0, 65) i64 @llvm.ctlz.i64(i64 [[AND]], i1 true)
 ; CHECK-NEXT:    [[SUB11:%.*]] = or disjoint i64 [[LZ]], 64
 ; CHECK-NEXT:    ret i64 [[SUB11]]
 ;
@@ -171,7 +171,7 @@ define i64 @select_clz_to_ctz_i64_wrong_xor(i64 %a) {
 define i64 @select_clz_to_ctz_i64_wrong_icmp_cst(i64 %a) {
 ; CHECK-LABEL: @select_clz_to_ctz_i64_wrong_icmp_cst(
 ; CHECK-NEXT:    [[TOBOOL:%.*]] = icmp eq i64 [[A:%.*]], 1
-; CHECK-NEXT:    [[SUB1:%.*]] = call i64 @llvm.cttz.i64(i64 [[A]], i1 true), !range [[RNG1]]
+; CHECK-NEXT:    [[SUB1:%.*]] = call range(i64 0, 65) i64 @llvm.cttz.i64(i64 [[A]], i1 true)
 ; CHECK-NEXT:    [[COND:%.*]] = select i1 [[TOBOOL]], i64 63, i64 [[SUB1]]
 ; CHECK-NEXT:    ret i64 [[COND]]
 ;
@@ -188,7 +188,7 @@ define i64 @select_clz_to_ctz_i64_wrong_icmp_pred(i64 %a) {
 ; CHECK-LABEL: @select_clz_to_ctz_i64_wrong_icmp_pred(
 ; CHECK-NEXT:    [[SUB:%.*]] = sub i64 0, [[A:%.*]]
 ; CHECK-NEXT:    [[AND:%.*]] = and i64 [[SUB]], [[A]]
-; CHECK-NEXT:    [[LZ:%.*]] = tail call i64 @llvm.ctlz.i64(i64 [[AND]], i1 true), !range [[RNG1]]
+; CHECK-NEXT:    [[LZ:%.*]] = tail call range(i64 0, 65) i64 @llvm.ctlz.i64(i64 [[AND]], i1 true)
 ; CHECK-NEXT:    [[TOBOOL:%.*]] = icmp slt i64 [[A]], 0
 ; CHECK-NEXT:    [[SUB1:%.*]] = xor i64 [[LZ]], 63
 ; CHECK-NEXT:    [[COND:%.*]] = select i1 [[TOBOOL]], i64 [[LZ]], i64 [[SUB1]]
@@ -207,7 +207,7 @@ define <2 x i32> @select_clz_to_ctz_vec_with_undef(<2 x i32> %a) {
 ; CHECK-LABEL: @select_clz_to_ctz_vec_with_undef(
 ; CHECK-NEXT:    [[SUB:%.*]] = sub <2 x i32> zeroinitializer, [[A:%.*]]
 ; CHECK-NEXT:    [[AND:%.*]] = and <2 x i32> [[SUB]], [[A]]
-; CHECK-NEXT:    [[LZ:%.*]] = tail call <2 x i32> @llvm.ctlz.v2i32(<2 x i32> [[AND]], i1 true), !range [[RNG0]]
+; CHECK-NEXT:    [[LZ:%.*]] = tail call range(i32 0, 33) <2 x i32> @llvm.ctlz.v2i32(<2 x i32> [[AND]], i1 true)
 ; CHECK-NEXT:    [[TOBOOL:%.*]] = icmp eq <2 x i32> [[A]], zeroinitializer
 ; CHECK-NEXT:    [[SUB1:%.*]] = xor <2 x i32> [[LZ]], <i32 31, i32 undef>
 ; CHECK-NEXT:    [[COND:%.*]] = select <2 x i1> [[TOBOOL]], <2 x i32> [[LZ]], <2 x i32> [[SUB1]]
@@ -226,7 +226,7 @@ define i32 @select_clz_to_ctz_wrong_constant_for_zero(i32 %a) {
 ; CHECK-LABEL: @select_clz_to_ctz_wrong_constant_for_zero(
 ; CHECK-NEXT:    [[SUB:%.*]] = sub i32 0, [[A:%.*]]
 ; CHECK-NEXT:    [[AND:%.*]] = and i32 [[SUB]], [[A]]
-; CHECK-NEXT:    [[LZ:%.*]] = tail call i32 @llvm.ctlz.i32(i32 [[AND]], i1 false), !range [[RNG0]]
+; CHECK-NEXT:    [[LZ:%.*]] = tail call range(i32 0, 33) i32 @llvm.ctlz.i32(i32 [[AND]], i1 false)
 ; CHECK-NEXT:    [[TOBOOL:%.*]] = icmp eq i32 [[A]], 0
 ; CHECK-NEXT:    [[SUB1:%.*]] = xor i32 [[LZ]], 31
 ; CHECK-NEXT:    [[COND:%.*]] = select i1 [[TOBOOL]], i32 31, i32 [[SUB1]]
@@ -243,7 +243,7 @@ define i32 @select_clz_to_ctz_wrong_constant_for_zero(i32 %a) {
 
 define i4 @PR45762(i3 %x4) {
 ; CHECK-LABEL: @PR45762(
-; CHECK-NEXT:    [[T4:%.*]] = call i3 @llvm.cttz.i3(i3 [[X4:%.*]], i1 false), !range [[RNG2:![0-9]+]]
+; CHECK-NEXT:    [[T4:%.*]] = call range(i3 0, -4) i3 @llvm.cttz.i3(i3 [[X4:%.*]], i1 false)
 ; CHECK-NEXT:    [[T7:%.*]] = zext nneg i3 [[T4]] to i4
 ; CHECK-NEXT:    [[ONE_HOT_16:%.*]] = shl nuw i4 1, [[T7]]
 ; CHECK-NEXT:    [[OR_69_NOT:%.*]] = icmp eq i3 [[X4]], 0
@@ -272,7 +272,7 @@ define i4 @PR45762(i3 %x4) {
 
 define i4 @PR45762_logical(i3 %x4) {
 ; CHECK-LABEL: @PR45762_logical(
-; CHECK-NEXT:    [[T4:%.*]] = call i3 @llvm.cttz.i3(i3 [[X4:%.*]], i1 false), !range [[RNG2]]
+; CHECK-NEXT:    [[T4:%.*]] = call range(i3 0, -4) i3 @llvm.cttz.i3(i3 [[X4:%.*]], i1 false)
 ; CHECK-NEXT:    [[T7:%.*]] = zext nneg i3 [[T4]] to i4
 ; CHECK-NEXT:    [[ONE_HOT_16:%.*]] = shl nuw i4 1, [[T7]]
 ; CHECK-NEXT:    [[OR_69_NOT:%.*]] = icmp eq i3 [[X4]], 0
diff --git a/llvm/test/Transforms/InstCombine/select-select.ll b/llvm/test/Transforms/InstCombine/select-select.ll
index 785766136fb7..84fe973093e3 100644
--- a/llvm/test/Transforms/InstCombine/select-select.ll
+++ b/llvm/test/Transforms/InstCombine/select-select.ll
@@ -18,11 +18,10 @@ define float @foo1(float %a) {
 
 define float @foo2(float %a) {
 ; CHECK-LABEL: @foo2(
-; CHECK-NEXT:    [[B:%.*]] = fcmp ogt float [[A:%.*]], 0.000000e+00
-; CHECK-NEXT:    [[C:%.*]] = select i1 [[B]], float [[A]], float 0.000000e+00
+; CHECK-NEXT:    [[B:%.*]] = fcmp ule float [[C:%.*]], 0.000000e+00
 ; CHECK-NEXT:    [[D:%.*]] = fcmp olt float [[C]], 1.000000e+00
-; CHECK-NEXT:    [[E:%.*]] = select i1 [[B]], float [[A]], float 0.000000e+00
-; CHECK-NEXT:    [[F:%.*]] = select i1 [[D]], float [[E]], float 1.000000e+00
+; CHECK-NEXT:    [[E:%.*]] = select i1 [[D]], float [[C]], float 1.000000e+00
+; CHECK-NEXT:    [[F:%.*]] = select i1 [[B]], float 0.000000e+00, float [[E]]
 ; CHECK-NEXT:    ret float [[F]]
 ;
   %b = fcmp ogt float %a, 0.0
diff --git a/llvm/test/Transforms/InstCombine/select.ll b/llvm/test/Transforms/InstCombine/select.ll
index 2ec092a745c5..2efe2742ca49 100644
--- a/llvm/test/Transforms/InstCombine/select.ll
+++ b/llvm/test/Transforms/InstCombine/select.ll
@@ -2736,7 +2736,7 @@ define void @select_freeze_icmp_multuses(i32 %x, i32 %y) {
 define i32 @pr47322_more_poisonous_replacement(i32 %arg) {
 ; CHECK-LABEL: @pr47322_more_poisonous_replacement(
 ; CHECK-NEXT:    [[CMP:%.*]] = icmp eq i32 [[ARG:%.*]], 0
-; CHECK-NEXT:    [[TRAILING:%.*]] = call i32 @llvm.cttz.i32(i32 [[ARG]], i1 immarg true), !range [[RNG0:![0-9]+]]
+; CHECK-NEXT:    [[TRAILING:%.*]] = call range(i32 0, 33) i32 @llvm.cttz.i32(i32 [[ARG]], i1 immarg true)
 ; CHECK-NEXT:    [[SHIFTED:%.*]] = lshr i32 [[ARG]], [[TRAILING]]
 ; CHECK-NEXT:    [[R1_SROA_0_1:%.*]] = select i1 [[CMP]], i32 0, i32 [[SHIFTED]]
 ; CHECK-NEXT:    ret i32 [[R1_SROA_0_1]]
@@ -3830,14 +3830,17 @@ entry:
   ret i32 %cond
 }
 
-; FIXME: This is a miscompile.
 define <2 x i32> @src_and_eq_C_xor_OrAndNotC_vec_poison(<2 x i32> %0, <2 x i32> %1, <2 x i32> %2) {
 ; CHECK-LABEL: @src_and_eq_C_xor_OrAndNotC_vec_poison(
 ; CHECK-NEXT:  entry:
-; CHECK-NEXT:    [[OR:%.*]] = or <2 x i32> [[TMP1:%.*]], [[TMP0:%.*]]
-; CHECK-NEXT:    [[NOT:%.*]] = xor <2 x i32> [[TMP2:%.*]], <i32 -1, i32 poison>
+; CHECK-NEXT:    [[AND:%.*]] = and <2 x i32> [[TMP1:%.*]], [[TMP0:%.*]]
+; CHECK-NEXT:    [[CMP:%.*]] = icmp eq <2 x i32> [[AND]], [[TMP2:%.*]]
+; CHECK-NEXT:    [[XOR:%.*]] = xor <2 x i32> [[TMP1]], [[TMP0]]
+; CHECK-NEXT:    [[OR:%.*]] = or <2 x i32> [[TMP1]], [[TMP0]]
+; CHECK-NEXT:    [[NOT:%.*]] = xor <2 x i32> [[TMP2]], <i32 -1, i32 poison>
 ; CHECK-NEXT:    [[AND1:%.*]] = and <2 x i32> [[OR]], [[NOT]]
-; CHECK-NEXT:    ret <2 x i32> [[AND1]]
+; CHECK-NEXT:    [[COND:%.*]] = select <2 x i1> [[CMP]], <2 x i32> [[XOR]], <2 x i32> [[AND1]]
+; CHECK-NEXT:    ret <2 x i32> [[COND]]
 ;
 entry:
   %and = and <2 x i32> %1, %0
diff --git a/llvm/test/Transforms/InstCombine/sext.ll b/llvm/test/Transforms/InstCombine/sext.ll
index 6d263cfcda05..a554f2b28d6f 100644
--- a/llvm/test/Transforms/InstCombine/sext.ll
+++ b/llvm/test/Transforms/InstCombine/sext.ll
@@ -11,7 +11,7 @@ declare void @use_vec(<2 x i5>)
 
 define i64 @test1(i32 %x) {
 ; CHECK-LABEL: @test1(
-; CHECK-NEXT:    [[T:%.*]] = call i32 @llvm.ctpop.i32(i32 [[X:%.*]]), !range [[RNG0:![0-9]+]]
+; CHECK-NEXT:    [[T:%.*]] = call range(i32 0, 33) i32 @llvm.ctpop.i32(i32 [[X:%.*]])
 ; CHECK-NEXT:    [[S:%.*]] = zext nneg i32 [[T]] to i64
 ; CHECK-NEXT:    ret i64 [[S]]
 ;
@@ -22,7 +22,7 @@ define i64 @test1(i32 %x) {
 
 define i64 @test2(i32 %x) {
 ; CHECK-LABEL: @test2(
-; CHECK-NEXT:    [[T:%.*]] = call i32 @llvm.ctlz.i32(i32 [[X:%.*]], i1 true), !range [[RNG0]]
+; CHECK-NEXT:    [[T:%.*]] = call range(i32 0, 33) i32 @llvm.ctlz.i32(i32 [[X:%.*]], i1 true)
 ; CHECK-NEXT:    [[S:%.*]] = zext nneg i32 [[T]] to i64
 ; CHECK-NEXT:    ret i64 [[S]]
 ;
@@ -33,7 +33,7 @@ define i64 @test2(i32 %x) {
 
 define i64 @test3(i32 %x) {
 ; CHECK-LABEL: @test3(
-; CHECK-NEXT:    [[T:%.*]] = call i32 @llvm.cttz.i32(i32 [[X:%.*]], i1 true), !range [[RNG0]]
+; CHECK-NEXT:    [[T:%.*]] = call range(i32 0, 33) i32 @llvm.cttz.i32(i32 [[X:%.*]], i1 true)
 ; CHECK-NEXT:    [[S:%.*]] = zext nneg i32 [[T]] to i64
 ; CHECK-NEXT:    ret i64 [[S]]
 ;
diff --git a/llvm/test/Transforms/InstCombine/shift-cttz-ctlz.ll b/llvm/test/Transforms/InstCombine/shift-cttz-ctlz.ll
index 2b2f820c9a09..1c381d083907 100644
--- a/llvm/test/Transforms/InstCombine/shift-cttz-ctlz.ll
+++ b/llvm/test/Transforms/InstCombine/shift-cttz-ctlz.ll
@@ -5,7 +5,7 @@ define i32 @shl_cttz_false(i32 %x, i32 %y) {
 ; CHECK-LABEL: define i32 @shl_cttz_false(
 ; CHECK-SAME: i32 [[X:%.*]], i32 [[Y:%.*]]) {
 ; CHECK-NEXT:  entry:
-; CHECK-NEXT:    [[CTTZ:%.*]] = call i32 @llvm.cttz.i32(i32 [[Y]], i1 true), !range [[RNG0:![0-9]+]]
+; CHECK-NEXT:    [[CTTZ:%.*]] = call range(i32 0, 33) i32 @llvm.cttz.i32(i32 [[Y]], i1 true)
 ; CHECK-NEXT:    [[RES:%.*]] = shl i32 [[X]], [[CTTZ]]
 ; CHECK-NEXT:    ret i32 [[RES]]
 ;
@@ -19,7 +19,7 @@ define i32 @shl_ctlz_false(i32 %x, i32 %y) {
 ; CHECK-LABEL: define i32 @shl_ctlz_false(
 ; CHECK-SAME: i32 [[X:%.*]], i32 [[Y:%.*]]) {
 ; CHECK-NEXT:  entry:
-; CHECK-NEXT:    [[CTTZ:%.*]] = call i32 @llvm.ctlz.i32(i32 [[Y]], i1 true), !range [[RNG0]]
+; CHECK-NEXT:    [[CTTZ:%.*]] = call range(i32 0, 33) i32 @llvm.ctlz.i32(i32 [[Y]], i1 true)
 ; CHECK-NEXT:    [[RES:%.*]] = shl i32 [[X]], [[CTTZ]]
 ; CHECK-NEXT:    ret i32 [[RES]]
 ;
@@ -33,7 +33,7 @@ define i32 @lshr_cttz_false(i32 %x, i32 %y) {
 ; CHECK-LABEL: define i32 @lshr_cttz_false(
 ; CHECK-SAME: i32 [[X:%.*]], i32 [[Y:%.*]]) {
 ; CHECK-NEXT:  entry:
-; CHECK-NEXT:    [[CTTZ:%.*]] = call i32 @llvm.cttz.i32(i32 [[Y]], i1 true), !range [[RNG0]]
+; CHECK-NEXT:    [[CTTZ:%.*]] = call range(i32 0, 33) i32 @llvm.cttz.i32(i32 [[Y]], i1 true)
 ; CHECK-NEXT:    [[RES:%.*]] = lshr i32 [[X]], [[CTTZ]]
 ; CHECK-NEXT:    ret i32 [[RES]]
 ;
@@ -47,7 +47,7 @@ define i32 @ashr_cttz_false(i32 %x, i32 %y) {
 ; CHECK-LABEL: define i32 @ashr_cttz_false(
 ; CHECK-SAME: i32 [[X:%.*]], i32 [[Y:%.*]]) {
 ; CHECK-NEXT:  entry:
-; CHECK-NEXT:    [[CTTZ:%.*]] = call i32 @llvm.cttz.i32(i32 [[Y]], i1 true), !range [[RNG0]]
+; CHECK-NEXT:    [[CTTZ:%.*]] = call range(i32 0, 33) i32 @llvm.cttz.i32(i32 [[Y]], i1 true)
 ; CHECK-NEXT:    [[RES:%.*]] = ashr i32 [[X]], [[CTTZ]]
 ; CHECK-NEXT:    ret i32 [[RES]]
 ;
@@ -61,7 +61,7 @@ define i32 @shl_cttz_false_multiuse(i32 %x, i32 %y) {
 ; CHECK-LABEL: define i32 @shl_cttz_false_multiuse(
 ; CHECK-SAME: i32 [[X:%.*]], i32 [[Y:%.*]]) {
 ; CHECK-NEXT:  entry:
-; CHECK-NEXT:    [[CTTZ:%.*]] = call i32 @llvm.cttz.i32(i32 [[Y]], i1 false), !range [[RNG0]]
+; CHECK-NEXT:    [[CTTZ:%.*]] = call range(i32 0, 33) i32 @llvm.cttz.i32(i32 [[Y]], i1 false)
 ; CHECK-NEXT:    call void @use(i32 [[CTTZ]])
 ; CHECK-NEXT:    [[RES:%.*]] = shl i32 [[X]], [[CTTZ]]
 ; CHECK-NEXT:    ret i32 [[RES]]
@@ -77,7 +77,7 @@ define i32 @shl_cttz_as_lhs(i32 %x, i32 %y) {
 ; CHECK-LABEL: define i32 @shl_cttz_as_lhs(
 ; CHECK-SAME: i32 [[X:%.*]], i32 [[Y:%.*]]) {
 ; CHECK-NEXT:  entry:
-; CHECK-NEXT:    [[CTTZ:%.*]] = call i32 @llvm.cttz.i32(i32 [[Y]], i1 false), !range [[RNG0]]
+; CHECK-NEXT:    [[CTTZ:%.*]] = call range(i32 0, 33) i32 @llvm.cttz.i32(i32 [[Y]], i1 false)
 ; CHECK-NEXT:    [[RES:%.*]] = shl i32 [[CTTZ]], [[X]]
 ; CHECK-NEXT:    ret i32 [[RES]]
 ;
@@ -88,6 +88,3 @@ entry:
 }
 
 declare void @use(i32)
-;.
-; CHECK: [[RNG0]] = !{i32 0, i32 33}
-;.
diff --git a/llvm/test/Transforms/InstCombine/shift.ll b/llvm/test/Transforms/InstCombine/shift.ll
index bb8661919c89..8da52e074637 100644
--- a/llvm/test/Transforms/InstCombine/shift.ll
+++ b/llvm/test/Transforms/InstCombine/shift.ll
@@ -366,7 +366,7 @@ define i32 @test26(i32 %A) {
   ret i32 %D
 }
 
-define i1 @test27(i32 %x) nounwind {
+define i1 @test27(i32 %x) {
 ; CHECK-LABEL: @test27(
 ; CHECK-NEXT:    [[TMP1:%.*]] = and i32 [[X:%.*]], 8
 ; CHECK-NEXT:    [[Z:%.*]] = icmp ne i32 [[TMP1]], 0
@@ -605,7 +605,7 @@ define <2 x i32> @shl_nuw_nsw_splat_vec(<2 x i8> %x) {
   ret <2 x i32> %t3
 }
 
-define i32 @test38(i32 %x) nounwind readnone {
+define i32 @test38(i32 %x) {
 ; CHECK-LABEL: @test38(
 ; CHECK-NEXT:    [[REM1:%.*]] = and i32 [[X:%.*]], 31
 ; CHECK-NEXT:    [[SHL:%.*]] = shl nuw i32 1, [[REM1]]
@@ -616,7 +616,7 @@ define i32 @test38(i32 %x) nounwind readnone {
   ret i32 %shl
 }
 
-define <2 x i32> @test38_uniform(<2 x i32> %x) nounwind readnone {
+define <2 x i32> @test38_uniform(<2 x i32> %x) {
 ; CHECK-LABEL: @test38_uniform(
 ; CHECK-NEXT:    [[REM1:%.*]] = and <2 x i32> [[X:%.*]], <i32 31, i32 31>
 ; CHECK-NEXT:    [[SHL:%.*]] = shl nuw <2 x i32> <i32 1, i32 1>, [[REM1]]
@@ -627,7 +627,7 @@ define <2 x i32> @test38_uniform(<2 x i32> %x) nounwind readnone {
   ret <2 x i32> %shl
 }
 
-define <3 x i32> @test38_nonuniform(<3 x i32> %x) nounwind readnone {
+define <3 x i32> @test38_nonuniform(<3 x i32> %x) {
 ; CHECK-LABEL: @test38_nonuniform(
 ; CHECK-NEXT:    [[REM1:%.*]] = and <3 x i32> [[X:%.*]], <i32 31, i32 15, i32 0>
 ; CHECK-NEXT:    [[SHL:%.*]] = shl nuw <3 x i32> <i32 1, i32 1, i32 1>, [[REM1]]
@@ -638,7 +638,7 @@ define <3 x i32> @test38_nonuniform(<3 x i32> %x) nounwind readnone {
   ret <3 x i32> %shl
 }
 
-define <2 x i32> @test38_poison(<2 x i32> %x) nounwind readnone {
+define <2 x i32> @test38_poison(<2 x i32> %x) {
 ; CHECK-LABEL: @test38_poison(
 ; CHECK-NEXT:    ret <2 x i32> poison
 ;
@@ -658,8 +658,8 @@ define i8 @test39(i32 %a0) {
 ; CHECK-NEXT:    [[I51:%.*]] = xor i8 [[I50]], [[I5]]
 ; CHECK-NEXT:    [[TMP0:%.*]] = lshr exact i8 [[I5]], 3
 ; CHECK-NEXT:    [[I54:%.*]] = and i8 [[TMP0]], 16
-; CHECK-NEXT:    [[I551:%.*]] = or disjoint i8 [[I54]], [[I51]]
-; CHECK-NEXT:    ret i8 [[I551]]
+; CHECK-NEXT:    [[I55:%.*]] = or disjoint i8 [[I54]], [[I51]]
+; CHECK-NEXT:    ret i8 [[I55]]
 ;
 entry:
   %i4 = trunc i32 %a0 to i8
@@ -675,7 +675,7 @@ entry:
   ret i8 %i55
 }
 
-define i32 @test42(i32 %a, i32 %b) nounwind {
+define i32 @test42(i32 %a, i32 %b) {
 ; CHECK-LABEL: @test42(
 ; CHECK-NEXT:    [[DIV:%.*]] = lshr exact i32 4096, [[B:%.*]]
 ; CHECK-NEXT:    [[DIV2:%.*]] = udiv i32 [[A:%.*]], [[DIV]]
@@ -697,7 +697,7 @@ define <2 x i32> @test42vec(<2 x i32> %a, <2 x i32> %b) {
   ret <2 x i32> %div2
 }
 
-define i32 @test43(i32 %a, i32 %b) nounwind {
+define i32 @test43(i32 %a, i32 %b) {
 ; CHECK-LABEL: @test43(
 ; CHECK-NEXT:    [[TMP1:%.*]] = add i32 [[B:%.*]], 12
 ; CHECK-NEXT:    [[DIV21:%.*]] = lshr i32 [[A:%.*]], [[TMP1]]
@@ -708,7 +708,7 @@ define i32 @test43(i32 %a, i32 %b) nounwind {
   ret i32 %div2
 }
 
-define i32 @test44(i32 %a) nounwind {
+define i32 @test44(i32 %a) {
 ; CHECK-LABEL: @test44(
 ; CHECK-NEXT:    [[Y:%.*]] = shl i32 [[A:%.*]], 5
 ; CHECK-NEXT:    ret i32 [[Y]]
@@ -718,7 +718,20 @@ define i32 @test44(i32 %a) nounwind {
   ret i32 %z
 }
 
-define i32 @test45(i32 %a) nounwind {
+define i32 @test44_multiuse(i32 %a) {
+; CHECK-LABEL: @test44_multiuse(
+; CHECK-NEXT:    [[Y:%.*]] = shl nuw i32 [[A:%.*]], 1
+; CHECK-NEXT:    call void @use_i32(i32 [[Y]])
+; CHECK-NEXT:    [[Z:%.*]] = shl i32 [[A]], 5
+; CHECK-NEXT:    ret i32 [[Z]]
+;
+  %y = shl nuw i32 %a, 1
+  call void @use_i32(i32 %y)
+  %z = shl i32 %y, 4
+  ret i32 %z
+}
+
+define i32 @test45(i32 %a) {
 ; CHECK-LABEL: @test45(
 ; CHECK-NEXT:    [[Y:%.*]] = lshr i32 [[A:%.*]], 5
 ; CHECK-NEXT:    ret i32 [[Y]]
@@ -728,6 +741,19 @@ define i32 @test45(i32 %a) nounwind {
   ret i32 %z
 }
 
+define i32 @test45_multiuse(i32 %a) {
+; CHECK-LABEL: @test45_multiuse(
+; CHECK-NEXT:    [[Y:%.*]] = lshr exact i32 [[A:%.*]], 1
+; CHECK-NEXT:    call void @use_i32(i32 [[Y]])
+; CHECK-NEXT:    [[Z:%.*]] = lshr i32 [[A]], 5
+; CHECK-NEXT:    ret i32 [[Z]]
+;
+  %y = lshr exact i32 %a, 1
+  call void @use_i32(i32 %y)
+  %z = lshr i32 %y, 4
+  ret i32 %z
+}
+
 ; (X >>?exact C1) << C2 --> X >>?exact (C1-C2)
 
 define i32 @test46(i32 %a) {
@@ -1751,14 +1777,14 @@ define void @ashr_out_of_range_1(ptr %A) {
 ; CHECK-NEXT:    [[L:%.*]] = load i177, ptr [[A:%.*]], align 4
 ; CHECK-NEXT:    [[L_FROZEN:%.*]] = freeze i177 [[L]]
 ; CHECK-NEXT:    [[TMP1:%.*]] = icmp eq i177 [[L_FROZEN]], -1
-; CHECK-NEXT:    [[TMP6:%.*]] = trunc i177 [[L_FROZEN]] to i64
-; CHECK-NEXT:    [[TMP2:%.*]] = select i1 [[TMP1]], i64 0, i64 [[TMP6]]
-; CHECK-NEXT:    [[TMP3:%.*]] = getelementptr i177, ptr [[A]], i64 [[TMP2]]
-; CHECK-NEXT:    [[G11:%.*]] = getelementptr i8, ptr [[TMP3]], i64 -24
-; CHECK-NEXT:    [[TMP4:%.*]] = sext i1 [[TMP1]] to i64
-; CHECK-NEXT:    [[G62:%.*]] = getelementptr i177, ptr [[G11]], i64 [[TMP4]]
-; CHECK-NEXT:    [[TMP5:%.*]] = icmp eq i177 [[L_FROZEN]], -1
-; CHECK-NEXT:    [[B28:%.*]] = select i1 [[TMP5]], i177 0, i177 [[L_FROZEN]]
+; CHECK-NEXT:    [[TMP2:%.*]] = trunc i177 [[L_FROZEN]] to i64
+; CHECK-NEXT:    [[TMP3:%.*]] = select i1 [[TMP1]], i64 0, i64 [[TMP2]]
+; CHECK-NEXT:    [[TMP4:%.*]] = getelementptr i177, ptr [[A]], i64 [[TMP3]]
+; CHECK-NEXT:    [[G11:%.*]] = getelementptr i8, ptr [[TMP4]], i64 -24
+; CHECK-NEXT:    [[TMP5:%.*]] = sext i1 [[TMP1]] to i64
+; CHECK-NEXT:    [[G62:%.*]] = getelementptr i177, ptr [[G11]], i64 [[TMP5]]
+; CHECK-NEXT:    [[TMP6:%.*]] = icmp eq i177 [[L_FROZEN]], -1
+; CHECK-NEXT:    [[B28:%.*]] = select i1 [[TMP6]], i177 0, i177 [[L_FROZEN]]
 ; CHECK-NEXT:    store i177 [[B28]], ptr [[G62]], align 4
 ; CHECK-NEXT:    ret void
 ;
@@ -2055,7 +2081,7 @@ define <2 x i8> @shl1_cttz_vec_poison(<2 x i8> %x) {
 
 define i32 @shl1_cttz_extra_use(i32 %x) {
 ; CHECK-LABEL: @shl1_cttz_extra_use(
-; CHECK-NEXT:    [[TZ:%.*]] = call i32 @llvm.cttz.i32(i32 [[X:%.*]], i1 false), !range [[RNG0:![0-9]+]]
+; CHECK-NEXT:    [[TZ:%.*]] = call range(i32 0, 33) i32 @llvm.cttz.i32(i32 [[X:%.*]], i1 false)
 ; CHECK-NEXT:    call void @use_i32(i32 [[TZ]])
 ; CHECK-NEXT:    [[SHL:%.*]] = shl nuw i32 1, [[TZ]]
 ; CHECK-NEXT:    ret i32 [[SHL]]
@@ -2070,7 +2096,7 @@ define i32 @shl1_cttz_extra_use(i32 %x) {
 
 define i32 @shl2_cttz(i32 %x) {
 ; CHECK-LABEL: @shl2_cttz(
-; CHECK-NEXT:    [[TZ:%.*]] = call i32 @llvm.cttz.i32(i32 [[X:%.*]], i1 true), !range [[RNG0]]
+; CHECK-NEXT:    [[TZ:%.*]] = call range(i32 0, 33) i32 @llvm.cttz.i32(i32 [[X:%.*]], i1 true)
 ; CHECK-NEXT:    [[SHL:%.*]] = shl i32 2, [[TZ]]
 ; CHECK-NEXT:    ret i32 [[SHL]]
 ;
diff --git a/llvm/test/Transforms/InstCombine/simplify-demanded-fpclass.ll b/llvm/test/Transforms/InstCombine/simplify-demanded-fpclass.ll
index a7d01b4f824d..e4fb7764ba9e 100644
--- a/llvm/test/Transforms/InstCombine/simplify-demanded-fpclass.ll
+++ b/llvm/test/Transforms/InstCombine/simplify-demanded-fpclass.ll
@@ -364,8 +364,8 @@ define nofpclass(inf) float @ret_nofpclass_inf__select_chain_inf_nan_1(i1 %cond,
 define nofpclass(inf) float @ret_nofpclass_inf__fabs_select_ninf_rhs(i1 %cond, float %x) {
 ; CHECK-LABEL: define nofpclass(inf) float @ret_nofpclass_inf__fabs_select_ninf_rhs
 ; CHECK-SAME: (i1 [[COND:%.*]], float [[X:%.*]]) {
-; CHECK-NEXT:    [[FABS:%.*]] = call float @llvm.fabs.f32(float [[X]])
-; CHECK-NEXT:    ret float [[FABS]]
+; CHECK-NEXT:    [[TMP1:%.*]] = call float @llvm.fabs.f32(float [[X]])
+; CHECK-NEXT:    ret float [[TMP1]]
 ;
   %select = select i1 %cond, float %x, float 0xFFF0000000000000
   %fabs = call float @llvm.fabs.f32(float %select)
@@ -376,8 +376,8 @@ define nofpclass(inf) float @ret_nofpclass_inf__fabs_select_ninf_rhs(i1 %cond, f
 define nofpclass(inf) float @ret_nofpclass_inf__fabs_select_pinf_rhs(i1 %cond, float %x) {
 ; CHECK-LABEL: define nofpclass(inf) float @ret_nofpclass_inf__fabs_select_pinf_rhs
 ; CHECK-SAME: (i1 [[COND:%.*]], float [[X:%.*]]) {
-; CHECK-NEXT:    [[FABS:%.*]] = call float @llvm.fabs.f32(float [[X]])
-; CHECK-NEXT:    ret float [[FABS]]
+; CHECK-NEXT:    [[TMP1:%.*]] = call float @llvm.fabs.f32(float [[X]])
+; CHECK-NEXT:    ret float [[TMP1]]
 ;
   %select = select i1 %cond, float %x, float 0x7FF0000000000000
   %fabs = call float @llvm.fabs.f32(float %select)
@@ -400,8 +400,8 @@ define nofpclass(ninf nnorm nsub nzero) float @ret_nofpclass_no_negatives__fabs_
 define nofpclass(pinf pnorm psub pzero) float @ret_nofpclass_no_positives__fabs_select_pinf_rhs(i1 %cond, float %x) {
 ; CHECK-LABEL: define nofpclass(pinf pzero psub pnorm) float @ret_nofpclass_no_positives__fabs_select_pinf_rhs
 ; CHECK-SAME: (i1 [[COND:%.*]], float [[X:%.*]]) {
-; CHECK-NEXT:    [[FABS:%.*]] = call float @llvm.fabs.f32(float [[X]])
-; CHECK-NEXT:    ret float [[FABS]]
+; CHECK-NEXT:    [[TMP1:%.*]] = call float @llvm.fabs.f32(float [[X]])
+; CHECK-NEXT:    ret float [[TMP1]]
 ;
   %select = select i1 %cond, float %x, float 0x7FF0000000000000
   %fabs = call float @llvm.fabs.f32(float %select)
@@ -435,8 +435,8 @@ define nofpclass(nan pinf pnorm psub pzero) float @ret_nofpclass_no_positives_na
 define nofpclass(inf) float @ret_nofpclass_inf__fneg_select_ninf_rhs(i1 %cond, float %x) {
 ; CHECK-LABEL: define nofpclass(inf) float @ret_nofpclass_inf__fneg_select_ninf_rhs
 ; CHECK-SAME: (i1 [[COND:%.*]], float [[X:%.*]]) {
-; CHECK-NEXT:    [[FNEG:%.*]] = fneg float [[X]]
-; CHECK-NEXT:    ret float [[FNEG]]
+; CHECK-NEXT:    [[X_NEG:%.*]] = fneg float [[X]]
+; CHECK-NEXT:    ret float [[X_NEG]]
 ;
   %select = select i1 %cond, float %x, float 0xFFF0000000000000
   %fneg = fneg float %select
@@ -447,8 +447,8 @@ define nofpclass(inf) float @ret_nofpclass_inf__fneg_select_ninf_rhs(i1 %cond, f
 define nofpclass(inf nnorm nsub nzero) float @ret_nofpclass_nonegatives_noinf___fneg_select_pinf_rhs(i1 %cond, float %x) {
 ; CHECK-LABEL: define nofpclass(inf nzero nsub nnorm) float @ret_nofpclass_nonegatives_noinf___fneg_select_pinf_rhs
 ; CHECK-SAME: (i1 [[COND:%.*]], float [[X:%.*]]) {
-; CHECK-NEXT:    [[FNEG:%.*]] = fneg float [[X]]
-; CHECK-NEXT:    ret float [[FNEG]]
+; CHECK-NEXT:    [[X_NEG:%.*]] = fneg float [[X]]
+; CHECK-NEXT:    ret float [[X_NEG]]
 ;
   %select = select i1 %cond, float %x, float 0x7FF0000000000000
   %fneg = fneg float %select
@@ -459,8 +459,8 @@ define nofpclass(inf nnorm nsub nzero) float @ret_nofpclass_nonegatives_noinf___
 define nofpclass(inf nnorm nsub nzero) float @ret_nofpclass_nonegatives_noinf___fneg_select_ninf_lhs(i1 %cond, float %x) {
 ; CHECK-LABEL: define nofpclass(inf nzero nsub nnorm) float @ret_nofpclass_nonegatives_noinf___fneg_select_ninf_lhs
 ; CHECK-SAME: (i1 [[COND:%.*]], float [[X:%.*]]) {
-; CHECK-NEXT:    [[FNEG:%.*]] = fneg float [[X]]
-; CHECK-NEXT:    ret float [[FNEG]]
+; CHECK-NEXT:    [[X_NEG:%.*]] = fneg float [[X]]
+; CHECK-NEXT:    ret float [[X_NEG]]
 ;
   %select = select i1 %cond, float 0xFFF0000000000000, float %x
   %fneg = fneg float %select
@@ -470,8 +470,8 @@ define nofpclass(inf nnorm nsub nzero) float @ret_nofpclass_nonegatives_noinf___
 define nofpclass(pzero psub pnorm pinf) float @ret_nofpclass_nopositives___fneg_select_pinf_rhs(i1 %cond, float %x) {
 ; CHECK-LABEL: define nofpclass(pinf pzero psub pnorm) float @ret_nofpclass_nopositives___fneg_select_pinf_rhs
 ; CHECK-SAME: (i1 [[COND:%.*]], float [[X:%.*]]) {
-; CHECK-NEXT:    [[SELECT:%.*]] = select i1 [[COND]], float [[X]], float 0x7FF0000000000000
-; CHECK-NEXT:    [[FNEG:%.*]] = fneg float [[SELECT]]
+; CHECK-NEXT:    [[X_NEG:%.*]] = fneg float [[X]]
+; CHECK-NEXT:    [[FNEG:%.*]] = select i1 [[COND]], float [[X_NEG]], float 0xFFF0000000000000
 ; CHECK-NEXT:    ret float [[FNEG]]
 ;
   %select = select i1 %cond, float %x, float 0x7FF0000000000000
@@ -483,9 +483,9 @@ define nofpclass(pzero psub pnorm pinf) float @ret_nofpclass_nopositives___fneg_
 define nofpclass(inf) float @ret_nofpclass_inf__fneg_fabs_select_pinf_rhs(i1 %cond, float %x) {
 ; CHECK-LABEL: define nofpclass(inf) float @ret_nofpclass_inf__fneg_fabs_select_pinf_rhs
 ; CHECK-SAME: (i1 [[COND:%.*]], float [[X:%.*]]) {
-; CHECK-NEXT:    [[FABS:%.*]] = call float @llvm.fabs.f32(float [[X]])
-; CHECK-NEXT:    [[FNEG:%.*]] = fneg float [[FABS]]
-; CHECK-NEXT:    ret float [[FNEG]]
+; CHECK-NEXT:    [[TMP1:%.*]] = call float @llvm.fabs.f32(float [[X]])
+; CHECK-NEXT:    [[DOTNEG:%.*]] = fneg float [[TMP1]]
+; CHECK-NEXT:    ret float [[DOTNEG]]
 ;
   %select = select i1 %cond, float %x, float 0x7FF0000000000000
   %fabs = call float @llvm.fabs.f32(float %select)
@@ -497,9 +497,9 @@ define nofpclass(inf) float @ret_nofpclass_inf__fneg_fabs_select_pinf_rhs(i1 %co
 define nofpclass(ninf nnorm nsub nzero) float @ret_nofpclass_nonegatives__fneg_fabs_select_pinf_rhs(i1 %cond, float %x) {
 ; CHECK-LABEL: define nofpclass(ninf nzero nsub nnorm) float @ret_nofpclass_nonegatives__fneg_fabs_select_pinf_rhs
 ; CHECK-SAME: (i1 [[COND:%.*]], float [[X:%.*]]) {
-; CHECK-NEXT:    [[FABS:%.*]] = call float @llvm.fabs.f32(float [[X]])
-; CHECK-NEXT:    [[FNEG:%.*]] = fneg float [[FABS]]
-; CHECK-NEXT:    ret float [[FNEG]]
+; CHECK-NEXT:    [[TMP1:%.*]] = call float @llvm.fabs.f32(float [[X]])
+; CHECK-NEXT:    [[DOTNEG:%.*]] = fneg float [[TMP1]]
+; CHECK-NEXT:    ret float [[DOTNEG]]
 ;
   %select = select i1 %cond, float %x, float 0x7FF0000000000000
   %fabs = call float @llvm.fabs.f32(float %select)
@@ -535,8 +535,8 @@ define nofpclass(inf) float @ret_nofpclass_inf__copysign_unknown_select_pinf_rhs
 define nofpclass(inf) float @ret_nofpclass_inf__copysign_positive_select_pinf_rhs(i1 %cond, float %x) {
 ; CHECK-LABEL: define nofpclass(inf) float @ret_nofpclass_inf__copysign_positive_select_pinf_rhs
 ; CHECK-SAME: (i1 [[COND:%.*]], float [[X:%.*]]) {
-; CHECK-NEXT:    [[COPYSIGN:%.*]] = call float @llvm.fabs.f32(float [[X]])
-; CHECK-NEXT:    ret float [[COPYSIGN]]
+; CHECK-NEXT:    [[TMP1:%.*]] = call float @llvm.fabs.f32(float [[X]])
+; CHECK-NEXT:    ret float [[TMP1]]
 ;
   %select = select i1 %cond, float %x, float 0x7FF0000000000000
   %copysign = call float @llvm.copysign.f32(float %select, float 1.0)
@@ -547,8 +547,8 @@ define nofpclass(inf) float @ret_nofpclass_inf__copysign_negative_select_pinf_rh
 ; CHECK-LABEL: define nofpclass(inf) float @ret_nofpclass_inf__copysign_negative_select_pinf_rhs
 ; CHECK-SAME: (i1 [[COND:%.*]], float [[X:%.*]]) {
 ; CHECK-NEXT:    [[TMP1:%.*]] = call float @llvm.fabs.f32(float [[X]])
-; CHECK-NEXT:    [[COPYSIGN:%.*]] = fneg float [[TMP1]]
-; CHECK-NEXT:    ret float [[COPYSIGN]]
+; CHECK-NEXT:    [[DOTNEG:%.*]] = fneg float [[TMP1]]
+; CHECK-NEXT:    ret float [[DOTNEG]]
 ;
   %select = select i1 %cond, float %x, float 0x7FF0000000000000
   %copysign = call float @llvm.copysign.f32(float %select, float -1.0)
@@ -627,8 +627,8 @@ define nofpclass(nan ninf nnorm nsub nzero) float @ret_nofpclass_nonegatives_non
 define nofpclass(pinf pnorm psub pzero) float @ret_nofpclass_nopositives__copysign_fabs_select_pinf_rhs(i1 %cond, float %x, float %sign) {
 ; CHECK-LABEL: define nofpclass(pinf pzero psub pnorm) float @ret_nofpclass_nopositives__copysign_fabs_select_pinf_rhs
 ; CHECK-SAME: (i1 [[COND:%.*]], float [[X:%.*]], float [[SIGN:%.*]]) {
-; CHECK-NEXT:    [[COPYSIGN:%.*]] = call float @llvm.fabs.f32(float [[X]])
-; CHECK-NEXT:    ret float [[COPYSIGN]]
+; CHECK-NEXT:    [[TMP1:%.*]] = call float @llvm.fabs.f32(float [[X]])
+; CHECK-NEXT:    ret float [[TMP1]]
 ;
   %select = select i1 %cond, float %x, float 0x7FF0000000000000
   %fabs.sign = call float @llvm.fabs.f32(float %sign)
@@ -678,9 +678,9 @@ define nofpclass(ninf nnorm nsub nzero) float @ret_nofpclass_no_negatives__copys
 define nofpclass(pinf pnorm psub pzero) float @ret_nofpclass_no_positives__copysign_unknown_select_pinf_rhs(i1 %cond, float %x, float %unknown.sign) {
 ; CHECK-LABEL: define nofpclass(pinf pzero psub pnorm) float @ret_nofpclass_no_positives__copysign_unknown_select_pinf_rhs
 ; CHECK-SAME: (i1 [[COND:%.*]], float [[X:%.*]], float [[UNKNOWN_SIGN:%.*]]) {
-; CHECK-NEXT:    [[TMP2:%.*]] = call float @llvm.fabs.f32(float [[X]])
-; CHECK-NEXT:    [[TMP1:%.*]] = select i1 [[COND]], float [[TMP2]], float 0x7FF0000000000000
-; CHECK-NEXT:    [[COPYSIGN:%.*]] = fneg float [[TMP1]]
+; CHECK-NEXT:    [[TMP1:%.*]] = call float @llvm.fabs.f32(float [[X]])
+; CHECK-NEXT:    [[DOTNEG:%.*]] = fneg float [[TMP1]]
+; CHECK-NEXT:    [[COPYSIGN:%.*]] = select i1 [[COND]], float [[DOTNEG]], float 0xFFF0000000000000
 ; CHECK-NEXT:    ret float [[COPYSIGN]]
 ;
   %select = select i1 %cond, float %x, float 0x7FF0000000000000
@@ -705,9 +705,9 @@ define nofpclass(nan ninf nnorm nsub nzero) float @ret_nofpclass_no_negatives_no
 define nofpclass(nan pinf pnorm psub pzero) float @ret_nofpclass_no_positives_nonan__copysign_unknown_select_pinf_rhs(i1 %cond, float %x, float %unknown.sign) {
 ; CHECK-LABEL: define nofpclass(nan pinf pzero psub pnorm) float @ret_nofpclass_no_positives_nonan__copysign_unknown_select_pinf_rhs
 ; CHECK-SAME: (i1 [[COND:%.*]], float [[X:%.*]], float [[UNKNOWN_SIGN:%.*]]) {
-; CHECK-NEXT:    [[TMP2:%.*]] = call float @llvm.fabs.f32(float [[X]])
-; CHECK-NEXT:    [[TMP1:%.*]] = select i1 [[COND]], float [[TMP2]], float 0x7FF0000000000000
-; CHECK-NEXT:    [[COPYSIGN:%.*]] = fneg float [[TMP1]]
+; CHECK-NEXT:    [[TMP1:%.*]] = call float @llvm.fabs.f32(float [[X]])
+; CHECK-NEXT:    [[DOTNEG:%.*]] = fneg float [[TMP1]]
+; CHECK-NEXT:    [[COPYSIGN:%.*]] = select i1 [[COND]], float [[DOTNEG]], float 0xFFF0000000000000
 ; CHECK-NEXT:    ret float [[COPYSIGN]]
 ;
   %select = select i1 %cond, float %x, float 0x7FF0000000000000
diff --git a/llvm/test/Transforms/InstCombine/sub-of-negatible.ll b/llvm/test/Transforms/InstCombine/sub-of-negatible.ll
index 72fd7f7be2b0..b2e14ceaca1b 100644
--- a/llvm/test/Transforms/InstCombine/sub-of-negatible.ll
+++ b/llvm/test/Transforms/InstCombine/sub-of-negatible.ll
@@ -1385,12 +1385,12 @@ define i8 @dont_negate_ordinary_select(i8 %x, i8 %y, i8 %z, i1 %c) {
   ret i8 %t1
 }
 
-; FIXME: This is a miscompile.
 define <2 x i32> @negate_select_of_negation_poison(<2 x i1> %c, <2 x i32> %x) {
 ; CHECK-LABEL: @negate_select_of_negation_poison(
 ; CHECK-NEXT:    [[NEG:%.*]] = sub <2 x i32> <i32 0, i32 poison>, [[X:%.*]]
-; CHECK-NEXT:    [[TMP1:%.*]] = select <2 x i1> [[C:%.*]], <2 x i32> [[X]], <2 x i32> [[NEG]]
-; CHECK-NEXT:    ret <2 x i32> [[TMP1]]
+; CHECK-NEXT:    [[SEL:%.*]] = select <2 x i1> [[C:%.*]], <2 x i32> [[NEG]], <2 x i32> [[X]]
+; CHECK-NEXT:    [[NEG2:%.*]] = sub <2 x i32> zeroinitializer, [[SEL]]
+; CHECK-NEXT:    ret <2 x i32> [[NEG2]]
 ;
   %neg = sub <2 x i32> <i32 0, i32 poison>, %x
   %sel = select <2 x i1> %c, <2 x i32> %neg, <2 x i32> %x
diff --git a/llvm/test/Transforms/InstCombine/sub-xor.ll b/llvm/test/Transforms/InstCombine/sub-xor.ll
index 2976598e043f..b4add9698b16 100644
--- a/llvm/test/Transforms/InstCombine/sub-xor.ll
+++ b/llvm/test/Transforms/InstCombine/sub-xor.ll
@@ -97,7 +97,7 @@ declare i32 @llvm.ctlz.i32(i32, i1)
 
 define i32 @range_masked_sub(i32 %x) {
 ; CHECK-LABEL: @range_masked_sub(
-; CHECK-NEXT:    [[COUNT:%.*]] = tail call i32 @llvm.ctlz.i32(i32 [[X:%.*]], i1 true) #[[ATTR1:[0-9]+]], !range [[RNG0:![0-9]+]]
+; CHECK-NEXT:    [[COUNT:%.*]] = tail call range(i32 0, 33) i32 @llvm.ctlz.i32(i32 [[X:%.*]], i1 true) #[[ATTR1:[0-9]+]]
 ; CHECK-NEXT:    [[SUB:%.*]] = xor i32 [[COUNT]], 31
 ; CHECK-NEXT:    ret i32 [[SUB]]
 ;
diff --git a/llvm/test/Transforms/InstCombine/sub.ll b/llvm/test/Transforms/InstCombine/sub.ll
index a84e389f13c3..32ed4a787e92 100644
--- a/llvm/test/Transforms/InstCombine/sub.ll
+++ b/llvm/test/Transforms/InstCombine/sub.ll
@@ -1123,7 +1123,8 @@ define i64 @test58(ptr %foo, i64 %i, i64 %j) {
 
 define i64 @test59(ptr %foo, i64 %i) {
 ; CHECK-LABEL: @test59(
-; CHECK-NEXT:    [[GEP1:%.*]] = getelementptr inbounds [100 x [100 x i8]], ptr [[FOO:%.*]], i64 0, i64 42, i64 [[I:%.*]]
+; CHECK-NEXT:    [[TMP1:%.*]] = getelementptr i8, ptr [[FOO:%.*]], i64 [[I:%.*]]
+; CHECK-NEXT:    [[GEP1:%.*]] = getelementptr i8, ptr [[TMP1]], i64 4200
 ; CHECK-NEXT:    [[GEP2:%.*]] = getelementptr inbounds i8, ptr [[FOO]], i64 4200
 ; CHECK-NEXT:    store ptr [[GEP1]], ptr @dummy_global1, align 8
 ; CHECK-NEXT:    store ptr [[GEP2]], ptr @dummy_global2, align 8
@@ -1142,13 +1143,12 @@ define i64 @test59(ptr %foo, i64 %i) {
 
 define i64 @test60(ptr %foo, i64 %i, i64 %j) {
 ; CHECK-LABEL: @test60(
-; CHECK-NEXT:    [[GEP1:%.*]] = getelementptr inbounds [100 x [100 x i8]], ptr [[FOO:%.*]], i64 0, i64 [[J:%.*]], i64 [[I:%.*]]
-; CHECK-NEXT:    [[GEP2:%.*]] = getelementptr inbounds i8, ptr [[FOO]], i64 4200
-; CHECK-NEXT:    [[CAST1:%.*]] = ptrtoint ptr [[GEP1]] to i64
-; CHECK-NEXT:    [[CAST2:%.*]] = ptrtoint ptr [[GEP2]] to i64
-; CHECK-NEXT:    [[SUB:%.*]] = sub i64 [[CAST1]], [[CAST2]]
+; CHECK-NEXT:    [[GEP1_IDX:%.*]] = mul nsw i64 [[J:%.*]], 100
+; CHECK-NEXT:    [[GEP1_OFFS:%.*]] = add nsw i64 [[GEP1_IDX]], [[I:%.*]]
+; CHECK-NEXT:    [[GEP1:%.*]] = getelementptr inbounds i8, ptr [[FOO:%.*]], i64 [[GEP1_OFFS]]
+; CHECK-NEXT:    [[GEPDIFF:%.*]] = add nsw i64 [[GEP1_OFFS]], -4200
 ; CHECK-NEXT:    store ptr [[GEP1]], ptr @dummy_global1, align 8
-; CHECK-NEXT:    ret i64 [[SUB]]
+; CHECK-NEXT:    ret i64 [[GEPDIFF]]
 ;
 ; gep1 has a non-constant index and more than one uses. Shouldn't duplicate the arithmetic.
   %gep1 = getelementptr inbounds [100 x [100 x i8]], ptr %foo, i64 0, i64 %j, i64 %i
@@ -1162,13 +1162,12 @@ define i64 @test60(ptr %foo, i64 %i, i64 %j) {
 
 define i64 @test61(ptr %foo, i64 %i, i64 %j) {
 ; CHECK-LABEL: @test61(
-; CHECK-NEXT:    [[GEP1:%.*]] = getelementptr inbounds i8, ptr [[FOO:%.*]], i64 4200
-; CHECK-NEXT:    [[GEP2:%.*]] = getelementptr inbounds [100 x [100 x i8]], ptr [[FOO]], i64 0, i64 [[J:%.*]], i64 [[I:%.*]]
-; CHECK-NEXT:    [[CAST1:%.*]] = ptrtoint ptr [[GEP1]] to i64
-; CHECK-NEXT:    [[CAST2:%.*]] = ptrtoint ptr [[GEP2]] to i64
-; CHECK-NEXT:    [[SUB:%.*]] = sub i64 [[CAST1]], [[CAST2]]
+; CHECK-NEXT:    [[GEP2_IDX:%.*]] = mul nsw i64 [[J:%.*]], 100
+; CHECK-NEXT:    [[GEP2_OFFS:%.*]] = add nsw i64 [[GEP2_IDX]], [[I:%.*]]
+; CHECK-NEXT:    [[GEP2:%.*]] = getelementptr inbounds i8, ptr [[FOO:%.*]], i64 [[GEP2_OFFS]]
+; CHECK-NEXT:    [[GEPDIFF:%.*]] = sub nsw i64 4200, [[GEP2_OFFS]]
 ; CHECK-NEXT:    store ptr [[GEP2]], ptr @dummy_global2, align 8
-; CHECK-NEXT:    ret i64 [[SUB]]
+; CHECK-NEXT:    ret i64 [[GEPDIFF]]
 ;
 ; gep2 has a non-constant index and more than one uses. Shouldn't duplicate the arithmetic.
   %gep1 = getelementptr inbounds [100 x [100 x i8]], ptr %foo, i64 0, i64 42, i64 0
@@ -1180,6 +1179,24 @@ define i64 @test61(ptr %foo, i64 %i, i64 %j) {
   ret i64 %sub
 }
 
+declare void @use.ptr(ptr)
+
+define i64 @test_sub_ptradd_multiuse(ptr %p, i64 %idx1, i64 %idx2) {
+; CHECK-LABEL: @test_sub_ptradd_multiuse(
+; CHECK-NEXT:    [[P1:%.*]] = getelementptr inbounds i8, ptr [[P:%.*]], i64 [[IDX1:%.*]]
+; CHECK-NEXT:    call void @use.ptr(ptr [[P1]])
+; CHECK-NEXT:    [[GEPDIFF:%.*]] = sub nsw i64 [[IDX1]], [[IDX2:%.*]]
+; CHECK-NEXT:    ret i64 [[GEPDIFF]]
+;
+  %p1 = getelementptr inbounds i8, ptr %p, i64 %idx1
+  call void @use.ptr(ptr %p1)
+  %p2 = getelementptr inbounds i8, ptr %p, i64 %idx2
+  %p1.int = ptrtoint ptr %p1 to i64
+  %p2.int = ptrtoint ptr %p2 to i64
+  %sub = sub i64 %p1.int, %p2.int
+  ret i64 %sub
+}
+
 define i32 @test62(i32 %A) {
 ; CHECK-LABEL: @test62(
 ; CHECK-NEXT:    [[TMP1:%.*]] = shl i32 [[A:%.*]], 1
diff --git a/llvm/test/Transforms/InstCombine/vector-reverse.ll b/llvm/test/Transforms/InstCombine/vector-reverse.ll
index 5e6672658f9a..a1a6ee949a13 100644
--- a/llvm/test/Transforms/InstCombine/vector-reverse.ll
+++ b/llvm/test/Transforms/InstCombine/vector-reverse.ll
@@ -8,11 +8,11 @@
 define <vscale x 4 x i32> @binop_reverse(<vscale x 4 x i32> %a, <vscale x 4 x i32> %b) {
 ; CHECK-LABEL: @binop_reverse(
 ; CHECK-NEXT:    [[ADD1:%.*]] = add nsw <vscale x 4 x i32> [[A:%.*]], [[B:%.*]]
-; CHECK-NEXT:    [[ADD:%.*]] = call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> [[ADD1]])
+; CHECK-NEXT:    [[ADD:%.*]] = call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> [[ADD1]])
 ; CHECK-NEXT:    ret <vscale x 4 x i32> [[ADD]]
 ;
-  %a.rev = tail call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> %a)
-  %b.rev = tail call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> %b)
+  %a.rev = tail call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> %a)
+  %b.rev = tail call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> %b)
   %add = add nsw <vscale x 4 x i32> %a.rev, %b.rev
   ret <vscale x 4 x i32> %add
 }
@@ -20,14 +20,14 @@ define <vscale x 4 x i32> @binop_reverse(<vscale x 4 x i32> %a, <vscale x 4 x i3
 ; %a.rev has multiple uses
 define <vscale x 4 x i32> @binop_reverse_1(<vscale x 4 x i32> %a, <vscale x 4 x i32> %b) {
 ; CHECK-LABEL: @binop_reverse_1(
-; CHECK-NEXT:    [[A_REV:%.*]] = tail call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> [[A:%.*]])
+; CHECK-NEXT:    [[A_REV:%.*]] = tail call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> [[A:%.*]])
 ; CHECK-NEXT:    call void @use_nxv4i32(<vscale x 4 x i32> [[A_REV]])
 ; CHECK-NEXT:    [[ADD1:%.*]] = add <vscale x 4 x i32> [[A]], [[B:%.*]]
-; CHECK-NEXT:    [[ADD:%.*]] = call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> [[ADD1]])
+; CHECK-NEXT:    [[ADD:%.*]] = call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> [[ADD1]])
 ; CHECK-NEXT:    ret <vscale x 4 x i32> [[ADD]]
 ;
-  %a.rev = tail call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> %a)
-  %b.rev = tail call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> %b)
+  %a.rev = tail call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> %a)
+  %b.rev = tail call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> %b)
   call void @use_nxv4i32(<vscale x 4 x i32>  %a.rev)
   %add = add <vscale x 4 x i32> %a.rev, %b.rev
   ret <vscale x 4 x i32> %add
@@ -36,14 +36,14 @@ define <vscale x 4 x i32> @binop_reverse_1(<vscale x 4 x i32> %a, <vscale x 4 x
 ; %b.rev has multiple uses
 define <vscale x 4 x i32> @binop_reverse_2(<vscale x 4 x i32> %a, <vscale x 4 x i32> %b) {
 ; CHECK-LABEL: @binop_reverse_2(
-; CHECK-NEXT:    [[B_REV:%.*]] = tail call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> [[B:%.*]])
+; CHECK-NEXT:    [[B_REV:%.*]] = tail call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> [[B:%.*]])
 ; CHECK-NEXT:    call void @use_nxv4i32(<vscale x 4 x i32> [[B_REV]])
 ; CHECK-NEXT:    [[ADD1:%.*]] = add <vscale x 4 x i32> [[A:%.*]], [[B]]
-; CHECK-NEXT:    [[ADD:%.*]] = call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> [[ADD1]])
+; CHECK-NEXT:    [[ADD:%.*]] = call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> [[ADD1]])
 ; CHECK-NEXT:    ret <vscale x 4 x i32> [[ADD]]
 ;
-  %a.rev = tail call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> %a)
-  %b.rev = tail call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> %b)
+  %a.rev = tail call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> %a)
+  %b.rev = tail call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> %b)
   call void @use_nxv4i32(<vscale x 4 x i32>  %b.rev)
   %add = add <vscale x 4 x i32> %a.rev, %b.rev
   ret <vscale x 4 x i32> %add
@@ -52,15 +52,15 @@ define <vscale x 4 x i32> @binop_reverse_2(<vscale x 4 x i32> %a, <vscale x 4 x
 ; %a.rev and %b.rev have multiple uses
 define <vscale x 4 x i32> @binop_reverse_3(<vscale x 4 x i32> %a, <vscale x 4 x i32> %b) {
 ; CHECK-LABEL: @binop_reverse_3(
-; CHECK-NEXT:    [[A_REV:%.*]] = tail call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> [[A:%.*]])
-; CHECK-NEXT:    [[B_REV:%.*]] = tail call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> [[B:%.*]])
+; CHECK-NEXT:    [[A_REV:%.*]] = tail call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> [[A:%.*]])
+; CHECK-NEXT:    [[B_REV:%.*]] = tail call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> [[B:%.*]])
 ; CHECK-NEXT:    call void @use_nxv4i32(<vscale x 4 x i32> [[A_REV]])
 ; CHECK-NEXT:    call void @use_nxv4i32(<vscale x 4 x i32> [[B_REV]])
 ; CHECK-NEXT:    [[ADD:%.*]] = add <vscale x 4 x i32> [[A_REV]], [[B_REV]]
 ; CHECK-NEXT:    ret <vscale x 4 x i32> [[ADD]]
 ;
-  %a.rev = tail call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> %a)
-  %b.rev = tail call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> %b)
+  %a.rev = tail call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> %a)
+  %b.rev = tail call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> %b)
   call void @use_nxv4i32(<vscale x 4 x i32> %a.rev)
   call void @use_nxv4i32(<vscale x 4 x i32> %b.rev)
   %add = add <vscale x 4 x i32> %a.rev, %b.rev
@@ -71,10 +71,10 @@ define <vscale x 4 x i32> @binop_reverse_3(<vscale x 4 x i32> %a, <vscale x 4 x
 define <vscale x 4 x i32> @binop_reverse_4(<vscale x 4 x i32> %a) {
 ; CHECK-LABEL: @binop_reverse_4(
 ; CHECK-NEXT:    [[MUL1:%.*]] = mul <vscale x 4 x i32> [[A:%.*]], [[A]]
-; CHECK-NEXT:    [[MUL:%.*]] = call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> [[MUL1]])
+; CHECK-NEXT:    [[MUL:%.*]] = call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> [[MUL1]])
 ; CHECK-NEXT:    ret <vscale x 4 x i32> [[MUL]]
 ;
-  %a.rev = tail call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> %a)
+  %a.rev = tail call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> %a)
   %mul = mul <vscale x 4 x i32> %a.rev, %a.rev
   ret <vscale x 4 x i32> %mul
 }
@@ -82,12 +82,12 @@ define <vscale x 4 x i32> @binop_reverse_4(<vscale x 4 x i32> %a) {
 ; %a.rev used as both operands along with a third use
 define <vscale x 4 x i32> @binop_reverse_5(<vscale x 4 x i32> %a) {
 ; CHECK-LABEL: @binop_reverse_5(
-; CHECK-NEXT:    [[A_REV:%.*]] = tail call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> [[A:%.*]])
+; CHECK-NEXT:    [[A_REV:%.*]] = tail call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> [[A:%.*]])
 ; CHECK-NEXT:    call void @use_nxv4i32(<vscale x 4 x i32> [[A_REV]])
 ; CHECK-NEXT:    [[MUL:%.*]] = mul <vscale x 4 x i32> [[A_REV]], [[A_REV]]
 ; CHECK-NEXT:    ret <vscale x 4 x i32> [[MUL]]
 ;
-  %a.rev = tail call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> %a)
+  %a.rev = tail call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> %a)
   call void @use_nxv4i32(<vscale x 4 x i32> %a.rev)
   %mul = mul <vscale x 4 x i32> %a.rev, %a.rev
   ret <vscale x 4 x i32> %mul
@@ -98,10 +98,10 @@ define <vscale x 4 x i32> @binop_reverse_splat_RHS(<vscale x 4 x i32> %a, i32 %b
 ; CHECK-NEXT:    [[B_INSERT:%.*]] = insertelement <vscale x 4 x i32> poison, i32 [[B:%.*]], i64 0
 ; CHECK-NEXT:    [[B_SPLAT:%.*]] = shufflevector <vscale x 4 x i32> [[B_INSERT]], <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer
 ; CHECK-NEXT:    [[DIV1:%.*]] = udiv <vscale x 4 x i32> [[A:%.*]], [[B_SPLAT]]
-; CHECK-NEXT:    [[DIV:%.*]] = call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> [[DIV1]])
+; CHECK-NEXT:    [[DIV:%.*]] = call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> [[DIV1]])
 ; CHECK-NEXT:    ret <vscale x 4 x i32> [[DIV]]
 ;
-  %a.rev = tail call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> %a)
+  %a.rev = tail call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> %a)
   %b.insert = insertelement <vscale x 4 x i32> poison, i32 %b, i32 0
   %b.splat = shufflevector <vscale x 4 x i32> %b.insert, <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer
   %div = udiv <vscale x 4 x i32> %a.rev, %b.splat
@@ -111,14 +111,14 @@ define <vscale x 4 x i32> @binop_reverse_splat_RHS(<vscale x 4 x i32> %a, i32 %b
 ; %a.rev has multiple uses
 define <vscale x 4 x i32> @binop_reverse_splat_RHS_1(<vscale x 4 x i32> %a, i32 %b) {
 ; CHECK-LABEL: @binop_reverse_splat_RHS_1(
-; CHECK-NEXT:    [[A_REV:%.*]] = tail call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> [[A:%.*]])
+; CHECK-NEXT:    [[A_REV:%.*]] = tail call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> [[A:%.*]])
 ; CHECK-NEXT:    [[B_INSERT:%.*]] = insertelement <vscale x 4 x i32> poison, i32 [[B:%.*]], i64 0
 ; CHECK-NEXT:    [[B_SPLAT:%.*]] = shufflevector <vscale x 4 x i32> [[B_INSERT]], <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer
 ; CHECK-NEXT:    call void @use_nxv4i32(<vscale x 4 x i32> [[A_REV]])
 ; CHECK-NEXT:    [[DIV:%.*]] = udiv <vscale x 4 x i32> [[A_REV]], [[B_SPLAT]]
 ; CHECK-NEXT:    ret <vscale x 4 x i32> [[DIV]]
 ;
-  %a.rev = tail call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> %a)
+  %a.rev = tail call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> %a)
   %b.insert = insertelement <vscale x 4 x i32> poison, i32 %b, i32 0
   %b.splat = shufflevector <vscale x 4 x i32> %b.insert, <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer
   call void @use_nxv4i32(<vscale x 4 x i32> %a.rev)
@@ -131,10 +131,10 @@ define <vscale x 4 x i32> @binop_reverse_splat_LHS(<vscale x 4 x i32> %a, i32 %b
 ; CHECK-NEXT:    [[B_INSERT:%.*]] = insertelement <vscale x 4 x i32> poison, i32 [[B:%.*]], i64 0
 ; CHECK-NEXT:    [[B_SPLAT:%.*]] = shufflevector <vscale x 4 x i32> [[B_INSERT]], <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer
 ; CHECK-NEXT:    [[DIV1:%.*]] = udiv <vscale x 4 x i32> [[B_SPLAT]], [[A:%.*]]
-; CHECK-NEXT:    [[DIV:%.*]] = call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> [[DIV1]])
+; CHECK-NEXT:    [[DIV:%.*]] = call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> [[DIV1]])
 ; CHECK-NEXT:    ret <vscale x 4 x i32> [[DIV]]
 ;
-  %a.rev = tail call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> %a)
+  %a.rev = tail call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> %a)
   %b.insert = insertelement <vscale x 4 x i32> poison, i32 %b, i32 0
   %b.splat = shufflevector <vscale x 4 x i32> %b.insert, <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer
   %div = udiv <vscale x 4 x i32> %b.splat, %a.rev
@@ -144,14 +144,14 @@ define <vscale x 4 x i32> @binop_reverse_splat_LHS(<vscale x 4 x i32> %a, i32 %b
 ; %a.rev has multiple uses
 define <vscale x 4 x i32> @binop_reverse_splat_LHS_1(<vscale x 4 x i32> %a, i32 %b) {
 ; CHECK-LABEL: @binop_reverse_splat_LHS_1(
-; CHECK-NEXT:    [[A_REV:%.*]] = tail call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> [[A:%.*]])
+; CHECK-NEXT:    [[A_REV:%.*]] = tail call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> [[A:%.*]])
 ; CHECK-NEXT:    [[B_INSERT:%.*]] = insertelement <vscale x 4 x i32> poison, i32 [[B:%.*]], i64 0
 ; CHECK-NEXT:    [[B_SPLAT:%.*]] = shufflevector <vscale x 4 x i32> [[B_INSERT]], <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer
 ; CHECK-NEXT:    call void @use_nxv4i32(<vscale x 4 x i32> [[A_REV]])
 ; CHECK-NEXT:    [[DIV:%.*]] = udiv <vscale x 4 x i32> [[B_SPLAT]], [[A_REV]]
 ; CHECK-NEXT:    ret <vscale x 4 x i32> [[DIV]]
 ;
-  %a.rev = tail call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> %a)
+  %a.rev = tail call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> %a)
   %b.insert = insertelement <vscale x 4 x i32> poison, i32 %b, i32 0
   %b.splat = shufflevector <vscale x 4 x i32> %b.insert, <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer
   call void @use_nxv4i32(<vscale x 4 x i32> %a.rev)
@@ -161,11 +161,11 @@ define <vscale x 4 x i32> @binop_reverse_splat_LHS_1(<vscale x 4 x i32> %a, i32
 
 define <vscale x 4 x float> @unop_reverse(<vscale x 4 x float> %a) {
 ; CHECK-LABEL: @unop_reverse(
-; CHECK-NEXT:    [[A_REV:%.*]] = tail call <vscale x 4 x float> @llvm.experimental.vector.reverse.nxv4f32(<vscale x 4 x float> [[A:%.*]])
+; CHECK-NEXT:    [[A_REV:%.*]] = tail call <vscale x 4 x float> @llvm.vector.reverse.nxv4f32(<vscale x 4 x float> [[A:%.*]])
 ; CHECK-NEXT:    [[NEG:%.*]] = fneg fast <vscale x 4 x float> [[A_REV]]
 ; CHECK-NEXT:    ret <vscale x 4 x float> [[NEG]]
 ;
-  %a.rev = tail call <vscale x 4 x float> @llvm.experimental.vector.reverse.nxv4f32(<vscale x 4 x float> %a)
+  %a.rev = tail call <vscale x 4 x float> @llvm.vector.reverse.nxv4f32(<vscale x 4 x float> %a)
   %neg = fneg fast <vscale x 4 x float> %a.rev
   ret <vscale x 4 x float> %neg
 }
@@ -173,12 +173,12 @@ define <vscale x 4 x float> @unop_reverse(<vscale x 4 x float> %a) {
 ; %a.rev has multiple uses
 define <vscale x 4 x float> @unop_reverse_1(<vscale x 4 x float> %a) {
 ; CHECK-LABEL: @unop_reverse_1(
-; CHECK-NEXT:    [[A_REV:%.*]] = tail call <vscale x 4 x float> @llvm.experimental.vector.reverse.nxv4f32(<vscale x 4 x float> [[A:%.*]])
+; CHECK-NEXT:    [[A_REV:%.*]] = tail call <vscale x 4 x float> @llvm.vector.reverse.nxv4f32(<vscale x 4 x float> [[A:%.*]])
 ; CHECK-NEXT:    call void @use_nxv4f32(<vscale x 4 x float> [[A_REV]])
 ; CHECK-NEXT:    [[NEG:%.*]] = fneg fast <vscale x 4 x float> [[A_REV]]
 ; CHECK-NEXT:    ret <vscale x 4 x float> [[NEG]]
 ;
-  %a.rev = tail call <vscale x 4 x float> @llvm.experimental.vector.reverse.nxv4f32(<vscale x 4 x float> %a)
+  %a.rev = tail call <vscale x 4 x float> @llvm.vector.reverse.nxv4f32(<vscale x 4 x float> %a)
   call void @use_nxv4f32(<vscale x 4 x float> %a.rev)
   %neg = fneg fast <vscale x 4 x float> %a.rev
   ret <vscale x 4 x float> %neg
@@ -187,11 +187,11 @@ define <vscale x 4 x float> @unop_reverse_1(<vscale x 4 x float> %a) {
 define <vscale x 4 x i1> @icmp_reverse(<vscale x 4 x i32> %a, <vscale x 4 x i32> %b) {
 ; CHECK-LABEL: @icmp_reverse(
 ; CHECK-NEXT:    [[CMP1:%.*]] = icmp eq <vscale x 4 x i32> [[A:%.*]], [[B:%.*]]
-; CHECK-NEXT:    [[CMP:%.*]] = call <vscale x 4 x i1> @llvm.experimental.vector.reverse.nxv4i1(<vscale x 4 x i1> [[CMP1]])
+; CHECK-NEXT:    [[CMP:%.*]] = call <vscale x 4 x i1> @llvm.vector.reverse.nxv4i1(<vscale x 4 x i1> [[CMP1]])
 ; CHECK-NEXT:    ret <vscale x 4 x i1> [[CMP]]
 ;
-  %a.rev = tail call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> %a)
-  %b.rev = tail call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> %b)
+  %a.rev = tail call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> %a)
+  %b.rev = tail call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> %b)
   %cmp = icmp eq <vscale x 4 x i32> %a.rev, %b.rev
   ret <vscale x 4 x i1> %cmp
 }
@@ -199,14 +199,14 @@ define <vscale x 4 x i1> @icmp_reverse(<vscale x 4 x i32> %a, <vscale x 4 x i32>
 ; %a.rev has multiple uses
 define <vscale x 4 x i1> @icmp_reverse_1(<vscale x 4 x i32> %a, <vscale x 4 x i32> %b) {
 ; CHECK-LABEL: @icmp_reverse_1(
-; CHECK-NEXT:    [[A_REV:%.*]] = tail call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> [[A:%.*]])
+; CHECK-NEXT:    [[A_REV:%.*]] = tail call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> [[A:%.*]])
 ; CHECK-NEXT:    call void @use_nxv4i32(<vscale x 4 x i32> [[A_REV]])
 ; CHECK-NEXT:    [[CMP1:%.*]] = icmp eq <vscale x 4 x i32> [[A]], [[B:%.*]]
-; CHECK-NEXT:    [[CMP:%.*]] = call <vscale x 4 x i1> @llvm.experimental.vector.reverse.nxv4i1(<vscale x 4 x i1> [[CMP1]])
+; CHECK-NEXT:    [[CMP:%.*]] = call <vscale x 4 x i1> @llvm.vector.reverse.nxv4i1(<vscale x 4 x i1> [[CMP1]])
 ; CHECK-NEXT:    ret <vscale x 4 x i1> [[CMP]]
 ;
-  %a.rev = tail call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> %a)
-  %b.rev = tail call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> %b)
+  %a.rev = tail call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> %a)
+  %b.rev = tail call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> %b)
   call void @use_nxv4i32(<vscale x 4 x i32> %a.rev)
   %cmp = icmp eq <vscale x 4 x i32> %a.rev, %b.rev
   ret <vscale x 4 x i1> %cmp
@@ -215,14 +215,14 @@ define <vscale x 4 x i1> @icmp_reverse_1(<vscale x 4 x i32> %a, <vscale x 4 x i3
 ; %b.rev has multiple uses
 define <vscale x 4 x i1> @icmp_reverse_2(<vscale x 4 x i32> %a, <vscale x 4 x i32> %b) {
 ; CHECK-LABEL: @icmp_reverse_2(
-; CHECK-NEXT:    [[B_REV:%.*]] = tail call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> [[B:%.*]])
+; CHECK-NEXT:    [[B_REV:%.*]] = tail call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> [[B:%.*]])
 ; CHECK-NEXT:    call void @use_nxv4i32(<vscale x 4 x i32> [[B_REV]])
 ; CHECK-NEXT:    [[CMP1:%.*]] = icmp eq <vscale x 4 x i32> [[A:%.*]], [[B]]
-; CHECK-NEXT:    [[CMP:%.*]] = call <vscale x 4 x i1> @llvm.experimental.vector.reverse.nxv4i1(<vscale x 4 x i1> [[CMP1]])
+; CHECK-NEXT:    [[CMP:%.*]] = call <vscale x 4 x i1> @llvm.vector.reverse.nxv4i1(<vscale x 4 x i1> [[CMP1]])
 ; CHECK-NEXT:    ret <vscale x 4 x i1> [[CMP]]
 ;
-  %a.rev = tail call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> %a)
-  %b.rev = tail call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> %b)
+  %a.rev = tail call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> %a)
+  %b.rev = tail call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> %b)
   call void @use_nxv4i32(<vscale x 4 x i32> %b.rev)
   %cmp = icmp eq <vscale x 4 x i32> %a.rev, %b.rev
   ret <vscale x 4 x i1> %cmp
@@ -231,15 +231,15 @@ define <vscale x 4 x i1> @icmp_reverse_2(<vscale x 4 x i32> %a, <vscale x 4 x i3
 ; %a.rev and %b.rev have multiple uses
 define <vscale x 4 x i1> @icmp_reverse_3(<vscale x 4 x i32> %a, <vscale x 4 x i32> %b) {
 ; CHECK-LABEL: @icmp_reverse_3(
-; CHECK-NEXT:    [[A_REV:%.*]] = tail call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> [[A:%.*]])
-; CHECK-NEXT:    [[B_REV:%.*]] = tail call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> [[B:%.*]])
+; CHECK-NEXT:    [[A_REV:%.*]] = tail call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> [[A:%.*]])
+; CHECK-NEXT:    [[B_REV:%.*]] = tail call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> [[B:%.*]])
 ; CHECK-NEXT:    call void @use_nxv4i32(<vscale x 4 x i32> [[A_REV]])
 ; CHECK-NEXT:    call void @use_nxv4i32(<vscale x 4 x i32> [[B_REV]])
 ; CHECK-NEXT:    [[CMP:%.*]] = icmp eq <vscale x 4 x i32> [[A_REV]], [[B_REV]]
 ; CHECK-NEXT:    ret <vscale x 4 x i1> [[CMP]]
 ;
-  %a.rev = tail call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> %a)
-  %b.rev = tail call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> %b)
+  %a.rev = tail call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> %a)
+  %b.rev = tail call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> %b)
   call void @use_nxv4i32(<vscale x 4 x i32> %a.rev)
   call void @use_nxv4i32(<vscale x 4 x i32> %b.rev)
   %cmp = icmp eq <vscale x 4 x i32> %a.rev, %b.rev
@@ -251,10 +251,10 @@ define <vscale x 4 x i1> @icmp_reverse_splat_RHS(<vscale x 4 x i32> %a, i32 %b)
 ; CHECK-NEXT:    [[B_INSERT:%.*]] = insertelement <vscale x 4 x i32> poison, i32 [[B:%.*]], i64 0
 ; CHECK-NEXT:    [[B_SPLAT:%.*]] = shufflevector <vscale x 4 x i32> [[B_INSERT]], <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer
 ; CHECK-NEXT:    [[CMP1:%.*]] = icmp slt <vscale x 4 x i32> [[B_SPLAT]], [[A:%.*]]
-; CHECK-NEXT:    [[CMP:%.*]] = call <vscale x 4 x i1> @llvm.experimental.vector.reverse.nxv4i1(<vscale x 4 x i1> [[CMP1]])
+; CHECK-NEXT:    [[CMP:%.*]] = call <vscale x 4 x i1> @llvm.vector.reverse.nxv4i1(<vscale x 4 x i1> [[CMP1]])
 ; CHECK-NEXT:    ret <vscale x 4 x i1> [[CMP]]
 ;
-  %a.rev = tail call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> %a)
+  %a.rev = tail call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> %a)
   %b.insert = insertelement <vscale x 4 x i32> poison, i32 %b, i32 0
   %b.splat = shufflevector <vscale x 4 x i32> %b.insert, <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer
   %cmp = icmp sgt <vscale x 4 x i32> %a.rev, %b.splat
@@ -264,14 +264,14 @@ define <vscale x 4 x i1> @icmp_reverse_splat_RHS(<vscale x 4 x i32> %a, i32 %b)
 ; %a.rev has multiple uses
 define <vscale x 4 x i1> @icmp_reverse_splat_RHS_1(<vscale x 4 x i32> %a, i32 %b) {
 ; CHECK-LABEL: @icmp_reverse_splat_RHS_1(
-; CHECK-NEXT:    [[A_REV:%.*]] = tail call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> [[A:%.*]])
+; CHECK-NEXT:    [[A_REV:%.*]] = tail call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> [[A:%.*]])
 ; CHECK-NEXT:    [[B_INSERT:%.*]] = insertelement <vscale x 4 x i32> poison, i32 [[B:%.*]], i64 0
 ; CHECK-NEXT:    [[B_SPLAT:%.*]] = shufflevector <vscale x 4 x i32> [[B_INSERT]], <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer
 ; CHECK-NEXT:    call void @use_nxv4i32(<vscale x 4 x i32> [[A_REV]])
 ; CHECK-NEXT:    [[CMP:%.*]] = icmp sgt <vscale x 4 x i32> [[A_REV]], [[B_SPLAT]]
 ; CHECK-NEXT:    ret <vscale x 4 x i1> [[CMP]]
 ;
-  %a.rev = tail call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> %a)
+  %a.rev = tail call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> %a)
   %b.insert = insertelement <vscale x 4 x i32> poison, i32 %b, i32 0
   %b.splat = shufflevector <vscale x 4 x i32> %b.insert, <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer
   call void @use_nxv4i32(<vscale x 4 x i32> %a.rev)
@@ -284,10 +284,10 @@ define <vscale x 4 x i1> @icmp_reverse_splat_LHS(<vscale x 4 x i32> %a, i32 %b)
 ; CHECK-NEXT:    [[B_INSERT:%.*]] = insertelement <vscale x 4 x i32> poison, i32 [[B:%.*]], i64 0
 ; CHECK-NEXT:    [[B_SPLAT:%.*]] = shufflevector <vscale x 4 x i32> [[B_INSERT]], <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer
 ; CHECK-NEXT:    [[CMP1:%.*]] = icmp ult <vscale x 4 x i32> [[B_SPLAT]], [[A:%.*]]
-; CHECK-NEXT:    [[CMP:%.*]] = call <vscale x 4 x i1> @llvm.experimental.vector.reverse.nxv4i1(<vscale x 4 x i1> [[CMP1]])
+; CHECK-NEXT:    [[CMP:%.*]] = call <vscale x 4 x i1> @llvm.vector.reverse.nxv4i1(<vscale x 4 x i1> [[CMP1]])
 ; CHECK-NEXT:    ret <vscale x 4 x i1> [[CMP]]
 ;
-  %a.rev = tail call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> %a)
+  %a.rev = tail call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> %a)
   %b.insert = insertelement <vscale x 4 x i32> poison, i32 %b, i32 0
   %b.splat = shufflevector <vscale x 4 x i32> %b.insert, <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer
   %cmp = icmp ult <vscale x 4 x i32> %b.splat, %a.rev
@@ -297,14 +297,14 @@ define <vscale x 4 x i1> @icmp_reverse_splat_LHS(<vscale x 4 x i32> %a, i32 %b)
 ; %a.rev has multiple uses
 define <vscale x 4 x i1> @icmp_reverse_splat_LHS_1(<vscale x 4 x i32> %a, i32 %b) {
 ; CHECK-LABEL: @icmp_reverse_splat_LHS_1(
-; CHECK-NEXT:    [[A_REV:%.*]] = tail call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> [[A:%.*]])
+; CHECK-NEXT:    [[A_REV:%.*]] = tail call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> [[A:%.*]])
 ; CHECK-NEXT:    [[B_INSERT:%.*]] = insertelement <vscale x 4 x i32> poison, i32 [[B:%.*]], i64 0
 ; CHECK-NEXT:    [[B_SPLAT:%.*]] = shufflevector <vscale x 4 x i32> [[B_INSERT]], <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer
 ; CHECK-NEXT:    call void @use_nxv4i32(<vscale x 4 x i32> [[A_REV]])
 ; CHECK-NEXT:    [[CMP:%.*]] = icmp ult <vscale x 4 x i32> [[B_SPLAT]], [[A_REV]]
 ; CHECK-NEXT:    ret <vscale x 4 x i1> [[CMP]]
 ;
-  %a.rev = tail call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> %a)
+  %a.rev = tail call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> %a)
   %b.insert = insertelement <vscale x 4 x i32> poison, i32 %b, i32 0
   %b.splat = shufflevector <vscale x 4 x i32> %b.insert, <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer
   call void @use_nxv4i32(<vscale x 4 x i32> %a.rev)
@@ -315,12 +315,12 @@ define <vscale x 4 x i1> @icmp_reverse_splat_LHS_1(<vscale x 4 x i32> %a, i32 %b
 define <vscale x 4 x i32> @select_reverse(<vscale x 4 x i1> %a, <vscale x 4 x i32> %b, <vscale x 4 x i32> %c) {
 ; CHECK-LABEL: @select_reverse(
 ; CHECK-NEXT:    [[SELECT1:%.*]] = select <vscale x 4 x i1> [[A:%.*]], <vscale x 4 x i32> [[B:%.*]], <vscale x 4 x i32> [[C:%.*]]
-; CHECK-NEXT:    [[SELECT:%.*]] = call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> [[SELECT1]])
+; CHECK-NEXT:    [[SELECT:%.*]] = call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> [[SELECT1]])
 ; CHECK-NEXT:    ret <vscale x 4 x i32> [[SELECT]]
 ;
-  %a.rev = tail call <vscale x 4 x i1> @llvm.experimental.vector.reverse.nxv4i1(<vscale x 4 x i1> %a)
-  %b.rev = tail call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> %b)
-  %c.rev = tail call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> %c)
+  %a.rev = tail call <vscale x 4 x i1> @llvm.vector.reverse.nxv4i1(<vscale x 4 x i1> %a)
+  %b.rev = tail call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> %b)
+  %c.rev = tail call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> %c)
   %select = select <vscale x 4 x i1> %a.rev, <vscale x 4 x i32> %b.rev, <vscale x 4 x i32> %c.rev
   ret <vscale x 4 x i32> %select
 }
@@ -328,15 +328,15 @@ define <vscale x 4 x i32> @select_reverse(<vscale x 4 x i1> %a, <vscale x 4 x i3
 ; %a.rev has multiple uses
 define <vscale x 4 x i32> @select_reverse_1(<vscale x 4 x i1> %a, <vscale x 4 x i32> %b, <vscale x 4 x i32> %c) {
 ; CHECK-LABEL: @select_reverse_1(
-; CHECK-NEXT:    [[A_REV:%.*]] = tail call <vscale x 4 x i1> @llvm.experimental.vector.reverse.nxv4i1(<vscale x 4 x i1> [[A:%.*]])
+; CHECK-NEXT:    [[A_REV:%.*]] = tail call <vscale x 4 x i1> @llvm.vector.reverse.nxv4i1(<vscale x 4 x i1> [[A:%.*]])
 ; CHECK-NEXT:    call void @use_nxv4i1(<vscale x 4 x i1> [[A_REV]])
 ; CHECK-NEXT:    [[SELECT1:%.*]] = select <vscale x 4 x i1> [[A]], <vscale x 4 x i32> [[B:%.*]], <vscale x 4 x i32> [[C:%.*]]
-; CHECK-NEXT:    [[SELECT:%.*]] = call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> [[SELECT1]])
+; CHECK-NEXT:    [[SELECT:%.*]] = call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> [[SELECT1]])
 ; CHECK-NEXT:    ret <vscale x 4 x i32> [[SELECT]]
 ;
-  %a.rev = tail call <vscale x 4 x i1> @llvm.experimental.vector.reverse.nxv4i1(<vscale x 4 x i1> %a)
-  %b.rev = tail call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> %b)
-  %c.rev = tail call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> %c)
+  %a.rev = tail call <vscale x 4 x i1> @llvm.vector.reverse.nxv4i1(<vscale x 4 x i1> %a)
+  %b.rev = tail call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> %b)
+  %c.rev = tail call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> %c)
   call void @use_nxv4i1(<vscale x 4 x i1> %a.rev)
   %select = select <vscale x 4 x i1> %a.rev, <vscale x 4 x i32> %b.rev, <vscale x 4 x i32> %c.rev
   ret <vscale x 4 x i32> %select
@@ -345,15 +345,15 @@ define <vscale x 4 x i32> @select_reverse_1(<vscale x 4 x i1> %a, <vscale x 4 x
 ; %b.rev has multiple uses
 define <vscale x 4 x i32> @select_reverse_2(<vscale x 4 x i1> %a, <vscale x 4 x i32> %b, <vscale x 4 x i32> %c) {
 ; CHECK-LABEL: @select_reverse_2(
-; CHECK-NEXT:    [[B_REV:%.*]] = tail call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> [[B:%.*]])
+; CHECK-NEXT:    [[B_REV:%.*]] = tail call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> [[B:%.*]])
 ; CHECK-NEXT:    call void @use_nxv4i32(<vscale x 4 x i32> [[B_REV]])
 ; CHECK-NEXT:    [[SELECT1:%.*]] = select <vscale x 4 x i1> [[A:%.*]], <vscale x 4 x i32> [[B]], <vscale x 4 x i32> [[C:%.*]]
-; CHECK-NEXT:    [[SELECT:%.*]] = call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> [[SELECT1]])
+; CHECK-NEXT:    [[SELECT:%.*]] = call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> [[SELECT1]])
 ; CHECK-NEXT:    ret <vscale x 4 x i32> [[SELECT]]
 ;
-  %a.rev = tail call <vscale x 4 x i1> @llvm.experimental.vector.reverse.nxv4i1(<vscale x 4 x i1> %a)
-  %b.rev = tail call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> %b)
-  %c.rev = tail call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> %c)
+  %a.rev = tail call <vscale x 4 x i1> @llvm.vector.reverse.nxv4i1(<vscale x 4 x i1> %a)
+  %b.rev = tail call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> %b)
+  %c.rev = tail call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> %c)
   call void @use_nxv4i32(<vscale x 4 x i32> %b.rev)
   %select = select <vscale x 4 x i1> %a.rev, <vscale x 4 x i32> %b.rev, <vscale x 4 x i32> %c.rev
   ret <vscale x 4 x i32> %select
@@ -362,15 +362,15 @@ define <vscale x 4 x i32> @select_reverse_2(<vscale x 4 x i1> %a, <vscale x 4 x
 ; %c.rev has multiple uses
 define <vscale x 4 x i32> @select_reverse_3(<vscale x 4 x i1> %a, <vscale x 4 x i32> %b, <vscale x 4 x i32> %c) {
 ; CHECK-LABEL: @select_reverse_3(
-; CHECK-NEXT:    [[C_REV:%.*]] = tail call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> [[C:%.*]])
+; CHECK-NEXT:    [[C_REV:%.*]] = tail call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> [[C:%.*]])
 ; CHECK-NEXT:    call void @use_nxv4i32(<vscale x 4 x i32> [[C_REV]])
 ; CHECK-NEXT:    [[SELECT1:%.*]] = select <vscale x 4 x i1> [[A:%.*]], <vscale x 4 x i32> [[B:%.*]], <vscale x 4 x i32> [[C]]
-; CHECK-NEXT:    [[SELECT:%.*]] = call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> [[SELECT1]])
+; CHECK-NEXT:    [[SELECT:%.*]] = call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> [[SELECT1]])
 ; CHECK-NEXT:    ret <vscale x 4 x i32> [[SELECT]]
 ;
-  %a.rev = tail call <vscale x 4 x i1> @llvm.experimental.vector.reverse.nxv4i1(<vscale x 4 x i1> %a)
-  %b.rev = tail call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> %b)
-  %c.rev = tail call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> %c)
+  %a.rev = tail call <vscale x 4 x i1> @llvm.vector.reverse.nxv4i1(<vscale x 4 x i1> %a)
+  %b.rev = tail call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> %b)
+  %c.rev = tail call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> %c)
   call void @use_nxv4i32(<vscale x 4 x i32> %c.rev)
   %select = select <vscale x 4 x i1> %a.rev, <vscale x 4 x i32> %b.rev, <vscale x 4 x i32> %c.rev
   ret <vscale x 4 x i32> %select
@@ -379,17 +379,17 @@ define <vscale x 4 x i32> @select_reverse_3(<vscale x 4 x i1> %a, <vscale x 4 x
 ; %a.rev and %b.rev have multiple uses
 define <vscale x 4 x i32> @select_reverse_4(<vscale x 4 x i1> %a, <vscale x 4 x i32> %b, <vscale x 4 x i32> %c) {
 ; CHECK-LABEL: @select_reverse_4(
-; CHECK-NEXT:    [[A_REV:%.*]] = tail call <vscale x 4 x i1> @llvm.experimental.vector.reverse.nxv4i1(<vscale x 4 x i1> [[A:%.*]])
-; CHECK-NEXT:    [[B_REV:%.*]] = tail call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> [[B:%.*]])
+; CHECK-NEXT:    [[A_REV:%.*]] = tail call <vscale x 4 x i1> @llvm.vector.reverse.nxv4i1(<vscale x 4 x i1> [[A:%.*]])
+; CHECK-NEXT:    [[B_REV:%.*]] = tail call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> [[B:%.*]])
 ; CHECK-NEXT:    call void @use_nxv4i1(<vscale x 4 x i1> [[A_REV]])
 ; CHECK-NEXT:    call void @use_nxv4i32(<vscale x 4 x i32> [[B_REV]])
 ; CHECK-NEXT:    [[SELECT1:%.*]] = select <vscale x 4 x i1> [[A]], <vscale x 4 x i32> [[B]], <vscale x 4 x i32> [[C:%.*]]
-; CHECK-NEXT:    [[SELECT:%.*]] = call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> [[SELECT1]])
+; CHECK-NEXT:    [[SELECT:%.*]] = call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> [[SELECT1]])
 ; CHECK-NEXT:    ret <vscale x 4 x i32> [[SELECT]]
 ;
-  %a.rev = tail call <vscale x 4 x i1> @llvm.experimental.vector.reverse.nxv4i1(<vscale x 4 x i1> %a)
-  %b.rev = tail call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> %b)
-  %c.rev = tail call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> %c)
+  %a.rev = tail call <vscale x 4 x i1> @llvm.vector.reverse.nxv4i1(<vscale x 4 x i1> %a)
+  %b.rev = tail call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> %b)
+  %c.rev = tail call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> %c)
   call void @use_nxv4i1(<vscale x 4 x i1> %a.rev)
   call void @use_nxv4i32(<vscale x 4 x i32> %b.rev)
   %select = select <vscale x 4 x i1> %a.rev, <vscale x 4 x i32> %b.rev, <vscale x 4 x i32> %c.rev
@@ -399,17 +399,17 @@ define <vscale x 4 x i32> @select_reverse_4(<vscale x 4 x i1> %a, <vscale x 4 x
 ; %a.rev and %c.rev have multiple uses
 define <vscale x 4 x i32> @select_reverse_5(<vscale x 4 x i1> %a, <vscale x 4 x i32> %b, <vscale x 4 x i32> %c) {
 ; CHECK-LABEL: @select_reverse_5(
-; CHECK-NEXT:    [[A_REV:%.*]] = tail call <vscale x 4 x i1> @llvm.experimental.vector.reverse.nxv4i1(<vscale x 4 x i1> [[A:%.*]])
-; CHECK-NEXT:    [[C_REV:%.*]] = tail call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> [[C:%.*]])
+; CHECK-NEXT:    [[A_REV:%.*]] = tail call <vscale x 4 x i1> @llvm.vector.reverse.nxv4i1(<vscale x 4 x i1> [[A:%.*]])
+; CHECK-NEXT:    [[C_REV:%.*]] = tail call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> [[C:%.*]])
 ; CHECK-NEXT:    call void @use_nxv4i1(<vscale x 4 x i1> [[A_REV]])
 ; CHECK-NEXT:    call void @use_nxv4i32(<vscale x 4 x i32> [[C_REV]])
 ; CHECK-NEXT:    [[SELECT1:%.*]] = select <vscale x 4 x i1> [[A]], <vscale x 4 x i32> [[B:%.*]], <vscale x 4 x i32> [[C]]
-; CHECK-NEXT:    [[SELECT:%.*]] = call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> [[SELECT1]])
+; CHECK-NEXT:    [[SELECT:%.*]] = call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> [[SELECT1]])
 ; CHECK-NEXT:    ret <vscale x 4 x i32> [[SELECT]]
 ;
-  %a.rev = tail call <vscale x 4 x i1> @llvm.experimental.vector.reverse.nxv4i1(<vscale x 4 x i1> %a)
-  %b.rev = tail call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> %b)
-  %c.rev = tail call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> %c)
+  %a.rev = tail call <vscale x 4 x i1> @llvm.vector.reverse.nxv4i1(<vscale x 4 x i1> %a)
+  %b.rev = tail call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> %b)
+  %c.rev = tail call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> %c)
   call void @use_nxv4i1(<vscale x 4 x i1> %a.rev)
   call void @use_nxv4i32(<vscale x 4 x i32> %c.rev)
   %select = select <vscale x 4 x i1> %a.rev, <vscale x 4 x i32> %b.rev, <vscale x 4 x i32> %c.rev
@@ -419,17 +419,17 @@ define <vscale x 4 x i32> @select_reverse_5(<vscale x 4 x i1> %a, <vscale x 4 x
 ; %b.rev and %c.rev have multiple uses
 define <vscale x 4 x i32> @select_reverse_6(<vscale x 4 x i1> %a, <vscale x 4 x i32> %b, <vscale x 4 x i32> %c) {
 ; CHECK-LABEL: @select_reverse_6(
-; CHECK-NEXT:    [[B_REV:%.*]] = tail call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> [[B:%.*]])
-; CHECK-NEXT:    [[C_REV:%.*]] = tail call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> [[C:%.*]])
+; CHECK-NEXT:    [[B_REV:%.*]] = tail call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> [[B:%.*]])
+; CHECK-NEXT:    [[C_REV:%.*]] = tail call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> [[C:%.*]])
 ; CHECK-NEXT:    call void @use_nxv4i32(<vscale x 4 x i32> [[B_REV]])
 ; CHECK-NEXT:    call void @use_nxv4i32(<vscale x 4 x i32> [[C_REV]])
 ; CHECK-NEXT:    [[SELECT1:%.*]] = select <vscale x 4 x i1> [[A:%.*]], <vscale x 4 x i32> [[B]], <vscale x 4 x i32> [[C]]
-; CHECK-NEXT:    [[SELECT:%.*]] = call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> [[SELECT1]])
+; CHECK-NEXT:    [[SELECT:%.*]] = call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> [[SELECT1]])
 ; CHECK-NEXT:    ret <vscale x 4 x i32> [[SELECT]]
 ;
-  %a.rev = tail call <vscale x 4 x i1> @llvm.experimental.vector.reverse.nxv4i1(<vscale x 4 x i1> %a)
-  %b.rev = tail call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> %b)
-  %c.rev = tail call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> %c)
+  %a.rev = tail call <vscale x 4 x i1> @llvm.vector.reverse.nxv4i1(<vscale x 4 x i1> %a)
+  %b.rev = tail call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> %b)
+  %c.rev = tail call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> %c)
   call void @use_nxv4i32(<vscale x 4 x i32> %b.rev)
   call void @use_nxv4i32(<vscale x 4 x i32> %c.rev)
   %select = select <vscale x 4 x i1> %a.rev, <vscale x 4 x i32> %b.rev, <vscale x 4 x i32> %c.rev
@@ -439,18 +439,18 @@ define <vscale x 4 x i32> @select_reverse_6(<vscale x 4 x i1> %a, <vscale x 4 x
 ; %a.rev, %b.rev and %c.rev have multiple uses
 define <vscale x 4 x i32> @select_reverse_7(<vscale x 4 x i1> %a, <vscale x 4 x i32> %b, <vscale x 4 x i32> %c) {
 ; CHECK-LABEL: @select_reverse_7(
-; CHECK-NEXT:    [[A_REV:%.*]] = tail call <vscale x 4 x i1> @llvm.experimental.vector.reverse.nxv4i1(<vscale x 4 x i1> [[A:%.*]])
-; CHECK-NEXT:    [[B_REV:%.*]] = tail call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> [[B:%.*]])
-; CHECK-NEXT:    [[C_REV:%.*]] = tail call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> [[C:%.*]])
+; CHECK-NEXT:    [[A_REV:%.*]] = tail call <vscale x 4 x i1> @llvm.vector.reverse.nxv4i1(<vscale x 4 x i1> [[A:%.*]])
+; CHECK-NEXT:    [[B_REV:%.*]] = tail call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> [[B:%.*]])
+; CHECK-NEXT:    [[C_REV:%.*]] = tail call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> [[C:%.*]])
 ; CHECK-NEXT:    call void @use_nxv4i1(<vscale x 4 x i1> [[A_REV]])
 ; CHECK-NEXT:    call void @use_nxv4i32(<vscale x 4 x i32> [[B_REV]])
 ; CHECK-NEXT:    call void @use_nxv4i32(<vscale x 4 x i32> [[C_REV]])
 ; CHECK-NEXT:    [[SELECT:%.*]] = select <vscale x 4 x i1> [[A_REV]], <vscale x 4 x i32> [[B_REV]], <vscale x 4 x i32> [[C_REV]]
 ; CHECK-NEXT:    ret <vscale x 4 x i32> [[SELECT]]
 ;
-  %a.rev = tail call <vscale x 4 x i1> @llvm.experimental.vector.reverse.nxv4i1(<vscale x 4 x i1> %a)
-  %b.rev = tail call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> %b)
-  %c.rev = tail call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> %c)
+  %a.rev = tail call <vscale x 4 x i1> @llvm.vector.reverse.nxv4i1(<vscale x 4 x i1> %a)
+  %b.rev = tail call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> %b)
+  %c.rev = tail call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> %c)
   call void @use_nxv4i1(<vscale x 4 x i1> %a.rev)
   call void @use_nxv4i32(<vscale x 4 x i32> %b.rev)
   call void @use_nxv4i32(<vscale x 4 x i32> %c.rev)
@@ -463,11 +463,11 @@ define <vscale x 4 x i32> @select_reverse_splat_false(<vscale x 4 x i1> %a, <vsc
 ; CHECK-NEXT:    [[C_INSERT:%.*]] = insertelement <vscale x 4 x i32> poison, i32 [[C:%.*]], i64 0
 ; CHECK-NEXT:    [[C_SPLAT:%.*]] = shufflevector <vscale x 4 x i32> [[C_INSERT]], <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer
 ; CHECK-NEXT:    [[SELECT1:%.*]] = select <vscale x 4 x i1> [[A:%.*]], <vscale x 4 x i32> [[B:%.*]], <vscale x 4 x i32> [[C_SPLAT]]
-; CHECK-NEXT:    [[SELECT:%.*]] = call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> [[SELECT1]])
+; CHECK-NEXT:    [[SELECT:%.*]] = call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> [[SELECT1]])
 ; CHECK-NEXT:    ret <vscale x 4 x i32> [[SELECT]]
 ;
-  %a.rev = tail call <vscale x 4 x i1> @llvm.experimental.vector.reverse.nxv4i1(<vscale x 4 x i1> %a)
-  %b.rev = tail call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> %b)
+  %a.rev = tail call <vscale x 4 x i1> @llvm.vector.reverse.nxv4i1(<vscale x 4 x i1> %a)
+  %b.rev = tail call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> %b)
   %c.insert = insertelement <vscale x 4 x i32> poison, i32 %c, i32 0
   %c.splat = shufflevector <vscale x 4 x i32> %c.insert, <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer
   %select = select <vscale x 4 x i1> %a.rev, <vscale x 4 x i32> %b.rev, <vscale x 4 x i32> %c.splat
@@ -477,16 +477,16 @@ define <vscale x 4 x i32> @select_reverse_splat_false(<vscale x 4 x i1> %a, <vsc
 ; %a.rev has multiple uses
 define <vscale x 4 x i32> @select_reverse_splat_false_1(<vscale x 4 x i1> %a, <vscale x 4 x i32> %b, i32 %c) {
 ; CHECK-LABEL: @select_reverse_splat_false_1(
-; CHECK-NEXT:    [[A_REV:%.*]] = tail call <vscale x 4 x i1> @llvm.experimental.vector.reverse.nxv4i1(<vscale x 4 x i1> [[A:%.*]])
+; CHECK-NEXT:    [[A_REV:%.*]] = tail call <vscale x 4 x i1> @llvm.vector.reverse.nxv4i1(<vscale x 4 x i1> [[A:%.*]])
 ; CHECK-NEXT:    [[C_INSERT:%.*]] = insertelement <vscale x 4 x i32> poison, i32 [[C:%.*]], i64 0
 ; CHECK-NEXT:    [[C_SPLAT:%.*]] = shufflevector <vscale x 4 x i32> [[C_INSERT]], <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer
 ; CHECK-NEXT:    call void @use_nxv4i1(<vscale x 4 x i1> [[A_REV]])
 ; CHECK-NEXT:    [[SELECT1:%.*]] = select <vscale x 4 x i1> [[A]], <vscale x 4 x i32> [[B:%.*]], <vscale x 4 x i32> [[C_SPLAT]]
-; CHECK-NEXT:    [[SELECT:%.*]] = call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> [[SELECT1]])
+; CHECK-NEXT:    [[SELECT:%.*]] = call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> [[SELECT1]])
 ; CHECK-NEXT:    ret <vscale x 4 x i32> [[SELECT]]
 ;
-  %a.rev = tail call <vscale x 4 x i1> @llvm.experimental.vector.reverse.nxv4i1(<vscale x 4 x i1> %a)
-  %b.rev = tail call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> %b)
+  %a.rev = tail call <vscale x 4 x i1> @llvm.vector.reverse.nxv4i1(<vscale x 4 x i1> %a)
+  %b.rev = tail call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> %b)
   %c.insert = insertelement <vscale x 4 x i32> poison, i32 %c, i32 0
   %c.splat = shufflevector <vscale x 4 x i32> %c.insert, <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer
   call void @use_nxv4i1(<vscale x 4 x i1> %a.rev)
@@ -497,16 +497,16 @@ define <vscale x 4 x i32> @select_reverse_splat_false_1(<vscale x 4 x i1> %a, <v
 ; %b.rev has multiple uses
 define <vscale x 4 x i32> @select_reverse_splat_false_2(<vscale x 4 x i1> %a, <vscale x 4 x i32> %b, i32 %c) {
 ; CHECK-LABEL: @select_reverse_splat_false_2(
-; CHECK-NEXT:    [[B_REV:%.*]] = tail call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> [[B:%.*]])
+; CHECK-NEXT:    [[B_REV:%.*]] = tail call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> [[B:%.*]])
 ; CHECK-NEXT:    [[C_INSERT:%.*]] = insertelement <vscale x 4 x i32> poison, i32 [[C:%.*]], i64 0
 ; CHECK-NEXT:    [[C_SPLAT:%.*]] = shufflevector <vscale x 4 x i32> [[C_INSERT]], <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer
 ; CHECK-NEXT:    call void @use_nxv4i32(<vscale x 4 x i32> [[B_REV]])
 ; CHECK-NEXT:    [[SELECT1:%.*]] = select <vscale x 4 x i1> [[A:%.*]], <vscale x 4 x i32> [[B]], <vscale x 4 x i32> [[C_SPLAT]]
-; CHECK-NEXT:    [[SELECT:%.*]] = call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> [[SELECT1]])
+; CHECK-NEXT:    [[SELECT:%.*]] = call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> [[SELECT1]])
 ; CHECK-NEXT:    ret <vscale x 4 x i32> [[SELECT]]
 ;
-  %a.rev = tail call <vscale x 4 x i1> @llvm.experimental.vector.reverse.nxv4i1(<vscale x 4 x i1> %a)
-  %b.rev = tail call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> %b)
+  %a.rev = tail call <vscale x 4 x i1> @llvm.vector.reverse.nxv4i1(<vscale x 4 x i1> %a)
+  %b.rev = tail call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> %b)
   %c.insert = insertelement <vscale x 4 x i32> poison, i32 %c, i32 0
   %c.splat = shufflevector <vscale x 4 x i32> %c.insert, <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer
   call void @use_nxv4i32(<vscale x 4 x i32> %b.rev)
@@ -517,8 +517,8 @@ define <vscale x 4 x i32> @select_reverse_splat_false_2(<vscale x 4 x i1> %a, <v
 ; %a.rev and %b.rev have multiple uses
 define <vscale x 4 x i32> @select_reverse_splat_false_3(<vscale x 4 x i1> %a, <vscale x 4 x i32> %b, i32 %c) {
 ; CHECK-LABEL: @select_reverse_splat_false_3(
-; CHECK-NEXT:    [[A_REV:%.*]] = tail call <vscale x 4 x i1> @llvm.experimental.vector.reverse.nxv4i1(<vscale x 4 x i1> [[A:%.*]])
-; CHECK-NEXT:    [[B_REV:%.*]] = tail call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> [[B:%.*]])
+; CHECK-NEXT:    [[A_REV:%.*]] = tail call <vscale x 4 x i1> @llvm.vector.reverse.nxv4i1(<vscale x 4 x i1> [[A:%.*]])
+; CHECK-NEXT:    [[B_REV:%.*]] = tail call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> [[B:%.*]])
 ; CHECK-NEXT:    [[C_INSERT:%.*]] = insertelement <vscale x 4 x i32> poison, i32 [[C:%.*]], i64 0
 ; CHECK-NEXT:    [[C_SPLAT:%.*]] = shufflevector <vscale x 4 x i32> [[C_INSERT]], <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer
 ; CHECK-NEXT:    call void @use_nxv4i1(<vscale x 4 x i1> [[A_REV]])
@@ -526,8 +526,8 @@ define <vscale x 4 x i32> @select_reverse_splat_false_3(<vscale x 4 x i1> %a, <v
 ; CHECK-NEXT:    [[SELECT:%.*]] = select <vscale x 4 x i1> [[A_REV]], <vscale x 4 x i32> [[B_REV]], <vscale x 4 x i32> [[C_SPLAT]]
 ; CHECK-NEXT:    ret <vscale x 4 x i32> [[SELECT]]
 ;
-  %a.rev = tail call <vscale x 4 x i1> @llvm.experimental.vector.reverse.nxv4i1(<vscale x 4 x i1> %a)
-  %b.rev = tail call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> %b)
+  %a.rev = tail call <vscale x 4 x i1> @llvm.vector.reverse.nxv4i1(<vscale x 4 x i1> %a)
+  %b.rev = tail call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> %b)
   %c.insert = insertelement <vscale x 4 x i32> poison, i32 %c, i32 0
   %c.splat = shufflevector <vscale x 4 x i32> %c.insert, <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer
   call void @use_nxv4i1(<vscale x 4 x i1> %a.rev)
@@ -541,11 +541,11 @@ define <vscale x 4 x i32> @select_reverse_splat_true(<vscale x 4 x i1> %a, <vsca
 ; CHECK-NEXT:    [[C_INSERT:%.*]] = insertelement <vscale x 4 x i32> poison, i32 [[C:%.*]], i64 0
 ; CHECK-NEXT:    [[C_SPLAT:%.*]] = shufflevector <vscale x 4 x i32> [[C_INSERT]], <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer
 ; CHECK-NEXT:    [[SELECT1:%.*]] = select <vscale x 4 x i1> [[A:%.*]], <vscale x 4 x i32> [[C_SPLAT]], <vscale x 4 x i32> [[B:%.*]]
-; CHECK-NEXT:    [[SELECT:%.*]] = call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> [[SELECT1]])
+; CHECK-NEXT:    [[SELECT:%.*]] = call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> [[SELECT1]])
 ; CHECK-NEXT:    ret <vscale x 4 x i32> [[SELECT]]
 ;
-  %a.rev = tail call <vscale x 4 x i1> @llvm.experimental.vector.reverse.nxv4i1(<vscale x 4 x i1> %a)
-  %b.rev = tail call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> %b)
+  %a.rev = tail call <vscale x 4 x i1> @llvm.vector.reverse.nxv4i1(<vscale x 4 x i1> %a)
+  %b.rev = tail call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> %b)
   %c.insert = insertelement <vscale x 4 x i32> poison, i32 %c, i32 0
   %c.splat = shufflevector <vscale x 4 x i32> %c.insert, <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer
   %select = select <vscale x 4 x i1> %a.rev, <vscale x 4 x i32> %c.splat, <vscale x 4 x i32> %b.rev
@@ -555,16 +555,16 @@ define <vscale x 4 x i32> @select_reverse_splat_true(<vscale x 4 x i1> %a, <vsca
 ; %a.rev has multiple uses
 define <vscale x 4 x i32> @select_reverse_splat_true_1(<vscale x 4 x i1> %a, <vscale x 4 x i32> %b, i32 %c) {
 ; CHECK-LABEL: @select_reverse_splat_true_1(
-; CHECK-NEXT:    [[A_REV:%.*]] = tail call <vscale x 4 x i1> @llvm.experimental.vector.reverse.nxv4i1(<vscale x 4 x i1> [[A:%.*]])
+; CHECK-NEXT:    [[A_REV:%.*]] = tail call <vscale x 4 x i1> @llvm.vector.reverse.nxv4i1(<vscale x 4 x i1> [[A:%.*]])
 ; CHECK-NEXT:    [[C_INSERT:%.*]] = insertelement <vscale x 4 x i32> poison, i32 [[C:%.*]], i64 0
 ; CHECK-NEXT:    [[C_SPLAT:%.*]] = shufflevector <vscale x 4 x i32> [[C_INSERT]], <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer
 ; CHECK-NEXT:    call void @use_nxv4i1(<vscale x 4 x i1> [[A_REV]])
 ; CHECK-NEXT:    [[SELECT1:%.*]] = select <vscale x 4 x i1> [[A]], <vscale x 4 x i32> [[C_SPLAT]], <vscale x 4 x i32> [[B:%.*]]
-; CHECK-NEXT:    [[SELECT:%.*]] = call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> [[SELECT1]])
+; CHECK-NEXT:    [[SELECT:%.*]] = call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> [[SELECT1]])
 ; CHECK-NEXT:    ret <vscale x 4 x i32> [[SELECT]]
 ;
-  %a.rev = tail call <vscale x 4 x i1> @llvm.experimental.vector.reverse.nxv4i1(<vscale x 4 x i1> %a)
-  %b.rev = tail call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> %b)
+  %a.rev = tail call <vscale x 4 x i1> @llvm.vector.reverse.nxv4i1(<vscale x 4 x i1> %a)
+  %b.rev = tail call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> %b)
   %c.insert = insertelement <vscale x 4 x i32> poison, i32 %c, i32 0
   %c.splat = shufflevector <vscale x 4 x i32> %c.insert, <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer
   call void @use_nxv4i1(<vscale x 4 x i1> %a.rev)
@@ -575,16 +575,16 @@ define <vscale x 4 x i32> @select_reverse_splat_true_1(<vscale x 4 x i1> %a, <vs
 ; %b.rev has multiple uses
 define <vscale x 4 x i32> @select_reverse_splat_true_2(<vscale x 4 x i1> %a, <vscale x 4 x i32> %b, i32 %c) {
 ; CHECK-LABEL: @select_reverse_splat_true_2(
-; CHECK-NEXT:    [[B_REV:%.*]] = tail call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> [[B:%.*]])
+; CHECK-NEXT:    [[B_REV:%.*]] = tail call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> [[B:%.*]])
 ; CHECK-NEXT:    [[C_INSERT:%.*]] = insertelement <vscale x 4 x i32> poison, i32 [[C:%.*]], i64 0
 ; CHECK-NEXT:    [[C_SPLAT:%.*]] = shufflevector <vscale x 4 x i32> [[C_INSERT]], <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer
 ; CHECK-NEXT:    call void @use_nxv4i32(<vscale x 4 x i32> [[B_REV]])
 ; CHECK-NEXT:    [[SELECT1:%.*]] = select <vscale x 4 x i1> [[A:%.*]], <vscale x 4 x i32> [[C_SPLAT]], <vscale x 4 x i32> [[B]]
-; CHECK-NEXT:    [[SELECT:%.*]] = call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> [[SELECT1]])
+; CHECK-NEXT:    [[SELECT:%.*]] = call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> [[SELECT1]])
 ; CHECK-NEXT:    ret <vscale x 4 x i32> [[SELECT]]
 ;
-  %a.rev = tail call <vscale x 4 x i1> @llvm.experimental.vector.reverse.nxv4i1(<vscale x 4 x i1> %a)
-  %b.rev = tail call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> %b)
+  %a.rev = tail call <vscale x 4 x i1> @llvm.vector.reverse.nxv4i1(<vscale x 4 x i1> %a)
+  %b.rev = tail call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> %b)
   %c.insert = insertelement <vscale x 4 x i32> poison, i32 %c, i32 0
   %c.splat = shufflevector <vscale x 4 x i32> %c.insert, <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer
   call void @use_nxv4i32(<vscale x 4 x i32> %b.rev)
@@ -595,8 +595,8 @@ define <vscale x 4 x i32> @select_reverse_splat_true_2(<vscale x 4 x i1> %a, <vs
 ; %a.rev and %b.rev have multiple uses
 define <vscale x 4 x i32> @select_reverse_splat_true_3(<vscale x 4 x i1> %a, <vscale x 4 x i32> %b, i32 %c) {
 ; CHECK-LABEL: @select_reverse_splat_true_3(
-; CHECK-NEXT:    [[A_REV:%.*]] = tail call <vscale x 4 x i1> @llvm.experimental.vector.reverse.nxv4i1(<vscale x 4 x i1> [[A:%.*]])
-; CHECK-NEXT:    [[B_REV:%.*]] = tail call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> [[B:%.*]])
+; CHECK-NEXT:    [[A_REV:%.*]] = tail call <vscale x 4 x i1> @llvm.vector.reverse.nxv4i1(<vscale x 4 x i1> [[A:%.*]])
+; CHECK-NEXT:    [[B_REV:%.*]] = tail call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> [[B:%.*]])
 ; CHECK-NEXT:    [[C_INSERT:%.*]] = insertelement <vscale x 4 x i32> poison, i32 [[C:%.*]], i64 0
 ; CHECK-NEXT:    [[C_SPLAT:%.*]] = shufflevector <vscale x 4 x i32> [[C_INSERT]], <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer
 ; CHECK-NEXT:    call void @use_nxv4i1(<vscale x 4 x i1> [[A_REV]])
@@ -604,8 +604,8 @@ define <vscale x 4 x i32> @select_reverse_splat_true_3(<vscale x 4 x i1> %a, <vs
 ; CHECK-NEXT:    [[SELECT:%.*]] = select <vscale x 4 x i1> [[A_REV]], <vscale x 4 x i32> [[C_SPLAT]], <vscale x 4 x i32> [[B_REV]]
 ; CHECK-NEXT:    ret <vscale x 4 x i32> [[SELECT]]
 ;
-  %a.rev = tail call <vscale x 4 x i1> @llvm.experimental.vector.reverse.nxv4i1(<vscale x 4 x i1> %a)
-  %b.rev = tail call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> %b)
+  %a.rev = tail call <vscale x 4 x i1> @llvm.vector.reverse.nxv4i1(<vscale x 4 x i1> %a)
+  %b.rev = tail call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> %b)
   %c.insert = insertelement <vscale x 4 x i32> poison, i32 %c, i32 0
   %c.splat = shufflevector <vscale x 4 x i32> %c.insert, <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer
   call void @use_nxv4i1(<vscale x 4 x i1> %a.rev)
@@ -622,10 +622,10 @@ define <vscale x 4 x float> @reverse_binop_reverse(<vscale x 4 x float> %a, <vsc
 ; CHECK-NEXT:    [[ADD1:%.*]] = fadd <vscale x 4 x float> [[A:%.*]], [[B:%.*]]
 ; CHECK-NEXT:    ret <vscale x 4 x float> [[ADD1]]
 ;
-  %a.rev = tail call <vscale x 4 x float> @llvm.experimental.vector.reverse.nxv4f32(<vscale x 4 x float> %a)
-  %b.rev = tail call <vscale x 4 x float> @llvm.experimental.vector.reverse.nxv4f32(<vscale x 4 x float> %b)
+  %a.rev = tail call <vscale x 4 x float> @llvm.vector.reverse.nxv4f32(<vscale x 4 x float> %a)
+  %b.rev = tail call <vscale x 4 x float> @llvm.vector.reverse.nxv4f32(<vscale x 4 x float> %b)
   %add = fadd <vscale x 4 x float> %a.rev, %b.rev
-  %add.rev = tail call <vscale x 4 x float> @llvm.experimental.vector.reverse.nxv4f32(<vscale x 4 x float> %add)
+  %add.rev = tail call <vscale x 4 x float> @llvm.vector.reverse.nxv4f32(<vscale x 4 x float> %add)
   ret <vscale x 4 x float> %add.rev
 }
 
@@ -636,11 +636,11 @@ define <vscale x 4 x float> @reverse_binop_reverse_splat_RHS(<vscale x 4 x float
 ; CHECK-NEXT:    [[DIV1:%.*]] = fdiv <vscale x 4 x float> [[A:%.*]], [[B_SPLAT]]
 ; CHECK-NEXT:    ret <vscale x 4 x float> [[DIV1]]
 ;
-  %a.rev = tail call <vscale x 4 x float> @llvm.experimental.vector.reverse.nxv4f32(<vscale x 4 x float> %a)
+  %a.rev = tail call <vscale x 4 x float> @llvm.vector.reverse.nxv4f32(<vscale x 4 x float> %a)
   %b.insert = insertelement <vscale x 4 x float> poison, float %b, i32 0
   %b.splat = shufflevector <vscale x 4 x float> %b.insert, <vscale x 4 x float> poison, <vscale x 4 x i32> zeroinitializer
   %div = fdiv <vscale x 4 x float> %a.rev, %b.splat
-  %div.rev = tail call <vscale x 4 x float> @llvm.experimental.vector.reverse.nxv4f32(<vscale x 4 x float> %div)
+  %div.rev = tail call <vscale x 4 x float> @llvm.vector.reverse.nxv4f32(<vscale x 4 x float> %div)
   ret <vscale x 4 x float> %div.rev
 }
 
@@ -651,11 +651,11 @@ define <vscale x 4 x float> @reverse_binop_reverse_splat_LHS(<vscale x 4 x float
 ; CHECK-NEXT:    [[DIV1:%.*]] = fdiv <vscale x 4 x float> [[B_SPLAT]], [[A:%.*]]
 ; CHECK-NEXT:    ret <vscale x 4 x float> [[DIV1]]
 ;
-  %a.rev = tail call <vscale x 4 x float> @llvm.experimental.vector.reverse.nxv4f32(<vscale x 4 x float> %a)
+  %a.rev = tail call <vscale x 4 x float> @llvm.vector.reverse.nxv4f32(<vscale x 4 x float> %a)
   %b.insert = insertelement <vscale x 4 x float> poison, float %b, i32 0
   %b.splat = shufflevector <vscale x 4 x float> %b.insert, <vscale x 4 x float> poison, <vscale x 4 x i32> zeroinitializer
   %div = fdiv <vscale x 4 x float> %b.splat, %a.rev
-  %div.rev = tail call <vscale x 4 x float> @llvm.experimental.vector.reverse.nxv4f32(<vscale x 4 x float> %div)
+  %div.rev = tail call <vscale x 4 x float> @llvm.vector.reverse.nxv4f32(<vscale x 4 x float> %div)
   ret <vscale x 4 x float> %div.rev
 }
 
@@ -664,10 +664,10 @@ define <vscale x 4 x i1> @reverse_fcmp_reverse(<vscale x 4 x float> %a, <vscale
 ; CHECK-NEXT:    [[CMP1:%.*]] = fcmp fast olt <vscale x 4 x float> [[A:%.*]], [[B:%.*]]
 ; CHECK-NEXT:    ret <vscale x 4 x i1> [[CMP1]]
 ;
-  %a.rev = tail call <vscale x 4 x float> @llvm.experimental.vector.reverse.nxv4f32(<vscale x 4 x float> %a)
-  %b.rev = tail call <vscale x 4 x float> @llvm.experimental.vector.reverse.nxv4f32(<vscale x 4 x float> %b)
+  %a.rev = tail call <vscale x 4 x float> @llvm.vector.reverse.nxv4f32(<vscale x 4 x float> %a)
+  %b.rev = tail call <vscale x 4 x float> @llvm.vector.reverse.nxv4f32(<vscale x 4 x float> %b)
   %cmp = fcmp fast olt <vscale x 4 x float> %a.rev, %b.rev
-  %cmp.rev = tail call <vscale x 4 x i1> @llvm.experimental.vector.reverse.nxv4i1(<vscale x 4 x i1> %cmp)
+  %cmp.rev = tail call <vscale x 4 x i1> @llvm.vector.reverse.nxv4i1(<vscale x 4 x i1> %cmp)
   ret <vscale x 4 x i1> %cmp.rev
 }
 
@@ -676,11 +676,11 @@ define <vscale x 4 x float> @reverse_select_reverse(<vscale x 4 x i1> %a, <vscal
 ; CHECK-NEXT:    [[SELECT1:%.*]] = select fast <vscale x 4 x i1> [[A:%.*]], <vscale x 4 x float> [[B:%.*]], <vscale x 4 x float> [[C:%.*]]
 ; CHECK-NEXT:    ret <vscale x 4 x float> [[SELECT1]]
 ;
-  %a.rev = tail call <vscale x 4 x i1> @llvm.experimental.vector.reverse.nxv4i1(<vscale x 4 x i1> %a)
-  %b.rev = tail call <vscale x 4 x float> @llvm.experimental.vector.reverse.nxv4f32(<vscale x 4 x float> %b)
-  %c.rev = tail call <vscale x 4 x float> @llvm.experimental.vector.reverse.nxv4f32(<vscale x 4 x float> %c)
+  %a.rev = tail call <vscale x 4 x i1> @llvm.vector.reverse.nxv4i1(<vscale x 4 x i1> %a)
+  %b.rev = tail call <vscale x 4 x float> @llvm.vector.reverse.nxv4f32(<vscale x 4 x float> %b)
+  %c.rev = tail call <vscale x 4 x float> @llvm.vector.reverse.nxv4f32(<vscale x 4 x float> %c)
   %select = select fast <vscale x 4 x i1> %a.rev, <vscale x 4 x float> %b.rev, <vscale x 4 x float> %c.rev
-  %select.rev = tail call <vscale x 4 x float> @llvm.experimental.vector.reverse.nxv4f32(<vscale x 4 x float> %select)
+  %select.rev = tail call <vscale x 4 x float> @llvm.vector.reverse.nxv4f32(<vscale x 4 x float> %select)
   ret <vscale x 4 x float> %select.rev
 }
 
@@ -689,9 +689,9 @@ define <vscale x 4 x float> @reverse_unop_reverse(<vscale x 4 x float> %a) {
 ; CHECK-NEXT:    [[NEG1:%.*]] = fneg <vscale x 4 x float> [[A:%.*]]
 ; CHECK-NEXT:    ret <vscale x 4 x float> [[NEG1]]
 ;
-  %a.rev = tail call <vscale x 4 x float> @llvm.experimental.vector.reverse.nxv4f32(<vscale x 4 x float> %a)
+  %a.rev = tail call <vscale x 4 x float> @llvm.vector.reverse.nxv4f32(<vscale x 4 x float> %a)
   %neg = fneg <vscale x 4 x float> %a.rev
-  %neg.rev = tail call <vscale x 4 x float> @llvm.experimental.vector.reverse.nxv4f32(<vscale x 4 x float> %neg)
+  %neg.rev = tail call <vscale x 4 x float> @llvm.vector.reverse.nxv4f32(<vscale x 4 x float> %neg)
   ret <vscale x 4 x float> %neg.rev
 }
 
@@ -700,6 +700,6 @@ declare void @use_nxv4i1(<vscale x 4 x i1>)
 declare void @use_nxv4i32(<vscale x 4 x i32>)
 declare void @use_nxv4f32(<vscale x 4 x float>)
 
-declare <vscale x 4 x i1> @llvm.experimental.vector.reverse.nxv4i1(<vscale x 4 x i1>)
-declare <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32>)
-declare <vscale x 4 x float> @llvm.experimental.vector.reverse.nxv4f32(<vscale x 4 x float>)
+declare <vscale x 4 x i1> @llvm.vector.reverse.nxv4i1(<vscale x 4 x i1>)
+declare <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32>)
+declare <vscale x 4 x float> @llvm.vector.reverse.nxv4f32(<vscale x 4 x float>)
diff --git a/llvm/test/Transforms/InstCombine/xor.ll b/llvm/test/Transforms/InstCombine/xor.ll
index ba9e992a9443..9a59db40ef8b 100644
--- a/llvm/test/Transforms/InstCombine/xor.ll
+++ b/llvm/test/Transforms/InstCombine/xor.ll
@@ -1336,7 +1336,7 @@ define i32 @xor_orn_2use(i32 %a, i32 %b, ptr %s1, ptr %s2) {
 
 define i32 @ctlz_pow2(i32 %x) {
 ; CHECK-LABEL: @ctlz_pow2(
-; CHECK-NEXT:    [[R:%.*]] = call i32 @llvm.cttz.i32(i32 [[X:%.*]], i1 true), !range [[RNG0:![0-9]+]]
+; CHECK-NEXT:    [[R:%.*]] = call range(i32 0, 33) i32 @llvm.cttz.i32(i32 [[X:%.*]], i1 true)
 ; CHECK-NEXT:    ret i32 [[R]]
 ;
   %n = sub i32 0, %x
@@ -1352,7 +1352,7 @@ define <2 x i8> @cttz_pow2(<2 x i8> %x, <2 x i8> %y) {
 ; CHECK-LABEL: @cttz_pow2(
 ; CHECK-NEXT:    [[S:%.*]] = shl nuw <2 x i8> <i8 1, i8 1>, [[X:%.*]]
 ; CHECK-NEXT:    [[D:%.*]] = udiv exact <2 x i8> [[S]], [[Y:%.*]]
-; CHECK-NEXT:    [[R:%.*]] = call <2 x i8> @llvm.ctlz.v2i8(<2 x i8> [[D]], i1 true), !range [[RNG1:![0-9]+]]
+; CHECK-NEXT:    [[R:%.*]] = call range(i8 0, 9) <2 x i8> @llvm.ctlz.v2i8(<2 x i8> [[D]], i1 true)
 ; CHECK-NEXT:    ret <2 x i8> [[R]]
 ;
   %s = shl <2 x i8> <i8 1, i8 1>, %x
@@ -1368,7 +1368,7 @@ define i32 @ctlz_pow2_or_zero(i32 %x) {
 ; CHECK-LABEL: @ctlz_pow2_or_zero(
 ; CHECK-NEXT:    [[N:%.*]] = sub i32 0, [[X:%.*]]
 ; CHECK-NEXT:    [[A:%.*]] = and i32 [[N]], [[X]]
-; CHECK-NEXT:    [[Z:%.*]] = call i32 @llvm.ctlz.i32(i32 [[A]], i1 false), !range [[RNG0]]
+; CHECK-NEXT:    [[Z:%.*]] = call range(i32 0, 33) i32 @llvm.ctlz.i32(i32 [[A]], i1 false)
 ; CHECK-NEXT:    [[R:%.*]] = xor i32 [[Z]], 31
 ; CHECK-NEXT:    ret i32 [[R]]
 ;
@@ -1385,7 +1385,7 @@ define i32 @ctlz_pow2_wrong_const(i32 %x) {
 ; CHECK-LABEL: @ctlz_pow2_wrong_const(
 ; CHECK-NEXT:    [[N:%.*]] = sub i32 0, [[X:%.*]]
 ; CHECK-NEXT:    [[A:%.*]] = and i32 [[N]], [[X]]
-; CHECK-NEXT:    [[Z:%.*]] = call i32 @llvm.ctlz.i32(i32 [[A]], i1 true), !range [[RNG0]]
+; CHECK-NEXT:    [[Z:%.*]] = call range(i32 0, 33) i32 @llvm.ctlz.i32(i32 [[A]], i1 true)
 ; CHECK-NEXT:    [[R:%.*]] = xor i32 [[Z]], 30
 ; CHECK-NEXT:    ret i32 [[R]]
 ;
diff --git a/llvm/test/Transforms/InstCombine/zext-ctlz-trunc-to-ctlz-add.ll b/llvm/test/Transforms/InstCombine/zext-ctlz-trunc-to-ctlz-add.ll
index c8eb513a8440..384ff8d2b7a3 100644
--- a/llvm/test/Transforms/InstCombine/zext-ctlz-trunc-to-ctlz-add.ll
+++ b/llvm/test/Transforms/InstCombine/zext-ctlz-trunc-to-ctlz-add.ll
@@ -13,7 +13,7 @@ declare void @use1(<vscale x 2 x i63>)
 
 define i16 @trunc_ctlz_zext_i16_i32(i16 %x) {
 ; CHECK-LABEL: @trunc_ctlz_zext_i16_i32(
-; CHECK-NEXT:    [[TMP1:%.*]] = call i16 @llvm.ctlz.i16(i16 [[X:%.*]], i1 false), !range [[RNG0:![0-9]+]]
+; CHECK-NEXT:    [[TMP1:%.*]] = call range(i16 0, 17) i16 @llvm.ctlz.i16(i16 [[X:%.*]], i1 false)
 ; CHECK-NEXT:    [[ZZ:%.*]] = add nuw nsw i16 [[TMP1]], 16
 ; CHECK-NEXT:    ret i16 [[ZZ]]
 ;
@@ -27,7 +27,7 @@ define i16 @trunc_ctlz_zext_i16_i32(i16 %x) {
 
 define <2 x i8> @trunc_ctlz_zext_v2i8_v2i33(<2 x i8> %x) {
 ; CHECK-LABEL: @trunc_ctlz_zext_v2i8_v2i33(
-; CHECK-NEXT:    [[TMP1:%.*]] = call <2 x i8> @llvm.ctlz.v2i8(<2 x i8> [[X:%.*]], i1 true), !range [[RNG1:![0-9]+]]
+; CHECK-NEXT:    [[TMP1:%.*]] = call range(i8 0, 9) <2 x i8> @llvm.ctlz.v2i8(<2 x i8> [[X:%.*]], i1 true)
 ; CHECK-NEXT:    [[ZZ:%.*]] = add nuw nsw <2 x i8> [[TMP1]], <i8 25, i8 25>
 ; CHECK-NEXT:    ret <2 x i8> [[ZZ]]
 ;
@@ -41,7 +41,7 @@ define <2 x i8> @trunc_ctlz_zext_v2i8_v2i33(<2 x i8> %x) {
 
 define <vscale x 2 x i16> @trunc_ctlz_zext_nxv2i16_nxv2i64(<vscale x 2 x i16> %x) {
 ; CHECK-LABEL: @trunc_ctlz_zext_nxv2i16_nxv2i64(
-; CHECK-NEXT:    [[TMP1:%.*]] = call <vscale x 2 x i16> @llvm.ctlz.nxv2i16(<vscale x 2 x i16> [[X:%.*]], i1 false), !range [[RNG0]]
+; CHECK-NEXT:    [[TMP1:%.*]] = call range(i16 0, 17) <vscale x 2 x i16> @llvm.ctlz.nxv2i16(<vscale x 2 x i16> [[X:%.*]], i1 false)
 ; CHECK-NEXT:    [[ZZ:%.*]] = add nuw nsw <vscale x 2 x i16> [[TMP1]], shufflevector (<vscale x 2 x i16> insertelement (<vscale x 2 x i16> poison, i16 48, i64 0), <vscale x 2 x i16> poison, <vscale x 2 x i32> zeroinitializer)
 ; CHECK-NEXT:    ret <vscale x 2 x i16> [[ZZ]]
 ;
@@ -56,7 +56,7 @@ define <vscale x 2 x i16> @trunc_ctlz_zext_nxv2i16_nxv2i64(<vscale x 2 x i16> %x
 define <2 x i17> @trunc_ctlz_zext_v2i17_v2i32_multiple_uses(<2 x i17> %x) {
 ; CHECK-LABEL: @trunc_ctlz_zext_v2i17_v2i32_multiple_uses(
 ; CHECK-NEXT:    [[Z:%.*]] = zext <2 x i17> [[X:%.*]] to <2 x i32>
-; CHECK-NEXT:    [[P:%.*]] = call <2 x i32> @llvm.ctlz.v2i32(<2 x i32> [[Z]], i1 false), !range [[RNG2:![0-9]+]]
+; CHECK-NEXT:    [[P:%.*]] = call range(i32 15, 33) <2 x i32> @llvm.ctlz.v2i32(<2 x i32> [[Z]], i1 false)
 ; CHECK-NEXT:    [[ZZ:%.*]] = trunc nuw nsw <2 x i32> [[P]] to <2 x i17>
 ; CHECK-NEXT:    call void @use(<2 x i32> [[P]])
 ; CHECK-NEXT:    ret <2 x i17> [[ZZ]]
@@ -73,7 +73,7 @@ define <2 x i17> @trunc_ctlz_zext_v2i17_v2i32_multiple_uses(<2 x i17> %x) {
 define <vscale x 2 x i16> @trunc_ctlz_zext_nxv2i16_nxv2i63_multiple_uses(<vscale x 2 x i16> %x) {
 ; CHECK-LABEL: @trunc_ctlz_zext_nxv2i16_nxv2i63_multiple_uses(
 ; CHECK-NEXT:    [[Z:%.*]] = zext <vscale x 2 x i16> [[X:%.*]] to <vscale x 2 x i63>
-; CHECK-NEXT:    [[TMP1:%.*]] = call <vscale x 2 x i16> @llvm.ctlz.nxv2i16(<vscale x 2 x i16> [[X]], i1 true), !range [[RNG0]]
+; CHECK-NEXT:    [[TMP1:%.*]] = call range(i16 0, 17) <vscale x 2 x i16> @llvm.ctlz.nxv2i16(<vscale x 2 x i16> [[X]], i1 true)
 ; CHECK-NEXT:    [[ZZ:%.*]] = add nuw nsw <vscale x 2 x i16> [[TMP1]], shufflevector (<vscale x 2 x i16> insertelement (<vscale x 2 x i16> poison, i16 47, i64 0), <vscale x 2 x i16> poison, <vscale x 2 x i32> zeroinitializer)
 ; CHECK-NEXT:    call void @use1(<vscale x 2 x i63> [[Z]])
 ; CHECK-NEXT:    ret <vscale x 2 x i16> [[ZZ]]
@@ -90,7 +90,7 @@ define <vscale x 2 x i16> @trunc_ctlz_zext_nxv2i16_nxv2i63_multiple_uses(<vscale
 define i16 @trunc_ctlz_zext_i10_i32(i10 %x) {
 ; CHECK-LABEL: @trunc_ctlz_zext_i10_i32(
 ; CHECK-NEXT:    [[Z:%.*]] = zext i10 [[X:%.*]] to i32
-; CHECK-NEXT:    [[P:%.*]] = call i32 @llvm.ctlz.i32(i32 [[Z]], i1 false), !range [[RNG3:![0-9]+]]
+; CHECK-NEXT:    [[P:%.*]] = call range(i32 22, 33) i32 @llvm.ctlz.i32(i32 [[Z]], i1 false)
 ; CHECK-NEXT:    [[ZZ:%.*]] = trunc nuw nsw i32 [[P]] to i16
 ; CHECK-NEXT:    ret i16 [[ZZ]]
 ;
@@ -108,7 +108,7 @@ define i16 @trunc_ctlz_zext_i10_i32(i10 %x) {
 define i3 @trunc_ctlz_zext_i3_i34(i3 %x) {
 ; CHECK-LABEL: @trunc_ctlz_zext_i3_i34(
 ; CHECK-NEXT:    [[Z:%.*]] = zext i3 [[X:%.*]] to i34
-; CHECK-NEXT:    [[P:%.*]] = call i34 @llvm.ctlz.i34(i34 [[Z]], i1 false), !range [[RNG4:![0-9]+]]
+; CHECK-NEXT:    [[P:%.*]] = call range(i34 31, 35) i34 @llvm.ctlz.i34(i34 [[Z]], i1 false)
 ; CHECK-NEXT:    [[T:%.*]] = trunc i34 [[P]] to i3
 ; CHECK-NEXT:    ret i3 [[T]]
 ;
diff --git a/llvm/test/Transforms/InstSimplify/ConstProp/gep-alias.ll b/llvm/test/Transforms/InstSimplify/ConstProp/gep-alias.ll
index bf0969162e27..f77a49e90be7 100644
--- a/llvm/test/Transforms/InstSimplify/ConstProp/gep-alias.ll
+++ b/llvm/test/Transforms/InstSimplify/ConstProp/gep-alias.ll
@@ -1,17 +1,20 @@
-; RUN: opt -passes=instcombine -S -o - %s | FileCheck %s

-; Test that we don't replace an alias with its aliasee when simplifying GEPs.

-; In this test case the transformation is invalid because it replaces the

-; reference to the symbol "b" (which refers to whichever instance of "b"

-; was chosen by the linker) with a reference to "a" (which refers to the

-; specific instance of "b" in this module).

-

-target datalayout = "e-m:e-i64:64-f80:128-n8:16:32:64-S128"

-target triple = "x86_64-unknown-linux-gnu"

-

-@a = internal global [3 x ptr] zeroinitializer

-@b = linkonce_odr alias [3 x ptr], ptr @a

-

-define ptr @f() {

-  ; CHECK: ret ptr getelementptr ([3 x ptr], ptr @b, i64 0, i64 1)

-  ret ptr getelementptr ([3 x ptr], ptr @b, i64 0, i64 1)

-}

+; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --version 4
+; RUN: opt -passes=instcombine -S -o - %s | FileCheck %s
+; Test that we don't replace an alias with its aliasee when simplifying GEPs.
+; In this test case the transformation is invalid because it replaces the
+; reference to the symbol "b" (which refers to whichever instance of "b"
+; was chosen by the linker) with a reference to "a" (which refers to the
+; specific instance of "b" in this module).
+
+target datalayout = "e-m:e-i64:64-f80:128-n8:16:32:64-S128"
+target triple = "x86_64-unknown-linux-gnu"
+
+@a = internal global [3 x ptr] zeroinitializer
+@b = linkonce_odr alias [3 x ptr], ptr @a
+
+define ptr @f() {
+; CHECK-LABEL: define ptr @f() {
+; CHECK-NEXT:    ret ptr getelementptr ([3 x ptr], ptr @b, i64 0, i64 1)
+;
+  ret ptr getelementptr ([3 x ptr], ptr @b, i64 0, i64 1)
+}
diff --git a/llvm/test/Transforms/InstSimplify/named-vector-shuffle-reverse.ll b/llvm/test/Transforms/InstSimplify/named-vector-shuffle-reverse.ll
index a26f0a9d87f8..25e99ff0e715 100644
--- a/llvm/test/Transforms/InstSimplify/named-vector-shuffle-reverse.ll
+++ b/llvm/test/Transforms/InstSimplify/named-vector-shuffle-reverse.ll
@@ -6,8 +6,8 @@ define <vscale x 4 x i32> @shuffle_b2b_reverse(<vscale x 4 x i32> %a) {
 ; CHECK-LABEL: @shuffle_b2b_reverse(
 ; CHECK-NEXT:    ret <vscale x 4 x i32> [[A:%.*]]
 ;
-  %rev = tail call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> %a)
-  %rev.rev = tail call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> %rev)
+  %rev = tail call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> %a)
+  %rev.rev = tail call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> %rev)
   ret <vscale x 4 x i32> %rev.rev
 }
 
@@ -20,8 +20,8 @@ define <vscale x 4 x i32> @splat_reverse(i32 %a) {
 ;
   %splat_insert = insertelement <vscale x 4 x i32> poison, i32 %a, i32 0
   %splat = shufflevector <vscale x 4 x i32> %splat_insert, <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer
-  %rev = tail call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> %splat)
+  %rev = tail call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> %splat)
   ret <vscale x 4 x i32> %rev
 }
 
-declare <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32>)
+declare <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32>)
diff --git a/llvm/test/Transforms/InstSimplify/past-the-end.ll b/llvm/test/Transforms/InstSimplify/past-the-end.ll
index 3138130e353b..96339c1cdcf2 100644
--- a/llvm/test/Transforms/InstSimplify/past-the-end.ll
+++ b/llvm/test/Transforms/InstSimplify/past-the-end.ll
@@ -1,4 +1,4 @@
-; NOTE: Assertions have been autogenerated by update_test_checks.py
+; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
 ; RUN: opt < %s -passes=instsimplify -S | FileCheck %s
 target datalayout = "p:32:32"
 
@@ -11,7 +11,7 @@ target datalayout = "p:32:32"
 
 define zeroext i1 @no_offsets() {
 ; CHECK-LABEL: @no_offsets(
-; CHECK:         ret i1 false
+; CHECK-NEXT:    ret i1 false
 ;
   %t = icmp eq ptr @opte_a, @opte_b
   ret i1 %t
@@ -21,7 +21,7 @@ define zeroext i1 @no_offsets() {
 
 define zeroext i1 @both_past_the_end() {
 ; CHECK-LABEL: @both_past_the_end(
-; CHECK:         ret i1 icmp eq (ptr getelementptr inbounds (i32, ptr @opte_a, i32 1), ptr getelementptr inbounds (i32, ptr @opte_b, i32 1))
+; CHECK-NEXT:    ret i1 icmp eq (ptr getelementptr inbounds (i32, ptr @opte_a, i32 1), ptr getelementptr inbounds (i32, ptr @opte_b, i32 1))
 ;
   %x = getelementptr i32, ptr @opte_a, i32 1
   %y = getelementptr i32, ptr @opte_b, i32 1
@@ -35,7 +35,7 @@ define zeroext i1 @both_past_the_end() {
 
 define zeroext i1 @just_one_past_the_end() {
 ; CHECK-LABEL: @just_one_past_the_end(
-; CHECK:         ret i1 icmp eq (ptr getelementptr inbounds (i32, ptr @opte_a, i32 1), ptr @opte_b)
+; CHECK-NEXT:    ret i1 icmp eq (ptr getelementptr inbounds (i32, ptr @opte_a, i32 1), ptr @opte_b)
 ;
   %x = getelementptr i32, ptr @opte_a, i32 1
   %t = icmp eq ptr %x, @opte_b
@@ -46,7 +46,7 @@ define zeroext i1 @just_one_past_the_end() {
 
 define zeroext i1 @no_alloca_offsets() {
 ; CHECK-LABEL: @no_alloca_offsets(
-; CHECK:         ret i1 false
+; CHECK-NEXT:    ret i1 false
 ;
   %m = alloca i32
   %n = alloca i32
@@ -73,8 +73,8 @@ define zeroext i1 @both_past_the_end_alloca() {
 
 define zeroext i1 @just_one_past_the_end_alloca() {
 ; CHECK-LABEL: @just_one_past_the_end_alloca(
-; CHECK:         [[M:%.*]] = alloca i32
-; CHECK-NEXT:    [[N:%.*]] = alloca i32
+; CHECK-NEXT:    [[M:%.*]] = alloca i32, align 4
+; CHECK-NEXT:    [[N:%.*]] = alloca i32, align 4
 ; CHECK-NEXT:    [[X:%.*]] = getelementptr i32, ptr [[M]], i32 1
 ; CHECK-NEXT:    [[T:%.*]] = icmp eq ptr [[X]], [[N]]
 ; CHECK-NEXT:    ret i1 [[T]]
diff --git a/llvm/test/Transforms/InstSimplify/select.ll b/llvm/test/Transforms/InstSimplify/select.ll
index 40c1460e3ebc..4eb6491eec5a 100644
--- a/llvm/test/Transforms/InstSimplify/select.ll
+++ b/llvm/test/Transforms/InstSimplify/select.ll
@@ -1105,19 +1105,19 @@ define <2 x i32> @select_ctpop_zero_vec(<2 x i32> %x) {
 define <2 x i32> @select_vector_reverse(<2 x i32> %x) {
 ; CHECK-LABEL: @select_vector_reverse(
 ; CHECK-NEXT:    [[CMP:%.*]] = icmp eq <2 x i32> [[X:%.*]], zeroinitializer
-; CHECK-NEXT:    [[REV:%.*]] = call <2 x i32> @llvm.experimental.vector.reverse.v2i32(<2 x i32> [[X]])
+; CHECK-NEXT:    [[REV:%.*]] = call <2 x i32> @llvm.vector.reverse.v2i32(<2 x i32> [[X]])
 ; CHECK-NEXT:    [[SEL:%.*]] = select <2 x i1> [[CMP]], <2 x i32> zeroinitializer, <2 x i32> [[REV]]
 ; CHECK-NEXT:    ret <2 x i32> [[SEL]]
 ;
   %cmp = icmp eq <2 x i32> %x, zeroinitializer
-  %rev = call <2 x i32> @llvm.experimental.vector.reverse.v2i32(<2 x i32> %x)
+  %rev = call <2 x i32> @llvm.vector.reverse.v2i32(<2 x i32> %x)
   %sel = select <2 x i1> %cmp, <2 x i32> zeroinitializer, <2 x i32> %rev
   ret <2 x i32> %sel
 }
 
 declare i32 @llvm.ctpop.i32(i32)
 declare <2 x i32> @llvm.ctpop.v2i32(<2 x i32>)
-declare <2 x i32> @llvm.experimental.vector.reverse.v2i32(<2 x i32>)
+declare <2 x i32> @llvm.vector.reverse.v2i32(<2 x i32>)
 
 define <2 x i32> @vec_select_no_equivalence(<2 x i32> %x, <2 x i32> %y) {
 ; CHECK-LABEL: @vec_select_no_equivalence(
diff --git a/llvm/test/Transforms/InterleavedAccess/AArch64/fixed-deinterleave-intrinsics.ll b/llvm/test/Transforms/InterleavedAccess/AArch64/fixed-deinterleave-intrinsics.ll
index 224a0693bf21..54348d1e2a48 100644
--- a/llvm/test/Transforms/InterleavedAccess/AArch64/fixed-deinterleave-intrinsics.ll
+++ b/llvm/test/Transforms/InterleavedAccess/AArch64/fixed-deinterleave-intrinsics.ll
@@ -15,11 +15,11 @@ define { <16 x i8>, <16 x i8> } @deinterleave_i8_factor2(ptr %ptr) {
 ; SVE-FIXED-LABEL: define { <16 x i8>, <16 x i8> } @deinterleave_i8_factor2
 ; SVE-FIXED-SAME: (ptr [[PTR:%.*]]) #[[ATTR0:[0-9]+]] {
 ; SVE-FIXED-NEXT:    [[LOAD:%.*]] = load <32 x i8>, ptr [[PTR]], align 1
-; SVE-FIXED-NEXT:    [[DEINTERLEAVE:%.*]] = tail call { <16 x i8>, <16 x i8> } @llvm.experimental.vector.deinterleave2.v32i8(<32 x i8> [[LOAD]])
+; SVE-FIXED-NEXT:    [[DEINTERLEAVE:%.*]] = tail call { <16 x i8>, <16 x i8> } @llvm.vector.deinterleave2.v32i8(<32 x i8> [[LOAD]])
 ; SVE-FIXED-NEXT:    ret { <16 x i8>, <16 x i8> } [[DEINTERLEAVE]]
 ;
   %load = load <32 x i8>, ptr %ptr, align 1
-  %deinterleave = tail call { <16 x i8>, <16 x i8> } @llvm.experimental.vector.deinterleave2.v32i8(<32 x i8> %load)
+  %deinterleave = tail call { <16 x i8>, <16 x i8> } @llvm.vector.deinterleave2.v32i8(<32 x i8> %load)
   ret { <16 x i8>, <16 x i8> } %deinterleave
 }
 
@@ -32,11 +32,11 @@ define { <8 x i16>, <8 x i16> } @deinterleave_i16_factor2(ptr %ptr) {
 ; SVE-FIXED-LABEL: define { <8 x i16>, <8 x i16> } @deinterleave_i16_factor2
 ; SVE-FIXED-SAME: (ptr [[PTR:%.*]]) #[[ATTR0]] {
 ; SVE-FIXED-NEXT:    [[LOAD:%.*]] = load <16 x i16>, ptr [[PTR]], align 2
-; SVE-FIXED-NEXT:    [[DEINTERLEAVE:%.*]] = tail call { <8 x i16>, <8 x i16> } @llvm.experimental.vector.deinterleave2.v16i16(<16 x i16> [[LOAD]])
+; SVE-FIXED-NEXT:    [[DEINTERLEAVE:%.*]] = tail call { <8 x i16>, <8 x i16> } @llvm.vector.deinterleave2.v16i16(<16 x i16> [[LOAD]])
 ; SVE-FIXED-NEXT:    ret { <8 x i16>, <8 x i16> } [[DEINTERLEAVE]]
 ;
   %load = load <16 x i16>, ptr %ptr, align 2
-  %deinterleave = tail call { <8 x i16>, <8 x i16> } @llvm.experimental.vector.deinterleave2.v16i16(<16 x i16> %load)
+  %deinterleave = tail call { <8 x i16>, <8 x i16> } @llvm.vector.deinterleave2.v16i16(<16 x i16> %load)
   ret { <8 x i16>, <8 x i16> } %deinterleave
 }
 
@@ -49,11 +49,11 @@ define { <4 x i32>, <4 x i32> } @deinterleave_8xi32_factor2(ptr %ptr) {
 ; SVE-FIXED-LABEL: define { <4 x i32>, <4 x i32> } @deinterleave_8xi32_factor2
 ; SVE-FIXED-SAME: (ptr [[PTR:%.*]]) #[[ATTR0]] {
 ; SVE-FIXED-NEXT:    [[LOAD:%.*]] = load <8 x i32>, ptr [[PTR]], align 4
-; SVE-FIXED-NEXT:    [[DEINTERLEAVE:%.*]] = tail call { <4 x i32>, <4 x i32> } @llvm.experimental.vector.deinterleave2.v8i32(<8 x i32> [[LOAD]])
+; SVE-FIXED-NEXT:    [[DEINTERLEAVE:%.*]] = tail call { <4 x i32>, <4 x i32> } @llvm.vector.deinterleave2.v8i32(<8 x i32> [[LOAD]])
 ; SVE-FIXED-NEXT:    ret { <4 x i32>, <4 x i32> } [[DEINTERLEAVE]]
 ;
   %load = load <8 x i32>, ptr %ptr, align 4
-  %deinterleave = tail call { <4 x i32>, <4 x i32> } @llvm.experimental.vector.deinterleave2.v8i32(<8 x i32> %load)
+  %deinterleave = tail call { <4 x i32>, <4 x i32> } @llvm.vector.deinterleave2.v8i32(<8 x i32> %load)
   ret { <4 x i32>, <4 x i32> } %deinterleave
 }
 
@@ -66,11 +66,11 @@ define { <2 x i64>, <2 x i64> } @deinterleave_i64_factor2(ptr %ptr) {
 ; SVE-FIXED-LABEL: define { <2 x i64>, <2 x i64> } @deinterleave_i64_factor2
 ; SVE-FIXED-SAME: (ptr [[PTR:%.*]]) #[[ATTR0]] {
 ; SVE-FIXED-NEXT:    [[LOAD:%.*]] = load <4 x i64>, ptr [[PTR]], align 8
-; SVE-FIXED-NEXT:    [[DEINTERLEAVE:%.*]] = tail call { <2 x i64>, <2 x i64> } @llvm.experimental.vector.deinterleave2.v4i64(<4 x i64> [[LOAD]])
+; SVE-FIXED-NEXT:    [[DEINTERLEAVE:%.*]] = tail call { <2 x i64>, <2 x i64> } @llvm.vector.deinterleave2.v4i64(<4 x i64> [[LOAD]])
 ; SVE-FIXED-NEXT:    ret { <2 x i64>, <2 x i64> } [[DEINTERLEAVE]]
 ;
   %load = load <4 x i64>, ptr %ptr, align 8
-  %deinterleave = tail call { <2 x i64>, <2 x i64> } @llvm.experimental.vector.deinterleave2.v4i64(<4 x i64> %load)
+  %deinterleave = tail call { <2 x i64>, <2 x i64> } @llvm.vector.deinterleave2.v4i64(<4 x i64> %load)
   ret { <2 x i64>, <2 x i64> } %deinterleave
 }
 
@@ -83,11 +83,11 @@ define { <4 x float>, <4 x float> } @deinterleave_float_factor2(ptr %ptr) {
 ; SVE-FIXED-LABEL: define { <4 x float>, <4 x float> } @deinterleave_float_factor2
 ; SVE-FIXED-SAME: (ptr [[PTR:%.*]]) #[[ATTR0]] {
 ; SVE-FIXED-NEXT:    [[LOAD:%.*]] = load <8 x float>, ptr [[PTR]], align 4
-; SVE-FIXED-NEXT:    [[DEINTERLEAVE:%.*]] = tail call { <4 x float>, <4 x float> } @llvm.experimental.vector.deinterleave2.v8f32(<8 x float> [[LOAD]])
+; SVE-FIXED-NEXT:    [[DEINTERLEAVE:%.*]] = tail call { <4 x float>, <4 x float> } @llvm.vector.deinterleave2.v8f32(<8 x float> [[LOAD]])
 ; SVE-FIXED-NEXT:    ret { <4 x float>, <4 x float> } [[DEINTERLEAVE]]
 ;
   %load = load <8 x float>, ptr %ptr, align 4
-  %deinterleave = tail call { <4 x float>, <4 x float> } @llvm.experimental.vector.deinterleave2.v8f32(<8 x float> %load)
+  %deinterleave = tail call { <4 x float>, <4 x float> } @llvm.vector.deinterleave2.v8f32(<8 x float> %load)
   ret { <4 x float>, <4 x float> } %deinterleave
 }
 
@@ -100,11 +100,11 @@ define { <2 x double>, <2 x double> } @deinterleave_double_factor2(ptr %ptr) {
 ; SVE-FIXED-LABEL: define { <2 x double>, <2 x double> } @deinterleave_double_factor2
 ; SVE-FIXED-SAME: (ptr [[PTR:%.*]]) #[[ATTR0]] {
 ; SVE-FIXED-NEXT:    [[LOAD:%.*]] = load <4 x double>, ptr [[PTR]], align 8
-; SVE-FIXED-NEXT:    [[DEINTERLEAVE:%.*]] = tail call { <2 x double>, <2 x double> } @llvm.experimental.vector.deinterleave2.v4f64(<4 x double> [[LOAD]])
+; SVE-FIXED-NEXT:    [[DEINTERLEAVE:%.*]] = tail call { <2 x double>, <2 x double> } @llvm.vector.deinterleave2.v4f64(<4 x double> [[LOAD]])
 ; SVE-FIXED-NEXT:    ret { <2 x double>, <2 x double> } [[DEINTERLEAVE]]
 ;
   %load = load <4 x double>, ptr %ptr, align 8
-  %deinterleave = tail call { <2 x double>, <2 x double> } @llvm.experimental.vector.deinterleave2.v4f64(<4 x double> %load)
+  %deinterleave = tail call { <2 x double>, <2 x double> } @llvm.vector.deinterleave2.v4f64(<4 x double> %load)
   ret { <2 x double>, <2 x double> } %deinterleave
 }
 
@@ -117,11 +117,11 @@ define { <2 x ptr>, <2 x ptr> } @deinterleave_ptr_factor2(ptr %ptr) {
 ; SVE-FIXED-LABEL: define { <2 x ptr>, <2 x ptr> } @deinterleave_ptr_factor2
 ; SVE-FIXED-SAME: (ptr [[PTR:%.*]]) #[[ATTR0]] {
 ; SVE-FIXED-NEXT:    [[LOAD:%.*]] = load <4 x ptr>, ptr [[PTR]], align 8
-; SVE-FIXED-NEXT:    [[DEINTERLEAVE:%.*]] = tail call { <2 x ptr>, <2 x ptr> } @llvm.experimental.vector.deinterleave2.v4p0(<4 x ptr> [[LOAD]])
+; SVE-FIXED-NEXT:    [[DEINTERLEAVE:%.*]] = tail call { <2 x ptr>, <2 x ptr> } @llvm.vector.deinterleave2.v4p0(<4 x ptr> [[LOAD]])
 ; SVE-FIXED-NEXT:    ret { <2 x ptr>, <2 x ptr> } [[DEINTERLEAVE]]
 ;
   %load = load <4 x ptr>, ptr %ptr, align 8
-  %deinterleave = tail call { <2 x ptr>, <2 x ptr> } @llvm.experimental.vector.deinterleave2.v4p0(<4 x ptr> %load)
+  %deinterleave = tail call { <2 x ptr>, <2 x ptr> } @llvm.vector.deinterleave2.v4p0(<4 x ptr> %load)
   ret { <2 x ptr>, <2 x ptr> } %deinterleave
 }
 
@@ -133,11 +133,11 @@ define void @interleave_i8_factor2(ptr %ptr, <16 x i8> %l, <16 x i8> %r) {
 ;
 ; SVE-FIXED-LABEL: define void @interleave_i8_factor2
 ; SVE-FIXED-SAME: (ptr [[PTR:%.*]], <16 x i8> [[L:%.*]], <16 x i8> [[R:%.*]]) #[[ATTR0]] {
-; SVE-FIXED-NEXT:    [[INTERLEAVE:%.*]] = tail call <32 x i8> @llvm.experimental.vector.interleave2.v32i8(<16 x i8> [[L]], <16 x i8> [[R]])
+; SVE-FIXED-NEXT:    [[INTERLEAVE:%.*]] = tail call <32 x i8> @llvm.vector.interleave2.v32i8(<16 x i8> [[L]], <16 x i8> [[R]])
 ; SVE-FIXED-NEXT:    store <32 x i8> [[INTERLEAVE]], ptr [[PTR]], align 1
 ; SVE-FIXED-NEXT:    ret void
 ;
-  %interleave = tail call <32 x i8> @llvm.experimental.vector.interleave2.v32i8(<16 x i8> %l, <16 x i8> %r)
+  %interleave = tail call <32 x i8> @llvm.vector.interleave2.v32i8(<16 x i8> %l, <16 x i8> %r)
   store <32 x i8> %interleave, ptr %ptr, align 1
   ret void
 }
@@ -150,11 +150,11 @@ define void @interleave_i16_factor2(ptr %ptr, <8 x i16> %l, <8 x i16> %r) {
 ;
 ; SVE-FIXED-LABEL: define void @interleave_i16_factor2
 ; SVE-FIXED-SAME: (ptr [[PTR:%.*]], <8 x i16> [[L:%.*]], <8 x i16> [[R:%.*]]) #[[ATTR0]] {
-; SVE-FIXED-NEXT:    [[INTERLEAVE:%.*]] = tail call <16 x i16> @llvm.experimental.vector.interleave2.v16i16(<8 x i16> [[L]], <8 x i16> [[R]])
+; SVE-FIXED-NEXT:    [[INTERLEAVE:%.*]] = tail call <16 x i16> @llvm.vector.interleave2.v16i16(<8 x i16> [[L]], <8 x i16> [[R]])
 ; SVE-FIXED-NEXT:    store <16 x i16> [[INTERLEAVE]], ptr [[PTR]], align 2
 ; SVE-FIXED-NEXT:    ret void
 ;
-  %interleave = tail call <16 x i16> @llvm.experimental.vector.interleave2.v16i16(<8 x i16> %l, <8 x i16> %r)
+  %interleave = tail call <16 x i16> @llvm.vector.interleave2.v16i16(<8 x i16> %l, <8 x i16> %r)
   store <16 x i16> %interleave, ptr %ptr, align 2
   ret void
 }
@@ -167,11 +167,11 @@ define void @interleave_i32_factor2(ptr %ptr, <4 x i32> %l, <4 x i32> %r) {
 ;
 ; SVE-FIXED-LABEL: define void @interleave_i32_factor2
 ; SVE-FIXED-SAME: (ptr [[PTR:%.*]], <4 x i32> [[L:%.*]], <4 x i32> [[R:%.*]]) #[[ATTR0]] {
-; SVE-FIXED-NEXT:    [[INTERLEAVE:%.*]] = tail call <8 x i32> @llvm.experimental.vector.interleave2.v8i32(<4 x i32> [[L]], <4 x i32> [[R]])
+; SVE-FIXED-NEXT:    [[INTERLEAVE:%.*]] = tail call <8 x i32> @llvm.vector.interleave2.v8i32(<4 x i32> [[L]], <4 x i32> [[R]])
 ; SVE-FIXED-NEXT:    store <8 x i32> [[INTERLEAVE]], ptr [[PTR]], align 4
 ; SVE-FIXED-NEXT:    ret void
 ;
-  %interleave = tail call <8 x i32> @llvm.experimental.vector.interleave2.v8i32(<4 x i32> %l, <4 x i32> %r)
+  %interleave = tail call <8 x i32> @llvm.vector.interleave2.v8i32(<4 x i32> %l, <4 x i32> %r)
   store <8 x i32> %interleave, ptr %ptr, align 4
   ret void
 }
@@ -184,11 +184,11 @@ define void @interleave_i64_factor2(ptr %ptr, <2 x i64> %l, <2 x i64> %r) {
 ;
 ; SVE-FIXED-LABEL: define void @interleave_i64_factor2
 ; SVE-FIXED-SAME: (ptr [[PTR:%.*]], <2 x i64> [[L:%.*]], <2 x i64> [[R:%.*]]) #[[ATTR0]] {
-; SVE-FIXED-NEXT:    [[INTERLEAVE:%.*]] = tail call <4 x i64> @llvm.experimental.vector.interleave2.v4i64(<2 x i64> [[L]], <2 x i64> [[R]])
+; SVE-FIXED-NEXT:    [[INTERLEAVE:%.*]] = tail call <4 x i64> @llvm.vector.interleave2.v4i64(<2 x i64> [[L]], <2 x i64> [[R]])
 ; SVE-FIXED-NEXT:    store <4 x i64> [[INTERLEAVE]], ptr [[PTR]], align 8
 ; SVE-FIXED-NEXT:    ret void
 ;
-  %interleave = tail call <4 x i64> @llvm.experimental.vector.interleave2.v4i64(<2 x i64> %l, <2 x i64> %r)
+  %interleave = tail call <4 x i64> @llvm.vector.interleave2.v4i64(<2 x i64> %l, <2 x i64> %r)
   store <4 x i64> %interleave, ptr %ptr, align 8
   ret void
 }
@@ -201,11 +201,11 @@ define void @interleave_float_factor2(ptr %ptr, <4 x float> %l, <4 x float> %r)
 ;
 ; SVE-FIXED-LABEL: define void @interleave_float_factor2
 ; SVE-FIXED-SAME: (ptr [[PTR:%.*]], <4 x float> [[L:%.*]], <4 x float> [[R:%.*]]) #[[ATTR0]] {
-; SVE-FIXED-NEXT:    [[INTERLEAVE:%.*]] = tail call <8 x float> @llvm.experimental.vector.interleave2.v8f32(<4 x float> [[L]], <4 x float> [[R]])
+; SVE-FIXED-NEXT:    [[INTERLEAVE:%.*]] = tail call <8 x float> @llvm.vector.interleave2.v8f32(<4 x float> [[L]], <4 x float> [[R]])
 ; SVE-FIXED-NEXT:    store <8 x float> [[INTERLEAVE]], ptr [[PTR]], align 4
 ; SVE-FIXED-NEXT:    ret void
 ;
-  %interleave = tail call <8 x float> @llvm.experimental.vector.interleave2.v8f32(<4 x float> %l, <4 x float> %r)
+  %interleave = tail call <8 x float> @llvm.vector.interleave2.v8f32(<4 x float> %l, <4 x float> %r)
   store <8 x float> %interleave, ptr %ptr, align 4
   ret void
 }
@@ -218,11 +218,11 @@ define void @interleave_double_factor2(ptr %ptr, <2 x double> %l, <2 x double> %
 ;
 ; SVE-FIXED-LABEL: define void @interleave_double_factor2
 ; SVE-FIXED-SAME: (ptr [[PTR:%.*]], <2 x double> [[L:%.*]], <2 x double> [[R:%.*]]) #[[ATTR0]] {
-; SVE-FIXED-NEXT:    [[INTERLEAVE:%.*]] = tail call <4 x double> @llvm.experimental.vector.interleave2.v4f64(<2 x double> [[L]], <2 x double> [[R]])
+; SVE-FIXED-NEXT:    [[INTERLEAVE:%.*]] = tail call <4 x double> @llvm.vector.interleave2.v4f64(<2 x double> [[L]], <2 x double> [[R]])
 ; SVE-FIXED-NEXT:    store <4 x double> [[INTERLEAVE]], ptr [[PTR]], align 4
 ; SVE-FIXED-NEXT:    ret void
 ;
-  %interleave = tail call <4 x double> @llvm.experimental.vector.interleave2.v4f64(<2 x double> %l, <2 x double> %r)
+  %interleave = tail call <4 x double> @llvm.vector.interleave2.v4f64(<2 x double> %l, <2 x double> %r)
   store <4 x double> %interleave, ptr %ptr, align 4
   ret void
 }
@@ -235,11 +235,11 @@ define void @interleave_ptr_factor2(ptr %ptr, <2 x ptr> %l, <2 x ptr> %r) {
 ;
 ; SVE-FIXED-LABEL: define void @interleave_ptr_factor2
 ; SVE-FIXED-SAME: (ptr [[PTR:%.*]], <2 x ptr> [[L:%.*]], <2 x ptr> [[R:%.*]]) #[[ATTR0]] {
-; SVE-FIXED-NEXT:    [[INTERLEAVE:%.*]] = tail call <4 x ptr> @llvm.experimental.vector.interleave2.v4p0(<2 x ptr> [[L]], <2 x ptr> [[R]])
+; SVE-FIXED-NEXT:    [[INTERLEAVE:%.*]] = tail call <4 x ptr> @llvm.vector.interleave2.v4p0(<2 x ptr> [[L]], <2 x ptr> [[R]])
 ; SVE-FIXED-NEXT:    store <4 x ptr> [[INTERLEAVE]], ptr [[PTR]], align 4
 ; SVE-FIXED-NEXT:    ret void
 ;
-  %interleave = tail call <4 x ptr> @llvm.experimental.vector.interleave2.v4p0(<2 x ptr> %l, <2 x ptr> %r)
+  %interleave = tail call <4 x ptr> @llvm.vector.interleave2.v4p0(<2 x ptr> %l, <2 x ptr> %r)
   store <4 x ptr> %interleave, ptr %ptr, align 4
   ret void
 }
@@ -266,11 +266,11 @@ define { <16 x i16>, <16 x i16> } @deinterleave_wide_i16_factor2(ptr %ptr) #0 {
 ; SVE-FIXED-LABEL: define { <16 x i16>, <16 x i16> } @deinterleave_wide_i16_factor2
 ; SVE-FIXED-SAME: (ptr [[PTR:%.*]]) #[[ATTR0]] {
 ; SVE-FIXED-NEXT:    [[LOAD:%.*]] = load <32 x i16>, ptr [[PTR]], align 2
-; SVE-FIXED-NEXT:    [[DEINTERLEAVE:%.*]] = tail call { <16 x i16>, <16 x i16> } @llvm.experimental.vector.deinterleave2.v32i16(<32 x i16> [[LOAD]])
+; SVE-FIXED-NEXT:    [[DEINTERLEAVE:%.*]] = tail call { <16 x i16>, <16 x i16> } @llvm.vector.deinterleave2.v32i16(<32 x i16> [[LOAD]])
 ; SVE-FIXED-NEXT:    ret { <16 x i16>, <16 x i16> } [[DEINTERLEAVE]]
 ;
   %load = load <32 x i16>, ptr %ptr, align 2
-  %deinterleave = tail call { <16 x i16>, <16 x i16> } @llvm.experimental.vector.deinterleave2.v32i16(<32 x i16> %load)
+  %deinterleave = tail call { <16 x i16>, <16 x i16> } @llvm.vector.deinterleave2.v32i16(<32 x i16> %load)
   ret { <16 x i16>, <16 x i16> } %deinterleave
 }
 
@@ -297,29 +297,29 @@ define void @interleave_wide_ptr_factor2(ptr %ptr, <8 x ptr> %l, <8 x ptr> %r) {
 ;
 ; SVE-FIXED-LABEL: define void @interleave_wide_ptr_factor2
 ; SVE-FIXED-SAME: (ptr [[PTR:%.*]], <8 x ptr> [[L:%.*]], <8 x ptr> [[R:%.*]]) #[[ATTR0]] {
-; SVE-FIXED-NEXT:    [[INTERLEAVE:%.*]] = tail call <16 x ptr> @llvm.experimental.vector.interleave2.v16p0(<8 x ptr> [[L]], <8 x ptr> [[R]])
+; SVE-FIXED-NEXT:    [[INTERLEAVE:%.*]] = tail call <16 x ptr> @llvm.vector.interleave2.v16p0(<8 x ptr> [[L]], <8 x ptr> [[R]])
 ; SVE-FIXED-NEXT:    store <16 x ptr> [[INTERLEAVE]], ptr [[PTR]], align 4
 ; SVE-FIXED-NEXT:    ret void
 ;
-  %interleave = tail call <16 x ptr> @llvm.experimental.vector.interleave2.v16p0(<8 x ptr> %l, <8 x ptr> %r)
+  %interleave = tail call <16 x ptr> @llvm.vector.interleave2.v16p0(<8 x ptr> %l, <8 x ptr> %r)
   store <16 x ptr> %interleave, ptr %ptr, align 4
   ret void
 }
 
-declare { <16 x i8>, <16 x i8> } @llvm.experimental.vector.deinterleave2.v32i8(<32 x i8>)
-declare { <8 x i16>, <8 x i16> } @llvm.experimental.vector.deinterleave2.v16i16(<16 x i16>)
-declare { <4 x i32>, <4 x i32> } @llvm.experimental.vector.deinterleave2.v8i32(<8 x i32>)
-declare { <2 x i64>, <2 x i64> } @llvm.experimental.vector.deinterleave2.v4i64(<4 x i64>)
-declare { <4 x float>, <4 x float> } @llvm.experimental.vector.deinterleave2.v8f32(<8 x float>)
-declare { <2 x double>, <2 x double> } @llvm.experimental.vector.deinterleave2.v4f64(<4 x double>)
-declare { <2 x ptr>, <2 x ptr> } @llvm.experimental.vector.deinterleave2.v4p0(<4 x ptr>)
-declare { <16 x i16>, <16 x i16> } @llvm.experimental.vector.deinterleave2.v32i16(<32 x i16>)
+declare { <16 x i8>, <16 x i8> } @llvm.vector.deinterleave2.v32i8(<32 x i8>)
+declare { <8 x i16>, <8 x i16> } @llvm.vector.deinterleave2.v16i16(<16 x i16>)
+declare { <4 x i32>, <4 x i32> } @llvm.vector.deinterleave2.v8i32(<8 x i32>)
+declare { <2 x i64>, <2 x i64> } @llvm.vector.deinterleave2.v4i64(<4 x i64>)
+declare { <4 x float>, <4 x float> } @llvm.vector.deinterleave2.v8f32(<8 x float>)
+declare { <2 x double>, <2 x double> } @llvm.vector.deinterleave2.v4f64(<4 x double>)
+declare { <2 x ptr>, <2 x ptr> } @llvm.vector.deinterleave2.v4p0(<4 x ptr>)
+declare { <16 x i16>, <16 x i16> } @llvm.vector.deinterleave2.v32i16(<32 x i16>)
 
-declare <32 x i8> @llvm.experimental.vector.interleave2.v32i8(<16 x i8>, <16 x i8>)
-declare <16 x i16> @llvm.experimental.vector.interleave2.v16i16(<8 x i16>, <8 x i16>)
-declare <8 x i32> @llvm.experimental.vector.interleave2.v8i32(<4 x i32>, <4 x i32>)
-declare <4 x i64> @llvm.experimental.vector.interleave2.v4i64(<2 x i64>, <2 x i64>)
-declare <8 x float> @llvm.experimental.vector.interleave2.v8f32(<4 x float>, <4 x float>)
-declare <4 x double> @llvm.experimental.vector.interleave2.v4f64(<2 x double>, <2 x double>)
-declare <4 x ptr> @llvm.experimental.vector.interleave2.v4p0(<2 x ptr>, <2 x ptr>)
-declare <16 x ptr> @llvm.experimental.vector.interleave2.v16p0(<8 x ptr>, <8 x ptr>)
+declare <32 x i8> @llvm.vector.interleave2.v32i8(<16 x i8>, <16 x i8>)
+declare <16 x i16> @llvm.vector.interleave2.v16i16(<8 x i16>, <8 x i16>)
+declare <8 x i32> @llvm.vector.interleave2.v8i32(<4 x i32>, <4 x i32>)
+declare <4 x i64> @llvm.vector.interleave2.v4i64(<2 x i64>, <2 x i64>)
+declare <8 x float> @llvm.vector.interleave2.v8f32(<4 x float>, <4 x float>)
+declare <4 x double> @llvm.vector.interleave2.v4f64(<2 x double>, <2 x double>)
+declare <4 x ptr> @llvm.vector.interleave2.v4p0(<2 x ptr>, <2 x ptr>)
+declare <16 x ptr> @llvm.vector.interleave2.v16p0(<8 x ptr>, <8 x ptr>)
diff --git a/llvm/test/Transforms/InterleavedAccess/AArch64/scalable-deinterleave-intrinsics.ll b/llvm/test/Transforms/InterleavedAccess/AArch64/scalable-deinterleave-intrinsics.ll
index 6353bf10d57c..2a05718cc416 100644
--- a/llvm/test/Transforms/InterleavedAccess/AArch64/scalable-deinterleave-intrinsics.ll
+++ b/llvm/test/Transforms/InterleavedAccess/AArch64/scalable-deinterleave-intrinsics.ll
@@ -11,7 +11,7 @@ define { <vscale x 16 x i8>, <vscale x 16 x i8> } @deinterleave_nxi8_factor2(ptr
 ; CHECK-NEXT:    ret { <vscale x 16 x i8>, <vscale x 16 x i8> } [[LDN]]
 ;
   %load = load <vscale x 32 x i8>, ptr %ptr, align 1
-  %deinterleave = tail call { <vscale x 16 x i8>, <vscale x 16 x i8> } @llvm.experimental.vector.deinterleave2.nxv32i8(<vscale x 32 x i8> %load)
+  %deinterleave = tail call { <vscale x 16 x i8>, <vscale x 16 x i8> } @llvm.vector.deinterleave2.nxv32i8(<vscale x 32 x i8> %load)
   ret { <vscale x 16 x i8>, <vscale x 16 x i8> } %deinterleave
 }
 
@@ -22,7 +22,7 @@ define { <vscale x 8 x i16>, <vscale x 8 x i16> } @deinterleave_nxi16_factor2(pt
 ; CHECK-NEXT:    ret { <vscale x 8 x i16>, <vscale x 8 x i16> } [[LDN]]
 ;
   %load = load <vscale x 16 x i16>, ptr %ptr, align 2
-  %deinterleave = tail call { <vscale x 8 x i16>, <vscale x 8 x i16> } @llvm.experimental.vector.deinterleave2.nxv16i16(<vscale x 16 x i16> %load)
+  %deinterleave = tail call { <vscale x 8 x i16>, <vscale x 8 x i16> } @llvm.vector.deinterleave2.nxv16i16(<vscale x 16 x i16> %load)
   ret { <vscale x 8 x i16>, <vscale x 8 x i16> } %deinterleave
 }
 
@@ -33,7 +33,7 @@ define { <vscale x 4 x i32>, <vscale x 4 x i32> } @deinterleave_nx8xi32_factor2(
 ; CHECK-NEXT:    ret { <vscale x 4 x i32>, <vscale x 4 x i32> } [[LDN]]
 ;
   %load = load <vscale x 8 x i32>, ptr %ptr, align 4
-  %deinterleave = tail call { <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.experimental.vector.deinterleave2.nxv8i32(<vscale x 8 x i32> %load)
+  %deinterleave = tail call { <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.vector.deinterleave2.nxv8i32(<vscale x 8 x i32> %load)
   ret { <vscale x 4 x i32>, <vscale x 4 x i32> } %deinterleave
 }
 
@@ -44,7 +44,7 @@ define { <vscale x 2 x i64>, <vscale x 2 x i64> } @deinterleave_nxi64_factor2(pt
 ; CHECK-NEXT:    ret { <vscale x 2 x i64>, <vscale x 2 x i64> } [[LDN]]
 ;
   %load = load <vscale x 4 x i64>, ptr %ptr, align 8
-  %deinterleave = tail call { <vscale x 2 x i64>, <vscale x 2 x i64> } @llvm.experimental.vector.deinterleave2.nxv4i64(<vscale x 4 x i64> %load)
+  %deinterleave = tail call { <vscale x 2 x i64>, <vscale x 2 x i64> } @llvm.vector.deinterleave2.nxv4i64(<vscale x 4 x i64> %load)
   ret { <vscale x 2 x i64>, <vscale x 2 x i64> } %deinterleave
 }
 
@@ -55,7 +55,7 @@ define { <vscale x 4 x float>, <vscale x 4 x float> } @deinterleave_nxfloat_fact
 ; CHECK-NEXT:    ret { <vscale x 4 x float>, <vscale x 4 x float> } [[LDN]]
 ;
   %load = load <vscale x 8 x float>, ptr %ptr, align 4
-  %deinterleave = tail call { <vscale x 4 x float>, <vscale x 4 x float> } @llvm.experimental.vector.deinterleave2.nxv8f32(<vscale x 8 x float> %load)
+  %deinterleave = tail call { <vscale x 4 x float>, <vscale x 4 x float> } @llvm.vector.deinterleave2.nxv8f32(<vscale x 8 x float> %load)
   ret { <vscale x 4 x float>, <vscale x 4 x float> } %deinterleave
 }
 
@@ -66,7 +66,7 @@ define { <vscale x 2 x double>, <vscale x 2 x double> } @deinterleave_nxdouble_f
 ; CHECK-NEXT:    ret { <vscale x 2 x double>, <vscale x 2 x double> } [[LDN]]
 ;
   %load = load <vscale x 4 x double>, ptr %ptr, align 8
-  %deinterleave = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.experimental.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %load)
+  %deinterleave = tail call { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.vector.deinterleave2.nxv4f64(<vscale x 4 x double> %load)
   ret { <vscale x 2 x double>, <vscale x 2 x double> } %deinterleave
 }
 
@@ -77,7 +77,7 @@ define { <vscale x 2 x ptr>, <vscale x 2 x ptr> } @deinterleave_nxptr_factor2(pt
 ; CHECK-NEXT:    ret { <vscale x 2 x ptr>, <vscale x 2 x ptr> } [[LDN]]
 ;
   %load = load <vscale x 4 x ptr>, ptr %ptr, align 8
-  %deinterleave = tail call { <vscale x 2 x ptr>, <vscale x 2 x ptr> } @llvm.experimental.vector.deinterleave2.nxv4p0(<vscale x 4 x ptr> %load)
+  %deinterleave = tail call { <vscale x 2 x ptr>, <vscale x 2 x ptr> } @llvm.vector.deinterleave2.nxv4p0(<vscale x 4 x ptr> %load)
   ret { <vscale x 2 x ptr>, <vscale x 2 x ptr> } %deinterleave
 }
 
@@ -87,7 +87,7 @@ define void @interleave_nxi8_factor2(ptr %ptr, <vscale x 16 x i8> %l, <vscale x
 ; CHECK-NEXT:    call void @llvm.aarch64.sve.st2.nxv16i8(<vscale x 16 x i8> [[L]], <vscale x 16 x i8> [[R]], <vscale x 16 x i1> shufflevector (<vscale x 16 x i1> insertelement (<vscale x 16 x i1> poison, i1 true, i64 0), <vscale x 16 x i1> poison, <vscale x 16 x i32> zeroinitializer), ptr [[PTR]])
 ; CHECK-NEXT:    ret void
 ;
-  %interleave = tail call <vscale x 32 x i8> @llvm.experimental.vector.interleave2.nxv32i8(<vscale x 16 x i8> %l, <vscale x 16 x i8> %r)
+  %interleave = tail call <vscale x 32 x i8> @llvm.vector.interleave2.nxv32i8(<vscale x 16 x i8> %l, <vscale x 16 x i8> %r)
   store <vscale x 32 x i8> %interleave, ptr %ptr, align 1
   ret void
 }
@@ -98,7 +98,7 @@ define void @interleave_nxi16_factor2(ptr %ptr, <vscale x 8 x i16> %l, <vscale x
 ; CHECK-NEXT:    call void @llvm.aarch64.sve.st2.nxv8i16(<vscale x 8 x i16> [[L]], <vscale x 8 x i16> [[R]], <vscale x 8 x i1> shufflevector (<vscale x 8 x i1> insertelement (<vscale x 8 x i1> poison, i1 true, i64 0), <vscale x 8 x i1> poison, <vscale x 8 x i32> zeroinitializer), ptr [[PTR]])
 ; CHECK-NEXT:    ret void
 ;
-  %interleave = tail call <vscale x 16 x i16> @llvm.experimental.vector.interleave2.nxv16i16(<vscale x 8 x i16> %l, <vscale x 8 x i16> %r)
+  %interleave = tail call <vscale x 16 x i16> @llvm.vector.interleave2.nxv16i16(<vscale x 8 x i16> %l, <vscale x 8 x i16> %r)
   store <vscale x 16 x i16> %interleave, ptr %ptr, align 2
   ret void
 }
@@ -109,7 +109,7 @@ define void @interleave_nxi32_factor2(ptr %ptr, <vscale x 4 x i32> %l, <vscale x
 ; CHECK-NEXT:    call void @llvm.aarch64.sve.st2.nxv4i32(<vscale x 4 x i32> [[L]], <vscale x 4 x i32> [[R]], <vscale x 4 x i1> shufflevector (<vscale x 4 x i1> insertelement (<vscale x 4 x i1> poison, i1 true, i64 0), <vscale x 4 x i1> poison, <vscale x 4 x i32> zeroinitializer), ptr [[PTR]])
 ; CHECK-NEXT:    ret void
 ;
-  %interleave = tail call <vscale x 8 x i32> @llvm.experimental.vector.interleave2.nxv8i32(<vscale x 4 x i32> %l, <vscale x 4 x i32> %r)
+  %interleave = tail call <vscale x 8 x i32> @llvm.vector.interleave2.nxv8i32(<vscale x 4 x i32> %l, <vscale x 4 x i32> %r)
   store <vscale x 8 x i32> %interleave, ptr %ptr, align 4
   ret void
 }
@@ -120,7 +120,7 @@ define void @interleave_nxi64_factor2(ptr %ptr, <vscale x 2 x i64> %l, <vscale x
 ; CHECK-NEXT:    call void @llvm.aarch64.sve.st2.nxv2i64(<vscale x 2 x i64> [[L]], <vscale x 2 x i64> [[R]], <vscale x 2 x i1> shufflevector (<vscale x 2 x i1> insertelement (<vscale x 2 x i1> poison, i1 true, i64 0), <vscale x 2 x i1> poison, <vscale x 2 x i32> zeroinitializer), ptr [[PTR]])
 ; CHECK-NEXT:    ret void
 ;
-  %interleave = tail call <vscale x 4 x i64> @llvm.experimental.vector.interleave2.nxv4i64(<vscale x 2 x i64> %l, <vscale x 2 x i64> %r)
+  %interleave = tail call <vscale x 4 x i64> @llvm.vector.interleave2.nxv4i64(<vscale x 2 x i64> %l, <vscale x 2 x i64> %r)
   store <vscale x 4 x i64> %interleave, ptr %ptr, align 8
   ret void
 }
@@ -131,7 +131,7 @@ define void @interleave_nxfloat_factor2(ptr %ptr, <vscale x 4 x float> %l, <vsca
 ; CHECK-NEXT:    call void @llvm.aarch64.sve.st2.nxv4f32(<vscale x 4 x float> [[L]], <vscale x 4 x float> [[R]], <vscale x 4 x i1> shufflevector (<vscale x 4 x i1> insertelement (<vscale x 4 x i1> poison, i1 true, i64 0), <vscale x 4 x i1> poison, <vscale x 4 x i32> zeroinitializer), ptr [[PTR]])
 ; CHECK-NEXT:    ret void
 ;
-  %interleave = tail call <vscale x 8 x float> @llvm.experimental.vector.interleave2.nxv8f32(<vscale x 4 x float> %l, <vscale x 4 x float> %r)
+  %interleave = tail call <vscale x 8 x float> @llvm.vector.interleave2.nxv8f32(<vscale x 4 x float> %l, <vscale x 4 x float> %r)
   store <vscale x 8 x float> %interleave, ptr %ptr, align 4
   ret void
 }
@@ -142,7 +142,7 @@ define void @interleave_nxdouble_factor2(ptr %ptr, <vscale x 2 x double> %l, <vs
 ; CHECK-NEXT:    call void @llvm.aarch64.sve.st2.nxv2f64(<vscale x 2 x double> [[L]], <vscale x 2 x double> [[R]], <vscale x 2 x i1> shufflevector (<vscale x 2 x i1> insertelement (<vscale x 2 x i1> poison, i1 true, i64 0), <vscale x 2 x i1> poison, <vscale x 2 x i32> zeroinitializer), ptr [[PTR]])
 ; CHECK-NEXT:    ret void
 ;
-  %interleave = tail call <vscale x 4 x double> @llvm.experimental.vector.interleave2.nxv4f64(<vscale x 2 x double> %l, <vscale x 2 x double> %r)
+  %interleave = tail call <vscale x 4 x double> @llvm.vector.interleave2.nxv4f64(<vscale x 2 x double> %l, <vscale x 2 x double> %r)
   store <vscale x 4 x double> %interleave, ptr %ptr, align 4
   ret void
 }
@@ -153,7 +153,7 @@ define void @interleave_nxptr_factor2(ptr %ptr, <vscale x 2 x ptr> %l, <vscale x
 ; CHECK-NEXT:    call void @llvm.aarch64.sve.st2.nxv2p0(<vscale x 2 x ptr> [[L]], <vscale x 2 x ptr> [[R]], <vscale x 2 x i1> shufflevector (<vscale x 2 x i1> insertelement (<vscale x 2 x i1> poison, i1 true, i64 0), <vscale x 2 x i1> poison, <vscale x 2 x i32> zeroinitializer), ptr [[PTR]])
 ; CHECK-NEXT:    ret void
 ;
-  %interleave = tail call <vscale x 4 x ptr> @llvm.experimental.vector.interleave2.nxv4p0(<vscale x 2 x ptr> %l, <vscale x 2 x ptr> %r)
+  %interleave = tail call <vscale x 4 x ptr> @llvm.vector.interleave2.nxv4p0(<vscale x 2 x ptr> %l, <vscale x 2 x ptr> %r)
   store <vscale x 4 x ptr> %interleave, ptr %ptr, align 4
   ret void
 }
@@ -192,7 +192,7 @@ define { <vscale x 16 x i32>, <vscale x 16 x i32> } @deinterleave_wide_nxi32_fac
 ; CHECK-NEXT:    ret { <vscale x 16 x i32>, <vscale x 16 x i32> } [[TMP22]]
 ;
   %load = load <vscale x 32 x i32>, ptr %ptr, align 4
-  %deinterleave = tail call { <vscale x 16 x i32>, <vscale x 16 x i32> } @llvm.experimental.vector.deinterleave2.nxv32i32(<vscale x 32 x i32> %load)
+  %deinterleave = tail call { <vscale x 16 x i32>, <vscale x 16 x i32> } @llvm.vector.deinterleave2.nxv32i32(<vscale x 32 x i32> %load)
   ret { <vscale x 16 x i32>, <vscale x 16 x i32> } %deinterleave
 }
 
@@ -216,7 +216,7 @@ define { <vscale x 4 x double>, <vscale x 4 x double> } @deinterleave_wide_nxdou
 ; CHECK-NEXT:    ret { <vscale x 4 x double>, <vscale x 4 x double> } [[TMP12]]
 ;
   %load = load <vscale x 8 x double>, ptr %ptr, align 8
-  %deinterleave = tail call { <vscale x 4 x double>, <vscale x 4 x double> } @llvm.experimental.vector.deinterleave2.nxv8f64(<vscale x 8 x double> %load)
+  %deinterleave = tail call { <vscale x 4 x double>, <vscale x 4 x double> } @llvm.vector.deinterleave2.nxv8f64(<vscale x 8 x double> %load)
   ret { <vscale x 4 x double>, <vscale x 4 x double> } %deinterleave
 }
 
@@ -233,32 +233,32 @@ define void @interleave_wide_nxdouble_factor2(ptr %ptr, <vscale x 4 x double> %l
 ; CHECK-NEXT:    call void @llvm.aarch64.sve.st2.nxv2f64(<vscale x 2 x double> [[TMP5]], <vscale x 2 x double> [[TMP6]], <vscale x 2 x i1> shufflevector (<vscale x 2 x i1> insertelement (<vscale x 2 x i1> poison, i1 true, i64 0), <vscale x 2 x i1> poison, <vscale x 2 x i32> zeroinitializer), ptr [[TMP4]])
 ; CHECK-NEXT:    ret void
 ;
-  %interleave = tail call <vscale x 8 x double> @llvm.experimental.vector.interleave2.nxv8f64(<vscale x 4 x double> %l, <vscale x 4 x double> %r)
+  %interleave = tail call <vscale x 8 x double> @llvm.vector.interleave2.nxv8f64(<vscale x 4 x double> %l, <vscale x 4 x double> %r)
   store <vscale x 8 x double> %interleave, ptr %ptr, align 4
   ret void
 }
 
-declare { <vscale x 16 x i8>, <vscale x 16 x i8> } @llvm.experimental.vector.deinterleave2.nxv32i8(<vscale x 32 x i8>)
-declare { <vscale x 8 x i16>, <vscale x 8 x i16> } @llvm.experimental.vector.deinterleave2.nxv16i16(<vscale x 16 x i16>)
-declare { <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.experimental.vector.deinterleave2.nxv8i32(<vscale x 8 x i32>)
-declare { <vscale x 2 x i64>, <vscale x 2 x i64> } @llvm.experimental.vector.deinterleave2.nxv4i64(<vscale x 4 x i64>)
-declare { <vscale x 4 x float>, <vscale x 4 x float> } @llvm.experimental.vector.deinterleave2.nxv8f32(<vscale x 8 x float>)
-declare { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.experimental.vector.deinterleave2.nxv4f64(<vscale x 4 x double>)
-declare { <vscale x 2 x ptr>, <vscale x 2 x ptr> } @llvm.experimental.vector.deinterleave2.nxv4p0(<vscale x 4 x ptr>)
+declare { <vscale x 16 x i8>, <vscale x 16 x i8> } @llvm.vector.deinterleave2.nxv32i8(<vscale x 32 x i8>)
+declare { <vscale x 8 x i16>, <vscale x 8 x i16> } @llvm.vector.deinterleave2.nxv16i16(<vscale x 16 x i16>)
+declare { <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.vector.deinterleave2.nxv8i32(<vscale x 8 x i32>)
+declare { <vscale x 2 x i64>, <vscale x 2 x i64> } @llvm.vector.deinterleave2.nxv4i64(<vscale x 4 x i64>)
+declare { <vscale x 4 x float>, <vscale x 4 x float> } @llvm.vector.deinterleave2.nxv8f32(<vscale x 8 x float>)
+declare { <vscale x 2 x double>, <vscale x 2 x double> } @llvm.vector.deinterleave2.nxv4f64(<vscale x 4 x double>)
+declare { <vscale x 2 x ptr>, <vscale x 2 x ptr> } @llvm.vector.deinterleave2.nxv4p0(<vscale x 4 x ptr>)
 
 ; Larger deinterleaves to test 'legalization'
-declare { <vscale x 16 x i32>, <vscale x 16 x i32> } @llvm.experimental.vector.deinterleave2.nxv32i32(<vscale x 32 x i32>)
-declare { <vscale x 4 x double>, <vscale x 4 x double> } @llvm.experimental.vector.deinterleave2.nxv8f64(<vscale x 8 x double>)
+declare { <vscale x 16 x i32>, <vscale x 16 x i32> } @llvm.vector.deinterleave2.nxv32i32(<vscale x 32 x i32>)
+declare { <vscale x 4 x double>, <vscale x 4 x double> } @llvm.vector.deinterleave2.nxv8f64(<vscale x 8 x double>)
 
-declare <vscale x 32 x i8> @llvm.experimental.vector.interleave2.nxv32i8(<vscale x 16 x i8>, <vscale x 16 x i8>)
-declare <vscale x 16 x i16> @llvm.experimental.vector.interleave2.nxv16i16(<vscale x 8 x i16>, <vscale x 8 x i16>)
-declare <vscale x 8 x i32> @llvm.experimental.vector.interleave2.nxv8i32(<vscale x 4 x i32>, <vscale x 4 x i32>)
-declare <vscale x 4 x i64> @llvm.experimental.vector.interleave2.nxv4i64(<vscale x 2 x i64>, <vscale x 2 x i64>)
-declare <vscale x 8 x float> @llvm.experimental.vector.interleave2.nxv8f32(<vscale x 4 x float>, <vscale x 4 x float>)
-declare <vscale x 4 x double> @llvm.experimental.vector.interleave2.nxv4f64(<vscale x 2 x double>, <vscale x 2 x double>)
-declare <vscale x 4 x ptr> @llvm.experimental.vector.interleave2.nxv4p0(<vscale x 2 x ptr>, <vscale x 2 x ptr>)
+declare <vscale x 32 x i8> @llvm.vector.interleave2.nxv32i8(<vscale x 16 x i8>, <vscale x 16 x i8>)
+declare <vscale x 16 x i16> @llvm.vector.interleave2.nxv16i16(<vscale x 8 x i16>, <vscale x 8 x i16>)
+declare <vscale x 8 x i32> @llvm.vector.interleave2.nxv8i32(<vscale x 4 x i32>, <vscale x 4 x i32>)
+declare <vscale x 4 x i64> @llvm.vector.interleave2.nxv4i64(<vscale x 2 x i64>, <vscale x 2 x i64>)
+declare <vscale x 8 x float> @llvm.vector.interleave2.nxv8f32(<vscale x 4 x float>, <vscale x 4 x float>)
+declare <vscale x 4 x double> @llvm.vector.interleave2.nxv4f64(<vscale x 2 x double>, <vscale x 2 x double>)
+declare <vscale x 4 x ptr> @llvm.vector.interleave2.nxv4p0(<vscale x 2 x ptr>, <vscale x 2 x ptr>)
 
 ; Larger interleaves to test 'legalization'
-declare <vscale x 8 x double> @llvm.experimental.vector.interleave2.nxv8f64(<vscale x 4 x double>, <vscale x 4 x double>)
+declare <vscale x 8 x double> @llvm.vector.interleave2.nxv8f64(<vscale x 4 x double>, <vscale x 4 x double>)
 
 attributes #0 = { vscale_range(1,16) "target-features"="+sve" }
diff --git a/llvm/test/Transforms/InterleavedAccess/AArch64/sve-interleaved-accesses.ll b/llvm/test/Transforms/InterleavedAccess/AArch64/sve-interleaved-accesses.ll
index 45e2c36836ff..73f26814f3a4 100644
--- a/llvm/test/Transforms/InterleavedAccess/AArch64/sve-interleaved-accesses.ll
+++ b/llvm/test/Transforms/InterleavedAccess/AArch64/sve-interleaved-accesses.ll
@@ -511,7 +511,7 @@ define { <vscale x 4 x double>, <vscale x 4 x double> } @deinterleave_nxptr_fact
 ; CHECK-NEXT:    ret { <vscale x 4 x double>, <vscale x 4 x double> } [[TMP12]]
 ;
   %wide.vec = load <vscale x 8 x double>, ptr %ptr, align 8
-  %ldN = tail call { <vscale x 4 x double>, <vscale x 4 x double> } @llvm.experimental.vector.deinterleave2.nxv8f64(<vscale x 8 x double> %wide.vec)
+  %ldN = tail call { <vscale x 4 x double>, <vscale x 4 x double> } @llvm.vector.deinterleave2.nxv8f64(<vscale x 8 x double> %wide.vec)
   ret { <vscale x 4 x double>, <vscale x 4 x double> } %ldN
 }
 
diff --git a/llvm/test/Transforms/LoopUnroll/unroll-loads-cse.ll b/llvm/test/Transforms/LoopUnroll/unroll-loads-cse.ll
new file mode 100644
index 000000000000..109a1834c302
--- /dev/null
+++ b/llvm/test/Transforms/LoopUnroll/unroll-loads-cse.ll
@@ -0,0 +1,480 @@
+; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --version 4
+; RUN: opt -p loop-unroll -S %s | FileCheck %s
+
+target datalayout = "e-m:o-i64:64-i128:128-n32:64-S128"
+
+define void @cse_matching_load_from_previous_unrolled_iteration(ptr %src, ptr noalias %dst, i64 %N) {
+; CHECK-LABEL: define void @cse_matching_load_from_previous_unrolled_iteration(
+; CHECK-SAME: ptr [[SRC:%.*]], ptr noalias [[DST:%.*]], i64 [[N:%.*]]) {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[SRC_4:%.*]] = getelementptr i8, ptr [[SRC]], i64 4
+; CHECK-NEXT:    [[SRC_12:%.*]] = getelementptr i8, ptr [[SRC]], i64 12
+; CHECK-NEXT:    [[TMP0:%.*]] = add i64 [[N]], -1
+; CHECK-NEXT:    [[XTRAITER:%.*]] = and i64 [[N]], 1
+; CHECK-NEXT:    [[TMP1:%.*]] = icmp ult i64 [[TMP0]], 1
+; CHECK-NEXT:    br i1 [[TMP1]], label [[EXIT_UNR_LCSSA:%.*]], label [[ENTRY_NEW:%.*]]
+; CHECK:       entry.new:
+; CHECK-NEXT:    [[UNROLL_ITER:%.*]] = sub i64 [[N]], [[XTRAITER]]
+; CHECK-NEXT:    br label [[LOOP:%.*]]
+; CHECK:       loop:
+; CHECK-NEXT:    [[IV:%.*]] = phi i64 [ 0, [[ENTRY_NEW]] ], [ [[IV_NEXT_1:%.*]], [[LOOP]] ]
+; CHECK-NEXT:    [[NITER:%.*]] = phi i64 [ 0, [[ENTRY_NEW]] ], [ [[NITER_NEXT_1:%.*]], [[LOOP]] ]
+; CHECK-NEXT:    [[GEP_SRC_12:%.*]] = getelementptr i64, ptr [[SRC_12]], i64 [[IV]]
+; CHECK-NEXT:    [[L_12:%.*]] = load i64, ptr [[GEP_SRC_12]], align 8
+; CHECK-NEXT:    [[GEP_SRC_4:%.*]] = getelementptr i64, ptr [[SRC_4]], i64 [[IV]]
+; CHECK-NEXT:    [[L_4:%.*]] = load i64, ptr [[GEP_SRC_4]], align 8
+; CHECK-NEXT:    [[MUL:%.*]] = mul i64 [[L_12]], [[L_4]]
+; CHECK-NEXT:    [[GEP_DST:%.*]] = getelementptr i64, ptr [[DST]], i64 [[IV]]
+; CHECK-NEXT:    store i64 [[MUL]], ptr [[GEP_DST]], align 8
+; CHECK-NEXT:    [[IV_NEXT:%.*]] = add nuw nsw i64 [[IV]], 1
+; CHECK-NEXT:    [[GEP_SRC_12_1:%.*]] = getelementptr i64, ptr [[SRC_12]], i64 [[IV_NEXT]]
+; CHECK-NEXT:    [[L_12_1:%.*]] = load i64, ptr [[GEP_SRC_12_1]], align 8
+; CHECK-NEXT:    [[GEP_SRC_4_1:%.*]] = getelementptr i64, ptr [[SRC_4]], i64 [[IV_NEXT]]
+; CHECK-NEXT:    [[L_4_1:%.*]] = load i64, ptr [[GEP_SRC_4_1]], align 8
+; CHECK-NEXT:    [[MUL_1:%.*]] = mul i64 [[L_12_1]], [[L_4_1]]
+; CHECK-NEXT:    [[GEP_DST_1:%.*]] = getelementptr i64, ptr [[DST]], i64 [[IV_NEXT]]
+; CHECK-NEXT:    store i64 [[MUL_1]], ptr [[GEP_DST_1]], align 8
+; CHECK-NEXT:    [[IV_NEXT_1]] = add nuw nsw i64 [[IV]], 2
+; CHECK-NEXT:    [[NITER_NEXT_1]] = add i64 [[NITER]], 2
+; CHECK-NEXT:    [[NITER_NCMP_1:%.*]] = icmp eq i64 [[NITER_NEXT_1]], [[UNROLL_ITER]]
+; CHECK-NEXT:    br i1 [[NITER_NCMP_1]], label [[EXIT_UNR_LCSSA_LOOPEXIT:%.*]], label [[LOOP]], !llvm.loop [[LOOP0:![0-9]+]]
+; CHECK:       exit.unr-lcssa.loopexit:
+; CHECK-NEXT:    [[IV_UNR_PH:%.*]] = phi i64 [ [[IV_NEXT_1]], [[LOOP]] ]
+; CHECK-NEXT:    br label [[EXIT_UNR_LCSSA]]
+; CHECK:       exit.unr-lcssa:
+; CHECK-NEXT:    [[IV_UNR:%.*]] = phi i64 [ 0, [[ENTRY:%.*]] ], [ [[IV_UNR_PH]], [[EXIT_UNR_LCSSA_LOOPEXIT]] ]
+; CHECK-NEXT:    [[LCMP_MOD:%.*]] = icmp ne i64 [[XTRAITER]], 0
+; CHECK-NEXT:    br i1 [[LCMP_MOD]], label [[LOOP_EPIL_PREHEADER:%.*]], label [[EXIT:%.*]]
+; CHECK:       loop.epil.preheader:
+; CHECK-NEXT:    br label [[LOOP_EPIL:%.*]]
+; CHECK:       loop.epil:
+; CHECK-NEXT:    [[GEP_SRC_12_EPIL:%.*]] = getelementptr i64, ptr [[SRC_12]], i64 [[IV_UNR]]
+; CHECK-NEXT:    [[L_12_EPIL:%.*]] = load i64, ptr [[GEP_SRC_12_EPIL]], align 8
+; CHECK-NEXT:    [[GEP_SRC_4_EPIL:%.*]] = getelementptr i64, ptr [[SRC_4]], i64 [[IV_UNR]]
+; CHECK-NEXT:    [[L_4_EPIL:%.*]] = load i64, ptr [[GEP_SRC_4_EPIL]], align 8
+; CHECK-NEXT:    [[MUL_EPIL:%.*]] = mul i64 [[L_12_EPIL]], [[L_4_EPIL]]
+; CHECK-NEXT:    [[GEP_DST_EPIL:%.*]] = getelementptr i64, ptr [[DST]], i64 [[IV_UNR]]
+; CHECK-NEXT:    store i64 [[MUL_EPIL]], ptr [[GEP_DST_EPIL]], align 8
+; CHECK-NEXT:    br label [[EXIT]]
+; CHECK:       exit:
+; CHECK-NEXT:    ret void
+;
+entry:
+  %src.4 = getelementptr i8, ptr %src, i64 4
+  %src.12 = getelementptr i8, ptr %src, i64 12
+  br label %loop
+
+loop:
+  %iv = phi i64 [ 0, %entry ], [ %iv.next, %loop ]
+  %gep.src.12 = getelementptr i64, ptr %src.12, i64 %iv
+  %l.12 = load i64, ptr %gep.src.12, align 8
+  %gep.src.4 = getelementptr i64, ptr %src.4, i64 %iv
+  %l.4 = load i64, ptr %gep.src.4, align 8
+  %mul = mul i64 %l.12, %l.4
+  %gep.dst = getelementptr i64, ptr %dst, i64 %iv
+  store i64 %mul, ptr %gep.dst
+  %iv.next = add nuw nsw i64 %iv, 1
+  %c = icmp eq i64 %iv.next, %N
+  br i1 %c, label %exit, label %loop, !llvm.loop !1
+
+exit:
+  ret void
+}
+
+define void @cse_different_load_types(ptr %src, ptr noalias %dst, i64 %N) {
+; CHECK-LABEL: define void @cse_different_load_types(
+; CHECK-SAME: ptr [[SRC:%.*]], ptr noalias [[DST:%.*]], i64 [[N:%.*]]) {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[SRC_4:%.*]] = getelementptr i8, ptr [[SRC]], i64 4
+; CHECK-NEXT:    [[SRC_12:%.*]] = getelementptr i8, ptr [[SRC]], i64 12
+; CHECK-NEXT:    [[TMP0:%.*]] = add i64 [[N]], -1
+; CHECK-NEXT:    [[XTRAITER:%.*]] = and i64 [[N]], 1
+; CHECK-NEXT:    [[TMP1:%.*]] = icmp ult i64 [[TMP0]], 1
+; CHECK-NEXT:    br i1 [[TMP1]], label [[EXIT_UNR_LCSSA:%.*]], label [[ENTRY_NEW:%.*]]
+; CHECK:       entry.new:
+; CHECK-NEXT:    [[UNROLL_ITER:%.*]] = sub i64 [[N]], [[XTRAITER]]
+; CHECK-NEXT:    br label [[LOOP:%.*]]
+; CHECK:       loop:
+; CHECK-NEXT:    [[IV:%.*]] = phi i64 [ 0, [[ENTRY_NEW]] ], [ [[IV_NEXT_1:%.*]], [[LOOP]] ]
+; CHECK-NEXT:    [[NITER:%.*]] = phi i64 [ 0, [[ENTRY_NEW]] ], [ [[NITER_NEXT_1:%.*]], [[LOOP]] ]
+; CHECK-NEXT:    [[GEP_SRC_12:%.*]] = getelementptr i64, ptr [[SRC_12]], i64 [[IV]]
+; CHECK-NEXT:    [[L_12:%.*]] = load i32, ptr [[GEP_SRC_12]], align 8
+; CHECK-NEXT:    [[L_12_EXT:%.*]] = zext i32 [[L_12]] to i64
+; CHECK-NEXT:    [[GEP_SRC_4:%.*]] = getelementptr i64, ptr [[SRC_4]], i64 [[IV]]
+; CHECK-NEXT:    [[L_4:%.*]] = load i64, ptr [[GEP_SRC_4]], align 8
+; CHECK-NEXT:    [[MUL:%.*]] = mul i64 [[L_12_EXT]], [[L_4]]
+; CHECK-NEXT:    [[GEP_DST:%.*]] = getelementptr i64, ptr [[DST]], i64 [[IV]]
+; CHECK-NEXT:    store i64 [[MUL]], ptr [[GEP_DST]], align 8
+; CHECK-NEXT:    [[IV_NEXT:%.*]] = add nuw nsw i64 [[IV]], 1
+; CHECK-NEXT:    [[GEP_SRC_12_1:%.*]] = getelementptr i64, ptr [[SRC_12]], i64 [[IV_NEXT]]
+; CHECK-NEXT:    [[L_12_1:%.*]] = load i32, ptr [[GEP_SRC_12_1]], align 8
+; CHECK-NEXT:    [[L_12_EXT_1:%.*]] = zext i32 [[L_12_1]] to i64
+; CHECK-NEXT:    [[GEP_SRC_4_1:%.*]] = getelementptr i64, ptr [[SRC_4]], i64 [[IV_NEXT]]
+; CHECK-NEXT:    [[L_4_1:%.*]] = load i64, ptr [[GEP_SRC_4_1]], align 8
+; CHECK-NEXT:    [[MUL_1:%.*]] = mul i64 [[L_12_EXT_1]], [[L_4_1]]
+; CHECK-NEXT:    [[GEP_DST_1:%.*]] = getelementptr i64, ptr [[DST]], i64 [[IV_NEXT]]
+; CHECK-NEXT:    store i64 [[MUL_1]], ptr [[GEP_DST_1]], align 8
+; CHECK-NEXT:    [[IV_NEXT_1]] = add nuw nsw i64 [[IV]], 2
+; CHECK-NEXT:    [[NITER_NEXT_1]] = add i64 [[NITER]], 2
+; CHECK-NEXT:    [[NITER_NCMP_1:%.*]] = icmp eq i64 [[NITER_NEXT_1]], [[UNROLL_ITER]]
+; CHECK-NEXT:    br i1 [[NITER_NCMP_1]], label [[EXIT_UNR_LCSSA_LOOPEXIT:%.*]], label [[LOOP]], !llvm.loop [[LOOP3:![0-9]+]]
+; CHECK:       exit.unr-lcssa.loopexit:
+; CHECK-NEXT:    [[IV_UNR_PH:%.*]] = phi i64 [ [[IV_NEXT_1]], [[LOOP]] ]
+; CHECK-NEXT:    br label [[EXIT_UNR_LCSSA]]
+; CHECK:       exit.unr-lcssa:
+; CHECK-NEXT:    [[IV_UNR:%.*]] = phi i64 [ 0, [[ENTRY:%.*]] ], [ [[IV_UNR_PH]], [[EXIT_UNR_LCSSA_LOOPEXIT]] ]
+; CHECK-NEXT:    [[LCMP_MOD:%.*]] = icmp ne i64 [[XTRAITER]], 0
+; CHECK-NEXT:    br i1 [[LCMP_MOD]], label [[LOOP_EPIL_PREHEADER:%.*]], label [[EXIT:%.*]]
+; CHECK:       loop.epil.preheader:
+; CHECK-NEXT:    br label [[LOOP_EPIL:%.*]]
+; CHECK:       loop.epil:
+; CHECK-NEXT:    [[GEP_SRC_12_EPIL:%.*]] = getelementptr i64, ptr [[SRC_12]], i64 [[IV_UNR]]
+; CHECK-NEXT:    [[L_12_EPIL:%.*]] = load i32, ptr [[GEP_SRC_12_EPIL]], align 8
+; CHECK-NEXT:    [[L_12_EXT_EPIL:%.*]] = zext i32 [[L_12_EPIL]] to i64
+; CHECK-NEXT:    [[GEP_SRC_4_EPIL:%.*]] = getelementptr i64, ptr [[SRC_4]], i64 [[IV_UNR]]
+; CHECK-NEXT:    [[L_4_EPIL:%.*]] = load i64, ptr [[GEP_SRC_4_EPIL]], align 8
+; CHECK-NEXT:    [[MUL_EPIL:%.*]] = mul i64 [[L_12_EXT_EPIL]], [[L_4_EPIL]]
+; CHECK-NEXT:    [[GEP_DST_EPIL:%.*]] = getelementptr i64, ptr [[DST]], i64 [[IV_UNR]]
+; CHECK-NEXT:    store i64 [[MUL_EPIL]], ptr [[GEP_DST_EPIL]], align 8
+; CHECK-NEXT:    br label [[EXIT]]
+; CHECK:       exit:
+; CHECK-NEXT:    ret void
+;
+entry:
+  %src.4 = getelementptr i8, ptr %src, i64 4
+  %src.12 = getelementptr i8, ptr %src, i64 12
+  br label %loop
+
+loop:
+  %iv = phi i64 [ 0, %entry ], [ %iv.next, %loop ]
+  %gep.src.12 = getelementptr i64, ptr %src.12, i64 %iv
+  %l.12 = load i32, ptr %gep.src.12, align 8
+  %l.12.ext = zext i32 %l.12 to i64
+  %gep.src.4 = getelementptr i64, ptr %src.4, i64 %iv
+  %l.4 = load i64, ptr %gep.src.4, align 8
+  %mul = mul i64 %l.12.ext, %l.4
+  %gep.dst = getelementptr i64, ptr %dst, i64 %iv
+  store i64 %mul, ptr %gep.dst
+  %iv.next = add nuw nsw i64 %iv, 1
+  %c = icmp eq i64 %iv.next, %N
+  br i1 %c, label %exit, label %loop, !llvm.loop !1
+
+exit:
+  ret void
+}
+
+define void @cse_volatile_loads(ptr %src, ptr noalias %dst, i64 %N) {
+; CHECK-LABEL: define void @cse_volatile_loads(
+; CHECK-SAME: ptr [[SRC:%.*]], ptr noalias [[DST:%.*]], i64 [[N:%.*]]) {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[SRC_4:%.*]] = getelementptr i8, ptr [[SRC]], i64 4
+; CHECK-NEXT:    [[SRC_12:%.*]] = getelementptr i8, ptr [[SRC]], i64 12
+; CHECK-NEXT:    [[TMP0:%.*]] = add i64 [[N]], -1
+; CHECK-NEXT:    [[XTRAITER:%.*]] = and i64 [[N]], 1
+; CHECK-NEXT:    [[TMP1:%.*]] = icmp ult i64 [[TMP0]], 1
+; CHECK-NEXT:    br i1 [[TMP1]], label [[EXIT_UNR_LCSSA:%.*]], label [[ENTRY_NEW:%.*]]
+; CHECK:       entry.new:
+; CHECK-NEXT:    [[UNROLL_ITER:%.*]] = sub i64 [[N]], [[XTRAITER]]
+; CHECK-NEXT:    br label [[LOOP:%.*]]
+; CHECK:       loop:
+; CHECK-NEXT:    [[IV:%.*]] = phi i64 [ 0, [[ENTRY_NEW]] ], [ [[IV_NEXT_1:%.*]], [[LOOP]] ]
+; CHECK-NEXT:    [[NITER:%.*]] = phi i64 [ 0, [[ENTRY_NEW]] ], [ [[NITER_NEXT_1:%.*]], [[LOOP]] ]
+; CHECK-NEXT:    [[GEP_SRC_12:%.*]] = getelementptr i64, ptr [[SRC_12]], i64 [[IV]]
+; CHECK-NEXT:    [[L_12:%.*]] = load i64, ptr [[GEP_SRC_12]], align 8
+; CHECK-NEXT:    [[GEP_SRC_4:%.*]] = getelementptr i64, ptr [[SRC_4]], i64 [[IV]]
+; CHECK-NEXT:    [[L_4:%.*]] = load volatile i64, ptr [[GEP_SRC_4]], align 8
+; CHECK-NEXT:    [[MUL:%.*]] = mul i64 [[L_12]], [[L_4]]
+; CHECK-NEXT:    [[GEP_DST:%.*]] = getelementptr i64, ptr [[DST]], i64 [[IV]]
+; CHECK-NEXT:    store i64 [[MUL]], ptr [[GEP_DST]], align 8
+; CHECK-NEXT:    [[IV_NEXT:%.*]] = add nuw nsw i64 [[IV]], 1
+; CHECK-NEXT:    [[GEP_SRC_12_1:%.*]] = getelementptr i64, ptr [[SRC_12]], i64 [[IV_NEXT]]
+; CHECK-NEXT:    [[L_12_1:%.*]] = load i64, ptr [[GEP_SRC_12_1]], align 8
+; CHECK-NEXT:    [[GEP_SRC_4_1:%.*]] = getelementptr i64, ptr [[SRC_4]], i64 [[IV_NEXT]]
+; CHECK-NEXT:    [[L_4_1:%.*]] = load volatile i64, ptr [[GEP_SRC_4_1]], align 8
+; CHECK-NEXT:    [[MUL_1:%.*]] = mul i64 [[L_12_1]], [[L_4_1]]
+; CHECK-NEXT:    [[GEP_DST_1:%.*]] = getelementptr i64, ptr [[DST]], i64 [[IV_NEXT]]
+; CHECK-NEXT:    store i64 [[MUL_1]], ptr [[GEP_DST_1]], align 8
+; CHECK-NEXT:    [[IV_NEXT_1]] = add nuw nsw i64 [[IV]], 2
+; CHECK-NEXT:    [[NITER_NEXT_1]] = add i64 [[NITER]], 2
+; CHECK-NEXT:    [[NITER_NCMP_1:%.*]] = icmp eq i64 [[NITER_NEXT_1]], [[UNROLL_ITER]]
+; CHECK-NEXT:    br i1 [[NITER_NCMP_1]], label [[EXIT_UNR_LCSSA_LOOPEXIT:%.*]], label [[LOOP]], !llvm.loop [[LOOP4:![0-9]+]]
+; CHECK:       exit.unr-lcssa.loopexit:
+; CHECK-NEXT:    [[IV_UNR_PH:%.*]] = phi i64 [ [[IV_NEXT_1]], [[LOOP]] ]
+; CHECK-NEXT:    br label [[EXIT_UNR_LCSSA]]
+; CHECK:       exit.unr-lcssa:
+; CHECK-NEXT:    [[IV_UNR:%.*]] = phi i64 [ 0, [[ENTRY:%.*]] ], [ [[IV_UNR_PH]], [[EXIT_UNR_LCSSA_LOOPEXIT]] ]
+; CHECK-NEXT:    [[LCMP_MOD:%.*]] = icmp ne i64 [[XTRAITER]], 0
+; CHECK-NEXT:    br i1 [[LCMP_MOD]], label [[LOOP_EPIL_PREHEADER:%.*]], label [[EXIT:%.*]]
+; CHECK:       loop.epil.preheader:
+; CHECK-NEXT:    br label [[LOOP_EPIL:%.*]]
+; CHECK:       loop.epil:
+; CHECK-NEXT:    [[GEP_SRC_12_EPIL:%.*]] = getelementptr i64, ptr [[SRC_12]], i64 [[IV_UNR]]
+; CHECK-NEXT:    [[L_12_EPIL:%.*]] = load i64, ptr [[GEP_SRC_12_EPIL]], align 8
+; CHECK-NEXT:    [[GEP_SRC_4_EPIL:%.*]] = getelementptr i64, ptr [[SRC_4]], i64 [[IV_UNR]]
+; CHECK-NEXT:    [[L_4_EPIL:%.*]] = load volatile i64, ptr [[GEP_SRC_4_EPIL]], align 8
+; CHECK-NEXT:    [[MUL_EPIL:%.*]] = mul i64 [[L_12_EPIL]], [[L_4_EPIL]]
+; CHECK-NEXT:    [[GEP_DST_EPIL:%.*]] = getelementptr i64, ptr [[DST]], i64 [[IV_UNR]]
+; CHECK-NEXT:    store i64 [[MUL_EPIL]], ptr [[GEP_DST_EPIL]], align 8
+; CHECK-NEXT:    br label [[EXIT]]
+; CHECK:       exit:
+; CHECK-NEXT:    ret void
+;
+entry:
+  %src.4 = getelementptr i8, ptr %src, i64 4
+  %src.12 = getelementptr i8, ptr %src, i64 12
+  br label %loop
+
+loop:
+  %iv = phi i64 [ 0, %entry ], [ %iv.next, %loop ]
+  %gep.src.12 = getelementptr i64, ptr %src.12, i64 %iv
+  %l.12 = load i64, ptr %gep.src.12, align 8
+  %gep.src.4 = getelementptr i64, ptr %src.4, i64 %iv
+  %l.4 = load volatile i64, ptr %gep.src.4, align 8
+  %mul = mul i64 %l.12, %l.4
+  %gep.dst = getelementptr i64, ptr %dst, i64 %iv
+  store i64 %mul, ptr %gep.dst
+  %iv.next = add nuw nsw i64 %iv, 1
+  %c = icmp eq i64 %iv.next, %N
+  br i1 %c, label %exit, label %loop, !llvm.loop !1
+
+exit:
+  ret void
+}
+
+define void @cse_atomic_loads(ptr %src, ptr noalias %dst, i64 %N) {
+; CHECK-LABEL: define void @cse_atomic_loads(
+; CHECK-SAME: ptr [[SRC:%.*]], ptr noalias [[DST:%.*]], i64 [[N:%.*]]) {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[SRC_4:%.*]] = getelementptr i8, ptr [[SRC]], i64 4
+; CHECK-NEXT:    [[SRC_12:%.*]] = getelementptr i8, ptr [[SRC]], i64 12
+; CHECK-NEXT:    [[TMP0:%.*]] = add i64 [[N]], -1
+; CHECK-NEXT:    [[XTRAITER:%.*]] = and i64 [[N]], 1
+; CHECK-NEXT:    [[TMP1:%.*]] = icmp ult i64 [[TMP0]], 1
+; CHECK-NEXT:    br i1 [[TMP1]], label [[EXIT_UNR_LCSSA:%.*]], label [[ENTRY_NEW:%.*]]
+; CHECK:       entry.new:
+; CHECK-NEXT:    [[UNROLL_ITER:%.*]] = sub i64 [[N]], [[XTRAITER]]
+; CHECK-NEXT:    br label [[LOOP:%.*]]
+; CHECK:       loop:
+; CHECK-NEXT:    [[IV:%.*]] = phi i64 [ 0, [[ENTRY_NEW]] ], [ [[IV_NEXT_1:%.*]], [[LOOP]] ]
+; CHECK-NEXT:    [[NITER:%.*]] = phi i64 [ 0, [[ENTRY_NEW]] ], [ [[NITER_NEXT_1:%.*]], [[LOOP]] ]
+; CHECK-NEXT:    [[GEP_SRC_12:%.*]] = getelementptr i64, ptr [[SRC_12]], i64 [[IV]]
+; CHECK-NEXT:    [[L_12:%.*]] = load i64, ptr [[GEP_SRC_12]], align 8
+; CHECK-NEXT:    [[GEP_SRC_4:%.*]] = getelementptr i64, ptr [[SRC_4]], i64 [[IV]]
+; CHECK-NEXT:    [[L_4:%.*]] = load atomic i64, ptr [[GEP_SRC_4]] unordered, align 8
+; CHECK-NEXT:    [[MUL:%.*]] = mul i64 [[L_12]], [[L_4]]
+; CHECK-NEXT:    [[GEP_DST:%.*]] = getelementptr i64, ptr [[DST]], i64 [[IV]]
+; CHECK-NEXT:    store i64 [[MUL]], ptr [[GEP_DST]], align 8
+; CHECK-NEXT:    [[IV_NEXT:%.*]] = add nuw nsw i64 [[IV]], 1
+; CHECK-NEXT:    [[GEP_SRC_12_1:%.*]] = getelementptr i64, ptr [[SRC_12]], i64 [[IV_NEXT]]
+; CHECK-NEXT:    [[L_12_1:%.*]] = load i64, ptr [[GEP_SRC_12_1]], align 8
+; CHECK-NEXT:    [[GEP_SRC_4_1:%.*]] = getelementptr i64, ptr [[SRC_4]], i64 [[IV_NEXT]]
+; CHECK-NEXT:    [[L_4_1:%.*]] = load atomic i64, ptr [[GEP_SRC_4_1]] unordered, align 8
+; CHECK-NEXT:    [[MUL_1:%.*]] = mul i64 [[L_12_1]], [[L_4_1]]
+; CHECK-NEXT:    [[GEP_DST_1:%.*]] = getelementptr i64, ptr [[DST]], i64 [[IV_NEXT]]
+; CHECK-NEXT:    store i64 [[MUL_1]], ptr [[GEP_DST_1]], align 8
+; CHECK-NEXT:    [[IV_NEXT_1]] = add nuw nsw i64 [[IV]], 2
+; CHECK-NEXT:    [[NITER_NEXT_1]] = add i64 [[NITER]], 2
+; CHECK-NEXT:    [[NITER_NCMP_1:%.*]] = icmp eq i64 [[NITER_NEXT_1]], [[UNROLL_ITER]]
+; CHECK-NEXT:    br i1 [[NITER_NCMP_1]], label [[EXIT_UNR_LCSSA_LOOPEXIT:%.*]], label [[LOOP]], !llvm.loop [[LOOP5:![0-9]+]]
+; CHECK:       exit.unr-lcssa.loopexit:
+; CHECK-NEXT:    [[IV_UNR_PH:%.*]] = phi i64 [ [[IV_NEXT_1]], [[LOOP]] ]
+; CHECK-NEXT:    br label [[EXIT_UNR_LCSSA]]
+; CHECK:       exit.unr-lcssa:
+; CHECK-NEXT:    [[IV_UNR:%.*]] = phi i64 [ 0, [[ENTRY:%.*]] ], [ [[IV_UNR_PH]], [[EXIT_UNR_LCSSA_LOOPEXIT]] ]
+; CHECK-NEXT:    [[LCMP_MOD:%.*]] = icmp ne i64 [[XTRAITER]], 0
+; CHECK-NEXT:    br i1 [[LCMP_MOD]], label [[LOOP_EPIL_PREHEADER:%.*]], label [[EXIT:%.*]]
+; CHECK:       loop.epil.preheader:
+; CHECK-NEXT:    br label [[LOOP_EPIL:%.*]]
+; CHECK:       loop.epil:
+; CHECK-NEXT:    [[GEP_SRC_12_EPIL:%.*]] = getelementptr i64, ptr [[SRC_12]], i64 [[IV_UNR]]
+; CHECK-NEXT:    [[L_12_EPIL:%.*]] = load i64, ptr [[GEP_SRC_12_EPIL]], align 8
+; CHECK-NEXT:    [[GEP_SRC_4_EPIL:%.*]] = getelementptr i64, ptr [[SRC_4]], i64 [[IV_UNR]]
+; CHECK-NEXT:    [[L_4_EPIL:%.*]] = load atomic i64, ptr [[GEP_SRC_4_EPIL]] unordered, align 8
+; CHECK-NEXT:    [[MUL_EPIL:%.*]] = mul i64 [[L_12_EPIL]], [[L_4_EPIL]]
+; CHECK-NEXT:    [[GEP_DST_EPIL:%.*]] = getelementptr i64, ptr [[DST]], i64 [[IV_UNR]]
+; CHECK-NEXT:    store i64 [[MUL_EPIL]], ptr [[GEP_DST_EPIL]], align 8
+; CHECK-NEXT:    br label [[EXIT]]
+; CHECK:       exit:
+; CHECK-NEXT:    ret void
+;
+entry:
+  %src.4 = getelementptr i8, ptr %src, i64 4
+  %src.12 = getelementptr i8, ptr %src, i64 12
+  br label %loop
+
+loop:
+  %iv = phi i64 [ 0, %entry ], [ %iv.next, %loop ]
+  %gep.src.12 = getelementptr i64, ptr %src.12, i64 %iv
+  %l.12 = load i64, ptr %gep.src.12, align 8
+  %gep.src.4 = getelementptr i64, ptr %src.4, i64 %iv
+  %l.4 = load atomic i64, ptr %gep.src.4 unordered, align 8
+  %mul = mul i64 %l.12, %l.4
+  %gep.dst = getelementptr i64, ptr %dst, i64 %iv
+  store i64 %mul, ptr %gep.dst
+  %iv.next = add nuw nsw i64 %iv, 1
+  %c = icmp eq i64 %iv.next, %N
+  br i1 %c, label %exit, label %loop, !llvm.loop !1
+
+exit:
+  ret void
+}
+
+define void @cse_load_may_be_clobbered(ptr %src, ptr %dst, i64 %N) {
+; CHECK-LABEL: define void @cse_load_may_be_clobbered(
+; CHECK-SAME: ptr [[SRC:%.*]], ptr [[DST:%.*]], i64 [[N:%.*]]) {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[SRC_4:%.*]] = getelementptr i8, ptr [[SRC]], i64 4
+; CHECK-NEXT:    [[SRC_12:%.*]] = getelementptr i8, ptr [[SRC]], i64 12
+; CHECK-NEXT:    [[TMP0:%.*]] = add i64 [[N]], -1
+; CHECK-NEXT:    [[XTRAITER:%.*]] = and i64 [[N]], 1
+; CHECK-NEXT:    [[TMP1:%.*]] = icmp ult i64 [[TMP0]], 1
+; CHECK-NEXT:    br i1 [[TMP1]], label [[EXIT_UNR_LCSSA:%.*]], label [[ENTRY_NEW:%.*]]
+; CHECK:       entry.new:
+; CHECK-NEXT:    [[UNROLL_ITER:%.*]] = sub i64 [[N]], [[XTRAITER]]
+; CHECK-NEXT:    br label [[LOOP:%.*]]
+; CHECK:       loop:
+; CHECK-NEXT:    [[IV:%.*]] = phi i64 [ 0, [[ENTRY_NEW]] ], [ [[IV_NEXT_1:%.*]], [[LOOP]] ]
+; CHECK-NEXT:    [[NITER:%.*]] = phi i64 [ 0, [[ENTRY_NEW]] ], [ [[NITER_NEXT_1:%.*]], [[LOOP]] ]
+; CHECK-NEXT:    [[GEP_SRC_12:%.*]] = getelementptr i64, ptr [[SRC_12]], i64 [[IV]]
+; CHECK-NEXT:    [[L_12:%.*]] = load i64, ptr [[GEP_SRC_12]], align 8
+; CHECK-NEXT:    [[GEP_SRC_4:%.*]] = getelementptr i64, ptr [[SRC_4]], i64 [[IV]]
+; CHECK-NEXT:    [[L_4:%.*]] = load i64, ptr [[GEP_SRC_4]], align 8
+; CHECK-NEXT:    [[MUL:%.*]] = mul i64 [[L_12]], [[L_4]]
+; CHECK-NEXT:    [[GEP_DST:%.*]] = getelementptr i64, ptr [[DST]], i64 [[IV]]
+; CHECK-NEXT:    store i64 [[MUL]], ptr [[GEP_DST]], align 8
+; CHECK-NEXT:    [[IV_NEXT:%.*]] = add nuw nsw i64 [[IV]], 1
+; CHECK-NEXT:    [[GEP_SRC_12_1:%.*]] = getelementptr i64, ptr [[SRC_12]], i64 [[IV_NEXT]]
+; CHECK-NEXT:    [[L_12_1:%.*]] = load i64, ptr [[GEP_SRC_12_1]], align 8
+; CHECK-NEXT:    [[GEP_SRC_4_1:%.*]] = getelementptr i64, ptr [[SRC_4]], i64 [[IV_NEXT]]
+; CHECK-NEXT:    [[L_4_1:%.*]] = load i64, ptr [[GEP_SRC_4_1]], align 8
+; CHECK-NEXT:    [[MUL_1:%.*]] = mul i64 [[L_12_1]], [[L_4_1]]
+; CHECK-NEXT:    [[GEP_DST_1:%.*]] = getelementptr i64, ptr [[DST]], i64 [[IV_NEXT]]
+; CHECK-NEXT:    store i64 [[MUL_1]], ptr [[GEP_DST_1]], align 8
+; CHECK-NEXT:    [[IV_NEXT_1]] = add nuw nsw i64 [[IV]], 2
+; CHECK-NEXT:    [[NITER_NEXT_1]] = add i64 [[NITER]], 2
+; CHECK-NEXT:    [[NITER_NCMP_1:%.*]] = icmp eq i64 [[NITER_NEXT_1]], [[UNROLL_ITER]]
+; CHECK-NEXT:    br i1 [[NITER_NCMP_1]], label [[EXIT_UNR_LCSSA_LOOPEXIT:%.*]], label [[LOOP]], !llvm.loop [[LOOP6:![0-9]+]]
+; CHECK:       exit.unr-lcssa.loopexit:
+; CHECK-NEXT:    [[IV_UNR_PH:%.*]] = phi i64 [ [[IV_NEXT_1]], [[LOOP]] ]
+; CHECK-NEXT:    br label [[EXIT_UNR_LCSSA]]
+; CHECK:       exit.unr-lcssa:
+; CHECK-NEXT:    [[IV_UNR:%.*]] = phi i64 [ 0, [[ENTRY:%.*]] ], [ [[IV_UNR_PH]], [[EXIT_UNR_LCSSA_LOOPEXIT]] ]
+; CHECK-NEXT:    [[LCMP_MOD:%.*]] = icmp ne i64 [[XTRAITER]], 0
+; CHECK-NEXT:    br i1 [[LCMP_MOD]], label [[LOOP_EPIL_PREHEADER:%.*]], label [[EXIT:%.*]]
+; CHECK:       loop.epil.preheader:
+; CHECK-NEXT:    br label [[LOOP_EPIL:%.*]]
+; CHECK:       loop.epil:
+; CHECK-NEXT:    [[GEP_SRC_12_EPIL:%.*]] = getelementptr i64, ptr [[SRC_12]], i64 [[IV_UNR]]
+; CHECK-NEXT:    [[L_12_EPIL:%.*]] = load i64, ptr [[GEP_SRC_12_EPIL]], align 8
+; CHECK-NEXT:    [[GEP_SRC_4_EPIL:%.*]] = getelementptr i64, ptr [[SRC_4]], i64 [[IV_UNR]]
+; CHECK-NEXT:    [[L_4_EPIL:%.*]] = load i64, ptr [[GEP_SRC_4_EPIL]], align 8
+; CHECK-NEXT:    [[MUL_EPIL:%.*]] = mul i64 [[L_12_EPIL]], [[L_4_EPIL]]
+; CHECK-NEXT:    [[GEP_DST_EPIL:%.*]] = getelementptr i64, ptr [[DST]], i64 [[IV_UNR]]
+; CHECK-NEXT:    store i64 [[MUL_EPIL]], ptr [[GEP_DST_EPIL]], align 8
+; CHECK-NEXT:    br label [[EXIT]]
+; CHECK:       exit:
+; CHECK-NEXT:    ret void
+;
+entry:
+  %src.4 = getelementptr i8, ptr %src, i64 4
+  %src.12 = getelementptr i8, ptr %src, i64 12
+  br label %loop
+
+loop:
+  %iv = phi i64 [ 0, %entry ], [ %iv.next, %loop ]
+  %gep.src.12 = getelementptr i64, ptr %src.12, i64 %iv
+  %l.12 = load i64, ptr %gep.src.12, align 8
+  %gep.src.4 = getelementptr i64, ptr %src.4, i64 %iv
+  %l.4 = load i64, ptr %gep.src.4, align 8
+  %mul = mul i64 %l.12, %l.4
+  %gep.dst = getelementptr i64, ptr %dst, i64 %iv
+  store i64 %mul, ptr %gep.dst
+  %iv.next = add nuw nsw i64 %iv, 1
+  %c = icmp eq i64 %iv.next, %N
+  br i1 %c, label %exit, label %loop, !llvm.loop !1
+
+exit:
+  ret void
+}
+
+
+declare void @foo()
+
+define void @loop_body_with_dead_blocks(ptr %src) {
+; CHECK-LABEL: define void @loop_body_with_dead_blocks(
+; CHECK-SAME: ptr [[SRC:%.*]]) {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    br label [[OUTER_HEADER:%.*]]
+; CHECK:       outer.header.loopexit:
+; CHECK-NEXT:    br label [[OUTER_HEADER]]
+; CHECK:       outer.header:
+; CHECK-NEXT:    br label [[LOOP_HEADER:%.*]]
+; CHECK:       loop.header:
+; CHECK-NEXT:    br label [[LOOP_BB:%.*]]
+; CHECK:       loop.bb.dead:
+; CHECK-NEXT:    unreachable
+; CHECK:       loop.bb:
+; CHECK-NEXT:    [[L_1:%.*]] = load i32, ptr [[SRC]], align 8
+; CHECK-NEXT:    [[C_1:%.*]] = icmp eq i32 [[L_1]], 0
+; CHECK-NEXT:    br i1 [[C_1]], label [[OUTER_HEADER_LOOPEXIT:%.*]], label [[LOOP_LATCH:%.*]]
+; CHECK:       loop.latch:
+; CHECK-NEXT:    call void @foo()
+; CHECK-NEXT:    [[L_2:%.*]] = load i32, ptr [[SRC]], align 8
+; CHECK-NEXT:    [[C_2:%.*]] = icmp eq i32 [[L_2]], 1
+; CHECK-NEXT:    br i1 [[C_2]], label [[EXIT:%.*]], label [[LOOP_HEADER_1:%.*]], !llvm.loop [[LOOP7:![0-9]+]]
+; CHECK:       loop.header.1:
+; CHECK-NEXT:    br label [[LOOP_BB_1:%.*]]
+; CHECK:       loop.bb.1:
+; CHECK-NEXT:    [[L_1_1:%.*]] = load i32, ptr [[SRC]], align 8
+; CHECK-NEXT:    [[C_1_1:%.*]] = icmp eq i32 [[L_1_1]], 0
+; CHECK-NEXT:    br i1 [[C_1_1]], label [[OUTER_HEADER_LOOPEXIT]], label [[LOOP_LATCH_1:%.*]]
+; CHECK:       loop.latch.1:
+; CHECK-NEXT:    call void @foo()
+; CHECK-NEXT:    [[L_2_1:%.*]] = load i32, ptr [[SRC]], align 8
+; CHECK-NEXT:    [[C_2_1:%.*]] = icmp eq i32 [[L_2_1]], 1
+; CHECK-NEXT:    br i1 [[C_2_1]], label [[EXIT]], label [[LOOP_HEADER]], !llvm.loop [[LOOP9:![0-9]+]]
+; CHECK:       exit:
+; CHECK-NEXT:    ret void
+;
+entry:
+  br label %outer.header
+
+outer.header:
+  br label %loop.header
+
+loop.header:
+  br label %loop.bb
+
+loop.bb.dead:
+  br label %loop.bb
+
+loop.bb:
+  %l.1 = load i32, ptr %src, align 8
+  %c.1 = icmp eq i32 %l.1, 0
+  br i1 %c.1, label %outer.header, label %loop.latch
+
+loop.latch:
+  call void @foo()
+  %l.2 = load i32, ptr %src, align 8
+  %c.2 = icmp eq i32 %l.2, 1
+  br i1 %c.2, label %exit, label %loop.header, !llvm.loop !1
+
+exit:
+  ret void
+}
+
+!0 = !{!"llvm.loop.mustprogress"}
+!1 = distinct !{!1, !0, !2}
+!2 = !{!"llvm.loop.unroll.count", i32 2}
+;.
+; CHECK: [[LOOP0]] = distinct !{[[LOOP0]], [[META1:![0-9]+]], [[META2:![0-9]+]]}
+; CHECK: [[META1]] = !{!"llvm.loop.mustprogress"}
+; CHECK: [[META2]] = !{!"llvm.loop.unroll.disable"}
+; CHECK: [[LOOP3]] = distinct !{[[LOOP3]], [[META1]], [[META2]]}
+; CHECK: [[LOOP4]] = distinct !{[[LOOP4]], [[META1]], [[META2]]}
+; CHECK: [[LOOP5]] = distinct !{[[LOOP5]], [[META1]], [[META2]]}
+; CHECK: [[LOOP6]] = distinct !{[[LOOP6]], [[META1]], [[META2]]}
+; CHECK: [[LOOP7]] = distinct !{[[LOOP7]], [[META1]], [[META8:![0-9]+]]}
+; CHECK: [[META8]] = !{!"llvm.loop.unroll.count", i32 2}
+; CHECK: [[LOOP9]] = distinct !{[[LOOP9]], [[META1]], [[META2]]}
+;.
diff --git a/llvm/test/Transforms/LoopUnroll/unroll-remove-redundant-dbg.ll b/llvm/test/Transforms/LoopUnroll/unroll-remove-redundant-dbg.ll
new file mode 100644
index 000000000000..66cd4d454443
--- /dev/null
+++ b/llvm/test/Transforms/LoopUnroll/unroll-remove-redundant-dbg.ll
@@ -0,0 +1,45 @@
+; RUN: opt < %s -S -passes=loop-unroll | FileCheck %s
+
+define i64 @d(i1 %tobool.not, i32 %add, i64 %conv23) !dbg !14{
+entry:
+  br label %for.body
+
+for.body:                                         ; preds = %for.body, %entry
+  ; There should be only one "llvm.dbg.vale" after loop unrolling
+  ; CHECK: call void @llvm.dbg.value
+  ; CHECK-NOT: call void @llvm.dbg.value
+
+  %k.045 = phi i64 [ 0, %entry ], [ %k.046, %for.body ]
+  tail call void @llvm.dbg.value(metadata i32 0, metadata !13, metadata !DIExpression()), !dbg !17
+  %k.046 = add nuw nsw i64 %k.045, 1
+  %exitcond = icmp ne i64 %k.046, 5
+  br i1 %exitcond, label %for.body, label %for.end22
+
+for.end22:                                        ; preds = %for.body
+  ret i64 %k.046
+}
+
+; Function Attrs: nocallback nofree nosync nounwind speculatable willreturn memory(none)
+declare void @llvm.dbg.value(metadata, metadata, metadata)
+
+!llvm.dbg.cu = !{!0}
+!llvm.module.flags = !{!12}
+
+!0 = distinct !DICompileUnit(language: DW_LANG_C89, file: !1, producer: "clang version 19.0.0git (https://github.com/llvm/llvm-project.git ec062f5b33ed22c61742e3c1486f6cba915801e0)", isOptimized: true, runtimeVersion: 0, emissionKind: FullDebug, globals: !2, splitDebugInlining: false, nameTableKind: None)
+!1 = !DIFile(filename: "unroll-remove-redundant-dbg.c", directory: "", checksumkind: CSK_MD5, checksum: "aa30a1d8c04deb9b0f3885c258d2b674")
+!2 = !{!3, !8, !10}
+!3 = !DIGlobalVariableExpression(var: !4, expr: !DIExpression())
+!4 = distinct !DIGlobalVariable(name: "a", scope: !0, file: !1, line: 2, type: !5, isLocal: false, isDefinition: true)
+!5 = !DIDerivedType(tag: DW_TAG_typedef, name: "uint32_t", file: !6, line: 198, baseType: !7)
+!6 = !DIFile(filename: "/usr/include/stdint.h", directory: "", checksumkind: CSK_MD5, checksum: "da031bcff2d0c1d65aa92e7e68a44ef3")
+!7 = !DIBasicType(name: "unsigned int", size: 32, encoding: DW_ATE_unsigned)
+!8 = !DIGlobalVariableExpression(var: !9, expr: !DIExpression())
+!9 = distinct !DIGlobalVariable(name: "c", scope: !0, file: !1, line: 2, type: !5, isLocal: false, isDefinition: true)
+!10 = !DIGlobalVariableExpression(var: !11, expr: !DIExpression())
+!11 = distinct !DIGlobalVariable(name: "b", scope: !0, file: !1, line: 2, type: !5, isLocal: false, isDefinition: true)
+!12 = !{i32 2, !"Debug Info Version", i32 3}
+!13 = !DILocalVariable(name: "f", scope: !14, file: !1, line: 4, type: !5)
+!14 = distinct !DISubprogram(name: "d", scope: !1, file: !1, line: 3, type: !15, scopeLine: 3, flags: DIFlagAllCallsDescribed, spFlags: DISPFlagDefinition | DISPFlagOptimized, unit: !0, retainedNodes: !16)
+!15 = !DISubroutineType(types: !16)
+!16 = !{}
+!17 = !DILocation(line: 0, scope: !14)
diff --git a/llvm/test/Transforms/LoopVectorize/AArch64/conditional-branches-cost.ll b/llvm/test/Transforms/LoopVectorize/AArch64/conditional-branches-cost.ll
new file mode 100644
index 000000000000..a78696cadaaf
--- /dev/null
+++ b/llvm/test/Transforms/LoopVectorize/AArch64/conditional-branches-cost.ll
@@ -0,0 +1,857 @@
+; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --version 4
+; RUN: opt -p loop-vectorize -S %s | FileCheck --check-prefix=DEFAULT %s
+; RUN: opt -p loop-vectorize  -prefer-predicate-over-epilogue=predicate-else-scalar-epilogue -S %s | FileCheck --check-prefix=PRED %s
+
+target datalayout = "e-m:o-i64:64-i128:128-n32:64-S128"
+target triple = "arm64-apple-macosx14.0.0"
+
+define void @invar_cond_gep_store(ptr %dst, i32 %0) {
+; DEFAULT-LABEL: define void @invar_cond_gep_store(
+; DEFAULT-SAME: ptr [[DST:%.*]], i32 [[TMP0:%.*]]) {
+; DEFAULT-NEXT:  entry:
+; DEFAULT-NEXT:    br i1 false, label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
+; DEFAULT:       vector.ph:
+; DEFAULT-NEXT:    [[BROADCAST_SPLATINSERT:%.*]] = insertelement <4 x i32> poison, i32 [[TMP0]], i64 0
+; DEFAULT-NEXT:    [[BROADCAST_SPLAT:%.*]] = shufflevector <4 x i32> [[BROADCAST_SPLATINSERT]], <4 x i32> poison, <4 x i32> zeroinitializer
+; DEFAULT-NEXT:    br label [[VECTOR_BODY:%.*]]
+; DEFAULT:       vector.body:
+; DEFAULT-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[PRED_STORE_CONTINUE14:%.*]] ]
+; DEFAULT-NEXT:    [[OFFSET_IDX:%.*]] = add i64 1, [[INDEX]]
+; DEFAULT-NEXT:    [[TMP1:%.*]] = icmp eq <4 x i32> [[BROADCAST_SPLAT]], zeroinitializer
+; DEFAULT-NEXT:    [[TMP2:%.*]] = icmp eq <4 x i32> [[BROADCAST_SPLAT]], zeroinitializer
+; DEFAULT-NEXT:    [[TMP3:%.*]] = extractelement <4 x i1> [[TMP1]], i32 0
+; DEFAULT-NEXT:    br i1 [[TMP3]], label [[PRED_STORE_IF:%.*]], label [[PRED_STORE_CONTINUE:%.*]]
+; DEFAULT:       pred.store.if:
+; DEFAULT-NEXT:    [[TMP4:%.*]] = add i64 [[OFFSET_IDX]], 0
+; DEFAULT-NEXT:    [[TMP5:%.*]] = add i64 [[TMP4]], 1
+; DEFAULT-NEXT:    [[TMP6:%.*]] = getelementptr i32, ptr [[DST]], i64 [[TMP5]]
+; DEFAULT-NEXT:    store i32 1, ptr [[TMP6]], align 4
+; DEFAULT-NEXT:    br label [[PRED_STORE_CONTINUE]]
+; DEFAULT:       pred.store.continue:
+; DEFAULT-NEXT:    [[TMP7:%.*]] = extractelement <4 x i1> [[TMP1]], i32 1
+; DEFAULT-NEXT:    br i1 [[TMP7]], label [[PRED_STORE_IF1:%.*]], label [[PRED_STORE_CONTINUE2:%.*]]
+; DEFAULT:       pred.store.if1:
+; DEFAULT-NEXT:    [[TMP8:%.*]] = add i64 [[OFFSET_IDX]], 1
+; DEFAULT-NEXT:    [[TMP9:%.*]] = add i64 [[TMP8]], 1
+; DEFAULT-NEXT:    [[TMP10:%.*]] = getelementptr i32, ptr [[DST]], i64 [[TMP9]]
+; DEFAULT-NEXT:    store i32 1, ptr [[TMP10]], align 4
+; DEFAULT-NEXT:    br label [[PRED_STORE_CONTINUE2]]
+; DEFAULT:       pred.store.continue2:
+; DEFAULT-NEXT:    [[TMP11:%.*]] = extractelement <4 x i1> [[TMP1]], i32 2
+; DEFAULT-NEXT:    br i1 [[TMP11]], label [[PRED_STORE_IF3:%.*]], label [[PRED_STORE_CONTINUE4:%.*]]
+; DEFAULT:       pred.store.if3:
+; DEFAULT-NEXT:    [[TMP12:%.*]] = add i64 [[OFFSET_IDX]], 2
+; DEFAULT-NEXT:    [[TMP13:%.*]] = add i64 [[TMP12]], 1
+; DEFAULT-NEXT:    [[TMP14:%.*]] = getelementptr i32, ptr [[DST]], i64 [[TMP13]]
+; DEFAULT-NEXT:    store i32 1, ptr [[TMP14]], align 4
+; DEFAULT-NEXT:    br label [[PRED_STORE_CONTINUE4]]
+; DEFAULT:       pred.store.continue4:
+; DEFAULT-NEXT:    [[TMP15:%.*]] = extractelement <4 x i1> [[TMP1]], i32 3
+; DEFAULT-NEXT:    br i1 [[TMP15]], label [[PRED_STORE_IF5:%.*]], label [[PRED_STORE_CONTINUE6:%.*]]
+; DEFAULT:       pred.store.if5:
+; DEFAULT-NEXT:    [[TMP16:%.*]] = add i64 [[OFFSET_IDX]], 3
+; DEFAULT-NEXT:    [[TMP17:%.*]] = add i64 [[TMP16]], 1
+; DEFAULT-NEXT:    [[TMP18:%.*]] = getelementptr i32, ptr [[DST]], i64 [[TMP17]]
+; DEFAULT-NEXT:    store i32 1, ptr [[TMP18]], align 4
+; DEFAULT-NEXT:    br label [[PRED_STORE_CONTINUE6]]
+; DEFAULT:       pred.store.continue6:
+; DEFAULT-NEXT:    [[TMP19:%.*]] = extractelement <4 x i1> [[TMP2]], i32 0
+; DEFAULT-NEXT:    br i1 [[TMP19]], label [[PRED_STORE_IF7:%.*]], label [[PRED_STORE_CONTINUE8:%.*]]
+; DEFAULT:       pred.store.if7:
+; DEFAULT-NEXT:    [[TMP20:%.*]] = add i64 [[OFFSET_IDX]], 4
+; DEFAULT-NEXT:    [[TMP21:%.*]] = add i64 [[TMP20]], 1
+; DEFAULT-NEXT:    [[TMP22:%.*]] = getelementptr i32, ptr [[DST]], i64 [[TMP21]]
+; DEFAULT-NEXT:    store i32 1, ptr [[TMP22]], align 4
+; DEFAULT-NEXT:    br label [[PRED_STORE_CONTINUE8]]
+; DEFAULT:       pred.store.continue8:
+; DEFAULT-NEXT:    [[TMP23:%.*]] = extractelement <4 x i1> [[TMP2]], i32 1
+; DEFAULT-NEXT:    br i1 [[TMP23]], label [[PRED_STORE_IF9:%.*]], label [[PRED_STORE_CONTINUE10:%.*]]
+; DEFAULT:       pred.store.if9:
+; DEFAULT-NEXT:    [[TMP24:%.*]] = add i64 [[OFFSET_IDX]], 5
+; DEFAULT-NEXT:    [[TMP25:%.*]] = add i64 [[TMP24]], 1
+; DEFAULT-NEXT:    [[TMP26:%.*]] = getelementptr i32, ptr [[DST]], i64 [[TMP25]]
+; DEFAULT-NEXT:    store i32 1, ptr [[TMP26]], align 4
+; DEFAULT-NEXT:    br label [[PRED_STORE_CONTINUE10]]
+; DEFAULT:       pred.store.continue10:
+; DEFAULT-NEXT:    [[TMP27:%.*]] = extractelement <4 x i1> [[TMP2]], i32 2
+; DEFAULT-NEXT:    br i1 [[TMP27]], label [[PRED_STORE_IF11:%.*]], label [[PRED_STORE_CONTINUE12:%.*]]
+; DEFAULT:       pred.store.if11:
+; DEFAULT-NEXT:    [[TMP28:%.*]] = add i64 [[OFFSET_IDX]], 6
+; DEFAULT-NEXT:    [[TMP29:%.*]] = add i64 [[TMP28]], 1
+; DEFAULT-NEXT:    [[TMP30:%.*]] = getelementptr i32, ptr [[DST]], i64 [[TMP29]]
+; DEFAULT-NEXT:    store i32 1, ptr [[TMP30]], align 4
+; DEFAULT-NEXT:    br label [[PRED_STORE_CONTINUE12]]
+; DEFAULT:       pred.store.continue12:
+; DEFAULT-NEXT:    [[TMP31:%.*]] = extractelement <4 x i1> [[TMP2]], i32 3
+; DEFAULT-NEXT:    br i1 [[TMP31]], label [[PRED_STORE_IF13:%.*]], label [[PRED_STORE_CONTINUE14]]
+; DEFAULT:       pred.store.if13:
+; DEFAULT-NEXT:    [[TMP32:%.*]] = add i64 [[OFFSET_IDX]], 7
+; DEFAULT-NEXT:    [[TMP33:%.*]] = add i64 [[TMP32]], 1
+; DEFAULT-NEXT:    [[TMP34:%.*]] = getelementptr i32, ptr [[DST]], i64 [[TMP33]]
+; DEFAULT-NEXT:    store i32 1, ptr [[TMP34]], align 4
+; DEFAULT-NEXT:    br label [[PRED_STORE_CONTINUE14]]
+; DEFAULT:       pred.store.continue14:
+; DEFAULT-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 8
+; DEFAULT-NEXT:    [[TMP35:%.*]] = icmp eq i64 [[INDEX_NEXT]], 96
+; DEFAULT-NEXT:    br i1 [[TMP35]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP0:![0-9]+]]
+; DEFAULT:       middle.block:
+; DEFAULT-NEXT:    br i1 false, label [[EXIT:%.*]], label [[SCALAR_PH]]
+; DEFAULT:       scalar.ph:
+; DEFAULT-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ 97, [[MIDDLE_BLOCK]] ], [ 1, [[ENTRY:%.*]] ]
+; DEFAULT-NEXT:    br label [[LOOP_HEADER:%.*]]
+; DEFAULT:       loop.header:
+; DEFAULT-NEXT:    [[IV:%.*]] = phi i64 [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ], [ [[IV_NEXT:%.*]], [[LOOP_LATCH:%.*]] ]
+; DEFAULT-NEXT:    [[IV_NEXT]] = add i64 [[IV]], 1
+; DEFAULT-NEXT:    [[CMP9:%.*]] = icmp eq i32 [[TMP0]], 0
+; DEFAULT-NEXT:    br i1 [[CMP9]], label [[THEN:%.*]], label [[LOOP_LATCH]]
+; DEFAULT:       then:
+; DEFAULT-NEXT:    [[GEP:%.*]] = getelementptr i32, ptr [[DST]], i64 [[IV_NEXT]]
+; DEFAULT-NEXT:    store i32 1, ptr [[GEP]], align 4
+; DEFAULT-NEXT:    br label [[LOOP_LATCH]]
+; DEFAULT:       loop.latch:
+; DEFAULT-NEXT:    [[EC:%.*]] = icmp eq i64 [[IV]], 100
+; DEFAULT-NEXT:    br i1 [[EC]], label [[EXIT]], label [[LOOP_HEADER]], !llvm.loop [[LOOP3:![0-9]+]]
+; DEFAULT:       exit:
+; DEFAULT-NEXT:    ret void
+;
+; PRED-LABEL: define void @invar_cond_gep_store(
+; PRED-SAME: ptr [[DST:%.*]], i32 [[TMP0:%.*]]) {
+; PRED-NEXT:  entry:
+; PRED-NEXT:    br i1 false, label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
+; PRED:       vector.ph:
+; PRED-NEXT:    [[BROADCAST_SPLATINSERT:%.*]] = insertelement <4 x i32> poison, i32 [[TMP0]], i64 0
+; PRED-NEXT:    [[BROADCAST_SPLAT:%.*]] = shufflevector <4 x i32> [[BROADCAST_SPLATINSERT]], <4 x i32> poison, <4 x i32> zeroinitializer
+; PRED-NEXT:    br label [[VECTOR_BODY:%.*]]
+; PRED:       vector.body:
+; PRED-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[PRED_STORE_CONTINUE6:%.*]] ]
+; PRED-NEXT:    [[OFFSET_IDX:%.*]] = add i64 1, [[INDEX]]
+; PRED-NEXT:    [[TMP1:%.*]] = icmp eq <4 x i32> [[BROADCAST_SPLAT]], zeroinitializer
+; PRED-NEXT:    [[TMP2:%.*]] = extractelement <4 x i1> [[TMP1]], i32 0
+; PRED-NEXT:    br i1 [[TMP2]], label [[PRED_STORE_IF:%.*]], label [[PRED_STORE_CONTINUE:%.*]]
+; PRED:       pred.store.if:
+; PRED-NEXT:    [[TMP3:%.*]] = add i64 [[OFFSET_IDX]], 0
+; PRED-NEXT:    [[TMP4:%.*]] = add i64 [[TMP3]], 1
+; PRED-NEXT:    [[TMP5:%.*]] = getelementptr i32, ptr [[DST]], i64 [[TMP4]]
+; PRED-NEXT:    store i32 1, ptr [[TMP5]], align 4
+; PRED-NEXT:    br label [[PRED_STORE_CONTINUE]]
+; PRED:       pred.store.continue:
+; PRED-NEXT:    [[TMP6:%.*]] = extractelement <4 x i1> [[TMP1]], i32 1
+; PRED-NEXT:    br i1 [[TMP6]], label [[PRED_STORE_IF1:%.*]], label [[PRED_STORE_CONTINUE2:%.*]]
+; PRED:       pred.store.if1:
+; PRED-NEXT:    [[TMP7:%.*]] = add i64 [[OFFSET_IDX]], 1
+; PRED-NEXT:    [[TMP8:%.*]] = add i64 [[TMP7]], 1
+; PRED-NEXT:    [[TMP9:%.*]] = getelementptr i32, ptr [[DST]], i64 [[TMP8]]
+; PRED-NEXT:    store i32 1, ptr [[TMP9]], align 4
+; PRED-NEXT:    br label [[PRED_STORE_CONTINUE2]]
+; PRED:       pred.store.continue2:
+; PRED-NEXT:    [[TMP10:%.*]] = extractelement <4 x i1> [[TMP1]], i32 2
+; PRED-NEXT:    br i1 [[TMP10]], label [[PRED_STORE_IF3:%.*]], label [[PRED_STORE_CONTINUE4:%.*]]
+; PRED:       pred.store.if3:
+; PRED-NEXT:    [[TMP11:%.*]] = add i64 [[OFFSET_IDX]], 2
+; PRED-NEXT:    [[TMP12:%.*]] = add i64 [[TMP11]], 1
+; PRED-NEXT:    [[TMP13:%.*]] = getelementptr i32, ptr [[DST]], i64 [[TMP12]]
+; PRED-NEXT:    store i32 1, ptr [[TMP13]], align 4
+; PRED-NEXT:    br label [[PRED_STORE_CONTINUE4]]
+; PRED:       pred.store.continue4:
+; PRED-NEXT:    [[TMP14:%.*]] = extractelement <4 x i1> [[TMP1]], i32 3
+; PRED-NEXT:    br i1 [[TMP14]], label [[PRED_STORE_IF5:%.*]], label [[PRED_STORE_CONTINUE6]]
+; PRED:       pred.store.if5:
+; PRED-NEXT:    [[TMP15:%.*]] = add i64 [[OFFSET_IDX]], 3
+; PRED-NEXT:    [[TMP16:%.*]] = add i64 [[TMP15]], 1
+; PRED-NEXT:    [[TMP17:%.*]] = getelementptr i32, ptr [[DST]], i64 [[TMP16]]
+; PRED-NEXT:    store i32 1, ptr [[TMP17]], align 4
+; PRED-NEXT:    br label [[PRED_STORE_CONTINUE6]]
+; PRED:       pred.store.continue6:
+; PRED-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 4
+; PRED-NEXT:    [[TMP18:%.*]] = icmp eq i64 [[INDEX_NEXT]], 100
+; PRED-NEXT:    br i1 [[TMP18]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP0:![0-9]+]]
+; PRED:       middle.block:
+; PRED-NEXT:    br i1 true, label [[EXIT:%.*]], label [[SCALAR_PH]]
+; PRED:       scalar.ph:
+; PRED-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ 101, [[MIDDLE_BLOCK]] ], [ 1, [[ENTRY:%.*]] ]
+; PRED-NEXT:    br label [[LOOP_HEADER:%.*]]
+; PRED:       loop.header:
+; PRED-NEXT:    [[IV:%.*]] = phi i64 [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ], [ [[IV_NEXT:%.*]], [[LOOP_LATCH:%.*]] ]
+; PRED-NEXT:    [[IV_NEXT]] = add i64 [[IV]], 1
+; PRED-NEXT:    [[CMP9:%.*]] = icmp eq i32 [[TMP0]], 0
+; PRED-NEXT:    br i1 [[CMP9]], label [[THEN:%.*]], label [[LOOP_LATCH]]
+; PRED:       then:
+; PRED-NEXT:    [[GEP:%.*]] = getelementptr i32, ptr [[DST]], i64 [[IV_NEXT]]
+; PRED-NEXT:    store i32 1, ptr [[GEP]], align 4
+; PRED-NEXT:    br label [[LOOP_LATCH]]
+; PRED:       loop.latch:
+; PRED-NEXT:    [[EC:%.*]] = icmp eq i64 [[IV]], 100
+; PRED-NEXT:    br i1 [[EC]], label [[EXIT]], label [[LOOP_HEADER]], !llvm.loop [[LOOP3:![0-9]+]]
+; PRED:       exit:
+; PRED-NEXT:    ret void
+;
+entry:
+  br label %loop.header
+
+loop.header:
+  %iv = phi i64 [ 1, %entry ], [ %iv.next, %loop.latch ]
+  %iv.next = add i64 %iv, 1
+  %cmp9 = icmp eq i32 %0, 0
+  br i1 %cmp9, label %then, label %loop.latch
+
+then:
+  %gep = getelementptr i32, ptr %dst, i64 %iv.next
+  store i32 1, ptr %gep, align 4
+  br label %loop.latch
+
+loop.latch:
+  %ec = icmp eq i64 %iv, 100
+  br i1 %ec, label %exit, label %loop.header
+
+exit:
+  ret void
+}
+
+declare double @llvm.fabs.f64(double) #0
+
+define void @loop_dependent_cond(ptr %src, ptr noalias %dst, i64 %N) {
+; DEFAULT-LABEL: define void @loop_dependent_cond(
+; DEFAULT-SAME: ptr [[SRC:%.*]], ptr noalias [[DST:%.*]], i64 [[N:%.*]]) {
+; DEFAULT-NEXT:  entry:
+; DEFAULT-NEXT:    [[TMP0:%.*]] = add i64 [[N]], 1
+; DEFAULT-NEXT:    [[MIN_ITERS_CHECK:%.*]] = icmp ult i64 [[TMP0]], 4
+; DEFAULT-NEXT:    br i1 [[MIN_ITERS_CHECK]], label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
+; DEFAULT:       vector.ph:
+; DEFAULT-NEXT:    [[N_MOD_VF:%.*]] = urem i64 [[TMP0]], 4
+; DEFAULT-NEXT:    [[N_VEC:%.*]] = sub i64 [[TMP0]], [[N_MOD_VF]]
+; DEFAULT-NEXT:    br label [[VECTOR_BODY:%.*]]
+; DEFAULT:       vector.body:
+; DEFAULT-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[PRED_STORE_CONTINUE7:%.*]] ]
+; DEFAULT-NEXT:    [[TMP1:%.*]] = add i64 [[INDEX]], 0
+; DEFAULT-NEXT:    [[TMP2:%.*]] = add i64 [[INDEX]], 2
+; DEFAULT-NEXT:    [[TMP3:%.*]] = getelementptr double, ptr [[SRC]], i64 [[TMP1]]
+; DEFAULT-NEXT:    [[TMP4:%.*]] = getelementptr double, ptr [[SRC]], i64 [[TMP2]]
+; DEFAULT-NEXT:    [[TMP5:%.*]] = getelementptr double, ptr [[TMP3]], i32 0
+; DEFAULT-NEXT:    [[TMP6:%.*]] = getelementptr double, ptr [[TMP3]], i32 2
+; DEFAULT-NEXT:    [[WIDE_LOAD:%.*]] = load <2 x double>, ptr [[TMP5]], align 8
+; DEFAULT-NEXT:    [[WIDE_LOAD1:%.*]] = load <2 x double>, ptr [[TMP6]], align 8
+; DEFAULT-NEXT:    [[TMP7:%.*]] = call <2 x double> @llvm.fabs.v2f64(<2 x double> [[WIDE_LOAD]])
+; DEFAULT-NEXT:    [[TMP8:%.*]] = call <2 x double> @llvm.fabs.v2f64(<2 x double> [[WIDE_LOAD1]])
+; DEFAULT-NEXT:    [[TMP9:%.*]] = fcmp ogt <2 x double> [[TMP7]], <double 1.000000e+00, double 1.000000e+00>
+; DEFAULT-NEXT:    [[TMP10:%.*]] = fcmp ogt <2 x double> [[TMP8]], <double 1.000000e+00, double 1.000000e+00>
+; DEFAULT-NEXT:    [[TMP11:%.*]] = extractelement <2 x i1> [[TMP9]], i32 0
+; DEFAULT-NEXT:    br i1 [[TMP11]], label [[PRED_STORE_IF:%.*]], label [[PRED_STORE_CONTINUE:%.*]]
+; DEFAULT:       pred.store.if:
+; DEFAULT-NEXT:    store i32 0, ptr [[DST]], align 4
+; DEFAULT-NEXT:    br label [[PRED_STORE_CONTINUE]]
+; DEFAULT:       pred.store.continue:
+; DEFAULT-NEXT:    [[TMP12:%.*]] = extractelement <2 x i1> [[TMP9]], i32 1
+; DEFAULT-NEXT:    br i1 [[TMP12]], label [[PRED_STORE_IF2:%.*]], label [[PRED_STORE_CONTINUE3:%.*]]
+; DEFAULT:       pred.store.if2:
+; DEFAULT-NEXT:    store i32 0, ptr [[DST]], align 4
+; DEFAULT-NEXT:    br label [[PRED_STORE_CONTINUE3]]
+; DEFAULT:       pred.store.continue3:
+; DEFAULT-NEXT:    [[TMP13:%.*]] = extractelement <2 x i1> [[TMP10]], i32 0
+; DEFAULT-NEXT:    br i1 [[TMP13]], label [[PRED_STORE_IF4:%.*]], label [[PRED_STORE_CONTINUE5:%.*]]
+; DEFAULT:       pred.store.if4:
+; DEFAULT-NEXT:    store i32 0, ptr [[DST]], align 4
+; DEFAULT-NEXT:    br label [[PRED_STORE_CONTINUE5]]
+; DEFAULT:       pred.store.continue5:
+; DEFAULT-NEXT:    [[TMP14:%.*]] = extractelement <2 x i1> [[TMP10]], i32 1
+; DEFAULT-NEXT:    br i1 [[TMP14]], label [[PRED_STORE_IF6:%.*]], label [[PRED_STORE_CONTINUE7]]
+; DEFAULT:       pred.store.if6:
+; DEFAULT-NEXT:    store i32 0, ptr [[DST]], align 4
+; DEFAULT-NEXT:    br label [[PRED_STORE_CONTINUE7]]
+; DEFAULT:       pred.store.continue7:
+; DEFAULT-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 4
+; DEFAULT-NEXT:    [[TMP15:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
+; DEFAULT-NEXT:    br i1 [[TMP15]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP4:![0-9]+]]
+; DEFAULT:       middle.block:
+; DEFAULT-NEXT:    [[CMP_N:%.*]] = icmp eq i64 [[TMP0]], [[N_VEC]]
+; DEFAULT-NEXT:    br i1 [[CMP_N]], label [[FOR_END123:%.*]], label [[SCALAR_PH]]
+; DEFAULT:       scalar.ph:
+; DEFAULT-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ [[N_VEC]], [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ]
+; DEFAULT-NEXT:    br label [[FOR_BODY112:%.*]]
+; DEFAULT:       loop.header:
+; DEFAULT-NEXT:    [[IV175:%.*]] = phi i64 [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ], [ [[IV_NEXT176:%.*]], [[FOR_INC121:%.*]] ]
+; DEFAULT-NEXT:    [[ARRAYIDX114:%.*]] = getelementptr double, ptr [[SRC]], i64 [[IV175]]
+; DEFAULT-NEXT:    [[TMP16:%.*]] = load double, ptr [[ARRAYIDX114]], align 8
+; DEFAULT-NEXT:    [[TMP17:%.*]] = tail call double @llvm.fabs.f64(double [[TMP16]])
+; DEFAULT-NEXT:    [[CMP115:%.*]] = fcmp ogt double [[TMP17]], 1.000000e+00
+; DEFAULT-NEXT:    br i1 [[CMP115]], label [[IF_THEN117:%.*]], label [[FOR_INC121]]
+; DEFAULT:       then:
+; DEFAULT-NEXT:    store i32 0, ptr [[DST]], align 4
+; DEFAULT-NEXT:    br label [[FOR_INC121]]
+; DEFAULT:       loop.latch:
+; DEFAULT-NEXT:    [[IV_NEXT176]] = add i64 [[IV175]], 1
+; DEFAULT-NEXT:    [[EXITCOND180_NOT:%.*]] = icmp eq i64 [[IV175]], [[N]]
+; DEFAULT-NEXT:    br i1 [[EXITCOND180_NOT]], label [[FOR_END123]], label [[FOR_BODY112]], !llvm.loop [[LOOP5:![0-9]+]]
+; DEFAULT:       exit:
+; DEFAULT-NEXT:    ret void
+;
+; PRED-LABEL: define void @loop_dependent_cond(
+; PRED-SAME: ptr [[SRC:%.*]], ptr noalias [[DST:%.*]], i64 [[N:%.*]]) {
+; PRED-NEXT:  entry:
+; PRED-NEXT:    br label [[FOR_BODY112:%.*]]
+; PRED:       loop.header:
+; PRED-NEXT:    [[IV175:%.*]] = phi i64 [ 0, [[ENTRY:%.*]] ], [ [[IV_NEXT176:%.*]], [[FOR_INC121:%.*]] ]
+; PRED-NEXT:    [[ARRAYIDX114:%.*]] = getelementptr double, ptr [[SRC]], i64 [[IV175]]
+; PRED-NEXT:    [[TMP0:%.*]] = load double, ptr [[ARRAYIDX114]], align 8
+; PRED-NEXT:    [[TMP1:%.*]] = tail call double @llvm.fabs.f64(double [[TMP0]])
+; PRED-NEXT:    [[CMP115:%.*]] = fcmp ogt double [[TMP1]], 1.000000e+00
+; PRED-NEXT:    br i1 [[CMP115]], label [[IF_THEN117:%.*]], label [[FOR_INC121]]
+; PRED:       then:
+; PRED-NEXT:    store i32 0, ptr [[DST]], align 4
+; PRED-NEXT:    br label [[FOR_INC121]]
+; PRED:       loop.latch:
+; PRED-NEXT:    [[IV_NEXT176]] = add i64 [[IV175]], 1
+; PRED-NEXT:    [[EXITCOND180_NOT:%.*]] = icmp eq i64 [[IV175]], [[N]]
+; PRED-NEXT:    br i1 [[EXITCOND180_NOT]], label [[FOR_END123:%.*]], label [[FOR_BODY112]]
+; PRED:       exit:
+; PRED-NEXT:    ret void
+;
+entry:
+  br label %loop.header
+
+loop.header:
+  %iv = phi i64 [ 0, %entry ], [ %iv.next, %loop.latch ]
+  %gep = getelementptr double, ptr %src, i64 %iv
+  %l = load double, ptr %gep, align 8
+  %abs = tail call double @llvm.fabs.f64(double %l)
+  %cmp = fcmp ogt double %abs, 1.000000e+00
+  br i1 %cmp, label %then, label %loop.latch
+
+then:
+  store i32 0, ptr %dst, align 4
+  br label %loop.latch
+
+loop.latch:
+  %iv.next = add i64 %iv, 1
+  %ec = icmp eq i64 %iv, %N
+  br i1 %ec, label %exit, label %loop.header
+
+exit:
+  ret void
+}
+
+define void @invar_cond_chain_1(ptr %I, ptr noalias %src, i1 %c) {
+; DEFAULT-LABEL: define void @invar_cond_chain_1(
+; DEFAULT-SAME: ptr [[I:%.*]], ptr noalias [[SRC:%.*]], i1 [[C:%.*]]) {
+; DEFAULT-NEXT:  entry:
+; DEFAULT-NEXT:    [[SRC2:%.*]] = ptrtoint ptr [[SRC]] to i64
+; DEFAULT-NEXT:    [[I1:%.*]] = ptrtoint ptr [[I]] to i64
+; DEFAULT-NEXT:    [[TMP29:%.*]] = sub i64 [[I1]], [[SRC2]]
+; DEFAULT-NEXT:    [[TMP0:%.*]] = lshr i64 [[TMP29]], 2
+; DEFAULT-NEXT:    [[TMP1:%.*]] = add nuw nsw i64 [[TMP0]], 1
+; DEFAULT-NEXT:    [[MIN_ITERS_CHECK:%.*]] = icmp ult i64 [[TMP1]], 8
+; DEFAULT-NEXT:    br i1 [[MIN_ITERS_CHECK]], label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
+; DEFAULT:       vector.ph:
+; DEFAULT-NEXT:    [[N_MOD_VF:%.*]] = urem i64 [[TMP1]], 8
+; DEFAULT-NEXT:    [[N_VEC:%.*]] = sub i64 [[TMP1]], [[N_MOD_VF]]
+; DEFAULT-NEXT:    [[TMP2:%.*]] = mul i64 [[N_VEC]], 4
+; DEFAULT-NEXT:    [[IND_END:%.*]] = getelementptr i8, ptr [[SRC]], i64 [[TMP2]]
+; DEFAULT-NEXT:    [[BROADCAST_SPLATINSERT:%.*]] = insertelement <4 x i1> poison, i1 [[C]], i64 0
+; DEFAULT-NEXT:    [[BROADCAST_SPLAT:%.*]] = shufflevector <4 x i1> [[BROADCAST_SPLATINSERT]], <4 x i1> poison, <4 x i32> zeroinitializer
+; DEFAULT-NEXT:    br label [[VECTOR_BODY:%.*]]
+; DEFAULT:       vector.body:
+; DEFAULT-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[PRED_STORE_CONTINUE17:%.*]] ]
+; DEFAULT-NEXT:    [[OFFSET_IDX:%.*]] = mul i64 [[INDEX]], 4
+; DEFAULT-NEXT:    [[TMP3:%.*]] = add i64 [[OFFSET_IDX]], 0
+; DEFAULT-NEXT:    [[TMP4:%.*]] = add i64 [[OFFSET_IDX]], 16
+; DEFAULT-NEXT:    [[NEXT_GEP:%.*]] = getelementptr i8, ptr [[SRC]], i64 [[TMP3]]
+; DEFAULT-NEXT:    [[NEXT_GEP3:%.*]] = getelementptr i8, ptr [[SRC]], i64 [[TMP4]]
+; DEFAULT-NEXT:    [[TMP5:%.*]] = getelementptr i32, ptr [[NEXT_GEP]], i32 0
+; DEFAULT-NEXT:    [[TMP6:%.*]] = getelementptr i32, ptr [[NEXT_GEP]], i32 4
+; DEFAULT-NEXT:    [[WIDE_LOAD:%.*]] = load <4 x i32>, ptr [[TMP5]], align 4
+; DEFAULT-NEXT:    [[WIDE_LOAD3:%.*]] = load <4 x i32>, ptr [[TMP6]], align 4
+; DEFAULT-NEXT:    [[TMP7:%.*]] = select <4 x i1> <i1 true, i1 true, i1 true, i1 true>, <4 x i1> [[BROADCAST_SPLAT]], <4 x i1> zeroinitializer
+; DEFAULT-NEXT:    [[TMP8:%.*]] = select <4 x i1> <i1 true, i1 true, i1 true, i1 true>, <4 x i1> [[BROADCAST_SPLAT]], <4 x i1> zeroinitializer
+; DEFAULT-NEXT:    [[TMP9:%.*]] = or <4 x i1> [[TMP7]], zeroinitializer
+; DEFAULT-NEXT:    [[TMP10:%.*]] = or <4 x i1> [[TMP8]], zeroinitializer
+; DEFAULT-NEXT:    [[TMP11:%.*]] = extractelement <4 x i1> [[TMP9]], i32 0
+; DEFAULT-NEXT:    br i1 [[TMP11]], label [[PRED_STORE_IF:%.*]], label [[PRED_STORE_CONTINUE:%.*]]
+; DEFAULT:       pred.store.if:
+; DEFAULT-NEXT:    [[TMP12:%.*]] = extractelement <4 x i32> [[WIDE_LOAD]], i32 0
+; DEFAULT-NEXT:    store i32 [[TMP12]], ptr [[I]], align 4
+; DEFAULT-NEXT:    br label [[PRED_STORE_CONTINUE]]
+; DEFAULT:       pred.store.continue:
+; DEFAULT-NEXT:    [[TMP13:%.*]] = extractelement <4 x i1> [[TMP9]], i32 1
+; DEFAULT-NEXT:    br i1 [[TMP13]], label [[PRED_STORE_IF4:%.*]], label [[PRED_STORE_CONTINUE5:%.*]]
+; DEFAULT:       pred.store.if5:
+; DEFAULT-NEXT:    [[TMP14:%.*]] = extractelement <4 x i32> [[WIDE_LOAD]], i32 1
+; DEFAULT-NEXT:    store i32 [[TMP14]], ptr [[I]], align 4
+; DEFAULT-NEXT:    br label [[PRED_STORE_CONTINUE5]]
+; DEFAULT:       pred.store.continue6:
+; DEFAULT-NEXT:    [[TMP15:%.*]] = extractelement <4 x i1> [[TMP9]], i32 2
+; DEFAULT-NEXT:    br i1 [[TMP15]], label [[PRED_STORE_IF6:%.*]], label [[PRED_STORE_CONTINUE7:%.*]]
+; DEFAULT:       pred.store.if7:
+; DEFAULT-NEXT:    [[TMP16:%.*]] = extractelement <4 x i32> [[WIDE_LOAD]], i32 2
+; DEFAULT-NEXT:    store i32 [[TMP16]], ptr [[I]], align 4
+; DEFAULT-NEXT:    br label [[PRED_STORE_CONTINUE7]]
+; DEFAULT:       pred.store.continue8:
+; DEFAULT-NEXT:    [[TMP17:%.*]] = extractelement <4 x i1> [[TMP9]], i32 3
+; DEFAULT-NEXT:    br i1 [[TMP17]], label [[PRED_STORE_IF8:%.*]], label [[PRED_STORE_CONTINUE9:%.*]]
+; DEFAULT:       pred.store.if9:
+; DEFAULT-NEXT:    [[TMP18:%.*]] = extractelement <4 x i32> [[WIDE_LOAD]], i32 3
+; DEFAULT-NEXT:    store i32 [[TMP18]], ptr [[I]], align 4
+; DEFAULT-NEXT:    br label [[PRED_STORE_CONTINUE9]]
+; DEFAULT:       pred.store.continue10:
+; DEFAULT-NEXT:    [[TMP19:%.*]] = extractelement <4 x i1> [[TMP10]], i32 0
+; DEFAULT-NEXT:    br i1 [[TMP19]], label [[PRED_STORE_IF10:%.*]], label [[PRED_STORE_CONTINUE11:%.*]]
+; DEFAULT:       pred.store.if11:
+; DEFAULT-NEXT:    [[TMP20:%.*]] = extractelement <4 x i32> [[WIDE_LOAD3]], i32 0
+; DEFAULT-NEXT:    store i32 [[TMP20]], ptr [[I]], align 4
+; DEFAULT-NEXT:    br label [[PRED_STORE_CONTINUE11]]
+; DEFAULT:       pred.store.continue12:
+; DEFAULT-NEXT:    [[TMP21:%.*]] = extractelement <4 x i1> [[TMP10]], i32 1
+; DEFAULT-NEXT:    br i1 [[TMP21]], label [[PRED_STORE_IF12:%.*]], label [[PRED_STORE_CONTINUE13:%.*]]
+; DEFAULT:       pred.store.if13:
+; DEFAULT-NEXT:    [[TMP22:%.*]] = extractelement <4 x i32> [[WIDE_LOAD3]], i32 1
+; DEFAULT-NEXT:    store i32 [[TMP22]], ptr [[I]], align 4
+; DEFAULT-NEXT:    br label [[PRED_STORE_CONTINUE13]]
+; DEFAULT:       pred.store.continue14:
+; DEFAULT-NEXT:    [[TMP23:%.*]] = extractelement <4 x i1> [[TMP10]], i32 2
+; DEFAULT-NEXT:    br i1 [[TMP23]], label [[PRED_STORE_IF14:%.*]], label [[PRED_STORE_CONTINUE15:%.*]]
+; DEFAULT:       pred.store.if15:
+; DEFAULT-NEXT:    [[TMP24:%.*]] = extractelement <4 x i32> [[WIDE_LOAD3]], i32 2
+; DEFAULT-NEXT:    store i32 [[TMP24]], ptr [[I]], align 4
+; DEFAULT-NEXT:    br label [[PRED_STORE_CONTINUE15]]
+; DEFAULT:       pred.store.continue16:
+; DEFAULT-NEXT:    [[TMP25:%.*]] = extractelement <4 x i1> [[TMP10]], i32 3
+; DEFAULT-NEXT:    br i1 [[TMP25]], label [[PRED_STORE_IF16:%.*]], label [[PRED_STORE_CONTINUE17]]
+; DEFAULT:       pred.store.if17:
+; DEFAULT-NEXT:    [[TMP26:%.*]] = extractelement <4 x i32> [[WIDE_LOAD3]], i32 3
+; DEFAULT-NEXT:    store i32 [[TMP26]], ptr [[I]], align 4
+; DEFAULT-NEXT:    br label [[PRED_STORE_CONTINUE17]]
+; DEFAULT:       pred.store.continue18:
+; DEFAULT-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 8
+; DEFAULT-NEXT:    [[TMP27:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
+; DEFAULT-NEXT:    br i1 [[TMP27]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP6:![0-9]+]]
+; DEFAULT:       middle.block:
+; DEFAULT-NEXT:    [[CMP_N:%.*]] = icmp eq i64 [[TMP1]], [[N_VEC]]
+; DEFAULT-NEXT:    br i1 [[CMP_N]], label [[FOR_COND_CLEANUP312_LOOPEXIT:%.*]], label [[SCALAR_PH]]
+; DEFAULT:       scalar.ph:
+; DEFAULT-NEXT:    [[BC_RESUME_VAL:%.*]] = phi ptr [ [[IND_END]], [[MIDDLE_BLOCK]] ], [ [[SRC]], [[ENTRY:%.*]] ]
+; DEFAULT-NEXT:    br label [[FOR_BODY313:%.*]]
+; DEFAULT:       loop.header:
+; DEFAULT-NEXT:    [[__BEGIN3_011973:%.*]] = phi ptr [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ], [ [[INCDEC_PTR329:%.*]], [[IF_END327:%.*]] ]
+; DEFAULT-NEXT:    [[TMP28:%.*]] = load i32, ptr [[__BEGIN3_011973]], align 4
+; DEFAULT-NEXT:    br i1 true, label [[IF_ELSE321:%.*]], label [[IF_THEN316:%.*]]
+; DEFAULT:       if:
+; DEFAULT-NEXT:    br label [[IF_END327_SINK_SPLIT:%.*]]
+; DEFAULT:       else.1:
+; DEFAULT-NEXT:    br i1 [[C]], label [[IF_THEN323:%.*]], label [[IF_END327]]
+; DEFAULT:       else.2:
+; DEFAULT-NEXT:    br label [[IF_END327_SINK_SPLIT]]
+; DEFAULT:       split:
+; DEFAULT-NEXT:    store i32 [[TMP28]], ptr [[I]], align 4
+; DEFAULT-NEXT:    br label [[IF_END327]]
+; DEFAULT:       loop.latch:
+; DEFAULT-NEXT:    [[INCDEC_PTR329]] = getelementptr inbounds i8, ptr [[__BEGIN3_011973]], i64 4
+; DEFAULT-NEXT:    [[CMP311_NOT:%.*]] = icmp eq ptr [[__BEGIN3_011973]], [[I]]
+; DEFAULT-NEXT:    br i1 [[CMP311_NOT]], label [[FOR_COND_CLEANUP312_LOOPEXIT]], label [[FOR_BODY313]], !llvm.loop [[LOOP7:![0-9]+]]
+; DEFAULT:       exit:
+; DEFAULT-NEXT:    ret void
+;
+; PRED-LABEL: define void @invar_cond_chain_1(
+; PRED-SAME: ptr [[I:%.*]], ptr noalias [[SRC:%.*]], i1 [[C:%.*]]) {
+; PRED-NEXT:  entry:
+; PRED-NEXT:    br label [[FOR_BODY313:%.*]]
+; PRED:       loop.header:
+; PRED-NEXT:    [[__BEGIN3_011973:%.*]] = phi ptr [ [[SRC]], [[ENTRY:%.*]] ], [ [[INCDEC_PTR329:%.*]], [[IF_END327:%.*]] ]
+; PRED-NEXT:    [[TMP0:%.*]] = load i32, ptr [[__BEGIN3_011973]], align 4
+; PRED-NEXT:    br i1 true, label [[IF_ELSE321:%.*]], label [[IF_THEN316:%.*]]
+; PRED:       if:
+; PRED-NEXT:    br label [[IF_END327_SINK_SPLIT:%.*]]
+; PRED:       else.1:
+; PRED-NEXT:    br i1 [[C]], label [[IF_THEN323:%.*]], label [[IF_END327]]
+; PRED:       else.2:
+; PRED-NEXT:    br label [[IF_END327_SINK_SPLIT]]
+; PRED:       split:
+; PRED-NEXT:    store i32 [[TMP0]], ptr [[I]], align 4
+; PRED-NEXT:    br label [[IF_END327]]
+; PRED:       loop.latch:
+; PRED-NEXT:    [[INCDEC_PTR329]] = getelementptr inbounds i8, ptr [[__BEGIN3_011973]], i64 4
+; PRED-NEXT:    [[CMP311_NOT:%.*]] = icmp eq ptr [[__BEGIN3_011973]], [[I]]
+; PRED-NEXT:    br i1 [[CMP311_NOT]], label [[FOR_COND_CLEANUP312_LOOPEXIT:%.*]], label [[FOR_BODY313]]
+; PRED:       exit:
+; PRED-NEXT:    ret void
+;
+entry:
+  br label %loop.header
+
+loop.header:
+  %ptr.iv = phi ptr [ %src, %entry ], [ %ptr.iv.next, %loop.latch ]
+  %l = load i32, ptr %ptr.iv, align 4
+  br i1 true, label %else.1, label %if
+
+if:
+  br label %split
+
+else.1:
+  br i1 %c, label %else.2, label %loop.latch
+
+else.2:
+  br label %split
+
+split:
+  store i32 %l, ptr %I, align 4
+  br label %loop.latch
+
+loop.latch:
+  %ptr.iv.next = getelementptr inbounds i8, ptr %ptr.iv, i64 4
+  %ec = icmp eq ptr %ptr.iv, %I
+  br i1 %ec, label %exit, label %loop.header
+
+exit:
+  ret void
+}
+
+define void @invar_cond_chain_2(ptr %I, ptr noalias %src, ptr noalias %dst, i32 %a) {
+; DEFAULT-LABEL: define void @invar_cond_chain_2(
+; DEFAULT-SAME: ptr [[I:%.*]], ptr noalias [[SRC:%.*]], ptr noalias [[DST:%.*]], i32 [[A:%.*]]) {
+; DEFAULT-NEXT:  entry:
+; DEFAULT-NEXT:    [[SRC2:%.*]] = ptrtoint ptr [[SRC]] to i64
+; DEFAULT-NEXT:    [[I1:%.*]] = ptrtoint ptr [[I]] to i64
+; DEFAULT-NEXT:    [[TMP0:%.*]] = sub i64 [[I1]], [[SRC2]]
+; DEFAULT-NEXT:    [[TMP1:%.*]] = lshr i64 [[TMP0]], 2
+; DEFAULT-NEXT:    [[TMP2:%.*]] = add nuw nsw i64 [[TMP1]], 1
+; DEFAULT-NEXT:    [[MIN_ITERS_CHECK:%.*]] = icmp ult i64 [[TMP2]], 8
+; DEFAULT-NEXT:    br i1 [[MIN_ITERS_CHECK]], label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
+; DEFAULT:       vector.ph:
+; DEFAULT-NEXT:    [[N_MOD_VF:%.*]] = urem i64 [[TMP2]], 8
+; DEFAULT-NEXT:    [[N_VEC:%.*]] = sub i64 [[TMP2]], [[N_MOD_VF]]
+; DEFAULT-NEXT:    [[TMP3:%.*]] = mul i64 [[N_VEC]], 4
+; DEFAULT-NEXT:    [[IND_END:%.*]] = getelementptr i8, ptr [[SRC]], i64 [[TMP3]]
+; DEFAULT-NEXT:    [[BROADCAST_SPLATINSERT:%.*]] = insertelement <4 x i32> poison, i32 [[A]], i64 0
+; DEFAULT-NEXT:    [[BROADCAST_SPLAT:%.*]] = shufflevector <4 x i32> [[BROADCAST_SPLATINSERT]], <4 x i32> poison, <4 x i32> zeroinitializer
+; DEFAULT-NEXT:    br label [[VECTOR_BODY:%.*]]
+; DEFAULT:       vector.body:
+; DEFAULT-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[PRED_STORE_CONTINUE15:%.*]] ]
+; DEFAULT-NEXT:    [[TMP4:%.*]] = icmp sgt <4 x i32> [[BROADCAST_SPLAT]], zeroinitializer
+; DEFAULT-NEXT:    [[TMP5:%.*]] = icmp sgt <4 x i32> [[BROADCAST_SPLAT]], zeroinitializer
+; DEFAULT-NEXT:    [[TMP6:%.*]] = xor <4 x i1> [[TMP4]], <i1 true, i1 true, i1 true, i1 true>
+; DEFAULT-NEXT:    [[TMP7:%.*]] = xor <4 x i1> [[TMP5]], <i1 true, i1 true, i1 true, i1 true>
+; DEFAULT-NEXT:    [[TMP8:%.*]] = extractelement <4 x i1> [[TMP6]], i32 0
+; DEFAULT-NEXT:    br i1 [[TMP8]], label [[PRED_STORE_IF:%.*]], label [[PRED_STORE_CONTINUE:%.*]]
+; DEFAULT:       pred.store.if:
+; DEFAULT-NEXT:    store i32 0, ptr [[DST]], align 4
+; DEFAULT-NEXT:    br label [[PRED_STORE_CONTINUE]]
+; DEFAULT:       pred.store.continue:
+; DEFAULT-NEXT:    [[TMP9:%.*]] = extractelement <4 x i1> [[TMP6]], i32 1
+; DEFAULT-NEXT:    br i1 [[TMP9]], label [[PRED_STORE_IF2:%.*]], label [[PRED_STORE_CONTINUE3:%.*]]
+; DEFAULT:       pred.store.if3:
+; DEFAULT-NEXT:    store i32 0, ptr [[DST]], align 4
+; DEFAULT-NEXT:    br label [[PRED_STORE_CONTINUE3]]
+; DEFAULT:       pred.store.continue4:
+; DEFAULT-NEXT:    [[TMP10:%.*]] = extractelement <4 x i1> [[TMP6]], i32 2
+; DEFAULT-NEXT:    br i1 [[TMP10]], label [[PRED_STORE_IF4:%.*]], label [[PRED_STORE_CONTINUE5:%.*]]
+; DEFAULT:       pred.store.if5:
+; DEFAULT-NEXT:    store i32 0, ptr [[DST]], align 4
+; DEFAULT-NEXT:    br label [[PRED_STORE_CONTINUE5]]
+; DEFAULT:       pred.store.continue6:
+; DEFAULT-NEXT:    [[TMP11:%.*]] = extractelement <4 x i1> [[TMP6]], i32 3
+; DEFAULT-NEXT:    br i1 [[TMP11]], label [[PRED_STORE_IF6:%.*]], label [[PRED_STORE_CONTINUE7:%.*]]
+; DEFAULT:       pred.store.if7:
+; DEFAULT-NEXT:    store i32 0, ptr [[DST]], align 4
+; DEFAULT-NEXT:    br label [[PRED_STORE_CONTINUE7]]
+; DEFAULT:       pred.store.continue8:
+; DEFAULT-NEXT:    [[TMP12:%.*]] = extractelement <4 x i1> [[TMP7]], i32 0
+; DEFAULT-NEXT:    br i1 [[TMP12]], label [[PRED_STORE_IF8:%.*]], label [[PRED_STORE_CONTINUE9:%.*]]
+; DEFAULT:       pred.store.if9:
+; DEFAULT-NEXT:    store i32 0, ptr [[DST]], align 4
+; DEFAULT-NEXT:    br label [[PRED_STORE_CONTINUE9]]
+; DEFAULT:       pred.store.continue10:
+; DEFAULT-NEXT:    [[TMP13:%.*]] = extractelement <4 x i1> [[TMP7]], i32 1
+; DEFAULT-NEXT:    br i1 [[TMP13]], label [[PRED_STORE_IF10:%.*]], label [[PRED_STORE_CONTINUE11:%.*]]
+; DEFAULT:       pred.store.if11:
+; DEFAULT-NEXT:    store i32 0, ptr [[DST]], align 4
+; DEFAULT-NEXT:    br label [[PRED_STORE_CONTINUE11]]
+; DEFAULT:       pred.store.continue12:
+; DEFAULT-NEXT:    [[TMP14:%.*]] = extractelement <4 x i1> [[TMP7]], i32 2
+; DEFAULT-NEXT:    br i1 [[TMP14]], label [[PRED_STORE_IF12:%.*]], label [[PRED_STORE_CONTINUE13:%.*]]
+; DEFAULT:       pred.store.if13:
+; DEFAULT-NEXT:    store i32 0, ptr [[DST]], align 4
+; DEFAULT-NEXT:    br label [[PRED_STORE_CONTINUE13]]
+; DEFAULT:       pred.store.continue14:
+; DEFAULT-NEXT:    [[TMP15:%.*]] = extractelement <4 x i1> [[TMP7]], i32 3
+; DEFAULT-NEXT:    br i1 [[TMP15]], label [[PRED_STORE_IF14:%.*]], label [[PRED_STORE_CONTINUE15]]
+; DEFAULT:       pred.store.if15:
+; DEFAULT-NEXT:    store i32 0, ptr [[DST]], align 4
+; DEFAULT-NEXT:    br label [[PRED_STORE_CONTINUE15]]
+; DEFAULT:       pred.store.continue16:
+; DEFAULT-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 8
+; DEFAULT-NEXT:    [[TMP16:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
+; DEFAULT-NEXT:    br i1 [[TMP16]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP8:![0-9]+]]
+; DEFAULT:       middle.block:
+; DEFAULT-NEXT:    [[CMP_N:%.*]] = icmp eq i64 [[TMP2]], [[N_VEC]]
+; DEFAULT-NEXT:    br i1 [[CMP_N]], label [[FOR_COND_CLEANUP312_LOOPEXIT:%.*]], label [[SCALAR_PH]]
+; DEFAULT:       scalar.ph:
+; DEFAULT-NEXT:    [[BC_RESUME_VAL:%.*]] = phi ptr [ [[IND_END]], [[MIDDLE_BLOCK]] ], [ [[SRC]], [[ENTRY:%.*]] ]
+; DEFAULT-NEXT:    br label [[FOR_BODY313:%.*]]
+; DEFAULT:       loop.header:
+; DEFAULT-NEXT:    [[__BEGIN3_01197:%.*]] = phi ptr [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ], [ [[INCDEC_PTR329:%.*]], [[IF_END327:%.*]] ]
+; DEFAULT-NEXT:    [[CMP315_NOT:%.*]] = icmp sgt i32 [[A]], 0
+; DEFAULT-NEXT:    br i1 [[CMP315_NOT]], label [[IF_END327]], label [[IF_THEN316:%.*]]
+; DEFAULT:       if:
+; DEFAULT-NEXT:    br label [[IF_END327_SINK_SPLIT:%.*]]
+; DEFAULT:       else:
+; DEFAULT-NEXT:    store i32 0, ptr [[DST]], align 4
+; DEFAULT-NEXT:    br label [[IF_END327]]
+; DEFAULT:       loop.latch:
+; DEFAULT-NEXT:    [[INCDEC_PTR329]] = getelementptr inbounds i8, ptr [[__BEGIN3_01197]], i64 4
+; DEFAULT-NEXT:    [[CMP311_NOT:%.*]] = icmp eq ptr [[__BEGIN3_01197]], [[I]]
+; DEFAULT-NEXT:    br i1 [[CMP311_NOT]], label [[FOR_COND_CLEANUP312_LOOPEXIT]], label [[FOR_BODY313]], !llvm.loop [[LOOP9:![0-9]+]]
+; DEFAULT:       exit:
+; DEFAULT-NEXT:    ret void
+;
+; PRED-LABEL: define void @invar_cond_chain_2(
+; PRED-SAME: ptr [[I:%.*]], ptr noalias [[SRC:%.*]], ptr noalias [[DST:%.*]], i32 [[A:%.*]]) {
+; PRED-NEXT:  entry:
+; PRED-NEXT:    [[SRC2:%.*]] = ptrtoint ptr [[SRC]] to i64
+; PRED-NEXT:    [[I1:%.*]] = ptrtoint ptr [[I]] to i64
+; PRED-NEXT:    [[TMP0:%.*]] = sub i64 [[I1]], [[SRC2]]
+; PRED-NEXT:    [[TMP1:%.*]] = lshr i64 [[TMP0]], 2
+; PRED-NEXT:    [[TMP2:%.*]] = add nuw nsw i64 [[TMP1]], 1
+; PRED-NEXT:    br i1 false, label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
+; PRED:       vector.ph:
+; PRED-NEXT:    [[N_RND_UP:%.*]] = add i64 [[TMP2]], 3
+; PRED-NEXT:    [[N_MOD_VF:%.*]] = urem i64 [[N_RND_UP]], 4
+; PRED-NEXT:    [[N_VEC:%.*]] = sub i64 [[N_RND_UP]], [[N_MOD_VF]]
+; PRED-NEXT:    [[TMP3:%.*]] = mul i64 [[N_VEC]], 4
+; PRED-NEXT:    [[IND_END:%.*]] = getelementptr i8, ptr [[SRC]], i64 [[TMP3]]
+; PRED-NEXT:    [[TRIP_COUNT_MINUS_1:%.*]] = sub i64 [[TMP2]], 1
+; PRED-NEXT:    [[BROADCAST_SPLATINSERT2:%.*]] = insertelement <4 x i64> poison, i64 [[TRIP_COUNT_MINUS_1]], i64 0
+; PRED-NEXT:    [[BROADCAST_SPLAT3:%.*]] = shufflevector <4 x i64> [[BROADCAST_SPLATINSERT2]], <4 x i64> poison, <4 x i32> zeroinitializer
+; PRED-NEXT:    [[BROADCAST_SPLATINSERT4:%.*]] = insertelement <4 x i32> poison, i32 [[A]], i64 0
+; PRED-NEXT:    [[BROADCAST_SPLAT5:%.*]] = shufflevector <4 x i32> [[BROADCAST_SPLATINSERT4]], <4 x i32> poison, <4 x i32> zeroinitializer
+; PRED-NEXT:    br label [[VECTOR_BODY:%.*]]
+; PRED:       vector.body:
+; PRED-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[PRED_STORE_CONTINUE11:%.*]] ]
+; PRED-NEXT:    [[BROADCAST_SPLATINSERT:%.*]] = insertelement <4 x i64> poison, i64 [[INDEX]], i64 0
+; PRED-NEXT:    [[BROADCAST_SPLAT:%.*]] = shufflevector <4 x i64> [[BROADCAST_SPLATINSERT]], <4 x i64> poison, <4 x i32> zeroinitializer
+; PRED-NEXT:    [[VEC_IV:%.*]] = add <4 x i64> [[BROADCAST_SPLAT]], <i64 0, i64 1, i64 2, i64 3>
+; PRED-NEXT:    [[TMP4:%.*]] = icmp ule <4 x i64> [[VEC_IV]], [[BROADCAST_SPLAT3]]
+; PRED-NEXT:    [[TMP5:%.*]] = icmp sgt <4 x i32> [[BROADCAST_SPLAT5]], zeroinitializer
+; PRED-NEXT:    [[TMP6:%.*]] = xor <4 x i1> [[TMP5]], <i1 true, i1 true, i1 true, i1 true>
+; PRED-NEXT:    [[TMP7:%.*]] = select <4 x i1> [[TMP4]], <4 x i1> [[TMP6]], <4 x i1> zeroinitializer
+; PRED-NEXT:    [[TMP8:%.*]] = extractelement <4 x i1> [[TMP7]], i32 0
+; PRED-NEXT:    br i1 [[TMP8]], label [[PRED_STORE_IF:%.*]], label [[PRED_STORE_CONTINUE:%.*]]
+; PRED:       pred.store.if:
+; PRED-NEXT:    store i32 0, ptr [[DST]], align 4
+; PRED-NEXT:    br label [[PRED_STORE_CONTINUE]]
+; PRED:       pred.store.continue:
+; PRED-NEXT:    [[TMP9:%.*]] = extractelement <4 x i1> [[TMP7]], i32 1
+; PRED-NEXT:    br i1 [[TMP9]], label [[PRED_STORE_IF6:%.*]], label [[PRED_STORE_CONTINUE7:%.*]]
+; PRED:       pred.store.if7:
+; PRED-NEXT:    store i32 0, ptr [[DST]], align 4
+; PRED-NEXT:    br label [[PRED_STORE_CONTINUE7]]
+; PRED:       pred.store.continue8:
+; PRED-NEXT:    [[TMP10:%.*]] = extractelement <4 x i1> [[TMP7]], i32 2
+; PRED-NEXT:    br i1 [[TMP10]], label [[PRED_STORE_IF8:%.*]], label [[PRED_STORE_CONTINUE9:%.*]]
+; PRED:       pred.store.if9:
+; PRED-NEXT:    store i32 0, ptr [[DST]], align 4
+; PRED-NEXT:    br label [[PRED_STORE_CONTINUE9]]
+; PRED:       pred.store.continue10:
+; PRED-NEXT:    [[TMP11:%.*]] = extractelement <4 x i1> [[TMP7]], i32 3
+; PRED-NEXT:    br i1 [[TMP11]], label [[PRED_STORE_IF10:%.*]], label [[PRED_STORE_CONTINUE11]]
+; PRED:       pred.store.if11:
+; PRED-NEXT:    store i32 0, ptr [[DST]], align 4
+; PRED-NEXT:    br label [[PRED_STORE_CONTINUE11]]
+; PRED:       pred.store.continue12:
+; PRED-NEXT:    [[INDEX_NEXT]] = add i64 [[INDEX]], 4
+; PRED-NEXT:    [[TMP12:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
+; PRED-NEXT:    br i1 [[TMP12]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP4:![0-9]+]]
+; PRED:       middle.block:
+; PRED-NEXT:    br i1 true, label [[FOR_COND_CLEANUP312_LOOPEXIT:%.*]], label [[SCALAR_PH]]
+; PRED:       scalar.ph:
+; PRED-NEXT:    [[BC_RESUME_VAL:%.*]] = phi ptr [ [[IND_END]], [[MIDDLE_BLOCK]] ], [ [[SRC]], [[ENTRY:%.*]] ]
+; PRED-NEXT:    br label [[FOR_BODY313:%.*]]
+; PRED:       loop.header:
+; PRED-NEXT:    [[__BEGIN3_01197:%.*]] = phi ptr [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ], [ [[INCDEC_PTR329:%.*]], [[IF_END327:%.*]] ]
+; PRED-NEXT:    [[CMP315_NOT:%.*]] = icmp sgt i32 [[A]], 0
+; PRED-NEXT:    br i1 [[CMP315_NOT]], label [[IF_END327]], label [[IF_THEN316:%.*]]
+; PRED:       if:
+; PRED-NEXT:    br label [[IF_END327_SINK_SPLIT:%.*]]
+; PRED:       else:
+; PRED-NEXT:    store i32 0, ptr [[DST]], align 4
+; PRED-NEXT:    br label [[IF_END327]]
+; PRED:       loop.latch:
+; PRED-NEXT:    [[INCDEC_PTR329]] = getelementptr inbounds i8, ptr [[__BEGIN3_01197]], i64 4
+; PRED-NEXT:    [[CMP311_NOT:%.*]] = icmp eq ptr [[__BEGIN3_01197]], [[I]]
+; PRED-NEXT:    br i1 [[CMP311_NOT]], label [[FOR_COND_CLEANUP312_LOOPEXIT]], label [[FOR_BODY313]], !llvm.loop [[LOOP5:![0-9]+]]
+; PRED:       exit:
+; PRED-NEXT:    ret void
+;
+entry:
+  br label %loop.header
+
+loop.header:
+  %ptr.iv = phi ptr [ %src, %entry ], [ %ptr.iv.next, %loop.latch ]
+  %cmp315.not = icmp sgt i32 %a, 0
+  br i1 %cmp315.not, label %loop.latch, label %if
+
+if:
+  br label %else
+
+else:
+  store i32 0, ptr %dst, align 4
+  br label %loop.latch
+
+loop.latch:
+  %ptr.iv.next = getelementptr inbounds i8, ptr %ptr.iv, i64 4
+  %cmp311.not = icmp eq ptr %ptr.iv, %I
+  br i1 %cmp311.not, label %exit, label %loop.header
+
+exit:
+  ret void
+}
+
+define void @latch_branch_cost(ptr %dst) {
+; DEFAULT-LABEL: define void @latch_branch_cost(
+; DEFAULT-SAME: ptr [[DST:%.*]]) {
+; DEFAULT-NEXT:  iter.check:
+; DEFAULT-NEXT:    br i1 false, label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
+; DEFAULT:       vector.main.loop.iter.check:
+; DEFAULT-NEXT:    br i1 false, label [[VEC_EPILOG_PH:%.*]], label [[VECTOR_PH1:%.*]]
+; DEFAULT:       vector.ph:
+; DEFAULT-NEXT:    br label [[VECTOR_BODY:%.*]]
+; DEFAULT:       vector.body:
+; DEFAULT-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH1]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
+; DEFAULT-NEXT:    [[TMP0:%.*]] = add i64 [[INDEX]], 0
+; DEFAULT-NEXT:    [[TMP1:%.*]] = add i64 [[INDEX]], 16
+; DEFAULT-NEXT:    [[TMP2:%.*]] = getelementptr i8, ptr [[DST]], i64 [[TMP0]]
+; DEFAULT-NEXT:    [[TMP3:%.*]] = getelementptr i8, ptr [[DST]], i64 [[TMP1]]
+; DEFAULT-NEXT:    [[TMP6:%.*]] = getelementptr i8, ptr [[TMP2]], i32 0
+; DEFAULT-NEXT:    [[TMP5:%.*]] = getelementptr i8, ptr [[TMP2]], i32 16
+; DEFAULT-NEXT:    store <16 x i8> zeroinitializer, ptr [[TMP6]], align 1
+; DEFAULT-NEXT:    store <16 x i8> zeroinitializer, ptr [[TMP5]], align 1
+; DEFAULT-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 32
+; DEFAULT-NEXT:    [[TMP4:%.*]] = icmp eq i64 [[INDEX_NEXT]], 96
+; DEFAULT-NEXT:    br i1 [[TMP4]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP10:![0-9]+]]
+; DEFAULT:       middle.block:
+; DEFAULT-NEXT:    br i1 false, label [[FOR_END:%.*]], label [[VEC_EPILOG_ITER_CHECK:%.*]]
+; DEFAULT:       vec.epilog.iter.check:
+; DEFAULT-NEXT:    br i1 false, label [[SCALAR_PH]], label [[VEC_EPILOG_PH]]
+; DEFAULT:       vec.epilog.ph:
+; DEFAULT-NEXT:    [[VEC_EPILOG_RESUME_VAL:%.*]] = phi i64 [ 96, [[VEC_EPILOG_ITER_CHECK]] ], [ 0, [[VECTOR_PH]] ]
+; DEFAULT-NEXT:    br label [[VEC_EPILOG_VECTOR_BODY:%.*]]
+; DEFAULT:       vec.epilog.vector.body:
+; DEFAULT-NEXT:    [[INDEX1:%.*]] = phi i64 [ [[VEC_EPILOG_RESUME_VAL]], [[VEC_EPILOG_PH]] ], [ [[INDEX_NEXT2:%.*]], [[VEC_EPILOG_VECTOR_BODY]] ]
+; DEFAULT-NEXT:    [[TMP7:%.*]] = add i64 [[INDEX1]], 0
+; DEFAULT-NEXT:    [[TMP8:%.*]] = getelementptr i8, ptr [[DST]], i64 [[TMP7]]
+; DEFAULT-NEXT:    [[TMP9:%.*]] = getelementptr i8, ptr [[TMP8]], i32 0
+; DEFAULT-NEXT:    store <4 x i8> zeroinitializer, ptr [[TMP9]], align 1
+; DEFAULT-NEXT:    [[INDEX_NEXT2]] = add nuw i64 [[INDEX1]], 4
+; DEFAULT-NEXT:    [[TMP10:%.*]] = icmp eq i64 [[INDEX_NEXT2]], 100
+; DEFAULT-NEXT:    br i1 [[TMP10]], label [[VEC_EPILOG_MIDDLE_BLOCK:%.*]], label [[VEC_EPILOG_VECTOR_BODY]], !llvm.loop [[LOOP11:![0-9]+]]
+; DEFAULT:       vec.epilog.middle.block:
+; DEFAULT-NEXT:    br i1 true, label [[FOR_END]], label [[SCALAR_PH]]
+; DEFAULT:       vec.epilog.scalar.ph:
+; DEFAULT-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ 100, [[VEC_EPILOG_MIDDLE_BLOCK]] ], [ 96, [[VEC_EPILOG_ITER_CHECK]] ], [ 0, [[ITER_CHECK:%.*]] ]
+; DEFAULT-NEXT:    br label [[FOR_BODY:%.*]]
+; DEFAULT:       loop:
+; DEFAULT-NEXT:    [[INDVARS_IV:%.*]] = phi i64 [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ], [ [[INDVARS_IV_NEXT:%.*]], [[FOR_BODY]] ]
+; DEFAULT-NEXT:    [[ARRAYIDX:%.*]] = getelementptr i8, ptr [[DST]], i64 [[INDVARS_IV]]
+; DEFAULT-NEXT:    store i8 0, ptr [[ARRAYIDX]], align 1
+; DEFAULT-NEXT:    [[INDVARS_IV_NEXT]] = add i64 [[INDVARS_IV]], 1
+; DEFAULT-NEXT:    [[EXITCOND_NOT:%.*]] = icmp eq i64 [[INDVARS_IV_NEXT]], 100
+; DEFAULT-NEXT:    br i1 [[EXITCOND_NOT]], label [[FOR_END]], label [[FOR_BODY]], !llvm.loop [[LOOP12:![0-9]+]]
+; DEFAULT:       exit:
+; DEFAULT-NEXT:    ret void
+;
+; PRED-LABEL: define void @latch_branch_cost(
+; PRED-SAME: ptr [[DST:%.*]]) {
+; PRED-NEXT:  entry:
+; PRED-NEXT:    br i1 false, label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
+; PRED:       vector.ph:
+; PRED-NEXT:    br label [[VECTOR_BODY:%.*]]
+; PRED:       vector.body:
+; PRED-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[PRED_STORE_CONTINUE6:%.*]] ]
+; PRED-NEXT:    [[VEC_IND:%.*]] = phi <4 x i64> [ <i64 0, i64 1, i64 2, i64 3>, [[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], [[PRED_STORE_CONTINUE6]] ]
+; PRED-NEXT:    [[TMP0:%.*]] = icmp ule <4 x i64> [[VEC_IND]], <i64 99, i64 99, i64 99, i64 99>
+; PRED-NEXT:    [[TMP1:%.*]] = extractelement <4 x i1> [[TMP0]], i32 0
+; PRED-NEXT:    br i1 [[TMP1]], label [[PRED_STORE_IF:%.*]], label [[PRED_STORE_CONTINUE:%.*]]
+; PRED:       pred.store.if:
+; PRED-NEXT:    [[TMP2:%.*]] = add i64 [[INDEX]], 0
+; PRED-NEXT:    [[TMP3:%.*]] = getelementptr i8, ptr [[DST]], i64 [[TMP2]]
+; PRED-NEXT:    store i8 0, ptr [[TMP3]], align 1
+; PRED-NEXT:    br label [[PRED_STORE_CONTINUE]]
+; PRED:       pred.store.continue:
+; PRED-NEXT:    [[TMP4:%.*]] = extractelement <4 x i1> [[TMP0]], i32 1
+; PRED-NEXT:    br i1 [[TMP4]], label [[PRED_STORE_IF1:%.*]], label [[PRED_STORE_CONTINUE2:%.*]]
+; PRED:       pred.store.if1:
+; PRED-NEXT:    [[TMP5:%.*]] = add i64 [[INDEX]], 1
+; PRED-NEXT:    [[TMP6:%.*]] = getelementptr i8, ptr [[DST]], i64 [[TMP5]]
+; PRED-NEXT:    store i8 0, ptr [[TMP6]], align 1
+; PRED-NEXT:    br label [[PRED_STORE_CONTINUE2]]
+; PRED:       pred.store.continue2:
+; PRED-NEXT:    [[TMP7:%.*]] = extractelement <4 x i1> [[TMP0]], i32 2
+; PRED-NEXT:    br i1 [[TMP7]], label [[PRED_STORE_IF3:%.*]], label [[PRED_STORE_CONTINUE4:%.*]]
+; PRED:       pred.store.if3:
+; PRED-NEXT:    [[TMP8:%.*]] = add i64 [[INDEX]], 2
+; PRED-NEXT:    [[TMP9:%.*]] = getelementptr i8, ptr [[DST]], i64 [[TMP8]]
+; PRED-NEXT:    store i8 0, ptr [[TMP9]], align 1
+; PRED-NEXT:    br label [[PRED_STORE_CONTINUE4]]
+; PRED:       pred.store.continue4:
+; PRED-NEXT:    [[TMP10:%.*]] = extractelement <4 x i1> [[TMP0]], i32 3
+; PRED-NEXT:    br i1 [[TMP10]], label [[PRED_STORE_IF5:%.*]], label [[PRED_STORE_CONTINUE6]]
+; PRED:       pred.store.if5:
+; PRED-NEXT:    [[TMP11:%.*]] = add i64 [[INDEX]], 3
+; PRED-NEXT:    [[TMP12:%.*]] = getelementptr i8, ptr [[DST]], i64 [[TMP11]]
+; PRED-NEXT:    store i8 0, ptr [[TMP12]], align 1
+; PRED-NEXT:    br label [[PRED_STORE_CONTINUE6]]
+; PRED:       pred.store.continue6:
+; PRED-NEXT:    [[INDEX_NEXT]] = add i64 [[INDEX]], 4
+; PRED-NEXT:    [[VEC_IND_NEXT]] = add <4 x i64> [[VEC_IND]], <i64 4, i64 4, i64 4, i64 4>
+; PRED-NEXT:    [[TMP13:%.*]] = icmp eq i64 [[INDEX_NEXT]], 100
+; PRED-NEXT:    br i1 [[TMP13]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP6:![0-9]+]]
+; PRED:       middle.block:
+; PRED-NEXT:    br i1 true, label [[FOR_END:%.*]], label [[SCALAR_PH]]
+; PRED:       scalar.ph:
+; PRED-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ 100, [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ]
+; PRED-NEXT:    br label [[FOR_BODY:%.*]]
+; PRED:       loop:
+; PRED-NEXT:    [[INDVARS_IV:%.*]] = phi i64 [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ], [ [[INDVARS_IV_NEXT:%.*]], [[FOR_BODY]] ]
+; PRED-NEXT:    [[ARRAYIDX:%.*]] = getelementptr i8, ptr [[DST]], i64 [[INDVARS_IV]]
+; PRED-NEXT:    store i8 0, ptr [[ARRAYIDX]], align 1
+; PRED-NEXT:    [[INDVARS_IV_NEXT]] = add i64 [[INDVARS_IV]], 1
+; PRED-NEXT:    [[EXITCOND_NOT:%.*]] = icmp eq i64 [[INDVARS_IV_NEXT]], 100
+; PRED-NEXT:    br i1 [[EXITCOND_NOT]], label [[FOR_END]], label [[FOR_BODY]], !llvm.loop [[LOOP7:![0-9]+]]
+; PRED:       exit:
+; PRED-NEXT:    ret void
+;
+entry:
+  br label %loop
+
+loop:
+  %iv = phi i64 [ 0, %entry ], [ %iv.next, %loop ]
+  %gep = getelementptr i8, ptr %dst, i64 %iv
+  store i8 0, ptr %gep, align 1
+  %iv.next = add i64 %iv, 1
+  %ec = icmp eq i64 %iv.next, 100
+  br i1 %ec, label %exit, label %loop
+
+exit:
+  ret void
+}
+
+;.
+; DEFAULT: [[LOOP0]] = distinct !{[[LOOP0]], [[META1:![0-9]+]], [[META2:![0-9]+]]}
+; DEFAULT: [[META1]] = !{!"llvm.loop.isvectorized", i32 1}
+; DEFAULT: [[META2]] = !{!"llvm.loop.unroll.runtime.disable"}
+; DEFAULT: [[LOOP3]] = distinct !{[[LOOP3]], [[META2]], [[META1]]}
+; DEFAULT: [[LOOP4]] = distinct !{[[LOOP4]], [[META1]], [[META2]]}
+; DEFAULT: [[LOOP5]] = distinct !{[[LOOP5]], [[META2]], [[META1]]}
+; DEFAULT: [[LOOP6]] = distinct !{[[LOOP6]], [[META1]], [[META2]]}
+; DEFAULT: [[LOOP7]] = distinct !{[[LOOP7]], [[META2]], [[META1]]}
+; DEFAULT: [[LOOP8]] = distinct !{[[LOOP8]], [[META1]], [[META2]]}
+; DEFAULT: [[LOOP9]] = distinct !{[[LOOP9]], [[META2]], [[META1]]}
+; DEFAULT: [[LOOP10]] = distinct !{[[LOOP10]], [[META1]], [[META2]]}
+; DEFAULT: [[LOOP11]] = distinct !{[[LOOP11]], [[META1]], [[META2]]}
+; DEFAULT: [[LOOP12]] = distinct !{[[LOOP12]], [[META2]], [[META1]]}
+;.
+; PRED: [[LOOP0]] = distinct !{[[LOOP0]], [[META1:![0-9]+]], [[META2:![0-9]+]]}
+; PRED: [[META1]] = !{!"llvm.loop.isvectorized", i32 1}
+; PRED: [[META2]] = !{!"llvm.loop.unroll.runtime.disable"}
+; PRED: [[LOOP3]] = distinct !{[[LOOP3]], [[META2]], [[META1]]}
+; PRED: [[LOOP4]] = distinct !{[[LOOP4]], [[META1]], [[META2]]}
+; PRED: [[LOOP5]] = distinct !{[[LOOP5]], [[META2]], [[META1]]}
+; PRED: [[LOOP6]] = distinct !{[[LOOP6]], [[META1]], [[META2]]}
+; PRED: [[LOOP7]] = distinct !{[[LOOP7]], [[META2]], [[META1]]}
+;.
diff --git a/llvm/test/Transforms/LoopVectorize/AArch64/first-order-recurrence.ll b/llvm/test/Transforms/LoopVectorize/AArch64/first-order-recurrence.ll
index 2be525a2abc0..2cc0aa2ffca5 100644
--- a/llvm/test/Transforms/LoopVectorize/AArch64/first-order-recurrence.ll
+++ b/llvm/test/Transforms/LoopVectorize/AArch64/first-order-recurrence.ll
@@ -22,8 +22,8 @@ define i32 @PR33613(ptr %b, double %j, i32 %d) #0 {
 ; CHECK-VF4UF2-LABEL: @PR33613
 ; CHECK-VF4UF2: vector.body
 ; CHECK-VF4UF2: %[[VEC_RECUR:.*]] = phi <vscale x 4 x double> [ {{.*}}, %vector.ph ], [ {{.*}}, %vector.body ]
-; CHECK-VF4UF2: %[[SPLICE1:.*]] = call <vscale x 4 x double> @llvm.experimental.vector.splice.nxv4f64(<vscale x 4 x double> %[[VEC_RECUR]], <vscale x 4 x double> {{.*}}, i32 -1)
-; CHECK-VF4UF2-NEXT: %[[SPLICE2:.*]] = call <vscale x 4 x double> @llvm.experimental.vector.splice.nxv4f64(<vscale x 4 x double> %{{.*}}, <vscale x 4 x double> %{{.*}}, i32 -1)
+; CHECK-VF4UF2: %[[SPLICE1:.*]] = call <vscale x 4 x double> @llvm.vector.splice.nxv4f64(<vscale x 4 x double> %[[VEC_RECUR]], <vscale x 4 x double> {{.*}}, i32 -1)
+; CHECK-VF4UF2-NEXT: %[[SPLICE2:.*]] = call <vscale x 4 x double> @llvm.vector.splice.nxv4f64(<vscale x 4 x double> %{{.*}}, <vscale x 4 x double> %{{.*}}, i32 -1)
 ; CHECK-VF4UF2-NOT: insertelement <vscale x 4 x double>
 ; CHECK-VF4UF2: middle.block
 entry:
@@ -71,7 +71,7 @@ define void @PR34711(ptr %a, ptr %b, ptr %c, i64 %n) #0 {
 ; CHECK-VF4UF1: vector.body
 ; CHECK-VF4UF1: %[[VEC_RECUR:.*]] = phi <vscale x 4 x i16> [ %vector.recur.init, %vector.ph ], [ %[[MGATHER:.*]], %vector.body ]
 ; CHECK-VF4UF1: %[[MGATHER]] = call <vscale x 4 x i16> @llvm.masked.gather.nxv4i16.nxv4p0(<vscale x 4 x ptr> {{.*}}, i32 2, <vscale x 4 x i1> shufflevector (<vscale x 4 x i1> insertelement (<vscale x 4 x i1> poison, i1 true, i64 0), <vscale x 4 x i1> poison, <vscale x 4 x i32> zeroinitializer), <vscale x 4 x i16> poison)
-; CHECK-VF4UF1-NEXT: %[[SPLICE:.*]] = call <vscale x 4 x i16> @llvm.experimental.vector.splice.nxv4i16(<vscale x 4 x i16> %[[VEC_RECUR]], <vscale x 4 x i16> %[[MGATHER]], i32 -1)
+; CHECK-VF4UF1-NEXT: %[[SPLICE:.*]] = call <vscale x 4 x i16> @llvm.vector.splice.nxv4i16(<vscale x 4 x i16> %[[VEC_RECUR]], <vscale x 4 x i16> %[[MGATHER]], i32 -1)
 ; CHECK-VF4UF1-NEXT: %[[SXT1:.*]] = sext <vscale x 4 x i16> %[[SPLICE]] to <vscale x 4 x i32>
 ; CHECK-VF4UF1-NEXT: %[[SXT2:.*]] = sext <vscale x 4 x i16> %[[MGATHER]] to <vscale x 4 x i32>
 ; CHECK-VF4UF1-NEXT: mul nsw <vscale x 4 x i32> %[[SXT2]], %[[SXT1]]
diff --git a/llvm/test/Transforms/LoopVectorize/AArch64/induction-costs-sve.ll b/llvm/test/Transforms/LoopVectorize/AArch64/induction-costs-sve.ll
new file mode 100644
index 000000000000..af5decb0d340
--- /dev/null
+++ b/llvm/test/Transforms/LoopVectorize/AArch64/induction-costs-sve.ll
@@ -0,0 +1,916 @@
+; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --version 4
+; RUN: opt -p loop-vectorize -S %s | FileCheck --check-prefixes=DEFAULT %s
+; RUN: opt -p loop-vectorize -prefer-predicate-over-epilogue=predicate-else-scalar-epilogue -S %s | FileCheck --check-prefixes=PRED %s
+
+target datalayout = "e-m:o-i64:64-i128:128-n32:64-S128"
+target triple = "arm64-apple-macosx14.0.0"
+
+define void @iv_casts(ptr %dst, ptr %src, i32 %x, i64 %N) #0 {
+; DEFAULT-LABEL: define void @iv_casts(
+; DEFAULT-SAME: ptr [[DST:%.*]], ptr [[SRC:%.*]], i32 [[X:%.*]], i64 [[N:%.*]]) #[[ATTR0:[0-9]+]] {
+; DEFAULT-NEXT:  iter.check:
+; DEFAULT-NEXT:    [[SRC2:%.*]] = ptrtoint ptr [[SRC]] to i64
+; DEFAULT-NEXT:    [[DST1:%.*]] = ptrtoint ptr [[DST]] to i64
+; DEFAULT-NEXT:    [[TMP0:%.*]] = add i64 [[N]], 1
+; DEFAULT-NEXT:    [[TMP1:%.*]] = call i64 @llvm.vscale.i64()
+; DEFAULT-NEXT:    [[TMP2:%.*]] = mul i64 [[TMP1]], 4
+; DEFAULT-NEXT:    [[MIN_ITERS_CHECK:%.*]] = icmp ult i64 [[TMP0]], [[TMP2]]
+; DEFAULT-NEXT:    br i1 [[MIN_ITERS_CHECK]], label [[VEC_EPILOG_SCALAR_PH:%.*]], label [[VECTOR_MEMCHECK:%.*]]
+; DEFAULT:       vector.memcheck:
+; DEFAULT-NEXT:    [[TMP3:%.*]] = call i64 @llvm.vscale.i64()
+; DEFAULT-NEXT:    [[TMP4:%.*]] = mul i64 [[TMP3]], 8
+; DEFAULT-NEXT:    [[TMP5:%.*]] = mul i64 [[TMP4]], 2
+; DEFAULT-NEXT:    [[TMP6:%.*]] = sub i64 [[DST1]], [[SRC2]]
+; DEFAULT-NEXT:    [[DIFF_CHECK:%.*]] = icmp ult i64 [[TMP6]], [[TMP5]]
+; DEFAULT-NEXT:    br i1 [[DIFF_CHECK]], label [[VEC_EPILOG_SCALAR_PH]], label [[VECTOR_MAIN_LOOP_ITER_CHECK:%.*]]
+; DEFAULT:       vector.main.loop.iter.check:
+; DEFAULT-NEXT:    [[TMP7:%.*]] = call i64 @llvm.vscale.i64()
+; DEFAULT-NEXT:    [[TMP8:%.*]] = mul i64 [[TMP7]], 16
+; DEFAULT-NEXT:    [[MIN_ITERS_CHECK3:%.*]] = icmp ult i64 [[TMP0]], [[TMP8]]
+; DEFAULT-NEXT:    br i1 [[MIN_ITERS_CHECK3]], label [[VEC_EPILOG_PH:%.*]], label [[VECTOR_PH:%.*]]
+; DEFAULT:       vector.ph:
+; DEFAULT-NEXT:    [[TMP9:%.*]] = call i64 @llvm.vscale.i64()
+; DEFAULT-NEXT:    [[TMP10:%.*]] = mul i64 [[TMP9]], 16
+; DEFAULT-NEXT:    [[N_MOD_VF:%.*]] = urem i64 [[TMP0]], [[TMP10]]
+; DEFAULT-NEXT:    [[N_VEC:%.*]] = sub i64 [[TMP0]], [[N_MOD_VF]]
+; DEFAULT-NEXT:    [[TMP11:%.*]] = call i64 @llvm.vscale.i64()
+; DEFAULT-NEXT:    [[TMP12:%.*]] = mul i64 [[TMP11]], 16
+; DEFAULT-NEXT:    [[BROADCAST_SPLATINSERT:%.*]] = insertelement <vscale x 8 x i32> poison, i32 [[X]], i64 0
+; DEFAULT-NEXT:    [[BROADCAST_SPLAT:%.*]] = shufflevector <vscale x 8 x i32> [[BROADCAST_SPLATINSERT]], <vscale x 8 x i32> poison, <vscale x 8 x i32> zeroinitializer
+; DEFAULT-NEXT:    [[TMP13:%.*]] = trunc <vscale x 8 x i32> [[BROADCAST_SPLAT]] to <vscale x 8 x i16>
+; DEFAULT-NEXT:    br label [[VECTOR_BODY:%.*]]
+; DEFAULT:       vector.body:
+; DEFAULT-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
+; DEFAULT-NEXT:    [[TMP14:%.*]] = add i64 [[INDEX]], 0
+; DEFAULT-NEXT:    [[TMP15:%.*]] = call i64 @llvm.vscale.i64()
+; DEFAULT-NEXT:    [[TMP16:%.*]] = mul i64 [[TMP15]], 8
+; DEFAULT-NEXT:    [[TMP17:%.*]] = add i64 [[TMP16]], 0
+; DEFAULT-NEXT:    [[TMP18:%.*]] = mul i64 [[TMP17]], 1
+; DEFAULT-NEXT:    [[TMP19:%.*]] = add i64 [[INDEX]], [[TMP18]]
+; DEFAULT-NEXT:    [[TMP20:%.*]] = getelementptr i8, ptr [[SRC]], i64 [[TMP14]]
+; DEFAULT-NEXT:    [[TMP21:%.*]] = getelementptr i8, ptr [[SRC]], i64 [[TMP19]]
+; DEFAULT-NEXT:    [[TMP22:%.*]] = getelementptr i8, ptr [[TMP20]], i32 0
+; DEFAULT-NEXT:    [[TMP23:%.*]] = call i64 @llvm.vscale.i64()
+; DEFAULT-NEXT:    [[TMP24:%.*]] = mul i64 [[TMP23]], 8
+; DEFAULT-NEXT:    [[TMP25:%.*]] = getelementptr i8, ptr [[TMP20]], i64 [[TMP24]]
+; DEFAULT-NEXT:    [[WIDE_LOAD:%.*]] = load <vscale x 8 x i8>, ptr [[TMP22]], align 1
+; DEFAULT-NEXT:    [[WIDE_LOAD4:%.*]] = load <vscale x 8 x i8>, ptr [[TMP25]], align 1
+; DEFAULT-NEXT:    [[TMP26:%.*]] = zext <vscale x 8 x i8> [[WIDE_LOAD]] to <vscale x 8 x i16>
+; DEFAULT-NEXT:    [[TMP27:%.*]] = zext <vscale x 8 x i8> [[WIDE_LOAD4]] to <vscale x 8 x i16>
+; DEFAULT-NEXT:    [[TMP28:%.*]] = mul <vscale x 8 x i16> [[TMP26]], [[TMP13]]
+; DEFAULT-NEXT:    [[TMP29:%.*]] = mul <vscale x 8 x i16> [[TMP27]], [[TMP13]]
+; DEFAULT-NEXT:    [[TMP30:%.*]] = zext <vscale x 8 x i8> [[WIDE_LOAD]] to <vscale x 8 x i16>
+; DEFAULT-NEXT:    [[TMP31:%.*]] = zext <vscale x 8 x i8> [[WIDE_LOAD4]] to <vscale x 8 x i16>
+; DEFAULT-NEXT:    [[TMP32:%.*]] = or <vscale x 8 x i16> [[TMP28]], [[TMP30]]
+; DEFAULT-NEXT:    [[TMP33:%.*]] = or <vscale x 8 x i16> [[TMP29]], [[TMP31]]
+; DEFAULT-NEXT:    [[TMP34:%.*]] = lshr <vscale x 8 x i16> [[TMP32]], trunc (<vscale x 8 x i32> shufflevector (<vscale x 8 x i32> insertelement (<vscale x 8 x i32> poison, i32 1, i64 0), <vscale x 8 x i32> poison, <vscale x 8 x i32> zeroinitializer) to <vscale x 8 x i16>)
+; DEFAULT-NEXT:    [[TMP35:%.*]] = lshr <vscale x 8 x i16> [[TMP33]], trunc (<vscale x 8 x i32> shufflevector (<vscale x 8 x i32> insertelement (<vscale x 8 x i32> poison, i32 1, i64 0), <vscale x 8 x i32> poison, <vscale x 8 x i32> zeroinitializer) to <vscale x 8 x i16>)
+; DEFAULT-NEXT:    [[TMP36:%.*]] = trunc <vscale x 8 x i16> [[TMP34]] to <vscale x 8 x i8>
+; DEFAULT-NEXT:    [[TMP37:%.*]] = trunc <vscale x 8 x i16> [[TMP35]] to <vscale x 8 x i8>
+; DEFAULT-NEXT:    [[TMP38:%.*]] = getelementptr i8, ptr [[DST]], i64 [[TMP14]]
+; DEFAULT-NEXT:    [[TMP39:%.*]] = getelementptr i8, ptr [[DST]], i64 [[TMP19]]
+; DEFAULT-NEXT:    [[TMP40:%.*]] = getelementptr i8, ptr [[TMP38]], i32 0
+; DEFAULT-NEXT:    [[TMP41:%.*]] = call i64 @llvm.vscale.i64()
+; DEFAULT-NEXT:    [[TMP42:%.*]] = mul i64 [[TMP41]], 8
+; DEFAULT-NEXT:    [[TMP43:%.*]] = getelementptr i8, ptr [[TMP38]], i64 [[TMP42]]
+; DEFAULT-NEXT:    store <vscale x 8 x i8> [[TMP36]], ptr [[TMP40]], align 1
+; DEFAULT-NEXT:    store <vscale x 8 x i8> [[TMP37]], ptr [[TMP43]], align 1
+; DEFAULT-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], [[TMP12]]
+; DEFAULT-NEXT:    [[TMP44:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
+; DEFAULT-NEXT:    br i1 [[TMP44]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP0:![0-9]+]]
+; DEFAULT:       middle.block:
+; DEFAULT-NEXT:    [[CMP_N:%.*]] = icmp eq i64 [[TMP0]], [[N_VEC]]
+; DEFAULT-NEXT:    br i1 [[CMP_N]], label [[EXIT:%.*]], label [[VEC_EPILOG_ITER_CHECK:%.*]]
+; DEFAULT:       vec.epilog.iter.check:
+; DEFAULT-NEXT:    [[N_VEC_REMAINING:%.*]] = sub i64 [[TMP0]], [[N_VEC]]
+; DEFAULT-NEXT:    [[TMP45:%.*]] = call i64 @llvm.vscale.i64()
+; DEFAULT-NEXT:    [[TMP46:%.*]] = mul i64 [[TMP45]], 4
+; DEFAULT-NEXT:    [[MIN_EPILOG_ITERS_CHECK:%.*]] = icmp ult i64 [[N_VEC_REMAINING]], [[TMP46]]
+; DEFAULT-NEXT:    br i1 [[MIN_EPILOG_ITERS_CHECK]], label [[VEC_EPILOG_SCALAR_PH]], label [[VEC_EPILOG_PH]]
+; DEFAULT:       vec.epilog.ph:
+; DEFAULT-NEXT:    [[VEC_EPILOG_RESUME_VAL:%.*]] = phi i64 [ [[N_VEC]], [[VEC_EPILOG_ITER_CHECK]] ], [ 0, [[VECTOR_MAIN_LOOP_ITER_CHECK]] ]
+; DEFAULT-NEXT:    [[TMP47:%.*]] = call i64 @llvm.vscale.i64()
+; DEFAULT-NEXT:    [[TMP48:%.*]] = mul i64 [[TMP47]], 4
+; DEFAULT-NEXT:    [[N_MOD_VF5:%.*]] = urem i64 [[TMP0]], [[TMP48]]
+; DEFAULT-NEXT:    [[N_VEC6:%.*]] = sub i64 [[TMP0]], [[N_MOD_VF5]]
+; DEFAULT-NEXT:    [[TMP49:%.*]] = call i64 @llvm.vscale.i64()
+; DEFAULT-NEXT:    [[TMP50:%.*]] = mul i64 [[TMP49]], 4
+; DEFAULT-NEXT:    [[BROADCAST_SPLATINSERT8:%.*]] = insertelement <vscale x 4 x i32> poison, i32 [[X]], i64 0
+; DEFAULT-NEXT:    [[BROADCAST_SPLAT9:%.*]] = shufflevector <vscale x 4 x i32> [[BROADCAST_SPLATINSERT8]], <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer
+; DEFAULT-NEXT:    [[TMP51:%.*]] = trunc <vscale x 4 x i32> [[BROADCAST_SPLAT9]] to <vscale x 4 x i16>
+; DEFAULT-NEXT:    br label [[VEC_EPILOG_VECTOR_BODY:%.*]]
+; DEFAULT:       vec.epilog.vector.body:
+; DEFAULT-NEXT:    [[INDEX10:%.*]] = phi i64 [ [[VEC_EPILOG_RESUME_VAL]], [[VEC_EPILOG_PH]] ], [ [[INDEX_NEXT12:%.*]], [[VEC_EPILOG_VECTOR_BODY]] ]
+; DEFAULT-NEXT:    [[TMP52:%.*]] = add i64 [[INDEX10]], 0
+; DEFAULT-NEXT:    [[TMP53:%.*]] = getelementptr i8, ptr [[SRC]], i64 [[TMP52]]
+; DEFAULT-NEXT:    [[TMP54:%.*]] = getelementptr i8, ptr [[TMP53]], i32 0
+; DEFAULT-NEXT:    [[WIDE_LOAD11:%.*]] = load <vscale x 4 x i8>, ptr [[TMP54]], align 1
+; DEFAULT-NEXT:    [[TMP55:%.*]] = zext <vscale x 4 x i8> [[WIDE_LOAD11]] to <vscale x 4 x i16>
+; DEFAULT-NEXT:    [[TMP56:%.*]] = mul <vscale x 4 x i16> [[TMP55]], [[TMP51]]
+; DEFAULT-NEXT:    [[TMP57:%.*]] = zext <vscale x 4 x i8> [[WIDE_LOAD11]] to <vscale x 4 x i16>
+; DEFAULT-NEXT:    [[TMP58:%.*]] = or <vscale x 4 x i16> [[TMP56]], [[TMP57]]
+; DEFAULT-NEXT:    [[TMP59:%.*]] = lshr <vscale x 4 x i16> [[TMP58]], trunc (<vscale x 4 x i32> shufflevector (<vscale x 4 x i32> insertelement (<vscale x 4 x i32> poison, i32 1, i64 0), <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer) to <vscale x 4 x i16>)
+; DEFAULT-NEXT:    [[TMP60:%.*]] = trunc <vscale x 4 x i16> [[TMP59]] to <vscale x 4 x i8>
+; DEFAULT-NEXT:    [[TMP61:%.*]] = getelementptr i8, ptr [[DST]], i64 [[TMP52]]
+; DEFAULT-NEXT:    [[TMP62:%.*]] = getelementptr i8, ptr [[TMP61]], i32 0
+; DEFAULT-NEXT:    store <vscale x 4 x i8> [[TMP60]], ptr [[TMP62]], align 1
+; DEFAULT-NEXT:    [[INDEX_NEXT12]] = add nuw i64 [[INDEX10]], [[TMP50]]
+; DEFAULT-NEXT:    [[TMP63:%.*]] = icmp eq i64 [[INDEX_NEXT12]], [[N_VEC6]]
+; DEFAULT-NEXT:    br i1 [[TMP63]], label [[VEC_EPILOG_MIDDLE_BLOCK:%.*]], label [[VEC_EPILOG_VECTOR_BODY]], !llvm.loop [[LOOP3:![0-9]+]]
+; DEFAULT:       vec.epilog.middle.block:
+; DEFAULT-NEXT:    [[CMP_N7:%.*]] = icmp eq i64 [[TMP0]], [[N_VEC6]]
+; DEFAULT-NEXT:    br i1 [[CMP_N7]], label [[EXIT]], label [[VEC_EPILOG_SCALAR_PH]]
+; DEFAULT:       vec.epilog.scalar.ph:
+; DEFAULT-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ [[N_VEC6]], [[VEC_EPILOG_MIDDLE_BLOCK]] ], [ [[N_VEC]], [[VEC_EPILOG_ITER_CHECK]] ], [ 0, [[VECTOR_MEMCHECK]] ], [ 0, [[ITER_CHECK:%.*]] ]
+; DEFAULT-NEXT:    br label [[LOOP:%.*]]
+; DEFAULT:       loop:
+; DEFAULT-NEXT:    [[IV:%.*]] = phi i64 [ [[BC_RESUME_VAL]], [[VEC_EPILOG_SCALAR_PH]] ], [ [[IV_NEXT:%.*]], [[LOOP]] ]
+; DEFAULT-NEXT:    [[GEP_SRC:%.*]] = getelementptr i8, ptr [[SRC]], i64 [[IV]]
+; DEFAULT-NEXT:    [[L:%.*]] = load i8, ptr [[GEP_SRC]], align 1
+; DEFAULT-NEXT:    [[L_EXT:%.*]] = zext i8 [[L]] to i32
+; DEFAULT-NEXT:    [[MUL16_US:%.*]] = mul i32 [[L_EXT]], [[X]]
+; DEFAULT-NEXT:    [[IV_NEXT]] = add i64 [[IV]], 1
+; DEFAULT-NEXT:    [[CONV25_US:%.*]] = zext i8 [[L]] to i32
+; DEFAULT-NEXT:    [[ADD34_US:%.*]] = or i32 [[MUL16_US]], [[CONV25_US]]
+; DEFAULT-NEXT:    [[SHR35_US:%.*]] = lshr i32 [[ADD34_US]], 1
+; DEFAULT-NEXT:    [[CONV36_US:%.*]] = trunc i32 [[SHR35_US]] to i8
+; DEFAULT-NEXT:    [[GEP_DST:%.*]] = getelementptr i8, ptr [[DST]], i64 [[IV]]
+; DEFAULT-NEXT:    store i8 [[CONV36_US]], ptr [[GEP_DST]], align 1
+; DEFAULT-NEXT:    [[EC:%.*]] = icmp eq i64 [[IV]], [[N]]
+; DEFAULT-NEXT:    br i1 [[EC]], label [[EXIT]], label [[LOOP]], !llvm.loop [[LOOP4:![0-9]+]]
+; DEFAULT:       exit:
+; DEFAULT-NEXT:    ret void
+;
+; PRED-LABEL: define void @iv_casts(
+; PRED-SAME: ptr [[DST:%.*]], ptr [[SRC:%.*]], i32 [[X:%.*]], i64 [[N:%.*]]) #[[ATTR0:[0-9]+]] {
+; PRED-NEXT:  entry:
+; PRED-NEXT:    [[SRC2:%.*]] = ptrtoint ptr [[SRC]] to i64
+; PRED-NEXT:    [[DST1:%.*]] = ptrtoint ptr [[DST]] to i64
+; PRED-NEXT:    [[TMP0:%.*]] = add i64 [[N]], 1
+; PRED-NEXT:    br i1 false, label [[SCALAR_PH:%.*]], label [[VECTOR_MEMCHECK:%.*]]
+; PRED:       vector.memcheck:
+; PRED-NEXT:    [[TMP1:%.*]] = call i64 @llvm.vscale.i64()
+; PRED-NEXT:    [[TMP2:%.*]] = mul i64 [[TMP1]], 8
+; PRED-NEXT:    [[TMP3:%.*]] = sub i64 [[DST1]], [[SRC2]]
+; PRED-NEXT:    [[DIFF_CHECK:%.*]] = icmp ult i64 [[TMP3]], [[TMP2]]
+; PRED-NEXT:    br i1 [[DIFF_CHECK]], label [[SCALAR_PH]], label [[VECTOR_PH:%.*]]
+; PRED:       vector.ph:
+; PRED-NEXT:    [[TMP4:%.*]] = call i64 @llvm.vscale.i64()
+; PRED-NEXT:    [[TMP5:%.*]] = mul i64 [[TMP4]], 8
+; PRED-NEXT:    [[TMP6:%.*]] = call i64 @llvm.vscale.i64()
+; PRED-NEXT:    [[TMP7:%.*]] = mul i64 [[TMP6]], 8
+; PRED-NEXT:    [[TMP8:%.*]] = sub i64 [[TMP7]], 1
+; PRED-NEXT:    [[N_RND_UP:%.*]] = add i64 [[TMP0]], [[TMP8]]
+; PRED-NEXT:    [[N_MOD_VF:%.*]] = urem i64 [[N_RND_UP]], [[TMP5]]
+; PRED-NEXT:    [[N_VEC:%.*]] = sub i64 [[N_RND_UP]], [[N_MOD_VF]]
+; PRED-NEXT:    [[TMP9:%.*]] = call i64 @llvm.vscale.i64()
+; PRED-NEXT:    [[TMP10:%.*]] = mul i64 [[TMP9]], 8
+; PRED-NEXT:    [[TMP11:%.*]] = call i64 @llvm.vscale.i64()
+; PRED-NEXT:    [[TMP12:%.*]] = mul i64 [[TMP11]], 8
+; PRED-NEXT:    [[TMP13:%.*]] = sub i64 [[TMP0]], [[TMP12]]
+; PRED-NEXT:    [[TMP14:%.*]] = icmp ugt i64 [[TMP0]], [[TMP12]]
+; PRED-NEXT:    [[TMP15:%.*]] = select i1 [[TMP14]], i64 [[TMP13]], i64 0
+; PRED-NEXT:    [[ACTIVE_LANE_MASK_ENTRY:%.*]] = call <vscale x 8 x i1> @llvm.get.active.lane.mask.nxv8i1.i64(i64 0, i64 [[TMP0]])
+; PRED-NEXT:    [[BROADCAST_SPLATINSERT:%.*]] = insertelement <vscale x 8 x i32> poison, i32 [[X]], i64 0
+; PRED-NEXT:    [[BROADCAST_SPLAT:%.*]] = shufflevector <vscale x 8 x i32> [[BROADCAST_SPLATINSERT]], <vscale x 8 x i32> poison, <vscale x 8 x i32> zeroinitializer
+; PRED-NEXT:    [[TMP16:%.*]] = trunc <vscale x 8 x i32> [[BROADCAST_SPLAT]] to <vscale x 8 x i16>
+; PRED-NEXT:    br label [[VECTOR_BODY:%.*]]
+; PRED:       vector.body:
+; PRED-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
+; PRED-NEXT:    [[ACTIVE_LANE_MASK:%.*]] = phi <vscale x 8 x i1> [ [[ACTIVE_LANE_MASK_ENTRY]], [[VECTOR_PH]] ], [ [[ACTIVE_LANE_MASK_NEXT:%.*]], [[VECTOR_BODY]] ]
+; PRED-NEXT:    [[TMP17:%.*]] = add i64 [[INDEX]], 0
+; PRED-NEXT:    [[TMP18:%.*]] = getelementptr i8, ptr [[SRC]], i64 [[TMP17]]
+; PRED-NEXT:    [[TMP19:%.*]] = getelementptr i8, ptr [[TMP18]], i32 0
+; PRED-NEXT:    [[WIDE_MASKED_LOAD:%.*]] = call <vscale x 8 x i8> @llvm.masked.load.nxv8i8.p0(ptr [[TMP19]], i32 1, <vscale x 8 x i1> [[ACTIVE_LANE_MASK]], <vscale x 8 x i8> poison)
+; PRED-NEXT:    [[TMP20:%.*]] = zext <vscale x 8 x i8> [[WIDE_MASKED_LOAD]] to <vscale x 8 x i16>
+; PRED-NEXT:    [[TMP21:%.*]] = mul <vscale x 8 x i16> [[TMP20]], [[TMP16]]
+; PRED-NEXT:    [[TMP22:%.*]] = zext <vscale x 8 x i8> [[WIDE_MASKED_LOAD]] to <vscale x 8 x i16>
+; PRED-NEXT:    [[TMP23:%.*]] = or <vscale x 8 x i16> [[TMP21]], [[TMP22]]
+; PRED-NEXT:    [[TMP24:%.*]] = lshr <vscale x 8 x i16> [[TMP23]], trunc (<vscale x 8 x i32> shufflevector (<vscale x 8 x i32> insertelement (<vscale x 8 x i32> poison, i32 1, i64 0), <vscale x 8 x i32> poison, <vscale x 8 x i32> zeroinitializer) to <vscale x 8 x i16>)
+; PRED-NEXT:    [[TMP25:%.*]] = trunc <vscale x 8 x i16> [[TMP24]] to <vscale x 8 x i8>
+; PRED-NEXT:    [[TMP26:%.*]] = getelementptr i8, ptr [[DST]], i64 [[TMP17]]
+; PRED-NEXT:    [[TMP27:%.*]] = getelementptr i8, ptr [[TMP26]], i32 0
+; PRED-NEXT:    call void @llvm.masked.store.nxv8i8.p0(<vscale x 8 x i8> [[TMP25]], ptr [[TMP27]], i32 1, <vscale x 8 x i1> [[ACTIVE_LANE_MASK]])
+; PRED-NEXT:    [[INDEX_NEXT]] = add i64 [[INDEX]], [[TMP10]]
+; PRED-NEXT:    [[ACTIVE_LANE_MASK_NEXT]] = call <vscale x 8 x i1> @llvm.get.active.lane.mask.nxv8i1.i64(i64 [[INDEX]], i64 [[TMP15]])
+; PRED-NEXT:    [[TMP28:%.*]] = xor <vscale x 8 x i1> [[ACTIVE_LANE_MASK_NEXT]], shufflevector (<vscale x 8 x i1> insertelement (<vscale x 8 x i1> poison, i1 true, i64 0), <vscale x 8 x i1> poison, <vscale x 8 x i32> zeroinitializer)
+; PRED-NEXT:    [[TMP29:%.*]] = extractelement <vscale x 8 x i1> [[TMP28]], i32 0
+; PRED-NEXT:    br i1 [[TMP29]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP0:![0-9]+]]
+; PRED:       middle.block:
+; PRED-NEXT:    br i1 true, label [[EXIT:%.*]], label [[SCALAR_PH]]
+; PRED:       scalar.ph:
+; PRED-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ [[N_VEC]], [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ], [ 0, [[VECTOR_MEMCHECK]] ]
+; PRED-NEXT:    br label [[LOOP:%.*]]
+; PRED:       loop:
+; PRED-NEXT:    [[IV:%.*]] = phi i64 [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ], [ [[IV_NEXT:%.*]], [[LOOP]] ]
+; PRED-NEXT:    [[GEP_SRC:%.*]] = getelementptr i8, ptr [[SRC]], i64 [[IV]]
+; PRED-NEXT:    [[L:%.*]] = load i8, ptr [[GEP_SRC]], align 1
+; PRED-NEXT:    [[L_EXT:%.*]] = zext i8 [[L]] to i32
+; PRED-NEXT:    [[MUL16_US:%.*]] = mul i32 [[L_EXT]], [[X]]
+; PRED-NEXT:    [[IV_NEXT]] = add i64 [[IV]], 1
+; PRED-NEXT:    [[CONV25_US:%.*]] = zext i8 [[L]] to i32
+; PRED-NEXT:    [[ADD34_US:%.*]] = or i32 [[MUL16_US]], [[CONV25_US]]
+; PRED-NEXT:    [[SHR35_US:%.*]] = lshr i32 [[ADD34_US]], 1
+; PRED-NEXT:    [[CONV36_US:%.*]] = trunc i32 [[SHR35_US]] to i8
+; PRED-NEXT:    [[GEP_DST:%.*]] = getelementptr i8, ptr [[DST]], i64 [[IV]]
+; PRED-NEXT:    store i8 [[CONV36_US]], ptr [[GEP_DST]], align 1
+; PRED-NEXT:    [[EC:%.*]] = icmp eq i64 [[IV]], [[N]]
+; PRED-NEXT:    br i1 [[EC]], label [[EXIT]], label [[LOOP]], !llvm.loop [[LOOP3:![0-9]+]]
+; PRED:       exit:
+; PRED-NEXT:    ret void
+;
+entry:
+  br label %loop
+
+loop:
+  %iv = phi i64 [ 0, %entry ], [ %iv.next, %loop ]
+  %gep.src  = getelementptr i8, ptr %src, i64 %iv
+  %l = load i8, ptr %gep.src, align 1
+  %l.ext = zext i8 %l to i32
+  %mul = mul i32 %l.ext, %x
+  %iv.next = add i64 %iv, 1
+  %l.ext.2 = zext i8 %l to i32
+  %or = or i32 %mul, %l.ext.2
+  %lshr  = lshr i32 %or, 1
+  %trunc = trunc i32 %lshr to i8
+  %gep.dst = getelementptr i8, ptr %dst, i64 %iv
+  store i8 %trunc, ptr %gep.dst, align 1
+  %ec = icmp eq i64 %iv, %N
+  br i1 %ec, label %exit, label %loop
+
+exit:
+  ret void
+}
+
+define void @iv_trunc(i32 %x, ptr %dst, i64 %N) #0 {
+; DEFAULT-LABEL: define void @iv_trunc(
+; DEFAULT-SAME: i32 [[X:%.*]], ptr [[DST:%.*]], i64 [[N:%.*]]) #[[ATTR0]] {
+; DEFAULT-NEXT:  entry:
+; DEFAULT-NEXT:    [[MUL_X:%.*]] = add i32 [[X]], 1
+; DEFAULT-NEXT:    [[TMP0:%.*]] = add i64 [[N]], 1
+; DEFAULT-NEXT:    [[MIN_ITERS_CHECK:%.*]] = icmp ult i64 [[TMP0]], 2
+; DEFAULT-NEXT:    br i1 [[MIN_ITERS_CHECK]], label [[SCALAR_PH:%.*]], label [[VECTOR_SCEVCHECK:%.*]]
+; DEFAULT:       vector.scevcheck:
+; DEFAULT-NEXT:    [[TMP1:%.*]] = sub i32 -1, [[X]]
+; DEFAULT-NEXT:    [[TMP2:%.*]] = icmp slt i32 [[MUL_X]], 0
+; DEFAULT-NEXT:    [[TMP3:%.*]] = select i1 [[TMP2]], i32 [[TMP1]], i32 [[MUL_X]]
+; DEFAULT-NEXT:    [[TMP4:%.*]] = trunc i64 [[N]] to i32
+; DEFAULT-NEXT:    [[MUL:%.*]] = call { i32, i1 } @llvm.umul.with.overflow.i32(i32 [[TMP3]], i32 [[TMP4]])
+; DEFAULT-NEXT:    [[MUL_RESULT:%.*]] = extractvalue { i32, i1 } [[MUL]], 0
+; DEFAULT-NEXT:    [[MUL_OVERFLOW:%.*]] = extractvalue { i32, i1 } [[MUL]], 1
+; DEFAULT-NEXT:    [[TMP5:%.*]] = sub i32 0, [[MUL_RESULT]]
+; DEFAULT-NEXT:    [[TMP6:%.*]] = icmp ugt i32 [[TMP5]], 0
+; DEFAULT-NEXT:    [[TMP7:%.*]] = select i1 [[TMP2]], i1 [[TMP6]], i1 false
+; DEFAULT-NEXT:    [[TMP8:%.*]] = or i1 [[TMP7]], [[MUL_OVERFLOW]]
+; DEFAULT-NEXT:    [[TMP9:%.*]] = icmp ugt i64 [[N]], 4294967295
+; DEFAULT-NEXT:    [[TMP10:%.*]] = icmp ne i32 [[MUL_X]], 0
+; DEFAULT-NEXT:    [[TMP11:%.*]] = and i1 [[TMP9]], [[TMP10]]
+; DEFAULT-NEXT:    [[TMP12:%.*]] = or i1 [[TMP8]], [[TMP11]]
+; DEFAULT-NEXT:    br i1 [[TMP12]], label [[SCALAR_PH]], label [[VECTOR_PH:%.*]]
+; DEFAULT:       vector.ph:
+; DEFAULT-NEXT:    [[N_MOD_VF:%.*]] = urem i64 [[TMP0]], 2
+; DEFAULT-NEXT:    [[N_VEC:%.*]] = sub i64 [[TMP0]], [[N_MOD_VF]]
+; DEFAULT-NEXT:    br label [[VECTOR_BODY:%.*]]
+; DEFAULT:       vector.body:
+; DEFAULT-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
+; DEFAULT-NEXT:    [[TMP13:%.*]] = trunc i64 [[INDEX]] to i32
+; DEFAULT-NEXT:    [[TMP14:%.*]] = add i32 [[TMP13]], 0
+; DEFAULT-NEXT:    [[TMP15:%.*]] = add i32 [[TMP13]], 1
+; DEFAULT-NEXT:    [[TMP16:%.*]] = mul i32 [[MUL_X]], [[TMP14]]
+; DEFAULT-NEXT:    [[TMP17:%.*]] = mul i32 [[MUL_X]], [[TMP15]]
+; DEFAULT-NEXT:    [[TMP18:%.*]] = zext i32 [[TMP16]] to i64
+; DEFAULT-NEXT:    [[TMP19:%.*]] = zext i32 [[TMP17]] to i64
+; DEFAULT-NEXT:    [[TMP20:%.*]] = getelementptr i32, ptr [[DST]], i64 [[TMP18]]
+; DEFAULT-NEXT:    [[TMP21:%.*]] = getelementptr i32, ptr [[DST]], i64 [[TMP19]]
+; DEFAULT-NEXT:    store i32 1, ptr [[TMP20]], align 4
+; DEFAULT-NEXT:    store i32 1, ptr [[TMP21]], align 4
+; DEFAULT-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 2
+; DEFAULT-NEXT:    [[TMP22:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
+; DEFAULT-NEXT:    br i1 [[TMP22]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP5:![0-9]+]]
+; DEFAULT:       middle.block:
+; DEFAULT-NEXT:    [[CMP_N:%.*]] = icmp eq i64 [[TMP0]], [[N_VEC]]
+; DEFAULT-NEXT:    br i1 [[CMP_N]], label [[EXIT:%.*]], label [[SCALAR_PH]]
+; DEFAULT:       scalar.ph:
+; DEFAULT-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ [[N_VEC]], [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ], [ 0, [[VECTOR_SCEVCHECK]] ]
+; DEFAULT-NEXT:    br label [[FOR_BODY:%.*]]
+; DEFAULT:       for.body:
+; DEFAULT-NEXT:    [[IV:%.*]] = phi i64 [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ], [ [[IV_NEXT:%.*]], [[FOR_BODY]] ]
+; DEFAULT-NEXT:    [[TRUNC_IV:%.*]] = trunc i64 [[IV]] to i32
+; DEFAULT-NEXT:    [[ADD_I:%.*]] = mul i32 [[MUL_X]], [[TRUNC_IV]]
+; DEFAULT-NEXT:    [[IV_MUL:%.*]] = zext i32 [[ADD_I]] to i64
+; DEFAULT-NEXT:    [[GEP:%.*]] = getelementptr i32, ptr [[DST]], i64 [[IV_MUL]]
+; DEFAULT-NEXT:    store i32 1, ptr [[GEP]], align 4
+; DEFAULT-NEXT:    [[IV_NEXT]] = add i64 [[IV]], 1
+; DEFAULT-NEXT:    [[EC:%.*]] = icmp eq i64 [[IV]], [[N]]
+; DEFAULT-NEXT:    br i1 [[EC]], label [[EXIT]], label [[FOR_BODY]], !llvm.loop [[LOOP6:![0-9]+]]
+; DEFAULT:       exit:
+; DEFAULT-NEXT:    ret void
+;
+; PRED-LABEL: define void @iv_trunc(
+; PRED-SAME: i32 [[X:%.*]], ptr [[DST:%.*]], i64 [[N:%.*]]) #[[ATTR0]] {
+; PRED-NEXT:  entry:
+; PRED-NEXT:    [[MUL_X:%.*]] = add i32 [[X]], 1
+; PRED-NEXT:    [[TMP0:%.*]] = add i64 [[N]], 1
+; PRED-NEXT:    br i1 false, label [[SCALAR_PH:%.*]], label [[VECTOR_SCEVCHECK:%.*]]
+; PRED:       vector.scevcheck:
+; PRED-NEXT:    [[TMP1:%.*]] = sub i32 -1, [[X]]
+; PRED-NEXT:    [[TMP2:%.*]] = icmp slt i32 [[MUL_X]], 0
+; PRED-NEXT:    [[TMP3:%.*]] = select i1 [[TMP2]], i32 [[TMP1]], i32 [[MUL_X]]
+; PRED-NEXT:    [[TMP4:%.*]] = trunc i64 [[N]] to i32
+; PRED-NEXT:    [[MUL:%.*]] = call { i32, i1 } @llvm.umul.with.overflow.i32(i32 [[TMP3]], i32 [[TMP4]])
+; PRED-NEXT:    [[MUL_RESULT:%.*]] = extractvalue { i32, i1 } [[MUL]], 0
+; PRED-NEXT:    [[MUL_OVERFLOW:%.*]] = extractvalue { i32, i1 } [[MUL]], 1
+; PRED-NEXT:    [[TMP5:%.*]] = sub i32 0, [[MUL_RESULT]]
+; PRED-NEXT:    [[TMP6:%.*]] = icmp ugt i32 [[TMP5]], 0
+; PRED-NEXT:    [[TMP7:%.*]] = select i1 [[TMP2]], i1 [[TMP6]], i1 false
+; PRED-NEXT:    [[TMP8:%.*]] = or i1 [[TMP7]], [[MUL_OVERFLOW]]
+; PRED-NEXT:    [[TMP9:%.*]] = icmp ugt i64 [[N]], 4294967295
+; PRED-NEXT:    [[TMP10:%.*]] = icmp ne i32 [[MUL_X]], 0
+; PRED-NEXT:    [[TMP11:%.*]] = and i1 [[TMP9]], [[TMP10]]
+; PRED-NEXT:    [[TMP12:%.*]] = or i1 [[TMP8]], [[TMP11]]
+; PRED-NEXT:    br i1 [[TMP12]], label [[SCALAR_PH]], label [[VECTOR_PH:%.*]]
+; PRED:       vector.ph:
+; PRED-NEXT:    [[N_RND_UP:%.*]] = add i64 [[TMP0]], 1
+; PRED-NEXT:    [[N_MOD_VF:%.*]] = urem i64 [[N_RND_UP]], 2
+; PRED-NEXT:    [[N_VEC:%.*]] = sub i64 [[N_RND_UP]], [[N_MOD_VF]]
+; PRED-NEXT:    [[TMP13:%.*]] = sub i64 [[TMP0]], 2
+; PRED-NEXT:    [[TMP14:%.*]] = icmp ugt i64 [[TMP0]], 2
+; PRED-NEXT:    [[TMP15:%.*]] = select i1 [[TMP14]], i64 [[TMP13]], i64 0
+; PRED-NEXT:    [[ACTIVE_LANE_MASK_ENTRY:%.*]] = call <2 x i1> @llvm.get.active.lane.mask.v2i1.i64(i64 0, i64 [[TMP0]])
+; PRED-NEXT:    [[BROADCAST_SPLATINSERT:%.*]] = insertelement <2 x i32> poison, i32 [[MUL_X]], i64 0
+; PRED-NEXT:    [[BROADCAST_SPLAT:%.*]] = shufflevector <2 x i32> [[BROADCAST_SPLATINSERT]], <2 x i32> poison, <2 x i32> zeroinitializer
+; PRED-NEXT:    br label [[VECTOR_BODY:%.*]]
+; PRED:       vector.body:
+; PRED-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[PRED_STORE_CONTINUE2:%.*]] ]
+; PRED-NEXT:    [[ACTIVE_LANE_MASK:%.*]] = phi <2 x i1> [ [[ACTIVE_LANE_MASK_ENTRY]], [[VECTOR_PH]] ], [ [[ACTIVE_LANE_MASK_NEXT:%.*]], [[PRED_STORE_CONTINUE2]] ]
+; PRED-NEXT:    [[VEC_IND:%.*]] = phi <2 x i32> [ <i32 0, i32 1>, [[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], [[PRED_STORE_CONTINUE2]] ]
+; PRED-NEXT:    [[TMP16:%.*]] = mul <2 x i32> [[BROADCAST_SPLAT]], [[VEC_IND]]
+; PRED-NEXT:    [[TMP17:%.*]] = zext <2 x i32> [[TMP16]] to <2 x i64>
+; PRED-NEXT:    [[TMP18:%.*]] = extractelement <2 x i1> [[ACTIVE_LANE_MASK]], i32 0
+; PRED-NEXT:    br i1 [[TMP18]], label [[PRED_STORE_IF:%.*]], label [[PRED_STORE_CONTINUE:%.*]]
+; PRED:       pred.store.if:
+; PRED-NEXT:    [[TMP19:%.*]] = extractelement <2 x i64> [[TMP17]], i32 0
+; PRED-NEXT:    [[TMP20:%.*]] = getelementptr i32, ptr [[DST]], i64 [[TMP19]]
+; PRED-NEXT:    store i32 1, ptr [[TMP20]], align 4
+; PRED-NEXT:    br label [[PRED_STORE_CONTINUE]]
+; PRED:       pred.store.continue:
+; PRED-NEXT:    [[TMP21:%.*]] = extractelement <2 x i1> [[ACTIVE_LANE_MASK]], i32 1
+; PRED-NEXT:    br i1 [[TMP21]], label [[PRED_STORE_IF1:%.*]], label [[PRED_STORE_CONTINUE2]]
+; PRED:       pred.store.if1:
+; PRED-NEXT:    [[TMP22:%.*]] = extractelement <2 x i64> [[TMP17]], i32 1
+; PRED-NEXT:    [[TMP23:%.*]] = getelementptr i32, ptr [[DST]], i64 [[TMP22]]
+; PRED-NEXT:    store i32 1, ptr [[TMP23]], align 4
+; PRED-NEXT:    br label [[PRED_STORE_CONTINUE2]]
+; PRED:       pred.store.continue2:
+; PRED-NEXT:    [[INDEX_NEXT]] = add i64 [[INDEX]], 2
+; PRED-NEXT:    [[ACTIVE_LANE_MASK_NEXT]] = call <2 x i1> @llvm.get.active.lane.mask.v2i1.i64(i64 [[INDEX]], i64 [[TMP15]])
+; PRED-NEXT:    [[TMP24:%.*]] = xor <2 x i1> [[ACTIVE_LANE_MASK_NEXT]], <i1 true, i1 true>
+; PRED-NEXT:    [[VEC_IND_NEXT]] = add <2 x i32> [[VEC_IND]], <i32 2, i32 2>
+; PRED-NEXT:    [[TMP25:%.*]] = extractelement <2 x i1> [[TMP24]], i32 0
+; PRED-NEXT:    br i1 [[TMP25]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP4:![0-9]+]]
+; PRED:       middle.block:
+; PRED-NEXT:    br i1 true, label [[EXIT:%.*]], label [[SCALAR_PH]]
+; PRED:       scalar.ph:
+; PRED-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ [[N_VEC]], [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ], [ 0, [[VECTOR_SCEVCHECK]] ]
+; PRED-NEXT:    br label [[FOR_BODY:%.*]]
+; PRED:       for.body:
+; PRED-NEXT:    [[IV:%.*]] = phi i64 [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ], [ [[IV_NEXT:%.*]], [[FOR_BODY]] ]
+; PRED-NEXT:    [[TRUNC_IV:%.*]] = trunc i64 [[IV]] to i32
+; PRED-NEXT:    [[ADD_I:%.*]] = mul i32 [[MUL_X]], [[TRUNC_IV]]
+; PRED-NEXT:    [[IV_MUL:%.*]] = zext i32 [[ADD_I]] to i64
+; PRED-NEXT:    [[GEP:%.*]] = getelementptr i32, ptr [[DST]], i64 [[IV_MUL]]
+; PRED-NEXT:    store i32 1, ptr [[GEP]], align 4
+; PRED-NEXT:    [[IV_NEXT]] = add i64 [[IV]], 1
+; PRED-NEXT:    [[EC:%.*]] = icmp eq i64 [[IV]], [[N]]
+; PRED-NEXT:    br i1 [[EC]], label [[EXIT]], label [[FOR_BODY]], !llvm.loop [[LOOP5:![0-9]+]]
+; PRED:       exit:
+; PRED-NEXT:    ret void
+;
+entry:
+  %mul.x = add i32 %x, 1
+  br label %for.body
+
+for.body:                                         ; preds = %for.body, %entry
+  %iv = phi i64 [ 0, %entry ], [ %iv.next, %for.body ]
+  %trunc.iv = trunc i64 %iv to i32
+  %add.i = mul i32 %mul.x, %trunc.iv
+  %iv.mul = zext i32 %add.i to i64
+  %gep = getelementptr i32, ptr %dst, i64 %iv.mul
+  store i32 1, ptr %gep, align 4
+  %iv.next = add i64 %iv, 1
+  %ec = icmp eq i64 %iv, %N
+  br i1 %ec, label %exit, label %for.body
+
+exit:
+  ret void
+}
+
+define void @trunc_ivs_and_store(i32 %x, ptr %dst, i64 %N) #0 {
+; DEFAULT-LABEL: define void @trunc_ivs_and_store(
+; DEFAULT-SAME: i32 [[X:%.*]], ptr [[DST:%.*]], i64 [[N:%.*]]) #[[ATTR0]] {
+; DEFAULT-NEXT:  entry:
+; DEFAULT-NEXT:    [[MUL:%.*]] = mul i32 [[X]], [[X]]
+; DEFAULT-NEXT:    [[TMP0:%.*]] = add i64 [[N]], 1
+; DEFAULT-NEXT:    [[MIN_ITERS_CHECK:%.*]] = icmp ult i64 [[TMP0]], 2
+; DEFAULT-NEXT:    br i1 [[MIN_ITERS_CHECK]], label [[SCALAR_PH:%.*]], label [[VECTOR_SCEVCHECK:%.*]]
+; DEFAULT:       vector.scevcheck:
+; DEFAULT-NEXT:    [[TMP1:%.*]] = mul i32 [[X]], [[X]]
+; DEFAULT-NEXT:    [[TMP2:%.*]] = sub i32 0, [[TMP1]]
+; DEFAULT-NEXT:    [[TMP3:%.*]] = icmp slt i32 [[MUL]], 0
+; DEFAULT-NEXT:    [[TMP4:%.*]] = select i1 [[TMP3]], i32 [[TMP2]], i32 [[MUL]]
+; DEFAULT-NEXT:    [[TMP5:%.*]] = trunc i64 [[N]] to i32
+; DEFAULT-NEXT:    [[MUL1:%.*]] = call { i32, i1 } @llvm.umul.with.overflow.i32(i32 [[TMP4]], i32 [[TMP5]])
+; DEFAULT-NEXT:    [[MUL_RESULT:%.*]] = extractvalue { i32, i1 } [[MUL1]], 0
+; DEFAULT-NEXT:    [[MUL_OVERFLOW:%.*]] = extractvalue { i32, i1 } [[MUL1]], 1
+; DEFAULT-NEXT:    [[TMP6:%.*]] = sub i32 0, [[MUL_RESULT]]
+; DEFAULT-NEXT:    [[TMP7:%.*]] = icmp ugt i32 [[TMP6]], 0
+; DEFAULT-NEXT:    [[TMP8:%.*]] = select i1 [[TMP3]], i1 [[TMP7]], i1 false
+; DEFAULT-NEXT:    [[TMP9:%.*]] = or i1 [[TMP8]], [[MUL_OVERFLOW]]
+; DEFAULT-NEXT:    [[TMP10:%.*]] = icmp ugt i64 [[N]], 4294967295
+; DEFAULT-NEXT:    [[TMP11:%.*]] = icmp ne i32 [[MUL]], 0
+; DEFAULT-NEXT:    [[TMP12:%.*]] = and i1 [[TMP10]], [[TMP11]]
+; DEFAULT-NEXT:    [[TMP13:%.*]] = or i1 [[TMP9]], [[TMP12]]
+; DEFAULT-NEXT:    br i1 [[TMP13]], label [[SCALAR_PH]], label [[VECTOR_PH:%.*]]
+; DEFAULT:       vector.ph:
+; DEFAULT-NEXT:    [[N_MOD_VF:%.*]] = urem i64 [[TMP0]], 2
+; DEFAULT-NEXT:    [[N_VEC:%.*]] = sub i64 [[TMP0]], [[N_MOD_VF]]
+; DEFAULT-NEXT:    [[IND_END:%.*]] = trunc i64 [[N_VEC]] to i32
+; DEFAULT-NEXT:    br label [[VECTOR_BODY:%.*]]
+; DEFAULT:       vector.body:
+; DEFAULT-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
+; DEFAULT-NEXT:    [[OFFSET_IDX:%.*]] = trunc i64 [[INDEX]] to i32
+; DEFAULT-NEXT:    [[TMP14:%.*]] = add i32 [[OFFSET_IDX]], 0
+; DEFAULT-NEXT:    [[TMP15:%.*]] = add i32 [[OFFSET_IDX]], 1
+; DEFAULT-NEXT:    [[TMP16:%.*]] = trunc i64 [[INDEX]] to i32
+; DEFAULT-NEXT:    [[TMP17:%.*]] = add i32 [[TMP16]], 0
+; DEFAULT-NEXT:    [[TMP18:%.*]] = add i32 [[TMP16]], 1
+; DEFAULT-NEXT:    [[TMP19:%.*]] = mul i32 [[MUL]], [[TMP17]]
+; DEFAULT-NEXT:    [[TMP20:%.*]] = mul i32 [[MUL]], [[TMP18]]
+; DEFAULT-NEXT:    [[TMP21:%.*]] = zext i32 [[TMP19]] to i64
+; DEFAULT-NEXT:    [[TMP22:%.*]] = zext i32 [[TMP20]] to i64
+; DEFAULT-NEXT:    [[TMP23:%.*]] = getelementptr i32, ptr [[DST]], i64 [[TMP21]]
+; DEFAULT-NEXT:    [[TMP24:%.*]] = getelementptr i32, ptr [[DST]], i64 [[TMP22]]
+; DEFAULT-NEXT:    store i32 [[TMP14]], ptr [[TMP23]], align 4
+; DEFAULT-NEXT:    store i32 [[TMP15]], ptr [[TMP24]], align 4
+; DEFAULT-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 2
+; DEFAULT-NEXT:    [[TMP25:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
+; DEFAULT-NEXT:    br i1 [[TMP25]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP7:![0-9]+]]
+; DEFAULT:       middle.block:
+; DEFAULT-NEXT:    [[CMP_N:%.*]] = icmp eq i64 [[TMP0]], [[N_VEC]]
+; DEFAULT-NEXT:    br i1 [[CMP_N]], label [[EXIT:%.*]], label [[SCALAR_PH]]
+; DEFAULT:       scalar.ph:
+; DEFAULT-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ [[N_VEC]], [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ], [ 0, [[VECTOR_SCEVCHECK]] ]
+; DEFAULT-NEXT:    [[BC_RESUME_VAL2:%.*]] = phi i32 [ [[IND_END]], [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY]] ], [ 0, [[VECTOR_SCEVCHECK]] ]
+; DEFAULT-NEXT:    br label [[LOOP:%.*]]
+; DEFAULT:       loop:
+; DEFAULT-NEXT:    [[IV_1:%.*]] = phi i64 [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ], [ [[IV_1_NEXT:%.*]], [[LOOP]] ]
+; DEFAULT-NEXT:    [[IV_2:%.*]] = phi i32 [ [[BC_RESUME_VAL2]], [[SCALAR_PH]] ], [ [[IV_2_NEXT:%.*]], [[LOOP]] ]
+; DEFAULT-NEXT:    [[IV_1_TRUNC:%.*]] = trunc i64 [[IV_1]] to i32
+; DEFAULT-NEXT:    [[IV_1_MUL:%.*]] = mul i32 [[MUL]], [[IV_1_TRUNC]]
+; DEFAULT-NEXT:    [[IV_2_NEXT]] = add i32 [[IV_2]], 1
+; DEFAULT-NEXT:    [[MUL_EXT:%.*]] = zext i32 [[IV_1_MUL]] to i64
+; DEFAULT-NEXT:    [[GEP:%.*]] = getelementptr i32, ptr [[DST]], i64 [[MUL_EXT]]
+; DEFAULT-NEXT:    store i32 [[IV_2]], ptr [[GEP]], align 4
+; DEFAULT-NEXT:    [[IV_1_NEXT]] = add i64 [[IV_1]], 1
+; DEFAULT-NEXT:    [[EXITCOND_3_NOT:%.*]] = icmp eq i64 [[IV_1]], [[N]]
+; DEFAULT-NEXT:    br i1 [[EXITCOND_3_NOT]], label [[EXIT]], label [[LOOP]], !llvm.loop [[LOOP8:![0-9]+]]
+; DEFAULT:       exit:
+; DEFAULT-NEXT:    ret void
+;
+; PRED-LABEL: define void @trunc_ivs_and_store(
+; PRED-SAME: i32 [[X:%.*]], ptr [[DST:%.*]], i64 [[N:%.*]]) #[[ATTR0]] {
+; PRED-NEXT:  entry:
+; PRED-NEXT:    [[MUL:%.*]] = mul i32 [[X]], [[X]]
+; PRED-NEXT:    [[TMP0:%.*]] = add i64 [[N]], 1
+; PRED-NEXT:    br i1 false, label [[SCALAR_PH:%.*]], label [[VECTOR_SCEVCHECK:%.*]]
+; PRED:       vector.scevcheck:
+; PRED-NEXT:    [[TMP1:%.*]] = mul i32 [[X]], [[X]]
+; PRED-NEXT:    [[TMP2:%.*]] = sub i32 0, [[TMP1]]
+; PRED-NEXT:    [[TMP3:%.*]] = icmp slt i32 [[MUL]], 0
+; PRED-NEXT:    [[TMP4:%.*]] = select i1 [[TMP3]], i32 [[TMP2]], i32 [[MUL]]
+; PRED-NEXT:    [[TMP5:%.*]] = trunc i64 [[N]] to i32
+; PRED-NEXT:    [[MUL1:%.*]] = call { i32, i1 } @llvm.umul.with.overflow.i32(i32 [[TMP4]], i32 [[TMP5]])
+; PRED-NEXT:    [[MUL_RESULT:%.*]] = extractvalue { i32, i1 } [[MUL1]], 0
+; PRED-NEXT:    [[MUL_OVERFLOW:%.*]] = extractvalue { i32, i1 } [[MUL1]], 1
+; PRED-NEXT:    [[TMP6:%.*]] = sub i32 0, [[MUL_RESULT]]
+; PRED-NEXT:    [[TMP7:%.*]] = icmp ugt i32 [[TMP6]], 0
+; PRED-NEXT:    [[TMP8:%.*]] = select i1 [[TMP3]], i1 [[TMP7]], i1 false
+; PRED-NEXT:    [[TMP9:%.*]] = or i1 [[TMP8]], [[MUL_OVERFLOW]]
+; PRED-NEXT:    [[TMP10:%.*]] = icmp ugt i64 [[N]], 4294967295
+; PRED-NEXT:    [[TMP11:%.*]] = icmp ne i32 [[MUL]], 0
+; PRED-NEXT:    [[TMP12:%.*]] = and i1 [[TMP10]], [[TMP11]]
+; PRED-NEXT:    [[TMP13:%.*]] = or i1 [[TMP9]], [[TMP12]]
+; PRED-NEXT:    br i1 [[TMP13]], label [[SCALAR_PH]], label [[VECTOR_PH:%.*]]
+; PRED:       vector.ph:
+; PRED-NEXT:    [[N_RND_UP:%.*]] = add i64 [[TMP0]], 3
+; PRED-NEXT:    [[N_MOD_VF:%.*]] = urem i64 [[N_RND_UP]], 4
+; PRED-NEXT:    [[N_VEC:%.*]] = sub i64 [[N_RND_UP]], [[N_MOD_VF]]
+; PRED-NEXT:    [[IND_END:%.*]] = trunc i64 [[N_VEC]] to i32
+; PRED-NEXT:    [[TMP14:%.*]] = sub i64 [[TMP0]], 4
+; PRED-NEXT:    [[TMP15:%.*]] = icmp ugt i64 [[TMP0]], 4
+; PRED-NEXT:    [[TMP16:%.*]] = select i1 [[TMP15]], i64 [[TMP14]], i64 0
+; PRED-NEXT:    [[ACTIVE_LANE_MASK_ENTRY:%.*]] = call <4 x i1> @llvm.get.active.lane.mask.v4i1.i64(i64 0, i64 [[TMP0]])
+; PRED-NEXT:    [[BROADCAST_SPLATINSERT:%.*]] = insertelement <4 x i32> poison, i32 [[MUL]], i64 0
+; PRED-NEXT:    [[BROADCAST_SPLAT:%.*]] = shufflevector <4 x i32> [[BROADCAST_SPLATINSERT]], <4 x i32> poison, <4 x i32> zeroinitializer
+; PRED-NEXT:    br label [[VECTOR_BODY:%.*]]
+; PRED:       vector.body:
+; PRED-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[PRED_STORE_CONTINUE8:%.*]] ]
+; PRED-NEXT:    [[ACTIVE_LANE_MASK:%.*]] = phi <4 x i1> [ [[ACTIVE_LANE_MASK_ENTRY]], [[VECTOR_PH]] ], [ [[ACTIVE_LANE_MASK_NEXT:%.*]], [[PRED_STORE_CONTINUE8]] ]
+; PRED-NEXT:    [[VEC_IND:%.*]] = phi <4 x i32> [ <i32 0, i32 1, i32 2, i32 3>, [[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], [[PRED_STORE_CONTINUE8]] ]
+; PRED-NEXT:    [[OFFSET_IDX:%.*]] = trunc i64 [[INDEX]] to i32
+; PRED-NEXT:    [[TMP17:%.*]] = mul <4 x i32> [[BROADCAST_SPLAT]], [[VEC_IND]]
+; PRED-NEXT:    [[TMP18:%.*]] = zext <4 x i32> [[TMP17]] to <4 x i64>
+; PRED-NEXT:    [[TMP19:%.*]] = extractelement <4 x i1> [[ACTIVE_LANE_MASK]], i32 0
+; PRED-NEXT:    br i1 [[TMP19]], label [[PRED_STORE_IF:%.*]], label [[PRED_STORE_CONTINUE:%.*]]
+; PRED:       pred.store.if:
+; PRED-NEXT:    [[TMP20:%.*]] = extractelement <4 x i64> [[TMP18]], i32 0
+; PRED-NEXT:    [[TMP21:%.*]] = getelementptr i32, ptr [[DST]], i64 [[TMP20]]
+; PRED-NEXT:    [[TMP22:%.*]] = add i32 [[OFFSET_IDX]], 0
+; PRED-NEXT:    store i32 [[TMP22]], ptr [[TMP21]], align 4
+; PRED-NEXT:    br label [[PRED_STORE_CONTINUE]]
+; PRED:       pred.store.continue:
+; PRED-NEXT:    [[TMP23:%.*]] = extractelement <4 x i1> [[ACTIVE_LANE_MASK]], i32 1
+; PRED-NEXT:    br i1 [[TMP23]], label [[PRED_STORE_IF3:%.*]], label [[PRED_STORE_CONTINUE4:%.*]]
+; PRED:       pred.store.if3:
+; PRED-NEXT:    [[TMP24:%.*]] = extractelement <4 x i64> [[TMP18]], i32 1
+; PRED-NEXT:    [[TMP25:%.*]] = getelementptr i32, ptr [[DST]], i64 [[TMP24]]
+; PRED-NEXT:    [[TMP26:%.*]] = add i32 [[OFFSET_IDX]], 1
+; PRED-NEXT:    store i32 [[TMP26]], ptr [[TMP25]], align 4
+; PRED-NEXT:    br label [[PRED_STORE_CONTINUE4]]
+; PRED:       pred.store.continue4:
+; PRED-NEXT:    [[TMP27:%.*]] = extractelement <4 x i1> [[ACTIVE_LANE_MASK]], i32 2
+; PRED-NEXT:    br i1 [[TMP27]], label [[PRED_STORE_IF5:%.*]], label [[PRED_STORE_CONTINUE6:%.*]]
+; PRED:       pred.store.if5:
+; PRED-NEXT:    [[TMP28:%.*]] = extractelement <4 x i64> [[TMP18]], i32 2
+; PRED-NEXT:    [[TMP29:%.*]] = getelementptr i32, ptr [[DST]], i64 [[TMP28]]
+; PRED-NEXT:    [[TMP30:%.*]] = add i32 [[OFFSET_IDX]], 2
+; PRED-NEXT:    store i32 [[TMP30]], ptr [[TMP29]], align 4
+; PRED-NEXT:    br label [[PRED_STORE_CONTINUE6]]
+; PRED:       pred.store.continue6:
+; PRED-NEXT:    [[TMP31:%.*]] = extractelement <4 x i1> [[ACTIVE_LANE_MASK]], i32 3
+; PRED-NEXT:    br i1 [[TMP31]], label [[PRED_STORE_IF7:%.*]], label [[PRED_STORE_CONTINUE8]]
+; PRED:       pred.store.if7:
+; PRED-NEXT:    [[TMP32:%.*]] = extractelement <4 x i64> [[TMP18]], i32 3
+; PRED-NEXT:    [[TMP33:%.*]] = getelementptr i32, ptr [[DST]], i64 [[TMP32]]
+; PRED-NEXT:    [[TMP34:%.*]] = add i32 [[OFFSET_IDX]], 3
+; PRED-NEXT:    store i32 [[TMP34]], ptr [[TMP33]], align 4
+; PRED-NEXT:    br label [[PRED_STORE_CONTINUE8]]
+; PRED:       pred.store.continue8:
+; PRED-NEXT:    [[INDEX_NEXT]] = add i64 [[INDEX]], 4
+; PRED-NEXT:    [[ACTIVE_LANE_MASK_NEXT]] = call <4 x i1> @llvm.get.active.lane.mask.v4i1.i64(i64 [[INDEX]], i64 [[TMP16]])
+; PRED-NEXT:    [[TMP35:%.*]] = xor <4 x i1> [[ACTIVE_LANE_MASK_NEXT]], <i1 true, i1 true, i1 true, i1 true>
+; PRED-NEXT:    [[VEC_IND_NEXT]] = add <4 x i32> [[VEC_IND]], <i32 4, i32 4, i32 4, i32 4>
+; PRED-NEXT:    [[TMP36:%.*]] = extractelement <4 x i1> [[TMP35]], i32 0
+; PRED-NEXT:    br i1 [[TMP36]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP6:![0-9]+]]
+; PRED:       middle.block:
+; PRED-NEXT:    br i1 true, label [[EXIT:%.*]], label [[SCALAR_PH]]
+; PRED:       scalar.ph:
+; PRED-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ [[N_VEC]], [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ], [ 0, [[VECTOR_SCEVCHECK]] ]
+; PRED-NEXT:    [[BC_RESUME_VAL2:%.*]] = phi i32 [ [[IND_END]], [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY]] ], [ 0, [[VECTOR_SCEVCHECK]] ]
+; PRED-NEXT:    br label [[LOOP:%.*]]
+; PRED:       loop:
+; PRED-NEXT:    [[IV_1:%.*]] = phi i64 [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ], [ [[IV_1_NEXT:%.*]], [[LOOP]] ]
+; PRED-NEXT:    [[IV_2:%.*]] = phi i32 [ [[BC_RESUME_VAL2]], [[SCALAR_PH]] ], [ [[IV_2_NEXT:%.*]], [[LOOP]] ]
+; PRED-NEXT:    [[IV_1_TRUNC:%.*]] = trunc i64 [[IV_1]] to i32
+; PRED-NEXT:    [[IV_1_MUL:%.*]] = mul i32 [[MUL]], [[IV_1_TRUNC]]
+; PRED-NEXT:    [[IV_2_NEXT]] = add i32 [[IV_2]], 1
+; PRED-NEXT:    [[MUL_EXT:%.*]] = zext i32 [[IV_1_MUL]] to i64
+; PRED-NEXT:    [[GEP:%.*]] = getelementptr i32, ptr [[DST]], i64 [[MUL_EXT]]
+; PRED-NEXT:    store i32 [[IV_2]], ptr [[GEP]], align 4
+; PRED-NEXT:    [[IV_1_NEXT]] = add i64 [[IV_1]], 1
+; PRED-NEXT:    [[EXITCOND_3_NOT:%.*]] = icmp eq i64 [[IV_1]], [[N]]
+; PRED-NEXT:    br i1 [[EXITCOND_3_NOT]], label [[EXIT]], label [[LOOP]], !llvm.loop [[LOOP7:![0-9]+]]
+; PRED:       exit:
+; PRED-NEXT:    ret void
+;
+entry:
+  %mul = mul i32 %x, %x
+  br label %loop
+
+loop:
+  %iv.1 = phi i64 [ 0, %entry ], [ %iv.1.next, %loop ]
+  %iv.2 = phi i32 [ 0, %entry ], [ %iv.2.next, %loop ]
+  %iv.1.trunc = trunc i64 %iv.1 to i32
+  %iv.1.mul = mul i32 %mul, %iv.1.trunc
+  %iv.2.next = add i32 %iv.2, 1
+  %mul.ext = zext i32 %iv.1.mul to i64
+  %gep = getelementptr i32, ptr %dst, i64 %mul.ext
+  store i32 %iv.2, ptr %gep, align 4
+  %iv.1.next = add i64 %iv.1, 1
+  %exitcond.3.not = icmp eq i64 %iv.1, %N
+  br i1 %exitcond.3.not, label %exit, label %loop
+
+exit:
+  ret void
+}
+
+define void @ivs_trunc_and_ext(i32 %x, ptr %dst, i64 %N) #0 {
+; DEFAULT-LABEL: define void @ivs_trunc_and_ext(
+; DEFAULT-SAME: i32 [[X:%.*]], ptr [[DST:%.*]], i64 [[N:%.*]]) #[[ATTR0]] {
+; DEFAULT-NEXT:  entry:
+; DEFAULT-NEXT:    [[ADD:%.*]] = add i32 [[X]], 1
+; DEFAULT-NEXT:    [[TMP0:%.*]] = add i64 [[N]], 1
+; DEFAULT-NEXT:    [[MIN_ITERS_CHECK:%.*]] = icmp ult i64 [[TMP0]], 2
+; DEFAULT-NEXT:    br i1 [[MIN_ITERS_CHECK]], label [[SCALAR_PH:%.*]], label [[VECTOR_SCEVCHECK:%.*]]
+; DEFAULT:       vector.scevcheck:
+; DEFAULT-NEXT:    [[TMP1:%.*]] = sub i32 -1, [[X]]
+; DEFAULT-NEXT:    [[TMP2:%.*]] = icmp slt i32 [[ADD]], 0
+; DEFAULT-NEXT:    [[TMP3:%.*]] = select i1 [[TMP2]], i32 [[TMP1]], i32 [[ADD]]
+; DEFAULT-NEXT:    [[TMP4:%.*]] = trunc i64 [[N]] to i32
+; DEFAULT-NEXT:    [[MUL:%.*]] = call { i32, i1 } @llvm.umul.with.overflow.i32(i32 [[TMP3]], i32 [[TMP4]])
+; DEFAULT-NEXT:    [[MUL_RESULT:%.*]] = extractvalue { i32, i1 } [[MUL]], 0
+; DEFAULT-NEXT:    [[MUL_OVERFLOW:%.*]] = extractvalue { i32, i1 } [[MUL]], 1
+; DEFAULT-NEXT:    [[TMP5:%.*]] = sub i32 0, [[MUL_RESULT]]
+; DEFAULT-NEXT:    [[TMP6:%.*]] = icmp ugt i32 [[TMP5]], 0
+; DEFAULT-NEXT:    [[TMP7:%.*]] = select i1 [[TMP2]], i1 [[TMP6]], i1 false
+; DEFAULT-NEXT:    [[TMP8:%.*]] = or i1 [[TMP7]], [[MUL_OVERFLOW]]
+; DEFAULT-NEXT:    [[TMP9:%.*]] = icmp ugt i64 [[N]], 4294967295
+; DEFAULT-NEXT:    [[TMP10:%.*]] = icmp ne i32 [[ADD]], 0
+; DEFAULT-NEXT:    [[TMP11:%.*]] = and i1 [[TMP9]], [[TMP10]]
+; DEFAULT-NEXT:    [[TMP12:%.*]] = or i1 [[TMP8]], [[TMP11]]
+; DEFAULT-NEXT:    br i1 [[TMP12]], label [[SCALAR_PH]], label [[VECTOR_PH:%.*]]
+; DEFAULT:       vector.ph:
+; DEFAULT-NEXT:    [[N_MOD_VF:%.*]] = urem i64 [[TMP0]], 2
+; DEFAULT-NEXT:    [[N_VEC:%.*]] = sub i64 [[TMP0]], [[N_MOD_VF]]
+; DEFAULT-NEXT:    [[IND_END:%.*]] = trunc i64 [[N_VEC]] to i32
+; DEFAULT-NEXT:    br label [[VECTOR_BODY:%.*]]
+; DEFAULT:       vector.body:
+; DEFAULT-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
+; DEFAULT-NEXT:    [[OFFSET_IDX:%.*]] = trunc i64 [[INDEX]] to i32
+; DEFAULT-NEXT:    [[TMP13:%.*]] = add i32 [[OFFSET_IDX]], 0
+; DEFAULT-NEXT:    [[TMP14:%.*]] = add i32 [[OFFSET_IDX]], 1
+; DEFAULT-NEXT:    [[TMP15:%.*]] = trunc i64 [[INDEX]] to i32
+; DEFAULT-NEXT:    [[TMP16:%.*]] = add i32 [[TMP15]], 0
+; DEFAULT-NEXT:    [[TMP17:%.*]] = add i32 [[TMP15]], 1
+; DEFAULT-NEXT:    [[TMP18:%.*]] = mul i32 [[ADD]], [[TMP16]]
+; DEFAULT-NEXT:    [[TMP19:%.*]] = mul i32 [[ADD]], [[TMP17]]
+; DEFAULT-NEXT:    [[TMP20:%.*]] = zext i32 [[TMP18]] to i64
+; DEFAULT-NEXT:    [[TMP21:%.*]] = zext i32 [[TMP19]] to i64
+; DEFAULT-NEXT:    [[TMP22:%.*]] = getelementptr i32, ptr [[DST]], i64 [[TMP20]]
+; DEFAULT-NEXT:    [[TMP23:%.*]] = getelementptr i32, ptr [[DST]], i64 [[TMP21]]
+; DEFAULT-NEXT:    store i32 [[TMP13]], ptr [[TMP22]], align 4
+; DEFAULT-NEXT:    store i32 [[TMP14]], ptr [[TMP23]], align 4
+; DEFAULT-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 2
+; DEFAULT-NEXT:    [[TMP24:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
+; DEFAULT-NEXT:    br i1 [[TMP24]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP9:![0-9]+]]
+; DEFAULT:       middle.block:
+; DEFAULT-NEXT:    [[CMP_N:%.*]] = icmp eq i64 [[TMP0]], [[N_VEC]]
+; DEFAULT-NEXT:    br i1 [[CMP_N]], label [[EXIT:%.*]], label [[SCALAR_PH]]
+; DEFAULT:       scalar.ph:
+; DEFAULT-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ [[N_VEC]], [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ], [ 0, [[VECTOR_SCEVCHECK]] ]
+; DEFAULT-NEXT:    [[BC_RESUME_VAL1:%.*]] = phi i32 [ [[IND_END]], [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY]] ], [ 0, [[VECTOR_SCEVCHECK]] ]
+; DEFAULT-NEXT:    br label [[LOOP:%.*]]
+; DEFAULT:       loop:
+; DEFAULT-NEXT:    [[IV_1:%.*]] = phi i64 [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ], [ [[IV_1_NEXT:%.*]], [[LOOP]] ]
+; DEFAULT-NEXT:    [[IV_2:%.*]] = phi i32 [ [[BC_RESUME_VAL1]], [[SCALAR_PH]] ], [ [[IV_2_NEXT:%.*]], [[LOOP]] ]
+; DEFAULT-NEXT:    [[IV_TRUNC:%.*]] = trunc i64 [[IV_1]] to i32
+; DEFAULT-NEXT:    [[IV_MUL:%.*]] = mul i32 [[ADD]], [[IV_TRUNC]]
+; DEFAULT-NEXT:    [[IV_2_NEXT]] = add i32 [[IV_2]], 1
+; DEFAULT-NEXT:    [[EXT:%.*]] = zext i32 [[IV_MUL]] to i64
+; DEFAULT-NEXT:    [[GEP:%.*]] = getelementptr i32, ptr [[DST]], i64 [[EXT]]
+; DEFAULT-NEXT:    store i32 [[IV_2]], ptr [[GEP]], align 4
+; DEFAULT-NEXT:    [[IV_1_NEXT]] = add i64 [[IV_1]], 1
+; DEFAULT-NEXT:    [[EC:%.*]] = icmp eq i64 [[IV_1]], [[N]]
+; DEFAULT-NEXT:    br i1 [[EC]], label [[EXIT]], label [[LOOP]], !llvm.loop [[LOOP10:![0-9]+]]
+; DEFAULT:       exit:
+; DEFAULT-NEXT:    ret void
+;
+; PRED-LABEL: define void @ivs_trunc_and_ext(
+; PRED-SAME: i32 [[X:%.*]], ptr [[DST:%.*]], i64 [[N:%.*]]) #[[ATTR0]] {
+; PRED-NEXT:  entry:
+; PRED-NEXT:    [[ADD:%.*]] = add i32 [[X]], 1
+; PRED-NEXT:    [[TMP0:%.*]] = add i64 [[N]], 1
+; PRED-NEXT:    br i1 false, label [[SCALAR_PH:%.*]], label [[VECTOR_SCEVCHECK:%.*]]
+; PRED:       vector.scevcheck:
+; PRED-NEXT:    [[TMP1:%.*]] = sub i32 -1, [[X]]
+; PRED-NEXT:    [[TMP2:%.*]] = icmp slt i32 [[ADD]], 0
+; PRED-NEXT:    [[TMP3:%.*]] = select i1 [[TMP2]], i32 [[TMP1]], i32 [[ADD]]
+; PRED-NEXT:    [[TMP4:%.*]] = trunc i64 [[N]] to i32
+; PRED-NEXT:    [[MUL:%.*]] = call { i32, i1 } @llvm.umul.with.overflow.i32(i32 [[TMP3]], i32 [[TMP4]])
+; PRED-NEXT:    [[MUL_RESULT:%.*]] = extractvalue { i32, i1 } [[MUL]], 0
+; PRED-NEXT:    [[MUL_OVERFLOW:%.*]] = extractvalue { i32, i1 } [[MUL]], 1
+; PRED-NEXT:    [[TMP5:%.*]] = sub i32 0, [[MUL_RESULT]]
+; PRED-NEXT:    [[TMP6:%.*]] = icmp ugt i32 [[TMP5]], 0
+; PRED-NEXT:    [[TMP7:%.*]] = select i1 [[TMP2]], i1 [[TMP6]], i1 false
+; PRED-NEXT:    [[TMP8:%.*]] = or i1 [[TMP7]], [[MUL_OVERFLOW]]
+; PRED-NEXT:    [[TMP9:%.*]] = icmp ugt i64 [[N]], 4294967295
+; PRED-NEXT:    [[TMP10:%.*]] = icmp ne i32 [[ADD]], 0
+; PRED-NEXT:    [[TMP11:%.*]] = and i1 [[TMP9]], [[TMP10]]
+; PRED-NEXT:    [[TMP12:%.*]] = or i1 [[TMP8]], [[TMP11]]
+; PRED-NEXT:    br i1 [[TMP12]], label [[SCALAR_PH]], label [[VECTOR_PH:%.*]]
+; PRED:       vector.ph:
+; PRED-NEXT:    [[N_RND_UP:%.*]] = add i64 [[TMP0]], 3
+; PRED-NEXT:    [[N_MOD_VF:%.*]] = urem i64 [[N_RND_UP]], 4
+; PRED-NEXT:    [[N_VEC:%.*]] = sub i64 [[N_RND_UP]], [[N_MOD_VF]]
+; PRED-NEXT:    [[IND_END:%.*]] = trunc i64 [[N_VEC]] to i32
+; PRED-NEXT:    [[TMP13:%.*]] = sub i64 [[TMP0]], 4
+; PRED-NEXT:    [[TMP14:%.*]] = icmp ugt i64 [[TMP0]], 4
+; PRED-NEXT:    [[TMP15:%.*]] = select i1 [[TMP14]], i64 [[TMP13]], i64 0
+; PRED-NEXT:    [[ACTIVE_LANE_MASK_ENTRY:%.*]] = call <4 x i1> @llvm.get.active.lane.mask.v4i1.i64(i64 0, i64 [[TMP0]])
+; PRED-NEXT:    [[BROADCAST_SPLATINSERT:%.*]] = insertelement <4 x i32> poison, i32 [[ADD]], i64 0
+; PRED-NEXT:    [[BROADCAST_SPLAT:%.*]] = shufflevector <4 x i32> [[BROADCAST_SPLATINSERT]], <4 x i32> poison, <4 x i32> zeroinitializer
+; PRED-NEXT:    br label [[VECTOR_BODY:%.*]]
+; PRED:       vector.body:
+; PRED-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[PRED_STORE_CONTINUE7:%.*]] ]
+; PRED-NEXT:    [[ACTIVE_LANE_MASK:%.*]] = phi <4 x i1> [ [[ACTIVE_LANE_MASK_ENTRY]], [[VECTOR_PH]] ], [ [[ACTIVE_LANE_MASK_NEXT:%.*]], [[PRED_STORE_CONTINUE7]] ]
+; PRED-NEXT:    [[VEC_IND:%.*]] = phi <4 x i32> [ <i32 0, i32 1, i32 2, i32 3>, [[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], [[PRED_STORE_CONTINUE7]] ]
+; PRED-NEXT:    [[OFFSET_IDX:%.*]] = trunc i64 [[INDEX]] to i32
+; PRED-NEXT:    [[TMP16:%.*]] = mul <4 x i32> [[BROADCAST_SPLAT]], [[VEC_IND]]
+; PRED-NEXT:    [[TMP17:%.*]] = zext <4 x i32> [[TMP16]] to <4 x i64>
+; PRED-NEXT:    [[TMP18:%.*]] = extractelement <4 x i1> [[ACTIVE_LANE_MASK]], i32 0
+; PRED-NEXT:    br i1 [[TMP18]], label [[PRED_STORE_IF:%.*]], label [[PRED_STORE_CONTINUE:%.*]]
+; PRED:       pred.store.if:
+; PRED-NEXT:    [[TMP19:%.*]] = extractelement <4 x i64> [[TMP17]], i32 0
+; PRED-NEXT:    [[TMP20:%.*]] = getelementptr i32, ptr [[DST]], i64 [[TMP19]]
+; PRED-NEXT:    [[TMP21:%.*]] = add i32 [[OFFSET_IDX]], 0
+; PRED-NEXT:    store i32 [[TMP21]], ptr [[TMP20]], align 4
+; PRED-NEXT:    br label [[PRED_STORE_CONTINUE]]
+; PRED:       pred.store.continue:
+; PRED-NEXT:    [[TMP22:%.*]] = extractelement <4 x i1> [[ACTIVE_LANE_MASK]], i32 1
+; PRED-NEXT:    br i1 [[TMP22]], label [[PRED_STORE_IF2:%.*]], label [[PRED_STORE_CONTINUE3:%.*]]
+; PRED:       pred.store.if2:
+; PRED-NEXT:    [[TMP23:%.*]] = extractelement <4 x i64> [[TMP17]], i32 1
+; PRED-NEXT:    [[TMP24:%.*]] = getelementptr i32, ptr [[DST]], i64 [[TMP23]]
+; PRED-NEXT:    [[TMP25:%.*]] = add i32 [[OFFSET_IDX]], 1
+; PRED-NEXT:    store i32 [[TMP25]], ptr [[TMP24]], align 4
+; PRED-NEXT:    br label [[PRED_STORE_CONTINUE3]]
+; PRED:       pred.store.continue3:
+; PRED-NEXT:    [[TMP26:%.*]] = extractelement <4 x i1> [[ACTIVE_LANE_MASK]], i32 2
+; PRED-NEXT:    br i1 [[TMP26]], label [[PRED_STORE_IF4:%.*]], label [[PRED_STORE_CONTINUE5:%.*]]
+; PRED:       pred.store.if4:
+; PRED-NEXT:    [[TMP27:%.*]] = extractelement <4 x i64> [[TMP17]], i32 2
+; PRED-NEXT:    [[TMP28:%.*]] = getelementptr i32, ptr [[DST]], i64 [[TMP27]]
+; PRED-NEXT:    [[TMP29:%.*]] = add i32 [[OFFSET_IDX]], 2
+; PRED-NEXT:    store i32 [[TMP29]], ptr [[TMP28]], align 4
+; PRED-NEXT:    br label [[PRED_STORE_CONTINUE5]]
+; PRED:       pred.store.continue5:
+; PRED-NEXT:    [[TMP30:%.*]] = extractelement <4 x i1> [[ACTIVE_LANE_MASK]], i32 3
+; PRED-NEXT:    br i1 [[TMP30]], label [[PRED_STORE_IF6:%.*]], label [[PRED_STORE_CONTINUE7]]
+; PRED:       pred.store.if6:
+; PRED-NEXT:    [[TMP31:%.*]] = extractelement <4 x i64> [[TMP17]], i32 3
+; PRED-NEXT:    [[TMP32:%.*]] = getelementptr i32, ptr [[DST]], i64 [[TMP31]]
+; PRED-NEXT:    [[TMP33:%.*]] = add i32 [[OFFSET_IDX]], 3
+; PRED-NEXT:    store i32 [[TMP33]], ptr [[TMP32]], align 4
+; PRED-NEXT:    br label [[PRED_STORE_CONTINUE7]]
+; PRED:       pred.store.continue7:
+; PRED-NEXT:    [[INDEX_NEXT]] = add i64 [[INDEX]], 4
+; PRED-NEXT:    [[ACTIVE_LANE_MASK_NEXT]] = call <4 x i1> @llvm.get.active.lane.mask.v4i1.i64(i64 [[INDEX]], i64 [[TMP15]])
+; PRED-NEXT:    [[TMP34:%.*]] = xor <4 x i1> [[ACTIVE_LANE_MASK_NEXT]], <i1 true, i1 true, i1 true, i1 true>
+; PRED-NEXT:    [[VEC_IND_NEXT]] = add <4 x i32> [[VEC_IND]], <i32 4, i32 4, i32 4, i32 4>
+; PRED-NEXT:    [[TMP35:%.*]] = extractelement <4 x i1> [[TMP34]], i32 0
+; PRED-NEXT:    br i1 [[TMP35]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP8:![0-9]+]]
+; PRED:       middle.block:
+; PRED-NEXT:    br i1 true, label [[EXIT:%.*]], label [[SCALAR_PH]]
+; PRED:       scalar.ph:
+; PRED-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ [[N_VEC]], [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ], [ 0, [[VECTOR_SCEVCHECK]] ]
+; PRED-NEXT:    [[BC_RESUME_VAL1:%.*]] = phi i32 [ [[IND_END]], [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY]] ], [ 0, [[VECTOR_SCEVCHECK]] ]
+; PRED-NEXT:    br label [[LOOP:%.*]]
+; PRED:       loop:
+; PRED-NEXT:    [[IV_1:%.*]] = phi i64 [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ], [ [[IV_1_NEXT:%.*]], [[LOOP]] ]
+; PRED-NEXT:    [[IV_2:%.*]] = phi i32 [ [[BC_RESUME_VAL1]], [[SCALAR_PH]] ], [ [[IV_2_NEXT:%.*]], [[LOOP]] ]
+; PRED-NEXT:    [[IV_TRUNC:%.*]] = trunc i64 [[IV_1]] to i32
+; PRED-NEXT:    [[IV_MUL:%.*]] = mul i32 [[ADD]], [[IV_TRUNC]]
+; PRED-NEXT:    [[IV_2_NEXT]] = add i32 [[IV_2]], 1
+; PRED-NEXT:    [[EXT:%.*]] = zext i32 [[IV_MUL]] to i64
+; PRED-NEXT:    [[GEP:%.*]] = getelementptr i32, ptr [[DST]], i64 [[EXT]]
+; PRED-NEXT:    store i32 [[IV_2]], ptr [[GEP]], align 4
+; PRED-NEXT:    [[IV_1_NEXT]] = add i64 [[IV_1]], 1
+; PRED-NEXT:    [[EC:%.*]] = icmp eq i64 [[IV_1]], [[N]]
+; PRED-NEXT:    br i1 [[EC]], label [[EXIT]], label [[LOOP]], !llvm.loop [[LOOP9:![0-9]+]]
+; PRED:       exit:
+; PRED-NEXT:    ret void
+;
+entry:
+  %add = add i32 %x, 1
+  br label %loop
+
+loop:
+  %iv.1 = phi i64 [ 0, %entry ], [ %iv.1.next, %loop ]
+  %iv.2 = phi i32 [ 0, %entry ], [ %iv.2.next, %loop ]
+  %iv.trunc = trunc i64 %iv.1 to i32
+  %iv.mul = mul i32 %add, %iv.trunc
+  %iv.2.next = add i32 %iv.2, 1
+  %ext = zext i32 %iv.mul to i64
+  %gep = getelementptr i32, ptr %dst, i64 %ext
+  store i32 %iv.2, ptr %gep, align 4
+  %iv.1.next = add i64 %iv.1, 1
+  %ec = icmp eq i64 %iv.1, %N
+  br i1 %ec, label %exit, label %loop
+
+exit:
+  ret void
+}
+
+define void @exit_cond_zext_iv(ptr %dst, i64 %N) {
+; DEFAULT-LABEL: define void @exit_cond_zext_iv(
+; DEFAULT-SAME: ptr [[DST:%.*]], i64 [[N:%.*]]) {
+; DEFAULT-NEXT:  entry:
+; DEFAULT-NEXT:    [[UMAX1:%.*]] = call i64 @llvm.umax.i64(i64 [[N]], i64 1)
+; DEFAULT-NEXT:    [[MIN_ITERS_CHECK:%.*]] = icmp ult i64 [[UMAX1]], 2
+; DEFAULT-NEXT:    br i1 [[MIN_ITERS_CHECK]], label [[SCALAR_PH:%.*]], label [[VECTOR_SCEVCHECK:%.*]]
+; DEFAULT:       vector.scevcheck:
+; DEFAULT-NEXT:    [[UMAX:%.*]] = call i64 @llvm.umax.i64(i64 [[N]], i64 1)
+; DEFAULT-NEXT:    [[TMP0:%.*]] = add i64 [[UMAX]], -1
+; DEFAULT-NEXT:    [[TMP1:%.*]] = icmp ugt i64 [[TMP0]], 4294967295
+; DEFAULT-NEXT:    [[TMP2:%.*]] = trunc i64 [[TMP0]] to i32
+; DEFAULT-NEXT:    [[TMP3:%.*]] = add i32 1, [[TMP2]]
+; DEFAULT-NEXT:    [[TMP4:%.*]] = icmp ult i32 [[TMP3]], 1
+; DEFAULT-NEXT:    [[TMP5:%.*]] = icmp ugt i64 [[TMP0]], 4294967295
+; DEFAULT-NEXT:    [[TMP6:%.*]] = or i1 [[TMP4]], [[TMP5]]
+; DEFAULT-NEXT:    br i1 [[TMP6]], label [[SCALAR_PH]], label [[VECTOR_PH:%.*]]
+; DEFAULT:       vector.ph:
+; DEFAULT-NEXT:    [[N_MOD_VF:%.*]] = urem i64 [[UMAX1]], 2
+; DEFAULT-NEXT:    [[N_VEC:%.*]] = sub i64 [[UMAX1]], [[N_MOD_VF]]
+; DEFAULT-NEXT:    [[IND_END:%.*]] = trunc i64 [[N_VEC]] to i32
+; DEFAULT-NEXT:    br label [[VECTOR_BODY:%.*]]
+; DEFAULT:       vector.body:
+; DEFAULT-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
+; DEFAULT-NEXT:    [[TMP7:%.*]] = add i64 [[INDEX]], 0
+; DEFAULT-NEXT:    [[TMP8:%.*]] = add i64 [[INDEX]], 1
+; DEFAULT-NEXT:    [[TMP9:%.*]] = getelementptr { [100 x i32], i32, i32 }, ptr [[DST]], i64 [[TMP7]], i32 2
+; DEFAULT-NEXT:    [[TMP10:%.*]] = getelementptr { [100 x i32], i32, i32 }, ptr [[DST]], i64 [[TMP8]], i32 2
+; DEFAULT-NEXT:    store i32 0, ptr [[TMP9]], align 8
+; DEFAULT-NEXT:    store i32 0, ptr [[TMP10]], align 8
+; DEFAULT-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 2
+; DEFAULT-NEXT:    [[TMP11:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
+; DEFAULT-NEXT:    br i1 [[TMP11]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP11:![0-9]+]]
+; DEFAULT:       middle.block:
+; DEFAULT-NEXT:    [[CMP_N:%.*]] = icmp eq i64 [[UMAX1]], [[N_VEC]]
+; DEFAULT-NEXT:    br i1 [[CMP_N]], label [[EXIT:%.*]], label [[SCALAR_PH]]
+; DEFAULT:       scalar.ph:
+; DEFAULT-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i32 [ [[IND_END]], [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ], [ 0, [[VECTOR_SCEVCHECK]] ]
+; DEFAULT-NEXT:    [[BC_RESUME_VAL2:%.*]] = phi i64 [ [[N_VEC]], [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY]] ], [ 0, [[VECTOR_SCEVCHECK]] ]
+; DEFAULT-NEXT:    br label [[LOOP:%.*]]
+; DEFAULT:       loop:
+; DEFAULT-NEXT:    [[IV_1:%.*]] = phi i32 [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ], [ [[IV_1_NEXT:%.*]], [[LOOP]] ]
+; DEFAULT-NEXT:    [[IV_CONV:%.*]] = phi i64 [ [[BC_RESUME_VAL2]], [[SCALAR_PH]] ], [ [[IV_EXT:%.*]], [[LOOP]] ]
+; DEFAULT-NEXT:    [[GEP:%.*]] = getelementptr { [100 x i32], i32, i32 }, ptr [[DST]], i64 [[IV_CONV]], i32 2
+; DEFAULT-NEXT:    store i32 0, ptr [[GEP]], align 8
+; DEFAULT-NEXT:    [[IV_1_NEXT]] = add i32 [[IV_1]], 1
+; DEFAULT-NEXT:    [[IV_EXT]] = zext i32 [[IV_1_NEXT]] to i64
+; DEFAULT-NEXT:    [[C:%.*]] = icmp ult i64 [[IV_EXT]], [[N]]
+; DEFAULT-NEXT:    br i1 [[C]], label [[LOOP]], label [[EXIT]], !llvm.loop [[LOOP12:![0-9]+]]
+; DEFAULT:       exit:
+; DEFAULT-NEXT:    ret void
+;
+; PRED-LABEL: define void @exit_cond_zext_iv(
+; PRED-SAME: ptr [[DST:%.*]], i64 [[N:%.*]]) {
+; PRED-NEXT:  entry:
+; PRED-NEXT:    br label [[LOOP:%.*]]
+; PRED:       loop:
+; PRED-NEXT:    [[IV_1:%.*]] = phi i32 [ 0, [[ENTRY:%.*]] ], [ [[IV_1_NEXT:%.*]], [[LOOP]] ]
+; PRED-NEXT:    [[IV_CONV:%.*]] = phi i64 [ 0, [[ENTRY]] ], [ [[IV_EXT:%.*]], [[LOOP]] ]
+; PRED-NEXT:    [[GEP:%.*]] = getelementptr { [100 x i32], i32, i32 }, ptr [[DST]], i64 [[IV_CONV]], i32 2
+; PRED-NEXT:    store i32 0, ptr [[GEP]], align 8
+; PRED-NEXT:    [[IV_1_NEXT]] = add i32 [[IV_1]], 1
+; PRED-NEXT:    [[IV_EXT]] = zext i32 [[IV_1_NEXT]] to i64
+; PRED-NEXT:    [[C:%.*]] = icmp ult i64 [[IV_EXT]], [[N]]
+; PRED-NEXT:    br i1 [[C]], label [[LOOP]], label [[EXIT:%.*]]
+; PRED:       exit:
+; PRED-NEXT:    ret void
+;
+entry:
+  br label %loop
+
+loop:
+  %iv.1 = phi i32 [ 0, %entry ], [ %iv.1.next, %loop ]
+  %iv.conv = phi i64 [ 0, %entry ], [ %iv.ext, %loop ]
+  %gep = getelementptr {[100 x i32], i32, i32}, ptr %dst, i64 %iv.conv, i32 2
+  store i32 0, ptr %gep, align 8
+  %iv.1.next = add i32 %iv.1, 1
+  %iv.ext = zext i32 %iv.1.next to i64
+  %c = icmp ult i64 %iv.ext, %N
+  br i1 %c, label %loop, label %exit
+
+exit:
+  ret void
+}
+
+attributes #0 = { "target-features"="+sve" }
+
+;.
+; DEFAULT: [[LOOP0]] = distinct !{[[LOOP0]], [[META1:![0-9]+]], [[META2:![0-9]+]]}
+; DEFAULT: [[META1]] = !{!"llvm.loop.isvectorized", i32 1}
+; DEFAULT: [[META2]] = !{!"llvm.loop.unroll.runtime.disable"}
+; DEFAULT: [[LOOP3]] = distinct !{[[LOOP3]], [[META1]], [[META2]]}
+; DEFAULT: [[LOOP4]] = distinct !{[[LOOP4]], [[META1]]}
+; DEFAULT: [[LOOP5]] = distinct !{[[LOOP5]], [[META1]], [[META2]]}
+; DEFAULT: [[LOOP6]] = distinct !{[[LOOP6]], [[META1]]}
+; DEFAULT: [[LOOP7]] = distinct !{[[LOOP7]], [[META1]], [[META2]]}
+; DEFAULT: [[LOOP8]] = distinct !{[[LOOP8]], [[META1]]}
+; DEFAULT: [[LOOP9]] = distinct !{[[LOOP9]], [[META1]], [[META2]]}
+; DEFAULT: [[LOOP10]] = distinct !{[[LOOP10]], [[META1]]}
+; DEFAULT: [[LOOP11]] = distinct !{[[LOOP11]], [[META1]], [[META2]]}
+; DEFAULT: [[LOOP12]] = distinct !{[[LOOP12]], [[META1]]}
+;.
+; PRED: [[LOOP0]] = distinct !{[[LOOP0]], [[META1:![0-9]+]], [[META2:![0-9]+]]}
+; PRED: [[META1]] = !{!"llvm.loop.isvectorized", i32 1}
+; PRED: [[META2]] = !{!"llvm.loop.unroll.runtime.disable"}
+; PRED: [[LOOP3]] = distinct !{[[LOOP3]], [[META1]]}
+; PRED: [[LOOP4]] = distinct !{[[LOOP4]], [[META1]], [[META2]]}
+; PRED: [[LOOP5]] = distinct !{[[LOOP5]], [[META1]]}
+; PRED: [[LOOP6]] = distinct !{[[LOOP6]], [[META1]], [[META2]]}
+; PRED: [[LOOP7]] = distinct !{[[LOOP7]], [[META1]]}
+; PRED: [[LOOP8]] = distinct !{[[LOOP8]], [[META1]], [[META2]]}
+; PRED: [[LOOP9]] = distinct !{[[LOOP9]], [[META1]]}
+;.
diff --git a/llvm/test/Transforms/LoopVectorize/AArch64/induction-costs.ll b/llvm/test/Transforms/LoopVectorize/AArch64/induction-costs.ll
index 931ab4f77618..4d9c850abdf3 100644
--- a/llvm/test/Transforms/LoopVectorize/AArch64/induction-costs.ll
+++ b/llvm/test/Transforms/LoopVectorize/AArch64/induction-costs.ll
@@ -9,37 +9,37 @@ define i32 @multi_exit_iv_uniform(i32 %a, i64 %N, ptr %dst) {
 ; CHECK-NEXT:  entry:
 ; CHECK-NEXT:    [[UMIN:%.*]] = call i64 @llvm.umin.i64(i64 [[N]], i64 2147483648)
 ; CHECK-NEXT:    [[TMP0:%.*]] = add nuw nsw i64 [[UMIN]], 1
-; CHECK-NEXT:    [[MIN_ITERS_CHECK:%.*]] = icmp ule i64 [[TMP0]], 4
+; CHECK-NEXT:    [[MIN_ITERS_CHECK:%.*]] = icmp ule i64 [[TMP0]], 8
 ; CHECK-NEXT:    br i1 [[MIN_ITERS_CHECK]], label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
 ; CHECK:       vector.ph:
-; CHECK-NEXT:    [[N_MOD_VF:%.*]] = urem i64 [[TMP0]], 4
+; CHECK-NEXT:    [[N_MOD_VF:%.*]] = urem i64 [[TMP0]], 8
 ; CHECK-NEXT:    [[TMP1:%.*]] = icmp eq i64 [[N_MOD_VF]], 0
-; CHECK-NEXT:    [[TMP2:%.*]] = select i1 [[TMP1]], i64 4, i64 [[N_MOD_VF]]
+; CHECK-NEXT:    [[TMP2:%.*]] = select i1 [[TMP1]], i64 8, i64 [[N_MOD_VF]]
 ; CHECK-NEXT:    [[N_VEC:%.*]] = sub i64 [[TMP0]], [[TMP2]]
-; CHECK-NEXT:    [[BROADCAST_SPLATINSERT:%.*]] = insertelement <2 x i32> poison, i32 [[A]], i64 0
-; CHECK-NEXT:    [[BROADCAST_SPLAT:%.*]] = shufflevector <2 x i32> [[BROADCAST_SPLATINSERT]], <2 x i32> poison, <2 x i32> zeroinitializer
+; CHECK-NEXT:    [[BROADCAST_SPLATINSERT:%.*]] = insertelement <4 x i32> poison, i32 [[A]], i64 0
+; CHECK-NEXT:    [[BROADCAST_SPLAT:%.*]] = shufflevector <4 x i32> [[BROADCAST_SPLATINSERT]], <4 x i32> poison, <4 x i32> zeroinitializer
 ; CHECK-NEXT:    br label [[VECTOR_BODY:%.*]]
 ; CHECK:       vector.body:
 ; CHECK-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
-; CHECK-NEXT:    [[VEC_PHI:%.*]] = phi <2 x i32> [ zeroinitializer, [[VECTOR_PH]] ], [ [[TMP10:%.*]], [[VECTOR_BODY]] ]
-; CHECK-NEXT:    [[VEC_PHI1:%.*]] = phi <2 x i32> [ zeroinitializer, [[VECTOR_PH]] ], [ [[TMP11:%.*]], [[VECTOR_BODY]] ]
+; CHECK-NEXT:    [[VEC_PHI:%.*]] = phi <4 x i32> [ zeroinitializer, [[VECTOR_PH]] ], [ [[TMP10:%.*]], [[VECTOR_BODY]] ]
+; CHECK-NEXT:    [[VEC_PHI1:%.*]] = phi <4 x i32> [ zeroinitializer, [[VECTOR_PH]] ], [ [[TMP11:%.*]], [[VECTOR_BODY]] ]
 ; CHECK-NEXT:    [[TMP3:%.*]] = add i64 [[INDEX]], 0
-; CHECK-NEXT:    [[TMP4:%.*]] = add i64 [[INDEX]], 2
+; CHECK-NEXT:    [[TMP4:%.*]] = add i64 [[INDEX]], 4
 ; CHECK-NEXT:    [[TMP5:%.*]] = getelementptr i64, ptr [[DST]], i64 [[TMP3]]
 ; CHECK-NEXT:    [[TMP6:%.*]] = getelementptr i64, ptr [[DST]], i64 [[TMP4]]
-; CHECK-NEXT:    [[TMP7:%.*]] = zext <2 x i32> [[BROADCAST_SPLAT]] to <2 x i64>
+; CHECK-NEXT:    [[TMP7:%.*]] = zext <4 x i32> [[BROADCAST_SPLAT]] to <4 x i64>
 ; CHECK-NEXT:    [[TMP8:%.*]] = getelementptr i64, ptr [[TMP5]], i32 0
-; CHECK-NEXT:    [[TMP9:%.*]] = getelementptr i64, ptr [[TMP5]], i32 2
-; CHECK-NEXT:    store <2 x i64> [[TMP7]], ptr [[TMP8]], align 8
-; CHECK-NEXT:    store <2 x i64> [[TMP7]], ptr [[TMP9]], align 8
-; CHECK-NEXT:    [[TMP10]] = add <2 x i32> [[VEC_PHI]], <i32 -1, i32 -1>
-; CHECK-NEXT:    [[TMP11]] = add <2 x i32> [[VEC_PHI1]], <i32 -1, i32 -1>
-; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 4
+; CHECK-NEXT:    [[TMP9:%.*]] = getelementptr i64, ptr [[TMP5]], i32 4
+; CHECK-NEXT:    store <4 x i64> [[TMP7]], ptr [[TMP8]], align 8
+; CHECK-NEXT:    store <4 x i64> [[TMP7]], ptr [[TMP9]], align 8
+; CHECK-NEXT:    [[TMP10]] = add <4 x i32> [[VEC_PHI]], <i32 -1, i32 -1, i32 -1, i32 -1>
+; CHECK-NEXT:    [[TMP11]] = add <4 x i32> [[VEC_PHI1]], <i32 -1, i32 -1, i32 -1, i32 -1>
+; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 8
 ; CHECK-NEXT:    [[TMP12:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
 ; CHECK-NEXT:    br i1 [[TMP12]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP0:![0-9]+]]
 ; CHECK:       middle.block:
-; CHECK-NEXT:    [[BIN_RDX:%.*]] = add <2 x i32> [[TMP11]], [[TMP10]]
-; CHECK-NEXT:    [[TMP13:%.*]] = call i32 @llvm.vector.reduce.add.v2i32(<2 x i32> [[BIN_RDX]])
+; CHECK-NEXT:    [[BIN_RDX:%.*]] = add <4 x i32> [[TMP11]], [[TMP10]]
+; CHECK-NEXT:    [[TMP13:%.*]] = call i32 @llvm.vector.reduce.add.v4i32(<4 x i32> [[BIN_RDX]])
 ; CHECK-NEXT:    br label [[SCALAR_PH]]
 ; CHECK:       scalar.ph:
 ; CHECK-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ [[N_VEC]], [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ]
@@ -285,6 +285,264 @@ loop:
 exit:
   ret void
 }
+
+define i64 @test_ptr_ivs_and_widened_ivs(ptr %src, i32 %N) {
+; DEFAULT-LABEL: define i64 @test_ptr_ivs_and_widened_ivs(
+; DEFAULT-SAME: ptr [[SRC:%.*]], i32 [[N:%.*]]) {
+; DEFAULT-NEXT:  entry:
+; DEFAULT-NEXT:    [[TMP0:%.*]] = add i32 [[N]], -1
+; DEFAULT-NEXT:    [[TMP1:%.*]] = zext i32 [[TMP0]] to i64
+; DEFAULT-NEXT:    [[TMP2:%.*]] = add nuw nsw i64 [[TMP1]], 1
+; DEFAULT-NEXT:    [[MIN_ITERS_CHECK:%.*]] = icmp ult i64 [[TMP2]], 8
+; DEFAULT-NEXT:    br i1 [[MIN_ITERS_CHECK]], label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
+; DEFAULT:       vector.ph:
+; DEFAULT-NEXT:    [[N_MOD_VF:%.*]] = urem i64 [[TMP2]], 8
+; DEFAULT-NEXT:    [[N_VEC:%.*]] = sub i64 [[TMP2]], [[N_MOD_VF]]
+; DEFAULT-NEXT:    [[TMP3:%.*]] = mul i64 [[N_VEC]], 4
+; DEFAULT-NEXT:    [[IND_END:%.*]] = getelementptr i8, ptr [[SRC]], i64 [[TMP3]]
+; DEFAULT-NEXT:    [[IND_END1:%.*]] = trunc i64 [[N_VEC]] to i32
+; DEFAULT-NEXT:    [[IND_END3:%.*]] = trunc i64 [[N_VEC]] to i32
+; DEFAULT-NEXT:    br label [[VECTOR_BODY:%.*]]
+; DEFAULT:       vector.body:
+; DEFAULT-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
+; DEFAULT-NEXT:    [[VECTOR_RECUR:%.*]] = phi <4 x i64> [ <i64 poison, i64 poison, i64 poison, i64 0>, [[VECTOR_PH]] ], [ [[TMP15:%.*]], [[VECTOR_BODY]] ]
+; DEFAULT-NEXT:    [[VEC_IND:%.*]] = phi <4 x i32> [ <i32 0, i32 1, i32 2, i32 3>, [[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], [[VECTOR_BODY]] ]
+; DEFAULT-NEXT:    [[STEP_ADD:%.*]] = add <4 x i32> [[VEC_IND]], <i32 4, i32 4, i32 4, i32 4>
+; DEFAULT-NEXT:    [[OFFSET_IDX:%.*]] = mul i64 [[INDEX]], 4
+; DEFAULT-NEXT:    [[TMP4:%.*]] = add i64 [[OFFSET_IDX]], 0
+; DEFAULT-NEXT:    [[TMP5:%.*]] = add i64 [[OFFSET_IDX]], 16
+; DEFAULT-NEXT:    [[NEXT_GEP:%.*]] = getelementptr i8, ptr [[SRC]], i64 [[TMP4]]
+; DEFAULT-NEXT:    [[NEXT_GEP6:%.*]] = getelementptr i8, ptr [[SRC]], i64 [[TMP5]]
+; DEFAULT-NEXT:    [[TMP6:%.*]] = getelementptr i32, ptr [[NEXT_GEP]], i32 0
+; DEFAULT-NEXT:    [[TMP7:%.*]] = getelementptr i32, ptr [[NEXT_GEP]], i32 4
+; DEFAULT-NEXT:    [[WIDE_LOAD:%.*]] = load <4 x i32>, ptr [[TMP6]], align 4
+; DEFAULT-NEXT:    [[WIDE_LOAD7:%.*]] = load <4 x i32>, ptr [[TMP7]], align 4
+; DEFAULT-NEXT:    [[TMP8:%.*]] = xor <4 x i32> [[WIDE_LOAD]], <i32 1, i32 1, i32 1, i32 1>
+; DEFAULT-NEXT:    [[TMP9:%.*]] = xor <4 x i32> [[WIDE_LOAD7]], <i32 1, i32 1, i32 1, i32 1>
+; DEFAULT-NEXT:    [[TMP10:%.*]] = zext <4 x i32> [[TMP8]] to <4 x i64>
+; DEFAULT-NEXT:    [[TMP11:%.*]] = zext <4 x i32> [[TMP9]] to <4 x i64>
+; DEFAULT-NEXT:    [[TMP12:%.*]] = zext <4 x i32> [[VEC_IND]] to <4 x i64>
+; DEFAULT-NEXT:    [[TMP13:%.*]] = zext <4 x i32> [[STEP_ADD]] to <4 x i64>
+; DEFAULT-NEXT:    [[TMP14:%.*]] = shl <4 x i64> [[TMP10]], [[TMP12]]
+; DEFAULT-NEXT:    [[TMP15]] = shl <4 x i64> [[TMP11]], [[TMP13]]
+; DEFAULT-NEXT:    [[TMP16:%.*]] = shufflevector <4 x i64> [[VECTOR_RECUR]], <4 x i64> [[TMP14]], <4 x i32> <i32 3, i32 4, i32 5, i32 6>
+; DEFAULT-NEXT:    [[TMP17:%.*]] = shufflevector <4 x i64> [[TMP14]], <4 x i64> [[TMP15]], <4 x i32> <i32 3, i32 4, i32 5, i32 6>
+; DEFAULT-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 8
+; DEFAULT-NEXT:    [[VEC_IND_NEXT]] = add <4 x i32> [[STEP_ADD]], <i32 4, i32 4, i32 4, i32 4>
+; DEFAULT-NEXT:    [[TMP18:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
+; DEFAULT-NEXT:    br i1 [[TMP18]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP6:![0-9]+]]
+; DEFAULT:       middle.block:
+; DEFAULT-NEXT:    [[CMP_N:%.*]] = icmp eq i64 [[TMP2]], [[N_VEC]]
+; DEFAULT-NEXT:    [[VECTOR_RECUR_EXTRACT:%.*]] = extractelement <4 x i64> [[TMP15]], i32 3
+; DEFAULT-NEXT:    [[VECTOR_RECUR_EXTRACT_FOR_PHI:%.*]] = extractelement <4 x i64> [[TMP15]], i32 2
+; DEFAULT-NEXT:    br i1 [[CMP_N]], label [[EXIT:%.*]], label [[SCALAR_PH]]
+; DEFAULT:       scalar.ph:
+; DEFAULT-NEXT:    [[SCALAR_RECUR_INIT:%.*]] = phi i64 [ 0, [[ENTRY:%.*]] ], [ [[VECTOR_RECUR_EXTRACT]], [[MIDDLE_BLOCK]] ]
+; DEFAULT-NEXT:    [[BC_RESUME_VAL:%.*]] = phi ptr [ [[IND_END]], [[MIDDLE_BLOCK]] ], [ [[SRC]], [[ENTRY]] ]
+; DEFAULT-NEXT:    [[BC_RESUME_VAL2:%.*]] = phi i32 [ [[IND_END1]], [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY]] ]
+; DEFAULT-NEXT:    [[BC_RESUME_VAL4:%.*]] = phi i32 [ [[IND_END3]], [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY]] ]
+; DEFAULT-NEXT:    br label [[LOOP:%.*]]
+; DEFAULT:       loop:
+; DEFAULT-NEXT:    [[SCALAR_RECUR:%.*]] = phi i64 [ [[SCALAR_RECUR_INIT]], [[SCALAR_PH]] ], [ [[SHL:%.*]], [[LOOP]] ]
+; DEFAULT-NEXT:    [[PTR_IV:%.*]] = phi ptr [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ], [ [[PTR_IV_NEXT:%.*]], [[LOOP]] ]
+; DEFAULT-NEXT:    [[IV_1:%.*]] = phi i32 [ [[BC_RESUME_VAL2]], [[SCALAR_PH]] ], [ [[IV_1_NEXT:%.*]], [[LOOP]] ]
+; DEFAULT-NEXT:    [[IV_2:%.*]] = phi i32 [ [[BC_RESUME_VAL4]], [[SCALAR_PH]] ], [ [[IV_2_NEXT:%.*]], [[LOOP]] ]
+; DEFAULT-NEXT:    [[L:%.*]] = load i32, ptr [[PTR_IV]], align 4
+; DEFAULT-NEXT:    [[NOT:%.*]] = xor i32 [[L]], 1
+; DEFAULT-NEXT:    [[NOT_EXT:%.*]] = zext i32 [[NOT]] to i64
+; DEFAULT-NEXT:    [[IV_EXT:%.*]] = zext i32 [[IV_1]] to i64
+; DEFAULT-NEXT:    [[SHL]] = shl i64 [[NOT_EXT]], [[IV_EXT]]
+; DEFAULT-NEXT:    [[PTR_IV_NEXT]] = getelementptr i8, ptr [[PTR_IV]], i64 4
+; DEFAULT-NEXT:    [[IV_1_NEXT]] = add i32 [[IV_1]], 1
+; DEFAULT-NEXT:    [[IV_2_NEXT]] = add i32 [[IV_2]], 1
+; DEFAULT-NEXT:    [[EC:%.*]] = icmp eq i32 [[IV_2_NEXT]], [[N]]
+; DEFAULT-NEXT:    br i1 [[EC]], label [[EXIT]], label [[LOOP]], !llvm.loop [[LOOP7:![0-9]+]]
+; DEFAULT:       exit:
+; DEFAULT-NEXT:    [[P_LCSSA:%.*]] = phi i64 [ [[SCALAR_RECUR]], [[LOOP]] ], [ [[VECTOR_RECUR_EXTRACT_FOR_PHI]], [[MIDDLE_BLOCK]] ]
+; DEFAULT-NEXT:    ret i64 [[P_LCSSA]]
+;
+; PRED-LABEL: define i64 @test_ptr_ivs_and_widened_ivs(
+; PRED-SAME: ptr [[SRC:%.*]], i32 [[N:%.*]]) {
+; PRED-NEXT:  entry:
+; PRED-NEXT:    [[TMP0:%.*]] = add i32 [[N]], -1
+; PRED-NEXT:    [[TMP1:%.*]] = zext i32 [[TMP0]] to i64
+; PRED-NEXT:    [[TMP2:%.*]] = add nuw nsw i64 [[TMP1]], 1
+; PRED-NEXT:    [[MIN_ITERS_CHECK:%.*]] = icmp ult i64 [[TMP2]], 8
+; PRED-NEXT:    br i1 [[MIN_ITERS_CHECK]], label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
+; PRED:       vector.ph:
+; PRED-NEXT:    [[N_MOD_VF:%.*]] = urem i64 [[TMP2]], 8
+; PRED-NEXT:    [[N_VEC:%.*]] = sub i64 [[TMP2]], [[N_MOD_VF]]
+; PRED-NEXT:    [[TMP3:%.*]] = mul i64 [[N_VEC]], 4
+; PRED-NEXT:    [[IND_END:%.*]] = getelementptr i8, ptr [[SRC]], i64 [[TMP3]]
+; PRED-NEXT:    [[IND_END1:%.*]] = trunc i64 [[N_VEC]] to i32
+; PRED-NEXT:    [[IND_END3:%.*]] = trunc i64 [[N_VEC]] to i32
+; PRED-NEXT:    br label [[VECTOR_BODY:%.*]]
+; PRED:       vector.body:
+; PRED-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
+; PRED-NEXT:    [[VECTOR_RECUR:%.*]] = phi <4 x i64> [ <i64 poison, i64 poison, i64 poison, i64 0>, [[VECTOR_PH]] ], [ [[TMP15:%.*]], [[VECTOR_BODY]] ]
+; PRED-NEXT:    [[VEC_IND:%.*]] = phi <4 x i32> [ <i32 0, i32 1, i32 2, i32 3>, [[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], [[VECTOR_BODY]] ]
+; PRED-NEXT:    [[STEP_ADD:%.*]] = add <4 x i32> [[VEC_IND]], <i32 4, i32 4, i32 4, i32 4>
+; PRED-NEXT:    [[OFFSET_IDX:%.*]] = mul i64 [[INDEX]], 4
+; PRED-NEXT:    [[TMP4:%.*]] = add i64 [[OFFSET_IDX]], 0
+; PRED-NEXT:    [[TMP5:%.*]] = add i64 [[OFFSET_IDX]], 16
+; PRED-NEXT:    [[NEXT_GEP:%.*]] = getelementptr i8, ptr [[SRC]], i64 [[TMP4]]
+; PRED-NEXT:    [[NEXT_GEP6:%.*]] = getelementptr i8, ptr [[SRC]], i64 [[TMP5]]
+; PRED-NEXT:    [[TMP6:%.*]] = getelementptr i32, ptr [[NEXT_GEP]], i32 0
+; PRED-NEXT:    [[TMP7:%.*]] = getelementptr i32, ptr [[NEXT_GEP]], i32 4
+; PRED-NEXT:    [[WIDE_LOAD:%.*]] = load <4 x i32>, ptr [[TMP6]], align 4
+; PRED-NEXT:    [[WIDE_LOAD7:%.*]] = load <4 x i32>, ptr [[TMP7]], align 4
+; PRED-NEXT:    [[TMP8:%.*]] = xor <4 x i32> [[WIDE_LOAD]], <i32 1, i32 1, i32 1, i32 1>
+; PRED-NEXT:    [[TMP9:%.*]] = xor <4 x i32> [[WIDE_LOAD7]], <i32 1, i32 1, i32 1, i32 1>
+; PRED-NEXT:    [[TMP10:%.*]] = zext <4 x i32> [[TMP8]] to <4 x i64>
+; PRED-NEXT:    [[TMP11:%.*]] = zext <4 x i32> [[TMP9]] to <4 x i64>
+; PRED-NEXT:    [[TMP12:%.*]] = zext <4 x i32> [[VEC_IND]] to <4 x i64>
+; PRED-NEXT:    [[TMP13:%.*]] = zext <4 x i32> [[STEP_ADD]] to <4 x i64>
+; PRED-NEXT:    [[TMP14:%.*]] = shl <4 x i64> [[TMP10]], [[TMP12]]
+; PRED-NEXT:    [[TMP15]] = shl <4 x i64> [[TMP11]], [[TMP13]]
+; PRED-NEXT:    [[TMP16:%.*]] = shufflevector <4 x i64> [[VECTOR_RECUR]], <4 x i64> [[TMP14]], <4 x i32> <i32 3, i32 4, i32 5, i32 6>
+; PRED-NEXT:    [[TMP17:%.*]] = shufflevector <4 x i64> [[TMP14]], <4 x i64> [[TMP15]], <4 x i32> <i32 3, i32 4, i32 5, i32 6>
+; PRED-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 8
+; PRED-NEXT:    [[VEC_IND_NEXT]] = add <4 x i32> [[STEP_ADD]], <i32 4, i32 4, i32 4, i32 4>
+; PRED-NEXT:    [[TMP18:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
+; PRED-NEXT:    br i1 [[TMP18]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP4:![0-9]+]]
+; PRED:       middle.block:
+; PRED-NEXT:    [[CMP_N:%.*]] = icmp eq i64 [[TMP2]], [[N_VEC]]
+; PRED-NEXT:    [[VECTOR_RECUR_EXTRACT:%.*]] = extractelement <4 x i64> [[TMP15]], i32 3
+; PRED-NEXT:    [[VECTOR_RECUR_EXTRACT_FOR_PHI:%.*]] = extractelement <4 x i64> [[TMP15]], i32 2
+; PRED-NEXT:    br i1 [[CMP_N]], label [[EXIT:%.*]], label [[SCALAR_PH]]
+; PRED:       scalar.ph:
+; PRED-NEXT:    [[SCALAR_RECUR_INIT:%.*]] = phi i64 [ 0, [[ENTRY:%.*]] ], [ [[VECTOR_RECUR_EXTRACT]], [[MIDDLE_BLOCK]] ]
+; PRED-NEXT:    [[BC_RESUME_VAL:%.*]] = phi ptr [ [[IND_END]], [[MIDDLE_BLOCK]] ], [ [[SRC]], [[ENTRY]] ]
+; PRED-NEXT:    [[BC_RESUME_VAL2:%.*]] = phi i32 [ [[IND_END1]], [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY]] ]
+; PRED-NEXT:    [[BC_RESUME_VAL4:%.*]] = phi i32 [ [[IND_END3]], [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY]] ]
+; PRED-NEXT:    br label [[LOOP:%.*]]
+; PRED:       loop:
+; PRED-NEXT:    [[SCALAR_RECUR:%.*]] = phi i64 [ [[SCALAR_RECUR_INIT]], [[SCALAR_PH]] ], [ [[SHL:%.*]], [[LOOP]] ]
+; PRED-NEXT:    [[PTR_IV:%.*]] = phi ptr [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ], [ [[PTR_IV_NEXT:%.*]], [[LOOP]] ]
+; PRED-NEXT:    [[IV_1:%.*]] = phi i32 [ [[BC_RESUME_VAL2]], [[SCALAR_PH]] ], [ [[IV_1_NEXT:%.*]], [[LOOP]] ]
+; PRED-NEXT:    [[IV_2:%.*]] = phi i32 [ [[BC_RESUME_VAL4]], [[SCALAR_PH]] ], [ [[IV_2_NEXT:%.*]], [[LOOP]] ]
+; PRED-NEXT:    [[L:%.*]] = load i32, ptr [[PTR_IV]], align 4
+; PRED-NEXT:    [[NOT:%.*]] = xor i32 [[L]], 1
+; PRED-NEXT:    [[NOT_EXT:%.*]] = zext i32 [[NOT]] to i64
+; PRED-NEXT:    [[IV_EXT:%.*]] = zext i32 [[IV_1]] to i64
+; PRED-NEXT:    [[SHL]] = shl i64 [[NOT_EXT]], [[IV_EXT]]
+; PRED-NEXT:    [[PTR_IV_NEXT]] = getelementptr i8, ptr [[PTR_IV]], i64 4
+; PRED-NEXT:    [[IV_1_NEXT]] = add i32 [[IV_1]], 1
+; PRED-NEXT:    [[IV_2_NEXT]] = add i32 [[IV_2]], 1
+; PRED-NEXT:    [[EC:%.*]] = icmp eq i32 [[IV_2_NEXT]], [[N]]
+; PRED-NEXT:    br i1 [[EC]], label [[EXIT]], label [[LOOP]], !llvm.loop [[LOOP5:![0-9]+]]
+; PRED:       exit:
+; PRED-NEXT:    [[P_LCSSA:%.*]] = phi i64 [ [[SCALAR_RECUR]], [[LOOP]] ], [ [[VECTOR_RECUR_EXTRACT_FOR_PHI]], [[MIDDLE_BLOCK]] ]
+; PRED-NEXT:    ret i64 [[P_LCSSA]]
+;
+entry:
+  br label %loop
+
+loop:
+  %p = phi i64 [ 0, %entry ], [ %shl, %loop ]
+  %ptr.iv = phi ptr [ %src, %entry ], [ %ptr.iv.next, %loop ]
+  %iv.1 = phi i32 [ 0, %entry ], [ %iv.1.next, %loop ]
+  %iv.2 = phi i32 [ 0, %entry ], [ %iv.2.next, %loop ]
+  %l = load i32, ptr %ptr.iv, align 4
+  %not = xor i32 %l, 1
+  %not.ext = zext i32 %not to i64
+  %iv.ext = zext i32 %iv.1 to i64
+  %shl = shl i64 %not.ext , %iv.ext
+  %ptr.iv.next = getelementptr i8, ptr %ptr.iv, i64 4
+  %iv.1.next = add i32 %iv.1, 1
+  %iv.2.next = add i32 %iv.2, 1
+  %ec = icmp eq i32 %iv.2.next, %N
+  br i1 %ec, label %exit, label %loop
+
+exit:
+  ret i64 %p
+}
+
+define void @zext_iv_increment(ptr %dst, i64 %N) {
+; DEFAULT-LABEL: define void @zext_iv_increment(
+; DEFAULT-SAME: ptr [[DST:%.*]], i64 [[N:%.*]]) {
+; DEFAULT-NEXT:  entry:
+; DEFAULT-NEXT:    [[UMAX1:%.*]] = call i64 @llvm.umax.i64(i64 [[N]], i64 1)
+; DEFAULT-NEXT:    [[MIN_ITERS_CHECK:%.*]] = icmp ult i64 [[UMAX1]], 2
+; DEFAULT-NEXT:    br i1 [[MIN_ITERS_CHECK]], label [[SCALAR_PH:%.*]], label [[VECTOR_SCEVCHECK:%.*]]
+; DEFAULT:       vector.scevcheck:
+; DEFAULT-NEXT:    [[UMAX:%.*]] = call i64 @llvm.umax.i64(i64 [[N]], i64 1)
+; DEFAULT-NEXT:    [[TMP0:%.*]] = add i64 [[UMAX]], -1
+; DEFAULT-NEXT:    [[TMP1:%.*]] = icmp ugt i64 [[TMP0]], 4294967295
+; DEFAULT-NEXT:    [[TMP2:%.*]] = trunc i64 [[TMP0]] to i32
+; DEFAULT-NEXT:    [[TMP3:%.*]] = add i32 1, [[TMP2]]
+; DEFAULT-NEXT:    [[TMP4:%.*]] = icmp ult i32 [[TMP3]], 1
+; DEFAULT-NEXT:    [[TMP5:%.*]] = icmp ugt i64 [[TMP0]], 4294967295
+; DEFAULT-NEXT:    [[TMP6:%.*]] = or i1 [[TMP4]], [[TMP5]]
+; DEFAULT-NEXT:    br i1 [[TMP6]], label [[SCALAR_PH]], label [[VECTOR_PH:%.*]]
+; DEFAULT:       vector.ph:
+; DEFAULT-NEXT:    [[N_MOD_VF:%.*]] = urem i64 [[UMAX1]], 2
+; DEFAULT-NEXT:    [[N_VEC:%.*]] = sub i64 [[UMAX1]], [[N_MOD_VF]]
+; DEFAULT-NEXT:    [[IND_END:%.*]] = trunc i64 [[N_VEC]] to i32
+; DEFAULT-NEXT:    br label [[VECTOR_BODY:%.*]]
+; DEFAULT:       vector.body:
+; DEFAULT-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
+; DEFAULT-NEXT:    [[TMP7:%.*]] = add i64 [[INDEX]], 0
+; DEFAULT-NEXT:    [[TMP8:%.*]] = add i64 [[INDEX]], 1
+; DEFAULT-NEXT:    [[TMP9:%.*]] = getelementptr { i32, i32, i32 }, ptr [[DST]], i64 [[TMP7]], i32 2
+; DEFAULT-NEXT:    [[TMP10:%.*]] = getelementptr { i32, i32, i32 }, ptr [[DST]], i64 [[TMP8]], i32 2
+; DEFAULT-NEXT:    store i32 0, ptr [[TMP9]], align 8
+; DEFAULT-NEXT:    store i32 0, ptr [[TMP10]], align 8
+; DEFAULT-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 2
+; DEFAULT-NEXT:    [[TMP11:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
+; DEFAULT-NEXT:    br i1 [[TMP11]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP8:![0-9]+]]
+; DEFAULT:       middle.block:
+; DEFAULT-NEXT:    [[CMP_N:%.*]] = icmp eq i64 [[UMAX1]], [[N_VEC]]
+; DEFAULT-NEXT:    br i1 [[CMP_N]], label [[FOR_COND_CLEANUP173_LOOPEXIT:%.*]], label [[SCALAR_PH]]
+; DEFAULT:       scalar.ph:
+; DEFAULT-NEXT:    [[BC_RESUME_VAL2:%.*]] = phi i32 [ [[IND_END]], [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ], [ 0, [[VECTOR_SCEVCHECK]] ]
+; DEFAULT-NEXT:    [[BC_RESUME_VAL3:%.*]] = phi i64 [ [[N_VEC]], [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY]] ], [ 0, [[VECTOR_SCEVCHECK]] ]
+; DEFAULT-NEXT:    br label [[FOR_BODY174:%.*]]
+; DEFAULT:       loop:
+; DEFAULT-NEXT:    [[I167_0800:%.*]] = phi i32 [ [[BC_RESUME_VAL2]], [[SCALAR_PH]] ], [ [[INC179:%.*]], [[FOR_BODY174]] ]
+; DEFAULT-NEXT:    [[CONV169801:%.*]] = phi i64 [ [[BC_RESUME_VAL3]], [[SCALAR_PH]] ], [ [[CONV169:%.*]], [[FOR_BODY174]] ]
+; DEFAULT-NEXT:    [[PATCH_INDEX:%.*]] = getelementptr { i32, i32, i32 }, ptr [[DST]], i64 [[CONV169801]], i32 2
+; DEFAULT-NEXT:    store i32 0, ptr [[PATCH_INDEX]], align 8
+; DEFAULT-NEXT:    [[INC179]] = add i32 [[I167_0800]], 1
+; DEFAULT-NEXT:    [[CONV169]] = zext i32 [[INC179]] to i64
+; DEFAULT-NEXT:    [[CMP172:%.*]] = icmp ult i64 [[CONV169]], [[N]]
+; DEFAULT-NEXT:    br i1 [[CMP172]], label [[FOR_BODY174]], label [[FOR_COND_CLEANUP173_LOOPEXIT]], !llvm.loop [[LOOP9:![0-9]+]]
+; DEFAULT:       exit:
+; DEFAULT-NEXT:    ret void
+;
+; PRED-LABEL: define void @zext_iv_increment(
+; PRED-SAME: ptr [[DST:%.*]], i64 [[N:%.*]]) {
+; PRED-NEXT:  entry:
+; PRED-NEXT:    br label [[FOR_BODY174:%.*]]
+; PRED:       loop:
+; PRED-NEXT:    [[I167_0800:%.*]] = phi i32 [ 0, [[ENTRY:%.*]] ], [ [[INC179:%.*]], [[FOR_BODY174]] ]
+; PRED-NEXT:    [[CONV169801:%.*]] = phi i64 [ 0, [[ENTRY]] ], [ [[CONV169:%.*]], [[FOR_BODY174]] ]
+; PRED-NEXT:    [[PATCH_INDEX:%.*]] = getelementptr { i32, i32, i32 }, ptr [[DST]], i64 [[CONV169801]], i32 2
+; PRED-NEXT:    store i32 0, ptr [[PATCH_INDEX]], align 8
+; PRED-NEXT:    [[INC179]] = add i32 [[I167_0800]], 1
+; PRED-NEXT:    [[CONV169]] = zext i32 [[INC179]] to i64
+; PRED-NEXT:    [[CMP172:%.*]] = icmp ult i64 [[CONV169]], [[N]]
+; PRED-NEXT:    br i1 [[CMP172]], label [[FOR_BODY174]], label [[FOR_COND_CLEANUP173_LOOPEXIT:%.*]]
+; PRED:       exit:
+; PRED-NEXT:    ret void
+;
+entry:
+  br label %loop
+
+loop:
+  %iv = phi i32 [ 0, %entry ], [ %iv.next, %loop ]
+  %iv.wide = phi i64 [ 0, %entry ], [ %iv.next.ext, %loop ]
+  %patch_index = getelementptr { i32, i32, i32 }, ptr %dst, i64 %iv.wide, i32 2
+  store i32 0, ptr %patch_index, align 8
+  %iv.next = add i32 %iv, 1
+  %iv.next.ext = zext i32 %iv.next to i64
+  %ec = icmp ult i64 %iv.next.ext, %N
+  br i1 %ec, label %loop, label %exit
+
+exit:
+  ret void
+}
+
 ;.
 ; CHECK: [[LOOP0]] = distinct !{[[LOOP0]], [[META1:![0-9]+]], [[META2:![0-9]+]]}
 ; CHECK: [[META1]] = !{!"llvm.loop.isvectorized", i32 1}
diff --git a/llvm/test/Transforms/LoopVectorize/AArch64/reduction-recurrence-costs-sve.ll b/llvm/test/Transforms/LoopVectorize/AArch64/reduction-recurrence-costs-sve.ll
new file mode 100644
index 000000000000..c24c1a38177d
--- /dev/null
+++ b/llvm/test/Transforms/LoopVectorize/AArch64/reduction-recurrence-costs-sve.ll
@@ -0,0 +1,482 @@
+; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --version 4
+; RUN: opt -p loop-vectorize -S %s | FileCheck --check-prefixes=DEFAULT %s
+; RUN: opt -p loop-vectorize -prefer-predicate-over-epilogue=predicate-else-scalar-epilogue -S %s | FileCheck --check-prefixes=PRED %s
+
+target datalayout = "e-m:o-i64:64-i128:128-n32:64-S128"
+target triple = "arm64-apple-macosx14.0.0"
+
+define i32 @chained_recurrences(i32 %x, i64 %y, ptr %src.1, i32 %z, ptr %src.2) #0 {
+; DEFAULT-LABEL: define i32 @chained_recurrences(
+; DEFAULT-SAME: i32 [[X:%.*]], i64 [[Y:%.*]], ptr [[SRC_1:%.*]], i32 [[Z:%.*]], ptr [[SRC_2:%.*]]) #[[ATTR0:[0-9]+]] {
+; DEFAULT-NEXT:  entry:
+; DEFAULT-NEXT:    [[TMP0:%.*]] = add i64 [[Y]], 1
+; DEFAULT-NEXT:    [[TMP1:%.*]] = call i64 @llvm.vscale.i64()
+; DEFAULT-NEXT:    [[TMP2:%.*]] = mul i64 [[TMP1]], 8
+; DEFAULT-NEXT:    [[MIN_ITERS_CHECK:%.*]] = icmp ult i64 [[TMP0]], [[TMP2]]
+; DEFAULT-NEXT:    br i1 [[MIN_ITERS_CHECK]], label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
+; DEFAULT:       vector.ph:
+; DEFAULT-NEXT:    [[TMP3:%.*]] = call i64 @llvm.vscale.i64()
+; DEFAULT-NEXT:    [[TMP4:%.*]] = mul i64 [[TMP3]], 8
+; DEFAULT-NEXT:    [[N_MOD_VF:%.*]] = urem i64 [[TMP0]], [[TMP4]]
+; DEFAULT-NEXT:    [[N_VEC:%.*]] = sub i64 [[TMP0]], [[N_MOD_VF]]
+; DEFAULT-NEXT:    [[TMP5:%.*]] = call i64 @llvm.vscale.i64()
+; DEFAULT-NEXT:    [[TMP6:%.*]] = mul i64 [[TMP5]], 8
+; DEFAULT-NEXT:    [[TMP7:%.*]] = call i32 @llvm.vscale.i32()
+; DEFAULT-NEXT:    [[TMP8:%.*]] = mul i32 [[TMP7]], 4
+; DEFAULT-NEXT:    [[TMP9:%.*]] = sub i32 [[TMP8]], 1
+; DEFAULT-NEXT:    [[VECTOR_RECUR_INIT:%.*]] = insertelement <vscale x 4 x i32> poison, i32 0, i32 [[TMP9]]
+; DEFAULT-NEXT:    [[TMP10:%.*]] = call i32 @llvm.vscale.i32()
+; DEFAULT-NEXT:    [[TMP11:%.*]] = mul i32 [[TMP10]], 4
+; DEFAULT-NEXT:    [[TMP12:%.*]] = sub i32 [[TMP11]], 1
+; DEFAULT-NEXT:    [[VECTOR_RECUR_INIT1:%.*]] = insertelement <vscale x 4 x i32> poison, i32 0, i32 [[TMP12]]
+; DEFAULT-NEXT:    [[BROADCAST_SPLATINSERT6:%.*]] = insertelement <vscale x 4 x i32> poison, i32 [[X]], i64 0
+; DEFAULT-NEXT:    [[BROADCAST_SPLAT7:%.*]] = shufflevector <vscale x 4 x i32> [[BROADCAST_SPLATINSERT6]], <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer
+; DEFAULT-NEXT:    [[BROADCAST_SPLATINSERT8:%.*]] = insertelement <vscale x 4 x i32> poison, i32 [[Z]], i64 0
+; DEFAULT-NEXT:    [[BROADCAST_SPLAT9:%.*]] = shufflevector <vscale x 4 x i32> [[BROADCAST_SPLATINSERT8]], <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer
+; DEFAULT-NEXT:    br label [[VECTOR_BODY:%.*]]
+; DEFAULT:       vector.body:
+; DEFAULT-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
+; DEFAULT-NEXT:    [[VECTOR_RECUR:%.*]] = phi <vscale x 4 x i32> [ [[VECTOR_RECUR_INIT]], [[VECTOR_PH]] ], [ [[BROADCAST_SPLAT5:%.*]], [[VECTOR_BODY]] ]
+; DEFAULT-NEXT:    [[VECTOR_RECUR2:%.*]] = phi <vscale x 4 x i32> [ [[VECTOR_RECUR_INIT1]], [[VECTOR_PH]] ], [ [[TMP20:%.*]], [[VECTOR_BODY]] ]
+; DEFAULT-NEXT:    [[VEC_PHI:%.*]] = phi <vscale x 4 x i32> [ zeroinitializer, [[VECTOR_PH]] ], [ [[TMP57:%.*]], [[VECTOR_BODY]] ]
+; DEFAULT-NEXT:    [[VEC_PHI3:%.*]] = phi <vscale x 4 x i32> [ zeroinitializer, [[VECTOR_PH]] ], [ [[TMP58:%.*]], [[VECTOR_BODY]] ]
+; DEFAULT-NEXT:    [[TMP13:%.*]] = add i64 [[Y]], 1
+; DEFAULT-NEXT:    [[TMP14:%.*]] = add i64 [[Y]], 1
+; DEFAULT-NEXT:    [[TMP15:%.*]] = getelementptr i32, ptr [[SRC_1]], i64 [[TMP13]]
+; DEFAULT-NEXT:    [[TMP16:%.*]] = getelementptr i32, ptr [[SRC_1]], i64 [[TMP14]]
+; DEFAULT-NEXT:    [[TMP17:%.*]] = load i32, ptr [[TMP15]], align 4
+; DEFAULT-NEXT:    [[BROADCAST_SPLATINSERT:%.*]] = insertelement <vscale x 4 x i32> poison, i32 [[TMP17]], i64 0
+; DEFAULT-NEXT:    [[BROADCAST_SPLAT:%.*]] = shufflevector <vscale x 4 x i32> [[BROADCAST_SPLATINSERT]], <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer
+; DEFAULT-NEXT:    [[TMP18:%.*]] = load i32, ptr [[TMP16]], align 4
+; DEFAULT-NEXT:    [[BROADCAST_SPLATINSERT4:%.*]] = insertelement <vscale x 4 x i32> poison, i32 [[TMP18]], i64 0
+; DEFAULT-NEXT:    [[BROADCAST_SPLAT5]] = shufflevector <vscale x 4 x i32> [[BROADCAST_SPLATINSERT4]], <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer
+; DEFAULT-NEXT:    [[TMP19:%.*]] = call <vscale x 4 x i32> @llvm.vector.splice.nxv4i32(<vscale x 4 x i32> [[VECTOR_RECUR]], <vscale x 4 x i32> [[BROADCAST_SPLAT]], i32 -1)
+; DEFAULT-NEXT:    [[TMP20]] = call <vscale x 4 x i32> @llvm.vector.splice.nxv4i32(<vscale x 4 x i32> [[BROADCAST_SPLAT]], <vscale x 4 x i32> [[BROADCAST_SPLAT5]], i32 -1)
+; DEFAULT-NEXT:    [[TMP21:%.*]] = call <vscale x 4 x i32> @llvm.vector.splice.nxv4i32(<vscale x 4 x i32> [[VECTOR_RECUR2]], <vscale x 4 x i32> [[TMP19]], i32 -1)
+; DEFAULT-NEXT:    [[TMP22:%.*]] = call <vscale x 4 x i32> @llvm.vector.splice.nxv4i32(<vscale x 4 x i32> [[TMP19]], <vscale x 4 x i32> [[TMP20]], i32 -1)
+; DEFAULT-NEXT:    [[TMP23:%.*]] = or <vscale x 4 x i32> [[TMP21]], [[BROADCAST_SPLAT7]]
+; DEFAULT-NEXT:    [[TMP24:%.*]] = or <vscale x 4 x i32> [[TMP22]], [[BROADCAST_SPLAT7]]
+; DEFAULT-NEXT:    [[TMP25:%.*]] = lshr <vscale x 4 x i32> [[BROADCAST_SPLAT7]], shufflevector (<vscale x 4 x i32> insertelement (<vscale x 4 x i32> poison, i32 1, i64 0), <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer)
+; DEFAULT-NEXT:    [[TMP26:%.*]] = lshr <vscale x 4 x i32> [[BROADCAST_SPLAT7]], shufflevector (<vscale x 4 x i32> insertelement (<vscale x 4 x i32> poison, i32 1, i64 0), <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer)
+; DEFAULT-NEXT:    [[TMP27:%.*]] = shl <vscale x 4 x i32> [[TMP23]], shufflevector (<vscale x 4 x i32> insertelement (<vscale x 4 x i32> poison, i32 1, i64 0), <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer)
+; DEFAULT-NEXT:    [[TMP28:%.*]] = shl <vscale x 4 x i32> [[TMP24]], shufflevector (<vscale x 4 x i32> insertelement (<vscale x 4 x i32> poison, i32 1, i64 0), <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer)
+; DEFAULT-NEXT:    [[TMP29:%.*]] = or <vscale x 4 x i32> [[TMP27]], shufflevector (<vscale x 4 x i32> insertelement (<vscale x 4 x i32> poison, i32 2, i64 0), <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer)
+; DEFAULT-NEXT:    [[TMP30:%.*]] = or <vscale x 4 x i32> [[TMP28]], shufflevector (<vscale x 4 x i32> insertelement (<vscale x 4 x i32> poison, i32 2, i64 0), <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer)
+; DEFAULT-NEXT:    [[TMP31:%.*]] = shl <vscale x 4 x i32> [[BROADCAST_SPLAT7]], shufflevector (<vscale x 4 x i32> insertelement (<vscale x 4 x i32> poison, i32 1, i64 0), <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer)
+; DEFAULT-NEXT:    [[TMP32:%.*]] = shl <vscale x 4 x i32> [[BROADCAST_SPLAT7]], shufflevector (<vscale x 4 x i32> insertelement (<vscale x 4 x i32> poison, i32 1, i64 0), <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer)
+; DEFAULT-NEXT:    [[TMP33:%.*]] = or <vscale x 4 x i32> [[TMP25]], [[TMP31]]
+; DEFAULT-NEXT:    [[TMP34:%.*]] = or <vscale x 4 x i32> [[TMP26]], [[TMP32]]
+; DEFAULT-NEXT:    [[TMP35:%.*]] = or <vscale x 4 x i32> [[TMP33]], [[TMP29]]
+; DEFAULT-NEXT:    [[TMP36:%.*]] = or <vscale x 4 x i32> [[TMP34]], [[TMP30]]
+; DEFAULT-NEXT:    [[TMP37:%.*]] = or <vscale x 4 x i32> [[TMP35]], [[BROADCAST_SPLAT7]]
+; DEFAULT-NEXT:    [[TMP38:%.*]] = or <vscale x 4 x i32> [[TMP36]], [[BROADCAST_SPLAT7]]
+; DEFAULT-NEXT:    [[TMP39:%.*]] = or <vscale x 4 x i32> [[BROADCAST_SPLAT9]], [[BROADCAST_SPLAT7]]
+; DEFAULT-NEXT:    [[TMP40:%.*]] = or <vscale x 4 x i32> [[BROADCAST_SPLAT9]], [[BROADCAST_SPLAT7]]
+; DEFAULT-NEXT:    [[TMP41:%.*]] = and <vscale x 4 x i32> [[TMP39]], shufflevector (<vscale x 4 x i32> insertelement (<vscale x 4 x i32> poison, i32 1, i64 0), <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer)
+; DEFAULT-NEXT:    [[TMP42:%.*]] = and <vscale x 4 x i32> [[TMP40]], shufflevector (<vscale x 4 x i32> insertelement (<vscale x 4 x i32> poison, i32 1, i64 0), <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer)
+; DEFAULT-NEXT:    [[TMP43:%.*]] = xor <vscale x 4 x i32> [[TMP41]], shufflevector (<vscale x 4 x i32> insertelement (<vscale x 4 x i32> poison, i32 1, i64 0), <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer)
+; DEFAULT-NEXT:    [[TMP44:%.*]] = xor <vscale x 4 x i32> [[TMP42]], shufflevector (<vscale x 4 x i32> insertelement (<vscale x 4 x i32> poison, i32 1, i64 0), <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer)
+; DEFAULT-NEXT:    [[TMP45:%.*]] = zext <vscale x 4 x i32> [[TMP43]] to <vscale x 4 x i64>
+; DEFAULT-NEXT:    [[TMP46:%.*]] = zext <vscale x 4 x i32> [[TMP44]] to <vscale x 4 x i64>
+; DEFAULT-NEXT:    [[TMP47:%.*]] = getelementptr i32, ptr [[SRC_2]], <vscale x 4 x i64> [[TMP45]]
+; DEFAULT-NEXT:    [[TMP48:%.*]] = getelementptr i32, ptr [[SRC_2]], <vscale x 4 x i64> [[TMP46]]
+; DEFAULT-NEXT:    [[WIDE_MASKED_GATHER:%.*]] = call <vscale x 4 x i32> @llvm.masked.gather.nxv4i32.nxv4p0(<vscale x 4 x ptr> [[TMP47]], i32 4, <vscale x 4 x i1> shufflevector (<vscale x 4 x i1> insertelement (<vscale x 4 x i1> poison, i1 true, i64 0), <vscale x 4 x i1> poison, <vscale x 4 x i32> zeroinitializer), <vscale x 4 x i32> poison)
+; DEFAULT-NEXT:    [[WIDE_MASKED_GATHER10:%.*]] = call <vscale x 4 x i32> @llvm.masked.gather.nxv4i32.nxv4p0(<vscale x 4 x ptr> [[TMP48]], i32 4, <vscale x 4 x i1> shufflevector (<vscale x 4 x i1> insertelement (<vscale x 4 x i1> poison, i1 true, i64 0), <vscale x 4 x i1> poison, <vscale x 4 x i32> zeroinitializer), <vscale x 4 x i32> poison)
+; DEFAULT-NEXT:    [[TMP49:%.*]] = lshr <vscale x 4 x i32> [[TMP37]], shufflevector (<vscale x 4 x i32> insertelement (<vscale x 4 x i32> poison, i32 1, i64 0), <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer)
+; DEFAULT-NEXT:    [[TMP50:%.*]] = lshr <vscale x 4 x i32> [[TMP38]], shufflevector (<vscale x 4 x i32> insertelement (<vscale x 4 x i32> poison, i32 1, i64 0), <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer)
+; DEFAULT-NEXT:    [[TMP51:%.*]] = zext <vscale x 4 x i32> [[TMP49]] to <vscale x 4 x i64>
+; DEFAULT-NEXT:    [[TMP52:%.*]] = zext <vscale x 4 x i32> [[TMP50]] to <vscale x 4 x i64>
+; DEFAULT-NEXT:    [[TMP53:%.*]] = getelementptr i32, ptr [[SRC_2]], <vscale x 4 x i64> [[TMP51]]
+; DEFAULT-NEXT:    [[TMP54:%.*]] = getelementptr i32, ptr [[SRC_2]], <vscale x 4 x i64> [[TMP52]]
+; DEFAULT-NEXT:    [[WIDE_MASKED_GATHER11:%.*]] = call <vscale x 4 x i32> @llvm.masked.gather.nxv4i32.nxv4p0(<vscale x 4 x ptr> [[TMP53]], i32 4, <vscale x 4 x i1> shufflevector (<vscale x 4 x i1> insertelement (<vscale x 4 x i1> poison, i1 true, i64 0), <vscale x 4 x i1> poison, <vscale x 4 x i32> zeroinitializer), <vscale x 4 x i32> poison)
+; DEFAULT-NEXT:    [[WIDE_MASKED_GATHER12:%.*]] = call <vscale x 4 x i32> @llvm.masked.gather.nxv4i32.nxv4p0(<vscale x 4 x ptr> [[TMP54]], i32 4, <vscale x 4 x i1> shufflevector (<vscale x 4 x i1> insertelement (<vscale x 4 x i1> poison, i1 true, i64 0), <vscale x 4 x i1> poison, <vscale x 4 x i32> zeroinitializer), <vscale x 4 x i32> poison)
+; DEFAULT-NEXT:    [[TMP55:%.*]] = or <vscale x 4 x i32> [[WIDE_MASKED_GATHER]], [[VEC_PHI]]
+; DEFAULT-NEXT:    [[TMP56:%.*]] = or <vscale x 4 x i32> [[WIDE_MASKED_GATHER10]], [[VEC_PHI3]]
+; DEFAULT-NEXT:    [[TMP57]] = or <vscale x 4 x i32> [[TMP55]], [[WIDE_MASKED_GATHER11]]
+; DEFAULT-NEXT:    [[TMP58]] = or <vscale x 4 x i32> [[TMP56]], [[WIDE_MASKED_GATHER12]]
+; DEFAULT-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], [[TMP6]]
+; DEFAULT-NEXT:    [[TMP59:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
+; DEFAULT-NEXT:    br i1 [[TMP59]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP0:![0-9]+]]
+; DEFAULT:       middle.block:
+; DEFAULT-NEXT:    [[BIN_RDX:%.*]] = or <vscale x 4 x i32> [[TMP58]], [[TMP57]]
+; DEFAULT-NEXT:    [[TMP60:%.*]] = call i32 @llvm.vector.reduce.or.nxv4i32(<vscale x 4 x i32> [[BIN_RDX]])
+; DEFAULT-NEXT:    [[CMP_N:%.*]] = icmp eq i64 [[TMP0]], [[N_VEC]]
+; DEFAULT-NEXT:    [[TMP61:%.*]] = call i32 @llvm.vscale.i32()
+; DEFAULT-NEXT:    [[TMP62:%.*]] = mul i32 [[TMP61]], 4
+; DEFAULT-NEXT:    [[TMP63:%.*]] = sub i32 [[TMP62]], 1
+; DEFAULT-NEXT:    [[VECTOR_RECUR_EXTRACT:%.*]] = extractelement <vscale x 4 x i32> [[BROADCAST_SPLAT5]], i32 [[TMP63]]
+; DEFAULT-NEXT:    [[TMP64:%.*]] = call i32 @llvm.vscale.i32()
+; DEFAULT-NEXT:    [[TMP65:%.*]] = mul i32 [[TMP64]], 4
+; DEFAULT-NEXT:    [[TMP66:%.*]] = sub i32 [[TMP65]], 1
+; DEFAULT-NEXT:    [[VECTOR_RECUR_EXTRACT13:%.*]] = extractelement <vscale x 4 x i32> [[TMP20]], i32 [[TMP66]]
+; DEFAULT-NEXT:    br i1 [[CMP_N]], label [[EXIT:%.*]], label [[SCALAR_PH]]
+; DEFAULT:       scalar.ph:
+; DEFAULT-NEXT:    [[SCALAR_RECUR_INIT14:%.*]] = phi i32 [ 0, [[ENTRY:%.*]] ], [ [[VECTOR_RECUR_EXTRACT13]], [[MIDDLE_BLOCK]] ]
+; DEFAULT-NEXT:    [[SCALAR_RECUR_INIT:%.*]] = phi i32 [ 0, [[ENTRY]] ], [ [[VECTOR_RECUR_EXTRACT]], [[MIDDLE_BLOCK]] ]
+; DEFAULT-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ [[N_VEC]], [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY]] ]
+; DEFAULT-NEXT:    [[BC_MERGE_RDX:%.*]] = phi i32 [ 0, [[ENTRY]] ], [ [[TMP60]], [[MIDDLE_BLOCK]] ]
+; DEFAULT-NEXT:    br label [[LOOP:%.*]]
+; DEFAULT:       loop:
+; DEFAULT-NEXT:    [[SCALAR_RECUR:%.*]] = phi i32 [ [[SCALAR_RECUR_INIT]], [[SCALAR_PH]] ], [ [[TMP68:%.*]], [[LOOP]] ]
+; DEFAULT-NEXT:    [[SCALAR_RECUR15:%.*]] = phi i32 [ [[SCALAR_RECUR_INIT14]], [[SCALAR_PH]] ], [ [[SCALAR_RECUR]], [[LOOP]] ]
+; DEFAULT-NEXT:    [[IV:%.*]] = phi i64 [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ], [ [[IV_NEXT:%.*]], [[LOOP]] ]
+; DEFAULT-NEXT:    [[SUM_RED:%.*]] = phi i32 [ [[BC_MERGE_RDX]], [[SCALAR_PH]] ], [ [[RED_2:%.*]], [[LOOP]] ]
+; DEFAULT-NEXT:    [[TMP67:%.*]] = add i64 [[Y]], 1
+; DEFAULT-NEXT:    [[GEP_1:%.*]] = getelementptr i32, ptr [[SRC_1]], i64 [[TMP67]]
+; DEFAULT-NEXT:    [[TMP68]] = load i32, ptr [[GEP_1]], align 4
+; DEFAULT-NEXT:    [[OR3:%.*]] = or i32 [[SCALAR_RECUR15]], [[X]]
+; DEFAULT-NEXT:    [[IV_NEXT]] = add i64 [[IV]], 1
+; DEFAULT-NEXT:    [[SHR:%.*]] = lshr i32 [[X]], 1
+; DEFAULT-NEXT:    [[TMP69:%.*]] = shl i32 [[OR3]], 1
+; DEFAULT-NEXT:    [[TMP70:%.*]] = or i32 [[TMP69]], 2
+; DEFAULT-NEXT:    [[SHL19:%.*]] = shl i32 [[X]], 1
+; DEFAULT-NEXT:    [[TMP71:%.*]] = or i32 [[SHR]], [[SHL19]]
+; DEFAULT-NEXT:    [[TMP72:%.*]] = or i32 [[TMP71]], [[TMP70]]
+; DEFAULT-NEXT:    [[TMP73:%.*]] = or i32 [[TMP72]], [[X]]
+; DEFAULT-NEXT:    [[OR20:%.*]] = or i32 [[Z]], [[X]]
+; DEFAULT-NEXT:    [[NOT:%.*]] = and i32 [[OR20]], 1
+; DEFAULT-NEXT:    [[AND:%.*]] = xor i32 [[NOT]], 1
+; DEFAULT-NEXT:    [[IDX_EXT_1:%.*]] = zext i32 [[AND]] to i64
+; DEFAULT-NEXT:    [[GEP_2:%.*]] = getelementptr i32, ptr [[SRC_2]], i64 [[IDX_EXT_1]]
+; DEFAULT-NEXT:    [[TMP74:%.*]] = load i32, ptr [[GEP_2]], align 4
+; DEFAULT-NEXT:    [[SHR24:%.*]] = lshr i32 [[TMP73]], 1
+; DEFAULT-NEXT:    [[IDX_EXT_2:%.*]] = zext i32 [[SHR24]] to i64
+; DEFAULT-NEXT:    [[GEP_3:%.*]] = getelementptr i32, ptr [[SRC_2]], i64 [[IDX_EXT_2]]
+; DEFAULT-NEXT:    [[TMP75:%.*]] = load i32, ptr [[GEP_3]], align 4
+; DEFAULT-NEXT:    [[RED_1:%.*]] = or i32 [[TMP74]], [[SUM_RED]]
+; DEFAULT-NEXT:    [[RED_2]] = or i32 [[RED_1]], [[TMP75]]
+; DEFAULT-NEXT:    [[EC:%.*]] = icmp eq i64 [[IV]], [[Y]]
+; DEFAULT-NEXT:    br i1 [[EC]], label [[EXIT]], label [[LOOP]], !llvm.loop [[LOOP3:![0-9]+]]
+; DEFAULT:       exit:
+; DEFAULT-NEXT:    [[RED_2_LCSSA:%.*]] = phi i32 [ [[RED_2]], [[LOOP]] ], [ [[TMP60]], [[MIDDLE_BLOCK]] ]
+; DEFAULT-NEXT:    ret i32 [[RED_2_LCSSA]]
+;
+; PRED-LABEL: define i32 @chained_recurrences(
+; PRED-SAME: i32 [[X:%.*]], i64 [[Y:%.*]], ptr [[SRC_1:%.*]], i32 [[Z:%.*]], ptr [[SRC_2:%.*]]) #[[ATTR0:[0-9]+]] {
+; PRED-NEXT:  entry:
+; PRED-NEXT:    [[TMP0:%.*]] = add i64 [[Y]], 1
+; PRED-NEXT:    br i1 false, label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
+; PRED:       vector.ph:
+; PRED-NEXT:    [[TMP1:%.*]] = call i64 @llvm.vscale.i64()
+; PRED-NEXT:    [[TMP2:%.*]] = mul i64 [[TMP1]], 4
+; PRED-NEXT:    [[TMP3:%.*]] = call i64 @llvm.vscale.i64()
+; PRED-NEXT:    [[TMP4:%.*]] = mul i64 [[TMP3]], 4
+; PRED-NEXT:    [[TMP5:%.*]] = sub i64 [[TMP4]], 1
+; PRED-NEXT:    [[N_RND_UP:%.*]] = add i64 [[TMP0]], [[TMP5]]
+; PRED-NEXT:    [[N_MOD_VF:%.*]] = urem i64 [[N_RND_UP]], [[TMP2]]
+; PRED-NEXT:    [[N_VEC:%.*]] = sub i64 [[N_RND_UP]], [[N_MOD_VF]]
+; PRED-NEXT:    [[TMP6:%.*]] = call i64 @llvm.vscale.i64()
+; PRED-NEXT:    [[TMP7:%.*]] = mul i64 [[TMP6]], 4
+; PRED-NEXT:    [[TMP8:%.*]] = call i64 @llvm.vscale.i64()
+; PRED-NEXT:    [[TMP9:%.*]] = mul i64 [[TMP8]], 4
+; PRED-NEXT:    [[TMP10:%.*]] = sub i64 [[TMP0]], [[TMP9]]
+; PRED-NEXT:    [[TMP11:%.*]] = icmp ugt i64 [[TMP0]], [[TMP9]]
+; PRED-NEXT:    [[TMP12:%.*]] = select i1 [[TMP11]], i64 [[TMP10]], i64 0
+; PRED-NEXT:    [[ACTIVE_LANE_MASK_ENTRY:%.*]] = call <vscale x 4 x i1> @llvm.get.active.lane.mask.nxv4i1.i64(i64 0, i64 [[TMP0]])
+; PRED-NEXT:    [[TMP13:%.*]] = call i32 @llvm.vscale.i32()
+; PRED-NEXT:    [[TMP14:%.*]] = mul i32 [[TMP13]], 4
+; PRED-NEXT:    [[TMP15:%.*]] = sub i32 [[TMP14]], 1
+; PRED-NEXT:    [[VECTOR_RECUR_INIT:%.*]] = insertelement <vscale x 4 x i32> poison, i32 0, i32 [[TMP15]]
+; PRED-NEXT:    [[TMP16:%.*]] = call i32 @llvm.vscale.i32()
+; PRED-NEXT:    [[TMP17:%.*]] = mul i32 [[TMP16]], 4
+; PRED-NEXT:    [[TMP18:%.*]] = sub i32 [[TMP17]], 1
+; PRED-NEXT:    [[VECTOR_RECUR_INIT1:%.*]] = insertelement <vscale x 4 x i32> poison, i32 0, i32 [[TMP18]]
+; PRED-NEXT:    [[BROADCAST_SPLATINSERT3:%.*]] = insertelement <vscale x 4 x i32> poison, i32 [[X]], i64 0
+; PRED-NEXT:    [[BROADCAST_SPLAT4:%.*]] = shufflevector <vscale x 4 x i32> [[BROADCAST_SPLATINSERT3]], <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer
+; PRED-NEXT:    [[BROADCAST_SPLATINSERT5:%.*]] = insertelement <vscale x 4 x i32> poison, i32 [[Z]], i64 0
+; PRED-NEXT:    [[BROADCAST_SPLAT6:%.*]] = shufflevector <vscale x 4 x i32> [[BROADCAST_SPLATINSERT5]], <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer
+; PRED-NEXT:    br label [[VECTOR_BODY:%.*]]
+; PRED:       vector.body:
+; PRED-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
+; PRED-NEXT:    [[ACTIVE_LANE_MASK:%.*]] = phi <vscale x 4 x i1> [ [[ACTIVE_LANE_MASK_ENTRY]], [[VECTOR_PH]] ], [ [[ACTIVE_LANE_MASK_NEXT:%.*]], [[VECTOR_BODY]] ]
+; PRED-NEXT:    [[VECTOR_RECUR:%.*]] = phi <vscale x 4 x i32> [ [[VECTOR_RECUR_INIT]], [[VECTOR_PH]] ], [ [[BROADCAST_SPLAT:%.*]], [[VECTOR_BODY]] ]
+; PRED-NEXT:    [[VECTOR_RECUR2:%.*]] = phi <vscale x 4 x i32> [ [[VECTOR_RECUR_INIT1]], [[VECTOR_PH]] ], [ [[TMP22:%.*]], [[VECTOR_BODY]] ]
+; PRED-NEXT:    [[VEC_PHI:%.*]] = phi <vscale x 4 x i32> [ zeroinitializer, [[VECTOR_PH]] ], [ [[TMP42:%.*]], [[VECTOR_BODY]] ]
+; PRED-NEXT:    [[TMP19:%.*]] = add i64 [[Y]], 1
+; PRED-NEXT:    [[TMP20:%.*]] = getelementptr i32, ptr [[SRC_1]], i64 [[TMP19]]
+; PRED-NEXT:    [[TMP21:%.*]] = load i32, ptr [[TMP20]], align 4
+; PRED-NEXT:    [[BROADCAST_SPLATINSERT:%.*]] = insertelement <vscale x 4 x i32> poison, i32 [[TMP21]], i64 0
+; PRED-NEXT:    [[BROADCAST_SPLAT]] = shufflevector <vscale x 4 x i32> [[BROADCAST_SPLATINSERT]], <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer
+; PRED-NEXT:    [[TMP22]] = call <vscale x 4 x i32> @llvm.vector.splice.nxv4i32(<vscale x 4 x i32> [[VECTOR_RECUR]], <vscale x 4 x i32> [[BROADCAST_SPLAT]], i32 -1)
+; PRED-NEXT:    [[TMP23:%.*]] = call <vscale x 4 x i32> @llvm.vector.splice.nxv4i32(<vscale x 4 x i32> [[VECTOR_RECUR2]], <vscale x 4 x i32> [[TMP22]], i32 -1)
+; PRED-NEXT:    [[TMP24:%.*]] = or <vscale x 4 x i32> [[TMP23]], [[BROADCAST_SPLAT4]]
+; PRED-NEXT:    [[TMP25:%.*]] = lshr <vscale x 4 x i32> [[BROADCAST_SPLAT4]], shufflevector (<vscale x 4 x i32> insertelement (<vscale x 4 x i32> poison, i32 1, i64 0), <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer)
+; PRED-NEXT:    [[TMP26:%.*]] = shl <vscale x 4 x i32> [[TMP24]], shufflevector (<vscale x 4 x i32> insertelement (<vscale x 4 x i32> poison, i32 1, i64 0), <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer)
+; PRED-NEXT:    [[TMP27:%.*]] = or <vscale x 4 x i32> [[TMP26]], shufflevector (<vscale x 4 x i32> insertelement (<vscale x 4 x i32> poison, i32 2, i64 0), <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer)
+; PRED-NEXT:    [[TMP28:%.*]] = shl <vscale x 4 x i32> [[BROADCAST_SPLAT4]], shufflevector (<vscale x 4 x i32> insertelement (<vscale x 4 x i32> poison, i32 1, i64 0), <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer)
+; PRED-NEXT:    [[TMP29:%.*]] = or <vscale x 4 x i32> [[TMP25]], [[TMP28]]
+; PRED-NEXT:    [[TMP30:%.*]] = or <vscale x 4 x i32> [[TMP29]], [[TMP27]]
+; PRED-NEXT:    [[TMP31:%.*]] = or <vscale x 4 x i32> [[TMP30]], [[BROADCAST_SPLAT4]]
+; PRED-NEXT:    [[TMP32:%.*]] = or <vscale x 4 x i32> [[BROADCAST_SPLAT6]], [[BROADCAST_SPLAT4]]
+; PRED-NEXT:    [[TMP33:%.*]] = and <vscale x 4 x i32> [[TMP32]], shufflevector (<vscale x 4 x i32> insertelement (<vscale x 4 x i32> poison, i32 1, i64 0), <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer)
+; PRED-NEXT:    [[TMP34:%.*]] = xor <vscale x 4 x i32> [[TMP33]], shufflevector (<vscale x 4 x i32> insertelement (<vscale x 4 x i32> poison, i32 1, i64 0), <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer)
+; PRED-NEXT:    [[TMP35:%.*]] = zext <vscale x 4 x i32> [[TMP34]] to <vscale x 4 x i64>
+; PRED-NEXT:    [[TMP36:%.*]] = getelementptr i32, ptr [[SRC_2]], <vscale x 4 x i64> [[TMP35]]
+; PRED-NEXT:    [[WIDE_MASKED_GATHER:%.*]] = call <vscale x 4 x i32> @llvm.masked.gather.nxv4i32.nxv4p0(<vscale x 4 x ptr> [[TMP36]], i32 4, <vscale x 4 x i1> [[ACTIVE_LANE_MASK]], <vscale x 4 x i32> poison)
+; PRED-NEXT:    [[TMP37:%.*]] = lshr <vscale x 4 x i32> [[TMP31]], shufflevector (<vscale x 4 x i32> insertelement (<vscale x 4 x i32> poison, i32 1, i64 0), <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer)
+; PRED-NEXT:    [[TMP38:%.*]] = zext <vscale x 4 x i32> [[TMP37]] to <vscale x 4 x i64>
+; PRED-NEXT:    [[TMP39:%.*]] = getelementptr i32, ptr [[SRC_2]], <vscale x 4 x i64> [[TMP38]]
+; PRED-NEXT:    [[WIDE_MASKED_GATHER7:%.*]] = call <vscale x 4 x i32> @llvm.masked.gather.nxv4i32.nxv4p0(<vscale x 4 x ptr> [[TMP39]], i32 4, <vscale x 4 x i1> [[ACTIVE_LANE_MASK]], <vscale x 4 x i32> poison)
+; PRED-NEXT:    [[TMP40:%.*]] = or <vscale x 4 x i32> [[WIDE_MASKED_GATHER]], [[VEC_PHI]]
+; PRED-NEXT:    [[TMP41:%.*]] = or <vscale x 4 x i32> [[TMP40]], [[WIDE_MASKED_GATHER7]]
+; PRED-NEXT:    [[TMP42]] = select <vscale x 4 x i1> [[ACTIVE_LANE_MASK]], <vscale x 4 x i32> [[TMP41]], <vscale x 4 x i32> [[VEC_PHI]]
+; PRED-NEXT:    [[INDEX_NEXT]] = add i64 [[INDEX]], [[TMP7]]
+; PRED-NEXT:    [[ACTIVE_LANE_MASK_NEXT]] = call <vscale x 4 x i1> @llvm.get.active.lane.mask.nxv4i1.i64(i64 [[INDEX]], i64 [[TMP12]])
+; PRED-NEXT:    [[TMP43:%.*]] = xor <vscale x 4 x i1> [[ACTIVE_LANE_MASK_NEXT]], shufflevector (<vscale x 4 x i1> insertelement (<vscale x 4 x i1> poison, i1 true, i64 0), <vscale x 4 x i1> poison, <vscale x 4 x i32> zeroinitializer)
+; PRED-NEXT:    [[TMP44:%.*]] = extractelement <vscale x 4 x i1> [[TMP43]], i32 0
+; PRED-NEXT:    br i1 [[TMP44]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP0:![0-9]+]]
+; PRED:       middle.block:
+; PRED-NEXT:    [[TMP45:%.*]] = call i32 @llvm.vector.reduce.or.nxv4i32(<vscale x 4 x i32> [[TMP42]])
+; PRED-NEXT:    [[TMP46:%.*]] = call i32 @llvm.vscale.i32()
+; PRED-NEXT:    [[TMP47:%.*]] = mul i32 [[TMP46]], 4
+; PRED-NEXT:    [[TMP48:%.*]] = sub i32 [[TMP47]], 1
+; PRED-NEXT:    [[VECTOR_RECUR_EXTRACT:%.*]] = extractelement <vscale x 4 x i32> [[BROADCAST_SPLAT]], i32 [[TMP48]]
+; PRED-NEXT:    [[TMP49:%.*]] = call i32 @llvm.vscale.i32()
+; PRED-NEXT:    [[TMP50:%.*]] = mul i32 [[TMP49]], 4
+; PRED-NEXT:    [[TMP51:%.*]] = sub i32 [[TMP50]], 1
+; PRED-NEXT:    [[VECTOR_RECUR_EXTRACT8:%.*]] = extractelement <vscale x 4 x i32> [[TMP22]], i32 [[TMP51]]
+; PRED-NEXT:    br i1 true, label [[EXIT:%.*]], label [[SCALAR_PH]]
+; PRED:       scalar.ph:
+; PRED-NEXT:    [[SCALAR_RECUR_INIT9:%.*]] = phi i32 [ 0, [[ENTRY:%.*]] ], [ [[VECTOR_RECUR_EXTRACT8]], [[MIDDLE_BLOCK]] ]
+; PRED-NEXT:    [[SCALAR_RECUR_INIT:%.*]] = phi i32 [ 0, [[ENTRY]] ], [ [[VECTOR_RECUR_EXTRACT]], [[MIDDLE_BLOCK]] ]
+; PRED-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ [[N_VEC]], [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY]] ]
+; PRED-NEXT:    [[BC_MERGE_RDX:%.*]] = phi i32 [ 0, [[ENTRY]] ], [ [[TMP45]], [[MIDDLE_BLOCK]] ]
+; PRED-NEXT:    br label [[LOOP:%.*]]
+; PRED:       loop:
+; PRED-NEXT:    [[SCALAR_RECUR:%.*]] = phi i32 [ [[SCALAR_RECUR_INIT]], [[SCALAR_PH]] ], [ [[TMP53:%.*]], [[LOOP]] ]
+; PRED-NEXT:    [[SCALAR_RECUR10:%.*]] = phi i32 [ [[SCALAR_RECUR_INIT9]], [[SCALAR_PH]] ], [ [[SCALAR_RECUR]], [[LOOP]] ]
+; PRED-NEXT:    [[IV:%.*]] = phi i64 [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ], [ [[IV_NEXT:%.*]], [[LOOP]] ]
+; PRED-NEXT:    [[SUM_RED:%.*]] = phi i32 [ [[BC_MERGE_RDX]], [[SCALAR_PH]] ], [ [[RED_2:%.*]], [[LOOP]] ]
+; PRED-NEXT:    [[TMP52:%.*]] = add i64 [[Y]], 1
+; PRED-NEXT:    [[GEP_1:%.*]] = getelementptr i32, ptr [[SRC_1]], i64 [[TMP52]]
+; PRED-NEXT:    [[TMP53]] = load i32, ptr [[GEP_1]], align 4
+; PRED-NEXT:    [[OR3:%.*]] = or i32 [[SCALAR_RECUR10]], [[X]]
+; PRED-NEXT:    [[IV_NEXT]] = add i64 [[IV]], 1
+; PRED-NEXT:    [[SHR:%.*]] = lshr i32 [[X]], 1
+; PRED-NEXT:    [[TMP54:%.*]] = shl i32 [[OR3]], 1
+; PRED-NEXT:    [[TMP55:%.*]] = or i32 [[TMP54]], 2
+; PRED-NEXT:    [[SHL19:%.*]] = shl i32 [[X]], 1
+; PRED-NEXT:    [[TMP56:%.*]] = or i32 [[SHR]], [[SHL19]]
+; PRED-NEXT:    [[TMP57:%.*]] = or i32 [[TMP56]], [[TMP55]]
+; PRED-NEXT:    [[TMP58:%.*]] = or i32 [[TMP57]], [[X]]
+; PRED-NEXT:    [[OR20:%.*]] = or i32 [[Z]], [[X]]
+; PRED-NEXT:    [[NOT:%.*]] = and i32 [[OR20]], 1
+; PRED-NEXT:    [[AND:%.*]] = xor i32 [[NOT]], 1
+; PRED-NEXT:    [[IDX_EXT_1:%.*]] = zext i32 [[AND]] to i64
+; PRED-NEXT:    [[GEP_2:%.*]] = getelementptr i32, ptr [[SRC_2]], i64 [[IDX_EXT_1]]
+; PRED-NEXT:    [[TMP59:%.*]] = load i32, ptr [[GEP_2]], align 4
+; PRED-NEXT:    [[SHR24:%.*]] = lshr i32 [[TMP58]], 1
+; PRED-NEXT:    [[IDX_EXT_2:%.*]] = zext i32 [[SHR24]] to i64
+; PRED-NEXT:    [[GEP_3:%.*]] = getelementptr i32, ptr [[SRC_2]], i64 [[IDX_EXT_2]]
+; PRED-NEXT:    [[TMP60:%.*]] = load i32, ptr [[GEP_3]], align 4
+; PRED-NEXT:    [[RED_1:%.*]] = or i32 [[TMP59]], [[SUM_RED]]
+; PRED-NEXT:    [[RED_2]] = or i32 [[RED_1]], [[TMP60]]
+; PRED-NEXT:    [[EC:%.*]] = icmp eq i64 [[IV]], [[Y]]
+; PRED-NEXT:    br i1 [[EC]], label [[EXIT]], label [[LOOP]], !llvm.loop [[LOOP3:![0-9]+]]
+; PRED:       exit:
+; PRED-NEXT:    [[RED_2_LCSSA:%.*]] = phi i32 [ [[RED_2]], [[LOOP]] ], [ [[TMP45]], [[MIDDLE_BLOCK]] ]
+; PRED-NEXT:    ret i32 [[RED_2_LCSSA]]
+;
+entry:
+  br label %loop
+
+loop:
+  %2 = phi i32 [ 0, %entry ], [ %5, %loop ]
+  %3 = phi i32 [ 0, %entry ], [ %2, %loop ]
+  %iv = phi i64 [ 0, %entry ], [ %iv.next, %loop ]
+  %sum.red = phi i32 [ 0, %entry ], [ %red.2, %loop ]
+  %4 = add i64 %y, 1
+  %gep.1 = getelementptr i32, ptr %src.1, i64 %4
+  %5 = load i32, ptr %gep.1, align 4
+  %or3 = or i32 %3, %x
+  %iv.next = add i64 %iv, 1
+  %shr = lshr i32 %x, 1
+  %6 = shl i32 %or3, 1
+  %7 = or i32 %6, 2
+  %shl19 = shl i32 %x, 1
+  %8 = or i32 %shr, %shl19
+  %9 = or i32 %8, %7
+  %10 = or i32 %9, %x
+  %or20 = or i32 %z, %x
+  %not = and i32 %or20, 1
+  %and = xor i32 %not, 1
+  %idx.ext.1 = zext i32 %and to i64
+  %gep.2 = getelementptr i32, ptr %src.2, i64 %idx.ext.1
+  %11 = load i32, ptr %gep.2, align 4
+  %shr24 = lshr i32 %10, 1
+  %idx.ext.2 = zext i32 %shr24 to i64
+  %gep.3 = getelementptr i32, ptr %src.2, i64 %idx.ext.2
+  %12 = load i32, ptr %gep.3, align 4
+  %red.1 = or i32 %11, %sum.red
+  %red.2 = or i32 %red.1, %12
+  %ec = icmp eq i64 %iv, %y
+  br i1 %ec, label %exit, label %loop
+
+exit:
+  ret i32 %red.2
+}
+
+define i16 @reduce_udiv(ptr %src, i16 %x, i64 %N) #0 {
+; DEFAULT-LABEL: define i16 @reduce_udiv(
+; DEFAULT-SAME: ptr [[SRC:%.*]], i16 [[X:%.*]], i64 [[N:%.*]]) #[[ATTR0]] {
+; DEFAULT-NEXT:  entry:
+; DEFAULT-NEXT:    [[TMP0:%.*]] = add i64 [[N]], 1
+; DEFAULT-NEXT:    [[TMP1:%.*]] = call i64 @llvm.vscale.i64()
+; DEFAULT-NEXT:    [[TMP2:%.*]] = mul i64 [[TMP1]], 8
+; DEFAULT-NEXT:    [[MIN_ITERS_CHECK:%.*]] = icmp ult i64 [[TMP0]], [[TMP2]]
+; DEFAULT-NEXT:    br i1 [[MIN_ITERS_CHECK]], label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
+; DEFAULT:       vector.ph:
+; DEFAULT-NEXT:    [[TMP3:%.*]] = call i64 @llvm.vscale.i64()
+; DEFAULT-NEXT:    [[TMP4:%.*]] = mul i64 [[TMP3]], 8
+; DEFAULT-NEXT:    [[N_MOD_VF:%.*]] = urem i64 [[TMP0]], [[TMP4]]
+; DEFAULT-NEXT:    [[N_VEC:%.*]] = sub i64 [[TMP0]], [[N_MOD_VF]]
+; DEFAULT-NEXT:    [[TMP5:%.*]] = call i64 @llvm.vscale.i64()
+; DEFAULT-NEXT:    [[TMP6:%.*]] = mul i64 [[TMP5]], 8
+; DEFAULT-NEXT:    [[BROADCAST_SPLATINSERT:%.*]] = insertelement <vscale x 4 x i16> poison, i16 [[X]], i64 0
+; DEFAULT-NEXT:    [[BROADCAST_SPLAT:%.*]] = shufflevector <vscale x 4 x i16> [[BROADCAST_SPLATINSERT]], <vscale x 4 x i16> poison, <vscale x 4 x i32> zeroinitializer
+; DEFAULT-NEXT:    br label [[VECTOR_BODY:%.*]]
+; DEFAULT:       vector.body:
+; DEFAULT-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
+; DEFAULT-NEXT:    [[VEC_PHI:%.*]] = phi <vscale x 4 x i16> [ zeroinitializer, [[VECTOR_PH]] ], [ [[TMP21:%.*]], [[VECTOR_BODY]] ]
+; DEFAULT-NEXT:    [[VEC_PHI1:%.*]] = phi <vscale x 4 x i16> [ zeroinitializer, [[VECTOR_PH]] ], [ [[TMP22:%.*]], [[VECTOR_BODY]] ]
+; DEFAULT-NEXT:    [[TMP7:%.*]] = add i64 [[INDEX]], 0
+; DEFAULT-NEXT:    [[TMP8:%.*]] = call i64 @llvm.vscale.i64()
+; DEFAULT-NEXT:    [[TMP9:%.*]] = mul i64 [[TMP8]], 4
+; DEFAULT-NEXT:    [[TMP10:%.*]] = add i64 [[TMP9]], 0
+; DEFAULT-NEXT:    [[TMP11:%.*]] = mul i64 [[TMP10]], 1
+; DEFAULT-NEXT:    [[TMP12:%.*]] = add i64 [[INDEX]], [[TMP11]]
+; DEFAULT-NEXT:    [[TMP13:%.*]] = getelementptr i16, ptr [[SRC]], i64 [[TMP7]]
+; DEFAULT-NEXT:    [[TMP14:%.*]] = getelementptr i16, ptr [[SRC]], i64 [[TMP12]]
+; DEFAULT-NEXT:    [[TMP15:%.*]] = getelementptr i16, ptr [[TMP13]], i32 0
+; DEFAULT-NEXT:    [[TMP16:%.*]] = call i64 @llvm.vscale.i64()
+; DEFAULT-NEXT:    [[TMP17:%.*]] = mul i64 [[TMP16]], 4
+; DEFAULT-NEXT:    [[TMP18:%.*]] = getelementptr i16, ptr [[TMP13]], i64 [[TMP17]]
+; DEFAULT-NEXT:    [[WIDE_LOAD:%.*]] = load <vscale x 4 x i16>, ptr [[TMP15]], align 2
+; DEFAULT-NEXT:    [[WIDE_LOAD2:%.*]] = load <vscale x 4 x i16>, ptr [[TMP18]], align 2
+; DEFAULT-NEXT:    [[TMP19:%.*]] = udiv <vscale x 4 x i16> [[WIDE_LOAD]], [[BROADCAST_SPLAT]]
+; DEFAULT-NEXT:    [[TMP20:%.*]] = udiv <vscale x 4 x i16> [[WIDE_LOAD2]], [[BROADCAST_SPLAT]]
+; DEFAULT-NEXT:    [[TMP21]] = or <vscale x 4 x i16> [[TMP19]], [[VEC_PHI]]
+; DEFAULT-NEXT:    [[TMP22]] = or <vscale x 4 x i16> [[TMP20]], [[VEC_PHI1]]
+; DEFAULT-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], [[TMP6]]
+; DEFAULT-NEXT:    [[TMP23:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
+; DEFAULT-NEXT:    br i1 [[TMP23]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP4:![0-9]+]]
+; DEFAULT:       middle.block:
+; DEFAULT-NEXT:    [[BIN_RDX:%.*]] = or <vscale x 4 x i16> [[TMP22]], [[TMP21]]
+; DEFAULT-NEXT:    [[TMP24:%.*]] = call i16 @llvm.vector.reduce.or.nxv4i16(<vscale x 4 x i16> [[BIN_RDX]])
+; DEFAULT-NEXT:    [[CMP_N:%.*]] = icmp eq i64 [[TMP0]], [[N_VEC]]
+; DEFAULT-NEXT:    br i1 [[CMP_N]], label [[EXIT:%.*]], label [[SCALAR_PH]]
+; DEFAULT:       scalar.ph:
+; DEFAULT-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ [[N_VEC]], [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ]
+; DEFAULT-NEXT:    [[BC_MERGE_RDX:%.*]] = phi i16 [ 0, [[ENTRY]] ], [ [[TMP24]], [[MIDDLE_BLOCK]] ]
+; DEFAULT-NEXT:    br label [[LOOP:%.*]]
+; DEFAULT:       loop:
+; DEFAULT-NEXT:    [[IV:%.*]] = phi i64 [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ], [ [[IV_NEXT:%.*]], [[LOOP]] ]
+; DEFAULT-NEXT:    [[RED:%.*]] = phi i16 [ [[BC_MERGE_RDX]], [[SCALAR_PH]] ], [ [[RED_NEXT:%.*]], [[LOOP]] ]
+; DEFAULT-NEXT:    [[GEP:%.*]] = getelementptr i16, ptr [[SRC]], i64 [[IV]]
+; DEFAULT-NEXT:    [[L:%.*]] = load i16, ptr [[GEP]], align 2
+; DEFAULT-NEXT:    [[DIV:%.*]] = udiv i16 [[L]], [[X]]
+; DEFAULT-NEXT:    [[RED_NEXT]] = or i16 [[DIV]], [[RED]]
+; DEFAULT-NEXT:    [[IV_NEXT]] = add i64 [[IV]], 1
+; DEFAULT-NEXT:    [[EC:%.*]] = icmp eq i64 [[IV]], [[N]]
+; DEFAULT-NEXT:    br i1 [[EC]], label [[EXIT]], label [[LOOP]], !llvm.loop [[LOOP5:![0-9]+]]
+; DEFAULT:       exit:
+; DEFAULT-NEXT:    [[RED_NEXT_LCSSA:%.*]] = phi i16 [ [[RED_NEXT]], [[LOOP]] ], [ [[TMP24]], [[MIDDLE_BLOCK]] ]
+; DEFAULT-NEXT:    ret i16 [[RED_NEXT_LCSSA]]
+;
+; PRED-LABEL: define i16 @reduce_udiv(
+; PRED-SAME: ptr [[SRC:%.*]], i16 [[X:%.*]], i64 [[N:%.*]]) #[[ATTR0]] {
+; PRED-NEXT:  entry:
+; PRED-NEXT:    [[TMP0:%.*]] = add i64 [[N]], 1
+; PRED-NEXT:    br i1 false, label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
+; PRED:       vector.ph:
+; PRED-NEXT:    [[TMP1:%.*]] = call i64 @llvm.vscale.i64()
+; PRED-NEXT:    [[TMP2:%.*]] = mul i64 [[TMP1]], 8
+; PRED-NEXT:    [[TMP3:%.*]] = call i64 @llvm.vscale.i64()
+; PRED-NEXT:    [[TMP4:%.*]] = mul i64 [[TMP3]], 8
+; PRED-NEXT:    [[TMP5:%.*]] = sub i64 [[TMP4]], 1
+; PRED-NEXT:    [[N_RND_UP:%.*]] = add i64 [[TMP0]], [[TMP5]]
+; PRED-NEXT:    [[N_MOD_VF:%.*]] = urem i64 [[N_RND_UP]], [[TMP2]]
+; PRED-NEXT:    [[N_VEC:%.*]] = sub i64 [[N_RND_UP]], [[N_MOD_VF]]
+; PRED-NEXT:    [[TMP6:%.*]] = call i64 @llvm.vscale.i64()
+; PRED-NEXT:    [[TMP7:%.*]] = mul i64 [[TMP6]], 8
+; PRED-NEXT:    [[TMP8:%.*]] = call i64 @llvm.vscale.i64()
+; PRED-NEXT:    [[TMP9:%.*]] = mul i64 [[TMP8]], 8
+; PRED-NEXT:    [[TMP10:%.*]] = sub i64 [[TMP0]], [[TMP9]]
+; PRED-NEXT:    [[TMP11:%.*]] = icmp ugt i64 [[TMP0]], [[TMP9]]
+; PRED-NEXT:    [[TMP12:%.*]] = select i1 [[TMP11]], i64 [[TMP10]], i64 0
+; PRED-NEXT:    [[ACTIVE_LANE_MASK_ENTRY:%.*]] = call <vscale x 8 x i1> @llvm.get.active.lane.mask.nxv8i1.i64(i64 0, i64 [[TMP0]])
+; PRED-NEXT:    [[BROADCAST_SPLATINSERT:%.*]] = insertelement <vscale x 8 x i16> poison, i16 [[X]], i64 0
+; PRED-NEXT:    [[BROADCAST_SPLAT:%.*]] = shufflevector <vscale x 8 x i16> [[BROADCAST_SPLATINSERT]], <vscale x 8 x i16> poison, <vscale x 8 x i32> zeroinitializer
+; PRED-NEXT:    br label [[VECTOR_BODY:%.*]]
+; PRED:       vector.body:
+; PRED-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
+; PRED-NEXT:    [[ACTIVE_LANE_MASK:%.*]] = phi <vscale x 8 x i1> [ [[ACTIVE_LANE_MASK_ENTRY]], [[VECTOR_PH]] ], [ [[ACTIVE_LANE_MASK_NEXT:%.*]], [[VECTOR_BODY]] ]
+; PRED-NEXT:    [[VEC_PHI:%.*]] = phi <vscale x 8 x i16> [ zeroinitializer, [[VECTOR_PH]] ], [ [[TMP19:%.*]], [[VECTOR_BODY]] ]
+; PRED-NEXT:    [[TMP13:%.*]] = add i64 [[INDEX]], 0
+; PRED-NEXT:    [[TMP14:%.*]] = getelementptr i16, ptr [[SRC]], i64 [[TMP13]]
+; PRED-NEXT:    [[TMP15:%.*]] = getelementptr i16, ptr [[TMP14]], i32 0
+; PRED-NEXT:    [[WIDE_MASKED_LOAD:%.*]] = call <vscale x 8 x i16> @llvm.masked.load.nxv8i16.p0(ptr [[TMP15]], i32 2, <vscale x 8 x i1> [[ACTIVE_LANE_MASK]], <vscale x 8 x i16> poison)
+; PRED-NEXT:    [[TMP16:%.*]] = select <vscale x 8 x i1> [[ACTIVE_LANE_MASK]], <vscale x 8 x i16> [[BROADCAST_SPLAT]], <vscale x 8 x i16> shufflevector (<vscale x 8 x i16> insertelement (<vscale x 8 x i16> poison, i16 1, i64 0), <vscale x 8 x i16> poison, <vscale x 8 x i32> zeroinitializer)
+; PRED-NEXT:    [[TMP17:%.*]] = udiv <vscale x 8 x i16> [[WIDE_MASKED_LOAD]], [[TMP16]]
+; PRED-NEXT:    [[TMP18:%.*]] = or <vscale x 8 x i16> [[TMP17]], [[VEC_PHI]]
+; PRED-NEXT:    [[TMP19]] = select <vscale x 8 x i1> [[ACTIVE_LANE_MASK]], <vscale x 8 x i16> [[TMP18]], <vscale x 8 x i16> [[VEC_PHI]]
+; PRED-NEXT:    [[INDEX_NEXT]] = add i64 [[INDEX]], [[TMP7]]
+; PRED-NEXT:    [[ACTIVE_LANE_MASK_NEXT]] = call <vscale x 8 x i1> @llvm.get.active.lane.mask.nxv8i1.i64(i64 [[INDEX]], i64 [[TMP12]])
+; PRED-NEXT:    [[TMP20:%.*]] = xor <vscale x 8 x i1> [[ACTIVE_LANE_MASK_NEXT]], shufflevector (<vscale x 8 x i1> insertelement (<vscale x 8 x i1> poison, i1 true, i64 0), <vscale x 8 x i1> poison, <vscale x 8 x i32> zeroinitializer)
+; PRED-NEXT:    [[TMP21:%.*]] = extractelement <vscale x 8 x i1> [[TMP20]], i32 0
+; PRED-NEXT:    br i1 [[TMP21]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP4:![0-9]+]]
+; PRED:       middle.block:
+; PRED-NEXT:    [[TMP22:%.*]] = call i16 @llvm.vector.reduce.or.nxv8i16(<vscale x 8 x i16> [[TMP19]])
+; PRED-NEXT:    br i1 true, label [[EXIT:%.*]], label [[SCALAR_PH]]
+; PRED:       scalar.ph:
+; PRED-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ [[N_VEC]], [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ]
+; PRED-NEXT:    [[BC_MERGE_RDX:%.*]] = phi i16 [ 0, [[ENTRY]] ], [ [[TMP22]], [[MIDDLE_BLOCK]] ]
+; PRED-NEXT:    br label [[LOOP:%.*]]
+; PRED:       loop:
+; PRED-NEXT:    [[IV:%.*]] = phi i64 [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ], [ [[IV_NEXT:%.*]], [[LOOP]] ]
+; PRED-NEXT:    [[RED:%.*]] = phi i16 [ [[BC_MERGE_RDX]], [[SCALAR_PH]] ], [ [[RED_NEXT:%.*]], [[LOOP]] ]
+; PRED-NEXT:    [[GEP:%.*]] = getelementptr i16, ptr [[SRC]], i64 [[IV]]
+; PRED-NEXT:    [[L:%.*]] = load i16, ptr [[GEP]], align 2
+; PRED-NEXT:    [[DIV:%.*]] = udiv i16 [[L]], [[X]]
+; PRED-NEXT:    [[RED_NEXT]] = or i16 [[DIV]], [[RED]]
+; PRED-NEXT:    [[IV_NEXT]] = add i64 [[IV]], 1
+; PRED-NEXT:    [[EC:%.*]] = icmp eq i64 [[IV]], [[N]]
+; PRED-NEXT:    br i1 [[EC]], label [[EXIT]], label [[LOOP]], !llvm.loop [[LOOP5:![0-9]+]]
+; PRED:       exit:
+; PRED-NEXT:    [[RED_NEXT_LCSSA:%.*]] = phi i16 [ [[RED_NEXT]], [[LOOP]] ], [ [[TMP22]], [[MIDDLE_BLOCK]] ]
+; PRED-NEXT:    ret i16 [[RED_NEXT_LCSSA]]
+;
+entry:
+  br label %loop
+
+loop:
+  %iv = phi i64 [ 0, %entry ], [ %iv.next, %loop ]
+  %red = phi i16 [ 0, %entry ], [ %red.next, %loop ]
+  %gep = getelementptr i16, ptr %src, i64 %iv
+  %l = load i16, ptr %gep, align 2
+  %div = udiv i16 %l, %x
+  %red.next = or i16 %div, %red
+  %iv.next = add i64 %iv, 1
+  %ec = icmp eq i64 %iv, %N
+  br i1 %ec, label %exit, label %loop
+
+exit:
+  ret i16 %red.next
+}
+
+attributes #0 = { "target-features"="+sve" }
+;.
+; DEFAULT: [[LOOP0]] = distinct !{[[LOOP0]], [[META1:![0-9]+]], [[META2:![0-9]+]]}
+; DEFAULT: [[META1]] = !{!"llvm.loop.isvectorized", i32 1}
+; DEFAULT: [[META2]] = !{!"llvm.loop.unroll.runtime.disable"}
+; DEFAULT: [[LOOP3]] = distinct !{[[LOOP3]], [[META2]], [[META1]]}
+; DEFAULT: [[LOOP4]] = distinct !{[[LOOP4]], [[META1]], [[META2]]}
+; DEFAULT: [[LOOP5]] = distinct !{[[LOOP5]], [[META2]], [[META1]]}
+;.
+; PRED: [[LOOP0]] = distinct !{[[LOOP0]], [[META1:![0-9]+]], [[META2:![0-9]+]]}
+; PRED: [[META1]] = !{!"llvm.loop.isvectorized", i32 1}
+; PRED: [[META2]] = !{!"llvm.loop.unroll.runtime.disable"}
+; PRED: [[LOOP3]] = distinct !{[[LOOP3]], [[META2]], [[META1]]}
+; PRED: [[LOOP4]] = distinct !{[[LOOP4]], [[META1]], [[META2]]}
+; PRED: [[LOOP5]] = distinct !{[[LOOP5]], [[META2]], [[META1]]}
+;.
diff --git a/llvm/test/Transforms/LoopVectorize/AArch64/scalable-strict-fadd.ll b/llvm/test/Transforms/LoopVectorize/AArch64/scalable-strict-fadd.ll
index ad6e8534f318..ddc004657ed5 100644
--- a/llvm/test/Transforms/LoopVectorize/AArch64/scalable-strict-fadd.ll
+++ b/llvm/test/Transforms/LoopVectorize/AArch64/scalable-strict-fadd.ll
@@ -588,7 +588,7 @@ define void @fadd_strict_interleave(ptr noalias nocapture readonly %a, ptr noali
 ; CHECK-UNORDERED-NEXT:    [[TMP10:%.*]] = getelementptr inbounds float, ptr [[B]], i64 [[TMP9]]
 ; CHECK-UNORDERED-NEXT:    [[TMP11:%.*]] = getelementptr inbounds float, ptr [[TMP10]], i32 0
 ; CHECK-UNORDERED-NEXT:    [[WIDE_VEC:%.*]] = load <vscale x 8 x float>, ptr [[TMP11]], align 4
-; CHECK-UNORDERED-NEXT:    [[STRIDED_VEC:%.*]] = call { <vscale x 4 x float>, <vscale x 4 x float> } @llvm.experimental.vector.deinterleave2.nxv8f32(<vscale x 8 x float> [[WIDE_VEC]])
+; CHECK-UNORDERED-NEXT:    [[STRIDED_VEC:%.*]] = call { <vscale x 4 x float>, <vscale x 4 x float> } @llvm.vector.deinterleave2.nxv8f32(<vscale x 8 x float> [[WIDE_VEC]])
 ; CHECK-UNORDERED-NEXT:    [[TMP12:%.*]] = extractvalue { <vscale x 4 x float>, <vscale x 4 x float> } [[STRIDED_VEC]], 0
 ; CHECK-UNORDERED-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 4 x float>, <vscale x 4 x float> } [[STRIDED_VEC]], 1
 ; CHECK-UNORDERED-NEXT:    [[TMP14]] = fadd <vscale x 4 x float> [[TMP12]], [[VEC_PHI1]]
@@ -658,7 +658,7 @@ define void @fadd_strict_interleave(ptr noalias nocapture readonly %a, ptr noali
 ; CHECK-ORDERED-NEXT:    [[TMP8:%.*]] = getelementptr inbounds float, ptr [[B]], i64 [[TMP7]]
 ; CHECK-ORDERED-NEXT:    [[TMP9:%.*]] = getelementptr inbounds float, ptr [[TMP8]], i32 0
 ; CHECK-ORDERED-NEXT:    [[WIDE_VEC:%.*]] = load <vscale x 8 x float>, ptr [[TMP9]], align 4
-; CHECK-ORDERED-NEXT:    [[STRIDED_VEC:%.*]] = call { <vscale x 4 x float>, <vscale x 4 x float> } @llvm.experimental.vector.deinterleave2.nxv8f32(<vscale x 8 x float> [[WIDE_VEC]])
+; CHECK-ORDERED-NEXT:    [[STRIDED_VEC:%.*]] = call { <vscale x 4 x float>, <vscale x 4 x float> } @llvm.vector.deinterleave2.nxv8f32(<vscale x 8 x float> [[WIDE_VEC]])
 ; CHECK-ORDERED-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 4 x float>, <vscale x 4 x float> } [[STRIDED_VEC]], 0
 ; CHECK-ORDERED-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 4 x float>, <vscale x 4 x float> } [[STRIDED_VEC]], 1
 ; CHECK-ORDERED-NEXT:    [[TMP12]] = call float @llvm.vector.reduce.fadd.nxv4f32(float [[VEC_PHI]], <vscale x 4 x float> [[TMP11]])
@@ -733,9 +733,9 @@ define void @fadd_strict_interleave(ptr noalias nocapture readonly %a, ptr noali
 ; CHECK-ORDERED-TF-NEXT:    [[TMP13:%.*]] = add i64 [[OFFSET_IDX]], 0
 ; CHECK-ORDERED-TF-NEXT:    [[TMP14:%.*]] = getelementptr inbounds float, ptr [[B]], i64 [[TMP13]]
 ; CHECK-ORDERED-TF-NEXT:    [[TMP15:%.*]] = getelementptr inbounds float, ptr [[TMP14]], i32 0
-; CHECK-ORDERED-TF-NEXT:    [[INTERLEAVED_MASK:%.*]] = call <vscale x 8 x i1> @llvm.experimental.vector.interleave2.nxv8i1(<vscale x 4 x i1> [[ACTIVE_LANE_MASK]], <vscale x 4 x i1> [[ACTIVE_LANE_MASK]])
+; CHECK-ORDERED-TF-NEXT:    [[INTERLEAVED_MASK:%.*]] = call <vscale x 8 x i1> @llvm.vector.interleave2.nxv8i1(<vscale x 4 x i1> [[ACTIVE_LANE_MASK]], <vscale x 4 x i1> [[ACTIVE_LANE_MASK]])
 ; CHECK-ORDERED-TF-NEXT:    [[WIDE_MASKED_VEC:%.*]] = call <vscale x 8 x float> @llvm.masked.load.nxv8f32.p0(ptr [[TMP15]], i32 4, <vscale x 8 x i1> [[INTERLEAVED_MASK]], <vscale x 8 x float> poison)
-; CHECK-ORDERED-TF-NEXT:    [[STRIDED_VEC:%.*]] = call { <vscale x 4 x float>, <vscale x 4 x float> } @llvm.experimental.vector.deinterleave2.nxv8f32(<vscale x 8 x float> [[WIDE_MASKED_VEC]])
+; CHECK-ORDERED-TF-NEXT:    [[STRIDED_VEC:%.*]] = call { <vscale x 4 x float>, <vscale x 4 x float> } @llvm.vector.deinterleave2.nxv8f32(<vscale x 8 x float> [[WIDE_MASKED_VEC]])
 ; CHECK-ORDERED-TF-NEXT:    [[TMP16:%.*]] = extractvalue { <vscale x 4 x float>, <vscale x 4 x float> } [[STRIDED_VEC]], 0
 ; CHECK-ORDERED-TF-NEXT:    [[TMP17:%.*]] = extractvalue { <vscale x 4 x float>, <vscale x 4 x float> } [[STRIDED_VEC]], 1
 ; CHECK-ORDERED-TF-NEXT:    [[TMP18:%.*]] = select <vscale x 4 x i1> [[ACTIVE_LANE_MASK]], <vscale x 4 x float> [[TMP17]], <vscale x 4 x float> shufflevector (<vscale x 4 x float> insertelement (<vscale x 4 x float> poison, float -0.000000e+00, i64 0), <vscale x 4 x float> poison, <vscale x 4 x i32> zeroinitializer)
diff --git a/llvm/test/Transforms/LoopVectorize/AArch64/store-costs-sve.ll b/llvm/test/Transforms/LoopVectorize/AArch64/store-costs-sve.ll
new file mode 100644
index 000000000000..1e13d70083ff
--- /dev/null
+++ b/llvm/test/Transforms/LoopVectorize/AArch64/store-costs-sve.ll
@@ -0,0 +1,341 @@
+; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --version 4
+; RUN: opt -p loop-vectorize -S %s | FileCheck --check-prefixes=DEFAULT %s
+; RUN: opt -p loop-vectorize -prefer-predicate-over-epilogue=predicate-else-scalar-epilogue -S %s | FileCheck --check-prefixes=PRED %s
+
+target datalayout = "e-m:o-i64:64-i128:128-n32:64-S128"
+target triple = "arm64-apple-macosx14.0.0"
+
+define void @cost_store_i8(ptr %dst) #0 {
+; DEFAULT-LABEL: define void @cost_store_i8(
+; DEFAULT-SAME: ptr [[DST:%.*]]) #[[ATTR0:[0-9]+]] {
+; DEFAULT-NEXT:  iter.check:
+; DEFAULT-NEXT:    [[TMP0:%.*]] = call i64 @llvm.vscale.i64()
+; DEFAULT-NEXT:    [[TMP1:%.*]] = mul i64 [[TMP0]], 8
+; DEFAULT-NEXT:    [[MIN_ITERS_CHECK:%.*]] = icmp ult i64 101, [[TMP1]]
+; DEFAULT-NEXT:    br i1 [[MIN_ITERS_CHECK]], label [[VEC_EPILOG_SCALAR_PH:%.*]], label [[VECTOR_MAIN_LOOP_ITER_CHECK:%.*]]
+; DEFAULT:       vector.main.loop.iter.check:
+; DEFAULT-NEXT:    [[TMP2:%.*]] = call i64 @llvm.vscale.i64()
+; DEFAULT-NEXT:    [[TMP3:%.*]] = mul i64 [[TMP2]], 16
+; DEFAULT-NEXT:    [[MIN_ITERS_CHECK1:%.*]] = icmp ult i64 101, [[TMP3]]
+; DEFAULT-NEXT:    br i1 [[MIN_ITERS_CHECK1]], label [[VEC_EPILOG_PH:%.*]], label [[VECTOR_PH:%.*]]
+; DEFAULT:       vector.ph:
+; DEFAULT-NEXT:    [[TMP4:%.*]] = call i64 @llvm.vscale.i64()
+; DEFAULT-NEXT:    [[TMP5:%.*]] = mul i64 [[TMP4]], 16
+; DEFAULT-NEXT:    [[N_MOD_VF:%.*]] = urem i64 101, [[TMP5]]
+; DEFAULT-NEXT:    [[N_VEC:%.*]] = sub i64 101, [[N_MOD_VF]]
+; DEFAULT-NEXT:    [[TMP6:%.*]] = call i64 @llvm.vscale.i64()
+; DEFAULT-NEXT:    [[TMP7:%.*]] = mul i64 [[TMP6]], 16
+; DEFAULT-NEXT:    br label [[VECTOR_BODY:%.*]]
+; DEFAULT:       vector.body:
+; DEFAULT-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
+; DEFAULT-NEXT:    [[TMP8:%.*]] = add i64 [[INDEX]], 0
+; DEFAULT-NEXT:    [[TMP9:%.*]] = getelementptr i8, ptr [[DST]], i64 [[TMP8]]
+; DEFAULT-NEXT:    [[TMP10:%.*]] = getelementptr i8, ptr [[TMP9]], i32 0
+; DEFAULT-NEXT:    store <vscale x 16 x i8> zeroinitializer, ptr [[TMP10]], align 1
+; DEFAULT-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], [[TMP7]]
+; DEFAULT-NEXT:    [[TMP11:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
+; DEFAULT-NEXT:    br i1 [[TMP11]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP0:![0-9]+]]
+; DEFAULT:       middle.block:
+; DEFAULT-NEXT:    [[CMP_N:%.*]] = icmp eq i64 101, [[N_VEC]]
+; DEFAULT-NEXT:    br i1 [[CMP_N]], label [[EXIT:%.*]], label [[VEC_EPILOG_ITER_CHECK:%.*]]
+; DEFAULT:       vec.epilog.iter.check:
+; DEFAULT-NEXT:    [[N_VEC_REMAINING:%.*]] = sub i64 101, [[N_VEC]]
+; DEFAULT-NEXT:    [[TMP12:%.*]] = call i64 @llvm.vscale.i64()
+; DEFAULT-NEXT:    [[TMP13:%.*]] = mul i64 [[TMP12]], 8
+; DEFAULT-NEXT:    [[MIN_EPILOG_ITERS_CHECK:%.*]] = icmp ult i64 [[N_VEC_REMAINING]], [[TMP13]]
+; DEFAULT-NEXT:    br i1 [[MIN_EPILOG_ITERS_CHECK]], label [[VEC_EPILOG_SCALAR_PH]], label [[VEC_EPILOG_PH]]
+; DEFAULT:       vec.epilog.ph:
+; DEFAULT-NEXT:    [[VEC_EPILOG_RESUME_VAL:%.*]] = phi i64 [ [[N_VEC]], [[VEC_EPILOG_ITER_CHECK]] ], [ 0, [[VECTOR_MAIN_LOOP_ITER_CHECK]] ]
+; DEFAULT-NEXT:    [[TMP14:%.*]] = call i64 @llvm.vscale.i64()
+; DEFAULT-NEXT:    [[TMP15:%.*]] = mul i64 [[TMP14]], 8
+; DEFAULT-NEXT:    [[N_MOD_VF2:%.*]] = urem i64 101, [[TMP15]]
+; DEFAULT-NEXT:    [[N_VEC3:%.*]] = sub i64 101, [[N_MOD_VF2]]
+; DEFAULT-NEXT:    [[TMP16:%.*]] = call i64 @llvm.vscale.i64()
+; DEFAULT-NEXT:    [[TMP17:%.*]] = mul i64 [[TMP16]], 8
+; DEFAULT-NEXT:    br label [[VEC_EPILOG_VECTOR_BODY:%.*]]
+; DEFAULT:       vec.epilog.vector.body:
+; DEFAULT-NEXT:    [[INDEX5:%.*]] = phi i64 [ [[VEC_EPILOG_RESUME_VAL]], [[VEC_EPILOG_PH]] ], [ [[INDEX_NEXT6:%.*]], [[VEC_EPILOG_VECTOR_BODY]] ]
+; DEFAULT-NEXT:    [[TMP18:%.*]] = add i64 [[INDEX5]], 0
+; DEFAULT-NEXT:    [[TMP19:%.*]] = getelementptr i8, ptr [[DST]], i64 [[TMP18]]
+; DEFAULT-NEXT:    [[TMP20:%.*]] = getelementptr i8, ptr [[TMP19]], i32 0
+; DEFAULT-NEXT:    store <vscale x 8 x i8> zeroinitializer, ptr [[TMP20]], align 1
+; DEFAULT-NEXT:    [[INDEX_NEXT6]] = add nuw i64 [[INDEX5]], [[TMP17]]
+; DEFAULT-NEXT:    [[TMP21:%.*]] = icmp eq i64 [[INDEX_NEXT6]], [[N_VEC3]]
+; DEFAULT-NEXT:    br i1 [[TMP21]], label [[VEC_EPILOG_MIDDLE_BLOCK:%.*]], label [[VEC_EPILOG_VECTOR_BODY]], !llvm.loop [[LOOP3:![0-9]+]]
+; DEFAULT:       vec.epilog.middle.block:
+; DEFAULT-NEXT:    [[CMP_N4:%.*]] = icmp eq i64 101, [[N_VEC3]]
+; DEFAULT-NEXT:    br i1 [[CMP_N4]], label [[EXIT]], label [[VEC_EPILOG_SCALAR_PH]]
+; DEFAULT:       vec.epilog.scalar.ph:
+; DEFAULT-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ [[N_VEC3]], [[VEC_EPILOG_MIDDLE_BLOCK]] ], [ [[N_VEC]], [[VEC_EPILOG_ITER_CHECK]] ], [ 0, [[ITER_CHECK:%.*]] ]
+; DEFAULT-NEXT:    br label [[LOOP:%.*]]
+; DEFAULT:       loop:
+; DEFAULT-NEXT:    [[IV:%.*]] = phi i64 [ [[BC_RESUME_VAL]], [[VEC_EPILOG_SCALAR_PH]] ], [ [[IV_NEXT:%.*]], [[LOOP]] ]
+; DEFAULT-NEXT:    [[GEP:%.*]] = getelementptr i8, ptr [[DST]], i64 [[IV]]
+; DEFAULT-NEXT:    store i8 0, ptr [[GEP]], align 1
+; DEFAULT-NEXT:    [[IV_NEXT]] = add i64 [[IV]], 1
+; DEFAULT-NEXT:    [[EC:%.*]] = icmp eq i64 [[IV]], 100
+; DEFAULT-NEXT:    br i1 [[EC]], label [[EXIT]], label [[LOOP]], !llvm.loop [[LOOP4:![0-9]+]]
+; DEFAULT:       exit:
+; DEFAULT-NEXT:    ret void
+;
+; PRED-LABEL: define void @cost_store_i8(
+; PRED-SAME: ptr [[DST:%.*]]) #[[ATTR0:[0-9]+]] {
+; PRED-NEXT:  entry:
+; PRED-NEXT:    br i1 false, label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
+; PRED:       vector.ph:
+; PRED-NEXT:    [[TMP0:%.*]] = call i64 @llvm.vscale.i64()
+; PRED-NEXT:    [[TMP1:%.*]] = mul i64 [[TMP0]], 16
+; PRED-NEXT:    [[TMP2:%.*]] = call i64 @llvm.vscale.i64()
+; PRED-NEXT:    [[TMP3:%.*]] = mul i64 [[TMP2]], 16
+; PRED-NEXT:    [[TMP4:%.*]] = sub i64 [[TMP3]], 1
+; PRED-NEXT:    [[N_RND_UP:%.*]] = add i64 101, [[TMP4]]
+; PRED-NEXT:    [[N_MOD_VF:%.*]] = urem i64 [[N_RND_UP]], [[TMP1]]
+; PRED-NEXT:    [[N_VEC:%.*]] = sub i64 [[N_RND_UP]], [[N_MOD_VF]]
+; PRED-NEXT:    [[TMP5:%.*]] = call i64 @llvm.vscale.i64()
+; PRED-NEXT:    [[TMP6:%.*]] = mul i64 [[TMP5]], 16
+; PRED-NEXT:    [[TMP7:%.*]] = call i64 @llvm.vscale.i64()
+; PRED-NEXT:    [[TMP8:%.*]] = mul i64 [[TMP7]], 16
+; PRED-NEXT:    [[TMP9:%.*]] = sub i64 101, [[TMP8]]
+; PRED-NEXT:    [[TMP10:%.*]] = icmp ugt i64 101, [[TMP8]]
+; PRED-NEXT:    [[TMP11:%.*]] = select i1 [[TMP10]], i64 [[TMP9]], i64 0
+; PRED-NEXT:    [[ACTIVE_LANE_MASK_ENTRY:%.*]] = call <vscale x 16 x i1> @llvm.get.active.lane.mask.nxv16i1.i64(i64 0, i64 101)
+; PRED-NEXT:    br label [[VECTOR_BODY:%.*]]
+; PRED:       vector.body:
+; PRED-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
+; PRED-NEXT:    [[ACTIVE_LANE_MASK:%.*]] = phi <vscale x 16 x i1> [ [[ACTIVE_LANE_MASK_ENTRY]], [[VECTOR_PH]] ], [ [[ACTIVE_LANE_MASK_NEXT:%.*]], [[VECTOR_BODY]] ]
+; PRED-NEXT:    [[TMP12:%.*]] = add i64 [[INDEX]], 0
+; PRED-NEXT:    [[TMP13:%.*]] = getelementptr i8, ptr [[DST]], i64 [[TMP12]]
+; PRED-NEXT:    [[TMP14:%.*]] = getelementptr i8, ptr [[TMP13]], i32 0
+; PRED-NEXT:    call void @llvm.masked.store.nxv16i8.p0(<vscale x 16 x i8> zeroinitializer, ptr [[TMP14]], i32 1, <vscale x 16 x i1> [[ACTIVE_LANE_MASK]])
+; PRED-NEXT:    [[INDEX_NEXT]] = add i64 [[INDEX]], [[TMP6]]
+; PRED-NEXT:    [[ACTIVE_LANE_MASK_NEXT]] = call <vscale x 16 x i1> @llvm.get.active.lane.mask.nxv16i1.i64(i64 [[INDEX]], i64 [[TMP11]])
+; PRED-NEXT:    [[TMP15:%.*]] = xor <vscale x 16 x i1> [[ACTIVE_LANE_MASK_NEXT]], shufflevector (<vscale x 16 x i1> insertelement (<vscale x 16 x i1> poison, i1 true, i64 0), <vscale x 16 x i1> poison, <vscale x 16 x i32> zeroinitializer)
+; PRED-NEXT:    [[TMP16:%.*]] = extractelement <vscale x 16 x i1> [[TMP15]], i32 0
+; PRED-NEXT:    br i1 [[TMP16]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP0:![0-9]+]]
+; PRED:       middle.block:
+; PRED-NEXT:    br i1 true, label [[EXIT:%.*]], label [[SCALAR_PH]]
+; PRED:       scalar.ph:
+; PRED-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ [[N_VEC]], [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ]
+; PRED-NEXT:    br label [[LOOP:%.*]]
+; PRED:       loop:
+; PRED-NEXT:    [[IV:%.*]] = phi i64 [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ], [ [[IV_NEXT:%.*]], [[LOOP]] ]
+; PRED-NEXT:    [[GEP:%.*]] = getelementptr i8, ptr [[DST]], i64 [[IV]]
+; PRED-NEXT:    store i8 0, ptr [[GEP]], align 1
+; PRED-NEXT:    [[IV_NEXT]] = add i64 [[IV]], 1
+; PRED-NEXT:    [[EC:%.*]] = icmp eq i64 [[IV]], 100
+; PRED-NEXT:    br i1 [[EC]], label [[EXIT]], label [[LOOP]], !llvm.loop [[LOOP3:![0-9]+]]
+; PRED:       exit:
+; PRED-NEXT:    ret void
+;
+entry:
+  br label %loop
+
+loop:
+  %iv = phi i64 [ 0, %entry ], [ %iv.next, %loop ]
+  %gep = getelementptr i8, ptr %dst, i64 %iv
+  store i8 0, ptr %gep, align 1
+  %iv.next = add i64 %iv, 1
+  %ec = icmp eq i64 %iv, 100
+  br i1 %ec, label %exit, label %loop
+
+exit:
+  ret void
+}
+
+define void @trunc_store(ptr %dst, ptr %src, i16 %x) #1 {
+; DEFAULT-LABEL: define void @trunc_store(
+; DEFAULT-SAME: ptr [[DST:%.*]], ptr [[SRC:%.*]], i16 [[X:%.*]]) #[[ATTR1:[0-9]+]] {
+; DEFAULT-NEXT:  iter.check:
+; DEFAULT-NEXT:    [[TMP0:%.*]] = call i64 @llvm.vscale.i64()
+; DEFAULT-NEXT:    [[TMP1:%.*]] = mul i64 [[TMP0]], 2
+; DEFAULT-NEXT:    [[MIN_ITERS_CHECK:%.*]] = icmp ult i64 0, [[TMP1]]
+; DEFAULT-NEXT:    br i1 [[MIN_ITERS_CHECK]], label [[VEC_EPILOG_SCALAR_PH:%.*]], label [[VECTOR_MEMCHECK:%.*]]
+; DEFAULT:       vector.memcheck:
+; DEFAULT-NEXT:    [[SCEVGEP:%.*]] = getelementptr i8, ptr [[SRC]], i64 8
+; DEFAULT-NEXT:    [[BOUND0:%.*]] = icmp ult ptr [[DST]], [[SCEVGEP]]
+; DEFAULT-NEXT:    [[BOUND1:%.*]] = icmp ult ptr [[SRC]], [[DST]]
+; DEFAULT-NEXT:    [[FOUND_CONFLICT:%.*]] = and i1 [[BOUND0]], [[BOUND1]]
+; DEFAULT-NEXT:    br i1 [[FOUND_CONFLICT]], label [[VEC_EPILOG_SCALAR_PH]], label [[VECTOR_MAIN_LOOP_ITER_CHECK:%.*]]
+; DEFAULT:       vector.main.loop.iter.check:
+; DEFAULT-NEXT:    br i1 true, label [[VEC_EPILOG_PH:%.*]], label [[VECTOR_PH:%.*]]
+; DEFAULT:       vector.ph:
+; DEFAULT-NEXT:    [[BROADCAST_SPLATINSERT3:%.*]] = insertelement <16 x i16> poison, i16 [[X]], i64 0
+; DEFAULT-NEXT:    [[BROADCAST_SPLAT4:%.*]] = shufflevector <16 x i16> [[BROADCAST_SPLATINSERT3]], <16 x i16> poison, <16 x i32> zeroinitializer
+; DEFAULT-NEXT:    br label [[VECTOR_BODY:%.*]]
+; DEFAULT:       vector.body:
+; DEFAULT-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
+; DEFAULT-NEXT:    [[TMP2:%.*]] = add i64 [[INDEX]], 0
+; DEFAULT-NEXT:    [[TMP3:%.*]] = add i64 [[INDEX]], 16
+; DEFAULT-NEXT:    [[TMP4:%.*]] = load i64, ptr [[SRC]], align 8, !alias.scope [[META5:![0-9]+]]
+; DEFAULT-NEXT:    [[BROADCAST_SPLATINSERT1:%.*]] = insertelement <16 x i64> poison, i64 [[TMP4]], i64 0
+; DEFAULT-NEXT:    [[BROADCAST_SPLAT2:%.*]] = shufflevector <16 x i64> [[BROADCAST_SPLATINSERT1]], <16 x i64> poison, <16 x i32> zeroinitializer
+; DEFAULT-NEXT:    [[TMP5:%.*]] = trunc <16 x i64> [[BROADCAST_SPLAT2]] to <16 x i8>
+; DEFAULT-NEXT:    [[TMP6:%.*]] = trunc <16 x i64> [[BROADCAST_SPLAT2]] to <16 x i8>
+; DEFAULT-NEXT:    [[TMP7:%.*]] = trunc <16 x i16> [[BROADCAST_SPLAT4]] to <16 x i8>
+; DEFAULT-NEXT:    [[TMP8:%.*]] = and <16 x i8> [[TMP5]], [[TMP7]]
+; DEFAULT-NEXT:    [[TMP9:%.*]] = and <16 x i8> [[TMP6]], [[TMP7]]
+; DEFAULT-NEXT:    [[TMP10:%.*]] = getelementptr i8, ptr [[DST]], i64 [[TMP2]]
+; DEFAULT-NEXT:    [[TMP11:%.*]] = getelementptr i8, ptr [[DST]], i64 [[TMP3]]
+; DEFAULT-NEXT:    [[TMP12:%.*]] = getelementptr i8, ptr [[TMP10]], i32 0
+; DEFAULT-NEXT:    [[TMP13:%.*]] = getelementptr i8, ptr [[TMP10]], i32 16
+; DEFAULT-NEXT:    store <16 x i8> [[TMP8]], ptr [[TMP12]], align 1, !alias.scope [[META8:![0-9]+]], !noalias [[META5]]
+; DEFAULT-NEXT:    store <16 x i8> [[TMP9]], ptr [[TMP13]], align 1, !alias.scope [[META8]], !noalias [[META5]]
+; DEFAULT-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 32
+; DEFAULT-NEXT:    [[TMP14:%.*]] = icmp eq i64 [[INDEX_NEXT]], 0
+; DEFAULT-NEXT:    br i1 [[TMP14]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP10:![0-9]+]]
+; DEFAULT:       middle.block:
+; DEFAULT-NEXT:    br i1 true, label [[EXIT:%.*]], label [[VEC_EPILOG_ITER_CHECK:%.*]]
+; DEFAULT:       vec.epilog.iter.check:
+; DEFAULT-NEXT:    [[TMP15:%.*]] = call i64 @llvm.vscale.i64()
+; DEFAULT-NEXT:    [[TMP16:%.*]] = mul i64 [[TMP15]], 2
+; DEFAULT-NEXT:    [[MIN_EPILOG_ITERS_CHECK:%.*]] = icmp ult i64 0, [[TMP16]]
+; DEFAULT-NEXT:    br i1 [[MIN_EPILOG_ITERS_CHECK]], label [[VEC_EPILOG_SCALAR_PH]], label [[VEC_EPILOG_PH]]
+; DEFAULT:       vec.epilog.ph:
+; DEFAULT-NEXT:    [[VEC_EPILOG_RESUME_VAL:%.*]] = phi i64 [ 0, [[VEC_EPILOG_ITER_CHECK]] ], [ 0, [[VECTOR_MAIN_LOOP_ITER_CHECK]] ]
+; DEFAULT-NEXT:    [[TMP17:%.*]] = call i64 @llvm.vscale.i64()
+; DEFAULT-NEXT:    [[TMP18:%.*]] = mul i64 [[TMP17]], 2
+; DEFAULT-NEXT:    [[N_MOD_VF:%.*]] = urem i64 0, [[TMP18]]
+; DEFAULT-NEXT:    [[N_VEC:%.*]] = sub i64 0, [[N_MOD_VF]]
+; DEFAULT-NEXT:    [[TMP19:%.*]] = call i64 @llvm.vscale.i64()
+; DEFAULT-NEXT:    [[TMP20:%.*]] = mul i64 [[TMP19]], 2
+; DEFAULT-NEXT:    [[BROADCAST_SPLATINSERT6:%.*]] = insertelement <vscale x 2 x i16> poison, i16 [[X]], i64 0
+; DEFAULT-NEXT:    [[BROADCAST_SPLAT7:%.*]] = shufflevector <vscale x 2 x i16> [[BROADCAST_SPLATINSERT6]], <vscale x 2 x i16> poison, <vscale x 2 x i32> zeroinitializer
+; DEFAULT-NEXT:    br label [[VEC_EPILOG_VECTOR_BODY:%.*]]
+; DEFAULT:       vec.epilog.vector.body:
+; DEFAULT-NEXT:    [[INDEX5:%.*]] = phi i64 [ [[VEC_EPILOG_RESUME_VAL]], [[VEC_EPILOG_PH]] ], [ [[INDEX_NEXT8:%.*]], [[VEC_EPILOG_VECTOR_BODY]] ]
+; DEFAULT-NEXT:    [[TMP21:%.*]] = add i64 [[INDEX5]], 0
+; DEFAULT-NEXT:    [[TMP22:%.*]] = load i64, ptr [[SRC]], align 8, !alias.scope [[META11:![0-9]+]]
+; DEFAULT-NEXT:    [[BROADCAST_SPLATINSERT:%.*]] = insertelement <vscale x 2 x i64> poison, i64 [[TMP22]], i64 0
+; DEFAULT-NEXT:    [[BROADCAST_SPLAT:%.*]] = shufflevector <vscale x 2 x i64> [[BROADCAST_SPLATINSERT]], <vscale x 2 x i64> poison, <vscale x 2 x i32> zeroinitializer
+; DEFAULT-NEXT:    [[TMP23:%.*]] = trunc <vscale x 2 x i64> [[BROADCAST_SPLAT]] to <vscale x 2 x i8>
+; DEFAULT-NEXT:    [[TMP24:%.*]] = trunc <vscale x 2 x i16> [[BROADCAST_SPLAT7]] to <vscale x 2 x i8>
+; DEFAULT-NEXT:    [[TMP25:%.*]] = and <vscale x 2 x i8> [[TMP23]], [[TMP24]]
+; DEFAULT-NEXT:    [[TMP26:%.*]] = getelementptr i8, ptr [[DST]], i64 [[TMP21]]
+; DEFAULT-NEXT:    [[TMP27:%.*]] = getelementptr i8, ptr [[TMP26]], i32 0
+; DEFAULT-NEXT:    store <vscale x 2 x i8> [[TMP25]], ptr [[TMP27]], align 1, !alias.scope [[META14:![0-9]+]], !noalias [[META11]]
+; DEFAULT-NEXT:    [[INDEX_NEXT8]] = add nuw i64 [[INDEX5]], [[TMP20]]
+; DEFAULT-NEXT:    [[TMP28:%.*]] = icmp eq i64 [[INDEX_NEXT8]], [[N_VEC]]
+; DEFAULT-NEXT:    br i1 [[TMP28]], label [[VEC_EPILOG_MIDDLE_BLOCK:%.*]], label [[VEC_EPILOG_VECTOR_BODY]], !llvm.loop [[LOOP16:![0-9]+]]
+; DEFAULT:       vec.epilog.middle.block:
+; DEFAULT-NEXT:    [[CMP_N:%.*]] = icmp eq i64 0, [[N_VEC]]
+; DEFAULT-NEXT:    br i1 [[CMP_N]], label [[EXIT]], label [[VEC_EPILOG_SCALAR_PH]]
+; DEFAULT:       vec.epilog.scalar.ph:
+; DEFAULT-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ [[N_VEC]], [[VEC_EPILOG_MIDDLE_BLOCK]] ], [ 0, [[VEC_EPILOG_ITER_CHECK]] ], [ 0, [[VECTOR_MEMCHECK]] ], [ 0, [[ITER_CHECK:%.*]] ]
+; DEFAULT-NEXT:    br label [[LOOP:%.*]]
+; DEFAULT:       loop:
+; DEFAULT-NEXT:    [[IV:%.*]] = phi i64 [ [[BC_RESUME_VAL]], [[VEC_EPILOG_SCALAR_PH]] ], [ [[IV_NEXT:%.*]], [[LOOP]] ]
+; DEFAULT-NEXT:    [[X_EXT:%.*]] = zext i16 [[X]] to i64
+; DEFAULT-NEXT:    [[L:%.*]] = load i64, ptr [[SRC]], align 8
+; DEFAULT-NEXT:    [[AND:%.*]] = and i64 [[L]], [[X_EXT]]
+; DEFAULT-NEXT:    [[TRUNC:%.*]] = trunc i64 [[AND]] to i8
+; DEFAULT-NEXT:    [[GEP:%.*]] = getelementptr i8, ptr [[DST]], i64 [[IV]]
+; DEFAULT-NEXT:    store i8 [[TRUNC]], ptr [[GEP]], align 1
+; DEFAULT-NEXT:    [[IV_NEXT]] = add i64 [[IV]], 1
+; DEFAULT-NEXT:    [[EC:%.*]] = icmp eq i64 [[IV_NEXT]], 0
+; DEFAULT-NEXT:    br i1 [[EC]], label [[EXIT]], label [[LOOP]], !llvm.loop [[LOOP17:![0-9]+]]
+; DEFAULT:       exit:
+; DEFAULT-NEXT:    ret void
+;
+; PRED-LABEL: define void @trunc_store(
+; PRED-SAME: ptr [[DST:%.*]], ptr [[SRC:%.*]], i16 [[X:%.*]]) #[[ATTR1:[0-9]+]] {
+; PRED-NEXT:  entry:
+; PRED-NEXT:    br i1 true, label [[SCALAR_PH:%.*]], label [[VECTOR_MEMCHECK:%.*]]
+; PRED:       vector.memcheck:
+; PRED-NEXT:    [[SCEVGEP:%.*]] = getelementptr i8, ptr [[SRC]], i64 8
+; PRED-NEXT:    [[BOUND0:%.*]] = icmp ult ptr [[DST]], [[SCEVGEP]]
+; PRED-NEXT:    [[BOUND1:%.*]] = icmp ult ptr [[SRC]], [[DST]]
+; PRED-NEXT:    [[FOUND_CONFLICT:%.*]] = and i1 [[BOUND0]], [[BOUND1]]
+; PRED-NEXT:    br i1 [[FOUND_CONFLICT]], label [[SCALAR_PH]], label [[VECTOR_PH:%.*]]
+; PRED:       vector.ph:
+; PRED-NEXT:    [[BROADCAST_SPLATINSERT1:%.*]] = insertelement <16 x i16> poison, i16 [[X]], i64 0
+; PRED-NEXT:    [[BROADCAST_SPLAT2:%.*]] = shufflevector <16 x i16> [[BROADCAST_SPLATINSERT1]], <16 x i16> poison, <16 x i32> zeroinitializer
+; PRED-NEXT:    br label [[VECTOR_BODY:%.*]]
+; PRED:       vector.body:
+; PRED-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
+; PRED-NEXT:    [[TMP0:%.*]] = add i64 [[INDEX]], 0
+; PRED-NEXT:    [[TMP1:%.*]] = load i64, ptr [[SRC]], align 8, !alias.scope [[META4:![0-9]+]]
+; PRED-NEXT:    [[BROADCAST_SPLATINSERT:%.*]] = insertelement <16 x i64> poison, i64 [[TMP1]], i64 0
+; PRED-NEXT:    [[BROADCAST_SPLAT:%.*]] = shufflevector <16 x i64> [[BROADCAST_SPLATINSERT]], <16 x i64> poison, <16 x i32> zeroinitializer
+; PRED-NEXT:    [[TMP2:%.*]] = trunc <16 x i64> [[BROADCAST_SPLAT]] to <16 x i8>
+; PRED-NEXT:    [[TMP3:%.*]] = trunc <16 x i16> [[BROADCAST_SPLAT2]] to <16 x i8>
+; PRED-NEXT:    [[TMP4:%.*]] = and <16 x i8> [[TMP2]], [[TMP3]]
+; PRED-NEXT:    [[TMP5:%.*]] = getelementptr i8, ptr [[DST]], i64 [[TMP0]]
+; PRED-NEXT:    [[TMP6:%.*]] = getelementptr i8, ptr [[TMP5]], i32 0
+; PRED-NEXT:    store <16 x i8> [[TMP4]], ptr [[TMP6]], align 1, !alias.scope [[META7:![0-9]+]], !noalias [[META4]]
+; PRED-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 16
+; PRED-NEXT:    [[TMP7:%.*]] = icmp eq i64 [[INDEX_NEXT]], 0
+; PRED-NEXT:    br i1 [[TMP7]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP9:![0-9]+]]
+; PRED:       middle.block:
+; PRED-NEXT:    br i1 true, label [[EXIT:%.*]], label [[SCALAR_PH]]
+; PRED:       scalar.ph:
+; PRED-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ 0, [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ], [ 0, [[VECTOR_MEMCHECK]] ]
+; PRED-NEXT:    br label [[LOOP:%.*]]
+; PRED:       loop:
+; PRED-NEXT:    [[IV:%.*]] = phi i64 [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ], [ [[IV_NEXT:%.*]], [[LOOP]] ]
+; PRED-NEXT:    [[X_EXT:%.*]] = zext i16 [[X]] to i64
+; PRED-NEXT:    [[L:%.*]] = load i64, ptr [[SRC]], align 8
+; PRED-NEXT:    [[AND:%.*]] = and i64 [[L]], [[X_EXT]]
+; PRED-NEXT:    [[TRUNC:%.*]] = trunc i64 [[AND]] to i8
+; PRED-NEXT:    [[GEP:%.*]] = getelementptr i8, ptr [[DST]], i64 [[IV]]
+; PRED-NEXT:    store i8 [[TRUNC]], ptr [[GEP]], align 1
+; PRED-NEXT:    [[IV_NEXT]] = add i64 [[IV]], 1
+; PRED-NEXT:    [[EC:%.*]] = icmp eq i64 [[IV_NEXT]], 0
+; PRED-NEXT:    br i1 [[EC]], label [[EXIT]], label [[LOOP]], !llvm.loop [[LOOP10:![0-9]+]]
+; PRED:       exit:
+; PRED-NEXT:    ret void
+;
+entry:
+  br label %loop
+
+loop:
+  %iv = phi i64 [ 0, %entry ], [ %iv.next, %loop ]
+  %x.ext = zext i16 %x to i64
+  %l = load i64, ptr %src, align 8
+  %and = and i64 %l, %x.ext
+  %trunc = trunc i64 %and to i8
+  %gep = getelementptr i8, ptr %dst, i64 %iv
+  store i8 %trunc, ptr %gep, align 1
+  %iv.next = add i64 %iv, 1
+  %ec = icmp eq i64 %iv.next, 0
+  br i1 %ec, label %exit, label %loop
+
+exit:
+  ret void
+}
+
+attributes #0 = { "target-features"="+sve" }
+attributes #1 = { vscale_range(1,16) "target-features"="+sve" }
+
+
+;.
+; DEFAULT: [[LOOP0]] = distinct !{[[LOOP0]], [[META1:![0-9]+]], [[META2:![0-9]+]]}
+; DEFAULT: [[META1]] = !{!"llvm.loop.isvectorized", i32 1}
+; DEFAULT: [[META2]] = !{!"llvm.loop.unroll.runtime.disable"}
+; DEFAULT: [[LOOP3]] = distinct !{[[LOOP3]], [[META1]], [[META2]]}
+; DEFAULT: [[LOOP4]] = distinct !{[[LOOP4]], [[META2]], [[META1]]}
+; DEFAULT: [[META5]] = !{[[META6:![0-9]+]]}
+; DEFAULT: [[META6]] = distinct !{[[META6]], [[META7:![0-9]+]]}
+; DEFAULT: [[META7]] = distinct !{[[META7]], !"LVerDomain"}
+; DEFAULT: [[META8]] = !{[[META9:![0-9]+]]}
+; DEFAULT: [[META9]] = distinct !{[[META9]], [[META7]]}
+; DEFAULT: [[LOOP10]] = distinct !{[[LOOP10]], [[META1]], [[META2]]}
+; DEFAULT: [[META11]] = !{[[META12:![0-9]+]]}
+; DEFAULT: [[META12]] = distinct !{[[META12]], [[META13:![0-9]+]]}
+; DEFAULT: [[META13]] = distinct !{[[META13]], !"LVerDomain"}
+; DEFAULT: [[META14]] = !{[[META15:![0-9]+]]}
+; DEFAULT: [[META15]] = distinct !{[[META15]], [[META13]]}
+; DEFAULT: [[LOOP16]] = distinct !{[[LOOP16]], [[META1]], [[META2]]}
+; DEFAULT: [[LOOP17]] = distinct !{[[LOOP17]], [[META1]]}
+;.
+; PRED: [[LOOP0]] = distinct !{[[LOOP0]], [[META1:![0-9]+]], [[META2:![0-9]+]]}
+; PRED: [[META1]] = !{!"llvm.loop.isvectorized", i32 1}
+; PRED: [[META2]] = !{!"llvm.loop.unroll.runtime.disable"}
+; PRED: [[LOOP3]] = distinct !{[[LOOP3]], [[META2]], [[META1]]}
+; PRED: [[META4]] = !{[[META5:![0-9]+]]}
+; PRED: [[META5]] = distinct !{[[META5]], [[META6:![0-9]+]]}
+; PRED: [[META6]] = distinct !{[[META6]], !"LVerDomain"}
+; PRED: [[META7]] = !{[[META8:![0-9]+]]}
+; PRED: [[META8]] = distinct !{[[META8]], [[META6]]}
+; PRED: [[LOOP9]] = distinct !{[[LOOP9]], [[META1]], [[META2]]}
+; PRED: [[LOOP10]] = distinct !{[[LOOP10]], [[META1]]}
+;.
diff --git a/llvm/test/Transforms/LoopVectorize/AArch64/sve-interleaved-accesses.ll b/llvm/test/Transforms/LoopVectorize/AArch64/sve-interleaved-accesses.ll
index c07b3c8d4922..1853e551806b 100644
--- a/llvm/test/Transforms/LoopVectorize/AArch64/sve-interleaved-accesses.ll
+++ b/llvm/test/Transforms/LoopVectorize/AArch64/sve-interleaved-accesses.ll
@@ -38,7 +38,7 @@ define void @test_array_load2_store2(i32 %C, i32 %D) #1 {
 ; CHECK-NEXT:    [[OFFSET_IDX:%.*]] = shl i64 [[INDEX]], 1
 ; CHECK-NEXT:    [[TMP2:%.*]] = getelementptr inbounds [1024 x i32], ptr @AB, i64 0, i64 [[OFFSET_IDX]]
 ; CHECK-NEXT:    [[WIDE_VEC:%.*]] = load <vscale x 8 x i32>, ptr [[TMP2]], align 4
-; CHECK-NEXT:    [[STRIDED_VEC:%.*]] = call { <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.experimental.vector.deinterleave2.nxv8i32(<vscale x 8 x i32> [[WIDE_VEC]])
+; CHECK-NEXT:    [[STRIDED_VEC:%.*]] = call { <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.vector.deinterleave2.nxv8i32(<vscale x 8 x i32> [[WIDE_VEC]])
 ; CHECK-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x i32>, <vscale x 4 x i32> } [[STRIDED_VEC]], 0
 ; CHECK-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x i32>, <vscale x 4 x i32> } [[STRIDED_VEC]], 1
 ; CHECK-NEXT:    [[TMP5:%.*]] = or disjoint i64 [[OFFSET_IDX]], 1
@@ -46,7 +46,7 @@ define void @test_array_load2_store2(i32 %C, i32 %D) #1 {
 ; CHECK-NEXT:    [[TMP7:%.*]] = mul nsw <vscale x 4 x i32> [[TMP4]], [[BROADCAST_SPLAT2]]
 ; CHECK-NEXT:    [[TMP8:%.*]] = getelementptr inbounds [1024 x i32], ptr @CD, i64 0, i64 [[TMP5]]
 ; CHECK-NEXT:    [[TMP9:%.*]] = getelementptr inbounds i8, ptr [[TMP8]], i64 -4
-; CHECK-NEXT:    [[INTERLEAVED_VEC:%.*]] = call <vscale x 8 x i32> @llvm.experimental.vector.interleave2.nxv8i32(<vscale x 4 x i32> [[TMP6]], <vscale x 4 x i32> [[TMP7]])
+; CHECK-NEXT:    [[INTERLEAVED_VEC:%.*]] = call <vscale x 8 x i32> @llvm.vector.interleave2.nxv8i32(<vscale x 4 x i32> [[TMP6]], <vscale x 4 x i32> [[TMP7]])
 ; CHECK-NEXT:    store <vscale x 8 x i32> [[INTERLEAVED_VEC]], ptr [[TMP9]], align 4
 ; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], [[TMP1]]
 ; CHECK-NEXT:    [[TMP10:%.*]] = icmp eq i64 [[INDEX_NEXT]], 512
@@ -134,7 +134,7 @@ define void @test_array_load2_i16_store2(i32 %C, i32 %D) #1 {
 ; CHECK-NEXT:    [[TMP13:%.*]] = extractelement <vscale x 4 x i64> [[TMP7]], i64 0
 ; CHECK-NEXT:    [[TMP14:%.*]] = getelementptr inbounds [1024 x i32], ptr @CD, i64 0, i64 [[TMP13]]
 ; CHECK-NEXT:    [[TMP15:%.*]] = getelementptr inbounds i8, ptr [[TMP14]], i64 -4
-; CHECK-NEXT:    [[INTERLEAVED_VEC:%.*]] = call <vscale x 8 x i32> @llvm.experimental.vector.interleave2.nxv8i32(<vscale x 4 x i32> [[TMP10]], <vscale x 4 x i32> [[TMP12]])
+; CHECK-NEXT:    [[INTERLEAVED_VEC:%.*]] = call <vscale x 8 x i32> @llvm.vector.interleave2.nxv8i32(<vscale x 4 x i32> [[TMP10]], <vscale x 4 x i32> [[TMP12]])
 ; CHECK-NEXT:    store <vscale x 8 x i32> [[INTERLEAVED_VEC]], ptr [[TMP15]], align 4
 ; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], [[TMP1]]
 ; CHECK-NEXT:    [[VEC_IND_NEXT]] = add <vscale x 4 x i64> [[VEC_IND]], [[DOTSPLAT]]
@@ -216,7 +216,7 @@ define void @test_array_load2_store2_i16(i32 noundef %C, i32 noundef %D) #1 {
 ; CHECK-NEXT:    [[OFFSET_IDX:%.*]] = shl i64 [[INDEX]], 1
 ; CHECK-NEXT:    [[TMP6:%.*]] = getelementptr inbounds [1024 x i32], ptr @AB, i64 0, i64 [[OFFSET_IDX]]
 ; CHECK-NEXT:    [[WIDE_VEC:%.*]] = load <vscale x 8 x i32>, ptr [[TMP6]], align 4
-; CHECK-NEXT:    [[STRIDED_VEC:%.*]] = call { <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.experimental.vector.deinterleave2.nxv8i32(<vscale x 8 x i32> [[WIDE_VEC]])
+; CHECK-NEXT:    [[STRIDED_VEC:%.*]] = call { <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.vector.deinterleave2.nxv8i32(<vscale x 8 x i32> [[WIDE_VEC]])
 ; CHECK-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 4 x i32>, <vscale x 4 x i32> } [[STRIDED_VEC]], 0
 ; CHECK-NEXT:    [[TMP8:%.*]] = extractvalue { <vscale x 4 x i32>, <vscale x 4 x i32> } [[STRIDED_VEC]], 1
 ; CHECK-NEXT:    [[TMP9:%.*]] = or disjoint <vscale x 4 x i64> [[VEC_IND]], shufflevector (<vscale x 4 x i64> insertelement (<vscale x 4 x i64> poison, i64 1, i64 0), <vscale x 4 x i64> poison, <vscale x 4 x i32> zeroinitializer)
@@ -401,11 +401,11 @@ define void @test_reversed_load2_store2(ptr noalias nocapture readonly %A, ptr n
 ; CHECK-NEXT:    [[TMP8:%.*]] = sext i32 [[TMP7]] to i64
 ; CHECK-NEXT:    [[TMP9:%.*]] = getelementptr inbounds i32, ptr [[TMP4]], i64 [[TMP8]]
 ; CHECK-NEXT:    [[WIDE_VEC:%.*]] = load <vscale x 8 x i32>, ptr [[TMP9]], align 4
-; CHECK-NEXT:    [[STRIDED_VEC:%.*]] = call { <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.experimental.vector.deinterleave2.nxv8i32(<vscale x 8 x i32> [[WIDE_VEC]])
+; CHECK-NEXT:    [[STRIDED_VEC:%.*]] = call { <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.vector.deinterleave2.nxv8i32(<vscale x 8 x i32> [[WIDE_VEC]])
 ; CHECK-NEXT:    [[TMP10:%.*]] = extractvalue { <vscale x 4 x i32>, <vscale x 4 x i32> } [[STRIDED_VEC]], 0
-; CHECK-NEXT:    [[REVERSE:%.*]] = call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> [[TMP10]])
+; CHECK-NEXT:    [[REVERSE:%.*]] = call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> [[TMP10]])
 ; CHECK-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 4 x i32>, <vscale x 4 x i32> } [[STRIDED_VEC]], 1
-; CHECK-NEXT:    [[REVERSE1:%.*]] = call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> [[TMP11]])
+; CHECK-NEXT:    [[REVERSE1:%.*]] = call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> [[TMP11]])
 ; CHECK-NEXT:    [[TMP12:%.*]] = add nsw <vscale x 4 x i32> [[REVERSE]], [[VEC_IND]]
 ; CHECK-NEXT:    [[TMP13:%.*]] = sub nsw <vscale x 4 x i32> [[REVERSE1]], [[VEC_IND]]
 ; CHECK-NEXT:    [[TMP14:%.*]] = getelementptr inbounds [[STRUCT_ST2]], ptr [[B:%.*]], i64 [[OFFSET_IDX]], i32 1
@@ -414,9 +414,9 @@ define void @test_reversed_load2_store2(ptr noalias nocapture readonly %A, ptr n
 ; CHECK-NEXT:    [[TMP17:%.*]] = sub nsw i32 1, [[TMP16]]
 ; CHECK-NEXT:    [[TMP18:%.*]] = sext i32 [[TMP17]] to i64
 ; CHECK-NEXT:    [[TMP19:%.*]] = getelementptr inbounds i32, ptr [[TMP14]], i64 [[TMP18]]
-; CHECK-NEXT:    [[REVERSE2:%.*]] = call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> [[TMP12]])
-; CHECK-NEXT:    [[REVERSE3:%.*]] = call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> [[TMP13]])
-; CHECK-NEXT:    [[INTERLEAVED_VEC:%.*]] = call <vscale x 8 x i32> @llvm.experimental.vector.interleave2.nxv8i32(<vscale x 4 x i32> [[REVERSE2]], <vscale x 4 x i32> [[REVERSE3]])
+; CHECK-NEXT:    [[REVERSE2:%.*]] = call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> [[TMP12]])
+; CHECK-NEXT:    [[REVERSE3:%.*]] = call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> [[TMP13]])
+; CHECK-NEXT:    [[INTERLEAVED_VEC:%.*]] = call <vscale x 8 x i32> @llvm.vector.interleave2.nxv8i32(<vscale x 4 x i32> [[REVERSE2]], <vscale x 4 x i32> [[REVERSE3]])
 ; CHECK-NEXT:    store <vscale x 8 x i32> [[INTERLEAVED_VEC]], ptr [[TMP19]], align 4
 ; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], [[TMP1]]
 ; CHECK-NEXT:    [[VEC_IND_NEXT]] = add <vscale x 4 x i32> [[VEC_IND]], [[DOTSPLAT]]
@@ -483,7 +483,7 @@ define void @even_load_static_tc(ptr noalias nocapture readonly %A, ptr noalias
 ; CHECK-NEXT:    [[OFFSET_IDX:%.*]] = shl i64 [[INDEX]], 1
 ; CHECK-NEXT:    [[TMP4:%.*]] = getelementptr inbounds i32, ptr [[A:%.*]], i64 [[OFFSET_IDX]]
 ; CHECK-NEXT:    [[WIDE_VEC:%.*]] = load <vscale x 8 x i32>, ptr [[TMP4]], align 4
-; CHECK-NEXT:    [[STRIDED_VEC:%.*]] = call { <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.experimental.vector.deinterleave2.nxv8i32(<vscale x 8 x i32> [[WIDE_VEC]])
+; CHECK-NEXT:    [[STRIDED_VEC:%.*]] = call { <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.vector.deinterleave2.nxv8i32(<vscale x 8 x i32> [[WIDE_VEC]])
 ; CHECK-NEXT:    [[TMP5:%.*]] = extractvalue { <vscale x 4 x i32>, <vscale x 4 x i32> } [[STRIDED_VEC]], 0
 ; CHECK-NEXT:    [[TMP6:%.*]] = shl nsw <vscale x 4 x i32> [[TMP5]], shufflevector (<vscale x 4 x i32> insertelement (<vscale x 4 x i32> poison, i32 1, i64 0), <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer)
 ; CHECK-NEXT:    [[TMP7:%.*]] = and i64 [[INDEX]], 9223372036854775804
@@ -569,7 +569,7 @@ define void @even_load_dynamic_tc(ptr noalias nocapture readonly %A, ptr noalias
 ; CHECK-NEXT:    [[OFFSET_IDX:%.*]] = shl i64 [[INDEX]], 1
 ; CHECK-NEXT:    [[TMP12:%.*]] = getelementptr inbounds i32, ptr [[A:%.*]], i64 [[OFFSET_IDX]]
 ; CHECK-NEXT:    [[WIDE_VEC:%.*]] = load <vscale x 8 x i32>, ptr [[TMP12]], align 4
-; CHECK-NEXT:    [[STRIDED_VEC:%.*]] = call { <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.experimental.vector.deinterleave2.nxv8i32(<vscale x 8 x i32> [[WIDE_VEC]])
+; CHECK-NEXT:    [[STRIDED_VEC:%.*]] = call { <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.vector.deinterleave2.nxv8i32(<vscale x 8 x i32> [[WIDE_VEC]])
 ; CHECK-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 4 x i32>, <vscale x 4 x i32> } [[STRIDED_VEC]], 0
 ; CHECK-NEXT:    [[TMP14:%.*]] = shl nsw <vscale x 4 x i32> [[TMP13]], shufflevector (<vscale x 4 x i32> insertelement (<vscale x 4 x i32> poison, i32 1, i64 0), <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer)
 ; CHECK-NEXT:    [[TMP15:%.*]] = and i64 [[INDEX]], 9223372036854775804
@@ -717,18 +717,18 @@ define void @mixed_load2_store2(ptr noalias nocapture readonly %A, ptr noalias n
 ; CHECK-NEXT:    [[OFFSET_IDX:%.*]] = shl i64 [[INDEX]], 1
 ; CHECK-NEXT:    [[TMP2:%.*]] = getelementptr inbounds i32, ptr [[A:%.*]], i64 [[OFFSET_IDX]]
 ; CHECK-NEXT:    [[WIDE_VEC:%.*]] = load <vscale x 8 x i32>, ptr [[TMP2]], align 4
-; CHECK-NEXT:    [[STRIDED_VEC:%.*]] = call { <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.experimental.vector.deinterleave2.nxv8i32(<vscale x 8 x i32> [[WIDE_VEC]])
+; CHECK-NEXT:    [[STRIDED_VEC:%.*]] = call { <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.vector.deinterleave2.nxv8i32(<vscale x 8 x i32> [[WIDE_VEC]])
 ; CHECK-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x i32>, <vscale x 4 x i32> } [[STRIDED_VEC]], 0
 ; CHECK-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x i32>, <vscale x 4 x i32> } [[STRIDED_VEC]], 1
 ; CHECK-NEXT:    [[TMP5:%.*]] = or disjoint i64 [[OFFSET_IDX]], 1
 ; CHECK-NEXT:    [[TMP6:%.*]] = mul nsw <vscale x 4 x i32> [[TMP4]], [[TMP3]]
-; CHECK-NEXT:    [[STRIDED_VEC2:%.*]] = call { <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.experimental.vector.deinterleave2.nxv8i32(<vscale x 8 x i32> [[WIDE_VEC]])
+; CHECK-NEXT:    [[STRIDED_VEC2:%.*]] = call { <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.vector.deinterleave2.nxv8i32(<vscale x 8 x i32> [[WIDE_VEC]])
 ; CHECK-NEXT:    [[TMP7:%.*]] = extractvalue { <vscale x 4 x i32>, <vscale x 4 x i32> } [[STRIDED_VEC2]], 0
 ; CHECK-NEXT:    [[TMP8:%.*]] = extractvalue { <vscale x 4 x i32>, <vscale x 4 x i32> } [[STRIDED_VEC2]], 1
 ; CHECK-NEXT:    [[TMP9:%.*]] = add nsw <vscale x 4 x i32> [[TMP8]], [[TMP7]]
 ; CHECK-NEXT:    [[TMP10:%.*]] = getelementptr inbounds i32, ptr [[B:%.*]], i64 [[TMP5]]
 ; CHECK-NEXT:    [[TMP11:%.*]] = getelementptr inbounds i8, ptr [[TMP10]], i64 -4
-; CHECK-NEXT:    [[INTERLEAVED_VEC:%.*]] = call <vscale x 8 x i32> @llvm.experimental.vector.interleave2.nxv8i32(<vscale x 4 x i32> [[TMP6]], <vscale x 4 x i32> [[TMP9]])
+; CHECK-NEXT:    [[INTERLEAVED_VEC:%.*]] = call <vscale x 8 x i32> @llvm.vector.interleave2.nxv8i32(<vscale x 4 x i32> [[TMP6]], <vscale x 4 x i32> [[TMP9]])
 ; CHECK-NEXT:    store <vscale x 8 x i32> [[INTERLEAVED_VEC]], ptr [[TMP11]], align 4
 ; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], [[TMP1]]
 ; CHECK-NEXT:    [[TMP12:%.*]] = icmp eq i64 [[INDEX_NEXT]], 512
@@ -811,7 +811,7 @@ define void @int_float_struct(ptr nocapture readonly %p) #0 {
 ; CHECK-NEXT:    [[VEC_PHI1:%.*]] = phi <vscale x 4 x i32> [ insertelement (<vscale x 4 x i32> zeroinitializer, i32 undef, i32 0), [[VECTOR_PH]] ], [ [[TMP6:%.*]], [[VECTOR_BODY]] ]
 ; CHECK-NEXT:    [[TMP2:%.*]] = getelementptr inbounds [[STRUCT_INTFLOAT:%.*]], ptr [[P:%.*]], i64 [[INDEX]], i32 0
 ; CHECK-NEXT:    [[WIDE_VEC:%.*]] = load <vscale x 8 x i32>, ptr [[TMP2]], align 4
-; CHECK-NEXT:    [[STRIDED_VEC:%.*]] = call { <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.experimental.vector.deinterleave2.nxv8i32(<vscale x 8 x i32> [[WIDE_VEC]])
+; CHECK-NEXT:    [[STRIDED_VEC:%.*]] = call { <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.vector.deinterleave2.nxv8i32(<vscale x 8 x i32> [[WIDE_VEC]])
 ; CHECK-NEXT:    [[TMP3:%.*]] = extractvalue { <vscale x 4 x i32>, <vscale x 4 x i32> } [[STRIDED_VEC]], 0
 ; CHECK-NEXT:    [[TMP4:%.*]] = extractvalue { <vscale x 4 x i32>, <vscale x 4 x i32> } [[STRIDED_VEC]], 1
 ; CHECK-NEXT:    [[TMP5:%.*]] = bitcast <vscale x 4 x i32> [[TMP4]] to <vscale x 4 x float>
@@ -910,7 +910,7 @@ define void @PR27626_0(ptr %p, i32 %z, i64 %n) #1 {
 ; CHECK-NEXT:    call void @llvm.masked.scatter.nxv4i32.nxv4p0(<vscale x 4 x i32> [[BROADCAST_SPLAT]], <vscale x 4 x ptr> [[TMP12]], i32 4, <vscale x 4 x i1> shufflevector (<vscale x 4 x i1> insertelement (<vscale x 4 x i1> poison, i1 true, i64 0), <vscale x 4 x i1> poison, <vscale x 4 x i32> zeroinitializer))
 ; CHECK-NEXT:    [[TMP14:%.*]] = extractelement <vscale x 4 x ptr> [[TMP12]], i64 0
 ; CHECK-NEXT:    [[WIDE_VEC:%.*]] = load <vscale x 8 x i32>, ptr [[TMP14]], align 4
-; CHECK-NEXT:    [[STRIDED_VEC:%.*]] = call { <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.experimental.vector.deinterleave2.nxv8i32(<vscale x 8 x i32> [[WIDE_VEC]])
+; CHECK-NEXT:    [[STRIDED_VEC:%.*]] = call { <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.vector.deinterleave2.nxv8i32(<vscale x 8 x i32> [[WIDE_VEC]])
 ; CHECK-NEXT:    [[TMP15:%.*]] = extractvalue { <vscale x 4 x i32>, <vscale x 4 x i32> } [[STRIDED_VEC]], 0
 ; CHECK-NEXT:    call void @llvm.masked.scatter.nxv4i32.nxv4p0(<vscale x 4 x i32> [[TMP15]], <vscale x 4 x ptr> [[TMP13]], i32 4, <vscale x 4 x i1> shufflevector (<vscale x 4 x i1> insertelement (<vscale x 4 x i1> poison, i1 true, i64 0), <vscale x 4 x i1> poison, <vscale x 4 x i32> zeroinitializer))
 ; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], [[TMP8]]
@@ -994,12 +994,12 @@ define i32 @PR27626_1(ptr %p, i64 %n) #1 {
 ; CHECK-NEXT:    [[TMP12:%.*]] = getelementptr inbounds [[PAIR_I32:%.*]], ptr [[P:%.*]], i64 [[INDEX]], i32 0
 ; CHECK-NEXT:    [[TMP13:%.*]] = getelementptr inbounds [[PAIR_I32]], ptr [[P]], <vscale x 4 x i64> [[VEC_IND]], i32 1
 ; CHECK-NEXT:    [[WIDE_VEC:%.*]] = load <vscale x 8 x i32>, ptr [[TMP12]], align 4
-; CHECK-NEXT:    [[STRIDED_VEC:%.*]] = call { <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.experimental.vector.deinterleave2.nxv8i32(<vscale x 8 x i32> [[WIDE_VEC]])
+; CHECK-NEXT:    [[STRIDED_VEC:%.*]] = call { <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.vector.deinterleave2.nxv8i32(<vscale x 8 x i32> [[WIDE_VEC]])
 ; CHECK-NEXT:    [[TMP14:%.*]] = extractvalue { <vscale x 4 x i32>, <vscale x 4 x i32> } [[STRIDED_VEC]], 0
 ; CHECK-NEXT:    call void @llvm.masked.scatter.nxv4i32.nxv4p0(<vscale x 4 x i32> [[TMP14]], <vscale x 4 x ptr> [[TMP13]], i32 4, <vscale x 4 x i1> shufflevector (<vscale x 4 x i1> insertelement (<vscale x 4 x i1> poison, i1 true, i64 0), <vscale x 4 x i1> poison, <vscale x 4 x i32> zeroinitializer))
 ; CHECK-NEXT:    [[TMP15:%.*]] = extractelement <vscale x 4 x ptr> [[TMP13]], i64 0
 ; CHECK-NEXT:    [[WIDE_VEC1:%.*]] = load <vscale x 8 x i32>, ptr [[TMP15]], align 4
-; CHECK-NEXT:    [[STRIDED_VEC2:%.*]] = call { <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.experimental.vector.deinterleave2.nxv8i32(<vscale x 8 x i32> [[WIDE_VEC1]])
+; CHECK-NEXT:    [[STRIDED_VEC2:%.*]] = call { <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.vector.deinterleave2.nxv8i32(<vscale x 8 x i32> [[WIDE_VEC1]])
 ; CHECK-NEXT:    [[TMP16:%.*]] = extractvalue { <vscale x 4 x i32>, <vscale x 4 x i32> } [[STRIDED_VEC2]], 0
 ; CHECK-NEXT:    [[TMP17]] = add <vscale x 4 x i32> [[TMP16]], [[VEC_PHI]]
 ; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], [[TMP8]]
@@ -1092,7 +1092,7 @@ define void @PR27626_2(ptr %p, i64 %n, i32 %z) #1 {
 ; CHECK-NEXT:    [[TMP14:%.*]] = getelementptr inbounds [[PAIR_I32]], ptr [[P]], <vscale x 4 x i64> [[VEC_IND]], i32 1
 ; CHECK-NEXT:    call void @llvm.masked.scatter.nxv4i32.nxv4p0(<vscale x 4 x i32> [[BROADCAST_SPLAT]], <vscale x 4 x ptr> [[TMP12]], i32 4, <vscale x 4 x i1> shufflevector (<vscale x 4 x i1> insertelement (<vscale x 4 x i1> poison, i1 true, i64 0), <vscale x 4 x i1> poison, <vscale x 4 x i32> zeroinitializer))
 ; CHECK-NEXT:    [[WIDE_VEC:%.*]] = load <vscale x 8 x i32>, ptr [[TMP13]], align 4
-; CHECK-NEXT:    [[STRIDED_VEC:%.*]] = call { <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.experimental.vector.deinterleave2.nxv8i32(<vscale x 8 x i32> [[WIDE_VEC]])
+; CHECK-NEXT:    [[STRIDED_VEC:%.*]] = call { <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.vector.deinterleave2.nxv8i32(<vscale x 8 x i32> [[WIDE_VEC]])
 ; CHECK-NEXT:    [[TMP15:%.*]] = extractvalue { <vscale x 4 x i32>, <vscale x 4 x i32> } [[STRIDED_VEC]], 0
 ; CHECK-NEXT:    call void @llvm.masked.scatter.nxv4i32.nxv4p0(<vscale x 4 x i32> [[TMP15]], <vscale x 4 x ptr> [[TMP14]], i32 4, <vscale x 4 x i1> shufflevector (<vscale x 4 x i1> insertelement (<vscale x 4 x i1> poison, i1 true, i64 0), <vscale x 4 x i1> poison, <vscale x 4 x i32> zeroinitializer))
 ; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], [[TMP8]]
@@ -1181,11 +1181,11 @@ define i32 @PR27626_3(ptr %p, i64 %n, i32 %z) #1 {
 ; CHECK-NEXT:    [[TMP14:%.*]] = getelementptr inbounds [[PAIR_I32]], ptr [[P]], i64 [[INDEX]], i32 1
 ; CHECK-NEXT:    [[TMP15:%.*]] = getelementptr inbounds [[PAIR_I32]], ptr [[P]], <vscale x 4 x i64> [[TMP12]], i32 1
 ; CHECK-NEXT:    [[WIDE_VEC:%.*]] = load <vscale x 8 x i32>, ptr [[TMP13]], align 4
-; CHECK-NEXT:    [[STRIDED_VEC:%.*]] = call { <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.experimental.vector.deinterleave2.nxv8i32(<vscale x 8 x i32> [[WIDE_VEC]])
+; CHECK-NEXT:    [[STRIDED_VEC:%.*]] = call { <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.vector.deinterleave2.nxv8i32(<vscale x 8 x i32> [[WIDE_VEC]])
 ; CHECK-NEXT:    [[TMP16:%.*]] = extractvalue { <vscale x 4 x i32>, <vscale x 4 x i32> } [[STRIDED_VEC]], 0
 ; CHECK-NEXT:    call void @llvm.masked.scatter.nxv4i32.nxv4p0(<vscale x 4 x i32> [[TMP16]], <vscale x 4 x ptr> [[TMP15]], i32 4, <vscale x 4 x i1> shufflevector (<vscale x 4 x i1> insertelement (<vscale x 4 x i1> poison, i1 true, i64 0), <vscale x 4 x i1> poison, <vscale x 4 x i32> zeroinitializer))
 ; CHECK-NEXT:    [[WIDE_VEC1:%.*]] = load <vscale x 8 x i32>, ptr [[TMP14]], align 4
-; CHECK-NEXT:    [[STRIDED_VEC2:%.*]] = call { <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.experimental.vector.deinterleave2.nxv8i32(<vscale x 8 x i32> [[WIDE_VEC1]])
+; CHECK-NEXT:    [[STRIDED_VEC2:%.*]] = call { <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.vector.deinterleave2.nxv8i32(<vscale x 8 x i32> [[WIDE_VEC1]])
 ; CHECK-NEXT:    [[TMP17:%.*]] = extractvalue { <vscale x 4 x i32>, <vscale x 4 x i32> } [[STRIDED_VEC2]], 0
 ; CHECK-NEXT:    [[TMP18]] = add <vscale x 4 x i32> [[TMP17]], [[VEC_PHI]]
 ; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], [[TMP8]]
@@ -1291,7 +1291,7 @@ define void @PR27626_4(ptr %a, i32 %x, i32 %y, i32 %z, i64 %n) #1 {
 ; CHECK-NEXT:    [[TMP14:%.*]] = getelementptr inbounds i32, ptr [[A]], i64 [[TMP12]]
 ; CHECK-NEXT:    call void @llvm.masked.scatter.nxv4i32.nxv4p0(<vscale x 4 x i32> [[BROADCAST_SPLAT]], <vscale x 4 x ptr> [[TMP13]], i32 4, <vscale x 4 x i1> shufflevector (<vscale x 4 x i1> insertelement (<vscale x 4 x i1> poison, i1 true, i64 0), <vscale x 4 x i1> poison, <vscale x 4 x i32> zeroinitializer))
 ; CHECK-NEXT:    [[TMP15:%.*]] = getelementptr inbounds i8, ptr [[TMP14]], i64 -4
-; CHECK-NEXT:    [[INTERLEAVED_VEC:%.*]] = call <vscale x 8 x i32> @llvm.experimental.vector.interleave2.nxv8i32(<vscale x 4 x i32> [[BROADCAST_SPLAT2]], <vscale x 4 x i32> [[BROADCAST_SPLAT4]])
+; CHECK-NEXT:    [[INTERLEAVED_VEC:%.*]] = call <vscale x 8 x i32> @llvm.vector.interleave2.nxv8i32(<vscale x 4 x i32> [[BROADCAST_SPLAT2]], <vscale x 4 x i32> [[BROADCAST_SPLAT4]])
 ; CHECK-NEXT:    store <vscale x 8 x i32> [[INTERLEAVED_VEC]], ptr [[TMP15]], align 4
 ; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], [[TMP7]]
 ; CHECK-NEXT:    [[VEC_IND_NEXT]] = add <vscale x 4 x i64> [[VEC_IND]], [[DOTSPLAT]]
@@ -1497,7 +1497,7 @@ define void @PR34743(ptr %a, ptr %b, i64 %n) #1 {
 ; CHECK-NEXT:    [[TMP21:%.*]] = sext <vscale x 4 x i16> [[WIDE_MASKED_GATHER]] to <vscale x 4 x i32>
 ; CHECK-NEXT:    [[TMP22:%.*]] = getelementptr inbounds i16, ptr [[A]], <vscale x 4 x i64> [[TMP19]]
 ; CHECK-NEXT:    [[WIDE_MASKED_GATHER4]] = call <vscale x 4 x i16> @llvm.masked.gather.nxv4i16.nxv4p0(<vscale x 4 x ptr> [[TMP22]], i32 4, <vscale x 4 x i1> shufflevector (<vscale x 4 x i1> insertelement (<vscale x 4 x i1> poison, i1 true, i64 0), <vscale x 4 x i1> poison, <vscale x 4 x i32> zeroinitializer), <vscale x 4 x i16> poison), !alias.scope [[META34]]
-; CHECK-NEXT:    [[TMP23:%.*]] = call <vscale x 4 x i16> @llvm.experimental.vector.splice.nxv4i16(<vscale x 4 x i16> [[VECTOR_RECUR]], <vscale x 4 x i16> [[WIDE_MASKED_GATHER4]], i32 -1)
+; CHECK-NEXT:    [[TMP23:%.*]] = call <vscale x 4 x i16> @llvm.vector.splice.nxv4i16(<vscale x 4 x i16> [[VECTOR_RECUR]], <vscale x 4 x i16> [[WIDE_MASKED_GATHER4]], i32 -1)
 ; CHECK-NEXT:    [[TMP24:%.*]] = sext <vscale x 4 x i16> [[TMP23]] to <vscale x 4 x i32>
 ; CHECK-NEXT:    [[TMP25:%.*]] = sext <vscale x 4 x i16> [[WIDE_MASKED_GATHER4]] to <vscale x 4 x i32>
 ; CHECK-NEXT:    [[TMP26:%.*]] = mul nsw <vscale x 4 x i32> [[TMP24]], [[TMP21]]
diff --git a/llvm/test/Transforms/LoopVectorize/AArch64/sve-interleaved-masked-accesses.ll b/llvm/test/Transforms/LoopVectorize/AArch64/sve-interleaved-masked-accesses.ll
index 3ba91360850e..726d98f4d37d 100644
--- a/llvm/test/Transforms/LoopVectorize/AArch64/sve-interleaved-masked-accesses.ll
+++ b/llvm/test/Transforms/LoopVectorize/AArch64/sve-interleaved-masked-accesses.ll
@@ -52,9 +52,9 @@ define dso_local void @masked_strided1(ptr noalias nocapture readonly %p, ptr no
 ; SCALAR_TAIL_FOLDING-NEXT:    [[TMP8:%.*]] = shl i32 [[INDEX]], 1
 ; SCALAR_TAIL_FOLDING-NEXT:    [[TMP9:%.*]] = sext i32 [[TMP8]] to i64
 ; SCALAR_TAIL_FOLDING-NEXT:    [[TMP10:%.*]] = getelementptr i8, ptr [[P]], i64 [[TMP9]]
-; SCALAR_TAIL_FOLDING-NEXT:    [[INTERLEAVED_MASK:%.*]] = call <vscale x 32 x i1> @llvm.experimental.vector.interleave2.nxv32i1(<vscale x 16 x i1> [[TMP7]], <vscale x 16 x i1> [[TMP7]])
+; SCALAR_TAIL_FOLDING-NEXT:    [[INTERLEAVED_MASK:%.*]] = call <vscale x 32 x i1> @llvm.vector.interleave2.nxv32i1(<vscale x 16 x i1> [[TMP7]], <vscale x 16 x i1> [[TMP7]])
 ; SCALAR_TAIL_FOLDING-NEXT:    [[WIDE_MASKED_VEC:%.*]] = call <vscale x 32 x i8> @llvm.masked.load.nxv32i8.p0(ptr [[TMP10]], i32 1, <vscale x 32 x i1> [[INTERLEAVED_MASK]], <vscale x 32 x i8> poison)
-; SCALAR_TAIL_FOLDING-NEXT:    [[STRIDED_VEC:%.*]] = call { <vscale x 16 x i8>, <vscale x 16 x i8> } @llvm.experimental.vector.deinterleave2.nxv32i8(<vscale x 32 x i8> [[WIDE_MASKED_VEC]])
+; SCALAR_TAIL_FOLDING-NEXT:    [[STRIDED_VEC:%.*]] = call { <vscale x 16 x i8>, <vscale x 16 x i8> } @llvm.vector.deinterleave2.nxv32i8(<vscale x 32 x i8> [[WIDE_MASKED_VEC]])
 ; SCALAR_TAIL_FOLDING-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 16 x i8>, <vscale x 16 x i8> } [[STRIDED_VEC]], 0
 ; SCALAR_TAIL_FOLDING-NEXT:    [[TMP12:%.*]] = extractvalue { <vscale x 16 x i8>, <vscale x 16 x i8> } [[STRIDED_VEC]], 1
 ; SCALAR_TAIL_FOLDING-NEXT:    [[TMP13:%.*]] = or disjoint i32 [[TMP8]], 1
@@ -63,8 +63,8 @@ define dso_local void @masked_strided1(ptr noalias nocapture readonly %p, ptr no
 ; SCALAR_TAIL_FOLDING-NEXT:    [[TMP16:%.*]] = sext i32 [[TMP13]] to i64
 ; SCALAR_TAIL_FOLDING-NEXT:    [[TMP17:%.*]] = getelementptr i8, ptr [[Q]], i64 [[TMP16]]
 ; SCALAR_TAIL_FOLDING-NEXT:    [[TMP18:%.*]] = getelementptr i8, ptr [[TMP17]], i64 -1
-; SCALAR_TAIL_FOLDING-NEXT:    [[INTERLEAVED_VEC:%.*]] = call <vscale x 32 x i8> @llvm.experimental.vector.interleave2.nxv32i8(<vscale x 16 x i8> [[TMP14]], <vscale x 16 x i8> [[TMP15]])
-; SCALAR_TAIL_FOLDING-NEXT:    [[INTERLEAVED_MASK1:%.*]] = call <vscale x 32 x i1> @llvm.experimental.vector.interleave2.nxv32i1(<vscale x 16 x i1> [[TMP7]], <vscale x 16 x i1> [[TMP7]])
+; SCALAR_TAIL_FOLDING-NEXT:    [[INTERLEAVED_VEC:%.*]] = call <vscale x 32 x i8> @llvm.vector.interleave2.nxv32i8(<vscale x 16 x i8> [[TMP14]], <vscale x 16 x i8> [[TMP15]])
+; SCALAR_TAIL_FOLDING-NEXT:    [[INTERLEAVED_MASK1:%.*]] = call <vscale x 32 x i1> @llvm.vector.interleave2.nxv32i1(<vscale x 16 x i1> [[TMP7]], <vscale x 16 x i1> [[TMP7]])
 ; SCALAR_TAIL_FOLDING-NEXT:    call void @llvm.masked.store.nxv32i8.p0(<vscale x 32 x i8> [[INTERLEAVED_VEC]], ptr [[TMP18]], i32 1, <vscale x 32 x i1> [[INTERLEAVED_MASK1]])
 ; SCALAR_TAIL_FOLDING-NEXT:    [[INDEX_NEXT]] = add nuw i32 [[INDEX]], [[TMP20]]
 ; SCALAR_TAIL_FOLDING-NEXT:    [[VEC_IND_NEXT]] = add <vscale x 16 x i32> [[VEC_IND]], [[DOTSPLAT]]
@@ -134,9 +134,9 @@ define dso_local void @masked_strided1(ptr noalias nocapture readonly %p, ptr no
 ; PREDICATED_TAIL_FOLDING-NEXT:    [[TMP7:%.*]] = shl i32 [[INDEX]], 1
 ; PREDICATED_TAIL_FOLDING-NEXT:    [[TMP8:%.*]] = sext i32 [[TMP7]] to i64
 ; PREDICATED_TAIL_FOLDING-NEXT:    [[TMP9:%.*]] = getelementptr i8, ptr [[P]], i64 [[TMP8]]
-; PREDICATED_TAIL_FOLDING-NEXT:    [[INTERLEAVED_MASK:%.*]] = call <vscale x 32 x i1> @llvm.experimental.vector.interleave2.nxv32i1(<vscale x 16 x i1> [[TMP10]], <vscale x 16 x i1> [[TMP10]])
+; PREDICATED_TAIL_FOLDING-NEXT:    [[INTERLEAVED_MASK:%.*]] = call <vscale x 32 x i1> @llvm.vector.interleave2.nxv32i1(<vscale x 16 x i1> [[TMP10]], <vscale x 16 x i1> [[TMP10]])
 ; PREDICATED_TAIL_FOLDING-NEXT:    [[WIDE_MASKED_VEC:%.*]] = call <vscale x 32 x i8> @llvm.masked.load.nxv32i8.p0(ptr [[TMP9]], i32 1, <vscale x 32 x i1> [[INTERLEAVED_MASK]], <vscale x 32 x i8> poison)
-; PREDICATED_TAIL_FOLDING-NEXT:    [[STRIDED_VEC:%.*]] = call { <vscale x 16 x i8>, <vscale x 16 x i8> } @llvm.experimental.vector.deinterleave2.nxv32i8(<vscale x 32 x i8> [[WIDE_MASKED_VEC]])
+; PREDICATED_TAIL_FOLDING-NEXT:    [[STRIDED_VEC:%.*]] = call { <vscale x 16 x i8>, <vscale x 16 x i8> } @llvm.vector.deinterleave2.nxv32i8(<vscale x 32 x i8> [[WIDE_MASKED_VEC]])
 ; PREDICATED_TAIL_FOLDING-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 16 x i8>, <vscale x 16 x i8> } [[STRIDED_VEC]], 0
 ; PREDICATED_TAIL_FOLDING-NEXT:    [[TMP12:%.*]] = extractvalue { <vscale x 16 x i8>, <vscale x 16 x i8> } [[STRIDED_VEC]], 1
 ; PREDICATED_TAIL_FOLDING-NEXT:    [[TMP13:%.*]] = or disjoint i32 [[TMP7]], 1
@@ -145,8 +145,8 @@ define dso_local void @masked_strided1(ptr noalias nocapture readonly %p, ptr no
 ; PREDICATED_TAIL_FOLDING-NEXT:    [[TMP16:%.*]] = sext i32 [[TMP13]] to i64
 ; PREDICATED_TAIL_FOLDING-NEXT:    [[TMP17:%.*]] = getelementptr i8, ptr [[Q]], i64 [[TMP16]]
 ; PREDICATED_TAIL_FOLDING-NEXT:    [[TMP18:%.*]] = getelementptr i8, ptr [[TMP17]], i64 -1
-; PREDICATED_TAIL_FOLDING-NEXT:    [[INTERLEAVED_VEC:%.*]] = call <vscale x 32 x i8> @llvm.experimental.vector.interleave2.nxv32i8(<vscale x 16 x i8> [[TMP14]], <vscale x 16 x i8> [[TMP15]])
-; PREDICATED_TAIL_FOLDING-NEXT:    [[INTERLEAVED_MASK1:%.*]] = call <vscale x 32 x i1> @llvm.experimental.vector.interleave2.nxv32i1(<vscale x 16 x i1> [[TMP10]], <vscale x 16 x i1> [[TMP10]])
+; PREDICATED_TAIL_FOLDING-NEXT:    [[INTERLEAVED_VEC:%.*]] = call <vscale x 32 x i8> @llvm.vector.interleave2.nxv32i8(<vscale x 16 x i8> [[TMP14]], <vscale x 16 x i8> [[TMP15]])
+; PREDICATED_TAIL_FOLDING-NEXT:    [[INTERLEAVED_MASK1:%.*]] = call <vscale x 32 x i1> @llvm.vector.interleave2.nxv32i1(<vscale x 16 x i1> [[TMP10]], <vscale x 16 x i1> [[TMP10]])
 ; PREDICATED_TAIL_FOLDING-NEXT:    call void @llvm.masked.store.nxv32i8.p0(<vscale x 32 x i8> [[INTERLEAVED_VEC]], ptr [[TMP18]], i32 1, <vscale x 32 x i1> [[INTERLEAVED_MASK1]])
 ; PREDICATED_TAIL_FOLDING-NEXT:    [[INDEX_NEXT]] = add i32 [[INDEX]], [[TMP20]]
 ; PREDICATED_TAIL_FOLDING-NEXT:    [[ACTIVE_LANE_MASK_NEXT]] = call <vscale x 16 x i1> @llvm.get.active.lane.mask.nxv16i1.i32(i32 [[INDEX]], i32 [[TMP2]])
diff --git a/llvm/test/Transforms/LoopVectorize/AArch64/sve-tail-folding-option.ll b/llvm/test/Transforms/LoopVectorize/AArch64/sve-tail-folding-option.ll
index 1dfa7f8fe18b..cf4d65318b7e 100644
--- a/llvm/test/Transforms/LoopVectorize/AArch64/sve-tail-folding-option.ll
+++ b/llvm/test/Transforms/LoopVectorize/AArch64/sve-tail-folding-option.ll
@@ -178,7 +178,7 @@ define void @add_recur(ptr noalias %dst, ptr noalias %src, i64 %n) #0 {
 ; CHECK-NOTF-NOT:     %{{.*}} = phi <vscale x 4 x i1>
 ; CHECK-NOTF:         %[[VECTOR_RECUR:.*]] = phi <vscale x 4 x i32> [ %[[RECUR_INIT]], %vector.ph ], [ %[[LOAD:.*]], %vector.body ]
 ; CHECK-NOTF:         %[[LOAD]] = load <vscale x 4 x i32>
-; CHECK-NOTF:         %[[SPLICE:.*]] = call <vscale x 4 x i32> @llvm.experimental.vector.splice.nxv4i32(<vscale x 4 x i32> %[[VECTOR_RECUR]], <vscale x 4 x i32> %[[LOAD]], i32 -1)
+; CHECK-NOTF:         %[[SPLICE:.*]] = call <vscale x 4 x i32> @llvm.vector.splice.nxv4i32(<vscale x 4 x i32> %[[VECTOR_RECUR]], <vscale x 4 x i32> %[[LOAD]], i32 -1)
 ; CHECK-NOTF:         %[[ADD:.*]] = add nsw <vscale x 4 x i32> %[[LOAD]], %[[SPLICE]]
 ; CHECK-NOTF:         store <vscale x 4 x i32> %[[ADD]]
 
@@ -191,7 +191,7 @@ define void @add_recur(ptr noalias %dst, ptr noalias %src, i64 %n) #0 {
 ; CHECK-TF-NORED:         %[[ACTIVE_LANE_MASK:.*]] = phi <vscale x 4 x i1>
 ; CHECK-TF-NORED:         %[[VECTOR_RECUR:.*]] = phi <vscale x 4 x i32> [ %[[RECUR_INIT]], %vector.ph ], [ %[[LOAD:.*]], %vector.body ]
 ; CHECK-TF-NORED:         %[[LOAD]] = call <vscale x 4 x i32> @llvm.masked.load.nxv4i32.p0({{.*}} %[[ACTIVE_LANE_MASK]]
-; CHECK-TF-NORED:         %[[SPLICE:.*]] = call <vscale x 4 x i32> @llvm.experimental.vector.splice.nxv4i32(<vscale x 4 x i32> %[[VECTOR_RECUR]], <vscale x 4 x i32> %[[LOAD]], i32 -1)
+; CHECK-TF-NORED:         %[[SPLICE:.*]] = call <vscale x 4 x i32> @llvm.vector.splice.nxv4i32(<vscale x 4 x i32> %[[VECTOR_RECUR]], <vscale x 4 x i32> %[[LOAD]], i32 -1)
 ; CHECK-TF-NORED:         %[[ADD:.*]] = add nsw <vscale x 4 x i32> %[[LOAD]], %[[SPLICE]]
 ; CHECK-TF-NORED:         call void @llvm.masked.store.nxv4i32.p0(<vscale x 4 x i32> %[[ADD]], {{.*}} <vscale x 4 x i1> %[[ACTIVE_LANE_MASK]])
 
@@ -204,7 +204,7 @@ define void @add_recur(ptr noalias %dst, ptr noalias %src, i64 %n) #0 {
 ; CHECK-TF-NOREC-NOT:     %{{.*}} = phi <vscale x 4 x i1>
 ; CHECK-TF-NOREC:         %[[VECTOR_RECUR:.*]] = phi <vscale x 4 x i32> [ %[[RECUR_INIT]], %vector.ph ], [ %[[LOAD:.*]], %vector.body ]
 ; CHECK-TF-NOREC:         %[[LOAD]] = load <vscale x 4 x i32>
-; CHECK-TF-NOREC:         %[[SPLICE:.*]] = call <vscale x 4 x i32> @llvm.experimental.vector.splice.nxv4i32(<vscale x 4 x i32> %[[VECTOR_RECUR]], <vscale x 4 x i32> %[[LOAD]], i32 -1)
+; CHECK-TF-NOREC:         %[[SPLICE:.*]] = call <vscale x 4 x i32> @llvm.vector.splice.nxv4i32(<vscale x 4 x i32> %[[VECTOR_RECUR]], <vscale x 4 x i32> %[[LOAD]], i32 -1)
 ; CHECK-TF-NOREC:         %[[ADD:.*]] = add nsw <vscale x 4 x i32> %[[LOAD]], %[[SPLICE]]
 ; CHECK-TF-NOREC:         store <vscale x 4 x i32> %[[ADD]]
 
@@ -217,7 +217,7 @@ define void @add_recur(ptr noalias %dst, ptr noalias %src, i64 %n) #0 {
 ; CHECK-TF-NOREV:         %[[ACTIVE_LANE_MASK:.*]] = phi <vscale x 4 x i1>
 ; CHECK-TF-NOREV:         %[[VECTOR_RECUR:.*]] = phi <vscale x 4 x i32> [ %[[RECUR_INIT]], %vector.ph ], [ %[[LOAD:.*]], %vector.body ]
 ; CHECK-TF-NOREV:         %[[LOAD]] = call <vscale x 4 x i32> @llvm.masked.load.nxv4i32.p0({{.*}} %[[ACTIVE_LANE_MASK]]
-; CHECK-TF-NOREV:         %[[SPLICE:.*]] = call <vscale x 4 x i32> @llvm.experimental.vector.splice.nxv4i32(<vscale x 4 x i32> %[[VECTOR_RECUR]], <vscale x 4 x i32> %[[LOAD]], i32 -1)
+; CHECK-TF-NOREV:         %[[SPLICE:.*]] = call <vscale x 4 x i32> @llvm.vector.splice.nxv4i32(<vscale x 4 x i32> %[[VECTOR_RECUR]], <vscale x 4 x i32> %[[LOAD]], i32 -1)
 ; CHECK-TF-NOREV:         %[[ADD:.*]] = add nsw <vscale x 4 x i32> %[[LOAD]], %[[SPLICE]]
 ; CHECK-TF-NOREV:         call void @llvm.masked.store.nxv4i32.p0(<vscale x 4 x i32> %[[ADD]], {{.*}} <vscale x 4 x i1> %[[ACTIVE_LANE_MASK]])
 
@@ -230,7 +230,7 @@ define void @add_recur(ptr noalias %dst, ptr noalias %src, i64 %n) #0 {
 ; CHECK-TF:         %[[ACTIVE_LANE_MASK:.*]] = phi <vscale x 4 x i1>
 ; CHECK-TF:         %[[VECTOR_RECUR:.*]] = phi <vscale x 4 x i32> [ %[[RECUR_INIT]], %vector.ph ], [ %[[LOAD:.*]], %vector.body ]
 ; CHECK-TF:         %[[LOAD]] = call <vscale x 4 x i32> @llvm.masked.load.nxv4i32.p0({{.*}} %[[ACTIVE_LANE_MASK]]
-; CHECK-TF:         %[[SPLICE:.*]] = call <vscale x 4 x i32> @llvm.experimental.vector.splice.nxv4i32(<vscale x 4 x i32> %[[VECTOR_RECUR]], <vscale x 4 x i32> %[[LOAD]], i32 -1)
+; CHECK-TF:         %[[SPLICE:.*]] = call <vscale x 4 x i32> @llvm.vector.splice.nxv4i32(<vscale x 4 x i32> %[[VECTOR_RECUR]], <vscale x 4 x i32> %[[LOAD]], i32 -1)
 ; CHECK-TF:         %[[ADD:.*]] = add nsw <vscale x 4 x i32> %[[LOAD]], %[[SPLICE]]
 ; CHECK-TF:         call void @llvm.masked.store.nxv4i32.p0(<vscale x 4 x i32> %[[ADD]], {{.*}} <vscale x 4 x i1> %[[ACTIVE_LANE_MASK]])
 
@@ -243,7 +243,7 @@ define void @add_recur(ptr noalias %dst, ptr noalias %src, i64 %n) #0 {
 ; CHECK-TF-ONLYRED-NOT:     %{{.*}} = phi <vscale x 4 x i1>
 ; CHECK-TF-ONLYRED:         %[[VECTOR_RECUR:.*]] = phi <vscale x 4 x i32> [ %[[RECUR_INIT]], %vector.ph ], [ %[[LOAD:.*]], %vector.body ]
 ; CHECK-TF-ONLYRED:         %[[LOAD]] = load <vscale x 4 x i32>
-; CHECK-TF-ONLYRED:         %[[SPLICE:.*]] = call <vscale x 4 x i32> @llvm.experimental.vector.splice.nxv4i32(<vscale x 4 x i32> %[[VECTOR_RECUR]], <vscale x 4 x i32> %[[LOAD]], i32 -1)
+; CHECK-TF-ONLYRED:         %[[SPLICE:.*]] = call <vscale x 4 x i32> @llvm.vector.splice.nxv4i32(<vscale x 4 x i32> %[[VECTOR_RECUR]], <vscale x 4 x i32> %[[LOAD]], i32 -1)
 ; CHECK-TF-ONLYRED:         %[[ADD:.*]] = add nsw <vscale x 4 x i32> %[[LOAD]], %[[SPLICE]]
 ; CHECK-TF-ONLYRED:         store <vscale x 4 x i32> %[[ADD]]
 
@@ -256,7 +256,7 @@ define void @add_recur(ptr noalias %dst, ptr noalias %src, i64 %n) #0 {
 ; CHECK-NEOVERSE-V1-NOT:     %{{.*}} = phi <vscale x 4 x i1>
 ; CHECK-NEOVERSE-V1:         %[[VECTOR_RECUR:.*]] = phi <vscale x 4 x i32> [ %[[RECUR_INIT]], %vector.ph ], [ %[[LOAD:.*]], %vector.body ]
 ; CHECK-NEOVERSE-V1:         %[[LOAD]] = load <vscale x 4 x i32>
-; CHECK-NEOVERSE-V1:         %[[SPLICE:.*]] = call <vscale x 4 x i32> @llvm.experimental.vector.splice.nxv4i32(<vscale x 4 x i32> %[[VECTOR_RECUR]], <vscale x 4 x i32> %[[LOAD]], i32 -1)
+; CHECK-NEOVERSE-V1:         %[[SPLICE:.*]] = call <vscale x 4 x i32> @llvm.vector.splice.nxv4i32(<vscale x 4 x i32> %[[VECTOR_RECUR]], <vscale x 4 x i32> %[[LOAD]], i32 -1)
 ; CHECK-NEOVERSE-V1:         %[[ADD:.*]] = add nsw <vscale x 4 x i32> %[[LOAD]], %[[SPLICE]]
 ; CHECK-NEOVERSE-V1:         store <vscale x 4 x i32> %[[ADD]]
 
@@ -350,30 +350,30 @@ define void @reverse(ptr noalias %dst, ptr noalias %src) #0 {
 ; CHECK-NOTF:       vector.body:
 ; CHECK-NOTF-NOT:     %{{.*}} = phi <vscale x 4 x i1>
 ; CHECK-NOTF:         %[[LOAD:.*]] = load <vscale x 2 x double>, ptr
-; CHECK-NOTF:         %{{.*}} = call <vscale x 2 x double> @llvm.experimental.vector.reverse.nxv2f64(<vscale x 2 x double> %[[LOAD]])
+; CHECK-NOTF:         %{{.*}} = call <vscale x 2 x double> @llvm.vector.reverse.nxv2f64(<vscale x 2 x double> %[[LOAD]])
 
 ; CHECK-TF-NOREV-LABEL: @reverse(
 ; CHECK-TF-NOREV:       vector.body:
 ; CHECK-TF-NOREV-NOT:     %{{.*}} = phi <vscale x 4 x i1>
 ; CHECK-TF-NOREV:         %[[LOAD:.*]] = load <vscale x 2 x double>, ptr
-; CHECK-TF-NOREV:         %{{.*}} = call <vscale x 2 x double> @llvm.experimental.vector.reverse.nxv2f64(<vscale x 2 x double> %[[LOAD]])
+; CHECK-TF-NOREV:         %{{.*}} = call <vscale x 2 x double> @llvm.vector.reverse.nxv2f64(<vscale x 2 x double> %[[LOAD]])
 
 ; CHECK-TF-LABEL: @reverse(
 ; CHECK-TF:       vector.body:
 ; CHECK-TF:         %[[ACTIVE_LANE_MASK:.*]] = phi <vscale x 2 x i1>
-; CHECK-TF:         %[[REVERSE_MASK:.*]] = call <vscale x 2 x i1> @llvm.experimental.vector.reverse.nxv2i1(<vscale x 2 x i1> %[[ACTIVE_LANE_MASK]])
+; CHECK-TF:         %[[REVERSE_MASK:.*]] = call <vscale x 2 x i1> @llvm.vector.reverse.nxv2i1(<vscale x 2 x i1> %[[ACTIVE_LANE_MASK]])
 ; CHECK-TF:         %[[MASKED_LOAD:.*]] = call <vscale x 2 x double> @llvm.masked.load.nxv2f64.p0({{.*}} <vscale x 2 x i1> %reverse
 
 ; CHECK-TF-NORED-LABEL: @reverse(
 ; CHECK-TF-NORED:       vector.body:
 ; CHECK-TF-NORED:         %[[ACTIVE_LANE_MASK:.*]] = phi <vscale x 2 x i1>
-; CHECK-TF-NORED:         %[[REVERSE_MASK:.*]] = call <vscale x 2 x i1> @llvm.experimental.vector.reverse.nxv2i1(<vscale x 2 x i1> %[[ACTIVE_LANE_MASK]])
+; CHECK-TF-NORED:         %[[REVERSE_MASK:.*]] = call <vscale x 2 x i1> @llvm.vector.reverse.nxv2i1(<vscale x 2 x i1> %[[ACTIVE_LANE_MASK]])
 ; CHECK-TF-NORED:         %[[MASKED_LOAD:.*]] = call <vscale x 2 x double> @llvm.masked.load.nxv2f64.p0({{.*}} <vscale x 2 x i1> %reverse
 
 ; CHECK-TF-NOREC-LABEL: @reverse(
 ; CHECK-TF-NOREC:       vector.body:
 ; CHECK-TF-NOREC:         %[[ACTIVE_LANE_MASK:.*]] = phi <vscale x 2 x i1>
-; CHECK-TF-NOREC:         %[[REVERSE_MASK:.*]] = call <vscale x 2 x i1> @llvm.experimental.vector.reverse.nxv2i1(<vscale x 2 x i1> %[[ACTIVE_LANE_MASK]])
+; CHECK-TF-NOREC:         %[[REVERSE_MASK:.*]] = call <vscale x 2 x i1> @llvm.vector.reverse.nxv2i1(<vscale x 2 x i1> %[[ACTIVE_LANE_MASK]])
 ; CHECK-TF-NOREC:         %[[MASKED_LOAD:.*]] = call <vscale x 2 x double> @llvm.masked.load.nxv2f64.p0({{.*}} <vscale x 2 x i1> %reverse
 
 entry:
diff --git a/llvm/test/Transforms/LoopVectorize/AArch64/sve-vector-reverse-mask4.ll b/llvm/test/Transforms/LoopVectorize/AArch64/sve-vector-reverse-mask4.ll
index 70833e44b075..9485d827ced4 100644
--- a/llvm/test/Transforms/LoopVectorize/AArch64/sve-vector-reverse-mask4.ll
+++ b/llvm/test/Transforms/LoopVectorize/AArch64/sve-vector-reverse-mask4.ll
@@ -1,5 +1,5 @@
 ; This is the loop in c++ being vectorize in this file with
-; experimental.vector.reverse
+; vector.reverse
 
 ;#pragma clang loop vectorize_width(4, scalable)
 ;  for (long int i = N - 1; i >= 0; i--)
@@ -18,12 +18,12 @@ target triple = "aarch64-unknown-linux-gnu"
 
 define void @vector_reverse_mask_nxv4i1(ptr %a, ptr %cond, i64 %N) #0 {
 ; CHECK-LABEL: vector.body:
-; CHECK: %[[REVERSE6:.*]] = call <vscale x 4 x i1> @llvm.experimental.vector.reverse.nxv4i1(<vscale x 4 x i1> %{{.*}})
+; CHECK: %[[REVERSE6:.*]] = call <vscale x 4 x i1> @llvm.vector.reverse.nxv4i1(<vscale x 4 x i1> %{{.*}})
 ; CHECK: %[[WIDEMSKLOAD:.*]] = call <vscale x 4 x double> @llvm.masked.load.nxv4f64.p0(ptr %{{.*}}, i32 8, <vscale x 4 x i1> %[[REVERSE6]], <vscale x 4 x double> poison)
-; CHECK: %[[REVERSE7:.*]] = call <vscale x 4 x double> @llvm.experimental.vector.reverse.nxv4f64(<vscale x 4 x double> %[[WIDEMSKLOAD]])
+; CHECK: %[[REVERSE7:.*]] = call <vscale x 4 x double> @llvm.vector.reverse.nxv4f64(<vscale x 4 x double> %[[WIDEMSKLOAD]])
 ; CHECK: %[[FADD:.*]] = fadd <vscale x 4 x double> %[[REVERSE7]]
-; CHECK: %[[REVERSE9:.*]] = call <vscale x 4 x i1> @llvm.experimental.vector.reverse.nxv4i1(<vscale x 4 x i1> %{{.*}})
-; CHECK: %[[REVERSE8:.*]] = call <vscale x 4 x double> @llvm.experimental.vector.reverse.nxv4f64(<vscale x 4 x double> %[[FADD]])
+; CHECK: %[[REVERSE9:.*]] = call <vscale x 4 x i1> @llvm.vector.reverse.nxv4i1(<vscale x 4 x i1> %{{.*}})
+; CHECK: %[[REVERSE8:.*]] = call <vscale x 4 x double> @llvm.vector.reverse.nxv4f64(<vscale x 4 x double> %[[FADD]])
 ; CHECK: call void @llvm.masked.store.nxv4f64.p0(<vscale x 4 x double> %[[REVERSE8]], ptr %{{.*}}, i32 8, <vscale x 4 x i1> %[[REVERSE9]]
 
 entry:
diff --git a/llvm/test/Transforms/LoopVectorize/AArch64/sve-vector-reverse.ll b/llvm/test/Transforms/LoopVectorize/AArch64/sve-vector-reverse.ll
index e35a4db36905..e3bba1338e1d 100644
--- a/llvm/test/Transforms/LoopVectorize/AArch64/sve-vector-reverse.ll
+++ b/llvm/test/Transforms/LoopVectorize/AArch64/sve-vector-reverse.ll
@@ -1,6 +1,6 @@
 ; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
 ; This is the loop in c++ being vectorize in this file with
-;experimental.vector.reverse
+;vector.reverse
 ;  #pragma clang loop vectorize_width(8, scalable) interleave_count(2)
 ;  for (int i = N-1; i >= 0; --i)
 ;    a[i] = b[i] + 1.0;
diff --git a/llvm/test/Transforms/LoopVectorize/AArch64/sve-widen-phi.ll b/llvm/test/Transforms/LoopVectorize/AArch64/sve-widen-phi.ll
index 126ceac7325a..61105e51cb94 100644
--- a/llvm/test/Transforms/LoopVectorize/AArch64/sve-widen-phi.ll
+++ b/llvm/test/Transforms/LoopVectorize/AArch64/sve-widen-phi.ll
@@ -40,10 +40,10 @@ define void @widen_ptr_phi_unrolled(ptr noalias nocapture %a, ptr noalias nocapt
 ; CHECK-NEXT:    [[TMP10:%.*]] = getelementptr i8, ptr [[TMP9]], i64 [[TMP7]]
 ; CHECK-NEXT:    [[WIDE_VEC:%.*]] = load <vscale x 8 x i32>, ptr [[TMP8]], align 4
 ; CHECK-NEXT:    [[WIDE_VEC2:%.*]] = load <vscale x 8 x i32>, ptr [[TMP10]], align 4
-; CHECK-NEXT:    [[STRIDED_VEC:%.*]] = call { <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.experimental.vector.deinterleave2.nxv8i32(<vscale x 8 x i32> [[WIDE_VEC]])
+; CHECK-NEXT:    [[STRIDED_VEC:%.*]] = call { <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.vector.deinterleave2.nxv8i32(<vscale x 8 x i32> [[WIDE_VEC]])
 ; CHECK-NEXT:    [[TMP11:%.*]] = extractvalue { <vscale x 4 x i32>, <vscale x 4 x i32> } [[STRIDED_VEC]], 0
 ; CHECK-NEXT:    [[TMP12:%.*]] = extractvalue { <vscale x 4 x i32>, <vscale x 4 x i32> } [[STRIDED_VEC]], 1
-; CHECK-NEXT:    [[STRIDED_VEC3:%.*]] = call { <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.experimental.vector.deinterleave2.nxv8i32(<vscale x 8 x i32> [[WIDE_VEC2]])
+; CHECK-NEXT:    [[STRIDED_VEC3:%.*]] = call { <vscale x 4 x i32>, <vscale x 4 x i32> } @llvm.vector.deinterleave2.nxv8i32(<vscale x 8 x i32> [[WIDE_VEC2]])
 ; CHECK-NEXT:    [[TMP13:%.*]] = extractvalue { <vscale x 4 x i32>, <vscale x 4 x i32> } [[STRIDED_VEC3]], 0
 ; CHECK-NEXT:    [[TMP14:%.*]] = extractvalue { <vscale x 4 x i32>, <vscale x 4 x i32> } [[STRIDED_VEC3]], 1
 ; CHECK-NEXT:    [[TMP15:%.*]] = add nsw <vscale x 4 x i32> [[TMP11]], shufflevector (<vscale x 4 x i32> insertelement (<vscale x 4 x i32> poison, i32 1, i64 0), <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer)
diff --git a/llvm/test/Transforms/LoopVectorize/ARM/tail-folding-counting-down.ll b/llvm/test/Transforms/LoopVectorize/ARM/tail-folding-counting-down.ll
index 5c1966fa7a2d..0f524561eadc 100644
--- a/llvm/test/Transforms/LoopVectorize/ARM/tail-folding-counting-down.ll
+++ b/llvm/test/Transforms/LoopVectorize/ARM/tail-folding-counting-down.ll
@@ -107,9 +107,10 @@ while.body:
   %1 = load i8, ptr %b.addr.07, align 1
   %add = add i8 %1, %0
   %incdec.ptr4 = getelementptr inbounds i8, ptr %c.addr.08, i32 1
-  store i8 %add, ptr %c.addr.08, align 1
   %cmp = icmp sgt i32 %N.addr.09, 1
   %select = select i1 %cmp, i8 %0, i8 %1
+  %add2 = add i8 %add, %select
+  store i8 %add2, ptr %c.addr.08, align 1
   br i1 %cmp, label %while.body, label %while.end.loopexit
 
 while.end.loopexit:
diff --git a/llvm/test/Transforms/LoopVectorize/RISCV/riscv-vector-reverse.ll b/llvm/test/Transforms/LoopVectorize/RISCV/riscv-vector-reverse.ll
index 72d9691b2bb8..c3374fceb1fb 100644
--- a/llvm/test/Transforms/LoopVectorize/RISCV/riscv-vector-reverse.ll
+++ b/llvm/test/Transforms/LoopVectorize/RISCV/riscv-vector-reverse.ll
@@ -1,6 +1,6 @@
 ; NOTE: Assertions have been autogenerated by utils/update_analyze_test_checks.py
 ; This is the loop in c++ being vectorize in this file with
-;experimental.vector.reverse
+;vector.reverse
 ;  #pragma clang loop vectorize_width(4, scalable)
 ;  for (int i = N-1; i >= 0; --i)
 ;    a[i] = b[i] + 1.0;
diff --git a/llvm/test/Transforms/LoopVectorize/RISCV/vectorize-force-tail-with-evl-reverse-load-store.ll b/llvm/test/Transforms/LoopVectorize/RISCV/vectorize-force-tail-with-evl-reverse-load-store.ll
index f2222e0a1f93..0dee4a9b8585 100644
--- a/llvm/test/Transforms/LoopVectorize/RISCV/vectorize-force-tail-with-evl-reverse-load-store.ll
+++ b/llvm/test/Transforms/LoopVectorize/RISCV/vectorize-force-tail-with-evl-reverse-load-store.ll
@@ -46,9 +46,9 @@ define void @reverse_load_store(i64 %startval, ptr noalias %ptr, ptr noalias %pt
 ; IF-EVL-NEXT:    [[TMP16:%.*]] = sub i64 1, [[TMP14]]
 ; IF-EVL-NEXT:    [[TMP17:%.*]] = getelementptr inbounds i32, ptr [[TMP12]], i64 [[TMP15]]
 ; IF-EVL-NEXT:    [[TMP18:%.*]] = getelementptr inbounds i32, ptr [[TMP17]], i64 [[TMP16]]
-; IF-EVL-NEXT:    [[REVERSE:%.*]] = call <vscale x 4 x i1> @llvm.experimental.vector.reverse.nxv4i1(<vscale x 4 x i1> [[TMP10]])
+; IF-EVL-NEXT:    [[REVERSE:%.*]] = call <vscale x 4 x i1> @llvm.vector.reverse.nxv4i1(<vscale x 4 x i1> [[TMP10]])
 ; IF-EVL-NEXT:    [[WIDE_MASKED_LOAD:%.*]] = call <vscale x 4 x i32> @llvm.masked.load.nxv4i32.p0(ptr [[TMP18]], i32 4, <vscale x 4 x i1> [[REVERSE]], <vscale x 4 x i32> poison)
-; IF-EVL-NEXT:    [[REVERSE3:%.*]] = call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> [[WIDE_MASKED_LOAD]])
+; IF-EVL-NEXT:    [[REVERSE3:%.*]] = call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> [[WIDE_MASKED_LOAD]])
 ; IF-EVL-NEXT:    [[TMP19:%.*]] = getelementptr inbounds i32, ptr [[PTR2:%.*]], i64 [[TMP11]]
 ; IF-EVL-NEXT:    [[TMP20:%.*]] = call i64 @llvm.vscale.i64()
 ; IF-EVL-NEXT:    [[TMP21:%.*]] = mul i64 [[TMP20]], 4
@@ -56,8 +56,8 @@ define void @reverse_load_store(i64 %startval, ptr noalias %ptr, ptr noalias %pt
 ; IF-EVL-NEXT:    [[TMP23:%.*]] = sub i64 1, [[TMP21]]
 ; IF-EVL-NEXT:    [[TMP24:%.*]] = getelementptr inbounds i32, ptr [[TMP19]], i64 [[TMP22]]
 ; IF-EVL-NEXT:    [[TMP25:%.*]] = getelementptr inbounds i32, ptr [[TMP24]], i64 [[TMP23]]
-; IF-EVL-NEXT:    [[REVERSE4:%.*]] = call <vscale x 4 x i1> @llvm.experimental.vector.reverse.nxv4i1(<vscale x 4 x i1> [[TMP10]])
-; IF-EVL-NEXT:    [[REVERSE5:%.*]] = call <vscale x 4 x i32> @llvm.experimental.vector.reverse.nxv4i32(<vscale x 4 x i32> [[REVERSE3]])
+; IF-EVL-NEXT:    [[REVERSE4:%.*]] = call <vscale x 4 x i1> @llvm.vector.reverse.nxv4i1(<vscale x 4 x i1> [[TMP10]])
+; IF-EVL-NEXT:    [[REVERSE5:%.*]] = call <vscale x 4 x i32> @llvm.vector.reverse.nxv4i32(<vscale x 4 x i32> [[REVERSE3]])
 ; IF-EVL-NEXT:    call void @llvm.masked.store.nxv4i32.p0(<vscale x 4 x i32> [[REVERSE5]], ptr [[TMP25]], i32 4, <vscale x 4 x i1> [[REVERSE4]])
 ; IF-EVL-NEXT:    [[INDEX_NEXT]] = add i64 [[INDEX]], [[TMP6]]
 ; IF-EVL-NEXT:    [[TMP26:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
diff --git a/llvm/test/Transforms/LoopVectorize/X86/induction-costs.ll b/llvm/test/Transforms/LoopVectorize/X86/induction-costs.ll
index 8ce87d0ef171..9ae36ae4d587 100644
--- a/llvm/test/Transforms/LoopVectorize/X86/induction-costs.ll
+++ b/llvm/test/Transforms/LoopVectorize/X86/induction-costs.ll
@@ -213,6 +213,79 @@ exit:
   ret void
 }
 
+define void @multiple_pointer_ivs_with_scalar_uses_only(ptr %A, ptr %B) #0 {
+entry:
+  br label %loop
+
+loop:
+  %iv.1 = phi i32 [ 100, %entry ], [ %dec, %loop ]
+  %iv.2 = phi i32 [ 2048, %entry ], [ %add38, %loop ]
+  %ptr.iv.1 = phi ptr [ %A, %entry ], [ %outptr.0, %loop ]
+  %ptr.iv.2 = phi ptr [ %B, %entry ], [ %incdec.ptr36, %loop ]
+  %ptr.iv.3 = phi ptr [ %B, %entry ], [ %incdec.ptr33, %loop ]
+  %incdec.ptr33 = getelementptr i8, ptr %ptr.iv.3, i64 1
+  %0 = load i8, ptr %ptr.iv.3, align 1
+  %conv34 = zext i8 %0 to i32
+  %incdec.ptr36 = getelementptr i8, ptr %ptr.iv.2, i64 1
+  %1 = load i8, ptr %ptr.iv.2, align 1
+  %conv37 = zext i8 %1 to i32
+  %add38 = add i32 %conv34, %conv37
+  %shr42 = lshr i32 %iv.2, 1
+  %conv43 = trunc i32 %shr42 to i8
+  store i8 %conv43, ptr %ptr.iv.1, align 1
+  %dec = add i32 %iv.1, 1
+  %outptr.0 = getelementptr i8, ptr %ptr.iv.1, i64 2
+  %cmp30.not = icmp eq i32 %dec, 0
+  br i1 %cmp30.not, label %exit, label %loop
+
+exit:
+  ret void
+}
+
+define i16 @iv_and_step_trunc() {
+entry:
+  br label %loop
+
+loop:
+  %iv = phi i64 [ 0, %entry ], [ %iv.next, %loop ]
+  %rec = phi i16 [ 0, %entry ], [ %rec.next, %loop ]
+  %iv.next = add i64 %iv, 1
+  %0 = trunc i64 %iv to i16
+  %1 = trunc i64 %iv.next to i16
+  %rec.next = mul i16 %0, %1
+  %ec = icmp eq i64 %iv, 1
+  br i1 %ec, label %exit, label %loop
+
+exit:
+  ret i16 %rec
+}
+
+define i32 @test_scalar_predicated_cost(i64 %x, i64 %y, ptr %A) #0 {
+entry:
+  br label %loop.header
+
+loop.header:
+  %iv = phi i64 [ 0, %entry ], [ %iv.next, %loop.latch ]
+  %cmp9.not = icmp ule i64 %iv, %y
+  br i1 %cmp9.not, label %loop.latch, label %if.then
+
+if.then:
+  %or = or i64 %x, %iv
+  %gep = getelementptr i32, ptr %A, i64 %iv
+  %t = trunc i64 %or to i32
+  store i32 %t, ptr %gep, align 4
+  br label %loop.latch
+
+loop.latch:
+  %iv.next = add i64 %iv, 1
+  %ec = icmp eq i64 %iv, 100
+  br i1 %ec, label %exit, label %loop.header
+
+exit:
+  ret i32 0
+}
+
+
 attributes #0 = { "min-legal-vector-width"="0" "target-cpu"="skylake-avx512" }
 ;.
 ; CHECK: [[LOOP0]] = distinct !{[[LOOP0]], [[META1:![0-9]+]], [[META2:![0-9]+]]}
diff --git a/llvm/test/Transforms/LoopVectorize/X86/masked-store-cost.ll b/llvm/test/Transforms/LoopVectorize/X86/masked-store-cost.ll
new file mode 100644
index 000000000000..2c8841caf870
--- /dev/null
+++ b/llvm/test/Transforms/LoopVectorize/X86/masked-store-cost.ll
@@ -0,0 +1,111 @@
+; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --version 4
+; RUN: opt -p loop-vectorize -mtriple=x86_64-apple-macosx -S %s | FileCheck %s
+
+target datalayout = "e-m:o-p270:32:32-p271:32:32-p272:64:64-i64:64-i128:128-f80:128-n8:16:32:64-S128"
+
+define i32 @test_scalar_predicated_cost(i64 %x, i64 %y, ptr %A) #0 {
+; CHECK-LABEL: define i32 @test_scalar_predicated_cost(
+; CHECK-SAME: i64 [[X:%.*]], i64 [[Y:%.*]], ptr [[A:%.*]]) #[[ATTR0:[0-9]+]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    br i1 false, label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
+; CHECK:       vector.ph:
+; CHECK-NEXT:    [[BROADCAST_SPLATINSERT:%.*]] = insertelement <8 x i64> poison, i64 [[Y]], i64 0
+; CHECK-NEXT:    [[BROADCAST_SPLAT:%.*]] = shufflevector <8 x i64> [[BROADCAST_SPLATINSERT]], <8 x i64> poison, <8 x i32> zeroinitializer
+; CHECK-NEXT:    [[BROADCAST_SPLATINSERT4:%.*]] = insertelement <8 x i64> poison, i64 [[X]], i64 0
+; CHECK-NEXT:    [[BROADCAST_SPLAT5:%.*]] = shufflevector <8 x i64> [[BROADCAST_SPLATINSERT4]], <8 x i64> poison, <8 x i32> zeroinitializer
+; CHECK-NEXT:    br label [[VECTOR_BODY:%.*]]
+; CHECK:       vector.body:
+; CHECK-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
+; CHECK-NEXT:    [[VEC_IND:%.*]] = phi <8 x i64> [ <i64 0, i64 1, i64 2, i64 3, i64 4, i64 5, i64 6, i64 7>, [[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], [[VECTOR_BODY]] ]
+; CHECK-NEXT:    [[STEP_ADD:%.*]] = add <8 x i64> [[VEC_IND]], <i64 8, i64 8, i64 8, i64 8, i64 8, i64 8, i64 8, i64 8>
+; CHECK-NEXT:    [[STEP_ADD1:%.*]] = add <8 x i64> [[STEP_ADD]], <i64 8, i64 8, i64 8, i64 8, i64 8, i64 8, i64 8, i64 8>
+; CHECK-NEXT:    [[STEP_ADD2:%.*]] = add <8 x i64> [[STEP_ADD1]], <i64 8, i64 8, i64 8, i64 8, i64 8, i64 8, i64 8, i64 8>
+; CHECK-NEXT:    [[TMP0:%.*]] = add i64 [[INDEX]], 0
+; CHECK-NEXT:    [[TMP1:%.*]] = add i64 [[INDEX]], 8
+; CHECK-NEXT:    [[TMP2:%.*]] = add i64 [[INDEX]], 16
+; CHECK-NEXT:    [[TMP3:%.*]] = add i64 [[INDEX]], 24
+; CHECK-NEXT:    [[TMP4:%.*]] = icmp ule <8 x i64> [[VEC_IND]], [[BROADCAST_SPLAT]]
+; CHECK-NEXT:    [[TMP5:%.*]] = icmp ule <8 x i64> [[STEP_ADD]], [[BROADCAST_SPLAT]]
+; CHECK-NEXT:    [[TMP6:%.*]] = icmp ule <8 x i64> [[STEP_ADD1]], [[BROADCAST_SPLAT]]
+; CHECK-NEXT:    [[TMP7:%.*]] = icmp ule <8 x i64> [[STEP_ADD2]], [[BROADCAST_SPLAT]]
+; CHECK-NEXT:    [[TMP8:%.*]] = xor <8 x i1> [[TMP4]], <i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true>
+; CHECK-NEXT:    [[TMP9:%.*]] = xor <8 x i1> [[TMP5]], <i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true>
+; CHECK-NEXT:    [[TMP10:%.*]] = xor <8 x i1> [[TMP6]], <i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true>
+; CHECK-NEXT:    [[TMP11:%.*]] = xor <8 x i1> [[TMP7]], <i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true, i1 true>
+; CHECK-NEXT:    [[TMP12:%.*]] = or <8 x i64> [[BROADCAST_SPLAT5]], [[VEC_IND]]
+; CHECK-NEXT:    [[TMP13:%.*]] = or <8 x i64> [[BROADCAST_SPLAT5]], [[STEP_ADD]]
+; CHECK-NEXT:    [[TMP14:%.*]] = or <8 x i64> [[BROADCAST_SPLAT5]], [[STEP_ADD1]]
+; CHECK-NEXT:    [[TMP15:%.*]] = or <8 x i64> [[BROADCAST_SPLAT5]], [[STEP_ADD2]]
+; CHECK-NEXT:    [[TMP16:%.*]] = getelementptr i32, ptr [[A]], i64 [[TMP0]]
+; CHECK-NEXT:    [[TMP17:%.*]] = getelementptr i32, ptr [[A]], i64 [[TMP1]]
+; CHECK-NEXT:    [[TMP18:%.*]] = getelementptr i32, ptr [[A]], i64 [[TMP2]]
+; CHECK-NEXT:    [[TMP19:%.*]] = getelementptr i32, ptr [[A]], i64 [[TMP3]]
+; CHECK-NEXT:    [[TMP20:%.*]] = trunc <8 x i64> [[TMP12]] to <8 x i32>
+; CHECK-NEXT:    [[TMP21:%.*]] = trunc <8 x i64> [[TMP13]] to <8 x i32>
+; CHECK-NEXT:    [[TMP22:%.*]] = trunc <8 x i64> [[TMP14]] to <8 x i32>
+; CHECK-NEXT:    [[TMP23:%.*]] = trunc <8 x i64> [[TMP15]] to <8 x i32>
+; CHECK-NEXT:    [[TMP24:%.*]] = getelementptr i32, ptr [[TMP16]], i32 0
+; CHECK-NEXT:    [[TMP25:%.*]] = getelementptr i32, ptr [[TMP16]], i32 8
+; CHECK-NEXT:    [[TMP26:%.*]] = getelementptr i32, ptr [[TMP16]], i32 16
+; CHECK-NEXT:    [[TMP27:%.*]] = getelementptr i32, ptr [[TMP16]], i32 24
+; CHECK-NEXT:    call void @llvm.masked.store.v8i32.p0(<8 x i32> [[TMP20]], ptr [[TMP24]], i32 4, <8 x i1> [[TMP8]])
+; CHECK-NEXT:    call void @llvm.masked.store.v8i32.p0(<8 x i32> [[TMP21]], ptr [[TMP25]], i32 4, <8 x i1> [[TMP9]])
+; CHECK-NEXT:    call void @llvm.masked.store.v8i32.p0(<8 x i32> [[TMP22]], ptr [[TMP26]], i32 4, <8 x i1> [[TMP10]])
+; CHECK-NEXT:    call void @llvm.masked.store.v8i32.p0(<8 x i32> [[TMP23]], ptr [[TMP27]], i32 4, <8 x i1> [[TMP11]])
+; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 32
+; CHECK-NEXT:    [[VEC_IND_NEXT]] = add <8 x i64> [[STEP_ADD2]], <i64 8, i64 8, i64 8, i64 8, i64 8, i64 8, i64 8, i64 8>
+; CHECK-NEXT:    [[TMP28:%.*]] = icmp eq i64 [[INDEX_NEXT]], 96
+; CHECK-NEXT:    br i1 [[TMP28]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP0:![0-9]+]]
+; CHECK:       middle.block:
+; CHECK-NEXT:    br i1 false, label [[EXIT:%.*]], label [[SCALAR_PH]]
+; CHECK:       scalar.ph:
+; CHECK-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ 96, [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ]
+; CHECK-NEXT:    br label [[LOOP_HEADER:%.*]]
+; CHECK:       loop.header:
+; CHECK-NEXT:    [[IV:%.*]] = phi i64 [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ], [ [[IV_NEXT:%.*]], [[LOOP_LATCH:%.*]] ]
+; CHECK-NEXT:    [[CMP9_NOT:%.*]] = icmp ule i64 [[IV]], [[Y]]
+; CHECK-NEXT:    br i1 [[CMP9_NOT]], label [[LOOP_LATCH]], label [[IF_THEN:%.*]]
+; CHECK:       if.then:
+; CHECK-NEXT:    [[OR:%.*]] = or i64 [[X]], [[IV]]
+; CHECK-NEXT:    [[GEP:%.*]] = getelementptr i32, ptr [[A]], i64 [[IV]]
+; CHECK-NEXT:    [[T:%.*]] = trunc i64 [[OR]] to i32
+; CHECK-NEXT:    store i32 [[T]], ptr [[GEP]], align 4
+; CHECK-NEXT:    br label [[LOOP_LATCH]]
+; CHECK:       loop.latch:
+; CHECK-NEXT:    [[IV_NEXT]] = add i64 [[IV]], 1
+; CHECK-NEXT:    [[EC:%.*]] = icmp eq i64 [[IV]], 100
+; CHECK-NEXT:    br i1 [[EC]], label [[EXIT]], label [[LOOP_HEADER]], !llvm.loop [[LOOP3:![0-9]+]]
+; CHECK:       exit:
+; CHECK-NEXT:    ret i32 0
+;
+entry:
+  br label %loop.header
+
+loop.header:
+  %iv = phi i64 [ 0, %entry ], [ %iv.next, %loop.latch ]
+  %cmp9.not = icmp ule i64 %iv, %y
+  br i1 %cmp9.not, label %loop.latch, label %if.then
+
+if.then:
+  %or = or i64 %x, %iv
+  %gep = getelementptr i32, ptr %A, i64 %iv
+  %t = trunc i64 %or to i32
+  store i32 %t, ptr %gep, align 4
+  br label %loop.latch
+
+loop.latch:
+  %iv.next = add i64 %iv, 1
+  %ec = icmp eq i64 %iv, 100
+  br i1 %ec, label %exit, label %loop.header
+
+exit:
+  ret i32 0
+}
+
+attributes #0 = { "min-legal-vector-width"="0" "target-cpu"="skylake-avx512" }
+;.
+; CHECK: [[LOOP0]] = distinct !{[[LOOP0]], [[META1:![0-9]+]], [[META2:![0-9]+]]}
+; CHECK: [[META1]] = !{!"llvm.loop.isvectorized", i32 1}
+; CHECK: [[META2]] = !{!"llvm.loop.unroll.runtime.disable"}
+; CHECK: [[LOOP3]] = distinct !{[[LOOP3]], [[META2]], [[META1]]}
+;.
diff --git a/llvm/test/Transforms/LoopVectorize/reduction-inloop.ll b/llvm/test/Transforms/LoopVectorize/reduction-inloop.ll
index 204d021f0657..e6936b19415d 100644
--- a/llvm/test/Transforms/LoopVectorize/reduction-inloop.ll
+++ b/llvm/test/Transforms/LoopVectorize/reduction-inloop.ll
@@ -1357,12 +1357,12 @@ define i32 @predicated_or_dominates_reduction(ptr %b) {
 ; CHECK-NEXT:    [[TMP46:%.*]] = xor <4 x i1> [[TMP19]], <i1 true, i1 true, i1 true, i1 true>
 ; CHECK-NEXT:    [[TMP47:%.*]] = select <4 x i1> [[TMP46]], <4 x i1> <i1 true, i1 true, i1 true, i1 true>, <4 x i1> [[TMP44]]
 ; CHECK-NEXT:    [[TMP48:%.*]] = bitcast <4 x i1> [[TMP47]] to i4
-; CHECK-NEXT:    [[TMP49:%.*]] = call i4 @llvm.ctpop.i4(i4 [[TMP48]]), !range [[RNG42:![0-9]+]]
+; CHECK-NEXT:    [[TMP49:%.*]] = call range(i4 0, 5) i4 @llvm.ctpop.i4(i4 [[TMP48]])
 ; CHECK-NEXT:    [[TMP50:%.*]] = zext nneg i4 [[TMP49]] to i32
 ; CHECK-NEXT:    [[TMP51]] = add i32 [[VEC_PHI]], [[TMP50]]
 ; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i32 [[INDEX]], 4
 ; CHECK-NEXT:    [[TMP52:%.*]] = icmp eq i32 [[INDEX_NEXT]], 1000
-; CHECK-NEXT:    br i1 [[TMP52]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP43:![0-9]+]]
+; CHECK-NEXT:    br i1 [[TMP52]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP42:![0-9]+]]
 ; CHECK:       middle.block:
 ; CHECK-NEXT:    br i1 true, label [[FOR_COND_CLEANUP:%.*]], label [[SCALAR_PH]]
 ; CHECK:       scalar.ph:
@@ -1377,7 +1377,7 @@ define i32 @predicated_or_dominates_reduction(ptr %b) {
 ; CHECK:       if.then:
 ; CHECK-NEXT:    br label [[FOR_INC]]
 ; CHECK:       for.inc:
-; CHECK-NEXT:    br i1 poison, label [[FOR_COND_CLEANUP]], label [[FOR_BODY]], !llvm.loop [[LOOP44:![0-9]+]]
+; CHECK-NEXT:    br i1 poison, label [[FOR_COND_CLEANUP]], label [[FOR_BODY]], !llvm.loop [[LOOP43:![0-9]+]]
 ;
 entry:
   br label %for.body
diff --git a/llvm/test/Transforms/LoopVectorize/runtime-checks-hoist.ll b/llvm/test/Transforms/LoopVectorize/runtime-checks-hoist.ll
index 0b9b592627c6..c4f9c404a926 100644
--- a/llvm/test/Transforms/LoopVectorize/runtime-checks-hoist.ll
+++ b/llvm/test/Transforms/LoopVectorize/runtime-checks-hoist.ll
@@ -1328,13 +1328,11 @@ define void @unknown_inner_stride(ptr nocapture noundef %dst, ptr nocapture noun
 ; CHECK:       vector.body:
 ; CHECK-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
 ; CHECK-NEXT:    [[TMP14:%.*]] = add i64 [[INDEX]], 0
-; CHECK-NEXT:    [[TMP15:%.*]] = mul nsw i64 [[TMP14]], [[TMP0]]
-; CHECK-NEXT:    [[TMP16:%.*]] = add nsw i64 [[TMP15]], [[TMP11]]
+; CHECK-NEXT:    [[TMP16:%.*]] = add nsw i64 [[TMP14]], [[TMP11]]
 ; CHECK-NEXT:    [[TMP17:%.*]] = getelementptr inbounds i32, ptr [[SRC]], i64 [[TMP16]]
 ; CHECK-NEXT:    [[TMP18:%.*]] = getelementptr inbounds i32, ptr [[TMP17]], i32 0
 ; CHECK-NEXT:    [[WIDE_LOAD:%.*]] = load <4 x i32>, ptr [[TMP18]], align 4, !alias.scope [[META60:![0-9]+]]
-; CHECK-NEXT:    [[TMP19:%.*]] = mul nsw i64 [[TMP14]], [[TMP1]]
-; CHECK-NEXT:    [[TMP20:%.*]] = add nsw i64 [[TMP19]], [[TMP12]]
+; CHECK-NEXT:    [[TMP20:%.*]] = add nsw i64 [[TMP14]], [[TMP12]]
 ; CHECK-NEXT:    [[TMP21:%.*]] = getelementptr inbounds i32, ptr [[DST]], i64 [[TMP20]]
 ; CHECK-NEXT:    [[TMP22:%.*]] = getelementptr inbounds i32, ptr [[TMP21]], i32 0
 ; CHECK-NEXT:    [[WIDE_LOAD3:%.*]] = load <4 x i32>, ptr [[TMP22]], align 4, !alias.scope [[META63:![0-9]+]], !noalias [[META60]]
diff --git a/llvm/test/Transforms/LoopVectorize/scalable-first-order-recurrence.ll b/llvm/test/Transforms/LoopVectorize/scalable-first-order-recurrence.ll
index 3be31c011eaa..d64755999635 100644
--- a/llvm/test/Transforms/LoopVectorize/scalable-first-order-recurrence.ll
+++ b/llvm/test/Transforms/LoopVectorize/scalable-first-order-recurrence.ll
@@ -21,7 +21,7 @@ define i32 @recurrence_1(ptr nocapture readonly %a, ptr nocapture %b, i32 %n) {
 ; CHECK-VF4UF1: %[[INDEX:.*]] = phi i64 [ 0, %vector.ph ], [ %[[NEXT_IDX:.*]], %vector.body ]
 ; CHECK-VF4UF1: %[[VEC_RECUR:.*]] = phi <vscale x 4 x i32> [ %[[VEC_RECUR_INIT]], %vector.ph ], [ %[[LOAD:.*]], %vector.body ]
 ; CHECK-VF4UF1: %[[LOAD]] = load <vscale x 4 x i32>, ptr
-; CHECK-VF4UF1: %[[SPLICE:.*]] = call <vscale x 4 x i32> @llvm.experimental.vector.splice.nxv4i32(<vscale x 4 x i32> %[[VEC_RECUR]], <vscale x 4 x i32> %[[LOAD]], i32 -1)
+; CHECK-VF4UF1: %[[SPLICE:.*]] = call <vscale x 4 x i32> @llvm.vector.splice.nxv4i32(<vscale x 4 x i32> %[[VEC_RECUR]], <vscale x 4 x i32> %[[LOAD]], i32 -1)
 ; CHECK-VF4UF1: middle.block:
 ; CHECK-VF4UF1: %[[VSCALE2:.*]] = call i32 @llvm.vscale.i32()
 ; CHECK-VF4UF1: %[[MUL2:.*]] = mul i32 %[[VSCALE2]], 4
@@ -70,7 +70,7 @@ define i32 @recurrence_2(ptr nocapture readonly %a, i32 %n) {
 ; CHECK-VF4UF1: vector.body:
 ; CHECK-VF4UF1: %[[VEC_RECUR:.*]] = phi <vscale x 4 x i32> [ %[[VEC_RECUR_INIT]], %vector.ph ], [ %[[LOAD:.*]], %vector.body ]
 ; CHECK-VF4UF1: %[[LOAD]] = load <vscale x 4 x i32>, ptr
-; CHECK-VF4UF1: %[[REVERSE:.*]] = call <vscale x 4 x i32> @llvm.experimental.vector.splice.nxv4i32(<vscale x 4 x i32> %[[VEC_RECUR]], <vscale x 4 x i32> %[[LOAD]], i32 -1)
+; CHECK-VF4UF1: %[[REVERSE:.*]] = call <vscale x 4 x i32> @llvm.vector.splice.nxv4i32(<vscale x 4 x i32> %[[VEC_RECUR]], <vscale x 4 x i32> %[[LOAD]], i32 -1)
 ; CHECK-VF4UF1: middle.block:
 ; CHECK-VF4UF1: %[[VSCALE2:.*]] = call i32 @llvm.vscale.i32()
 ; CHECK-VF4UF1: %[[MUL2:.*]] = mul i32 %[[VSCALE2]], 4
@@ -119,7 +119,7 @@ define void @recurrence_3(ptr nocapture readonly %a, ptr nocapture %b, i32 %n, f
 ; CHECK-VF4UF1: vector.body:
 ; CHECK-VF4UF1: %vector.recur = phi <vscale x 4 x i16> [ %vector.recur.init, %vector.ph ], [ %[[L1:.*]], %vector.body ]
 ; CHECK-VF4UF1: %[[L1]] = load <vscale x 4 x i16>, ptr
-; CHECK-VF4UF1: %[[SPLICE:.*]] = call <vscale x 4 x i16> @llvm.experimental.vector.splice.nxv4i16(<vscale x 4 x i16> %vector.recur, <vscale x 4 x i16> %[[L1]], i32 -1)
+; CHECK-VF4UF1: %[[SPLICE:.*]] = call <vscale x 4 x i16> @llvm.vector.splice.nxv4i16(<vscale x 4 x i16> %vector.recur, <vscale x 4 x i16> %[[L1]], i32 -1)
 ; Check also that the casts were not moved needlessly.
 ; CHECK-VF4UF1: sitofp <vscale x 4 x i16> %[[L1]] to <vscale x 4 x double>
 ; CHECK-VF4UF1: sitofp <vscale x 4 x i16> %[[SPLICE]] to <vscale x 4 x double>
@@ -169,8 +169,8 @@ define i64 @constant_folded_previous_value() {
 ; CHECK-VF4UF2-LABEL: @constant_folded_previous_value
 ; CHECK-VF4UF2: vector.body
 ; CHECK-VF4UF2: %[[VECTOR_RECUR:.*]] = phi <vscale x 4 x i64> [ %vector.recur.init, %vector.ph ], [ shufflevector (<vscale x 4 x i64> insertelement (<vscale x 4 x i64> poison, i64 1, i64 0), <vscale x 4 x i64> poison, <vscale x 4 x i32> zeroinitializer), %vector.body ]
-; CHECK-VF4UF2: %[[SPLICE1:.*]] = call <vscale x 4 x i64> @llvm.experimental.vector.splice.nxv4i64(<vscale x 4 x i64> %vector.recur, <vscale x 4 x i64> shufflevector (<vscale x 4 x i64> insertelement (<vscale x 4 x i64> poison, i64 1, i64 0), <vscale x 4 x i64> poison, <vscale x 4 x i32> zeroinitializer), i32 -1)
-; CHECK-VF4UF2: %[[SPLICE2:.*]] = call <vscale x 4 x i64> @llvm.experimental.vector.splice.nxv4i64(<vscale x 4 x i64> shufflevector (<vscale x 4 x i64> insertelement (<vscale x 4 x i64> poison, i64 1, i64 0), <vscale x 4 x i64> poison, <vscale x 4 x i32> zeroinitializer), <vscale x 4 x i64> shufflevector (<vscale x 4 x i64> insertelement (<vscale x 4 x i64> poison, i64 1, i64 0), <vscale x 4 x i64> poison, <vscale x 4 x i32> zeroinitializer), i32 -1)
+; CHECK-VF4UF2: %[[SPLICE1:.*]] = call <vscale x 4 x i64> @llvm.vector.splice.nxv4i64(<vscale x 4 x i64> %vector.recur, <vscale x 4 x i64> shufflevector (<vscale x 4 x i64> insertelement (<vscale x 4 x i64> poison, i64 1, i64 0), <vscale x 4 x i64> poison, <vscale x 4 x i32> zeroinitializer), i32 -1)
+; CHECK-VF4UF2: %[[SPLICE2:.*]] = call <vscale x 4 x i64> @llvm.vector.splice.nxv4i64(<vscale x 4 x i64> shufflevector (<vscale x 4 x i64> insertelement (<vscale x 4 x i64> poison, i64 1, i64 0), <vscale x 4 x i64> poison, <vscale x 4 x i32> zeroinitializer), <vscale x 4 x i64> shufflevector (<vscale x 4 x i64> insertelement (<vscale x 4 x i64> poison, i64 1, i64 0), <vscale x 4 x i64> poison, <vscale x 4 x i32> zeroinitializer), i32 -1)
 ; CHECK-VF4UF2: br i1 {{.*}}, label %middle.block, label %vector.body
 entry:
   br label %scalar.body
@@ -242,7 +242,7 @@ define void @sink_after(ptr %a, ptr %b, i64 %n) {
 ; CHECK-VF4UF1: vector.body
 ; CHECK-VF4UF1: %[[VEC_RECUR:.*]] = phi <vscale x 4 x i16> [ %vector.recur.init, %vector.ph ], [ %[[LOAD:.*]], %vector.body ]
 ; CHECK-VF4UF1: %[[LOAD]] = load <vscale x 4 x i16>, ptr
-; CHECK-VF4UF1-NEXT: %[[SPLICE:.*]] = call <vscale x 4 x i16> @llvm.experimental.vector.splice.nxv4i16(<vscale x 4 x i16> %[[VEC_RECUR]], <vscale x 4 x i16> %[[LOAD]], i32 -1)
+; CHECK-VF4UF1-NEXT: %[[SPLICE:.*]] = call <vscale x 4 x i16> @llvm.vector.splice.nxv4i16(<vscale x 4 x i16> %[[VEC_RECUR]], <vscale x 4 x i16> %[[LOAD]], i32 -1)
 ; CHECK-VF4UF1-NEXT: sext <vscale x 4 x i16> %[[SPLICE]] to <vscale x 4 x i32>
 ; CHECK-VF4UF1-NEXT: sext <vscale x 4 x i16> %[[LOAD]] to <vscale x 4 x i32>
 entry:
diff --git a/llvm/test/Transforms/LoopVectorize/trunc-extended-icmps.ll b/llvm/test/Transforms/LoopVectorize/trunc-extended-icmps.ll
new file mode 100644
index 000000000000..33fc1f70bf3c
--- /dev/null
+++ b/llvm/test/Transforms/LoopVectorize/trunc-extended-icmps.ll
@@ -0,0 +1,211 @@
+; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --version 4
+; RUN: opt -p loop-vectorize -force-vector-width=4 -force-vector-interleave=1 -S %s | FileCheck %s
+
+target datalayout = "e-m:e-p:64:64-i64:64-i128:128-n32:64-S128"
+
+define i32 @test_icmp_constant_op_zext(ptr %dst) {
+; CHECK-LABEL: define i32 @test_icmp_constant_op_zext(
+; CHECK-SAME: ptr [[DST:%.*]]) {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    br i1 false, label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
+; CHECK:       vector.ph:
+; CHECK-NEXT:    br label [[VECTOR_BODY:%.*]]
+; CHECK:       vector.body:
+; CHECK-NEXT:    [[INDEX:%.*]] = phi i32 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
+; CHECK-NEXT:    [[DOTCAST:%.*]] = trunc i32 [[INDEX]] to i16
+; CHECK-NEXT:    [[OFFSET_IDX:%.*]] = add i16 1, [[DOTCAST]]
+; CHECK-NEXT:    [[TMP0:%.*]] = add i16 [[OFFSET_IDX]], 0
+; CHECK-NEXT:    [[TMP1:%.*]] = getelementptr i8, ptr [[DST]], i16 [[TMP0]]
+; CHECK-NEXT:    [[TMP2:%.*]] = getelementptr i8, ptr [[TMP1]], i32 0
+; CHECK-NEXT:    store <4 x i8> <i8 109, i8 109, i8 109, i8 109>, ptr [[TMP2]], align 1
+; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i32 [[INDEX]], 4
+; CHECK-NEXT:    [[TMP3:%.*]] = icmp eq i32 [[INDEX_NEXT]], 996
+; CHECK-NEXT:    br i1 [[TMP3]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP0:![0-9]+]]
+; CHECK:       middle.block:
+; CHECK-NEXT:    br i1 false, label [[EXIT:%.*]], label [[SCALAR_PH]]
+; CHECK:       scalar.ph:
+; CHECK-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i16 [ 997, [[MIDDLE_BLOCK]] ], [ 1, [[ENTRY:%.*]] ]
+; CHECK-NEXT:    br label [[LOOP:%.*]]
+; CHECK:       loop:
+; CHECK-NEXT:    [[IV:%.*]] = phi i16 [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ], [ [[IV_NEXT:%.*]], [[LOOP]] ]
+; CHECK-NEXT:    [[C:%.*]] = icmp ne i64 7304878031173690989, 0
+; CHECK-NEXT:    [[C_EXT:%.*]] = zext i1 [[C]] to i64
+; CHECK-NEXT:    [[OR:%.*]] = or i64 7304878031173690989, [[C_EXT]]
+; CHECK-NEXT:    [[OR_TRUNC:%.*]] = trunc i64 [[OR]] to i8
+; CHECK-NEXT:    [[GEP:%.*]] = getelementptr i8, ptr [[DST]], i16 [[IV]]
+; CHECK-NEXT:    store i8 [[OR_TRUNC]], ptr [[GEP]], align 1
+; CHECK-NEXT:    [[IV_NEXT]] = add i16 [[IV]], 1
+; CHECK-NEXT:    [[EC:%.*]] = icmp eq i16 [[IV_NEXT]], 1000
+; CHECK-NEXT:    br i1 [[EC]], label [[EXIT]], label [[LOOP]], !llvm.loop [[LOOP3:![0-9]+]]
+; CHECK:       exit:
+; CHECK-NEXT:    ret i32 0
+;
+entry:
+  br label %loop
+
+loop:
+  %iv = phi i16 [ 1, %entry ], [ %iv.next, %loop ]
+  %c = icmp ne i64 7304878031173690989, 0
+  %c.ext = zext i1 %c to i64
+  %or = or i64 7304878031173690989, %c.ext
+  %or.trunc = trunc i64 %or to i8
+  %gep = getelementptr i8, ptr %dst, i16 %iv
+  store i8 %or.trunc, ptr %gep, align 1
+  %iv.next = add i16 %iv, 1
+  %ec = icmp eq i16 %iv.next, 1000
+  br i1 %ec, label %exit, label %loop
+
+exit:
+  ret i32 0
+}
+
+
+define i32 @test_icmp_and_op_zext(ptr %dst, i64 %a) {
+; CHECK-LABEL: define i32 @test_icmp_and_op_zext(
+; CHECK-SAME: ptr [[DST:%.*]], i64 [[A:%.*]]) {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[AND:%.*]] = and i64 [[A]], 7304878031173690989
+; CHECK-NEXT:    br i1 false, label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
+; CHECK:       vector.ph:
+; CHECK-NEXT:    [[BROADCAST_SPLATINSERT:%.*]] = insertelement <4 x i64> poison, i64 [[AND]], i64 0
+; CHECK-NEXT:    [[BROADCAST_SPLAT:%.*]] = shufflevector <4 x i64> [[BROADCAST_SPLATINSERT]], <4 x i64> poison, <4 x i32> zeroinitializer
+; CHECK-NEXT:    br label [[VECTOR_BODY:%.*]]
+; CHECK:       vector.body:
+; CHECK-NEXT:    [[INDEX:%.*]] = phi i32 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
+; CHECK-NEXT:    [[DOTCAST:%.*]] = trunc i32 [[INDEX]] to i16
+; CHECK-NEXT:    [[OFFSET_IDX:%.*]] = add i16 1, [[DOTCAST]]
+; CHECK-NEXT:    [[TMP0:%.*]] = add i16 [[OFFSET_IDX]], 0
+; CHECK-NEXT:    [[TMP1:%.*]] = icmp ne <4 x i64> [[BROADCAST_SPLAT]], zeroinitializer
+; CHECK-NEXT:    [[TMP2:%.*]] = zext <4 x i1> [[TMP1]] to <4 x i64>
+; CHECK-NEXT:    [[TMP3:%.*]] = or <4 x i64> [[BROADCAST_SPLAT]], [[TMP2]]
+; CHECK-NEXT:    [[TMP4:%.*]] = trunc <4 x i64> [[TMP3]] to <4 x i8>
+; CHECK-NEXT:    [[TMP5:%.*]] = getelementptr i8, ptr [[DST]], i16 [[TMP0]]
+; CHECK-NEXT:    [[TMP6:%.*]] = getelementptr i8, ptr [[TMP5]], i32 0
+; CHECK-NEXT:    store <4 x i8> [[TMP4]], ptr [[TMP6]], align 1
+; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i32 [[INDEX]], 4
+; CHECK-NEXT:    [[TMP7:%.*]] = icmp eq i32 [[INDEX_NEXT]], 996
+; CHECK-NEXT:    br i1 [[TMP7]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP4:![0-9]+]]
+; CHECK:       middle.block:
+; CHECK-NEXT:    br i1 false, label [[EXIT:%.*]], label [[SCALAR_PH]]
+; CHECK:       scalar.ph:
+; CHECK-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i16 [ 997, [[MIDDLE_BLOCK]] ], [ 1, [[ENTRY:%.*]] ]
+; CHECK-NEXT:    br label [[LOOP:%.*]]
+; CHECK:       loop:
+; CHECK-NEXT:    [[IV:%.*]] = phi i16 [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ], [ [[IV_NEXT:%.*]], [[LOOP]] ]
+; CHECK-NEXT:    [[C:%.*]] = icmp ne i64 [[AND]], 0
+; CHECK-NEXT:    [[C_EXT:%.*]] = zext i1 [[C]] to i64
+; CHECK-NEXT:    [[OR:%.*]] = or i64 [[AND]], [[C_EXT]]
+; CHECK-NEXT:    [[OR_TRUNC:%.*]] = trunc i64 [[OR]] to i8
+; CHECK-NEXT:    [[GEP:%.*]] = getelementptr i8, ptr [[DST]], i16 [[IV]]
+; CHECK-NEXT:    store i8 [[OR_TRUNC]], ptr [[GEP]], align 1
+; CHECK-NEXT:    [[IV_NEXT]] = add i16 [[IV]], 1
+; CHECK-NEXT:    [[EC:%.*]] = icmp eq i16 [[IV_NEXT]], 1000
+; CHECK-NEXT:    br i1 [[EC]], label [[EXIT]], label [[LOOP]], !llvm.loop [[LOOP5:![0-9]+]]
+; CHECK:       exit:
+; CHECK-NEXT:    ret i32 0
+;
+entry:
+  %and = and i64 %a, 7304878031173690989
+  br label %loop
+
+loop:
+  %iv = phi i16 [ 1, %entry ], [ %iv.next, %loop ]
+  %c = icmp ne i64 %and, 0
+  %c.ext = zext i1 %c to i64
+  %or = or i64 %and, %c.ext
+  %or.trunc = trunc i64 %or to i8
+  %gep = getelementptr i8, ptr %dst, i16 %iv
+  store i8 %or.trunc, ptr %gep, align 1
+  %iv.next = add i16 %iv, 1
+  %ec = icmp eq i16 %iv.next, 1000
+  br i1 %ec, label %exit, label %loop
+
+exit:
+  ret i32 0
+}
+
+define void @ext_cmp(ptr %src.1, ptr %src.2, ptr noalias %dst) {
+; CHECK-LABEL: define void @ext_cmp(
+; CHECK-SAME: ptr [[SRC_1:%.*]], ptr [[SRC_2:%.*]], ptr noalias [[DST:%.*]]) {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    br i1 false, label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
+; CHECK:       vector.ph:
+; CHECK-NEXT:    br label [[VECTOR_BODY:%.*]]
+; CHECK:       vector.body:
+; CHECK-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
+; CHECK-NEXT:    [[TMP0:%.*]] = add i64 [[INDEX]], 0
+; CHECK-NEXT:    [[TMP1:%.*]] = getelementptr inbounds i16, ptr [[SRC_1]], i64 [[TMP0]]
+; CHECK-NEXT:    [[TMP2:%.*]] = getelementptr inbounds i16, ptr [[TMP1]], i32 0
+; CHECK-NEXT:    [[WIDE_LOAD:%.*]] = load <4 x i16>, ptr [[TMP2]], align 2
+; CHECK-NEXT:    [[TMP3:%.*]] = icmp sgt <4 x i16> zeroinitializer, [[WIDE_LOAD]]
+; CHECK-NEXT:    [[TMP4:%.*]] = getelementptr inbounds i8, ptr [[SRC_2]], i64 [[TMP0]]
+; CHECK-NEXT:    [[TMP5:%.*]] = getelementptr inbounds i8, ptr [[TMP4]], i32 0
+; CHECK-NEXT:    [[WIDE_LOAD1:%.*]] = load <4 x i8>, ptr [[TMP5]], align 2
+; CHECK-NEXT:    [[TMP6:%.*]] = zext <4 x i8> [[WIDE_LOAD1]] to <4 x i16>
+; CHECK-NEXT:    [[TMP7:%.*]] = select <4 x i1> [[TMP3]], <4 x i16> zeroinitializer, <4 x i16> [[TMP6]]
+; CHECK-NEXT:    [[TMP8:%.*]] = and <4 x i16> [[TMP7]], zeroinitializer
+; CHECK-NEXT:    [[TMP9:%.*]] = getelementptr inbounds i16, ptr [[DST]], i64 [[TMP0]]
+; CHECK-NEXT:    [[TMP10:%.*]] = getelementptr inbounds i16, ptr [[TMP9]], i32 0
+; CHECK-NEXT:    store <4 x i16> [[TMP8]], ptr [[TMP10]], align 2
+; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 4
+; CHECK-NEXT:    [[TMP11:%.*]] = icmp eq i64 [[INDEX_NEXT]], 1000
+; CHECK-NEXT:    br i1 [[TMP11]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP6:![0-9]+]]
+; CHECK:       middle.block:
+; CHECK-NEXT:    br i1 true, label [[EXIT:%.*]], label [[SCALAR_PH]]
+; CHECK:       scalar.ph:
+; CHECK-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ 1000, [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ]
+; CHECK-NEXT:    br label [[LOOP:%.*]]
+; CHECK:       loop:
+; CHECK-NEXT:    [[IV:%.*]] = phi i64 [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ], [ [[IV_NEXT:%.*]], [[LOOP]] ]
+; CHECK-NEXT:    [[GEP_SRC_1:%.*]] = getelementptr inbounds i16, ptr [[SRC_1]], i64 [[IV]]
+; CHECK-NEXT:    [[I2:%.*]] = load i16, ptr [[GEP_SRC_1]], align 2
+; CHECK-NEXT:    [[I3:%.*]] = sext i16 [[I2]] to i32
+; CHECK-NEXT:    [[C_1:%.*]] = icmp sgt i32 0, [[I3]]
+; CHECK-NEXT:    [[GEP_SRC_2:%.*]] = getelementptr inbounds i8, ptr [[SRC_2]], i64 [[IV]]
+; CHECK-NEXT:    [[I4:%.*]] = load i8, ptr [[GEP_SRC_2]], align 2
+; CHECK-NEXT:    [[I5:%.*]] = zext i8 [[I4]] to i32
+; CHECK-NEXT:    [[I6:%.*]] = select i1 [[C_1]], i32 0, i32 [[I5]]
+; CHECK-NEXT:    [[I7:%.*]] = and i32 [[I6]], 0
+; CHECK-NEXT:    [[I8:%.*]] = trunc nuw nsw i32 [[I7]] to i16
+; CHECK-NEXT:    [[GEP_DST:%.*]] = getelementptr inbounds i16, ptr [[DST]], i64 [[IV]]
+; CHECK-NEXT:    store i16 [[I8]], ptr [[GEP_DST]], align 2
+; CHECK-NEXT:    [[IV_NEXT]] = add nsw i64 [[IV]], 1
+; CHECK-NEXT:    [[EC:%.*]] = icmp eq i64 [[IV_NEXT]], 1000
+; CHECK-NEXT:    br i1 [[EC]], label [[EXIT]], label [[LOOP]], !llvm.loop [[LOOP7:![0-9]+]]
+; CHECK:       exit:
+; CHECK-NEXT:    ret void
+;
+entry:
+  br label %loop
+
+loop:
+  %iv = phi i64 [ 0, %entry ], [ %iv.next, %loop ]
+  %gep.src.1 = getelementptr inbounds i16, ptr %src.1, i64 %iv
+  %i2 = load i16, ptr %gep.src.1, align 2
+  %i3 = sext i16 %i2 to i32
+  %c.1 = icmp sgt i32 0, %i3
+  %gep.src.2 = getelementptr inbounds i8, ptr %src.2, i64 %iv
+  %i4 = load i8, ptr %gep.src.2, align 2
+  %i5 = zext i8 %i4 to i32
+  %i6 = select i1 %c.1, i32 0, i32 %i5
+  %i7 = and i32 %i6, 0
+  %i8 = trunc nuw nsw i32 %i7 to i16
+  %gep.dst = getelementptr inbounds i16, ptr %dst, i64 %iv
+  store i16 %i8, ptr %gep.dst, align 2
+  %iv.next = add nsw i64 %iv, 1
+  %ec = icmp eq i64 %iv.next, 1000
+  br i1 %ec, label %exit, label %loop
+
+exit:
+  ret void
+}
+;.
+; CHECK: [[LOOP0]] = distinct !{[[LOOP0]], [[META1:![0-9]+]], [[META2:![0-9]+]]}
+; CHECK: [[META1]] = !{!"llvm.loop.isvectorized", i32 1}
+; CHECK: [[META2]] = !{!"llvm.loop.unroll.runtime.disable"}
+; CHECK: [[LOOP3]] = distinct !{[[LOOP3]], [[META2]], [[META1]]}
+; CHECK: [[LOOP4]] = distinct !{[[LOOP4]], [[META1]], [[META2]]}
+; CHECK: [[LOOP5]] = distinct !{[[LOOP5]], [[META2]], [[META1]]}
+; CHECK: [[LOOP6]] = distinct !{[[LOOP6]], [[META1]], [[META2]]}
+; CHECK: [[LOOP7]] = distinct !{[[LOOP7]], [[META2]], [[META1]]}
+;.
diff --git a/llvm/test/Transforms/LoopVectorize/unused-blend-mask-for-first-operand.ll b/llvm/test/Transforms/LoopVectorize/unused-blend-mask-for-first-operand.ll
index c622925510dd..d0c74897f264 100644
--- a/llvm/test/Transforms/LoopVectorize/unused-blend-mask-for-first-operand.ll
+++ b/llvm/test/Transforms/LoopVectorize/unused-blend-mask-for-first-operand.ll
@@ -10,17 +10,13 @@ define void @test_not_first_lane_only_constant(ptr %A, ptr noalias %B)  {
 ; CHECK-NEXT:  entry:
 ; CHECK-NEXT:    br i1 false, label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
 ; CHECK:       vector.ph:
-; CHECK-NEXT:    [[BROADCAST_SPLATINSERT3:%.*]] = insertelement <4 x ptr> poison, ptr [[B]], i64 0
-; CHECK-NEXT:    [[BROADCAST_SPLAT4:%.*]] = shufflevector <4 x ptr> [[BROADCAST_SPLATINSERT3]], <4 x ptr> poison, <4 x i32> zeroinitializer
 ; CHECK-NEXT:    br label [[VECTOR_BODY:%.*]]
 ; CHECK:       vector.body:
 ; CHECK-NEXT:    [[INDEX:%.*]] = phi i32 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
 ; CHECK-NEXT:    [[OFFSET_IDX:%.*]] = trunc i32 [[INDEX]] to i16
 ; CHECK-NEXT:    [[TMP0:%.*]] = add i16 [[OFFSET_IDX]], 0
 ; CHECK-NEXT:    [[TMP1:%.*]] = getelementptr inbounds i16, ptr [[A]], i16 [[TMP0]]
-; CHECK-NEXT:    [[PREDPHI:%.*]] = select <4 x i1> zeroinitializer, <4 x ptr> poison, <4 x ptr> [[BROADCAST_SPLAT4]]
-; CHECK-NEXT:    [[TMP12:%.*]] = extractelement <4 x ptr> [[PREDPHI]], i32 0
-; CHECK-NEXT:    [[TMP13:%.*]] = load i16, ptr [[TMP12]], align 2
+; CHECK-NEXT:    [[TMP13:%.*]] = load i16, ptr %B, align 2
 ; CHECK-NEXT:    [[BROADCAST_SPLATINSERT5:%.*]] = insertelement <4 x i16> poison, i16 [[TMP13]], i64 0
 ; CHECK-NEXT:    [[BROADCAST_SPLAT6:%.*]] = shufflevector <4 x i16> [[BROADCAST_SPLATINSERT5]], <4 x i16> poison, <4 x i32> zeroinitializer
 ; CHECK-NEXT:    [[TMP2:%.*]] = getelementptr inbounds i16, ptr [[TMP1]], i32 0
diff --git a/llvm/test/Transforms/LoopVectorize/version-stride-with-integer-casts.ll b/llvm/test/Transforms/LoopVectorize/version-stride-with-integer-casts.ll
new file mode 100644
index 000000000000..d09066fa2d70
--- /dev/null
+++ b/llvm/test/Transforms/LoopVectorize/version-stride-with-integer-casts.ll
@@ -0,0 +1,428 @@
+; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --version 4
+; RUN: opt -passes=loop-vectorize -force-vector-width=4 -S %s | FileCheck %s
+
+target datalayout = "e-m:o-p270:32:32-p271:32:32-p272:64:64-i64:64-i128:128-f80:128-n8:16:32:64-S128"
+
+define void @test_versioned_with_sext_use(i32 %offset, ptr %dst) {
+; CHECK-LABEL: define void @test_versioned_with_sext_use(
+; CHECK-SAME: i32 [[OFFSET:%.*]], ptr [[DST:%.*]]) {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[OFFSET_EXT:%.*]] = sext i32 [[OFFSET]] to i64
+; CHECK-NEXT:    br label [[OUTER_HEADER:%.*]]
+; CHECK:       outer.header.loopexit:
+; CHECK-NEXT:    [[IV_2_NEXT_LCSSA:%.*]] = phi i64 [ [[IV_2_NEXT:%.*]], [[INNER_LOOP:%.*]] ], [ [[IND_END:%.*]], [[MIDDLE_BLOCK:%.*]] ]
+; CHECK-NEXT:    br label [[OUTER_HEADER]]
+; CHECK:       outer.header:
+; CHECK-NEXT:    [[IV_1:%.*]] = phi i64 [ 0, [[ENTRY:%.*]] ], [ [[IV_2_NEXT_LCSSA]], [[OUTER_HEADER_LOOPEXIT:%.*]] ]
+; CHECK-NEXT:    [[C:%.*]] = call i1 @cond()
+; CHECK-NEXT:    br i1 [[C]], label [[INNER_LOOP_PREHEADER:%.*]], label [[EXIT:%.*]]
+; CHECK:       inner.loop.preheader:
+; CHECK-NEXT:    br i1 false, label [[SCALAR_PH:%.*]], label [[VECTOR_SCEVCHECK:%.*]]
+; CHECK:       vector.scevcheck:
+; CHECK-NEXT:    [[IDENT_CHECK:%.*]] = icmp ne i32 [[OFFSET]], 1
+; CHECK-NEXT:    br i1 [[IDENT_CHECK]], label [[SCALAR_PH]], label [[VECTOR_PH:%.*]]
+; CHECK:       vector.ph:
+; CHECK-NEXT:    [[TMP0:%.*]] = mul i64 200, [[OFFSET_EXT]]
+; CHECK-NEXT:    [[IND_END]] = add i64 [[IV_1]], [[TMP0]]
+; CHECK-NEXT:    br label [[VECTOR_BODY:%.*]]
+; CHECK:       vector.body:
+; CHECK-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
+; CHECK-NEXT:    [[TMP1:%.*]] = mul i64 [[INDEX]], [[OFFSET_EXT]]
+; CHECK-NEXT:    [[OFFSET_IDX:%.*]] = add i64 [[IV_1]], [[TMP1]]
+; CHECK-NEXT:    [[TMP2:%.*]] = mul i64 0, [[OFFSET_EXT]]
+; CHECK-NEXT:    [[TMP3:%.*]] = add i64 [[OFFSET_IDX]], [[TMP2]]
+; CHECK-NEXT:    [[TMP4:%.*]] = getelementptr i32, ptr [[DST]], i64 [[TMP3]]
+; CHECK-NEXT:    [[TMP5:%.*]] = getelementptr i32, ptr [[TMP4]], i32 0
+; CHECK-NEXT:    store <4 x i32> zeroinitializer, ptr [[TMP5]], align 8
+; CHECK-NEXT:    [[TMP6:%.*]] = add i64 [[TMP3]], 1
+; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 4
+; CHECK-NEXT:    [[TMP7:%.*]] = icmp eq i64 [[INDEX_NEXT]], 200
+; CHECK-NEXT:    br i1 [[TMP7]], label [[MIDDLE_BLOCK]], label [[VECTOR_BODY]], !llvm.loop [[LOOP0:![0-9]+]]
+; CHECK:       middle.block:
+; CHECK-NEXT:    br i1 false, label [[OUTER_HEADER_LOOPEXIT]], label [[SCALAR_PH]]
+; CHECK:       scalar.ph:
+; CHECK-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ [[IND_END]], [[MIDDLE_BLOCK]] ], [ [[IV_1]], [[INNER_LOOP_PREHEADER]] ], [ [[IV_1]], [[VECTOR_SCEVCHECK]] ]
+; CHECK-NEXT:    [[BC_RESUME_VAL1:%.*]] = phi i32 [ 200, [[MIDDLE_BLOCK]] ], [ 0, [[INNER_LOOP_PREHEADER]] ], [ 0, [[VECTOR_SCEVCHECK]] ]
+; CHECK-NEXT:    br label [[INNER_LOOP]]
+; CHECK:       inner.loop:
+; CHECK-NEXT:    [[IV_2:%.*]] = phi i64 [ [[IV_2_NEXT]], [[INNER_LOOP]] ], [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ]
+; CHECK-NEXT:    [[IV_3:%.*]] = phi i32 [ [[IV_3_NEXT:%.*]], [[INNER_LOOP]] ], [ [[BC_RESUME_VAL1]], [[SCALAR_PH]] ]
+; CHECK-NEXT:    [[GEP:%.*]] = getelementptr i32, ptr [[DST]], i64 [[IV_2]]
+; CHECK-NEXT:    store i32 0, ptr [[GEP]], align 8
+; CHECK-NEXT:    [[IV_2_NEXT]] = add i64 [[IV_2]], [[OFFSET_EXT]]
+; CHECK-NEXT:    [[IV_3_NEXT]] = add i32 [[IV_3]], 1
+; CHECK-NEXT:    [[EC:%.*]] = icmp eq i32 [[IV_3]], 200
+; CHECK-NEXT:    br i1 [[EC]], label [[OUTER_HEADER_LOOPEXIT]], label [[INNER_LOOP]], !llvm.loop [[LOOP3:![0-9]+]]
+; CHECK:       exit:
+; CHECK-NEXT:    ret void
+;
+entry:
+  %offset.ext = sext i32 %offset to i64
+  br label %outer.header
+
+outer.header:
+  %iv.1 = phi i64 [ 0, %entry ], [ %iv.2.next, %inner.loop ]
+  %c = call i1 @cond()
+  br i1 %c, label %inner.loop, label %exit
+
+inner.loop:
+  %iv.2 = phi i64 [ %iv.1, %outer.header ], [ %iv.2.next, %inner.loop ]
+  %iv.3 = phi i32 [ 0, %outer.header ], [ %iv.3.next, %inner.loop ]
+  %gep = getelementptr i32, ptr %dst, i64 %iv.2
+  store i32 0, ptr %gep, align 8
+  %iv.2.next = add i64 %iv.2, %offset.ext
+  %iv.3.next = add i32 %iv.3, 1
+  %ec = icmp eq i32 %iv.3, 200
+  br i1 %ec, label %outer.header, label %inner.loop
+
+exit:
+  ret void
+}
+
+define void @test_versioned_with_zext_use(i32 %offset, ptr %dst) {
+; CHECK-LABEL: define void @test_versioned_with_zext_use(
+; CHECK-SAME: i32 [[OFFSET:%.*]], ptr [[DST:%.*]]) {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[OFFSET_EXT:%.*]] = zext i32 [[OFFSET]] to i64
+; CHECK-NEXT:    br label [[OUTER_HEADER:%.*]]
+; CHECK:       outer.header.loopexit:
+; CHECK-NEXT:    [[IV_2_NEXT_LCSSA:%.*]] = phi i64 [ [[IV_2_NEXT:%.*]], [[INNER_LOOP:%.*]] ], [ [[IND_END:%.*]], [[MIDDLE_BLOCK:%.*]] ]
+; CHECK-NEXT:    br label [[OUTER_HEADER]]
+; CHECK:       outer.header:
+; CHECK-NEXT:    [[IV_1:%.*]] = phi i64 [ 0, [[ENTRY:%.*]] ], [ [[IV_2_NEXT_LCSSA]], [[OUTER_HEADER_LOOPEXIT:%.*]] ]
+; CHECK-NEXT:    [[C:%.*]] = call i1 @cond()
+; CHECK-NEXT:    br i1 [[C]], label [[INNER_LOOP_PREHEADER:%.*]], label [[EXIT:%.*]]
+; CHECK:       inner.loop.preheader:
+; CHECK-NEXT:    br i1 false, label [[SCALAR_PH:%.*]], label [[VECTOR_SCEVCHECK:%.*]]
+; CHECK:       vector.scevcheck:
+; CHECK-NEXT:    [[IDENT_CHECK:%.*]] = icmp ne i32 [[OFFSET]], 1
+; CHECK-NEXT:    br i1 [[IDENT_CHECK]], label [[SCALAR_PH]], label [[VECTOR_PH:%.*]]
+; CHECK:       vector.ph:
+; CHECK-NEXT:    [[TMP0:%.*]] = mul i64 200, [[OFFSET_EXT]]
+; CHECK-NEXT:    [[IND_END]] = add i64 [[IV_1]], [[TMP0]]
+; CHECK-NEXT:    br label [[VECTOR_BODY:%.*]]
+; CHECK:       vector.body:
+; CHECK-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
+; CHECK-NEXT:    [[TMP1:%.*]] = mul i64 [[INDEX]], [[OFFSET_EXT]]
+; CHECK-NEXT:    [[OFFSET_IDX:%.*]] = add i64 [[IV_1]], [[TMP1]]
+; CHECK-NEXT:    [[TMP2:%.*]] = mul i64 0, [[OFFSET_EXT]]
+; CHECK-NEXT:    [[TMP3:%.*]] = add i64 [[OFFSET_IDX]], [[TMP2]]
+; CHECK-NEXT:    [[TMP4:%.*]] = getelementptr i32, ptr [[DST]], i64 [[TMP3]]
+; CHECK-NEXT:    [[TMP5:%.*]] = getelementptr i32, ptr [[TMP4]], i32 0
+; CHECK-NEXT:    store <4 x i32> zeroinitializer, ptr [[TMP5]], align 8
+; CHECK-NEXT:    [[TMP6:%.*]] = add i64 [[TMP3]], 1
+; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 4
+; CHECK-NEXT:    [[TMP7:%.*]] = icmp eq i64 [[INDEX_NEXT]], 200
+; CHECK-NEXT:    br i1 [[TMP7]], label [[MIDDLE_BLOCK]], label [[VECTOR_BODY]], !llvm.loop [[LOOP4:![0-9]+]]
+; CHECK:       middle.block:
+; CHECK-NEXT:    br i1 false, label [[OUTER_HEADER_LOOPEXIT]], label [[SCALAR_PH]]
+; CHECK:       scalar.ph:
+; CHECK-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ [[IND_END]], [[MIDDLE_BLOCK]] ], [ [[IV_1]], [[INNER_LOOP_PREHEADER]] ], [ [[IV_1]], [[VECTOR_SCEVCHECK]] ]
+; CHECK-NEXT:    [[BC_RESUME_VAL1:%.*]] = phi i32 [ 200, [[MIDDLE_BLOCK]] ], [ 0, [[INNER_LOOP_PREHEADER]] ], [ 0, [[VECTOR_SCEVCHECK]] ]
+; CHECK-NEXT:    br label [[INNER_LOOP]]
+; CHECK:       inner.loop:
+; CHECK-NEXT:    [[IV_2:%.*]] = phi i64 [ [[IV_2_NEXT]], [[INNER_LOOP]] ], [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ]
+; CHECK-NEXT:    [[IV_3:%.*]] = phi i32 [ [[IV_3_NEXT:%.*]], [[INNER_LOOP]] ], [ [[BC_RESUME_VAL1]], [[SCALAR_PH]] ]
+; CHECK-NEXT:    [[GEP:%.*]] = getelementptr i32, ptr [[DST]], i64 [[IV_2]]
+; CHECK-NEXT:    store i32 0, ptr [[GEP]], align 8
+; CHECK-NEXT:    [[IV_2_NEXT]] = add i64 [[IV_2]], [[OFFSET_EXT]]
+; CHECK-NEXT:    [[IV_3_NEXT]] = add i32 [[IV_3]], 1
+; CHECK-NEXT:    [[EC:%.*]] = icmp eq i32 [[IV_3]], 200
+; CHECK-NEXT:    br i1 [[EC]], label [[OUTER_HEADER_LOOPEXIT]], label [[INNER_LOOP]], !llvm.loop [[LOOP5:![0-9]+]]
+; CHECK:       exit:
+; CHECK-NEXT:    ret void
+;
+entry:
+  %offset.ext = zext i32 %offset to i64
+  br label %outer.header
+
+outer.header:
+  %iv.1 = phi i64 [ 0, %entry ], [ %iv.2.next, %inner.loop ]
+  %c = call i1 @cond()
+  br i1 %c, label %inner.loop, label %exit
+
+inner.loop:
+  %iv.2 = phi i64 [ %iv.1, %outer.header ], [ %iv.2.next, %inner.loop ]
+  %iv.3 = phi i32 [ 0, %outer.header ], [ %iv.3.next, %inner.loop ]
+  %gep = getelementptr i32, ptr %dst, i64 %iv.2
+  store i32 0, ptr %gep, align 8
+  %iv.2.next = add i64 %iv.2, %offset.ext
+  %iv.3.next = add i32 %iv.3, 1
+  %ec = icmp eq i32 %iv.3, 200
+  br i1 %ec, label %outer.header, label %inner.loop
+
+exit:
+  ret void
+}
+
+define void @versioned_sext_use_in_gep(i32 %scale, ptr %dst, i64 %scale.2) {
+; CHECK-LABEL: define void @versioned_sext_use_in_gep(
+; CHECK-SAME: i32 [[SCALE:%.*]], ptr [[DST:%.*]], i64 [[SCALE_2:%.*]]) {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[SCALE_EXT:%.*]] = sext i32 [[SCALE]] to i64
+; CHECK-NEXT:    br i1 false, label [[SCALAR_PH:%.*]], label [[VECTOR_SCEVCHECK:%.*]]
+; CHECK:       vector.scevcheck:
+; CHECK-NEXT:    [[IDENT_CHECK:%.*]] = icmp ne i32 [[SCALE]], 1
+; CHECK-NEXT:    br i1 [[IDENT_CHECK]], label [[SCALAR_PH]], label [[VECTOR_PH:%.*]]
+; CHECK:       vector.ph:
+; CHECK-NEXT:    br label [[VECTOR_BODY:%.*]]
+; CHECK:       vector.body:
+; CHECK-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
+; CHECK-NEXT:    [[TMP10:%.*]] = add i64 [[INDEX]], 0
+; CHECK-NEXT:    [[TMP12:%.*]] = add i64 [[INDEX]], 1
+; CHECK-NEXT:    [[TMP14:%.*]] = add i64 [[INDEX]], 2
+; CHECK-NEXT:    [[TMP16:%.*]] = add i64 [[INDEX]], 3
+; CHECK-NEXT:    [[TMP11:%.*]] = getelementptr i8, ptr [[DST]], i64 [[TMP10]]
+; CHECK-NEXT:    [[TMP13:%.*]] = getelementptr i8, ptr [[DST]], i64 [[TMP12]]
+; CHECK-NEXT:    [[TMP15:%.*]] = getelementptr i8, ptr [[DST]], i64 [[TMP14]]
+; CHECK-NEXT:    [[TMP17:%.*]] = getelementptr i8, ptr [[DST]], i64 [[TMP16]]
+; CHECK-NEXT:    [[TMP8:%.*]] = getelementptr i8, ptr [[DST]], i64 [[SCALE_2]]
+; CHECK-NEXT:    store ptr [[TMP8]], ptr [[TMP11]], align 8
+; CHECK-NEXT:    store ptr [[TMP8]], ptr [[TMP13]], align 8
+; CHECK-NEXT:    store ptr [[TMP8]], ptr [[TMP15]], align 8
+; CHECK-NEXT:    store ptr [[TMP8]], ptr [[TMP17]], align 8
+; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 4
+; CHECK-NEXT:    [[TMP18:%.*]] = icmp eq i64 [[INDEX_NEXT]], 256
+; CHECK-NEXT:    br i1 [[TMP18]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP6:![0-9]+]]
+; CHECK:       middle.block:
+; CHECK-NEXT:    br i1 true, label [[EXIT:%.*]], label [[SCALAR_PH]]
+; CHECK:       scalar.ph:
+; CHECK-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ 256, [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ], [ 0, [[VECTOR_SCEVCHECK]] ]
+; CHECK-NEXT:    br label [[LOOP:%.*]]
+; CHECK:       loop:
+; CHECK-NEXT:    [[IV:%.*]] = phi i64 [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ], [ [[IV_NEXT:%.*]], [[LOOP]] ]
+; CHECK-NEXT:    [[IV_MUL:%.*]] = mul i64 [[IV]], [[SCALE_EXT]]
+; CHECK-NEXT:    [[GEP_1:%.*]] = getelementptr i8, ptr [[DST]], i64 [[IV_MUL]]
+; CHECK-NEXT:    [[IV_NEXT]] = add i64 [[IV]], 1
+; CHECK-NEXT:    [[SCALE_MUL:%.*]] = mul i64 [[SCALE_EXT]], [[SCALE_2]]
+; CHECK-NEXT:    [[GEP_2:%.*]] = getelementptr i8, ptr [[DST]], i64 [[SCALE_MUL]]
+; CHECK-NEXT:    store ptr [[GEP_2]], ptr [[GEP_1]], align 8
+; CHECK-NEXT:    [[EC:%.*]] = icmp eq i64 [[IV_NEXT]], 256
+; CHECK-NEXT:    br i1 [[EC]], label [[EXIT]], label [[LOOP]], !llvm.loop [[LOOP7:![0-9]+]]
+; CHECK:       exit:
+; CHECK-NEXT:    ret void
+;
+entry:
+  %scale.ext = sext i32 %scale to i64
+  br label %loop
+
+loop:
+  %iv = phi i64 [ 0, %entry ], [ %iv.next, %loop ]
+  %iv.mul  = mul i64 %iv, %scale.ext
+  %gep.1 = getelementptr i8, ptr %dst, i64 %iv.mul
+  %iv.next = add i64 %iv, 1
+  %scale.mul = mul i64 %scale.ext, %scale.2
+  %gep.2 = getelementptr i8, ptr %dst, i64 %scale.mul
+  store ptr %gep.2, ptr %gep.1, align 8
+  %ec = icmp eq i64 %iv.next, 256
+  br i1 %ec, label %exit, label %loop
+
+exit:
+  ret void
+}
+
+declare i1 @cond()
+
+define void @test_versioned_with_different_uses(i32 %offset, ptr noalias %dst.1, ptr %dst.2) {
+; CHECK-LABEL: define void @test_versioned_with_different_uses(
+; CHECK-SAME: i32 [[OFFSET:%.*]], ptr noalias [[DST_1:%.*]], ptr [[DST_2:%.*]]) {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[OFFSET_EXT:%.*]] = zext i32 [[OFFSET]] to i64
+; CHECK-NEXT:    br label [[OUTER_HEADER:%.*]]
+; CHECK:       outer.header.loopexit:
+; CHECK-NEXT:    [[IV_2_NEXT_LCSSA:%.*]] = phi i64 [ [[IV_2_NEXT:%.*]], [[INNER_LOOP:%.*]] ], [ [[IND_END:%.*]], [[MIDDLE_BLOCK:%.*]] ]
+; CHECK-NEXT:    br label [[OUTER_HEADER]]
+; CHECK:       outer.header:
+; CHECK-NEXT:    [[IV_1:%.*]] = phi i64 [ 0, [[ENTRY:%.*]] ], [ [[IV_2_NEXT_LCSSA]], [[OUTER_HEADER_LOOPEXIT:%.*]] ]
+; CHECK-NEXT:    [[C:%.*]] = call i1 @cond()
+; CHECK-NEXT:    br i1 [[C]], label [[INNER_LOOP_PREHEADER:%.*]], label [[EXIT:%.*]]
+; CHECK:       inner.loop.preheader:
+; CHECK-NEXT:    br i1 false, label [[SCALAR_PH:%.*]], label [[VECTOR_SCEVCHECK:%.*]]
+; CHECK:       vector.scevcheck:
+; CHECK-NEXT:    [[IDENT_CHECK:%.*]] = icmp ne i32 [[OFFSET]], 1
+; CHECK-NEXT:    br i1 [[IDENT_CHECK]], label [[SCALAR_PH]], label [[VECTOR_PH:%.*]]
+; CHECK:       vector.ph:
+; CHECK-NEXT:    [[TMP0:%.*]] = mul i64 200, [[OFFSET_EXT]]
+; CHECK-NEXT:    [[IND_END]] = add i64 [[IV_1]], [[TMP0]]
+; CHECK-NEXT:    br label [[VECTOR_BODY:%.*]]
+; CHECK:       vector.body:
+; CHECK-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
+; CHECK-NEXT:    [[TMP1:%.*]] = mul i64 [[INDEX]], [[OFFSET_EXT]]
+; CHECK-NEXT:    [[OFFSET_IDX:%.*]] = add i64 [[IV_1]], [[TMP1]]
+; CHECK-NEXT:    [[TMP2:%.*]] = mul i64 0, [[OFFSET_EXT]]
+; CHECK-NEXT:    [[TMP3:%.*]] = add i64 [[OFFSET_IDX]], [[TMP2]]
+; CHECK-NEXT:    [[OFFSET_IDX2:%.*]] = trunc i64 [[INDEX]] to i32
+; CHECK-NEXT:    [[TMP4:%.*]] = add i32 [[OFFSET_IDX2]], 0
+; CHECK-NEXT:    [[TMP5:%.*]] = add i32 [[OFFSET_IDX2]], 1
+; CHECK-NEXT:    [[TMP6:%.*]] = add i32 [[OFFSET_IDX2]], 2
+; CHECK-NEXT:    [[TMP7:%.*]] = add i32 [[OFFSET_IDX2]], 3
+; CHECK-NEXT:    [[TMP8:%.*]] = getelementptr i8, ptr [[DST_1]], i32 [[TMP4]]
+; CHECK-NEXT:    [[TMP9:%.*]] = getelementptr i8, ptr [[DST_1]], i32 [[TMP5]]
+; CHECK-NEXT:    [[TMP10:%.*]] = getelementptr i8, ptr [[DST_1]], i32 [[TMP6]]
+; CHECK-NEXT:    [[TMP11:%.*]] = getelementptr i8, ptr [[DST_1]], i32 [[TMP7]]
+; CHECK-NEXT:    store i32 0, ptr [[TMP8]], align 8
+; CHECK-NEXT:    store i32 0, ptr [[TMP9]], align 8
+; CHECK-NEXT:    store i32 0, ptr [[TMP10]], align 8
+; CHECK-NEXT:    store i32 0, ptr [[TMP11]], align 8
+; CHECK-NEXT:    [[TMP12:%.*]] = getelementptr i32, ptr [[DST_2]], i64 [[TMP3]]
+; CHECK-NEXT:    [[TMP13:%.*]] = getelementptr i32, ptr [[TMP12]], i32 0
+; CHECK-NEXT:    store <4 x i32> zeroinitializer, ptr [[TMP13]], align 8
+; CHECK-NEXT:    [[TMP14:%.*]] = add i64 [[TMP3]], 1
+; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 4
+; CHECK-NEXT:    [[TMP15:%.*]] = icmp eq i64 [[INDEX_NEXT]], 200
+; CHECK-NEXT:    br i1 [[TMP15]], label [[MIDDLE_BLOCK]], label [[VECTOR_BODY]], !llvm.loop [[LOOP8:![0-9]+]]
+; CHECK:       middle.block:
+; CHECK-NEXT:    br i1 false, label [[OUTER_HEADER_LOOPEXIT]], label [[SCALAR_PH]]
+; CHECK:       scalar.ph:
+; CHECK-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ [[IND_END]], [[MIDDLE_BLOCK]] ], [ [[IV_1]], [[INNER_LOOP_PREHEADER]] ], [ [[IV_1]], [[VECTOR_SCEVCHECK]] ]
+; CHECK-NEXT:    [[BC_RESUME_VAL1:%.*]] = phi i32 [ 200, [[MIDDLE_BLOCK]] ], [ 0, [[INNER_LOOP_PREHEADER]] ], [ 0, [[VECTOR_SCEVCHECK]] ]
+; CHECK-NEXT:    br label [[INNER_LOOP]]
+; CHECK:       inner.loop:
+; CHECK-NEXT:    [[IV_2:%.*]] = phi i64 [ [[IV_2_NEXT]], [[INNER_LOOP]] ], [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ]
+; CHECK-NEXT:    [[IV_3:%.*]] = phi i32 [ [[IV_3_NEXT:%.*]], [[INNER_LOOP]] ], [ [[BC_RESUME_VAL1]], [[SCALAR_PH]] ]
+; CHECK-NEXT:    [[IV_MUL:%.*]] = mul i32 [[IV_3]], [[OFFSET]]
+; CHECK-NEXT:    [[GEP_MUL:%.*]] = getelementptr i8, ptr [[DST_1]], i32 [[IV_MUL]]
+; CHECK-NEXT:    store i32 0, ptr [[GEP_MUL]], align 8
+; CHECK-NEXT:    [[GEP:%.*]] = getelementptr i32, ptr [[DST_2]], i64 [[IV_2]]
+; CHECK-NEXT:    store i32 0, ptr [[GEP]], align 8
+; CHECK-NEXT:    [[IV_2_NEXT]] = add i64 [[IV_2]], [[OFFSET_EXT]]
+; CHECK-NEXT:    [[IV_3_NEXT]] = add i32 [[IV_3]], 1
+; CHECK-NEXT:    [[EC:%.*]] = icmp eq i32 [[IV_3]], 200
+; CHECK-NEXT:    br i1 [[EC]], label [[OUTER_HEADER_LOOPEXIT]], label [[INNER_LOOP]], !llvm.loop [[LOOP9:![0-9]+]]
+; CHECK:       exit:
+; CHECK-NEXT:    ret void
+;
+entry:
+  %offset.ext = zext i32 %offset to i64
+  br label %outer.header
+
+outer.header:
+  %iv.1 = phi i64 [ 0, %entry ], [ %iv.2.next, %inner.loop ]
+  %c = call i1 @cond()
+  br i1 %c, label %inner.loop, label %exit
+
+inner.loop:
+  %iv.2 = phi i64 [ %iv.1, %outer.header ], [ %iv.2.next, %inner.loop ]
+  %iv.3 = phi i32 [ 0, %outer.header ], [ %iv.3.next, %inner.loop ]
+  %iv.mul  = mul i32 %iv.3, %offset
+  %gep.mul = getelementptr i8, ptr %dst.1, i32 %iv.mul
+  store i32 0, ptr %gep.mul, align 8
+  %gep = getelementptr i32, ptr %dst.2, i64 %iv.2
+  store i32 0, ptr %gep, align 8
+  %iv.2.next = add i64 %iv.2, %offset.ext
+  %iv.3.next = add i32 %iv.3, 1
+  %ec = icmp eq i32 %iv.3, 200
+  br i1 %ec, label %outer.header, label %inner.loop
+
+exit:
+  ret void
+}
+
+define void @test_versioned_with_non_ex_use(i32 %offset, ptr noalias %dst.1, ptr %dst.2) {
+; CHECK-LABEL: define void @test_versioned_with_non_ex_use(
+; CHECK-SAME: i32 [[OFFSET:%.*]], ptr noalias [[DST_1:%.*]], ptr [[DST_2:%.*]]) {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[OFFSET_EXT:%.*]] = zext i32 [[OFFSET]] to i64
+; CHECK-NEXT:    [[ADD:%.*]] = add i32 [[OFFSET]], 3
+; CHECK-NEXT:    br i1 false, label [[SCALAR_PH:%.*]], label [[VECTOR_SCEVCHECK:%.*]]
+; CHECK:       vector.scevcheck:
+; CHECK-NEXT:    [[TMP0:%.*]] = sub i32 -3, [[OFFSET]]
+; CHECK-NEXT:    [[TMP1:%.*]] = icmp slt i32 [[ADD]], 0
+; CHECK-NEXT:    [[TMP2:%.*]] = select i1 [[TMP1]], i32 [[TMP0]], i32 [[ADD]]
+; CHECK-NEXT:    [[MUL:%.*]] = call { i32, i1 } @llvm.umul.with.overflow.i32(i32 [[TMP2]], i32 200)
+; CHECK-NEXT:    [[MUL_RESULT:%.*]] = extractvalue { i32, i1 } [[MUL]], 0
+; CHECK-NEXT:    [[MUL_OVERFLOW:%.*]] = extractvalue { i32, i1 } [[MUL]], 1
+; CHECK-NEXT:    [[TMP3:%.*]] = sub i32 0, [[MUL_RESULT]]
+; CHECK-NEXT:    [[TMP4:%.*]] = icmp slt i32 [[MUL_RESULT]], 0
+; CHECK-NEXT:    [[TMP5:%.*]] = icmp sgt i32 [[TMP3]], 0
+; CHECK-NEXT:    [[TMP6:%.*]] = select i1 [[TMP1]], i1 [[TMP5]], i1 [[TMP4]]
+; CHECK-NEXT:    [[TMP7:%.*]] = or i1 [[TMP6]], [[MUL_OVERFLOW]]
+; CHECK-NEXT:    [[IDENT_CHECK:%.*]] = icmp ne i32 [[OFFSET]], 1
+; CHECK-NEXT:    [[TMP8:%.*]] = or i1 [[TMP7]], [[IDENT_CHECK]]
+; CHECK-NEXT:    br i1 [[TMP8]], label [[SCALAR_PH]], label [[VECTOR_PH:%.*]]
+; CHECK:       vector.ph:
+; CHECK-NEXT:    [[BROADCAST_SPLATINSERT:%.*]] = insertelement <4 x i32> poison, i32 [[ADD]], i64 0
+; CHECK-NEXT:    [[BROADCAST_SPLAT:%.*]] = shufflevector <4 x i32> [[BROADCAST_SPLATINSERT]], <4 x i32> poison, <4 x i32> zeroinitializer
+; CHECK-NEXT:    br label [[VECTOR_BODY:%.*]]
+; CHECK:       vector.body:
+; CHECK-NEXT:    [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
+; CHECK-NEXT:    [[VEC_IND:%.*]] = phi <4 x i32> [ <i32 0, i32 1, i32 2, i32 3>, [[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], [[VECTOR_BODY]] ]
+; CHECK-NEXT:    [[TMP9:%.*]] = add i64 [[INDEX]], 0
+; CHECK-NEXT:    [[TMP10:%.*]] = mul <4 x i32> [[VEC_IND]], [[BROADCAST_SPLAT]]
+; CHECK-NEXT:    [[TMP11:%.*]] = extractelement <4 x i32> [[TMP10]], i32 0
+; CHECK-NEXT:    [[TMP12:%.*]] = getelementptr i8, ptr [[DST_1]], i32 [[TMP11]]
+; CHECK-NEXT:    [[TMP13:%.*]] = extractelement <4 x i32> [[TMP10]], i32 1
+; CHECK-NEXT:    [[TMP14:%.*]] = getelementptr i8, ptr [[DST_1]], i32 [[TMP13]]
+; CHECK-NEXT:    [[TMP15:%.*]] = extractelement <4 x i32> [[TMP10]], i32 2
+; CHECK-NEXT:    [[TMP16:%.*]] = getelementptr i8, ptr [[DST_1]], i32 [[TMP15]]
+; CHECK-NEXT:    [[TMP17:%.*]] = extractelement <4 x i32> [[TMP10]], i32 3
+; CHECK-NEXT:    [[TMP18:%.*]] = getelementptr i8, ptr [[DST_1]], i32 [[TMP17]]
+; CHECK-NEXT:    store i32 0, ptr [[TMP12]], align 8
+; CHECK-NEXT:    store i32 0, ptr [[TMP14]], align 8
+; CHECK-NEXT:    store i32 0, ptr [[TMP16]], align 8
+; CHECK-NEXT:    store i32 0, ptr [[TMP18]], align 8
+; CHECK-NEXT:    [[TMP20:%.*]] = getelementptr i32, ptr [[DST_2]], i64 [[TMP9]]
+; CHECK-NEXT:    [[TMP21:%.*]] = getelementptr i32, ptr [[TMP20]], i32 0
+; CHECK-NEXT:    store <4 x i32> zeroinitializer, ptr [[TMP21]], align 8
+; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 4
+; CHECK-NEXT:    [[VEC_IND_NEXT]] = add <4 x i32> [[VEC_IND]], <i32 4, i32 4, i32 4, i32 4>
+; CHECK-NEXT:    [[TMP22:%.*]] = icmp eq i64 [[INDEX_NEXT]], 200
+; CHECK-NEXT:    br i1 [[TMP22]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP10:![0-9]+]]
+; CHECK:       middle.block:
+; CHECK-NEXT:    br i1 false, label [[EXIT:%.*]], label [[SCALAR_PH]]
+; CHECK:       scalar.ph:
+; CHECK-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i64 [ 200, [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ], [ 0, [[VECTOR_SCEVCHECK]] ]
+; CHECK-NEXT:    [[BC_RESUME_VAL1:%.*]] = phi i32 [ 200, [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY]] ], [ 0, [[VECTOR_SCEVCHECK]] ]
+; CHECK-NEXT:    br label [[LOOP:%.*]]
+; CHECK:       loop:
+; CHECK-NEXT:    [[IV_2:%.*]] = phi i64 [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ], [ [[IV_2_NEXT:%.*]], [[LOOP]] ]
+; CHECK-NEXT:    [[IV_3:%.*]] = phi i32 [ [[BC_RESUME_VAL1]], [[SCALAR_PH]] ], [ [[IV_3_NEXT:%.*]], [[LOOP]] ]
+; CHECK-NEXT:    [[IV_MUL:%.*]] = mul i32 [[IV_3]], [[ADD]]
+; CHECK-NEXT:    [[GEP_MUL:%.*]] = getelementptr i8, ptr [[DST_1]], i32 [[IV_MUL]]
+; CHECK-NEXT:    store i32 0, ptr [[GEP_MUL]], align 8
+; CHECK-NEXT:    [[IV_2_MUL:%.*]] = mul i64 [[IV_2]], [[OFFSET_EXT]]
+; CHECK-NEXT:    [[GEP:%.*]] = getelementptr i32, ptr [[DST_2]], i64 [[IV_2_MUL]]
+; CHECK-NEXT:    store i32 0, ptr [[GEP]], align 8
+; CHECK-NEXT:    [[IV_2_NEXT]] = add i64 [[IV_2]], 1
+; CHECK-NEXT:    [[IV_3_NEXT]] = add i32 [[IV_3]], 1
+; CHECK-NEXT:    [[EC:%.*]] = icmp eq i32 [[IV_3]], 200
+; CHECK-NEXT:    br i1 [[EC]], label [[EXIT]], label [[LOOP]], !llvm.loop [[LOOP11:![0-9]+]]
+; CHECK:       exit:
+; CHECK-NEXT:    ret void
+;
+entry:
+  %offset.ext = zext i32 %offset to i64
+  %add = add i32 %offset, 3
+  br label %loop
+
+loop:
+  %iv.2 = phi i64 [ 0, %entry ], [ %iv.2.next, %loop ]
+  %iv.3 = phi i32 [ 0, %entry ], [ %iv.3.next, %loop ]
+  %iv.mul  = mul i32 %iv.3, %add
+  %gep.mul = getelementptr i8, ptr %dst.1, i32 %iv.mul
+  store i32 0, ptr %gep.mul, align 8
+  %iv.2.mul  = mul i64 %iv.2, %offset.ext
+  %gep = getelementptr i32, ptr %dst.2, i64 %iv.2.mul
+  store i32 0, ptr %gep, align 8
+  %iv.2.next = add i64 %iv.2, 1
+  %iv.3.next = add i32 %iv.3, 1
+  %ec = icmp eq i32 %iv.3, 200
+  br i1 %ec, label %exit, label %loop
+
+exit:
+  ret void
+}
+;.
+; CHECK: [[LOOP0]] = distinct !{[[LOOP0]], [[META1:![0-9]+]], [[META2:![0-9]+]]}
+; CHECK: [[META1]] = !{!"llvm.loop.isvectorized", i32 1}
+; CHECK: [[META2]] = !{!"llvm.loop.unroll.runtime.disable"}
+; CHECK: [[LOOP3]] = distinct !{[[LOOP3]], [[META1]]}
+; CHECK: [[LOOP4]] = distinct !{[[LOOP4]], [[META1]], [[META2]]}
+; CHECK: [[LOOP5]] = distinct !{[[LOOP5]], [[META1]]}
+; CHECK: [[LOOP6]] = distinct !{[[LOOP6]], [[META1]], [[META2]]}
+; CHECK: [[LOOP7]] = distinct !{[[LOOP7]], [[META1]]}
+; CHECK: [[LOOP8]] = distinct !{[[LOOP8]], [[META1]], [[META2]]}
+; CHECK: [[LOOP9]] = distinct !{[[LOOP9]], [[META1]]}
+; CHECK: [[LOOP10]] = distinct !{[[LOOP10]], [[META1]], [[META2]]}
+; CHECK: [[LOOP11]] = distinct !{[[LOOP11]], [[META1]]}
+;.
diff --git a/llvm/test/Transforms/LoopVectorize/vplan-infer-not-or-type.ll b/llvm/test/Transforms/LoopVectorize/vplan-infer-not-or-type.ll
new file mode 100644
index 000000000000..102ef699cb37
--- /dev/null
+++ b/llvm/test/Transforms/LoopVectorize/vplan-infer-not-or-type.ll
@@ -0,0 +1,64 @@
+; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --version 4
+; RUN: opt < %s -passes=loop-vectorize -S | FileCheck %s
+
+; This test used to crash due to missing Or/Not cases in
+; inferScalarTypeForRecipe.
+
+define void @foo(i8 %arg.0, i8 %arg.1) {
+; CHECK-LABEL: define void @foo(
+; CHECK-SAME: i8 [[ARG_0:%.*]], i8 [[ARG_1:%.*]]) {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    br i1 false, label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]]
+; CHECK:       vector.ph:
+; CHECK-NEXT:    br label [[VECTOR_BODY:%.*]]
+; CHECK:       vector.body:
+; CHECK-NEXT:    [[INDEX:%.*]] = phi i32 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ]
+; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i32 [[INDEX]], 2
+; CHECK-NEXT:    br i1 true, label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP0:![0-9]+]]
+; CHECK:       middle.block:
+; CHECK-NEXT:    br i1 true, label [[LOOPEXIT:%.*]], label [[SCALAR_PH]]
+; CHECK:       scalar.ph:
+; CHECK-NEXT:    [[BC_RESUME_VAL:%.*]] = phi i8 [ 2, [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ]
+; CHECK-NEXT:    br label [[LOOP_HEADER:%.*]]
+; CHECK:       loop.header:
+; CHECK-NEXT:    [[INCREMENTOR:%.*]] = phi i8 [ [[ADD:%.*]], [[LATCH:%.*]] ], [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ]
+; CHECK-NEXT:    [[AND:%.*]] = and i8 [[ARG_0]], [[ARG_1]]
+; CHECK-NEXT:    [[EXTRACT_T:%.*]] = trunc i8 [[AND]] to i1
+; CHECK-NEXT:    br i1 [[EXTRACT_T]], label [[LATCH]], label [[INDIRECT_LATCH:%.*]]
+; CHECK:       indirect.latch:
+; CHECK-NEXT:    br label [[LATCH]]
+; CHECK:       latch:
+; CHECK-NEXT:    [[ADD]] = add i8 [[INCREMENTOR]], 1
+; CHECK-NEXT:    [[CONV:%.*]] = zext i8 [[INCREMENTOR]] to i32
+; CHECK-NEXT:    [[CMP:%.*]] = icmp ult i32 [[CONV]], 1
+; CHECK-NEXT:    br i1 [[CMP]], label [[LOOP_HEADER]], label [[LOOPEXIT]], !llvm.loop [[LOOP3:![0-9]+]]
+; CHECK:       loop.exit:
+; CHECK-NEXT:    ret void
+;
+entry:
+  br label %loop.header
+
+loop.header:                                         ; preds = %latch, %entry
+  %incrementor = phi i8 [ %add, %latch ], [ 0, %entry ]
+  %and = and i8 %arg.0, %arg.1
+  %extract.t = trunc i8 %and to i1
+  br i1 %extract.t, label %latch, label %indirect.latch
+
+indirect.latch:                                       ; preds = %loop.header
+  br label %latch
+
+latch:                               ; preds = %loop.header16, %loop.header
+  %add = add i8 %incrementor, 1
+  %conv = zext i8 %incrementor to i32
+  %cmp = icmp ult i32 %conv, 1
+  br i1 %cmp, label %loop.header, label %loop.exit
+
+loop.exit:
+  ret void
+}
+;.
+; CHECK: [[LOOP0]] = distinct !{[[LOOP0]], [[META1:![0-9]+]], [[META2:![0-9]+]]}
+; CHECK: [[META1]] = !{!"llvm.loop.isvectorized", i32 1}
+; CHECK: [[META2]] = !{!"llvm.loop.unroll.runtime.disable"}
+; CHECK: [[LOOP3]] = distinct !{[[LOOP3]], [[META2]], [[META1]]}
+;.
diff --git a/llvm/test/Transforms/LowerTypeTests/cfi-nounwind-direct-call.ll b/llvm/test/Transforms/LowerTypeTests/cfi-nounwind-direct-call.ll
index 4c88f4acc12f..2795333effd7 100644
--- a/llvm/test/Transforms/LowerTypeTests/cfi-nounwind-direct-call.ll
+++ b/llvm/test/Transforms/LowerTypeTests/cfi-nounwind-direct-call.ll
@@ -109,8 +109,8 @@ attributes #6 = { noreturn nounwind }
 !11 = !{}
 !12 = !{!"branch_weights", i32 1048575, i32 1}
 ; CHECK: Function Attrs: minsize mustprogress nofree norecurse nosync nounwind optsize willreturn memory(none)
-; CHECK-LABEL: define dso_local noundef i32 @_Z9nothrow_ei
-; CHECK-SAME: (i32 noundef [[NUM:%.*]]) #[[ATTR0:[0-9]+]] !type !4 !type !5 !type !6 {
+; CHECK-LABEL: define dso_local noundef range(i32 0, 2) i32 @_Z9nothrow_ei
+; CHECK-SAME: (i32 noundef [[NUM:%.*]]) #[[ATTR0:[0-9]+]] !type [[META4:![0-9]+]] !type [[META5:![0-9]+]] !type [[META6:![0-9]+]] {
 ; CHECK-NEXT:  entry:
 ; CHECK-NEXT:    [[TOBOOL_NOT:%.*]] = icmp ne i32 [[NUM]], 0
 ; CHECK-NEXT:    [[DOT:%.*]] = zext i1 [[TOBOOL_NOT]] to i32
@@ -118,8 +118,8 @@ attributes #6 = { noreturn nounwind }
 ;
 ;
 ; CHECK: Function Attrs: minsize mustprogress nofree norecurse nosync nounwind optsize willreturn memory(write, argmem: none, inaccessiblemem: none)
-; CHECK-LABEL: define dso_local noundef i32 @_Z10call_catchi
-; CHECK-SAME: (i32 noundef [[NUM:%.*]]) local_unnamed_addr #[[ATTR1:[0-9]+]] !type !4 !type !5 !type !6 {
+; CHECK-LABEL: define dso_local noundef range(i32 0, 2) i32 @_Z10call_catchi
+; CHECK-SAME: (i32 noundef [[NUM:%.*]]) local_unnamed_addr #[[ATTR1:[0-9]+]] !type [[META4]] !type [[META5]] !type [[META6]] {
 ; CHECK-NEXT:  entry:
 ; CHECK-NEXT:    store ptr @_Z9nothrow_ei.cfi_jt, ptr @catch_ptr, align 8, !tbaa [[TBAA7:![0-9]+]]
 ; CHECK-NEXT:    [[TOBOOL_NOT_I:%.*]] = icmp ne i32 [[NUM]], 0
@@ -131,17 +131,17 @@ attributes #6 = { noreturn nounwind }
 ; CHECK-LABEL: define weak_odr hidden void @__cfi_check_fail
 ; CHECK-SAME: (ptr noundef [[TMP0:%.*]], ptr noundef [[TMP1:%.*]]) #[[ATTR2:[0-9]+]] {
 ; CHECK-NEXT:  entry:
-; CHECK-NEXT:    [[DOTNOT:%.*]] = icmp eq ptr [[TMP0]], null, !nosanitize !11
-; CHECK-NEXT:    br i1 [[DOTNOT]], label [[TRAP:%.*]], label [[CONT:%.*]], !nosanitize !11
+; CHECK-NEXT:    [[DOTNOT:%.*]] = icmp eq ptr [[TMP0]], null, !nosanitize [[META11:![0-9]+]]
+; CHECK-NEXT:    br i1 [[DOTNOT]], label [[TRAP:%.*]], label [[CONT:%.*]], !nosanitize [[META11]]
 ; CHECK:       trap:
-; CHECK-NEXT:    tail call void @llvm.ubsantrap(i8 2) #[[ATTR5:[0-9]+]], !nosanitize !11
-; CHECK-NEXT:    unreachable, !nosanitize !11
+; CHECK-NEXT:    tail call void @llvm.ubsantrap(i8 2) #[[ATTR6:[0-9]+]], !nosanitize [[META11]]
+; CHECK-NEXT:    unreachable, !nosanitize [[META11]]
 ; CHECK:       cont:
-; CHECK-NEXT:    [[TMP2:%.*]] = load i8, ptr [[TMP0]], align 4, !nosanitize !11
+; CHECK-NEXT:    [[TMP2:%.*]] = load i8, ptr [[TMP0]], align 4, !nosanitize [[META11]]
 ; CHECK-NEXT:    [[SWITCH:%.*]] = icmp ult i8 [[TMP2]], 5
 ; CHECK-NEXT:    br i1 [[SWITCH]], label [[TRAP]], label [[CONT6:%.*]]
 ; CHECK:       cont6:
-; CHECK-NEXT:    ret void, !nosanitize !11
+; CHECK-NEXT:    ret void, !nosanitize [[META11]]
 ;
 ;
 ; CHECK-LABEL: define weak void @__cfi_check
@@ -153,8 +153,8 @@ attributes #6 = { noreturn nounwind }
 ;
 ; CHECK: Function Attrs: naked nocf_check noinline nounwind
 ; CHECK-LABEL: define internal void @_Z9nothrow_ei.cfi_jt
-; CHECK-SAME: () #[[ATTR4:[0-9]+]] align 8 {
+; CHECK-SAME: () #[[ATTR5:[0-9]+]] align 8 {
 ; CHECK-NEXT:  entry:
-; CHECK-NEXT:    tail call void asm sideeffect "jmp ${0:c}@plt\0Aint3\0Aint3\0Aint3\0A", "s"(ptr nonnull @_Z9nothrow_ei) #[[ATTR6:[0-9]+]]
+; CHECK-NEXT:    tail call void asm sideeffect "jmp ${0:c}@plt\0Aint3\0Aint3\0Aint3\0A", "s"(ptr nonnull @_Z9nothrow_ei) #[[ATTR7:[0-9]+]]
 ; CHECK-NEXT:    unreachable
 ;
diff --git a/llvm/test/Transforms/OpenMP/add_attributes.ll b/llvm/test/Transforms/OpenMP/add_attributes.ll
index 47ff5cad4e7e..ebcca3067f04 100644
--- a/llvm/test/Transforms/OpenMP/add_attributes.ll
+++ b/llvm/test/Transforms/OpenMP/add_attributes.ll
@@ -641,8 +641,6 @@ declare i32 @__tgt_target_teams_mapper(ptr, i64, ptr, i32, ptr, ptr, ptr, ptr, p
 
 declare i32 @__tgt_target_teams_nowait_mapper(ptr, i64, ptr, i32, ptr, ptr, ptr, ptr, ptr, ptr, i32, i32, i32, ptr, i32, ptr)
 
-declare void @__tgt_register_requires(i64)
-
 declare void @__tgt_target_data_begin_mapper(ptr, i64, i32, ptr, ptr, ptr, ptr, ptr, ptr)
 
 declare void @__tgt_target_data_begin_nowait_mapper(ptr, i64, i32, ptr, ptr, ptr, ptr, ptr, ptr)
@@ -1249,9 +1247,6 @@ declare i32 @__tgt_target_kernel_nowait(ptr, i64, i32, i32, ptr, ptr, i32, ptr,
 ; CHECK-NEXT: declare i32 @__tgt_target_teams_nowait_mapper(ptr, i64, ptr, i32, ptr, ptr, ptr, ptr, ptr, ptr, i32, i32, i32, ptr, i32, ptr)
 
 ; CHECK: ; Function Attrs: nounwind
-; CHECK-NEXT: declare void @__tgt_register_requires(i64)
-
-; CHECK: ; Function Attrs: nounwind
 ; CHECK-NEXT: declare void @__tgt_target_data_begin_mapper(ptr, i64, i32, ptr, ptr, ptr, ptr, ptr, ptr)
 
 ; CHECK: ; Function Attrs: nounwind
@@ -1894,9 +1889,6 @@ declare i32 @__tgt_target_kernel_nowait(ptr, i64, i32, i32, ptr, ptr, i32, ptr,
 ; OPTIMISTIC-NEXT: declare i32 @__tgt_target_teams_nowait_mapper(ptr, i64, ptr, i32, ptr, ptr, ptr, ptr, ptr, ptr, i32, i32, i32, ptr, i32, ptr)
 
 ; OPTIMISTIC: ; Function Attrs: nounwind
-; OPTIMISTIC-NEXT: declare void @__tgt_register_requires(i64)
-
-; OPTIMISTIC: ; Function Attrs: nounwind
 ; OPTIMISTIC-NEXT: declare void @__tgt_target_data_begin_mapper(ptr, i64, i32, ptr, ptr, ptr, ptr, ptr, ptr)
 
 ; OPTIMISTIC: ; Function Attrs: nounwind
@@ -2552,9 +2544,6 @@ declare i32 @__tgt_target_kernel_nowait(ptr, i64, i32, i32, ptr, ptr, i32, ptr,
 ; EXT-NEXT: declare signext i32 @__tgt_target_teams_nowait_mapper(ptr, i64, ptr, i32 signext, ptr, ptr, ptr, ptr, ptr, ptr, i32 signext, i32 signext, i32 signext, ptr, i32 signext, ptr)
 
 ; EXT: ; Function Attrs: nounwind
-; EXT-NEXT: declare void @__tgt_register_requires(i64)
-
-; EXT: ; Function Attrs: nounwind
 ; EXT-NEXT: declare void @__tgt_target_data_begin_mapper(ptr, i64, i32 signext, ptr, ptr, ptr, ptr, ptr, ptr)
 
 ; EXT: ; Function Attrs: nounwind
diff --git a/llvm/test/Transforms/PhaseOrdering/AArch64/extra-unroll-simplifications.ll b/llvm/test/Transforms/PhaseOrdering/AArch64/extra-unroll-simplifications.ll
index 6132c35c96ca..b32f4e2a258c 100644
--- a/llvm/test/Transforms/PhaseOrdering/AArch64/extra-unroll-simplifications.ll
+++ b/llvm/test/Transforms/PhaseOrdering/AArch64/extra-unroll-simplifications.ll
@@ -72,6 +72,86 @@ exit:
   ret void
 }
 
+define void @cse_matching_load_from_previous_unrolled_iteration(i32 %N, ptr %src, ptr noalias %dst) {
+; CHECK-LABEL: define void @cse_matching_load_from_previous_unrolled_iteration(
+; CHECK-SAME: i32 [[N:%.*]], ptr nocapture readonly [[SRC:%.*]], ptr noalias nocapture writeonly [[DST:%.*]]) local_unnamed_addr #[[ATTR0]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[SRC_4:%.*]] = getelementptr i8, ptr [[SRC]], i64 4
+; CHECK-NEXT:    [[SRC_12:%.*]] = getelementptr i8, ptr [[SRC]], i64 12
+; CHECK-NEXT:    [[CMP141:%.*]] = icmp sgt i32 [[N]], 0
+; CHECK-NEXT:    br i1 [[CMP141]], label [[LOOP_LATCH_PREHEADER:%.*]], label [[EXIT:%.*]]
+; CHECK:       loop.latch.preheader:
+; CHECK-NEXT:    [[WIDE_TRIP_COUNT:%.*]] = zext nneg i32 [[N]] to i64
+; CHECK-NEXT:    [[XTRAITER:%.*]] = and i64 [[WIDE_TRIP_COUNT]], 1
+; CHECK-NEXT:    [[TMP0:%.*]] = icmp eq i32 [[N]], 1
+; CHECK-NEXT:    br i1 [[TMP0]], label [[EXIT_LOOPEXIT_UNR_LCSSA:%.*]], label [[LOOP_LATCH_PREHEADER_NEW:%.*]]
+; CHECK:       loop.latch.preheader.new:
+; CHECK-NEXT:    [[UNROLL_ITER:%.*]] = and i64 [[WIDE_TRIP_COUNT]], 2147483646
+; CHECK-NEXT:    br label [[LOOP_LATCH:%.*]]
+; CHECK:       loop.latch:
+; CHECK-NEXT:    [[INDVARS_IV:%.*]] = phi i64 [ 0, [[LOOP_LATCH_PREHEADER_NEW]] ], [ [[INDVARS_IV_NEXT_1:%.*]], [[LOOP_LATCH]] ]
+; CHECK-NEXT:    [[NITER:%.*]] = phi i64 [ 0, [[LOOP_LATCH_PREHEADER_NEW]] ], [ [[NITER_NEXT_1:%.*]], [[LOOP_LATCH]] ]
+; CHECK-NEXT:    [[GEP_SRC_12:%.*]] = getelementptr <2 x i32>, ptr [[SRC_12]], i64 [[INDVARS_IV]]
+; CHECK-NEXT:    [[L_12:%.*]] = load <2 x i32>, ptr [[GEP_SRC_12]], align 8
+; CHECK-NEXT:    [[GEP_SRC_4:%.*]] = getelementptr <2 x i32>, ptr [[SRC_4]], i64 [[INDVARS_IV]]
+; CHECK-NEXT:    [[L_4:%.*]] = load <2 x i32>, ptr [[GEP_SRC_4]], align 8
+; CHECK-NEXT:    [[MUL:%.*]] = mul <2 x i32> [[L_4]], [[L_12]]
+; CHECK-NEXT:    [[GEP_DST:%.*]] = getelementptr <2 x i32>, ptr [[DST]], i64 [[INDVARS_IV]]
+; CHECK-NEXT:    store <2 x i32> [[MUL]], ptr [[GEP_DST]], align 8
+; CHECK-NEXT:    [[INDVARS_IV_NEXT:%.*]] = or disjoint i64 [[INDVARS_IV]], 1
+; CHECK-NEXT:    [[GEP_SRC_12_1:%.*]] = getelementptr <2 x i32>, ptr [[SRC_12]], i64 [[INDVARS_IV_NEXT]]
+; CHECK-NEXT:    [[L_12_1:%.*]] = load <2 x i32>, ptr [[GEP_SRC_12_1]], align 8
+; CHECK-NEXT:    [[GEP_SRC_4_1:%.*]] = getelementptr <2 x i32>, ptr [[SRC_4]], i64 [[INDVARS_IV_NEXT]]
+; CHECK-NEXT:    [[L_4_1:%.*]] = load <2 x i32>, ptr [[GEP_SRC_4_1]], align 8
+; CHECK-NEXT:    [[MUL_1:%.*]] = mul <2 x i32> [[L_4_1]], [[L_12_1]]
+; CHECK-NEXT:    [[GEP_DST_1:%.*]] = getelementptr <2 x i32>, ptr [[DST]], i64 [[INDVARS_IV_NEXT]]
+; CHECK-NEXT:    store <2 x i32> [[MUL_1]], ptr [[GEP_DST_1]], align 8
+; CHECK-NEXT:    [[INDVARS_IV_NEXT_1]] = add nuw nsw i64 [[INDVARS_IV]], 2
+; CHECK-NEXT:    [[NITER_NEXT_1]] = add i64 [[NITER]], 2
+; CHECK-NEXT:    [[NITER_NCMP_1:%.*]] = icmp eq i64 [[NITER_NEXT_1]], [[UNROLL_ITER]]
+; CHECK-NEXT:    br i1 [[NITER_NCMP_1]], label [[EXIT_LOOPEXIT_UNR_LCSSA]], label [[LOOP_LATCH]], !llvm.loop [[LOOP3:![0-9]+]]
+; CHECK:       exit.loopexit.unr-lcssa:
+; CHECK-NEXT:    [[INDVARS_IV_UNR:%.*]] = phi i64 [ 0, [[LOOP_LATCH_PREHEADER]] ], [ [[INDVARS_IV_NEXT_1]], [[LOOP_LATCH]] ]
+; CHECK-NEXT:    [[LCMP_MOD_NOT:%.*]] = icmp eq i64 [[XTRAITER]], 0
+; CHECK-NEXT:    br i1 [[LCMP_MOD_NOT]], label [[EXIT]], label [[LOOP_LATCH_EPIL:%.*]]
+; CHECK:       loop.latch.epil:
+; CHECK-NEXT:    [[GEP_SRC_12_EPIL:%.*]] = getelementptr <2 x i32>, ptr [[SRC_12]], i64 [[INDVARS_IV_UNR]]
+; CHECK-NEXT:    [[L_12_EPIL:%.*]] = load <2 x i32>, ptr [[GEP_SRC_12_EPIL]], align 8
+; CHECK-NEXT:    [[GEP_SRC_4_EPIL:%.*]] = getelementptr <2 x i32>, ptr [[SRC_4]], i64 [[INDVARS_IV_UNR]]
+; CHECK-NEXT:    [[L_4_EPIL:%.*]] = load <2 x i32>, ptr [[GEP_SRC_4_EPIL]], align 8
+; CHECK-NEXT:    [[MUL_EPIL:%.*]] = mul <2 x i32> [[L_4_EPIL]], [[L_12_EPIL]]
+; CHECK-NEXT:    [[GEP_DST_EPIL:%.*]] = getelementptr <2 x i32>, ptr [[DST]], i64 [[INDVARS_IV_UNR]]
+; CHECK-NEXT:    store <2 x i32> [[MUL_EPIL]], ptr [[GEP_DST_EPIL]], align 8
+; CHECK-NEXT:    br label [[EXIT]]
+; CHECK:       exit:
+; CHECK-NEXT:    ret void
+;
+entry:
+  %src.4 = getelementptr i8, ptr %src, i64 4
+  %src.12 = getelementptr i8, ptr %src, i64 12
+  br label %loop.header
+
+loop.header:
+  %iv = phi i32 [ 0, %entry ], [ %iv.next, %loop.latch ]
+  %cmp14 = icmp slt i32 %iv, %N
+  br i1 %cmp14, label %loop.latch, label %exit
+
+loop.latch:
+  %iv.ext = zext i32 %iv to i64
+  %gep.src.12 = getelementptr <2 x i32>, ptr %src.12, i64 %iv.ext
+  %l.12 = load <2 x i32>, ptr %gep.src.12, align 8
+  %gep.src.4 = getelementptr <2 x i32>, ptr %src.4, i64 %iv.ext
+  %l.4 = load <2 x i32>, ptr %gep.src.4, align 8
+  %mul = mul <2 x i32> %l.12, %l.4
+  %gep.dst = getelementptr <2 x i32>, ptr %dst, i64 %iv.ext
+  store <2 x i32> %mul, ptr %gep.dst
+  %iv.next = add nuw nsw i32 %iv, 1
+  br label %loop.header, !llvm.loop !0
+
+exit:
+  ret void
+}
+
 !0 = distinct !{!0, !1, !2}
 !1 = !{!"llvm.loop.mustprogress"}
 !2 = !{!"llvm.loop.unroll.count", i32 2}
@@ -79,4 +159,5 @@ exit:
 ; CHECK: [[LOOP0]] = distinct !{[[LOOP0]], [[META1:![0-9]+]], [[META2:![0-9]+]]}
 ; CHECK: [[META1]] = !{!"llvm.loop.mustprogress"}
 ; CHECK: [[META2]] = !{!"llvm.loop.unroll.disable"}
+; CHECK: [[LOOP3]] = distinct !{[[LOOP3]], [[META1]], [[META2]]}
 ;.
diff --git a/llvm/test/Transforms/PhaseOrdering/AArch64/quant_4x4.ll b/llvm/test/Transforms/PhaseOrdering/AArch64/quant_4x4.ll
index 9206893cb234..c133852f6693 100644
--- a/llvm/test/Transforms/PhaseOrdering/AArch64/quant_4x4.ll
+++ b/llvm/test/Transforms/PhaseOrdering/AArch64/quant_4x4.ll
@@ -7,7 +7,7 @@ target triple = "aarch64"
 ; Check that the function gets vectorized.
 
 define i32 @quant_4x4(ptr noundef %dct, ptr noundef %mf, ptr noundef %bias) {
-; CHECK-LABEL: define i32 @quant_4x4
+; CHECK-LABEL: define range(i32 0, 2) i32 @quant_4x4
 ; CHECK-SAME: (ptr nocapture noundef [[DCT:%.*]], ptr nocapture noundef readonly [[MF:%.*]], ptr nocapture noundef readonly [[BIAS:%.*]]) local_unnamed_addr #[[ATTR0:[0-9]+]] {
 ; CHECK-NEXT:  entry:
 ; CHECK-NEXT:    [[SCEVGEP:%.*]] = getelementptr i8, ptr [[DCT]], i64 32
diff --git a/llvm/test/Transforms/PhaseOrdering/X86/loop-idiom-vs-indvars.ll b/llvm/test/Transforms/PhaseOrdering/X86/loop-idiom-vs-indvars.ll
index 4b5f3107466c..7382c3c80d42 100644
--- a/llvm/test/Transforms/PhaseOrdering/X86/loop-idiom-vs-indvars.ll
+++ b/llvm/test/Transforms/PhaseOrdering/X86/loop-idiom-vs-indvars.ll
@@ -12,7 +12,7 @@ define i32 @cttz(i32 %n, ptr %p1) {
 ; ALL-LABEL: @cttz(
 ; ALL-NEXT:  entry:
 ; ALL-NEXT:    [[TMP0:%.*]] = shl i32 [[N:%.*]], 1
-; ALL-NEXT:    [[TMP1:%.*]] = tail call i32 @llvm.cttz.i32(i32 [[TMP0]], i1 false), !range [[RNG0:![0-9]+]]
+; ALL-NEXT:    [[TMP1:%.*]] = tail call range(i32 1, 33) i32 @llvm.cttz.i32(i32 [[TMP0]], i1 false)
 ; ALL-NEXT:    [[TMP2:%.*]] = sub nuw nsw i32 32, [[TMP1]]
 ; ALL-NEXT:    [[TMP3:%.*]] = sub nuw nsw i32 75, [[TMP1]]
 ; ALL-NEXT:    store i32 [[TMP3]], ptr [[P1:%.*]], align 4
diff --git a/llvm/test/Transforms/PhaseOrdering/icmp-ashr-breaking-select-idiom.ll b/llvm/test/Transforms/PhaseOrdering/icmp-ashr-breaking-select-idiom.ll
index 67d721b23d6f..35d5ceeb9195 100644
--- a/llvm/test/Transforms/PhaseOrdering/icmp-ashr-breaking-select-idiom.ll
+++ b/llvm/test/Transforms/PhaseOrdering/icmp-ashr-breaking-select-idiom.ll
@@ -2,7 +2,7 @@
 ; RUN: opt -O1 -S < %s  | FileCheck %s
 
 define i32 @testa(i32 %mul) {
-; CHECK-LABEL: define i32 @testa(
+; CHECK-LABEL: define range(i32 -65536, 65536) i32 @testa(
 ; CHECK-SAME: i32 [[MUL:%.*]]) local_unnamed_addr #[[ATTR0:[0-9]+]] {
 ; CHECK-NEXT:    [[SHR:%.*]] = ashr i32 [[MUL]], 15
 ; CHECK-NEXT:    [[SPEC_SELECT_I:%.*]] = tail call i32 @llvm.smin.i32(i32 [[SHR]], i32 32767)
@@ -16,7 +16,7 @@ define i32 @testa(i32 %mul) {
 }
 
 define i32 @testb(i32 %mul) {
-; CHECK-LABEL: define i32 @testb(
+; CHECK-LABEL: define range(i32 -16777216, 16777216) i32 @testb(
 ; CHECK-SAME: i32 [[MUL:%.*]]) local_unnamed_addr #[[ATTR0]] {
 ; CHECK-NEXT:    [[SHR102:%.*]] = ashr i32 [[MUL]], 7
 ; CHECK-NEXT:    [[TMP1:%.*]] = tail call i32 @llvm.smax.i32(i32 [[SHR102]], i32 -128)
diff --git a/llvm/test/Transforms/PhaseOrdering/lower-table-based-cttz.ll b/llvm/test/Transforms/PhaseOrdering/lower-table-based-cttz.ll
index 284873a97376..19fbc1f1ae64 100644
--- a/llvm/test/Transforms/PhaseOrdering/lower-table-based-cttz.ll
+++ b/llvm/test/Transforms/PhaseOrdering/lower-table-based-cttz.ll
@@ -15,7 +15,7 @@
 
 ; RUN: opt -O3 -S < %s | FileCheck %s
 
-; CHECK: call i32 @llvm.cttz.i32
+; CHECK: call range(i32 0, 33) i32 @llvm.cttz.i32
 
 @ctz1.table = internal constant [32 x i8] c"\00\01\1C\02\1D\0E\18\03\1E\16\14\0F\19\11\04\08\1F\1B\0D\17\15\13\10\07\1A\0C\12\06\0B\05\0A\09", align 16
 
diff --git a/llvm/test/Transforms/PhaseOrdering/min_max_loop.ll b/llvm/test/Transforms/PhaseOrdering/min_max_loop.ll
index fb338a6507eb..63cfef6f3d09 100644
--- a/llvm/test/Transforms/PhaseOrdering/min_max_loop.ll
+++ b/llvm/test/Transforms/PhaseOrdering/min_max_loop.ll
@@ -19,7 +19,7 @@
 ;; }
 
 define i16 @vecreduce_smin_v2i16(i32 %n, ptr %v) {
-; CHECK-LABEL: define i16 @vecreduce_smin_v2i16(
+; CHECK-LABEL: define range(i16 -32768, 1) i16 @vecreduce_smin_v2i16(
 ; CHECK:    @llvm.smin.v2i16
 
 entry:
@@ -65,7 +65,7 @@ for.end:                                          ; preds = %for.cond
 }
 
 define i16 @vecreduce_smax_v2i16(i32 %n, ptr %v) {
-; CHECK-LABEL: define i16 @vecreduce_smax_v2i16(
+; CHECK-LABEL: define range(i16 0, -32768) i16 @vecreduce_smax_v2i16(
 ; CHECK:  @llvm.smax.v2i16
 
 entry:
diff --git a/llvm/test/Transforms/SCCP/and-add-shl.ll b/llvm/test/Transforms/SCCP/and-add-shl.ll
index 7c037ffa6bf6..7af563f13a18 100644
--- a/llvm/test/Transforms/SCCP/and-add-shl.ll
+++ b/llvm/test/Transforms/SCCP/and-add-shl.ll
@@ -59,7 +59,7 @@ define i8 @and_not_shl_1(i8 %x) {
 
 ; Negative test: https://alive2.llvm.org/ce/z/Zv4Pyu
 define i8 @and_add_shl_overlap(i8 %x) {
-; CHECK-LABEL: define i8 @and_add_shl_overlap
+; CHECK-LABEL: define range(i8 0, 33) i8 @and_add_shl_overlap
 ; CHECK-SAME: (i8 [[X:%.*]]) {
 ; CHECK-NEXT:    [[OP1_P2:%.*]] = icmp ule i8 [[X]], 6
 ; CHECK-NEXT:    call void @llvm.assume(i1 [[OP1_P2]])
@@ -77,7 +77,7 @@ define i8 @and_add_shl_overlap(i8 %x) {
 }
 
 define i8 @and_not_shl_overlap(i8 %x) {
-; CHECK-LABEL: define i8 @and_not_shl_overlap
+; CHECK-LABEL: define range(i8 0, 5) i8 @and_not_shl_overlap
 ; CHECK-SAME: (i8 [[X:%.*]]) {
 ; CHECK-NEXT:    [[OP1_P2:%.*]] = icmp ule i8 [[X]], 3
 ; CHECK-NEXT:    call void @llvm.assume(i1 [[OP1_P2]])
diff --git a/llvm/test/Transforms/SCCP/ip-add-range-to-call.ll b/llvm/test/Transforms/SCCP/ip-add-range-to-call.ll
index 64c1b9020a05..c24c554102dd 100644
--- a/llvm/test/Transforms/SCCP/ip-add-range-to-call.ll
+++ b/llvm/test/Transforms/SCCP/ip-add-range-to-call.ll
@@ -1,20 +1,21 @@
-; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
+; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --version 4
 ; RUN: opt -passes=ipsccp -S %s | FileCheck %s
 
 ; Test 1.
 ; Both arguments and return value of @callee can be tracked. The inferred range
 ; can be added to call sites.
 define internal i32 @callee(i32 %x) {
-; CHECK-LABEL: @callee(
-; CHECK-NEXT:    ret i32 [[X:%.*]]
+; CHECK-LABEL: define internal range(i32 0, 21) i32 @callee(
+; CHECK-SAME: i32 [[X:%.*]]) {
+; CHECK-NEXT:    ret i32 [[X]]
 ;
   ret i32 %x
 }
 
 define i32 @caller1() {
-; CHECK-LABEL: @caller1(
-; CHECK-NEXT:    [[C1:%.*]] = call i32 @callee(i32 10), !range [[RNG0:![0-9]+]]
-; CHECK-NEXT:    [[C2:%.*]] = call i32 @callee(i32 20), !range [[RNG0]]
+; CHECK-LABEL: define range(i32 0, 41) i32 @caller1() {
+; CHECK-NEXT:    [[C1:%.*]] = call i32 @callee(i32 10)
+; CHECK-NEXT:    [[C2:%.*]] = call i32 @callee(i32 20)
 ; CHECK-NEXT:    [[A:%.*]] = add nuw nsw i32 [[C1]], [[C2]]
 ; CHECK-NEXT:    ret i32 [[A]]
 ;
@@ -25,9 +26,10 @@ define i32 @caller1() {
 }
 
 define i32 @caller2(i32 %x) {
-; CHECK-LABEL: @caller2(
-; CHECK-NEXT:    [[X_15:%.*]] = and i32 [[X:%.*]], 15
-; CHECK-NEXT:    [[C:%.*]] = call i32 @callee(i32 [[X_15]]), !range [[RNG0]]
+; CHECK-LABEL: define range(i32 0, 21) i32 @caller2(
+; CHECK-SAME: i32 [[X:%.*]]) {
+; CHECK-NEXT:    [[X_15:%.*]] = and i32 [[X]], 15
+; CHECK-NEXT:    [[C:%.*]] = call i32 @callee(i32 [[X_15]])
 ; CHECK-NEXT:    ret i32 [[C]]
 ;
   %x.15 = and i32 %x, 15
@@ -43,14 +45,15 @@ define i32 @caller2(i32 %x) {
 declare void @use_cb1(ptr)
 
 define internal i32 @callee2(i32 %x) {
-; CHECK-LABEL: @callee2(
-; CHECK-NEXT:    ret i32 [[X:%.*]]
+; CHECK-LABEL: define internal i32 @callee2(
+; CHECK-SAME: i32 [[X:%.*]]) {
+; CHECK-NEXT:    ret i32 [[X]]
 ;
   ret i32 %x
 }
 
 define void @caller_cb1() {
-; CHECK-LABEL: @caller_cb1(
+; CHECK-LABEL: define void @caller_cb1() {
 ; CHECK-NEXT:    [[C1:%.*]] = call i32 @callee2(i32 9)
 ; CHECK-NEXT:    [[C2:%.*]] = call i32 @callee2(i32 10)
 ; CHECK-NEXT:    call void @use_cb1(ptr @callee2)
@@ -70,8 +73,9 @@ define void @caller_cb1() {
 declare void @use_cb2(ptr)
 
 define internal i32 @callee3(i32 %x) {
-; CHECK-LABEL: @callee3(
-; CHECK-NEXT:    [[C:%.*]] = icmp eq i32 [[X:%.*]], 10
+; CHECK-LABEL: define internal range(i32 500, 601) i32 @callee3(
+; CHECK-SAME: i32 [[X:%.*]]) {
+; CHECK-NEXT:    [[C:%.*]] = icmp eq i32 [[X]], 10
 ; CHECK-NEXT:    [[S:%.*]] = select i1 [[C]], i32 500, i32 600
 ; CHECK-NEXT:    ret i32 [[S]]
 ;
@@ -81,9 +85,9 @@ define internal i32 @callee3(i32 %x) {
 }
 
 define void @caller_cb2() {
-; CHECK-LABEL: @caller_cb2(
-; CHECK-NEXT:    [[C1:%.*]] = call i32 @callee3(i32 9), !range [[RNG1:![0-9]+]]
-; CHECK-NEXT:    [[C2:%.*]] = call i32 @callee3(i32 10), !range [[RNG1]]
+; CHECK-LABEL: define void @caller_cb2() {
+; CHECK-NEXT:    [[C1:%.*]] = call i32 @callee3(i32 9)
+; CHECK-NEXT:    [[C2:%.*]] = call i32 @callee3(i32 10)
 ; CHECK-NEXT:    call void @use_cb2(ptr @callee3)
 ; CHECK-NEXT:    ret void
 ;
@@ -100,9 +104,10 @@ define void @caller_cb2() {
 declare void @use_cb3(ptr)
 
 define internal i32 @callee4(i32 %x, i32 %y) {
-; CHECK-LABEL: @callee4(
-; CHECK-NEXT:    [[C:%.*]] = icmp eq i32 [[X:%.*]], 10
-; CHECK-NEXT:    [[S:%.*]] = select i1 [[C]], i32 500, i32 [[Y:%.*]]
+; CHECK-LABEL: define internal i32 @callee4(
+; CHECK-SAME: i32 [[X:%.*]], i32 [[Y:%.*]]) {
+; CHECK-NEXT:    [[C:%.*]] = icmp eq i32 [[X]], 10
+; CHECK-NEXT:    [[S:%.*]] = select i1 [[C]], i32 500, i32 [[Y]]
 ; CHECK-NEXT:    ret i32 [[S]]
 ;
   %c = icmp eq i32 %x, 10
@@ -111,11 +116,9 @@ define internal i32 @callee4(i32 %x, i32 %y) {
 }
 
 define void @caller_cb3() {
-; CHECK-LABEL: @caller_cb3(
+; CHECK-LABEL: define void @caller_cb3() {
 ; CHECK-NEXT:    [[C1:%.*]] = call i32 @callee4(i32 11, i32 30)
-; CHECK-NOT:   !range
 ; CHECK-NEXT:    [[C2:%.*]] = call i32 @callee4(i32 12, i32 40)
-; CHECK-NOT:   !range
 ; CHECK-NEXT:    call void @use_cb3(ptr @callee4)
 ; CHECK-NEXT:    ret void
 ;
@@ -129,15 +132,16 @@ define void @caller_cb3() {
 ; Range for the return value of callee5 includes undef. No range metadata
 ; should be added at call sites.
 define internal i32 @callee5(i32 %x, i32 %y) {
-; CHECK-LABEL: @callee5(
-; CHECK-NEXT:    [[C:%.*]] = icmp slt i32 [[X:%.*]], 15
+; CHECK-LABEL: define internal i32 @callee5(
+; CHECK-SAME: i32 [[X:%.*]], i32 [[Y:%.*]]) {
+; CHECK-NEXT:    [[C:%.*]] = icmp slt i32 [[X]], 15
 ; CHECK-NEXT:    br i1 [[C]], label [[BB1:%.*]], label [[BB2:%.*]]
 ; CHECK:       bb1:
 ; CHECK-NEXT:    br label [[EXIT:%.*]]
 ; CHECK:       bb2:
 ; CHECK-NEXT:    br label [[EXIT]]
 ; CHECK:       exit:
-; CHECK-NEXT:    [[RES:%.*]] = phi i32 [ [[Y:%.*]], [[BB1]] ], [ undef, [[BB2]] ]
+; CHECK-NEXT:    [[RES:%.*]] = phi i32 [ [[Y]], [[BB1]] ], [ undef, [[BB2]] ]
 ; CHECK-NEXT:    ret i32 [[RES]]
 ;
   %c = icmp slt i32 %x, 15
@@ -155,11 +159,9 @@ exit:
 }
 
 define i32 @caller5() {
-; CHECK-LABEL: @caller5(
+; CHECK-LABEL: define range(i32 200, 401) i32 @caller5() {
 ; CHECK-NEXT:    [[C1:%.*]] = call i32 @callee5(i32 10, i32 100)
-; CHECK-NOT:   !range
 ; CHECK-NEXT:    [[C2:%.*]] = call i32 @callee5(i32 20, i32 200)
-; CHECK-NOT:   !range
 ; CHECK-NEXT:    [[A:%.*]] = add i32 [[C1]], [[C2]]
 ; CHECK-NEXT:    ret i32 [[A]]
 ;
@@ -170,8 +172,9 @@ define i32 @caller5() {
 }
 
 define internal <2 x i64> @ctlz(<2 x i64> %arg) {
-; CHECK-LABEL: @ctlz(
-; CHECK-NEXT:    [[RES:%.*]] = call <2 x i64> @llvm.ctlz.v2i64(<2 x i64> [[ARG:%.*]], i1 false)
+; CHECK-LABEL: define internal range(i64 0, 65) <2 x i64> @ctlz(
+; CHECK-SAME: <2 x i64> [[ARG:%.*]]) {
+; CHECK-NEXT:    [[RES:%.*]] = call <2 x i64> @llvm.ctlz.v2i64(<2 x i64> [[ARG]], i1 false)
 ; CHECK-NEXT:    ret <2 x i64> [[RES]]
 ;
   %res = call <2 x i64> @llvm.ctlz.v2i64(<2 x i64> %arg, i1 false)
@@ -179,8 +182,9 @@ define internal <2 x i64> @ctlz(<2 x i64> %arg) {
 }
 
 define <2 x i64> @ctlz_caller(<2 x i64> %arg) {
-; CHECK-LABEL: @ctlz_caller(
-; CHECK-NEXT:    [[RES:%.*]] = call <2 x i64> @ctlz(<2 x i64> [[ARG:%.*]]), !range [[RNG2:![0-9]+]]
+; CHECK-LABEL: define range(i64 0, 65) <2 x i64> @ctlz_caller(
+; CHECK-SAME: <2 x i64> [[ARG:%.*]]) {
+; CHECK-NEXT:    [[RES:%.*]] = call <2 x i64> @ctlz(<2 x i64> [[ARG]])
 ; CHECK-NEXT:    ret <2 x i64> [[RES]]
 ;
   %res = call <2 x i64> @ctlz(<2 x i64> %arg)
@@ -189,6 +193,3 @@ define <2 x i64> @ctlz_caller(<2 x i64> %arg) {
 
 declare <2 x i64> @llvm.ctlz.v2i64(<2 x i64>, i1)
 
-; CHECK: [[RNG0]] = !{i32 0, i32 21}
-; CHECK: [[RNG1]] = !{i32 500, i32 601}
-; CHECK: [[RNG2]] = !{i64 0, i64 65}
diff --git a/llvm/test/Transforms/SCCP/ip-ranges-casts.ll b/llvm/test/Transforms/SCCP/ip-ranges-casts.ll
index 80d90922c2fb..05fa04a9fbe0 100644
--- a/llvm/test/Transforms/SCCP/ip-ranges-casts.ll
+++ b/llvm/test/Transforms/SCCP/ip-ranges-casts.ll
@@ -1,10 +1,11 @@
-; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
+; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --version 4
 ; RUN: opt < %s -passes=ipsccp -S | FileCheck %s
 
 ; x = [100, 301)
 define internal i1 @f.trunc(i32 %x) {
-; CHECK-LABEL: @f.trunc(
-; CHECK-NEXT:    [[T_1:%.*]] = trunc nuw nsw i32 [[X:%.*]] to i16
+; CHECK-LABEL: define internal i1 @f.trunc(
+; CHECK-SAME: i32 [[X:%.*]]) {
+; CHECK-NEXT:    [[T_1:%.*]] = trunc nuw nsw i32 [[X]] to i16
 ; CHECK-NEXT:    [[C_2:%.*]] = icmp sgt i16 [[T_1]], 299
 ; CHECK-NEXT:    [[C_4:%.*]] = icmp slt i16 [[T_1]], 101
 ; CHECK-NEXT:    [[RES_1:%.*]] = add nuw nsw i1 false, [[C_2]]
@@ -43,7 +44,7 @@ define internal i1 @f.trunc(i32 %x) {
 }
 
 define i1 @caller1() {
-; CHECK-LABEL: @caller1(
+; CHECK-LABEL: define i1 @caller1() {
 ; CHECK-NEXT:    [[CALL_1:%.*]] = tail call i1 @f.trunc(i32 100)
 ; CHECK-NEXT:    [[CALL_2:%.*]] = tail call i1 @f.trunc(i32 300)
 ; CHECK-NEXT:    [[RES:%.*]] = and i1 [[CALL_1]], [[CALL_2]]
@@ -58,14 +59,15 @@ define i1 @caller1() {
 
 ; x = [100, 301)
 define internal i1 @f.zext(i32 %x, i32 %y) {
-; CHECK-LABEL: @f.zext(
-; CHECK-NEXT:    [[T_1:%.*]] = zext nneg i32 [[X:%.*]] to i64
+; CHECK-LABEL: define internal i1 @f.zext(
+; CHECK-SAME: i32 [[X:%.*]], i32 [[Y:%.*]]) {
+; CHECK-NEXT:    [[T_1:%.*]] = zext nneg i32 [[X]] to i64
 ; CHECK-NEXT:    [[C_2:%.*]] = icmp sgt i64 [[T_1]], 299
 ; CHECK-NEXT:    [[C_4:%.*]] = icmp slt i64 [[T_1]], 101
 ; CHECK-NEXT:    [[RES_1:%.*]] = add nuw nsw i1 false, [[C_2]]
 ; CHECK-NEXT:    [[RES_2:%.*]] = add nuw nsw i1 [[RES_1]], false
 ; CHECK-NEXT:    [[RES_3:%.*]] = add i1 [[RES_2]], [[C_4]]
-; CHECK-NEXT:    [[T_2:%.*]] = zext i32 [[Y:%.*]] to i64
+; CHECK-NEXT:    [[T_2:%.*]] = zext i32 [[Y]] to i64
 ; CHECK-NEXT:    [[C_5:%.*]] = icmp sgt i64 [[T_2]], 300
 ; CHECK-NEXT:    [[C_6:%.*]] = icmp sgt i64 [[T_2]], 299
 ; CHECK-NEXT:    [[C_8:%.*]] = icmp slt i64 [[T_2]], 1
@@ -97,7 +99,7 @@ define internal i1 @f.zext(i32 %x, i32 %y) {
 }
 
 define i1 @caller.zext() {
-; CHECK-LABEL: @caller.zext(
+; CHECK-LABEL: define i1 @caller.zext() {
 ; CHECK-NEXT:    [[CALL_1:%.*]] = tail call i1 @f.zext(i32 100, i32 -120)
 ; CHECK-NEXT:    [[CALL_2:%.*]] = tail call i1 @f.zext(i32 300, i32 900)
 ; CHECK-NEXT:    [[RES:%.*]] = and i1 [[CALL_1]], [[CALL_2]]
@@ -111,14 +113,15 @@ define i1 @caller.zext() {
 
 ; x = [100, 301)
 define internal i1 @f.sext(i32 %x, i32 %y) {
-; CHECK-LABEL: @f.sext(
-; CHECK-NEXT:    [[T_1:%.*]] = zext nneg i32 [[X:%.*]] to i64
+; CHECK-LABEL: define internal i1 @f.sext(
+; CHECK-SAME: i32 [[X:%.*]], i32 [[Y:%.*]]) {
+; CHECK-NEXT:    [[T_1:%.*]] = zext nneg i32 [[X]] to i64
 ; CHECK-NEXT:    [[C_2:%.*]] = icmp sgt i64 [[T_1]], 299
 ; CHECK-NEXT:    [[C_4:%.*]] = icmp slt i64 [[T_1]], 101
 ; CHECK-NEXT:    [[RES_1:%.*]] = add nuw nsw i1 false, [[C_2]]
 ; CHECK-NEXT:    [[RES_2:%.*]] = add nuw nsw i1 [[RES_1]], false
 ; CHECK-NEXT:    [[RES_3:%.*]] = add i1 [[RES_2]], [[C_4]]
-; CHECK-NEXT:    [[T_2:%.*]] = sext i32 [[Y:%.*]] to i64
+; CHECK-NEXT:    [[T_2:%.*]] = sext i32 [[Y]] to i64
 ; CHECK-NEXT:    [[C_6:%.*]] = icmp sgt i64 [[T_2]], 899
 ; CHECK-NEXT:    [[C_8:%.*]] = icmp slt i64 [[T_2]], -119
 ; CHECK-NEXT:    [[RES_4:%.*]] = add nuw nsw i1 [[RES_3]], false
@@ -148,7 +151,7 @@ define internal i1 @f.sext(i32 %x, i32 %y) {
 }
 
 define i1 @caller.sext() {
-; CHECK-LABEL: @caller.sext(
+; CHECK-LABEL: define i1 @caller.sext() {
 ; CHECK-NEXT:    [[CALL_1:%.*]] = tail call i1 @f.sext(i32 100, i32 -120)
 ; CHECK-NEXT:    [[CALL_2:%.*]] = tail call i1 @f.sext(i32 300, i32 900)
 ; CHECK-NEXT:    [[RES:%.*]] = and i1 [[CALL_1]], [[CALL_2]]
@@ -162,8 +165,9 @@ define i1 @caller.sext() {
 
 ; There's nothing we can do besides going to the full range or overdefined.
 define internal i1 @f.fptosi(i32 %x) {
-; CHECK-LABEL: @f.fptosi(
-; CHECK-NEXT:    [[TO_DOUBLE:%.*]] = sitofp i32 [[X:%.*]] to double
+; CHECK-LABEL: define internal i1 @f.fptosi(
+; CHECK-SAME: i32 [[X:%.*]]) {
+; CHECK-NEXT:    [[TO_DOUBLE:%.*]] = sitofp i32 [[X]] to double
 ; CHECK-NEXT:    [[ADD:%.*]] = fadd double 0.000000e+00, [[TO_DOUBLE]]
 ; CHECK-NEXT:    [[TO_I32:%.*]] = fptosi double [[ADD]] to i32
 ; CHECK-NEXT:    [[C_1:%.*]] = icmp sgt i32 [[TO_I32]], 300
@@ -189,7 +193,7 @@ define internal i1 @f.fptosi(i32 %x) {
 }
 
 define i1 @caller.fptosi() {
-; CHECK-LABEL: @caller.fptosi(
+; CHECK-LABEL: define i1 @caller.fptosi() {
 ; CHECK-NEXT:    [[CALL_1:%.*]] = tail call i1 @f.fptosi(i32 100)
 ; CHECK-NEXT:    [[CALL_2:%.*]] = tail call i1 @f.fptosi(i32 300)
 ; CHECK-NEXT:    [[RES:%.*]] = and i1 [[CALL_1]], [[CALL_2]]
@@ -203,8 +207,9 @@ define i1 @caller.fptosi() {
 
 ; There's nothing we can do besides going to the full range or overdefined.
 define internal i1 @f.fpext(i16 %x) {
-; CHECK-LABEL: @f.fpext(
-; CHECK-NEXT:    [[TO_FLOAT:%.*]] = sitofp i16 [[X:%.*]] to float
+; CHECK-LABEL: define internal i1 @f.fpext(
+; CHECK-SAME: i16 [[X:%.*]]) {
+; CHECK-NEXT:    [[TO_FLOAT:%.*]] = sitofp i16 [[X]] to float
 ; CHECK-NEXT:    [[TO_DOUBLE:%.*]] = fpext float [[TO_FLOAT]] to double
 ; CHECK-NEXT:    [[TO_I64:%.*]] = fptoui float [[TO_FLOAT]] to i64
 ; CHECK-NEXT:    [[C_1:%.*]] = icmp sgt i64 [[TO_I64]], 300
@@ -231,7 +236,7 @@ define internal i1 @f.fpext(i16 %x) {
 
 ; There's nothing we can do besides going to the full range or overdefined.
 define i1 @caller.fpext() {
-; CHECK-LABEL: @caller.fpext(
+; CHECK-LABEL: define i1 @caller.fpext() {
 ; CHECK-NEXT:    [[CALL_1:%.*]] = tail call i1 @f.fpext(i16 100)
 ; CHECK-NEXT:    [[CALL_2:%.*]] = tail call i1 @f.fpext(i16 300)
 ; CHECK-NEXT:    [[RES:%.*]] = and i1 [[CALL_1]], [[CALL_2]]
@@ -245,8 +250,9 @@ define i1 @caller.fpext() {
 
 ; There's nothing we can do besides going to the full range or overdefined.
 define internal i1 @f.inttoptr.ptrtoint(i64 %x) {
-; CHECK-LABEL: @f.inttoptr.ptrtoint(
-; CHECK-NEXT:    [[TO_PTR:%.*]] = inttoptr i64 [[X:%.*]] to ptr
+; CHECK-LABEL: define internal i1 @f.inttoptr.ptrtoint(
+; CHECK-SAME: i64 [[X:%.*]]) {
+; CHECK-NEXT:    [[TO_PTR:%.*]] = inttoptr i64 [[X]] to ptr
 ; CHECK-NEXT:    [[TO_I64:%.*]] = ptrtoint ptr [[TO_PTR]] to i64
 ; CHECK-NEXT:    [[C_1:%.*]] = icmp sgt i64 [[TO_I64]], 300
 ; CHECK-NEXT:    [[C_2:%.*]] = icmp sgt i64 [[TO_I64]], 299
@@ -270,7 +276,7 @@ define internal i1 @f.inttoptr.ptrtoint(i64 %x) {
 }
 
 define i1 @caller.inttoptr.ptrtoint() {
-; CHECK-LABEL: @caller.inttoptr.ptrtoint(
+; CHECK-LABEL: define i1 @caller.inttoptr.ptrtoint() {
 ; CHECK-NEXT:    [[CALL_1:%.*]] = tail call i1 @f.inttoptr.ptrtoint(i64 100)
 ; CHECK-NEXT:    [[CALL_2:%.*]] = tail call i1 @f.inttoptr.ptrtoint(i64 300)
 ; CHECK-NEXT:    [[RES:%.*]] = and i1 [[CALL_1]], [[CALL_2]]
@@ -284,8 +290,9 @@ define i1 @caller.inttoptr.ptrtoint() {
 
 ; Make sure we do not create constant ranges for int to fp casts.
 define i1 @int_range_to_double_cast(i32 %a) {
-; CHECK-LABEL: @int_range_to_double_cast(
-; CHECK-NEXT:    [[R:%.*]] = and i32 [[A:%.*]], 255
+; CHECK-LABEL: define i1 @int_range_to_double_cast(
+; CHECK-SAME: i32 [[A:%.*]]) {
+; CHECK-NEXT:    [[R:%.*]] = and i32 [[A]], 255
 ; CHECK-NEXT:    [[T4:%.*]] = sitofp i32 [[R]] to double
 ; CHECK-NEXT:    [[T10:%.*]] = fadd double 0.000000e+00, [[T4]]
 ; CHECK-NEXT:    [[T11:%.*]] = fcmp olt double [[T4]], [[T10]]
@@ -300,7 +307,7 @@ define i1 @int_range_to_double_cast(i32 %a) {
 
 ; Make sure we do not use ranges to propagate info from vectors.
 define i16 @vector_binop_and_cast() {
-; CHECK-LABEL: @vector_binop_and_cast(
+; CHECK-LABEL: define i16 @vector_binop_and_cast() {
 ; CHECK-NEXT:  entry:
 ; CHECK-NEXT:    [[VECINIT7:%.*]] = insertelement <8 x i16> <i16 undef, i16 1, i16 2, i16 3, i16 4, i16 5, i16 6, i16 7>, i16 undef, i32 0
 ; CHECK-NEXT:    [[REM:%.*]] = srem <8 x i16> <i16 2, i16 2, i16 2, i16 2, i16 2, i16 2, i16 2, i16 2>, [[VECINIT7]]
@@ -317,8 +324,9 @@ entry:
 }
 
 define internal i64 @f.sext_to_zext(i32 %t) {
-; CHECK-LABEL: @f.sext_to_zext(
-; CHECK-NEXT:    [[A:%.*]] = zext nneg i32 [[T:%.*]] to i64
+; CHECK-LABEL: define internal range(i64 0, 2) i64 @f.sext_to_zext(
+; CHECK-SAME: i32 [[T:%.*]]) {
+; CHECK-NEXT:    [[A:%.*]] = zext nneg i32 [[T]] to i64
 ; CHECK-NEXT:    ret i64 [[A]]
 ;
   %a = sext i32 %t to i64
@@ -326,10 +334,11 @@ define internal i64 @f.sext_to_zext(i32 %t) {
 }
 
 define i64 @caller.sext_to_zext(i32 %i) {
-; CHECK-LABEL: @caller.sext_to_zext(
-; CHECK-NEXT:    [[CMP:%.*]] = icmp sle i32 [[I:%.*]], 9
+; CHECK-LABEL: define range(i64 0, 2) i64 @caller.sext_to_zext(
+; CHECK-SAME: i32 [[I:%.*]]) {
+; CHECK-NEXT:    [[CMP:%.*]] = icmp sle i32 [[I]], 9
 ; CHECK-NEXT:    [[CONV:%.*]] = zext i1 [[CMP]] to i32
-; CHECK-NEXT:    [[T:%.*]] = call i64 @f.sext_to_zext(i32 [[CONV]]), !range [[RNG0:![0-9]+]]
+; CHECK-NEXT:    [[T:%.*]] = call i64 @f.sext_to_zext(i32 [[CONV]])
 ; CHECK-NEXT:    ret i64 [[T]]
 ;
   %cmp = icmp sle i32 %i, 9
diff --git a/llvm/test/Transforms/SCCP/ipsccp-basic.ll b/llvm/test/Transforms/SCCP/ipsccp-basic.ll
index 71c042b9b294..6a7ab8ac2864 100644
--- a/llvm/test/Transforms/SCCP/ipsccp-basic.ll
+++ b/llvm/test/Transforms/SCCP/ipsccp-basic.ll
@@ -71,7 +71,7 @@ define void @test3a() {
 }
 
 define i32 @test3b() {
-; CHECK-LABEL: define i32 @test3b() {
+; CHECK-LABEL: define range(i32 0, 18) i32 @test3b() {
 ; CHECK-NEXT:    [[V:%.*]] = load i32, ptr @G, align 4
 ; CHECK-NEXT:    [[C:%.*]] = icmp eq i32 [[V]], 17
 ; CHECK-NEXT:    br i1 [[C]], label [[T:%.*]], label [[F:%.*]]
@@ -105,7 +105,7 @@ define internal {i64,i64} @test4a() {
 }
 
 define i64 @test4b() personality ptr @__gxx_personality_v0 {
-; CHECK-LABEL: define i64 @test4b() personality ptr @__gxx_personality_v0 {
+; CHECK-LABEL: define range(i64 0, 6) i64 @test4b() personality ptr @__gxx_personality_v0 {
 ; CHECK-NEXT:    [[A:%.*]] = invoke { i64, i64 } @test4a()
 ; CHECK-NEXT:    to label [[A:%.*]] unwind label [[B:%.*]]
 ; CHECK:       A:
@@ -149,7 +149,7 @@ define internal {i64,i64} @test5a() {
 }
 
 define i64 @test5b() personality ptr @__gxx_personality_v0 {
-; CHECK-LABEL: define i64 @test5b() personality ptr @__gxx_personality_v0 {
+; CHECK-LABEL: define range(i64 0, 6) i64 @test5b() personality ptr @__gxx_personality_v0 {
 ; CHECK-NEXT:    [[A:%.*]] = invoke { i64, i64 } @test5a()
 ; CHECK-NEXT:    to label [[A:%.*]] unwind label [[B:%.*]]
 ; CHECK:       A:
diff --git a/llvm/test/Transforms/SCCP/switch.ll b/llvm/test/Transforms/SCCP/switch.ll
index 306f0eebf2b4..5208213de210 100644
--- a/llvm/test/Transforms/SCCP/switch.ll
+++ b/llvm/test/Transforms/SCCP/switch.ll
@@ -1,4 +1,4 @@
-; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
+; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --version 4
 ; RUN: opt -S -passes=ipsccp < %s | FileCheck %s
 
 ; Make sure we always consider the default edge executable for a switch
@@ -7,7 +7,7 @@ declare void @foo()
 declare i32 @g(i32)
 
 define void @test1() {
-; CHECK-LABEL: @test1(
+; CHECK-LABEL: define void @test1() {
 ; CHECK-NEXT:    switch i32 undef, label [[D:%.*]] [
 ; CHECK-NEXT:    ]
 ; CHECK:       d:
@@ -21,15 +21,16 @@ d:
 }
 
 define i32 @test_duplicate_successors_phi(i1 %c, i32 %x) {
-; CHECK-LABEL: @test_duplicate_successors_phi(
+; CHECK-LABEL: define i32 @test_duplicate_successors_phi(
+; CHECK-SAME: i1 [[C:%.*]], i32 [[X:%.*]]) {
 ; CHECK-NEXT:  entry:
-; CHECK-NEXT:    br i1 [[C:%.*]], label [[SWITCH:%.*]], label [[END:%.*]]
+; CHECK-NEXT:    br i1 [[C]], label [[SWITCH:%.*]], label [[END:%.*]]
 ; CHECK:       switch:
 ; CHECK-NEXT:    br label [[SWITCH_DEFAULT:%.*]]
 ; CHECK:       switch.default:
 ; CHECK-NEXT:    ret i32 -1
 ; CHECK:       end:
-; CHECK-NEXT:    ret i32 [[X:%.*]]
+; CHECK-NEXT:    ret i32 [[X]]
 ;
 entry:
   br i1 %c, label %switch, label %end
@@ -49,13 +50,14 @@ end:
 }
 
 define i32 @test_duplicate_successors_phi_2(i1 %c, i32 %x) {
-; CHECK-LABEL: @test_duplicate_successors_phi_2(
+; CHECK-LABEL: define i32 @test_duplicate_successors_phi_2(
+; CHECK-SAME: i1 [[C:%.*]], i32 [[X:%.*]]) {
 ; CHECK-NEXT:  entry:
-; CHECK-NEXT:    br i1 [[C:%.*]], label [[SWITCH:%.*]], label [[END:%.*]]
+; CHECK-NEXT:    br i1 [[C]], label [[SWITCH:%.*]], label [[END:%.*]]
 ; CHECK:       switch:
 ; CHECK-NEXT:    br label [[END]]
 ; CHECK:       end:
-; CHECK-NEXT:    [[PHI:%.*]] = phi i32 [ [[X:%.*]], [[ENTRY:%.*]] ], [ 1, [[SWITCH]] ]
+; CHECK-NEXT:    [[PHI:%.*]] = phi i32 [ [[X]], [[ENTRY:%.*]] ], [ 1, [[SWITCH]] ]
 ; CHECK-NEXT:    ret i32 [[PHI]]
 ;
 entry:
@@ -76,22 +78,23 @@ end:
 }
 
 define i32 @test_duplicate_successors_phi_3(i1 %c1, ptr %p, i32 %y) {
-; CHECK-LABEL: @test_duplicate_successors_phi_3(
+; CHECK-LABEL: define i32 @test_duplicate_successors_phi_3(
+; CHECK-SAME: i1 [[C1:%.*]], ptr [[P:%.*]], i32 [[Y:%.*]]) {
 ; CHECK-NEXT:  entry:
-; CHECK-NEXT:    br i1 [[C1:%.*]], label [[SWITCH:%.*]], label [[SWITCH_1:%.*]]
+; CHECK-NEXT:    br i1 [[C1]], label [[SWITCH:%.*]], label [[SWITCH_1:%.*]]
 ; CHECK:       switch:
-; CHECK-NEXT:    [[X:%.*]] = load i32, ptr [[P:%.*]], align 4, !range [[RNG0:![0-9]+]]
+; CHECK-NEXT:    [[X:%.*]] = load i32, ptr [[P]], align 4, !range [[RNG0:![0-9]+]]
 ; CHECK-NEXT:    switch i32 [[X]], label [[SWITCH_DEFAULT:%.*]] [
-; CHECK-NEXT:    i32 0, label [[SWITCH_DEFAULT]]
-; CHECK-NEXT:    i32 1, label [[SWITCH_0:%.*]]
-; CHECK-NEXT:    i32 2, label [[SWITCH_0]]
+; CHECK-NEXT:      i32 0, label [[SWITCH_DEFAULT]]
+; CHECK-NEXT:      i32 1, label [[SWITCH_0:%.*]]
+; CHECK-NEXT:      i32 2, label [[SWITCH_0]]
 ; CHECK-NEXT:    ]
 ; CHECK:       switch.default:
 ; CHECK-NEXT:    ret i32 -1
 ; CHECK:       switch.0:
 ; CHECK-NEXT:    ret i32 0
 ; CHECK:       switch.1:
-; CHECK-NEXT:    ret i32 [[Y:%.*]]
+; CHECK-NEXT:    ret i32 [[Y]]
 ;
 entry:
   br i1 %c1, label %switch, label %switch.1
@@ -118,12 +121,13 @@ switch.1:
 }
 
 define i32 @test_local_range(ptr %p) {
-; CHECK-LABEL: @test_local_range(
-; CHECK-NEXT:    [[X:%.*]] = load i32, ptr [[P:%.*]], align 4, !range [[RNG0]]
+; CHECK-LABEL: define range(i32 0, 3) i32 @test_local_range(
+; CHECK-SAME: ptr [[P:%.*]]) {
+; CHECK-NEXT:    [[X:%.*]] = load i32, ptr [[P]], align 4, !range [[RNG0]]
 ; CHECK-NEXT:    switch i32 [[X]], label [[DEFAULT_UNREACHABLE:%.*]] [
-; CHECK-NEXT:    i32 0, label [[SWITCH_0:%.*]]
-; CHECK-NEXT:    i32 1, label [[SWITCH_1:%.*]]
-; CHECK-NEXT:    i32 2, label [[SWITCH_2:%.*]]
+; CHECK-NEXT:      i32 0, label [[SWITCH_0:%.*]]
+; CHECK-NEXT:      i32 1, label [[SWITCH_1:%.*]]
+; CHECK-NEXT:      i32 2, label [[SWITCH_2:%.*]]
 ; CHECK-NEXT:    ]
 ; CHECK:       default.unreachable:
 ; CHECK-NEXT:    unreachable
@@ -160,13 +164,14 @@ switch.3:
 
 ; TODO: Determine that case i3 is dead, even though the edge is shared?
 define i32 @test_duplicate_successors(ptr %p) {
-; CHECK-LABEL: @test_duplicate_successors(
-; CHECK-NEXT:    [[X:%.*]] = load i32, ptr [[P:%.*]], align 4, !range [[RNG0]]
+; CHECK-LABEL: define range(i32 0, 2) i32 @test_duplicate_successors(
+; CHECK-SAME: ptr [[P:%.*]]) {
+; CHECK-NEXT:    [[X:%.*]] = load i32, ptr [[P]], align 4, !range [[RNG0]]
 ; CHECK-NEXT:    switch i32 [[X]], label [[DEFAULT_UNREACHABLE:%.*]] [
-; CHECK-NEXT:    i32 0, label [[SWITCH_0:%.*]]
-; CHECK-NEXT:    i32 1, label [[SWITCH_0]]
-; CHECK-NEXT:    i32 2, label [[SWITCH_1:%.*]]
-; CHECK-NEXT:    i32 3, label [[SWITCH_1]]
+; CHECK-NEXT:      i32 0, label [[SWITCH_0:%.*]]
+; CHECK-NEXT:      i32 1, label [[SWITCH_0]]
+; CHECK-NEXT:      i32 2, label [[SWITCH_1:%.*]]
+; CHECK-NEXT:      i32 3, label [[SWITCH_1]]
 ; CHECK-NEXT:    ]
 ; CHECK:       default.unreachable:
 ; CHECK-NEXT:    unreachable
@@ -201,11 +206,12 @@ switch.2:
 ; Case i32 2 is dead as well, but this cannot be determined based on
 ; range information.
 define internal i32 @test_ip_range(i32 %x) {
-; CHECK-LABEL: @test_ip_range(
-; CHECK-NEXT:    switch i32 [[X:%.*]], label [[DEFAULT_UNREACHABLE:%.*]] [
-; CHECK-NEXT:    i32 3, label [[SWITCH_3:%.*]]
-; CHECK-NEXT:    i32 1, label [[SWITCH_1:%.*]]
-; CHECK-NEXT:    i32 2, label [[SWITCH_2:%.*]]
+; CHECK-LABEL: define internal range(i32 1, 4) i32 @test_ip_range(
+; CHECK-SAME: i32 [[X:%.*]]) {
+; CHECK-NEXT:    switch i32 [[X]], label [[DEFAULT_UNREACHABLE:%.*]] [
+; CHECK-NEXT:      i32 3, label [[SWITCH_3:%.*]]
+; CHECK-NEXT:      i32 1, label [[SWITCH_1:%.*]]
+; CHECK-NEXT:      i32 2, label [[SWITCH_2:%.*]]
 ; CHECK-NEXT:    ], !prof [[PROF1:![0-9]+]]
 ; CHECK:       default.unreachable:
 ; CHECK-NEXT:    unreachable
@@ -240,9 +246,9 @@ switch.3:
 }
 
 define void @call_test_ip_range() {
-; CHECK-LABEL: @call_test_ip_range(
-; CHECK-NEXT:    [[TMP1:%.*]] = call i32 @test_ip_range(i32 1), !range [[RNG2:![0-9]+]]
-; CHECK-NEXT:    [[TMP2:%.*]] = call i32 @test_ip_range(i32 3), !range [[RNG2]]
+; CHECK-LABEL: define void @call_test_ip_range() {
+; CHECK-NEXT:    [[TMP1:%.*]] = call i32 @test_ip_range(i32 1)
+; CHECK-NEXT:    [[TMP2:%.*]] = call i32 @test_ip_range(i32 3)
 ; CHECK-NEXT:    ret void
 ;
   call i32 @test_ip_range(i32 1)
@@ -251,11 +257,12 @@ define void @call_test_ip_range() {
 }
 
 define i32 @test_switch_range_may_include_undef(i1 %c.1, i1 %c.2, i32 %x) {
-; CHECK-LABEL: @test_switch_range_may_include_undef(
+; CHECK-LABEL: define range(i32 -1, 21) i32 @test_switch_range_may_include_undef(
+; CHECK-SAME: i1 [[C_1:%.*]], i1 [[C_2:%.*]], i32 [[X:%.*]]) {
 ; CHECK-NEXT:  entry:
-; CHECK-NEXT:    br i1 [[C_1:%.*]], label [[THEN_1:%.*]], label [[ELSE_1:%.*]]
+; CHECK-NEXT:    br i1 [[C_1]], label [[THEN_1:%.*]], label [[ELSE_1:%.*]]
 ; CHECK:       then.1:
-; CHECK-NEXT:    br i1 [[C_2:%.*]], label [[SWITCH:%.*]], label [[ELSE_2:%.*]]
+; CHECK-NEXT:    br i1 [[C_2]], label [[SWITCH:%.*]], label [[ELSE_2:%.*]]
 ; CHECK:       else.1:
 ; CHECK-NEXT:    br label [[SWITCH]]
 ; CHECK:       else.2:
@@ -263,8 +270,8 @@ define i32 @test_switch_range_may_include_undef(i1 %c.1, i1 %c.2, i32 %x) {
 ; CHECK:       switch:
 ; CHECK-NEXT:    [[P:%.*]] = phi i32 [ 0, [[THEN_1]] ], [ 2, [[ELSE_1]] ], [ undef, [[ELSE_2]] ]
 ; CHECK-NEXT:    switch i32 [[P]], label [[SWITCH_DEFAULT:%.*]] [
-; CHECK-NEXT:    i32 0, label [[END_1:%.*]]
-; CHECK-NEXT:    i32 3, label [[END_2:%.*]]
+; CHECK-NEXT:      i32 0, label [[END_1:%.*]]
+; CHECK-NEXT:      i32 3, label [[END_2:%.*]]
 ; CHECK-NEXT:    ]
 ; CHECK:       switch.default:
 ; CHECK-NEXT:    ret i32 -1
@@ -303,9 +310,10 @@ end.2:
 }
 
 define i32 @test_default_unreachable_by_dom_cond(i32 %x) {
-; CHECK-LABEL: @test_default_unreachable_by_dom_cond(
+; CHECK-LABEL: define i32 @test_default_unreachable_by_dom_cond(
+; CHECK-SAME: i32 [[X:%.*]]) {
 ; CHECK-NEXT:  entry:
-; CHECK-NEXT:    [[OR_COND:%.*]] = icmp ult i32 [[X:%.*]], 4
+; CHECK-NEXT:    [[OR_COND:%.*]] = icmp ult i32 [[X]], 4
 ; CHECK-NEXT:    br i1 [[OR_COND]], label [[IF_THEN:%.*]], label [[RETURN:%.*]]
 ; CHECK:       if.then:
 ; CHECK-NEXT:    switch i32 [[X]], label [[DEFAULT_UNREACHABLE:%.*]] [
@@ -371,4 +379,7 @@ return:
 
 declare void @llvm.assume(i1)
 
-; CHECK: !1 = !{!"branch_weights", i32 1, i32 5, i32 3, i32 4}
+;.
+; CHECK: [[RNG0]] = !{i32 0, i32 3}
+; CHECK: [[PROF1]] = !{!"branch_weights", i32 1, i32 5, i32 3, i32 4}
+;.
diff --git a/llvm/test/Transforms/SCCP/trunc-nuw-nsw-flags.ll b/llvm/test/Transforms/SCCP/trunc-nuw-nsw-flags.ll
index fc3e56011d46..d3bac0d68a97 100644
--- a/llvm/test/Transforms/SCCP/trunc-nuw-nsw-flags.ll
+++ b/llvm/test/Transforms/SCCP/trunc-nuw-nsw-flags.ll
@@ -16,7 +16,7 @@ entry:
 }
 
 define i8 @range_from_or_nsw(i16 %a) {
-; CHECK-LABEL: define i8 @range_from_or_nsw(
+; CHECK-LABEL: define range(i8 -128, 0) i8 @range_from_or_nsw(
 ; CHECK-SAME: i16 [[A:%.*]]) {
 ; CHECK-NEXT:  entry:
 ; CHECK-NEXT:    [[AND1:%.*]] = or i16 [[A]], -128
@@ -30,7 +30,7 @@ entry:
 }
 
 define i16 @range_from_and_nuw_nsw(i32 %a) {
-; CHECK-LABEL: define i16 @range_from_and_nuw_nsw(
+; CHECK-LABEL: define range(i16 0, -32768) i16 @range_from_and_nuw_nsw(
 ; CHECK-SAME: i32 [[A:%.*]]) {
 ; CHECK-NEXT:  entry:
 ; CHECK-NEXT:    [[AND1:%.*]] = and i32 [[A]], 32767
diff --git a/llvm/test/Transforms/SLPVectorizer/AArch64/gather-buildvector-with-minbitwidth-user.ll b/llvm/test/Transforms/SLPVectorizer/AArch64/gather-buildvector-with-minbitwidth-user.ll
index 690772472975..3771ec4bda88 100644
--- a/llvm/test/Transforms/SLPVectorizer/AArch64/gather-buildvector-with-minbitwidth-user.ll
+++ b/llvm/test/Transforms/SLPVectorizer/AArch64/gather-buildvector-with-minbitwidth-user.ll
@@ -5,12 +5,7 @@ define void @h() {
 ; CHECK-LABEL: define void @h() {
 ; CHECK-NEXT:  entry:
 ; CHECK-NEXT:    [[ARRAYIDX2:%.*]] = getelementptr i8, ptr null, i64 16
-; CHECK-NEXT:    [[TMP0:%.*]] = insertelement <8 x i32> <i32 undef, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0>, i32 0, i32 0
-; CHECK-NEXT:    [[TMP1:%.*]] = trunc <8 x i32> [[TMP0]] to <8 x i1>
-; CHECK-NEXT:    [[TMP2:%.*]] = or <8 x i1> zeroinitializer, [[TMP1]]
-; CHECK-NEXT:    [[TMP4:%.*]] = or <8 x i1> [[TMP2]], zeroinitializer
-; CHECK-NEXT:    [[TMP3:%.*]] = zext <8 x i1> [[TMP4]] to <8 x i16>
-; CHECK-NEXT:    store <8 x i16> [[TMP3]], ptr [[ARRAYIDX2]], align 2
+; CHECK-NEXT:    store <8 x i16> zeroinitializer, ptr [[ARRAYIDX2]], align 2
 ; CHECK-NEXT:    ret void
 ;
 entry:
diff --git a/llvm/test/Transforms/SLPVectorizer/AArch64/gather-with-minbith-user.ll b/llvm/test/Transforms/SLPVectorizer/AArch64/gather-with-minbith-user.ll
index d51ef0bce3a4..76bb882171b1 100644
--- a/llvm/test/Transforms/SLPVectorizer/AArch64/gather-with-minbith-user.ll
+++ b/llvm/test/Transforms/SLPVectorizer/AArch64/gather-with-minbith-user.ll
@@ -5,7 +5,8 @@ define void @h() {
 ; CHECK-LABEL: define void @h() {
 ; CHECK-NEXT:  entry:
 ; CHECK-NEXT:    [[ARRAYIDX2:%.*]] = getelementptr i8, ptr null, i64 16
-; CHECK-NEXT:    [[TMP0:%.*]] = trunc <8 x i32> zeroinitializer to <8 x i1>
+; CHECK-NEXT:    [[TMP6:%.*]] = trunc i32 0 to i1
+; CHECK-NEXT:    [[TMP0:%.*]] = insertelement <8 x i1> <i1 false, i1 false, i1 false, i1 false, i1 poison, i1 false, i1 false, i1 false>, i1 [[TMP6]], i32 4
 ; CHECK-NEXT:    [[TMP1:%.*]] = sub <8 x i1> [[TMP0]], zeroinitializer
 ; CHECK-NEXT:    [[TMP2:%.*]] = add <8 x i1> [[TMP0]], zeroinitializer
 ; CHECK-NEXT:    [[TMP3:%.*]] = shufflevector <8 x i1> [[TMP1]], <8 x i1> [[TMP2]], <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 12, i32 13, i32 14, i32 15>
diff --git a/llvm/test/Transforms/SLPVectorizer/AArch64/user-node-not-in-bitwidths.ll b/llvm/test/Transforms/SLPVectorizer/AArch64/user-node-not-in-bitwidths.ll
index 6404cf4a2cd1..2ab6e919c23b 100644
--- a/llvm/test/Transforms/SLPVectorizer/AArch64/user-node-not-in-bitwidths.ll
+++ b/llvm/test/Transforms/SLPVectorizer/AArch64/user-node-not-in-bitwidths.ll
@@ -5,7 +5,12 @@ define void @h() {
 ; CHECK-LABEL: define void @h() {
 ; CHECK-NEXT:  entry:
 ; CHECK-NEXT:    [[ARRAYIDX2:%.*]] = getelementptr i8, ptr null, i64 16
-; CHECK-NEXT:    store <8 x i16> zeroinitializer, ptr [[ARRAYIDX2]], align 2
+; CHECK-NEXT:    [[TMP0:%.*]] = trunc i32 0 to i1
+; CHECK-NEXT:    [[TMP1:%.*]] = insertelement <8 x i1> <i1 false, i1 false, i1 false, i1 false, i1 poison, i1 false, i1 false, i1 false>, i1 [[TMP0]], i32 4
+; CHECK-NEXT:    [[TMP2:%.*]] = or <8 x i1> zeroinitializer, [[TMP1]]
+; CHECK-NEXT:    [[TMP3:%.*]] = or <8 x i1> zeroinitializer, [[TMP2]]
+; CHECK-NEXT:    [[TMP4:%.*]] = zext <8 x i1> [[TMP3]] to <8 x i16>
+; CHECK-NEXT:    store <8 x i16> [[TMP4]], ptr [[ARRAYIDX2]], align 2
 ; CHECK-NEXT:    ret void
 ;
 entry:
diff --git a/llvm/test/Transforms/SLPVectorizer/AArch64/vec3-reorder-reshuffle.ll b/llvm/test/Transforms/SLPVectorizer/AArch64/vec3-reorder-reshuffle.ll
index 47d918eabdfe..9bbd314a27cb 100644
--- a/llvm/test/Transforms/SLPVectorizer/AArch64/vec3-reorder-reshuffle.ll
+++ b/llvm/test/Transforms/SLPVectorizer/AArch64/vec3-reorder-reshuffle.ll
@@ -537,24 +537,18 @@ entry:
 }
 
 define void @vec3_extract(<3 x i16> %pixel.sroa.0.4.vec.insert606, ptr %call3.i536) {
-; NON-POW2-LABEL: define void @vec3_extract(
-; NON-POW2-SAME: <3 x i16> [[PIXEL_SROA_0_4_VEC_INSERT606:%.*]], ptr [[CALL3_I536:%.*]]) {
-; NON-POW2-NEXT:  entry:
-; NON-POW2-NEXT:    store <3 x i16> [[PIXEL_SROA_0_4_VEC_INSERT606]], ptr [[CALL3_I536]], align 2
-; NON-POW2-NEXT:    ret void
-;
-; POW2-ONLY-LABEL: define void @vec3_extract(
-; POW2-ONLY-SAME: <3 x i16> [[PIXEL_SROA_0_4_VEC_INSERT606:%.*]], ptr [[CALL3_I536:%.*]]) {
-; POW2-ONLY-NEXT:  entry:
-; POW2-ONLY-NEXT:    [[PIXEL_SROA_0_4_VEC_EXTRACT:%.*]] = extractelement <3 x i16> [[PIXEL_SROA_0_4_VEC_INSERT606]], i64 2
-; POW2-ONLY-NEXT:    [[RED668:%.*]] = getelementptr i16, ptr [[CALL3_I536]], i64 2
-; POW2-ONLY-NEXT:    store i16 [[PIXEL_SROA_0_4_VEC_EXTRACT]], ptr [[RED668]], align 2
-; POW2-ONLY-NEXT:    [[PIXEL_SROA_0_2_VEC_EXTRACT:%.*]] = extractelement <3 x i16> [[PIXEL_SROA_0_4_VEC_INSERT606]], i64 1
-; POW2-ONLY-NEXT:    [[GREEN670:%.*]] = getelementptr i16, ptr [[CALL3_I536]], i64 1
-; POW2-ONLY-NEXT:    store i16 [[PIXEL_SROA_0_2_VEC_EXTRACT]], ptr [[GREEN670]], align 2
-; POW2-ONLY-NEXT:    [[PIXEL_SROA_0_0_VEC_EXTRACT:%.*]] = extractelement <3 x i16> [[PIXEL_SROA_0_4_VEC_INSERT606]], i64 0
-; POW2-ONLY-NEXT:    store i16 [[PIXEL_SROA_0_0_VEC_EXTRACT]], ptr [[CALL3_I536]], align 2
-; POW2-ONLY-NEXT:    ret void
+; CHECK-LABEL: define void @vec3_extract(
+; CHECK-SAME: <3 x i16> [[PIXEL_SROA_0_4_VEC_INSERT606:%.*]], ptr [[CALL3_I536:%.*]]) {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[PIXEL_SROA_0_4_VEC_EXTRACT:%.*]] = extractelement <3 x i16> [[PIXEL_SROA_0_4_VEC_INSERT606]], i64 2
+; CHECK-NEXT:    [[RED668:%.*]] = getelementptr i16, ptr [[CALL3_I536]], i64 2
+; CHECK-NEXT:    store i16 [[PIXEL_SROA_0_4_VEC_EXTRACT]], ptr [[RED668]], align 2
+; CHECK-NEXT:    [[PIXEL_SROA_0_2_VEC_EXTRACT:%.*]] = extractelement <3 x i16> [[PIXEL_SROA_0_4_VEC_INSERT606]], i64 1
+; CHECK-NEXT:    [[GREEN670:%.*]] = getelementptr i16, ptr [[CALL3_I536]], i64 1
+; CHECK-NEXT:    store i16 [[PIXEL_SROA_0_2_VEC_EXTRACT]], ptr [[GREEN670]], align 2
+; CHECK-NEXT:    [[PIXEL_SROA_0_0_VEC_EXTRACT:%.*]] = extractelement <3 x i16> [[PIXEL_SROA_0_4_VEC_INSERT606]], i64 0
+; CHECK-NEXT:    store i16 [[PIXEL_SROA_0_0_VEC_EXTRACT]], ptr [[CALL3_I536]], align 2
+; CHECK-NEXT:    ret void
 ;
 entry:
   %pixel.sroa.0.4.vec.extract = extractelement <3 x i16> %pixel.sroa.0.4.vec.insert606, i64 2
diff --git a/llvm/test/Transforms/SLPVectorizer/RISCV/minbw-with-and-and-scalar-trunc.ll b/llvm/test/Transforms/SLPVectorizer/RISCV/minbw-with-and-and-scalar-trunc.ll
new file mode 100644
index 000000000000..fc977585614b
--- /dev/null
+++ b/llvm/test/Transforms/SLPVectorizer/RISCV/minbw-with-and-and-scalar-trunc.ll
@@ -0,0 +1,44 @@
+; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --version 4
+; RUN: opt -S --passes=slp-vectorizer -mtriple=riscv64-unknown-linux-gnu -mattr=+v < %s | FileCheck %s
+@c = global [12 x i64] zeroinitializer
+
+; FIXME: after minbitwidth analysis and i32 conv.., 65535 is transformed to
+; and <4 x i16> , -1, which must be dropped.
+; FIXME: need to adjust the cost of the final transformation, since the user is
+; just a trunc to i16 (it must be free).
+define i16 @test() {
+; CHECK-LABEL: define i16 @test(
+; CHECK-SAME: ) #[[ATTR0:[0-9]+]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP0:%.*]] = call <4 x i64> @llvm.experimental.vp.strided.load.v4i64.p0.i64(ptr align 8 @c, i64 24, <4 x i1> <i1 true, i1 true, i1 true, i1 true>, i32 4)
+; CHECK-NEXT:    [[TMP1:%.*]] = trunc <4 x i64> [[TMP0]] to <4 x i16>
+; CHECK-NEXT:    [[TMP2:%.*]] = and <4 x i16> [[TMP1]], <i16 -1, i16 -1, i16 -1, i16 -1>
+; CHECK-NEXT:    [[TMP3:%.*]] = xor <4 x i16> [[TMP2]], <i16 -1, i16 -1, i16 -1, i16 -1>
+; CHECK-NEXT:    [[TMP4:%.*]] = call i16 @llvm.vector.reduce.umax.v4i16(<4 x i16> [[TMP3]])
+; CHECK-NEXT:    [[TMP5:%.*]] = zext i16 [[TMP4]] to i32
+; CHECK-NEXT:    [[T:%.*]] = trunc i32 [[TMP5]] to i16
+; CHECK-NEXT:    ret i16 [[T]]
+;
+entry:
+  %0 = load i64, ptr @c, align 8
+  %conv = trunc i64 %0 to i32
+  %conv3 = and i32 %conv, 65535
+  %conv4 = xor i32 %conv3, 65535
+  %1 = load i64, ptr getelementptr inbounds ([12 x i64], ptr @c, i64 0, i64 3), align 8
+  %conv.1 = trunc i64 %1 to i32
+  %conv3.1 = and i32 %conv.1, 65535
+  %conv4.1 = xor i32 %conv3.1, 65535
+  %.conv4.1 = tail call i32 @llvm.umax.i32(i32 %conv4, i32 %conv4.1)
+  %2 = load i64, ptr getelementptr inbounds ([12 x i64], ptr @c, i64 0, i64 6), align 8
+  %conv.2 = trunc i64 %2 to i32
+  %conv3.2 = and i32 %conv.2, 65535
+  %conv4.2 = xor i32 %conv3.2, 65535
+  %.conv4.2 = tail call i32 @llvm.umax.i32(i32 %.conv4.1, i32 %conv4.2)
+  %3 = load i64, ptr getelementptr inbounds ([12 x i64], ptr @c, i64 0, i64 9), align 8
+  %conv.3 = trunc i64 %3 to i32
+  %conv3.3 = and i32 %conv.3, 65535
+  %conv4.3 = xor i32 %conv3.3, 65535
+  %.conv4.3 = tail call i32 @llvm.umax.i32(i32 %.conv4.2, i32 %conv4.3)
+  %t = trunc i32 %.conv4.3 to i16
+  ret i16 %t
+}
diff --git a/llvm/test/Transforms/SLPVectorizer/RISCV/strided-stores-vectorized.ll b/llvm/test/Transforms/SLPVectorizer/RISCV/strided-stores-vectorized.ll
new file mode 100644
index 000000000000..0dfa45da9d87
--- /dev/null
+++ b/llvm/test/Transforms/SLPVectorizer/RISCV/strided-stores-vectorized.ll
@@ -0,0 +1,65 @@
+; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
+; RUN: opt -passes=slp-vectorizer -S < %s -mtriple=riscv64-unknown-linux -mattr=+v | FileCheck %s
+
+define void @store_reverse(ptr %p3) {
+; CHECK-LABEL: @store_reverse(
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP0:%.*]] = load i64, ptr [[P3:%.*]], align 8
+; CHECK-NEXT:    [[ARRAYIDX1:%.*]] = getelementptr inbounds i64, ptr [[P3]], i64 8
+; CHECK-NEXT:    [[TMP1:%.*]] = load i64, ptr [[ARRAYIDX1]], align 8
+; CHECK-NEXT:    [[SHL:%.*]] = shl i64 [[TMP0]], [[TMP1]]
+; CHECK-NEXT:    [[ARRAYIDX2:%.*]] = getelementptr inbounds i64, ptr [[P3]], i64 7
+; CHECK-NEXT:    store i64 [[SHL]], ptr [[ARRAYIDX2]], align 8
+; CHECK-NEXT:    [[ARRAYIDX3:%.*]] = getelementptr inbounds i64, ptr [[P3]], i64 1
+; CHECK-NEXT:    [[TMP2:%.*]] = load i64, ptr [[ARRAYIDX3]], align 8
+; CHECK-NEXT:    [[ARRAYIDX4:%.*]] = getelementptr inbounds i64, ptr [[P3]], i64 9
+; CHECK-NEXT:    [[TMP3:%.*]] = load i64, ptr [[ARRAYIDX4]], align 8
+; CHECK-NEXT:    [[SHL5:%.*]] = shl i64 [[TMP2]], [[TMP3]]
+; CHECK-NEXT:    [[ARRAYIDX6:%.*]] = getelementptr inbounds i64, ptr [[P3]], i64 6
+; CHECK-NEXT:    store i64 [[SHL5]], ptr [[ARRAYIDX6]], align 8
+; CHECK-NEXT:    [[ARRAYIDX7:%.*]] = getelementptr inbounds i64, ptr [[P3]], i64 2
+; CHECK-NEXT:    [[TMP4:%.*]] = load i64, ptr [[ARRAYIDX7]], align 8
+; CHECK-NEXT:    [[ARRAYIDX8:%.*]] = getelementptr inbounds i64, ptr [[P3]], i64 10
+; CHECK-NEXT:    [[TMP5:%.*]] = load i64, ptr [[ARRAYIDX8]], align 8
+; CHECK-NEXT:    [[SHL9:%.*]] = shl i64 [[TMP4]], [[TMP5]]
+; CHECK-NEXT:    [[ARRAYIDX10:%.*]] = getelementptr inbounds i64, ptr [[P3]], i64 5
+; CHECK-NEXT:    store i64 [[SHL9]], ptr [[ARRAYIDX10]], align 8
+; CHECK-NEXT:    [[ARRAYIDX11:%.*]] = getelementptr inbounds i64, ptr [[P3]], i64 3
+; CHECK-NEXT:    [[TMP6:%.*]] = load i64, ptr [[ARRAYIDX11]], align 8
+; CHECK-NEXT:    [[ARRAYIDX12:%.*]] = getelementptr inbounds i64, ptr [[P3]], i64 11
+; CHECK-NEXT:    [[TMP7:%.*]] = load i64, ptr [[ARRAYIDX12]], align 8
+; CHECK-NEXT:    [[SHL13:%.*]] = shl i64 [[TMP6]], [[TMP7]]
+; CHECK-NEXT:    [[ARRAYIDX14:%.*]] = getelementptr inbounds i64, ptr [[P3]], i64 4
+; CHECK-NEXT:    store i64 [[SHL13]], ptr [[ARRAYIDX14]], align 8
+; CHECK-NEXT:    ret void
+;
+entry:
+  %0 = load i64, ptr %p3, align 8
+  %arrayidx1 = getelementptr inbounds i64, ptr %p3, i64 8
+  %1 = load i64, ptr %arrayidx1, align 8
+  %shl = shl i64 %0, %1
+  %arrayidx2 = getelementptr inbounds i64, ptr %p3, i64 7
+  store i64 %shl, ptr %arrayidx2, align 8
+  %arrayidx3 = getelementptr inbounds i64, ptr %p3, i64 1
+  %2 = load i64, ptr %arrayidx3, align 8
+  %arrayidx4 = getelementptr inbounds i64, ptr %p3, i64 9
+  %3 = load i64, ptr %arrayidx4, align 8
+  %shl5 = shl i64 %2, %3
+  %arrayidx6 = getelementptr inbounds i64, ptr %p3, i64 6
+  store i64 %shl5, ptr %arrayidx6, align 8
+  %arrayidx7 = getelementptr inbounds i64, ptr %p3, i64 2
+  %4 = load i64, ptr %arrayidx7, align 8
+  %arrayidx8 = getelementptr inbounds i64, ptr %p3, i64 10
+  %5 = load i64, ptr %arrayidx8, align 8
+  %shl9 = shl i64 %4, %5
+  %arrayidx10 = getelementptr inbounds i64, ptr %p3, i64 5
+  store i64 %shl9, ptr %arrayidx10, align 8
+  %arrayidx11 = getelementptr inbounds i64, ptr %p3, i64 3
+  %6 = load i64, ptr %arrayidx11, align 8
+  %arrayidx12 = getelementptr inbounds i64, ptr %p3, i64 11
+  %7 = load i64, ptr %arrayidx12, align 8
+  %shl13 = shl i64 %6, %7
+  %arrayidx14 = getelementptr inbounds i64, ptr %p3, i64 4
+  store i64 %shl13, ptr %arrayidx14, align 8
+  ret void
+}
diff --git a/llvm/test/Transforms/SLPVectorizer/RISCV/unsigned-icmp-signed-op.ll b/llvm/test/Transforms/SLPVectorizer/RISCV/unsigned-icmp-signed-op.ll
new file mode 100644
index 000000000000..5ec6b4f1040d
--- /dev/null
+++ b/llvm/test/Transforms/SLPVectorizer/RISCV/unsigned-icmp-signed-op.ll
@@ -0,0 +1,43 @@
+; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --version 4
+; RUN: opt -S --passes=slp-vectorizer -mtriple=riscv64-unknown-linux-gnu -mattr=+v < %s | FileCheck %s
+
+define i32 @test(ptr %f, i16 %0) {
+; CHECK-LABEL: define i32 @test(
+; CHECK-SAME: ptr [[F:%.*]], i16 [[TMP0:%.*]]) #[[ATTR0:[0-9]+]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP1:%.*]] = load i16, ptr [[F]], align 2
+; CHECK-NEXT:    [[TMP2:%.*]] = insertelement <4 x i16> <i16 0, i16 poison, i16 0, i16 0>, i16 [[TMP0]], i32 1
+; CHECK-NEXT:    [[TMP3:%.*]] = insertelement <4 x i16> <i16 0, i16 poison, i16 0, i16 0>, i16 [[TMP1]], i32 1
+; CHECK-NEXT:    [[TMP6:%.*]] = zext <4 x i16> [[TMP3]] to <4 x i32>
+; CHECK-NEXT:    [[TMP7:%.*]] = sext <4 x i16> [[TMP2]] to <4 x i32>
+; CHECK-NEXT:    [[TMP4:%.*]] = icmp ule <4 x i32> [[TMP6]], [[TMP7]]
+; CHECK-NEXT:    [[TMP5:%.*]] = call i1 @llvm.vector.reduce.and.v4i1(<4 x i1> [[TMP4]])
+; CHECK-NEXT:    [[ZEXT_4:%.*]] = zext i1 [[TMP5]] to i32
+; CHECK-NEXT:    ret i32 [[ZEXT_4]]
+;
+entry:
+  %1 = load i16, ptr %f, align 2
+
+  %zext.0 = zext i16 %1 to i32
+  %sext.0 = sext i16 %0 to i32
+
+  %zext.1 = zext i16 0 to i32
+  %sext.1 = sext i16 0 to i32
+  %zext.2 = zext i16 0 to i32
+  %sext.2 = sext i16 0 to i32
+  %zext.3 = zext i16 0 to i32
+  %sext.3 = sext i16 0 to i32
+
+  %cmp.0 = icmp ule i32 %zext.0, %sext.0
+  %cmp.1 = icmp ule i32 %zext.1, %sext.1
+  %cmp.2 = icmp ule i32 %zext.2, %sext.2
+  %cmp.3 = icmp ule i32 %zext.3, %sext.3
+
+  %and.0 = and i1 %cmp.0, %cmp.1
+  %and.1 = and i1 %and.0, %cmp.2
+  %and.2 = and i1 %and.1, %cmp.3
+
+  %zext.4 = zext i1 %and.2 to i32
+
+  ret i32 %zext.4
+}
diff --git a/llvm/test/Transforms/SLPVectorizer/SystemZ/minbitwidth-root-trunc.ll b/llvm/test/Transforms/SLPVectorizer/SystemZ/minbitwidth-root-trunc.ll
index 7b4e2b0ce911..1bb87bf6205f 100644
--- a/llvm/test/Transforms/SLPVectorizer/SystemZ/minbitwidth-root-trunc.ll
+++ b/llvm/test/Transforms/SLPVectorizer/SystemZ/minbitwidth-root-trunc.ll
@@ -7,9 +7,9 @@ define void @test(ptr %a, i8 %0, i16 %b.promoted.i) {
 ; CHECK-NEXT:    [[TMP2:%.*]] = zext i8 [[TMP0]] to i128
 ; CHECK-NEXT:    [[TMP3:%.*]] = insertelement <4 x i16> poison, i16 [[B_PROMOTED_I]], i32 0
 ; CHECK-NEXT:    [[TMP4:%.*]] = shufflevector <4 x i16> [[TMP3]], <4 x i16> poison, <4 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP5:%.*]] = insertelement <4 x i128> poison, i128 [[TMP2]], i32 0
-; CHECK-NEXT:    [[TMP6:%.*]] = shufflevector <4 x i128> [[TMP5]], <4 x i128> poison, <4 x i32> zeroinitializer
-; CHECK-NEXT:    [[TMP7:%.*]] = trunc <4 x i128> [[TMP6]] to <4 x i16>
+; CHECK-NEXT:    [[TMP5:%.*]] = trunc i128 [[TMP2]] to i16
+; CHECK-NEXT:    [[TMP6:%.*]] = insertelement <4 x i16> poison, i16 [[TMP5]], i32 0
+; CHECK-NEXT:    [[TMP7:%.*]] = shufflevector <4 x i16> [[TMP6]], <4 x i16> poison, <4 x i32> zeroinitializer
 ; CHECK-NEXT:    [[TMP8:%.*]] = or <4 x i16> [[TMP4]], [[TMP7]]
 ; CHECK-NEXT:    [[TMP9:%.*]] = call i16 @llvm.vector.reduce.and.v4i16(<4 x i16> [[TMP8]])
 ; CHECK-NEXT:    [[TMP11:%.*]] = zext i16 [[TMP9]] to i64
diff --git a/llvm/test/Transforms/SLPVectorizer/X86/alternate-opcode-sindle-bv.ll b/llvm/test/Transforms/SLPVectorizer/X86/alternate-opcode-sindle-bv.ll
new file mode 100644
index 000000000000..89268837c9d8
--- /dev/null
+++ b/llvm/test/Transforms/SLPVectorizer/X86/alternate-opcode-sindle-bv.ll
@@ -0,0 +1,25 @@
+; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --version 4
+; RUN: opt -S --passes=slp-vectorizer -mtriple=x86_64-unknown-linux-gnu < %s | FileCheck %s
+
+define <2 x i32> @test(i32 %arg) {
+; CHECK-LABEL: define <2 x i32> @test(
+; CHECK-SAME: i32 [[ARG:%.*]]) {
+; CHECK-NEXT:  bb:
+; CHECK-NEXT:    [[OR:%.*]] = or i32 [[ARG]], 0
+; CHECK-NEXT:    [[MUL:%.*]] = mul i32 0, 1
+; CHECK-NEXT:    [[MUL1:%.*]] = mul i32 [[OR]], [[MUL]]
+; CHECK-NEXT:    [[CMP:%.*]] = icmp ugt i32 0, [[MUL1]]
+; CHECK-NEXT:    [[TMP0:%.*]] = insertelement <2 x i32> poison, i32 [[OR]], i32 0
+; CHECK-NEXT:    [[TMP1:%.*]] = insertelement <2 x i32> [[TMP0]], i32 [[MUL]], i32 1
+; CHECK-NEXT:    ret <2 x i32> [[TMP1]]
+;
+bb:
+  %or = or i32 %arg, 0
+  %mul = mul i32 0, 1
+  %mul1 = mul i32 %or, %mul
+  %cmp = icmp ugt i32 0, %mul1
+  %0 = insertelement <2 x i32> poison, i32 %or, i32 0
+  %1 = insertelement <2 x i32> %0, i32 %mul, i32 1
+  ret <2 x i32> %1
+}
+
diff --git a/llvm/test/Transforms/SLPVectorizer/X86/gather-node-same-as-vect-but-order.ll b/llvm/test/Transforms/SLPVectorizer/X86/gather-node-same-as-vect-but-order.ll
new file mode 100644
index 000000000000..d80d7b5ecd4e
--- /dev/null
+++ b/llvm/test/Transforms/SLPVectorizer/X86/gather-node-same-as-vect-but-order.ll
@@ -0,0 +1,143 @@
+; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --version 4
+; RUN: opt -S --passes=slp-vectorizer -mtriple=x86_64-unknown-linux-gnu -mcpu=x86-64-v4 < %s | FileCheck %s
+
+%struct.rect = type { float, float, float, float }
+
+define void @foo(ptr %i7, i32 %0, i1 %tobool62.not) {
+; CHECK-LABEL: define void @foo(
+; CHECK-SAME: ptr [[I7:%.*]], i32 [[TMP0:%.*]], i1 [[TOBOOL62_NOT:%.*]]) #[[ATTR0:[0-9]+]] {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[RC21:%.*]] = alloca [0 x [0 x %struct.rect]], i32 0, align 4
+; CHECK-NEXT:    [[TMP1:%.*]] = load float, ptr [[RC21]], align 4
+; CHECK-NEXT:    [[TMP2:%.*]] = insertelement <2 x i32> poison, i32 [[TMP0]], i32 0
+; CHECK-NEXT:    [[TMP3:%.*]] = shufflevector <2 x i32> [[TMP2]], <2 x i32> poison, <2 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP4:%.*]] = sitofp <2 x i32> [[TMP3]] to <2 x float>
+; CHECK-NEXT:    [[TMP5:%.*]] = shufflevector <2 x float> [[TMP4]], <2 x float> poison, <4 x i32> <i32 1, i32 0, i32 1, i32 0>
+; CHECK-NEXT:    [[X1:%.*]] = getelementptr i8, ptr [[RC21]], i64 4
+; CHECK-NEXT:    [[TMP6:%.*]] = load <2 x float>, ptr [[X1]], align 4
+; CHECK-NEXT:    [[TMP7:%.*]] = load float, ptr [[I7]], align 4
+; CHECK-NEXT:    [[TMP8:%.*]] = insertelement <4 x float> poison, float [[TMP1]], i32 0
+; CHECK-NEXT:    [[TMP9:%.*]] = shufflevector <2 x float> [[TMP6]], <2 x float> poison, <4 x i32> <i32 0, i32 1, i32 poison, i32 poison>
+; CHECK-NEXT:    [[TMP10:%.*]] = shufflevector <4 x float> [[TMP8]], <4 x float> [[TMP9]], <4 x i32> <i32 0, i32 4, i32 5, i32 poison>
+; CHECK-NEXT:    [[TMP11:%.*]] = insertelement <4 x float> [[TMP10]], float [[TMP7]], i32 3
+; CHECK-NEXT:    [[TMP12:%.*]] = fcmp olt <4 x float> [[TMP11]], zeroinitializer
+; CHECK-NEXT:    [[TMP14:%.*]] = fcmp olt <4 x float> [[TMP5]], zeroinitializer
+; CHECK-NEXT:    [[TMP15:%.*]] = select <4 x i1> [[TMP14]], <4 x float> [[TMP5]], <4 x float> zeroinitializer
+; CHECK-NEXT:    [[TMP16:%.*]] = select <4 x i1> [[TMP12]], <4 x float> zeroinitializer, <4 x float> [[TMP15]]
+; CHECK-NEXT:    store <4 x float> [[TMP16]], ptr [[RC21]], align 4
+; CHECK-NEXT:    br label [[IF_END:%.*]]
+; CHECK:       entry.if.end72_crit_edge:
+; CHECK-NEXT:    br label [[IF_END72:%.*]]
+; CHECK:       if.then63:
+; CHECK-NEXT:    br label [[IF_END]]
+; CHECK:       if.end:
+; CHECK-NEXT:    [[TMP17:%.*]] = phi <4 x float> [ poison, [[IF_THEN63:%.*]] ], [ [[TMP16]], [[ENTRY:%.*]] ]
+; CHECK-NEXT:    [[TMP18:%.*]] = call <4 x float> @llvm.round.v4f32(<4 x float> [[TMP17]])
+; CHECK-NEXT:    [[TMP19:%.*]] = fptosi <4 x float> [[TMP18]] to <4 x i32>
+; CHECK-NEXT:    br label [[IF_END72]]
+; CHECK:       if.end72:
+; CHECK-NEXT:    [[TMP20:%.*]] = phi <4 x i32> [ poison, [[ENTRY_IF_END72_CRIT_EDGE:%.*]] ], [ [[TMP19]], [[IF_END]] ]
+; CHECK-NEXT:    [[TMP21:%.*]] = shufflevector <4 x i32> [[TMP20]], <4 x i32> poison, <4 x i32> <i32 1, i32 0, i32 3, i32 2>
+; CHECK-NEXT:    br i1 [[TOBOOL62_NOT]], label [[IF_END75:%.*]], label [[IF_THEN74:%.*]]
+; CHECK:       if.then74:
+; CHECK-NEXT:    br label [[IF_END75]]
+; CHECK:       if.end75:
+; CHECK-NEXT:    [[TMP22:%.*]] = phi <4 x i32> [ [[TMP20]], [[IF_THEN74]] ], [ [[TMP21]], [[IF_END72]] ]
+; CHECK-NEXT:    [[TMP23:%.*]] = or <4 x i32> [[TMP22]], <i32 1, i32 1, i32 1, i32 1>
+; CHECK-NEXT:    [[TMP24:%.*]] = shufflevector <2 x i32> [[TMP3]], <2 x i32> poison, <4 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP25:%.*]] = mul <4 x i32> [[TMP23]], [[TMP24]]
+; CHECK-NEXT:    [[TMP26:%.*]] = sitofp <4 x i32> [[TMP25]] to <4 x float>
+; CHECK-NEXT:    [[TMP27:%.*]] = shufflevector <4 x float> [[TMP26]], <4 x float> poison, <4 x i32> <i32 1, i32 0, i32 3, i32 2>
+; CHECK-NEXT:    store <4 x float> [[TMP27]], ptr [[RC21]], align 4
+; CHECK-NEXT:    ret void
+;
+entry:
+  %rc21 = alloca [0 x [0 x %struct.rect]], i32 0, align 4
+  %1 = load float, ptr %rc21, align 4
+  %cmp = fcmp olt float %1, 0.000000e+00
+  %conv = sitofp i32 %0 to float
+  %cmp2 = fcmp olt float %conv, 0.000000e+00
+  %cond = select i1 %cmp2, float %conv, float 0.000000e+00
+  %cond9 = select i1 %cmp, float 0.000000e+00, float %cond
+  store float %cond9, ptr %rc21, align 4
+  %x1 = getelementptr i8, ptr %rc21, i64 4
+  %2 = load float, ptr %x1, align 4
+  %cmp11 = fcmp olt float %2, 0.000000e+00
+  %conv16 = sitofp i32 %0 to float
+  %cmp17 = fcmp olt float %conv16, 0.000000e+00
+  %cond24 = select i1 %cmp17, float %conv16, float 0.000000e+00
+  %cond26 = select i1 %cmp11, float 0.000000e+00, float %cond24
+  store float %cond26, ptr %x1, align 4
+  %y0 = getelementptr i8, ptr %rc21, i64 8
+  %3 = load float, ptr %y0, align 4
+  %cmp28 = fcmp olt float %3, 0.000000e+00
+  %cmp34 = fcmp olt float %conv, 0.000000e+00
+  %cond41 = select i1 %cmp34, float %conv, float 0.000000e+00
+  %cond43 = select i1 %cmp28, float 0.000000e+00, float %cond41
+  store float %cond43, ptr %y0, align 4
+  %y11 = getelementptr i8, ptr %rc21, i64 12
+  %4 = load float, ptr %i7, align 4
+  %cmp45 = fcmp olt float %4, 0.000000e+00
+  %cmp51 = fcmp olt float %conv16, 0.000000e+00
+  %cond58 = select i1 %cmp51, float %conv16, float 0.000000e+00
+  %cond60 = select i1 %cmp45, float 0.000000e+00, float %cond58
+  store float %cond60, ptr %y11, align 4
+  br label %if.end
+
+entry.if.end72_crit_edge:
+  br label %if.end72
+
+if.then63:
+  br label %if.end
+
+if.end:
+  %5 = phi float [ 0.000000e+00, %if.then63 ], [ %cond60, %entry ]
+  %6 = phi float [ 0.000000e+00, %if.then63 ], [ %cond26, %entry ]
+  %7 = phi float [ 0.000000e+00, %if.then63 ], [ %cond43, %entry ]
+  %8 = phi float [ 0.000000e+00, %if.then63 ], [ %cond9, %entry ]
+  %9 = call float @llvm.round.f32(float %8)
+  %conv65 = fptosi float %9 to i32
+  %10 = call float @llvm.round.f32(float %7)
+  %conv67 = fptosi float %10 to i32
+  %11 = call float @llvm.round.f32(float %6)
+  %conv69 = fptosi float %11 to i32
+  %12 = call float @llvm.round.f32(float %5)
+  %conv71 = fptosi float %12 to i32
+  br label %if.end72
+
+if.end72:
+  %.pre100 = phi i32 [ 0, %entry.if.end72_crit_edge ], [ %conv71, %if.end ]
+  %.pre99 = phi i32 [ 0, %entry.if.end72_crit_edge ], [ %conv67, %if.end ]
+  %.pre98 = phi i32 [ 0, %entry.if.end72_crit_edge ], [ %conv69, %if.end ]
+  %.pre97 = phi i32 [ 0, %entry.if.end72_crit_edge ], [ %conv65, %if.end ]
+  br i1 %tobool62.not, label %if.end75, label %if.then74
+
+if.then74:
+  br label %if.end75
+
+if.end75:
+  %13 = phi i32 [ %.pre99, %if.then74 ], [ %.pre100, %if.end72 ]
+  %14 = phi i32 [ %.pre100, %if.then74 ], [ %.pre99, %if.end72 ]
+  %15 = phi i32 [ %.pre97, %if.then74 ], [ %.pre98, %if.end72 ]
+  %16 = phi i32 [ %.pre98, %if.then74 ], [ %.pre97, %if.end72 ]
+  %sub = or i32 %16, 1
+  %mul = mul i32 %sub, %0
+  %conv77 = sitofp i32 %mul to float
+  store float %conv77, ptr %rc21, align 4
+  %x178 = getelementptr i8, ptr %rc21, i64 4
+  %sub79 = or i32 %15, 1
+  %mul80 = mul i32 %sub79, %0
+  %conv81 = sitofp i32 %mul80 to float
+  store float %conv81, ptr %x178, align 4
+  %y082 = getelementptr i8, ptr %rc21, i64 8
+  %sub83 = or i32 %14, 1
+  %mul84 = mul i32 %sub83, %0
+  %conv85 = sitofp i32 %mul84 to float
+  store float %conv85, ptr %y082, align 4
+  %y186 = getelementptr i8, ptr %rc21, i64 12
+  %sub87 = or i32 %13, 1
+  %mul88 = mul i32 %sub87, %0
+  %conv89 = sitofp i32 %mul88 to float
+  store float %conv89, ptr %y186, align 4
+  ret void
+}
diff --git a/llvm/test/Transforms/SLPVectorizer/X86/gep-with-extractelement-many-users.ll b/llvm/test/Transforms/SLPVectorizer/X86/gep-with-extractelement-many-users.ll
new file mode 100644
index 000000000000..37d166953c33
--- /dev/null
+++ b/llvm/test/Transforms/SLPVectorizer/X86/gep-with-extractelement-many-users.ll
@@ -0,0 +1,26 @@
+; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --version 4
+; RUN: opt -S --passes=slp-vectorizer -mtriple=x86_64-unknown-linux-gnu -slp-threshold=-99999 < %s | FileCheck %s
+
+define void @test() {
+; CHECK-LABEL: define void @test() {
+; CHECK-NEXT:  bb:
+; CHECK-NEXT:    [[TMP0:%.*]] = extractelement <2 x ptr> zeroinitializer, i32 0
+; CHECK-NEXT:    [[GETELEMENTPTR6:%.*]] = getelementptr i8, ptr [[TMP0]], i64 872
+; CHECK-NEXT:    store double 0.000000e+00, ptr [[GETELEMENTPTR6]], align 8
+; CHECK-NEXT:    br label [[BB9:%.*]]
+; CHECK:       bb9:
+; CHECK-NEXT:    [[TMP1:%.*]] = phi <2 x ptr> [ getelementptr (i8, <2 x ptr> zeroinitializer, <2 x i64> <i64 32, i64 872>), [[BB:%.*]] ]
+; CHECK-NEXT:    ret void
+;
+bb:
+  %getelementptr = getelementptr i8, ptr null, i64 32
+  %0 = extractelement <2 x ptr> zeroinitializer, i32 0
+  %getelementptr6 = getelementptr i8, ptr %0, i64 872
+  store double 0.000000e+00, ptr %getelementptr6, align 8
+  br label %bb9
+
+bb9:
+  %phi10 = phi ptr [ %getelementptr, %bb ]
+  %phi11 = phi ptr [ %getelementptr6, %bb ]
+  ret void
+}
diff --git a/llvm/test/Transforms/SLPVectorizer/X86/minbitwidth-node-with-multi-users.ll b/llvm/test/Transforms/SLPVectorizer/X86/minbitwidth-node-with-multi-users.ll
index 668d3c3c8c82..0ab56279fe47 100644
--- a/llvm/test/Transforms/SLPVectorizer/X86/minbitwidth-node-with-multi-users.ll
+++ b/llvm/test/Transforms/SLPVectorizer/X86/minbitwidth-node-with-multi-users.ll
@@ -16,8 +16,7 @@ define void @test() {
 ; CHECK-NEXT:    [[TMP9:%.*]] = trunc <4 x i8> [[TMP8]] to <4 x i1>
 ; CHECK-NEXT:    [[TMP10:%.*]] = or <4 x i1> zeroinitializer, [[TMP15]]
 ; CHECK-NEXT:    [[TMP11:%.*]] = icmp eq <4 x i1> [[TMP9]], [[TMP10]]
-; CHECK-NEXT:    [[TMP16:%.*]] = shufflevector <4 x i1> [[TMP15]], <4 x i1> poison, <4 x i32> <i32 0, i32 1, i32 2, i32 3>
-; CHECK-NEXT:    [[TMP6:%.*]] = zext <4 x i1> [[TMP16]] to <4 x i32>
+; CHECK-NEXT:    [[TMP6:%.*]] = zext <4 x i1> [[TMP15]] to <4 x i32>
 ; CHECK-NEXT:    [[TMP12:%.*]] = shufflevector <4 x i32> [[TMP6]], <4 x i32> <i32 0, i32 0, i32 poison, i32 0>, <4 x i32> <i32 4, i32 5, i32 2, i32 7>
 ; CHECK-NEXT:    [[TMP13:%.*]] = select <4 x i1> [[TMP11]], <4 x i32> [[TMP12]], <4 x i32> zeroinitializer
 ; CHECK-NEXT:    [[TMP14:%.*]] = call i32 @llvm.vector.reduce.and.v4i32(<4 x i32> [[TMP13]])
diff --git a/llvm/test/Transforms/SLPVectorizer/X86/pr46983.ll b/llvm/test/Transforms/SLPVectorizer/X86/pr46983.ll
index 75505f632a43..29021150ccd2 100644
--- a/llvm/test/Transforms/SLPVectorizer/X86/pr46983.ll
+++ b/llvm/test/Transforms/SLPVectorizer/X86/pr46983.ll
@@ -1,9 +1,9 @@
 ; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
-; RUN: opt < %s -passes=slp-vectorizer,instcombine -S -mtriple=x86_64-unknown-linux-gnu -mattr=+sse2   | FileCheck %s --check-prefixes=CHECK,SSE
-; RUN: opt < %s -passes=slp-vectorizer,instcombine -S -mtriple=x86_64-unknown-linux-gnu -mattr=+sse4.2 | FileCheck %s --check-prefixes=CHECK,SSE
-; RUN: opt < %s -passes=slp-vectorizer,instcombine -S -mtriple=x86_64-unknown-linux-gnu -mattr=+avx    | FileCheck %s --check-prefixes=CHECK,AVX
-; RUN: opt < %s -passes=slp-vectorizer,instcombine -S -mtriple=x86_64-unknown-linux-gnu -mattr=+avx2   | FileCheck %s --check-prefixes=CHECK,AVX
-; RUN: opt < %s -passes=slp-vectorizer,instcombine -S -mtriple=x86_64-unknown-linux-gnu -mattr=+avx512bw,+avx512vl | FileCheck %s --check-prefixes=CHECK,AVX
+; RUN: opt < %s -passes=slp-vectorizer,instcombine -S -mtriple=x86_64-unknown-linux-gnu -mattr=+sse2   -slp-threshold=-1 | FileCheck %s
+; RUN: opt < %s -passes=slp-vectorizer,instcombine -S -mtriple=x86_64-unknown-linux-gnu -mattr=+sse4.2 | FileCheck %s
+; RUN: opt < %s -passes=slp-vectorizer,instcombine -S -mtriple=x86_64-unknown-linux-gnu -mattr=+avx    | FileCheck %s
+; RUN: opt < %s -passes=slp-vectorizer,instcombine -S -mtriple=x86_64-unknown-linux-gnu -mattr=+avx2   | FileCheck %s
+; RUN: opt < %s -passes=slp-vectorizer,instcombine -S -mtriple=x86_64-unknown-linux-gnu -mattr=+avx512bw,+avx512vl | FileCheck %s
 
 define void @store_i32(ptr nocapture %0, i32 %1, i32 %2) {
 ; CHECK-LABEL: @store_i32(
@@ -98,58 +98,19 @@ define void @store_i8(ptr nocapture %0, i32 %1, i32 %2) {
 }
 
 define void @store_i64(ptr nocapture %0, i32 %1, i32 %2) {
-; SSE-LABEL: @store_i64(
-; SSE-NEXT:    [[TMP4:%.*]] = zext i32 [[TMP1:%.*]] to i64
-; SSE-NEXT:    [[TMP5:%.*]] = load i64, ptr [[TMP0:%.*]], align 8, !tbaa [[TBAA5:![0-9]+]]
-; SSE-NEXT:    [[TMP6:%.*]] = mul i64 [[TMP5]], [[TMP4]]
-; SSE-NEXT:    [[TMP7:%.*]] = lshr i64 [[TMP6]], 15
-; SSE-NEXT:    [[TMP8:%.*]] = trunc i64 [[TMP7]] to i32
-; SSE-NEXT:    [[TMP9:%.*]] = icmp ult i32 [[TMP8]], 255
-; SSE-NEXT:    [[TMP10:%.*]] = and i64 [[TMP7]], 4294967295
-; SSE-NEXT:    [[TMP11:%.*]] = select i1 [[TMP9]], i64 [[TMP10]], i64 255
-; SSE-NEXT:    store i64 [[TMP11]], ptr [[TMP0]], align 8, !tbaa [[TBAA5]]
-; SSE-NEXT:    [[TMP12:%.*]] = getelementptr inbounds i8, ptr [[TMP0]], i64 8
-; SSE-NEXT:    [[TMP13:%.*]] = load i64, ptr [[TMP12]], align 8, !tbaa [[TBAA5]]
-; SSE-NEXT:    [[TMP14:%.*]] = mul i64 [[TMP13]], [[TMP4]]
-; SSE-NEXT:    [[TMP15:%.*]] = lshr i64 [[TMP14]], 15
-; SSE-NEXT:    [[TMP16:%.*]] = trunc i64 [[TMP15]] to i32
-; SSE-NEXT:    [[TMP17:%.*]] = icmp ult i32 [[TMP16]], 255
-; SSE-NEXT:    [[TMP18:%.*]] = and i64 [[TMP15]], 4294967295
-; SSE-NEXT:    [[TMP19:%.*]] = select i1 [[TMP17]], i64 [[TMP18]], i64 255
-; SSE-NEXT:    store i64 [[TMP19]], ptr [[TMP12]], align 8, !tbaa [[TBAA5]]
-; SSE-NEXT:    [[TMP20:%.*]] = getelementptr inbounds i8, ptr [[TMP0]], i64 16
-; SSE-NEXT:    [[TMP21:%.*]] = load i64, ptr [[TMP20]], align 8, !tbaa [[TBAA5]]
-; SSE-NEXT:    [[TMP22:%.*]] = mul i64 [[TMP21]], [[TMP4]]
-; SSE-NEXT:    [[TMP23:%.*]] = lshr i64 [[TMP22]], 15
-; SSE-NEXT:    [[TMP24:%.*]] = trunc i64 [[TMP23]] to i32
-; SSE-NEXT:    [[TMP25:%.*]] = icmp ult i32 [[TMP24]], 255
-; SSE-NEXT:    [[TMP26:%.*]] = and i64 [[TMP23]], 4294967295
-; SSE-NEXT:    [[TMP27:%.*]] = select i1 [[TMP25]], i64 [[TMP26]], i64 255
-; SSE-NEXT:    store i64 [[TMP27]], ptr [[TMP20]], align 8, !tbaa [[TBAA5]]
-; SSE-NEXT:    [[TMP28:%.*]] = getelementptr inbounds i8, ptr [[TMP0]], i64 24
-; SSE-NEXT:    [[TMP29:%.*]] = load i64, ptr [[TMP28]], align 8, !tbaa [[TBAA5]]
-; SSE-NEXT:    [[TMP30:%.*]] = mul i64 [[TMP29]], [[TMP4]]
-; SSE-NEXT:    [[TMP31:%.*]] = lshr i64 [[TMP30]], 15
-; SSE-NEXT:    [[TMP32:%.*]] = trunc i64 [[TMP31]] to i32
-; SSE-NEXT:    [[TMP33:%.*]] = icmp ult i32 [[TMP32]], 255
-; SSE-NEXT:    [[TMP34:%.*]] = and i64 [[TMP31]], 4294967295
-; SSE-NEXT:    [[TMP35:%.*]] = select i1 [[TMP33]], i64 [[TMP34]], i64 255
-; SSE-NEXT:    store i64 [[TMP35]], ptr [[TMP28]], align 8, !tbaa [[TBAA5]]
-; SSE-NEXT:    ret void
-;
-; AVX-LABEL: @store_i64(
-; AVX-NEXT:    [[TMP4:%.*]] = zext i32 [[TMP1:%.*]] to i64
-; AVX-NEXT:    [[TMP5:%.*]] = load <4 x i64>, ptr [[TMP0:%.*]], align 8, !tbaa [[TBAA5:![0-9]+]]
-; AVX-NEXT:    [[TMP6:%.*]] = insertelement <4 x i64> poison, i64 [[TMP4]], i64 0
-; AVX-NEXT:    [[TMP7:%.*]] = shufflevector <4 x i64> [[TMP6]], <4 x i64> poison, <4 x i32> zeroinitializer
-; AVX-NEXT:    [[TMP8:%.*]] = mul <4 x i64> [[TMP5]], [[TMP7]]
-; AVX-NEXT:    [[TMP9:%.*]] = lshr <4 x i64> [[TMP8]], <i64 15, i64 15, i64 15, i64 15>
-; AVX-NEXT:    [[TMP10:%.*]] = trunc <4 x i64> [[TMP9]] to <4 x i32>
-; AVX-NEXT:    [[TMP11:%.*]] = icmp ult <4 x i32> [[TMP10]], <i32 255, i32 255, i32 255, i32 255>
-; AVX-NEXT:    [[TMP12:%.*]] = and <4 x i64> [[TMP9]], <i64 4294967295, i64 4294967295, i64 4294967295, i64 4294967295>
-; AVX-NEXT:    [[TMP13:%.*]] = select <4 x i1> [[TMP11]], <4 x i64> [[TMP12]], <4 x i64> <i64 255, i64 255, i64 255, i64 255>
-; AVX-NEXT:    store <4 x i64> [[TMP13]], ptr [[TMP0]], align 8, !tbaa [[TBAA5]]
-; AVX-NEXT:    ret void
+; CHECK-LABEL: @store_i64(
+; CHECK-NEXT:    [[TMP4:%.*]] = zext i32 [[TMP1:%.*]] to i64
+; CHECK-NEXT:    [[TMP5:%.*]] = load <4 x i64>, ptr [[TMP0:%.*]], align 8, !tbaa [[TBAA5:![0-9]+]]
+; CHECK-NEXT:    [[TMP6:%.*]] = insertelement <4 x i64> poison, i64 [[TMP4]], i64 0
+; CHECK-NEXT:    [[TMP7:%.*]] = shufflevector <4 x i64> [[TMP6]], <4 x i64> poison, <4 x i32> zeroinitializer
+; CHECK-NEXT:    [[TMP8:%.*]] = mul <4 x i64> [[TMP5]], [[TMP7]]
+; CHECK-NEXT:    [[TMP9:%.*]] = lshr <4 x i64> [[TMP8]], <i64 15, i64 15, i64 15, i64 15>
+; CHECK-NEXT:    [[TMP10:%.*]] = trunc <4 x i64> [[TMP9]] to <4 x i32>
+; CHECK-NEXT:    [[TMP11:%.*]] = icmp ult <4 x i32> [[TMP10]], <i32 255, i32 255, i32 255, i32 255>
+; CHECK-NEXT:    [[TMP12:%.*]] = and <4 x i64> [[TMP9]], <i64 4294967295, i64 4294967295, i64 4294967295, i64 4294967295>
+; CHECK-NEXT:    [[TMP13:%.*]] = select <4 x i1> [[TMP11]], <4 x i64> [[TMP12]], <4 x i64> <i64 255, i64 255, i64 255, i64 255>
+; CHECK-NEXT:    store <4 x i64> [[TMP13]], ptr [[TMP0]], align 8, !tbaa [[TBAA5]]
+; CHECK-NEXT:    ret void
 ;
   %4 = zext i32 %1 to i64
   %5 = load i64, ptr %0, align 8, !tbaa !7
diff --git a/llvm/test/Transforms/SampleProfile/pseudo-probe-profile.ll b/llvm/test/Transforms/SampleProfile/pseudo-probe-profile.ll
index 867a49dbaed2..7258ffca1278 100644
--- a/llvm/test/Transforms/SampleProfile/pseudo-probe-profile.ll
+++ b/llvm/test/Transforms/SampleProfile/pseudo-probe-profile.ll
@@ -1,5 +1,9 @@
-; RUN: opt < %s -passes=pseudo-probe,sample-profile -sample-profile-file=%S/Inputs/pseudo-probe-profile.prof -pass-remarks=sample-profile -pass-remarks-output=%t.opt.yaml -sample-profile-use-profi=0 -S | FileCheck %s
-; RUN: FileCheck %s -check-prefix=YAML < %t.opt.yaml
+; RUN: opt < %s -passes=pseudo-probe,sample-profile -sample-profile-file=%S/Inputs/pseudo-probe-profile.prof -pass-remarks=sample-profile -pass-remarks-output=%t.opt.yaml -sample-profile-use-profi=0 -S -o %t
+; RUN: FileCheck %s --input-file %t
+; RUN: FileCheck %s -check-prefix=YAML --input-file %t.opt.yaml
+; RUN: opt < %t -passes=sample-profile -sample-profile-file=%S/Inputs/pseudo-probe-profile.prof -sample-profile-remove-probe -S | FileCheck %s -check-prefix=REMOVE-PROBE
+
+; REMOVE-PROBE-NOT: call void @llvm.pseudoprobe
 
 define dso_local i32 @foo(i32 %x, ptr %f) #0 !dbg !4 {
 entry:
diff --git a/llvm/test/Transforms/SimpleLoopUnswitch/endless-unswitch.ll b/llvm/test/Transforms/SimpleLoopUnswitch/endless-unswitch.ll
index 0d3aa8b24310..e70bea2d2f7a 100644
--- a/llvm/test/Transforms/SimpleLoopUnswitch/endless-unswitch.ll
+++ b/llvm/test/Transforms/SimpleLoopUnswitch/endless-unswitch.ll
@@ -106,3 +106,96 @@ for.inc:                                          ; preds = %for.cond5
   store i8 0, ptr @b, align 1
   br label %for.cond5
 }
+
+define void @e(ptr %p) {
+; CHECK-LABEL: @e(
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    br label [[FOR_COND:%.*]]
+; CHECK:       for.cond:
+; CHECK-NEXT:    br i1 false, label [[FOR_END:%.*]], label [[FOR_COND]]
+; CHECK:       for.end:
+; CHECK-NEXT:    [[TMP0:%.*]] = load i16, ptr [[P:%.*]], align 2
+; CHECK-NEXT:    [[TMP1:%.*]] = trunc i16 [[TMP0]] to i1
+; CHECK-NEXT:    br i1 [[TMP1]], label [[FOR_END_SPLIT:%.*]], label [[FOR_END_SPLIT_US:%.*]]
+; CHECK:       for.end.split.us:
+; CHECK-NEXT:    br label [[G_US:%.*]]
+; CHECK:       g.us:
+; CHECK-NEXT:    br label [[G_SPLIT_US6:%.*]]
+; CHECK:       for.cond1.us1:
+; CHECK-NEXT:    [[TMP2:%.*]] = load i16, ptr [[P]], align 2
+; CHECK-NEXT:    [[TOBOOL4_NOT_US:%.*]] = trunc i16 [[TMP2]] to i1
+; CHECK-NEXT:    br i1 [[TOBOOL4_NOT_US]], label [[FOR_COND5_PREHEADER_US4:%.*]], label [[G_LOOPEXIT_US:%.*]]
+; CHECK:       for.cond5.us2:
+; CHECK-NEXT:    br i1 false, label [[FOR_COND1_LOOPEXIT_US5:%.*]], label [[FOR_INC_US3:%.*]]
+; CHECK:       for.inc.us3:
+; CHECK-NEXT:    store i8 0, ptr @b, align 1
+; CHECK-NEXT:    br label [[FOR_COND5_US2:%.*]]
+; CHECK:       for.cond5.preheader.us4:
+; CHECK-NEXT:    br label [[FOR_COND5_US2]]
+; CHECK:       for.cond1.loopexit.us5:
+; CHECK-NEXT:    br label [[FOR_COND1_US1:%.*]], !llvm.loop [[LOOP3:![0-9]+]]
+; CHECK:       g.loopexit.us:
+; CHECK-NEXT:    br label [[G_US]]
+; CHECK:       g.split.us6:
+; CHECK-NEXT:    br label [[FOR_COND1_US1]]
+; CHECK:       for.end.split:
+; CHECK-NEXT:    br label [[G:%.*]]
+; CHECK:       g.loopexit:
+; CHECK-NEXT:    br label [[G]], !llvm.loop [[LOOP4:![0-9]+]]
+; CHECK:       g:
+; CHECK-NEXT:    [[TMP3:%.*]] = load i16, ptr [[P]], align 2
+; CHECK-NEXT:    [[TMP4:%.*]] = trunc i16 [[TMP3]] to i1
+; CHECK-NEXT:    br i1 [[TMP4]], label [[G_SPLIT_US:%.*]], label [[G_SPLIT:%.*]]
+; CHECK:       g.split.us:
+; CHECK-NEXT:    br label [[FOR_COND1_US:%.*]]
+; CHECK:       for.cond1.us:
+; CHECK-NEXT:    br label [[FOR_COND5_PREHEADER_US:%.*]]
+; CHECK:       for.cond5.us:
+; CHECK-NEXT:    br i1 false, label [[FOR_COND1_LOOPEXIT_US:%.*]], label [[FOR_INC_US:%.*]]
+; CHECK:       for.inc.us:
+; CHECK-NEXT:    store i8 0, ptr @b, align 1
+; CHECK-NEXT:    br label [[FOR_COND5_US:%.*]]
+; CHECK:       for.cond5.preheader.us:
+; CHECK-NEXT:    br label [[FOR_COND5_US]]
+; CHECK:       for.cond1.loopexit.us:
+; CHECK-NEXT:    br label [[FOR_COND1_US]]
+; CHECK:       g.split:
+; CHECK-NEXT:    br label [[FOR_COND1:%.*]]
+; CHECK:       for.cond1.loopexit:
+; CHECK-NEXT:    br label [[FOR_COND1]], !llvm.loop [[LOOP3]]
+; CHECK:       for.cond1:
+; CHECK-NEXT:    [[TMP5:%.*]] = load i16, ptr [[P]], align 2
+; CHECK-NEXT:    [[TOBOOL4_NOT:%.*]] = trunc i16 [[TMP5]] to i1
+; CHECK-NEXT:    br i1 [[TOBOOL4_NOT]], label [[FOR_COND5_PREHEADER:%.*]], label [[G_LOOPEXIT:%.*]]
+; CHECK:       for.cond5.preheader:
+; CHECK-NEXT:    br label [[FOR_COND5:%.*]]
+; CHECK:       for.cond5:
+; CHECK-NEXT:    br i1 false, label [[FOR_COND1_LOOPEXIT:%.*]], label [[FOR_INC:%.*]]
+; CHECK:       for.inc:
+; CHECK-NEXT:    store i8 0, ptr @b, align 1
+; CHECK-NEXT:    br label [[FOR_COND5]]
+;
+entry:
+  br label %for.cond
+
+for.cond:                                         ; preds = %for.cond, %entry
+  br i1 false, label %for.end, label %for.cond
+
+for.end:                                          ; preds = %for.cond
+  br label %g
+
+g:                                                ; preds = %for.cond1, %for.end
+  br label %for.cond1
+
+for.cond1:                                        ; preds = %for.cond5, %g
+  %0 = load i16, ptr %p, align 2
+  %tobool4.not = trunc i16 %0 to i1
+  br i1 %tobool4.not, label %for.cond5, label %g
+
+for.cond5:                                        ; preds = %for.inc, %for.cond1
+  br i1 false, label %for.cond1, label %for.inc
+
+for.inc:                                          ; preds = %for.cond5
+  store i8 0, ptr @b, align 1
+  br label %for.cond5
+}
diff --git a/llvm/test/Transforms/SimpleLoopUnswitch/partial-unswitch.ll b/llvm/test/Transforms/SimpleLoopUnswitch/partial-unswitch.ll
index f97e5c3eec9d..1d8942079ffd 100644
--- a/llvm/test/Transforms/SimpleLoopUnswitch/partial-unswitch.ll
+++ b/llvm/test/Transforms/SimpleLoopUnswitch/partial-unswitch.ll
@@ -1326,6 +1326,136 @@ exit:
   ret i32 10
 }
 
+define i32 @partial_unswitch_true_successor_trunc(ptr %ptr, i32 %N) {
+; CHECK-LABEL: @partial_unswitch_true_successor_trunc(
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP0:%.*]] = load i32, ptr [[PTR:%.*]], align 4
+; CHECK-NEXT:    [[TMP1:%.*]] = trunc i32 [[TMP0]] to i1
+; CHECK-NEXT:    br i1 [[TMP1]], label [[ENTRY_SPLIT_US:%.*]], label [[ENTRY_SPLIT:%.*]]
+; CHECK:       entry.split.us:
+; CHECK-NEXT:    br label [[LOOP_HEADER_US:%.*]]
+; CHECK:       loop.header.us:
+; CHECK-NEXT:    [[IV_US:%.*]] = phi i32 [ 0, [[ENTRY_SPLIT_US]] ], [ [[IV_NEXT_US:%.*]], [[LOOP_LATCH_US:%.*]] ]
+; CHECK-NEXT:    br label [[NOCLOBBER_US:%.*]]
+; CHECK:       noclobber.us:
+; CHECK-NEXT:    br label [[LOOP_LATCH_US]]
+; CHECK:       loop.latch.us:
+; CHECK-NEXT:    [[C_US:%.*]] = icmp ult i32 [[IV_US]], [[N:%.*]]
+; CHECK-NEXT:    [[IV_NEXT_US]] = add i32 [[IV_US]], 1
+; CHECK-NEXT:    br i1 [[C_US]], label [[LOOP_HEADER_US]], label [[EXIT_SPLIT_US:%.*]]
+; CHECK:       exit.split.us:
+; CHECK-NEXT:    br label [[EXIT:%.*]]
+; CHECK:       entry.split:
+; CHECK-NEXT:    br label [[LOOP_HEADER:%.*]]
+; CHECK:       loop.header:
+; CHECK-NEXT:    [[IV:%.*]] = phi i32 [ 0, [[ENTRY_SPLIT]] ], [ [[IV_NEXT:%.*]], [[LOOP_LATCH:%.*]] ]
+; CHECK-NEXT:    [[LV:%.*]] = load i32, ptr [[PTR]], align 4
+; CHECK-NEXT:    [[SC:%.*]] = trunc i32 [[LV]] to i1
+; CHECK-NEXT:    br i1 [[SC]], label [[NOCLOBBER:%.*]], label [[CLOBBER:%.*]]
+; CHECK:       noclobber:
+; CHECK-NEXT:    br label [[LOOP_LATCH]]
+; CHECK:       clobber:
+; CHECK-NEXT:    call void @clobber()
+; CHECK-NEXT:    br label [[LOOP_LATCH]]
+; CHECK:       loop.latch:
+; CHECK-NEXT:    [[C:%.*]] = icmp ult i32 [[IV]], [[N]]
+; CHECK-NEXT:    [[IV_NEXT]] = add i32 [[IV]], 1
+; CHECK-NEXT:    br i1 [[C]], label [[LOOP_HEADER]], label [[EXIT_SPLIT:%.*]], !llvm.loop [[LOOP12:![0-9]+]]
+; CHECK:       exit.split:
+; CHECK-NEXT:    br label [[EXIT]]
+; CHECK:       exit:
+; CHECK-NEXT:    ret i32 10
+;
+entry:
+  br label %loop.header
+
+loop.header:
+  %iv = phi i32 [ 0, %entry ], [ %iv.next, %loop.latch ]
+  %lv = load i32, ptr %ptr
+  %sc = trunc i32 %lv to i1
+  br i1 %sc, label %noclobber, label %clobber
+
+noclobber:
+  br label %loop.latch
+
+clobber:
+  call void @clobber()
+  br label %loop.latch
+
+loop.latch:
+  %c = icmp ult i32 %iv, %N
+  %iv.next = add i32 %iv, 1
+  br i1 %c, label %loop.header, label %exit
+
+exit:
+  ret i32 10
+}
+
+define i32 @partial_unswitch_false_successor_trunc(ptr %ptr, i32 %N) {
+; CHECK-LABEL: @partial_unswitch_false_successor_trunc(
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP0:%.*]] = load i32, ptr [[PTR:%.*]], align 4
+; CHECK-NEXT:    [[TMP1:%.*]] = trunc i32 [[TMP0]] to i1
+; CHECK-NEXT:    br i1 [[TMP1]], label [[ENTRY_SPLIT:%.*]], label [[ENTRY_SPLIT_US:%.*]]
+; CHECK:       entry.split.us:
+; CHECK-NEXT:    br label [[LOOP_HEADER_US:%.*]]
+; CHECK:       loop.header.us:
+; CHECK-NEXT:    [[IV_US:%.*]] = phi i32 [ 0, [[ENTRY_SPLIT_US]] ], [ [[IV_NEXT_US:%.*]], [[LOOP_LATCH_US:%.*]] ]
+; CHECK-NEXT:    br label [[NOCLOBBER_US:%.*]]
+; CHECK:       noclobber.us:
+; CHECK-NEXT:    br label [[LOOP_LATCH_US]]
+; CHECK:       loop.latch.us:
+; CHECK-NEXT:    [[C_US:%.*]] = icmp ult i32 [[IV_US]], [[N:%.*]]
+; CHECK-NEXT:    [[IV_NEXT_US]] = add i32 [[IV_US]], 1
+; CHECK-NEXT:    br i1 [[C_US]], label [[LOOP_HEADER_US]], label [[EXIT_SPLIT_US:%.*]]
+; CHECK:       exit.split.us:
+; CHECK-NEXT:    br label [[EXIT:%.*]]
+; CHECK:       entry.split:
+; CHECK-NEXT:    br label [[LOOP_HEADER:%.*]]
+; CHECK:       loop.header:
+; CHECK-NEXT:    [[IV:%.*]] = phi i32 [ 0, [[ENTRY_SPLIT]] ], [ [[IV_NEXT:%.*]], [[LOOP_LATCH:%.*]] ]
+; CHECK-NEXT:    [[LV:%.*]] = load i32, ptr [[PTR]], align 4
+; CHECK-NEXT:    [[SC:%.*]] = trunc i32 [[LV]] to i1
+; CHECK-NEXT:    br i1 [[SC]], label [[CLOBBER:%.*]], label [[NOCLOBBER:%.*]]
+; CHECK:       clobber:
+; CHECK-NEXT:    call void @clobber()
+; CHECK-NEXT:    br label [[LOOP_LATCH]]
+; CHECK:       noclobber:
+; CHECK-NEXT:    br label [[LOOP_LATCH]]
+; CHECK:       loop.latch:
+; CHECK-NEXT:    [[C:%.*]] = icmp ult i32 [[IV]], [[N]]
+; CHECK-NEXT:    [[IV_NEXT]] = add i32 [[IV]], 1
+; CHECK-NEXT:    br i1 [[C]], label [[LOOP_HEADER]], label [[EXIT_SPLIT:%.*]], !llvm.loop [[LOOP13:![0-9]+]]
+; CHECK:       exit.split:
+; CHECK-NEXT:    br label [[EXIT]]
+; CHECK:       exit:
+; CHECK-NEXT:    ret i32 10
+;
+entry:
+  br label %loop.header
+
+loop.header:
+  %iv = phi i32 [ 0, %entry ], [ %iv.next, %loop.latch ]
+  %lv = load i32, ptr %ptr
+  %sc = trunc i32 %lv to i1
+  br i1 %sc, label %clobber, label %noclobber
+
+clobber:
+  call void @clobber()
+  br label %loop.latch
+
+noclobber:
+  br label %loop.latch
+
+loop.latch:
+  %c = icmp ult i32 %iv, %N
+  %iv.next = add i32 %iv, 1
+  br i1 %c, label %loop.header, label %exit
+
+exit:
+  ret i32 10
+}
+
 ; CHECK: [[LOOP0]] = distinct !{[[LOOP0]], [[UNSWITCH_PARTIAL_DISABLE:![0-9]+]]}
 ; CHECK: [[UNSWITCH_PARTIAL_DISABLE]] = !{!"llvm.loop.unswitch.partial.disable"}
 ; CHECK: [[LOOP2]] = distinct !{[[LOOP2]], [[UNSWITCH_PARTIAL_DISABLE]]}
diff --git a/llvm/test/Transforms/SimplifyCFG/UnreachableEliminate.ll b/llvm/test/Transforms/SimplifyCFG/UnreachableEliminate.ll
index 757340527ec0..ef2d3219cca9 100644
--- a/llvm/test/Transforms/SimplifyCFG/UnreachableEliminate.ll
+++ b/llvm/test/Transforms/SimplifyCFG/UnreachableEliminate.ll
@@ -627,7 +627,233 @@ else:
   ret void
 }
 
+define i32 @test_assume_false(i32 %cond) {
+; CHECK-LABEL: @test_assume_false(
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    switch i32 [[COND:%.*]], label [[DEFAULT:%.*]] [
+; CHECK-NEXT:      i32 0, label [[EXIT:%.*]]
+; CHECK-NEXT:      i32 1, label [[CASE1:%.*]]
+; CHECK-NEXT:      i32 2, label [[CASE2:%.*]]
+; CHECK-NEXT:    ]
+; CHECK:       case1:
+; CHECK-NEXT:    br label [[EXIT]]
+; CHECK:       case2:
+; CHECK-NEXT:    br label [[EXIT]]
+; CHECK:       default:
+; CHECK-NEXT:    unreachable
+; CHECK:       exit:
+; CHECK-NEXT:    [[RES:%.*]] = phi i32 [ 2, [[CASE1]] ], [ 3, [[CASE2]] ], [ 1, [[ENTRY:%.*]] ]
+; CHECK-NEXT:    call void @llvm.assume(i1 true)
+; CHECK-NEXT:    ret i32 [[RES]]
+;
+entry:
+  switch i32 %cond, label %default [
+  i32 0, label %case0
+  i32 1, label %case1
+  i32 2, label %case2
+  ]
+
+case0:
+  br label %exit
+
+case1:
+  br label %exit
+
+case2:
+  br label %exit
+
+default:
+  br label %exit
+
+exit:
+  %bool = phi i1 [ false, %default ], [ true, %case0 ], [ true, %case1 ], [ true, %case2 ]
+  %res = phi i32 [ 0, %default ], [ 1, %case0 ], [ 2, %case1 ], [ 3, %case2 ]
+  call void @llvm.assume(i1 %bool)
+  ret i32 %res
+}
+
+define i32 @test_assume_undef(i32 %cond) {
+; CHECK-LABEL: @test_assume_undef(
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    switch i32 [[COND:%.*]], label [[DEFAULT:%.*]] [
+; CHECK-NEXT:      i32 0, label [[EXIT:%.*]]
+; CHECK-NEXT:      i32 1, label [[CASE1:%.*]]
+; CHECK-NEXT:      i32 2, label [[CASE2:%.*]]
+; CHECK-NEXT:    ]
+; CHECK:       case1:
+; CHECK-NEXT:    br label [[EXIT]]
+; CHECK:       case2:
+; CHECK-NEXT:    br label [[EXIT]]
+; CHECK:       default:
+; CHECK-NEXT:    unreachable
+; CHECK:       exit:
+; CHECK-NEXT:    [[RES:%.*]] = phi i32 [ 2, [[CASE1]] ], [ 3, [[CASE2]] ], [ 1, [[ENTRY:%.*]] ]
+; CHECK-NEXT:    call void @llvm.assume(i1 true)
+; CHECK-NEXT:    ret i32 [[RES]]
+;
+entry:
+  switch i32 %cond, label %default [
+  i32 0, label %case0
+  i32 1, label %case1
+  i32 2, label %case2
+  ]
+
+case0:
+  br label %exit
+
+case1:
+  br label %exit
+
+case2:
+  br label %exit
+
+default:
+  br label %exit
+
+exit:
+  %bool = phi i1 [ undef, %default ], [ true, %case0 ], [ true, %case1 ], [ true, %case2 ]
+  %res = phi i32 [ 0, %default ], [ 1, %case0 ], [ 2, %case1 ], [ 3, %case2 ]
+  call void @llvm.assume(i1 %bool)
+  ret i32 %res
+}
+
+define i32 @test_assume_var(i32 %cond, i1 %var) {
+; CHECK-LABEL: @test_assume_var(
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    switch i32 [[COND:%.*]], label [[DEFAULT:%.*]] [
+; CHECK-NEXT:      i32 0, label [[EXIT:%.*]]
+; CHECK-NEXT:      i32 1, label [[CASE1:%.*]]
+; CHECK-NEXT:      i32 2, label [[CASE2:%.*]]
+; CHECK-NEXT:    ]
+; CHECK:       case1:
+; CHECK-NEXT:    br label [[EXIT]]
+; CHECK:       case2:
+; CHECK-NEXT:    br label [[EXIT]]
+; CHECK:       default:
+; CHECK-NEXT:    br label [[EXIT]]
+; CHECK:       exit:
+; CHECK-NEXT:    [[BOOL:%.*]] = phi i1 [ [[VAR:%.*]], [[DEFAULT]] ], [ true, [[CASE1]] ], [ true, [[CASE2]] ], [ true, [[ENTRY:%.*]] ]
+; CHECK-NEXT:    [[RES:%.*]] = phi i32 [ 0, [[DEFAULT]] ], [ 2, [[CASE1]] ], [ 3, [[CASE2]] ], [ 1, [[ENTRY]] ]
+; CHECK-NEXT:    call void @llvm.assume(i1 [[BOOL]])
+; CHECK-NEXT:    ret i32 [[RES]]
+;
+entry:
+  switch i32 %cond, label %default [
+  i32 0, label %case0
+  i32 1, label %case1
+  i32 2, label %case2
+  ]
+
+case0:
+  br label %exit
 
+case1:
+  br label %exit
+
+case2:
+  br label %exit
+
+default:
+  br label %exit
+
+exit:
+  %bool = phi i1 [ %var, %default ], [ true, %case0 ], [ true, %case1 ], [ true, %case2 ]
+  %res = phi i32 [ 0, %default ], [ 1, %case0 ], [ 2, %case1 ], [ 3, %case2 ]
+  call void @llvm.assume(i1 %bool)
+  ret i32 %res
+}
+
+define i32 @test_assume_bundle_nonnull(i32 %cond, ptr nonnull %p) {
+; CHECK-LABEL: @test_assume_bundle_nonnull(
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    switch i32 [[COND:%.*]], label [[DEFAULT:%.*]] [
+; CHECK-NEXT:      i32 0, label [[EXIT:%.*]]
+; CHECK-NEXT:      i32 1, label [[CASE1:%.*]]
+; CHECK-NEXT:      i32 2, label [[CASE2:%.*]]
+; CHECK-NEXT:    ]
+; CHECK:       case1:
+; CHECK-NEXT:    br label [[EXIT]]
+; CHECK:       case2:
+; CHECK-NEXT:    br label [[EXIT]]
+; CHECK:       default:
+; CHECK-NEXT:    br label [[EXIT]]
+; CHECK:       exit:
+; CHECK-NEXT:    [[PTR:%.*]] = phi ptr [ null, [[DEFAULT]] ], [ [[P:%.*]], [[CASE1]] ], [ [[P]], [[CASE2]] ], [ [[P]], [[ENTRY:%.*]] ]
+; CHECK-NEXT:    [[RES:%.*]] = phi i32 [ 0, [[DEFAULT]] ], [ 2, [[CASE1]] ], [ 3, [[CASE2]] ], [ 1, [[ENTRY]] ]
+; CHECK-NEXT:    call void @llvm.assume(i1 true) [ "nonnull"(ptr [[PTR]]) ]
+; CHECK-NEXT:    ret i32 [[RES]]
+;
+entry:
+  switch i32 %cond, label %default [
+  i32 0, label %case0
+  i32 1, label %case1
+  i32 2, label %case2
+  ]
+
+case0:
+  br label %exit
+
+case1:
+  br label %exit
+
+case2:
+  br label %exit
+
+default:
+  br label %exit
+
+exit:
+  %ptr = phi ptr [ null, %default ], [ %p, %case0 ], [ %p, %case1 ], [ %p, %case2 ]
+  %res = phi i32 [ 0, %default ], [ 1, %case0 ], [ 2, %case1 ], [ 3, %case2 ]
+  call void @llvm.assume(i1 true) [ "nonnull"(ptr %ptr) ]
+  ret i32 %res
+}
+
+define i32 @test_assume_bundle_align(i32 %cond, ptr nonnull %p) {
+; CHECK-LABEL: @test_assume_bundle_align(
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    switch i32 [[COND:%.*]], label [[DEFAULT:%.*]] [
+; CHECK-NEXT:      i32 0, label [[EXIT:%.*]]
+; CHECK-NEXT:      i32 1, label [[CASE1:%.*]]
+; CHECK-NEXT:      i32 2, label [[CASE2:%.*]]
+; CHECK-NEXT:    ]
+; CHECK:       case1:
+; CHECK-NEXT:    br label [[EXIT]]
+; CHECK:       case2:
+; CHECK-NEXT:    br label [[EXIT]]
+; CHECK:       default:
+; CHECK-NEXT:    br label [[EXIT]]
+; CHECK:       exit:
+; CHECK-NEXT:    [[PTR:%.*]] = phi ptr [ null, [[DEFAULT]] ], [ [[P:%.*]], [[CASE1]] ], [ [[P]], [[CASE2]] ], [ [[P]], [[ENTRY:%.*]] ]
+; CHECK-NEXT:    [[RES:%.*]] = phi i32 [ 0, [[DEFAULT]] ], [ 2, [[CASE1]] ], [ 3, [[CASE2]] ], [ 1, [[ENTRY]] ]
+; CHECK-NEXT:    call void @llvm.assume(i1 true) [ "align"(ptr [[PTR]], i32 8) ]
+; CHECK-NEXT:    ret i32 [[RES]]
+;
+entry:
+  switch i32 %cond, label %default [
+  i32 0, label %case0
+  i32 1, label %case1
+  i32 2, label %case2
+  ]
+
+case0:
+  br label %exit
+
+case1:
+  br label %exit
+
+case2:
+  br label %exit
+
+default:
+  br label %exit
+
+exit:
+  %ptr = phi ptr [ null, %default ], [ %p, %case0 ], [ %p, %case1 ], [ %p, %case2 ]
+  %res = phi i32 [ 0, %default ], [ 1, %case0 ], [ 2, %case1 ], [ 3, %case2 ]
+  call void @llvm.assume(i1 true) [ "align"(ptr %ptr, i32 8) ]
+  ret i32 %res
+}
 
 attributes #0 = { null_pointer_is_valid }
 ;.
diff --git a/llvm/test/Transforms/SimplifyCFG/hoist-dbgvalue-inlined.ll b/llvm/test/Transforms/SimplifyCFG/hoist-dbgvalue-inlined.ll
index e00d1daf71de..5af73e789f11 100644
--- a/llvm/test/Transforms/SimplifyCFG/hoist-dbgvalue-inlined.ll
+++ b/llvm/test/Transforms/SimplifyCFG/hoist-dbgvalue-inlined.ll
@@ -9,7 +9,6 @@ init:
 
 ; CHECK:  %vala = load i64, ptr %ptr
 ; CHECK-NEXT:  call void @llvm.dbg.value(metadata i64 %vala, metadata [[MD:![0-9]*]]
-; CHECK-NEXT:  call void @llvm.dbg.value(metadata i64 %vala, metadata [[MD]]
 ; CHECK-NEXT:  %valbmasked = and i64 %vala, 1
 
 a:                                              ; preds = %init
diff --git a/llvm/test/Transforms/SimplifyCFG/hoist-dbgvalue.ll b/llvm/test/Transforms/SimplifyCFG/hoist-dbgvalue.ll
index af7da45ec089..c5d723c4e3dd 100644
--- a/llvm/test/Transforms/SimplifyCFG/hoist-dbgvalue.ll
+++ b/llvm/test/Transforms/SimplifyCFG/hoist-dbgvalue.ll
@@ -47,7 +47,6 @@ define i1 @hoist_with_debug2(i32 %x) !dbg !22 {
 ; CHECK-NEXT:  entry:
 ; CHECK-NEXT:    [[TOBOOL_NOT:%.*]] = icmp ugt i32 [[X:%.*]], 2
 ; CHECK-NEXT:    call void @llvm.dbg.value(metadata i32 [[X]], metadata [[META21:![0-9]+]], metadata !DIExpression()), !dbg [[DBG23:![0-9]+]]
-; CHECK-NEXT:    call void @llvm.dbg.value(metadata i32 [[X]], metadata [[META21]], metadata !DIExpression()), !dbg [[DBG23]]
 ; CHECK-NEXT:    [[DOT:%.*]] = select i1 [[TOBOOL_NOT]], i1 false, i1 true
 ; CHECK-NEXT:    ret i1 [[DOT]]
 ;
diff --git a/llvm/test/Transforms/SimplifyCFG/mmra.ll b/llvm/test/Transforms/SimplifyCFG/mmra.ll
new file mode 100644
index 000000000000..667065747137
--- /dev/null
+++ b/llvm/test/Transforms/SimplifyCFG/mmra.ll
@@ -0,0 +1,150 @@
+; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --version 4
+; RUN: opt -passes=simplifycfg -simplifycfg-require-and-preserve-domtree=1 -S %s | FileCheck %s
+
+; RUN: opt -passes='simplifycfg<sink-common-insts;hoist-common-insts>,verify' -S %s | FileCheck %s
+
+declare void @clobber1()
+declare void @clobber2()
+
+define void @sink(ptr %arg, i1 %c) {
+; CHECK-LABEL: define void @sink(
+; CHECK-SAME: ptr [[ARG:%.*]], i1 [[C:%.*]]) {
+; CHECK-NEXT:  bb:
+; CHECK-NEXT:    br i1 [[C]], label [[THEN:%.*]], label [[ELSE:%.*]]
+; CHECK:       then:
+; CHECK-NEXT:    call void @clobber1()
+; CHECK-NEXT:    store ptr null, ptr [[ARG]], align 8
+; CHECK-NEXT:    br label [[EXIT:%.*]]
+; CHECK:       else:
+; CHECK-NEXT:    call void @clobber2()
+; CHECK-NEXT:    store ptr null, ptr [[ARG]], align 8, !mmra [[META0:![0-9]+]]
+; CHECK-NEXT:    br label [[EXIT]]
+; CHECK:       exit:
+; CHECK-NEXT:    ret void
+;
+bb:
+  br i1 %c, label %then, label %else
+
+then:
+  call void @clobber1()
+  store ptr null, ptr %arg, align 8
+  br label %exit
+
+else:
+  call void @clobber2()
+  store ptr null, ptr %arg, align 8, !mmra !0
+  br label %exit
+
+exit:
+  ret void
+}
+
+define void @hoist_store(ptr %arg, i1 %c) {
+; CHECK-LABEL: define void @hoist_store(
+; CHECK-SAME: ptr [[ARG:%.*]], i1 [[C:%.*]]) {
+; CHECK-NEXT:  bb:
+; CHECK-NEXT:    br i1 [[C]], label [[THEN:%.*]], label [[ELSE:%.*]]
+; CHECK:       then:
+; CHECK-NEXT:    store ptr null, ptr [[ARG]], align 8
+; CHECK-NEXT:    call void @clobber1()
+; CHECK-NEXT:    br label [[EXIT:%.*]]
+; CHECK:       else:
+; CHECK-NEXT:    store ptr null, ptr [[ARG]], align 8, !mmra [[META0]]
+; CHECK-NEXT:    call void @clobber2()
+; CHECK-NEXT:    br label [[EXIT]]
+; CHECK:       exit:
+; CHECK-NEXT:    ret void
+;
+bb:
+  br i1 %c, label %then, label %else
+
+then:
+  store ptr null, ptr %arg, align 8
+  call void @clobber1()
+  br label %exit
+
+else:
+  store ptr null, ptr %arg, align 8, !mmra !0
+  call void @clobber2()
+  br label %exit
+
+exit:
+  ret void
+}
+
+define ptr @sink_load(ptr %arg, i1 %c) {
+; CHECK-LABEL: define ptr @sink_load(
+; CHECK-SAME: ptr [[ARG:%.*]], i1 [[C:%.*]]) {
+; CHECK-NEXT:  bb:
+; CHECK-NEXT:    br i1 [[C]], label [[THEN:%.*]], label [[ELSE:%.*]]
+; CHECK:       then:
+; CHECK-NEXT:    call void @clobber1()
+; CHECK-NEXT:    [[L1:%.*]] = load ptr, ptr [[ARG]], align 8
+; CHECK-NEXT:    br label [[EXIT:%.*]]
+; CHECK:       else:
+; CHECK-NEXT:    call void @clobber2()
+; CHECK-NEXT:    [[L2:%.*]] = load ptr, ptr [[ARG]], align 8, !mmra [[META0]]
+; CHECK-NEXT:    br label [[EXIT]]
+; CHECK:       exit:
+; CHECK-NEXT:    [[P:%.*]] = phi ptr [ [[L1]], [[THEN]] ], [ [[L2]], [[ELSE]] ]
+; CHECK-NEXT:    ret ptr [[P]]
+;
+bb:
+  br i1 %c, label %then, label %else
+
+then:
+  call void @clobber1()
+  %l1 = load ptr, ptr %arg, align 8
+  br label %exit
+
+else:
+  call void @clobber2()
+  %l2 = load ptr, ptr %arg, align 8, !mmra !0
+  br label %exit
+
+exit:
+  %p = phi ptr [ %l1, %then ], [ %l2, %else ]
+  ret ptr %p
+}
+
+define ptr @hoist_load(ptr %arg, i1 %c) {
+; CHECK-LABEL: define ptr @hoist_load(
+; CHECK-SAME: ptr [[ARG:%.*]], i1 [[C:%.*]]) {
+; CHECK-NEXT:  bb:
+; CHECK-NEXT:    br i1 [[C]], label [[THEN:%.*]], label [[ELSE:%.*]]
+; CHECK:       then:
+; CHECK-NEXT:    [[L1:%.*]] = load ptr, ptr [[ARG]], align 8
+; CHECK-NEXT:    call void @clobber1()
+; CHECK-NEXT:    br label [[EXIT:%.*]]
+; CHECK:       else:
+; CHECK-NEXT:    [[L2:%.*]] = load ptr, ptr [[ARG]], align 8, !mmra [[META0]]
+; CHECK-NEXT:    call void @clobber2()
+; CHECK-NEXT:    br label [[EXIT]]
+; CHECK:       exit:
+; CHECK-NEXT:    [[P:%.*]] = phi ptr [ [[L1]], [[THEN]] ], [ [[L2]], [[ELSE]] ]
+; CHECK-NEXT:    ret ptr [[P]]
+;
+bb:
+  br i1 %c, label %then, label %else
+
+then:
+  %l1 = load ptr, ptr %arg, align 8
+  call void @clobber1()
+  br label %exit
+
+else:
+  %l2 = load ptr, ptr %arg, align 8, !mmra !0
+  call void @clobber2()
+  br label %exit
+
+exit:
+  %p = phi ptr [ %l1, %then ], [ %l2, %else ]
+  ret ptr %p
+}
+
+
+!0 = !{!"foo", !"bar"}
+
+;.
+; CHECK: [[META0]] = !{!"foo", !"bar"}
+;.
diff --git a/llvm/test/Transforms/SimplifyCFG/speculate-store.ll b/llvm/test/Transforms/SimplifyCFG/speculate-store.ll
index 6fbae7a0b2a7..c7ebeff5246d 100644
--- a/llvm/test/Transforms/SimplifyCFG/speculate-store.ll
+++ b/llvm/test/Transforms/SimplifyCFG/speculate-store.ll
@@ -120,14 +120,14 @@ define i32 @load_before_store_noescape(i64 %i, i32 %b)  {
 ; CHECK-NEXT:    [[A:%.*]] = alloca [2 x i32], align 8
 ; CHECK-NEXT:    store i64 4294967296, ptr [[A]], align 8
 ; CHECK-NEXT:    [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x i32], ptr [[A]], i64 0, i64 [[I:%.*]]
-; CHECK-NEXT:    [[TMP1:%.*]] = load i32, ptr [[ARRAYIDX]], align 4
-; CHECK-NEXT:    [[CMP:%.*]] = icmp slt i32 [[TMP1]], [[B:%.*]]
-; CHECK-NEXT:    [[SPEC_STORE_SELECT:%.*]] = select i1 [[CMP]], i32 [[B]], i32 [[TMP1]]
+; CHECK-NEXT:    [[TMP0:%.*]] = load i32, ptr [[ARRAYIDX]], align 4
+; CHECK-NEXT:    [[CMP:%.*]] = icmp slt i32 [[TMP0]], [[B:%.*]]
+; CHECK-NEXT:    [[SPEC_STORE_SELECT:%.*]] = select i1 [[CMP]], i32 [[B]], i32 [[TMP0]]
 ; CHECK-NEXT:    store i32 [[SPEC_STORE_SELECT]], ptr [[ARRAYIDX]], align 4
-; CHECK-NEXT:    [[TMP2:%.*]] = load i32, ptr [[A]], align 4
+; CHECK-NEXT:    [[TMP1:%.*]] = load i32, ptr [[A]], align 4
 ; CHECK-NEXT:    [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x i32], ptr [[A]], i64 0, i64 1
-; CHECK-NEXT:    [[TMP3:%.*]] = load i32, ptr [[ARRAYIDX2]], align 4
-; CHECK-NEXT:    [[ADD:%.*]] = add nsw i32 [[TMP2]], [[TMP3]]
+; CHECK-NEXT:    [[TMP2:%.*]] = load i32, ptr [[ARRAYIDX2]], align 4
+; CHECK-NEXT:    [[ADD:%.*]] = add nsw i32 [[TMP1]], [[TMP2]]
 ; CHECK-NEXT:    ret i32 [[ADD]]
 ;
 entry:
@@ -158,17 +158,17 @@ define i32 @load_before_store_escape(i64 %i, i32 %b)  {
 ; CHECK-NEXT:    store i64 4294967296, ptr [[A]], align 8
 ; CHECK-NEXT:    call void @fork_some_threads(ptr [[A]])
 ; CHECK-NEXT:    [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x i32], ptr [[A]], i64 0, i64 [[I:%.*]]
-; CHECK-NEXT:    [[TMP1:%.*]] = load i32, ptr [[ARRAYIDX]], align 4
-; CHECK-NEXT:    [[CMP:%.*]] = icmp slt i32 [[TMP1]], [[B:%.*]]
+; CHECK-NEXT:    [[TMP0:%.*]] = load i32, ptr [[ARRAYIDX]], align 4
+; CHECK-NEXT:    [[CMP:%.*]] = icmp slt i32 [[TMP0]], [[B:%.*]]
 ; CHECK-NEXT:    br i1 [[CMP]], label [[IF_THEN:%.*]], label [[IF_END:%.*]]
 ; CHECK:       if.then:
 ; CHECK-NEXT:    store i32 [[B]], ptr [[ARRAYIDX]], align 4
 ; CHECK-NEXT:    br label [[IF_END]]
 ; CHECK:       if.end:
-; CHECK-NEXT:    [[TMP2:%.*]] = load i32, ptr [[A]], align 4
+; CHECK-NEXT:    [[TMP1:%.*]] = load i32, ptr [[A]], align 4
 ; CHECK-NEXT:    [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x i32], ptr [[A]], i64 0, i64 1
-; CHECK-NEXT:    [[TMP3:%.*]] = load i32, ptr [[ARRAYIDX2]], align 4
-; CHECK-NEXT:    [[ADD:%.*]] = add nsw i32 [[TMP2]], [[TMP3]]
+; CHECK-NEXT:    [[TMP2:%.*]] = load i32, ptr [[ARRAYIDX2]], align 4
+; CHECK-NEXT:    [[ADD:%.*]] = add nsw i32 [[TMP1]], [[TMP2]]
 ; CHECK-NEXT:    call void @join_some_threads()
 ; CHECK-NEXT:    ret i32 [[ADD]]
 ;
@@ -205,18 +205,18 @@ define i32 @not_alone_in_block(i64 %i, i32 %b)  {
 ; CHECK-NEXT:    [[A:%.*]] = alloca [2 x i32], align 8
 ; CHECK-NEXT:    store i64 4294967296, ptr [[A]], align 8
 ; CHECK-NEXT:    [[ARRAYIDX:%.*]] = getelementptr inbounds [2 x i32], ptr [[A]], i64 0, i64 [[I:%.*]]
-; CHECK-NEXT:    [[TMP1:%.*]] = load i32, ptr [[ARRAYIDX]], align 4
-; CHECK-NEXT:    [[CMP:%.*]] = icmp slt i32 [[TMP1]], [[B:%.*]]
+; CHECK-NEXT:    [[TMP0:%.*]] = load i32, ptr [[ARRAYIDX]], align 4
+; CHECK-NEXT:    [[CMP:%.*]] = icmp slt i32 [[TMP0]], [[B:%.*]]
 ; CHECK-NEXT:    br i1 [[CMP]], label [[IF_THEN:%.*]], label [[IF_END:%.*]]
 ; CHECK:       if.then:
 ; CHECK-NEXT:    store i32 [[B]], ptr [[ARRAYIDX]], align 4
 ; CHECK-NEXT:    store i32 [[B]], ptr [[A]], align 4
 ; CHECK-NEXT:    br label [[IF_END]]
 ; CHECK:       if.end:
-; CHECK-NEXT:    [[TMP2:%.*]] = load i32, ptr [[A]], align 4
+; CHECK-NEXT:    [[TMP1:%.*]] = load i32, ptr [[A]], align 4
 ; CHECK-NEXT:    [[ARRAYIDX2:%.*]] = getelementptr inbounds [2 x i32], ptr [[A]], i64 0, i64 1
-; CHECK-NEXT:    [[TMP3:%.*]] = load i32, ptr [[ARRAYIDX2]], align 4
-; CHECK-NEXT:    [[ADD:%.*]] = add nsw i32 [[TMP2]], [[TMP3]]
+; CHECK-NEXT:    [[TMP2:%.*]] = load i32, ptr [[ARRAYIDX2]], align 4
+; CHECK-NEXT:    [[ADD:%.*]] = add nsw i32 [[TMP1]], [[TMP2]]
 ; CHECK-NEXT:    ret i32 [[ADD]]
 ;
 entry:
@@ -240,6 +240,58 @@ if.end:
   ret i32 %add
 }
 
+define void @wrong_align_store(ptr %A, i32 %B, i32 %C, i32 %D) {
+; CHECK-LABEL: @wrong_align_store(
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    store i32 [[B:%.*]], ptr [[A:%.*]], align 4
+; CHECK-NEXT:    [[CMP:%.*]] = icmp sgt i32 [[D:%.*]], 42
+; CHECK-NEXT:    br i1 [[CMP]], label [[IF_THEN:%.*]], label [[RET_END:%.*]]
+; CHECK:       if.then:
+; CHECK-NEXT:    store i32 [[C:%.*]], ptr [[A]], align 8
+; CHECK-NEXT:    br label [[RET_END]]
+; CHECK:       ret.end:
+; CHECK-NEXT:    ret void
+;
+entry:
+  store i32 %B, ptr %A, align 4
+  %cmp = icmp sgt i32 %D, 42
+  br i1 %cmp, label %if.then, label %ret.end
+
+if.then:
+  store i32 %C, ptr %A, align 8
+  br label %ret.end
+
+ret.end:
+  ret void
+}
+
+define void @wrong_align_load(i32 %C, i32 %D) {
+; CHECK-LABEL: @wrong_align_load(
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[A:%.*]] = alloca i32, align 4
+; CHECK-NEXT:    [[TMP0:%.*]] = load i32, ptr [[A]], align 4
+; CHECK-NEXT:    [[CMP:%.*]] = icmp sgt i32 [[D:%.*]], 42
+; CHECK-NEXT:    br i1 [[CMP]], label [[IF_THEN:%.*]], label [[RET_END:%.*]]
+; CHECK:       if.then:
+; CHECK-NEXT:    store i32 [[C:%.*]], ptr [[A]], align 8
+; CHECK-NEXT:    br label [[RET_END]]
+; CHECK:       ret.end:
+; CHECK-NEXT:    ret void
+;
+entry:
+  %A = alloca i32, align 4
+  load i32, ptr %A, align 4
+  %cmp = icmp sgt i32 %D, 42
+  br i1 %cmp, label %if.then, label %ret.end
+
+if.then:
+  store i32 %C, ptr %A, align 8
+  br label %ret.end
+
+ret.end:
+  ret void
+}
+
 ; CHECK: !0 = !{!"branch_weights", i32 3, i32 5}
 !0 = !{!"branch_weights", i32 3, i32 5}
 
diff --git a/llvm/test/Transforms/VectorCombine/AArch64/shuffletoidentity.ll b/llvm/test/Transforms/VectorCombine/AArch64/shuffletoidentity.ll
index 4cc7d5f25328..6a81964b917e 100644
--- a/llvm/test/Transforms/VectorCombine/AArch64/shuffletoidentity.ll
+++ b/llvm/test/Transforms/VectorCombine/AArch64/shuffletoidentity.ll
@@ -5,7 +5,7 @@ target triple = "aarch64"
 
 define <8 x i8> @trivial(<8 x i8> %a) {
 ; CHECK-LABEL: @trivial(
-; CHECK-NEXT:    ret <8 x i8> [[R:%.*]]
+; CHECK-NEXT:    ret <8 x i8> [[A:%.*]]
 ;
   %ab = shufflevector <8 x i8> %a, <8 x i8> poison, <4 x i32> <i32 3, i32 2, i32 1, i32 0>
   %at = shufflevector <8 x i8> %a, <8 x i8> poison, <4 x i32> <i32 7, i32 6, i32 5, i32 4>
diff --git a/llvm/test/Transforms/VectorCombine/X86/shuffle-of-binops.ll b/llvm/test/Transforms/VectorCombine/X86/shuffle-of-binops.ll
index 5c4ad4f1fcc4..c423053a9a48 100644
--- a/llvm/test/Transforms/VectorCombine/X86/shuffle-of-binops.ll
+++ b/llvm/test/Transforms/VectorCombine/X86/shuffle-of-binops.ll
@@ -25,9 +25,9 @@ define <4 x float> @shuf_fdiv_v4f32_yy(<4 x float> %x, <4 x float> %y, <4 x floa
 define <4 x i32> @shuf_add_v4i32_xx(<4 x i32> %x, <4 x i32> %y, <4 x i32> %z) {
 ; CHECK-LABEL: define <4 x i32> @shuf_add_v4i32_xx(
 ; CHECK-SAME: <4 x i32> [[X:%.*]], <4 x i32> [[Y:%.*]], <4 x i32> [[Z:%.*]]) #[[ATTR0]] {
-; CHECK-NEXT:    [[TMP1:%.*]] = shufflevector <4 x i32> [[X]], <4 x i32> poison, <4 x i32> <i32 poison, i32 poison, i32 2, i32 0>
-; CHECK-NEXT:    [[R1:%.*]] = shufflevector <4 x i32> [[Y]], <4 x i32> [[Z]], <4 x i32> <i32 poison, i32 poison, i32 6, i32 0>
-; CHECK-NEXT:    [[R2:%.*]] = add <4 x i32> [[TMP1]], [[R1]]
+; CHECK-NEXT:    [[B0:%.*]] = add <4 x i32> [[X]], [[Y]]
+; CHECK-NEXT:    [[B1:%.*]] = add <4 x i32> [[X]], [[Z]]
+; CHECK-NEXT:    [[R2:%.*]] = shufflevector <4 x i32> [[B0]], <4 x i32> [[B1]], <4 x i32> <i32 poison, i32 poison, i32 6, i32 0>
 ; CHECK-NEXT:    ret <4 x i32> [[R2]]
 ;
   %b0 = add <4 x i32> %x, %y
@@ -36,15 +36,22 @@ define <4 x i32> @shuf_add_v4i32_xx(<4 x i32> %x, <4 x i32> %y, <4 x i32> %z) {
   ret <4 x i32> %r
 }
 
-; For commutative instructions, common operand may be swapped.
+; For commutative instructions, common operand may be swapped (SSE - expensive fmul vs AVX - cheap fmul)
 
 define <4 x float> @shuf_fmul_v4f32_xx_swap(<4 x float> %x, <4 x float> %y, <4 x float> %z) {
-; CHECK-LABEL: define <4 x float> @shuf_fmul_v4f32_xx_swap(
-; CHECK-SAME: <4 x float> [[X:%.*]], <4 x float> [[Y:%.*]], <4 x float> [[Z:%.*]]) #[[ATTR0]] {
-; CHECK-NEXT:    [[TMP1:%.*]] = shufflevector <4 x float> [[Y]], <4 x float> [[Z]], <4 x i32> <i32 0, i32 3, i32 4, i32 7>
-; CHECK-NEXT:    [[TMP2:%.*]] = shufflevector <4 x float> [[X]], <4 x float> poison, <4 x i32> <i32 0, i32 3, i32 0, i32 3>
-; CHECK-NEXT:    [[R:%.*]] = fmul <4 x float> [[TMP1]], [[TMP2]]
-; CHECK-NEXT:    ret <4 x float> [[R]]
+; SSE-LABEL: define <4 x float> @shuf_fmul_v4f32_xx_swap(
+; SSE-SAME: <4 x float> [[X:%.*]], <4 x float> [[Y:%.*]], <4 x float> [[Z:%.*]]) #[[ATTR0]] {
+; SSE-NEXT:    [[TMP1:%.*]] = shufflevector <4 x float> [[Y]], <4 x float> [[Z]], <4 x i32> <i32 0, i32 3, i32 4, i32 7>
+; SSE-NEXT:    [[TMP2:%.*]] = shufflevector <4 x float> [[X]], <4 x float> poison, <4 x i32> <i32 0, i32 3, i32 0, i32 3>
+; SSE-NEXT:    [[R:%.*]] = fmul <4 x float> [[TMP1]], [[TMP2]]
+; SSE-NEXT:    ret <4 x float> [[R]]
+;
+; AVX-LABEL: define <4 x float> @shuf_fmul_v4f32_xx_swap(
+; AVX-SAME: <4 x float> [[X:%.*]], <4 x float> [[Y:%.*]], <4 x float> [[Z:%.*]]) #[[ATTR0]] {
+; AVX-NEXT:    [[B0:%.*]] = fmul <4 x float> [[X]], [[Y]]
+; AVX-NEXT:    [[B1:%.*]] = fmul <4 x float> [[Z]], [[X]]
+; AVX-NEXT:    [[R:%.*]] = shufflevector <4 x float> [[B0]], <4 x float> [[B1]], <4 x i32> <i32 0, i32 3, i32 4, i32 7>
+; AVX-NEXT:    ret <4 x float> [[R]]
 ;
   %b0 = fmul <4 x float> %x, %y
   %b1 = fmul <4 x float> %z, %x
@@ -57,9 +64,9 @@ define <4 x float> @shuf_fmul_v4f32_xx_swap(<4 x float> %x, <4 x float> %y, <4 x
 define <2 x i64> @shuf_and_v2i64_yy_swap(<2 x i64> %x, <2 x i64> %y, <2 x i64> %z) {
 ; CHECK-LABEL: define <2 x i64> @shuf_and_v2i64_yy_swap(
 ; CHECK-SAME: <2 x i64> [[X:%.*]], <2 x i64> [[Y:%.*]], <2 x i64> [[Z:%.*]]) #[[ATTR0]] {
-; CHECK-NEXT:    [[TMP1:%.*]] = shufflevector <2 x i64> [[Y]], <2 x i64> poison, <2 x i32> <i32 1, i32 0>
-; CHECK-NEXT:    [[TMP2:%.*]] = shufflevector <2 x i64> [[X]], <2 x i64> [[Z]], <2 x i32> <i32 3, i32 0>
-; CHECK-NEXT:    [[R:%.*]] = and <2 x i64> [[TMP1]], [[TMP2]]
+; CHECK-NEXT:    [[B0:%.*]] = and <2 x i64> [[X]], [[Y]]
+; CHECK-NEXT:    [[B1:%.*]] = and <2 x i64> [[Y]], [[Z]]
+; CHECK-NEXT:    [[R:%.*]] = shufflevector <2 x i64> [[B0]], <2 x i64> [[B1]], <2 x i32> <i32 3, i32 0>
 ; CHECK-NEXT:    ret <2 x i64> [[R]]
 ;
   %b0 = and <2 x i64> %x, %y
@@ -84,15 +91,22 @@ define <4 x i32> @shuf_shl_v4i32_xx(<4 x i32> %x, <4 x i32> %y, <4 x i32> %z) {
   ret <4 x i32> %r
 }
 
-; negative test - common operand, but not commutable
+; common operand, but not commutable (SSE - expensive vector shift vs AVX2 - cheap vector shift)
 
 define <4 x i32> @shuf_shl_v4i32_xx_swap(<4 x i32> %x, <4 x i32> %y, <4 x i32> %z) {
-; CHECK-LABEL: define <4 x i32> @shuf_shl_v4i32_xx_swap(
-; CHECK-SAME: <4 x i32> [[X:%.*]], <4 x i32> [[Y:%.*]], <4 x i32> [[Z:%.*]]) #[[ATTR0]] {
-; CHECK-NEXT:    [[B0:%.*]] = shl <4 x i32> [[X]], [[Y]]
-; CHECK-NEXT:    [[B1:%.*]] = shl <4 x i32> [[Z]], [[X]]
-; CHECK-NEXT:    [[R1:%.*]] = shufflevector <4 x i32> [[B0]], <4 x i32> [[B1]], <4 x i32> <i32 3, i32 2, i32 2, i32 5>
-; CHECK-NEXT:    ret <4 x i32> [[R1]]
+; SSE-LABEL: define <4 x i32> @shuf_shl_v4i32_xx_swap(
+; SSE-SAME: <4 x i32> [[X:%.*]], <4 x i32> [[Y:%.*]], <4 x i32> [[Z:%.*]]) #[[ATTR0]] {
+; SSE-NEXT:    [[TMP1:%.*]] = shufflevector <4 x i32> [[X]], <4 x i32> [[Z]], <4 x i32> <i32 3, i32 2, i32 2, i32 5>
+; SSE-NEXT:    [[TMP2:%.*]] = shufflevector <4 x i32> [[Y]], <4 x i32> [[X]], <4 x i32> <i32 3, i32 2, i32 2, i32 5>
+; SSE-NEXT:    [[R:%.*]] = shl <4 x i32> [[TMP1]], [[TMP2]]
+; SSE-NEXT:    ret <4 x i32> [[R]]
+;
+; AVX-LABEL: define <4 x i32> @shuf_shl_v4i32_xx_swap(
+; AVX-SAME: <4 x i32> [[X:%.*]], <4 x i32> [[Y:%.*]], <4 x i32> [[Z:%.*]]) #[[ATTR0]] {
+; AVX-NEXT:    [[B0:%.*]] = shl <4 x i32> [[X]], [[Y]]
+; AVX-NEXT:    [[B1:%.*]] = shl <4 x i32> [[Z]], [[X]]
+; AVX-NEXT:    [[R:%.*]] = shufflevector <4 x i32> [[B0]], <4 x i32> [[B1]], <4 x i32> <i32 3, i32 2, i32 2, i32 5>
+; AVX-NEXT:    ret <4 x i32> [[R]]
 ;
   %b0 = shl <4 x i32> %x, %y
   %b1 = shl <4 x i32> %z, %x
@@ -116,7 +130,7 @@ define <2 x i64> @shuf_sub_add_v2i64_yy(<2 x i64> %x, <2 x i64> %y, <2 x i64> %z
   ret <2 x i64> %r
 }
 
-; negative test - type change via shuffle
+; type change via shuffle
 
 define <8 x float> @shuf_fmul_v4f32_xx_type(<4 x float> %x, <4 x float> %y, <4 x float> %z) {
 ; CHECK-LABEL: define <8 x float> @shuf_fmul_v4f32_xx_type(
@@ -168,18 +182,18 @@ define <4 x i32> @shuf_mul_v4i32_yy_use2(<4 x i32> %x, <4 x i32> %y, <4 x i32> %
   ret <4 x i32> %r
 }
 
-; negative test - must have matching operand
+; non-matching operands (not commutable)
 
-define <4 x float> @shuf_fadd_v4f32_no_common_op(<4 x float> %x, <4 x float> %y, <4 x float> %z, <4 x float> %w) {
-; CHECK-LABEL: define <4 x float> @shuf_fadd_v4f32_no_common_op(
+define <4 x float> @shuf_fdiv_v4f32_no_common_op(<4 x float> %x, <4 x float> %y, <4 x float> %z, <4 x float> %w) {
+; CHECK-LABEL: define <4 x float> @shuf_fdiv_v4f32_no_common_op(
 ; CHECK-SAME: <4 x float> [[X:%.*]], <4 x float> [[Y:%.*]], <4 x float> [[Z:%.*]], <4 x float> [[W:%.*]]) #[[ATTR0]] {
-; CHECK-NEXT:    [[B0:%.*]] = fadd <4 x float> [[X]], [[Y]]
-; CHECK-NEXT:    [[B1:%.*]] = fadd <4 x float> [[Z]], [[W]]
-; CHECK-NEXT:    [[R:%.*]] = shufflevector <4 x float> [[B0]], <4 x float> [[B1]], <4 x i32> <i32 1, i32 3, i32 5, i32 7>
+; CHECK-NEXT:    [[TMP1:%.*]] = shufflevector <4 x float> [[X]], <4 x float> [[Z]], <4 x i32> <i32 1, i32 3, i32 5, i32 7>
+; CHECK-NEXT:    [[TMP2:%.*]] = shufflevector <4 x float> [[Y]], <4 x float> [[W]], <4 x i32> <i32 1, i32 3, i32 5, i32 7>
+; CHECK-NEXT:    [[R:%.*]] = fdiv <4 x float> [[TMP1]], [[TMP2]]
 ; CHECK-NEXT:    ret <4 x float> [[R]]
 ;
-  %b0 = fadd <4 x float> %x, %y
-  %b1 = fadd <4 x float> %z, %w
+  %b0 = fdiv <4 x float> %x, %y
+  %b1 = fdiv <4 x float> %z, %w
   %r = shufflevector <4 x float> %b0, <4 x float> %b1, <4 x i32> <i32 1, i32 3, i32 5, i32 7>
   ret <4 x float> %r
 }
@@ -216,6 +230,3 @@ define <4 x i32> @shuf_srem_v4i32_poison(<4 x i32> %a0, <4 x i32> %a1) {
   ret <4 x i32> %r
 }
 
-;; NOTE: These prefixes are unused and the list is autogenerated. Do not add tests below this line:
-; AVX: {{.*}}
-; SSE: {{.*}}
diff --git a/llvm/test/Transforms/VectorCombine/X86/shuffle-of-shuffles.ll b/llvm/test/Transforms/VectorCombine/X86/shuffle-of-shuffles.ll
index 9079a4f693ab..57df36a3874a 100644
--- a/llvm/test/Transforms/VectorCombine/X86/shuffle-of-shuffles.ll
+++ b/llvm/test/Transforms/VectorCombine/X86/shuffle-of-shuffles.ll
@@ -43,3 +43,23 @@ define <8 x i32> @concat_extract_subvectors_poison(<8 x i32> %x) {
   %concat = shufflevector <4 x i32> %lo, <4 x i32> %hi, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>
   ret <8 x i32> %concat
 }
+
+; broadcast loads are free on AVX (and blends are much cheap than general 2-operand shuffles)
+
+define  <4 x double> @blend_broadcasts_v4f64(ptr %p0, ptr %p1)  {
+; CHECK-LABEL: define <4 x double> @blend_broadcasts_v4f64(
+; CHECK-SAME: ptr [[P0:%.*]], ptr [[P1:%.*]]) #[[ATTR0]] {
+; CHECK-NEXT:    [[LD0:%.*]] = load <4 x double>, ptr [[P0]], align 32
+; CHECK-NEXT:    [[LD1:%.*]] = load <4 x double>, ptr [[P1]], align 32
+; CHECK-NEXT:    [[BCST0:%.*]] = shufflevector <4 x double> [[LD0]], <4 x double> undef, <4 x i32> zeroinitializer
+; CHECK-NEXT:    [[BCST1:%.*]] = shufflevector <4 x double> [[LD1]], <4 x double> undef, <4 x i32> zeroinitializer
+; CHECK-NEXT:    [[BLEND:%.*]] = shufflevector <4 x double> [[BCST0]], <4 x double> [[BCST1]], <4 x i32> <i32 0, i32 5, i32 6, i32 3>
+; CHECK-NEXT:    ret <4 x double> [[BLEND]]
+;
+  %ld0 = load <4 x double>, ptr %p0, align 32
+  %ld1 = load <4 x double>, ptr %p1, align 32
+  %bcst0 = shufflevector <4 x double> %ld0, <4 x double> undef, <4 x i32> zeroinitializer
+  %bcst1 = shufflevector <4 x double> %ld1, <4 x double> undef, <4 x i32> zeroinitializer
+  %blend = shufflevector <4 x double> %bcst0, <4 x double> %bcst1, <4 x i32> <i32 0, i32 5, i32 6, i32 3>
+  ret <4 x double> %blend
+}
diff --git a/llvm/test/Verifier/invalid-splice.ll b/llvm/test/Verifier/invalid-splice.ll
index d5096bdf17ca..2239386df562 100644
--- a/llvm/test/Verifier/invalid-splice.ll
+++ b/llvm/test/Verifier/invalid-splice.ll
@@ -2,36 +2,36 @@
 
 ; CHECK: The splice index exceeds the range [-VL, VL-1] where VL is the known minimum number of elements in the vector
 define <2 x double> @splice_v2f64_idx_neg3(<2 x double> %a, <2 x double> %b) #0 {
-  %res = call <2 x double> @llvm.experimental.vector.splice.v2f64(<2 x double> %a, <2 x double> %b, i32 -3)
+  %res = call <2 x double> @llvm.vector.splice.v2f64(<2 x double> %a, <2 x double> %b, i32 -3)
   ret <2 x double> %res
 }
 
 ; CHECK: The splice index exceeds the range [-VL, VL-1] where VL is the known minimum number of elements in the vector
 define <vscale x 2 x double> @splice_nxv2f64_idx_neg3_vscale_min1(<vscale x 2 x double> %a, <vscale x 2 x double> %b) #0 {
-  %res = call <vscale x 2 x double> @llvm.experimental.vector.splice.nxv2f64(<vscale x 2 x double> %a, <vscale x 2 x double> %b, i32 -3)
+  %res = call <vscale x 2 x double> @llvm.vector.splice.nxv2f64(<vscale x 2 x double> %a, <vscale x 2 x double> %b, i32 -3)
   ret <vscale x 2 x double> %res
 }
 
 ; CHECK: The splice index exceeds the range [-VL, VL-1] where VL is the known minimum number of elements in the vector
 define <vscale x 2 x double> @splice_nxv2f64_idx_neg5_vscale_min2(<vscale x 2 x double> %a, <vscale x 2 x double> %b) #1 {
-  %res = call <vscale x 2 x double> @llvm.experimental.vector.splice.nxv2f64(<vscale x 2 x double> %a, <vscale x 2 x double> %b, i32 -5)
+  %res = call <vscale x 2 x double> @llvm.vector.splice.nxv2f64(<vscale x 2 x double> %a, <vscale x 2 x double> %b, i32 -5)
   ret <vscale x 2 x double> %res
 }
 
 ; CHECK: The splice index exceeds the range [-VL, VL-1] where VL is the known minimum number of elements in the vector
 define <2 x double> @splice_v2f64_idx2(<2 x double> %a, <2 x double> %b) #0 {
-  %res = call <2 x double> @llvm.experimental.vector.splice.v2f64(<2 x double> %a, <2 x double> %b, i32 2)
+  %res = call <2 x double> @llvm.vector.splice.v2f64(<2 x double> %a, <2 x double> %b, i32 2)
   ret <2 x double> %res
 }
 
 ; CHECK: The splice index exceeds the range [-VL, VL-1] where VL is the known minimum number of elements in the vector
 define <2 x double> @splice_v2f64_idx3(<2 x double> %a, <2 x double> %b) #1 {
-  %res = call <2 x double> @llvm.experimental.vector.splice.v2f64(<2 x double> %a, <2 x double> %b, i32 4)
+  %res = call <2 x double> @llvm.vector.splice.v2f64(<2 x double> %a, <2 x double> %b, i32 4)
   ret <2 x double> %res
 }
 
 attributes #0 = { vscale_range(1,16) }
 attributes #1 = { vscale_range(2,16) }
 
-declare <2 x double> @llvm.experimental.vector.splice.v2f64(<2 x double>, <2 x double>, i32)
-declare <vscale x 2 x double> @llvm.experimental.vector.splice.nxv2f64(<vscale x 2 x double>, <vscale x 2 x double>, i32)
+declare <2 x double> @llvm.vector.splice.v2f64(<2 x double>, <2 x double>, i32)
+declare <vscale x 2 x double> @llvm.vector.splice.nxv2f64(<vscale x 2 x double>, <vscale x 2 x double>, i32)
diff --git a/llvm/test/Verifier/mmra-allowed.ll b/llvm/test/Verifier/mmra-allowed.ll
new file mode 100644
index 000000000000..76dff3f207cd
--- /dev/null
+++ b/llvm/test/Verifier/mmra-allowed.ll
@@ -0,0 +1,31 @@
+; RUN: opt -S -passes=verify < %s
+
+; This file contains MMRA metadata that is okay and should pass the verifier.
+
+define void @test(ptr %ptr) {
+  %ld = load i8, ptr %ptr,                                !mmra !0
+  store i8 1, ptr %ptr,                                   !mmra !1
+  call void @writesMem(),                                 !mmra !2
+  call void @readsMem(),                                  !mmra !2
+  fence release,                                          !mmra !0
+  %rmw.1 = atomicrmw add ptr %ptr, i8 0 release,          !mmra !0
+  %rmw.2 = atomicrmw add ptr %ptr, i8 0 acquire,          !mmra !0
+  %pair = cmpxchg ptr %ptr, i8 0, i8 1 acquire acquire,   !mmra !1
+  %ld.atomic = load atomic i8, ptr %ptr acquire, align 4, !mmra !1
+  store atomic i8 1, ptr %ptr release, align 4,           !mmra !2
+  %mld = call <2 x i64> @llvm.vp.load.v2i64.p0(ptr undef, <2 x i1> undef, i32 undef), !mmra !2
+  ; TODO: barrier
+  ret void
+}
+
+declare <2 x i64> @llvm.vp.load.v2i64.p0(ptr, <2 x i1>, i32)
+
+declare void @readsMem(ptr) #0
+declare void @writesMem(ptr) #1
+
+attributes #0 = { memory(read) }
+attributes #1 = { memory(write) }
+
+!0 = !{!"scope", !"workgroup"}
+!1 = !{!"as", !"private"}
+!2 = !{!0, !1}
diff --git a/llvm/test/Verifier/mmra.ll b/llvm/test/Verifier/mmra.ll
new file mode 100644
index 000000000000..b506d593a1c4
--- /dev/null
+++ b/llvm/test/Verifier/mmra.ll
@@ -0,0 +1,43 @@
+; RUN: not opt -S -passes=verify < %s 2>&1 | FileCheck %s
+
+define void @foo(ptr %ptr, i32 %x) {
+
+  ; CHECK: !mmra metadata attached to unexpected instruction kind
+  ; CHECK-NEXT: %bad.add
+  %bad.add = add i32 %x, 42, !mmra !{}
+
+  ; CHECK: !mmra metadata attached to unexpected instruction kind
+  ; CHECK-NEXT: %bad.sub
+  %bad.sub = sub i32 %x, 42, !mmra !{}
+
+  ; CHECK: !mmra metadata attached to unexpected instruction kind
+  ; CHECK-NEXT: %bad.sqrt
+  %bad.sqrt = call float @llvm.sqrt.f32(float undef), !mmra !{}
+
+  ; CHECK: !mmra expected to be a metadata tuple
+  ; CHECK-NEXT: %bad.md0
+  ; CHECK-NEXT: !DIFile
+  %bad.md0 = load atomic i32, ptr %ptr acquire, align 4, !mmra !0
+
+  ; CHECK: !mmra expected to be a metadata tuple
+  ; CHECK-NEXT: %bad.md1
+  ; CHECK-NEXT: !DIFile
+  %bad.md1 = load atomic i32, ptr %ptr acquire, align 4, !mmra !0
+
+  ; CHECK: !mmra metadata tuple operand is not an MMRA tag
+  ; CHECK-NEXT: %bad.md2
+  ; CHECK-NEXT: !"foo"
+  %bad.md2 = load atomic i32, ptr %ptr acquire, align 4, !mmra !1
+
+  ; CHECK: !mmra metadata tuple operand is not an MMRA tag
+  ; CHECK-NEXT: %bad.md3
+  ; CHECK-NEXT: !"baz"
+  %bad.md3 = load atomic i32, ptr %ptr acquire, align 4, !mmra !2
+  ret void
+}
+
+declare float @llvm.sqrt.f32(float)
+
+!0 = !DIFile(filename: "test.c", directory: "")
+!1 = !{!"foo", !"bar", !"bux"}
+!2 = !{!"baz", !0}
diff --git a/llvm/test/lit.cfg.py b/llvm/test/lit.cfg.py
index 4c05317036d1..affd87b98c14 100644
--- a/llvm/test/lit.cfg.py
+++ b/llvm/test/lit.cfg.py
@@ -306,6 +306,9 @@ def enable_ptxas(ptxas_executable):
             (11, 8),
             (12, 0),
             (12, 1),
+            (12, 2),
+            (12, 3),
+            (12, 4),
         ]
 
         def version_int(ver):
diff --git a/llvm/test/tools/llvm-mca/X86/BtVer2/skip-unsupported-instructions-none-remain.s b/llvm/test/tools/llvm-mca/X86/BtVer2/skip-unsupported-instructions-none-remain.s
new file mode 100644
index 000000000000..0d67f53e12f1
--- /dev/null
+++ b/llvm/test/tools/llvm-mca/X86/BtVer2/skip-unsupported-instructions-none-remain.s
@@ -0,0 +1,14 @@
+# RUN: not llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=btver2 -skip-unsupported-instructions %s 2>&1 | FileCheck --check-prefixes=CHECK-ALL,CHECK-SKIP %s
+# RUN: not llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=btver2 %s 2>&1 | FileCheck --check-prefixes=CHECK-ALL,CHECK-ERROR %s
+
+# Test defends that if all instructions are skipped leaving an empty input, an error is printed.
+
+bzhi %eax, %ebx, %ecx
+
+# CHECK-ALL-NOT: error
+
+# CHECK-ERROR: error: found an unsupported instruction in the input assembly sequence, use -skip-unsupported-instructions to ignore.
+
+# CHECK-SKIP: warning: found an unsupported instruction in the input assembly sequence, skipping with -skip-unsupported-instructions, note accuracy will be impacted:
+# CHECK-SKIP: note: instruction:      bzhil   %eax, %ebx, %ecx
+# CHECK-SKIP: error: no assembly instructions found.
diff --git a/llvm/test/tools/llvm-mca/X86/BtVer2/unsupported-instruction.s b/llvm/test/tools/llvm-mca/X86/BtVer2/unsupported-instruction.s
index bb88e951c129..3690a1101be9 100644
--- a/llvm/test/tools/llvm-mca/X86/BtVer2/unsupported-instruction.s
+++ b/llvm/test/tools/llvm-mca/X86/BtVer2/unsupported-instruction.s
@@ -1,6 +1,55 @@
-# RUN: not llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=btver2 %s 2>&1 | FileCheck %s
+# RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=btver2 -skip-unsupported-instructions -timeline %s 2>&1 | FileCheck --check-prefix=CHECK-SKIP %s
+# RUN: not llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=btver2 %s 2>&1 | FileCheck --check-prefix=CHECK-ERROR %s
+
+# Test checks that unsupported instructions exit with an error, unless -skip-unsupported-instructions is passed, in which case the remaining instructions should be analysed.
+
+# CHECK-SKIP: warning: found an unsupported instruction in the input assembly sequence, skipping with -skip-unsupported-instructions, note accuracy will be impacted:
+# CHECK-ERROR: error: found an unsupported instruction in the input assembly sequence, use -skip-unsupported-instructions to ignore.
 
 bzhi %eax, %ebx, %ecx
 
-# CHECK: error: found an unsupported instruction in the input assembly sequence.
-# CHECK-NEXT: note: instruction: 	bzhil	%eax, %ebx, %ecx
+# Supported instruction that may be analysed.
+add %eax, %eax
+
+# CHECK-SKIP: Iterations:        100
+# CHECK-SKIP: Instructions:      100
+# CHECK-SKIP: Total Cycles:      103
+# CHECK-SKIP: Total uOps:        100
+
+# CHECK-SKIP: Dispatch Width:    2
+# CHECK-SKIP: uOps Per Cycle:    0.97
+# CHECK-SKIP: IPC:               0.97
+# CHECK-SKIP: Block RThroughput: 0.5
+
+# CHECK-SKIP: Instruction Info:
+# CHECK-SKIP: [1]: #uOps
+# CHECK-SKIP: [2]: Latency
+# CHECK-SKIP: [3]: RThroughput
+# CHECK-SKIP: [4]: MayLoad
+# CHECK-SKIP: [5]: MayStore
+# CHECK-SKIP: [6]: HasSideEffects (U)
+
+# CHECK-SKIP: [1]    [2]    [3]    [4]    [5]    [6]    Instructions:
+# CHECK-SKIP:  1      1     0.50                        addl  %eax, %eax
+
+# CHECK-SKIP: Timeline view:
+
+# CHECK-SKIP: [0,0]     DeER .    . .   addl  %eax, %eax
+# CHECK-SKIP: [1,0]     D=eER.    . .   addl  %eax, %eax
+# CHECK-SKIP: [2,0]     .D=eER    . .   addl  %eax, %eax
+# CHECK-SKIP: [3,0]     .D==eER   . .   addl  %eax, %eax
+# CHECK-SKIP: [4,0]     . D==eER  . .   addl  %eax, %eax
+# CHECK-SKIP: [5,0]     . D===eER . .   addl  %eax, %eax
+# CHECK-SKIP: [6,0]     .  D===eER. .   addl  %eax, %eax
+# CHECK-SKIP: [7,0]     .  D====eER .   addl  %eax, %eax
+# CHECK-SKIP: [8,0]     .   D====eER.   addl  %eax, %eax
+# CHECK-SKIP: [9,0]     .   D=====eER   addl  %eax, %eax
+
+# CHECK-SKIP: Average Wait times (based on the timeline view):
+# CHECK-SKIP: [0]: Executions
+# CHECK-SKIP: [1]: Average time spent waiting in a scheduler's queue
+# CHECK-SKIP: [2]: Average time spent waiting in a scheduler's queue while ready
+# CHECK-SKIP: [3]: Average time elapsed from WB until retire stage
+
+# CHECK-SKIP:       [0]    [1]    [2]    [3]
+# CHECK-SKIP: 0.     10    3.5    0.1    0.0       addl       %eax, %eax
diff --git a/llvm/test/tools/llvm-profdata/memprof-merge-v0.test b/llvm/test/tools/llvm-profdata/memprof-merge-v0.test
index 03ccbdd42efd..28f65e0781bc 100644
--- a/llvm/test/tools/llvm-profdata/memprof-merge-v0.test
+++ b/llvm/test/tools/llvm-profdata/memprof-merge-v0.test
@@ -16,6 +16,9 @@ RUN: llvm-profdata show %t.prof.v1 | FileCheck %s
 RUN: llvm-profdata merge %t.proftext %p/Inputs/basic.memprofraw --memprof-version=2 --profiled-binary %p/Inputs/basic.memprofexe -o %t.prof.v2
 RUN: llvm-profdata show %t.prof.v2 | FileCheck %s
 
+RUN: llvm-profdata merge %t.proftext %p/Inputs/basic.memprofraw --memprof-version=2 --memprof-full-schema --profiled-binary %p/Inputs/basic.memprofexe -o %t.prof.v2
+RUN: llvm-profdata show %t.prof.v2 | FileCheck %s
+
 For now we only check the validity of the instrumented profile since we don't
 have a way to display the contents of the memprof indexed format yet.
 
diff --git a/llvm/test/tools/llvm-rc/Inputs/dialog-with-menu.rc b/llvm/test/tools/llvm-rc/Inputs/dialog-with-menu.rc
new file mode 100644
index 000000000000..bb79dca399c2
--- /dev/null
+++ b/llvm/test/tools/llvm-rc/Inputs/dialog-with-menu.rc
@@ -0,0 +1,16 @@
+101 DIALOG 0, 0, 362, 246

+STYLE 0x40l | 0x0004l | 0x0008l | 0x0800l | 0x00020000l |

+    0x00010000l | 0x80000000l | 0x10000000l | 0x02000000l | 0x00C00000l |

+    0x00080000l | 0x00040000l

+CAPTION "MakeNSISW"

+MENU 104

+FONT 8, "MS Shell Dlg"

+BEGIN

+    CONTROL "",202,"RichEdit20A",0x0004l | 0x0040l |

+                    0x0100l | 0x0800l | 0x00008000 |

+                    0x00010000l | 0x00800000l | 0x00200000l,7,22,348,190

+    CONTROL "",-1,"Static",0x00000010l,7,220,346,1

+    LTEXT "",200,7,230,200,12,0x08000000l

+    DEFPUSHBUTTON "Test &Installer",203,230,226,60,15,0x08000000l | 0x00010000l

+    PUSHBUTTON "&Close",2,296,226,49,15,0x00010000l

+END

diff --git a/llvm/test/tools/llvm-rc/dialog-with-menu.test b/llvm/test/tools/llvm-rc/dialog-with-menu.test
new file mode 100644
index 000000000000..2529e9c1722b
--- /dev/null
+++ b/llvm/test/tools/llvm-rc/dialog-with-menu.test
@@ -0,0 +1,32 @@
+; RUN: llvm-rc -no-preprocess /FO %t -- %p/Inputs/dialog-with-menu.rc
+; RUN: llvm-readobj %t | FileCheck %s
+
+CHECK: Resource type (int): DIALOG (ID 5)
+CHECK-NEXT: Resource name (int): 101
+CHECK-NEXT: Data version: 0
+CHECK-NEXT: Memory flags: 0x1030
+CHECK-NEXT: Language ID: 1033
+CHECK-NEXT: Version (major): 0
+CHECK-NEXT: Version (minor): 0
+CHECK-NEXT: Characteristics: 0
+CHECK-NEXT: Data size: 278
+CHECK-NEXT: Data: (
+CHECK-NEXT:   0000: 4C08CF92 00000000 05000000 00006A01  |L.............j.|
+CHECK-NEXT:   0010: F600FFFF 68000000 4D006100 6B006500  |....h...M.a.k.e.|
+CHECK-NEXT:   0020: 4E005300 49005300 57000000 08004D00  |N.S.I.S.W.....M.|
+CHECK-NEXT:   0030: 53002000 53006800 65006C00 6C002000  |S. .S.h.e.l.l. .|
+CHECK-NEXT:   0040: 44006C00 67000000 4489A150 00000000  |D.l.g...D..P....|
+CHECK-NEXT:   0050: 07001600 5C01BE00 CA005200 69006300  |....\.....R.i.c.|
+CHECK-NEXT:   0060: 68004500 64006900 74003200 30004100  |h.E.d.i.t.2.0.A.|
+CHECK-NEXT:   0070: 00000000 00000000 10000050 00000000  |...........P....|
+CHECK-NEXT:   0080: 0700DC00 5A010100 FFFF5300 74006100  |....Z.....S.t.a.|
+CHECK-NEXT:   0090: 74006900 63000000 00000000 00000258  |t.i.c..........X|
+CHECK-NEXT:   00A0: 00000000 0700E600 C8000C00 C800FFFF  |................|
+CHECK-NEXT:   00B0: 82000000 00000000 01000158 00000000  |...........X....|
+CHECK-NEXT:   00C0: E600E200 3C000F00 CB00FFFF 80005400  |....<.........T.|
+CHECK-NEXT:   00D0: 65007300 74002000 26004900 6E007300  |e.s.t. .&.I.n.s.|
+CHECK-NEXT:   00E0: 74006100 6C006C00 65007200 00000000  |t.a.l.l.e.r.....|
+CHECK-NEXT:   00F0: 00000150 00000000 2801E200 31000F00  |...P....(...1...|
+CHECK-NEXT:   0100: 0200FFFF 80002600 43006C00 6F007300  |......&.C.l.o.s.|
+CHECK-NEXT:   0110: 65000000 0000                        |e.....|
+CHECK-NEXT: )
diff --git a/llvm/tools/gold/gold-plugin.cpp b/llvm/tools/gold/gold-plugin.cpp
index b8a33f74bd57..5503f7343cb6 100644
--- a/llvm/tools/gold/gold-plugin.cpp
+++ b/llvm/tools/gold/gold-plugin.cpp
@@ -434,8 +434,10 @@ ld_plugin_status onload(ld_plugin_tv *tv) {
       // FIXME: When binutils 2.31 (containing gold 1.16) is the minimum
       // required version, this should be changed to:
       // get_wrap_symbols = tv->tv_u.tv_get_wrap_symbols;
-      get_wrap_symbols =
-          (ld_plugin_get_wrap_symbols)tv->tv_u.tv_message;
+#pragma GCC diagnostic push
+#pragma GCC diagnostic ignored "-Wcast-function-type"
+      get_wrap_symbols = (ld_plugin_get_wrap_symbols)tv->tv_u.tv_message;
+#pragma GCC diagnostic pop
       break;
     default:
       break;
diff --git a/llvm/tools/llvm-exegesis/lib/BenchmarkRunner.cpp b/llvm/tools/llvm-exegesis/lib/BenchmarkRunner.cpp
index 498308e2edbe..ed53f8fabb17 100644
--- a/llvm/tools/llvm-exegesis/lib/BenchmarkRunner.cpp
+++ b/llvm/tools/llvm-exegesis/lib/BenchmarkRunner.cpp
@@ -18,6 +18,7 @@
 #include "PerfHelper.h"
 #include "SubprocessMemory.h"
 #include "Target.h"
+#include "llvm/ADT/ScopeExit.h"
 #include "llvm/ADT/StringExtras.h"
 #include "llvm/ADT/StringRef.h"
 #include "llvm/ADT/Twine.h"
@@ -283,6 +284,7 @@ private:
                    SmallVectorImpl<int64_t> &CounterValues,
                    ArrayRef<const char *> ValidationCounters,
                    SmallVectorImpl<int64_t> &ValidationCounterValues) const {
+    auto WriteFDClose = make_scope_exit([WriteFD]() { close(WriteFD); });
     const ExegesisTarget &ET = State.getExegesisTarget();
     auto CounterOrError =
         ET.createCounter(CounterName, State, ValidationCounters, ChildPID);
diff --git a/llvm/tools/llvm-exegesis/lib/SubprocessMemory.cpp b/llvm/tools/llvm-exegesis/lib/SubprocessMemory.cpp
index 1d44e09ad61e..89d7b197079e 100644
--- a/llvm/tools/llvm-exegesis/lib/SubprocessMemory.cpp
+++ b/llvm/tools/llvm-exegesis/lib/SubprocessMemory.cpp
@@ -8,6 +8,7 @@
 
 #include "SubprocessMemory.h"
 #include "Error.h"
+#include "llvm/ADT/ScopeExit.h"
 #include "llvm/Support/Error.h"
 #include "llvm/Support/FormatVariadic.h"
 #include <cerrno>
@@ -56,6 +57,8 @@ Error SubprocessMemory::initializeSubprocessMemory(pid_t ProcessID) {
     return make_error<Failure>(
         "Failed to create shared memory object for auxiliary memory: " +
         Twine(strerror(errno)));
+  auto AuxiliaryMemoryFDClose =
+      make_scope_exit([AuxiliaryMemoryFD]() { close(AuxiliaryMemoryFD); });
   if (ftruncate(AuxiliaryMemoryFD, AuxiliaryMemorySize) != 0) {
     return make_error<Failure>("Truncating the auxiliary memory failed: " +
                                Twine(strerror(errno)));
@@ -78,6 +81,8 @@ Error SubprocessMemory::addMemoryDefinition(
       return make_error<Failure>(
           "Failed to create shared memory object for memory definition: " +
           Twine(strerror(errno)));
+    auto SharedMemoryFDClose =
+        make_scope_exit([SharedMemoryFD]() { close(SharedMemoryFD); });
     if (ftruncate(SharedMemoryFD, MemVal.SizeBytes) != 0) {
       return make_error<Failure>("Truncating a memory definiton failed: " +
                                  Twine(strerror(errno)));
@@ -138,7 +143,7 @@ Expected<int> SubprocessMemory::setupAuxiliaryMemoryInSubprocess(
 }
 
 SubprocessMemory::~SubprocessMemory() {
-  for (std::string SharedMemoryName : SharedMemoryNames) {
+  for (const std::string &SharedMemoryName : SharedMemoryNames) {
     if (shm_unlink(SharedMemoryName.c_str()) != 0) {
       errs() << "Failed to unlink shared memory section: " << strerror(errno)
              << "\n";
diff --git a/llvm/tools/llvm-jitlink/llvm-jitlink.cpp b/llvm/tools/llvm-jitlink/llvm-jitlink.cpp
index 09b2a5900eb0..bff05b9ca4be 100644
--- a/llvm/tools/llvm-jitlink/llvm-jitlink.cpp
+++ b/llvm/tools/llvm-jitlink/llvm-jitlink.cpp
@@ -807,8 +807,8 @@ static Expected<std::unique_ptr<ExecutorProcessControl>> launchExecutor() {
     S.CreateMemoryManager = createSharedMemoryManager;
 
   return SimpleRemoteEPC::Create<FDSimpleRemoteEPCTransport>(
-      std::make_unique<DynamicThreadPoolTaskDispatcher>(), std::move(S),
-      FromExecutor[ReadEnd], ToExecutor[WriteEnd]);
+      std::make_unique<DynamicThreadPoolTaskDispatcher>(std::nullopt),
+      std::move(S), FromExecutor[ReadEnd], ToExecutor[WriteEnd]);
 #endif
 }
 
@@ -897,7 +897,7 @@ static Expected<std::unique_ptr<ExecutorProcessControl>> connectToExecutor() {
     S.CreateMemoryManager = createSharedMemoryManager;
 
   return SimpleRemoteEPC::Create<FDSimpleRemoteEPCTransport>(
-      std::make_unique<DynamicThreadPoolTaskDispatcher>(),
+      std::make_unique<DynamicThreadPoolTaskDispatcher>(std::nullopt),
       std::move(S), *SockFD, *SockFD);
 #endif
 }
diff --git a/llvm/tools/llvm-lto2/llvm-lto2.cpp b/llvm/tools/llvm-lto2/llvm-lto2.cpp
index faed9ff9939b..5dd961a603c9 100644
--- a/llvm/tools/llvm-lto2/llvm-lto2.cpp
+++ b/llvm/tools/llvm-lto2/llvm-lto2.cpp
@@ -251,10 +251,9 @@ static int run(int argc, char **argv) {
   // resolutions and apply them in the order observed.
   std::map<std::pair<std::string, std::string>, std::list<SymbolResolution>>
       CommandLineResolutions;
-  for (std::string R : SymbolResolutions) {
-    StringRef Rest = R;
-    StringRef FileName, SymbolName;
-    std::tie(FileName, Rest) = Rest.split(',');
+  for (StringRef R : SymbolResolutions) {
+    StringRef Rest, FileName, SymbolName;
+    std::tie(FileName, Rest) = R.split(',');
     if (Rest.empty()) {
       llvm::errs() << "invalid resolution: " << R << '\n';
       return 1;
diff --git a/llvm/tools/llvm-mca/CodeRegion.h b/llvm/tools/llvm-mca/CodeRegion.h
index ce107fd8f3b6..5a2e8baa1f3e 100644
--- a/llvm/tools/llvm-mca/CodeRegion.h
+++ b/llvm/tools/llvm-mca/CodeRegion.h
@@ -59,6 +59,7 @@
 #define LLVM_TOOLS_LLVM_MCA_CODEREGION_H
 
 #include "llvm/ADT/ArrayRef.h"
+#include "llvm/ADT/SmallPtrSet.h"
 #include "llvm/ADT/SmallVector.h"
 #include "llvm/ADT/StringMap.h"
 #include "llvm/ADT/StringRef.h"
@@ -97,6 +98,20 @@ public:
     Instructions.emplace_back(Instruction);
   }
 
+  // Remove the given instructions from the set, for unsupported instructions
+  // being skipped. Returns an ArrayRef for the updated vector of Instructions.
+  [[nodiscard]] llvm::ArrayRef<llvm::MCInst>
+  dropInstructions(const llvm::SmallPtrSetImpl<const llvm::MCInst *> &Insts) {
+    if (Insts.empty())
+      return Instructions;
+    Instructions.erase(std::remove_if(Instructions.begin(), Instructions.end(),
+                                      [&Insts](const llvm::MCInst &Inst) {
+                                        return Insts.contains(&Inst);
+                                      }),
+                       Instructions.end());
+    return Instructions;
+  }
+
   llvm::SMLoc startLoc() const { return RangeStart; }
   llvm::SMLoc endLoc() const { return RangeEnd; }
 
diff --git a/llvm/tools/llvm-mca/llvm-mca.cpp b/llvm/tools/llvm-mca/llvm-mca.cpp
index eb71cffba6dd..e037c06b12a3 100644
--- a/llvm/tools/llvm-mca/llvm-mca.cpp
+++ b/llvm/tools/llvm-mca/llvm-mca.cpp
@@ -237,6 +237,11 @@ static cl::opt<bool> DisableInstrumentManager(
              "ignores instruments.)."),
     cl::cat(ViewOptions), cl::init(false));
 
+static cl::opt<bool> SkipUnsupportedInstructions(
+    "skip-unsupported-instructions",
+    cl::desc("Make unsupported instruction errors into warnings."),
+    cl::cat(ViewOptions), cl::init(false));
+
 namespace {
 
 const Target *getTarget(const char *ProgName) {
@@ -558,6 +563,7 @@ int main(int argc, char **argv) {
   assert(MAB && "Unable to create asm backend!");
 
   json::Object JSONOutput;
+  int NonEmptyRegions = 0;
   for (const std::unique_ptr<mca::AnalysisRegion> &Region : Regions) {
     // Skip empty code regions.
     if (Region->empty())
@@ -571,14 +577,13 @@ int main(int argc, char **argv) {
 
     IPP->resetState();
 
-    DenseMap<const MCInst *, SmallVector<mca::Instrument *>>
-        InstToInstruments;
+    DenseMap<const MCInst *, SmallVector<mca::Instrument *>> InstToInstruments;
     SmallVector<std::unique_ptr<mca::Instruction>> LoweredSequence;
+    SmallPtrSet<const MCInst *, 16> DroppedInsts;
     for (const MCInst &MCI : Insts) {
       SMLoc Loc = MCI.getLoc();
       const SmallVector<mca::Instrument *> Instruments =
           InstrumentRegions.getActiveInstruments(Loc);
-      InstToInstruments.insert({&MCI, Instruments});
 
       Expected<std::unique_ptr<mca::Instruction>> Inst =
           IB.createInstruction(MCI, Instruments);
@@ -588,7 +593,15 @@ int main(int argc, char **argv) {
                 [&IP, &STI](const mca::InstructionError<MCInst> &IE) {
                   std::string InstructionStr;
                   raw_string_ostream SS(InstructionStr);
-                  WithColor::error() << IE.Message << '\n';
+                  if (SkipUnsupportedInstructions)
+                    WithColor::warning()
+                        << IE.Message
+                        << ", skipping with -skip-unsupported-instructions, "
+                           "note accuracy will be impacted:\n";
+                  else
+                    WithColor::error()
+                        << IE.Message
+                        << ", use -skip-unsupported-instructions to ignore.\n";
                   IP->printInst(&IE.Inst, 0, "", *STI, SS);
                   SS.flush();
                   WithColor::note()
@@ -597,14 +610,25 @@ int main(int argc, char **argv) {
           // Default case.
           WithColor::error() << toString(std::move(NewE));
         }
+        if (SkipUnsupportedInstructions) {
+          DroppedInsts.insert(&MCI);
+          continue;
+        }
         return 1;
       }
 
       IPP->postProcessInstruction(Inst.get(), MCI);
-
+      InstToInstruments.insert({&MCI, Instruments});
       LoweredSequence.emplace_back(std::move(Inst.get()));
     }
 
+    Insts = Region->dropInstructions(DroppedInsts);
+
+    // Skip empty regions.
+    if (Insts.empty())
+      continue;
+    NonEmptyRegions++;
+
     mca::CircularSourceMgr S(LoweredSequence,
                              PrintInstructionTables ? 1 : Iterations);
 
@@ -759,6 +783,11 @@ int main(int argc, char **argv) {
     ++RegionIdx;
   }
 
+  if (NonEmptyRegions == 0) {
+    WithColor::error() << "no assembly instructions found.\n";
+    return 1;
+  }
+
   if (PrintJson)
     TOF->os() << formatv("{0:2}", json::Value(std::move(JSONOutput))) << "\n";
 
diff --git a/llvm/tools/llvm-objdump/llvm-objdump.cpp b/llvm/tools/llvm-objdump/llvm-objdump.cpp
index 339822e4adcd..675364a1c1bc 100644
--- a/llvm/tools/llvm-objdump/llvm-objdump.cpp
+++ b/llvm/tools/llvm-objdump/llvm-objdump.cpp
@@ -947,6 +947,55 @@ public:
 };
 AArch64PrettyPrinter AArch64PrettyPrinterInst;
 
+class RISCVPrettyPrinter : public PrettyPrinter {
+public:
+  void printInst(MCInstPrinter &IP, const MCInst *MI, ArrayRef<uint8_t> Bytes,
+                 object::SectionedAddress Address, formatted_raw_ostream &OS,
+                 StringRef Annot, MCSubtargetInfo const &STI, SourcePrinter *SP,
+                 StringRef ObjectFilename, std::vector<RelocationRef> *Rels,
+                 LiveVariablePrinter &LVP) override {
+    if (SP && (PrintSource || PrintLines))
+      SP->printSourceLine(OS, Address, ObjectFilename, LVP);
+    LVP.printBetweenInsts(OS, false);
+
+    size_t Start = OS.tell();
+    if (LeadingAddr)
+      OS << format("%8" PRIx64 ":", Address.Address);
+    if (ShowRawInsn) {
+      size_t Pos = 0, End = Bytes.size();
+      if (End % 4 == 0) {
+        // 32-bit and 64-bit instructions.
+        for (; Pos + 4 <= End; Pos += 4)
+          OS << ' '
+             << format_hex_no_prefix(
+                    llvm::support::endian::read<uint32_t>(
+                        Bytes.data() + Pos, llvm::endianness::little),
+                    8);
+      } else if (End % 2 == 0) {
+        // 16-bit and 48-bits instructions.
+        for (; Pos + 2 <= End; Pos += 2)
+          OS << ' '
+             << format_hex_no_prefix(
+                    llvm::support::endian::read<uint16_t>(
+                        Bytes.data() + Pos, llvm::endianness::little),
+                    4);
+      }
+      if (Pos < End) {
+        OS << ' ';
+        dumpBytes(Bytes.slice(Pos), OS);
+      }
+    }
+
+    AlignToInstStartColumn(Start, STI, OS);
+
+    if (MI) {
+      IP.printInst(MI, Address.Address, "", STI, OS);
+    } else
+      OS << "\t<unknown>";
+  }
+};
+RISCVPrettyPrinter RISCVPrettyPrinterInst;
+
 PrettyPrinter &selectPrettyPrinter(Triple const &Triple) {
   switch(Triple.getArch()) {
   default:
@@ -967,6 +1016,9 @@ PrettyPrinter &selectPrettyPrinter(Triple const &Triple) {
   case Triple::aarch64_be:
   case Triple::aarch64_32:
     return AArch64PrettyPrinterInst;
+  case Triple::riscv32:
+  case Triple::riscv64:
+    return RISCVPrettyPrinterInst;
   }
 }
 
diff --git a/llvm/tools/llvm-profdata/llvm-profdata.cpp b/llvm/tools/llvm-profdata/llvm-profdata.cpp
index 78daf9f7dc10..ec046ebfab13 100644
--- a/llvm/tools/llvm-profdata/llvm-profdata.cpp
+++ b/llvm/tools/llvm-profdata/llvm-profdata.cpp
@@ -308,6 +308,10 @@ cl::opt<memprof::IndexedVersion> MemProfVersionRequested(
                clEnumValN(memprof::Version1, "1", "version 1"),
                clEnumValN(memprof::Version2, "2", "version 2")));
 
+cl::opt<bool> MemProfFullSchema(
+    "memprof-full-schema", cl::Hidden, cl::sub(MergeSubcommand),
+    cl::desc("Use the full schema for serialization"), cl::init(false));
+
 // Options specific to overlap subcommand.
 cl::opt<std::string> BaseFilename(cl::Positional, cl::Required,
                                   cl::desc("<base profile file>"),
@@ -600,7 +604,7 @@ struct WriterContext {
                 SmallSet<instrprof_error, 4> &WriterErrorCodes,
                 uint64_t ReservoirSize = 0, uint64_t MaxTraceLength = 0)
       : Writer(IsSparse, ReservoirSize, MaxTraceLength, DoWritePrevVersion,
-               MemProfVersionRequested),
+               MemProfVersionRequested, MemProfFullSchema),
         ErrLock(ErrLock), WriterErrorCodes(WriterErrorCodes) {}
 };
 
diff --git a/llvm/tools/llvm-rc/ResourceFileWriter.cpp b/llvm/tools/llvm-rc/ResourceFileWriter.cpp
index d507525970ec..85b59532bb83 100644
--- a/llvm/tools/llvm-rc/ResourceFileWriter.cpp
+++ b/llvm/tools/llvm-rc/ResourceFileWriter.cpp
@@ -550,6 +550,11 @@ Error ResourceFileWriter::visitVersionStmt(const VersionStmt *Stmt) {
   return Error::success();
 }
 
+Error ResourceFileWriter::visitMenuStmt(const MenuStmt *Stmt) {
+  ObjectData.Menu = Stmt->Value;
+  return Error::success();
+}
+
 Error ResourceFileWriter::writeResource(
     const RCResource *Res,
     Error (ResourceFileWriter::*BodyWriter)(const RCResource *)) {
@@ -1132,9 +1137,8 @@ Error ResourceFileWriter::writeDialogBody(const RCResource *Base) {
            ulittle16_t(Res->Height)};
   writeObject(Middle);
 
-  // MENU field. As of now, we don't keep them in the state and can peacefully
-  // think there is no menu attached to the dialog.
-  writeInt<uint16_t>(0);
+  // MENU field.
+  RETURN_IF_ERROR(writeIntOrString(ObjectData.Menu));
 
   // Window CLASS field.
   RETURN_IF_ERROR(writeIntOrString(ObjectData.Class));
diff --git a/llvm/tools/llvm-rc/ResourceFileWriter.h b/llvm/tools/llvm-rc/ResourceFileWriter.h
index 9413a0eecdac..82d3e3b9e9e8 100644
--- a/llvm/tools/llvm-rc/ResourceFileWriter.h
+++ b/llvm/tools/llvm-rc/ResourceFileWriter.h
@@ -16,6 +16,7 @@
 #include "ResourceScriptStmt.h"
 #include "ResourceVisitor.h"
 
+#include "llvm/ADT/StringRef.h"
 #include "llvm/Support/Endian.h"
 
 namespace llvm {
@@ -68,6 +69,7 @@ public:
   Error visitLanguageStmt(const LanguageResource *) override;
   Error visitStyleStmt(const StyleStmt *) override;
   Error visitVersionStmt(const VersionStmt *) override;
+  Error visitMenuStmt(const MenuStmt *) override;
 
   // Stringtables are output at the end of .res file. We need a separate
   // function to do it.
@@ -92,10 +94,11 @@ public:
     };
     std::optional<FontInfo> Font;
     IntOrString Class;
+    IntOrString Menu;
 
     ObjectInfo()
         : LanguageInfo(0), Characteristics(0), VersionInfo(0),
-          Class(StringRef()) {}
+          Class(StringRef()), Menu(StringRef()) {}
   } ObjectData;
 
   struct StringTableInfo {
diff --git a/llvm/tools/llvm-rc/ResourceScriptParser.cpp b/llvm/tools/llvm-rc/ResourceScriptParser.cpp
index 4f02fa502d24..69798152c1f2 100644
--- a/llvm/tools/llvm-rc/ResourceScriptParser.cpp
+++ b/llvm/tools/llvm-rc/ResourceScriptParser.cpp
@@ -430,6 +430,8 @@ RCParser::parseSingleOptionalStatement(OptStmtType StmtsType) {
       return parseFontStmt(StmtsType);
     if (TypeToken->equals_insensitive("STYLE"))
       return parseStyleStmt();
+    if (TypeToken->equals_insensitive("MENU"))
+      return parseMenuStmt();
   }
 
   return getExpectedError("optional statement type, BEGIN or '{'",
@@ -965,6 +967,11 @@ RCParser::ParseOptionType RCParser::parseExStyleStmt() {
   return std::make_unique<ExStyleStmt>(*Arg);
 }
 
+RCParser::ParseOptionType RCParser::parseMenuStmt() {
+  ASSIGN_OR_RETURN(Arg, readIntOrString());
+  return std::make_unique<MenuStmt>(*Arg);
+}
+
 Error RCParser::getExpectedError(const Twine &Message, bool IsAlreadyRead) {
   return make_error<ParserError>(
       Message, IsAlreadyRead ? std::prev(CurLoc) : CurLoc, End);
diff --git a/llvm/tools/llvm-rc/ResourceScriptParser.h b/llvm/tools/llvm-rc/ResourceScriptParser.h
index 603afd8d73fb..aa7f847187c4 100644
--- a/llvm/tools/llvm-rc/ResourceScriptParser.h
+++ b/llvm/tools/llvm-rc/ResourceScriptParser.h
@@ -176,6 +176,7 @@ private:
   ParseOptionType parseExStyleStmt();
   ParseOptionType parseFontStmt(OptStmtType DialogType);
   ParseOptionType parseStyleStmt();
+  ParseOptionType parseMenuStmt();
 
   // Raises an error. If IsAlreadyRead = false (default), this complains about
   // the token that couldn't be parsed. If the flag is on, this complains about
diff --git a/llvm/tools/llvm-rc/ResourceScriptStmt.cpp b/llvm/tools/llvm-rc/ResourceScriptStmt.cpp
index 62df7999252f..a7f3df0863e7 100644
--- a/llvm/tools/llvm-rc/ResourceScriptStmt.cpp
+++ b/llvm/tools/llvm-rc/ResourceScriptStmt.cpp
@@ -309,5 +309,9 @@ raw_ostream &ExStyleStmt::log(raw_ostream &OS) const {
   return OS << "ExStyle: " << Value << "\n";
 }
 
+raw_ostream &MenuStmt::log(raw_ostream &OS) const {
+  return OS << "Menu: " << Value << "\n";
+}
+
 } // namespace rc
 } // namespace llvm
diff --git a/llvm/tools/llvm-rc/ResourceScriptStmt.h b/llvm/tools/llvm-rc/ResourceScriptStmt.h
index 70e7cec9cb84..05865e582859 100644
--- a/llvm/tools/llvm-rc/ResourceScriptStmt.h
+++ b/llvm/tools/llvm-rc/ResourceScriptStmt.h
@@ -993,6 +993,19 @@ public:
   Error visit(Visitor *V) const override { return V->visitExStyleStmt(this); }
 };
 
+// MENU optional statement.
+//
+// Ref: https://learn.microsoft.com/en-us/windows/win32/menurc/menu-statement
+class MenuStmt : public OptionalStmt {
+public:
+  IntOrString Value;
+
+  MenuStmt(IntOrString NameOrId) : Value(NameOrId) {}
+  raw_ostream &log(raw_ostream &) const override;
+  Twine getResourceTypeName() const override { return "MENU"; }
+  Error visit(Visitor *V) const override { return V->visitMenuStmt(this); }
+};
+
 // CLASS optional statement.
 //
 // Ref: msdn.microsoft.com/en-us/library/windows/desktop/aa380883(v=vs.85).aspx
diff --git a/llvm/tools/llvm-rc/ResourceVisitor.h b/llvm/tools/llvm-rc/ResourceVisitor.h
index a950cd7555ec..a121a0a507c2 100644
--- a/llvm/tools/llvm-rc/ResourceVisitor.h
+++ b/llvm/tools/llvm-rc/ResourceVisitor.h
@@ -28,6 +28,7 @@ class FontStmt;
 class LanguageResource;
 class StyleStmt;
 class VersionStmt;
+class MenuStmt;
 
 class Visitor {
 public:
@@ -52,6 +53,7 @@ public:
   virtual Error visitLanguageStmt(const LanguageResource *) = 0;
   virtual Error visitStyleStmt(const StyleStmt *) = 0;
   virtual Error visitVersionStmt(const VersionStmt *) = 0;
+  virtual Error visitMenuStmt(const MenuStmt *) = 0;
 
   virtual ~Visitor() {}
 };
diff --git a/llvm/tools/llvm-readtapi/llvm-readtapi.cpp b/llvm/tools/llvm-readtapi/llvm-readtapi.cpp
index 80064ed98485..1f183975d948 100644
--- a/llvm/tools/llvm-readtapi/llvm-readtapi.cpp
+++ b/llvm/tools/llvm-readtapi/llvm-readtapi.cpp
@@ -133,9 +133,7 @@ getInterfaceFile(const StringRef Filename, bool ResetBanner = true) {
   std::unique_ptr<InterfaceFile> IF;
   switch (identify_magic(Buffer->getBuffer())) {
   case file_magic::macho_dynamically_linked_shared_lib:
-    LLVM_FALLTHROUGH;
   case file_magic::macho_dynamically_linked_shared_lib_stub:
-    LLVM_FALLTHROUGH;
   case file_magic::macho_universal_binary:
     IF = ExitOnErr(DylibReader::get(Buffer->getMemBufferRef()));
     break;
diff --git a/llvm/unittests/ADT/StringRefTest.cpp b/llvm/unittests/ADT/StringRefTest.cpp
index 8df71e8ad033..fa537e816fc8 100644
--- a/llvm/unittests/ADT/StringRefTest.cpp
+++ b/llvm/unittests/ADT/StringRefTest.cpp
@@ -368,6 +368,8 @@ TEST(StringRefTest, StartsWith) {
   EXPECT_TRUE(Str.starts_with("he"));
   EXPECT_FALSE(Str.starts_with("helloworld"));
   EXPECT_FALSE(Str.starts_with("hi"));
+  EXPECT_TRUE(Str.starts_with('h'));
+  EXPECT_FALSE(Str.starts_with('i'));
 }
 
 TEST(StringRefTest, StartsWithInsensitive) {
@@ -421,6 +423,8 @@ TEST(StringRefTest, EndsWith) {
   EXPECT_FALSE(Str.ends_with("helloworld"));
   EXPECT_FALSE(Str.ends_with("worldhello"));
   EXPECT_FALSE(Str.ends_with("so"));
+  EXPECT_TRUE(Str.ends_with('o'));
+  EXPECT_FALSE(Str.ends_with('p'));
 }
 
 TEST(StringRefTest, EndsWithInsensitive) {
diff --git a/llvm/unittests/Analysis/LoadsTest.cpp b/llvm/unittests/Analysis/LoadsTest.cpp
index 0111cfeefa41..5da3feaf762f 100644
--- a/llvm/unittests/Analysis/LoadsTest.cpp
+++ b/llvm/unittests/Analysis/LoadsTest.cpp
@@ -68,35 +68,49 @@ TEST(LoadsTest, CanReplacePointersIfEqual) {
                                       R"IR(
 @y = common global [1 x i32] zeroinitializer, align 4
 @x = common global [1 x i32] zeroinitializer, align 4
-
 declare void @use(i32*)
 
-define void @f(i32* %p) {
+define void @f(i32* %p1, i32* %p2, i64 %i) {
   call void @use(i32* getelementptr inbounds ([1 x i32], [1 x i32]* @y, i64 0, i64 0))
-  call void @use(i32* getelementptr inbounds (i32, i32* getelementptr inbounds ([1 x i32], [1 x i32]* @x, i64 0, i64 0), i64 1))
+
+  %p1_idx = getelementptr inbounds i32, i32* %p1, i64 %i
+  call void @use(i32* %p1_idx)
+
+  %icmp = icmp eq i32* %p1, getelementptr inbounds ([1 x i32], [1 x i32]* @y, i64 0, i64 0)
+  %ptrInt = ptrtoint i32* %p1 to i64
   ret void
 }
 )IR");
-  const auto &DL = M->getDataLayout();
+  const DataLayout &DL = M->getDataLayout();
   auto *GV = M->getNamedValue("f");
   ASSERT_TRUE(GV);
   auto *F = dyn_cast<Function>(GV);
   ASSERT_TRUE(F);
 
-  // NOTE: the implementation of canReplacePointersIfEqual is incomplete.
-  // Currently the only the cases it returns false for are really sound and
-  // returning true means unknown.
-  Value *P = &*F->arg_begin();
+  Value *P1 = &*F->arg_begin();
+  Value *P2 = F->getArg(1);
+  Value *NullPtr = Constant::getNullValue(P1->getType());
   auto InstIter = F->front().begin();
-  Value *ConstDerefPtr = *cast<CallInst>(&*InstIter)->arg_begin();
-  // ConstDerefPtr is a constant pointer that is provably de-referenceable. We
-  // can replace an arbitrary pointer with it.
-  EXPECT_TRUE(canReplacePointersIfEqual(P, ConstDerefPtr, DL, nullptr));
+  CallInst *UserOfY = cast<CallInst>(&*InstIter);
+  Value *ConstDerefPtr = UserOfY->getArgOperand(0);
+  // We cannot replace two pointers in arbitrary instructions unless we are
+  // replacing with null, a constant dereferencable pointer or they have the
+  // same underlying object.
+  EXPECT_FALSE(canReplacePointersIfEqual(ConstDerefPtr, P1, DL));
+  EXPECT_FALSE(canReplacePointersIfEqual(P1, P2, DL));
+  EXPECT_TRUE(canReplacePointersIfEqual(P1, ConstDerefPtr, DL));
+  EXPECT_TRUE(canReplacePointersIfEqual(P1, NullPtr, DL));
+
+  GetElementPtrInst *BasedOnP1 = cast<GetElementPtrInst>(&*++InstIter);
+  EXPECT_TRUE(canReplacePointersIfEqual(BasedOnP1, P1, DL));
+  EXPECT_FALSE(canReplacePointersIfEqual(BasedOnP1, P2, DL));
 
-  ++InstIter;
-  Value *ConstUnDerefPtr = *cast<CallInst>(&*InstIter)->arg_begin();
-  // ConstUndDerefPtr is a constant pointer that is provably not
-  // de-referenceable. We cannot replace an arbitrary pointer with it.
-  EXPECT_FALSE(
-      canReplacePointersIfEqual(ConstDerefPtr, ConstUnDerefPtr, DL, nullptr));
+  // We can replace two arbitrary pointers in icmp and ptrtoint instructions.
+  auto P1UseIter = P1->use_begin();
+  const Use &PtrToIntUse = *P1UseIter;
+  const Use &IcmpUse = *++P1UseIter;
+  const Use &GEPUse = *++P1UseIter;
+  EXPECT_FALSE(canReplacePointersInUseIfEqual(GEPUse, P2, DL));
+  EXPECT_TRUE(canReplacePointersInUseIfEqual(PtrToIntUse, P2, DL));
+  EXPECT_TRUE(canReplacePointersInUseIfEqual(IcmpUse, P2, DL));
 }
diff --git a/llvm/unittests/BinaryFormat/DwarfTest.cpp b/llvm/unittests/BinaryFormat/DwarfTest.cpp
index 2fff8657939b..684e59fa2785 100644
--- a/llvm/unittests/BinaryFormat/DwarfTest.cpp
+++ b/llvm/unittests/BinaryFormat/DwarfTest.cpp
@@ -204,4 +204,19 @@ TEST(DwarfTest, format_provider) {
   EXPECT_EQ("DW_OP_lit0", formatv("{0}", DW_OP_lit0).str());
   EXPECT_EQ("DW_OP_unknown_ff", formatv("{0}", DW_OP_hi_user).str());
 }
+
+TEST(DwarfTest, lname) {
+  auto roundtrip = [](llvm::dwarf::SourceLanguage sl) {
+    auto name_version = toDW_LNAME(sl);
+    // Ignore ones without a defined mapping.
+    if (sl == DW_LANG_Mips_Assembler || sl == DW_LANG_GOOGLE_RenderScript ||
+        !name_version.has_value())
+      return sl;
+    return dwarf::toDW_LANG(name_version->first, name_version->second)
+        .value_or(sl);
+  };
+#define HANDLE_DW_LANG(ID, NAME, LOWER_BOUND, VERSION, VENDOR)                 \
+  EXPECT_EQ(roundtrip(DW_LANG_##NAME), DW_LANG_##NAME);
+#include "llvm/BinaryFormat/Dwarf.def"
+}
 } // end namespace
diff --git a/llvm/unittests/CodeGen/GlobalISel/KnownBitsTest.cpp b/llvm/unittests/CodeGen/GlobalISel/KnownBitsTest.cpp
index ef80eed8d180..34a36ba68d7c 100644
--- a/llvm/unittests/CodeGen/GlobalISel/KnownBitsTest.cpp
+++ b/llvm/unittests/CodeGen/GlobalISel/KnownBitsTest.cpp
@@ -745,6 +745,120 @@ TEST_F(AArch64GISelMITest, TestNumSignBitsConstant) {
   EXPECT_EQ(3u, Info.computeNumSignBits(CopyRegNeg32));
 }
 
+TEST_F(AArch64GISelMITest, TestNumSignBitsXOR) {
+  StringRef MIRString = "  %c1:_(s8) = G_CONSTANT i8 1\n"
+                        "  %cn1:_(s8) = G_CONSTANT i8 -1\n"
+                        "  %c127:_(s8) = G_CONSTANT i8 127\n"
+                        "  %c32:_(s8) = G_CONSTANT i8 32\n"
+                        "  %cn32:_(s8) = G_CONSTANT i8 -32\n"
+
+                        "  %xor1:_(s8) = G_XOR %c1, %cn1\n"
+                        "  %Copy1:_(s8) = COPY %xor1\n"
+
+                        "  %xor2:_(s8) = G_XOR %c1, %c32\n"
+                        "  %Copy2:_(s8) = COPY %xor2\n"
+
+                        "  %xor3:_(s8) = G_XOR %c32, %c127\n"
+                        "  %Copy3:_(s8) = COPY %xor3\n"
+
+                        "  %xor4:_(s8) = G_XOR %cn32, %c127\n"
+                        "  %Copy4:_(s8) = COPY %xor4\n"
+
+                        "  %xor5:_(s8) = G_XOR %c127, %cn32\n"
+                        "  %Copy5:_(s8) = COPY %xor5\n";
+  setUp(MIRString);
+  if (!TM)
+    GTEST_SKIP();
+  Register Copy1 = Copies[Copies.size() - 5];
+  Register Copy2 = Copies[Copies.size() - 4];
+  Register Copy3 = Copies[Copies.size() - 3];
+  Register Copy4 = Copies[Copies.size() - 2];
+  Register Copy5 = Copies[Copies.size() - 1];
+
+  GISelKnownBits Info(*MF);
+  EXPECT_EQ(7u, Info.computeNumSignBits(Copy1));
+  EXPECT_EQ(2u, Info.computeNumSignBits(Copy2));
+  EXPECT_EQ(1u, Info.computeNumSignBits(Copy3));
+  EXPECT_EQ(1u, Info.computeNumSignBits(Copy4));
+  EXPECT_EQ(1u, Info.computeNumSignBits(Copy5));
+}
+
+TEST_F(AArch64GISelMITest, TestNumSignBitsOR) {
+  StringRef MIRString = "  %c1:_(s8) = G_CONSTANT i8 1\n"
+                        "  %cn1:_(s8) = G_CONSTANT i8 -1\n"
+                        "  %c127:_(s8) = G_CONSTANT i8 127\n"
+                        "  %c32:_(s8) = G_CONSTANT i8 32\n"
+                        "  %cn32:_(s8) = G_CONSTANT i8 -32\n"
+
+                        "  %or1:_(s8) = G_OR %c1, %cn1\n"
+                        "  %Copy1:_(s8) = COPY %or1\n"
+
+                        "  %or2:_(s8) = G_OR %c1, %c32\n"
+                        "  %Copy2:_(s8) = COPY %or2\n"
+
+                        "  %or3:_(s8) = G_OR %c32, %c127\n"
+                        "  %Copy3:_(s8) = COPY %or3\n"
+
+                        "  %or4:_(s8) = G_OR %cn32, %c127\n"
+                        "  %Copy4:_(s8) = COPY %or4\n"
+
+                        "  %or5:_(s8) = G_OR %c127, %cn32\n"
+                        "  %Copy5:_(s8) = COPY %or5\n";
+  setUp(MIRString);
+  if (!TM)
+    GTEST_SKIP();
+  Register Copy1 = Copies[Copies.size() - 5];
+  Register Copy2 = Copies[Copies.size() - 4];
+  Register Copy3 = Copies[Copies.size() - 3];
+  Register Copy4 = Copies[Copies.size() - 2];
+  Register Copy5 = Copies[Copies.size() - 1];
+
+  GISelKnownBits Info(*MF);
+  EXPECT_EQ(8u, Info.computeNumSignBits(Copy1));
+  EXPECT_EQ(2u, Info.computeNumSignBits(Copy2));
+  EXPECT_EQ(1u, Info.computeNumSignBits(Copy3));
+  EXPECT_EQ(8u, Info.computeNumSignBits(Copy4));
+  EXPECT_EQ(8u, Info.computeNumSignBits(Copy5));
+}
+
+TEST_F(AArch64GISelMITest, TestNumSignBitsAND) {
+  StringRef MIRString = "  %c1:_(s8) = G_CONSTANT i8 1\n"
+                        "  %cn1:_(s8) = G_CONSTANT i8 -1\n"
+                        "  %c127:_(s8) = G_CONSTANT i8 127\n"
+                        "  %c32:_(s8) = G_CONSTANT i8 32\n"
+                        "  %cn32:_(s8) = G_CONSTANT i8 -32\n"
+
+                        "  %and1:_(s8) = G_AND %c1, %cn1\n"
+                        "  %Copy1:_(s8) = COPY %and1\n"
+
+                        "  %and2:_(s8) = G_AND %c1, %c32\n"
+                        "  %Copy2:_(s8) = COPY %and2\n"
+
+                        "  %and3:_(s8) = G_AND %c32, %c127\n"
+                        "  %Copy3:_(s8) = COPY %and3\n"
+
+                        "  %and4:_(s8) = G_AND %cn32, %c127\n"
+                        "  %Copy4:_(s8) = COPY %and4\n"
+
+                        "  %and5:_(s8) = G_AND %c127, %cn32\n"
+                        "  %Copy5:_(s8) = COPY %and5\n";
+  setUp(MIRString);
+  if (!TM)
+    GTEST_SKIP();
+  Register Copy1 = Copies[Copies.size() - 5];
+  Register Copy2 = Copies[Copies.size() - 4];
+  Register Copy3 = Copies[Copies.size() - 3];
+  Register Copy4 = Copies[Copies.size() - 2];
+  Register Copy5 = Copies[Copies.size() - 1];
+
+  GISelKnownBits Info(*MF);
+  EXPECT_EQ(7u, Info.computeNumSignBits(Copy1));
+  EXPECT_EQ(8u, Info.computeNumSignBits(Copy2));
+  EXPECT_EQ(2u, Info.computeNumSignBits(Copy3));
+  EXPECT_EQ(1u, Info.computeNumSignBits(Copy4));
+  EXPECT_EQ(1u, Info.computeNumSignBits(Copy5));
+}
+
 TEST_F(AArch64GISelMITest, TestNumSignBitsSext) {
   StringRef MIRString = "  %3:_(p0) = G_IMPLICIT_DEF\n"
                         "  %4:_(s8) = G_LOAD %3 :: (load (s8))\n"
diff --git a/llvm/unittests/CodeGen/MachineInstrTest.cpp b/llvm/unittests/CodeGen/MachineInstrTest.cpp
index 49da0c38eefd..8ea12a6ec645 100644
--- a/llvm/unittests/CodeGen/MachineInstrTest.cpp
+++ b/llvm/unittests/CodeGen/MachineInstrTest.cpp
@@ -18,6 +18,7 @@
 #include "llvm/CodeGen/TargetSubtargetInfo.h"
 #include "llvm/IR/DebugInfoMetadata.h"
 #include "llvm/IR/IRBuilder.h"
+#include "llvm/IR/MemoryModelRelaxationAnnotations.h"
 #include "llvm/IR/ModuleSlotTracker.h"
 #include "llvm/MC/MCAsmInfo.h"
 #include "llvm/MC/MCSymbol.h"
@@ -277,12 +278,14 @@ TEST(MachineInstrExtraInfo, AddExtraInfo) {
   MCSymbol *Sym2 = MC->createTempSymbol("post_label", false);
   MDNode *HAM = MDNode::getDistinct(Ctx, std::nullopt);
   MDNode *PCS = MDNode::getDistinct(Ctx, std::nullopt);
+  MDNode *MMRA = MMRAMetadata::getTagMD(Ctx, "foo", "bar");
 
   ASSERT_TRUE(MI->memoperands_empty());
   ASSERT_FALSE(MI->getPreInstrSymbol());
   ASSERT_FALSE(MI->getPostInstrSymbol());
   ASSERT_FALSE(MI->getHeapAllocMarker());
   ASSERT_FALSE(MI->getPCSections());
+  ASSERT_FALSE(MI->getMMRAMetadata());
 
   MI->setMemRefs(*MF, MMOs);
   ASSERT_TRUE(MI->memoperands().size() == 1);
@@ -290,6 +293,7 @@ TEST(MachineInstrExtraInfo, AddExtraInfo) {
   ASSERT_FALSE(MI->getPostInstrSymbol());
   ASSERT_FALSE(MI->getHeapAllocMarker());
   ASSERT_FALSE(MI->getPCSections());
+  ASSERT_FALSE(MI->getMMRAMetadata());
 
   MI->setPreInstrSymbol(*MF, Sym1);
   ASSERT_TRUE(MI->memoperands().size() == 1);
@@ -297,6 +301,7 @@ TEST(MachineInstrExtraInfo, AddExtraInfo) {
   ASSERT_FALSE(MI->getPostInstrSymbol());
   ASSERT_FALSE(MI->getHeapAllocMarker());
   ASSERT_FALSE(MI->getPCSections());
+  ASSERT_FALSE(MI->getMMRAMetadata());
 
   MI->setPostInstrSymbol(*MF, Sym2);
   ASSERT_TRUE(MI->memoperands().size() == 1);
@@ -304,6 +309,7 @@ TEST(MachineInstrExtraInfo, AddExtraInfo) {
   ASSERT_TRUE(MI->getPostInstrSymbol() == Sym2);
   ASSERT_FALSE(MI->getHeapAllocMarker());
   ASSERT_FALSE(MI->getPCSections());
+  ASSERT_FALSE(MI->getMMRAMetadata());
 
   MI->setHeapAllocMarker(*MF, HAM);
   ASSERT_TRUE(MI->memoperands().size() == 1);
@@ -311,6 +317,7 @@ TEST(MachineInstrExtraInfo, AddExtraInfo) {
   ASSERT_TRUE(MI->getPostInstrSymbol() == Sym2);
   ASSERT_TRUE(MI->getHeapAllocMarker() == HAM);
   ASSERT_FALSE(MI->getPCSections());
+  ASSERT_FALSE(MI->getMMRAMetadata());
 
   MI->setPCSections(*MF, PCS);
   ASSERT_TRUE(MI->memoperands().size() == 1);
@@ -318,6 +325,21 @@ TEST(MachineInstrExtraInfo, AddExtraInfo) {
   ASSERT_TRUE(MI->getPostInstrSymbol() == Sym2);
   ASSERT_TRUE(MI->getHeapAllocMarker() == HAM);
   ASSERT_TRUE(MI->getPCSections() == PCS);
+  ASSERT_FALSE(MI->getMMRAMetadata());
+
+  MI->setMMRAMetadata(*MF, MMRA);
+  ASSERT_TRUE(MI->memoperands().size() == 1);
+  ASSERT_TRUE(MI->getPreInstrSymbol() == Sym1);
+  ASSERT_TRUE(MI->getPostInstrSymbol() == Sym2);
+  ASSERT_TRUE(MI->getHeapAllocMarker() == HAM);
+  ASSERT_TRUE(MI->getPCSections() == PCS);
+  ASSERT_TRUE(MI->getMMRAMetadata() == MMRA);
+
+  // Check with nothing but MMRAs.
+  MachineInstr *MMRAMI = MF->CreateMachineInstr(MCID, DebugLoc());
+  ASSERT_FALSE(MMRAMI->getMMRAMetadata());
+  MMRAMI->setMMRAMetadata(*MF, MMRA);
+  ASSERT_TRUE(MMRAMI->getMMRAMetadata() == MMRA);
 }
 
 TEST(MachineInstrExtraInfo, ChangeExtraInfo) {
@@ -338,11 +360,15 @@ TEST(MachineInstrExtraInfo, ChangeExtraInfo) {
   MDNode *HAM = MDNode::getDistinct(Ctx, std::nullopt);
   MDNode *PCS = MDNode::getDistinct(Ctx, std::nullopt);
 
+  MDNode *MMRA1 = MMRAMetadata::getTagMD(Ctx, "foo", "bar");
+  MDNode *MMRA2 = MMRAMetadata::getTagMD(Ctx, "bar", "bux");
+
   MI->setMemRefs(*MF, MMOs);
   MI->setPreInstrSymbol(*MF, Sym1);
   MI->setPostInstrSymbol(*MF, Sym2);
   MI->setHeapAllocMarker(*MF, HAM);
   MI->setPCSections(*MF, PCS);
+  MI->setMMRAMetadata(*MF, MMRA1);
 
   MMOs.push_back(MMO);
 
@@ -352,6 +378,7 @@ TEST(MachineInstrExtraInfo, ChangeExtraInfo) {
   ASSERT_TRUE(MI->getPostInstrSymbol() == Sym2);
   ASSERT_TRUE(MI->getHeapAllocMarker() == HAM);
   ASSERT_TRUE(MI->getPCSections() == PCS);
+  ASSERT_TRUE(MI->getMMRAMetadata() == MMRA1);
 
   MI->setPostInstrSymbol(*MF, Sym1);
   ASSERT_TRUE(MI->memoperands().size() == 2);
@@ -359,6 +386,15 @@ TEST(MachineInstrExtraInfo, ChangeExtraInfo) {
   ASSERT_TRUE(MI->getPostInstrSymbol() == Sym1);
   ASSERT_TRUE(MI->getHeapAllocMarker() == HAM);
   ASSERT_TRUE(MI->getPCSections() == PCS);
+  ASSERT_TRUE(MI->getMMRAMetadata() == MMRA1);
+
+  MI->setMMRAMetadata(*MF, MMRA2);
+  ASSERT_TRUE(MI->memoperands().size() == 2);
+  ASSERT_TRUE(MI->getPreInstrSymbol() == Sym1);
+  ASSERT_TRUE(MI->getPostInstrSymbol() == Sym1);
+  ASSERT_TRUE(MI->getHeapAllocMarker() == HAM);
+  ASSERT_TRUE(MI->getPCSections() == PCS);
+  ASSERT_TRUE(MI->getMMRAMetadata() == MMRA2);
 }
 
 TEST(MachineInstrExtraInfo, RemoveExtraInfo) {
@@ -380,11 +416,14 @@ TEST(MachineInstrExtraInfo, RemoveExtraInfo) {
   MDNode *HAM = MDNode::getDistinct(Ctx, std::nullopt);
   MDNode *PCS = MDNode::getDistinct(Ctx, std::nullopt);
 
+  MDNode *MMRA = MDTuple::get(Ctx, {});
+
   MI->setMemRefs(*MF, MMOs);
   MI->setPreInstrSymbol(*MF, Sym1);
   MI->setPostInstrSymbol(*MF, Sym2);
   MI->setHeapAllocMarker(*MF, HAM);
   MI->setPCSections(*MF, PCS);
+  MI->setMMRAMetadata(*MF, MMRA);
 
   MI->setPostInstrSymbol(*MF, nullptr);
   ASSERT_TRUE(MI->memoperands().size() == 2);
@@ -392,6 +431,7 @@ TEST(MachineInstrExtraInfo, RemoveExtraInfo) {
   ASSERT_FALSE(MI->getPostInstrSymbol());
   ASSERT_TRUE(MI->getHeapAllocMarker() == HAM);
   ASSERT_TRUE(MI->getPCSections() == PCS);
+  ASSERT_TRUE(MI->getMMRAMetadata() == MMRA);
 
   MI->setHeapAllocMarker(*MF, nullptr);
   ASSERT_TRUE(MI->memoperands().size() == 2);
@@ -399,6 +439,7 @@ TEST(MachineInstrExtraInfo, RemoveExtraInfo) {
   ASSERT_FALSE(MI->getPostInstrSymbol());
   ASSERT_FALSE(MI->getHeapAllocMarker());
   ASSERT_TRUE(MI->getPCSections() == PCS);
+  ASSERT_TRUE(MI->getMMRAMetadata() == MMRA);
 
   MI->setPCSections(*MF, nullptr);
   ASSERT_TRUE(MI->memoperands().size() == 2);
@@ -406,6 +447,7 @@ TEST(MachineInstrExtraInfo, RemoveExtraInfo) {
   ASSERT_FALSE(MI->getPostInstrSymbol());
   ASSERT_FALSE(MI->getHeapAllocMarker());
   ASSERT_FALSE(MI->getPCSections());
+  ASSERT_TRUE(MI->getMMRAMetadata() == MMRA);
 
   MI->setPreInstrSymbol(*MF, nullptr);
   ASSERT_TRUE(MI->memoperands().size() == 2);
@@ -413,6 +455,7 @@ TEST(MachineInstrExtraInfo, RemoveExtraInfo) {
   ASSERT_FALSE(MI->getPostInstrSymbol());
   ASSERT_FALSE(MI->getHeapAllocMarker());
   ASSERT_FALSE(MI->getPCSections());
+  ASSERT_TRUE(MI->getMMRAMetadata() == MMRA);
 
   MI->setMemRefs(*MF, {});
   ASSERT_TRUE(MI->memoperands_empty());
@@ -420,6 +463,15 @@ TEST(MachineInstrExtraInfo, RemoveExtraInfo) {
   ASSERT_FALSE(MI->getPostInstrSymbol());
   ASSERT_FALSE(MI->getHeapAllocMarker());
   ASSERT_FALSE(MI->getPCSections());
+  ASSERT_TRUE(MI->getMMRAMetadata() == MMRA);
+
+  MI->setMMRAMetadata(*MF, nullptr);
+  ASSERT_TRUE(MI->memoperands_empty());
+  ASSERT_FALSE(MI->getPreInstrSymbol());
+  ASSERT_FALSE(MI->getPostInstrSymbol());
+  ASSERT_FALSE(MI->getHeapAllocMarker());
+  ASSERT_FALSE(MI->getPCSections());
+  ASSERT_FALSE(MI->getMMRAMetadata());
 }
 
 TEST(MachineInstrDebugValue, AddDebugValueOperand) {
diff --git a/llvm/unittests/ExecutionEngine/Orc/CoreAPIsTest.cpp b/llvm/unittests/ExecutionEngine/Orc/CoreAPIsTest.cpp
index 5e2b5f35bcf4..3b24e29e1ed3 100644
--- a/llvm/unittests/ExecutionEngine/Orc/CoreAPIsTest.cpp
+++ b/llvm/unittests/ExecutionEngine/Orc/CoreAPIsTest.cpp
@@ -1005,11 +1005,11 @@ TEST_F(CoreAPIsStandardTest, RedefineBoundWeakSymbol) {
 
 TEST_F(CoreAPIsStandardTest, DefineMaterializingSymbol) {
   bool ExpectNoMoreMaterialization = false;
-  ES.setDispatchTask([&](std::unique_ptr<Task> T) {
+  DispatchOverride = [&](std::unique_ptr<Task> T) {
     if (ExpectNoMoreMaterialization && isa<MaterializationTask>(*T))
       ADD_FAILURE() << "Unexpected materialization";
     T->run();
-  });
+  };
 
   auto MU = std::make_unique<SimpleMaterializationUnit>(
       SymbolFlagsMap({{Foo, FooSym.getFlags()}}),
@@ -1403,7 +1403,7 @@ TEST_F(CoreAPIsStandardTest, TestLookupWithThreadedMaterialization) {
 
   std::mutex WorkThreadsMutex;
   std::vector<std::thread> WorkThreads;
-  ES.setDispatchTask([&](std::unique_ptr<Task> T) {
+  DispatchOverride = [&](std::unique_ptr<Task> T) {
     std::promise<void> WaitP;
     std::lock_guard<std::mutex> Lock(WorkThreadsMutex);
     WorkThreads.push_back(
@@ -1412,7 +1412,7 @@ TEST_F(CoreAPIsStandardTest, TestLookupWithThreadedMaterialization) {
           T->run();
         }));
     WaitP.set_value();
-  });
+  };
 
   cantFail(JD.define(absoluteSymbols({{Foo, FooSym}})));
 
diff --git a/llvm/unittests/ExecutionEngine/Orc/OrcTestCommon.cpp b/llvm/unittests/ExecutionEngine/Orc/OrcTestCommon.cpp
index bc87df1fe8c6..307f14dfe24d 100644
--- a/llvm/unittests/ExecutionEngine/Orc/OrcTestCommon.cpp
+++ b/llvm/unittests/ExecutionEngine/Orc/OrcTestCommon.cpp
@@ -22,3 +22,18 @@ ModuleBuilder::ModuleBuilder(LLVMContext &Context, StringRef Triple,
   if (Triple != "")
     M->setTargetTriple(Triple);
 }
+
+void llvm::orc::CoreAPIsBasedStandardTest::OverridableDispatcher::dispatch(
+    std::unique_ptr<Task> T) {
+  if (Parent.DispatchOverride)
+    Parent.DispatchOverride(std::move(T));
+  else
+    InPlaceTaskDispatcher::dispatch(std::move(T));
+}
+
+std::unique_ptr<llvm::orc::ExecutorProcessControl>
+llvm::orc::CoreAPIsBasedStandardTest::makeEPC(
+    std::shared_ptr<SymbolStringPool> SSP) {
+  return std::make_unique<UnsupportedExecutorProcessControl>(
+      std::move(SSP), std::make_unique<OverridableDispatcher>(*this));
+}
diff --git a/llvm/unittests/ExecutionEngine/Orc/OrcTestCommon.h b/llvm/unittests/ExecutionEngine/Orc/OrcTestCommon.h
index ce7da76c9653..0981f4b8132b 100644
--- a/llvm/unittests/ExecutionEngine/Orc/OrcTestCommon.h
+++ b/llvm/unittests/ExecutionEngine/Orc/OrcTestCommon.h
@@ -52,8 +52,20 @@ public:
   }
 
 protected:
+  class OverridableDispatcher : public InPlaceTaskDispatcher {
+  public:
+    OverridableDispatcher(CoreAPIsBasedStandardTest &Parent) : Parent(Parent) {}
+    void dispatch(std::unique_ptr<Task> T) override;
+
+  private:
+    CoreAPIsBasedStandardTest &Parent;
+  };
+
+  std::unique_ptr<llvm::orc::ExecutorProcessControl>
+  makeEPC(std::shared_ptr<SymbolStringPool> SSP);
+
   std::shared_ptr<SymbolStringPool> SSP = std::make_shared<SymbolStringPool>();
-  ExecutionSession ES{std::make_unique<UnsupportedExecutorProcessControl>(SSP)};
+  ExecutionSession ES{makeEPC(SSP)};
   JITDylib &JD = ES.createBareJITDylib("JD");
   SymbolStringPtr Foo = ES.intern("foo");
   SymbolStringPtr Bar = ES.intern("bar");
@@ -67,6 +79,7 @@ protected:
   ExecutorSymbolDef BarSym{BarAddr, JITSymbolFlags::Exported};
   ExecutorSymbolDef BazSym{BazAddr, JITSymbolFlags::Exported};
   ExecutorSymbolDef QuxSym{QuxAddr, JITSymbolFlags::Exported};
+  unique_function<void(std::unique_ptr<Task>)> DispatchOverride;
 };
 
 } // end namespace orc
diff --git a/llvm/unittests/ExecutionEngine/Orc/TaskDispatchTest.cpp b/llvm/unittests/ExecutionEngine/Orc/TaskDispatchTest.cpp
index 83d386c631dd..6af0d60cf8ae 100644
--- a/llvm/unittests/ExecutionEngine/Orc/TaskDispatchTest.cpp
+++ b/llvm/unittests/ExecutionEngine/Orc/TaskDispatchTest.cpp
@@ -24,7 +24,7 @@ TEST(InPlaceTaskDispatchTest, GenericNamedTask) {
 
 #if LLVM_ENABLE_THREADS
 TEST(DynamicThreadPoolDispatchTest, GenericNamedTask) {
-  auto D = std::make_unique<DynamicThreadPoolTaskDispatcher>();
+  auto D = std::make_unique<DynamicThreadPoolTaskDispatcher>(std::nullopt);
   std::promise<bool> P;
   auto F = P.get_future();
   D->dispatch(makeGenericNamedTask(
diff --git a/llvm/unittests/Frontend/CMakeLists.txt b/llvm/unittests/Frontend/CMakeLists.txt
index ddb6a16cbb98..3f290b63ba64 100644
--- a/llvm/unittests/Frontend/CMakeLists.txt
+++ b/llvm/unittests/Frontend/CMakeLists.txt
@@ -14,7 +14,7 @@ add_llvm_unittest(LLVMFrontendTests
   OpenMPContextTest.cpp
   OpenMPIRBuilderTest.cpp
   OpenMPParsingTest.cpp
-  OpenMPComposeTest.cpp
+  OpenMPCompositionTest.cpp
 
   DEPENDS
   acc_gen
diff --git a/llvm/unittests/Frontend/OpenMPComposeTest.cpp b/llvm/unittests/Frontend/OpenMPComposeTest.cpp
deleted file mode 100644
index c5fbe6ec6adf..000000000000
--- a/llvm/unittests/Frontend/OpenMPComposeTest.cpp
+++ /dev/null
@@ -1,41 +0,0 @@
-//===- llvm/unittests/Frontend/OpenMPComposeTest.cpp ----------------------===//
-//
-// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
-// See https://llvm.org/LICENSE.txt for license information.
-// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
-//
-//===----------------------------------------------------------------------===//
-
-#include "llvm/ADT/ArrayRef.h"
-#include "llvm/Frontend/OpenMP/OMP.h"
-#include "gtest/gtest.h"
-
-using namespace llvm;
-using namespace llvm::omp;
-
-TEST(Composition, GetLeafConstructs) {
-  ArrayRef<Directive> L1 = getLeafConstructs(OMPD_loop);
-  ASSERT_EQ(L1, (ArrayRef<Directive>{}));
-  ArrayRef<Directive> L2 = getLeafConstructs(OMPD_parallel_for);
-  ASSERT_EQ(L2, (ArrayRef<Directive>{OMPD_parallel, OMPD_for}));
-  ArrayRef<Directive> L3 = getLeafConstructs(OMPD_parallel_for_simd);
-  ASSERT_EQ(L3, (ArrayRef<Directive>{OMPD_parallel, OMPD_for, OMPD_simd}));
-}
-
-TEST(Composition, GetCompoundConstruct) {
-  Directive C1 =
-      getCompoundConstruct({OMPD_target, OMPD_teams, OMPD_distribute});
-  ASSERT_EQ(C1, OMPD_target_teams_distribute);
-  Directive C2 = getCompoundConstruct({OMPD_target});
-  ASSERT_EQ(C2, OMPD_target);
-  Directive C3 = getCompoundConstruct({OMPD_target, OMPD_masked});
-  ASSERT_EQ(C3, OMPD_unknown);
-  Directive C4 = getCompoundConstruct({OMPD_target, OMPD_teams_distribute});
-  ASSERT_EQ(C4, OMPD_target_teams_distribute);
-  Directive C5 = getCompoundConstruct({});
-  ASSERT_EQ(C5, OMPD_unknown);
-  Directive C6 = getCompoundConstruct({OMPD_parallel_for, OMPD_simd});
-  ASSERT_EQ(C6, OMPD_parallel_for_simd);
-  Directive C7 = getCompoundConstruct({OMPD_do, OMPD_simd});
-  ASSERT_EQ(C7, OMPD_do_simd); // Make sure it's not OMPD_end_do_simd
-}
diff --git a/llvm/unittests/Frontend/OpenMPCompositionTest.cpp b/llvm/unittests/Frontend/OpenMPCompositionTest.cpp
new file mode 100644
index 000000000000..920b445427e7
--- /dev/null
+++ b/llvm/unittests/Frontend/OpenMPCompositionTest.cpp
@@ -0,0 +1,100 @@
+//===- llvm/unittests/Frontend/OpenMPCompositionTest.cpp ------------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#include "llvm/ADT/ArrayRef.h"
+#include "llvm/ADT/SmallVector.h"
+#include "llvm/Frontend/OpenMP/OMP.h"
+#include "gtest/gtest.h"
+
+using namespace llvm;
+using namespace llvm::omp;
+
+TEST(Composition, GetLeafConstructs) {
+  ArrayRef<Directive> L1 = getLeafConstructs(OMPD_loop);
+  ASSERT_EQ(L1, (ArrayRef<Directive>{}));
+  ArrayRef<Directive> L2 = getLeafConstructs(OMPD_parallel_for);
+  ASSERT_EQ(L2, (ArrayRef<Directive>{OMPD_parallel, OMPD_for}));
+  ArrayRef<Directive> L3 = getLeafConstructs(OMPD_parallel_for_simd);
+  ASSERT_EQ(L3, (ArrayRef<Directive>{OMPD_parallel, OMPD_for, OMPD_simd}));
+}
+
+TEST(Composition, GetCompoundConstruct) {
+  Directive C1 =
+      getCompoundConstruct({OMPD_target, OMPD_teams, OMPD_distribute});
+  ASSERT_EQ(C1, OMPD_target_teams_distribute);
+  Directive C2 = getCompoundConstruct({OMPD_target});
+  ASSERT_EQ(C2, OMPD_target);
+  Directive C3 = getCompoundConstruct({OMPD_target, OMPD_masked});
+  ASSERT_EQ(C3, OMPD_unknown);
+  Directive C4 = getCompoundConstruct({OMPD_target, OMPD_teams_distribute});
+  ASSERT_EQ(C4, OMPD_target_teams_distribute);
+  Directive C5 = getCompoundConstruct({});
+  ASSERT_EQ(C5, OMPD_unknown);
+  Directive C6 = getCompoundConstruct({OMPD_parallel_for, OMPD_simd});
+  ASSERT_EQ(C6, OMPD_parallel_for_simd);
+  Directive C7 = getCompoundConstruct({OMPD_do, OMPD_simd});
+  ASSERT_EQ(C7, OMPD_do_simd); // Make sure it's not OMPD_end_do_simd
+}
+
+TEST(Composition, GetLeafOrCompositeConstructs) {
+  SmallVector<Directive> Out1;
+  auto Ret1 = getLeafOrCompositeConstructs(
+      OMPD_target_teams_distribute_parallel_for, Out1);
+  ASSERT_EQ(Ret1, ArrayRef<Directive>(Out1));
+  ASSERT_EQ((ArrayRef<Directive>(Out1)),
+            (ArrayRef<Directive>{OMPD_target, OMPD_teams,
+                                 OMPD_distribute_parallel_for}));
+
+  SmallVector<Directive> Out2;
+  auto Ret2 =
+      getLeafOrCompositeConstructs(OMPD_parallel_masked_taskloop_simd, Out2);
+  ASSERT_EQ(Ret2, ArrayRef<Directive>(Out2));
+  ASSERT_EQ(
+      (ArrayRef<Directive>(Out2)),
+      (ArrayRef<Directive>{OMPD_parallel, OMPD_masked, OMPD_taskloop_simd}));
+
+  SmallVector<Directive> Out3;
+  auto Ret3 =
+      getLeafOrCompositeConstructs(OMPD_distribute_parallel_do_simd, Out3);
+  ASSERT_EQ(Ret3, ArrayRef<Directive>(Out3));
+  ASSERT_EQ((ArrayRef<Directive>(Out3)),
+            (ArrayRef<Directive>{OMPD_distribute_parallel_do_simd}));
+
+  SmallVector<Directive> Out4;
+  auto Ret4 = getLeafOrCompositeConstructs(OMPD_target_parallel_loop, Out4);
+  ASSERT_EQ(Ret4, ArrayRef<Directive>(Out4));
+  ASSERT_EQ((ArrayRef<Directive>(Out4)),
+            (ArrayRef<Directive>{OMPD_target, OMPD_parallel, OMPD_loop}));
+}
+
+TEST(Composition, IsLeafConstruct) {
+  ASSERT_TRUE(isLeafConstruct(OMPD_loop));
+  ASSERT_TRUE(isLeafConstruct(OMPD_teams));
+  ASSERT_FALSE(isLeafConstruct(OMPD_for_simd));
+  ASSERT_FALSE(isLeafConstruct(OMPD_distribute_simd));
+  ASSERT_FALSE(isLeafConstruct(OMPD_parallel_for));
+}
+
+TEST(Composition, IsCompositeConstruct) {
+  ASSERT_TRUE(isCompositeConstruct(OMPD_distribute_simd));
+  ASSERT_FALSE(isCompositeConstruct(OMPD_for));
+  ASSERT_TRUE(isCompositeConstruct(OMPD_for_simd));
+  // directive-name-A = "parallel", directive-name-B = "for simd",
+  // only directive-name-B is loop-associated, so this is not a
+  // composite construct, even though "for simd" is.
+  ASSERT_FALSE(isCompositeConstruct(OMPD_parallel_for_simd));
+}
+
+TEST(Composition, IsCombinedConstruct) {
+  // "parallel for simd" is a combined construct, see comment in
+  // IsCompositeConstruct.
+  ASSERT_TRUE(isCombinedConstruct(OMPD_parallel_for_simd));
+  ASSERT_FALSE(isCombinedConstruct(OMPD_for_simd));
+  ASSERT_TRUE(isCombinedConstruct(OMPD_parallel_for));
+  ASSERT_FALSE(isCombinedConstruct(OMPD_parallel));
+}
diff --git a/llvm/unittests/IR/CMakeLists.txt b/llvm/unittests/IR/CMakeLists.txt
index 803164b8f1ea..a03b0711ba33 100644
--- a/llvm/unittests/IR/CMakeLists.txt
+++ b/llvm/unittests/IR/CMakeLists.txt
@@ -31,6 +31,7 @@ add_llvm_unittest(IRTests
   IntrinsicsTest.cpp
   LegacyPassManagerTest.cpp
   MDBuilderTest.cpp
+  MemoryModelRelaxationAnnotationsTest.cpp
   ManglerTest.cpp
   MetadataTest.cpp
   ModuleTest.cpp
diff --git a/llvm/unittests/IR/IntrinsicsTest.cpp b/llvm/unittests/IR/IntrinsicsTest.cpp
index a500346b66a5..dddd2f73d444 100644
--- a/llvm/unittests/IR/IntrinsicsTest.cpp
+++ b/llvm/unittests/IR/IntrinsicsTest.cpp
@@ -6,7 +6,12 @@
 //
 //===----------------------------------------------------------------------===//
 
+#include "llvm/IR/Intrinsics.h"
+#include "llvm/ADT/SmallVector.h"
+#include "llvm/IR/Constant.h"
+#include "llvm/IR/IRBuilder.h"
 #include "llvm/IR/IntrinsicInst.h"
+#include "llvm/IR/Module.h"
 #include "gtest/gtest.h"
 
 using namespace llvm;
@@ -14,14 +19,41 @@ using namespace llvm;
 namespace {
 
 static const char *const NameTable1[] = {
-  "llvm.foo",
-  "llvm.foo.a",
-  "llvm.foo.b",
-  "llvm.foo.b.a",
-  "llvm.foo.c",
+    "llvm.foo", "llvm.foo.a", "llvm.foo.b", "llvm.foo.b.a", "llvm.foo.c",
 };
 
-TEST(IntrinNameLookup, Basic) {
+class IntrinsicsTest : public ::testing::Test {
+  LLVMContext Context;
+  std::unique_ptr<Module> M;
+  BasicBlock *BB = nullptr;
+
+  void TearDown() override { M.reset(); }
+
+  void SetUp() override {
+    M = std::make_unique<Module>("Test", Context);
+    auto F = M->getOrInsertFunction(
+        "test", FunctionType::get(Type::getVoidTy(Context), false));
+    BB = BasicBlock::Create(Context, "", cast<Function>(F.getCallee()));
+    EXPECT_NE(BB, nullptr);
+  }
+
+public:
+  Instruction *makeIntrinsic(Intrinsic::ID ID) const {
+    IRBuilder<> Builder(BB);
+    SmallVector<Value *, 4> ProcessedArgs;
+    auto *Decl = Intrinsic::getDeclaration(M.get(), ID);
+    for (auto *Ty : Decl->getFunctionType()->params()) {
+      auto *Val = Constant::getNullValue(Ty);
+      ProcessedArgs.push_back(Val);
+    }
+    return Builder.CreateCall(Decl, ProcessedArgs);
+  }
+  template <typename T> void checkIsa(const Instruction &I) {
+    EXPECT_TRUE(isa<T>(I));
+  }
+};
+
+TEST(IntrinsicNameLookup, Basic) {
   int I = Intrinsic::lookupLLVMIntrinsicByName(NameTable1, "llvm.foo");
   EXPECT_EQ(0, I);
   I = Intrinsic::lookupLLVMIntrinsicByName(NameTable1, "llvm.foo.f64");
@@ -36,4 +68,45 @@ TEST(IntrinNameLookup, Basic) {
   EXPECT_EQ(4, I);
 }
 
+TEST_F(IntrinsicsTest, InstrProfInheritance) {
+  auto isInstrProfInstBase = [](const Instruction &I) {
+    return isa<InstrProfInstBase>(I);
+  };
+#define __ISA(TYPE, PARENT)                                                    \
+  auto is##TYPE = [&](const Instruction &I) -> bool {                          \
+    return isa<TYPE>(I) && is##PARENT(I);                                      \
+  }
+  __ISA(InstrProfCntrInstBase, InstrProfInstBase);
+  __ISA(InstrProfMCDCCondBitmapUpdate, InstrProfInstBase);
+  __ISA(InstrProfCoverInst, InstrProfCntrInstBase);
+  __ISA(InstrProfIncrementInst, InstrProfCntrInstBase);
+  __ISA(InstrProfIncrementInstStep, InstrProfIncrementInst);
+  __ISA(InstrProfCallsite, InstrProfCntrInstBase);
+  __ISA(InstrProfTimestampInst, InstrProfCntrInstBase);
+  __ISA(InstrProfValueProfileInst, InstrProfCntrInstBase);
+  __ISA(InstrProfMCDCBitmapInstBase, InstrProfInstBase);
+  __ISA(InstrProfMCDCBitmapParameters, InstrProfMCDCBitmapInstBase);
+  __ISA(InstrProfMCDCTVBitmapUpdate, InstrProfMCDCBitmapInstBase);
+#undef __ISA
+
+  std::vector<
+      std::pair<Intrinsic::ID, std::function<bool(const Instruction &)>>>
+      LeafIDs = {
+          {Intrinsic::instrprof_cover, isInstrProfCoverInst},
+          {Intrinsic::instrprof_increment, isInstrProfIncrementInst},
+          {Intrinsic::instrprof_increment_step, isInstrProfIncrementInstStep},
+          {Intrinsic::instrprof_callsite, isInstrProfCallsite},
+          {Intrinsic::instrprof_mcdc_condbitmap_update,
+           isInstrProfMCDCCondBitmapUpdate},
+          {Intrinsic::instrprof_mcdc_parameters,
+           isInstrProfMCDCBitmapParameters},
+          {Intrinsic::instrprof_mcdc_tvbitmap_update,
+           isInstrProfMCDCTVBitmapUpdate},
+          {Intrinsic::instrprof_timestamp, isInstrProfTimestampInst},
+          {Intrinsic::instrprof_value_profile, isInstrProfValueProfileInst}};
+  for (const auto &[ID, Checker] : LeafIDs) {
+    auto *Intr = makeIntrinsic(ID);
+    EXPECT_TRUE(Checker(*Intr));
+  }
+}
 } // end namespace
diff --git a/llvm/unittests/IR/MemoryModelRelaxationAnnotationsTest.cpp b/llvm/unittests/IR/MemoryModelRelaxationAnnotationsTest.cpp
new file mode 100644
index 000000000000..8feeb8af65a7
--- /dev/null
+++ b/llvm/unittests/IR/MemoryModelRelaxationAnnotationsTest.cpp
@@ -0,0 +1,212 @@
+//===- llvm/unittests/IR/MemoryModelRelaxationAnnotationsTest.cpp ---------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#include "llvm/IR/MemoryModelRelaxationAnnotations.h"
+#include "llvm/ADT/STLExtras.h"
+#include "llvm/IR/Metadata.h"
+#include "llvm/IR/Module.h"
+#include "gtest/gtest.h"
+
+using namespace llvm;
+
+namespace {
+
+void checkMMRA(const MMRAMetadata &MMRA,
+               ArrayRef<MMRAMetadata::TagT> Expected) {
+  EXPECT_EQ(MMRA.size(), Expected.size());
+  for (const auto &E : Expected)
+    EXPECT_TRUE(MMRA.hasTag(E.first, E.second));
+}
+
+MMRAMetadata createFromMD(LLVMContext &Ctx,
+                          ArrayRef<MMRAMetadata::TagT> Expected) {
+  SmallVector<Metadata *> MD;
+  for (const auto &Tag : Expected)
+    MD.push_back(MMRAMetadata::getTagMD(Ctx, Tag));
+  return MDTuple::get(Ctx, MD);
+}
+
+TEST(MMRATest, MDParse) {
+  LLVMContext Ctx;
+
+  // No nesting:
+  // !{!"foo", "!bar"}
+  MDNode *FooBar =
+      MDTuple::get(Ctx, {MDString::get(Ctx, "foo"), MDString::get(Ctx, "bar")});
+  MMRAMetadata FooBarMMRA(FooBar);
+
+  checkMMRA(FooBarMMRA, {{"foo", "bar"}});
+
+  // Nested:
+  // !{!{!"foo", "!bar"}, !{!"bux", !"qux"}}
+  MDNode *BuxQux =
+      MDTuple::get(Ctx, {MDString::get(Ctx, "bux"), MDString::get(Ctx, "qux")});
+  MDNode *Nested = MDTuple::get(Ctx, {FooBar, BuxQux});
+  MMRAMetadata NestedMMRA(Nested);
+
+  checkMMRA(NestedMMRA, {{"foo", "bar"}, {"bux", "qux"}});
+}
+
+TEST(MMRATest, GetMD) {
+  LLVMContext Ctx;
+
+  EXPECT_EQ(MMRAMetadata::getMD(Ctx, {}), nullptr);
+
+  MDTuple *SingleMD = MMRAMetadata::getMD(Ctx, {{"foo", "bar"}});
+  EXPECT_EQ(SingleMD->getNumOperands(), 2u);
+  EXPECT_EQ(cast<MDString>(SingleMD->getOperand(0))->getString(), "foo");
+  EXPECT_EQ(cast<MDString>(SingleMD->getOperand(1))->getString(), "bar");
+
+  MDTuple *MultiMD = MMRAMetadata::getMD(Ctx, {{"foo", "bar"}, {"bux", "qux"}});
+  EXPECT_EQ(MultiMD->getNumOperands(), 2u);
+
+  MDTuple *FooBar = cast<MDTuple>(MultiMD->getOperand(0));
+  EXPECT_EQ(cast<MDString>(FooBar->getOperand(0))->getString(), "foo");
+  EXPECT_EQ(cast<MDString>(FooBar->getOperand(1))->getString(), "bar");
+  MDTuple *BuxQux = cast<MDTuple>(MultiMD->getOperand(1));
+  EXPECT_EQ(cast<MDString>(BuxQux->getOperand(0))->getString(), "bux");
+  EXPECT_EQ(cast<MDString>(BuxQux->getOperand(1))->getString(), "qux");
+}
+
+TEST(MMRATest, Utility) {
+  LLVMContext Ctx;
+  MMRAMetadata MMRA =
+      createFromMD(Ctx, {{"foo", "0"}, {"foo", "1"}, {"bar", "x"}});
+
+  EXPECT_TRUE(MMRA.hasTagWithPrefix("foo"));
+  EXPECT_TRUE(MMRA.hasTagWithPrefix("bar"));
+  EXPECT_FALSE(MMRA.hasTagWithPrefix("x"));
+
+  EXPECT_TRUE(MMRA.hasTag("foo", "0"));
+  EXPECT_TRUE(MMRA.hasTag("foo", "1"));
+  EXPECT_TRUE(MMRA.hasTag("bar", "x"));
+}
+
+TEST(MMRATest, Operators) {
+  LLVMContext Ctx;
+
+  MMRAMetadata A = createFromMD(Ctx, {{"foo", "0"}, {"bar", "x"}});
+  MMRAMetadata B = createFromMD(Ctx, {{"foo", "0"}, {"bar", "y"}});
+
+  // ensure we have different objects by creating copies.
+  EXPECT_EQ(MMRAMetadata(A), MMRAMetadata(A));
+  EXPECT_TRUE((bool)A);
+
+  EXPECT_EQ(MMRAMetadata(B), MMRAMetadata(B));
+  EXPECT_TRUE((bool)B);
+
+  EXPECT_NE(A, B);
+
+  EXPECT_EQ(MMRAMetadata(), MMRAMetadata());
+  EXPECT_NE(A, MMRAMetadata());
+  EXPECT_NE(B, MMRAMetadata());
+
+  MMRAMetadata Empty;
+  EXPECT_FALSE((bool)Empty);
+}
+
+TEST(MMRATest, Compatibility) {
+  LLVMContext Ctx;
+
+  MMRAMetadata Foo0 = createFromMD(Ctx, {{"foo", "0"}});
+  MMRAMetadata Foo1 = createFromMD(Ctx, {{"foo", "1"}});
+  MMRAMetadata Foo10 = createFromMD(Ctx, {{"foo", "0"}, {"foo", "1"}});
+
+  MMRAMetadata Bar = createFromMD(Ctx, {{"bar", "y"}});
+
+  MMRAMetadata Empty;
+
+  // Other set has no tag with same prefix
+  EXPECT_TRUE(Foo0.isCompatibleWith(Bar));
+  EXPECT_TRUE(Bar.isCompatibleWith(Foo0));
+
+  EXPECT_TRUE(Foo0.isCompatibleWith(Empty));
+  EXPECT_TRUE(Empty.isCompatibleWith(Foo0));
+
+  EXPECT_TRUE(Empty.isCompatibleWith(MMRAMetadata()));
+  EXPECT_TRUE(MMRAMetadata().isCompatibleWith(Empty));
+
+  // Other set has conflicting tags.
+  EXPECT_FALSE(Foo1.isCompatibleWith(Foo0));
+  EXPECT_FALSE(Foo0.isCompatibleWith(Foo1));
+
+  // Both have common tags.
+  EXPECT_TRUE(Foo0.isCompatibleWith(Foo0));
+  EXPECT_TRUE(Foo0.isCompatibleWith(Foo10));
+  EXPECT_TRUE(Foo10.isCompatibleWith(Foo0));
+
+  EXPECT_TRUE(Foo1.isCompatibleWith(Foo1));
+  EXPECT_TRUE(Foo1.isCompatibleWith(Foo10));
+  EXPECT_TRUE(Foo10.isCompatibleWith(Foo1));
+
+  // Try with more prefixes now:
+  MMRAMetadata Multiple0 =
+      createFromMD(Ctx, {{"foo", "y"}, {"foo", "x"}, {"bar", "z"}});
+  MMRAMetadata Multiple1 =
+      createFromMD(Ctx, {{"foo", "z"}, {"foo", "x"}, {"bar", "y"}});
+  MMRAMetadata Multiple2 =
+      createFromMD(Ctx, {{"foo", "z"}, {"foo", "x"}, {"bux", "y"}});
+
+  // Multiple0 and Multiple1 are not compatible because "bar" is getting in the
+  // way.
+  EXPECT_FALSE(Multiple0.isCompatibleWith(Multiple1));
+  EXPECT_FALSE(Multiple1.isCompatibleWith(Multiple0));
+
+  EXPECT_TRUE(Multiple0.isCompatibleWith(Empty));
+  EXPECT_TRUE(Empty.isCompatibleWith(Multiple0));
+  EXPECT_TRUE(Multiple1.isCompatibleWith(Empty));
+  EXPECT_TRUE(Empty.isCompatibleWith(Multiple1));
+
+  // Multiple2 is compatible with both 1/0 because there is always "foo:x" in
+  // common, and the other prefixes are unique to each set.
+  EXPECT_TRUE(Multiple2.isCompatibleWith(Multiple0));
+  EXPECT_TRUE(Multiple0.isCompatibleWith(Multiple2));
+  EXPECT_TRUE(Multiple2.isCompatibleWith(Multiple1));
+  EXPECT_TRUE(Multiple1.isCompatibleWith(Multiple2));
+}
+
+TEST(MMRATest, Combine) {
+  LLVMContext Ctx;
+
+  MMRAMetadata Foo0 = createFromMD(Ctx, {{"foo", "0"}});
+  MMRAMetadata Foo10 = createFromMD(Ctx, {{"foo", "0"}, {"foo", "1"}});
+  MMRAMetadata Bar0 = createFromMD(Ctx, {{"bar", "0"}});
+  MMRAMetadata BarFoo0 = createFromMD(Ctx, {{"bar", "0"}, {"foo", "0"}});
+
+  {
+    // foo is common to both sets
+    MMRAMetadata Combined = MMRAMetadata::combine(Ctx, Foo0, Foo10);
+    EXPECT_EQ(Combined, Foo10);
+  }
+
+  {
+    // nothing is common
+    MMRAMetadata Combined = MMRAMetadata::combine(Ctx, Foo0, Bar0);
+    EXPECT_TRUE(Combined.empty());
+  }
+
+  {
+    // only foo is common.
+    MMRAMetadata Combined = MMRAMetadata::combine(Ctx, BarFoo0, Foo0);
+    EXPECT_EQ(Combined, Foo0);
+  }
+
+  {
+    // only bar is common.
+    MMRAMetadata Combined = MMRAMetadata::combine(Ctx, BarFoo0, Bar0);
+    EXPECT_EQ(Combined, Bar0);
+  }
+
+  {
+    // only foo is common
+    MMRAMetadata Combined = MMRAMetadata::combine(Ctx, BarFoo0, Foo10);
+    EXPECT_EQ(Combined, Foo10);
+  }
+}
+
+} // namespace
diff --git a/llvm/unittests/IR/PatternMatch.cpp b/llvm/unittests/IR/PatternMatch.cpp
index f0377eae9989..a25885faa3a4 100644
--- a/llvm/unittests/IR/PatternMatch.cpp
+++ b/llvm/unittests/IR/PatternMatch.cpp
@@ -1995,7 +1995,7 @@ TEST_F(PatternMatchTest, VScale) {
   EXPECT_TRUE(match(PtrToInt2, m_VScale()));
 }
 
-TEST_F(PatternMatchTest, NotForbidUndef) {
+TEST_F(PatternMatchTest, NotForbidPoison) {
   Type *ScalarTy = IRB.getInt8Ty();
   Type *VectorTy = FixedVectorType::get(ScalarTy, 3);
   Constant *ScalarUndef = UndefValue::get(ScalarTy);
@@ -2020,23 +2020,33 @@ TEST_F(PatternMatchTest, NotForbidUndef) {
   Value *X;
   EXPECT_TRUE(match(Not, m_Not(m_Value(X))));
   EXPECT_TRUE(match(X, m_Zero()));
+  X = nullptr;
+  EXPECT_TRUE(match(Not, m_NotForbidPoison(m_Value(X))));
+  EXPECT_TRUE(match(X, m_Zero()));
 
   Value *NotCommute = IRB.CreateXor(VectorOnes, VectorZero);
   Value *Y;
   EXPECT_TRUE(match(NotCommute, m_Not(m_Value(Y))));
   EXPECT_TRUE(match(Y, m_Zero()));
+  Y = nullptr;
+  EXPECT_TRUE(match(NotCommute, m_NotForbidPoison(m_Value(Y))));
+  EXPECT_TRUE(match(Y, m_Zero()));
 
   Value *NotWithUndefs = IRB.CreateXor(VectorZero, VectorMixedUndef);
   EXPECT_FALSE(match(NotWithUndefs, m_Not(m_Value())));
+  EXPECT_FALSE(match(NotWithUndefs, m_NotForbidPoison(m_Value())));
 
   Value *NotWithPoisons = IRB.CreateXor(VectorZero, VectorMixedPoison);
   EXPECT_TRUE(match(NotWithPoisons, m_Not(m_Value())));
+  EXPECT_FALSE(match(NotWithPoisons, m_NotForbidPoison(m_Value())));
 
   Value *NotWithUndefsCommute = IRB.CreateXor(VectorMixedUndef, VectorZero);
   EXPECT_FALSE(match(NotWithUndefsCommute, m_Not(m_Value())));
+  EXPECT_FALSE(match(NotWithUndefsCommute, m_NotForbidPoison(m_Value())));
 
   Value *NotWithPoisonsCommute = IRB.CreateXor(VectorMixedPoison, VectorZero);
   EXPECT_TRUE(match(NotWithPoisonsCommute, m_Not(m_Value())));
+  EXPECT_FALSE(match(NotWithPoisonsCommute, m_NotForbidPoison(m_Value())));
 }
 
 template <typename T> struct MutableConstTest : PatternMatchTest { };
diff --git a/llvm/unittests/Object/ELFObjectFileTest.cpp b/llvm/unittests/Object/ELFObjectFileTest.cpp
index c4d2b4ae8b9a..c13dc0e3fab8 100644
--- a/llvm/unittests/Object/ELFObjectFileTest.cpp
+++ b/llvm/unittests/Object/ELFObjectFileTest.cpp
@@ -1504,3 +1504,46 @@ Sections:
                "SHT_RELA section with index 1: failed to get a "
                "relocated section: invalid section index: 255");
 }
+
+TEST(ELFObjectFileTest, ELFSymbolRefLess) {
+  SmallString<0> Storage;
+  Expected<ELFObjectFile<ELF64LE>> ElfOrErr = toBinary<ELF64LE>(Storage, R"(
+--- !ELF
+FileHeader:
+  Class:   ELFCLASS64
+  Data:    ELFDATA2LSB
+  Type:    ET_DYN
+  Machine: EM_X86_64
+)");
+
+  ASSERT_THAT_EXPECTED(ElfOrErr, Succeeded());
+  const ELFObjectFile<ELF64LE> &Obj = *ElfOrErr;
+
+  const uint32_t ValLow = 0x00000001;
+  const uint32_t ValHigh = 0x00000100;
+
+  auto MakeSymbol = [&Obj](size_t SymtabIndex, size_t SymbolIndex) {
+    DataRefImpl Data;
+    Data.d.a = SymtabIndex;
+    Data.d.b = SymbolIndex;
+    SymbolRef Sym(Data, &Obj);
+    return ELFSymbolRef(Sym);
+  };
+
+  ELFSymbolRef ELFSymLowLow = MakeSymbol(ValLow, ValLow);
+  ELFSymbolRef ELFSymLowHigh = MakeSymbol(ValLow, ValHigh);
+  ELFSymbolRef ELFSymHighLow = MakeSymbol(ValHigh, ValLow);
+  ELFSymbolRef ELFSymHighHigh = MakeSymbol(ValHigh, ValHigh);
+
+  EXPECT_TRUE(ELFSymLowLow < ELFSymLowHigh);
+  EXPECT_FALSE(ELFSymLowHigh < ELFSymLowLow);
+  EXPECT_FALSE(ELFSymLowLow < ELFSymLowLow);
+
+  EXPECT_TRUE(ELFSymLowLow < ELFSymHighHigh);
+  EXPECT_TRUE(ELFSymLowHigh < ELFSymHighLow);
+  EXPECT_TRUE(ELFSymLowLow < ELFSymHighLow);
+
+  EXPECT_FALSE(ELFSymHighLow < ELFSymLowHigh);
+  EXPECT_FALSE(ELFSymHighHigh < ELFSymLowLow);
+  EXPECT_FALSE(ELFSymHighLow < ELFSymLowLow);
+}
diff --git a/llvm/unittests/ProfileData/InstrProfTest.cpp b/llvm/unittests/ProfileData/InstrProfTest.cpp
index 73ba0a23ea5f..402de64fe99b 100644
--- a/llvm/unittests/ProfileData/InstrProfTest.cpp
+++ b/llvm/unittests/ProfileData/InstrProfTest.cpp
@@ -370,12 +370,31 @@ static CallStackIdMapTy getCallStackMapping() {
   return Mapping;
 }
 
+// Populate all of the fields of MIB.
+MemInfoBlock makeFullMIB() {
+  MemInfoBlock MIB;
+#define MIBEntryDef(NameTag, Name, Type) MIB.NameTag;
+#include "llvm/ProfileData/MIBEntryDef.inc"
+#undef MIBEntryDef
+  return MIB;
+}
+
+// Populate those fields returned by getHotColdSchema.
+MemInfoBlock makePartialMIB() {
+  MemInfoBlock MIB;
+  MIB.AllocCount = 1;
+  MIB.TotalSize = 5;
+  MIB.TotalLifetime = 10;
+  MIB.TotalLifetimeAccessDensity = 23;
+  return MIB;
+}
+
 IndexedMemProfRecord makeRecord(
     std::initializer_list<std::initializer_list<::llvm::memprof::FrameId>>
         AllocFrames,
     std::initializer_list<std::initializer_list<::llvm::memprof::FrameId>>
         CallSiteFrames,
-    const MemInfoBlock &Block = MemInfoBlock()) {
+    const MemInfoBlock &Block = makeFullMIB()) {
   llvm::memprof::IndexedMemProfRecord MR;
   for (const auto &Frames : AllocFrames)
     MR.AllocSites.emplace_back(Frames, llvm::memprof::hashCallStack(Frames),
@@ -388,13 +407,13 @@ IndexedMemProfRecord makeRecord(
 IndexedMemProfRecord
 makeRecordV2(std::initializer_list<::llvm::memprof::CallStackId> AllocFrames,
              std::initializer_list<::llvm::memprof::CallStackId> CallSiteFrames,
-             const MemInfoBlock &Block = MemInfoBlock()) {
+             const MemInfoBlock &Block, const memprof::MemProfSchema &Schema) {
   llvm::memprof::IndexedMemProfRecord MR;
   for (const auto &CSId : AllocFrames)
     // We don't populate IndexedAllocationInfo::CallStack because we use it only
     // in Version0 and Version1.
     MR.AllocSites.emplace_back(::llvm::SmallVector<memprof::FrameId>(), CSId,
-                               Block);
+                               Block, Schema);
   for (const auto &CSId : CallSiteFrames)
     MR.CallSiteIds.push_back(CSId);
   return MR;
@@ -476,15 +495,18 @@ TEST_F(InstrProfTest, test_memprof_v0) {
   EXPECT_THAT(WantRecord, EqualsRecord(Record));
 }
 
-TEST_F(InstrProfTest, test_memprof_v2) {
+TEST_F(InstrProfTest, test_memprof_v2_full_schema) {
+  const MemInfoBlock MIB = makeFullMIB();
+
   Writer.setMemProfVersionRequested(memprof::Version2);
+  Writer.setMemProfFullSchema(true);
 
   ASSERT_THAT_ERROR(Writer.mergeProfileKind(InstrProfKind::MemProf),
                     Succeeded());
 
   const IndexedMemProfRecord IndexedMR = makeRecordV2(
       /*AllocFrames=*/{0x111, 0x222},
-      /*CallSiteFrames=*/{0x333});
+      /*CallSiteFrames=*/{0x333}, MIB, memprof::getFullSchema());
   const FrameIdMapTy IdToFrameMap = getFrameMapping();
   const auto CSIdToCallStackMap = getCallStackMapping();
   for (const auto &I : IdToFrameMap) {
@@ -502,38 +524,58 @@ TEST_F(InstrProfTest, test_memprof_v2) {
   ASSERT_THAT_ERROR(RecordOr.takeError(), Succeeded());
   const memprof::MemProfRecord &Record = RecordOr.get();
 
-  std::optional<memprof::FrameId> LastUnmappedFrameId;
-  auto IdToFrameCallback = [&](const memprof::FrameId Id) {
-    auto Iter = IdToFrameMap.find(Id);
-    if (Iter == IdToFrameMap.end()) {
-      LastUnmappedFrameId = Id;
-      return memprof::Frame(0, 0, 0, false);
-    }
-    return Iter->second;
-  };
+  memprof::FrameIdConverter<decltype(IdToFrameMap)> FrameIdConv(IdToFrameMap);
+  memprof::CallStackIdConverter<decltype(CSIdToCallStackMap)> CSIdConv(
+      CSIdToCallStackMap, FrameIdConv);
 
-  std::optional<::llvm::memprof::CallStackId> LastUnmappedCSId;
-  auto CSIdToCallStackCallback = [&](::llvm::memprof::CallStackId CSId) {
-    llvm::SmallVector<memprof::Frame> Frames;
-    auto CSIter = CSIdToCallStackMap.find(CSId);
-    if (CSIter == CSIdToCallStackMap.end()) {
-      LastUnmappedCSId = CSId;
-    } else {
-      const ::llvm::SmallVector<::llvm::memprof::FrameId> &CS =
-          CSIter->getSecond();
-      Frames.reserve(CS.size());
-      for (::llvm::memprof::FrameId Id : CS)
-        Frames.push_back(IdToFrameCallback(Id));
-    }
-    return Frames;
-  };
+  const ::llvm::memprof::MemProfRecord WantRecord =
+      IndexedMR.toMemProfRecord(CSIdConv);
+  ASSERT_EQ(FrameIdConv.LastUnmappedId, std::nullopt)
+      << "could not map frame id: " << *FrameIdConv.LastUnmappedId;
+  ASSERT_EQ(CSIdConv.LastUnmappedId, std::nullopt)
+      << "could not map call stack id: " << *CSIdConv.LastUnmappedId;
+  EXPECT_THAT(WantRecord, EqualsRecord(Record));
+}
+
+TEST_F(InstrProfTest, test_memprof_v2_partial_schema) {
+  const MemInfoBlock MIB = makePartialMIB();
+
+  Writer.setMemProfVersionRequested(memprof::Version2);
+  Writer.setMemProfFullSchema(false);
+
+  ASSERT_THAT_ERROR(Writer.mergeProfileKind(InstrProfKind::MemProf),
+                    Succeeded());
+
+  const IndexedMemProfRecord IndexedMR = makeRecordV2(
+      /*AllocFrames=*/{0x111, 0x222},
+      /*CallSiteFrames=*/{0x333}, MIB, memprof::getHotColdSchema());
+  const FrameIdMapTy IdToFrameMap = getFrameMapping();
+  const auto CSIdToCallStackMap = getCallStackMapping();
+  for (const auto &I : IdToFrameMap) {
+    Writer.addMemProfFrame(I.first, I.getSecond(), Err);
+  }
+  for (const auto &I : CSIdToCallStackMap) {
+    Writer.addMemProfCallStack(I.first, I.getSecond(), Err);
+  }
+  Writer.addMemProfRecord(/*Id=*/0x9999, IndexedMR);
+
+  auto Profile = Writer.writeBuffer();
+  readProfile(std::move(Profile));
+
+  auto RecordOr = Reader->getMemProfRecord(0x9999);
+  ASSERT_THAT_ERROR(RecordOr.takeError(), Succeeded());
+  const memprof::MemProfRecord &Record = RecordOr.get();
+
+  memprof::FrameIdConverter<decltype(IdToFrameMap)> FrameIdConv(IdToFrameMap);
+  memprof::CallStackIdConverter<decltype(CSIdToCallStackMap)> CSIdConv(
+      CSIdToCallStackMap, FrameIdConv);
 
   const ::llvm::memprof::MemProfRecord WantRecord =
-      IndexedMR.toMemProfRecord(CSIdToCallStackCallback);
-  ASSERT_EQ(LastUnmappedFrameId, std::nullopt)
-      << "could not map frame id: " << *LastUnmappedFrameId;
-  ASSERT_EQ(LastUnmappedCSId, std::nullopt)
-      << "could not map call stack id: " << *LastUnmappedCSId;
+      IndexedMR.toMemProfRecord(CSIdConv);
+  ASSERT_EQ(FrameIdConv.LastUnmappedId, std::nullopt)
+      << "could not map frame id: " << *FrameIdConv.LastUnmappedId;
+  ASSERT_EQ(CSIdConv.LastUnmappedId, std::nullopt)
+      << "could not map call stack id: " << *CSIdConv.LastUnmappedId;
   EXPECT_THAT(WantRecord, EqualsRecord(Record));
 }
 
diff --git a/llvm/unittests/ProfileData/MemProfTest.cpp b/llvm/unittests/ProfileData/MemProfTest.cpp
index 7e00a80cacf9..40335d191ba7 100644
--- a/llvm/unittests/ProfileData/MemProfTest.cpp
+++ b/llvm/unittests/ProfileData/MemProfTest.cpp
@@ -1,6 +1,7 @@
 #include "llvm/ProfileData/MemProf.h"
 #include "llvm/ADT/DenseMap.h"
 #include "llvm/ADT/MapVector.h"
+#include "llvm/ADT/STLForwardCompat.h"
 #include "llvm/DebugInfo/DIContext.h"
 #include "llvm/DebugInfo/Symbolize/SymbolizableModule.h"
 #include "llvm/IR/Value.h"
@@ -122,14 +123,6 @@ MATCHER_P4(FrameContains, FunctionName, LineOffset, Column, Inline, "") {
   return false;
 }
 
-MemProfSchema getFullSchema() {
-  MemProfSchema Schema;
-#define MIBEntryDef(NameTag, Name, Type) Schema.push_back(Meta::Name);
-#include "llvm/ProfileData/MIBEntryDef.inc"
-#undef MIBEntryDef
-  return Schema;
-}
-
 TEST(MemProf, FillsValue) {
   std::unique_ptr<MockSymbolizer> Symbolizer(new MockSymbolizer());
 
@@ -187,7 +180,7 @@ TEST(MemProf, FillsValue) {
 
   // Check the memprof record for foo.
   const llvm::GlobalValue::GUID FooId = IndexedMemProfRecord::getGUID("foo");
-  ASSERT_EQ(Records.count(FooId), 1U);
+  ASSERT_TRUE(Records.contains(FooId));
   const MemProfRecord &Foo = Records[FooId];
   ASSERT_THAT(Foo.AllocSites, SizeIs(1));
   EXPECT_EQ(Foo.AllocSites[0].Info.getAllocCount(), 1U);
@@ -203,7 +196,7 @@ TEST(MemProf, FillsValue) {
 
   // Check the memprof record for bar.
   const llvm::GlobalValue::GUID BarId = IndexedMemProfRecord::getGUID("bar");
-  ASSERT_EQ(Records.count(BarId), 1U);
+  ASSERT_TRUE(Records.contains(BarId));
   const MemProfRecord &Bar = Records[BarId];
   ASSERT_THAT(Bar.AllocSites, SizeIs(1));
   EXPECT_EQ(Bar.AllocSites[0].Info.getAllocCount(), 1U);
@@ -223,7 +216,7 @@ TEST(MemProf, FillsValue) {
 
   // Check the memprof record for xyz.
   const llvm::GlobalValue::GUID XyzId = IndexedMemProfRecord::getGUID("xyz");
-  ASSERT_EQ(Records.count(XyzId), 1U);
+  ASSERT_TRUE(Records.contains(XyzId));
   const MemProfRecord &Xyz = Records[XyzId];
   ASSERT_THAT(Xyz.CallSites, SizeIs(1));
   ASSERT_THAT(Xyz.CallSites[0], SizeIs(2));
@@ -234,7 +227,7 @@ TEST(MemProf, FillsValue) {
 
   // Check the memprof record for abc.
   const llvm::GlobalValue::GUID AbcId = IndexedMemProfRecord::getGUID("abc");
-  ASSERT_EQ(Records.count(AbcId), 1U);
+  ASSERT_TRUE(Records.contains(AbcId));
   const MemProfRecord &Abc = Records[AbcId];
   EXPECT_TRUE(Abc.AllocSites.empty());
   ASSERT_THAT(Abc.CallSites, SizeIs(1));
@@ -248,8 +241,8 @@ TEST(MemProf, PortableWrapper) {
                     /*dealloc_timestamp=*/2000, /*alloc_cpu=*/3,
                     /*dealloc_cpu=*/4);
 
-  const auto Schema = getFullSchema();
-  PortableMemInfoBlock WriteBlock(Info);
+  const auto Schema = llvm::memprof::getFullSchema();
+  PortableMemInfoBlock WriteBlock(Info, Schema);
 
   std::string Buffer;
   llvm::raw_string_ostream OS(Buffer);
@@ -271,7 +264,7 @@ TEST(MemProf, PortableWrapper) {
 // Version0 and Version1 serialize IndexedMemProfRecord in the same format, so
 // we share one test.
 TEST(MemProf, RecordSerializationRoundTripVersion0And1) {
-  const MemProfSchema Schema = getFullSchema();
+  const auto Schema = llvm::memprof::getFullSchema();
 
   MemInfoBlock Info(/*size=*/16, /*access_count=*/7, /*alloc_timestamp=*/1000,
                     /*dealloc_timestamp=*/2000, /*alloc_cpu=*/3,
@@ -305,7 +298,7 @@ TEST(MemProf, RecordSerializationRoundTripVersion0And1) {
 }
 
 TEST(MemProf, RecordSerializationRoundTripVerion2) {
-  const MemProfSchema Schema = getFullSchema();
+  const auto Schema = llvm::memprof::getFullSchema();
 
   MemInfoBlock Info(/*size=*/16, /*access_count=*/7, /*alloc_timestamp=*/1000,
                     /*dealloc_timestamp=*/2000, /*alloc_cpu=*/3,
@@ -334,6 +327,65 @@ TEST(MemProf, RecordSerializationRoundTripVerion2) {
   EXPECT_EQ(Record, GotRecord);
 }
 
+TEST(MemProf, RecordSerializationRoundTripVersion2HotColdSchema) {
+  const auto Schema = llvm::memprof::getHotColdSchema();
+
+  MemInfoBlock Info;
+  Info.AllocCount = 11;
+  Info.TotalSize = 22;
+  Info.TotalLifetime = 33;
+  Info.TotalLifetimeAccessDensity = 44;
+
+  llvm::SmallVector<llvm::memprof::CallStackId> CallStackIds = {0x123, 0x456};
+
+  llvm::SmallVector<llvm::memprof::CallStackId> CallSiteIds = {0x333, 0x444};
+
+  IndexedMemProfRecord Record;
+  for (const auto &CSId : CallStackIds) {
+    // Use the same info block for both allocation sites.
+    Record.AllocSites.emplace_back(llvm::SmallVector<FrameId>(), CSId, Info,
+                                   Schema);
+  }
+  Record.CallSiteIds.assign(CallSiteIds);
+
+  std::bitset<llvm::to_underlying(Meta::Size)> SchemaBitSet;
+  for (auto Id : Schema)
+    SchemaBitSet.set(llvm::to_underlying(Id));
+
+  // Verify that SchemaBitSet has the fields we expect and nothing else, which
+  // we check with count().
+  EXPECT_EQ(SchemaBitSet.count(), 4U);
+  EXPECT_TRUE(SchemaBitSet[llvm::to_underlying(Meta::AllocCount)]);
+  EXPECT_TRUE(SchemaBitSet[llvm::to_underlying(Meta::TotalSize)]);
+  EXPECT_TRUE(SchemaBitSet[llvm::to_underlying(Meta::TotalLifetime)]);
+  EXPECT_TRUE(
+      SchemaBitSet[llvm::to_underlying(Meta::TotalLifetimeAccessDensity)]);
+
+  // Verify that Schema has propagated all the way to the Info field in each
+  // IndexedAllocationInfo.
+  ASSERT_THAT(Record.AllocSites, ::SizeIs(2));
+  EXPECT_EQ(Record.AllocSites[0].Info.getSchema(), SchemaBitSet);
+  EXPECT_EQ(Record.AllocSites[1].Info.getSchema(), SchemaBitSet);
+
+  std::string Buffer;
+  llvm::raw_string_ostream OS(Buffer);
+  Record.serialize(Schema, OS, llvm::memprof::Version2);
+  OS.flush();
+
+  const IndexedMemProfRecord GotRecord = IndexedMemProfRecord::deserialize(
+      Schema, reinterpret_cast<const unsigned char *>(Buffer.data()),
+      llvm::memprof::Version2);
+
+  // Verify that Schema comes back correctly after deserialization. Technically,
+  // the comparison between Record and GotRecord below includes the comparison
+  // of their Schemas, but we'll verify the Schemas on our own.
+  ASSERT_THAT(GotRecord.AllocSites, ::SizeIs(2));
+  EXPECT_EQ(GotRecord.AllocSites[0].Info.getSchema(), SchemaBitSet);
+  EXPECT_EQ(GotRecord.AllocSites[1].Info.getSchema(), SchemaBitSet);
+
+  EXPECT_EQ(Record, GotRecord);
+}
+
 TEST(MemProf, SymbolizationFilter) {
   std::unique_ptr<MockSymbolizer> Symbolizer(new MockSymbolizer());
 
@@ -510,37 +562,15 @@ TEST(MemProf, IndexedMemProfRecordToMemProfRecord) {
   IndexedRecord.CallSiteIds.push_back(llvm::memprof::hashCallStack(CS3));
   IndexedRecord.CallSiteIds.push_back(llvm::memprof::hashCallStack(CS4));
 
-  bool CSIdMissing = false;
-  bool FrameIdMissing = false;
-
-  auto Callback = [&](CallStackId CSId) -> llvm::SmallVector<Frame> {
-    llvm::SmallVector<Frame> CallStack;
-    llvm::SmallVector<FrameId> FrameIds;
-
-    auto Iter = CallStackIdMap.find(CSId);
-    if (Iter == CallStackIdMap.end())
-      CSIdMissing = true;
-    else
-      FrameIds = Iter->second;
-
-    for (FrameId Id : FrameIds) {
-      Frame F(0, 0, 0, false);
-      auto Iter = FrameIdMap.find(Id);
-      if (Iter == FrameIdMap.end())
-        FrameIdMissing = true;
-      else
-        F = Iter->second;
-      CallStack.push_back(F);
-    }
-
-    return CallStack;
-  };
+  llvm::memprof::FrameIdConverter<decltype(FrameIdMap)> FrameIdConv(FrameIdMap);
+  llvm::memprof::CallStackIdConverter<decltype(CallStackIdMap)> CSIdConv(
+      CallStackIdMap, FrameIdConv);
 
-  MemProfRecord Record = IndexedRecord.toMemProfRecord(Callback);
+  MemProfRecord Record = IndexedRecord.toMemProfRecord(CSIdConv);
 
   // Make sure that all lookups are successful.
-  ASSERT_FALSE(CSIdMissing);
-  ASSERT_FALSE(FrameIdMissing);
+  ASSERT_EQ(FrameIdConv.LastUnmappedId, std::nullopt);
+  ASSERT_EQ(CSIdConv.LastUnmappedId, std::nullopt);
 
   // Verify the contents of Record.
   ASSERT_THAT(Record.AllocSites, SizeIs(2));
diff --git a/llvm/unittests/Support/CMakeLists.txt b/llvm/unittests/Support/CMakeLists.txt
index 15a126279125..2718be8450f8 100644
--- a/llvm/unittests/Support/CMakeLists.txt
+++ b/llvm/unittests/Support/CMakeLists.txt
@@ -71,7 +71,6 @@ add_llvm_unittest(SupportTests
   ReverseIterationTest.cpp
   ReplaceFileTest.cpp
   RISCVAttributeParserTest.cpp
-  RISCVISAInfoTest.cpp
   ScaledNumberTest.cpp
   ScopedPrinterTest.cpp
   SHA256.cpp
diff --git a/llvm/unittests/Support/MathExtrasTest.cpp b/llvm/unittests/Support/MathExtrasTest.cpp
index d09f987b9d0f..67239a24c47a 100644
--- a/llvm/unittests/Support/MathExtrasTest.cpp
+++ b/llvm/unittests/Support/MathExtrasTest.cpp
@@ -77,7 +77,7 @@ TEST(MathExtras, maxUIntN) {
   EXPECT_EQ(0xffffffffffffffffULL, maxUIntN(64));
   EXPECT_EQ(1ULL, maxUIntN(1));
   EXPECT_EQ(0x0fULL, maxUIntN(4));
-  EXPECT_EQ(0, maxUIntN(0));
+  EXPECT_EQ(0ULL, maxUIntN(0));
 }
 
 TEST(MathExtras, reverseBits) {
diff --git a/llvm/unittests/Support/YAMLIOTest.cpp b/llvm/unittests/Support/YAMLIOTest.cpp
index 401981f3841e..6ac0d1b412f0 100644
--- a/llvm/unittests/Support/YAMLIOTest.cpp
+++ b/llvm/unittests/Support/YAMLIOTest.cpp
@@ -2906,6 +2906,87 @@ TEST(YAMLIO, Numeric) {
 }
 
 //===----------------------------------------------------------------------===//
+//  Test writing and reading escaped keys
+//===----------------------------------------------------------------------===//
+
+// Struct with dynamic string key
+struct QuotedKeyStruct {
+  int unquoted_bool;
+  int unquoted_null;
+  int unquoted_numeric;
+  int unquoted_str;
+  int colon;
+  int just_space;
+  int unprintable;
+};
+
+namespace llvm {
+namespace yaml {
+template <> struct MappingTraits<QuotedKeyStruct> {
+  static void mapping(IO &io, QuotedKeyStruct &map) {
+    io.mapRequired("true", map.unquoted_bool);
+    io.mapRequired("null", map.unquoted_null);
+    io.mapRequired("42", map.unquoted_numeric);
+    io.mapRequired("unquoted", map.unquoted_str);
+    io.mapRequired(":", map.colon);
+    io.mapRequired(" ", map.just_space);
+    char unprintableKey[] = {/* \f, form-feed */ 0xC, 0};
+    io.mapRequired(unprintableKey, map.unprintable);
+  }
+};
+} // namespace yaml
+} // namespace llvm
+
+TEST(YAMLIO, TestQuotedKeyRead) {
+  QuotedKeyStruct map = {};
+  Input yin("---\ntrue:  1\nnull:  2\n42:  3\nunquoted:  4\n':':  5\n' ':  "
+            "6\n\"\\f\":  7\n...\n");
+  yin >> map;
+
+  EXPECT_FALSE(yin.error());
+  EXPECT_EQ(map.unquoted_bool, 1);
+  EXPECT_EQ(map.unquoted_null, 2);
+  EXPECT_EQ(map.unquoted_numeric, 3);
+  EXPECT_EQ(map.unquoted_str, 4);
+  EXPECT_EQ(map.colon, 5);
+  EXPECT_EQ(map.just_space, 6);
+  EXPECT_EQ(map.unprintable, 7);
+}
+
+TEST(YAMLIO, TestQuotedKeyWriteRead) {
+  std::string intermediate;
+  {
+    QuotedKeyStruct map = {1, 2, 3, 4, 5, 6, 7};
+    llvm::raw_string_ostream ostr(intermediate);
+    Output yout(ostr);
+    yout << map;
+  }
+
+  EXPECT_NE(std::string::npos, intermediate.find("true:"));
+  EXPECT_NE(std::string::npos, intermediate.find("null:"));
+  EXPECT_NE(std::string::npos, intermediate.find("42:"));
+  EXPECT_NE(std::string::npos, intermediate.find("unquoted:"));
+  EXPECT_NE(std::string::npos, intermediate.find("':':"));
+  EXPECT_NE(std::string::npos, intermediate.find("' '"));
+  EXPECT_NE(std::string::npos, intermediate.find("\"\\f\":"));
+
+  {
+    Input yin(intermediate);
+    QuotedKeyStruct map;
+    yin >> map;
+
+    EXPECT_FALSE(yin.error());
+    EXPECT_EQ(map.unquoted_bool, 1);
+    EXPECT_EQ(map.unquoted_null, 2);
+    EXPECT_EQ(map.unquoted_numeric, 3);
+    EXPECT_EQ(map.unquoted_str, 4);
+    EXPECT_EQ(map.colon, 5);
+    EXPECT_EQ(map.just_space, 6);
+    EXPECT_EQ(map.unprintable, 7);
+  }
+}
+
+//===----------------------------------------------------------------------===//
 //  Test PolymorphicTraits and TaggedScalarTraits
 //===----------------------------------------------------------------------===//
 
diff --git a/llvm/unittests/TargetParser/CMakeLists.txt b/llvm/unittests/TargetParser/CMakeLists.txt
index 3bbc74f3f8d3..086c57903716 100644
--- a/llvm/unittests/TargetParser/CMakeLists.txt
+++ b/llvm/unittests/TargetParser/CMakeLists.txt
@@ -6,6 +6,7 @@ set(LLVM_LINK_COMPONENTS
 add_llvm_unittest(TargetParserTests
   CSKYTargetParserTest.cpp
   Host.cpp
+  RISCVISAInfoTest.cpp
   RISCVTargetParserTest.cpp
   TargetParserTest.cpp
   TripleTest.cpp
diff --git a/llvm/unittests/Support/RISCVISAInfoTest.cpp b/llvm/unittests/TargetParser/RISCVISAInfoTest.cpp
index caf7bf0a3171..9f23000d733d 100644
--- a/llvm/unittests/Support/RISCVISAInfoTest.cpp
+++ b/llvm/unittests/TargetParser/RISCVISAInfoTest.cpp
@@ -6,8 +6,8 @@
 //
 //===----------------------------------------------------------------------===//
 
+#include "llvm/TargetParser/RISCVISAInfo.h"
 #include "llvm/ADT/StringMap.h"
-#include "llvm/Support/RISCVISAInfo.h"
 #include "llvm/Testing/Support/Error.h"
 #include "gtest/gtest.h"
 
@@ -15,8 +15,8 @@ using ::testing::ElementsAre;
 
 using namespace llvm;
 
-bool operator==(const RISCVISAInfo::ExtensionVersion &A,
-                const RISCVISAInfo::ExtensionVersion &B) {
+bool operator==(const RISCVISAUtils::ExtensionVersion &A,
+                const RISCVISAUtils::ExtensionVersion &B) {
   return A.Major == B.Major && A.Minor == B.Minor;
 }
 
@@ -51,7 +51,7 @@ TEST(ParseNormalizedArchString, AcceptsValidBaseISAsAndSetsXLen) {
   RISCVISAInfo &InfoRV32I = **MaybeRV32I;
   EXPECT_EQ(InfoRV32I.getExtensions().size(), 1UL);
   EXPECT_TRUE(InfoRV32I.getExtensions().at("i") ==
-              (RISCVISAInfo::ExtensionVersion{2, 0}));
+              (RISCVISAUtils::ExtensionVersion{2, 0}));
   EXPECT_EQ(InfoRV32I.getXLen(), 32U);
 
   auto MaybeRV32E = RISCVISAInfo::parseNormalizedArchString("rv32e2p0");
@@ -59,7 +59,7 @@ TEST(ParseNormalizedArchString, AcceptsValidBaseISAsAndSetsXLen) {
   RISCVISAInfo &InfoRV32E = **MaybeRV32E;
   EXPECT_EQ(InfoRV32E.getExtensions().size(), 1UL);
   EXPECT_TRUE(InfoRV32E.getExtensions().at("e") ==
-              (RISCVISAInfo::ExtensionVersion{2, 0}));
+              (RISCVISAUtils::ExtensionVersion{2, 0}));
   EXPECT_EQ(InfoRV32E.getXLen(), 32U);
 
   auto MaybeRV64I = RISCVISAInfo::parseNormalizedArchString("rv64i2p0");
@@ -67,7 +67,7 @@ TEST(ParseNormalizedArchString, AcceptsValidBaseISAsAndSetsXLen) {
   RISCVISAInfo &InfoRV64I = **MaybeRV64I;
   EXPECT_EQ(InfoRV64I.getExtensions().size(), 1UL);
   EXPECT_TRUE(InfoRV64I.getExtensions().at("i") ==
-              (RISCVISAInfo::ExtensionVersion{2, 0}));
+              (RISCVISAUtils::ExtensionVersion{2, 0}));
   EXPECT_EQ(InfoRV64I.getXLen(), 64U);
 
   auto MaybeRV64E = RISCVISAInfo::parseNormalizedArchString("rv64e2p0");
@@ -75,7 +75,7 @@ TEST(ParseNormalizedArchString, AcceptsValidBaseISAsAndSetsXLen) {
   RISCVISAInfo &InfoRV64E = **MaybeRV64E;
   EXPECT_EQ(InfoRV64E.getExtensions().size(), 1UL);
   EXPECT_TRUE(InfoRV64E.getExtensions().at("e") ==
-              (RISCVISAInfo::ExtensionVersion{2, 0}));
+              (RISCVISAUtils::ExtensionVersion{2, 0}));
   EXPECT_EQ(InfoRV64E.getXLen(), 64U);
 }
 
@@ -86,15 +86,15 @@ TEST(ParseNormalizedArchString, AcceptsArbitraryExtensionsAndVersions) {
   RISCVISAInfo &Info = **MaybeISAInfo;
   EXPECT_EQ(Info.getExtensions().size(), 5UL);
   EXPECT_TRUE(Info.getExtensions().at("i") ==
-              (RISCVISAInfo::ExtensionVersion{5, 1}));
+              (RISCVISAUtils::ExtensionVersion{5, 1}));
   EXPECT_TRUE(Info.getExtensions().at("m") ==
-              (RISCVISAInfo::ExtensionVersion{3, 2}));
+              (RISCVISAUtils::ExtensionVersion{3, 2}));
   EXPECT_TRUE(Info.getExtensions().at("zmadeup") ==
-              (RISCVISAInfo::ExtensionVersion{11, 12}));
+              (RISCVISAUtils::ExtensionVersion{11, 12}));
   EXPECT_TRUE(Info.getExtensions().at("sfoo") ==
-              (RISCVISAInfo::ExtensionVersion{2, 0}));
+              (RISCVISAUtils::ExtensionVersion{2, 0}));
   EXPECT_TRUE(Info.getExtensions().at("xbar") ==
-              (RISCVISAInfo::ExtensionVersion{3, 0}));
+              (RISCVISAUtils::ExtensionVersion{3, 0}));
 }
 
 TEST(ParseNormalizedArchString, UpdatesFLenMinVLenMaxELen) {
@@ -120,10 +120,14 @@ TEST(ParseArchString, RejectsInvalidBaseISA) {
     EXPECT_EQ(toString(RISCVISAInfo::parseArchString(Input, true).takeError()),
               "string must begin with rv32{i,e,g} or rv64{i,e,g}");
   }
-  for (StringRef Input : {"rv32j", "rv64k", "rv32_i"}) {
+
+  for (StringRef Input : {"rv32j", "rv32_i"}) {
     EXPECT_EQ(toString(RISCVISAInfo::parseArchString(Input, true).takeError()),
-              "first letter should be 'e', 'i' or 'g'");
+              "first letter after 'rv32' should be 'e', 'i' or 'g'");
   }
+
+  EXPECT_EQ(toString(RISCVISAInfo::parseArchString("rv64k", true).takeError()),
+            "first letter after 'rv64' should be 'e', 'i' or 'g'");
 }
 
 TEST(ParseArchString, RejectsUnsupportedBaseISA) {
@@ -137,68 +141,68 @@ TEST(ParseArchString, AcceptsSupportedBaseISAsAndSetsXLenAndFLen) {
   auto MaybeRV32I = RISCVISAInfo::parseArchString("rv32i", true);
   ASSERT_THAT_EXPECTED(MaybeRV32I, Succeeded());
   RISCVISAInfo &InfoRV32I = **MaybeRV32I;
-  RISCVISAInfo::OrderedExtensionMap ExtsRV32I = InfoRV32I.getExtensions();
+  const auto &ExtsRV32I = InfoRV32I.getExtensions();
   EXPECT_EQ(ExtsRV32I.size(), 1UL);
-  EXPECT_TRUE(ExtsRV32I.at("i") == (RISCVISAInfo::ExtensionVersion{2, 1}));
+  EXPECT_TRUE(ExtsRV32I.at("i") == (RISCVISAUtils::ExtensionVersion{2, 1}));
   EXPECT_EQ(InfoRV32I.getXLen(), 32U);
   EXPECT_EQ(InfoRV32I.getFLen(), 0U);
 
   auto MaybeRV32E = RISCVISAInfo::parseArchString("rv32e", true);
   ASSERT_THAT_EXPECTED(MaybeRV32E, Succeeded());
   RISCVISAInfo &InfoRV32E = **MaybeRV32E;
-  RISCVISAInfo::OrderedExtensionMap ExtsRV32E = InfoRV32E.getExtensions();
+  const auto &ExtsRV32E = InfoRV32E.getExtensions();
   EXPECT_EQ(ExtsRV32E.size(), 1UL);
-  EXPECT_TRUE(ExtsRV32E.at("e") == (RISCVISAInfo::ExtensionVersion{2, 0}));
+  EXPECT_TRUE(ExtsRV32E.at("e") == (RISCVISAUtils::ExtensionVersion{2, 0}));
   EXPECT_EQ(InfoRV32E.getXLen(), 32U);
   EXPECT_EQ(InfoRV32E.getFLen(), 0U);
 
   auto MaybeRV32G = RISCVISAInfo::parseArchString("rv32g", true);
   ASSERT_THAT_EXPECTED(MaybeRV32G, Succeeded());
   RISCVISAInfo &InfoRV32G = **MaybeRV32G;
-  RISCVISAInfo::OrderedExtensionMap ExtsRV32G = InfoRV32G.getExtensions();
+  const auto &ExtsRV32G = InfoRV32G.getExtensions();
   EXPECT_EQ(ExtsRV32G.size(), 7UL);
-  EXPECT_TRUE(ExtsRV32G.at("i") == (RISCVISAInfo::ExtensionVersion{2, 1}));
-  EXPECT_TRUE(ExtsRV32G.at("m") == (RISCVISAInfo::ExtensionVersion{2, 0}));
-  EXPECT_TRUE(ExtsRV32G.at("a") == (RISCVISAInfo::ExtensionVersion{2, 1}));
-  EXPECT_TRUE(ExtsRV32G.at("f") == (RISCVISAInfo::ExtensionVersion{2, 2}));
-  EXPECT_TRUE(ExtsRV32G.at("d") == (RISCVISAInfo::ExtensionVersion{2, 2}));
-  EXPECT_TRUE(ExtsRV32G.at("zicsr") == (RISCVISAInfo::ExtensionVersion{2, 0}));
+  EXPECT_TRUE(ExtsRV32G.at("i") == (RISCVISAUtils::ExtensionVersion{2, 1}));
+  EXPECT_TRUE(ExtsRV32G.at("m") == (RISCVISAUtils::ExtensionVersion{2, 0}));
+  EXPECT_TRUE(ExtsRV32G.at("a") == (RISCVISAUtils::ExtensionVersion{2, 1}));
+  EXPECT_TRUE(ExtsRV32G.at("f") == (RISCVISAUtils::ExtensionVersion{2, 2}));
+  EXPECT_TRUE(ExtsRV32G.at("d") == (RISCVISAUtils::ExtensionVersion{2, 2}));
+  EXPECT_TRUE(ExtsRV32G.at("zicsr") == (RISCVISAUtils::ExtensionVersion{2, 0}));
   EXPECT_TRUE(ExtsRV32G.at("zifencei") ==
-              (RISCVISAInfo::ExtensionVersion{2, 0}));
+              (RISCVISAUtils::ExtensionVersion{2, 0}));
   EXPECT_EQ(InfoRV32G.getXLen(), 32U);
   EXPECT_EQ(InfoRV32G.getFLen(), 64U);
 
   auto MaybeRV64I = RISCVISAInfo::parseArchString("rv64i", true);
   ASSERT_THAT_EXPECTED(MaybeRV64I, Succeeded());
   RISCVISAInfo &InfoRV64I = **MaybeRV64I;
-  RISCVISAInfo::OrderedExtensionMap ExtsRV64I = InfoRV64I.getExtensions();
+  const auto &ExtsRV64I = InfoRV64I.getExtensions();
   EXPECT_EQ(ExtsRV64I.size(), 1UL);
-  EXPECT_TRUE(ExtsRV64I.at("i") == (RISCVISAInfo::ExtensionVersion{2, 1}));
+  EXPECT_TRUE(ExtsRV64I.at("i") == (RISCVISAUtils::ExtensionVersion{2, 1}));
   EXPECT_EQ(InfoRV64I.getXLen(), 64U);
   EXPECT_EQ(InfoRV64I.getFLen(), 0U);
 
   auto MaybeRV64E = RISCVISAInfo::parseArchString("rv64e", true);
   ASSERT_THAT_EXPECTED(MaybeRV64E, Succeeded());
   RISCVISAInfo &InfoRV64E = **MaybeRV64E;
-  RISCVISAInfo::OrderedExtensionMap ExtsRV64E = InfoRV64E.getExtensions();
+  const auto &ExtsRV64E = InfoRV64E.getExtensions();
   EXPECT_EQ(ExtsRV64E.size(), 1UL);
-  EXPECT_TRUE(ExtsRV64E.at("e") == (RISCVISAInfo::ExtensionVersion{2, 0}));
+  EXPECT_TRUE(ExtsRV64E.at("e") == (RISCVISAUtils::ExtensionVersion{2, 0}));
   EXPECT_EQ(InfoRV64E.getXLen(), 64U);
   EXPECT_EQ(InfoRV64E.getFLen(), 0U);
 
   auto MaybeRV64G = RISCVISAInfo::parseArchString("rv64g", true);
   ASSERT_THAT_EXPECTED(MaybeRV64G, Succeeded());
   RISCVISAInfo &InfoRV64G = **MaybeRV64G;
-  RISCVISAInfo::OrderedExtensionMap ExtsRV64G = InfoRV64G.getExtensions();
+  const auto &ExtsRV64G = InfoRV64G.getExtensions();
   EXPECT_EQ(ExtsRV64G.size(), 7UL);
-  EXPECT_TRUE(ExtsRV64G.at("i") == (RISCVISAInfo::ExtensionVersion{2, 1}));
-  EXPECT_TRUE(ExtsRV64G.at("m") == (RISCVISAInfo::ExtensionVersion{2, 0}));
-  EXPECT_TRUE(ExtsRV64G.at("a") == (RISCVISAInfo::ExtensionVersion{2, 1}));
-  EXPECT_TRUE(ExtsRV64G.at("f") == (RISCVISAInfo::ExtensionVersion{2, 2}));
-  EXPECT_TRUE(ExtsRV64G.at("d") == (RISCVISAInfo::ExtensionVersion{2, 2}));
-  EXPECT_TRUE(ExtsRV64G.at("zicsr") == (RISCVISAInfo::ExtensionVersion{2, 0}));
+  EXPECT_TRUE(ExtsRV64G.at("i") == (RISCVISAUtils::ExtensionVersion{2, 1}));
+  EXPECT_TRUE(ExtsRV64G.at("m") == (RISCVISAUtils::ExtensionVersion{2, 0}));
+  EXPECT_TRUE(ExtsRV64G.at("a") == (RISCVISAUtils::ExtensionVersion{2, 1}));
+  EXPECT_TRUE(ExtsRV64G.at("f") == (RISCVISAUtils::ExtensionVersion{2, 2}));
+  EXPECT_TRUE(ExtsRV64G.at("d") == (RISCVISAUtils::ExtensionVersion{2, 2}));
+  EXPECT_TRUE(ExtsRV64G.at("zicsr") == (RISCVISAUtils::ExtensionVersion{2, 0}));
   EXPECT_TRUE(ExtsRV64G.at("zifencei") ==
-              (RISCVISAInfo::ExtensionVersion{2, 0}));
+              (RISCVISAUtils::ExtensionVersion{2, 0}));
   EXPECT_EQ(InfoRV64G.getXLen(), 64U);
   EXPECT_EQ(InfoRV64G.getFLen(), 64U);
 }
@@ -241,9 +245,9 @@ TEST(ParseArchString, IgnoresUnrecognizedExtensionNamesWithIgnoreUnknown) {
     auto MaybeISAInfo = RISCVISAInfo::parseArchString(Input, true, false, true);
     ASSERT_THAT_EXPECTED(MaybeISAInfo, Succeeded());
     RISCVISAInfo &Info = **MaybeISAInfo;
-    RISCVISAInfo::OrderedExtensionMap Exts = Info.getExtensions();
+    const auto &Exts = Info.getExtensions();
     EXPECT_EQ(Exts.size(), 1UL);
-    EXPECT_TRUE(Exts.at("i") == (RISCVISAInfo::ExtensionVersion{2, 1}));
+    EXPECT_TRUE(Exts.at("i") == (RISCVISAUtils::ExtensionVersion{2, 1}));
   }
 
   // Checks that supported extensions aren't incorrectly ignored when a
@@ -251,22 +255,22 @@ TEST(ParseArchString, IgnoresUnrecognizedExtensionNamesWithIgnoreUnknown) {
   auto MaybeISAInfo =
       RISCVISAInfo::parseArchString("rv32i_zbc1p0_xmadeup", true, false, true);
   ASSERT_THAT_EXPECTED(MaybeISAInfo, Succeeded());
-  RISCVISAInfo::OrderedExtensionMap Exts = (*MaybeISAInfo)->getExtensions();
-  EXPECT_TRUE(Exts.at("zbc") == (RISCVISAInfo::ExtensionVersion{1, 0}));
+  const auto &Exts = (*MaybeISAInfo)->getExtensions();
+  EXPECT_TRUE(Exts.at("zbc") == (RISCVISAUtils::ExtensionVersion{1, 0}));
 }
 
 TEST(ParseArchString, AcceptsVersionInLongOrShortForm) {
   for (StringRef Input : {"rv64i2p1"}) {
     auto MaybeISAInfo = RISCVISAInfo::parseArchString(Input, true);
     ASSERT_THAT_EXPECTED(MaybeISAInfo, Succeeded());
-    RISCVISAInfo::OrderedExtensionMap Exts = (*MaybeISAInfo)->getExtensions();
-    EXPECT_TRUE(Exts.at("i") == (RISCVISAInfo::ExtensionVersion{2, 1}));
+    const auto &Exts = (*MaybeISAInfo)->getExtensions();
+    EXPECT_TRUE(Exts.at("i") == (RISCVISAUtils::ExtensionVersion{2, 1}));
   }
   for (StringRef Input : {"rv32i_zfinx1", "rv32i_zfinx1p0"}) {
     auto MaybeISAInfo = RISCVISAInfo::parseArchString(Input, true);
     ASSERT_THAT_EXPECTED(MaybeISAInfo, Succeeded());
-    RISCVISAInfo::OrderedExtensionMap Exts = (*MaybeISAInfo)->getExtensions();
-    EXPECT_TRUE(Exts.at("zfinx") == (RISCVISAInfo::ExtensionVersion{1, 0}));
+    const auto &Exts = (*MaybeISAInfo)->getExtensions();
+    EXPECT_TRUE(Exts.at("zfinx") == (RISCVISAUtils::ExtensionVersion{1, 0}));
   }
 }
 
@@ -293,16 +297,16 @@ TEST(ParseArchString,
   for (StringRef Input : {"rv32i0p1", "rv32i99p99", "rv64i0p1", "rv64i99p99"}) {
     auto MaybeISAInfo = RISCVISAInfo::parseArchString(Input, true, false, true);
     ASSERT_THAT_EXPECTED(MaybeISAInfo, Succeeded());
-    RISCVISAInfo::OrderedExtensionMap Exts = (*MaybeISAInfo)->getExtensions();
+    const auto &Exts = (*MaybeISAInfo)->getExtensions();
     EXPECT_EQ(Exts.size(), 1UL);
-    EXPECT_TRUE(Exts.at("i") == (RISCVISAInfo::ExtensionVersion{2, 1}));
+    EXPECT_TRUE(Exts.at("i") == (RISCVISAUtils::ExtensionVersion{2, 1}));
   }
   for (StringRef Input : {"rv32e0p1", "rv32e99p99", "rv64e0p1", "rv64e99p99"}) {
     auto MaybeISAInfo = RISCVISAInfo::parseArchString(Input, true, false, true);
     ASSERT_THAT_EXPECTED(MaybeISAInfo, Succeeded());
-    RISCVISAInfo::OrderedExtensionMap Exts = (*MaybeISAInfo)->getExtensions();
+    const auto &Exts = (*MaybeISAInfo)->getExtensions();
     EXPECT_EQ(Exts.size(), 1UL);
-    EXPECT_TRUE(Exts.at("e") == (RISCVISAInfo::ExtensionVersion{2, 0}));
+    EXPECT_TRUE(Exts.at("e") == (RISCVISAUtils::ExtensionVersion{2, 0}));
   }
 }
 
@@ -311,9 +315,9 @@ TEST(ParseArchString,
   for (StringRef Input : {"rv32im1p1", "rv64i_svnapot10p9", "rv32i_zicsr0p5"}) {
     auto MaybeISAInfo = RISCVISAInfo::parseArchString(Input, true, false, true);
     ASSERT_THAT_EXPECTED(MaybeISAInfo, Succeeded());
-    RISCVISAInfo::OrderedExtensionMap Exts = (*MaybeISAInfo)->getExtensions();
+    const auto &Exts = (*MaybeISAInfo)->getExtensions();
     EXPECT_EQ(Exts.size(), 1UL);
-    EXPECT_TRUE(Exts.at("i") == (RISCVISAInfo::ExtensionVersion{2, 1}));
+    EXPECT_TRUE(Exts.at("i") == (RISCVISAUtils::ExtensionVersion{2, 1}));
   }
 }
 
@@ -321,7 +325,7 @@ TEST(ParseArchString, AcceptsUnderscoreSplittingExtensions) {
   for (StringRef Input : {"rv32imafdczifencei", "rv32i_m_a_f_d_c_zifencei"}) {
     auto MaybeISAInfo = RISCVISAInfo::parseArchString(Input, true);
     ASSERT_THAT_EXPECTED(MaybeISAInfo, Succeeded());
-    RISCVISAInfo::OrderedExtensionMap Exts = (*MaybeISAInfo)->getExtensions();
+    const auto &Exts = (*MaybeISAInfo)->getExtensions();
     EXPECT_EQ(Exts.size(), 8UL);
     EXPECT_EQ(Exts.count("i"), 1U);
     EXPECT_EQ(Exts.count("m"), 1U);
@@ -339,7 +343,7 @@ TEST(ParseArchString, AcceptsRelaxSingleLetterExtensions) {
        {"rv32imfad", "rv32im_fa_d", "rv32im2p0fad", "rv32i2p1m2p0fad"}) {
     auto MaybeISAInfo = RISCVISAInfo::parseArchString(Input, true);
     ASSERT_THAT_EXPECTED(MaybeISAInfo, Succeeded());
-    RISCVISAInfo::OrderedExtensionMap Exts = (*MaybeISAInfo)->getExtensions();
+    const auto &Exts = (*MaybeISAInfo)->getExtensions();
     EXPECT_EQ(Exts.size(), 6UL);
     EXPECT_EQ(Exts.count("i"), 1U);
     EXPECT_EQ(Exts.count("m"), 1U);
@@ -356,7 +360,7 @@ TEST(ParseArchString, AcceptsRelaxMixedLetterExtensions) {
         "rv32i_zihintntl_mafd_svinval"}) {
     auto MaybeISAInfo = RISCVISAInfo::parseArchString(Input, true);
     ASSERT_THAT_EXPECTED(MaybeISAInfo, Succeeded());
-    RISCVISAInfo::OrderedExtensionMap Exts = (*MaybeISAInfo)->getExtensions();
+    const auto &Exts = (*MaybeISAInfo)->getExtensions();
     EXPECT_EQ(Exts.size(), 8UL);
     EXPECT_EQ(Exts.count("i"), 1U);
     EXPECT_EQ(Exts.count("m"), 1U);
@@ -373,7 +377,7 @@ TEST(ParseArchString, AcceptsAmbiguousFromRelaxExtensions) {
   for (StringRef Input : {"rv32i_zba_m", "rv32izba_m", "rv32izba1p0_m2p0"}) {
     auto MaybeISAInfo = RISCVISAInfo::parseArchString(Input, true);
     ASSERT_THAT_EXPECTED(MaybeISAInfo, Succeeded());
-    RISCVISAInfo::OrderedExtensionMap Exts = (*MaybeISAInfo)->getExtensions();
+    const auto &Exts = (*MaybeISAInfo)->getExtensions();
     EXPECT_EQ(Exts.size(), 3UL);
     EXPECT_EQ(Exts.count("i"), 1U);
     EXPECT_EQ(Exts.count("zba"), 1U);
@@ -383,7 +387,7 @@ TEST(ParseArchString, AcceptsAmbiguousFromRelaxExtensions) {
        {"rv32ia_zba_m", "rv32iazba_m", "rv32ia2p1zba1p0_m2p0"}) {
     auto MaybeISAInfo = RISCVISAInfo::parseArchString(Input, true);
     ASSERT_THAT_EXPECTED(MaybeISAInfo, Succeeded());
-    RISCVISAInfo::OrderedExtensionMap Exts = (*MaybeISAInfo)->getExtensions();
+    const auto &Exts = (*MaybeISAInfo)->getExtensions();
     EXPECT_EQ(Exts.size(), 4UL);
     EXPECT_EQ(Exts.count("i"), 1U);
     EXPECT_EQ(Exts.count("zba"), 1U);
@@ -395,7 +399,7 @@ TEST(ParseArchString, AcceptsAmbiguousFromRelaxExtensions) {
 TEST(ParseArchString, RejectsRelaxExtensionsNotStartWithEorIorG) {
   EXPECT_EQ(
       toString(RISCVISAInfo::parseArchString("rv32zba_im", true).takeError()),
-      "first letter should be 'e', 'i' or 'g'");
+      "first letter after 'rv32' should be 'e', 'i' or 'g'");
 }
 
 TEST(ParseArchString,
@@ -457,12 +461,12 @@ TEST(ParseArchString,
   // hopefully serve as a reminder to update.
   auto MaybeISAInfo = RISCVISAInfo::parseArchString("rv64iztso", true, false);
   ASSERT_THAT_EXPECTED(MaybeISAInfo, Succeeded());
-  RISCVISAInfo::OrderedExtensionMap Exts = (*MaybeISAInfo)->getExtensions();
+  const auto &Exts = (*MaybeISAInfo)->getExtensions();
   EXPECT_EQ(Exts.size(), 2UL);
   EXPECT_EQ(Exts.count("ztso"), 1U);
   auto MaybeISAInfo2 = RISCVISAInfo::parseArchString("rv64iztso0p1", true);
   ASSERT_THAT_EXPECTED(MaybeISAInfo2, Succeeded());
-  RISCVISAInfo::OrderedExtensionMap Exts2 = (*MaybeISAInfo2)->getExtensions();
+  const auto &Exts2 = (*MaybeISAInfo2)->getExtensions();
   EXPECT_EQ(Exts2.size(), 2UL);
   EXPECT_EQ(Exts2.count("ztso"), 1U);
 }
@@ -479,9 +483,9 @@ TEST(ParseArchString,
   auto MaybeISAInfo =
       RISCVISAInfo::parseArchString("rv64iztso9p9", true, false);
   ASSERT_THAT_EXPECTED(MaybeISAInfo, Succeeded());
-  RISCVISAInfo::OrderedExtensionMap Exts = (*MaybeISAInfo)->getExtensions();
+  const auto &Exts = (*MaybeISAInfo)->getExtensions();
   EXPECT_EQ(Exts.size(), 2UL);
-  EXPECT_TRUE(Exts.at("ztso") == (RISCVISAInfo::ExtensionVersion{9, 9}));
+  EXPECT_TRUE(Exts.at("ztso") == (RISCVISAUtils::ExtensionVersion{9, 9}));
 }
 
 TEST(ParseArchString, RejectsUnrecognizedVersionForExperimentalExtension) {
@@ -502,8 +506,7 @@ TEST(ParseArchString, AddsImpliedExtensions) {
   // Does not attempt to exhaustively test all implications.
   auto MaybeRV64ID = RISCVISAInfo::parseArchString("rv64id", true);
   ASSERT_THAT_EXPECTED(MaybeRV64ID, Succeeded());
-  RISCVISAInfo::OrderedExtensionMap ExtsRV64ID =
-      (*MaybeRV64ID)->getExtensions();
+  const auto &ExtsRV64ID = (*MaybeRV64ID)->getExtensions();
   EXPECT_EQ(ExtsRV64ID.size(), 4UL);
   EXPECT_EQ(ExtsRV64ID.count("i"), 1U);
   EXPECT_EQ(ExtsRV64ID.count("f"), 1U);
@@ -512,8 +515,7 @@ TEST(ParseArchString, AddsImpliedExtensions) {
 
   auto MaybeRV32IZKN = RISCVISAInfo::parseArchString("rv64izkn", true);
   ASSERT_THAT_EXPECTED(MaybeRV32IZKN, Succeeded());
-  RISCVISAInfo::OrderedExtensionMap ExtsRV32IZKN =
-      (*MaybeRV32IZKN)->getExtensions();
+  const auto &ExtsRV32IZKN = (*MaybeRV32IZKN)->getExtensions();
   EXPECT_EQ(ExtsRV32IZKN.size(), 8UL);
   EXPECT_EQ(ExtsRV32IZKN.count("i"), 1U);
   EXPECT_EQ(ExtsRV32IZKN.count("zbkb"), 1U);
@@ -603,7 +605,7 @@ TEST(ToFeatures, AddAllExtensionsAddsNegativeExtensions) {
 }
 
 TEST(OrderedExtensionMap, ExtensionsAreCorrectlyOrdered) {
-  RISCVISAInfo::OrderedExtensionMap Exts;
+  RISCVISAUtils::OrderedExtensionMap Exts;
   for (auto ExtName : {"y", "l", "m", "c", "i", "xfoo", "xbar", "sfoo", "sbar",
                        "zmfoo", "zzfoo", "zfinx", "zicsr"})
     Exts[ExtName] = {1, 0};
@@ -621,8 +623,7 @@ TEST(OrderedExtensionMap, ExtensionsAreCorrectlyOrdered) {
 TEST(ParseArchString, ZceImplication) {
   auto MaybeRV32IZce = RISCVISAInfo::parseArchString("rv32izce", true);
   ASSERT_THAT_EXPECTED(MaybeRV32IZce, Succeeded());
-  RISCVISAInfo::OrderedExtensionMap ExtsRV32IZce =
-      (*MaybeRV32IZce)->getExtensions();
+  const auto &ExtsRV32IZce = (*MaybeRV32IZce)->getExtensions();
   EXPECT_EQ(ExtsRV32IZce.size(), 7UL);
   EXPECT_EQ(ExtsRV32IZce.count("i"), 1U);
   EXPECT_EQ(ExtsRV32IZce.count("zicsr"), 1U);
@@ -634,8 +635,7 @@ TEST(ParseArchString, ZceImplication) {
 
   auto MaybeRV32IFZce = RISCVISAInfo::parseArchString("rv32ifzce", true);
   ASSERT_THAT_EXPECTED(MaybeRV32IFZce, Succeeded());
-  RISCVISAInfo::OrderedExtensionMap ExtsRV32IFZce =
-      (*MaybeRV32IFZce)->getExtensions();
+  const auto &ExtsRV32IFZce = (*MaybeRV32IFZce)->getExtensions();
   EXPECT_EQ(ExtsRV32IFZce.size(), 9UL);
   EXPECT_EQ(ExtsRV32IFZce.count("i"), 1U);
   EXPECT_EQ(ExtsRV32IFZce.count("zicsr"), 1U);
@@ -649,8 +649,7 @@ TEST(ParseArchString, ZceImplication) {
 
   auto MaybeRV32IDZce = RISCVISAInfo::parseArchString("rv32idzce", true);
   ASSERT_THAT_EXPECTED(MaybeRV32IDZce, Succeeded());
-  RISCVISAInfo::OrderedExtensionMap ExtsRV32IDZce =
-      (*MaybeRV32IDZce)->getExtensions();
+  const auto &ExtsRV32IDZce = (*MaybeRV32IDZce)->getExtensions();
   EXPECT_EQ(ExtsRV32IDZce.size(), 10UL);
   EXPECT_EQ(ExtsRV32IDZce.count("i"), 1U);
   EXPECT_EQ(ExtsRV32IDZce.count("zicsr"), 1U);
@@ -665,8 +664,7 @@ TEST(ParseArchString, ZceImplication) {
 
   auto MaybeRV64IZce = RISCVISAInfo::parseArchString("rv64izce", true);
   ASSERT_THAT_EXPECTED(MaybeRV64IZce, Succeeded());
-  RISCVISAInfo::OrderedExtensionMap ExtsRV64IZce =
-      (*MaybeRV64IZce)->getExtensions();
+  const auto &ExtsRV64IZce = (*MaybeRV64IZce)->getExtensions();
   EXPECT_EQ(ExtsRV64IZce.size(), 7UL);
   EXPECT_EQ(ExtsRV64IZce.count("i"), 1U);
   EXPECT_EQ(ExtsRV64IZce.count("zicsr"), 1U);
@@ -678,8 +676,7 @@ TEST(ParseArchString, ZceImplication) {
 
   auto MaybeRV64IFZce = RISCVISAInfo::parseArchString("rv64ifzce", true);
   ASSERT_THAT_EXPECTED(MaybeRV64IFZce, Succeeded());
-  RISCVISAInfo::OrderedExtensionMap ExtsRV64IFZce =
-      (*MaybeRV64IFZce)->getExtensions();
+  const auto &ExtsRV64IFZce = (*MaybeRV64IFZce)->getExtensions();
   EXPECT_EQ(ExtsRV64IFZce.size(), 8UL);
   EXPECT_EQ(ExtsRV64IFZce.count("i"), 1U);
   EXPECT_EQ(ExtsRV64IFZce.count("zicsr"), 1U);
@@ -698,8 +695,7 @@ TEST(ParseArchString, ZceImplication) {
 
   auto MaybeRV64IDZce = RISCVISAInfo::parseArchString("rv64idzce", true);
   ASSERT_THAT_EXPECTED(MaybeRV64IDZce, Succeeded());
-  RISCVISAInfo::OrderedExtensionMap ExtsRV64IDZce =
-      (*MaybeRV64IDZce)->getExtensions();
+  const auto &ExtsRV64IDZce = (*MaybeRV64IDZce)->getExtensions();
   EXPECT_EQ(ExtsRV64IDZce.size(), 9UL);
   EXPECT_EQ(ExtsRV64IDZce.count("i"), 1U);
   EXPECT_EQ(ExtsRV64IDZce.count("zicsr"), 1U);
diff --git a/llvm/unittests/TargetParser/TargetParserTest.cpp b/llvm/unittests/TargetParser/TargetParserTest.cpp
index 2c72a7229b52..75e235008b4f 100644
--- a/llvm/unittests/TargetParser/TargetParserTest.cpp
+++ b/llvm/unittests/TargetParser/TargetParserTest.cpp
@@ -1346,6 +1346,44 @@ INSTANTIATE_TEST_SUITE_P(
                  AArch64::AEK_PAUTH}),
             "9-A"),
         ARMCPUTestParams<AArch64::ExtensionBitset>(
+            "neoverse-v3", "armv9.2-a", "neon-fp-armv8",
+            AArch64::ExtensionBitset(
+                {AArch64::AEK_BF16,        AArch64::AEK_I8MM,
+                 AArch64::AEK_SVE,         AArch64::AEK_SVE2,
+                 AArch64::AEK_FP16,        AArch64::AEK_DOTPROD,
+                 AArch64::AEK_LSE,         AArch64::AEK_RDM,
+                 AArch64::AEK_SIMD,        AArch64::AEK_RCPC,
+                 AArch64::AEK_RAS,         AArch64::AEK_CRC,
+                 AArch64::AEK_FP,          AArch64::AEK_PROFILE,
+                 AArch64::AEK_MTE,         AArch64::AEK_SSBS,
+                 AArch64::AEK_SB,          AArch64::AEK_PREDRES,
+                 AArch64::AEK_LS64,        AArch64::AEK_BRBE,
+                 AArch64::AEK_PAUTH,       AArch64::AEK_FLAGM,
+                 AArch64::AEK_PERFMON,     AArch64::AEK_RAND,
+                 AArch64::AEK_SVE2BITPERM, AArch64::AEK_FP16FML,
+                 AArch64::AEK_PROFILE,     AArch64::AEK_JSCVT,
+                 AArch64::AEK_FCMA}),
+            "9.2-A"),
+        ARMCPUTestParams<AArch64::ExtensionBitset>(
+            "neoverse-v3ae", "armv9.2-a", "neon-fp-armv8",
+            AArch64::ExtensionBitset(
+                {AArch64::AEK_BF16,        AArch64::AEK_I8MM,
+                 AArch64::AEK_SVE,         AArch64::AEK_SVE2,
+                 AArch64::AEK_FP16,        AArch64::AEK_DOTPROD,
+                 AArch64::AEK_LSE,         AArch64::AEK_RDM,
+                 AArch64::AEK_SIMD,        AArch64::AEK_RCPC,
+                 AArch64::AEK_RAS,         AArch64::AEK_CRC,
+                 AArch64::AEK_FP,          AArch64::AEK_PROFILE,
+                 AArch64::AEK_MTE,         AArch64::AEK_SSBS,
+                 AArch64::AEK_SB,          AArch64::AEK_PREDRES,
+                 AArch64::AEK_LS64,        AArch64::AEK_BRBE,
+                 AArch64::AEK_PAUTH,       AArch64::AEK_FLAGM,
+                 AArch64::AEK_PERFMON,     AArch64::AEK_RAND,
+                 AArch64::AEK_SVE2BITPERM, AArch64::AEK_FP16FML,
+                 AArch64::AEK_PROFILE,     AArch64::AEK_JSCVT,
+                 AArch64::AEK_FCMA}),
+            "9.2-A"),
+        ARMCPUTestParams<AArch64::ExtensionBitset>(
             "cortex-r82", "armv8-r", "crypto-neon-fp-armv8",
             AArch64::ExtensionBitset(
                 {AArch64::AEK_CRC, AArch64::AEK_RDM, AArch64::AEK_SSBS,
@@ -1637,6 +1675,24 @@ INSTANTIATE_TEST_SUITE_P(
                  AArch64::AEK_FP16FML}),
             "9-A"),
         ARMCPUTestParams<AArch64::ExtensionBitset>(
+            "neoverse-n3", "armv9.2-a", "neon-fp-armv8",
+            AArch64::ExtensionBitset(
+                {AArch64::AEK_BF16,    AArch64::AEK_I8MM,
+                 AArch64::AEK_SVE,     AArch64::AEK_SVE2,
+                 AArch64::AEK_FP16,    AArch64::AEK_DOTPROD,
+                 AArch64::AEK_LSE,     AArch64::AEK_RDM,
+                 AArch64::AEK_SIMD,    AArch64::AEK_RCPC,
+                 AArch64::AEK_RAS,     AArch64::AEK_CRC,
+                 AArch64::AEK_FP,      AArch64::AEK_PROFILE,
+                 AArch64::AEK_MTE,     AArch64::AEK_SSBS,
+                 AArch64::AEK_SB,      AArch64::AEK_PREDRES,
+                 AArch64::AEK_FCMA,    AArch64::AEK_PAUTH,
+                 AArch64::AEK_FLAGM,   AArch64::AEK_PERFMON,
+                 AArch64::AEK_RAND,    AArch64::AEK_SVE2BITPERM,
+                 AArch64::AEK_FP16FML, AArch64::AEK_PROFILE,
+                 AArch64::AEK_JSCVT}),
+            "9.2-A"),
+        ARMCPUTestParams<AArch64::ExtensionBitset>(
             "ampere1", "armv8.6-a", "crypto-neon-fp-armv8",
             AArch64::ExtensionBitset(
                 {AArch64::AEK_CRC,  AArch64::AEK_FP,   AArch64::AEK_FP16,
@@ -1750,7 +1806,7 @@ INSTANTIATE_TEST_SUITE_P(
     ARMCPUTestParams<AArch64::ExtensionBitset>::PrintToStringParamName);
 
 // Note: number of CPUs includes aliases.
-static constexpr unsigned NumAArch64CPUArchs = 72;
+static constexpr unsigned NumAArch64CPUArchs = 75;
 
 TEST(TargetParserTest, testAArch64CPUArchList) {
   SmallVector<StringRef, NumAArch64CPUArchs> List;
diff --git a/llvm/utils/LLVMVisualizers/llvm.natvis b/llvm/utils/LLVMVisualizers/llvm.natvis
index 0fc50f79466a..d83ae8013c51 100644
--- a/llvm/utils/LLVMVisualizers/llvm.natvis
+++ b/llvm/utils/LLVMVisualizers/llvm.natvis
@@ -92,11 +92,11 @@ For later versions of Visual Studio, no setup is required.
 

   <Type Name="llvm::PointerIntPair&lt;*&gt;">

     <DisplayString IncludeView="ptr">{($T1)(*(intptr_t *)Value.Data &amp; $T5::PointerBitMask)}</DisplayString>

-    <DisplayString IncludeView="int">{($T3)((*(intptr_t *)Value.Data &gt;&gt; $T5::IntShift) &amp; $T5::IntMask)}</DisplayString>

-    <DisplayString>{$T5::IntMask}: {($T1)(*(intptr_t *)Value.Data &amp; $T5::PointerBitMask)} [{($T3)((*(intptr_t *)Value.Data &gt;&gt; $T5::IntShift) &amp; $T5::IntMask)}]</DisplayString>

+    <DisplayString IncludeView="int">{((*(intptr_t *)Value.Data &gt;&gt; $T5::IntShift) &amp; $T5::IntMask)}</DisplayString>

+    <DisplayString>{$T5::IntMask}: {($T1)(*(intptr_t *)Value.Data &amp; $T5::PointerBitMask)} [{((*(intptr_t *)Value.Data &gt;&gt; $T5::IntShift) &amp; $T5::IntMask)}]</DisplayString>

     <Expand>

       <Item Name="[ptr]">($T1)(*(intptr_t *)Value.Data &amp; $T5::PointerBitMask)</Item>

-      <Item Name="[int]">($T3)((*(intptr_t *)Value.Data &gt;&gt; $T5::IntShift) &amp; $T5::IntMask)</Item>

+      <Item Name="[int]">((*(intptr_t *)Value.Data &gt;&gt; $T5::IntShift) &amp; $T5::IntMask)</Item>

     </Expand>

   </Type>

   <!-- PointerUnion types -->

diff --git a/llvm/utils/TableGen/ARMTargetDefEmitter.cpp b/llvm/utils/TableGen/ARMTargetDefEmitter.cpp
new file mode 100644
index 000000000000..db87ac3336c1
--- /dev/null
+++ b/llvm/utils/TableGen/ARMTargetDefEmitter.cpp
@@ -0,0 +1,63 @@
+//===- ARMTargetDefEmitter.cpp - Generate data about ARM Architectures ----===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This tablegen backend exports information about CPUs, FPUs, architectures,
+// and features into a common format that can be used by both TargetParser and
+// the ARM and AArch64 backends.
+//
+//===----------------------------------------------------------------------===//
+
+#include "llvm/ADT/StringSet.h"
+#include "llvm/TableGen/Record.h"
+#include "llvm/TableGen/TableGenBackend.h"
+
+using namespace llvm;
+
+static void EmitARMTargetDef(RecordKeeper &RK, raw_ostream &OS) {
+  OS << "// Autogenerated by ARMTargetDefEmitter.cpp\n\n";
+
+  // Look through all SubtargetFeature defs with the given FieldName, and
+  // collect the set of all Values that that FieldName is set to.
+  auto gatherSubtargetFeatureFieldValues = [&RK](StringRef FieldName) {
+    llvm::StringSet<> Set;
+    for (const Record *Rec : RK.getAllDerivedDefinitions("SubtargetFeature")) {
+      if (Rec->getValueAsString("FieldName") == FieldName) {
+        Set.insert(Rec->getValueAsString("Value"));
+      }
+    }
+    return Set;
+  };
+
+  // The ARMProcFamilyEnum values are initialised by SubtargetFeature defs
+  // which set the ARMProcFamily field. We can generate the enum from these defs
+  // which look like this:
+  //
+  // def ProcA5      : SubtargetFeature<"a5", "ARMProcFamily", "CortexA5",
+  //                                    "Cortex-A5 ARM processors", []>;
+  OS << "#ifndef ARM_PROCESSOR_FAMILY\n"
+     << "#define ARM_PROCESSOR_FAMILY(ENUM)\n"
+     << "#endif\n\n";
+  const StringSet<> ARMProcFamilyVals =
+      gatherSubtargetFeatureFieldValues("ARMProcFamily");
+  for (const StringRef &Family : ARMProcFamilyVals.keys())
+    OS << "ARM_PROCESSOR_FAMILY(" << Family << ")\n";
+  OS << "\n#undef ARM_PROCESSOR_FAMILY\n\n";
+
+  OS << "#ifndef ARM_ARCHITECTURE\n"
+     << "#define ARM_ARCHITECTURE(ENUM)\n"
+     << "#endif\n\n";
+  // This should correspond to instances of the Architecture tablegen class.
+  const StringSet<> ARMArchVals = gatherSubtargetFeatureFieldValues("ARMArch");
+  for (const StringRef &Arch : ARMArchVals.keys())
+    OS << "ARM_ARCHITECTURE(" << Arch << ")\n";
+  OS << "\n#undef ARM_ARCHITECTURE\n\n";
+}
+
+static TableGen::Emitter::Opt
+    X("gen-arm-target-def", EmitARMTargetDef,
+      "Generate the ARM or AArch64 Architecture information header.");
diff --git a/llvm/utils/TableGen/CMakeLists.txt b/llvm/utils/TableGen/CMakeLists.txt
index 577aeded4be7..202f33fdf8b4 100644
--- a/llvm/utils/TableGen/CMakeLists.txt
+++ b/llvm/utils/TableGen/CMakeLists.txt
@@ -13,6 +13,7 @@ set(LLVM_LINK_COMPONENTS Support)
 # ValueType definitions.
 add_tablegen(llvm-min-tblgen LLVM_HEADERS
   TableGen.cpp
+  ARMTargetDefEmitter.cpp
   Attributes.cpp
   DirectiveEmitter.cpp
   IntrinsicEmitter.cpp
@@ -32,6 +33,7 @@ set(LLVM_LINK_COMPONENTS
 add_tablegen(llvm-tblgen LLVM
   DESTINATION "${LLVM_TOOLS_INSTALL_DIR}"
   EXPORT LLVM
+  ARMTargetDefEmitter.cpp
   AsmMatcherEmitter.cpp
   AsmWriterEmitter.cpp
   Attributes.cpp
diff --git a/llvm/utils/TableGen/Common/CodeGenDAGPatterns.cpp b/llvm/utils/TableGen/Common/CodeGenDAGPatterns.cpp
index 7a5d2be3ae95..88d353e89a46 100644
--- a/llvm/utils/TableGen/Common/CodeGenDAGPatterns.cpp
+++ b/llvm/utils/TableGen/Common/CodeGenDAGPatterns.cpp
@@ -4246,7 +4246,7 @@ static TreePatternNodePtr PromoteXForms(TreePatternNodePtr N) {
 
 void CodeGenDAGPatterns::ParseOnePattern(
     Record *TheDef, TreePattern &Pattern, TreePattern &Result,
-    const std::vector<Record *> &InstImpResults) {
+    const std::vector<Record *> &InstImpResults, bool ShouldIgnore) {
 
   // Inline pattern fragments and expand multiple alternatives.
   Pattern.InlinePatternFragments();
@@ -4332,7 +4332,7 @@ void CodeGenDAGPatterns::ParseOnePattern(
         AddPatternToMatch(&Pattern,
                           PatternToMatch(TheDef, Preds, T, Temp.getOnlyTree(),
                                          InstImpResults, Complexity,
-                                         TheDef->getID()));
+                                         TheDef->getID(), ShouldIgnore));
     }
   } else {
     // Show a message about a dropped pattern with some info to make it
@@ -4378,7 +4378,8 @@ void CodeGenDAGPatterns::ParsePatterns() {
       FindPatternInputsAndOutputs(Pattern, Pattern.getTree(j), InstInputs,
                                   InstResults, InstImpResults);
 
-    ParseOnePattern(CurPattern, Pattern, Result, InstImpResults);
+    ParseOnePattern(CurPattern, Pattern, Result, InstImpResults,
+                    CurPattern->getValueAsBit("GISelShouldIgnore"));
   }
 }
 
@@ -4407,10 +4408,10 @@ void CodeGenDAGPatterns::ExpandHwModeBasedTypes() {
       return;
     }
 
-    PatternsToMatch.emplace_back(P.getSrcRecord(), P.getPredicates(),
-                                 std::move(NewSrc), std::move(NewDst),
-                                 P.getDstRegs(), P.getAddedComplexity(),
-                                 Record::getNewUID(Records), Check);
+    PatternsToMatch.emplace_back(
+        P.getSrcRecord(), P.getPredicates(), std::move(NewSrc),
+        std::move(NewDst), P.getDstRegs(), P.getAddedComplexity(),
+        Record::getNewUID(Records), P.getGISelShouldIgnore(), Check);
   };
 
   for (PatternToMatch &P : Copy) {
@@ -4781,6 +4782,7 @@ void CodeGenDAGPatterns::GenerateVariants() {
           Variant, PatternsToMatch[i].getDstPatternShared(),
           PatternsToMatch[i].getDstRegs(),
           PatternsToMatch[i].getAddedComplexity(), Record::getNewUID(Records),
+          PatternsToMatch[i].getGISelShouldIgnore(),
           PatternsToMatch[i].getHwModeFeatures());
     }
 
diff --git a/llvm/utils/TableGen/Common/CodeGenDAGPatterns.h b/llvm/utils/TableGen/Common/CodeGenDAGPatterns.h
index 7fcd39a9e940..7f94db0b7d5d 100644
--- a/llvm/utils/TableGen/Common/CodeGenDAGPatterns.h
+++ b/llvm/utils/TableGen/Common/CodeGenDAGPatterns.h
@@ -1057,17 +1057,19 @@ class PatternToMatch {
   TreePatternNodePtr DstPattern; // Resulting pattern.
   std::vector<Record *> Dstregs; // Physical register defs being matched.
   std::string HwModeFeatures;
-  int AddedComplexity; // Add to matching pattern complexity.
-  unsigned ID;         // Unique ID for the record.
+  int AddedComplexity;    // Add to matching pattern complexity.
+  bool GISelShouldIgnore; // Should GlobalISel ignore importing this pattern.
+  unsigned ID;            // Unique ID for the record.
 
 public:
   PatternToMatch(Record *srcrecord, ListInit *preds, TreePatternNodePtr src,
                  TreePatternNodePtr dst, std::vector<Record *> dstregs,
-                 int complexity, unsigned uid, const Twine &hwmodefeatures = "")
+                 int complexity, unsigned uid, bool ignore,
+                 const Twine &hwmodefeatures = "")
       : SrcRecord(srcrecord), Predicates(preds), SrcPattern(src),
         DstPattern(dst), Dstregs(std::move(dstregs)),
         HwModeFeatures(hwmodefeatures.str()), AddedComplexity(complexity),
-        ID(uid) {}
+        GISelShouldIgnore(ignore), ID(uid) {}
 
   Record *getSrcRecord() const { return SrcRecord; }
   ListInit *getPredicates() const { return Predicates; }
@@ -1078,6 +1080,7 @@ public:
   const std::vector<Record *> &getDstRegs() const { return Dstregs; }
   StringRef getHwModeFeatures() const { return HwModeFeatures; }
   int getAddedComplexity() const { return AddedComplexity; }
+  bool getGISelShouldIgnore() const { return GISelShouldIgnore; }
   unsigned getID() const { return ID; }
 
   std::string getPredicateCheck() const;
@@ -1240,7 +1243,8 @@ private:
 
   void ParseOnePattern(Record *TheDef, TreePattern &Pattern,
                        TreePattern &Result,
-                       const std::vector<Record *> &InstImpResults);
+                       const std::vector<Record *> &InstImpResults,
+                       bool ShouldIgnore = false);
   void AddPatternToMatch(TreePattern *Pattern, PatternToMatch &&PTM);
   void FindPatternInputsAndOutputs(
       TreePattern &I, TreePatternNodePtr Pat,
diff --git a/llvm/utils/TableGen/Common/GlobalISel/GlobalISelMatchTable.cpp b/llvm/utils/TableGen/Common/GlobalISel/GlobalISelMatchTable.cpp
index 19d42b7688da..8af219f34e18 100644
--- a/llvm/utils/TableGen/Common/GlobalISel/GlobalISelMatchTable.cpp
+++ b/llvm/utils/TableGen/Common/GlobalISel/GlobalISelMatchTable.cpp
@@ -723,6 +723,29 @@ void RuleMatcher::optimize() {
     return std::tuple(L->getKind(), L->getInsnVarID(), L->getOpIdx()) <
            std::tuple(R->getKind(), R->getInsnVarID(), R->getOpIdx());
   });
+
+  // Deduplicate EraseInst actions, and if an EraseInst erases the root, place
+  // it at the end to favor generation of GIR_EraseRootFromParent_Done
+  DenseSet<unsigned> AlreadySeenEraseInsts;
+  auto EraseRootIt = Actions.end();
+  auto It = Actions.begin();
+  while (It != Actions.end()) {
+    if (const auto *EI = dyn_cast<EraseInstAction>(It->get())) {
+      unsigned InstID = EI->getInsnID();
+      if (!AlreadySeenEraseInsts.insert(InstID).second) {
+        It = Actions.erase(It);
+        continue;
+      }
+
+      if (InstID == 0)
+        EraseRootIt = It;
+    }
+
+    ++It;
+  }
+
+  if (EraseRootIt != Actions.end())
+    Actions.splice(Actions.end(), Actions, EraseRootIt);
 }
 
 bool RuleMatcher::hasFirstCondition() const {
@@ -966,66 +989,60 @@ void RuleMatcher::emit(MatchTable &Table) {
 
   // We must also check if it's safe to fold the matched instructions.
   if (InsnVariableIDs.size() >= 2) {
-    // Invert the map to create stable ordering (by var names)
-    SmallVector<unsigned, 2> InsnIDs;
-    for (const auto &Pair : InsnVariableIDs) {
-      // Skip the root node since it isn't moving anywhere. Everything else is
-      // sinking to meet it.
-      if (Pair.first == Matchers.front().get())
-        continue;
-
-      InsnIDs.push_back(Pair.second);
-    }
-    llvm::sort(InsnIDs);
-
-    for (const auto &InsnID : InsnIDs) {
-      // Reject the difficult cases until we have a more accurate check.
-      Table << MatchTable::Opcode("GIM_CheckIsSafeToFold")
-            << MatchTable::Comment("InsnID") << MatchTable::ULEB128Value(InsnID)
-            << MatchTable::LineBreak;
 
-      // FIXME: Emit checks to determine it's _actually_ safe to fold and/or
-      //        account for unsafe cases.
-      //
-      //        Example:
-      //          MI1--> %0 = ...
-      //                 %1 = ... %0
-      //          MI0--> %2 = ... %0
-      //          It's not safe to erase MI1. We currently handle this by not
-      //          erasing %0 (even when it's dead).
-      //
-      //        Example:
-      //          MI1--> %0 = load volatile @a
-      //                 %1 = load volatile @a
-      //          MI0--> %2 = ... %0
-      //          It's not safe to sink %0's def past %1. We currently handle
-      //          this by rejecting all loads.
-      //
-      //        Example:
-      //          MI1--> %0 = load @a
-      //                 %1 = store @a
-      //          MI0--> %2 = ... %0
-      //          It's not safe to sink %0's def past %1. We currently handle
-      //          this by rejecting all loads.
-      //
-      //        Example:
-      //                   G_CONDBR %cond, @BB1
-      //                 BB0:
-      //          MI1-->   %0 = load @a
-      //                   G_BR @BB1
-      //                 BB1:
-      //          MI0-->   %2 = ... %0
-      //          It's not always safe to sink %0 across control flow. In this
-      //          case it may introduce a memory fault. We currentl handle
-      //          this by rejecting all loads.
-    }
+    // FIXME: Emit checks to determine it's _actually_ safe to fold and/or
+    //        account for unsafe cases.
+    //
+    //        Example:
+    //          MI1--> %0 = ...
+    //                 %1 = ... %0
+    //          MI0--> %2 = ... %0
+    //          It's not safe to erase MI1. We currently handle this by not
+    //          erasing %0 (even when it's dead).
+    //
+    //        Example:
+    //          MI1--> %0 = load volatile @a
+    //                 %1 = load volatile @a
+    //          MI0--> %2 = ... %0
+    //          It's not safe to sink %0's def past %1. We currently handle
+    //          this by rejecting all loads.
+    //
+    //        Example:
+    //          MI1--> %0 = load @a
+    //                 %1 = store @a
+    //          MI0--> %2 = ... %0
+    //          It's not safe to sink %0's def past %1. We currently handle
+    //          this by rejecting all loads.
+    //
+    //        Example:
+    //                   G_CONDBR %cond, @BB1
+    //                 BB0:
+    //          MI1-->   %0 = load @a
+    //                   G_BR @BB1
+    //                 BB1:
+    //          MI0-->   %2 = ... %0
+    //          It's not always safe to sink %0 across control flow. In this
+    //          case it may introduce a memory fault. We currentl handle
+    //          this by rejecting all loads.
+
+    Table << MatchTable::Opcode("GIM_CheckIsSafeToFold")
+          << MatchTable::Comment("NumInsns")
+          << MatchTable::IntValue(1, InsnVariableIDs.size() - 1)
+          << MatchTable::LineBreak;
   }
 
   for (const auto &PM : EpilogueMatchers)
     PM->emitPredicateOpcodes(Table, *this);
 
-  for (const auto &MA : Actions)
-    MA->emitActionOpcodes(Table, *this);
+  // Emit all actions except the last one, then emit coverage and emit the
+  // final action.
+  //
+  // This is because some actions, such as GIR_EraseRootFromParent_Done, also
+  // double as a GIR_Done and terminate execution of the rule.
+  if (!Actions.empty()) {
+    for (const auto &MA : drop_end(Actions))
+      MA->emitActionOpcodes(Table, *this);
+  }
 
   assert((Table.isWithCoverage() ? !Table.isCombiner() : true) &&
          "Combiner tables don't support coverage!");
@@ -1036,8 +1053,13 @@ void RuleMatcher::emit(MatchTable &Table) {
     Table << MatchTable::Comment(("GIR_Coverage, " + Twine(RuleID) + ",").str())
           << MatchTable::LineBreak;
 
-  Table << MatchTable::Opcode("GIR_Done", -1) << MatchTable::LineBreak
-        << MatchTable::Label(LabelID);
+  if (Actions.empty() ||
+      !Actions.back()->emitActionOpcodesAndDone(Table, *this)) {
+    Table << MatchTable::Opcode("GIR_Done", -1) << MatchTable::LineBreak;
+  }
+
+  Table << MatchTable::Label(LabelID);
+
   ++NumPatternEmitted;
 }
 
@@ -1140,10 +1162,14 @@ bool LLTOperandMatcher::hasValue() const {
 
 void LLTOperandMatcher::emitPredicateOpcodes(MatchTable &Table,
                                              RuleMatcher &Rule) const {
-  Table << MatchTable::Opcode("GIM_CheckType") << MatchTable::Comment("MI")
-        << MatchTable::ULEB128Value(InsnVarID) << MatchTable::Comment("Op")
-        << MatchTable::ULEB128Value(OpIdx) << MatchTable::Comment("Type")
-        << getValue() << MatchTable::LineBreak;
+  if (InsnVarID == 0) {
+    Table << MatchTable::Opcode("GIM_RootCheckType");
+  } else {
+    Table << MatchTable::Opcode("GIM_CheckType") << MatchTable::Comment("MI")
+          << MatchTable::ULEB128Value(InsnVarID);
+  }
+  Table << MatchTable::Comment("Op") << MatchTable::ULEB128Value(OpIdx)
+        << MatchTable::Comment("Type") << getValue() << MatchTable::LineBreak;
 }
 
 //===- PointerToAnyOperandMatcher -----------------------------------------===//
@@ -1205,9 +1231,14 @@ bool RegisterBankOperandMatcher::isIdentical(const PredicateMatcher &B) const {
 
 void RegisterBankOperandMatcher::emitPredicateOpcodes(MatchTable &Table,
                                                       RuleMatcher &Rule) const {
-  Table << MatchTable::Opcode("GIM_CheckRegBankForClass")
-        << MatchTable::Comment("MI") << MatchTable::ULEB128Value(InsnVarID)
-        << MatchTable::Comment("Op") << MatchTable::ULEB128Value(OpIdx)
+  if (InsnVarID == 0) {
+    Table << MatchTable::Opcode("GIM_RootCheckRegBankForClass");
+  } else {
+    Table << MatchTable::Opcode("GIM_CheckRegBankForClass")
+          << MatchTable::Comment("MI") << MatchTable::ULEB128Value(InsnVarID);
+  }
+
+  Table << MatchTable::Comment("Op") << MatchTable::ULEB128Value(OpIdx)
         << MatchTable::Comment("RC")
         << MatchTable::NamedValue(2, RC.getQualifiedIdName())
         << MatchTable::LineBreak;
@@ -1810,17 +1841,28 @@ OperandRenderer::~OperandRenderer() {}
 
 //===- CopyRenderer -------------------------------------------------------===//
 
+void CopyRenderer::emitRenderOpcodes(MatchTable &Table, RuleMatcher &Rule,
+                                     unsigned NewInsnID, unsigned OldInsnID,
+                                     unsigned OpIdx, StringRef Name) {
+  if (NewInsnID == 0 && OldInsnID == 0) {
+    Table << MatchTable::Opcode("GIR_RootToRootCopy");
+  } else {
+    Table << MatchTable::Opcode("GIR_Copy") << MatchTable::Comment("NewInsnID")
+          << MatchTable::ULEB128Value(NewInsnID)
+          << MatchTable::Comment("OldInsnID")
+          << MatchTable::ULEB128Value(OldInsnID);
+  }
+
+  Table << MatchTable::Comment("OpIdx") << MatchTable::ULEB128Value(OpIdx)
+        << MatchTable::Comment(Name) << MatchTable::LineBreak;
+}
+
 void CopyRenderer::emitRenderOpcodes(MatchTable &Table,
                                      RuleMatcher &Rule) const {
   const OperandMatcher &Operand = Rule.getOperandMatcher(SymbolicName);
   unsigned OldInsnVarID = Rule.getInsnVarID(Operand.getInstructionMatcher());
-  Table << MatchTable::Opcode("GIR_Copy") << MatchTable::Comment("NewInsnID")
-        << MatchTable::ULEB128Value(NewInsnID)
-        << MatchTable::Comment("OldInsnID")
-        << MatchTable::ULEB128Value(OldInsnVarID)
-        << MatchTable::Comment("OpIdx")
-        << MatchTable::ULEB128Value(Operand.getOpIdx())
-        << MatchTable::Comment(SymbolicName) << MatchTable::LineBreak;
+  emitRenderOpcodes(Table, Rule, NewInsnID, OldInsnVarID, Operand.getOpIdx(),
+                    SymbolicName);
 }
 
 //===- CopyPhysRegRenderer ------------------------------------------------===//
@@ -1829,13 +1871,8 @@ void CopyPhysRegRenderer::emitRenderOpcodes(MatchTable &Table,
                                             RuleMatcher &Rule) const {
   const OperandMatcher &Operand = Rule.getPhysRegOperandMatcher(PhysReg);
   unsigned OldInsnVarID = Rule.getInsnVarID(Operand.getInstructionMatcher());
-  Table << MatchTable::Opcode("GIR_Copy") << MatchTable::Comment("NewInsnID")
-        << MatchTable::ULEB128Value(NewInsnID)
-        << MatchTable::Comment("OldInsnID")
-        << MatchTable::ULEB128Value(OldInsnVarID)
-        << MatchTable::Comment("OpIdx")
-        << MatchTable::ULEB128Value(Operand.getOpIdx())
-        << MatchTable::Comment(PhysReg->getName()) << MatchTable::LineBreak;
+  CopyRenderer::emitRenderOpcodes(Table, Rule, NewInsnID, OldInsnVarID,
+                                  Operand.getOpIdx(), PhysReg->getName());
 }
 
 //===- CopyOrAddZeroRegRenderer -------------------------------------------===//
@@ -2185,10 +2222,17 @@ void BuildMIAction::emitActionOpcodes(MatchTable &Table,
   // TODO: Simple permutation looks like it could be almost as common as
   //       mutation due to commutative operations.
 
-  Table << MatchTable::Opcode("GIR_BuildMI") << MatchTable::Comment("InsnID")
-        << MatchTable::ULEB128Value(InsnID) << MatchTable::Comment("Opcode")
+  if (InsnID == 0) {
+    Table << MatchTable::Opcode("GIR_BuildRootMI");
+  } else {
+    Table << MatchTable::Opcode("GIR_BuildMI") << MatchTable::Comment("InsnID")
+          << MatchTable::ULEB128Value(InsnID);
+  }
+
+  Table << MatchTable::Comment("Opcode")
         << MatchTable::NamedValue(2, I->Namespace, I->TheDef->getName())
         << MatchTable::LineBreak;
+
   for (const auto &Renderer : OperandRenderers)
     Renderer->emitRenderOpcodes(Table, Rule);
 
@@ -2244,8 +2288,8 @@ void BuildConstantAction::emitActionOpcodes(MatchTable &Table,
 
 //===- EraseInstAction ----------------------------------------------------===//
 
-void EraseInstAction::emitActionOpcodes(MatchTable &Table, RuleMatcher &Rule,
-                                        unsigned InsnID) {
+void EraseInstAction::emitActionOpcodes(MatchTable &Table,
+                                        RuleMatcher &Rule) const {
   // Avoid erasing the same inst twice.
   if (!Rule.tryEraseInsnID(InsnID))
     return;
@@ -2255,9 +2299,19 @@ void EraseInstAction::emitActionOpcodes(MatchTable &Table, RuleMatcher &Rule,
         << MatchTable::LineBreak;
 }
 
-void EraseInstAction::emitActionOpcodes(MatchTable &Table,
-                                        RuleMatcher &Rule) const {
-  emitActionOpcodes(Table, Rule, InsnID);
+bool EraseInstAction::emitActionOpcodesAndDone(MatchTable &Table,
+                                               RuleMatcher &Rule) const {
+  if (InsnID != 0) {
+    emitActionOpcodes(Table, Rule);
+    return false;
+  }
+
+  if (!Rule.tryEraseInsnID(0))
+    return false;
+
+  Table << MatchTable::Opcode("GIR_EraseRootFromParent_Done", -1)
+        << MatchTable::LineBreak;
+  return true;
 }
 
 //===- ReplaceRegAction ---------------------------------------------------===//
diff --git a/llvm/utils/TableGen/Common/GlobalISel/GlobalISelMatchTable.h b/llvm/utils/TableGen/Common/GlobalISel/GlobalISelMatchTable.h
index aa86fad763d1..30301c28ce6c 100644
--- a/llvm/utils/TableGen/Common/GlobalISel/GlobalISelMatchTable.h
+++ b/llvm/utils/TableGen/Common/GlobalISel/GlobalISelMatchTable.h
@@ -1884,6 +1884,10 @@ public:
 
   StringRef getSymbolicName() const { return SymbolicName; }
 
+  static void emitRenderOpcodes(MatchTable &Table, RuleMatcher &Rule,
+                                unsigned NewInsnID, unsigned OldInsnID,
+                                unsigned OpIdx, StringRef Name);
+
   void emitRenderOpcodes(MatchTable &Table, RuleMatcher &Rule) const override;
 };
 
@@ -2226,6 +2230,15 @@ public:
   virtual void emitActionOpcodes(MatchTable &Table,
                                  RuleMatcher &Rule) const = 0;
 
+  /// If this opcode has an overload that can call GIR_Done directly, emit that
+  /// instead of the usual opcode and return "true". Return "false" if GIR_Done
+  /// still needs to be emitted.
+  virtual bool emitActionOpcodesAndDone(MatchTable &Table,
+                                        RuleMatcher &Rule) const {
+    emitActionOpcodes(Table, Rule);
+    return false;
+  }
+
 private:
   ActionKind Kind;
 };
@@ -2334,13 +2347,15 @@ public:
   EraseInstAction(unsigned InsnID)
       : MatchAction(AK_EraseInst), InsnID(InsnID) {}
 
+  unsigned getInsnID() const { return InsnID; }
+
   static bool classof(const MatchAction *A) {
     return A->getKind() == AK_EraseInst;
   }
 
   void emitActionOpcodes(MatchTable &Table, RuleMatcher &Rule) const override;
-  static void emitActionOpcodes(MatchTable &Table, RuleMatcher &Rule,
-                                unsigned InsnID);
+  bool emitActionOpcodesAndDone(MatchTable &Table,
+                                RuleMatcher &Rule) const override;
 };
 
 class ReplaceRegAction : public MatchAction {
@@ -2351,11 +2366,11 @@ class ReplaceRegAction : public MatchAction {
 public:
   ReplaceRegAction(unsigned OldInsnID, unsigned OldOpIdx, unsigned NewInsnId,
                    unsigned NewOpIdx)
-      : MatchAction(AK_EraseInst), OldInsnID(OldInsnID), OldOpIdx(OldOpIdx),
+      : MatchAction(AK_ReplaceReg), OldInsnID(OldInsnID), OldOpIdx(OldOpIdx),
         NewInsnId(NewInsnId), NewOpIdx(NewOpIdx) {}
 
   ReplaceRegAction(unsigned OldInsnID, unsigned OldOpIdx, unsigned TempRegID)
-      : MatchAction(AK_EraseInst), OldInsnID(OldInsnID), OldOpIdx(OldOpIdx),
+      : MatchAction(AK_ReplaceReg), OldInsnID(OldInsnID), OldOpIdx(OldOpIdx),
         TempRegID(TempRegID) {}
 
   static bool classof(const MatchAction *A) {
@@ -2381,9 +2396,14 @@ public:
   }
 
   void emitActionOpcodes(MatchTable &Table, RuleMatcher &Rule) const override {
-    Table << MatchTable::Opcode("GIR_ConstrainSelectedInstOperands")
-          << MatchTable::Comment("InsnID") << MatchTable::ULEB128Value(InsnID)
-          << MatchTable::LineBreak;
+    if (InsnID == 0) {
+      Table << MatchTable::Opcode("GIR_RootConstrainSelectedInstOperands")
+            << MatchTable::LineBreak;
+    } else {
+      Table << MatchTable::Opcode("GIR_ConstrainSelectedInstOperands")
+            << MatchTable::Comment("InsnID") << MatchTable::ULEB128Value(InsnID)
+            << MatchTable::LineBreak;
+    }
   }
 };
 
diff --git a/llvm/utils/TableGen/Common/SubtargetFeatureInfo.cpp b/llvm/utils/TableGen/Common/SubtargetFeatureInfo.cpp
index 819abfa965a5..4f57234d6fe2 100644
--- a/llvm/utils/TableGen/Common/SubtargetFeatureInfo.cpp
+++ b/llvm/utils/TableGen/Common/SubtargetFeatureInfo.cpp
@@ -108,7 +108,7 @@ void SubtargetFeatureInfo::emitComputeAvailableFeatures(
   if (!ExtraParams.empty())
     OS << ", " << ExtraParams;
   OS << ") const {\n";
-  OS << "  PredicateBitset Features;\n";
+  OS << "  PredicateBitset Features{};\n";
   for (const auto &SF : SubtargetFeatures) {
     const SubtargetFeatureInfo &SFI = SF.second;
     StringRef CondStr = SFI.TheDef->getValueAsString("CondString");
diff --git a/llvm/utils/TableGen/DAGISelMatcherEmitter.cpp b/llvm/utils/TableGen/DAGISelMatcherEmitter.cpp
index dcecac4380ce..ff508d648733 100644
--- a/llvm/utils/TableGen/DAGISelMatcherEmitter.cpp
+++ b/llvm/utils/TableGen/DAGISelMatcherEmitter.cpp
@@ -1352,7 +1352,7 @@ void llvm::EmitMatcherTable(Matcher *TheMatcher, const CodeGenDAGPatterns &CGP,
   MatcherEmitter.EmitHistogram(TheMatcher, OS);
 
   OS << "  #undef TARGET_VAL\n";
-  OS << "  SelectCodeCommon(N, MatcherTable,sizeof(MatcherTable));\n";
+  OS << "  SelectCodeCommon(N, MatcherTable, sizeof(MatcherTable));\n";
   OS << "}\n";
   EndEmitFunction(OS);
 
diff --git a/llvm/utils/TableGen/DXILEmitter.cpp b/llvm/utils/TableGen/DXILEmitter.cpp
index f2504775d557..0439df8067ed 100644
--- a/llvm/utils/TableGen/DXILEmitter.cpp
+++ b/llvm/utils/TableGen/DXILEmitter.cpp
@@ -97,7 +97,7 @@ static ParameterKind getParameterKind(const Record *R) {
     if (R->getValueAsInt("isHalfOrFloat") || R->getValueAsInt("isI16OrI32")) {
       return ParameterKind::Overload;
     }
-    LLVM_FALLTHROUGH;
+    [[fallthrough]];
   default:
     llvm_unreachable("Support for specified DXIL Type not yet implemented");
   }
@@ -272,7 +272,7 @@ static std::string getOverloadKindStr(const Record *R) {
         return "OverloadKind::I16 | OverloadKind::I32";
       }
     }
-    LLVM_FALLTHROUGH;
+    [[fallthrough]];
   default:
     llvm_unreachable(
         "Support for specified parameter OverloadKind not yet implemented");
diff --git a/llvm/utils/TableGen/GlobalISelEmitter.cpp b/llvm/utils/TableGen/GlobalISelEmitter.cpp
index 25e302ce1ca4..0eb258ff89a2 100644
--- a/llvm/utils/TableGen/GlobalISelEmitter.cpp
+++ b/llvm/utils/TableGen/GlobalISelEmitter.cpp
@@ -834,6 +834,11 @@ Expected<InstructionMatcher &> GlobalISelEmitter::createAndImportSelDAGMatcher(
       return InsnMatcher;
     }
 
+    if (SrcGIOrNull->TheDef->getName() == "G_FRAME_INDEX") {
+      InsnMatcher.addOperand(OpIdx++, Src.getName(), TempOpIdx);
+      return InsnMatcher;
+    }
+
     // Special case because the operand order is changed from setcc. The
     // predicate operand needs to be swapped from the last operand to the first
     // source.
@@ -1223,6 +1228,10 @@ Expected<action_iterator> GlobalISelEmitter::importExplicitUseRenderer(
     if (DstChild.getOperator()->getName() == "timm") {
       DstMIBuilder.addRenderer<CopyRenderer>(DstChild.getName());
       return InsertPt;
+    }
+    if (DstChild.getOperator()->getName() == "tframeindex") {
+      DstMIBuilder.addRenderer<CopyRenderer>(DstChild.getName());
+      return InsertPt;
     } else if (DstChild.getOperator()->getName() == "imm") {
       DstMIBuilder.addRenderer<CopyConstantAsImmRenderer>(DstChild.getName());
       return InsertPt;
@@ -2411,6 +2420,8 @@ void GlobalISelEmitter::run(raw_ostream &OS) {
   for (const PatternToMatch &Pat : CGP.ptms()) {
     ++NumPatternTotal;
 
+    if (Pat.getGISelShouldIgnore())
+      continue; // skip without warning
     auto MatcherOrErr = runOnPattern(Pat);
 
     // The pattern analysis can fail, indicating an unsupported pattern.
diff --git a/llvm/utils/TableGen/RISCVTargetDefEmitter.cpp b/llvm/utils/TableGen/RISCVTargetDefEmitter.cpp
index 62916bd62c01..097e1deb3ed1 100644
--- a/llvm/utils/TableGen/RISCVTargetDefEmitter.cpp
+++ b/llvm/utils/TableGen/RISCVTargetDefEmitter.cpp
@@ -6,18 +6,82 @@
 //
 //===----------------------------------------------------------------------===//
 //
-// This tablegen backend emits the include file needed by the target
-// parser to parse the RISC-V CPUs.
+// This tablegen backend emits the include file needed by RISCVTargetParser.cpp
+// and RISCVISAInfo.cpp to parse the RISC-V CPUs and extensions.
 //
 //===----------------------------------------------------------------------===//
 
-#include "llvm/Support/RISCVISAInfo.h"
+#include "llvm/Support/RISCVISAUtils.h"
 #include "llvm/TableGen/Record.h"
 #include "llvm/TableGen/TableGenBackend.h"
 
 using namespace llvm;
 
-using ISAInfoTy = llvm::Expected<std::unique_ptr<RISCVISAInfo>>;
+static StringRef getExtensionName(const Record *R) {
+  StringRef Name = R->getValueAsString("Name");
+  Name.consume_front("experimental-");
+  return Name;
+}
+
+static void printExtensionTable(raw_ostream &OS,
+                                const std::vector<Record *> &Extensions,
+                                bool Experimental) {
+  OS << "static const RISCVSupportedExtension Supported";
+  if (Experimental)
+    OS << "Experimental";
+  OS << "Extensions[] = {\n";
+
+  for (Record *R : Extensions) {
+    if (R->getValueAsBit("Experimental") != Experimental)
+      continue;
+
+    OS << "    {\"" << getExtensionName(R) << "\", {"
+       << R->getValueAsInt("MajorVersion") << ", "
+       << R->getValueAsInt("MinorVersion") << "}},\n";
+  }
+
+  OS << "};\n\n";
+}
+
+static void emitRISCVExtensions(RecordKeeper &Records, raw_ostream &OS) {
+  OS << "#ifdef GET_SUPPORTED_EXTENSIONS\n";
+  OS << "#undef GET_SUPPORTED_EXTENSIONS\n\n";
+
+  std::vector<Record *> Extensions =
+      Records.getAllDerivedDefinitions("RISCVExtension");
+  llvm::sort(Extensions, [](const Record *Rec1, const Record *Rec2) {
+    return getExtensionName(Rec1) < getExtensionName(Rec2);
+  });
+
+  printExtensionTable(OS, Extensions, /*Experimental=*/false);
+  printExtensionTable(OS, Extensions, /*Experimental=*/true);
+
+  OS << "#endif // GET_SUPPORTED_EXTENSIONS\n\n";
+
+  OS << "#ifdef GET_IMPLIED_EXTENSIONS\n";
+  OS << "#undef GET_IMPLIED_EXTENSIONS\n\n";
+
+  OS << "\nstatic constexpr ImpliedExtsEntry ImpliedExts[] = {\n";
+  for (Record *Ext : Extensions) {
+    auto ImpliesList = Ext->getValueAsListOfDefs("Implies");
+    if (ImpliesList.empty())
+      continue;
+
+    StringRef Name = getExtensionName(Ext);
+
+    for (auto *ImpliedExt : ImpliesList) {
+      if (!ImpliedExt->isSubClassOf("RISCVExtension"))
+        continue;
+
+      OS << "    { {\"" << Name << "\"}, \"" << getExtensionName(ImpliedExt)
+         << "\"},\n";
+    }
+  }
+
+  OS << "};\n\n";
+
+  OS << "#endif // GET_IMPLIED_EXTENSIONS\n\n";
+}
 
 // We can generate march string from target features as what has been described
 // in RISC-V ISA specification (version 20191213) 'Chapter 27. ISA Extension
@@ -25,57 +89,83 @@ using ISAInfoTy = llvm::Expected<std::unique_ptr<RISCVISAInfo>>;
 //
 // This is almost the same as RISCVFeatures::parseFeatureBits, except that we
 // get feature name from feature records instead of feature bits.
-static std::string getMArch(const Record &Rec) {
-  std::vector<std::string> FeatureVector;
-  unsigned XLen = 32;
+static void printMArch(raw_ostream &OS, const std::vector<Record *> &Features) {
+  RISCVISAUtils::OrderedExtensionMap Extensions;
+  unsigned XLen = 0;
 
   // Convert features to FeatureVector.
-  for (auto *Feature : Rec.getValueAsListOfDefs("Features")) {
-    StringRef FeatureName = Feature->getValueAsString("Name");
-    if (llvm::RISCVISAInfo::isSupportedExtensionFeature(FeatureName))
-      FeatureVector.push_back((Twine("+") + FeatureName).str());
-    else if (FeatureName == "64bit")
+  for (auto *Feature : Features) {
+    StringRef FeatureName = getExtensionName(Feature);
+    if (Feature->isSubClassOf("RISCVExtension")) {
+      unsigned Major = Feature->getValueAsInt("MajorVersion");
+      unsigned Minor = Feature->getValueAsInt("MinorVersion");
+      Extensions[FeatureName.str()] = {Major, Minor};
+    } else if (FeatureName == "64bit") {
+      assert(XLen == 0 && "Already determined XLen");
       XLen = 64;
+    } else if (FeatureName == "32bit") {
+      assert(XLen == 0 && "Already determined XLen");
+      XLen = 32;
+    }
+  }
+
+  assert(XLen != 0 && "Unable to determine XLen");
+
+  OS << "rv" << XLen;
+
+  ListSeparator LS("_");
+  for (auto const &Ext : Extensions)
+    OS << LS << Ext.first << Ext.second.Major << 'p' << Ext.second.Minor;
+}
+
+static void emitRISCVProfiles(RecordKeeper &Records, raw_ostream &OS) {
+  OS << "#ifdef GET_SUPPORTED_PROFILES\n";
+  OS << "#undef GET_SUPPORTED_PROFILES\n\n";
+
+  OS << "static constexpr RISCVProfile SupportedProfiles[] = {\n";
+
+  for (const Record *Rec : Records.getAllDerivedDefinitions("RISCVProfile")) {
+    OS.indent(4) << "{\"" << Rec->getValueAsString("Name") << "\",\"";
+    printMArch(OS, Rec->getValueAsListOfDefs("Implies"));
+    OS << "\"},\n";
   }
 
-  ISAInfoTy ISAInfo = llvm::RISCVISAInfo::parseFeatures(XLen, FeatureVector);
-  if (!ISAInfo)
-    report_fatal_error("Invalid features");
+  OS << "};\n\n";
 
-  // RISCVISAInfo::toString will generate a march string with all the extensions
-  // we have added to it.
-  return (*ISAInfo)->toString();
+  OS << "#endif // GET_SUPPORTED_PROFILES\n\n";
 }
 
-static void EmitRISCVTargetDef(RecordKeeper &RK, raw_ostream &OS) {
+static void emitRISCVProcs(RecordKeeper &RK, raw_ostream &OS) {
   OS << "#ifndef PROC\n"
      << "#define PROC(ENUM, NAME, DEFAULT_MARCH, FAST_UNALIGNED_ACCESS)\n"
      << "#endif\n\n";
 
   // Iterate on all definition records.
   for (const Record *Rec : RK.getAllDerivedDefinitions("RISCVProcessorModel")) {
-    std::string MArch = Rec->getValueAsString("DefaultMarch").str();
+    const std::vector<Record *> &Features =
+        Rec->getValueAsListOfDefs("Features");
+    bool FastScalarUnalignedAccess = any_of(Features, [&](auto &Feature) {
+      return Feature->getValueAsString("Name") == "unaligned-scalar-mem";
+    });
 
-    // Compute MArch from features if we don't specify it.
-    if (MArch.empty())
-      MArch = getMArch(*Rec);
-
-    bool FastScalarUnalignedAccess =
-        any_of(Rec->getValueAsListOfDefs("Features"), [&](auto &Feature) {
-          return Feature->getValueAsString("Name") == "unaligned-scalar-mem";
-        });
-
-    bool FastVectorUnalignedAccess =
-        any_of(Rec->getValueAsListOfDefs("Features"), [&](auto &Feature) {
-          return Feature->getValueAsString("Name") == "unaligned-vector-mem";
-        });
+    bool FastVectorUnalignedAccess = any_of(Features, [&](auto &Feature) {
+      return Feature->getValueAsString("Name") == "unaligned-vector-mem";
+    });
 
     bool FastUnalignedAccess =
         FastScalarUnalignedAccess && FastVectorUnalignedAccess;
 
-    OS << "PROC(" << Rec->getName() << ", "
-       << "{\"" << Rec->getValueAsString("Name") << "\"}, "
-       << "{\"" << MArch << "\"}, " << FastUnalignedAccess << ")\n";
+    OS << "PROC(" << Rec->getName() << ", {\"" << Rec->getValueAsString("Name")
+       << "\"}, {\"";
+
+    StringRef MArch = Rec->getValueAsString("DefaultMarch");
+
+    // Compute MArch from features if we don't specify it.
+    if (MArch.empty())
+      printMArch(OS, Features);
+    else
+      OS << MArch;
+    OS << "\"}, " << FastUnalignedAccess << ")\n";
   }
   OS << "\n#undef PROC\n";
   OS << "\n";
@@ -92,5 +182,12 @@ static void EmitRISCVTargetDef(RecordKeeper &RK, raw_ostream &OS) {
   OS << "\n#undef TUNE_PROC\n";
 }
 
+static void EmitRISCVTargetDef(RecordKeeper &RK, raw_ostream &OS) {
+  emitRISCVExtensions(RK, OS);
+  emitRISCVProfiles(RK, OS);
+  emitRISCVProcs(RK, OS);
+}
+
 static TableGen::Emitter::Opt X("gen-riscv-target-def", EmitRISCVTargetDef,
-                                "Generate the list of CPU for RISCV");
+                                "Generate the list of CPUs and extensions for "
+                                "RISC-V");
diff --git a/llvm/utils/gn/secondary/clang-tools-extra/clang-tidy/bugprone/BUILD.gn b/llvm/utils/gn/secondary/clang-tools-extra/clang-tidy/bugprone/BUILD.gn
index cd90c7752e16..bbf2b84334d9 100644
--- a/llvm/utils/gn/secondary/clang-tools-extra/clang-tidy/bugprone/BUILD.gn
+++ b/llvm/utils/gn/secondary/clang-tools-extra/clang-tidy/bugprone/BUILD.gn
@@ -61,6 +61,7 @@ static_library("bugprone") {
     "PosixReturnCheck.cpp",
     "RedundantBranchConditionCheck.cpp",
     "ReservedIdentifierCheck.cpp",
+    "ReturnConstRefFromParameterCheck.cpp",
     "SharedPtrArrayMismatchCheck.cpp",
     "SignalHandlerCheck.cpp",
     "SignedCharMisuseCheck.cpp",
diff --git a/llvm/utils/gn/secondary/clang-tools-extra/clang-tidy/modernize/BUILD.gn b/llvm/utils/gn/secondary/clang-tools-extra/clang-tidy/modernize/BUILD.gn
index c9e081383fa0..0d27b786da1f 100644
--- a/llvm/utils/gn/secondary/clang-tools-extra/clang-tidy/modernize/BUILD.gn
+++ b/llvm/utils/gn/secondary/clang-tools-extra/clang-tidy/modernize/BUILD.gn
@@ -25,6 +25,7 @@ static_library("modernize") {
     "MakeSharedCheck.cpp",
     "MakeSmartPtrCheck.cpp",
     "MakeUniqueCheck.cpp",
+    "MinMaxUseInitializerListCheck.cpp",
     "ModernizeTidyModule.cpp",
     "PassByValueCheck.cpp",
     "RawStringLiteralCheck.cpp",
diff --git a/llvm/utils/gn/secondary/clang-tools-extra/clang-tidy/readability/BUILD.gn b/llvm/utils/gn/secondary/clang-tools-extra/clang-tidy/readability/BUILD.gn
index 59dc38c8c4d8..815c5a93c72f 100644
--- a/llvm/utils/gn/secondary/clang-tools-extra/clang-tidy/readability/BUILD.gn
+++ b/llvm/utils/gn/secondary/clang-tools-extra/clang-tidy/readability/BUILD.gn
@@ -35,6 +35,7 @@ static_library("readability") {
     "IsolateDeclarationCheck.cpp",
     "MagicNumbersCheck.cpp",
     "MakeMemberFunctionConstCheck.cpp",
+    "MathMissingParenthesesCheck.cpp",
     "MisleadingIndentationCheck.cpp",
     "MisplacedArrayIndexCheck.cpp",
     "NamedParameterCheck.cpp",
diff --git a/llvm/utils/gn/secondary/llvm/include/llvm/TargetParser/BUILD.gn b/llvm/utils/gn/secondary/llvm/include/llvm/TargetParser/BUILD.gn
index aecb65ab6c72..a71dfa518b1d 100644
--- a/llvm/utils/gn/secondary/llvm/include/llvm/TargetParser/BUILD.gn
+++ b/llvm/utils/gn/secondary/llvm/include/llvm/TargetParser/BUILD.gn
@@ -1,5 +1,19 @@
 import("//llvm/utils/TableGen/tablegen.gni")
 
+tablegen("ARMTargetParserDef") {
+  visibility = [ ":gen" ]
+  args = [ "-gen-arm-target-def" ]
+  td_file = "//llvm/lib/Target/ARM/ARM.td"
+  tblgen_target = "//llvm/utils/TableGen:llvm-min-tblgen"
+}
+
+tablegen("AArch64TargetParserDef") {
+  visibility = [ ":gen" ]
+  args = [ "-gen-arm-target-def" ]
+  td_file = "//llvm/lib/Target/AArch64/AArch64.td"
+  tblgen_target = "//llvm/utils/TableGen:llvm-min-tblgen"
+}
+
 tablegen("RISCVTargetParserDef") {
   visibility = [ ":gen" ]
   args = [ "-gen-riscv-target-def" ]
@@ -8,5 +22,9 @@ tablegen("RISCVTargetParserDef") {
 }
 
 group("gen") {
-  deps = [ ":RISCVTargetParserDef" ]
+  deps = [
+    ":ARMTargetParserDef",
+    ":AArch64TargetParserDef",
+    ":RISCVTargetParserDef",
+  ]
 }
diff --git a/llvm/utils/gn/secondary/llvm/lib/IR/BUILD.gn b/llvm/utils/gn/secondary/llvm/lib/IR/BUILD.gn
index 896b031a7bd3..247ef480f5f1 100644
--- a/llvm/utils/gn/secondary/llvm/lib/IR/BUILD.gn
+++ b/llvm/utils/gn/secondary/llvm/lib/IR/BUILD.gn
@@ -31,8 +31,8 @@ static_library("IR") {
     "DataLayout.cpp",
     "DebugInfo.cpp",
     "DebugInfoMetadata.cpp",
-    "DebugProgramInstruction.cpp",
     "DebugLoc.cpp",
+    "DebugProgramInstruction.cpp",
     "DiagnosticHandler.cpp",
     "DiagnosticInfo.cpp",
     "DiagnosticPrinter.cpp",
@@ -55,6 +55,7 @@ static_library("IR") {
     "LegacyPassManager.cpp",
     "MDBuilder.cpp",
     "Mangler.cpp",
+    "MemoryModelRelaxationAnnotations.cpp",
     "Metadata.cpp",
     "Module.cpp",
     "ModuleSummaryIndex.cpp",
@@ -79,10 +80,10 @@ static_library("IR") {
     "TypedPointerType.cpp",
     "Use.cpp",
     "User.cpp",
+    "VFABIDemangler.cpp",
     "Value.cpp",
     "ValueSymbolTable.cpp",
     "VectorBuilder.cpp",
     "Verifier.cpp",
-    "VFABIDemangler.cpp",
   ]
 }
diff --git a/llvm/utils/gn/secondary/llvm/lib/Support/BUILD.gn b/llvm/utils/gn/secondary/llvm/lib/Support/BUILD.gn
index ba0f6d8c0f8c..941d448b3367 100644
--- a/llvm/utils/gn/secondary/llvm/lib/Support/BUILD.gn
+++ b/llvm/utils/gn/secondary/llvm/lib/Support/BUILD.gn
@@ -119,7 +119,7 @@ static_library("Support") {
     "PrettyStackTrace.cpp",
     "RISCVAttributeParser.cpp",
     "RISCVAttributes.cpp",
-    "RISCVISAInfo.cpp",
+    "RISCVISAUtils.cpp",
     "RWMutex.cpp",
     "RandomNumberGenerator.cpp",
     "Regex.cpp",
diff --git a/llvm/utils/gn/secondary/llvm/lib/Target/Hexagon/BUILD.gn b/llvm/utils/gn/secondary/llvm/lib/Target/Hexagon/BUILD.gn
index cae491a34331..7e873532b9ab 100644
--- a/llvm/utils/gn/secondary/llvm/lib/Target/Hexagon/BUILD.gn
+++ b/llvm/utils/gn/secondary/llvm/lib/Target/Hexagon/BUILD.gn
@@ -50,6 +50,7 @@ static_library("LLVMHexagonCodeGen") {
     "HexagonCommonGEP.cpp",
     "HexagonConstExtenders.cpp",
     "HexagonConstPropagation.cpp",
+    "HexagonCopyHoisting.cpp",
     "HexagonCopyToCombine.cpp",
     "HexagonEarlyIfConv.cpp",
     "HexagonExpandCondsets.cpp",
diff --git a/llvm/utils/gn/secondary/llvm/lib/Target/WebAssembly/BUILD.gn b/llvm/utils/gn/secondary/llvm/lib/Target/WebAssembly/BUILD.gn
index a8d6290f1b99..2ece91331c5d 100644
--- a/llvm/utils/gn/secondary/llvm/lib/Target/WebAssembly/BUILD.gn
+++ b/llvm/utils/gn/secondary/llvm/lib/Target/WebAssembly/BUILD.gn
@@ -37,6 +37,7 @@ static_library("LLVMWebAssemblyCodeGen") {
     "WebAssemblyAsmPrinter.cpp",
     "WebAssemblyCFGSort.cpp",
     "WebAssemblyCFGStackify.cpp",
+    "WebAssemblyCleanCodeAfterTrap.cpp",
     "WebAssemblyDebugFixup.cpp",
     "WebAssemblyDebugValueManager.cpp",
     "WebAssemblyExceptionInfo.cpp",
diff --git a/llvm/utils/gn/secondary/llvm/lib/TargetParser/BUILD.gn b/llvm/utils/gn/secondary/llvm/lib/TargetParser/BUILD.gn
index 1a91bb7e6fa9..31919badac7b 100644
--- a/llvm/utils/gn/secondary/llvm/lib/TargetParser/BUILD.gn
+++ b/llvm/utils/gn/secondary/llvm/lib/TargetParser/BUILD.gn
@@ -13,6 +13,7 @@ static_library("TargetParser") {
     "CSKYTargetParser.cpp",
     "Host.cpp",
     "LoongArchTargetParser.cpp",
+    "RISCVISAInfo.cpp",
     "RISCVTargetParser.cpp",
     "SubtargetFeature.cpp",
     "TargetParser.cpp",
diff --git a/llvm/utils/gn/secondary/llvm/unittests/Frontend/BUILD.gn b/llvm/utils/gn/secondary/llvm/unittests/Frontend/BUILD.gn
index c78ea923aae8..80ac77feec9a 100644
--- a/llvm/utils/gn/secondary/llvm/unittests/Frontend/BUILD.gn
+++ b/llvm/utils/gn/secondary/llvm/unittests/Frontend/BUILD.gn
@@ -13,7 +13,7 @@ unittest("LLVMFrontendTests") {
   ]
   sources = [
     "OpenACCTest.cpp",
-    "OpenMPComposeTest.cpp",
+    "OpenMPCompositionTest.cpp",
     "OpenMPContextTest.cpp",
     "OpenMPIRBuilderTest.cpp",
     "OpenMPParsingTest.cpp",
diff --git a/llvm/utils/gn/secondary/llvm/unittests/IR/BUILD.gn b/llvm/utils/gn/secondary/llvm/unittests/IR/BUILD.gn
index 4ffd6d4d182e..8f6caf2e575b 100644
--- a/llvm/utils/gn/secondary/llvm/unittests/IR/BUILD.gn
+++ b/llvm/utils/gn/secondary/llvm/unittests/IR/BUILD.gn
@@ -33,6 +33,7 @@ unittest("IRTests") {
     "LegacyPassManagerTest.cpp",
     "MDBuilderTest.cpp",
     "ManglerTest.cpp",
+    "MemoryModelRelaxationAnnotationsTest.cpp",
     "MetadataTest.cpp",
     "ModuleSummaryIndexTest.cpp",
     "ModuleTest.cpp",
@@ -45,6 +46,7 @@ unittest("IRTests") {
     "TypesTest.cpp",
     "UseTest.cpp",
     "UserTest.cpp",
+    "VFABIDemanglerTest.cpp",
     "VPIntrinsicTest.cpp",
     "ValueHandleTest.cpp",
     "ValueMapTest.cpp",
@@ -52,6 +54,5 @@ unittest("IRTests") {
     "VectorBuilderTest.cpp",
     "VectorTypesTest.cpp",
     "VerifierTest.cpp",
-    "VFABIDemanglerTest.cpp",
   ]
 }
diff --git a/llvm/utils/gn/secondary/llvm/unittests/Support/BUILD.gn b/llvm/utils/gn/secondary/llvm/unittests/Support/BUILD.gn
index 7a152fdcc059..c7356dd33a37 100644
--- a/llvm/utils/gn/secondary/llvm/unittests/Support/BUILD.gn
+++ b/llvm/utils/gn/secondary/llvm/unittests/Support/BUILD.gn
@@ -71,7 +71,6 @@ unittest("SupportTests") {
     "ProcessTest.cpp",
     "ProgramTest.cpp",
     "RISCVAttributeParserTest.cpp",
-    "RISCVISAInfoTest.cpp",
     "RegexTest.cpp",
     "ReplaceFileTest.cpp",
     "ReverseIterationTest.cpp",
diff --git a/llvm/utils/gn/secondary/llvm/unittests/TargetParser/BUILD.gn b/llvm/utils/gn/secondary/llvm/unittests/TargetParser/BUILD.gn
index eebaf67767f4..3739614c3160 100644
--- a/llvm/utils/gn/secondary/llvm/unittests/TargetParser/BUILD.gn
+++ b/llvm/utils/gn/secondary/llvm/unittests/TargetParser/BUILD.gn
@@ -10,6 +10,7 @@ unittest("TargetParserTests") {
   sources = [
     "CSKYTargetParserTest.cpp",
     "Host.cpp",
+    "RISCVISAInfoTest.cpp",
     "RISCVTargetParserTest.cpp",
     "TargetParserTest.cpp",
     "TripleTest.cpp",
diff --git a/llvm/utils/gn/secondary/llvm/utils/TableGen/BUILD.gn b/llvm/utils/gn/secondary/llvm/utils/TableGen/BUILD.gn
index 53a9d8d01519..f3ae5b5899ac 100644
--- a/llvm/utils/gn/secondary/llvm/utils/TableGen/BUILD.gn
+++ b/llvm/utils/gn/secondary/llvm/utils/TableGen/BUILD.gn
@@ -1,5 +1,6 @@
 source_set("llvm-min-tblgen-sources") {
   sources = [
+    "ARMTargetDefEmitter.cpp",
     "Attributes.cpp",
     "DirectiveEmitter.cpp",
     "IntrinsicEmitter.cpp",
diff --git a/llvm/utils/release/test-release.sh b/llvm/utils/release/test-release.sh
index 4314b565e11b..050004aa08c4 100755
--- a/llvm/utils/release/test-release.sh
+++ b/llvm/utils/release/test-release.sh
@@ -353,8 +353,7 @@ function build_with_cmake_cache() {
   env CC="$c_compiler" CXX="$cxx_compiler" \
   cmake -G "$generator" -B $CMakeBuildDir -S $SrcDir/llvm \
         -C $SrcDir/clang/cmake/caches/Release.cmake \
-	-DCLANG_BOOTSTRAP_PASSTHROUGH="CMAKE_POSITION_INDEPENDENT_CODE;LLVM_LIT_ARGS" \
-        -DCMAKE_POSITION_INDEPENDENT_CODE=ON \
+	-DCLANG_BOOTSTRAP_PASSTHROUGH="LLVM_LIT_ARGS" \
         -DLLVM_LIT_ARGS="-j $NumJobs $LitVerbose" \
         $ExtraConfigureFlags
         2>&1 | tee $LogDir/llvm.configure-$Flavor.log
diff --git a/mlir/CMakeLists.txt b/mlir/CMakeLists.txt
index 5c4301af040b..4c0ef8387b8d 100644
--- a/mlir/CMakeLists.txt
+++ b/mlir/CMakeLists.txt
@@ -185,10 +185,13 @@ include_directories( ${MLIR_INCLUDE_DIR})
 add_subdirectory(tools/mlir-linalg-ods-gen)
 add_subdirectory(tools/mlir-pdll)
 add_subdirectory(tools/mlir-tblgen)
+add_subdirectory(tools/mlir-src-sharder)
 set(MLIR_TABLEGEN_EXE "${MLIR_TABLEGEN_EXE}" CACHE INTERNAL "")
 set(MLIR_TABLEGEN_TARGET "${MLIR_TABLEGEN_TARGET}" CACHE INTERNAL "")
 set(MLIR_PDLL_TABLEGEN_EXE "${MLIR_PDLL_TABLEGEN_EXE}" CACHE INTERNAL "")
 set(MLIR_PDLL_TABLEGEN_TARGET "${MLIR_PDLL_TABLEGEN_TARGET}" CACHE INTERNAL "")
+set(MLIR_SRC_SHARDER_TABLEGEN_EXE "${MLIR_SRC_SHARDER_TABLEGEN_EXE}" CACHE INTERNAL "")
+set(MLIR_SRC_SHARDER_TABLEGEN_TARGET "${MLIR_SRC_SHARDER_TABLEGEN_TARGET}" CACHE INTERNAL "")
 
 add_subdirectory(include/mlir)
 add_subdirectory(lib)
diff --git a/mlir/cmake/modules/AddMLIR.cmake b/mlir/cmake/modules/AddMLIR.cmake
index 1d2ed748bc2f..afb74fb2d000 100644
--- a/mlir/cmake/modules/AddMLIR.cmake
+++ b/mlir/cmake/modules/AddMLIR.cmake
@@ -5,6 +5,28 @@ function(mlir_tablegen ofn)
   tablegen(MLIR ${ARGV})
   set(TABLEGEN_OUTPUT ${TABLEGEN_OUTPUT} ${CMAKE_CURRENT_BINARY_DIR}/${ofn}
       PARENT_SCOPE)
+
+  # Get the current set of include paths for this td file.
+  cmake_parse_arguments(ARG "" "" "DEPENDS;EXTRA_INCLUDES" ${ARGN})
+  get_directory_property(tblgen_includes INCLUDE_DIRECTORIES)
+  list(APPEND tblgen_includes ${ARG_EXTRA_INCLUDES})
+  # Filter out any empty include items.
+  list(REMOVE_ITEM tblgen_includes "")
+
+  # Build the absolute path for the current input file.
+  if (IS_ABSOLUTE ${LLVM_TARGET_DEFINITIONS})
+    set(LLVM_TARGET_DEFINITIONS_ABSOLUTE ${LLVM_TARGET_DEFINITIONS})
+  else()
+    set(LLVM_TARGET_DEFINITIONS_ABSOLUTE ${CMAKE_CURRENT_SOURCE_DIR}/${LLVM_TARGET_DEFINITIONS})
+  endif()
+
+  # Append the includes used for this file to the tablegen_compile_commands
+  # file.
+  file(APPEND ${CMAKE_BINARY_DIR}/tablegen_compile_commands.yml
+      "--- !FileInfo:\n"
+      "  filepath: \"${LLVM_TARGET_DEFINITIONS_ABSOLUTE}\"\n"
+      "  includes: \"${CMAKE_CURRENT_SOURCE_DIR};${tblgen_includes}\"\n"
+  )
 endfunction()
 
 # Clear out any pre-existing compile_commands file before processing. This
@@ -149,6 +171,22 @@ function(add_mlir_dialect dialect dialect_namespace)
   add_dependencies(mlir-headers MLIR${dialect}IncGen)
 endfunction()
 
+# Declare sharded dialect operation declarations and definitions
+function(add_sharded_ops ops_target shard_count)
+  set(LLVM_TARGET_DEFINITIONS ${ops_target}.td)
+  mlir_tablegen(${ops_target}.h.inc -gen-op-decls -op-shard-count=${shard_count})
+  mlir_tablegen(${ops_target}.cpp.inc -gen-op-defs -op-shard-count=${shard_count})
+  set(LLVM_TARGET_DEFINITIONS ${ops_target}.cpp)
+  foreach(index RANGE ${shard_count})
+    set(SHARDED_SRC ${ops_target}.${index}.cpp)
+    list(APPEND SHARDED_SRCS ${SHARDED_SRC})
+    tablegen(MLIR_SRC_SHARDER ${SHARDED_SRC} -op-shard-index=${index})
+    set(TABLEGEN_OUTPUT ${TABLEGEN_OUTPUT} ${CMAKE_CURRENT_BINARY_DIR}/${SHARDED_SRC})
+  endforeach()
+  add_public_tablegen_target(MLIR${ops_target}ShardGen)
+  set(SHARDED_SRCS ${SHARDED_SRCS} PARENT_SCOPE)
+endfunction()
+
 # Declare a dialect in the include directory
 function(add_mlir_interface interface)
   set(LLVM_TARGET_DEFINITIONS ${interface}.td)
diff --git a/mlir/cmake/modules/CMakeLists.txt b/mlir/cmake/modules/CMakeLists.txt
index 8d2904ef46df..3ac1c79b090e 100644
--- a/mlir/cmake/modules/CMakeLists.txt
+++ b/mlir/cmake/modules/CMakeLists.txt
@@ -39,6 +39,7 @@ set(MLIR_CONFIG_INCLUDE_DIRS
 # Refer to the best host mlir-tbgen, which might be a host-optimized version
 set(MLIR_CONFIG_TABLEGEN_EXE "${MLIR_TABLEGEN_EXE}")
 set(MLIR_CONFIG_PDLL_TABLEGEN_EXE "${MLIR_PDLL_TABLEGEN_EXE}")
+set(MLIR_CONFIG_SRC_SHARDER_TABLEGEN_EXE "${MLIR_SRC_SHARDER_TABLEGEN_EXE}")
 
 configure_file(
   ${CMAKE_CURRENT_SOURCE_DIR}/MLIRConfig.cmake.in
@@ -77,6 +78,7 @@ set(MLIR_CONFIG_INCLUDE_DIRS
 # if we're building with a host-optimized mlir-tblgen (with LLVM_OPTIMIZED_TABLEGEN).
 set(MLIR_CONFIG_TABLEGEN_EXE mlir-tblgen)
 set(MLIR_CONFIG_PDLL_TABLEGEN_EXE mlir-pdll)
+set(MLIR_CONFIG_SRC_SHARDER_TABLEGEN_EXE mlir-src-sharder)
 
 configure_file(
   ${CMAKE_CURRENT_SOURCE_DIR}/MLIRConfig.cmake.in
diff --git a/mlir/cmake/modules/MLIRConfig.cmake.in b/mlir/cmake/modules/MLIRConfig.cmake.in
index d4da3cd98cce..7076d94a32f2 100644
--- a/mlir/cmake/modules/MLIRConfig.cmake.in
+++ b/mlir/cmake/modules/MLIRConfig.cmake.in
@@ -11,6 +11,7 @@ set(MLIR_CMAKE_DIR "@MLIR_CONFIG_CMAKE_DIR@")
 set(MLIR_INCLUDE_DIRS "@MLIR_CONFIG_INCLUDE_DIRS@")
 set(MLIR_TABLEGEN_EXE "@MLIR_CONFIG_TABLEGEN_EXE@")
 set(MLIR_PDLL_TABLEGEN_EXE "@MLIR_CONFIG_PDLL_TABLEGEN_EXE@")
+set(MLIR_SRC_SHARDER_TABLEGEN_EXE "@MLIR_CONFIG_SRC_SHARDER_TABLEGEN_EXE@")
 set(MLIR_INSTALL_AGGREGATE_OBJECTS "@MLIR_INSTALL_AGGREGATE_OBJECTS@")
 set(MLIR_ENABLE_BINDINGS_PYTHON "@MLIR_ENABLE_BINDINGS_PYTHON@")
 set(MLIR_ENABLE_EXECUTION_ENGINE "@MLIR_ENABLE_EXECUTION_ENGINE@")
diff --git a/mlir/docs/DefiningDialects/Operations.md b/mlir/docs/DefiningDialects/Operations.md
index 729393d53626..79a0cc55f138 100644
--- a/mlir/docs/DefiningDialects/Operations.md
+++ b/mlir/docs/DefiningDialects/Operations.md
@@ -1114,6 +1114,100 @@ void process(AddOp op, ArrayRef<Value> newOperands) {
 }
 ```
 
+#### Sharded Operation Definitions
+
+Large dialects with many operations may struggle with C++ compile time of
+generated op definitions, due to large compilation units. `mlir-tblgen`
+provides the ability to shard op definitions by splitting them up evenly
+by passing `-op-shard-count` to `-gen-op-defs` and `-gen-op-decls`. The tool
+will generate a single include file for the definitions broken up by
+`GET_OP_DEFS_${N}` where `${N}` is the shard number. A shard can be compiled in
+a single compilation unit by adding a file like this to your dialect library:
+
+```c++
+#include "mlir/IR/Operation.h"
+// Add any other required includes.
+
+// Utilities shared by generated op definitions: custom directive parsers,
+// printers, etc.
+#include "OpUtils.h"
+
+#define GET_OP_DEFS_0
+#include "MyDialectOps.cpp.inc"
+```
+
+Note: this requires restructing shared utility functions within the dialect
+library so they can be shared by multiple compilation units. I.e. instead of
+defining `static` methods in the same source file, you should declare them in a
+shared header and define them in their own source file.
+
+The op registration hooks are also sharded, because the template instantiation
+can take a very long time to compile. Operations should be registered in your
+dialect like:
+
+```c++
+void MyDialect::initialize() {
+  registerMyDialectOperations(this);
+}
+```
+
+CMake and Bazel functions are included to make sharding dialects easier.
+Assuming you have organized your operation utility functions into their own
+header, define a file that looks like the one above, but without the `#define`:
+
+```c++
+// MyDialectOps.cpp
+#include "mlir/IR/Operation.h"
+
+#include "OpUtils.h"
+
+#include "MyDialectOps.cpp.inc"
+```
+
+In CMake, remove the manual `mlir_tablegen` invocations and replace them with:
+
+```cmake
+set(LLVM_TARGET_DEFINITIONS MyDialectOps.td)
+add_sharded_ops(MyDialectOps 8) # shard the op definitions by 8
+
+add_mlir_library(MyDialect
+  MyDialect.cpp
+  MyDialectOpDefs.cpp
+  ${SHARDED_SRCS}
+
+  DEPENDS
+  MLIRTestOpsShardGen
+)
+```
+
+This will automatically duplicate the `MyDialectOps.cpp` source file and add the
+`#define` up the number of shards indicated.
+
+It is recommended that any out-of-line op member functions (like verifiers) be
+defined in a separate source file. In this example, it is called
+`MyDialectOpDefs.cpp`.
+
+In Bazel, remove the `-gen-op-defs` and `-gen-op-decls` invocations, and add
+
+```bazel
+gentbl_sharded_ops(
+    name = "MyDialectOpSrcs",
+    hdr_out = "MyDialectOps.h.inc",
+    shard_count = 8,
+    sharder = "//mlir:mlir-src-sharder",
+    src_file = "MyDialectOps.cpp",
+    src_out = "MyDialectOps.cpp.inc",
+    tblgen = "//mlir:mlir-tblgen",
+    td_file = "MyDialectOps.td",
+    deps = [":MyDialectOpsTdFiles"],
+)
+
+cc_library(
+    name = "MyDialect",
+    srcs = glob(["MyDialect/*.cpp"]) + [":MyDialectOpSrcs"]
+)
+```
+
 ## Constraints
 
 Constraint is a core concept in table-driven operation definition: operation
diff --git a/mlir/docs/Tutorials/UnderstandingTheIRStructure.md b/mlir/docs/Tutorials/UnderstandingTheIRStructure.md
index 067a11dc435a..ed323fc42336 100644
--- a/mlir/docs/Tutorials/UnderstandingTheIRStructure.md
+++ b/mlir/docs/Tutorials/UnderstandingTheIRStructure.md
@@ -40,8 +40,8 @@ the nested regions and print them individually:
     if (!op->getAttrs().empty()) {
       printIndent() << op->getAttrs().size() << " attributes:\n";
       for (NamedAttribute attr : op->getAttrs())
-        printIndent() << " - '" << attr.first << "' : '" << attr.second
-                      << "'\n";
+        printIndent() << " - '" << attr.getName() << "' : '"
+                      << attr.getValue() << "'\n";
     }
 
     // Recurse into each of the regions attached to the operation.
diff --git a/mlir/include/mlir-c/Dialect/LLVM.h b/mlir/include/mlir-c/Dialect/LLVM.h
index bd9b7dd26f5e..b3e64bd68f7b 100644
--- a/mlir/include/mlir-c/Dialect/LLVM.h
+++ b/mlir/include/mlir-c/Dialect/LLVM.h
@@ -23,6 +23,13 @@ MLIR_DECLARE_CAPI_DIALECT_REGISTRATION(LLVM, llvm);
 MLIR_CAPI_EXPORTED MlirType mlirLLVMPointerTypeGet(MlirContext ctx,
                                                    unsigned addressSpace);
 
+/// Returns `true` if the type is an LLVM dialect pointer type.
+MLIR_CAPI_EXPORTED bool mlirTypeIsALLVMPointerType(MlirType type);
+
+/// Returns address space of llvm.ptr
+MLIR_CAPI_EXPORTED unsigned
+mlirLLVMPointerTypeGetAddressSpace(MlirType pointerType);
+
 /// Creates an llmv.void type.
 MLIR_CAPI_EXPORTED MlirType mlirLLVMVoidTypeGet(MlirContext ctx);
 
diff --git a/mlir/include/mlir-c/Dialect/SparseTensor.h b/mlir/include/mlir-c/Dialect/SparseTensor.h
index 52ca7ba8a161..125469f57c5f 100644
--- a/mlir/include/mlir-c/Dialect/SparseTensor.h
+++ b/mlir/include/mlir-c/Dialect/SparseTensor.h
@@ -53,7 +53,8 @@ mlirAttributeIsASparseTensorEncodingAttr(MlirAttribute attr);
 MLIR_CAPI_EXPORTED MlirAttribute mlirSparseTensorEncodingAttrGet(
     MlirContext ctx, intptr_t lvlRank,
     MlirSparseTensorLevelType const *lvlTypes, MlirAffineMap dimToLvl,
-    MlirAffineMap lvlTodim, int posWidth, int crdWidth);
+    MlirAffineMap lvlTodim, int posWidth, int crdWidth,
+    MlirAttribute explicitVal, MlirAttribute implicitVal);
 
 /// Returns the level-rank of the `sparse_tensor.encoding` attribute.
 MLIR_CAPI_EXPORTED intptr_t
@@ -85,6 +86,14 @@ mlirSparseTensorEncodingAttrGetPosWidth(MlirAttribute attr);
 MLIR_CAPI_EXPORTED int
 mlirSparseTensorEncodingAttrGetCrdWidth(MlirAttribute attr);
 
+/// Returns the explicit value of the `sparse_tensor.encoding` attribute.
+MLIR_CAPI_EXPORTED MlirAttribute
+mlirSparseTensorEncodingAttrGetExplicitVal(MlirAttribute attr);
+
+/// Returns the implicit value of the `sparse_tensor.encoding` attribute.
+MLIR_CAPI_EXPORTED MlirAttribute
+mlirSparseTensorEncodingAttrGetImplicitVal(MlirAttribute attr);
+
 MLIR_CAPI_EXPORTED unsigned
 mlirSparseTensorEncodingAttrGetStructuredN(MlirSparseTensorLevelType lvlType);
 
diff --git a/mlir/include/mlir/Dialect/EmitC/IR/EmitC.td b/mlir/include/mlir/Dialect/EmitC/IR/EmitC.td
index c1a1e77c34ab..5da8593f5956 100644
--- a/mlir/include/mlir/Dialect/EmitC/IR/EmitC.td
+++ b/mlir/include/mlir/Dialect/EmitC/IR/EmitC.td
@@ -1014,6 +1014,75 @@ def EmitC_VariableOp : EmitC_Op<"variable", []> {
   let hasVerifier = 1;
 }
 
+def EmitC_GlobalOp : EmitC_Op<"global", [Symbol]> {
+  let summary = "A global variable";
+  let description = [{
+    The `emitc.global` operation declares or defines a named global variable.
+    The backing memory for the variable is allocated statically and is
+    described by the type of the variable.
+    Optionally, an `initial_value` can be provided.
+    Internal linkage can be specified using the `static_specifier` unit attribute
+    and external linkage can be specified using the `extern_specifier` unit attribute.
+    Note that the default linkage without those two keywords depends on whether
+    the target is C or C++ and whether the global variable is `const`.
+    The global variable can also be marked constant using the `const_specifier`
+    unit attribute. Writing to such constant global variables is
+    undefined.
+
+    The global variable can be accessed by using the `emitc.get_global` to
+    retrieve the value for the global variable.
+
+    Example:
+
+    ```mlir
+    // Global variable with an initial value.
+    emitc.global @x : emitc.array<2xf32> = dense<0.0, 2.0>
+    // External global variable
+    emitc.global extern @x : emitc.array<2xf32>
+    // Constant global variable with internal linkage
+    emitc.global static const @x : i32 = 0
+    ```
+  }];
+
+  let arguments = (ins SymbolNameAttr:$sym_name,
+                       TypeAttr:$type,
+                       OptionalAttr<EmitC_OpaqueOrTypedAttr>:$initial_value,
+                       UnitAttr:$extern_specifier,
+                       UnitAttr:$static_specifier,
+                       UnitAttr:$const_specifier);
+
+  let assemblyFormat = [{
+       (`extern` $extern_specifier^)?
+       (`static` $static_specifier^)?
+       (`const` $const_specifier^)?
+       $sym_name
+       `:` custom<EmitCGlobalOpTypeAndInitialValue>($type, $initial_value)
+       attr-dict
+  }];
+
+  let hasVerifier = 1;
+}
+
+def EmitC_GetGlobalOp : EmitC_Op<"get_global",
+    [Pure, DeclareOpInterfaceMethods<SymbolUserOpInterface>]> {
+  let summary = "Obtain access to a global variable";
+  let description = [{
+     The `emitc.get_global` operation retrieves the lvalue of a
+     named global variable. If the global variable is marked constant, assigning
+     to that lvalue is undefined.
+
+     Example:
+
+     ```mlir
+     %x = emitc.get_global @foo : !emitc.array<2xf32>
+     ```
+  }];
+
+  let arguments = (ins FlatSymbolRefAttr:$name);
+  let results = (outs EmitCType:$result);
+  let assemblyFormat = "$name `:` type($result) attr-dict";
+}
+
 def EmitC_VerbatimOp : EmitC_Op<"verbatim"> {
   let summary = "Verbatim operation";
   let description = [{
diff --git a/mlir/include/mlir/Dialect/GPU/Transforms/Passes.td b/mlir/include/mlir/Dialect/GPU/Transforms/Passes.td
index a8235bed6f27..4a9ddafdd177 100644
--- a/mlir/include/mlir/Dialect/GPU/Transforms/Passes.td
+++ b/mlir/include/mlir/Dialect/GPU/Transforms/Passes.td
@@ -88,8 +88,6 @@ def GpuModuleToBinaryPass
     4. `fatbinary`, `fatbin`: produces fatbinaries.
   }];
   let options = [
-    Option<"offloadingHandler", "handler", "Attribute", "nullptr",
-           "Offloading handler to be attached to the resulting binary op.">,
     Option<"toolkitPath", "toolkit", "std::string", [{""}],
            "Toolkit path.">,
     ListOption<"linkFiles", "l", "std::string",
diff --git a/mlir/include/mlir/Dialect/LLVMIR/LLVMIntrinsicOps.td b/mlir/include/mlir/Dialect/LLVMIR/LLVMIntrinsicOps.td
index a52cca3c95de..759cbe6c1564 100644
--- a/mlir/include/mlir/Dialect/LLVMIR/LLVMIntrinsicOps.td
+++ b/mlir/include/mlir/Dialect/LLVMIR/LLVMIntrinsicOps.td
@@ -1060,8 +1060,8 @@ def LLVM_vector_extract
   }];
 }
 
-def LLVM_experimental_vector_interleave2
-    : LLVM_OneResultIntrOp<"experimental.vector.interleave2",
+def LLVM_vector_interleave2
+    : LLVM_OneResultIntrOp<"vector.interleave2",
         /*overloadedResults=*/[0], /*overloadedOperands=*/[],
         /*traits=*/[
           Pure, AllTypesMatch<["vec1", "vec2"]>,
diff --git a/mlir/include/mlir/Dialect/LLVMIR/VCIXOps.td b/mlir/include/mlir/Dialect/LLVMIR/VCIXOps.td
index 25c1d027768a..27d9a32dd8e0 100644
--- a/mlir/include/mlir/Dialect/LLVMIR/VCIXOps.td
+++ b/mlir/include/mlir/Dialect/LLVMIR/VCIXOps.td
@@ -78,7 +78,7 @@ def VCIX_BinaryImmOp : VCIX_Op<"v.iv">,
           xlen, $opcode, $_location, moduleTranslation);
       llvm::Value *immConst = mlir::LLVM::detail::getLLVMConstant(
           xlen, $imm, $_location, moduleTranslation);
-      VectorType vt = op.getResult().getType().cast<VectorType>();
+      VectorType vt = mlir::cast<VectorType>(op.getResult().getType());
       llvm::Value *vl =
           createVL(builder, $vl, vt, xlen, $_location, moduleTranslation);
       $res = createIntrinsicCall(
@@ -120,7 +120,7 @@ def VCIX_BinaryOp : VCIX_Op<"v.sv">,
       } else {
         id = llvm::Intrinsic::riscv_sf_vc_v_fv_se;
       }
-      VectorType vt = op.getResult().getType().cast<VectorType>();
+      VectorType vt = mlir::cast<VectorType>(op.getResult().getType());
       llvm::Value *vl =
           createVL(builder, $vl, vt, xlen, $_location, moduleTranslation);
       $res = createIntrinsicCall(
diff --git a/mlir/include/mlir/Dialect/Linalg/IR/LinalgEnums.td b/mlir/include/mlir/Dialect/Linalg/IR/LinalgEnums.td
index 59f909aed8f6..6b4b073fc672 100644
--- a/mlir/include/mlir/Dialect/Linalg/IR/LinalgEnums.td
+++ b/mlir/include/mlir/Dialect/Linalg/IR/LinalgEnums.td
@@ -22,7 +22,14 @@ def UnaryFn : I32EnumAttr<"UnaryFn", "", [
   I32EnumAttrCase<"abs", 2>,
   I32EnumAttrCase<"ceil", 3>,
   I32EnumAttrCase<"floor", 4>,
-  I32EnumAttrCase<"negf", 5>
+  I32EnumAttrCase<"negf", 5>,
+  I32EnumAttrCase<"reciprocal", 6>,
+  I32EnumAttrCase<"round", 7>,
+  I32EnumAttrCase<"sqrt", 8>,
+  I32EnumAttrCase<"rsqrt", 9>,
+  I32EnumAttrCase<"square", 10>,
+  I32EnumAttrCase<"tanh", 11>,
+  I32EnumAttrCase<"erf", 12>
 ]> {
   let genSpecializedAttr = 0;
   let cppNamespace = "::mlir::linalg";
@@ -36,7 +43,8 @@ def BinaryFn : I32EnumAttr<"BinaryFn", "", [
   I32EnumAttrCase<"max_signed", 5>,
   I32EnumAttrCase<"min_signed", 6>,
   I32EnumAttrCase<"max_unsigned", 7>,
-  I32EnumAttrCase<"min_unsigned", 8>
+  I32EnumAttrCase<"min_unsigned", 8>,
+  I32EnumAttrCase<"powf", 9>
 ]> {
   let genSpecializedAttr = 0;
   let cppNamespace = "::mlir::linalg";
diff --git a/mlir/include/mlir/Dialect/Linalg/IR/LinalgNamedStructuredOps.yaml b/mlir/include/mlir/Dialect/Linalg/IR/LinalgNamedStructuredOps.yaml
index 1ff6c4086cf3..584bfcd8b59d 100644
--- a/mlir/include/mlir/Dialect/Linalg/IR/LinalgNamedStructuredOps.yaml
+++ b/mlir/include/mlir/Dialect/Linalg/IR/LinalgNamedStructuredOps.yaml
@@ -305,6 +305,251 @@ structured_op: !LinalgStructuredOpConfig
           scalar_arg: I
 --- !LinalgOpConfig
 metadata: !LinalgOpMetadata
+  name: reciprocal
+  cpp_class_name: ReciprocalOp
+  doc: |-
+    Applies reciprocal(x) elementwise.
+
+    No numeric casting is performed on the input operand.
+structured_op: !LinalgStructuredOpConfig
+  args:
+  - !LinalgOperandDefConfig
+    name: I
+    kind: input_tensor
+    type_var: T1
+    shape_map: affine_map<() -> ()>
+  - !LinalgOperandDefConfig
+    name: O
+    kind: output_tensor
+    type_var: T1
+    shape_map: affine_map<() -> ()>
+  indexing_maps: !LinalgIndexingMapsConfig
+    static_indexing_maps:
+    - affine_map<() -> ()>
+    - affine_map<() -> ()>
+  iterator_types: []
+  assignments:
+  - !ScalarAssign
+    arg: O
+    value: !ScalarExpression
+      scalar_fn:
+        kind: unary
+        fn_name: reciprocal
+        operands:
+        - !ScalarExpression
+          scalar_arg: I
+--- !LinalgOpConfig
+metadata: !LinalgOpMetadata
+  name: round
+  cpp_class_name: RoundOp
+  doc: |-
+    Applies round(x) elementwise.
+
+    No numeric casting is performed on the input operand.
+structured_op: !LinalgStructuredOpConfig
+  args:
+  - !LinalgOperandDefConfig
+    name: I
+    kind: input_tensor
+    type_var: T1
+    shape_map: affine_map<() -> ()>
+  - !LinalgOperandDefConfig
+    name: O
+    kind: output_tensor
+    type_var: T1
+    shape_map: affine_map<() -> ()>
+  indexing_maps: !LinalgIndexingMapsConfig
+    static_indexing_maps:
+    - affine_map<() -> ()>
+    - affine_map<() -> ()>
+  iterator_types: []
+  assignments:
+  - !ScalarAssign
+    arg: O
+    value: !ScalarExpression
+      scalar_fn:
+        kind: unary
+        fn_name: round
+        operands:
+        - !ScalarExpression
+          scalar_arg: I
+--- !LinalgOpConfig
+metadata: !LinalgOpMetadata
+  name: sqrt
+  cpp_class_name: SqrtOp
+  doc: |-
+    Applies sqrt(x) elementwise.
+
+    No numeric casting is performed on the input operand.
+structured_op: !LinalgStructuredOpConfig
+  args:
+  - !LinalgOperandDefConfig
+    name: I
+    kind: input_tensor
+    type_var: T1
+    shape_map: affine_map<() -> ()>
+  - !LinalgOperandDefConfig
+    name: O
+    kind: output_tensor
+    type_var: T1
+    shape_map: affine_map<() -> ()>
+  indexing_maps: !LinalgIndexingMapsConfig
+    static_indexing_maps:
+    - affine_map<() -> ()>
+    - affine_map<() -> ()>
+  iterator_types: []
+  assignments:
+  - !ScalarAssign
+    arg: O
+    value: !ScalarExpression
+      scalar_fn:
+        kind: unary
+        fn_name: sqrt
+        operands:
+        - !ScalarExpression
+          scalar_arg: I
+--- !LinalgOpConfig
+metadata: !LinalgOpMetadata
+  name: rsqrt
+  cpp_class_name: RsqrtOp
+  doc: |-
+    Applies rsqrt(x) elementwise.
+
+    No numeric casting is performed on the input operand.
+structured_op: !LinalgStructuredOpConfig
+  args:
+  - !LinalgOperandDefConfig
+    name: I
+    kind: input_tensor
+    type_var: T1
+    shape_map: affine_map<() -> ()>
+  - !LinalgOperandDefConfig
+    name: O
+    kind: output_tensor
+    type_var: T1
+    shape_map: affine_map<() -> ()>
+  indexing_maps: !LinalgIndexingMapsConfig
+    static_indexing_maps:
+    - affine_map<() -> ()>
+    - affine_map<() -> ()>
+  iterator_types: []
+  assignments:
+  - !ScalarAssign
+    arg: O
+    value: !ScalarExpression
+      scalar_fn:
+        kind: unary
+        fn_name: rsqrt
+        operands:
+        - !ScalarExpression
+          scalar_arg: I
+--- !LinalgOpConfig
+metadata: !LinalgOpMetadata
+  name: square
+  cpp_class_name: SquareOp
+  doc: |-
+    Applies square(x) elementwise.
+
+    No numeric casting is performed on the input operand.
+structured_op: !LinalgStructuredOpConfig
+  args:
+  - !LinalgOperandDefConfig
+    name: I
+    kind: input_tensor
+    type_var: T1
+    shape_map: affine_map<() -> ()>
+  - !LinalgOperandDefConfig
+    name: O
+    kind: output_tensor
+    type_var: T1
+    shape_map: affine_map<() -> ()>
+  indexing_maps: !LinalgIndexingMapsConfig
+    static_indexing_maps:
+    - affine_map<() -> ()>
+    - affine_map<() -> ()>
+  iterator_types: []
+  assignments:
+  - !ScalarAssign
+    arg: O
+    value: !ScalarExpression
+      scalar_fn:
+        kind: unary
+        fn_name: square
+        operands:
+        - !ScalarExpression
+          scalar_arg: I
+--- !LinalgOpConfig
+metadata: !LinalgOpMetadata
+  name: tanh
+  cpp_class_name: TanhOp
+  doc: |-
+    Applies tanh(x) elementwise.
+
+    No numeric casting is performed on the input operand.
+structured_op: !LinalgStructuredOpConfig
+  args:
+  - !LinalgOperandDefConfig
+    name: I
+    kind: input_tensor
+    type_var: T1
+    shape_map: affine_map<() -> ()>
+  - !LinalgOperandDefConfig
+    name: O
+    kind: output_tensor
+    type_var: T1
+    shape_map: affine_map<() -> ()>
+  indexing_maps: !LinalgIndexingMapsConfig
+    static_indexing_maps:
+    - affine_map<() -> ()>
+    - affine_map<() -> ()>
+  iterator_types: []
+  assignments:
+  - !ScalarAssign
+    arg: O
+    value: !ScalarExpression
+      scalar_fn:
+        kind: unary
+        fn_name: tanh
+        operands:
+        - !ScalarExpression
+          scalar_arg: I
+--- !LinalgOpConfig
+metadata: !LinalgOpMetadata
+  name: erf
+  cpp_class_name: erfOp
+  doc: |-
+    Applies erf(x) elementwise.
+
+    No numeric casting is performed on the input operand.
+structured_op: !LinalgStructuredOpConfig
+  args:
+  - !LinalgOperandDefConfig
+    name: I
+    kind: input_tensor
+    type_var: T1
+    shape_map: affine_map<() -> ()>
+  - !LinalgOperandDefConfig
+    name: O
+    kind: output_tensor
+    type_var: T1
+    shape_map: affine_map<() -> ()>
+  indexing_maps: !LinalgIndexingMapsConfig
+    static_indexing_maps:
+    - affine_map<() -> ()>
+    - affine_map<() -> ()>
+  iterator_types: []
+  assignments:
+  - !ScalarAssign
+    arg: O
+    value: !ScalarExpression
+      scalar_fn:
+        kind: unary
+        fn_name: erf
+        operands:
+        - !ScalarExpression
+          scalar_arg: I
+--- !LinalgOpConfig
+metadata: !LinalgOpMetadata
   name: elemwise_binary
   cpp_class_name: ElemwiseBinaryOp
   doc: |-
@@ -625,7 +870,7 @@ metadata: !LinalgOpMetadata
 
     This means reduction/broadcast/element cast semantics is explicit. Further
     passes can take that into account when lowering this code. For example,
-    a `linalg.broadcast` + `linalg.div` sequence can be lowered to a
+    a `linalg.broadcast` + `linalg.max` sequence can be lowered to a
     `linalg.generic` with different affine maps for the two operands.
 structured_op: !LinalgStructuredOpConfig
   args:
@@ -664,6 +909,106 @@ structured_op: !LinalgStructuredOpConfig
           scalar_arg: rhs
 --- !LinalgOpConfig
 metadata: !LinalgOpMetadata
+  name: min
+  cpp_class_name: MinOp
+  doc: |-
+    Takes the min (signed) between two inputs, elementwise.
+
+    The shapes and element types must be identical. The appropriate casts,
+    broadcasts and reductions should be done previously to calling this op.
+
+    This means reduction/broadcast/element cast semantics is explicit. Further
+    passes can take that into account when lowering this code. For example,
+    a `linalg.broadcast` + `linalg.min` sequence can be lowered to a
+    `linalg.generic` with different affine maps for the two operands.
+structured_op: !LinalgStructuredOpConfig
+  args:
+  - !LinalgOperandDefConfig
+    name: lhs
+    kind: input_tensor
+    type_var: T1
+    shape_map: affine_map<() -> ()>
+  - !LinalgOperandDefConfig
+    name: rhs
+    kind: input_tensor
+    type_var: T1
+    shape_map: affine_map<() -> ()>
+  - !LinalgOperandDefConfig
+    name: O
+    kind: output_tensor
+    type_var: T1
+    shape_map: affine_map<() -> ()>
+  indexing_maps: !LinalgIndexingMapsConfig
+    static_indexing_maps:
+    - affine_map<() -> ()>
+    - affine_map<() -> ()>
+    - affine_map<() -> ()>
+  iterator_types: []
+  assignments:
+  - !ScalarAssign
+    arg: O
+    value: !ScalarExpression
+      scalar_fn:
+        kind: binary
+        fn_name: min_signed
+        operands:
+        - !ScalarExpression
+          scalar_arg: lhs
+        - !ScalarExpression
+          scalar_arg: rhs
+--- !LinalgOpConfig
+metadata: !LinalgOpMetadata
+  name: powf
+  cpp_class_name: PowFOp
+  doc: |-
+    Takes the powf(lhs, rhs) between two inputs, elementwise. For powf(arg, 2) use `linalg.square`.
+
+    Only applies to floating point values.
+
+    The shapes and element types must be identical. The appropriate casts,
+    broadcasts and reductions should be done previously to calling this op.
+
+    This means reduction/broadcast/element cast semantics is explicit. Further
+    passes can take that into account when lowering this code. For example,
+    a `linalg.broadcast` + `linalg.powf` sequence can be lowered to a
+    `linalg.generic` with different affine maps for the two operands.
+structured_op: !LinalgStructuredOpConfig
+  args:
+  - !LinalgOperandDefConfig
+    name: lhs
+    kind: input_tensor
+    type_var: T1
+    shape_map: affine_map<() -> ()>
+  - !LinalgOperandDefConfig
+    name: rhs
+    kind: input_tensor
+    type_var: T1
+    shape_map: affine_map<() -> ()>
+  - !LinalgOperandDefConfig
+    name: O
+    kind: output_tensor
+    type_var: T1
+    shape_map: affine_map<() -> ()>
+  indexing_maps: !LinalgIndexingMapsConfig
+    static_indexing_maps:
+    - affine_map<() -> ()>
+    - affine_map<() -> ()>
+    - affine_map<() -> ()>
+  iterator_types: []
+  assignments:
+  - !ScalarAssign
+    arg: O
+    value: !ScalarExpression
+      scalar_fn:
+        kind: binary
+        fn_name: powf
+        operands:
+        - !ScalarExpression
+          scalar_arg: lhs
+        - !ScalarExpression
+          scalar_arg: rhs
+--- !LinalgOpConfig
+metadata: !LinalgOpMetadata
   name: matmul
   cpp_class_name: MatmulOp
   doc: |-
diff --git a/mlir/include/mlir/Dialect/Linalg/IR/LinalgOps.td b/mlir/include/mlir/Dialect/Linalg/IR/LinalgOps.td
index da12e7c83b22..64c538367267 100644
--- a/mlir/include/mlir/Dialect/Linalg/IR/LinalgOps.td
+++ b/mlir/include/mlir/Dialect/Linalg/IR/LinalgOps.td
@@ -138,10 +138,10 @@ def Linalg_SoftmaxOp : Linalg_Op<"softmax",
 
   let extraClassDeclaration = [{
     ShapedType getInputOperandType() {
-      return getInput().getType().cast<ShapedType>();
+      return cast<ShapedType>(getInput().getType());
     }
     ShapedType getOutputOperandType() {
-      return getOutput().getType().cast<ShapedType>();
+      return cast<ShapedType>(getOutput().getType());
     }
     int64_t getInputOperandRank() {
       return getInputOperandType().getRank();
diff --git a/mlir/include/mlir/Dialect/Linalg/TransformOps/LinalgTransformEnums.td b/mlir/include/mlir/Dialect/Linalg/TransformOps/LinalgTransformEnums.td
index e4cf4e194b7a..c72c04517db8 100644
--- a/mlir/include/mlir/Dialect/Linalg/TransformOps/LinalgTransformEnums.td
+++ b/mlir/include/mlir/Dialect/Linalg/TransformOps/LinalgTransformEnums.td
@@ -8,3 +8,12 @@ def MatchInterfaceEnum : I32EnumAttr<"MatchInterfaceEnum", "An interface to matc
     ]>{
   let cppNamespace = "mlir::transform";
 }
+
+def TransposeMatmulInput : I32EnumAttr<"TransposeMatmulInput",
+    "Input to transpose when converting matmul ops to transposed variants",
+    [
+      I32EnumAttrCase<"lhs", 0>,
+      I32EnumAttrCase<"rhs", 1>,
+    ]>{
+  let cppNamespace = "mlir::transform";
+}
diff --git a/mlir/include/mlir/Dialect/Linalg/TransformOps/LinalgTransformOps.td b/mlir/include/mlir/Dialect/Linalg/TransformOps/LinalgTransformOps.td
index 157dc671f720..d0ad4ccdf031 100644
--- a/mlir/include/mlir/Dialect/Linalg/TransformOps/LinalgTransformOps.td
+++ b/mlir/include/mlir/Dialect/Linalg/TransformOps/LinalgTransformOps.td
@@ -2413,6 +2413,52 @@ def TransposeConv2DOp : Op<Transform_Dialect,
 }
 
 //===----------------------------------------------------------------------===//
+// TransposeMatmulOp
+//===----------------------------------------------------------------------===//
+
+def TransposeMatmulOp : Op<Transform_Dialect,
+    "structured.transpose_matmul",
+    [FunctionalStyleTransformOpTrait, MemoryEffectsOpInterface,
+     TransformOpInterface, TransformEachOpTrait,
+     ReportTrackingListenerFailuresOpTrait]> {
+  let description = [{
+    Convert Linalg matmul ops to transposed variants.
+
+    By default the LHS matrix is transposed. Specify `<rhs>` to instead
+    transpose RHS matrix.
+
+    #### Return modes:
+
+    This operation fails if `target` is unsupported, i.e., not a
+    `linalg.matmul` or `linalg.batch_matmul`. Otherwise, the operation succeeds
+    and returns a handle to the transposed matmul op.
+  }];
+
+  let arguments = (ins
+    TransformHandleTypeInterface:$target,
+    DefaultValuedAttr<TransposeMatmulInput,
+                      "TransposeMatmulInput::lhs">:$inputToTranspose);
+  let results = (outs TransformHandleTypeInterface:$transformed);
+
+  let assemblyFormat = [{
+    $target (`<` $inputToTranspose^ `>`)?
+    attr-dict `:` functional-type($target, results)
+  }];
+
+  let builders = [
+    OpBuilder<(ins "Value":$target)>
+  ];
+
+  let extraClassDeclaration = [{
+    ::mlir::DiagnosedSilenceableFailure applyToOne(
+        ::mlir::transform::TransformRewriter &rewriter,
+        ::mlir::linalg::LinalgOp target,
+        ::mlir::transform::ApplyToEachResultList &results,
+        ::mlir::transform::TransformState &state);
+  }];
+}
+
+//===----------------------------------------------------------------------===//
 // InsertSliceToCopyOp
 //===----------------------------------------------------------------------===//
 
diff --git a/mlir/include/mlir/Dialect/Linalg/Transforms/RuntimeOpVerification.h b/mlir/include/mlir/Dialect/Linalg/Transforms/RuntimeOpVerification.h
new file mode 100644
index 000000000000..6c3643f7835c
--- /dev/null
+++ b/mlir/include/mlir/Dialect/Linalg/Transforms/RuntimeOpVerification.h
@@ -0,0 +1,21 @@
+//===- RuntimeOpVerification.h - Op Verification ----------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef MLIR_DIALECT_LINALG_RUNTIMEOPVERIFICATION_H
+#define MLIR_DIALECT_LINALG_RUNTIMEOPVERIFICATION_H
+
+namespace mlir {
+class DialectRegistry;
+
+namespace linalg {
+void registerRuntimeVerifiableOpInterfaceExternalModels(
+    DialectRegistry &registry);
+} // namespace linalg
+} // namespace mlir
+
+#endif // MLIR_DIALECT_LINALG_RUNTIMEOPVERIFICATION_H
diff --git a/mlir/include/mlir/Dialect/Linalg/Transforms/Transforms.h b/mlir/include/mlir/Dialect/Linalg/Transforms/Transforms.h
index feb3b3f03cf5..5ecf84fa9c70 100644
--- a/mlir/include/mlir/Dialect/Linalg/Transforms/Transforms.h
+++ b/mlir/include/mlir/Dialect/Linalg/Transforms/Transforms.h
@@ -1244,6 +1244,14 @@ FailureOr<Operation *> transposeConv2D(RewriterBase &rewriter,
 FailureOr<Operation *> transposeConv2D(RewriterBase &rewriter,
                                        linalg::Conv2DNhwcFhwcQOp op);
 
+/// Convert Linalg matmul ops to transposed variants.
+FailureOr<Operation *> transposeMatmul(RewriterBase &rewriter,
+                                       linalg::MatmulOp op,
+                                       bool transposeLHS = true);
+FailureOr<Operation *> transposeBatchMatmul(RewriterBase &rewriter,
+                                            linalg::BatchMatmulOp op,
+                                            bool transposeLHS = true);
+
 //===----------------------------------------------------------------------===//
 // Rewrite patterns wrapping transformations.
 // TODO: every single such pattern should be a close to noop wrapper around a
@@ -1616,6 +1624,10 @@ void populateSplitReductionPattern(
     const ControlSplitReductionFn &controlSplitReductionFn,
     bool useAlloc = false);
 
+/// Patterns to convert Linalg matmul ops to transposed variants.
+void populateTransposeMatmulPatterns(RewritePatternSet &patterns,
+                                     bool transposeLHS = true);
+
 } // namespace linalg
 } // namespace mlir
 
diff --git a/mlir/include/mlir/Dialect/OpenMP/OpenMPClauseOperands.h b/mlir/include/mlir/Dialect/OpenMP/OpenMPClauseOperands.h
index 3c5fa23bd4a7..244cee1dd635 100644
--- a/mlir/include/mlir/Dialect/OpenMP/OpenMPClauseOperands.h
+++ b/mlir/include/mlir/Dialect/OpenMP/OpenMPClauseOperands.h
@@ -295,10 +295,9 @@ using TeamsClauseOps =
                     PrivateClauseOps, ReductionClauseOps, ThreadLimitClauseOps>;
 
 using WsloopClauseOps =
-    detail::Clauses<AllocateClauseOps, CollapseClauseOps, LinearClauseOps,
-                    LoopRelatedOps, NowaitClauseOps, OrderClauseOps,
-                    OrderedClauseOps, PrivateClauseOps, ReductionClauseOps,
-                    ScheduleClauseOps>;
+    detail::Clauses<AllocateClauseOps, LinearClauseOps, NowaitClauseOps,
+                    OrderClauseOps, OrderedClauseOps, PrivateClauseOps,
+                    ReductionClauseOps, ScheduleClauseOps>;
 
 } // namespace omp
 } // namespace mlir
diff --git a/mlir/include/mlir/Dialect/OpenMP/OpenMPOps.td b/mlir/include/mlir/Dialect/OpenMP/OpenMPOps.td
index 10771f6e854d..8ab116ce391e 100644
--- a/mlir/include/mlir/Dialect/OpenMP/OpenMPOps.td
+++ b/mlir/include/mlir/Dialect/OpenMP/OpenMPOps.td
@@ -600,29 +600,30 @@ def LoopNestOp : OpenMP_Op<"loop_nest", [SameVariadicOperandSize,
 //===----------------------------------------------------------------------===//
 
 def WsloopOp : OpenMP_Op<"wsloop", [AttrSizedOperandSegments,
-                         AllTypesMatch<["lowerBound", "upperBound", "step"]>,
                          DeclareOpInterfaceMethods<LoopWrapperInterface>,
-                         RecursiveMemoryEffects, ReductionClauseInterface]> {
+                         RecursiveMemoryEffects, ReductionClauseInterface,
+                         SingleBlockImplicitTerminator<"TerminatorOp">]> {
   let summary = "worksharing-loop construct";
   let description = [{
     The worksharing-loop construct specifies that the iterations of the loop(s)
     will be executed in parallel by threads in the current context. These
     iterations are spread across threads that already exist in the enclosing
-    parallel region. The lower and upper bounds specify a half-open range: the
-    range includes the lower bound but does not include the upper bound. If the
-    `inclusive` attribute is specified then the upper bound is also included.
+    parallel region.
 
-    The body region can contain any number of blocks. The region is terminated
-    by "omp.yield" instruction without operands.
+    The body region can only contain a single block which must contain a single
+    operation and a terminator. The operation must be another compatible loop
+    wrapper or an `omp.loop_nest`.
 
     ```
-    omp.wsloop <clauses>
-    for (%i1, %i2) : index = (%c0, %c0) to (%c10, %c10) step (%c1, %c1) {
-      %a = load %arrA[%i1, %i2] : memref<?x?xf32>
-      %b = load %arrB[%i1, %i2] : memref<?x?xf32>
-      %sum = arith.addf %a, %b : f32
-      store %sum, %arrC[%i1, %i2] : memref<?x?xf32>
-      omp.yield
+    omp.wsloop <clauses> {
+      omp.loop_nest (%i1, %i2) : index = (%c0, %c0) to (%c10, %c10) step (%c1, %c1) {
+        %a = load %arrA[%i1, %i2] : memref<?x?xf32>
+        %b = load %arrB[%i1, %i2] : memref<?x?xf32>
+        %sum = arith.addf %a, %b : f32
+        store %sum, %arrC[%i1, %i2] : memref<?x?xf32>
+        omp.yield
+      }
+      omp.terminator
     }
     ```
 
@@ -665,10 +666,7 @@ def WsloopOp : OpenMP_Op<"wsloop", [AttrSizedOperandSegments,
     passed by reference.
   }];
 
-  let arguments = (ins Variadic<IntLikeType>:$lowerBound,
-             Variadic<IntLikeType>:$upperBound,
-             Variadic<IntLikeType>:$step,
-             Variadic<AnyType>:$linear_vars,
+  let arguments = (ins Variadic<AnyType>:$linear_vars,
              Variadic<I32>:$linear_step_vars,
              Variadic<OpenMP_PointerLikeType>:$reduction_vars,
              OptionalAttr<SymbolRefArrayAttr>:$reductions,
@@ -679,22 +677,16 @@ def WsloopOp : OpenMP_Op<"wsloop", [AttrSizedOperandSegments,
              UnitAttr:$nowait,
              UnitAttr:$byref,
              ConfinedAttr<OptionalAttr<I64Attr>, [IntMinValue<0>]>:$ordered_val,
-             OptionalAttr<OrderKindAttr>:$order_val,
-             UnitAttr:$inclusive);
+             OptionalAttr<OrderKindAttr>:$order_val);
 
   let builders = [
-    OpBuilder<(ins "ValueRange":$lowerBound, "ValueRange":$upperBound,
-               "ValueRange":$step,
-               CArg<"ArrayRef<NamedAttribute>", "{}">:$attributes)>,
+    OpBuilder<(ins CArg<"ArrayRef<NamedAttribute>", "{}">:$attributes)>,
     OpBuilder<(ins CArg<"const WsloopClauseOps &">:$clauses)>
   ];
 
   let regions = (region AnyRegion:$region);
 
   let extraClassDeclaration = [{
-    /// Returns the number of loops in the worksharing-loop nest.
-    unsigned getNumLoops() { return getLowerBound().size(); }
-
     /// Returns the number of reduction variables.
     unsigned getNumReductionVars() { return getReductionVars().size(); }
   }];
@@ -711,9 +703,8 @@ def WsloopOp : OpenMP_Op<"wsloop", [AttrSizedOperandSegments,
           |`byref` $byref
           |`ordered` `(` $ordered_val `)`
           |`order` `(` custom<ClauseAttr>($order_val) `)`
-    ) custom<Wsloop>($region, $lowerBound, $upperBound, $step, type($step),
-                     $reduction_vars, type($reduction_vars), $reductions,
-                     $inclusive) attr-dict
+    ) custom<Wsloop>($region, $reduction_vars, type($reduction_vars),
+                     $reductions) attr-dict
   }];
   let hasVerifier = 1;
 }
@@ -732,7 +723,7 @@ def SimdOp : OpenMP_Op<"simd", [AttrSizedOperandSegments,
     transformed into a SIMD loop (that is, multiple iterations of the loop can
     be executed concurrently using SIMD instructions).
 
-    The body region can contain a single block which must contain a single
+    The body region can only contain a single block which must contain a single
     operation and a terminator. The operation must be another compatible loop
     wrapper or an `omp.loop_nest`.
 
@@ -766,6 +757,7 @@ def SimdOp : OpenMP_Op<"simd", [AttrSizedOperandSegments,
         store %sum, %arrC[%i1, %i2] : memref<?x?xf32>
         omp.yield
       }
+      omp.terminator
     }
     ```
   }];
@@ -805,8 +797,8 @@ def SimdOp : OpenMP_Op<"simd", [AttrSizedOperandSegments,
 
 def YieldOp : OpenMP_Op<"yield",
     [Pure, ReturnLike, Terminator,
-     ParentOneOf<["LoopNestOp", "WsloopOp", "DeclareReductionOp",
-     "AtomicUpdateOp", "PrivateClauseOp"]>]> {
+     ParentOneOf<["AtomicUpdateOp", "DeclareReductionOp", "LoopNestOp",
+                  "PrivateClauseOp"]>]> {
   let summary = "loop yield and termination operation";
   let description = [{
     "omp.yield" yields SSA values from the OpenMP dialect op region and
@@ -846,7 +838,7 @@ def DistributeOp : OpenMP_Op<"distribute", [AttrSizedOperandSegments,
     iterations are spread across threads that already exist in the enclosing
     region.
     
-    The body region can contain a single block which must contain a single
+    The body region can only contain a single block which must contain a single
     operation and a terminator. The operation must be another compatible loop
     wrapper or an `omp.loop_nest`.
 
@@ -864,6 +856,7 @@ def DistributeOp : OpenMP_Op<"distribute", [AttrSizedOperandSegments,
         store %sum, %arrC[%i1, %i2] : memref<?x?xf32>
         omp.yield
       }
+      omp.terminator
     }
     ```
     // TODO: private_var, firstprivate_var, lastprivate_var, collapse
@@ -1029,7 +1022,7 @@ def TaskloopOp : OpenMP_Op<"taskloop", [AttrSizedOperandSegments,
     iterations are distributed across tasks generated by the construct and
     scheduled to be executed.
 
-    The body region can contain a single block which must contain a single
+    The body region can only contain a single block which must contain a single
     operation and a terminator. The operation must be another compatible loop
     wrapper or an `omp.loop_nest`.
 
@@ -1042,6 +1035,7 @@ def TaskloopOp : OpenMP_Op<"taskloop", [AttrSizedOperandSegments,
         store %sum, %arrC[%i1, %i2] : memref<?x?xf32>
         omp.yield
       }
+      omp.terminator
     }
     ```
 
diff --git a/mlir/include/mlir/Dialect/OpenMP/OpenMPOpsInterfaces.td b/mlir/include/mlir/Dialect/OpenMP/OpenMPOpsInterfaces.td
index ab9b78e755d9..d9569d9d294d 100644
--- a/mlir/include/mlir/Dialect/OpenMP/OpenMPOpsInterfaces.td
+++ b/mlir/include/mlir/Dialect/OpenMP/OpenMPOpsInterfaces.td
@@ -234,8 +234,8 @@ def OffloadModuleInterface : OpInterface<"OffloadModuleInterface"> {
       /*methodName=*/"getIsTargetDevice",
       (ins), [{}], [{
         if (Attribute isTargetDevice = $_op->getAttr("omp.is_target_device"))
-          if (isTargetDevice.isa<mlir::BoolAttr>())
-           return isTargetDevice.dyn_cast<BoolAttr>().getValue();
+          if (::llvm::isa<mlir::BoolAttr>(isTargetDevice))
+           return ::llvm::dyn_cast<BoolAttr>(isTargetDevice).getValue();
         return false;
       }]>,
     InterfaceMethod<
@@ -259,7 +259,7 @@ def OffloadModuleInterface : OpInterface<"OffloadModuleInterface"> {
       /*methodName=*/"getIsGPU",
       (ins), [{}], [{
         if (Attribute isTargetCGAttr = $_op->getAttr("omp.is_gpu"))
-          if (auto isTargetCGVal = isTargetCGAttr.dyn_cast<BoolAttr>())
+          if (auto isTargetCGVal = ::llvm::dyn_cast<BoolAttr>(isTargetCGAttr))
            return isTargetCGVal.getValue();
         return false;
       }]>,
@@ -332,7 +332,7 @@ def OffloadModuleInterface : OpInterface<"OffloadModuleInterface"> {
       /*methodName=*/"getRequires",
       (ins), [{}], [{
         if (Attribute requiresAttr = $_op->getAttr("omp.requires"))
-          if (auto requiresVal = requiresAttr.dyn_cast<mlir::omp::ClauseRequiresAttr>())
+          if (auto requiresVal = ::llvm::dyn_cast<mlir::omp::ClauseRequiresAttr>(requiresAttr))
             return requiresVal.getValue();
         return mlir::omp::ClauseRequires::none;
       }]>,
diff --git a/mlir/include/mlir/Dialect/Polynomial/IR/Polynomial.h b/mlir/include/mlir/Dialect/Polynomial/IR/Polynomial.h
index 39b05b9d3ad1..3325a6fa3f9f 100644
--- a/mlir/include/mlir/Dialect/Polynomial/IR/Polynomial.h
+++ b/mlir/include/mlir/Dialect/Polynomial/IR/Polynomial.h
@@ -51,9 +51,6 @@ public:
     return (exponent.ult(other.exponent));
   }
 
-  // Prints polynomial to 'os'.
-  void print(raw_ostream &os) const;
-
   friend ::llvm::hash_code hash_value(const Monomial &arg);
 
 public:
@@ -102,6 +99,8 @@ public:
 
   unsigned getDegree() const;
 
+  ArrayRef<Monomial> getTerms() const { return terms; }
+
   friend ::llvm::hash_code hash_value(const Polynomial &arg);
 
 private:
diff --git a/mlir/include/mlir/Dialect/Polynomial/IR/Polynomial.td b/mlir/include/mlir/Dialect/Polynomial/IR/Polynomial.td
index 5d8da8399b01..d3e3ac55677f 100644
--- a/mlir/include/mlir/Dialect/Polynomial/IR/Polynomial.td
+++ b/mlir/include/mlir/Dialect/Polynomial/IR/Polynomial.td
@@ -35,18 +35,18 @@ def Polynomial_Dialect : Dialect {
 
     ```mlir
     // A constant polynomial in a ring with i32 coefficients and no polynomial modulus
-    #ring = #polynomial.ring<ctype=i32>
+    #ring = #polynomial.ring<coefficientType=i32>
     %a = polynomial.constant <1 + x**2 - 3x**3> : polynomial.polynomial<#ring>
 
     // A constant polynomial in a ring with i32 coefficients, modulo (x^1024 + 1)
     #modulus = #polynomial.polynomial<1 + x**1024>
-    #ring = #polynomial.ring<ctype=i32, ideal=#modulus>
+    #ring = #polynomial.ring<coefficientType=i32, polynomialModulus=#modulus>
     %a = polynomial.constant <1 + x**2 - 3x**3> : polynomial.polynomial<#ring>
 
     // A constant polynomial in a ring with i32 coefficients, with a polynomial
     // modulus of (x^1024 + 1) and a coefficient modulus of 17.
     #modulus = #polynomial.polynomial<1 + x**1024>
-    #ring = #polynomial.ring<ctype=i32, cmod=17, ideal=#modulus>
+    #ring = #polynomial.ring<coefficientType=i32, coefficientModulus=17, polynomialModulus=#modulus>
     %a = polynomial.constant <1 + x**2 - 3x**3> : polynomial.polynomial<#ring>
     ```
   }];
@@ -63,7 +63,21 @@ class Polynomial_Attr<string name, string attrMnemonic, list<Trait> traits = []>
 def Polynomial_PolynomialAttr : Polynomial_Attr<"Polynomial", "polynomial"> {
   let summary = "An attribute containing a single-variable polynomial.";
   let description = [{
-     #poly = #polynomial.poly<x**1024 + 1>
+    A polynomial attribute represents a single-variable polynomial, which
+    is used to define the modulus of a `RingAttr`, as well as to define constants
+    and perform constant folding for `polynomial` ops.
+
+    The polynomial must be expressed as a list of monomial terms, with addition
+    or subtraction between them. The choice of variable name is arbitrary, but
+    must be consistent across all the monomials used to define a single
+    attribute. The order of monomial terms is arbitrary, each monomial degree
+    must occur at most once.
+
+    Example:
+
+    ```mlir
+    #poly = #polynomial.polynomial<x**1024 + 1>
+    ```
   }];
   let parameters = (ins "Polynomial":$polynomial);
   let hasCustomAssemblyFormat = 1;
@@ -79,10 +93,10 @@ def Polynomial_RingAttr : Polynomial_Attr<"Ring", "ring"> {
     integral, whose coefficients are taken modulo some statically known modulus
     (`coefficientModulus`).
 
-    Additionally, a polynomial ring can specify an _ideal_, which converts
+    Additionally, a polynomial ring can specify a _polynomialModulus_, which converts
     polynomial arithmetic to the analogue of modular integer arithmetic, where
     each polynomial is represented as its remainder when dividing by the
-    modulus. For single-variable polynomials, an "ideal" is always specificed
+    modulus. For single-variable polynomials, an "polynomialModulus" is always specificed
     via a single polynomial, which we call `polynomialModulus`.
 
     An expressive example is polynomials with i32 coefficients, whose
@@ -122,32 +136,284 @@ class Polynomial_Type<string name, string typeMnemonic>
 
 def Polynomial_PolynomialType : Polynomial_Type<"Polynomial", "polynomial"> {
   let summary = "An element of a polynomial ring.";
-
   let description = [{
     A type for polynomials in a polynomial quotient ring.
   }];
-
   let parameters = (ins Polynomial_RingAttr:$ring);
   let assemblyFormat = "`<` $ring `>`";
 }
 
+def PolynomialLike: TypeOrContainer<Polynomial_PolynomialType, "polynomial-like">;
+
 class Polynomial_Op<string mnemonic, list<Trait> traits = []> :
-    Op<Polynomial_Dialect, mnemonic, traits # [Pure]>;
+    Op<Polynomial_Dialect, mnemonic, traits # [Pure]> {
+  let assemblyFormat = "operands attr-dict `:` functional-type(operands, results)";
+}
 
 class Polynomial_UnaryOp<string mnemonic, list<Trait> traits = []> :
     Polynomial_Op<mnemonic, traits # [SameOperandsAndResultType]> {
   let arguments = (ins Polynomial_PolynomialType:$operand);
   let results = (outs Polynomial_PolynomialType:$result);
-
-  let assemblyFormat = "$operand attr-dict `:` qualified(type($result))";
 }
 
 class Polynomial_BinaryOp<string mnemonic, list<Trait> traits = []> :
-    Polynomial_Op<mnemonic, traits # [SameOperandsAndResultType]> {
-  let arguments = (ins Polynomial_PolynomialType:$lhs, Polynomial_PolynomialType:$rhs);
-  let results = (outs Polynomial_PolynomialType:$result);
+    Polynomial_Op<mnemonic, !listconcat(traits, [Pure, SameOperandsAndResultType, ElementwiseMappable])> {
+  let arguments = (ins PolynomialLike:$lhs, PolynomialLike:$rhs);
+  let results = (outs PolynomialLike:$result);
+  let assemblyFormat = "operands attr-dict `:` type($result)";
+}
+
+def Polynomial_AddOp : Polynomial_BinaryOp<"add", [Commutative]> {
+  let summary = "Addition operation between polynomials.";
+  let description = [{
+    Performs polynomial addition on the operands. The operands may be single
+    polynomials or containers of identically-typed polynomials, i.e., polynomials
+    from the same underlying ring with the same coefficient types.
+
+    Addition is defined to occur in the ring defined by the ring attribute of
+    the two operands, meaning the addition is taken modulo the coefficientModulus
+    and the polynomialModulus of the ring.
+
+    Example:
+
+    ```mlir
+    // add two polynomials modulo x^1024 - 1
+    #poly = #polynomial.polynomial<x**1024 - 1>
+    #ring = #polynomial.ring<coefficientType=i32, coefficientModulus=65536, polynomialModulus=#poly>
+    %0 = polynomial.constant #polynomial.polynomial<1 + x**2> : !polynomial.polynomial<#ring>
+    %1 = polynomial.constant #polynomial.polynomial<x**5 - x + 1> : !polynomial.polynomial<#ring>
+    %2 = polynomial.add %0, %1 : !polynomial.polynomial<#ring>
+    ```
+  }];
+}
+
+def Polynomial_SubOp : Polynomial_BinaryOp<"sub"> {
+  let summary = "Subtraction operation between polynomials.";
+  let description = [{
+    Performs polynomial subtraction on the operands. The operands may be single
+    polynomials or containers of identically-typed polynomials, i.e., polynomials
+    from the same underlying ring with the same coefficient types.
+
+    Subtraction is defined to occur in the ring defined by the ring attribute of
+    the two operands, meaning the subtraction is taken modulo the coefficientModulus
+    and the polynomialModulus of the ring.
+
+    Example:
 
-  let assemblyFormat = "$lhs `,` $rhs attr-dict `:` qualified(type($result))";
+    ```mlir
+    // subtract two polynomials modulo x^1024 - 1
+    #poly = #polynomial.polynomial<x**1024 - 1>
+    #ring = #polynomial.ring<coefficientType=i32, coefficientModulus=65536, polynomialModulus=#poly>
+    %0 = polynomial.constant #polynomial.polynomial<1 + x**2> : !polynomial.polynomial<#ring>
+    %1 = polynomial.constant #polynomial.polynomial<x**5 - x + 1> : !polynomial.polynomial<#ring>
+    %2 = polynomial.sub %0, %1 : !polynomial.polynomial<#ring>
+    ```
+  }];
+}
+
+def Polynomial_MulOp : Polynomial_BinaryOp<"mul", [Commutative]> {
+  let summary = "Multiplication operation between polynomials.";
+  let description = [{
+    Performs polynomial multiplication on the operands. The operands may be single
+    polynomials or containers of identically-typed polynomials, i.e., polynomials
+    from the same underlying ring with the same coefficient types.
+
+    Multiplication is defined to occur in the ring defined by the ring attribute of
+    the two operands, meaning the multiplication is taken modulo the coefficientModulus
+    and the polynomialModulus of the ring.
+
+    Example:
+
+    ```mlir
+    // multiply two polynomials modulo x^1024 - 1
+    #poly = #polynomial.polynomial<x**1024 - 1>
+    #ring = #polynomial.ring<coefficientType=i32, coefficientModulus=65536, polynomialModulus=#poly>
+    %0 = polynomial.constant #polynomial.polynomial<1 + x**2> : !polynomial.polynomial<#ring>
+    %1 = polynomial.constant #polynomial.polynomial<x**5 - x + 1> : !polynomial.polynomial<#ring>
+    %2 = polynomial.mul %0, %1 : !polynomial.polynomial<#ring>
+    ```
+  }];
+}
+
+def Polynomial_MulScalarOp : Polynomial_Op<"mul_scalar", [
+      ElementwiseMappable, AllTypesMatch<["polynomial", "output"]>]> {
+  let summary = "Multiplication by a scalar of the field.";
+  let description = [{
+    Multiplies the polynomial operand's coefficients by a given scalar value.
+    The operation is defined to occur in the ring defined by the ring attribute
+    of the two operands, meaning the multiplication is taken modulo the
+    coefficientModulus of the ring.
+
+    The `scalar` input must have the same type as the polynomial ring's
+    coefficientType.
+
+    Example:
+
+    ```mlir
+    // multiply two polynomials modulo x^1024 - 1
+    #poly = #polynomial.polynomial<x**1024 - 1>
+    #ring = #polynomial.ring<coefficientType=i32, coefficientModulus=65536, polynomialModulus=#poly>
+    %0 = polynomial.constant #polynomial.polynomial<1 + x**2> : !polynomial.polynomial<#ring>
+    %1 = arith.constant 3 : i32
+    %2 = polynomial.mul_scalar %0, %1 : !polynomial.polynomial<#ring>, i32
+    ```
+  }];
+
+  let arguments = (ins
+    PolynomialLike:$polynomial,
+    AnyInteger:$scalar
+  );
+  let results = (outs
+    PolynomialLike:$output
+  );
+  let assemblyFormat = "operands attr-dict `:` type($polynomial) `,` type($scalar)";
+  let hasVerifier = 1;
+}
+
+def Polynomial_LeadingTermOp: Polynomial_Op<"leading_term"> {
+  let summary = "Compute the leading term of the polynomial.";
+  let description = [{
+    The degree of a polynomial is the largest $k$ for which the coefficient
+    `a_k` of `x^k` is nonzero. The leading term is the term `a_k * x^k`, which
+    this op represents as a pair of results. The first is the degree `k` as an
+    index, and the second is the coefficient, whose type matches the
+    coefficient type of the polynomial's ring attribute.
+
+    Example:
+
+    ```mlir
+    #poly = #polynomial.polynomial<x**1024 - 1>
+    #ring = #polynomial.ring<coefficientType=i32, coefficientModulus=65536, polynomialModulus=#poly>
+    %0 = polynomial.constant #polynomial.polynomial<1 + x**2> : !polynomial.polynomial<#ring>
+    %1, %2 = polynomial.leading_term %0 : !polynomial.polynomial<#ring> -> (index, i32)
+    ```
+  }];
+  let arguments = (ins Polynomial_PolynomialType:$input);
+  let results = (outs Index:$degree, AnyInteger:$coefficient);
+  let assemblyFormat = "operands attr-dict `:` type($input) `->` `(` type($degree) `,` type($coefficient) `)`";
+}
+
+def Polynomial_MonomialOp: Polynomial_Op<"monomial"> {
+  let summary = "Create a polynomial that consists of a single monomial.";
+  let description = [{
+    Construct a polynomial that consists of a single monomial term, from its
+    degree and coefficient as dynamic inputs.
+
+    The coefficient type of the output polynomial's ring attribute must match
+    the `coefficient` input type.
+
+    Example:
+
+    ```mlir
+    #poly = #polynomial.polynomial<x**1024 - 1>
+    #ring = #polynomial.ring<coefficientType=i32, coefficientModulus=65536, polynomialModulus=#poly>
+    %deg = arith.constant 1023 : index
+    %five = arith.constant 5 : i32
+    %0 = polynomial.monomial %five, %deg : (i32, index) -> !polynomial.polynomial<#ring>
+    ```
+  }];
+  let arguments = (ins AnyInteger:$coefficient, Index:$degree);
+  let results = (outs Polynomial_PolynomialType:$output);
+}
+
+def Polynomial_MonicMonomialMulOp: Polynomial_Op<"monic_monomial_mul", [AllTypesMatch<["input", "output"]>]> {
+  let summary = "Multiply a polynomial by a monic monomial.";
+  let description = [{
+    Multiply a polynomial by a monic monomial, meaning a polynomial of the form
+    `1 * x^k` for an index operand `k`.
+
+    In some special rings of polynomials, such as a ring of polynomials
+    modulo `x^n - 1`, `monomial_mul` can be interpreted as a cyclic shift of
+    the coefficients of the polynomial. For some rings, this results in
+    optimized lowerings that involve rotations and rescaling of the
+    coefficients of the input.
+  }];
+  let arguments = (ins PolynomialLike:$input, Index:$monomialDegree);
+  let results = (outs PolynomialLike:$output);
+}
+
+def Polynomial_FromTensorOp : Polynomial_Op<"from_tensor", [Pure]> {
+  let summary = "Creates a polynomial from integer coefficients stored in a tensor.";
+  let description = [{
+    `polynomial.from_tensor` creates a polynomial value from a tensor of coefficients.
+    The input tensor must list the coefficients in degree-increasing order.
+
+    The input one-dimensional tensor may have size at most the degree of the
+    ring's polynomialModulus generator polynomial, with smaller dimension implying that
+    all higher-degree terms have coefficient zero.
+
+    Example:
+
+    ```mlir
+    #poly = #polynomial.polynomial<x**1024 - 1>
+    #ring = #polynomial.ring<coefficientType=i32, coefficientModulus=65536, polynomialModulus=#poly>
+    %two = arith.constant 2 : i32
+    %five = arith.constant 5 : i32
+    %coeffs = tensor.from_elements %two, %two, %five : tensor<3xi32>
+    %poly = polynomial.from_tensor %coeffs : tensor<3xi32> -> !polynomial.polynomial<#ring>
+    ```
+  }];
+  let arguments = (ins RankedTensorOf<[AnyInteger]>:$input);
+  let results = (outs Polynomial_PolynomialType:$output);
+
+  let assemblyFormat = "$input attr-dict `:` type($input) `->` type($output)";
+
+  let builders = [
+    // Builder that infers coefficient modulus from tensor bit width,
+    // and uses whatever input ring is provided by the caller.
+    OpBuilder<(ins "::mlir::Value":$input, "::mlir::polynomial::RingAttr":$ring)>
+  ];
+  let hasVerifier = 1;
+}
+
+def Polynomial_ToTensorOp : Polynomial_Op<"to_tensor", [Pure]> {
+  let summary = "Creates a tensor containing the coefficients of a polynomial.";
+  let description = [{
+    `polynomial.to_tensor` creates a dense tensor value containing the
+    coefficients of the input polynomial. The output tensor contains the
+    coefficients in degree-increasing order.
+
+    Operations that act on the coefficients of a polynomial, such as extracting
+    a specific coefficient or extracting a range of coefficients, should be
+    implemented by composing `to_tensor` with the relevant `tensor` dialect
+    ops.
+
+    The output tensor has shape equal to the degree of the polynomial ring
+    attribute's polynomialModulus, including zeroes.
+
+    Example:
+
+    ```mlir
+    #poly = #polynomial.polynomial<x**1024 - 1>
+    #ring = #polynomial.ring<coefficientType=i32, coefficientModulus=65536, polynomialModulus=#poly>
+    %two = arith.constant 2 : i32
+    %five = arith.constant 5 : i32
+    %coeffs = tensor.from_elements %two, %two, %five : tensor<3xi32>
+    %poly = polynomial.from_tensor %coeffs : tensor<3xi32> -> !polynomial.polynomial<#ring>
+    %tensor = polynomial.to_tensor %poly : !polynomial.polynomial<#ring> -> tensor<1024xi32>
+    ```
+  }];
+  let arguments = (ins Polynomial_PolynomialType:$input);
+  let results = (outs RankedTensorOf<[AnyInteger]>:$output);
+  let assemblyFormat = "$input attr-dict `:` type($input) `->` type($output)";
+
+  let hasVerifier = 1;
+}
+
+def Polynomial_ConstantOp : Polynomial_Op<"constant", [Pure]> {
+  let summary = "Define a constant polynomial via an attribute.";
+  let description = [{
+    Example:
+
+    ```mlir
+    #poly = #polynomial.polynomial<x**1024 - 1>
+    #ring = #polynomial.ring<coefficientType=i32, coefficientModulus=65536, polynomialModulus=#poly>
+    %0 = polynomial.constant #polynomial.polynomial<1 + x**2> : !polynomial.polynomial<#ring>
+    ```
+  }];
+  let arguments = (ins Polynomial_PolynomialAttr:$input);
+  let results = (outs Polynomial_PolynomialType:$output);
+  let assemblyFormat = "$input attr-dict `:` type($output)";
 }
 
 #endif // POLYNOMIAL_OPS
diff --git a/mlir/include/mlir/Dialect/SCF/Transforms/Passes.h b/mlir/include/mlir/Dialect/SCF/Transforms/Passes.h
index 90b315e83a8c..31c3d0eb629d 100644
--- a/mlir/include/mlir/Dialect/SCF/Transforms/Passes.h
+++ b/mlir/include/mlir/Dialect/SCF/Transforms/Passes.h
@@ -59,6 +59,9 @@ createParallelLoopTilingPass(llvm::ArrayRef<int64_t> tileSize = {},
 /// loop range.
 std::unique_ptr<Pass> createForLoopRangeFoldingPass();
 
+/// Creates a pass that converts SCF forall loops to SCF for loops.
+std::unique_ptr<Pass> createForallToForLoopPass();
+
 // Creates a pass which lowers for loops into while loops.
 std::unique_ptr<Pass> createForToWhileLoopPass();
 
diff --git a/mlir/include/mlir/Dialect/SCF/Transforms/Passes.td b/mlir/include/mlir/Dialect/SCF/Transforms/Passes.td
index 350611ad8687..a7aeb42d60c0 100644
--- a/mlir/include/mlir/Dialect/SCF/Transforms/Passes.td
+++ b/mlir/include/mlir/Dialect/SCF/Transforms/Passes.td
@@ -120,6 +120,11 @@ def SCFForLoopRangeFolding : Pass<"scf-for-loop-range-folding"> {
   let constructor = "mlir::createForLoopRangeFoldingPass()";
 }
 
+def SCFForallToForLoop : Pass<"scf-forall-to-for"> {
+  let summary = "Convert SCF forall loops to SCF for loops";
+  let constructor = "mlir::createForallToForLoopPass()";
+}
+
 def SCFForToWhileLoop : Pass<"scf-for-to-while"> {
   let summary = "Convert SCF for loops to SCF while loops";
   let constructor = "mlir::createForToWhileLoopPass()";
diff --git a/mlir/include/mlir/Dialect/SCF/Transforms/Transforms.h b/mlir/include/mlir/Dialect/SCF/Transforms/Transforms.h
index 220dcb35571d..b063e6e775e6 100644
--- a/mlir/include/mlir/Dialect/SCF/Transforms/Transforms.h
+++ b/mlir/include/mlir/Dialect/SCF/Transforms/Transforms.h
@@ -28,10 +28,17 @@ class Value;
 namespace scf {
 
 class IfOp;
+class ForallOp;
 class ForOp;
 class ParallelOp;
 class WhileOp;
 
+/// Try converting scf.forall into a set of nested scf.for loops.
+/// The newly created scf.for ops will be returned through the `results`
+/// vector if provided.
+LogicalResult forallToForLoop(RewriterBase &rewriter, ForallOp forallOp,
+                              SmallVectorImpl<Operation *> *results = nullptr);
+
 /// Fuses all adjacent scf.parallel operations with identical bounds and step
 /// into one scf.parallel operations. Uses a naive aliasing and dependency
 /// analysis.
diff --git a/mlir/include/mlir/Dialect/SparseTensor/IR/SparseTensor.h b/mlir/include/mlir/Dialect/SparseTensor/IR/SparseTensor.h
index 5e523ec428ae..b182b4c72b95 100644
--- a/mlir/include/mlir/Dialect/SparseTensor/IR/SparseTensor.h
+++ b/mlir/include/mlir/Dialect/SparseTensor/IR/SparseTensor.h
@@ -89,18 +89,21 @@ inline MemRefType getMemRefType(T &&t) {
 /// Returns null-attribute for any type without an encoding.
 SparseTensorEncodingAttr getSparseTensorEncoding(Type type);
 
+/// Returns true iff the type range has any sparse tensor type.
+inline bool hasAnySparseType(TypeRange types) {
+  return llvm::any_of(types, [](Type type) {
+    return getSparseTensorEncoding(type) != nullptr;
+  });
+}
+
 /// Returns true iff MLIR operand has any sparse operand.
 inline bool hasAnySparseOperand(Operation *op) {
-  return llvm::any_of(op->getOperands().getTypes(), [](Type t) {
-    return getSparseTensorEncoding(t) != nullptr;
-  });
+  return hasAnySparseType(op->getOperands().getTypes());
 }
 
 /// Returns true iff MLIR operand has any sparse result.
 inline bool hasAnySparseResult(Operation *op) {
-  return llvm::any_of(op->getResults().getTypes(), [](Type t) {
-    return getSparseTensorEncoding(t) != nullptr;
-  });
+  return hasAnySparseType(op->getResults().getTypes());
 }
 
 /// Returns true iff MLIR operand has any sparse operand or result.
diff --git a/mlir/include/mlir/Dialect/SparseTensor/IR/SparseTensorAttrDefs.td b/mlir/include/mlir/Dialect/SparseTensor/IR/SparseTensorAttrDefs.td
index 4a9b9169ae4b..eefa4c71bbd2 100644
--- a/mlir/include/mlir/Dialect/SparseTensor/IR/SparseTensorAttrDefs.td
+++ b/mlir/include/mlir/Dialect/SparseTensor/IR/SparseTensorAttrDefs.td
@@ -167,7 +167,7 @@ def SparseTensorEncodingAttr : SparseTensor_Attr<"SparseTensorEncoding",
     - **soa** : only applicable to singleton levels, fuses the singleton
       level in SoA (structure of arrays) scheme.
 
-    In addition to the map, the following two fields are optional:
+    In addition to the map, the following fields are optional:
 
     - The required bitwidth for position storage (integral offsets
       into the sparse storage scheme).  A narrow width reduces the memory
@@ -183,6 +183,23 @@ def SparseTensorEncodingAttr : SparseTensor_Attr<"SparseTensorEncoding",
       coordinate over all levels).  The choices are `8`, `16`, `32`,
       `64`, or, the default, `0` to indicate a native bitwidth.
 
+    - The explicit value for the sparse tensor. If explicitVal is set,
+      then all the non-zero values in the tensor have the same explicit value.
+      The default value Attribute() indicates that it is not set. This
+      is useful for binary-valued sparse tensors whose values can either
+      be an implicit value (0 by default) or an explicit value (such as 1).
+      In this approach, we don't store explicit/implicit values, and metadata
+      (such as position and coordinate arrays) alone fully defines the original tensor.
+      This yields additional savings for the storage requirements,
+      as well as for the computational time, since we skip operating on
+      implicit values and can constant fold the explicit values where they are used.
+
+    - The implicit value for the sparse tensor. If implicitVal is set,
+      then the "zero" value in the tensor is equal to the implicit value.
+      For now, we only support `0` as the implicit value but it could be
+      extended in the future. The default value Attribute() indicates that
+      the implicit value is `0` (same type as the tensor element type).
+
     Examples:
 
     ```mlir
@@ -226,6 +243,15 @@ def SparseTensorEncodingAttr : SparseTensor_Attr<"SparseTensorEncoding",
     }>
     ... tensor<8x8xf64, #DCSC> ...
 
+    // Doubly compressed sparse column storage with specific
+    // explicit and implicit values.
+    #DCSC = #sparse_tensor.encoding<{
+      map = (i, j) -> (j : compressed, i : compressed),
+      explicitVal = 1 : i64,
+      implicitVal = 0 : i64
+    }>
+    ... tensor<8x8xi64, #DCSC> ...
+
     // Block sparse row storage (2x3 blocks).
     #BSR = #sparse_tensor.encoding<{
       map = ( i, j ) ->
@@ -307,6 +333,12 @@ def SparseTensorEncodingAttr : SparseTensor_Attr<"SparseTensorEncoding",
     // The required bitwidth for coordinate storage.
     "unsigned":$crdWidth,
 
+    // The required explicit value.
+    "::mlir::Attribute":$explicitVal,
+
+    // The required implicit value.
+    "::mlir::Attribute":$implicitVal,
+
     // A slice attribute for each dimension of the tensor type.
     ArrayRefParameter<
       "::mlir::sparse_tensor::SparseTensorDimSliceAttr",
@@ -319,7 +351,9 @@ def SparseTensorEncodingAttr : SparseTensor_Attr<"SparseTensorEncoding",
                      CArg<"AffineMap", "{}">:$dimToLvl,
                      CArg<"AffineMap", "{}">:$lvlToDim,
                      CArg<"unsigned", "0">:$posWidth,
-                     CArg<"unsigned", "0">:$crdWidth), [{
+                     CArg<"unsigned", "0">:$crdWidth,
+                     CArg<"::mlir::Attribute", "{}">:$explicitVal,
+                     CArg<"::mlir::Attribute", "{}">:$implicitVal), [{
       if (!dimToLvl) {
         dimToLvl = ::mlir::AffineMap::getMultiDimIdentityMap(lvlTypes.size(), $_ctxt);
       }
@@ -327,6 +361,7 @@ def SparseTensorEncodingAttr : SparseTensor_Attr<"SparseTensorEncoding",
         lvlToDim = ::mlir::sparse_tensor::inferLvlToDim(dimToLvl, $_ctxt);
       }
       return $_get($_ctxt, lvlTypes, dimToLvl, lvlToDim, posWidth, crdWidth,
+        explicitVal, implicitVal,
         ArrayRef<::mlir::sparse_tensor::SparseTensorDimSliceAttr>{});
     }]>
   ];
@@ -353,6 +388,22 @@ def SparseTensorEncodingAttr : SparseTensor_Attr<"SparseTensorEncoding",
     /// reset to the default, and all other fields inherited from `this`.
     SparseTensorEncodingAttr withoutBitWidths() const;
 
+    /// Constructs a new encoding with the given explicit value
+    /// and all other fields inherited from `this`.
+    SparseTensorEncodingAttr withExplicitVal(Attribute explicitVal) const;
+
+    /// Constructs a new encoding with the explicit value
+    /// reset to the default, and all other fields inherited from `this`.
+    SparseTensorEncodingAttr withoutExplicitVal() const;
+
+    /// Constructs a new encoding with the given implicit value
+    /// and all other fields inherited from `this`.
+    SparseTensorEncodingAttr withImplicitVal(Attribute implicitVal) const;
+
+    /// Constructs a new encoding with the implicit value
+    /// reset to the default, and all other fields inherited from `this`.
+    SparseTensorEncodingAttr withoutImplicitVal() const;
+
     /// Constructs a new encoding with the given dimSlices, and all
     /// other fields inherited from `this`.
     SparseTensorEncodingAttr withDimSlices(ArrayRef<::mlir::sparse_tensor::SparseTensorDimSliceAttr> dimSlices) const;
diff --git a/mlir/include/mlir/Dialect/SparseTensor/IR/SparseTensorType.h b/mlir/include/mlir/Dialect/SparseTensor/IR/SparseTensorType.h
index 825d89a408fe..34d99913fbd5 100644
--- a/mlir/include/mlir/Dialect/SparseTensor/IR/SparseTensorType.h
+++ b/mlir/include/mlir/Dialect/SparseTensor/IR/SparseTensorType.h
@@ -115,6 +115,22 @@ public:
     return withEncoding(enc.withoutBitWidths());
   }
 
+  SparseTensorType withExplicitVal(Attribute explicitVal) const {
+    return withEncoding(enc.withExplicitVal(explicitVal));
+  }
+
+  SparseTensorType withoutExplicitVal() const {
+    return withEncoding(enc.withoutExplicitVal());
+  }
+
+  SparseTensorType withImplicitVal(Attribute implicitVal) const {
+    return withEncoding(enc.withImplicitVal(implicitVal));
+  }
+
+  SparseTensorType withoutImplicitVal() const {
+    return withEncoding(enc.withoutImplicitVal());
+  }
+
   SparseTensorType
   withDimSlices(ArrayRef<SparseTensorDimSliceAttr> dimSlices) const {
     return withEncoding(enc.withDimSlices(dimSlices));
@@ -327,6 +343,12 @@ public:
   /// Returns the position-overhead bitwidth, defaulting to zero.
   unsigned getPosWidth() const { return enc ? enc.getPosWidth() : 0; }
 
+  /// Returns the explicit value, defaulting to null Attribute for unset.
+  Attribute getExplicitVal() const { return enc.getExplicitVal(); }
+
+  /// Returns the implicit value, defaulting to null Attribute for 0.
+  Attribute getImplicitVal() const { return enc.getImplicitVal(); }
+
   /// Returns the coordinate-overhead MLIR type, defaulting to `IndexType`.
   Type getCrdType() const { return enc.getCrdElemType(); }
 
diff --git a/mlir/include/mlir/Dialect/Vector/Utils/VectorUtils.h b/mlir/include/mlir/Dialect/Vector/Utils/VectorUtils.h
index 8a57c6094c41..030be328e97f 100644
--- a/mlir/include/mlir/Dialect/Vector/Utils/VectorUtils.h
+++ b/mlir/include/mlir/Dialect/Vector/Utils/VectorUtils.h
@@ -194,7 +194,7 @@ bool isLinearizableVector(VectorType type);
 /// for each dimension of the passed in tensor.
 Value createReadOrMaskedRead(OpBuilder &builder, Location loc, Value source,
                              ArrayRef<int64_t> readShape, Value padValue,
-                             bool useInBoundsInsteadOfMasking = true);
+                             bool useInBoundsInsteadOfMasking);
 
 /// Returns success if `inputVectorSizes` is a valid masking configuraion for
 /// given `shape`, i.e., it meets:
diff --git a/mlir/include/mlir/Dialect/XeGPU/IR/CMakeLists.txt b/mlir/include/mlir/Dialect/XeGPU/IR/CMakeLists.txt
index f1740e9ed929..3f8cac4dc07c 100644
--- a/mlir/include/mlir/Dialect/XeGPU/IR/CMakeLists.txt
+++ b/mlir/include/mlir/Dialect/XeGPU/IR/CMakeLists.txt
@@ -2,12 +2,12 @@ add_mlir_dialect(XeGPU xegpu)
 add_mlir_doc(XeGPU XeGPU Dialects/ -gen-dialect-doc -dialect=xegpu)
 
 set(LLVM_TARGET_DEFINITIONS XeGPU.td)
-mlir_tablegen(XeGPUAttrs.h.inc -gen-attrdef-decls)
-mlir_tablegen(XeGPUAttrs.cpp.inc -gen-attrdef-defs)
+mlir_tablegen(XeGPUAttrs.h.inc -gen-attrdef-decls -attrdefs-dialect=xegpu)
+mlir_tablegen(XeGPUAttrs.cpp.inc -gen-attrdef-defs -attrdefs-dialect=xegpu)
 add_public_tablegen_target(MLIRXeGPUAttrsIncGen)
 add_dependencies(mlir-headers MLIRXeGPUAttrsIncGen)
 
-set(LLVM_TARGET_DEFINITIONS XeGPU.td)
+set(LLVM_TARGET_DEFINITIONS XeGPUAttrs.td)
 mlir_tablegen(XeGPUEnums.h.inc -gen-enum-decls)
 mlir_tablegen(XeGPUEnums.cpp.inc -gen-enum-defs)
 add_public_tablegen_target(MLIRXeGPUEnumsIncGen)
diff --git a/mlir/include/mlir/Dialect/XeGPU/IR/XeGPU.h b/mlir/include/mlir/Dialect/XeGPU/IR/XeGPU.h
index eca9255ff397..7ac0cf77fe59 100644
--- a/mlir/include/mlir/Dialect/XeGPU/IR/XeGPU.h
+++ b/mlir/include/mlir/Dialect/XeGPU/IR/XeGPU.h
@@ -10,6 +10,7 @@
 #define MLIR_DIALECT_XEGPU_IR_XEGPU_H
 
 #include "mlir/Bytecode/BytecodeOpInterface.h"
+#include "mlir/Dialect/Arith/IR/Arith.h"
 #include "mlir/IR/BuiltinTypes.h"
 #include "mlir/IR/Dialect.h"
 #include "mlir/IR/TypeUtilities.h"
@@ -19,7 +20,7 @@
 
 namespace mlir {
 namespace xegpu {
-// placeholder
+class TensorDescType;
 } // namespace xegpu
 } // namespace mlir
 
diff --git a/mlir/include/mlir/Dialect/XeGPU/IR/XeGPUAttrs.td b/mlir/include/mlir/Dialect/XeGPU/IR/XeGPUAttrs.td
index 6579d07ec262..f3ca09a6a68e 100644
--- a/mlir/include/mlir/Dialect/XeGPU/IR/XeGPUAttrs.td
+++ b/mlir/include/mlir/Dialect/XeGPU/IR/XeGPUAttrs.td
@@ -10,6 +10,7 @@
 #define MLIR_DIALECT_XEGPU_IR_XEGPUATTRS_TD
 
 include "mlir/Dialect/XeGPU/IR/XeGPUDialect.td"
+include "mlir/IR/AttrTypeBase.td"
 include "mlir/IR/EnumAttr.td"
 
 class XeGPUAttr<string name, string attrMnemonic, list<Trait> traits = [],
@@ -98,4 +99,21 @@ def XeGPU_CacheHintAttr
     let assemblyFormat = "`<` $value `>`";
 }
 
+def XeGPU_FenceScopeWorkgroup: I32EnumAttrCase<"Workgroup", 0, "workgroup">;
+def XeGPU_FenceScopeGPU: I32EnumAttrCase<"GPU", 1, "gpu">;
+def XeGPU_FenceScope: I32EnumAttr<"FenceScope",
+      "The enumeration for the scope of fence operation.",
+      [XeGPU_FenceScopeWorkgroup, XeGPU_FenceScopeGPU]> {
+  let genSpecializedAttr = 0;
+  let cppNamespace = "::mlir::xegpu";
+}
+
+def XeGPU_FenceScopeAttr:
+  EnumAttr<XeGPU_Dialect, XeGPU_FenceScope, "fence_scope"> {
+    let summary = [{Describes the scope of fence.
+                    "workgroup" means that the scope is within each work group.
+                    "gpu" means the scope is across work groups within the gpu.}];
+    let assemblyFormat = "$value";
+}
+
 #endif // MLIR_DIALECT_XEGPU_IR_XEGPUATTRS_TD
\ No newline at end of file
diff --git a/mlir/include/mlir/Dialect/XeGPU/IR/XeGPUDialect.td b/mlir/include/mlir/Dialect/XeGPU/IR/XeGPUDialect.td
index c2f09319c790..765f218f95d2 100644
--- a/mlir/include/mlir/Dialect/XeGPU/IR/XeGPUDialect.td
+++ b/mlir/include/mlir/Dialect/XeGPU/IR/XeGPUDialect.td
@@ -17,12 +17,14 @@ def XeGPU_Dialect : Dialect {
     let summary = "The XeGPU dialect that models Intel GPU's ISA";
     let description = [{
       The XeGPU dialect models Intel Xe ISA semantics but works at vector and
-      TensorDesc data type. It provides 1:1 mappings to match Xe instructions 
+      TensorDesc data type. It provides 1:1 mappings to match Xe instructions
       like DPAS and 2D block load. The matrix size being processed at this level
       exactly matches the hardware instructions or the intrinsic supported by
       the lower-level GPU compiler.
     }];
 
+    let dependentDialects = ["arith::ArithDialect"];
+
     let useDefaultTypePrinterParser = true;
     let useDefaultAttributePrinterParser = true;
 }
diff --git a/mlir/include/mlir/Dialect/XeGPU/IR/XeGPUOps.td b/mlir/include/mlir/Dialect/XeGPU/IR/XeGPUOps.td
index c6f7f83441b9..e477d9a0ca3f 100644
--- a/mlir/include/mlir/Dialect/XeGPU/IR/XeGPUOps.td
+++ b/mlir/include/mlir/Dialect/XeGPU/IR/XeGPUOps.td
@@ -9,7 +9,7 @@
 #ifndef MLIR_DIALECT_XEGPU_IR_XEGPUOPS_TD
 #define MLIR_DIALECT_XEGPU_IR_XEGPUOPS_TD
 
-include "mlir/IR/AttrTypeBase.td"
+include "mlir/Dialect/Arith/IR/ArithBase.td"
 include "mlir/Dialect/XeGPU/IR/XeGPUAttrs.td"
 include "mlir/Dialect/XeGPU/IR/XeGPUDialect.td"
 include "mlir/Dialect/XeGPU/IR/XeGPUTypes.td"
@@ -36,7 +36,7 @@ class XeGPU_Op<string mnemonic, list<Trait> traits = []>:
 
     static ::mlir::ParseResult parseProperties(::mlir::OpAsmParser &parser,
                                      ::mlir::OperationState &result) {
-      if (mlir::succeeded(parser.parseLess())) {
+      if (mlir::succeeded(parser.parseOptionalLess())) {
         if (parser.parseAttribute(result.propertiesAttr) || parser.parseGreater())
           return failure();
       }
@@ -164,10 +164,10 @@ def XeGPU_CreateNdDescOp: XeGPU_Op<"create_nd_tdesc", [Pure, ViewLikeOpInterface
     /// source operand. They overide static shape from source memref type.
     ArrayRef<int64_t> getStaticSizes() {
       auto attr = getConstShapeAttr();
-      if (getSourceType().isa<IntegerType>() || attr)
+      if (llvm::isa<IntegerType>(getSourceType()) || attr)
         return attr;
 
-      auto memrefType = getSourceType().dyn_cast<MemRefType>();
+      auto memrefType = llvm::dyn_cast<MemRefType>(getSourceType());
       assert(memrefType && "Incorrect use of getStaticSizes");
       return memrefType.getShape();
     }
@@ -179,10 +179,10 @@ def XeGPU_CreateNdDescOp: XeGPU_Op<"create_nd_tdesc", [Pure, ViewLikeOpInterface
     /// source operand. They overide static strides from source memref type.
     ArrayRef<int64_t> getStaticStrides() {
       auto attr = getConstStridesAttr();
-      if (getSourceType().isa<IntegerType>() || attr)
+      if (llvm::isa<IntegerType>(getSourceType()) || attr)
         return attr;
 
-      auto memrefType = getSourceType().dyn_cast<MemRefType>();
+      auto memrefType = llvm::dyn_cast<MemRefType>(getSourceType());
       assert(memrefType && "Incorrect use of getStaticStrides");
       auto [strides, offset] = getStridesAndOffset(memrefType);
       // reuse the storage of ConstStridesAttr since strides from
@@ -196,7 +196,7 @@ def XeGPU_CreateNdDescOp: XeGPU_Op<"create_nd_tdesc", [Pure, ViewLikeOpInterface
     /// `static_shape` and `static_strides` attributes.
     std::array<unsigned, 3> getArrayAttrMaxRanks() {
       unsigned rank;
-      if (auto ty = getSourceType().dyn_cast<MemRefType>()) {
+      if (auto ty = llvm::dyn_cast<MemRefType>(getSourceType())) {
         rank = ty.getRank();
       } else {
         rank = (unsigned)getMixedOffsets().size();
@@ -254,7 +254,7 @@ def XeGPU_LoadNdOp : XeGPU_Op<"load_nd", [AllElementTypesMatch<["value", "Tensor
     a block of data from memory to register. It takes a set of optional cache
     hints for each level of cache, L1, L2 and L3. If hardware does not have a
     correspoding cache, Corresponding cache hint attribute will be masked.
-    vnni transform is an hardware feature for Intel GPU, which is used to
+    VNNI transformation is an hardware feature for Intel GPU, which is used to
     do data packing during the load for B operand of matrix operation, if
     the bit width of the data type is less then 32 bits, e.g., fp16. And
     transpose is another Intel hardware feature, which will do transpose
@@ -425,10 +425,6 @@ def XeGPU_CreateDescOp: XeGPU_Op<"create_tdesc", [Pure, ViewLikeOpInterface]> {
     %0 = memref.alloc() : memref<1024xf32>
     %1 = xegpu.create_tdesc %0[0, 4, 8, 12] {chunk_size = 8}: memref<1024xf32> -> TensorDesc<4x8xf32>
     ```
-
-
-
-
   }];
 
   let arguments = (ins XeGPU_BaseAddrType: $source,
@@ -663,4 +659,153 @@ def XeGPU_UpdateOffsetOp: XeGPU_Op<"update_offset",
   }];
 }
 
+def XeGPU_DpasOp : XeGPU_Op<"dpas", [Pure, AllElementTypesMatch<["lhs", "rhs"]>]> {
+  let summary = "It performs mma computation";
+
+  let description = [{DPAS performs matrix multiplication on matrix A of `mxk`
+    size, B of `kxn` size, and accumulate on matrix C of `mxn` to the same size
+    matrix , `m=8`, `n=16` and `k=8 * 32/bit_width_of_elem_type`. So for fp16
+    data type, the matrices are `A: vector<8x16xf16>`, `B: vector<16x16xf16>`,
+    and `C/D: vector<8x16xf32>`. Besides the matrix size requirements, DPAS
+    also requires A and B to be loaded with the required data layout. Specially,
+    VNNI layout is required for B operand. It is achieved via setting `vnni_axis = 0`
+    of the corresponding `load_nd` operator. To keep both operands as 3D vector,
+    operand A is loaded via setting `vnni_axis = 1` without impacting the
+    physical layouts change in register. Due to the VNNI transformation, A and B operands
+    are represented as 3D vector, with the last dimension representing the VNNI factor,
+    which is computed as `32/bit_width_of_elem_type`. Therefore, `A: vector<8x16xf16>`
+    is represented as `A: vector<8x8x2xf16>`, and `B: vector<16x16xf16>` is
+    represented as `B: vector<8x16x2xf16>`.
+
+    Note: on PVC, the hardware can perform load with VNNI transformation when data
+          element type is 16-bit or lower precision, taking 2 or 4 elements from
+          the first dimension and inserted into the newly added innermost dimension.
+  }];
+
+  let arguments = (ins
+    XeGPU_DpasOpType : $lhs,
+    XeGPU_DpasOpType : $rhs,
+    Optional<XeGPU_Vector2DType>: $acc);
+  let results = (outs XeGPU_Vector2DType: $result);
+
+  let extraClassDeclaration = [{
+    VectorType getLhsType() {
+      return getLhs().getType();
+    }
+
+    VectorType getRhsType() {
+      return getRhs().getType();
+    }
+
+    VectorType getAccType() {
+      if (getAcc())
+        return getAcc().getType();
+      return {};
+    }
+
+    VectorType getResultType() {
+      return getResult().getType();
+    }
+  }];
+
+  let assemblyFormat = [{
+    $lhs `,` $rhs (`,` $acc^)? attr-dict `:` type($lhs)`,` type($rhs) (`,` type($acc)^)?  `->` type($result)
+  }];
+
+  let hasVerifier = 1;
+}
+
+def XeGPU_AtomicRMWOp: XeGPU_Op<"atomic_rmw", [Pure,
+      AllElementTypesMatch<["tensorDesc", "value", "result"]>,
+      AllShapesMatch<["tensorDesc", "mask", "value", "result"]>]> {
+  let summary = "Atomic ready-modify-write operation on the TensorDesc. ";
+
+  let description = [{
+    The `xegpu.atomic_rmw` operation provides a way to perform a read-modify-write
+    operation on the region described by the `TensorDesc` free from data races. The
+    `kind` enumeration specifies the modification to be performed, The `mask` operand
+    has the same shape with `TensorDesc`, and is used to enable or disable specific
+    data points of the `TensorDesc`. The `value` operand represents the new value to
+    be applied during the modification.
+  }];
+
+  let arguments = (ins
+    AtomicRMWKindAttr:$kind,
+    XeGPU_TensorDesc:$tensorDesc,
+    XeGPU_MaskType:$mask,
+    XeGPU_ValueType:$value);
+
+  let results = (outs XeGPU_ValueType:$result);
+
+  let assemblyFormat = [{
+    $kind $tensorDesc `,` $mask `,` $value attr-dict `:`
+    type($tensorDesc) `,` type($mask) `,` type($value) `->` type($result)
+  }];
+}
+
+def XeGPU_AllocNbarrierOp: XeGPU_Op<"alloc_nbarrier", []> {
+  let summary = "It allocates a set of named barriers.";
+  let description = [{AllocNbarrier is to create a set of named barriers as
+  specified by `nbarrier_num`. Named barriers are workgroup level resources,
+    and are shared by all threads in the workgroup. For example, there are
+    up to 32 barriers (range 0-31) for each XeCore on PVC. A typical use case
+    is that a workgroup is partitioned into N subgroups of threads (N <= 32),
+    and each subgroup coordinating their work with a separate barrier with id
+    range from 0 to N respectively.}];
+  let arguments = (ins I64Attr: $nbarrier_num);
+  let assemblyFormat = "$nbarrier_num attr-dict";
+}
+
+def XeGPU_InitNbarrierOp: XeGPU_Op<"init_nbarrier", []> {
+  let summary = "It assigns a named barrier to the current thread.";
+  let description = [{InitNbarrierOp assigns the named barrier with the specified
+      barrier ID (0~31) to the current thread. Multiple threads may bind to the
+      same named barrier, and the `participant_thread_num` specifies the total
+      number of threads associated with the nbarrier. It returns an object of
+      NbarrierType representing the barrier}];
+
+  let arguments = (ins I8: $nbarrier_id,
+                       I8: $participant_thread_num);
+  let results = (outs XeGPU_Nbarrier: $result);
+  let assemblyFormat = [{
+    $nbarrier_id `,` $participant_thread_num attr-dict `:`
+    type($nbarrier_id) `,` type($participant_thread_num) `->` qualified(type($result))
+  }];
+}
+
+def XeGPU_NbarrierArriveOp: XeGPU_Op<"nbarrier_arrive", []> {
+  let summary = "It signals the arrival at the named barrier.";
+  let description = [{NbarrierArriveOp signals the hardware (or other threads)
+    that the current thread has produced its data for the consumer threads. When
+    the hardware signalled by `participant_thread_num` threads for the named barrier,
+    it will notify the threads waiting for the named barrier to continue their work.}];
+
+  let arguments = (ins XeGPU_Nbarrier: $nbarrier);
+  let assemblyFormat = [{ $nbarrier attr-dict `:` qualified(type($nbarrier))}];
+}
+
+def XeGPU_NbarrierWaitOp: XeGPU_Op<"nbarrier_wait", []> {
+  let summary = "It waits for a named barrier.";
+  let description = [{NbarrierWaitOp signals the hardware which named barrier
+    the current thread is waiting for, such that it can get notified when the
+    named barrier is completed.}];
+  let arguments = (ins XeGPU_Nbarrier: $nbarrier);
+  let assemblyFormat = [{ $nbarrier attr-dict `:` qualified(type($nbarrier)) }];
+}
+
+def XeGPU_FenceOp: XeGPU_Op<"fence", []> {
+  let summary = "It synchronizes memory accesses.";
+  let description = [{It synchronizes the memory access between
+    write and following read or write.
+    1. `Memory_kind` describes the memory kind. "global" means the global memory,
+        "slm" means the share local memory.
+    2. `Fence_scope` describes the scope of fence. "Workgroup" means that the scope would be
+        within each workgroup. "GPU" means the scope would be across workgroups within the GPU.
+  }];
+  let arguments = (ins XeGPU_MemoryScopeAttr: $memory_kind,
+                       XeGPU_FenceScopeAttr: $fence_scope);
+  let assemblyFormat = [{`memory_kind` `=` `` $memory_kind `,` `fence_scope` `=` `` $fence_scope attr-dict}];
+  let extraClassDeclaration = extraBaseClassDeclaration;
+}
+
 #endif // MLIR_DIALECT_XEGPU_IR_XEGPUOPS_TD
diff --git a/mlir/include/mlir/Dialect/XeGPU/IR/XeGPUTypes.td b/mlir/include/mlir/Dialect/XeGPU/IR/XeGPUTypes.td
index 4cd4e5411653..bab0e4afb1e5 100644
--- a/mlir/include/mlir/Dialect/XeGPU/IR/XeGPUTypes.td
+++ b/mlir/include/mlir/Dialect/XeGPU/IR/XeGPUTypes.td
@@ -151,4 +151,15 @@ def XeGPU_TensorDesc: XeGPUTypeDef<"TensorDesc", "tensor_desc",
 
 }
 
+
+def XeGPU_Nbarrier: XeGPUTypeDef<"Nbarrier", "nbarrier", [], "mlir::Type"> {
+  let summary = "!xegpu.nbarrier a custom XeGPU type representing a barrier.";
+
+  let extraClassDeclaration = [{
+    static NbarrierType get(mlir::MLIRContext *context) {
+      return Base::get(context);
+    };
+  }];
+}
+
 #endif // MLIR_DIALECT_XEGPU_IR_XEGPUTYPES_TD
diff --git a/mlir/include/mlir/IR/Builders.h b/mlir/include/mlir/IR/Builders.h
index 3beade017d1a..0d5fa719d0de 100644
--- a/mlir/include/mlir/IR/Builders.h
+++ b/mlir/include/mlir/IR/Builders.h
@@ -517,7 +517,7 @@ public:
 
   /// Create an operation of specific op type at the current insertion point,
   /// and immediately try to fold it. This functions populates 'results' with
-  /// the results after folding the operation.
+  /// the results of the operation.
   template <typename OpTy, typename... Args>
   void createOrFold(SmallVectorImpl<Value> &results, Location location,
                     Args &&...args) {
@@ -530,10 +530,17 @@ public:
     if (block)
       block->getOperations().insert(insertPoint, op);
 
-    // Fold the operation. If successful erase it, otherwise notify.
-    if (succeeded(tryFold(op, results)))
+    // Attempt to fold the operation.
+    if (succeeded(tryFold(op, results)) && !results.empty()) {
+      // Erase the operation, if the fold removed the need for this operation.
+      // Note: The fold already populated the results in this case.
       op->erase();
-    else if (block && listener)
+      return;
+    }
+
+    ResultRange opResults = op->getResults();
+    results.assign(opResults.begin(), opResults.end());
+    if (block && listener)
       listener->notifyOperationInserted(op, /*previous=*/{});
   }
 
@@ -560,7 +567,8 @@ public:
   }
 
   /// Attempts to fold the given operation and places new results within
-  /// 'results'. Returns success if the operation was folded, failure otherwise.
+  /// `results`. Returns success if the operation was folded, failure otherwise.
+  /// If the fold was in-place, `results` will not be filled.
   /// Note: This function does not erase the operation on a successful fold.
   LogicalResult tryFold(Operation *op, SmallVectorImpl<Value> &results);
 
diff --git a/mlir/include/mlir/IR/BuiltinLocationAttributes.td b/mlir/include/mlir/IR/BuiltinLocationAttributes.td
index dfcc180071f7..5a72404dea15 100644
--- a/mlir/include/mlir/IR/BuiltinLocationAttributes.td
+++ b/mlir/include/mlir/IR/BuiltinLocationAttributes.td
@@ -228,7 +228,8 @@ def OpaqueLoc : Builtin_LocationAttr<"OpaqueLoc"> {
     template <typename T> static T getUnderlyingLocation(Location location) {
       assert(isa<T>(location));
       return reinterpret_cast<T>(
-          location.cast<mlir::OpaqueLoc>().getUnderlyingLocation());
+          mlir::cast<mlir::OpaqueLoc>(static_cast<LocationAttr>(location))
+              .getUnderlyingLocation());
     }
 
     /// Returns a pointer to some data structure that opaque location stores.
@@ -237,15 +238,17 @@ def OpaqueLoc : Builtin_LocationAttr<"OpaqueLoc"> {
     template <typename T>
     static T getUnderlyingLocationOrNull(Location location) {
       return isa<T>(location)
-                 ? reinterpret_cast<T>(
-                       location.cast<mlir::OpaqueLoc>().getUnderlyingLocation())
-                 : T(nullptr);
+                ? reinterpret_cast<T>(mlir::cast<mlir::OpaqueLoc>(
+                                          static_cast<LocationAttr>(location))
+                                          .getUnderlyingLocation())
+                : T(nullptr);
     }
 
     /// Checks whether provided location is opaque location and contains a
     /// pointer to an object of particular type.
     template <typename T> static bool isa(Location location) {
-      auto opaque_loc = location.dyn_cast<OpaqueLoc>();
+      auto opaque_loc =
+          mlir::dyn_cast<OpaqueLoc>(static_cast<LocationAttr>(location));
       return opaque_loc && opaque_loc.getUnderlyingTypeID() == TypeID::get<T>();
     }
   }];
diff --git a/mlir/include/mlir/IR/OperationSupport.h b/mlir/include/mlir/IR/OperationSupport.h
index 2c1c490aac49..cdb75a3777ad 100644
--- a/mlir/include/mlir/IR/OperationSupport.h
+++ b/mlir/include/mlir/IR/OperationSupport.h
@@ -960,9 +960,12 @@ struct OperationState {
   /// Regions that the op will hold.
   SmallVector<std::unique_ptr<Region>, 1> regions;
 
-  // If we're creating an unregistered operation, this Attribute is used to
-  // build the properties. Otherwise it is ignored. For registered operations
-  // see the `getOrAddProperties` method.
+  /// This Attribute is used to opaquely construct the properties of the
+  /// operation. If we're creating an unregistered operation, the Attribute is
+  /// used as-is as the Properties storage of the operation. Otherwise, the
+  /// operation properties are constructed opaquely using its
+  /// `setPropertiesFromAttr` hook. Note that `getOrAddProperties` is the
+  /// preferred method to construct properties from C++.
   Attribute propertiesAttr;
 
 private:
diff --git a/mlir/include/mlir/IR/Value.h b/mlir/include/mlir/IR/Value.h
index cdbc6cc37436..a7344c64e673 100644
--- a/mlir/include/mlir/IR/Value.h
+++ b/mlir/include/mlir/IR/Value.h
@@ -98,25 +98,25 @@ public:
   constexpr Value(detail::ValueImpl *impl = nullptr) : impl(impl) {}
 
   template <typename U>
-  [[deprecated("Use isa<U>() instead")]]
+  [[deprecated("Use mlir::isa<U>() instead")]]
   bool isa() const {
     return llvm::isa<U>(*this);
   }
 
   template <typename U>
-  [[deprecated("Use dyn_cast<U>() instead")]]
+  [[deprecated("Use mlir::dyn_cast<U>() instead")]]
   U dyn_cast() const {
     return llvm::dyn_cast<U>(*this);
   }
 
   template <typename U>
-  [[deprecated("Use dyn_cast_or_null<U>() instead")]]
+  [[deprecated("Use mlir::dyn_cast_or_null<U>() instead")]]
   U dyn_cast_or_null() const {
     return llvm::dyn_cast_or_null<U>(*this);
   }
 
   template <typename U>
-  [[deprecated("Use cast<U>() instead")]]
+  [[deprecated("Use mlir::cast<U>() instead")]]
   U cast() const {
     return llvm::cast<U>(*this);
   }
diff --git a/mlir/include/mlir/InitAllDialects.h b/mlir/include/mlir/InitAllDialects.h
index c4d788cf8ed3..d9db21073e15 100644
--- a/mlir/include/mlir/InitAllDialects.h
+++ b/mlir/include/mlir/InitAllDialects.h
@@ -45,6 +45,7 @@
 #include "mlir/Dialect/LLVMIR/ROCDLDialect.h"
 #include "mlir/Dialect/Linalg/IR/Linalg.h"
 #include "mlir/Dialect/Linalg/Transforms/AllInterfaces.h"
+#include "mlir/Dialect/Linalg/Transforms/RuntimeOpVerification.h"
 #include "mlir/Dialect/MLProgram/IR/MLProgram.h"
 #include "mlir/Dialect/MLProgram/Transforms/BufferizableOpInterfaceImpl.h"
 #include "mlir/Dialect/MPI/IR/MPI.h"
@@ -161,6 +162,7 @@ inline void registerAllDialects(DialectRegistry &registry) {
   cf::registerBufferDeallocationOpInterfaceExternalModels(registry);
   gpu::registerBufferDeallocationOpInterfaceExternalModels(registry);
   linalg::registerAllDialectInterfaceImplementations(registry);
+  linalg::registerRuntimeVerifiableOpInterfaceExternalModels(registry);
   memref::registerAllocationOpInterfaceExternalModels(registry);
   memref::registerBufferViewFlowOpInterfaceExternalModels(registry);
   memref::registerRuntimeVerifiableOpInterfaceExternalModels(registry);
diff --git a/mlir/include/mlir/Interfaces/MemorySlotInterfaces.td b/mlir/include/mlir/Interfaces/MemorySlotInterfaces.td
index 8c642c0ed26a..764fa6d547b2 100644
--- a/mlir/include/mlir/Interfaces/MemorySlotInterfaces.td
+++ b/mlir/include/mlir/Interfaces/MemorySlotInterfaces.td
@@ -128,6 +128,7 @@ def PromotableMemOpInterface : OpInterface<"PromotableMemOpInterface"> {
       "::mlir::Value", "getStored",
       (ins "const ::mlir::MemorySlot &":$slot,
            "::mlir::RewriterBase &":$rewriter,
+           "::mlir::Value":$reachingDef,
            "const ::mlir::DataLayout &":$dataLayout)
     >,
     InterfaceMethod<[{
diff --git a/mlir/include/mlir/Interfaces/RuntimeVerifiableOpInterface.td b/mlir/include/mlir/Interfaces/RuntimeVerifiableOpInterface.td
index d5f11d00cc3d..6fd0df59d9d2 100644
--- a/mlir/include/mlir/Interfaces/RuntimeVerifiableOpInterface.td
+++ b/mlir/include/mlir/Interfaces/RuntimeVerifiableOpInterface.td
@@ -35,6 +35,12 @@ def RuntimeVerifiableOpInterface : OpInterface<"RuntimeVerifiableOpInterface"> {
                     "::mlir::Location":$loc)
     >,
   ];
+
+  let extraClassDeclaration = [{
+    /// Generate the error message that will be printed to the user when 
+    /// verification fails.
+    static std::string generateErrorMessage(Operation *op, const std::string &msg);
+  }];
 }
 
 #endif // MLIR_INTERFACES_RUNTIMEVERIFIABLEOPINTERFACE
diff --git a/mlir/include/mlir/Interfaces/TilingInterface.td b/mlir/include/mlir/Interfaces/TilingInterface.td
index 84f7dec2f400..66382f29c242 100644
--- a/mlir/include/mlir/Interfaces/TilingInterface.td
+++ b/mlir/include/mlir/Interfaces/TilingInterface.td
@@ -63,7 +63,7 @@ def TilingInterface : OpInterface<"TilingInterface"> {
           The method returns the operation that is the tiled
           implementation.
         }],
-        /*retType=*/"FailureOr<::mlir::TilingResult>",
+        /*retType=*/"FailureOr<TilingResult>",
         /*methodName=*/"getTiledImplementation",
         /*args=*/(ins
             "OpBuilder &":$b,
@@ -82,34 +82,15 @@ def TilingInterface : OpInterface<"TilingInterface"> {
           by the tiled implementation. Expects the same `offsets` and `sizes` as
           used to obtain the tiled implementation of the operation.
         }],
-        /*retType=*/"::mlir::LogicalResult",
+        /*retType=*/"LogicalResult",
         /*methodName=*/"getResultTilePosition",
         /*args=*/(ins
           "OpBuilder &":$b,
           "unsigned":$resultNumber,
           "ArrayRef<OpFoldResult> ":$offsets,
           "ArrayRef<OpFoldResult> ":$sizes,
-          "SmallVectorImpl<OpFoldResult> &":$resultOffsets,
-          "SmallVectorImpl<OpFoldResult> &":$resultSizes),
-        /*methodBody=*/"",
-        /*defaultImplementation=*/[{
-          return failure();
-        }]
-      >,
-      InterfaceMethod<
-        /*desc=*/[{
-          Method to return the position of iteration domain tile computed by the
-          tiled operation.
-        }],
-        /*retType=*/"::mlir::LogicalResult",
-        /*methodName=*/"getIterationDomainTileFromOperandTile",
-        /*args=*/(ins
-          "OpBuilder &":$b,
-          "unsigned":$operandNumber,
-          "ArrayRef<OpFoldResult> ":$offsets,
-          "ArrayRef<OpFoldResult> ":$sizes,
-          "SmallVectorImpl<OpFoldResult> &":$iterDomainOffsets,
-          "SmallVectorImpl<OpFoldResult> &":$iterDomainSizes),
+          "SmallVector<OpFoldResult> &":$resultOffsets,
+          "SmallVector<OpFoldResult> &":$resultSizes),
         /*methodBody=*/"",
         /*defaultImplementation=*/[{
           return failure();
@@ -138,7 +119,7 @@ def TilingInterface : OpInterface<"TilingInterface"> {
             iteration space).
           - `sizes` provides the size of the tile.
         }],
-        /*retType=*/"FailureOr<::mlir::TilingResult>",
+        /*retType=*/"FailureOr<TilingResult>",
         /*methodName=*/"generateResultTileValue",
         /*args=*/(ins
           "OpBuilder &":$b,
@@ -152,42 +133,6 @@ def TilingInterface : OpInterface<"TilingInterface"> {
       >,
       InterfaceMethod<
         /*desc=*/[{
-          Method to generate the tiled implementation of an operation from
-          operand tile position.
-
-          Generates the IR that computes the tiled implementation of an
-          operation from operand tile.  The `offsets` and `sizes`
-          describe the tile of the operand required. This is different from
-          `getTiledImplementation` which generates the tiled
-          implementation of the operation given a tile of the
-          iteration space. This method generates a tiled
-          implementation of the operation based on the tile of the
-          operand required. This method enables consumer fusion by using
-          tile and fuse. The method returns failure if the operation
-          can't be tiled to generate the operand tile. In practical terms
-          this implies it cannot be tiled and fused with its producers.
-
-          - `offsets` provides the offset of the tile in the coordinate system
-            of the original iteration space, i.e., if an iteration space
-            dimension had non-zero offset, it must be included in the offset
-            provided here (as opposed to zero-based offset "relative" to the
-            iteration space).
-          - `sizes` provides the size of the tile.
-        }],
-        /*retType=*/"FailureOr<::mlir::TilingResult>",
-        /*methodName=*/"getTiledImplementationFromOperandTile",
-        /*args=*/(ins
-          "OpBuilder &":$b,
-          "unsigned":$operandNumber,
-          "ArrayRef<OpFoldResult>":$offsets,
-          "ArrayRef<OpFoldResult>":$sizes),
-        /*methodBody=*/"",
-        /*defaultImplementation=*/[{
-          return failure();
-        }]
-      >,
-      InterfaceMethod<
-        /*desc=*/[{
           Generates the scalar implementation of the operation.
 
           Given the list `ivs` that represent points in the iteration space
@@ -197,7 +142,7 @@ def TilingInterface : OpInterface<"TilingInterface"> {
           transformations are done, this method can be used to lower to scalar
           code that can then be lowered to LLVM or SPIR-V dialects.
         }],
-        /*retType=*/"::mlir::LogicalResult",
+        /*retType=*/"LogicalResult",
         /*methodName=*/"generateScalarImplementation",
         /*args=*/(ins
             "OpBuilder &":$b,
diff --git a/mlir/include/mlir/TableGen/CodeGenHelpers.h b/mlir/include/mlir/TableGen/CodeGenHelpers.h
index dd17a44c889b..c263c69c53d1 100644
--- a/mlir/include/mlir/TableGen/CodeGenHelpers.h
+++ b/mlir/include/mlir/TableGen/CodeGenHelpers.h
@@ -99,8 +99,14 @@ private:
 ///
 class StaticVerifierFunctionEmitter {
 public:
+  /// Create a constraint uniquer with a unique prefix derived from the record
+  /// keeper with an optional tag.
   StaticVerifierFunctionEmitter(raw_ostream &os,
-                                const llvm::RecordKeeper &records);
+                                const llvm::RecordKeeper &records,
+                                StringRef tag = "");
+
+  /// Collect and unique all the constraints used by operations.
+  void collectOpConstraints(ArrayRef<llvm::Record *> opDefs);
 
   /// Collect and unique all compatible type, attribute, successor, and region
   /// constraints from the operations in the file and emit them at the top of
@@ -108,7 +114,7 @@ public:
   ///
   /// Constraints that do not meet the restriction that they can only reference
   /// `$_self` and `$_op` are not uniqued.
-  void emitOpConstraints(ArrayRef<llvm::Record *> opDefs, bool emitDecl);
+  void emitOpConstraints(ArrayRef<llvm::Record *> opDefs);
 
   /// Unique all compatible type and attribute constraints from a pattern file
   /// and emit them at the top of the generated file.
@@ -177,8 +183,6 @@ private:
   /// Emit pattern constraints.
   void emitPatternConstraints();
 
-  /// Collect and unique all the constraints used by operations.
-  void collectOpConstraints(ArrayRef<llvm::Record *> opDefs);
   /// Collect and unique all pattern constraints.
   void collectPatternConstraints(ArrayRef<DagLeaf> constraints);
 
diff --git a/mlir/include/mlir/Tools/lsp-server-support/Transport.h b/mlir/include/mlir/Tools/lsp-server-support/Transport.h
index ce742be7a941..44c71058cf71 100644
--- a/mlir/include/mlir/Tools/lsp-server-support/Transport.h
+++ b/mlir/include/mlir/Tools/lsp-server-support/Transport.h
@@ -147,9 +147,15 @@ public:
                     void (ThisT::*handler)(const Param &)) {
     notificationHandlers[method] = [method, handler,
                                     thisPtr](llvm::json::Value rawParams) {
-      llvm::Expected<Param> param = parse<Param>(rawParams, method, "request");
-      if (!param)
-        return llvm::consumeError(param.takeError());
+      llvm::Expected<Param> param =
+          parse<Param>(rawParams, method, "notification");
+      if (!param) {
+        return llvm::consumeError(
+            llvm::handleErrors(param.takeError(), [](const LSPError &lspError) {
+              Logger::error("JSON parsing error: {0}",
+                            lspError.message.c_str());
+            }));
+      }
       (thisPtr->*handler)(*param);
     };
   }
diff --git a/mlir/lib/Bindings/Python/DialectLLVM.cpp b/mlir/lib/Bindings/Python/DialectLLVM.cpp
index 843707751dd8..42a4c8c0793b 100644
--- a/mlir/lib/Bindings/Python/DialectLLVM.cpp
+++ b/mlir/lib/Bindings/Python/DialectLLVM.cpp
@@ -19,6 +19,11 @@ using namespace mlir::python;
 using namespace mlir::python::adaptors;
 
 void populateDialectLLVMSubmodule(const pybind11::module &m) {
+
+  //===--------------------------------------------------------------------===//
+  // StructType
+  //===--------------------------------------------------------------------===//
+
   auto llvmStructType =
       mlir_type_subclass(m, "StructType", mlirTypeIsALLVMStructType);
 
@@ -35,8 +40,8 @@ void populateDialectLLVMSubmodule(const pybind11::module &m) {
         }
         return cls(type);
       },
-      py::arg("cls"), py::arg("elements"), py::kw_only(),
-      py::arg("packed") = false, py::arg("loc") = py::none());
+      "cls"_a, "elements"_a, py::kw_only(), "packed"_a = false,
+      "loc"_a = py::none());
 
   llvmStructType.def_classmethod(
       "get_identified",
@@ -44,8 +49,7 @@ void populateDialectLLVMSubmodule(const pybind11::module &m) {
         return cls(mlirLLVMStructTypeIdentifiedGet(
             context, mlirStringRefCreate(name.data(), name.size())));
       },
-      py::arg("cls"), py::arg("name"), py::kw_only(),
-      py::arg("context") = py::none());
+      "cls"_a, "name"_a, py::kw_only(), "context"_a = py::none());
 
   llvmStructType.def_classmethod(
       "get_opaque",
@@ -53,7 +57,7 @@ void populateDialectLLVMSubmodule(const pybind11::module &m) {
         return cls(mlirLLVMStructTypeOpaqueGet(
             context, mlirStringRefCreate(name.data(), name.size())));
       },
-      py::arg("cls"), py::arg("name"), py::arg("context") = py::none());
+      "cls"_a, "name"_a, "context"_a = py::none());
 
   llvmStructType.def(
       "set_body",
@@ -65,7 +69,7 @@ void populateDialectLLVMSubmodule(const pybind11::module &m) {
               "Struct body already set to different content.");
         }
       },
-      py::arg("elements"), py::kw_only(), py::arg("packed") = false);
+      "elements"_a, py::kw_only(), "packed"_a = false);
 
   llvmStructType.def_classmethod(
       "new_identified",
@@ -75,8 +79,8 @@ void populateDialectLLVMSubmodule(const pybind11::module &m) {
             ctx, mlirStringRefCreate(name.data(), name.length()),
             elements.size(), elements.data(), packed));
       },
-      py::arg("cls"), py::arg("name"), py::arg("elements"), py::kw_only(),
-      py::arg("packed") = false, py::arg("context") = py::none());
+      "cls"_a, "name"_a, "elements"_a, py::kw_only(), "packed"_a = false,
+      "context"_a = py::none());
 
   llvmStructType.def_property_readonly(
       "name", [](MlirType type) -> std::optional<std::string> {
@@ -105,6 +109,29 @@ void populateDialectLLVMSubmodule(const pybind11::module &m) {
 
   llvmStructType.def_property_readonly(
       "opaque", [](MlirType type) { return mlirLLVMStructTypeIsOpaque(type); });
+
+  //===--------------------------------------------------------------------===//
+  // PointerType
+  //===--------------------------------------------------------------------===//
+
+  mlir_type_subclass(m, "PointerType", mlirTypeIsALLVMPointerType)
+      .def_classmethod(
+          "get",
+          [](py::object cls, std::optional<unsigned> addressSpace,
+             MlirContext context) {
+            CollectDiagnosticsToStringScope scope(context);
+            MlirType type = mlirLLVMPointerTypeGet(
+                context, addressSpace.has_value() ? *addressSpace : 0);
+            if (mlirTypeIsNull(type)) {
+              throw py::value_error(scope.takeMessage());
+            }
+            return cls(type);
+          },
+          "cls"_a, "address_space"_a = py::none(), py::kw_only(),
+          "context"_a = py::none())
+      .def_property_readonly("address_space", [](MlirType type) {
+        return mlirLLVMPointerTypeGetAddressSpace(type);
+      });
 }
 
 PYBIND11_MODULE(_mlirDialectsLLVM, m) {
diff --git a/mlir/lib/Bindings/Python/DialectSparseTensor.cpp b/mlir/lib/Bindings/Python/DialectSparseTensor.cpp
index 171faf9e0087..584981cfe99b 100644
--- a/mlir/lib/Bindings/Python/DialectSparseTensor.cpp
+++ b/mlir/lib/Bindings/Python/DialectSparseTensor.cpp
@@ -42,16 +42,19 @@ static void populateDialectSparseTensorSubmodule(const py::module &m) {
           [](py::object cls, std::vector<MlirSparseTensorLevelType> lvlTypes,
              std::optional<MlirAffineMap> dimToLvl,
              std::optional<MlirAffineMap> lvlToDim, int posWidth, int crdWidth,
-             MlirContext context) {
+             std::optional<MlirAttribute> explicitVal,
+             std::optional<MlirAttribute> implicitVal, MlirContext context) {
             return cls(mlirSparseTensorEncodingAttrGet(
                 context, lvlTypes.size(), lvlTypes.data(),
                 dimToLvl ? *dimToLvl : MlirAffineMap{nullptr},
                 lvlToDim ? *lvlToDim : MlirAffineMap{nullptr}, posWidth,
-                crdWidth));
+                crdWidth, explicitVal ? *explicitVal : MlirAttribute{nullptr},
+                implicitVal ? *implicitVal : MlirAttribute{nullptr}));
           },
           py::arg("cls"), py::arg("lvl_types"), py::arg("dim_to_lvl"),
           py::arg("lvl_to_dim"), py::arg("pos_width"), py::arg("crd_width"),
-          py::arg("context") = py::none(),
+          py::arg("explicit_val") = py::none(),
+          py::arg("implicit_val") = py::none(), py::arg("context") = py::none(),
           "Gets a sparse_tensor.encoding from parameters.")
       .def_classmethod(
           "build_level_type",
@@ -98,6 +101,24 @@ static void populateDialectSparseTensorSubmodule(const py::module &m) {
       .def_property_readonly("crd_width",
                              mlirSparseTensorEncodingAttrGetCrdWidth)
       .def_property_readonly(
+          "explicit_val",
+          [](MlirAttribute self) -> std::optional<MlirAttribute> {
+            MlirAttribute ret =
+                mlirSparseTensorEncodingAttrGetExplicitVal(self);
+            if (mlirAttributeIsNull(ret))
+              return {};
+            return ret;
+          })
+      .def_property_readonly(
+          "implicit_val",
+          [](MlirAttribute self) -> std::optional<MlirAttribute> {
+            MlirAttribute ret =
+                mlirSparseTensorEncodingAttrGetImplicitVal(self);
+            if (mlirAttributeIsNull(ret))
+              return {};
+            return ret;
+          })
+      .def_property_readonly(
           "structured_n",
           [](MlirAttribute self) -> unsigned {
             const int lvlRank = mlirSparseTensorEncodingGetLvlRank(self);
diff --git a/mlir/lib/CAPI/Dialect/LLVM.cpp b/mlir/lib/CAPI/Dialect/LLVM.cpp
index 21c66f38a8af..108ebe5367d5 100644
--- a/mlir/lib/CAPI/Dialect/LLVM.cpp
+++ b/mlir/lib/CAPI/Dialect/LLVM.cpp
@@ -27,6 +27,14 @@ MlirType mlirLLVMPointerTypeGet(MlirContext ctx, unsigned addressSpace) {
   return wrap(LLVMPointerType::get(unwrap(ctx), addressSpace));
 }
 
+bool mlirTypeIsALLVMPointerType(MlirType type) {
+  return isa<LLVM::LLVMPointerType>(unwrap(type));
+}
+
+unsigned mlirLLVMPointerTypeGetAddressSpace(MlirType pointerType) {
+  return cast<LLVM::LLVMPointerType>(unwrap(pointerType)).getAddressSpace();
+}
+
 MlirType mlirLLVMVoidTypeGet(MlirContext ctx) {
   return wrap(LLVMVoidType::get(unwrap(ctx)));
 }
diff --git a/mlir/lib/CAPI/Dialect/SparseTensor.cpp b/mlir/lib/CAPI/Dialect/SparseTensor.cpp
index 3ae06f220c52..19171d64d409 100644
--- a/mlir/lib/CAPI/Dialect/SparseTensor.cpp
+++ b/mlir/lib/CAPI/Dialect/SparseTensor.cpp
@@ -44,18 +44,20 @@ bool mlirAttributeIsASparseTensorEncodingAttr(MlirAttribute attr) {
   return isa<SparseTensorEncodingAttr>(unwrap(attr));
 }
 
-MlirAttribute
-mlirSparseTensorEncodingAttrGet(MlirContext ctx, intptr_t lvlRank,
-                                MlirSparseTensorLevelType const *lvlTypes,
-                                MlirAffineMap dimToLvl, MlirAffineMap lvlToDim,
-                                int posWidth, int crdWidth) {
+MlirAttribute mlirSparseTensorEncodingAttrGet(
+    MlirContext ctx, intptr_t lvlRank,
+    MlirSparseTensorLevelType const *lvlTypes, MlirAffineMap dimToLvl,
+    MlirAffineMap lvlToDim, int posWidth, int crdWidth,
+    MlirAttribute explicitVal, MlirAttribute implicitVal) {
   SmallVector<LevelType> cppLvlTypes;
+
   cppLvlTypes.reserve(lvlRank);
   for (intptr_t l = 0; l < lvlRank; ++l)
     cppLvlTypes.push_back(static_cast<LevelType>(lvlTypes[l]));
-  return wrap(SparseTensorEncodingAttr::get(unwrap(ctx), cppLvlTypes,
-                                            unwrap(dimToLvl), unwrap(lvlToDim),
-                                            posWidth, crdWidth));
+
+  return wrap(SparseTensorEncodingAttr::get(
+      unwrap(ctx), cppLvlTypes, unwrap(dimToLvl), unwrap(lvlToDim), posWidth,
+      crdWidth, unwrap(explicitVal), unwrap(implicitVal)));
 }
 
 MlirAffineMap mlirSparseTensorEncodingAttrGetDimToLvl(MlirAttribute attr) {
@@ -91,6 +93,14 @@ int mlirSparseTensorEncodingAttrGetCrdWidth(MlirAttribute attr) {
   return cast<SparseTensorEncodingAttr>(unwrap(attr)).getCrdWidth();
 }
 
+MlirAttribute mlirSparseTensorEncodingAttrGetExplicitVal(MlirAttribute attr) {
+  return wrap(cast<SparseTensorEncodingAttr>(unwrap(attr)).getExplicitVal());
+}
+
+MlirAttribute mlirSparseTensorEncodingAttrGetImplicitVal(MlirAttribute attr) {
+  return wrap(cast<SparseTensorEncodingAttr>(unwrap(attr)).getImplicitVal());
+}
+
 MlirSparseTensorLevelType mlirSparseTensorEncodingAttrBuildLvlType(
     enum MlirSparseTensorLevelFormat lvlFmt,
     const enum MlirSparseTensorLevelPropertyNondefault *properties,
diff --git a/mlir/lib/Conversion/ComplexToStandard/ComplexToStandard.cpp b/mlir/lib/Conversion/ComplexToStandard/ComplexToStandard.cpp
index 4a15976d40c7..c2a83f90bcbe 100644
--- a/mlir/lib/Conversion/ComplexToStandard/ComplexToStandard.cpp
+++ b/mlir/lib/Conversion/ComplexToStandard/ComplexToStandard.cpp
@@ -857,7 +857,7 @@ struct SqrtOpConversion : public OpConversionPattern<complex::SqrtOp> {
     ImplicitLocOpBuilder b(op.getLoc(), rewriter);
 
     auto type = cast<ComplexType>(op.getType());
-    auto elementType = type.getElementType().cast<FloatType>();
+    auto elementType = cast<FloatType>(type.getElementType());
     arith::FastMathFlags fmf = op.getFastMathFlagsAttr().getValue();
 
     auto cst = [&](APFloat v) {
diff --git a/mlir/lib/Conversion/MemRefToEmitC/MemRefToEmitC.cpp b/mlir/lib/Conversion/MemRefToEmitC/MemRefToEmitC.cpp
index 25fa15892203..e0c421741b30 100644
--- a/mlir/lib/Conversion/MemRefToEmitC/MemRefToEmitC.cpp
+++ b/mlir/lib/Conversion/MemRefToEmitC/MemRefToEmitC.cpp
@@ -50,6 +50,72 @@ struct ConvertAlloca final : public OpConversionPattern<memref::AllocaOp> {
   }
 };
 
+struct ConvertGlobal final : public OpConversionPattern<memref::GlobalOp> {
+  using OpConversionPattern::OpConversionPattern;
+
+  LogicalResult
+  matchAndRewrite(memref::GlobalOp op, OpAdaptor operands,
+                  ConversionPatternRewriter &rewriter) const override {
+
+    if (!op.getType().hasStaticShape()) {
+      return rewriter.notifyMatchFailure(
+          op.getLoc(), "cannot transform global with dynamic shape");
+    }
+
+    if (op.getAlignment().value_or(1) > 1) {
+      // TODO: Extend GlobalOp to specify alignment via the `alignas` specifier.
+      return rewriter.notifyMatchFailure(
+          op.getLoc(), "global variable with alignment requirement is "
+                       "currently not supported");
+    }
+    auto resultTy = getTypeConverter()->convertType(op.getType());
+    if (!resultTy) {
+      return rewriter.notifyMatchFailure(op.getLoc(),
+                                         "cannot convert result type");
+    }
+
+    SymbolTable::Visibility visibility = SymbolTable::getSymbolVisibility(op);
+    if (visibility != SymbolTable::Visibility::Public &&
+        visibility != SymbolTable::Visibility::Private) {
+      return rewriter.notifyMatchFailure(
+          op.getLoc(),
+          "only public and private visibility is currently supported");
+    }
+    // We are explicit in specifing the linkage because the default linkage
+    // for constants is different in C and C++.
+    bool staticSpecifier = visibility == SymbolTable::Visibility::Private;
+    bool externSpecifier = !staticSpecifier;
+
+    Attribute initialValue = operands.getInitialValueAttr();
+    if (isa_and_present<UnitAttr>(initialValue))
+      initialValue = {};
+
+    rewriter.replaceOpWithNewOp<emitc::GlobalOp>(
+        op, operands.getSymName(), resultTy, initialValue, externSpecifier,
+        staticSpecifier, operands.getConstant());
+    return success();
+  }
+};
+
+struct ConvertGetGlobal final
+    : public OpConversionPattern<memref::GetGlobalOp> {
+  using OpConversionPattern::OpConversionPattern;
+
+  LogicalResult
+  matchAndRewrite(memref::GetGlobalOp op, OpAdaptor operands,
+                  ConversionPatternRewriter &rewriter) const override {
+
+    auto resultTy = getTypeConverter()->convertType(op.getType());
+    if (!resultTy) {
+      return rewriter.notifyMatchFailure(op.getLoc(),
+                                         "cannot convert result type");
+    }
+    rewriter.replaceOpWithNewOp<emitc::GetGlobalOp>(op, resultTy,
+                                                    operands.getNameAttr());
+    return success();
+  }
+};
+
 struct ConvertLoad final : public OpConversionPattern<memref::LoadOp> {
   using OpConversionPattern::OpConversionPattern;
 
@@ -120,6 +186,6 @@ void mlir::populateMemRefToEmitCTypeConversion(TypeConverter &typeConverter) {
 
 void mlir::populateMemRefToEmitCConversionPatterns(RewritePatternSet &patterns,
                                                    TypeConverter &converter) {
-  patterns.add<ConvertAlloca, ConvertLoad, ConvertStore>(converter,
-                                                         patterns.getContext());
+  patterns.add<ConvertAlloca, ConvertGlobal, ConvertGetGlobal, ConvertLoad,
+               ConvertStore>(converter, patterns.getContext());
 }
diff --git a/mlir/lib/Conversion/SCFToOpenMP/SCFToOpenMP.cpp b/mlir/lib/Conversion/SCFToOpenMP/SCFToOpenMP.cpp
index 7f91367ad427..d6f85451ee5d 100644
--- a/mlir/lib/Conversion/SCFToOpenMP/SCFToOpenMP.cpp
+++ b/mlir/lib/Conversion/SCFToOpenMP/SCFToOpenMP.cpp
@@ -461,18 +461,50 @@ struct ParallelOpLowering : public OpRewritePattern<scf::ParallelOp> {
       // Replace the loop.
       {
         OpBuilder::InsertionGuard allocaGuard(rewriter);
-        auto loop = rewriter.create<omp::WsloopOp>(
+        // Create worksharing loop wrapper.
+        auto wsloopOp = rewriter.create<omp::WsloopOp>(parallelOp.getLoc());
+        if (!reductionVariables.empty()) {
+          wsloopOp.setReductionsAttr(
+              ArrayAttr::get(rewriter.getContext(), reductionDeclSymbols));
+          wsloopOp.getReductionVarsMutable().append(reductionVariables);
+        }
+        rewriter.create<omp::TerminatorOp>(loc); // omp.parallel terminator.
+
+        // The wrapper's entry block arguments will define the reduction
+        // variables.
+        llvm::SmallVector<mlir::Type> reductionTypes;
+        reductionTypes.reserve(reductionVariables.size());
+        llvm::transform(reductionVariables, std::back_inserter(reductionTypes),
+                        [](mlir::Value v) { return v.getType(); });
+        rewriter.createBlock(
+            &wsloopOp.getRegion(), {}, reductionTypes,
+            llvm::SmallVector<mlir::Location>(reductionVariables.size(),
+                                              parallelOp.getLoc()));
+
+        rewriter.setInsertionPoint(
+            rewriter.create<omp::TerminatorOp>(parallelOp.getLoc()));
+
+        // Create loop nest and populate region with contents of scf.parallel.
+        auto loopOp = rewriter.create<omp::LoopNestOp>(
             parallelOp.getLoc(), parallelOp.getLowerBound(),
             parallelOp.getUpperBound(), parallelOp.getStep());
-        rewriter.create<omp::TerminatorOp>(loc);
 
-        rewriter.inlineRegionBefore(parallelOp.getRegion(), loop.getRegion(),
-                                    loop.getRegion().begin());
+        rewriter.inlineRegionBefore(parallelOp.getRegion(), loopOp.getRegion(),
+                                    loopOp.getRegion().begin());
 
-        Block *ops = rewriter.splitBlock(&*loop.getRegion().begin(),
-                                         loop.getRegion().begin()->begin());
+        // Remove reduction-related block arguments from omp.loop_nest and
+        // redirect uses to the corresponding omp.wsloop block argument.
+        mlir::Block &loopOpEntryBlock = loopOp.getRegion().front();
+        unsigned numLoops = parallelOp.getNumLoops();
+        rewriter.replaceAllUsesWith(
+            loopOpEntryBlock.getArguments().drop_front(numLoops),
+            wsloopOp.getRegion().getArguments());
+        loopOpEntryBlock.eraseArguments(
+            numLoops, loopOpEntryBlock.getNumArguments() - numLoops);
 
-        rewriter.setInsertionPointToStart(&*loop.getRegion().begin());
+        Block *ops =
+            rewriter.splitBlock(&loopOpEntryBlock, loopOpEntryBlock.begin());
+        rewriter.setInsertionPointToStart(&loopOpEntryBlock);
 
         auto scope = rewriter.create<memref::AllocaScopeOp>(parallelOp.getLoc(),
                                                             TypeRange());
@@ -481,11 +513,6 @@ struct ParallelOpLowering : public OpRewritePattern<scf::ParallelOp> {
         rewriter.mergeBlocks(ops, scopeBlock);
         rewriter.setInsertionPointToEnd(&*scope.getBodyRegion().begin());
         rewriter.create<memref::AllocaScopeReturnOp>(loc, ValueRange());
-        if (!reductionVariables.empty()) {
-          loop.setReductionsAttr(
-              ArrayAttr::get(rewriter.getContext(), reductionDeclSymbols));
-          loop.getReductionVarsMutable().append(reductionVariables);
-        }
       }
     }
 
diff --git a/mlir/lib/Conversion/TosaToLinalg/TosaToLinalgNamed.cpp b/mlir/lib/Conversion/TosaToLinalg/TosaToLinalgNamed.cpp
index 8fb8d1648656..d8fb3abc0bef 100644
--- a/mlir/lib/Conversion/TosaToLinalg/TosaToLinalgNamed.cpp
+++ b/mlir/lib/Conversion/TosaToLinalg/TosaToLinalgNamed.cpp
@@ -101,9 +101,18 @@ static mlir::Value linalgBroadcastAndMaybeExtSI(PatternRewriter &rewriter,
   // The source tensor is broadcast to all the outer dimensions of the
   // result tensor.
   SmallVector<AffineExpr> sourceDims;
-  for (auto dim : llvm::seq<int64_t>(0, sourceRank)) {
-    auto expr = rewriter.getAffineDimExpr(dim + resultRank - sourceRank);
-    sourceDims.push_back(expr);
+  // In the case of a rank one source tensor with a single element TOSA
+  // specifies that the value be broadcast meaning we need an edge case for a
+  // constant map.
+  assert(sourceTy.hasStaticShape() &&
+         "Dynamic broadcasting shapes not supported!");
+  if (sourceRank == 1 && sourceTy.getDimSize(0) == 1) {
+    sourceDims.push_back(rewriter.getAffineConstantExpr(0));
+  } else {
+    for (auto dim : llvm::seq<int64_t>(0, sourceRank)) {
+      auto expr = rewriter.getAffineDimExpr(dim + resultRank - sourceRank);
+      sourceDims.push_back(expr);
+    }
   }
 
   // Creating maps for the input and output of the broacast-like generic op.
diff --git a/mlir/lib/Conversion/VectorToLLVM/ConvertVectorToLLVM.cpp b/mlir/lib/Conversion/VectorToLLVM/ConvertVectorToLLVM.cpp
index 1b9975237c69..fe6bcc1c8b66 100644
--- a/mlir/lib/Conversion/VectorToLLVM/ConvertVectorToLLVM.cpp
+++ b/mlir/lib/Conversion/VectorToLLVM/ConvertVectorToLLVM.cpp
@@ -1738,7 +1738,7 @@ struct VectorInterleaveOpLowering
                                          "InterleaveOp not rank 1");
     // If the result is rank 1, then this directly maps to LLVM.
     if (resultType.isScalable()) {
-      rewriter.replaceOpWithNewOp<LLVM::experimental_vector_interleave2>(
+      rewriter.replaceOpWithNewOp<LLVM::vector_interleave2>(
           interleaveOp, typeConverter->convertType(resultType),
           adaptor.getLhs(), adaptor.getRhs());
       return success();
diff --git a/mlir/lib/Dialect/Arith/IR/ArithCanonicalization.td b/mlir/lib/Dialect/Arith/IR/ArithCanonicalization.td
index caca2ff81964..02d05780a7ac 100644
--- a/mlir/lib/Dialect/Arith/IR/ArithCanonicalization.td
+++ b/mlir/lib/Dialect/Arith/IR/ArithCanonicalization.td
@@ -175,6 +175,7 @@ def MulSIExtendedToMulI :
 def IsScalarOrSplatOne :
     Constraint<And<[
       CPred<"succeeded(getIntOrSplatIntValue($0))">,
+      CPred<"getIntOrSplatIntValue($0)->isStrictlyPositive()">,
       CPred<"*getIntOrSplatIntValue($0) == 1">]>>;
 
 // mulsi_extended(x, 1) -> [x, extsi(cmpi slt, x, 0)]
diff --git a/mlir/lib/Dialect/ArmSME/Transforms/OuterProductFusion.cpp b/mlir/lib/Dialect/ArmSME/Transforms/OuterProductFusion.cpp
index d3751d4ba7e7..39292c4533d6 100644
--- a/mlir/lib/Dialect/ArmSME/Transforms/OuterProductFusion.cpp
+++ b/mlir/lib/Dialect/ArmSME/Transforms/OuterProductFusion.cpp
@@ -86,8 +86,7 @@ static Value createInterleave2Intrinsic(RewriterBase &rewriter, Location loc,
   auto inputType = cast<VectorType>(lhs.getType());
   VectorType inputTypeX2 =
       VectorType::Builder(inputType).setDim(0, inputType.getShape()[0] * 2);
-  return rewriter.create<LLVM::experimental_vector_interleave2>(
-      loc, inputTypeX2, lhs, rhs);
+  return rewriter.create<LLVM::vector_interleave2>(loc, inputTypeX2, lhs, rhs);
 }
 
 // Fuse two 'arm_sme.outerproduct' operations that are chained via the
diff --git a/mlir/lib/Dialect/Bufferization/Transforms/OneShotAnalysis.cpp b/mlir/lib/Dialect/Bufferization/Transforms/OneShotAnalysis.cpp
index 531016130d1d..2d329a1f3d88 100644
--- a/mlir/lib/Dialect/Bufferization/Transforms/OneShotAnalysis.cpp
+++ b/mlir/lib/Dialect/Bufferization/Transforms/OneShotAnalysis.cpp
@@ -1382,14 +1382,27 @@ LogicalResult
 bufferization::runOneShotBufferize(Operation *op,
                                    const OneShotBufferizationOptions &options,
                                    BufferizationStatistics *statistics) {
+  // copy-before-write deactivates the analysis. It cannot be used together with
+  // test-analysis-only.
   assert(!(options.copyBeforeWrite && options.testAnalysisOnly) &&
          "invalid combination of bufferization flags");
-  if (!options.copyBeforeWrite) {
-    // If a buffer is copied before every write, no analysis is needed.
+
+  if (options.copyBeforeWrite) {
+    // Copy buffer before each write. No analysis is needed.
+  } else {
+    // Run One-Shot Analysis and insert buffer copies (on the tensor level)
+    // only where needed. This is the default and much more efficient than
+    // copy-before-write.
     if (failed(insertTensorCopies(op, options, statistics)))
       return failure();
+
+    // If test-analysis-only is set, the IR was annotated with RaW conflict
+    // markers (attributes) during One-Shot Analysis.
+    if (options.testAnalysisOnly)
+      return success();
   }
-  if (options.testAnalysisOnly)
-    return success();
+
+  // Bufferize the op and its nested ops. If options.copyBeforeWrite is set,
+  // a new buffer copy is allocated every time a buffer is written to.
   return bufferizeOp(op, options, statistics);
 }
diff --git a/mlir/lib/Dialect/EmitC/IR/EmitC.cpp b/mlir/lib/Dialect/EmitC/IR/EmitC.cpp
index 66a71df29a9b..ef7b7a19489d 100644
--- a/mlir/lib/Dialect/EmitC/IR/EmitC.cpp
+++ b/mlir/lib/Dialect/EmitC/IR/EmitC.cpp
@@ -882,13 +882,6 @@ LogicalResult emitc::SubscriptOp::verify() {
 }
 
 //===----------------------------------------------------------------------===//
-// TableGen'd op method definitions
-//===----------------------------------------------------------------------===//
-
-#define GET_OP_CLASSES
-#include "mlir/Dialect/EmitC/IR/EmitC.cpp.inc"
-
-//===----------------------------------------------------------------------===//
 // EmitC Enums
 //===----------------------------------------------------------------------===//
 
@@ -987,3 +980,114 @@ LogicalResult mlir::emitc::OpaqueType::verify(
   }
   return success();
 }
+
+//===----------------------------------------------------------------------===//
+// GlobalOp
+//===----------------------------------------------------------------------===//
+static void printEmitCGlobalOpTypeAndInitialValue(OpAsmPrinter &p, GlobalOp op,
+                                                  TypeAttr type,
+                                                  Attribute initialValue) {
+  p << type;
+  if (initialValue) {
+    p << " = ";
+    p.printAttributeWithoutType(initialValue);
+  }
+}
+
+static Type getInitializerTypeForGlobal(Type type) {
+  if (auto array = llvm::dyn_cast<ArrayType>(type))
+    return RankedTensorType::get(array.getShape(), array.getElementType());
+  return type;
+}
+
+static ParseResult
+parseEmitCGlobalOpTypeAndInitialValue(OpAsmParser &parser, TypeAttr &typeAttr,
+                                      Attribute &initialValue) {
+  Type type;
+  if (parser.parseType(type))
+    return failure();
+
+  typeAttr = TypeAttr::get(type);
+
+  if (parser.parseOptionalEqual())
+    return success();
+
+  if (parser.parseAttribute(initialValue, getInitializerTypeForGlobal(type)))
+    return failure();
+
+  if (!llvm::isa<ElementsAttr, IntegerAttr, FloatAttr, emitc::OpaqueAttr>(
+          initialValue))
+    return parser.emitError(parser.getNameLoc())
+           << "initial value should be a integer, float, elements or opaque "
+              "attribute";
+  return success();
+}
+
+LogicalResult GlobalOp::verify() {
+  if (!isSupportedEmitCType(getType())) {
+    return emitOpError("expected valid emitc type");
+  }
+  if (getInitialValue().has_value()) {
+    Attribute initValue = getInitialValue().value();
+    // Check that the type of the initial value is compatible with the type of
+    // the global variable.
+    if (auto elementsAttr = llvm::dyn_cast<ElementsAttr>(initValue)) {
+      auto arrayType = llvm::dyn_cast<ArrayType>(getType());
+      if (!arrayType)
+        return emitOpError("expected array type, but got ") << getType();
+
+      Type initType = elementsAttr.getType();
+      Type tensorType = getInitializerTypeForGlobal(getType());
+      if (initType != tensorType) {
+        return emitOpError("initial value expected to be of type ")
+               << getType() << ", but was of type " << initType;
+      }
+    } else if (auto intAttr = dyn_cast<IntegerAttr>(initValue)) {
+      if (intAttr.getType() != getType()) {
+        return emitOpError("initial value expected to be of type ")
+               << getType() << ", but was of type " << intAttr.getType();
+      }
+    } else if (auto floatAttr = dyn_cast<FloatAttr>(initValue)) {
+      if (floatAttr.getType() != getType()) {
+        return emitOpError("initial value expected to be of type ")
+               << getType() << ", but was of type " << floatAttr.getType();
+      }
+    } else if (!isa<emitc::OpaqueAttr>(initValue)) {
+      return emitOpError("initial value should be a integer, float, elements "
+                         "or opaque attribute, but got ")
+             << initValue;
+    }
+  }
+  if (getStaticSpecifier() && getExternSpecifier()) {
+    return emitOpError("cannot have both static and extern specifiers");
+  }
+  return success();
+}
+
+//===----------------------------------------------------------------------===//
+// GetGlobalOp
+//===----------------------------------------------------------------------===//
+
+LogicalResult
+GetGlobalOp::verifySymbolUses(SymbolTableCollection &symbolTable) {
+  // Verify that the type matches the type of the global variable.
+  auto global =
+      symbolTable.lookupNearestSymbolFrom<GlobalOp>(*this, getNameAttr());
+  if (!global)
+    return emitOpError("'")
+           << getName() << "' does not reference a valid emitc.global";
+
+  Type resultType = getResult().getType();
+  if (global.getType() != resultType)
+    return emitOpError("result type ")
+           << resultType << " does not match type " << global.getType()
+           << " of the global @" << getName();
+  return success();
+}
+
+//===----------------------------------------------------------------------===//
+// TableGen'd op method definitions
+//===----------------------------------------------------------------------===//
+
+#define GET_OP_CLASSES
+#include "mlir/Dialect/EmitC/IR/EmitC.cpp.inc"
diff --git a/mlir/lib/Dialect/GPU/Transforms/ModuleToBinary.cpp b/mlir/lib/Dialect/GPU/Transforms/ModuleToBinary.cpp
index 01613ab5268b..836e939a8295 100644
--- a/mlir/lib/Dialect/GPU/Transforms/ModuleToBinary.cpp
+++ b/mlir/lib/Dialect/GPU/Transforms/ModuleToBinary.cpp
@@ -88,10 +88,7 @@ void GpuModuleToBinaryPass::runOnOperation() {
   TargetOptions targetOptions(toolkitPath, linkFiles, cmdOptions, *targetFormat,
                               lazyTableBuilder);
   if (failed(transformGpuModulesToBinaries(
-          getOperation(),
-          offloadingHandler ? dyn_cast<OffloadingLLVMTranslationAttrInterface>(
-                                  offloadingHandler.getValue())
-                            : OffloadingLLVMTranslationAttrInterface(nullptr),
+          getOperation(), OffloadingLLVMTranslationAttrInterface(nullptr),
           targetOptions)))
     return signalPassFailure();
 }
diff --git a/mlir/lib/Dialect/LLVMIR/IR/LLVMDialect.cpp b/mlir/lib/Dialect/LLVMIR/IR/LLVMDialect.cpp
index 78ff24dae68b..4e06b9c127e7 100644
--- a/mlir/lib/Dialect/LLVMIR/IR/LLVMDialect.cpp
+++ b/mlir/lib/Dialect/LLVMIR/IR/LLVMDialect.cpp
@@ -2831,7 +2831,8 @@ LogicalResult SExtOp::verify() { return verifyExtOp<SExtOp>(*this); }
 
 /// Folds a cast op that can be chained.
 template <typename T>
-static Value foldChainableCast(T castOp, typename T::FoldAdaptor adaptor) {
+static OpFoldResult foldChainableCast(T castOp,
+                                      typename T::FoldAdaptor adaptor) {
   // cast(x : T0, T0) -> x
   if (castOp.getArg().getType() == castOp.getType())
     return castOp.getArg();
diff --git a/mlir/lib/Dialect/LLVMIR/IR/LLVMMemorySlot.cpp b/mlir/lib/Dialect/LLVMIR/IR/LLVMMemorySlot.cpp
index f2ab3eae2c34..6eeb13ebffb5 100644
--- a/mlir/lib/Dialect/LLVMIR/IR/LLVMMemorySlot.cpp
+++ b/mlir/lib/Dialect/LLVMIR/IR/LLVMMemorySlot.cpp
@@ -113,7 +113,7 @@ bool LLVM::LoadOp::loadsFrom(const MemorySlot &slot) {
 bool LLVM::LoadOp::storesTo(const MemorySlot &slot) { return false; }
 
 Value LLVM::LoadOp::getStored(const MemorySlot &slot, RewriterBase &rewriter,
-                              const DataLayout &dataLayout) {
+                              Value reachingDef, const DataLayout &dataLayout) {
   llvm_unreachable("getStored should not be called on LoadOp");
 }
 
@@ -142,9 +142,10 @@ static bool isSupportedTypeForConversion(Type type) {
 }
 
 /// Checks that `rhs` can be converted to `lhs` by a sequence of casts and
-/// truncations.
+/// truncations. Checks for narrowing or widening conversion compatibility
+/// depending on `narrowingConversion`.
 static bool areConversionCompatible(const DataLayout &layout, Type targetType,
-                                    Type srcType) {
+                                    Type srcType, bool narrowingConversion) {
   if (targetType == srcType)
     return true;
 
@@ -152,13 +153,18 @@ static bool areConversionCompatible(const DataLayout &layout, Type targetType,
       !isSupportedTypeForConversion(srcType))
     return false;
 
+  uint64_t targetSize = layout.getTypeSize(targetType);
+  uint64_t srcSize = layout.getTypeSize(srcType);
+
   // Pointer casts will only be sane when the bitsize of both pointer types is
   // the same.
   if (isa<LLVM::LLVMPointerType>(targetType) &&
       isa<LLVM::LLVMPointerType>(srcType))
-    return layout.getTypeSize(targetType) == layout.getTypeSize(srcType);
+    return targetSize == srcSize;
 
-  return layout.getTypeSize(targetType) <= layout.getTypeSize(srcType);
+  if (narrowingConversion)
+    return targetSize <= srcSize;
+  return targetSize >= srcSize;
 }
 
 /// Checks if `dataLayout` describes a little endian layout.
@@ -167,22 +173,49 @@ static bool isBigEndian(const DataLayout &dataLayout) {
   return endiannessStr && endiannessStr == "big";
 }
 
-/// The size of a byte in bits.
-constexpr const static uint64_t kBitsInByte = 8;
+/// Converts a value to an integer type of the same size.
+/// Assumes that the type can be converted.
+static Value castToSameSizedInt(RewriterBase &rewriter, Location loc, Value val,
+                                const DataLayout &dataLayout) {
+  Type type = val.getType();
+  assert(isSupportedTypeForConversion(type) &&
+         "expected value to have a convertible type");
+
+  if (isa<IntegerType>(type))
+    return val;
+
+  uint64_t typeBitSize = dataLayout.getTypeSizeInBits(type);
+  IntegerType valueSizeInteger = rewriter.getIntegerType(typeBitSize);
+
+  if (isa<LLVM::LLVMPointerType>(type))
+    return rewriter.createOrFold<LLVM::PtrToIntOp>(loc, valueSizeInteger, val);
+  return rewriter.createOrFold<LLVM::BitcastOp>(loc, valueSizeInteger, val);
+}
+
+/// Converts a value with an integer type to `targetType`.
+static Value castIntValueToSameSizedType(RewriterBase &rewriter, Location loc,
+                                         Value val, Type targetType) {
+  assert(isa<IntegerType>(val.getType()) &&
+         "expected value to have an integer type");
+  assert(isSupportedTypeForConversion(targetType) &&
+         "expected the target type to be supported for conversions");
+  if (val.getType() == targetType)
+    return val;
+  if (isa<LLVM::LLVMPointerType>(targetType))
+    return rewriter.createOrFold<LLVM::IntToPtrOp>(loc, targetType, val);
+  return rewriter.createOrFold<LLVM::BitcastOp>(loc, targetType, val);
+}
 
-/// Constructs operations that convert `inputValue` into a new value of type
-/// `targetType`. Assumes that this conversion is possible.
-static Value createConversionSequence(RewriterBase &rewriter, Location loc,
-                                      Value srcValue, Type targetType,
-                                      const DataLayout &dataLayout) {
-  // Get the types of the source and target values.
+/// Constructs operations that convert `srcValue` into a new value of type
+/// `targetType`. Assumes the types have the same bitsize.
+static Value castSameSizedTypes(RewriterBase &rewriter, Location loc,
+                                Value srcValue, Type targetType,
+                                const DataLayout &dataLayout) {
   Type srcType = srcValue.getType();
-  assert(areConversionCompatible(dataLayout, targetType, srcType) &&
+  assert(areConversionCompatible(dataLayout, targetType, srcType,
+                                 /*narrowingConversion=*/true) &&
          "expected that the compatibility was checked before");
 
-  uint64_t srcTypeSize = dataLayout.getTypeSize(srcType);
-  uint64_t targetTypeSize = dataLayout.getTypeSize(targetType);
-
   // Nothing has to be done if the types are already the same.
   if (srcType == targetType)
     return srcValue;
@@ -196,48 +229,117 @@ static Value createConversionSequence(RewriterBase &rewriter, Location loc,
     return rewriter.createOrFold<LLVM::AddrSpaceCastOp>(loc, targetType,
                                                         srcValue);
 
-  IntegerType valueSizeInteger =
-      rewriter.getIntegerType(srcTypeSize * kBitsInByte);
-  Value replacement = srcValue;
+  // For all other castable types, casting through integers is necessary.
+  Value replacement = castToSameSizedInt(rewriter, loc, srcValue, dataLayout);
+  return castIntValueToSameSizedType(rewriter, loc, replacement, targetType);
+}
+
+/// Constructs operations that convert `srcValue` into a new value of type
+/// `targetType`. Performs bit-level extraction if the source type is larger
+/// than the target type. Assumes that this conversion is possible.
+static Value createExtractAndCast(RewriterBase &rewriter, Location loc,
+                                  Value srcValue, Type targetType,
+                                  const DataLayout &dataLayout) {
+  // Get the types of the source and target values.
+  Type srcType = srcValue.getType();
+  assert(areConversionCompatible(dataLayout, targetType, srcType,
+                                 /*narrowingConversion=*/true) &&
+         "expected that the compatibility was checked before");
+
+  uint64_t srcTypeSize = dataLayout.getTypeSizeInBits(srcType);
+  uint64_t targetTypeSize = dataLayout.getTypeSizeInBits(targetType);
+  if (srcTypeSize == targetTypeSize)
+    return castSameSizedTypes(rewriter, loc, srcValue, targetType, dataLayout);
 
   // First, cast the value to a same-sized integer type.
-  if (isa<LLVM::LLVMPointerType>(srcType))
-    replacement = rewriter.createOrFold<LLVM::PtrToIntOp>(loc, valueSizeInteger,
-                                                          replacement);
-  else if (replacement.getType() != valueSizeInteger)
-    replacement = rewriter.createOrFold<LLVM::BitcastOp>(loc, valueSizeInteger,
-                                                         replacement);
+  Value replacement = castToSameSizedInt(rewriter, loc, srcValue, dataLayout);
 
   // Truncate the integer if the size of the target is less than the value.
-  if (targetTypeSize != srcTypeSize) {
-    if (isBigEndian(dataLayout)) {
-      uint64_t shiftAmount = (srcTypeSize - targetTypeSize) * kBitsInByte;
-      auto shiftConstant = rewriter.create<LLVM::ConstantOp>(
-          loc, rewriter.getIntegerAttr(srcType, shiftAmount));
-      replacement =
-          rewriter.createOrFold<LLVM::LShrOp>(loc, srcValue, shiftConstant);
-    }
-
-    replacement = rewriter.create<LLVM::TruncOp>(
-        loc, rewriter.getIntegerType(targetTypeSize * kBitsInByte),
-        replacement);
+  if (isBigEndian(dataLayout)) {
+    uint64_t shiftAmount = srcTypeSize - targetTypeSize;
+    auto shiftConstant = rewriter.create<LLVM::ConstantOp>(
+        loc, rewriter.getIntegerAttr(srcType, shiftAmount));
+    replacement =
+        rewriter.createOrFold<LLVM::LShrOp>(loc, srcValue, shiftConstant);
   }
 
+  replacement = rewriter.create<LLVM::TruncOp>(
+      loc, rewriter.getIntegerType(targetTypeSize), replacement);
+
   // Now cast the integer to the actual target type if required.
-  if (isa<LLVM::LLVMPointerType>(targetType))
-    replacement =
-        rewriter.createOrFold<LLVM::IntToPtrOp>(loc, targetType, replacement);
-  else if (replacement.getType() != targetType)
-    replacement =
-        rewriter.createOrFold<LLVM::BitcastOp>(loc, targetType, replacement);
+  return castIntValueToSameSizedType(rewriter, loc, replacement, targetType);
+}
+
+/// Constructs operations that insert the bits of `srcValue` into the
+/// "beginning" of `reachingDef` (beginning is endianness dependent).
+/// Assumes that this conversion is possible.
+static Value createInsertAndCast(RewriterBase &rewriter, Location loc,
+                                 Value srcValue, Value reachingDef,
+                                 const DataLayout &dataLayout) {
+
+  assert(areConversionCompatible(dataLayout, reachingDef.getType(),
+                                 srcValue.getType(),
+                                 /*narrowingConversion=*/false) &&
+         "expected that the compatibility was checked before");
+  uint64_t valueTypeSize = dataLayout.getTypeSizeInBits(srcValue.getType());
+  uint64_t slotTypeSize = dataLayout.getTypeSizeInBits(reachingDef.getType());
+  if (slotTypeSize == valueTypeSize)
+    return castSameSizedTypes(rewriter, loc, srcValue, reachingDef.getType(),
+                              dataLayout);
+
+  // In the case where the store only overwrites parts of the memory,
+  // bit fiddling is required to construct the new value.
+
+  // First convert both values to integers of the same size.
+  Value defAsInt = castToSameSizedInt(rewriter, loc, reachingDef, dataLayout);
+  Value valueAsInt = castToSameSizedInt(rewriter, loc, srcValue, dataLayout);
+  // Extend the value to the size of the reaching definition.
+  valueAsInt =
+      rewriter.createOrFold<LLVM::ZExtOp>(loc, defAsInt.getType(), valueAsInt);
+  uint64_t sizeDifference = slotTypeSize - valueTypeSize;
+  if (isBigEndian(dataLayout)) {
+    // On big endian systems, a store to the base pointer overwrites the most
+    // significant bits. To accomodate for this, the stored value needs to be
+    // shifted into the according position.
+    Value bigEndianShift = rewriter.create<LLVM::ConstantOp>(
+        loc, rewriter.getIntegerAttr(defAsInt.getType(), sizeDifference));
+    valueAsInt =
+        rewriter.createOrFold<LLVM::ShlOp>(loc, valueAsInt, bigEndianShift);
+  }
+
+  // Construct the mask that is used to erase the bits that are overwritten by
+  // the store.
+  APInt maskValue;
+  if (isBigEndian(dataLayout)) {
+    // Build a mask that has the most significant bits set to zero.
+    // Note: This is the same as 2^sizeDifference - 1
+    maskValue = APInt::getAllOnes(sizeDifference).zext(slotTypeSize);
+  } else {
+    // Build a mask that has the least significant bits set to zero.
+    // Note: This is the same as -(2^valueTypeSize)
+    maskValue = APInt::getAllOnes(valueTypeSize).zext(slotTypeSize);
+    maskValue.flipAllBits();
+  }
+
+  // Mask out the affected bits ...
+  Value mask = rewriter.create<LLVM::ConstantOp>(
+      loc, rewriter.getIntegerAttr(defAsInt.getType(), maskValue));
+  Value masked = rewriter.createOrFold<LLVM::AndOp>(loc, defAsInt, mask);
+
+  // ... and combine the result with the new value.
+  Value combined = rewriter.createOrFold<LLVM::OrOp>(loc, masked, valueAsInt);
 
-  return replacement;
+  return castIntValueToSameSizedType(rewriter, loc, combined,
+                                     reachingDef.getType());
 }
 
 Value LLVM::StoreOp::getStored(const MemorySlot &slot, RewriterBase &rewriter,
+                               Value reachingDef,
                                const DataLayout &dataLayout) {
-  return createConversionSequence(rewriter, getLoc(), getValue(), slot.elemType,
-                                  dataLayout);
+  assert(reachingDef && reachingDef.getType() == slot.elemType &&
+         "expected the reaching definition's type to match the slot's type");
+  return createInsertAndCast(rewriter, getLoc(), getValue(), reachingDef,
+                             dataLayout);
 }
 
 bool LLVM::LoadOp::canUsesBeRemoved(
@@ -249,11 +351,10 @@ bool LLVM::LoadOp::canUsesBeRemoved(
   Value blockingUse = (*blockingUses.begin())->get();
   // If the blocking use is the slot ptr itself, there will be enough
   // context to reconstruct the result of the load at removal time, so it can
-  // be removed (provided it loads the exact stored value and is not
-  // volatile).
+  // be removed (provided it is not volatile).
   return blockingUse == slot.ptr && getAddr() == slot.ptr &&
          areConversionCompatible(dataLayout, getResult().getType(),
-                                 slot.elemType) &&
+                                 slot.elemType, /*narrowingConversion=*/true) &&
          !getVolatile_();
 }
 
@@ -263,9 +364,8 @@ DeletionKind LLVM::LoadOp::removeBlockingUses(
     const DataLayout &dataLayout) {
   // `canUsesBeRemoved` checked this blocking use must be the loaded slot
   // pointer.
-  Value newResult =
-      createConversionSequence(rewriter, getLoc(), reachingDefinition,
-                               getResult().getType(), dataLayout);
+  Value newResult = createExtractAndCast(rewriter, getLoc(), reachingDefinition,
+                                         getResult().getType(), dataLayout);
   rewriter.replaceAllUsesWith(getResult(), newResult);
   return DeletionKind::Delete;
 }
@@ -283,7 +383,8 @@ bool LLVM::StoreOp::canUsesBeRemoved(
   return blockingUse == slot.ptr && getAddr() == slot.ptr &&
          getValue() != slot.ptr &&
          areConversionCompatible(dataLayout, slot.elemType,
-                                 getValue().getType()) &&
+                                 getValue().getType(),
+                                 /*narrowingConversion=*/false) &&
          !getVolatile_();
 }
 
@@ -838,6 +939,7 @@ bool LLVM::MemsetOp::storesTo(const MemorySlot &slot) {
 }
 
 Value LLVM::MemsetOp::getStored(const MemorySlot &slot, RewriterBase &rewriter,
+                                Value reachingDef,
                                 const DataLayout &dataLayout) {
   // TODO: Support non-integer types.
   return TypeSwitch<Type, Value>(slot.elemType)
@@ -1149,6 +1251,7 @@ bool LLVM::MemcpyOp::storesTo(const MemorySlot &slot) {
 }
 
 Value LLVM::MemcpyOp::getStored(const MemorySlot &slot, RewriterBase &rewriter,
+                                Value reachingDef,
                                 const DataLayout &dataLayout) {
   return memcpyGetStored(*this, slot, rewriter);
 }
@@ -1199,7 +1302,7 @@ bool LLVM::MemcpyInlineOp::storesTo(const MemorySlot &slot) {
 }
 
 Value LLVM::MemcpyInlineOp::getStored(const MemorySlot &slot,
-                                      RewriterBase &rewriter,
+                                      RewriterBase &rewriter, Value reachingDef,
                                       const DataLayout &dataLayout) {
   return memcpyGetStored(*this, slot, rewriter);
 }
@@ -1252,6 +1355,7 @@ bool LLVM::MemmoveOp::storesTo(const MemorySlot &slot) {
 }
 
 Value LLVM::MemmoveOp::getStored(const MemorySlot &slot, RewriterBase &rewriter,
+                                 Value reachingDef,
                                  const DataLayout &dataLayout) {
   return memcpyGetStored(*this, slot, rewriter);
 }
diff --git a/mlir/lib/Dialect/Linalg/IR/LinalgOps.cpp b/mlir/lib/Dialect/Linalg/IR/LinalgOps.cpp
index e9999c34d0fa..036005ce9d92 100644
--- a/mlir/lib/Dialect/Linalg/IR/LinalgOps.cpp
+++ b/mlir/lib/Dialect/Linalg/IR/LinalgOps.cpp
@@ -395,6 +395,24 @@ public:
       return builder.create<math::FloorOp>(arg.getLoc(), arg);
     case UnaryFn::negf:
       return builder.create<arith::NegFOp>(arg.getLoc(), arg);
+    case UnaryFn::reciprocal: {
+      Attribute oneAttr = builder.getOneAttr(arg.getType());
+      auto one = builder.create<arith::ConstantOp>(arg.getLoc(),
+                                                   ::cast<TypedAttr>(oneAttr));
+      return builder.create<arith::DivFOp>(arg.getLoc(), one, arg);
+    }
+    case UnaryFn::round:
+      return builder.create<math::RoundOp>(arg.getLoc(), arg);
+    case UnaryFn::sqrt:
+      return builder.create<math::SqrtOp>(arg.getLoc(), arg);
+    case UnaryFn::rsqrt:
+      return builder.create<math::RsqrtOp>(arg.getLoc(), arg);
+    case UnaryFn::square:
+      return builder.create<arith::MulFOp>(arg.getLoc(), arg, arg);
+    case UnaryFn::tanh:
+      return builder.create<math::TanhOp>(arg.getLoc(), arg);
+    case UnaryFn::erf:
+      return builder.create<math::ErfOp>(arg.getLoc(), arg);
     }
     llvm_unreachable("unsupported unary function");
   }
@@ -467,6 +485,9 @@ public:
       if (allFloatingPoint)
         return builder.create<arith::MinimumFOp>(arg0.getLoc(), arg0, arg1);
       return builder.create<arith::MinUIOp>(arg0.getLoc(), arg0, arg1);
+    case BinaryFn::powf:
+      assert(allFloatingPoint);
+      return builder.create<math::PowFOp>(arg0.getLoc(), arg0, arg1);
     }
     llvm_unreachable("unsupported binary function");
   }
@@ -2425,8 +2446,8 @@ SoftmaxOp::getTiledImplementation(OpBuilder &builder,
 
 LogicalResult SoftmaxOp::getResultTilePosition(
     OpBuilder &builder, unsigned resultNumber, ArrayRef<OpFoldResult> offsets,
-    ArrayRef<OpFoldResult> sizes, SmallVectorImpl<OpFoldResult> &resultOffsets,
-    SmallVectorImpl<OpFoldResult> &resultSizes) {
+    ArrayRef<OpFoldResult> sizes, SmallVector<OpFoldResult> &resultOffsets,
+    SmallVector<OpFoldResult> &resultSizes) {
   if (resultNumber == 0) {
     resultOffsets.assign(offsets.begin(), offsets.end());
     resultSizes.assign(sizes.begin(), sizes.end());
diff --git a/mlir/lib/Dialect/Linalg/TransformOps/LinalgTransformOps.cpp b/mlir/lib/Dialect/Linalg/TransformOps/LinalgTransformOps.cpp
index 82020c035271..156784f0e674 100644
--- a/mlir/lib/Dialect/Linalg/TransformOps/LinalgTransformOps.cpp
+++ b/mlir/lib/Dialect/Linalg/TransformOps/LinalgTransformOps.cpp
@@ -3417,6 +3417,32 @@ DiagnosedSilenceableFailure transform::TransposeConv2DOp::applyToOne(
 }
 
 //===----------------------------------------------------------------------===//
+// TransposeMatmulOp
+//===----------------------------------------------------------------------===//
+
+DiagnosedSilenceableFailure transform::TransposeMatmulOp::applyToOne(
+    transform::TransformRewriter &rewriter, linalg::LinalgOp target,
+    transform::ApplyToEachResultList &results,
+    transform::TransformState &state) {
+  rewriter.setInsertionPoint(target);
+  bool transposeLHS = getInputToTranspose() == TransposeMatmulInput::lhs;
+  auto maybeTransformed =
+      TypeSwitch<Operation *, FailureOr<Operation *>>(target)
+          .Case([&](linalg::MatmulOp op) {
+            return transposeMatmul(rewriter, op, transposeLHS);
+          })
+          .Case([&](linalg::BatchMatmulOp op) {
+            return transposeBatchMatmul(rewriter, op, transposeLHS);
+          })
+          .Default([&](Operation *op) { return failure(); });
+  if (failed(maybeTransformed))
+    return emitSilenceableFailure(target->getLoc()) << "not supported";
+  // Handle to the new Matmul operation with transposed filters
+  results.push_back(*maybeTransformed);
+  return DiagnosedSilenceableFailure::success();
+}
+
+//===----------------------------------------------------------------------===//
 // InsertSliceToCopyOp
 //===----------------------------------------------------------------------===//
 template <typename OpTy>
diff --git a/mlir/lib/Dialect/Linalg/Transforms/CMakeLists.txt b/mlir/lib/Dialect/Linalg/Transforms/CMakeLists.txt
index 513c54de5d7b..3b5282a09569 100644
--- a/mlir/lib/Dialect/Linalg/Transforms/CMakeLists.txt
+++ b/mlir/lib/Dialect/Linalg/Transforms/CMakeLists.txt
@@ -22,10 +22,12 @@ add_mlir_dialect_library(MLIRLinalgTransforms
   InlineScalarOperands.cpp
   Interchange.cpp
   Loops.cpp
+  TransposeMatmul.cpp
   MeshShardingInterfaceImpl.cpp
   NamedOpConversions.cpp
   Padding.cpp
   Promotion.cpp
+  RuntimeOpVerification.cpp
   Specialize.cpp
   Split.cpp
   SplitReduction.cpp
@@ -59,6 +61,7 @@ add_mlir_dialect_library(MLIRLinalgTransforms
   MLIRFuncDialect
   MLIRFuncToLLVM
   MLIRFuncTransforms
+  MLIRIndexDialect
   MLIRInferTypeOpInterface
   MLIRIR
   MLIRMemRefDialect
diff --git a/mlir/lib/Dialect/Linalg/Transforms/RuntimeOpVerification.cpp b/mlir/lib/Dialect/Linalg/Transforms/RuntimeOpVerification.cpp
new file mode 100644
index 000000000000..b30182dc8407
--- /dev/null
+++ b/mlir/lib/Dialect/Linalg/Transforms/RuntimeOpVerification.cpp
@@ -0,0 +1,135 @@
+//===- RuntimeOpVerification.cpp - Op Verification ------------------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#include "mlir/Dialect/Linalg/Transforms/RuntimeOpVerification.h"
+
+#include "mlir/Dialect/Affine/IR/AffineOps.h"
+#include "mlir/Dialect/Arith/IR/Arith.h"
+#include "mlir/Dialect/Arith/Utils/Utils.h"
+#include "mlir/Dialect/ControlFlow/IR/ControlFlowOps.h"
+#include "mlir/Dialect/Index/IR/IndexAttrs.h"
+#include "mlir/Dialect/Index/IR/IndexDialect.h"
+#include "mlir/Dialect/Index/IR/IndexOps.h"
+#include "mlir/Dialect/Linalg/IR/Linalg.h"
+#include "mlir/Dialect/MemRef/IR/MemRef.h"
+#include "mlir/Dialect/Tensor/IR/Tensor.h"
+#include "mlir/Interfaces/RuntimeVerifiableOpInterface.h"
+
+namespace mlir {
+namespace linalg {
+namespace {
+/// Verify that the runtime sizes of the operands to linalg structured ops are
+/// compatible with the runtime sizes inferred by composing the loop ranges with
+/// the linalg op's indexing maps. This is similar to the verifier except that
+/// here we insert IR to perform the verification at runtime.
+template <typename T>
+struct StructuredOpInterface
+    : public RuntimeVerifiableOpInterface::ExternalModel<
+          StructuredOpInterface<T>, T> {
+  void generateRuntimeVerification(Operation *op, OpBuilder &builder,
+                                   Location loc) const {
+    auto linalgOp = llvm::cast<LinalgOp>(op);
+
+    SmallVector<Range> loopRanges = linalgOp.createLoopRanges(builder, loc);
+    auto [starts, ends, _] = getOffsetsSizesAndStrides(loopRanges);
+
+    auto zero = builder.create<arith::ConstantIndexOp>(loc, 0);
+    auto one = builder.create<arith::ConstantIndexOp>(loc, 1);
+
+    // Subtract one from the loop ends before composing with the indexing map
+    transform(ends, ends.begin(), [&](OpFoldResult end) {
+      auto endValue = getValueOrCreateConstantIndexOp(builder, loc, end);
+      return builder.createOrFold<index::SubOp>(loc, endValue, one);
+    });
+
+    for (OpOperand &opOperand : linalgOp->getOpOperands()) {
+      AffineMap indexingMap = linalgOp.getMatchingIndexingMap(&opOperand);
+      auto startIndices = affine::makeComposedFoldedMultiResultAffineApply(
+          builder, loc, indexingMap, starts);
+      auto endIndices = affine::makeComposedFoldedMultiResultAffineApply(
+          builder, loc, indexingMap, ends);
+
+      for (auto dim : llvm::seq(linalgOp.getRank(&opOperand))) {
+        auto startIndex =
+            getValueOrCreateConstantIndexOp(builder, loc, startIndices[dim]);
+        auto endIndex =
+            getValueOrCreateConstantIndexOp(builder, loc, endIndices[dim]);
+
+        // Generate:
+        //   minIndex = min(startIndex, endIndex)
+        //   assert(minIndex >= 0)
+        // To ensure we do not generate a negative index. We take the minimum of
+        // the start and end indices in order to handle reverse loops such as
+        // `affine_map<(i) -> (3 - i)>`
+        auto min =
+            builder.createOrFold<index::MinSOp>(loc, startIndex, endIndex);
+        auto cmpOp = builder.createOrFold<index::CmpOp>(
+            loc, index::IndexCmpPredicate::SGE, min, zero);
+        auto msg = RuntimeVerifiableOpInterface::generateErrorMessage(
+            linalgOp, "unexpected negative result on dimension #" +
+                          std::to_string(dim) + " of input/output operand #" +
+                          std::to_string(opOperand.getOperandNumber()));
+        builder.createOrFold<cf::AssertOp>(loc, cmpOp, msg);
+
+        // Generate:
+        //   inferredDimSize = max(startIndex, endIndex) + 1
+        //   actualDimSize = dim(operand)
+        //   assert(inferredDimSize <= actualDimSize)
+        // To ensure that we do not index past the bounds of the operands.
+        auto max =
+            builder.createOrFold<index::MaxSOp>(loc, startIndex, endIndex);
+
+        auto inferredDimSize =
+            builder.createOrFold<index::AddOp>(loc, max, one);
+
+        auto actualDimSize =
+            createOrFoldDimOp(builder, loc, opOperand.get(), dim);
+
+        // Similar to the verifier, when the affine expression in the indexing
+        // map is complicated, we just check that the inferred dimension sizes
+        // are in the boundary of the operands' size. Being more precise than
+        // that is difficult.
+        auto predicate = isa<AffineDimExpr>(indexingMap.getResult(dim))
+                             ? index::IndexCmpPredicate::EQ
+                             : index::IndexCmpPredicate::SLE;
+
+        cmpOp = builder.createOrFold<index::CmpOp>(
+            loc, predicate, inferredDimSize, actualDimSize);
+        msg = RuntimeVerifiableOpInterface::generateErrorMessage(
+            linalgOp, "dimension #" + std::to_string(dim) +
+                          " of input/output operand #" +
+                          std::to_string(opOperand.getOperandNumber()) +
+                          " is incompatible with inferred dimension size");
+        builder.createOrFold<cf::AssertOp>(loc, cmpOp, msg);
+      }
+    }
+  }
+};
+
+template <typename... OpTs>
+void attachInterface(MLIRContext *ctx) {
+  (OpTs::template attachInterface<StructuredOpInterface<OpTs>>(*ctx), ...);
+}
+} // namespace
+} // namespace linalg
+} // namespace mlir
+
+void mlir::linalg::registerRuntimeVerifiableOpInterfaceExternalModels(
+    DialectRegistry &registry) {
+  registry.addExtension(+[](MLIRContext *ctx, LinalgDialect *) {
+    attachInterface<
+#define GET_OP_LIST
+#include "mlir/Dialect/Linalg/IR/LinalgStructuredOps.cpp.inc"
+        >(ctx);
+
+    // Load additional dialects of which ops may get created.
+    ctx->loadDialect<affine::AffineDialect, arith::ArithDialect,
+                     cf::ControlFlowDialect, index::IndexDialect,
+                     tensor::TensorDialect>();
+  });
+}
diff --git a/mlir/lib/Dialect/Linalg/Transforms/TilingInterfaceImpl.cpp b/mlir/lib/Dialect/Linalg/Transforms/TilingInterfaceImpl.cpp
index 71e9c3771dcd..bd870d4f982e 100644
--- a/mlir/lib/Dialect/Linalg/Transforms/TilingInterfaceImpl.cpp
+++ b/mlir/lib/Dialect/Linalg/Transforms/TilingInterfaceImpl.cpp
@@ -110,7 +110,7 @@ struct LinalgOpTilingInterface
         }));
   }
 
-  /// Instantiate the tiled implementation of the operation.
+  // Instantiate the tiled implementation of the operation.
   FailureOr<TilingResult>
   getTiledImplementation(Operation *op, OpBuilder &b,
                          ArrayRef<OpFoldResult> offsets,
@@ -132,66 +132,14 @@ struct LinalgOpTilingInterface
     return TilingResult{{tiledOp}, SmallVector<Value>(tiledOp->getResults())};
   }
 
-  void
-  getMappedOffsetAndSize(LinalgOp linalgOp, OpBuilder &b, AffineMap indexingMap,
-                         ArrayRef<OpFoldResult> offsets,
-                         ArrayRef<OpFoldResult> sizes,
-                         SmallVectorImpl<OpFoldResult> &mappedOffsets,
-                         SmallVectorImpl<OpFoldResult> &mappedSizes) const {
-    unsigned numLoops = linalgOp.getNumLoops();
-    auto tilingInterfaceOp = cast<TilingInterface>(linalgOp.getOperation());
-    mappedOffsets.resize(numLoops);
-    mappedSizes.resize(numLoops);
-    if (!indexingMap.isPermutation()) {
-      SmallVector<Range> iterationDomain =
-          tilingInterfaceOp.getIterationDomain(b);
-      for (const auto &&[index, value] : llvm::enumerate(iterationDomain)) {
-        mappedOffsets[index] = value.offset;
-        mappedSizes[index] = value.size;
-      }
-    }
-    for (const auto &&[index, value] :
-         llvm::enumerate(indexingMap.getResults())) {
-      unsigned dimPosition = cast<AffineDimExpr>(value).getPosition();
-      mappedOffsets[dimPosition] = offsets[index];
-      mappedSizes[dimPosition] = sizes[index];
-    }
-  }
-
-  /// Return the details of the output tile generated by the tiled
-  /// implementation.
-  LogicalResult getIterationDomainTileFromOperandTile(
-      Operation *op, OpBuilder &b, unsigned operandNumber,
-      ArrayRef<OpFoldResult> offsets, ArrayRef<OpFoldResult> sizes,
-      SmallVectorImpl<OpFoldResult> &iterDomainOffsets,
-      SmallVectorImpl<OpFoldResult> &iterDomainSizes) const {
-    auto linalgOp = cast<LinalgOp>(op);
-
-    // Check that the indexing map used for the operand is a projected
-    // permutation. This could be relaxed with a more general approach that can
-    // map the offsets and sizes from the operand to iteration space tiles
-    // (filling in full extent for dimensions not used to access the result).
-    AffineMap indexingMap =
-        linalgOp.getMatchingIndexingMap(&op->getOpOperand(operandNumber));
-    if (!indexingMap.isProjectedPermutation()) {
-      return emitError(op->getLoc(),
-                       "unhandled get iter domain position when operand is not "
-                       "accessed using a permuted projection");
-    }
-
-    getMappedOffsetAndSize(linalgOp, b, indexingMap, offsets, sizes,
-                           iterDomainOffsets, iterDomainSizes);
-    return success();
-  }
-
-  /// Return the details of the output tile generated by the tiled
-  /// implementation.
+  // Return the details of the output tile generated by the tiled
+  // implementation.
   LogicalResult
   getResultTilePosition(Operation *op, OpBuilder &b, unsigned resultNumber,
                         ArrayRef<OpFoldResult> offsets,
                         ArrayRef<OpFoldResult> sizes,
-                        SmallVectorImpl<OpFoldResult> &resultOffsets,
-                        SmallVectorImpl<OpFoldResult> &resultSizes) const {
+                        SmallVector<OpFoldResult> &resultOffsets,
+                        SmallVector<OpFoldResult> &resultSizes) const {
     Location loc = op->getLoc();
     LinalgOp linalgOp = cast<LinalgOp>(op);
 
@@ -212,21 +160,6 @@ struct LinalgOpTilingInterface
     return success();
   }
 
-  FailureOr<TilingResult> getTiledImplementationFromOperandTile(
-      Operation *op, OpBuilder &b, unsigned operandNumber,
-      ArrayRef<OpFoldResult> offsets, ArrayRef<OpFoldResult> sizes) const {
-    SmallVector<OpFoldResult> mappedOffsets, mappedSizes;
-    auto tilingInterfaceOp = cast<TilingInterface>(op);
-    if (failed(tilingInterfaceOp.getIterationDomainTileFromOperandTile(
-            b, operandNumber, offsets, sizes, mappedOffsets, mappedSizes))) {
-      return emitError(
-          op->getLoc(),
-          "unable to obtain the iter domain position of the operation.");
-    }
-    return tilingInterfaceOp.getTiledImplementation(b, mappedOffsets,
-                                                    mappedSizes);
-  }
-
   FailureOr<TilingResult>
   generateResultTileValue(Operation *op, OpBuilder &b, unsigned resultNumber,
                           ArrayRef<OpFoldResult> offsets,
@@ -244,16 +177,29 @@ struct LinalgOpTilingInterface
           "unhandled tiled implementation generation when result is not "
           "accessed using a permuted projection");
     }
-    SmallVector<OpFoldResult> mappedOffsets, mappedSizes;
-    getMappedOffsetAndSize(linalgOp, b, indexingMap, offsets, sizes,
-                           mappedOffsets, mappedSizes);
-    auto tilingInterfaceOp = cast<TilingInterface>(op);
-    FailureOr<TilingResult> tilingResult =
-        tilingInterfaceOp.getTiledImplementation(b, mappedOffsets, mappedSizes);
 
-    if (failed(tilingResult))
-      return failure();
+    auto numLoops = linalgOp.getNumLoops();
+    auto tilingInterfaceOp = cast<TilingInterface>(op);
+    SmallVector<OpFoldResult> iterationTileOffsets(numLoops),
+        iterationTileSizes(numLoops);
+    if (!indexingMap.isPermutation()) {
+      SmallVector<Range> iterationDomain =
+          tilingInterfaceOp.getIterationDomain(b);
+      for (const auto &range : llvm::enumerate(iterationDomain)) {
+        iterationTileOffsets[range.index()] = range.value().offset;
+        iterationTileSizes[range.index()] = range.value().size;
+      }
+    }
+    for (const auto &resultExpr : llvm::enumerate(indexingMap.getResults())) {
+      unsigned dimPosition =
+          cast<AffineDimExpr>(resultExpr.value()).getPosition();
+      iterationTileOffsets[dimPosition] = offsets[resultExpr.index()];
+      iterationTileSizes[dimPosition] = sizes[resultExpr.index()];
+    }
 
+    FailureOr<TilingResult> tilingResult =
+        tilingInterfaceOp.getTiledImplementation(b, iterationTileOffsets,
+                                                 iterationTileSizes);
     if (tilingResult->tiledOps.size() != 1)
       return op->emitOpError("failed to generate tiled implementation");
 
diff --git a/mlir/lib/Dialect/Linalg/Transforms/TransposeMatmul.cpp b/mlir/lib/Dialect/Linalg/Transforms/TransposeMatmul.cpp
new file mode 100644
index 000000000000..aa0052ce47fa
--- /dev/null
+++ b/mlir/lib/Dialect/Linalg/Transforms/TransposeMatmul.cpp
@@ -0,0 +1,161 @@
+//===- TransposeMatmul.cpp - Convert Linalg matmul to transposed variants -===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+// This is intended to be a simple high-level (target-agnostic) matmul
+// transposition transformation.
+//===----------------------------------------------------------------------===//
+
+#include "mlir/Dialect/Linalg/Transforms/Transforms.h"
+#include "mlir/IR/PatternMatch.h"
+#include "mlir/Transforms/GreedyPatternRewriteDriver.h"
+
+#define DEBUG_TYPE "linalg-transpose-matmul"
+
+using namespace mlir;
+using namespace mlir::linalg;
+
+/// Pattern to replace
+///
+///   linalg.matmul(a, b)
+///
+/// with
+///
+///   linalg.matmul_transpose_a(linalg.transpose(a), b)
+///
+/// By default the LHS is transposed. Set `transposeLHS=false` to
+/// transpose RHS instead.
+FailureOr<Operation *> mlir::linalg::transposeMatmul(RewriterBase &rewriter,
+                                                     linalg::MatmulOp matmulOp,
+                                                     bool transposeLHS) {
+  if (!bufferization::hasTensorSemantics(matmulOp))
+    return rewriter.notifyMatchFailure(
+        matmulOp, "only matmul ops with tensors are supported");
+
+  Location loc = matmulOp.getLoc();
+  Value input = matmulOp.getInputs()[transposeLHS ? 0 : 1];
+  auto type = cast<ShapedType>(input.getType());
+
+  SmallVector<Value> dynamicDims;
+  if (type.isDynamicDim(1))
+    dynamicDims.push_back(rewriter.create<tensor::DimOp>(loc, input, 1));
+  if (type.isDynamicDim(0))
+    dynamicDims.push_back(rewriter.create<tensor::DimOp>(loc, input, 0));
+
+  ArrayRef<int64_t> shape = type.getShape();
+  Value empty = rewriter.create<tensor::EmptyOp>(
+      loc, ArrayRef<int64_t>{shape[1], shape[0]}, type.getElementType(),
+      dynamicDims);
+  auto transposeOp = rewriter.create<linalg::TransposeOp>(
+      loc, input, empty, ArrayRef<int64_t>{1, 0});
+  Operation *newMatmulOp;
+  if (transposeLHS) {
+    newMatmulOp = rewriter.create<linalg::MatmulTransposeAOp>(
+        loc, matmulOp.getResultTypes(),
+        ValueRange{transposeOp->getResult(0), matmulOp.getInputs()[1]},
+        matmulOp.getOutputs());
+  } else {
+    newMatmulOp = rewriter.create<linalg::MatmulTransposeBOp>(
+        loc, matmulOp.getResultTypes(),
+        ValueRange{matmulOp.getInputs()[0], transposeOp->getResult(0)},
+        matmulOp.getOutputs());
+  }
+  rewriter.replaceOp(matmulOp, newMatmulOp);
+  return newMatmulOp;
+}
+
+/// Pattern to replace
+///
+///   linalg.batch_matmul(a, b)
+///
+/// with
+///
+///   linalg.batch_matmul_transpose_a(linalg.transpose(a), b)
+///
+/// Only the non-batch dimensions are transposed. By default the LHS is
+/// transposed. Set `transposeLHS=false` to transpose RHS instead.
+FailureOr<Operation *>
+mlir::linalg::transposeBatchMatmul(RewriterBase &rewriter,
+                                   linalg::BatchMatmulOp batchMatmulOp,
+                                   bool transposeLHS) {
+  if (!bufferization::hasTensorSemantics(batchMatmulOp))
+    return rewriter.notifyMatchFailure(
+        batchMatmulOp, "only matmul ops with tensors are supported");
+
+  Location loc = batchMatmulOp.getLoc();
+  Value input = batchMatmulOp.getInputs()[transposeLHS ? 0 : 1];
+  auto type = cast<ShapedType>(input.getType());
+
+  SmallVector<Value> dynamicDims;
+  if (type.isDynamicDim(0))
+    dynamicDims.push_back(rewriter.create<tensor::DimOp>(loc, input, 0));
+  if (type.isDynamicDim(2))
+    dynamicDims.push_back(rewriter.create<tensor::DimOp>(loc, input, 2));
+  if (type.isDynamicDim(1))
+    dynamicDims.push_back(rewriter.create<tensor::DimOp>(loc, input, 1));
+
+  ArrayRef<int64_t> shape = type.getShape();
+  Value empty = rewriter.create<tensor::EmptyOp>(
+      loc, ArrayRef<int64_t>{shape[0], shape[2], shape[1]},
+      type.getElementType(), dynamicDims);
+  auto transposeOp = rewriter.create<linalg::TransposeOp>(
+      loc, input, empty, ArrayRef<int64_t>{0, 2, 1});
+  Operation *newMatmulOp;
+  if (transposeLHS) {
+    newMatmulOp = rewriter.create<linalg::BatchMatmulTransposeAOp>(
+        loc, batchMatmulOp.getResultTypes(),
+        ValueRange{transposeOp->getResult(0), batchMatmulOp.getInputs()[1]},
+        batchMatmulOp.getOutputs());
+  } else {
+    newMatmulOp = rewriter.create<linalg::BatchMatmulTransposeBOp>(
+        loc, batchMatmulOp.getResultTypes(),
+        ValueRange{batchMatmulOp.getInputs()[0], transposeOp->getResult(0)},
+        batchMatmulOp.getOutputs());
+  }
+  rewriter.replaceOp(batchMatmulOp, newMatmulOp);
+  return newMatmulOp;
+}
+
+namespace {
+struct TransposeMatmul final : public OpRewritePattern<linalg::MatmulOp> {
+  TransposeMatmul(MLIRContext *ctx, bool transposeLHS)
+      : OpRewritePattern(ctx), transposeLHS(transposeLHS) {}
+
+  LogicalResult matchAndRewrite(linalg::MatmulOp op,
+                                PatternRewriter &rewriter) const override {
+    if (failed(transposeMatmul(rewriter, op, transposeLHS))) {
+      return failure();
+    }
+    return success();
+  }
+
+private:
+  bool transposeLHS;
+};
+
+struct TransposeBatchMatmul final
+    : public OpRewritePattern<linalg::BatchMatmulOp> {
+  TransposeBatchMatmul(MLIRContext *ctx, bool transposeLHS)
+      : OpRewritePattern(ctx), transposeLHS(transposeLHS) {}
+
+  LogicalResult matchAndRewrite(linalg::BatchMatmulOp op,
+                                PatternRewriter &rewriter) const override {
+    if (failed(transposeBatchMatmul(rewriter, op, transposeLHS))) {
+      return failure();
+    }
+    return success();
+  }
+
+private:
+  bool transposeLHS;
+};
+} // namespace
+
+void mlir::linalg::populateTransposeMatmulPatterns(RewritePatternSet &patterns,
+                                                   bool transposeLHS) {
+  patterns.add<TransposeMatmul, TransposeBatchMatmul>(patterns.getContext(),
+                                                      transposeLHS);
+}
diff --git a/mlir/lib/Dialect/Linalg/Transforms/Vectorization.cpp b/mlir/lib/Dialect/Linalg/Transforms/Vectorization.cpp
index e836f0dc63b4..ef9a30be9a01 100644
--- a/mlir/lib/Dialect/Linalg/Transforms/Vectorization.cpp
+++ b/mlir/lib/Dialect/Linalg/Transforms/Vectorization.cpp
@@ -1499,11 +1499,11 @@ vectorizeAsTensorPackOp(RewriterBase &rewriter, tensor::PackOp packOp,
   // If the input vector sizes are not provided, then the vector sizes are
   // determined by the result tensor shape. In case the vector sizes aren't
   // provided, we update the inBounds attribute instead of masking.
-  bool useInBoundsInsteadOfMasking = true;
+  bool useInBoundsInsteadOfMasking = false;
   if (inputVectorSizes.empty()) {
     ArrayRef<int64_t> resultTensorShape = packOp.getDestType().getShape();
     inputVectorSizes = resultTensorShape.take_front(packOp.getSourceRank());
-    useInBoundsInsteadOfMasking = false;
+    useInBoundsInsteadOfMasking = true;
   }
 
   // Create masked TransferReadOp.
@@ -1612,7 +1612,8 @@ vectorizeAsTensorUnpackOp(RewriterBase &rewriter, tensor::UnPackOp unpackOp,
   // to shape of source, then a mask is necessary.
   Value readResult = vector::createReadOrMaskedRead(
       rewriter, loc, unpackOp.getSource(),
-      ArrayRef<int64_t>(readMaskShape.begin(), readMaskShape.end()), padValue);
+      ArrayRef<int64_t>(readMaskShape.begin(), readMaskShape.end()), padValue,
+      /*useInBoundsInsteadOfMasking=*/false);
 
   PackingMetadata packMetadata;
   SmallVector<int64_t> lastDimToInsertPosPerm =
@@ -1669,7 +1670,8 @@ vectorizeAsTensorPadOp(RewriterBase &rewriter, tensor::PadOp padOp,
   (void)status; // prevent unused variable warning on non-assert builds
   assert(succeeded(status) && "failed to reify result shapes");
   auto maskedRead = vector::createReadOrMaskedRead(
-      rewriter, loc, padOp.getSource(), inputVectorSizes, padValue);
+      rewriter, loc, padOp.getSource(), inputVectorSizes, padValue,
+      /*useInBoundsInsteadOfMasking=*/false);
   Operation *write = createWriteOrMaskedWrite(
       rewriter, loc, maskedRead, reifiedReturnShapes[0], inputVectorSizes);
   newResults.push_back(write->getResult(0));
diff --git a/mlir/lib/Dialect/MemRef/IR/MemRefMemorySlot.cpp b/mlir/lib/Dialect/MemRef/IR/MemRefMemorySlot.cpp
index ebbf20f1b76b..958c5f0c8dbc 100644
--- a/mlir/lib/Dialect/MemRef/IR/MemRefMemorySlot.cpp
+++ b/mlir/lib/Dialect/MemRef/IR/MemRefMemorySlot.cpp
@@ -161,6 +161,7 @@ bool memref::LoadOp::loadsFrom(const MemorySlot &slot) {
 bool memref::LoadOp::storesTo(const MemorySlot &slot) { return false; }
 
 Value memref::LoadOp::getStored(const MemorySlot &slot, RewriterBase &rewriter,
+                                Value reachingDef,
                                 const DataLayout &dataLayout) {
   llvm_unreachable("getStored should not be called on LoadOp");
 }
@@ -242,6 +243,7 @@ bool memref::StoreOp::storesTo(const MemorySlot &slot) {
 }
 
 Value memref::StoreOp::getStored(const MemorySlot &slot, RewriterBase &rewriter,
+                                 Value reachingDef,
                                  const DataLayout &dataLayout) {
   return getValue();
 }
diff --git a/mlir/lib/Dialect/MemRef/Transforms/EmulateNarrowType.cpp b/mlir/lib/Dialect/MemRef/Transforms/EmulateNarrowType.cpp
index 4449733f0daf..77c108aab480 100644
--- a/mlir/lib/Dialect/MemRef/Transforms/EmulateNarrowType.cpp
+++ b/mlir/lib/Dialect/MemRef/Transforms/EmulateNarrowType.cpp
@@ -13,7 +13,6 @@
 #include "mlir/Dialect/Arith/Transforms/Passes.h"
 #include "mlir/Dialect/Arith/Utils/Utils.h"
 #include "mlir/Dialect/MemRef/IR/MemRef.h"
-#include "mlir/Dialect/MemRef/Transforms/Passes.h"
 #include "mlir/Dialect/MemRef/Transforms/Transforms.h"
 #include "mlir/Dialect/MemRef/Utils/MemRefUtils.h"
 #include "mlir/Dialect/Vector/IR/VectorOps.h"
@@ -24,7 +23,6 @@
 #include "mlir/Support/MathExtras.h"
 #include "mlir/Transforms/DialectConversion.h"
 #include "llvm/Support/FormatVariadic.h"
-#include "llvm/Support/MathExtras.h"
 #include <cassert>
 #include <type_traits>
 
@@ -430,6 +428,33 @@ struct ConvertMemRefSubview final : OpConversionPattern<memref::SubViewOp> {
   }
 };
 
+//===----------------------------------------------------------------------===//
+// ConvertMemRefCollapseShape
+//===----------------------------------------------------------------------===//
+
+/// Emulating a `memref.collapse_shape` becomes a no-op after emulation given
+/// that we flatten memrefs to a single dimension as part of the emulation and
+/// there is no dimension to collapse any further.
+struct ConvertMemRefCollapseShape final
+    : OpConversionPattern<memref::CollapseShapeOp> {
+  using OpConversionPattern::OpConversionPattern;
+
+  LogicalResult
+  matchAndRewrite(memref::CollapseShapeOp collapseShapeOp, OpAdaptor adaptor,
+                  ConversionPatternRewriter &rewriter) const override {
+    Value srcVal = adaptor.getSrc();
+    auto newTy = dyn_cast<MemRefType>(srcVal.getType());
+    if (!newTy)
+      return failure();
+
+    if (newTy.getRank() != 1)
+      return failure();
+
+    rewriter.replaceOp(collapseShapeOp, srcVal);
+    return success();
+  }
+};
+
 } // end anonymous namespace
 
 //===----------------------------------------------------------------------===//
@@ -442,7 +467,8 @@ void memref::populateMemRefNarrowTypeEmulationPatterns(
 
   // Populate `memref.*` conversion patterns.
   patterns.add<ConvertMemRefAllocation<memref::AllocOp>,
-               ConvertMemRefAllocation<memref::AllocaOp>, ConvertMemRefLoad,
+               ConvertMemRefAllocation<memref::AllocaOp>,
+               ConvertMemRefCollapseShape, ConvertMemRefLoad,
                ConvertMemrefStore, ConvertMemRefAssumeAlignment,
                ConvertMemRefSubview, ConvertMemRefReinterpretCast>(
       typeConverter, patterns.getContext());
diff --git a/mlir/lib/Dialect/MemRef/Transforms/ExpandStridedMetadata.cpp b/mlir/lib/Dialect/MemRef/Transforms/ExpandStridedMetadata.cpp
index 96eb7cfd2db6..585c5b738142 100644
--- a/mlir/lib/Dialect/MemRef/Transforms/ExpandStridedMetadata.cpp
+++ b/mlir/lib/Dialect/MemRef/Transforms/ExpandStridedMetadata.cpp
@@ -550,6 +550,89 @@ getCollapsedStride(memref::CollapseShapeOp collapseShape, OpBuilder &builder,
   return {makeComposedFoldedAffineMin(builder, collapseShape.getLoc(), minMap,
                                       groupStrides)};
 }
+
+/// From `reshape_like(memref, subSizes, subStrides))` compute
+///
+/// \verbatim
+/// baseBuffer, baseOffset, baseSizes, baseStrides =
+///     extract_strided_metadata(memref)
+/// strides#i = baseStrides#i * subStrides#i
+/// sizes = subSizes
+/// \endverbatim
+///
+/// and return {baseBuffer, baseOffset, sizes, strides}
+template <typename ReassociativeReshapeLikeOp>
+static FailureOr<StridedMetadata> resolveReshapeStridedMetadata(
+    RewriterBase &rewriter, ReassociativeReshapeLikeOp reshape,
+    function_ref<SmallVector<OpFoldResult>(
+        ReassociativeReshapeLikeOp, OpBuilder &,
+        ArrayRef<OpFoldResult> /*origSizes*/, unsigned /*groupId*/)>
+        getReshapedSizes,
+    function_ref<SmallVector<OpFoldResult>(
+        ReassociativeReshapeLikeOp, OpBuilder &,
+        ArrayRef<OpFoldResult> /*origSizes*/,
+        ArrayRef<OpFoldResult> /*origStrides*/, unsigned /*groupId*/)>
+        getReshapedStrides) {
+  // Build a plain extract_strided_metadata(memref) from
+  // extract_strided_metadata(reassociative_reshape_like(memref)).
+  Location origLoc = reshape.getLoc();
+  Value source = reshape.getSrc();
+  auto sourceType = cast<MemRefType>(source.getType());
+  unsigned sourceRank = sourceType.getRank();
+
+  auto newExtractStridedMetadata =
+      rewriter.create<memref::ExtractStridedMetadataOp>(origLoc, source);
+
+  // Collect statically known information.
+  auto [strides, offset] = getStridesAndOffset(sourceType);
+  MemRefType reshapeType = reshape.getResultType();
+  unsigned reshapeRank = reshapeType.getRank();
+
+  OpFoldResult offsetOfr =
+      ShapedType::isDynamic(offset)
+          ? getAsOpFoldResult(newExtractStridedMetadata.getOffset())
+          : rewriter.getIndexAttr(offset);
+
+  // Get the special case of 0-D out of the way.
+  if (sourceRank == 0) {
+    SmallVector<OpFoldResult> ones(reshapeRank, rewriter.getIndexAttr(1));
+    return StridedMetadata{newExtractStridedMetadata.getBaseBuffer(), offsetOfr,
+                           /*sizes=*/ones, /*strides=*/ones};
+  }
+
+  SmallVector<OpFoldResult> finalSizes;
+  finalSizes.reserve(reshapeRank);
+  SmallVector<OpFoldResult> finalStrides;
+  finalStrides.reserve(reshapeRank);
+
+  // Compute the reshaped strides and sizes from the base strides and sizes.
+  SmallVector<OpFoldResult> origSizes =
+      getAsOpFoldResult(newExtractStridedMetadata.getSizes());
+  SmallVector<OpFoldResult> origStrides =
+      getAsOpFoldResult(newExtractStridedMetadata.getStrides());
+  unsigned idx = 0, endIdx = reshape.getReassociationIndices().size();
+  for (; idx != endIdx; ++idx) {
+    SmallVector<OpFoldResult> reshapedSizes =
+        getReshapedSizes(reshape, rewriter, origSizes, /*groupId=*/idx);
+    SmallVector<OpFoldResult> reshapedStrides = getReshapedStrides(
+        reshape, rewriter, origSizes, origStrides, /*groupId=*/idx);
+
+    unsigned groupSize = reshapedSizes.size();
+    for (unsigned i = 0; i < groupSize; ++i) {
+      finalSizes.push_back(reshapedSizes[i]);
+      finalStrides.push_back(reshapedStrides[i]);
+    }
+  }
+  assert(((isa<memref::ExpandShapeOp>(reshape) && idx == sourceRank) ||
+          (isa<memref::CollapseShapeOp>(reshape) && idx == reshapeRank)) &&
+         "We should have visited all the input dimensions");
+  assert(finalSizes.size() == reshapeRank &&
+         "We should have populated all the values");
+
+  return StridedMetadata{newExtractStridedMetadata.getBaseBuffer(), offsetOfr,
+                         finalSizes, finalStrides};
+}
+
 /// Replace `baseBuffer, offset, sizes, strides =
 ///              extract_strided_metadata(reshapeLike(memref))`
 /// With
@@ -580,68 +663,65 @@ public:
 
   LogicalResult matchAndRewrite(ReassociativeReshapeLikeOp reshape,
                                 PatternRewriter &rewriter) const override {
-    // Build a plain extract_strided_metadata(memref) from
-    // extract_strided_metadata(reassociative_reshape_like(memref)).
-    Location origLoc = reshape.getLoc();
-    Value source = reshape.getSrc();
-    auto sourceType = cast<MemRefType>(source.getType());
-    unsigned sourceRank = sourceType.getRank();
-
-    auto newExtractStridedMetadata =
-        rewriter.create<memref::ExtractStridedMetadataOp>(origLoc, source);
-
-    // Collect statically known information.
-    auto [strides, offset] = getStridesAndOffset(sourceType);
-    MemRefType reshapeType = reshape.getResultType();
-    unsigned reshapeRank = reshapeType.getRank();
-
-    OpFoldResult offsetOfr =
-        ShapedType::isDynamic(offset)
-            ? getAsOpFoldResult(newExtractStridedMetadata.getOffset())
-            : rewriter.getIndexAttr(offset);
-
-    // Get the special case of 0-D out of the way.
-    if (sourceRank == 0) {
-      SmallVector<OpFoldResult> ones(reshapeRank, rewriter.getIndexAttr(1));
-      auto memrefDesc = rewriter.create<memref::ReinterpretCastOp>(
-          origLoc, reshapeType, newExtractStridedMetadata.getBaseBuffer(),
-          offsetOfr, /*sizes=*/ones, /*strides=*/ones);
-      rewriter.replaceOp(reshape, memrefDesc.getResult());
-      return success();
+    FailureOr<StridedMetadata> stridedMetadata =
+        resolveReshapeStridedMetadata<ReassociativeReshapeLikeOp>(
+            rewriter, reshape, getReshapedSizes, getReshapedStrides);
+    if (failed(stridedMetadata)) {
+      return rewriter.notifyMatchFailure(reshape,
+                                         "failed to resolve reshape metadata");
     }
 
-    SmallVector<OpFoldResult> finalSizes;
-    finalSizes.reserve(reshapeRank);
-    SmallVector<OpFoldResult> finalStrides;
-    finalStrides.reserve(reshapeRank);
-
-    // Compute the reshaped strides and sizes from the base strides and sizes.
-    SmallVector<OpFoldResult> origSizes =
-        getAsOpFoldResult(newExtractStridedMetadata.getSizes());
-    SmallVector<OpFoldResult> origStrides =
-        getAsOpFoldResult(newExtractStridedMetadata.getStrides());
-    unsigned idx = 0, endIdx = reshape.getReassociationIndices().size();
-    for (; idx != endIdx; ++idx) {
-      SmallVector<OpFoldResult> reshapedSizes =
-          getReshapedSizes(reshape, rewriter, origSizes, /*groupId=*/idx);
-      SmallVector<OpFoldResult> reshapedStrides = getReshapedStrides(
-          reshape, rewriter, origSizes, origStrides, /*groupId=*/idx);
-
-      unsigned groupSize = reshapedSizes.size();
-      for (unsigned i = 0; i < groupSize; ++i) {
-        finalSizes.push_back(reshapedSizes[i]);
-        finalStrides.push_back(reshapedStrides[i]);
-      }
+    rewriter.replaceOpWithNewOp<memref::ReinterpretCastOp>(
+        reshape, reshape.getType(), stridedMetadata->basePtr,
+        stridedMetadata->offset, stridedMetadata->sizes,
+        stridedMetadata->strides);
+    return success();
+  }
+};
+
+/// Pattern to replace `extract_strided_metadata(collapse_shape)`
+/// With
+///
+/// \verbatim
+/// baseBuffer, baseOffset, baseSizes, baseStrides =
+///     extract_strided_metadata(memref)
+/// strides#i = baseStrides#i * subSizes#i
+/// offset = baseOffset + sum(subOffset#i * baseStrides#i)
+/// sizes = subSizes
+/// \verbatim
+///
+/// with `baseBuffer`, `offset`, `sizes` and `strides` being
+/// the replacements for the original `extract_strided_metadata`.
+struct ExtractStridedMetadataOpCollapseShapeFolder
+    : OpRewritePattern<memref::ExtractStridedMetadataOp> {
+  using OpRewritePattern::OpRewritePattern;
+
+  LogicalResult matchAndRewrite(memref::ExtractStridedMetadataOp op,
+                                PatternRewriter &rewriter) const override {
+    auto collapseShapeOp =
+        op.getSource().getDefiningOp<memref::CollapseShapeOp>();
+    if (!collapseShapeOp)
+      return failure();
+
+    FailureOr<StridedMetadata> stridedMetadata =
+        resolveReshapeStridedMetadata<memref::CollapseShapeOp>(
+            rewriter, collapseShapeOp, getCollapsedSize, getCollapsedStride);
+    if (failed(stridedMetadata)) {
+      return rewriter.notifyMatchFailure(
+          op,
+          "failed to resolve metadata in terms of source collapse_shape op");
     }
-    assert(((isa<memref::ExpandShapeOp>(reshape) && idx == sourceRank) ||
-            (isa<memref::CollapseShapeOp>(reshape) && idx == reshapeRank)) &&
-           "We should have visited all the input dimensions");
-    assert(finalSizes.size() == reshapeRank &&
-           "We should have populated all the values");
-    auto memrefDesc = rewriter.create<memref::ReinterpretCastOp>(
-        origLoc, reshapeType, newExtractStridedMetadata.getBaseBuffer(),
-        offsetOfr, finalSizes, finalStrides);
-    rewriter.replaceOp(reshape, memrefDesc.getResult());
+
+    Location loc = collapseShapeOp.getLoc();
+    SmallVector<Value> results;
+    results.push_back(stridedMetadata->basePtr);
+    results.push_back(getValueOrCreateConstantIndexOp(rewriter, loc,
+                                                      stridedMetadata->offset));
+    results.append(
+        getValueOrCreateConstantIndexOp(rewriter, loc, stridedMetadata->sizes));
+    results.append(getValueOrCreateConstantIndexOp(rewriter, loc,
+                                                   stridedMetadata->strides));
+    rewriter.replaceOp(op, results);
     return success();
   }
 };
@@ -1018,9 +1098,11 @@ void memref::populateExpandStridedMetadataPatterns(
                              getCollapsedStride>,
                ExtractStridedMetadataOpAllocFolder<memref::AllocOp>,
                ExtractStridedMetadataOpAllocFolder<memref::AllocaOp>,
+               ExtractStridedMetadataOpCollapseShapeFolder,
                ExtractStridedMetadataOpGetGlobalFolder,
                RewriteExtractAlignedPointerAsIndexOfViewLikeOp,
                ExtractStridedMetadataOpReinterpretCastFolder,
+               ExtractStridedMetadataOpSubviewFolder,
                ExtractStridedMetadataOpCastFolder,
                ExtractStridedMetadataOpExtractStridedMetadataFolder>(
       patterns.getContext());
@@ -1030,6 +1112,7 @@ void memref::populateResolveExtractStridedMetadataPatterns(
     RewritePatternSet &patterns) {
   patterns.add<ExtractStridedMetadataOpAllocFolder<memref::AllocOp>,
                ExtractStridedMetadataOpAllocFolder<memref::AllocaOp>,
+               ExtractStridedMetadataOpCollapseShapeFolder,
                ExtractStridedMetadataOpGetGlobalFolder,
                ExtractStridedMetadataOpSubviewFolder,
                RewriteExtractAlignedPointerAsIndexOfViewLikeOp,
diff --git a/mlir/lib/Dialect/MemRef/Transforms/RuntimeOpVerification.cpp b/mlir/lib/Dialect/MemRef/Transforms/RuntimeOpVerification.cpp
index 05b813a3b1e9..450bfa0cec0c 100644
--- a/mlir/lib/Dialect/MemRef/Transforms/RuntimeOpVerification.cpp
+++ b/mlir/lib/Dialect/MemRef/Transforms/RuntimeOpVerification.cpp
@@ -20,25 +20,6 @@
 
 using namespace mlir;
 
-/// Generate an error message string for the given op and the specified error.
-static std::string generateErrorMessage(Operation *op, const std::string &msg) {
-  std::string buffer;
-  llvm::raw_string_ostream stream(buffer);
-  OpPrintingFlags flags;
-  // We may generate a lot of error messages and so we need to ensure the
-  // printing is fast.
-  flags.elideLargeElementsAttrs();
-  flags.printGenericOpForm();
-  flags.skipRegions();
-  flags.useLocalScope();
-  stream << "ERROR: Runtime op verification failed\n";
-  op->print(stream, flags);
-  stream << "\n^ " << msg;
-  stream << "\nLocation: ";
-  op->getLoc().print(stream);
-  return stream.str();
-}
-
 namespace mlir {
 namespace memref {
 namespace {
@@ -62,8 +43,10 @@ struct CastOpInterface
           builder.create<arith::ConstantIndexOp>(loc, resultType.getRank());
       Value isSameRank = builder.create<arith::CmpIOp>(
           loc, arith::CmpIPredicate::eq, srcRank, resultRank);
-      builder.create<cf::AssertOp>(loc, isSameRank,
-                                   generateErrorMessage(op, "rank mismatch"));
+      builder.create<cf::AssertOp>(
+          loc, isSameRank,
+          RuntimeVerifiableOpInterface::generateErrorMessage(op,
+                                                             "rank mismatch"));
     }
 
     // Get source offset and strides. We do not have an op to get offsets and
@@ -101,8 +84,8 @@ struct CastOpInterface
           loc, arith::CmpIPredicate::eq, srcDimSz, resultDimSz);
       builder.create<cf::AssertOp>(
           loc, isSameSz,
-          generateErrorMessage(op, "size mismatch of dim " +
-                                       std::to_string(it.index())));
+          RuntimeVerifiableOpInterface::generateErrorMessage(
+              op, "size mismatch of dim " + std::to_string(it.index())));
     }
 
     // Get result offset and strides.
@@ -119,8 +102,10 @@ struct CastOpInterface
           builder.create<arith::ConstantIndexOp>(loc, resultOffset);
       Value isSameOffset = builder.create<arith::CmpIOp>(
           loc, arith::CmpIPredicate::eq, srcOffset, resultOffsetVal);
-      builder.create<cf::AssertOp>(loc, isSameOffset,
-                                   generateErrorMessage(op, "offset mismatch"));
+      builder.create<cf::AssertOp>(
+          loc, isSameOffset,
+          RuntimeVerifiableOpInterface::generateErrorMessage(
+              op, "offset mismatch"));
     }
 
     // Check strides.
@@ -137,8 +122,8 @@ struct CastOpInterface
           loc, arith::CmpIPredicate::eq, srcStride, resultStrideVal);
       builder.create<cf::AssertOp>(
           loc, isSameStride,
-          generateErrorMessage(op, "stride mismatch of dim " +
-                                       std::to_string(it.index())));
+          RuntimeVerifiableOpInterface::generateErrorMessage(
+              op, "stride mismatch of dim " + std::to_string(it.index())));
     }
   }
 };
@@ -178,7 +163,9 @@ struct LoadStoreOpInterface
                 : andOp;
     }
     builder.create<cf::AssertOp>(
-        loc, assertCond, generateErrorMessage(op, "out-of-bounds access"));
+        loc, assertCond,
+        RuntimeVerifiableOpInterface::generateErrorMessage(
+            op, "out-of-bounds access"));
   }
 };
 
@@ -248,7 +235,7 @@ struct ReinterpretCastOpInterface
 
     builder.create<cf::AssertOp>(
         loc, assertCond,
-        generateErrorMessage(
+        RuntimeVerifiableOpInterface::generateErrorMessage(
             op,
             "result of reinterpret_cast is out-of-bounds of the base memref"));
   }
@@ -293,8 +280,8 @@ struct SubViewOpInterface
 
     builder.create<cf::AssertOp>(
         loc, assertCond,
-        generateErrorMessage(op,
-                             "subview is out-of-bounds of the base memref"));
+        RuntimeVerifiableOpInterface::generateErrorMessage(
+            op, "subview is out-of-bounds of the base memref"));
   }
 };
 
@@ -334,8 +321,9 @@ struct ExpandShapeOpInterface
           builder.create<arith::ConstantIndexOp>(loc, 0));
       builder.create<cf::AssertOp>(
           loc, isModZero,
-          generateErrorMessage(op, "static result dims in reassoc group do not "
-                                   "divide src dim evenly"));
+          RuntimeVerifiableOpInterface::generateErrorMessage(
+              op, "static result dims in reassoc group do not "
+                  "divide src dim evenly"));
     }
   }
 };
diff --git a/mlir/lib/Dialect/OpenMP/IR/OpenMPDialect.cpp b/mlir/lib/Dialect/OpenMP/IR/OpenMPDialect.cpp
index 528a0d05b101..f60668dd0cf9 100644
--- a/mlir/lib/Dialect/OpenMP/IR/OpenMPDialect.cpp
+++ b/mlir/lib/Dialect/OpenMP/IR/OpenMPDialect.cpp
@@ -1484,86 +1484,72 @@ LogicalResult SingleOp::verify() {
 // WsloopOp
 //===----------------------------------------------------------------------===//
 
-/// loop-control ::= `(` ssa-id-list `)` `:` type `=`  loop-bounds
-/// loop-bounds := `(` ssa-id-list `)` to `(` ssa-id-list `)` inclusive? steps
-/// steps := `step` `(`ssa-id-list`)`
 ParseResult
 parseWsloop(OpAsmParser &parser, Region &region,
-            SmallVectorImpl<OpAsmParser::UnresolvedOperand> &lowerBound,
-            SmallVectorImpl<OpAsmParser::UnresolvedOperand> &upperBound,
-            SmallVectorImpl<OpAsmParser::UnresolvedOperand> &steps,
-            SmallVectorImpl<Type> &loopVarTypes,
             SmallVectorImpl<OpAsmParser::UnresolvedOperand> &reductionOperands,
-            SmallVectorImpl<Type> &reductionTypes, ArrayAttr &reductionSymbols,
-            UnitAttr &inclusive) {
-
+            SmallVectorImpl<Type> &reductionTypes,
+            ArrayAttr &reductionSymbols) {
   // Parse an optional reduction clause
   llvm::SmallVector<OpAsmParser::Argument> privates;
-  bool hasReduction = succeeded(parser.parseOptionalKeyword("reduction")) &&
-                      succeeded(parseClauseWithRegionArgs(
-                          parser, region, reductionOperands, reductionTypes,
-                          reductionSymbols, privates));
-
-  if (parser.parseKeyword("for"))
-    return failure();
-
-  // Parse an opening `(` followed by induction variables followed by `)`
-  SmallVector<OpAsmParser::Argument> ivs;
-  Type loopVarType;
-  if (parser.parseArgumentList(ivs, OpAsmParser::Delimiter::Paren) ||
-      parser.parseColonType(loopVarType) ||
-      // Parse loop bounds.
-      parser.parseEqual() ||
-      parser.parseOperandList(lowerBound, ivs.size(),
-                              OpAsmParser::Delimiter::Paren) ||
-      parser.parseKeyword("to") ||
-      parser.parseOperandList(upperBound, ivs.size(),
-                              OpAsmParser::Delimiter::Paren))
-    return failure();
-
-  if (succeeded(parser.parseOptionalKeyword("inclusive")))
-    inclusive = UnitAttr::get(parser.getBuilder().getContext());
-
-  // Parse step values.
-  if (parser.parseKeyword("step") ||
-      parser.parseOperandList(steps, ivs.size(), OpAsmParser::Delimiter::Paren))
-    return failure();
-
-  // Now parse the body.
-  loopVarTypes = SmallVector<Type>(ivs.size(), loopVarType);
-  for (auto &iv : ivs)
-    iv.type = loopVarType;
-
-  SmallVector<OpAsmParser::Argument> regionArgs{ivs};
-  if (hasReduction)
-    llvm::copy(privates, std::back_inserter(regionArgs));
-
-  return parser.parseRegion(region, regionArgs);
+  if (succeeded(parser.parseOptionalKeyword("reduction"))) {
+    if (failed(parseClauseWithRegionArgs(parser, region, reductionOperands,
+                                         reductionTypes, reductionSymbols,
+                                         privates)))
+      return failure();
+  }
+  return parser.parseRegion(region, privates);
 }
 
 void printWsloop(OpAsmPrinter &p, Operation *op, Region &region,
-                 ValueRange lowerBound, ValueRange upperBound, ValueRange steps,
-                 TypeRange loopVarTypes, ValueRange reductionOperands,
-                 TypeRange reductionTypes, ArrayAttr reductionSymbols,
-                 UnitAttr inclusive) {
+                 ValueRange reductionOperands, TypeRange reductionTypes,
+                 ArrayAttr reductionSymbols) {
   if (reductionSymbols) {
-    auto reductionArgs =
-        region.front().getArguments().drop_front(loopVarTypes.size());
+    auto reductionArgs = region.front().getArguments();
     printClauseWithRegionArgs(p, op, reductionArgs, "reduction",
                               reductionOperands, reductionTypes,
                               reductionSymbols);
   }
-
-  p << " for ";
-  auto args = region.front().getArguments().drop_back(reductionOperands.size());
-  p << " (" << args << ") : " << args[0].getType() << " = (" << lowerBound
-    << ") to (" << upperBound << ") ";
-  if (inclusive)
-    p << "inclusive ";
-  p << "step (" << steps << ") ";
   p.printRegion(region, /*printEntryBlockArgs=*/false);
 }
 
+void WsloopOp::build(OpBuilder &builder, OperationState &state,
+                     ArrayRef<NamedAttribute> attributes) {
+  build(builder, state, /*linear_vars=*/ValueRange(),
+        /*linear_step_vars=*/ValueRange(), /*reduction_vars=*/ValueRange(),
+        /*reductions=*/nullptr, /*schedule_val=*/nullptr,
+        /*schedule_chunk_var=*/nullptr, /*schedule_modifier=*/nullptr,
+        /*simd_modifier=*/false, /*nowait=*/false, /*byref=*/false,
+        /*ordered_val=*/nullptr, /*order_val=*/nullptr);
+  state.addAttributes(attributes);
+}
+
+void WsloopOp::build(OpBuilder &builder, OperationState &state,
+                     const WsloopClauseOps &clauses) {
+  MLIRContext *ctx = builder.getContext();
+  // TODO: Store clauses in op: allocateVars, allocatorVars, privateVars,
+  // privatizers.
+  WsloopOp::build(
+      builder, state, clauses.linearVars, clauses.linearStepVars,
+      clauses.reductionVars, makeArrayAttr(ctx, clauses.reductionDeclSymbols),
+      clauses.scheduleValAttr, clauses.scheduleChunkVar,
+      clauses.scheduleModAttr, clauses.scheduleSimdAttr, clauses.nowaitAttr,
+      clauses.reductionByRefAttr, clauses.orderedAttr, clauses.orderAttr);
+}
+
+LogicalResult WsloopOp::verify() {
+  if (!isWrapper())
+    return emitOpError() << "must be a loop wrapper";
+
+  if (LoopWrapperInterface nested = getNestedWrapper()) {
+    // Check for the allowed leaf constructs that may appear in a composite
+    // construct directly after DO/FOR.
+    if (!isa<SimdOp>(nested))
+      return emitError() << "only supported nested wrapper is 'omp.simd'";
+  }
+
+  return verifyReductionVarList(*this, getReductions(), getReductionVars());
+}
+
 //===----------------------------------------------------------------------===//
 // Simd construct [2.9.3.1]
 //===----------------------------------------------------------------------===//
@@ -1948,42 +1934,6 @@ void LoopNestOp::gatherWrappers(
 }
 
 //===----------------------------------------------------------------------===//
-// WsloopOp
-//===----------------------------------------------------------------------===//
-
-void WsloopOp::build(OpBuilder &builder, OperationState &state,
-                     ValueRange lowerBound, ValueRange upperBound,
-                     ValueRange step, ArrayRef<NamedAttribute> attributes) {
-  build(builder, state, lowerBound, upperBound, step,
-        /*linear_vars=*/ValueRange(),
-        /*linear_step_vars=*/ValueRange(), /*reduction_vars=*/ValueRange(),
-        /*reductions=*/nullptr, /*schedule_val=*/nullptr,
-        /*schedule_chunk_var=*/nullptr, /*schedule_modifier=*/nullptr,
-        /*simd_modifier=*/false, /*nowait=*/false, /*byref=*/false,
-        /*ordered_val=*/nullptr,
-        /*order_val=*/nullptr, /*inclusive=*/false);
-  state.addAttributes(attributes);
-}
-
-void WsloopOp::build(OpBuilder &builder, OperationState &state,
-                     const WsloopClauseOps &clauses) {
-  MLIRContext *ctx = builder.getContext();
-  // TODO Store clauses in op: allocateVars, allocatorVars, privateVars,
-  // privatizers.
-  WsloopOp::build(
-      builder, state, clauses.loopLBVar, clauses.loopUBVar, clauses.loopStepVar,
-      clauses.linearVars, clauses.linearStepVars, clauses.reductionVars,
-      makeArrayAttr(ctx, clauses.reductionDeclSymbols), clauses.scheduleValAttr,
-      clauses.scheduleChunkVar, clauses.scheduleModAttr,
-      clauses.scheduleSimdAttr, clauses.nowaitAttr, clauses.reductionByRefAttr,
-      clauses.orderedAttr, clauses.orderAttr, clauses.loopInclusiveAttr);
-}
-
-LogicalResult WsloopOp::verify() {
-  return verifyReductionVarList(*this, getReductions(), getReductionVars());
-}
-
-//===----------------------------------------------------------------------===//
 // Critical construct (2.17.1)
 //===----------------------------------------------------------------------===//
 
@@ -2014,6 +1964,39 @@ LogicalResult CriticalOp::verifySymbolUses(SymbolTableCollection &symbolTable) {
 // Ordered construct
 //===----------------------------------------------------------------------===//
 
+static LogicalResult verifyOrderedParent(Operation &op) {
+  bool hasRegion = op.getNumRegions() > 0;
+  auto loopOp = op.getParentOfType<LoopNestOp>();
+  if (!loopOp) {
+    if (hasRegion)
+      return success();
+
+    // TODO: Consider if this needs to be the case only for the standalone
+    // variant of the ordered construct.
+    return op.emitOpError() << "must be nested inside of a loop";
+  }
+
+  Operation *wrapper = loopOp->getParentOp();
+  if (auto wsloopOp = dyn_cast<WsloopOp>(wrapper)) {
+    IntegerAttr orderedAttr = wsloopOp.getOrderedValAttr();
+    if (!orderedAttr)
+      return op.emitOpError() << "the enclosing worksharing-loop region must "
+                                 "have an ordered clause";
+
+    if (hasRegion && orderedAttr.getInt() != 0)
+      return op.emitOpError() << "the enclosing loop's ordered clause must not "
+                                 "have a parameter present";
+
+    if (!hasRegion && orderedAttr.getInt() == 0)
+      return op.emitOpError() << "the enclosing loop's ordered clause must "
+                                 "have a parameter present";
+  } else if (!isa<SimdOp>(wrapper)) {
+    return op.emitOpError() << "must be nested inside of a worksharing, simd "
+                               "or worksharing simd loop";
+  }
+  return success();
+}
+
 void OrderedOp::build(OpBuilder &builder, OperationState &state,
                       const OrderedOpClauseOps &clauses) {
   OrderedOp::build(builder, state, clauses.doacrossDependTypeAttr,
@@ -2021,14 +2004,11 @@ void OrderedOp::build(OpBuilder &builder, OperationState &state,
 }
 
 LogicalResult OrderedOp::verify() {
-  auto container = (*this)->getParentOfType<WsloopOp>();
-  if (!container || !container.getOrderedValAttr() ||
-      container.getOrderedValAttr().getInt() == 0)
-    return emitOpError() << "ordered depend directive must be closely "
-                         << "nested inside a worksharing-loop with ordered "
-                         << "clause with parameter present";
-
-  if (container.getOrderedValAttr().getInt() != (int64_t)*getNumLoopsVal())
+  if (failed(verifyOrderedParent(**this)))
+    return failure();
+
+  auto wrapper = (*this)->getParentOfType<WsloopOp>();
+  if (!wrapper || *wrapper.getOrderedVal() != *getNumLoopsVal())
     return emitOpError() << "number of variables in depend clause does not "
                          << "match number of iteration variables in the "
                          << "doacross loop";
@@ -2046,15 +2026,7 @@ LogicalResult OrderedRegionOp::verify() {
   if (getSimd())
     return failure();
 
-  if (auto container = (*this)->getParentOfType<WsloopOp>()) {
-    if (!container.getOrderedValAttr() ||
-        container.getOrderedValAttr().getInt() != 0)
-      return emitOpError() << "ordered region must be closely nested inside "
-                           << "a worksharing-loop region with an ordered "
-                           << "clause without parameter present";
-  }
-
-  return success();
+  return verifyOrderedParent(**this);
 }
 
 //===----------------------------------------------------------------------===//
@@ -2199,15 +2171,19 @@ LogicalResult CancelOp::verify() {
                          << "inside a parallel region";
   }
   if (cct == ClauseCancellationConstructType::Loop) {
-    if (!isa<WsloopOp>(parentOp)) {
-      return emitOpError() << "cancel loop must appear "
-                           << "inside a worksharing-loop region";
+    auto loopOp = dyn_cast<LoopNestOp>(parentOp);
+    auto wsloopOp = llvm::dyn_cast_if_present<WsloopOp>(
+        loopOp ? loopOp->getParentOp() : nullptr);
+
+    if (!wsloopOp) {
+      return emitOpError()
+             << "cancel loop must appear inside a worksharing-loop region";
     }
-    if (cast<WsloopOp>(parentOp).getNowaitAttr()) {
+    if (wsloopOp.getNowaitAttr()) {
       return emitError() << "A worksharing construct that is canceled "
                          << "must not have a nowait clause";
     }
-    if (cast<WsloopOp>(parentOp).getOrderedValAttr()) {
+    if (wsloopOp.getOrderedValAttr()) {
       return emitError() << "A worksharing construct that is canceled "
                          << "must not have an ordered clause";
     }
@@ -2245,7 +2221,7 @@ LogicalResult CancellationPointOp::verify() {
                          << "inside a parallel region";
   }
   if ((cct == ClauseCancellationConstructType::Loop) &&
-      !isa<WsloopOp>(parentOp)) {
+      (!isa<LoopNestOp>(parentOp) || !isa<WsloopOp>(parentOp->getParentOp()))) {
     return emitOpError() << "cancellation point loop must appear "
                          << "inside a worksharing-loop region";
   }
diff --git a/mlir/lib/Dialect/Polynomial/IR/CMakeLists.txt b/mlir/lib/Dialect/Polynomial/IR/CMakeLists.txt
index 7f5b3255d5d9..d6e703b8b359 100644
--- a/mlir/lib/Dialect/Polynomial/IR/CMakeLists.txt
+++ b/mlir/lib/Dialect/Polynomial/IR/CMakeLists.txt
@@ -16,4 +16,5 @@ add_mlir_dialect_library(MLIRPolynomialDialect
   MLIRSupport
   MLIRDialect
   MLIRIR
+  MLIRInferTypeOpInterface
   )
diff --git a/mlir/lib/Dialect/Polynomial/IR/PolynomialAttributes.cpp b/mlir/lib/Dialect/Polynomial/IR/PolynomialAttributes.cpp
index ee09c73bb3c4..f1ec2be72a33 100644
--- a/mlir/lib/Dialect/Polynomial/IR/PolynomialAttributes.cpp
+++ b/mlir/lib/Dialect/Polynomial/IR/PolynomialAttributes.cpp
@@ -172,7 +172,7 @@ Attribute RingAttr::parse(AsmParser &parser, Type type) {
     if (failed(parser.parseEqual()))
       return {};
 
-    IntegerType iType = ty.dyn_cast<IntegerType>();
+    IntegerType iType = mlir::dyn_cast<IntegerType>(ty);
     if (!iType) {
       parser.emitError(parser.getCurrentLocation(),
                        "coefficientType must specify an integer type");
diff --git a/mlir/lib/Dialect/Polynomial/IR/PolynomialDialect.cpp b/mlir/lib/Dialect/Polynomial/IR/PolynomialDialect.cpp
index a672a59b8a46..825b80d70f80 100644
--- a/mlir/lib/Dialect/Polynomial/IR/PolynomialDialect.cpp
+++ b/mlir/lib/Dialect/Polynomial/IR/PolynomialDialect.cpp
@@ -8,9 +8,18 @@
 
 #include "mlir/Dialect/Polynomial/IR/Polynomial.h"
 
+#include "mlir/Dialect/Arith/IR/Arith.h"
 #include "mlir/Dialect/Polynomial/IR/PolynomialAttributes.h"
 #include "mlir/Dialect/Polynomial/IR/PolynomialOps.h"
 #include "mlir/Dialect/Polynomial/IR/PolynomialTypes.h"
+#include "mlir/IR/Builders.h"
+#include "mlir/IR/BuiltinOps.h"
+#include "mlir/IR/BuiltinTypes.h"
+#include "mlir/IR/Dialect.h"
+#include "mlir/IR/PatternMatch.h"
+#include "mlir/Interfaces/InferTypeOpInterface.h"
+#include "mlir/Support/LogicalResult.h"
+#include "llvm/ADT/APInt.h"
 #include "llvm/ADT/TypeSwitch.h"
 
 using namespace mlir;
diff --git a/mlir/lib/Dialect/Polynomial/IR/PolynomialOps.cpp b/mlir/lib/Dialect/Polynomial/IR/PolynomialOps.cpp
index 96c59a28b8fd..8e2bb5f27dc6 100644
--- a/mlir/lib/Dialect/Polynomial/IR/PolynomialOps.cpp
+++ b/mlir/lib/Dialect/Polynomial/IR/PolynomialOps.cpp
@@ -6,10 +6,101 @@
 //
 //===----------------------------------------------------------------------===//
 
+#include "mlir/Dialect/Polynomial/IR/PolynomialOps.h"
 #include "mlir/Dialect/Polynomial/IR/Polynomial.h"
+#include "mlir/Dialect/Polynomial/IR/PolynomialAttributes.h"
+#include "mlir/Dialect/Polynomial/IR/PolynomialTypes.h"
+#include "mlir/IR/Builders.h"
+#include "mlir/IR/BuiltinTypes.h"
+#include "mlir/IR/Dialect.h"
+#include "mlir/Support/LogicalResult.h"
+#include "llvm/ADT/APInt.h"
 
 using namespace mlir;
 using namespace mlir::polynomial;
 
-#define GET_OP_CLASSES
-#include "mlir/Dialect/Polynomial/IR/Polynomial.cpp.inc"
+void FromTensorOp::build(OpBuilder &builder, OperationState &result,
+                         Value input, RingAttr ring) {
+  TensorType tensorType = dyn_cast<TensorType>(input.getType());
+  auto bitWidth = tensorType.getElementTypeBitWidth();
+  APInt cmod(1 + bitWidth, 1);
+  cmod = cmod << bitWidth;
+  Type resultType = PolynomialType::get(builder.getContext(), ring);
+  build(builder, result, resultType, input);
+}
+
+LogicalResult FromTensorOp::verify() {
+  ArrayRef<int64_t> tensorShape = getInput().getType().getShape();
+  RingAttr ring = getOutput().getType().getRing();
+  unsigned polyDegree = ring.getPolynomialModulus().getPolynomial().getDegree();
+  bool compatible = tensorShape.size() == 1 && tensorShape[0] <= polyDegree;
+  if (!compatible) {
+    InFlightDiagnostic diag = emitOpError()
+                              << "input type " << getInput().getType()
+                              << " does not match output type "
+                              << getOutput().getType();
+    diag.attachNote() << "the input type must be a tensor of shape [d] where d "
+                         "is at most the degree of the polynomialModulus of "
+                         "the output type's ring attribute";
+    return diag;
+  }
+
+  APInt coefficientModulus = ring.getCoefficientModulus().getValue();
+  unsigned cmodBitWidth = coefficientModulus.ceilLogBase2();
+  unsigned inputBitWidth = getInput().getType().getElementTypeBitWidth();
+
+  if (inputBitWidth > cmodBitWidth) {
+    InFlightDiagnostic diag = emitOpError()
+                              << "input tensor element type "
+                              << getInput().getType().getElementType()
+                              << " is too large to fit in the coefficients of "
+                              << getOutput().getType();
+    diag.attachNote() << "the input tensor's elements must be rescaled"
+                         " to fit before using from_tensor";
+    return diag;
+  }
+
+  return success();
+}
+
+LogicalResult ToTensorOp::verify() {
+  ArrayRef<int64_t> tensorShape = getOutput().getType().getShape();
+  unsigned polyDegree = getInput()
+                            .getType()
+                            .getRing()
+                            .getPolynomialModulus()
+                            .getPolynomial()
+                            .getDegree();
+  bool compatible = tensorShape.size() == 1 && tensorShape[0] == polyDegree;
+
+  if (compatible)
+    return success();
+
+  InFlightDiagnostic diag =
+      emitOpError() << "input type " << getInput().getType()
+                    << " does not match output type " << getOutput().getType();
+  diag.attachNote() << "the output type must be a tensor of shape [d] where d "
+                       "is at most the degree of the polynomialModulus of "
+                       "the input type's ring attribute";
+  return diag;
+}
+
+LogicalResult MulScalarOp::verify() {
+  Type argType = getPolynomial().getType();
+  PolynomialType polyType;
+
+  if (auto shapedPolyType = dyn_cast<ShapedType>(argType)) {
+    polyType = cast<PolynomialType>(shapedPolyType.getElementType());
+  } else {
+    polyType = cast<PolynomialType>(argType);
+  }
+
+  Type coefficientType = polyType.getRing().getCoefficientType();
+
+  if (coefficientType != getScalar().getType())
+    return emitOpError() << "polynomial coefficient type " << coefficientType
+                         << " does not match scalar type "
+                         << getScalar().getType();
+
+  return success();
+}
diff --git a/mlir/lib/Dialect/SCF/TransformOps/SCFTransformOps.cpp b/mlir/lib/Dialect/SCF/TransformOps/SCFTransformOps.cpp
index 7e4faf8b73af..69f83d8bd70d 100644
--- a/mlir/lib/Dialect/SCF/TransformOps/SCFTransformOps.cpp
+++ b/mlir/lib/Dialect/SCF/TransformOps/SCFTransformOps.cpp
@@ -69,16 +69,12 @@ transform::ForallToForOp::apply(transform::TransformRewriter &rewriter,
     return diag;
   }
 
-  rewriter.setInsertionPoint(target);
-
   if (!target.getOutputs().empty()) {
     return emitSilenceableError()
            << "unsupported shared outputs (didn't bufferize?)";
   }
 
   SmallVector<OpFoldResult> lbs = target.getMixedLowerBound();
-  SmallVector<OpFoldResult> ubs = target.getMixedUpperBound();
-  SmallVector<OpFoldResult> steps = target.getMixedStep();
 
   if (getNumResults() != lbs.size()) {
     DiagnosedSilenceableFailure diag =
@@ -89,28 +85,15 @@ transform::ForallToForOp::apply(transform::TransformRewriter &rewriter,
     return diag;
   }
 
-  auto loc = target.getLoc();
-  SmallVector<Value> ivs;
-  for (auto &&[lb, ub, step] : llvm::zip(lbs, ubs, steps)) {
-    Value lbValue = getValueOrCreateConstantIndexOp(rewriter, loc, lb);
-    Value ubValue = getValueOrCreateConstantIndexOp(rewriter, loc, ub);
-    Value stepValue = getValueOrCreateConstantIndexOp(rewriter, loc, step);
-    auto loop = rewriter.create<scf::ForOp>(
-        loc, lbValue, ubValue, stepValue, ValueRange(),
-        [](OpBuilder &, Location, Value, ValueRange) {});
-    ivs.push_back(loop.getInductionVar());
-    rewriter.setInsertionPointToStart(loop.getBody());
-    rewriter.create<scf::YieldOp>(loc);
-    rewriter.setInsertionPointToStart(loop.getBody());
+  SmallVector<Operation *> opResults;
+  if (failed(scf::forallToForLoop(rewriter, target, &opResults))) {
+    DiagnosedSilenceableFailure diag = emitSilenceableError()
+                                       << "failed to convert forall into for";
+    return diag;
   }
-  rewriter.eraseOp(target.getBody()->getTerminator());
-  rewriter.inlineBlockBefore(target.getBody(), &*rewriter.getInsertionPoint(),
-                             ivs);
-  rewriter.eraseOp(target);
-
-  for (auto &&[i, iv] : llvm::enumerate(ivs)) {
-    results.set(cast<OpResult>(getTransformed()[i]),
-                {iv.getParentBlock()->getParentOp()});
+
+  for (auto &&[i, res] : llvm::enumerate(opResults)) {
+    results.set(cast<OpResult>(getTransformed()[i]), {res});
   }
   return DiagnosedSilenceableFailure::success();
 }
diff --git a/mlir/lib/Dialect/SCF/Transforms/CMakeLists.txt b/mlir/lib/Dialect/SCF/Transforms/CMakeLists.txt
index a2925aef17ca..e7671c9cc28f 100644
--- a/mlir/lib/Dialect/SCF/Transforms/CMakeLists.txt
+++ b/mlir/lib/Dialect/SCF/Transforms/CMakeLists.txt
@@ -2,6 +2,7 @@ add_mlir_dialect_library(MLIRSCFTransforms
   BufferDeallocationOpInterfaceImpl.cpp
   BufferizableOpInterfaceImpl.cpp
   Bufferize.cpp
+  ForallToFor.cpp
   ForToWhile.cpp
   LoopCanonicalization.cpp
   LoopPipelining.cpp
diff --git a/mlir/lib/Dialect/SCF/Transforms/ForallToFor.cpp b/mlir/lib/Dialect/SCF/Transforms/ForallToFor.cpp
new file mode 100644
index 000000000000..198cb2e6cc69
--- /dev/null
+++ b/mlir/lib/Dialect/SCF/Transforms/ForallToFor.cpp
@@ -0,0 +1,79 @@
+//===- ForallToFor.cpp - scf.forall to scf.for loop conversion ------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// Transforms SCF.ForallOp's into SCF.ForOp's.
+//
+//===----------------------------------------------------------------------===//
+
+#include "mlir/Dialect/SCF/Transforms/Passes.h"
+
+#include "mlir/Dialect/SCF/IR/SCF.h"
+#include "mlir/Dialect/SCF/Transforms/Transforms.h"
+#include "mlir/IR/PatternMatch.h"
+
+namespace mlir {
+#define GEN_PASS_DEF_SCFFORALLTOFORLOOP
+#include "mlir/Dialect/SCF/Transforms/Passes.h.inc"
+} // namespace mlir
+
+using namespace llvm;
+using namespace mlir;
+using scf::ForallOp;
+using scf::ForOp;
+using scf::LoopNest;
+
+LogicalResult
+mlir::scf::forallToForLoop(RewriterBase &rewriter, scf::ForallOp forallOp,
+                           SmallVectorImpl<Operation *> *results) {
+  OpBuilder::InsertionGuard guard(rewriter);
+  rewriter.setInsertionPoint(forallOp);
+
+  Location loc = forallOp.getLoc();
+  SmallVector<Value> lbs = getValueOrCreateConstantIndexOp(
+      rewriter, loc, forallOp.getMixedLowerBound());
+  SmallVector<Value> ubs = getValueOrCreateConstantIndexOp(
+      rewriter, loc, forallOp.getMixedUpperBound());
+  SmallVector<Value> steps =
+      getValueOrCreateConstantIndexOp(rewriter, loc, forallOp.getMixedStep());
+  LoopNest loopNest = scf::buildLoopNest(rewriter, loc, lbs, ubs, steps);
+
+  SmallVector<Value> ivs = llvm::map_to_vector(
+      loopNest.loops, [](scf::ForOp loop) { return loop.getInductionVar(); });
+
+  Block *innermostBlock = loopNest.loops.back().getBody();
+  rewriter.eraseOp(forallOp.getBody()->getTerminator());
+  rewriter.inlineBlockBefore(forallOp.getBody(), innermostBlock,
+                             innermostBlock->getTerminator()->getIterator(),
+                             ivs);
+  rewriter.eraseOp(forallOp);
+
+  if (results) {
+    llvm::move(loopNest.loops, std::back_inserter(*results));
+  }
+
+  return success();
+}
+
+namespace {
+struct ForallToForLoop : public impl::SCFForallToForLoopBase<ForallToForLoop> {
+  void runOnOperation() override {
+    Operation *parentOp = getOperation();
+    IRRewriter rewriter(parentOp->getContext());
+
+    parentOp->walk([&](scf::ForallOp forallOp) {
+      if (failed(scf::forallToForLoop(rewriter, forallOp))) {
+        return signalPassFailure();
+      }
+    });
+  }
+};
+} // namespace
+
+std::unique_ptr<Pass> mlir::createForallToForLoopPass() {
+  return std::make_unique<ForallToForLoop>();
+}
diff --git a/mlir/lib/Dialect/SparseTensor/IR/SparseTensorDialect.cpp b/mlir/lib/Dialect/SparseTensor/IR/SparseTensorDialect.cpp
index b1d44559fa5a..028a69da10c1 100644
--- a/mlir/lib/Dialect/SparseTensor/IR/SparseTensorDialect.cpp
+++ b/mlir/lib/Dialect/SparseTensor/IR/SparseTensorDialect.cpp
@@ -326,9 +326,9 @@ SparseTensorDimSliceAttr::verify(function_ref<InFlightDiagnostic()> emitError,
 SparseTensorEncodingAttr
 SparseTensorEncodingAttr::withDimToLvl(AffineMap dimToLvl) const {
   assert(getImpl() && "Uninitialized SparseTensorEncodingAttr");
-  return SparseTensorEncodingAttr::get(getContext(), getLvlTypes(), dimToLvl,
-                                       AffineMap(), getPosWidth(),
-                                       getCrdWidth());
+  return SparseTensorEncodingAttr::get(
+      getContext(), getLvlTypes(), dimToLvl, AffineMap(), getPosWidth(),
+      getCrdWidth(), getExplicitVal(), getImplicitVal());
 }
 
 SparseTensorEncodingAttr
@@ -344,20 +344,44 @@ SparseTensorEncodingAttr
 SparseTensorEncodingAttr::withBitWidths(unsigned posWidth,
                                         unsigned crdWidth) const {
   assert(getImpl() && "Uninitialized SparseTensorEncodingAttr");
-  return SparseTensorEncodingAttr::get(getContext(), getLvlTypes(),
-                                       getDimToLvl(), getLvlToDim(), posWidth,
-                                       crdWidth);
+  return SparseTensorEncodingAttr::get(
+      getContext(), getLvlTypes(), getDimToLvl(), getLvlToDim(), posWidth,
+      crdWidth, getExplicitVal(), getImplicitVal());
 }
 
 SparseTensorEncodingAttr SparseTensorEncodingAttr::withoutBitWidths() const {
   return withBitWidths(0, 0);
 }
 
+SparseTensorEncodingAttr
+SparseTensorEncodingAttr::withExplicitVal(Attribute explicitVal) const {
+  assert(getImpl() && "Uninitialized SparseTensorEncodingAttr");
+  return SparseTensorEncodingAttr::get(
+      getContext(), getLvlTypes(), getDimToLvl(), getLvlToDim(), getPosWidth(),
+      getCrdWidth(), explicitVal, getImplicitVal());
+}
+
+SparseTensorEncodingAttr SparseTensorEncodingAttr::withoutExplicitVal() const {
+  return withExplicitVal(Attribute());
+}
+
+SparseTensorEncodingAttr
+SparseTensorEncodingAttr::withImplicitVal(Attribute implicitVal) const {
+  assert(getImpl() && "Uninitialized SparseTensorEncodingAttr");
+  return SparseTensorEncodingAttr::get(
+      getContext(), getLvlTypes(), getDimToLvl(), getLvlToDim(), getPosWidth(),
+      getCrdWidth(), getExplicitVal(), implicitVal);
+}
+
+SparseTensorEncodingAttr SparseTensorEncodingAttr::withoutImplicitVal() const {
+  return withImplicitVal(Attribute());
+}
+
 SparseTensorEncodingAttr SparseTensorEncodingAttr::withDimSlices(
     ArrayRef<SparseTensorDimSliceAttr> dimSlices) const {
-  return SparseTensorEncodingAttr::get(getContext(), getLvlTypes(),
-                                       getDimToLvl(), getLvlToDim(),
-                                       getPosWidth(), getCrdWidth(), dimSlices);
+  return SparseTensorEncodingAttr::get(
+      getContext(), getLvlTypes(), getDimToLvl(), getLvlToDim(), getPosWidth(),
+      getCrdWidth(), getExplicitVal(), getImplicitVal(), dimSlices);
 }
 
 SparseTensorEncodingAttr SparseTensorEncodingAttr::withoutDimSlices() const {
@@ -553,8 +577,11 @@ Attribute SparseTensorEncodingAttr::parse(AsmParser &parser, Type type) {
   AffineMap lvlToDim = {};
   unsigned posWidth = 0;
   unsigned crdWidth = 0;
+  Attribute explicitVal;
+  Attribute implicitVal;
   StringRef attrName;
-  SmallVector<StringRef, 3> keys = {"map", "posWidth", "crdWidth"};
+  SmallVector<StringRef, 5> keys = {"map", "posWidth", "crdWidth",
+                                    "explicitVal", "implicitVal"};
   while (succeeded(parser.parseOptionalKeyword(&attrName))) {
     // Detect admissible keyword.
     auto *it = find(keys, attrName);
@@ -628,6 +655,36 @@ Attribute SparseTensorEncodingAttr::parse(AsmParser &parser, Type type) {
       crdWidth = intAttr.getInt();
       break;
     }
+    case 3: { // explicitVal
+      Attribute attr;
+      if (failed(parser.parseAttribute(attr)))
+        return {};
+      if (auto result = llvm::dyn_cast<FloatAttr>(attr)) {
+        explicitVal = result;
+      } else if (auto result = llvm::dyn_cast<IntegerAttr>(attr)) {
+        explicitVal = result;
+      } else {
+        parser.emitError(parser.getNameLoc(),
+                         "expected a numeric value for explicitVal");
+        return {};
+      }
+      break;
+    }
+    case 4: { // implicitVal
+      Attribute attr;
+      if (failed(parser.parseAttribute(attr)))
+        return {};
+      if (auto result = llvm::dyn_cast<FloatAttr>(attr)) {
+        implicitVal = result;
+      } else if (auto result = llvm::dyn_cast<IntegerAttr>(attr)) {
+        implicitVal = result;
+      } else {
+        parser.emitError(parser.getNameLoc(),
+                         "expected a numeric value for implicitVal");
+        return {};
+      }
+      break;
+    }
     } // switch
     // Only last item can omit the comma.
     if (parser.parseOptionalComma().failed())
@@ -646,7 +703,7 @@ Attribute SparseTensorEncodingAttr::parse(AsmParser &parser, Type type) {
   }
   return parser.getChecked<SparseTensorEncodingAttr>(
       parser.getContext(), lvlTypes, dimToLvl, lvlToDim, posWidth, crdWidth,
-      dimSlices);
+      explicitVal, implicitVal, dimSlices);
 }
 
 void SparseTensorEncodingAttr::print(AsmPrinter &printer) const {
@@ -666,6 +723,11 @@ void SparseTensorEncodingAttr::print(AsmPrinter &printer) const {
     printer << ", posWidth = " << getPosWidth();
   if (getCrdWidth())
     printer << ", crdWidth = " << getCrdWidth();
+  if (getExplicitVal()) {
+    printer << ", explicitVal = " << getExplicitVal();
+  }
+  if (getImplicitVal())
+    printer << ", implicitVal = " << getImplicitVal();
   printer << " }>";
 }
 
@@ -715,7 +777,8 @@ void SparseTensorEncodingAttr::printLevels(AffineMap &map, AsmPrinter &printer,
 LogicalResult SparseTensorEncodingAttr::verify(
     function_ref<InFlightDiagnostic()> emitError, ArrayRef<LevelType> lvlTypes,
     AffineMap dimToLvl, AffineMap lvlToDim, unsigned posWidth,
-    unsigned crdWidth, ArrayRef<SparseTensorDimSliceAttr> dimSlices) {
+    unsigned crdWidth, Attribute explicitVal, Attribute implicitVal,
+    ArrayRef<SparseTensorDimSliceAttr> dimSlices) {
   if (!acceptBitWidth(posWidth))
     return emitError() << "unexpected position bitwidth: " << posWidth;
   if (!acceptBitWidth(crdWidth))
@@ -831,7 +894,8 @@ LogicalResult SparseTensorEncodingAttr::verifyEncoding(
   // Check structural integrity.  In particular, this ensures that the
   // level-rank is coherent across all the fields.
   if (failed(verify(emitError, getLvlTypes(), getDimToLvl(), getLvlToDim(),
-                    getPosWidth(), getCrdWidth(), getDimSlices())))
+                    getPosWidth(), getCrdWidth(), getExplicitVal(),
+                    getImplicitVal(), getDimSlices())))
     return failure();
   // Check integrity with tensor type specifics.  In particular, we
   // need only check that the dimension-rank of the tensor agrees with
@@ -921,9 +985,9 @@ mlir::sparse_tensor::SparseTensorType::getCOOType(bool ordered) const {
     // Ends by a unique singleton level.
     lvlTypes.push_back(*buildLevelType(LevelFormat::Singleton, ordered, true));
   }
-  auto enc = SparseTensorEncodingAttr::get(getContext(), lvlTypes,
-                                           getDimToLvl(), getLvlToDim(),
-                                           getPosWidth(), getCrdWidth());
+  auto enc = SparseTensorEncodingAttr::get(
+      getContext(), lvlTypes, getDimToLvl(), getLvlToDim(), getPosWidth(),
+      getCrdWidth(), getExplicitVal(), getImplicitVal());
   return RankedTensorType::get(getDimShape(), getElementType(), enc);
 }
 
@@ -1115,7 +1179,10 @@ getNormalizedEncodingForSpecifier(SparseTensorEncodingAttr enc) {
       // `getPosWidth` and `getCrdWidth`. It allows us to reuse the same SSA
       // value for different bitwidth, it also avoids casting between index and
       // integer (returned by DimOp)
-      0, 0, enc.getDimSlices());
+      0, 0,
+      Attribute(), // explicitVal (irrelevant to storage specifier)
+      Attribute(), // implicitVal (irrelevant to storage specifier)
+      enc.getDimSlices());
 }
 
 StorageSpecifierType
diff --git a/mlir/lib/Dialect/SparseTensor/Transforms/SparseTensorRewriting.cpp b/mlir/lib/Dialect/SparseTensor/Transforms/SparseTensorRewriting.cpp
index 02375f54d715..9a8c6422a7ff 100644
--- a/mlir/lib/Dialect/SparseTensor/Transforms/SparseTensorRewriting.cpp
+++ b/mlir/lib/Dialect/SparseTensor/Transforms/SparseTensorRewriting.cpp
@@ -209,6 +209,117 @@ static void concatSizesFromInputs(OpBuilder &builder,
 
 namespace {
 
+/// TODO: move it to tensor dialect instead.
+///
+/// Fold `tensor.concat` and `tensor.extract_slice`
+///
+/// %concat = tensor.concat dim(2) %t0, %t1
+///   : (tensor<1x64x1xf32>, tensor<1x64x1xf32>) -> tensor<1x64x2xf32>
+/// %extracted0 = tensor.extract_slice %concat[0, 0, 0][1, 64, 1][1, 1, 1]
+///   : tensor<1x64x2xf32> to tensor<1x64x1xf32>
+/// %extracted1 = tensor.extract_slice %concat[0, 0, 1][1, 64, 1][1, 1, 1]
+///   : tensor<1x64x2xf32> to tensor<1x64x1xf32>
+///
+/// Becomes
+///
+/// %extract0, %extract1 = %t0, %t1
+struct FuseExtractSliceWithConcat
+    : public OpRewritePattern<tensor::ExtractSliceOp> {
+  using OpRewritePattern<tensor::ExtractSliceOp>::OpRewritePattern;
+
+  LogicalResult matchAndRewrite(tensor::ExtractSliceOp extractOp,
+                                PatternRewriter &rewriter) const override {
+    auto concatOp = extractOp.getSource().getDefiningOp<tensor::ConcatOp>();
+    if (!concatOp)
+      return failure();
+
+    Location loc = extractOp.getLoc();
+    int64_t dim = concatOp.getDim();
+    int64_t rank = extractOp.getResultType().getRank();
+
+    SmallVector<OpFoldResult> srcStrides(rank, rewriter.getIndexAttr(1));
+    SmallVector<OpFoldResult> srcOffsets(rank, rewriter.getIndexAttr(0));
+
+    // Compute the partial sums for the slice offsets.
+    AffineExpr sum = rewriter.getAffineDimExpr(0);
+    SmallVector<AffineExpr> partialSums = {sum};
+    SmallVector<OpFoldResult> offsetStrides = {rewriter.getIndexAttr(0)};
+    for (auto [idx, input] :
+         llvm::enumerate(concatOp.getInputs().drop_back())) {
+      sum = sum + rewriter.getAffineDimExpr(idx + 1);
+      partialSums.push_back(sum);
+      offsetStrides.push_back(
+          rewriter.createOrFold<tensor::DimOp>(loc, input, dim));
+    }
+    auto partialSumMap = AffineMap::get(concatOp.getInputs().size(), 0,
+                                        partialSums, rewriter.getContext());
+    SmallVector<OpFoldResult> dimOffsets =
+        affine::makeComposedFoldedMultiResultAffineApply(
+            rewriter, loc, partialSumMap, offsetStrides);
+
+    auto allEqual = [](ArrayRef<OpFoldResult> lhs, ArrayRef<OpFoldResult> rhs) {
+      for (auto [l, r] : llvm::zip(lhs, rhs)) {
+        std::optional<int64_t> staticVal = getConstantIntValue(l);
+        if (!staticVal.has_value() || staticVal != getConstantIntValue(r))
+          return false;
+      }
+      return lhs.size() == rhs.size();
+    };
+
+    for (auto [i, input, offset] :
+         llvm::enumerate(concatOp.getInputs(), dimOffsets)) {
+      SmallVector<OpFoldResult> srcSizes =
+          tensor::getMixedSizes(rewriter, loc, input);
+      srcOffsets[dim] = offset;
+
+      SmallVector<OpFoldResult> dstSizes = extractOp.getMixedSizes();
+      SmallVector<OpFoldResult> dstOffsets = extractOp.getMixedOffsets();
+      SmallVector<OpFoldResult> dstStrides = extractOp.getMixedStrides();
+
+      if (allEqual(srcSizes, dstSizes) && allEqual(srcOffsets, dstOffsets) &&
+          allEqual(srcStrides, dstStrides)) {
+        Value operand = concatOp.getOperand(i);
+        if (operand.getType() == extractOp.getResultType())
+          rewriter.replaceOp(extractOp, operand);
+        break;
+      }
+    }
+
+    return success();
+  }
+};
+
+/// Rewriting rule that fuses sparse_tensor.convert into producer.
+struct FoldConvertIntoProducer : public OpRewritePattern<ConvertOp> {
+public:
+  using OpRewritePattern::OpRewritePattern;
+
+  LogicalResult matchAndRewrite(ConvertOp op,
+                                PatternRewriter &rewriter) const override {
+    auto producer = op.getSource().getDefiningOp<GenericOp>();
+    if (!producer || producer.getDpsInits().size() != 1 ||
+        !isMaterializing(producer.getDpsInitOperand(0), false) ||
+        !producer.getResult(0).hasOneUse()) {
+      return failure();
+    }
+    rewriter.modifyOpInPlace(producer, [&]() {
+      producer.getResult(0).setType(op.getResult().getType());
+    });
+
+    Operation *materializeOp =
+        producer.getDpsInitOperand(0)->get().getDefiningOp();
+
+    rewriter.modifyOpInPlace(materializeOp, [&]() {
+      materializeOp->getResult(0).setType(op.getResult().getType());
+    });
+
+    rewriter.replaceAllOpUsesWith(op, producer);
+    op->erase();
+
+    return success();
+  }
+};
+
 /// Rewriting rule that converts direct yield of zero with initial allocation.
 struct FoldInvariantYield : public OpRewritePattern<GenericOp> {
 public:
@@ -1426,7 +1537,8 @@ struct OutRewriter : public OpRewritePattern<OutOp> {
 //===---------------------------------------------------------------------===//
 
 void mlir::populatePreSparsificationRewriting(RewritePatternSet &patterns) {
-  patterns.add<FoldInvariantYield, FuseSparseMultiplyOverAdd, FuseTensorCast,
+  patterns.add<FuseExtractSliceWithConcat, FoldConvertIntoProducer,
+               FoldInvariantYield, FuseSparseMultiplyOverAdd, FuseTensorCast,
                GenSemiRingReduction, GenSemiRingSelect, PrintRewriter>(
       patterns.getContext());
 }
diff --git a/mlir/lib/Dialect/SparseTensor/Transforms/Sparsification.cpp b/mlir/lib/Dialect/SparseTensor/Transforms/Sparsification.cpp
index cd046b670d9a..0a9bb40b458d 100644
--- a/mlir/lib/Dialect/SparseTensor/Transforms/Sparsification.cpp
+++ b/mlir/lib/Dialect/SparseTensor/Transforms/Sparsification.cpp
@@ -403,6 +403,22 @@ static Value genInsertionLoadReduce(CodegenEnv &env, OpBuilder &builder,
   return builder.create<arith::SelectOp>(loc, isFilled, valAtIndex, identity);
 }
 
+static Value genConditionalInsert(Location loc, OpBuilder &builder, Value cond,
+                                  Value sparseOut, ValueRange ivs, Value v) {
+  scf::IfOp condInsert =
+      builder.create<scf::IfOp>(loc, sparseOut.getType(), cond, true);
+  // True branch.
+  builder.setInsertionPointToStart(condInsert.thenBlock());
+  Value res = builder.create<tensor::InsertOp>(loc, v, sparseOut, ivs);
+  builder.create<scf::YieldOp>(loc, res);
+  // False branch.
+  builder.setInsertionPointToStart(condInsert.elseBlock());
+  builder.create<scf::YieldOp>(loc, sparseOut);
+  // Value assignment.
+  builder.setInsertionPointAfter(condInsert);
+  return condInsert.getResult(0);
+}
+
 /// Generates insertion code to implement dynamic tensor store.
 static void genInsertionStore(CodegenEnv &env, OpBuilder &builder, OpOperand *t,
                               Value rhs) {
@@ -423,23 +439,21 @@ static void genInsertionStore(CodegenEnv &env, OpBuilder &builder, OpOperand *t,
       //     return updated chain
       //   else
       //     return unmodified chain
-      scf::IfOp ifValidLexInsert = builder.create<scf::IfOp>(
-          loc, chain.getType(), env.getValidLexInsert(),
-          /*else=*/true);
-      // True branch.
-      builder.setInsertionPointToStart(ifValidLexInsert.thenBlock());
-      Value res = builder.create<tensor::InsertOp>(loc, rhs, chain, ivs);
-      builder.create<scf::YieldOp>(loc, res);
-      // False branch.
-      builder.setInsertionPointToStart(ifValidLexInsert.elseBlock());
-      builder.create<scf::YieldOp>(loc, chain);
-      // Value assignment.
-      builder.setInsertionPointAfter(ifValidLexInsert);
-      env.updateInsertionChain(ifValidLexInsert.getResult(0));
+      Value out = genConditionalInsert(loc, builder, env.getValidLexInsert(),
+                                       chain, ivs, rhs);
+      env.updateInsertionChain(out);
     } else {
+      Value sparseOut;
+      if (!hasAnySparseType(env.op().getInputs().getTypes())) {
+        // This is an all-dense -> sparse kernel, test rhs != 0 before
+        // insertion.
+        Value nz = genIsNonzero(builder, loc, rhs);
+        sparseOut = genConditionalInsert(loc, builder, nz, chain, ivs, rhs);
+      } else {
+        sparseOut = builder.create<tensor::InsertOp>(loc, rhs, chain, ivs);
+      }
       // Generates regular insertion chain.
-      env.updateInsertionChain(
-          builder.create<tensor::InsertOp>(loc, rhs, chain, ivs));
+      env.updateInsertionChain(sparseOut);
     }
     return;
   }
diff --git a/mlir/lib/Dialect/SparseTensor/Transforms/Utils/SparseTensorIterator.h b/mlir/lib/Dialect/SparseTensor/Transforms/Utils/SparseTensorIterator.h
index 9d69a2335559..46b923250dd8 100644
--- a/mlir/lib/Dialect/SparseTensor/Transforms/Utils/SparseTensorIterator.h
+++ b/mlir/lib/Dialect/SparseTensor/Transforms/Utils/SparseTensorIterator.h
@@ -284,9 +284,10 @@ private:
 };
 
 /// Helper function to create a TensorLevel object from given `tensor`.
-std::unique_ptr<SparseTensorLevel> makeSparseTensorLevel(OpBuilder &builder,
-                                                         Location loc, Value t,
-                                                         unsigned tid, Level l);
+std::unique_ptr<SparseTensorLevel> makeSparseTensorLevel(OpBuilder &b,
+                                                         Location l, Value t,
+                                                         unsigned tid,
+                                                         Level lvl);
 
 /// Helper function to create a simple SparseIterator object that iterate over
 /// the SparseTensorLevel.
diff --git a/mlir/lib/Dialect/Tensor/IR/TensorOps.cpp b/mlir/lib/Dialect/Tensor/IR/TensorOps.cpp
index 3ff41ab22fbc..5029ed4aa038 100644
--- a/mlir/lib/Dialect/Tensor/IR/TensorOps.cpp
+++ b/mlir/lib/Dialect/Tensor/IR/TensorOps.cpp
@@ -1609,6 +1609,9 @@ OpFoldResult ReshapeOp::fold(FoldAdaptor adaptor) {
             cst.has_value() && cst.value() == static_cast<int64_t>(id);
         continue;
       }
+
+      dynamicNoop = false;
+      break;
     }
 
     if (dynamicNoop)
diff --git a/mlir/lib/Dialect/Tensor/IR/TensorTilingInterfaceImpl.cpp b/mlir/lib/Dialect/Tensor/IR/TensorTilingInterfaceImpl.cpp
index 296c5fc7a5c2..d25efcf50ec5 100644
--- a/mlir/lib/Dialect/Tensor/IR/TensorTilingInterfaceImpl.cpp
+++ b/mlir/lib/Dialect/Tensor/IR/TensorTilingInterfaceImpl.cpp
@@ -61,8 +61,8 @@ struct PadOpTiling : public TilingInterface::ExternalModel<PadOpTiling, PadOp> {
   getResultTilePosition(Operation *op, OpBuilder &b, unsigned resultNumber,
                         ArrayRef<OpFoldResult> offsets,
                         ArrayRef<OpFoldResult> sizes,
-                        SmallVectorImpl<OpFoldResult> &resultOffsets,
-                        SmallVectorImpl<OpFoldResult> &resultSizes) const {
+                        SmallVector<OpFoldResult> &resultOffsets,
+                        SmallVector<OpFoldResult> &resultSizes) const {
     resultOffsets.assign(offsets.begin(), offsets.end());
     resultSizes.assign(sizes.begin(), sizes.end());
     return success();
@@ -199,8 +199,8 @@ struct PackOpTiling
   getResultTilePosition(Operation *op, OpBuilder &b, unsigned resultNumber,
                         ArrayRef<OpFoldResult> offsets,
                         ArrayRef<OpFoldResult> sizes,
-                        SmallVectorImpl<OpFoldResult> &resultOffsets,
-                        SmallVectorImpl<OpFoldResult> &resultSizes) const {
+                        SmallVector<OpFoldResult> &resultOffsets,
+                        SmallVector<OpFoldResult> &resultSizes) const {
     // The iteration domain is over outer dimensions of packed layout. In this
     // context, the outer dimensions of `resultOffsets` are `offsets`. The
     // inner dimensions of `resultOffsets` are zeros because tiling is not
@@ -452,8 +452,8 @@ struct UnPackOpTiling
   getResultTilePosition(Operation *op, OpBuilder &b, unsigned resultNumber,
                         ArrayRef<OpFoldResult> offsets,
                         ArrayRef<OpFoldResult> sizes,
-                        SmallVectorImpl<OpFoldResult> &resultOffsets,
-                        SmallVectorImpl<OpFoldResult> &resultSizes) const {
+                        SmallVector<OpFoldResult> &resultOffsets,
+                        SmallVector<OpFoldResult> &resultSizes) const {
     resultOffsets = llvm::to_vector(offsets);
     resultSizes = llvm::to_vector(sizes);
     return success();
diff --git a/mlir/lib/Dialect/Vector/Transforms/LowerVectorMultiReduction.cpp b/mlir/lib/Dialect/Vector/Transforms/LowerVectorMultiReduction.cpp
index 2f21c50c6347..ac576ed0b4f0 100644
--- a/mlir/lib/Dialect/Vector/Transforms/LowerVectorMultiReduction.cpp
+++ b/mlir/lib/Dialect/Vector/Transforms/LowerVectorMultiReduction.cpp
@@ -437,8 +437,10 @@ struct OneDimMultiReductionToTwoDim
     auto loc = multiReductionOp.getLoc();
     auto srcVectorType = multiReductionOp.getSourceVectorType();
     auto srcShape = srcVectorType.getShape();
-    auto castedType = VectorType::get(ArrayRef<int64_t>{1, srcShape.back()},
-                                      srcVectorType.getElementType());
+    auto castedType = VectorType::get(
+        ArrayRef<int64_t>{1, srcShape.back()}, srcVectorType.getElementType(),
+        ArrayRef<bool>{false, srcVectorType.getScalableDims().back()});
+
     auto accType =
         VectorType::get(ArrayRef<int64_t>{1}, srcVectorType.getElementType());
     assert(!llvm::isa<VectorType>(multiReductionOp.getDestType()) &&
@@ -455,10 +457,11 @@ struct OneDimMultiReductionToTwoDim
         loc, accType, multiReductionOp.getAcc());
     Value castMask;
     if (maskableOp.isMasked()) {
-      auto maskType = llvm::cast<ShapedType>(mask.getType());
-      auto castMaskType =
-          VectorType::get(ArrayRef<int64_t>{1, maskType.getShape().back()},
-                          maskType.getElementType());
+      auto maskType = llvm::cast<VectorType>(mask.getType());
+      auto castMaskType = VectorType::get(
+          ArrayRef<int64_t>{1, maskType.getShape().back()},
+          maskType.getElementType(),
+          ArrayRef<bool>{false, maskType.getScalableDims().back()});
       castMask = rewriter.create<vector::BroadcastOp>(loc, castMaskType, mask);
     }
 
diff --git a/mlir/lib/Dialect/Vector/Transforms/VectorLinearize.cpp b/mlir/lib/Dialect/Vector/Transforms/VectorLinearize.cpp
index 69999f0918c1..802a64b0805e 100644
--- a/mlir/lib/Dialect/Vector/Transforms/VectorLinearize.cpp
+++ b/mlir/lib/Dialect/Vector/Transforms/VectorLinearize.cpp
@@ -140,7 +140,7 @@ struct LinearizeVectorExtractStridedSlice final
                   ConversionPatternRewriter &rewriter) const override {
     Type dstType = getTypeConverter()->convertType(extractOp.getType());
     assert(!(extractOp.getVector().getType().isScalable() ||
-             dstType.cast<VectorType>().isScalable()) &&
+             cast<VectorType>(dstType).isScalable()) &&
            "scalable vectors are not supported.");
     if (!isLessThanTargetBitWidth(extractOp, targetVectorBitWidth))
       return rewriter.notifyMatchFailure(
@@ -172,7 +172,7 @@ struct LinearizeVectorExtractStridedSlice final
     // Get total number of extracted slices.
     int64_t nExtractedSlices = 1;
     for (Attribute size : sizes) {
-      nExtractedSlices *= size.cast<IntegerAttr>().getInt();
+      nExtractedSlices *= cast<IntegerAttr>(size).getInt();
     }
     // Compute the strides of the source vector considering first k dimensions.
     llvm::SmallVector<int64_t, 4> sourceStrides(kD, extractGranularitySize);
@@ -189,7 +189,7 @@ struct LinearizeVectorExtractStridedSlice final
     // Compute extractedStrides.
     for (int i = kD - 2; i >= 0; --i) {
       extractedStrides[i] =
-          extractedStrides[i + 1] * sizes[i + 1].cast<IntegerAttr>().getInt();
+          extractedStrides[i + 1] * cast<IntegerAttr>(sizes[i + 1]).getInt();
     }
     // Iterate over all extracted slices from 0 to nExtractedSlices - 1
     // and compute the multi-dimensional index and the corresponding linearized
@@ -207,7 +207,7 @@ struct LinearizeVectorExtractStridedSlice final
       int64_t linearizedIndex = 0;
       for (int64_t j = 0; j < kD; ++j) {
         linearizedIndex +=
-            (offsets[j].cast<IntegerAttr>().getInt() + multiDimIndex[j]) *
+            (cast<IntegerAttr>(offsets[j]).getInt() + multiDimIndex[j]) *
             sourceStrides[j];
       }
       // Fill the indices array form linearizedIndex to linearizedIndex +
@@ -254,7 +254,7 @@ struct LinearizeVectorShuffle final
     Type dstType = getTypeConverter()->convertType(shuffleOp.getType());
     assert(!(shuffleOp.getV1VectorType().isScalable() ||
              shuffleOp.getV2VectorType().isScalable() ||
-             dstType.cast<VectorType>().isScalable()) &&
+             cast<VectorType>(dstType).isScalable()) &&
            "scalable vectors are not supported.");
     if (!isLessThanTargetBitWidth(shuffleOp, targetVectorBitWidth))
       return rewriter.notifyMatchFailure(
@@ -324,7 +324,7 @@ struct LinearizeVectorExtract final
                   ConversionPatternRewriter &rewriter) const override {
     Type dstTy = getTypeConverter()->convertType(extractOp.getType());
     assert(!(extractOp.getVector().getType().isScalable() ||
-             dstTy.cast<VectorType>().isScalable()) &&
+             cast<VectorType>(dstTy).isScalable()) &&
            "scalable vectors are not supported.");
     if (!isLessThanTargetBitWidth(extractOp, targetVectorBitWidth))
       return rewriter.notifyMatchFailure(
@@ -405,9 +405,7 @@ void mlir::vector::populateVectorLinearizeShuffleLikeOpsPatterns(
       [=](vector::ShuffleOp shuffleOp) -> bool {
         return isLessThanTargetBitWidth(shuffleOp, targetBitWidth)
                    ? (typeConverter.isLegal(shuffleOp) &&
-                      shuffleOp.getResult()
-                              .getType()
-                              .cast<mlir::VectorType>()
+                      cast<mlir::VectorType>(shuffleOp.getResult().getType())
                               .getRank() == 1)
                    : true;
       });
diff --git a/mlir/lib/Dialect/Vector/Utils/VectorUtils.cpp b/mlir/lib/Dialect/Vector/Utils/VectorUtils.cpp
index fcaf1ec944b4..6727f3f46172 100644
--- a/mlir/lib/Dialect/Vector/Utils/VectorUtils.cpp
+++ b/mlir/lib/Dialect/Vector/Utils/VectorUtils.cpp
@@ -345,7 +345,7 @@ Value vector::createReadOrMaskedRead(OpBuilder &builder, Location loc,
   int64_t readRank = readShape.size();
   auto zero = builder.create<arith::ConstantIndexOp>(loc, 0);
   SmallVector<bool> inBoundsVal(readRank, true);
-  if (!useInBoundsInsteadOfMasking) {
+  if (useInBoundsInsteadOfMasking) {
     // Update the inBounds attribute.
     for (unsigned i = 0; i < readRank; i++)
       inBoundsVal[i] = (sourceShape[i] == readShape[i]) &&
@@ -359,7 +359,7 @@ Value vector::createReadOrMaskedRead(OpBuilder &builder, Location loc,
       /*padding=*/padValue,
       /*inBounds=*/inBoundsVal);
 
-  if (llvm::equal(readShape, sourceShape) || !useInBoundsInsteadOfMasking)
+  if (llvm::equal(readShape, sourceShape) || useInBoundsInsteadOfMasking)
     return transferReadOp;
   SmallVector<OpFoldResult> mixedSourceDims =
       tensor::getMixedSizes(builder, loc, source);
diff --git a/mlir/lib/Dialect/XeGPU/IR/CMakeLists.txt b/mlir/lib/Dialect/XeGPU/IR/CMakeLists.txt
index 617c89a84ee0..a0ce7f9706ce 100644
--- a/mlir/lib/Dialect/XeGPU/IR/CMakeLists.txt
+++ b/mlir/lib/Dialect/XeGPU/IR/CMakeLists.txt
@@ -11,6 +11,7 @@ add_mlir_dialect_library(MLIRXeGPUDialect
   MLIRXeGPUEnumsIncGen
 
   LINK_LIBS PUBLIC
+  MLIRArithDialect
   MLIRDialectUtils
   MLIRIR
   MLIRViewLikeInterface
diff --git a/mlir/lib/Dialect/XeGPU/IR/XeGPUOps.cpp b/mlir/lib/Dialect/XeGPU/IR/XeGPUOps.cpp
index 23c5749c2309..22959224d56c 100644
--- a/mlir/lib/Dialect/XeGPU/IR/XeGPUOps.cpp
+++ b/mlir/lib/Dialect/XeGPU/IR/XeGPUOps.cpp
@@ -406,6 +406,28 @@ LogicalResult StoreScatterOp::verify() {
 
   return success();
 }
+//===----------------------------------------------------------------------===//
+// XeGPU_DpasOp
+//===----------------------------------------------------------------------===//
+LogicalResult DpasOp::verify() {
+  int64_t lhsRank = getLhsType().getRank();
+  int64_t rhsRank = getRhsType().getRank();
+
+  if (lhsRank != rhsRank || lhsRank != 3)
+    return emitOpError(
+        "lhs and rhs rank does not match for dpas op, or their rank is not 3.");
+
+  if (getAcc() && getAccType() != getResultType())
+    return emitOpError("Accumulator and Result for dpas op should have the "
+                       "same type (both shape and element type).");
+
+  auto lhsShape = getLhsType().getShape();
+  auto rhsShape = getRhsType().getShape();
+  if (lhsShape[1] != rhsShape[0] || lhsShape[2] != rhsShape[2])
+    return emitOpError("K-dimension or vnni-factor mismatch.");
+
+  return success();
+}
 
 } // namespace xegpu
 } // namespace mlir
diff --git a/mlir/lib/IR/Builders.cpp b/mlir/lib/IR/Builders.cpp
index 18ca3c332e02..d49f69a7b7ae 100644
--- a/mlir/lib/IR/Builders.cpp
+++ b/mlir/lib/IR/Builders.cpp
@@ -476,16 +476,14 @@ Operation *OpBuilder::create(Location loc, StringAttr opName,
   return create(state);
 }
 
-/// Attempts to fold the given operation and places new results within
-/// 'results'. Returns success if the operation was folded, failure otherwise.
-/// Note: This function does not erase the operation on a successful fold.
 LogicalResult OpBuilder::tryFold(Operation *op,
                                  SmallVectorImpl<Value> &results) {
+  assert(results.empty() && "expected empty results");
   ResultRange opResults = op->getResults();
 
   results.reserve(opResults.size());
   auto cleanupFailure = [&] {
-    results.assign(opResults.begin(), opResults.end());
+    results.clear();
     return failure();
   };
 
@@ -495,20 +493,24 @@ LogicalResult OpBuilder::tryFold(Operation *op,
 
   // Try to fold the operation.
   SmallVector<OpFoldResult, 4> foldResults;
-  if (failed(op->fold(foldResults)) || foldResults.empty())
+  if (failed(op->fold(foldResults)))
     return cleanupFailure();
 
+  // An in-place fold does not require generation of any constants.
+  if (foldResults.empty())
+    return success();
+
   // A temporary builder used for creating constants during folding.
   OpBuilder cstBuilder(context);
   SmallVector<Operation *, 1> generatedConstants;
 
   // Populate the results with the folded results.
   Dialect *dialect = op->getDialect();
-  for (auto it : llvm::zip_equal(foldResults, opResults.getTypes())) {
-    Type expectedType = std::get<1>(it);
+  for (auto [foldResult, expectedType] :
+       llvm::zip_equal(foldResults, opResults.getTypes())) {
 
     // Normal values get pushed back directly.
-    if (auto value = llvm::dyn_cast_if_present<Value>(std::get<0>(it))) {
+    if (auto value = llvm::dyn_cast_if_present<Value>(foldResult)) {
       results.push_back(value);
       continue;
     }
@@ -518,7 +520,7 @@ LogicalResult OpBuilder::tryFold(Operation *op,
       return cleanupFailure();
 
     // Ask the dialect to materialize a constant operation for this value.
-    Attribute attr = std::get<0>(it).get<Attribute>();
+    Attribute attr = foldResult.get<Attribute>();
     auto *constOp = dialect->materializeConstant(cstBuilder, attr, expectedType,
                                                  op->getLoc());
     if (!constOp) {
diff --git a/mlir/lib/Interfaces/RuntimeVerifiableOpInterface.cpp b/mlir/lib/Interfaces/RuntimeVerifiableOpInterface.cpp
index 9205d8d8c34a..561e8d338687 100644
--- a/mlir/lib/Interfaces/RuntimeVerifiableOpInterface.cpp
+++ b/mlir/lib/Interfaces/RuntimeVerifiableOpInterface.cpp
@@ -11,6 +11,27 @@
 namespace mlir {
 class Location;
 class OpBuilder;
+
+/// Generate an error message string for the given op and the specified error.
+std::string
+RuntimeVerifiableOpInterface::generateErrorMessage(Operation *op,
+                                                   const std::string &msg) {
+  std::string buffer;
+  llvm::raw_string_ostream stream(buffer);
+  OpPrintingFlags flags;
+  // We may generate a lot of error messages and so we need to ensure the
+  // printing is fast.
+  flags.elideLargeElementsAttrs();
+  flags.printGenericOpForm();
+  flags.skipRegions();
+  flags.useLocalScope();
+  stream << "ERROR: Runtime op verification failed\n";
+  op->print(stream, flags);
+  stream << "\n^ " << msg;
+  stream << "\nLocation: ";
+  op->getLoc().print(stream);
+  return stream.str();
+}
 } // namespace mlir
 
 /// Include the definitions of the interface.
diff --git a/mlir/lib/TableGen/CodeGenHelpers.cpp b/mlir/lib/TableGen/CodeGenHelpers.cpp
index d906de6b56af..59865146e20b 100644
--- a/mlir/lib/TableGen/CodeGenHelpers.cpp
+++ b/mlir/lib/TableGen/CodeGenHelpers.cpp
@@ -24,7 +24,8 @@ using namespace mlir::tblgen;
 
 /// Generate a unique label based on the current file name to prevent name
 /// collisions if multiple generated files are included at once.
-static std::string getUniqueOutputLabel(const llvm::RecordKeeper &records) {
+static std::string getUniqueOutputLabel(const llvm::RecordKeeper &records,
+                                        StringRef tag) {
   // Use the input file name when generating a unique name.
   std::string inputFilename = records.getInputFilename();
 
@@ -33,7 +34,7 @@ static std::string getUniqueOutputLabel(const llvm::RecordKeeper &records) {
   nameRef.consume_back(".td");
 
   // Sanitize any invalid characters.
-  std::string uniqueName;
+  std::string uniqueName(tag);
   for (char c : nameRef) {
     if (llvm::isAlnum(c) || c == '_')
       uniqueName.push_back(c);
@@ -44,15 +45,11 @@ static std::string getUniqueOutputLabel(const llvm::RecordKeeper &records) {
 }
 
 StaticVerifierFunctionEmitter::StaticVerifierFunctionEmitter(
-    raw_ostream &os, const llvm::RecordKeeper &records)
-    : os(os), uniqueOutputLabel(getUniqueOutputLabel(records)) {}
+    raw_ostream &os, const llvm::RecordKeeper &records, StringRef tag)
+    : os(os), uniqueOutputLabel(getUniqueOutputLabel(records, tag)) {}
 
 void StaticVerifierFunctionEmitter::emitOpConstraints(
-    ArrayRef<llvm::Record *> opDefs, bool emitDecl) {
-  collectOpConstraints(opDefs);
-  if (emitDecl)
-    return;
-
+    ArrayRef<llvm::Record *> opDefs) {
   NamespaceEmitter namespaceEmitter(os, Operator(*opDefs[0]).getCppNamespace());
   emitTypeConstraints();
   emitAttrConstraints();
diff --git a/mlir/lib/Target/Cpp/TranslateToCpp.cpp b/mlir/lib/Target/Cpp/TranslateToCpp.cpp
index ee87c1d4b934..1bdb74cd8bf2 100644
--- a/mlir/lib/Target/Cpp/TranslateToCpp.cpp
+++ b/mlir/lib/Target/Cpp/TranslateToCpp.cpp
@@ -154,6 +154,9 @@ struct CppEmitter {
   /// any result type could not be converted.
   LogicalResult emitAssignPrefix(Operation &op);
 
+  /// Emits a global variable declaration or definition.
+  LogicalResult emitGlobalVariable(GlobalOp op);
+
   /// Emits a label for the block.
   LogicalResult emitLabel(Block &block);
 
@@ -345,6 +348,12 @@ static LogicalResult printOperation(CppEmitter &emitter,
 }
 
 static LogicalResult printOperation(CppEmitter &emitter,
+                                    emitc::GlobalOp globalOp) {
+
+  return emitter.emitGlobalVariable(globalOp);
+}
+
+static LogicalResult printOperation(CppEmitter &emitter,
                                     emitc::AssignOp assignOp) {
   OpResult result = assignOp.getVar().getDefiningOp()->getResult(0);
 
@@ -355,6 +364,13 @@ static LogicalResult printOperation(CppEmitter &emitter,
 }
 
 static LogicalResult printOperation(CppEmitter &emitter,
+                                    emitc::GetGlobalOp op) {
+  // Add name to cache so that `hasValueInScope` works.
+  emitter.getOrCreateName(op.getResult());
+  return success();
+}
+
+static LogicalResult printOperation(CppEmitter &emitter,
                                     emitc::SubscriptOp subscriptOp) {
   // Add name to cache so that `hasValueInScope` works.
   emitter.getOrCreateName(subscriptOp.getResult());
@@ -1119,6 +1135,9 @@ StringRef CppEmitter::getOrCreateName(Value val) {
     if (auto subscript =
             dyn_cast_if_present<emitc::SubscriptOp>(val.getDefiningOp())) {
       valueMapper.insert(val, getSubscriptName(subscript));
+    } else if (auto getGlobal = dyn_cast_if_present<emitc::GetGlobalOp>(
+                   val.getDefiningOp())) {
+      valueMapper.insert(val, getGlobal.getName().str());
     } else {
       valueMapper.insert(val, formatv("v{0}", ++valueInScopeCount.top()));
     }
@@ -1384,6 +1403,30 @@ LogicalResult CppEmitter::emitVariableDeclaration(OpResult result,
   return success();
 }
 
+LogicalResult CppEmitter::emitGlobalVariable(GlobalOp op) {
+  if (op.getExternSpecifier())
+    os << "extern ";
+  else if (op.getStaticSpecifier())
+    os << "static ";
+  if (op.getConstSpecifier())
+    os << "const ";
+
+  if (failed(emitVariableDeclaration(op->getLoc(), op.getType(),
+                                     op.getSymName()))) {
+    return failure();
+  }
+
+  std::optional<Attribute> initialValue = op.getInitialValue();
+  if (initialValue) {
+    os << " = ";
+    if (failed(emitAttribute(op->getLoc(), *initialValue)))
+      return failure();
+  }
+
+  os << ";";
+  return success();
+}
+
 LogicalResult CppEmitter::emitAssignPrefix(Operation &op) {
   // If op is being emitted as part of an expression, bail out.
   if (getEmittedExpression())
@@ -1444,11 +1487,11 @@ LogicalResult CppEmitter::emitOperation(Operation &op, bool trailingSemicolon) {
                 emitc::CallOpaqueOp, emitc::CastOp, emitc::CmpOp,
                 emitc::ConditionalOp, emitc::ConstantOp, emitc::DeclareFuncOp,
                 emitc::DivOp, emitc::ExpressionOp, emitc::ForOp, emitc::FuncOp,
-                emitc::IfOp, emitc::IncludeOp, emitc::LogicalAndOp,
-                emitc::LogicalNotOp, emitc::LogicalOrOp, emitc::MulOp,
-                emitc::RemOp, emitc::ReturnOp, emitc::SubOp, emitc::SubscriptOp,
-                emitc::UnaryMinusOp, emitc::UnaryPlusOp, emitc::VariableOp,
-                emitc::VerbatimOp>(
+                emitc::GlobalOp, emitc::GetGlobalOp, emitc::IfOp,
+                emitc::IncludeOp, emitc::LogicalAndOp, emitc::LogicalNotOp,
+                emitc::LogicalOrOp, emitc::MulOp, emitc::RemOp, emitc::ReturnOp,
+                emitc::SubOp, emitc::SubscriptOp, emitc::UnaryMinusOp,
+                emitc::UnaryPlusOp, emitc::VariableOp, emitc::VerbatimOp>(
               [&](auto op) { return printOperation(*this, op); })
           // Func ops.
           .Case<func::CallOp, func::FuncOp, func::ReturnOp>(
@@ -1461,7 +1504,7 @@ LogicalResult CppEmitter::emitOperation(Operation &op, bool trailingSemicolon) {
   if (failed(status))
     return failure();
 
-  if (isa<emitc::LiteralOp, emitc::SubscriptOp>(op))
+  if (isa<emitc::LiteralOp, emitc::SubscriptOp, emitc::GetGlobalOp>(op))
     return success();
 
   if (getEmittedExpression() ||
diff --git a/mlir/lib/Target/LLVMIR/ConvertFromLLVMIR.cpp b/mlir/lib/Target/LLVMIR/ConvertFromLLVMIR.cpp
index 101add70c51e..7e2da1e0303c 100644
--- a/mlir/lib/Target/LLVMIR/ConvertFromLLVMIR.cpp
+++ b/mlir/lib/Target/LLVMIR/ConvertFromLLVMIR.cpp
@@ -56,6 +56,10 @@ void registerFromLLVMIRTranslation() {
         if (llvm::verifyModule(*llvmModule, &llvm::errs()))
           return nullptr;
 
+        // Debug records are not currently supported in the LLVM IR translator.
+        if (llvmModule->IsNewDbgInfoFormat)
+          llvmModule->convertFromNewDbgValues();
+
         return translateLLVMIRToModule(std::move(llvmModule), context,
                                        emitExpensiveWarnings,
                                        dropDICompositeTypeElements);
diff --git a/mlir/lib/Target/LLVMIR/Dialect/OpenMP/OpenMPToLLVMIRTranslation.cpp b/mlir/lib/Target/LLVMIR/Dialect/OpenMP/OpenMPToLLVMIRTranslation.cpp
index ebcdbc02aadd..9f87f89d8c63 100644
--- a/mlir/lib/Target/LLVMIR/Dialect/OpenMP/OpenMPToLLVMIRTranslation.cpp
+++ b/mlir/lib/Target/LLVMIR/Dialect/OpenMP/OpenMPToLLVMIRTranslation.cpp
@@ -916,35 +916,37 @@ static LogicalResult inlineReductionCleanup(
 static LogicalResult
 convertOmpWsloop(Operation &opInst, llvm::IRBuilderBase &builder,
                  LLVM::ModuleTranslation &moduleTranslation) {
-  auto loop = cast<omp::WsloopOp>(opInst);
-  const bool isByRef = loop.getByref();
+  auto wsloopOp = cast<omp::WsloopOp>(opInst);
+  auto loopOp = cast<omp::LoopNestOp>(wsloopOp.getWrappedLoop());
+  const bool isByRef = wsloopOp.getByref();
+
   // TODO: this should be in the op verifier instead.
-  if (loop.getLowerBound().empty())
+  if (loopOp.getLowerBound().empty())
     return failure();
 
   // Static is the default.
   auto schedule =
-      loop.getScheduleVal().value_or(omp::ClauseScheduleKind::Static);
+      wsloopOp.getScheduleVal().value_or(omp::ClauseScheduleKind::Static);
 
   // Find the loop configuration.
-  llvm::Value *step = moduleTranslation.lookupValue(loop.getStep()[0]);
+  llvm::Value *step = moduleTranslation.lookupValue(loopOp.getStep()[0]);
   llvm::Type *ivType = step->getType();
   llvm::Value *chunk = nullptr;
-  if (loop.getScheduleChunkVar()) {
+  if (wsloopOp.getScheduleChunkVar()) {
     llvm::Value *chunkVar =
-        moduleTranslation.lookupValue(loop.getScheduleChunkVar());
+        moduleTranslation.lookupValue(wsloopOp.getScheduleChunkVar());
     chunk = builder.CreateSExtOrTrunc(chunkVar, ivType);
   }
 
   SmallVector<omp::DeclareReductionOp> reductionDecls;
-  collectReductionDecls(loop, reductionDecls);
+  collectReductionDecls(wsloopOp, reductionDecls);
   llvm::OpenMPIRBuilder::InsertPointTy allocaIP =
       findAllocaInsertPoint(builder, moduleTranslation);
 
   SmallVector<llvm::Value *> privateReductionVariables;
   DenseMap<Value, llvm::Value *> reductionVariableMap;
   if (!isByRef) {
-    allocByValReductionVars(loop, builder, moduleTranslation, allocaIP,
+    allocByValReductionVars(wsloopOp, builder, moduleTranslation, allocaIP,
                             reductionDecls, privateReductionVariables,
                             reductionVariableMap);
   }
@@ -952,13 +954,12 @@ convertOmpWsloop(Operation &opInst, llvm::IRBuilderBase &builder,
   // Before the loop, store the initial values of reductions into reduction
   // variables. Although this could be done after allocas, we don't want to mess
   // up with the alloca insertion point.
-  MutableArrayRef<BlockArgument> reductionArgs =
-      loop.getRegion().getArguments().take_back(loop.getNumReductionVars());
-  for (unsigned i = 0; i < loop.getNumReductionVars(); ++i) {
+  ArrayRef<BlockArgument> reductionArgs = wsloopOp.getRegion().getArguments();
+  for (unsigned i = 0; i < wsloopOp.getNumReductionVars(); ++i) {
     SmallVector<llvm::Value *> phis;
 
     // map block argument to initializer region
-    mapInitializationArg(loop, moduleTranslation, reductionDecls, i);
+    mapInitializationArg(wsloopOp, moduleTranslation, reductionDecls, i);
 
     if (failed(inlineConvertOmpRegions(reductionDecls[i].getInitializerRegion(),
                                        "omp.reduction.neutral", builder,
@@ -977,7 +978,7 @@ convertOmpWsloop(Operation &opInst, llvm::IRBuilderBase &builder,
 
       privateReductionVariables.push_back(var);
       moduleTranslation.mapValue(reductionArgs[i], phis[0]);
-      reductionVariableMap.try_emplace(loop.getReductionVars()[i], phis[0]);
+      reductionVariableMap.try_emplace(wsloopOp.getReductionVars()[i], phis[0]);
     } else {
       // for by-ref case the store is inside of the reduction region
       builder.CreateStore(phis[0], privateReductionVariables[i]);
@@ -1008,33 +1009,34 @@ convertOmpWsloop(Operation &opInst, llvm::IRBuilderBase &builder,
   auto bodyGen = [&](llvm::OpenMPIRBuilder::InsertPointTy ip, llvm::Value *iv) {
     // Make sure further conversions know about the induction variable.
     moduleTranslation.mapValue(
-        loop.getRegion().front().getArgument(loopInfos.size()), iv);
+        loopOp.getRegion().front().getArgument(loopInfos.size()), iv);
 
     // Capture the body insertion point for use in nested loops. BodyIP of the
     // CanonicalLoopInfo always points to the beginning of the entry block of
     // the body.
     bodyInsertPoints.push_back(ip);
 
-    if (loopInfos.size() != loop.getNumLoops() - 1)
+    if (loopInfos.size() != loopOp.getNumLoops() - 1)
       return;
 
     // Convert the body of the loop.
     builder.restoreIP(ip);
-    convertOmpOpRegions(loop.getRegion(), "omp.wsloop.region", builder,
+    convertOmpOpRegions(loopOp.getRegion(), "omp.wsloop.region", builder,
                         moduleTranslation, bodyGenStatus);
   };
 
   // Delegate actual loop construction to the OpenMP IRBuilder.
-  // TODO: this currently assumes Wsloop is semantically similar to SCF loop,
-  // i.e. it has a positive step, uses signed integer semantics. Reconsider
-  // this code when Wsloop clearly supports more cases.
+  // TODO: this currently assumes omp.loop_nest is semantically similar to SCF
+  // loop, i.e. it has a positive step, uses signed integer semantics.
+  // Reconsider this code when the nested loop operation clearly supports more
+  // cases.
   llvm::OpenMPIRBuilder *ompBuilder = moduleTranslation.getOpenMPBuilder();
-  for (unsigned i = 0, e = loop.getNumLoops(); i < e; ++i) {
+  for (unsigned i = 0, e = loopOp.getNumLoops(); i < e; ++i) {
     llvm::Value *lowerBound =
-        moduleTranslation.lookupValue(loop.getLowerBound()[i]);
+        moduleTranslation.lookupValue(loopOp.getLowerBound()[i]);
     llvm::Value *upperBound =
-        moduleTranslation.lookupValue(loop.getUpperBound()[i]);
-    llvm::Value *step = moduleTranslation.lookupValue(loop.getStep()[i]);
+        moduleTranslation.lookupValue(loopOp.getUpperBound()[i]);
+    llvm::Value *step = moduleTranslation.lookupValue(loopOp.getStep()[i]);
 
     // Make sure loop trip count are emitted in the preheader of the outermost
     // loop at the latest so that they are all available for the new collapsed
@@ -1047,7 +1049,7 @@ convertOmpWsloop(Operation &opInst, llvm::IRBuilderBase &builder,
     }
     loopInfos.push_back(ompBuilder->createCanonicalLoop(
         loc, bodyGen, lowerBound, upperBound, step,
-        /*IsSigned=*/true, loop.getInclusive(), computeIP));
+        /*IsSigned=*/true, loopOp.getInclusive(), computeIP));
 
     if (failed(bodyGenStatus))
       return failure();
@@ -1062,13 +1064,13 @@ convertOmpWsloop(Operation &opInst, llvm::IRBuilderBase &builder,
   allocaIP = findAllocaInsertPoint(builder, moduleTranslation);
 
   // TODO: Handle doacross loops when the ordered clause has a parameter.
-  bool isOrdered = loop.getOrderedVal().has_value();
+  bool isOrdered = wsloopOp.getOrderedVal().has_value();
   std::optional<omp::ScheduleModifier> scheduleModifier =
-      loop.getScheduleModifier();
-  bool isSimd = loop.getSimdModifier();
+      wsloopOp.getScheduleModifier();
+  bool isSimd = wsloopOp.getSimdModifier();
 
   ompBuilder->applyWorkshareLoop(
-      ompLoc.DL, loopInfo, allocaIP, !loop.getNowait(),
+      ompLoc.DL, loopInfo, allocaIP, !wsloopOp.getNowait(),
       convertToScheduleKind(schedule), chunk, isSimd,
       scheduleModifier == omp::ScheduleModifier::monotonic,
       scheduleModifier == omp::ScheduleModifier::nonmonotonic, isOrdered);
@@ -1080,7 +1082,7 @@ convertOmpWsloop(Operation &opInst, llvm::IRBuilderBase &builder,
   builder.restoreIP(afterIP);
 
   // Process the reductions if required.
-  if (loop.getNumReductionVars() == 0)
+  if (wsloopOp.getNumReductionVars() == 0)
     return success();
 
   // Create the reduction generators. We need to own them here because
@@ -1088,7 +1090,7 @@ convertOmpWsloop(Operation &opInst, llvm::IRBuilderBase &builder,
   SmallVector<OwningReductionGen> owningReductionGens;
   SmallVector<OwningAtomicReductionGen> owningAtomicReductionGens;
   SmallVector<llvm::OpenMPIRBuilder::ReductionInfo> reductionInfos;
-  collectReductionInfo(loop, builder, moduleTranslation, reductionDecls,
+  collectReductionInfo(wsloopOp, builder, moduleTranslation, reductionDecls,
                        owningReductionGens, owningAtomicReductionGens,
                        privateReductionVariables, reductionInfos);
 
@@ -1099,9 +1101,9 @@ convertOmpWsloop(Operation &opInst, llvm::IRBuilderBase &builder,
   builder.SetInsertPoint(tempTerminator);
   llvm::OpenMPIRBuilder::InsertPointTy contInsertPoint =
       ompBuilder->createReductions(builder.saveIP(), allocaIP, reductionInfos,
-                                   loop.getNowait(), isByRef);
+                                   wsloopOp.getNowait(), isByRef);
   if (!contInsertPoint.getBlock())
-    return loop->emitOpError() << "failed to convert reductions";
+    return wsloopOp->emitOpError() << "failed to convert reductions";
   auto nextInsertionPoint =
       ompBuilder->createBarrier(contInsertPoint, llvm::omp::OMPD_for);
   tempTerminator->eraseFromParent();
diff --git a/mlir/lib/Tools/lsp-server-support/Transport.cpp b/mlir/lib/Tools/lsp-server-support/Transport.cpp
index 64dea35614c0..339c5f382516 100644
--- a/mlir/lib/Tools/lsp-server-support/Transport.cpp
+++ b/mlir/lib/Tools/lsp-server-support/Transport.cpp
@@ -51,12 +51,12 @@ private:
 
 Reply::Reply(const llvm::json::Value &id, llvm::StringRef method,
              JSONTransport &transport, std::mutex &transportOutputMutex)
-    : id(id), transport(&transport),
+    : method(method), id(id), transport(&transport),
       transportOutputMutex(transportOutputMutex) {}
 
 Reply::Reply(Reply &&other)
-    : replied(other.replied.load()), id(std::move(other.id)),
-      transport(other.transport),
+    : method(other.method), replied(other.replied.load()),
+      id(std::move(other.id)), transport(other.transport),
       transportOutputMutex(other.transportOutputMutex) {
   other.transport = nullptr;
 }
diff --git a/mlir/lib/Transforms/Mem2Reg.cpp b/mlir/lib/Transforms/Mem2Reg.cpp
index 0c1ce70f0708..71ba5bc076f0 100644
--- a/mlir/lib/Transforms/Mem2Reg.cpp
+++ b/mlir/lib/Transforms/Mem2Reg.cpp
@@ -191,13 +191,13 @@ private:
 
   /// Lazily-constructed default value representing the content of the slot when
   /// no store has been executed. This function may mutate IR.
-  Value getLazyDefaultValue();
+  Value getOrCreateDefaultValue();
 
   MemorySlot slot;
   PromotableAllocationOpInterface allocator;
   RewriterBase &rewriter;
-  /// Potentially non-initialized default value. Use `getLazyDefaultValue` to
-  /// initialize it on demand.
+  /// Potentially non-initialized default value. Use `getOrCreateDefaultValue`
+  /// to initialize it on demand.
   Value defaultValue;
   /// Contains the reaching definition at this operation. Reaching definitions
   /// are only computed for promotable memory operations with blocking uses.
@@ -232,7 +232,7 @@ MemorySlotPromoter::MemorySlotPromoter(
 #endif // NDEBUG
 }
 
-Value MemorySlotPromoter::getLazyDefaultValue() {
+Value MemorySlotPromoter::getOrCreateDefaultValue() {
   if (defaultValue)
     return defaultValue;
 
@@ -438,7 +438,7 @@ Value MemorySlotPromoter::computeReachingDefInBlock(Block *block,
 
       if (memOp.storesTo(slot)) {
         rewriter.setInsertionPointAfter(memOp);
-        Value stored = memOp.getStored(slot, rewriter, dataLayout);
+        Value stored = memOp.getStored(slot, rewriter, reachingDef, dataLayout);
         assert(stored && "a memory operation storing to a slot must provide a "
                          "new definition of the slot");
         reachingDef = stored;
@@ -452,6 +452,7 @@ Value MemorySlotPromoter::computeReachingDefInBlock(Block *block,
 
 void MemorySlotPromoter::computeReachingDefInRegion(Region *region,
                                                     Value reachingDef) {
+  assert(reachingDef && "expected an initial reaching def to be provided");
   if (region->hasOneBlock()) {
     computeReachingDefInBlock(&region->front(), reachingDef);
     return;
@@ -508,12 +509,11 @@ void MemorySlotPromoter::computeReachingDefInRegion(Region *region,
     }
 
     job.reachingDef = computeReachingDefInBlock(block, job.reachingDef);
+    assert(job.reachingDef);
 
     if (auto terminator = dyn_cast<BranchOpInterface>(block->getTerminator())) {
       for (BlockOperand &blockOperand : terminator->getBlockOperands()) {
         if (info.mergePoints.contains(blockOperand.get())) {
-          if (!job.reachingDef)
-            job.reachingDef = getLazyDefaultValue();
           rewriter.modifyOpInPlace(terminator, [&]() {
             terminator.getSuccessorOperands(blockOperand.getOperandNumber())
                 .append(job.reachingDef);
@@ -567,7 +567,7 @@ void MemorySlotPromoter::removeBlockingUses() {
       // If no reaching definition is known, this use is outside the reach of
       // the slot. The default value should thus be used.
       if (!reachingDef)
-        reachingDef = getLazyDefaultValue();
+        reachingDef = getOrCreateDefaultValue();
 
       rewriter.setInsertionPointAfter(toPromote);
       if (toPromoteMemOp.removeBlockingUses(
@@ -601,7 +601,8 @@ void MemorySlotPromoter::removeBlockingUses() {
 }
 
 void MemorySlotPromoter::promoteSlot() {
-  computeReachingDefInRegion(slot.ptr.getParentRegion(), {});
+  computeReachingDefInRegion(slot.ptr.getParentRegion(),
+                             getOrCreateDefaultValue());
 
   // Now that reaching definitions are known, remove all users.
   removeBlockingUses();
@@ -617,7 +618,7 @@ void MemorySlotPromoter::promoteSlot() {
              succOperands.size() + 1 == mergePoint->getNumArguments());
       if (succOperands.size() + 1 == mergePoint->getNumArguments())
         rewriter.modifyOpInPlace(
-            user, [&]() { succOperands.append(getLazyDefaultValue()); });
+            user, [&]() { succOperands.append(getOrCreateDefaultValue()); });
     }
   }
 
diff --git a/mlir/lib/Transforms/Utils/DialectConversion.cpp b/mlir/lib/Transforms/Utils/DialectConversion.cpp
index d85938847c77..d407d60334c7 100644
--- a/mlir/lib/Transforms/Utils/DialectConversion.cpp
+++ b/mlir/lib/Transforms/Utils/DialectConversion.cpp
@@ -2072,6 +2072,10 @@ OperationLegalizer::legalizeWithFold(Operation *op,
     LLVM_DEBUG(logFailure(rewriterImpl.logger, "unable to fold"));
     return failure();
   }
+  // An empty list of replacement values indicates that the fold was in-place.
+  // As the operation changed, a new legalization needs to be attempted.
+  if (replacementValues.empty())
+    return legalize(op, rewriter);
 
   // Insert a replacement for 'op' with the folded replacement values.
   rewriter.replaceOp(op, replacementValues);
diff --git a/mlir/python/mlir/dialects/LLVMOps.td b/mlir/python/mlir/dialects/LLVMOps.td
index dcf2f4245cf4..30f047f21698 100644
--- a/mlir/python/mlir/dialects/LLVMOps.td
+++ b/mlir/python/mlir/dialects/LLVMOps.td
@@ -10,5 +10,6 @@
 #define PYTHON_BINDINGS_LLVM_OPS
 
 include "mlir/Dialect/LLVMIR/LLVMOps.td"
+include "mlir/Dialect/LLVMIR/LLVMIntrinsicOps.td"
 
 #endif
diff --git a/mlir/python/mlir/dialects/linalg/opdsl/lang/comprehension.py b/mlir/python/mlir/dialects/linalg/opdsl/lang/comprehension.py
index 23d6d26b7e29..bb43ebf2b692 100644
--- a/mlir/python/mlir/dialects/linalg/opdsl/lang/comprehension.py
+++ b/mlir/python/mlir/dialects/linalg/opdsl/lang/comprehension.py
@@ -291,6 +291,12 @@ class UnaryFn:
     ceil = UnaryFnType("ceil")
     floor = UnaryFnType("floor")
     negf = UnaryFnType("negf")
+    round = UnaryFnType("round")
+    sqrt = UnaryFnType("sqrt")
+    rsqrt = UnaryFnType("rsqrt")
+    square = UnaryFnType("square")
+    tanh = UnaryFnType("tanh")
+    erf = UnaryFnType("erf")
 
 
 class BinaryFnType:
@@ -330,6 +336,7 @@ class BinaryFn:
     min_signed = BinaryFnType("min_signed")
     max_unsigned = BinaryFnType("max_unsigned")
     min_unsigned = BinaryFnType("min_unsigned")
+    powf = BinaryFnType("powf")
 
 
 class TypeFnType:
diff --git a/mlir/python/mlir/dialects/linalg/opdsl/ops/core_named_ops.py b/mlir/python/mlir/dialects/linalg/opdsl/ops/core_named_ops.py
index 5b05364f6d35..ca2bb0c5f7f8 100644
--- a/mlir/python/mlir/dialects/linalg/opdsl/ops/core_named_ops.py
+++ b/mlir/python/mlir/dialects/linalg/opdsl/ops/core_named_ops.py
@@ -109,6 +109,78 @@ def negf(
 
 
 @linalg_structured_op
+def round(
+    I=TensorDef(T1),
+    O=TensorDef(T1, output=True),
+):
+    """Applies round(x) elementwise.
+
+    No numeric casting is performed on the input operand.
+    """
+    O[None] = UnaryFn.round(I[None])
+
+
+@linalg_structured_op
+def sqrt(
+    I=TensorDef(T1),
+    O=TensorDef(T1, output=True),
+):
+    """Applies sqrt(x) elementwise.
+
+    No numeric casting is performed on the input operand.
+    """
+    O[None] = UnaryFn.sqrt(I[None])
+
+
+@linalg_structured_op
+def rsqrt(
+    I=TensorDef(T1),
+    O=TensorDef(T1, output=True),
+):
+    """Applies rsqrt(x) elementwise.
+
+    No numeric casting is performed on the input operand.
+    """
+    O[None] = UnaryFn.rsqrt(I[None])
+
+
+@linalg_structured_op
+def square(
+    I=TensorDef(T1),
+    O=TensorDef(T1, output=True),
+):
+    """Applies square(x) elementwise.
+
+    No numeric casting is performed on the input operand.
+    """
+    O[None] = UnaryFn.square(I[None])
+
+
+@linalg_structured_op
+def tanh(
+    I=TensorDef(T1),
+    O=TensorDef(T1, output=True),
+):
+    """Applies tanh(x) elementwise.
+
+    No numeric casting is performed on the input operand.
+    """
+    O[None] = UnaryFn.tanh(I[None])
+
+
+@linalg_structured_op
+def erf(
+    I=TensorDef(T1),
+    O=TensorDef(T1, output=True),
+):
+    """Applies erf(x) elementwise.
+
+    No numeric casting is performed on the input operand.
+    """
+    O[None] = UnaryFn.erf(I[None])
+
+
+@linalg_structured_op
 def elemwise_binary(
     lhs=TensorDef(T1),
     rhs=TensorDef(T2),
@@ -233,13 +305,53 @@ def max(
 
     This means reduction/broadcast/element cast semantics is explicit. Further
     passes can take that into account when lowering this code. For example,
-    a `linalg.broadcast` + `linalg.div` sequence can be lowered to a
+    a `linalg.broadcast` + `linalg.max` sequence can be lowered to a
     `linalg.generic` with different affine maps for the two operands.
     """
     O[None] = BinaryFn.max_signed(lhs[None], rhs[None])
 
 
 @linalg_structured_op
+def min(
+    lhs=TensorDef(T1),
+    rhs=TensorDef(T1),
+    O=TensorDef(T1, output=True),
+):
+    """Takes the min (signed) between two inputs, elementwise.
+
+    The shapes and element types must be identical. The appropriate casts,
+    broadcasts and reductions should be done previously to calling this op.
+
+    This means reduction/broadcast/element cast semantics is explicit. Further
+    passes can take that into account when lowering this code. For example,
+    a `linalg.broadcast` + `linalg.min` sequence can be lowered to a
+    `linalg.generic` with different affine maps for the two operands.
+    """
+    O[None] = BinaryFn.min_signed(lhs[None], rhs[None])
+
+
+@linalg_structured_op
+def powf(
+    lhs=TensorDef(T1),
+    rhs=TensorDef(T1),
+    O=TensorDef(T1, output=True),
+):
+    """Takes the powf(lhs, rhs) between two inputs, elementwise. For powf(arg, 2) use `linalg.square`.
+
+    Only applies to floating point values.
+
+    The shapes and element types must be identical. The appropriate casts,
+    broadcasts and reductions should be done previously to calling this op.
+
+    This means reduction/broadcast/element cast semantics is explicit. Further
+    passes can take that into account when lowering this code. For example,
+    a `linalg.broadcast` + `linalg.powf` sequence can be lowered to a
+    `linalg.generic` with different affine maps for the two operands.
+    """
+    O[None] = BinaryFn.powf(lhs[None], rhs[None])
+
+
+@linalg_structured_op
 def matmul(
     A=TensorDef(T1, S.M, S.K),
     B=TensorDef(T2, S.K, S.N),
diff --git a/mlir/python/mlir/dialects/llvm.py b/mlir/python/mlir/dialects/llvm.py
index 8aa16e4a2560..941a584966dc 100644
--- a/mlir/python/mlir/dialects/llvm.py
+++ b/mlir/python/mlir/dialects/llvm.py
@@ -5,3 +5,11 @@
 from ._llvm_ops_gen import *
 from ._llvm_enum_gen import *
 from .._mlir_libs._mlirDialectsLLVM import *
+from ..ir import Value
+from ._ods_common import get_op_result_or_op_results as _get_op_result_or_op_results
+
+
+def mlir_constant(value, *, loc=None, ip=None) -> Value:
+    return _get_op_result_or_op_results(
+        ConstantOp(res=value.type, value=value, loc=loc, ip=ip)
+    )
diff --git a/mlir/python/mlir/dialects/transform/interpreter/__init__.py b/mlir/python/mlir/dialects/transform/interpreter/__init__.py
index 34cdc43cb617..e69aa963038f 100644
--- a/mlir/python/mlir/dialects/transform/interpreter/__init__.py
+++ b/mlir/python/mlir/dialects/transform/interpreter/__init__.py
@@ -29,7 +29,7 @@ def apply_named_sequence(
     if transform_options is None:
         _cextTransformInterpreter.apply_named_sequence(*args)
     else:
-        _cextTransformInterpreter(*args, transform_options)
+        _cextTransformInterpreter.apply_named_sequence(*args, transform_options)
 
 
 def copy_symbols_and_merge_into(target, other):
diff --git a/mlir/test/CAPI/execution_engine.c b/mlir/test/CAPI/execution_engine.c
index 38a8fb8c3e21..81ff8477ffd7 100644
--- a/mlir/test/CAPI/execution_engine.c
+++ b/mlir/test/CAPI/execution_engine.c
@@ -99,8 +99,11 @@ void testOmpCreation(void) {
 "    %1 = arith.constant 1 : i32                                                \n"
 "    %2 = arith.constant 2 : i32                                                \n"
 "    omp.parallel {                                                             \n"
-"      omp.wsloop for (%3) : i32 = (%0) to (%2) step (%1) {                     \n"
-"        omp.yield                                                              \n"
+"      omp.wsloop {                                                             \n"
+"        omp.loop_nest (%3) : i32 = (%0) to (%2) step (%1) {                    \n"
+"          omp.yield                                                            \n"
+"        }                                                                      \n"
+"        omp.terminator                                                         \n"
 "      }                                                                        \n"
 "      omp.terminator                                                           \n"
 "    }                                                                          \n"
diff --git a/mlir/test/CAPI/sparse_tensor.c b/mlir/test/CAPI/sparse_tensor.c
index f241e0e5c2fb..22b7052b732a 100644
--- a/mlir/test/CAPI/sparse_tensor.c
+++ b/mlir/test/CAPI/sparse_tensor.c
@@ -27,7 +27,7 @@ static int testRoundtripEncoding(MlirContext ctx) {
   const char *originalAsm =
     "#sparse_tensor.encoding<{ "
     "map = [s0](d0, d1) -> (s0 : dense, d0 : compressed, d1 : compressed), "
-    "posWidth = 32, crdWidth = 64 }>";
+    "posWidth = 32, crdWidth = 64, explicitVal = 1 : i64}>";
   // clang-format on
   MlirAttribute originalAttr =
       mlirAttributeParseGet(ctx, mlirStringRefCreateFromCString(originalAsm));
@@ -56,8 +56,21 @@ static int testRoundtripEncoding(MlirContext ctx) {
   // CHECK: crdWidth: 64
   int crdWidth = mlirSparseTensorEncodingAttrGetCrdWidth(originalAttr);
   fprintf(stderr, "crdWidth: %d\n", crdWidth);
+
+  // CHECK: explicitVal: 1 : i64
+  MlirAttribute explicitVal =
+      mlirSparseTensorEncodingAttrGetExplicitVal(originalAttr);
+  fprintf(stderr, "explicitVal: ");
+  mlirAttributeDump(explicitVal);
+  // CHECK: implicitVal: <<NULL ATTRIBUTE>>
+  MlirAttribute implicitVal =
+      mlirSparseTensorEncodingAttrGetImplicitVal(originalAttr);
+  fprintf(stderr, "implicitVal: ");
+  mlirAttributeDump(implicitVal);
+
   MlirAttribute newAttr = mlirSparseTensorEncodingAttrGet(
-      ctx, lvlRank, lvlTypes, dimToLvl, lvlToDim, posWidth, crdWidth);
+      ctx, lvlRank, lvlTypes, dimToLvl, lvlToDim, posWidth, crdWidth,
+      explicitVal, implicitVal);
   mlirAttributeDump(newAttr); // For debugging filecheck output.
   // CHECK: equal: 1
   fprintf(stderr, "equal: %d\n", mlirAttributeEqual(originalAttr, newAttr));
diff --git a/mlir/test/Conversion/MemRefToEmitC/memref-to-emitc-failed.mlir b/mlir/test/Conversion/MemRefToEmitC/memref-to-emitc-failed.mlir
index 390190d341e5..89dafa7529ed 100644
--- a/mlir/test/Conversion/MemRefToEmitC/memref-to-emitc-failed.mlir
+++ b/mlir/test/Conversion/MemRefToEmitC/memref-to-emitc-failed.mlir
@@ -38,3 +38,8 @@ func.func @zero_rank() {
   %0 = memref.alloca() : memref<f32>
   return
 }
+
+// -----
+
+// expected-error@+1 {{failed to legalize operation 'memref.global'}}
+memref.global "nested" constant @nested_global : memref<3x7xf32>
diff --git a/mlir/test/Conversion/MemRefToEmitC/memref-to-emitc.mlir b/mlir/test/Conversion/MemRefToEmitC/memref-to-emitc.mlir
index 7aa2ba88843a..bc40ef48268e 100644
--- a/mlir/test/Conversion/MemRefToEmitC/memref-to-emitc.mlir
+++ b/mlir/test/Conversion/MemRefToEmitC/memref-to-emitc.mlir
@@ -11,6 +11,7 @@ func.func @memref_store(%v : f32, %i: index, %j: index) {
   memref.store %v, %0[%i, %j] : memref<4x8xf32>
   return
 }
+
 // -----
 
 // CHECK-LABEL: memref_load
@@ -26,3 +27,21 @@ func.func @memref_load(%i: index, %j: index) -> f32 {
   // CHECK: return %[[VAR]] : f32
   return %1 : f32
 }
+
+// -----
+
+// CHECK-LABEL: globals
+module @globals {
+  memref.global "private" constant @internal_global : memref<3x7xf32> = dense<4.0>
+  // CHECK: emitc.global static const @internal_global : !emitc.array<3x7xf32> = dense<4.000000e+00>
+  memref.global @public_global : memref<3x7xf32>
+  // CHECK: emitc.global extern @public_global : !emitc.array<3x7xf32>
+  memref.global @uninitialized_global : memref<3x7xf32> = uninitialized
+  // CHECK: emitc.global extern @uninitialized_global : !emitc.array<3x7xf32>
+
+  func.func @use_global() {
+    // CHECK: emitc.get_global @public_global : !emitc.array<3x7xf32>
+    %0 = memref.get_global @public_global : memref<3x7xf32>
+    return
+  }
+}
diff --git a/mlir/test/Conversion/OpenMPToLLVM/convert-to-llvmir.mlir b/mlir/test/Conversion/OpenMPToLLVM/convert-to-llvmir.mlir
index 9f45d139b81f..3aeb9e70522d 100644
--- a/mlir/test/Conversion/OpenMPToLLVM/convert-to-llvmir.mlir
+++ b/mlir/test/Conversion/OpenMPToLLVM/convert-to-llvmir.mlir
@@ -71,15 +71,18 @@ func.func @branch_loop() {
 func.func @wsloop(%arg0: index, %arg1: index, %arg2: index, %arg3: index, %arg4: index, %arg5: index) {
   // CHECK: omp.parallel
   omp.parallel {
-    // CHECK: omp.wsloop for (%[[ARG6:.*]], %[[ARG7:.*]]) : i64 = (%[[ARG0]], %[[ARG1]]) to (%[[ARG2]], %[[ARG3]]) step (%[[ARG4]], %[[ARG5]]) {
-    "omp.wsloop"(%arg0, %arg1, %arg2, %arg3, %arg4, %arg5) ({
-    ^bb0(%arg6: index, %arg7: index):
-      // CHECK-DAG: %[[CAST_ARG6:.*]] = builtin.unrealized_conversion_cast %[[ARG6]] : i64 to index
-      // CHECK-DAG: %[[CAST_ARG7:.*]] = builtin.unrealized_conversion_cast %[[ARG7]] : i64 to index
-      // CHECK: "test.payload"(%[[CAST_ARG6]], %[[CAST_ARG7]]) : (index, index) -> ()
-      "test.payload"(%arg6, %arg7) : (index, index) -> ()
-      omp.yield
-    }) {operandSegmentSizes = array<i32: 2, 2, 2, 0, 0, 0, 0>} : (index, index, index, index, index, index) -> ()
+    // CHECK: omp.wsloop {
+    "omp.wsloop"() ({
+      // CHECK: omp.loop_nest (%[[ARG6:.*]], %[[ARG7:.*]]) : i64 = (%[[ARG0]], %[[ARG1]]) to (%[[ARG2]], %[[ARG3]]) step (%[[ARG4]], %[[ARG5]]) {
+      omp.loop_nest (%arg6, %arg7) : index = (%arg0, %arg1) to (%arg2, %arg3) step (%arg4, %arg5) {
+        // CHECK-DAG: %[[CAST_ARG6:.*]] = builtin.unrealized_conversion_cast %[[ARG6]] : i64 to index
+        // CHECK-DAG: %[[CAST_ARG7:.*]] = builtin.unrealized_conversion_cast %[[ARG7]] : i64 to index
+        // CHECK: "test.payload"(%[[CAST_ARG6]], %[[CAST_ARG7]]) : (index, index) -> ()
+        "test.payload"(%arg6, %arg7) : (index, index) -> ()
+        omp.yield
+      }
+      omp.terminator
+    }) : () -> ()
     omp.terminator
   }
   return
@@ -323,12 +326,14 @@ llvm.func @_QPsb() {
 // CHECK-LABEL:  @_QPsimple_reduction
 // CHECK:    %[[RED_ACCUMULATOR:.*]] = llvm.alloca %{{.*}} x i32 {bindc_name = "x", uniq_name = "_QFsimple_reductionEx"} : (i64) -> !llvm.ptr
 // CHECK:    omp.parallel
-// CHECK:      omp.wsloop reduction(@eqv_reduction %{{.+}} -> %[[PRV:.+]] : !llvm.ptr) for
-// CHECK:        %[[LPRV:.+]] = llvm.load %[[PRV]] : !llvm.ptr -> i32
-// CHECK:        %[[CMP:.+]] = llvm.icmp "eq" %{{.*}}, %[[LPRV]] : i32
-// CHECK:        %[[ZEXT:.+]] = llvm.zext %[[CMP]] : i1 to i32
-// CHECK:        llvm.store %[[ZEXT]], %[[PRV]] : i32, !llvm.ptr
-// CHECK:        omp.yield
+// CHECK:      omp.wsloop reduction(@eqv_reduction %{{.+}} -> %[[PRV:.+]] : !llvm.ptr)
+// CHECK-NEXT:   omp.loop_nest {{.*}}{
+// CHECK:          %[[LPRV:.+]] = llvm.load %[[PRV]] : !llvm.ptr -> i32
+// CHECK:          %[[CMP:.+]] = llvm.icmp "eq" %{{.*}}, %[[LPRV]] : i32
+// CHECK:          %[[ZEXT:.+]] = llvm.zext %[[CMP]] : i1 to i32
+// CHECK:          llvm.store %[[ZEXT]], %[[PRV]] : i32, !llvm.ptr
+// CHECK:          omp.yield
+// CHECK:        omp.terminator
 // CHECK:      omp.terminator
 // CHECK:    llvm.return
 
@@ -354,20 +359,23 @@ llvm.func @_QPsimple_reduction(%arg0: !llvm.ptr {fir.bindc_name = "y"}) {
   %4 = llvm.alloca %3 x i32 {bindc_name = "x", uniq_name = "_QFsimple_reductionEx"} : (i64) -> !llvm.ptr
   %5 = llvm.zext %2 : i1 to i32
   llvm.store %5, %4 : i32, !llvm.ptr
-  omp.parallel   {
+  omp.parallel {
     %6 = llvm.alloca %3 x i32 {adapt.valuebyref, in_type = i32, operandSegmentSizes = array<i32: 0, 0>, pinned} : (i64) -> !llvm.ptr
-    omp.wsloop   reduction(@eqv_reduction %4 -> %prv : !llvm.ptr) for  (%arg1) : i32 = (%1) to (%0) inclusive step (%1) {
-      llvm.store %arg1, %6 : i32, !llvm.ptr
-      %7 = llvm.load %6 : !llvm.ptr -> i32
-      %8 = llvm.sext %7 : i32 to i64
-      %9 = llvm.sub %8, %3  : i64
-      %10 = llvm.getelementptr %arg0[0, %9] : (!llvm.ptr, i64) -> !llvm.ptr, !llvm.array<100 x i32>
-      %11 = llvm.load %10 : !llvm.ptr -> i32
-      %12 = llvm.load %prv : !llvm.ptr -> i32
-      %13 = llvm.icmp "eq" %11, %12 : i32
-      %14 = llvm.zext %13 : i1 to i32
-      llvm.store %14, %prv : i32, !llvm.ptr
-      omp.yield
+    omp.wsloop reduction(@eqv_reduction %4 -> %prv : !llvm.ptr) {
+      omp.loop_nest (%arg1) : i32 = (%1) to (%0) inclusive step (%1) {
+        llvm.store %arg1, %6 : i32, !llvm.ptr
+        %7 = llvm.load %6 : !llvm.ptr -> i32
+        %8 = llvm.sext %7 : i32 to i64
+        %9 = llvm.sub %8, %3  : i64
+        %10 = llvm.getelementptr %arg0[0, %9] : (!llvm.ptr, i64) -> !llvm.ptr, !llvm.array<100 x i32>
+        %11 = llvm.load %10 : !llvm.ptr -> i32
+        %12 = llvm.load %prv : !llvm.ptr -> i32
+        %13 = llvm.icmp "eq" %11, %12 : i32
+        %14 = llvm.zext %13 : i1 to i32
+        llvm.store %14, %prv : i32, !llvm.ptr
+        omp.yield
+      }
+      omp.terminator
     }
     omp.terminator
   }
diff --git a/mlir/test/Conversion/SCFToOpenMP/reductions.mlir b/mlir/test/Conversion/SCFToOpenMP/reductions.mlir
index 3b6c145d62f1..fc6d56559c26 100644
--- a/mlir/test/Conversion/SCFToOpenMP/reductions.mlir
+++ b/mlir/test/Conversion/SCFToOpenMP/reductions.mlir
@@ -28,6 +28,7 @@ func.func @reduction1(%arg0 : index, %arg1 : index, %arg2 : index,
   // CHECK: omp.parallel
   // CHECK: omp.wsloop
   // CHECK-SAME: reduction(@[[$REDF]] %[[BUF]] -> %[[PVT_BUF:[a-z0-9]+]]
+  // CHECK: omp.loop_nest
   // CHECK: memref.alloca_scope
   scf.parallel (%i0, %i1) = (%arg0, %arg1) to (%arg2, %arg3)
                             step (%arg4, %step) init (%zero) -> (f32) {
@@ -43,6 +44,7 @@ func.func @reduction1(%arg0 : index, %arg1 : index, %arg2 : index,
     }
     // CHECK: omp.yield
   }
+  // CHECK:   omp.terminator
   // CHECK: omp.terminator
   // CHECK: llvm.load %[[BUF]]
   return
@@ -107,6 +109,7 @@ func.func @reduction_muli(%arg0 : index, %arg1 : index, %arg2 : index,
   %one = arith.constant 1 : i32
   // CHECK: %[[RED_VAR:.*]] = llvm.alloca %{{.*}} x i32 : (i64) -> !llvm.ptr
   // CHECK: omp.wsloop reduction(@[[$REDI]] %[[RED_VAR]] -> %[[RED_PVT_VAR:.*]] : !llvm.ptr)
+  // CHECK: omp.loop_nest
   scf.parallel (%i0, %i1) = (%arg0, %arg1) to (%arg2, %arg3)
                             step (%arg4, %step) init (%one) -> (i32) {
     // CHECK: %[[C2:.*]] = arith.constant 2 : i32
@@ -208,6 +211,7 @@ func.func @reduction4(%arg0 : index, %arg1 : index, %arg2 : index,
   // CHECK: omp.wsloop
   // CHECK-SAME: reduction(@[[$REDF1]] %[[BUF1]] -> %[[PVT_BUF1:[a-z0-9]+]]
   // CHECK-SAME:           @[[$REDF2]] %[[BUF2]] -> %[[PVT_BUF2:[a-z0-9]+]]
+  // CHECK: omp.loop_nest
   // CHECK: memref.alloca_scope
   %res:2 = scf.parallel (%i0, %i1) = (%arg0, %arg1) to (%arg2, %arg3)
                         step (%arg4, %step) init (%zero, %ione) -> (f32, i64) {
@@ -236,6 +240,7 @@ func.func @reduction4(%arg0 : index, %arg1 : index, %arg2 : index,
     }
     // CHECK: omp.yield
   }
+  // CHECK:   omp.terminator
   // CHECK: omp.terminator
   // CHECK: %[[RES1:.*]] = llvm.load %[[BUF1]] : !llvm.ptr -> f32
   // CHECK: %[[RES2:.*]] = llvm.load %[[BUF2]] : !llvm.ptr -> i64
diff --git a/mlir/test/Conversion/SCFToOpenMP/scf-to-openmp.mlir b/mlir/test/Conversion/SCFToOpenMP/scf-to-openmp.mlir
index acd2690c56e2..b2f19d294cb5 100644
--- a/mlir/test/Conversion/SCFToOpenMP/scf-to-openmp.mlir
+++ b/mlir/test/Conversion/SCFToOpenMP/scf-to-openmp.mlir
@@ -2,10 +2,11 @@
 
 // CHECK-LABEL: @parallel
 func.func @parallel(%arg0: index, %arg1: index, %arg2: index,
-          %arg3: index, %arg4: index, %arg5: index) {
+                    %arg3: index, %arg4: index, %arg5: index) {
   // CHECK: %[[FOUR:.+]] = llvm.mlir.constant(4 : i32) : i32
   // CHECK: omp.parallel num_threads(%[[FOUR]] : i32) {
-  // CHECK: omp.wsloop for (%[[LVAR1:.*]], %[[LVAR2:.*]]) : index = (%arg0, %arg1) to (%arg2, %arg3) step (%arg4, %arg5) {
+  // CHECK: omp.wsloop {
+  // CHECK: omp.loop_nest (%[[LVAR1:.*]], %[[LVAR2:.*]]) : index = (%arg0, %arg1) to (%arg2, %arg3) step (%arg4, %arg5) {
   // CHECK: memref.alloca_scope
   scf.parallel (%i, %j) = (%arg0, %arg1) to (%arg2, %arg3) step (%arg4, %arg5) {
     // CHECK: "test.payload"(%[[LVAR1]], %[[LVAR2]]) : (index, index) -> ()
@@ -13,6 +14,8 @@ func.func @parallel(%arg0: index, %arg1: index, %arg2: index,
     // CHECK:   omp.yield
     // CHECK: }
   }
+  // CHECK:     omp.terminator
+  // CHECK:   }
   // CHECK:   omp.terminator
   // CHECK: }
   return
@@ -23,20 +26,26 @@ func.func @nested_loops(%arg0: index, %arg1: index, %arg2: index,
                    %arg3: index, %arg4: index, %arg5: index) {
   // CHECK: %[[FOUR:.+]] = llvm.mlir.constant(4 : i32) : i32
   // CHECK: omp.parallel num_threads(%[[FOUR]] : i32) {
-  // CHECK: omp.wsloop for (%[[LVAR_OUT1:.*]]) : index = (%arg0) to (%arg2) step (%arg4) {
-    // CHECK: memref.alloca_scope
+  // CHECK: omp.wsloop {
+  // CHECK: omp.loop_nest (%[[LVAR_OUT1:.*]]) : index = (%arg0) to (%arg2) step (%arg4) {
+  // CHECK: memref.alloca_scope
   scf.parallel (%i) = (%arg0) to (%arg2) step (%arg4) {
     // CHECK: omp.parallel
-    // CHECK: omp.wsloop for (%[[LVAR_IN1:.*]]) : index = (%arg1) to (%arg3) step (%arg5) {
+    // CHECK: omp.wsloop {
+    // CHECK: omp.loop_nest (%[[LVAR_IN1:.*]]) : index = (%arg1) to (%arg3) step (%arg5) {
     // CHECK: memref.alloca_scope
     scf.parallel (%j) = (%arg1) to (%arg3) step (%arg5) {
       // CHECK: "test.payload"(%[[LVAR_OUT1]], %[[LVAR_IN1]]) : (index, index) -> ()
       "test.payload"(%i, %j) : (index, index) -> ()
       // CHECK: }
     }
-    // CHECK:   omp.yield
+    // CHECK:     omp.yield
+    // CHECK:   }
+    // CHECK:   omp.terminator
     // CHECK: }
   }
+  // CHECK:     omp.terminator
+  // CHECK:   }
   // CHECK:   omp.terminator
   // CHECK: }
   return
@@ -47,7 +56,8 @@ func.func @adjacent_loops(%arg0: index, %arg1: index, %arg2: index,
                      %arg3: index, %arg4: index, %arg5: index) {
   // CHECK: %[[FOUR:.+]] = llvm.mlir.constant(4 : i32) : i32
   // CHECK: omp.parallel num_threads(%[[FOUR]] : i32) {
-  // CHECK: omp.wsloop for (%[[LVAR_AL1:.*]]) : index = (%arg0) to (%arg2) step (%arg4) {
+  // CHECK: omp.wsloop {
+  // CHECK: omp.loop_nest (%[[LVAR_AL1:.*]]) : index = (%arg0) to (%arg2) step (%arg4) {
   // CHECK: memref.alloca_scope
   scf.parallel (%i) = (%arg0) to (%arg2) step (%arg4) {
     // CHECK: "test.payload1"(%[[LVAR_AL1]]) : (index) -> ()
@@ -55,12 +65,15 @@ func.func @adjacent_loops(%arg0: index, %arg1: index, %arg2: index,
     // CHECK:   omp.yield
     // CHECK: }
   }
+  // CHECK:     omp.terminator
+  // CHECK:   }
   // CHECK:   omp.terminator
   // CHECK: }
 
   // CHECK: %[[FOUR:.+]] = llvm.mlir.constant(4 : i32) : i32
   // CHECK: omp.parallel num_threads(%[[FOUR]] : i32) {
-  // CHECK: omp.wsloop for (%[[LVAR_AL2:.*]]) : index = (%arg1) to (%arg3) step (%arg5) {
+  // CHECK: omp.wsloop {
+  // CHECK: omp.loop_nest (%[[LVAR_AL2:.*]]) : index = (%arg1) to (%arg3) step (%arg5) {
   // CHECK: memref.alloca_scope
   scf.parallel (%j) = (%arg1) to (%arg3) step (%arg5) {
     // CHECK: "test.payload2"(%[[LVAR_AL2]]) : (index) -> ()
@@ -68,6 +81,8 @@ func.func @adjacent_loops(%arg0: index, %arg1: index, %arg2: index,
     // CHECK:   omp.yield
     // CHECK: }
   }
+  // CHECK:     omp.terminator
+  // CHECK:   }
   // CHECK:   omp.terminator
   // CHECK: }
   return
diff --git a/mlir/test/Conversion/TosaToLinalg/tosa-to-linalg-named.mlir b/mlir/test/Conversion/TosaToLinalg/tosa-to-linalg-named.mlir
index b4049000c50d..39699ee315e6 100644
--- a/mlir/test/Conversion/TosaToLinalg/tosa-to-linalg-named.mlir
+++ b/mlir/test/Conversion/TosaToLinalg/tosa-to-linalg-named.mlir
@@ -503,6 +503,19 @@ func.func @avg_pool_dyn(%arg0: tensor<?x6x34x62xf32>) -> (tensor<?x5x33x62xf32>)
 
 // -----
 
+// CHECK: #[[$MAP1:.+]] = affine_map<(d0, d1, d2, d3) -> (0)>
+// CHECK: #[[$MAP2:.+]] = affine_map<(d0, d1, d2, d3) -> (d0, d1, d2, d3)>
+
+// CHECK-LABEL: @conv2d_scalar_bias_f32
+func.func @conv2d_scalar_bias_f32(%input: tensor<1x49x42x27xf32>, %weights: tensor<28x3x3x27xf32>, %bias: tensor<1xf32>) -> () {
+  // CHECK: %[[INIT:.+]] = tensor.empty() : tensor<1x45x40x28xf32>
+  // CHECK: %[[BROADCAST:.+]] = linalg.generic {indexing_maps = [#[[$MAP1]], #[[$MAP2]]], iterator_types = ["parallel", "parallel", "parallel", "parallel"]} ins(%arg2 : tensor<1xf32>) outs(%[[INIT]] : tensor<1x45x40x28xf32>) {
+  %0 = tosa.conv2d %input, %weights, %bias {pad = array<i64: 0, 0, 0, 0>, stride = array<i64: 1, 1>, dilation = array<i64: 2, 1>} : (tensor<1x49x42x27xf32>, tensor<28x3x3x27xf32>, tensor<1xf32>) -> tensor<1x45x40x28xf32>
+  return
+}
+
+// -----
+
 // CHECK: #[[$MAP1:.+]] = affine_map<(d0, d1, d2, d3) -> (d3)>
 // CHECK: #[[$MAP2:.+]] = affine_map<(d0, d1, d2, d3) -> (d0, d1, d2, d3)>
 
diff --git a/mlir/test/Conversion/VectorToArmSME/vector-to-arm-sme.mlir b/mlir/test/Conversion/VectorToArmSME/vector-to-arm-sme.mlir
index d3f02c6288a2..ce0b46e0f061 100644
--- a/mlir/test/Conversion/VectorToArmSME/vector-to-arm-sme.mlir
+++ b/mlir/test/Conversion/VectorToArmSME/vector-to-arm-sme.mlir
@@ -620,3 +620,626 @@ func.func @vector_print_tile(%tile: vector<[4]x[4]xf32>)
 // CHECK-NEXT:      scf.for %[[TILE_SLICE_INDEX:.*]] = %[[C0]] to %[[NUM_TILE_SLICES]] step %[[C1]] {
 // CHECK-NEXT:        %[[TILE_SLICE:.*]] = arm_sme.move_tile_slice_to_vector %[[TILE]][%[[TILE_SLICE_INDEX]]] : vector<[4]xf32> from vector<[4]x[4]xf32>
 // CHECK-NEXT:        vector.print %[[TILE_SLICE]] : vector<[4]xf32>
+
+//===----------------------------------------------------------------------===//
+// vector.load
+//===----------------------------------------------------------------------===//
+
+// -----
+
+// CHECK-LABEL: @vector_load_i8_with_offset(
+// CHECK-SAME:                              %[[MEMREF:.*]]: memref<?x?xi8>)
+// CHECK: %[[C0:.*]] = arith.constant 0 : index
+// CHECK: %[[C123:.*]] = arith.constant 123 : index
+// CHECK: arm_sme.tile_load %[[MEMREF]][%[[C123]], %[[C0]]] : memref<?x?xi8>, vector<[16]x[16]xi8>
+func.func @vector_load_i8_with_offset(%arg0 : memref<?x?xi8>) -> vector<[16]x[16]xi8> {
+  %c0 = arith.constant 0 : index
+  %c123 = arith.constant 123 : index
+  %tile = vector.load %arg0[%c123, %c0] : memref<?x?xi8>, vector<[16]x[16]xi8>
+  return %tile : vector<[16]x[16]xi8>
+}
+
+// -----
+
+// CHECK-LABEL: @vector_load_i8_from_rank_1_memref(
+// CHECK-SAME:                                     %[[MEMREF:.*]]: memref<?xi8>)
+// CHECK: %[[C0:.*]] = arith.constant 0 : index
+// CHECK: arm_sme.tile_load %[[MEMREF]][%[[C0]]] : memref<?xi8>, vector<[16]x[16]xi8>
+func.func @vector_load_i8_from_rank_1_memref(%arg0 : memref<?xi8>) -> vector<[16]x[16]xi8> {
+  %c0 = arith.constant 0 : index
+  %tile = vector.load %arg0[%c0] : memref<?xi8>, vector<[16]x[16]xi8>
+  return %tile : vector<[16]x[16]xi8>
+}
+
+// -----
+
+// CHECK-LABEL: @vector_load_i16(
+// CHECK: arm_sme.tile_load {{.*}} : memref<?x?xi16>, vector<[8]x[8]xi16>
+func.func @vector_load_i16(%arg0 : memref<?x?xi16>) -> vector<[8]x[8]xi16> {
+  %c0 = arith.constant 0 : index
+  %tile = vector.load %arg0[%c0, %c0] : memref<?x?xi16>, vector<[8]x[8]xi16>
+  return %tile : vector<[8]x[8]xi16>
+}
+
+// -----
+
+// CHECK-LABEL: @vector_load_i32(
+// CHECK: arm_sme.tile_load {{.*}} : memref<?x?xi32>, vector<[4]x[4]xi32>
+func.func @vector_load_i32(%arg0 : memref<?x?xi32>) -> vector<[4]x[4]xi32> {
+  %c0 = arith.constant 0 : index
+  %tile = vector.load %arg0[%c0, %c0] : memref<?x?xi32>, vector<[4]x[4]xi32>
+  return %tile : vector<[4]x[4]xi32>
+}
+
+// -----
+
+// CHECK-LABEL: @vector_load_i64(
+// CHECK: arm_sme.tile_load {{.*}} : memref<?x?xi64>, vector<[2]x[2]xi64>
+func.func @vector_load_i64(%arg0 : memref<?x?xi64>) -> vector<[2]x[2]xi64> {
+  %c0 = arith.constant 0 : index
+  %tile = vector.load %arg0[%c0, %c0] : memref<?x?xi64>, vector<[2]x[2]xi64>
+  return %tile : vector<[2]x[2]xi64>
+}
+
+// -----
+
+// CHECK-LABEL: @vector_load_f16(
+// CHECK: arm_sme.tile_load {{.*}} : memref<?x?xf16>, vector<[8]x[8]xf16>
+func.func @vector_load_f16(%arg0 : memref<?x?xf16>) -> vector<[8]x[8]xf16> {
+  %c0 = arith.constant 0 : index
+  %tile = vector.load %arg0[%c0, %c0] : memref<?x?xf16>, vector<[8]x[8]xf16>
+  return %tile : vector<[8]x[8]xf16>
+}
+
+// -----
+
+// CHECK-LABEL: @vector_load_bf16(
+// CHECK: arm_sme.tile_load {{.*}} : memref<?x?xbf16>, vector<[8]x[8]xbf16>
+func.func @vector_load_bf16(%arg0 : memref<?x?xbf16>) -> vector<[8]x[8]xbf16> {
+  %c0 = arith.constant 0 : index
+  %tile = vector.load %arg0[%c0, %c0] : memref<?x?xbf16>, vector<[8]x[8]xbf16>
+  return %tile : vector<[8]x[8]xbf16>
+}
+
+// -----
+
+// CHECK-LABEL: @vector_load_f32(
+// CHECK: arm_sme.tile_load {{.*}} : memref<?x?xf32>, vector<[4]x[4]xf32>
+func.func @vector_load_f32(%arg0 : memref<?x?xf32>) -> vector<[4]x[4]xf32> {
+  %c0 = arith.constant 0 : index
+  %tile = vector.load %arg0[%c0, %c0] : memref<?x?xf32>, vector<[4]x[4]xf32>
+  return %tile : vector<[4]x[4]xf32>
+}
+
+// -----
+
+// CHECK-LABEL: @vector_load_f64(
+// CHECK: arm_sme.tile_load {{.*}} : memref<?x?xf64>, vector<[2]x[2]xf64>
+func.func @vector_load_f64(%arg0 : memref<?x?xf64>) -> vector<[2]x[2]xf64> {
+  %c0 = arith.constant 0 : index
+  %tile = vector.load %arg0[%c0, %c0] : memref<?x?xf64>, vector<[2]x[2]xf64>
+  return %tile : vector<[2]x[2]xf64>
+}
+
+// -----
+
+// CHECK-LABEL: @vector_load_i128(
+// CHECK: arm_sme.tile_load {{.*}} : memref<?x?xi128>, vector<[1]x[1]xi128>
+func.func @vector_load_i128(%arg0 : memref<?x?xi128>) -> vector<[1]x[1]xi128> {
+  %c0 = arith.constant 0 : index
+  %tile = vector.load %arg0[%c0, %c0] : memref<?x?xi128>, vector<[1]x[1]xi128>
+  return %tile : vector<[1]x[1]xi128>
+}
+
+
+//===----------------------------------------------------------------------===//
+// vector.store
+//===----------------------------------------------------------------------===//
+
+// -----
+
+// CHECK-LABEL: @vector_store_i8(
+// CHECK-SAME:                   %[[MEMREF:.*]]: memref<?x?xi8>) {
+// CHECK: %[[C0:.*]] = arith.constant 0 : index
+// CHECK: %[[TILE:.*]] = arm_sme.get_tile : vector<[16]x[16]xi8>
+// CHECK: arm_sme.tile_store %[[TILE]], %[[MEMREF]][%[[C0]], %[[C0]]] : memref<?x?xi8>, vector<[16]x[16]xi8>
+func.func @vector_store_i8(%arg0 : memref<?x?xi8>) {
+  %c0 = arith.constant 0 : index
+  %tile = arm_sme.get_tile : vector<[16]x[16]xi8>
+  vector.store %tile, %arg0[%c0, %c0] : memref<?x?xi8>, vector<[16]x[16]xi8>
+  return
+}
+
+// -----
+
+// CHECK-LABEL: @vector_store_i16
+// CHECK: arm_sme.tile_store {{.*}} : memref<?x?xi16>, vector<[8]x[8]xi16>
+func.func @vector_store_i16(%arg0 : memref<?x?xi16>) {
+  %c0 = arith.constant 0 : index
+  %tile = arm_sme.get_tile : vector<[8]x[8]xi16>
+  vector.store %tile, %arg0[%c0, %c0] : memref<?x?xi16>, vector<[8]x[8]xi16>
+  return
+}
+
+// -----
+
+// CHECK-LABEL: @vector_store_i32
+// CHECK: arm_sme.tile_store {{.*}} : memref<?x?xi32>, vector<[4]x[4]xi32>
+func.func @vector_store_i32(%arg0 : memref<?x?xi32>) {
+  %c0 = arith.constant 0 : index
+  %tile = arm_sme.get_tile : vector<[4]x[4]xi32>
+  vector.store %tile, %arg0[%c0, %c0] : memref<?x?xi32>, vector<[4]x[4]xi32>
+  return
+}
+
+// -----
+
+// CHECK-LABEL: @vector_store_i64
+// CHECK: arm_sme.tile_store {{.*}} : memref<?x?xi64>, vector<[2]x[2]xi64>
+func.func @vector_store_i64(%arg0 : memref<?x?xi64>) {
+  %c0 = arith.constant 0 : index
+  %tile = arm_sme.get_tile : vector<[2]x[2]xi64>
+  vector.store %tile, %arg0[%c0, %c0] : memref<?x?xi64>, vector<[2]x[2]xi64>
+  return
+}
+
+// -----
+
+// CHECK-LABEL: @vector_store_f16
+// CHECK: arm_sme.tile_store {{.*}} : memref<?x?xf16>, vector<[8]x[8]xf16>
+func.func @vector_store_f16(%arg0 : memref<?x?xf16>) {
+  %c0 = arith.constant 0 : index
+  %tile = arm_sme.get_tile : vector<[8]x[8]xf16>
+  vector.store %tile, %arg0[%c0, %c0] : memref<?x?xf16>, vector<[8]x[8]xf16>
+  return
+}
+
+// -----
+
+// CHECK-LABEL: @vector_store_bf16
+// CHECK: arm_sme.tile_store {{.*}} : memref<?x?xbf16>, vector<[8]x[8]xbf16>
+func.func @vector_store_bf16(%arg0 : memref<?x?xbf16>) {
+  %c0 = arith.constant 0 : index
+  %tile = arm_sme.get_tile : vector<[8]x[8]xbf16>
+  vector.store %tile, %arg0[%c0, %c0] : memref<?x?xbf16>, vector<[8]x[8]xbf16>
+  return
+}
+// -----
+
+// CHECK-LABEL: @vector_store_f32
+// CHECK: arm_sme.tile_store {{.*}} : memref<?x?xf32>, vector<[4]x[4]xf32>
+func.func @vector_store_f32(%arg0 : memref<?x?xf32>) {
+  %c0 = arith.constant 0 : index
+  %tile = arm_sme.get_tile : vector<[4]x[4]xf32>
+  vector.store %tile, %arg0[%c0, %c0] : memref<?x?xf32>, vector<[4]x[4]xf32>
+  return
+}
+
+// -----
+
+// CHECK-LABEL: @vector_store_f64
+// CHECK: arm_sme.tile_store {{.*}} : memref<?x?xf64>, vector<[2]x[2]xf64>
+func.func @vector_store_f64(%arg0 : memref<?x?xf64>) {
+  %c0 = arith.constant 0 : index
+  %tile = arm_sme.get_tile : vector<[2]x[2]xf64>
+  vector.store %tile, %arg0[%c0, %c0] : memref<?x?xf64>, vector<[2]x[2]xf64>
+  return
+}
+
+// -----
+
+// CHECK-LABEL: @vector_store_i128
+// CHECK: arm_sme.tile_store {{.*}} : memref<?x?xi128>, vector<[1]x[1]xi128>
+func.func @vector_store_i128(%arg0 : memref<?x?xi128>) {
+  %c0 = arith.constant 0 : index
+  %tile = arm_sme.get_tile : vector<[1]x[1]xi128>
+  vector.store %tile, %arg0[%c0, %c0] : memref<?x?xi128>, vector<[1]x[1]xi128>
+  return
+}
+
+//===----------------------------------------------------------------------===//
+// vector.insert
+//===----------------------------------------------------------------------===//
+
+// -----
+
+// CHECK-LABEL: @vector_insert_slice_i32(
+// CHECK-SAME:                       %[[SLICE:.*]]: vector<[4]xi32>,
+// CHECK-SAME:                       %[[INDEX:.*]]: index)
+func.func @vector_insert_slice_i32(%slice: vector<[4]xi32>, %row: index) -> vector<[4]x[4]xi32>{
+  // CHECK-NEXT: %[[TILE:.*]] = arm_sme.get_tile : vector<[4]x[4]xi32>
+  // CHECK-NEXT: arm_sme.move_vector_to_tile_slice %[[SLICE]], %[[TILE]], %[[INDEX]] : vector<[4]xi32> into vector<[4]x[4]xi32>
+  %tile = arm_sme.get_tile : vector<[4]x[4]xi32>
+  %new_tile = vector.insert %slice, %tile[%row] : vector<[4]xi32> into vector<[4]x[4]xi32>
+  return %new_tile : vector<[4]x[4]xi32>
+}
+
+// -----
+
+// CHECK-LABEL: @vector_insert_slice_i8
+func.func @vector_insert_slice_i8(%slice: vector<[16]xi8>, %row: index) -> vector<[16]x[16]xi8> {
+  // CHECK: arm_sme.move_vector_to_tile_slice %{{.*}} : vector<[16]xi8> into vector<[16]x[16]xi8>
+  %tile = arm_sme.get_tile : vector<[16]x[16]xi8>
+  %new_tile = vector.insert %slice, %tile[%row] : vector<[16]xi8> into vector<[16]x[16]xi8>
+  return %new_tile : vector<[16]x[16]xi8>
+}
+
+// -----
+
+// CHECK-LABEL: @vector_insert_slice_i16
+func.func @vector_insert_slice_i16(%slice: vector<[8]xi16>, %row: index) -> vector<[8]x[8]xi16> {
+  // CHECK: arm_sme.move_vector_to_tile_slice %{{.*}} : vector<[8]xi16> into vector<[8]x[8]xi16>
+  %tile = arm_sme.get_tile : vector<[8]x[8]xi16>
+  %new_tile = vector.insert %slice, %tile[%row] : vector<[8]xi16> into vector<[8]x[8]xi16>
+  return %new_tile : vector<[8]x[8]xi16>
+}
+
+// -----
+
+// CHECK-LABEL: @vector_insert_slice_i64
+func.func @vector_insert_slice_i64(%slice: vector<[2]xi64>, %row: index) -> vector<[2]x[2]xi64> {
+  // CHECK: arm_sme.move_vector_to_tile_slice %{{.*}} : vector<[2]xi64> into vector<[2]x[2]xi64>
+  %tile = arm_sme.get_tile : vector<[2]x[2]xi64>
+  %new_tile = vector.insert %slice, %tile[%row] : vector<[2]xi64> into vector<[2]x[2]xi64>
+  return %new_tile : vector<[2]x[2]xi64>
+}
+
+// -----
+
+// CHECK-LABEL: @vector_insert_slice_i128
+func.func @vector_insert_slice_i128(%slice: vector<[1]xi128>, %row: index) -> vector<[1]x[1]xi128> {
+  // CHECK: arm_sme.move_vector_to_tile_slice %{{.*}} : vector<[1]xi128> into vector<[1]x[1]xi128>
+  %tile = arm_sme.get_tile : vector<[1]x[1]xi128>
+  %new_tile = vector.insert %slice, %tile[%row] : vector<[1]xi128> into vector<[1]x[1]xi128>
+  return %new_tile : vector<[1]x[1]xi128>
+}
+
+// -----
+
+// CHECK-LABEL: @vector_insert_slice_f16
+func.func @vector_insert_slice_f16(%slice: vector<[8]xf16>, %row: index) -> vector<[8]x[8]xf16> {
+  // CHECK: arm_sme.move_vector_to_tile_slice %{{.*}} : vector<[8]xf16> into vector<[8]x[8]xf16>
+  %tile = arm_sme.get_tile : vector<[8]x[8]xf16>
+  %new_tile = vector.insert %slice, %tile[%row] : vector<[8]xf16> into vector<[8]x[8]xf16>
+  return %new_tile : vector<[8]x[8]xf16>
+}
+
+// -----
+
+// CHECK-LABEL: @vector_insert_slice_bf16
+func.func @vector_insert_slice_bf16(%slice: vector<[8]xbf16>, %row: index) -> vector<[8]x[8]xbf16> {
+  // CHECK: arm_sme.move_vector_to_tile_slice %{{.*}} : vector<[8]xbf16> into vector<[8]x[8]xbf16>
+  %tile = arm_sme.get_tile : vector<[8]x[8]xbf16>
+  %new_tile = vector.insert %slice, %tile[%row] : vector<[8]xbf16> into vector<[8]x[8]xbf16>
+  return %new_tile : vector<[8]x[8]xbf16>
+}
+
+// -----
+
+// CHECK-LABEL: @vector_insert_slice_f32
+func.func @vector_insert_slice_f32(%slice: vector<[4]xf32>, %row: index) -> vector<[4]x[4]xf32> {
+  // CHECK: arm_sme.move_vector_to_tile_slice %{{.*}} : vector<[4]xf32> into vector<[4]x[4]xf32>
+  %tile = arm_sme.get_tile : vector<[4]x[4]xf32>
+  %new_tile = vector.insert %slice, %tile[%row] : vector<[4]xf32> into vector<[4]x[4]xf32>
+  return %new_tile : vector<[4]x[4]xf32>
+}
+
+// -----
+
+// CHECK-LABEL: @vector_insert_slice_f64
+func.func @vector_insert_slice_f64(%slice: vector<[2]xf64>, %row: index) -> vector<[2]x[2]xf64> {
+  // CHECK: arm_sme.move_vector_to_tile_slice %{{.*}} : vector<[2]xf64> into vector<[2]x[2]xf64>
+  %tile = arm_sme.get_tile : vector<[2]x[2]xf64>
+  %new_tile = vector.insert %slice, %tile[%row] : vector<[2]xf64> into vector<[2]x[2]xf64>
+  return %new_tile : vector<[2]x[2]xf64>
+}
+
+// -----
+
+// CHECK-LABEL: @vector_insert_element_i32(
+// CHECK-SAME:                         %[[EL:.*]]: i32,
+// CHECK-SAME:                         %[[ROW:.*]]: index,
+// CHECK-SAME:                         %[[COL:.*]]: index)
+func.func @vector_insert_element_i32(%el: i32, %row: index, %col: index) -> vector<[4]x[4]xi32> {
+  // CHECK-NEXT: %[[TILE:.*]] = arm_sme.get_tile : vector<[4]x[4]xi32>
+  // CHECK-NEXT: %[[SLICE:.*]] = arm_sme.move_tile_slice_to_vector %[[TILE]][%[[ROW]]] : vector<[4]xi32> from vector<[4]x[4]xi32>
+  // CHECK-NEXT: %[[NEW_SLICE:.*]] = vector.insert %[[EL]], %[[SLICE]] [%[[COL]]] : i32 into vector<[4]xi32>
+  // CHECK-NEXT: arm_sme.move_vector_to_tile_slice %[[NEW_SLICE]], %[[TILE]], %[[ROW]] : vector<[4]xi32> into vector<[4]x[4]xi32>
+  %tile = arm_sme.get_tile : vector<[4]x[4]xi32>
+  %new_tile = vector.insert %el, %tile[%row, %col] : i32 into vector<[4]x[4]xi32>
+  return %new_tile : vector<[4]x[4]xi32>
+}
+
+// -----
+
+// CHECK-LABEL: @vector_insert_element_i8
+func.func @vector_insert_element_i8(%el: i8, %row: index, %col: index) -> vector<[16]x[16]xi8> {
+  // CHECK: %[[TILE:.*]] = arm_sme.get_tile : vector<[16]x[16]xi8>
+  // CHECK: arm_sme.move_tile_slice_to_vector %[[TILE]]{{.*}} : vector<[16]xi8> from vector<[16]x[16]xi8>
+  // CHECK: arm_sme.move_vector_to_tile_slice %{{.*}}, %[[TILE]], %{{.*}} : vector<[16]xi8> into vector<[16]x[16]xi8>
+  %tile = arm_sme.get_tile : vector<[16]x[16]xi8>
+  %new_tile = vector.insert %el, %tile[%row, %col] : i8 into vector<[16]x[16]xi8>
+  return %new_tile : vector<[16]x[16]xi8>
+}
+
+// -----
+
+// CHECK-LABEL: @vector_insert_element_i16
+func.func @vector_insert_element_i16(%el: i16, %row: index, %col: index) -> vector<[8]x[8]xi16> {
+  // CHECK: %[[TILE:.*]] = arm_sme.get_tile : vector<[8]x[8]xi16>
+  // CHECK: arm_sme.move_tile_slice_to_vector %[[TILE]]{{.*}} : vector<[8]xi16> from vector<[8]x[8]xi16>
+  // CHECK: arm_sme.move_vector_to_tile_slice %{{.*}}, %[[TILE]], %{{.*}} : vector<[8]xi16> into vector<[8]x[8]xi16>
+  %tile = arm_sme.get_tile : vector<[8]x[8]xi16>
+  %new_tile = vector.insert %el, %tile[%row, %col] : i16 into vector<[8]x[8]xi16>
+  return %new_tile : vector<[8]x[8]xi16>
+}
+
+// -----
+
+// CHECK-LABEL: @vector_insert_element_i64
+func.func @vector_insert_element_i64(%el: i64, %row: index, %col: index) -> vector<[2]x[2]xi64> {
+  // CHECK: %[[TILE:.*]] = arm_sme.get_tile : vector<[2]x[2]xi64>
+  // CHECK: arm_sme.move_tile_slice_to_vector %[[TILE]]{{.*}} : vector<[2]xi64> from vector<[2]x[2]xi64>
+  // CHECK: arm_sme.move_vector_to_tile_slice %{{.*}}, %[[TILE]], %{{.*}} : vector<[2]xi64> into vector<[2]x[2]xi64>
+  %tile = arm_sme.get_tile : vector<[2]x[2]xi64>
+  %new_tile = vector.insert %el, %tile[%row, %col] : i64 into vector<[2]x[2]xi64>
+  return %new_tile : vector<[2]x[2]xi64>
+}
+
+// -----
+
+// CHECK-LABEL: @vector_insert_element_i128
+func.func @vector_insert_element_i128(%el: i128, %row: index, %col: index) -> vector<[1]x[1]xi128> {
+  // CHECK: %[[TILE:.*]] = arm_sme.get_tile : vector<[1]x[1]xi128>
+  // CHECK: arm_sme.move_tile_slice_to_vector %[[TILE]]{{.*}} : vector<[1]xi128> from vector<[1]x[1]xi128>
+  // CHECK: arm_sme.move_vector_to_tile_slice %{{.*}}, %[[TILE]], %{{.*}} : vector<[1]xi128> into vector<[1]x[1]xi128>
+  %tile = arm_sme.get_tile : vector<[1]x[1]xi128>
+  %new_tile = vector.insert %el, %tile[%row, %col] : i128 into vector<[1]x[1]xi128>
+  return %new_tile : vector<[1]x[1]xi128>
+}
+
+// -----
+
+// CHECK-LABEL: @vector_insert_element_f16
+func.func @vector_insert_element_f16(%el: f16, %row: index, %col: index) -> vector<[8]x[8]xf16> {
+  // CHECK: %[[TILE:.*]] = arm_sme.get_tile : vector<[8]x[8]xf16>
+  // CHECK: arm_sme.move_tile_slice_to_vector %[[TILE]]{{.*}} : vector<[8]xf16> from vector<[8]x[8]xf16>
+  // CHECK: arm_sme.move_vector_to_tile_slice %{{.*}}, %[[TILE]], %{{.*}} : vector<[8]xf16> into vector<[8]x[8]xf16>
+  %tile = arm_sme.get_tile : vector<[8]x[8]xf16>
+  %new_tile = vector.insert %el, %tile[%row, %col] : f16 into vector<[8]x[8]xf16>
+  return %new_tile : vector<[8]x[8]xf16>
+}
+
+// -----
+
+// CHECK-LABEL: @vector_insert_element_bf16
+func.func @vector_insert_element_bf16(%el: bf16, %row: index, %col: index) -> vector<[8]x[8]xbf16> {
+  // CHECK: %[[TILE:.*]] = arm_sme.get_tile : vector<[8]x[8]xbf16>
+  // CHECK: arm_sme.move_tile_slice_to_vector %[[TILE]]{{.*}} : vector<[8]xbf16> from vector<[8]x[8]xbf16>
+  // CHECK: arm_sme.move_vector_to_tile_slice %{{.*}}, %[[TILE]], %{{.*}} : vector<[8]xbf16> into vector<[8]x[8]xbf16>
+  %tile = arm_sme.get_tile : vector<[8]x[8]xbf16>
+  %new_tile = vector.insert %el, %tile[%row, %col] : bf16 into vector<[8]x[8]xbf16>
+  return %new_tile : vector<[8]x[8]xbf16>
+}
+
+// -----
+
+// CHECK-LABEL: @vector_insert_element_f32
+func.func @vector_insert_element_f32(%el: f32, %row: index, %col: index) -> vector<[4]x[4]xf32> {
+  // CHECK: %[[TILE:.*]] = arm_sme.get_tile : vector<[4]x[4]xf32>
+  // CHECK: arm_sme.move_tile_slice_to_vector %[[TILE]]{{.*}} : vector<[4]xf32> from vector<[4]x[4]xf32>
+  // CHECK: arm_sme.move_vector_to_tile_slice %{{.*}}, %[[TILE]], %{{.*}} : vector<[4]xf32> into vector<[4]x[4]xf32>
+  %tile = arm_sme.get_tile : vector<[4]x[4]xf32>
+  %new_tile = vector.insert %el, %tile[%row, %col] : f32 into vector<[4]x[4]xf32>
+  return %new_tile : vector<[4]x[4]xf32>
+}
+
+// -----
+
+// CHECK-LABEL: @vector_insert_element_f64
+func.func @vector_insert_element_f64(%el: f64, %row: index, %col: index) -> vector<[2]x[2]xf64> {
+  // CHECK: %[[TILE:.*]] = arm_sme.get_tile : vector<[2]x[2]xf64>
+  // CHECK: arm_sme.move_tile_slice_to_vector %[[TILE]]{{.*}} : vector<[2]xf64> from vector<[2]x[2]xf64>
+  // CHECK: arm_sme.move_vector_to_tile_slice %{{.*}}, %[[TILE]], %{{.*}} : vector<[2]xf64> into vector<[2]x[2]xf64>
+  %tile = arm_sme.get_tile : vector<[2]x[2]xf64>
+  %new_tile = vector.insert %el, %tile[%row, %col] : f64 into vector<[2]x[2]xf64>
+  return %new_tile : vector<[2]x[2]xf64>
+}
+
+//===----------------------------------------------------------------------===//
+// vector.extract
+//===----------------------------------------------------------------------===//
+
+// -----
+
+// CHECK-LABEL: @vector_extract_slice_i32(
+// CHECK-SAME:                            %[[INDEX:.*]]: index)
+func.func @vector_extract_slice_i32(%row: index) -> vector<[4]xi32> {
+  // CHECK: %[[TILE:.*]] = arm_sme.get_tile : vector<[4]x[4]xi32>
+  // CHECK: arm_sme.move_tile_slice_to_vector %[[TILE]][%[[INDEX]]] : vector<[4]xi32> from vector<[4]x[4]xi32>
+  %tile = arm_sme.get_tile : vector<[4]x[4]xi32>
+  %slice = vector.extract %tile[%row] : vector<[4]xi32> from vector<[4]x[4]xi32>
+  return %slice : vector<[4]xi32>
+}
+
+// -----
+
+// CHECK-LABEL: @vector_extract_slice_i8
+func.func @vector_extract_slice_i8(%row: index) -> vector<[16]xi8> {
+  // CHECK: arm_sme.move_tile_slice_to_vector {{.*}} : vector<[16]xi8> from vector<[16]x[16]xi8>
+  %tile = arm_sme.get_tile : vector<[16]x[16]xi8>
+  %slice = vector.extract %tile[%row] : vector<[16]xi8> from vector<[16]x[16]xi8>
+  return %slice : vector<[16]xi8>
+}
+
+// -----
+
+// CHECK-LABEL: @vector_extract_slice_i16
+func.func @vector_extract_slice_i16(%row: index) -> vector<[8]xi16> {
+  // CHECK: arm_sme.move_tile_slice_to_vector {{.*}} : vector<[8]xi16> from vector<[8]x[8]xi16>
+  %tile = arm_sme.get_tile : vector<[8]x[8]xi16>
+  %slice = vector.extract %tile[%row] : vector<[8]xi16> from vector<[8]x[8]xi16>
+  return %slice : vector<[8]xi16>
+}
+
+// -----
+
+// CHECK-LABEL: @vector_extract_slice_i64
+func.func @vector_extract_slice_i64(%row: index) -> vector<[2]xi64> {
+  // CHECK: arm_sme.move_tile_slice_to_vector {{.*}} : vector<[2]xi64> from vector<[2]x[2]xi64>
+  %tile = arm_sme.get_tile : vector<[2]x[2]xi64>
+  %slice = vector.extract %tile[%row] : vector<[2]xi64> from vector<[2]x[2]xi64>
+  return %slice : vector<[2]xi64>
+}
+
+// -----
+
+// CHECK-LABEL: @vector_extract_slice_i128
+func.func @vector_extract_slice_i128(%row: index) -> vector<[1]xi128> {
+  // CHECK: arm_sme.move_tile_slice_to_vector {{.*}} : vector<[1]xi128> from vector<[1]x[1]xi128>
+  %tile = arm_sme.get_tile : vector<[1]x[1]xi128>
+  %slice = vector.extract %tile[%row] : vector<[1]xi128> from vector<[1]x[1]xi128>
+  return %slice : vector<[1]xi128>
+}
+
+// -----
+
+// CHECK-LABEL: @vector_extract_slice_f16
+func.func @vector_extract_slice_f16(%row: index) -> vector<[8]xf16> {
+  // CHECK: arm_sme.move_tile_slice_to_vector {{.*}} : vector<[8]xf16> from vector<[8]x[8]xf16>
+  %tile = arm_sme.get_tile : vector<[8]x[8]xf16>
+  %slice = vector.extract %tile[%row] : vector<[8]xf16> from vector<[8]x[8]xf16>
+  return %slice : vector<[8]xf16>
+}
+
+// -----
+
+// CHECK-LABEL: @vector_extract_slice_bf16
+func.func @vector_extract_slice_bf16(%row: index) -> vector<[8]xbf16> {
+  // CHECK: arm_sme.move_tile_slice_to_vector {{.*}} : vector<[8]xbf16> from vector<[8]x[8]xbf16>
+  %tile = arm_sme.get_tile : vector<[8]x[8]xbf16>
+  %slice = vector.extract %tile[%row] : vector<[8]xbf16> from vector<[8]x[8]xbf16>
+  return %slice : vector<[8]xbf16>
+}
+
+// -----
+
+// CHECK-LABEL: @vector_extract_slice_f32
+func.func @vector_extract_slice_f32(%row: index) -> vector<[4]xf32> {
+  // CHECK: arm_sme.move_tile_slice_to_vector {{.*}} : vector<[4]xf32> from vector<[4]x[4]xf32>
+  %tile = arm_sme.get_tile : vector<[4]x[4]xf32>
+  %slice = vector.extract %tile[%row] : vector<[4]xf32> from vector<[4]x[4]xf32>
+  return %slice : vector<[4]xf32>
+}
+
+// -----
+
+// CHECK-LABEL: @vector_extract_slice_f64
+func.func @vector_extract_slice_f64(%row: index) -> vector<[2]xf64> {
+  // CHECK: arm_sme.move_tile_slice_to_vector {{.*}} : vector<[2]xf64> from vector<[2]x[2]xf64>
+  %tile = arm_sme.get_tile : vector<[2]x[2]xf64>
+  %slice = vector.extract %tile[%row] : vector<[2]xf64> from vector<[2]x[2]xf64>
+  return %slice : vector<[2]xf64>
+}
+
+// -----
+
+// CHECK-LABEL: @vector_extract_element(
+// CHECK-SAME:                          %[[ROW:.*]]: index,
+// CHECK-SAME:                          %[[COL:.*]]: index)
+func.func @vector_extract_element(%row: index, %col: index) -> i32 {
+  // CHECK-NEXT: %[[TILE:.*]] = arm_sme.get_tile : vector<[4]x[4]xi32>
+  // CHECK-NEXT: %[[SLICE:.*]] = arm_sme.move_tile_slice_to_vector %[[TILE]][%[[ROW]]] : vector<[4]xi32> from vector<[4]x[4]xi32>
+  // CHECK-NEXT: %[[EL:.*]] = vector.extract %[[SLICE]]{{\[}}%[[COL]]] : i32 from vector<[4]xi32>
+  %tile = arm_sme.get_tile : vector<[4]x[4]xi32>
+  %el = vector.extract %tile[%row, %col] : i32 from vector<[4]x[4]xi32>
+  return %el : i32
+}
+
+// -----
+
+// CHECK-LABEL: @vector_extract_element_i8
+func.func @vector_extract_element_i8(%row: index, %col: index) -> i8 {
+  // CHECK: %[[SLICE:.*]] = arm_sme.move_tile_slice_to_vector %{{.*}} : vector<[16]xi8> from vector<[16]x[16]xi8>
+  // CHECK-NEXT: %{{.*}} = vector.extract %[[SLICE]]{{\[}}%{{.*}}] : i8 from vector<[16]xi8>
+  %tile = arm_sme.get_tile : vector<[16]x[16]xi8>
+  %el = vector.extract %tile[%row, %col] : i8 from vector<[16]x[16]xi8>
+  return %el : i8
+}
+
+// -----
+
+// CHECK-LABEL: @vector_extract_element_i16
+func.func @vector_extract_element_i16(%row: index, %col: index) -> i16 {
+  // CHECK: %[[SLICE:.*]] = arm_sme.move_tile_slice_to_vector %{{.*}} : vector<[8]xi16> from vector<[8]x[8]xi16>
+  // CHECK-NEXT: %{{.*}} = vector.extract %[[SLICE]]{{\[}}%{{.*}}] : i16 from vector<[8]xi16>
+  %tile = arm_sme.get_tile : vector<[8]x[8]xi16>
+  %el = vector.extract %tile[%row, %col] : i16 from vector<[8]x[8]xi16>
+  return %el : i16
+}
+
+// -----
+
+// CHECK-LABEL: @vector_extract_element_i64
+func.func @vector_extract_element_i64(%row: index, %col: index) -> i64 {
+  // CHECK: %[[SLICE:.*]] = arm_sme.move_tile_slice_to_vector %{{.*}} : vector<[2]xi64> from vector<[2]x[2]xi64>
+  // CHECK-NEXT: %{{.*}} = vector.extract %[[SLICE]]{{\[}}%{{.*}}] : i64 from vector<[2]xi64>
+  %tile = arm_sme.get_tile : vector<[2]x[2]xi64>
+  %el = vector.extract %tile[%row, %col] : i64 from vector<[2]x[2]xi64>
+  return %el : i64
+}
+
+// -----
+
+// CHECK-LABEL: @vector_extract_element_i128
+func.func @vector_extract_element_i128(%row: index, %col: index) -> i128 {
+  // CHECK: %[[SLICE:.*]] = arm_sme.move_tile_slice_to_vector %{{.*}} : vector<[1]xi128> from vector<[1]x[1]xi128>
+  // CHECK-NEXT: %{{.*}} = vector.extract %[[SLICE]]{{\[}}%{{.*}}] : i128 from vector<[1]xi128>
+  %tile = arm_sme.get_tile : vector<[1]x[1]xi128>
+  %el = vector.extract %tile[%row, %col] : i128 from vector<[1]x[1]xi128>
+  return %el : i128
+}
+
+// -----
+
+// CHECK-LABEL: @vector_extract_element_f16
+func.func @vector_extract_element_f16(%row: index, %col: index) -> f16 {
+  // CHECK: %[[SLICE:.*]] = arm_sme.move_tile_slice_to_vector %{{.*}} : vector<[8]xf16> from vector<[8]x[8]xf16>
+  // CHECK-NEXT: %{{.*}} = vector.extract %[[SLICE]]{{\[}}%{{.*}}] : f16 from vector<[8]xf16>
+  %tile = arm_sme.get_tile : vector<[8]x[8]xf16>
+  %el = vector.extract %tile[%row, %col] : f16 from vector<[8]x[8]xf16>
+  return %el : f16
+}
+
+// -----
+
+// CHECK-LABEL: @vector_extract_element_bf16
+func.func @vector_extract_element_bf16(%row: index, %col: index) -> bf16 {
+  // CHECK: %[[SLICE:.*]] = arm_sme.move_tile_slice_to_vector %{{.*}} : vector<[8]xbf16> from vector<[8]x[8]xbf16>
+  // CHECK-NEXT: %{{.*}} = vector.extract %[[SLICE]]{{\[}}%{{.*}}] : bf16 from vector<[8]xbf16>
+  %tile = arm_sme.get_tile : vector<[8]x[8]xbf16>
+  %el = vector.extract %tile[%row, %col] : bf16 from vector<[8]x[8]xbf16>
+  return %el : bf16
+}
+
+// -----
+
+// CHECK-LABEL: @vector_extract_element_f32
+func.func @vector_extract_element_f32(%row: index, %col: index) -> f32 {
+  // CHECK: %[[SLICE:.*]] = arm_sme.move_tile_slice_to_vector %{{.*}} : vector<[4]xf32> from vector<[4]x[4]xf32>
+  // CHECK-NEXT: %{{.*}} = vector.extract %[[SLICE]]{{\[}}%{{.*}}] : f32 from vector<[4]xf32>
+  %tile = arm_sme.get_tile : vector<[4]x[4]xf32>
+  %el = vector.extract %tile[%row, %col] : f32 from vector<[4]x[4]xf32>
+  return %el : f32
+}
+
+// -----
+
+// CHECK-LABEL: @vector_extract_element_f64
+func.func @vector_extract_element_f64(%row: index, %col: index) -> f64 {
+  // CHECK: %[[SLICE:.*]] = arm_sme.move_tile_slice_to_vector %{{.*}} : vector<[2]xf64> from vector<[2]x[2]xf64>
+  // CHECK-NEXT: %{{.*}} = vector.extract %[[SLICE]]{{\[}}%{{.*}}] : f64 from vector<[2]xf64>
+  %tile = arm_sme.get_tile : vector<[2]x[2]xf64>
+  %el = vector.extract %tile[%row, %col] : f64 from vector<[2]x[2]xf64>
+  return %el : f64
+}
diff --git a/mlir/test/Conversion/VectorToLLVM/vector-to-llvm.mlir b/mlir/test/Conversion/VectorToLLVM/vector-to-llvm.mlir
index 1712d3d745b7..439f1e920e39 100644
--- a/mlir/test/Conversion/VectorToLLVM/vector-to-llvm.mlir
+++ b/mlir/test/Conversion/VectorToLLVM/vector-to-llvm.mlir
@@ -2517,7 +2517,7 @@ func.func @vector_interleave_1d(%a: vector<8xf32>, %b: vector<8xf32>) -> vector<
 //  CHECK-SAME:     %[[LHS:.*]]: vector<[4]xi32>, %[[RHS:.*]]: vector<[4]xi32>)
 func.func @vector_interleave_1d_scalable(%a: vector<[4]xi32>, %b: vector<[4]xi32>) -> vector<[8]xi32>
 {
-  // CHECK: %[[ZIP:.*]] = "llvm.intr.experimental.vector.interleave2"(%[[LHS]], %[[RHS]]) : (vector<[4]xi32>, vector<[4]xi32>) -> vector<[8]xi32>
+  // CHECK: %[[ZIP:.*]] = "llvm.intr.vector.interleave2"(%[[LHS]], %[[RHS]]) : (vector<[4]xi32>, vector<[4]xi32>) -> vector<[8]xi32>
   // CHECK: return %[[ZIP]]
   %0 = vector.interleave %a, %b : vector<[4]xi32>
   return %0 : vector<[8]xi32>
@@ -2541,7 +2541,7 @@ func.func @vector_interleave_2d(%a: vector<2x3xi8>, %b: vector<2x3xi8>) -> vecto
 //  CHECK-SAME:     %[[LHS:.*]]: vector<2x[8]xi16>, %[[RHS:.*]]: vector<2x[8]xi16>)
 func.func @vector_interleave_2d_scalable(%a: vector<2x[8]xi16>, %b: vector<2x[8]xi16>) -> vector<2x[16]xi16>
 {
-  // CHECK: llvm.intr.experimental.vector.interleave2
+  // CHECK: llvm.intr.vector.interleave2
   // CHECK-NOT: vector.interleave {{.*}} : vector<2x[8]xi16>
   %0 = vector.interleave %a, %b : vector<2x[8]xi16>
   return %0 : vector<2x[16]xi16>
diff --git a/mlir/test/Dialect/Arith/canonicalize.mlir b/mlir/test/Dialect/Arith/canonicalize.mlir
index 79a318565e98..f7ce2123a93c 100644
--- a/mlir/test/Dialect/Arith/canonicalize.mlir
+++ b/mlir/test/Dialect/Arith/canonicalize.mlir
@@ -1223,6 +1223,28 @@ func.func @mulsiExtendedOneRhsSplat(%arg0: vector<3xi32>) -> (vector<3xi32>, vec
   return %low, %high : vector<3xi32>, vector<3xi32>
 }
 
+// CHECK-LABEL: @mulsiExtendedOneRhsI1
+//  CHECK-SAME:   (%[[ARG:.+]]: i1) -> (i1, i1)
+//  CHECK-NEXT:   %[[T:.+]]  = arith.constant true
+//  CHECK-NEXT:   %[[LOW:.+]], %[[HIGH:.+]] = arith.mulsi_extended %[[ARG]], %[[T]] : i1
+//  CHECK-NEXT:   return %[[LOW]], %[[HIGH]] : i1, i1
+func.func @mulsiExtendedOneRhsI1(%arg0: i1) -> (i1, i1) {
+  %one = arith.constant true
+  %low, %high = arith.mulsi_extended %arg0, %one: i1
+  return %low, %high : i1, i1
+}
+
+// CHECK-LABEL: @mulsiExtendedOneRhsSplatI1
+//  CHECK-SAME:   (%[[ARG:.+]]: vector<3xi1>) -> (vector<3xi1>, vector<3xi1>)
+//  CHECK-NEXT:   %[[TS:.+]]  = arith.constant dense<true> : vector<3xi1>
+//  CHECK-NEXT:   %[[LOW:.+]], %[[HIGH:.+]] = arith.mulsi_extended %[[ARG]], %[[TS]] : vector<3xi1>
+//  CHECK-NEXT:   return %[[LOW]], %[[HIGH]] : vector<3xi1>, vector<3xi1>
+func.func @mulsiExtendedOneRhsSplatI1(%arg0: vector<3xi1>) -> (vector<3xi1>, vector<3xi1>) {
+  %one = arith.constant dense<true> : vector<3xi1>
+  %low, %high = arith.mulsi_extended %arg0, %one: vector<3xi1>
+  return %low, %high : vector<3xi1>, vector<3xi1>
+}
+
 // CHECK-LABEL: @mulsiExtendedUnusedHigh
 //  CHECK-SAME:   (%[[ARG:.+]]: i32) -> i32
 //  CHECK-NEXT:   %[[RES:.+]] = arith.muli %[[ARG]], %[[ARG]] : i32
@@ -2809,6 +2831,87 @@ func.func @unsignedExtendConstantResource() -> tensor<i16> {
   return %ext : tensor<i16>
 }
 
+// CHECK-LABEL: @extsi_i0
+//       CHECK:   %[[ZERO:.*]] = arith.constant 0 : i16
+//       CHECK:   return %[[ZERO]] : i16
+func.func @extsi_i0() -> i16 {
+  %c0 = arith.constant 0 : i0
+  %extsi = arith.extsi %c0 : i0 to i16
+  return %extsi : i16
+}
+
+// CHECK-LABEL: @extui_i0
+//       CHECK:   %[[ZERO:.*]] = arith.constant 0 : i16
+//       CHECK:   return %[[ZERO]] : i16
+func.func @extui_i0() -> i16 {
+  %c0 = arith.constant 0 : i0
+  %extui = arith.extui %c0 : i0 to i16
+  return %extui : i16
+}
+
+// CHECK-LABEL: @trunc_i0
+//       CHECK:   %[[ZERO:.*]] = arith.constant 0 : i0
+//       CHECK:   return %[[ZERO]] : i0
+func.func @trunc_i0() -> i0 {
+  %cFF = arith.constant 0xFF : i8
+  %trunc = arith.trunci %cFF : i8 to i0
+  return %trunc : i0
+}
+
+// CHECK-LABEL: @shli_i0
+//       CHECK:   %[[ZERO:.*]] = arith.constant 0 : i0
+//       CHECK:   return %[[ZERO]] : i0
+func.func @shli_i0() -> i0 {
+  %c0 = arith.constant 0 : i0
+  %shli = arith.shli %c0, %c0 : i0
+  return %shli : i0
+}
+
+// CHECK-LABEL: @shrsi_i0
+//       CHECK:   %[[ZERO:.*]] = arith.constant 0 : i0
+//       CHECK:   return %[[ZERO]] : i0
+func.func @shrsi_i0() -> i0 {
+  %c0 = arith.constant 0 : i0
+  %shrsi = arith.shrsi %c0, %c0 : i0
+  return %shrsi : i0
+}
+
+// CHECK-LABEL: @shrui_i0
+//       CHECK:   %[[ZERO:.*]] = arith.constant 0 : i0
+//       CHECK:   return %[[ZERO]] : i0
+func.func @shrui_i0() -> i0 {
+  %c0 = arith.constant 0 : i0
+  %shrui = arith.shrui %c0, %c0 : i0
+  return %shrui : i0
+}
+
+// CHECK-LABEL: @maxsi_i0
+//       CHECK:   %[[ZERO:.*]] = arith.constant 0 : i0
+//       CHECK:   return %[[ZERO]] : i0
+func.func @maxsi_i0() -> i0 {
+  %c0 = arith.constant 0 : i0
+  %maxsi = arith.maxsi %c0, %c0 : i0
+  return %maxsi : i0
+}
+
+// CHECK-LABEL: @minsi_i0
+//       CHECK:   %[[ZERO:.*]] = arith.constant 0 : i0
+//       CHECK:   return %[[ZERO]] : i0
+func.func @minsi_i0() -> i0 {
+  %c0 = arith.constant 0 : i0
+  %minsi = arith.minsi %c0, %c0 : i0
+  return %minsi : i0
+}
+
+// CHECK-LABEL: @mulsi_extended_i0
+//       CHECK:   %[[ZERO:.*]] = arith.constant 0 : i0
+//       CHECK:   return %[[ZERO]], %[[ZERO]] : i0
+func.func @mulsi_extended_i0() -> (i0, i0) {
+  %c0 = arith.constant 0 : i0
+  %mulsi_extended:2 = arith.mulsi_extended %c0, %c0 : i0
+  return %mulsi_extended#0, %mulsi_extended#1 : i0, i0
+}
+
 {-#
   dialect_resources: {
     builtin: {
diff --git a/mlir/test/Dialect/ArmSME/outer-product-fusion.mlir b/mlir/test/Dialect/ArmSME/outer-product-fusion.mlir
index de9de86003e6..01f54a4cf186 100644
--- a/mlir/test/Dialect/ArmSME/outer-product-fusion.mlir
+++ b/mlir/test/Dialect/ArmSME/outer-product-fusion.mlir
@@ -4,10 +4,10 @@
 // CHECK-SAME:    %[[A0:.*]]: vector<[4]xf16>, %[[B0:.*]]: vector<[4]xf16>, %[[A1:.*]]: vector<[4]xf16>, %[[B1:.*]]: vector<[4]xf16>,
 // CHECK-SAME:    %[[A0_MASK:.*]]: vector<[4]xi1>, %[[B0_MASK:.*]]: vector<[4]xi1>, %[[A1_MASK:.*]]: vector<[4]xi1>, %[[B1_MASK:.*]]: vector<[4]xi1>
 // CHECK-DAG: %[[ACC:.*]] = arith.constant dense<0.000000e+00> : vector<[4]x[4]xf32>
-// CHECK-DAG: %[[LHS:.*]] = "llvm.intr.experimental.vector.interleave2"(%[[A0]], %[[A1]]) : (vector<[4]xf16>, vector<[4]xf16>) -> vector<[8]xf16>
-// CHECK-DAG: %[[RHS:.*]] = "llvm.intr.experimental.vector.interleave2"(%[[B0]], %[[B1]]) : (vector<[4]xf16>, vector<[4]xf16>) -> vector<[8]xf16>
-// CHECK-DAG: %[[LHS_MASK:.*]] = "llvm.intr.experimental.vector.interleave2"(%[[A0_MASK]], %[[A1_MASK]]) : (vector<[4]xi1>, vector<[4]xi1>) -> vector<[8]xi1>
-// CHECK-DAG: %[[RHS_MASK:.*]] = "llvm.intr.experimental.vector.interleave2"(%[[B0_MASK]], %[[B1_MASK]]) : (vector<[4]xi1>, vector<[4]xi1>) -> vector<[8]xi1>
+// CHECK-DAG: %[[LHS:.*]] = "llvm.intr.vector.interleave2"(%[[A0]], %[[A1]]) : (vector<[4]xf16>, vector<[4]xf16>) -> vector<[8]xf16>
+// CHECK-DAG: %[[RHS:.*]] = "llvm.intr.vector.interleave2"(%[[B0]], %[[B1]]) : (vector<[4]xf16>, vector<[4]xf16>) -> vector<[8]xf16>
+// CHECK-DAG: %[[LHS_MASK:.*]] = "llvm.intr.vector.interleave2"(%[[A0_MASK]], %[[A1_MASK]]) : (vector<[4]xi1>, vector<[4]xi1>) -> vector<[8]xi1>
+// CHECK-DAG: %[[RHS_MASK:.*]] = "llvm.intr.vector.interleave2"(%[[B0_MASK]], %[[B1_MASK]]) : (vector<[4]xi1>, vector<[4]xi1>) -> vector<[8]xi1>
 // CHECK-DAG: arm_sme.fmopa_2way %[[LHS]], %[[RHS]] acc(%[[ACC]]) masks(%[[LHS_MASK]], %[[RHS_MASK]]) : vector<[8]xf16>, vector<[8]xf16> into vector<[4]x[4]xf32>
 func.func @outerproduct_add_widening_2way_f16f16f32(
     %a0 : vector<[4]xf16>, %b0 : vector<[4]xf16>,
@@ -225,18 +225,18 @@ func.func @outerproduct_sub_widening_2way_unsigned_i16i16i32(
 // CHECK-SAME:    %[[A2_MASK:[a-z0-9]+]]: vector<[4]xi1>, %[[B2_MASK:[a-z0-9]+]]: vector<[4]xi1>,
 // CHECK-SAME:    %[[A3_MASK:[a-z0-9]+]]: vector<[4]xi1>, %[[B3_MASK:[a-z0-9]+]]: vector<[4]xi1>
 // CHECK-DAG: %[[ACC:.*]] = arith.constant dense<0> : vector<[4]x[4]xi32>
-// CHECK-DAG: %[[LHS0:.*]] = "llvm.intr.experimental.vector.interleave2"(%[[A0]], %[[A2]]) : (vector<[4]xi8>, vector<[4]xi8>) -> vector<[8]xi8>
-// CHECK-DAG: %[[LHS1:.*]] = "llvm.intr.experimental.vector.interleave2"(%[[A1]], %[[A3]]) : (vector<[4]xi8>, vector<[4]xi8>) -> vector<[8]xi8>
-// CHECK-DAG: %[[RHS0:.*]] = "llvm.intr.experimental.vector.interleave2"(%[[B0]], %[[B2]]) : (vector<[4]xi8>, vector<[4]xi8>) -> vector<[8]xi8>
-// CHECK-DAG: %[[RHS1:.*]] = "llvm.intr.experimental.vector.interleave2"(%[[B1]], %[[B3]]) : (vector<[4]xi8>, vector<[4]xi8>) -> vector<[8]xi8>
-// CHECK-DAG: %[[LHS:.*]] = "llvm.intr.experimental.vector.interleave2"(%[[LHS0]], %[[LHS1]]) : (vector<[8]xi8>, vector<[8]xi8>) -> vector<[16]xi8>
-// CHECK-DAG: %[[RHS:.*]] = "llvm.intr.experimental.vector.interleave2"(%[[RHS0]], %[[RHS1]]) : (vector<[8]xi8>, vector<[8]xi8>) -> vector<[16]xi8>
-// CHECK-DAG: %[[LHS0_MASK:.*]] = "llvm.intr.experimental.vector.interleave2"(%[[A0_MASK]], %[[A2_MASK]]) : (vector<[4]xi1>, vector<[4]xi1>) -> vector<[8]xi1>
-// CHECK-DAG: %[[LHS1_MASK:.*]] = "llvm.intr.experimental.vector.interleave2"(%[[A1_MASK]], %[[A3_MASK]]) : (vector<[4]xi1>, vector<[4]xi1>) -> vector<[8]xi1>
-// CHECK-DAG: %[[RHS0_MASK:.*]] = "llvm.intr.experimental.vector.interleave2"(%[[B0_MASK]], %[[B2_MASK]]) : (vector<[4]xi1>, vector<[4]xi1>) -> vector<[8]xi1>
-// CHECK-DAG: %[[RHS1_MASK:.*]] = "llvm.intr.experimental.vector.interleave2"(%[[B1_MASK]], %[[B3_MASK]]) : (vector<[4]xi1>, vector<[4]xi1>) -> vector<[8]xi1>
-// CHECK-DAG: %[[LHS_MASK:.*]] = "llvm.intr.experimental.vector.interleave2"(%[[LHS0_MASK]], %[[LHS1_MASK]]) : (vector<[8]xi1>, vector<[8]xi1>) -> vector<[16]xi1>
-// CHECK-DAG: %[[RHS_MASK:.*]] = "llvm.intr.experimental.vector.interleave2"(%[[RHS0_MASK]], %[[RHS1_MASK]]) : (vector<[8]xi1>, vector<[8]xi1>) -> vector<[16]xi1>
+// CHECK-DAG: %[[LHS0:.*]] = "llvm.intr.vector.interleave2"(%[[A0]], %[[A2]]) : (vector<[4]xi8>, vector<[4]xi8>) -> vector<[8]xi8>
+// CHECK-DAG: %[[LHS1:.*]] = "llvm.intr.vector.interleave2"(%[[A1]], %[[A3]]) : (vector<[4]xi8>, vector<[4]xi8>) -> vector<[8]xi8>
+// CHECK-DAG: %[[RHS0:.*]] = "llvm.intr.vector.interleave2"(%[[B0]], %[[B2]]) : (vector<[4]xi8>, vector<[4]xi8>) -> vector<[8]xi8>
+// CHECK-DAG: %[[RHS1:.*]] = "llvm.intr.vector.interleave2"(%[[B1]], %[[B3]]) : (vector<[4]xi8>, vector<[4]xi8>) -> vector<[8]xi8>
+// CHECK-DAG: %[[LHS:.*]] = "llvm.intr.vector.interleave2"(%[[LHS0]], %[[LHS1]]) : (vector<[8]xi8>, vector<[8]xi8>) -> vector<[16]xi8>
+// CHECK-DAG: %[[RHS:.*]] = "llvm.intr.vector.interleave2"(%[[RHS0]], %[[RHS1]]) : (vector<[8]xi8>, vector<[8]xi8>) -> vector<[16]xi8>
+// CHECK-DAG: %[[LHS0_MASK:.*]] = "llvm.intr.vector.interleave2"(%[[A0_MASK]], %[[A2_MASK]]) : (vector<[4]xi1>, vector<[4]xi1>) -> vector<[8]xi1>
+// CHECK-DAG: %[[LHS1_MASK:.*]] = "llvm.intr.vector.interleave2"(%[[A1_MASK]], %[[A3_MASK]]) : (vector<[4]xi1>, vector<[4]xi1>) -> vector<[8]xi1>
+// CHECK-DAG: %[[RHS0_MASK:.*]] = "llvm.intr.vector.interleave2"(%[[B0_MASK]], %[[B2_MASK]]) : (vector<[4]xi1>, vector<[4]xi1>) -> vector<[8]xi1>
+// CHECK-DAG: %[[RHS1_MASK:.*]] = "llvm.intr.vector.interleave2"(%[[B1_MASK]], %[[B3_MASK]]) : (vector<[4]xi1>, vector<[4]xi1>) -> vector<[8]xi1>
+// CHECK-DAG: %[[LHS_MASK:.*]] = "llvm.intr.vector.interleave2"(%[[LHS0_MASK]], %[[LHS1_MASK]]) : (vector<[8]xi1>, vector<[8]xi1>) -> vector<[16]xi1>
+// CHECK-DAG: %[[RHS_MASK:.*]] = "llvm.intr.vector.interleave2"(%[[RHS0_MASK]], %[[RHS1_MASK]]) : (vector<[8]xi1>, vector<[8]xi1>) -> vector<[16]xi1>
 // CHECK-DAG: arm_sme.smopa_4way %[[LHS]], %[[RHS]] acc(%[[ACC]]) masks(%[[LHS_MASK]], %[[RHS_MASK]]) : vector<[16]xi8>, vector<[16]xi8> into vector<[4]x[4]xi32>
 func.func @outerproduct_add_widening_4way_signed_i8i8i32(
     %a0 : vector<[4]xi8>, %b0 : vector<[4]xi8>,
diff --git a/mlir/test/Dialect/ArmSME/vector-ops-to-llvm.mlir b/mlir/test/Dialect/ArmSME/vector-ops-to-llvm.mlir
deleted file mode 100644
index 17a070999c20..000000000000
--- a/mlir/test/Dialect/ArmSME/vector-ops-to-llvm.mlir
+++ /dev/null
@@ -1,877 +0,0 @@
-// RUN: mlir-opt %s -convert-vector-to-arm-sme -convert-arith-to-arm-sme -allocate-arm-sme-tiles -convert-arm-sme-to-scf -convert-arm-sme-to-llvm -cse -canonicalize -split-input-file -allow-unregistered-dialect -verify-diagnostics | FileCheck %s
-
-//===----------------------------------------------------------------------===//
-// vector.transfer_write
-//===----------------------------------------------------------------------===//
-
-// CHECK-LABEL: @transfer_write_2d_zero_i8(
-// CHECK-SAME:                             %[[ARG0:.*]]: memref<?x?xi8>)
-// CHECK-DAG:  %[[MEM_DESC:.*]] = builtin.unrealized_conversion_cast %[[ARG0]] : memref<?x?xi8> to !llvm.struct<(ptr, ptr, i64, array<2 x i64>, array<2 x i64>)>
-// CHECK-DAG:  %[[C0:.*]] = arith.constant 0 : index
-// CHECK-DAG:  %[[C1:.*]] = arith.constant 1 : index
-// CHECK-DAG:  %[[MIN_SVL_B:.*]] = arith.constant 16 : index
-// CHECK-DAG:  %[[PTRUE_ALL:.*]] = arith.constant dense<true> : vector<[16]xi1>
-// CHECK-DAG:  %[[C0_I64:.*]] = builtin.unrealized_conversion_cast %[[C0]] : index to i64
-// CHECK-DAG:  "arm_sme.intr.zero"() <{tile_mask = 255 : i32}> : () -> ()
-// CHECK-DAG:  %[[VSCALE:.*]] = vector.vscale
-// CHECK-NEXT: %[[SVL_B:.*]] = arith.muli %[[VSCALE]], %[[MIN_SVL_B]] : index
-// CHECK-NEXT: scf.for %[[TILE_SLICE:.*]] = %[[C0]] to %[[SVL_B]] step %[[C1]] {
-// CHECK:        %[[TILE_SLICE_I64:.*]] = builtin.unrealized_conversion_cast %[[TILE_SLICE]] : index to i64
-// CHECK-NEXT:   %[[ALIGNED_BASE:.*]] = llvm.extractvalue %[[MEM_DESC]][1] : !llvm.struct<(ptr, ptr, i64, array<2 x i64>, array<2 x i64>)>
-// CHECK-NEXT:   %[[STRIDE0:.*]] = llvm.extractvalue %[[MEM_DESC]][4, 0] : !llvm.struct<(ptr, ptr, i64, array<2 x i64>, array<2 x i64>)>
-// CHECK-NEXT:   %[[OFF0:.*]] = llvm.mul %[[TILE_SLICE_I64]], %[[STRIDE0]]  : i64
-// CHECK-NEXT:   %[[OFF1:.*]] = llvm.add %[[OFF0]], %[[C0_I64]]  : i64
-// CHECK-NEXT:   %[[GEP:.*]] = llvm.getelementptr %[[ALIGNED_BASE]]{{\[}}%[[OFF1]]] : (!llvm.ptr, i64) -> !llvm.ptr, i8
-// CHECK-NEXT:   %[[TILE_SLICE_I32:.*]] = arith.index_castui %[[TILE_SLICE]] : index to i32
-// CHECK-NEXT:   "arm_sme.intr.st1b.horiz"(%[[PTRUE_ALL]], %[[GEP]], %[[TILE_SLICE_I32]]) <{tile_id = 0 : i32}> : (vector<[16]xi1>, !llvm.ptr, i32) -> ()
-func.func @transfer_write_2d_zero_i8(%arg0 : memref<?x?xi8>) {
-  %c0 = arith.constant 0 : index
-  %cst = arith.constant dense<0> : vector<[16]x[16]xi8>
-  vector.transfer_write %cst, %arg0[%c0, %c0] {in_bounds = [true, true]} : vector<[16]x[16]xi8>, memref<?x?xi8>
-  return
-}
-
-//===----------------------------------------------------------------------===//
-// vector.load
-//===----------------------------------------------------------------------===//
-
-// -----
-
-// Load an 8-bit tile from a rank 2 memref with a non-zero offset for the first
-// memref index. This verifies the offset is preserved when materializing the
-// loop of tile slice loads.
-
-// CHECK-LABEL: @vector_load_i8_with_offset(
-// CHECK-SAME:                              %[[ARG0:.*]]: memref<?x?xi8>)
-// CHECK-DAG:  %[[MEM_DESC:.*]] = builtin.unrealized_conversion_cast %[[ARG0]] : memref<?x?xi8> to !llvm.struct<(ptr, ptr, i64, array<2 x i64>, array<2 x i64>)>
-// CHECK-DAG:  %[[C0:.*]] = arith.constant 0 : index
-// CHECK-DAG:  %[[C1:.*]] = arith.constant 1 : index
-// CHECK-DAG:  %[[C123:.*]] = arith.constant 123 : index
-// CHECK-DAG:  %[[MIN_SVL_B:.*]] = arith.constant 16 : index
-// CHECK-DAG:  %[[PTRUE_ALL:.*]] = arith.constant dense<true> : vector<[16]xi1>
-// CHECK-DAG:  %[[C0_I64:.*]] = builtin.unrealized_conversion_cast %[[C0]] : index to i64
-// CHECK-DAG:  %[[VSCALE:.*]] = vector.vscale
-// CHECK-NEXT: %[[SVL_B:.*]] = arith.muli %[[VSCALE]], %[[MIN_SVL_B]] : index
-// CHECK-NEXT: scf.for %[[TILE_SLICE:.*]] = %[[C0]] to %[[SVL_B]] step %[[C1]] {
-// CHECK-NEXT:   %[[TILE_SLICE_PLUS_OFF0:.*]] = arith.addi %[[TILE_SLICE]], %[[C123]] : index
-// CHECK-NEXT:   %[[TILE_SLICE_PLUS_OFF0_I64:.*]] = builtin.unrealized_conversion_cast %[[TILE_SLICE_PLUS_OFF0]] : index to i64
-// CHECK-NEXT:   %[[ALIGNED_BASE:.*]] = llvm.extractvalue %[[MEM_DESC]][1] : !llvm.struct<(ptr, ptr, i64, array<2 x i64>, array<2 x i64>)>
-// CHECK-NEXT:   %[[STRIDE0:.*]] = llvm.extractvalue %[[MEM_DESC]][4, 0] : !llvm.struct<(ptr, ptr, i64, array<2 x i64>, array<2 x i64>)>
-// CHECK-NEXT:   %[[OFF0:.*]] = llvm.mul %[[TILE_SLICE_PLUS_OFF0_I64]], %[[STRIDE0]]  : i64
-// CHECK-NEXT:   %[[OFF1:.*]] = llvm.add %[[OFF0]], %[[C0_I64]]  : i64
-// CHECK-NEXT:   %[[GEP:.*]] = llvm.getelementptr %[[ALIGNED_BASE]]{{\[}}%[[OFF1]]] : (!llvm.ptr, i64) -> !llvm.ptr, i8
-// CHECK-NEXT:   %[[TILE_SLICE_I32:.*]] = arith.index_castui %[[TILE_SLICE]] : index to i32
-// CHECK-NEXT:   "arm_sme.intr.ld1b.horiz"(%[[PTRUE_ALL]], %[[GEP]], %[[TILE_SLICE_I32]]) <{tile_id = 0 : i32}> : (vector<[16]xi1>, !llvm.ptr, i32) -> ()
-// CHECK-NEXT: }
-func.func @vector_load_i8_with_offset(%arg0 : memref<?x?xi8>) -> vector<[16]x[16]xi8> {
-  %c0 = arith.constant 0 : index
-  %c123 = arith.constant 123 : index
-  %tile = vector.load %arg0[%c123, %c0] : memref<?x?xi8>, vector<[16]x[16]xi8>
-  return %tile : vector<[16]x[16]xi8>
-}
-
-// -----
-
-// CHECK-LABEL: @vector_load_i8_from_rank_1_memref(
-// CHECK-SAME:                                     %[[ARG0:.*]]: memref<?xi8>)
-// CHECK-DAG:  %[[MEM_DESC:.*]] = builtin.unrealized_conversion_cast %[[ARG0]] : memref<?xi8> to !llvm.struct<(ptr, ptr, i64, array<1 x i64>, array<1 x i64>)>
-// CHECK-DAG:  %[[C0:.*]] = arith.constant 0 : index
-// CHECK-DAG:  %[[C1:.*]] = arith.constant 1 : index
-// CHECK-DAG:  %[[MIN_SVL_B:.*]] = arith.constant 16 : index
-// CHECK-DAG:  %[[PTRUE_ALL:.*]] = arith.constant dense<true> : vector<[16]xi1>
-// CHECK-DAG:  %[[VSCALE:.*]] = vector.vscale
-// CHECK-NEXT: %[[SVL_B:.*]] = arith.muli %[[VSCALE]], %[[MIN_SVL_B]] : index
-// CHECK-NEXT: scf.for %[[TILE_SLICE:.*]] = %[[C0]] to %[[SVL_B]] step %[[C1]] {
-// CHECK-NEXT:   %[[TILE_SLICE_IDX:.*]] = arith.muli %[[TILE_SLICE]], %[[SVL_B]] : index
-// CHECK-NEXT:   %[[TILE_SLICE_IDX_I64:.*]] = builtin.unrealized_conversion_cast %[[TILE_SLICE_IDX]] : index to i64
-// CHECK-NEXT:   %[[ALIGNED_BASE:.*]] = llvm.extractvalue %[[MEM_DESC]][1] : !llvm.struct<(ptr, ptr, i64, array<1 x i64>, array<1 x i64>)>
-// CHECK-NEXT:   %[[GEP:.*]] = llvm.getelementptr %[[ALIGNED_BASE]]{{\[}}%[[TILE_SLICE_IDX_I64]]] : (!llvm.ptr, i64) -> !llvm.ptr, i8
-// CHECK-NEXT:   %[[TILE_SLICE_I32:.*]] = arith.index_castui %[[TILE_SLICE]] : index to i32
-// CHECK-NEXT:   "arm_sme.intr.ld1b.horiz"(%[[PTRUE_ALL]], %[[GEP]], %[[TILE_SLICE_I32]]) <{tile_id = 0 : i32}> : (vector<[16]xi1>, !llvm.ptr, i32) -> ()
-// CHECK-NEXT: }
-func.func @vector_load_i8_from_rank_1_memref(%arg0 : memref<?xi8>) -> vector<[16]x[16]xi8> {
-  %c0 = arith.constant 0 : index
-  %tile = vector.load %arg0[%c0] : memref<?xi8>, vector<[16]x[16]xi8>
-  return %tile : vector<[16]x[16]xi8>
-}
-
-
-// -----
-
-// CHECK-LABEL: @vector_load_i16(
-// CHECK-SAME:                   %[[ARG0:.*]]: memref<?x?xi16>)
-// CHECK-DAG: %[[MIN_SVL_H:.*]] = arith.constant 8 : index
-// CHECK:     %[[SVL_H:.*]] = arith.muli %{{.*}}, %[[MIN_SVL_H]] : index
-// CHECK:       arm_sme.intr.ld1h.horiz
-// CHECK-SAME:  tile_id = 0
-func.func @vector_load_i16(%arg0 : memref<?x?xi16>) -> vector<[8]x[8]xi16> {
-  %c0 = arith.constant 0 : index
-  %tile = vector.load %arg0[%c0, %c0] : memref<?x?xi16>, vector<[8]x[8]xi16>
-  return %tile : vector<[8]x[8]xi16>
-}
-
-// -----
-
-// CHECK-LABEL: @vector_load_i32(
-// CHECK-SAME:                   %[[ARG0:.*]]: memref<?x?xi32>)
-// CHECK-DAG: %[[MIN_SVL_S:.*]] = arith.constant 4 : index
-// CHECK:     %[[SVL_S:.*]] = arith.muli %{{.*}}, %[[MIN_SVL_S]] : index
-// CHECK:       arm_sme.intr.ld1w.horiz
-// CHECK-SAME:  tile_id = 0
-func.func @vector_load_i32(%arg0 : memref<?x?xi32>) -> vector<[4]x[4]xi32> {
-  %c0 = arith.constant 0 : index
-  %tile = vector.load %arg0[%c0, %c0] : memref<?x?xi32>, vector<[4]x[4]xi32>
-  return %tile : vector<[4]x[4]xi32>
-}
-
-// -----
-
-// CHECK-LABEL: @vector_load_i64(
-// CHECK-SAME:                   %[[ARG0:.*]]: memref<?x?xi64>)
-// CHECK-DAG: %[[MIN_SVL_D:.*]] = arith.constant 2 : index
-// CHECK:     %[[SVL_D:.*]] = arith.muli %{{.*}}, %[[MIN_SVL_D]] : index
-// CHECK:       arm_sme.intr.ld1d.horiz
-// CHECK-SAME:  tile_id = 0
-func.func @vector_load_i64(%arg0 : memref<?x?xi64>) -> vector<[2]x[2]xi64> {
-  %c0 = arith.constant 0 : index
-  %tile = vector.load %arg0[%c0, %c0] : memref<?x?xi64>, vector<[2]x[2]xi64>
-  return %tile : vector<[2]x[2]xi64>
-}
-
-// -----
-
-// CHECK-LABEL: @vector_load_f16(
-// CHECK-SAME:                   %[[ARG0:.*]]: memref<?x?xf16>)
-// CHECK-DAG: %[[MIN_SVL_H:.*]] = arith.constant 8 : index
-// CHECK:     %[[SVL_H:.*]] = arith.muli %{{.*}}, %[[MIN_SVL_H]] : index
-// CHECK:       arm_sme.intr.ld1h.horiz
-// CHECK-SAME:  tile_id = 0
-func.func @vector_load_f16(%arg0 : memref<?x?xf16>) -> vector<[8]x[8]xf16> {
-  %c0 = arith.constant 0 : index
-  %tile = vector.load %arg0[%c0, %c0] : memref<?x?xf16>, vector<[8]x[8]xf16>
-  return %tile : vector<[8]x[8]xf16>
-}
-
-// -----
-
-// CHECK-LABEL: @vector_load_bf16(
-// CHECK-SAME:                    %[[ARG0:.*]]: memref<?x?xbf16>)
-// CHECK-DAG: %[[MIN_SVL_H:.*]] = arith.constant 8 : index
-// CHECK:     %[[SVL_H:.*]] = arith.muli %{{.*}}, %[[MIN_SVL_H]] : index
-// CHECK:       arm_sme.intr.ld1h.horiz
-// CHECK-SAME:  tile_id = 0
-func.func @vector_load_bf16(%arg0 : memref<?x?xbf16>) -> vector<[8]x[8]xbf16> {
-  %c0 = arith.constant 0 : index
-  %tile = vector.load %arg0[%c0, %c0] : memref<?x?xbf16>, vector<[8]x[8]xbf16>
-  return %tile : vector<[8]x[8]xbf16>
-}
-
-// -----
-
-// CHECK-LABEL: @vector_load_f32(
-// CHECK-SAME:                   %[[ARG0:.*]]: memref<?x?xf32>)
-// CHECK-DAG: %[[MIN_SVL_S:.*]] = arith.constant 4 : index
-// CHECK:     %[[SVL_S:.*]] = arith.muli %{{.*}}, %[[MIN_SVL_S]] : index
-// CHECK:       arm_sme.intr.ld1w.horiz
-// CHECK-SAME:  tile_id = 0
-func.func @vector_load_f32(%arg0 : memref<?x?xf32>) -> vector<[4]x[4]xf32> {
-  %c0 = arith.constant 0 : index
-  %tile = vector.load %arg0[%c0, %c0] : memref<?x?xf32>, vector<[4]x[4]xf32>
-  return %tile : vector<[4]x[4]xf32>
-}
-
-// -----
-
-// CHECK-LABEL: @vector_load_f64(
-// CHECK-SAME:                   %[[ARG0:.*]]: memref<?x?xf64>)
-// CHECK-DAG: %[[MIN_SVL_D:.*]] = arith.constant 2 : index
-// CHECK:     %[[SVL_D:.*]] = arith.muli %{{.*}}, %[[MIN_SVL_D]] : index
-// CHECK:       arm_sme.intr.ld1d.horiz
-// CHECK-SAME:  tile_id = 0
-func.func @vector_load_f64(%arg0 : memref<?x?xf64>) -> vector<[2]x[2]xf64> {
-  %c0 = arith.constant 0 : index
-  %tile = vector.load %arg0[%c0, %c0] : memref<?x?xf64>, vector<[2]x[2]xf64>
-  return %tile : vector<[2]x[2]xf64>
-}
-
-// -----
-
-// CHECK-LABEL: @vector_load_i128(
-// CHECK-SAME:                    %[[ARG0:.*]]: memref<?x?xi128>)
-// CHECK:       arm_sme.intr.ld1q.horiz
-// CHECK-SAME:  tile_id = 0
-func.func @vector_load_i128(%arg0 : memref<?x?xi128>) -> vector<[1]x[1]xi128> {
-  %c0 = arith.constant 0 : index
-  %tile = vector.load %arg0[%c0, %c0] : memref<?x?xi128>, vector<[1]x[1]xi128>
-  return %tile : vector<[1]x[1]xi128>
-}
-
-//===----------------------------------------------------------------------===//
-// vector.store
-//===----------------------------------------------------------------------===//
-
-// -----
-
-// CHECK-LABEL: @vector_store_i8(
-// CHECK-SAME:                   %[[ARG0:.*]]: memref<?x?xi8>)
-// CHECK-DAG:  %[[MEM_DESC:.*]] = builtin.unrealized_conversion_cast %[[ARG0]] : memref<?x?xi8> to !llvm.struct<(ptr, ptr, i64, array<2 x i64>, array<2 x i64>)>
-// CHECK-DAG:  %[[C0:.*]] = arith.constant 0 : index
-// CHECK-DAG:  %[[C1:.*]] = arith.constant 1 : index
-// CHECK-DAG:  %[[MIN_SVL_B:.*]] = arith.constant 16 : index
-// CHECK-DAG:  %[[C0_I64:.*]] = builtin.unrealized_conversion_cast %[[C0]] : index to i64
-// CHECK-DAG:  %[[PTRUE_ALL:.*]] = arith.constant dense<true> : vector<[16]xi1>
-// CHECK-DAG:  %[[VSCALE:.*]] = vector.vscale
-// CHECK-NEXT: %[[SVL_B:.*]] = arith.muli %[[VSCALE]], %[[MIN_SVL_B]] : index
-// CHECK-NEXT: scf.for %[[TILE_SLICE:.*]] = %[[C0]] to %[[SVL_B]] step %[[C1]] {
-// CHECK:        %[[TILE_SLICE_I64:.*]] = builtin.unrealized_conversion_cast %[[TILE_SLICE]] : index to i64
-// CHECK-NEXT:   %[[ALIGNED_BASE:.*]] = llvm.extractvalue %[[MEM_DESC]][1] : !llvm.struct<(ptr, ptr, i64, array<2 x i64>, array<2 x i64>)>
-// CHECK-NEXT:   %[[STRIDE0:.*]] = llvm.extractvalue %[[MEM_DESC]][4, 0] : !llvm.struct<(ptr, ptr, i64, array<2 x i64>, array<2 x i64>)>
-// CHECK-NEXT:   %[[OFF0:.*]] = llvm.mul %[[TILE_SLICE_I64]], %[[STRIDE0]]  : i64
-// CHECK-NEXT:   %[[OFF1:.*]] = llvm.add %[[OFF0]], %[[C0_I64]]  : i64
-// CHECK-NEXT:   %[[GEP:.*]] = llvm.getelementptr %[[ALIGNED_BASE]]{{\[}}%[[OFF1]]] : (!llvm.ptr, i64) -> !llvm.ptr, i8
-// CHECK-NEXT:   %[[TILE_SLICE_I32:.*]] = arith.index_castui %[[TILE_SLICE]] : index to i32
-// CHECK-NEXT:   "arm_sme.intr.st1b.horiz"(%[[PTRUE_ALL]], %[[GEP]], %[[TILE_SLICE_I32]]) <{tile_id = 0 : i32}> : (vector<[16]xi1>, !llvm.ptr, i32) -> ()
-// CHECK-NEXT: }
-// CHECK-NEXT: return
-func.func @vector_store_i8(%arg0 : memref<?x?xi8>) {
-  %c0 = arith.constant 0 : index
-  %tile = arm_sme.get_tile : vector<[16]x[16]xi8>
-  vector.store %tile, %arg0[%c0, %c0] : memref<?x?xi8>, vector<[16]x[16]xi8>
-  return
-}
-
-// -----
-
-// CHECK-LABEL: @vector_store_i16(
-// CHECK-SAME:                   %[[ARG0:.*]]: memref<?x?xi16>)
-// CHECK:     %[[MIN_SVL_H:.*]] = arith.constant 8 : index
-// CHECK:     %[[SVL_H:.*]] = arith.muli %{{.*}}, %[[MIN_SVL_H]] : index
-// CHECK:       arm_sme.intr.st1h.horiz
-// CHECK-SAME:  tile_id = 0
-func.func @vector_store_i16(%arg0 : memref<?x?xi16>) {
-  %c0 = arith.constant 0 : index
-  %tile = arm_sme.get_tile : vector<[8]x[8]xi16>
-  vector.store %tile, %arg0[%c0, %c0] : memref<?x?xi16>, vector<[8]x[8]xi16>
-  return
-}
-
-// -----
-
-// CHECK-LABEL: @vector_store_i32(
-// CHECK-SAME:                   %[[ARG0:.*]]: memref<?x?xi32>)
-// CHECK:     %[[MIN_SVL_S:.*]] = arith.constant 4 : index
-// CHECK:     %[[SVL_S:.*]] = arith.muli %{{.*}}, %[[MIN_SVL_S]] : index
-// CHECK:       arm_sme.intr.st1w.horiz
-// CHECK-SAME:  tile_id = 0
-func.func @vector_store_i32(%arg0 : memref<?x?xi32>) {
-  %c0 = arith.constant 0 : index
-  %tile = arm_sme.get_tile : vector<[4]x[4]xi32>
-  vector.store %tile, %arg0[%c0, %c0] : memref<?x?xi32>, vector<[4]x[4]xi32>
-  return
-}
-
-// -----
-
-// CHECK-LABEL: @vector_store_i64(
-// CHECK-SAME:                   %[[ARG0:.*]]: memref<?x?xi64>)
-// CHECK:     %[[MIN_SVL_D:.*]] = arith.constant 2 : index
-// CHECK:     %[[SVL_D:.*]] = arith.muli %{{.*}}, %[[MIN_SVL_D]] : index
-// CHECK:       arm_sme.intr.st1d.horiz
-// CHECK-SAME:  tile_id = 0
-func.func @vector_store_i64(%arg0 : memref<?x?xi64>) {
-  %c0 = arith.constant 0 : index
-  %tile = arm_sme.get_tile : vector<[2]x[2]xi64>
-  vector.store %tile, %arg0[%c0, %c0] : memref<?x?xi64>, vector<[2]x[2]xi64>
-  return
-}
-
-// -----
-
-// CHECK-LABEL: @vector_store_f16(
-// CHECK-SAME:                   %[[ARG0:.*]]: memref<?x?xf16>)
-// CHECK:     %[[MIN_SVL_H:.*]] = arith.constant 8 : index
-// CHECK:     %[[SVL_H:.*]] = arith.muli %{{.*}}, %[[MIN_SVL_H]] : index
-// CHECK:       arm_sme.intr.st1h.horiz
-// CHECK-SAME:  tile_id = 0
-func.func @vector_store_f16(%arg0 : memref<?x?xf16>) {
-  %c0 = arith.constant 0 : index
-  %tile = arm_sme.get_tile : vector<[8]x[8]xf16>
-  vector.store %tile, %arg0[%c0, %c0] : memref<?x?xf16>, vector<[8]x[8]xf16>
-  return
-}
-
-// -----
-
-// CHECK-LABEL: @vector_store_bf16(
-// CHECK-SAME:                   %[[ARG0:.*]]: memref<?x?xbf16>)
-// CHECK:     %[[MIN_SVL_H:.*]] = arith.constant 8 : index
-// CHECK:     %[[SVL_H:.*]] = arith.muli %{{.*}}, %[[MIN_SVL_H]] : index
-// CHECK:       arm_sme.intr.st1h.horiz
-// CHECK-SAME:  tile_id = 0
-func.func @vector_store_bf16(%arg0 : memref<?x?xbf16>) {
-  %c0 = arith.constant 0 : index
-  %tile = arm_sme.get_tile : vector<[8]x[8]xbf16>
-  vector.store %tile, %arg0[%c0, %c0] : memref<?x?xbf16>, vector<[8]x[8]xbf16>
-  return
-}
-// -----
-
-// CHECK-LABEL: @vector_store_f32(
-// CHECK-SAME:                   %[[ARG0:.*]]: memref<?x?xf32>)
-// CHECK:     %[[MIN_SVL_S:.*]] = arith.constant 4 : index
-// CHECK:     %[[SVL_S:.*]] = arith.muli %{{.*}}, %[[MIN_SVL_S]] : index
-// CHECK:       arm_sme.intr.st1w.horiz
-// CHECK-SAME:  tile_id = 0
-func.func @vector_store_f32(%arg0 : memref<?x?xf32>) {
-  %c0 = arith.constant 0 : index
-  %tile = arm_sme.get_tile : vector<[4]x[4]xf32>
-  vector.store %tile, %arg0[%c0, %c0] : memref<?x?xf32>, vector<[4]x[4]xf32>
-  return
-}
-
-// -----
-
-// CHECK-LABEL: @vector_store_f64(
-// CHECK-SAME:                   %[[ARG0:.*]]: memref<?x?xf64>)
-// CHECK:     %[[MIN_SVL_D:.*]] = arith.constant 2 : index
-// CHECK:     %[[SVL_D:.*]] = arith.muli %{{.*}}, %[[MIN_SVL_D]] : index
-// CHECK:       arm_sme.intr.st1d.horiz
-// CHECK-SAME:  tile_id = 0
-func.func @vector_store_f64(%arg0 : memref<?x?xf64>) {
-  %c0 = arith.constant 0 : index
-  %tile = arm_sme.get_tile : vector<[2]x[2]xf64>
-  vector.store %tile, %arg0[%c0, %c0] : memref<?x?xf64>, vector<[2]x[2]xf64>
-  return
-}
-
-// -----
-
-// CHECK-LABEL: @vector_store_i128(
-// CHECK-SAME:                     %[[ARG0:.*]]: memref<?x?xi128>)
-// CHECK:       arm_sme.intr.st1q.horiz
-// CHECK-SAME:  tile_id = 0
-func.func @vector_store_i128(%arg0 : memref<?x?xi128>) {
-  %c0 = arith.constant 0 : index
-  %tile = arm_sme.get_tile : vector<[1]x[1]xi128>
-  vector.store %tile, %arg0[%c0, %c0] : memref<?x?xi128>, vector<[1]x[1]xi128>
-  return
-}
-
-//===----------------------------------------------------------------------===//
-// vector.outerproduct
-//===----------------------------------------------------------------------===//
-
-// -----
-
-// CHECK-LABEL: @vector_outerproduct_add_f16
-// CHECK-SAME: (%[[LHS:.*]]: vector<[8]xf16>, %[[RHS:.*]]: vector<[8]xf16>)
-func.func @vector_outerproduct_add_f16(%lhs : vector<[8]xf16>, %rhs : vector<[8]xf16>) {
-  // CHECK: %[[PTRUE_ALL:.*]] = arith.constant dense<true> : vector<[8]xi1>
-  // CHECK: "arm_sme.intr.mopa"(%[[PTRUE_ALL]], %[[PTRUE_ALL]], %[[LHS]], %[[RHS]]) <{tile_id = 0 : i32}> : (vector<[8]xi1>, vector<[8]xi1>, vector<[8]xf16>, vector<[8]xf16>)
-  %acc = arm_sme.get_tile : vector<[8]x[8]xf16>
-  %0 = vector.outerproduct %lhs, %rhs, %acc {kind = #vector.kind<add>} : vector<[8]xf16>, vector<[8]xf16>
-  "prevent.dce"(%0) : (vector<[8]x[8]xf16>) -> ()
-}
-
-// -----
-
-// CHECK-LABEL: @vector_outerproduct_add_bf16
-func.func @vector_outerproduct_add_bf16(%lhs : vector<[8]xbf16>, %rhs : vector<[8]xbf16>) {
-  // CHECK: "arm_sme.intr.mopa"({{.*}}) <{tile_id = 0 : i32}> : (vector<[8]xi1>, vector<[8]xi1>, vector<[8]xbf16>, vector<[8]xbf16>)
-  %acc = arm_sme.get_tile : vector<[8]x[8]xbf16>
-  %0 = vector.outerproduct %lhs, %rhs, %acc {kind = #vector.kind<add>} : vector<[8]xbf16>, vector<[8]xbf16>
-  "prevent.dce"(%0) : (vector<[8]x[8]xbf16>) -> ()
-}
-
-// -----
-
-// CHECK-LABEL: @vector_outerproduct_add_f32
-func.func @vector_outerproduct_add_f32(%lhs : vector<[4]xf32>, %rhs : vector<[4]xf32>) {
-  // CHECK: "arm_sme.intr.mopa"({{.*}}) <{tile_id = 0 : i32}> : (vector<[4]xi1>, vector<[4]xi1>, vector<[4]xf32>, vector<[4]xf32>)
-  %acc = arm_sme.get_tile : vector<[4]x[4]xf32>
-  %0 = vector.outerproduct %lhs, %rhs, %acc {kind = #vector.kind<add>} : vector<[4]xf32>, vector<[4]xf32>
-  "prevent.dce"(%0) : (vector<[4]x[4]xf32>) -> ()
-}
-
-// -----
-
-// CHECK-LABEL: @vector_outerproduct_add_f64
-func.func @vector_outerproduct_add_f64(%lhs : vector<[2]xf64>, %rhs : vector<[2]xf64>) {
-  // CHECK: "arm_sme.intr.mopa"({{.*}}) <{tile_id = 0 : i32}> : (vector<[2]xi1>, vector<[2]xi1>, vector<[2]xf64>, vector<[2]xf64>)
-  %acc = arm_sme.get_tile : vector<[2]x[2]xf64>
-  %0 = vector.outerproduct %lhs, %rhs, %acc {kind = #vector.kind<add>} : vector<[2]xf64>, vector<[2]xf64>
-  "prevent.dce"(%0) : (vector<[2]x[2]xf64>) -> ()
-}
-
-// -----
-
-// CHECK-LABEL: @vector_outerproduct_no_accumulator
-func.func @vector_outerproduct_no_accumulator(%lhs : vector<[2]xf64>, %rhs : vector<[2]xf64>) {
-  // CHECK: "arm_sme.intr.zero"() <{tile_mask = 1 : i32}> : () -> ()
-  // CHECK: "arm_sme.intr.mopa"({{.*}}) <{tile_id = 0 : i32}> : (vector<[2]xi1>, vector<[2]xi1>, vector<[2]xf64>, vector<[2]xf64>)
-  %0 = vector.outerproduct %lhs, %rhs {kind = #vector.kind<add>} : vector<[2]xf64>, vector<[2]xf64>
-  "prevent.dce"(%0) : (vector<[2]x[2]xf64>) -> ()
-}
-
-// -----
-
-// CHECK-LABEL: @vector_outerproduct_masked_f32
-// CHECK-SAME: (%[[LHS:.*]]: vector<[4]xf32>, %[[RHS:.*]]: vector<[4]xf32>, %[[DIM0:.*]]: index, %[[DIM1:.*]]: index
-func.func @vector_outerproduct_masked_f32(%lhs : vector<[4]xf32>, %rhs : vector<[4]xf32>, %dim0 : index, %dim1 : index) {
-  // CHECK: %[[LHS_MASK:.*]] = vector.create_mask %[[DIM0]] : vector<[4]xi1>
-  // CHECK: %[[RHS_MASK:.*]] = vector.create_mask %[[DIM1]] : vector<[4]xi1>
-  // CHECK: "arm_sme.intr.mopa"(%[[LHS_MASK]], %[[RHS_MASK]], %[[LHS]], %[[RHS]]) <{tile_id = 0 : i32}> : (vector<[4]xi1>, vector<[4]xi1>, vector<[4]xf32>, vector<[4]xf32>)
-  %acc = arm_sme.get_tile : vector<[4]x[4]xf32>
-  %mask = vector.create_mask %dim0, %dim1 : vector<[4]x[4]xi1>
-  %result = vector.mask %mask { vector.outerproduct %lhs, %rhs, %acc {kind = #vector.kind<add>} : vector<[4]xf32>, vector<[4]xf32> } : vector<[4]x[4]xi1> -> vector<[4]x[4]xf32>
-  "prevent.dce"(%result) : (vector<[4]x[4]xf32>) -> ()
-}
-
-// -----
-
-// CHECK-LABEL: @vector_outerproduct_masked_f16
-// CHECK-SAME: (%[[LHS:.*]]: vector<[8]xf16>, %[[RHS:.*]]: vector<[8]xf16>,
-func.func @vector_outerproduct_masked_f16(%lhs : vector<[8]xf16>, %rhs : vector<[8]xf16>, %dim0 : index, %dim1 : index) {
-  // CHECK: vector.create_mask {{.*}} : vector<[8]xi1>
-  // CHECK: vector.create_mask {{.*}} : vector<[8]xi1>
-  // CHECK: "arm_sme.intr.mopa"({{.*}}) <{tile_id = 0 : i32}> : (vector<[8]xi1>, vector<[8]xi1>, vector<[8]xf16>, vector<[8]xf16>)
-  %acc = arm_sme.get_tile : vector<[8]x[8]xf16>
-  %mask = vector.create_mask %dim0, %dim1 : vector<[8]x[8]xi1>
-  %result = vector.mask %mask { vector.outerproduct %lhs, %rhs, %acc {kind = #vector.kind<add>} : vector<[8]xf16>, vector<[8]xf16> } : vector<[8]x[8]xi1> -> vector<[8]x[8]xf16>
-  "prevent.dce"(%result) : (vector<[8]x[8]xf16>) -> ()
-}
-
-// -----
-
-// CHECK-LABEL: @vector_outerproduct_masked_bf16
-// CHECK-SAME: (%[[LHS:.*]]: vector<[8]xbf16>, %[[RHS:.*]]: vector<[8]xbf16>
-func.func @vector_outerproduct_masked_bf16(%lhs : vector<[8]xbf16>, %rhs : vector<[8]xbf16>, %dim0 : index, %dim1 : index) {
-  // CHECK: vector.create_mask {{.*}} : vector<[8]xi1>
-  // CHECK: vector.create_mask {{.*}} : vector<[8]xi1>
-  // CHECK: "arm_sme.intr.mopa"({{.*}}) <{tile_id = 0 : i32}> : (vector<[8]xi1>, vector<[8]xi1>, vector<[8]xbf16>, vector<[8]xbf16>)
-  %acc = arm_sme.get_tile : vector<[8]x[8]xbf16>
-  %mask = vector.create_mask %dim0, %dim1 : vector<[8]x[8]xi1>
-  %result = vector.mask %mask { vector.outerproduct %lhs, %rhs, %acc {kind = #vector.kind<add>} : vector<[8]xbf16>, vector<[8]xbf16> } : vector<[8]x[8]xi1> -> vector<[8]x[8]xbf16>
-  "prevent.dce"(%result) : (vector<[8]x[8]xbf16>) -> ()
-}
-
-// -----
-
-// CHECK-LABEL: @vector_outerproduct_masked_f64
-// CHECK-SAME: (%[[LHS:.*]]: vector<[2]xf64>, %[[RHS:.*]]: vector<[2]xf64>,
-func.func @vector_outerproduct_masked_f64(%lhs : vector<[2]xf64>, %rhs : vector<[2]xf64>, %dim0 : index, %dim1 : index) {
-  // CHECK: vector.create_mask {{.*}} : vector<[2]xi1>
-  // CHECK: vector.create_mask {{.*}} : vector<[2]xi1>
-  // CHECK: "arm_sme.intr.mopa"({{.*}}) <{tile_id = 0 : i32}> : (vector<[2]xi1>, vector<[2]xi1>, vector<[2]xf64>, vector<[2]xf64>)
-  %acc = arm_sme.get_tile : vector<[2]x[2]xf64>
-  %mask = vector.create_mask %dim0, %dim1 : vector<[2]x[2]xi1>
-  %result = vector.mask %mask { vector.outerproduct %lhs, %rhs, %acc {kind = #vector.kind<add>} : vector<[2]xf64>, vector<[2]xf64> } : vector<[2]x[2]xi1> -> vector<[2]x[2]xf64>
-  "prevent.dce"(%result) : (vector<[2]x[2]xf64>) -> ()
-}
-
-//===----------------------------------------------------------------------===//
-// vector.insert
-//===----------------------------------------------------------------------===//
-
-// -----
-
-// CHECK-LABEL: @vector_insert_slice_i32(
-// CHECK-SAME:                       %[[SLICE:.*]]: vector<[4]xi32>,
-// CHECK-SAME:                       %[[INDEX:.*]]: index)
-func.func @vector_insert_slice_i32(%slice: vector<[4]xi32>, %row: index) -> vector<[4]x[4]xi32>{
-  // CHECK-NEXT: %[[PTRUE:.*]] = arith.constant dense<true> : vector<[4]xi1>
-  // CHECK: %[[TILE_SLICE_INDEX:.*]] = arith.index_castui %[[INDEX]] : index to i32
-  // CHECK-NEXT: "arm_sme.intr.write.horiz"(%[[TILE_SLICE_INDEX]], %[[PTRUE]], %[[SLICE]]) <{tile_id = 0 : i32}> : (i32, vector<[4]xi1>, vector<[4]xi32>) -> ()
-  %tile = arm_sme.get_tile : vector<[4]x[4]xi32>
-  %new_tile = vector.insert %slice, %tile[%row] : vector<[4]xi32> into vector<[4]x[4]xi32>
-  return %new_tile : vector<[4]x[4]xi32>
-}
-
-// -----
-
-// CHECK-LABEL: @vector_insert_slice_i8
-func.func @vector_insert_slice_i8(%slice: vector<[16]xi8>, %row: index) -> vector<[16]x[16]xi8> {
-  // CHECK: "arm_sme.intr.write.horiz"(%{{.*}}) <{tile_id = 0 : i32}> : (i32, vector<[16]xi1>, vector<[16]xi8>) -> ()
-  %tile = arm_sme.get_tile : vector<[16]x[16]xi8>
-  %new_tile = vector.insert %slice, %tile[%row] : vector<[16]xi8> into vector<[16]x[16]xi8>
-  return %new_tile : vector<[16]x[16]xi8>
-}
-
-// -----
-
-// CHECK-LABEL: @vector_insert_slice_i16
-func.func @vector_insert_slice_i16(%slice: vector<[8]xi16>, %row: index) -> vector<[8]x[8]xi16> {
-  // CHECK: "arm_sme.intr.write.horiz"(%{{.*}}) <{tile_id = 0 : i32}> : (i32, vector<[8]xi1>, vector<[8]xi16>) -> ()
-  %tile = arm_sme.get_tile : vector<[8]x[8]xi16>
-  %new_tile = vector.insert %slice, %tile[%row] : vector<[8]xi16> into vector<[8]x[8]xi16>
-  return %new_tile : vector<[8]x[8]xi16>
-}
-
-// -----
-
-// CHECK-LABEL: @vector_insert_slice_i64
-func.func @vector_insert_slice_i64(%slice: vector<[2]xi64>, %row: index) -> vector<[2]x[2]xi64> {
-  // CHECK: "arm_sme.intr.write.horiz"(%{{.*}}) <{tile_id = 0 : i32}> : (i32, vector<[2]xi1>, vector<[2]xi64>) -> ()
-  %tile = arm_sme.get_tile : vector<[2]x[2]xi64>
-  %new_tile = vector.insert %slice, %tile[%row] : vector<[2]xi64> into vector<[2]x[2]xi64>
-  return %new_tile : vector<[2]x[2]xi64>
-}
-
-// -----
-
-// CHECK-LABEL: @vector_insert_slice_i128
-func.func @vector_insert_slice_i128(%slice: vector<[1]xi128>, %row: index) -> vector<[1]x[1]xi128> {
-  // CHECK: "arm_sme.intr.write.horiz"(%{{.*}}) <{tile_id = 0 : i32}> : (i32, vector<[1]xi1>, vector<[1]xi128>) -> ()
-  %tile = arm_sme.get_tile : vector<[1]x[1]xi128>
-  %new_tile = vector.insert %slice, %tile[%row] : vector<[1]xi128> into vector<[1]x[1]xi128>
-  return %new_tile : vector<[1]x[1]xi128>
-}
-
-// -----
-
-// CHECK-LABEL: @vector_insert_slice_f16
-func.func @vector_insert_slice_f16(%slice: vector<[8]xf16>, %row: index) -> vector<[8]x[8]xf16> {
-  // CHECK: "arm_sme.intr.write.horiz"(%{{.*}}) <{tile_id = 0 : i32}> : (i32, vector<[8]xi1>, vector<[8]xf16>) -> ()
-  %tile = arm_sme.get_tile : vector<[8]x[8]xf16>
-  %new_tile = vector.insert %slice, %tile[%row] : vector<[8]xf16> into vector<[8]x[8]xf16>
-  return %new_tile : vector<[8]x[8]xf16>
-}
-
-// -----
-
-// CHECK-LABEL: @vector_insert_slice_bf16
-func.func @vector_insert_slice_bf16(%slice: vector<[8]xbf16>, %row: index) -> vector<[8]x[8]xbf16> {
-  // CHECK: "arm_sme.intr.write.horiz"(%{{.*}}) <{tile_id = 0 : i32}> : (i32, vector<[8]xi1>, vector<[8]xbf16>) -> ()
-  %tile = arm_sme.get_tile : vector<[8]x[8]xbf16>
-  %new_tile = vector.insert %slice, %tile[%row] : vector<[8]xbf16> into vector<[8]x[8]xbf16>
-  return %new_tile : vector<[8]x[8]xbf16>
-}
-
-// -----
-
-// CHECK-LABEL: @vector_insert_slice_f32
-func.func @vector_insert_slice_f32(%slice: vector<[4]xf32>, %row: index) -> vector<[4]x[4]xf32> {
-  // CHECK: "arm_sme.intr.write.horiz"(%{{.*}}) <{tile_id = 0 : i32}> : (i32, vector<[4]xi1>, vector<[4]xf32>) -> ()
-  %tile = arm_sme.get_tile : vector<[4]x[4]xf32>
-  %new_tile = vector.insert %slice, %tile[%row] : vector<[4]xf32> into vector<[4]x[4]xf32>
-  return %new_tile : vector<[4]x[4]xf32>
-}
-
-// -----
-
-// CHECK-LABEL: @vector_insert_slice_f64
-func.func @vector_insert_slice_f64(%slice: vector<[2]xf64>, %row: index) -> vector<[2]x[2]xf64> {
-  // CHECK: "arm_sme.intr.write.horiz"(%{{.*}}) <{tile_id = 0 : i32}> : (i32, vector<[2]xi1>, vector<[2]xf64>) -> ()
-  %tile = arm_sme.get_tile : vector<[2]x[2]xf64>
-  %new_tile = vector.insert %slice, %tile[%row] : vector<[2]xf64> into vector<[2]x[2]xf64>
-  return %new_tile : vector<[2]x[2]xf64>
-}
-
-// -----
-
-// CHECK-LABEL: @vector_insert_element_i32(
-// CHECK-SAME:                         %[[EL:.*]]: i32,
-// CHECK-SAME:                         %[[ROW:.*]]: index,
-// CHECK-SAME:                         %[[COL:.*]]: index)
-func.func @vector_insert_element_i32(%el: i32, %row: index, %col: index) -> vector<[4]x[4]xi32> {
-  // CHECK-DAG:  %[[ZERO_VEC:.*]] = arith.constant dense<0> : vector<[4]xi32>
-  // CHECK-DAG:  %[[PTRUE:.*]] = arith.constant dense<true> : vector<[4]xi1>
-  // CHECK-DAG:  %[[ROW_I32:.*]] = arith.index_cast %[[ROW]] : index to i32
-  // CHECK-NEXT: %[[SLICE:.*]] = "arm_sme.intr.read.horiz"(%[[ZERO_VEC]], %[[PTRUE]], %[[ROW_I32]]) <{tile_id = 0 : i32}> : (vector<[4]xi32>, vector<[4]xi1>, i32) -> vector<[4]xi32>
-  // CHECK-NEXT: %[[NEW_SLICE:.*]] = vector.insert %[[EL]], %[[SLICE]] [%[[COL]]] : i32 into vector<[4]xi32>
-  // CHECK-NEXT: %[[SLICE_INDEX:.*]] = arith.index_castui %[[ROW]] : index to i32
-  // CHECK-NEXT: "arm_sme.intr.write.horiz"(%[[SLICE_INDEX]], %[[PTRUE]], %[[NEW_SLICE]]) <{tile_id = 0 : i32}> : (i32, vector<[4]xi1>, vector<[4]xi32>) -> ()
-  %tile = arm_sme.get_tile : vector<[4]x[4]xi32>
-  %new_tile = vector.insert %el, %tile[%row, %col] : i32 into vector<[4]x[4]xi32>
-  return %new_tile : vector<[4]x[4]xi32>
-}
-
-// -----
-
-// CHECK-LABEL: @vector_insert_element_i8
-func.func @vector_insert_element_i8(%el: i8, %row: index, %col: index) -> vector<[16]x[16]xi8> {
-  // CHECK: %{{.*}} = "arm_sme.intr.read.horiz"(%{{.*}}) <{tile_id = 0 : i32}> : (vector<[16]xi8>, vector<[16]xi1>, i32) -> vector<[16]xi8>
-  // CHECK: "arm_sme.intr.write.horiz"(%{{.*}}) <{tile_id = 0 : i32}> : (i32, vector<[16]xi1>, vector<[16]xi8>) -> ()
-  %tile = arm_sme.get_tile : vector<[16]x[16]xi8>
-  %new_tile = vector.insert %el, %tile[%row, %col] : i8 into vector<[16]x[16]xi8>
-  return %new_tile : vector<[16]x[16]xi8>
-}
-
-// -----
-
-// CHECK-LABEL: @vector_insert_element_i16
-func.func @vector_insert_element_i16(%el: i16, %row: index, %col: index) -> vector<[8]x[8]xi16> {
-  // CHECK: %{{.*}} = "arm_sme.intr.read.horiz"(%{{.*}}) <{tile_id = 0 : i32}> : (vector<[8]xi16>, vector<[8]xi1>, i32) -> vector<[8]xi16>
-  // CHECK: "arm_sme.intr.write.horiz"(%{{.*}}) <{tile_id = 0 : i32}> : (i32, vector<[8]xi1>, vector<[8]xi16>) -> ()
-  %tile = arm_sme.get_tile : vector<[8]x[8]xi16>
-  %new_tile = vector.insert %el, %tile[%row, %col] : i16 into vector<[8]x[8]xi16>
-  return %new_tile : vector<[8]x[8]xi16>
-}
-
-// -----
-
-// CHECK-LABEL: @vector_insert_element_i64
-func.func @vector_insert_element_i64(%el: i64, %row: index, %col: index) -> vector<[2]x[2]xi64> {
-  // CHECK: %{{.*}} = "arm_sme.intr.read.horiz"(%{{.*}}) <{tile_id = 0 : i32}> : (vector<[2]xi64>, vector<[2]xi1>, i32) -> vector<[2]xi64>
-  // CHECK: "arm_sme.intr.write.horiz"(%{{.*}}) <{tile_id = 0 : i32}> : (i32, vector<[2]xi1>, vector<[2]xi64>) -> ()
-  %tile = arm_sme.get_tile : vector<[2]x[2]xi64>
-  %new_tile = vector.insert %el, %tile[%row, %col] : i64 into vector<[2]x[2]xi64>
-  return %new_tile : vector<[2]x[2]xi64>
-}
-
-// -----
-
-// CHECK-LABEL: @vector_insert_element_i128
-func.func @vector_insert_element_i128(%el: i128, %row: index, %col: index) -> vector<[1]x[1]xi128> {
-  // CHECK: %{{.*}} = "arm_sme.intr.read.horiz"(%{{.*}}) <{tile_id = 0 : i32}> : (vector<[1]xi128>, vector<[1]xi1>, i32) -> vector<[1]xi128>
-  // CHECK: "arm_sme.intr.write.horiz"(%{{.*}}) <{tile_id = 0 : i32}> : (i32, vector<[1]xi1>, vector<[1]xi128>) -> ()
-  %tile = arm_sme.get_tile : vector<[1]x[1]xi128>
-  %new_tile = vector.insert %el, %tile[%row, %col] : i128 into vector<[1]x[1]xi128>
-  return %new_tile : vector<[1]x[1]xi128>
-}
-
-// -----
-
-// CHECK-LABEL: @vector_insert_element_f16
-func.func @vector_insert_element_f16(%el: f16, %row: index, %col: index) -> vector<[8]x[8]xf16> {
-  // CHECK: %{{.*}} = "arm_sme.intr.read.horiz"(%{{.*}}) <{tile_id = 0 : i32}> : (vector<[8]xf16>, vector<[8]xi1>, i32) -> vector<[8]xf16>
-  // CHECK: "arm_sme.intr.write.horiz"(%{{.*}}) <{tile_id = 0 : i32}> : (i32, vector<[8]xi1>, vector<[8]xf16>) -> ()
-  %tile = arm_sme.get_tile : vector<[8]x[8]xf16>
-  %new_tile = vector.insert %el, %tile[%row, %col] : f16 into vector<[8]x[8]xf16>
-  return %new_tile : vector<[8]x[8]xf16>
-}
-
-// -----
-
-// CHECK-LABEL: @vector_insert_element_bf16
-func.func @vector_insert_element_bf16(%el: bf16, %row: index, %col: index) -> vector<[8]x[8]xbf16> {
-  // CHECK: %{{.*}} = "arm_sme.intr.read.horiz"(%{{.*}}) <{tile_id = 0 : i32}> : (vector<[8]xbf16>, vector<[8]xi1>, i32) -> vector<[8]xbf16>
-  // CHECK: "arm_sme.intr.write.horiz"(%{{.*}}) <{tile_id = 0 : i32}> : (i32, vector<[8]xi1>, vector<[8]xbf16>) -> ()
-  %tile = arm_sme.get_tile : vector<[8]x[8]xbf16>
-  %new_tile = vector.insert %el, %tile[%row, %col] : bf16 into vector<[8]x[8]xbf16>
-  return %new_tile : vector<[8]x[8]xbf16>
-}
-
-// -----
-
-// CHECK-LABEL: @vector_insert_element_f32
-func.func @vector_insert_element_f32(%el: f32, %row: index, %col: index) -> vector<[4]x[4]xf32> {
-  // CHECK: %{{.*}} = "arm_sme.intr.read.horiz"(%{{.*}}) <{tile_id = 0 : i32}> : (vector<[4]xf32>, vector<[4]xi1>, i32) -> vector<[4]xf32>
-  // CHECK: "arm_sme.intr.write.horiz"(%{{.*}}) <{tile_id = 0 : i32}> : (i32, vector<[4]xi1>, vector<[4]xf32>) -> ()
-  %tile = arm_sme.get_tile : vector<[4]x[4]xf32>
-  %new_tile = vector.insert %el, %tile[%row, %col] : f32 into vector<[4]x[4]xf32>
-  return %new_tile : vector<[4]x[4]xf32>
-}
-
-// -----
-
-// CHECK-LABEL: @vector_insert_element_f64
-func.func @vector_insert_element_f64(%el: f64, %row: index, %col: index) -> vector<[2]x[2]xf64> {
-  // CHECK: %{{.*}} = "arm_sme.intr.read.horiz"(%{{.*}}) <{tile_id = 0 : i32}> : (vector<[2]xf64>, vector<[2]xi1>, i32) -> vector<[2]xf64>
-  // CHECK: "arm_sme.intr.write.horiz"(%{{.*}}) <{tile_id = 0 : i32}> : (i32, vector<[2]xi1>, vector<[2]xf64>) -> ()
-  %tile = arm_sme.get_tile : vector<[2]x[2]xf64>
-  %new_tile = vector.insert %el, %tile[%row, %col] : f64 into vector<[2]x[2]xf64>
-  return %new_tile : vector<[2]x[2]xf64>
-}
-
-//===----------------------------------------------------------------------===//
-// vector.extract
-//===----------------------------------------------------------------------===//
-
-// -----
-
-// CHECK-LABEL: @vector_extract_slice_i32(
-// CHECK-SAME:                        %[[INDEX:.*]]: index)
-func.func @vector_extract_slice_i32(%row: index) -> vector<[4]xi32> {
-  // CHECK-NEXT: %[[PTRUE:.*]] = arith.constant dense<true> : vector<[4]xi1>
-  // CHECK-NEXT: %[[ZERO_VEC:.*]] = arith.constant dense<0> : vector<[4]xi32>
-  // CHECK-NEXT: %[[TILE_SLICE_INDEX:.*]] = arith.index_cast %[[INDEX]] : index to i32
-  // CHECK-NEXT: %[[SLICE:.*]] = "arm_sme.intr.read.horiz"(%[[ZERO_VEC]], %[[PTRUE]], %[[TILE_SLICE_INDEX]]) <{tile_id = 0 : i32}> : (vector<[4]xi32>, vector<[4]xi1>, i32) -> vector<[4]xi32>
-  %tile = arm_sme.get_tile : vector<[4]x[4]xi32>
-  %slice = vector.extract %tile[%row] : vector<[4]xi32> from vector<[4]x[4]xi32>
-  return %slice : vector<[4]xi32>
-}
-
-// -----
-
-// CHECK-LABEL: @vector_extract_slice_i8
-func.func @vector_extract_slice_i8(%row: index) -> vector<[16]xi8> {
-  // CHECK: %{{.*}} = "arm_sme.intr.read.horiz"(%{{.*}}) <{tile_id = 0 : i32}> : (vector<[16]xi8>, vector<[16]xi1>, i32) -> vector<[16]xi8>
-  %tile = arm_sme.get_tile : vector<[16]x[16]xi8>
-  %slice = vector.extract %tile[%row] : vector<[16]xi8> from vector<[16]x[16]xi8>
-  return %slice : vector<[16]xi8>
-}
-
-// -----
-
-// CHECK-LABEL: @vector_extract_slice_i16
-func.func @vector_extract_slice_i16(%row: index) -> vector<[8]xi16> {
-  // CHECK: %{{.*}} = "arm_sme.intr.read.horiz"(%{{.*}}) <{tile_id = 0 : i32}> : (vector<[8]xi16>, vector<[8]xi1>, i32) -> vector<[8]xi16>
-  %tile = arm_sme.get_tile : vector<[8]x[8]xi16>
-  %slice = vector.extract %tile[%row] : vector<[8]xi16> from vector<[8]x[8]xi16>
-  return %slice : vector<[8]xi16>
-}
-
-// -----
-
-// CHECK-LABEL: @vector_extract_slice_i64
-func.func @vector_extract_slice_i64(%row: index) -> vector<[2]xi64> {
-  // CHECK: %{{.*}} = "arm_sme.intr.read.horiz"(%{{.*}}) <{tile_id = 0 : i32}> : (vector<[2]xi64>, vector<[2]xi1>, i32) -> vector<[2]xi64>
-  %tile = arm_sme.get_tile : vector<[2]x[2]xi64>
-  %slice = vector.extract %tile[%row] : vector<[2]xi64> from vector<[2]x[2]xi64>
-  return %slice : vector<[2]xi64>
-}
-
-// -----
-
-// CHECK-LABEL: @vector_extract_slice_i128
-func.func @vector_extract_slice_i128(%row: index) -> vector<[1]xi128> {
-  // CHECK: %{{.*}} = "arm_sme.intr.read.horiz"(%{{.*}}) <{tile_id = 0 : i32}> : (vector<[1]xi128>, vector<[1]xi1>, i32) -> vector<[1]xi128>
-  %tile = arm_sme.get_tile : vector<[1]x[1]xi128>
-  %slice = vector.extract %tile[%row] : vector<[1]xi128> from vector<[1]x[1]xi128>
-  return %slice : vector<[1]xi128>
-}
-
-// -----
-
-// CHECK-LABEL: @vector_extract_slice_f16
-func.func @vector_extract_slice_f16(%row: index) -> vector<[8]xf16> {
-  // CHECK: %{{.*}} = "arm_sme.intr.read.horiz"(%{{.*}}) <{tile_id = 0 : i32}> : (vector<[8]xf16>, vector<[8]xi1>, i32) -> vector<[8]xf16>
-  %tile = arm_sme.get_tile : vector<[8]x[8]xf16>
-  %slice = vector.extract %tile[%row] : vector<[8]xf16> from vector<[8]x[8]xf16>
-  return %slice : vector<[8]xf16>
-}
-
-// -----
-
-// CHECK-LABEL: @vector_extract_slice_bf16
-func.func @vector_extract_slice_bf16(%row: index) -> vector<[8]xbf16> {
-  // CHECK: %{{.*}} = "arm_sme.intr.read.horiz"(%{{.*}}) <{tile_id = 0 : i32}> : (vector<[8]xbf16>, vector<[8]xi1>, i32) -> vector<[8]xbf16>
-  %tile = arm_sme.get_tile : vector<[8]x[8]xbf16>
-  %slice = vector.extract %tile[%row] : vector<[8]xbf16> from vector<[8]x[8]xbf16>
-  return %slice : vector<[8]xbf16>
-}
-
-// -----
-
-// CHECK-LABEL: @vector_extract_slice_f32
-func.func @vector_extract_slice_f32(%row: index) -> vector<[4]xf32> {
-  // CHECK: %{{.*}} = "arm_sme.intr.read.horiz"(%{{.*}}) <{tile_id = 0 : i32}> : (vector<[4]xf32>, vector<[4]xi1>, i32) -> vector<[4]xf32>
-  %tile = arm_sme.get_tile : vector<[4]x[4]xf32>
-  %slice = vector.extract %tile[%row] : vector<[4]xf32> from vector<[4]x[4]xf32>
-  return %slice : vector<[4]xf32>
-}
-
-// -----
-
-// CHECK-LABEL: @vector_extract_slice_f64
-func.func @vector_extract_slice_f64(%row: index) -> vector<[2]xf64> {
-  // CHECK: %{{.*}} = "arm_sme.intr.read.horiz"(%{{.*}}) <{tile_id = 0 : i32}> : (vector<[2]xf64>, vector<[2]xi1>, i32) -> vector<[2]xf64>
-  %tile = arm_sme.get_tile : vector<[2]x[2]xf64>
-  %slice = vector.extract %tile[%row] : vector<[2]xf64> from vector<[2]x[2]xf64>
-  return %slice : vector<[2]xf64>
-}
-
-// -----
-
-// CHECK-LABEL: @vector_extract_element(
-// CHECK-SAME:                          %[[ROW:.*]]: index,
-// CHECK-SAME:                          %[[COL:.*]]: index)
-func.func @vector_extract_element(%row: index, %col: index) -> i32 {
-  // CHECK-NEXT: %[[PTRUE:.*]] = arith.constant dense<true> : vector<[4]xi1>
-  // CHECK-NEXT: %[[ZERO_VEC:.*]] = arith.constant dense<0> : vector<[4]xi32>
-  // CHECK-NEXT: %[[ROW_I32:.*]] = arith.index_cast %[[ROW]] : index to i32
-  // CHECK-NEXT: %[[SLICE:.*]] = "arm_sme.intr.read.horiz"(%[[ZERO_VEC]], %[[PTRUE]], %[[ROW_I32]]) <{tile_id = 0 : i32}> : (vector<[4]xi32>, vector<[4]xi1>, i32) -> vector<[4]xi32>
-  // CHECK-NEXT: %[[EL:.*]] = vector.extract %[[SLICE]]{{\[}}%[[COL]]] : i32 from vector<[4]xi32>
-  %tile = arm_sme.get_tile : vector<[4]x[4]xi32>
-  %el = vector.extract %tile[%row, %col] : i32 from vector<[4]x[4]xi32>
-  return %el : i32
-}
-
-// -----
-
-// CHECK-LABEL: @vector_extract_element_i8
-func.func @vector_extract_element_i8(%row: index, %col: index) -> i8 {
-  // CHECK: %[[SLICE:.*]] = "arm_sme.intr.read.horiz"(%{{.*}}) <{tile_id = 0 : i32}> : (vector<[16]xi8>, vector<[16]xi1>, i32) -> vector<[16]xi8>
-  // CHECK-NEXT: %{{.*}} = vector.extract %[[SLICE]]{{\[}}%{{.*}}] : i8 from vector<[16]xi8>
-  %tile = arm_sme.get_tile : vector<[16]x[16]xi8>
-  %el = vector.extract %tile[%row, %col] : i8 from vector<[16]x[16]xi8>
-  return %el : i8
-}
-
-// -----
-
-// CHECK-LABEL: @vector_extract_element_i16
-func.func @vector_extract_element_i16(%row: index, %col: index) -> i16 {
-  // CHECK: %[[SLICE:.*]] = "arm_sme.intr.read.horiz"(%{{.*}}) <{tile_id = 0 : i32}> : (vector<[8]xi16>, vector<[8]xi1>, i32) -> vector<[8]xi16>
-  // CHECK-NEXT: %{{.*}} = vector.extract %[[SLICE]]{{\[}}%{{.*}}] : i16 from vector<[8]xi16>
-  %tile = arm_sme.get_tile : vector<[8]x[8]xi16>
-  %el = vector.extract %tile[%row, %col] : i16 from vector<[8]x[8]xi16>
-  return %el : i16
-}
-
-// -----
-
-// CHECK-LABEL: @vector_extract_element_i64
-func.func @vector_extract_element_i64(%row: index, %col: index) -> i64 {
-  // CHECK: %[[SLICE:.*]] = "arm_sme.intr.read.horiz"(%{{.*}}) <{tile_id = 0 : i32}> : (vector<[2]xi64>, vector<[2]xi1>, i32) -> vector<[2]xi64>
-  // CHECK-NEXT: %{{.*}} = vector.extract %[[SLICE]]{{\[}}%{{.*}}] : i64 from vector<[2]xi64>
-  %tile = arm_sme.get_tile : vector<[2]x[2]xi64>
-  %el = vector.extract %tile[%row, %col] : i64 from vector<[2]x[2]xi64>
-  return %el : i64
-}
-
-// -----
-
-// CHECK-LABEL: @vector_extract_element_i128
-func.func @vector_extract_element_i128(%row: index, %col: index) -> i128 {
-  // CHECK: %[[SLICE:.*]] = "arm_sme.intr.read.horiz"(%{{.*}}) <{tile_id = 0 : i32}> : (vector<[1]xi128>, vector<[1]xi1>, i32) -> vector<[1]xi128>
-  // CHECK-NEXT: %{{.*}} = vector.extract %[[SLICE]]{{\[}}%{{.*}}] : i128 from vector<[1]xi128>
-  %tile = arm_sme.get_tile : vector<[1]x[1]xi128>
-  %el = vector.extract %tile[%row, %col] : i128 from vector<[1]x[1]xi128>
-  return %el : i128
-}
-
-// -----
-
-// CHECK-LABEL: @vector_extract_element_f16
-func.func @vector_extract_element_f16(%row: index, %col: index) -> f16 {
-  // CHECK: %[[SLICE:.*]] = "arm_sme.intr.read.horiz"(%{{.*}}) <{tile_id = 0 : i32}> : (vector<[8]xf16>, vector<[8]xi1>, i32) -> vector<[8]xf16>
-  // CHECK-NEXT: %{{.*}} = vector.extract %[[SLICE]]{{\[}}%{{.*}}] : f16 from vector<[8]xf16>
-  %tile = arm_sme.get_tile : vector<[8]x[8]xf16>
-  %el = vector.extract %tile[%row, %col] : f16 from vector<[8]x[8]xf16>
-  return %el : f16
-}
-
-// -----
-
-// CHECK-LABEL: @vector_extract_element_bf16
-func.func @vector_extract_element_bf16(%row: index, %col: index) -> bf16 {
-  // CHECK: %[[SLICE:.*]] = "arm_sme.intr.read.horiz"(%{{.*}}) <{tile_id = 0 : i32}> : (vector<[8]xbf16>, vector<[8]xi1>, i32) -> vector<[8]xbf16>
-  // CHECK-NEXT: %{{.*}} = vector.extract %[[SLICE]]{{\[}}%{{.*}}] : bf16 from vector<[8]xbf16>
-  %tile = arm_sme.get_tile : vector<[8]x[8]xbf16>
-  %el = vector.extract %tile[%row, %col] : bf16 from vector<[8]x[8]xbf16>
-  return %el : bf16
-}
-
-// -----
-
-// CHECK-LABEL: @vector_extract_element_f32
-func.func @vector_extract_element_f32(%row: index, %col: index) -> f32 {
-  // CHECK: %[[SLICE:.*]] = "arm_sme.intr.read.horiz"(%{{.*}}) <{tile_id = 0 : i32}> : (vector<[4]xf32>, vector<[4]xi1>, i32) -> vector<[4]xf32>
-  // CHECK-NEXT: %{{.*}} = vector.extract %[[SLICE]]{{\[}}%{{.*}}] : f32 from vector<[4]xf32>
-  %tile = arm_sme.get_tile : vector<[4]x[4]xf32>
-  %el = vector.extract %tile[%row, %col] : f32 from vector<[4]x[4]xf32>
-  return %el : f32
-}
-
-// -----
-
-// CHECK-LABEL: @vector_extract_element_f64
-func.func @vector_extract_element_f64(%row: index, %col: index) -> f64 {
-  // CHECK: %[[SLICE:.*]] = "arm_sme.intr.read.horiz"(%{{.*}}) <{tile_id = 0 : i32}> : (vector<[2]xf64>, vector<[2]xi1>, i32) -> vector<[2]xf64>
-  // CHECK-NEXT: %{{.*}} = vector.extract %[[SLICE]]{{\[}}%{{.*}}] : f64 from vector<[2]xf64>
-  %tile = arm_sme.get_tile : vector<[2]x[2]xf64>
-  %el = vector.extract %tile[%row, %col] : f64 from vector<[2]x[2]xf64>
-  return %el : f64
-}
diff --git a/mlir/test/Dialect/EmitC/invalid_ops.mlir b/mlir/test/Dialect/EmitC/invalid_ops.mlir
index bbaab0d5b6f3..21ea6a5df91b 100644
--- a/mlir/test/Dialect/EmitC/invalid_ops.mlir
+++ b/mlir/test/Dialect/EmitC/invalid_ops.mlir
@@ -435,3 +435,18 @@ func.func @test_subscript_ptr_type_mismatch(%arg0: !emitc.ptr<f32>, %arg1: index
   %0 = emitc.subscript %arg0[%arg1] : (!emitc.ptr<f32>, index) -> f64
   return
 }
+
+// -----
+
+// expected-error @+1 {{'emitc.global' op cannot have both static and extern specifiers}}
+emitc.global extern static @uninit : i32
+
+// -----
+
+emitc.global @myglobal : !emitc.array<2xf32>
+
+func.func @use_global() {
+  // expected-error @+1 {{'emitc.get_global' op result type 'f32' does not match type '!emitc.array<2xf32>' of the global @myglobal}}
+  %0 = emitc.get_global @myglobal : f32
+  return
+}
diff --git a/mlir/test/Dialect/EmitC/ops.mlir b/mlir/test/Dialect/EmitC/ops.mlir
index ace3670426af..05510e6dddbf 100644
--- a/mlir/test/Dialect/EmitC/ops.mlir
+++ b/mlir/test/Dialect/EmitC/ops.mlir
@@ -231,3 +231,17 @@ emitc.verbatim "#endif  // __cplusplus"
 
 emitc.verbatim "typedef int32_t i32;"
 emitc.verbatim "typedef float f32;"
+
+
+emitc.global @uninit : i32
+emitc.global @myglobal_int : i32 = 4
+emitc.global extern @external_linkage : i32
+emitc.global static @internal_linkage : i32
+emitc.global @myglobal : !emitc.array<2xf32> = dense<4.000000e+00>
+emitc.global const @myconstant : !emitc.array<2xi16> = dense<2>
+
+func.func @use_global(%i: index) -> f32 {
+  %0 = emitc.get_global @myglobal : !emitc.array<2xf32>
+  %1 = emitc.subscript %0[%i] : (!emitc.array<2xf32>, index) -> f32
+  return %1 : f32
+}
diff --git a/mlir/test/Dialect/LLVMIR/invalid.mlir b/mlir/test/Dialect/LLVMIR/invalid.mlir
index de1ab9db8e8d..0914f0023210 100644
--- a/mlir/test/Dialect/LLVMIR/invalid.mlir
+++ b/mlir/test/Dialect/LLVMIR/invalid.mlir
@@ -1221,17 +1221,17 @@ func.func @extract_scalable_from_fixed_length_vector(%arg0 : vector<16xf32>) {
 
 // -----
 
-func.func @experimental_vector_interleave2_bad_type0(%vec1: vector<[2]xf16>, %vec2 : vector<[4]xf16>) {
+func.func @vector_interleave2_bad_type0(%vec1: vector<[2]xf16>, %vec2 : vector<[4]xf16>) {
   // expected-error@+1 {{op failed to verify that all of {vec1, vec2} have same type}}
-  %0 = "llvm.intr.experimental.vector.interleave2"(%vec1, %vec2) : (vector<[2]xf16>, vector<[4]xf16>) -> vector<[8]xf16>
+  %0 = "llvm.intr.vector.interleave2"(%vec1, %vec2) : (vector<[2]xf16>, vector<[4]xf16>) -> vector<[8]xf16>
   return
 }
 
 // -----
 
-func.func @experimental_vector_interleave2_bad_type1(%vec1: vector<[2]xf16>, %vec2 : vector<[2]xf16>) {
+func.func @vector_interleave2_bad_type1(%vec1: vector<[2]xf16>, %vec2 : vector<[2]xf16>) {
   // expected-error@+1 {{op failed to verify that result has twice as many elements as 'vec1'}}
-  %0 = "llvm.intr.experimental.vector.interleave2"(%vec1, %vec2) : (vector<[2]xf16>, vector<[2]xf16>) -> vector<[8]xf16>
+  %0 = "llvm.intr.vector.interleave2"(%vec1, %vec2) : (vector<[2]xf16>, vector<[2]xf16>) -> vector<[8]xf16>
   return
 }
 
@@ -1239,9 +1239,9 @@ func.func @experimental_vector_interleave2_bad_type1(%vec1: vector<[2]xf16>, %ve
 
 /// result vector type is not scalable.
 
-func.func @experimental_vector_interleave2_bad_type2(%vec1: vector<[2]xf16>, %vec2 : vector<[2]xf16>) {
+func.func @vector_interleave2_bad_type2(%vec1: vector<[2]xf16>, %vec2 : vector<[2]xf16>) {
   // expected-error@+1 {{op failed to verify that result has twice as many elements as 'vec1'}}
-  %0 = "llvm.intr.experimental.vector.interleave2"(%vec1, %vec2) : (vector<[2]xf16>, vector<[2]xf16>) -> vector<4xf16>
+  %0 = "llvm.intr.vector.interleave2"(%vec1, %vec2) : (vector<[2]xf16>, vector<[2]xf16>) -> vector<4xf16>
   return
 }
 
@@ -1250,9 +1250,9 @@ func.func @experimental_vector_interleave2_bad_type2(%vec1: vector<[2]xf16>, %ve
 
 /// element type doesn't match.
 
-func.func @experimental_vector_interleave2_bad_type3(%vec1: vector<[2]xf16>, %vec2 : vector<[2]xf16>) {
+func.func @vector_interleave2_bad_type3(%vec1: vector<[2]xf16>, %vec2 : vector<[2]xf16>) {
   // expected-error@+1 {{op failed to verify that result has twice as many elements as 'vec1'}}
-  %0 = "llvm.intr.experimental.vector.interleave2"(%vec1, %vec2) : (vector<[2]xf16>, vector<[2]xf16>) -> vector<[4]xf32>
+  %0 = "llvm.intr.vector.interleave2"(%vec1, %vec2) : (vector<[2]xf16>, vector<[2]xf16>) -> vector<[4]xf32>
   return
 }
 
diff --git a/mlir/test/Dialect/LLVMIR/legalize-for-export.mlir b/mlir/test/Dialect/LLVMIR/legalize-for-export.mlir
index 37720e98d92a..b1b06740f194 100644
--- a/mlir/test/Dialect/LLVMIR/legalize-for-export.mlir
+++ b/mlir/test/Dialect/LLVMIR/legalize-for-export.mlir
@@ -32,14 +32,17 @@ llvm.func @repeated_successor_no_args(%arg0: i1) {
 
 // CHECK: @repeated_successor_openmp
 llvm.func @repeated_successor_openmp(%arg0: i64, %arg1: i64, %arg2: i64, %arg3: i1) {
-  omp.wsloop for (%arg4) : i64 = (%arg0) to (%arg1) step (%arg2) {
-    // CHECK: llvm.cond_br %{{.*}}, ^[[BB1:.*]]({{.*}}), ^[[BB2:.*]]({{.*}})
-    llvm.cond_br %arg3, ^bb1(%arg0 : i64), ^bb1(%arg1 : i64)
-  // CHECK: ^[[BB1]]
-  ^bb1(%0: i64):  // 2 preds: ^bb0, ^bb0
-    omp.yield
-  // CHECK: ^[[BB2]](%[[ARG:.*]]: i64):
-  // CHECK:  llvm.br ^[[BB1]](%[[ARG]] : i64)
+  omp.wsloop {
+    omp.loop_nest (%arg4) : i64 = (%arg0) to (%arg1) step (%arg2) {
+      // CHECK: llvm.cond_br %{{.*}}, ^[[BB1:.*]]({{.*}}), ^[[BB2:.*]]({{.*}})
+      llvm.cond_br %arg3, ^bb1(%arg0 : i64), ^bb1(%arg1 : i64)
+    // CHECK: ^[[BB1]]
+    ^bb1(%0: i64):  // 2 preds: ^bb0, ^bb0
+      omp.yield
+    // CHECK: ^[[BB2]](%[[ARG:.*]]: i64):
+    // CHECK:  llvm.br ^[[BB1]](%[[ARG]] : i64)
+    }
+    omp.terminator
   }
   llvm.return
 }
diff --git a/mlir/test/Dialect/LLVMIR/mem2reg.mlir b/mlir/test/Dialect/LLVMIR/mem2reg.mlir
index 644d30f9f9f1..38c836c139da 100644
--- a/mlir/test/Dialect/LLVMIR/mem2reg.mlir
+++ b/mlir/test/Dialect/LLVMIR/mem2reg.mlir
@@ -856,28 +856,6 @@ llvm.func @stores_with_different_types(%arg0: i64, %arg1: f64, %cond: i1) -> f64
 
 // -----
 
-// Verifies that stores with smaller bitsize inputs are not replaced. A trivial
-// implementation will be incorrect due to endianness considerations.
-
-// CHECK-LABEL: @stores_with_different_type_sizes
-llvm.func @stores_with_different_type_sizes(%arg0: i64, %arg1: f32, %cond: i1) -> f64 {
-  %0 = llvm.mlir.constant(1 : i32) : i32
-  // CHECK: llvm.alloca
-  %1 = llvm.alloca %0 x i64 {alignment = 4 : i64} : (i32) -> !llvm.ptr
-  llvm.cond_br %cond, ^bb1, ^bb2
-^bb1:
-  llvm.store %arg0, %1 {alignment = 4 : i64} : i64, !llvm.ptr
-  llvm.br ^bb3
-^bb2:
-  llvm.store %arg1, %1 {alignment = 4 : i64} : f32, !llvm.ptr
-  llvm.br ^bb3
-^bb3:
-  %2 = llvm.load %1 {alignment = 4 : i64} : !llvm.ptr -> f64
-  llvm.return %2 : f64
-}
-
-// -----
-
 // CHECK-LABEL: @load_smaller_int
 llvm.func @load_smaller_int() -> i16 {
   %0 = llvm.mlir.constant(1 : i32) : i32
@@ -1047,3 +1025,135 @@ llvm.func @scalable_llvm_vector() -> i16 {
   %2 = llvm.load %1 : !llvm.ptr -> i16
   llvm.return %2 : i16
 }
+
+// -----
+
+// CHECK-LABEL: @smaller_store_forwarding
+// CHECK-SAME: %[[ARG:.+]]: i16
+llvm.func @smaller_store_forwarding(%arg : i16) {
+  %0 = llvm.mlir.constant(1 : i32) : i32
+  // CHECK-NOT: llvm.alloca
+  // CHECK: %[[UNDEF:.+]] = llvm.mlir.undef : i32
+  %1 = llvm.alloca %0 x i32 : (i32) -> !llvm.ptr
+
+  // CHECK: %[[ZEXT:.+]] = llvm.zext %[[ARG]] : i16 to i32
+  // CHECK: %[[MASK:.+]] = llvm.mlir.constant(-65536 : i32) : i32
+  // CHECK: %[[MASKED:.+]] = llvm.and %[[UNDEF]], %[[MASK]]
+  // CHECK: %[[NEW_DEF:.+]] = llvm.or %[[MASKED]], %[[ZEXT]]
+  llvm.store %arg, %1 : i16, !llvm.ptr
+  llvm.return
+}
+
+// -----
+
+module attributes { dlti.dl_spec = #dlti.dl_spec<
+  #dlti.dl_entry<"dlti.endianness", "big">
+>} {
+  // CHECK-LABEL: @smaller_store_forwarding_big_endian
+  // CHECK-SAME: %[[ARG:.+]]: i16
+  llvm.func @smaller_store_forwarding_big_endian(%arg : i16) {
+    %0 = llvm.mlir.constant(1 : i32) : i32
+    // CHECK-NOT: llvm.alloca
+    // CHECK: %[[UNDEF:.+]] = llvm.mlir.undef : i32
+    %1 = llvm.alloca %0 x i32 : (i32) -> !llvm.ptr
+
+    // CHECK: %[[ZEXT:.+]] = llvm.zext %[[ARG]] : i16 to i32
+    // CHECK: %[[SHIFT_WIDTH:.+]] = llvm.mlir.constant(16 : i32) : i32
+    // CHECK: %[[SHIFTED:.+]] = llvm.shl %[[ZEXT]], %[[SHIFT_WIDTH]]
+    // CHECK: %[[MASK:.+]] = llvm.mlir.constant(65535 : i32) : i32
+    // CHECK: %[[MASKED:.+]] = llvm.and %[[UNDEF]], %[[MASK]]
+    // CHECK: %[[NEW_DEF:.+]] = llvm.or %[[MASKED]], %[[SHIFTED]]
+    llvm.store %arg, %1 : i16, !llvm.ptr
+    llvm.return
+  }
+}
+
+// -----
+
+// CHECK-LABEL: @smaller_store_forwarding_type_mix
+// CHECK-SAME: %[[ARG:.+]]: vector<1xi8>
+llvm.func @smaller_store_forwarding_type_mix(%arg : vector<1xi8>) {
+  %0 = llvm.mlir.constant(1 : i32) : i32
+  // CHECK-NOT: llvm.alloca
+  // CHECK: %[[UNDEF:.+]] = llvm.mlir.undef : f32
+  %1 = llvm.alloca %0 x f32 : (i32) -> !llvm.ptr
+
+  // CHECK: %[[CASTED_DEF:.+]] = llvm.bitcast %[[UNDEF]] : f32 to i32
+  // CHECK: %[[CASTED_ARG:.+]] = llvm.bitcast %[[ARG]] : vector<1xi8> to i8
+  // CHECK: %[[ZEXT:.+]] = llvm.zext %[[CASTED_ARG]] : i8 to i32
+  // CHECK: %[[MASK:.+]] = llvm.mlir.constant(-256 : i32) : i32
+  // CHECK: %[[MASKED:.+]] = llvm.and %[[CASTED_DEF]], %[[MASK]]
+  // CHECK: %[[NEW_DEF:.+]] = llvm.or %[[MASKED]], %[[ZEXT]]
+  // CHECK: %[[CASTED_NEW_DEF:.+]] = llvm.bitcast %[[NEW_DEF]] : i32 to f32
+  llvm.store %arg, %1 : vector<1xi8>, !llvm.ptr
+  llvm.return
+}
+
+// -----
+
+module attributes { dlti.dl_spec = #dlti.dl_spec<
+  #dlti.dl_entry<"dlti.endianness", "big">
+>} {
+  // CHECK-LABEL: @smaller_store_forwarding_type_mix
+  // CHECK-SAME: %[[ARG:.+]]: vector<1xi8>
+  llvm.func @smaller_store_forwarding_type_mix(%arg : vector<1xi8>) {
+    %0 = llvm.mlir.constant(1 : i32) : i32
+    // CHECK-NOT: llvm.alloca
+    // CHECK: %[[UNDEF:.+]] = llvm.mlir.undef : f32
+    %1 = llvm.alloca %0 x f32 : (i32) -> !llvm.ptr
+
+    // CHECK: %[[CASTED_DEF:.+]] = llvm.bitcast %[[UNDEF]] : f32 to i32
+    // CHECK: %[[CASTED_ARG:.+]] = llvm.bitcast %[[ARG]] : vector<1xi8> to i8
+    // CHECK: %[[ZEXT:.+]] = llvm.zext %[[CASTED_ARG]] : i8 to i32
+    // CHECK: %[[SHIFT_WIDTH:.+]] = llvm.mlir.constant(24 : i32) : i32
+    // CHECK: %[[SHIFTED:.+]] = llvm.shl %[[ZEXT]], %[[SHIFT_WIDTH]]
+    // CHECK: %[[MASK:.+]] = llvm.mlir.constant(16777215 : i32) : i32
+    // CHECK: %[[MASKED:.+]] = llvm.and %[[CASTED_DEF]], %[[MASK]]
+    // CHECK: %[[NEW_DEF:.+]] = llvm.or %[[MASKED]], %[[SHIFTED]]
+    // CHECK: %[[CASTED_NEW_DEF:.+]] = llvm.bitcast %[[NEW_DEF]] : i32 to f32
+    llvm.store %arg, %1 : vector<1xi8>, !llvm.ptr
+    llvm.return
+  }
+}
+
+// -----
+
+// CHECK-LABEL: @stores_with_different_types_branches
+// CHECK-SAME: %[[ARG0:.+]]: i64
+// CHECK-SAME: %[[ARG1:.+]]: f32
+llvm.func @stores_with_different_types_branches(%arg0: i64, %arg1: f32, %cond: i1) -> f64 {
+  %0 = llvm.mlir.constant(1 : i32) : i32
+  // CHECK-NOT: llvm.alloca
+  // CHECK: %[[UNDEF:.+]] = llvm.mlir.undef : i64
+  %1 = llvm.alloca %0 x i64 {alignment = 4 : i64} : (i32) -> !llvm.ptr
+  llvm.cond_br %cond, ^bb1, ^bb2
+^bb1:
+  llvm.store %arg0, %1 {alignment = 4 : i64} : i64, !llvm.ptr
+  // CHECK: llvm.br ^[[BB3:.+]](%[[ARG0]] : i64)
+  llvm.br ^bb3
+^bb2:
+  llvm.store %arg1, %1 {alignment = 4 : i64} : f32, !llvm.ptr
+  // CHECK: %[[CAST:.+]] = llvm.bitcast %[[ARG1]] : f32 to i32
+  // CHECK: %[[ZEXT:.+]] = llvm.zext %[[CAST]] : i32 to i64
+  // CHECK: %[[MASK:.+]] = llvm.mlir.constant(-4294967296 : i64) : i64
+  // CHECK: %[[MASKED:.+]] = llvm.and %[[UNDEF]], %[[MASK]]
+  // CHECK: %[[NEW_DEF:.+]] = llvm.or %[[MASKED]], %[[ZEXT]]
+  // CHECK: llvm.br ^[[BB3]](%[[NEW_DEF]] : i64)
+  llvm.br ^bb3
+^bb3:
+  %2 = llvm.load %1 {alignment = 4 : i64} : !llvm.ptr -> f64
+  llvm.return %2 : f64
+}
+
+// -----
+
+// Verifiy that mem2reg does not touch stores with undefined semantics.
+
+// CHECK-LABEL: @store_out_of_bounds
+llvm.func @store_out_of_bounds(%arg : i64) {
+  %0 = llvm.mlir.constant(1 : i32) : i32
+  // CHECK: llvm.alloca
+  %1 = llvm.alloca %0 x i32 : (i32) -> !llvm.ptr
+  llvm.store %arg, %1 : i64, !llvm.ptr
+  llvm.return
+}
diff --git a/mlir/test/Dialect/LLVMIR/roundtrip.mlir b/mlir/test/Dialect/LLVMIR/roundtrip.mlir
index 31acf2b95e46..3b94db389f54 100644
--- a/mlir/test/Dialect/LLVMIR/roundtrip.mlir
+++ b/mlir/test/Dialect/LLVMIR/roundtrip.mlir
@@ -342,10 +342,10 @@ func.func @mixed_vect(%arg0: vector<8xf32>, %arg1: vector<4xf32>, %arg2: vector<
   return
 }
 
-// CHECK-LABEL: @experimental_vector_interleave2
-func.func @experimental_vector_interleave2(%vec1: vector<[4]xf16>, %vec2 : vector<[4]xf16>) {
-  // CHECK: = "llvm.intr.experimental.vector.interleave2"({{.*}}) : (vector<[4]xf16>, vector<[4]xf16>) -> vector<[8]xf16>
-  %0 = "llvm.intr.experimental.vector.interleave2"(%vec1, %vec2) : (vector<[4]xf16>, vector<[4]xf16>) -> vector<[8]xf16>
+// CHECK-LABEL: @vector_interleave2
+func.func @vector_interleave2(%vec1: vector<[4]xf16>, %vec2 : vector<[4]xf16>) {
+  // CHECK: = "llvm.intr.vector.interleave2"({{.*}}) : (vector<[4]xf16>, vector<[4]xf16>) -> vector<[8]xf16>
+  %0 = "llvm.intr.vector.interleave2"(%vec1, %vec2) : (vector<[4]xf16>, vector<[4]xf16>) -> vector<[8]xf16>
   return
 }
 
diff --git a/mlir/test/Dialect/Linalg/generalize-named-ops.mlir b/mlir/test/Dialect/Linalg/generalize-named-ops.mlir
index e852824cdb73..667ea3c18c8a 100644
--- a/mlir/test/Dialect/Linalg/generalize-named-ops.mlir
+++ b/mlir/test/Dialect/Linalg/generalize-named-ops.mlir
@@ -565,6 +565,155 @@ func.func @generalize_negf(%arg: memref<7x14x21xf32>, %out: memref<7x14x21xf32>)
 
 // -----
 
+func.func @generalize_reciprocal(%arg: memref<7x14x21xf32>, %out: memref<7x14x21xf32>) {
+  linalg.reciprocal ins(%arg : memref<7x14x21xf32>) outs(%out : memref<7x14x21xf32>)
+  return
+}
+
+// CHECK: #[[MAP:.+]] = affine_map<(d0, d1, d2) -> (d0, d1, d2)>
+
+// CHECK: func @generalize_reciprocal
+// CHECK-SAME: (%[[ARG:.+]]: memref<7x14x21xf32>, %[[OUT:.+]]: memref<7x14x21xf32>)
+
+// CHECK: %[[one:.+]] = arith.constant 1.000000e+00 : f32
+
+// CHECK: linalg.generic
+// CHECK-SAME: indexing_maps = [#[[MAP]], #[[MAP]]]
+// CHECK-SAME: iterator_types = ["parallel", "parallel", "parallel"]}
+// CHECK-SAME:  ins(%[[LHS]] : memref<7x14x21xf32>) outs(%[[OUT]] : memref<7x14x21xf32>)
+
+// CHECK:         ^{{.+}}(%[[BBARG0:.+]]: f32, %[[BBARG1:.+]]: f32)
+// CHECK-NEXT:      %[[reciprocal:.+]] = arith.divf %[[one]], %[[BBARG0]] : f32
+// CHECK-NEXT:      linalg.yield %[[reciprocal]] : f32
+
+// -----
+
+func.func @generalize_round(%arg: memref<7x14x21xf32>, %out: memref<7x14x21xf32>) {
+  linalg.round ins(%arg : memref<7x14x21xf32>) outs(%out : memref<7x14x21xf32>)
+  return
+}
+
+// CHECK: #[[MAP:.+]] = affine_map<(d0, d1, d2) -> (d0, d1, d2)>
+
+// CHECK: func @generalize_round
+// CHECK-SAME: (%[[ARG:.+]]: memref<7x14x21xf32>, %[[OUT:.+]]: memref<7x14x21xf32>)
+
+// CHECK: linalg.generic
+// CHECK-SAME: indexing_maps = [#[[MAP]], #[[MAP]]]
+// CHECK-SAME: iterator_types = ["parallel", "parallel", "parallel"]}
+// CHECK-SAME:  ins(%[[LHS]] : memref<7x14x21xf32>) outs(%[[OUT]] : memref<7x14x21xf32>)
+
+// CHECK:         ^{{.+}}(%[[BBARG0:.+]]: f32, %[[BBARG1:.+]]: f32)
+// CHECK-NEXT:      %[[round:.+]] = math.round %[[BBARG0]] : f32
+// CHECK-NEXT:      linalg.yield %[[round]] : f32
+
+// -----
+
+func.func @generalize_sqrt(%arg: memref<7x14x21xf32>, %out: memref<7x14x21xf32>) {
+  linalg.sqrt ins(%arg : memref<7x14x21xf32>) outs(%out : memref<7x14x21xf32>)
+  return
+}
+
+// CHECK: #[[MAP:.+]] = affine_map<(d0, d1, d2) -> (d0, d1, d2)>
+
+// CHECK: func @generalize_sqrt
+// CHECK-SAME: (%[[ARG:.+]]: memref<7x14x21xf32>, %[[OUT:.+]]: memref<7x14x21xf32>)
+
+// CHECK: linalg.generic
+// CHECK-SAME: indexing_maps = [#[[MAP]], #[[MAP]]]
+// CHECK-SAME: iterator_types = ["parallel", "parallel", "parallel"]}
+// CHECK-SAME:  ins(%[[LHS]] : memref<7x14x21xf32>) outs(%[[OUT]] : memref<7x14x21xf32>)
+
+// CHECK:         ^{{.+}}(%[[BBARG0:.+]]: f32, %[[BBARG1:.+]]: f32)
+// CHECK-NEXT:      %[[sqrt:.+]] = math.sqrt %[[BBARG0]] : f32
+// CHECK-NEXT:      linalg.yield %[[sqrt]] : f32
+
+// -----
+
+func.func @generalize_rsqrt(%arg: memref<7x14x21xf32>, %out: memref<7x14x21xf32>) {
+  linalg.rsqrt ins(%arg : memref<7x14x21xf32>) outs(%out : memref<7x14x21xf32>)
+  return
+}
+
+// CHECK: #[[MAP:.+]] = affine_map<(d0, d1, d2) -> (d0, d1, d2)>
+
+// CHECK: func @generalize_rsqrt
+// CHECK-SAME: (%[[ARG:.+]]: memref<7x14x21xf32>, %[[OUT:.+]]: memref<7x14x21xf32>)
+
+// CHECK: linalg.generic
+// CHECK-SAME: indexing_maps = [#[[MAP]], #[[MAP]]]
+// CHECK-SAME: iterator_types = ["parallel", "parallel", "parallel"]}
+// CHECK-SAME:  ins(%[[LHS]] : memref<7x14x21xf32>) outs(%[[OUT]] : memref<7x14x21xf32>)
+
+// CHECK:         ^{{.+}}(%[[BBARG0:.+]]: f32, %[[BBARG1:.+]]: f32)
+// CHECK-NEXT:      %[[rsqrt:.+]] = math.rsqrt %[[BBARG0]] : f32
+// CHECK-NEXT:      linalg.yield %[[rsqrt]] : f32
+
+// -----
+
+func.func @generalize_square(%arg: memref<7x14x21xf32>, %out: memref<7x14x21xf32>) {
+  linalg.square ins(%arg : memref<7x14x21xf32>) outs(%out : memref<7x14x21xf32>)
+  return
+}
+
+// CHECK: #[[MAP:.+]] = affine_map<(d0, d1, d2) -> (d0, d1, d2)>
+
+// CHECK: func @generalize_square
+// CHECK-SAME: (%[[ARG:.+]]: memref<7x14x21xf32>, %[[OUT:.+]]: memref<7x14x21xf32>)
+
+// CHECK: linalg.generic
+// CHECK-SAME: indexing_maps = [#[[MAP]], #[[MAP]]]
+// CHECK-SAME: iterator_types = ["parallel", "parallel", "parallel"]}
+// CHECK-SAME:  ins(%[[LHS]] : memref<7x14x21xf32>) outs(%[[OUT]] : memref<7x14x21xf32>)
+
+// CHECK:         ^{{.+}}(%[[BBARG0:.+]]: f32, %[[BBARG1:.+]]: f32)
+// CHECK-NEXT:      %[[square:.+]] = arith.mulf %[[BBARG0]], %[[BBARG0]] : f32
+// CHECK-NEXT:      linalg.yield %[[square]] : f32
+
+// -----
+
+func.func @generalize_tanh(%arg: memref<7x14x21xf32>, %out: memref<7x14x21xf32>) {
+  linalg.tanh ins(%arg : memref<7x14x21xf32>) outs(%out : memref<7x14x21xf32>)
+  return
+}
+
+// CHECK: #[[MAP:.+]] = affine_map<(d0, d1, d2) -> (d0, d1, d2)>
+
+// CHECK: func @generalize_tanh
+// CHECK-SAME: (%[[ARG:.+]]: memref<7x14x21xf32>, %[[OUT:.+]]: memref<7x14x21xf32>)
+
+// CHECK: linalg.generic
+// CHECK-SAME: indexing_maps = [#[[MAP]], #[[MAP]]]
+// CHECK-SAME: iterator_types = ["parallel", "parallel", "parallel"]}
+// CHECK-SAME:  ins(%[[LHS]] : memref<7x14x21xf32>) outs(%[[OUT]] : memref<7x14x21xf32>)
+
+// CHECK:         ^{{.+}}(%[[BBARG0:.+]]: f32, %[[BBARG1:.+]]: f32)
+// CHECK-NEXT:      %[[tanh:.+]] = math.tanh %[[BBARG0]] : f32
+// CHECK-NEXT:      linalg.yield %[[tanh]] : f32
+
+// -----
+
+func.func @generalize_erf(%arg: memref<7x14x21xf32>, %out: memref<7x14x21xf32>) {
+  linalg.erf ins(%arg : memref<7x14x21xf32>) outs(%out : memref<7x14x21xf32>)
+  return
+}
+
+// CHECK: #[[MAP:.+]] = affine_map<(d0, d1, d2) -> (d0, d1, d2)>
+
+// CHECK: func @generalize_erf
+// CHECK-SAME: (%[[ARG:.+]]: memref<7x14x21xf32>, %[[OUT:.+]]: memref<7x14x21xf32>)
+
+// CHECK: linalg.generic
+// CHECK-SAME: indexing_maps = [#[[MAP]], #[[MAP]]]
+// CHECK-SAME: iterator_types = ["parallel", "parallel", "parallel"]}
+// CHECK-SAME:  ins(%[[LHS]] : memref<7x14x21xf32>) outs(%[[OUT]] : memref<7x14x21xf32>)
+
+// CHECK:         ^{{.+}}(%[[BBARG0:.+]]: f32, %[[BBARG1:.+]]: f32)
+// CHECK-NEXT:      %[[erf:.+]] = math.erf %[[BBARG0]] : f32
+// CHECK-NEXT:      linalg.yield %[[erf]] : f32
+
+// -----
+
 func.func @generalize_max(%lhs: memref<7x14x21xf32>, %rhs: memref<7x14x21xf32>,
                           %out: memref<7x14x21xf32>) {
   linalg.max ins(%lhs, %rhs : memref<7x14x21xf32>, memref<7x14x21xf32>)
@@ -590,6 +739,58 @@ func.func @generalize_max(%lhs: memref<7x14x21xf32>, %rhs: memref<7x14x21xf32>,
 
 // -----
 
+func.func @generalize_min(%lhs: memref<7x14x21xf32>, %rhs: memref<7x14x21xf32>,
+                          %out: memref<7x14x21xf32>) {
+  linalg.min ins(%lhs, %rhs : memref<7x14x21xf32>, memref<7x14x21xf32>)
+             outs(%out : memref<7x14x21xf32>)
+  return
+}
+
+// CHECK: #[[MAP:.+]] = affine_map<(d0, d1, d2) -> (d0, d1, d2)>
+
+// CHECK: func @generalize_min
+// CHECK-SAME: (%[[LHS:.+]]: memref<7x14x21xf32>, %[[RHS:.+]]: memref<7x14x21xf32>,
+// CHECK-SAME:  %[[OUT:.+]]: memref<7x14x21xf32>)
+
+// CHECK: linalg.generic
+// CHECK-SAME: indexing_maps = [#[[MAP]], #[[MAP]], #[[MAP]]]
+// CHECK-SAME: iterator_types = ["parallel", "parallel", "parallel"]}
+// CHECK-SAME:  ins(%[[LHS]], %[[RHS]] : memref<7x14x21xf32>, memref<7x14x21xf32>)
+// CHECK-SAME: outs(%[[OUT]] : memref<7x14x21xf32>)
+
+// CHECK:         ^{{.+}}(%[[BBARG0:.+]]: f32, %[[BBARG1:.+]]: f32, %[[BBARG2:.+]]: f32)
+// CHECK-NEXT:      %[[min:.+]] = arith.minimumf %[[BBARG0]], %[[BBARG1]] : f32
+// CHECK-NEXT:      linalg.yield %[[min]] : f32
+
+
+// -----
+
+func.func @generalize_powf(%lhs: memref<7x14x21xf32>, %rhs: memref<7x14x21xf32>,
+                          %out: memref<7x14x21xf32>) {
+  linalg.powf ins(%lhs, %rhs : memref<7x14x21xf32>, memref<7x14x21xf32>)
+             outs(%out : memref<7x14x21xf32>)
+  return
+}
+
+// CHECK: #[[MAP:.+]] = affine_map<(d0, d1, d2) -> (d0, d1, d2)>
+
+// CHECK: func @generalize_powf
+// CHECK-SAME: (%[[LHS:.+]]: memref<7x14x21xf32>, %[[RHS:.+]]: memref<7x14x21xf32>,
+// CHECK-SAME:  %[[OUT:.+]]: memref<7x14x21xf32>)
+
+// CHECK: linalg.generic
+// CHECK-SAME: indexing_maps = [#[[MAP]], #[[MAP]], #[[MAP]]]
+// CHECK-SAME: iterator_types = ["parallel", "parallel", "parallel"]}
+// CHECK-SAME:  ins(%[[LHS]], %[[RHS]] : memref<7x14x21xf32>, memref<7x14x21xf32>)
+// CHECK-SAME: outs(%[[OUT]] : memref<7x14x21xf32>)
+
+// CHECK:         ^{{.+}}(%[[BBARG0:.+]]: f32, %[[BBARG1:.+]]: f32, %[[BBARG2:.+]]: f32)
+// CHECK-NEXT:      %[[powf:.+]] = math.powf %[[BBARG0]], %[[BBARG1]] : f32
+// CHECK-NEXT:      linalg.yield %[[powf]] : f32
+
+
+// -----
+
 
 // CHECK-LABEL: func @fill_tensor
 func.func @fill_tensor(%f: f32, %v: vector<2x4xf32>) -> (tensor<f32>, tensor<vector<2x4xf32>>) {
diff --git a/mlir/test/Dialect/Linalg/match-ops-interpreter.mlir b/mlir/test/Dialect/Linalg/match-ops-interpreter.mlir
index 24c7bdd9e105..4bfed475d44f 100644
--- a/mlir/test/Dialect/Linalg/match-ops-interpreter.mlir
+++ b/mlir/test/Dialect/Linalg/match-ops-interpreter.mlir
@@ -1062,6 +1062,28 @@ module attributes { transform.target_tag = "start_here" } {
     return %result : tensor<10x18x15xf64>
   }
 
+  func.func @convolution_depthwise(%input: tensor<1x10x196x48xf32>, %filter: tensor<1x4x48xf32>) -> tensor<1x10x191x48xf32> {
+    %cst = arith.constant 0.0 : f32 
+    %empty = tensor.empty() : tensor<1x10x191x48xf32>
+    %fill = linalg.fill ins(%cst : f32) outs(%empty : tensor<1x10x191x48xf32>) -> tensor<1x10x191x48xf32>
+    // expected-remark @below {{convolution}}
+    // expected-remark @below {{batch dims 0}}
+    // expected-remark @below {{output image dims 1 : i64, 2 : i64}}
+    // expected-remark @below {{output channel dims}}
+    // expected-remark @below {{filter loop dims 4 : i64, 5 : i64}}
+    // expected-remark @below {{input channel dims}}
+    // expected-remark @below {{depth dims 3}}
+    // expected-remark @below {{strides 1 : i64, 1 : i64}}
+    // expected-remark @below {{dilations 1 : i64, 1 : i64}}
+    %result = linalg.depthwise_conv_2d_nhwc_hwc {
+      dilations = dense<1> : tensor<2xi64>,
+      strides = dense<1> : tensor<2xi64>}
+      ins(%input, %filter : tensor<1x10x196x48xf32>, tensor<1x4x48xf32>)
+      outs(%fill : tensor<1x10x191x48xf32>) -> tensor<1x10x191x48xf32>
+
+    return %result : tensor<1x10x191x48xf32>
+  }
+
   func.func @convolution_multi_channel(%input: tensor<2x34x68x16xf32>, %filter: tensor<8x2x3x5x16x16xf32>) -> tensor<8x32x32x16xf32> {
     %cst = arith.constant 0.0 : f32
     %empty = tensor.empty() : tensor<8x32x32x16xf32>
diff --git a/mlir/test/Dialect/Linalg/named-ops-fail.mlir b/mlir/test/Dialect/Linalg/named-ops-fail.mlir
index c351e139a97e..e92a77aa7ad0 100644
--- a/mlir/test/Dialect/Linalg/named-ops-fail.mlir
+++ b/mlir/test/Dialect/Linalg/named-ops-fail.mlir
@@ -176,6 +176,118 @@ func.func @negf_broadcast(%arg: memref<8x16xf32>, %out: memref<4x8x16xf32>) {
 
 // -----
 
+func.func @reciprocal_type_cast(%arg: memref<4x8x16xf16>, %out: memref<4x8x16xf32>) {
+  // CHECK: operand 1 ('f16') doesn't match the element type of the enclosing linalg.generic op ('f32')
+  linalg.reciprocal ins(%arg : memref<4x8x16xf16>) outs(%out: memref<4x8x16xf32>)
+  return
+}
+
+// -----
+
+func.func @reciprocal_broadcast(%arg: memref<8x16xf32>, %out: memref<4x8x16xf32>) {
+  // CHECK: op expected operand rank (2) to match the result rank of indexing_map #0 (3)
+  linalg.reciprocal ins(%arg : memref<8x16xf32>) outs(%out: memref<4x8x16xf32>)
+  return
+}
+
+// -----
+
+func.func @round_type_cast(%arg: memref<4x8x16xf16>, %out: memref<4x8x16xf32>) {
+  // CHECK: operand 1 ('f16') doesn't match the element type of the enclosing linalg.generic op ('f32')
+  linalg.round ins(%arg : memref<4x8x16xf16>) outs(%out: memref<4x8x16xf32>)
+  return
+}
+
+// -----
+
+func.func @round_broadcast(%arg: memref<8x16xf32>, %out: memref<4x8x16xf32>) {
+  // CHECK: op expected operand rank (2) to match the result rank of indexing_map #0 (3)
+  linalg.round ins(%arg : memref<8x16xf32>) outs(%out: memref<4x8x16xf32>)
+  return
+}
+
+// -----
+
+func.func @sqrt_type_cast(%arg: memref<4x8x16xf16>, %out: memref<4x8x16xf32>) {
+  // CHECK: operand 1 ('f16') doesn't match the element type of the enclosing linalg.generic op ('f32')
+  linalg.sqrt ins(%arg : memref<4x8x16xf16>) outs(%out: memref<4x8x16xf32>)
+  return
+}
+
+// -----
+
+func.func @sqrt_broadcast(%arg: memref<8x16xf32>, %out: memref<4x8x16xf32>) {
+  // CHECK: op expected operand rank (2) to match the result rank of indexing_map #0 (3)
+  linalg.sqrt ins(%arg : memref<8x16xf32>) outs(%out: memref<4x8x16xf32>)
+  return
+}
+
+// -----
+
+func.func @rsqrt_type_cast(%arg: memref<4x8x16xf16>, %out: memref<4x8x16xf32>) {
+  // CHECK: operand 1 ('f16') doesn't match the element type of the enclosing linalg.generic op ('f32')
+  linalg.rsqrt ins(%arg : memref<4x8x16xf16>) outs(%out: memref<4x8x16xf32>)
+  return
+}
+
+// -----
+
+func.func @rsqrt_broadcast(%arg: memref<8x16xf32>, %out: memref<4x8x16xf32>) {
+  // CHECK: op expected operand rank (2) to match the result rank of indexing_map #0 (3)
+  linalg.rsqrt ins(%arg : memref<8x16xf32>) outs(%out: memref<4x8x16xf32>)
+  return
+}
+
+// -----
+
+func.func @square_type_cast(%arg: memref<4x8x16xf16>, %out: memref<4x8x16xf32>) {
+  // CHECK: operand 1 ('f16') doesn't match the element type of the enclosing linalg.generic op ('f32')
+  linalg.square ins(%arg : memref<4x8x16xf16>) outs(%out: memref<4x8x16xf32>)
+  return
+}
+
+// -----
+
+func.func @square_broadcast(%arg: memref<8x16xf32>, %out: memref<4x8x16xf32>) {
+  // CHECK: op expected operand rank (2) to match the result rank of indexing_map #0 (3)
+  linalg.square ins(%arg : memref<8x16xf32>) outs(%out: memref<4x8x16xf32>)
+  return
+}
+
+// -----
+
+func.func @tanh_type_cast(%arg: memref<4x8x16xf16>, %out: memref<4x8x16xf32>) {
+  // CHECK: operand 1 ('f16') doesn't match the element type of the enclosing linalg.generic op ('f32')
+  linalg.tanh ins(%arg : memref<4x8x16xf16>) outs(%out: memref<4x8x16xf32>)
+  return
+}
+
+// -----
+
+func.func @tanh_broadcast(%arg: memref<8x16xf32>, %out: memref<4x8x16xf32>) {
+  // CHECK: op expected operand rank (2) to match the result rank of indexing_map #0 (3)
+  linalg.tanh ins(%arg : memref<8x16xf32>) outs(%out: memref<4x8x16xf32>)
+  return
+}
+
+// -----
+
+func.func @erf_type_cast(%arg: memref<4x8x16xf16>, %out: memref<4x8x16xf32>) {
+  // CHECK: operand 1 ('f16') doesn't match the element type of the enclosing linalg.generic op ('f32')
+  linalg.erf ins(%arg : memref<4x8x16xf16>) outs(%out: memref<4x8x16xf32>)
+  return
+}
+
+// -----
+
+func.func @erf_broadcast(%arg: memref<8x16xf32>, %out: memref<4x8x16xf32>) {
+  // CHECK: op expected operand rank (2) to match the result rank of indexing_map #0 (3)
+  linalg.erf ins(%arg : memref<8x16xf32>) outs(%out: memref<4x8x16xf32>)
+  return
+}
+
+// -----
+
 func.func @max_type_cast(%arg0: memref<4x8x16xf32>, %arg1: memref<4x8x16xf16>, %arg2: memref<4x8x16xf32>) {
   // CHECK: op requires the same type for all operands and results
   linalg.max ins(%arg0, %arg1 : memref<4x8x16xf32>, memref<4x8x16xf16>) outs(%arg2: memref<4x8x16xf32>)
@@ -189,3 +301,36 @@ func.func @max_broadcast(%arg0: memref<8x16xf32>, %arg1: memref<4x8x16xf32>, %ar
   linalg.max ins(%arg0, %arg1 : memref<8x16xf32>, memref<4x8x16xf32>) outs(%arg2: memref<4x8x16xf32>)
   return
 }
+
+// -----
+
+func.func @min_type_cast(%arg0: memref<4x8x16xf32>, %arg1: memref<4x8x16xf16>, %arg2: memref<4x8x16xf32>) {
+  // CHECK: op requires the same type for all operands and results
+  linalg.min ins(%arg0, %arg1 : memref<4x8x16xf32>, memref<4x8x16xf16>) outs(%arg2: memref<4x8x16xf32>)
+  return
+}
+
+// -----
+
+func.func @min_broadcast(%arg0: memref<8x16xf32>, %arg1: memref<4x8x16xf32>, %arg2: memref<4x8x16xf32>) {
+  // CHECK: op expected operand rank (2) to match the result rank of indexing_map #0 (3)
+  linalg.min ins(%arg0, %arg1 : memref<8x16xf32>, memref<4x8x16xf32>) outs(%arg2: memref<4x8x16xf32>)
+  return
+}
+
+// -----
+
+func.func @powf_type_cast(%arg0: memref<4x8x16xf32>, %arg1: memref<4x8x16xf16>, %arg2: memref<4x8x16xf32>) {
+  // CHECK: op requires the same type for all operands and results
+  linalg.powf ins(%arg0, %arg1 : memref<4x8x16xf32>, memref<4x8x16xf16>) outs(%arg2: memref<4x8x16xf32>)
+  return
+}
+
+// -----
+
+func.func @powf_broadcast(%arg0: memref<8x16xf32>, %arg1: memref<4x8x16xf32>, %arg2: memref<4x8x16xf32>) {
+  // CHECK: op expected operand rank (2) to match the result rank of indexing_map #0 (3)
+  linalg.powf ins(%arg0, %arg1 : memref<8x16xf32>, memref<4x8x16xf32>) outs(%arg2: memref<4x8x16xf32>)
+  return
+}
+
diff --git a/mlir/test/Dialect/Linalg/named-ops.mlir b/mlir/test/Dialect/Linalg/named-ops.mlir
index 7064e1b3f9dc..fefe5578947f 100644
--- a/mlir/test/Dialect/Linalg/named-ops.mlir
+++ b/mlir/test/Dialect/Linalg/named-ops.mlir
@@ -1597,6 +1597,223 @@ func.func @negf_tensor(%arg0: tensor<4x8x16xf32>) -> tensor<4x8x16xf32> {
 
 // -----
 
+// CHECK-LABEL: func @reciprocal_dynamic
+func.func @reciprocal_dynamic(%arg0: memref<?x?x?xf32>, %arg1: memref<?x?x?xf32>) {
+  // CHECK: linalg.reciprocal
+  // CHECK-SAME: ins(%{{.+}} : memref<?x?x?xf32>) outs(%{{.+}} : memref<?x?x?xf32>)
+  linalg.reciprocal ins(%arg0 : memref<?x?x?xf32>) outs(%arg1: memref<?x?x?xf32>)
+  return
+}
+
+// -----
+
+// CHECK-LABEL: func @reciprocal_static
+func.func @reciprocal_static(%arg0: memref<4x8x16xf32>, %arg1: memref<4x8x16xf32>) {
+  // CHECK: linalg.reciprocal
+  // CHECK-SAME: ins(%{{.+}} : memref<4x8x16xf32>) outs(%{{.+}} : memref<4x8x16xf32>)
+  linalg.reciprocal ins(%arg0 : memref<4x8x16xf32>) outs(%arg1: memref<4x8x16xf32>)
+  return
+}
+
+// -----
+
+// CHECK-LABEL: func @reciprocal_tensor
+func.func @reciprocal_tensor(%arg0: tensor<4x8x16xf32>) -> tensor<4x8x16xf32> {
+  %0 = tensor.empty() : tensor<4x8x16xf32>
+  // CHECK: linalg.reciprocal
+  // CHECK-SAME: ins(%{{.+}} : tensor<4x8x16xf32>) outs(%{{.+}} : tensor<4x8x16xf32>)
+  %1 = linalg.reciprocal ins(%arg0 : tensor<4x8x16xf32>) outs(%0: tensor<4x8x16xf32>) -> tensor<4x8x16xf32>
+  return %1 : tensor<4x8x16xf32>
+}
+
+// -----
+
+// CHECK-LABEL: func @round_dynamic
+func.func @round_dynamic(%arg0: memref<?x?x?xf32>, %arg1: memref<?x?x?xf32>) {
+  // CHECK: linalg.round
+  // CHECK-SAME: ins(%{{.+}} : memref<?x?x?xf32>) outs(%{{.+}} : memref<?x?x?xf32>)
+  linalg.round ins(%arg0 : memref<?x?x?xf32>) outs(%arg1: memref<?x?x?xf32>)
+  return
+}
+
+// -----
+
+// CHECK-LABEL: func @round_static
+func.func @round_static(%arg0: memref<4x8x16xf32>, %arg1: memref<4x8x16xf32>) {
+  // CHECK: linalg.round
+  // CHECK-SAME: ins(%{{.+}} : memref<4x8x16xf32>) outs(%{{.+}} : memref<4x8x16xf32>)
+  linalg.round ins(%arg0 : memref<4x8x16xf32>) outs(%arg1: memref<4x8x16xf32>)
+  return
+}
+
+// -----
+
+// CHECK-LABEL: func @round_tensor
+func.func @round_tensor(%arg0: tensor<4x8x16xf32>) -> tensor<4x8x16xf32> {
+  %0 = tensor.empty() : tensor<4x8x16xf32>
+  // CHECK: linalg.round
+  // CHECK-SAME: ins(%{{.+}} : tensor<4x8x16xf32>) outs(%{{.+}} : tensor<4x8x16xf32>)
+  %1 = linalg.round ins(%arg0 : tensor<4x8x16xf32>) outs(%0: tensor<4x8x16xf32>) -> tensor<4x8x16xf32>
+  return %1 : tensor<4x8x16xf32>
+}
+
+// -----
+
+// CHECK-LABEL: func @sqrt_dynamic
+func.func @sqrt_dynamic(%arg0: memref<?x?x?xf32>, %arg1: memref<?x?x?xf32>) {
+  // CHECK: linalg.sqrt
+  // CHECK-SAME: ins(%{{.+}} : memref<?x?x?xf32>) outs(%{{.+}} : memref<?x?x?xf32>)
+  linalg.sqrt ins(%arg0 : memref<?x?x?xf32>) outs(%arg1: memref<?x?x?xf32>)
+  return
+}
+
+// -----
+
+// CHECK-LABEL: func @sqrt_static
+func.func @sqrt_static(%arg0: memref<4x8x16xf32>, %arg1: memref<4x8x16xf32>) {
+  // CHECK: linalg.sqrt
+  // CHECK-SAME: ins(%{{.+}} : memref<4x8x16xf32>) outs(%{{.+}} : memref<4x8x16xf32>)
+  linalg.sqrt ins(%arg0 : memref<4x8x16xf32>) outs(%arg1: memref<4x8x16xf32>)
+  return
+}
+
+// -----
+
+// CHECK-LABEL: func @sqrt_tensor
+func.func @sqrt_tensor(%arg0: tensor<4x8x16xf32>) -> tensor<4x8x16xf32> {
+  %0 = tensor.empty() : tensor<4x8x16xf32>
+  // CHECK: linalg.sqrt
+  // CHECK-SAME: ins(%{{.+}} : tensor<4x8x16xf32>) outs(%{{.+}} : tensor<4x8x16xf32>)
+  %1 = linalg.sqrt ins(%arg0 : tensor<4x8x16xf32>) outs(%0: tensor<4x8x16xf32>) -> tensor<4x8x16xf32>
+  return %1 : tensor<4x8x16xf32>
+}
+
+// -----
+
+// CHECK-LABEL: func @rsqrt_dynamic
+func.func @rsqrt_dynamic(%arg0: memref<?x?x?xf32>, %arg1: memref<?x?x?xf32>) {
+  // CHECK: linalg.rsqrt
+  // CHECK-SAME: ins(%{{.+}} : memref<?x?x?xf32>) outs(%{{.+}} : memref<?x?x?xf32>)
+  linalg.rsqrt ins(%arg0 : memref<?x?x?xf32>) outs(%arg1: memref<?x?x?xf32>)
+  return
+}
+
+// -----
+
+// CHECK-LABEL: func @rsqrt_static
+func.func @rsqrt_static(%arg0: memref<4x8x16xf32>, %arg1: memref<4x8x16xf32>) {
+  // CHECK: linalg.rsqrt
+  // CHECK-SAME: ins(%{{.+}} : memref<4x8x16xf32>) outs(%{{.+}} : memref<4x8x16xf32>)
+  linalg.rsqrt ins(%arg0 : memref<4x8x16xf32>) outs(%arg1: memref<4x8x16xf32>)
+  return
+}
+
+// -----
+
+// CHECK-LABEL: func @rsqrt_tensor
+func.func @rsqrt_tensor(%arg0: tensor<4x8x16xf32>) -> tensor<4x8x16xf32> {
+  %0 = tensor.empty() : tensor<4x8x16xf32>
+  // CHECK: linalg.rsqrt
+  // CHECK-SAME: ins(%{{.+}} : tensor<4x8x16xf32>) outs(%{{.+}} : tensor<4x8x16xf32>)
+  %1 = linalg.rsqrt ins(%arg0 : tensor<4x8x16xf32>) outs(%0: tensor<4x8x16xf32>) -> tensor<4x8x16xf32>
+  return %1 : tensor<4x8x16xf32>
+}
+
+// -----
+
+// CHECK-LABEL: func @square_dynamic
+func.func @square_dynamic(%arg0: memref<?x?x?xf32>, %arg1: memref<?x?x?xf32>) {
+  // CHECK: linalg.square
+  // CHECK-SAME: ins(%{{.+}} : memref<?x?x?xf32>) outs(%{{.+}} : memref<?x?x?xf32>)
+  linalg.square ins(%arg0 : memref<?x?x?xf32>) outs(%arg1: memref<?x?x?xf32>)
+  return
+}
+
+// -----
+
+// CHECK-LABEL: func @square_static
+func.func @square_static(%arg0: memref<4x8x16xf32>, %arg1: memref<4x8x16xf32>) {
+  // CHECK: linalg.square
+  // CHECK-SAME: ins(%{{.+}} : memref<4x8x16xf32>) outs(%{{.+}} : memref<4x8x16xf32>)
+  linalg.square ins(%arg0 : memref<4x8x16xf32>) outs(%arg1: memref<4x8x16xf32>)
+  return
+}
+
+// -----
+
+// CHECK-LABEL: func @square_tensor
+func.func @square_tensor(%arg0: tensor<4x8x16xf32>) -> tensor<4x8x16xf32> {
+  %0 = tensor.empty() : tensor<4x8x16xf32>
+  // CHECK: linalg.square
+  // CHECK-SAME: ins(%{{.+}} : tensor<4x8x16xf32>) outs(%{{.+}} : tensor<4x8x16xf32>)
+  %1 = linalg.square ins(%arg0 : tensor<4x8x16xf32>) outs(%0: tensor<4x8x16xf32>) -> tensor<4x8x16xf32>
+  return %1 : tensor<4x8x16xf32>
+}
+
+// -----
+
+// CHECK-LABEL: func @tanh_dynamic
+func.func @tanh_dynamic(%arg0: memref<?x?x?xf32>, %arg1: memref<?x?x?xf32>) {
+  // CHECK: linalg.tanh
+  // CHECK-SAME: ins(%{{.+}} : memref<?x?x?xf32>) outs(%{{.+}} : memref<?x?x?xf32>)
+  linalg.tanh ins(%arg0 : memref<?x?x?xf32>) outs(%arg1: memref<?x?x?xf32>)
+  return
+}
+
+// -----
+
+// CHECK-LABEL: func @tanh_static
+func.func @tanh_static(%arg0: memref<4x8x16xf32>, %arg1: memref<4x8x16xf32>) {
+  // CHECK: linalg.tanh
+  // CHECK-SAME: ins(%{{.+}} : memref<4x8x16xf32>) outs(%{{.+}} : memref<4x8x16xf32>)
+  linalg.tanh ins(%arg0 : memref<4x8x16xf32>) outs(%arg1: memref<4x8x16xf32>)
+  return
+}
+
+// -----
+
+// CHECK-LABEL: func @tanh_tensor
+func.func @tanh_tensor(%arg0: tensor<4x8x16xf32>) -> tensor<4x8x16xf32> {
+  %0 = tensor.empty() : tensor<4x8x16xf32>
+  // CHECK: linalg.tanh
+  // CHECK-SAME: ins(%{{.+}} : tensor<4x8x16xf32>) outs(%{{.+}} : tensor<4x8x16xf32>)
+  %1 = linalg.tanh ins(%arg0 : tensor<4x8x16xf32>) outs(%0: tensor<4x8x16xf32>) -> tensor<4x8x16xf32>
+  return %1 : tensor<4x8x16xf32>
+}
+
+// -----
+
+// CHECK-LABEL: func @erf_dynamic
+func.func @erf_dynamic(%arg0: memref<?x?x?xf32>, %arg1: memref<?x?x?xf32>) {
+  // CHECK: linalg.erf
+  // CHECK-SAME: ins(%{{.+}} : memref<?x?x?xf32>) outs(%{{.+}} : memref<?x?x?xf32>)
+  linalg.erf ins(%arg0 : memref<?x?x?xf32>) outs(%arg1: memref<?x?x?xf32>)
+  return
+}
+
+// -----
+
+// CHECK-LABEL: func @erf_static
+func.func @erf_static(%arg0: memref<4x8x16xf32>, %arg1: memref<4x8x16xf32>) {
+  // CHECK: linalg.erf
+  // CHECK-SAME: ins(%{{.+}} : memref<4x8x16xf32>) outs(%{{.+}} : memref<4x8x16xf32>)
+  linalg.erf ins(%arg0 : memref<4x8x16xf32>) outs(%arg1: memref<4x8x16xf32>)
+  return
+}
+
+// -----
+
+// CHECK-LABEL: func @erf_tensor
+func.func @erf_tensor(%arg0: tensor<4x8x16xf32>) -> tensor<4x8x16xf32> {
+  %0 = tensor.empty() : tensor<4x8x16xf32>
+  // CHECK: linalg.erf
+  // CHECK-SAME: ins(%{{.+}} : tensor<4x8x16xf32>) outs(%{{.+}} : tensor<4x8x16xf32>)
+  %1 = linalg.erf ins(%arg0 : tensor<4x8x16xf32>) outs(%0: tensor<4x8x16xf32>) -> tensor<4x8x16xf32>
+  return %1 : tensor<4x8x16xf32>
+}
+
+// -----
+
 // CHECK-LABEL: func @max_dynamic
 func.func @max_dynamic(%arg0: memref<?x?x?xf32>, %arg1: memref<?x?x?xf32>, %arg2: memref<?x?x?xf32>) {
   // CHECK: linalg.max
@@ -1631,6 +1848,74 @@ func.func @max_tensor(%arg0: tensor<4x8x16xf32>, %arg1: tensor<4x8x16xf32>) -> t
 
 // -----
 
+// CHECK-LABEL: func @min_dynamic
+func.func @min_dynamic(%arg0: memref<?x?x?xf32>, %arg1: memref<?x?x?xf32>, %arg2: memref<?x?x?xf32>) {
+  // CHECK: linalg.min
+  // CHECK-SAME: ins(%{{.+}}, %{{.+}} : memref<?x?x?xf32>, memref<?x?x?xf32>)
+  // CHECK-SAME: outs(%{{.+}} : memref<?x?x?xf32>)
+  linalg.min ins(%arg0, %arg1 : memref<?x?x?xf32>, memref<?x?x?xf32>) outs(%arg2: memref<?x?x?xf32>)
+  return
+}
+
+// -----
+
+// CHECK-LABEL: func @min_static
+func.func @min_static(%arg0: memref<4x8x16xf32>, %arg1: memref<4x8x16xf32>, %arg2: memref<4x8x16xf32>) {
+  // CHECK: linalg.min
+  // CHECK-SAME: ins(%{{.+}}, %{{.+}} : memref<4x8x16xf32>, memref<4x8x16xf32>)
+  // CHECK-SAME: outs(%{{.+}} : memref<4x8x16xf32>)
+  linalg.min ins(%arg0, %arg1 : memref<4x8x16xf32>, memref<4x8x16xf32>) outs(%arg2: memref<4x8x16xf32>)
+  return
+}
+
+// -----
+
+// CHECK-LABEL: func @min_tensor
+func.func @min_tensor(%arg0: tensor<4x8x16xf32>, %arg1: tensor<4x8x16xf32>) -> tensor<4x8x16xf32> {
+  %0 = tensor.empty() : tensor<4x8x16xf32>
+  // CHECK: linalg.min
+  // CHECK-SAME: ins(%{{.+}}, %{{.+}} : tensor<4x8x16xf32>, tensor<4x8x16xf32>)
+  // CHECK-SAME: outs(%{{.+}} : tensor<4x8x16xf32>)
+  %1 = linalg.min ins(%arg0, %arg1 : tensor<4x8x16xf32>, tensor<4x8x16xf32>) outs(%0: tensor<4x8x16xf32>) -> tensor<4x8x16xf32>
+  return %1 : tensor<4x8x16xf32>
+}
+
+// -----
+
+// CHECK-LABEL: func @powf_dynamic
+func.func @powf_dynamic(%arg0: memref<?x?x?xf32>, %arg1: memref<?x?x?xf32>, %arg2: memref<?x?x?xf32>) {
+  // CHECK: linalg.powf
+  // CHECK-SAME: ins(%{{.+}}, %{{.+}} : memref<?x?x?xf32>, memref<?x?x?xf32>)
+  // CHECK-SAME: outs(%{{.+}} : memref<?x?x?xf32>)
+  linalg.powf ins(%arg0, %arg1 : memref<?x?x?xf32>, memref<?x?x?xf32>) outs(%arg2: memref<?x?x?xf32>)
+  return
+}
+
+// -----
+
+// CHECK-LABEL: func @powf_static
+func.func @powf_static(%arg0: memref<4x8x16xf32>, %arg1: memref<4x8x16xf32>, %arg2: memref<4x8x16xf32>) {
+  // CHECK: linalg.powf
+  // CHECK-SAME: ins(%{{.+}}, %{{.+}} : memref<4x8x16xf32>, memref<4x8x16xf32>)
+  // CHECK-SAME: outs(%{{.+}} : memref<4x8x16xf32>)
+  linalg.powf ins(%arg0, %arg1 : memref<4x8x16xf32>, memref<4x8x16xf32>) outs(%arg2: memref<4x8x16xf32>)
+  return
+}
+
+// -----
+
+// CHECK-LABEL: func @powf_tensor
+func.func @powf_tensor(%arg0: tensor<4x8x16xf32>, %arg1: tensor<4x8x16xf32>) -> tensor<4x8x16xf32> {
+  %0 = tensor.empty() : tensor<4x8x16xf32>
+  // CHECK: linalg.powf
+  // CHECK-SAME: ins(%{{.+}}, %{{.+}} : tensor<4x8x16xf32>, tensor<4x8x16xf32>)
+  // CHECK-SAME: outs(%{{.+}} : tensor<4x8x16xf32>)
+  %1 = linalg.powf ins(%arg0, %arg1 : tensor<4x8x16xf32>, tensor<4x8x16xf32>) outs(%0: tensor<4x8x16xf32>) -> tensor<4x8x16xf32>
+  return %1 : tensor<4x8x16xf32>
+}
+
+// -----
+
 // CHECK-LABEL: func @fill_tensor
 func.func @fill_tensor(%f: f32, %v: vector<2x4xf32>) -> (tensor<f32>, tensor<vector<2x4xf32>>) {
   %e0 = tensor.empty() : tensor<f32>
diff --git a/mlir/test/Dialect/Linalg/runtime-verification.mlir b/mlir/test/Dialect/Linalg/runtime-verification.mlir
new file mode 100644
index 000000000000..a4f29d8457e5
--- /dev/null
+++ b/mlir/test/Dialect/Linalg/runtime-verification.mlir
@@ -0,0 +1,43 @@
+// RUN: mlir-opt %s -generate-runtime-verification | FileCheck %s
+
+// Most of the tests for linalg runtime-verification are implemented as integration tests.
+
+#identity = affine_map<(d0) -> (d0)>
+
+// CHECK-LABEL: @static_dims
+func.func @static_dims(%arg0: tensor<5xf32>, %arg1: tensor<5xf32>) -> (tensor<5xf32>) {
+    // CHECK: %[[TRUE:.*]] = index.bool.constant true
+    // CHECK: cf.assert %[[TRUE]]
+    %result = tensor.empty() : tensor<5xf32> 
+    %0 = linalg.generic {
+      indexing_maps = [#identity, #identity, #identity],
+      iterator_types = ["parallel"]
+    } ins(%arg0, %arg1 : tensor<5xf32>, tensor<5xf32>)
+      outs(%result : tensor<5xf32>) {
+      ^bb0(%gen_arg1: f32, %gen_arg2: f32, %out: f32) :
+        %tmp1 = arith.addf %gen_arg1, %gen_arg2 : f32
+        linalg.yield %tmp1 : f32
+    } -> tensor<5xf32>
+    return %0 : tensor<5xf32>
+}
+
+// -----
+
+#map = affine_map<() -> ()>
+
+// CHECK-LABEL: @scalars
+func.func @scalars(%arg0: tensor<f32>, %arg1: tensor<f32>) -> (tensor<f32>) {
+    // No runtime checks are required if the operands are all scalars
+    // CHECK-NOT: cf.assert
+    %result = tensor.empty() : tensor<f32> 
+    %0 = linalg.generic {
+      indexing_maps = [#map, #map, #map],
+      iterator_types = []
+    } ins(%arg0, %arg1 : tensor<f32>, tensor<f32>)
+      outs(%result : tensor<f32>) {
+      ^bb0(%gen_arg1: f32, %gen_arg2: f32, %out: f32) :
+        %tmp1 = arith.addf %gen_arg1, %gen_arg2 : f32
+        linalg.yield %tmp1 : f32
+    } -> tensor<f32>
+    return %0 : tensor<f32>
+}
diff --git a/mlir/test/Dialect/Linalg/transpose-matmul-a.mlir b/mlir/test/Dialect/Linalg/transpose-matmul-a.mlir
new file mode 100644
index 000000000000..b1f33cfa5632
--- /dev/null
+++ b/mlir/test/Dialect/Linalg/transpose-matmul-a.mlir
@@ -0,0 +1,14 @@
+// RUN: mlir-opt %s
+
+module attributes {transform.with_named_sequence} {
+  transform.named_sequence @__transform_main(%arg1: !transform.any_op {transform.readonly}) {
+    %matmul = transform.structured.match ops{["linalg.matmul", "linalg.batch_matmul"]} in %arg1 : (!transform.any_op) -> !transform.any_op
+    transform.structured.transpose_matmul %matmul : (!transform.any_op) -> (!transform.any_op)
+    %0 = transform.structured.match ops{["func.func"]} in %arg1 : (!transform.any_op) -> !transform.any_op
+    transform.apply_cse to %0 : !transform.any_op
+    transform.apply_patterns to %0 {
+      transform.apply_patterns.canonicalization
+    } : !transform.any_op
+    transform.yield
+  }
+}
diff --git a/mlir/test/Dialect/Linalg/transpose-matmul-b.mlir b/mlir/test/Dialect/Linalg/transpose-matmul-b.mlir
new file mode 100644
index 000000000000..41e64c04dc6e
--- /dev/null
+++ b/mlir/test/Dialect/Linalg/transpose-matmul-b.mlir
@@ -0,0 +1,14 @@
+// RUN: mlir-opt %s
+
+module attributes {transform.with_named_sequence} {
+  transform.named_sequence @__transform_main(%arg1: !transform.any_op {transform.readonly}) {
+    %matmul = transform.structured.match ops{["linalg.matmul", "linalg.batch_matmul"]} in %arg1 : (!transform.any_op) -> !transform.any_op
+    transform.structured.transpose_matmul %matmul <rhs> : (!transform.any_op) -> (!transform.any_op)
+    %0 = transform.structured.match ops{["func.func"]} in %arg1 : (!transform.any_op) -> !transform.any_op
+    transform.apply_cse to %0 : !transform.any_op
+    transform.apply_patterns to %0 {
+      transform.apply_patterns.canonicalization
+    } : !transform.any_op
+    transform.yield
+  }
+}
diff --git a/mlir/test/Dialect/Linalg/transpose-matmul.mlir b/mlir/test/Dialect/Linalg/transpose-matmul.mlir
new file mode 100644
index 000000000000..d2b7e9f7f199
--- /dev/null
+++ b/mlir/test/Dialect/Linalg/transpose-matmul.mlir
@@ -0,0 +1,179 @@
+// RUN: mlir-opt -transform-preload-library='transform-library-paths=%p/transpose-matmul-a.mlir' -transform-interpreter -split-input-file %s | FileCheck %s --check-prefixes=CHECK,TRANSPOSE-A
+// RUN: mlir-opt -transform-preload-library='transform-library-paths=%p/transpose-matmul-b.mlir' -transform-interpreter -split-input-file %s | FileCheck %s --check-prefixes=CHECK,TRANSPOSE-B
+
+// CHECK-LABEL:   func.func @matmul_static(
+// CHECK-SAME:                             %[[A:.*]]: tensor<16x8xf32>,
+// CHECK-SAME:                             %[[B:.*]]: tensor<8x16xf32>) -> tensor<16x16xf32> {
+// CHECK:           %[[C0_F32:.*]] = arith.constant 0.000000e+00 : f32
+// CHECK:           %[[C_INIT:.*]] = tensor.empty() : tensor<16x16xf32>
+// CHECK:           %[[C_ZERO:.*]] = linalg.fill ins(%[[C0_F32]] : f32) outs(%[[C_INIT]] : tensor<16x16xf32>) -> tensor<16x16xf32>
+// TRANSPOSE-A:     %[[A_TRANSP_INIT:.*]] = tensor.empty() : tensor<8x16xf32>
+// TRANSPOSE-A:     %[[A_TRANSP:.*]] = linalg.transpose ins(%[[A]] : tensor<16x8xf32>) outs(%[[A_TRANSP_INIT]] : tensor<8x16xf32>) permutation = [1, 0]
+// TRANSPOSE-A:     %[[C:.*]] = linalg.matmul_transpose_a ins(%[[A_TRANSP]], %[[B]] : tensor<8x16xf32>, tensor<8x16xf32>) outs(%[[C_ZERO]] : tensor<16x16xf32>) -> tensor<16x16xf32>
+// TRANSPOSE-B:     %[[B_TRANSP_INIT:.*]] = tensor.empty() : tensor<16x8xf32>
+// TRANSPOSE-B:     %[[B_TRANSP:.*]] = linalg.transpose ins(%[[B]] : tensor<8x16xf32>) outs(%[[B_TRANSP_INIT]] : tensor<16x8xf32>) permutation = [1, 0]
+// TRANSPOSE-B:     %[[C:.*]] = linalg.matmul_transpose_b ins(%[[A]], %[[B_TRANSP]] : tensor<16x8xf32>, tensor<16x8xf32>) outs(%[[C_ZERO]] : tensor<16x16xf32>) -> tensor<16x16xf32>
+// CHECK:           return %[[C]] : tensor<16x16xf32>
+// CHECK:         }
+func.func @matmul_static(%A: tensor<16x8xf32>, %B: tensor<8x16xf32>) -> (tensor<16x16xf32>) {
+  %cst = arith.constant 0.0 : f32
+  %init = tensor.empty() : tensor<16x16xf32>
+  %C = linalg.fill ins(%cst : f32) outs(%init : tensor<16x16xf32>) -> tensor<16x16xf32>
+  %0 = linalg.matmul ins(%A, %B : tensor<16x8xf32>, tensor<8x16xf32>) outs(%C : tensor<16x16xf32>) -> tensor<16x16xf32>
+  return %0 : tensor<16x16xf32>
+}
+
+//-----
+
+// CHECK-LABEL:   func.func @matmul_dynamic(
+// CHECK-SAME:                              %[[A:.*]]: tensor<?x?xf32>,
+// CHECK-SAME:                              %[[B:.*]]: tensor<?x?xf32>) -> tensor<?x?xf32> {
+// CHECK:           %[[C0_F32:.*]] = arith.constant 0.000000e+00 : f32
+// CHECK:           %[[C0:.*]] = arith.constant 0 : index
+// CHECK:           %[[C1:.*]] = arith.constant 1 : index
+// CHECK:           %[[A_DIM0:.*]] = tensor.dim %[[A]], %[[C0]] : tensor<?x?xf32>
+// CHECK:           %[[B_DIM1:.*]] = tensor.dim %[[B]], %[[C1]] : tensor<?x?xf32>
+// CHECK:           %[[C_INIT:.*]] = tensor.empty(%[[A_DIM0]], %[[B_DIM1]]) : tensor<?x?xf32>
+// CHECK:           %[[C_ZERO:.*]] = linalg.fill ins(%[[C0_F32]] : f32) outs(%[[C_INIT]] : tensor<?x?xf32>) -> tensor<?x?xf32>
+// TRANSPOSE-A:     %[[A_DIM1:.*]] = tensor.dim %[[A]], %[[C1]] : tensor<?x?xf32>
+// TRANSPOSE-A:     %[[A_TRANSP_INIT:.*]] = tensor.empty(%[[A_DIM1]], %[[A_DIM0]]) : tensor<?x?xf32>
+// TRANSPOSE-A:     %[[A_TRANSP:.*]] = linalg.transpose ins(%[[A]] : tensor<?x?xf32>) outs(%[[A_TRANSP_INIT]] : tensor<?x?xf32>) permutation = [1, 0]
+// TRANSPOSE-A:     %[[C:.*]] = linalg.matmul_transpose_a ins(%[[A_TRANSP]], %[[B]] : tensor<?x?xf32>, tensor<?x?xf32>) outs(%[[C_ZERO]] : tensor<?x?xf32>) -> tensor<?x?xf32>
+// TRANSPOSE-B:     %[[B_DIM0:.*]] = tensor.dim %[[B]], %[[C0]] : tensor<?x?xf32>
+// TRANSPOSE-B:     %[[B_TRANSP_INIT:.*]] = tensor.empty(%[[B_DIM1]], %[[B_DIM0]]) : tensor<?x?xf32>
+// TRANSPOSE-B:     %[[B_TRANSP:.*]] = linalg.transpose ins(%[[B]] : tensor<?x?xf32>) outs(%[[B_TRANSP_INIT]] : tensor<?x?xf32>) permutation = [1, 0]
+// TRANSPOSE-B:     %[[C:.*]] = linalg.matmul_transpose_b ins(%[[A]], %[[B_TRANSP]] : tensor<?x?xf32>, tensor<?x?xf32>) outs(%[[C_ZERO]] : tensor<?x?xf32>) -> tensor<?x?xf32>
+// CHECK:           return %[[C]] : tensor<?x?xf32>
+// CHECK:         }
+func.func @matmul_dynamic(%A: tensor<?x?xf32>, %B: tensor<?x?xf32>) -> (tensor<?x?xf32>) {
+  %cst = arith.constant 0.0 : f32
+  %c0 = arith.constant 0 : index
+  %c1 = arith.constant 1 : index
+  %d0 = tensor.dim %A, %c0 : tensor<?x?xf32>
+  %d1 = tensor.dim %B, %c1 : tensor<?x?xf32>
+  %init = tensor.empty(%d0, %d1) : tensor<?x?xf32>
+  %C = linalg.fill ins(%cst : f32) outs(%init : tensor<?x?xf32>) -> tensor<?x?xf32>
+  %0 = linalg.matmul ins(%A, %B : tensor<?x?xf32>, tensor<?x?xf32>) outs(%C : tensor<?x?xf32>) -> tensor<?x?xf32>
+  return %0 : tensor<?x?xf32>
+}
+
+//-----
+
+// CHECK-LABEL:   func.func @matmul_mixed(
+// CHECK-SAME:                            %[[A:.*]]: tensor<?x8xf32>,
+// CHECK-SAME:                            %[[B:.*]]: tensor<8x16xf32>) -> tensor<?x16xf32> {
+// CHECK:           %[[C0_F32:.*]] = arith.constant 0.000000e+00 : f32
+// CHECK:           %[[C0:.*]] = arith.constant 0 : index
+// CHECK:           %[[A_DIM0:.*]] = tensor.dim %[[A]], %[[C0]] : tensor<?x8xf32>
+// CHECK:           %[[C_INIT:.*]] = tensor.empty(%[[A_DIM0]]) : tensor<?x16xf32>
+// CHECK:           %[[C_ZERO:.*]] = linalg.fill ins(%[[C0_F32]] : f32) outs(%[[C_INIT]] : tensor<?x16xf32>) -> tensor<?x16xf32>
+// TRANSPOSE-A:     %[[A_TRANSP_INIT:.*]] = tensor.empty(%[[A_DIM0]]) : tensor<8x?xf32>
+// TRANSPOSE-A:     %[[A_TRANSP:.*]] = linalg.transpose ins(%[[A]] : tensor<?x8xf32>) outs(%[[A_TRANSP_INIT]] : tensor<8x?xf32>) permutation = [1, 0]
+// TRANSPOSE-A:     %[[B0:.*]] = linalg.matmul_transpose_a ins(%[[A_TRANSP]], %[[B]] : tensor<8x?xf32>, tensor<8x16xf32>) outs(%[[C_ZERO]] : tensor<?x16xf32>) -> tensor<?x16xf32>
+// TRANSPOSE-B:     %[[B_TRANSP_INIT:.*]] = tensor.empty() : tensor<16x8xf32>
+// TRANSPOSE-B:     %[[B_TRANSP:.*]] = linalg.transpose ins(%[[B]] : tensor<8x16xf32>) outs(%[[B_TRANSP_INIT]] : tensor<16x8xf32>) permutation = [1, 0]
+// TRANSPOSE-B:     %[[B0:.*]] = linalg.matmul_transpose_b ins(%[[A]], %[[B_TRANSP]] : tensor<?x8xf32>, tensor<16x8xf32>) outs(%[[C_ZERO]] : tensor<?x16xf32>) -> tensor<?x16xf32>
+// CHECK:           return %[[B0]] : tensor<?x16xf32>
+// CHECK:         }
+func.func @matmul_mixed(%A: tensor<?x8xf32>, %B: tensor<8x16xf32>) -> (tensor<?x16xf32>) {
+  %cst = arith.constant 0.0 : f32
+  %c0 = arith.constant 0 : index
+  %c1 = arith.constant 1 : index
+  %d0 = tensor.dim %A, %c0 : tensor<?x8xf32>
+  %init = tensor.empty(%d0) : tensor<?x16xf32>
+  %C = linalg.fill ins(%cst : f32) outs(%init : tensor<?x16xf32>) -> tensor<?x16xf32>
+  %0 = linalg.matmul ins(%A, %B : tensor<?x8xf32>, tensor<8x16xf32>) outs(%C : tensor<?x16xf32>) -> tensor<?x16xf32>
+  return %0 : tensor<?x16xf32>
+}
+
+//-----
+
+// CHECK-LABEL:   func.func @batch_matmul_static(
+// CHECK-SAME:                                   %[[A:.*]]: tensor<2x16x8xf32>,
+// CHECK-SAME:                                   %[[B:.*]]: tensor<2x8x16xf32>) -> tensor<2x16x16xf32> {
+// CHECK:           %[[C0_F32:.*]] = arith.constant 0.000000e+00 : f32
+// CHECK:           %[[C_INIT:.*]] = tensor.empty() : tensor<2x16x16xf32>
+// CHECK:           %[[C_ZERO:.*]] = linalg.fill ins(%[[C0_F32]] : f32) outs(%[[C_INIT]] : tensor<2x16x16xf32>) -> tensor<2x16x16xf32>
+// TRANSPOSE-A:     %[[A_TRANSP_INIT:.*]] = tensor.empty() : tensor<2x8x16xf32>
+// TRANSPOSE-A:     %[[A_TRANSP:.*]] = linalg.transpose ins(%[[A]] : tensor<2x16x8xf32>) outs(%[[A_TRANSP_INIT]] : tensor<2x8x16xf32>) permutation = [0, 2, 1]
+// TRANSPOSE-A:     %[[C:.*]] = linalg.batch_matmul_transpose_a ins(%[[A_TRANSP]], %[[B]] : tensor<2x8x16xf32>, tensor<2x8x16xf32>) outs(%[[C_ZERO]] : tensor<2x16x16xf32>) -> tensor<2x16x16xf32>
+// TRANSPOSE-B:     %[[B_TRANSP_INIT:.*]] = tensor.empty() : tensor<2x16x8xf32>
+// TRANSPOSE-B:     %[[B_TRANSP:.*]] = linalg.transpose ins(%[[B]] : tensor<2x8x16xf32>) outs(%[[B_TRANSP_INIT]] : tensor<2x16x8xf32>) permutation = [0, 2, 1]
+// TRANSPOSE-B:     %[[C:.*]] = linalg.batch_matmul_transpose_b ins(%[[A]], %[[B_TRANSP]] : tensor<2x16x8xf32>, tensor<2x16x8xf32>) outs(%[[C_ZERO]] : tensor<2x16x16xf32>) -> tensor<2x16x16xf32>
+// CHECK:           return %[[C]] : tensor<2x16x16xf32>
+// CHECK:         }
+func.func @batch_matmul_static(%A: tensor<2x16x8xf32>, %B: tensor<2x8x16xf32>) -> (tensor<2x16x16xf32>) {
+  %cst = arith.constant 0.0 : f32
+  %init = tensor.empty() : tensor<2x16x16xf32>
+  %C = linalg.fill ins(%cst : f32) outs(%init : tensor<2x16x16xf32>) -> tensor<2x16x16xf32>
+  %0 = linalg.batch_matmul ins(%A, %B : tensor<2x16x8xf32>, tensor<2x8x16xf32>) outs(%C : tensor<2x16x16xf32>) -> tensor<2x16x16xf32>
+  return %0 : tensor<2x16x16xf32>
+}
+
+//-----
+
+// CHECK-LABEL:   func.func @batch_matmul_dynamic(
+// CHECK-SAME:                                    %[[A:.*]]: tensor<?x?x?xf32>,
+// CHECK-SAME:                                    %[[B:.*]]: tensor<?x?x?xf32>) -> tensor<?x?x?xf32> {
+// CHECK:           %[[C0_F32:.*]] = arith.constant 0.000000e+00 : f32
+// CHECK:           %[[C0:.*]] = arith.constant 0 : index
+// CHECK:           %[[C1:.*]] = arith.constant 1 : index
+// CHECK:           %[[C2:.*]] = arith.constant 2 : index
+// CHECK:           %[[A_DIM0:.*]] = tensor.dim %[[A]], %[[C0]] : tensor<?x?x?xf32>
+// CHECK:           %[[A_DIM1:.*]] = tensor.dim %[[A]], %[[C1]] : tensor<?x?x?xf32>
+// CHECK:           %[[B_DIM2:.*]] = tensor.dim %[[B]], %[[C2]] : tensor<?x?x?xf32>
+// CHECK:           %[[C_INIT:.*]] = tensor.empty(%[[A_DIM0]], %[[A_DIM1]], %[[B_DIM2]]) : tensor<?x?x?xf32>
+// CHECK:           %[[C_ZERO:.*]] = linalg.fill ins(%[[C0_F32]] : f32) outs(%[[C_INIT]] : tensor<?x?x?xf32>) -> tensor<?x?x?xf32>
+// TRANSPOSE-A:     %[[A_DIM2:.*]] = tensor.dim %[[A]], %[[C2]] : tensor<?x?x?xf32>
+// TRANSPOSE-A:     %[[A_TRANSP_INIT:.*]] = tensor.empty(%[[A_DIM0]], %[[A_DIM2]], %[[A_DIM1]]) : tensor<?x?x?xf32>
+// TRANSPOSE-A:     %[[A_TRANSP:.*]] = linalg.transpose ins(%[[A]] : tensor<?x?x?xf32>) outs(%[[A_TRANSP_INIT]] : tensor<?x?x?xf32>) permutation = [0, 2, 1]
+// TRANSPOSE-A:     %[[C:.*]] = linalg.batch_matmul_transpose_a ins(%[[A_TRANSP]], %[[B]] : tensor<?x?x?xf32>, tensor<?x?x?xf32>) outs(%[[C_ZERO]] : tensor<?x?x?xf32>) -> tensor<?x?x?xf32>
+// TRANSPOSE-B:     %[[B_DIM0:.*]] = tensor.dim %[[B]], %[[C0]] : tensor<?x?x?xf32>
+// TRANSPOSE-B:     %[[B_DIM1:.*]] = tensor.dim %[[B]], %[[C1]] : tensor<?x?x?xf32>
+// TRANSPOSE-B:     %[[B_TRANSP_INIT:.*]] = tensor.empty(%[[B_DIM0]], %[[B_DIM2]], %[[B_DIM1]]) : tensor<?x?x?xf32>
+// TRANSPOSE-B:     %[[B_TRANSP:.*]] = linalg.transpose ins(%[[B]] : tensor<?x?x?xf32>) outs(%[[B_TRANSP_INIT]] : tensor<?x?x?xf32>) permutation = [0, 2, 1]
+// TRANSPOSE-B:     %[[C:.*]] = linalg.batch_matmul_transpose_b ins(%[[A]], %[[B_TRANSP]] : tensor<?x?x?xf32>, tensor<?x?x?xf32>) outs(%[[C_ZERO]] : tensor<?x?x?xf32>) -> tensor<?x?x?xf32>
+// CHECK:           return %[[C]] : tensor<?x?x?xf32>
+// CHECK:         }
+func.func @batch_matmul_dynamic(%A: tensor<?x?x?xf32>, %B: tensor<?x?x?xf32>) -> (tensor<?x?x?xf32>) {
+  %cst = arith.constant 0.0 : f32
+  %c0 = arith.constant 0 : index
+  %c1 = arith.constant 1 : index
+  %c2 = arith.constant 2 : index
+  %d0 = tensor.dim %A, %c0 : tensor<?x?x?xf32>
+  %d1 = tensor.dim %A, %c1 : tensor<?x?x?xf32>
+  %d2 = tensor.dim %B, %c2 : tensor<?x?x?xf32>
+  %init = tensor.empty(%d0, %d1, %d2) : tensor<?x?x?xf32>
+  %C = linalg.fill ins(%cst : f32) outs(%init : tensor<?x?x?xf32>) -> tensor<?x?x?xf32>
+  %0 = linalg.batch_matmul ins(%A, %B : tensor<?x?x?xf32>, tensor<?x?x?xf32>) outs(%C : tensor<?x?x?xf32>) -> tensor<?x?x?xf32>
+  return %0 : tensor<?x?x?xf32>
+}
+
+//-----
+
+// CHECK-LABEL:   func.func @batch_matmul_mixed(
+// CHECK-SAME:                                  %[[A:.*]]: tensor<2x?x8xf32>,
+// CHECK-SAME:                                  %[[B:.*]]: tensor<2x8x16xf32>) -> tensor<2x?x16xf32> {
+// CHECK:           %[[C0_F32:.*]] = arith.constant 0.000000e+00 : f32
+// CHECK:           %[[C1:.*]] = arith.constant 1 : index
+// CHECK:           %[[A_DIM1:.*]] = tensor.dim %[[A]], %[[C1]] : tensor<2x?x8xf32>
+// CHECK:           %[[C_INIT:.*]] = tensor.empty(%[[A_DIM1]]) : tensor<2x?x16xf32>
+// CHECK:           %[[C_ZERO:.*]] = linalg.fill ins(%[[C0_F32]] : f32) outs(%[[C_INIT]] : tensor<2x?x16xf32>) -> tensor<2x?x16xf32>
+// TRANSPOSE-A:     %[[A_TRANSP_INIT:.*]] = tensor.empty(%[[A_DIM1]]) : tensor<2x8x?xf32>
+// TRANSPOSE-A:     %[[A_TRANSP:.*]] = linalg.transpose ins(%[[A]] : tensor<2x?x8xf32>) outs(%[[A_TRANSP_INIT]] : tensor<2x8x?xf32>) permutation = [0, 2, 1]
+// TRANSPOSE-A:     %[[B0:.*]] = linalg.batch_matmul_transpose_a ins(%[[A_TRANSP]], %[[B]] : tensor<2x8x?xf32>, tensor<2x8x16xf32>) outs(%[[C_ZERO]] : tensor<2x?x16xf32>) -> tensor<2x?x16xf32>
+// TRANSPOSE-B:     %[[B_TRANSP_INIT:.*]] = tensor.empty() : tensor<2x16x8xf32>
+// TRANSPOSE-B:     %[[B_TRANSP:.*]] = linalg.transpose ins(%[[B]] : tensor<2x8x16xf32>) outs(%[[B_TRANSP_INIT]] : tensor<2x16x8xf32>) permutation = [0, 2, 1]
+// TRANSPOSE-B:     %[[B0:.*]] = linalg.batch_matmul_transpose_b ins(%[[A]], %[[B_TRANSP]] : tensor<2x?x8xf32>, tensor<2x16x8xf32>) outs(%[[C_ZERO]] : tensor<2x?x16xf32>) -> tensor<2x?x16xf32>
+// CHECK:           return %[[B0]] : tensor<2x?x16xf32>
+// CHECK:         }
+func.func @batch_matmul_mixed(%A: tensor<2x?x8xf32>, %B: tensor<2x8x16xf32>) -> (tensor<2x?x16xf32>) {
+  %cst = arith.constant 0.0 : f32
+  %c0 = arith.constant 0 : index
+  %c1 = arith.constant 1 : index
+  %d1 = tensor.dim %A, %c1 : tensor<2x?x8xf32>
+  %init = tensor.empty(%d1) : tensor<2x?x16xf32>
+  %C = linalg.fill ins(%cst : f32) outs(%init : tensor<2x?x16xf32>) -> tensor<2x?x16xf32>
+  %0 = linalg.batch_matmul ins(%A, %B : tensor<2x?x8xf32>, tensor<2x8x16xf32>) outs(%C : tensor<2x?x16xf32>) -> tensor<2x?x16xf32>
+  return %0 : tensor<2x?x16xf32>
+}
diff --git a/mlir/test/Dialect/MemRef/emulate-narrow-type.mlir b/mlir/test/Dialect/MemRef/emulate-narrow-type.mlir
index fd37b7ff0a27..435dcc944778 100644
--- a/mlir/test/Dialect/MemRef/emulate-narrow-type.mlir
+++ b/mlir/test/Dialect/MemRef/emulate-narrow-type.mlir
@@ -430,3 +430,23 @@ func.func @rank_zero_memref_store(%arg0: i4) -> () {
 //       CHECK32:   %[[EXTUI:.+]] = arith.extui %[[ARG0]] : i4 to i32
 //       CHECK32:   %[[WRITE_RMW:.+]] = memref.atomic_rmw assign %[[EXTUI]], %[[ALLOC]][] : (i32, memref<i32>) -> i32
 //       CHECK32:   return
+
+// -----
+
+func.func @memref_collapse_shape_i4(%idx0 : index, %idx1 : index) -> i4 {
+  %arr = memref.alloc() : memref<32x8x128xi4>
+  %collapse = memref.collapse_shape %arr[[0, 1], [2]] : memref<32x8x128xi4> into memref<256x128xi4>
+  %1 = memref.load %collapse[%idx0, %idx1] : memref<256x128xi4>
+  return %1 : i4
+}
+
+// CHECK-LABEL:   func.func @memref_collapse_shape_i4(
+//       CHECK:     %[[ALLOC:.*]] = memref.alloc() : memref<16384xi8>
+//   CHECK-NOT:     memref.collapse_shape
+//       CHECK:     memref.load %[[ALLOC]][%{{.*}}] : memref<16384xi8>
+
+// CHECK32-LABEL:   func.func @memref_collapse_shape_i4(
+//       CHECK32:     %[[ALLOC:.*]] = memref.alloc() : memref<4096xi32>
+//   CHECK32-NOT:     memref.collapse_shape
+//       CHECK32:     memref.load %[[ALLOC]][%{{.*}}] : memref<4096xi32>
+
diff --git a/mlir/test/Dialect/MemRef/expand-strided-metadata.mlir b/mlir/test/Dialect/MemRef/expand-strided-metadata.mlir
index 28b700430059..0705b30ca45d 100644
--- a/mlir/test/Dialect/MemRef/expand-strided-metadata.mlir
+++ b/mlir/test/Dialect/MemRef/expand-strided-metadata.mlir
@@ -1513,4 +1513,26 @@ func.func @zero_sized_memred(%arg0: f32) -> (memref<f16, 3>, index,index,index)
     %sizes, %strides :
       memref<f16,3>, index,
       index, index
-}
\ No newline at end of file
+}
+
+// -----
+
+func.func @extract_strided_metadata_of_collapse_shape(%base: memref<5x4xf32>)
+    -> (memref<f32>, index, index, index) {
+
+  %collapse = memref.collapse_shape %base[[0, 1]] :
+    memref<5x4xf32> into memref<20xf32>
+
+  %base_buffer, %offset, %size, %stride = memref.extract_strided_metadata %collapse :
+    memref<20xf32> -> memref<f32>, index, index, index
+
+  return %base_buffer, %offset, %size, %stride :
+    memref<f32>, index, index, index
+}
+
+// CHECK-LABEL:  func @extract_strided_metadata_of_collapse_shape
+//   CHECK-DAG:    %[[OFFSET:.*]] = arith.constant 0 : index
+//   CHECK-DAG:    %[[SIZE:.*]] = arith.constant 20 : index
+//   CHECK-DAG:    %[[STEP:.*]] = arith.constant 1 : index
+//       CHECK:    %[[BASE:.*]], %{{.*}}, %{{.*}}, %{{.*}} = memref.extract_strided_metadata
+//       CHECK:    return %[[BASE]], %[[OFFSET]], %[[SIZE]], %[[STEP]] : memref<f32>, index, index, index
diff --git a/mlir/test/Dialect/OpenMP/invalid.mlir b/mlir/test/Dialect/OpenMP/invalid.mlir
index 2f24dce4233e..e329b3010017 100644
--- a/mlir/test/Dialect/OpenMP/invalid.mlir
+++ b/mlir/test/Dialect/OpenMP/invalid.mlir
@@ -149,50 +149,74 @@ func.func @invalid_parent(%lb : index, %ub : index, %step : index) {
 // -----
 
 func.func @invalid_wrapper(%lb : index, %ub : index, %step : index) {
-  // TODO Remove induction variables from omp.wsloop.
-  omp.wsloop for (%iv) : index = (%lb) to (%ub) step (%step) {
+  omp.parallel {
     %0 = arith.constant 0 : i32
     // expected-error@+1 {{op expects parent op to be a valid loop wrapper}}
     omp.loop_nest (%iv2) : index = (%lb) to (%ub) step (%step) {
       omp.yield
     }
-    omp.yield
+    omp.terminator
   }
 }
 
 // -----
 
 func.func @type_mismatch(%lb : index, %ub : index, %step : index) {
-  // TODO Remove induction variables from omp.wsloop.
-  omp.wsloop for (%iv) : index = (%lb) to (%ub) step (%step) {
+  omp.wsloop {
     // expected-error@+1 {{range argument type does not match corresponding IV type}}
     "omp.loop_nest" (%lb, %ub, %step) ({
     ^bb0(%iv2: i32):
       omp.yield
     }) : (index, index, index) -> ()
-    omp.yield
+    omp.terminator
   }
 }
 
 // -----
 
 func.func @iv_number_mismatch(%lb : index, %ub : index, %step : index) {
-  // TODO Remove induction variables from omp.wsloop.
-  omp.wsloop for (%iv) : index = (%lb) to (%ub) step (%step) {
+  omp.wsloop {
     // expected-error@+1 {{number of range arguments and IVs do not match}}
     "omp.loop_nest" (%lb, %ub, %step) ({
     ^bb0(%iv1 : index, %iv2 : index):
       omp.yield
     }) : (index, index, index) -> ()
-    omp.yield
+    omp.terminator
+  }
+}
+
+// -----
+
+func.func @no_wrapper(%lb : index, %ub : index, %step : index) {
+  // expected-error @below {{op must be a loop wrapper}}
+  omp.wsloop {
+    %0 = arith.constant 0 : i32
+    omp.loop_nest (%iv) : index = (%lb) to (%ub) step (%step) {
+      omp.yield
+    }
+    omp.terminator
+  }
+}
+
+// -----
+
+func.func @invalid_nested_wrapper(%lb : index, %ub : index, %step : index) {
+  // expected-error @below {{only supported nested wrapper is 'omp.simd'}}
+  omp.wsloop {
+    omp.distribute {
+      omp.loop_nest (%iv) : index = (%lb) to (%ub) step (%step) {
+        omp.yield
+      }
+      omp.terminator
+    }
+    omp.terminator
   }
 }
 
 // -----
 
 func.func @no_loops(%lb : index, %ub : index, %step : index) {
-  // TODO Remove induction variables from omp.wsloop.
-  omp.wsloop for (%iv) : index = (%lb) to (%ub) step (%step) {
+  omp.wsloop {
     // expected-error@+1 {{op must represent at least one loop}}
     "omp.loop_nest" () ({
     ^bb0():
@@ -205,10 +229,12 @@ func.func @no_loops(%lb : index, %ub : index, %step : index) {
 // -----
 
 func.func @inclusive_not_a_clause(%lb : index, %ub : index, %step : index) {
-  // expected-error @below {{expected 'for'}}
-  omp.wsloop nowait inclusive
-  for (%iv) : index = (%lb) to (%ub) step (%step) {
-    omp.yield
+  // expected-error @below {{expected '{'}}
+  omp.wsloop nowait inclusive {
+    omp.loop_nest (%iv) : index = (%lb) to (%ub) step (%step) {
+      omp.yield
+    }
+    omp.terminator
   }
 }
 
@@ -216,39 +242,47 @@ func.func @inclusive_not_a_clause(%lb : index, %ub : index, %step : index) {
 
 func.func @order_value(%lb : index, %ub : index, %step : index) {
   // expected-error @below {{invalid clause value: 'default'}}
-  omp.wsloop order(default)
-  for (%iv) : index = (%lb) to (%ub) step (%step) {
-    omp.yield
+  omp.wsloop order(default) {
+    omp.loop_nest (%iv) : index = (%lb) to (%ub) step (%step) {
+      omp.yield
+    }
+    omp.terminator
   }
 }
 
 // -----
 
 func.func @if_not_allowed(%lb : index, %ub : index, %step : index, %bool_var : i1) {
-  // expected-error @below {{expected 'for'}}
-  omp.wsloop if(%bool_var: i1)
-  for (%iv) : index = (%lb) to (%ub) step (%step) {
-    omp.yield
+  // expected-error @below {{expected '{'}}
+  omp.wsloop if(%bool_var: i1) {
+    omp.loop_nest (%iv) : index = (%lb) to (%ub) step (%step) {
+      omp.yield
+    }
+    omp.terminator
   }
 }
 
 // -----
 
 func.func @num_threads_not_allowed(%lb : index, %ub : index, %step : index, %int_var : i32) {
-  // expected-error @below {{expected 'for'}}
-  omp.wsloop num_threads(%int_var: i32)
-  for (%iv) : index = (%lb) to (%ub) step (%step) {
-    omp.yield
+  // expected-error @below {{expected '{'}}
+  omp.wsloop num_threads(%int_var: i32) {
+    omp.loop_nest (%iv) : index = (%lb) to (%ub) step (%step) {
+      omp.yield
+    }
+    omp.terminator
   }
 }
 
 // -----
 
 func.func @proc_bind_not_allowed(%lb : index, %ub : index, %step : index) {
-  // expected-error @below {{expected 'for'}}
-  omp.wsloop proc_bind(close)
-  for (%iv) : index = (%lb) to (%ub) step (%step) {
-    omp.yield
+  // expected-error @below {{expected '{'}}
+  omp.wsloop proc_bind(close) {
+    omp.loop_nest (%iv) : index = (%lb) to (%ub) step (%step) {
+      omp.yield
+    }
+    omp.terminator
   }
 }
 
@@ -256,9 +290,11 @@ func.func @proc_bind_not_allowed(%lb : index, %ub : index, %step : index) {
 
 llvm.func @test_omp_wsloop_dynamic_bad_modifier(%lb : i64, %ub : i64, %step : i64) -> () {
   // expected-error @+1 {{unknown modifier type: ginandtonic}}
-  omp.wsloop schedule(dynamic, ginandtonic)
-  for (%iv) : i64 = (%lb) to (%ub) step (%step) {
-    omp.yield
+  omp.wsloop schedule(dynamic, ginandtonic) {
+    omp.loop_nest (%iv) : i64 = (%lb) to (%ub) step (%step) {
+      omp.yield
+    }
+    omp.terminator
   }
   llvm.return
 }
@@ -267,9 +303,11 @@ llvm.func @test_omp_wsloop_dynamic_bad_modifier(%lb : i64, %ub : i64, %step : i6
 
 llvm.func @test_omp_wsloop_dynamic_many_modifier(%lb : i64, %ub : i64, %step : i64) -> () {
   // expected-error @+1 {{unexpected modifier(s)}}
-  omp.wsloop schedule(dynamic, monotonic, monotonic, monotonic)
-  for (%iv) : i64 = (%lb) to (%ub) step (%step) {
-    omp.yield
+  omp.wsloop schedule(dynamic, monotonic, monotonic, monotonic) {
+    omp.loop_nest (%iv) : i64 = (%lb) to (%ub) step (%step) {
+      omp.yield
+    }
+    omp.terminator
   }
   llvm.return
 }
@@ -278,9 +316,11 @@ llvm.func @test_omp_wsloop_dynamic_many_modifier(%lb : i64, %ub : i64, %step : i
 
 llvm.func @test_omp_wsloop_dynamic_wrong_modifier(%lb : i64, %ub : i64, %step : i64) -> () {
   // expected-error @+1 {{incorrect modifier order}}
-  omp.wsloop schedule(dynamic, simd, monotonic)
-  for (%iv) : i64 = (%lb) to (%ub) step (%step) {
-    omp.yield
+  omp.wsloop schedule(dynamic, simd, monotonic) {
+    omp.loop_nest (%iv) : i64 = (%lb) to (%ub) step (%step) {
+      omp.yield
+    }
+    omp.terminator
   }
   llvm.return
 }
@@ -289,9 +329,11 @@ llvm.func @test_omp_wsloop_dynamic_wrong_modifier(%lb : i64, %ub : i64, %step :
 
 llvm.func @test_omp_wsloop_dynamic_wrong_modifier2(%lb : i64, %ub : i64, %step : i64) -> () {
   // expected-error @+1 {{incorrect modifier order}}
-  omp.wsloop schedule(dynamic, monotonic, monotonic)
-  for (%iv) : i64 = (%lb) to (%ub) step (%step) {
-    omp.yield
+  omp.wsloop schedule(dynamic, monotonic, monotonic) {
+    omp.loop_nest (%iv) : i64 = (%lb) to (%ub) step (%step) {
+      omp.yield
+    }
+    omp.terminator
   }
   llvm.return
 }
@@ -300,9 +342,11 @@ llvm.func @test_omp_wsloop_dynamic_wrong_modifier2(%lb : i64, %ub : i64, %step :
 
 llvm.func @test_omp_wsloop_dynamic_wrong_modifier3(%lb : i64, %ub : i64, %step : i64) -> () {
   // expected-error @+1 {{incorrect modifier order}}
-  omp.wsloop schedule(dynamic, simd, simd)
-  for (%iv) : i64 = (%lb) to (%ub) step (%step) {
-    omp.yield
+  omp.wsloop schedule(dynamic, simd, simd) {
+    omp.loop_nest (%iv) : i64 = (%lb) to (%ub) step (%step) {
+      omp.yield
+    }
+    omp.terminator
   }
   llvm.return
 }
@@ -601,11 +645,13 @@ func.func @foo(%lb : index, %ub : index, %step : index) {
   %1 = llvm.alloca %c1 x i32 : (i32) -> !llvm.ptr
 
   // expected-error @below {{expected symbol reference @foo to point to a reduction declaration}}
-  omp.wsloop reduction(@foo %0 -> %prv : !llvm.ptr)
-  for (%iv) : index = (%lb) to (%ub) step (%step) {
-    %2 = arith.constant 2.0 : f32
-    omp.reduction %2, %1 : f32, !llvm.ptr
-    omp.yield
+  omp.wsloop reduction(@foo %0 -> %prv : !llvm.ptr) {
+    omp.loop_nest (%iv) : index = (%lb) to (%ub) step (%step) {
+      %2 = arith.constant 2.0 : f32
+      omp.reduction %2, %1 : f32, !llvm.ptr
+      omp.yield
+    }
+    omp.terminator
   }
   return
 }
@@ -629,11 +675,13 @@ func.func @foo(%lb : index, %ub : index, %step : index) {
   %0 = llvm.alloca %c1 x i32 : (i32) -> !llvm.ptr
 
   // expected-error @below {{accumulator variable used more than once}}
-  omp.wsloop reduction(@add_f32 %0 -> %prv : !llvm.ptr, @add_f32 %0 -> %prv1 : !llvm.ptr)
-  for (%iv) : index = (%lb) to (%ub) step (%step) {
-    %2 = arith.constant 2.0 : f32
-    omp.reduction %2, %0 : f32, !llvm.ptr
-    omp.yield
+  omp.wsloop reduction(@add_f32 %0 -> %prv : !llvm.ptr, @add_f32 %0 -> %prv1 : !llvm.ptr) {
+    omp.loop_nest (%iv) : index = (%lb) to (%ub) step (%step) {
+      %2 = arith.constant 2.0 : f32
+      omp.reduction %2, %0 : f32, !llvm.ptr
+      omp.yield
+    }
+    omp.terminator
   }
   return
 }
@@ -662,11 +710,13 @@ func.func @foo(%lb : index, %ub : index, %step : index, %mem : memref<1xf32>) {
   %c1 = arith.constant 1 : i32
 
   // expected-error @below {{expected accumulator ('memref<1xf32>') to be the same type as reduction declaration ('!llvm.ptr')}}
-  omp.wsloop reduction(@add_f32 %mem -> %prv : memref<1xf32>)
-  for (%iv) : index = (%lb) to (%ub) step (%step) {
-    %2 = arith.constant 2.0 : f32
-    omp.reduction %2, %mem : f32, memref<1xf32>
-    omp.yield
+  omp.wsloop reduction(@add_f32 %mem -> %prv : memref<1xf32>) {
+    omp.loop_nest (%iv) : index = (%lb) to (%ub) step (%step) {
+      %2 = arith.constant 2.0 : f32
+      omp.reduction %2, %mem : f32, memref<1xf32>
+      omp.yield
+    }
+    omp.terminator
   }
   return
 }
@@ -698,60 +748,112 @@ omp.critical.declare @mutex hint(invalid_hint)
 
 // -----
 
-func.func @omp_ordered1(%arg1 : i32, %arg2 : i32, %arg3 : i32) -> () {
-  omp.wsloop ordered(1)
-  for (%0) : i32 = (%arg1) to (%arg2) step (%arg3) {
-    // expected-error @below {{ordered region must be closely nested inside a worksharing-loop region with an ordered clause without parameter present}}
-    omp.ordered.region {
-      omp.terminator
+func.func @omp_ordered_region1(%x : i32) -> () {
+  omp.distribute {
+    omp.loop_nest (%i) : i32 = (%x) to (%x) step (%x) {
+      // expected-error @below {{op must be nested inside of a worksharing, simd or worksharing simd loop}}
+      omp.ordered.region {
+        omp.terminator
+      }
+      omp.yield
     }
-    omp.yield
+    omp.terminator
   }
   return
 }
 
 // -----
 
-func.func @omp_ordered2(%arg1 : i32, %arg2 : i32, %arg3 : i32) -> () {
-  omp.wsloop for (%0) : i32 = (%arg1) to (%arg2) step (%arg3) {
-    // expected-error @below {{ordered region must be closely nested inside a worksharing-loop region with an ordered clause without parameter present}}
-    omp.ordered.region {
-      omp.terminator
+func.func @omp_ordered_region2(%x : i32) -> () {
+  omp.wsloop {
+    omp.loop_nest (%i) : i32 = (%x) to (%x) step (%x) {
+      // expected-error @below {{the enclosing worksharing-loop region must have an ordered clause}}
+      omp.ordered.region {
+        omp.terminator
+      }
+      omp.yield
     }
-    omp.yield
+    omp.terminator
   }
   return
 }
 
 // -----
 
-func.func @omp_ordered3(%vec0 : i64) -> () {
-  // expected-error @below {{ordered depend directive must be closely nested inside a worksharing-loop with ordered clause with parameter present}}
+func.func @omp_ordered_region3(%x : i32) -> () {
+  omp.wsloop ordered(1) {
+    omp.loop_nest (%i) : i32 = (%x) to (%x) step (%x) {
+      // expected-error @below {{the enclosing loop's ordered clause must not have a parameter present}}
+      omp.ordered.region {
+        omp.terminator
+      }
+      omp.yield
+    }
+    omp.terminator
+  }
+  return
+}
+
+// -----
+
+func.func @omp_ordered1(%vec0 : i64) -> () {
+  // expected-error @below {{op must be nested inside of a loop}}
   omp.ordered depend_type(dependsink) depend_vec(%vec0 : i64) {num_loops_val = 1 : i64}
   return
 }
 
 // -----
 
-func.func @omp_ordered4(%arg1 : i32, %arg2 : i32, %arg3 : i32, %vec0 : i64) -> () {
-  omp.wsloop ordered(0)
-  for (%0) : i32 = (%arg1) to (%arg2) step (%arg3) {
-    // expected-error @below {{ordered depend directive must be closely nested inside a worksharing-loop with ordered clause with parameter present}}
-    omp.ordered depend_type(dependsink) depend_vec(%vec0 : i64) {num_loops_val = 1 : i64}
+func.func @omp_ordered2(%arg1 : i32, %arg2 : i32, %arg3 : i32, %vec0 : i64) -> () {
+  omp.distribute {
+    omp.loop_nest (%0) : i32 = (%arg1) to (%arg2) step (%arg3) {
+      // expected-error @below {{op must be nested inside of a worksharing, simd or worksharing simd loop}}
+      omp.ordered depend_type(dependsink) depend_vec(%vec0 : i64) {num_loops_val = 1 : i64}
+      omp.yield
+    }
+    omp.terminator
+  }
+  return
+}
 
-    omp.yield
+// -----
+
+func.func @omp_ordered3(%arg1 : i32, %arg2 : i32, %arg3 : i32, %vec0 : i64) -> () {
+  omp.wsloop {
+    omp.loop_nest (%0) : i32 = (%arg1) to (%arg2) step (%arg3) {
+      // expected-error @below {{the enclosing worksharing-loop region must have an ordered clause}}
+      omp.ordered depend_type(dependsink) depend_vec(%vec0 : i64) {num_loops_val = 1 : i64}
+      omp.yield
+    }
+    omp.terminator
   }
   return
 }
+
 // -----
 
-func.func @omp_ordered5(%arg1 : i32, %arg2 : i32, %arg3 : i32, %vec0 : i64, %vec1 : i64) -> () {
-  omp.wsloop ordered(1)
-  for (%0) : i32 = (%arg1) to (%arg2) step (%arg3) {
-    // expected-error @below {{number of variables in depend clause does not match number of iteration variables in the doacross loop}}
-    omp.ordered depend_type(dependsource) depend_vec(%vec0, %vec1 : i64, i64) {num_loops_val = 2 : i64}
+func.func @omp_ordered4(%arg1 : i32, %arg2 : i32, %arg3 : i32, %vec0 : i64) -> () {
+  omp.wsloop ordered(0) {
+    omp.loop_nest (%0) : i32 = (%arg1) to (%arg2) step (%arg3) {
+      // expected-error @below {{the enclosing loop's ordered clause must have a parameter present}}
+      omp.ordered depend_type(dependsink) depend_vec(%vec0 : i64) {num_loops_val = 1 : i64}
+      omp.yield
+    }
+    omp.terminator
+  }
+  return
+}
 
-    omp.yield
+// -----
+
+func.func @omp_ordered5(%arg1 : i32, %arg2 : i32, %arg3 : i32, %vec0 : i64, %vec1 : i64) -> () {
+  omp.wsloop ordered(1) {
+    omp.loop_nest (%0) : i32 = (%arg1) to (%arg2) step (%arg3) {
+      // expected-error @below {{number of variables in depend clause does not match number of iteration variables in the doacross loop}}
+      omp.ordered depend_type(dependsource) depend_vec(%vec0, %vec1 : i64, i64) {num_loops_val = 2 : i64}
+      omp.yield
+    }
+    omp.terminator
   }
   return
 }
@@ -1590,11 +1692,13 @@ func.func @omp_cancel2() {
 // -----
 
 func.func @omp_cancel3(%arg1 : i32, %arg2 : i32, %arg3 : i32) -> () {
-  omp.wsloop nowait
-    for (%0) : i32 = (%arg1) to (%arg2) step (%arg3) {
-    // expected-error @below {{A worksharing construct that is canceled must not have a nowait clause}}
-    omp.cancel cancellation_construct_type(loop)
-    // CHECK: omp.terminator
+  omp.wsloop nowait {
+    omp.loop_nest (%0) : i32 = (%arg1) to (%arg2) step (%arg3) {
+      // expected-error @below {{A worksharing construct that is canceled must not have a nowait clause}}
+      omp.cancel cancellation_construct_type(loop)
+      // CHECK: omp.yield
+      omp.yield
+    }
     omp.terminator
   }
   return
@@ -1603,11 +1707,13 @@ func.func @omp_cancel3(%arg1 : i32, %arg2 : i32, %arg3 : i32) -> () {
 // -----
 
 func.func @omp_cancel4(%arg1 : i32, %arg2 : i32, %arg3 : i32) -> () {
-  omp.wsloop ordered(1)
-    for (%0) : i32 = (%arg1) to (%arg2) step (%arg3) {
-    // expected-error @below {{A worksharing construct that is canceled must not have an ordered clause}}
-    omp.cancel cancellation_construct_type(loop)
-    // CHECK: omp.terminator
+  omp.wsloop ordered(1) {
+    omp.loop_nest (%0) : i32 = (%arg1) to (%arg2) step (%arg3) {
+      // expected-error @below {{A worksharing construct that is canceled must not have an ordered clause}}
+      omp.cancel cancellation_construct_type(loop)
+      // CHECK: omp.yield
+      omp.yield
+    }
     omp.terminator
   }
   return
@@ -2143,4 +2249,4 @@ func.func @undefined_privatizer(%arg0: !llvm.ptr) {
       omp.terminator
     }) : (!llvm.ptr) -> ()
   return
-}
+}
\ No newline at end of file
diff --git a/mlir/test/Dialect/OpenMP/ops.mlir b/mlir/test/Dialect/OpenMP/ops.mlir
index c10fc88211c3..a012588f0b55 100644
--- a/mlir/test/Dialect/OpenMP/ops.mlir
+++ b/mlir/test/Dialect/OpenMP/ops.mlir
@@ -90,10 +90,9 @@ func.func @omp_parallel(%data_var : memref<i32>, %if_cond : i1, %num_threads : i
     // CHECK-NEXT: omp.parallel
     omp.parallel {
       // CHECK-NEXT: omp.wsloop
-      // TODO Remove induction variables from omp.wsloop.
-      omp.wsloop for (%iv) : index = (%idx) to (%idx) step (%idx) {
+      omp.wsloop {
         // CHECK-NEXT: omp.loop_nest
-        omp.loop_nest (%iv2) : index = (%idx) to (%idx) step (%idx) {
+        omp.loop_nest (%iv) : index = (%idx) to (%idx) step (%idx) {
           omp.yield
         }
         omp.terminator
@@ -153,49 +152,45 @@ func.func @omp_parallel_pretty(%data_var : memref<i32>, %if_cond : i1, %num_thre
 
 // CHECK-LABEL: omp_loop_nest
 func.func @omp_loop_nest(%lb : index, %ub : index, %step : index) -> () {
-  // TODO Remove induction variables from omp.wsloop.
-  omp.wsloop for (%iv) : index = (%lb) to (%ub) step (%step) {
+  omp.wsloop {
     // CHECK: omp.loop_nest
     // CHECK-SAME: (%{{.*}}) : index =
     // CHECK-SAME: (%{{.*}}) to (%{{.*}}) step (%{{.*}})
     "omp.loop_nest" (%lb, %ub, %step) ({
-    ^bb0(%iv2: index):
+    ^bb0(%iv: index):
       omp.yield
     }) : (index, index, index) -> ()
-    omp.yield
+    omp.terminator
   }
 
-  // TODO Remove induction variables from omp.wsloop.
-  omp.wsloop for (%iv) : index = (%lb) to (%ub) step (%step) {
+  omp.wsloop {
     // CHECK: omp.loop_nest
     // CHECK-SAME: (%{{.*}}) : index =
     // CHECK-SAME: (%{{.*}}) to (%{{.*}}) inclusive step (%{{.*}})
     "omp.loop_nest" (%lb, %ub, %step) ({
-    ^bb0(%iv2: index):
+    ^bb0(%iv: index):
       omp.yield
     }) {inclusive} : (index, index, index) -> ()
-    omp.yield
+    omp.terminator
   }
 
-  // TODO Remove induction variables from omp.wsloop.
-  omp.wsloop for (%iv) : index = (%lb) to (%ub) step (%step) {
+  omp.wsloop {
     // CHECK: omp.loop_nest
     // CHECK-SAME: (%{{.*}}, %{{.*}}) : index =
     // CHECK-SAME: (%{{.*}}, %{{.*}}) to (%{{.*}}, %{{.*}}) step (%{{.*}}, %{{.*}})
     "omp.loop_nest" (%lb, %lb, %ub, %ub, %step, %step) ({
-    ^bb0(%iv2: index, %iv3: index):
+    ^bb0(%iv: index, %iv3: index):
       omp.yield
     }) : (index, index, index, index, index, index) -> ()
-    omp.yield
+    omp.terminator
   }
 
-  // TODO Remove induction variables from omp.wsloop.
-  omp.wsloop for (%iv) : index = (%lb) to (%ub) step (%step) {
+  omp.wsloop {
     // CHECK: omp.loop_nest
     // CHECK-SAME: (%{{.*}}) : index =
     // CHECK-SAME: (%{{.*}}) to (%{{.*}}) step (%{{.*}})
     "omp.loop_nest" (%lb, %ub, %step) ({
-    ^bb0(%iv2: index):
+    ^bb0(%iv: index):
       // CHECK: test.op1
       "test.op1"(%lb) : (index) -> ()
       // CHECK: test.op2
@@ -203,7 +198,7 @@ func.func @omp_loop_nest(%lb : index, %ub : index, %step : index) -> () {
       // CHECK: omp.yield
       omp.yield
     }) : (index, index, index) -> ()
-    omp.yield
+    omp.terminator
   }
 
   return
@@ -211,45 +206,41 @@ func.func @omp_loop_nest(%lb : index, %ub : index, %step : index) -> () {
 
 // CHECK-LABEL: omp_loop_nest_pretty
 func.func @omp_loop_nest_pretty(%lb : index, %ub : index, %step : index) -> () {
-  // TODO Remove induction variables from omp.wsloop.
-  omp.wsloop for (%iv) : index = (%lb) to (%ub) step (%step) {
+  omp.wsloop {
     // CHECK: omp.loop_nest
     // CHECK-SAME: (%{{.*}}) : index =
     // CHECK-SAME: (%{{.*}}) to (%{{.*}}) step (%{{.*}})
-    omp.loop_nest (%iv2) : index = (%lb) to (%ub) step (%step) {
+    omp.loop_nest (%iv) : index = (%lb) to (%ub) step (%step) {
       omp.yield
     }
-    omp.yield
+    omp.terminator
   }
 
-  // TODO Remove induction variables from omp.wsloop.
-  omp.wsloop for (%iv) : index = (%lb) to (%ub) step (%step) {
+  omp.wsloop {
     // CHECK: omp.loop_nest
     // CHECK-SAME: (%{{.*}}) : index =
     // CHECK-SAME: (%{{.*}}) to (%{{.*}}) inclusive step (%{{.*}})
-    omp.loop_nest (%iv2) : index = (%lb) to (%ub) inclusive step (%step) {
+    omp.loop_nest (%iv) : index = (%lb) to (%ub) inclusive step (%step) {
       omp.yield
     }
-    omp.yield
+    omp.terminator
   }
 
-  // TODO Remove induction variables from omp.wsloop.
-  omp.wsloop for (%iv) : index = (%lb) to (%ub) step (%step) {
+  omp.wsloop {
     // CHECK: omp.loop_nest
     // CHECK-SAME: (%{{.*}}) : index =
     // CHECK-SAME: (%{{.*}}, %{{.*}}) to (%{{.*}}, %{{.*}}) step (%{{.*}}, %{{.*}})
-    omp.loop_nest (%iv2, %iv3) : index = (%lb, %lb) to (%ub, %ub) step (%step, %step) {
+    omp.loop_nest (%iv1, %iv2) : index = (%lb, %lb) to (%ub, %ub) step (%step, %step) {
       omp.yield
     }
-    omp.yield
+    omp.terminator
   }
 
-  // TODO Remove induction variables from omp.wsloop.
-  omp.wsloop for (%iv) : index = (%lb) to (%ub) step (%step) {
+  omp.wsloop {
     // CHECK: omp.loop_nest
     // CHECK-SAME: (%{{.*}}) : index =
     // CHECK-SAME: (%{{.*}}) to (%{{.*}}) step (%{{.*}})
-    omp.loop_nest (%iv2) : index = (%lb) to (%ub) step (%step)  {
+    omp.loop_nest (%iv) : index = (%lb) to (%ub) step (%step)  {
       // CHECK: test.op1
       "test.op1"(%lb) : (index) -> ()
       // CHECK: test.op2
@@ -257,201 +248,271 @@ func.func @omp_loop_nest_pretty(%lb : index, %ub : index, %step : index) -> () {
       // CHECK: omp.yield
       omp.yield
     }
-    omp.yield
+    omp.terminator
   }
 
   return
 }
 
-// CHECK-LABEL: omp_wsloop
-func.func @omp_wsloop(%lb : index, %ub : index, %step : index, %data_var : memref<i32>, %linear_var : i32, %chunk_var : i32) -> () {
+// CHECK-LABEL: omp_loop_nest_pretty_multi_block
+func.func @omp_loop_nest_pretty_multi_block(%lb : index, %ub : index,
+    %step : index, %data1 : memref<?xi32>, %data2 : memref<?xi32>) -> () {
 
-  // CHECK: omp.wsloop ordered(1)
-  // CHECK-SAME: for (%{{.*}}) : index = (%{{.*}}) to (%{{.*}}) step (%{{.*}})
-  "omp.wsloop" (%lb, %ub, %step) ({
-    ^bb0(%iv: index):
+  omp.wsloop {
+    // CHECK: omp.loop_nest (%{{.*}}) : index = (%{{.*}}) to (%{{.*}}) step (%{{.*}})
+    omp.loop_nest (%iv) : index = (%lb) to (%ub) step (%step) {
+      %1 = "test.payload"(%iv) : (index) -> (i32)
+      cf.br ^bb1(%1: i32)
+    ^bb1(%arg: i32):
+      memref.store %arg, %data1[%iv] : memref<?xi32>
       omp.yield
-  }) {operandSegmentSizes = array<i32: 1,1,1,0,0,0,0>, ordered_val = 1} :
-    (index, index, index) -> ()
+    }
+    omp.terminator
+  }
 
-  // CHECK: omp.wsloop linear(%{{.*}} = %{{.*}} : memref<i32>) schedule(static)
-  // CHECK-SAME: for (%{{.*}}) : index = (%{{.*}}) to (%{{.*}}) step (%{{.*}})
-  "omp.wsloop" (%lb, %ub, %step, %data_var, %linear_var) ({
-    ^bb0(%iv: index):
+  omp.wsloop {
+    // CHECK: omp.loop_nest (%{{.*}}) : index = (%{{.*}}) to (%{{.*}}) step (%{{.*}})
+    omp.loop_nest (%iv) : index = (%lb) to (%ub) step (%step) {
+      %c = "test.condition"(%iv) : (index) -> (i1)
+      %v1 = "test.payload"(%iv) : (index) -> (i32)
+      cf.cond_br %c, ^bb1(%v1: i32), ^bb2(%v1: i32)
+    ^bb1(%arg0: i32):
+      memref.store %arg0, %data1[%iv] : memref<?xi32>
+      cf.br ^bb3
+    ^bb2(%arg1: i32):
+      memref.store %arg1, %data2[%iv] : memref<?xi32>
+      cf.br ^bb3
+    ^bb3:
       omp.yield
-  }) {operandSegmentSizes = array<i32: 1,1,1,1,1,0,0>, schedule_val = #omp<schedulekind static>} :
-    (index, index, index, memref<i32>, i32) -> ()
+    }
+    omp.terminator
+  }
 
-  // CHECK: omp.wsloop linear(%{{.*}} = %{{.*}} : memref<i32>, %{{.*}} = %{{.*}} : memref<i32>) schedule(static)
-  // CHECK-SAME: for (%{{.*}}) : index = (%{{.*}}) to (%{{.*}}) step (%{{.*}})
-  "omp.wsloop" (%lb, %ub, %step, %data_var, %data_var, %linear_var, %linear_var) ({
-    ^bb0(%iv: index):
+  omp.wsloop {
+    // CHECK: omp.loop_nest (%{{.*}}) : index = (%{{.*}}) to (%{{.*}}) step (%{{.*}})
+    omp.loop_nest (%iv) : index = (%lb) to (%ub) step (%step) {
+      %c = "test.condition"(%iv) : (index) -> (i1)
+      %v1 = "test.payload"(%iv) : (index) -> (i32)
+      cf.cond_br %c, ^bb1(%v1: i32), ^bb2(%v1: i32)
+    ^bb1(%arg0: i32):
+      memref.store %arg0, %data1[%iv] : memref<?xi32>
+      omp.yield
+    ^bb2(%arg1: i32):
+      memref.store %arg1, %data2[%iv] : memref<?xi32>
       omp.yield
-  }) {operandSegmentSizes = array<i32: 1,1,1,2,2,0,0>, schedule_val = #omp<schedulekind static>} :
-    (index, index, index, memref<i32>, memref<i32>, i32, i32) -> ()
+    }
+    omp.terminator
+  }
 
-  // CHECK: omp.wsloop linear(%{{.*}} = %{{.*}} : memref<i32>) schedule(dynamic = %{{.*}}) ordered(2)
-  // CHECK-SAME: for (%{{.*}}) : index = (%{{.*}}) to (%{{.*}}) step (%{{.*}})
-  "omp.wsloop" (%lb, %ub, %step, %data_var, %linear_var, %chunk_var) ({
-    ^bb0(%iv: index):
+  return
+}
+
+// CHECK-LABEL: omp_loop_nest_pretty_non_index
+func.func @omp_loop_nest_pretty_non_index(%lb1 : i32, %ub1 : i32, %step1 : i32,
+    %lb2 : i64, %ub2 : i64, %step2 : i64, %data1 : memref<?xi32>,
+    %data2 : memref<?xi64>) -> () {
+
+  omp.wsloop {
+    // CHECK: omp.loop_nest (%{{.*}}) : i32 = (%{{.*}}) to (%{{.*}}) step (%{{.*}})
+    omp.loop_nest (%iv1) : i32 = (%lb1) to (%ub1) step (%step1) {
+      %1 = "test.payload"(%iv1) : (i32) -> (index)
+      cf.br ^bb1(%1: index)
+    ^bb1(%arg1: index):
+      memref.store %iv1, %data1[%arg1] : memref<?xi32>
       omp.yield
-  }) {operandSegmentSizes = array<i32: 1,1,1,1,1,0,1>, schedule_val = #omp<schedulekind dynamic>, ordered_val = 2} :
-    (index, index, index, memref<i32>, i32, i32) -> ()
+    }
+    omp.terminator
+  }
 
-  // CHECK: omp.wsloop schedule(auto) nowait
-  // CHECK-SAME: for (%{{.*}}) : index = (%{{.*}}) to (%{{.*}}) step (%{{.*}})
-  "omp.wsloop" (%lb, %ub, %step) ({
-    ^bb0(%iv: index):
+  omp.wsloop {
+    // CHECK: omp.loop_nest (%{{.*}}) : i64 = (%{{.*}}) to (%{{.*}}) step (%{{.*}})
+    omp.loop_nest (%iv) : i64 = (%lb2) to (%ub2) step (%step2) {
+      %2 = "test.payload"(%iv) : (i64) -> (index)
+      cf.br ^bb1(%2: index)
+    ^bb1(%arg2: index):
+      memref.store %iv, %data2[%arg2] : memref<?xi64>
       omp.yield
-  }) {operandSegmentSizes = array<i32: 1,1,1,0,0,0,0>, nowait, schedule_val = #omp<schedulekind auto>} :
-    (index, index, index) -> ()
+    }
+    omp.terminator
+  }
 
   return
 }
 
-// CHECK-LABEL: omp_wsloop_pretty
-func.func @omp_wsloop_pretty(%lb : index, %ub : index, %step : index, %data_var : memref<i32>, %linear_var : i32, %chunk_var : i32, %chunk_var2 : i16) -> () {
+// CHECK-LABEL: omp_loop_nest_pretty_multiple
+func.func @omp_loop_nest_pretty_multiple(%lb1 : i32, %ub1 : i32, %step1 : i32,
+    %lb2 : i32, %ub2 : i32, %step2 : i32, %data1 : memref<?xi32>) -> () {
 
-  // CHECK: omp.wsloop ordered(2)
-  // CHECK-SAME: for (%{{.*}}) : index = (%{{.*}}) to (%{{.*}}) step (%{{.*}})
-  omp.wsloop ordered(2)
-  for (%iv) : index = (%lb) to (%ub) step (%step) {
-    omp.yield
+  omp.wsloop {
+    // CHECK: omp.loop_nest (%{{.*}}, %{{.*}}) : i32 = (%{{.*}}, %{{.*}}) to (%{{.*}}, %{{.*}}) step (%{{.*}}, %{{.*}})
+    omp.loop_nest (%iv1, %iv2) : i32 = (%lb1, %lb2) to (%ub1, %ub2) step (%step1, %step2) {
+      %1 = "test.payload"(%iv1) : (i32) -> (index)
+      %2 = "test.payload"(%iv2) : (i32) -> (index)
+      memref.store %iv1, %data1[%1] : memref<?xi32>
+      memref.store %iv2, %data1[%2] : memref<?xi32>
+      omp.yield
+    }
+    omp.terminator
   }
 
-  // CHECK: omp.wsloop linear(%{{.*}} = %{{.*}} : memref<i32>) schedule(static)
-  // CHECK-SAME: for (%{{.*}}) : index = (%{{.*}}) to (%{{.*}}) step (%{{.*}})
-  omp.wsloop schedule(static) linear(%data_var = %linear_var : memref<i32>)
-  for (%iv) : index = (%lb) to (%ub) step (%step) {
-    omp.yield
-  }
+  return
+}
 
-  // CHECK: omp.wsloop linear(%{{.*}} = %{{.*}} : memref<i32>) schedule(static = %{{.*}} : i32) ordered(2)
-  // CHECK-SAME: for (%{{.*}}) : index = (%{{.*}}) to (%{{.*}}) step (%{{.*}})
-  omp.wsloop ordered(2) linear(%data_var = %linear_var : memref<i32>) schedule(static = %chunk_var : i32)
-  for (%iv) : index = (%lb) to (%ub) step (%step) {
-    omp.yield
-  }
+// CHECK-LABEL: omp_wsloop
+func.func @omp_wsloop(%lb : index, %ub : index, %step : index, %data_var : memref<i32>, %linear_var : i32, %chunk_var : i32) -> () {
 
-  // CHECK: omp.wsloop linear(%{{.*}} = %{{.*}} : memref<i32>) schedule(dynamic = %{{.*}} : i32, nonmonotonic) ordered(2)
-  // CHECK-SAME: for (%{{.*}}) : index = (%{{.*}}) to (%{{.*}}) step (%{{.*}})
-  omp.wsloop ordered(2) linear(%data_var = %linear_var : memref<i32>) schedule(dynamic = %chunk_var : i32, nonmonotonic)
-  for (%iv) : index = (%lb) to (%ub) step (%step)  {
-    omp.yield
-  }
+  // CHECK: omp.wsloop ordered(1) {
+  // CHECK-NEXT: omp.loop_nest
+  "omp.wsloop" () ({
+    omp.loop_nest (%iv) : index = (%lb) to (%ub) step (%step) {
+      omp.yield
+    }
+    omp.terminator
+  }) {operandSegmentSizes = array<i32: 0,0,0,0>, ordered_val = 1} :
+    () -> ()
 
-  // CHECK: omp.wsloop linear(%{{.*}} = %{{.*}} : memref<i32>) schedule(dynamic = %{{.*}} : i16, monotonic) ordered(2)
-  // CHECK-SAME: for (%{{.*}}) : index = (%{{.*}}) to (%{{.*}}) step (%{{.*}})
-  omp.wsloop ordered(2) linear(%data_var = %linear_var : memref<i32>) schedule(dynamic = %chunk_var2 : i16, monotonic)
-  for (%iv) : index = (%lb) to (%ub) step (%step) {
-    omp.yield
-  }
+  // CHECK: omp.wsloop linear(%{{.*}} = %{{.*}} : memref<i32>) schedule(static) {
+  // CHECK-NEXT: omp.loop_nest
+  "omp.wsloop" (%data_var, %linear_var) ({
+    omp.loop_nest (%iv) : index = (%lb) to (%ub) step (%step) {
+      omp.yield
+    }
+    omp.terminator
+  }) {operandSegmentSizes = array<i32: 1,1,0,0>, schedule_val = #omp<schedulekind static>} :
+    (memref<i32>, i32) -> ()
 
-  // CHECK: omp.wsloop for (%{{.*}}) : index = (%{{.*}}) to (%{{.*}}) step (%{{.*}})
-  omp.wsloop for (%iv) : index = (%lb) to (%ub) step (%step) {
-    omp.yield
-  }
+  // CHECK: omp.wsloop linear(%{{.*}} = %{{.*}} : memref<i32>, %{{.*}} = %{{.*}} : memref<i32>) schedule(static) {
+  // CHECK-NEXT: omp.loop_nest
+  "omp.wsloop" (%data_var, %data_var, %linear_var, %linear_var) ({
+    omp.loop_nest (%iv) : index = (%lb) to (%ub) step (%step) {
+      omp.yield
+    }
+    omp.terminator
+  }) {operandSegmentSizes = array<i32: 2,2,0,0>, schedule_val = #omp<schedulekind static>} :
+    (memref<i32>, memref<i32>, i32, i32) -> ()
 
-  // CHECK: omp.wsloop for (%{{.*}}) : index = (%{{.*}}) to (%{{.*}}) inclusive step (%{{.*}})
-  omp.wsloop for (%iv) : index = (%lb) to (%ub) inclusive step (%step) {
-    omp.yield
-  }
+  // CHECK: omp.wsloop linear(%{{.*}} = %{{.*}} : memref<i32>) schedule(dynamic = %{{.*}}) ordered(2) {
+  // CHECK-NEXT: omp.loop_nest
+  "omp.wsloop" (%data_var, %linear_var, %chunk_var) ({
+    omp.loop_nest (%iv) : index = (%lb) to (%ub) step (%step) {
+      omp.yield
+    }
+    omp.terminator
+  }) {operandSegmentSizes = array<i32: 1,1,0,1>, schedule_val = #omp<schedulekind dynamic>, ordered_val = 2} :
+    (memref<i32>, i32, i32) -> ()
 
-  // CHECK: omp.wsloop nowait
-  // CHECK-SAME: for (%{{.*}}) : index = (%{{.*}}) to (%{{.*}}) step (%{{.*}})
-  omp.wsloop nowait
-  for (%iv) : index = (%lb) to (%ub) step (%step) {
-    omp.yield
-  }
+  // CHECK: omp.wsloop schedule(auto) nowait {
+  // CHECK-NEXT: omp.loop_nest
+  "omp.wsloop" () ({
+    omp.loop_nest (%iv) : index = (%lb) to (%ub) step (%step) {
+      omp.yield
+    }
+    omp.terminator
+  }) {operandSegmentSizes = array<i32: 0,0,0,0>, nowait, schedule_val = #omp<schedulekind auto>} :
+    () -> ()
 
-  // CHECK: omp.wsloop nowait order(concurrent)
-  // CHECK-SAME: for (%{{.*}}) : index = (%{{.*}}) to (%{{.*}}) step (%{{.*}})
-  omp.wsloop order(concurrent) nowait
-  for (%iv) : index = (%lb) to (%ub) step (%step) {
-    omp.yield
-  }
+  // CHECK: omp.wsloop {
+  // CHECK-NEXT: omp.simd
+  // CHECK-NEXT: omp.loop_nest
+  "omp.wsloop" () ({
+    omp.simd {
+      omp.loop_nest (%iv) : index = (%lb) to (%ub) step (%step) {
+        omp.yield
+      }
+      omp.terminator
+    }
+    omp.terminator
+  }) : () -> ()
 
   return
 }
 
-// CHECK-LABEL: omp_wsloop_pretty_multi_block
-func.func @omp_wsloop_pretty_multi_block(%lb : index, %ub : index, %step : index, %data1 : memref<?xi32>, %data2 : memref<?xi32>) -> () {
+// CHECK-LABEL: omp_wsloop_pretty
+func.func @omp_wsloop_pretty(%lb : index, %ub : index, %step : index, %data_var : memref<i32>, %linear_var : i32, %chunk_var : i32, %chunk_var2 : i16) -> () {
 
-  // CHECK: omp.wsloop for (%{{.*}}) : index = (%{{.*}}) to (%{{.*}}) step (%{{.*}})
-  omp.wsloop for (%iv) : index = (%lb) to (%ub) step (%step) {
-    %1 = "test.payload"(%iv) : (index) -> (i32)
-    cf.br ^bb1(%1: i32)
-  ^bb1(%arg: i32):
-    memref.store %arg, %data1[%iv] : memref<?xi32>
-    omp.yield
+  // CHECK: omp.wsloop ordered(2) {
+  // CHECK-NEXT: omp.loop_nest
+  omp.wsloop ordered(2) {
+    omp.loop_nest (%iv) : index = (%lb) to (%ub) step (%step) {
+      omp.yield
+    }
+    omp.terminator
   }
 
-  // CHECK: omp.wsloop for (%{{.*}}) : index = (%{{.*}}) to (%{{.*}}) step (%{{.*}})
-  omp.wsloop for (%iv) : index = (%lb) to (%ub) step (%step) {
-    %c = "test.condition"(%iv) : (index) -> (i1)
-    %v1 = "test.payload"(%iv) : (index) -> (i32)
-    cf.cond_br %c, ^bb1(%v1: i32), ^bb2(%v1: i32)
-  ^bb1(%arg0: i32):
-    memref.store %arg0, %data1[%iv] : memref<?xi32>
-    cf.br ^bb3
-  ^bb2(%arg1: i32):
-    memref.store %arg1, %data2[%iv] : memref<?xi32>
-    cf.br ^bb3
-  ^bb3:
-    omp.yield
+  // CHECK: omp.wsloop linear(%{{.*}} = %{{.*}} : memref<i32>) schedule(static) {
+  // CHECK-NEXT: omp.loop_nest
+  omp.wsloop schedule(static) linear(%data_var = %linear_var : memref<i32>) {
+    omp.loop_nest (%iv) : index = (%lb) to (%ub) step (%step) {
+      omp.yield
+    }
+    omp.terminator
   }
 
-  // CHECK: omp.wsloop for (%{{.*}}) : index = (%{{.*}}) to (%{{.*}}) step (%{{.*}})
-  omp.wsloop for (%iv) : index = (%lb) to (%ub) step (%step) {
-    %c = "test.condition"(%iv) : (index) -> (i1)
-    %v1 = "test.payload"(%iv) : (index) -> (i32)
-    cf.cond_br %c, ^bb1(%v1: i32), ^bb2(%v1: i32)
-  ^bb1(%arg0: i32):
-    memref.store %arg0, %data1[%iv] : memref<?xi32>
-    omp.yield
-  ^bb2(%arg1: i32):
-    memref.store %arg1, %data2[%iv] : memref<?xi32>
-    omp.yield
+  // CHECK: omp.wsloop linear(%{{.*}} = %{{.*}} : memref<i32>) schedule(static = %{{.*}} : i32) ordered(2) {
+  // CHECK-NEXT: omp.loop_nest
+  omp.wsloop ordered(2) linear(%data_var = %linear_var : memref<i32>) schedule(static = %chunk_var : i32) {
+    omp.loop_nest (%iv) : index = (%lb) to (%ub) step (%step) {
+      omp.yield
+    }
+    omp.terminator
   }
 
-  return
-}
-
-// CHECK-LABEL: omp_wsloop_pretty_non_index
-func.func @omp_wsloop_pretty_non_index(%lb1 : i32, %ub1 : i32, %step1 : i32, %lb2 : i64, %ub2 : i64, %step2 : i64,
-                           %data1 : memref<?xi32>, %data2 : memref<?xi64>) -> () {
+  // CHECK: omp.wsloop linear(%{{.*}} = %{{.*}} : memref<i32>) schedule(dynamic = %{{.*}} : i32, nonmonotonic) ordered(2) {
+  // CHECK-NEXT: omp.loop_nest
+  omp.wsloop ordered(2) linear(%data_var = %linear_var : memref<i32>) schedule(dynamic = %chunk_var : i32, nonmonotonic) {
+    omp.loop_nest (%iv) : index = (%lb) to (%ub) step (%step)  {
+      omp.yield
+    }
+    omp.terminator
+  }
 
-  // CHECK: omp.wsloop for (%{{.*}}) : i32 = (%{{.*}}) to (%{{.*}}) step (%{{.*}})
-  omp.wsloop for (%iv1) : i32 = (%lb1) to (%ub1) step (%step1) {
-    %1 = "test.payload"(%iv1) : (i32) -> (index)
-    cf.br ^bb1(%1: index)
-  ^bb1(%arg1: index):
-    memref.store %iv1, %data1[%arg1] : memref<?xi32>
-    omp.yield
+  // CHECK: omp.wsloop linear(%{{.*}} = %{{.*}} : memref<i32>) schedule(dynamic = %{{.*}} : i16, monotonic) ordered(2) {
+  // CHECK-NEXT: omp.loop_nest
+  omp.wsloop ordered(2) linear(%data_var = %linear_var : memref<i32>) schedule(dynamic = %chunk_var2 : i16, monotonic) {
+    omp.loop_nest (%iv) : index = (%lb) to (%ub) step (%step) {
+      omp.yield
+    }
+    omp.terminator
   }
 
-  // CHECK: omp.wsloop for (%{{.*}}) : i64 = (%{{.*}}) to (%{{.*}}) step (%{{.*}})
-  omp.wsloop for (%iv2) : i64 = (%lb2) to (%ub2) step (%step2) {
-    %2 = "test.payload"(%iv2) : (i64) -> (index)
-    cf.br ^bb1(%2: index)
-  ^bb1(%arg2: index):
-    memref.store %iv2, %data2[%arg2] : memref<?xi64>
-    omp.yield
+  // CHECK: omp.wsloop {
+  // CHECK-NEXT: omp.loop_nest
+  omp.wsloop {
+    omp.loop_nest (%iv) : index = (%lb) to (%ub) step (%step) {
+      omp.yield
+    }
+    omp.terminator
   }
 
-  return
-}
+  // CHECK: omp.wsloop nowait {
+  // CHECK-NEXT: omp.loop_nest
+  omp.wsloop nowait {
+    omp.loop_nest (%iv) : index = (%lb) to (%ub) step (%step) {
+      omp.yield
+    }
+    omp.terminator
+  }
 
-// CHECK-LABEL: omp_wsloop_pretty_multiple
-func.func @omp_wsloop_pretty_multiple(%lb1 : i32, %ub1 : i32, %step1 : i32, %lb2 : i32, %ub2 : i32, %step2 : i32, %data1 : memref<?xi32>) -> () {
+  // CHECK: omp.wsloop nowait order(concurrent) {
+  // CHECK-NEXT: omp.loop_nest
+  omp.wsloop order(concurrent) nowait {
+    omp.loop_nest (%iv) : index = (%lb) to (%ub) step (%step) {
+      omp.yield
+    }
+    omp.terminator
+  }
 
-  // CHECK: omp.wsloop for (%{{.*}}, %{{.*}}) : i32 = (%{{.*}}, %{{.*}}) to (%{{.*}}, %{{.*}}) step (%{{.*}}, %{{.*}})
-  omp.wsloop for (%iv1, %iv2) : i32 = (%lb1, %lb2) to (%ub1, %ub2) step (%step1, %step2) {
-    %1 = "test.payload"(%iv1) : (i32) -> (index)
-    %2 = "test.payload"(%iv2) : (i32) -> (index)
-    memref.store %iv1, %data1[%1] : memref<?xi32>
-    memref.store %iv2, %data1[%2] : memref<?xi32>
-    omp.yield
+  // CHECK: omp.wsloop {
+  // CHECK-NEXT: omp.simd
+  // CHECK-NEXT: omp.loop_nest
+  omp.wsloop {
+    omp.simd {
+      omp.loop_nest (%iv) : index = (%lb) to (%ub) step (%step) {
+        omp.yield
+      }
+      omp.terminator
+    }
+    omp.terminator
   }
 
   return
@@ -659,7 +720,7 @@ func.func @omp_distribute(%chunk_size : i32, %data_var : memref<i32>, %arg0 : i3
   // CHECK: omp.distribute
   omp.distribute {
     omp.simd {
-      omp.loop_nest (%iv2) : i32 = (%arg0) to (%arg0) step (%arg0) {
+      omp.loop_nest (%iv) : i32 = (%arg0) to (%arg0) step (%arg0) {
         omp.yield
       }
     }
@@ -791,17 +852,19 @@ func.func @wsloop_reduction(%lb : index, %ub : index, %step : index) {
   %c1 = arith.constant 1 : i32
   %0 = llvm.alloca %c1 x i32 : (i32) -> !llvm.ptr
   // CHECK: reduction(@add_f32 %{{.+}} -> %[[PRV:.+]] : !llvm.ptr)
-  omp.wsloop reduction(@add_f32 %0 -> %prv : !llvm.ptr)
-  for (%iv) : index = (%lb) to (%ub) step (%step) {
-    // CHECK: %[[CST:.+]] = arith.constant 2.0{{.*}} : f32
-    %cst = arith.constant 2.0 : f32
-    // CHECK: %[[LPRV:.+]] = llvm.load %[[PRV]] : !llvm.ptr -> f32
-    %lprv = llvm.load %prv : !llvm.ptr -> f32
-    // CHECK: %[[RES:.+]] = llvm.fadd %[[LPRV]], %[[CST]] : f32
-    %res = llvm.fadd %lprv, %cst: f32
-    // CHECK: llvm.store %[[RES]], %[[PRV]] :  f32, !llvm.ptr
-    llvm.store %res, %prv :  f32, !llvm.ptr
-    omp.yield
+  omp.wsloop reduction(@add_f32 %0 -> %prv : !llvm.ptr) {
+    omp.loop_nest (%iv) : index = (%lb) to (%ub) step (%step) {
+      // CHECK: %[[CST:.+]] = arith.constant 2.0{{.*}} : f32
+      %cst = arith.constant 2.0 : f32
+      // CHECK: %[[LPRV:.+]] = llvm.load %[[PRV]] : !llvm.ptr -> f32
+      %lprv = llvm.load %prv : !llvm.ptr -> f32
+      // CHECK: %[[RES:.+]] = llvm.fadd %[[LPRV]], %[[CST]] : f32
+      %res = llvm.fadd %lprv, %cst: f32
+      // CHECK: llvm.store %[[RES]], %[[PRV]] :  f32, !llvm.ptr
+      llvm.store %res, %prv :  f32, !llvm.ptr
+      omp.yield
+    }
+    omp.terminator
   }
   return
 }
@@ -828,14 +891,19 @@ func.func @parallel_wsloop_reduction(%lb : index, %ub : index, %step : index) {
   %0 = llvm.alloca %c1 x i32 : (i32) -> !llvm.ptr
   // CHECK: omp.parallel reduction(@add_f32 %{{.*}} -> %{{.+}} : !llvm.ptr) {
   omp.parallel reduction(@add_f32 %0 -> %prv : !llvm.ptr) {
-    // CHECK: omp.wsloop for (%{{.+}}) : index = (%{{.+}}) to (%{{.+}}) step (%{{.+}})
-    omp.wsloop for (%iv) : index = (%lb) to (%ub) step (%step) {
-      %1 = arith.constant 2.0 : f32
-      %2 = llvm.load %prv : !llvm.ptr -> f32
-      // CHECK: llvm.fadd %{{.+}}, %{{.+}} : f32
-      llvm.fadd %1, %2 : f32
-      // CHECK: omp.yield
-      omp.yield
+    // CHECK: omp.wsloop {
+    omp.wsloop {
+      // CHECK: omp.loop_nest (%{{.+}}) : index = (%{{.+}}) to (%{{.+}}) step (%{{.+}}) {
+      omp.loop_nest (%iv) : index = (%lb) to (%ub) step (%step) {
+        %1 = arith.constant 2.0 : f32
+        %2 = llvm.load %prv : !llvm.ptr -> f32
+        // CHECK: llvm.fadd %{{.+}}, %{{.+}} : f32
+        llvm.fadd %1, %2 : f32
+        // CHECK: omp.yield
+        omp.yield
+      }
+      // CHECK: omp.terminator
+      omp.terminator
     }
     // CHECK: omp.terminator
     omp.terminator
@@ -959,16 +1027,18 @@ combiner {
 // CHECK-LABEL: func @wsloop_reduction2
 func.func @wsloop_reduction2(%lb : index, %ub : index, %step : index) {
   %0 = memref.alloca() : memref<1xf32>
-  // CHECK: omp.wsloop reduction(@add2_f32 %{{.+}} -> %{{.+}} : memref<1xf32>)
-  omp.wsloop reduction(@add2_f32 %0 -> %prv : memref<1xf32>)
-  for (%iv) : index = (%lb) to (%ub) step (%step) {
-    %1 = arith.constant 2.0 : f32
-    %2 = arith.constant 0 : index
-    %3 = memref.load %prv[%2] : memref<1xf32>
-    // CHECK: llvm.fadd
-    %4 = llvm.fadd %1, %3 : f32
-    memref.store %4, %prv[%2] : memref<1xf32>
-    omp.yield
+  // CHECK: omp.wsloop reduction(@add2_f32 %{{.+}} -> %{{.+}} : memref<1xf32>) {
+  omp.wsloop reduction(@add2_f32 %0 -> %prv : memref<1xf32>) {
+    omp.loop_nest (%iv) : index = (%lb) to (%ub) step (%step) {
+      %1 = arith.constant 2.0 : f32
+      %2 = arith.constant 0 : index
+      %3 = memref.load %prv[%2] : memref<1xf32>
+      // CHECK: llvm.fadd
+      %4 = llvm.fadd %1, %3 : f32
+      memref.store %4, %prv[%2] : memref<1xf32>
+      omp.yield
+    }
+    omp.terminator
   }
   return
 }
@@ -995,14 +1065,19 @@ func.func @parallel_wsloop_reduction2(%lb : index, %ub : index, %step : index) {
   %0 = llvm.alloca %c1 x i32 : (i32) -> !llvm.ptr
   // CHECK: omp.parallel reduction(@add2_f32 %{{.*}} -> %{{.+}} : !llvm.ptr) {
   omp.parallel reduction(@add2_f32 %0 -> %prv : !llvm.ptr) {
-    // CHECK: omp.wsloop for (%{{.+}}) : index = (%{{.+}}) to (%{{.+}}) step (%{{.+}})
-    omp.wsloop for (%iv) : index = (%lb) to (%ub) step (%step) {
-      %1 = arith.constant 2.0 : f32
-      %2 = llvm.load %prv : !llvm.ptr -> f32
-      // CHECK: llvm.fadd %{{.+}}, %{{.+}} : f32
-      %3 = llvm.fadd %1, %2 : f32
-      // CHECK: omp.yield
-      omp.yield
+    // CHECK: omp.wsloop {
+    omp.wsloop {
+      // CHECK: omp.loop_nest (%{{.+}}) : index = (%{{.+}}) to (%{{.+}}) step (%{{.+}}) {
+      omp.loop_nest (%iv) : index = (%lb) to (%ub) step (%step) {
+        %1 = arith.constant 2.0 : f32
+        %2 = llvm.load %prv : !llvm.ptr -> f32
+        // CHECK: llvm.fadd %{{.+}}, %{{.+}} : f32
+        %3 = llvm.fadd %1, %2 : f32
+        // CHECK: omp.yield
+        omp.yield
+      }
+      // CHECK: omp.terminator
+      omp.terminator
     }
     // CHECK: omp.terminator
     omp.terminator
@@ -1076,36 +1151,44 @@ func.func @omp_ordered(%arg1 : i32, %arg2 : i32, %arg3 : i32,
     omp.terminator
   }
 
-  omp.wsloop ordered(0)
-  for (%0) : i32 = (%arg1) to (%arg2) step (%arg3)  {
-    omp.ordered.region {
-      omp.terminator
+  omp.wsloop ordered(0) {
+    omp.loop_nest (%0) : i32 = (%arg1) to (%arg2) step (%arg3)  {
+      // CHECK: omp.ordered.region
+      omp.ordered.region {
+        // CHECK: omp.terminator
+        omp.terminator
+      }
+      omp.yield
     }
-    omp.yield
+    omp.terminator
   }
 
-  omp.wsloop ordered(1)
-  for (%0) : i32 = (%arg1) to (%arg2) step (%arg3) {
-    // Only one DEPEND(SINK: vec) clause
-    // CHECK: omp.ordered depend_type(dependsink) depend_vec(%{{.*}} : i64) {num_loops_val = 1 : i64}
-    omp.ordered depend_type(dependsink) depend_vec(%vec0 : i64) {num_loops_val = 1 : i64}
+  omp.wsloop ordered(1) {
+    omp.loop_nest (%0) : i32 = (%arg1) to (%arg2) step (%arg3) {
+      // Only one DEPEND(SINK: vec) clause
+      // CHECK: omp.ordered depend_type(dependsink) depend_vec(%{{.*}} : i64) {num_loops_val = 1 : i64}
+      omp.ordered depend_type(dependsink) depend_vec(%vec0 : i64) {num_loops_val = 1 : i64}
 
-    // CHECK: omp.ordered depend_type(dependsource) depend_vec(%{{.*}} : i64) {num_loops_val = 1 : i64}
-    omp.ordered depend_type(dependsource) depend_vec(%vec0 : i64) {num_loops_val = 1 : i64}
+      // CHECK: omp.ordered depend_type(dependsource) depend_vec(%{{.*}} : i64) {num_loops_val = 1 : i64}
+      omp.ordered depend_type(dependsource) depend_vec(%vec0 : i64) {num_loops_val = 1 : i64}
 
-    omp.yield
+      omp.yield
+    }
+    omp.terminator
   }
 
-  omp.wsloop ordered(2)
-  for (%0) : i32 = (%arg1) to (%arg2) step (%arg3) {
-    // Multiple DEPEND(SINK: vec) clauses
-    // CHECK: omp.ordered depend_type(dependsink) depend_vec(%{{.*}}, %{{.*}}, %{{.*}}, %{{.*}} : i64, i64, i64, i64) {num_loops_val = 2 : i64}
-    omp.ordered depend_type(dependsink) depend_vec(%vec0, %vec1, %vec2, %vec3 : i64, i64, i64, i64) {num_loops_val = 2 : i64}
+  omp.wsloop ordered(2) {
+    omp.loop_nest (%0) : i32 = (%arg1) to (%arg2) step (%arg3) {
+      // Multiple DEPEND(SINK: vec) clauses
+      // CHECK: omp.ordered depend_type(dependsink) depend_vec(%{{.*}}, %{{.*}}, %{{.*}}, %{{.*}} : i64, i64, i64, i64) {num_loops_val = 2 : i64}
+      omp.ordered depend_type(dependsink) depend_vec(%vec0, %vec1, %vec2, %vec3 : i64, i64, i64, i64) {num_loops_val = 2 : i64}
 
-    // CHECK: omp.ordered depend_type(dependsource) depend_vec(%{{.*}}, %{{.*}} : i64, i64) {num_loops_val = 2 : i64}
-    omp.ordered depend_type(dependsource) depend_vec(%vec0, %vec1 : i64, i64) {num_loops_val = 2 : i64}
+      // CHECK: omp.ordered depend_type(dependsource) depend_vec(%{{.*}}, %{{.*}} : i64, i64) {num_loops_val = 2 : i64}
+      omp.ordered depend_type(dependsource) depend_vec(%vec0, %vec1 : i64, i64) {num_loops_val = 2 : i64}
 
-    omp.yield
+      omp.yield
+    }
+    omp.terminator
   }
 
   return
@@ -1956,11 +2039,13 @@ func.func @omp_cancel_parallel(%if_cond : i1) -> () {
 }
 
 func.func @omp_cancel_wsloop(%lb : index, %ub : index, %step : index) {
-  omp.wsloop
-  for (%iv) : index = (%lb) to (%ub) step (%step) {
-    // CHECK: omp.cancel cancellation_construct_type(loop)
-    omp.cancel cancellation_construct_type(loop)
-    // CHECK: omp.terminator
+  omp.wsloop {
+    omp.loop_nest (%iv) : index = (%lb) to (%ub) step (%step) {
+      // CHECK: omp.cancel cancellation_construct_type(loop)
+      omp.cancel cancellation_construct_type(loop)
+      // CHECK: omp.yield
+      omp.yield
+    }
     omp.terminator
   }
   return
@@ -1991,13 +2076,15 @@ func.func @omp_cancellationpoint_parallel() -> () {
 }
 
 func.func @omp_cancellationpoint_wsloop(%lb : index, %ub : index, %step : index) {
-  omp.wsloop
-  for (%iv) : index = (%lb) to (%ub) step (%step) {
-    // CHECK: omp.cancellation_point cancellation_construct_type(loop)
-    omp.cancellation_point cancellation_construct_type(loop)
-    // CHECK: omp.cancel cancellation_construct_type(loop)
-    omp.cancel cancellation_construct_type(loop)
-    // CHECK: omp.terminator
+  omp.wsloop {
+    omp.loop_nest (%iv) : index = (%lb) to (%ub) step (%step) {
+      // CHECK: omp.cancellation_point cancellation_construct_type(loop)
+      omp.cancellation_point cancellation_construct_type(loop)
+      // CHECK: omp.cancel cancellation_construct_type(loop)
+      omp.cancel cancellation_construct_type(loop)
+      // CHECK: omp.yield
+      omp.yield
+    }
     omp.terminator
   }
   return
diff --git a/mlir/test/Dialect/Polynomial/ops.mlir b/mlir/test/Dialect/Polynomial/ops.mlir
new file mode 100644
index 000000000000..ea1b279fa1ff
--- /dev/null
+++ b/mlir/test/Dialect/Polynomial/ops.mlir
@@ -0,0 +1,82 @@
+// RUN: mlir-opt %s | FileCheck %s
+
+// This simply tests for syntax.
+
+#my_poly = #polynomial.polynomial<1 + x**1024>
+#my_poly_2 = #polynomial.polynomial<2>
+#my_poly_3 = #polynomial.polynomial<3x>
+#my_poly_4 = #polynomial.polynomial<t**3 + 4t + 2>
+#ring1 = #polynomial.ring<coefficientType=i32, coefficientModulus=2837465, polynomialModulus=#my_poly>
+#one_plus_x_squared = #polynomial.polynomial<1 + x**2>
+
+#ideal = #polynomial.polynomial<-1 + x**1024>
+#ring = #polynomial.ring<coefficientType=i32, coefficientModulus=18, polynomialModulus=#ideal>
+!poly_ty = !polynomial.polynomial<#ring>
+
+module {
+  func.func @test_multiply() -> !polynomial.polynomial<#ring1> {
+    %c0 = arith.constant 0 : index
+    %two = arith.constant 2 : i16
+    %five = arith.constant 5 : i16
+    %coeffs1 = tensor.from_elements %two, %two, %five : tensor<3xi16>
+    %coeffs2 = tensor.from_elements %five, %five, %two : tensor<3xi16>
+
+    %poly1 = polynomial.from_tensor %coeffs1 : tensor<3xi16> -> !polynomial.polynomial<#ring1>
+    %poly2 = polynomial.from_tensor %coeffs2 : tensor<3xi16> -> !polynomial.polynomial<#ring1>
+
+    %3 = polynomial.mul %poly1, %poly2 : !polynomial.polynomial<#ring1>
+
+    return %3 : !polynomial.polynomial<#ring1>
+  }
+
+  func.func @test_elementwise(%p0 : !polynomial.polynomial<#ring1>, %p1: !polynomial.polynomial<#ring1>) {
+    %tp0 = tensor.from_elements %p0, %p1 : tensor<2x!polynomial.polynomial<#ring1>>
+    %tp1 = tensor.from_elements %p1, %p0 : tensor<2x!polynomial.polynomial<#ring1>>
+
+    %c = arith.constant 2 : i32
+    %mul_const_sclr = polynomial.mul_scalar %tp0, %c : tensor<2x!polynomial.polynomial<#ring1>>, i32
+
+    %add = polynomial.add %tp0, %tp1 : tensor<2x!polynomial.polynomial<#ring1>>
+    %sub = polynomial.sub %tp0, %tp1 : tensor<2x!polynomial.polynomial<#ring1>>
+    %mul = polynomial.mul %tp0, %tp1 : tensor<2x!polynomial.polynomial<#ring1>>
+
+    return
+  }
+
+  func.func @test_to_from_tensor(%p0 : !polynomial.polynomial<#ring1>) {
+    %c0 = arith.constant 0 : index
+    %two = arith.constant 2 : i16
+    %coeffs1 = tensor.from_elements %two, %two : tensor<2xi16>
+    // CHECK: from_tensor
+    %poly = polynomial.from_tensor %coeffs1 : tensor<2xi16> -> !polynomial.polynomial<#ring1>
+    // CHECK: to_tensor
+    %tensor = polynomial.to_tensor %poly : !polynomial.polynomial<#ring1> -> tensor<1024xi16>
+
+    return
+  }
+
+  func.func @test_degree(%p0 : !polynomial.polynomial<#ring1>) {
+    %0, %1 = polynomial.leading_term %p0 : !polynomial.polynomial<#ring1> -> (index, i32)
+    return
+  }
+
+  func.func @test_monomial() {
+    %deg = arith.constant 1023 : index
+    %five = arith.constant 5 : i16
+    %0 = polynomial.monomial %five, %deg : (i16, index) -> !polynomial.polynomial<#ring1>
+    return
+  }
+
+  func.func @test_monic_monomial_mul() {
+    %five = arith.constant 5 : index
+    %0 = polynomial.constant #one_plus_x_squared : !polynomial.polynomial<#ring1>
+    %1 = polynomial.monic_monomial_mul %0, %five : (!polynomial.polynomial<#ring1>, index) -> !polynomial.polynomial<#ring1>
+    return
+  }
+
+  func.func @test_constant() {
+    %0 = polynomial.constant #one_plus_x_squared : !polynomial.polynomial<#ring1>
+    %1 = polynomial.constant <1 + x**2> : !polynomial.polynomial<#ring1>
+    return
+  }
+}
diff --git a/mlir/test/Dialect/Polynomial/ops_errors.mlir b/mlir/test/Dialect/Polynomial/ops_errors.mlir
new file mode 100644
index 000000000000..c34a7de30e5f
--- /dev/null
+++ b/mlir/test/Dialect/Polynomial/ops_errors.mlir
@@ -0,0 +1,53 @@
+// RUN: mlir-opt --split-input-file --verify-diagnostics %s
+
+#my_poly = #polynomial.polynomial<1 + x**1024>
+#ring = #polynomial.ring<coefficientType=i16, coefficientModulus=256, polynomialModulus=#my_poly>
+!ty = !polynomial.polynomial<#ring>
+
+func.func @test_from_tensor_too_large_coeffs() {
+  %two = arith.constant 2 : i32
+  %coeffs1 = tensor.from_elements %two, %two : tensor<2xi32>
+  // expected-error@below {{is too large to fit in the coefficients}}
+  // expected-note@below {{rescaled to fit}}
+  %poly = polynomial.from_tensor %coeffs1 : tensor<2xi32> -> !ty
+  return
+}
+
+// -----
+
+#my_poly = #polynomial.polynomial<1 + x**4>
+#ring = #polynomial.ring<coefficientType=i32, coefficientModulus=256, polynomialModulus=#my_poly>
+!ty = !polynomial.polynomial<#ring>
+func.func @test_from_tensor_wrong_tensor_type() {
+  %two = arith.constant 2 : i32
+  %coeffs1 = tensor.from_elements %two, %two, %two, %two, %two : tensor<5xi32>
+  // expected-error@below {{input type 'tensor<5xi32>' does not match output type '!polynomial.polynomial<#polynomial.ring<coefficientType=i32, coefficientModulus=256 : i32, polynomialModulus=#polynomial.polynomial<1 + x**4>>>'}}
+  // expected-note@below {{at most the degree of the polynomialModulus of the output type's ring attribute}}
+  %poly = polynomial.from_tensor %coeffs1 : tensor<5xi32> -> !ty
+  return
+}
+
+// -----
+
+#my_poly = #polynomial.polynomial<1 + x**4>
+#ring = #polynomial.ring<coefficientType=i32, coefficientModulus=256, polynomialModulus=#my_poly>
+!ty = !polynomial.polynomial<#ring>
+func.func @test_to_tensor_wrong_output_tensor_type(%arg0 : !ty) {
+  // expected-error@below {{input type '!polynomial.polynomial<#polynomial.ring<coefficientType=i32, coefficientModulus=256 : i32, polynomialModulus=#polynomial.polynomial<1 + x**4>>>' does not match output type 'tensor<5xi32>'}}
+  // expected-note@below {{at most the degree of the polynomialModulus of the input type's ring attribute}}
+  %tensor = polynomial.to_tensor %arg0 : !ty -> tensor<5xi32>
+  return
+}
+
+// -----
+
+#my_poly = #polynomial.polynomial<1 + x**1024>
+#ring = #polynomial.ring<coefficientType=i16, coefficientModulus=256, polynomialModulus=#my_poly>
+!ty = !polynomial.polynomial<#ring>
+
+func.func @test_mul_scalar_wrong_type(%arg0: !ty) -> !ty {
+  %scalar = arith.constant 2 : i32  // should be i16
+  // expected-error@below {{polynomial coefficient type 'i16' does not match scalar type 'i32'}}
+  %poly = polynomial.mul_scalar %arg0, %scalar : !ty, i32
+  return %poly : !ty
+}
diff --git a/mlir/test/Dialect/Polynomial/types.mlir b/mlir/test/Dialect/Polynomial/types.mlir
index 64b74d9d36bb..00296a36e890 100644
--- a/mlir/test/Dialect/Polynomial/types.mlir
+++ b/mlir/test/Dialect/Polynomial/types.mlir
@@ -40,3 +40,17 @@ func.func @test_non_x_variable_64_bit(%0: !ty2) -> !ty2 {
 func.func @test_linear_poly(%0: !ty3) -> !ty3 {
   return %0 : !ty3
 }
+
+// CHECK-LABEL: func @test_negative_leading_1
+// CHECK-SAME:  !polynomial.polynomial<
+// CHECK-SAME:    #polynomial.ring<
+// CHECK-SAME:       coefficientType=i32,
+// CHECK-SAME:       coefficientModulus=2837465 : i32,
+// CHECK-SAME:       polynomialModulus=#polynomial.polynomial<-1 + x**1024>>>
+#my_poly_4 = #polynomial.polynomial<-1 + x**1024>
+#ring4 = #polynomial.ring<coefficientType=i32, coefficientModulus=2837465, polynomialModulus=#my_poly_4>
+!ty4 = !polynomial.polynomial<#ring4>
+func.func @test_negative_leading_1(%0: !ty4) -> !ty4 {
+  return %0 : !ty4
+}
+
diff --git a/mlir/test/Dialect/SCF/forall-to-for.mlir b/mlir/test/Dialect/SCF/forall-to-for.mlir
new file mode 100644
index 000000000000..e7d183fb9d2b
--- /dev/null
+++ b/mlir/test/Dialect/SCF/forall-to-for.mlir
@@ -0,0 +1,57 @@
+// RUN: mlir-opt %s -pass-pipeline='builtin.module(func.func(scf-forall-to-for))' -split-input-file | FileCheck %s
+
+func.func private @callee(%i: index, %j: index)
+
+// CHECK-LABEL: @two_iters
+// CHECK-SAME: %[[UB1:.+]]: index, %[[UB2:.+]]: index
+func.func @two_iters(%ub1: index, %ub2: index) {
+  scf.forall (%i, %j) in (%ub1, %ub2) {
+    func.call @callee(%i, %j) : (index, index) -> ()
+  }
+  // CHECK: scf.for %[[IV1:.+]] = %{{.*}} to %[[UB1]]
+  // CHECK:   scf.for %[[IV2:.+]] = %{{.*}} to %[[UB2]]
+  // CHECK:     func.call @callee(%[[IV1]], %[[IV2]])
+  return
+}
+
+// -----
+
+func.func private @callee(%i: index, %j: index)
+
+// CHECK-LABEL: @repeated
+// CHECK-SAME: %[[UB1:.+]]: index, %[[UB2:.+]]: index
+func.func @repeated(%ub1: index, %ub2: index) {
+  scf.forall (%i, %j) in (%ub1, %ub2) {
+    func.call @callee(%i, %j) : (index, index) -> ()
+  }
+  // CHECK: scf.for %[[IV1:.+]] = %{{.*}} to %[[UB1]]
+  // CHECK:   scf.for %[[IV2:.+]] = %{{.*}} to %[[UB2]]
+  // CHECK:     func.call @callee(%[[IV1]], %[[IV2]])
+  scf.forall (%i, %j) in (%ub1, %ub2) {
+    func.call @callee(%i, %j) : (index, index) -> ()
+  }
+  // CHECK: scf.for %[[IV1:.+]] = %{{.*}} to %[[UB1]]
+  // CHECK:   scf.for %[[IV2:.+]] = %{{.*}} to %[[UB2]]
+  // CHECK:     func.call @callee(%[[IV1]], %[[IV2]])
+  return
+}
+
+// -----
+
+func.func private @callee(%i: index, %j: index, %k: index, %l: index)
+
+// CHECK-LABEL: @nested
+// CHECK-SAME: %[[UB1:.+]]: index, %[[UB2:.+]]: index, %[[UB3:.+]]: index, %[[UB4:.+]]: index
+func.func @nested(%ub1: index, %ub2: index, %ub3: index, %ub4: index) {
+  // CHECK: scf.for %[[IV1:.+]] = %{{.*}} to %[[UB1]]
+  // CHECK:   scf.for %[[IV2:.+]] = %{{.*}} to %[[UB2]]
+  // CHECK:     scf.for %[[IV3:.+]] = %{{.*}} to %[[UB3]]
+  // CHECK:       scf.for %[[IV4:.+]] = %{{.*}} to %[[UB4]]
+  // CHECK:         func.call @callee(%[[IV1]], %[[IV2]], %[[IV3]], %[[IV4]])
+  scf.forall (%i, %j) in (%ub1, %ub2) {
+    scf.forall (%k, %l) in (%ub3, %ub4) {
+      func.call @callee(%i, %j, %k, %l) : (index, index, index, index) -> ()
+    }
+  }
+  return
+}
diff --git a/mlir/test/Dialect/SparseTensor/fuse_sparse_concat_with_extract_slice.mlir b/mlir/test/Dialect/SparseTensor/fuse_sparse_concat_with_extract_slice.mlir
new file mode 100644
index 000000000000..5d93301bc8ca
--- /dev/null
+++ b/mlir/test/Dialect/SparseTensor/fuse_sparse_concat_with_extract_slice.mlir
@@ -0,0 +1,23 @@
+// RUN: mlir-opt %s --pre-sparsification-rewrite | FileCheck %s
+
+#CCCD = #sparse_tensor.encoding<{ map = (d0, d1, d2, d3) -> (d0 : compressed, d1 : compressed, d2 : compressed, d3 : dense) }>
+
+
+
+// CHECK-LABEL: func.func @fuse_concat_with_extract(
+// CHECK-SAME:    %[[VAL_0:.*0]]: tensor<128x32x32x1xf32, #sparse{{[0-9]*}}>,
+// CHECK-SAME:    %[[VAL_1:.*1]]: tensor<128x32x32x1xf32, #sparse{{[0-9]*}}>,
+// CHECK-SAME:    %[[VAL_2:.*2]]: tensor<128x32x32x1xf32, #sparse{{[0-9]*}}>)
+// CHECK-NOT:     tensor.concat
+// CHECK-NOT:     tensor.extract_slice
+// CHECK:         return %[[VAL_0]], %[[VAL_1]], %[[VAL_2]]
+// CHECK:       }
+func.func @fuse_concat_with_extract(%t0 : tensor<128x32x32x1xf32, #CCCD>,
+                                    %t1 : tensor<128x32x32x1xf32, #CCCD>,
+                                    %t2 : tensor<128x32x32x1xf32, #CCCD>) -> (tensor<128x32x32x1xf32, #CCCD>, tensor<128x32x32x1xf32, #CCCD>, tensor<128x32x32x1xf32, #CCCD>) {
+  %concat = tensor.concat dim(3) %t0, %t1, %t2 : (tensor<128x32x32x1xf32, #CCCD>, tensor<128x32x32x1xf32, #CCCD>, tensor<128x32x32x1xf32, #CCCD>) -> tensor<128x32x32x3xf32, #CCCD>
+  %r0 = tensor.extract_slice %concat[0, 0, 0, 0] [128, 32, 32, 1] [1, 1, 1, 1] : tensor<128x32x32x3xf32, #CCCD> to tensor<128x32x32x1xf32, #CCCD>
+  %r1 = tensor.extract_slice %concat[0, 0, 0, 1] [128, 32, 32, 1] [1, 1, 1, 1] : tensor<128x32x32x3xf32, #CCCD> to tensor<128x32x32x1xf32, #CCCD>
+  %r2 = tensor.extract_slice %concat[0, 0, 0, 2] [128, 32, 32, 1] [1, 1, 1, 1] : tensor<128x32x32x3xf32, #CCCD> to tensor<128x32x32x1xf32, #CCCD>
+  return %r0, %r1, %r2 : tensor<128x32x32x1xf32, #CCCD>, tensor<128x32x32x1xf32, #CCCD>, tensor<128x32x32x1xf32, #CCCD>
+}
diff --git a/mlir/test/Dialect/SparseTensor/fuse_sparse_convert_into_producer.mlir b/mlir/test/Dialect/SparseTensor/fuse_sparse_convert_into_producer.mlir
new file mode 100644
index 000000000000..efa92e565ba5
--- /dev/null
+++ b/mlir/test/Dialect/SparseTensor/fuse_sparse_convert_into_producer.mlir
@@ -0,0 +1,78 @@
+// RUN: mlir-opt %s --pre-sparsification-rewrite --sparse-reinterpret-map  | FileCheck %s --check-prefix=CHECK-FOLD
+// RUN: mlir-opt %s --pre-sparsification-rewrite --sparse-reinterpret-map --sparsification | FileCheck %s
+
+#trait = {
+  indexing_maps = [
+      affine_map<(d0, d1, d2, d3) -> (d0, d1, d2, d3)>,
+      affine_map<(d0, d1, d2, d3) -> (d0, d1, d2, d3)>,
+      affine_map<(d0, d1, d2, d3) -> (d0, d1, d2, d3)>,
+      affine_map<(d0, d1, d2, d3) -> (d0, d1, d2, d3)>
+  ],
+  iterator_types = ["parallel", "parallel", "parallel", "parallel"]
+}
+
+#map = affine_map<(d0, d1, d2) -> (d0, d1, d2)>
+
+#COO = #sparse_tensor.encoding<{map = (d0, d1, d2) -> (d0 : compressed(nonunique), d1 : singleton(nonunique, soa), d2 : singleton(soa))}>
+#CCCD = #sparse_tensor.encoding<{ map = (d0, d1, d2, d3) -> (d0 : compressed, d1 : compressed, d2 : compressed, d3 : dense) }>
+
+// CHECK-LABEL:   func.func @fold_convert(
+// CHECK:           scf.for
+// CHECK:             scf.for
+// CHECK:               scf.for
+// CHECK:                 scf.if
+// CHECK-NEXT:               tensor.insert
+// CHECK-NEXT:               scf.yield
+// CHECK-NEXT:             else
+// CHECK-NEXT:               scf.yield
+// CHECK:                 scf.yield
+// CHECK:               scf.yield
+// CHECK:             scf.yield
+// CHECK:           sparse_tensor.load
+
+// CHECK-FOLD-LABEL:   func.func @fold_convert(
+// CHECK-FOLD-NOT:     sparse_tensor.convert
+func.func @fold_convert(%arg0: tensor<128x32x32x1xf32>, %arg1: tensor<128x32x32x1xf32>, %arg2: tensor<128x32x32x1xf32>) -> tensor<128x32x32x1xf32, #CCCD> {
+  %cst = arith.constant 0.000000e+00 : f32
+  %cst_0 = arith.constant 1.000000e+00 : f32
+  %cst_1 = arith.constant 1.000000e+00 : f32
+  %0 = tensor.empty() : tensor<128x32x32x1xf32>
+  %1 = linalg.generic #trait
+  ins(%arg0, %arg1, %arg2 : tensor<128x32x32x1xf32>, tensor<128x32x32x1xf32>, tensor<128x32x32x1xf32>)
+  outs(%0 : tensor<128x32x32x1xf32>) {
+    ^bb0(%in: f32, %in_2: f32, %in_3: f32, %out: f32):
+      %3 = arith.subf %cst_0, %in_2 : f32
+      %4 = arith.mulf %in, %3 : f32
+      %5 = arith.mulf %4, %cst_1 : f32
+      %6 = arith.addf %5, %in_3 : f32
+      %7 = arith.subf %6, %cst_0 : f32
+      %8 = arith.cmpf uge, %7, %cst : f32
+      %9 = arith.uitofp %8 : i1 to f32
+      linalg.yield %9 : f32
+    } -> tensor<128x32x32x1xf32>
+  %2 = sparse_tensor.convert %1 : tensor<128x32x32x1xf32> to tensor<128x32x32x1xf32, #CCCD>
+  return %2 : tensor<128x32x32x1xf32, #CCCD>
+}
+
+
+// FIXME: The following kernel is not sparsifiable because `arith.select`
+// operations is not handled by the sparse compiler at the moment.
+//
+// CHECK-FOLD-LABEL:   func.func @fold_cast(
+// CHECK-FOLD-NOT:     sparse_tensor.convert
+func.func @fold_cast(%0: tensor<10x20x30xf64, #COO>) -> tensor<10x20x30xf64, #COO> {
+  %cst = arith.constant 0.000000e+00 : f64
+  %1 = tensor.empty() : tensor<10x20x30xf64>
+  %2 = linalg.generic { indexing_maps = [#map, #map],
+                        iterator_types = ["parallel", "parallel", "parallel"]
+                      }
+  ins (%0 : tensor<10x20x30xf64, #COO>)
+  outs(%1 : tensor<10x20x30xf64>) {
+      ^bb0(%in: f64, %out: f64):
+        %4 = arith.cmpf ugt, %in, %cst : f64
+        %5 = arith.select %4, %in, %cst : f64
+        linalg.yield %5 : f64
+  } -> tensor<10x20x30xf64>
+  %cast = tensor.cast %2 : tensor<10x20x30xf64> to tensor<10x20x30xf64, #COO>
+  return %cast : tensor<10x20x30xf64, #COO>
+}
diff --git a/mlir/test/Dialect/SparseTensor/no_fold_into_consumer.mlir b/mlir/test/Dialect/SparseTensor/no_fold_into_consumer.mlir
deleted file mode 100644
index bbc7f397e793..000000000000
--- a/mlir/test/Dialect/SparseTensor/no_fold_into_consumer.mlir
+++ /dev/null
@@ -1,47 +0,0 @@
-// RUN: mlir-opt %s --canonicalize --pre-sparsification-rewrite | FileCheck %s
-
-#map = affine_map<(d0, d1, d2) -> (d0, d1, d2)>
-
-#sparse = #sparse_tensor.encoding<{
-  map = (d0, d1, d2) ->
-          (d0 : compressed(nonunique),
-	   d1 : singleton(nonunique, soa),
-	   d2 : singleton(soa)),
-  posWidth = 64,
-  crdWidth = 64
-}>
-
-
-module {
-  //
-  // This IR should not end up in an infinite loop trying to fold
-  // the linalg producer into the tensor cast consumer (even though
-  // static sizes can fold, the different encodings cannot). The
-  // cast was sloppy to begin with (but it has been observed by
-  // external sources) and can be easily repaired by the sparsifier.
-  //
-  // CHECK-LABEL: func @avoid_fold
-  // CHECK:       arith.constant
-  // CHECK:       tensor.empty()
-  // CHECK:       linalg.generic
-  // CHECK:       sparse_tensor.convert
-  // CHECK:       return
-  //
-  func.func @avoid_fold(%0: tensor<10x20x30xf64, #sparse>) -> tensor<10x20x30xf64, #sparse> {
-    %1 = tensor.empty() : tensor<10x20x30xf64>
-    %2 = linalg.generic { indexing_maps = [#map, #map],
-                          iterator_types = ["parallel", "parallel", "parallel"]
-                        }
-    ins (%0 : tensor<10x20x30xf64, #sparse>)
-    outs(%1 : tensor<10x20x30xf64>) {
-        ^bb0(%in: f64, %out: f64):
-          %cst = arith.constant 0.000000e+00 : f64
-          %4 = arith.cmpf ugt, %in, %cst : f64
-          %5 = arith.select %4, %in, %cst : f64
-          linalg.yield %5 : f64
-    } -> tensor<10x20x30xf64>
-    %cast = tensor.cast %2 : tensor<10x20x30xf64> to tensor<10x20x30xf64, #sparse>
-    return %cast : tensor<10x20x30xf64, #sparse>
-  }
-}
-
diff --git a/mlir/test/Dialect/SparseTensor/roundtrip_encoding.mlir b/mlir/test/Dialect/SparseTensor/roundtrip_encoding.mlir
index 66e61afd897d..7eeda9a98802 100644
--- a/mlir/test/Dialect/SparseTensor/roundtrip_encoding.mlir
+++ b/mlir/test/Dialect/SparseTensor/roundtrip_encoding.mlir
@@ -22,6 +22,64 @@ func.func private @sparse_csr(tensor<?x?xf32, #CSR>)
 
 // -----
 
+#CSR_OnlyOnes = #sparse_tensor.encoding<{
+  map = (d0, d1) -> (d0 : dense, d1 : compressed),
+  posWidth = 64,
+  crdWidth = 64,
+  explicitVal = 1.0 : f32,
+  implicitVal = 0.0 : f32
+}>
+
+// CHECK: #[[$CSR_OnlyOnes:.*]] = #sparse_tensor.encoding<{ map = (d0, d1) -> (d0 : dense, d1 : compressed), posWidth = 64, crdWidth = 64, explicitVal = 1.000000e+00 : f32, implicitVal = 0.000000e+00 : f32 }>
+// CHECK-LABEL: func private @sparse_csr(
+// CHECK-SAME: tensor<?x?xf32, #[[$CSR_OnlyOnes]]>)
+func.func private @sparse_csr(tensor<?x?xf32, #CSR_OnlyOnes>)
+
+// -----
+
+#CSR_OnlyOnes = #sparse_tensor.encoding<{
+  map = (d0, d1) -> (d0 : dense, d1 : compressed),
+  explicitVal = 1.0 : f64,
+  implicitVal = 0.0 : f64
+}>
+
+// CHECK: #[[$CSR_OnlyOnes:.*]] = #sparse_tensor.encoding<{ map = (d0, d1) -> (d0 : dense, d1 : compressed), explicitVal = 1.000000e+00 : f64, implicitVal = 0.000000e+00 : f64 }>
+// CHECK-LABEL: func private @sparse_csr(
+// CHECK-SAME: tensor<?x?xf64, #[[$CSR_OnlyOnes]]>)
+func.func private @sparse_csr(tensor<?x?xf64, #CSR_OnlyOnes>)
+
+// -----
+
+#CSR_OnlyOnes = #sparse_tensor.encoding<{
+  map = (d0, d1) -> (d0 : dense, d1 : compressed),
+  posWidth = 64,
+  crdWidth = 64,
+  explicitVal = 1 : i32,
+  implicitVal = 0 : i32
+}>
+
+// CHECK: #[[$CSR_OnlyOnes:.*]] = #sparse_tensor.encoding<{ map = (d0, d1) -> (d0 : dense, d1 : compressed), posWidth = 64, crdWidth = 64, explicitVal = 1 : i32, implicitVal = 0 : i32 }>
+// CHECK-LABEL: func private @sparse_csr(
+// CHECK-SAME: tensor<?x?xi32, #[[$CSR_OnlyOnes]]>)
+func.func private @sparse_csr(tensor<?x?xi32, #CSR_OnlyOnes>)
+
+// -----
+
+#CSR_OnlyOnes = #sparse_tensor.encoding<{
+  map = (d0, d1) -> (d0 : dense, d1 : compressed),
+  posWidth = 64,
+  crdWidth = 64,
+  explicitVal = 1 : i64,
+  implicitVal = 0 : i64
+}>
+
+// CHECK: #[[$CSR_OnlyOnes:.*]] = #sparse_tensor.encoding<{ map = (d0, d1) -> (d0 : dense, d1 : compressed), posWidth = 64, crdWidth = 64, explicitVal = 1 : i64, implicitVal = 0 : i64 }>
+// CHECK-LABEL: func private @sparse_csr(
+// CHECK-SAME: tensor<?x?xi64, #[[$CSR_OnlyOnes]]>)
+func.func private @sparse_csr(tensor<?x?xi64, #CSR_OnlyOnes>)
+
+// -----
+
 #BCSR = #sparse_tensor.encoding<{
   map = (d0, d1, d2) -> (d0 : batch, d1: dense, d2 : compressed),
 }>
diff --git a/mlir/test/Dialect/Tensor/canonicalize.mlir b/mlir/test/Dialect/Tensor/canonicalize.mlir
index 751c57eacd7a..9a4dd2f3b5cc 100644
--- a/mlir/test/Dialect/Tensor/canonicalize.mlir
+++ b/mlir/test/Dialect/Tensor/canonicalize.mlir
@@ -2431,6 +2431,15 @@ func.func @reshape_nofold_2d(%arg0 : tensor<?x?xi32>) -> tensor<?x?xi32> {
   return %reshape : tensor<?x?xi32>
 }
 
+// -----
+
+// CHECK-LABEL: @reshape_nofold_2d_ins
+func.func @reshape_nofold_2d_ins(%arg0 : tensor<?x?xi32>, %arg1: index, %arg2: index) -> tensor<?x?xi32> {
+  %ds = tensor.from_elements %arg1, %arg2 : tensor<2xindex>
+  // CHECK: tensor.reshape
+  %reshape = tensor.reshape %arg0(%ds) : (tensor<?x?xi32>, tensor<2xindex>) -> tensor<?x?xi32>
+  return %reshape : tensor<?x?xi32>
+}
 
 // -----
 
diff --git a/mlir/test/Dialect/Vector/vector-multi-reduction-lowering.mlir b/mlir/test/Dialect/Vector/vector-multi-reduction-lowering.mlir
index 22808aa7d6ac..f70d23a19322 100644
--- a/mlir/test/Dialect/Vector/vector-multi-reduction-lowering.mlir
+++ b/mlir/test/Dialect/Vector/vector-multi-reduction-lowering.mlir
@@ -281,6 +281,23 @@ func.func private @scalable_dims(%A : vector<8x[4]x2xf32>, %B: vector<8x[4]xf32>
 // CHECK:           %[[VAL_163:.*]] = vector.shape_cast %[[VAL_162]] : vector<[32]xf32> to vector<8x[4]xf32>
 // CHECK:           return %[[VAL_163]] : vector<8x[4]xf32>
 
+// Check that OneDimMultiReductionToTwoDim handles scalable dim
+func.func @scalable_dim_1d(%A: vector<[4]xf32>, %B: f32, %C: vector<[4]xi1>) -> f32 {
+    %0 = vector.mask %C { vector.multi_reduction <add>, %A, %B [0] : vector<[4]xf32> to f32 } : vector<[4]xi1> -> f32
+    return %0 : f32
+}
+
+// CHECK-LABEL:  func.func @scalable_dim_1d(
+// CHECK-SAME:                                      %[[ARG_0:.*]]: vector<[4]xf32>,
+// CHECK-SAME:                                      %[[ARG_1:.*]]: f32,
+// CHECK-SAME:                                      %[[ARG_2:.*]]: vector<[4]xi1>) -> f32 {
+// CHECK-DAG:      %[[VAL_0:.*]] = arith.constant 0 : index
+// CHECK-DAG:      %[[VAL_1:.*]] = arith.constant dense<0.000000e+00> : vector<1xf32>
+// CHECK:          %[[VAL_2:.*]] = vector.mask %[[ARG_2]] { vector.reduction <add>, %[[ARG_0]], %[[ARG_1]] : vector<[4]xf32> into f32 } : vector<[4]xi1> -> f32
+// CHECK:          %[[VAL_3:.*]] = vector.insertelement %[[VAL_2]], %[[VAL_1]][%[[VAL_0]] : index] : vector<1xf32>
+// CHECK:          %[[VAL_4:.*]] = vector.extract %[[VAL_3]][0] : f32 from vector<1xf32>
+// CHECK:          return %[[VAL_4]] : f32
+
 module attributes {transform.with_named_sequence} {
   transform.named_sequence @__transform_main(%root : !transform.any_op {transform.readonly}) {
     %func_op = transform.structured.match ops{["func.func"]} in %root : (!transform.any_op) -> !transform.op<"func.func">
diff --git a/mlir/test/Dialect/XeGPU/XeGPUOps.mlir b/mlir/test/Dialect/XeGPU/XeGPUOps.mlir
index f0945c79a94a..00d32d2a2ee9 100644
--- a/mlir/test/Dialect/XeGPU/XeGPUOps.mlir
+++ b/mlir/test/Dialect/XeGPU/XeGPUOps.mlir
@@ -80,7 +80,7 @@ gpu.func @test_prefetch_vc(%src: ui64) {
   //CHECK: %[[R0:.*]] = xegpu.create_tdesc %arg0 [0, 8, 16, 24] {chunk_size = 2 : i64} : ui64 -> !xegpu.tensor_desc<4x2xf32, #xegpu.tdesc_attr<scattered = true>>
   %1 = xegpu.create_tdesc %src[0, 8, 16, 24] {chunk_size = 2} : ui64  -> !xegpu.tensor_desc<4x2xf32, #xegpu.tdesc_attr<scattered = true>>
   // CHECK: xegpu.prefetch %[[R0]] <{l1_hint = #xegpu.cache_hint<cached>, l2_hint = #xegpu.cache_hint<uncached>}> : !xegpu.tensor_desc<4x2xf32, #xegpu.tdesc_attr<scattered = true>>
-  xegpu.prefetch %1 <{l1_hint = #xegpu.cache_hint<cached>, l2_hint = #xegpu.cache_hint<uncached>}>: !xegpu.tensor_desc<4x2xf32, #xegpu.tdesc_attr<scattered = true>> 
+  xegpu.prefetch %1 <{l1_hint = #xegpu.cache_hint<cached>, l2_hint = #xegpu.cache_hint<uncached>}>: !xegpu.tensor_desc<4x2xf32, #xegpu.tdesc_attr<scattered = true>>
   gpu.return
 }
 
@@ -121,4 +121,59 @@ gpu.func @test_create_update_tdesc_vc(%src: ui64) {
   gpu.return
 }
 
-}
\ No newline at end of file
+// CHECK: gpu.func @test_dpas_vc(%[[arg0:.*]]: vector<8x8x2xf16>, %[[arg1:.*]]: vector<8x16x2xf16>)
+gpu.func @test_dpas_vc(%a : vector<8x8x2xf16>, %b: vector<8x16x2xf16>) {
+  // CHECK: %0 = xegpu.dpas %[[arg0]], %[[arg1]] : vector<8x8x2xf16>, vector<8x16x2xf16> -> vector<8x16xf32>
+  %1 = xegpu.dpas %a, %b: vector<8x8x2xf16>, vector<8x16x2xf16> -> vector<8x16xf32>
+  gpu.return
+}
+
+// CHECK: gpu.func @test_atomic_rmw(%[[arg0:.*]]: ui64, %[[arg1:.*]]: vector<16xf32>, %[[arg2:.*]]: vector<16xi1>)
+gpu.func @test_atomic_rmw(%src: ui64, %value : vector<16xf32>, %mask : vector<16xi1>) {
+  //CHECK: %[[R0:.*]] = xegpu.create_tdesc %[[arg0]] [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15] : ui64 -> !xegpu.tensor_desc<16xf32, #xegpu.tdesc_attr<scattered = true>>
+  %1 = xegpu.create_tdesc %src[0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15]: ui64 -> !xegpu.tensor_desc<16xf32, #xegpu.tdesc_attr<scattered = true>>
+  //CHECK: %[[R1:.*]] = xegpu.atomic_rmw addf %[[R0]], %[[arg2]], %[[arg1]] : <16xf32, #xegpu.tdesc_attr<scattered = true>>, vector<16xi1>, vector<16xf32> -> vector<16xf32>
+  xegpu.atomic_rmw addf %1, %mask, %value: !xegpu.tensor_desc<16xf32, #xegpu.tdesc_attr<scattered = true>>, vector<16xi1>, vector<16xf32> -> vector<16xf32>
+  gpu.return
+}
+
+// CHECK: gpu.func @alloc_nbarrier({{.*}}) {
+gpu.func @alloc_nbarrier() {
+  // CHECK: xegpu.alloc_nbarrier
+  xegpu.alloc_nbarrier 8
+  gpu.return
+}
+
+// CHECK: gpu.func @init_nbarrier({{.*}}) {
+gpu.func @init_nbarrier() {
+  //CHECK: %[[c1:.*]] = arith.constant 1 : i8
+  //CHECK: %[[c16:.*]] = arith.constant 16 : i8
+  %nbarrier_id = arith.constant 1 : i8
+  %threads_count = arith.constant 16 : i8
+  //CHECK: xegpu.init_nbarrier %[[c1]], %[[c16]] : i8, i8 -> !xegpu.nbarrier
+  %nbarrier = xegpu.init_nbarrier %nbarrier_id, %threads_count : i8, i8 -> !xegpu.nbarrier
+  gpu.return
+}
+
+// CHECK: gpu.func @nbarrier_arrive(%[[arg0:.*]]: !xegpu.nbarrier) {
+gpu.func @nbarrier_arrive(%nbarrier : !xegpu.nbarrier) {
+  //CHECK: xegpu.nbarrier_arrive %[[arg0]] : !xegpu.nbarrier
+  xegpu.nbarrier_arrive %nbarrier : !xegpu.nbarrier
+  gpu.return
+}
+
+// CHECK: gpu.func @nbarrier_wait(%[[arg0:.*]]: !xegpu.nbarrier) {
+gpu.func @nbarrier_wait(%nbarrier : !xegpu.nbarrier) {
+  //CHECK: xegpu.nbarrier_wait %[[arg0]] : !xegpu.nbarrier
+  xegpu.nbarrier_wait %nbarrier : !xegpu.nbarrier
+  gpu.return
+}
+
+// CHECK-LABEL: gpu.func @fence({{.*}}) {
+gpu.func @fence() {
+  //CHECK: xegpu.fence memory_kind = global, fence_scope = workgroup
+  xegpu.fence memory_kind = global, fence_scope = workgroup
+  gpu.return
+}
+
+}
diff --git a/mlir/test/Dialect/XeGPU/invalid.mlir b/mlir/test/Dialect/XeGPU/invalid.mlir
index 5e29361ec690..7819ad60b97d 100644
--- a/mlir/test/Dialect/XeGPU/invalid.mlir
+++ b/mlir/test/Dialect/XeGPU/invalid.mlir
@@ -156,4 +156,32 @@ func.func @test_store_scatter_vc_2(%src: ui64) {
   xegpu.store %1, %2, %0 <{l1_hint = #xegpu.cache_hint<streaming>}> : vector<4x2xf32>,
           !xegpu.tensor_desc<4x2xf32, #xegpu.tdesc_attr<scattered = true>>, vector<4xi1>
   return
+}
+
+// -----
+func.func @test_dpas_vc_1(%a : vector<8x4x2xf16>, %b: vector<8x16x2xf16>) {
+  // expected-error@+1 {{K-dimension or vnni-factor mismatch}}
+  %1 = xegpu.dpas %a, %b : vector<8x4x2xf16>, vector<8x16x2xf16> -> vector<8x16xf32>
+  return
+}
+
+// -----
+func.func @test_dpas_vc_2(%a : vector<8x16xf16>, %b: vector<8x16x2xf16>) {
+  // expected-error@+1 {{lhs and rhs rank does not match for dpas op, or their rank is not 3}}
+  %1 = xegpu.dpas %a, %b : vector<8x16xf16>, vector<8x16x2xf16> -> vector<8x16xf32>
+  return
+}
+
+// -----
+func.func @test_dpas_vc_3(%a : vector<8x16xf16>, %b: vector<16x16xf16>) {
+  // expected-error@+1 {{lhs and rhs rank does not match for dpas op, or their rank is not 3}}
+  %1 = xegpu.dpas %a, %b : vector<8x16xf16>, vector<16x16xf16> -> vector<8x16xf32>
+  return
+}
+
+// -----
+func.func @test_dpas_vc_4(%a : vector<8x8x2xf16>, %b: vector<8x16x2xf16>, %c : vector<8x16xf16>) {
+  // expected-error@+1 {{Accumulator and Result for dpas op should have the same type}}
+  %1 = xegpu.dpas %a, %b, %c : vector<8x8x2xf16>, vector<8x16x2xf16>, vector<8x16xf16> -> vector<8x16xf32>
+  return
 }
\ No newline at end of file
diff --git a/mlir/test/Examples/NVGPU/Ch0.py b/mlir/test/Examples/NVGPU/Ch0.py
new file mode 100644
index 000000000000..8f60088178d1
--- /dev/null
+++ b/mlir/test/Examples/NVGPU/Ch0.py
@@ -0,0 +1,50 @@
+# RUN: env SUPPORT_LIB=%mlir_cuda_runtime \
+# RUN:   %PYTHON %s | FileCheck %s
+
+# ===----------------------------------------------------------------------===//
+#  Chapter 0 : Hello World
+# ===----------------------------------------------------------------------===//
+#
+# This program demonstrates Hello World:
+#   1. Build MLIR function with arguments
+#   2. Build MLIR GPU kernel
+#   3. Print from a GPU thread
+#   4. Pass arguments, JIT compile and run the MLIR function
+#
+# ===----------------------------------------------------------------------===//
+
+
+from mlir.dialects import gpu
+from tools.nvdsl import *
+
+
+# 1. The decorator generates a MLIR func.func.
+# Everything inside the Python function becomes the body of the func.
+# The decorator also translates `alpha` to an `index` type.
+@NVDSL.mlir_func
+def main(alpha):
+    # 2. The decorator generates a MLIR gpu.launch.
+    # Everything inside the Python function becomes the body of the gpu.launch.
+    # This allows for late outlining of the GPU kernel, enabling optimizations
+    # like constant folding from host to device.
+    @NVDSL.mlir_gpu_launch(grid=(1, 1, 1), block=(4, 1, 1))
+    def kernel():
+        tidx = gpu.thread_id(gpu.Dimension.x)
+        # + operator generates arith.addi
+        myValue = alpha + tidx
+        # Print from a GPU thread
+        gpu.printf("GPU thread %llu has %llu\n", [tidx, myValue])
+
+    # 3. Call the GPU kernel
+    kernel()
+
+
+alpha = 100
+# 4. The `mlir_func` decorator JIT compiles the IR and executes the MLIR function.
+main(alpha)
+
+
+# CHECK: GPU thread 0 has 100
+# CHECK: GPU thread 1 has 101
+# CHECK: GPU thread 2 has 102
+# CHECK: GPU thread 3 has 103
diff --git a/mlir/test/Examples/NVGPU/Ch1.py b/mlir/test/Examples/NVGPU/Ch1.py
new file mode 100644
index 000000000000..da65aa2ef6a1
--- /dev/null
+++ b/mlir/test/Examples/NVGPU/Ch1.py
@@ -0,0 +1,66 @@
+# RUN: env SUPPORT_LIB=%mlir_cuda_runtime \
+# RUN:   %PYTHON %s | FileCheck %s
+
+# ===----------------------------------------------------------------------===//
+#  Chapter 1 : 2D Saxpy
+# ===----------------------------------------------------------------------===//
+#
+# This program demonstrates 2D Saxpy:
+#  1. Use GPU dialect to allocate and copy memory host to gpu and vice versa
+#  2. Computes 2D SAXPY kernel using operator overloading
+#  3. Pass numpy arrays to MLIR as memref arguments
+#  4. Verify MLIR program with reference computation in python
+#
+# ===----------------------------------------------------------------------===//
+
+
+from mlir import ir
+from mlir.dialects import gpu, memref
+from tools.nvdsl import *
+import numpy as np
+
+
+@NVDSL.mlir_func
+def saxpy(x, y, alpha):
+    # 1. Use MLIR GPU dialect to allocate and copy memory
+    token_ty = ir.Type.parse("!gpu.async.token")
+    t1 = gpu.wait(token_ty, [])
+    x_dev, t2 = gpu.alloc(x.type, token_ty, [t1], [], [])
+    y_dev, t3 = gpu.alloc(y.type, token_ty, [t2], [], [])
+    t4 = gpu.memcpy(token_ty, [t3], x_dev, x)
+    t5 = gpu.memcpy(token_ty, [t4], y_dev, y)
+    t6 = gpu.wait(token_ty, [t5])
+
+    # 2. Compute 2D SAXPY kernel
+    @NVDSL.mlir_gpu_launch(grid=(M, 1, 1), block=(N, 1, 1))
+    def saxpy_kernel():
+        bidx = gpu.block_id(gpu.Dimension.x)
+        tidx = gpu.thread_id(gpu.Dimension.x)
+        x_val = memref.load(x_dev, [bidx, tidx])
+        y_val = memref.load(y_dev, [bidx, tidx])
+
+        # SAXPY: y[i] += a * x[i];
+        y_val += x_val * alpha
+
+        memref.store(y_val, y_dev, [bidx, tidx])
+
+    saxpy_kernel()
+
+    t7 = gpu.memcpy(token_ty, [t6], y, y_dev)
+    gpu.wait(token_ty, [t7])
+
+
+# 3. Pass numpy arrays to MLIR
+M = 256
+N = 32
+alpha = 2.0
+x = np.random.randn(M, N).astype(np.float32)
+y = np.ones((M, N), np.float32)
+saxpy(x, y, alpha)
+
+#  4. Verify MLIR with reference computation
+ref = np.ones((M, N), np.float32)
+ref += x * alpha
+np.testing.assert_allclose(y, ref, rtol=5e-03, atol=1e-01)
+print("PASS")
+# CHECK-NOT: Mismatched elements
diff --git a/mlir/test/Examples/NVGPU/Ch2.py b/mlir/test/Examples/NVGPU/Ch2.py
new file mode 100644
index 000000000000..78c14cb2c7ad
--- /dev/null
+++ b/mlir/test/Examples/NVGPU/Ch2.py
@@ -0,0 +1,93 @@
+# RUN: env SUPPORT_LIB=%mlir_cuda_runtime \
+# RUN:   %PYTHON %s | FileCheck %s
+
+# ===----------------------------------------------------------------------===//
+#  Chapter 2 : 2D Saxpy with TMA
+# ===----------------------------------------------------------------------===//
+#
+# This program demonstrates 2D Saxpy. It is same as Chapter 1,
+# but it loads data using TMA (Tensor Memory Accelerator)
+#
+# This chapter introduces demonstrates:
+#  1. Computes 2D SAXPY in the same way as Ch1.py but loads data using TMA
+#  2. Create and initialize 1 asynchronous transactional barrier (mbarrier)
+#  3. Thread-0 Load request data load from TMA for each thread block
+#  4. Each thread block loads <1x32xf32> for x and y.
+#  5. Wait for completion of TMA load with mbarrier
+#
+# ===----------------------------------------------------------------------===//
+
+from mlir import ir
+from mlir.dialects import nvgpu, scf, arith, memref, vector, gpu
+from tools.nvdsl import *
+from mlir import runtime as rt
+from mlir.extras import types as T
+import numpy as np
+
+
+@NVDSL.mlir_func
+def saxpy(x, y, alpha):
+    token_ty = ir.Type.parse("!gpu.async.token")
+    t1 = gpu.wait(token_ty, [])
+    x_dev, t2 = gpu.alloc(x.type, token_ty, [t1], [], [])
+    y_dev, t3 = gpu.alloc(y.type, token_ty, [t2], [], [])
+    t4 = gpu.memcpy(token_ty, [t3], x_dev, x)
+    t5 = gpu.memcpy(token_ty, [t4], y_dev, y)
+    t6 = gpu.wait(token_ty, [t5])
+
+    x_tma = TMA([1, N], x.type)
+    y_tma = TMA([1, N], y.type)
+    x_tma.create_descriptor(x_dev)
+    y_tma.create_descriptor(y_dev)
+    sz_x = get_type_size(x_tma.tma_memref)
+    sz_y = get_type_size(x_tma.tma_memref)
+    sz = sz_x + sz_y
+
+    @NVDSL.mlir_gpu_launch(grid=(M, 1, 1), block=(N, 1, 1), smem=sz)
+    def saxpy_tma_kernel():
+        bidx = gpu.block_id(gpu.Dimension.x)
+        tidx = gpu.thread_id(gpu.Dimension.x)
+        isThread0 = tidx == 0
+
+        # 1. Create and initialize asynchronous transactional barrier (mbarrier)
+        mbar_group = Mbarriers(number_of_barriers=1)
+        mbar_group[0].init(1, predicate=isThread0)
+
+        # 2. Execute Tensor Memory Accelerator (TMA) Load
+        x_smem = get_dynamic_shared_memory([1, N], T.f32())
+        y_smem = get_dynamic_shared_memory([1, N], T.f32(), offset=sz_x)
+        x_tma.load(x_smem, mbar_group[0], coords=[0, bidx], predicate=isThread0)
+        y_tma.load(y_smem, mbar_group[0], coords=[0, bidx], predicate=isThread0)
+        mbar_group[0].arrive(txcount=sz, predicate=isThread0)
+
+        # 3. Wait for completion of TMA load with mbarrier
+        mbar_group[0].try_wait()
+
+        x_val = memref.load(x_smem, [const(0), tidx])
+        y_val = memref.load(y_smem, [const(0), tidx])
+
+        # SAXPY: y[i] += a * x[i];
+        y_val += x_val * alpha
+
+        memref.store(y_val, y_dev, [bidx, tidx])
+
+    saxpy_tma_kernel()
+
+    t7 = gpu.memcpy(token_ty, [t6], y, y_dev)
+    gpu.wait(token_ty, [t7])
+
+
+# 3. Pass numpy arrays to MLIR
+M = 256
+N = 32
+alpha = 2.0
+x = np.random.randn(M, N).astype(np.float32)
+y = np.ones((M, N), np.float32)
+saxpy(x, y, alpha)
+
+#  4. Verify MLIR with reference computation
+ref = np.ones((M, N), np.float32)
+ref += x * alpha
+np.testing.assert_allclose(y, ref, rtol=5e-03, atol=1e-01)
+print("PASS")
+# CHECK-NOT: Mismatched elements
diff --git a/mlir/test/Examples/NVGPU/Ch3.py b/mlir/test/Examples/NVGPU/Ch3.py
new file mode 100644
index 000000000000..a417014de8b4
--- /dev/null
+++ b/mlir/test/Examples/NVGPU/Ch3.py
@@ -0,0 +1,129 @@
+# RUN: env SUPPORT_LIB=%mlir_cuda_runtime \
+# RUN:   %PYTHON %s | FileCheck %s
+
+# ===----------------------------------------------------------------------===//
+#  Chapter 3 : GEMM 128x128x64 with Tensor Core
+# ===----------------------------------------------------------------------===//
+#
+# This program demonstrates a GEMM operation with 128x128x64 matrix multiplication
+#
+# This chapter introduces demonstrates:
+# 1. Execute TMA Load for two input matrices
+# 2. Performs Tensor Core GEMM 128x128x64 by warpgroup
+# 3. Stores fragmented registers to global memory by warpgroup
+#
+# ===----------------------------------------------------------------------===//
+
+
+from mlir import ir
+from mlir.dialects import nvgpu, scf, arith, memref, vector, gpu
+from tools.nvdsl import *
+from mlir.extras import types as T
+import numpy as np
+
+
+def tma_load(
+    mbar_group: Mbarriers,
+    a_tma: TMA,
+    b_tma: TMA,
+    p,
+):
+    """
+    TMA loads two input matrices from global memory to shared memory. It performs the following operations:
+
+       - tma.load a_shared_memory[0] at coordinate [0, 0]  (Loads 128x64)
+       - tma.load b_shared_memory[0] at coordinate [0, 0]  (Loads 64x64)
+       - tma.load b_shared_memory[0] at coordinate [64, 0] (Loads 64x64)
+
+       mbarrier.arrive ta_count = 128x64xf16 + 64x128xf16
+    """
+
+    size_tma_a = get_type_size(a_tma.tma_memref)
+    size_tma_b = get_type_size(b_tma.tma_memref)
+    ta_count = size_tma_a + (size_tma_b * 2)
+
+    off_b = size_tma_a
+    off_b2 = off_b + size_tma_b
+    a_elem_ty = a_tma.tma_memref.element_type
+    b_elem_ty = b_tma.tma_memref.element_type
+    a = get_dynamic_shared_memory(a_tma.tma_memref.shape, a_elem_ty)
+    b1 = get_dynamic_shared_memory(b_tma.tma_memref.shape, b_elem_ty, off_b)
+    b2 = get_dynamic_shared_memory(b_tma.tma_memref.shape, b_elem_ty, off_b2)
+
+    mbar_group[0].arrive(ta_count, predicate=p)
+
+    a_tma.load(a, mbar_group[0], coords=[0, 0], predicate=p)
+    b_tma.load(b1, mbar_group[0], coords=[0, 0], predicate=p)
+    b_tma.load(b2, mbar_group[0], coords=[64, 0], predicate=p)
+
+
+@NVDSL.mlir_func
+def gemm_128_128_64(a, b, d):
+    token_ty = ir.Type.parse("!gpu.async.token")
+    t1 = gpu.wait(token_ty, [])
+    a_dev, t2 = gpu.alloc(a.type, token_ty, [t1], [], [])
+    b_dev, t3 = gpu.alloc(b.type, token_ty, [t2], [], [])
+    d_dev, t4 = gpu.alloc(d.type, token_ty, [t3], [], [])
+    t5 = gpu.memcpy(token_ty, [t4], a_dev, a)
+    t6 = gpu.memcpy(token_ty, [t5], b_dev, b)
+    t7 = gpu.wait(token_ty, [t6])
+
+    sw = nvgpu.TensorMapSwizzleKind.SWIZZLE_128B
+    a_tma = TMA([128, 64], a.type, swizzle=sw)
+    b_tma = TMA([64, 64], b.type, swizzle=sw)
+    a_tma.create_descriptor(a_dev)
+    b_tma.create_descriptor(b_dev)
+    a_size = get_type_size(a.type)
+    b_size = get_type_size(b.type)
+    smem_size_in_bytes = a_size + b_size
+
+    @NVDSL.mlir_gpu_launch(grid=(1, 1, 1), block=(128, 1, 1), smem=smem_size_in_bytes)
+    def gemm_tma_kernel():
+        tidx = gpu.thread_id(gpu.Dimension.x)
+
+        mbar_group = Mbarriers(number_of_barriers=1)
+        isThread0 = tidx == 0
+
+        mbar_group[0].init(1, predicate=isThread0)
+        a_tma.prefetch(predicate=isThread0)
+        b_tma.prefetch(predicate=isThread0)
+
+        a_smem = get_dynamic_shared_memory((M, K), T.f16())
+        b_smem = get_dynamic_shared_memory((K, N), T.f16(), offset=a_size)
+
+        # 1. TMA Load for two input matrices
+        tma_load(mbar_group, a_tma, b_tma, isThread0)
+
+        # 2. All threads wait TMA load completion
+        mbar_group[0].try_wait()
+
+        # 3. Performs Tensor Core GEMM 128x128x64 by warpgroup
+        A = WGMMAMatrix(WGMMAType.Descriptor, [M, K], desc=a_tma, smem=a_smem)
+        B = WGMMAMatrix(WGMMAType.Descriptor, [K, N], desc=b_tma, smem=b_smem)
+        D = WGMMAMatrix(WGMMAType.Accumulator, shape=[M, N], ty=T.f32())
+
+        # Matrix Multiply
+        D += A @ B
+
+        # 4. Stores fragmented registers to global memory by warpgroup
+        D.store_accumulator(d_dev)
+
+    gemm_tma_kernel()
+
+    t8 = gpu.memcpy(token_ty, [t7], d, d_dev)
+    gpu.wait(None, [t8])
+
+
+# Python pass arguments to MLIR
+M = 128
+N = 128
+K = 64
+a = np.random.randn(M, K).astype(np.float16)
+b = np.random.randn(K, N).astype(np.float16)
+d = np.zeros((M, N), np.float32)
+gemm_128_128_64(a, b, d)
+
+ref_d = a.astype(np.float16) @ b.astype(np.float16)
+np.testing.assert_allclose(d, ref_d, rtol=5e-03, atol=1e-01)
+print("PASS")
+# CHECK-NOT: Mismatched elements
diff --git a/mlir/test/Examples/NVGPU/Ch4.py b/mlir/test/Examples/NVGPU/Ch4.py
new file mode 100644
index 000000000000..8f38d8a90add
--- /dev/null
+++ b/mlir/test/Examples/NVGPU/Ch4.py
@@ -0,0 +1,323 @@
+# RUN: env SUPPORT_LIB=%mlir_cuda_runtime \
+# RUN:   %PYTHON %s | FileCheck %s
+
+# ===----------------------------------------------------------------------===//
+#  Chapter 4 : Multistage GEMM with Tensor Core
+# ===----------------------------------------------------------------------===//
+#
+# This program exemplifies a GEMM operation for `f32+=f16*f16`, utilizing the
+# Multistage method with a tile size of 128x128x64. The code completely
+# parallelizes the two outermost loops into thread blocks. It launches one Warp
+# Groups (128 threads in total) and allocates multiple slots/stage in the
+# shared memory. The program consists of three main parts: prologue, mainloop,
+# and epilogue. In the prologue, thread0 requests for TMA to load data into
+# shared memory slots. The mainloop executes MMA while simultaneously loading
+# TMA for the utilized slots. This overlap of TMA and MMA operations enhances
+# performance by maximizing computational throughput.
+#
+# Loops illustration:
+#
+#  for s in range(num_stages):
+#    TMA_128x64_64x128...
+#  for ti in range(M//128):  # -> blockIdx.x
+#   for tj in range(N//128): # -> blockIdx.y
+#    for tk in range(K//64):
+#      MMA_128x128x64...
+#      TMA_128x64_64x128...
+#  Epilogue...
+#
+# This chapter introduces demonstrates:
+#  1. Partition shape based on block IDs
+#  2. Prologue
+#    2.1 Execute TMA Load for two input matrices for each stage
+#  3. Main loop
+#    3.1 Wait for completion of TMA load with mbarrier
+#    3.2 Performs Tensor Core GEMM 64x128x64 by warpgroup
+#    3.3 Load next stage if needed
+#  4. Epilogue
+#    4.1 Store fragmented registers to shared memory
+#    4.2 Store shared memory to global
+#
+# ===----------------------------------------------------------------------===//
+
+
+from mlir import ir
+from mlir.dialects import gpu, scf, nvgpu, nvvm
+from mlir.extras import types as T
+from tools.nvdsl import *
+import numpy as np
+
+
+def partition_shape():
+    """
+    Calculate the partition shape based on the block IDs.
+
+    It partitions the shape like below:
+    for(.. i < M ...)   --> blockIdx.x
+     for(.. j < N ...)  --> blockIdx.y
+      for(.. k < K ...)
+
+    Returns:
+        dimX (int): Dimension along the x-axis.
+        dimY (int): Dimension along the y-axis.
+    """
+    bidx = gpu.block_id(gpu.Dimension.x)
+    bidy = gpu.block_id(gpu.Dimension.y)
+    dimX = bidx * TILE_M
+    dimY = bidy * TILE_N
+    return dimX, dimY
+
+
+def tma_load(
+    mbar_group: Mbarriers,
+    a_tma: TMA,
+    b_tma: TMA,
+    slot,
+    stage,
+    num_stages,
+    p=None,
+):
+    """
+    TMA loads two input matrices from global memory to shared memory. It performs the following operations:
+
+       - tma.load a_shared_memory[off_x]  at coordinate [x, z]      (Loads 128x64)
+       - tma.load b_shared_memory[off_y1] at coordinate [y, x]      (Loads 64x64)
+       - tma.load b_shared_memory[off_y2] at coordinate [y + 64, x] (Loads 64x64)
+
+       mbarrier.arrive ta_count = 128x64x2x4
+    """
+    dimX, dimY = partition_shape()
+
+    tidx = gpu.thread_id(gpu.Dimension.x)
+    begin_b = num_stages * get_type_size(a_tma.tma_memref)
+    size_tma_a = get_type_size(a_tma.tma_memref)
+    size_tma_b = get_type_size(b_tma.tma_memref)
+    ta_count = size_tma_a + (size_tma_b * 2)
+    tidx = gpu.thread_id(gpu.Dimension.x)
+
+    p = tidx == 0 if p is None else p
+
+    off_a = slot * size_tma_a
+    off_b = (slot * size_tma_a) + begin_b
+    off_b2 = off_b + size_tma_b
+    a_elem_ty = a_tma.tma_memref.element_type
+    b_elem_ty = b_tma.tma_memref.element_type
+    a = get_dynamic_shared_memory(a_tma.tma_memref.shape, a_elem_ty, off_a)
+    b1 = get_dynamic_shared_memory(b_tma.tma_memref.shape, b_elem_ty, off_b)
+    b2 = get_dynamic_shared_memory(b_tma.tma_memref.shape, b_elem_ty, off_b2)
+
+    mbar_group[slot].arrive(ta_count, predicate=p)
+
+    c1 = stage * 64
+    a_tma.load(a, mbar_group[slot], coords=[c1, dimX], predicate=p)
+    b_tma.load(b1, mbar_group[slot], coords=[dimY, c1], predicate=p)
+    b_tma.load(b2, mbar_group[slot], coords=[dimY + 64, c1], predicate=p)
+
+
+def initialize(a_tma: TMA, b_tma: TMA, num_stages):
+    """
+    Initialize mbarriers and prefetch TMA descriptors.
+    """
+    tidx = gpu.thread_id(gpu.Dimension.x)
+    mbar_group = Mbarriers(number_of_barriers=num_stages)
+    isThread0 = tidx == const(0)
+    with ir.InsertionPoint(scf.IfOp(isThread0).then_block):
+        for i in scf.for_(0, num_stages, 1):
+            mbar_group[i].init(1)
+            scf.yield_([])
+        a_tma.prefetch()
+        b_tma.prefetch()
+        scf.yield_([])
+
+    return mbar_group
+
+
+def prologue(mbar_group: Mbarriers, a_tma: TMA, b_tma: TMA, num_stages):
+    """
+    Prologue of the GEMM kernel. It loads 2 input matrices for each stage in loop like below:
+
+    for stage in range(NUM_STAGES):
+        tma_load x, y, stage
+
+    """
+    ns = num_stages if num_stages == 1 else num_stages - 1
+    for iv in scf.for_(0, ns, 1):
+        tma_load(mbar_group, a_tma, b_tma, iv, iv, num_stages)
+        scf.yield_([])
+
+
+def mainloop(mbar_group: Mbarriers, a_tma: TMA, b_tma: TMA, num_stages):
+    """
+    Main loop of the Multistage GEMM kernel. It iterates through
+    stages and performs matrix multiplication, loading data by TMA to shared memory. It like following
+
+    MatrixAccumulator D
+    for k in range(K // TILE_K):
+
+        try_wait(stage, ...)    # Wait TMA load
+
+        Matrix A(stage, ...)    # Find shared memory slot
+        Matrix B(stage, ...)    # Find shared memory slot
+        D += A @ B              # Multiply and accumulate
+
+        if(needLoad)            # Load next stage if needed
+            tma_load(x, y, nextSlot, nextStage)
+
+    """
+    ns = num_stages if num_stages == 1 else num_stages - 1
+
+    tidx = gpu.thread_id(gpu.Dimension.x)
+    begin_b = num_stages * get_type_size(a_tma.tma_memref)
+
+    size_a = TILE_M * TILE_K * get_type_size(T.f16())
+
+    # Initialize A and B (input matrices) and C (accumulator)
+    A = WGMMAMatrix(WGMMAType.Descriptor, [TILE_M, TILE_K], desc=a_tma)
+    B = WGMMAMatrix(WGMMAType.Descriptor, [TILE_K, TILE_N], desc=b_tma)
+    D = WGMMAMatrix(WGMMAType.Accumulator, shape=[TILE_M, TILE_N], ty=T.f32())
+
+    phase = const(False, ty=T.bool())
+
+    # Main Loop
+    for_op = scf.ForOp(const(0), const(K // TILE_K), const(1), [D.acc_op, phase])
+    with ir.InsertionPoint(for_op.body):
+        phase = for_op.inner_iter_args[1]
+        iv = for_op.induction_variable
+        stage = iv % num_stages
+
+        # Wait for current stage
+        mbar_group[stage].try_wait(phase=phase)
+
+        # Find shared memory slot
+        offset_a = stage * size_a
+        offset_b = offset_a + begin_b
+        a_smem = get_dynamic_shared_memory([TILE_M, TILE_K], T.f16(), offset_a)
+        b_smem = get_dynamic_shared_memory([TILE_K, TILE_N], T.f16(), offset_b)
+
+        # Iterate input matrices, update accumulator
+        A.update_smem(a_smem)
+        B.update_smem(b_smem)
+        D.update_accumulator(for_op.inner_iter_args[0])
+
+        # Matrix Multiply
+        D += A @ B
+
+        # Wait Tensor Core for single stage
+        if num_stages == 1:
+            nvvm.WgmmaWaitGroupSyncOp(0)
+
+        # Load next stage
+        pred = ((iv + ns) < const(K // TILE_K)) & (tidx == 0)
+        nextStage = iv + ns
+        nextSlot = nextStage % num_stages
+        tma_load(mbar_group, a_tma, b_tma, nextSlot, nextStage, num_stages, pred)
+
+        # Switch phase parity for the mbarrier
+        newPhase = arith.select(
+            stage == (num_stages - 1),
+            (phase ^ const(True, ty=T.bool())),
+            phase,
+        )
+        scf.yield_([D.acc_op, newPhase])
+
+    nvvm.WgmmaWaitGroupSyncOp(0)
+
+    D.update_accumulator(for_op.results[0])
+    return D
+
+
+def epilogue(D: WGMMAMatrix, d_dev):
+    """
+    Epilogue of the GEMM kernel. It stores the fragmented registers to global memory.
+
+    MatrixAccumulator D               # Fragmented results
+    store D -> Shared Memory          # Store Shared Memory
+    Shared Memory -> Z[dimX][dimY]    # Store Shared Memory to Global Memory
+
+    """
+    tidx = gpu.thread_id(gpu.Dimension.x)
+    dimX, dimY = partition_shape()
+
+    d_smem = get_dynamic_shared_memory([TILE_M, TILE_N], T.f32())
+    d_gmem = memref.subview(d_dev, [dimX, dimY], [TILE_M, TILE_N], [1, 1])
+
+    # Store (registers -> shared memory)
+    D.store_accumulator(d_smem)
+    gpu.barrier()
+
+    # Store (shared memory --> global memory)
+    for i in scf.for_(0, TILE_M, 1):
+        val = memref.load(d_smem, [i, tidx])
+        memref.store(val, d_gmem, [i, tidx])
+        scf.yield_([])
+
+
+# The decorator generates
+#   a -> memref<MxKxf16>
+#   b -> memref<NxKf16>
+#   d -> memref<MxNxf32>
+@NVDSL.mlir_func
+def gemm_multistage(a, b, d, num_stages):
+    token_ty = ir.Type.parse("!gpu.async.token")
+    t1 = gpu.wait(token_ty, [])
+    a_dev, t2 = gpu.alloc(a.type, token_ty, [t1], [], [])
+    b_dev, t3 = gpu.alloc(b.type, token_ty, [t2], [], [])
+    d_dev, t4 = gpu.alloc(d.type, token_ty, [t3], [], [])
+    t5 = gpu.memcpy(token_ty, [t4], a_dev, a)
+    t6 = gpu.memcpy(token_ty, [t5], b_dev, b)
+    t7 = gpu.wait(token_ty, [t6])
+
+    sw = nvgpu.TensorMapSwizzleKind.SWIZZLE_128B
+    a_tma = TMA([128, 64], a.type, swizzle=sw)
+    b_tma = TMA([64, 64], b.type, swizzle=sw)
+    a_tma.create_descriptor(a_dev)
+    b_tma.create_descriptor(b_dev)
+
+    grid = [(M // TILE_M), (N // TILE_N), 1]
+    block = [128, 1, 1]
+
+    size_a = get_type_size(a.type.element_type) * TILE_M * TILE_K
+    size_b = get_type_size(b.type.element_type) * TILE_N * TILE_K
+    smem_size_in_bytes = (size_a + size_b) * num_stages
+
+    @NVDSL.mlir_gpu_launch(grid=grid, block=block, smem=smem_size_in_bytes)
+    def gemm_multistage_kernel():
+        # Initialize mbarriers and prefetch TMA descriptors
+        mbar_group = initialize(a_tma, b_tma, num_stages)
+
+        # Fill the pipeline stages
+        prologue(mbar_group, a_tma, b_tma, num_stages)
+
+        # Main loop
+        D = mainloop(mbar_group, a_tma, b_tma, num_stages)
+
+        # Store registers to global memory
+        epilogue(D, d_dev)
+
+    gemm_multistage_kernel()
+
+    t8 = gpu.memcpy(token_ty, [t7], d, d_dev)
+    gpu.wait(None, [t8])
+
+
+# Python pass arguments to MLIR
+N = 256
+M = 512
+K = 1024
+TILE_M = 128
+TILE_N = 128
+TILE_K = 64
+a = np.random.randn(M, K).astype(np.float16)
+b = np.random.randn(K, N).astype(np.float16)
+d = np.zeros((M, N), np.float32)
+
+gemm_multistage(a, b, d, num_stages=7)
+
+
+# Verify MLIR with reference computation
+ref_d = a.astype(np.float16) @ b.astype(np.float16)
+np.testing.assert_allclose(d, ref_d, rtol=5e-03, atol=1e-01)
+
+
+print("PASS")
+# CHECK-NOT: Mismatched elements
diff --git a/mlir/test/Examples/NVGPU/Ch5.py b/mlir/test/Examples/NVGPU/Ch5.py
new file mode 100644
index 000000000000..92e9314e1b81
--- /dev/null
+++ b/mlir/test/Examples/NVGPU/Ch5.py
@@ -0,0 +1,321 @@
+# RUN: env SUPPORT_LIB=%mlir_cuda_runtime \
+# RUN:   %PYTHON %s | FileCheck %s
+
+# ===----------------------------------------------------------------------===//
+#  Chapter 5 : Warp Specialized GEMM with Tensor Core
+# ===----------------------------------------------------------------------===//
+#
+# This program demonstrates a GEMM operation for `f32+=f16*f16`, utilizing the
+# Warp Specialized method with a tile size of 128x128x64. The code completely
+# parallelizes the two outermost loops into thread blocks. It launches two Warp
+# Groups (256 threads in total): one for the producer and the other for the consumer.
+# Each group takes a different control-flow. The producer thread group is responsible
+# for loading data into shared memory, while the consumer group executes the Tensor
+# Core GEMM operation and epilogue.
+#
+#  for ti in range(M//128):  # -> blockIdx.x
+#   for tj in range(N//128): # -> blockIdx.y
+#    with wg_producer:
+#     for tk in range(K//64):
+#        TMA_128x64_64x128...
+#    with wg_consumer:
+#     for tk in range(K//64):
+#        MMA_128x128x64...
+#     Epilogue..
+#
+# This chapter demonstrates:
+#  2 WG (warpgroups)
+#    Producer:
+#       2.1.1 Wait MMA Barrier
+#       2.1.1 Load TMA with TMA barrier
+#       2.1.1 Arrive TMA barrier with txcount
+#    Consumer:
+#       Loop
+#           Wait TMA barrier
+#           Performs Tensor Core GEMM 64x128x64 by warpgroup
+#           Arrive MMA Barrier
+#       Epilogue
+#           Store fragmented registers to shared memory
+#           Store shared memory to global
+#
+# ===----------------------------------------------------------------------===//
+
+
+from mlir import ir
+from mlir.dialects import gpu, scf, nvgpu, nvvm
+from mlir.extras import types as T
+from tools.nvdsl import *
+import numpy as np
+
+
+def partition_shape():
+    """
+    Calculate the partition shape based on the block IDs.
+
+    It parallelizes the two outermost loops into thread blocks.
+    for ti in range(M//128):    # -> blockIdx.x
+     for tj in range(N//128):   # -> blockIdx.y
+      D = 0
+      for tk in range(K//64):
+       for i in range(128):
+        for j in range(128):
+         for k in range(64):
+           FMA
+
+    Returns:
+        dimX (int): Dimension along the x-axis.
+        dimY (int): Dimension along the y-axis.
+    """
+    bidx = gpu.block_id(gpu.Dimension.x)
+    bidy = gpu.block_id(gpu.Dimension.y)
+    dimX = bidx * TILE_M
+    dimY = bidy * TILE_N
+    return dimX, dimY
+
+
+def tma_load(
+    mbar_group: Mbarriers,
+    a_tma: TMA,
+    b_tma: TMA,
+    slot,
+    stage,
+    num_stages,
+    p=None,
+):
+    """
+    TMA loads two input matrices from global memory to shared memory. It performs the following operations:
+
+       - tma.load a_shared_memory[off_x]  at coordinate [x, z]      (Loads 128x64)
+       - tma.load b_shared_memory[off_y1] at coordinate [y, x]      (Loads 64x64)
+       - tma.load b_shared_memory[off_y2] at coordinate [y + 64, x] (Loads 64x64)
+
+       mbarrier.arrive ta_count = 128x64x2x4
+    """
+    dimX, dimY = partition_shape()
+
+    tidx = gpu.thread_id(gpu.Dimension.x)
+    begin_b = num_stages * get_type_size(a_tma.tma_memref)
+    size_tma_a = get_type_size(a_tma.tma_memref)
+    size_tma_b = get_type_size(b_tma.tma_memref)
+    ta_count = size_tma_a + (size_tma_b * 2)
+
+    off_a = slot * size_tma_a
+    off_b = (slot * size_tma_a) + begin_b
+    off_b2 = off_b + size_tma_b
+    a_elem_ty = a_tma.tma_memref.element_type
+    b_elem_ty = b_tma.tma_memref.element_type
+    a = get_dynamic_shared_memory(a_tma.tma_memref.shape, a_elem_ty, off_a)
+    b1 = get_dynamic_shared_memory(b_tma.tma_memref.shape, b_elem_ty, off_b)
+    b2 = get_dynamic_shared_memory(b_tma.tma_memref.shape, b_elem_ty, off_b2)
+
+    mbar_group[slot].arrive(ta_count, predicate=p)
+    p = (tidx % WARP_GROUP_SIZE) == 0
+    c1 = stage * 64
+    a_tma.load(a, mbar_group[slot], coords=[c1, dimX], predicate=p)
+    b_tma.load(b1, mbar_group[slot], coords=[dimY, c1], predicate=p)
+    b_tma.load(b2, mbar_group[slot], coords=[dimY + 64, c1], predicate=p)
+
+
+def initialize(a_tma: TMA, b_tma: TMA, num_stages):
+    """
+    Initialize mbarriers and prefetch TMA descriptors.
+    """
+    tidx = gpu.thread_id(gpu.Dimension.x)
+    mbar_group_tma = Mbarriers(number_of_barriers=num_stages)
+    mbar_group_mma = Mbarriers(number_of_barriers=num_stages)
+    isThread0 = tidx == const(0)
+    with ir.InsertionPoint(scf.IfOp(isThread0).then_block):
+        for i in scf.for_(0, num_stages, 1):
+            mbar_group_tma[i].init(1)
+            mbar_group_mma[i].init(1)
+            scf.yield_([])
+        a_tma.prefetch()
+        b_tma.prefetch()
+        scf.yield_([])
+
+    return mbar_group_tma, mbar_group_mma
+
+
+def switch_phase(stage, phase, num_stages):
+    p = stage == (num_stages - 1)
+    phase = arith.select(
+        p,
+        (phase ^ const(True, ty=T.bool())),
+        phase,
+    )
+    return phase
+
+
+def producer_loop(
+    mbar_tma: Mbarriers,
+    mbar_mma: Mbarriers,
+    a_tma: TMA,
+    b_tma: TMA,
+    wg_me: Warpgroup,
+    num_stages,
+):
+    phase = const(True, ty=T.bool())
+
+    for iv, phase in scf.for_(0, (K // TILE_K), 1, [phase]):
+        stage = iv % num_stages
+        # Wait MMA to be done
+        mbar_mma[stage].try_wait(phase)
+        # New phase for mbarrier
+        phase = switch_phase(stage, phase, num_stages)
+        # TMA Load
+        tma_load(mbar_tma, a_tma, b_tma, stage, iv, num_stages, wg_me.is_wg_primary)
+        scf.yield_([phase])
+
+
+def consumer_loop(
+    mbar_tma: Mbarriers,
+    mbar_mma: Mbarriers,
+    a_tma: TMA,
+    b_tma: TMA,
+    wg_me: Warpgroup,
+    num_stages,
+):
+    begin_b = num_stages * get_type_size(a_tma.tma_memref)
+
+    size_a = TILE_M * TILE_K * get_type_size(T.f16())
+
+    phase = const(False, ty=T.bool())
+    A = WGMMAMatrix(WGMMAType.Descriptor, [TILE_M, TILE_K], desc=a_tma)
+    B = WGMMAMatrix(WGMMAType.Descriptor, [TILE_K, TILE_N], desc=b_tma)
+    D = WGMMAMatrix(WGMMAType.Accumulator, shape=[TILE_M, TILE_N], ty=T.f32())
+
+    for_op = scf.ForOp(const(0), const(K // TILE_K), const(1), [D.acc_op, phase])
+    with ir.InsertionPoint(for_op.body):
+        phase = for_op.inner_iter_args[1]
+        iv = for_op.induction_variable
+        stage = iv % num_stages
+
+        # Wait TMA for current stage
+        mbar_tma[stage].try_wait(phase)
+
+        # Find shared memory slot
+        offset_a = stage * size_a
+        offset_b = offset_a + begin_b
+        a_smem = get_dynamic_shared_memory([TILE_M, TILE_K], T.f16(), offset_a)
+        b_smem = get_dynamic_shared_memory([TILE_K, TILE_N], T.f16(), offset_b)
+
+        # Iterate input matrices, update accumulator
+        A.update_smem(a_smem)
+        B.update_smem(b_smem)
+        D.update_accumulator(for_op.inner_iter_args[0])
+
+        # Matrix Multiply
+        D += A @ B
+
+        # MMA Barrier Arrive
+        p_arrive = (iv > 0) & wg_me.is_wg_primary
+        with ir.InsertionPoint(scf.IfOp(p_arrive).then_block):
+            barId = arith.select((stage == 0), const(num_stages - 1), (stage - 1))
+            mbar_mma[barId].arrive()
+            scf.yield_([])
+
+        phase = switch_phase(stage, phase, num_stages)
+        scf.yield_([D.acc_op, phase])
+
+    nvvm.WgmmaWaitGroupSyncOp(0)
+    D.update_accumulator(for_op.results[0])
+    return D
+
+
+def epilogue(D: WGMMAMatrix, d_dev):
+    """
+    Epilogue of the GEMM kernel. It stores the fragmented registers to global memory.
+
+    MatrixAccumulator D               # Fragmented results
+    store D -> Shared Memory          # Store Shared Memory
+    Shared Memory -> Z[dimX][dimY]    # Store Shared Memory to Global Memory
+
+    """
+    tidx = gpu.thread_id(gpu.Dimension.x)
+    dimX, dimY = partition_shape()
+    # s = tidx - WARP_GROUP_SIZE
+    # debug_print("[Epilogue] store to global memory @ s={}", s)
+
+    d_smem = get_dynamic_shared_memory([TILE_M, TILE_N], T.f32())
+    d_gmem = memref.subview(d_dev, [dimX, dimY], [TILE_M, TILE_N], [1, 1])
+
+    # Store (registers -> shared memory)
+    D.store_accumulator(d_smem)
+    gpu.barrier()
+
+    # Store (shared memory --> global memory)
+    for i in scf.for_(0, TILE_M, 1):
+        val = memref.load(d_smem, [i, tidx])
+        memref.store(val, d_gmem, [i, tidx])
+        scf.yield_([])
+
+
+@NVDSL.mlir_func
+def gemm_warp_specialized(a, b, d, num_stages):
+    token_ty = ir.Type.parse("!gpu.async.token")
+    t1 = gpu.wait(token_ty, [])
+    a_dev, t2 = gpu.alloc(a.type, token_ty, [t1], [], [])
+    b_dev, t3 = gpu.alloc(b.type, token_ty, [t2], [], [])
+    d_dev, t4 = gpu.alloc(d.type, token_ty, [t3], [], [])
+    t5 = gpu.memcpy(token_ty, [t4], a_dev, a)
+    t6 = gpu.memcpy(token_ty, [t5], b_dev, b)
+    t7 = gpu.wait(token_ty, [t6])
+
+    sw = nvgpu.TensorMapSwizzleKind.SWIZZLE_128B
+    a_tma = TMA([128, 64], a.type, swizzle=sw)
+    b_tma = TMA([64, 64], b.type, swizzle=sw)
+    a_tma.create_descriptor(a_dev)
+    b_tma.create_descriptor(b_dev)
+
+    grid = [(M // TILE_M), (N // TILE_N), 1]
+    block = [256, 1, 1]
+
+    size_a = get_type_size(a.type.element_type) * TILE_M * TILE_K
+    size_b = get_type_size(b.type.element_type) * TILE_N * TILE_K
+    smem_size_in_bytes = (size_a + size_b) * num_stages
+
+    @NVDSL.mlir_gpu_launch(grid=grid, block=block, smem=smem_size_in_bytes)
+    def gemm_warp_specialized_kernel():
+        # Init Warpgroups
+        wg_producer = Warpgroup(primary_thread=128, register_size=40)
+        wg_consumer = Warpgroup(primary_thread=0, register_size=232)
+
+        # Initialize mbarriers and prefetch TMA descriptors
+        mbar_mma, mbar_tma = initialize(a_tma, b_tma, num_stages)
+
+        # Producer performs TMA
+        with wg_producer:
+            producer_loop(mbar_tma, mbar_mma, a_tma, b_tma, wg_producer, num_stages)
+
+        # Consumer performs MMA/Tensor Core
+        with wg_consumer:
+            D = consumer_loop(mbar_tma, mbar_mma, a_tma, b_tma, wg_consumer, num_stages)
+            epilogue(D, d_dev)
+
+    gemm_warp_specialized_kernel()
+
+    t8 = gpu.memcpy(token_ty, [t7], d, d_dev)
+    gpu.wait(None, [t8])
+
+
+# Python pass arguments to MLIR
+N = 256
+M = 512
+K = 1024
+TILE_M = 128
+TILE_N = 128
+TILE_K = 64
+a = np.random.randn(M, K).astype(np.float16)
+b = np.random.randn(K, N).astype(np.float16)
+d = np.zeros((M, N), np.float32)
+
+gemm_warp_specialized(a, b, d, num_stages=7)
+
+
+# Verify MLIR with reference computation
+ref_d = a.astype(np.float16) @ b.astype(np.float16)
+np.testing.assert_allclose(d, ref_d, rtol=5e-03, atol=1e-01)
+
+
+print("PASS")
+# CHECK-NOT: Mismatched elements
diff --git a/mlir/test/Examples/NVGPU/lit.local.cfg b/mlir/test/Examples/NVGPU/lit.local.cfg
new file mode 100644
index 000000000000..689cd252e7a2
--- /dev/null
+++ b/mlir/test/Examples/NVGPU/lit.local.cfg
@@ -0,0 +1,4 @@
+config.unsupported = False
+if not config.enable_cuda_runner or not config.mlir_run_cuda_sm90_tests:
+  config.unsupported = True
+  
\ No newline at end of file
diff --git a/mlir/test/Examples/NVGPU/tools/lit.local.cfg b/mlir/test/Examples/NVGPU/tools/lit.local.cfg
new file mode 100644
index 000000000000..d9f34f219c4d
--- /dev/null
+++ b/mlir/test/Examples/NVGPU/tools/lit.local.cfg
@@ -0,0 +1,3 @@
+# Files in this directory are tools, not tests.
+config.unsupported = True
+
diff --git a/mlir/test/Examples/NVGPU/tools/nvdsl.py b/mlir/test/Examples/NVGPU/tools/nvdsl.py
new file mode 100644
index 000000000000..600cae5b47ee
--- /dev/null
+++ b/mlir/test/Examples/NVGPU/tools/nvdsl.py
@@ -0,0 +1,456 @@
+from enum import Enum
+import functools, sys, ctypes, os, errno
+import numpy as np
+from functools import partialmethod
+from mlir import ir
+from mlir.dialects import arith, func, gpu, memref, nvgpu, scf, nvvm
+from mlir.extras import types as T
+from mlir import runtime as rt
+from tools import nvgpucompiler
+
+MLIR_DYNAMIC = -9223372036854775808
+
+
+def const(value: int, ty=None):
+    ty = T.index() if ty is None else ty
+    if isinstance(value, ir.Value) and (
+        value.type.isinstance(value.type) or T.bool().isinstance(value.type)
+    ):
+        return value
+    return arith.constant(ty, value)
+
+
+def get_type_size(ty):
+    if ir.MemRefType.isinstance(ty):
+        size = get_type_size(ty.element_type)
+        for sz in ty.shape:
+            size *= sz
+        return size
+    if ir.FloatType.isinstance(ty):
+        return ir.FloatType(ty).width // 8
+    if ir.IntegerType.isinstance(ty):
+        return ir.IntegerType(ty).width // 8
+    raise NotImplementedError(ty)
+
+
+def get_mlir_func_obj_ty(inputArgs):
+    args = []
+    c_int_p = ctypes.c_int * 1
+    c_float_p = ctypes.c_float * 1
+    c_bool_p = ctypes.c_bool * 1
+    for arg in inputArgs:
+        if isinstance(arg, bool):
+            args.append(c_bool_p(arg))
+        elif isinstance(arg, int):
+            args.append(c_int_p(arg))
+        elif isinstance(arg, float):
+            args.append(c_float_p(arg))
+        elif isinstance(arg, np.ndarray):
+            args.append(
+                ctypes.pointer(ctypes.pointer(rt.get_ranked_memref_descriptor(arg)))
+            )
+        else:
+            raise NotImplementedError(arg)
+    return args
+
+
+class Mbarriers:
+    def __init__(self, number_of_barriers=1):
+        self.mbar_ty = ir.Type.parse(
+            "!nvgpu.mbarrier.group<memorySpace=#gpu.address_space<workgroup>, num_barriers = "
+            + str(number_of_barriers)
+            + ">"
+        )
+        self.mbar_group_op = nvgpu.mbarrier_create(self.mbar_ty)
+        self.number_of_barriers = number_of_barriers
+
+    def __getitem__(self, key):
+        self.id_op = const(key)
+        return self
+
+    def init(self, count: int, predicate=None):
+        count_op = const(count)
+        if predicate is None:
+            nvgpu.mbarrier_init(self.mbar_group_op, count_op, self.id_op)
+        else:
+            nvgpu.mbarrier_init(
+                self.mbar_group_op, count_op, self.id_op, predicate=predicate
+            )
+
+    def arrive(self, txcount: int = 0, predicate=None):
+        if txcount != 0:
+            txcount_op = const(txcount)
+            nvgpu.mbarrier_arrive_expect_tx(
+                self.mbar_group_op, txcount_op, self.id_op, predicate=predicate
+            )
+        else:
+            nvgpu.mbarrier_arrive(
+                ir.Type.parse("!nvgpu.mbarrier.token"), self.mbar_group_op, self.id_op
+            )
+
+    def try_wait(self, phase: bool = False, ticks: int = 10000000):
+        ticks_op = const(ticks)
+        phase_op = const(phase, T.bool())
+        nvgpu.MBarrierTryWaitParityOp(
+            self.mbar_group_op,
+            phase_op,
+            ticks_op,
+            mbarId=self.id_op,
+        )
+
+
+class TMA:
+    """A class that builds a TMA descriptor."""
+
+    def __init__(
+        self,
+        tma_box_shape,
+        memref_ty,
+        swizzle=nvgpu.TensorMapSwizzleKind.SWIZZLE_NONE,
+        l2promo=nvgpu.TensorMapL2PromoKind.L2PROMO_NONE,
+        oob=nvgpu.TensorMapOOBKind.OOB_ZERO,
+        interleave=nvgpu.TensorMapInterleaveKind.INTERLEAVE_NONE,
+    ):
+        self.swizzle = swizzle  # mlir.nvgpu.TensorMapSwizzleKind
+        self.l2promo = l2promo  # mlir.nvgpu.TensorMapL2PromoKind
+        self.oob = oob  # mlir.nvgpu.TensorMapOOBKind
+        self.interleave = interleave  # mlir.nvgpu.TensorMapInterleaveKind
+        self.tma_box_shape = tma_box_shape
+        self.memref_ty = memref_ty  # MemRefType
+        self.tma_memref = ir.MemRefType.get(tma_box_shape, memref_ty.element_type)
+
+    @property
+    def tensormap_descriptor_ty(self):
+        """Returns a tensormap descriptor type."""
+        tensorMemrefType = ir.MemRefType.get(
+            self.tma_box_shape,
+            self.memref_ty.element_type,
+            memory_space=ir.Attribute.parse("3"),
+        )
+        return nvgpu.TensorMapDescriptorType.get(
+            tensorMemrefType,
+            self.swizzle,
+            self.l2promo,
+            self.oob,
+            self.interleave,
+        )
+
+    def create_descriptor(self, device_ptr):
+        tma_descriptor_ty = self.tensormap_descriptor_ty
+        device_unranked_memref = memref.CastOp(
+            ir.UnrankedMemRefType.get(
+                self.memref_ty.element_type, self.memref_ty.memory_space
+            ),
+            device_ptr,
+        )
+        self.tma_descriptor = nvgpu.TmaCreateDescriptorOp(
+            tma_descriptor_ty, device_unranked_memref, map(const, self.tma_box_shape)
+        )
+        return self.tma_descriptor.result
+
+    def prefetch(self, predicate=None):
+        nvgpu.tma_prefetch_descriptor(self.tma_descriptor, predicate=predicate)
+
+    def load(self, dest, mbarrier: Mbarriers, coords=[0], predicate=None):
+        nvgpu.TmaAsyncLoadOp(
+            dest,
+            mbarrier.mbar_group_op,
+            self.tma_descriptor,
+            coordinates=map(const, coords),
+            mbarId=mbarrier.id_op,
+            predicate=predicate,
+        )
+
+
+WARP_GROUP_SIZE = 128  # Number of threads in a warpgroup
+
+
+class Warpgroup:
+    def __init__(self, primary_thread, register_size):
+        assert (primary_thread % WARP_GROUP_SIZE) == 0
+        tidx = gpu.thread_id(gpu.Dimension.x)
+        self.primary_thread = primary_thread
+        self.register_size = register_size
+        self.is_wg_primary = (tidx % WARP_GROUP_SIZE) == 0
+        self.wg_id = tidx / WARP_GROUP_SIZE
+        self.is_me = self.wg_id == (primary_thread // WARP_GROUP_SIZE)
+
+    def __enter__(self):
+        if_op = scf.IfOp(self.is_me)
+        self.ipoint_op = ir.InsertionPoint(if_op.then_block)
+        self.ipoint_op.__enter__()
+        if self.register_size < 64:
+            nvvm.setmaxregister(self.register_size, nvvm.SetMaxRegisterAction.decrease)
+        else:
+            nvvm.setmaxregister(self.register_size, nvvm.SetMaxRegisterAction.increase)
+
+    def __exit__(self, exc_type, exc_value, traceback):
+        scf.yield_([])
+        self.ipoint_op.__exit__(exc_type, exc_value, traceback)
+        return True
+
+
+class WGMMAType(Enum):
+    Accumulator = 1
+    Descriptor = 2
+
+
+class WGMMAMatrix:
+    def __init__(
+        self,
+        matrix_type: WGMMAType,
+        shape: list = None,
+        desc: TMA = None,
+        smem=None,
+        ty=None,
+        acc_op=None,
+    ):
+        if acc_op is None:
+            self.M = shape[0]
+            self.N = shape[1]
+            self.ty = ty
+            self.matrix_type = matrix_type
+            self.desc = desc
+            self.smem = smem
+            if matrix_type is WGMMAType.Accumulator:
+                self.acc_op = nvgpu.warpgroup_mma_init_accumulator(self.acc_ty)
+        elif acc_op:
+            self.acc_op = acc_op
+            self.matrix_type = WGMMAType.Accumulator
+
+    @property
+    def acc_ty(self):
+        parse_str = f"!nvgpu.warpgroup.accumulator<fragmented=vector<{self.M}x{self.N}x{self.ty}>>"
+        return ir.Type.parse(parse_str)
+
+    @property
+    def wgmma_ty(self):
+        parse_str = f"!nvgpu.warpgroup.descriptor<tensor=memref<{self.M}x{self.N}x{self.desc.memref_ty.element_type}, #gpu.address_space<workgroup>>>"
+        return ir.Type.parse(parse_str)
+
+    def store_accumulator(self, dest):
+        assert self.matrix_type == WGMMAType.Accumulator
+        nvgpu.warpgroup_mma_store(self.acc_op, dest)
+
+    def update_smem(self, smem):
+        self.smem = smem
+
+    def update_accumulator(self, acc_op):
+        self.acc_op = acc_op
+
+    def __matmul__(self, rhs):
+        lhs = nvgpu.warpgroup_generate_descriptor(
+            self.wgmma_ty, self.smem, self.desc.tma_descriptor
+        )
+        rhs = nvgpu.warpgroup_generate_descriptor(
+            rhs.wgmma_ty, rhs.smem, rhs.desc.tma_descriptor
+        )
+        return [lhs, rhs]
+
+    def __iadd__(self, matmulResult):
+        lhs = matmulResult[0]
+        rhs = matmulResult[1]
+        acc_op = nvgpu.WarpgroupMmaOp(
+            self.acc_op.type, lhs, rhs, self.acc_op, transposeB=True
+        )
+        return WGMMAMatrix(WGMMAType.Accumulator, acc_op=acc_op)
+
+
+def get_dynamic_shared_memory(shape=None, ty=None, offset: int = 0):
+    smem_space_str = "#gpu.address_space<workgroup>"
+    smem_space = ir.Attribute.parse(smem_space_str)
+    dynamic_smem = gpu.dynamic_shared_memory(
+        ir.MemRefType.get((MLIR_DYNAMIC,), T.i8(), memory_space=smem_space)
+    )
+    if shape is None:
+        return dynamic_smem
+    memref_ty = ir.MemRefType.get(shape, ty, memory_space=smem_space)
+    return memref.view(
+        ir.MemRefType.get(
+            memref_ty.shape, memref_ty.element_type, memory_space=smem_space
+        ),
+        dynamic_smem,
+        const(offset),
+        [],
+    )
+
+
+def get_mlir_ty(arg):
+    def get_mlir_ty_from_np(dtype):
+        if dtype == np.float16:
+            return T.f16()
+        if dtype == np.float32:
+            return T.f32()
+        if dtype == np.float64:
+            return T.f64()
+        if dtype == np.int32:
+            return T.i32()
+        if dtype == np.int64:
+            return T.i64()
+        raise NotImplementedError(dtype)
+
+    if isinstance(arg, bool):
+        return T.bool()
+    elif isinstance(arg, int):
+        return T.index()
+    elif isinstance(arg, float):
+        return T.f32()
+    elif isinstance(arg, np.ndarray):
+        descriptor = rt.get_ranked_memref_descriptor(arg)
+        dtype = get_mlir_ty_from_np(arg.dtype)
+        shape = descriptor.shape
+        return memref.MemRefType.get(shape, dtype)
+    raise NotImplementedError(arg)
+
+
+class NVDSL:
+    @staticmethod
+    def mlir_gpu_launch(grid=(1, 1, 1), block=(1, 1, 1), smem=0):
+        def decorator(func):
+            @functools.wraps(func)
+            def wrapper(*args, **kwargs):
+                launch_op = gpu.LaunchOp(
+                    None,
+                    [],
+                    *map(const, grid),
+                    *map(const, block),
+                    dynamicSharedMemorySize=arith.constant(T.i32(), smem),
+                )
+                launch_op.body.blocks.append(*([T.index()] * 12))
+                with ir.InsertionPoint(launch_op.body.blocks[0]):
+                    result = func(*args, **kwargs)
+                    gpu.terminator()
+                    return result
+
+            return wrapper
+
+        return decorator
+
+    @staticmethod
+    def mlir_func(funcBody):
+        @functools.wraps(funcBody)
+        def wrapper(*args, **kwargs):
+            function_name = funcBody.__name__
+
+            def saveIR(module):
+                """Save generated IR"""
+                if True:  # self.saveIR:
+                    # print(mlir_nvgpu_module)
+                    original_stdout = sys.stdout
+                    with open("nvdsl.mlir", "w") as f:
+                        sys.stdout = f
+                        print(module)
+                        sys.stdout = original_stdout
+
+            def _binary_op(lhs, rhs, op: str, predAtt="") -> "ArithValue":
+                """Generate MLIR's Arith dialects binary operations."""
+                rhs = const(rhs)
+                if arith._is_float_type(lhs.type) and arith._is_float_type(rhs.type):
+                    op += "F"
+                    if op.startswith("Cmp"):
+                        predicateAttr = getattr(arith, f"CmpFPredicate").__dict__[
+                            predAtt
+                        ]
+                elif arith._is_integer_like_type(
+                    lhs.type
+                ) and arith._is_integer_like_type(lhs.type):
+                    if op == "Div" or op == "Rem":
+                        op += "U"
+                    op += "I"
+                    if op.startswith("Cmp"):
+                        predicateAttr = getattr(arith, f"CmpIPredicate").__dict__[
+                            predAtt
+                        ]
+                else:
+                    raise NotImplementedError(
+                        f"Unsupported '{op}' operands: {lhs}, {rhs}"
+                    )
+
+                if op.startswith("Cmp"):
+                    op = getattr(arith, f"{op}Op")
+
+                    return op(predicateAttr, lhs, rhs).result
+                else:
+                    op = getattr(arith, f"{op}Op")
+                    return op(lhs, rhs).result
+
+            @ir.register_value_caster(ir.IndexType.static_typeid)
+            @ir.register_value_caster(ir.F32Type.static_typeid)
+            @ir.register_value_caster(ir.F16Type.static_typeid)
+            @ir.register_value_caster(ir.F64Type.static_typeid)
+            @ir.register_value_caster(ir.IntegerType.static_typeid)
+            class ArithValue(ir.Value):
+                """Overloads operators for MLIR's Arith dialects binary operations."""
+
+                def __init__(self, v):
+                    super().__init__(v)
+
+                __add__ = partialmethod(_binary_op, op="Add")
+                __sub__ = partialmethod(_binary_op, op="Sub")
+                __mul__ = partialmethod(_binary_op, op="Mul")
+                __truediv__ = partialmethod(_binary_op, op="Div")
+                __mod__ = partialmethod(_binary_op, op="Rem")
+                __xor__ = partialmethod(_binary_op, op="XOr")
+                __lt__ = partialmethod(_binary_op, op="Cmp", predAtt="ult")
+                __le__ = partialmethod(_binary_op, op="Cmp", predAtt="ule")
+                __eq__ = partialmethod(_binary_op, op="Cmp", predAtt="eq")
+                __ne__ = partialmethod(_binary_op, op="Cmp", predAtt="ne")
+                __gt__ = partialmethod(_binary_op, op="Cmp", predAtt="ugt")
+                __ge__ = partialmethod(_binary_op, op="Cmp", predAtt="uge")
+                __and__ = partialmethod(_binary_op, op="And")
+                __or__ = partialmethod(_binary_op, op="Or")
+
+                def __str__(self):
+                    return (
+                        super()
+                        .__str__()
+                        .replace(ir.Value.__name__, ArithValue.__name__)
+                    )
+
+            # Generate MLIR Context and start generating IR
+            with ir.Context(), ir.Location.unknown():
+                types = []
+                for arg in args:
+                    types.append(get_mlir_ty(arg))
+
+                # Build IR
+                module = ir.Module.create()
+                with ir.InsertionPoint(module.body):
+                    fop = func.FuncOp(function_name, (types, []))
+                    fop.attributes["llvm.emit_c_interface"] = ir.UnitAttr.get()
+                    with ir.InsertionPoint(fop.add_entry_block()):
+                        fargs = []
+                        for i, a in enumerate(types):
+                            fargs.append(fop.arguments[i])
+
+                        # Call user function body
+                        result = funcBody(*fargs, **kwargs)
+                        func.ReturnOp([])
+
+                # Save IR in a file
+                # saveIR(module)
+
+                # Verify the module
+                # module.operation.verify()
+
+                # Compile and JIT MLIR module
+                options = f"cubin-chip=sm_90a cubin-features=+ptx80 opt-level=3"
+                support_lib = os.getenv("SUPPORT_LIB")
+                if not os.path.exists(support_lib):
+                    raise FileNotFoundError(
+                        errno.ENOENT, os.strerror(errno.ENOENT), support_lib
+                    )
+                compiler = nvgpucompiler.NvgpuCompiler(
+                    options, opt_level=3, shared_libs=[support_lib]
+                )
+                engine = compiler.compile_and_jit(module)
+
+            # Convert input arguments to MLIR arguments
+            newArgs = get_mlir_func_obj_ty(args)
+
+            # Run the compiled program
+            engine.invoke(function_name, *newArgs)
+
+            return result
+
+        return wrapper
diff --git a/mlir/test/Examples/NVGPU/tools/nvgpucompiler.py b/mlir/test/Examples/NVGPU/tools/nvgpucompiler.py
new file mode 100644
index 000000000000..1c9cc74fcd16
--- /dev/null
+++ b/mlir/test/Examples/NVGPU/tools/nvgpucompiler.py
@@ -0,0 +1,45 @@
+#  Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+#  See https://llvm.org/LICENSE.txt for license information.
+#  SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+
+#  This file contains the Nvgpu class.
+
+from mlir import execution_engine
+from mlir import ir
+from mlir import passmanager
+from typing import Sequence
+import errno
+import os
+import sys
+
+_SCRIPT_PATH = os.path.dirname(os.path.abspath(__file__))
+sys.path.append(_SCRIPT_PATH)
+
+
+class NvgpuCompiler:
+    """Nvgpu class for compiling and building MLIR modules."""
+
+    def __init__(self, options: str, opt_level: int, shared_libs: Sequence[str]):
+        pipeline = f"builtin.module(gpu-lower-to-nvvm-pipeline{{{options}}})"
+        self.pipeline = pipeline
+        self.shared_libs = shared_libs
+        self.opt_level = opt_level
+
+    def __call__(self, module: ir.Module):
+        """Convenience application method."""
+        self.compile(module)
+
+    def compile(self, module: ir.Module):
+        """Compiles the module by invoking the nvgpu pipeline."""
+        passmanager.PassManager.parse(self.pipeline).run(module.operation)
+
+    def jit(self, module: ir.Module) -> execution_engine.ExecutionEngine:
+        """Wraps the module in a JIT execution engine."""
+        return execution_engine.ExecutionEngine(
+            module, opt_level=self.opt_level, shared_libs=self.shared_libs
+        )
+
+    def compile_and_jit(self, module: ir.Module) -> execution_engine.ExecutionEngine:
+        """Compiles and jits the module."""
+        self.compile(module)
+        return self.jit(module)
diff --git a/mlir/test/Integration/Dialect/Linalg/CPU/runtime-verification.mlir b/mlir/test/Integration/Dialect/Linalg/CPU/runtime-verification.mlir
new file mode 100644
index 000000000000..b05ef9422e59
--- /dev/null
+++ b/mlir/test/Integration/Dialect/Linalg/CPU/runtime-verification.mlir
@@ -0,0 +1,298 @@
+// RUN: mlir-opt %s -generate-runtime-verification \
+// RUN: -one-shot-bufferize="bufferize-function-boundaries" \
+// RUN: -convert-linalg-to-loops \
+// RUN: -expand-strided-metadata \
+// RUN: -lower-affine \
+// RUN: -convert-scf-to-cf \
+// RUN: -test-cf-assert \
+// RUN: -convert-index-to-llvm \
+// RUN: -finalize-memref-to-llvm \
+// RUN: -convert-func-to-llvm \
+// RUN: -reconcile-unrealized-casts | \
+// RUN: mlir-cpu-runner -e main -entry-point-result=void \
+// RUN:     -shared-libs=%mlir_runner_utils \
+// RUN:     -shared-libs=%mlir_c_runner_utils 2>&1 | \
+// RUN: FileCheck %s
+
+func.func @main() {
+  %c5x = arith.constant dense<0.0> : tensor<5xf32>
+  %c4x = arith.constant dense<0.0> : tensor<4xf32>
+  %d5x = tensor.cast %c5x : tensor<5xf32> to tensor<?xf32>
+  %d4x = tensor.cast %c4x : tensor<4xf32> to tensor<?xf32>
+
+  // CHECK-NOT: ERROR: Runtime op verification failed
+  func.call @simple_add(%d5x, %d5x) : (tensor<?xf32>, tensor<?xf32>) -> (tensor<?xf32>)
+
+  // CHECK: ERROR: Runtime op verification failed
+  // CHECK: linalg.generic
+  // CHECK: ^ dimension #0 of input/output operand #1 is incompatible with inferred dimension size
+  func.call @simple_add(%d5x, %d4x) : (tensor<?xf32>, tensor<?xf32>) -> (tensor<?xf32>)
+
+  // CHECK: ERROR: Runtime op verification failed
+  // CHECK: linalg.generic
+  // CHECK: ^ dimension #0 of input/output operand #1 is incompatible with inferred dimension size
+  func.call @simple_add(%d4x, %d5x) : (tensor<?xf32>, tensor<?xf32>) -> (tensor<?xf32>)
+
+  %c1x1 = arith.constant dense<0.0> : tensor<1x1xf32>
+  %c1x4 = arith.constant dense<0.0> : tensor<1x4xf32>
+  %c4x4 = arith.constant dense<0.0> : tensor<4x4xf32>
+  %c4x5 = arith.constant dense<0.0> : tensor<4x5xf32>
+  %c5x4 = arith.constant dense<0.0> : tensor<5x4xf32>
+  %d1x1 = tensor.cast %c1x1 : tensor<1x1xf32> to tensor<?x?xf32>
+  %d1x4 = tensor.cast %c1x4 : tensor<1x4xf32> to tensor<?x?xf32>
+  %d4x4 = tensor.cast %c4x4 : tensor<4x4xf32> to tensor<?x?xf32>
+  %d4x5 = tensor.cast %c4x5 : tensor<4x5xf32> to tensor<?x?xf32>
+  %d5x4 = tensor.cast %c5x4 : tensor<5x4xf32> to tensor<?x?xf32>
+
+  // CHECK-NOT: ERROR: Runtime op verification failed
+  func.call @broadcast_add(%d1x1, %d1x1) : (tensor<?x?xf32>, tensor<?x?xf32>) -> (tensor<?x?xf32>)
+
+  // CHECK-NOT: ERROR: Runtime op verification failed
+  func.call @broadcast_add(%d1x1, %d4x5) : (tensor<?x?xf32>, tensor<?x?xf32>) -> (tensor<?x?xf32>)
+
+  // CHECK-NOT: ERROR: Runtime op verification failed
+  func.call @broadcast_add(%d4x4, %d1x4) : (tensor<?x?xf32>, tensor<?x?xf32>) -> (tensor<?x?xf32>)
+
+  // CHECK: ERROR: Runtime op verification failed
+  // CHECK: linalg.generic
+  // CHECK: ^ dimension #1 of input/output operand #1 is incompatible with inferred dimension size
+  func.call @broadcast_add(%d1x4, %d4x5) : (tensor<?x?xf32>, tensor<?x?xf32>) -> (tensor<?x?xf32>)
+
+  // CHECK: ERROR: Runtime op verification failed
+  // CHECK: linalg.generic
+  // CHECK: ^ dimension #0 of input/output operand #1 is incompatible with inferred dimension size
+  // CHECK: ERROR: Runtime op verification failed
+  // CHECK: linalg.generic
+  // CHECK: ^ dimension #1 of input/output operand #1 is incompatible with inferred dimension size
+  // CHECK: ERROR: Runtime op verification failed
+  // CHECK: linalg.generic
+  // CHECK: ^ dimension #1 of input/output operand #2 is incompatible with inferred dimension size
+  func.call @broadcast_add(%d5x4, %d4x5) : (tensor<?x?xf32>, tensor<?x?xf32>) -> (tensor<?x?xf32>)
+
+  // CHECK-NOT: ERROR: Runtime op verification failed
+  func.call @matmul_generic(%d5x4, %d4x5) : (tensor<?x?xf32>, tensor<?x?xf32>) -> (tensor<?x?xf32>)
+
+  // CHECK: ERROR: Runtime op verification failed
+  // CHECK: linalg.generic
+  // CHECK: ^ dimension #0 of input/output operand #1 is incompatible with inferred dimension size
+  func.call @matmul_generic(%d4x5, %d4x5) : (tensor<?x?xf32>, tensor<?x?xf32>) -> (tensor<?x?xf32>)
+
+  // CHECK-NOT: ERROR: Runtime op verification failed
+  func.call @matmul_named(%d5x4, %d4x5) : (tensor<?x?xf32>, tensor<?x?xf32>) -> (tensor<?x?xf32>)
+
+  // CHECK: ERROR: Runtime op verification failed
+  // CHECK: linalg.matmul
+  // CHECK: ^ dimension #0 of input/output operand #1 is incompatible with inferred dimension size
+  func.call @matmul_named(%d4x5, %d4x5) : (tensor<?x?xf32>, tensor<?x?xf32>) -> (tensor<?x?xf32>)
+
+  %c64x57 = arith.constant dense<0.0> : tensor<16x29xf32>
+  %c3x4 = arith.constant dense<0.0> : tensor<3x4xf32>
+
+  // CHECK-NOT: ERROR: Runtime op verification failed
+  func.call @conv(%c64x57, %c3x4) : (tensor<16x29xf32>, tensor<3x4xf32>) -> (tensor<5x7xf32>)
+
+  // CHECK-NOT: ERROR: Runtime op verification failed
+  func.call @reverse_from_3(%d4x) : (tensor<?xf32>) -> (tensor<?xf32>)
+
+  // CHECK: ERROR: Runtime op verification failed
+  // CHECK: linalg.generic
+  // CHECK: unexpected negative result on dimension #0 of input/output operand #0
+  func.call @reverse_from_3(%d5x) : (tensor<?xf32>) -> (tensor<?xf32>)
+
+  return
+}
+
+
+#identity1D = affine_map<(d0) -> (d0)>
+
+func.func @simple_add(%arg0: tensor<?xf32>, %arg1: tensor<?xf32>) -> (tensor<?xf32>) {
+    %c0 = arith.constant 0 : index
+    %dim = tensor.dim %arg0, %c0 : tensor<?xf32>
+    %result = tensor.empty(%dim) : tensor<?xf32> 
+    %0 = linalg.generic {
+      indexing_maps = [#identity1D, #identity1D, #identity1D],
+      iterator_types = ["parallel"]
+    } ins(%arg0, %arg1 : tensor<?xf32>, tensor<?xf32>)
+      outs(%result : tensor<?xf32>) {
+      ^bb0(%gen_arg1: f32, %gen_arg2: f32, %out: f32) :
+        %tmp1 = arith.addf %gen_arg1, %gen_arg2 : f32
+        linalg.yield %tmp1 : f32
+    } -> tensor<?xf32>
+    return %0 : tensor<?xf32>
+}
+
+#broadcastD0 = affine_map<(d0, d1) -> (0, d1)>
+#broadcastD1 = affine_map<(d0, d1) -> (d0, 0)>
+#identity2D = affine_map<(d0, d1) -> (d0, d1)>
+
+func.func @broadcast_add(%arg0: tensor<?x?xf32>, %arg1: tensor<?x?xf32>) -> tensor<?x?xf32> {
+ // Calculate maximum dimension 0
+  %c0 = arith.constant 0 : index
+  %dim = tensor.dim %arg0, %c0 : tensor<?x?xf32>
+  %dim_0 = tensor.dim %arg1, %c0 : tensor<?x?xf32>
+  %0 = arith.maxui %dim, %dim_0 : index
+
+  // Calculate maximum dimension 1
+  %c1 = arith.constant 1 : index
+  %dim_1 = tensor.dim %arg0, %c1 : tensor<?x?xf32>
+  %dim_2 = tensor.dim %arg1, %c1 : tensor<?x?xf32>
+  %1 = arith.maxui %dim_1, %dim_2 : index
+
+  // Broadcast dimension 0 of %arg0
+  %dim_3 = tensor.dim %arg0, %c0 : tensor<?x?xf32>
+  %2 = arith.cmpi eq, %dim_3, %c1 : index
+  %3 = scf.if %2 -> (tensor<?x?xf32>) {
+    %dim_7 = tensor.dim %arg0, %c1 : tensor<?x?xf32>
+    %12 = tensor.empty(%0, %dim_7) : tensor<?x?xf32>
+    %13 = linalg.generic {
+      indexing_maps = [#broadcastD0, #identity2D],
+      iterator_types = ["parallel", "parallel"]
+    } ins(%arg0 : tensor<?x?xf32>) outs(%12 : tensor<?x?xf32>) {
+    ^bb0(%in: f32, %out: f32):
+      linalg.yield %in : f32
+    } -> tensor<?x?xf32>
+    scf.yield %13 : tensor<?x?xf32>
+  } else {
+    scf.yield %arg0 : tensor<?x?xf32>
+  }
+
+  // Broadcast dimension 1 of %arg0
+  %dim_4 = tensor.dim %3, %c1 : tensor<?x?xf32>
+  %4 = arith.cmpi eq, %dim_4, %c1 : index
+  %5 = scf.if %4 -> (tensor<?x?xf32>) {
+    %dim_7 = tensor.dim %3, %c0 : tensor<?x?xf32>
+    %12 = tensor.empty(%dim_7, %1) : tensor<?x?xf32>
+    %13 = linalg.generic {
+      indexing_maps = [#broadcastD1, #identity2D],
+      iterator_types = ["parallel", "parallel"]
+    } ins(%3 : tensor<?x?xf32>) outs(%12 : tensor<?x?xf32>) {
+    ^bb0(%in: f32, %out: f32):
+      linalg.yield %in : f32
+    } -> tensor<?x?xf32>
+    scf.yield %13 : tensor<?x?xf32>
+  } else {
+    scf.yield %3 : tensor<?x?xf32>
+  }
+
+  // Broadcast dimension 0 of %arg1
+  %dim_5 = tensor.dim %arg1, %c0 : tensor<?x?xf32>
+  %6 = arith.cmpi eq, %dim_5, %c1 : index
+  %7 = scf.if %6 -> (tensor<?x?xf32>) {
+    %dim_7 = tensor.dim %arg1, %c1 : tensor<?x?xf32>
+    %12 = tensor.empty(%0, %dim_7) : tensor<?x?xf32>
+    %13 = linalg.generic {
+      indexing_maps = [#broadcastD0, #identity2D],
+      iterator_types = ["parallel", "parallel"]
+    } ins(%arg1 : tensor<?x?xf32>) outs(%12 : tensor<?x?xf32>) {
+    ^bb0(%in: f32, %out: f32):
+      linalg.yield %in : f32
+    } -> tensor<?x?xf32>
+    scf.yield %13 : tensor<?x?xf32>
+  } else {
+    scf.yield %arg1 : tensor<?x?xf32>
+  }
+
+  // Broadcast dimension 1 of %arg1
+  %dim_6 = tensor.dim %7, %c1 : tensor<?x?xf32>
+  %8 = arith.cmpi eq, %dim_6, %c1 : index
+  %9 = scf.if %8 -> (tensor<?x?xf32>) {
+    %dim_7 = tensor.dim %7, %c0 : tensor<?x?xf32>
+    %12 = tensor.empty(%dim_7, %1) : tensor<?x?xf32>
+    %13 = linalg.generic {
+      indexing_maps = [#broadcastD1, #identity2D],
+      iterator_types = ["parallel", "parallel"]
+    } ins(%7 : tensor<?x?xf32>) outs(%12 : tensor<?x?xf32>) {
+    ^bb0(%in: f32, %out: f32):
+      linalg.yield %in : f32
+    } -> tensor<?x?xf32>
+    scf.yield %13 : tensor<?x?xf32>
+  } else {
+    scf.yield %7 : tensor<?x?xf32>
+  }
+
+  // Perform element-wise computation
+  %10 = tensor.empty(%0, %1) : tensor<?x?xf32>
+  %11 = linalg.generic {
+    indexing_maps = [#identity2D, #identity2D, #identity2D],
+    iterator_types = ["parallel", "parallel"]
+  } ins(%5, %9 : tensor<?x?xf32>, tensor<?x?xf32>) outs(%10 : tensor<?x?xf32>) {
+  ^bb0(%in: f32, %in_7: f32, %out: f32):
+    %12 = arith.addf %in, %in_7 : f32
+    linalg.yield %12 : f32
+  } -> tensor<?x?xf32>
+  return %11 : tensor<?x?xf32>
+}
+
+#matmul_accesses = [
+  affine_map<(m, n, k) -> (m, k)>,
+  affine_map<(m, n, k) -> (k, n)>,
+  affine_map<(m, n, k) -> (m, n)>
+]
+#matmul_trait = {
+  iterator_types = ["parallel", "parallel", "reduction"],
+  indexing_maps = #matmul_accesses
+}
+
+func.func @matmul_generic(%arg0: tensor<?x?xf32>, %arg1: tensor<?x?xf32>) -> tensor<?x?xf32> {
+  %cf0 = arith.constant 0.0 : f32 
+  %ci0 = arith.constant 0 : index 
+  %ci1 = arith.constant 1 : index 
+  %d0 = tensor.dim %arg0, %ci0 : tensor<?x?xf32>
+  %d1 = tensor.dim %arg1, %ci1 : tensor<?x?xf32>
+  %splat = tensor.splat %cf0[%d0, %d1] : tensor<?x?xf32>
+  %0 = linalg.generic #matmul_trait ins(%arg0, %arg1 : tensor<?x?xf32>, tensor<?x?xf32>) outs(%splat : tensor<?x?xf32>) {
+  ^bb0(%in: f32, %in_0: f32, %out: f32):
+    %1 = arith.mulf %in, %in_0 : f32
+    %2 = arith.addf %out, %1 : f32
+    linalg.yield %2 : f32
+  } -> tensor<?x?xf32>
+  return %0 : tensor<?x?xf32>
+}
+
+func.func @matmul_named(%arg0: tensor<?x?xf32>, %arg1: tensor<?x?xf32>) -> tensor<?x?xf32> {
+  %cf0 = arith.constant 0.0 : f32 
+  %ci0 = arith.constant 0 : index 
+  %ci1 = arith.constant 1 : index 
+  %d0 = tensor.dim %arg0, %ci0 : tensor<?x?xf32>
+  %d1 = tensor.dim %arg1, %ci1 : tensor<?x?xf32>
+  %splat = tensor.splat %cf0[%d0, %d1] : tensor<?x?xf32>
+  %0 = linalg.matmul  ins(%arg0, %arg1 : tensor<?x?xf32>, tensor<?x?xf32>) outs(%splat : tensor<?x?xf32>) -> tensor<?x?xf32>
+  return %0 : tensor<?x?xf32>
+}
+
+#conv_trait = {
+  indexing_maps = [affine_map<(d0, d1, d2, d3) -> (d0 * 3 + d2, d1 * 4 + d3)>, affine_map<(d0, d1, d2, d3) -> (d2, d3)>, affine_map<(d0, d1, d2, d3) -> (d0, d1)>],
+  iterator_types = ["parallel", "parallel", "reduction", "reduction"]
+}
+
+func.func @conv(%arg0: tensor<16x29xf32>, %arg1: tensor<3x4xf32>) -> (tensor<5x7xf32>) {
+  %c0 = arith.constant 0.0 : f32 
+  %splat = tensor.splat %c0 : tensor<5x7xf32>
+  %result = linalg.generic #conv_trait ins(%arg0, %arg1 : tensor<16x29xf32>, tensor<3x4xf32>) outs(%splat : tensor<5x7xf32>) {
+  ^bb0(%in: f32, %in_64: f32, %out: f32):
+    %5 = arith.mulf %in, %in_64 : f32
+    %6 = arith.addf %out, %5 : f32
+    linalg.yield %6 : f32
+  } -> tensor<5x7xf32>
+  return %result : tensor<5x7xf32>
+}
+
+#reverse_trait = {
+  indexing_maps = [
+          affine_map<(i) -> (3 - i)>,
+          affine_map<(i) -> (i)>
+  ],
+  iterator_types = ["parallel"]
+}
+
+func.func @reverse_from_3(%arg0: tensor<?xf32>) -> (tensor<?xf32>) {
+  %cf0 = arith.constant 0.0 : f32 
+  %ci0 = arith.constant 0 : index 
+  %d0 = tensor.dim %arg0, %ci0 : tensor<?xf32>
+  %splat = tensor.splat %cf0[%d0] : tensor<?xf32>
+  %result = linalg.generic #reverse_trait ins(%arg0: tensor<?xf32>) outs(%splat: tensor<?xf32>) {
+    ^bb0(%a: f32, %b: f32):
+    linalg.yield %a : f32
+  } -> tensor<?xf32>
+  return %result : tensor<?xf32>
+}
diff --git a/mlir/test/Integration/lit.local.cfg b/mlir/test/Integration/lit.local.cfg
index 86008241f17c..1dc40f986ead 100644
--- a/mlir/test/Integration/lit.local.cfg
+++ b/mlir/test/Integration/lit.local.cfg
@@ -4,15 +4,14 @@ if not config.mlir_include_integration_tests:
     config.unsupported = True
 
 
-def configure_aarch64_lli_and_mcr_cmd():
-    lli_cmd = "lli"
+def configure_aarch64_mcr_cmd():
     mcr_cmd = "mlir-cpu-runner"
 
     # NOTE: If the SVE tests are disabled and the SME tests are enabled to run
     # under emulation, the SVE specific RUN lines in the SparseTensor tests
     # will run under emulation.
     if not (config.mlir_run_arm_sve_tests or config.mlir_run_arm_sme_tests):
-        return (lli_cmd, mcr_cmd)
+        return mcr_cmd
 
     config.substitutions.append(
         (
@@ -22,20 +21,6 @@ def configure_aarch64_lli_and_mcr_cmd():
     )
 
     if config.arm_emulator_executable:
-        if config.arm_emulator_lli_executable:
-            lli_cmd = config.arm_emulator_lli_executable
-        else:
-            # Top-level lit config adds llvm_tools_dir to PATH but this is lost
-            # when running under an emulator. If the user didn't specify an lli
-            # executable, use absolute path %llvm_tools_dir/lli.
-            lli_cmd = llvm_config.use_llvm_tool(
-                "lli",
-                search_env="LLI",
-                required=True,
-                search_paths=[config.llvm_tools_dir],
-                use_installed=False,
-            )
-
         if config.arm_emulator_mlir_cpu_runner_executable:
             mcr_cmd = config.arm_emulator_mlir_cpu_runner_executable
         else:
@@ -56,35 +41,20 @@ def configure_aarch64_lli_and_mcr_cmd():
             f"{config.arm_emulator_executable} {config.arm_emulator_options}"
         )
 
-        lli_cmd = f"{emulation_cmd} {lli_cmd}"
         mcr_cmd = f"{emulation_cmd} {mcr_cmd}"
 
-    return (lli_cmd, mcr_cmd)
+    return mcr_cmd
 
 
-aarch64_lli_cmd, aarch64_mcr_cmd = configure_aarch64_lli_and_mcr_cmd()
+aarch64_mcr_cmd = configure_aarch64_mcr_cmd()
 
-# Configure the following AArch64 substitutions:
-#
-# * %lli_aarch64_cmd         - Invokes lli. For tests that _will_ run on AArch64 (ArmSVE, ArmSME).
-# * %lli_host_or_aarch64_cmd - Invokes lli. For tests that _may_ run on AArch64 (SparseTensor).
-# * %mcr_aarch64_cmd         - Invokes mlir-cpu-runner. For tests that _will_
-#                              run on AArch64. May invoke mlir-cpu-runner under
-#                              an AArch64 emulator (when
-#                              `config.arm_emulator_executable` is set).
-#
 # AArch64 tests will run under emulation if configured at build time by the
 # following CMake options:
 #
 # * ARM_EMULATOR_EXECUTABLE     - emulator to use.
 # * ARM_EMULATOR_OPTIONS        - options for emulator.
-# * ARM_EMULATOR_LLI_EXECUTABLE - AArch64 native lli to support cross-compilation.
-# * ARM_EMULATOR_UTILS_LIB_DIR  - AArch64 native utilites library to support cross-compilation.
-#
-# Functionally the two substitutions are equivalent, i.e. %lli_aarch64_cmd
-# could be used in the SparseTensor tests where necessary, but the meaning
-# conveyed by the substitution name would be a misnomer if the host target
-# is not AArch64 and MLIR_RUN_ARM_SVE_TESTS=OFF.
-config.substitutions.append(("%lli_aarch64_cmd", aarch64_lli_cmd))
-config.substitutions.append(("%lli_host_or_aarch64_cmd", aarch64_lli_cmd))
+# * ARM_EMULATOR_MLIR_CPU_RUNNER_EXECUTABLE - AArch64 native mlir-cpu-runner to
+#                                             support cross-compilation
+# * ARM_EMULATOR_UTILS_LIB_DIR - AArch64 native utilites library to support
+#                                cross-compilation.
 config.substitutions.append(("%mcr_aarch64_cmd", aarch64_mcr_cmd))
diff --git a/mlir/test/Target/Cpp/global.mlir b/mlir/test/Target/Cpp/global.mlir
new file mode 100644
index 000000000000..f0d92e862ae3
--- /dev/null
+++ b/mlir/test/Target/Cpp/global.mlir
@@ -0,0 +1,38 @@
+// RUN: mlir-translate -mlir-to-cpp %s | FileCheck %s
+// RUN: mlir-translate -mlir-to-cpp -declare-variables-at-top %s | FileCheck %s
+
+emitc.global extern @decl : i8
+// CHECK: extern int8_t decl;
+
+emitc.global @uninit : i32
+// CHECK: int32_t uninit;
+
+emitc.global @myglobal_int : i32 = 4
+// CHECK: int32_t myglobal_int = 4;
+
+emitc.global @myglobal : !emitc.array<2xf32> = dense<4.000000e+00>
+// CHECK: float myglobal[2] = {4.000000000e+00f, 4.000000000e+00f};
+
+emitc.global const @myconstant : !emitc.array<2xi16> = dense<2>
+// CHECK: const int16_t myconstant[2] = {2, 2};
+
+emitc.global extern const @extern_constant : !emitc.array<2xi16>
+// CHECK: extern const int16_t extern_constant[2];
+
+emitc.global static @static_var : f32
+// CHECK: static float static_var;
+
+emitc.global static @static_const : f32 = 3.0
+// CHECK: static float static_const = 3.000000000e+00f;
+
+emitc.global @opaque_init : !emitc.opaque<"char"> = #emitc.opaque<"CHAR_MIN">
+// CHECK: char opaque_init = CHAR_MIN;
+
+func.func @use_global(%i: index) -> f32 {
+  %0 = emitc.get_global @myglobal : !emitc.array<2xf32>
+  %1 = emitc.subscript %0[%i] : (!emitc.array<2xf32>, index) -> f32
+  return %1 : f32
+  // CHECK-LABEL: use_global
+  // CHECK-SAME: (size_t [[V1:.*]])
+  // CHECK:   return myglobal[[[V1]]];
+}
diff --git a/mlir/test/Target/LLVMIR/Import/import-failure.ll b/mlir/test/Target/LLVMIR/Import/import-failure.ll
index 3f4efab70e1c..9c24ed2b7574 100644
--- a/mlir/test/Target/LLVMIR/Import/import-failure.ll
+++ b/mlir/test/Target/LLVMIR/Import/import-failure.ll
@@ -64,12 +64,12 @@ define void @unhandled_intrinsic() gc "example" {
 declare void @llvm.dbg.value(metadata, metadata, metadata)
 
 ; CHECK:      import-failure.ll
-; CHECK-SAME: warning: dropped intrinsic: call void @llvm.dbg.value(metadata !DIArgList(i64 %{{.*}}, i64 undef), metadata !3, metadata !DIExpression(DW_OP_LLVM_arg, 0, DW_OP_LLVM_arg, 1, DW_OP_constu, 1, DW_OP_mul, DW_OP_plus, DW_OP_stack_value))
+; CHECK-SAME: warning: dropped intrinsic: tail call void @llvm.dbg.value(metadata !DIArgList(i64 %{{.*}}, i64 undef), metadata !3, metadata !DIExpression(DW_OP_LLVM_arg, 0, DW_OP_LLVM_arg, 1, DW_OP_constu, 1, DW_OP_mul, DW_OP_plus, DW_OP_stack_value))
 ; CHECK:      import-failure.ll
-; CHECK-SAME: warning: dropped intrinsic: call void @llvm.dbg.value(metadata !6, metadata !3, metadata !DIExpression())
+; CHECK-SAME: warning: dropped intrinsic: tail call void @llvm.dbg.value(metadata !6, metadata !3, metadata !DIExpression())
 define void @unsupported_argument(i64 %arg1) {
-  call void @llvm.dbg.value(metadata !DIArgList(i64 %arg1, i64 undef), metadata !3, metadata !DIExpression(DW_OP_LLVM_arg, 0, DW_OP_LLVM_arg, 1, DW_OP_constu, 1, DW_OP_mul, DW_OP_plus, DW_OP_stack_value)), !dbg !5
-  call void @llvm.dbg.value(metadata !6, metadata !3, metadata !DIExpression()), !dbg !5
+  tail call void @llvm.dbg.value(metadata !DIArgList(i64 %arg1, i64 undef), metadata !3, metadata !DIExpression(DW_OP_LLVM_arg, 0, DW_OP_LLVM_arg, 1, DW_OP_constu, 1, DW_OP_mul, DW_OP_plus, DW_OP_stack_value)), !dbg !5
+  tail call void @llvm.dbg.value(metadata !6, metadata !3, metadata !DIExpression()), !dbg !5
   ret void
 }
 
diff --git a/mlir/test/Target/LLVMIR/omptarget-parallel-wsloop.mlir b/mlir/test/Target/LLVMIR/omptarget-parallel-wsloop.mlir
index b0fe642238f1..360b3b0c0e60 100644
--- a/mlir/test/Target/LLVMIR/omptarget-parallel-wsloop.mlir
+++ b/mlir/test/Target/LLVMIR/omptarget-parallel-wsloop.mlir
@@ -12,10 +12,13 @@ module attributes {dlti.dl_spec = #dlti.dl_spec<#dlti.dl_entry<"dlti.alloca_memo
       %loop_ub = llvm.mlir.constant(9 : i32) : i32
       %loop_lb = llvm.mlir.constant(0 : i32) : i32
       %loop_step = llvm.mlir.constant(1 : i32) : i32
-      omp.wsloop for  (%loop_cnt) : i32 = (%loop_lb) to (%loop_ub) inclusive step (%loop_step) {
-        %gep = llvm.getelementptr %arg0[0, %loop_cnt] : (!llvm.ptr, i32) -> !llvm.ptr, !llvm.array<10 x i32>
-        llvm.store %loop_cnt, %gep : i32, !llvm.ptr
-        omp.yield
+      omp.wsloop {
+        omp.loop_nest (%loop_cnt) : i32 = (%loop_lb) to (%loop_ub) inclusive step (%loop_step) {
+          %gep = llvm.getelementptr %arg0[0, %loop_cnt] : (!llvm.ptr, i32) -> !llvm.ptr, !llvm.array<10 x i32>
+          llvm.store %loop_cnt, %gep : i32, !llvm.ptr
+          omp.yield
+        }
+        omp.terminator
       }
      omp.terminator
     }
diff --git a/mlir/test/Target/LLVMIR/omptarget-wsloop-collapsed.mlir b/mlir/test/Target/LLVMIR/omptarget-wsloop-collapsed.mlir
index 0d77423abcb4..13d34b7e58f7 100644
--- a/mlir/test/Target/LLVMIR/omptarget-wsloop-collapsed.mlir
+++ b/mlir/test/Target/LLVMIR/omptarget-wsloop-collapsed.mlir
@@ -8,13 +8,16 @@ module attributes {dlti.dl_spec = #dlti.dl_spec<#dlti.dl_entry<"dlti.alloca_memo
     %loop_ub = llvm.mlir.constant(99 : i32) : i32
     %loop_lb = llvm.mlir.constant(0 : i32) : i32
     %loop_step = llvm.mlir.constant(1 : index) : i32
-    omp.wsloop for  (%arg1, %arg2) : i32 = (%loop_lb, %loop_lb) to (%loop_ub, %loop_ub) inclusive step (%loop_step, %loop_step) {
-      %1 = llvm.add %arg1, %arg2  : i32
-      %2 = llvm.mul %arg2, %loop_ub overflow<nsw>  : i32
-      %3 = llvm.add %arg1, %2 :i32
-      %4 = llvm.getelementptr %arg0[%3] : (!llvm.ptr, i32) -> !llvm.ptr, i32
-      llvm.store %1, %4 : i32, !llvm.ptr
-      omp.yield
+    omp.wsloop {
+      omp.loop_nest (%arg1, %arg2) : i32 = (%loop_lb, %loop_lb) to (%loop_ub, %loop_ub) inclusive step (%loop_step, %loop_step) {
+        %1 = llvm.add %arg1, %arg2  : i32
+        %2 = llvm.mul %arg2, %loop_ub overflow<nsw>  : i32
+        %3 = llvm.add %arg1, %2 :i32
+        %4 = llvm.getelementptr %arg0[%3] : (!llvm.ptr, i32) -> !llvm.ptr, i32
+        llvm.store %1, %4 : i32, !llvm.ptr
+        omp.yield
+      }
+      omp.terminator
     }
     llvm.return
   }
diff --git a/mlir/test/Target/LLVMIR/omptarget-wsloop.mlir b/mlir/test/Target/LLVMIR/omptarget-wsloop.mlir
index 0f3f503dfa53..ee851eaf71ac 100644
--- a/mlir/test/Target/LLVMIR/omptarget-wsloop.mlir
+++ b/mlir/test/Target/LLVMIR/omptarget-wsloop.mlir
@@ -8,10 +8,13 @@ module attributes {dlti.dl_spec = #dlti.dl_spec<#dlti.dl_entry<"dlti.alloca_memo
       %loop_ub = llvm.mlir.constant(9 : i32) : i32
       %loop_lb = llvm.mlir.constant(0 : i32) : i32
       %loop_step = llvm.mlir.constant(1 : i32) : i32
-      omp.wsloop for  (%loop_cnt) : i32 = (%loop_lb) to (%loop_ub) inclusive step (%loop_step) {
-        %gep = llvm.getelementptr %arg0[0, %loop_cnt] : (!llvm.ptr, i32) -> !llvm.ptr, !llvm.array<10 x i32>
-        llvm.store %loop_cnt, %gep : i32, !llvm.ptr
-        omp.yield
+      omp.wsloop {
+        omp.loop_nest (%loop_cnt) : i32 = (%loop_lb) to (%loop_ub) inclusive step (%loop_step) {
+          %gep = llvm.getelementptr %arg0[0, %loop_cnt] : (!llvm.ptr, i32) -> !llvm.ptr, !llvm.array<10 x i32>
+          llvm.store %loop_cnt, %gep : i32, !llvm.ptr
+          omp.yield
+        }
+        omp.terminator
       }
     llvm.return
   }
@@ -20,8 +23,11 @@ module attributes {dlti.dl_spec = #dlti.dl_spec<#dlti.dl_entry<"dlti.alloca_memo
       %loop_ub = llvm.mlir.constant(9 : i32) : i32
       %loop_lb = llvm.mlir.constant(0 : i32) : i32
       %loop_step = llvm.mlir.constant(1 : i32) : i32
-      omp.wsloop for  (%loop_cnt) : i32 = (%loop_lb) to (%loop_ub) inclusive step (%loop_step) {
-        omp.yield
+      omp.wsloop {
+        omp.loop_nest (%loop_cnt) : i32 = (%loop_lb) to (%loop_ub) inclusive step (%loop_step) {
+          omp.yield
+        }
+        omp.terminator
       }
     llvm.return
   }
diff --git a/mlir/test/Target/LLVMIR/openmp-data-target-device.mlir b/mlir/test/Target/LLVMIR/openmp-data-target-device.mlir
index d41429a6de06..4ea9df369af6 100644
--- a/mlir/test/Target/LLVMIR/openmp-data-target-device.mlir
+++ b/mlir/test/Target/LLVMIR/openmp-data-target-device.mlir
@@ -31,20 +31,23 @@ module attributes { } {
           %18 = llvm.mlir.constant(1 : i64) : i64
           %19 = llvm.alloca %18 x i32 {pinned} : (i64) -> !llvm.ptr<5>
           %20 = llvm.addrspacecast %19 : !llvm.ptr<5> to !llvm.ptr
-          omp.wsloop  for  (%arg2) : i32 = (%16) to (%15) inclusive step (%16) {
-            llvm.store %arg2, %20 : i32, !llvm.ptr
-            %21 = llvm.load %20 : !llvm.ptr -> i32
-            %22 = llvm.sext %21 : i32 to i64
-            %23 = llvm.mlir.constant(1 : i64) : i64
-            %24 = llvm.mlir.constant(0 : i64) : i64
-            %25 = llvm.sub %22, %23 overflow<nsw>  : i64
-            %26 = llvm.mul %25, %23 overflow<nsw>  : i64
-            %27 = llvm.mul %26, %23 overflow<nsw>  : i64
-            %28 = llvm.add %27, %24 overflow<nsw>  : i64
-            %29 = llvm.mul %23, %17 overflow<nsw>  : i64
-            %30 = llvm.getelementptr %arg0[%28] : (!llvm.ptr, i64) -> !llvm.ptr, i32
-            llvm.store %21, %30 : i32, !llvm.ptr
-            omp.yield
+          omp.wsloop {
+            omp.loop_nest (%arg2) : i32 = (%16) to (%15) inclusive step (%16) {
+              llvm.store %arg2, %20 : i32, !llvm.ptr
+              %21 = llvm.load %20 : !llvm.ptr -> i32
+              %22 = llvm.sext %21 : i32 to i64
+              %23 = llvm.mlir.constant(1 : i64) : i64
+              %24 = llvm.mlir.constant(0 : i64) : i64
+              %25 = llvm.sub %22, %23 overflow<nsw>  : i64
+              %26 = llvm.mul %25, %23 overflow<nsw>  : i64
+              %27 = llvm.mul %26, %23 overflow<nsw>  : i64
+              %28 = llvm.add %27, %24 overflow<nsw>  : i64
+              %29 = llvm.mul %23, %17 overflow<nsw>  : i64
+              %30 = llvm.getelementptr %arg0[%28] : (!llvm.ptr, i64) -> !llvm.ptr, i32
+              llvm.store %21, %30 : i32, !llvm.ptr
+              omp.yield
+            }
+            omp.terminator
           }
           omp.terminator
         }
diff --git a/mlir/test/Target/LLVMIR/openmp-llvm.mlir b/mlir/test/Target/LLVMIR/openmp-llvm.mlir
index d1390022c1dc..ad40ca26bec9 100644
--- a/mlir/test/Target/LLVMIR/openmp-llvm.mlir
+++ b/mlir/test/Target/LLVMIR/openmp-llvm.mlir
@@ -320,18 +320,20 @@ llvm.func @wsloop_simple(%arg0: !llvm.ptr) {
   %1 = llvm.mlir.constant(10 : index) : i64
   %2 = llvm.mlir.constant(1 : index) : i64
   omp.parallel {
-    "omp.wsloop"(%1, %0, %2) ({
-    ^bb0(%arg1: i64):
-      // The form of the emitted IR is controlled by OpenMPIRBuilder and
-      // tested there. Just check that the right functions are called.
-      // CHECK: call i32 @__kmpc_global_thread_num
-      // CHECK: call void @__kmpc_for_static_init_{{.*}}(ptr @[[$loc_struct]],
-      %3 = llvm.mlir.constant(2.000000e+00 : f32) : f32
-      %4 = llvm.getelementptr %arg0[%arg1] : (!llvm.ptr, i64) -> !llvm.ptr, f32
-      llvm.store %3, %4 : f32, !llvm.ptr
-      omp.yield
+    "omp.wsloop"() ({
+      omp.loop_nest (%arg1) : i64 = (%1) to (%0) step (%2) {
+        // The form of the emitted IR is controlled by OpenMPIRBuilder and
+        // tested there. Just check that the right functions are called.
+        // CHECK: call i32 @__kmpc_global_thread_num
+        // CHECK: call void @__kmpc_for_static_init_{{.*}}(ptr @[[$loc_struct]],
+        %3 = llvm.mlir.constant(2.000000e+00 : f32) : f32
+        %4 = llvm.getelementptr %arg0[%arg1] : (!llvm.ptr, i64) -> !llvm.ptr, f32
+        llvm.store %3, %4 : f32, !llvm.ptr
+        omp.yield
+      }
+      omp.terminator
       // CHECK: call void @__kmpc_for_static_fini(ptr @[[$loc_struct]],
-    }) {operandSegmentSizes = array<i32: 1, 1, 1, 0, 0, 0, 0>} : (i64, i64, i64) -> ()
+    }) : () -> ()
     omp.terminator
   }
   llvm.return
@@ -345,13 +347,15 @@ llvm.func @wsloop_inclusive_1(%arg0: !llvm.ptr) {
   %1 = llvm.mlir.constant(10 : index) : i64
   %2 = llvm.mlir.constant(1 : index) : i64
   // CHECK: store i64 31, ptr %{{.*}}upperbound
-  "omp.wsloop"(%1, %0, %2) ({
-  ^bb0(%arg1: i64):
-    %3 = llvm.mlir.constant(2.000000e+00 : f32) : f32
-    %4 = llvm.getelementptr %arg0[%arg1] : (!llvm.ptr, i64) -> !llvm.ptr, f32
-    llvm.store %3, %4 : f32, !llvm.ptr
-    omp.yield
-  }) {operandSegmentSizes = array<i32: 1, 1, 1, 0, 0, 0, 0>} : (i64, i64, i64) -> ()
+  "omp.wsloop"() ({
+    omp.loop_nest (%arg1) : i64 = (%1) to (%0) step (%2) {
+      %3 = llvm.mlir.constant(2.000000e+00 : f32) : f32
+      %4 = llvm.getelementptr %arg0[%arg1] : (!llvm.ptr, i64) -> !llvm.ptr, f32
+      llvm.store %3, %4 : f32, !llvm.ptr
+      omp.yield
+    }
+    omp.terminator
+  }) : () -> ()
   llvm.return
 }
 
@@ -363,13 +367,15 @@ llvm.func @wsloop_inclusive_2(%arg0: !llvm.ptr) {
   %1 = llvm.mlir.constant(10 : index) : i64
   %2 = llvm.mlir.constant(1 : index) : i64
   // CHECK: store i64 32, ptr %{{.*}}upperbound
-  "omp.wsloop"(%1, %0, %2) ({
-  ^bb0(%arg1: i64):
-    %3 = llvm.mlir.constant(2.000000e+00 : f32) : f32
-    %4 = llvm.getelementptr %arg0[%arg1] : (!llvm.ptr, i64) -> !llvm.ptr, f32
-    llvm.store %3, %4 : f32, !llvm.ptr
-    omp.yield
-  }) {inclusive, operandSegmentSizes = array<i32: 1, 1, 1, 0, 0, 0, 0>} : (i64, i64, i64) -> ()
+  "omp.wsloop"() ({
+    omp.loop_nest (%arg1) : i64 = (%1) to (%0) inclusive step (%2) {
+      %3 = llvm.mlir.constant(2.000000e+00 : f32) : f32
+      %4 = llvm.getelementptr %arg0[%arg1] : (!llvm.ptr, i64) -> !llvm.ptr, f32
+      llvm.store %3, %4 : f32, !llvm.ptr
+      omp.yield
+    }
+    omp.terminator
+  }) : () -> ()
   llvm.return
 }
 
@@ -379,14 +385,16 @@ llvm.func @body(i32)
 
 // CHECK-LABEL: @test_omp_wsloop_static_defchunk
 llvm.func @test_omp_wsloop_static_defchunk(%lb : i32, %ub : i32, %step : i32) -> () {
- omp.wsloop schedule(static)
- for (%iv) : i32 = (%lb) to (%ub) step (%step) {
-   // CHECK: call void @__kmpc_for_static_init_4u(ptr @{{.*}}, i32 %{{.*}}, i32 34, ptr %{{.*}}, ptr %{{.*}}, ptr %{{.*}}, ptr %{{.*}}, i32 1, i32 0)
-   // CHECK: call void @__kmpc_for_static_fini
-   llvm.call @body(%iv) : (i32) -> ()
-   omp.yield
- }
- llvm.return
+  omp.wsloop schedule(static) {
+    omp.loop_nest (%iv) : i32 = (%lb) to (%ub) step (%step) {
+      // CHECK: call void @__kmpc_for_static_init_4u(ptr @{{.*}}, i32 %{{.*}}, i32 34, ptr %{{.*}}, ptr %{{.*}}, ptr %{{.*}}, ptr %{{.*}}, i32 1, i32 0)
+      // CHECK: call void @__kmpc_for_static_fini
+      llvm.call @body(%iv) : (i32) -> ()
+      omp.yield
+    }
+    omp.terminator
+  }
+  llvm.return
 }
 
 // -----
@@ -395,15 +403,17 @@ llvm.func @body(i32)
 
 // CHECK-LABEL: @test_omp_wsloop_static_1
 llvm.func @test_omp_wsloop_static_1(%lb : i32, %ub : i32, %step : i32) -> () {
- %static_chunk_size = llvm.mlir.constant(1 : i32) : i32
- omp.wsloop schedule(static = %static_chunk_size : i32)
- for (%iv) : i32 = (%lb) to (%ub) step (%step) {
-   // CHECK: call void @__kmpc_for_static_init_4u(ptr @{{.*}}, i32 %{{.*}}, i32 33, ptr %{{.*}}, ptr %{{.*}}, ptr %{{.*}}, ptr %{{.*}}, i32 1, i32 1)
-   // CHECK: call void @__kmpc_for_static_fini
-   llvm.call @body(%iv) : (i32) -> ()
-   omp.yield
- }
- llvm.return
+  %static_chunk_size = llvm.mlir.constant(1 : i32) : i32
+  omp.wsloop schedule(static = %static_chunk_size : i32) {
+    omp.loop_nest (%iv) : i32 = (%lb) to (%ub) step (%step) {
+      // CHECK: call void @__kmpc_for_static_init_4u(ptr @{{.*}}, i32 %{{.*}}, i32 33, ptr %{{.*}}, ptr %{{.*}}, ptr %{{.*}}, ptr %{{.*}}, i32 1, i32 1)
+      // CHECK: call void @__kmpc_for_static_fini
+      llvm.call @body(%iv) : (i32) -> ()
+      omp.yield
+    }
+    omp.terminator
+  }
+  llvm.return
 }
 
 // -----
@@ -412,15 +422,17 @@ llvm.func @body(i32)
 
 // CHECK-LABEL: @test_omp_wsloop_static_2
 llvm.func @test_omp_wsloop_static_2(%lb : i32, %ub : i32, %step : i32) -> () {
- %static_chunk_size = llvm.mlir.constant(2 : i32) : i32
- omp.wsloop schedule(static = %static_chunk_size : i32)
- for (%iv) : i32 = (%lb) to (%ub) step (%step) {
-   // CHECK: call void @__kmpc_for_static_init_4u(ptr @{{.*}}, i32 %{{.*}}, i32 33, ptr %{{.*}}, ptr %{{.*}}, ptr %{{.*}}, ptr %{{.*}}, i32 1, i32 2)
-   // CHECK: call void @__kmpc_for_static_fini
-   llvm.call @body(%iv) : (i32) -> ()
-   omp.yield
- }
- llvm.return
+  %static_chunk_size = llvm.mlir.constant(2 : i32) : i32
+  omp.wsloop schedule(static = %static_chunk_size : i32) {
+    omp.loop_nest (%iv) : i32 = (%lb) to (%ub) step (%step) {
+      // CHECK: call void @__kmpc_for_static_init_4u(ptr @{{.*}}, i32 %{{.*}}, i32 33, ptr %{{.*}}, ptr %{{.*}}, ptr %{{.*}}, ptr %{{.*}}, i32 1, i32 2)
+      // CHECK: call void @__kmpc_for_static_fini
+      llvm.call @body(%iv) : (i32) -> ()
+      omp.yield
+    }
+    omp.terminator
+  }
+  llvm.return
 }
 
 // -----
@@ -428,16 +440,18 @@ llvm.func @test_omp_wsloop_static_2(%lb : i32, %ub : i32, %step : i32) -> () {
 llvm.func @body(i64)
 
 llvm.func @test_omp_wsloop_dynamic(%lb : i64, %ub : i64, %step : i64) -> () {
- omp.wsloop schedule(dynamic)
- for (%iv) : i64 = (%lb) to (%ub) step (%step)  {
-  // CHECK: call void @__kmpc_dispatch_init_8u
-  // CHECK: %[[continue:.*]] = call i32 @__kmpc_dispatch_next_8u
-  // CHECK: %[[cond:.*]] = icmp ne i32 %[[continue]], 0
-  // CHECK: br i1 %[[cond]], label %omp_loop.header{{.*}}, label %omp_loop.exit{{.*}}
-   llvm.call @body(%iv) : (i64) -> ()
-   omp.yield
- }
- llvm.return
+  omp.wsloop schedule(dynamic) {
+    omp.loop_nest (%iv) : i64 = (%lb) to (%ub) step (%step)  {
+      // CHECK: call void @__kmpc_dispatch_init_8u
+      // CHECK: %[[continue:.*]] = call i32 @__kmpc_dispatch_next_8u
+      // CHECK: %[[cond:.*]] = icmp ne i32 %[[continue]], 0
+      // CHECK: br i1 %[[cond]], label %omp_loop.header{{.*}}, label %omp_loop.exit{{.*}}
+      llvm.call @body(%iv) : (i64) -> ()
+      omp.yield
+    }
+    omp.terminator
+  }
+  llvm.return
 }
 
 // -----
@@ -445,17 +459,19 @@ llvm.func @test_omp_wsloop_dynamic(%lb : i64, %ub : i64, %step : i64) -> () {
 llvm.func @body(i64)
 
 llvm.func @test_omp_wsloop_dynamic_chunk_const(%lb : i64, %ub : i64, %step : i64) -> () {
- %chunk_size_const = llvm.mlir.constant(2 : i16) : i16
- omp.wsloop schedule(dynamic = %chunk_size_const : i16)
- for (%iv) : i64 = (%lb) to (%ub) step (%step)  {
-  // CHECK: call void @__kmpc_dispatch_init_8u(ptr @{{.*}}, i32 %{{.*}}, i32 1073741859, i64 {{.*}}, i64 %{{.*}}, i64 {{.*}}, i64 2)
-  // CHECK: %[[continue:.*]] = call i32 @__kmpc_dispatch_next_8u
-  // CHECK: %[[cond:.*]] = icmp ne i32 %[[continue]], 0
-  // CHECK: br i1 %[[cond]], label %omp_loop.header{{.*}}, label %omp_loop.exit{{.*}}
-   llvm.call @body(%iv) : (i64) -> ()
-   omp.yield
- }
- llvm.return
+  %chunk_size_const = llvm.mlir.constant(2 : i16) : i16
+  omp.wsloop schedule(dynamic = %chunk_size_const : i16) {
+    omp.loop_nest (%iv) : i64 = (%lb) to (%ub) step (%step)  {
+      // CHECK: call void @__kmpc_dispatch_init_8u(ptr @{{.*}}, i32 %{{.*}}, i32 1073741859, i64 {{.*}}, i64 %{{.*}}, i64 {{.*}}, i64 2)
+      // CHECK: %[[continue:.*]] = call i32 @__kmpc_dispatch_next_8u
+      // CHECK: %[[cond:.*]] = icmp ne i32 %[[continue]], 0
+      // CHECK: br i1 %[[cond]], label %omp_loop.header{{.*}}, label %omp_loop.exit{{.*}}
+      llvm.call @body(%iv) : (i64) -> ()
+      omp.yield
+    }
+    omp.terminator
+  }
+  llvm.return
 }
 
 // -----
@@ -463,20 +479,22 @@ llvm.func @test_omp_wsloop_dynamic_chunk_const(%lb : i64, %ub : i64, %step : i64
 llvm.func @body(i32)
 
 llvm.func @test_omp_wsloop_dynamic_chunk_var(%lb : i32, %ub : i32, %step : i32) -> () {
- %1 = llvm.mlir.constant(1 : i64) : i64
- %chunk_size_alloca = llvm.alloca %1 x i16 {bindc_name = "chunk_size", in_type = i16, uniq_name = "_QFsub1Echunk_size"} : (i64) -> !llvm.ptr
- %chunk_size_var = llvm.load %chunk_size_alloca : !llvm.ptr -> i16
- omp.wsloop schedule(dynamic = %chunk_size_var : i16)
- for (%iv) : i32 = (%lb) to (%ub) step (%step) {
-  // CHECK: %[[CHUNK_SIZE:.*]] = sext i16 %{{.*}} to i32
-  // CHECK: call void @__kmpc_dispatch_init_4u(ptr @{{.*}}, i32 %{{.*}}, i32 1073741859, i32 {{.*}}, i32 %{{.*}}, i32 {{.*}}, i32 %[[CHUNK_SIZE]])
-  // CHECK: %[[continue:.*]] = call i32 @__kmpc_dispatch_next_4u
-  // CHECK: %[[cond:.*]] = icmp ne i32 %[[continue]], 0
-  // CHECK: br i1 %[[cond]], label %omp_loop.header{{.*}}, label %omp_loop.exit{{.*}}
-   llvm.call @body(%iv) : (i32) -> ()
-   omp.yield
- }
- llvm.return
+  %1 = llvm.mlir.constant(1 : i64) : i64
+  %chunk_size_alloca = llvm.alloca %1 x i16 {bindc_name = "chunk_size", in_type = i16, uniq_name = "_QFsub1Echunk_size"} : (i64) -> !llvm.ptr
+  %chunk_size_var = llvm.load %chunk_size_alloca : !llvm.ptr -> i16
+  omp.wsloop schedule(dynamic = %chunk_size_var : i16) {
+    omp.loop_nest (%iv) : i32 = (%lb) to (%ub) step (%step) {
+      // CHECK: %[[CHUNK_SIZE:.*]] = sext i16 %{{.*}} to i32
+      // CHECK: call void @__kmpc_dispatch_init_4u(ptr @{{.*}}, i32 %{{.*}}, i32 1073741859, i32 {{.*}}, i32 %{{.*}}, i32 {{.*}}, i32 %[[CHUNK_SIZE]])
+      // CHECK: %[[continue:.*]] = call i32 @__kmpc_dispatch_next_4u
+      // CHECK: %[[cond:.*]] = icmp ne i32 %[[continue]], 0
+      // CHECK: br i1 %[[cond]], label %omp_loop.header{{.*}}, label %omp_loop.exit{{.*}}
+      llvm.call @body(%iv) : (i32) -> ()
+      omp.yield
+    }
+    omp.terminator
+  }
+  llvm.return
 }
 
 // -----
@@ -484,20 +502,22 @@ llvm.func @test_omp_wsloop_dynamic_chunk_var(%lb : i32, %ub : i32, %step : i32)
 llvm.func @body(i32)
 
 llvm.func @test_omp_wsloop_dynamic_chunk_var2(%lb : i32, %ub : i32, %step : i32) -> () {
- %1 = llvm.mlir.constant(1 : i64) : i64
- %chunk_size_alloca = llvm.alloca %1 x i64 {bindc_name = "chunk_size", in_type = i64, uniq_name = "_QFsub1Echunk_size"} : (i64) -> !llvm.ptr
- %chunk_size_var = llvm.load %chunk_size_alloca : !llvm.ptr -> i64
- omp.wsloop schedule(dynamic = %chunk_size_var : i64)
- for (%iv) : i32 = (%lb) to (%ub) step (%step) {
-  // CHECK: %[[CHUNK_SIZE:.*]] = trunc i64 %{{.*}} to i32
-  // CHECK: call void @__kmpc_dispatch_init_4u(ptr @{{.*}}, i32 %{{.*}}, i32 1073741859, i32 {{.*}}, i32 %{{.*}}, i32 {{.*}}, i32 %[[CHUNK_SIZE]])
-  // CHECK: %[[continue:.*]] = call i32 @__kmpc_dispatch_next_4u
-  // CHECK: %[[cond:.*]] = icmp ne i32 %[[continue]], 0
-  // CHECK: br i1 %[[cond]], label %omp_loop.header{{.*}}, label %omp_loop.exit{{.*}}
-   llvm.call @body(%iv) : (i32) -> ()
-   omp.yield
- }
- llvm.return
+  %1 = llvm.mlir.constant(1 : i64) : i64
+  %chunk_size_alloca = llvm.alloca %1 x i64 {bindc_name = "chunk_size", in_type = i64, uniq_name = "_QFsub1Echunk_size"} : (i64) -> !llvm.ptr
+  %chunk_size_var = llvm.load %chunk_size_alloca : !llvm.ptr -> i64
+  omp.wsloop schedule(dynamic = %chunk_size_var : i64) {
+    omp.loop_nest (%iv) : i32 = (%lb) to (%ub) step (%step) {
+      // CHECK: %[[CHUNK_SIZE:.*]] = trunc i64 %{{.*}} to i32
+      // CHECK: call void @__kmpc_dispatch_init_4u(ptr @{{.*}}, i32 %{{.*}}, i32 1073741859, i32 {{.*}}, i32 %{{.*}}, i32 {{.*}}, i32 %[[CHUNK_SIZE]])
+      // CHECK: %[[continue:.*]] = call i32 @__kmpc_dispatch_next_4u
+      // CHECK: %[[cond:.*]] = icmp ne i32 %[[continue]], 0
+      // CHECK: br i1 %[[cond]], label %omp_loop.header{{.*}}, label %omp_loop.exit{{.*}}
+      llvm.call @body(%iv) : (i32) -> ()
+      omp.yield
+    }
+    omp.terminator
+  }
+  llvm.return
 }
 
 // -----
@@ -505,16 +525,18 @@ llvm.func @test_omp_wsloop_dynamic_chunk_var2(%lb : i32, %ub : i32, %step : i32)
 llvm.func @body(i32)
 
 llvm.func @test_omp_wsloop_dynamic_chunk_var3(%lb : i32, %ub : i32, %step : i32, %chunk_size : i32) -> () {
- omp.wsloop schedule(dynamic = %chunk_size : i32)
- for (%iv) : i32 = (%lb) to (%ub) step (%step) {
-  // CHECK: call void @__kmpc_dispatch_init_4u(ptr @{{.*}}, i32 %{{.*}}, i32 1073741859, i32 {{.*}}, i32 %{{.*}}, i32 {{.*}}, i32 %{{.*}})
-  // CHECK: %[[continue:.*]] = call i32 @__kmpc_dispatch_next_4u
-  // CHECK: %[[cond:.*]] = icmp ne i32 %[[continue]], 0
-  // CHECK: br i1 %[[cond]], label %omp_loop.header{{.*}}, label %omp_loop.exit{{.*}}
-   llvm.call @body(%iv) : (i32) -> ()
-   omp.yield
- }
- llvm.return
+  omp.wsloop schedule(dynamic = %chunk_size : i32) {
+    omp.loop_nest (%iv) : i32 = (%lb) to (%ub) step (%step) {
+      // CHECK: call void @__kmpc_dispatch_init_4u(ptr @{{.*}}, i32 %{{.*}}, i32 1073741859, i32 {{.*}}, i32 %{{.*}}, i32 {{.*}}, i32 %{{.*}})
+      // CHECK: %[[continue:.*]] = call i32 @__kmpc_dispatch_next_4u
+      // CHECK: %[[cond:.*]] = icmp ne i32 %[[continue]], 0
+      // CHECK: br i1 %[[cond]], label %omp_loop.header{{.*}}, label %omp_loop.exit{{.*}}
+      llvm.call @body(%iv) : (i32) -> ()
+      omp.yield
+    }
+    omp.terminator
+  }
+  llvm.return
 }
 
 // -----
@@ -522,16 +544,18 @@ llvm.func @test_omp_wsloop_dynamic_chunk_var3(%lb : i32, %ub : i32, %step : i32,
 llvm.func @body(i64)
 
 llvm.func @test_omp_wsloop_auto(%lb : i64, %ub : i64, %step : i64) -> () {
- omp.wsloop schedule(auto)
- for (%iv) : i64 = (%lb) to (%ub) step (%step) {
-  // CHECK: call void @__kmpc_dispatch_init_8u
-  // CHECK: %[[continue:.*]] = call i32 @__kmpc_dispatch_next_8u
-  // CHECK: %[[cond:.*]] = icmp ne i32 %[[continue]], 0
-  // CHECK  br i1 %[[cond]], label %omp_loop.header{{.*}}, label %omp_loop.exit{{.*}}
-   llvm.call @body(%iv) : (i64) -> ()
-   omp.yield
- }
- llvm.return
+  omp.wsloop schedule(auto) {
+    omp.loop_nest (%iv) : i64 = (%lb) to (%ub) step (%step) {
+      // CHECK: call void @__kmpc_dispatch_init_8u
+      // CHECK: %[[continue:.*]] = call i32 @__kmpc_dispatch_next_8u
+      // CHECK: %[[cond:.*]] = icmp ne i32 %[[continue]], 0
+      // CHECK  br i1 %[[cond]], label %omp_loop.header{{.*}}, label %omp_loop.exit{{.*}}
+      llvm.call @body(%iv) : (i64) -> ()
+      omp.yield
+    }
+    omp.terminator
+  }
+  llvm.return
 }
 
 // -----
@@ -539,14 +563,16 @@ llvm.func @test_omp_wsloop_auto(%lb : i64, %ub : i64, %step : i64) -> () {
 llvm.func @body(i64)
 
 llvm.func @test_omp_wsloop_runtime(%lb : i64, %ub : i64, %step : i64) -> () {
-  omp.wsloop schedule(runtime)
-  for (%iv) : i64 = (%lb) to (%ub) step (%step) {
-    // CHECK: call void @__kmpc_dispatch_init_8u
-    // CHECK: %[[continue:.*]] = call i32 @__kmpc_dispatch_next_8u
-    // CHECK: %[[cond:.*]] = icmp ne i32 %[[continue]], 0
-    // CHECK  br i1 %[[cond]], label %omp_loop.header{{.*}}, label %omp_loop.exit{{.*}}
-    llvm.call @body(%iv) : (i64) -> ()
-    omp.yield
+  omp.wsloop schedule(runtime) {
+    omp.loop_nest (%iv) : i64 = (%lb) to (%ub) step (%step) {
+      // CHECK: call void @__kmpc_dispatch_init_8u
+      // CHECK: %[[continue:.*]] = call i32 @__kmpc_dispatch_next_8u
+      // CHECK: %[[cond:.*]] = icmp ne i32 %[[continue]], 0
+      // CHECK  br i1 %[[cond]], label %omp_loop.header{{.*}}, label %omp_loop.exit{{.*}}
+      llvm.call @body(%iv) : (i64) -> ()
+      omp.yield
+    }
+    omp.terminator
   }
   llvm.return
 }
@@ -556,14 +582,16 @@ llvm.func @test_omp_wsloop_runtime(%lb : i64, %ub : i64, %step : i64) -> () {
 llvm.func @body(i64)
 
 llvm.func @test_omp_wsloop_guided(%lb : i64, %ub : i64, %step : i64) -> () {
-  omp.wsloop schedule(guided)
-  for (%iv) : i64 = (%lb) to (%ub) step (%step) {
-    // CHECK: call void @__kmpc_dispatch_init_8u
-    // CHECK: %[[continue:.*]] = call i32 @__kmpc_dispatch_next_8u
-    // CHECK: %[[cond:.*]] = icmp ne i32 %[[continue]], 0
-    // CHECK  br i1 %[[cond]], label %omp_loop.header{{.*}}, label %omp_loop.exit{{.*}}
-    llvm.call @body(%iv) : (i64) -> ()
-    omp.yield
+  omp.wsloop schedule(guided) {
+    omp.loop_nest (%iv) : i64 = (%lb) to (%ub) step (%step) {
+      // CHECK: call void @__kmpc_dispatch_init_8u
+      // CHECK: %[[continue:.*]] = call i32 @__kmpc_dispatch_next_8u
+      // CHECK: %[[cond:.*]] = icmp ne i32 %[[continue]], 0
+      // CHECK  br i1 %[[cond]], label %omp_loop.header{{.*}}, label %omp_loop.exit{{.*}}
+      llvm.call @body(%iv) : (i64) -> ()
+      omp.yield
+    }
+    omp.terminator
   }
   llvm.return
 }
@@ -573,14 +601,16 @@ llvm.func @test_omp_wsloop_guided(%lb : i64, %ub : i64, %step : i64) -> () {
 llvm.func @body(i64)
 
 llvm.func @test_omp_wsloop_dynamic_nonmonotonic(%lb : i64, %ub : i64, %step : i64) -> () {
-  omp.wsloop schedule(dynamic, nonmonotonic)
-  for (%iv) : i64 = (%lb) to (%ub) step (%step) {
-    // CHECK: call void @__kmpc_dispatch_init_8u(ptr @{{.*}}, i32 %{{.*}}, i32 1073741859
-    // CHECK: %[[continue:.*]] = call i32 @__kmpc_dispatch_next_8u
-    // CHECK: %[[cond:.*]] = icmp ne i32 %[[continue]], 0
-    // CHECK  br i1 %[[cond]], label %omp_loop.header{{.*}}, label %omp_loop.exit{{.*}}
-    llvm.call @body(%iv) : (i64) -> ()
-    omp.yield
+  omp.wsloop schedule(dynamic, nonmonotonic) {
+    omp.loop_nest (%iv) : i64 = (%lb) to (%ub) step (%step) {
+      // CHECK: call void @__kmpc_dispatch_init_8u(ptr @{{.*}}, i32 %{{.*}}, i32 1073741859
+      // CHECK: %[[continue:.*]] = call i32 @__kmpc_dispatch_next_8u
+      // CHECK: %[[cond:.*]] = icmp ne i32 %[[continue]], 0
+      // CHECK  br i1 %[[cond]], label %omp_loop.header{{.*}}, label %omp_loop.exit{{.*}}
+      llvm.call @body(%iv) : (i64) -> ()
+      omp.yield
+    }
+    omp.terminator
   }
   llvm.return
 }
@@ -590,14 +620,16 @@ llvm.func @test_omp_wsloop_dynamic_nonmonotonic(%lb : i64, %ub : i64, %step : i6
 llvm.func @body(i64)
 
 llvm.func @test_omp_wsloop_dynamic_monotonic(%lb : i64, %ub : i64, %step : i64) -> () {
-  omp.wsloop schedule(dynamic, monotonic)
-  for (%iv) : i64 = (%lb) to (%ub) step (%step) {
-    // CHECK: call void @__kmpc_dispatch_init_8u(ptr @{{.*}}, i32 %{{.*}}, i32 536870947
-    // CHECK: %[[continue:.*]] = call i32 @__kmpc_dispatch_next_8u
-    // CHECK: %[[cond:.*]] = icmp ne i32 %[[continue]], 0
-    // CHECK  br i1 %[[cond]], label %omp_loop.header{{.*}}, label %omp_loop.exit{{.*}}
-    llvm.call @body(%iv) : (i64) -> ()
-    omp.yield
+  omp.wsloop schedule(dynamic, monotonic) {
+    omp.loop_nest (%iv) : i64 = (%lb) to (%ub) step (%step) {
+      // CHECK: call void @__kmpc_dispatch_init_8u(ptr @{{.*}}, i32 %{{.*}}, i32 536870947
+      // CHECK: %[[continue:.*]] = call i32 @__kmpc_dispatch_next_8u
+      // CHECK: %[[cond:.*]] = icmp ne i32 %[[continue]], 0
+      // CHECK  br i1 %[[cond]], label %omp_loop.header{{.*}}, label %omp_loop.exit{{.*}}
+      llvm.call @body(%iv) : (i64) -> ()
+      omp.yield
+    }
+    omp.terminator
   }
   llvm.return
 }
@@ -607,14 +639,16 @@ llvm.func @test_omp_wsloop_dynamic_monotonic(%lb : i64, %ub : i64, %step : i64)
 llvm.func @body(i64)
 
 llvm.func @test_omp_wsloop_runtime_simd(%lb : i64, %ub : i64, %step : i64) -> () {
-  omp.wsloop schedule(runtime, simd)
-  for (%iv) : i64 = (%lb) to (%ub) step (%step) {
-    // CHECK: call void @__kmpc_dispatch_init_8u(ptr @{{.*}}, i32 %{{.*}}, i32 1073741871
-    // CHECK: %[[continue:.*]] = call i32 @__kmpc_dispatch_next_8u
-    // CHECK: %[[cond:.*]] = icmp ne i32 %[[continue]], 0
-    // CHECK  br i1 %[[cond]], label %omp_loop.header{{.*}}, label %omp_loop.exit{{.*}}
-    llvm.call @body(%iv) : (i64) -> ()
-    omp.yield
+  omp.wsloop schedule(runtime, simd) {
+    omp.loop_nest (%iv) : i64 = (%lb) to (%ub) step (%step) {
+      // CHECK: call void @__kmpc_dispatch_init_8u(ptr @{{.*}}, i32 %{{.*}}, i32 1073741871
+      // CHECK: %[[continue:.*]] = call i32 @__kmpc_dispatch_next_8u
+      // CHECK: %[[cond:.*]] = icmp ne i32 %[[continue]], 0
+      // CHECK  br i1 %[[cond]], label %omp_loop.header{{.*}}, label %omp_loop.exit{{.*}}
+      llvm.call @body(%iv) : (i64) -> ()
+      omp.yield
+    }
+    omp.terminator
   }
   llvm.return
 }
@@ -624,14 +658,16 @@ llvm.func @test_omp_wsloop_runtime_simd(%lb : i64, %ub : i64, %step : i64) -> ()
 llvm.func @body(i64)
 
 llvm.func @test_omp_wsloop_guided_simd(%lb : i64, %ub : i64, %step : i64) -> () {
-  omp.wsloop schedule(guided, simd)
-  for (%iv) : i64 = (%lb) to (%ub) step (%step) {
-    // CHECK: call void @__kmpc_dispatch_init_8u(ptr @{{.*}}, i32 %{{.*}}, i32 1073741870
-    // CHECK: %[[continue:.*]] = call i32 @__kmpc_dispatch_next_8u
-    // CHECK: %[[cond:.*]] = icmp ne i32 %[[continue]], 0
-    // CHECK  br i1 %[[cond]], label %omp_loop.header{{.*}}, label %omp_loop.exit{{.*}}
-    llvm.call @body(%iv) : (i64) -> ()
-    omp.yield
+  omp.wsloop schedule(guided, simd) {
+    omp.loop_nest (%iv) : i64 = (%lb) to (%ub) step (%step) {
+      // CHECK: call void @__kmpc_dispatch_init_8u(ptr @{{.*}}, i32 %{{.*}}, i32 1073741870
+      // CHECK: %[[continue:.*]] = call i32 @__kmpc_dispatch_next_8u
+      // CHECK: %[[cond:.*]] = icmp ne i32 %[[continue]], 0
+      // CHECK  br i1 %[[cond]], label %omp_loop.header{{.*}}, label %omp_loop.exit{{.*}}
+      llvm.call @body(%iv) : (i64) -> ()
+      omp.yield
+    }
+    omp.terminator
   }
   llvm.return
 }
@@ -793,17 +829,19 @@ llvm.func @simd_if(%arg0: !llvm.ptr {fir.bindc_name = "n"}, %arg1: !llvm.ptr {fi
 llvm.func @body(i64)
 
 llvm.func @test_omp_wsloop_ordered(%lb : i64, %ub : i64, %step : i64) -> () {
- omp.wsloop ordered(0)
- for (%iv) : i64 = (%lb) to (%ub) step (%step) {
-  // CHECK: call void @__kmpc_dispatch_init_8u(ptr @{{.*}}, i32 %{{.*}}, i32 66, i64 1, i64 %{{.*}}, i64 1, i64 1)
-  // CHECK: call void @__kmpc_dispatch_fini_8u
-  // CHECK: %[[continue:.*]] = call i32 @__kmpc_dispatch_next_8u
-  // CHECK: %[[cond:.*]] = icmp ne i32 %[[continue]], 0
-  // CHECK  br i1 %[[cond]], label %omp_loop.header{{.*}}, label %omp_loop.exit{{.*}}
-   llvm.call @body(%iv) : (i64) -> ()
-   omp.yield
- }
- llvm.return
+  omp.wsloop ordered(0) {
+    omp.loop_nest (%iv) : i64 = (%lb) to (%ub) step (%step) {
+      // CHECK: call void @__kmpc_dispatch_init_8u(ptr @{{.*}}, i32 %{{.*}}, i32 66, i64 1, i64 %{{.*}}, i64 1, i64 1)
+      // CHECK: call void @__kmpc_dispatch_fini_8u
+      // CHECK: %[[continue:.*]] = call i32 @__kmpc_dispatch_next_8u
+      // CHECK: %[[cond:.*]] = icmp ne i32 %[[continue]], 0
+      // CHECK  br i1 %[[cond]], label %omp_loop.header{{.*}}, label %omp_loop.exit{{.*}}
+      llvm.call @body(%iv) : (i64) -> ()
+      omp.yield
+    }
+    omp.terminator
+  }
+  llvm.return
 }
 
 // -----
@@ -811,17 +849,19 @@ llvm.func @test_omp_wsloop_ordered(%lb : i64, %ub : i64, %step : i64) -> () {
 llvm.func @body(i64)
 
 llvm.func @test_omp_wsloop_static_ordered(%lb : i64, %ub : i64, %step : i64) -> () {
- omp.wsloop schedule(static) ordered(0)
- for (%iv) : i64 = (%lb) to (%ub) step (%step) {
-  // CHECK: call void @__kmpc_dispatch_init_8u(ptr @{{.*}}, i32 %{{.*}}, i32 66, i64 1, i64 %{{.*}}, i64 1, i64 1)
-  // CHECK: call void @__kmpc_dispatch_fini_8u
-  // CHECK: %[[continue:.*]] = call i32 @__kmpc_dispatch_next_8u
-  // CHECK: %[[cond:.*]] = icmp ne i32 %[[continue]], 0
-  // CHECK  br i1 %[[cond]], label %omp_loop.header{{.*}}, label %omp_loop.exit{{.*}}
-   llvm.call @body(%iv) : (i64) -> ()
-   omp.yield
- }
- llvm.return
+  omp.wsloop schedule(static) ordered(0) {
+    omp.loop_nest (%iv) : i64 = (%lb) to (%ub) step (%step) {
+      // CHECK: call void @__kmpc_dispatch_init_8u(ptr @{{.*}}, i32 %{{.*}}, i32 66, i64 1, i64 %{{.*}}, i64 1, i64 1)
+      // CHECK: call void @__kmpc_dispatch_fini_8u
+      // CHECK: %[[continue:.*]] = call i32 @__kmpc_dispatch_next_8u
+      // CHECK: %[[cond:.*]] = icmp ne i32 %[[continue]], 0
+      // CHECK  br i1 %[[cond]], label %omp_loop.header{{.*}}, label %omp_loop.exit{{.*}}
+      llvm.call @body(%iv) : (i64) -> ()
+      omp.yield
+    }
+    omp.terminator
+  }
+  llvm.return
 }
 
 // -----
@@ -829,18 +869,20 @@ llvm.func @test_omp_wsloop_static_ordered(%lb : i64, %ub : i64, %step : i64) ->
 llvm.func @body(i32)
 
 llvm.func @test_omp_wsloop_static_chunk_ordered(%lb : i32, %ub : i32, %step : i32) -> () {
- %static_chunk_size = llvm.mlir.constant(1 : i32) : i32
- omp.wsloop schedule(static = %static_chunk_size : i32) ordered(0)
- for (%iv) : i32 = (%lb) to (%ub) step (%step) {
-  // CHECK: call void @__kmpc_dispatch_init_4u(ptr @{{.*}}, i32 %{{.*}}, i32 65, i32 1, i32 %{{.*}}, i32 1, i32 1)
-  // CHECK: call void @__kmpc_dispatch_fini_4u
-  // CHECK: %[[continue:.*]] = call i32 @__kmpc_dispatch_next_4u
-  // CHECK: %[[cond:.*]] = icmp ne i32 %[[continue]], 0
-  // CHECK  br i1 %[[cond]], label %omp_loop.header{{.*}}, label %omp_loop.exit{{.*}}
-   llvm.call @body(%iv) : (i32) -> ()
-   omp.yield
- }
- llvm.return
+  %static_chunk_size = llvm.mlir.constant(1 : i32) : i32
+  omp.wsloop schedule(static = %static_chunk_size : i32) ordered(0) {
+    omp.loop_nest (%iv) : i32 = (%lb) to (%ub) step (%step) {
+      // CHECK: call void @__kmpc_dispatch_init_4u(ptr @{{.*}}, i32 %{{.*}}, i32 65, i32 1, i32 %{{.*}}, i32 1, i32 1)
+      // CHECK: call void @__kmpc_dispatch_fini_4u
+      // CHECK: %[[continue:.*]] = call i32 @__kmpc_dispatch_next_4u
+      // CHECK: %[[cond:.*]] = icmp ne i32 %[[continue]], 0
+      // CHECK  br i1 %[[cond]], label %omp_loop.header{{.*}}, label %omp_loop.exit{{.*}}
+      llvm.call @body(%iv) : (i32) -> ()
+      omp.yield
+    }
+    omp.terminator
+  }
+  llvm.return
 }
 
 // -----
@@ -848,17 +890,19 @@ llvm.func @test_omp_wsloop_static_chunk_ordered(%lb : i32, %ub : i32, %step : i3
 llvm.func @body(i64)
 
 llvm.func @test_omp_wsloop_dynamic_ordered(%lb : i64, %ub : i64, %step : i64) -> () {
- omp.wsloop schedule(dynamic) ordered(0)
- for (%iv) : i64 = (%lb) to (%ub) step (%step) {
-  // CHECK: call void @__kmpc_dispatch_init_8u(ptr @{{.*}}, i32 %{{.*}}, i32 67, i64 1, i64 %{{.*}}, i64 1, i64 1)
-  // CHECK: call void @__kmpc_dispatch_fini_8u
-  // CHECK: %[[continue:.*]] = call i32 @__kmpc_dispatch_next_8u
-  // CHECK: %[[cond:.*]] = icmp ne i32 %[[continue]], 0
-  // CHECK  br i1 %[[cond]], label %omp_loop.header{{.*}}, label %omp_loop.exit{{.*}}
-   llvm.call @body(%iv) : (i64) -> ()
-   omp.yield
- }
- llvm.return
+  omp.wsloop schedule(dynamic) ordered(0) {
+    omp.loop_nest (%iv) : i64 = (%lb) to (%ub) step (%step) {
+      // CHECK: call void @__kmpc_dispatch_init_8u(ptr @{{.*}}, i32 %{{.*}}, i32 67, i64 1, i64 %{{.*}}, i64 1, i64 1)
+      // CHECK: call void @__kmpc_dispatch_fini_8u
+      // CHECK: %[[continue:.*]] = call i32 @__kmpc_dispatch_next_8u
+      // CHECK: %[[cond:.*]] = icmp ne i32 %[[continue]], 0
+      // CHECK  br i1 %[[cond]], label %omp_loop.header{{.*}}, label %omp_loop.exit{{.*}}
+      llvm.call @body(%iv) : (i64) -> ()
+      omp.yield
+    }
+    omp.terminator
+  }
+  llvm.return
 }
 
 // -----
@@ -866,17 +910,19 @@ llvm.func @test_omp_wsloop_dynamic_ordered(%lb : i64, %ub : i64, %step : i64) ->
 llvm.func @body(i64)
 
 llvm.func @test_omp_wsloop_auto_ordered(%lb : i64, %ub : i64, %step : i64) -> () {
- omp.wsloop schedule(auto) ordered(0)
- for (%iv) : i64 = (%lb) to (%ub) step (%step) {
-  // CHECK: call void @__kmpc_dispatch_init_8u(ptr @{{.*}}, i32 %{{.*}}, i32 70, i64 1, i64 %{{.*}}, i64 1, i64 1)
-  // CHECK: call void @__kmpc_dispatch_fini_8u
-  // CHECK: %[[continue:.*]] = call i32 @__kmpc_dispatch_next_8u
-  // CHECK: %[[cond:.*]] = icmp ne i32 %[[continue]], 0
-  // CHECK  br i1 %[[cond]], label %omp_loop.header{{.*}}, label %omp_loop.exit{{.*}}
-   llvm.call @body(%iv) : (i64) -> ()
-   omp.yield
- }
- llvm.return
+  omp.wsloop schedule(auto) ordered(0) {
+    omp.loop_nest (%iv) : i64 = (%lb) to (%ub) step (%step) {
+      // CHECK: call void @__kmpc_dispatch_init_8u(ptr @{{.*}}, i32 %{{.*}}, i32 70, i64 1, i64 %{{.*}}, i64 1, i64 1)
+      // CHECK: call void @__kmpc_dispatch_fini_8u
+      // CHECK: %[[continue:.*]] = call i32 @__kmpc_dispatch_next_8u
+      // CHECK: %[[cond:.*]] = icmp ne i32 %[[continue]], 0
+      // CHECK  br i1 %[[cond]], label %omp_loop.header{{.*}}, label %omp_loop.exit{{.*}}
+      llvm.call @body(%iv) : (i64) -> ()
+      omp.yield
+    }
+    omp.terminator
+  }
+  llvm.return
 }
 
 // -----
@@ -884,17 +930,19 @@ llvm.func @test_omp_wsloop_auto_ordered(%lb : i64, %ub : i64, %step : i64) -> ()
 llvm.func @body(i64)
 
 llvm.func @test_omp_wsloop_runtime_ordered(%lb : i64, %ub : i64, %step : i64) -> () {
- omp.wsloop schedule(runtime) ordered(0)
- for (%iv) : i64 = (%lb) to (%ub) step (%step) {
-  // CHECK: call void @__kmpc_dispatch_init_8u(ptr @{{.*}}, i32 %{{.*}}, i32 69, i64 1, i64 %{{.*}}, i64 1, i64 1)
-  // CHECK: call void @__kmpc_dispatch_fini_8u
-  // CHECK: %[[continue:.*]] = call i32 @__kmpc_dispatch_next_8u
-  // CHECK: %[[cond:.*]] = icmp ne i32 %[[continue]], 0
-  // CHECK  br i1 %[[cond]], label %omp_loop.header{{.*}}, label %omp_loop.exit{{.*}}
-   llvm.call @body(%iv) : (i64) -> ()
-   omp.yield
- }
- llvm.return
+  omp.wsloop schedule(runtime) ordered(0) {
+    omp.loop_nest (%iv) : i64 = (%lb) to (%ub) step (%step) {
+      // CHECK: call void @__kmpc_dispatch_init_8u(ptr @{{.*}}, i32 %{{.*}}, i32 69, i64 1, i64 %{{.*}}, i64 1, i64 1)
+      // CHECK: call void @__kmpc_dispatch_fini_8u
+      // CHECK: %[[continue:.*]] = call i32 @__kmpc_dispatch_next_8u
+      // CHECK: %[[cond:.*]] = icmp ne i32 %[[continue]], 0
+      // CHECK  br i1 %[[cond]], label %omp_loop.header{{.*}}, label %omp_loop.exit{{.*}}
+      llvm.call @body(%iv) : (i64) -> ()
+      omp.yield
+    }
+    omp.terminator
+  }
+  llvm.return
 }
 
 // -----
@@ -902,17 +950,19 @@ llvm.func @test_omp_wsloop_runtime_ordered(%lb : i64, %ub : i64, %step : i64) ->
 llvm.func @body(i64)
 
 llvm.func @test_omp_wsloop_guided_ordered(%lb : i64, %ub : i64, %step : i64) -> () {
- omp.wsloop schedule(guided) ordered(0)
- for (%iv) : i64 = (%lb) to (%ub) step (%step) {
-  // CHECK: call void @__kmpc_dispatch_init_8u(ptr @{{.*}}, i32 %{{.*}}, i32 68, i64 1, i64 %{{.*}}, i64 1, i64 1)
-  // CHECK: call void @__kmpc_dispatch_fini_8u
-  // CHECK: %[[continue:.*]] = call i32 @__kmpc_dispatch_next_8u
-  // CHECK: %[[cond:.*]] = icmp ne i32 %[[continue]], 0
-  // CHECK  br i1 %[[cond]], label %omp_loop.header{{.*}}, label %omp_loop.exit{{.*}}
-   llvm.call @body(%iv) : (i64) -> ()
-   omp.yield
- }
- llvm.return
+  omp.wsloop schedule(guided) ordered(0) {
+    omp.loop_nest (%iv) : i64 = (%lb) to (%ub) step (%step) {
+      // CHECK: call void @__kmpc_dispatch_init_8u(ptr @{{.*}}, i32 %{{.*}}, i32 68, i64 1, i64 %{{.*}}, i64 1, i64 1)
+      // CHECK: call void @__kmpc_dispatch_fini_8u
+      // CHECK: %[[continue:.*]] = call i32 @__kmpc_dispatch_next_8u
+      // CHECK: %[[cond:.*]] = icmp ne i32 %[[continue]], 0
+      // CHECK  br i1 %[[cond]], label %omp_loop.header{{.*}}, label %omp_loop.exit{{.*}}
+      llvm.call @body(%iv) : (i64) -> ()
+      omp.yield
+    }
+    omp.terminator
+  }
+  llvm.return
 }
 
 // -----
@@ -920,17 +970,19 @@ llvm.func @test_omp_wsloop_guided_ordered(%lb : i64, %ub : i64, %step : i64) ->
 llvm.func @body(i64)
 
 llvm.func @test_omp_wsloop_dynamic_nonmonotonic_ordered(%lb : i64, %ub : i64, %step : i64) -> () {
- omp.wsloop schedule(dynamic, nonmonotonic) ordered(0)
- for (%iv) : i64 = (%lb) to (%ub) step (%step) {
-  // CHECK: call void @__kmpc_dispatch_init_8u(ptr @{{.*}}, i32 %{{.*}}, i32 1073741891, i64 1, i64 %{{.*}}, i64 1, i64 1)
-  // CHECK: call void @__kmpc_dispatch_fini_8u
-  // CHECK: %[[continue:.*]] = call i32 @__kmpc_dispatch_next_8u
-  // CHECK: %[[cond:.*]] = icmp ne i32 %[[continue]], 0
-  // CHECK  br i1 %[[cond]], label %omp_loop.header{{.*}}, label %omp_loop.exit{{.*}}
-   llvm.call @body(%iv) : (i64) -> ()
-   omp.yield
- }
- llvm.return
+  omp.wsloop schedule(dynamic, nonmonotonic) ordered(0) {
+    omp.loop_nest (%iv) : i64 = (%lb) to (%ub) step (%step) {
+      // CHECK: call void @__kmpc_dispatch_init_8u(ptr @{{.*}}, i32 %{{.*}}, i32 1073741891, i64 1, i64 %{{.*}}, i64 1, i64 1)
+      // CHECK: call void @__kmpc_dispatch_fini_8u
+      // CHECK: %[[continue:.*]] = call i32 @__kmpc_dispatch_next_8u
+      // CHECK: %[[cond:.*]] = icmp ne i32 %[[continue]], 0
+      // CHECK  br i1 %[[cond]], label %omp_loop.header{{.*}}, label %omp_loop.exit{{.*}}
+      llvm.call @body(%iv) : (i64) -> ()
+      omp.yield
+    }
+    omp.terminator
+  }
+  llvm.return
 }
 
 // -----
@@ -938,17 +990,19 @@ llvm.func @test_omp_wsloop_dynamic_nonmonotonic_ordered(%lb : i64, %ub : i64, %s
 llvm.func @body(i64)
 
 llvm.func @test_omp_wsloop_dynamic_monotonic_ordered(%lb : i64, %ub : i64, %step : i64) -> () {
- omp.wsloop schedule(dynamic, monotonic) ordered(0)
- for (%iv) : i64 = (%lb) to (%ub) step (%step) {
-  // CHECK: call void @__kmpc_dispatch_init_8u(ptr @{{.*}}, i32 %{{.*}}, i32 536870979, i64 1, i64 %{{.*}}, i64 1, i64 1)
-  // CHECK: call void @__kmpc_dispatch_fini_8u
-  // CHECK: %[[continue:.*]] = call i32 @__kmpc_dispatch_next_8u
-  // CHECK: %[[cond:.*]] = icmp ne i32 %[[continue]], 0
-  // CHECK  br i1 %[[cond]], label %omp_loop.header{{.*}}, label %omp_loop.exit{{.*}}
-   llvm.call @body(%iv) : (i64) -> ()
-   omp.yield
- }
- llvm.return
+  omp.wsloop schedule(dynamic, monotonic) ordered(0) {
+    omp.loop_nest (%iv) : i64 = (%lb) to (%ub) step (%step) {
+      // CHECK: call void @__kmpc_dispatch_init_8u(ptr @{{.*}}, i32 %{{.*}}, i32 536870979, i64 1, i64 %{{.*}}, i64 1, i64 1)
+      // CHECK: call void @__kmpc_dispatch_fini_8u
+      // CHECK: %[[continue:.*]] = call i32 @__kmpc_dispatch_next_8u
+      // CHECK: %[[cond:.*]] = icmp ne i32 %[[continue]], 0
+      // CHECK  br i1 %[[cond]], label %omp_loop.header{{.*}}, label %omp_loop.exit{{.*}}
+      llvm.call @body(%iv) : (i64) -> ()
+      omp.yield
+    }
+    omp.terminator
+  }
+  llvm.return
 }
 
 // -----
@@ -1114,14 +1168,16 @@ llvm.func @collapse_wsloop(
     // CHECK: %[[TOTAL_SUB_1:.*]] = sub i32 %[[TOTAL]], 1
     // CHECK: store i32 %[[TOTAL_SUB_1]], ptr
     // CHECK: call void @__kmpc_for_static_init_4u
-    omp.wsloop
-    for (%arg0, %arg1, %arg2) : i32 = (%0, %1, %2) to (%3, %4, %5) step (%6, %7, %8) {
-      %31 = llvm.load %20 : !llvm.ptr -> i32
-      %32 = llvm.add %31, %arg0 : i32
-      %33 = llvm.add %32, %arg1 : i32
-      %34 = llvm.add %33, %arg2 : i32
-      llvm.store %34, %20 : i32, !llvm.ptr
-      omp.yield
+    omp.wsloop {
+      omp.loop_nest (%arg0, %arg1, %arg2) : i32 = (%0, %1, %2) to (%3, %4, %5) step (%6, %7, %8) {
+        %31 = llvm.load %20 : !llvm.ptr -> i32
+        %32 = llvm.add %31, %arg0 : i32
+        %33 = llvm.add %32, %arg1 : i32
+        %34 = llvm.add %33, %arg2 : i32
+        llvm.store %34, %20 : i32, !llvm.ptr
+        omp.yield
+      }
+      omp.terminator
     }
     omp.terminator
   }
@@ -1175,14 +1231,16 @@ llvm.func @collapse_wsloop_dynamic(
     // CHECK: store i32 1, ptr
     // CHECK: store i32 %[[TOTAL]], ptr
     // CHECK: call void @__kmpc_dispatch_init_4u
-    omp.wsloop schedule(dynamic)
-    for (%arg0, %arg1, %arg2) : i32 = (%0, %1, %2) to (%3, %4, %5) step (%6, %7, %8) {
-      %31 = llvm.load %20 : !llvm.ptr -> i32
-      %32 = llvm.add %31, %arg0 : i32
-      %33 = llvm.add %32, %arg1 : i32
-      %34 = llvm.add %33, %arg2 : i32
-      llvm.store %34, %20 : i32, !llvm.ptr
-      omp.yield
+    omp.wsloop schedule(dynamic) {
+      omp.loop_nest (%arg0, %arg1, %arg2) : i32 = (%0, %1, %2) to (%3, %4, %5) step (%6, %7, %8) {
+        %31 = llvm.load %20 : !llvm.ptr -> i32
+        %32 = llvm.add %31, %arg0 : i32
+        %33 = llvm.add %32, %arg1 : i32
+        %34 = llvm.add %33, %arg2 : i32
+        llvm.store %34, %20 : i32, !llvm.ptr
+        omp.yield
+      }
+      omp.terminator
     }
     omp.terminator
   }
@@ -1207,63 +1265,69 @@ llvm.func @omp_ordered(%arg0 : i32, %arg1 : i32, %arg2 : i32, %arg3 : i64,
   // CHECK: call void @__kmpc_end_ordered(ptr @[[GLOB1]], i32 [[OMP_THREAD]])
   }
 
-  omp.wsloop ordered(0)
-  for (%arg7) : i32 = (%arg0) to (%arg1) step (%arg2) {
-    // CHECK:  call void @__kmpc_ordered(ptr @[[GLOB3:[0-9]+]], i32 [[OMP_THREAD2:%.*]])
-    omp.ordered.region  {
-      omp.terminator
-    // CHECK: call void @__kmpc_end_ordered(ptr @[[GLOB3]], i32 [[OMP_THREAD2]])
+  omp.wsloop ordered(0) {
+    omp.loop_nest (%arg7) : i32 = (%arg0) to (%arg1) step (%arg2) {
+      // CHECK:  call void @__kmpc_ordered(ptr @[[GLOB3:[0-9]+]], i32 [[OMP_THREAD2:%.*]])
+      omp.ordered.region  {
+        omp.terminator
+      // CHECK: call void @__kmpc_end_ordered(ptr @[[GLOB3]], i32 [[OMP_THREAD2]])
+      }
+      omp.yield
     }
-    omp.yield
+    omp.terminator
   }
 
-  omp.wsloop ordered(1)
-  for (%arg7) : i32 = (%arg0) to (%arg1) step (%arg2) {
-    // CHECK: [[TMP:%.*]] = getelementptr inbounds [1 x i64], ptr [[ADDR]], i64 0, i64 0
-    // CHECK: store i64 [[ARG0:%.*]], ptr [[TMP]], align 8
-    // CHECK: [[TMP2:%.*]] = getelementptr inbounds [1 x i64], ptr [[ADDR]], i64 0, i64 0
-    // CHECK: [[OMP_THREAD2:%.*]] = call i32 @__kmpc_global_thread_num(ptr @[[GLOB3:[0-9]+]])
-    // CHECK: call void @__kmpc_doacross_wait(ptr @[[GLOB3]], i32 [[OMP_THREAD2]], ptr [[TMP2]])
-    omp.ordered depend_type(dependsink) depend_vec(%arg3 : i64) {num_loops_val = 1 : i64}
+  omp.wsloop ordered(1) {
+    omp.loop_nest (%arg7) : i32 = (%arg0) to (%arg1) step (%arg2) {
+      // CHECK: [[TMP:%.*]] = getelementptr inbounds [1 x i64], ptr [[ADDR]], i64 0, i64 0
+      // CHECK: store i64 [[ARG0:%.*]], ptr [[TMP]], align 8
+      // CHECK: [[TMP2:%.*]] = getelementptr inbounds [1 x i64], ptr [[ADDR]], i64 0, i64 0
+      // CHECK: [[OMP_THREAD2:%.*]] = call i32 @__kmpc_global_thread_num(ptr @[[GLOB3:[0-9]+]])
+      // CHECK: call void @__kmpc_doacross_wait(ptr @[[GLOB3]], i32 [[OMP_THREAD2]], ptr [[TMP2]])
+      omp.ordered depend_type(dependsink) depend_vec(%arg3 : i64) {num_loops_val = 1 : i64}
 
-    // CHECK: [[TMP3:%.*]] = getelementptr inbounds [1 x i64], ptr [[ADDR3]], i64 0, i64 0
-    // CHECK: store i64 [[ARG0]], ptr [[TMP3]], align 8
-    // CHECK: [[TMP4:%.*]] = getelementptr inbounds [1 x i64], ptr [[ADDR3]], i64 0, i64 0
-    // CHECK: [[OMP_THREAD4:%.*]] = call i32 @__kmpc_global_thread_num(ptr @[[GLOB5:[0-9]+]])
-    // CHECK: call void @__kmpc_doacross_post(ptr @[[GLOB5]], i32 [[OMP_THREAD4]], ptr [[TMP4]])
-    omp.ordered depend_type(dependsource) depend_vec(%arg3 : i64) {num_loops_val = 1 : i64}
+      // CHECK: [[TMP3:%.*]] = getelementptr inbounds [1 x i64], ptr [[ADDR3]], i64 0, i64 0
+      // CHECK: store i64 [[ARG0]], ptr [[TMP3]], align 8
+      // CHECK: [[TMP4:%.*]] = getelementptr inbounds [1 x i64], ptr [[ADDR3]], i64 0, i64 0
+      // CHECK: [[OMP_THREAD4:%.*]] = call i32 @__kmpc_global_thread_num(ptr @[[GLOB5:[0-9]+]])
+      // CHECK: call void @__kmpc_doacross_post(ptr @[[GLOB5]], i32 [[OMP_THREAD4]], ptr [[TMP4]])
+      omp.ordered depend_type(dependsource) depend_vec(%arg3 : i64) {num_loops_val = 1 : i64}
 
-    omp.yield
+      omp.yield
+    }
+    omp.terminator
   }
 
-  omp.wsloop ordered(2)
-  for (%arg7) : i32 = (%arg0) to (%arg1) step (%arg2) {
-    // CHECK: [[TMP5:%.*]] = getelementptr inbounds [2 x i64], ptr [[ADDR5]], i64 0, i64 0
-    // CHECK: store i64 [[ARG0]], ptr [[TMP5]], align 8
-    // CHECK: [[TMP6:%.*]] = getelementptr inbounds [2 x i64], ptr [[ADDR5]], i64 0, i64 1
-    // CHECK: store i64 [[ARG1:%.*]], ptr [[TMP6]], align 8
-    // CHECK: [[TMP7:%.*]] = getelementptr inbounds [2 x i64], ptr [[ADDR5]], i64 0, i64 0
-    // CHECK: [[OMP_THREAD6:%.*]] = call i32 @__kmpc_global_thread_num(ptr @[[GLOB7:[0-9]+]])
-    // CHECK: call void @__kmpc_doacross_wait(ptr @[[GLOB7]], i32 [[OMP_THREAD6]], ptr [[TMP7]])
-    // CHECK: [[TMP8:%.*]] = getelementptr inbounds [2 x i64], ptr [[ADDR7]], i64 0, i64 0
-    // CHECK: store i64 [[ARG2:%.*]], ptr [[TMP8]], align 8
-    // CHECK: [[TMP9:%.*]] = getelementptr inbounds [2 x i64], ptr [[ADDR7]], i64 0, i64 1
-    // CHECK: store i64 [[ARG3:%.*]], ptr [[TMP9]], align 8
-    // CHECK: [[TMP10:%.*]] = getelementptr inbounds [2 x i64], ptr [[ADDR7]], i64 0, i64 0
-    // CHECK: [[OMP_THREAD8:%.*]] = call i32 @__kmpc_global_thread_num(ptr @[[GLOB7]])
-    // CHECK: call void @__kmpc_doacross_wait(ptr @[[GLOB7]], i32 [[OMP_THREAD8]], ptr [[TMP10]])
-    omp.ordered depend_type(dependsink) depend_vec(%arg3, %arg4, %arg5, %arg6 : i64, i64, i64, i64) {num_loops_val = 2 : i64}
+  omp.wsloop ordered(2) {
+    omp.loop_nest (%arg7) : i32 = (%arg0) to (%arg1) step (%arg2) {
+      // CHECK: [[TMP5:%.*]] = getelementptr inbounds [2 x i64], ptr [[ADDR5]], i64 0, i64 0
+      // CHECK: store i64 [[ARG0]], ptr [[TMP5]], align 8
+      // CHECK: [[TMP6:%.*]] = getelementptr inbounds [2 x i64], ptr [[ADDR5]], i64 0, i64 1
+      // CHECK: store i64 [[ARG1:%.*]], ptr [[TMP6]], align 8
+      // CHECK: [[TMP7:%.*]] = getelementptr inbounds [2 x i64], ptr [[ADDR5]], i64 0, i64 0
+      // CHECK: [[OMP_THREAD6:%.*]] = call i32 @__kmpc_global_thread_num(ptr @[[GLOB7:[0-9]+]])
+      // CHECK: call void @__kmpc_doacross_wait(ptr @[[GLOB7]], i32 [[OMP_THREAD6]], ptr [[TMP7]])
+      // CHECK: [[TMP8:%.*]] = getelementptr inbounds [2 x i64], ptr [[ADDR7]], i64 0, i64 0
+      // CHECK: store i64 [[ARG2:%.*]], ptr [[TMP8]], align 8
+      // CHECK: [[TMP9:%.*]] = getelementptr inbounds [2 x i64], ptr [[ADDR7]], i64 0, i64 1
+      // CHECK: store i64 [[ARG3:%.*]], ptr [[TMP9]], align 8
+      // CHECK: [[TMP10:%.*]] = getelementptr inbounds [2 x i64], ptr [[ADDR7]], i64 0, i64 0
+      // CHECK: [[OMP_THREAD8:%.*]] = call i32 @__kmpc_global_thread_num(ptr @[[GLOB7]])
+      // CHECK: call void @__kmpc_doacross_wait(ptr @[[GLOB7]], i32 [[OMP_THREAD8]], ptr [[TMP10]])
+      omp.ordered depend_type(dependsink) depend_vec(%arg3, %arg4, %arg5, %arg6 : i64, i64, i64, i64) {num_loops_val = 2 : i64}
+
+      // CHECK: [[TMP11:%.*]] = getelementptr inbounds [2 x i64], ptr [[ADDR9]], i64 0, i64 0
+      // CHECK: store i64 [[ARG0]], ptr [[TMP11]], align 8
+      // CHECK: [[TMP12:%.*]] = getelementptr inbounds [2 x i64], ptr [[ADDR9]], i64 0, i64 1
+      // CHECK: store i64 [[ARG1]], ptr [[TMP12]], align 8
+      // CHECK: [[TMP13:%.*]] = getelementptr inbounds [2 x i64], ptr [[ADDR9]], i64 0, i64 0
+      // CHECK: [[OMP_THREAD10:%.*]] = call i32 @__kmpc_global_thread_num(ptr @[[GLOB9:[0-9]+]])
+      // CHECK: call void @__kmpc_doacross_post(ptr @[[GLOB9]], i32 [[OMP_THREAD10]], ptr [[TMP13]])
+      omp.ordered depend_type(dependsource) depend_vec(%arg3, %arg4 : i64, i64) {num_loops_val = 2 : i64}
 
-    // CHECK: [[TMP11:%.*]] = getelementptr inbounds [2 x i64], ptr [[ADDR9]], i64 0, i64 0
-    // CHECK: store i64 [[ARG0]], ptr [[TMP11]], align 8
-    // CHECK: [[TMP12:%.*]] = getelementptr inbounds [2 x i64], ptr [[ADDR9]], i64 0, i64 1
-    // CHECK: store i64 [[ARG1]], ptr [[TMP12]], align 8
-    // CHECK: [[TMP13:%.*]] = getelementptr inbounds [2 x i64], ptr [[ADDR9]], i64 0, i64 0
-    // CHECK: [[OMP_THREAD10:%.*]] = call i32 @__kmpc_global_thread_num(ptr @[[GLOB9:[0-9]+]])
-    // CHECK: call void @__kmpc_doacross_post(ptr @[[GLOB9]], i32 [[OMP_THREAD10]], ptr [[TMP13]])
-    omp.ordered depend_type(dependsource) depend_vec(%arg3, %arg4 : i64, i64) {num_loops_val = 2 : i64}
-
-    omp.yield
+      omp.yield
+    }
+    omp.terminator
   }
 
   llvm.return
@@ -2133,10 +2197,13 @@ llvm.func @omp_sections_with_clauses() -> () {
 // introduction mechanism itself is tested elsewhere.
 // CHECK-LABEL: @repeated_successor
 llvm.func @repeated_successor(%arg0: i64, %arg1: i64, %arg2: i64, %arg3: i1) {
-  omp.wsloop for (%arg4) : i64 = (%arg0) to (%arg1) step (%arg2)  {
-    llvm.cond_br %arg3, ^bb1(%arg0 : i64), ^bb1(%arg1 : i64)
-  ^bb1(%0: i64):  // 2 preds: ^bb0, ^bb0
-    omp.yield
+  omp.wsloop {
+    omp.loop_nest (%arg4) : i64 = (%arg0) to (%arg1) step (%arg2)  {
+      llvm.cond_br %arg3, ^bb1(%arg0 : i64), ^bb1(%arg1 : i64)
+    ^bb1(%0: i64):  // 2 preds: ^bb0, ^bb0
+      omp.yield
+    }
+    omp.terminator
   }
   llvm.return
 }
diff --git a/mlir/test/Target/LLVMIR/openmp-nested.mlir b/mlir/test/Target/LLVMIR/openmp-nested.mlir
index e1fdfdd24a3c..ce5f22f10d7d 100644
--- a/mlir/test/Target/LLVMIR/openmp-nested.mlir
+++ b/mlir/test/Target/LLVMIR/openmp-nested.mlir
@@ -11,20 +11,26 @@ module {
       %2 = llvm.mlir.constant(0 : index) : i64
       %4 = llvm.mlir.constant(0 : i32) : i32
       %12 = llvm.alloca %0 x i64 : (i64) -> !llvm.ptr
-      omp.wsloop for (%arg2) : i64 = (%2) to (%1) step (%0)  {
-        omp.parallel   {
-          omp.wsloop for (%arg3) : i64 = (%2) to (%0) step (%0)  {
-            llvm.store %2, %12 : i64, !llvm.ptr
-            omp.yield
+      omp.wsloop {
+        omp.loop_nest (%arg2) : i64 = (%2) to (%1) step (%0) {
+          omp.parallel {
+            omp.wsloop {
+              omp.loop_nest (%arg3) : i64 = (%2) to (%0) step (%0) {
+                llvm.store %2, %12 : i64, !llvm.ptr
+                omp.yield
+              }
+              omp.terminator
+            }
+            omp.terminator
           }
-          omp.terminator
+          %19 = llvm.load %12 : !llvm.ptr -> i64
+          %20 = llvm.trunc %19 : i64 to i32
+          %5 = llvm.mlir.addressof @str0 : !llvm.ptr
+          %6 = llvm.getelementptr %5[%4, %4] : (!llvm.ptr, i32, i32) -> !llvm.ptr, !llvm.array<29 x i8>
+          %21 = llvm.call @printf(%6, %20, %20) vararg(!llvm.func<i32 (ptr, ...)>): (!llvm.ptr, i32, i32) -> i32
+          omp.yield
         }
-        %19 = llvm.load %12 : !llvm.ptr -> i64
-        %20 = llvm.trunc %19 : i64 to i32
-        %5 = llvm.mlir.addressof @str0 : !llvm.ptr
-        %6 = llvm.getelementptr %5[%4, %4] : (!llvm.ptr, i32, i32) -> !llvm.ptr, !llvm.array<29 x i8>
-        %21 = llvm.call @printf(%6, %20, %20) vararg(!llvm.func<i32 (ptr, ...)>): (!llvm.ptr, i32, i32) -> i32
-        omp.yield
+        omp.terminator
       }
       omp.terminator
     }
diff --git a/mlir/test/Target/LLVMIR/openmp-reduction.mlir b/mlir/test/Target/LLVMIR/openmp-reduction.mlir
index 39b64d71a227..bfdad8c19335 100644
--- a/mlir/test/Target/LLVMIR/openmp-reduction.mlir
+++ b/mlir/test/Target/LLVMIR/openmp-reduction.mlir
@@ -26,13 +26,15 @@ llvm.func @simple_reduction(%lb : i64, %ub : i64, %step : i64) {
   %c1 = llvm.mlir.constant(1 : i32) : i32
   %0 = llvm.alloca %c1 x i32 : (i32) -> !llvm.ptr
   omp.parallel {
-    omp.wsloop reduction(@add_f32 %0 -> %prv : !llvm.ptr)
-    for (%iv) : i64 = (%lb) to (%ub) step (%step) {
-      %1 = llvm.mlir.constant(2.0 : f32) : f32
-      %2 = llvm.load %prv : !llvm.ptr -> f32
-      %3 = llvm.fadd %1, %2 : f32
-      llvm.store %3, %prv : f32, !llvm.ptr
-      omp.yield
+    omp.wsloop reduction(@add_f32 %0 -> %prv : !llvm.ptr) {
+      omp.loop_nest (%iv) : i64 = (%lb) to (%ub) step (%step) {
+        %1 = llvm.mlir.constant(2.0 : f32) : f32
+        %2 = llvm.load %prv : !llvm.ptr -> f32
+        %3 = llvm.fadd %1, %2 : f32
+        llvm.store %3, %prv : f32, !llvm.ptr
+        omp.yield
+      }
+      omp.terminator
     }
     omp.terminator
   }
@@ -105,16 +107,18 @@ llvm.func @reuse_declaration(%lb : i64, %ub : i64, %step : i64) {
   %0 = llvm.alloca %c1 x i32 : (i32) -> !llvm.ptr
   %2 = llvm.alloca %c1 x i32 : (i32) -> !llvm.ptr
   omp.parallel {
-    omp.wsloop reduction(@add_f32 %0 -> %prv0 : !llvm.ptr, @add_f32 %2 -> %prv1 : !llvm.ptr)
-    for (%iv) : i64 = (%lb) to (%ub) step (%step) {
-      %1 = llvm.mlir.constant(2.0 : f32) : f32
-      %3 = llvm.load %prv0 : !llvm.ptr -> f32
-      %4 = llvm.fadd %3, %1 : f32
-      llvm.store %4, %prv0 : f32, !llvm.ptr
-      %5 = llvm.load %prv1 : !llvm.ptr -> f32
-      %6 = llvm.fadd %5, %1 : f32
-      llvm.store %6, %prv1 : f32, !llvm.ptr
-      omp.yield
+    omp.wsloop reduction(@add_f32 %0 -> %prv0 : !llvm.ptr, @add_f32 %2 -> %prv1 : !llvm.ptr) {
+      omp.loop_nest (%iv) : i64 = (%lb) to (%ub) step (%step) {
+        %1 = llvm.mlir.constant(2.0 : f32) : f32
+        %3 = llvm.load %prv0 : !llvm.ptr -> f32
+        %4 = llvm.fadd %3, %1 : f32
+        llvm.store %4, %prv0 : f32, !llvm.ptr
+        %5 = llvm.load %prv1 : !llvm.ptr -> f32
+        %6 = llvm.fadd %5, %1 : f32
+        llvm.store %6, %prv1 : f32, !llvm.ptr
+        omp.yield
+      }
+      omp.terminator
     }
     omp.terminator
   }
@@ -195,13 +199,15 @@ llvm.func @missing_omp_reduction(%lb : i64, %ub : i64, %step : i64) {
   %0 = llvm.alloca %c1 x i32 : (i32) -> !llvm.ptr
   %2 = llvm.alloca %c1 x i32 : (i32) -> !llvm.ptr
   omp.parallel {
-    omp.wsloop reduction(@add_f32 %0 -> %prv0 : !llvm.ptr, @add_f32 %2 -> %prv1 : !llvm.ptr)
-    for (%iv) : i64 = (%lb) to (%ub) step (%step) {
-      %1 = llvm.mlir.constant(2.0 : f32) : f32
-      %3 = llvm.load %prv0 : !llvm.ptr -> f32
-      %4 = llvm.fadd %3, %1 : f32
-      llvm.store %4, %prv0 : f32, !llvm.ptr
-      omp.yield
+    omp.wsloop reduction(@add_f32 %0 -> %prv0 : !llvm.ptr, @add_f32 %2 -> %prv1 : !llvm.ptr) {
+      omp.loop_nest (%iv) : i64 = (%lb) to (%ub) step (%step) {
+        %1 = llvm.mlir.constant(2.0 : f32) : f32
+        %3 = llvm.load %prv0 : !llvm.ptr -> f32
+        %4 = llvm.fadd %3, %1 : f32
+        llvm.store %4, %prv0 : f32, !llvm.ptr
+        omp.yield
+      }
+      omp.terminator
     }
     omp.terminator
   }
@@ -280,16 +286,18 @@ llvm.func @double_reference(%lb : i64, %ub : i64, %step : i64) {
   %c1 = llvm.mlir.constant(1 : i32) : i32
   %0 = llvm.alloca %c1 x i32 : (i32) -> !llvm.ptr
   omp.parallel {
-    omp.wsloop reduction(@add_f32 %0 -> %prv : !llvm.ptr)
-    for (%iv) : i64 = (%lb) to (%ub) step (%step) {
-      %1 = llvm.mlir.constant(2.0 : f32) : f32
-      %2 = llvm.load %prv : !llvm.ptr -> f32
-      %3 = llvm.fadd %2, %1 : f32
-      llvm.store %3, %prv : f32, !llvm.ptr
-      %4 = llvm.load %prv : !llvm.ptr -> f32
-      %5 = llvm.fadd %4, %1 : f32
-      llvm.store %5, %prv : f32, !llvm.ptr
-      omp.yield
+    omp.wsloop reduction(@add_f32 %0 -> %prv : !llvm.ptr) {
+      omp.loop_nest (%iv) : i64 = (%lb) to (%ub) step (%step) {
+        %1 = llvm.mlir.constant(2.0 : f32) : f32
+        %2 = llvm.load %prv : !llvm.ptr -> f32
+        %3 = llvm.fadd %2, %1 : f32
+        llvm.store %3, %prv : f32, !llvm.ptr
+        %4 = llvm.load %prv : !llvm.ptr -> f32
+        %5 = llvm.fadd %4, %1 : f32
+        llvm.store %5, %prv : f32, !llvm.ptr
+        omp.yield
+      }
+      omp.terminator
     }
     omp.terminator
   }
@@ -374,16 +382,18 @@ llvm.func @no_atomic(%lb : i64, %ub : i64, %step : i64) {
   %0 = llvm.alloca %c1 x i32 : (i32) -> !llvm.ptr
   %2 = llvm.alloca %c1 x i32 : (i32) -> !llvm.ptr
   omp.parallel {
-    omp.wsloop reduction(@add_f32 %0 -> %prv0 : !llvm.ptr, @mul_f32 %2 -> %prv1 : !llvm.ptr)
-    for (%iv) : i64 = (%lb) to (%ub) step (%step) {
-      %1 = llvm.mlir.constant(2.0 : f32) : f32
-      %3 = llvm.load %prv0 : !llvm.ptr -> f32
-      %4 = llvm.fadd %3, %1 : f32
-      llvm.store %4, %prv0 : f32, !llvm.ptr
-      %5 = llvm.load %prv1 : !llvm.ptr -> f32
-      %6 = llvm.fmul %5, %1 : f32
-      llvm.store %6, %prv1 : f32, !llvm.ptr
-      omp.yield
+    omp.wsloop reduction(@add_f32 %0 -> %prv0 : !llvm.ptr, @mul_f32 %2 -> %prv1 : !llvm.ptr) {
+      omp.loop_nest (%iv) : i64 = (%lb) to (%ub) step (%step) {
+        %1 = llvm.mlir.constant(2.0 : f32) : f32
+        %3 = llvm.load %prv0 : !llvm.ptr -> f32
+        %4 = llvm.fadd %3, %1 : f32
+        llvm.store %4, %prv0 : f32, !llvm.ptr
+        %5 = llvm.load %prv1 : !llvm.ptr -> f32
+        %6 = llvm.fmul %5, %1 : f32
+        llvm.store %6, %prv1 : f32, !llvm.ptr
+        omp.yield
+      }
+      omp.terminator
     }
     omp.terminator
   }
@@ -531,12 +541,15 @@ llvm.func @parallel_nested_workshare_reduction(%ub : i64) {
   %step = llvm.mlir.constant(1 : i64) : i64
   
   omp.parallel reduction(@add_i32 %0 -> %prv : !llvm.ptr) {
-    omp.wsloop for (%iv) : i64 = (%lb) to (%ub) step (%step) {
-      %ival = llvm.trunc %iv : i64 to i32
-      %lprv = llvm.load %prv : !llvm.ptr -> i32
-      %add = llvm.add %lprv, %ival : i32
-      llvm.store %add, %prv : i32, !llvm.ptr
-      omp.yield
+    omp.wsloop {
+      omp.loop_nest (%iv) : i64 = (%lb) to (%ub) step (%step) {
+        %ival = llvm.trunc %iv : i64 to i32
+        %lprv = llvm.load %prv : !llvm.ptr -> i32
+        %add = llvm.add %lprv, %ival : i32
+        llvm.store %add, %prv : i32, !llvm.ptr
+        omp.yield
+      }
+      omp.terminator
     }
     omp.terminator
   }
diff --git a/mlir/test/Target/LLVMIR/openmp-wsloop-reduction-cleanup.mlir b/mlir/test/Target/LLVMIR/openmp-wsloop-reduction-cleanup.mlir
index 3842522934e4..7a1a31830ce9 100644
--- a/mlir/test/Target/LLVMIR/openmp-wsloop-reduction-cleanup.mlir
+++ b/mlir/test/Target/LLVMIR/openmp-wsloop-reduction-cleanup.mlir
@@ -30,9 +30,12 @@
     %loop_ub = llvm.mlir.constant(9 : i32) : i32
     %loop_lb = llvm.mlir.constant(0 : i32) : i32
     %loop_step = llvm.mlir.constant(1 : i32) : i32 
-    omp.wsloop byref reduction(@add_reduction_i_32 %1 -> %arg0 : !llvm.ptr, @add_reduction_i_32 %2 -> %arg1 : !llvm.ptr) for (%loop_cnt) : i32 = (%loop_lb) to (%loop_ub) inclusive step (%loop_step) {
-      llvm.store %0, %arg0 : i32, !llvm.ptr
-      llvm.store %0, %arg1 : i32, !llvm.ptr
+    omp.wsloop byref reduction(@add_reduction_i_32 %1 -> %arg0 : !llvm.ptr, @add_reduction_i_32 %2 -> %arg1 : !llvm.ptr) {
+      omp.loop_nest (%loop_cnt) : i32 = (%loop_lb) to (%loop_ub) inclusive step (%loop_step) {
+        llvm.store %0, %arg0 : i32, !llvm.ptr
+        llvm.store %0, %arg1 : i32, !llvm.ptr
+        omp.yield
+      }
       omp.terminator
     }
     llvm.return
diff --git a/mlir/test/Transforms/test-legalizer.mlir b/mlir/test/Transforms/test-legalizer.mlir
index d552f0346644..65c947198e06 100644
--- a/mlir/test/Transforms/test-legalizer.mlir
+++ b/mlir/test/Transforms/test-legalizer.mlir
@@ -427,3 +427,13 @@ func.func @use_of_replaced_bbarg(%arg0: i64) {
   }) : (i64) -> (i64)
   "test.invalid"(%0) : (i64) -> ()
 }
+
+// -----
+
+// CHECK-LABEL: @fold_legalization
+func.func @fold_legalization() -> i32 {
+  // CHECK: op_in_place_self_fold
+  // CHECK-SAME: folded
+  %1 = "test.op_in_place_self_fold"() : () -> (i32)
+  "test.return"(%1) : (i32) -> ()
+}
diff --git a/mlir/test/lib/Dialect/Test/CMakeLists.txt b/mlir/test/lib/Dialect/Test/CMakeLists.txt
index f63e4d330e6a..fab893780933 100644
--- a/mlir/test/lib/Dialect/Test/CMakeLists.txt
+++ b/mlir/test/lib/Dialect/Test/CMakeLists.txt
@@ -31,8 +31,6 @@ mlir_tablegen(TestOpEnums.cpp.inc -gen-enum-defs)
 add_public_tablegen_target(MLIRTestEnumDefIncGen)
 
 set(LLVM_TARGET_DEFINITIONS TestOps.td)
-mlir_tablegen(TestOps.h.inc -gen-op-decls)
-mlir_tablegen(TestOps.cpp.inc -gen-op-defs)
 mlir_tablegen(TestOpsDialect.h.inc -gen-dialect-decls -dialect=test)
 mlir_tablegen(TestOpsDialect.cpp.inc -gen-dialect-defs -dialect=test)
 mlir_tablegen(TestPatterns.inc -gen-rewriters)
@@ -43,6 +41,8 @@ mlir_tablegen(TestOpsSyntax.h.inc -gen-op-decls)
 mlir_tablegen(TestOpsSyntax.cpp.inc -gen-op-defs)
 add_public_tablegen_target(MLIRTestOpsSyntaxIncGen)
 
+add_sharded_ops(TestOps 20)
+
 # Exclude tests from libMLIR.so
 add_mlir_library(MLIRTestDialect
   TestAttributes.cpp
@@ -56,6 +56,7 @@ add_mlir_library(MLIRTestDialect
   TestTypes.cpp
   TestOpsSyntax.cpp
   TestDialectInterfaces.cpp
+  ${SHARDED_SRCS}
 
   EXCLUDE_FROM_LIBMLIR
 
@@ -66,6 +67,7 @@ add_mlir_library(MLIRTestDialect
   MLIRTestTypeDefIncGen
   MLIRTestOpsIncGen
   MLIRTestOpsSyntaxIncGen
+  MLIRTestOpsShardGen
 
   LINK_LIBS PUBLIC
   MLIRControlFlowInterfaces
diff --git a/mlir/test/lib/Dialect/Test/TestDialect.cpp b/mlir/test/lib/Dialect/Test/TestDialect.cpp
index 77fd7e61bd3a..bfb9592e6382 100644
--- a/mlir/test/lib/Dialect/Test/TestDialect.cpp
+++ b/mlir/test/lib/Dialect/Test/TestDialect.cpp
@@ -326,12 +326,9 @@ struct TestOpEffectInterfaceFallback
 void TestDialect::initialize() {
   registerAttributes();
   registerTypes();
-  addOperations<
-#define GET_OP_LIST
-#include "TestOps.cpp.inc"
-      >();
   registerOpsSyntax();
   addOperations<ManualCppOpWithFold>();
+  registerTestDialectOperations(this);
   registerDynamicOp(getDynamicGenericOp(this));
   registerDynamicOp(getDynamicOneOperandTwoResultsOp(this));
   registerDynamicOp(getDynamicCustomParserPrinterOp(this));
diff --git a/mlir/test/lib/Dialect/Test/TestOpDefs.cpp b/mlir/test/lib/Dialect/Test/TestOpDefs.cpp
index 7263774ca158..08df2e5e1228 100644
--- a/mlir/test/lib/Dialect/Test/TestOpDefs.cpp
+++ b/mlir/test/lib/Dialect/Test/TestOpDefs.cpp
@@ -826,6 +826,19 @@ LogicalResult CompareOp::verify() {
 }
 
 //===----------------------------------------------------------------------===//
+// TestOpInPlaceSelfFold
+//===----------------------------------------------------------------------===//
+
+OpFoldResult TestOpInPlaceSelfFold::fold(FoldAdaptor adaptor) {
+  if (!getFolded()) {
+    // The folder adds the "folded" if not present.
+    setFolded(true);
+    return getResult();
+  }
+  return {};
+}
+
+//===----------------------------------------------------------------------===//
 // TestOpFoldWithFoldAdaptor
 //===----------------------------------------------------------------------===//
 
diff --git a/mlir/test/lib/Dialect/Test/TestOps.cpp b/mlir/test/lib/Dialect/Test/TestOps.cpp
index ce7e476be74e..47d5b1b19121 100644
--- a/mlir/test/lib/Dialect/Test/TestOps.cpp
+++ b/mlir/test/lib/Dialect/Test/TestOps.cpp
@@ -14,5 +14,4 @@
 using namespace mlir;
 using namespace test;
 
-#define GET_OP_CLASSES
 #include "TestOps.cpp.inc"
diff --git a/mlir/test/lib/Dialect/Test/TestOps.td b/mlir/test/lib/Dialect/Test/TestOps.td
index b641b3da719c..5352d574ac39 100644
--- a/mlir/test/lib/Dialect/Test/TestOps.td
+++ b/mlir/test/lib/Dialect/Test/TestOps.td
@@ -1351,6 +1351,12 @@ def TestOpInPlaceFold : TEST_Op<"op_in_place_fold"> {
   let hasFolder = 1;
 }
 
+def TestOpInPlaceSelfFold : TEST_Op<"op_in_place_self_fold"> {
+  let arguments = (ins UnitAttr:$folded);
+  let results = (outs I32);
+  let hasFolder = 1;
+}
+
 // Test op that simply returns success.
 def TestOpInPlaceFoldSuccess : TEST_Op<"op_in_place_fold_success"> {
   let results = (outs Variadic<I1>);
diff --git a/mlir/test/lib/Dialect/Test/TestPatterns.cpp b/mlir/test/lib/Dialect/Test/TestPatterns.cpp
index 0c1731ba5f07..f9f7d4eacf94 100644
--- a/mlir/test/lib/Dialect/Test/TestPatterns.cpp
+++ b/mlir/test/lib/Dialect/Test/TestPatterns.cpp
@@ -1168,6 +1168,10 @@ struct TestLegalizePatternDriver
     target.addDynamicallyLegalOp<TestRecursiveRewriteOp>(
         [](TestRecursiveRewriteOp op) { return op.getDepth() == 0; });
 
+    // Create a dynamically legal rule that can only be legalized by folding it.
+    target.addDynamicallyLegalOp<TestOpInPlaceSelfFold>(
+        [](TestOpInPlaceSelfFold op) { return op.getFolded(); });
+
     // Handle a partial conversion.
     if (mode == ConversionMode::Partial) {
       DenseSet<Operation *> unlegalizedOps;
diff --git a/mlir/test/mlir-tblgen/shard-op-defs.td b/mlir/test/mlir-tblgen/shard-op-defs.td
new file mode 100644
index 000000000000..84ac6b0fbe9e
--- /dev/null
+++ b/mlir/test/mlir-tblgen/shard-op-defs.td
@@ -0,0 +1,33 @@
+// RUN: mlir-tblgen -gen-op-defs -op-shard-count=2 -I %S/../../include %s | FileCheck %s --check-prefix=DEFS
+// RUN: mlir-tblgen -gen-op-decls -op-shard-count=2 -I %S/../../include %s | FileCheck %s --check-prefix=DECLS
+
+include "mlir/IR/OpBase.td"
+
+def Test_Dialect : Dialect {
+  let name = "test";
+  let cppNamespace = "test";
+}
+
+class Test_Op<string mnemonic, list<Trait> traits = []> 
+    : Op<Test_Dialect, mnemonic, traits>;
+
+def OpA : Test_Op<"a">;
+def OpB : Test_Op<"b">;
+def OpC : Test_Op<"c">;
+
+// DECLS: OpA
+// DECLS: OpB
+// DECLS: OpC
+// DECLS: registerTestDialectOperations(
+// DECLS: registerTestDialectOperations0(
+// DECLS: registerTestDialectOperations1(
+
+// DEFS-LABEL: GET_OP_DEFS_0
+// DEFS: void test::registerTestDialectOperations(
+// DEFS: void test::registerTestDialectOperations0(
+// DEFS: OpAAdaptor
+// DEFS: OpBAdaptor
+
+// DEFS-LABEL: GET_OP_DEFS_1
+// DEFS: void test::registerTestDialectOperations1(
+// DEFS: OpCAdaptor
diff --git a/mlir/test/python/dialects/llvm.py b/mlir/test/python/dialects/llvm.py
index fb4b343b170b..d9ffdeb65bfd 100644
--- a/mlir/test/python/dialects/llvm.py
+++ b/mlir/test/python/dialects/llvm.py
@@ -107,3 +107,46 @@ def testSmoke():
     )
     result = llvm.UndefOp(mat64f32_t)
     # CHECK: %0 = llvm.mlir.undef : !llvm.struct<(f32, f32, f32, f32, f32, f32, f32, f32, f32, f32, f32, f32, f32, f32, f32, f32)>
+
+
+# CHECK-LABEL: testPointerType
+@constructAndPrintInModule
+def testPointerType():
+    ptr = llvm.PointerType.get()
+    # CHECK: !llvm.ptr
+    print(ptr)
+
+    ptr_with_addr = llvm.PointerType.get(1)
+    # CHECK: !llvm.ptr<1>
+    print(ptr_with_addr)
+
+
+# CHECK-LABEL: testConstant
+@constructAndPrintInModule
+def testConstant():
+    i32 = IntegerType.get_signless(32)
+    c_128 = llvm.mlir_constant(IntegerAttr.get(i32, 128))
+    # CHECK: %{{.*}} = llvm.mlir.constant(128 : i32) : i32
+    print(c_128.owner)
+
+
+# CHECK-LABEL: testIntrinsics
+@constructAndPrintInModule
+def testIntrinsics():
+    i32 = IntegerType.get_signless(32)
+    ptr = llvm.PointerType.get()
+    c_128 = llvm.mlir_constant(IntegerAttr.get(i32, 128))
+    # CHECK: %[[CST128:.*]] = llvm.mlir.constant(128 : i32) : i32
+    print(c_128.owner)
+
+    alloca = llvm.alloca(ptr, c_128, i32)
+    # CHECK: %[[ALLOCA:.*]] = llvm.alloca %[[CST128]] x i32 : (i32) -> !llvm.ptr
+    print(alloca.owner)
+
+    c_0 = llvm.mlir_constant(IntegerAttr.get(IntegerType.get_signless(8), 0))
+    # CHECK: %[[CST0:.+]] = llvm.mlir.constant(0 : i8) : i8
+    print(c_0.owner)
+
+    result = llvm.intr_memset(alloca, c_0, c_128, False)
+    # CHECK: "llvm.intr.memset"(%[[ALLOCA]], %[[CST0]], %[[CST128]]) <{isVolatile = false}> : (!llvm.ptr, i8, i32) -> ()
+    print(result)
diff --git a/mlir/test/python/dialects/sparse_tensor/dialect.py b/mlir/test/python/dialects/sparse_tensor/dialect.py
index 5666d090c3d5..3cc4575eb3e2 100644
--- a/mlir/test/python/dialects/sparse_tensor/dialect.py
+++ b/mlir/test/python/dialects/sparse_tensor/dialect.py
@@ -2,6 +2,7 @@
 
 from mlir.ir import *
 from mlir.dialects import sparse_tensor as st
+import textwrap
 
 
 def run(f):
@@ -15,13 +16,18 @@ def run(f):
 def testEncodingAttr1D():
     with Context() as ctx:
         parsed = Attribute.parse(
-            "#sparse_tensor.encoding<{"
-            "  map = (d0) -> (d0 : compressed),"
-            "  posWidth = 16,"
-            "  crdWidth = 32"
-            "}>"
+            textwrap.dedent(
+                """\
+                #sparse_tensor.encoding<{
+                    map = (d0) -> (d0 : compressed),
+                    posWidth = 16,
+                    crdWidth = 32,
+                    explicitVal = 1.0 : f64
+                }>\
+            """
+            )
         )
-        # CHECK: #sparse_tensor.encoding<{ map = (d0) -> (d0 : compressed), posWidth = 16, crdWidth = 32 }>
+        # CHECK: #sparse_tensor.encoding<{ map = (d0) -> (d0 : compressed), posWidth = 16, crdWidth = 32, explicitVal = 1.000000e+00 : f64 }>
         print(parsed)
 
         casted = st.EncodingAttr(parsed)
@@ -38,9 +44,16 @@ def testEncodingAttr1D():
         print(f"pos_width: {casted.pos_width}")
         # CHECK: crd_width: 32
         print(f"crd_width: {casted.crd_width}")
+        # CHECK: explicit_val: 1.000000e+00
+        print(f"explicit_val: {casted.explicit_val}")
+        # CHECK: implicit_val: None
+        print(f"implicit_val: {casted.implicit_val}")
 
-        created = st.EncodingAttr.get(casted.lvl_types, None, None, 0, 0)
-        # CHECK: #sparse_tensor.encoding<{ map = (d0) -> (d0 : compressed) }>
+        new_explicit_val = FloatAttr.get_f64(1.0)
+        created = st.EncodingAttr.get(
+            casted.lvl_types, None, None, 0, 0, new_explicit_val
+        )
+        # CHECK: #sparse_tensor.encoding<{ map = (d0) -> (d0 : compressed), explicitVal = 1.000000e+00 : f64 }>
         print(created)
         # CHECK: created_equal: False
         print(f"created_equal: {created == casted}")
@@ -57,12 +70,16 @@ def testEncodingAttr1D():
 def testEncodingAttrStructure():
     with Context() as ctx:
         parsed = Attribute.parse(
-            "#sparse_tensor.encoding<{"
-            "  map = (d0, d1) -> (d0 : dense, d1 floordiv 4 : dense,"
-            "  d1 mod 4 : structured[2, 4]),"
-            "  posWidth = 16,"
-            "  crdWidth = 32"
-            "}>"
+            textwrap.dedent(
+                """\
+                #sparse_tensor.encoding<{
+                    map = (d0, d1) -> (d0 : dense, d1 floordiv 4 : dense,
+                    d1 mod 4 : structured[2, 4]),
+                    posWidth = 16,
+                    crdWidth = 32,
+                }>\
+            """
+            )
         )
         # CHECK: #sparse_tensor.encoding<{ map = (d0, d1) -> (d0 : dense, d1 floordiv 4 : dense, d1 mod 4 : structured[2, 4]), posWidth = 16, crdWidth = 32 }>
         print(parsed)
@@ -144,11 +161,15 @@ def testEncodingAttrStructure():
 def testEncodingAttr2D():
     with Context() as ctx:
         parsed = Attribute.parse(
-            "#sparse_tensor.encoding<{"
-            "  map = (d0, d1) -> (d1 : dense, d0 : compressed),"
-            "  posWidth = 8,"
-            "  crdWidth = 32"
-            "}>"
+            textwrap.dedent(
+                """\
+                #sparse_tensor.encoding<{
+                    map = (d0, d1) -> (d1 : dense, d0 : compressed),
+                    posWidth = 8,
+                    crdWidth = 32,
+                }>\
+            """
+            )
         )
         # CHECK: #sparse_tensor.encoding<{ map = (d0, d1) -> (d1 : dense, d0 : compressed), posWidth = 8, crdWidth = 32 }>
         print(parsed)
@@ -187,11 +208,15 @@ def testEncodingAttrOnTensorType():
     with Context() as ctx, Location.unknown():
         encoding = st.EncodingAttr(
             Attribute.parse(
-                "#sparse_tensor.encoding<{"
-                "  map = (d0) -> (d0 : compressed), "
-                "  posWidth = 64,"
-                "  crdWidth = 32"
-                "}>"
+                textwrap.dedent(
+                    """\
+                    #sparse_tensor.encoding<{
+                        map = (d0) -> (d0 : compressed),
+                        posWidth = 64,
+                        crdWidth = 32,
+                    }>\
+                """
+                )
             )
         )
         tt = RankedTensorType.get((1024,), F32Type.get(), encoding=encoding)
diff --git a/mlir/test/python/dialects/transform_interpreter.py b/mlir/test/python/dialects/transform_interpreter.py
index 807a98c49327..819a3be1db9d 100644
--- a/mlir/test/python/dialects/transform_interpreter.py
+++ b/mlir/test/python/dialects/transform_interpreter.py
@@ -46,6 +46,21 @@ def print_other():
 
 
 @test_in_context
+def transform_options():
+    options = interp.TransformOptions()
+    options.expensive_checks = False
+    options.enforce_single_top_level_transform_op = True
+    m = ir.Module.parse(
+        print_root_module.replace("from interpreter", "transform_options")
+    )
+    payload = ir.Module.parse("module attributes { this.is.payload } {}")
+    interp.apply_named_sequence(payload, m.body.operations[0], m, options)
+
+
+# CHECK-LABEL: transform_options
+
+
+@test_in_context
 def failed():
     payload = ir.Module.parse("module attributes { this.is.payload } {}")
     try:
diff --git a/mlir/tools/mlir-src-sharder/CMakeLists.txt b/mlir/tools/mlir-src-sharder/CMakeLists.txt
new file mode 100644
index 000000000000..4ef870b61124
--- /dev/null
+++ b/mlir/tools/mlir-src-sharder/CMakeLists.txt
@@ -0,0 +1,14 @@
+set(LLVM_LINK_COMPONENTS Support)
+set(LIBS MLIRSupport)
+
+add_tablegen(mlir-src-sharder MLIR_SRC_SHARDER
+  mlir-src-sharder.cpp
+
+  DEPENDS
+  ${LIBS}
+  )
+
+set_target_properties(mlir-src-sharder PROPERTIES FOLDER "Tablegenning")
+target_link_libraries(mlir-src-sharder PRIVATE ${LIBS})
+
+mlir_check_all_link_libraries(mlir-src-sharder)
diff --git a/mlir/tools/mlir-src-sharder/mlir-src-sharder.cpp b/mlir/tools/mlir-src-sharder/mlir-src-sharder.cpp
new file mode 100644
index 000000000000..dc1e2939c7d2
--- /dev/null
+++ b/mlir/tools/mlir-src-sharder/mlir-src-sharder.cpp
@@ -0,0 +1,114 @@
+//===- mlir-src-sharder.cpp - A tool for sharder generated source files ---===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#include "mlir/Support/FileUtilities.h"
+#include "mlir/Support/LogicalResult.h"
+#include "llvm/Support/CommandLine.h"
+#include "llvm/Support/InitLLVM.h"
+#include "llvm/Support/MemoryBuffer.h"
+#include "llvm/Support/ToolOutputFile.h"
+
+using namespace mlir;
+
+/// Create a dependency file for `-d` option.
+///
+/// This functionality is generally only for the benefit of the build system,
+/// and is modeled after the same option in TableGen.
+static LogicalResult createDependencyFile(StringRef outputFilename,
+                                          StringRef dependencyFile) {
+  if (outputFilename == "-") {
+    llvm::errs() << "error: the option -d must be used together with -o\n";
+    return failure();
+  }
+
+  std::string errorMessage;
+  std::unique_ptr<llvm::ToolOutputFile> outputFile =
+      openOutputFile(dependencyFile, &errorMessage);
+  if (!outputFile) {
+    llvm::errs() << errorMessage << "\n";
+    return failure();
+  }
+
+  outputFile->os() << outputFilename << ":\n";
+  outputFile->keep();
+  return success();
+}
+
+int main(int argc, char **argv) {
+  // FIXME: This is necessary because we link in TableGen, which defines its
+  // options as static variables.. some of which overlap with our options.
+  llvm::cl::ResetCommandLineParser();
+
+  llvm::cl::opt<unsigned> opShardIndex(
+      "op-shard-index", llvm::cl::desc("The current shard index"));
+  llvm::cl::opt<std::string> inputFilename(llvm::cl::Positional,
+                                           llvm::cl::desc("<input file>"),
+                                           llvm::cl::init("-"));
+  llvm::cl::opt<std::string> outputFilename(
+      "o", llvm::cl::desc("Output filename"), llvm::cl::value_desc("filename"),
+      llvm::cl::init("-"));
+  llvm::cl::list<std::string> includeDirs(
+      "I", llvm::cl::desc("Directory of include files"),
+      llvm::cl::value_desc("directory"), llvm::cl::Prefix);
+  llvm::cl::opt<std::string> dependencyFilename(
+      "d", llvm::cl::desc("Dependency filename"),
+      llvm::cl::value_desc("filename"), llvm::cl::init(""));
+  llvm::cl::opt<bool> writeIfChanged(
+      "write-if-changed",
+      llvm::cl::desc("Only write to the output file if it changed"));
+
+  llvm::InitLLVM y(argc, argv);
+  llvm::cl::ParseCommandLineOptions(argc, argv);
+
+  // Open the input file.
+  std::string errorMessage;
+  std::unique_ptr<llvm::MemoryBuffer> inputFile =
+      openInputFile(inputFilename, &errorMessage);
+  if (!inputFile) {
+    llvm::errs() << errorMessage << "\n";
+    return 1;
+  }
+
+  // Write the output to a buffer.
+  std::string outputStr;
+  llvm::raw_string_ostream os(outputStr);
+  os << "#define GET_OP_DEFS_" << opShardIndex << "\n"
+     << inputFile->getBuffer();
+
+  // Determine whether we need to write the output file.
+  bool shouldWriteOutput = true;
+  if (writeIfChanged) {
+    // Only update the real output file if there are any differences. This
+    // prevents recompilation of all the files depending on it if there aren't
+    // any.
+    if (auto existingOrErr =
+            llvm::MemoryBuffer::getFile(outputFilename, /*IsText=*/true))
+      if (std::move(existingOrErr.get())->getBuffer() == os.str())
+        shouldWriteOutput = false;
+  }
+
+  // Populate the output file if necessary.
+  if (shouldWriteOutput) {
+    std::unique_ptr<llvm::ToolOutputFile> outputFile =
+        openOutputFile(outputFilename, &errorMessage);
+    if (!outputFile) {
+      llvm::errs() << errorMessage << "\n";
+      return 1;
+    }
+    outputFile->os() << os.str();
+    outputFile->keep();
+  }
+
+  // Always write the depfile, even if the main output hasn't changed. If it's
+  // missing, Ninja considers the output dirty.
+  if (!dependencyFilename.empty())
+    if (failed(createDependencyFile(outputFilename, dependencyFilename)))
+      return 1;
+
+  return 0;
+}
diff --git a/mlir/tools/mlir-tblgen/OpDefinitionsGen.cpp b/mlir/tools/mlir-tblgen/OpDefinitionsGen.cpp
index 53ed5cb7c043..63fe5a809907 100644
--- a/mlir/tools/mlir-tblgen/OpDefinitionsGen.cpp
+++ b/mlir/tools/mlir-tblgen/OpDefinitionsGen.cpp
@@ -4303,32 +4303,15 @@ void OpOperandAdaptorEmitter::emitDef(
   emitter.adaptor.writeDefTo(os);
 }
 
-// Emits the opcode enum and op classes.
-static void emitOpClasses(const RecordKeeper &recordKeeper,
-                          const std::vector<Record *> &defs, raw_ostream &os,
-                          bool emitDecl) {
-  // First emit forward declaration for each class, this allows them to refer
-  // to each others in traits for example.
-  if (emitDecl) {
-    os << "#if defined(GET_OP_CLASSES) || defined(GET_OP_FWD_DEFINES)\n";
-    os << "#undef GET_OP_FWD_DEFINES\n";
-    for (auto *def : defs) {
-      Operator op(*def);
-      NamespaceEmitter emitter(os, op.getCppNamespace());
-      os << "class " << op.getCppClassName() << ";\n";
-    }
-    os << "#endif\n\n";
-  }
-
-  IfDefScope scope("GET_OP_CLASSES", os);
+/// Emit the class declarations or definitions for the given op defs.
+static void
+emitOpClasses(const RecordKeeper &recordKeeper,
+              const std::vector<Record *> &defs, raw_ostream &os,
+              const StaticVerifierFunctionEmitter &staticVerifierEmitter,
+              bool emitDecl) {
   if (defs.empty())
     return;
 
-  // Generate all of the locally instantiated methods first.
-  StaticVerifierFunctionEmitter staticVerifierEmitter(os, recordKeeper);
-  os << formatv(opCommentHeader, "Local Utility Method", "Definitions");
-  staticVerifierEmitter.emitOpConstraints(defs, emitDecl);
-
   for (auto *def : defs) {
     Operator op(*def);
     if (emitDecl) {
@@ -4358,34 +4341,145 @@ static void emitOpClasses(const RecordKeeper &recordKeeper,
   }
 }
 
-// Emits a comma-separated list of the ops.
-static void emitOpList(const std::vector<Record *> &defs, raw_ostream &os) {
-  IfDefScope scope("GET_OP_LIST", os);
+/// Emit the declarations for the provided op classes.
+static void emitOpClassDecls(const RecordKeeper &recordKeeper,
+                             const std::vector<Record *> &defs,
+                             raw_ostream &os) {
+  // First emit forward declaration for each class, this allows them to refer
+  // to each others in traits for example.
+  for (auto *def : defs) {
+    Operator op(*def);
+    NamespaceEmitter emitter(os, op.getCppNamespace());
+    os << "class " << op.getCppClassName() << ";\n";
+  }
+
+  // Emit the op class declarations.
+  IfDefScope scope("GET_OP_CLASSES", os);
+  if (defs.empty())
+    return;
+  StaticVerifierFunctionEmitter staticVerifierEmitter(os, recordKeeper);
+  staticVerifierEmitter.collectOpConstraints(defs);
+  emitOpClasses(recordKeeper, defs, os, staticVerifierEmitter,
+                /*emitDecl=*/true);
+}
+
+/// Emit the definitions for the provided op classes.
+static void emitOpClassDefs(const RecordKeeper &recordKeeper,
+                            ArrayRef<Record *> defs, raw_ostream &os,
+                            StringRef constraintPrefix = "") {
+  if (defs.empty())
+    return;
+
+  // Generate all of the locally instantiated methods first.
+  StaticVerifierFunctionEmitter staticVerifierEmitter(os, recordKeeper,
+                                                      constraintPrefix);
+  os << formatv(opCommentHeader, "Local Utility Method", "Definitions");
+  staticVerifierEmitter.collectOpConstraints(defs);
+  staticVerifierEmitter.emitOpConstraints(defs);
 
-  interleave(
-      // TODO: We are constructing the Operator wrapper instance just for
-      // getting it's qualified class name here. Reduce the overhead by having a
-      // lightweight version of Operator class just for that purpose.
-      defs, [&os](Record *def) { os << Operator(def).getQualCppClassName(); },
-      [&os]() { os << ",\n"; });
+  // Emit the classes.
+  emitOpClasses(recordKeeper, defs, os, staticVerifierEmitter,
+                /*emitDecl=*/false);
 }
 
+/// Emit op declarations for all op records.
 static bool emitOpDecls(const RecordKeeper &recordKeeper, raw_ostream &os) {
   emitSourceFileHeader("Op Declarations", os, recordKeeper);
 
   std::vector<Record *> defs = getRequestedOpDefinitions(recordKeeper);
-  emitOpClasses(recordKeeper, defs, os, /*emitDecl=*/true);
+  emitOpClassDecls(recordKeeper, defs, os);
+
+  // If we are generating sharded op definitions, emit the sharded op
+  // registration hooks.
+  SmallVector<ArrayRef<Record *>, 4> shardedDefs;
+  shardOpDefinitions(defs, shardedDefs);
+  if (defs.empty() || shardedDefs.size() <= 1)
+    return false;
+
+  Dialect dialect = Operator(defs.front()).getDialect();
+  NamespaceEmitter ns(os, dialect);
+
+  const char *const opRegistrationHook =
+      "void register{0}Operations{1}({2}::{0} *dialect);\n";
+  os << formatv(opRegistrationHook, dialect.getCppClassName(), "",
+                dialect.getCppNamespace());
+  for (unsigned i = 0; i < shardedDefs.size(); ++i) {
+    os << formatv(opRegistrationHook, dialect.getCppClassName(), i,
+                  dialect.getCppNamespace());
+  }
 
   return false;
 }
 
+/// Generate the dialect op registration hook and the op class definitions for a
+/// shard of ops.
+static void emitOpDefShard(const RecordKeeper &recordKeeper,
+                           ArrayRef<Record *> defs, const Dialect &dialect,
+                           unsigned shardIndex, unsigned shardCount,
+                           raw_ostream &os) {
+  std::string shardGuard = "GET_OP_DEFS_";
+  std::string indexStr = std::to_string(shardIndex);
+  shardGuard += indexStr;
+  IfDefScope scope(shardGuard, os);
+
+  // Emit the op registration hook in the first shard.
+  const char *const opRegistrationHook =
+      "void {0}::register{1}Operations{2}({0}::{1} *dialect) {{\n";
+  if (shardIndex == 0) {
+    os << formatv(opRegistrationHook, dialect.getCppNamespace(),
+                  dialect.getCppClassName(), "");
+    for (unsigned i = 0; i < shardCount; ++i) {
+      os << formatv("  {0}::register{1}Operations{2}(dialect);\n",
+                    dialect.getCppNamespace(), dialect.getCppClassName(), i);
+    }
+    os << "}\n";
+  }
+
+  // Generate the per-shard op registration hook.
+  os << formatv(opCommentHeader, dialect.getCppClassName(),
+                "Op Registration Hook")
+     << formatv(opRegistrationHook, dialect.getCppNamespace(),
+                dialect.getCppClassName(), shardIndex);
+  for (Record *def : defs) {
+    os << formatv("  ::mlir::RegisteredOperationName::insert<{0}>(*dialect);\n",
+                  Operator(def).getQualCppClassName());
+  }
+  os << "}\n";
+
+  // Generate the per-shard op definitions.
+  emitOpClassDefs(recordKeeper, defs, os, indexStr);
+}
+
+/// Emit op definitions for all op records.
 static bool emitOpDefs(const RecordKeeper &recordKeeper, raw_ostream &os) {
   emitSourceFileHeader("Op Definitions", os, recordKeeper);
 
   std::vector<Record *> defs = getRequestedOpDefinitions(recordKeeper);
-  emitOpList(defs, os);
-  emitOpClasses(recordKeeper, defs, os, /*emitDecl=*/false);
+  SmallVector<ArrayRef<Record *>, 4> shardedDefs;
+  shardOpDefinitions(defs, shardedDefs);
+
+  // If no shard was requested, emit the regular op list and class definitions.
+  if (shardedDefs.size() == 1) {
+    {
+      IfDefScope scope("GET_OP_LIST", os);
+      interleave(
+          defs, os,
+          [&](Record *def) { os << Operator(def).getQualCppClassName(); },
+          ",\n");
+    }
+    {
+      IfDefScope scope("GET_OP_CLASSES", os);
+      emitOpClassDefs(recordKeeper, defs, os);
+    }
+    return false;
+  }
 
+  if (defs.empty())
+    return false;
+  Dialect dialect = Operator(defs.front()).getDialect();
+  for (auto [idx, value] : llvm::enumerate(shardedDefs)) {
+    emitOpDefShard(recordKeeper, value, dialect, idx, shardedDefs.size(), os);
+  }
   return false;
 }
 
diff --git a/mlir/tools/mlir-tblgen/OpGenHelpers.cpp b/mlir/tools/mlir-tblgen/OpGenHelpers.cpp
index 7fd34df8460d..c2a2423a2402 100644
--- a/mlir/tools/mlir-tblgen/OpGenHelpers.cpp
+++ b/mlir/tools/mlir-tblgen/OpGenHelpers.cpp
@@ -31,6 +31,10 @@ static cl::opt<std::string> opExcFilter(
     "op-exclude-regex",
     cl::desc("Regex of name of op's to exclude (no filter if empty)"),
     cl::cat(opDefGenCat));
+static cl::opt<unsigned> opShardCount(
+    "op-shard-count",
+    cl::desc("The number of shards into which the op classes will be divided"),
+    cl::cat(opDefGenCat), cl::init(1));
 
 static std::string getOperationName(const Record &def) {
   auto prefix = def.getValueAsDef("opDialect")->getValueAsString("name");
@@ -79,4 +83,23 @@ bool mlir::tblgen::isPythonReserved(StringRef str) {
   reserved.insert("issubclass");
   reserved.insert("type");
   return reserved.contains(str);
-}
\ No newline at end of file
+}
+
+void mlir::tblgen::shardOpDefinitions(
+    ArrayRef<llvm::Record *> defs,
+    SmallVectorImpl<ArrayRef<llvm::Record *>> &shardedDefs) {
+  assert(opShardCount > 0 && "expected a positive shard count");
+  if (opShardCount == 1) {
+    shardedDefs.push_back(defs);
+    return;
+  }
+
+  unsigned minShardSize = defs.size() / opShardCount;
+  unsigned numMissing = defs.size() - minShardSize * opShardCount;
+  shardedDefs.reserve(opShardCount);
+  for (unsigned i = 0, start = 0; i < opShardCount; ++i) {
+    unsigned size = minShardSize + (i < numMissing);
+    shardedDefs.push_back(defs.slice(start, size));
+    start += size;
+  }
+}
diff --git a/mlir/tools/mlir-tblgen/OpGenHelpers.h b/mlir/tools/mlir-tblgen/OpGenHelpers.h
index 3dcff14d1221..1b43d5d3ce3a 100644
--- a/mlir/tools/mlir-tblgen/OpGenHelpers.h
+++ b/mlir/tools/mlir-tblgen/OpGenHelpers.h
@@ -13,6 +13,7 @@
 #ifndef MLIR_TOOLS_MLIRTBLGEN_OPGENHELPERS_H_
 #define MLIR_TOOLS_MLIRTBLGEN_OPGENHELPERS_H_
 
+#include "mlir/Support/LLVM.h"
 #include "llvm/TableGen/Record.h"
 #include <vector>
 
@@ -28,6 +29,10 @@ getRequestedOpDefinitions(const llvm::RecordKeeper &recordKeeper);
 /// Regenerate using python -c"print(set(sorted(__import__('keyword').kwlist)))"
 bool isPythonReserved(llvm::StringRef str);
 
+/// Shard the op defintions into the number of shards set by "op-shard-count".
+void shardOpDefinitions(ArrayRef<llvm::Record *> defs,
+                        SmallVectorImpl<ArrayRef<llvm::Record *>> &shardedDefs);
+
 } // namespace tblgen
 } // namespace mlir
 
diff --git a/mlir/unittests/CMakeLists.txt b/mlir/unittests/CMakeLists.txt
index 6fad249a0b2f..6d8aa290e82f 100644
--- a/mlir/unittests/CMakeLists.txt
+++ b/mlir/unittests/CMakeLists.txt
@@ -20,6 +20,7 @@ add_subdirectory(Support)
 add_subdirectory(Rewrite)
 add_subdirectory(TableGen)
 add_subdirectory(Target)
+add_subdirectory(Tools)
 add_subdirectory(Transforms)
 
 if(MLIR_ENABLE_EXECUTION_ENGINE)
diff --git a/mlir/unittests/Tools/CMakeLists.txt b/mlir/unittests/Tools/CMakeLists.txt
new file mode 100644
index 000000000000..a97588d92866
--- /dev/null
+++ b/mlir/unittests/Tools/CMakeLists.txt
@@ -0,0 +1 @@
+add_subdirectory(lsp-server-support)
diff --git a/mlir/unittests/Tools/lsp-server-support/CMakeLists.txt b/mlir/unittests/Tools/lsp-server-support/CMakeLists.txt
new file mode 100644
index 000000000000..3aa8b9c4bc77
--- /dev/null
+++ b/mlir/unittests/Tools/lsp-server-support/CMakeLists.txt
@@ -0,0 +1,6 @@
+add_mlir_unittest(MLIRLspServerSupportTests
+  Transport.cpp
+)
+target_link_libraries(MLIRLspServerSupportTests
+  PRIVATE
+  MLIRLspServerSupportLib)
diff --git a/mlir/unittests/Tools/lsp-server-support/Transport.cpp b/mlir/unittests/Tools/lsp-server-support/Transport.cpp
new file mode 100644
index 000000000000..a086964cd366
--- /dev/null
+++ b/mlir/unittests/Tools/lsp-server-support/Transport.cpp
@@ -0,0 +1,134 @@
+//===- Transport.cpp - LSP JSON transport unit tests ----------------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#include "mlir/Tools/lsp-server-support/Transport.h"
+#include "mlir/Tools/lsp-server-support/Logging.h"
+#include "mlir/Tools/lsp-server-support/Protocol.h"
+#include "llvm/Support/FileSystem.h"
+#include "gmock/gmock.h"
+#include "gtest/gtest.h"
+
+using namespace mlir;
+using namespace mlir::lsp;
+using namespace testing;
+
+namespace {
+
+TEST(TransportTest, SendReply) {
+  std::string out;
+  llvm::raw_string_ostream os(out);
+  JSONTransport transport(nullptr, os);
+  MessageHandler handler(transport);
+
+  transport.reply(1989, nullptr);
+  EXPECT_THAT(out, HasSubstr("\"id\":1989"));
+  EXPECT_THAT(out, HasSubstr("\"result\":null"));
+}
+
+class TransportInputTest : public Test {
+  llvm::SmallVector<char> inputPath;
+  std::FILE *in = nullptr;
+  std::string output = "";
+  llvm::raw_string_ostream os;
+  std::optional<JSONTransport> transport = std::nullopt;
+  std::optional<MessageHandler> messageHandler = std::nullopt;
+
+protected:
+  TransportInputTest() : os(output) {}
+
+  void SetUp() override {
+    std::error_code ec =
+        llvm::sys::fs::createTemporaryFile("lsp-unittest", "json", inputPath);
+    ASSERT_FALSE(ec) << "Could not create temporary file: " << ec.message();
+
+    in = std::fopen(inputPath.data(), "r");
+    ASSERT_TRUE(in) << "Could not open temporary file: "
+                    << std::strerror(errno);
+    transport.emplace(in, os, JSONStreamStyle::Delimited);
+    messageHandler.emplace(*transport);
+  }
+
+  void TearDown() override {
+    EXPECT_EQ(std::fclose(in), 0)
+        << "Could not close temporary file FD: " << std::strerror(errno);
+    std::error_code ec =
+        llvm::sys::fs::remove(inputPath, /*IgnoreNonExisting=*/false);
+    EXPECT_FALSE(ec) << "Could not remove temporary file '" << inputPath.data()
+                     << "': " << ec.message();
+  }
+
+  void writeInput(StringRef buffer) {
+    std::error_code ec;
+    llvm::raw_fd_ostream os(inputPath.data(), ec);
+    ASSERT_FALSE(ec) << "Could not write to '" << inputPath.data()
+                     << "': " << ec.message();
+    os << buffer;
+    os.close();
+  }
+
+  StringRef getOutput() const { return output; }
+  MessageHandler &getMessageHandler() { return *messageHandler; }
+
+  void runTransport() {
+    bool gotEOF = false;
+    llvm::Error err = llvm::handleErrors(
+        transport->run(*messageHandler), [&](const llvm::ECError &ecErr) {
+          gotEOF = ecErr.convertToErrorCode() == std::errc::io_error;
+        });
+    llvm::consumeError(std::move(err));
+    EXPECT_TRUE(gotEOF);
+  }
+};
+
+TEST_F(TransportInputTest, RequestWithInvalidParams) {
+  struct Handler {
+    void onMethod(const TextDocumentItem &params,
+                  mlir::lsp::Callback<TextDocumentIdentifier> callback) {}
+  } handler;
+  getMessageHandler().method("invalid-params-request", &handler,
+                             &Handler::onMethod);
+
+  writeInput("{\"jsonrpc\":\"2.0\",\"id\":92,"
+             "\"method\":\"invalid-params-request\",\"params\":{}}\n");
+  runTransport();
+  EXPECT_THAT(getOutput(), HasSubstr("error"));
+  EXPECT_THAT(getOutput(), HasSubstr("missing value at (root).uri"));
+}
+
+TEST_F(TransportInputTest, NotificationWithInvalidParams) {
+  // JSON parsing errors are only reported via error logging. As a result, this
+  // test can't make any expectations -- but it prints the output anyway, by way
+  // of demonstration.
+  Logger::setLogLevel(Logger::Level::Error);
+
+  struct Handler {
+    void onNotification(const TextDocumentItem &params) {}
+  } handler;
+  getMessageHandler().notification("invalid-params-notification", &handler,
+                                   &Handler::onNotification);
+
+  writeInput("{\"jsonrpc\":\"2.0\",\"method\":\"invalid-params-notification\","
+             "\"params\":{}}\n");
+  runTransport();
+}
+
+TEST_F(TransportInputTest, MethodNotFound) {
+  writeInput("{\"jsonrpc\":\"2.0\",\"id\":29,\"method\":\"ack\"}\n");
+  runTransport();
+  EXPECT_THAT(getOutput(), HasSubstr("\"id\":29"));
+  EXPECT_THAT(getOutput(), HasSubstr("\"error\""));
+  EXPECT_THAT(getOutput(), HasSubstr("\"message\":\"method not found: ack\""));
+}
+
+TEST_F(TransportInputTest, OutgoingNotification) {
+  auto notifyFn = getMessageHandler().outgoingNotification<CompletionList>(
+      "outgoing-notification");
+  notifyFn(CompletionList{});
+  EXPECT_THAT(getOutput(), HasSubstr("\"method\":\"outgoing-notification\""));
+}
+} // namespace
diff --git a/offload/CMakeLists.txt b/offload/CMakeLists.txt
index abc8baa0805f..a416ac29873f 100644
--- a/offload/CMakeLists.txt
+++ b/offload/CMakeLists.txt
@@ -151,6 +151,25 @@ if (NOT LIBOMPTARGET_LLVM_INCLUDE_DIRS)
   message(FATAL_ERROR "Missing definition for LIBOMPTARGET_LLVM_INCLUDE_DIRS")
 endif()
 
+set(LIBOMPTARGET_ALL_PLUGIN_TARGETS amdgpu cuda host)
+set(LIBOMPTARGET_PLUGINS_TO_BUILD "all" CACHE STRING
+    "Semicolon-separated list of plugins to use: cuda, amdgpu, host or \"all\".")
+
+if(LIBOMPTARGET_PLUGINS_TO_BUILD STREQUAL "all")
+  set(LIBOMPTARGET_PLUGINS_TO_BUILD ${LIBOMPTARGET_ALL_PLUGIN_TARGETS})
+endif()
+
+set(LIBOMPTARGET_ENUM_PLUGIN_TARGETS "")
+foreach(plugin IN LISTS LIBOMPTARGET_PLUGINS_TO_BUILD)
+  set(LIBOMPTARGET_ENUM_PLUGIN_TARGETS
+      "${LIBOMPTARGET_ENUM_PLUGIN_TARGETS}PLUGIN_TARGET(${plugin})\n")
+endforeach()
+string(STRIP ${LIBOMPTARGET_ENUM_PLUGIN_TARGETS} LIBOMPTARGET_ENUM_PLUGIN_TARGETS)
+configure_file(
+  ${CMAKE_CURRENT_SOURCE_DIR}/include/Shared/Targets.def.in
+  ${CMAKE_CURRENT_BINARY_DIR}/include/Shared/Targets.def
+)
+
 include_directories(${LIBOMPTARGET_LLVM_INCLUDE_DIRS})
 
 # This is a list of all the targets that are supported/tested right now.
@@ -288,6 +307,7 @@ set(LIBOMPTARGET_GPU_LIBC_SUPPORT ${LLVM_LIBC_GPU_BUILD} CACHE BOOL
 pythonize_bool(LIBOMPTARGET_GPU_LIBC_SUPPORT)
 
 set(LIBOMPTARGET_INCLUDE_DIR ${CMAKE_CURRENT_SOURCE_DIR}/include)
+set(LIBOMPTARGET_BINARY_INCLUDE_DIR ${CMAKE_CURRENT_BINARY_DIR}/include)
 message(STATUS "OpenMP tools dir in libomptarget: ${LIBOMP_OMP_TOOLS_INCLUDE_DIR}")
 if(LIBOMP_OMP_TOOLS_INCLUDE_DIR)
   include_directories(${LIBOMP_OMP_TOOLS_INCLUDE_DIR})
diff --git a/offload/include/Shared/Targets.def.in b/offload/include/Shared/Targets.def.in
new file mode 100644
index 000000000000..f34b523b4542
--- /dev/null
+++ b/offload/include/Shared/Targets.def.in
@@ -0,0 +1,20 @@
+//===-- Shared/Targets.def - Target plugin enumerator -----------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// Enumerates over all of the supported target plugins that are available to
+// the offloading library.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef PLUGIN_TARGET
+#  error Please define the macro PLUGIN_TARGET(TargetName)
+#endif
+
+@LIBOMPTARGET_ENUM_PLUGIN_TARGETS@
+
+#undef PLUGIN_TARGET
diff --git a/offload/plugins-nextgen/CMakeLists.txt b/offload/plugins-nextgen/CMakeLists.txt
index dbd82ac94517..df625e97c7eb 100644
--- a/offload/plugins-nextgen/CMakeLists.txt
+++ b/offload/plugins-nextgen/CMakeLists.txt
@@ -69,9 +69,12 @@ function(add_target_library target_name lib_name)
   set_target_properties(${target_name} PROPERTIES CXX_VISIBILITY_PRESET protected)
 endfunction()
 
-add_subdirectory(amdgpu)
-add_subdirectory(cuda)
-add_subdirectory(host)
+foreach(plugin IN LISTS LIBOMPTARGET_PLUGINS_TO_BUILD)
+  if(NOT EXISTS ${CMAKE_CURRENT_SOURCE_DIR}/${plugin})
+    message(FATAL_ERROR "Unknown plugin target '${plugin}'")
+  endif()
+  add_subdirectory(${plugin})
+endforeach()
 
 # Make sure the parent scope can see the plugins that will be created.
 set(LIBOMPTARGET_SYSTEM_TARGETS "${LIBOMPTARGET_SYSTEM_TARGETS}" PARENT_SCOPE)
diff --git a/offload/plugins-nextgen/common/CMakeLists.txt b/offload/plugins-nextgen/common/CMakeLists.txt
index a7350e662a7c..acf0af63f050 100644
--- a/offload/plugins-nextgen/common/CMakeLists.txt
+++ b/offload/plugins-nextgen/common/CMakeLists.txt
@@ -62,6 +62,7 @@ target_link_options(PluginCommon PUBLIC ${offload_link_flags})
 target_include_directories(PluginCommon PUBLIC 
   ${CMAKE_CURRENT_SOURCE_DIR}/include
   ${LIBOMPTARGET_LLVM_INCLUDE_DIRS}
+  ${LIBOMPTARGET_BINARY_INCLUDE_DIR}
   ${LIBOMPTARGET_INCLUDE_DIR}
 )
 
diff --git a/offload/plugins-nextgen/host/CMakeLists.txt b/offload/plugins-nextgen/host/CMakeLists.txt
index 7da18ee278d4..6407f72e8db0 100644
--- a/offload/plugins-nextgen/host/CMakeLists.txt
+++ b/offload/plugins-nextgen/host/CMakeLists.txt
@@ -14,36 +14,36 @@ if(CMAKE_SYSTEM_PROCESSOR MATCHES "ppc64le$")
 endif()
 
 # Create the library and add the default arguments.
-add_target_library(omptarget.rtl.${machine} ${machine})
+add_target_library(omptarget.rtl.host ${machine})
 
-target_sources(omptarget.rtl.${machine} PRIVATE src/rtl.cpp)
+target_sources(omptarget.rtl.host PRIVATE src/rtl.cpp)
 
 if(LIBOMPTARGET_DEP_LIBFFI_FOUND)
   libomptarget_say("Building ${machine} plugin linked with libffi")
   if(FFI_STATIC_LIBRARIES)
-    target_link_libraries(omptarget.rtl.${machine} PRIVATE FFI::ffi_static)
+    target_link_libraries(omptarget.rtl.host PRIVATE FFI::ffi_static)
   else()
-    target_link_libraries(omptarget.rtl.${machine} PRIVATE FFI::ffi)
+    target_link_libraries(omptarget.rtl.host PRIVATE FFI::ffi)
   endif()
 else()
   libomptarget_say("Building ${machine} plugin for dlopened libffi")
-  target_sources(omptarget.rtl.${machine} PRIVATE dynamic_ffi/ffi.cpp)
-  target_include_directories(omptarget.rtl.${machine} PRIVATE dynamic_ffi)
+  target_sources(omptarget.rtl.host PRIVATE dynamic_ffi/ffi.cpp)
+  target_include_directories(omptarget.rtl.host PRIVATE dynamic_ffi)
 endif()
 
 # Install plugin under the lib destination folder.
-install(TARGETS omptarget.rtl.${machine}
+install(TARGETS omptarget.rtl.host
         LIBRARY DESTINATION "${OFFLOAD_INSTALL_LIBDIR}")
-set_target_properties(omptarget.rtl.${machine} PROPERTIES
+set_target_properties(omptarget.rtl.host PROPERTIES
   INSTALL_RPATH "$ORIGIN" BUILD_RPATH "$ORIGIN:${CMAKE_CURRENT_BINARY_DIR}/.."
   POSITION_INDEPENDENT_CODE ON
   CXX_VISIBILITY_PRESET protected)
 
-target_include_directories(omptarget.rtl.${machine} PRIVATE
+target_include_directories(omptarget.rtl.host PRIVATE
                            ${LIBOMPTARGET_INCLUDE_DIR})
 
 if(LIBOMPTARGET_DEP_LIBFFI_FOUND)
-  list(APPEND LIBOMPTARGET_TESTED_PLUGINS omptarget.rtl.${machine})
+  list(APPEND LIBOMPTARGET_TESTED_PLUGINS omptarget.rtl.host)
   set(LIBOMPTARGET_TESTED_PLUGINS
       "${LIBOMPTARGET_TESTED_PLUGINS}" PARENT_SCOPE)
 else()
@@ -52,36 +52,36 @@ endif()
 
 # Define the target specific triples and ELF machine values.
 if(CMAKE_SYSTEM_PROCESSOR MATCHES "ppc64le$")
-  target_compile_definitions(omptarget.rtl.${machine} PRIVATE TARGET_ELF_ID=EM_PPC64)
-  target_compile_definitions(omptarget.rtl.${machine} PRIVATE
+  target_compile_definitions(omptarget.rtl.host PRIVATE TARGET_ELF_ID=EM_PPC64)
+  target_compile_definitions(omptarget.rtl.host PRIVATE
       LIBOMPTARGET_NEXTGEN_GENERIC_PLUGIN_TRIPLE="powerpc64le-ibm-linux-gnu")
   list(APPEND LIBOMPTARGET_SYSTEM_TARGETS 
        "powerpc64le-ibm-linux-gnu" "powerpc64le-ibm-linux-gnu-LTO")
   set(LIBOMPTARGET_SYSTEM_TARGETS "${LIBOMPTARGET_SYSTEM_TARGETS}" PARENT_SCOPE)
 elseif(CMAKE_SYSTEM_PROCESSOR MATCHES "ppc64$")
-  target_compile_definitions(omptarget.rtl.${machine} PRIVATE TARGET_ELF_ID=EM_PPC64)
-  target_compile_definitions(omptarget.rtl.${machine} PRIVATE
+  target_compile_definitions(omptarget.rtl.host PRIVATE TARGET_ELF_ID=EM_PPC64)
+  target_compile_definitions(omptarget.rtl.host PRIVATE
       LIBOMPTARGET_NEXTGEN_GENERIC_PLUGIN_TRIPLE="powerpc64-ibm-linux-gnu")
   list(APPEND LIBOMPTARGET_SYSTEM_TARGETS 
        "powerpc64-ibm-linux-gnu" "powerpc64-ibm-linux-gnu-LTO")
   set(LIBOMPTARGET_SYSTEM_TARGETS "${LIBOMPTARGET_SYSTEM_TARGETS}" PARENT_SCOPE)
 elseif(CMAKE_SYSTEM_PROCESSOR MATCHES "x86_64$")
-  target_compile_definitions(omptarget.rtl.${machine} PRIVATE TARGET_ELF_ID=EM_X86_64)
-  target_compile_definitions(omptarget.rtl.${machine} PRIVATE
+  target_compile_definitions(omptarget.rtl.host PRIVATE TARGET_ELF_ID=EM_X86_64)
+  target_compile_definitions(omptarget.rtl.host PRIVATE
       LIBOMPTARGET_NEXTGEN_GENERIC_PLUGIN_TRIPLE="x86_64-pc-linux-gnu")
   list(APPEND LIBOMPTARGET_SYSTEM_TARGETS 
        "x86_64-pc-linux-gnu" "x86_64-pc-linux-gnu-LTO")
   set(LIBOMPTARGET_SYSTEM_TARGETS "${LIBOMPTARGET_SYSTEM_TARGETS}" PARENT_SCOPE)
 elseif(CMAKE_SYSTEM_PROCESSOR MATCHES "aarch64$")
-  target_compile_definitions(omptarget.rtl.${machine} PRIVATE TARGET_ELF_ID=EM_AARCH64)
-  target_compile_definitions(omptarget.rtl.${machine} PRIVATE
+  target_compile_definitions(omptarget.rtl.host PRIVATE TARGET_ELF_ID=EM_AARCH64)
+  target_compile_definitions(omptarget.rtl.host PRIVATE
       LIBOMPTARGET_NEXTGEN_GENERIC_PLUGIN_TRIPLE="aarch64-unknown-linux-gnu")
   list(APPEND LIBOMPTARGET_SYSTEM_TARGETS 
        "aarch64-unknown-linux-gnu" "aarch64-unknown-linux-gnu-LTO")
   set(LIBOMPTARGET_SYSTEM_TARGETS "${LIBOMPTARGET_SYSTEM_TARGETS}" PARENT_SCOPE)
 elseif(CMAKE_SYSTEM_PROCESSOR MATCHES "s390x$")
-  target_compile_definitions(omptarget.rtl.${machine} PRIVATE TARGET_ELF_ID=EM_S390)
-  target_compile_definitions(omptarget.rtl.${machine} PRIVATE
+  target_compile_definitions(omptarget.rtl.host PRIVATE TARGET_ELF_ID=EM_S390)
+  target_compile_definitions(omptarget.rtl.host PRIVATE
       LIBOMPTARGET_NEXTGEN_GENERIC_PLUGIN_TRIPLE="s390x-ibm-linux-gnu")
   list(APPEND LIBOMPTARGET_SYSTEM_TARGETS 
        "s390x-ibm-linux-gnu" "s390x-ibm-linux-gnu-LTO")
diff --git a/offload/src/CMakeLists.txt b/offload/src/CMakeLists.txt
index fb1ad3d7ae70..eda5a85ff1ab 100644
--- a/offload/src/CMakeLists.txt
+++ b/offload/src/CMakeLists.txt
@@ -37,6 +37,7 @@ add_llvm_library(omptarget
 
   ADDITIONAL_HEADER_DIRS
   ${LIBOMPTARGET_INCLUDE_DIR}
+  ${LIBOMPTARGET_BINARY_INCLUDE_DIR}
 
   LINK_COMPONENTS
   Support
@@ -49,7 +50,9 @@ add_llvm_library(omptarget
   NO_INSTALL_RPATH
   BUILDTREE_ONLY
 )
-target_include_directories(omptarget PRIVATE ${LIBOMPTARGET_INCLUDE_DIR})
+target_include_directories(omptarget PRIVATE 
+  ${LIBOMPTARGET_INCLUDE_DIR} ${LIBOMPTARGET_BINARY_INCLUDE_DIR}
+)
 
 if (LIBOMP_HAVE_VERSION_SCRIPT_FLAG)
   target_link_libraries(omptarget PRIVATE
@@ -65,23 +68,6 @@ target_compile_definitions(omptarget PRIVATE
 target_compile_options(omptarget PUBLIC ${offload_compile_flags})
 target_link_options(omptarget PUBLIC ${offload_link_flags})
 
-macro(check_plugin_target target)
-if (TARGET omptarget.rtl.${target})
-	list(APPEND LIBOMPTARGET_PLUGINS_TO_LOAD ${target})
-endif()
-endmacro()
-
-set(LIBOMPTARGET_PLUGINS_TO_LOAD "" CACHE STRING
-  "Comma separated list of plugin names to look for at runtime")
-if (NOT LIBOMPTARGET_PLUGINS_TO_LOAD)
-	check_plugin_target(ppc64)
-	check_plugin_target(x86_64)
-	check_plugin_target(cuda)
-	check_plugin_target(aarch64)
-	check_plugin_target(amdgpu)
-	check_plugin_target(s390x)
-endif()
-
 list(TRANSFORM LIBOMPTARGET_PLUGINS_TO_LOAD PREPEND "\"libomptarget.rtl.")
 list(TRANSFORM LIBOMPTARGET_PLUGINS_TO_LOAD APPEND "\"")
 list(JOIN LIBOMPTARGET_PLUGINS_TO_LOAD "," ENABLED_OFFLOAD_PLUGINS)
diff --git a/offload/src/PluginManager.cpp b/offload/src/PluginManager.cpp
index 792cae3e3dd5..dbb556c179e5 100644
--- a/offload/src/PluginManager.cpp
+++ b/offload/src/PluginManager.cpp
@@ -23,9 +23,6 @@ using namespace llvm::sys;
 
 PluginManager *PM = nullptr;
 
-// List of all plugins that can support offloading.
-static const char *RTLNames[] = {ENABLED_OFFLOAD_PLUGINS};
-
 Expected<std::unique_ptr<PluginAdaptorTy>>
 PluginAdaptorTy::create(const std::string &Name) {
   DP("Attempting to load library '%s'...\n", Name.c_str());
@@ -95,17 +92,19 @@ void PluginManager::init() {
 
   // Attempt to open all the plugins and, if they exist, check if the interface
   // is correct and if they are supporting any devices.
-  for (const char *Name : RTLNames) {
-    auto PluginAdaptorOrErr =
-        PluginAdaptorTy::create(std::string(Name) + ".so");
-    if (!PluginAdaptorOrErr) {
-      [[maybe_unused]] std::string InfoMsg =
-          toString(PluginAdaptorOrErr.takeError());
-      DP("%s", InfoMsg.c_str());
-    } else {
-      PluginAdaptors.push_back(std::move(*PluginAdaptorOrErr));
-    }
-  }
+#define PLUGIN_TARGET(Name)                                                    \
+  do {                                                                         \
+    auto PluginAdaptorOrErr =                                                  \
+        PluginAdaptorTy::create("libomptarget.rtl." #Name ".so");              \
+    if (!PluginAdaptorOrErr) {                                                 \
+      [[maybe_unused]] std::string InfoMsg =                                   \
+          toString(PluginAdaptorOrErr.takeError());                            \
+      DP("%s", InfoMsg.c_str());                                               \
+    } else {                                                                   \
+      PluginAdaptors.push_back(std::move(*PluginAdaptorOrErr));                \
+    }                                                                          \
+  } while (false);
+#include "Shared/Targets.def"
 
   DP("RTLs loaded!\n");
 }
diff --git a/offload/test/unified_shared_memory/api.c b/offload/test/unified_shared_memory/api.c
index c7ab055abb51..b938971b4b03 100644
--- a/offload/test/unified_shared_memory/api.c
+++ b/offload/test/unified_shared_memory/api.c
@@ -9,11 +9,6 @@
 #include <omp.h>
 #include <stdio.h>
 
-// ---------------------------------------------------------------------------
-// Various definitions copied from OpenMP RTL
-
-extern void __tgt_register_requires(int64_t);
-
 // End of definitions copied from OpenMP RTL.
 // ---------------------------------------------------------------------------
 
@@ -32,10 +27,6 @@ void init(int A[], int B[], int C[]) {
 int main(int argc, char *argv[]) {
   const int device = omp_get_default_device();
 
-  // Manual registration of requires flags for Clang versions
-  // that do not support requires.
-  __tgt_register_requires(8);
-
   // CHECK: Initial device: [[INITIAL_DEVICE:[0-9]+]]
   printf("Initial device: %d\n", omp_get_initial_device());
   // CHECK: Num devices: [[INITIAL_DEVICE]]
diff --git a/offload/test/unified_shared_memory/close_manual.c b/offload/test/unified_shared_memory/close_manual.c
index 9985e822c05d..c588cb1c403a 100644
--- a/offload/test/unified_shared_memory/close_manual.c
+++ b/offload/test/unified_shared_memory/close_manual.c
@@ -8,8 +8,6 @@
 // ---------------------------------------------------------------------------
 // Various definitions copied from OpenMP RTL
 
-extern void __tgt_register_requires(int64_t);
-
 extern void __tgt_target_data_begin(int64_t device_id, int32_t arg_num,
                                     void **args_base, void **args,
                                     int64_t *arg_sizes, int64_t *arg_types);
@@ -30,10 +28,6 @@ int main(int argc, char *argv[]) {
   void *host_alloc = 0, *device_alloc = 0;
   int *a = (int *)malloc(N * sizeof(int));
 
-  // Manual registration of requires flags for Clang versions
-  // that do not support requires.
-  __tgt_register_requires(8);
-
   // Init
   for (int i = 0; i < N; ++i) {
     a[i] = 10;
diff --git a/offload/test/unified_shared_memory/shared_update.c b/offload/test/unified_shared_memory/shared_update.c
index 65db9e4f6bdc..f8eb11d56a6c 100644
--- a/offload/test/unified_shared_memory/shared_update.c
+++ b/offload/test/unified_shared_memory/shared_update.c
@@ -11,11 +11,6 @@
 #include <omp.h>
 #include <stdio.h>
 
-// ---------------------------------------------------------------------------
-// Various definitions copied from OpenMP RTL
-
-extern void __tgt_register_requires(int64_t);
-
 // End of definitions copied from OpenMP RTL.
 // ---------------------------------------------------------------------------
 
@@ -30,10 +25,6 @@ int main(int argc, char *argv[]) {
   int *alloc = (int *)malloc(N * sizeof(int));
   int data[N];
 
-  // Manual registration of requires flags for Clang versions
-  // that do not support requires.
-  __tgt_register_requires(8);
-
   for (int i = 0; i < N; ++i) {
     alloc[i] = 10;
     data[i] = 1;
diff --git a/openmp/runtime/src/kmp_affinity.cpp b/openmp/runtime/src/kmp_affinity.cpp
index 378e5aa296c4..f34e55555545 100644
--- a/openmp/runtime/src/kmp_affinity.cpp
+++ b/openmp/runtime/src/kmp_affinity.cpp
@@ -3038,15 +3038,7 @@ static bool __kmp_affinity_create_cpuinfo_map(int *line,
     KMP_INFORM(AffParseFilename, "KMP_AFFINITY", "system info for topology");
 
   // Get the number of SMT threads per core.
-  int retval =
-      lpar_get_info(LPAR_INFO_FORMAT1, &cpuinfo, sizeof(lpar_info_format1_t));
-  if (!retval)
-    smt_threads = cpuinfo.smt_threads;
-  else {
-    CLEANUP_THREAD_INFO;
-    *msg_id = kmp_i18n_str_UnknownTopology;
-    return false;
-  }
+  smt_threads = syssmt(GET_NUMBER_SMT_SETS, 0, 0, NULL);
 
   // Allocate a resource set containing available system resourses.
   rsethandle_t sys_rset = rs_alloc(RS_SYSTEM);
diff --git a/openmp/runtime/src/kmp_affinity.h b/openmp/runtime/src/kmp_affinity.h
index 8e9e7667eb90..3dc2c84d53f7 100644
--- a/openmp/runtime/src/kmp_affinity.h
+++ b/openmp/runtime/src/kmp_affinity.h
@@ -322,6 +322,8 @@ public:
 #include <sys/dr.h>
 #include <sys/rset.h>
 #define VMI_MAXRADS 64 // Maximum number of RADs allowed by AIX.
+#define GET_NUMBER_SMT_SETS 0x0004
+extern "C" int syssmt(int flags, int, int, int *);
 #endif
 class KMPNativeAffinity : public KMPAffinity {
   class Mask : public KMPAffinity::Mask {
diff --git a/utils/bazel/llvm-project-overlay/libc/BUILD.bazel b/utils/bazel/llvm-project-overlay/libc/BUILD.bazel
index 3df8341712aa..aa9f665c350a 100644
--- a/utils/bazel/llvm-project-overlay/libc/BUILD.bazel
+++ b/utils/bazel/llvm-project-overlay/libc/BUILD.bazel
@@ -23,17 +23,6 @@ package(
 
 licenses(["notice"])
 
-PRINTF_COPTS = [
-    "LIBC_COPT_STDIO_USE_SYSTEM_FILE",
-    "LIBC_COPT_PRINTF_DISABLE_WRITE_INT",
-]
-
-MEMORY_COPTS = [
-    # "LIBC_COPT_MEMCPY_X86_USE_REPMOVSB_FROM_SIZE=0",
-    # "LIBC_COPT_MEMCPY_X86_USE_SOFTWARE_PREFETCHING",
-    # "LIBC_COPT_MEMSET_X86_USE_SOFTWARE_PREFETCHING",
-]
-
 # A flag to pick which `mpfr` to use for math tests.
 # Usage: `--@llvm-project//libc:mpfr=<disable|external|system>`.
 # Flag documentation: https://bazel.build/extending/config
@@ -2421,7 +2410,6 @@ libc_support_library(
         "src/string/memory_utils/op_x86.h",
         "src/string/memory_utils/utils.h",
     ],
-    defines = MEMORY_COPTS,
     textual_hdrs = [
         "src/string/memory_utils/aarch64/inline_bcmp.h",
         "src/string/memory_utils/aarch64/inline_memcmp.h",
@@ -3191,7 +3179,6 @@ libc_function(
 libc_support_library(
     name = "printf_config",
     hdrs = ["src/stdio/printf_core/printf_config.h"],
-    defines = PRINTF_COPTS,
     deps = [
     ],
 )
@@ -3199,7 +3186,6 @@ libc_support_library(
 libc_support_library(
     name = "printf_core_structs",
     hdrs = ["src/stdio/printf_core/core_structs.h"],
-    defines = PRINTF_COPTS,
     deps = [
         ":__support_cpp_string_view",
         ":__support_fputil_fp_bits",
@@ -3210,7 +3196,6 @@ libc_support_library(
 libc_support_library(
     name = "printf_parser",
     hdrs = ["src/stdio/printf_core/parser.h"],
-    defines = PRINTF_COPTS,
     deps = [
         ":__support_arg_list",
         ":__support_common",
@@ -3231,7 +3216,7 @@ libc_support_library(
 libc_support_library(
     name = "printf_mock_parser",
     hdrs = ["src/stdio/printf_core/parser.h"],
-    defines = PRINTF_COPTS + ["LIBC_COPT_MOCK_ARG_LIST"],
+    local_defines = ["LIBC_COPT_MOCK_ARG_LIST"],
     deps = [
         ":__support_arg_list",
         ":__support_common",
@@ -3251,7 +3236,6 @@ libc_support_library(
     name = "printf_writer",
     srcs = ["src/stdio/printf_core/writer.cpp"],
     hdrs = ["src/stdio/printf_core/writer.h"],
-    defines = PRINTF_COPTS,
     deps = [
         ":__support_cpp_string_view",
         ":__support_macros_optimization",
@@ -3276,7 +3260,6 @@ libc_support_library(
         "src/stdio/printf_core/string_converter.h",
         "src/stdio/printf_core/write_int_converter.h",
     ],
-    defines = PRINTF_COPTS,
     deps = [
         ":__support_big_int",
         ":__support_common",
@@ -3300,7 +3283,6 @@ libc_support_library(
     name = "printf_main",
     srcs = ["src/stdio/printf_core/printf_main.cpp"],
     hdrs = ["src/stdio/printf_core/printf_main.h"],
-    defines = PRINTF_COPTS,
     deps = [
         ":__support_arg_list",
         ":printf_converter",
@@ -3313,7 +3295,6 @@ libc_support_library(
 libc_support_library(
     name = "vfprintf_internal",
     hdrs = ["src/stdio/printf_core/vfprintf_internal.h"],
-    defines = PRINTF_COPTS,
     deps = [
         ":__support_arg_list",
         ":__support_file_file",
@@ -3327,7 +3308,6 @@ libc_function(
     name = "sprintf",
     srcs = ["src/stdio/sprintf.cpp"],
     hdrs = ["src/stdio/sprintf.h"],
-    defines = PRINTF_COPTS,
     deps = [
         ":__support_arg_list",
         ":__support_cpp_limits",
@@ -3341,7 +3321,6 @@ libc_function(
     name = "snprintf",
     srcs = ["src/stdio/snprintf.cpp"],
     hdrs = ["src/stdio/snprintf.h"],
-    defines = PRINTF_COPTS,
     deps = [
         ":__support_arg_list",
         ":errno",
@@ -3354,7 +3333,6 @@ libc_function(
     name = "printf",
     srcs = ["src/stdio/printf.cpp"],
     hdrs = ["src/stdio/printf.h"],
-    defines = PRINTF_COPTS,
     deps = [
         ":__support_arg_list",
         ":__support_file_file",
@@ -3367,7 +3345,6 @@ libc_function(
     name = "fprintf",
     srcs = ["src/stdio/fprintf.cpp"],
     hdrs = ["src/stdio/fprintf.h"],
-    defines = PRINTF_COPTS,
     deps = [
         ":__support_arg_list",
         ":__support_file_file",
@@ -3380,7 +3357,6 @@ libc_function(
     name = "vsprintf",
     srcs = ["src/stdio/vsprintf.cpp"],
     hdrs = ["src/stdio/vsprintf.h"],
-    defines = PRINTF_COPTS,
     deps = [
         ":__support_arg_list",
         ":__support_cpp_limits",
@@ -3394,7 +3370,6 @@ libc_function(
     name = "vsnprintf",
     srcs = ["src/stdio/vsnprintf.cpp"],
     hdrs = ["src/stdio/vsnprintf.h"],
-    defines = PRINTF_COPTS,
     deps = [
         ":__support_arg_list",
         ":errno",
@@ -3407,7 +3382,6 @@ libc_function(
     name = "vprintf",
     srcs = ["src/stdio/vprintf.cpp"],
     hdrs = ["src/stdio/vprintf.h"],
-    defines = PRINTF_COPTS,
     deps = [
         ":__support_arg_list",
         ":__support_file_file",
@@ -3420,7 +3394,6 @@ libc_function(
     name = "vfprintf",
     srcs = ["src/stdio/vfprintf.cpp"],
     hdrs = ["src/stdio/vfprintf.h"],
-    defines = PRINTF_COPTS,
     deps = [
         ":__support_arg_list",
         ":__support_file_file",
diff --git a/utils/bazel/llvm-project-overlay/libc/libc_build_rules.bzl b/utils/bazel/llvm-project-overlay/libc/libc_build_rules.bzl
index be59e18ffd89..ec3714407cb9 100644
--- a/utils/bazel/llvm-project-overlay/libc/libc_build_rules.bzl
+++ b/utils/bazel/llvm-project-overlay/libc/libc_build_rules.bzl
@@ -6,6 +6,7 @@
 
 load("@bazel_skylib//lib:paths.bzl", "paths")
 load("@bazel_skylib//lib:selects.bzl", "selects")
+load(":libc_configure_options.bzl", "LIBC_CONFIGURE_OPTIONS")
 load(":libc_namespace.bzl", "LIBC_NAMESPACE")
 load(":platforms.bzl", "PLATFORM_CPU_ARM64", "PLATFORM_CPU_X86_64")
 
@@ -21,13 +22,14 @@ def libc_common_copts():
         "-DLIBC_NAMESPACE=" + LIBC_NAMESPACE,
     ]
 
-def _libc_library(name, hidden, copts = [], deps = [], **kwargs):
+def _libc_library(name, hidden, copts = [], deps = [], local_defines = [], **kwargs):
     """Internal macro to serve as a base for all other libc library rules.
 
     Args:
       name: Target name.
       copts: The special compiler options for the target.
       deps: The list of target dependencies if any.
+      local_defines: The list of target local_defines if any.
       hidden: Whether the symbols should be explicitly hidden or not.
       **kwargs: All other attributes relevant for the cc_library rule.
     """
@@ -40,6 +42,7 @@ def _libc_library(name, hidden, copts = [], deps = [], **kwargs):
     native.cc_library(
         name = name,
         copts = copts + libc_common_copts(),
+        local_defines = local_defines + LIBC_CONFIGURE_OPTIONS,
         deps = deps,
         linkstatic = 1,
         **kwargs
diff --git a/utils/bazel/llvm-project-overlay/libc/libc_configure_options.bzl b/utils/bazel/llvm-project-overlay/libc/libc_configure_options.bzl
new file mode 100644
index 000000000000..f780c323d9a9
--- /dev/null
+++ b/utils/bazel/llvm-project-overlay/libc/libc_configure_options.bzl
@@ -0,0 +1,49 @@
+# This file is licensed under the Apache License v2.0 with LLVM Exceptions.
+# See https://llvm.org/LICENSE.txt for license information.
+# SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+
+"""LLVM libc configuration options.
+The canonical list of user options is in 'libc/config/config.json'.
+These options are then processed by CMake and turned into preprocessor
+definitions. We don't have this logic in Bazel yet but the list of definitions
+is discoverable with the following command:
+
+> git grep -hoE '\bLIBC_COPT_\\w*'  -- '*.h' '*.cpp' | sort -u
+"""
+
+# This list of definitions is used to customize LLVM libc.
+LIBC_CONFIGURE_OPTIONS = [
+    # Documentation in libc/docs/dev/printf_behavior.rst
+    # "LIBC_COPT_FLOAT_TO_STR_NO_SPECIALIZE_LD",
+    # "LIBC_COPT_FLOAT_TO_STR_NO_TABLE",
+    # "LIBC_COPT_FLOAT_TO_STR_USE_DYADIC_FLOAT",
+    # "LIBC_COPT_FLOAT_TO_STR_USE_DYADIC_FLOAT_LD",
+    # "LIBC_COPT_FLOAT_TO_STR_USE_INT_CALC",
+    # "LIBC_COPT_FLOAT_TO_STR_USE_MEGA_LONG_DOUBLE_TABLE",
+
+    # Documentation in libc/src/string/memory_utils/...
+    # "LIBC_COPT_MEMCPY_USE_EMBEDDED_TINY",
+    # "LIBC_COPT_MEMCPY_X86_USE_REPMOVSB_FROM_SIZE",
+    # "LIBC_COPT_MEMCPY_X86_USE_SOFTWARE_PREFETCHING",
+    # "LIBC_COPT_MEMSET_X86_USE_SOFTWARE_PREFETCHING",
+
+    # Documentation in libc/docs/dev/printf_behavior.rst
+    # "LIBC_COPT_PRINTF_CONV_ATLAS",
+    # "LIBC_COPT_PRINTF_DISABLE_FIXED_POINT",
+    # "LIBC_COPT_PRINTF_DISABLE_FLOAT",
+    # "LIBC_COPT_PRINTF_DISABLE_INDEX_MODE",
+    "LIBC_COPT_PRINTF_DISABLE_WRITE_INT",
+    # "LIBC_COPT_PRINTF_HEX_LONG_DOUBLE",
+    # "LIBC_COPT_PRINTF_INDEX_ARR_LEN",
+    # "LIBC_COPT_PRINTF_NO_NULLPTR_CHECKS",
+    # "LIBC_COPT_SCANF_DISABLE_FLOAT",
+    # "LIBC_COPT_SCANF_DISABLE_INDEX_MODE",
+    "LIBC_COPT_STDIO_USE_SYSTEM_FILE",
+    # "LIBC_COPT_STRING_UNSAFE_WIDE_READ",
+    # "LIBC_COPT_STRTOFLOAT_DISABLE_CLINGER_FAST_PATH",
+    # "LIBC_COPT_STRTOFLOAT_DISABLE_EISEL_LEMIRE",
+    # "LIBC_COPT_STRTOFLOAT_DISABLE_SIMPLE_DECIMAL_CONVERSION",
+
+    # Documentation in libc/src/__support/libc_assert.h
+    # "LIBC_COPT_USE_C_ASSERT",
+]
diff --git a/utils/bazel/llvm-project-overlay/libc/test/UnitTest/BUILD.bazel b/utils/bazel/llvm-project-overlay/libc/test/UnitTest/BUILD.bazel
index e55804b1ea79..6126a4a8fca8 100644
--- a/utils/bazel/llvm-project-overlay/libc/test/UnitTest/BUILD.bazel
+++ b/utils/bazel/llvm-project-overlay/libc/test/UnitTest/BUILD.bazel
@@ -65,10 +65,12 @@ libc_support_library(
 libc_support_library(
     name = "fp_test_helpers",
     srcs = [
+        "FEnvSafeTest.cpp",
         "FPExceptMatcher.cpp",
         "RoundingModeUtils.cpp",
     ],
     hdrs = [
+        "FEnvSafeTest.h",
         "FPExceptMatcher.h",
         "FPMatcher.h",
         "RoundingModeUtils.h",
@@ -82,13 +84,15 @@ libc_support_library(
         "//libc:__support_cpp_bitset",
         "//libc:__support_cpp_span",
         "//libc:__support_cpp_type_traits",
+        "//libc:__support_cpp_utility",
         "//libc:__support_fputil_fenv_impl",
         "//libc:__support_fputil_fp_bits",
         "//libc:__support_fputil_fpbits_str",
         "//libc:__support_fputil_rounding_mode",
+        "//libc:__support_macros_properties_architectures",
         "//libc:hdr_math_macros",
-	"//libc:hdr_fenv_macros",
-	"//libc:types_fenv_t",
+	    "//libc:hdr_fenv_macros",
+	    "//libc:types_fenv_t",
     ],
 )
 
diff --git a/utils/bazel/llvm-project-overlay/libc/test/libc_test_rules.bzl b/utils/bazel/llvm-project-overlay/libc/test/libc_test_rules.bzl
index 18056bacfd50..ae24a41c6041 100644
--- a/utils/bazel/llvm-project-overlay/libc/test/libc_test_rules.bzl
+++ b/utils/bazel/llvm-project-overlay/libc/test/libc_test_rules.bzl
@@ -13,8 +13,9 @@ When performing tests we make sure to always use the internal version.
 """
 
 load("//libc:libc_build_rules.bzl", "libc_common_copts", "libc_internal_target")
+load("//libc:libc_configure_options.bzl", "LIBC_CONFIGURE_OPTIONS")
 
-def libc_test(name, srcs, libc_function_deps = [], copts = [], deps = [], **kwargs):
+def libc_test(name, srcs, libc_function_deps = [], copts = [], deps = [], local_defines = [], **kwargs):
     """Add target for a libc test.
 
     Args:
@@ -23,12 +24,14 @@ def libc_test(name, srcs, libc_function_deps = [], copts = [], deps = [], **kwar
       libc_function_deps: List of libc_function targets used by this test.
       copts: The list of options to add to the C++ compilation command.
       deps: The list of other libraries to be linked in to the test target.
+      local_defines: The list of target local_defines if any.
       **kwargs: Attributes relevant for a libc_test. For example, name, srcs.
     """
     all_function_deps = libc_function_deps + ["//libc:errno"]
     native.cc_test(
         name = name,
         srcs = srcs,
+        local_defines = local_defines + LIBC_CONFIGURE_OPTIONS,
         deps = [libc_internal_target(d) for d in all_function_deps] + [
             "//libc/test/UnitTest:LibcUnitTest",
         ] + deps,
diff --git a/utils/bazel/llvm-project-overlay/libc/test/src/fenv/BUILD.bazel b/utils/bazel/llvm-project-overlay/libc/test/src/fenv/BUILD.bazel
index fc3ab3da3587..03c94d1db23a 100644
--- a/utils/bazel/llvm-project-overlay/libc/test/src/fenv/BUILD.bazel
+++ b/utils/bazel/llvm-project-overlay/libc/test/src/fenv/BUILD.bazel
@@ -12,7 +12,10 @@ licenses(["notice"])
 
 libc_test(
     name = "exception_status_test",
-    srcs = ["exception_status_test.cpp"],
+    srcs = [
+        "exception_status_test.cpp",
+        "excepts.h",
+    ],
     libc_function_deps = [
         "//libc:feclearexcept",
         "//libc:feraiseexcept",
@@ -21,23 +24,33 @@ libc_test(
     ],
     deps = [
         "//libc:__support_fputil_fenv_impl",
-	"//libc:hdr_fenv_macros",
+        "//libc:hdr_fenv_macros",
+        "//libc/test/UnitTest:fp_test_helpers",
     ],
 )
 
 libc_test(
     name = "rounding_mode_test",
-    srcs = ["rounding_mode_test.cpp"],
+    srcs = [
+        "excepts.h",
+        "rounding_mode_test.cpp",
+    ],
     libc_function_deps = [
         "//libc:fegetround",
         "//libc:fesetround",
     ],
-    deps = ["//libc:hdr_fenv_macros"],
+    deps = [
+        "//libc:hdr_fenv_macros",
+        "//libc/test/UnitTest:fp_test_helpers",
+    ],
 )
 
 libc_test(
     name = "enabled_exceptions_test",
-    srcs = ["enabled_exceptions_test.cpp"],
+    srcs = [
+        "enabled_exceptions_test.cpp",
+        "excepts.h",
+    ],
     libc_function_deps = [
         "//libc:feclearexcept",
         "//libc:feraiseexcept",
@@ -48,14 +61,17 @@ libc_test(
         "//libc:__support_common",
         "//libc:__support_fputil_fenv_impl",
         "//libc:__support_macros_properties_architectures",
+        "//libc:hdr_fenv_macros",
         "//libc/test/UnitTest:fp_test_helpers",
-	"//libc:hdr_fenv_macros",
     ],
 )
 
 libc_test(
     name = "feholdexcept_test",
-    srcs = ["feholdexcept_test.cpp"],
+    srcs = [
+        "excepts.h",
+        "feholdexcept_test.cpp",
+    ],
     libc_function_deps = [
         "//libc:feholdexcept",
     ],
@@ -64,14 +80,18 @@ libc_test(
         "//libc:__support_common",
         "//libc:__support_fputil_fenv_impl",
         "//libc:__support_macros_properties_architectures",
+        "//libc:hdr_fenv_macros",
+        "//libc:types_fenv_t",
         "//libc/test/UnitTest:fp_test_helpers",
-	"//libc:types_fenv_t",
     ],
 )
 
 libc_test(
     name = "exception_flags_test",
-    srcs = ["exception_flags_test.cpp"],
+    srcs = [
+        "exception_flags_test.cpp",
+        "excepts.h",
+    ],
     libc_function_deps = [
         "//libc:fegetexceptflag",
         "//libc:fesetexceptflag",
@@ -79,25 +99,34 @@ libc_test(
     ],
     deps = [
         "//libc:__support_fputil_fenv_impl",
-	"//libc:types_fexcept_t",
+        "//libc:hdr_fenv_macros",
+        "//libc:types_fexcept_t",
+        "//libc/test/UnitTest:fp_test_helpers",
     ],
 )
 
 libc_test(
     name = "feclearexcept_test",
-    srcs = ["feclearexcept_test.cpp"],
+    srcs = [
+        "excepts.h",
+        "feclearexcept_test.cpp",
+    ],
     libc_function_deps = [
         "//libc:feclearexcept",
     ],
     deps = [
         "//libc:__support_fputil_fenv_impl",
-	"//libc:hdr_fenv_macros",
+        "//libc:hdr_fenv_macros",
+        "//libc/test/UnitTest:fp_test_helpers",
     ],
 )
 
 libc_test(
     name = "feenableexcept_test",
-    srcs = ["feenableexcept_test.cpp"],
+    srcs = [
+        "excepts.h",
+        "feenableexcept_test.cpp",
+    ],
     libc_function_deps = [
         "//libc:fedisableexcept",
         "//libc:feenableexcept",
@@ -106,25 +135,34 @@ libc_test(
     deps = [
         "//libc:__support_common",
         "//libc:__support_macros_properties_architectures",
-	"//libc:hdr_fenv_macros",
+        "//libc:hdr_fenv_macros",
+        "//libc/test/UnitTest:fp_test_helpers",
     ],
 )
 
 libc_test(
     name = "feupdateenv_test",
-    srcs = ["feupdateenv_test.cpp"],
+    srcs = [
+        "excepts.h",
+        "feupdateenv_test.cpp",
+    ],
     libc_function_deps = [
         "//libc:feupdateenv",
     ],
     deps = [
         "//libc:__support_fputil_fenv_impl",
-	"//libc:types_fenv_t",
+        "//libc:hdr_fenv_macros",
+        "//libc:types_fenv_t",
+        "//libc/test/UnitTest:fp_test_helpers",
     ],
 )
 
 libc_test(
     name = "getenv_and_setenv_test",
-    srcs = ["getenv_and_setenv_test.cpp"],
+    srcs = [
+        "excepts.h",
+        "getenv_and_setenv_test.cpp",
+    ],
     libc_function_deps = [
         "//libc:fegetenv",
         "//libc:fegetround",
@@ -133,6 +171,8 @@ libc_test(
     ],
     deps = [
         "//libc:__support_fputil_fenv_impl",
-	"//libc:types_fenv_t",
+        "//libc:hdr_fenv_macros",
+        "//libc:types_fenv_t",
+        "//libc/test/UnitTest:fp_test_helpers",
     ],
 )
diff --git a/utils/bazel/llvm-project-overlay/libc/test/src/math/BUILD.bazel b/utils/bazel/llvm-project-overlay/libc/test/src/math/BUILD.bazel
index e30c8bf023cf..4f72a0a8e186 100644
--- a/utils/bazel/llvm-project-overlay/libc/test/src/math/BUILD.bazel
+++ b/utils/bazel/llvm-project-overlay/libc/test/src/math/BUILD.bazel
@@ -298,6 +298,7 @@ libc_support_library(
         "//libc:__support_fputil_fp_bits",
         "//libc:__support_fputil_manipulation_functions",
         "//libc:hdr_math_macros",
+        "//libc/test/UnitTest:fp_test_helpers",
         "//libc/test/UnitTest:LibcUnitTest",
     ],
 )
diff --git a/utils/bazel/llvm-project-overlay/lldb/source/Plugins/BUILD.bazel b/utils/bazel/llvm-project-overlay/lldb/source/Plugins/BUILD.bazel
index e0907a838148..13fec77fe567 100644
--- a/utils/bazel/llvm-project-overlay/lldb/source/Plugins/BUILD.bazel
+++ b/utils/bazel/llvm-project-overlay/lldb/source/Plugins/BUILD.bazel
@@ -2092,6 +2092,10 @@ cc_library(
     srcs = glob(["Process/Linux/*.cpp"]),
     hdrs = glob(["Process/Linux/*.h"]),
     include_prefix = "Plugins",
+    target_compatible_with = select({
+        "@platforms//os:linux": [],
+        "//conditions:default": ["@platforms//:incompatible"],
+    }),
     deps = [
         ":PluginProcessPOSIX",
         ":PluginProcessUtility",
diff --git a/utils/bazel/llvm-project-overlay/llvm/BUILD.bazel b/utils/bazel/llvm-project-overlay/llvm/BUILD.bazel
index c9c0edb1dc31..f45f057b63c2 100644
--- a/utils/bazel/llvm-project-overlay/llvm/BUILD.bazel
+++ b/utils/bazel/llvm-project-overlay/llvm/BUILD.bazel
@@ -636,6 +636,7 @@ cc_binary(
     name = "llvm-min-tblgen",
     srcs = [
         "utils/TableGen/Attributes.cpp",
+        "utils/TableGen/ARMTargetDefEmitter.cpp",
         "utils/TableGen/Basic/CodeGenIntrinsics.cpp",
         "utils/TableGen/Basic/CodeGenIntrinsics.h",
         "utils/TableGen/Basic/SDNodeProperties.cpp",
@@ -1232,6 +1233,30 @@ cc_library(
 )
 
 gentbl(
+    name = "ARMTargetParserDefGen",
+    tbl_outs = [("-gen-arm-target-def", "include/llvm/TargetParser/ARMTargetParserDef.inc")],
+    tblgen = ":llvm-min-tblgen",
+    td_file = "lib/Target/ARM/ARM.td",
+    td_srcs = [
+        ":common_target_td_sources",
+    ] + glob([
+        "lib/Target/ARM/**/*.td",
+    ]),
+)
+
+gentbl(
+    name = "AArch64TargetParserDefGen",
+    tbl_outs = [("-gen-arm-target-def", "include/llvm/TargetParser/AArch64TargetParserDef.inc")],
+    tblgen = ":llvm-min-tblgen",
+    td_file = "lib/Target/AArch64/AArch64.td",
+    td_srcs = [
+        ":common_target_td_sources",
+    ] + glob([
+        "lib/Target/AArch64/**/*.td",
+    ]),
+)
+
+gentbl(
     name = "RISCVTargetParserDefGen",
     tbl_outs = [("-gen-riscv-target-def", "include/llvm/TargetParser/RISCVTargetParserDef.inc")],
     tblgen = ":llvm-min-tblgen",
@@ -1261,6 +1286,8 @@ cc_library(
     copts = llvm_copts,
     includes = ["include"],
     textual_hdrs = [
+        "include/llvm/TargetParser/ARMTargetParserDef.inc",
+        "include/llvm/TargetParser/AArch64TargetParserDef.inc",
         "include/llvm/TargetParser/RISCVTargetParserDef.inc",
     ] + glob([
         "include/llvm/TargetParser/*.def",
@@ -3084,12 +3111,26 @@ cc_library(
 )
 
 cc_library(
-    name = "llvm-mca-headers",
+    name = "MCAApplication",
+    srcs = glob([
+        "tools/llvm-mca/Views/*.cpp",
+    ]) + [
+        mca_source
+        for mca_source in glob(["tools/llvm-mca/*.cpp"])
+        if mca_source != "tools/llvm-mca/llvm-mca.cpp"
+    ],
     hdrs = glob([
         "tools/llvm-mca/*.h",
         "tools/llvm-mca/Views/*.h",
     ]),
     strip_include_prefix = "tools/llvm-mca",
+    deps = [
+        ":MC",
+        ":MCA",
+        ":MCParser",
+        ":Support",
+        ":TargetParser",
+    ],
 )
 
 cc_library(
@@ -4007,12 +4048,9 @@ cc_binary(
 
 cc_binary(
     name = "llvm-mca",
-    srcs = glob([
-        "tools/llvm-mca/*.cpp",
-        "tools/llvm-mca/*.h",
-        "tools/llvm-mca/Views/*.cpp",
-        "tools/llvm-mca/Views/*.h",
-    ]),
+    srcs =[
+        "tools/llvm-mca/llvm-mca.cpp",
+    ],
     copts = llvm_copts,
     stamp = 0,
     deps = [
@@ -4022,10 +4060,10 @@ cc_binary(
         ":AllTargetsMCAs",
         ":MC",
         ":MCA",
+        ":MCAApplication",
         ":MCParser",
         ":Support",
         ":TargetParser",
-        ":llvm-mca-headers",
     ],
 )
 
diff --git a/utils/bazel/llvm-project-overlay/llvm/unittests/BUILD.bazel b/utils/bazel/llvm-project-overlay/llvm/unittests/BUILD.bazel
index dd42f84d16dc..9be26ab551b0 100644
--- a/utils/bazel/llvm-project-overlay/llvm/unittests/BUILD.bazel
+++ b/utils/bazel/llvm-project-overlay/llvm/unittests/BUILD.bazel
@@ -811,3 +811,29 @@ cc_test(
         "//third-party/unittest:gtest_main",
     ],
 )
+
+cc_test(
+    name = "llvm_mca_tests",
+    size = "small",
+    srcs = glob(
+        [
+            "tools/llvm-mca/*.cpp",
+            "tools/llvm-mca/*.h",
+            "tools/llvm-mca/X86/*.cpp",
+            "tools/llvm-mca/X86/*.h",
+        ],
+        allow_empty = False,
+    ),
+    includes = ["tools/llvm-mca"],
+    deps = [
+        "//llvm:MC",
+        "//llvm:MCA",
+        "//llvm:MCAApplication",
+        "//llvm:Support",
+        "//llvm:TargetParser",
+        "//llvm:X86CodeGen",
+        "//llvm:X86UtilsAndDesc",
+        "//third-party/unittest:gtest",
+        "//third-party/unittest:gtest_main",
+    ],
+)
diff --git a/utils/bazel/llvm-project-overlay/mlir/BUILD.bazel b/utils/bazel/llvm-project-overlay/mlir/BUILD.bazel
index 6c732b8f1349..52c874c344c5 100644
--- a/utils/bazel/llvm-project-overlay/mlir/BUILD.bazel
+++ b/utils/bazel/llvm-project-overlay/mlir/BUILD.bazel
@@ -5,6 +5,7 @@
 # Description:
 #   The MLIR "Multi-Level Intermediate Representation" Compiler Infrastructure
 
+load("@bazel_skylib//rules:common_settings.bzl", "bool_flag")
 load("@bazel_skylib//rules:expand_template.bzl", "expand_template")
 load(
     ":build_defs.bzl",
@@ -28,6 +29,18 @@ exports_files([
     "utils/textmate/mlir.json",
 ])
 
+bool_flag(
+    name = "enable_cuda",
+    build_setting_default = False,
+)
+
+config_setting(
+    name = "enable_cuda_config",
+    flag_values = {
+        ":enable_cuda": "True",
+    },
+)
+
 expand_template(
     name = "mlir_config_h_gen",
     out = "include/mlir/Config/mlir-config.h",
@@ -3640,14 +3653,6 @@ gentbl_cc_library(
             "include/mlir/Dialect/XeGPU/IR/XeGPUTypes.cpp.inc",
         ),
         (
-            ["-gen-enum-decls"],
-            "include/mlir/Dialect/XeGPU/IR/XeGPUEnums.h.inc",
-        ),
-        (
-            ["-gen-enum-defs"],
-            "include/mlir/Dialect/XeGPU/IR/XeGPUEnums.cpp.inc",
-        ),
-        (
             [
                 "-gen-attrdef-decls",
                 "-attrdefs-dialect=xegpu",
@@ -3664,7 +3669,44 @@ gentbl_cc_library(
     ],
     tblgen = ":mlir-tblgen",
     td_file = "include/mlir/Dialect/XeGPU/IR/XeGPU.td",
-    deps = [":XeGPUTdFiles"],
+    deps = [
+        ":ArithOpsTdFiles",
+        ":XeGPUTdFiles",
+    ],
+)
+
+td_library(
+    name = "XeGPUAttrTdFiles",
+    srcs = [
+        "include/mlir/Dialect/XeGPU/IR/XeGPUAttrs.td",
+        "include/mlir/Dialect/XeGPU/IR/XeGPUDialect.td",
+    ],
+    includes = ["include"],
+    deps = [
+        ":BuiltinDialectTdFiles",
+        ":OpBaseTdFiles",
+        ":ShapedOpInterfacesTdFiles",
+        ":ViewLikeInterfaceTdFiles",
+    ],
+)
+
+# Separated from the XeGPUIncGen target because the enum declaration causes
+# duplicate declarations with the Arith enums.
+gentbl_cc_library(
+    name = "XeGPUEnumsIncGen",
+    tbl_outs = [
+        (
+            ["-gen-enum-decls"],
+            "include/mlir/Dialect/XeGPU/IR/XeGPUEnums.h.inc",
+        ),
+        (
+            ["-gen-enum-defs"],
+            "include/mlir/Dialect/XeGPU/IR/XeGPUEnums.cpp.inc",
+        ),
+    ],
+    tblgen = ":mlir-tblgen",
+    td_file = "include/mlir/Dialect/XeGPU/IR/XeGPUAttrs.td",
+    deps = [":XeGPUAttrTdFiles"],
 )
 
 cc_library(
@@ -3676,12 +3718,14 @@ cc_library(
     hdrs = ["include/mlir/Dialect/XeGPU/IR/XeGPU.h"],
     includes = ["include"],
     deps = [
+        ":ArithDialect",
         ":BytecodeOpInterface",
         ":DialectUtils",
         ":IR",
         ":ShapedOpInterfaces",
         ":SideEffectInterfaces",
         ":ViewLikeInterface",
+        ":XeGPUEnumsIncGen",
         ":XeGPUIncGen",
         "//llvm:Support",
     ],
@@ -6653,6 +6697,7 @@ cc_library(
     ]),
     includes = ["include"],
     deps = [
+        ":ArithDialect",
         ":IR",
         ":InferTypeOpInterface",
         ":PolynomialAttributesIncGen",
@@ -9728,6 +9773,15 @@ cc_binary(
 )
 
 cc_binary(
+    name = "mlir-src-sharder",
+    srcs = ["tools/mlir-src-sharder/mlir-src-sharder.cpp"],
+    deps = [
+        ":Support",
+        "//llvm:Support",
+    ],
+)
+
+cc_binary(
     name = "mlir-linalg-ods-yaml-gen",
     srcs = [
         "tools/mlir-linalg-ods-gen/mlir-linalg-ods-yaml-gen.cpp",
@@ -11051,6 +11105,7 @@ cc_library(
         ":FuncTransforms",
         ":GPUDialect",
         ":IR",
+        ":IndexDialect",
         ":LinalgDialect",
         ":LinalgPassIncGen",
         ":LinalgStructuredOpsIncGen",
@@ -11062,6 +11117,7 @@ cc_library(
         ":MeshShardingInterface",
         ":MeshTransforms",
         ":Pass",
+        ":RuntimeVerifiableOpInterface",
         ":SCFDialect",
         ":SCFTransforms",
         ":SCFUtils",
diff --git a/utils/bazel/llvm-project-overlay/mlir/build_defs.bzl b/utils/bazel/llvm-project-overlay/mlir/build_defs.bzl
index fe39dbe21e33..53767fd03c44 100644
--- a/utils/bazel/llvm-project-overlay/mlir/build_defs.bzl
+++ b/utils/bazel/llvm-project-overlay/mlir/build_defs.bzl
@@ -6,7 +6,8 @@
 
 def if_cuda_available(if_true, if_false = []):
     return select({
-        # CUDA is not yet supported.
+        # CUDA auto-detection is not yet supported.
+        "//mlir:enable_cuda_config": if_true,
         "//conditions:default": if_false,
     })
 
diff --git a/utils/bazel/llvm-project-overlay/mlir/python/BUILD.bazel b/utils/bazel/llvm-project-overlay/mlir/python/BUILD.bazel
index 3cc672476014..add150de69fa 100644
--- a/utils/bazel/llvm-project-overlay/mlir/python/BUILD.bazel
+++ b/utils/bazel/llvm-project-overlay/mlir/python/BUILD.bazel
@@ -340,13 +340,6 @@ gentbl_filegroup(
     tbl_outs = [
         (
             [
-                "-gen-python-enum-bindings",
-                "-bind-dialect=async",
-            ],
-            "mlir/dialects/_async_enum_gen.py",
-        ),
-        (
-            [
                 "-gen-python-op-bindings",
                 "-bind-dialect=async",
             ],
diff --git a/utils/bazel/llvm-project-overlay/mlir/tblgen.bzl b/utils/bazel/llvm-project-overlay/mlir/tblgen.bzl
index fdf6a57107ac..e45ba1fe0ef7 100644
--- a/utils/bazel/llvm-project-overlay/mlir/tblgen.bzl
+++ b/utils/bazel/llvm-project-overlay/mlir/tblgen.bzl
@@ -432,3 +432,136 @@ def gentbl_cc_library(
         copts = copts,
         **kwargs
     )
+
+def _gentbl_shard_impl(ctx):
+    args = ctx.actions.args()
+    args.add(ctx.file.src_file)
+    args.add("-op-shard-index", ctx.attr.index)
+    args.add("-o", ctx.outputs.out.path)
+    ctx.actions.run(
+        outputs = [ctx.outputs.out],
+        inputs = [ctx.file.src_file],
+        executable = ctx.executable.sharder,
+        arguments = [args],
+        use_default_shell_env = True,
+        mnemonic = "ShardGenerate",
+    )
+
+gentbl_shard_rule = rule(
+    _gentbl_shard_impl,
+    doc = "",
+    output_to_genfiles = True,
+    attrs = {
+        "index": attr.int(mandatory = True, doc = ""),
+        "sharder": attr.label(
+            doc = "",
+            executable = True,
+            cfg = "exec",
+        ),
+        "src_file": attr.label(
+            doc = "",
+            allow_single_file = True,
+            mandatory = True,
+        ),
+        "out": attr.output(
+            doc = "",
+            mandatory = True,
+        ),
+    },
+)
+
+def gentbl_sharded_ops(
+        name,
+        tblgen,
+        sharder,
+        td_file,
+        shard_count,
+        src_file,
+        src_out,
+        hdr_out,
+        test = False,
+        includes = [],
+        strip_include_prefix = None,
+        deps = []):
+    """Generate sharded op declarations and definitions.
+
+    This special build rule shards op definitions in a TableGen file and generates multiple copies
+    of a template source file for including and compiling each shard. The rule defines a filegroup
+    consisting of the source shards, the generated source file, and the generated header file.
+
+    Args:
+      name: The name of the filegroup.
+      tblgen: The binary used to produce the output.
+      sharder: The source file sharder to use.
+      td_file: The primary table definitions file.
+      shard_count: The number of op definition shards to produce.
+      src_file: The source file template.
+      src_out: The generated source file.
+      hdr_out: The generated header file.
+      test: Whether this is a test target.
+      includes: See gentbl_rule.includes
+      deps: See gentbl_rule.deps
+      strip_include_prefix: Attribute to pass through to cc_library.
+    """
+    cc_lib_name = name + "__gentbl_cc_lib"
+    gentbl_cc_library(
+        name = cc_lib_name,
+        strip_include_prefix = strip_include_prefix,
+        includes = includes,
+        tbl_outs = [
+            (
+                [
+                    "-gen-op-defs",
+                    "-op-shard-count=" + str(shard_count),
+                ],
+                src_out,
+            ),
+            (
+                [
+                    "-gen-op-decls",
+                    "-op-shard-count=" + str(shard_count),
+                ],
+                hdr_out,
+            ),
+        ],
+        tblgen = tblgen,
+        td_file = td_file,
+        test = test,
+        deps = deps,
+    )
+    all_files = [hdr_out, src_out]
+    for i in range(0, shard_count):
+        out_file = "shard_copy_" + str(i) + "_" + src_file
+        gentbl_shard_rule(
+            index = i,
+            name = name + "__src_shard" + str(i),
+            testonly = test,
+            out = out_file,
+            sharder = sharder,
+            src_file = src_file,
+        )
+        all_files.append(out_file)
+    native.filegroup(name = name, srcs = all_files)
+
+def gentbl_sharded_op_defs(name, source_file, shard_count):
+    """Generates multiple copies of a source file that includes sharded op definitions.
+
+    Args:
+      name: The name of the rule.
+      source_file: The source to copy.
+      shard_count: The number of shards.
+
+    Returns:
+      A list of the copied filenames to be included in the dialect library.
+    """
+    copies = []
+    for i in range(0, shard_count):
+        out_file = "shard_copy_" + str(i) + "_" + source_file
+        copies.append(out_file)
+        native.genrule(
+            name = name + "_shard_" + str(i),
+            srcs = [source_file],
+            outs = [out_file],
+            cmd = "echo -e \"#define GET_OP_DEFS_" + str(i) + "\n$$(cat $(SRCS))\" > $(OUTS)",
+        )
+    return copies
diff --git a/utils/bazel/llvm-project-overlay/mlir/test/BUILD.bazel b/utils/bazel/llvm-project-overlay/mlir/test/BUILD.bazel
index dc5f4047c286..0ebfcbe284bd 100644
--- a/utils/bazel/llvm-project-overlay/mlir/test/BUILD.bazel
+++ b/utils/bazel/llvm-project-overlay/mlir/test/BUILD.bazel
@@ -4,7 +4,7 @@
 
 load("@bazel_skylib//rules:expand_template.bzl", "expand_template")
 load("//llvm:lit_test.bzl", "package_path")
-load("//mlir:tblgen.bzl", "gentbl_cc_library", "td_library")
+load("//mlir:tblgen.bzl", "gentbl_cc_library", "gentbl_sharded_ops", "td_library")
 
 package(
     default_visibility = ["//visibility:public"],
@@ -152,14 +152,6 @@ gentbl_cc_library(
     strip_include_prefix = "lib/Dialect/Test",
     tbl_outs = [
         (
-            ["-gen-op-decls"],
-            "lib/Dialect/Test/TestOps.h.inc",
-        ),
-        (
-            ["-gen-op-defs"],
-            "lib/Dialect/Test/TestOps.cpp.inc",
-        ),
-        (
             [
                 "-gen-dialect-decls",
                 "-dialect=test",
@@ -370,12 +362,25 @@ cc_library(
     ],
 )
 
+gentbl_sharded_ops(
+    name = "TestDialectOpSrcs",
+    hdr_out = "lib/Dialect/Test/TestOps.h.inc",
+    shard_count = 20,
+    sharder = "//mlir:mlir-src-sharder",
+    src_file = "lib/Dialect/Test/TestOps.cpp",
+    src_out = "lib/Dialect/Test/TestOps.cpp.inc",
+    tblgen = "//mlir:mlir-tblgen",
+    td_file = "lib/Dialect/Test/TestOps.td",
+    test = True,
+    deps = [":TestOpTdFiles"],
+)
+
 cc_library(
     name = "TestDialect",
     srcs = glob(
         ["lib/Dialect/Test/*.cpp"],
         exclude = ["lib/Dialect/Test/TestToLLVMIRTranslation.cpp"],
-    ),
+    ) + [":TestDialectOpSrcs"],
     hdrs = glob(["lib/Dialect/Test/*.h"]),
     includes = [
         "lib/Dialect/Test",
@@ -418,9 +423,9 @@ cc_library(
         "//mlir:SideEffectInterfaces",
         "//mlir:Support",
         "//mlir:TensorDialect",
-        "//mlir:TranslateLib",
         "//mlir:TransformUtils",
         "//mlir:Transforms",
+        "//mlir:TranslateLib",
         "//mlir:ValueBoundsOpInterface",
         "//mlir:ViewLikeInterface",
     ],
diff --git a/utils/bazel/llvm-project-overlay/mlir/test/Dialect/BUILD.bazel b/utils/bazel/llvm-project-overlay/mlir/test/Dialect/BUILD.bazel
index 1fd6885db8bc..396b89c0a4a3 100644
--- a/utils/bazel/llvm-project-overlay/mlir/test/Dialect/BUILD.bazel
+++ b/utils/bazel/llvm-project-overlay/mlir/test/Dialect/BUILD.bazel
@@ -18,6 +18,7 @@ package(default_visibility = ["//visibility:public"])
         ] + glob([
             "IRDL/*.irdl.mlir",
             "LLVM/*-symbol-def.mlir",
+            "Linalg/transpose-matmul-*.mlir",
             "Transform/include/**/*.mlir",
         ]),
     )
@@ -26,6 +27,7 @@ package(default_visibility = ["//visibility:public"])
         exclude = [
             "IRDL/*.irdl.mlir",
             "LLVM/*-symbol-def.mlir",
+            "Linalg/transpose-matmul-*.mlir",
             "Transform/*-symbol-decl-and-schedule.mlir",
             "Transform/include/**/*.mlir",
         ],